{"contract": {"schema": "noise-screen-v1", "formal_base": {"status": "selected", "method": "hnm", "version": "hnm-v2", "reason": "固定120録音で音量変動の再現誤差が平均約7.2%減少。振幅超過の頻度・程度も踏まえて事後に採否を見直し、雑音包絡を分離する実装を採用。日本語12録音の補助評価も実施。予測MOSや知覚品質の改善を保証するものではない。", "comparison_reference": "hnm", "comparison_reference_meaning": "実験開始時点の唯一の自作ベース。過去結果は当時の版を保持する。既存比較対象はWORLD（Harvest）。", "previous_version": "hnm-v1", "selection_report": "experiments/2026-10-12/jvs-supplement/results/run-001/index.html"}, "baseline_sha256": "ca2f7f56fe4bbf4c3d987c90ac81cd90cf394e4b636ef36988bdf34f220a7153", "sources": {"audio_report.py": "f46ff0cf69e7cb956ea94ee0c7013d44052c0400c5ffc64d7f9f90b4489aa980", "audio_samples.py": "b9b975c1f305cdaff47909b99bd2a3dce64996e90774d2a8305adc1c67ed6c17", "benchmark.py": "10a84680e1641fe5922b741ce05b6538b43b98be9daf27245661fe80bab732ba", "benchmark_io.py": "b27cca4d9457de305be7bbbd84aedf9b0d097791a933e2e9af55eb67e1edbb3d", "benchmark_pipeline.py": "a049034dcd272bbafda47cbf5719cc6b1cd8dae6f80682200cd3fc7b90e522cd", "benchmark_report.py": "c1a8153022a620bc62095f30c5709a34e217739cb9473860dc5b3daff5b87c88", "benchmark_summary.py": "8221086fe08bfc70cbfb9308a3a01d6fea976afb7de86af8cbacd7999e8281da", "build_world.py": "7d29d81c9a48d633edfeae0425c5cc055bc0a10d230232a51eab5720b2cc6974", "calibrate_pitch.py": "282c917bf87c6f549f32467647802852907e4b0b88128eb8017157b84566adec", "control.py": "bc41e73ff34b5f9cb52280a6f2317c989abc53f8b15f8f82db5ee3102b4b58bd", "corpus.py": "bc14cee2c3dd1a651cf2e5f0e9fe871fbd3a712d1a90d7c18a18547a625802e8", "extended_metrics.py": "ab3537d2c8e5d568a0cd2fa8aa36309de09c3bce7cf2f849214331a22fe3624f", "fixtures.py": "a1646c4d0f7c6966ea49a4673ec2c71e75f864943c8908395089d027325eac69", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_diagnostics.py": "a3f47468d10db7e8040c78fc24546043b2eeae6b518f59da1f5904c674aae436", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "937721cd363f06afa3bf4c20c1ff182a8c337c4b19ac9f4b57314f6adea6b44c", "logging_utils.py": "c80eb5d9647516a4bc03980cb178f8b4bb130ac15877d1b3289797e7ad81e381", "methods.py": "b675f7bddf6d19515609713de11fcf4bbdf4c54a3a536030d8da16c3282dd97c", "metrics.py": "b8767feb49494058bd9ec08fd95ae2fb50ce7897fcdd6cebe319ff8a0ac51f22", "mos.py": "fb632aa5440245525289962372db4f3eb007cd2b3b493b2d05ecccb878e85f03", "mos_batch.py": "c7a732dfe52b08fa155201a7134d2f5561a10b16d1866e9f0526f61b2b244f64", "noise_screen.py": "ada7ed38f5c67b7892f1a18af226d738400934607dd4adb9e403b5ca7bdbf667", "noise_stability.py": "624dc434add0086c53c41b9fb7fd0e65b2c51b16c08d4485e4147a77f5534213", "report.py": "cefe7b865cbf768cc497c2fbf8085f0492875abab89dce34ff99831deb180bce", "run.py": "8b2713600f3890dae76d4b2b1b0ef992fdb621679bff3aecc73ca9b68d92e9d2", "site_summary.py": "9c1b03304cc94162deb95bbd465b05c727fa825aed0d1d4ad43c3febc3d76ad7", "supplement.py": "8b72b63b7059f3e94ecb13f575ab93060e00fee0cb89c5abcaa51719061b890d", "supplement_contract.py": "748bff65db3bd3709cdd472b02358af35518c5d1a96fbf212f6b02ba5fa45cd4", "supplement_report.py": "378abee381348bbe0e833d710b899c66d134fd78834e5afbd7a3a71ecdde510f", "temporal_metrics.py": "ce21b6a4767071a7b98ea0cdbfe54bdba15b596bf8d6b6127281a97095e7be91", "world.py": "d14558eed772e07031a9183f57facc55acb1c27deccb3dd4fa098ffcbe0cde3a", "__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d"}, "candidate_sources": {"A": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "2168d3b18d284b2aeac1f57ef4ab847fb4c14bf64fdd1e84b2c72597c790cbd2"}, "B": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "e2478a1afd758cced692e05b83fa88443184d63406af4b5a1985cf8e3b86758b"}, "C": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "640a1f8ae71a296357162c2bbf941bd39465946982c67f7bdb85bdfb1cd8419c"}, "D": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "a4fda7d4cfcaff18f49862b09f5c289d3947a8517f4885b35b281102d953ef0c"}, "E": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "96d7a9ae6af173785b8321ef4dbfc9d19b3d643d538ef074f6adca4357306ee5"}}, "manifest_sha256": "644f4d57483366bb20d8dbcfdaa254e354aed4af8318a6a2984bb2e8b3ddc9c1", "input_records": [{"id": "librispeech_1089-134686-0008", "corpus": "librispeech_test_clean", "speaker": "librispeech_1089", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1089-134686-0008", "transcript": "THE CHAOS IN WHICH HIS ARDOUR EXTINGUISHED ITSELF WAS A COLD INDIFFERENT KNOWLEDGE OF HIMSELF", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1089/134686/1089-134686-0008.flac"}, "sample_rate": 16000, "source_samples": 107680, "source_duration_seconds": 6.73, "source_peak": 0.45428466796875, "source_rms_dbfs": -27.667836551289334, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.30287974095976, "decoded_pcm_sha256": "40f044b90ed06d7cffc4145524ea3e1b8bd36b6599e24e1e905262fbb3e50d8d", "source_sha256": "1bdf7eede41ca2cc8883defb66bbd2e59ec93bf2d587327cb051d0ae427495ae", "wav_sha256": "17a52b57314c689b9e8d2dccefec59b6b2777d0f240fd26d5e940b551cf97c95", "local_source": "audio/librispeech_1089-134686-0008.flac", "local_wav": "audio/librispeech_1089-134686-0008.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1089-134686-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_1089", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1089-134686-0011", "transcript": "ON SATURDAY MORNINGS WHEN THE SODALITY MET IN THE CHAPEL TO RECITE THE LITTLE OFFICE HIS PLACE WAS A CUSHIONED KNEELING DESK AT THE RIGHT OF THE ALTAR FROM WHICH HE LED HIS WING OF BOYS THROUGH THE RESPONSES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1089/134686/1089-134686-0011.flac"}, "sample_rate": 16000, "source_samples": 199120, "source_duration_seconds": 12.445, "source_peak": 0.5145263671875, "source_rms_dbfs": -27.02531260265946, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.932448099275575, "decoded_pcm_sha256": "3db35e4a6ed6a12218e0ab1bfe72f1810bad8f201648ec0e2a9af7dae7aa32eb", "source_sha256": "c65e16be3ef92c11a31a9c2c4c570542eb2ed66153150a7c4cca00c01b30a4e2", "wav_sha256": "f7033475e19dc33371d8718f1495a3ec181845217fe2434773cf6c766ae39dc5", "local_source": "audio/librispeech_1089-134686-0011.flac", "local_wav": "audio/librispeech_1089-134686-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1089-134686-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_1089", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1089-134686-0032", "transcript": "HE IS CALLED AS YOU KNOW THE APOSTLE OF THE INDIES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1089/134686/1089-134686-0032.flac"}, "sample_rate": 16000, "source_samples": 65440, "source_duration_seconds": 4.09, "source_peak": 0.43682861328125, "source_rms_dbfs": -26.667394249066906, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.667394249066906, "decoded_pcm_sha256": "643f787b905d5160255daf650b50e8a272ea88d166a6e7aca2a7ee357cf232fb", "source_sha256": "ba1d2185ad72cd3142e6309865859a8e9e5dfdc91ab046adfcc07338c5a0f0f0", "wav_sha256": "520cdcc0a8dc585f41bbeb35f00cb5ea8279a28f7472c297a5ef1f0757880352", "local_source": "audio/librispeech_1089-134686-0032.flac", "local_wav": "audio/librispeech_1089-134686-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 65440}, {"id": "librispeech_1188-133604-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_1188", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1188-133604-0003", "transcript": "MY FIRST AND PRINCIPAL REASON WAS THAT THEY ENFORCED BEYOND ALL RESISTANCE ON ANY STUDENT WHO MIGHT ATTEMPT TO COPY THEM THIS METHOD OF LAYING PORTIONS OF DISTINCT HUE SIDE BY SIDE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1188/133604/1188-133604-0003.flac"}, "sample_rate": 16000, "source_samples": 201760, "source_duration_seconds": 12.61, "source_peak": 0.29107666015625, "source_rms_dbfs": -29.84254246966504, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.312803528459913, "decoded_pcm_sha256": "81283c3eb08ea201c5ec2d52e3a8e0ed4d0258d76164b522d1ee5ec43edebe81", "source_sha256": "a013400c991a40062cb4f458d247cfd6dac24fa0016bf6e2eac7a3e8e08b8ae8", "wav_sha256": "64ef54500e885de74afaed253f49333ec5284415ca720897ca3b5cded31c617e", "local_source": "audio/librispeech_1188-133604-0003.flac", "local_wav": "audio/librispeech_1188-133604-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1188-133604-0031", "corpus": "librispeech_test_clean", "speaker": "librispeech_1188", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1188-133604-0031", "transcript": "THERE'S ONE AND THERE'S ANOTHER THE DUDLEY AND THE FLINT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1188/133604/1188-133604-0031.flac"}, "sample_rate": 16000, "source_samples": 68000, "source_duration_seconds": 4.25, "source_peak": 0.299102783203125, "source_rms_dbfs": -26.69687768186568, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.69687768186568, "decoded_pcm_sha256": "b739cb4e3f2719b98dd98461d6859b487da84f7d1b9f40d280b7f0eb538a6458", "source_sha256": "08584f8b87cd859e9212f1730979ed141623f35e38995a88ba50747d04cdc222", "wav_sha256": "94599f5aded76bb1de8d31d5c918c6c759fd0febba014767c838658f4127d63e", "local_source": "audio/librispeech_1188-133604-0031.flac", "local_wav": "audio/librispeech_1188-133604-0031.wav", "excerpt_start_sample": 0, "excerpt_samples": 68000}, {"id": "librispeech_1188-133604-0037", "corpus": "librispeech_test_clean", "speaker": "librispeech_1188", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1188-133604-0037", "transcript": "EVERY LINE IN WHICH THE MASTER TRACES IT EVEN WHERE SEEMINGLY NEGLIGENT IS LOVELY AND SET DOWN WITH A MEDITATIVE CALMNESS WHICH MAKES THESE TWO ETCHINGS CAPABLE OF BEING PLACED BESIDE THE MOST TRANQUIL WORK OF HOLBEIN OR DUERER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1188/133604/1188-133604-0037.flac"}, "sample_rate": 16000, "source_samples": 232160, "source_duration_seconds": 14.51, "source_peak": 0.42877197265625, "source_rms_dbfs": -26.333011899815872, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.701166289573592, "decoded_pcm_sha256": "55ff59cb08cac135da664331bc418f7f430ca702cefbb6986e7121d9fe94ca8b", "source_sha256": "94c0626f16e39947864688eb2721a57f838b43caeb1857767a9b4381807368c8", "wav_sha256": "f05a798d7d0c465a9fbd5936dce78da714f9df3c37fdcb61a19556c5153a0050", "local_source": "audio/librispeech_1188-133604-0037.flac", "local_wav": "audio/librispeech_1188-133604-0037.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_121-121726-0012", "corpus": "librispeech_test_clean", "speaker": "librispeech_121", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "121-121726-0012", "transcript": "HUSSY WOMAN AND BOND TIE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/121/121726/121-121726-0012.flac"}, "sample_rate": 16000, "source_samples": 64720, "source_duration_seconds": 4.045, "source_peak": 0.755340576171875, "source_rms_dbfs": -24.740950707596877, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.740950707596877, "decoded_pcm_sha256": "9c9a8032f3e173a5a354dda74e6cb24f7034610d2cd1b6d15dff6a794afd035e", "source_sha256": "6d1fab44fd2e6930a69e1f6f7e52a141ed5290b9a45f56df5e0c3177fa05c6ba", "wav_sha256": "e6e9fea185053d1e5f5604b3e61bf6a51d50cae5fcc82a4e5e7e14838ced2a4d", "local_source": "audio/librispeech_121-121726-0012.flac", "local_wav": "audio/librispeech_121-121726-0012.wav", "excerpt_start_sample": 0, "excerpt_samples": 64720}, {"id": "librispeech_121-123852-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_121", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "121-123852-0000", "transcript": "THOSE PRETTY WRONGS THAT LIBERTY COMMITS WHEN I AM SOMETIME ABSENT FROM THY HEART THY BEAUTY AND THY YEARS FULL WELL BEFITS FOR STILL TEMPTATION FOLLOWS WHERE THOU ART", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/121/123852/121-123852-0000.flac"}, "sample_rate": 16000, "source_samples": 283120, "source_duration_seconds": 17.695, "source_peak": 0.656524658203125, "source_rms_dbfs": -26.637611816594372, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.994738916713484, "decoded_pcm_sha256": "085897f5f5cc183ef4daa8868cb997e49f924fa58877dcfe167b6419325ce812", "source_sha256": "58d6350b52675d6d19616bc0347afc52a1535f8ea8be82aabb7275e6919a2598", "wav_sha256": "feffeb168d26185309000ded59165b27456151ac9fc91fc180467e41eab7e4ee", "local_source": "audio/librispeech_121-123852-0000.flac", "local_wav": "audio/librispeech_121-123852-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_121-127105-0035", "corpus": "librispeech_test_clean", "speaker": "librispeech_121", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "121-127105-0035", "transcript": "SHE PROMISED TO DO THIS AND SHE MENTIONED TO ME THAT WHEN FOR A MOMENT DISBURDENED DELIGHTED HE HELD HER HAND THANKING HER FOR THE SACRIFICE SHE ALREADY FELT REWARDED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/121/127105/121-127105-0035.flac"}, "sample_rate": 16000, "source_samples": 226400, "source_duration_seconds": 14.15, "source_peak": 0.419677734375, "source_rms_dbfs": -29.41904633729958, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.828167012475838, "decoded_pcm_sha256": "48306d9610578521a036cf3d52224eb0076f9d14ddb6850ac09278e898fc7303", "source_sha256": "7e293a1a5f64d336563f850f69929dd6b112e3ce37913f4422f2c29e9c55042d", "wav_sha256": "471c8af18dbbc4b37815971c9bd8c0e583a204a80ab851f2ea450c08140fd69f", "local_source": "audio/librispeech_121-127105-0035.flac", "local_wav": "audio/librispeech_121-127105-0035.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1221-135766-0005", "corpus": "librispeech_test_clean", "speaker": "librispeech_1221", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1221-135766-0005", "transcript": "HESTER COULD ONLY ACCOUNT FOR THE CHILD'S CHARACTER AND EVEN THEN MOST VAGUELY AND IMPERFECTLY BY RECALLING WHAT SHE HERSELF HAD BEEN DURING THAT MOMENTOUS PERIOD WHILE PEARL WAS IMBIBING HER SOUL FROM THE SPIRITUAL WORLD AND HER BODILY FRAME FROM ITS MATERIAL OF EARTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1221/135766/1221-135766-0005.flac"}, "sample_rate": 16000, "source_samples": 266320, "source_duration_seconds": 16.645, "source_peak": 0.37432861328125, "source_rms_dbfs": -29.831601026934116, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.793345815556208, "decoded_pcm_sha256": "ae61e68356f4a720a3d9f729ddcb9fa3e695cdbc220194eb938e92185005daf4", "source_sha256": "57ad0fad287d98afb900ccaa898a6e75348b80a3388201cd3e1feee0a0afc6b1", "wav_sha256": "7795716b196680446728c19c7a356f14b470482cf25a7298ec2b64caba670fd6", "local_source": "audio/librispeech_1221-135766-0005.flac", "local_wav": "audio/librispeech_1221-135766-0005.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1221-135767-0006", "corpus": "librispeech_test_clean", "speaker": "librispeech_1221", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1221-135767-0006", "transcript": "THE MOTHER HERSELF AS IF THE RED IGNOMINY WERE SO DEEPLY SCORCHED INTO HER BRAIN THAT ALL HER CONCEPTIONS ASSUMED ITS FORM HAD CAREFULLY WROUGHT OUT THE SIMILITUDE LAVISHING MANY HOURS OF MORBID INGENUITY TO CREATE AN ANALOGY BETWEEN THE OBJECT OF HER AFFECTION AND THE EMBLEM OF HER GUILT AND TORTURE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1221/135767/1221-135767-0006.flac"}, "sample_rate": 16000, "source_samples": 328960, "source_duration_seconds": 20.56, "source_peak": 0.466552734375, "source_rms_dbfs": -29.808508935576647, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.67998387340972, "decoded_pcm_sha256": "1bf36f00ad8a7e2e9cd6b41efd7f023aab063334b9ad526bda961251493c65cb", "source_sha256": "d70493003bb4a6833163a79d2df440c45bad786d8b50b5c194b896f5a4afd905", "wav_sha256": "135604e3b27df5b5dc13e00735ce8fe6b63247ccc1698c3d65cb047fb9a1e5fc", "local_source": "audio/librispeech_1221-135767-0006.flac", "local_wav": "audio/librispeech_1221-135767-0006.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1221-135767-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_1221", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1221-135767-0019", "transcript": "MOTHER CRIED SHE I SEE YOU HERE LOOK LOOK", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1221/135767/1221-135767-0019.flac"}, "sample_rate": 16000, "source_samples": 60480, "source_duration_seconds": 3.78, "source_peak": 0.397979736328125, "source_rms_dbfs": -26.54945814985232, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.54945814985232, "decoded_pcm_sha256": "0ed57e185be961c721604990dce623ab138f9937e79722c597c3874498c02937", "source_sha256": "245fd9a32160c5fa9dd1a76624a71c7e1398de763f8e9ab88a7dd25aae1002f4", "wav_sha256": "dffedfb5e6881aaaa7fffc0aea511d5ae9815d0718a7632999004dca5c04ce36", "local_source": "audio/librispeech_1221-135767-0019.flac", "local_wav": "audio/librispeech_1221-135767-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 60480}, {"id": "librispeech_1284-1180-0020", "corpus": "librispeech_test_clean", "speaker": "librispeech_1284", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1284-1180-0020", "transcript": "THE FIRST LOT WE TESTED ON OUR GLASS CAT WHICH NOT ONLY BEGAN TO LIVE BUT HAS LIVED EVER SINCE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1284/1180/1284-1180-0020.flac"}, "sample_rate": 16000, "source_samples": 93920, "source_duration_seconds": 5.87, "source_peak": 0.4365234375, "source_rms_dbfs": -24.258485416618605, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.258485416618605, "decoded_pcm_sha256": "0bce114d9020d75b12c3d76b3c232bd8cd315dc052825ca64d389fcb87293422", "source_sha256": "b67810620f4f39a1bca2214b4f6475b98f19452b1cd0a4cc67efee2a8a805016", "wav_sha256": "e2fa9e1870ad76ff19bf8817e638d6c60e1fc83bbbc3577cf3cd7f2b51ced390", "local_source": "audio/librispeech_1284-1180-0020.flac", "local_wav": "audio/librispeech_1284-1180-0020.wav", "excerpt_start_sample": 0, "excerpt_samples": 93920}, {"id": "librispeech_1284-1180-0027", "corpus": "librispeech_test_clean", "speaker": "librispeech_1284", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1284-1180-0027", "transcript": "YET THAT TASK WAS NOT SO EASY AS YOU MAY SUPPOSE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1284/1180/1284-1180-0027.flac"}, "sample_rate": 16000, "source_samples": 52320, "source_duration_seconds": 3.27, "source_peak": 0.374481201171875, "source_rms_dbfs": -23.509845909817045, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.509845909817045, "decoded_pcm_sha256": "96e7e523ecd8732e2ed1542dd9b6cbde8714adfeef42d0c1bfb37fb1a7c1cf83", "source_sha256": "9c4d5811c87e0273494850f3401459de8eafd6731f43a615c07f9ff944262a95", "wav_sha256": "19f2fc30155fbf7c6c393466ad62366f3db8a378533c3fceb06a2668b540c1fc", "local_source": "audio/librispeech_1284-1180-0027.flac", "local_wav": "audio/librispeech_1284-1180-0027.wav", "excerpt_start_sample": 0, "excerpt_samples": 52320}, {"id": "librispeech_1284-1180-0028", "corpus": "librispeech_test_clean", "speaker": "librispeech_1284", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1284-1180-0028", "transcript": "A BED QUILT MADE OF PATCHES OF DIFFERENT KINDS AND COLORS OF CLOTH ALL NEATLY SEWED TOGETHER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1284/1180/1284-1180-0028.flac"}, "sample_rate": 16000, "source_samples": 96720, "source_duration_seconds": 6.045, "source_peak": 0.508331298828125, "source_rms_dbfs": -25.041413930656645, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.009063765320747, "decoded_pcm_sha256": "0bfea5c92c4d2ee444f3984a4d2b53407bcff068a4ef104efccbb0731c86faf4", "source_sha256": "1ee65c4ba89a7fe76bef27dade9b0ab41fe2109b77c8e68d5b80e34194031ee4", "wav_sha256": "877171668e0347c683e8c5d855a388bd8b3df6a214587ba1670ff202b603832a", "local_source": "audio/librispeech_1284-1180-0028.flac", "local_wav": "audio/librispeech_1284-1180-0028.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1320-122617-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_1320", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1320-122617-0019", "transcript": "UNCAS OCCUPIED A DISTANT CORNER IN A RECLINING ATTITUDE BEING RIGIDLY BOUND BOTH HANDS AND FEET BY STRONG AND PAINFUL WITHES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1320/122617/1320-122617-0019.flac"}, "sample_rate": 16000, "source_samples": 131680, "source_duration_seconds": 8.23, "source_peak": 0.569244384765625, "source_rms_dbfs": -21.522908425452997, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.91267260169434, "decoded_pcm_sha256": "8e4bd7f99207ae6f91c9d166201c4877840bfbd64affa6d81029b74db79e47d6", "source_sha256": "3f0ea5a1e33f502dca87255f51487aa60fcb4f12605e9feba2adeb14df1d89f7", "wav_sha256": "dccd68b1d51eb31bdebba14d8c1c3bdd16cb9a1d378941d4fde54eed91ca6ca2", "local_source": "audio/librispeech_1320-122617-0019.flac", "local_wav": "audio/librispeech_1320-122617-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1320-122617-0023", "corpus": "librispeech_test_clean", "speaker": "librispeech_1320", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1320-122617-0023", "transcript": "UNCAS WHO HAD ALREADY APPROACHED THE DOOR IN READINESS TO LEAD THE WAY NOW RECOILED AND PLACED HIMSELF ONCE MORE IN THE BOTTOM OF THE LODGE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1320/122617/1320-122617-0023.flac"}, "sample_rate": 16000, "source_samples": 125040, "source_duration_seconds": 7.815, "source_peak": 0.526458740234375, "source_rms_dbfs": -22.664914502509315, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.27270752547399, "decoded_pcm_sha256": "b3661757a6ed833feab8313fa697ecafb680dc14c418cd468619e460e550745e", "source_sha256": "553fc6594343774831ac876fee34c4ec2b47b53d451f313b10e64eea5cadd8c2", "wav_sha256": "eadf1d183380287ca406621110169ede44028d7ccf2ec2b7c0454cd0ce9baeb5", "local_source": "audio/librispeech_1320-122617-0023.flac", "local_wav": "audio/librispeech_1320-122617-0023.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1320-122617-0039", "corpus": "librispeech_test_clean", "speaker": "librispeech_1320", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1320-122617-0039", "transcript": "THE MOHICAN STARTED ON HIS FEET AND SHOOK HIS SHAGGY COVERING AS THOUGH THE ANIMAL HE COUNTERFEITED WAS ABOUT TO MAKE SOME DESPERATE EFFORT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1320/122617/1320-122617-0039.flac"}, "sample_rate": 16000, "source_samples": 112880, "source_duration_seconds": 7.055, "source_peak": 0.5504150390625, "source_rms_dbfs": -21.698218264348213, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.2918213919449, "decoded_pcm_sha256": "9cf3ed8985625af93ca84569280377a536aa4892a8aefbabb265fc0a052a863c", "source_sha256": "264a53a8c31bd50758ba7eeeb3b5efa9c5bc7908757d44a635bec4c9a4f2c3c0", "wav_sha256": "736879dad172de3ddbbc6cba1cf6b9db0a9b9d447ce01399ec8cd3579d63e366", "local_source": "audio/librispeech_1320-122617-0039.flac", "local_wav": "audio/librispeech_1320-122617-0039.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1580-141083-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_1580", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1580-141083-0010", "transcript": "I GAVE HIM A LITTLE BRANDY AND LEFT HIM COLLAPSED IN A CHAIR WHILE I MADE A MOST CAREFUL EXAMINATION OF THE ROOM", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1580/141083/1580-141083-0010.flac"}, "sample_rate": 16000, "source_samples": 85120, "source_duration_seconds": 5.32, "source_peak": 0.542816162109375, "source_rms_dbfs": -25.39547002291151, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.39547002291151, "decoded_pcm_sha256": "d43cef78fec64a97f03dd03591e102c104a281997670664d424ccf19276cb4a0", "source_sha256": "15b52d8b6b3ab1e0d95b51380278481420b17cdd350b6fe62c29aca97d46a6a7", "wav_sha256": "4498e89acf526da02a12bb3c73380b07b96d180fccbe89e5eb32c295e313221a", "local_source": "audio/librispeech_1580-141083-0010.flac", "local_wav": "audio/librispeech_1580-141083-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 85120}, {"id": "librispeech_1580-141084-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_1580", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1580-141084-0016", "transcript": "MY FRIEND DID NOT APPEAR TO BE DEPRESSED BY HIS FAILURE BUT SHRUGGED HIS SHOULDERS IN HALF HUMOROUS RESIGNATION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1580/141084/1580-141084-0016.flac"}, "sample_rate": 16000, "source_samples": 95360, "source_duration_seconds": 5.96, "source_peak": 0.5615234375, "source_rms_dbfs": -26.925662267221572, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.925662267221572, "decoded_pcm_sha256": "0698365674d63379a8ffa4e1a7bae8ef5f6ed226e14107811f5311d9de011ca9", "source_sha256": "86f57be846b5ad8817f2752d90d5f2aba1e97ec394b2f4ccf28bc5dc39a181ef", "wav_sha256": "0cf7a7fe6f0728e339cc5584e7b2224d39f0e765e107f75e333ec6c19316e34c", "local_source": "audio/librispeech_1580-141084-0016.flac", "local_wav": "audio/librispeech_1580-141084-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 95360}, {"id": "librispeech_1580-141084-0047", "corpus": "librispeech_test_clean", "speaker": "librispeech_1580", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1580-141084-0047", "transcript": "I HAVE A LETTER HERE MISTER SOAMES WHICH I WROTE TO YOU EARLY THIS MORNING IN THE MIDDLE OF A RESTLESS NIGHT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1580/141084/1580-141084-0047.flac"}, "sample_rate": 16000, "source_samples": 84000, "source_duration_seconds": 5.25, "source_peak": 0.592376708984375, "source_rms_dbfs": -27.064546771845936, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.064546771845936, "decoded_pcm_sha256": "ca185084384947ca6386941bb27f596abb096a46a774b4716da42a2e461e4cab", "source_sha256": "0e78a5bcc7c122a563c24dc2d7a65f1be8d851e0d45c8caa4226571676cc5d4c", "wav_sha256": "d8dbb036d18173b12b42c70eff46d203272d1fd2479f615e6e6ef5fb71b45e84", "local_source": "audio/librispeech_1580-141084-0047.flac", "local_wav": "audio/librispeech_1580-141084-0047.wav", "excerpt_start_sample": 0, "excerpt_samples": 84000}, {"id": "librispeech_1995-1836-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_1995", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1995-1836-0000", "transcript": "THE HON CHARLES SMITH MISS SARAH'S BROTHER WAS WALKING SWIFTLY UPTOWN FROM MISTER EASTERLY'S WALL STREET OFFICE AND HIS FACE WAS PALE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1995/1836/1995-1836-0000.flac"}, "sample_rate": 16000, "source_samples": 143280, "source_duration_seconds": 8.955, "source_peak": 0.295013427734375, "source_rms_dbfs": -23.750101874704374, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.46393888832139, "decoded_pcm_sha256": "3feb5dcb82941d2a5e9f89efbf0e32e0138e5eec693c118bb9e8031d23d08925", "source_sha256": "8120c55c7231b13d3d497b04b2fe3d999cfdac785a0134eb0fc9e41aa90ce591", "wav_sha256": "4aafb5b71621c159d7c4925ab8452b0bacf56fb3a1c0a62391bd0c29221d989c", "local_source": "audio/librispeech_1995-1836-0000.flac", "local_wav": "audio/librispeech_1995-1836-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1995-1836-0014", "corpus": "librispeech_test_clean", "speaker": "librispeech_1995", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1995-1836-0014", "transcript": "FORTUNATELY SAID MISTER VANDERPOOL NORTHERNERS AND SOUTHERNERS ARE ARRIVING AT A BETTER MUTUAL UNDERSTANDING ON MOST OF THESE MATTERS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1995/1836/1995-1836-0014.flac"}, "sample_rate": 16000, "source_samples": 144720, "source_duration_seconds": 9.045, "source_peak": 0.529449462890625, "source_rms_dbfs": -22.041190562551304, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.513469437558737, "decoded_pcm_sha256": "af7f0befa255d17d0806307d2e5a0582ff2d024a50adfd0a534360c964de3b82", "source_sha256": "c90576861bc6bc0b6837f0b1c340ecc276ac0339ee0a25442a4784f5fc4a4821", "wav_sha256": "44c2d7a7fad4ab4db4dd9d0c3b94288237e031c318c2b8997522fc519b69d61c", "local_source": "audio/librispeech_1995-1836-0014.flac", "local_wav": "audio/librispeech_1995-1836-0014.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1995-1837-0023", "corpus": "librispeech_test_clean", "speaker": "librispeech_1995", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1995-1837-0023", "transcript": "THE NET AND WEB OF ENDLESS THINGS HAD BEEN CRAWLING AND CREEPING AROUND HER SHE HAD STRUGGLED IN DUMB SPEECHLESS TERROR AGAINST SOME MIGHTY GRASPING THAT STROVE FOR HER LIFE WITH GNARLED AND CREEPING FINGERS BUT NOW AT LAST WEAKLY SHE OPENED HER EYES AND QUESTIONED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1995/1837/1995-1837-0023.flac"}, "sample_rate": 16000, "source_samples": 271440, "source_duration_seconds": 16.965, "source_peak": 0.366790771484375, "source_rms_dbfs": -23.594041351148842, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.511532890985304, "decoded_pcm_sha256": "2fff476dd7493d67e46c49e37f104d20891e347d86560cf73c09f434075020cb", "source_sha256": "7f584164eef2f5976f874ee9c027933c892ba7f920f7233bfd2a219d0ebffcd4", "wav_sha256": "b1024b177f01004b24a07a5fe535254d4fe201bd8f6ca6820677b0b8d9569c92", "local_source": "audio/librispeech_1995-1837-0023.flac", "local_wav": "audio/librispeech_1995-1837-0023.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2094-142345-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_2094", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2094-142345-0032", "transcript": "I OFTEN HEARD HER TALK OF YOU IN THE SAME SORT OF WAY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2094/142345/2094-142345-0032.flac"}, "sample_rate": 16000, "source_samples": 51840, "source_duration_seconds": 3.24, "source_peak": 0.49334716796875, "source_rms_dbfs": -21.689633787280776, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.689633787280776, "decoded_pcm_sha256": "973bdb09bc9497c63b4dfbb3bc77778c87a6ced88d08e7f2f6e22739dd6d5444", "source_sha256": "b92bf752cc5264a2a47e0153c236b90edb48dd32b819cfd12d0e5250ca3c7a1e", "wav_sha256": "320b33604dfc6f64e4624c4baa3515e073f94afc1760147678affda9973624a2", "local_source": "audio/librispeech_2094-142345-0032.flac", "local_wav": "audio/librispeech_2094-142345-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 51840}, {"id": "librispeech_2094-142345-0039", "corpus": "librispeech_test_clean", "speaker": "librispeech_2094", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2094-142345-0039", "transcript": "I'VE STRONG ASSURANCE THAT NO EVIL WILL HAPPEN TO YOU AND MY UNCLE AND THE CHILDREN FROM ANYTHING I'VE DONE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2094/142345/2094-142345-0039.flac"}, "sample_rate": 16000, "source_samples": 100480, "source_duration_seconds": 6.28, "source_peak": 0.509735107421875, "source_rms_dbfs": -21.82028978216153, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.62361673126311, "decoded_pcm_sha256": "c411d502f97ab6fc22d81b52c4ba95e5e562514c255c2379c8837148877e0ea1", "source_sha256": "21d7d00350a4ca2b0109da67cfad724f36979e7c1b2e098f9327bc8c4f433435", "wav_sha256": "c7a482fbc5c97605308c5c4ecd86d44207027acba93121762e33b0b573212e24", "local_source": "audio/librispeech_2094-142345-0039.flac", "local_wav": "audio/librispeech_2094-142345-0039.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2094-142345-0040", "corpus": "librispeech_test_clean", "speaker": "librispeech_2094", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2094-142345-0040", "transcript": "I DIDN'T PREACH WITHOUT DIRECTION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2094/142345/2094-142345-0040.flac"}, "sample_rate": 16000, "source_samples": 38160, "source_duration_seconds": 2.385, "source_peak": 0.6126708984375, "source_rms_dbfs": -22.307817294696616, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.307817294696616, "decoded_pcm_sha256": "9cbc49d67edc85ca6ba00d13dae38ec545b7aaf549562d1af9d9e72881814d19", "source_sha256": "8b3f5a025e4853fabdb000f2820a651f4a66a5ef234d912581ae29fafea72e48", "wav_sha256": "6bd7f0881080a71cb6c4ab59396378576c0114a4c9e05c3c1de5edbdeeeb0664", "local_source": "audio/librispeech_2094-142345-0040.flac", "local_wav": "audio/librispeech_2094-142345-0040.wav", "excerpt_start_sample": 0, "excerpt_samples": 38160}, {"id": "librispeech_2300-131720-0022", "corpus": "librispeech_test_clean", "speaker": "librispeech_2300", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2300-131720-0022", "transcript": "MEANWHILE HE HAD CALLED UPON ME TO MAKE A REPORT OF THE THREE WIRE SYSTEM KNOWN IN ENGLAND AS THE HOPKINSON BOTH DOCTOR JOHN HOPKINSON AND MISTER EDISON BEING INDEPENDENT INVENTORS AT PRACTICALLY THE SAME TIME", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2300/131720/2300-131720-0022.flac"}, "sample_rate": 16000, "source_samples": 218640, "source_duration_seconds": 13.665, "source_peak": 0.64959716796875, "source_rms_dbfs": -22.580480288302475, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.225644794441955, "decoded_pcm_sha256": "b7f1420236bac4a5c404bc851efe87bc6142a76f2a2935e40197b946d8182998", "source_sha256": "93212060c39673d7a3ee80b52a8f2612e70e3c121fd588c690e3d51be10e6963", "wav_sha256": "47955be189936046241d85bf424da25472f89de1b610988282b8cb71c08fcad7", "local_source": "audio/librispeech_2300-131720-0022.flac", "local_wav": "audio/librispeech_2300-131720-0022.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2300-131720-0028", "corpus": "librispeech_test_clean", "speaker": "librispeech_2300", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2300-131720-0028", "transcript": "THERE WAS INFINITE SCEPTICISM AROUND HIM ON THE SUBJECT AND WHILE OTHER INVENTORS WERE ALSO GIVING THE SUBJECT THEIR THOUGHT THE PUBLIC TOOK IT FOR GRANTED THAT ANYTHING SO UTTERLY INTANGIBLE AS ELECTRICITY THAT COULD NOT BE SEEN OR WEIGHED AND ONLY GAVE SECONDARY EVIDENCE OF ITSELF AT THE EXACT POINT OF USE COULD NOT BE BROUGHT TO ACCURATE REGISTRATION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2300/131720/2300-131720-0028.flac"}, "sample_rate": 16000, "source_samples": 415040, "source_duration_seconds": 25.94, "source_peak": 0.58966064453125, "source_rms_dbfs": -23.395115497521118, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.870200927979273, "decoded_pcm_sha256": "9eaf55b481a213cff53140518cebb83838d1187d3e2187282fcf13a3699505c4", "source_sha256": "72f44e8aee815c88d0ad1b8f6d9a15fea55954dbd8e4d996b97112fe34bda0a1", "wav_sha256": "1aa80ada2f91d3419088daa04ecfc0ddc106837f4ce52f87063757534aef4e1d", "local_source": "audio/librispeech_2300-131720-0028.flac", "local_wav": "audio/librispeech_2300-131720-0028.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2300-131720-0041", "corpus": "librispeech_test_clean", "speaker": "librispeech_2300", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2300-131720-0041", "transcript": "WE HAD METERS IN WHICH THERE WERE TWO BOTTLES OF LIQUID", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2300/131720/2300-131720-0041.flac"}, "sample_rate": 16000, "source_samples": 60000, "source_duration_seconds": 3.75, "source_peak": 0.49029541015625, "source_rms_dbfs": -23.37033603041045, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.37033603041045, "decoded_pcm_sha256": "976092b4e8947717eb460c0f8df2d08babdc9b088a74a4efec5fca50e34afea2", "source_sha256": "4ede50e44cbe5dfb258311f80fd68b7efe2e080265efa5e6bc07aea531a22e28", "wav_sha256": "ceb4b44d00ec8e36a1e67a7f31a0bc8f7440b700ab4fd9d88011defa7894bcd0", "local_source": "audio/librispeech_2300-131720-0041.flac", "local_wav": "audio/librispeech_2300-131720-0041.wav", "excerpt_start_sample": 0, "excerpt_samples": 60000}, {"id": "librispeech_237-126133-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_237", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "237-126133-0003", "transcript": "SOMEHOW OF ALL THE DAYS WHEN THE HOME FEELING WAS THE STRONGEST THIS DAY IT SEEMED AS IF SHE COULD BEAR IT NO LONGER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/237/126133/237-126133-0003.flac"}, "sample_rate": 16000, "source_samples": 104960, "source_duration_seconds": 6.56, "source_peak": 0.233489990234375, "source_rms_dbfs": -30.677490302294746, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.31670425464334, "decoded_pcm_sha256": "15627325ed4198a6a91f3a9dd72c16085a4986165c018e492c56a9076ad969f4", "source_sha256": "18c38673f71d4012a9c023d7548bff3c0f78172415f67c69abf403590c4e84bc", "wav_sha256": "0496d3168b5687d428050d9c03d395d3225f8e034222090d10a32baed9384b2f", "local_source": "audio/librispeech_237-126133-0003.flac", "local_wav": "audio/librispeech_237-126133-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_237-134500-0029", "corpus": "librispeech_test_clean", "speaker": "librispeech_237", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "237-134500-0029", "transcript": "I DON'T WANT TO STAND AROUND AND LOOK ON", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/237/134500/237-134500-0029.flac"}, "sample_rate": 16000, "source_samples": 33040, "source_duration_seconds": 2.065, "source_peak": 0.935546875, "source_rms_dbfs": -21.343679960921527, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.343679960921527, "decoded_pcm_sha256": "8562d47701848db95c8610733017a69a5928db4edb524e0f205ab4182cf7b4f7", "source_sha256": "a71914689325868d93f40677ab17bdfbb8c8209801864e31a4f4303cf7fd0fa0", "wav_sha256": "d1f7208286ea0672b525054ddc0371bc4a14a529a808bb5cd1163899ca3e9af3", "local_source": "audio/librispeech_237-134500-0029.flac", "local_wav": "audio/librispeech_237-134500-0029.wav", "excerpt_start_sample": 0, "excerpt_samples": 33040}, {"id": "librispeech_237-134500-0031", "corpus": "librispeech_test_clean", "speaker": "librispeech_237", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "237-134500-0031", "transcript": "SOMETIMES I DON'T WANT TO DO ANYTHING AT ALL AND SOMETIMES I WANT TO PULL THE FOUR CORNERS OF THE DIVIDE TOGETHER HE THREW OUT HIS ARM AND BROUGHT IT BACK WITH A JERK SO LIKE A TABLE CLOTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/237/134500/237-134500-0031.flac"}, "sample_rate": 16000, "source_samples": 169760, "source_duration_seconds": 10.61, "source_peak": 0.700347900390625, "source_rms_dbfs": -21.801426992576367, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.291181411176545, "decoded_pcm_sha256": "66fcbe0885d7123cd5c5c14ff8b544f94ed65d51857cb70d13de972dd00f9421", "source_sha256": "7456acb116b3a8c6d43c44ebf79a8f7b59f6ed546ffb63ce78a6dc58bb6d5abe", "wav_sha256": "7fcf9d87377483efb1617dbb6ddf6de5d613cf60dc9c5826d5c3b2d1f5efeb32", "local_source": "audio/librispeech_237-134500-0031.flac", "local_wav": "audio/librispeech_237-134500-0031.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_260-123286-0008", "corpus": "librispeech_test_clean", "speaker": "librispeech_260", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "260-123286-0008", "transcript": "THEREFORE DON'T TALK TO ME ABOUT VIEWS AND PROSPECTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/260/123286/260-123286-0008.flac"}, "sample_rate": 16000, "source_samples": 59760, "source_duration_seconds": 3.735, "source_peak": 0.46673583984375, "source_rms_dbfs": -23.64591439734617, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.64591439734617, "decoded_pcm_sha256": "931193da803cf63a31b439032de16719f19aca2c3130d571b4b8044d43a1413c", "source_sha256": "735769549e4dbbfb3f4ba43f89866cde8085933d8b16e67550cc65264bf84fc3", "wav_sha256": "a76c87fb51f5ea1aaa5040037a8d8364b9103e68aace0946bee820b613b70f25", "local_source": "audio/librispeech_260-123286-0008.flac", "local_wav": "audio/librispeech_260-123286-0008.wav", "excerpt_start_sample": 0, "excerpt_samples": 59760}, {"id": "librispeech_260-123288-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_260", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "260-123288-0003", "transcript": "THE ELECTRIC LIGHT CAN SCARCELY PENETRATE THROUGH THE DENSE CURTAIN WHICH HAS DROPPED OVER THE THEATRE ON WHICH THE BATTLE OF THE ELEMENTS IS ABOUT TO BE WAGED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/260/123288/260-123288-0003.flac"}, "sample_rate": 16000, "source_samples": 142480, "source_duration_seconds": 8.905, "source_peak": 0.39129638671875, "source_rms_dbfs": -25.639248238099196, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.560371051032, "decoded_pcm_sha256": "d480f00d02d595a6352c610e02536f099ff03473f6693b5628683647e9479766", "source_sha256": "5e8ca953fe1b2dd885f0cfbf4d706040e55637eee883bea8f0dc6e08afac3240", "wav_sha256": "b92004dac924d397e6ba0829351dde9494b14ae7a6782ccdbd4d8cbd2680f841", "local_source": "audio/librispeech_260-123288-0003.flac", "local_wav": "audio/librispeech_260-123288-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_260-123440-0020", "corpus": "librispeech_test_clean", "speaker": "librispeech_260", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "260-123440-0020", "transcript": "WE WON'T TALK ABOUT HER ANY MORE IF YOU'D RATHER NOT WE INDEED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/260/123440/260-123440-0020.flac"}, "sample_rate": 16000, "source_samples": 79920, "source_duration_seconds": 4.995, "source_peak": 0.46844482421875, "source_rms_dbfs": -22.68064988825362, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.68064988825362, "decoded_pcm_sha256": "c10a9cdd8fd4960bfb88b1611c0123695ebcd137816c47ff78925a8699616669", "source_sha256": "44be2bb2c15059058798a25f87de6310601074e1bc4b68bda0ad3ad1860a9040", "wav_sha256": "dc0e2b8720fca680ed9e8a46c6c222626f95be9b07311fc744a3761bb1f0981f", "local_source": "audio/librispeech_260-123440-0020.flac", "local_wav": "audio/librispeech_260-123440-0020.wav", "excerpt_start_sample": 0, "excerpt_samples": 79920}, {"id": "librispeech_2830-3980-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_2830", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2830-3980-0000", "transcript": "IN EVERY WAY THEY SOUGHT TO UNDERMINE THE AUTHORITY OF SAINT PAUL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2830/3980/2830-3980-0000.flac"}, "sample_rate": 16000, "source_samples": 59680, "source_duration_seconds": 3.73, "source_peak": 0.89697265625, "source_rms_dbfs": -16.80410950587789, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -16.80410950587789, "decoded_pcm_sha256": "8f71058a78243905dd373f5302bc82f25ff2a7e7631413b5ce924104f0f14bea", "source_sha256": "a73adb84b9560fdfc8f6c7f5801e567267ac8beb6fd51df4f5dec7217eb43ad8", "wav_sha256": "77aacbae6ecdcf8ad5434acd4a1388d34de3c30a9ad3cd826055b781b1bc977f", "local_source": "audio/librispeech_2830-3980-0000.flac", "local_wav": "audio/librispeech_2830-3980-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 59680}, {"id": "librispeech_2830-3980-0038", "corpus": "librispeech_test_clean", "speaker": "librispeech_2830", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2830-3980-0038", "transcript": "GRACE BE TO YOU AND PEACE FROM GOD THE FATHER AND FROM OUR LORD JESUS CHRIST", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2830/3980/2830-3980-0038.flac"}, "sample_rate": 16000, "source_samples": 88640, "source_duration_seconds": 5.54, "source_peak": 0.566497802734375, "source_rms_dbfs": -24.04910591648981, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.04910591648981, "decoded_pcm_sha256": "77d64124560df236e40a3ec37cc728b327f805fee264fb8362c33791320222e1", "source_sha256": "2c8978e75dd10716d066062bb2fbfd1a002ac3190e42b0f251f15d0aa2d6d272", "wav_sha256": "eb598d45742c7992ecb5c2ef39512cf5ba5471f617decb4918e42ed52e064291", "local_source": "audio/librispeech_2830-3980-0038.flac", "local_wav": "audio/librispeech_2830-3980-0038.wav", "excerpt_start_sample": 0, "excerpt_samples": 88640}, {"id": "librispeech_2830-3980-0042", "corpus": "librispeech_test_clean", "speaker": "librispeech_2830", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2830-3980-0042", "transcript": "THE WORLD BRANDS THIS A PERNICIOUS DOCTRINE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2830/3980/2830-3980-0042.flac"}, "sample_rate": 16000, "source_samples": 48320, "source_duration_seconds": 3.02, "source_peak": 0.554443359375, "source_rms_dbfs": -22.785804418379104, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.785804418379104, "decoded_pcm_sha256": "3b4a08b0741e0ca396487185085d64365ab56b16d5868e61a4550a5dc24466f2", "source_sha256": "ff30195a24b3f6c24a6ad4748f2ab49f4a036975c0b62f5b1c2f4203488ee9ae", "wav_sha256": "9226d5602341de020d92f33ac9700e8ea29f2fd9ff33db456d1a37078a94b50a", "local_source": "audio/librispeech_2830-3980-0042.flac", "local_wav": "audio/librispeech_2830-3980-0042.wav", "excerpt_start_sample": 0, "excerpt_samples": 48320}, {"id": "librispeech_2961-960-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_2961", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2961-960-0010", "transcript": "BUT HE HAS NOT AS YET DEFINED THIS INTERMEDIATE TERRITORY WHICH LIES SOMEWHERE BETWEEN MEDICINE AND MATHEMATICS AND HE WOULD HAVE FELT THAT THERE WAS AS GREAT AN IMPIETY IN RANKING THEORIES OF PHYSICS FIRST IN THE ORDER OF KNOWLEDGE AS IN PLACING THE BODY BEFORE THE SOUL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2961/960/2961-960-0010.flac"}, "sample_rate": 16000, "source_samples": 327360, "source_duration_seconds": 20.46, "source_peak": 0.380584716796875, "source_rms_dbfs": -28.229074807392955, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.157181381609547, "decoded_pcm_sha256": "ab9593d38d133538cf3475c5a7bf30b265737c427a5c4a501ecf3384b1e2aee8", "source_sha256": "d93e37326bfa987a0cf66c0edd79dfa034ea013e5967c8f2ba9edf54cde8d2be", "wav_sha256": "52ca5b9bf9daf5a48184c8d90a1033ba402bb03cb88b7ea9362dd6fa4b24abd9", "local_source": "audio/librispeech_2961-960-0010.flac", "local_wav": "audio/librispeech_2961-960-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2961-960-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_2961", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2961-960-0019", "transcript": "HE COULD WRITE IN ONE STYLE BUT NOT IN ANOTHER AND THE GREEK LANGUAGE HAD NOT AS YET BEEN FASHIONED BY ANY POET OR PHILOSOPHER TO DESCRIBE PHYSICAL PHENOMENA", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2961/960/2961-960-0019.flac"}, "sample_rate": 16000, "source_samples": 194080, "source_duration_seconds": 12.13, "source_peak": 0.625213623046875, "source_rms_dbfs": -25.64813842860343, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.64907732157032, "decoded_pcm_sha256": "966508e2ccc362392d8ec80fc3d05a564460cb6f0dd69f613c7b86089cb21caa", "source_sha256": "f5052549b9cbc3b8433ae187d0e876b214f5250ddb4fbb637dea220e8a4726cf", "wav_sha256": "834322db0896bb0348dd0ab0f7fc4d268c5c91a6c91e78f489697f63ba92e7f7", "local_source": "audio/librispeech_2961-960-0019.flac", "local_wav": "audio/librispeech_2961-960-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2961-961-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_2961", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2961-961-0011", "transcript": "THE GENEALOGIES WHICH YOU HAVE RECITED TO US OUT OF YOUR OWN ANNALS SOLON ARE A MERE CHILDREN'S STORY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2961/961/2961-961-0011.flac"}, "sample_rate": 16000, "source_samples": 125040, "source_duration_seconds": 7.815, "source_peak": 0.61773681640625, "source_rms_dbfs": -27.948512195945238, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.258152323716782, "decoded_pcm_sha256": "00bd4132c58a261e6169cd035b93bf1134d0bd1b082f44ceb99cf14faa9b7062", "source_sha256": "f859855c45fc07e8d561ed7a846fd14ca4ae1515749b60c4e8d4ce825df30180", "wav_sha256": "ddf0f4bb5b9f2e544585e0723cf46b3096a44e71b8c13ef37d1d0a6e688f07e4", "local_source": "audio/librispeech_2961-961-0011.flac", "local_wav": "audio/librispeech_2961-961-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3570-5694-0001", "corpus": "librispeech_test_clean", "speaker": "librispeech_3570", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3570-5694-0001", "transcript": "THE UTILITY OF CONSUMPTION AS AN EVIDENCE OF WEALTH IS TO BE CLASSED AS A DERIVATIVE GROWTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3570/5694/3570-5694-0001.flac"}, "sample_rate": 16000, "source_samples": 90800, "source_duration_seconds": 5.675, "source_peak": 0.429901123046875, "source_rms_dbfs": -22.81140123891472, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.81140123891472, "decoded_pcm_sha256": "d7ebe73ddd2095245c2316a87024f8d15117a6bca41809d6c43a3d2cfe68c3a9", "source_sha256": "83d42febda01d467c35c0b90dc7bb5d5d80b18ed95d65748a831afca53f2e3cd", "wav_sha256": "5c52d70d17699b53a204ff0cdbe488ace62913f5304818fe714ac7198b7b65a2", "local_source": "audio/librispeech_3570-5694-0001.flac", "local_wav": "audio/librispeech_3570-5694-0001.wav", "excerpt_start_sample": 0, "excerpt_samples": 90800}, {"id": "librispeech_3570-5694-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_3570", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3570-5694-0004", "transcript": "IN THE NATURE OF THINGS LUXURIES AND THE COMFORTS OF LIFE BELONG TO THE LEISURE CLASS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3570/5694/3570-5694-0004.flac"}, "sample_rate": 16000, "source_samples": 85280, "source_duration_seconds": 5.33, "source_peak": 0.596405029296875, "source_rms_dbfs": -23.986191830522884, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.986191830522884, "decoded_pcm_sha256": "86a8946d8572399656832193f585d37dfcbd327696eb662dcf0e41d4b67d0bd1", "source_sha256": "e7c4c68060bd5da1df857aa90f0fb6601cfee83e0001d75b64164988b070e7d8", "wav_sha256": "75ab8457c235b0d72c8db7fd3f4b5e5feb1e2b50ce0c9835a6af02d2ab282d69", "local_source": "audio/librispeech_3570-5694-0004.flac", "local_wav": "audio/librispeech_3570-5694-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 85280}, {"id": "librispeech_3570-5696-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_3570", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3570-5696-0010", "transcript": "AN ARTICLE MAY BE USEFUL AND WASTEFUL BOTH AND ITS UTILITY TO THE CONSUMER MAY BE MADE UP OF USE AND WASTE IN THE MOST VARYING PROPORTIONS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3570/5696/3570-5696-0010.flac"}, "sample_rate": 16000, "source_samples": 121120, "source_duration_seconds": 7.57, "source_peak": 0.74517822265625, "source_rms_dbfs": -22.057355240612, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.485928141030136, "decoded_pcm_sha256": "0efec23fe6eeee5dfee017e6be4b06b64a152a981c62a71c533addc819701d30", "source_sha256": "426a986fec027f14754878775f6b605d58d61a0fd0d77bf122657dc1cef2ce9d", "wav_sha256": "9bdc7d45ba3123b85371d1a0bfaec30557f59f961d83177b590797de0d75f4c6", "local_source": "audio/librispeech_3570-5696-0010.flac", "local_wav": "audio/librispeech_3570-5696-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3575-170457-0034", "corpus": "librispeech_test_clean", "speaker": "librispeech_3575", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3575-170457-0034", "transcript": "IN THIS MONOTONOUS LIFE OF MINE THAT WAS A PLEASANT EVENT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3575/170457/3575-170457-0034.flac"}, "sample_rate": 16000, "source_samples": 55920, "source_duration_seconds": 3.495, "source_peak": 0.84869384765625, "source_rms_dbfs": -19.712774707346266, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -19.712774707346266, "decoded_pcm_sha256": "b1c60dfc008f07f56156adc48b00db15f16da435fb212895c0516ecbd9e5de30", "source_sha256": "7379f6afffbfc11aded5aae60cf94ff2f5aa5a2dc317f61b2cd70c9c4d4847df", "wav_sha256": "53cc005f386dbcd9738a8dd623cab86151cf87bd4f53bdfa78d7cea937f0764f", "local_source": "audio/librispeech_3575-170457-0034.flac", "local_wav": "audio/librispeech_3575-170457-0034.wav", "excerpt_start_sample": 0, "excerpt_samples": 55920}, {"id": "librispeech_3575-170457-0041", "corpus": "librispeech_test_clean", "speaker": "librispeech_3575", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3575-170457-0041", "transcript": "SHE WAS GONE OUT INTO THE VILLAGE ON SOME ERRAND WHEN AS SHE WAS DESCENDING THE STEEP STREET HER FOOT SLIPPED ON THE ICE AND SHE FELL IT WAS DARK AND NO ONE SAW HER MISCHANCE TILL AFTER A TIME HER GROANS ATTRACTED THE ATTENTION OF A PASSER BY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3575/170457/3575-170457-0041.flac"}, "sample_rate": 16000, "source_samples": 224160, "source_duration_seconds": 14.01, "source_peak": 0.66522216796875, "source_rms_dbfs": -21.83040488145807, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.483529355547855, "decoded_pcm_sha256": "ce9155c8469c627681fdbab9602fc52b17f46ce8849d53a2be4d450d58cfb2c4", "source_sha256": "90d1dd17384a0960d3acaf50b3f7f40b4699288aa74a04c25a221270e104ce55", "wav_sha256": "f49c7afeb6e5b22010dcb4ea4529e175cd6527a2dc81e2ef1d5509308999fa73", "local_source": "audio/librispeech_3575-170457-0041.flac", "local_wav": "audio/librispeech_3575-170457-0041.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3575-170457-0054", "corpus": "librispeech_test_clean", "speaker": "librispeech_3575", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3575-170457-0054", "transcript": "STUNG BY ANXIETY FOR THIS LITTLE SISTER SHE UPBRAIDED MISS W FOR HER FANCIED INDIFFERENCE TO ANNE'S STATE OF HEALTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3575/170457/3575-170457-0054.flac"}, "sample_rate": 16000, "source_samples": 128080, "source_duration_seconds": 8.005, "source_peak": 0.630035400390625, "source_rms_dbfs": -23.47235016030915, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.430811733763257, "decoded_pcm_sha256": "5ff0a2aa3e7e6a57cda2e60dda3bc719388bba59551820dff2445caca9cf7c6a", "source_sha256": "638053737df40babfd92ea956f1083964cfe38bf547b96334228a8f67b2cb78b", "wav_sha256": "cf42c27ee1f2f8a74c2815e7f8f0ba01f79488df60e88267cb947da17f444a20", "local_source": "audio/librispeech_3575-170457-0054.flac", "local_wav": "audio/librispeech_3575-170457-0054.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3729-6852-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_3729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3729-6852-0000", "transcript": "TO CELEBRATE THE ARRIVAL OF HER SON SILVIA GAVE A SPLENDID SUPPER TO WHICH SHE HAD INVITED ALL HER RELATIVES AND IT WAS A GOOD OPPORTUNITY FOR ME TO MAKE THEIR ACQUAINTANCE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3729/6852/3729-6852-0000.flac"}, "sample_rate": 16000, "source_samples": 178720, "source_duration_seconds": 11.17, "source_peak": 0.402435302734375, "source_rms_dbfs": -26.96578233204508, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.348026827773285, "decoded_pcm_sha256": "e1524c8eb759928cd93fd86e98694c9a5efccdfa2a0fd948935f558dbffb8843", "source_sha256": "3fa96f8e0276d79ec027bb1595215a8bbeb488cefcc1702ff798e91ec8504403", "wav_sha256": "f7b24fd424e1ff2458c9d5639915bae0d1ca941e15f073571afd8ddaf9139fe0", "local_source": "audio/librispeech_3729-6852-0000.flac", "local_wav": "audio/librispeech_3729-6852-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3729-6852-0007", "corpus": "librispeech_test_clean", "speaker": "librispeech_3729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3729-6852-0007", "transcript": "TWO YEARS BEFORE HER DEATH I SAW HER PERFORM THE CHARACTER OF MARIANNE IN THE COMEDY OF MARIVAUX AND IN SPITE OF HER AGE AND DECLINING HEALTH THE ILLUSION WAS COMPLETE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3729/6852/3729-6852-0007.flac"}, "sample_rate": 16000, "source_samples": 199840, "source_duration_seconds": 12.49, "source_peak": 0.485931396484375, "source_rms_dbfs": -28.592297306892682, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.01965963933016, "decoded_pcm_sha256": "d081e3d6cdc89188a9f3d229a815f9784efd7e72dfcfa5317b4ac90fa9d86ff1", "source_sha256": "9d9bbbeeff1bfb647e2698db36f186cb439a8b2894772ca3cdfe9425c3e55a1a", "wav_sha256": "96a00b5f0bef082a0cd43dbeb530fadd34a0cabe94dc7b7c87521a64a2ded2bc", "local_source": "audio/librispeech_3729-6852-0007.flac", "local_wav": "audio/librispeech_3729-6852-0007.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3729-6852-0017", "corpus": "librispeech_test_clean", "speaker": "librispeech_3729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3729-6852-0017", "transcript": "I SEE A QUANTITY OF CHAIRS FOR HIRE AT THE RATE OF ONE SOU MEN READING THE NEWSPAPER UNDER THE SHADE OF THE TREES GIRLS AND MEN BREAKFASTING EITHER ALONE OR IN COMPANY WAITERS WHO WERE RAPIDLY GOING UP AND DOWN A NARROW STAIRCASE HIDDEN UNDER THE FOLIAGE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3729/6852/3729-6852-0017.flac"}, "sample_rate": 16000, "source_samples": 287840, "source_duration_seconds": 17.99, "source_peak": 0.384185791015625, "source_rms_dbfs": -28.259280515739242, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.46214544153569, "decoded_pcm_sha256": "ba156f183286a1f4044b0b7794b3982603585d0e122bdb35978c337419392c08", "source_sha256": "12a9659e29190ce17dd89b5209d4ccaebc746bfb35093185d5970a91878e42b9", "wav_sha256": "73c66dd96fb1be6aa0bf20f4b8915715ca5a0ae34cdfa961c9ca3e910cbcba0f", "local_source": "audio/librispeech_3729-6852-0017.flac", "local_wav": "audio/librispeech_3729-6852-0017.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4077-13751-0001", "corpus": "librispeech_test_clean", "speaker": "librispeech_4077", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "4077-13751-0001", "transcript": "ITS ORIGIN WAS SMALL A GERM AN INSIGNIFICANT SEED HARDLY TO BE THOUGHT OF AS LIKELY TO AROUSE OPPOSITION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4077/13751/4077-13751-0001.flac"}, "sample_rate": 16000, "source_samples": 139920, "source_duration_seconds": 8.745, "source_peak": 0.941619873046875, "source_rms_dbfs": -23.904775017352993, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.869637117226276, "decoded_pcm_sha256": "ff5a7ca9328a329a2ca9ba953b0550653b86fe7bbf7bda2cae73908a1fbfde2a", "source_sha256": "1f85c5ad8dfb8fcf3c8d86e54afc055f756f147e7b426ff8b7f48780d8d43c0e", "wav_sha256": "67df7354e652c45bc30afaf081016690e05f0fc4c9c513350071df1b7f79cd44", "local_source": "audio/librispeech_4077-13751-0001.flac", "local_wav": "audio/librispeech_4077-13751-0001.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4077-13754-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_4077", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "4077-13754-0002", "transcript": "IT WAS THROUGH FLOYD'S ADVICE THAT BUCHANAN ORDERED THE MILITARY EXPEDITION TO UTAH OSTENSIBLY TO INSTALL CERTAIN FEDERAL OFFICIALS AND TO REPRESS AN ALLEGED INFANTILE REBELLION WHICH IN FACT HAD NEVER COME INTO EXISTENCE BUT IN REALITY TO FURTHER THE INTERESTS OF THE SECESSIONISTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4077/13754/4077-13754-0002.flac"}, "sample_rate": 16000, "source_samples": 284480, "source_duration_seconds": 17.78, "source_peak": 0.79345703125, "source_rms_dbfs": -25.576833348040672, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.41752474008618, "decoded_pcm_sha256": "e4ef84f14b8958b7a78512946198f223e31d6190beadfe364d4772e4a9963ced", "source_sha256": "58ff360d46e4fc6bcaf04da1a6ead97dc191c54426528cfa13efe7e780349ff5", "wav_sha256": "a850b52a73c2375cffe72a0ad6d0aacc3e356a76d7e143a5ae32b5d2c6aacec0", "local_source": "audio/librispeech_4077-13754-0002.flac", "local_wav": "audio/librispeech_4077-13754-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4077-13754-0012", "corpus": "librispeech_test_clean", "speaker": "librispeech_4077", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "4077-13754-0012", "transcript": "THIS MEANT THAT FOR AN ALLEGED MISDEMEANOR FOR WHICH CONGRESS PRESCRIBED A MAXIMUM PENALTY OF SIX MONTHS IMPRISONMENT AND A FINE OF THREE HUNDRED DOLLARS A MAN MIGHT BE IMPRISONED FOR LIFE AYE FOR MANY TERMS OF A MAN'S NATURAL LIFE DID THE COURT'S POWER TO ENFORCE ITS SENTENCES EXTEND SO FAR AND MIGHT BE FINED MILLIONS OF DOLLARS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4077/13754/4077-13754-0012.flac"}, "sample_rate": 16000, "source_samples": 353520, "source_duration_seconds": 22.095, "source_peak": 0.543731689453125, "source_rms_dbfs": -27.828520650204958, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.244250003732674, "decoded_pcm_sha256": "43e800e9f69ae92ed54ca36d54a78bd7c56d8ff0657f75ee376d88adab6b3764", "source_sha256": "65e48d9b9257be487e80c6938e34856d742870d390ad5d3d1b04d5f95f955d4a", "wav_sha256": "76f48fb7f3a6036c409df2023e580ecd2152586e46206da6014fee596e331fec", "local_source": "audio/librispeech_4077-13754-0012.flac", "local_wav": "audio/librispeech_4077-13754-0012.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4446-2271-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_4446", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4446-2271-0003", "transcript": "IT'S BEEN ON ONLY TWO WEEKS AND I'VE BEEN HALF A DOZEN TIMES ALREADY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4446/2271/4446-2271-0003.flac"}, "sample_rate": 16000, "source_samples": 59200, "source_duration_seconds": 3.7, "source_peak": 0.517730712890625, "source_rms_dbfs": -22.567726927886582, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.567726927886582, "decoded_pcm_sha256": "257baf7b0200414f4ef5d6ab5db2362889681a1c0de114633c54f50f9b628549", "source_sha256": "b8fecefeca8a7ae57ad53900dbba581245a09ad79d46ae1fc06140e0e5a5b281", "wav_sha256": "3d19719ae977a823291a0e50648779b441f15059218e7de8cff1ee0b8844309c", "local_source": "audio/librispeech_4446-2271-0003.flac", "local_wav": "audio/librispeech_4446-2271-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 59200}, {"id": "librispeech_4446-2271-0022", "corpus": "librispeech_test_clean", "speaker": "librispeech_4446", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4446-2271-0022", "transcript": "I'M GLAD SHE'S HELD HER OWN SINCE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4446/2271/4446-2271-0022.flac"}, "sample_rate": 16000, "source_samples": 37200, "source_duration_seconds": 2.325, "source_peak": 0.45782470703125, "source_rms_dbfs": -23.848490648398872, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.848490648398872, "decoded_pcm_sha256": "36c74fdf4f12eb94aa845b16c467769326a4db667ab611a8845b7f72c2cae5b1", "source_sha256": "ed740f49d11cc5c58f71f1925854cc9c1f30170b81143873aa8c58cd35430583", "wav_sha256": "0c393b4c2613903f50dbc975ca65d846b232b63f03096066c97bab44c6558f68", "local_source": "audio/librispeech_4446-2271-0022.flac", "local_wav": "audio/librispeech_4446-2271-0022.wav", "excerpt_start_sample": 0, "excerpt_samples": 37200}, {"id": "librispeech_4446-2273-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_4446", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4446-2273-0010", "transcript": "THERE WAS WATERCRESS SOUP AND SOLE AND A DELIGHTFUL OMELETTE STUFFED WITH MUSHROOMS AND TRUFFLES AND TWO SMALL RARE DUCKLINGS AND ARTICHOKES AND A DRY YELLOW RHONE WINE OF WHICH BARTLEY HAD ALWAYS BEEN VERY FOND", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4446/2273/4446-2273-0010.flac"}, "sample_rate": 16000, "source_samples": 199040, "source_duration_seconds": 12.44, "source_peak": 0.4254150390625, "source_rms_dbfs": -27.512108974962025, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.66417947925898, "decoded_pcm_sha256": "263c284aa66f0a837f0cdfcf012d15e0d4edfeb7a8a0717217e17395942a83fd", "source_sha256": "ee6acd6b5c33e2d7e114405b48aec2700f720d68dd8d59676172e81ca63317dc", "wav_sha256": "f67f7cae4b9d9ff9f64d921650ca4ba0ed2375ba7bc5d28c5460713d0e6548df", "local_source": "audio/librispeech_4446-2273-0010.flac", "local_wav": "audio/librispeech_4446-2273-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4507-16021-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_4507", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4507-16021-0016", "transcript": "CAN ONE IMAGINE A NATURALIST REFUSING TO STUDY THE VIPER THE BAT THE SCORPION THE CENTIPEDE THE TARANTULA AND ONE WHO WOULD CAST THEM BACK INTO THEIR DARKNESS SAYING OH HOW UGLY THAT IS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4507/16021/4507-16021-0016.flac"}, "sample_rate": 16000, "source_samples": 276240, "source_duration_seconds": 17.265, "source_peak": 0.553863525390625, "source_rms_dbfs": -26.07557497706036, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.407502540486163, "decoded_pcm_sha256": "0f759c362b7bb3f7ab6547020834b6f41d471a7cfe5c2018d0a8cc506e042103", "source_sha256": "e3cf855f3ed817d96563b7914780df19ad9175cd7deb4a04880dfb6093dcccf7", "wav_sha256": "97b09a28609b066912fc01e5a4c462962bf4af014153f52ff0be247c7b7efa39", "local_source": "audio/librispeech_4507-16021-0016.flac", "local_wav": "audio/librispeech_4507-16021-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4507-16021-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_4507", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4507-16021-0032", "transcript": "HE MUST DESCEND WITH HIS HEART FULL OF CHARITY AND SEVERITY AT THE SAME TIME AS A BROTHER AND AS A JUDGE TO THOSE IMPENETRABLE CASEMATES WHERE CRAWL PELL MELL THOSE WHO BLEED AND THOSE WHO DEAL THE BLOW THOSE WHO WEEP AND THOSE WHO CURSE THOSE WHO FAST AND THOSE WHO DEVOUR THOSE WHO ENDURE EVIL AND THOSE WHO INFLICT IT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4507/16021/4507-16021-0032.flac"}, "sample_rate": 16000, "source_samples": 449440, "source_duration_seconds": 28.09, "source_peak": 0.4244384765625, "source_rms_dbfs": -25.686727205489255, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.511214264972427, "decoded_pcm_sha256": "7576336f86f07936b03cf91b64bf80c306a8a013d42a9e0eb9b1dae02be902c2", "source_sha256": "cf872613e0d590501ff569c17e5f9d525920c4babeda14ac94577f60cfc54401", "wav_sha256": "56e8dbd76689127c0d2f1fd91f5b1b06d839e01bac6a98198046d6714ab13c63", "local_source": "audio/librispeech_4507-16021-0032.flac", "local_wav": "audio/librispeech_4507-16021-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4507-16021-0045", "corpus": "librispeech_test_clean", "speaker": "librispeech_4507", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4507-16021-0045", "transcript": "IT IS SO MADE THAT EVERYWHERE WE FEEL THE SENSE OF PUNISHMENT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4507/16021/4507-16021-0045.flac"}, "sample_rate": 16000, "source_samples": 78240, "source_duration_seconds": 4.89, "source_peak": 0.550750732421875, "source_rms_dbfs": -25.788181877969095, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.788181877969095, "decoded_pcm_sha256": "ef91d0d7b9ebfc4c08837a0dcebf79c72e3e4ff97d381f356a678c2a085d6e6c", "source_sha256": "86e41fe1ab9a5fdf5d806ab2728428f4b8d2720df04ce6431b65f617878b924e", "wav_sha256": "52e640aa0a3813985286616d54082fb10048f8acae7858ee0fdbef99208e9f66", "local_source": "audio/librispeech_4507-16021-0045.flac", "local_wav": "audio/librispeech_4507-16021-0045.wav", "excerpt_start_sample": 0, "excerpt_samples": 78240}, {"id": "librispeech_4970-29093-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_4970", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4970-29093-0004", "transcript": "HE WAS UNABLE TO DECIDE EXACTLY WHAT IT SHOULD BE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4970/29093/4970-29093-0004.flac"}, "sample_rate": 16000, "source_samples": 60000, "source_duration_seconds": 3.75, "source_peak": 0.652984619140625, "source_rms_dbfs": -25.09401331193901, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.09401331193901, "decoded_pcm_sha256": "222e257a563c92e3e220ca509960e218a09ffd8f995b3f1148bba432a24fc5cf", "source_sha256": "48c89f99c44d1be73c92460e6d61d4c1090efe126701f7492184112d37575229", "wav_sha256": "c1ad0bc6656e5a37b40d868e2b1df01441c4a9fbd5afe93f389122fc10be9f4d", "local_source": "audio/librispeech_4970-29093-0004.flac", "local_wav": "audio/librispeech_4970-29093-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 60000}, {"id": "librispeech_4970-29093-0005", "corpus": "librispeech_test_clean", "speaker": "librispeech_4970", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4970-29093-0005", "transcript": "SOMETIMES HE THOUGHT HE WOULD LIKE TO STAND IN A CONSPICUOUS PULPIT AND HUMBLY PREACH THE GOSPEL OF REPENTANCE AND IT EVEN CROSSED HIS MIND THAT IT WOULD BE NOBLE TO GIVE HIMSELF TO A MISSIONARY LIFE TO SOME BENIGHTED REGION WHERE THE DATE PALM GROWS AND THE NIGHTINGALE'S VOICE IS IN TUNE AND THE BUL BUL SINGS ON THE OFF NIGHTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4970/29093/4970-29093-0005.flac"}, "sample_rate": 16000, "source_samples": 354560, "source_duration_seconds": 22.16, "source_peak": 0.676025390625, "source_rms_dbfs": -25.92844507815476, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.632807866301704, "decoded_pcm_sha256": "e5b971d7dd1c4efc72382481eb6d7cbe047355a62af1c39714d28b6cb739b74e", "source_sha256": "f466147cca3169748874913da56b587fd3d7fed00a285a1c9124dac56a8323ba", "wav_sha256": "37a020e1d205b7fb675b2ff89e777850155dfad1f71f9d58087cea696b62bd39", "local_source": "audio/librispeech_4970-29093-0005.flac", "local_wav": "audio/librispeech_4970-29093-0005.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4970-29095-0008", "corpus": "librispeech_test_clean", "speaker": "librispeech_4970", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4970-29095-0008", "transcript": "MOTHER I'M GOING TO STUDY MEDICINE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4970/29095/4970-29095-0008.flac"}, "sample_rate": 16000, "source_samples": 48640, "source_duration_seconds": 3.04, "source_peak": 0.34185791015625, "source_rms_dbfs": -28.408655353194106, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.408655353194106, "decoded_pcm_sha256": "44ec7bab382a51da3a9f07cbc8ffdb491029ea5b6b84d63bfceb71c60641803b", "source_sha256": "cd2a3b352b93beb33caced744406acf8dace3f1bbb5fc63c8feb4c8a0ed55bb4", "wav_sha256": "1dc0a4e22ebfc2ff3d5048978b903074647fdb65e2c0030918f94bccfe72e63f", "local_source": "audio/librispeech_4970-29095-0008.flac", "local_wav": "audio/librispeech_4970-29095-0008.wav", "excerpt_start_sample": 0, "excerpt_samples": 48640}, {"id": "librispeech_4992-23283-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_4992", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4992-23283-0016", "transcript": "AGAIN HE SEARCHED HIS OWN THOUGHTS NOR INEFFECTUALLY AS BEFORE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4992/23283/4992-23283-0016.flac"}, "sample_rate": 16000, "source_samples": 71920, "source_duration_seconds": 4.495, "source_peak": 0.317962646484375, "source_rms_dbfs": -28.34849352834874, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.34849352834874, "decoded_pcm_sha256": "4af74b1b8682fb324d2e81cb2f0cc0dc50b77d4a3b44fd510cfb4df09ae57d6d", "source_sha256": "e7442a8f1bbd834ad37ee3c9cc5fa26b1cfb61446f827b54298716f59b010002", "wav_sha256": "e74da3b395ffef1f2fc80932d932ba92ecb3611d9ca1f06743f0a824ef2c9468", "local_source": "audio/librispeech_4992-23283-0016.flac", "local_wav": "audio/librispeech_4992-23283-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 71920}, {"id": "librispeech_4992-41797-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_4992", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4992-41797-0000", "transcript": "YES DEAD THESE FOUR YEARS AN A GOOD JOB FOR HER TOO", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4992/41797/4992-41797-0000.flac"}, "sample_rate": 16000, "source_samples": 87760, "source_duration_seconds": 5.485, "source_peak": 0.48004150390625, "source_rms_dbfs": -23.74627112170186, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.74627112170186, "decoded_pcm_sha256": "70049242cb761e6a91df355dd7a9dbdfb974c89071ee6b5dd83506a6ce8390e3", "source_sha256": "f7dbbd42eb06171dccf937d53a6c9144e51b88ea5b1b2ee69978900b87216fb5", "wav_sha256": "d8a54528a66df71baf1fe8b7d60083abd4034bbb2cb4331344d960b542806bbc", "local_source": "audio/librispeech_4992-41797-0000.flac", "local_wav": "audio/librispeech_4992-41797-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 87760}, {"id": "librispeech_4992-41806-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_4992", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4992-41806-0000", "transcript": "NATTY HARMON TRIED THE KITCHEN PUMP SECRETLY SEVERAL TIMES DURING THE EVENING FOR THE WATER HAD TO RUN UP HILL ALL THE WAY FROM THE WELL TO THE KITCHEN SINK AND HE BELIEVED THIS TO BE A CONTINUAL MIRACLE THAT MIGHT GIVE OUT AT ANY MOMENT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4992/41806/4992-41806-0000.flac"}, "sample_rate": 16000, "source_samples": 229760, "source_duration_seconds": 14.36, "source_peak": 0.582489013671875, "source_rms_dbfs": -21.657320978064327, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.055751743922045, "decoded_pcm_sha256": "9e17e42f2de437c240d5850615912f710137bb8bb0f2e42082d93bc758515234", "source_sha256": "ce4dd04d130b3b575f1826c6505b5519c09a7301eee9b76e58eeb1f591131035", "wav_sha256": "04109bfdf70350f438b7afe006744ec48bab6fdb23019457ec981bb2cb1518fb", "local_source": "audio/librispeech_4992-41806-0000.flac", "local_wav": "audio/librispeech_4992-41806-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5105-28240-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_5105", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5105-28240-0011", "transcript": "I LEFT YOU ON A CONTINENT AND HERE I HAVE THE HONOR OF FINDING YOU ON AN ISLAND", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5105/28240/5105-28240-0011.flac"}, "sample_rate": 16000, "source_samples": 96320, "source_duration_seconds": 6.02, "source_peak": 0.599212646484375, "source_rms_dbfs": -24.990381079106882, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.975943389244275, "decoded_pcm_sha256": "b9bc3828aa0284c0660c9ec7acd1980026fcc21a26b62a3c5c68458711b73ba2", "source_sha256": "2e162ff81e45f909e947ebfae4f39e339d73d5fc19afafb1c903db96f7f8532d", "wav_sha256": "94233ddbd1f16127374435a042a6df2adeba0af01cb759c656e84cd9293ac5ac", "local_source": "audio/librispeech_5105-28240-0011.flac", "local_wav": "audio/librispeech_5105-28240-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5105-28241-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_5105", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5105-28241-0004", "transcript": "ALTHOUGH ONLY A MODERATE BREEZE WAS BLOWING THE SEA WAS ROUGH A CIRCUMSTANCE TO BE ACCOUNTED FOR ONLY BY THE DIMINUTION IN THE FORCE OF THE EARTH'S ATTRACTION RENDERING THE LIQUID PARTICLES SO BUOYANT THAT BY THE MERE EFFECT OF OSCILLATION THEY WERE CARRIED TO A HEIGHT THAT WAS QUITE UNPRECEDENTED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5105/28241/5105-28241-0004.flac"}, "sample_rate": 16000, "source_samples": 278640, "source_duration_seconds": 17.415, "source_peak": 0.83734130859375, "source_rms_dbfs": -26.537292077408626, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.37366219728025, "decoded_pcm_sha256": "6704bc85da6844b255174407f722499ab1188299d4775a37e25cc238c7fb903e", "source_sha256": "70c67685f8e24f7a27f6de3b63d5b3ecbde28f017d1b7aac9304f6c6034fa1bc", "wav_sha256": "4497c937acc206dd49bddec67200ad51f5097466123879a49f6a3ee336f15a69", "local_source": "audio/librispeech_5105-28241-0004.flac", "local_wav": "audio/librispeech_5105-28241-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5105-28241-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_5105", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5105-28241-0019", "transcript": "NOTHING WAS TO BE DONE BUT TO PUT ABOUT AND RETURN IN DISAPPOINTMENT TOWARDS THE NORTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5105/28241/5105-28241-0019.flac"}, "sample_rate": 16000, "source_samples": 84640, "source_duration_seconds": 5.29, "source_peak": 0.58538818359375, "source_rms_dbfs": -29.753814832387818, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.753814832387818, "decoded_pcm_sha256": "3d0b0ee3168cffdc7dff07188417fdc9c6c2c31369b906ef6a97ab560d3e8f4b", "source_sha256": "96d523790a28d90ffd89ea67603421a3881a96d21cb523495d1a066c4c5ceecd", "wav_sha256": "9c73b0201f91008baedeb8630a30a0d9ff83ad92b23f14df405ae17c8cd5e664", "local_source": "audio/librispeech_5105-28241-0019.flac", "local_wav": "audio/librispeech_5105-28241-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 84640}, {"id": "librispeech_5142-33396-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_5142", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5142-33396-0019", "transcript": "OH IT IS BETTER TO LIVE ON THE SEA AND LET OTHER MEN RAISE YOUR CROPS AND COOK YOUR MEALS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5142/33396/5142-33396-0019.flac"}, "sample_rate": 16000, "source_samples": 79760, "source_duration_seconds": 4.985, "source_peak": 0.376739501953125, "source_rms_dbfs": -26.870219951708695, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.870219951708695, "decoded_pcm_sha256": "f3e12cfe88aca2fb306552ad3fe2c213b91a4bfc2f0aa979df8a1f01006ba85d", "source_sha256": "2025e96df477d9ebc46b91bae88c601221bbbb2ab12d9fa83ceced80b0f9b72f", "wav_sha256": "20b7da9ef469c7b32d3ebf290f98c5e8058c718ee8ccd7d3f172ccc870264c9a", "local_source": "audio/librispeech_5142-33396-0019.flac", "local_wav": "audio/librispeech_5142-33396-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 79760}, {"id": "librispeech_5142-33396-0030", "corpus": "librispeech_test_clean", "speaker": "librispeech_5142", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5142-33396-0030", "transcript": "THE THRALLS WERE BRINGING IN A GREAT POT OF MEAT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5142/33396/5142-33396-0030.flac"}, "sample_rate": 16000, "source_samples": 44240, "source_duration_seconds": 2.765, "source_peak": 0.41986083984375, "source_rms_dbfs": -28.781219695797876, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.781219695797876, "decoded_pcm_sha256": "cdc8b64c805f94ff6dd7f431b45e924a4edf992efb5d08cda66456142b76481e", "source_sha256": "87b13e7bd9a84ee5f9ba237f84dc992d0fbbad061a6df1750b87a47756642b33", "wav_sha256": "8732110f0034b47a150b49823c48d798c5010d9ea27a2236efaa6871fd045d5b", "local_source": "audio/librispeech_5142-33396-0030.flac", "local_wav": "audio/librispeech_5142-33396-0030.wav", "excerpt_start_sample": 0, "excerpt_samples": 44240}, {"id": "librispeech_5142-33396-0065", "corpus": "librispeech_test_clean", "speaker": "librispeech_5142", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5142-33396-0065", "transcript": "SOFT HEART HE SAID GENTLY TO HER THEN TO THORKEL WELL LET HIM GO THORKEL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5142/33396/5142-33396-0065.flac"}, "sample_rate": 16000, "source_samples": 83120, "source_duration_seconds": 5.195, "source_peak": 0.5650634765625, "source_rms_dbfs": -29.734936973090612, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.734936973090612, "decoded_pcm_sha256": "0b650128a600d9d1c16e5bf3b64c63018c80e6b3463b2d6fb23e3aead6248f75", "source_sha256": "67800641844bf8c9d654fc9a6099941ba97ff31b0ac0a579209378598c4ea09b", "wav_sha256": "0aace7ee3326bd6ad0ef18685074f6bd6f6749bd1b9c5fd08080a225d32a3ed8", "local_source": "audio/librispeech_5142-33396-0065.flac", "local_wav": "audio/librispeech_5142-33396-0065.wav", "excerpt_start_sample": 0, "excerpt_samples": 83120}, {"id": "librispeech_5639-40744-0013", "corpus": "librispeech_test_clean", "speaker": "librispeech_5639", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5639-40744-0013", "transcript": "SHE SAW THAT THE BED WAS GILDED AND SO RICH THAT IT SEEMED THAT OF A PRINCE RATHER THAN OF A PRIVATE GENTLEMAN", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5639/40744/5639-40744-0013.flac"}, "sample_rate": 16000, "source_samples": 109840, "source_duration_seconds": 6.865, "source_peak": 0.287109375, "source_rms_dbfs": -32.469388774245715, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -32.072413641742806, "decoded_pcm_sha256": "ed7f1ce67f421ec6afa8566971a1c9c8ec8d332f4950f7baa3aacb009b96f5ef", "source_sha256": "a2688cb66fd448ee0d578c6d82312afea0f9ad95b24350f251eb79ca5a4f759e", "wav_sha256": "57e719154dbc8c577990ab1a3b41317f9774b3a08c636666478a6d070c102aa9", "local_source": "audio/librispeech_5639-40744-0013.flac", "local_wav": "audio/librispeech_5639-40744-0013.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5639-40744-0023", "corpus": "librispeech_test_clean", "speaker": "librispeech_5639", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5639-40744-0023", "transcript": "WHEN THE BOY WALKED THROUGH THE STREETS BLESSINGS WERE SHOWERED UPON HIM BY ALL WHO SAW HIM BLESSINGS UPON HIS BEAUTY UPON THE MOTHER THAT BORE HIM UPON THE FATHER THAT BEGOT HIM UPON THOSE WHO BROUGHT HIM UP SO WELL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5639/40744/5639-40744-0023.flac"}, "sample_rate": 16000, "source_samples": 222720, "source_duration_seconds": 13.92, "source_peak": 0.2412109375, "source_rms_dbfs": -30.75070314546162, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.202866501530522, "decoded_pcm_sha256": "da6b45c38da036a67d46d62fc3e6e54310d46e3e14d8f0a3eff1b445bb1755f4", "source_sha256": "dd277d4b42ef738b14b59fa1d835988e71ec031e2f01d024279202edc26ad9c0", "wav_sha256": "3de21cbccb1dbd2ff7b98796f239d0fb4ae4abe216c2f77e5a5d068ec8ed4d89", "local_source": "audio/librispeech_5639-40744-0023.flac", "local_wav": "audio/librispeech_5639-40744-0023.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5639-40744-0030", "corpus": "librispeech_test_clean", "speaker": "librispeech_5639", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5639-40744-0030", "transcript": "JUST THEN LEOCADIA CAME TO HERSELF AND EMBRACING THE CROSS SEEMED CHANGED INTO A SEA OF TEARS AND THE GENTLEMAN REMAINED IN UTTER BEWILDERMENT UNTIL HIS WIFE HAD REPEATED TO HIM FROM BEGINNING TO END LEOCADIA'S WHOLE STORY AND HE BELIEVED IT THROUGH THE BLESSED DISPENSATION OF HEAVEN WHICH HAD CONFIRMED IT BY SO MANY CONVINCING TESTIMONIES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5639/40744/5639-40744-0030.flac"}, "sample_rate": 16000, "source_samples": 356560, "source_duration_seconds": 22.285, "source_peak": 0.222198486328125, "source_rms_dbfs": -34.05232832381272, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -33.056314847596, "decoded_pcm_sha256": "76c45c30d066507ae3cbb52fafb188ebe36da864c7c54942db2ea8356726ce3a", "source_sha256": "c7f95b2f012ab8f4629d15c8c78e78bf17b3445c96e5a9b0c525b0743e50d9c5", "wav_sha256": "257fa9e1dafacf00ce9d6925632b4e3904f46e59298d19f84b3d14211b4df0a2", "local_source": "audio/librispeech_5639-40744-0030.flac", "local_wav": "audio/librispeech_5639-40744-0030.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5683-32865-0017", "corpus": "librispeech_test_clean", "speaker": "librispeech_5683", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5683-32865-0017", "transcript": "ALL THE TIME HE WAS TALKING TO ME HIS ANGRY LITTLE EYES WERE FOLLOWING LAKE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5683/32865/5683-32865-0017.flac"}, "sample_rate": 16000, "source_samples": 87280, "source_duration_seconds": 5.455, "source_peak": 0.2510986328125, "source_rms_dbfs": -34.9940582978409, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -34.9940582978409, "decoded_pcm_sha256": "44477e160959a9526d3545567baa9b096405cf9d914595d737ffdf3d46ce1e6f", "source_sha256": "d0027b49e5e0c29b3b59d1d782b8d4d9753f90e4343f243a15698e787c8e64ca", "wav_sha256": "bfb05fe49b3082f400d278836c3aff46c9a56c7d7919fbc0de59a5c012bbdb55", "local_source": "audio/librispeech_5683-32865-0017.flac", "local_wav": "audio/librispeech_5683-32865-0017.wav", "excerpt_start_sample": 0, "excerpt_samples": 87280}, {"id": "librispeech_5683-32879-0009", "corpus": "librispeech_test_clean", "speaker": "librispeech_5683", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5683-32879-0009", "transcript": "ILL AND TROUBLED DEAR TROUBLED IN MIND AND MISERABLY NERVOUS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5683/32879/5683-32879-0009.flac"}, "sample_rate": 16000, "source_samples": 82160, "source_duration_seconds": 5.135, "source_peak": 0.744171142578125, "source_rms_dbfs": -22.065935064466796, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.065935064466796, "decoded_pcm_sha256": "c0eff412ac130cf1ccdb29b7cbb8149ed148614513d6a355a1677d911d2c6b5f", "source_sha256": "c66196c0436cce295ac3ae7695f4904f87672f1d63d89b8810937a146378411a", "wav_sha256": "eea99813322332eff96b5f9c8124d990192b34c98b8b51879f7e89d7170803d7", "local_source": "audio/librispeech_5683-32879-0009.flac", "local_wav": "audio/librispeech_5683-32879-0009.wav", "excerpt_start_sample": 0, "excerpt_samples": 82160}, {"id": "librispeech_5683-32879-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_5683", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5683-32879-0019", "transcript": "I HAVE VERY FEW TO LOVE ME NOW AND I THOUGHT YOU MIGHT LOVE ME AS I HAVE BEGUN TO LOVE YOU", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5683/32879/5683-32879-0019.flac"}, "sample_rate": 16000, "source_samples": 101600, "source_duration_seconds": 6.35, "source_peak": 0.799591064453125, "source_rms_dbfs": -22.192801992464837, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.95646015705721, "decoded_pcm_sha256": "cf57296e7013c02425f574c6c3f30623e8057ab1e199fc8fcd761f1b2ff63d52", "source_sha256": "0c9375f8c28b1576aaa73e07006a7fc838f49b250df910885e24a16abef0c19f", "wav_sha256": "c8d43ac515baa4c0d047816bffd947306c829e56c1cbd3aea016eb764930c45d", "local_source": "audio/librispeech_5683-32879-0019.flac", "local_wav": "audio/librispeech_5683-32879-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_61-70968-0042", "corpus": "librispeech_test_clean", "speaker": "librispeech_61", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "61-70968-0042", "transcript": "MONTFICHET CALLED OUT FOR ROBIN TO GIVE HIM AN ARM", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/61/70968/61-70968-0042.flac"}, "sample_rate": 16000, "source_samples": 44560, "source_duration_seconds": 2.785, "source_peak": 0.362152099609375, "source_rms_dbfs": -27.019043277748153, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.019043277748153, "decoded_pcm_sha256": "4d1c295afd61c34ca0f59a70b2baace0777ff6e94127ed717b67dc664c9d9ee4", "source_sha256": "9aa0aa0746140383c3458b53991ccd9ebfc2c0ded6c6817ec0748aa70e6b8088", "wav_sha256": "6341417e6e575c7204a790e5368cdbc7d10afd5dce8bd9816b8670f57a6efeb1", "local_source": "audio/librispeech_61-70968-0042.flac", "local_wav": "audio/librispeech_61-70968-0042.wav", "excerpt_start_sample": 0, "excerpt_samples": 44560}, {"id": "librispeech_61-70968-0043", "corpus": "librispeech_test_clean", "speaker": "librispeech_61", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "61-70968-0043", "transcript": "FRIENDS SAID MONTFICHET FAINTLY TO THE WRESTLERS BEAR US ESCORT SO FAR AS THE SHERIFF'S HOUSE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/61/70968/61-70968-0043.flac"}, "sample_rate": 16000, "source_samples": 107760, "source_duration_seconds": 6.735, "source_peak": 0.53997802734375, "source_rms_dbfs": -27.886439724320226, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.565805365496097, "decoded_pcm_sha256": "1731c4205d3ab140aea21c6e2df47e27d59a45b1eaa6b80c407cddcad99de7af", "source_sha256": "92cde1bfb6bab32cd1428505868f1ca27db04ed17bec9e47e7cbb24e34dc09c3", "wav_sha256": "86ec6218c4ceffd4402d269164b03d365c360d751eb3b8e6078808e4ace33916", "local_source": "audio/librispeech_61-70968-0043.flac", "local_wav": "audio/librispeech_61-70968-0043.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_61-70968-0054", "corpus": "librispeech_test_clean", "speaker": "librispeech_61", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "61-70968-0054", "transcript": "MISTRESS FITZOOTH HAD BEEN CARRIED OFF BY THE SHERIFF'S DAUGHTER AND HER MAIDS AS SOON AS THEY HAD ENTERED THE HOUSE SO THAT ROBIN ALONE HAD THE CARE OF MONTFICHET", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/61/70968/61-70968-0054.flac"}, "sample_rate": 16000, "source_samples": 125760, "source_duration_seconds": 7.86, "source_peak": 0.413848876953125, "source_rms_dbfs": -26.416451889769732, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.102361433729442, "decoded_pcm_sha256": "0c30dfc3b23081f52641ba6b1793c8bc5c9ea236aea529b533de133f00169e95", "source_sha256": "7b58da65dd63a6b09d43ee20c6aa4b5a8853199af7c350a87ab692fb4d0d874c", "wav_sha256": "6650129147c01f0d91d5e0a36f5c40ce37be0edd09c841daac604d1bf7ecfa24", "local_source": "audio/librispeech_61-70968-0054.flac", "local_wav": "audio/librispeech_61-70968-0054.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_672-122797-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_672", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "672-122797-0003", "transcript": "BUT THIS WAS WHAT THE TREE COULD NOT BEAR TO HEAR", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/672/122797/672-122797-0003.flac"}, "sample_rate": 16000, "source_samples": 76160, "source_duration_seconds": 4.76, "source_peak": 0.368072509765625, "source_rms_dbfs": -30.00537875055086, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.00537875055086, "decoded_pcm_sha256": "4d9f339e6dbd2372c553d9d5f6718849a37e2770b2e7e3a43beccd5922c086ec", "source_sha256": "b00cbf4705fcd94f9ea9130da79993f83b7e3a83313328314b3492d19295af94", "wav_sha256": "939b4401889c28674fc8462e95e87c8aa498a702648053a96efaa49dd3a4d6e9", "local_source": "audio/librispeech_672-122797-0003.flac", "local_wav": "audio/librispeech_672-122797-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 76160}, {"id": "librispeech_672-122797-0044", "corpus": "librispeech_test_clean", "speaker": "librispeech_672", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "672-122797-0044", "transcript": "AND HE LEANED AGAINST THE WALL LOST IN REVERIE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/672/122797/672-122797-0044.flac"}, "sample_rate": 16000, "source_samples": 59840, "source_duration_seconds": 3.74, "source_peak": 0.291778564453125, "source_rms_dbfs": -28.28836846886545, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.28836846886545, "decoded_pcm_sha256": "432f9a1ae86511f6d52017a356d8d76c16009ab41a0b0e212b50e3c7fe63423f", "source_sha256": "30dfd88ccc8b641d250263094e829de48032160f0e8f954c0487a8ecd0da85cc", "wav_sha256": "e521e1a275330db01220cdbf937c18de9ae32a2869b60191d0e45f37d008553a", "local_source": "audio/librispeech_672-122797-0044.flac", "local_wav": "audio/librispeech_672-122797-0044.wav", "excerpt_start_sample": 0, "excerpt_samples": 59840}, {"id": "librispeech_672-122797-0056", "corpus": "librispeech_test_clean", "speaker": "librispeech_672", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "672-122797-0056", "transcript": "SAID THE FIR TREE THINKING OVER WHAT HE HAD HIMSELF RELATED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/672/122797/672-122797-0056.flac"}, "sample_rate": 16000, "source_samples": 83600, "source_duration_seconds": 5.225, "source_peak": 0.34185791015625, "source_rms_dbfs": -29.44078110552336, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.44078110552336, "decoded_pcm_sha256": "1ee86311a9a7b6ee7bf0ef8ccad1d6c7f831fa6001565687e3d9e0eb90cd44f4", "source_sha256": "e58678d63f84c218fb28da622351b90e344380dbfc13763dd1123e71faf708d1", "wav_sha256": "c2d2cf8f5b61062744b09b2ded2dd8e540b5984a424be0cef6ae7903b9072300", "local_source": "audio/librispeech_672-122797-0056.flac", "local_wav": "audio/librispeech_672-122797-0056.wav", "excerpt_start_sample": 0, "excerpt_samples": 83600}, {"id": "librispeech_6829-68769-0014", "corpus": "librispeech_test_clean", "speaker": "librispeech_6829", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "6829-68769-0014", "transcript": "THEY FOLLOWED THE JAILER ALONG A SUCCESSION OF PASSAGES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6829/68769/6829-68769-0014.flac"}, "sample_rate": 16000, "source_samples": 58480, "source_duration_seconds": 3.655, "source_peak": 0.45867919921875, "source_rms_dbfs": -23.65210599160649, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.65210599160649, "decoded_pcm_sha256": "f8c838b07c2dd578b611a15c028e07c92a592c03766bbc5d9020f61b8ed5195e", "source_sha256": "c228d6cd297ffb4916b952d7d5861d8ac8225058219ec05b587d8dd4bd041179", "wav_sha256": "1297807e7c7832e33167b40d55c2795aab4db406c144ff8aa315638314ee3bf5", "local_source": "audio/librispeech_6829-68769-0014.flac", "local_wav": "audio/librispeech_6829-68769-0014.wav", "excerpt_start_sample": 0, "excerpt_samples": 58480}, {"id": "librispeech_6829-68769-0038", "corpus": "librispeech_test_clean", "speaker": "librispeech_6829", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "6829-68769-0038", "transcript": "AND IT RUINS A MAN'S DISPOSITION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6829/68769/6829-68769-0038.flac"}, "sample_rate": 16000, "source_samples": 40080, "source_duration_seconds": 2.505, "source_peak": 0.4256591796875, "source_rms_dbfs": -24.099666920041205, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.099666920041205, "decoded_pcm_sha256": "2491d28a0087423aef64b260b9a3bef96a9ffd38bdd5460aeb1d8818a5a4e0a8", "source_sha256": "0f141cbbe137a2927bbf38106a3b14d2b2f186b86e1a54bcbd354d8c2a8981c4", "wav_sha256": "4f8dddd42e85f0a1e91a1b93df0c6a50013c6a166ff3a1bcdf0e54f9924f9ea1", "local_source": "audio/librispeech_6829-68769-0038.flac", "local_wav": "audio/librispeech_6829-68769-0038.wav", "excerpt_start_sample": 0, "excerpt_samples": 40080}, {"id": "librispeech_6829-68769-0041", "corpus": "librispeech_test_clean", "speaker": "librispeech_6829", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "6829-68769-0041", "transcript": "I'M NOT ELECTIONEERING JUST NOW", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6829/68769/6829-68769-0041.flac"}, "sample_rate": 16000, "source_samples": 42560, "source_duration_seconds": 2.66, "source_peak": 0.617095947265625, "source_rms_dbfs": -23.43903741251144, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.43903741251144, "decoded_pcm_sha256": "d14d8bc17d20f03134a1360b706d2d290d10240225382c5adf8b0919adb152e4", "source_sha256": "e0e3964567d7a515d2443c6e1eff105f88da0e87ef44cc2954b36f1b9d93096c", "wav_sha256": "f34de7e7dd5bb7962b7644864833e52a9a705d8b11000ee7070e9b63e310f6ad", "local_source": "audio/librispeech_6829-68769-0041.flac", "local_wav": "audio/librispeech_6829-68769-0041.wav", "excerpt_start_sample": 0, "excerpt_samples": 42560}, {"id": "librispeech_6930-75918-0013", "corpus": "librispeech_test_clean", "speaker": "librispeech_6930", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "6930-75918-0013", "transcript": "IN THOSE VERY TERMS I EVEN ADDED MORE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6930/75918/6930-75918-0013.flac"}, "sample_rate": 16000, "source_samples": 47040, "source_duration_seconds": 2.94, "source_peak": 0.408905029296875, "source_rms_dbfs": -29.974578166391865, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.974578166391865, "decoded_pcm_sha256": "3f54b1539585ee3c865737cd0781dd64a40895a5d1930b99c37db78e450618a7", "source_sha256": "83f99c357f446d1fddffc61c09f59eb8a41c82c5e4261db42c71396c537c409e", "wav_sha256": "6b56ade8142254ab2fddf538d8164b7043ee7a100b957e3f941f9e2a2de64b9c", "local_source": "audio/librispeech_6930-75918-0013.flac", "local_wav": "audio/librispeech_6930-75918-0013.wav", "excerpt_start_sample": 0, "excerpt_samples": 47040}, {"id": "librispeech_6930-76324-0024", "corpus": "librispeech_test_clean", "speaker": "librispeech_6930", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "6930-76324-0024", "transcript": "THEY SAY ILLUMINATION BY CANDLE LIGHT IS THE PRETTIEST IN THE WORLD", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6930/76324/6930-76324-0024.flac"}, "sample_rate": 16000, "source_samples": 64800, "source_duration_seconds": 4.05, "source_peak": 0.373291015625, "source_rms_dbfs": -28.918530059065088, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.918530059065088, "decoded_pcm_sha256": "d8dd1ce759e26c72605c90e837ae65e87caae1e3095da47a3f97063f55664b19", "source_sha256": "5d8bd7d094fe646881a4848026dbf1a3748ec90f170b87d2f0836adcfd9b7f7f", "wav_sha256": "894af1a21bcf4de58645a4cc3b9ec88c579dc75aef0c896f4aeacea6cdb9640d", "local_source": "audio/librispeech_6930-76324-0024.flac", "local_wav": "audio/librispeech_6930-76324-0024.wav", "excerpt_start_sample": 0, "excerpt_samples": 64800}, {"id": "librispeech_6930-81414-0026", "corpus": "librispeech_test_clean", "speaker": "librispeech_6930", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "6930-81414-0026", "transcript": "MY OVERWROUGHT NERVES YIELDED AT LAST", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6930/81414/6930-81414-0026.flac"}, "sample_rate": 16000, "source_samples": 49200, "source_duration_seconds": 3.075, "source_peak": 0.434173583984375, "source_rms_dbfs": -27.195611560424933, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.195611560424933, "decoded_pcm_sha256": "6ea614ae159242a4cb64e926932b9c1cf0c8b84fb4912932e606a31b6d48c909", "source_sha256": "84048bb63a1968e9856f38f430d293fd5036cc1b8c31312f3aff96d087732145", "wav_sha256": "8e930ede632ab23724d64be8cf34e2f53d115e42f4dbb2c644589c9510008d66", "local_source": "audio/librispeech_6930-81414-0026.flac", "local_wav": "audio/librispeech_6930-81414-0026.wav", "excerpt_start_sample": 0, "excerpt_samples": 49200}, {"id": "librispeech_7021-79740-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_7021", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7021-79740-0002", "transcript": "NOW DELIA CONTRIVED TO OBTAIN A GREAT INFLUENCE AND ASCENDENCY OVER THE MINDS OF THE CHILDREN BY MEANS OF THESE DOLLS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7021/79740/7021-79740-0002.flac"}, "sample_rate": 16000, "source_samples": 147600, "source_duration_seconds": 9.225, "source_peak": 0.571044921875, "source_rms_dbfs": -24.145550365790243, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.686648589877173, "decoded_pcm_sha256": "32223cc598805a0301d5fd306db57f06a7d2c624ffce3ec523d6a45df0f1b192", "source_sha256": "df09e0bfaf8000a8183b367c2fbc6ce3404f4152ef14d9340515b37edb7155ba", "wav_sha256": "15dbc2fc45c2fb603bb8186e866988f9615538186eb94834e17d09a846b828ef", "local_source": "audio/librispeech_7021-79740-0002.flac", "local_wav": "audio/librispeech_7021-79740-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7021-79740-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_7021", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7021-79740-0011", "transcript": "SO SAYING SHE LED THE WAY ON TIPTOE FOLLOWED BY THE CHILDREN OUT OF THE ROOM AND ROUND BY A CIRCUITOUS ROUTE TO THE PIAZZA THERE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7021/79740/7021-79740-0011.flac"}, "sample_rate": 16000, "source_samples": 165040, "source_duration_seconds": 10.315, "source_peak": 0.536102294921875, "source_rms_dbfs": -24.714389336771436, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.228922794174977, "decoded_pcm_sha256": "6cda455d70e2130c6b8f8269b688ec0381f52c4c07376a3f0f4bbbf73ca24fc7", "source_sha256": "af3d9a4d3df12f73bc4432b8be4762df32f7381a8c20009735ca8e982084c2a6", "wav_sha256": "cf60e39e5c031f055d45d2c1943b8eae811ecff60ac6593bb402412220cc0cd6", "local_source": "audio/librispeech_7021-79740-0011.flac", "local_wav": "audio/librispeech_7021-79740-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7021-85628-0025", "corpus": "librispeech_test_clean", "speaker": "librispeech_7021", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7021-85628-0025", "transcript": "BUT HIS MOTHER HUGGED HIM CLOSE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7021/85628/7021-85628-0025.flac"}, "sample_rate": 16000, "source_samples": 44400, "source_duration_seconds": 2.775, "source_peak": 0.411346435546875, "source_rms_dbfs": -25.512275213051296, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.512275213051296, "decoded_pcm_sha256": "f8114fb1a86ed60fb5b4e5ba870142361984ad8304c63676d1f9c029942602ee", "source_sha256": "477a753670424f1fd835f77fdd15e54570114ec1a4f99c206e555e94137688c6", "wav_sha256": "4acccc54bcc7e762d4e6fc8d1cbd7625bb0ec4c79de17ddb8382b1ce7fe8cffe", "local_source": "audio/librispeech_7021-85628-0025.flac", "local_wav": "audio/librispeech_7021-85628-0025.wav", "excerpt_start_sample": 0, "excerpt_samples": 44400}, {"id": "librispeech_7127-75946-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_7127", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7127-75946-0002", "transcript": "LET HIM COME IN THEN SAID THE KING AND AS IF COLBERT HAD BEEN LISTENING AT THE DOOR FOR THE PURPOSE OF KEEPING HIMSELF AU COURANT WITH THE CONVERSATION HE ENTERED AS SOON AS THE KING HAD PRONOUNCED HIS NAME TO THE TWO COURTIERS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7127/75946/7127-75946-0002.flac"}, "sample_rate": 16000, "source_samples": 211040, "source_duration_seconds": 13.19, "source_peak": 0.515167236328125, "source_rms_dbfs": -26.998476985882505, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.14563278485655, "decoded_pcm_sha256": "93c54b2cc025d80c3059dc9d273712d4259e0ddbef769ef56e3f3d16805ad734", "source_sha256": "8a4077cfaf005d616cab7fe18bc2a434b627b24de056d78cafc33f09faa1cf6c", "wav_sha256": "ccc35cdc72cfa24627ef25ab7fdda0305587a1d469a9ca5b3adf519ca7bc2926", "local_source": "audio/librispeech_7127-75946-0002.flac", "local_wav": "audio/librispeech_7127-75946-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7127-75946-0029", "corpus": "librispeech_test_clean", "speaker": "librispeech_7127", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7127-75946-0029", "transcript": "THE KING WHO HAD FROM THIS MOMENT BECOME IN REALITY THE PRINCIPAL DANCER IN THE QUADRILLE CAST A LOOK UPON HIS VANQUISHED RIVAL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7127/75946/7127-75946-0029.flac"}, "sample_rate": 16000, "source_samples": 148560, "source_duration_seconds": 9.285, "source_peak": 0.531982421875, "source_rms_dbfs": -28.20061844499801, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.58805214770615, "decoded_pcm_sha256": "92db6e9b9692ce0f9a8c0ad4504635e4a6e649b6d8760fca3400f3ea75d63e54", "source_sha256": "e26b6b1a8cceb2a05b3aae15719c785d4d24a2a09d91dd6292da8a4b4e32401a", "wav_sha256": "a0738b9c79c82ef114bc52a64dee558b82ab4c4e6d068a5d9c0429add5aa4272", "local_source": "audio/librispeech_7127-75946-0029.flac", "local_wav": "audio/librispeech_7127-75946-0029.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7127-75947-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_7127", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7127-75947-0011", "transcript": "REMAIN I IMPLORE YOU THE EVENING IS MOST LOVELY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7127/75947/7127-75947-0011.flac"}, "sample_rate": 16000, "source_samples": 57920, "source_duration_seconds": 3.62, "source_peak": 0.337554931640625, "source_rms_dbfs": -25.11875836284418, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.11875836284418, "decoded_pcm_sha256": "aebe57a65af629b724d8ca28eb599d83587cc213e6e303682e713453ab9de80a", "source_sha256": "e1bbd2d0c4d8800ab59fe4f4a1f027601dc96c787ad8bf50fd97e051270ee72b", "wav_sha256": "c31a1ac369e0120b546cca7a9eb59600c80e55118335293e80c3aafefc23467e", "local_source": "audio/librispeech_7127-75947-0011.flac", "local_wav": "audio/librispeech_7127-75947-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 57920}, {"id": "librispeech_7176-88083-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_7176", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7176-88083-0002", "transcript": "HIS FEET WERE RED HIS LONG NARROW BEAK WITH ITS SAW TOOTHED EDGES AND SHARP HOOKED TIP WAS BRIGHT RED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7176/88083/7176-88083-0002.flac"}, "sample_rate": 16000, "source_samples": 120160, "source_duration_seconds": 7.51, "source_peak": 0.627410888671875, "source_rms_dbfs": -21.729688964800097, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.54323279506653, "decoded_pcm_sha256": "22dedd61331f17693761f502975e31bf01741aced72867d6ade55e38b2e09d33", "source_sha256": "3c042266b649c153544bdb4b6c17ac6faf27e9a2b589a3d673225c7ea91180cf", "wav_sha256": "31423e751d074a942799ab45f2654e651a29295a96fe071812f6e7002ee7d77b", "local_source": "audio/librispeech_7176-88083-0002.flac", "local_wav": "audio/librispeech_7176-88083-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7176-88083-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_7176", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7176-88083-0004", "transcript": "BUT SUDDENLY STRAIGHT AND SWIFT AS A DIVING CORMORANT HE SHOT DOWN INTO THE TORRENT AND DISAPPEARED BENEATH THE SURFACE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7176/88083/7176-88083-0004.flac"}, "sample_rate": 16000, "source_samples": 120000, "source_duration_seconds": 7.5, "source_peak": 0.60992431640625, "source_rms_dbfs": -21.15699985941999, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.87864531935906, "decoded_pcm_sha256": "43bbb6d389351da5d47cefcacfede3e848cb81497c672b951f8b533d7fbedfed", "source_sha256": "a0879f1d42b65a0bc10e3b725393bac049736d153e7a1c1b5d35e48bda4e3778", "wav_sha256": "f0dcd5085b4ebd5cc0be611f94aa76f39d4732e8894a1330b1b07532c0a56fd1", "local_source": "audio/librispeech_7176-88083-0004.flac", "local_wav": "audio/librispeech_7176-88083-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7176-88083-0026", "corpus": "librispeech_test_clean", "speaker": "librispeech_7176", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7176-88083-0026", "transcript": "THE DRAG UPON HIS BEAK AND THE LIGHT CHECK UPON HIS WINGS WERE INEXPLICABLE TO HIM AND APPALLING", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7176/88083/7176-88083-0026.flac"}, "sample_rate": 16000, "source_samples": 88480, "source_duration_seconds": 5.53, "source_peak": 0.62872314453125, "source_rms_dbfs": -22.34407403080596, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.34407403080596, "decoded_pcm_sha256": "bc9284b0dea002361e46bae8e07d7c2c20a9984ef0ce6f6e2c9fda86e70a579e", "source_sha256": "c637736d1dbd70e42d9ad3e319b273baf390cbc3f3ec8b31239ea7dca8e9d5d6", "wav_sha256": "50ee066cc5b8c5ac8d9942ca3a18ee3a2dfebe811603f1a35913c8be6b77fd0d", "local_source": "audio/librispeech_7176-88083-0026.flac", "local_wav": "audio/librispeech_7176-88083-0026.wav", "excerpt_start_sample": 0, "excerpt_samples": 88480}, {"id": "librispeech_7729-102255-0028", "corpus": "librispeech_test_clean", "speaker": "librispeech_7729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7729-102255-0028", "transcript": "PRIVATE PERSONS WHO HAD LEASED THE FREE STATE HOTEL VAINLY BESOUGHT THE VARIOUS AUTHORITIES TO PREVENT THE DESTRUCTION OF THEIR PROPERTY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7729/102255/7729-102255-0028.flac"}, "sample_rate": 16000, "source_samples": 153600, "source_duration_seconds": 9.6, "source_peak": 0.71356201171875, "source_rms_dbfs": -24.18336133917225, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.02753596641884, "decoded_pcm_sha256": "7a4ab1cb22b23d73175ff25f3f2557bc2724e5cd7a3952f1cc5626d2a9bfbd3f", "source_sha256": "4654ddc035bcd2a1c3def2c04709f8f93eff6f9d9da968a1f044f65edf3a5b15", "wav_sha256": "564baedf5982a21580b89a5281b3a5dcf3522711c4a8fa991f0f1b8e60b09b1c", "local_source": "audio/librispeech_7729-102255-0028.flac", "local_wav": "audio/librispeech_7729-102255-0028.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7729-102255-0038", "corpus": "librispeech_test_clean", "speaker": "librispeech_7729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7729-102255-0038", "transcript": "ATCHISON WHO HAD BEEN HARANGUING THE MOB PLANTED HIS TWO GUNS BEFORE THE BUILDING AND TRAINED THEM UPON IT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7729/102255/7729-102255-0038.flac"}, "sample_rate": 16000, "source_samples": 126720, "source_duration_seconds": 7.92, "source_peak": 0.547515869140625, "source_rms_dbfs": -23.868262038672686, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.65352971837625, "decoded_pcm_sha256": "4488be03955742e6e44fcdb33f0ae71643965b6765824a166c40ca31fbd9f66a", "source_sha256": "7bad471a79798adfc54e63acefeaf2a9579bc1f06779fd3aa49cc060df7c47f7", "wav_sha256": "b74c24750ddac32d162fd267caa8077dd1b1ce41e69840db61fc22a5aa5a4c04", "local_source": "audio/librispeech_7729-102255-0038.flac", "local_wav": "audio/librispeech_7729-102255-0038.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7729-102255-0042", "corpus": "librispeech_test_clean", "speaker": "librispeech_7729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7729-102255-0042", "transcript": "RELOCATED FOOTNOTE GOVERNOR ROBINSON BEING ON HIS WAY EAST THE STEAMBOAT ON WHICH HE WAS TRAVELING STOPPED AT LEXINGTON MISSOURI", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7729/102255/7729-102255-0042.flac"}, "sample_rate": 16000, "source_samples": 147600, "source_duration_seconds": 9.225, "source_peak": 0.69677734375, "source_rms_dbfs": -23.419483735767322, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.03270368618639, "decoded_pcm_sha256": "3e552debe1c0d524847e6550ec6f2d9550fc271ad1a9ac47e3543e154db1f552", "source_sha256": "258a5f83b294976c312b375ef48ed79028fb41459458f6ece40b90fab7bce317", "wav_sha256": "9b9665bfe15edaa592968b7b1014db2734b12dc98752aa4ad4d62e5499eac370", "local_source": "audio/librispeech_7729-102255-0042.flac", "local_wav": "audio/librispeech_7729-102255-0042.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8224-274381-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_8224", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8224-274381-0010", "transcript": "BY A QUICK AND UNEXPECTED MARCH MONTROSE HASTENED TO INNERLOCHY AND PRESENTED HIMSELF IN ORDER OF BATTLE BEFORE THE SURPRISED BUT NOT AFFRIGHTENED COVENANTERS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8224/274381/8224-274381-0010.flac"}, "sample_rate": 16000, "source_samples": 217920, "source_duration_seconds": 13.62, "source_peak": 0.44830322265625, "source_rms_dbfs": -23.72504999405364, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.951331441759642, "decoded_pcm_sha256": "34d9175c2a6517466fd56565f6c4abc5a8a75b16d60e1e96d0615d4383e19b19", "source_sha256": "e9f711f02a8f7cfe4527d6d16f4ad04dc95369de860a6c4f56a645dc7da5a02b", "wav_sha256": "18c326a6f8dc3490bccdd4d2cb16af9847410c33ce1a4fcf91c3002db72c8aae", "local_source": "audio/librispeech_8224-274381-0010.flac", "local_wav": "audio/librispeech_8224-274381-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8224-274381-0013", "corpus": "librispeech_test_clean", "speaker": "librispeech_8224", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8224-274381-0013", "transcript": "FROM THE SAME MEN NEW REGIMENTS AND NEW COMPANIES WERE FORMED DIFFERENT OFFICERS APPOINTED AND THE WHOLE MILITARY FORCE PUT INTO SUCH HANDS AS THE INDEPENDENTS COULD RELY ON", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8224/274381/8224-274381-0013.flac"}, "sample_rate": 16000, "source_samples": 217760, "source_duration_seconds": 13.61, "source_peak": 0.536865234375, "source_rms_dbfs": -21.803141664293197, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.810391566928214, "decoded_pcm_sha256": "31d48338892252220a9e3042986f534b91479b75a07438f2a0515b00bc8fe055", "source_sha256": "cd0ba8a197cc480b016a6386531b3e1ee14c9ff6ddc85f19abde2abc4f00070a", "wav_sha256": "a6a7de02099102496e51606fd6032ef6e7d8800ebd1ab894b29e2fe6066d3816", "local_source": "audio/librispeech_8224-274381-0013.flac", "local_wav": "audio/librispeech_8224-274381-0013.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8224-274384-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_8224", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8224-274384-0010", "transcript": "BEFORE THE SETTLEMENT OF TERMS THE ADMINISTRATION MUST BE POSSESSED ENTIRELY BY THE PARLIAMENTS OF BOTH KINGDOMS AND HOW INCOMPATIBLE THAT SCHEME WITH THE LIBERTY OF THE KING IS EASILY IMAGINED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8224/274384/8224-274384-0010.flac"}, "sample_rate": 16000, "source_samples": 233600, "source_duration_seconds": 14.6, "source_peak": 0.393951416015625, "source_rms_dbfs": -25.11362426001846, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.152240114273017, "decoded_pcm_sha256": "bfff7d3cf9d00e04b5e5f471ec3e7e23254f82646da0069c2f8d619c084a00f4", "source_sha256": "677a2c48f5ddd5929e5c68e682b66998c48322833dac3baa41a086f87ef15077", "wav_sha256": "5d22068dd2ce8f40fce798da990cdd48504bfbf705b82b996eeab016ef4c4c97", "local_source": "audio/librispeech_8224-274384-0010.flac", "local_wav": "audio/librispeech_8224-274384-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8230-279154-0018", "corpus": "librispeech_test_clean", "speaker": "librispeech_8230", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8230-279154-0018", "transcript": "THERE IS OF COURSE A DIFFERENCE BETWEEN KNOWING THE TEMPORAL RELATION OF A REMEMBERED EVENT TO THE PRESENT AND KNOWING THE TIME ORDER OF TWO REMEMBERED EVENTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8230/279154/8230-279154-0018.flac"}, "sample_rate": 16000, "source_samples": 189520, "source_duration_seconds": 11.845, "source_peak": 0.658111572265625, "source_rms_dbfs": -23.673484842402758, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.85204146467471, "decoded_pcm_sha256": "2becab040601a06405a880e563913b87f3eb167e942e97f7225c55643949f142", "source_sha256": "6fb6d05dd68dfe641a0f58b800e9d2e86662dab952355bc7573505bc35efb2fc", "wav_sha256": "7683ffe730b9a0eb95d05f98aba9d11f6ad6fa68a80ab310dc223d779cd0921a", "local_source": "audio/librispeech_8230-279154-0018.flac", "local_wav": "audio/librispeech_8230-279154-0018.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8230-279154-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_8230", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8230-279154-0032", "transcript": "IT IS THIS THAT IS OF INTEREST TO THEORY OF KNOWLEDGE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8230/279154/8230-279154-0032.flac"}, "sample_rate": 16000, "source_samples": 62080, "source_duration_seconds": 3.88, "source_peak": 0.763580322265625, "source_rms_dbfs": -23.179606504911124, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.179606504911124, "decoded_pcm_sha256": "d59c4c75b338357fc4c79c9980c56076c2ae54441e0d223325fd8514435d4576", "source_sha256": "d60da130333b0e1adba450d171b7f753f423dce0f66a00448dfb036d43a7231d", "wav_sha256": "46983a79c683e36b2b6ffd2611bd41f526680b8d83fa351d5b1ff14283d23e2e", "local_source": "audio/librispeech_8230-279154-0032.flac", "local_wav": "audio/librispeech_8230-279154-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 62080}, {"id": "librispeech_8230-279154-0033", "corpus": "librispeech_test_clean", "speaker": "librispeech_8230", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8230-279154-0033", "transcript": "IT IS BY NO MEANS ALWAYS RELIABLE ALMOST EVERYBODY HAS AT SOME TIME EXPERIENCED THE WELL KNOWN ILLUSION THAT ALL THAT IS HAPPENING NOW HAPPENED BEFORE AT SOME TIME", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8230/279154/8230-279154-0033.flac"}, "sample_rate": 16000, "source_samples": 187040, "source_duration_seconds": 11.69, "source_peak": 0.857177734375, "source_rms_dbfs": -21.919399299890486, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.434707149884012, "decoded_pcm_sha256": "bd287f891d96a4cd6b7a04337bf6be93b04db8cbec30940ae4945f8b4efde0f4", "source_sha256": "1bcf539698173ec1996fa39e5b4c5faf7a8ddf5b5f9636d269ca583befc17eb6", "wav_sha256": "4d90673ae367514c97b64216ae1e3fedf844226a86b71986ec389ed219babf2f", "local_source": "audio/librispeech_8230-279154-0033.flac", "local_wav": "audio/librispeech_8230-279154-0033.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8455-210777-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_8455", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8455-210777-0010", "transcript": "THEIR MASTERS SAID MISSUS NEVERBEND", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8455/210777/8455-210777-0010.flac"}, "sample_rate": 16000, "source_samples": 49600, "source_duration_seconds": 3.1, "source_peak": 0.97686767578125, "source_rms_dbfs": -24.180707212674086, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.180707212674086, "decoded_pcm_sha256": "10ec70ae1221ff8684ca0b1b6aa22fd7ae9343c0b701f90c315735ff27f3d5ad", "source_sha256": "28029242eb37055c0eb70761eebf5b7be98a5da6c30127b3b282bf7e11663a68", "wav_sha256": "c7e846edf284a6ab491b525b5058af16b5a3f2b4858d3189ffdb3c297723e488", "local_source": "audio/librispeech_8455-210777-0010.flac", "local_wav": "audio/librispeech_8455-210777-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 49600}, {"id": "librispeech_8455-210777-0049", "corpus": "librispeech_test_clean", "speaker": "librispeech_8455", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8455-210777-0049", "transcript": "LIEUTENANT CROSSTREES IS A VERY GALLANT OFFICER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8455/210777/8455-210777-0049.flac"}, "sample_rate": 16000, "source_samples": 65760, "source_duration_seconds": 4.11, "source_peak": 0.46630859375, "source_rms_dbfs": -24.60742082061149, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.60742082061149, "decoded_pcm_sha256": "6bab2226481307e41ffb714c528db253ee0755e0775a2890f32c07e3f5832d55", "source_sha256": "2c4c3f4fa85e1e234b108aeada818530820e7a515dc7d91f2a2c075af7105fed", "wav_sha256": "f86b45c1d7a339929b918658a770ad680ab2167df6561f4cc3f2581743927ba9", "local_source": "audio/librispeech_8455-210777-0049.flac", "local_wav": "audio/librispeech_8455-210777-0049.wav", "excerpt_start_sample": 0, "excerpt_samples": 65760}, {"id": "librispeech_8455-210777-0060", "corpus": "librispeech_test_clean", "speaker": "librispeech_8455", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8455-210777-0060", "transcript": "THE JOHN BRIGHT IS ARMED WITH A WEAPON OF GREAT POWER AGAINST WHICH IT IS IMPOSSIBLE THAT THE PEOPLE OF BRITANNULA SHOULD PREVAIL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8455/210777/8455-210777-0060.flac"}, "sample_rate": 16000, "source_samples": 113200, "source_duration_seconds": 7.075, "source_peak": 0.728973388671875, "source_rms_dbfs": -21.874934111772994, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.328040515972575, "decoded_pcm_sha256": "71c64b069cc8f43d09cfadd1defcda3beccfdf73919c8518a13182fa2c801f93", "source_sha256": "ef57db07976e02027d19da092bac4747c7928a93199441010b967c12ce13e724", "wav_sha256": "879b2509fba33c8e73d65f576d33bc56edca7a2ee0441e2e0485f96aabf7a688", "local_source": "audio/librispeech_8455-210777-0060.flac", "local_wav": "audio/librispeech_8455-210777-0060.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8463-294828-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_8463", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8463-294828-0003", "transcript": "I WANTED NOTHING MORE THAN TO SEE MY COUNTRY AGAIN MY FRIENDS MY MODEST QUARTERS BY THE BOTANICAL GARDENS MY DEARLY BELOVED COLLECTIONS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8463/294828/8463-294828-0003.flac"}, "sample_rate": 16000, "source_samples": 149440, "source_duration_seconds": 9.34, "source_peak": 0.445587158203125, "source_rms_dbfs": -24.97531570503334, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.82912392962218, "decoded_pcm_sha256": "cd223f40e942ab9ac205813cbe938225779742060f5fe247eb3e2fd12df52eba", "source_sha256": "3790b43d50403978fd6ba4d90f598e8e6afddfb630a3bf5b4110bd432517f1e9", "wav_sha256": "03177b8bf19007f5f7477015e45cc382bce9b710fdea1134a0efc765c379e895", "local_source": "audio/librispeech_8463-294828-0003.flac", "local_wav": "audio/librispeech_8463-294828-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8463-294828-0014", "corpus": "librispeech_test_clean", "speaker": "librispeech_8463", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8463-294828-0014", "transcript": "THERE WAS GOOD REASON TO STOP AND THINK EVEN FOR THE WORLD'S MOST EMOTIONLESS MAN", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8463/294828/8463-294828-0014.flac"}, "sample_rate": 16000, "source_samples": 91600, "source_duration_seconds": 5.725, "source_peak": 0.51129150390625, "source_rms_dbfs": -23.96932327924813, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.96932327924813, "decoded_pcm_sha256": "c608aedb889e9a3c3821520a449396c5b4cd1264e1d081a38be40b0b66f8b285", "source_sha256": "b61f11aa0c2b3d1adca49e6199328637c6ecc84c003dab03fc8c97b95cab4c63", "wav_sha256": "87e86c5001f0bc4f6cc029d8e282bfd78ad4fef425a55309bbfc646a7744e707", "local_source": "audio/librispeech_8463-294828-0014.flac", "local_wav": "audio/librispeech_8463-294828-0014.wav", "excerpt_start_sample": 0, "excerpt_samples": 91600}, {"id": "librispeech_8463-294828-0022", "corpus": "librispeech_test_clean", "speaker": "librispeech_8463", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8463-294828-0022", "transcript": "WE'RE LEAVING ON THE ABRAHAM LINCOLN", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8463/294828/8463-294828-0022.flac"}, "sample_rate": 16000, "source_samples": 37680, "source_duration_seconds": 2.355, "source_peak": 0.42718505859375, "source_rms_dbfs": -24.863923522138165, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.863923522138165, "decoded_pcm_sha256": "276e5d5f0e0b54af2b651a409e0f7fbcd7b42508e69037a0becabc956b358b07", "source_sha256": "4b0b7f62860ac9f04945b08f79183b52c26f80eedd14ff829e2ff2aaa5143e3f", "wav_sha256": "76e6aeb80f67faf23c266b54112340687389f0475ffad207ed027383328b49b4", "local_source": "audio/librispeech_8463-294828-0022.flac", "local_wav": "audio/librispeech_8463-294828-0022.wav", "excerpt_start_sample": 0, "excerpt_samples": 37680}, {"id": "librispeech_8555-284447-0007", "corpus": "librispeech_test_clean", "speaker": "librispeech_8555", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8555-284447-0007", "transcript": "THEREFORE HER MAJESTY PAID NO ATTENTION TO ANYONE AND NO ONE PAID ANY ATTENTION TO HER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8555/284447/8555-284447-0007.flac"}, "sample_rate": 16000, "source_samples": 101840, "source_duration_seconds": 6.365, "source_peak": 0.24127197265625, "source_rms_dbfs": -26.869887535083663, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.613422172170882, "decoded_pcm_sha256": "b2e9b095c084b8ff6903b45eed724068bad7267cf8c1f692479d68f2b6adbaed", "source_sha256": "73d10187e60efe5f030a76fa33d1eab2abae7e38817b356dd61e30b743edf22f", "wav_sha256": "cc3409e20c1284fb07dc358b49b0faf9bb0cc3d7d6726d4073f8961fc15b3155", "local_source": "audio/librispeech_8555-284447-0007.flac", "local_wav": "audio/librispeech_8555-284447-0007.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8555-284449-0020", "corpus": "librispeech_test_clean", "speaker": "librispeech_8555", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8555-284449-0020", "transcript": "THE COMBINED BANDS OF BOTH THE COUNTRIES PLAYED THE MUSIC AND A FINE SUPPER WAS SERVED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8555/284449/8555-284449-0020.flac"}, "sample_rate": 16000, "source_samples": 81520, "source_duration_seconds": 5.095, "source_peak": 0.29266357421875, "source_rms_dbfs": -27.14678667416795, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.14678667416795, "decoded_pcm_sha256": "fea5f67e0d192edbb42257fa9560e233f643cc196059371f1de72f17987367e7", "source_sha256": "5703c01afc67c6eb161bd9fe7fe92f5b373e9331629f60d1c163e5d893c97d7a", "wav_sha256": "a033b75742b985d12a5c13f94f380d6d7440bd6723b0faabac5379e81d1fec62", "local_source": "audio/librispeech_8555-284449-0020.flac", "local_wav": "audio/librispeech_8555-284449-0020.wav", "excerpt_start_sample": 0, "excerpt_samples": 81520}, {"id": "librispeech_8555-292519-0001", "corpus": "librispeech_test_clean", "speaker": "librispeech_8555", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8555-292519-0001", "transcript": "GUIDED BY YOU HOW WE MIGHT STROLL TOWARDS DEATH OUR ONLY MUSIC ONE ANOTHER'S BREATH THROUGH GARDENS INTIMATE WITH HOLLYHOCKS WHERE SILENT POPPIES BURN BETWEEN THE ROCKS BY POOLS WHERE BIRCHES BEND TO CONFIDANTS ABOVE GREEN WATERS SCUMMED WITH LILY PLANTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8555/292519/8555-292519-0001.flac"}, "sample_rate": 16000, "source_samples": 333120, "source_duration_seconds": 20.82, "source_peak": 0.60888671875, "source_rms_dbfs": -23.415033537556127, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.624461889815606, "decoded_pcm_sha256": "c2e62e46b5a18b9f8f30423189b31cc00703ba64084bb8bca7ffe75ae8d6977a", "source_sha256": "b082de228dbad73be95d916e5149c02035758a390482845def17215c0eed0fa9", "wav_sha256": "9f14d64bcc89c118ff39930d39a3240d88be1fe829cbbb61a05786867b5b1be1", "local_source": "audio/librispeech_8555-292519-0001.flac", "local_wav": "audio/librispeech_8555-292519-0001.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_908-157963-0018", "corpus": "librispeech_test_clean", "speaker": "librispeech_908", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "908-157963-0018", "transcript": "AND FEAREST THOU BECAUSE I VANISH AND AM SEEN NO MORE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/908/157963/908-157963-0018.flac"}, "sample_rate": 16000, "source_samples": 68080, "source_duration_seconds": 4.255, "source_peak": 0.513336181640625, "source_rms_dbfs": -23.538819799676702, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.538819799676702, "decoded_pcm_sha256": "8ae8f9234bac8c63b967d3c7b5075657f9771ac253a59a383e523ecff9c3b006", "source_sha256": "53efff3282f4691482a4e735887fab9130c78ec10f3bdb79bf68f8c212898134", "wav_sha256": "aeae977ce030a5c4dc6958737ff58b35c245d4fdadf9685b6e94dacfdcce7da4", "local_source": "audio/librispeech_908-157963-0018.flac", "local_wav": "audio/librispeech_908-157963-0018.wav", "excerpt_start_sample": 0, "excerpt_samples": 68080}, {"id": "librispeech_908-31957-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_908", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "908-31957-0002", "transcript": "I DID NOT WRONG MYSELF SO BUT I PLACED A WRONG ON THEE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/908/31957/908-31957-0002.flac"}, "sample_rate": 16000, "source_samples": 76640, "source_duration_seconds": 4.79, "source_peak": 0.42138671875, "source_rms_dbfs": -25.537578974520407, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.537578974520407, "decoded_pcm_sha256": "5e0a5e522c82eea67f456aec58b62300d4c743b9610bb9f4f869f3b8bd8e7f17", "source_sha256": "77ad707e976d262395c265cdfda5b9a3da7439f329aac0f5216a404acefe3fe7", "wav_sha256": "88f78a25a0c7082384d7a9653ba5e656504ddc299b790c6bd9d4a11b5119d742", "local_source": "audio/librispeech_908-31957-0002.flac", "local_wav": "audio/librispeech_908-31957-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 76640}, {"id": "librispeech_908-31957-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_908", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "908-31957-0016", "transcript": "DEAREST TEACH ME SO TO POUR OUT GRATITUDE AS THOU DOST GOOD", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/908/31957/908-31957-0016.flac"}, "sample_rate": 16000, "source_samples": 103680, "source_duration_seconds": 6.48, "source_peak": 0.563873291015625, "source_rms_dbfs": -24.06459814668036, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.73060181736992, "decoded_pcm_sha256": "a1a0fc0c8f705774ce9810918acda74ca07e00f2107a855ab611aa7ee1735b94", "source_sha256": "2a57896e33e6a87debcb741f00196495aa665bb2f0d7c205ac980a8417976d12", "wav_sha256": "fccef19835dc4eb3cde5b4e85b177f7b4375f6ef7db090434ea7ab00d5b00ca9", "local_source": "audio/librispeech_908-31957-0016.flac", "local_wav": "audio/librispeech_908-31957-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}], "criteria_sha256": "19c16dbc3ed6478cd79cbc8b04e99e6fee6a693f0ba9b36e95d944ec710b1a9d", "lock_sha256": "c8b2bc28ebb1523bf8cfe0f4627d06bd5d03e0ba6a2bb5dc0905be77a3dc8823", "python": "3.12.14", "platform": "Windows-11-10.0.26200-SP0", "dependencies": {"numpy": "2.3.5", "scipy": "1.17.0", "pystoi": "0.4.1", "torch": "2.8.0+cpu", "torchaudio": "2.8.0+cpu", "torchvision": "0.23.0+cpu", "transformers": "4.51.3", "timm": "1.0.15", "librosa": "0.11.0", "utmosv2": "1.3.1.dev0"}, "methods": ["hnm", "A", "B", "C", "D", "E"], "noise_seeds": [0, 1, 2, 3, 4, 5, 6, 7], "mos_noise_seed": 0, "mos_extraction_seed": 20261012, "mos_repetitions": 1, "mos_threads": 1, "workers": 2, "threads_per_worker": 1, "alignment": "none", "normalization": "none", "analysis": "single_current_base_encode_per_recording", "world": "not_run; historical_reference_only", "role": "candidate_screening_not_automatic_adoption"}, "acoustic_contract": {"schema": "noise-screen-v1", "formal_base": {"status": "selected", "method": "hnm", "version": "hnm-v2", "reason": "固定120録音で音量変動の再現誤差が平均約7.2%減少。振幅超過の頻度・程度も踏まえて事後に採否を見直し、雑音包絡を分離する実装を採用。日本語12録音の補助評価も実施。予測MOSや知覚品質の改善を保証するものではない。", "comparison_reference": "hnm", "comparison_reference_meaning": "実験開始時点の唯一の自作ベース。過去結果は当時の版を保持する。既存比較対象はWORLD（Harvest）。", "previous_version": "hnm-v1", "selection_report": "experiments/2026-10-12/jvs-supplement/results/run-001/index.html"}, "baseline_sha256": "ca2f7f56fe4bbf4c3d987c90ac81cd90cf394e4b636ef36988bdf34f220a7153", "sources": {"audio_report.py": "f46ff0cf69e7cb956ea94ee0c7013d44052c0400c5ffc64d7f9f90b4489aa980", "audio_samples.py": "b9b975c1f305cdaff47909b99bd2a3dce64996e90774d2a8305adc1c67ed6c17", "benchmark.py": "10a84680e1641fe5922b741ce05b6538b43b98be9daf27245661fe80bab732ba", "benchmark_io.py": "b27cca4d9457de305be7bbbd84aedf9b0d097791a933e2e9af55eb67e1edbb3d", "benchmark_pipeline.py": "a049034dcd272bbafda47cbf5719cc6b1cd8dae6f80682200cd3fc7b90e522cd", "benchmark_report.py": "c1a8153022a620bc62095f30c5709a34e217739cb9473860dc5b3daff5b87c88", "benchmark_summary.py": "8221086fe08bfc70cbfb9308a3a01d6fea976afb7de86af8cbacd7999e8281da", "build_world.py": "7d29d81c9a48d633edfeae0425c5cc055bc0a10d230232a51eab5720b2cc6974", "calibrate_pitch.py": "282c917bf87c6f549f32467647802852907e4b0b88128eb8017157b84566adec", "control.py": "bc41e73ff34b5f9cb52280a6f2317c989abc53f8b15f8f82db5ee3102b4b58bd", "corpus.py": "bc14cee2c3dd1a651cf2e5f0e9fe871fbd3a712d1a90d7c18a18547a625802e8", "extended_metrics.py": "ab3537d2c8e5d568a0cd2fa8aa36309de09c3bce7cf2f849214331a22fe3624f", "fixtures.py": "a1646c4d0f7c6966ea49a4673ec2c71e75f864943c8908395089d027325eac69", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_diagnostics.py": "a3f47468d10db7e8040c78fc24546043b2eeae6b518f59da1f5904c674aae436", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "937721cd363f06afa3bf4c20c1ff182a8c337c4b19ac9f4b57314f6adea6b44c", "logging_utils.py": "c80eb5d9647516a4bc03980cb178f8b4bb130ac15877d1b3289797e7ad81e381", "methods.py": "b675f7bddf6d19515609713de11fcf4bbdf4c54a3a536030d8da16c3282dd97c", "metrics.py": "b8767feb49494058bd9ec08fd95ae2fb50ce7897fcdd6cebe319ff8a0ac51f22", "mos.py": "fb632aa5440245525289962372db4f3eb007cd2b3b493b2d05ecccb878e85f03", "mos_batch.py": "c7a732dfe52b08fa155201a7134d2f5561a10b16d1866e9f0526f61b2b244f64", "noise_screen.py": "ada7ed38f5c67b7892f1a18af226d738400934607dd4adb9e403b5ca7bdbf667", "noise_stability.py": "624dc434add0086c53c41b9fb7fd0e65b2c51b16c08d4485e4147a77f5534213", "report.py": "cefe7b865cbf768cc497c2fbf8085f0492875abab89dce34ff99831deb180bce", "run.py": "8b2713600f3890dae76d4b2b1b0ef992fdb621679bff3aecc73ca9b68d92e9d2", "site_summary.py": "9c1b03304cc94162deb95bbd465b05c727fa825aed0d1d4ad43c3febc3d76ad7", "supplement.py": "8b72b63b7059f3e94ecb13f575ab93060e00fee0cb89c5abcaa51719061b890d", "supplement_contract.py": "748bff65db3bd3709cdd472b02358af35518c5d1a96fbf212f6b02ba5fa45cd4", "supplement_report.py": "378abee381348bbe0e833d710b899c66d134fd78834e5afbd7a3a71ecdde510f", "temporal_metrics.py": "ce21b6a4767071a7b98ea0cdbfe54bdba15b596bf8d6b6127281a97095e7be91", "world.py": "d14558eed772e07031a9183f57facc55acb1c27deccb3dd4fa098ffcbe0cde3a", "__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d"}, "candidate_sources": {"A": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "2168d3b18d284b2aeac1f57ef4ab847fb4c14bf64fdd1e84b2c72597c790cbd2"}, "B": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "e2478a1afd758cced692e05b83fa88443184d63406af4b5a1985cf8e3b86758b"}, "C": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "640a1f8ae71a296357162c2bbf941bd39465946982c67f7bdb85bdfb1cd8419c"}, "D": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "a4fda7d4cfcaff18f49862b09f5c289d3947a8517f4885b35b281102d953ef0c"}, "E": {"__init__.py": "73738d70706824f85fdfac302306af505437a50b0ca4cdc68fd57aec211a308d", "hnm.py": "58f0378793b7433f80bf854972882298649799d8e3d0868674816e9bf6b63f74", "hnm_analysis.py": "3f45703667237333d76c4145c1513e4eb3342bd40f01a714c10303082440e26a", "hnm_control.py": "b2f5efbbb1e550569c231a08f12997d94b0bacb130bcc816fe6bd458b17a98bc", "hnm_parameters.py": "635bf222bac8ce41897c0977677e67be4366724f4239a9e85e153169fcea2cf0", "hnm_synthesis.py": "96d7a9ae6af173785b8321ef4dbfc9d19b3d643d538ef074f6adca4357306ee5"}}, "manifest_sha256": "644f4d57483366bb20d8dbcfdaa254e354aed4af8318a6a2984bb2e8b3ddc9c1", "input_records": [{"id": "librispeech_1089-134686-0008", "corpus": "librispeech_test_clean", "speaker": "librispeech_1089", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1089-134686-0008", "transcript": "THE CHAOS IN WHICH HIS ARDOUR EXTINGUISHED ITSELF WAS A COLD INDIFFERENT KNOWLEDGE OF HIMSELF", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1089/134686/1089-134686-0008.flac"}, "sample_rate": 16000, "source_samples": 107680, "source_duration_seconds": 6.73, "source_peak": 0.45428466796875, "source_rms_dbfs": -27.667836551289334, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.30287974095976, "decoded_pcm_sha256": "40f044b90ed06d7cffc4145524ea3e1b8bd36b6599e24e1e905262fbb3e50d8d", "source_sha256": "1bdf7eede41ca2cc8883defb66bbd2e59ec93bf2d587327cb051d0ae427495ae", "wav_sha256": "17a52b57314c689b9e8d2dccefec59b6b2777d0f240fd26d5e940b551cf97c95", "local_source": "audio/librispeech_1089-134686-0008.flac", "local_wav": "audio/librispeech_1089-134686-0008.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1089-134686-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_1089", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1089-134686-0011", "transcript": "ON SATURDAY MORNINGS WHEN THE SODALITY MET IN THE CHAPEL TO RECITE THE LITTLE OFFICE HIS PLACE WAS A CUSHIONED KNEELING DESK AT THE RIGHT OF THE ALTAR FROM WHICH HE LED HIS WING OF BOYS THROUGH THE RESPONSES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1089/134686/1089-134686-0011.flac"}, "sample_rate": 16000, "source_samples": 199120, "source_duration_seconds": 12.445, "source_peak": 0.5145263671875, "source_rms_dbfs": -27.02531260265946, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.932448099275575, "decoded_pcm_sha256": "3db35e4a6ed6a12218e0ab1bfe72f1810bad8f201648ec0e2a9af7dae7aa32eb", "source_sha256": "c65e16be3ef92c11a31a9c2c4c570542eb2ed66153150a7c4cca00c01b30a4e2", "wav_sha256": "f7033475e19dc33371d8718f1495a3ec181845217fe2434773cf6c766ae39dc5", "local_source": "audio/librispeech_1089-134686-0011.flac", "local_wav": "audio/librispeech_1089-134686-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1089-134686-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_1089", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1089-134686-0032", "transcript": "HE IS CALLED AS YOU KNOW THE APOSTLE OF THE INDIES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1089/134686/1089-134686-0032.flac"}, "sample_rate": 16000, "source_samples": 65440, "source_duration_seconds": 4.09, "source_peak": 0.43682861328125, "source_rms_dbfs": -26.667394249066906, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.667394249066906, "decoded_pcm_sha256": "643f787b905d5160255daf650b50e8a272ea88d166a6e7aca2a7ee357cf232fb", "source_sha256": "ba1d2185ad72cd3142e6309865859a8e9e5dfdc91ab046adfcc07338c5a0f0f0", "wav_sha256": "520cdcc0a8dc585f41bbeb35f00cb5ea8279a28f7472c297a5ef1f0757880352", "local_source": "audio/librispeech_1089-134686-0032.flac", "local_wav": "audio/librispeech_1089-134686-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 65440}, {"id": "librispeech_1188-133604-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_1188", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1188-133604-0003", "transcript": "MY FIRST AND PRINCIPAL REASON WAS THAT THEY ENFORCED BEYOND ALL RESISTANCE ON ANY STUDENT WHO MIGHT ATTEMPT TO COPY THEM THIS METHOD OF LAYING PORTIONS OF DISTINCT HUE SIDE BY SIDE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1188/133604/1188-133604-0003.flac"}, "sample_rate": 16000, "source_samples": 201760, "source_duration_seconds": 12.61, "source_peak": 0.29107666015625, "source_rms_dbfs": -29.84254246966504, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.312803528459913, "decoded_pcm_sha256": "81283c3eb08ea201c5ec2d52e3a8e0ed4d0258d76164b522d1ee5ec43edebe81", "source_sha256": "a013400c991a40062cb4f458d247cfd6dac24fa0016bf6e2eac7a3e8e08b8ae8", "wav_sha256": "64ef54500e885de74afaed253f49333ec5284415ca720897ca3b5cded31c617e", "local_source": "audio/librispeech_1188-133604-0003.flac", "local_wav": "audio/librispeech_1188-133604-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1188-133604-0031", "corpus": "librispeech_test_clean", "speaker": "librispeech_1188", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1188-133604-0031", "transcript": "THERE'S ONE AND THERE'S ANOTHER THE DUDLEY AND THE FLINT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1188/133604/1188-133604-0031.flac"}, "sample_rate": 16000, "source_samples": 68000, "source_duration_seconds": 4.25, "source_peak": 0.299102783203125, "source_rms_dbfs": -26.69687768186568, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.69687768186568, "decoded_pcm_sha256": "b739cb4e3f2719b98dd98461d6859b487da84f7d1b9f40d280b7f0eb538a6458", "source_sha256": "08584f8b87cd859e9212f1730979ed141623f35e38995a88ba50747d04cdc222", "wav_sha256": "94599f5aded76bb1de8d31d5c918c6c759fd0febba014767c838658f4127d63e", "local_source": "audio/librispeech_1188-133604-0031.flac", "local_wav": "audio/librispeech_1188-133604-0031.wav", "excerpt_start_sample": 0, "excerpt_samples": 68000}, {"id": "librispeech_1188-133604-0037", "corpus": "librispeech_test_clean", "speaker": "librispeech_1188", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1188-133604-0037", "transcript": "EVERY LINE IN WHICH THE MASTER TRACES IT EVEN WHERE SEEMINGLY NEGLIGENT IS LOVELY AND SET DOWN WITH A MEDITATIVE CALMNESS WHICH MAKES THESE TWO ETCHINGS CAPABLE OF BEING PLACED BESIDE THE MOST TRANQUIL WORK OF HOLBEIN OR DUERER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1188/133604/1188-133604-0037.flac"}, "sample_rate": 16000, "source_samples": 232160, "source_duration_seconds": 14.51, "source_peak": 0.42877197265625, "source_rms_dbfs": -26.333011899815872, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.701166289573592, "decoded_pcm_sha256": "55ff59cb08cac135da664331bc418f7f430ca702cefbb6986e7121d9fe94ca8b", "source_sha256": "94c0626f16e39947864688eb2721a57f838b43caeb1857767a9b4381807368c8", "wav_sha256": "f05a798d7d0c465a9fbd5936dce78da714f9df3c37fdcb61a19556c5153a0050", "local_source": "audio/librispeech_1188-133604-0037.flac", "local_wav": "audio/librispeech_1188-133604-0037.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_121-121726-0012", "corpus": "librispeech_test_clean", "speaker": "librispeech_121", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "121-121726-0012", "transcript": "HUSSY WOMAN AND BOND TIE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/121/121726/121-121726-0012.flac"}, "sample_rate": 16000, "source_samples": 64720, "source_duration_seconds": 4.045, "source_peak": 0.755340576171875, "source_rms_dbfs": -24.740950707596877, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.740950707596877, "decoded_pcm_sha256": "9c9a8032f3e173a5a354dda74e6cb24f7034610d2cd1b6d15dff6a794afd035e", "source_sha256": "6d1fab44fd2e6930a69e1f6f7e52a141ed5290b9a45f56df5e0c3177fa05c6ba", "wav_sha256": "e6e9fea185053d1e5f5604b3e61bf6a51d50cae5fcc82a4e5e7e14838ced2a4d", "local_source": "audio/librispeech_121-121726-0012.flac", "local_wav": "audio/librispeech_121-121726-0012.wav", "excerpt_start_sample": 0, "excerpt_samples": 64720}, {"id": "librispeech_121-123852-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_121", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "121-123852-0000", "transcript": "THOSE PRETTY WRONGS THAT LIBERTY COMMITS WHEN I AM SOMETIME ABSENT FROM THY HEART THY BEAUTY AND THY YEARS FULL WELL BEFITS FOR STILL TEMPTATION FOLLOWS WHERE THOU ART", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/121/123852/121-123852-0000.flac"}, "sample_rate": 16000, "source_samples": 283120, "source_duration_seconds": 17.695, "source_peak": 0.656524658203125, "source_rms_dbfs": -26.637611816594372, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.994738916713484, "decoded_pcm_sha256": "085897f5f5cc183ef4daa8868cb997e49f924fa58877dcfe167b6419325ce812", "source_sha256": "58d6350b52675d6d19616bc0347afc52a1535f8ea8be82aabb7275e6919a2598", "wav_sha256": "feffeb168d26185309000ded59165b27456151ac9fc91fc180467e41eab7e4ee", "local_source": "audio/librispeech_121-123852-0000.flac", "local_wav": "audio/librispeech_121-123852-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_121-127105-0035", "corpus": "librispeech_test_clean", "speaker": "librispeech_121", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "121-127105-0035", "transcript": "SHE PROMISED TO DO THIS AND SHE MENTIONED TO ME THAT WHEN FOR A MOMENT DISBURDENED DELIGHTED HE HELD HER HAND THANKING HER FOR THE SACRIFICE SHE ALREADY FELT REWARDED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/121/127105/121-127105-0035.flac"}, "sample_rate": 16000, "source_samples": 226400, "source_duration_seconds": 14.15, "source_peak": 0.419677734375, "source_rms_dbfs": -29.41904633729958, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.828167012475838, "decoded_pcm_sha256": "48306d9610578521a036cf3d52224eb0076f9d14ddb6850ac09278e898fc7303", "source_sha256": "7e293a1a5f64d336563f850f69929dd6b112e3ce37913f4422f2c29e9c55042d", "wav_sha256": "471c8af18dbbc4b37815971c9bd8c0e583a204a80ab851f2ea450c08140fd69f", "local_source": "audio/librispeech_121-127105-0035.flac", "local_wav": "audio/librispeech_121-127105-0035.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1221-135766-0005", "corpus": "librispeech_test_clean", "speaker": "librispeech_1221", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1221-135766-0005", "transcript": "HESTER COULD ONLY ACCOUNT FOR THE CHILD'S CHARACTER AND EVEN THEN MOST VAGUELY AND IMPERFECTLY BY RECALLING WHAT SHE HERSELF HAD BEEN DURING THAT MOMENTOUS PERIOD WHILE PEARL WAS IMBIBING HER SOUL FROM THE SPIRITUAL WORLD AND HER BODILY FRAME FROM ITS MATERIAL OF EARTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1221/135766/1221-135766-0005.flac"}, "sample_rate": 16000, "source_samples": 266320, "source_duration_seconds": 16.645, "source_peak": 0.37432861328125, "source_rms_dbfs": -29.831601026934116, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.793345815556208, "decoded_pcm_sha256": "ae61e68356f4a720a3d9f729ddcb9fa3e695cdbc220194eb938e92185005daf4", "source_sha256": "57ad0fad287d98afb900ccaa898a6e75348b80a3388201cd3e1feee0a0afc6b1", "wav_sha256": "7795716b196680446728c19c7a356f14b470482cf25a7298ec2b64caba670fd6", "local_source": "audio/librispeech_1221-135766-0005.flac", "local_wav": "audio/librispeech_1221-135766-0005.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1221-135767-0006", "corpus": "librispeech_test_clean", "speaker": "librispeech_1221", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1221-135767-0006", "transcript": "THE MOTHER HERSELF AS IF THE RED IGNOMINY WERE SO DEEPLY SCORCHED INTO HER BRAIN THAT ALL HER CONCEPTIONS ASSUMED ITS FORM HAD CAREFULLY WROUGHT OUT THE SIMILITUDE LAVISHING MANY HOURS OF MORBID INGENUITY TO CREATE AN ANALOGY BETWEEN THE OBJECT OF HER AFFECTION AND THE EMBLEM OF HER GUILT AND TORTURE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1221/135767/1221-135767-0006.flac"}, "sample_rate": 16000, "source_samples": 328960, "source_duration_seconds": 20.56, "source_peak": 0.466552734375, "source_rms_dbfs": -29.808508935576647, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.67998387340972, "decoded_pcm_sha256": "1bf36f00ad8a7e2e9cd6b41efd7f023aab063334b9ad526bda961251493c65cb", "source_sha256": "d70493003bb4a6833163a79d2df440c45bad786d8b50b5c194b896f5a4afd905", "wav_sha256": "135604e3b27df5b5dc13e00735ce8fe6b63247ccc1698c3d65cb047fb9a1e5fc", "local_source": "audio/librispeech_1221-135767-0006.flac", "local_wav": "audio/librispeech_1221-135767-0006.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1221-135767-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_1221", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1221-135767-0019", "transcript": "MOTHER CRIED SHE I SEE YOU HERE LOOK LOOK", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1221/135767/1221-135767-0019.flac"}, "sample_rate": 16000, "source_samples": 60480, "source_duration_seconds": 3.78, "source_peak": 0.397979736328125, "source_rms_dbfs": -26.54945814985232, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.54945814985232, "decoded_pcm_sha256": "0ed57e185be961c721604990dce623ab138f9937e79722c597c3874498c02937", "source_sha256": "245fd9a32160c5fa9dd1a76624a71c7e1398de763f8e9ab88a7dd25aae1002f4", "wav_sha256": "dffedfb5e6881aaaa7fffc0aea511d5ae9815d0718a7632999004dca5c04ce36", "local_source": "audio/librispeech_1221-135767-0019.flac", "local_wav": "audio/librispeech_1221-135767-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 60480}, {"id": "librispeech_1284-1180-0020", "corpus": "librispeech_test_clean", "speaker": "librispeech_1284", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1284-1180-0020", "transcript": "THE FIRST LOT WE TESTED ON OUR GLASS CAT WHICH NOT ONLY BEGAN TO LIVE BUT HAS LIVED EVER SINCE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1284/1180/1284-1180-0020.flac"}, "sample_rate": 16000, "source_samples": 93920, "source_duration_seconds": 5.87, "source_peak": 0.4365234375, "source_rms_dbfs": -24.258485416618605, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.258485416618605, "decoded_pcm_sha256": "0bce114d9020d75b12c3d76b3c232bd8cd315dc052825ca64d389fcb87293422", "source_sha256": "b67810620f4f39a1bca2214b4f6475b98f19452b1cd0a4cc67efee2a8a805016", "wav_sha256": "e2fa9e1870ad76ff19bf8817e638d6c60e1fc83bbbc3577cf3cd7f2b51ced390", "local_source": "audio/librispeech_1284-1180-0020.flac", "local_wav": "audio/librispeech_1284-1180-0020.wav", "excerpt_start_sample": 0, "excerpt_samples": 93920}, {"id": "librispeech_1284-1180-0027", "corpus": "librispeech_test_clean", "speaker": "librispeech_1284", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1284-1180-0027", "transcript": "YET THAT TASK WAS NOT SO EASY AS YOU MAY SUPPOSE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1284/1180/1284-1180-0027.flac"}, "sample_rate": 16000, "source_samples": 52320, "source_duration_seconds": 3.27, "source_peak": 0.374481201171875, "source_rms_dbfs": -23.509845909817045, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.509845909817045, "decoded_pcm_sha256": "96e7e523ecd8732e2ed1542dd9b6cbde8714adfeef42d0c1bfb37fb1a7c1cf83", "source_sha256": "9c4d5811c87e0273494850f3401459de8eafd6731f43a615c07f9ff944262a95", "wav_sha256": "19f2fc30155fbf7c6c393466ad62366f3db8a378533c3fceb06a2668b540c1fc", "local_source": "audio/librispeech_1284-1180-0027.flac", "local_wav": "audio/librispeech_1284-1180-0027.wav", "excerpt_start_sample": 0, "excerpt_samples": 52320}, {"id": "librispeech_1284-1180-0028", "corpus": "librispeech_test_clean", "speaker": "librispeech_1284", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1284-1180-0028", "transcript": "A BED QUILT MADE OF PATCHES OF DIFFERENT KINDS AND COLORS OF CLOTH ALL NEATLY SEWED TOGETHER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1284/1180/1284-1180-0028.flac"}, "sample_rate": 16000, "source_samples": 96720, "source_duration_seconds": 6.045, "source_peak": 0.508331298828125, "source_rms_dbfs": -25.041413930656645, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.009063765320747, "decoded_pcm_sha256": "0bfea5c92c4d2ee444f3984a4d2b53407bcff068a4ef104efccbb0731c86faf4", "source_sha256": "1ee65c4ba89a7fe76bef27dade9b0ab41fe2109b77c8e68d5b80e34194031ee4", "wav_sha256": "877171668e0347c683e8c5d855a388bd8b3df6a214587ba1670ff202b603832a", "local_source": "audio/librispeech_1284-1180-0028.flac", "local_wav": "audio/librispeech_1284-1180-0028.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1320-122617-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_1320", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1320-122617-0019", "transcript": "UNCAS OCCUPIED A DISTANT CORNER IN A RECLINING ATTITUDE BEING RIGIDLY BOUND BOTH HANDS AND FEET BY STRONG AND PAINFUL WITHES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1320/122617/1320-122617-0019.flac"}, "sample_rate": 16000, "source_samples": 131680, "source_duration_seconds": 8.23, "source_peak": 0.569244384765625, "source_rms_dbfs": -21.522908425452997, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.91267260169434, "decoded_pcm_sha256": "8e4bd7f99207ae6f91c9d166201c4877840bfbd64affa6d81029b74db79e47d6", "source_sha256": "3f0ea5a1e33f502dca87255f51487aa60fcb4f12605e9feba2adeb14df1d89f7", "wav_sha256": "dccd68b1d51eb31bdebba14d8c1c3bdd16cb9a1d378941d4fde54eed91ca6ca2", "local_source": "audio/librispeech_1320-122617-0019.flac", "local_wav": "audio/librispeech_1320-122617-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1320-122617-0023", "corpus": "librispeech_test_clean", "speaker": "librispeech_1320", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1320-122617-0023", "transcript": "UNCAS WHO HAD ALREADY APPROACHED THE DOOR IN READINESS TO LEAD THE WAY NOW RECOILED AND PLACED HIMSELF ONCE MORE IN THE BOTTOM OF THE LODGE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1320/122617/1320-122617-0023.flac"}, "sample_rate": 16000, "source_samples": 125040, "source_duration_seconds": 7.815, "source_peak": 0.526458740234375, "source_rms_dbfs": -22.664914502509315, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.27270752547399, "decoded_pcm_sha256": "b3661757a6ed833feab8313fa697ecafb680dc14c418cd468619e460e550745e", "source_sha256": "553fc6594343774831ac876fee34c4ec2b47b53d451f313b10e64eea5cadd8c2", "wav_sha256": "eadf1d183380287ca406621110169ede44028d7ccf2ec2b7c0454cd0ce9baeb5", "local_source": "audio/librispeech_1320-122617-0023.flac", "local_wav": "audio/librispeech_1320-122617-0023.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1320-122617-0039", "corpus": "librispeech_test_clean", "speaker": "librispeech_1320", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "1320-122617-0039", "transcript": "THE MOHICAN STARTED ON HIS FEET AND SHOOK HIS SHAGGY COVERING AS THOUGH THE ANIMAL HE COUNTERFEITED WAS ABOUT TO MAKE SOME DESPERATE EFFORT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1320/122617/1320-122617-0039.flac"}, "sample_rate": 16000, "source_samples": 112880, "source_duration_seconds": 7.055, "source_peak": 0.5504150390625, "source_rms_dbfs": -21.698218264348213, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.2918213919449, "decoded_pcm_sha256": "9cf3ed8985625af93ca84569280377a536aa4892a8aefbabb265fc0a052a863c", "source_sha256": "264a53a8c31bd50758ba7eeeb3b5efa9c5bc7908757d44a635bec4c9a4f2c3c0", "wav_sha256": "736879dad172de3ddbbc6cba1cf6b9db0a9b9d447ce01399ec8cd3579d63e366", "local_source": "audio/librispeech_1320-122617-0039.flac", "local_wav": "audio/librispeech_1320-122617-0039.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1580-141083-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_1580", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1580-141083-0010", "transcript": "I GAVE HIM A LITTLE BRANDY AND LEFT HIM COLLAPSED IN A CHAIR WHILE I MADE A MOST CAREFUL EXAMINATION OF THE ROOM", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1580/141083/1580-141083-0010.flac"}, "sample_rate": 16000, "source_samples": 85120, "source_duration_seconds": 5.32, "source_peak": 0.542816162109375, "source_rms_dbfs": -25.39547002291151, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.39547002291151, "decoded_pcm_sha256": "d43cef78fec64a97f03dd03591e102c104a281997670664d424ccf19276cb4a0", "source_sha256": "15b52d8b6b3ab1e0d95b51380278481420b17cdd350b6fe62c29aca97d46a6a7", "wav_sha256": "4498e89acf526da02a12bb3c73380b07b96d180fccbe89e5eb32c295e313221a", "local_source": "audio/librispeech_1580-141083-0010.flac", "local_wav": "audio/librispeech_1580-141083-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 85120}, {"id": "librispeech_1580-141084-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_1580", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1580-141084-0016", "transcript": "MY FRIEND DID NOT APPEAR TO BE DEPRESSED BY HIS FAILURE BUT SHRUGGED HIS SHOULDERS IN HALF HUMOROUS RESIGNATION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1580/141084/1580-141084-0016.flac"}, "sample_rate": 16000, "source_samples": 95360, "source_duration_seconds": 5.96, "source_peak": 0.5615234375, "source_rms_dbfs": -26.925662267221572, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.925662267221572, "decoded_pcm_sha256": "0698365674d63379a8ffa4e1a7bae8ef5f6ed226e14107811f5311d9de011ca9", "source_sha256": "86f57be846b5ad8817f2752d90d5f2aba1e97ec394b2f4ccf28bc5dc39a181ef", "wav_sha256": "0cf7a7fe6f0728e339cc5584e7b2224d39f0e765e107f75e333ec6c19316e34c", "local_source": "audio/librispeech_1580-141084-0016.flac", "local_wav": "audio/librispeech_1580-141084-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 95360}, {"id": "librispeech_1580-141084-0047", "corpus": "librispeech_test_clean", "speaker": "librispeech_1580", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1580-141084-0047", "transcript": "I HAVE A LETTER HERE MISTER SOAMES WHICH I WROTE TO YOU EARLY THIS MORNING IN THE MIDDLE OF A RESTLESS NIGHT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1580/141084/1580-141084-0047.flac"}, "sample_rate": 16000, "source_samples": 84000, "source_duration_seconds": 5.25, "source_peak": 0.592376708984375, "source_rms_dbfs": -27.064546771845936, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.064546771845936, "decoded_pcm_sha256": "ca185084384947ca6386941bb27f596abb096a46a774b4716da42a2e461e4cab", "source_sha256": "0e78a5bcc7c122a563c24dc2d7a65f1be8d851e0d45c8caa4226571676cc5d4c", "wav_sha256": "d8dbb036d18173b12b42c70eff46d203272d1fd2479f615e6e6ef5fb71b45e84", "local_source": "audio/librispeech_1580-141084-0047.flac", "local_wav": "audio/librispeech_1580-141084-0047.wav", "excerpt_start_sample": 0, "excerpt_samples": 84000}, {"id": "librispeech_1995-1836-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_1995", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1995-1836-0000", "transcript": "THE HON CHARLES SMITH MISS SARAH'S BROTHER WAS WALKING SWIFTLY UPTOWN FROM MISTER EASTERLY'S WALL STREET OFFICE AND HIS FACE WAS PALE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1995/1836/1995-1836-0000.flac"}, "sample_rate": 16000, "source_samples": 143280, "source_duration_seconds": 8.955, "source_peak": 0.295013427734375, "source_rms_dbfs": -23.750101874704374, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.46393888832139, "decoded_pcm_sha256": "3feb5dcb82941d2a5e9f89efbf0e32e0138e5eec693c118bb9e8031d23d08925", "source_sha256": "8120c55c7231b13d3d497b04b2fe3d999cfdac785a0134eb0fc9e41aa90ce591", "wav_sha256": "4aafb5b71621c159d7c4925ab8452b0bacf56fb3a1c0a62391bd0c29221d989c", "local_source": "audio/librispeech_1995-1836-0000.flac", "local_wav": "audio/librispeech_1995-1836-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1995-1836-0014", "corpus": "librispeech_test_clean", "speaker": "librispeech_1995", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1995-1836-0014", "transcript": "FORTUNATELY SAID MISTER VANDERPOOL NORTHERNERS AND SOUTHERNERS ARE ARRIVING AT A BETTER MUTUAL UNDERSTANDING ON MOST OF THESE MATTERS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1995/1836/1995-1836-0014.flac"}, "sample_rate": 16000, "source_samples": 144720, "source_duration_seconds": 9.045, "source_peak": 0.529449462890625, "source_rms_dbfs": -22.041190562551304, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.513469437558737, "decoded_pcm_sha256": "af7f0befa255d17d0806307d2e5a0582ff2d024a50adfd0a534360c964de3b82", "source_sha256": "c90576861bc6bc0b6837f0b1c340ecc276ac0339ee0a25442a4784f5fc4a4821", "wav_sha256": "44c2d7a7fad4ab4db4dd9d0c3b94288237e031c318c2b8997522fc519b69d61c", "local_source": "audio/librispeech_1995-1836-0014.flac", "local_wav": "audio/librispeech_1995-1836-0014.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_1995-1837-0023", "corpus": "librispeech_test_clean", "speaker": "librispeech_1995", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "1995-1837-0023", "transcript": "THE NET AND WEB OF ENDLESS THINGS HAD BEEN CRAWLING AND CREEPING AROUND HER SHE HAD STRUGGLED IN DUMB SPEECHLESS TERROR AGAINST SOME MIGHTY GRASPING THAT STROVE FOR HER LIFE WITH GNARLED AND CREEPING FINGERS BUT NOW AT LAST WEAKLY SHE OPENED HER EYES AND QUESTIONED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/1995/1837/1995-1837-0023.flac"}, "sample_rate": 16000, "source_samples": 271440, "source_duration_seconds": 16.965, "source_peak": 0.366790771484375, "source_rms_dbfs": -23.594041351148842, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.511532890985304, "decoded_pcm_sha256": "2fff476dd7493d67e46c49e37f104d20891e347d86560cf73c09f434075020cb", "source_sha256": "7f584164eef2f5976f874ee9c027933c892ba7f920f7233bfd2a219d0ebffcd4", "wav_sha256": "b1024b177f01004b24a07a5fe535254d4fe201bd8f6ca6820677b0b8d9569c92", "local_source": "audio/librispeech_1995-1837-0023.flac", "local_wav": "audio/librispeech_1995-1837-0023.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2094-142345-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_2094", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2094-142345-0032", "transcript": "I OFTEN HEARD HER TALK OF YOU IN THE SAME SORT OF WAY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2094/142345/2094-142345-0032.flac"}, "sample_rate": 16000, "source_samples": 51840, "source_duration_seconds": 3.24, "source_peak": 0.49334716796875, "source_rms_dbfs": -21.689633787280776, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.689633787280776, "decoded_pcm_sha256": "973bdb09bc9497c63b4dfbb3bc77778c87a6ced88d08e7f2f6e22739dd6d5444", "source_sha256": "b92bf752cc5264a2a47e0153c236b90edb48dd32b819cfd12d0e5250ca3c7a1e", "wav_sha256": "320b33604dfc6f64e4624c4baa3515e073f94afc1760147678affda9973624a2", "local_source": "audio/librispeech_2094-142345-0032.flac", "local_wav": "audio/librispeech_2094-142345-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 51840}, {"id": "librispeech_2094-142345-0039", "corpus": "librispeech_test_clean", "speaker": "librispeech_2094", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2094-142345-0039", "transcript": "I'VE STRONG ASSURANCE THAT NO EVIL WILL HAPPEN TO YOU AND MY UNCLE AND THE CHILDREN FROM ANYTHING I'VE DONE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2094/142345/2094-142345-0039.flac"}, "sample_rate": 16000, "source_samples": 100480, "source_duration_seconds": 6.28, "source_peak": 0.509735107421875, "source_rms_dbfs": -21.82028978216153, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.62361673126311, "decoded_pcm_sha256": "c411d502f97ab6fc22d81b52c4ba95e5e562514c255c2379c8837148877e0ea1", "source_sha256": "21d7d00350a4ca2b0109da67cfad724f36979e7c1b2e098f9327bc8c4f433435", "wav_sha256": "c7a482fbc5c97605308c5c4ecd86d44207027acba93121762e33b0b573212e24", "local_source": "audio/librispeech_2094-142345-0039.flac", "local_wav": "audio/librispeech_2094-142345-0039.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2094-142345-0040", "corpus": "librispeech_test_clean", "speaker": "librispeech_2094", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2094-142345-0040", "transcript": "I DIDN'T PREACH WITHOUT DIRECTION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2094/142345/2094-142345-0040.flac"}, "sample_rate": 16000, "source_samples": 38160, "source_duration_seconds": 2.385, "source_peak": 0.6126708984375, "source_rms_dbfs": -22.307817294696616, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.307817294696616, "decoded_pcm_sha256": "9cbc49d67edc85ca6ba00d13dae38ec545b7aaf549562d1af9d9e72881814d19", "source_sha256": "8b3f5a025e4853fabdb000f2820a651f4a66a5ef234d912581ae29fafea72e48", "wav_sha256": "6bd7f0881080a71cb6c4ab59396378576c0114a4c9e05c3c1de5edbdeeeb0664", "local_source": "audio/librispeech_2094-142345-0040.flac", "local_wav": "audio/librispeech_2094-142345-0040.wav", "excerpt_start_sample": 0, "excerpt_samples": 38160}, {"id": "librispeech_2300-131720-0022", "corpus": "librispeech_test_clean", "speaker": "librispeech_2300", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2300-131720-0022", "transcript": "MEANWHILE HE HAD CALLED UPON ME TO MAKE A REPORT OF THE THREE WIRE SYSTEM KNOWN IN ENGLAND AS THE HOPKINSON BOTH DOCTOR JOHN HOPKINSON AND MISTER EDISON BEING INDEPENDENT INVENTORS AT PRACTICALLY THE SAME TIME", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2300/131720/2300-131720-0022.flac"}, "sample_rate": 16000, "source_samples": 218640, "source_duration_seconds": 13.665, "source_peak": 0.64959716796875, "source_rms_dbfs": -22.580480288302475, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.225644794441955, "decoded_pcm_sha256": "b7f1420236bac4a5c404bc851efe87bc6142a76f2a2935e40197b946d8182998", "source_sha256": "93212060c39673d7a3ee80b52a8f2612e70e3c121fd588c690e3d51be10e6963", "wav_sha256": "47955be189936046241d85bf424da25472f89de1b610988282b8cb71c08fcad7", "local_source": "audio/librispeech_2300-131720-0022.flac", "local_wav": "audio/librispeech_2300-131720-0022.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2300-131720-0028", "corpus": "librispeech_test_clean", "speaker": "librispeech_2300", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2300-131720-0028", "transcript": "THERE WAS INFINITE SCEPTICISM AROUND HIM ON THE SUBJECT AND WHILE OTHER INVENTORS WERE ALSO GIVING THE SUBJECT THEIR THOUGHT THE PUBLIC TOOK IT FOR GRANTED THAT ANYTHING SO UTTERLY INTANGIBLE AS ELECTRICITY THAT COULD NOT BE SEEN OR WEIGHED AND ONLY GAVE SECONDARY EVIDENCE OF ITSELF AT THE EXACT POINT OF USE COULD NOT BE BROUGHT TO ACCURATE REGISTRATION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2300/131720/2300-131720-0028.flac"}, "sample_rate": 16000, "source_samples": 415040, "source_duration_seconds": 25.94, "source_peak": 0.58966064453125, "source_rms_dbfs": -23.395115497521118, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.870200927979273, "decoded_pcm_sha256": "9eaf55b481a213cff53140518cebb83838d1187d3e2187282fcf13a3699505c4", "source_sha256": "72f44e8aee815c88d0ad1b8f6d9a15fea55954dbd8e4d996b97112fe34bda0a1", "wav_sha256": "1aa80ada2f91d3419088daa04ecfc0ddc106837f4ce52f87063757534aef4e1d", "local_source": "audio/librispeech_2300-131720-0028.flac", "local_wav": "audio/librispeech_2300-131720-0028.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2300-131720-0041", "corpus": "librispeech_test_clean", "speaker": "librispeech_2300", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2300-131720-0041", "transcript": "WE HAD METERS IN WHICH THERE WERE TWO BOTTLES OF LIQUID", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2300/131720/2300-131720-0041.flac"}, "sample_rate": 16000, "source_samples": 60000, "source_duration_seconds": 3.75, "source_peak": 0.49029541015625, "source_rms_dbfs": -23.37033603041045, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.37033603041045, "decoded_pcm_sha256": "976092b4e8947717eb460c0f8df2d08babdc9b088a74a4efec5fca50e34afea2", "source_sha256": "4ede50e44cbe5dfb258311f80fd68b7efe2e080265efa5e6bc07aea531a22e28", "wav_sha256": "ceb4b44d00ec8e36a1e67a7f31a0bc8f7440b700ab4fd9d88011defa7894bcd0", "local_source": "audio/librispeech_2300-131720-0041.flac", "local_wav": "audio/librispeech_2300-131720-0041.wav", "excerpt_start_sample": 0, "excerpt_samples": 60000}, {"id": "librispeech_237-126133-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_237", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "237-126133-0003", "transcript": "SOMEHOW OF ALL THE DAYS WHEN THE HOME FEELING WAS THE STRONGEST THIS DAY IT SEEMED AS IF SHE COULD BEAR IT NO LONGER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/237/126133/237-126133-0003.flac"}, "sample_rate": 16000, "source_samples": 104960, "source_duration_seconds": 6.56, "source_peak": 0.233489990234375, "source_rms_dbfs": -30.677490302294746, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.31670425464334, "decoded_pcm_sha256": "15627325ed4198a6a91f3a9dd72c16085a4986165c018e492c56a9076ad969f4", "source_sha256": "18c38673f71d4012a9c023d7548bff3c0f78172415f67c69abf403590c4e84bc", "wav_sha256": "0496d3168b5687d428050d9c03d395d3225f8e034222090d10a32baed9384b2f", "local_source": "audio/librispeech_237-126133-0003.flac", "local_wav": "audio/librispeech_237-126133-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_237-134500-0029", "corpus": "librispeech_test_clean", "speaker": "librispeech_237", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "237-134500-0029", "transcript": "I DON'T WANT TO STAND AROUND AND LOOK ON", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/237/134500/237-134500-0029.flac"}, "sample_rate": 16000, "source_samples": 33040, "source_duration_seconds": 2.065, "source_peak": 0.935546875, "source_rms_dbfs": -21.343679960921527, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.343679960921527, "decoded_pcm_sha256": "8562d47701848db95c8610733017a69a5928db4edb524e0f205ab4182cf7b4f7", "source_sha256": "a71914689325868d93f40677ab17bdfbb8c8209801864e31a4f4303cf7fd0fa0", "wav_sha256": "d1f7208286ea0672b525054ddc0371bc4a14a529a808bb5cd1163899ca3e9af3", "local_source": "audio/librispeech_237-134500-0029.flac", "local_wav": "audio/librispeech_237-134500-0029.wav", "excerpt_start_sample": 0, "excerpt_samples": 33040}, {"id": "librispeech_237-134500-0031", "corpus": "librispeech_test_clean", "speaker": "librispeech_237", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "237-134500-0031", "transcript": "SOMETIMES I DON'T WANT TO DO ANYTHING AT ALL AND SOMETIMES I WANT TO PULL THE FOUR CORNERS OF THE DIVIDE TOGETHER HE THREW OUT HIS ARM AND BROUGHT IT BACK WITH A JERK SO LIKE A TABLE CLOTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/237/134500/237-134500-0031.flac"}, "sample_rate": 16000, "source_samples": 169760, "source_duration_seconds": 10.61, "source_peak": 0.700347900390625, "source_rms_dbfs": -21.801426992576367, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.291181411176545, "decoded_pcm_sha256": "66fcbe0885d7123cd5c5c14ff8b544f94ed65d51857cb70d13de972dd00f9421", "source_sha256": "7456acb116b3a8c6d43c44ebf79a8f7b59f6ed546ffb63ce78a6dc58bb6d5abe", "wav_sha256": "7fcf9d87377483efb1617dbb6ddf6de5d613cf60dc9c5826d5c3b2d1f5efeb32", "local_source": "audio/librispeech_237-134500-0031.flac", "local_wav": "audio/librispeech_237-134500-0031.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_260-123286-0008", "corpus": "librispeech_test_clean", "speaker": "librispeech_260", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "260-123286-0008", "transcript": "THEREFORE DON'T TALK TO ME ABOUT VIEWS AND PROSPECTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/260/123286/260-123286-0008.flac"}, "sample_rate": 16000, "source_samples": 59760, "source_duration_seconds": 3.735, "source_peak": 0.46673583984375, "source_rms_dbfs": -23.64591439734617, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.64591439734617, "decoded_pcm_sha256": "931193da803cf63a31b439032de16719f19aca2c3130d571b4b8044d43a1413c", "source_sha256": "735769549e4dbbfb3f4ba43f89866cde8085933d8b16e67550cc65264bf84fc3", "wav_sha256": "a76c87fb51f5ea1aaa5040037a8d8364b9103e68aace0946bee820b613b70f25", "local_source": "audio/librispeech_260-123286-0008.flac", "local_wav": "audio/librispeech_260-123286-0008.wav", "excerpt_start_sample": 0, "excerpt_samples": 59760}, {"id": "librispeech_260-123288-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_260", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "260-123288-0003", "transcript": "THE ELECTRIC LIGHT CAN SCARCELY PENETRATE THROUGH THE DENSE CURTAIN WHICH HAS DROPPED OVER THE THEATRE ON WHICH THE BATTLE OF THE ELEMENTS IS ABOUT TO BE WAGED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/260/123288/260-123288-0003.flac"}, "sample_rate": 16000, "source_samples": 142480, "source_duration_seconds": 8.905, "source_peak": 0.39129638671875, "source_rms_dbfs": -25.639248238099196, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.560371051032, "decoded_pcm_sha256": "d480f00d02d595a6352c610e02536f099ff03473f6693b5628683647e9479766", "source_sha256": "5e8ca953fe1b2dd885f0cfbf4d706040e55637eee883bea8f0dc6e08afac3240", "wav_sha256": "b92004dac924d397e6ba0829351dde9494b14ae7a6782ccdbd4d8cbd2680f841", "local_source": "audio/librispeech_260-123288-0003.flac", "local_wav": "audio/librispeech_260-123288-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_260-123440-0020", "corpus": "librispeech_test_clean", "speaker": "librispeech_260", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "260-123440-0020", "transcript": "WE WON'T TALK ABOUT HER ANY MORE IF YOU'D RATHER NOT WE INDEED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/260/123440/260-123440-0020.flac"}, "sample_rate": 16000, "source_samples": 79920, "source_duration_seconds": 4.995, "source_peak": 0.46844482421875, "source_rms_dbfs": -22.68064988825362, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.68064988825362, "decoded_pcm_sha256": "c10a9cdd8fd4960bfb88b1611c0123695ebcd137816c47ff78925a8699616669", "source_sha256": "44be2bb2c15059058798a25f87de6310601074e1bc4b68bda0ad3ad1860a9040", "wav_sha256": "dc0e2b8720fca680ed9e8a46c6c222626f95be9b07311fc744a3761bb1f0981f", "local_source": "audio/librispeech_260-123440-0020.flac", "local_wav": "audio/librispeech_260-123440-0020.wav", "excerpt_start_sample": 0, "excerpt_samples": 79920}, {"id": "librispeech_2830-3980-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_2830", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2830-3980-0000", "transcript": "IN EVERY WAY THEY SOUGHT TO UNDERMINE THE AUTHORITY OF SAINT PAUL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2830/3980/2830-3980-0000.flac"}, "sample_rate": 16000, "source_samples": 59680, "source_duration_seconds": 3.73, "source_peak": 0.89697265625, "source_rms_dbfs": -16.80410950587789, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -16.80410950587789, "decoded_pcm_sha256": "8f71058a78243905dd373f5302bc82f25ff2a7e7631413b5ce924104f0f14bea", "source_sha256": "a73adb84b9560fdfc8f6c7f5801e567267ac8beb6fd51df4f5dec7217eb43ad8", "wav_sha256": "77aacbae6ecdcf8ad5434acd4a1388d34de3c30a9ad3cd826055b781b1bc977f", "local_source": "audio/librispeech_2830-3980-0000.flac", "local_wav": "audio/librispeech_2830-3980-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 59680}, {"id": "librispeech_2830-3980-0038", "corpus": "librispeech_test_clean", "speaker": "librispeech_2830", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2830-3980-0038", "transcript": "GRACE BE TO YOU AND PEACE FROM GOD THE FATHER AND FROM OUR LORD JESUS CHRIST", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2830/3980/2830-3980-0038.flac"}, "sample_rate": 16000, "source_samples": 88640, "source_duration_seconds": 5.54, "source_peak": 0.566497802734375, "source_rms_dbfs": -24.04910591648981, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.04910591648981, "decoded_pcm_sha256": "77d64124560df236e40a3ec37cc728b327f805fee264fb8362c33791320222e1", "source_sha256": "2c8978e75dd10716d066062bb2fbfd1a002ac3190e42b0f251f15d0aa2d6d272", "wav_sha256": "eb598d45742c7992ecb5c2ef39512cf5ba5471f617decb4918e42ed52e064291", "local_source": "audio/librispeech_2830-3980-0038.flac", "local_wav": "audio/librispeech_2830-3980-0038.wav", "excerpt_start_sample": 0, "excerpt_samples": 88640}, {"id": "librispeech_2830-3980-0042", "corpus": "librispeech_test_clean", "speaker": "librispeech_2830", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "2830-3980-0042", "transcript": "THE WORLD BRANDS THIS A PERNICIOUS DOCTRINE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2830/3980/2830-3980-0042.flac"}, "sample_rate": 16000, "source_samples": 48320, "source_duration_seconds": 3.02, "source_peak": 0.554443359375, "source_rms_dbfs": -22.785804418379104, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.785804418379104, "decoded_pcm_sha256": "3b4a08b0741e0ca396487185085d64365ab56b16d5868e61a4550a5dc24466f2", "source_sha256": "ff30195a24b3f6c24a6ad4748f2ab49f4a036975c0b62f5b1c2f4203488ee9ae", "wav_sha256": "9226d5602341de020d92f33ac9700e8ea29f2fd9ff33db456d1a37078a94b50a", "local_source": "audio/librispeech_2830-3980-0042.flac", "local_wav": "audio/librispeech_2830-3980-0042.wav", "excerpt_start_sample": 0, "excerpt_samples": 48320}, {"id": "librispeech_2961-960-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_2961", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2961-960-0010", "transcript": "BUT HE HAS NOT AS YET DEFINED THIS INTERMEDIATE TERRITORY WHICH LIES SOMEWHERE BETWEEN MEDICINE AND MATHEMATICS AND HE WOULD HAVE FELT THAT THERE WAS AS GREAT AN IMPIETY IN RANKING THEORIES OF PHYSICS FIRST IN THE ORDER OF KNOWLEDGE AS IN PLACING THE BODY BEFORE THE SOUL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2961/960/2961-960-0010.flac"}, "sample_rate": 16000, "source_samples": 327360, "source_duration_seconds": 20.46, "source_peak": 0.380584716796875, "source_rms_dbfs": -28.229074807392955, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.157181381609547, "decoded_pcm_sha256": "ab9593d38d133538cf3475c5a7bf30b265737c427a5c4a501ecf3384b1e2aee8", "source_sha256": "d93e37326bfa987a0cf66c0edd79dfa034ea013e5967c8f2ba9edf54cde8d2be", "wav_sha256": "52ca5b9bf9daf5a48184c8d90a1033ba402bb03cb88b7ea9362dd6fa4b24abd9", "local_source": "audio/librispeech_2961-960-0010.flac", "local_wav": "audio/librispeech_2961-960-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2961-960-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_2961", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2961-960-0019", "transcript": "HE COULD WRITE IN ONE STYLE BUT NOT IN ANOTHER AND THE GREEK LANGUAGE HAD NOT AS YET BEEN FASHIONED BY ANY POET OR PHILOSOPHER TO DESCRIBE PHYSICAL PHENOMENA", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2961/960/2961-960-0019.flac"}, "sample_rate": 16000, "source_samples": 194080, "source_duration_seconds": 12.13, "source_peak": 0.625213623046875, "source_rms_dbfs": -25.64813842860343, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.64907732157032, "decoded_pcm_sha256": "966508e2ccc362392d8ec80fc3d05a564460cb6f0dd69f613c7b86089cb21caa", "source_sha256": "f5052549b9cbc3b8433ae187d0e876b214f5250ddb4fbb637dea220e8a4726cf", "wav_sha256": "834322db0896bb0348dd0ab0f7fc4d268c5c91a6c91e78f489697f63ba92e7f7", "local_source": "audio/librispeech_2961-960-0019.flac", "local_wav": "audio/librispeech_2961-960-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_2961-961-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_2961", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "2961-961-0011", "transcript": "THE GENEALOGIES WHICH YOU HAVE RECITED TO US OUT OF YOUR OWN ANNALS SOLON ARE A MERE CHILDREN'S STORY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/2961/961/2961-961-0011.flac"}, "sample_rate": 16000, "source_samples": 125040, "source_duration_seconds": 7.815, "source_peak": 0.61773681640625, "source_rms_dbfs": -27.948512195945238, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.258152323716782, "decoded_pcm_sha256": "00bd4132c58a261e6169cd035b93bf1134d0bd1b082f44ceb99cf14faa9b7062", "source_sha256": "f859855c45fc07e8d561ed7a846fd14ca4ae1515749b60c4e8d4ce825df30180", "wav_sha256": "ddf0f4bb5b9f2e544585e0723cf46b3096a44e71b8c13ef37d1d0a6e688f07e4", "local_source": "audio/librispeech_2961-961-0011.flac", "local_wav": "audio/librispeech_2961-961-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3570-5694-0001", "corpus": "librispeech_test_clean", "speaker": "librispeech_3570", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3570-5694-0001", "transcript": "THE UTILITY OF CONSUMPTION AS AN EVIDENCE OF WEALTH IS TO BE CLASSED AS A DERIVATIVE GROWTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3570/5694/3570-5694-0001.flac"}, "sample_rate": 16000, "source_samples": 90800, "source_duration_seconds": 5.675, "source_peak": 0.429901123046875, "source_rms_dbfs": -22.81140123891472, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.81140123891472, "decoded_pcm_sha256": "d7ebe73ddd2095245c2316a87024f8d15117a6bca41809d6c43a3d2cfe68c3a9", "source_sha256": "83d42febda01d467c35c0b90dc7bb5d5d80b18ed95d65748a831afca53f2e3cd", "wav_sha256": "5c52d70d17699b53a204ff0cdbe488ace62913f5304818fe714ac7198b7b65a2", "local_source": "audio/librispeech_3570-5694-0001.flac", "local_wav": "audio/librispeech_3570-5694-0001.wav", "excerpt_start_sample": 0, "excerpt_samples": 90800}, {"id": "librispeech_3570-5694-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_3570", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3570-5694-0004", "transcript": "IN THE NATURE OF THINGS LUXURIES AND THE COMFORTS OF LIFE BELONG TO THE LEISURE CLASS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3570/5694/3570-5694-0004.flac"}, "sample_rate": 16000, "source_samples": 85280, "source_duration_seconds": 5.33, "source_peak": 0.596405029296875, "source_rms_dbfs": -23.986191830522884, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.986191830522884, "decoded_pcm_sha256": "86a8946d8572399656832193f585d37dfcbd327696eb662dcf0e41d4b67d0bd1", "source_sha256": "e7c4c68060bd5da1df857aa90f0fb6601cfee83e0001d75b64164988b070e7d8", "wav_sha256": "75ab8457c235b0d72c8db7fd3f4b5e5feb1e2b50ce0c9835a6af02d2ab282d69", "local_source": "audio/librispeech_3570-5694-0004.flac", "local_wav": "audio/librispeech_3570-5694-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 85280}, {"id": "librispeech_3570-5696-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_3570", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3570-5696-0010", "transcript": "AN ARTICLE MAY BE USEFUL AND WASTEFUL BOTH AND ITS UTILITY TO THE CONSUMER MAY BE MADE UP OF USE AND WASTE IN THE MOST VARYING PROPORTIONS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3570/5696/3570-5696-0010.flac"}, "sample_rate": 16000, "source_samples": 121120, "source_duration_seconds": 7.57, "source_peak": 0.74517822265625, "source_rms_dbfs": -22.057355240612, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.485928141030136, "decoded_pcm_sha256": "0efec23fe6eeee5dfee017e6be4b06b64a152a981c62a71c533addc819701d30", "source_sha256": "426a986fec027f14754878775f6b605d58d61a0fd0d77bf122657dc1cef2ce9d", "wav_sha256": "9bdc7d45ba3123b85371d1a0bfaec30557f59f961d83177b590797de0d75f4c6", "local_source": "audio/librispeech_3570-5696-0010.flac", "local_wav": "audio/librispeech_3570-5696-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3575-170457-0034", "corpus": "librispeech_test_clean", "speaker": "librispeech_3575", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3575-170457-0034", "transcript": "IN THIS MONOTONOUS LIFE OF MINE THAT WAS A PLEASANT EVENT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3575/170457/3575-170457-0034.flac"}, "sample_rate": 16000, "source_samples": 55920, "source_duration_seconds": 3.495, "source_peak": 0.84869384765625, "source_rms_dbfs": -19.712774707346266, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -19.712774707346266, "decoded_pcm_sha256": "b1c60dfc008f07f56156adc48b00db15f16da435fb212895c0516ecbd9e5de30", "source_sha256": "7379f6afffbfc11aded5aae60cf94ff2f5aa5a2dc317f61b2cd70c9c4d4847df", "wav_sha256": "53cc005f386dbcd9738a8dd623cab86151cf87bd4f53bdfa78d7cea937f0764f", "local_source": "audio/librispeech_3575-170457-0034.flac", "local_wav": "audio/librispeech_3575-170457-0034.wav", "excerpt_start_sample": 0, "excerpt_samples": 55920}, {"id": "librispeech_3575-170457-0041", "corpus": "librispeech_test_clean", "speaker": "librispeech_3575", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3575-170457-0041", "transcript": "SHE WAS GONE OUT INTO THE VILLAGE ON SOME ERRAND WHEN AS SHE WAS DESCENDING THE STEEP STREET HER FOOT SLIPPED ON THE ICE AND SHE FELL IT WAS DARK AND NO ONE SAW HER MISCHANCE TILL AFTER A TIME HER GROANS ATTRACTED THE ATTENTION OF A PASSER BY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3575/170457/3575-170457-0041.flac"}, "sample_rate": 16000, "source_samples": 224160, "source_duration_seconds": 14.01, "source_peak": 0.66522216796875, "source_rms_dbfs": -21.83040488145807, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.483529355547855, "decoded_pcm_sha256": "ce9155c8469c627681fdbab9602fc52b17f46ce8849d53a2be4d450d58cfb2c4", "source_sha256": "90d1dd17384a0960d3acaf50b3f7f40b4699288aa74a04c25a221270e104ce55", "wav_sha256": "f49c7afeb6e5b22010dcb4ea4529e175cd6527a2dc81e2ef1d5509308999fa73", "local_source": "audio/librispeech_3575-170457-0041.flac", "local_wav": "audio/librispeech_3575-170457-0041.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3575-170457-0054", "corpus": "librispeech_test_clean", "speaker": "librispeech_3575", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3575-170457-0054", "transcript": "STUNG BY ANXIETY FOR THIS LITTLE SISTER SHE UPBRAIDED MISS W FOR HER FANCIED INDIFFERENCE TO ANNE'S STATE OF HEALTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3575/170457/3575-170457-0054.flac"}, "sample_rate": 16000, "source_samples": 128080, "source_duration_seconds": 8.005, "source_peak": 0.630035400390625, "source_rms_dbfs": -23.47235016030915, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.430811733763257, "decoded_pcm_sha256": "5ff0a2aa3e7e6a57cda2e60dda3bc719388bba59551820dff2445caca9cf7c6a", "source_sha256": "638053737df40babfd92ea956f1083964cfe38bf547b96334228a8f67b2cb78b", "wav_sha256": "cf42c27ee1f2f8a74c2815e7f8f0ba01f79488df60e88267cb947da17f444a20", "local_source": "audio/librispeech_3575-170457-0054.flac", "local_wav": "audio/librispeech_3575-170457-0054.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3729-6852-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_3729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3729-6852-0000", "transcript": "TO CELEBRATE THE ARRIVAL OF HER SON SILVIA GAVE A SPLENDID SUPPER TO WHICH SHE HAD INVITED ALL HER RELATIVES AND IT WAS A GOOD OPPORTUNITY FOR ME TO MAKE THEIR ACQUAINTANCE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3729/6852/3729-6852-0000.flac"}, "sample_rate": 16000, "source_samples": 178720, "source_duration_seconds": 11.17, "source_peak": 0.402435302734375, "source_rms_dbfs": -26.96578233204508, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.348026827773285, "decoded_pcm_sha256": "e1524c8eb759928cd93fd86e98694c9a5efccdfa2a0fd948935f558dbffb8843", "source_sha256": "3fa96f8e0276d79ec027bb1595215a8bbeb488cefcc1702ff798e91ec8504403", "wav_sha256": "f7b24fd424e1ff2458c9d5639915bae0d1ca941e15f073571afd8ddaf9139fe0", "local_source": "audio/librispeech_3729-6852-0000.flac", "local_wav": "audio/librispeech_3729-6852-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3729-6852-0007", "corpus": "librispeech_test_clean", "speaker": "librispeech_3729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3729-6852-0007", "transcript": "TWO YEARS BEFORE HER DEATH I SAW HER PERFORM THE CHARACTER OF MARIANNE IN THE COMEDY OF MARIVAUX AND IN SPITE OF HER AGE AND DECLINING HEALTH THE ILLUSION WAS COMPLETE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3729/6852/3729-6852-0007.flac"}, "sample_rate": 16000, "source_samples": 199840, "source_duration_seconds": 12.49, "source_peak": 0.485931396484375, "source_rms_dbfs": -28.592297306892682, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.01965963933016, "decoded_pcm_sha256": "d081e3d6cdc89188a9f3d229a815f9784efd7e72dfcfa5317b4ac90fa9d86ff1", "source_sha256": "9d9bbbeeff1bfb647e2698db36f186cb439a8b2894772ca3cdfe9425c3e55a1a", "wav_sha256": "96a00b5f0bef082a0cd43dbeb530fadd34a0cabe94dc7b7c87521a64a2ded2bc", "local_source": "audio/librispeech_3729-6852-0007.flac", "local_wav": "audio/librispeech_3729-6852-0007.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_3729-6852-0017", "corpus": "librispeech_test_clean", "speaker": "librispeech_3729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "3729-6852-0017", "transcript": "I SEE A QUANTITY OF CHAIRS FOR HIRE AT THE RATE OF ONE SOU MEN READING THE NEWSPAPER UNDER THE SHADE OF THE TREES GIRLS AND MEN BREAKFASTING EITHER ALONE OR IN COMPANY WAITERS WHO WERE RAPIDLY GOING UP AND DOWN A NARROW STAIRCASE HIDDEN UNDER THE FOLIAGE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/3729/6852/3729-6852-0017.flac"}, "sample_rate": 16000, "source_samples": 287840, "source_duration_seconds": 17.99, "source_peak": 0.384185791015625, "source_rms_dbfs": -28.259280515739242, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.46214544153569, "decoded_pcm_sha256": "ba156f183286a1f4044b0b7794b3982603585d0e122bdb35978c337419392c08", "source_sha256": "12a9659e29190ce17dd89b5209d4ccaebc746bfb35093185d5970a91878e42b9", "wav_sha256": "73c66dd96fb1be6aa0bf20f4b8915715ca5a0ae34cdfa961c9ca3e910cbcba0f", "local_source": "audio/librispeech_3729-6852-0017.flac", "local_wav": "audio/librispeech_3729-6852-0017.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4077-13751-0001", "corpus": "librispeech_test_clean", "speaker": "librispeech_4077", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "4077-13751-0001", "transcript": "ITS ORIGIN WAS SMALL A GERM AN INSIGNIFICANT SEED HARDLY TO BE THOUGHT OF AS LIKELY TO AROUSE OPPOSITION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4077/13751/4077-13751-0001.flac"}, "sample_rate": 16000, "source_samples": 139920, "source_duration_seconds": 8.745, "source_peak": 0.941619873046875, "source_rms_dbfs": -23.904775017352993, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.869637117226276, "decoded_pcm_sha256": "ff5a7ca9328a329a2ca9ba953b0550653b86fe7bbf7bda2cae73908a1fbfde2a", "source_sha256": "1f85c5ad8dfb8fcf3c8d86e54afc055f756f147e7b426ff8b7f48780d8d43c0e", "wav_sha256": "67df7354e652c45bc30afaf081016690e05f0fc4c9c513350071df1b7f79cd44", "local_source": "audio/librispeech_4077-13751-0001.flac", "local_wav": "audio/librispeech_4077-13751-0001.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4077-13754-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_4077", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "4077-13754-0002", "transcript": "IT WAS THROUGH FLOYD'S ADVICE THAT BUCHANAN ORDERED THE MILITARY EXPEDITION TO UTAH OSTENSIBLY TO INSTALL CERTAIN FEDERAL OFFICIALS AND TO REPRESS AN ALLEGED INFANTILE REBELLION WHICH IN FACT HAD NEVER COME INTO EXISTENCE BUT IN REALITY TO FURTHER THE INTERESTS OF THE SECESSIONISTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4077/13754/4077-13754-0002.flac"}, "sample_rate": 16000, "source_samples": 284480, "source_duration_seconds": 17.78, "source_peak": 0.79345703125, "source_rms_dbfs": -25.576833348040672, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.41752474008618, "decoded_pcm_sha256": "e4ef84f14b8958b7a78512946198f223e31d6190beadfe364d4772e4a9963ced", "source_sha256": "58ff360d46e4fc6bcaf04da1a6ead97dc191c54426528cfa13efe7e780349ff5", "wav_sha256": "a850b52a73c2375cffe72a0ad6d0aacc3e356a76d7e143a5ae32b5d2c6aacec0", "local_source": "audio/librispeech_4077-13754-0002.flac", "local_wav": "audio/librispeech_4077-13754-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4077-13754-0012", "corpus": "librispeech_test_clean", "speaker": "librispeech_4077", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "4077-13754-0012", "transcript": "THIS MEANT THAT FOR AN ALLEGED MISDEMEANOR FOR WHICH CONGRESS PRESCRIBED A MAXIMUM PENALTY OF SIX MONTHS IMPRISONMENT AND A FINE OF THREE HUNDRED DOLLARS A MAN MIGHT BE IMPRISONED FOR LIFE AYE FOR MANY TERMS OF A MAN'S NATURAL LIFE DID THE COURT'S POWER TO ENFORCE ITS SENTENCES EXTEND SO FAR AND MIGHT BE FINED MILLIONS OF DOLLARS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4077/13754/4077-13754-0012.flac"}, "sample_rate": 16000, "source_samples": 353520, "source_duration_seconds": 22.095, "source_peak": 0.543731689453125, "source_rms_dbfs": -27.828520650204958, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.244250003732674, "decoded_pcm_sha256": "43e800e9f69ae92ed54ca36d54a78bd7c56d8ff0657f75ee376d88adab6b3764", "source_sha256": "65e48d9b9257be487e80c6938e34856d742870d390ad5d3d1b04d5f95f955d4a", "wav_sha256": "76f48fb7f3a6036c409df2023e580ecd2152586e46206da6014fee596e331fec", "local_source": "audio/librispeech_4077-13754-0012.flac", "local_wav": "audio/librispeech_4077-13754-0012.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4446-2271-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_4446", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4446-2271-0003", "transcript": "IT'S BEEN ON ONLY TWO WEEKS AND I'VE BEEN HALF A DOZEN TIMES ALREADY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4446/2271/4446-2271-0003.flac"}, "sample_rate": 16000, "source_samples": 59200, "source_duration_seconds": 3.7, "source_peak": 0.517730712890625, "source_rms_dbfs": -22.567726927886582, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.567726927886582, "decoded_pcm_sha256": "257baf7b0200414f4ef5d6ab5db2362889681a1c0de114633c54f50f9b628549", "source_sha256": "b8fecefeca8a7ae57ad53900dbba581245a09ad79d46ae1fc06140e0e5a5b281", "wav_sha256": "3d19719ae977a823291a0e50648779b441f15059218e7de8cff1ee0b8844309c", "local_source": "audio/librispeech_4446-2271-0003.flac", "local_wav": "audio/librispeech_4446-2271-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 59200}, {"id": "librispeech_4446-2271-0022", "corpus": "librispeech_test_clean", "speaker": "librispeech_4446", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4446-2271-0022", "transcript": "I'M GLAD SHE'S HELD HER OWN SINCE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4446/2271/4446-2271-0022.flac"}, "sample_rate": 16000, "source_samples": 37200, "source_duration_seconds": 2.325, "source_peak": 0.45782470703125, "source_rms_dbfs": -23.848490648398872, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.848490648398872, "decoded_pcm_sha256": "36c74fdf4f12eb94aa845b16c467769326a4db667ab611a8845b7f72c2cae5b1", "source_sha256": "ed740f49d11cc5c58f71f1925854cc9c1f30170b81143873aa8c58cd35430583", "wav_sha256": "0c393b4c2613903f50dbc975ca65d846b232b63f03096066c97bab44c6558f68", "local_source": "audio/librispeech_4446-2271-0022.flac", "local_wav": "audio/librispeech_4446-2271-0022.wav", "excerpt_start_sample": 0, "excerpt_samples": 37200}, {"id": "librispeech_4446-2273-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_4446", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4446-2273-0010", "transcript": "THERE WAS WATERCRESS SOUP AND SOLE AND A DELIGHTFUL OMELETTE STUFFED WITH MUSHROOMS AND TRUFFLES AND TWO SMALL RARE DUCKLINGS AND ARTICHOKES AND A DRY YELLOW RHONE WINE OF WHICH BARTLEY HAD ALWAYS BEEN VERY FOND", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4446/2273/4446-2273-0010.flac"}, "sample_rate": 16000, "source_samples": 199040, "source_duration_seconds": 12.44, "source_peak": 0.4254150390625, "source_rms_dbfs": -27.512108974962025, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.66417947925898, "decoded_pcm_sha256": "263c284aa66f0a837f0cdfcf012d15e0d4edfeb7a8a0717217e17395942a83fd", "source_sha256": "ee6acd6b5c33e2d7e114405b48aec2700f720d68dd8d59676172e81ca63317dc", "wav_sha256": "f67f7cae4b9d9ff9f64d921650ca4ba0ed2375ba7bc5d28c5460713d0e6548df", "local_source": "audio/librispeech_4446-2273-0010.flac", "local_wav": "audio/librispeech_4446-2273-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4507-16021-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_4507", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4507-16021-0016", "transcript": "CAN ONE IMAGINE A NATURALIST REFUSING TO STUDY THE VIPER THE BAT THE SCORPION THE CENTIPEDE THE TARANTULA AND ONE WHO WOULD CAST THEM BACK INTO THEIR DARKNESS SAYING OH HOW UGLY THAT IS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4507/16021/4507-16021-0016.flac"}, "sample_rate": 16000, "source_samples": 276240, "source_duration_seconds": 17.265, "source_peak": 0.553863525390625, "source_rms_dbfs": -26.07557497706036, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.407502540486163, "decoded_pcm_sha256": "0f759c362b7bb3f7ab6547020834b6f41d471a7cfe5c2018d0a8cc506e042103", "source_sha256": "e3cf855f3ed817d96563b7914780df19ad9175cd7deb4a04880dfb6093dcccf7", "wav_sha256": "97b09a28609b066912fc01e5a4c462962bf4af014153f52ff0be247c7b7efa39", "local_source": "audio/librispeech_4507-16021-0016.flac", "local_wav": "audio/librispeech_4507-16021-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4507-16021-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_4507", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4507-16021-0032", "transcript": "HE MUST DESCEND WITH HIS HEART FULL OF CHARITY AND SEVERITY AT THE SAME TIME AS A BROTHER AND AS A JUDGE TO THOSE IMPENETRABLE CASEMATES WHERE CRAWL PELL MELL THOSE WHO BLEED AND THOSE WHO DEAL THE BLOW THOSE WHO WEEP AND THOSE WHO CURSE THOSE WHO FAST AND THOSE WHO DEVOUR THOSE WHO ENDURE EVIL AND THOSE WHO INFLICT IT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4507/16021/4507-16021-0032.flac"}, "sample_rate": 16000, "source_samples": 449440, "source_duration_seconds": 28.09, "source_peak": 0.4244384765625, "source_rms_dbfs": -25.686727205489255, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.511214264972427, "decoded_pcm_sha256": "7576336f86f07936b03cf91b64bf80c306a8a013d42a9e0eb9b1dae02be902c2", "source_sha256": "cf872613e0d590501ff569c17e5f9d525920c4babeda14ac94577f60cfc54401", "wav_sha256": "56e8dbd76689127c0d2f1fd91f5b1b06d839e01bac6a98198046d6714ab13c63", "local_source": "audio/librispeech_4507-16021-0032.flac", "local_wav": "audio/librispeech_4507-16021-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4507-16021-0045", "corpus": "librispeech_test_clean", "speaker": "librispeech_4507", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4507-16021-0045", "transcript": "IT IS SO MADE THAT EVERYWHERE WE FEEL THE SENSE OF PUNISHMENT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4507/16021/4507-16021-0045.flac"}, "sample_rate": 16000, "source_samples": 78240, "source_duration_seconds": 4.89, "source_peak": 0.550750732421875, "source_rms_dbfs": -25.788181877969095, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.788181877969095, "decoded_pcm_sha256": "ef91d0d7b9ebfc4c08837a0dcebf79c72e3e4ff97d381f356a678c2a085d6e6c", "source_sha256": "86e41fe1ab9a5fdf5d806ab2728428f4b8d2720df04ce6431b65f617878b924e", "wav_sha256": "52e640aa0a3813985286616d54082fb10048f8acae7858ee0fdbef99208e9f66", "local_source": "audio/librispeech_4507-16021-0045.flac", "local_wav": "audio/librispeech_4507-16021-0045.wav", "excerpt_start_sample": 0, "excerpt_samples": 78240}, {"id": "librispeech_4970-29093-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_4970", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4970-29093-0004", "transcript": "HE WAS UNABLE TO DECIDE EXACTLY WHAT IT SHOULD BE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4970/29093/4970-29093-0004.flac"}, "sample_rate": 16000, "source_samples": 60000, "source_duration_seconds": 3.75, "source_peak": 0.652984619140625, "source_rms_dbfs": -25.09401331193901, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.09401331193901, "decoded_pcm_sha256": "222e257a563c92e3e220ca509960e218a09ffd8f995b3f1148bba432a24fc5cf", "source_sha256": "48c89f99c44d1be73c92460e6d61d4c1090efe126701f7492184112d37575229", "wav_sha256": "c1ad0bc6656e5a37b40d868e2b1df01441c4a9fbd5afe93f389122fc10be9f4d", "local_source": "audio/librispeech_4970-29093-0004.flac", "local_wav": "audio/librispeech_4970-29093-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 60000}, {"id": "librispeech_4970-29093-0005", "corpus": "librispeech_test_clean", "speaker": "librispeech_4970", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4970-29093-0005", "transcript": "SOMETIMES HE THOUGHT HE WOULD LIKE TO STAND IN A CONSPICUOUS PULPIT AND HUMBLY PREACH THE GOSPEL OF REPENTANCE AND IT EVEN CROSSED HIS MIND THAT IT WOULD BE NOBLE TO GIVE HIMSELF TO A MISSIONARY LIFE TO SOME BENIGHTED REGION WHERE THE DATE PALM GROWS AND THE NIGHTINGALE'S VOICE IS IN TUNE AND THE BUL BUL SINGS ON THE OFF NIGHTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4970/29093/4970-29093-0005.flac"}, "sample_rate": 16000, "source_samples": 354560, "source_duration_seconds": 22.16, "source_peak": 0.676025390625, "source_rms_dbfs": -25.92844507815476, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.632807866301704, "decoded_pcm_sha256": "e5b971d7dd1c4efc72382481eb6d7cbe047355a62af1c39714d28b6cb739b74e", "source_sha256": "f466147cca3169748874913da56b587fd3d7fed00a285a1c9124dac56a8323ba", "wav_sha256": "37a020e1d205b7fb675b2ff89e777850155dfad1f71f9d58087cea696b62bd39", "local_source": "audio/librispeech_4970-29093-0005.flac", "local_wav": "audio/librispeech_4970-29093-0005.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_4970-29095-0008", "corpus": "librispeech_test_clean", "speaker": "librispeech_4970", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4970-29095-0008", "transcript": "MOTHER I'M GOING TO STUDY MEDICINE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4970/29095/4970-29095-0008.flac"}, "sample_rate": 16000, "source_samples": 48640, "source_duration_seconds": 3.04, "source_peak": 0.34185791015625, "source_rms_dbfs": -28.408655353194106, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.408655353194106, "decoded_pcm_sha256": "44ec7bab382a51da3a9f07cbc8ffdb491029ea5b6b84d63bfceb71c60641803b", "source_sha256": "cd2a3b352b93beb33caced744406acf8dace3f1bbb5fc63c8feb4c8a0ed55bb4", "wav_sha256": "1dc0a4e22ebfc2ff3d5048978b903074647fdb65e2c0030918f94bccfe72e63f", "local_source": "audio/librispeech_4970-29095-0008.flac", "local_wav": "audio/librispeech_4970-29095-0008.wav", "excerpt_start_sample": 0, "excerpt_samples": 48640}, {"id": "librispeech_4992-23283-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_4992", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4992-23283-0016", "transcript": "AGAIN HE SEARCHED HIS OWN THOUGHTS NOR INEFFECTUALLY AS BEFORE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4992/23283/4992-23283-0016.flac"}, "sample_rate": 16000, "source_samples": 71920, "source_duration_seconds": 4.495, "source_peak": 0.317962646484375, "source_rms_dbfs": -28.34849352834874, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.34849352834874, "decoded_pcm_sha256": "4af74b1b8682fb324d2e81cb2f0cc0dc50b77d4a3b44fd510cfb4df09ae57d6d", "source_sha256": "e7442a8f1bbd834ad37ee3c9cc5fa26b1cfb61446f827b54298716f59b010002", "wav_sha256": "e74da3b395ffef1f2fc80932d932ba92ecb3611d9ca1f06743f0a824ef2c9468", "local_source": "audio/librispeech_4992-23283-0016.flac", "local_wav": "audio/librispeech_4992-23283-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 71920}, {"id": "librispeech_4992-41797-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_4992", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4992-41797-0000", "transcript": "YES DEAD THESE FOUR YEARS AN A GOOD JOB FOR HER TOO", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4992/41797/4992-41797-0000.flac"}, "sample_rate": 16000, "source_samples": 87760, "source_duration_seconds": 5.485, "source_peak": 0.48004150390625, "source_rms_dbfs": -23.74627112170186, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.74627112170186, "decoded_pcm_sha256": "70049242cb761e6a91df355dd7a9dbdfb974c89071ee6b5dd83506a6ce8390e3", "source_sha256": "f7dbbd42eb06171dccf937d53a6c9144e51b88ea5b1b2ee69978900b87216fb5", "wav_sha256": "d8a54528a66df71baf1fe8b7d60083abd4034bbb2cb4331344d960b542806bbc", "local_source": "audio/librispeech_4992-41797-0000.flac", "local_wav": "audio/librispeech_4992-41797-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 87760}, {"id": "librispeech_4992-41806-0000", "corpus": "librispeech_test_clean", "speaker": "librispeech_4992", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "4992-41806-0000", "transcript": "NATTY HARMON TRIED THE KITCHEN PUMP SECRETLY SEVERAL TIMES DURING THE EVENING FOR THE WATER HAD TO RUN UP HILL ALL THE WAY FROM THE WELL TO THE KITCHEN SINK AND HE BELIEVED THIS TO BE A CONTINUAL MIRACLE THAT MIGHT GIVE OUT AT ANY MOMENT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/4992/41806/4992-41806-0000.flac"}, "sample_rate": 16000, "source_samples": 229760, "source_duration_seconds": 14.36, "source_peak": 0.582489013671875, "source_rms_dbfs": -21.657320978064327, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.055751743922045, "decoded_pcm_sha256": "9e17e42f2de437c240d5850615912f710137bb8bb0f2e42082d93bc758515234", "source_sha256": "ce4dd04d130b3b575f1826c6505b5519c09a7301eee9b76e58eeb1f591131035", "wav_sha256": "04109bfdf70350f438b7afe006744ec48bab6fdb23019457ec981bb2cb1518fb", "local_source": "audio/librispeech_4992-41806-0000.flac", "local_wav": "audio/librispeech_4992-41806-0000.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5105-28240-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_5105", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5105-28240-0011", "transcript": "I LEFT YOU ON A CONTINENT AND HERE I HAVE THE HONOR OF FINDING YOU ON AN ISLAND", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5105/28240/5105-28240-0011.flac"}, "sample_rate": 16000, "source_samples": 96320, "source_duration_seconds": 6.02, "source_peak": 0.599212646484375, "source_rms_dbfs": -24.990381079106882, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.975943389244275, "decoded_pcm_sha256": "b9bc3828aa0284c0660c9ec7acd1980026fcc21a26b62a3c5c68458711b73ba2", "source_sha256": "2e162ff81e45f909e947ebfae4f39e339d73d5fc19afafb1c903db96f7f8532d", "wav_sha256": "94233ddbd1f16127374435a042a6df2adeba0af01cb759c656e84cd9293ac5ac", "local_source": "audio/librispeech_5105-28240-0011.flac", "local_wav": "audio/librispeech_5105-28240-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5105-28241-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_5105", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5105-28241-0004", "transcript": "ALTHOUGH ONLY A MODERATE BREEZE WAS BLOWING THE SEA WAS ROUGH A CIRCUMSTANCE TO BE ACCOUNTED FOR ONLY BY THE DIMINUTION IN THE FORCE OF THE EARTH'S ATTRACTION RENDERING THE LIQUID PARTICLES SO BUOYANT THAT BY THE MERE EFFECT OF OSCILLATION THEY WERE CARRIED TO A HEIGHT THAT WAS QUITE UNPRECEDENTED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5105/28241/5105-28241-0004.flac"}, "sample_rate": 16000, "source_samples": 278640, "source_duration_seconds": 17.415, "source_peak": 0.83734130859375, "source_rms_dbfs": -26.537292077408626, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.37366219728025, "decoded_pcm_sha256": "6704bc85da6844b255174407f722499ab1188299d4775a37e25cc238c7fb903e", "source_sha256": "70c67685f8e24f7a27f6de3b63d5b3ecbde28f017d1b7aac9304f6c6034fa1bc", "wav_sha256": "4497c937acc206dd49bddec67200ad51f5097466123879a49f6a3ee336f15a69", "local_source": "audio/librispeech_5105-28241-0004.flac", "local_wav": "audio/librispeech_5105-28241-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5105-28241-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_5105", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5105-28241-0019", "transcript": "NOTHING WAS TO BE DONE BUT TO PUT ABOUT AND RETURN IN DISAPPOINTMENT TOWARDS THE NORTH", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5105/28241/5105-28241-0019.flac"}, "sample_rate": 16000, "source_samples": 84640, "source_duration_seconds": 5.29, "source_peak": 0.58538818359375, "source_rms_dbfs": -29.753814832387818, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.753814832387818, "decoded_pcm_sha256": "3d0b0ee3168cffdc7dff07188417fdc9c6c2c31369b906ef6a97ab560d3e8f4b", "source_sha256": "96d523790a28d90ffd89ea67603421a3881a96d21cb523495d1a066c4c5ceecd", "wav_sha256": "9c73b0201f91008baedeb8630a30a0d9ff83ad92b23f14df405ae17c8cd5e664", "local_source": "audio/librispeech_5105-28241-0019.flac", "local_wav": "audio/librispeech_5105-28241-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 84640}, {"id": "librispeech_5142-33396-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_5142", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5142-33396-0019", "transcript": "OH IT IS BETTER TO LIVE ON THE SEA AND LET OTHER MEN RAISE YOUR CROPS AND COOK YOUR MEALS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5142/33396/5142-33396-0019.flac"}, "sample_rate": 16000, "source_samples": 79760, "source_duration_seconds": 4.985, "source_peak": 0.376739501953125, "source_rms_dbfs": -26.870219951708695, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.870219951708695, "decoded_pcm_sha256": "f3e12cfe88aca2fb306552ad3fe2c213b91a4bfc2f0aa979df8a1f01006ba85d", "source_sha256": "2025e96df477d9ebc46b91bae88c601221bbbb2ab12d9fa83ceced80b0f9b72f", "wav_sha256": "20b7da9ef469c7b32d3ebf290f98c5e8058c718ee8ccd7d3f172ccc870264c9a", "local_source": "audio/librispeech_5142-33396-0019.flac", "local_wav": "audio/librispeech_5142-33396-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 79760}, {"id": "librispeech_5142-33396-0030", "corpus": "librispeech_test_clean", "speaker": "librispeech_5142", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5142-33396-0030", "transcript": "THE THRALLS WERE BRINGING IN A GREAT POT OF MEAT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5142/33396/5142-33396-0030.flac"}, "sample_rate": 16000, "source_samples": 44240, "source_duration_seconds": 2.765, "source_peak": 0.41986083984375, "source_rms_dbfs": -28.781219695797876, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.781219695797876, "decoded_pcm_sha256": "cdc8b64c805f94ff6dd7f431b45e924a4edf992efb5d08cda66456142b76481e", "source_sha256": "87b13e7bd9a84ee5f9ba237f84dc992d0fbbad061a6df1750b87a47756642b33", "wav_sha256": "8732110f0034b47a150b49823c48d798c5010d9ea27a2236efaa6871fd045d5b", "local_source": "audio/librispeech_5142-33396-0030.flac", "local_wav": "audio/librispeech_5142-33396-0030.wav", "excerpt_start_sample": 0, "excerpt_samples": 44240}, {"id": "librispeech_5142-33396-0065", "corpus": "librispeech_test_clean", "speaker": "librispeech_5142", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5142-33396-0065", "transcript": "SOFT HEART HE SAID GENTLY TO HER THEN TO THORKEL WELL LET HIM GO THORKEL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5142/33396/5142-33396-0065.flac"}, "sample_rate": 16000, "source_samples": 83120, "source_duration_seconds": 5.195, "source_peak": 0.5650634765625, "source_rms_dbfs": -29.734936973090612, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.734936973090612, "decoded_pcm_sha256": "0b650128a600d9d1c16e5bf3b64c63018c80e6b3463b2d6fb23e3aead6248f75", "source_sha256": "67800641844bf8c9d654fc9a6099941ba97ff31b0ac0a579209378598c4ea09b", "wav_sha256": "0aace7ee3326bd6ad0ef18685074f6bd6f6749bd1b9c5fd08080a225d32a3ed8", "local_source": "audio/librispeech_5142-33396-0065.flac", "local_wav": "audio/librispeech_5142-33396-0065.wav", "excerpt_start_sample": 0, "excerpt_samples": 83120}, {"id": "librispeech_5639-40744-0013", "corpus": "librispeech_test_clean", "speaker": "librispeech_5639", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5639-40744-0013", "transcript": "SHE SAW THAT THE BED WAS GILDED AND SO RICH THAT IT SEEMED THAT OF A PRINCE RATHER THAN OF A PRIVATE GENTLEMAN", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5639/40744/5639-40744-0013.flac"}, "sample_rate": 16000, "source_samples": 109840, "source_duration_seconds": 6.865, "source_peak": 0.287109375, "source_rms_dbfs": -32.469388774245715, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -32.072413641742806, "decoded_pcm_sha256": "ed7f1ce67f421ec6afa8566971a1c9c8ec8d332f4950f7baa3aacb009b96f5ef", "source_sha256": "a2688cb66fd448ee0d578c6d82312afea0f9ad95b24350f251eb79ca5a4f759e", "wav_sha256": "57e719154dbc8c577990ab1a3b41317f9774b3a08c636666478a6d070c102aa9", "local_source": "audio/librispeech_5639-40744-0013.flac", "local_wav": "audio/librispeech_5639-40744-0013.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5639-40744-0023", "corpus": "librispeech_test_clean", "speaker": "librispeech_5639", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5639-40744-0023", "transcript": "WHEN THE BOY WALKED THROUGH THE STREETS BLESSINGS WERE SHOWERED UPON HIM BY ALL WHO SAW HIM BLESSINGS UPON HIS BEAUTY UPON THE MOTHER THAT BORE HIM UPON THE FATHER THAT BEGOT HIM UPON THOSE WHO BROUGHT HIM UP SO WELL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5639/40744/5639-40744-0023.flac"}, "sample_rate": 16000, "source_samples": 222720, "source_duration_seconds": 13.92, "source_peak": 0.2412109375, "source_rms_dbfs": -30.75070314546162, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.202866501530522, "decoded_pcm_sha256": "da6b45c38da036a67d46d62fc3e6e54310d46e3e14d8f0a3eff1b445bb1755f4", "source_sha256": "dd277d4b42ef738b14b59fa1d835988e71ec031e2f01d024279202edc26ad9c0", "wav_sha256": "3de21cbccb1dbd2ff7b98796f239d0fb4ae4abe216c2f77e5a5d068ec8ed4d89", "local_source": "audio/librispeech_5639-40744-0023.flac", "local_wav": "audio/librispeech_5639-40744-0023.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5639-40744-0030", "corpus": "librispeech_test_clean", "speaker": "librispeech_5639", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "5639-40744-0030", "transcript": "JUST THEN LEOCADIA CAME TO HERSELF AND EMBRACING THE CROSS SEEMED CHANGED INTO A SEA OF TEARS AND THE GENTLEMAN REMAINED IN UTTER BEWILDERMENT UNTIL HIS WIFE HAD REPEATED TO HIM FROM BEGINNING TO END LEOCADIA'S WHOLE STORY AND HE BELIEVED IT THROUGH THE BLESSED DISPENSATION OF HEAVEN WHICH HAD CONFIRMED IT BY SO MANY CONVINCING TESTIMONIES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5639/40744/5639-40744-0030.flac"}, "sample_rate": 16000, "source_samples": 356560, "source_duration_seconds": 22.285, "source_peak": 0.222198486328125, "source_rms_dbfs": -34.05232832381272, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -33.056314847596, "decoded_pcm_sha256": "76c45c30d066507ae3cbb52fafb188ebe36da864c7c54942db2ea8356726ce3a", "source_sha256": "c7f95b2f012ab8f4629d15c8c78e78bf17b3445c96e5a9b0c525b0743e50d9c5", "wav_sha256": "257fa9e1dafacf00ce9d6925632b4e3904f46e59298d19f84b3d14211b4df0a2", "local_source": "audio/librispeech_5639-40744-0030.flac", "local_wav": "audio/librispeech_5639-40744-0030.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_5683-32865-0017", "corpus": "librispeech_test_clean", "speaker": "librispeech_5683", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5683-32865-0017", "transcript": "ALL THE TIME HE WAS TALKING TO ME HIS ANGRY LITTLE EYES WERE FOLLOWING LAKE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5683/32865/5683-32865-0017.flac"}, "sample_rate": 16000, "source_samples": 87280, "source_duration_seconds": 5.455, "source_peak": 0.2510986328125, "source_rms_dbfs": -34.9940582978409, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -34.9940582978409, "decoded_pcm_sha256": "44477e160959a9526d3545567baa9b096405cf9d914595d737ffdf3d46ce1e6f", "source_sha256": "d0027b49e5e0c29b3b59d1d782b8d4d9753f90e4343f243a15698e787c8e64ca", "wav_sha256": "bfb05fe49b3082f400d278836c3aff46c9a56c7d7919fbc0de59a5c012bbdb55", "local_source": "audio/librispeech_5683-32865-0017.flac", "local_wav": "audio/librispeech_5683-32865-0017.wav", "excerpt_start_sample": 0, "excerpt_samples": 87280}, {"id": "librispeech_5683-32879-0009", "corpus": "librispeech_test_clean", "speaker": "librispeech_5683", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5683-32879-0009", "transcript": "ILL AND TROUBLED DEAR TROUBLED IN MIND AND MISERABLY NERVOUS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5683/32879/5683-32879-0009.flac"}, "sample_rate": 16000, "source_samples": 82160, "source_duration_seconds": 5.135, "source_peak": 0.744171142578125, "source_rms_dbfs": -22.065935064466796, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.065935064466796, "decoded_pcm_sha256": "c0eff412ac130cf1ccdb29b7cbb8149ed148614513d6a355a1677d911d2c6b5f", "source_sha256": "c66196c0436cce295ac3ae7695f4904f87672f1d63d89b8810937a146378411a", "wav_sha256": "eea99813322332eff96b5f9c8124d990192b34c98b8b51879f7e89d7170803d7", "local_source": "audio/librispeech_5683-32879-0009.flac", "local_wav": "audio/librispeech_5683-32879-0009.wav", "excerpt_start_sample": 0, "excerpt_samples": 82160}, {"id": "librispeech_5683-32879-0019", "corpus": "librispeech_test_clean", "speaker": "librispeech_5683", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "5683-32879-0019", "transcript": "I HAVE VERY FEW TO LOVE ME NOW AND I THOUGHT YOU MIGHT LOVE ME AS I HAVE BEGUN TO LOVE YOU", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/5683/32879/5683-32879-0019.flac"}, "sample_rate": 16000, "source_samples": 101600, "source_duration_seconds": 6.35, "source_peak": 0.799591064453125, "source_rms_dbfs": -22.192801992464837, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.95646015705721, "decoded_pcm_sha256": "cf57296e7013c02425f574c6c3f30623e8057ab1e199fc8fcd761f1b2ff63d52", "source_sha256": "0c9375f8c28b1576aaa73e07006a7fc838f49b250df910885e24a16abef0c19f", "wav_sha256": "c8d43ac515baa4c0d047816bffd947306c829e56c1cbd3aea016eb764930c45d", "local_source": "audio/librispeech_5683-32879-0019.flac", "local_wav": "audio/librispeech_5683-32879-0019.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_61-70968-0042", "corpus": "librispeech_test_clean", "speaker": "librispeech_61", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "61-70968-0042", "transcript": "MONTFICHET CALLED OUT FOR ROBIN TO GIVE HIM AN ARM", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/61/70968/61-70968-0042.flac"}, "sample_rate": 16000, "source_samples": 44560, "source_duration_seconds": 2.785, "source_peak": 0.362152099609375, "source_rms_dbfs": -27.019043277748153, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.019043277748153, "decoded_pcm_sha256": "4d1c295afd61c34ca0f59a70b2baace0777ff6e94127ed717b67dc664c9d9ee4", "source_sha256": "9aa0aa0746140383c3458b53991ccd9ebfc2c0ded6c6817ec0748aa70e6b8088", "wav_sha256": "6341417e6e575c7204a790e5368cdbc7d10afd5dce8bd9816b8670f57a6efeb1", "local_source": "audio/librispeech_61-70968-0042.flac", "local_wav": "audio/librispeech_61-70968-0042.wav", "excerpt_start_sample": 0, "excerpt_samples": 44560}, {"id": "librispeech_61-70968-0043", "corpus": "librispeech_test_clean", "speaker": "librispeech_61", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "61-70968-0043", "transcript": "FRIENDS SAID MONTFICHET FAINTLY TO THE WRESTLERS BEAR US ESCORT SO FAR AS THE SHERIFF'S HOUSE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/61/70968/61-70968-0043.flac"}, "sample_rate": 16000, "source_samples": 107760, "source_duration_seconds": 6.735, "source_peak": 0.53997802734375, "source_rms_dbfs": -27.886439724320226, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.565805365496097, "decoded_pcm_sha256": "1731c4205d3ab140aea21c6e2df47e27d59a45b1eaa6b80c407cddcad99de7af", "source_sha256": "92cde1bfb6bab32cd1428505868f1ca27db04ed17bec9e47e7cbb24e34dc09c3", "wav_sha256": "86ec6218c4ceffd4402d269164b03d365c360d751eb3b8e6078808e4ace33916", "local_source": "audio/librispeech_61-70968-0043.flac", "local_wav": "audio/librispeech_61-70968-0043.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_61-70968-0054", "corpus": "librispeech_test_clean", "speaker": "librispeech_61", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "61-70968-0054", "transcript": "MISTRESS FITZOOTH HAD BEEN CARRIED OFF BY THE SHERIFF'S DAUGHTER AND HER MAIDS AS SOON AS THEY HAD ENTERED THE HOUSE SO THAT ROBIN ALONE HAD THE CARE OF MONTFICHET", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/61/70968/61-70968-0054.flac"}, "sample_rate": 16000, "source_samples": 125760, "source_duration_seconds": 7.86, "source_peak": 0.413848876953125, "source_rms_dbfs": -26.416451889769732, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.102361433729442, "decoded_pcm_sha256": "0c30dfc3b23081f52641ba6b1793c8bc5c9ea236aea529b533de133f00169e95", "source_sha256": "7b58da65dd63a6b09d43ee20c6aa4b5a8853199af7c350a87ab692fb4d0d874c", "wav_sha256": "6650129147c01f0d91d5e0a36f5c40ce37be0edd09c841daac604d1bf7ecfa24", "local_source": "audio/librispeech_61-70968-0054.flac", "local_wav": "audio/librispeech_61-70968-0054.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_672-122797-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_672", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "672-122797-0003", "transcript": "BUT THIS WAS WHAT THE TREE COULD NOT BEAR TO HEAR", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/672/122797/672-122797-0003.flac"}, "sample_rate": 16000, "source_samples": 76160, "source_duration_seconds": 4.76, "source_peak": 0.368072509765625, "source_rms_dbfs": -30.00537875055086, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -30.00537875055086, "decoded_pcm_sha256": "4d9f339e6dbd2372c553d9d5f6718849a37e2770b2e7e3a43beccd5922c086ec", "source_sha256": "b00cbf4705fcd94f9ea9130da79993f83b7e3a83313328314b3492d19295af94", "wav_sha256": "939b4401889c28674fc8462e95e87c8aa498a702648053a96efaa49dd3a4d6e9", "local_source": "audio/librispeech_672-122797-0003.flac", "local_wav": "audio/librispeech_672-122797-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 76160}, {"id": "librispeech_672-122797-0044", "corpus": "librispeech_test_clean", "speaker": "librispeech_672", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "672-122797-0044", "transcript": "AND HE LEANED AGAINST THE WALL LOST IN REVERIE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/672/122797/672-122797-0044.flac"}, "sample_rate": 16000, "source_samples": 59840, "source_duration_seconds": 3.74, "source_peak": 0.291778564453125, "source_rms_dbfs": -28.28836846886545, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.28836846886545, "decoded_pcm_sha256": "432f9a1ae86511f6d52017a356d8d76c16009ab41a0b0e212b50e3c7fe63423f", "source_sha256": "30dfd88ccc8b641d250263094e829de48032160f0e8f954c0487a8ecd0da85cc", "wav_sha256": "e521e1a275330db01220cdbf937c18de9ae32a2869b60191d0e45f37d008553a", "local_source": "audio/librispeech_672-122797-0044.flac", "local_wav": "audio/librispeech_672-122797-0044.wav", "excerpt_start_sample": 0, "excerpt_samples": 59840}, {"id": "librispeech_672-122797-0056", "corpus": "librispeech_test_clean", "speaker": "librispeech_672", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "672-122797-0056", "transcript": "SAID THE FIR TREE THINKING OVER WHAT HE HAD HIMSELF RELATED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/672/122797/672-122797-0056.flac"}, "sample_rate": 16000, "source_samples": 83600, "source_duration_seconds": 5.225, "source_peak": 0.34185791015625, "source_rms_dbfs": -29.44078110552336, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.44078110552336, "decoded_pcm_sha256": "1ee86311a9a7b6ee7bf0ef8ccad1d6c7f831fa6001565687e3d9e0eb90cd44f4", "source_sha256": "e58678d63f84c218fb28da622351b90e344380dbfc13763dd1123e71faf708d1", "wav_sha256": "c2d2cf8f5b61062744b09b2ded2dd8e540b5984a424be0cef6ae7903b9072300", "local_source": "audio/librispeech_672-122797-0056.flac", "local_wav": "audio/librispeech_672-122797-0056.wav", "excerpt_start_sample": 0, "excerpt_samples": 83600}, {"id": "librispeech_6829-68769-0014", "corpus": "librispeech_test_clean", "speaker": "librispeech_6829", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "6829-68769-0014", "transcript": "THEY FOLLOWED THE JAILER ALONG A SUCCESSION OF PASSAGES", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6829/68769/6829-68769-0014.flac"}, "sample_rate": 16000, "source_samples": 58480, "source_duration_seconds": 3.655, "source_peak": 0.45867919921875, "source_rms_dbfs": -23.65210599160649, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.65210599160649, "decoded_pcm_sha256": "f8c838b07c2dd578b611a15c028e07c92a592c03766bbc5d9020f61b8ed5195e", "source_sha256": "c228d6cd297ffb4916b952d7d5861d8ac8225058219ec05b587d8dd4bd041179", "wav_sha256": "1297807e7c7832e33167b40d55c2795aab4db406c144ff8aa315638314ee3bf5", "local_source": "audio/librispeech_6829-68769-0014.flac", "local_wav": "audio/librispeech_6829-68769-0014.wav", "excerpt_start_sample": 0, "excerpt_samples": 58480}, {"id": "librispeech_6829-68769-0038", "corpus": "librispeech_test_clean", "speaker": "librispeech_6829", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "6829-68769-0038", "transcript": "AND IT RUINS A MAN'S DISPOSITION", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6829/68769/6829-68769-0038.flac"}, "sample_rate": 16000, "source_samples": 40080, "source_duration_seconds": 2.505, "source_peak": 0.4256591796875, "source_rms_dbfs": -24.099666920041205, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.099666920041205, "decoded_pcm_sha256": "2491d28a0087423aef64b260b9a3bef96a9ffd38bdd5460aeb1d8818a5a4e0a8", "source_sha256": "0f141cbbe137a2927bbf38106a3b14d2b2f186b86e1a54bcbd354d8c2a8981c4", "wav_sha256": "4f8dddd42e85f0a1e91a1b93df0c6a50013c6a166ff3a1bcdf0e54f9924f9ea1", "local_source": "audio/librispeech_6829-68769-0038.flac", "local_wav": "audio/librispeech_6829-68769-0038.wav", "excerpt_start_sample": 0, "excerpt_samples": 40080}, {"id": "librispeech_6829-68769-0041", "corpus": "librispeech_test_clean", "speaker": "librispeech_6829", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "6829-68769-0041", "transcript": "I'M NOT ELECTIONEERING JUST NOW", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6829/68769/6829-68769-0041.flac"}, "sample_rate": 16000, "source_samples": 42560, "source_duration_seconds": 2.66, "source_peak": 0.617095947265625, "source_rms_dbfs": -23.43903741251144, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.43903741251144, "decoded_pcm_sha256": "d14d8bc17d20f03134a1360b706d2d290d10240225382c5adf8b0919adb152e4", "source_sha256": "e0e3964567d7a515d2443c6e1eff105f88da0e87ef44cc2954b36f1b9d93096c", "wav_sha256": "f34de7e7dd5bb7962b7644864833e52a9a705d8b11000ee7070e9b63e310f6ad", "local_source": "audio/librispeech_6829-68769-0041.flac", "local_wav": "audio/librispeech_6829-68769-0041.wav", "excerpt_start_sample": 0, "excerpt_samples": 42560}, {"id": "librispeech_6930-75918-0013", "corpus": "librispeech_test_clean", "speaker": "librispeech_6930", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "6930-75918-0013", "transcript": "IN THOSE VERY TERMS I EVEN ADDED MORE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6930/75918/6930-75918-0013.flac"}, "sample_rate": 16000, "source_samples": 47040, "source_duration_seconds": 2.94, "source_peak": 0.408905029296875, "source_rms_dbfs": -29.974578166391865, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -29.974578166391865, "decoded_pcm_sha256": "3f54b1539585ee3c865737cd0781dd64a40895a5d1930b99c37db78e450618a7", "source_sha256": "83f99c357f446d1fddffc61c09f59eb8a41c82c5e4261db42c71396c537c409e", "wav_sha256": "6b56ade8142254ab2fddf538d8164b7043ee7a100b957e3f941f9e2a2de64b9c", "local_source": "audio/librispeech_6930-75918-0013.flac", "local_wav": "audio/librispeech_6930-75918-0013.wav", "excerpt_start_sample": 0, "excerpt_samples": 47040}, {"id": "librispeech_6930-76324-0024", "corpus": "librispeech_test_clean", "speaker": "librispeech_6930", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "6930-76324-0024", "transcript": "THEY SAY ILLUMINATION BY CANDLE LIGHT IS THE PRETTIEST IN THE WORLD", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6930/76324/6930-76324-0024.flac"}, "sample_rate": 16000, "source_samples": 64800, "source_duration_seconds": 4.05, "source_peak": 0.373291015625, "source_rms_dbfs": -28.918530059065088, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -28.918530059065088, "decoded_pcm_sha256": "d8dd1ce759e26c72605c90e837ae65e87caae1e3095da47a3f97063f55664b19", "source_sha256": "5d8bd7d094fe646881a4848026dbf1a3748ec90f170b87d2f0836adcfd9b7f7f", "wav_sha256": "894af1a21bcf4de58645a4cc3b9ec88c579dc75aef0c896f4aeacea6cdb9640d", "local_source": "audio/librispeech_6930-76324-0024.flac", "local_wav": "audio/librispeech_6930-76324-0024.wav", "excerpt_start_sample": 0, "excerpt_samples": 64800}, {"id": "librispeech_6930-81414-0026", "corpus": "librispeech_test_clean", "speaker": "librispeech_6930", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "6930-81414-0026", "transcript": "MY OVERWROUGHT NERVES YIELDED AT LAST", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/6930/81414/6930-81414-0026.flac"}, "sample_rate": 16000, "source_samples": 49200, "source_duration_seconds": 3.075, "source_peak": 0.434173583984375, "source_rms_dbfs": -27.195611560424933, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.195611560424933, "decoded_pcm_sha256": "6ea614ae159242a4cb64e926932b9c1cf0c8b84fb4912932e606a31b6d48c909", "source_sha256": "84048bb63a1968e9856f38f430d293fd5036cc1b8c31312f3aff96d087732145", "wav_sha256": "8e930ede632ab23724d64be8cf34e2f53d115e42f4dbb2c644589c9510008d66", "local_source": "audio/librispeech_6930-81414-0026.flac", "local_wav": "audio/librispeech_6930-81414-0026.wav", "excerpt_start_sample": 0, "excerpt_samples": 49200}, {"id": "librispeech_7021-79740-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_7021", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7021-79740-0002", "transcript": "NOW DELIA CONTRIVED TO OBTAIN A GREAT INFLUENCE AND ASCENDENCY OVER THE MINDS OF THE CHILDREN BY MEANS OF THESE DOLLS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7021/79740/7021-79740-0002.flac"}, "sample_rate": 16000, "source_samples": 147600, "source_duration_seconds": 9.225, "source_peak": 0.571044921875, "source_rms_dbfs": -24.145550365790243, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.686648589877173, "decoded_pcm_sha256": "32223cc598805a0301d5fd306db57f06a7d2c624ffce3ec523d6a45df0f1b192", "source_sha256": "df09e0bfaf8000a8183b367c2fbc6ce3404f4152ef14d9340515b37edb7155ba", "wav_sha256": "15dbc2fc45c2fb603bb8186e866988f9615538186eb94834e17d09a846b828ef", "local_source": "audio/librispeech_7021-79740-0002.flac", "local_wav": "audio/librispeech_7021-79740-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7021-79740-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_7021", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7021-79740-0011", "transcript": "SO SAYING SHE LED THE WAY ON TIPTOE FOLLOWED BY THE CHILDREN OUT OF THE ROOM AND ROUND BY A CIRCUITOUS ROUTE TO THE PIAZZA THERE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7021/79740/7021-79740-0011.flac"}, "sample_rate": 16000, "source_samples": 165040, "source_duration_seconds": 10.315, "source_peak": 0.536102294921875, "source_rms_dbfs": -24.714389336771436, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.228922794174977, "decoded_pcm_sha256": "6cda455d70e2130c6b8f8269b688ec0381f52c4c07376a3f0f4bbbf73ca24fc7", "source_sha256": "af3d9a4d3df12f73bc4432b8be4762df32f7381a8c20009735ca8e982084c2a6", "wav_sha256": "cf60e39e5c031f055d45d2c1943b8eae811ecff60ac6593bb402412220cc0cd6", "local_source": "audio/librispeech_7021-79740-0011.flac", "local_wav": "audio/librispeech_7021-79740-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7021-85628-0025", "corpus": "librispeech_test_clean", "speaker": "librispeech_7021", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7021-85628-0025", "transcript": "BUT HIS MOTHER HUGGED HIM CLOSE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7021/85628/7021-85628-0025.flac"}, "sample_rate": 16000, "source_samples": 44400, "source_duration_seconds": 2.775, "source_peak": 0.411346435546875, "source_rms_dbfs": -25.512275213051296, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.512275213051296, "decoded_pcm_sha256": "f8114fb1a86ed60fb5b4e5ba870142361984ad8304c63676d1f9c029942602ee", "source_sha256": "477a753670424f1fd835f77fdd15e54570114ec1a4f99c206e555e94137688c6", "wav_sha256": "4acccc54bcc7e762d4e6fc8d1cbd7625bb0ec4c79de17ddb8382b1ce7fe8cffe", "local_source": "audio/librispeech_7021-85628-0025.flac", "local_wav": "audio/librispeech_7021-85628-0025.wav", "excerpt_start_sample": 0, "excerpt_samples": 44400}, {"id": "librispeech_7127-75946-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_7127", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7127-75946-0002", "transcript": "LET HIM COME IN THEN SAID THE KING AND AS IF COLBERT HAD BEEN LISTENING AT THE DOOR FOR THE PURPOSE OF KEEPING HIMSELF AU COURANT WITH THE CONVERSATION HE ENTERED AS SOON AS THE KING HAD PRONOUNCED HIS NAME TO THE TWO COURTIERS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7127/75946/7127-75946-0002.flac"}, "sample_rate": 16000, "source_samples": 211040, "source_duration_seconds": 13.19, "source_peak": 0.515167236328125, "source_rms_dbfs": -26.998476985882505, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.14563278485655, "decoded_pcm_sha256": "93c54b2cc025d80c3059dc9d273712d4259e0ddbef769ef56e3f3d16805ad734", "source_sha256": "8a4077cfaf005d616cab7fe18bc2a434b627b24de056d78cafc33f09faa1cf6c", "wav_sha256": "ccc35cdc72cfa24627ef25ab7fdda0305587a1d469a9ca5b3adf519ca7bc2926", "local_source": "audio/librispeech_7127-75946-0002.flac", "local_wav": "audio/librispeech_7127-75946-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7127-75946-0029", "corpus": "librispeech_test_clean", "speaker": "librispeech_7127", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7127-75946-0029", "transcript": "THE KING WHO HAD FROM THIS MOMENT BECOME IN REALITY THE PRINCIPAL DANCER IN THE QUADRILLE CAST A LOOK UPON HIS VANQUISHED RIVAL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7127/75946/7127-75946-0029.flac"}, "sample_rate": 16000, "source_samples": 148560, "source_duration_seconds": 9.285, "source_peak": 0.531982421875, "source_rms_dbfs": -28.20061844499801, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.58805214770615, "decoded_pcm_sha256": "92db6e9b9692ce0f9a8c0ad4504635e4a6e649b6d8760fca3400f3ea75d63e54", "source_sha256": "e26b6b1a8cceb2a05b3aae15719c785d4d24a2a09d91dd6292da8a4b4e32401a", "wav_sha256": "a0738b9c79c82ef114bc52a64dee558b82ab4c4e6d068a5d9c0429add5aa4272", "local_source": "audio/librispeech_7127-75946-0029.flac", "local_wav": "audio/librispeech_7127-75946-0029.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7127-75947-0011", "corpus": "librispeech_test_clean", "speaker": "librispeech_7127", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7127-75947-0011", "transcript": "REMAIN I IMPLORE YOU THE EVENING IS MOST LOVELY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7127/75947/7127-75947-0011.flac"}, "sample_rate": 16000, "source_samples": 57920, "source_duration_seconds": 3.62, "source_peak": 0.337554931640625, "source_rms_dbfs": -25.11875836284418, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.11875836284418, "decoded_pcm_sha256": "aebe57a65af629b724d8ca28eb599d83587cc213e6e303682e713453ab9de80a", "source_sha256": "e1bbd2d0c4d8800ab59fe4f4a1f027601dc96c787ad8bf50fd97e051270ee72b", "wav_sha256": "c31a1ac369e0120b546cca7a9eb59600c80e55118335293e80c3aafefc23467e", "local_source": "audio/librispeech_7127-75947-0011.flac", "local_wav": "audio/librispeech_7127-75947-0011.wav", "excerpt_start_sample": 0, "excerpt_samples": 57920}, {"id": "librispeech_7176-88083-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_7176", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7176-88083-0002", "transcript": "HIS FEET WERE RED HIS LONG NARROW BEAK WITH ITS SAW TOOTHED EDGES AND SHARP HOOKED TIP WAS BRIGHT RED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7176/88083/7176-88083-0002.flac"}, "sample_rate": 16000, "source_samples": 120160, "source_duration_seconds": 7.51, "source_peak": 0.627410888671875, "source_rms_dbfs": -21.729688964800097, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.54323279506653, "decoded_pcm_sha256": "22dedd61331f17693761f502975e31bf01741aced72867d6ade55e38b2e09d33", "source_sha256": "3c042266b649c153544bdb4b6c17ac6faf27e9a2b589a3d673225c7ea91180cf", "wav_sha256": "31423e751d074a942799ab45f2654e651a29295a96fe071812f6e7002ee7d77b", "local_source": "audio/librispeech_7176-88083-0002.flac", "local_wav": "audio/librispeech_7176-88083-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7176-88083-0004", "corpus": "librispeech_test_clean", "speaker": "librispeech_7176", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7176-88083-0004", "transcript": "BUT SUDDENLY STRAIGHT AND SWIFT AS A DIVING CORMORANT HE SHOT DOWN INTO THE TORRENT AND DISAPPEARED BENEATH THE SURFACE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7176/88083/7176-88083-0004.flac"}, "sample_rate": 16000, "source_samples": 120000, "source_duration_seconds": 7.5, "source_peak": 0.60992431640625, "source_rms_dbfs": -21.15699985941999, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.87864531935906, "decoded_pcm_sha256": "43bbb6d389351da5d47cefcacfede3e848cb81497c672b951f8b533d7fbedfed", "source_sha256": "a0879f1d42b65a0bc10e3b725393bac049736d153e7a1c1b5d35e48bda4e3778", "wav_sha256": "f0dcd5085b4ebd5cc0be611f94aa76f39d4732e8894a1330b1b07532c0a56fd1", "local_source": "audio/librispeech_7176-88083-0004.flac", "local_wav": "audio/librispeech_7176-88083-0004.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7176-88083-0026", "corpus": "librispeech_test_clean", "speaker": "librispeech_7176", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7176-88083-0026", "transcript": "THE DRAG UPON HIS BEAK AND THE LIGHT CHECK UPON HIS WINGS WERE INEXPLICABLE TO HIM AND APPALLING", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7176/88083/7176-88083-0026.flac"}, "sample_rate": 16000, "source_samples": 88480, "source_duration_seconds": 5.53, "source_peak": 0.62872314453125, "source_rms_dbfs": -22.34407403080596, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.34407403080596, "decoded_pcm_sha256": "bc9284b0dea002361e46bae8e07d7c2c20a9984ef0ce6f6e2c9fda86e70a579e", "source_sha256": "c637736d1dbd70e42d9ad3e319b273baf390cbc3f3ec8b31239ea7dca8e9d5d6", "wav_sha256": "50ee066cc5b8c5ac8d9942ca3a18ee3a2dfebe811603f1a35913c8be6b77fd0d", "local_source": "audio/librispeech_7176-88083-0026.flac", "local_wav": "audio/librispeech_7176-88083-0026.wav", "excerpt_start_sample": 0, "excerpt_samples": 88480}, {"id": "librispeech_7729-102255-0028", "corpus": "librispeech_test_clean", "speaker": "librispeech_7729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7729-102255-0028", "transcript": "PRIVATE PERSONS WHO HAD LEASED THE FREE STATE HOTEL VAINLY BESOUGHT THE VARIOUS AUTHORITIES TO PREVENT THE DESTRUCTION OF THEIR PROPERTY", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7729/102255/7729-102255-0028.flac"}, "sample_rate": 16000, "source_samples": 153600, "source_duration_seconds": 9.6, "source_peak": 0.71356201171875, "source_rms_dbfs": -24.18336133917225, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.02753596641884, "decoded_pcm_sha256": "7a4ab1cb22b23d73175ff25f3f2557bc2724e5cd7a3952f1cc5626d2a9bfbd3f", "source_sha256": "4654ddc035bcd2a1c3def2c04709f8f93eff6f9d9da968a1f044f65edf3a5b15", "wav_sha256": "564baedf5982a21580b89a5281b3a5dcf3522711c4a8fa991f0f1b8e60b09b1c", "local_source": "audio/librispeech_7729-102255-0028.flac", "local_wav": "audio/librispeech_7729-102255-0028.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7729-102255-0038", "corpus": "librispeech_test_clean", "speaker": "librispeech_7729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7729-102255-0038", "transcript": "ATCHISON WHO HAD BEEN HARANGUING THE MOB PLANTED HIS TWO GUNS BEFORE THE BUILDING AND TRAINED THEM UPON IT", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7729/102255/7729-102255-0038.flac"}, "sample_rate": 16000, "source_samples": 126720, "source_duration_seconds": 7.92, "source_peak": 0.547515869140625, "source_rms_dbfs": -23.868262038672686, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.65352971837625, "decoded_pcm_sha256": "4488be03955742e6e44fcdb33f0ae71643965b6765824a166c40ca31fbd9f66a", "source_sha256": "7bad471a79798adfc54e63acefeaf2a9579bc1f06779fd3aa49cc060df7c47f7", "wav_sha256": "b74c24750ddac32d162fd267caa8077dd1b1ce41e69840db61fc22a5aa5a4c04", "local_source": "audio/librispeech_7729-102255-0038.flac", "local_wav": "audio/librispeech_7729-102255-0038.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_7729-102255-0042", "corpus": "librispeech_test_clean", "speaker": "librispeech_7729", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "7729-102255-0042", "transcript": "RELOCATED FOOTNOTE GOVERNOR ROBINSON BEING ON HIS WAY EAST THE STEAMBOAT ON WHICH HE WAS TRAVELING STOPPED AT LEXINGTON MISSOURI", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/7729/102255/7729-102255-0042.flac"}, "sample_rate": 16000, "source_samples": 147600, "source_duration_seconds": 9.225, "source_peak": 0.69677734375, "source_rms_dbfs": -23.419483735767322, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.03270368618639, "decoded_pcm_sha256": "3e552debe1c0d524847e6550ec6f2d9550fc271ad1a9ac47e3543e154db1f552", "source_sha256": "258a5f83b294976c312b375ef48ed79028fb41459458f6ece40b90fab7bce317", "wav_sha256": "9b9665bfe15edaa592968b7b1014db2734b12dc98752aa4ad4d62e5499eac370", "local_source": "audio/librispeech_7729-102255-0042.flac", "local_wav": "audio/librispeech_7729-102255-0042.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8224-274381-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_8224", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8224-274381-0010", "transcript": "BY A QUICK AND UNEXPECTED MARCH MONTROSE HASTENED TO INNERLOCHY AND PRESENTED HIMSELF IN ORDER OF BATTLE BEFORE THE SURPRISED BUT NOT AFFRIGHTENED COVENANTERS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8224/274381/8224-274381-0010.flac"}, "sample_rate": 16000, "source_samples": 217920, "source_duration_seconds": 13.62, "source_peak": 0.44830322265625, "source_rms_dbfs": -23.72504999405364, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.951331441759642, "decoded_pcm_sha256": "34d9175c2a6517466fd56565f6c4abc5a8a75b16d60e1e96d0615d4383e19b19", "source_sha256": "e9f711f02a8f7cfe4527d6d16f4ad04dc95369de860a6c4f56a645dc7da5a02b", "wav_sha256": "18c326a6f8dc3490bccdd4d2cb16af9847410c33ce1a4fcf91c3002db72c8aae", "local_source": "audio/librispeech_8224-274381-0010.flac", "local_wav": "audio/librispeech_8224-274381-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8224-274381-0013", "corpus": "librispeech_test_clean", "speaker": "librispeech_8224", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8224-274381-0013", "transcript": "FROM THE SAME MEN NEW REGIMENTS AND NEW COMPANIES WERE FORMED DIFFERENT OFFICERS APPOINTED AND THE WHOLE MILITARY FORCE PUT INTO SUCH HANDS AS THE INDEPENDENTS COULD RELY ON", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8224/274381/8224-274381-0013.flac"}, "sample_rate": 16000, "source_samples": 217760, "source_duration_seconds": 13.61, "source_peak": 0.536865234375, "source_rms_dbfs": -21.803141664293197, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.810391566928214, "decoded_pcm_sha256": "31d48338892252220a9e3042986f534b91479b75a07438f2a0515b00bc8fe055", "source_sha256": "cd0ba8a197cc480b016a6386531b3e1ee14c9ff6ddc85f19abde2abc4f00070a", "wav_sha256": "a6a7de02099102496e51606fd6032ef6e7d8800ebd1ab894b29e2fe6066d3816", "local_source": "audio/librispeech_8224-274381-0013.flac", "local_wav": "audio/librispeech_8224-274381-0013.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8224-274384-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_8224", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8224-274384-0010", "transcript": "BEFORE THE SETTLEMENT OF TERMS THE ADMINISTRATION MUST BE POSSESSED ENTIRELY BY THE PARLIAMENTS OF BOTH KINGDOMS AND HOW INCOMPATIBLE THAT SCHEME WITH THE LIBERTY OF THE KING IS EASILY IMAGINED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8224/274384/8224-274384-0010.flac"}, "sample_rate": 16000, "source_samples": 233600, "source_duration_seconds": 14.6, "source_peak": 0.393951416015625, "source_rms_dbfs": -25.11362426001846, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.152240114273017, "decoded_pcm_sha256": "bfff7d3cf9d00e04b5e5f471ec3e7e23254f82646da0069c2f8d619c084a00f4", "source_sha256": "677a2c48f5ddd5929e5c68e682b66998c48322833dac3baa41a086f87ef15077", "wav_sha256": "5d22068dd2ce8f40fce798da990cdd48504bfbf705b82b996eeab016ef4c4c97", "local_source": "audio/librispeech_8224-274384-0010.flac", "local_wav": "audio/librispeech_8224-274384-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8230-279154-0018", "corpus": "librispeech_test_clean", "speaker": "librispeech_8230", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8230-279154-0018", "transcript": "THERE IS OF COURSE A DIFFERENCE BETWEEN KNOWING THE TEMPORAL RELATION OF A REMEMBERED EVENT TO THE PRESENT AND KNOWING THE TIME ORDER OF TWO REMEMBERED EVENTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8230/279154/8230-279154-0018.flac"}, "sample_rate": 16000, "source_samples": 189520, "source_duration_seconds": 11.845, "source_peak": 0.658111572265625, "source_rms_dbfs": -23.673484842402758, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.85204146467471, "decoded_pcm_sha256": "2becab040601a06405a880e563913b87f3eb167e942e97f7225c55643949f142", "source_sha256": "6fb6d05dd68dfe641a0f58b800e9d2e86662dab952355bc7573505bc35efb2fc", "wav_sha256": "7683ffe730b9a0eb95d05f98aba9d11f6ad6fa68a80ab310dc223d779cd0921a", "local_source": "audio/librispeech_8230-279154-0018.flac", "local_wav": "audio/librispeech_8230-279154-0018.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8230-279154-0032", "corpus": "librispeech_test_clean", "speaker": "librispeech_8230", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8230-279154-0032", "transcript": "IT IS THIS THAT IS OF INTEREST TO THEORY OF KNOWLEDGE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8230/279154/8230-279154-0032.flac"}, "sample_rate": 16000, "source_samples": 62080, "source_duration_seconds": 3.88, "source_peak": 0.763580322265625, "source_rms_dbfs": -23.179606504911124, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.179606504911124, "decoded_pcm_sha256": "d59c4c75b338357fc4c79c9980c56076c2ae54441e0d223325fd8514435d4576", "source_sha256": "d60da130333b0e1adba450d171b7f753f423dce0f66a00448dfb036d43a7231d", "wav_sha256": "46983a79c683e36b2b6ffd2611bd41f526680b8d83fa351d5b1ff14283d23e2e", "local_source": "audio/librispeech_8230-279154-0032.flac", "local_wav": "audio/librispeech_8230-279154-0032.wav", "excerpt_start_sample": 0, "excerpt_samples": 62080}, {"id": "librispeech_8230-279154-0033", "corpus": "librispeech_test_clean", "speaker": "librispeech_8230", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8230-279154-0033", "transcript": "IT IS BY NO MEANS ALWAYS RELIABLE ALMOST EVERYBODY HAS AT SOME TIME EXPERIENCED THE WELL KNOWN ILLUSION THAT ALL THAT IS HAPPENING NOW HAPPENED BEFORE AT SOME TIME", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8230/279154/8230-279154-0033.flac"}, "sample_rate": 16000, "source_samples": 187040, "source_duration_seconds": 11.69, "source_peak": 0.857177734375, "source_rms_dbfs": -21.919399299890486, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -20.434707149884012, "decoded_pcm_sha256": "bd287f891d96a4cd6b7a04337bf6be93b04db8cbec30940ae4945f8b4efde0f4", "source_sha256": "1bcf539698173ec1996fa39e5b4c5faf7a8ddf5b5f9636d269ca583befc17eb6", "wav_sha256": "4d90673ae367514c97b64216ae1e3fedf844226a86b71986ec389ed219babf2f", "local_source": "audio/librispeech_8230-279154-0033.flac", "local_wav": "audio/librispeech_8230-279154-0033.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8455-210777-0010", "corpus": "librispeech_test_clean", "speaker": "librispeech_8455", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8455-210777-0010", "transcript": "THEIR MASTERS SAID MISSUS NEVERBEND", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8455/210777/8455-210777-0010.flac"}, "sample_rate": 16000, "source_samples": 49600, "source_duration_seconds": 3.1, "source_peak": 0.97686767578125, "source_rms_dbfs": -24.180707212674086, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.180707212674086, "decoded_pcm_sha256": "10ec70ae1221ff8684ca0b1b6aa22fd7ae9343c0b701f90c315735ff27f3d5ad", "source_sha256": "28029242eb37055c0eb70761eebf5b7be98a5da6c30127b3b282bf7e11663a68", "wav_sha256": "c7e846edf284a6ab491b525b5058af16b5a3f2b4858d3189ffdb3c297723e488", "local_source": "audio/librispeech_8455-210777-0010.flac", "local_wav": "audio/librispeech_8455-210777-0010.wav", "excerpt_start_sample": 0, "excerpt_samples": 49600}, {"id": "librispeech_8455-210777-0049", "corpus": "librispeech_test_clean", "speaker": "librispeech_8455", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8455-210777-0049", "transcript": "LIEUTENANT CROSSTREES IS A VERY GALLANT OFFICER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8455/210777/8455-210777-0049.flac"}, "sample_rate": 16000, "source_samples": 65760, "source_duration_seconds": 4.11, "source_peak": 0.46630859375, "source_rms_dbfs": -24.60742082061149, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.60742082061149, "decoded_pcm_sha256": "6bab2226481307e41ffb714c528db253ee0755e0775a2890f32c07e3f5832d55", "source_sha256": "2c4c3f4fa85e1e234b108aeada818530820e7a515dc7d91f2a2c075af7105fed", "wav_sha256": "f86b45c1d7a339929b918658a770ad680ab2167df6561f4cc3f2581743927ba9", "local_source": "audio/librispeech_8455-210777-0049.flac", "local_wav": "audio/librispeech_8455-210777-0049.wav", "excerpt_start_sample": 0, "excerpt_samples": 65760}, {"id": "librispeech_8455-210777-0060", "corpus": "librispeech_test_clean", "speaker": "librispeech_8455", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "8455-210777-0060", "transcript": "THE JOHN BRIGHT IS ARMED WITH A WEAPON OF GREAT POWER AGAINST WHICH IT IS IMPOSSIBLE THAT THE PEOPLE OF BRITANNULA SHOULD PREVAIL", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8455/210777/8455-210777-0060.flac"}, "sample_rate": 16000, "source_samples": 113200, "source_duration_seconds": 7.075, "source_peak": 0.728973388671875, "source_rms_dbfs": -21.874934111772994, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -21.328040515972575, "decoded_pcm_sha256": "71c64b069cc8f43d09cfadd1defcda3beccfdf73919c8518a13182fa2c801f93", "source_sha256": "ef57db07976e02027d19da092bac4747c7928a93199441010b967c12ce13e724", "wav_sha256": "879b2509fba33c8e73d65f576d33bc56edca7a2ee0441e2e0485f96aabf7a688", "local_source": "audio/librispeech_8455-210777-0060.flac", "local_wav": "audio/librispeech_8455-210777-0060.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8463-294828-0003", "corpus": "librispeech_test_clean", "speaker": "librispeech_8463", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8463-294828-0003", "transcript": "I WANTED NOTHING MORE THAN TO SEE MY COUNTRY AGAIN MY FRIENDS MY MODEST QUARTERS BY THE BOTANICAL GARDENS MY DEARLY BELOVED COLLECTIONS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8463/294828/8463-294828-0003.flac"}, "sample_rate": 16000, "source_samples": 149440, "source_duration_seconds": 9.34, "source_peak": 0.445587158203125, "source_rms_dbfs": -24.97531570503334, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.82912392962218, "decoded_pcm_sha256": "cd223f40e942ab9ac205813cbe938225779742060f5fe247eb3e2fd12df52eba", "source_sha256": "3790b43d50403978fd6ba4d90f598e8e6afddfb630a3bf5b4110bd432517f1e9", "wav_sha256": "03177b8bf19007f5f7477015e45cc382bce9b710fdea1134a0efc765c379e895", "local_source": "audio/librispeech_8463-294828-0003.flac", "local_wav": "audio/librispeech_8463-294828-0003.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8463-294828-0014", "corpus": "librispeech_test_clean", "speaker": "librispeech_8463", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8463-294828-0014", "transcript": "THERE WAS GOOD REASON TO STOP AND THINK EVEN FOR THE WORLD'S MOST EMOTIONLESS MAN", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8463/294828/8463-294828-0014.flac"}, "sample_rate": 16000, "source_samples": 91600, "source_duration_seconds": 5.725, "source_peak": 0.51129150390625, "source_rms_dbfs": -23.96932327924813, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.96932327924813, "decoded_pcm_sha256": "c608aedb889e9a3c3821520a449396c5b4cd1264e1d081a38be40b0b66f8b285", "source_sha256": "b61f11aa0c2b3d1adca49e6199328637c6ecc84c003dab03fc8c97b95cab4c63", "wav_sha256": "87e86c5001f0bc4f6cc029d8e282bfd78ad4fef425a55309bbfc646a7744e707", "local_source": "audio/librispeech_8463-294828-0014.flac", "local_wav": "audio/librispeech_8463-294828-0014.wav", "excerpt_start_sample": 0, "excerpt_samples": 91600}, {"id": "librispeech_8463-294828-0022", "corpus": "librispeech_test_clean", "speaker": "librispeech_8463", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8463-294828-0022", "transcript": "WE'RE LEAVING ON THE ABRAHAM LINCOLN", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8463/294828/8463-294828-0022.flac"}, "sample_rate": 16000, "source_samples": 37680, "source_duration_seconds": 2.355, "source_peak": 0.42718505859375, "source_rms_dbfs": -24.863923522138165, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -24.863923522138165, "decoded_pcm_sha256": "276e5d5f0e0b54af2b651a409e0f7fbcd7b42508e69037a0becabc956b358b07", "source_sha256": "4b0b7f62860ac9f04945b08f79183b52c26f80eedd14ff829e2ff2aaa5143e3f", "wav_sha256": "76e6aeb80f67faf23c266b54112340687389f0475ffad207ed027383328b49b4", "local_source": "audio/librispeech_8463-294828-0022.flac", "local_wav": "audio/librispeech_8463-294828-0022.wav", "excerpt_start_sample": 0, "excerpt_samples": 37680}, {"id": "librispeech_8555-284447-0007", "corpus": "librispeech_test_clean", "speaker": "librispeech_8555", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8555-284447-0007", "transcript": "THEREFORE HER MAJESTY PAID NO ATTENTION TO ANYONE AND NO ONE PAID ANY ATTENTION TO HER", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8555/284447/8555-284447-0007.flac"}, "sample_rate": 16000, "source_samples": 101840, "source_duration_seconds": 6.365, "source_peak": 0.24127197265625, "source_rms_dbfs": -26.869887535083663, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -26.613422172170882, "decoded_pcm_sha256": "b2e9b095c084b8ff6903b45eed724068bad7267cf8c1f692479d68f2b6adbaed", "source_sha256": "73d10187e60efe5f030a76fa33d1eab2abae7e38817b356dd61e30b743edf22f", "wav_sha256": "cc3409e20c1284fb07dc358b49b0faf9bb0cc3d7d6726d4073f8961fc15b3155", "local_source": "audio/librispeech_8555-284447-0007.flac", "local_wav": "audio/librispeech_8555-284447-0007.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_8555-284449-0020", "corpus": "librispeech_test_clean", "speaker": "librispeech_8555", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8555-284449-0020", "transcript": "THE COMBINED BANDS OF BOTH THE COUNTRIES PLAYED THE MUSIC AND A FINE SUPPER WAS SERVED", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8555/284449/8555-284449-0020.flac"}, "sample_rate": 16000, "source_samples": 81520, "source_duration_seconds": 5.095, "source_peak": 0.29266357421875, "source_rms_dbfs": -27.14678667416795, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -27.14678667416795, "decoded_pcm_sha256": "fea5f67e0d192edbb42257fa9560e233f643cc196059371f1de72f17987367e7", "source_sha256": "5703c01afc67c6eb161bd9fe7fe92f5b373e9331629f60d1c163e5d893c97d7a", "wav_sha256": "a033b75742b985d12a5c13f94f380d6d7440bd6723b0faabac5379e81d1fec62", "local_source": "audio/librispeech_8555-284449-0020.flac", "local_wav": "audio/librispeech_8555-284449-0020.wav", "excerpt_start_sample": 0, "excerpt_samples": 81520}, {"id": "librispeech_8555-292519-0001", "corpus": "librispeech_test_clean", "speaker": "librispeech_8555", "language": "en", "style": "normal", "speaker_gender_as_distributed": "F", "source_utterance_id": "8555-292519-0001", "transcript": "GUIDED BY YOU HOW WE MIGHT STROLL TOWARDS DEATH OUR ONLY MUSIC ONE ANOTHER'S BREATH THROUGH GARDENS INTIMATE WITH HOLLYHOCKS WHERE SILENT POPPIES BURN BETWEEN THE ROCKS BY POOLS WHERE BIRCHES BEND TO CONFIDANTS ABOVE GREEN WATERS SCUMMED WITH LILY PLANTS", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/8555/292519/8555-292519-0001.flac"}, "sample_rate": 16000, "source_samples": 333120, "source_duration_seconds": 20.82, "source_peak": 0.60888671875, "source_rms_dbfs": -23.415033537556127, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -22.624461889815606, "decoded_pcm_sha256": "c2e62e46b5a18b9f8f30423189b31cc00703ba64084bb8bca7ffe75ae8d6977a", "source_sha256": "b082de228dbad73be95d916e5149c02035758a390482845def17215c0eed0fa9", "wav_sha256": "9f14d64bcc89c118ff39930d39a3240d88be1fe829cbbb61a05786867b5b1be1", "local_source": "audio/librispeech_8555-292519-0001.flac", "local_wav": "audio/librispeech_8555-292519-0001.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}, {"id": "librispeech_908-157963-0018", "corpus": "librispeech_test_clean", "speaker": "librispeech_908", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "908-157963-0018", "transcript": "AND FEAREST THOU BECAUSE I VANISH AND AM SEEN NO MORE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/908/157963/908-157963-0018.flac"}, "sample_rate": 16000, "source_samples": 68080, "source_duration_seconds": 4.255, "source_peak": 0.513336181640625, "source_rms_dbfs": -23.538819799676702, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.538819799676702, "decoded_pcm_sha256": "8ae8f9234bac8c63b967d3c7b5075657f9771ac253a59a383e523ecff9c3b006", "source_sha256": "53efff3282f4691482a4e735887fab9130c78ec10f3bdb79bf68f8c212898134", "wav_sha256": "aeae977ce030a5c4dc6958737ff58b35c245d4fdadf9685b6e94dacfdcce7da4", "local_source": "audio/librispeech_908-157963-0018.flac", "local_wav": "audio/librispeech_908-157963-0018.wav", "excerpt_start_sample": 0, "excerpt_samples": 68080}, {"id": "librispeech_908-31957-0002", "corpus": "librispeech_test_clean", "speaker": "librispeech_908", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "908-31957-0002", "transcript": "I DID NOT WRONG MYSELF SO BUT I PLACED A WRONG ON THEE", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/908/31957/908-31957-0002.flac"}, "sample_rate": 16000, "source_samples": 76640, "source_duration_seconds": 4.79, "source_peak": 0.42138671875, "source_rms_dbfs": -25.537578974520407, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -25.537578974520407, "decoded_pcm_sha256": "5e0a5e522c82eea67f456aec58b62300d4c743b9610bb9f4f869f3b8bd8e7f17", "source_sha256": "77ad707e976d262395c265cdfda5b9a3da7439f329aac0f5216a404acefe3fe7", "wav_sha256": "88f78a25a0c7082384d7a9653ba5e656504ddc299b790c6bd9d4a11b5119d742", "local_source": "audio/librispeech_908-31957-0002.flac", "local_wav": "audio/librispeech_908-31957-0002.wav", "excerpt_start_sample": 0, "excerpt_samples": 76640}, {"id": "librispeech_908-31957-0016", "corpus": "librispeech_test_clean", "speaker": "librispeech_908", "language": "en", "style": "normal", "speaker_gender_as_distributed": "M", "source_utterance_id": "908-31957-0016", "transcript": "DEAREST TEACH ME SO TO POUR OUT GRATITUDE AS THOU DOST GOOD", "source": {"archive_url": "https://www.openslr.org/resources/12/test-clean.tar.gz", "member": "LibriSpeech/test-clean/908/31957/908-31957-0016.flac"}, "sample_rate": 16000, "source_samples": 103680, "source_duration_seconds": 6.48, "source_peak": 0.563873291015625, "source_rms_dbfs": -24.06459814668036, "full_scale_fraction": 0.0, "reasons": [], "excerpt_full_scale_fraction": 0.0, "excerpt_rms_dbfs": -23.73060181736992, "decoded_pcm_sha256": "a1a0fc0c8f705774ce9810918acda74ca07e00f2107a855ab611aa7ee1735b94", "source_sha256": "2a57896e33e6a87debcb741f00196495aa665bb2f0d7c205ac980a8417976d12", "wav_sha256": "fccef19835dc4eb3cde5b4e85b177f7b4375f6ef7db090434ea7ab00d5b00ca9", "local_source": "audio/librispeech_908-31957-0016.flac", "local_wav": "audio/librispeech_908-31957-0016.wav", "excerpt_start_sample": 0, "excerpt_samples": 96000}], "criteria_sha256": "19c16dbc3ed6478cd79cbc8b04e99e6fee6a693f0ba9b36e95d944ec710b1a9d", "lock_sha256": "c8b2bc28ebb1523bf8cfe0f4627d06bd5d03e0ba6a2bb5dc0905be77a3dc8823", "python": "3.12.14", "platform": "Windows-11-10.0.26200-SP0", "dependencies": {"numpy": "2.3.5", "scipy": "1.17.0", "pystoi": "0.4.1", "torch": "2.8.0+cpu", "torchaudio": "2.8.0+cpu", "torchvision": "0.23.0+cpu", "transformers": "4.51.3", "timm": "1.0.15", "librosa": "0.11.0", "utmosv2": "1.3.1.dev0"}, "methods": ["hnm", "A", "B", "C", "D", "E"], "noise_seeds": [0, 1, 2, 3, 4, 5, 6, 7], "mos_noise_seed": 0, "mos_extraction_seed": 20261012, "mos_repetitions": 1, "mos_threads": 1, "workers": 2, "threads_per_worker": 1, "alignment": "none", "normalization": "none", "analysis": "single_current_base_encode_per_recording", "world": "not_run; historical_reference_only", "role": "candidate_screening_not_automatic_adoption"}, "effective_mos_contract": {"original_contract_sha256": "7919752c875190bf5db026ad8e6a081abbfc4b1d0bcc4c15974480a240061a18", "planned": {"mos_noise_seed": 0, "mos_extraction_seed": 20261012, "mos_repetitions": 1, "mos_threads": 1}, "effective": {"mos_noise_seed": 0, "mos_extraction_seed": 20261012, "mos_repetitions": 1, "mos_threads": 4}, "amendment": "英語MOS開始前にCPUスレッド数のみ変更。音響契約原本は維持。", "cache_reuse_policy": {"performed_at": "2026-10-12T06:58:46.617698+00:00", "script_sha256": "043cd433757d86bc1620397487fe2db290df5ea68c435b0e21620eec3fa90df7", "skip_first_sorted_recordings": 12, "matching_required": ["input_float32_sha256", "full_mos_provenance", "mos.py_source_sha256"], "reuse_count": 214, "limitation": "完全一致入力の旧Linux測定を引用。新Windows再測定とは数えない。"}}, "mos_provenance": {"name": "UTMOSv2", "quantity": "predicted_mos_not_human_rating", "source_repository": "https://github.com/sarulab-speech/UTMOSv2", "source_commit": "cc2700db57bb83ee13dc31ebe1b868c254e15d09", "checkpoint_revision": "506474f2b33dc77c234d668cc419be1861899cad", "checkpoint_sha256": "c8149d988e4bbf3f347e6966b5d769de347a5f8c59ffca1dc4bd4bf5b8585e57", "fold": 0, "device": "cpu", "configuration": "fusion_stage3", "predict_dataset": "bvcc", "repetitions": 1, "extraction_seed": 20261012, "cpu_threads": 4, "remove_silent_section": false, "nnpack_enabled": false, "internal_sample_rate": 16000, "ssl_crop_seconds": 3.0, "spectrogram_crop_seconds": 1.4, "spectrogram_frames_per_repetition": 2, "spectrogram_inner_mixup": true, "short_input_extension": "tile", "code_license": "MIT", "weights_license": "MIT", "independent_human_rating_calibration": "not_performed_for_this_project", "scope": "JVS/LibriSpeechの再合成への適用は未校正。言語・発声別に集計する。", "paper_protocol_difference": "論文の5fold×5抽出平均ではなく、fold0と上記反復数。", "initialization": "official_architecture_then_complete_strict_state_dict", "packages": {"torch": "2.8.0+cpu", "torchaudio": "2.8.0+cpu", "torchvision": "0.23.0+cpu", "transformers": "4.51.3", "timm": "1.0.15", "librosa": "0.11.0", "numpy": "2.3.5", "utmosv2": "1.3.1.dev0"}, "checkpoint": {"url": "https://huggingface.co/sarulab-speech/UTMOSv2/resolve/506474f2b33dc77c234d668cc419be1861899cad/fold0_s42_best_model.pth", "sha256": "c8149d988e4bbf3f347e6966b5d769de347a5f8c59ffca1dc4bd4bf5b8585e57", "bytes": 818531314}, "feature_configuration": {"config.json": {"url": "https://huggingface.co/facebook/wav2vec2-base/resolve/0b5b8e868dd84f03fd87d01f9c4ff0f080fecfe8/config.json", "sha256": "4937977e24d12d1bba70cdce8709c3c04807a8e4ae8ddac4229c48c436ae99ae", "bytes": 1842}, "preprocessor_config.json": {"url": "https://huggingface.co/facebook/wav2vec2-base/resolve/0b5b8e868dd84f03fd87d01f9c4ff0f080fecfe8/preprocessor_config.json", "sha256": "b225d617c025463b9e157e06afea8b90dc7078fc70b013c533328423e0486b4a", "bytes": 159}}, "initialization_status": "ready"}, "stage": "all", "finished_at": "2026-10-12T07:22:18.055941+00:00", "parts": [{"case": {"id": "librispeech_1089-134686-0008", "language": "en", "style": "normal", "speaker": "librispeech_1089", "source_utterance_id": "1089-134686-0008", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1089-134686-0008.wav", "source_file_sha256": "a1f82cc76d47fc74b1f6fd2d0b2668731331625a2f9a98ffa7ce164eb9f29b1f", "decoded_pcm_sha256": "40f044b90ed06d7cffc4145524ea3e1b8bd36b6599e24e1e905262fbb3e50d8d", "original_source_sha256": "1bdf7eede41ca2cc8883defb66bbd2e59ec93bf2d587327cb051d0ae427495ae", "corpus_entry_id": "librispeech_1089-134686-0008"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.073606014251709, "predicted_mos": 4.073606014251709, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "10cabd75d8a85c14e77bf1489974a434f5ed836d2c398b22362a8eaa3c9d988f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.012534500099719, "cpu_threads": 4}, "waveform_sha256": "2906c24c57cf16b7a1a29dac15e0d727cb8b4679724efec0f565946723cc9fd8"}, "rows": [{"case": "librispeech_1089-134686-0008", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.417376756668091, "predicted_mos": 2.417376756668091, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6cfab5b450a5dd209725fa1eba0bd1bd541039668d0910af1bc05194d6835f40", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.619196000043303, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.143506609631696, "全帯域_10ms_rms_envelope_nrmse": 0.10984710477420097, "全帯域_10ms_envelope_correlation": 0.9888262213151996, "全帯域_10ms_modulation_mae_db": 0.2733341449979953, "全帯域_40ms_log_rms_mae_db": 0.6041422432736383, "全帯域_40ms_rms_envelope_nrmse": 0.07484982339203639, "全帯域_40ms_envelope_correlation": 0.9947860045112915, "全帯域_40ms_modulation_mae_db": 0.10866976755601299, "300–1000 Hz_10ms_log_rms_mae_db": 1.8192610854240836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15350823786947534, "300–1000 Hz_10ms_envelope_correlation": 0.9835770960803077, "300–1000 Hz_10ms_modulation_mae_db": 0.757595572789404, "300–1000 Hz_40ms_log_rms_mae_db": 1.1962098102858976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11722505792572667, "300–1000 Hz_40ms_envelope_correlation": 0.9909703798109416, "300–1000 Hz_40ms_modulation_mae_db": 0.37369407833487545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8823493872571633, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19465173412647452, "1000–3000 Hz_10ms_envelope_correlation": 0.9709207373047798, "1000–3000 Hz_10ms_modulation_mae_db": 1.280695039240025, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0906793764080704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11266613026813384, "1000–3000 Hz_40ms_envelope_correlation": 0.989747259996758, "1000–3000 Hz_40ms_modulation_mae_db": 1.2436626786338445, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3084330112418825, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3504519756203869, "3000–6000 Hz_10ms_envelope_correlation": 0.9250056516353918, "3000–6000 Hz_10ms_modulation_mae_db": 0.34216076734215356, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7936023782174138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29915946211041133, "3000–6000 Hz_40ms_envelope_correlation": 0.9419540963314044, "3000–6000 Hz_40ms_modulation_mae_db": 0.6919316739762346, "mr_stft_db": 6.038273884152922, "band_mae_db": 0.41358601687889057, "rms_gain_error_db": -0.20794070153926258, "output_peak": 0.3895092075985285}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1610986093065958, "全帯域_10ms_rms_envelope_nrmse": 0.10196888162356714, "全帯域_10ms_envelope_correlation": 0.9900903207160376, "全帯域_10ms_modulation_mae_db": 0.5611563820205826, "全帯域_40ms_log_rms_mae_db": 0.5921720370988007, "全帯域_40ms_rms_envelope_nrmse": 0.04899553882711189, "全帯域_40ms_envelope_correlation": 0.9975487394726873, "全帯域_40ms_modulation_mae_db": 0.3080427538936715, "300–1000 Hz_10ms_log_rms_mae_db": 1.7444918749478244, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12685023202638288, "300–1000 Hz_10ms_envelope_correlation": 0.9882258750297209, "300–1000 Hz_10ms_modulation_mae_db": 1.3091373454630308, "300–1000 Hz_40ms_log_rms_mae_db": 1.1554902991038611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06499291415965434, "300–1000 Hz_40ms_envelope_correlation": 0.9973395692987312, "300–1000 Hz_40ms_modulation_mae_db": 0.6203527588048153, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8293068622593638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21404648225184358, "1000–3000 Hz_10ms_envelope_correlation": 0.9680857365335048, "1000–3000 Hz_10ms_modulation_mae_db": 0.5117579285326036, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0863676636245625, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12852413920801492, "1000–3000 Hz_40ms_envelope_correlation": 0.988613496144053, "1000–3000 Hz_40ms_modulation_mae_db": 0.29334686915458064, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3215547501763503, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.38791108161030846, "3000–6000 Hz_10ms_envelope_correlation": 0.9199610757873606, "3000–6000 Hz_10ms_modulation_mae_db": 0.6252099580806157, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8146781103937901, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.328138340379438, "3000–6000 Hz_40ms_envelope_correlation": 0.9408172550602663, "3000–6000 Hz_40ms_modulation_mae_db": 1.0429580306350077, "mr_stft_db": 6.024094897938498, "band_mae_db": 0.43599864302772967, "rms_gain_error_db": -0.04347021148137834, "output_peak": 0.4926831191228564}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1808896540906304, "全帯域_10ms_rms_envelope_nrmse": 0.12151095245023583, "全帯域_10ms_envelope_correlation": 0.9859646087870957, "全帯域_10ms_modulation_mae_db": 0.42439378768984953, "全帯域_40ms_log_rms_mae_db": 0.6310875281901189, "全帯域_40ms_rms_envelope_nrmse": 0.0545049895947333, "全帯域_40ms_envelope_correlation": 0.9969382784833999, "全帯域_40ms_modulation_mae_db": 0.14025377898029942, "300–1000 Hz_10ms_log_rms_mae_db": 1.8444199387501132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15182427633313328, "300–1000 Hz_10ms_envelope_correlation": 0.9825522318176507, "300–1000 Hz_10ms_modulation_mae_db": 1.112502258989245, "300–1000 Hz_40ms_log_rms_mae_db": 1.1867851809975796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07129390982627991, "300–1000 Hz_40ms_envelope_correlation": 0.995989698843579, "300–1000 Hz_40ms_modulation_mae_db": 0.2229857369442184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8770904985866885, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2071003504551098, "1000–3000 Hz_10ms_envelope_correlation": 0.9714519603492746, "1000–3000 Hz_10ms_modulation_mae_db": 0.9255314773524557, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1143934412755947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13783448472953824, "1000–3000 Hz_40ms_envelope_correlation": 0.9878511476383728, "1000–3000 Hz_40ms_modulation_mae_db": 1.2335101390441068, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3398705926957057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3618591069661715, "3000–6000 Hz_10ms_envelope_correlation": 0.9181229957462506, "3000–6000 Hz_10ms_modulation_mae_db": 0.5232393518009053, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7801690926940172, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30489062658686233, "3000–6000 Hz_40ms_envelope_correlation": 0.9381055650561322, "3000–6000 Hz_40ms_modulation_mae_db": 0.35567806193390683, "mr_stft_db": 6.038274636631589, "band_mae_db": 0.4852470451612893, "rms_gain_error_db": -0.018294326852565754, "output_peak": 0.5025552434410288}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1809718302843797, "全帯域_10ms_rms_envelope_nrmse": 0.10779181631628709, "全帯域_10ms_envelope_correlation": 0.9889854243247898, "全帯域_10ms_modulation_mae_db": 0.13251930505959417, "全帯域_40ms_log_rms_mae_db": 0.6119008267660536, "全帯域_40ms_rms_envelope_nrmse": 0.04332754617260792, "全帯域_40ms_envelope_correlation": 0.9980995317140544, "全帯域_40ms_modulation_mae_db": 0.36310924589013077, "300–1000 Hz_10ms_log_rms_mae_db": 1.7898676507010467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15111839047454934, "300–1000 Hz_10ms_envelope_correlation": 0.9826981236037927, "300–1000 Hz_10ms_modulation_mae_db": 0.6232045397498648, "300–1000 Hz_40ms_log_rms_mae_db": 1.1608695837706586, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0647022833709667, "300–1000 Hz_40ms_envelope_correlation": 0.9967620996733527, "300–1000 Hz_40ms_modulation_mae_db": 0.2080618836881217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8731562684757586, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20197336064770588, "1000–3000 Hz_10ms_envelope_correlation": 0.9686528173887679, "1000–3000 Hz_10ms_modulation_mae_db": 0.49477296812221166, "1000–3000 Hz_40ms_log_rms_mae_db": 1.118334235738605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12987047378335115, "1000–3000 Hz_40ms_envelope_correlation": 0.9865512467577503, "1000–3000 Hz_40ms_modulation_mae_db": 0.4971661674571825, "3000–6000 Hz_10ms_log_rms_mae_db": 2.373388839475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3605275119264488, "3000–6000 Hz_10ms_envelope_correlation": 0.9180285923008915, "3000–6000 Hz_10ms_modulation_mae_db": 0.2959888802692307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8149876460192744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30627022816448185, "3000–6000 Hz_40ms_envelope_correlation": 0.9369116430016664, "3000–6000 Hz_40ms_modulation_mae_db": 0.20455072039373048, "mr_stft_db": 6.034760482011673, "band_mae_db": 0.37151475806160555, "rms_gain_error_db": -0.03566259599858102, "output_peak": 0.4368925016895423}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.174202076543238, "全帯域_10ms_rms_envelope_nrmse": 0.12175414592320433, "全帯域_10ms_envelope_correlation": 0.9858785273205389, "全帯域_10ms_modulation_mae_db": 0.4352443187058547, "全帯域_40ms_log_rms_mae_db": 0.580226912105949, "全帯域_40ms_rms_envelope_nrmse": 0.0678376051009288, "全帯域_40ms_envelope_correlation": 0.9953677940384004, "全帯域_40ms_modulation_mae_db": 0.3360779090119628, "300–1000 Hz_10ms_log_rms_mae_db": 1.8202866336777894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1755003885506249, "300–1000 Hz_10ms_envelope_correlation": 0.976675023966562, "300–1000 Hz_10ms_modulation_mae_db": 1.0837673020698395, "300–1000 Hz_40ms_log_rms_mae_db": 1.2550205150636249, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11736347866123575, "300–1000 Hz_40ms_envelope_correlation": 0.9892573967777633, "300–1000 Hz_40ms_modulation_mae_db": 0.6647488748171371, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8374747358012435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1886615560623643, "1000–3000 Hz_10ms_envelope_correlation": 0.9738204897987419, "1000–3000 Hz_10ms_modulation_mae_db": 0.34792794833213975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0502066668227767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1263625267343196, "1000–3000 Hz_40ms_envelope_correlation": 0.9890365123486442, "1000–3000 Hz_40ms_modulation_mae_db": 0.1940610789645961, "3000–6000 Hz_10ms_log_rms_mae_db": 2.34390092319463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36072076399203834, "3000–6000 Hz_10ms_envelope_correlation": 0.9148610197481968, "3000–6000 Hz_10ms_modulation_mae_db": 0.5514578491956943, "3000–6000 Hz_40ms_log_rms_mae_db": 1.793068573573156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31104940928781805, "3000–6000 Hz_40ms_envelope_correlation": 0.9324463638675, "3000–6000 Hz_40ms_modulation_mae_db": 0.5763084653685225, "mr_stft_db": 6.044010565725976, "band_mae_db": 0.45890133373759784, "rms_gain_error_db": -0.11265470469247259, "output_peak": 0.3956115368969384}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1808879562194885, "全帯域_10ms_rms_envelope_nrmse": 0.09515275307680991, "全帯域_10ms_envelope_correlation": 0.9914254693248938, "全帯域_10ms_modulation_mae_db": 0.6007519158787692, "全帯域_40ms_log_rms_mae_db": 0.5827952074086061, "全帯域_40ms_rms_envelope_nrmse": 0.049541920950126286, "全帯域_40ms_envelope_correlation": 0.997491587681069, "全帯域_40ms_modulation_mae_db": 0.32692086894992195, "300–1000 Hz_10ms_log_rms_mae_db": 1.8489792484420202, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.133747377329969, "300–1000 Hz_10ms_envelope_correlation": 0.9864778766270474, "300–1000 Hz_10ms_modulation_mae_db": 1.267699441833031, "300–1000 Hz_40ms_log_rms_mae_db": 1.2532624964442458, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06892192707430826, "300–1000 Hz_40ms_envelope_correlation": 0.9962647950347736, "300–1000 Hz_40ms_modulation_mae_db": 0.815225561573321, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8792144043626589, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2132845203350511, "1000–3000 Hz_10ms_envelope_correlation": 0.9657157986607005, "1000–3000 Hz_10ms_modulation_mae_db": 0.6586933395517113, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0720126969196098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1315499900001786, "1000–3000 Hz_40ms_envelope_correlation": 0.9860840404872578, "1000–3000 Hz_40ms_modulation_mae_db": 0.4468535304191949, "3000–6000 Hz_10ms_log_rms_mae_db": 2.342570842246644, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37632166205865025, "3000–6000 Hz_10ms_envelope_correlation": 0.9147134929773396, "3000–6000 Hz_10ms_modulation_mae_db": 0.15518290926255904, "3000–6000 Hz_40ms_log_rms_mae_db": 1.808560109003132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3106982951783973, "3000–6000 Hz_40ms_envelope_correlation": 0.9384884403796627, "3000–6000 Hz_40ms_modulation_mae_db": 0.3233622044251698, "mr_stft_db": 6.043049753592482, "band_mae_db": 0.3425091611131208, "rms_gain_error_db": -0.00340200706659315, "output_peak": 0.4788640391063838}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1559212290071585, "全帯域_10ms_rms_envelope_nrmse": 0.11668582597071739, "全帯域_10ms_envelope_correlation": 0.9870430145034126, "全帯域_10ms_modulation_mae_db": 0.345016122135335, "全帯域_40ms_log_rms_mae_db": 0.5917097448871713, "全帯域_40ms_rms_envelope_nrmse": 0.05214017822399293, "全帯域_40ms_envelope_correlation": 0.9972201272942645, "全帯域_40ms_modulation_mae_db": 0.18209630615603822, "300–1000 Hz_10ms_log_rms_mae_db": 1.8834118680217673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15590418990847185, "300–1000 Hz_10ms_envelope_correlation": 0.9815621627096357, "300–1000 Hz_10ms_modulation_mae_db": 1.1903458167903054, "300–1000 Hz_40ms_log_rms_mae_db": 1.2191115508945891, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07244806485747816, "300–1000 Hz_40ms_envelope_correlation": 0.9958642067161829, "300–1000 Hz_40ms_modulation_mae_db": 0.38474593078814817, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8591771954050003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19258959338229414, "1000–3000 Hz_10ms_envelope_correlation": 0.9728601742823361, "1000–3000 Hz_10ms_modulation_mae_db": 0.41908716087074344, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1110705458995467, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1257445440973603, "1000–3000 Hz_40ms_envelope_correlation": 0.9895405044242769, "1000–3000 Hz_40ms_modulation_mae_db": 0.3095945483088158, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3488697085548482, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3658758178783502, "3000–6000 Hz_10ms_envelope_correlation": 0.9117137172962663, "3000–6000 Hz_10ms_modulation_mae_db": 0.982689790232956, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8389625211129248, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3128357992059926, "3000–6000 Hz_40ms_envelope_correlation": 0.9311591509791142, "3000–6000 Hz_40ms_modulation_mae_db": 0.9264488655531341, "mr_stft_db": 6.039021372403464, "band_mae_db": 0.39119486701403366, "rms_gain_error_db": -0.049599195418289596, "output_peak": 0.5006104560607434}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1580646494505427, "全帯域_10ms_rms_envelope_nrmse": 0.12465505179972283, "全帯域_10ms_envelope_correlation": 0.9851887245866415, "全帯域_10ms_modulation_mae_db": 0.7014540393366464, "全帯域_40ms_log_rms_mae_db": 0.5613022798778752, "全帯域_40ms_rms_envelope_nrmse": 0.051271219724820814, "全帯域_40ms_envelope_correlation": 0.9973891043171876, "全帯域_40ms_modulation_mae_db": 0.509596823469595, "300–1000 Hz_10ms_log_rms_mae_db": 1.734557755771412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15104860017872573, "300–1000 Hz_10ms_envelope_correlation": 0.9827786958704482, "300–1000 Hz_10ms_modulation_mae_db": 1.2501414806168232, "300–1000 Hz_40ms_log_rms_mae_db": 1.11853145758402, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07157491080250813, "300–1000 Hz_40ms_envelope_correlation": 0.9961951323205841, "300–1000 Hz_40ms_modulation_mae_db": 0.7087474529014139, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8701413771979751, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19155895679769208, "1000–3000 Hz_10ms_envelope_correlation": 0.9719346360526122, "1000–3000 Hz_10ms_modulation_mae_db": 0.4450701603406501, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0826142206006892, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10566616193345826, "1000–3000 Hz_40ms_envelope_correlation": 0.9909487710342187, "1000–3000 Hz_40ms_modulation_mae_db": 0.48261481790531846, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3049910843393233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3690765211027787, "3000–6000 Hz_10ms_envelope_correlation": 0.922631667716502, "3000–6000 Hz_10ms_modulation_mae_db": 0.42547267760327984, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8027513461824405, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.316366026485514, "3000–6000 Hz_40ms_envelope_correlation": 0.9402272391550087, "3000–6000 Hz_40ms_modulation_mae_db": 0.6736066565417601, "mr_stft_db": 6.032148902591547, "band_mae_db": 0.4068895053949418, "rms_gain_error_db": -0.0966280478361526, "output_peak": 0.450045509419914}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.376878197580313, "spectral_convergence": 0.3554782893267541, "spectral_convergence_status": "ok", "amplitude_floor": 1.4536470931091146e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.652592123013152, "spectral_convergence": 0.3296916378347613, "spectral_convergence_status": "ok", "amplitude_floor": 9.268471977763907e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0853513318653, "spectral_convergence": 0.3811671685885452, "spectral_convergence_status": "ok", "amplitude_floor": 5.673364127394706e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031816153358243563, "output_power": 0.0003274304319004605, "delta_db": 0.12471369623281923, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009151577935124143, "output_power": 0.000841743062693758, "delta_db": -0.36316436733999125, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00046151207587702967, "output_power": 0.00043515756440100063, "delta_db": -0.2553653189301105, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011338295265990407, "output_power": 0.00011012584542318836, "delta_db": -0.1265850701265666, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.877003374326731e-05, "output_power": 3.9834595770734726e-05, "delta_db": 0.11764235443708682, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3868434770644032e-05, "output_power": 1.9562843362811302e-05, "delta_db": 1.494045294206769, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.41358601687889057, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.20794070153926258, "pitch": null, "output_rms_dbfs": -27.51082044249902, "output_peak": 0.3895092075985285, "reference_rms_dbfs": -27.30287974095976, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.038273884152922}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016375851184643225, "log_rms_mae_db": 1.143506609631696, "rms_envelope_nrmse": 0.10984710477420097, "envelope_correlation": 0.9888262213151996, "rms_gain_error_db": -0.20793163583809482, "reference_mean_power": 0.0018623503698243853, "output_mean_power": 0.0017752855818887196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.48847287729811e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002826666045552883, "output_power": 0.0002781331386687704, "reference_active": true, "delta_db": -0.07021765370437143}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.155794557866922e-05, "output_power": 5.933897109813953e-05, "reference_active": true, "delta_db": -0.15944105710289594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.629343961311588e-06, "output_power": 9.88580756928647e-06, "reference_active": true, "delta_db": 0.5903437241867184}], "active_band_count": 3, "active_band_mae_db": 0.2733341449979953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014981717790874295, "log_rms_mae_db": 0.6041422432736383, "rms_envelope_nrmse": 0.07484982339203639, "envelope_correlation": 0.9947860045112915, "rms_gain_error_db": -0.2079616918451382, "reference_mean_power": 0.0018713077892111094, "output_mean_power": 0.001783811897177925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.792361294396736e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025909451544035513, "output_power": 0.00025527273639877054, "reference_active": true, "delta_db": -0.06453786033204488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9672561350575063e-05, "output_power": 2.866221431258421e-05, "reference_active": true, "delta_db": -0.15045297477442365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.579324289571111e-07, "output_power": 7.775571799227372e-07, "reference_active": true, "delta_db": 0.11101846756157038}], "active_band_count": 3, "active_band_mae_db": 0.10866976755601299}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014318840727902982, "log_rms_mae_db": 1.8192610854240836, "rms_envelope_nrmse": 0.15350823786947534, "envelope_correlation": 0.9835770960803077, "rms_gain_error_db": -0.47276471851942803, "reference_mean_power": 0.0010573416450716808, "output_mean_power": 0.0009482849675435141}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.91879361710777e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021964908416761717, "output_power": 0.00020729776184984276, "reference_active": true, "delta_db": -0.2513478366997757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.090309412479679e-05, "output_power": 3.413799394339488e-05, "reference_active": true, "delta_db": -0.7851816455500749}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.52813465615998e-06, "output_power": 6.0192870815072975e-06, "reference_active": true, "delta_db": 1.2362572361183612}], "active_band_count": 3, "active_band_mae_db": 0.757595572789404}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013022613407710025, "log_rms_mae_db": 1.1962098102858976, "rms_envelope_nrmse": 0.11722505792572667, "envelope_correlation": 0.9909703798109416, "rms_gain_error_db": -0.4727806603875801, "reference_mean_power": 0.001062644583935512, "output_mean_power": 0.0009530374505844853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.561667844043168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020949277098297487, "output_power": 0.0001968890726650899, "reference_active": true, "delta_db": -0.2694742777534873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.887160269235181e-05, "output_power": 1.5862107614286397e-05, "reference_active": true, "delta_db": -0.754478927219026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.293974094910667e-07, "output_power": 4.199006211224699e-07, "reference_active": true, "delta_db": -0.09712903003211307}], "active_band_count": 3, "active_band_mae_db": 0.37369407833487545}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.227046231131274e-05, "log_rms_mae_db": 1.8823493872571633, "rms_envelope_nrmse": 0.19465173412647452, "envelope_correlation": 0.9709207373047798, "rms_gain_error_db": -0.1152649735093886, "reference_mean_power": 0.00011346719055744512, "output_mean_power": 0.0001104952997859652}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5656809677774876e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2313251419233995e-05, "output_power": 2.315997713537058e-05, "reference_active": true, "delta_db": 0.16175267331403678}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.900047334435587e-06, "output_power": 3.659502193592514e-06, "reference_active": true, "delta_db": 1.0101692532988265}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0009489505393011e-06, "output_power": 1.851093039057729e-06, "reference_active": true, "delta_db": 2.6701631911072115}], "active_band_count": 3, "active_band_mae_db": 1.280695039240025}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.589293372832859e-05, "log_rms_mae_db": 1.0906793764080704, "rms_envelope_nrmse": 0.11266613026813384, "envelope_correlation": 0.989747259996758, "rms_gain_error_db": -0.11546250609612382, "reference_mean_power": 0.0001140327901646041, "output_mean_power": 0.00011104103476067902}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.160520396268688e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1030857564945278e-05, "output_power": 2.2667479973730186e-05, "reference_active": true, "delta_db": 0.32546258651479304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3790687257240336e-06, "output_power": 1.8504612958475742e-06, "reference_active": true, "delta_db": 1.2769409611686964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.278096740186873e-08, "output_power": 1.188163680230094e-07, "reference_active": true, "delta_db": 2.128584488218044}], "active_band_count": 3, "active_band_mae_db": 1.2436626786338445}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9795267482339013e-05, "log_rms_mae_db": 2.3084330112418825, "rms_envelope_nrmse": 0.3504519756203869, "envelope_correlation": 0.9250056516353918, "rms_gain_error_db": 0.4180083393474283, "reference_mean_power": 2.1804807525571507e-05, "output_mean_power": 2.4007840182466557e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7142973306691612e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0022096022650052e-05, "output_power": 9.553201582356977e-06, "reference_active": true, "delta_db": -0.2080961754121879}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6747496282854573e-06, "output_power": 2.8666561406615583e-06, "reference_active": true, "delta_db": 0.3009246601496582}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.037734768320834e-07, "output_power": 9.054819923304653e-07, "reference_active": true, "delta_db": 0.5174614664646146}], "active_band_count": 3, "active_band_mae_db": 0.34216076734215356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.774754919111929e-05, "log_rms_mae_db": 1.7936023782174138, "rms_envelope_nrmse": 0.29915946211041133, "envelope_correlation": 0.9419540963314044, "rms_gain_error_db": 0.4179679400250047, "reference_mean_power": 2.1914321405186452e-05, "output_mean_power": 2.412819426774e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.567366861761873e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0140844119742473e-05, "output_power": 9.58943508823097e-06, "reference_active": true, "delta_db": -0.24281083277957358}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3436726086917478e-06, "output_power": 1.5021426152773909e-06, "reference_active": true, "delta_db": 0.4841770309439686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.915436539179574e-08, "output_power": 6.705679762141562e-08, "reference_active": true, "delta_db": 1.3488071582051613}], "active_band_count": 3, "active_band_mae_db": 0.6919316739762346}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8607023328844143, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2a3e425ac19beb44a7f9296ecdf4d81f5fad9b69e102b575db6f7780d05e3756", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9704302937963397, "standard_deviation": 0.0027210283412564635, "minimum": 0.9657157986607005, "maximum": 0.9738204897987419, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8634888411682315, "standard_deviation": 0.019962341892443707, "minimum": 1.8293068622593638, "maximum": 1.8823493872571633, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6354420027928176, "standard_deviation": 0.3166441725413212, "minimum": 0.34792794833213975, "maximum": 1.280695039240025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2004833192573169, "standard_deviation": 0.010067998015940556, "minimum": 0.1886615560623643, "maximum": 0.21404648225184358, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9885466223539164, "standard_deviation": 0.0016463653999880027, "minimum": 0.9860840404872578, "maximum": 0.9909487710342187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0907098559111819, "standard_deviation": 0.023339308436281847, "minimum": 1.0502066668227767, "maximum": 1.118334235738605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.587601228735955, "standard_deviation": 0.41496517922479625, "minimum": 0.1940610789645961, "maximum": 1.2436626786338445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12477730634429438, "standard_deviation": 0.01049968136627549, "minimum": 0.10566616193345826, "maximum": 0.13783448472953824, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9181297766510249, "standard_deviation": 0.004382272799405276, "minimum": 0.9117137172962663, "maximum": 0.9250056516353918, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.335447468990548, "standard_deviation": 0.022715166099595122, "minimum": 2.3049910843393233, "maximum": 2.373388839475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4876752729734243, "standard_deviation": 0.2511762790015633, "minimum": 0.15518290926255904, "maximum": 0.982689790232956, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.36659305514439167, "standard_deviation": 0.011401788135557733, "minimum": 0.3504519756203869, "maximum": 0.38791108161030846, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9375137192288444, "standard_deviation": 0.0038861470963117623, "minimum": 0.9311591509791142, "maximum": 0.9419540963314044, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8058474721495186, "standard_deviation": 0.0179192456729577, "minimum": 1.7801690926940172, "maximum": 1.8389625211129248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5993555848534333, "standard_deviation": 0.29515367184871155, "minimum": 0.20455072039373048, "maximum": 1.0429580306350077, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.31117602342486445, "standard_deviation": 0.008665417520810011, "minimum": 0.29915946211041133, "maximum": 0.328138340379438, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9830683857131457, "standard_deviation": 0.003424998890878353, "minimum": 0.976675023966562, "maximum": 0.9882258750297209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.810659506967007, "standard_deviation": 0.05164394232163298, "minimum": 1.734557755771412, "maximum": 1.8834118680217673, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0742992197876928, "standard_deviation": 0.2513610470146372, "minimum": 0.6232045397498648, "maximum": 1.3091373454630308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14993771158391653, "standard_deviation": 0.014653833303939223, "minimum": 0.12685023202638288, "maximum": 0.1755003885506249, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9948304098094886, "standard_deviation": 0.0029837231901898007, "minimum": 0.9892573967777633, "maximum": 0.9973395692987312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1931601117680595, "standard_deviation": 0.0480974498070592, "minimum": 1.11853145758402, "maximum": 1.2550205150636249, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4998202847315064, "standard_deviation": 0.23169278479933508, "minimum": 0.2080618836881217, "maximum": 0.815225561573321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08106531833476974, "standard_deviation": 0.02254652829699793, "minimum": 0.0647022833709667, "maximum": 0.11736347866123575, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.41323016629865117, "standard_deviation": 0.046380739217672065, "minimum": 0.3425091611131208, "maximum": 0.4852470451612893, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.036704311881018, "standard_deviation": 0.006408384803115367, "minimum": 6.024094897938498, "maximum": 6.044010565725976, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45584645166699195, "standard_deviation": 0.0455036180316533, "minimum": 0.3895092075985285, "maximum": 0.5025552434410288, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07095647386066195, "standard_deviation": 0.06650096055751951, "minimum": -0.20794070153926258, "maximum": -0.00340200706659315, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9879252888598262, "standard_deviation": 0.0022416474785693415, "minimum": 0.9851887245866415, "maximum": 0.9914254693248938, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1669428268167161, "standard_deviation": 0.014253816911500789, "minimum": 1.143506609631696, "maximum": 1.1809718302843797, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.43423375197807834, "standard_deviation": 0.1852497378153121, "minimum": 0.13251930505959417, "maximum": 0.7014540393366464, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11242081649184318, "standard_deviation": 0.010509105036429486, "minimum": 0.09515275307680991, "maximum": 0.12465505179972283, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968551459390445, "standard_deviation": 0.0011559425295601358, "minimum": 0.9947860045112915, "maximum": 0.9980995317140544, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5944170974510267, "standard_deviation": 0.021348705640596242, "minimum": 0.5613022798778752, "maximum": 0.6310875281901189, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28434593173845407, "standard_deviation": 0.13313346423439298, "minimum": 0.10866976755601299, "maximum": 0.509596823469595, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055308602748294786, "standard_deviation": 0.010572710732614482, "minimum": 0.04332754617260792, "maximum": 0.07484982339203639, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0008", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.482732057571411, "predicted_mos": 2.482732057571411, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "965f8bd29b9f5191873e7ccdf37ec1ef0562de61ebe6da0c5d5da51c4085d965", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.0258169998414814, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.171353140311317, "全帯域_10ms_rms_envelope_nrmse": 0.11227720461849947, "全帯域_10ms_envelope_correlation": 0.9882011219712732, "全帯域_10ms_modulation_mae_db": 0.3678579337178826, "全帯域_40ms_log_rms_mae_db": 0.6281246946377331, "全帯域_40ms_rms_envelope_nrmse": 0.07594847406150121, "全帯域_40ms_envelope_correlation": 0.9944858028913846, "全帯域_40ms_modulation_mae_db": 0.23733318924249472, "300–1000 Hz_10ms_log_rms_mae_db": 1.8098860376355694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1538151430965251, "300–1000 Hz_10ms_envelope_correlation": 0.9832287455030805, "300–1000 Hz_10ms_modulation_mae_db": 0.7778218843583122, "300–1000 Hz_40ms_log_rms_mae_db": 1.1877231173272826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11653546612655204, "300–1000 Hz_40ms_envelope_correlation": 0.9907894382133913, "300–1000 Hz_40ms_modulation_mae_db": 0.309412978963319, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8515460395290577, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17884299651664906, "1000–3000 Hz_10ms_envelope_correlation": 0.9755594679796491, "1000–3000 Hz_10ms_modulation_mae_db": 0.8417198777078694, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0325765098155695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10525796089482749, "1000–3000 Hz_40ms_envelope_correlation": 0.9913601706855457, "1000–3000 Hz_40ms_modulation_mae_db": 0.8330501861587237, "3000–6000 Hz_10ms_log_rms_mae_db": 2.246585599499018, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3432220483360929, "3000–6000 Hz_10ms_envelope_correlation": 0.926987449854994, "3000–6000 Hz_10ms_modulation_mae_db": 0.14914102722790934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7307572275160126, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29467143190290446, "3000–6000 Hz_40ms_envelope_correlation": 0.9425123604812531, "3000–6000 Hz_40ms_modulation_mae_db": 0.33925456381827096, "mr_stft_db": 5.973760144027086, "band_mae_db": 0.3963032134111825, "rms_gain_error_db": -0.18257198181027928, "output_peak": 0.39738224956040896}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1613142538514416, "全帯域_10ms_rms_envelope_nrmse": 0.10653909130085354, "全帯域_10ms_envelope_correlation": 0.9891837173254604, "全帯域_10ms_modulation_mae_db": 0.6138676627325431, "全帯域_40ms_log_rms_mae_db": 0.589351410361731, "全帯域_40ms_rms_envelope_nrmse": 0.04837976067627176, "全帯域_40ms_envelope_correlation": 0.9976148720273726, "全帯域_40ms_modulation_mae_db": 0.2944565252244756, "300–1000 Hz_10ms_log_rms_mae_db": 1.7460135720663919, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13485454005423203, "300–1000 Hz_10ms_envelope_correlation": 0.9865271392805584, "300–1000 Hz_10ms_modulation_mae_db": 1.3959952360297905, "300–1000 Hz_40ms_log_rms_mae_db": 1.121623616643395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06388300668162343, "300–1000 Hz_40ms_envelope_correlation": 0.9973495359472944, "300–1000 Hz_40ms_modulation_mae_db": 0.6244060237592785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8445946943892328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17739115420177276, "1000–3000 Hz_10ms_envelope_correlation": 0.9766026521110336, "1000–3000 Hz_10ms_modulation_mae_db": 0.09764727378576704, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0708164121531238, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10149316362950625, "1000–3000 Hz_40ms_envelope_correlation": 0.991946641299415, "1000–3000 Hz_40ms_modulation_mae_db": 0.4766101604593662, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2645675234918934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3889292627761709, "3000–6000 Hz_10ms_envelope_correlation": 0.9194392798569115, "3000–6000 Hz_10ms_modulation_mae_db": 0.6030047414924984, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7348195271634121, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3289440872861246, "3000–6000 Hz_40ms_envelope_correlation": 0.9402984631280558, "3000–6000 Hz_40ms_modulation_mae_db": 1.0680166737234629, "mr_stft_db": 5.959535583529841, "band_mae_db": 0.37224388245027407, "rms_gain_error_db": -0.050631264380074446, "output_peak": 0.5054614327496575}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2239331570068686, "全帯域_10ms_rms_envelope_nrmse": 0.12813439181540667, "全帯域_10ms_envelope_correlation": 0.9843609260775623, "全帯域_10ms_modulation_mae_db": 0.5372160333086428, "全帯域_40ms_log_rms_mae_db": 0.6561133655489629, "全帯域_40ms_rms_envelope_nrmse": 0.060021103167166204, "全帯域_40ms_envelope_correlation": 0.9963056813444965, "全帯域_40ms_modulation_mae_db": 0.3035266724741848, "300–1000 Hz_10ms_log_rms_mae_db": 1.8680396384160531, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16027092251307454, "300–1000 Hz_10ms_envelope_correlation": 0.9805495566724314, "300–1000 Hz_10ms_modulation_mae_db": 1.28345410964169, "300–1000 Hz_40ms_log_rms_mae_db": 1.1639377460986404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07940131134575935, "300–1000 Hz_40ms_envelope_correlation": 0.995037516398381, "300–1000 Hz_40ms_modulation_mae_db": 0.36442548820425463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8788316180181157, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19709702846737948, "1000–3000 Hz_10ms_envelope_correlation": 0.9727625433079923, "1000–3000 Hz_10ms_modulation_mae_db": 0.5252636986887395, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0817604965372758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1240442592170252, "1000–3000 Hz_40ms_envelope_correlation": 0.9892171566779946, "1000–3000 Hz_40ms_modulation_mae_db": 0.7155881108109873, "3000–6000 Hz_10ms_log_rms_mae_db": 2.295161223211488, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3554463716604455, "3000–6000 Hz_10ms_envelope_correlation": 0.9195432450350547, "3000–6000 Hz_10ms_modulation_mae_db": 0.7217193531183064, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7264022544639412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.299135247742221, "3000–6000 Hz_40ms_envelope_correlation": 0.9390361894696421, "3000–6000 Hz_40ms_modulation_mae_db": 0.5251944475457307, "mr_stft_db": 5.98057360909543, "band_mae_db": 0.404898504919885, "rms_gain_error_db": -0.05340167861132615, "output_peak": 0.4978237008485972}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1910202703138866, "全帯域_10ms_rms_envelope_nrmse": 0.10894592842971788, "全帯域_10ms_envelope_correlation": 0.988731548749719, "全帯域_10ms_modulation_mae_db": 0.06636996367309429, "全帯域_40ms_log_rms_mae_db": 0.6370526728648476, "全帯域_40ms_rms_envelope_nrmse": 0.044709203091726435, "全帯域_40ms_envelope_correlation": 0.9979751811302169, "全帯域_40ms_modulation_mae_db": 0.4410588537942928, "300–1000 Hz_10ms_log_rms_mae_db": 1.809608588756072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15137538695150715, "300–1000 Hz_10ms_envelope_correlation": 0.9826321134503341, "300–1000 Hz_10ms_modulation_mae_db": 0.41561932144671837, "300–1000 Hz_40ms_log_rms_mae_db": 1.1765535339579944, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06655466914979175, "300–1000 Hz_40ms_envelope_correlation": 0.9965634490495774, "300–1000 Hz_40ms_modulation_mae_db": 0.1722864804386763, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8945634736039871, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19083899488593872, "1000–3000 Hz_10ms_envelope_correlation": 0.9725120154772795, "1000–3000 Hz_10ms_modulation_mae_db": 0.38414996796959483, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0967615434953395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12036176036016384, "1000–3000 Hz_40ms_envelope_correlation": 0.9892929270939457, "1000–3000 Hz_40ms_modulation_mae_db": 0.2906135128363024, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3056542159390547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35603144566405415, "3000–6000 Hz_10ms_envelope_correlation": 0.9196558197912802, "3000–6000 Hz_10ms_modulation_mae_db": 0.43168730036504205, "3000–6000 Hz_40ms_log_rms_mae_db": 1.763227442337754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3026824667616907, "3000–6000 Hz_40ms_envelope_correlation": 0.9378679063023712, "3000–6000 Hz_40ms_modulation_mae_db": 0.2929833380588172, "mr_stft_db": 5.970120444410786, "band_mae_db": 0.3201534911735879, "rms_gain_error_db": -0.04325780368118742, "output_peak": 0.4266135386906337}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1566969587017548, "全帯域_10ms_rms_envelope_nrmse": 0.1170133562349953, "全帯域_10ms_envelope_correlation": 0.9869845398791748, "全帯域_10ms_modulation_mae_db": 0.3560349510705075, "全帯域_40ms_log_rms_mae_db": 0.581356030738992, "全帯域_40ms_rms_envelope_nrmse": 0.06468748426460212, "全帯域_40ms_envelope_correlation": 0.9958347565213777, "全帯域_40ms_modulation_mae_db": 0.27466909880508833, "300–1000 Hz_10ms_log_rms_mae_db": 1.8401552713657072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16671472563760206, "300–1000 Hz_10ms_envelope_correlation": 0.9790103752996457, "300–1000 Hz_10ms_modulation_mae_db": 0.9793015511053623, "300–1000 Hz_40ms_log_rms_mae_db": 1.2495992124873965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11116652941168692, "300–1000 Hz_40ms_envelope_correlation": 0.9904292116548302, "300–1000 Hz_40ms_modulation_mae_db": 0.5943277723228101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8458818330939974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19264424447253306, "1000–3000 Hz_10ms_envelope_correlation": 0.9744952479196219, "1000–3000 Hz_10ms_modulation_mae_db": 0.33883528921855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0547812697578374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13643262830624597, "1000–3000 Hz_40ms_envelope_correlation": 0.9892261886549437, "1000–3000 Hz_40ms_modulation_mae_db": 0.1866061943263225, "3000–6000 Hz_10ms_log_rms_mae_db": 2.291146724016288, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3579742813602336, "3000–6000 Hz_10ms_envelope_correlation": 0.9157628548939564, "3000–6000 Hz_10ms_modulation_mae_db": 0.5872571537054797, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7714482176463844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3080184653833894, "3000–6000 Hz_40ms_envelope_correlation": 0.9332993963105471, "3000–6000 Hz_40ms_modulation_mae_db": 0.4860155155211125, "mr_stft_db": 5.971101266497674, "band_mae_db": 0.5008968982946578, "rms_gain_error_db": -0.12349890144833511, "output_peak": 0.40797407313601564}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1827739058195506, "全帯域_10ms_rms_envelope_nrmse": 0.09432919429441115, "全帯域_10ms_envelope_correlation": 0.991553105572976, "全帯域_10ms_modulation_mae_db": 0.5008560517545299, "全帯域_40ms_log_rms_mae_db": 0.588101661792419, "全帯域_40ms_rms_envelope_nrmse": 0.04888553847955418, "全帯域_40ms_envelope_correlation": 0.9975566973492153, "全帯域_40ms_modulation_mae_db": 0.2391622794778786, "300–1000 Hz_10ms_log_rms_mae_db": 1.8622569832313478, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13143268604695096, "300–1000 Hz_10ms_envelope_correlation": 0.9869416631374852, "300–1000 Hz_10ms_modulation_mae_db": 1.1488400695621872, "300–1000 Hz_40ms_log_rms_mae_db": 1.2370174335997626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06916138096678653, "300–1000 Hz_40ms_envelope_correlation": 0.9962476797992528, "300–1000 Hz_40ms_modulation_mae_db": 0.6986818493398593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.885923252969467, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2089471418273312, "1000–3000 Hz_10ms_envelope_correlation": 0.9665673570317892, "1000–3000 Hz_10ms_modulation_mae_db": 0.17282789120967876, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0242420677969521, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12498097512656592, "1000–3000 Hz_40ms_envelope_correlation": 0.987307712849731, "1000–3000 Hz_40ms_modulation_mae_db": 0.411773537657053, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2702973906008554, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3575279717939166, "3000–6000 Hz_10ms_envelope_correlation": 0.9207904107813478, "3000–6000 Hz_10ms_modulation_mae_db": 0.1712746489556375, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7540610846842446, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29832344748398676, "3000–6000 Hz_40ms_envelope_correlation": 0.9411846078707863, "3000–6000 Hz_40ms_modulation_mae_db": 0.11835706854796156, "mr_stft_db": 5.977111003980007, "band_mae_db": 0.34786040918898586, "rms_gain_error_db": -0.02671129967221323, "output_peak": 0.48942428353684897}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1634600918129796, "全帯域_10ms_rms_envelope_nrmse": 0.12112563968655675, "全帯域_10ms_envelope_correlation": 0.9860613882385395, "全帯域_10ms_modulation_mae_db": 0.2827126890151302, "全帯域_40ms_log_rms_mae_db": 0.5979004199040324, "全帯域_40ms_rms_envelope_nrmse": 0.05396410960568912, "全帯域_40ms_envelope_correlation": 0.9970121845817759, "全帯域_40ms_modulation_mae_db": 0.32484173322721205, "300–1000 Hz_10ms_log_rms_mae_db": 1.8905382095267604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1662476516973854, "300–1000 Hz_10ms_envelope_correlation": 0.9790493253205242, "300–1000 Hz_10ms_modulation_mae_db": 0.9449157073966928, "300–1000 Hz_40ms_log_rms_mae_db": 1.2093204846996963, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07718314433550334, "300–1000 Hz_40ms_envelope_correlation": 0.9952465524520585, "300–1000 Hz_40ms_modulation_mae_db": 0.6078134970350401, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8338612931269556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19117255867966768, "1000–3000 Hz_10ms_envelope_correlation": 0.9739130739666594, "1000–3000 Hz_10ms_modulation_mae_db": 0.1931709883816999, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0663648135131103, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12527291333086338, "1000–3000 Hz_40ms_envelope_correlation": 0.9904160886600535, "1000–3000 Hz_40ms_modulation_mae_db": 0.20346377411511193, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2774143580072677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3586241125658871, "3000–6000 Hz_10ms_envelope_correlation": 0.9144550594559461, "3000–6000 Hz_10ms_modulation_mae_db": 1.1403576865341336, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7901477225261142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30870714569864405, "3000–6000 Hz_40ms_envelope_correlation": 0.9322782843958721, "3000–6000 Hz_40ms_modulation_mae_db": 1.121527391255627, "mr_stft_db": 5.98088470068877, "band_mae_db": 0.41551755359176296, "rms_gain_error_db": -0.030491890882986906, "output_peak": 0.5022843079017211}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1764185986505202, "全帯域_10ms_rms_envelope_nrmse": 0.12003841827941639, "全帯域_10ms_envelope_correlation": 0.9862637070947489, "全帯域_10ms_modulation_mae_db": 0.7879677156122039, "全帯域_40ms_log_rms_mae_db": 0.5736751633948297, "全帯域_40ms_rms_envelope_nrmse": 0.04767163434033973, "全帯域_40ms_envelope_correlation": 0.9977464014650539, "全帯域_40ms_modulation_mae_db": 0.5781218285209798, "300–1000 Hz_10ms_log_rms_mae_db": 1.7485266914038287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15121432004324073, "300–1000 Hz_10ms_envelope_correlation": 0.9827272360428858, "300–1000 Hz_10ms_modulation_mae_db": 1.4045232741090004, "300–1000 Hz_40ms_log_rms_mae_db": 1.078279555757581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06810618406927943, "300–1000 Hz_40ms_envelope_correlation": 0.9965666591303479, "300–1000 Hz_40ms_modulation_mae_db": 0.8736650326350409, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8852695937883412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19196427568637953, "1000–3000 Hz_10ms_envelope_correlation": 0.9718370177070169, "1000–3000 Hz_10ms_modulation_mae_db": 0.2944508301655106, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0641876543425974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10037702426735944, "1000–3000 Hz_40ms_envelope_correlation": 0.9921608884803742, "1000–3000 Hz_40ms_modulation_mae_db": 0.5592805891951648, "3000–6000 Hz_10ms_log_rms_mae_db": 2.241861631565241, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3616992134202932, "3000–6000 Hz_10ms_envelope_correlation": 0.9251731835624141, "3000–6000 Hz_10ms_modulation_mae_db": 0.2812634764109166, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7298916686434564, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3119133588325476, "3000–6000 Hz_40ms_envelope_correlation": 0.9411863617598821, "3000–6000 Hz_40ms_modulation_mae_db": 0.499300486269912, "mr_stft_db": 5.966697684673573, "band_mae_db": 0.4140490503085748, "rms_gain_error_db": -0.08948127007226912, "output_peak": 0.4583891367453002}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.291533962716603, "spectral_convergence": 0.35617786668986734, "spectral_convergence_status": "ok", "amplitude_floor": 1.4536470931091146e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.610752019517614, "spectral_convergence": 0.3303314843359083, "spectral_convergence_status": "ok", "amplitude_floor": 9.268471977763907e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.018994449847042, "spectral_convergence": 0.38341552765207215, "spectral_convergence_status": "ok", "amplitude_floor": 5.673364127394706e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031816153358243563, "output_power": 0.0003287128562477775, "delta_db": 0.14169018709227338, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009151577935124143, "output_power": 0.0008538506228303927, "delta_db": -0.3011408298169493, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00046151207587702967, "output_power": 0.0004367802863237322, "delta_db": -0.23920040557655645, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011338295265990407, "output_power": 0.00010672539895726229, "delta_db": -0.26279975529413474, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.877003374326731e-05, "output_power": 3.900243559024926e-05, "delta_db": 0.02595549325249523, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3868434770644032e-05, "output_power": 1.917479409803508e-05, "delta_db": 1.4070326094346859, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3963032134111825, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.18257198181027928, "pitch": null, "output_rms_dbfs": -27.485451722770037, "output_peak": 0.39738224956040896, "reference_rms_dbfs": -27.30287974095976, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.973760144027086}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016375851184643225, "log_rms_mae_db": 1.171353140311317, "rms_envelope_nrmse": 0.11227720461849947, "envelope_correlation": 0.9882011219712732, "rms_gain_error_db": -0.1825877657645021, "reference_mean_power": 0.0018623503698243853, "output_mean_power": 0.0017856757998446794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.48847287729811e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002826666045552883, "output_power": 0.0002781420077332155, "reference_active": true, "delta_db": -0.07007916879188168}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.155794557866922e-05, "output_power": 6.355580477250342e-05, "reference_active": true, "delta_db": 0.13871104972905152}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.629343961311588e-06, "output_power": 1.060366841040187e-05, "reference_active": true, "delta_db": 0.8947835826327145}], "active_band_count": 3, "active_band_mae_db": 0.3678579337178826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014981717790874295, "log_rms_mae_db": 0.6281246946377331, "rms_envelope_nrmse": 0.07594847406150121, "envelope_correlation": 0.9944858028913846, "rms_gain_error_db": -0.18263788556320681, "reference_mean_power": 0.0018713077892111094, "output_mean_power": 0.001794243727946988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.792361294396736e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025909451544035513, "output_power": 0.00025597284512840257, "reference_active": true, "delta_db": -0.05264324220477683}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9672561350575063e-05, "output_power": 3.063598966966363e-05, "reference_active": true, "delta_db": 0.1387687795416163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.579324289571111e-07, "output_power": 8.544550951153158e-07, "reference_active": true, "delta_db": 0.5205875459810911}], "active_band_count": 3, "active_band_mae_db": 0.23733318924249472}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014318840727902982, "log_rms_mae_db": 1.8098860376355694, "rms_envelope_nrmse": 0.1538151430965251, "envelope_correlation": 0.9832287455030805, "rms_gain_error_db": -0.43241808762247264, "reference_mean_power": 0.0010573416450716808, "output_mean_power": 0.0009571357307820829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.91879361710777e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021964908416761717, "output_power": 0.0002051830118414442, "reference_active": true, "delta_db": -0.2958799635253191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.090309412479679e-05, "output_power": 3.7175179639646294e-05, "reference_active": true, "delta_db": -0.4150308574756875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.52813465615998e-06, "output_power": 6.579226175407306e-06, "reference_active": true, "delta_db": 1.62255483207393}], "active_band_count": 3, "active_band_mae_db": 0.7778218843583122}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013022613407710025, "log_rms_mae_db": 1.1877231173272826, "rms_envelope_nrmse": 0.11653546612655204, "envelope_correlation": 0.9907894382133913, "rms_gain_error_db": -0.43242941568346266, "reference_mean_power": 0.001062644583935512, "output_mean_power": 0.0009619335927776213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.561667844043168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020949277098297487, "output_power": 0.0001957020397154607, "reference_active": true, "delta_db": -0.29573689117452007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.887160269235181e-05, "output_power": 1.7254218841566807e-05, "reference_active": true, "delta_db": -0.38913482632170304}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.293974094910667e-07, "output_power": 4.5414667825499423e-07, "reference_active": true, "delta_db": 0.24336721939373385}], "active_band_count": 3, "active_band_mae_db": 0.309412978963319}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.227046231131274e-05, "log_rms_mae_db": 1.8515460395290577, "rms_envelope_nrmse": 0.17884299651664906, "envelope_correlation": 0.9755594679796491, "rms_gain_error_db": -0.22737361184496352, "reference_mean_power": 0.00011346719055744512, "output_mean_power": 0.00010767947778991626}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5656809677774876e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2313251419233995e-05, "output_power": 2.2622038685068096e-05, "reference_active": true, "delta_db": 0.05968881753058746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.900047334435587e-06, "output_power": 3.30820161378647e-06, "reference_active": true, "delta_db": 0.5718688267468685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0009489505393011e-06, "output_power": 1.5480041742819764e-06, "reference_active": true, "delta_db": 1.8936019888461524}], "active_band_count": 3, "active_band_mae_db": 0.8417198777078694}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.589293372832859e-05, "log_rms_mae_db": 1.0325765098155695, "rms_envelope_nrmse": 0.10525796089482749, "envelope_correlation": 0.9913601706855457, "rms_gain_error_db": -0.22757567984035193, "reference_mean_power": 0.0001140327901646041, "output_mean_power": 0.00010821119244641304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.160520396268688e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1030857564945278e-05, "output_power": 2.2052776887829643e-05, "reference_active": true, "delta_db": 0.20606301590899487}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3790687257240336e-06, "output_power": 1.6555859883871202e-06, "reference_active": true, "delta_db": 0.7936583247970906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.278096740186873e-08, "output_power": 1.0279233836479699e-07, "reference_active": true, "delta_db": 1.4994292177700856}], "active_band_count": 3, "active_band_mae_db": 0.8330501861587237}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9795267482339013e-05, "log_rms_mae_db": 2.246585599499018, "rms_envelope_nrmse": 0.3432220483360929, "envelope_correlation": 0.926987449854994, "rms_gain_error_db": 0.32230039631639495, "reference_mean_power": 2.1804807525571507e-05, "output_mean_power": 2.348455293208963e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7142973306691612e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0022096022650052e-05, "output_power": 9.780969096156556e-06, "reference_active": true, "delta_db": -0.10576672726619873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6747496282854573e-06, "output_power": 2.7959380930130818e-06, "reference_active": true, "delta_db": 0.19244415299927659}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.037734768320834e-07, "output_power": 8.318689058442198e-07, "reference_active": true, "delta_db": 0.1492122014182527}], "active_band_count": 3, "active_band_mae_db": 0.14914102722790934}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.774754919111929e-05, "log_rms_mae_db": 1.7307572275160126, "rms_envelope_nrmse": 0.29467143190290446, "envelope_correlation": 0.9425123604812531, "rms_gain_error_db": 0.3222639738515806, "reference_mean_power": 2.1914321405186452e-05, "output_mean_power": 2.360230533047577e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.567366861761873e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0140844119742473e-05, "output_power": 9.885417706849416e-06, "reference_active": true, "delta_db": -0.11079081912208653}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3436726086917478e-06, "output_power": 1.4420277722887843e-06, "reference_active": true, "delta_db": 0.3068016063812787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.915436539179574e-08, "output_power": 5.643898828957351e-08, "reference_active": true, "delta_db": 0.6001712659514478}], "active_band_count": 3, "active_band_mae_db": 0.33925456381827096}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8552979065664841, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d7deae631e77fc6a7050cc6839add8f937e637fd11023d3d35d4008bc7201e90", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9730311719376302, "standard_deviation": 0.0030616898557449734, "minimum": 0.9665673570317892, "maximum": 0.9766026521110336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8650589748148942, "standard_deviation": 0.02344121118416599, "minimum": 1.8338612931269556, "maximum": 1.8945634736039871, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3560082271409263, "standard_deviation": 0.23792579475202535, "minimum": 0.09764727378576704, "maximum": 0.8417198777078694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19111229934220642, "standard_deviation": 0.009970350696281677, "minimum": 0.17739115420177276, "maximum": 0.2089471418273312, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9901159718002505, "standard_deviation": 0.0016619454878005696, "minimum": 0.987307712849731, "maximum": 0.9921608884803742, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.061436345926476, "standard_deviation": 0.024027901398268166, "minimum": 1.0242420677969521, "maximum": 1.0967615434953395, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.459623258194879, "standard_deviation": 0.23488468386577735, "minimum": 0.1866061943263225, "maximum": 0.8330501861587237, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11727758564156969, "standard_deviation": 0.013227306555243679, "minimum": 0.10037702426735944, "maximum": 0.13643262830624597, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9202259129039881, "standard_deviation": 0.004228204553234587, "minimum": 0.9144550594559461, "maximum": 0.926987449854994, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2740860832913885, "standard_deviation": 0.022811339494915348, "minimum": 2.241861631565241, "maximum": 2.3056542159390547, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5107131734762405, "standard_deviation": 0.3295446460286396, "minimum": 0.14914102722790934, "maximum": 1.1403576865341336, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3599318384471367, "standard_deviation": 0.012927515203412409, "minimum": 0.3432220483360929, "maximum": 0.3889292627761709, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9384579462148012, "standard_deviation": 0.003783159722391588, "minimum": 0.9322782843958721, "maximum": 0.9425123604812531, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.750094393122665, "standard_deviation": 0.023376366747211396, "minimum": 1.7264022544639412, "maximum": 1.7901477225261142, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5563311855926119, "standard_deviation": 0.3583156541298071, "minimum": 0.11835706854796156, "maximum": 1.121527391255627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3065494563864386, "standard_deviation": 0.01079539147611627, "minimum": 0.29467143190290446, "maximum": 0.3289440872861246, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9825832693383681, "standard_deviation": 0.003032673917009615, "minimum": 0.9790103752996457, "maximum": 0.9869416631374852, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8218781240502162, "standard_deviation": 0.05377175796277404, "minimum": 1.7460135720663919, "maximum": 1.8905382095267604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0438088942062191, "standard_deviation": 0.3384679239405448, "minimum": 0.41561932144671837, "maximum": 1.4045232741090004, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15199067200506478, "standard_deviation": 0.013143219715971764, "minimum": 0.13143268604695096, "maximum": 0.16671472563760206, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947787553306418, "standard_deviation": 0.0026797482938199398, "minimum": 0.9904292116548302, "maximum": 0.9973495359472944, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1780068375714685, "standard_deviation": 0.05733026992156428, "minimum": 1.078279555757581, "maximum": 1.2495992124873965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5306273903372849, "standard_deviation": 0.22981175116719746, "minimum": 0.1722864804386763, "maximum": 0.8736650326350409, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08149896151087285, "standard_deviation": 0.020688104846373612, "minimum": 0.06388300668162343, "maximum": 0.11653546612655204, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.39649037541736387, "standard_deviation": 0.054013067461910376, "minimum": 0.3201534911735879, "maximum": 0.5008968982946578, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.9724730546128955, "standard_deviation": 0.007247576732746341, "minimum": 5.959535583529841, "maximum": 5.98088470068877, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4606690903961479, "standard_deviation": 0.044549289443769395, "minimum": 0.39738224956040896, "maximum": 0.5054614327496575, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07500576131983397, "standard_deviation": 0.05413507151538769, "minimum": -0.18257198181027928, "maximum": -0.02671129967221323, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9876675068636818, "standard_deviation": 0.0022285130513004054, "minimum": 0.9843609260775623, "maximum": 0.991553105572976, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1783712970585398, "standard_deviation": 0.021684075241383407, "minimum": 1.1566969587017548, "maximum": 1.2239331570068686, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.43911037511056683, "standard_deviation": 0.2207302456367437, "minimum": 0.06636996367309429, "maximum": 0.7879677156122039, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11355040308248214, "standard_deviation": 0.010458072097888172, "minimum": 0.09432919429441115, "maximum": 0.12813439181540667, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968164471638616, "standard_deviation": 0.001200460989314065, "minimum": 0.9944858028913846, "maximum": 0.9979751811302169, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6064594274054435, "standard_deviation": 0.029949435842378146, "minimum": 0.5736751633948297, "maximum": 0.6561133655489629, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3366462725958258, "standard_deviation": 0.11676352694810292, "minimum": 0.23733318924249472, "maximum": 0.5781218285209798, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05553341346085634, "standard_deviation": 0.010673825636858094, "minimum": 0.044709203091726435, "maximum": 0.07594847406150121, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0008", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.477627992630005, "predicted_mos": 2.477627992630005, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "eb153bdd3292614aa1bb7522b9adadcb08ecf0710e0fe2391dab2b2e19629e43", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.54082030011341, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1445730840702846, "全帯域_10ms_rms_envelope_nrmse": 0.0947943326965816, "全帯域_10ms_envelope_correlation": 0.9915008536690739, "全帯域_10ms_modulation_mae_db": 0.4371422807252601, "全帯域_40ms_log_rms_mae_db": 0.5666766044362118, "全帯域_40ms_rms_envelope_nrmse": 0.04982178661312173, "全帯域_40ms_envelope_correlation": 0.997583212280713, "全帯域_40ms_modulation_mae_db": 0.29689947450559, "300–1000 Hz_10ms_log_rms_mae_db": 2.005552799333584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1249477580608113, "300–1000 Hz_10ms_envelope_correlation": 0.9887723872598694, "300–1000 Hz_10ms_modulation_mae_db": 1.0275129081842478, "300–1000 Hz_40ms_log_rms_mae_db": 1.2869627029566129, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07280404114689455, "300–1000 Hz_40ms_envelope_correlation": 0.9966429748839163, "300–1000 Hz_40ms_modulation_mae_db": 0.5731482053973974, "1000–3000 Hz_10ms_log_rms_mae_db": 2.138449140536845, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20547231987517728, "1000–3000 Hz_10ms_envelope_correlation": 0.9687020677126447, "1000–3000 Hz_10ms_modulation_mae_db": 0.6699572926611488, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2604486145791083, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1150478504591069, "1000–3000 Hz_40ms_envelope_correlation": 0.9897131641535102, "1000–3000 Hz_40ms_modulation_mae_db": 0.5304839508579567, "3000–6000 Hz_10ms_log_rms_mae_db": 2.89676292105822, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3798224108130272, "3000–6000 Hz_10ms_envelope_correlation": 0.9202750636835468, "3000–6000 Hz_10ms_modulation_mae_db": 0.3730998795164571, "3000–6000 Hz_40ms_log_rms_mae_db": 2.230532617278121, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32352541407344415, "3000–6000 Hz_40ms_envelope_correlation": 0.9397880658061718, "3000–6000 Hz_40ms_modulation_mae_db": 0.5414327481772947, "mr_stft_db": 6.367914355998004, "band_mae_db": 0.4149547318669777, "rms_gain_error_db": -0.10662558514643417, "output_peak": 0.3801538195495765}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.140972007111288, "全帯域_10ms_rms_envelope_nrmse": 0.10564621095111129, "全帯域_10ms_envelope_correlation": 0.9894231560781844, "全帯域_10ms_modulation_mae_db": 0.0807577140841742, "全帯域_40ms_log_rms_mae_db": 0.5329172297404786, "全帯域_40ms_rms_envelope_nrmse": 0.04569917041726035, "全帯域_40ms_envelope_correlation": 0.9978563056993934, "全帯域_40ms_modulation_mae_db": 0.32361205870481197, "300–1000 Hz_10ms_log_rms_mae_db": 1.986068390025256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15328903865071455, "300–1000 Hz_10ms_envelope_correlation": 0.9822823033817244, "300–1000 Hz_10ms_modulation_mae_db": 0.5971880050497713, "300–1000 Hz_40ms_log_rms_mae_db": 1.3399510471915685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0714109703572609, "300–1000 Hz_40ms_envelope_correlation": 0.9959224189132347, "300–1000 Hz_40ms_modulation_mae_db": 0.2784596772628751, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1588678398035595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2329664826228081, "1000–3000 Hz_10ms_envelope_correlation": 0.9618718255553522, "1000–3000 Hz_10ms_modulation_mae_db": 0.6112482188997989, "1000–3000 Hz_40ms_log_rms_mae_db": 1.25332780232673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13515351598035213, "1000–3000 Hz_40ms_envelope_correlation": 0.9872018936136423, "1000–3000 Hz_40ms_modulation_mae_db": 0.5353155415143822, "3000–6000 Hz_10ms_log_rms_mae_db": 2.896504328784102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3742581867610989, "3000–6000 Hz_10ms_envelope_correlation": 0.9195310479287793, "3000–6000 Hz_10ms_modulation_mae_db": 0.15835614964286363, "3000–6000 Hz_40ms_log_rms_mae_db": 2.258211341831078, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3189790652753045, "3000–6000 Hz_40ms_envelope_correlation": 0.9388752686056532, "3000–6000 Hz_40ms_modulation_mae_db": 0.5687273445883455, "mr_stft_db": 6.350637564056327, "band_mae_db": 0.4184252145539067, "rms_gain_error_db": 0.0044180601820761846, "output_peak": 0.40813298351341437}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1615637713877986, "全帯域_10ms_rms_envelope_nrmse": 0.11956540471583568, "全帯域_10ms_envelope_correlation": 0.9863760777083808, "全帯域_10ms_modulation_mae_db": 0.35502695860897715, "全帯域_40ms_log_rms_mae_db": 0.5945608440685448, "全帯域_40ms_rms_envelope_nrmse": 0.061455448385350255, "全帯域_40ms_envelope_correlation": 0.9961043006538163, "全帯域_40ms_modulation_mae_db": 0.3502130149794689, "300–1000 Hz_10ms_log_rms_mae_db": 2.0494749194673894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15465147183893413, "300–1000 Hz_10ms_envelope_correlation": 0.9818605838228781, "300–1000 Hz_10ms_modulation_mae_db": 0.5617959061500402, "300–1000 Hz_40ms_log_rms_mae_db": 1.395973684962687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08194505345852021, "300–1000 Hz_40ms_envelope_correlation": 0.9946619907819048, "300–1000 Hz_40ms_modulation_mae_db": 0.334102791340572, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1463599409952763, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.205813111135956, "1000–3000 Hz_10ms_envelope_correlation": 0.968280068302331, "1000–3000 Hz_10ms_modulation_mae_db": 0.49597588577490276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2830388606842256, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12023345557913862, "1000–3000 Hz_40ms_envelope_correlation": 0.9885535259897192, "1000–3000 Hz_40ms_modulation_mae_db": 0.484392190199038, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8757564296676548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39291788729173416, "3000–6000 Hz_10ms_envelope_correlation": 0.9053627241275833, "3000–6000 Hz_10ms_modulation_mae_db": 0.44667618183295826, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2351947173048727, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32209182715515783, "3000–6000 Hz_40ms_envelope_correlation": 0.9328460990806102, "3000–6000 Hz_40ms_modulation_mae_db": 0.603709828642657, "mr_stft_db": 6.3484072027741325, "band_mae_db": 0.4222805749431506, "rms_gain_error_db": -0.01888416807786344, "output_peak": 0.3934790721621443}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1568957697889846, "全帯域_10ms_rms_envelope_nrmse": 0.09942503189523035, "全帯域_10ms_envelope_correlation": 0.990778539021747, "全帯域_10ms_modulation_mae_db": 0.27499512639386464, "全帯域_40ms_log_rms_mae_db": 0.6039479461805444, "全帯域_40ms_rms_envelope_nrmse": 0.052304159994227335, "全帯域_40ms_envelope_correlation": 0.9973052222034955, "全帯域_40ms_modulation_mae_db": 0.20261202032842943, "300–1000 Hz_10ms_log_rms_mae_db": 2.0891987519012662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12096454485003079, "300–1000 Hz_10ms_envelope_correlation": 0.988953276465308, "300–1000 Hz_10ms_modulation_mae_db": 0.754122976086837, "300–1000 Hz_40ms_log_rms_mae_db": 1.3958218082764944, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0703727607722249, "300–1000 Hz_40ms_envelope_correlation": 0.9960755851886212, "300–1000 Hz_40ms_modulation_mae_db": 0.49776784550462966, "1000–3000 Hz_10ms_log_rms_mae_db": 2.258788351729779, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2349392414096148, "1000–3000 Hz_10ms_envelope_correlation": 0.9589596687058108, "1000–3000 Hz_10ms_modulation_mae_db": 0.8861789788478598, "1000–3000 Hz_40ms_log_rms_mae_db": 1.360119438000905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14704408416533302, "1000–3000 Hz_40ms_envelope_correlation": 0.9829694979568291, "1000–3000 Hz_40ms_modulation_mae_db": 0.6252295646716806, "3000–6000 Hz_10ms_log_rms_mae_db": 2.867345340959095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3772451717307106, "3000–6000 Hz_10ms_envelope_correlation": 0.9137050322103741, "3000–6000 Hz_10ms_modulation_mae_db": 0.43273500505050416, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2173887717359713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3154750531572755, "3000–6000 Hz_40ms_envelope_correlation": 0.9364155084942113, "3000–6000 Hz_40ms_modulation_mae_db": 0.7300802792923643, "mr_stft_db": 6.357615361048498, "band_mae_db": 0.37741942323293537, "rms_gain_error_db": 0.053241728875081565, "output_peak": 0.4126673873180009}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1196597282371952, "全帯域_10ms_rms_envelope_nrmse": 0.10278898486786839, "全帯域_10ms_envelope_correlation": 0.9900284365887025, "全帯域_10ms_modulation_mae_db": 0.193265438343919, "全帯域_40ms_log_rms_mae_db": 0.5776935561222294, "全帯域_40ms_rms_envelope_nrmse": 0.04818619705811917, "全帯域_40ms_envelope_correlation": 0.9976607683740122, "全帯域_40ms_modulation_mae_db": 0.3992782297452417, "300–1000 Hz_10ms_log_rms_mae_db": 2.1261449277917066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1370299575777213, "300–1000 Hz_10ms_envelope_correlation": 0.9858337834612656, "300–1000 Hz_10ms_modulation_mae_db": 0.8417738818456243, "300–1000 Hz_40ms_log_rms_mae_db": 1.4640515701680903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06995938858804772, "300–1000 Hz_40ms_envelope_correlation": 0.9960908650654824, "300–1000 Hz_40ms_modulation_mae_db": 0.4931542648170561, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1895383603298555, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21030078807690591, "1000–3000 Hz_10ms_envelope_correlation": 0.9670520475765696, "1000–3000 Hz_10ms_modulation_mae_db": 0.9590710352288642, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3121877142787468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13175967059561275, "1000–3000 Hz_40ms_envelope_correlation": 0.9863236777942922, "1000–3000 Hz_40ms_modulation_mae_db": 1.0701325969107436, "3000–6000 Hz_10ms_log_rms_mae_db": 2.884207051663756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3808241848279394, "3000–6000 Hz_10ms_envelope_correlation": 0.9147092262978854, "3000–6000 Hz_10ms_modulation_mae_db": 0.4538269475902547, "3000–6000 Hz_40ms_log_rms_mae_db": 2.266654380250894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3199745568359475, "3000–6000 Hz_40ms_envelope_correlation": 0.9368792507292535, "3000–6000 Hz_40ms_modulation_mae_db": 0.7306201664150077, "mr_stft_db": 6.3602553343616774, "band_mae_db": 0.39897718347128136, "rms_gain_error_db": -0.00146188225925981, "output_peak": 0.45343465735027266}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0974149858099387, "全帯域_10ms_rms_envelope_nrmse": 0.09460111510391159, "全帯域_10ms_envelope_correlation": 0.9916695465803819, "全帯域_10ms_modulation_mae_db": 0.22220082965866758, "全帯域_40ms_log_rms_mae_db": 0.5555311359982699, "全帯域_40ms_rms_envelope_nrmse": 0.043414707777922695, "全帯域_40ms_envelope_correlation": 0.9981659977681734, "全帯域_40ms_modulation_mae_db": 0.24263924092122646, "300–1000 Hz_10ms_log_rms_mae_db": 1.9896911739739862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12333653705040788, "300–1000 Hz_10ms_envelope_correlation": 0.9885963381160756, "300–1000 Hz_10ms_modulation_mae_db": 0.6198195198759376, "300–1000 Hz_40ms_log_rms_mae_db": 1.3166485154304943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06224138330089746, "300–1000 Hz_40ms_envelope_correlation": 0.9969270239098594, "300–1000 Hz_40ms_modulation_mae_db": 0.33575551956497646, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1743971387170498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22668406118529302, "1000–3000 Hz_10ms_envelope_correlation": 0.9610136888566189, "1000–3000 Hz_10ms_modulation_mae_db": 0.646155358036912, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2987020813522996, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12405260179460847, "1000–3000 Hz_40ms_envelope_correlation": 0.9875927673485912, "1000–3000 Hz_40ms_modulation_mae_db": 0.6003047376282018, "3000–6000 Hz_10ms_log_rms_mae_db": 2.870393974735406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39574265947446985, "3000–6000 Hz_10ms_envelope_correlation": 0.9023644266027792, "3000–6000 Hz_10ms_modulation_mae_db": 0.36249123558696733, "3000–6000 Hz_40ms_log_rms_mae_db": 2.187227734073398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3186930713444972, "3000–6000 Hz_40ms_envelope_correlation": 0.9330866373174524, "3000–6000 Hz_40ms_modulation_mae_db": 0.636545764994724, "mr_stft_db": 6.359121094145319, "band_mae_db": 0.42272703799154615, "rms_gain_error_db": 0.07891454479761492, "output_peak": 0.3847294135744102}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1528449862370325, "全帯域_10ms_rms_envelope_nrmse": 0.09836425329125023, "全帯域_10ms_envelope_correlation": 0.9908630118528674, "全帯域_10ms_modulation_mae_db": 0.0630895226622466, "全帯域_40ms_log_rms_mae_db": 0.581994669649903, "全帯域_40ms_rms_envelope_nrmse": 0.04815250628122817, "全帯域_40ms_envelope_correlation": 0.9978213277053933, "全帯域_40ms_modulation_mae_db": 0.2555418066490827, "300–1000 Hz_10ms_log_rms_mae_db": 2.091981827871706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13590675671405938, "300–1000 Hz_10ms_envelope_correlation": 0.986822066876181, "300–1000 Hz_10ms_modulation_mae_db": 0.45079496730091, "300–1000 Hz_40ms_log_rms_mae_db": 1.3984501181034663, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07679274044647481, "300–1000 Hz_40ms_envelope_correlation": 0.9965339830032641, "300–1000 Hz_40ms_modulation_mae_db": 0.7540502835846453, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1600563850947565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21407370175176435, "1000–3000 Hz_10ms_envelope_correlation": 0.9655867883931653, "1000–3000 Hz_10ms_modulation_mae_db": 0.7171110837634078, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2945667034051842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11893791262651546, "1000–3000 Hz_40ms_envelope_correlation": 0.9887141228463647, "1000–3000 Hz_40ms_modulation_mae_db": 0.7218930441514746, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9250081261244993, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3833682492098112, "3000–6000 Hz_10ms_envelope_correlation": 0.9130500717081699, "3000–6000 Hz_10ms_modulation_mae_db": 0.5674777631929392, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2848417041681643, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3207285511122669, "3000–6000 Hz_40ms_envelope_correlation": 0.9360804847568085, "3000–6000 Hz_40ms_modulation_mae_db": 1.019270307627725, "mr_stft_db": 6.364189355662011, "band_mae_db": 0.39779506143681415, "rms_gain_error_db": -0.12041861968494902, "output_peak": 0.40563672751664903}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1095595197344386, "全帯域_10ms_rms_envelope_nrmse": 0.11433449509786293, "全帯域_10ms_envelope_correlation": 0.9878168667087828, "全帯域_10ms_modulation_mae_db": 0.17653449401507015, "全帯域_40ms_log_rms_mae_db": 0.5836690305804332, "全帯域_40ms_rms_envelope_nrmse": 0.06558157258785517, "全帯域_40ms_envelope_correlation": 0.9957119437684847, "全帯域_40ms_modulation_mae_db": 0.14150901888009618, "300–1000 Hz_10ms_log_rms_mae_db": 1.958530051893272, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14530747997213336, "300–1000 Hz_10ms_envelope_correlation": 0.9842191159430586, "300–1000 Hz_10ms_modulation_mae_db": 0.6434542473276815, "300–1000 Hz_40ms_log_rms_mae_db": 1.3415125717339744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08492870060461667, "300–1000 Hz_40ms_envelope_correlation": 0.9942907361431812, "300–1000 Hz_40ms_modulation_mae_db": 0.2321978372828306, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1710191049269145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1990786795928108, "1000–3000 Hz_10ms_envelope_correlation": 0.9701871653491927, "1000–3000 Hz_10ms_modulation_mae_db": 0.5246942128903351, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2705998667629432, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11525092263312109, "1000–3000 Hz_40ms_envelope_correlation": 0.9893765521950688, "1000–3000 Hz_40ms_modulation_mae_db": 0.06920596590809673, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8449136577263907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.38393147150260143, "3000–6000 Hz_10ms_envelope_correlation": 0.916238015733181, "3000–6000 Hz_10ms_modulation_mae_db": 0.5434415357199706, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1942415625031786, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31746425452037985, "3000–6000 Hz_40ms_envelope_correlation": 0.9405083030902939, "3000–6000 Hz_40ms_modulation_mae_db": 0.8067151757898817, "mr_stft_db": 6.334800134842259, "band_mae_db": 0.42720813528598534, "rms_gain_error_db": 0.06566761883470906, "output_peak": 0.554085999763889}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.765813761699185, "spectral_convergence": 0.3736169961563948, "spectral_convergence_status": "ok", "amplitude_floor": 1.4536470931091146e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.047368375911399, "spectral_convergence": 0.3173520143663204, "spectral_convergence_status": "ok", "amplitude_floor": 9.268471977763907e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.290560930383429, "spectral_convergence": 0.37033175064689505, "spectral_convergence_status": "ok", "amplitude_floor": 5.673364127394706e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031816153358243563, "output_power": 0.00033354879182571635, "delta_db": 0.205117005557672, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009151577935124143, "output_power": 0.0008455646016799578, "delta_db": -0.3434918894986827, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00046151207587702967, "output_power": 0.00045887363920270655, "delta_db": -0.024899596083446836, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011338295265990407, "output_power": 0.0001173260658398197, "delta_db": 0.1484674574468844, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.877003374326731e-05, "output_power": 4.042878356481611e-05, "delta_db": 0.18194495335111832, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3868434770644032e-05, "output_power": 1.998058488834455e-05, "delta_db": 1.5858074892640615, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4149547318669777, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10662558514643417, "pitch": null, "output_rms_dbfs": -27.409505326106192, "output_peak": 0.3801538195495765, "reference_rms_dbfs": -27.30287974095976, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.367914355998004}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016375851184643225, "log_rms_mae_db": 1.1445730840702846, "rms_envelope_nrmse": 0.0947943326965816, "envelope_correlation": 0.9915008536690739, "rms_gain_error_db": -0.10664807975168238, "reference_mean_power": 0.0018623503698243853, "output_mean_power": 0.0018171742812206134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.48847287729811e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002826666045552883, "output_power": 0.0002787571039153095, "reference_active": true, "delta_db": -0.06048558242181831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.155794557866922e-05, "output_power": 5.656842430721246e-05, "reference_active": true, "delta_db": -0.3671003555802247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.629343961311588e-06, "output_power": 1.057698456314168e-05, "reference_active": true, "delta_db": 0.8838409041737374}], "active_band_count": 3, "active_band_mae_db": 0.4371422807252601}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014981717790874295, "log_rms_mae_db": 0.5666766044362118, "rms_envelope_nrmse": 0.04982178661312173, "envelope_correlation": 0.997583212280713, "rms_gain_error_db": -0.10659563708376574, "reference_mean_power": 0.0018713077892111094, "output_mean_power": 0.001825936464133974}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.792361294396736e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025909451544035513, "output_power": 0.00025641268782998353, "reference_active": true, "delta_db": -0.04518708691407172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9672561350575063e-05, "output_power": 2.7743374834834746e-05, "reference_active": true, "delta_db": -0.2919574687751615}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.579324289571111e-07, "output_power": 8.609657516577968e-07, "reference_active": true, "delta_db": 0.5535538678275368}], "active_band_count": 3, "active_band_mae_db": 0.29689947450559}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014318840727902982, "log_rms_mae_db": 2.005552799333584, "rms_envelope_nrmse": 0.1249477580608113, "envelope_correlation": 0.9887723872598694, "rms_gain_error_db": -0.29852616268228305, "reference_mean_power": 0.0010573416450716808, "output_mean_power": 0.0009871035678444397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.91879361710777e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021964908416761717, "output_power": 0.0002097489149128595, "reference_active": true, "delta_db": -0.20029674018108737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.090309412479679e-05, "output_power": 3.306299235048971e-05, "reference_active": true, "delta_db": -0.9241400493384777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.52813465615998e-06, "output_power": 7.107707165918184e-06, "reference_active": true, "delta_db": 1.958101935033178}], "active_band_count": 3, "active_band_mae_db": 1.0275129081842478}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013022613407710025, "log_rms_mae_db": 1.2869627029566129, "rms_envelope_nrmse": 0.07280404114689455, "envelope_correlation": 0.9966429748839163, "rms_gain_error_db": -0.2985516449722802, "reference_mean_power": 0.001062644583935512, "output_mean_power": 0.0009920484172629789}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.561667844043168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020949277098297487, "output_power": 0.0002018168214413843, "reference_active": true, "delta_db": -0.16211679382919458}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.887160269235181e-05, "output_power": 1.567135620017561e-05, "reference_active": true, "delta_db": -0.8070220271728956}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.293974094910667e-07, "output_power": 5.103757129171022e-07, "reference_active": true, "delta_db": 0.7503057951901021}], "active_band_count": 3, "active_band_mae_db": 0.5731482053973974}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.227046231131274e-05, "log_rms_mae_db": 2.138449140536845, "rms_envelope_nrmse": 0.20547231987517728, "envelope_correlation": 0.9687020677126447, "rms_gain_error_db": 0.15637644014891663, "reference_mean_power": 0.00011346719055744512, "output_mean_power": 0.00011762725056969855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5656809677774876e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2313251419233995e-05, "output_power": 2.2304245715359807e-05, "reference_active": true, "delta_db": -0.0017531810623664062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.900047334435587e-06, "output_power": 3.4147731647081246e-06, "reference_active": true, "delta_db": 0.7095677333126972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0009489505393011e-06, "output_power": 1.3497925442288448e-06, "reference_active": true, "delta_db": 1.298550963608383}], "active_band_count": 3, "active_band_mae_db": 0.6699572926611488}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.589293372832859e-05, "log_rms_mae_db": 1.2604486145791083, "rms_envelope_nrmse": 0.1150478504591069, "envelope_correlation": 0.9897131641535102, "rms_gain_error_db": 0.15626843447470926, "reference_mean_power": 0.0001140327901646041, "output_mean_power": 0.00011821064697329285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.160520396268688e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1030857564945278e-05, "output_power": 2.145238608249939e-05, "reference_active": true, "delta_db": 0.08618622378170726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3790687257240336e-06, "output_power": 1.6504594767587082e-06, "reference_active": true, "delta_db": 0.7801895596540515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.278096740186873e-08, "output_power": 8.600534272584558e-08, "reference_active": true, "delta_db": 0.7250760691381117}], "active_band_count": 3, "active_band_mae_db": 0.5304839508579567}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9795267482339013e-05, "log_rms_mae_db": 2.89676292105822, "rms_envelope_nrmse": 0.3798224108130272, "envelope_correlation": 0.9202750636835468, "rms_gain_error_db": 0.827037101498763, "reference_mean_power": 2.1804807525571507e-05, "output_mean_power": 2.6378856636843302e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7142973306691612e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0022096022650052e-05, "output_power": 9.464674160284636e-06, "reference_active": true, "delta_db": -0.2485289231331973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6747496282854573e-06, "output_power": 2.80527305954789e-06, "reference_active": true, "delta_db": 0.20692005150733528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.037734768320834e-07, "output_power": 9.365236250393441e-07, "reference_active": true, "delta_db": 0.6638506639088387}], "active_band_count": 3, "active_band_mae_db": 0.3730998795164571}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.774754919111929e-05, "log_rms_mae_db": 2.230532617278121, "rms_envelope_nrmse": 0.32352541407344415, "envelope_correlation": 0.9397880658061718, "rms_gain_error_db": 0.8269893705095057, "reference_mean_power": 2.1914321405186452e-05, "output_mean_power": 2.651105214696575e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.567366861761873e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0140844119742473e-05, "output_power": 9.536284104046239e-06, "reference_active": true, "delta_db": -0.2669492595036161}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3436726086917478e-06, "output_power": 1.459865425576298e-06, "reference_active": true, "delta_db": 0.3601935917935662}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.915436539179574e-08, "output_power": 6.184116077821113e-08, "reference_active": true, "delta_db": 0.9971553932347019}], "active_band_count": 3, "active_band_mae_db": 0.5414327481772947}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8503914922072315, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bcffd336070fb71860aba4c7eb9e7d4ed35c6b76596ffae9dfcdc94d923e05e4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9652066650564606, "standard_deviation": 0.004101326059402614, "minimum": 0.9589596687058108, "maximum": 0.9701871653491927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1746845327667543, "standard_deviation": 0.037582372660807134, "minimum": 2.138449140536845, "maximum": 2.258788351729779, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6887990082629036, "standard_deviation": 0.162615300404756, "minimum": 0.49597588577490276, "maximum": 0.9590710352288642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21616604820629126, "standard_deviation": 0.013615183039877235, "minimum": 0.1990786795928108, "maximum": 0.2349392414096148, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9875556502372522, "standard_deviation": 0.0021707613973770946, "minimum": 0.9829694979568291, "maximum": 0.9897131641535102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.291623885173768, "standard_deviation": 0.03407444018729123, "minimum": 1.25332780232673, "maximum": 1.360119438000905, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5796196989801967, "standard_deviation": 0.2770560479408618, "minimum": 0.06920596590809673, "maximum": 1.0701325969107436, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12593500172922356, "standard_deviation": 0.011226187856542075, "minimum": 0.1150478504591069, "maximum": 0.14704408416533302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9131544510365375, "standard_deviation": 0.006332901387928401, "minimum": 0.9023644266027792, "maximum": 0.9202750636835468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.88261147883989, "standard_deviation": 0.02403490300803824, "minimum": 2.8449136577263907, "maximum": 2.9250081261244993, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4172630872666144, "standard_deviation": 0.12700394986900626, "minimum": 0.15835614964286363, "maximum": 0.5674777631929392, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.38351377770142403, "standard_deviation": 0.007406560384065437, "minimum": 0.3742581867610989, "maximum": 0.39574265947446985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9368099522350568, "standard_deviation": 0.0028555761979234535, "minimum": 0.9328460990806102, "maximum": 0.9405083030902939, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.23428660364321, "standard_deviation": 0.03442275115859035, "minimum": 2.187227734073398, "maximum": 2.2848417041681643, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.704637701941, "standard_deviation": 0.15614385770151282, "minimum": 0.5414327481772947, "maximum": 1.019270307627725, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3196164741842842, "standard_deviation": 0.0025583058622308257, "minimum": 0.3154750531572755, "maximum": 0.32352541407344415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9859174819157951, "standard_deviation": 0.002876641769304108, "minimum": 0.9818605838228781, "maximum": 0.988953276465308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0370803552822707, "standard_deviation": 0.06074385836950887, "minimum": 1.958530051893272, "maximum": 2.1261449277917066, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6870578014776312, "standard_deviation": 0.181404830949978, "minimum": 0.45079496730091, "maximum": 1.0275129081842478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1369291930893516, "standard_deviation": 0.013288878317304052, "minimum": 0.12096454485003079, "maximum": 0.15465147183893413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958931972361831, "standard_deviation": 0.0009410111007030264, "minimum": 0.9942907361431812, "maximum": 0.9969270239098594, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3674215023529235, "standard_deviation": 0.05656773778097839, "minimum": 1.2869627029566129, "maximum": 1.4640515701680903, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4373295530943728, "standard_deviation": 0.17477316726554035, "minimum": 0.2321978372828306, "maximum": 0.7540502835846453, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07380687983436715, "standard_deviation": 0.0072303257298893765, "minimum": 0.06224138330089746, "maximum": 0.08492870060461667, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4099734203478247, "standard_deviation": 0.01705618817469661, "minimum": 0.37741942323293537, "maximum": 0.42720813528598534, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.355367550361028, "standard_deviation": 0.010502493219772547, "minimum": 6.334800134842259, "maximum": 6.367914355998004, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42404000759354465, "standard_deviation": 0.057199334681577924, "minimum": 0.3801538195495765, "maximum": 0.554085999763889, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.005643537809878087, "standard_deviation": 0.0749929088263914, "minimum": -0.12041861968494902, "maximum": 0.07891454479761492, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.989807061026015, "standard_deviation": 0.001864154258553397, "minimum": 0.9863760777083808, "maximum": 0.9916695465803819, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1354354815471202, "standard_deviation": 0.023677804973351584, "minimum": 1.0974149858099387, "maximum": 1.1615637713877986, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.22537654556152242, "standard_deviation": 0.12805936031695597, "minimum": 0.0630895226622466, "maximum": 0.4371422807252601, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10368997857745652, "standard_deviation": 0.009088977600037036, "minimum": 0.09460111510391159, "maximum": 0.11956540471583568, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972761348066852, "standard_deviation": 0.0008854913220850981, "minimum": 0.9957119437684847, "maximum": 0.9981659977681734, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5746238770970769, "standard_deviation": 0.022583452641527984, "minimum": 0.5329172297404786, "maximum": 0.6039479461805444, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2765381080892434, "standard_deviation": 0.08316212281014092, "minimum": 0.14150901888009618, "maximum": 0.3992782297452417, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05182694363938561, "standard_deviation": 0.0077571916284286885, "minimum": 0.043414707777922695, "maximum": 0.06558157258785517, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0008", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.35406756401062, "predicted_mos": 2.35406756401062, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8ed187fcdc9c2ba858a65ae9df78d433e4879aea393fe7919b22b178d12e9d36", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.641571100335568, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1591009667289065, "全帯域_10ms_rms_envelope_nrmse": 0.110503948026214, "全帯域_10ms_envelope_correlation": 0.9890716802888099, "全帯域_10ms_modulation_mae_db": 0.3361364859175379, "全帯域_40ms_log_rms_mae_db": 0.6405969619652334, "全帯域_40ms_rms_envelope_nrmse": 0.06735974523137904, "全帯域_40ms_envelope_correlation": 0.9959071772754231, "全帯域_40ms_modulation_mae_db": 0.30682214512616346, "300–1000 Hz_10ms_log_rms_mae_db": 1.6658605908518667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13527391432241706, "300–1000 Hz_10ms_envelope_correlation": 0.9866912508241673, "300–1000 Hz_10ms_modulation_mae_db": 0.9483734849618434, "300–1000 Hz_40ms_log_rms_mae_db": 1.096371020919104, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08161733715183961, "300–1000 Hz_40ms_envelope_correlation": 0.9950227420111276, "300–1000 Hz_40ms_modulation_mae_db": 0.6964431876702658, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7442536939675528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19147633025221147, "1000–3000 Hz_10ms_envelope_correlation": 0.9726503841502886, "1000–3000 Hz_10ms_modulation_mae_db": 0.16371717873356642, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0311585697396335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10877665320972894, "1000–3000 Hz_40ms_envelope_correlation": 0.9906986397688541, "1000–3000 Hz_40ms_modulation_mae_db": 0.44982007928508433, "3000–6000 Hz_10ms_log_rms_mae_db": 2.14583821128646, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3644081883873057, "3000–6000 Hz_10ms_envelope_correlation": 0.9186942115723077, "3000–6000 Hz_10ms_modulation_mae_db": 0.530198994398586, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7035621591589174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30678782402712956, "3000–6000 Hz_40ms_envelope_correlation": 0.938690649912406, "3000–6000 Hz_40ms_modulation_mae_db": 0.4142093868270038, "mr_stft_db": 5.912122515545879, "band_mae_db": 0.3888607384566268, "rms_gain_error_db": 0.1571138453059574, "output_peak": 0.4765650381790079}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1978604660614673, "全帯域_10ms_rms_envelope_nrmse": 0.10397759493964621, "全帯域_10ms_envelope_correlation": 0.9898542606027301, "全帯域_10ms_modulation_mae_db": 0.3109830027018354, "全帯域_40ms_log_rms_mae_db": 0.6034763983512343, "全帯域_40ms_rms_envelope_nrmse": 0.050116181806346013, "全帯域_40ms_envelope_correlation": 0.9975329915335946, "全帯域_40ms_modulation_mae_db": 0.6568397715366289, "300–1000 Hz_10ms_log_rms_mae_db": 1.6465176975426756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12833991249686458, "300–1000 Hz_10ms_envelope_correlation": 0.9876478175578702, "300–1000 Hz_10ms_modulation_mae_db": 0.7789305498814519, "300–1000 Hz_40ms_log_rms_mae_db": 1.092273353198222, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06707307105985288, "300–1000 Hz_40ms_envelope_correlation": 0.99646666395979, "300–1000 Hz_40ms_modulation_mae_db": 0.8004631851548636, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7800363501818939, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1884259947952845, "1000–3000 Hz_10ms_envelope_correlation": 0.9728565800587068, "1000–3000 Hz_10ms_modulation_mae_db": 0.5301088421493239, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0041945685428637, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10216738437880976, "1000–3000 Hz_40ms_envelope_correlation": 0.9916169386713105, "1000–3000 Hz_40ms_modulation_mae_db": 0.3447605665019114, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1082829482241348, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34868629537631995, "3000–6000 Hz_10ms_envelope_correlation": 0.9228016926873635, "3000–6000 Hz_10ms_modulation_mae_db": 0.401641712332198, "3000–6000 Hz_40ms_log_rms_mae_db": 1.696056082800525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3040165337183578, "3000–6000 Hz_40ms_envelope_correlation": 0.9372623094208715, "3000–6000 Hz_40ms_modulation_mae_db": 0.5080010885436986, "mr_stft_db": 5.919720418728257, "band_mae_db": 0.32794045487402507, "rms_gain_error_db": -0.011684089913641177, "output_peak": 0.4677329431855905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1493471007678482, "全帯域_10ms_rms_envelope_nrmse": 0.09412564325339495, "全帯域_10ms_envelope_correlation": 0.9915731193326779, "全帯域_10ms_modulation_mae_db": 0.25980299059819034, "全帯域_40ms_log_rms_mae_db": 0.5675959728726521, "全帯域_40ms_rms_envelope_nrmse": 0.044793190939112194, "全帯域_40ms_envelope_correlation": 0.9979365785050971, "全帯域_40ms_modulation_mae_db": 0.1742039832435683, "300–1000 Hz_10ms_log_rms_mae_db": 1.6819040580633353, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13555471461219956, "300–1000 Hz_10ms_envelope_correlation": 0.9860982313859492, "300–1000 Hz_10ms_modulation_mae_db": 0.7297349134306769, "300–1000 Hz_40ms_log_rms_mae_db": 1.1595641740894223, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07173702890637307, "300–1000 Hz_40ms_envelope_correlation": 0.9959315374870386, "300–1000 Hz_40ms_modulation_mae_db": 0.3248439038556052, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7521916978631202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19098311185467862, "1000–3000 Hz_10ms_envelope_correlation": 0.9720245509384935, "1000–3000 Hz_10ms_modulation_mae_db": 0.46232496347287616, "1000–3000 Hz_40ms_log_rms_mae_db": 1.059146135145288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10907973231912187, "1000–3000 Hz_40ms_envelope_correlation": 0.9904920991599152, "1000–3000 Hz_40ms_modulation_mae_db": 0.18387347289746467, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1447414886389917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35203264275738205, "3000–6000 Hz_10ms_envelope_correlation": 0.9252471322531388, "3000–6000 Hz_10ms_modulation_mae_db": 0.6255566149768809, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7155526176969262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2985889854106486, "3000–6000 Hz_40ms_envelope_correlation": 0.9428888547670836, "3000–6000 Hz_40ms_modulation_mae_db": 0.6144131650922197, "mr_stft_db": 5.889644559807938, "band_mae_db": 0.3389289166189476, "rms_gain_error_db": -0.019600389474879576, "output_peak": 0.3920159047439441}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1835400086129944, "全帯域_10ms_rms_envelope_nrmse": 0.10928829629002172, "全帯域_10ms_envelope_correlation": 0.988809703150312, "全帯域_10ms_modulation_mae_db": 0.2444631261536353, "全帯域_40ms_log_rms_mae_db": 0.6263244944120185, "全帯域_40ms_rms_envelope_nrmse": 0.05769556080034728, "全帯域_40ms_envelope_correlation": 0.9969743858515205, "全帯域_40ms_modulation_mae_db": 0.363921976949472, "300–1000 Hz_10ms_log_rms_mae_db": 1.7154616189278284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14517778448958985, "300–1000 Hz_10ms_envelope_correlation": 0.9846967284185452, "300–1000 Hz_10ms_modulation_mae_db": 0.9515916765505562, "300–1000 Hz_40ms_log_rms_mae_db": 1.170080120847432, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08170383765364916, "300–1000 Hz_40ms_envelope_correlation": 0.995788498926703, "300–1000 Hz_40ms_modulation_mae_db": 0.8677668698365998, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7325443998340513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19940036529670932, "1000–3000 Hz_10ms_envelope_correlation": 0.969666062066802, "1000–3000 Hz_10ms_modulation_mae_db": 0.5773373812908708, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0762589307893304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12279896438290522, "1000–3000 Hz_40ms_envelope_correlation": 0.988118105474876, "1000–3000 Hz_40ms_modulation_mae_db": 0.3025599147999421, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1306335984336866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3499564501337264, "3000–6000 Hz_10ms_envelope_correlation": 0.9321018809874461, "3000–6000 Hz_10ms_modulation_mae_db": 0.6430562625407974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7189653821454418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3047114458393374, "3000–6000 Hz_40ms_envelope_correlation": 0.9459651557468713, "3000–6000 Hz_40ms_modulation_mae_db": 0.9125758934354301, "mr_stft_db": 5.8905610365195065, "band_mae_db": 0.45582154073123277, "rms_gain_error_db": -0.17900731464821631, "output_peak": 0.395491039186297}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1981629200925625, "全帯域_10ms_rms_envelope_nrmse": 0.10892044045050762, "全帯域_10ms_envelope_correlation": 0.9888342325023702, "全帯域_10ms_modulation_mae_db": 0.09827151255525872, "全帯域_40ms_log_rms_mae_db": 0.6401807641595021, "全帯域_40ms_rms_envelope_nrmse": 0.05633913822195639, "全帯域_40ms_envelope_correlation": 0.9970566115560454, "全帯域_40ms_modulation_mae_db": 0.5401453923664638, "300–1000 Hz_10ms_log_rms_mae_db": 1.7053672678981202, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1431465520456397, "300–1000 Hz_10ms_envelope_correlation": 0.9847756272607626, "300–1000 Hz_10ms_modulation_mae_db": 0.6730592710805591, "300–1000 Hz_40ms_log_rms_mae_db": 1.1466146627263702, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0818844704883755, "300–1000 Hz_40ms_envelope_correlation": 0.9952250090527913, "300–1000 Hz_40ms_modulation_mae_db": 0.7077768882355411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6976000528680306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18052668571725594, "1000–3000 Hz_10ms_envelope_correlation": 0.9753207213845103, "1000–3000 Hz_10ms_modulation_mae_db": 0.37044482316025923, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0422476092833648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10864365611040193, "1000–3000 Hz_40ms_envelope_correlation": 0.9911859364400065, "1000–3000 Hz_40ms_modulation_mae_db": 0.3961053103239727, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2036145134339282, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3675342928813772, "3000–6000 Hz_10ms_envelope_correlation": 0.9205465476550244, "3000–6000 Hz_10ms_modulation_mae_db": 0.4515136120046857, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7252176533633612, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31758884068381926, "3000–6000 Hz_40ms_envelope_correlation": 0.9369626040399501, "3000–6000 Hz_40ms_modulation_mae_db": 0.5130755232155955, "mr_stft_db": 5.925951245996561, "band_mae_db": 0.33596388055013743, "rms_gain_error_db": -0.16078039022083396, "output_peak": 0.3794492176902906}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1490835398481376, "全帯域_10ms_rms_envelope_nrmse": 0.10043875994935778, "全帯域_10ms_envelope_correlation": 0.9904295464336075, "全帯域_10ms_modulation_mae_db": 0.7765264099906922, "全帯域_40ms_log_rms_mae_db": 0.5894291379245129, "全帯域_40ms_rms_envelope_nrmse": 0.04824209862776462, "全帯域_40ms_envelope_correlation": 0.9976563571966172, "全帯域_40ms_modulation_mae_db": 0.3023948303507074, "300–1000 Hz_10ms_log_rms_mae_db": 1.735972485727386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11966890778467683, "300–1000 Hz_10ms_envelope_correlation": 0.9894066086973738, "300–1000 Hz_10ms_modulation_mae_db": 1.273066555401208, "300–1000 Hz_40ms_log_rms_mae_db": 1.1780452180754326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06457717357762023, "300–1000 Hz_40ms_envelope_correlation": 0.9970779406879609, "300–1000 Hz_40ms_modulation_mae_db": 0.5934844961542481, "1000–3000 Hz_10ms_log_rms_mae_db": 1.708254136465815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19913993441585273, "1000–3000 Hz_10ms_envelope_correlation": 0.9695296546235993, "1000–3000 Hz_10ms_modulation_mae_db": 0.4768938974625186, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9822881944619862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10843389121454378, "1000–3000 Hz_40ms_envelope_correlation": 0.9906776984572258, "1000–3000 Hz_40ms_modulation_mae_db": 0.2879578707667428, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0928867903236337, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.357385622358137, "3000–6000 Hz_10ms_envelope_correlation": 0.9224498865827828, "3000–6000 Hz_10ms_modulation_mae_db": 0.1506914572101572, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6602339999115103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30005519629542193, "3000–6000 Hz_40ms_envelope_correlation": 0.9418645803129844, "3000–6000 Hz_40ms_modulation_mae_db": 0.39968750354038535, "mr_stft_db": 5.917152149892444, "band_mae_db": 0.3963088570204185, "rms_gain_error_db": -0.0638218460566471, "output_peak": 0.35804095716614787}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2298715002232965, "全帯域_10ms_rms_envelope_nrmse": 0.09615474253737197, "全帯域_10ms_envelope_correlation": 0.9914336607149749, "全帯域_10ms_modulation_mae_db": 0.22290450104780815, "全帯域_40ms_log_rms_mae_db": 0.5982534703874471, "全帯域_40ms_rms_envelope_nrmse": 0.050935348165581665, "全帯域_40ms_envelope_correlation": 0.9977346743026264, "全帯域_40ms_modulation_mae_db": 0.24961054377618241, "300–1000 Hz_10ms_log_rms_mae_db": 1.633836715428894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.134997301545385, "300–1000 Hz_10ms_envelope_correlation": 0.9869180240557731, "300–1000 Hz_10ms_modulation_mae_db": 1.0555711820606712, "300–1000 Hz_40ms_log_rms_mae_db": 1.0669279940408523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07617522753943455, "300–1000 Hz_40ms_envelope_correlation": 0.9964977624040019, "300–1000 Hz_40ms_modulation_mae_db": 0.4390678865576129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7452675013793826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1835608680162631, "1000–3000 Hz_10ms_envelope_correlation": 0.9741950835711659, "1000–3000 Hz_10ms_modulation_mae_db": 0.08517893421870798, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0458374393795373, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10499465094287244, "1000–3000 Hz_40ms_envelope_correlation": 0.9910988308329414, "1000–3000 Hz_40ms_modulation_mae_db": 0.48582907822391735, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0928582316594455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35384567231505143, "3000–6000 Hz_10ms_envelope_correlation": 0.9192883510322929, "3000–6000 Hz_10ms_modulation_mae_db": 0.790904538830393, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6516035893362584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3050139315606037, "3000–6000 Hz_40ms_envelope_correlation": 0.9358471678675709, "3000–6000 Hz_40ms_modulation_mae_db": 0.5949632846924743, "mr_stft_db": 5.903807655668275, "band_mae_db": 0.3119666082718645, "rms_gain_error_db": -0.1771431362141078, "output_peak": 0.3631051755470165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1783491129622277, "全帯域_10ms_rms_envelope_nrmse": 0.11683955612425703, "全帯域_10ms_envelope_correlation": 0.9874425803931305, "全帯域_10ms_modulation_mae_db": 0.15687219337732197, "全帯域_40ms_log_rms_mae_db": 0.6301224849666519, "全帯域_40ms_rms_envelope_nrmse": 0.06778497444901678, "全帯域_40ms_envelope_correlation": 0.9960779555273791, "全帯域_40ms_modulation_mae_db": 0.29060487952266106, "300–1000 Hz_10ms_log_rms_mae_db": 1.7665522755103835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17440738703720282, "300–1000 Hz_10ms_envelope_correlation": 0.9781455722576805, "300–1000 Hz_10ms_modulation_mae_db": 0.6593217672597186, "300–1000 Hz_40ms_log_rms_mae_db": 1.1864332925369043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11300769155231724, "300–1000 Hz_40ms_envelope_correlation": 0.9918630967148373, "300–1000 Hz_40ms_modulation_mae_db": 0.6682730000644762, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7132011957979403, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20481889086078148, "1000–3000 Hz_10ms_envelope_correlation": 0.9678335357853607, "1000–3000 Hz_10ms_modulation_mae_db": 0.8737268728948102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9821470945867233, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11183171811609154, "1000–3000 Hz_40ms_envelope_correlation": 0.9902990949391232, "1000–3000 Hz_40ms_modulation_mae_db": 0.616778862200254, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1038104674517437, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3684105028758721, "3000–6000 Hz_10ms_envelope_correlation": 0.92154447482519, "3000–6000 Hz_10ms_modulation_mae_db": 0.2599900862334055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.688955073001661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30389937699877123, "3000–6000 Hz_40ms_envelope_correlation": 0.9439351821009748, "3000–6000 Hz_40ms_modulation_mae_db": 0.12718830738390127, "mr_stft_db": 5.916102972501274, "band_mae_db": 0.48108798408262476, "rms_gain_error_db": -0.24757620068765546, "output_peak": 0.39329435017210895}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.156173903263909, "spectral_convergence": 0.3644920746963887, "spectral_convergence_status": "ok", "amplitude_floor": 1.4536470931091146e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.527488789280487, "spectral_convergence": 0.32721396843955614, "spectral_convergence_status": "ok", "amplitude_floor": 9.268471977763907e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.052704854093243, "spectral_convergence": 0.37449067817752046, "spectral_convergence_status": "ok", "amplitude_floor": 5.673364127394706e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031816153358243563, "output_power": 0.000321280184115055, "delta_db": 0.042362686665725444, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009151577935124143, "output_power": 0.0009293790980173345, "delta_db": 0.06696918325261624, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00046151207587702967, "output_power": 0.0005022782855773791, "delta_db": 0.36761333993891043, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011338295265990407, "output_power": 0.00011586714544600564, "delta_db": 0.09412545153229537, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.877003374326731e-05, "output_power": 4.085295054617762e-05, "delta_db": 0.2272724930870878, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3868434770644032e-05, "output_power": 1.9747384201513058e-05, "delta_db": 1.5348212762631255, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3888607384566268, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.1571138453059574, "pitch": null, "output_rms_dbfs": -27.145765895653803, "output_peak": 0.4765650381790079, "reference_rms_dbfs": -27.30287974095976, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.912122515545879}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016375851184643225, "log_rms_mae_db": 1.1591009667289065, "rms_envelope_nrmse": 0.110503948026214, "envelope_correlation": 0.9890716802888099, "rms_gain_error_db": 0.15719678059820763, "reference_mean_power": 0.0018623503698243853, "output_mean_power": 0.0019309946368785337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.48847287729811e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002826666045552883, "output_power": 0.00028518910544006346, "reference_active": true, "delta_db": 0.03858428827735008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.155794557866922e-05, "output_power": 5.6348099258201725e-05, "reference_active": true, "delta_db": -0.38404846404534143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.629343961311588e-06, "output_power": 9.875417168362921e-06, "reference_active": true, "delta_db": 0.5857767054299222}], "active_band_count": 3, "active_band_mae_db": 0.3361364859175379}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014981717790874295, "log_rms_mae_db": 0.6405969619652334, "rms_envelope_nrmse": 0.06735974523137904, "envelope_correlation": 0.9959071772754231, "rms_gain_error_db": 0.15729385501901588, "reference_mean_power": 0.0018713077892111094, "output_mean_power": 0.001940325587363372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.792361294396736e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025909451544035513, "output_power": 0.000262814581574569, "reference_active": true, "delta_db": 0.06191237382369133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9672561350575063e-05, "output_power": 2.6911101611005604e-05, "reference_active": true, "delta_db": -0.42423560401400684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.579324289571111e-07, "output_power": 6.858018360189403e-07, "reference_active": true, "delta_db": -0.43431845754079224}], "active_band_count": 3, "active_band_mae_db": 0.30682214512616346}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014318840727902982, "log_rms_mae_db": 1.6658605908518667, "rms_envelope_nrmse": 0.13527391432241706, "envelope_correlation": 0.9866912508241673, "rms_gain_error_db": 0.13649257656255848, "reference_mean_power": 0.0010573416450716808, "output_mean_power": 0.0010911000999918976}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.91879361710777e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021964908416761717, "output_power": 0.0002210508589052026, "reference_active": true, "delta_db": 0.027628099167891638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.090309412479679e-05, "output_power": 3.478612137995405e-05, "reference_active": true, "delta_db": -0.7035015354695006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.52813465615998e-06, "output_power": 7.3674706650544405e-06, "reference_active": true, "delta_db": 2.113990820248138}], "active_band_count": 3, "active_band_mae_db": 0.9483734849618434}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013022613407710025, "log_rms_mae_db": 1.096371020919104, "rms_envelope_nrmse": 0.08161733715183961, "envelope_correlation": 0.9950227420111276, "rms_gain_error_db": 0.13646093498543632, "reference_mean_power": 0.001062644583935512, "output_mean_power": 0.0010965643600222617}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.561667844043168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020949277098297487, "output_power": 0.00021084579269369587, "reference_active": true, "delta_db": 0.02795898083614759}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.887160269235181e-05, "output_power": 1.5567608703111565e-05, "reference_active": true, "delta_db": -0.8358687775572706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.293974094910667e-07, "output_power": 5.693896919035545e-07, "reference_active": true, "delta_db": 1.2255018046173791}], "active_band_count": 3, "active_band_mae_db": 0.6964431876702658}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.227046231131274e-05, "log_rms_mae_db": 1.7442536939675528, "rms_envelope_nrmse": 0.19147633025221147, "envelope_correlation": 0.9726503841502886, "rms_gain_error_db": 0.08014479325508786, "reference_mean_power": 0.00011346719055744512, "output_mean_power": 0.0001155805565060502}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5656809677774876e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2313251419233995e-05, "output_power": 2.1742858829611465e-05, "reference_active": true, "delta_db": -0.11246212856334428}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.900047334435587e-06, "output_power": 2.854915879414941e-06, "reference_active": true, "delta_db": -0.06811770281679752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0009489505393011e-06, "output_power": 9.31868771602177e-07, "reference_active": true, "delta_db": -0.3105717048205575}], "active_band_count": 3, "active_band_mae_db": 0.16371717873356642}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.589293372832859e-05, "log_rms_mae_db": 1.0311585697396335, "rms_envelope_nrmse": 0.10877665320972894, "envelope_correlation": 0.9906986397688541, "rms_gain_error_db": 0.0800666967594685, "reference_mean_power": 0.0001140327901646041, "output_mean_power": 0.00011615460184763123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.160520396268688e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1030857564945278e-05, "output_power": 2.0678606543916043e-05, "reference_active": true, "delta_db": -0.07335712177717263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3790687257240336e-06, "output_power": 1.4377417682748932e-06, "reference_active": true, "delta_db": 0.18094980015163378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.278096740186873e-08, "output_power": 5.655910028838614e-08, "reference_active": true, "delta_db": -1.0951533159264464}], "active_band_count": 3, "active_band_mae_db": 0.44982007928508433}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9795267482339013e-05, "log_rms_mae_db": 2.14583821128646, "rms_envelope_nrmse": 0.3644081883873057, "envelope_correlation": 0.9186942115723077, "rms_gain_error_db": 0.4067620584172408, "reference_mean_power": 2.1804807525571507e-05, "output_mean_power": 2.3945751060932846e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7142973306691612e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0022096022650052e-05, "output_power": 9.655727758744239e-06, "reference_active": true, "delta_db": -0.16173547024478926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6747496282854573e-06, "output_power": 2.512166564729503e-06, "reference_active": true, "delta_db": -0.27234704694619627}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.037734768320834e-07, "output_power": 6.158603216922465e-07, "reference_active": true, "delta_db": -1.1565144660047724}], "active_band_count": 3, "active_band_mae_db": 0.530198994398586}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.774754919111929e-05, "log_rms_mae_db": 1.7035621591589174, "rms_envelope_nrmse": 0.30678782402712956, "envelope_correlation": 0.938690649912406, "rms_gain_error_db": 0.40672539327202517, "reference_mean_power": 2.1914321405186452e-05, "output_mean_power": 2.4065814578650145e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.567366861761873e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0140844119742473e-05, "output_power": 9.470631727347251e-06, "reference_active": true, "delta_db": -0.29695157926155646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3436726086917478e-06, "output_power": 1.2819111732774587e-06, "reference_active": true, "delta_db": -0.20435531074972535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.915436539179574e-08, "output_power": 4.144097516864562e-08, "reference_active": true, "delta_db": -0.7413212704697296}], "active_band_count": 3, "active_band_mae_db": 0.4142093868270038}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8655436304819218, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "06f336d2dc50347b4e0ce226f921c2b4cff5298fce593b44479544ff57ad90f9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9717595715723659, "standard_deviation": 0.0025481019283976397, "minimum": 0.9678335357853607, "maximum": 0.9753207213845103, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7341686285447233, "standard_deviation": 0.02700440551425594, "minimum": 1.6976000528680306, "maximum": 1.7800363501818939, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.44246661167286666, "standard_deviation": 0.24610658520174272, "minimum": 0.08517893421870798, "maximum": 0.8737268728948102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19229152265112964, "standard_deviation": 0.008335173691024246, "minimum": 0.18052668571725594, "maximum": 0.20481889086078148, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9905234179680316, "standard_deviation": 0.0010594576272057882, "minimum": 0.988118105474876, "maximum": 0.9916169386713105, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0279098177410908, "standard_deviation": 0.03502823945792895, "minimum": 0.9821470945867233, "maximum": 1.0762589307893304, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3834606443749111, "standard_deviation": 0.13441863819256858, "minimum": 0.18387347289746467, "maximum": 0.616778862200254, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10959083133430944, "standard_deviation": 0.006081218753052008, "minimum": 0.10216738437880976, "maximum": 0.12279896438290522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9228342721994434, "standard_deviation": 0.004284026807153348, "minimum": 0.9186942115723077, "maximum": 0.9321018809874461, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.127833281181503, "standard_deviation": 0.03734878911119367, "minimum": 2.0928582316594455, "maximum": 2.2036145134339282, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4816941598158879, "standard_deviation": 0.2106902455056507, "minimum": 0.1506914572101572, "maximum": 0.790904538830393, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3577824583856465, "standard_deviation": 0.007970092610879592, "minimum": 0.34868629537631995, "maximum": 0.3684105028758721, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9404270630210891, "standard_deviation": 0.003724685155754304, "minimum": 0.9358471678675709, "maximum": 0.9459651557468713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6950183196768251, "standard_deviation": 0.02702576487323919, "minimum": 1.6516035893362584, "maximum": 1.7252176533633612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5105142690913386, "standard_deviation": 0.2229955277978626, "minimum": 0.12718830738390127, "maximum": 0.9125758934354301, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.30508276681676116, "standard_deviation": 0.00572506578874955, "minimum": 0.2985889854106486, "maximum": 0.31758884068381926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9855474825572652, "standard_deviation": 0.0033573372758402306, "minimum": 0.9781455722576805, "maximum": 0.9894066086973738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6939340887438112, "standard_deviation": 0.04539923121999434, "minimum": 1.633836715428894, "maximum": 1.7665522755103835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8837061750783357, "standard_deviation": 0.21358670047970288, "minimum": 0.6593217672597186, "maximum": 1.273066555401208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13957080929174692, "standard_deviation": 0.016193205270374652, "minimum": 0.11966890778467683, "maximum": 0.17440738703720282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954841564055313, "standard_deviation": 0.0016132387403197056, "minimum": 0.9918630967148373, "maximum": 0.9970779406879609, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1370387295542175, "standard_deviation": 0.045332894583857655, "minimum": 1.0669279940408523, "maximum": 1.1864332925369043, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6372649271911516, "standard_deviation": 0.1804779407147241, "minimum": 0.3248439038556052, "maximum": 0.8677668698365998, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07972197974118278, "standard_deviation": 0.015051502312680414, "minimum": 0.06457717357762023, "maximum": 0.11300769155231724, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3796098725757347, "standard_deviation": 0.06242219477975551, "minimum": 0.3119666082718645, "maximum": 0.48108798408262476, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.909382819332517, "standard_deviation": 0.013466086831603233, "minimum": 5.889644559807938, "maximum": 5.925951245996561, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4032118282338004, "standard_deviation": 0.044786152489820236, "minimum": 0.35804095716614787, "maximum": 0.4765650381790079, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.087812440238753, "standard_deviation": 0.1299220111077282, "minimum": -0.24757620068765546, "maximum": 0.1571138453059574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9896810979273266, "standard_deviation": 0.0014192675332265782, "minimum": 0.9874425803931305, "maximum": 0.9915731193326779, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.18066445191218, "standard_deviation": 0.027979525417988583, "minimum": 1.1490835398481376, "maximum": 1.2298715002232965, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.300745027792785, "standard_deviation": 0.20713117389907557, "minimum": 0.09827151255525872, "maximum": 0.7765264099906922, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1050311226963464, "standard_deviation": 0.007769542730187775, "minimum": 0.09412564325339495, "maximum": 0.11683955612425703, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997109591468538, "standard_deviation": 0.000763826506244939, "minimum": 0.9959071772754231, "maximum": 0.9979365785050971, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6119974606299066, "standard_deviation": 0.026426008707889876, "minimum": 0.5675959728726521, "maximum": 0.6405969619652334, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3605679403589809, "standard_deviation": 0.15957395431195795, "minimum": 0.1742039832435683, "maximum": 0.6568397715366289, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055408279780188, "standard_deviation": 0.008570430318144456, "minimum": 0.044793190939112194, "maximum": 0.06778497444901678, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0008", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.4255356788635254, "predicted_mos": 2.4255356788635254, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "93a0b9d54173e3e8d45679c3447f63e46bd208b319c3e39b069793f8a9fd7b70", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.111765400040895, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1267817830356814, "全帯域_10ms_rms_envelope_nrmse": 0.10953294584064513, "全帯域_10ms_envelope_correlation": 0.9888973505049089, "全帯域_10ms_modulation_mae_db": 0.2635148272272375, "全帯域_40ms_log_rms_mae_db": 0.5968187586410617, "全帯域_40ms_rms_envelope_nrmse": 0.07448098587179376, "全帯域_40ms_envelope_correlation": 0.9948466614972955, "全帯域_40ms_modulation_mae_db": 0.10073649335358593, "300–1000 Hz_10ms_log_rms_mae_db": 1.85035255276408, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15345018398973995, "300–1000 Hz_10ms_envelope_correlation": 0.9835791043188611, "300–1000 Hz_10ms_modulation_mae_db": 0.7591148945823418, "300–1000 Hz_40ms_log_rms_mae_db": 1.2251272926255676, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11704625784440378, "300–1000 Hz_40ms_envelope_correlation": 0.9909925269203405, "300–1000 Hz_40ms_modulation_mae_db": 0.37033840793858763, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9081443287740312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19475029275688457, "1000–3000 Hz_10ms_envelope_correlation": 0.9708998246634313, "1000–3000 Hz_10ms_modulation_mae_db": 1.2883707143998213, "1000–3000 Hz_40ms_log_rms_mae_db": 1.123659603059192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1126294692674157, "1000–3000 Hz_40ms_envelope_correlation": 0.989753061152516, "1000–3000 Hz_40ms_modulation_mae_db": 1.2504336188923546, "3000–6000 Hz_10ms_log_rms_mae_db": 2.343635273266785, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3504776535736538, "3000–6000 Hz_10ms_envelope_correlation": 0.9253045220295366, "3000–6000 Hz_10ms_modulation_mae_db": 0.33989930230939086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8277212237492741, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2990801709865258, "3000–6000 Hz_40ms_envelope_correlation": 0.9422869724679456, "3000–6000 Hz_40ms_modulation_mae_db": 0.6874465767766154, "mr_stft_db": 6.051681444648096, "band_mae_db": 0.4134413167751776, "rms_gain_error_db": -0.20877726563894836, "output_peak": 0.3883210433659479}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1467200611208272, "全帯域_10ms_rms_envelope_nrmse": 0.10164991254708193, "全帯域_10ms_envelope_correlation": 0.9901516026385686, "全帯域_10ms_modulation_mae_db": 0.5746493715126454, "全帯域_40ms_log_rms_mae_db": 0.583032932711318, "全帯域_40ms_rms_envelope_nrmse": 0.048939724932485024, "全帯域_40ms_envelope_correlation": 0.997555681083583, "全帯域_40ms_modulation_mae_db": 0.3114503297407116, "300–1000 Hz_10ms_log_rms_mae_db": 1.7795223273510874, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12688291282553643, "300–1000 Hz_10ms_envelope_correlation": 0.988208563635072, "300–1000 Hz_10ms_modulation_mae_db": 1.3139021267140605, "300–1000 Hz_40ms_log_rms_mae_db": 1.192612993170445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0648674461596853, "300–1000 Hz_40ms_envelope_correlation": 0.9973404018499907, "300–1000 Hz_40ms_modulation_mae_db": 0.6220925002903487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8561858676405028, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21447131133559694, "1000–3000 Hz_10ms_envelope_correlation": 0.9680231864538595, "1000–3000 Hz_10ms_modulation_mae_db": 0.5177063542425061, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1182964368242208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1289858340358165, "1000–3000 Hz_40ms_envelope_correlation": 0.9885800045856362, "1000–3000 Hz_40ms_modulation_mae_db": 0.29710751449552486, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3523945000998006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3886074034417655, "3000–6000 Hz_10ms_envelope_correlation": 0.9201381273124947, "3000–6000 Hz_10ms_modulation_mae_db": 0.6240074950619788, "3000–6000 Hz_40ms_log_rms_mae_db": 1.850854880469803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32872726868629504, "3000–6000 Hz_40ms_envelope_correlation": 0.9410578227558337, "3000–6000 Hz_40ms_modulation_mae_db": 1.0441004802049505, "mr_stft_db": 6.03611170689516, "band_mae_db": 0.43849332535758134, "rms_gain_error_db": -0.04333699602381121, "output_peak": 0.4934046545571618}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1630471832393208, "全帯域_10ms_rms_envelope_nrmse": 0.12179943560614993, "全帯域_10ms_envelope_correlation": 0.985898388004634, "全帯域_10ms_modulation_mae_db": 0.41719335423606535, "全帯域_40ms_log_rms_mae_db": 0.6211289393340005, "全帯域_40ms_rms_envelope_nrmse": 0.054922807236795865, "全帯域_40ms_envelope_correlation": 0.9968922821118116, "全帯域_40ms_modulation_mae_db": 0.13784276847309204, "300–1000 Hz_10ms_log_rms_mae_db": 1.8718845661480132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15198293098742258, "300–1000 Hz_10ms_envelope_correlation": 0.9825122207439693, "300–1000 Hz_10ms_modulation_mae_db": 1.1135844563078907, "300–1000 Hz_40ms_log_rms_mae_db": 1.2141613727208478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07144273402889519, "300–1000 Hz_40ms_envelope_correlation": 0.9959674957696493, "300–1000 Hz_40ms_modulation_mae_db": 0.22142845655849375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9037628246859157, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20763633477512683, "1000–3000 Hz_10ms_envelope_correlation": 0.9713842751622743, "1000–3000 Hz_10ms_modulation_mae_db": 0.9336054887537165, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1416426159174564, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13842462895897184, "1000–3000 Hz_40ms_envelope_correlation": 0.9878097170231686, "1000–3000 Hz_40ms_modulation_mae_db": 1.2416569195265765, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3766554929101837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36266225958943027, "3000–6000 Hz_10ms_envelope_correlation": 0.9180927452946773, "3000–6000 Hz_10ms_modulation_mae_db": 0.5196542748029932, "3000–6000 Hz_40ms_log_rms_mae_db": 1.818758930756061, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30547570958710485, "3000–6000 Hz_40ms_envelope_correlation": 0.9381719132274327, "3000–6000 Hz_40ms_modulation_mae_db": 0.3555994323041786, "mr_stft_db": 6.050493705155926, "band_mae_db": 0.48757582853463816, "rms_gain_error_db": -0.02051855172395421, "output_peak": 0.507420688257283}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.162277518224212, "全帯域_10ms_rms_envelope_nrmse": 0.10788426865828554, "全帯域_10ms_envelope_correlation": 0.9889658920940751, "全帯域_10ms_modulation_mae_db": 0.13646995723776922, "全帯域_40ms_log_rms_mae_db": 0.5861640450643189, "全帯域_40ms_rms_envelope_nrmse": 0.043258769067862614, "全帯域_40ms_envelope_correlation": 0.9981036716777303, "全帯域_40ms_modulation_mae_db": 0.37386276815799063, "300–1000 Hz_10ms_log_rms_mae_db": 1.8218924694398506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15127718461065304, "300–1000 Hz_10ms_envelope_correlation": 0.9826575838105206, "300–1000 Hz_10ms_modulation_mae_db": 0.6254492414508585, "300–1000 Hz_40ms_log_rms_mae_db": 1.1945821644376167, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06477123506996918, "300–1000 Hz_40ms_envelope_correlation": 0.9967488771431416, "300–1000 Hz_40ms_modulation_mae_db": 0.20958018647083024, "1000–3000 Hz_10ms_log_rms_mae_db": 1.902867017701191, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20210527273969445, "1000–3000 Hz_10ms_envelope_correlation": 0.9686040027123386, "1000–3000 Hz_10ms_modulation_mae_db": 0.501655093495634, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1501610201001202, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12980775714419412, "1000–3000 Hz_40ms_envelope_correlation": 0.9865427972305283, "1000–3000 Hz_40ms_modulation_mae_db": 0.5046917243779702, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4041465811580722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36033893509189785, "3000–6000 Hz_10ms_envelope_correlation": 0.9183648367839629, "3000–6000 Hz_10ms_modulation_mae_db": 0.2933128220450941, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8505560856267316, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3058067867060718, "3000–6000 Hz_40ms_envelope_correlation": 0.937342008680781, "3000–6000 Hz_40ms_modulation_mae_db": 0.20195107328453474, "mr_stft_db": 6.046545477257669, "band_mae_db": 0.3733584797636437, "rms_gain_error_db": -0.03403929875014889, "output_peak": 0.43774918770677707}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1508714255833794, "全帯域_10ms_rms_envelope_nrmse": 0.12184626455804656, "全帯域_10ms_envelope_correlation": 0.9858571050454612, "全帯域_10ms_modulation_mae_db": 0.4378470685267861, "全帯域_40ms_log_rms_mae_db": 0.5807666764660222, "全帯域_40ms_rms_envelope_nrmse": 0.06798962145749354, "全帯域_40ms_envelope_correlation": 0.9953478002679578, "全帯域_40ms_modulation_mae_db": 0.33843962842369546, "300–1000 Hz_10ms_log_rms_mae_db": 1.8520362111290931, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17557993981768694, "300–1000 Hz_10ms_envelope_correlation": 0.9766449684249633, "300–1000 Hz_10ms_modulation_mae_db": 1.086860395650933, "300–1000 Hz_40ms_log_rms_mae_db": 1.2901540001330192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11728526889914172, "300–1000 Hz_40ms_envelope_correlation": 0.9892619876471019, "300–1000 Hz_40ms_modulation_mae_db": 0.6664265522757122, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8645793885803832, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18851397591689953, "1000–3000 Hz_10ms_envelope_correlation": 0.9738206324264994, "1000–3000 Hz_10ms_modulation_mae_db": 0.3448006304368168, "1000–3000 Hz_40ms_log_rms_mae_db": 1.079288463308591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1259800729769448, "1000–3000 Hz_40ms_envelope_correlation": 0.9890614305444219, "1000–3000 Hz_40ms_modulation_mae_db": 0.19686451213474906, "3000–6000 Hz_10ms_log_rms_mae_db": 2.381576611721327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36119237450360314, "3000–6000 Hz_10ms_envelope_correlation": 0.9148891211766708, "3000–6000 Hz_10ms_modulation_mae_db": 0.5477650890148039, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8388067899497385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.311369241901102, "3000–6000 Hz_40ms_envelope_correlation": 0.9325236978378844, "3000–6000 Hz_40ms_modulation_mae_db": 0.5735332261565125, "mr_stft_db": 6.055718530180994, "band_mae_db": 0.4539056737218363, "rms_gain_error_db": -0.113260149923013, "output_peak": 0.39701266490072673}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1736782710478615, "全帯域_10ms_rms_envelope_nrmse": 0.09566530555137698, "全帯域_10ms_envelope_correlation": 0.9913330058779506, "全帯域_10ms_modulation_mae_db": 0.5982668235056233, "全帯域_40ms_log_rms_mae_db": 0.5795646981373839, "全帯域_40ms_rms_envelope_nrmse": 0.04996814346979643, "全帯域_40ms_envelope_correlation": 0.9974486644445654, "全帯域_40ms_modulation_mae_db": 0.32798794644463397, "300–1000 Hz_10ms_log_rms_mae_db": 1.8798140171502518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13386489494816367, "300–1000 Hz_10ms_envelope_correlation": 0.9864501809874648, "300–1000 Hz_10ms_modulation_mae_db": 1.2716500890772855, "300–1000 Hz_40ms_log_rms_mae_db": 1.2911591504676516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06897373239395733, "300–1000 Hz_40ms_envelope_correlation": 0.9962526751893483, "300–1000 Hz_40ms_modulation_mae_db": 0.8172465293283584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9009922497048453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21344069479541108, "1000–3000 Hz_10ms_envelope_correlation": 0.9656965250517334, "1000–3000 Hz_10ms_modulation_mae_db": 0.6658567454664831, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1034093709498956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1315255802703595, "1000–3000 Hz_40ms_envelope_correlation": 0.9861099667702994, "1000–3000 Hz_40ms_modulation_mae_db": 0.44919460710405507, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3708271285541547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.377141950119232, "3000–6000 Hz_10ms_envelope_correlation": 0.9147331308484601, "3000–6000 Hz_10ms_modulation_mae_db": 0.15320038808266626, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8374881776890823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3113270646771873, "3000–6000 Hz_40ms_envelope_correlation": 0.9386017465549082, "3000–6000 Hz_40ms_modulation_mae_db": 0.3178292025827995, "mr_stft_db": 6.056390224279448, "band_mae_db": 0.3396459289437123, "rms_gain_error_db": -0.004073589452154109, "output_peak": 0.477438757702633}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.144374537595593, "全帯域_10ms_rms_envelope_nrmse": 0.11662829008888809, "全帯域_10ms_envelope_correlation": 0.9870520368171153, "全帯域_10ms_modulation_mae_db": 0.3499055987151804, "全帯域_40ms_log_rms_mae_db": 0.5875214735400297, "全帯域_40ms_rms_envelope_nrmse": 0.05201550733731709, "全帯域_40ms_envelope_correlation": 0.997233130617094, "全帯域_40ms_modulation_mae_db": 0.19347088017001055, "300–1000 Hz_10ms_log_rms_mae_db": 1.9158401931992055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15612163974853185, "300–1000 Hz_10ms_envelope_correlation": 0.9815091071311434, "300–1000 Hz_10ms_modulation_mae_db": 1.1957411540325176, "300–1000 Hz_40ms_log_rms_mae_db": 1.2535416448093863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07261647797076634, "300–1000 Hz_40ms_envelope_correlation": 0.9958398337287306, "300–1000 Hz_40ms_modulation_mae_db": 0.38879832462832997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8914890765257946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1925153992145311, "1000–3000 Hz_10ms_envelope_correlation": 0.9728398675338779, "1000–3000 Hz_10ms_modulation_mae_db": 0.42019341513132275, "1000–3000 Hz_40ms_log_rms_mae_db": 1.148435342570528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1254262393297328, "1000–3000 Hz_40ms_envelope_correlation": 0.9895533614876283, "1000–3000 Hz_40ms_modulation_mae_db": 0.31375745288864565, "3000–6000 Hz_10ms_log_rms_mae_db": 2.380877727909516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36646213355033974, "3000–6000 Hz_10ms_envelope_correlation": 0.9116685884857695, "3000–6000 Hz_10ms_modulation_mae_db": 0.9769809884000145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.871058770781636, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3130978608275913, "3000–6000 Hz_40ms_envelope_correlation": 0.931237567709273, "3000–6000 Hz_40ms_modulation_mae_db": 0.9246699837297649, "mr_stft_db": 6.0524083723529385, "band_mae_db": 0.38816678376508085, "rms_gain_error_db": -0.0500872507751672, "output_peak": 0.5045927522985976}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1348156322618688, "全帯域_10ms_rms_envelope_nrmse": 0.12447559189821664, "全帯域_10ms_envelope_correlation": 0.98522887898878, "全帯域_10ms_modulation_mae_db": 0.7047269433985077, "全帯域_40ms_log_rms_mae_db": 0.5535029860467412, "全帯域_40ms_rms_envelope_nrmse": 0.0512430416923165, "全帯域_40ms_envelope_correlation": 0.9973909864686213, "全帯域_40ms_modulation_mae_db": 0.5178723108319213, "300–1000 Hz_10ms_log_rms_mae_db": 1.7686925816290613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1511541422705571, "300–1000 Hz_10ms_envelope_correlation": 0.9827469487886592, "300–1000 Hz_10ms_modulation_mae_db": 1.2536668630196008, "300–1000 Hz_40ms_log_rms_mae_db": 1.1574729914580508, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07154349690728834, "300–1000 Hz_40ms_envelope_correlation": 0.9961896192195079, "300–1000 Hz_40ms_modulation_mae_db": 0.7112874181217862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8970039523881703, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19162793176413764, "1000–3000 Hz_10ms_envelope_correlation": 0.971916031979666, "1000–3000 Hz_10ms_modulation_mae_db": 0.4516480431049683, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1126173856952517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10544862095478548, "1000–3000 Hz_40ms_envelope_correlation": 0.990974857616661, "1000–3000 Hz_40ms_modulation_mae_db": 0.48338959360184797, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3358636549964817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3689693869794633, "3000–6000 Hz_10ms_envelope_correlation": 0.9230268297727809, "3000–6000 Hz_10ms_modulation_mae_db": 0.4235649618206838, "3000–6000 Hz_40ms_log_rms_mae_db": 1.833934420153068, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31603136636251794, "3000–6000 Hz_40ms_envelope_correlation": 0.9407130730238861, "3000–6000 Hz_40ms_modulation_mae_db": 0.668424589687232, "mr_stft_db": 6.04373864549092, "band_mae_db": 0.40736223079282197, "rms_gain_error_db": -0.0958837630753201, "output_peak": 0.4453233546960263}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.393300978191069, "spectral_convergence": 0.3549098768296807, "spectral_convergence_status": "ok", "amplitude_floor": 1.4536470931091146e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.666793491143508, "spectral_convergence": 0.3293487986418764, "spectral_convergence_status": "ok", "amplitude_floor": 9.268471977763907e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.094949864609713, "spectral_convergence": 0.38042465147280186, "spectral_convergence_status": "ok", "amplitude_floor": 5.673364127394706e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031816153358243563, "output_power": 0.00032587002735268167, "delta_db": 0.10396745723255543, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009151577935124143, "output_power": 0.000842232073479996, "delta_db": -0.36064206534797394, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00046151207587702967, "output_power": 0.000435523327878888, "delta_db": -0.25171647189829927, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011338295265990407, "output_power": 0.00011028422739288325, "delta_db": -0.12034357431544501, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.877003374326731e-05, "output_power": 3.9952211995856145e-05, "delta_db": 0.13044650489315612, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3868434770644032e-05, "output_power": 1.9650817887506544e-05, "delta_db": 1.5135318269636358, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4134413167751776, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.20877726563894836, "pitch": null, "output_rms_dbfs": -27.51165700659871, "output_peak": 0.3883210433659479, "reference_rms_dbfs": -27.30287974095976, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.051681444648096}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016375851184643225, "log_rms_mae_db": 1.1267817830356814, "rms_envelope_nrmse": 0.10953294584064513, "envelope_correlation": 0.9888973505049089, "rms_gain_error_db": -0.20876852419846925, "reference_mean_power": 0.0018623503698243853, "output_mean_power": 0.0017749435161334075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.48847287729811e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002826666045552883, "output_power": 0.00027850534693776445, "reference_active": true, "delta_db": -0.06440964633100951}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.155794557866922e-05, "output_power": 5.933714660098909e-05, "reference_active": true, "delta_db": -0.15957459181234046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.629343961311588e-06, "output_power": 9.831817408275825e-06, "reference_active": true, "delta_db": 0.5665602435383625}], "active_band_count": 3, "active_band_mae_db": 0.2635148272272375}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014981717790874295, "log_rms_mae_db": 0.5968187586410617, "rms_envelope_nrmse": 0.07448098587179376, "envelope_correlation": 0.9948466614972955, "rms_gain_error_db": -0.20880157779660544, "reference_mean_power": 0.0018713077892111094, "output_mean_power": 0.0017834669575677556}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.792361294396736e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025909451544035513, "output_power": 0.0002557576212475341, "reference_active": true, "delta_db": -0.056296358659155755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9672561350575063e-05, "output_power": 2.8645102113652126e-05, "reference_active": true, "delta_db": -0.15304661700154942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.579324289571111e-07, "output_power": 7.743140500232665e-07, "reference_active": true, "delta_db": 0.09286650440005263}], "active_band_count": 3, "active_band_mae_db": 0.10073649335358593}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014318840727902982, "log_rms_mae_db": 1.85035255276408, "rms_envelope_nrmse": 0.15345018398973995, "envelope_correlation": 0.9835791043188611, "rms_gain_error_db": -0.4700844403444357, "reference_mean_power": 0.0010573416450716808, "output_mean_power": 0.0009488703887434735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.91879361710777e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021964908416761717, "output_power": 0.0002072937269982486, "reference_active": true, "delta_db": -0.2514323687671163}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.090309412479679e-05, "output_power": 3.415108654925223e-05, "reference_active": true, "delta_db": -0.7835163583603139}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.52813465615998e-06, "output_power": 6.027801315430477e-06, "reference_active": true, "delta_db": 1.242395956619595}], "active_band_count": 3, "active_band_mae_db": 0.7591148945823418}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013022613407710025, "log_rms_mae_db": 1.2251272926255676, "rms_envelope_nrmse": 0.11704625784440378, "envelope_correlation": 0.9909925269203405, "rms_gain_error_db": -0.47010047406461697, "reference_mean_power": 0.001062644583935512, "output_mean_power": 0.0009536257855482864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.561667844043168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020949277098297487, "output_power": 0.00019689810889003383, "reference_active": true, "delta_db": -0.2692749628537197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.887160269235181e-05, "output_power": 1.5876798851038252e-05, "reference_active": true, "delta_db": -0.7504584210202319}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.293974094910667e-07, "output_power": 4.204663414848577e-07, "reference_active": true, "delta_db": -0.0912818399418113}], "active_band_count": 3, "active_band_mae_db": 0.37033840793858763}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.227046231131274e-05, "log_rms_mae_db": 1.9081443287740312, "rms_envelope_nrmse": 0.19475029275688457, "envelope_correlation": 0.9708998246634313, "rms_gain_error_db": -0.10653106382685035, "reference_mean_power": 0.00011346719055744512, "output_mean_power": 0.00011071773572541473}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5656809677774876e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2313251419233995e-05, "output_power": 2.316623132351765e-05, "reference_active": true, "delta_db": 0.1629252965473713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.900047334435587e-06, "output_power": 3.6720703645107296e-06, "reference_active": true, "delta_db": 1.0250590810427107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0009489505393011e-06, "output_power": 1.8540639306189422e-06, "reference_active": true, "delta_db": 2.677127765609382}], "active_band_count": 3, "active_band_mae_db": 1.2883707143998213}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.589293372832859e-05, "log_rms_mae_db": 1.123659603059192, "rms_envelope_nrmse": 0.1126294692674157, "envelope_correlation": 0.989753061152516, "rms_gain_error_db": -0.10672934464761868, "reference_mean_power": 0.0001140327901646041, "output_mean_power": 0.00011126455013912913}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.160520396268688e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1030857564945278e-05, "output_power": 2.2676493300326688e-05, "reference_active": true, "delta_db": 0.32718913925085336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3790687257240336e-06, "output_power": 1.8564678519278817e-06, "reference_active": true, "delta_db": 1.2910152331062443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.278096740186873e-08, "output_power": 1.1893987352253459e-07, "reference_active": true, "delta_db": 2.1330964843199665}], "active_band_count": 3, "active_band_mae_db": 1.2504336188923546}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9795267482339013e-05, "log_rms_mae_db": 2.343635273266785, "rms_envelope_nrmse": 0.3504776535736538, "envelope_correlation": 0.9253045220295366, "rms_gain_error_db": 0.43810437745474695, "reference_mean_power": 2.1804807525571507e-05, "output_mean_power": 2.4119188693773152e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7142973306691612e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0022096022650052e-05, "output_power": 9.569943219432434e-06, "reference_active": true, "delta_db": -0.20049198422627118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6747496282854573e-06, "output_power": 2.8650910360121875e-06, "reference_active": true, "delta_db": 0.29855290051419314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.037734768320834e-07, "output_power": 9.061476600934738e-07, "reference_active": true, "delta_db": 0.5206530221877084}], "active_band_count": 3, "active_band_mae_db": 0.33989930230939086}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.774754919111929e-05, "log_rms_mae_db": 1.8277212237492741, "rms_envelope_nrmse": 0.2990801709865258, "envelope_correlation": 0.9422869724679456, "rms_gain_error_db": 0.43806374534587195, "reference_mean_power": 2.1914321405186452e-05, "output_mean_power": 2.424009968274364e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.567366861761873e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0140844119742473e-05, "output_power": 9.603083781659844e-06, "reference_active": true, "delta_db": -0.23663389135694993}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3436726086917478e-06, "output_power": 1.500746962497604e-06, "reference_active": true, "delta_db": 0.4801400903315433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.915436539179574e-08, "output_power": 6.700676763939538e-08, "reference_active": true, "delta_db": 1.3455657486413526}], "active_band_count": 3, "active_band_mae_db": 0.6874465767766154}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8605243696289565, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cf3a66dee9ee9abfb952c56e5a439998ac5e5e2d6bd67cb37533f7610261e763", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9703980432479601, "standard_deviation": 0.0027299437430376187, "minimum": 0.9656965250517334, "maximum": 0.9738206324264994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8906280882501043, "standard_deviation": 0.019430246963435458, "minimum": 1.8561858676405028, "maximum": 1.9081443287740312, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6404795606289087, "standard_deviation": 0.3189905079047401, "minimum": 0.3448006304368168, "maximum": 1.2883707143998213, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20063265166228528, "standard_deviation": 0.010248441528174628, "minimum": 0.18851397591689953, "maximum": 0.21447131133559694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9885481495513575, "standard_deviation": 0.0016530103690746392, "minimum": 0.9861099667702994, "maximum": 0.990974857616661, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1221887798031571, "standard_deviation": 0.024353963057350388, "minimum": 1.079288463308591, "maximum": 1.1501610201001202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5921369928777155, "standard_deviation": 0.41677566350852785, "minimum": 0.19686451213474906, "maximum": 1.2504336188923546, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12477852536727758, "standard_deviation": 0.010676018053481845, "minimum": 0.10544862095478548, "maximum": 0.13842462895897184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9182772377130441, "standard_deviation": 0.004522329401197325, "minimum": 0.9116685884857695, "maximum": 0.9253045220295366, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.36824712132704, "standard_deviation": 0.02270642910606896, "minimum": 2.3358636549964817, "maximum": 2.4041465811580722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4847981651922032, "standard_deviation": 0.25019188605677, "minimum": 0.15320038808266626, "maximum": 0.9769809884000145, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3669815121061732, "standard_deviation": 0.011611609397550902, "minimum": 0.3504776535736538, "maximum": 0.3886074034417655, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.937741850282243, "standard_deviation": 0.003983914612421692, "minimum": 0.931237567709273, "maximum": 0.9422869724679456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8411474098969245, "standard_deviation": 0.016168089602373745, "minimum": 1.818758930756061, "maximum": 1.871058770781636, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5966943205908235, "standard_deviation": 0.29601464190823706, "minimum": 0.20195107328453474, "maximum": 1.0441004802049505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3113644337167995, "standard_deviation": 0.00880521039423887, "minimum": 0.2990801709865258, "maximum": 0.32872726868629504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9830385847300818, "standard_deviation": 0.003430633756089226, "minimum": 0.9766449684249633, "maximum": 0.988208563635072, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8425043648513304, "standard_deviation": 0.050179060618567056, "minimum": 1.7686925816290613, "maximum": 1.9158401931992055, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.077496152604436, "standard_deviation": 0.25233329016226114, "minimum": 0.6254492414508585, "maximum": 1.3139021267140605, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15003922864978644, "standard_deviation": 0.014664514250546816, "minimum": 0.12688291282553643, "maximum": 0.17557993981768694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9948241771834764, "standard_deviation": 0.002973673216713678, "minimum": 0.9892619876471019, "maximum": 0.9973404018499907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2273514512278232, "standard_deviation": 0.047874866189080545, "minimum": 1.1574729914580508, "maximum": 1.2911591504676516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5008997969515558, "standard_deviation": 0.2326890464140697, "minimum": 0.20958018647083024, "maximum": 0.8172465293283584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0810683311592634, "standard_deviation": 0.022472934619182008, "minimum": 0.06477123506996918, "maximum": 0.11728526889914172, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.41274369595681154, "standard_deviation": 0.04703365078906226, "minimum": 0.3396459289437123, "maximum": 0.48757582853463816, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.049136013282643, "standard_deviation": 0.006766081380132648, "minimum": 6.03611170689516, "maximum": 6.056390224279448, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45640788793564413, "standard_deviation": 0.04685809896169983, "minimum": 0.3883210433659479, "maximum": 0.507420688257283, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07124710817031463, "standard_deviation": 0.0665297009060517, "minimum": -0.20877726563894836, "maximum": -0.004073589452154109, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9879230324964368, "standard_deviation": 0.002236567306328146, "minimum": 0.98522887898878, "maximum": 0.9913330058779506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.150320801513593, "standard_deviation": 0.01555459273758132, "minimum": 1.1267817830356814, "maximum": 1.1736782710478615, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.43532174304497695, "standard_deviation": 0.18708416323867372, "minimum": 0.13646995723776922, "maximum": 0.7047269433985077, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11243525184358635, "standard_deviation": 0.010457368739522504, "minimum": 0.09566530555137698, "maximum": 0.12447559189821664, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968523597710823, "standard_deviation": 0.001142631388733393, "minimum": 0.9948466614972955, "maximum": 0.9981036716777303, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5860625637426096, "standard_deviation": 0.01884924623523103, "minimum": 0.5535029860467412, "maximum": 0.6211289393340005, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2877078906994552, "standard_deviation": 0.13703841544759293, "minimum": 0.10073649335358593, "maximum": 0.5178723108319213, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0553523251332326, "standard_deviation": 0.010489293322841105, "minimum": 0.043258769067862614, "maximum": 0.07448098587179376, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0008", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.3969738483428955, "predicted_mos": 2.3969738483428955, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "812e14a48304a2ad1522c3ed1ece78a60771f53f2756a4073e49e7fc6667b0ee", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.363090999890119, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1253569671769286, "全帯域_10ms_rms_envelope_nrmse": 0.0938181293585995, "全帯域_10ms_envelope_correlation": 0.9921105930587749, "全帯域_10ms_modulation_mae_db": 0.5735538516750555, "全帯域_40ms_log_rms_mae_db": 0.6160108993148273, "全帯域_40ms_rms_envelope_nrmse": 0.058482655565886485, "全帯域_40ms_envelope_correlation": 0.9971710505240822, "全帯域_40ms_modulation_mae_db": 0.46205491820330863, "300–1000 Hz_10ms_log_rms_mae_db": 1.7852689293460564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13021521927090438, "300–1000 Hz_10ms_envelope_correlation": 0.9885779536808291, "300–1000 Hz_10ms_modulation_mae_db": 1.0921798507625746, "300–1000 Hz_40ms_log_rms_mae_db": 1.2233331320834673, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09157507788633745, "300–1000 Hz_40ms_envelope_correlation": 0.9951213271449604, "300–1000 Hz_40ms_modulation_mae_db": 0.7695263961978794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8771824010627896, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1887362560414663, "1000–3000 Hz_10ms_envelope_correlation": 0.9726805144276206, "1000–3000 Hz_10ms_modulation_mae_db": 0.49731625189848727, "1000–3000 Hz_40ms_log_rms_mae_db": 1.034377358009375, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09972072159625212, "1000–3000 Hz_40ms_envelope_correlation": 0.9921039948552636, "1000–3000 Hz_40ms_modulation_mae_db": 0.49866500570601197, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3025169877645086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35556438935838186, "3000–6000 Hz_10ms_envelope_correlation": 0.9264993290437357, "3000–6000 Hz_10ms_modulation_mae_db": 0.2820999801430088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.799918929738733, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30132392625881793, "3000–6000 Hz_40ms_envelope_correlation": 0.9445016048059024, "3000–6000 Hz_40ms_modulation_mae_db": 0.5093486925283397, "mr_stft_db": 6.0168866087632935, "band_mae_db": 0.4361004762064567, "rms_gain_error_db": -0.22670826864223823, "output_peak": 0.37308530839627946}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1367980220367873, "全帯域_10ms_rms_envelope_nrmse": 0.10939700839905522, "全帯域_10ms_envelope_correlation": 0.9886965920146671, "全帯域_10ms_modulation_mae_db": 0.45083992330917066, "全帯域_40ms_log_rms_mae_db": 0.5722762540518102, "全帯域_40ms_rms_envelope_nrmse": 0.057831355629728645, "全帯域_40ms_envelope_correlation": 0.9965781453305991, "全帯域_40ms_modulation_mae_db": 0.2779136780288033, "300–1000 Hz_10ms_log_rms_mae_db": 1.7463565442553806, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13789797275676788, "300–1000 Hz_10ms_envelope_correlation": 0.9856414563639663, "300–1000 Hz_10ms_modulation_mae_db": 1.1165674878713419, "300–1000 Hz_40ms_log_rms_mae_db": 1.1342128301812966, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07199959208625754, "300–1000 Hz_40ms_envelope_correlation": 0.9958575163086698, "300–1000 Hz_40ms_modulation_mae_db": 0.8134170985585488, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8323340478327104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18520384509072157, "1000–3000 Hz_10ms_envelope_correlation": 0.9742216388999831, "1000–3000 Hz_10ms_modulation_mae_db": 0.3271100499736696, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1018110577557316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10493814134975923, "1000–3000 Hz_40ms_envelope_correlation": 0.9912198534717827, "1000–3000 Hz_40ms_modulation_mae_db": 0.4218281582685321, "3000–6000 Hz_10ms_log_rms_mae_db": 2.312560521212417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3655089958858881, "3000–6000 Hz_10ms_envelope_correlation": 0.9225653807963269, "3000–6000 Hz_10ms_modulation_mae_db": 0.3509413590179829, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7776296142436956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3056726416369398, "3000–6000 Hz_40ms_envelope_correlation": 0.9431562150073856, "3000–6000 Hz_40ms_modulation_mae_db": 0.5404568831076407, "mr_stft_db": 6.002609854028399, "band_mae_db": 0.38627514079466757, "rms_gain_error_db": 0.024366253686237405, "output_peak": 0.4368329107792162}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1105399139399696, "全帯域_10ms_rms_envelope_nrmse": 0.09874707860503334, "全帯域_10ms_envelope_correlation": 0.9907870407402085, "全帯域_10ms_modulation_mae_db": 0.578900992352054, "全帯域_40ms_log_rms_mae_db": 0.5921335100356022, "全帯域_40ms_rms_envelope_nrmse": 0.04934135432338302, "全帯域_40ms_envelope_correlation": 0.9975170702238667, "全帯域_40ms_modulation_mae_db": 0.3406090612252044, "300–1000 Hz_10ms_log_rms_mae_db": 1.8143344814814537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1391148057501598, "300–1000 Hz_10ms_envelope_correlation": 0.985340973257489, "300–1000 Hz_10ms_modulation_mae_db": 1.2778521255267448, "300–1000 Hz_40ms_log_rms_mae_db": 1.1861249613640228, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06371795201941186, "300–1000 Hz_40ms_envelope_correlation": 0.9968043484997586, "300–1000 Hz_40ms_modulation_mae_db": 0.7579688219878052, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8595735242923004, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18229185562839337, "1000–3000 Hz_10ms_envelope_correlation": 0.9745269053444229, "1000–3000 Hz_10ms_modulation_mae_db": 0.4422121245235074, "1000–3000 Hz_40ms_log_rms_mae_db": 1.072005374865909, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09766465480675901, "1000–3000 Hz_40ms_envelope_correlation": 0.9923219506804427, "1000–3000 Hz_40ms_modulation_mae_db": 0.27319876838021867, "3000–6000 Hz_10ms_log_rms_mae_db": 2.330730274472325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3679585360460388, "3000–6000 Hz_10ms_envelope_correlation": 0.9117891400747268, "3000–6000 Hz_10ms_modulation_mae_db": 0.391324800340665, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8365775542087845, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3140340841004521, "3000–6000 Hz_40ms_envelope_correlation": 0.9314908772891101, "3000–6000 Hz_40ms_modulation_mae_db": 0.43318545013513554, "mr_stft_db": 6.031105983801834, "band_mae_db": 0.306648672916114, "rms_gain_error_db": 0.023012448483675847, "output_peak": 0.3975735987428376}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1235362870806245, "全帯域_10ms_rms_envelope_nrmse": 0.12006304362029552, "全帯域_10ms_envelope_correlation": 0.9863143585669014, "全帯域_10ms_modulation_mae_db": 0.4739902146587036, "全帯域_40ms_log_rms_mae_db": 0.5641057768978284, "全帯域_40ms_rms_envelope_nrmse": 0.0524283881300745, "全帯域_40ms_envelope_correlation": 0.9973402919628187, "全帯域_40ms_modulation_mae_db": 0.3976697988789848, "300–1000 Hz_10ms_log_rms_mae_db": 1.7920540314112392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15702793364829176, "300–1000 Hz_10ms_envelope_correlation": 0.9815293257324447, "300–1000 Hz_10ms_modulation_mae_db": 1.2766034963435817, "300–1000 Hz_40ms_log_rms_mae_db": 1.1938816783734176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07930518862539543, "300–1000 Hz_40ms_envelope_correlation": 0.9955528396860038, "300–1000 Hz_40ms_modulation_mae_db": 0.9752916032445285, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8800361818721851, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22354236918892667, "1000–3000 Hz_10ms_envelope_correlation": 0.9628298166833458, "1000–3000 Hz_10ms_modulation_mae_db": 0.8203999491199316, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1405148009930821, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13933235017699533, "1000–3000 Hz_40ms_envelope_correlation": 0.9846997315454901, "1000–3000 Hz_40ms_modulation_mae_db": 0.9253488140469367, "3000–6000 Hz_10ms_log_rms_mae_db": 2.28784805486986, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35184948014473894, "3000–6000 Hz_10ms_envelope_correlation": 0.9200275522731208, "3000–6000 Hz_10ms_modulation_mae_db": 0.755764253423704, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7621019913365317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30247008637331796, "3000–6000 Hz_40ms_envelope_correlation": 0.93693701884033, "3000–6000 Hz_40ms_modulation_mae_db": 0.7883175634006054, "mr_stft_db": 6.0247190104371535, "band_mae_db": 0.29109376317638486, "rms_gain_error_db": -0.12468301818925204, "output_peak": 0.4609937316085491}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1719149339454187, "全帯域_10ms_rms_envelope_nrmse": 0.1106286011961294, "全帯域_10ms_envelope_correlation": 0.9885801653000155, "全帯域_10ms_modulation_mae_db": 0.18423326883423474, "全帯域_40ms_log_rms_mae_db": 0.606205039749985, "全帯域_40ms_rms_envelope_nrmse": 0.057801742227100696, "全帯域_40ms_envelope_correlation": 0.9966839217571877, "全帯域_40ms_modulation_mae_db": 0.23565938150964316, "300–1000 Hz_10ms_log_rms_mae_db": 1.7894542799230335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14120147216236623, "300–1000 Hz_10ms_envelope_correlation": 0.9850641629058311, "300–1000 Hz_10ms_modulation_mae_db": 0.8196292194293108, "300–1000 Hz_40ms_log_rms_mae_db": 1.1814255489378065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0812661571647061, "300–1000 Hz_40ms_envelope_correlation": 0.9947625703118703, "300–1000 Hz_40ms_modulation_mae_db": 0.2735410634871642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8779488728246365, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21145601477766263, "1000–3000 Hz_10ms_envelope_correlation": 0.9655766015358751, "1000–3000 Hz_10ms_modulation_mae_db": 0.5915819716837708, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0755687353636738, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11079060089892791, "1000–3000 Hz_40ms_envelope_correlation": 0.9902846407383452, "1000–3000 Hz_40ms_modulation_mae_db": 0.2861180316571279, "3000–6000 Hz_10ms_log_rms_mae_db": 2.358920709815311, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35918646860378073, "3000–6000 Hz_10ms_envelope_correlation": 0.923079756476708, "3000–6000 Hz_10ms_modulation_mae_db": 0.42505163644609123, "3000–6000 Hz_40ms_log_rms_mae_db": 1.813098945390852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3037050660329597, "3000–6000 Hz_40ms_envelope_correlation": 0.9418126344074247, "3000–6000 Hz_40ms_modulation_mae_db": 0.6231513553023735, "mr_stft_db": 6.006348428694307, "band_mae_db": 0.355082096297205, "rms_gain_error_db": 0.05473882734910776, "output_peak": 0.4868724263599038}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.179673028664901, "全帯域_10ms_rms_envelope_nrmse": 0.09935500630878642, "全帯域_10ms_envelope_correlation": 0.9909383622021197, "全帯域_10ms_modulation_mae_db": 0.17465059843063255, "全帯域_40ms_log_rms_mae_db": 0.6091683752981429, "全帯域_40ms_rms_envelope_nrmse": 0.053103371878908186, "全帯域_40ms_envelope_correlation": 0.9973343221295281, "全帯域_40ms_modulation_mae_db": 0.35629769871433264, "300–1000 Hz_10ms_log_rms_mae_db": 1.7230579181648573, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1287051164352933, "300–1000 Hz_10ms_envelope_correlation": 0.9877740038679627, "300–1000 Hz_10ms_modulation_mae_db": 0.7543586666817776, "300–1000 Hz_40ms_log_rms_mae_db": 1.194582972823547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06773518292640174, "300–1000 Hz_40ms_envelope_correlation": 0.9964791085364137, "300–1000 Hz_40ms_modulation_mae_db": 0.21956568159456383, "1000–3000 Hz_10ms_log_rms_mae_db": 1.895792558176423, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18613172022254665, "1000–3000 Hz_10ms_envelope_correlation": 0.973433311076964, "1000–3000 Hz_10ms_modulation_mae_db": 0.11942810842198746, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1061128113702352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10665741593794276, "1000–3000 Hz_40ms_envelope_correlation": 0.9909240250234184, "1000–3000 Hz_40ms_modulation_mae_db": 0.3456762738434656, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3389480708101718, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3667910012581013, "3000–6000 Hz_10ms_envelope_correlation": 0.9168463778128261, "3000–6000 Hz_10ms_modulation_mae_db": 0.269503382861449, "3000–6000 Hz_40ms_log_rms_mae_db": 1.78767161655206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.304552483182814, "3000–6000 Hz_40ms_envelope_correlation": 0.9390800139535687, "3000–6000 Hz_40ms_modulation_mae_db": 0.09956333734175236, "mr_stft_db": 6.025675012108221, "band_mae_db": 0.35852389922061384, "rms_gain_error_db": 0.08047096453187826, "output_peak": 0.45788048979840895}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1625705554824504, "全帯域_10ms_rms_envelope_nrmse": 0.11601570463544614, "全帯域_10ms_envelope_correlation": 0.9871533406487519, "全帯域_10ms_modulation_mae_db": 0.12111690199282725, "全帯域_40ms_log_rms_mae_db": 0.6139446283899384, "全帯域_40ms_rms_envelope_nrmse": 0.05944665211801353, "全帯域_40ms_envelope_correlation": 0.9963850115891509, "全帯域_40ms_modulation_mae_db": 0.3967209186971183, "300–1000 Hz_10ms_log_rms_mae_db": 1.800696815389204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1642554724481993, "300–1000 Hz_10ms_envelope_correlation": 0.9795561544785314, "300–1000 Hz_10ms_modulation_mae_db": 0.37395732689960365, "300–1000 Hz_40ms_log_rms_mae_db": 1.1802868640251039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0898083979425539, "300–1000 Hz_40ms_envelope_correlation": 0.993777512821652, "300–1000 Hz_40ms_modulation_mae_db": 1.0377652641106498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8265180763979023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19165746028719116, "1000–3000 Hz_10ms_envelope_correlation": 0.9719927516486504, "1000–3000 Hz_10ms_modulation_mae_db": 0.5828424264512473, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0083137124718378, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11195186269437836, "1000–3000 Hz_40ms_envelope_correlation": 0.9898048437451897, "1000–3000 Hz_40ms_modulation_mae_db": 0.5409081214608321, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2898996794551065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3622844539528858, "3000–6000 Hz_10ms_envelope_correlation": 0.9234052055456246, "3000–6000 Hz_10ms_modulation_mae_db": 0.19125468331873774, "3000–6000 Hz_40ms_log_rms_mae_db": 1.760292716066803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30198765918598197, "3000–6000 Hz_40ms_envelope_correlation": 0.9441338173507424, "3000–6000 Hz_40ms_modulation_mae_db": 0.19442121169359186, "mr_stft_db": 6.005555732364357, "band_mae_db": 0.46501219554969, "rms_gain_error_db": -0.06161142007523351, "output_peak": 0.42551274556636054}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0937685320392725, "全帯域_10ms_rms_envelope_nrmse": 0.09751040838569155, "全帯域_10ms_envelope_correlation": 0.9910116565749421, "全帯域_10ms_modulation_mae_db": 0.42661056054503727, "全帯域_40ms_log_rms_mae_db": 0.5398686066107017, "全帯域_40ms_rms_envelope_nrmse": 0.04734949212216734, "全帯域_40ms_envelope_correlation": 0.9977839231573409, "全帯域_40ms_modulation_mae_db": 0.18926712398400816, "300–1000 Hz_10ms_log_rms_mae_db": 1.7441888323796277, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13003491723717156, "300–1000 Hz_10ms_envelope_correlation": 0.9873468558217383, "300–1000 Hz_10ms_modulation_mae_db": 1.3161125630161405, "300–1000 Hz_40ms_log_rms_mae_db": 1.1912337535482975, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06641578685337125, "300–1000 Hz_40ms_envelope_correlation": 0.9967739039938637, "300–1000 Hz_40ms_modulation_mae_db": 0.6732104018142736, "1000–3000 Hz_10ms_log_rms_mae_db": 1.896418464933125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18700178684219362, "1000–3000 Hz_10ms_envelope_correlation": 0.9733690093751765, "1000–3000 Hz_10ms_modulation_mae_db": 0.45792904875780804, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0885289894154255, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10788488704821159, "1000–3000 Hz_40ms_envelope_correlation": 0.9912220943764295, "1000–3000 Hz_40ms_modulation_mae_db": 0.6092318789566714, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3139186982174027, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3619435308808064, "3000–6000 Hz_10ms_envelope_correlation": 0.919588348748488, "3000–6000 Hz_10ms_modulation_mae_db": 0.5589533780275348, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8168985297919322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3052778075735746, "3000–6000 Hz_40ms_envelope_correlation": 0.9393128741093847, "3000–6000 Hz_40ms_modulation_mae_db": 0.459701583955717, "mr_stft_db": 6.035870642823487, "band_mae_db": 0.3533330388124669, "rms_gain_error_db": -0.08621622985013755, "output_peak": 0.4760347567145317}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.366506802590621, "spectral_convergence": 0.3623713299167973, "spectral_convergence_status": "ok", "amplitude_floor": 1.4536470931091146e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6294423014707, "spectral_convergence": 0.34039292863475046, "spectral_convergence_status": "ok", "amplitude_floor": 9.268471977763907e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.054710722228561, "spectral_convergence": 0.3798815072805899, "spectral_convergence_status": "ok", "amplitude_floor": 5.673364127394706e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031816153358243563, "output_power": 0.0003243285633614068, "delta_db": 0.08337526988423138, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009151577935124143, "output_power": 0.0008193187306210624, "delta_db": -0.48043099057121763, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00046151207587702967, "output_power": 0.0004578502631693692, "delta_db": -0.03459601062366004, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011338295265990407, "output_power": 0.0001047393311156244, "delta_db": -0.3443796638236715, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.877003374326731e-05, "output_power": 4.0402849267847125e-05, "delta_db": 0.17915814277772224, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3868434770644032e-05, "output_power": 1.9565625030207986e-05, "delta_db": 1.4946627795582375, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4361004762064567, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.22670826864223823, "pitch": null, "output_rms_dbfs": -27.529588009601994, "output_peak": 0.37308530839627946, "reference_rms_dbfs": -27.30287974095976, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0168866087632935}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016375851184643225, "log_rms_mae_db": 1.1253569671769286, "rms_envelope_nrmse": 0.0938181293585995, "envelope_correlation": 0.9921105930587749, "rms_gain_error_db": -0.2266451573949975, "reference_mean_power": 0.0018623503698243853, "output_mean_power": 0.0017676524266247414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.48847287729811e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002826666045552883, "output_power": 0.00026122443208854095, "reference_active": true, "delta_db": -0.34260708429037}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.155794557866922e-05, "output_power": 5.176044726791082e-05, "reference_active": true, "delta_db": -0.7528609688795528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.629343961311588e-06, "output_power": 9.965454951637102e-06, "reference_active": true, "delta_db": 0.6251935018552435}], "active_band_count": 3, "active_band_mae_db": 0.5735538516750555}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014981717790874295, "log_rms_mae_db": 0.6160108993148273, "rms_envelope_nrmse": 0.058482655565886485, "envelope_correlation": 0.9971710505240822, "rms_gain_error_db": -0.2265590958773705, "reference_mean_power": 0.0018713077892111094, "output_mean_power": 0.0017761895709401286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.792361294396736e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025909451544035513, "output_power": 0.00023829449980879549, "reference_active": true, "delta_db": -0.3634420155879038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9672561350575063e-05, "output_power": 2.4748222940598044e-05, "reference_active": true, "delta_db": -0.7881101681403395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.579324289571111e-07, "output_power": 8.000032407897304e-07, "reference_active": true, "delta_db": 0.2346125708816825}], "active_band_count": 3, "active_band_mae_db": 0.46205491820330863}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014318840727902982, "log_rms_mae_db": 1.7852689293460564, "rms_envelope_nrmse": 0.13021521927090438, "envelope_correlation": 0.9885779536808291, "rms_gain_error_db": -0.4441983296304453, "reference_mean_power": 0.0010573416450716808, "output_mean_power": 0.0009545430171485205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.91879361710777e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021964908416761717, "output_power": 0.00018993266764726844, "reference_active": true, "delta_db": -0.6312972875516527}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.090309412479679e-05, "output_power": 2.9325406871167506e-05, "reference_active": true, "delta_db": -1.4451211517868865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.52813465615998e-06, "output_power": 5.969410491916929e-06, "reference_active": true, "delta_db": 1.2001211129491849}], "active_band_count": 3, "active_band_mae_db": 1.0921798507625746}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013022613407710025, "log_rms_mae_db": 1.2233331320834673, "rms_envelope_nrmse": 0.09157507788633745, "envelope_correlation": 0.9951213271449604, "rms_gain_error_db": -0.4442007514205175, "reference_mean_power": 0.001062644583935512, "output_mean_power": 0.0009593298499118995}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.561667844043168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020949277098297487, "output_power": 0.00018087204619737953, "reference_active": true, "delta_db": -0.6379758949070203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.887160269235181e-05, "output_power": 1.3064680677944018e-05, "reference_active": true, "delta_db": -1.5970998535548633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.293974094910667e-07, "output_power": 4.221911020606553e-07, "reference_active": true, "delta_db": -0.07350344013175462}], "active_band_count": 3, "active_band_mae_db": 0.7695263961978794}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.227046231131274e-05, "log_rms_mae_db": 1.8771824010627896, "rms_envelope_nrmse": 0.1887362560414663, "envelope_correlation": 0.9726805144276206, "rms_gain_error_db": -0.1754283477071529, "reference_mean_power": 0.00011346719055744512, "output_mean_power": 0.00010897514790073691}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5656809677774876e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2313251419233995e-05, "output_power": 2.2204185582695075e-05, "reference_active": true, "delta_db": -0.021280104576049175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.900047334435587e-06, "output_power": 2.9857738543443486e-06, "reference_active": true, "delta_db": 0.12651824222473967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0009489505393011e-06, "output_power": 1.3640395709988087e-06, "reference_active": true, "delta_db": 1.344150408894673}], "active_band_count": 3, "active_band_mae_db": 0.49731625189848727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.589293372832859e-05, "log_rms_mae_db": 1.034377358009375, "rms_envelope_nrmse": 0.09972072159625212, "envelope_correlation": 0.9921039948552636, "rms_gain_error_db": -0.17554875280870477, "reference_mean_power": 0.0001140327901646041, "output_mean_power": 0.00010951531975706659}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.160520396268688e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1030857564945278e-05, "output_power": 2.141530083335284e-05, "reference_active": true, "delta_db": 0.07867197505914889}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3790687257240336e-06, "output_power": 1.4510743898093725e-06, "reference_active": true, "delta_db": 0.22103767530542645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.278096740186873e-08, "output_power": 9.5861972411536e-08, "reference_active": true, "delta_db": 1.1962853667534605}], "active_band_count": 3, "active_band_mae_db": 0.49866500570601197}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9795267482339013e-05, "log_rms_mae_db": 2.3025169877645086, "rms_envelope_nrmse": 0.35556438935838186, "envelope_correlation": 0.9264993290437357, "rms_gain_error_db": 0.6229805723279411, "reference_mean_power": 2.1804807525571507e-05, "output_mean_power": 2.5168093266964617e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7142973306691612e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0022096022650052e-05, "output_power": 1.0261765349378187e-05, "reference_active": true, "delta_db": 0.1026352019673392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6747496282854573e-06, "output_power": 3.1212808490522255e-06, "reference_active": true, "delta_db": 0.6704971187595103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.037734768320834e-07, "output_power": 7.903453621470055e-07, "reference_active": true, "delta_db": -0.07316761970217676}], "active_band_count": 3, "active_band_mae_db": 0.2820999801430088}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.774754919111929e-05, "log_rms_mae_db": 1.799918929738733, "rms_envelope_nrmse": 0.30132392625881793, "envelope_correlation": 0.9445016048059024, "rms_gain_error_db": 0.6229329280008463, "reference_mean_power": 2.1914321405186452e-05, "output_mean_power": 2.529422163895843e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.567366861761873e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0140844119742473e-05, "output_power": 1.0371762927459407e-05, "reference_active": true, "delta_db": 0.09778474327803516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3436726086917478e-06, "output_power": 1.6119733416485455e-06, "reference_active": true, "delta_db": 0.7906439130488743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.915436539179574e-08, "output_power": 5.695394816337796e-08, "reference_active": true, "delta_db": 0.6396174212581096}], "active_band_count": 3, "active_band_mae_db": 0.5093486925283397}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8684604253685574, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2f97b089d34283fc6e2f376f62c59c616c2d356df871cf72b3df614ee15fbfed", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9710788186240047, "standard_deviation": 0.00437976448533772, "minimum": 0.9628298166833458, "maximum": 0.9745269053444229, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.868225515924009, "standard_deviation": 0.026650141881397724, "minimum": 1.8265180763979023, "maximum": 1.896418464933125, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4798524913538012, "standard_deviation": 0.20511369014365746, "minimum": 0.11942810842198746, "maximum": 0.8203999491199316, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19450266350988776, "standard_deviation": 0.014803353659701129, "minimum": 0.18229185562839337, "maximum": 0.22354236918892667, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9903226418045452, "standard_deviation": 0.0024211213118269, "minimum": 0.9846997315454901, "maximum": 0.9923219506804427, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0784041050306588, "standard_deviation": 0.04170630361883245, "minimum": 1.0083137124718378, "maximum": 1.1405148009930821, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.48762188153997454, "standard_deviation": 0.21394714815008267, "minimum": 0.27319876838021867, "maximum": 0.9253488140469367, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10986757931365329, "standard_deviation": 0.01289393319302598, "minimum": 0.09766465480675901, "maximum": 0.13933235017699533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9204751363464446, "standard_deviation": 0.0045611795862082975, "minimum": 0.9117891400747268, "maximum": 0.9264993290437357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.316917874577138, "standard_deviation": 0.024642456158350076, "minimum": 2.28784805486986, "maximum": 2.358920709815311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4031116841973967, "standard_deviation": 0.1810173245884568, "minimum": 0.19125468331873774, "maximum": 0.755764253423704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3613858570163277, "standard_deviation": 0.005608714183583368, "minimum": 0.35184948014473894, "maximum": 0.3679585360460388, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.940053131970481, "standard_deviation": 0.0043700798218275355, "minimum": 0.9314908772891101, "maximum": 0.9445016048059024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.794273737166174, "standard_deviation": 0.027250417272720998, "minimum": 1.760292716066803, "maximum": 1.8365775542087845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4560182596831445, "standard_deviation": 0.2217973162092764, "minimum": 0.09956333734175236, "maximum": 0.7883175634006054, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3048779692931073, "standard_deviation": 0.00401586978616199, "minimum": 0.30132392625881793, "maximum": 0.3140340841004521, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9851038607635991, "standard_deviation": 0.0031197794250028815, "minimum": 0.9795561544785314, "maximum": 0.9885779536808291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7744264790438566, "standard_deviation": 0.032244310510629225, "minimum": 1.7230579181648573, "maximum": 1.8143344814814537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0034075920663845, "standard_deviation": 0.3295957204672544, "minimum": 0.37395732689960365, "maximum": 1.3161125630161405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14105661371364428, "standard_deviation": 0.013085498154020169, "minimum": 0.1287051164352933, "maximum": 0.1642554724481993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956411409128991, "standard_deviation": 0.0010628944014989508, "minimum": 0.993777512821652, "maximum": 0.9968043484997586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1856352176671199, "standard_deviation": 0.024759906453193942, "minimum": 1.1342128301812966, "maximum": 1.2233331320834673, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6900357913744267, "standard_deviation": 0.29845706572519987, "minimum": 0.21956568159456383, "maximum": 1.0377652641106498, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07647791693805442, "standard_deviation": 0.01067106006392091, "minimum": 0.06371795201941186, "maximum": 0.09157507788633745, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3690086603716999, "standard_deviation": 0.059191246481064844, "minimum": 0.29109376317638486, "maximum": 0.46501219554969, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.018596409127632, "standard_deviation": 0.012655022219468294, "minimum": 6.002609854028399, "maximum": 6.035870642823487, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.43934824599576094, "standard_deviation": 0.039188308992020915, "minimum": 0.37308530839627946, "maximum": 0.4868724263599038, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03957880533824525, "standard_deviation": 0.10435861746123433, "minimum": -0.22670826864223823, "maximum": 0.08047096453187826, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9894490136382976, "standard_deviation": 0.0020685241894079966, "minimum": 0.9863143585669014, "maximum": 0.9921105930587749, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.138019780045794, "standard_deviation": 0.030639708678050957, "minimum": 1.0937685320392725, "maximum": 1.179673028664901, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3729870389747144, "standard_deviation": 0.1852294688867693, "minimum": 0.12111690199282725, "maximum": 0.578900992352054, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10569187256362964, "standard_deviation": 0.009617837284292776, "minimum": 0.0938181293585995, "maximum": 0.12006304362029552, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970992170843218, "standard_deviation": 0.0004951263396703748, "minimum": 0.9963850115891509, "maximum": 0.9977839231573409, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5892141362936045, "standard_deviation": 0.02771629693718829, "minimum": 0.5398686066107017, "maximum": 0.6160108993148273, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3320240724051754, "standard_deviation": 0.09154203554659139, "minimum": 0.18926712398400816, "maximum": 0.46205491820330863, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054473126499407803, "standard_deviation": 0.004571778351940039, "minimum": 0.04734949212216734, "maximum": 0.05944665211801353, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1089-134686-0011", "language": "en", "style": "normal", "speaker": "librispeech_1089", "source_utterance_id": "1089-134686-0011", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1089-134686-0011.wav", "source_file_sha256": "fd85a2e73fb95e38d373f7227e7c5b46767b4a9ff6a5180d40c3aadb305baa85", "decoded_pcm_sha256": "3db35e4a6ed6a12218e0ab1bfe72f1810bad8f201648ec0e2a9af7dae7aa32eb", "original_source_sha256": "c65e16be3ef92c11a31a9c2c4c570542eb2ed66153150a7c4cca00c01b30a4e2", "corpus_entry_id": "librispeech_1089-134686-0011"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.290397644042969, "predicted_mos": 4.290397644042969, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f4a352d9828bf71d7120a3cdb583d04065a14d4100cc6ad72a997590d4df31d6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.312535899691284, "cpu_threads": 4}, "waveform_sha256": "a66507e24c1f41aeef2407f91575332cabbbd6656e90a56803a001b4fcbf9333"}, "rows": [{"case": "librispeech_1089-134686-0011", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.207479238510132, "predicted_mos": 2.207479238510132, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "791f5b94ea89ea36eb907ba785a9d0aa6966c39eef57bf59704e668fa8ac77d1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.154300199821591, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.00848328800796, "全帯域_10ms_rms_envelope_nrmse": 0.07916988044589256, "全帯域_10ms_envelope_correlation": 0.9941523094775775, "全帯域_10ms_modulation_mae_db": 0.126610451293661, "全帯域_40ms_log_rms_mae_db": 0.5618148636591328, "全帯域_40ms_rms_envelope_nrmse": 0.045818492817202044, "全帯域_40ms_envelope_correlation": 0.9980537829402071, "全帯域_40ms_modulation_mae_db": 0.06366600244660087, "300–1000 Hz_10ms_log_rms_mae_db": 1.777124313663679, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10753583538139024, "300–1000 Hz_10ms_envelope_correlation": 0.991560497588342, "300–1000 Hz_10ms_modulation_mae_db": 0.6142836463830322, "300–1000 Hz_40ms_log_rms_mae_db": 1.0849458142041606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06288933382109214, "300–1000 Hz_40ms_envelope_correlation": 0.9972041572220411, "300–1000 Hz_40ms_modulation_mae_db": 0.9084412720041982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7995376851600937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14122590990607795, "1000–3000 Hz_10ms_envelope_correlation": 0.9854775345624205, "1000–3000 Hz_10ms_modulation_mae_db": 0.45950036681396694, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9480272181170573, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08694922934870165, "1000–3000 Hz_40ms_envelope_correlation": 0.99430326454631, "1000–3000 Hz_40ms_modulation_mae_db": 0.6001185725578266, "3000–6000 Hz_10ms_log_rms_mae_db": 2.269981760456873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1888967864378498, "3000–6000 Hz_10ms_envelope_correlation": 0.9747185528455676, "3000–6000 Hz_10ms_modulation_mae_db": 0.8925031053788443, "3000–6000 Hz_40ms_log_rms_mae_db": 1.42646120234389, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12041827698029486, "3000–6000 Hz_40ms_envelope_correlation": 0.9895003119093068, "3000–6000 Hz_40ms_modulation_mae_db": 0.8495651770233975, "mr_stft_db": 5.839719195685588, "band_mae_db": 0.16610301270123148, "rms_gain_error_db": -0.13927228994619356, "output_peak": 0.5006850034796816}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0560536352571401, "全帯域_10ms_rms_envelope_nrmse": 0.07305803508088822, "全帯域_10ms_envelope_correlation": 0.9949475148982655, "全帯域_10ms_modulation_mae_db": 0.15923286512739457, "全帯域_40ms_log_rms_mae_db": 0.5509387267812728, "全帯域_40ms_rms_envelope_nrmse": 0.038068001163431515, "全帯域_40ms_envelope_correlation": 0.9986026323203778, "全帯域_40ms_modulation_mae_db": 0.2554766956349835, "300–1000 Hz_10ms_log_rms_mae_db": 1.7637424504070682, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09739870604889862, "300–1000 Hz_10ms_envelope_correlation": 0.9929961412613828, "300–1000 Hz_10ms_modulation_mae_db": 0.42141081310098977, "300–1000 Hz_40ms_log_rms_mae_db": 1.0414201594699424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05250316049863951, "300–1000 Hz_40ms_envelope_correlation": 0.9979871586828764, "300–1000 Hz_40ms_modulation_mae_db": 0.7680407398504306, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8669745818388173, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18356906005471182, "1000–3000 Hz_10ms_envelope_correlation": 0.9752648678587625, "1000–3000 Hz_10ms_modulation_mae_db": 0.5201685274644512, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0531256833032518, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1180285142252902, "1000–3000 Hz_40ms_envelope_correlation": 0.9893722143509412, "1000–3000 Hz_40ms_modulation_mae_db": 0.4927070760321472, "3000–6000 Hz_10ms_log_rms_mae_db": 2.268262872404078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20725942827775212, "3000–6000 Hz_10ms_envelope_correlation": 0.9693460247763969, "3000–6000 Hz_10ms_modulation_mae_db": 0.8047600533588, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4743284910812653, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13743611397942804, "3000–6000 Hz_40ms_envelope_correlation": 0.9858661304003967, "3000–6000 Hz_40ms_modulation_mae_db": 0.7963555069171723, "mr_stft_db": 5.833486100734784, "band_mae_db": 0.13519224289517331, "rms_gain_error_db": -0.0892924158007552, "output_peak": 0.5270500930312775}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0446128196630335, "全帯域_10ms_rms_envelope_nrmse": 0.07904994438235768, "全帯域_10ms_envelope_correlation": 0.9940146229300101, "全帯域_10ms_modulation_mae_db": 0.37743504476424644, "全帯域_40ms_log_rms_mae_db": 0.5624581542935947, "全帯域_40ms_rms_envelope_nrmse": 0.04592178477839142, "全帯域_40ms_envelope_correlation": 0.9978171695116571, "全帯域_40ms_modulation_mae_db": 0.4868279707461367, "300–1000 Hz_10ms_log_rms_mae_db": 1.8011720413420316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10321377382849184, "300–1000 Hz_10ms_envelope_correlation": 0.991988665437289, "300–1000 Hz_10ms_modulation_mae_db": 0.655674647576161, "300–1000 Hz_40ms_log_rms_mae_db": 1.0405147236580583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05892186221388797, "300–1000 Hz_40ms_envelope_correlation": 0.9972339388093757, "300–1000 Hz_40ms_modulation_mae_db": 0.4817036347611652, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8319803114761406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15068584631772472, "1000–3000 Hz_10ms_envelope_correlation": 0.9835227671411683, "1000–3000 Hz_10ms_modulation_mae_db": 0.5211844762906669, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0035745050076312, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09327492257129237, "1000–3000 Hz_40ms_envelope_correlation": 0.9935840668223274, "1000–3000 Hz_40ms_modulation_mae_db": 0.810634082378586, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2815490245165666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20333796260985448, "3000–6000 Hz_10ms_envelope_correlation": 0.9706120256288921, "3000–6000 Hz_10ms_modulation_mae_db": 0.5575005485026757, "3000–6000 Hz_40ms_log_rms_mae_db": 1.430140042607415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13166924753347753, "3000–6000 Hz_40ms_envelope_correlation": 0.9868198377928408, "3000–6000 Hz_40ms_modulation_mae_db": 0.6615092613619907, "mr_stft_db": 5.839767253040512, "band_mae_db": 0.13939766612665846, "rms_gain_error_db": -0.031933388786284045, "output_peak": 0.49523599994080186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0442115228480624, "全帯域_10ms_rms_envelope_nrmse": 0.0701808669569854, "全帯域_10ms_envelope_correlation": 0.9953499488233358, "全帯域_10ms_modulation_mae_db": 0.3075841173581885, "全帯域_40ms_log_rms_mae_db": 0.5189974200886146, "全帯域_40ms_rms_envelope_nrmse": 0.03636354421332883, "全帯域_40ms_envelope_correlation": 0.9987283890982823, "全帯域_40ms_modulation_mae_db": 0.341076356006226, "300–1000 Hz_10ms_log_rms_mae_db": 1.7550605258695837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09266121581488934, "300–1000 Hz_10ms_envelope_correlation": 0.9936483986527604, "300–1000 Hz_10ms_modulation_mae_db": 0.533945654852035, "300–1000 Hz_40ms_log_rms_mae_db": 1.0474389815097913, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05240192385285011, "300–1000 Hz_40ms_envelope_correlation": 0.9979388670493735, "300–1000 Hz_40ms_modulation_mae_db": 0.8405961884707067, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8161281360654689, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1686488603789892, "1000–3000 Hz_10ms_envelope_correlation": 0.9794167347123041, "1000–3000 Hz_10ms_modulation_mae_db": 0.35535424602469273, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0237480606458575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11376822393125008, "1000–3000 Hz_40ms_envelope_correlation": 0.9905176219595364, "1000–3000 Hz_40ms_modulation_mae_db": 0.5441028641988049, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2368916210089536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19434710438712552, "3000–6000 Hz_10ms_envelope_correlation": 0.9731752052678513, "3000–6000 Hz_10ms_modulation_mae_db": 0.594211382703368, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3890235109360582, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12296396685861032, "3000–6000 Hz_40ms_envelope_correlation": 0.988500933332924, "3000–6000 Hz_40ms_modulation_mae_db": 0.8301068626845782, "mr_stft_db": 5.820791687211746, "band_mae_db": 0.14287422222340587, "rms_gain_error_db": -0.09115326634653753, "output_peak": 0.5198394308212653}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0073564218571942, "全帯域_10ms_rms_envelope_nrmse": 0.08933269122515339, "全帯域_10ms_envelope_correlation": 0.9923464529097611, "全帯域_10ms_modulation_mae_db": 0.37927587522322237, "全帯域_40ms_log_rms_mae_db": 0.5188776448505703, "全帯域_40ms_rms_envelope_nrmse": 0.048189390126060154, "全帯域_40ms_envelope_correlation": 0.9976055962264496, "全帯域_40ms_modulation_mae_db": 0.4551155879907603, "300–1000 Hz_10ms_log_rms_mae_db": 1.7989579809521148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10817345250343868, "300–1000 Hz_10ms_envelope_correlation": 0.9912360250097277, "300–1000 Hz_10ms_modulation_mae_db": 0.3503453481367165, "300–1000 Hz_40ms_log_rms_mae_db": 1.0300385362118665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055166625109481865, "300–1000 Hz_40ms_envelope_correlation": 0.99767017433723, "300–1000 Hz_40ms_modulation_mae_db": 0.07957943078988779, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8462674034784643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.163287594758124, "1000–3000 Hz_10ms_envelope_correlation": 0.9804702233080471, "1000–3000 Hz_10ms_modulation_mae_db": 0.18808315546046195, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0470046401496553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09681262402423248, "1000–3000 Hz_40ms_envelope_correlation": 0.992858299901449, "1000–3000 Hz_40ms_modulation_mae_db": 0.5958108234438718, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2476347786555886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2083042457607255, "3000–6000 Hz_10ms_envelope_correlation": 0.9696593552607622, "3000–6000 Hz_10ms_modulation_mae_db": 0.7300247170150255, "3000–6000 Hz_40ms_log_rms_mae_db": 1.421801090649624, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13599850919262887, "3000–6000 Hz_40ms_envelope_correlation": 0.9860967460036811, "3000–6000 Hz_40ms_modulation_mae_db": 1.0342349959230228, "mr_stft_db": 5.818907509435545, "band_mae_db": 0.12502848953546394, "rms_gain_error_db": -0.027909468992589805, "output_peak": 0.5137403087611481}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0001817943374673, "全帯域_10ms_rms_envelope_nrmse": 0.07260130271176253, "全帯域_10ms_envelope_correlation": 0.9949529437853847, "全帯域_10ms_modulation_mae_db": 0.06783075923793824, "全帯域_40ms_log_rms_mae_db": 0.5149769056190368, "全帯域_40ms_rms_envelope_nrmse": 0.03714741924563236, "全帯域_40ms_envelope_correlation": 0.9985685901532719, "全帯域_40ms_modulation_mae_db": 0.19835493827744344, "300–1000 Hz_10ms_log_rms_mae_db": 1.7697602717549001, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10207799496752178, "300–1000 Hz_10ms_envelope_correlation": 0.9921671438747939, "300–1000 Hz_10ms_modulation_mae_db": 0.42188586157691804, "300–1000 Hz_40ms_log_rms_mae_db": 1.0475108132371114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05314822211349991, "300–1000 Hz_40ms_envelope_correlation": 0.9977305622528208, "300–1000 Hz_40ms_modulation_mae_db": 0.2458739480616561, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8815397331441455, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15340108893486168, "1000–3000 Hz_10ms_envelope_correlation": 0.9828009477603532, "1000–3000 Hz_10ms_modulation_mae_db": 0.45369630654084564, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0221914871825184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09840690902951607, "1000–3000 Hz_40ms_envelope_correlation": 0.9926127451852922, "1000–3000 Hz_40ms_modulation_mae_db": 0.6707085336941411, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2543344852747493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21214198614126023, "3000–6000 Hz_10ms_envelope_correlation": 0.9694646758393247, "3000–6000 Hz_10ms_modulation_mae_db": 0.7782340968477747, "3000–6000 Hz_40ms_log_rms_mae_db": 1.433815920936933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13770699462037891, "3000–6000 Hz_40ms_envelope_correlation": 0.986412957804453, "3000–6000 Hz_40ms_modulation_mae_db": 0.6770924586653839, "mr_stft_db": 5.828626264550214, "band_mae_db": 0.1693716143233963, "rms_gain_error_db": -0.01567030490969483, "output_peak": 0.5085513093709324}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0423989476998108, "全帯域_10ms_rms_envelope_nrmse": 0.07769029645544022, "全帯域_10ms_envelope_correlation": 0.9943911627790457, "全帯域_10ms_modulation_mae_db": 0.21719873202389905, "全帯域_40ms_log_rms_mae_db": 0.5264936926730883, "全帯域_40ms_rms_envelope_nrmse": 0.04037988654159928, "全帯域_40ms_envelope_correlation": 0.9985759898251665, "全帯域_40ms_modulation_mae_db": 0.28399811383365475, "300–1000 Hz_10ms_log_rms_mae_db": 1.8011786445758988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10741524229694861, "300–1000 Hz_10ms_envelope_correlation": 0.9918205986993439, "300–1000 Hz_10ms_modulation_mae_db": 0.47708488072235616, "300–1000 Hz_40ms_log_rms_mae_db": 1.083052047204155, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06211609198351602, "300–1000 Hz_40ms_envelope_correlation": 0.99759409863193, "300–1000 Hz_40ms_modulation_mae_db": 0.8510661627594507, "1000–3000 Hz_10ms_log_rms_mae_db": 1.865294702087854, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16377793880502042, "1000–3000 Hz_10ms_envelope_correlation": 0.9805602358409444, "1000–3000 Hz_10ms_modulation_mae_db": 0.5555495961123171, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0509426453795647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10525228639356603, "1000–3000 Hz_40ms_envelope_correlation": 0.9918876318358134, "1000–3000 Hz_40ms_modulation_mae_db": 0.8487789867376664, "3000–6000 Hz_10ms_log_rms_mae_db": 2.27661465960248, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19928847461115679, "3000–6000 Hz_10ms_envelope_correlation": 0.9716480448267969, "3000–6000 Hz_10ms_modulation_mae_db": 0.7781944649648125, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4349669544545218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13316136327115896, "3000–6000 Hz_40ms_envelope_correlation": 0.9866410925476883, "3000–6000 Hz_40ms_modulation_mae_db": 0.6545051943208305, "mr_stft_db": 5.84355948314667, "band_mae_db": 0.19784349882224994, "rms_gain_error_db": -0.1420282277820266, "output_peak": 0.5012409453233807}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0282995799922447, "全帯域_10ms_rms_envelope_nrmse": 0.07544119088744, "全帯域_10ms_envelope_correlation": 0.9945531073477274, "全帯域_10ms_modulation_mae_db": 0.12199706563391971, "全帯域_40ms_log_rms_mae_db": 0.5424921377166705, "全帯域_40ms_rms_envelope_nrmse": 0.040184674473929476, "全帯域_40ms_envelope_correlation": 0.9983417814071264, "全帯域_40ms_modulation_mae_db": 0.432401237316268, "300–1000 Hz_10ms_log_rms_mae_db": 1.7380495843016128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10139664112545288, "300–1000 Hz_10ms_envelope_correlation": 0.9922724398079898, "300–1000 Hz_10ms_modulation_mae_db": 0.40447599449514343, "300–1000 Hz_40ms_log_rms_mae_db": 1.0556502701164912, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055033691883054695, "300–1000 Hz_40ms_envelope_correlation": 0.9975675727619192, "300–1000 Hz_40ms_modulation_mae_db": 0.3905975190370717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8450140201291343, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1693554735114134, "1000–3000 Hz_10ms_envelope_correlation": 0.9795565490702659, "1000–3000 Hz_10ms_modulation_mae_db": 0.5287903167494553, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0770990182948854, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11102907380402112, "1000–3000 Hz_40ms_envelope_correlation": 0.9914872066648716, "1000–3000 Hz_40ms_modulation_mae_db": 0.7273986802003267, "3000–6000 Hz_10ms_log_rms_mae_db": 2.266358541370239, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19541247278999166, "3000–6000 Hz_10ms_envelope_correlation": 0.9728111587728171, "3000–6000 Hz_10ms_modulation_mae_db": 0.6000469109273902, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4862812985351797, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12827141698104613, "3000–6000 Hz_40ms_envelope_correlation": 0.9874163255600934, "3000–6000 Hz_40ms_modulation_mae_db": 0.847608297453136, "mr_stft_db": 5.842022890268486, "band_mae_db": 0.12111881508080835, "rms_gain_error_db": -0.04355571897069865, "output_peak": 0.517974777150861}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.306960884400956, "spectral_convergence": 0.2486845068842256, "spectral_convergence_status": "ok", "amplitude_floor": 1.4412887028114478e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.495130353427159, "spectral_convergence": 0.24392775409474285, "spectral_convergence_status": "ok", "amplitude_floor": 8.082487629299873e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.717066349228648, "spectral_convergence": 0.2585823703093344, "spectral_convergence_status": "ok", "amplitude_floor": 6.705293681874596e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00036448625115270087, "output_power": 0.0003724946968578188, "delta_db": 0.09438943285730217, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008505212390156939, "output_power": 0.0008336023896741891, "delta_db": -0.08726212241282645, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010532918945599765, "output_power": 0.0009884692914952564, "delta_db": -0.2758556040230877, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002198202138287696, "output_power": 0.00021639191508712312, "delta_db": -0.06826595518137324, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.872413022951283e-05, "output_power": 3.661431153332603e-05, "delta_db": -0.2433079957098851, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.441903264074617e-05, "output_power": 2.317259966796558e-05, "delta_db": -0.22753696602291432, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16610301270123148, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13927228994619356, "pitch": null, "output_rms_dbfs": -26.071720389221767, "output_peak": 0.5006850034796816, "reference_rms_dbfs": -25.93244809927558, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.839719195685588}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019142337741905198, "log_rms_mae_db": 1.00848328800796, "rms_envelope_nrmse": 0.07916988044589256, "envelope_correlation": 0.9941523094775775, "rms_gain_error_db": -0.13928656069284065, "reference_mean_power": 0.0025533227171819564, "output_mean_power": 0.002472732022791361}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1748402838347296e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007147512141697323, "output_power": 0.000708927562929459, "reference_active": true, "delta_db": -0.035530398369276385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.175142428480134e-05, "output_power": 8.165823039568269e-05, "reference_active": true, "delta_db": -0.004953635849547487}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.431203959050675e-05, "output_power": 1.3236305946113477e-05, "reference_active": true, "delta_db": -0.33934731966215914}], "active_band_count": 3, "active_band_mae_db": 0.126610451293661}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016640454830055276, "log_rms_mae_db": 0.5618148636591328, "rms_envelope_nrmse": 0.045818492817202044, "envelope_correlation": 0.9980537829402071, "rms_gain_error_db": -0.13897029046467607, "reference_mean_power": 0.002564981061322552, "output_mean_power": 0.002484203296655418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0626539543337418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006440432662381922, "output_power": 0.0006385608128204947, "reference_active": true, "delta_db": -0.037127805950246215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1296244809426825e-05, "output_power": 4.082812473502977e-05, "reference_active": true, "delta_db": -0.04951128835807873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.585289215752233e-07, "output_power": 8.794088285763748e-07, "reference_active": true, "delta_db": 0.10435891303147767}], "active_band_count": 3, "active_band_mae_db": 0.06366600244660087}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016776043588712122, "log_rms_mae_db": 1.777124313663679, "rms_envelope_nrmse": 0.10753583538139024, "envelope_correlation": 0.991560497588342, "rms_gain_error_db": -0.2092470784614642, "reference_mean_power": 0.0015150798695545982, "output_mean_power": 0.0014438125750690345}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.503421266374213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005710315291680691, "output_power": 0.0005462957206910222, "reference_active": true, "delta_db": -0.19232289660341967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462070521392158e-05, "output_power": 5.810886284047859e-05, "reference_active": true, "delta_db": -0.4612931660800221}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.82379092220564e-06, "output_power": 7.470602403766131e-06, "reference_active": true, "delta_db": -1.189234876465655}], "active_band_count": 3, "active_band_mae_db": 0.6142836463830322}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015135634256581334, "log_rms_mae_db": 1.0849458142041606, "rms_envelope_nrmse": 0.06288933382109214, "envelope_correlation": 0.9972041572220411, "rms_gain_error_db": -0.20914937629855157, "reference_mean_power": 0.0015225180811750088, "output_mean_power": 0.0014509335442693691}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.758019890177841e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005253930575206644, "output_power": 0.0005039056949558665, "reference_active": true, "delta_db": -0.1813506315484391}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0513708673138778e-05, "output_power": 2.7887409852371428e-05, "reference_active": true, "delta_db": -0.39086816115472944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.334657094067365e-07, "output_power": 5.076650135043337e-07, "reference_active": true, "delta_db": -2.1531050233094264}], "active_band_count": 3, "active_band_mae_db": 0.9084412720041982}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.526806573516121e-05, "log_rms_mae_db": 1.7995376851600937, "rms_envelope_nrmse": 0.14122590990607795, "envelope_correlation": 0.9854775345624205, "rms_gain_error_db": -0.1638065245302776, "reference_mean_power": 0.00023682053730462446, "output_mean_power": 0.00022805453331918373}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1631666261710027e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014015200472863968, "output_power": 0.00013392847667181027, "reference_active": true, "delta_db": -0.197263851569423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3136457414027934e-05, "output_power": 1.4194874399329172e-05, "reference_active": true, "delta_db": 0.33653291848172884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.230915915111415e-06, "output_power": 2.6598382220690657e-06, "reference_active": true, "delta_db": -0.8447043303907489}], "active_band_count": 3, "active_band_mae_db": 0.45950036681396694}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.301677625713533e-05, "log_rms_mae_db": 0.9480272181170573, "rms_envelope_nrmse": 0.08694922934870165, "envelope_correlation": 0.99430326454631, "rms_gain_error_db": -0.164157542227784, "reference_mean_power": 0.0002379882918616051, "output_mean_power": 0.0002291605403825156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0209734833253925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001318054605596598, "output_power": 0.00012658667894843497, "reference_active": true, "delta_db": -0.17545396860889798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.132933188997116e-06, "output_power": 7.018220430161444e-06, "reference_active": true, "delta_db": 0.5855877093978502}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2319303983633163e-07, "output_power": 1.756908901615355e-07, "reference_active": true, "delta_db": -1.0393140396667317}], "active_band_count": 3, "active_band_mae_db": 0.6001185725578266}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.803319177623983e-05, "log_rms_mae_db": 2.269981760456873, "rms_envelope_nrmse": 0.1888967864378498, "envelope_correlation": 0.9747185528455676, "rms_gain_error_db": -0.2558165482632942, "reference_mean_power": 3.0517013147816785e-05, "output_mean_power": 2.8771358597529522e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.826326538605559e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1311662288856248e-05, "output_power": 1.03740899181549e-05, "reference_active": true, "delta_db": -0.3757642277643889}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5475498743040963e-06, "output_power": 2.2682490531172612e-06, "reference_active": true, "delta_db": -0.5043195684892925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.732573657220471e-07, "output_power": 8.67145941141375e-07, "reference_active": true, "delta_db": 1.7974255198828515}], "active_band_count": 3, "active_band_mae_db": 0.8925031053788443}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.459177046887288e-05, "log_rms_mae_db": 1.42646120234389, "rms_envelope_nrmse": 0.12041827698029486, "envelope_correlation": 0.9895003119093068, "rms_gain_error_db": -0.2559117835719611, "reference_mean_power": 3.0668913834731335e-05, "output_mean_power": 2.8913936106841652e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6363060360497894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1094596811060155e-05, "output_power": 1.0229600555221882e-05, "reference_active": true, "delta_db": -0.3525284841079552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1901156553217825e-06, "output_power": 1.2693971094885113e-06, "reference_active": true, "delta_db": 0.28008336920051785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.41782955712918e-08, "output_power": 6.867795016340386e-08, "reference_active": true, "delta_db": 1.9160836777617198}], "active_band_count": 3, "active_band_mae_db": 0.8495651770233975}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8771360095235333, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "403ec34fae5fb34602e5112e133cd5c7d66623e1785499daa9c0649e5f4679d3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9808837325317833, "standard_deviation": 0.0031053873405187723, "minimum": 0.9752648678587625, "maximum": 0.9854775345624205, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8440920716725149, "standard_deviation": 0.02750283102919872, "minimum": 1.7995376851600937, "maximum": 1.8815397331441455, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4477908739321072, "standard_deviation": 0.12245268072147866, "minimum": 0.18808315546046195, "maximum": 0.5555495961123171, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1617439715833654, "standard_deviation": 0.0131004002834014, "minimum": 0.14122590990607795, "maximum": 0.18356906005471182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9920778814083177, "standard_deviation": 0.001614938841028771, "minimum": 0.9893722143509412, "maximum": 0.99430326454631, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0282141572600527, "standard_deviation": 0.03959303047199091, "minimum": 0.9480272181170573, "maximum": 1.0770990182948854, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6612824524054214, "standard_deviation": 0.1265460583328606, "minimum": 0.4927070760321472, "maximum": 0.8487789867376664, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10294022291598376, "standard_deviation": 0.010842414042135996, "minimum": 0.08694922934870165, "maximum": 0.1180285142252902, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9714293804023011, "standard_deviation": 0.001993746578305866, "minimum": 0.9693460247763969, "maximum": 0.9747185528455676, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.262703467911191, "standard_deviation": 0.015162326656556289, "minimum": 2.2368916210089536, "maximum": 2.2815490245165666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7169344099623364, "standard_deviation": 0.1197391793908377, "minimum": 0.5575005485026757, "maximum": 0.8925031053788443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20112355762696452, "standard_deviation": 0.00799034076676443, "minimum": 0.1888967864378498, "maximum": 0.21214198614126023, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.987156791918923, "standard_deviation": 0.0012585954421673287, "minimum": 0.9858661304003967, "maximum": 0.9895003119093068, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.437102313943111, "standard_deviation": 0.030535988463384824, "minimum": 1.3890235109360582, "maximum": 1.4862812985351797, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.793872219293689, "standard_deviation": 0.12854309166262845, "minimum": 0.6545051943208305, "maximum": 1.0342349959230228, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13095323617712795, "standard_deviation": 0.0065518828129162805, "minimum": 0.12041827698029486, "maximum": 0.13770699462037891, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922112387914537, "standard_deviation": 0.000780460934630988, "minimum": 0.9912360250097277, "maximum": 0.9936483986527604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.775630726608361, "standard_deviation": 0.023490773979369765, "minimum": 1.7380495843016128, "maximum": 1.8011786445758988, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.484888355855419, "standard_deviation": 0.10753424236322312, "minimum": 0.3503453481367165, "maximum": 0.655674647576161, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10248410774587899, "standard_deviation": 0.00543053698967326, "minimum": 0.09266121581488934, "maximum": 0.10817345250343868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976158162184459, "standard_deviation": 0.00028697214704956244, "minimum": 0.9972041572220411, "maximum": 0.9979871586828764, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0538214182014471, "standard_deviation": 0.02001266186486095, "minimum": 1.0300385362118665, "maximum": 1.0849458142041606, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5707373619668208, "standard_deviation": 0.3143088324784383, "minimum": 0.07957943078988779, "maximum": 0.9084412720041982, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05652261393450278, "standard_deviation": 0.004245278360515747, "minimum": 0.05240192385285011, "maximum": 0.06288933382109214, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14961619521354846, "standard_deviation": 0.026083775554842988, "minimum": 0.12111881508080835, "maximum": 0.19784349882224994, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.833360048009193, "standard_deviation": 0.00963496985020331, "minimum": 5.818907509435545, "maximum": 5.84355948314667, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5105397334849185, "standard_deviation": 0.010995720952824336, "minimum": 0.49523599994080186, "maximum": 0.5270500930312775, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07260188519184753, "standard_deviation": 0.05017670310512286, "minimum": -0.1420282277820266, "maximum": -0.01567030490969483, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943385078688884, "standard_deviation": 0.0009202572305199079, "minimum": 0.9923464529097611, "maximum": 0.9953499488233358, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0289497512078642, "standard_deviation": 0.02106764905634987, "minimum": 1.0001817943374673, "maximum": 1.0560536352571401, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.21964561383280873, "standard_deviation": 0.12130641439520613, "minimum": 0.06783075923793824, "maximum": 0.37927587522322237, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07706552601824, "standard_deviation": 0.005919292579159553, "minimum": 0.0701808669569854, "maximum": 0.08933269122515339, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982867414353174, "standard_deviation": 0.00041398735079086857, "minimum": 0.9976055962264496, "maximum": 0.9987283890982823, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5371311932102476, "standard_deviation": 0.019770544054803777, "minimum": 0.5149769056190368, "maximum": 0.5624581542935947, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31461461278150915, "standard_deviation": 0.1438739813999206, "minimum": 0.06366600244660087, "maximum": 0.4868279707461367, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041509149169946885, "standard_deviation": 0.004520434873579799, "minimum": 0.03636354421332883, "maximum": 0.048189390126060154, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0011", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.329204797744751, "predicted_mos": 2.329204797744751, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d9fb6e23467c54bc99323fd93460240dbdd4b022d30f310f94564183b6702f85", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.974118000362068, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.020916304804663, "全帯域_10ms_rms_envelope_nrmse": 0.08177097061768537, "全帯域_10ms_envelope_correlation": 0.993708196239233, "全帯域_10ms_modulation_mae_db": 0.11303233586267296, "全帯域_40ms_log_rms_mae_db": 0.5735608673153647, "全帯域_40ms_rms_envelope_nrmse": 0.047649843833008014, "全帯域_40ms_envelope_correlation": 0.9978262631155506, "全帯域_40ms_modulation_mae_db": 0.08734584714834133, "300–1000 Hz_10ms_log_rms_mae_db": 1.8060395885289244, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11011214698911181, "300–1000 Hz_10ms_envelope_correlation": 0.9910446215076338, "300–1000 Hz_10ms_modulation_mae_db": 0.6234047320986763, "300–1000 Hz_40ms_log_rms_mae_db": 1.063842150136927, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06334668953730399, "300–1000 Hz_40ms_envelope_correlation": 0.9970396888652945, "300–1000 Hz_40ms_modulation_mae_db": 0.9172033597619521, "1000–3000 Hz_10ms_log_rms_mae_db": 1.787839957210094, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1391410621505066, "1000–3000 Hz_10ms_envelope_correlation": 0.9860136318102076, "1000–3000 Hz_10ms_modulation_mae_db": 0.3696190045728239, "1000–3000 Hz_40ms_log_rms_mae_db": 0.930464471432644, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08447351377444791, "1000–3000 Hz_40ms_envelope_correlation": 0.9948113076402161, "1000–3000 Hz_40ms_modulation_mae_db": 0.5598809345862598, "3000–6000 Hz_10ms_log_rms_mae_db": 2.162488341119269, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1810817404519999, "3000–6000 Hz_10ms_envelope_correlation": 0.9770944904396033, "3000–6000 Hz_10ms_modulation_mae_db": 0.9208212093563581, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3794599549503488, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11999555461882336, "3000–6000 Hz_40ms_envelope_correlation": 0.9899649589834817, "3000–6000 Hz_40ms_modulation_mae_db": 0.9720061775898969, "mr_stft_db": 5.77669852640014, "band_mae_db": 0.20621513972675473, "rms_gain_error_db": -0.12545648918460323, "output_peak": 0.4959105712890302}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0688038094020091, "全帯域_10ms_rms_envelope_nrmse": 0.07471432512151939, "全帯域_10ms_envelope_correlation": 0.9947215655979819, "全帯域_10ms_modulation_mae_db": 0.10541111890758549, "全帯域_40ms_log_rms_mae_db": 0.5656612283609896, "全帯域_40ms_rms_envelope_nrmse": 0.03931211652790762, "全帯域_40ms_envelope_correlation": 0.9985201833384867, "全帯域_40ms_modulation_mae_db": 0.14239875176126596, "300–1000 Hz_10ms_log_rms_mae_db": 1.7593167442303914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09934920446979151, "300–1000 Hz_10ms_envelope_correlation": 0.9927293061898599, "300–1000 Hz_10ms_modulation_mae_db": 0.36501798287605686, "300–1000 Hz_40ms_log_rms_mae_db": 1.0251344200009564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05299633515281616, "300–1000 Hz_40ms_envelope_correlation": 0.9979761328396272, "300–1000 Hz_40ms_modulation_mae_db": 0.6977723051473083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8367790684542684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17727609907294917, "1000–3000 Hz_10ms_envelope_correlation": 0.977127121054214, "1000–3000 Hz_10ms_modulation_mae_db": 0.5232934998600979, "1000–3000 Hz_40ms_log_rms_mae_db": 1.031694523728567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11426411804182131, "1000–3000 Hz_40ms_envelope_correlation": 0.9903586031526749, "1000–3000 Hz_40ms_modulation_mae_db": 0.4436363125169945, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1466992223558874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2007710001596627, "3000–6000 Hz_10ms_envelope_correlation": 0.9712061645719127, "3000–6000 Hz_10ms_modulation_mae_db": 0.9184724032443939, "3000–6000 Hz_40ms_log_rms_mae_db": 1.390701436760279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13844152822986602, "3000–6000 Hz_40ms_envelope_correlation": 0.9855077901657737, "3000–6000 Hz_40ms_modulation_mae_db": 1.0050371628469996, "mr_stft_db": 5.765406356041193, "band_mae_db": 0.1497184054293074, "rms_gain_error_db": -0.09468304852519141, "output_peak": 0.5214099511430544}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0368274458756919, "全帯域_10ms_rms_envelope_nrmse": 0.07823251602706827, "全帯域_10ms_envelope_correlation": 0.9941379041038401, "全帯域_10ms_modulation_mae_db": 0.30596647250575676, "全帯域_40ms_log_rms_mae_db": 0.5520815323112478, "全帯域_40ms_rms_envelope_nrmse": 0.04558045096465146, "全帯域_40ms_envelope_correlation": 0.9978440226493462, "全帯域_40ms_modulation_mae_db": 0.42501395389310254, "300–1000 Hz_10ms_log_rms_mae_db": 1.8159278807033719, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10294452934117473, "300–1000 Hz_10ms_envelope_correlation": 0.9920203179826091, "300–1000 Hz_10ms_modulation_mae_db": 0.5219734397355767, "300–1000 Hz_40ms_log_rms_mae_db": 1.047575914675342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057983920338892984, "300–1000 Hz_40ms_envelope_correlation": 0.9972922034938235, "300–1000 Hz_40ms_modulation_mae_db": 0.35155089939408063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8361803781435704, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14739050268101633, "1000–3000 Hz_10ms_envelope_correlation": 0.9844024531523562, "1000–3000 Hz_10ms_modulation_mae_db": 0.40920495585101757, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0192811834151727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09046199212094555, "1000–3000 Hz_40ms_envelope_correlation": 0.994208565020998, "1000–3000 Hz_40ms_modulation_mae_db": 0.7006443816836163, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1658292447103267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18024917528497544, "3000–6000 Hz_10ms_envelope_correlation": 0.9768748391697497, "3000–6000 Hz_10ms_modulation_mae_db": 0.49693149926628516, "3000–6000 Hz_40ms_log_rms_mae_db": 1.365299826190629, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12111149572943067, "3000–6000 Hz_40ms_envelope_correlation": 0.9889561365152211, "3000–6000 Hz_40ms_modulation_mae_db": 0.6624584482587239, "mr_stft_db": 5.769975471851526, "band_mae_db": 0.18546948594515456, "rms_gain_error_db": -0.01936422606514228, "output_peak": 0.4934397002436146}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.077621577123437, "全帯域_10ms_rms_envelope_nrmse": 0.07017553535929297, "全帯域_10ms_envelope_correlation": 0.9953420352238835, "全帯域_10ms_modulation_mae_db": 0.3438039139516369, "全帯域_40ms_log_rms_mae_db": 0.5244014942495301, "全帯域_40ms_rms_envelope_nrmse": 0.03680113079007969, "全帯域_40ms_envelope_correlation": 0.9986841318570164, "全帯域_40ms_modulation_mae_db": 0.37640295145256, "300–1000 Hz_10ms_log_rms_mae_db": 1.8115831380037044, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09490926660261621, "300–1000 Hz_10ms_envelope_correlation": 0.9932947832281386, "300–1000 Hz_10ms_modulation_mae_db": 0.5649250019917448, "300–1000 Hz_40ms_log_rms_mae_db": 1.0521784326244308, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05155582886230703, "300–1000 Hz_40ms_envelope_correlation": 0.9979552748934841, "300–1000 Hz_40ms_modulation_mae_db": 0.9200695875304717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8019186406882317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1532506831794639, "1000–3000 Hz_10ms_envelope_correlation": 0.9835336824462212, "1000–3000 Hz_10ms_modulation_mae_db": 0.46296111386964717, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9834839646142403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09950212955083954, "1000–3000 Hz_40ms_envelope_correlation": 0.9934587511802667, "1000–3000 Hz_40ms_modulation_mae_db": 0.6276068459939322, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1107815642697236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1874090106089289, "3000–6000 Hz_10ms_envelope_correlation": 0.974953336283033, "3000–6000 Hz_10ms_modulation_mae_db": 0.6340443308882809, "3000–6000 Hz_40ms_log_rms_mae_db": 1.327830891716619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12143037462199094, "3000–6000 Hz_40ms_envelope_correlation": 0.9887987265085926, "3000–6000 Hz_40ms_modulation_mae_db": 0.8824372840914565, "mr_stft_db": 5.756533614692458, "band_mae_db": 0.17895477923980987, "rms_gain_error_db": -0.08685232161249379, "output_peak": 0.5180162471911294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0097559506276204, "全帯域_10ms_rms_envelope_nrmse": 0.08854730033235693, "全帯域_10ms_envelope_correlation": 0.9924818017358014, "全帯域_10ms_modulation_mae_db": 0.3648763018319561, "全帯域_40ms_log_rms_mae_db": 0.5122425220848474, "全帯域_40ms_rms_envelope_nrmse": 0.04539186620766401, "全帯域_40ms_envelope_correlation": 0.9978858330362391, "全帯域_40ms_modulation_mae_db": 0.407878951155185, "300–1000 Hz_10ms_log_rms_mae_db": 1.7980150356739497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10847695121687445, "300–1000 Hz_10ms_envelope_correlation": 0.9911850202722895, "300–1000 Hz_10ms_modulation_mae_db": 0.31998827036745026, "300–1000 Hz_40ms_log_rms_mae_db": 1.010863052616551, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0542241569990496, "300–1000 Hz_40ms_envelope_correlation": 0.9977505248427493, "300–1000 Hz_40ms_modulation_mae_db": 0.185767862455261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8521659411688338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15976224771453723, "1000–3000 Hz_10ms_envelope_correlation": 0.98148337726295, "1000–3000 Hz_10ms_modulation_mae_db": 0.1914197220149202, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0299911164307303, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0901620850964477, "1000–3000 Hz_40ms_envelope_correlation": 0.9941672560284934, "1000–3000 Hz_40ms_modulation_mae_db": 0.5729467195007221, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1630956936410883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19767841814306777, "3000–6000 Hz_10ms_envelope_correlation": 0.9724157895119103, "3000–6000 Hz_10ms_modulation_mae_db": 0.7159368178083132, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3892800567360029, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1333194280799607, "3000–6000 Hz_40ms_envelope_correlation": 0.9864751979438658, "3000–6000 Hz_40ms_modulation_mae_db": 0.9862479537652464, "mr_stft_db": 5.7541337542394615, "band_mae_db": 0.15424482952055352, "rms_gain_error_db": -0.04417231453192981, "output_peak": 0.5146615839794062}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9811078865541488, "全帯域_10ms_rms_envelope_nrmse": 0.07277047372995607, "全帯域_10ms_envelope_correlation": 0.9949321117172589, "全帯域_10ms_modulation_mae_db": 0.11367053838968445, "全帯域_40ms_log_rms_mae_db": 0.5011042042190967, "全帯域_40ms_rms_envelope_nrmse": 0.035293669410824, "全帯域_40ms_envelope_correlation": 0.9987249358244891, "全帯域_40ms_modulation_mae_db": 0.1967688338916516, "300–1000 Hz_10ms_log_rms_mae_db": 1.766764287367348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10530045231999133, "300–1000 Hz_10ms_envelope_correlation": 0.9916671134890505, "300–1000 Hz_10ms_modulation_mae_db": 0.43624717439444644, "300–1000 Hz_40ms_log_rms_mae_db": 1.0210963491222376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05301487478380744, "300–1000 Hz_40ms_envelope_correlation": 0.9977746043177799, "300–1000 Hz_40ms_modulation_mae_db": 0.2291324533787427, "1000–3000 Hz_10ms_log_rms_mae_db": 1.871597402684066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14522765393662118, "1000–3000 Hz_10ms_envelope_correlation": 0.9847050795721219, "1000–3000 Hz_10ms_modulation_mae_db": 0.5779930013939473, "1000–3000 Hz_40ms_log_rms_mae_db": 0.99453878304295, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09131345415277085, "1000–3000 Hz_40ms_envelope_correlation": 0.993829715831432, "1000–3000 Hz_40ms_modulation_mae_db": 0.7703497005928069, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1727635682602413, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19944146642329746, "3000–6000 Hz_10ms_envelope_correlation": 0.9724247215613138, "3000–6000 Hz_10ms_modulation_mae_db": 0.8117981361395122, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3815692205626688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13481742838052926, "3000–6000 Hz_40ms_envelope_correlation": 0.9865241886224758, "3000–6000 Hz_40ms_modulation_mae_db": 0.647167994828952, "mr_stft_db": 5.75345137118011, "band_mae_db": 0.21538886188960268, "rms_gain_error_db": -0.040993879474448686, "output_peak": 0.5098733382665206}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0648848779933102, "全帯域_10ms_rms_envelope_nrmse": 0.07918263047819316, "全帯域_10ms_envelope_correlation": 0.9941380846155954, "全帯域_10ms_modulation_mae_db": 0.23408264338502172, "全帯域_40ms_log_rms_mae_db": 0.5363240272573895, "全帯域_40ms_rms_envelope_nrmse": 0.041953712153749954, "全帯域_40ms_envelope_correlation": 0.9984100308580662, "全帯域_40ms_modulation_mae_db": 0.3062503006298691, "300–1000 Hz_10ms_log_rms_mae_db": 1.8507850636021352, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11134815311636694, "300–1000 Hz_10ms_envelope_correlation": 0.9910638970688814, "300–1000 Hz_10ms_modulation_mae_db": 0.5517968208989333, "300–1000 Hz_40ms_log_rms_mae_db": 1.1174165664001567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06438460692030769, "300–1000 Hz_40ms_envelope_correlation": 0.9972373343184519, "300–1000 Hz_40ms_modulation_mae_db": 0.9323024276565196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8510364227802354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16086765677122206, "1000–3000 Hz_10ms_envelope_correlation": 0.9813873308290758, "1000–3000 Hz_10ms_modulation_mae_db": 0.6502153279523607, "1000–3000 Hz_40ms_log_rms_mae_db": 1.008990358636911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10107389267235985, "1000–3000 Hz_40ms_envelope_correlation": 0.9927555309838858, "1000–3000 Hz_40ms_modulation_mae_db": 1.054286026772812, "3000–6000 Hz_10ms_log_rms_mae_db": 2.212592936697444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1913676146611094, "3000–6000 Hz_10ms_envelope_correlation": 0.9739594697143433, "3000–6000 Hz_10ms_modulation_mae_db": 0.7896860008712601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4345273135376337, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13182831487265387, "3000–6000 Hz_40ms_envelope_correlation": 0.987038568830324, "3000–6000 Hz_40ms_modulation_mae_db": 0.6295496337692562, "mr_stft_db": 5.783630744676895, "band_mae_db": 0.21376568924137507, "rms_gain_error_db": -0.13212869019072776, "output_peak": 0.4991202835801101}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0419447161240154, "全帯域_10ms_rms_envelope_nrmse": 0.07530929520044355, "全帯域_10ms_envelope_correlation": 0.9945954327352993, "全帯域_10ms_modulation_mae_db": 0.08937419589969024, "全帯域_40ms_log_rms_mae_db": 0.541869449524915, "全帯域_40ms_rms_envelope_nrmse": 0.03896141572818532, "全帯域_40ms_envelope_correlation": 0.9984821409568474, "全帯域_40ms_modulation_mae_db": 0.4030596389566279, "300–1000 Hz_10ms_log_rms_mae_db": 1.7673319343484164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10340307135387619, "300–1000 Hz_10ms_envelope_correlation": 0.9919674938891228, "300–1000 Hz_10ms_modulation_mae_db": 0.45947488620081617, "300–1000 Hz_40ms_log_rms_mae_db": 1.0482073304852864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0549198645290708, "300–1000 Hz_40ms_envelope_correlation": 0.9975922886498465, "300–1000 Hz_40ms_modulation_mae_db": 0.3539194078126517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.844125765641948, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1587489263130705, "1000–3000 Hz_10ms_envelope_correlation": 0.9826666756711248, "1000–3000 Hz_10ms_modulation_mae_db": 0.5353171471295303, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0214470688900947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10016054383300543, "1000–3000 Hz_40ms_envelope_correlation": 0.9939435233428858, "1000–3000 Hz_40ms_modulation_mae_db": 0.774998230466866, "3000–6000 Hz_10ms_log_rms_mae_db": 2.176389597957274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18261461129552573, "3000–6000 Hz_10ms_envelope_correlation": 0.9762455772531451, "3000–6000 Hz_10ms_modulation_mae_db": 0.625480474887969, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4090950386057464, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12323782261413227, "3000–6000 Hz_40ms_envelope_correlation": 0.9884387577647157, "3000–6000 Hz_40ms_modulation_mae_db": 0.8548989237531549, "mr_stft_db": 5.771925877398533, "band_mae_db": 0.16297244825225415, "rms_gain_error_db": -0.07175001172705506, "output_peak": 0.5138744707760867}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.215831628079265, "spectral_convergence": 0.2508236331589619, "spectral_convergence_status": "ok", "amplitude_floor": 1.4412887028114478e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.441648087663989, "spectral_convergence": 0.2429615821434491, "spectral_convergence_status": "ok", "amplitude_floor": 8.082487629299873e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.672615863457168, "spectral_convergence": 0.25855820246783145, "spectral_convergence_status": "ok", "amplitude_floor": 6.705293681874596e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00036448625115270087, "output_power": 0.000373772637606411, "delta_db": 0.10926354384115106, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008505212390156939, "output_power": 0.0008376637177195084, "delta_db": -0.06615458216581284, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010532918945599765, "output_power": 0.0009965346142426193, "delta_db": -0.2405635380527281, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002198202138287696, "output_power": 0.0002125030793820009, "delta_db": -0.14702398195501865, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.872413022951283e-05, "output_power": 3.503414230532637e-05, "delta_db": -0.4349018228822496, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.441903264074617e-05, "output_power": 2.3109477082238053e-05, "delta_db": -0.2393833694635682, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20621513972675473, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12545648918460323, "pitch": null, "output_rms_dbfs": -26.05790458846018, "output_peak": 0.4959105712890302, "reference_rms_dbfs": -25.93244809927558, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.77669852640014}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019142337741905198, "log_rms_mae_db": 1.020916304804663, "rms_envelope_nrmse": 0.08177097061768537, "envelope_correlation": 0.993708196239233, "rms_gain_error_db": -0.12543292967157593, "reference_mean_power": 0.0025533227171819564, "output_mean_power": 0.002480632425404057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1748402838347296e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007147512141697323, "output_power": 0.0007108877542051384, "reference_active": true, "delta_db": -0.023538686699432213}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.175142428480134e-05, "output_power": 8.40765944565562e-05, "reference_active": true, "delta_db": 0.12179784432331112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.431203959050675e-05, "output_power": 1.3687542518592885e-05, "reference_active": true, "delta_db": -0.19376047656527556}], "active_band_count": 3, "active_band_mae_db": 0.11303233586267296}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016640454830055276, "log_rms_mae_db": 0.5735608673153647, "rms_envelope_nrmse": 0.047649843833008014, "envelope_correlation": 0.9978262631155506, "rms_gain_error_db": -0.12519663898119013, "reference_mean_power": 0.002564981061322552, "output_mean_power": 0.0024920944553413313}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0626539543337418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006440432662381922, "output_power": 0.0006408101652742063, "reference_active": true, "delta_db": -0.021856514326514247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1296244809426825e-05, "output_power": 4.1986269235682706e-05, "reference_active": true, "delta_db": 0.07196724586214806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.585289215752233e-07, "output_power": 8.924344118082968e-07, "reference_active": true, "delta_db": 0.16821378125636166}], "active_band_count": 3, "active_band_mae_db": 0.08734584714834133}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016776043588712122, "log_rms_mae_db": 1.8060395885289244, "rms_envelope_nrmse": 0.11011214698911181, "envelope_correlation": 0.9910446215076338, "rms_gain_error_db": -0.1774722500918834, "reference_mean_power": 0.0015150798695545982, "output_mean_power": 0.001454414859041397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.503421266374213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005710315291680691, "output_power": 0.0005494884495194655, "reference_active": true, "delta_db": -0.16701520419735094}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462070521392158e-05, "output_power": 5.972034150889734e-05, "reference_active": true, "delta_db": -0.3424941065860199}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.82379092220564e-06, "output_power": 7.181391814427604e-06, "reference_active": true, "delta_db": -1.360704885512658}], "active_band_count": 3, "active_band_mae_db": 0.6234047320986763}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015135634256581334, "log_rms_mae_db": 1.063842150136927, "rms_envelope_nrmse": 0.06334668953730399, "envelope_correlation": 0.9970396888652945, "rms_gain_error_db": -0.1775001343533657, "reference_mean_power": 0.0015225180811750088, "output_mean_power": 0.0014615458547010325}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.758019890177841e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005253930575206644, "output_power": 0.0005070861481983323, "reference_active": true, "delta_db": -0.15402582454687547}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0513708673138778e-05, "output_power": 2.855324977148335e-05, "reference_active": true, "delta_db": -0.28839451159858465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.334657094067365e-07, "output_power": 4.897435921975376e-07, "reference_active": true, "delta_db": -2.3091897431403963}], "active_band_count": 3, "active_band_mae_db": 0.9172033597619521}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.526806573516121e-05, "log_rms_mae_db": 1.787839957210094, "rms_envelope_nrmse": 0.1391410621505066, "envelope_correlation": 0.9860136318102076, "rms_gain_error_db": -0.21753786898384866, "reference_mean_power": 0.00023682053730462446, "output_mean_power": 0.00022525040225511642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1631666261710027e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014015200472863968, "output_power": 0.00013329700315268522, "reference_active": true, "delta_db": -0.21778928765275157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3136457414027934e-05, "output_power": 1.3831925157995333e-05, "reference_active": true, "delta_db": 0.22404368126775012}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.230915915111415e-06, "output_power": 2.770915302866003e-06, "reference_active": true, "delta_db": -0.6670240447979701}], "active_band_count": 3, "active_band_mae_db": 0.3696190045728239}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.301677625713533e-05, "log_rms_mae_db": 0.930464471432644, "rms_envelope_nrmse": 0.08447351377444791, "envelope_correlation": 0.9948113076402161, "rms_gain_error_db": -0.2177881350704032, "reference_mean_power": 0.0002379882918616051, "output_mean_power": 0.0002263480609575442}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0209734833253925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001318054605596598, "output_power": 0.00012595813915452936, "reference_active": true, "delta_db": -0.1970716703554546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.132933188997116e-06, "output_power": 6.851166111249584e-06, "reference_active": true, "delta_db": 0.4809626382428849}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2319303983633163e-07, "output_power": 1.7722288319357635e-07, "reference_active": true, "delta_db": -1.0016084951604398}], "active_band_count": 3, "active_band_mae_db": 0.5598809345862598}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.803319177623983e-05, "log_rms_mae_db": 2.162488341119269, "rms_envelope_nrmse": 0.1810817404519999, "envelope_correlation": 0.9770944904396033, "rms_gain_error_db": -0.3701920073241476, "reference_mean_power": 3.0517013147816785e-05, "output_mean_power": 2.8023528936357935e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.826326538605559e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1311662288856248e-05, "output_power": 1.0306480311747473e-05, "reference_active": true, "delta_db": -0.40416052754553927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5475498743040963e-06, "output_power": 2.3093414640628474e-06, "reference_active": true, "delta_db": -0.4263454174528874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.732573657220471e-07, "output_power": 8.944280527997345e-07, "reference_active": true, "delta_db": 1.9319576830706475}], "active_band_count": 3, "active_band_mae_db": 0.9208212093563581}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.459177046887288e-05, "log_rms_mae_db": 1.3794599549503488, "rms_envelope_nrmse": 0.11999555461882336, "envelope_correlation": 0.9899649589834817, "rms_gain_error_db": -0.3701941374445095, "reference_mean_power": 3.0668913834731335e-05, "output_mean_power": 2.8163004308095487e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6363060360497894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1094596811060155e-05, "output_power": 1.0172620293676604e-05, "reference_active": true, "delta_db": -0.37678689959140393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1901156553217825e-06, "output_power": 1.30131138939407e-06, "reference_active": true, "delta_db": 0.3879206271559519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.41782955712918e-08, "output_power": 7.250034230818954e-08, "reference_active": true, "delta_db": 2.151311006022335}], "active_band_count": 3, "active_band_mae_db": 0.9720061775898969}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8771005282439883, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b94857e2fe80343b1ddf90ffab82e2cd935cceb8f32728e45acf3db3eda18ed9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826649189747839, "standard_deviation": 0.002747606467104103, "minimum": 0.977127121054214, "maximum": 0.9860136318102076, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8352054470964059, "standard_deviation": 0.027508836378987284, "minimum": 1.787839957210094, "maximum": 1.871597402684066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46500297158054316, "standard_deviation": 0.1426663020115512, "minimum": 0.1914197220149202, "maximum": 0.6502153279523607, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15520810397742335, "standard_deviation": 0.01180199859374654, "minimum": 0.1391410621505066, "maximum": 0.17727609907294917, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934416566476065, "standard_deviation": 0.0013819926797579995, "minimum": 0.9903586031526749, "maximum": 0.9948113076402161, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0024864337739137, "standard_deviation": 0.033604660541549355, "minimum": 0.930464471432644, "maximum": 1.031694523728567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6880436440142512, "standard_deviation": 0.1858141086977957, "minimum": 0.4436363125169945, "maximum": 1.054286026772812, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09642646615532977, "standard_deviation": 0.009312179600593044, "minimum": 0.08447351377444791, "maximum": 0.11426411804182131, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9743967985631264, "standard_deviation": 0.002246611096638485, "minimum": 0.9712061645719127, "maximum": 0.9770944904396033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1638300211264068, "standard_deviation": 0.028608970740906087, "minimum": 2.1107815642697236, "maximum": 2.212592936697444, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7391463590577965, "standard_deviation": 0.1493649452662218, "minimum": 0.49693149926628516, "maximum": 0.9208212093563581, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19007662962857091, "standard_deviation": 0.008466812178250778, "minimum": 0.18024917528497544, "maximum": 0.2007710001596627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9877130406668063, "standard_deviation": 0.0015392701548431022, "minimum": 0.9855077901657737, "maximum": 0.9899649589834817, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.384720467382491, "standard_deviation": 0.03112782854003444, "minimum": 1.327830891716619, "maximum": 1.4345273135376337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8299754473629608, "standard_deviation": 0.16047468391713235, "minimum": 0.6295496337692562, "maximum": 1.0050371628469996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1280227433934234, "standard_deviation": 0.007327170193123473, "minimum": 0.11999555461882336, "maximum": 0.13844152822986602, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918715692034482, "standard_deviation": 0.000814327679320361, "minimum": 0.9910446215076338, "maximum": 0.9932947832281386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7969704590572801, "standard_deviation": 0.031078066056772734, "minimum": 1.7593167442303914, "maximum": 1.8507850636021352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4803535385704626, "standard_deviation": 0.10409675053023956, "minimum": 0.31998827036745026, "maximum": 0.6234047320986763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1044804719262254, "standard_deviation": 0.005567543856830544, "minimum": 0.09490926660261621, "maximum": 0.11134815311636694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975772565276321, "standard_deviation": 0.00034989295366252973, "minimum": 0.9970396888652945, "maximum": 0.9979761328396272, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0482892770077359, "standard_deviation": 0.03319302882049299, "minimum": 1.010863052616551, "maximum": 1.1174165664001567, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5734647878921235, "standard_deviation": 0.3270708778337267, "minimum": 0.185767862455261, "maximum": 0.9323024276565196, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056553284640444464, "standard_deviation": 0.0048955719394615975, "minimum": 0.05155582886230703, "maximum": 0.06438460692030769, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18334120490560152, "standard_deviation": 0.02643500580275849, "minimum": 0.1497184054293074, "maximum": 0.21538886188960268, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.766469464560039, "standard_deviation": 0.011101477290661809, "minimum": 5.75345137118011, "maximum": 5.783630744676895, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5082882683086191, "standard_deviation": 0.010683021857378244, "minimum": 0.4934397002436146, "maximum": 0.5214099511430544, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.076925122663949, "standard_deviation": 0.04056821395069689, "minimum": -0.13212869019072776, "maximum": -0.01936422606514228, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9942571414961117, "standard_deviation": 0.0008819033122421375, "minimum": 0.9924818017358014, "maximum": 0.9953420352238835, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0377328210631118, "standard_deviation": 0.03291590332683344, "minimum": 0.9811078865541488, "maximum": 1.077621577123437, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.20877719009175058, "standard_deviation": 0.11700991620270423, "minimum": 0.08937419589969024, "maximum": 0.3648763018319561, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07758788085831447, "standard_deviation": 0.0057590357377877805, "minimum": 0.07017553535929297, "maximum": 0.08854730033235693, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982971927045052, "standard_deviation": 0.00038279120857727297, "minimum": 0.9978262631155506, "maximum": 0.9987249358244891, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5384056656654226, "standard_deviation": 0.025225101363513677, "minimum": 0.5011042042190967, "maximum": 0.5735608673153647, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.29313990361107545, "standard_deviation": 0.13315096247098976, "minimum": 0.08734584714834133, "maximum": 0.42501395389310254, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04136802570200876, "standard_deviation": 0.004496235794625888, "minimum": 0.035293669410824, "maximum": 0.047649843833008014, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0011", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5505495071411133, "predicted_mos": 2.5505495071411133, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6138353a01d905a6ba1096ff49d8c4cd6f55647b148266f36bfc1a0af2eb01fb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.522523699793965, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0148254316436254, "全帯域_10ms_rms_envelope_nrmse": 0.07179544306080464, "全帯域_10ms_envelope_correlation": 0.9950683481597159, "全帯域_10ms_modulation_mae_db": 0.09061340951756612, "全帯域_40ms_log_rms_mae_db": 0.5243153701026344, "全帯域_40ms_rms_envelope_nrmse": 0.03746862009626833, "全帯域_40ms_envelope_correlation": 0.9985590965590017, "全帯域_40ms_modulation_mae_db": 0.1977848951097546, "300–1000 Hz_10ms_log_rms_mae_db": 1.9891985878095395, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09577448016530922, "300–1000 Hz_10ms_envelope_correlation": 0.9931690287392598, "300–1000 Hz_10ms_modulation_mae_db": 0.2173874851404531, "300–1000 Hz_40ms_log_rms_mae_db": 1.218439911292993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05545270134726016, "300–1000 Hz_40ms_envelope_correlation": 0.9976025441621656, "300–1000 Hz_40ms_modulation_mae_db": 0.6047304324879264, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1361012931092787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17498489211340187, "1000–3000 Hz_10ms_envelope_correlation": 0.9775783965377676, "1000–3000 Hz_10ms_modulation_mae_db": 0.40201458375918536, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1979100038683346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1093636548486379, "1000–3000 Hz_40ms_envelope_correlation": 0.9908972328038982, "1000–3000 Hz_40ms_modulation_mae_db": 0.45721355998103164, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8537204846711943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24859261359029736, "3000–6000 Hz_10ms_envelope_correlation": 0.958106756319067, "3000–6000 Hz_10ms_modulation_mae_db": 0.9873729634315879, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8284542439670182, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15673701922492303, "3000–6000 Hz_40ms_envelope_correlation": 0.9826279711686845, "3000–6000 Hz_40ms_modulation_mae_db": 0.842400034315844, "mr_stft_db": 6.171712355292349, "band_mae_db": 0.10771259830174891, "rms_gain_error_db": -0.035876882458188486, "output_peak": 0.4964136130725766}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0050336352935116, "全帯域_10ms_rms_envelope_nrmse": 0.07854861037777096, "全帯域_10ms_envelope_correlation": 0.9942232317337297, "全帯域_10ms_modulation_mae_db": 0.08587675776516623, "全帯域_40ms_log_rms_mae_db": 0.5347660030446959, "全帯域_40ms_rms_envelope_nrmse": 0.0455745173848626, "全帯域_40ms_envelope_correlation": 0.9980417354909377, "全帯域_40ms_modulation_mae_db": 0.12129554456965397, "300–1000 Hz_10ms_log_rms_mae_db": 2.0094032881397976, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11097838443298635, "300–1000 Hz_10ms_envelope_correlation": 0.9909275632812632, "300–1000 Hz_10ms_modulation_mae_db": 0.2677908066301617, "300–1000 Hz_40ms_log_rms_mae_db": 1.1927432770292472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06502617306789168, "300–1000 Hz_40ms_envelope_correlation": 0.9969241179323962, "300–1000 Hz_40ms_modulation_mae_db": 0.4226801681701535, "1000–3000 Hz_10ms_log_rms_mae_db": 2.179693028625686, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16936975214863612, "1000–3000 Hz_10ms_envelope_correlation": 0.979133843996083, "1000–3000 Hz_10ms_modulation_mae_db": 0.23482816728881695, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2001310422526352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10755694344838997, "1000–3000 Hz_40ms_envelope_correlation": 0.9914368749365339, "1000–3000 Hz_40ms_modulation_mae_db": 0.4604143132234016, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9528586733135858, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26625337456614834, "3000–6000 Hz_10ms_envelope_correlation": 0.9509026132995011, "3000–6000 Hz_10ms_modulation_mae_db": 0.8502183740195347, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9319708147173322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18594682913724678, "3000–6000 Hz_40ms_envelope_correlation": 0.9745348763228795, "3000–6000 Hz_40ms_modulation_mae_db": 0.7459345113239944, "mr_stft_db": 6.174497365819978, "band_mae_db": 0.1348388453460354, "rms_gain_error_db": -0.12939851635484217, "output_peak": 0.515904544825083}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9714449689268478, "全帯域_10ms_rms_envelope_nrmse": 0.07749992583504339, "全帯域_10ms_envelope_correlation": 0.9942738440131732, "全帯域_10ms_modulation_mae_db": 0.2339830877261427, "全帯域_40ms_log_rms_mae_db": 0.47941269591878477, "全帯域_40ms_rms_envelope_nrmse": 0.04312070638784033, "全帯域_40ms_envelope_correlation": 0.9980752882503405, "全帯域_40ms_modulation_mae_db": 0.43847150541156193, "300–1000 Hz_10ms_log_rms_mae_db": 1.9662060065487863, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09880303593420771, "300–1000 Hz_10ms_envelope_correlation": 0.9926764481431213, "300–1000 Hz_10ms_modulation_mae_db": 0.4544835066926076, "300–1000 Hz_40ms_log_rms_mae_db": 1.1337867427456643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051310282269811845, "300–1000 Hz_40ms_envelope_correlation": 0.9978733838697225, "300–1000 Hz_40ms_modulation_mae_db": 0.3302588271380071, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1193602962448463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16711076098037805, "1000–3000 Hz_10ms_envelope_correlation": 0.9797462907423617, "1000–3000 Hz_10ms_modulation_mae_db": 0.4492480945396811, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1413570018354449, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09370592732597702, "1000–3000 Hz_40ms_envelope_correlation": 0.9937633880607472, "1000–3000 Hz_40ms_modulation_mae_db": 0.4769795751614873, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9122050850998535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23585674922888086, "3000–6000 Hz_10ms_envelope_correlation": 0.9625068853899363, "3000–6000 Hz_10ms_modulation_mae_db": 0.7387907096741229, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8639359354185048, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1559075472984247, "3000–6000 Hz_40ms_envelope_correlation": 0.9829153212231827, "3000–6000 Hz_40ms_modulation_mae_db": 0.7831650644638546, "mr_stft_db": 6.205080788942094, "band_mae_db": 0.1517821482791912, "rms_gain_error_db": 0.0037236862948702427, "output_peak": 0.4941763637903457}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0408240731608611, "全帯域_10ms_rms_envelope_nrmse": 0.07334807305710443, "全帯域_10ms_envelope_correlation": 0.9948628120422517, "全帯域_10ms_modulation_mae_db": 0.08209534202497008, "全帯域_40ms_log_rms_mae_db": 0.5128329832748754, "全帯域_40ms_rms_envelope_nrmse": 0.0351414318496556, "全帯域_40ms_envelope_correlation": 0.9987547305667381, "全帯域_40ms_modulation_mae_db": 0.20865790394728342, "300–1000 Hz_10ms_log_rms_mae_db": 1.913214565248369, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09279851670980417, "300–1000 Hz_10ms_envelope_correlation": 0.9936032765614224, "300–1000 Hz_10ms_modulation_mae_db": 0.2639159959838515, "300–1000 Hz_40ms_log_rms_mae_db": 1.1017787891460449, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04756635792442484, "300–1000 Hz_40ms_envelope_correlation": 0.9983131124328909, "300–1000 Hz_40ms_modulation_mae_db": 0.6814838527731933, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1590577734326493, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16162231880144057, "1000–3000 Hz_10ms_envelope_correlation": 0.9809206246241241, "1000–3000 Hz_10ms_modulation_mae_db": 0.40887890368992713, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1866121987103955, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09597133149546773, "1000–3000 Hz_40ms_envelope_correlation": 0.993047515180844, "1000–3000 Hz_40ms_modulation_mae_db": 0.5951828258126609, "3000–6000 Hz_10ms_log_rms_mae_db": 2.948929190895949, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26386407061048534, "3000–6000 Hz_10ms_envelope_correlation": 0.9525863888327433, "3000–6000 Hz_10ms_modulation_mae_db": 0.8975113457179832, "3000–6000 Hz_40ms_log_rms_mae_db": 1.869465007564445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17495574344208337, "3000–6000 Hz_40ms_envelope_correlation": 0.9780092932567671, "3000–6000 Hz_40ms_modulation_mae_db": 0.5915086197273421, "mr_stft_db": 6.18435165617565, "band_mae_db": 0.11282131283006143, "rms_gain_error_db": -0.05711181409462706, "output_peak": 0.5309947233072053}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0234334707355528, "全帯域_10ms_rms_envelope_nrmse": 0.07424709394256743, "全帯域_10ms_envelope_correlation": 0.9947871101449653, "全帯域_10ms_modulation_mae_db": 0.13335066755085648, "全帯域_40ms_log_rms_mae_db": 0.5672479334385808, "全帯域_40ms_rms_envelope_nrmse": 0.04354733879731156, "全帯域_40ms_envelope_correlation": 0.9981430918323709, "全帯域_40ms_modulation_mae_db": 0.40020675003767114, "300–1000 Hz_10ms_log_rms_mae_db": 1.918008693811842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09529772009955909, "300–1000 Hz_10ms_envelope_correlation": 0.9933799630086667, "300–1000 Hz_10ms_modulation_mae_db": 0.5774613517342414, "300–1000 Hz_40ms_log_rms_mae_db": 1.1605147297516893, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0583508378539964, "300–1000 Hz_40ms_envelope_correlation": 0.997520218627777, "300–1000 Hz_40ms_modulation_mae_db": 0.7474956865786561, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1180024170121197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15816232625647078, "1000–3000 Hz_10ms_envelope_correlation": 0.9818250122869501, "1000–3000 Hz_10ms_modulation_mae_db": 0.5545990601928743, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1450373336197766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08729885032827298, "1000–3000 Hz_40ms_envelope_correlation": 0.9944795520316251, "1000–3000 Hz_40ms_modulation_mae_db": 0.4833425203701988, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9501408045690782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23835441295588228, "3000–6000 Hz_10ms_envelope_correlation": 0.9611791868206294, "3000–6000 Hz_10ms_modulation_mae_db": 0.6440324207760376, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8832720860262229, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15411279069106978, "3000–6000 Hz_40ms_envelope_correlation": 0.9829512555749185, "3000–6000 Hz_40ms_modulation_mae_db": 0.8161000449406162, "mr_stft_db": 6.1803130403473, "band_mae_db": 0.1453475442778123, "rms_gain_error_db": -0.0889963699734289, "output_peak": 0.5163366719487298}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9724513656513771, "全帯域_10ms_rms_envelope_nrmse": 0.0722322582149747, "全帯域_10ms_envelope_correlation": 0.995264491187733, "全帯域_10ms_modulation_mae_db": 0.12804377964907984, "全帯域_40ms_log_rms_mae_db": 0.5011587178847948, "全帯域_40ms_rms_envelope_nrmse": 0.04157864468449356, "全帯域_40ms_envelope_correlation": 0.9985526390926729, "全帯域_40ms_modulation_mae_db": 0.18292001132855618, "300–1000 Hz_10ms_log_rms_mae_db": 2.026022861897501, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10197131074246288, "300–1000 Hz_10ms_envelope_correlation": 0.9926419811338222, "300–1000 Hz_10ms_modulation_mae_db": 0.46959584684383415, "300–1000 Hz_40ms_log_rms_mae_db": 1.220379235839265, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06207609349541829, "300–1000 Hz_40ms_envelope_correlation": 0.9975029188652413, "300–1000 Hz_40ms_modulation_mae_db": 0.7103787922655233, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2351538898514707, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1755923894676218, "1000–3000 Hz_10ms_envelope_correlation": 0.9777666263267193, "1000–3000 Hz_10ms_modulation_mae_db": 0.5941735012815889, "1000–3000 Hz_40ms_log_rms_mae_db": 1.225204339678418, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10574598201815773, "1000–3000 Hz_40ms_envelope_correlation": 0.9921771089282373, "1000–3000 Hz_40ms_modulation_mae_db": 0.36505713307287757, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9882723548154213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2769482195710147, "3000–6000 Hz_10ms_envelope_correlation": 0.945880837362279, "3000–6000 Hz_10ms_modulation_mae_db": 0.9780242367095964, "3000–6000 Hz_40ms_log_rms_mae_db": 1.932487972230531, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19628210630519952, "3000–6000 Hz_40ms_envelope_correlation": 0.9708511556218264, "3000–6000 Hz_40ms_modulation_mae_db": 1.1394440967354715, "mr_stft_db": 6.173661621192175, "band_mae_db": 0.15134415467769138, "rms_gain_error_db": -0.1658293255189664, "output_peak": 0.49829463944126473}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0393429928412117, "全帯域_10ms_rms_envelope_nrmse": 0.07239439677477952, "全帯域_10ms_envelope_correlation": 0.9949879239832418, "全帯域_10ms_modulation_mae_db": 0.20234848251229542, "全帯域_40ms_log_rms_mae_db": 0.5275482241719268, "全帯域_40ms_rms_envelope_nrmse": 0.03555821653524225, "全帯域_40ms_envelope_correlation": 0.9987137569779191, "全帯域_40ms_modulation_mae_db": 0.23803608462875733, "300–1000 Hz_10ms_log_rms_mae_db": 1.9203950633655682, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0995439634431092, "300–1000 Hz_10ms_envelope_correlation": 0.9926340777741935, "300–1000 Hz_10ms_modulation_mae_db": 0.6292199706602585, "300–1000 Hz_40ms_log_rms_mae_db": 1.1106468550114792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05343633939155875, "300–1000 Hz_40ms_envelope_correlation": 0.997853917648649, "300–1000 Hz_40ms_modulation_mae_db": 0.6208344090503224, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1258876864482503, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15495638655544502, "1000–3000 Hz_10ms_envelope_correlation": 0.982469790059328, "1000–3000 Hz_10ms_modulation_mae_db": 0.6529919985794146, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1114647468231942, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09783665555546066, "1000–3000 Hz_40ms_envelope_correlation": 0.9926455911809522, "1000–3000 Hz_40ms_modulation_mae_db": 1.03205613611186, "3000–6000 Hz_10ms_log_rms_mae_db": 2.917976734216116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24971219288156765, "3000–6000 Hz_10ms_envelope_correlation": 0.9557613901227295, "3000–6000 Hz_10ms_modulation_mae_db": 1.6245399733938843, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8481998950995318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15239218394604812, "3000–6000 Hz_40ms_envelope_correlation": 0.9828491018619143, "3000–6000 Hz_40ms_modulation_mae_db": 1.0661929906235643, "mr_stft_db": 6.182558409001321, "band_mae_db": 0.11996908896623537, "rms_gain_error_db": -0.047756682340743234, "output_peak": 0.5158221216059544}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0142403579459667, "全帯域_10ms_rms_envelope_nrmse": 0.0703604791835686, "全帯域_10ms_envelope_correlation": 0.9953306288232184, "全帯域_10ms_modulation_mae_db": 0.26317173396549, "全帯域_40ms_log_rms_mae_db": 0.49081594742940854, "全帯域_40ms_rms_envelope_nrmse": 0.037769202703556375, "全帯域_40ms_envelope_correlation": 0.998633297806454, "全帯域_40ms_modulation_mae_db": 0.20446813673591788, "300–1000 Hz_10ms_log_rms_mae_db": 1.8711477724564543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09305090570108746, "300–1000 Hz_10ms_envelope_correlation": 0.99368327917014, "300–1000 Hz_10ms_modulation_mae_db": 0.38887975758388044, "300–1000 Hz_40ms_log_rms_mae_db": 1.092016170354704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05173466019962774, "300–1000 Hz_40ms_envelope_correlation": 0.9981021043320929, "300–1000 Hz_40ms_modulation_mae_db": 0.7244781966368373, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2039214371989004, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16643692741513214, "1000–3000 Hz_10ms_envelope_correlation": 0.9799438367294704, "1000–3000 Hz_10ms_modulation_mae_db": 0.3350772460370246, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2452173170338028, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10054886364234947, "1000–3000 Hz_40ms_envelope_correlation": 0.9927245825058294, "1000–3000 Hz_40ms_modulation_mae_db": 0.5680581258497969, "3000–6000 Hz_10ms_log_rms_mae_db": 2.923689927654318, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2472776723688011, "3000–6000 Hz_10ms_envelope_correlation": 0.9569358000546162, "3000–6000 Hz_10ms_modulation_mae_db": 1.0597813315335642, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8827252046970415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1701713112334739, "3000–6000 Hz_40ms_envelope_correlation": 0.9783836167680063, "3000–6000 Hz_40ms_modulation_mae_db": 0.6778717790362565, "mr_stft_db": 6.175382516195477, "band_mae_db": 0.17749642642320995, "rms_gain_error_db": -0.0950075576714944, "output_peak": 0.501455305527671}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.718866927222408, "spectral_convergence": 0.25199484456231036, "spectral_convergence_status": "ok", "amplitude_floor": 1.4412887028114478e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.894893888837155, "spectral_convergence": 0.2523778884085933, "spectral_convergence_status": "ok", "amplitude_floor": 8.082487629299873e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.901376249817481, "spectral_convergence": 0.2686447805945583, "spectral_convergence_status": "ok", "amplitude_floor": 6.705293681874596e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00036448625115270087, "output_power": 0.00038059129665518376, "delta_db": 0.18777701770064728, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008505212390156939, "output_power": 0.0008284522730003361, "delta_db": -0.11417669341232789, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010532918945599765, "output_power": 0.0010421174066497097, "delta_db": -0.04632092108060531, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002198202138287696, "output_power": 0.00021466461227063834, "delta_db": -0.10307169639337904, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.872413022951283e-05, "output_power": 3.9250752310391967e-05, "delta_db": 0.05866313196982499, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.441903264074617e-05, "output_power": 2.5197361591518835e-05, "delta_db": 0.13626612925370893, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10771259830174891, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.035876882458188486, "pitch": null, "output_rms_dbfs": -25.968324981733765, "output_peak": 0.4964136130725766, "reference_rms_dbfs": -25.93244809927558, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.171712355292349}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019142337741905198, "log_rms_mae_db": 1.0148254316436254, "rms_envelope_nrmse": 0.07179544306080464, "envelope_correlation": 0.9950683481597159, "rms_gain_error_db": -0.03594586533243159, "reference_mean_power": 0.0025533227171819564, "output_mean_power": 0.002532276488391273}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1748402838347296e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007147512141697323, "output_power": 0.0007213357690584386, "reference_active": true, "delta_db": 0.03982566413908487}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.175142428480134e-05, "output_power": 8.008024981124907e-05, "reference_active": true, "delta_db": -0.08969908477779105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.431203959050675e-05, "output_power": 1.3850644450280463e-05, "reference_active": true, "delta_db": -0.14231547963582242}], "active_band_count": 3, "active_band_mae_db": 0.09061340951756612}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016640454830055276, "log_rms_mae_db": 0.5243153701026344, "rms_envelope_nrmse": 0.03746862009626833, "envelope_correlation": 0.9985590965590017, "rms_gain_error_db": -0.03595836218412386, "reference_mean_power": 0.002564981061322552, "output_mean_power": 0.002543831416599394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0626539543337418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006440432662381922, "output_power": 0.0006529286822595348, "reference_active": true, "delta_db": 0.05950703136797858}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1296244809426825e-05, "output_power": 4.087838117654778e-05, "reference_active": true, "delta_db": -0.04416872775281352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.585289215752233e-07, "output_power": 7.669852946926284e-07, "reference_active": true, "delta_db": -0.4896789262084717}], "active_band_count": 3, "active_band_mae_db": 0.1977848951097546}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016776043588712122, "log_rms_mae_db": 1.9891985878095395, "rms_envelope_nrmse": 0.09577448016530922, "envelope_correlation": 0.9931690287392598, "rms_gain_error_db": -0.09630816919916622, "reference_mean_power": 0.0015150798695545982, "output_mean_power": 0.0014818515916560399}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.503421266374213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005710315291680691, "output_power": 0.0005667144210288194, "reference_active": true, "delta_db": -0.032958240642700413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462070521392158e-05, "output_power": 5.7926528385595695e-05, "reference_active": true, "delta_db": -0.47494191788857854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.82379092220564e-06, "output_power": 9.502828274457392e-06, "reference_active": true, "delta_db": -0.1442622968900804}], "active_band_count": 3, "active_band_mae_db": 0.2173874851404531}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015135634256581334, "log_rms_mae_db": 1.218439911292993, "rms_envelope_nrmse": 0.05545270134726016, "envelope_correlation": 0.9976025441621656, "rms_gain_error_db": -0.09640488457865695, "reference_mean_power": 0.0015225180811750088, "output_mean_power": 0.0014890935088517252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.758019890177841e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005253930575206644, "output_power": 0.0005254259984043656, "reference_active": true, "delta_db": 0.0002722836808962447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0513708673138778e-05, "output_power": 2.765887368440849e-05, "reference_active": true, "delta_db": -0.42660504592401005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.334657094067365e-07, "output_power": 6.055588282108219e-07, "reference_active": true, "delta_db": -1.3873139678588728}], "active_band_count": 3, "active_band_mae_db": 0.6047304324879264}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.526806573516121e-05, "log_rms_mae_db": 2.1361012931092787, "rms_envelope_nrmse": 0.17498489211340187, "envelope_correlation": 0.9775783965377676, "rms_gain_error_db": -0.16641261410574495, "reference_mean_power": 0.00023682053730462446, "output_mean_power": 0.00022791772470634147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1631666261710027e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014015200472863968, "output_power": 0.00012798010558586162, "reference_active": true, "delta_db": -0.39456850137652977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3136457414027934e-05, "output_power": 1.403783970936593e-05, "reference_active": true, "delta_db": 0.28822016868703254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.230915915111415e-06, "output_power": 3.6446108499955408e-06, "reference_active": true, "delta_db": 0.5232550812139938}], "active_band_count": 3, "active_band_mae_db": 0.40201458375918536}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.301677625713533e-05, "log_rms_mae_db": 1.1979100038683346, "rms_envelope_nrmse": 0.1093636548486379, "envelope_correlation": 0.9908972328038982, "rms_gain_error_db": -0.16665825065247897, "reference_mean_power": 0.0002379882918616051, "output_mean_power": 0.00022902862557315287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0209734833253925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001318054605596598, "output_power": 0.0001207299844084769, "reference_active": true, "delta_db": -0.3811825844204659}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.132933188997116e-06, "output_power": 7.366712578461049e-06, "reference_active": true, "delta_db": 0.7960549204891563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2319303983633163e-07, "output_power": 2.1342256233958574e-07, "reference_active": true, "delta_db": -0.19440317503347262}], "active_band_count": 3, "active_band_mae_db": 0.45721355998103164}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.803319177623983e-05, "log_rms_mae_db": 2.8537204846711943, "rms_envelope_nrmse": 0.24859261359029736, "envelope_correlation": 0.958106756319067, "rms_gain_error_db": 0.25252667765521547, "reference_mean_power": 3.0517013147816785e-05, "output_mean_power": 3.234407181146197e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.826326538605559e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1311662288856248e-05, "output_power": 1.1631660439094671e-05, "reference_active": true, "delta_db": 0.12115284693451886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5475498743040963e-06, "output_power": 2.3219183694061627e-06, "reference_active": true, "delta_db": -0.4027574771787162}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.732573657220471e-07, "output_power": 1.0050102711821324e-06, "reference_active": true, "delta_db": 2.4382085661815283}], "active_band_count": 3, "active_band_mae_db": 0.9873729634315879}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.459177046887288e-05, "log_rms_mae_db": 1.8284542439670182, "rms_envelope_nrmse": 0.15673701922492303, "envelope_correlation": 0.9826279711686845, "rms_gain_error_db": 0.2524734051985365, "reference_mean_power": 3.0668913834731335e-05, "output_mean_power": 3.2504668098929084e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6363060360497894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1094596811060155e-05, "output_power": 1.1529341887565795e-05, "reference_active": true, "delta_db": 0.1669299365891968}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1901156553217825e-06, "output_power": 1.295428456396272e-06, "reference_active": true, "delta_db": 0.3682426473723647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.41782955712918e-08, "output_power": 6.988946377604392e-08, "reference_active": true, "delta_db": 1.9920275189859704}], "active_band_count": 3, "active_band_mae_db": 0.842400034315844}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8518565118650924, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "07279739a037ba3408464524f7051b0670e54acbbdd8ebf9cc415da7edb89c98", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9799230526628505, "standard_deviation": 0.0017688004635083463, "minimum": 0.9775783965377676, "maximum": 0.982469790059328, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1596472277404004, "standard_deviation": 0.04328491656071292, "minimum": 2.1180024170121197, "maximum": 2.2351538898514707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4539764444210641, "standard_deviation": 0.13954337960265317, "minimum": 0.23482816728881695, "maximum": 0.6529919985794146, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1660294692173158, "standard_deviation": 0.007445706527673752, "minimum": 0.15495638655544502, "maximum": 0.1755923894676218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9926464807035833, "standard_deviation": 0.0011660608941839454, "minimum": 0.9908972328038982, "maximum": 0.9944795520316251, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1816167479777502, "standard_deviation": 0.04544748668637441, "minimum": 1.1114647468231942, "maximum": 1.2452173170338028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5547880236979144, "standard_deviation": 0.2053485835776091, "minimum": 0.36505713307287757, "maximum": 1.03205613611186, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09975352608283919, "standard_deviation": 0.007554201998484668, "minimum": 0.08729885032827298, "maximum": 0.1093636548486379, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9554824822751877, "standard_deviation": 0.005508766523911301, "minimum": 0.945880837362279, "maximum": 0.9625068853899363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.9309741569044396, "standard_deviation": 0.039630724179388815, "minimum": 2.8537204846711943, "maximum": 2.9882723548154213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9725339194070388, "standard_deviation": 0.29667513451473443, "minimum": 0.6440324207760376, "maximum": 1.6245399733938843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25335741322163474, "standard_deviation": 0.01433048332129455, "minimum": 0.23585674922888086, "maximum": 0.2769482195710147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9791403239747725, "standard_deviation": 0.004574226188915188, "minimum": 0.9708511556218264, "maximum": 0.9829512555749185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8800638949650783, "standard_deviation": 0.036861333421962586, "minimum": 1.8284542439670182, "maximum": 1.932487972230531, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.832827142645868, "standard_deviation": 0.18559073914413438, "minimum": 0.5915086197273421, "maximum": 1.1394440967354715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16831319140980866, "standard_deviation": 0.01640478638435848, "minimum": 0.15239218394604812, "maximum": 0.19628210630519952, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9928394522264861, "standard_deviation": 0.0008831200595497695, "minimum": 0.9909275632812632, "maximum": 0.99368327917014, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9516996049097322, "standard_deviation": 0.054188848082862916, "minimum": 1.8711477724564543, "maximum": 2.026022861897501, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.40859184015866107, "standard_deviation": 0.15152789479407217, "minimum": 0.2173874851404531, "maximum": 0.6292199706602585, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09852728965356576, "standard_deviation": 0.00596200354975998, "minimum": 0.09279851670980417, "maximum": 0.11097838443298635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997711539733867, "standard_deviation": 0.000426436224183364, "minimum": 0.9969241179323962, "maximum": 0.9983131124328909, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.153788213896386, "standard_deviation": 0.0520424970023848, "minimum": 1.092016170354704, "maximum": 1.220379235839265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6052925456375775, "standard_deviation": 0.15142554955986415, "minimum": 0.3302588271380071, "maximum": 0.7474956865786561, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05561918069374871, "standard_deviation": 0.005866877369377866, "minimum": 0.04756635792442484, "maximum": 0.06502617306789168, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13766401488774824, "standard_deviation": 0.0235052113829242, "minimum": 0.10771259830174891, "maximum": 0.17749642642320995, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.180944719120793, "standard_deviation": 0.010743414081489145, "minimum": 6.171712355292349, "maximum": 6.205080788942094, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5086747479398538, "standard_deviation": 0.012987064254244245, "minimum": 0.4941763637903457, "maximum": 0.5309947233072053, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07703168276467756, "standard_deviation": 0.0541987960798172, "minimum": -0.1658293255189664, "maximum": 0.0037236862948702427, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9948497987610037, "standard_deviation": 0.0004137665134377269, "minimum": 0.9942232317337297, "maximum": 0.9953306288232184, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0101995370248693, "standard_deviation": 0.026601812231144568, "minimum": 0.9714449689268478, "maximum": 1.0408240731608611, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.15243540758894586, "standard_deviation": 0.07127691483098045, "minimum": 0.08209534202497008, "maximum": 0.26317173396549, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07380328505582671, "standard_deviation": 0.0028516882976220674, "minimum": 0.0703604791835686, "maximum": 0.07854861037777096, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984342045720543, "standard_deviation": 0.0002970687399492998, "minimum": 0.9980417354909377, "maximum": 0.9987547305667381, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5172622344082127, "standard_deviation": 0.027701045626028917, "minimum": 0.47941269591878477, "maximum": 0.5672479334385808, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24898010397114456, "standard_deviation": 0.11070661888944265, "minimum": 0.12129554456965397, "maximum": 0.43847150541156193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03996983480490382, "standard_deviation": 0.003974956785358015, "minimum": 0.0351414318496556, "maximum": 0.0455745173848626, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0011", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.5468103885650635, "predicted_mos": 2.5468103885650635, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7cb716f3225abbecf4dd3fe187fb94d9c49b2bbe92f369c65ba059dc228b6a34", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.3918995000422, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0404243896557785, "全帯域_10ms_rms_envelope_nrmse": 0.07213076790155253, "全帯域_10ms_envelope_correlation": 0.9951628481221645, "全帯域_10ms_modulation_mae_db": 0.24846642311216952, "全帯域_40ms_log_rms_mae_db": 0.5367313530277263, "全帯域_40ms_rms_envelope_nrmse": 0.03782013744016778, "全帯域_40ms_envelope_correlation": 0.9987356755207006, "全帯域_40ms_modulation_mae_db": 0.27451332626508756, "300–1000 Hz_10ms_log_rms_mae_db": 1.6470946548820957, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10149438633791666, "300–1000 Hz_10ms_envelope_correlation": 0.9925122158902713, "300–1000 Hz_10ms_modulation_mae_db": 0.5276927252352105, "300–1000 Hz_40ms_log_rms_mae_db": 0.9496720033379139, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056629519178235006, "300–1000 Hz_40ms_envelope_correlation": 0.9978071220131185, "300–1000 Hz_40ms_modulation_mae_db": 0.8760404379180633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7061831867960542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15220060544752934, "1000–3000 Hz_10ms_envelope_correlation": 0.9832858484212681, "1000–3000 Hz_10ms_modulation_mae_db": 0.18509873775367103, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9684291979304676, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08781696709835937, "1000–3000 Hz_40ms_envelope_correlation": 0.9945478151276121, "1000–3000 Hz_40ms_modulation_mae_db": 0.668900476735486, "3000–6000 Hz_10ms_log_rms_mae_db": 2.015405408958817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20141274035080325, "3000–6000 Hz_10ms_envelope_correlation": 0.9714465884121913, "3000–6000 Hz_10ms_modulation_mae_db": 1.1184642185317673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.331471729805438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12765894655365073, "3000–6000 Hz_40ms_envelope_correlation": 0.9876127633430901, "3000–6000 Hz_40ms_modulation_mae_db": 0.9918201062908288, "mr_stft_db": 5.7205403498694585, "band_mae_db": 0.19270057855763587, "rms_gain_error_db": -0.12845831920105655, "output_peak": 0.5217923311017789}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1180783806030603, "全帯域_10ms_rms_envelope_nrmse": 0.07748509328360992, "全帯域_10ms_envelope_correlation": 0.9942523580544975, "全帯域_10ms_modulation_mae_db": 0.12284037385371804, "全帯域_40ms_log_rms_mae_db": 0.5764220599040981, "全帯域_40ms_rms_envelope_nrmse": 0.04010761265671762, "全帯域_40ms_envelope_correlation": 0.9983295746882297, "全帯域_40ms_modulation_mae_db": 0.21269724421977998, "300–1000 Hz_10ms_log_rms_mae_db": 1.571350138388482, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10139252441399808, "300–1000 Hz_10ms_envelope_correlation": 0.9922687053644498, "300–1000 Hz_10ms_modulation_mae_db": 0.3647997241971143, "300–1000 Hz_40ms_log_rms_mae_db": 0.9315308300146692, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05481177499961263, "300–1000 Hz_40ms_envelope_correlation": 0.9975692079708571, "300–1000 Hz_40ms_modulation_mae_db": 0.5468886392726128, "1000–3000 Hz_10ms_log_rms_mae_db": 1.705535427246109, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1568256132995722, "1000–3000 Hz_10ms_envelope_correlation": 0.9821082967379318, "1000–3000 Hz_10ms_modulation_mae_db": 0.27541322895675696, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9685152469164304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10099431908259435, "1000–3000 Hz_40ms_envelope_correlation": 0.9923885621058346, "1000–3000 Hz_40ms_modulation_mae_db": 0.7098818527218675, "3000–6000 Hz_10ms_log_rms_mae_db": 2.013725607423312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20894125716254752, "3000–6000 Hz_10ms_envelope_correlation": 0.9688231808801075, "3000–6000 Hz_10ms_modulation_mae_db": 0.8774661734764674, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3139087064153792, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1424660873960409, "3000–6000 Hz_40ms_envelope_correlation": 0.9844600086234653, "3000–6000 Hz_40ms_modulation_mae_db": 1.1438048333246404, "mr_stft_db": 5.70535838528675, "band_mae_db": 0.10942345021665299, "rms_gain_error_db": -0.00775863964171643, "output_peak": 0.526401291726023}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0579656567543936, "全帯域_10ms_rms_envelope_nrmse": 0.07331345042207056, "全帯域_10ms_envelope_correlation": 0.9950161039719367, "全帯域_10ms_modulation_mae_db": 0.2716927102351124, "全帯域_40ms_log_rms_mae_db": 0.5225931742330013, "全帯域_40ms_rms_envelope_nrmse": 0.03999776478825163, "全帯域_40ms_envelope_correlation": 0.9985762143533711, "全帯域_40ms_modulation_mae_db": 0.14248627079414883, "300–1000 Hz_10ms_log_rms_mae_db": 1.667114527633837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09739422710225216, "300–1000 Hz_10ms_envelope_correlation": 0.9931324029445426, "300–1000 Hz_10ms_modulation_mae_db": 0.5399424656580852, "300–1000 Hz_40ms_log_rms_mae_db": 0.9803278869702027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05397743522137019, "300–1000 Hz_40ms_envelope_correlation": 0.9980334966246355, "300–1000 Hz_40ms_modulation_mae_db": 0.5316590389668565, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7247256406727465, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14966675228718793, "1000–3000 Hz_10ms_envelope_correlation": 0.9844448129528769, "1000–3000 Hz_10ms_modulation_mae_db": 0.26471447070742915, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9848984645664834, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1002952106938017, "1000–3000 Hz_40ms_envelope_correlation": 0.9934373518949469, "1000–3000 Hz_40ms_modulation_mae_db": 0.5732074034392068, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0102521636856507, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19974049636474048, "3000–6000 Hz_10ms_envelope_correlation": 0.9717969475428264, "3000–6000 Hz_10ms_modulation_mae_db": 0.6900695638121146, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3332756786007505, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13874438588167257, "3000–6000 Hz_40ms_envelope_correlation": 0.9853035072310924, "3000–6000 Hz_40ms_modulation_mae_db": 0.9852877582093518, "mr_stft_db": 5.713396658884243, "band_mae_db": 0.19768969423219684, "rms_gain_error_db": -0.13616195621664956, "output_peak": 0.5124403776968987}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0822628459517394, "全帯域_10ms_rms_envelope_nrmse": 0.07517991874869384, "全帯域_10ms_envelope_correlation": 0.9946848781926195, "全帯域_10ms_modulation_mae_db": 0.16706692015228605, "全帯域_40ms_log_rms_mae_db": 0.5155853253960517, "全帯域_40ms_rms_envelope_nrmse": 0.03773101018421525, "全帯域_40ms_envelope_correlation": 0.9986802306831141, "全帯域_40ms_modulation_mae_db": 0.1443857536959573, "300–1000 Hz_10ms_log_rms_mae_db": 1.6410440619791873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10671566417844684, "300–1000 Hz_10ms_envelope_correlation": 0.9916098263870414, "300–1000 Hz_10ms_modulation_mae_db": 0.41510160894369386, "300–1000 Hz_40ms_log_rms_mae_db": 0.9299608861090389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05725329867873111, "300–1000 Hz_40ms_envelope_correlation": 0.9976696599068335, "300–1000 Hz_40ms_modulation_mae_db": 0.6295881182963176, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6706361515244332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13470123908899082, "1000–3000 Hz_10ms_envelope_correlation": 0.9869907736072071, "1000–3000 Hz_10ms_modulation_mae_db": 0.28082566778423573, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9705040022887438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07625787944686681, "1000–3000 Hz_40ms_envelope_correlation": 0.9959608272523914, "1000–3000 Hz_40ms_modulation_mae_db": 0.6326443904775378, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9915584614732411, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19600471368285205, "3000–6000 Hz_10ms_envelope_correlation": 0.9726253246380896, "3000–6000 Hz_10ms_modulation_mae_db": 1.0332347771308317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3339228088843558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12804623229934808, "3000–6000 Hz_40ms_envelope_correlation": 0.9878367951111118, "3000–6000 Hz_40ms_modulation_mae_db": 1.3514911958900775, "mr_stft_db": 5.702181273590635, "band_mae_db": 0.17814995160114955, "rms_gain_error_db": -0.11336166320074062, "output_peak": 0.5278294040676141}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0638715591845451, "全帯域_10ms_rms_envelope_nrmse": 0.07040638957617949, "全帯域_10ms_envelope_correlation": 0.9953055828282557, "全帯域_10ms_modulation_mae_db": 0.5015453488233138, "全帯域_40ms_log_rms_mae_db": 0.5433644984639387, "全帯域_40ms_rms_envelope_nrmse": 0.03797410634063325, "全帯域_40ms_envelope_correlation": 0.9985885381263164, "全帯域_40ms_modulation_mae_db": 0.4535171757508554, "300–1000 Hz_10ms_log_rms_mae_db": 1.622706417771666, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10294642490157115, "300–1000 Hz_10ms_envelope_correlation": 0.9921116848187452, "300–1000 Hz_10ms_modulation_mae_db": 0.7617919434495182, "300–1000 Hz_40ms_log_rms_mae_db": 0.9985290801239869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057053509638767735, "300–1000 Hz_40ms_envelope_correlation": 0.9975354188688816, "300–1000 Hz_40ms_modulation_mae_db": 0.8501487446475907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7275529345335514, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14937495543669926, "1000–3000 Hz_10ms_envelope_correlation": 0.9838911415856137, "1000–3000 Hz_10ms_modulation_mae_db": 0.6788707434916533, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9877900336006985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08778167287436922, "1000–3000 Hz_40ms_envelope_correlation": 0.9944835292110293, "1000–3000 Hz_40ms_modulation_mae_db": 0.6512519787188955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9575319361525716, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18692654948470686, "3000–6000 Hz_10ms_envelope_correlation": 0.9752123342541669, "3000–6000 Hz_10ms_modulation_mae_db": 0.9593166526762685, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2455558036057053, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12012920867938202, "3000–6000 Hz_40ms_envelope_correlation": 0.9889595984332683, "3000–6000 Hz_40ms_modulation_mae_db": 0.7943044480160723, "mr_stft_db": 5.711215280911763, "band_mae_db": 0.18607827179870404, "rms_gain_error_db": -0.08514064858086967, "output_peak": 0.5145085738447776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.99814892798869, "全帯域_10ms_rms_envelope_nrmse": 0.07543862253853545, "全帯域_10ms_envelope_correlation": 0.9946034264039469, "全帯域_10ms_modulation_mae_db": 0.2462789039805687, "全帯域_40ms_log_rms_mae_db": 0.47403631355469045, "全帯域_40ms_rms_envelope_nrmse": 0.03843251495924028, "全帯域_40ms_envelope_correlation": 0.9985621804010932, "全帯域_40ms_modulation_mae_db": 0.24489725795810866, "300–1000 Hz_10ms_log_rms_mae_db": 1.6997035154969633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09744322310697884, "300–1000 Hz_10ms_envelope_correlation": 0.9929491035732102, "300–1000 Hz_10ms_modulation_mae_db": 0.22912232690486467, "300–1000 Hz_40ms_log_rms_mae_db": 1.0777750321863107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05449357733748005, "300–1000 Hz_40ms_envelope_correlation": 0.9977632096175538, "300–1000 Hz_40ms_modulation_mae_db": 0.5242076218378372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.707284935382972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14816471242221868, "1000–3000 Hz_10ms_envelope_correlation": 0.9842756742597086, "1000–3000 Hz_10ms_modulation_mae_db": 0.3498201563397548, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9726159543883962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08652928986066709, "1000–3000 Hz_40ms_envelope_correlation": 0.9948386918644263, "1000–3000 Hz_40ms_modulation_mae_db": 0.6853184358388326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.937118472053147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1877955784899684, "3000–6000 Hz_10ms_envelope_correlation": 0.9748949257391488, "3000–6000 Hz_10ms_modulation_mae_db": 0.8537441981463424, "3000–6000 Hz_40ms_log_rms_mae_db": 1.286679830146002, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12639858521929745, "3000–6000 Hz_40ms_envelope_correlation": 0.9880529725714744, "3000–6000 Hz_40ms_modulation_mae_db": 0.8890300090183022, "mr_stft_db": 5.70066077642699, "band_mae_db": 0.18045743326691618, "rms_gain_error_db": -0.09011398708692638, "output_peak": 0.5207916666361289}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.046946587730834, "全帯域_10ms_rms_envelope_nrmse": 0.08114263851152383, "全帯域_10ms_envelope_correlation": 0.9938604723388249, "全帯域_10ms_modulation_mae_db": 0.3096300984140506, "全帯域_40ms_log_rms_mae_db": 0.5219325975738174, "全帯域_40ms_rms_envelope_nrmse": 0.04078718306163192, "全帯域_40ms_envelope_correlation": 0.9985403267820113, "全帯域_40ms_modulation_mae_db": 0.17549072176617533, "300–1000 Hz_10ms_log_rms_mae_db": 1.6280433091543407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10888326721769673, "300–1000 Hz_10ms_envelope_correlation": 0.9913301814486507, "300–1000 Hz_10ms_modulation_mae_db": 0.39493226029376166, "300–1000 Hz_40ms_log_rms_mae_db": 0.9098183331076489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058335470459202926, "300–1000 Hz_40ms_envelope_correlation": 0.9976788652430391, "300–1000 Hz_40ms_modulation_mae_db": 0.7342882130421353, "1000–3000 Hz_10ms_log_rms_mae_db": 1.767864510917939, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14988127905696563, "1000–3000 Hz_10ms_envelope_correlation": 0.9843355708100747, "1000–3000 Hz_10ms_modulation_mae_db": 0.4849732725432263, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0230022042490376, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09599249213063174, "1000–3000 Hz_40ms_envelope_correlation": 0.9940396394632589, "1000–3000 Hz_40ms_modulation_mae_db": 0.9034962981174774, "3000–6000 Hz_10ms_log_rms_mae_db": 2.026495594285551, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2081482036482805, "3000–6000 Hz_10ms_envelope_correlation": 0.9712269937654918, "3000–6000 Hz_10ms_modulation_mae_db": 0.8850659907500278, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2952028852555626, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1357087749986049, "3000–6000 Hz_40ms_envelope_correlation": 0.9872608985470546, "3000–6000 Hz_40ms_modulation_mae_db": 0.9352859516000139, "mr_stft_db": 5.698251886356764, "band_mae_db": 0.2881168388357452, "rms_gain_error_db": -0.1449644505935196, "output_peak": 0.4939775212753851}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0611409323144132, "全帯域_10ms_rms_envelope_nrmse": 0.07672433072635644, "全帯域_10ms_envelope_correlation": 0.9943789134032089, "全帯域_10ms_modulation_mae_db": 0.1569107658221742, "全帯域_40ms_log_rms_mae_db": 0.50749883202071, "全帯域_40ms_rms_envelope_nrmse": 0.036790574398842785, "全帯域_40ms_envelope_correlation": 0.9986224971384879, "全帯域_40ms_modulation_mae_db": 0.23643002810154343, "300–1000 Hz_10ms_log_rms_mae_db": 1.6508278335818631, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10637260008010029, "300–1000 Hz_10ms_envelope_correlation": 0.9914812915188845, "300–1000 Hz_10ms_modulation_mae_db": 0.5518045417196601, "300–1000 Hz_40ms_log_rms_mae_db": 1.0184718469794323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05354852813602999, "300–1000 Hz_40ms_envelope_correlation": 0.997728977285589, "300–1000 Hz_40ms_modulation_mae_db": 0.5461412906824555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6908526933525907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14681594246329205, "1000–3000 Hz_10ms_envelope_correlation": 0.9845830837992826, "1000–3000 Hz_10ms_modulation_mae_db": 0.3358788836047922, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9767168204073604, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09421675006315698, "1000–3000 Hz_40ms_envelope_correlation": 0.9937342272812921, "1000–3000 Hz_40ms_modulation_mae_db": 0.38098667644487744, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9938670592974033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18526022573819945, "3000–6000 Hz_10ms_envelope_correlation": 0.9755855642449298, "3000–6000 Hz_10ms_modulation_mae_db": 0.7546017557336215, "3000–6000 Hz_40ms_log_rms_mae_db": 1.323327185109541, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12426339021307983, "3000–6000 Hz_40ms_envelope_correlation": 0.9885081078896184, "3000–6000 Hz_40ms_modulation_mae_db": 1.011885626355647, "mr_stft_db": 5.703446833182149, "band_mae_db": 0.1905396586322808, "rms_gain_error_db": -0.044648983251311836, "output_peak": 0.504709193337702}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.095525802268823, "spectral_convergence": 0.24556693182446462, "spectral_convergence_status": "ok", "amplitude_floor": 1.4412887028114478e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.361602153915904, "spectral_convergence": 0.2373413099292186, "spectral_convergence_status": "ok", "amplitude_floor": 8.082487629299873e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.704493093423647, "spectral_convergence": 0.2564960820297405, "spectral_convergence_status": "ok", "amplitude_floor": 6.705293681874596e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00036448625115270087, "output_power": 0.00038433370165986584, "delta_db": 0.23027317914098172, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008505212390156939, "output_power": 0.0008299713886430803, "delta_db": -0.10622041853762568, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010532918945599765, "output_power": 0.0009908454657208481, "delta_db": -0.265428158401217, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002198202138287696, "output_power": 0.0002100813475802888, "delta_db": -0.19680131424348418, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.872413022951283e-05, "output_power": 3.660797857524767e-05, "delta_db": -0.24405923382339795, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.441903264074617e-05, "output_power": 2.5065165245659537e-05, "delta_db": 0.1134211671991085, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19270057855763587, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12845831920105655, "pitch": null, "output_rms_dbfs": -26.060906418476634, "output_peak": 0.5217923311017789, "reference_rms_dbfs": -25.93244809927558, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7205403498694585}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019142337741905198, "log_rms_mae_db": 1.0404243896557785, "rms_envelope_nrmse": 0.07213076790155253, "envelope_correlation": 0.9951628481221645, "rms_gain_error_db": -0.12845509842817546, "reference_mean_power": 0.0025533227171819564, "output_mean_power": 0.0024789068032140945}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1748402838347296e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007147512141697323, "output_power": 0.0006997202390469126, "reference_active": true, "delta_db": -0.09230465942321128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.175142428480134e-05, "output_power": 7.873173203187361e-05, "reference_active": true, "delta_db": -0.16345522077349856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.431203959050675e-05, "output_power": 1.278608508172862e-05, "reference_active": true, "delta_db": -0.4896393891397987}], "active_band_count": 3, "active_band_mae_db": 0.24846642311216952}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016640454830055276, "log_rms_mae_db": 0.5367313530277263, "rms_envelope_nrmse": 0.03782013744016778, "envelope_correlation": 0.9987356755207006, "rms_gain_error_db": -0.1285492937134874, "reference_mean_power": 0.002564981061322552, "output_mean_power": 0.0024901713574295185}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0626539543337418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006440432662381922, "output_power": 0.0006301390350481723, "reference_active": true, "delta_db": -0.09478660277991831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1296244809426825e-05, "output_power": 4.0430434824514005e-05, "reference_active": true, "delta_db": -0.0920214961647181}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.585289215752233e-07, "output_power": 7.41449782287204e-07, "reference_active": true, "delta_db": -0.6367318798506262}], "active_band_count": 3, "active_band_mae_db": 0.27451332626508756}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016776043588712122, "log_rms_mae_db": 1.6470946548820957, "rms_envelope_nrmse": 0.10149438633791666, "envelope_correlation": 0.9925122158902713, "rms_gain_error_db": -0.2064248940169851, "reference_mean_power": 0.0015150798695545982, "output_mean_power": 0.0014447511155717848}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.503421266374213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005710315291680691, "output_power": 0.0005499636396883424, "reference_active": true, "delta_db": -0.16326110735085844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462070521392158e-05, "output_power": 5.622928222007332e-05, "reference_active": true, "delta_db": -0.6040915354380245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.82379092220564e-06, "output_power": 8.141539937416362e-06, "reference_active": true, "delta_db": -0.8157255329167485}], "active_band_count": 3, "active_band_mae_db": 0.5276927252352105}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015135634256581334, "log_rms_mae_db": 0.9496720033379139, "rms_envelope_nrmse": 0.056629519178235006, "envelope_correlation": 0.9978071220131185, "rms_gain_error_db": -0.20644708168548406, "reference_mean_power": 0.0015225180811750088, "output_mean_power": 0.0014518366342567144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.758019890177841e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005253930575206644, "output_power": 0.0005087238405537553, "reference_active": true, "delta_db": -0.14002239128453844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0513708673138778e-05, "output_power": 2.677024810476374e-05, "reference_active": true, "delta_db": -0.5684259931208777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.334657094067365e-07, "output_power": 5.356985166056577e-07, "reference_active": true, "delta_db": -1.9196729293487738}], "active_band_count": 3, "active_band_mae_db": 0.8760404379180633}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.526806573516121e-05, "log_rms_mae_db": 1.7061831867960542, "rms_envelope_nrmse": 0.15220060544752934, "envelope_correlation": 0.9832858484212681, "rms_gain_error_db": -0.26230748054712877, "reference_mean_power": 0.00023682053730462446, "output_mean_power": 0.00022294031686956865}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1631666261710027e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014015200472863968, "output_power": 0.00013005163358050664, "reference_active": true, "delta_db": -0.32483502525929625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3136457414027934e-05, "output_power": 1.2866611177955288e-05, "reference_active": true, "delta_db": -0.0901408524561953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.230915915111415e-06, "output_power": 3.1281935920216887e-06, "reference_active": true, "delta_db": -0.14032033554552148}], "active_band_count": 3, "active_band_mae_db": 0.18509873775367103}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.301677625713533e-05, "log_rms_mae_db": 0.9684291979304676, "rms_envelope_nrmse": 0.08781696709835937, "envelope_correlation": 0.9945478151276121, "rms_gain_error_db": -0.26280734921336557, "reference_mean_power": 0.0002379882918616051, "output_mean_power": 0.00022401384324353014}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0209734833253925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001318054605596598, "output_power": 0.00012275773851107283, "reference_active": true, "delta_db": -0.3088452388965047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.132933188997116e-06, "output_power": 6.444931379805374e-06, "reference_active": true, "delta_db": 0.21550064120495416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2319303983633163e-07, "output_power": 1.586518322981222e-07, "reference_active": true, "delta_db": -1.4823555501049992}], "active_band_count": 3, "active_band_mae_db": 0.668900476735486}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.803319177623983e-05, "log_rms_mae_db": 2.015405408958817, "rms_envelope_nrmse": 0.20141274035080325, "envelope_correlation": 0.9714465884121913, "rms_gain_error_db": 0.02661096747591676, "reference_mean_power": 3.0517013147816785e-05, "output_mean_power": 3.070457719942233e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.826326538605559e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1311662288856248e-05, "output_power": 1.0979395434842405e-05, "reference_active": true, "delta_db": -0.12948003778229525}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5475498743040963e-06, "output_power": 2.2411056628189627e-06, "reference_active": true, "delta_db": -0.5566036212863373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.732573657220471e-07, "output_power": 1.0599381974484097e-06, "reference_active": true, "delta_db": 2.669308996526669}], "active_band_count": 3, "active_band_mae_db": 1.1184642185317673}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.459177046887288e-05, "log_rms_mae_db": 1.331471729805438, "rms_envelope_nrmse": 0.12765894655365073, "envelope_correlation": 0.9876127633430901, "rms_gain_error_db": 0.026602580770104956, "reference_mean_power": 3.0668913834731335e-05, "output_mean_power": 3.085735191124708e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6363060360497894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1094596811060155e-05, "output_power": 1.0922552844371381e-05, "reference_active": true, "delta_db": -0.06787369605748421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1901156553217825e-06, "output_power": 1.2708351021970678e-06, "reference_active": true, "delta_db": 0.2850003398788647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.41782955712918e-08, "output_power": 8.081046108929491e-08, "reference_active": true, "delta_db": 2.6225862829361377}], "active_band_count": 3, "active_band_mae_db": 0.9918201062908288}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8824790944511773, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1602767455c6efa41a3ca78941e5bbb7a66eebfe1c41237849ce22e65e9663b2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9842394002717454, "standard_deviation": 0.0013790825137570287, "minimum": 0.9821082967379318, "maximum": 0.9869907736072071, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7125794350532995, "standard_deviation": 0.02873034163593128, "minimum": 1.6706361515244332, "maximum": 1.767864510917939, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.35694939514769, "standard_deviation": 0.1562969509819236, "minimum": 0.18509873775367103, "maximum": 0.6788707434916533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.148453887437807, "standard_deviation": 0.006328872236710033, "minimum": 0.13470123908899082, "maximum": 0.1568256132995722, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941788305250989, "standard_deviation": 0.0010563967069650874, "minimum": 0.9923885621058346, "maximum": 0.9959608272523914, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9815589905434523, "standard_deviation": 0.018266620302990134, "minimum": 0.9684291979304676, "maximum": 1.0230022042490376, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6507109390617727, "standard_deviation": 0.1454131265788875, "minimum": 0.38098667644487744, "maximum": 0.9034962981174774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09123557265630591, "standard_deviation": 0.008271487636906391, "minimum": 0.07625787944686681, "maximum": 0.10099431908259435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.972701482434619, "standard_deviation": 0.0023618267884220694, "minimum": 0.9688231808801075, "maximum": 0.9755855642449298, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9932443379162117, "standard_deviation": 0.031016645480006724, "minimum": 1.937118472053147, "maximum": 2.026495594285551, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8964954162821801, "standard_deviation": 0.13988779365375592, "minimum": 0.6900695638121146, "maximum": 1.1184642185317673, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1967787206152623, "standard_deviation": 0.009399662787226842, "minimum": 0.18526022573819945, "maximum": 0.20894125716254752, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9872493314687719, "standard_deviation": 0.001567493638374511, "minimum": 0.9844600086234653, "maximum": 0.9889595984332683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.307918078477842, "standard_deviation": 0.03082009832828608, "minimum": 1.2455558036057053, "maximum": 1.3339228088843558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.012863741088117, "standard_deviation": 0.16993260012459563, "minimum": 0.7943044480160723, "maximum": 1.3514911958900775, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13042695140513458, "standard_deviation": 0.007705365221818428, "minimum": 0.12012920867938202, "maximum": 0.1424660873960409, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921744264932245, "standard_deviation": 0.0006715365352693643, "minimum": 0.9913301814486507, "maximum": 0.9931324029445426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6409855573610543, "standard_deviation": 0.03706256659824066, "minimum": 1.571350138388482, "maximum": 1.6997035154969633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47314844955023855, "standard_deviation": 0.15933512247943235, "minimum": 0.22912232690486467, "maximum": 0.7617919434495182, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1028302896673701, "standard_deviation": 0.004254731680918042, "minimum": 0.09739422710225216, "maximum": 0.10888326721769673, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977232446913135, "standard_deviation": 0.00015527323409663066, "minimum": 0.9975354188688816, "maximum": 0.9980334966246355, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9745107373536503, "standard_deviation": 0.0558580411417171, "minimum": 0.9098183331076489, "maximum": 1.0777750321863107, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6548702630829836, "standard_deviation": 0.14625422157757564, "minimum": 0.5242076218378372, "maximum": 0.8760404379180633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05576288920617871, "standard_deviation": 0.001767292321344315, "minimum": 0.05354852813602999, "maximum": 0.058335470459202926, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19039448464266018, "standard_deviation": 0.048450017581312996, "minimum": 0.10942345021665299, "maximum": 0.2881168388357452, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.706881430563595, "standard_deviation": 0.007532498241220295, "minimum": 5.698251886356764, "maximum": 5.7205403498694585, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5153062949607885, "standard_deviation": 0.01151489825422734, "minimum": 0.4939775212753851, "maximum": 0.5278294040676141, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09382608097159884, "standard_deviation": 0.04766504836917736, "minimum": -0.1449644505935196, "maximum": -0.00775863964171643, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946580729144319, "standard_deviation": 0.000490576826590447, "minimum": 0.9938604723388249, "maximum": 0.9953055828282557, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0586049100229318, "standard_deviation": 0.034336034477882166, "minimum": 0.99814892798869, "maximum": 1.1180783806030603, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.25305394304917417, "standard_deviation": 0.11889734015518, "minimum": 0.12284037385371804, "maximum": 0.5015453488233138, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07522765146356526, "standard_deviation": 0.00335664311438205, "minimum": 0.07040638957617949, "maximum": 0.08114263851152383, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985794047116655, "standard_deviation": 0.00011998559392360805, "minimum": 0.9983295746882297, "maximum": 0.9987356755207006, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5247705192717542, "standard_deviation": 0.0295830851375457, "minimum": 0.47403631355469045, "maximum": 0.5764220599040981, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.23555222231895706, "standard_deviation": 0.10015496491960543, "minimum": 0.14248627079414883, "maximum": 0.4535171757508554, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03870511297871257, "standard_deviation": 0.0014131609045334628, "minimum": 0.036790574398842785, "maximum": 0.04078718306163192, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0011", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2612550258636475, "predicted_mos": 2.2612550258636475, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7582b41c3ab7b1f23aac94fce12946eabee336d8df3dafa01531e1b524b33c94", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.627853200305253, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0048357453418706, "全帯域_10ms_rms_envelope_nrmse": 0.07912208506829276, "全帯域_10ms_envelope_correlation": 0.994157127613546, "全帯域_10ms_modulation_mae_db": 0.1159298841232362, "全帯域_40ms_log_rms_mae_db": 0.5461995477716278, "全帯域_40ms_rms_envelope_nrmse": 0.045860012361831314, "全帯域_40ms_envelope_correlation": 0.9980469574233886, "全帯域_40ms_modulation_mae_db": 0.06889736097380425, "300–1000 Hz_10ms_log_rms_mae_db": 1.8023615760724012, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10764112532015811, "300–1000 Hz_10ms_envelope_correlation": 0.9915429392310849, "300–1000 Hz_10ms_modulation_mae_db": 0.6150806856868097, "300–1000 Hz_40ms_log_rms_mae_db": 1.1137416307790804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06292507639085029, "300–1000 Hz_40ms_envelope_correlation": 0.9971986104386358, "300–1000 Hz_40ms_modulation_mae_db": 0.9104041099861986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8236613213450235, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1413635831541276, "1000–3000 Hz_10ms_envelope_correlation": 0.9854484095266137, "1000–3000 Hz_10ms_modulation_mae_db": 0.46112800941447313, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9776601996483772, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0870078247343152, "1000–3000 Hz_40ms_envelope_correlation": 0.9942915075616826, "1000–3000 Hz_40ms_modulation_mae_db": 0.6023391671280792, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3097879787931266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18913176662114034, "3000–6000 Hz_10ms_envelope_correlation": 0.9746448081413417, "3000–6000 Hz_10ms_modulation_mae_db": 0.8867167565351656, "3000–6000 Hz_40ms_log_rms_mae_db": 1.467348054010481, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12040579318245657, "3000–6000 Hz_40ms_envelope_correlation": 0.9894759448421049, "3000–6000 Hz_40ms_modulation_mae_db": 0.8533502479722905, "mr_stft_db": 5.850828832677878, "band_mae_db": 0.1623342932242181, "rms_gain_error_db": -0.13834094153203133, "output_peak": 0.5007795649389142}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.04403266354351, "全帯域_10ms_rms_envelope_nrmse": 0.07300262520877629, "全帯域_10ms_envelope_correlation": 0.9949561426827246, "全帯域_10ms_modulation_mae_db": 0.163750631856564, "全帯域_40ms_log_rms_mae_db": 0.5435577884445492, "全帯域_40ms_rms_envelope_nrmse": 0.03820807951799945, "全帯域_40ms_envelope_correlation": 0.9985922078472131, "全帯域_40ms_modulation_mae_db": 0.2716513244253466, "300–1000 Hz_10ms_log_rms_mae_db": 1.790630509499792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09755144468764035, "300–1000 Hz_10ms_envelope_correlation": 0.9929740311369882, "300–1000 Hz_10ms_modulation_mae_db": 0.42149447754533736, "300–1000 Hz_40ms_log_rms_mae_db": 1.0697556684535445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052597364927584156, "300–1000 Hz_40ms_envelope_correlation": 0.9979784259231742, "300–1000 Hz_40ms_modulation_mae_db": 0.7687340215974204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9013950616728879, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18369358562681556, "1000–3000 Hz_10ms_envelope_correlation": 0.9752273771757363, "1000–3000 Hz_10ms_modulation_mae_db": 0.5215238844172223, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0955697121351626, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11806911198690426, "1000–3000 Hz_40ms_envelope_correlation": 0.9893599055261804, "1000–3000 Hz_40ms_modulation_mae_db": 0.4949138834071796, "3000–6000 Hz_10ms_log_rms_mae_db": 2.303095020389132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20761257442712172, "3000–6000 Hz_10ms_envelope_correlation": 0.9692467988076243, "3000–6000 Hz_10ms_modulation_mae_db": 0.7979210701348486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5151455203830977, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13744162648922972, "3000–6000 Hz_40ms_envelope_correlation": 0.9858519468273382, "3000–6000 Hz_40ms_modulation_mae_db": 0.7983674559789785, "mr_stft_db": 5.8450689420387905, "band_mae_db": 0.12981371895105162, "rms_gain_error_db": -0.08958516010175255, "output_peak": 0.5269533140510891}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.007345542515242, "全帯域_10ms_rms_envelope_nrmse": 0.0785905608952017, "全帯域_10ms_envelope_correlation": 0.99408429503634, "全帯域_10ms_modulation_mae_db": 0.37790854588097167, "全帯域_40ms_log_rms_mae_db": 0.5431182914902968, "全帯域_40ms_rms_envelope_nrmse": 0.04565150959158876, "全帯域_40ms_envelope_correlation": 0.9978432872135002, "全帯域_40ms_modulation_mae_db": 0.48426266384943023, "300–1000 Hz_10ms_log_rms_mae_db": 1.8257493052366587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10337811290335512, "300–1000 Hz_10ms_envelope_correlation": 0.9919634897416033, "300–1000 Hz_10ms_modulation_mae_db": 0.6580147376629011, "300–1000 Hz_40ms_log_rms_mae_db": 1.0623220087744316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05900277411013295, "300–1000 Hz_40ms_envelope_correlation": 0.9972256579700048, "300–1000 Hz_40ms_modulation_mae_db": 0.48344129590577883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8599745260800027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1508171798318315, "1000–3000 Hz_10ms_envelope_correlation": 0.9834889809283639, "1000–3000 Hz_10ms_modulation_mae_db": 0.52280849879588, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0327794142113913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09332951293721113, "1000–3000 Hz_40ms_envelope_correlation": 0.9935684618557223, "1000–3000 Hz_40ms_modulation_mae_db": 0.8132800755209028, "3000–6000 Hz_10ms_log_rms_mae_db": 2.318527315537261, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2040563850320313, "3000–6000 Hz_10ms_envelope_correlation": 0.970441719816448, "3000–6000 Hz_10ms_modulation_mae_db": 0.5567982247428368, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4720309870473618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1320995767654531, "3000–6000 Hz_40ms_envelope_correlation": 0.9867515463746627, "3000–6000 Hz_40ms_modulation_mae_db": 0.6712086038745081, "mr_stft_db": 5.8502890694772764, "band_mae_db": 0.13764270758963826, "rms_gain_error_db": -0.03216046051765612, "output_peak": 0.495431515965114}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0336489845466128, "全帯域_10ms_rms_envelope_nrmse": 0.07011945112820103, "全帯域_10ms_envelope_correlation": 0.9953577172056735, "全帯域_10ms_modulation_mae_db": 0.3151638698730617, "全帯域_40ms_log_rms_mae_db": 0.5060471660574161, "全帯域_40ms_rms_envelope_nrmse": 0.03647842234381488, "全帯域_40ms_envelope_correlation": 0.9987194198445134, "全帯域_40ms_modulation_mae_db": 0.3540693212038528, "300–1000 Hz_10ms_log_rms_mae_db": 1.7841318784529114, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09281744785444591, "300–1000 Hz_10ms_envelope_correlation": 0.9936276568836481, "300–1000 Hz_10ms_modulation_mae_db": 0.5350519515253577, "300–1000 Hz_40ms_log_rms_mae_db": 1.0746663308249214, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05249700119401327, "300–1000 Hz_40ms_envelope_correlation": 0.9979308045457249, "300–1000 Hz_40ms_modulation_mae_db": 0.8421831971988517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8387492409994444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16882506105536305, "1000–3000 Hz_10ms_envelope_correlation": 0.9793688740477022, "1000–3000 Hz_10ms_modulation_mae_db": 0.35441711649344404, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0440284000179498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11385412937996175, "1000–3000 Hz_40ms_envelope_correlation": 0.9904967095289167, "1000–3000 Hz_40ms_modulation_mae_db": 0.5427931718859853, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2680998877930003, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19481186029950226, "3000–6000 Hz_10ms_envelope_correlation": 0.973079801455834, "3000–6000 Hz_10ms_modulation_mae_db": 0.5954082114297313, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4279960362561803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12320143009721503, "3000–6000 Hz_40ms_envelope_correlation": 0.9884713564453568, "3000–6000 Hz_40ms_modulation_mae_db": 0.8325818731190049, "mr_stft_db": 5.8322767660155215, "band_mae_db": 0.14106814137301407, "rms_gain_error_db": -0.09109104506824217, "output_peak": 0.519862740376394}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9944245158404268, "全帯域_10ms_rms_envelope_nrmse": 0.08920881490824846, "全帯域_10ms_envelope_correlation": 0.9923675687784532, "全帯域_10ms_modulation_mae_db": 0.37199370784115393, "全帯域_40ms_log_rms_mae_db": 0.5163848521940826, "全帯域_40ms_rms_envelope_nrmse": 0.048221790533963, "全帯域_40ms_envelope_correlation": 0.9976018643432859, "全帯域_40ms_modulation_mae_db": 0.44368455104347954, "300–1000 Hz_10ms_log_rms_mae_db": 1.8308044165034938, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10832014959039865, "300–1000 Hz_10ms_envelope_correlation": 0.9912121890607682, "300–1000 Hz_10ms_modulation_mae_db": 0.352170747676613, "300–1000 Hz_40ms_log_rms_mae_db": 1.0571790116701931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05525662648222062, "300–1000 Hz_40ms_envelope_correlation": 0.997661675974734, "300–1000 Hz_40ms_modulation_mae_db": 0.08103848113326177, "1000–3000 Hz_10ms_log_rms_mae_db": 1.875934943296277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1634378804443739, "1000–3000 Hz_10ms_envelope_correlation": 0.9804341308686714, "1000–3000 Hz_10ms_modulation_mae_db": 0.1888958401152037, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0835754344752253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09688731693795512, "1000–3000 Hz_40ms_envelope_correlation": 0.9928453483199293, "1000–3000 Hz_40ms_modulation_mae_db": 0.5979443522753654, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2855339264632266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20890667374381788, "3000–6000 Hz_10ms_envelope_correlation": 0.9695445305651993, "3000–6000 Hz_10ms_modulation_mae_db": 0.7380748362117918, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4646823927108403, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13645013506924153, "3000–6000 Hz_40ms_envelope_correlation": 0.9860478925006738, "3000–6000 Hz_40ms_modulation_mae_db": 1.0437002496951455, "mr_stft_db": 5.829914990768419, "band_mae_db": 0.12205876421411754, "rms_gain_error_db": -0.028709674522556114, "output_peak": 0.5137941002178051}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9899984643612787, "全帯域_10ms_rms_envelope_nrmse": 0.0725783339427753, "全帯域_10ms_envelope_correlation": 0.9949560671070241, "全帯域_10ms_modulation_mae_db": 0.07628552045446675, "全帯域_40ms_log_rms_mae_db": 0.5100074403276421, "全帯域_40ms_rms_envelope_nrmse": 0.037181893413905945, "全帯域_40ms_envelope_correlation": 0.9985658573817653, "全帯域_40ms_modulation_mae_db": 0.2025312882892035, "300–1000 Hz_10ms_log_rms_mae_db": 1.7976297542783513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1022215885922675, "300–1000 Hz_10ms_envelope_correlation": 0.9921472003723691, "300–1000 Hz_10ms_modulation_mae_db": 0.4241718218846861, "300–1000 Hz_40ms_log_rms_mae_db": 1.0784736929865724, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05324288056128285, "300–1000 Hz_40ms_envelope_correlation": 0.9977241158697755, "300–1000 Hz_40ms_modulation_mae_db": 0.24732974899402138, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9053489528249838, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15355883634162565, "1000–3000 Hz_10ms_envelope_correlation": 0.9827633154146249, "1000–3000 Hz_10ms_modulation_mae_db": 0.45498157137239814, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0565754059782428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09849664282300215, "1000–3000 Hz_40ms_envelope_correlation": 0.9925942940469736, "1000–3000 Hz_40ms_modulation_mae_db": 0.6730729252318914, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2855091065739868, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21274037746356958, "3000–6000 Hz_10ms_envelope_correlation": 0.9693764394221613, "3000–6000 Hz_10ms_modulation_mae_db": 0.7712622720954866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4734508252883303, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13824017516689238, "3000–6000 Hz_40ms_envelope_correlation": 0.9863742815978779, "3000–6000 Hz_40ms_modulation_mae_db": 0.6805863909138427, "mr_stft_db": 5.8407933764351485, "band_mae_db": 0.16792364285386627, "rms_gain_error_db": -0.015170930613627252, "output_peak": 0.5083937855920473}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0295563903563234, "全帯域_10ms_rms_envelope_nrmse": 0.07774430437332158, "全帯域_10ms_envelope_correlation": 0.9943822046624765, "全帯域_10ms_modulation_mae_db": 0.21585444404055465, "全帯域_40ms_log_rms_mae_db": 0.527365601761848, "全帯域_40ms_rms_envelope_nrmse": 0.04033422513946324, "全帯域_40ms_envelope_correlation": 0.9985799062128101, "全帯域_40ms_modulation_mae_db": 0.284426023191213, "300–1000 Hz_10ms_log_rms_mae_db": 1.8253625390982038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10753953055623082, "300–1000 Hz_10ms_envelope_correlation": 0.9917998820549072, "300–1000 Hz_10ms_modulation_mae_db": 0.47757520011694105, "300–1000 Hz_40ms_log_rms_mae_db": 1.1111903990163094, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06217444193744418, "300–1000 Hz_40ms_envelope_correlation": 0.9975868525111189, "300–1000 Hz_40ms_modulation_mae_db": 0.8518450651758513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8983934778605824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16395094830647988, "1000–3000 Hz_10ms_envelope_correlation": 0.9805154592030145, "1000–3000 Hz_10ms_modulation_mae_db": 0.5549601814768942, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0865148363409247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10534292063902888, "1000–3000 Hz_40ms_envelope_correlation": 0.9918672182575791, "1000–3000 Hz_40ms_modulation_mae_db": 0.8471290277610364, "3000–6000 Hz_10ms_log_rms_mae_db": 2.305651233302755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19966200486007235, "3000–6000 Hz_10ms_envelope_correlation": 0.9715411712613086, "3000–6000 Hz_10ms_modulation_mae_db": 0.7697357675736282, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4739965029013158, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13330051809765522, "3000–6000 Hz_40ms_envelope_correlation": 0.9865927347872108, "3000–6000 Hz_40ms_modulation_mae_db": 0.6596799418316335, "mr_stft_db": 5.855751314204693, "band_mae_db": 0.19353443899450448, "rms_gain_error_db": -0.1414890454100063, "output_peak": 0.5013992029976044}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0244537883914542, "全帯域_10ms_rms_envelope_nrmse": 0.07571136018785005, "全帯域_10ms_envelope_correlation": 0.9945128567887702, "全帯域_10ms_modulation_mae_db": 0.1183175059354757, "全帯域_40ms_log_rms_mae_db": 0.5466455795111524, "全帯域_40ms_rms_envelope_nrmse": 0.04042346092897513, "全帯域_40ms_envelope_correlation": 0.9983203779454833, "全帯域_40ms_modulation_mae_db": 0.43290712965759814, "300–1000 Hz_10ms_log_rms_mae_db": 1.7663434420666395, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10156800483670671, "300–1000 Hz_10ms_envelope_correlation": 0.992248199703651, "300–1000 Hz_10ms_modulation_mae_db": 0.4066777138811934, "300–1000 Hz_40ms_log_rms_mae_db": 1.0801921332954243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05512774071492181, "300–1000 Hz_40ms_envelope_correlation": 0.9975594768868293, "300–1000 Hz_40ms_modulation_mae_db": 0.3909912083663236, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8747768208755646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16954807920308748, "1000–3000 Hz_10ms_envelope_correlation": 0.9795053936189825, "1000–3000 Hz_10ms_modulation_mae_db": 0.5283951087017099, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1120087195137374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11117443530314995, "1000–3000 Hz_40ms_envelope_correlation": 0.9914576029227741, "1000–3000 Hz_40ms_modulation_mae_db": 0.7268053971433021, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3005889819139194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19609399172212946, "3000–6000 Hz_10ms_envelope_correlation": 0.9726470924381106, "3000–6000 Hz_10ms_modulation_mae_db": 0.5933408454573365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5254515935056687, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1285841909888135, "3000–6000 Hz_40ms_envelope_correlation": 0.987361305753228, "3000–6000 Hz_40ms_modulation_mae_db": 0.8510387205555938, "mr_stft_db": 5.852918319835175, "band_mae_db": 0.12024182801837512, "rms_gain_error_db": -0.04241695056147751, "output_peak": 0.5182576014159885}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.318521514609895, "spectral_convergence": 0.24859356003857938, "spectral_convergence_status": "ok", "amplitude_floor": 1.4412887028114478e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5074379544350345, "spectral_convergence": 0.24368293809344063, "spectral_convergence_status": "ok", "amplitude_floor": 8.082487629299873e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7265270289887065, "spectral_convergence": 0.25842325394677523, "spectral_convergence_status": "ok", "amplitude_floor": 6.705293681874596e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00036448625115270087, "output_power": 0.0003723777885167991, "delta_db": 0.09302617541909329, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008505212390156939, "output_power": 0.0008339140348844994, "delta_db": -0.08563880051048343, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010532918945599765, "output_power": 0.000988596029523961, "delta_db": -0.2752988027277143, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002198202138287696, "output_power": 0.0002164794907061045, "delta_db": -0.06650868463484282, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.872413022951283e-05, "output_power": 3.669166506530922e-05, "delta_db": -0.23414251531562735, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.441903264074617e-05, "output_power": 2.3216105963367823e-05, "delta_db": -0.2193907807375475, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1623342932242181, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13834094153203133, "pitch": null, "output_rms_dbfs": -26.07078904080761, "output_peak": 0.5007795649389142, "reference_rms_dbfs": -25.93244809927558, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.850828832677878}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019142337741905198, "log_rms_mae_db": 1.0048357453418706, "rms_envelope_nrmse": 0.07912208506829276, "envelope_correlation": 0.994157127613546, "rms_gain_error_db": -0.1383541082348617, "reference_mean_power": 0.0025533227171819564, "output_mean_power": 0.0024732629879980835}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1748402838347296e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007147512141697323, "output_power": 0.0007084446331390502, "reference_active": true, "delta_db": -0.038489871566494804}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.175142428480134e-05, "output_power": 8.17461793676838e-05, "reference_active": true, "delta_db": -0.0002786387704871944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.431203959050675e-05, "output_power": 1.3329056679291608e-05, "reference_active": true, "delta_db": -0.3090211420327266}], "active_band_count": 3, "active_band_mae_db": 0.1159298841232362}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016640454830055276, "log_rms_mae_db": 0.5461995477716278, "rms_envelope_nrmse": 0.045860012361831314, "envelope_correlation": 0.9980469574233886, "rms_gain_error_db": -0.1380374870750631, "reference_mean_power": 0.002564981061322552, "output_mean_power": 0.00248473692584686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0626539543337418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006440432662381922, "output_power": 0.0006378044151576633, "reference_active": true, "delta_db": -0.04227522555001907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1296244809426825e-05, "output_power": 4.096194042639961e-05, "reference_active": true, "delta_db": -0.03530040147506434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.585289215752233e-07, "output_power": 8.844363336186086e-07, "reference_active": true, "delta_db": 0.12911645589632934}], "active_band_count": 3, "active_band_mae_db": 0.06889736097380425}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016776043588712122, "log_rms_mae_db": 1.8023615760724012, "rms_envelope_nrmse": 0.10764112532015811, "envelope_correlation": 0.9915429392310849, "rms_gain_error_db": -0.20840609512966562, "reference_mean_power": 0.0015150798695545982, "output_mean_power": 0.001444092187159732}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.503421266374213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005710315291680691, "output_power": 0.000546028773776027, "reference_active": true, "delta_db": -0.19444559137111408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462070521392158e-05, "output_power": 5.807812939563515e-05, "reference_active": true, "delta_db": -0.4635907323668854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.82379092220564e-06, "output_power": 7.4740936900152355e-06, "reference_active": true, "delta_db": -1.1872057333224297}], "active_band_count": 3, "active_band_mae_db": 0.6150806856868097}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015135634256581334, "log_rms_mae_db": 1.1137416307790804, "rms_envelope_nrmse": 0.06292507639085029, "envelope_correlation": 0.9971986104386358, "rms_gain_error_db": -0.20830960382991356, "reference_mean_power": 0.0015225180811750088, "output_mean_power": 0.0014512141308081594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.758019890177841e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005253930575206644, "output_power": 0.0005036654569650498, "reference_active": true, "delta_db": -0.18342163240100334}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0513708673138778e-05, "output_power": 2.788087558140453e-05, "reference_active": true, "delta_db": -0.39188587142324366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.334657094067365e-07, "output_power": 5.073378382958326e-07, "reference_active": true, "delta_db": -2.155904826134349}], "active_band_count": 3, "active_band_mae_db": 0.9104041099861986}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.526806573516121e-05, "log_rms_mae_db": 1.8236613213450235, "rms_envelope_nrmse": 0.1413635831541276, "envelope_correlation": 0.9854484095266137, "rms_gain_error_db": -0.16147072546468244, "reference_mean_power": 0.00023682053730462446, "output_mean_power": 0.00022817722261503286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1631666261710027e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014015200472863968, "output_power": 0.0001338328247455562, "reference_active": true, "delta_db": -0.20036669767173218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3136457414027934e-05, "output_power": 1.4197690399180164e-05, "reference_active": true, "delta_db": 0.33739439275079547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.230915915111415e-06, "output_power": 2.6592756800914804e-06, "reference_active": true, "delta_db": -0.8456229378208917}], "active_band_count": 3, "active_band_mae_db": 0.46112800941447313}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.301677625713533e-05, "log_rms_mae_db": 0.9776601996483772, "rms_envelope_nrmse": 0.0870078247343152, "envelope_correlation": 0.9942915075616826, "rms_gain_error_db": -0.16182535088627628, "reference_mean_power": 0.0002379882918616051, "output_mean_power": 0.00022928363422246021}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0209734833253925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001318054605596598, "output_power": 0.00012648666654380318, "reference_active": true, "delta_db": -0.17888655749969132}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.132933188997116e-06, "output_power": 7.0226073786690635e-06, "reference_active": true, "delta_db": 0.5883015488055962}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2319303983633163e-07, "output_power": 1.7567004304444068e-07, "reference_active": true, "delta_db": -1.0398293950789503}], "active_band_count": 3, "active_band_mae_db": 0.6023391671280792}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.803319177623983e-05, "log_rms_mae_db": 2.3097879787931266, "rms_envelope_nrmse": 0.18913176662114034, "envelope_correlation": 0.9746448081413417, "rms_gain_error_db": -0.24391258848236613, "reference_mean_power": 3.0517013147816785e-05, "output_mean_power": 2.8850328725811227e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.826326538605559e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1311662288856248e-05, "output_power": 1.0395638897433249e-05, "reference_active": true, "delta_db": -0.3667524527601629}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5475498743040963e-06, "output_power": 2.2749391667607283e-06, "reference_active": true, "delta_db": -0.49152907248446553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.732573657220471e-07, "output_power": 8.68033563676121e-07, "reference_active": true, "delta_db": 1.8018687443608683}], "active_band_count": 3, "active_band_mae_db": 0.8867167565351656}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.459177046887288e-05, "log_rms_mae_db": 1.467348054010481, "rms_envelope_nrmse": 0.12040579318245657, "envelope_correlation": 0.9894759448421049, "rms_gain_error_db": -0.24400972599891071, "reference_mean_power": 3.0668913834731335e-05, "output_mean_power": 2.8993284875404965e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6363060360497894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1094596811060155e-05, "output_power": 1.0250280980998175e-05, "reference_active": true, "delta_db": -0.34375753703495693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1901156553217825e-06, "output_power": 1.273154688720358e-06, "reference_active": true, "delta_db": 0.29292005646047803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.41782955712918e-08, "output_power": 6.879332036432784e-08, "reference_active": true, "delta_db": 1.9233731504214364}], "active_band_count": 3, "active_band_mae_db": 0.8533502479722905}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8767845500971785, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a93d04d646c600c2b1e30ca9eed52a44ca5a83b42fd815cf911876f7b17e85a6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9808439925979637, "standard_deviation": 0.003109215109709981, "minimum": 0.9752273771757363, "maximum": 0.9854484095266137, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.872279293119346, "standard_deviation": 0.02995773524203572, "minimum": 1.8236613213450235, "maximum": 1.9053489528249838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4483887763484032, "standard_deviation": 0.12248529156319696, "minimum": 0.1888958401152037, "maximum": 0.5549601814768942, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16189939424546307, "standard_deviation": 0.013104687022998728, "minimum": 0.1413635831541276, "maximum": 0.18369358562681556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9920601310024697, "standard_deviation": 0.0016165112100871763, "minimum": 0.9893599055261804, "maximum": 0.9942915075616826, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0610890152901264, "standard_deviation": 0.04310632073659293, "minimum": 0.9776601996483772, "maximum": 1.1120087195137374, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6622847500442177, "standard_deviation": 0.12607860240580923, "minimum": 0.4949138834071796, "maximum": 0.8471290277610364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10302023684269106, "standard_deviation": 0.010850350645546248, "minimum": 0.0870078247343152, "maximum": 0.11806911198690426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9713152952385035, "standard_deviation": 0.0019985860368588017, "minimum": 0.9692467988076243, "maximum": 0.9746448081413417, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.297099181345801, "standard_deviation": 0.016258979419633057, "minimum": 2.2680998877930003, "maximum": 2.318527315537261, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7136572480226032, "standard_deviation": 0.117821273567634, "minimum": 0.5567982247428368, "maximum": 0.8867167565351656, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2016269542711731, "standard_deviation": 0.008064757494683201, "minimum": 0.18913176662114034, "maximum": 0.21274037746356958, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9871158761410566, "standard_deviation": 0.0012627616397066636, "minimum": 0.9858519468273382, "maximum": 0.9894759448421049, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4775127390129095, "standard_deviation": 0.030456327344417975, "minimum": 1.4279960362561803, "maximum": 1.5254515935056687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7988141854926247, "standard_deviation": 0.12896874402017955, "minimum": 0.6596799418316335, "maximum": 1.0437002496951455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13121543073211964, "standard_deviation": 0.0066401363338529145, "minimum": 0.12040579318245657, "maximum": 0.13824017516689238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921894485231275, "standard_deviation": 0.0007805748339046408, "minimum": 0.9912121890607682, "maximum": 0.9936276568836481, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8028766776510565, "standard_deviation": 0.02290520409753955, "minimum": 1.7663434420666395, "maximum": 1.8308044165034938, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48627966699748, "standard_deviation": 0.10751300171995945, "minimum": 0.352170747676613, "maximum": 0.6580147376629011, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1026296755426504, "standard_deviation": 0.005418551189685995, "minimum": 0.09281744785444591, "maximum": 0.10832014959039865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976082025149997, "standard_deviation": 0.0002864483913456192, "minimum": 0.9971986104386358, "maximum": 0.9979784259231742, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0809401094750595, "standard_deviation": 0.020953478940443953, "minimum": 1.0571790116701931, "maximum": 1.1137416307790804, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5719958910447135, "standard_deviation": 0.31432345690887065, "minimum": 0.08103848113326177, "maximum": 0.9104041099861986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05660298828980627, "standard_deviation": 0.004224966334790474, "minimum": 0.05249700119401327, "maximum": 0.06292507639085029, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14682719190234816, "standard_deviation": 0.0256190041766682, "minimum": 0.12024182801837512, "maximum": 0.19353443899450448, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.844730201431613, "standard_deviation": 0.00961246500037497, "minimum": 5.829914990768419, "maximum": 5.855751314204693, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5106089781943696, "standard_deviation": 0.010942157209073237, "minimum": 0.495431515965114, "maximum": 0.5269533140510891, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07237052604091866, "standard_deviation": 0.04995483191856955, "minimum": -0.1414890454100063, "maximum": -0.015170930613627252, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994346747484376, "standard_deviation": 0.0009114766302976731, "minimum": 0.9923675687784532, "maximum": 0.9953577172056735, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0160370118620898, "standard_deviation": 0.019616573780592374, "minimum": 0.9899984643612787, "maximum": 1.04403266354351, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.21940051375068556, "standard_deviation": 0.1207686279297462, "minimum": 0.07628552045446675, "maximum": 0.37790854588097167, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07700969196408339, "standard_deviation": 0.005869706846840026, "minimum": 0.07011945112820103, "maximum": 0.08920881490824846, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998283734776495, "standard_deviation": 0.0004085925980303462, "minimum": 0.9976018643432859, "maximum": 0.9987194198445134, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5299157834448269, "standard_deviation": 0.017159337836714528, "minimum": 0.5060471660574161, "maximum": 0.5466455795111524, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.317803707829241, "standard_deviation": 0.13969598403823807, "minimum": 0.06889736097380425, "maximum": 0.48426266384943023, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041544924228942716, "standard_deviation": 0.004450169445871294, "minimum": 0.03647842234381488, "maximum": 0.048221790533963, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0011", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.706021547317505, "predicted_mos": 2.706021547317505, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "04e9239a89a322b105249c190b9b0a7fe6b2ed23fc2b4b5d67f40bec9b0f2fd8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.020470299758017, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0161590871024906, "全帯域_10ms_rms_envelope_nrmse": 0.07853793568092217, "全帯域_10ms_envelope_correlation": 0.9941038951799829, "全帯域_10ms_modulation_mae_db": 0.10187976235700712, "全帯域_40ms_log_rms_mae_db": 0.4936992116876541, "全帯域_40ms_rms_envelope_nrmse": 0.03981214871794313, "全帯域_40ms_envelope_correlation": 0.9983968319534743, "全帯域_40ms_modulation_mae_db": 0.18745639376550027, "300–1000 Hz_10ms_log_rms_mae_db": 1.708739951263849, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11116250637293294, "300–1000 Hz_10ms_envelope_correlation": 0.9907833644600448, "300–1000 Hz_10ms_modulation_mae_db": 0.30993194908748184, "300–1000 Hz_40ms_log_rms_mae_db": 0.984295444951148, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05923929109191062, "300–1000 Hz_40ms_envelope_correlation": 0.997367464865821, "300–1000 Hz_40ms_modulation_mae_db": 0.42514160883635926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8665508481990272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16790097850753596, "1000–3000 Hz_10ms_envelope_correlation": 0.9793472250544935, "1000–3000 Hz_10ms_modulation_mae_db": 0.4301454801060096, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0422721354604636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1046858743186533, "1000–3000 Hz_40ms_envelope_correlation": 0.9916473972768773, "1000–3000 Hz_40ms_modulation_mae_db": 0.9217807108827708, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2474418728194374, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19691393860621975, "3000–6000 Hz_10ms_envelope_correlation": 0.9736522966303698, "3000–6000 Hz_10ms_modulation_mae_db": 0.6227084589815207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.421944241975149, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1276701500745522, "3000–6000 Hz_40ms_envelope_correlation": 0.9883094957517926, "3000–6000 Hz_40ms_modulation_mae_db": 0.6753216318907967, "mr_stft_db": 5.82454171339359, "band_mae_db": 0.19354429808501447, "rms_gain_error_db": -0.06062368564387727, "output_peak": 0.5464301514911364}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0085349854865429, "全帯域_10ms_rms_envelope_nrmse": 0.07173156176004267, "全帯域_10ms_envelope_correlation": 0.9950771943306183, "全帯域_10ms_modulation_mae_db": 0.3429835362541978, "全帯域_40ms_log_rms_mae_db": 0.520683794565466, "全帯域_40ms_rms_envelope_nrmse": 0.0398048309888705, "全帯域_40ms_envelope_correlation": 0.9983723814912555, "全帯域_40ms_modulation_mae_db": 0.38921520367299456, "300–1000 Hz_10ms_log_rms_mae_db": 1.7073498831190377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09834892082014092, "300–1000 Hz_10ms_envelope_correlation": 0.992735684224498, "300–1000 Hz_10ms_modulation_mae_db": 0.4105367460519292, "300–1000 Hz_40ms_log_rms_mae_db": 0.9839995743932886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05739339354886981, "300–1000 Hz_40ms_envelope_correlation": 0.9973829336636031, "300–1000 Hz_40ms_modulation_mae_db": 0.19928459841618865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8266147082509354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15443727825547018, "1000–3000 Hz_10ms_envelope_correlation": 0.9828474498247848, "1000–3000 Hz_10ms_modulation_mae_db": 0.3576474824907317, "1000–3000 Hz_40ms_log_rms_mae_db": 1.036160134595524, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.096277809190066, "1000–3000 Hz_40ms_envelope_correlation": 0.9934043492810092, "1000–3000 Hz_40ms_modulation_mae_db": 0.36588074624876415, "3000–6000 Hz_10ms_log_rms_mae_db": 2.211317052427791, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19522547419101532, "3000–6000 Hz_10ms_envelope_correlation": 0.9728548156462775, "3000–6000 Hz_10ms_modulation_mae_db": 0.9620457534475987, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3817823464574108, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12274687309899977, "3000–6000 Hz_40ms_envelope_correlation": 0.9886326180228008, "3000–6000 Hz_40ms_modulation_mae_db": 0.8470861015235368, "mr_stft_db": 5.82517968419783, "band_mae_db": 0.14642161764465073, "rms_gain_error_db": -0.04274931554294663, "output_peak": 0.5122106659542183}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0356474796505637, "全帯域_10ms_rms_envelope_nrmse": 0.07475608437399978, "全帯域_10ms_envelope_correlation": 0.9948340558210519, "全帯域_10ms_modulation_mae_db": 0.20244989504160657, "全帯域_40ms_log_rms_mae_db": 0.5682823293049124, "全帯域_40ms_rms_envelope_nrmse": 0.0434973010984172, "全帯域_40ms_envelope_correlation": 0.9983012409213071, "全帯域_40ms_modulation_mae_db": 0.23766126678427035, "300–1000 Hz_10ms_log_rms_mae_db": 1.7427065152354653, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10015025178516966, "300–1000 Hz_10ms_envelope_correlation": 0.9927647748658865, "300–1000 Hz_10ms_modulation_mae_db": 0.25193235534495595, "300–1000 Hz_40ms_log_rms_mae_db": 1.0409950166748398, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06196904897988404, "300–1000 Hz_40ms_envelope_correlation": 0.9973348511040623, "300–1000 Hz_40ms_modulation_mae_db": 0.6078165664549054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8827084906149016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15348592205746137, "1000–3000 Hz_10ms_envelope_correlation": 0.9832170354206473, "1000–3000 Hz_10ms_modulation_mae_db": 0.5468050072049485, "1000–3000 Hz_40ms_log_rms_mae_db": 1.047950739068945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08475286745112444, "1000–3000 Hz_40ms_envelope_correlation": 0.9953078232549845, "1000–3000 Hz_40ms_modulation_mae_db": 0.3241061167730702, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2504430963549886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20936801060666854, "3000–6000 Hz_10ms_envelope_correlation": 0.9691095597672039, "3000–6000 Hz_10ms_modulation_mae_db": 0.6354825896387767, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4678040806297747, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1431201690478245, "3000–6000 Hz_40ms_envelope_correlation": 0.9844563318793117, "3000–6000 Hz_40ms_modulation_mae_db": 0.8656892124665876, "mr_stft_db": 5.836935067053363, "band_mae_db": 0.13483146786302555, "rms_gain_error_db": -0.146144842144237, "output_peak": 0.5116296884358328}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9601773705723882, "全帯域_10ms_rms_envelope_nrmse": 0.07198465725104589, "全帯域_10ms_envelope_correlation": 0.9950378393870853, "全帯域_10ms_modulation_mae_db": 0.18461459862832985, "全帯域_40ms_log_rms_mae_db": 0.47816191649467793, "全帯域_40ms_rms_envelope_nrmse": 0.03630347178148769, "全帯域_40ms_envelope_correlation": 0.9986347426535058, "全帯域_40ms_modulation_mae_db": 0.2118051585355548, "300–1000 Hz_10ms_log_rms_mae_db": 1.7302700662545754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10053130519402363, "300–1000 Hz_10ms_envelope_correlation": 0.9924156866270675, "300–1000 Hz_10ms_modulation_mae_db": 0.3640239578848079, "300–1000 Hz_40ms_log_rms_mae_db": 1.0375823268118654, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05461930526016974, "300–1000 Hz_40ms_envelope_correlation": 0.9976252735947905, "300–1000 Hz_40ms_modulation_mae_db": 0.5456465722214919, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7747066352840783, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14395791518384754, "1000–3000 Hz_10ms_envelope_correlation": 0.9851577665720619, "1000–3000 Hz_10ms_modulation_mae_db": 0.3516203294883414, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0037178863244238, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09319976302760576, "1000–3000 Hz_40ms_envelope_correlation": 0.9937829115750895, "1000–3000 Hz_40ms_modulation_mae_db": 0.5951761362920008, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2408480445290913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20987561389805107, "3000–6000 Hz_10ms_envelope_correlation": 0.970535198327229, "3000–6000 Hz_10ms_modulation_mae_db": 1.1436943290430566, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3897344879067235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13908604882216574, "3000–6000 Hz_40ms_envelope_correlation": 0.9864369142195607, "3000–6000 Hz_40ms_modulation_mae_db": 1.3397614285048, "mr_stft_db": 5.813245748315144, "band_mae_db": 0.15020133031033903, "rms_gain_error_db": -0.016633008525652405, "output_peak": 0.5240675022188944}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0527343408077696, "全帯域_10ms_rms_envelope_nrmse": 0.07366353960575547, "全帯域_10ms_envelope_correlation": 0.994903404171266, "全帯域_10ms_modulation_mae_db": 0.32126694296748515, "全帯域_40ms_log_rms_mae_db": 0.5575249233267867, "全帯域_40ms_rms_envelope_nrmse": 0.03852333617699332, "全帯域_40ms_envelope_correlation": 0.9986180658720656, "全帯域_40ms_modulation_mae_db": 0.418387665229865, "300–1000 Hz_10ms_log_rms_mae_db": 1.7358018905171444, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09628659718800944, "300–1000 Hz_10ms_envelope_correlation": 0.9932155563693875, "300–1000 Hz_10ms_modulation_mae_db": 0.35315466444651494, "300–1000 Hz_40ms_log_rms_mae_db": 1.0023515458205723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05246295687638903, "300–1000 Hz_40ms_envelope_correlation": 0.9980657328879239, "300–1000 Hz_40ms_modulation_mae_db": 0.5630646157495459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8636538021947062, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15047491281107322, "1000–3000 Hz_10ms_envelope_correlation": 0.9837008706521411, "1000–3000 Hz_10ms_modulation_mae_db": 0.22678486735031012, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9876522272555467, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09506088518326712, "1000–3000 Hz_40ms_envelope_correlation": 0.9935065942200608, "1000–3000 Hz_40ms_modulation_mae_db": 0.5007086343654766, "3000–6000 Hz_10ms_log_rms_mae_db": 2.263647009190761, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2139835372390764, "3000–6000 Hz_10ms_envelope_correlation": 0.9673684441812377, "3000–6000 Hz_10ms_modulation_mae_db": 0.980581498189241, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4418391890396303, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13764216596923293, "3000–6000 Hz_40ms_envelope_correlation": 0.9854800620687909, "3000–6000 Hz_40ms_modulation_mae_db": 0.789240016624745, "mr_stft_db": 5.815508097330966, "band_mae_db": 0.21137219924242814, "rms_gain_error_db": -0.11321090950044282, "output_peak": 0.5057740807691176}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0035047994354807, "全帯域_10ms_rms_envelope_nrmse": 0.07030362400330231, "全帯域_10ms_envelope_correlation": 0.9952983501717342, "全帯域_10ms_modulation_mae_db": 0.15404808907593331, "全帯域_40ms_log_rms_mae_db": 0.48273293261916955, "全帯域_40ms_rms_envelope_nrmse": 0.038544826739029356, "全帯域_40ms_envelope_correlation": 0.9985033908817735, "全帯域_40ms_modulation_mae_db": 0.1604561052283581, "300–1000 Hz_10ms_log_rms_mae_db": 1.7157980600994076, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09691124758474519, "300–1000 Hz_10ms_envelope_correlation": 0.9930167381738882, "300–1000 Hz_10ms_modulation_mae_db": 0.4178515121576702, "300–1000 Hz_40ms_log_rms_mae_db": 1.0170818704035072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05533360179341189, "300–1000 Hz_40ms_envelope_correlation": 0.9976685197153062, "300–1000 Hz_40ms_modulation_mae_db": 0.7075307347007799, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8136761739161698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15851645081453963, "1000–3000 Hz_10ms_envelope_correlation": 0.9817350100279817, "1000–3000 Hz_10ms_modulation_mae_db": 0.4465760037392675, "1000–3000 Hz_40ms_log_rms_mae_db": 1.017395257602131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0922321176250114, "1000–3000 Hz_40ms_envelope_correlation": 0.9937817389479439, "1000–3000 Hz_40ms_modulation_mae_db": 0.5447387652388241, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2528667216777722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19656154895406994, "3000–6000 Hz_10ms_envelope_correlation": 0.9724824760192666, "3000–6000 Hz_10ms_modulation_mae_db": 1.0391478231017077, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4352375979102414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12757669858063045, "3000–6000 Hz_40ms_envelope_correlation": 0.9878703118120139, "3000–6000 Hz_40ms_modulation_mae_db": 0.9590665992727351, "mr_stft_db": 5.825828776250738, "band_mae_db": 0.15890909162318412, "rms_gain_error_db": -0.0570176092123113, "output_peak": 0.5059916625338896}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9977820832025975, "全帯域_10ms_rms_envelope_nrmse": 0.07525691552072457, "全帯域_10ms_envelope_correlation": 0.9945826030396601, "全帯域_10ms_modulation_mae_db": 0.17528156030675868, "全帯域_40ms_log_rms_mae_db": 0.5080518905894809, "全帯域_40ms_rms_envelope_nrmse": 0.03653608102523929, "全帯域_40ms_envelope_correlation": 0.9986500328736356, "全帯域_40ms_modulation_mae_db": 0.3794799004975699, "300–1000 Hz_10ms_log_rms_mae_db": 1.6841613258570058, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0972540998434423, "300–1000 Hz_10ms_envelope_correlation": 0.9929399679529691, "300–1000 Hz_10ms_modulation_mae_db": 0.6430461482120088, "300–1000 Hz_40ms_log_rms_mae_db": 0.954365792280449, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049497927294205116, "300–1000 Hz_40ms_envelope_correlation": 0.9981231834702053, "300–1000 Hz_40ms_modulation_mae_db": 0.5936806268696649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8221721005610685, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1543082544842052, "1000–3000 Hz_10ms_envelope_correlation": 0.9833870376176389, "1000–3000 Hz_10ms_modulation_mae_db": 0.6949900205694023, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0145102915297477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09579232971519099, "1000–3000 Hz_40ms_envelope_correlation": 0.9941709318583214, "1000–3000 Hz_40ms_modulation_mae_db": 0.7371164068189767, "3000–6000 Hz_10ms_log_rms_mae_db": 2.219350833656446, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19219294857683378, "3000–6000 Hz_10ms_envelope_correlation": 0.973739082289514, "3000–6000 Hz_10ms_modulation_mae_db": 0.8527019027193088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.384572522778885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12922860200559677, "3000–6000 Hz_40ms_envelope_correlation": 0.9874665922230885, "3000–6000 Hz_40ms_modulation_mae_db": 0.7345613501394523, "mr_stft_db": 5.820896263838473, "band_mae_db": 0.1965816614618544, "rms_gain_error_db": -0.04923235799942316, "output_peak": 0.4933709848836422}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.981733169451474, "全帯域_10ms_rms_envelope_nrmse": 0.07366720116186742, "全帯域_10ms_envelope_correlation": 0.9949294028608127, "全帯域_10ms_modulation_mae_db": 0.21073496313103454, "全帯域_40ms_log_rms_mae_db": 0.47572489124717643, "全帯域_40ms_rms_envelope_nrmse": 0.038150700779555964, "全帯域_40ms_envelope_correlation": 0.9986832533060984, "全帯域_40ms_modulation_mae_db": 0.250661843549345, "300–1000 Hz_10ms_log_rms_mae_db": 1.6998227412794185, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10269682672412953, "300–1000 Hz_10ms_envelope_correlation": 0.9922067613487121, "300–1000 Hz_10ms_modulation_mae_db": 0.217251312996923, "300–1000 Hz_40ms_log_rms_mae_db": 0.9690193914236027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05751681904818655, "300–1000 Hz_40ms_envelope_correlation": 0.9975749302192158, "300–1000 Hz_40ms_modulation_mae_db": 0.48860976097661135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8475264921250223, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16363159645241634, "1000–3000 Hz_10ms_envelope_correlation": 0.9816185534711039, "1000–3000 Hz_10ms_modulation_mae_db": 0.5852464028544788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0674970487056272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10920148074478311, "1000–3000 Hz_40ms_envelope_correlation": 0.9925883121024388, "1000–3000 Hz_40ms_modulation_mae_db": 0.9173288871738952, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2321884287834095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1981511962827763, "3000–6000 Hz_10ms_envelope_correlation": 0.9723386267918654, "3000–6000 Hz_10ms_modulation_mae_db": 0.8053553438268687, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4423092716041808, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1253409631517676, "3000–6000 Hz_40ms_envelope_correlation": 0.9880863592377265, "3000–6000 Hz_40ms_modulation_mae_db": 0.7744806613771558, "mr_stft_db": 5.820694198736807, "band_mae_db": 0.18624810094023547, "rms_gain_error_db": -0.12348716384266768, "output_peak": 0.5047674194975562}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.300430775249182, "spectral_convergence": 0.2550960622409437, "spectral_convergence_status": "ok", "amplitude_floor": 1.4412887028114478e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.494151688367627, "spectral_convergence": 0.24824342844618646, "spectral_convergence_status": "ok", "amplitude_floor": 8.082487629299873e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.679042676563964, "spectral_convergence": 0.25622223863955956, "spectral_convergence_status": "ok", "amplitude_floor": 6.705293681874596e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00036448625115270087, "output_power": 0.0003753318232713862, "delta_db": 0.12734237646803034, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008505212390156939, "output_power": 0.0008580651394929171, "delta_db": 0.03835095118177391, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010532918945599765, "output_power": 0.0010047396604082222, "delta_db": -0.20495196511736993, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002198202138287696, "output_power": 0.00021586548519336704, "delta_db": -0.07884417597530342, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.872413022951283e-05, "output_power": 3.551267744651584e-05, "delta_db": -0.3759825520970617, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.441903264074617e-05, "output_power": 2.6382007055761106e-05, "delta_db": 0.3357937676705474, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19354429808501447, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06062368564387727, "pitch": null, "output_rms_dbfs": -25.99307178491945, "output_peak": 0.5464301514911364, "reference_rms_dbfs": -25.93244809927558, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.82454171339359}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019142337741905198, "log_rms_mae_db": 1.0161590871024906, "rms_envelope_nrmse": 0.07853793568092217, "envelope_correlation": 0.9941038951799829, "rms_gain_error_db": -0.060559751447941255, "reference_mean_power": 0.0025533227171819564, "output_mean_power": 0.002517965260904463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1748402838347296e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007147512141697323, "output_power": 0.0006897375321069612, "reference_active": true, "delta_db": -0.15471042989140665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.175142428480134e-05, "output_power": 7.933088330051715e-05, "reference_active": true, "delta_db": -0.13053037823098831}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.431203959050675e-05, "output_power": 1.4244974660213335e-05, "reference_active": true, "delta_db": -0.02039847894862641}], "active_band_count": 3, "active_band_mae_db": 0.10187976235700712}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016640454830055276, "log_rms_mae_db": 0.4936992116876541, "rms_envelope_nrmse": 0.03981214871794313, "envelope_correlation": 0.9983968319534743, "rms_gain_error_db": -0.06053698096532497, "reference_mean_power": 0.002564981061322552, "output_mean_power": 0.0025294754269149142}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0626539543337418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006440432662381922, "output_power": 0.0006232343942745666, "reference_active": true, "delta_db": -0.14263631223756273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1296244809426825e-05, "output_power": 3.9420269956893535e-05, "reference_active": true, "delta_db": -0.20190967682649805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.585289215752233e-07, "output_power": 9.026871271365366e-07, "reference_active": true, "delta_db": 0.21782319223244007}], "active_band_count": 3, "active_band_mae_db": 0.18745639376550027}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016776043588712122, "log_rms_mae_db": 1.708739951263849, "rms_envelope_nrmse": 0.11116250637293294, "envelope_correlation": 0.9907833644600448, "rms_gain_error_db": -0.14799757556212548, "reference_mean_power": 0.0015150798695545982, "output_mean_power": 0.0014643192454824954}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.503421266374213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005710315291680691, "output_power": 0.0005395596699990977, "reference_active": true, "delta_db": -0.24620607842518946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462070521392158e-05, "output_power": 5.643654070679692e-05, "reference_active": true, "delta_db": -0.5881130751900967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.82379092220564e-06, "output_power": 1.004215175955679e-05, "reference_active": true, "delta_db": 0.09547669364715922}], "active_band_count": 3, "active_band_mae_db": 0.30993194908748184}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015135634256581334, "log_rms_mae_db": 0.984295444951148, "rms_envelope_nrmse": 0.05923929109191062, "envelope_correlation": 0.997367464865821, "rms_gain_error_db": -0.14798958097711454, "reference_mean_power": 0.0015225180811750088, "output_mean_power": 0.0014715109590508748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.758019890177841e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005253930575206644, "output_power": 0.0004975668207305419, "reference_active": true, "delta_db": -0.23632917203126785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0513708673138778e-05, "output_power": 2.6814647708793322e-05, "reference_active": true, "delta_db": -0.5612289995842498}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.334657094067365e-07, "output_power": 7.466224959796816e-07, "reference_active": true, "delta_db": -0.47786665489356017}], "active_band_count": 3, "active_band_mae_db": 0.42514160883635926}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.526806573516121e-05, "log_rms_mae_db": 1.8665508481990272, "rms_envelope_nrmse": 0.16790097850753596, "envelope_correlation": 0.9793472250544935, "rms_gain_error_db": -0.17557112042576403, "reference_mean_power": 0.00023682053730462446, "output_mean_power": 0.00022743759277038297}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1631666261710027e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014015200472863968, "output_power": 0.00012445081204125744, "reference_active": true, "delta_db": -0.5160157954720344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3136457414027934e-05, "output_power": 1.3503075941817224e-05, "reference_active": true, "delta_db": 0.11954447896008286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.230915915111415e-06, "output_power": 2.7786768255241028e-06, "reference_active": true, "delta_db": -0.6548761658859117}], "active_band_count": 3, "active_band_mae_db": 0.4301454801060096}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.301677625713533e-05, "log_rms_mae_db": 1.0422721354604636, "rms_envelope_nrmse": 0.1046858743186533, "envelope_correlation": 0.9916473972768773, "rms_gain_error_db": -0.17587991275709744, "reference_mean_power": 0.0002379882918616051, "output_mean_power": 0.00022854282988136362}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0209734833253925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001318054605596598, "output_power": 0.00011776623812629225, "reference_active": true, "delta_db": -0.4891260062703212}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.132933188997116e-06, "output_power": 6.576638692712515e-06, "reference_active": true, "delta_db": 0.30335749658134425}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2319303983633163e-07, "output_power": 1.4170813560461244e-07, "reference_active": true, "delta_db": -1.9728586297966468}], "active_band_count": 3, "active_band_mae_db": 0.9217807108827708}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.803319177623983e-05, "log_rms_mae_db": 2.2474418728194374, "rms_envelope_nrmse": 0.19691393860621975, "envelope_correlation": 0.9736522966303698, "rms_gain_error_db": 0.2096116946606868, "reference_mean_power": 3.0517013147816785e-05, "output_mean_power": 3.20260364989446e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.826326538605559e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1311662288856248e-05, "output_power": 1.126235237018874e-05, "reference_active": true, "delta_db": -0.018973195476508442}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5475498743040963e-06, "output_power": 2.425480751109801e-06, "reference_active": true, "delta_db": -0.21324862736119024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.732573657220471e-07, "output_power": 8.35487581557394e-07, "reference_active": true, "delta_db": 1.6359035541068636}], "active_band_count": 3, "active_band_mae_db": 0.6227084589815207}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.459177046887288e-05, "log_rms_mae_db": 1.421944241975149, "rms_envelope_nrmse": 0.1276701500745522, "envelope_correlation": 0.9883094957517926, "rms_gain_error_db": 0.20959785744861736, "reference_mean_power": 3.0668913834731335e-05, "output_mean_power": 3.2185345913998106e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6363060360497894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1094596811060155e-05, "output_power": 1.1078700314893214e-05, "reference_active": true, "delta_db": -0.006227095330451378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1901156553217825e-06, "output_power": 1.337343393120742e-06, "reference_active": true, "delta_db": 0.506537683601815}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.41782955712918e-08, "output_power": 6.259346133968648e-08, "reference_active": true, "delta_db": 1.5132001167401237}], "active_band_count": 3, "active_band_mae_db": 0.6753216318907967}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8730923781871583, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b3958fb6e547b5482f5ad84ae0eb744c07f4e602578ea89f9018d8df3dfb4fa1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826263685801067, "standard_deviation": 0.0017362482433748445, "minimum": 0.9793472250544935, "maximum": 0.9851577665720619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8372011563932387, "standard_deviation": 0.03491952437414327, "minimum": 1.7747066352840783, "maximum": 1.8827084906149016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.45497694922543624, "standard_deviation": 0.14924726229104168, "minimum": 0.22678486735031012, "maximum": 0.6949900205694023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15583916357081867, "standard_deviation": 0.007494409505973667, "minimum": 0.14395791518384754, "maximum": 0.16790097850753596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9935237573145906, "standard_deviation": 0.0010787412777314355, "minimum": 0.9916473972768773, "maximum": 0.9953078232549845, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.027144465067801, "standard_deviation": 0.026006789165354007, "minimum": 0.9876522272555467, "maximum": 1.0674970487056272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6133545504742224, "standard_deviation": 0.22849811321518135, "minimum": 0.3241061167730702, "maximum": 0.9217807108827708, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09640039090696277, "standard_deviation": 0.0075401733620167315, "minimum": 0.08475286745112444, "maximum": 0.10920148074478311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9715100624566204, "standard_deviation": 0.0022948370616861415, "minimum": 0.9673684441812377, "maximum": 0.973739082289514, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.239762882429962, "standard_deviation": 0.017726343920497455, "minimum": 2.211317052427791, "maximum": 2.263647009190761, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8802147123685099, "standard_deviation": 0.18670649818549084, "minimum": 0.6227084589815207, "maximum": 1.1436943290430566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2015340335443389, "standard_deviation": 0.008198532615363867, "minimum": 0.19219294857683378, "maximum": 0.2139835372390764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9870923356518857, "standard_deviation": 0.0014911580119731443, "minimum": 0.9844563318793117, "maximum": 0.9886326180228008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4206529672877495, "standard_deviation": 0.03190193421319073, "minimum": 1.3817823464574108, "maximum": 1.4678040806297747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8731508752249761, "standard_deviation": 0.2073514195009418, "minimum": 0.6753216318907967, "maximum": 1.3397614285048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13155145884384623, "standard_deviation": 0.007368385584912082, "minimum": 0.12274687309899977, "maximum": 0.1431201690478245, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925098167528067, "standard_deviation": 0.000768805020838287, "minimum": 0.9907833644600448, "maximum": 0.9932155563693875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.715581304203238, "standard_deviation": 0.01967799222463191, "minimum": 1.6841613258570058, "maximum": 1.7427065152354653, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3709660807727865, "standard_deviation": 0.13075577386784154, "minimum": 0.217251312996923, "maximum": 0.6430461482120088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1004177194390742, "standard_deviation": 0.004845898848222613, "minimum": 0.09628659718800944, "maximum": 0.11116250637293294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997642861190116, "standard_deviation": 0.0003055178118197794, "minimum": 0.9973348511040623, "maximum": 0.9981231834702053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9987113703449091, "standard_deviation": 0.03144606244876699, "minimum": 0.954365792280449, "maximum": 1.0409950166748398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5163468855281934, "standard_deviation": 0.15285648577483685, "minimum": 0.19928459841618865, "maximum": 0.7075307347007799, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05600404298662835, "standard_deviation": 0.003918868922232864, "minimum": 0.049497927294205116, "maximum": 0.06196904897988404, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17226372089634148, "standard_deviation": 0.028039116125266644, "minimum": 0.13483146786302555, "maximum": 0.21137219924242814, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.822853693639614, "standard_deviation": 0.007271875221666901, "minimum": 5.813245748315144, "maximum": 5.836935067053363, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.513030269473036, "standard_deviation": 0.01602213836769006, "minimum": 0.4933709848836422, "maximum": 0.5464301514911364, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0761373615514448, "standard_deviation": 0.0455178881706139, "minimum": -0.146144842144237, "maximum": -0.016633008525652405, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9948458431202765, "standard_deviation": 0.00036358632537845313, "minimum": 0.9941038951799829, "maximum": 0.9952983501717342, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0070341644636636, "standard_deviation": 0.02909713169053513, "minimum": 0.9601773705723882, "maximum": 1.0527343408077696, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.21165741847029415, "standard_deviation": 0.08174080311864576, "minimum": 0.10187976235700712, "maximum": 0.3429835362541978, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07373768991970754, "standard_deviation": 0.0025435768527812008, "minimum": 0.07030362400330231, "maximum": 0.07853793568092217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985199924941395, "standard_deviation": 0.00014708402392246355, "minimum": 0.9983012409213071, "maximum": 0.9986832533060984, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5106077362294155, "standard_deviation": 0.03579323040720667, "minimum": 0.47572489124717643, "maximum": 0.5682823293049124, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2793904421579323, "standard_deviation": 0.10080481756250319, "minimum": 0.1604561052283581, "maximum": 0.418387665229865, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038896587163442055, "standard_deviation": 0.00226606529105722, "minimum": 0.03630347178148769, "maximum": 0.0434973010984172, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1089-134686-0032", "language": "en", "style": "normal", "speaker": "librispeech_1089", "source_utterance_id": "1089-134686-0032", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 65440, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1089-134686-0032.wav", "source_file_sha256": "c480d3d43e31e24fb4851550ef850b8899f3177462cb2fcdad79c88acbb5a361", "decoded_pcm_sha256": "643f787b905d5160255daf650b50e8a272ea88d166a6e7aca2a7ee357cf232fb", "original_source_sha256": "ba1d2185ad72cd3142e6309865859a8e9e5dfdc91ab046adfcc07338c5a0f0f0", "corpus_entry_id": "librispeech_1089-134686-0032"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.574302673339844, "predicted_mos": 4.574302673339844, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "bf9b5af891930f0a9556716b887be2e7c74c73daff10b5442892fda8a4088950", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.537529000081122, "cpu_threads": 4}, "waveform_sha256": "cf1d3a15f9864fa36824ce1cb46c161624f17c4e82ba1e8b14109293167293b6"}, "rows": [{"case": "librispeech_1089-134686-0032", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.047332525253296, "predicted_mos": 3.047332525253296, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "c3b6741a4c23c9ead0f4da722046d84e0c959421785ebbe9a1e8a2fe94036f47", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.750029400456697, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2653138220661322, "全帯域_10ms_rms_envelope_nrmse": 0.08491180640442397, "全帯域_10ms_envelope_correlation": 0.993652366360229, "全帯域_10ms_modulation_mae_db": 0.3029398633369284, "全帯域_40ms_log_rms_mae_db": 0.6878562269841925, "全帯域_40ms_rms_envelope_nrmse": 0.04284700791518867, "全帯域_40ms_envelope_correlation": 0.9982807601887297, "全帯域_40ms_modulation_mae_db": 0.07802012063780511, "300–1000 Hz_10ms_log_rms_mae_db": 1.327059171511584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1094698937320158, "300–1000 Hz_10ms_envelope_correlation": 0.9912910485980952, "300–1000 Hz_10ms_modulation_mae_db": 0.5922259994019062, "300–1000 Hz_40ms_log_rms_mae_db": 0.9272853067460387, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05655742207348494, "300–1000 Hz_40ms_envelope_correlation": 0.9975642606724682, "300–1000 Hz_40ms_modulation_mae_db": 0.30885530996561794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8311923029923274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23284096771326426, "1000–3000 Hz_10ms_envelope_correlation": 0.9570807700027005, "1000–3000 Hz_10ms_modulation_mae_db": 1.4207593242918088, "1000–3000 Hz_40ms_log_rms_mae_db": 1.108373594328475, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14345547747701348, "1000–3000 Hz_40ms_envelope_correlation": 0.9827156815822214, "1000–3000 Hz_40ms_modulation_mae_db": 1.3751408584663685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8932116452673078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32940087988116584, "3000–6000 Hz_10ms_envelope_correlation": 0.924303798798235, "3000–6000 Hz_10ms_modulation_mae_db": 3.556564185474974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.324678186699503, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2607311118393259, "3000–6000 Hz_40ms_envelope_correlation": 0.9491516791734556, "3000–6000 Hz_40ms_modulation_mae_db": 4.643919239375975, "mr_stft_db": 5.843034061797712, "band_mae_db": 0.22809505658873297, "rms_gain_error_db": -0.004594751560468198, "output_peak": 0.40215556821936427}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3413503406342375, "全帯域_10ms_rms_envelope_nrmse": 0.09182642376361953, "全帯域_10ms_envelope_correlation": 0.9925750665363747, "全帯域_10ms_modulation_mae_db": 0.43364634395890667, "全帯域_40ms_log_rms_mae_db": 0.6692459484519457, "全帯域_40ms_rms_envelope_nrmse": 0.04290541022694746, "全帯域_40ms_envelope_correlation": 0.9982802215702797, "全帯域_40ms_modulation_mae_db": 0.38100239677879993, "300–1000 Hz_10ms_log_rms_mae_db": 1.3464091733319103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10737354148388302, "300–1000 Hz_10ms_envelope_correlation": 0.9916393503484157, "300–1000 Hz_10ms_modulation_mae_db": 0.7302541504187103, "300–1000 Hz_40ms_log_rms_mae_db": 0.946019931593915, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044439038415247135, "300–1000 Hz_40ms_envelope_correlation": 0.9985807240707684, "300–1000 Hz_40ms_modulation_mae_db": 0.28102377183248306, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8179704076133867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22937202982132612, "1000–3000 Hz_10ms_envelope_correlation": 0.9590030278410834, "1000–3000 Hz_10ms_modulation_mae_db": 1.0966515184394479, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1000844725280283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13311648295460052, "1000–3000 Hz_40ms_envelope_correlation": 0.9852260887288019, "1000–3000 Hz_40ms_modulation_mae_db": 0.8917389821251667, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9045751388258827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3219880083076317, "3000–6000 Hz_10ms_envelope_correlation": 0.927681807476401, "3000–6000 Hz_10ms_modulation_mae_db": 3.4828139722104403, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3125970468040482, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25967209203515057, "3000–6000 Hz_40ms_envelope_correlation": 0.9495180415277127, "3000–6000 Hz_40ms_modulation_mae_db": 4.517890136407076, "mr_stft_db": 5.81588045813258, "band_mae_db": 0.10482689402292393, "rms_gain_error_db": -0.026813361119732943, "output_peak": 0.4264560457753801}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2886015464115586, "全帯域_10ms_rms_envelope_nrmse": 0.08806072036092852, "全帯域_10ms_envelope_correlation": 0.9931765512031274, "全帯域_10ms_modulation_mae_db": 0.8591427889984296, "全帯域_40ms_log_rms_mae_db": 0.7408234909866099, "全帯域_40ms_rms_envelope_nrmse": 0.04768669318596424, "全帯域_40ms_envelope_correlation": 0.9979116912112942, "全帯域_40ms_modulation_mae_db": 0.30002070999201114, "300–1000 Hz_10ms_log_rms_mae_db": 1.3489316192070464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11646540300889079, "300–1000 Hz_10ms_envelope_correlation": 0.9901355088012994, "300–1000 Hz_10ms_modulation_mae_db": 1.231065730593018, "300–1000 Hz_40ms_log_rms_mae_db": 0.9424826265360642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06491911447277242, "300–1000 Hz_40ms_envelope_correlation": 0.9968228868983937, "300–1000 Hz_40ms_modulation_mae_db": 0.522800142152099, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8498516938907588, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25357539651031025, "1000–3000 Hz_10ms_envelope_correlation": 0.949163030337961, "1000–3000 Hz_10ms_modulation_mae_db": 1.5214924123203468, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0445496209049059, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1534275990445808, "1000–3000 Hz_40ms_envelope_correlation": 0.9801520270435301, "1000–3000 Hz_40ms_modulation_mae_db": 1.5185261905708718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8608349877887402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32967310973362773, "3000–6000 Hz_10ms_envelope_correlation": 0.9245321677464563, "3000–6000 Hz_10ms_modulation_mae_db": 3.5457300598004493, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3262085418248413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2674056148574297, "3000–6000 Hz_40ms_envelope_correlation": 0.9468776327484603, "3000–6000 Hz_40ms_modulation_mae_db": 4.553240151534439, "mr_stft_db": 5.826141363949385, "band_mae_db": 0.27637703164384375, "rms_gain_error_db": -0.06763068076848026, "output_peak": 0.44901921080628576}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3000301754603931, "全帯域_10ms_rms_envelope_nrmse": 0.09777095572071269, "全帯域_10ms_envelope_correlation": 0.9915800467131436, "全帯域_10ms_modulation_mae_db": 0.7222877383465899, "全帯域_40ms_log_rms_mae_db": 0.7083150722326611, "全帯域_40ms_rms_envelope_nrmse": 0.04144508648350835, "全帯域_40ms_envelope_correlation": 0.9984684442528863, "全帯域_40ms_modulation_mae_db": 0.2727852894331933, "300–1000 Hz_10ms_log_rms_mae_db": 1.3547718401042779, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12469900705008885, "300–1000 Hz_10ms_envelope_correlation": 0.9887004834922342, "300–1000 Hz_10ms_modulation_mae_db": 1.2237421286855965, "300–1000 Hz_40ms_log_rms_mae_db": 0.9441108379821453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059516400824779325, "300–1000 Hz_40ms_envelope_correlation": 0.9974514566095443, "300–1000 Hz_40ms_modulation_mae_db": 0.41113098543264764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7501943608348067, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2295848513153337, "1000–3000 Hz_10ms_envelope_correlation": 0.9581904720412008, "1000–3000 Hz_10ms_modulation_mae_db": 1.2968334703403313, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0985940794861782, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14114288459413324, "1000–3000 Hz_40ms_envelope_correlation": 0.9833304072679281, "1000–3000 Hz_40ms_modulation_mae_db": 1.1458562760511244, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8062573529483097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31311764625702426, "3000–6000 Hz_10ms_envelope_correlation": 0.9311129217935189, "3000–6000 Hz_10ms_modulation_mae_db": 3.2813013269906612, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2223646139880793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25537214672561526, "3000–6000 Hz_40ms_envelope_correlation": 0.9510593849737083, "3000–6000 Hz_40ms_modulation_mae_db": 4.3203542337363245, "mr_stft_db": 5.81791136333532, "band_mae_db": 0.17447337423665435, "rms_gain_error_db": -0.0844224096607432, "output_peak": 0.4169209124041627}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.242311063107516, "全帯域_10ms_rms_envelope_nrmse": 0.07830074235455245, "全帯域_10ms_envelope_correlation": 0.9946862274168674, "全帯域_10ms_modulation_mae_db": 0.43246077870846866, "全帯域_40ms_log_rms_mae_db": 0.6429538995060919, "全帯域_40ms_rms_envelope_nrmse": 0.04622168207449303, "全帯域_40ms_envelope_correlation": 0.9981409528821709, "全帯域_40ms_modulation_mae_db": 0.8352069779339102, "300–1000 Hz_10ms_log_rms_mae_db": 1.3829745167580203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10239400499406388, "300–1000 Hz_10ms_envelope_correlation": 0.9926432512140303, "300–1000 Hz_10ms_modulation_mae_db": 0.5550222488611171, "300–1000 Hz_40ms_log_rms_mae_db": 0.9778749976824911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06829119714395206, "300–1000 Hz_40ms_envelope_correlation": 0.9968162087813743, "300–1000 Hz_40ms_modulation_mae_db": 1.0356135281164114, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7303680267324668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21841365480436523, "1000–3000 Hz_10ms_envelope_correlation": 0.9631324914388063, "1000–3000 Hz_10ms_modulation_mae_db": 0.6203320473588386, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0686139845448386, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14501321952667093, "1000–3000 Hz_40ms_envelope_correlation": 0.982700584211435, "1000–3000 Hz_40ms_modulation_mae_db": 0.26041971212838577, "3000–6000 Hz_10ms_log_rms_mae_db": 1.89021879701857, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31440853048088363, "3000–6000 Hz_10ms_envelope_correlation": 0.929733378578849, "3000–6000 Hz_10ms_modulation_mae_db": 3.1792198660606825, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2979519661194232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25615359185505576, "3000–6000 Hz_40ms_envelope_correlation": 0.9500847985771291, "3000–6000 Hz_40ms_modulation_mae_db": 4.213324216919616, "mr_stft_db": 5.825174290863668, "band_mae_db": 0.24387784943925137, "rms_gain_error_db": -0.11516710923691428, "output_peak": 0.4035462461535144}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2765099202515249, "全帯域_10ms_rms_envelope_nrmse": 0.08663736922611094, "全帯域_10ms_envelope_correlation": 0.9934236135463085, "全帯域_10ms_modulation_mae_db": 0.3552026784709474, "全帯域_40ms_log_rms_mae_db": 0.6821626668052192, "全帯域_40ms_rms_envelope_nrmse": 0.04718135735201184, "全帯域_40ms_envelope_correlation": 0.9979307070242998, "全帯域_40ms_modulation_mae_db": 0.27942605205300275, "300–1000 Hz_10ms_log_rms_mae_db": 1.3877075559369592, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1028140563319715, "300–1000 Hz_10ms_envelope_correlation": 0.9923338165845887, "300–1000 Hz_10ms_modulation_mae_db": 0.4137187387247581, "300–1000 Hz_40ms_log_rms_mae_db": 1.0727420991464527, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05747769063706521, "300–1000 Hz_40ms_envelope_correlation": 0.9974854778150357, "300–1000 Hz_40ms_modulation_mae_db": 0.30117075124461834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7979202421265128, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2161626454259378, "1000–3000 Hz_10ms_envelope_correlation": 0.9632107751063254, "1000–3000 Hz_10ms_modulation_mae_db": 0.9567636403205974, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1088614982892153, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14613365695701777, "1000–3000 Hz_40ms_envelope_correlation": 0.9821117069693744, "1000–3000 Hz_40ms_modulation_mae_db": 0.5756275005236886, "3000–6000 Hz_10ms_log_rms_mae_db": 2.001084911573628, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3240486594603257, "3000–6000 Hz_10ms_envelope_correlation": 0.9262102006353516, "3000–6000 Hz_10ms_modulation_mae_db": 3.372121889253453, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3774855090815041, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25938845844270103, "3000–6000 Hz_40ms_envelope_correlation": 0.9493322170051861, "3000–6000 Hz_40ms_modulation_mae_db": 4.306414218188823, "mr_stft_db": 5.831935977111947, "band_mae_db": 0.15410864839614827, "rms_gain_error_db": 0.01110485519128921, "output_peak": 0.44684811112816314}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2695855160652858, "全帯域_10ms_rms_envelope_nrmse": 0.08383502774221754, "全帯域_10ms_envelope_correlation": 0.9938151236391969, "全帯域_10ms_modulation_mae_db": 0.447093077476086, "全帯域_40ms_log_rms_mae_db": 0.7175801287789967, "全帯域_40ms_rms_envelope_nrmse": 0.052347638686495244, "全帯域_40ms_envelope_correlation": 0.9974673004539492, "全帯域_40ms_modulation_mae_db": 0.7074889695941063, "300–1000 Hz_10ms_log_rms_mae_db": 1.3644734818374427, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10788035261813751, "300–1000 Hz_10ms_envelope_correlation": 0.9915772266322179, "300–1000 Hz_10ms_modulation_mae_db": 0.37358674346088394, "300–1000 Hz_40ms_log_rms_mae_db": 0.9736371682895654, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07269884664132527, "300–1000 Hz_40ms_envelope_correlation": 0.9960676808112509, "300–1000 Hz_40ms_modulation_mae_db": 0.524450079196621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.812149353341706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21835064823860756, "1000–3000 Hz_10ms_envelope_correlation": 0.9638144050815924, "1000–3000 Hz_10ms_modulation_mae_db": 1.0397543048868203, "1000–3000 Hz_40ms_log_rms_mae_db": 1.117887892423992, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13564754019725692, "1000–3000 Hz_40ms_envelope_correlation": 0.9852387210014381, "1000–3000 Hz_40ms_modulation_mae_db": 0.6411795684785557, "3000–6000 Hz_10ms_log_rms_mae_db": 1.896183035845483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.323535500385116, "3000–6000 Hz_10ms_envelope_correlation": 0.9261613395833311, "3000–6000 Hz_10ms_modulation_mae_db": 3.1897416851112177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.312008514349208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25160848292950155, "3000–6000 Hz_40ms_envelope_correlation": 0.9523259841396533, "3000–6000 Hz_40ms_modulation_mae_db": 4.23572745470272, "mr_stft_db": 5.820639514073683, "band_mae_db": 0.1335931511527808, "rms_gain_error_db": -0.06561267306933852, "output_peak": 0.4482311771281752}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.207754196185114, "全帯域_10ms_rms_envelope_nrmse": 0.08328809889873531, "全帯域_10ms_envelope_correlation": 0.994027380470885, "全帯域_10ms_modulation_mae_db": 0.5833499693918692, "全帯域_40ms_log_rms_mae_db": 0.6465566781031573, "全帯域_40ms_rms_envelope_nrmse": 0.044467093137791354, "全帯域_40ms_envelope_correlation": 0.9983601521991425, "全帯域_40ms_modulation_mae_db": 0.19146556719399524, "300–1000 Hz_10ms_log_rms_mae_db": 1.3080045471693407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10755101987546031, "300–1000 Hz_10ms_envelope_correlation": 0.9917434035373756, "300–1000 Hz_10ms_modulation_mae_db": 0.8387506760532334, "300–1000 Hz_40ms_log_rms_mae_db": 0.8932258318413303, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060891702711827124, "300–1000 Hz_40ms_envelope_correlation": 0.9974388648428322, "300–1000 Hz_40ms_modulation_mae_db": 0.227114671013014, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7205079757073682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20837311934949374, "1000–3000 Hz_10ms_envelope_correlation": 0.9658973724812031, "1000–3000 Hz_10ms_modulation_mae_db": 0.9660569872676081, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0331051518287957, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12447692219110637, "1000–3000 Hz_40ms_envelope_correlation": 0.9869647115956595, "1000–3000 Hz_40ms_modulation_mae_db": 1.1533554903070016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.889841365054292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3284512866309529, "3000–6000 Hz_10ms_envelope_correlation": 0.923074831701173, "3000–6000 Hz_10ms_modulation_mae_db": 3.4175808584704783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.311554129264904, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26513929386113194, "3000–6000 Hz_40ms_envelope_correlation": 0.9463743875342489, "3000–6000 Hz_40ms_modulation_mae_db": 4.5093427509773525, "mr_stft_db": 5.8154067039755475, "band_mae_db": 0.16119310108980092, "rms_gain_error_db": -0.13939898134112186, "output_peak": 0.40119664684343337}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65440, "output_samples": 65440, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.290344441698296, "spectral_convergence": 0.2849745977894037, "spectral_convergence_status": "ok", "amplitude_floor": 1.239800719019687e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.404634352628396, "spectral_convergence": 0.25315678749268433, "spectral_convergence_status": "ok", "amplitude_floor": 7.959069981155994e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.834123391066445, "spectral_convergence": 0.28367970023474665, "spectral_convergence_status": "ok", "amplitude_floor": 5.70263552352201e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002917609391844852, "output_power": 0.0003004059824335296, "delta_db": 0.1268142886279859, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009442146735758444, "output_power": 0.0009641108997146976, "delta_db": 0.09056247482003907, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008095018592590047, "output_power": 0.0007832193898352315, "delta_db": -0.14334420000268144, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.140726670679603e-05, "output_power": 6.511482656580911e-05, "delta_db": -0.40062521529546413, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.8558923829027427e-05, "output_power": 2.925701185338883e-05, "delta_db": 0.10488129538034414, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.629692823024492e-06, "output_power": 9.68789947268988e-06, "delta_db": 0.502342865405883, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22809505658873297, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.004594751560468198, "pitch": null, "output_rms_dbfs": -26.67198900062737, "output_peak": 0.40215556821936427, "reference_rms_dbfs": -26.667394249066906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.843034061797712}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65440, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00016079151656388168, "log_rms_mae_db": 1.2653138220661322, "rms_envelope_nrmse": 0.08491180640442397, "envelope_correlation": 0.993652366360229, "rms_gain_error_db": -0.004595211813729829, "reference_mean_power": 0.0021567095381351967, "output_mean_power": 0.0021544287594685288}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.6620838184249172e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000592954618765589, "output_power": 0.0005929575110612054, "reference_active": true, "delta_db": 2.1183829638546746e-05}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014246995030046006, "output_power": 0.00014466980447032732, "reference_active": true, "delta_db": 0.06654621687516565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.93963265073239e-05, "output_power": 2.3547494553361286e-05, "reference_active": true, "delta_db": 0.8422521893059811}], "active_band_count": 3, "active_band_mae_db": 0.3029398633369284}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00012458719833187296, "log_rms_mae_db": 0.6878562269841925, "rms_envelope_nrmse": 0.04284700791518867, "envelope_correlation": 0.9982807601887297, "rms_gain_error_db": -0.0045993960677450105, "reference_mean_power": 0.0021726617564687806, "output_mean_power": 0.0021703620168415875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.5403145620848259e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005607371759141368, "output_power": 0.0005616830666278839, "reference_active": true, "delta_db": 0.007319811228284008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.842815402518841e-05, "output_power": 7.079680770716113e-05, "reference_active": true, "delta_db": 0.14778851130604212}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9268319788025257e-06, "output_power": 1.8921198486737165e-06, "reference_active": true, "delta_db": -0.07895203937908922}], "active_band_count": 3, "active_band_mae_db": 0.07802012063780511}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00013740128966162354, "log_rms_mae_db": 1.327059171511584, "rms_envelope_nrmse": 0.1094698937320158, "envelope_correlation": 0.9912910485980952, "rms_gain_error_db": -0.025726455692512248, "reference_mean_power": 0.001344215530062727, "output_mean_power": 0.0013362762912172445}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.2437342579426783e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004237422220408671, "output_power": 0.00041411486078139575, "reference_active": true, "delta_db": -0.09980923911926232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010753440950326828, "output_power": 0.00011188793421138412, "reference_active": true, "delta_db": 0.17235800901734077}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.157936749643158e-05, "output_power": 1.637328843240133e-05, "reference_active": true, "delta_db": 1.5045107500691157}], "active_band_count": 3, "active_band_mae_db": 0.5922259994019062}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00011336523032200158, "log_rms_mae_db": 0.9272853067460387, "rms_envelope_nrmse": 0.05655742207348494, "envelope_correlation": 0.9975642606724682, "rms_gain_error_db": -0.025726552975183985, "reference_mean_power": 0.0013541602801659725, "output_mean_power": 0.0013461622752384303}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.1658618871138855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004110855381392752, "output_power": 0.00040573096018896857, "reference_active": true, "delta_db": -0.05694049945171374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.54169136496196e-05, "output_power": 4.9815629629770654e-05, "reference_active": true, "delta_db": 0.40148005840065953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3416624074278087e-06, "output_power": 1.4943687892130754e-06, "reference_active": true, "delta_db": 0.4681453720444805}], "active_band_count": 3, "active_band_mae_db": 0.30885530996561794}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 4.166087007180048e-05, "log_rms_mae_db": 1.8311923029923274, "rms_envelope_nrmse": 0.23284096771326426, "envelope_correlation": 0.9570807700027005, "rms_gain_error_db": -0.15811344159929341, "reference_mean_power": 7.742789466080943e-05, "output_mean_power": 7.465967694342912e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.015275941663665e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8994572701205476e-05, "output_power": 3.553263683486879e-05, "reference_active": true, "delta_db": -0.403767285820155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.4522009847535795e-06, "output_power": 6.563204499087216e-06, "reference_active": true, "delta_db": -0.5516862284188143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.182412963771591e-06, "output_power": 2.5319296143604234e-06, "reference_active": true, "delta_db": 3.306824458636457}], "active_band_count": 3, "active_band_mae_db": 1.4207593242918088}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 3.308768412789438e-05, "log_rms_mae_db": 1.108373594328475, "rms_envelope_nrmse": 0.14345547747701348, "envelope_correlation": 0.9827156815822214, "rms_gain_error_db": -0.15811436034174542, "reference_mean_power": 7.800066882867515e-05, "output_mean_power": 7.521195726267993e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 6.505338847209246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.845568227980419e-05, "output_power": 3.540971529836241e-05, "reference_active": true, "delta_db": -0.3583808604955599}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.058614755593344e-06, "output_power": 2.7591327446825334e-06, "reference_active": true, "delta_db": -0.44752184021658253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905753913870012e-08, "output_power": 2.1273528144384793e-07, "reference_active": true, "delta_db": 3.3195198746869634}], "active_band_count": 3, "active_band_mae_db": 1.3751408584663685}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 1.92244611767394e-05, "log_rms_mae_db": 1.8932116452673078, "rms_envelope_nrmse": 0.32940087988116584, "envelope_correlation": 0.924303798798235, "rms_gain_error_db": 0.07984671425951058, "reference_mean_power": 1.4177032937223143e-05, "output_mean_power": 1.4440093980905691e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.931109844663442e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.831691262181727e-06, "output_power": 4.185912020729541e-06, "reference_active": true, "delta_db": 0.3839958750260372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0928531053800205e-06, "output_power": 1.4926764707153584e-06, "reference_active": true, "delta_db": 1.3540389635375694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4875100105629565e-07, "output_power": 1.1631230674508897e-06, "reference_active": true, "delta_db": 8.931657717861315}], "active_band_count": 3, "active_band_mae_db": 3.556564185474974}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 1.4099069206239137e-05, "log_rms_mae_db": 1.324678186699503, "rms_envelope_nrmse": 0.2607311118393259, "envelope_correlation": 0.9491516791734556, "rms_gain_error_db": 0.07984681403150323, "reference_mean_power": 1.4281884811465133e-05, "output_mean_power": 1.4546891761701391e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 7.397985811629507e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.714499411475195e-06, "output_power": 4.733251861829625e-06, "reference_active": true, "delta_db": 1.0525932052154436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.071953545728327e-07, "output_power": 8.396397157885122e-07, "reference_active": true, "delta_db": 2.189177057871004}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3628863245634334e-08, "output_power": 1.597564418717379e-07, "reference_active": true, "delta_db": 10.689987455041479}], "active_band_count": 3, "active_band_mae_db": 4.643919239375975}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8518526574789272, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4d8f79425c73b14b37b3edee49d94676bf494890ba94e5e12b5c8649e7141874", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9599365430413591, "standard_deviation": 0.0053405218254096954, "minimum": 0.949163030337961, "maximum": 0.9658973724812031, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7887692954049168, "standard_deviation": 0.04865649918895923, "minimum": 1.7205079757073682, "maximum": 1.8498516938907588, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1148304631532249, "standard_deviation": 0.29024761793158116, "minimum": 0.6203320473588386, "maximum": 1.5214924123203468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22583416414732982, "standard_deviation": 0.01391170496926616, "minimum": 0.20837311934949374, "maximum": 0.25357539651031025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9835549910500485, "standard_deviation": 0.0021514722690465712, "minimum": 0.9801520270435301, "maximum": 0.9869647115956595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0850087867918035, "standard_deviation": 0.03208795748639323, "minimum": 1.0331051518287957, "maximum": 1.117887892423992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9452305723313954, "standard_deviation": 0.43060390858172604, "minimum": 0.26041971212838577, "maximum": 1.5185261905708718, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14030172286779752, "standard_deviation": 0.008967125737769671, "minimum": 0.12447692219110637, "maximum": 0.1534275990445808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9266013057891644, "standard_deviation": 0.0027690758240867465, "minimum": 0.923074831701173, "maximum": 0.9311129217935189, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8927759042902768, "standard_deviation": 0.05398768854544794, "minimum": 1.8062573529483097, "maximum": 2.001084911573628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.3781342304215447, "standard_deviation": 0.1497192279863298, "minimum": 3.1792198660606825, "maximum": 3.556564185474974, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.32307795264209094, "standard_deviation": 0.0064260008481004036, "minimum": 0.31311764625702426, "maximum": 0.32967310973362773, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9493405157099443, "standard_deviation": 0.0019767376672435973, "minimum": 0.9463743875342489, "maximum": 0.9523259841396533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.310606063516439, "standard_deviation": 0.04287495360528353, "minimum": 1.2223646139880793, "maximum": 1.3774855090815041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.41252655023029, "standard_deviation": 0.16235446622221814, "minimum": 4.213324216919616, "maximum": 4.643919239375975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.25943384906823896, "standard_deviation": 0.005159412055511371, "minimum": 0.25160848292950155, "maximum": 0.2674056148574297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9912580111510321, "standard_deviation": 0.0012739826171832512, "minimum": 0.9887004834922342, "maximum": 0.9926432512140303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3525414882320725, "standard_deviation": 0.02670926139279193, "minimum": 1.3080045471693407, "maximum": 1.3877075559369592, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7447958020249029, "standard_deviation": 0.3341485423308426, "minimum": 0.37358674346088394, "maximum": 1.231065730593018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10983090988681396, "standard_deviation": 0.007414196466349328, "minimum": 0.10239400499406388, "maximum": 0.12469900705008885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972784450627084, "standard_deviation": 0.0007332215542004227, "minimum": 0.9960676808112509, "maximum": 0.9985807240707684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9596723499772504, "standard_deviation": 0.052778806390556356, "minimum": 0.8932258318413303, "maximum": 1.0727420991464527, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.451519904869189, "standard_deviation": 0.2606533009527617, "minimum": 0.227114671013014, "maximum": 1.0356135281164114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060598926615056684, "standard_deviation": 0.008558358931539348, "minimum": 0.044439038415247135, "maximum": 0.07269884664132527, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18456813832126706, "standard_deviation": 0.05900883693257657, "minimum": 0.10482689402292393, "maximum": 0.27637703164384375, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.82451546665498, "standard_deviation": 0.009390344978096508, "minimum": 5.8154067039755475, "maximum": 5.843034061797712, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42429673980730986, "standard_deviation": 0.02140132306944793, "minimum": 0.40119664684343337, "maximum": 0.44901921080628576, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06156688894568876, "standard_deviation": 0.05242838042786297, "minimum": -0.13939898134112186, "maximum": 0.01110485519128921, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933670469857665, "standard_deviation": 0.0009500788373084294, "minimum": 0.9915800467131436, "maximum": 0.9946862274168674, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2739320725227203, "standard_deviation": 0.039498171679480665, "minimum": 1.207754196185114, "maximum": 1.3413503406342375, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5170154048360283, "standard_deviation": 0.1907087776876642, "minimum": 0.3029398633369284, "maximum": 0.8591427889984296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08682889305891262, "standard_deviation": 0.005905967403713966, "minimum": 0.07830074235455245, "maximum": 0.09777095572071269, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998105028722844, "standard_deviation": 0.0003239242097546253, "minimum": 0.9974673004539492, "maximum": 0.9984684442528863, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6869367639811094, "standard_deviation": 0.03422921327563932, "minimum": 0.6429538995060919, "maximum": 0.7408234909866099, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.380677010452103, "standard_deviation": 0.258899769880838, "minimum": 0.07802012063780511, "maximum": 0.8352069779339102, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04563774613280003, "standard_deviation": 0.003508098010523071, "minimum": 0.04144508648350835, "maximum": 0.052347638686495244, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0032", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.1304566860198975, "predicted_mos": 3.1304566860198975, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "1b8b846fb4c752c785d0020db79f44ff40c26e10510bc548cb003161fd024579", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.959065599832684, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.235228947836046, "全帯域_10ms_rms_envelope_nrmse": 0.08291933367357664, "全帯域_10ms_envelope_correlation": 0.9939420317669506, "全帯域_10ms_modulation_mae_db": 0.29964080768849777, "全帯域_40ms_log_rms_mae_db": 0.662434156630392, "全帯域_40ms_rms_envelope_nrmse": 0.04028089924634026, "全帯域_40ms_envelope_correlation": 0.9984833745765024, "全帯域_40ms_modulation_mae_db": 0.12980880046133672, "300–1000 Hz_10ms_log_rms_mae_db": 1.3213806380918958, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10538586266343851, "300–1000 Hz_10ms_envelope_correlation": 0.9919277102496872, "300–1000 Hz_10ms_modulation_mae_db": 0.47006166028611124, "300–1000 Hz_40ms_log_rms_mae_db": 0.9019097239428031, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053090645031119356, "300–1000 Hz_40ms_envelope_correlation": 0.997856706982139, "300–1000 Hz_40ms_modulation_mae_db": 0.09168243827549971, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7916353343772213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2277447325948103, "1000–3000 Hz_10ms_envelope_correlation": 0.9589387696749726, "1000–3000 Hz_10ms_modulation_mae_db": 1.580931914264019, "1000–3000 Hz_40ms_log_rms_mae_db": 1.024336587763878, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13566623412017365, "1000–3000 Hz_40ms_envelope_correlation": 0.9848660561141185, "1000–3000 Hz_40ms_modulation_mae_db": 1.5185231697975305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.837848323789674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31879979667018626, "3000–6000 Hz_10ms_envelope_correlation": 0.9292241094665491, "3000–6000 Hz_10ms_modulation_mae_db": 3.523085641913413, "3000–6000 Hz_40ms_log_rms_mae_db": 1.266374379006433, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.251617588774721, "3000–6000 Hz_40ms_envelope_correlation": 0.952603960638042, "3000–6000 Hz_40ms_modulation_mae_db": 4.640435608692122, "mr_stft_db": 5.763119010760231, "band_mae_db": 0.22372122496803534, "rms_gain_error_db": -0.01814199036057522, "output_peak": 0.4054114248606039}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.371327119122534, "全帯域_10ms_rms_envelope_nrmse": 0.09574454811018855, "全帯域_10ms_envelope_correlation": 0.9919259029558855, "全帯域_10ms_modulation_mae_db": 0.3929799863212274, "全帯域_40ms_log_rms_mae_db": 0.6957100874201457, "全帯域_40ms_rms_envelope_nrmse": 0.04607509516179608, "全帯域_40ms_envelope_correlation": 0.9980196200425459, "全帯域_40ms_modulation_mae_db": 0.3227179147499751, "300–1000 Hz_10ms_log_rms_mae_db": 1.3320341749929598, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11432115511731852, "300–1000 Hz_10ms_envelope_correlation": 0.9905272987723518, "300–1000 Hz_10ms_modulation_mae_db": 0.7144989121833502, "300–1000 Hz_40ms_log_rms_mae_db": 0.9125603278329122, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04996541282248428, "300–1000 Hz_40ms_envelope_correlation": 0.9982098074361713, "300–1000 Hz_40ms_modulation_mae_db": 0.3408147320390047, "1000–3000 Hz_10ms_log_rms_mae_db": 1.75498906278379, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21919017264644794, "1000–3000 Hz_10ms_envelope_correlation": 0.9623210146571626, "1000–3000 Hz_10ms_modulation_mae_db": 1.0292781120797312, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0189011514122766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1294434428685899, "1000–3000 Hz_40ms_envelope_correlation": 0.9860143847392793, "1000–3000 Hz_40ms_modulation_mae_db": 0.896585088976678, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8601397015568009, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.318222751406438, "3000–6000 Hz_10ms_envelope_correlation": 0.9290703754683762, "3000–6000 Hz_10ms_modulation_mae_db": 3.534781633038195, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3032361138479545, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25670489587387135, "3000–6000 Hz_40ms_envelope_correlation": 0.9503241329071308, "3000–6000 Hz_40ms_modulation_mae_db": 4.64155685026011, "mr_stft_db": 5.751301787437353, "band_mae_db": 0.09598644293986557, "rms_gain_error_db": -0.03531471577775191, "output_peak": 0.4238585330994729}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2973103058384694, "全帯域_10ms_rms_envelope_nrmse": 0.08748360198046484, "全帯域_10ms_envelope_correlation": 0.9932586354151853, "全帯域_10ms_modulation_mae_db": 0.7250384798926905, "全帯域_40ms_log_rms_mae_db": 0.7357396992561216, "全帯域_40ms_rms_envelope_nrmse": 0.04566633246854831, "全帯域_40ms_envelope_correlation": 0.9980779827562939, "全帯域_40ms_modulation_mae_db": 0.24503883830407716, "300–1000 Hz_10ms_log_rms_mae_db": 1.3470982440043697, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1151310848576307, "300–1000 Hz_10ms_envelope_correlation": 0.9903570486086446, "300–1000 Hz_10ms_modulation_mae_db": 0.9703149621694064, "300–1000 Hz_40ms_log_rms_mae_db": 0.8788505215766023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06281287187321904, "300–1000 Hz_40ms_envelope_correlation": 0.9970061449067267, "300–1000 Hz_40ms_modulation_mae_db": 0.3487772763124554, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8072594554847525, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23341996411019705, "1000–3000 Hz_10ms_envelope_correlation": 0.9569154365345242, "1000–3000 Hz_10ms_modulation_mae_db": 1.5388821018983252, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9918016246974538, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1381304631219665, "1000–3000 Hz_40ms_envelope_correlation": 0.9849179202320902, "1000–3000 Hz_40ms_modulation_mae_db": 1.3183172296364711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7893299754269523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3213128928879685, "3000–6000 Hz_10ms_envelope_correlation": 0.9275907688836186, "3000–6000 Hz_10ms_modulation_mae_db": 3.399654979292444, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2434691783321243, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25606597134143433, "3000–6000 Hz_40ms_envelope_correlation": 0.9506522486005169, "3000–6000 Hz_40ms_modulation_mae_db": 4.472414400175169, "mr_stft_db": 5.759802955873319, "band_mae_db": 0.2635074213301268, "rms_gain_error_db": -0.059163511233480987, "output_peak": 0.44959439541161206}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3312380149401366, "全帯域_10ms_rms_envelope_nrmse": 0.0998651548414634, "全帯域_10ms_envelope_correlation": 0.9911990991326844, "全帯域_10ms_modulation_mae_db": 0.8798061397665206, "全帯域_40ms_log_rms_mae_db": 0.7395689950345065, "全帯域_40ms_rms_envelope_nrmse": 0.03914393261592256, "全帯域_40ms_envelope_correlation": 0.9985998965408671, "全帯域_40ms_modulation_mae_db": 0.39391087606276404, "300–1000 Hz_10ms_log_rms_mae_db": 1.3666990083571091, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12782572544845164, "300–1000 Hz_10ms_envelope_correlation": 0.9880976990742991, "300–1000 Hz_10ms_modulation_mae_db": 1.4394473785913877, "300–1000 Hz_40ms_log_rms_mae_db": 0.9097484598787288, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054362444724641024, "300–1000 Hz_40ms_envelope_correlation": 0.9978309917405835, "300–1000 Hz_40ms_modulation_mae_db": 0.5509893499819211, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7035491157049873, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21637913579542936, "1000–3000 Hz_10ms_envelope_correlation": 0.963017916075526, "1000–3000 Hz_10ms_modulation_mae_db": 1.2862554043436134, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0196578318180731, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1286193290949198, "1000–3000 Hz_40ms_envelope_correlation": 0.9866204356373122, "1000–3000 Hz_40ms_modulation_mae_db": 1.1878762120005504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.742033014345822, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3000740484283189, "3000–6000 Hz_10ms_envelope_correlation": 0.9359734129933202, "3000–6000 Hz_10ms_modulation_mae_db": 3.14960773480538, "3000–6000 Hz_40ms_log_rms_mae_db": 1.151328176912496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24579376816477178, "3000–6000 Hz_40ms_envelope_correlation": 0.9540256781410948, "3000–6000 Hz_40ms_modulation_mae_db": 4.232148192565584, "mr_stft_db": 5.740421925561882, "band_mae_db": 0.14166998950892273, "rms_gain_error_db": -0.05891533499544788, "output_peak": 0.42878972928644155}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2590866622581232, "全帯域_10ms_rms_envelope_nrmse": 0.07814055484685635, "全帯域_10ms_envelope_correlation": 0.9946768518505473, "全帯域_10ms_modulation_mae_db": 0.32126898148895444, "全帯域_40ms_log_rms_mae_db": 0.6627397805128826, "全帯域_40ms_rms_envelope_nrmse": 0.04509585781419469, "全帯域_40ms_envelope_correlation": 0.998198409447025, "全帯域_40ms_modulation_mae_db": 0.666631902367803, "300–1000 Hz_10ms_log_rms_mae_db": 1.3386527144696545, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10067109413014533, "300–1000 Hz_10ms_envelope_correlation": 0.9928146580395061, "300–1000 Hz_10ms_modulation_mae_db": 0.5610206626425961, "300–1000 Hz_40ms_log_rms_mae_db": 0.8898401435748768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06386790470663127, "300–1000 Hz_40ms_envelope_correlation": 0.9971637480560055, "300–1000 Hz_40ms_modulation_mae_db": 1.1052555083142688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7213834441234435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21170444848341277, "1000–3000 Hz_10ms_envelope_correlation": 0.9653093575658885, "1000–3000 Hz_10ms_modulation_mae_db": 0.5229007374479141, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0225776160117044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1420223523584799, "1000–3000 Hz_40ms_envelope_correlation": 0.9834162663510834, "1000–3000 Hz_40ms_modulation_mae_db": 0.19692279853187397, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8059132876244306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30751910580396596, "3000–6000 Hz_10ms_envelope_correlation": 0.9323806119747614, "3000–6000 Hz_10ms_modulation_mae_db": 3.1320796420161012, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2382311028172859, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.252102301643885, "3000–6000 Hz_40ms_envelope_correlation": 0.951364490053683, "3000–6000 Hz_40ms_modulation_mae_db": 4.230504532605598, "mr_stft_db": 5.758500105882064, "band_mae_db": 0.18594673637582748, "rms_gain_error_db": -0.09921121680178531, "output_peak": 0.4010598071466407}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2516544996208003, "全帯域_10ms_rms_envelope_nrmse": 0.08732785097961586, "全帯域_10ms_envelope_correlation": 0.9933113731579909, "全帯域_10ms_modulation_mae_db": 0.626234852920879, "全帯域_40ms_log_rms_mae_db": 0.6483148531462245, "全帯域_40ms_rms_envelope_nrmse": 0.045798850604450464, "全帯域_40ms_envelope_correlation": 0.9980433175210606, "全帯域_40ms_modulation_mae_db": 0.3252468044846552, "300–1000 Hz_10ms_log_rms_mae_db": 1.3658530445041992, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1063735093793627, "300–1000 Hz_10ms_envelope_correlation": 0.9917827026229611, "300–1000 Hz_10ms_modulation_mae_db": 0.6688302298565282, "300–1000 Hz_40ms_log_rms_mae_db": 0.9808929137564824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055624740373261275, "300–1000 Hz_40ms_envelope_correlation": 0.9976444844545036, "300–1000 Hz_40ms_modulation_mae_db": 0.20687794298572607, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7915227962060905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2149233495060893, "1000–3000 Hz_10ms_envelope_correlation": 0.9636346645385718, "1000–3000 Hz_10ms_modulation_mae_db": 0.936343090441739, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0649436104321952, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1475788310136465, "1000–3000 Hz_40ms_envelope_correlation": 0.9818818638304041, "1000–3000 Hz_40ms_modulation_mae_db": 0.5778904881165344, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8855980781235606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31215189679106997, "3000–6000 Hz_10ms_envelope_correlation": 0.9308518767743682, "3000–6000 Hz_10ms_modulation_mae_db": 3.1765197527871076, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2791778780001035, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2501255443877768, "3000–6000 Hz_40ms_envelope_correlation": 0.9523994647392842, "3000–6000 Hz_40ms_modulation_mae_db": 4.19935884909356, "mr_stft_db": 5.761458149098107, "band_mae_db": 0.13767251637930758, "rms_gain_error_db": 0.007659593351487415, "output_peak": 0.44122625447059627}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2726460766186307, "全帯域_10ms_rms_envelope_nrmse": 0.08545194438365933, "全帯域_10ms_envelope_correlation": 0.9935789740144086, "全帯域_10ms_modulation_mae_db": 0.5126491339724156, "全帯域_40ms_log_rms_mae_db": 0.7189293684385408, "全帯域_40ms_rms_envelope_nrmse": 0.054034542188709546, "全帯域_40ms_envelope_correlation": 0.9973085094988746, "全帯域_40ms_modulation_mae_db": 0.8288504313129387, "300–1000 Hz_10ms_log_rms_mae_db": 1.3477849271110611, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1069594014091282, "300–1000 Hz_10ms_envelope_correlation": 0.9917165794978768, "300–1000 Hz_10ms_modulation_mae_db": 0.37158624389905154, "300–1000 Hz_40ms_log_rms_mae_db": 0.9119882321341017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07107372294037274, "300–1000 Hz_40ms_envelope_correlation": 0.9962346146056495, "300–1000 Hz_40ms_modulation_mae_db": 0.5540868414434045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7568053485598099, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22256932519614905, "1000–3000 Hz_10ms_envelope_correlation": 0.9619636504539248, "1000–3000 Hz_10ms_modulation_mae_db": 0.9518536533947012, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0647403870799617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13641808901555966, "1000–3000 Hz_40ms_envelope_correlation": 0.9848061096294196, "1000–3000 Hz_40ms_modulation_mae_db": 0.5023450914011848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7580416232323208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3101767966077584, "3000–6000 Hz_10ms_envelope_correlation": 0.9313939706964, "3000–6000 Hz_10ms_modulation_mae_db": 3.005435721944661, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1462671542183194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24185215484276068, "3000–6000 Hz_40ms_envelope_correlation": 0.955385586068631, "3000–6000 Hz_40ms_modulation_mae_db": 4.1232341589939185, "mr_stft_db": 5.754154295634396, "band_mae_db": 0.07191626052146362, "rms_gain_error_db": -0.0711299665557911, "output_peak": 0.44985652678175725}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2084066749366698, "全帯域_10ms_rms_envelope_nrmse": 0.08547869588212485, "全帯域_10ms_envelope_correlation": 0.9936482082012431, "全帯域_10ms_modulation_mae_db": 0.5901128800634473, "全帯域_40ms_log_rms_mae_db": 0.6623105411664804, "全帯域_40ms_rms_envelope_nrmse": 0.04601905010263164, "全帯域_40ms_envelope_correlation": 0.9981592845245152, "全帯域_40ms_modulation_mae_db": 0.22032104712188239, "300–1000 Hz_10ms_log_rms_mae_db": 1.3070409442298454, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11364609382281118, "300–1000 Hz_10ms_envelope_correlation": 0.9906997735345306, "300–1000 Hz_10ms_modulation_mae_db": 0.8517652625202125, "300–1000 Hz_40ms_log_rms_mae_db": 0.859581851232598, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06402915011134451, "300–1000 Hz_40ms_envelope_correlation": 0.9970651273496486, "300–1000 Hz_40ms_modulation_mae_db": 0.21811454015047596, "1000–3000 Hz_10ms_log_rms_mae_db": 1.684044366686239, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2034460834914592, "1000–3000 Hz_10ms_envelope_correlation": 0.9675166293040738, "1000–3000 Hz_10ms_modulation_mae_db": 0.7827998292547695, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9525124061066647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11978916339054656, "1000–3000 Hz_40ms_envelope_correlation": 0.9879892165021945, "1000–3000 Hz_40ms_modulation_mae_db": 0.9152333084145582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8178348361634895, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3152572359978592, "3000–6000 Hz_10ms_envelope_correlation": 0.9288382068574658, "3000–6000 Hz_10ms_modulation_mae_db": 3.181359794473971, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2328251416761409, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25053426958265046, "3000–6000 Hz_40ms_envelope_correlation": 0.951929104505816, "3000–6000 Hz_40ms_modulation_mae_db": 4.2719854125863685, "mr_stft_db": 5.7452311987151035, "band_mae_db": 0.14051902764978363, "rms_gain_error_db": -0.11641072223775638, "output_peak": 0.39274402656022267}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65440, "output_samples": 65440, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.189902014355698, "spectral_convergence": 0.281144685559342, "spectral_convergence_status": "ok", "amplitude_floor": 1.239800719019687e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.338918499728481, "spectral_convergence": 0.25271786797494195, "spectral_convergence_status": "ok", "amplitude_floor": 7.959069981155994e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.760536518196512, "spectral_convergence": 0.2843578896880348, "spectral_convergence_status": "ok", "amplitude_floor": 5.70263552352201e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002917609391844852, "output_power": 0.0002979719817214619, "delta_db": 0.09148281014271081, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009442146735758444, "output_power": 0.0009656161266603648, "delta_db": 0.09733764954848925, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008095018592590047, "output_power": 0.0007799324486419112, "delta_db": -0.16160861285653724, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.140726670679603e-05, "output_power": 6.349539793717478e-05, "delta_db": -0.5100016033593919, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.8558923829027427e-05, "output_power": 2.846694061923404e-05, "delta_db": -0.014010426838042146, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.629692823024492e-06, "output_power": 9.611340469012197e-06, "delta_db": 0.46788624706304077, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22372122496803534, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.01814199036057522, "pitch": null, "output_rms_dbfs": -26.685536239427478, "output_peak": 0.4054114248606039, "reference_rms_dbfs": -26.667394249066906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.763119010760231}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65440, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00016079151656388168, "log_rms_mae_db": 1.235228947836046, "rms_envelope_nrmse": 0.08291933367357664, "envelope_correlation": 0.9939420317669506, "rms_gain_error_db": -0.018142402614935587, "reference_mean_power": 0.0021567095381351967, "output_mean_power": 0.002147718800096015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.6620838184249172e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000592954618765589, "output_power": 0.0005889970028373603, "reference_active": true, "delta_db": -0.029083714614796162}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014246995030046006, "output_power": 0.00013967067641329855, "reference_active": true, "delta_db": -0.08618036398438361}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.93963265073239e-05, "output_power": 2.3231931614413097e-05, "reference_active": true, "delta_db": 0.7836583444663136}], "active_band_count": 3, "active_band_mae_db": 0.29964080768849777}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00012458719833187296, "log_rms_mae_db": 0.662434156630392, "rms_envelope_nrmse": 0.04028089924634026, "envelope_correlation": 0.9984833745765024, "rms_gain_error_db": -0.01814726086856741, "reference_mean_power": 0.0021726617564687806, "output_mean_power": 0.0021636020976271288}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.5403145620848259e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005607371759141368, "output_power": 0.0005586880323296419, "reference_active": true, "delta_db": -0.01589981606328491}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.842815402518841e-05, "output_power": 6.762390678955118e-05, "reference_active": true, "delta_db": -0.0513456691353421}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9268319788025257e-06, "output_power": 1.7890634935773126e-06, "reference_active": true, "delta_db": -0.32218091618538314}], "active_band_count": 3, "active_band_mae_db": 0.12980880046133672}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00013740128966162354, "log_rms_mae_db": 1.3213806380918958, "rms_envelope_nrmse": 0.10538586266343851, "envelope_correlation": 0.9919277102496872, "rms_gain_error_db": -0.03351252855636184, "reference_mean_power": 0.001344215530062727, "output_mean_power": 0.0013338827485705376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.2437342579426783e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004237422220408671, "output_power": 0.0004155012386737379, "reference_active": true, "delta_db": -0.08529416812811796}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010753440950326828, "output_power": 0.0001058145868646406, "reference_active": true, "delta_db": -0.07001913985563099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.157936749643158e-05, "output_power": 1.545866567266265e-05, "reference_active": true, "delta_db": 1.2548716728745848}], "active_band_count": 3, "active_band_mae_db": 0.47006166028611124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00011336523032200158, "log_rms_mae_db": 0.9019097239428031, "rms_envelope_nrmse": 0.053090645031119356, "envelope_correlation": 0.997856706982139, "rms_gain_error_db": -0.03351259019011514, "reference_mean_power": 0.0013541602801659725, "output_mean_power": 0.00134375103581359}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.1658618871138855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004110855381392752, "output_power": 0.0004067402976916461, "reference_active": true, "delta_db": -0.046149965533268694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.54169136496196e-05, "output_power": 4.676649407674392e-05, "reference_active": true, "delta_db": 0.12717195580735374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3416624074278087e-06, "output_power": 1.3734592880280927e-06, "reference_active": true, "delta_db": 0.10172539348587673}], "active_band_count": 3, "active_band_mae_db": 0.09168243827549971}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 4.166087007180048e-05, "log_rms_mae_db": 1.7916353343772213, "rms_envelope_nrmse": 0.2277447325948103, "envelope_correlation": 0.9589387696749726, "rms_gain_error_db": -0.2577771655478898, "reference_mean_power": 7.742789466080943e-05, "output_mean_power": 7.29658647726521e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.015275941663665e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8994572701205476e-05, "output_power": 3.497889284887323e-05, "reference_active": true, "delta_db": -0.4719810666496369}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.4522009847535795e-06, "output_power": 5.980567690386611e-06, "reference_active": true, "delta_db": -0.9554214898039901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.182412963771591e-06, "output_power": 2.5369300997879323e-06, "reference_active": true, "delta_db": 3.3153931863384303}], "active_band_count": 3, "active_band_mae_db": 1.580931914264019}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 3.308768412789438e-05, "log_rms_mae_db": 1.024336587763878, "rms_envelope_nrmse": 0.13566623412017365, "envelope_correlation": 0.9848660561141185, "rms_gain_error_db": -0.25777781763362345, "reference_mean_power": 7.800066882867515e-05, "output_mean_power": 7.350561996177348e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 6.505338847209246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.845568227980419e-05, "output_power": 3.5106893267950316e-05, "reference_active": true, "delta_db": -0.3956812207709293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.058614755593344e-06, "output_power": 2.3700909962210438e-06, "reference_active": true, "delta_db": -1.107597589130598}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905753913870012e-08, "output_power": 2.0003988889482538e-07, "reference_active": true, "delta_db": 3.0522906994910644}], "active_band_count": 3, "active_band_mae_db": 1.5185231697975305}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 1.92244611767394e-05, "log_rms_mae_db": 1.837848323789674, "rms_envelope_nrmse": 0.31879979667018626, "envelope_correlation": 0.9292241094665491, "rms_gain_error_db": 0.05062261073055686, "reference_mean_power": 1.4177032937223143e-05, "output_mean_power": 1.4343251363137098e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.931109844663442e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.831691262181727e-06, "output_power": 4.152991811829842e-06, "reference_active": true, "delta_db": 0.34970566223706456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0928531053800205e-06, "output_power": 1.5624233977352436e-06, "reference_active": true, "delta_db": 1.5523694334336464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4875100105629565e-07, "output_power": 1.0944050291557291e-06, "reference_active": true, "delta_db": 8.667181830069529}], "active_band_count": 3, "active_band_mae_db": 3.523085641913413}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 1.4099069206239137e-05, "log_rms_mae_db": 1.266374379006433, "rms_envelope_nrmse": 0.251617588774721, "envelope_correlation": 0.952603960638042, "rms_gain_error_db": 0.050623510809588636, "reference_mean_power": 1.4281884811465133e-05, "output_mean_power": 1.444933556636029e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 7.397985811629507e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.714499411475195e-06, "output_power": 4.697573610991558e-06, "reference_active": true, "delta_db": 1.0197330034646552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.071953545728327e-07, "output_power": 8.718859806452563e-07, "reference_active": true, "delta_db": 2.3528442738183433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3628863245634334e-08, "output_power": 1.5464382697459137e-07, "reference_active": true, "delta_db": 10.548729548793368}], "active_band_count": 3, "active_band_mae_db": 4.640435608692122}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8564762233272172, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3087987d44b23be1caa061dbc14d07914db84c46329cb9af3ddfb39f9aeb91ef", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9624521798505805, "standard_deviation": 0.003354866043029601, "minimum": 0.9569154365345242, "maximum": 0.9675166293040738, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7513986154907917, "standard_deviation": 0.044910356613214, "minimum": 1.684044366686239, "maximum": 1.8072594554847525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0786556053906016, "standard_deviation": 0.3669082081335184, "minimum": 0.5229007374479141, "maximum": 1.580931914264019, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21867215147799937, "standard_deviation": 0.009372905350549713, "minimum": 0.2034460834914592, "maximum": 0.23341996411019705, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9850640316294877, "standard_deviation": 0.0018841693859958963, "minimum": 0.9818818638304041, "maximum": 0.9879892165021945, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0199339019152758, "standard_deviation": 0.03664129939080718, "minimum": 0.9525124061066647, "maximum": 1.0649436104321952, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8892116733594226, "standard_deviation": 0.4465657936159033, "minimum": 0.19692279853187397, "maximum": 1.5185231697975305, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1347084881229853, "standard_deviation": 0.00863607939610641, "minimum": 0.11978916339054656, "maximum": 0.1475788310136465, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9306654166393574, "standard_deviation": 0.0026483406082375123, "minimum": 0.9275907688836186, "maximum": 0.9359734129933202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8120923550328814, "standard_deviation": 0.048981169087508464, "minimum": 1.742033014345822, "maximum": 1.8855980781235606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.262815612533909, "standard_deviation": 0.19663794436514165, "minimum": 3.005435721944661, "maximum": 3.534781633038195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.31293931557419563, "standard_deviation": 0.00697848527422542, "minimum": 0.3000740484283189, "maximum": 0.3213128928879685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9523355832067748, "standard_deviation": 0.0016996991410680802, "minimum": 0.9503241329071308, "maximum": 0.955385586068631, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.232613640601357, "standard_deviation": 0.056704793223116216, "minimum": 1.1462671542183194, "maximum": 1.3032361138479545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.351454750621554, "standard_deviation": 0.20448422530550062, "minimum": 4.1232341589939185, "maximum": 4.64155685026011, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2505995618264839, "standard_deviation": 0.004932299954796719, "minimum": 0.24185215484276068, "maximum": 0.25670489587387135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9909904337999822, "standard_deviation": 0.0014357341138238268, "minimum": 0.9880976990742991, "maximum": 0.9928146580395061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3408179619701368, "standard_deviation": 0.020650507301449933, "minimum": 1.3070409442298454, "maximum": 1.3666990083571091, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7559406640185804, "standard_deviation": 0.3378661401805978, "minimum": 0.37158624389905154, "maximum": 1.4394473785913877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11128924085353585, "standard_deviation": 0.00839616568582425, "minimum": 0.10067109413014533, "maximum": 0.12782572544845164, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973764531914284, "standard_deviation": 0.0006309471211105675, "minimum": 0.9962346146056495, "maximum": 0.9982098074361713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9056715217411382, "standard_deviation": 0.035598240637892346, "minimum": 0.859581851232598, "maximum": 0.9808929137564824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4270748286878445, "standard_deviation": 0.3183947081795533, "minimum": 0.09168243827549971, "maximum": 1.1052555083142688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05935336157288419, "standard_deviation": 0.0071488885002643785, "minimum": 0.04996541282248428, "maximum": 0.07107372294037274, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15761745245916658, "standard_deviation": 0.06377695104175884, "minimum": 0.07191626052146362, "maximum": 0.2635074213301268, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.754248678620307, "standard_deviation": 0.008113225588768971, "minimum": 5.740421925561882, "maximum": 5.763119010760231, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4240675872021684, "standard_deviation": 0.02233519107213016, "minimum": 0.39274402656022267, "maximum": 0.44985652678175725, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.056328483076387666, "standard_deviation": 0.04084269121896941, "minimum": -0.11641072223775638, "maximum": 0.007659593351487415, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993192634561862, "standard_deviation": 0.001116022918372978, "minimum": 0.9911990991326844, "maximum": 0.9946768518505473, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2783622876464262, "standard_deviation": 0.05300826528310382, "minimum": 1.2084066749366698, "maximum": 1.371327119122534, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5434664077643291, "standard_deviation": 0.20267420319684581, "minimum": 0.29964080768849777, "maximum": 0.8798061397665206, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08780146058724372, "standard_deviation": 0.006938859764325725, "minimum": 0.07814055484685635, "maximum": 0.0998651548414634, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981112993634607, "standard_deviation": 0.00038649821659989604, "minimum": 0.9973085094988746, "maximum": 0.9985998965408671, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6907184352006618, "standard_deviation": 0.036687760719243206, "minimum": 0.6483148531462245, "maximum": 0.7395689950345065, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.39156582685817903, "standard_deviation": 0.237647575805981, "minimum": 0.12980880046133672, "maximum": 0.8288504313129387, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0452643200253242, "standard_deviation": 0.004486696394729116, "minimum": 0.03914393261592256, "maximum": 0.054034542188709546, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0032", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.6393470764160156, "predicted_mos": 2.6393470764160156, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "7c93acf94159ee7deabe8083372acdc9b39308e1bc870b5ae1730bfc971f4b16", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.737453700043261, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2846637218942407, "全帯域_10ms_rms_envelope_nrmse": 0.09078160045409511, "全帯域_10ms_envelope_correlation": 0.9927712478088127, "全帯域_10ms_modulation_mae_db": 0.23611893169617923, "全帯域_40ms_log_rms_mae_db": 0.6922401439331871, "全帯域_40ms_rms_envelope_nrmse": 0.046462011595119715, "全帯域_40ms_envelope_correlation": 0.9980384636321964, "全帯域_40ms_modulation_mae_db": 0.3698805784836814, "300–1000 Hz_10ms_log_rms_mae_db": 1.6351459871736282, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11677087213312513, "300–1000 Hz_10ms_envelope_correlation": 0.9900920561635959, "300–1000 Hz_10ms_modulation_mae_db": 0.8222804952049315, "300–1000 Hz_40ms_log_rms_mae_db": 1.2371879430881405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06310449918837976, "300–1000 Hz_40ms_envelope_correlation": 0.9970026685224436, "300–1000 Hz_40ms_modulation_mae_db": 0.10109356410330628, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1563517512769956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2568437663401912, "1000–3000 Hz_10ms_envelope_correlation": 0.9498748949268808, "1000–3000 Hz_10ms_modulation_mae_db": 1.749925574039656, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3147194213136548, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13731919766216324, "1000–3000 Hz_40ms_envelope_correlation": 0.9848165519198463, "1000–3000 Hz_40ms_modulation_mae_db": 1.9516276562503727, "3000–6000 Hz_10ms_log_rms_mae_db": 2.371074462319581, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34410540234104336, "3000–6000 Hz_10ms_envelope_correlation": 0.9167626929660196, "3000–6000 Hz_10ms_modulation_mae_db": 3.293062562601188, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6711378493094637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.276639833506684, "3000–6000 Hz_40ms_envelope_correlation": 0.9426023920627354, "3000–6000 Hz_40ms_modulation_mae_db": 4.166150782013784, "mr_stft_db": 6.115227052660625, "band_mae_db": 0.18421935893891672, "rms_gain_error_db": -0.05895485228483721, "output_peak": 0.3775407192428513}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2171256820272736, "全帯域_10ms_rms_envelope_nrmse": 0.09392318607392836, "全帯域_10ms_envelope_correlation": 0.9924201787853711, "全帯域_10ms_modulation_mae_db": 0.31231954957541536, "全帯域_40ms_log_rms_mae_db": 0.6665689207774231, "全帯域_40ms_rms_envelope_nrmse": 0.05983057429069842, "全帯域_40ms_envelope_correlation": 0.9969517191011263, "全帯域_40ms_modulation_mae_db": 0.6698380674596379, "300–1000 Hz_10ms_log_rms_mae_db": 1.5968231343996215, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1212165014523084, "300–1000 Hz_10ms_envelope_correlation": 0.9896263277559362, "300–1000 Hz_10ms_modulation_mae_db": 0.4385814196581305, "300–1000 Hz_40ms_log_rms_mae_db": 1.0859118422630467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08249988242135485, "300–1000 Hz_40ms_envelope_correlation": 0.9953007457673537, "300–1000 Hz_40ms_modulation_mae_db": 0.7777515190347076, "1000–3000 Hz_10ms_log_rms_mae_db": 2.272365669505664, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2734430388006005, "1000–3000 Hz_10ms_envelope_correlation": 0.9412690127508545, "1000–3000 Hz_10ms_modulation_mae_db": 2.019412886180261, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4044935812285035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15671696462417775, "1000–3000 Hz_40ms_envelope_correlation": 0.9793225694220218, "1000–3000 Hz_40ms_modulation_mae_db": 1.5911023931074961, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4512947563620413, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34881633033920073, "3000–6000 Hz_10ms_envelope_correlation": 0.913968496504689, "3000–6000 Hz_10ms_modulation_mae_db": 3.078553374446565, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6388932724925762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2753347408791335, "3000–6000 Hz_40ms_envelope_correlation": 0.9430427730084793, "3000–6000 Hz_40ms_modulation_mae_db": 3.8129284364641225, "mr_stft_db": 6.124150608599201, "band_mae_db": 0.1558732470807659, "rms_gain_error_db": -0.17277196331072187, "output_peak": 0.35953649683570726}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2119713550683353, "全帯域_10ms_rms_envelope_nrmse": 0.08491000391692535, "全帯域_10ms_envelope_correlation": 0.9936903906644063, "全帯域_10ms_modulation_mae_db": 0.11610010680762044, "全帯域_40ms_log_rms_mae_db": 0.6248715022205269, "全帯域_40ms_rms_envelope_nrmse": 0.046980779441307643, "全帯域_40ms_envelope_correlation": 0.9980167289178353, "全帯域_40ms_modulation_mae_db": 0.33539354540800836, "300–1000 Hz_10ms_log_rms_mae_db": 1.5536568878091628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11465499453354658, "300–1000 Hz_10ms_envelope_correlation": 0.990496204166861, "300–1000 Hz_10ms_modulation_mae_db": 0.5576374207259844, "300–1000 Hz_40ms_log_rms_mae_db": 1.086350435176539, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06305243853529505, "300–1000 Hz_40ms_envelope_correlation": 0.997112791140451, "300–1000 Hz_40ms_modulation_mae_db": 0.14879362336982002, "1000–3000 Hz_10ms_log_rms_mae_db": 2.108990534372082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25260844726641574, "1000–3000 Hz_10ms_envelope_correlation": 0.9557202565714145, "1000–3000 Hz_10ms_modulation_mae_db": 1.5870876925880297, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3188608521401413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17146849636508313, "1000–3000 Hz_40ms_envelope_correlation": 0.9796653437576355, "1000–3000 Hz_40ms_modulation_mae_db": 1.2872336242382156, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4598800171176443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3510589913718665, "3000–6000 Hz_10ms_envelope_correlation": 0.9152807356105034, "3000–6000 Hz_10ms_modulation_mae_db": 3.317116860109717, "3000–6000 Hz_40ms_log_rms_mae_db": 1.65277264778446, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26830450249973203, "3000–6000 Hz_40ms_envelope_correlation": 0.9474971898418025, "3000–6000 Hz_40ms_modulation_mae_db": 4.187267156171304, "mr_stft_db": 6.106085846273825, "band_mae_db": 0.2848547202903979, "rms_gain_error_db": -0.08882004511964128, "output_peak": 0.414518044868784}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3227485718201226, "全帯域_10ms_rms_envelope_nrmse": 0.09187643884773, "全帯域_10ms_envelope_correlation": 0.9928700437494263, "全帯域_10ms_modulation_mae_db": 0.15448529578484393, "全帯域_40ms_log_rms_mae_db": 0.6716081120327811, "全帯域_40ms_rms_envelope_nrmse": 0.06014724783888835, "全帯域_40ms_envelope_correlation": 0.9970481244980037, "全帯域_40ms_modulation_mae_db": 0.35643211425760274, "300–1000 Hz_10ms_log_rms_mae_db": 1.6219258464349573, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12430643687399345, "300–1000 Hz_10ms_envelope_correlation": 0.98928032352258, "300–1000 Hz_10ms_modulation_mae_db": 0.5372911667545378, "300–1000 Hz_40ms_log_rms_mae_db": 1.1255843310771902, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08414762705071466, "300–1000 Hz_40ms_envelope_correlation": 0.9953598936469975, "300–1000 Hz_40ms_modulation_mae_db": 0.4217111441241439, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2004783865318487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23239503670858536, "1000–3000 Hz_10ms_envelope_correlation": 0.9572428600098333, "1000–3000 Hz_10ms_modulation_mae_db": 0.7761682491627219, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3355238694084557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15904504405991401, "1000–3000 Hz_40ms_envelope_correlation": 0.978721543679593, "1000–3000 Hz_40ms_modulation_mae_db": 0.5672021591831203, "3000–6000 Hz_10ms_log_rms_mae_db": 2.359907376097266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3451524155196482, "3000–6000 Hz_10ms_envelope_correlation": 0.9224404451741922, "3000–6000 Hz_10ms_modulation_mae_db": 3.8936084305286713, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7029067509803613, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2743755832855225, "3000–6000 Hz_40ms_envelope_correlation": 0.9484739689816342, "3000–6000 Hz_40ms_modulation_mae_db": 4.8447888701919295, "mr_stft_db": 6.114986990842612, "band_mae_db": 0.28171982190232886, "rms_gain_error_db": -0.20321643365916092, "output_peak": 0.43015232013049487}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3216988239845606, "全帯域_10ms_rms_envelope_nrmse": 0.09015727399525651, "全帯域_10ms_envelope_correlation": 0.9929129864945812, "全帯域_10ms_modulation_mae_db": 0.3111533199531994, "全帯域_40ms_log_rms_mae_db": 0.7139494669314023, "全帯域_40ms_rms_envelope_nrmse": 0.04636842738227734, "全帯域_40ms_envelope_correlation": 0.9981423874827764, "全帯域_40ms_modulation_mae_db": 0.5395869096714002, "300–1000 Hz_10ms_log_rms_mae_db": 1.5575164527368575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10780773626980746, "300–1000 Hz_10ms_envelope_correlation": 0.9919403387447214, "300–1000 Hz_10ms_modulation_mae_db": 0.7741282335178474, "300–1000 Hz_40ms_log_rms_mae_db": 1.0713610262905768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06005716785201193, "300–1000 Hz_40ms_envelope_correlation": 0.9978208971304319, "300–1000 Hz_40ms_modulation_mae_db": 0.4292002443747589, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2251145082940726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25842147540613297, "1000–3000 Hz_10ms_envelope_correlation": 0.9505454802957743, "1000–3000 Hz_10ms_modulation_mae_db": 1.9837997298195755, "1000–3000 Hz_40ms_log_rms_mae_db": 1.394479814508647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15842179171696552, "1000–3000 Hz_40ms_envelope_correlation": 0.9806630761493395, "1000–3000 Hz_40ms_modulation_mae_db": 1.8728688881889077, "3000–6000 Hz_10ms_log_rms_mae_db": 2.377621273208665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3418559540738717, "3000–6000 Hz_10ms_envelope_correlation": 0.9224159010550659, "3000–6000 Hz_10ms_modulation_mae_db": 3.47122477528418, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6460098915217944, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2659529929479168, "3000–6000 Hz_40ms_envelope_correlation": 0.9506154645927041, "3000–6000 Hz_40ms_modulation_mae_db": 4.319605326833718, "mr_stft_db": 6.101326845368952, "band_mae_db": 0.3272130843565803, "rms_gain_error_db": -0.12197109425278149, "output_peak": 0.4387141769240377}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.181692852798527, "全帯域_10ms_rms_envelope_nrmse": 0.078312817378924, "全帯域_10ms_envelope_correlation": 0.9946127005407094, "全帯域_10ms_modulation_mae_db": 0.7518115523958139, "全帯域_40ms_log_rms_mae_db": 0.6074864019362013, "全帯域_40ms_rms_envelope_nrmse": 0.040341389325797346, "全帯域_40ms_envelope_correlation": 0.9985260081374204, "全帯域_40ms_modulation_mae_db": 0.469104500423639, "300–1000 Hz_10ms_log_rms_mae_db": 1.4989671722983624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08883656999205591, "300–1000 Hz_10ms_envelope_correlation": 0.9944515182818552, "300–1000 Hz_10ms_modulation_mae_db": 0.8167745618831885, "300–1000 Hz_40ms_log_rms_mae_db": 1.0542331018233675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04575454030685698, "300–1000 Hz_40ms_envelope_correlation": 0.9986774083526307, "300–1000 Hz_40ms_modulation_mae_db": 0.3603200219992974, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1529537928618714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23000451354661436, "1000–3000 Hz_10ms_envelope_correlation": 0.9602556913351511, "1000–3000 Hz_10ms_modulation_mae_db": 1.123011286971337, "1000–3000 Hz_40ms_log_rms_mae_db": 1.32073158457115, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13809976842739594, "1000–3000 Hz_40ms_envelope_correlation": 0.984982638038794, "1000–3000 Hz_40ms_modulation_mae_db": 1.2452297048682779, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4229441713612108, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.349446555269524, "3000–6000 Hz_10ms_envelope_correlation": 0.9169599719344134, "3000–6000 Hz_10ms_modulation_mae_db": 3.528910822280441, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6525931991751184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27603000126535626, "3000–6000 Hz_40ms_envelope_correlation": 0.9450013798472464, "3000–6000 Hz_40ms_modulation_mae_db": 4.3736580602501265, "mr_stft_db": 6.101262723920857, "band_mae_db": 0.25296113455745384, "rms_gain_error_db": -0.0696367978867151, "output_peak": 0.43283233343981586}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2162032813683479, "全帯域_10ms_rms_envelope_nrmse": 0.0808469052738979, "全帯域_10ms_envelope_correlation": 0.9943516996830195, "全帯域_10ms_modulation_mae_db": 0.7120018646102843, "全帯域_40ms_log_rms_mae_db": 0.7192992507199776, "全帯域_40ms_rms_envelope_nrmse": 0.04674873616146962, "全帯域_40ms_envelope_correlation": 0.9981197325719663, "全帯域_40ms_modulation_mae_db": 0.31173501913269364, "300–1000 Hz_10ms_log_rms_mae_db": 1.595052540978776, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11234228114640132, "300–1000 Hz_10ms_envelope_correlation": 0.9911499494013875, "300–1000 Hz_10ms_modulation_mae_db": 1.186273240225135, "300–1000 Hz_40ms_log_rms_mae_db": 1.193674935157061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07012908944026171, "300–1000 Hz_40ms_envelope_correlation": 0.9967403283943044, "300–1000 Hz_40ms_modulation_mae_db": 0.3689947106418847, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1608780124220126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24120453714906584, "1000–3000 Hz_10ms_envelope_correlation": 0.9560141482043262, "1000–3000 Hz_10ms_modulation_mae_db": 1.0634456712261233, "1000–3000 Hz_40ms_log_rms_mae_db": 1.336708357046128, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13837373710963927, "1000–3000 Hz_40ms_envelope_correlation": 0.9847410249859738, "1000–3000 Hz_40ms_modulation_mae_db": 0.9048111636037953, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4592419746504754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35847430254406565, "3000–6000 Hz_10ms_envelope_correlation": 0.9105489752981282, "3000–6000 Hz_10ms_modulation_mae_db": 3.2934964546659757, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7408220411315813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27732554842313917, "3000–6000 Hz_40ms_envelope_correlation": 0.9431417610910761, "3000–6000 Hz_40ms_modulation_mae_db": 4.143542136382461, "mr_stft_db": 6.105770544411375, "band_mae_db": 0.23534383197173478, "rms_gain_error_db": -0.12588063119800272, "output_peak": 0.4262264348066711}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2760404100681497, "全帯域_10ms_rms_envelope_nrmse": 0.0828797136579631, "全帯域_10ms_envelope_correlation": 0.9939657376416166, "全帯域_10ms_modulation_mae_db": 0.22534222553546437, "全帯域_40ms_log_rms_mae_db": 0.6605516566937061, "全帯域_40ms_rms_envelope_nrmse": 0.045367192316585356, "全帯域_40ms_envelope_correlation": 0.9981295057171482, "全帯域_40ms_modulation_mae_db": 0.7118495156004515, "300–1000 Hz_10ms_log_rms_mae_db": 1.622165297652909, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10458166998856819, "300–1000 Hz_10ms_envelope_correlation": 0.9920944677900383, "300–1000 Hz_10ms_modulation_mae_db": 0.43106359917790615, "300–1000 Hz_40ms_log_rms_mae_db": 1.1681016377756215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05834797126929595, "300–1000 Hz_40ms_envelope_correlation": 0.9975215039293819, "300–1000 Hz_40ms_modulation_mae_db": 0.4053481836708825, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2002310257972817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23812739285328502, "1000–3000 Hz_10ms_envelope_correlation": 0.9548428178924822, "1000–3000 Hz_10ms_modulation_mae_db": 1.3996833251260539, "1000–3000 Hz_40ms_log_rms_mae_db": 1.335988399713364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1355254080009194, "1000–3000 Hz_40ms_envelope_correlation": 0.9848990609615943, "1000–3000 Hz_40ms_modulation_mae_db": 0.8798522258070941, "3000–6000 Hz_10ms_log_rms_mae_db": 2.456098449673641, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3482040343882742, "3000–6000 Hz_10ms_envelope_correlation": 0.9133214762099342, "3000–6000 Hz_10ms_modulation_mae_db": 3.0098866415331, "3000–6000 Hz_40ms_log_rms_mae_db": 1.666975631708363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27111416569430086, "3000–6000 Hz_40ms_envelope_correlation": 0.944243652150007, "3000–6000 Hz_40ms_modulation_mae_db": 3.729298972576895, "mr_stft_db": 6.099459502471337, "band_mae_db": 0.17025858109066708, "rms_gain_error_db": -0.07772119711759334, "output_peak": 0.47286111719986584}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65440, "output_samples": 65440, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.629452320631676, "spectral_convergence": 0.2910933470843931, "spectral_convergence_status": "ok", "amplitude_floor": 1.239800719019687e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.719349789536507, "spectral_convergence": 0.268577254898041, "spectral_convergence_status": "ok", "amplitude_floor": 7.959069981155994e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.996879047813692, "spectral_convergence": 0.2900229934207286, "spectral_convergence_status": "ok", "amplitude_floor": 5.70263552352201e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002917609391844852, "output_power": 0.00029174766082983543, "delta_db": -0.0001976565950871387, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009442146735758444, "output_power": 0.0009617659819943889, "delta_db": 0.07998666554193157, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008095018592590047, "output_power": 0.0007591564671462663, "delta_db": -0.27886554502174277, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.140726670679603e-05, "output_power": 7.23156763110918e-05, "delta_db": 0.05490042492635526, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.8558923829027427e-05, "output_power": 3.060053623894172e-05, "delta_db": 0.2998719893493965, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.629692823024492e-06, "output_power": 9.443755276675873e-06, "delta_db": 0.391493872198987, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18421935893891672, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05895485228483721, "pitch": null, "output_rms_dbfs": -26.726349101351744, "output_peak": 0.3775407192428513, "reference_rms_dbfs": -26.667394249066906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.115227052660625}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65440, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00016079151656388168, "log_rms_mae_db": 1.2846637218942407, "rms_envelope_nrmse": 0.09078160045409511, "envelope_correlation": 0.9927712478088127, "rms_gain_error_db": -0.058954410154518626, "reference_mean_power": 0.0021567095381351967, "output_mean_power": 0.0021276305520047554}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.6620838184249172e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000592954618765589, "output_power": 0.0005943884465427686, "reference_active": true, "delta_db": 0.010489029034918601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014246995030046006, "output_power": 0.0001316251724382428, "reference_active": true, "delta_db": -0.34384319651174283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.93963265073239e-05, "output_power": 2.1043692309849546e-05, "reference_active": true, "delta_db": 0.3540245695418762}], "active_band_count": 3, "active_band_mae_db": 0.23611893169617923}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00012458719833187296, "log_rms_mae_db": 0.6922401439331871, "rms_envelope_nrmse": 0.046462011595119715, "envelope_correlation": 0.9980384636321964, "rms_gain_error_db": -0.058958462348660826, "reference_mean_power": 0.0021726617564687806, "output_mean_power": 0.002143365686177264}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.5403145620848259e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005607371759141368, "output_power": 0.0005656749937006381, "reference_active": true, "delta_db": 0.038076306555409827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.842815402518841e-05, "output_power": 6.127043470396776e-05, "reference_active": true, "delta_db": -0.4798386267207577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9268319788025257e-06, "output_power": 1.6814004731152103e-06, "reference_active": true, "delta_db": -0.5917268021748766}], "active_band_count": 3, "active_band_mae_db": 0.3698805784836814}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00013740128966162354, "log_rms_mae_db": 1.6351459871736282, "rms_envelope_nrmse": 0.11677087213312513, "envelope_correlation": 0.9900920561635959, "rms_gain_error_db": -0.07224567084905184, "reference_mean_power": 0.001344215530062727, "output_mean_power": 0.0013220392269281548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.2437342579426783e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004237422220408671, "output_power": 0.0004127224654887224, "reference_active": true, "delta_db": -0.11443630407263272}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010753440950326828, "output_power": 0.00010097502759799092, "reference_active": true, "delta_db": -0.27333474093886995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.157936749643158e-05, "output_power": 1.8689250344644497e-05, "reference_active": true, "delta_db": 2.079070440603292}], "active_band_count": 3, "active_band_mae_db": 0.8222804952049315}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00011336523032200158, "log_rms_mae_db": 1.2371879430881405, "rms_envelope_nrmse": 0.06310449918837976, "envelope_correlation": 0.9970026685224436, "rms_gain_error_db": -0.07224592913627621, "reference_mean_power": 0.0013541602801659725, "output_mean_power": 0.0013318198335147064}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.1658618871138855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004110855381392752, "output_power": 0.00040179520279656636, "reference_active": true, "delta_db": -0.09927451487909937}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.54169136496196e-05, "output_power": 4.410802370594943e-05, "reference_active": true, "delta_db": -0.1270001866924718}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3416624074278087e-06, "output_power": 1.365663965854995e-06, "reference_active": true, "delta_db": 0.07700599073834766}], "active_band_count": 3, "active_band_mae_db": 0.10109356410330628}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 4.166087007180048e-05, "log_rms_mae_db": 2.1563517512769956, "rms_envelope_nrmse": 0.2568437663401912, "envelope_correlation": 0.9498748949268808, "rms_gain_error_db": 0.15470780868248699, "reference_mean_power": 7.742789466080943e-05, "output_mean_power": 8.023580817559815e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.015275941663665e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8994572701205476e-05, "output_power": 3.983938722861873e-05, "reference_active": true, "delta_db": 0.09308484154322927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.4522009847535795e-06, "output_power": 7.914617338323159e-06, "reference_active": true, "delta_db": 0.2614536288058773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.182412963771591e-06, "output_power": 3.6500012045834354e-06, "reference_active": true, "delta_db": 4.895238251769861}], "active_band_count": 3, "active_band_mae_db": 1.749925574039656}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 3.308768412789438e-05, "log_rms_mae_db": 1.3147194213136548, "rms_envelope_nrmse": 0.13731919766216324, "envelope_correlation": 0.9848165519198463, "rms_gain_error_db": 0.15470740332892916, "reference_mean_power": 7.800066882867515e-05, "output_mean_power": 8.082934638715765e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 6.505338847209246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.845568227980419e-05, "output_power": 4.0248283008922874e-05, "reference_active": true, "delta_db": 0.19786837101995852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.058614755593344e-06, "output_power": 3.382822975200974e-06, "reference_active": true, "delta_db": 0.43754492184916716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905753913870012e-08, "output_power": 3.2948413089836225e-07, "reference_active": true, "delta_db": 5.219469675881992}], "active_band_count": 3, "active_band_mae_db": 1.9516276562503727}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 1.92244611767394e-05, "log_rms_mae_db": 2.371074462319581, "rms_envelope_nrmse": 0.34410540234104336, "envelope_correlation": 0.9167626929660196, "rms_gain_error_db": 0.07886222100711812, "reference_mean_power": 1.4177032937223143e-05, "output_mean_power": 1.443682095661356e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.931109844663442e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.831691262181727e-06, "output_power": 4.1151530362946145e-06, "reference_active": true, "delta_db": 0.30995481982866047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0928531053800205e-06, "output_power": 1.446527609505683e-06, "reference_active": true, "delta_db": 1.2176493666310881}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4875100105629565e-07, "output_power": 1.0176964210651396e-06, "reference_active": true, "delta_db": 8.351583501343814}], "active_band_count": 3, "active_band_mae_db": 3.293062562601188}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 1.4099069206239137e-05, "log_rms_mae_db": 1.6711378493094637, "rms_envelope_nrmse": 0.276639833506684, "envelope_correlation": 0.9426023920627354, "rms_gain_error_db": 0.07886050006944528, "reference_mean_power": 1.4281884811465133e-05, "output_mean_power": 1.4543588433216745e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 7.397985811629507e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.714499411475195e-06, "output_power": 4.618581257860364e-06, "reference_active": true, "delta_db": 0.946082949909257}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.071953545728327e-07, "output_power": 7.835301643470037e-07, "reference_active": true, "delta_db": 1.888804534295327}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3628863245634334e-08, "output_power": 1.2612933905892474e-07, "reference_active": true, "delta_db": 9.66356486183677}], "active_band_count": 3, "active_band_mae_db": 4.166150782013784}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8374720723986493, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fe6c16d405da7312a76659f6f45971e899c1a4ba1a3f96fdef5b9021cd3ae39e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9532206452483396, "standard_deviation": 0.005898490324311932, "minimum": 0.9412690127508545, "maximum": 0.9602556913351511, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1846704601327285, "standard_deviation": 0.05059590681455668, "minimum": 2.108990534372082, "maximum": 2.272365669505664, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.4628168018892198, "standard_deviation": 0.45190616478410334, "minimum": 0.7761682491627219, "maximum": 2.019412886180261, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24788102600886136, "standard_deviation": 0.014958472148415088, "minimum": 0.23000451354661436, "maximum": 0.2734430388006005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9822264761143498, "standard_deviation": 0.0028658439513554023, "minimum": 0.978721543679593, "maximum": 0.984982638038794, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3451882349912556, "standard_deviation": 0.034674685309814254, "minimum": 1.3147194213136548, "maximum": 1.4044935812285035, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.28749097690591, "standard_deviation": 0.49412837734802423, "minimum": 0.5672021591831203, "maximum": 1.9516276562503727, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14937130099578227, "standard_deviation": 0.013641812353177057, "minimum": 0.1355254080009194, "maximum": 0.17146849636508313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9164623368441183, "standard_deviation": 0.00420830525634113, "minimum": 0.9105489752981282, "maximum": 0.9224404451741922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.4197578100988153, "standard_deviation": 0.04346064763681296, "minimum": 2.359907376097266, "maximum": 2.4598800171176443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.3607324901812294, "standard_deviation": 0.2772748388676502, "minimum": 3.0098866415331, "maximum": 3.8936084305286713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3483892482309368, "standard_deviation": 0.0050945742084499614, "minimum": 0.3418559540738717, "maximum": 0.35847430254406565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9455773226969606, "standard_deviation": 0.0029465676241504546, "minimum": 0.9426023920627354, "maximum": 0.9506154645927041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6715139105129648, "standard_deviation": 0.03429630145562106, "minimum": 1.6388932724925762, "maximum": 1.7408220411315813, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.197154967610542, "standard_deviation": 0.3455905354391072, "minimum": 3.729298972576895, "maximum": 4.8447888701919295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.27313467106272316, "standard_deviation": 0.004202077631840036, "minimum": 0.2659529929479168, "maximum": 0.27732554842313917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9911413982283719, "standard_deviation": 0.0016809881095854792, "minimum": 0.98928032352258, "maximum": 0.9944515182818552, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5851566649355342, "standard_deviation": 0.04573124087604946, "minimum": 1.4989671722983624, "maximum": 1.6351459871736282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6955037671434576, "standard_deviation": 0.25573185686899547, "minimum": 0.43106359917790615, "maximum": 1.186273240225135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1113146327987258, "standard_deviation": 0.011159320385676502, "minimum": 0.08883656999205591, "maximum": 0.12430643687399345, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969420296104994, "standard_deviation": 0.001158214438671058, "minimum": 0.9953007457673537, "maximum": 0.9986774083526307, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.127800656581443, "standard_deviation": 0.06547026676403737, "minimum": 1.0542331018233675, "maximum": 1.2371879430881405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.37665162641485017, "standard_deviation": 0.20517351491645258, "minimum": 0.10109356410330628, "maximum": 0.7777515190347076, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06588665200802135, "standard_deviation": 0.012757913870655386, "minimum": 0.04575454030685698, "maximum": 0.08414762705071466, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23655547252360568, "standard_deviation": 0.061535147092332494, "minimum": 0.1558732470807659, "maximum": 0.3272130843565803, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.108533764318598, "standard_deviation": 0.00871048375710031, "minimum": 6.099459502471337, "maximum": 6.124150608599201, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4190477054310285, "standard_deviation": 0.0357220373265405, "minimum": 0.35953649683570726, "maximum": 0.47286111719986584, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11487162685368174, "standard_deviation": 0.05149281397245802, "minimum": -0.20321643365916092, "maximum": -0.05895485228483721, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9934493731709929, "standard_deviation": 0.0008136929692671424, "minimum": 0.9924201787853711, "maximum": 0.9946127005407094, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2540180873786948, "standard_deviation": 0.05414078947104207, "minimum": 1.181692852798527, "maximum": 1.3227485718201226, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35241660579485257, "standard_deviation": 0.2440209747128278, "minimum": 0.11610010680762044, "maximum": 0.7518115523958139, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08671099244984004, "standard_deviation": 0.0057319081736860065, "minimum": 0.078312817378924, "maximum": 0.09392318607392836, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978715837573091, "standard_deviation": 0.0005609761449334324, "minimum": 0.9969517191011263, "maximum": 0.9985260081374204, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6695719319056506, "standard_deviation": 0.03946660542030903, "minimum": 0.6074864019362013, "maximum": 0.7192992507199776, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4704775313046394, "standard_deviation": 0.15554329762555444, "minimum": 0.31173501913269364, "maximum": 0.7118495156004515, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04903079479401797, "standard_deviation": 0.007093616022133636, "minimum": 0.040341389325797346, "maximum": 0.06014724783888835, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0032", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8325209617614746, "predicted_mos": 2.8325209617614746, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "9f4acdcdf00b78aabe14862a58e5cfb506b03b31fd7d259d5410cad870e7c9f3", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0334777999669313, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2328270824781289, "全帯域_10ms_rms_envelope_nrmse": 0.07446408438663715, "全帯域_10ms_envelope_correlation": 0.9952961086348876, "全帯域_10ms_modulation_mae_db": 0.3277735109658273, "全帯域_40ms_log_rms_mae_db": 0.6492987020546791, "全帯域_40ms_rms_envelope_nrmse": 0.042672859835201436, "全帯域_40ms_envelope_correlation": 0.9985430591427065, "全帯域_40ms_modulation_mae_db": 0.9048521224882539, "300–1000 Hz_10ms_log_rms_mae_db": 1.236700722496432, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09324915271699931, "300–1000 Hz_10ms_envelope_correlation": 0.9939617867452739, "300–1000 Hz_10ms_modulation_mae_db": 0.46868176100298226, "300–1000 Hz_40ms_log_rms_mae_db": 0.9242588047561827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05350713078627791, "300–1000 Hz_40ms_envelope_correlation": 0.9982051627564063, "300–1000 Hz_40ms_modulation_mae_db": 0.8264298602825685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6311447596162454, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22846451442395363, "1000–3000 Hz_10ms_envelope_correlation": 0.960388404264175, "1000–3000 Hz_10ms_modulation_mae_db": 1.6487284085797371, "1000–3000 Hz_40ms_log_rms_mae_db": 0.988854507230608, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1364821482146777, "1000–3000 Hz_40ms_envelope_correlation": 0.9850321608878969, "1000–3000 Hz_40ms_modulation_mae_db": 1.5136810541554697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7003915396333718, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32835511514552235, "3000–6000 Hz_10ms_envelope_correlation": 0.923493449389494, "3000–6000 Hz_10ms_modulation_mae_db": 3.6290429120701493, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2460593712113104, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2638452225778226, "3000–6000 Hz_40ms_envelope_correlation": 0.9469944513414428, "3000–6000 Hz_40ms_modulation_mae_db": 4.643485800292066, "mr_stft_db": 5.720380967187375, "band_mae_db": 0.14931052361576058, "rms_gain_error_db": -0.14056392908815318, "output_peak": 0.445368001758485}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3467094221499802, "全帯域_10ms_rms_envelope_nrmse": 0.08847461062936242, "全帯域_10ms_envelope_correlation": 0.9933575555725582, "全帯域_10ms_modulation_mae_db": 0.6925793662537639, "全帯域_40ms_log_rms_mae_db": 0.7335067983319346, "全帯域_40ms_rms_envelope_nrmse": 0.04725812082612611, "全帯域_40ms_envelope_correlation": 0.9983116232189063, "全帯域_40ms_modulation_mae_db": 0.6477452677051235, "300–1000 Hz_10ms_log_rms_mae_db": 1.2739357873191768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11694645085312122, "300–1000 Hz_10ms_envelope_correlation": 0.9907944028577131, "300–1000 Hz_10ms_modulation_mae_db": 0.9809329027273607, "300–1000 Hz_40ms_log_rms_mae_db": 0.8724892851394461, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07036088234477009, "300–1000 Hz_40ms_envelope_correlation": 0.9972629941523524, "300–1000 Hz_40ms_modulation_mae_db": 0.5833862820972876, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6699483596631375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2790244571675517, "1000–3000 Hz_10ms_envelope_correlation": 0.9441560204644157, "1000–3000 Hz_10ms_modulation_mae_db": 2.0579033369940896, "1000–3000 Hz_40ms_log_rms_mae_db": 1.024796369645715, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19792158934091003, "1000–3000 Hz_40ms_envelope_correlation": 0.970864404571155, "1000–3000 Hz_40ms_modulation_mae_db": 1.61054577595959, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6713424893636042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31253475354362126, "3000–6000 Hz_10ms_envelope_correlation": 0.9347202175383604, "3000–6000 Hz_10ms_modulation_mae_db": 3.577869021686149, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1749368326813956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25317199692974984, "3000–6000 Hz_40ms_envelope_correlation": 0.9542443324272762, "3000–6000 Hz_40ms_modulation_mae_db": 4.724439482926418, "mr_stft_db": 5.713997071130386, "band_mae_db": 0.33095665988658923, "rms_gain_error_db": -0.18170658413111068, "output_peak": 0.3494355200791462}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3280395920127395, "全帯域_10ms_rms_envelope_nrmse": 0.0893621574722869, "全帯域_10ms_envelope_correlation": 0.9932392596213067, "全帯域_10ms_modulation_mae_db": 0.4670881044247581, "全帯域_40ms_log_rms_mae_db": 0.7519331402979129, "全帯域_40ms_rms_envelope_nrmse": 0.055873459587039556, "全帯域_40ms_envelope_correlation": 0.9974771088635952, "全帯域_40ms_modulation_mae_db": 0.13820069529118886, "300–1000 Hz_10ms_log_rms_mae_db": 1.3475827480284979, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10581361614033491, "300–1000 Hz_10ms_envelope_correlation": 0.9925980642022302, "300–1000 Hz_10ms_modulation_mae_db": 0.8907948356081903, "300–1000 Hz_40ms_log_rms_mae_db": 0.9651000812606566, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06442783385487393, "300–1000 Hz_40ms_envelope_correlation": 0.9977969240307951, "300–1000 Hz_40ms_modulation_mae_db": 0.8148940942751007, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5500954509600693, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21107349369342654, "1000–3000 Hz_10ms_envelope_correlation": 0.9652597848076401, "1000–3000 Hz_10ms_modulation_mae_db": 1.3360685230178742, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8990672504451104, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11827824636516802, "1000–3000 Hz_40ms_envelope_correlation": 0.9883753253541219, "1000–3000 Hz_40ms_modulation_mae_db": 1.313112961323184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7045120841206747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3402907700614592, "3000–6000 Hz_10ms_envelope_correlation": 0.9179739546320442, "3000–6000 Hz_10ms_modulation_mae_db": 3.6604741097427382, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1837121860046667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27112958086215444, "3000–6000 Hz_40ms_envelope_correlation": 0.9441468880408241, "3000–6000 Hz_40ms_modulation_mae_db": 4.885965797622749, "mr_stft_db": 5.7271620742074125, "band_mae_db": 0.2118115946502062, "rms_gain_error_db": -0.19170603914774223, "output_peak": 0.38608086466507596}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3223738777465175, "全帯域_10ms_rms_envelope_nrmse": 0.0819122906930158, "全帯域_10ms_envelope_correlation": 0.9941291998039139, "全帯域_10ms_modulation_mae_db": 0.35031836570022573, "全帯域_40ms_log_rms_mae_db": 0.7258864018987784, "全帯域_40ms_rms_envelope_nrmse": 0.04823103684769292, "全帯域_40ms_envelope_correlation": 0.9979165667132457, "全帯域_40ms_modulation_mae_db": 0.19559876956877006, "300–1000 Hz_10ms_log_rms_mae_db": 1.2793245766648205, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10887392927457316, "300–1000 Hz_10ms_envelope_correlation": 0.9915908903504954, "300–1000 Hz_10ms_modulation_mae_db": 0.625619792559788, "300–1000 Hz_40ms_log_rms_mae_db": 0.8924814907127272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061894982087396286, "300–1000 Hz_40ms_envelope_correlation": 0.9974266708753203, "300–1000 Hz_40ms_modulation_mae_db": 0.5758969724754598, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6077040518159382, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2532969994228245, "1000–3000 Hz_10ms_envelope_correlation": 0.9519242854652666, "1000–3000 Hz_10ms_modulation_mae_db": 1.3110237460547918, "1000–3000 Hz_40ms_log_rms_mae_db": 1.016098684062065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18871127968876455, "1000–3000 Hz_40ms_envelope_correlation": 0.971730051722937, "1000–3000 Hz_40ms_modulation_mae_db": 1.2924343494934398, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7008915410781928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.307621908290523, "3000–6000 Hz_10ms_envelope_correlation": 0.9322534153652245, "3000–6000 Hz_10ms_modulation_mae_db": 2.786579840867621, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1717457683646606, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24505426155104723, "3000–6000 Hz_40ms_envelope_correlation": 0.9540450504297474, "3000–6000 Hz_40ms_modulation_mae_db": 3.8376797587809937, "mr_stft_db": 5.735623950828938, "band_mae_db": 0.2096995604433347, "rms_gain_error_db": -0.09292961495585483, "output_peak": 0.4038315713601569}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3557854551626236, "全帯域_10ms_rms_envelope_nrmse": 0.10423620999739293, "全帯域_10ms_envelope_correlation": 0.990511295050275, "全帯域_10ms_modulation_mae_db": 1.4883655744898343, "全帯域_40ms_log_rms_mae_db": 0.7765067770523755, "全帯域_40ms_rms_envelope_nrmse": 0.05443379659213552, "全帯域_40ms_envelope_correlation": 0.9972355491826078, "全帯域_40ms_modulation_mae_db": 1.217256924793739, "300–1000 Hz_10ms_log_rms_mae_db": 1.3241740717414656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13654393306834922, "300–1000 Hz_10ms_envelope_correlation": 0.9864979414179231, "300–1000 Hz_10ms_modulation_mae_db": 1.7403989867034892, "300–1000 Hz_40ms_log_rms_mae_db": 0.9127784515642361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06994387097209336, "300–1000 Hz_40ms_envelope_correlation": 0.9962745881670616, "300–1000 Hz_40ms_modulation_mae_db": 1.149126402161563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.671452569925775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2381832517532072, "1000–3000 Hz_10ms_envelope_correlation": 0.9561294946126522, "1000–3000 Hz_10ms_modulation_mae_db": 1.755998439056018, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0507446148722057, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14912662724435838, "1000–3000 Hz_40ms_envelope_correlation": 0.9817323438686065, "1000–3000 Hz_40ms_modulation_mae_db": 1.42708524991749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7096600500338621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30989774135472664, "3000–6000 Hz_10ms_envelope_correlation": 0.9337821493994877, "3000–6000 Hz_10ms_modulation_mae_db": 3.5115417265855933, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2193369499126765, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24924676866381887, "3000–6000 Hz_40ms_envelope_correlation": 0.9539988677188349, "3000–6000 Hz_40ms_modulation_mae_db": 4.505540710144963, "mr_stft_db": 5.711847610876416, "band_mae_db": 0.14671542430680717, "rms_gain_error_db": 0.0009708350290092183, "output_peak": 0.4287331503660655}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.239325234919158, "全帯域_10ms_rms_envelope_nrmse": 0.07715591788659515, "全帯域_10ms_envelope_correlation": 0.994871839951887, "全帯域_10ms_modulation_mae_db": 0.450240639797222, "全帯域_40ms_log_rms_mae_db": 0.6880192457573084, "全帯域_40ms_rms_envelope_nrmse": 0.041900760681356675, "全帯域_40ms_envelope_correlation": 0.9985387638660621, "全帯域_40ms_modulation_mae_db": 0.4615779237936, "300–1000 Hz_10ms_log_rms_mae_db": 1.304863812468039, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09797946370663485, "300–1000 Hz_10ms_envelope_correlation": 0.9932175337552489, "300–1000 Hz_10ms_modulation_mae_db": 0.5974611111000344, "300–1000 Hz_40ms_log_rms_mae_db": 0.9406938255458991, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060928533442952036, "300–1000 Hz_40ms_envelope_correlation": 0.997459922905585, "300–1000 Hz_40ms_modulation_mae_db": 0.6505315480777615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5582056704663878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22244722868360253, "1000–3000 Hz_10ms_envelope_correlation": 0.9612824274678108, "1000–3000 Hz_10ms_modulation_mae_db": 1.7423135427546566, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9118520507168276, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12217961028225599, "1000–3000 Hz_40ms_envelope_correlation": 0.9874628958260304, "1000–3000 Hz_40ms_modulation_mae_db": 1.438183986537042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6883984186611511, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31344212372988733, "3000–6000 Hz_10ms_envelope_correlation": 0.9327775328762224, "3000–6000 Hz_10ms_modulation_mae_db": 3.382709184715958, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1638656041942985, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23904816298715503, "3000–6000 Hz_40ms_envelope_correlation": 0.9581261330602368, "3000–6000 Hz_40ms_modulation_mae_db": 4.492088872015148, "mr_stft_db": 5.698533083973866, "band_mae_db": 0.17470475411306538, "rms_gain_error_db": -0.12840340189924415, "output_peak": 0.399625265137812}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2618697613824474, "全帯域_10ms_rms_envelope_nrmse": 0.07943851811448338, "全帯域_10ms_envelope_correlation": 0.9946223685152458, "全帯域_10ms_modulation_mae_db": 0.19259449573852164, "全帯域_40ms_log_rms_mae_db": 0.644318395853014, "全帯域_40ms_rms_envelope_nrmse": 0.039411250214075114, "全帯域_40ms_envelope_correlation": 0.9988102031412315, "全帯域_40ms_modulation_mae_db": 0.48295256563308747, "300–1000 Hz_10ms_log_rms_mae_db": 1.278311764740016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09656680666428741, "300–1000 Hz_10ms_envelope_correlation": 0.9937650571492764, "300–1000 Hz_10ms_modulation_mae_db": 0.35441243494366415, "300–1000 Hz_40ms_log_rms_mae_db": 0.8586911432749978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0499448118447582, "300–1000 Hz_40ms_envelope_correlation": 0.998822449955398, "300–1000 Hz_40ms_modulation_mae_db": 1.012540816805253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5734348468673218, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2247482064337943, "1000–3000 Hz_10ms_envelope_correlation": 0.962583698449446, "1000–3000 Hz_10ms_modulation_mae_db": 1.7017167503106012, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9889496776390899, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14004855797914717, "1000–3000 Hz_40ms_envelope_correlation": 0.9848359515016816, "1000–3000 Hz_40ms_modulation_mae_db": 1.4483504886396197, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6126383223343426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3150049377931427, "3000–6000 Hz_10ms_envelope_correlation": 0.9284568006438622, "3000–6000 Hz_10ms_modulation_mae_db": 3.25921964252803, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1228314141529538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25450742087323547, "3000–6000 Hz_40ms_envelope_correlation": 0.950215990587271, "3000–6000 Hz_40ms_modulation_mae_db": 4.4367778725286975, "mr_stft_db": 5.702342257048116, "band_mae_db": 0.2086615356382083, "rms_gain_error_db": -0.15251095335738693, "output_peak": 0.40376369983401844}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3714826504126154, "全帯域_10ms_rms_envelope_nrmse": 0.11589856576021533, "全帯域_10ms_envelope_correlation": 0.9882989247196409, "全帯域_10ms_modulation_mae_db": 1.3151056958297709, "全帯域_40ms_log_rms_mae_db": 0.7543586916110806, "全帯域_40ms_rms_envelope_nrmse": 0.06662344169076424, "全帯域_40ms_envelope_correlation": 0.9959041276711198, "全帯域_40ms_modulation_mae_db": 0.44121937166064695, "300–1000 Hz_10ms_log_rms_mae_db": 1.3737236058757392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1583329799946318, "300–1000 Hz_10ms_envelope_correlation": 0.9821194781798029, "300–1000 Hz_10ms_modulation_mae_db": 1.860338963285422, "300–1000 Hz_40ms_log_rms_mae_db": 1.0408248115510257, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09794355592219585, "300–1000 Hz_40ms_envelope_correlation": 0.9928564767280053, "300–1000 Hz_40ms_modulation_mae_db": 0.7991217028489883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6110802994659028, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22397607063303937, "1000–3000 Hz_10ms_envelope_correlation": 0.9604795589610517, "1000–3000 Hz_10ms_modulation_mae_db": 1.129191430023222, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9945322923512393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14091994022203475, "1000–3000 Hz_40ms_envelope_correlation": 0.9833368764496803, "1000–3000 Hz_40ms_modulation_mae_db": 1.0096921184427956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6748523564967606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31349392416458755, "3000–6000 Hz_10ms_envelope_correlation": 0.9329511469438261, "3000–6000 Hz_10ms_modulation_mae_db": 3.234984418844787, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1600905836727329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2504725175443966, "3000–6000 Hz_40ms_envelope_correlation": 0.9541330604501551, "3000–6000 Hz_40ms_modulation_mae_db": 4.330177203342139, "mr_stft_db": 5.71793709492271, "band_mae_db": 0.19843272666999345, "rms_gain_error_db": 0.04925998274834548, "output_peak": 0.42216242970686807}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65440, "output_samples": 65440, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.099816547829754, "spectral_convergence": 0.2697923488049574, "spectral_convergence_status": "ok", "amplitude_floor": 1.239800719019687e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.273065995202006, "spectral_convergence": 0.2472832390708349, "spectral_convergence_status": "ok", "amplitude_floor": 7.959069981155994e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.788260358530365, "spectral_convergence": 0.27919121337624336, "spectral_convergence_status": "ok", "amplitude_floor": 5.70263552352201e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002917609391844852, "output_power": 0.0002922725703944233, "delta_db": 0.007609106821505377, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009442146735758444, "output_power": 0.0009054513053122512, "delta_db": -0.18205646151879057, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008095018592590047, "output_power": 0.0007722422082278701, "delta_db": -0.2046431552587955, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.140726670679603e-05, "output_power": 7.732172533067042e-05, "delta_db": 0.3455912648749149, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.8558923829027427e-05, "output_power": 2.9564707034835508e-05, "delta_db": 0.15031741686905295, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.629692823024492e-06, "output_power": 8.618481693685982e-06, "delta_db": -0.00564573635150428, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14931052361576058, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14056392908815318, "pitch": null, "output_rms_dbfs": -26.807958178155058, "output_peak": 0.445368001758485, "reference_rms_dbfs": -26.667394249066906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.720380967187375}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65440, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00016079151656388168, "log_rms_mae_db": 1.2328270824781289, "rms_envelope_nrmse": 0.07446408438663715, "envelope_correlation": 0.9952961086348876, "rms_gain_error_db": -0.1405649648960542, "reference_mean_power": 0.0021567095381351967, "output_mean_power": 0.0020880224438842897}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.6620838184249172e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000592954618765589, "output_power": 0.0005783194906569973, "reference_active": true, "delta_db": -0.10853627065695663}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014246995030046006, "output_power": 0.00012959095962430958, "reference_active": true, "delta_db": -0.41148566963008554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.93963265073239e-05, "output_power": 1.7433702351202446e-05, "reference_active": true, "delta_db": -0.4632985926104397}], "active_band_count": 3, "active_band_mae_db": 0.3277735109658273}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00012458719833187296, "log_rms_mae_db": 0.6492987020546791, "rms_envelope_nrmse": 0.042672859835201436, "envelope_correlation": 0.9985430591427065, "rms_gain_error_db": -0.1405674403476959, "reference_mean_power": 0.0021726617564687806, "output_mean_power": 0.0021034654154607935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.5403145620848259e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005607371759141368, "output_power": 0.000550799022895136, "reference_active": true, "delta_db": -0.07766188876042932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.842815402518841e-05, "output_power": 6.1025422379493346e-05, "reference_active": true, "delta_db": -0.4972403025290057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9268319788025257e-06, "output_power": 1.1772763722471413e-06, "reference_active": true, "delta_db": -2.139654176175327}], "active_band_count": 3, "active_band_mae_db": 0.9048521224882539}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00013740128966162354, "log_rms_mae_db": 1.236700722496432, "rms_envelope_nrmse": 0.09324915271699931, "envelope_correlation": 0.9939617867452739, "rms_gain_error_db": -0.2075407895654119, "reference_mean_power": 0.001344215530062727, "output_mean_power": 0.0012814888436014733}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.2437342579426783e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004237422220408671, "output_power": 0.0003823269650354195, "reference_active": true, "delta_db": -0.44666810296548815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010753440950326828, "output_power": 9.592440910252644e-05, "reference_active": true, "delta_db": -0.4961832202464172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.157936749643158e-05, "output_power": 1.2882621093357625e-05, "reference_active": true, "delta_db": 0.46319395979704137}], "active_band_count": 3, "active_band_mae_db": 0.46868176100298226}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00011336523032200158, "log_rms_mae_db": 0.9242588047561827, "rms_envelope_nrmse": 0.05350713078627791, "envelope_correlation": 0.9982051627564063, "rms_gain_error_db": -0.2075408890943926, "reference_mean_power": 0.0013541602801659725, "output_mean_power": 0.001290969500724679}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.1658618871138855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004110855381392752, "output_power": 0.00036957177183733256, "reference_active": true, "delta_db": -0.4623340662080655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.54169136496196e-05, "output_power": 4.2596182591313275e-05, "reference_active": true, "delta_db": -0.2784693798496495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3416624074278087e-06, "output_power": 8.990723421011747e-07, "reference_active": true, "delta_db": -1.7384861347899907}], "active_band_count": 3, "active_band_mae_db": 0.8264298602825685}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 4.166087007180048e-05, "log_rms_mae_db": 1.6311447596162454, "rms_envelope_nrmse": 0.22846451442395363, "envelope_correlation": 0.960388404264175, "rms_gain_error_db": 0.1118890047411658, "reference_mean_power": 7.742789466080943e-05, "output_mean_power": 7.944861871488907e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.015275941663665e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8994572701205476e-05, "output_power": 4.21637170024496e-05, "reference_active": true, "delta_db": 0.339347239939619}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.4522009847535795e-06, "output_power": 8.901316034073638e-06, "reference_active": true, "delta_db": 0.7716966134704235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.182412963771591e-06, "output_power": 2.8594553523186065e-06, "reference_active": true, "delta_db": 3.8351413723291685}], "active_band_count": 3, "active_band_mae_db": 1.6487284085797371}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 3.308768412789438e-05, "log_rms_mae_db": 0.988854507230608, "rms_envelope_nrmse": 0.1364821482146777, "envelope_correlation": 0.9850321608878969, "rms_gain_error_db": 0.11188931837841903, "reference_mean_power": 7.800066882867515e-05, "output_mean_power": 8.003634700335543e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 6.505338847209246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.845568227980419e-05, "output_power": 4.2323983713629246e-05, "reference_active": true, "delta_db": 0.41626017595827597}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.058614755593344e-06, "output_power": 3.785177449092539e-06, "reference_active": true, "delta_db": 0.9256146468292594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905753913870012e-08, "output_power": 2.0692090475543692e-07, "reference_active": true, "delta_db": 3.1991683396788737}], "active_band_count": 3, "active_band_mae_db": 1.5136810541554697}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 1.92244611767394e-05, "log_rms_mae_db": 1.7003915396333718, "rms_envelope_nrmse": 0.32835511514552235, "envelope_correlation": 0.923493449389494, "rms_gain_error_db": -0.08720853549428066, "reference_mean_power": 1.4177032937223143e-05, "output_mean_power": 1.3895190166090558e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.931109844663442e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.831691262181727e-06, "output_power": 4.174026287821828e-06, "reference_active": true, "delta_db": 0.3716467128606621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0928531053800205e-06, "output_power": 1.6625680526298886e-06, "reference_active": true, "delta_db": 1.8221764031506575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4875100105629565e-07, "output_power": 1.101007960767248e-06, "reference_active": true, "delta_db": 8.693305620199128}], "active_band_count": 3, "active_band_mae_db": 3.6290429120701493}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 1.4099069206239137e-05, "log_rms_mae_db": 1.2460593712113104, "rms_envelope_nrmse": 0.2638452225778226, "envelope_correlation": 0.9469944513414428, "rms_gain_error_db": -0.08720841951962721, "reference_mean_power": 1.4281884811465133e-05, "output_mean_power": 1.3997957934053044e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 7.397985811629507e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.714499411475195e-06, "output_power": 4.845700777744013e-06, "reference_active": true, "delta_db": 1.1545629904915535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.071953545728327e-07, "output_power": 9.239374858835251e-07, "reference_active": true, "delta_db": 2.6046732044417804}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3628863245634334e-08, "output_power": 1.4176915616315907e-07, "reference_active": true, "delta_db": 10.171221205942864}], "active_band_count": 3, "active_band_mae_db": 4.643485800292066}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8582907202098113, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "90bb76cb139c6855a7f4bbae93231de49d1debf4d38e8fff546b567c301b8d99", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9577754593115573, "standard_deviation": 0.006845303951383742, "minimum": 0.9441560204644157, "maximum": 0.9652597848076401, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6091332510975973, "standard_deviation": 0.04694369183677615, "minimum": 1.5500954509600693, "maximum": 1.671452569925775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.5853680220988737, "standard_deviation": 0.302395183681751, "minimum": 1.129191430023222, "maximum": 2.0579033369940896, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.235151777776425, "standard_deviation": 0.021654001444933298, "minimum": 0.21107349369342654, "maximum": 0.2790244571675517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9816712512727637, "standard_deviation": 0.006742108832411, "minimum": 0.970864404571155, "maximum": 0.9883753253541219, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9843619308703575, "standard_deviation": 0.053085319764724734, "minimum": 0.8990672504451104, "maximum": 1.0507446148722057, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3816357480585788, "standard_deviation": 0.18147975340068864, "minimum": 1.0096921184427956, "maximum": 1.61054577595959, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14920849991716456, "standard_deviation": 0.029118594058224666, "minimum": 0.11827824636516802, "maximum": 0.19792158934091003, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9295510833485652, "standard_deviation": 0.005925937763917668, "minimum": 0.9179739546320442, "maximum": 0.9347202175383604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6828358502152452, "standard_deviation": 0.03157321563703875, "minimum": 1.6126383223343426, "maximum": 1.7096600500338621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.3803026071301283, "standard_deviation": 0.28906622567236756, "minimum": 2.786579840867621, "maximum": 3.6604741097427382, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.31758015926043376, "standard_deviation": 0.01105806986711739, "minimum": 0.307621908290523, "maximum": 0.3402907700614592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9519880967569736, "standard_deviation": 0.004554473868330846, "minimum": 0.9441468880408241, "maximum": 0.9581261330602368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.180322338774337, "standard_deviation": 0.037688391652300646, "minimum": 1.1228314141529538, "maximum": 1.2460593712113104, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.482019437206647, "standard_deviation": 0.31372022459040344, "minimum": 3.8376797587809937, "maximum": 4.885965797622749, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.25330949149867255, "standard_deviation": 0.010184719676970195, "minimum": 0.23904816298715503, "maximum": 0.27112958086215444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9905681443322456, "standard_deviation": 0.00417339555548618, "minimum": 0.9821194781798029, "maximum": 0.9939617867452739, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3023271361667734, "standard_deviation": 0.04447792435028413, "minimum": 1.236700722496432, "maximum": 1.3737236058757392, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9398300984913663, "standard_deviation": 0.5698345936763982, "minimum": 0.35441243494366415, "maximum": 1.860338963285422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11428829155236649, "standard_deviation": 0.02257091575970156, "minimum": 0.09324915271699931, "maximum": 0.1583329799946318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970131486963656, "standard_deviation": 0.0018347906220180599, "minimum": 0.9928564767280053, "maximum": 0.998822449955398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9259147367256464, "standard_deviation": 0.058094276113616644, "minimum": 0.8586911432749978, "maximum": 1.0408248115510257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8014909598779978, "standard_deviation": 0.20264349082553385, "minimum": 0.5758969724754598, "maximum": 1.149126402161563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06611895015691471, "standard_deviation": 0.014705356090731881, "minimum": 0.0499448118447582, "maximum": 0.09794355592219585, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20378659741549562, "standard_deviation": 0.057810691648845836, "minimum": 0.14671542430680717, "maximum": 0.33095665988658923, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7159780137719025, "standard_deviation": 0.012221256933983763, "minimum": 5.698533083973866, "maximum": 5.735623950828938, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40487506286345354, "standard_deviation": 0.02918999011533032, "minimum": 0.3494355200791462, "maximum": 0.445368001758485, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10469871310026715, "standard_deviation": 0.08672004908599126, "minimum": -0.19170603914774223, "maximum": 0.04925998274834548, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930408189837143, "standard_deviation": 0.0024245351313454586, "minimum": 0.9882989247196409, "maximum": 0.9952961086348876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3073016345330264, "standard_deviation": 0.05465658722445506, "minimum": 1.2328270824781289, "maximum": 1.3714826504126154, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6605082191499905, "standard_deviation": 0.48142220781168404, "minimum": 0.19259449573852164, "maximum": 1.4883655744898343, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08886779436749863, "standard_deviation": 0.014388363271000492, "minimum": 0.07446408438663715, "maximum": 0.11589856576021533, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978421252249344, "standard_deviation": 0.0009559411566397783, "minimum": 0.9959041276711198, "maximum": 0.9988102031412315, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7154785191071354, "standard_deviation": 0.049563727208342796, "minimum": 0.644318395853014, "maximum": 0.7765067770523755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5611754551168012, "standard_deviation": 0.35837585818510864, "minimum": 0.13820069529118886, "maximum": 1.217256924793739, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049550590784298944, "standard_deviation": 0.009031156853949967, "minimum": 0.039411250214075114, "maximum": 0.06662344169076424, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0032", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.0739176273345947, "predicted_mos": 3.0739176273345947, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "94d40c5166df923073d285e081db3baf330ae5cd19bdbb19942ba58e88e41efc", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.956676300149411, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2469237118880128, "全帯域_10ms_rms_envelope_nrmse": 0.0850333006577494, "全帯域_10ms_envelope_correlation": 0.9936343097612489, "全帯域_10ms_modulation_mae_db": 0.34034928847913176, "全帯域_40ms_log_rms_mae_db": 0.6722753583319002, "全帯域_40ms_rms_envelope_nrmse": 0.04330466332144275, "全帯域_40ms_envelope_correlation": 0.9982438744040437, "全帯域_40ms_modulation_mae_db": 0.08430724902190616, "300–1000 Hz_10ms_log_rms_mae_db": 1.3738835089678214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10957104393511267, "300–1000 Hz_10ms_envelope_correlation": 0.9912752305095779, "300–1000 Hz_10ms_modulation_mae_db": 0.5968674782339666, "300–1000 Hz_40ms_log_rms_mae_db": 0.9973334403499221, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05665188328239574, "300–1000 Hz_40ms_envelope_correlation": 0.9975563314432107, "300–1000 Hz_40ms_modulation_mae_db": 0.31286089503657566, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8834156683192511, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2331879005040229, "1000–3000 Hz_10ms_envelope_correlation": 0.9569456839842425, "1000–3000 Hz_10ms_modulation_mae_db": 1.419219521634415, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1832301872499682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1436950762161321, "1000–3000 Hz_40ms_envelope_correlation": 0.9826406796775472, "1000–3000 Hz_40ms_modulation_mae_db": 1.3701322274165892, "3000–6000 Hz_10ms_log_rms_mae_db": 1.947681477661774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32973447331592043, "3000–6000 Hz_10ms_envelope_correlation": 0.9243498271258311, "3000–6000 Hz_10ms_modulation_mae_db": 3.5700782172306735, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4068648710279774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26106311900055124, "3000–6000 Hz_40ms_envelope_correlation": 0.9491797278968142, "3000–6000 Hz_40ms_modulation_mae_db": 4.653072698052632, "mr_stft_db": 5.859929941182103, "band_mae_db": 0.2312222883743412, "rms_gain_error_db": -0.004904831674896172, "output_peak": 0.40271667681768053}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3148985331429452, "全帯域_10ms_rms_envelope_nrmse": 0.0914967955814828, "全帯域_10ms_envelope_correlation": 0.9926283993098386, "全帯域_10ms_modulation_mae_db": 0.448866398579338, "全帯域_40ms_log_rms_mae_db": 0.6405743945947615, "全帯域_40ms_rms_envelope_nrmse": 0.04215727956493661, "全帯域_40ms_envelope_correlation": 0.9983391182593977, "全帯域_40ms_modulation_mae_db": 0.34735370133633586, "300–1000 Hz_10ms_log_rms_mae_db": 1.3917717867820856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10748096590598041, "300–1000 Hz_10ms_envelope_correlation": 0.9916205645894469, "300–1000 Hz_10ms_modulation_mae_db": 0.7327198807885941, "300–1000 Hz_40ms_log_rms_mae_db": 1.0135715842325441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044531720817287755, "300–1000 Hz_40ms_envelope_correlation": 0.9985731815678635, "300–1000 Hz_40ms_modulation_mae_db": 0.2783922257986471, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8694617708081431, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2296049577036256, "1000–3000 Hz_10ms_envelope_correlation": 0.9589198169000682, "1000–3000 Hz_10ms_modulation_mae_db": 1.1031529254497363, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1671497671917221, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13321834387760886, "1000–3000 Hz_40ms_envelope_correlation": 0.9851971483636216, "1000–3000 Hz_40ms_modulation_mae_db": 0.9003607251216824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9512344441979308, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32243223541711535, "3000–6000 Hz_10ms_envelope_correlation": 0.9276874098760316, "3000–6000 Hz_10ms_modulation_mae_db": 3.5029191549435565, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3784088145004016, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25990188833328776, "3000–6000 Hz_40ms_envelope_correlation": 0.949592537343942, "3000–6000 Hz_40ms_modulation_mae_db": 4.532584427660165, "mr_stft_db": 5.831547829315942, "band_mae_db": 0.11205105200677877, "rms_gain_error_db": -0.025513789889929913, "output_peak": 0.42652248159814143}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2418427649904449, "全帯域_10ms_rms_envelope_nrmse": 0.08806021134987947, "全帯域_10ms_envelope_correlation": 0.9931795694023211, "全帯域_10ms_modulation_mae_db": 0.8344142951565608, "全帯域_40ms_log_rms_mae_db": 0.7075254457179889, "全帯域_40ms_rms_envelope_nrmse": 0.04755621978327508, "全帯域_40ms_envelope_correlation": 0.9979289626961302, "全帯域_40ms_modulation_mae_db": 0.28482141724930377, "300–1000 Hz_10ms_log_rms_mae_db": 1.3897749204998187, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1166852645838973, "300–1000 Hz_10ms_envelope_correlation": 0.9900964995372684, "300–1000 Hz_10ms_modulation_mae_db": 1.2326000073227916, "300–1000 Hz_40ms_log_rms_mae_db": 1.0003254985131158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06513865693103536, "300–1000 Hz_40ms_envelope_correlation": 0.9967992626963463, "300–1000 Hz_40ms_modulation_mae_db": 0.5211497562812343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8859284066531499, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2538527933890296, "1000–3000 Hz_10ms_envelope_correlation": 0.9490464267166304, "1000–3000 Hz_10ms_modulation_mae_db": 1.5202470883598067, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1055572879011468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15364415761527803, "1000–3000 Hz_40ms_envelope_correlation": 0.9800838554612297, "1000–3000 Hz_40ms_modulation_mae_db": 1.5142543203562238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9119079060388917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3300936838392486, "3000–6000 Hz_10ms_envelope_correlation": 0.9245704837070164, "3000–6000 Hz_10ms_modulation_mae_db": 3.5611159502899965, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4054506021030457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2678353577928016, "3000–6000 Hz_40ms_envelope_correlation": 0.9468985037183952, "3000–6000 Hz_40ms_modulation_mae_db": 4.564968840447659, "mr_stft_db": 5.842192174844495, "band_mae_db": 0.27517737499520284, "rms_gain_error_db": -0.07085303253190538, "output_peak": 0.4490155345232677}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2575959963411187, "全帯域_10ms_rms_envelope_nrmse": 0.09799680927812103, "全帯域_10ms_envelope_correlation": 0.9915401802566689, "全帯域_10ms_modulation_mae_db": 0.7176379419402349, "全帯域_40ms_log_rms_mae_db": 0.6721994839109766, "全帯域_40ms_rms_envelope_nrmse": 0.04156825210888844, "全帯域_40ms_envelope_correlation": 0.9984590358866469, "全帯域_40ms_modulation_mae_db": 0.2571063778130403, "300–1000 Hz_10ms_log_rms_mae_db": 1.3935736762178275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12498969253266035, "300–1000 Hz_10ms_envelope_correlation": 0.9886473135736256, "300–1000 Hz_10ms_modulation_mae_db": 1.2320898224077779, "300–1000 Hz_40ms_log_rms_mae_db": 0.9956883829032112, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05988957349165662, "300–1000 Hz_40ms_envelope_correlation": 0.9974190802933378, "300–1000 Hz_40ms_modulation_mae_db": 0.42084075895069234, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7944838666894245, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2298676560878372, "1000–3000 Hz_10ms_envelope_correlation": 0.9580780244011147, "1000–3000 Hz_10ms_modulation_mae_db": 1.2941106629497323, "1000–3000 Hz_40ms_log_rms_mae_db": 1.162711345252653, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14133952254060975, "1000–3000 Hz_40ms_envelope_correlation": 0.9832689702684803, "1000–3000 Hz_40ms_modulation_mae_db": 1.1387710859562583, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8651219702628317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31361752110787616, "3000–6000 Hz_10ms_envelope_correlation": 0.9310880823603345, "3000–6000 Hz_10ms_modulation_mae_db": 3.299849385189701, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2986301392483905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25562802814364444, "3000–6000 Hz_40ms_envelope_correlation": 0.9511197851836584, "3000–6000 Hz_40ms_modulation_mae_db": 4.335330460144271, "mr_stft_db": 5.834548285681585, "band_mae_db": 0.17793882986450615, "rms_gain_error_db": -0.08438568940162094, "output_peak": 0.4166481785041401}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2147223102163993, "全帯域_10ms_rms_envelope_nrmse": 0.07776122872696008, "全帯域_10ms_envelope_correlation": 0.9947633868772644, "全帯域_10ms_modulation_mae_db": 0.4279807999156405, "全帯域_40ms_log_rms_mae_db": 0.6406856718185081, "全帯域_40ms_rms_envelope_nrmse": 0.04610442652186984, "全帯域_40ms_envelope_correlation": 0.9981534953850172, "全帯域_40ms_modulation_mae_db": 0.8142695469352267, "300–1000 Hz_10ms_log_rms_mae_db": 1.421090398580275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10254182178193902, "300–1000 Hz_10ms_envelope_correlation": 0.9926185594311341, "300–1000 Hz_10ms_modulation_mae_db": 0.5516497551326203, "300–1000 Hz_40ms_log_rms_mae_db": 1.0296801030864668, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06842398849936969, "300–1000 Hz_40ms_envelope_correlation": 0.9967994904382017, "300–1000 Hz_40ms_modulation_mae_db": 1.0332170498053157, "1000–3000 Hz_10ms_log_rms_mae_db": 1.779104897424582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21871173405190203, "1000–3000 Hz_10ms_envelope_correlation": 0.9630335463567249, "1000–3000 Hz_10ms_modulation_mae_db": 0.6171804645331798, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1290038535430271, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1451896774732129, "1000–3000 Hz_40ms_envelope_correlation": 0.9826504919173457, "1000–3000 Hz_40ms_modulation_mae_db": 0.25478889940883925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9474899706181845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31486557779858887, "3000–6000 Hz_10ms_envelope_correlation": 0.9296909697265876, "3000–6000 Hz_10ms_modulation_mae_db": 3.1991579996745174, "3000–6000 Hz_40ms_log_rms_mae_db": 1.375243420583286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25648706114298764, "3000–6000 Hz_40ms_envelope_correlation": 0.9500692318247833, "3000–6000 Hz_40ms_modulation_mae_db": 4.22922127505512, "mr_stft_db": 5.84306418805136, "band_mae_db": 0.2467311271573612, "rms_gain_error_db": -0.11593129659802728, "output_peak": 0.4037713922275177}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2625375554989782, "全帯域_10ms_rms_envelope_nrmse": 0.08679879834966461, "全帯域_10ms_envelope_correlation": 0.9934000103120603, "全帯域_10ms_modulation_mae_db": 0.40043345924976986, "全帯域_40ms_log_rms_mae_db": 0.674067860106364, "全帯域_40ms_rms_envelope_nrmse": 0.047544061024447505, "全帯域_40ms_envelope_correlation": 0.9978981638447363, "全帯域_40ms_modulation_mae_db": 0.2582080839382162, "300–1000 Hz_10ms_log_rms_mae_db": 1.4398736517442479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10306327673628222, "300–1000 Hz_10ms_envelope_correlation": 0.9922978418703264, "300–1000 Hz_10ms_modulation_mae_db": 0.41944803738981173, "300–1000 Hz_40ms_log_rms_mae_db": 1.1426886208416123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05776221509774142, "300–1000 Hz_40ms_envelope_correlation": 0.9974618465623308, "300–1000 Hz_40ms_modulation_mae_db": 0.30308012547002056, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8426704004566623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21642876221614604, "1000–3000 Hz_10ms_envelope_correlation": 0.9631091146867681, "1000–3000 Hz_10ms_modulation_mae_db": 0.9521954023910296, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1735059468720839, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14628813517752426, "1000–3000 Hz_40ms_envelope_correlation": 0.9820535784584006, "1000–3000 Hz_40ms_modulation_mae_db": 0.5699642215144269, "3000–6000 Hz_10ms_log_rms_mae_db": 2.057169024836575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3244630415868997, "3000–6000 Hz_10ms_envelope_correlation": 0.9262192407001348, "3000–6000 Hz_10ms_modulation_mae_db": 3.390565206018112, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4577169182121321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25964385293790126, "3000–6000 Hz_40ms_envelope_correlation": 0.9493818753752862, "3000–6000 Hz_40ms_modulation_mae_db": 4.321565790710426, "mr_stft_db": 5.849533592882099, "band_mae_db": 0.16067027829895408, "rms_gain_error_db": 0.012758365136396497, "output_peak": 0.4471590199142867}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2524833771425365, "全帯域_10ms_rms_envelope_nrmse": 0.08363607430573226, "全帯域_10ms_envelope_correlation": 0.9938424191848374, "全帯域_10ms_modulation_mae_db": 0.4149715191989489, "全帯域_40ms_log_rms_mae_db": 0.7087268607441067, "全帯域_40ms_rms_envelope_nrmse": 0.05199852525357148, "全帯域_40ms_envelope_correlation": 0.9974986438233115, "全帯域_40ms_modulation_mae_db": 0.6761582319461672, "300–1000 Hz_10ms_log_rms_mae_db": 1.4069315972516074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1080084394076065, "300–1000 Hz_10ms_envelope_correlation": 0.9915572670541284, "300–1000 Hz_10ms_modulation_mae_db": 0.3708376710526335, "300–1000 Hz_40ms_log_rms_mae_db": 1.0314147270287712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07280748281233025, "300–1000 Hz_40ms_envelope_correlation": 0.9960563102817407, "300–1000 Hz_40ms_modulation_mae_db": 0.522989397746699, "1000–3000 Hz_10ms_log_rms_mae_db": 1.865582547525593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21869009689035787, "1000–3000 Hz_10ms_envelope_correlation": 0.9637204155830593, "1000–3000 Hz_10ms_modulation_mae_db": 1.0454189723760925, "1000–3000 Hz_40ms_log_rms_mae_db": 1.192770697138043, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13603389885744982, "1000–3000 Hz_40ms_envelope_correlation": 0.9851648832853498, "1000–3000 Hz_40ms_modulation_mae_db": 0.6498552755476092, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9546637570332424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32382694854809785, "3000–6000 Hz_10ms_envelope_correlation": 0.9262288694365577, "3000–6000 Hz_10ms_modulation_mae_db": 3.208681608608021, "3000–6000 Hz_40ms_log_rms_mae_db": 1.381011882649297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25175315206706356, "3000–6000 Hz_40ms_envelope_correlation": 0.9524187849754341, "3000–6000 Hz_40ms_modulation_mae_db": 4.2518366385667905, "mr_stft_db": 5.839588373616319, "band_mae_db": 0.1406765275944608, "rms_gain_error_db": -0.06375622821236072, "output_peak": 0.44860343383646656}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1960210769472452, "全帯域_10ms_rms_envelope_nrmse": 0.08350206875285059, "全帯域_10ms_envelope_correlation": 0.9939952700044208, "全帯域_10ms_modulation_mae_db": 0.5749413216958342, "全帯域_40ms_log_rms_mae_db": 0.6455684783674056, "全帯域_40ms_rms_envelope_nrmse": 0.04463243292557301, "全帯域_40ms_envelope_correlation": 0.9983462341540863, "全帯域_40ms_modulation_mae_db": 0.18129820391357898, "300–1000 Hz_10ms_log_rms_mae_db": 1.348773111666501, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10769484025764149, "300–1000 Hz_10ms_envelope_correlation": 0.991716914419472, "300–1000 Hz_10ms_modulation_mae_db": 0.8403308403663455, "300–1000 Hz_40ms_log_rms_mae_db": 0.9488523964800556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06103001817939947, "300–1000 Hz_40ms_envelope_correlation": 0.9974220495861219, "300–1000 Hz_40ms_modulation_mae_db": 0.22427528757610363, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7807578607371086, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.208820149303122, "1000–3000 Hz_10ms_envelope_correlation": 0.9657547275973924, "1000–3000 Hz_10ms_modulation_mae_db": 0.976316725376367, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1077837595400721, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12498276052886398, "1000–3000 Hz_40ms_envelope_correlation": 0.9868555898400561, "1000–3000 Hz_40ms_modulation_mae_db": 1.1656325490576753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9409039197739228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3288644702661395, "3000–6000 Hz_10ms_envelope_correlation": 0.9230428924015813, "3000–6000 Hz_10ms_modulation_mae_db": 3.436262225235382, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3789809349284285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2654029821770083, "3000–6000 Hz_40ms_envelope_correlation": 0.9463778761996557, "3000–6000 Hz_40ms_modulation_mae_db": 4.523794561262209, "mr_stft_db": 5.830145948163332, "band_mae_db": 0.16676593844389723, "rms_gain_error_db": -0.13945207322578151, "output_peak": 0.40177190466534013}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65440, "output_samples": 65440, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.307369409989186, "spectral_convergence": 0.28457134499968206, "spectral_convergence_status": "ok", "amplitude_floor": 1.239800719019687e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.419722571029538, "spectral_convergence": 0.2522289149274661, "spectral_convergence_status": "ok", "amplitude_floor": 7.959069981155994e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.852697842527586, "spectral_convergence": 0.2826776818454911, "spectral_convergence_status": "ok", "amplitude_floor": 5.70263552352201e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002917609391844852, "output_power": 0.0002994041213247997, "delta_db": 0.1123062584011107, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009442146735758444, "output_power": 0.0009645056676986041, "delta_db": 0.09234038715714063, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008095018592590047, "output_power": 0.0007834776517845296, "delta_db": -0.14191237575712415, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.140726670679603e-05, "output_power": 6.518156452939091e-05, "delta_db": -0.3961762920968924, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.8558923829027427e-05, "output_power": 2.932410417259287e-05, "delta_db": 0.11482915527956426, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.629692823024492e-06, "output_power": 9.749273708628632e-06, "delta_db": 0.5297692615542148, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2312222883743412, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.004904831674896172, "pitch": null, "output_rms_dbfs": -26.6722990807418, "output_peak": 0.40271667681768053, "reference_rms_dbfs": -26.667394249066906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.859929941182103}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65440, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00016079151656388168, "log_rms_mae_db": 1.2469237118880128, "rms_envelope_nrmse": 0.0850333006577494, "envelope_correlation": 0.9936343097612489, "rms_gain_error_db": -0.004905661911683199, "reference_mean_power": 0.0021567095381351967, "output_mean_power": 0.0021542747582683867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.6620838184249172e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000592954618765589, "output_power": 0.0005923933371120274, "reference_active": true, "delta_db": -0.004112911182665185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014246995030046006, "output_power": 0.0001448810255378, "reference_active": true, "delta_db": 0.07288238661323557}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.93963265073239e-05, "output_power": 2.410597728266039e-05, "reference_active": true, "delta_db": 0.9440525676414947}], "active_band_count": 3, "active_band_mae_db": 0.34034928847913176}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00012458719833187296, "log_rms_mae_db": 0.6722753583319002, "rms_envelope_nrmse": 0.04330466332144275, "envelope_correlation": 0.9982438744040437, "rms_gain_error_db": -0.004911071071003548, "reference_mean_power": 0.0021726617564687806, "output_mean_power": 0.002170206264617111}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.5403145620848259e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005607371759141368, "output_power": 0.0005611803185653964, "reference_active": true, "delta_db": 0.0034308123207549886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.842815402518841e-05, "output_power": 7.100189967526026e-05, "reference_active": true, "delta_db": 0.16035144277175872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9268319788025257e-06, "output_power": 1.966789108811886e-06, "reference_active": true, "delta_db": 0.08913949197320478}], "active_band_count": 3, "active_band_mae_db": 0.08430724902190616}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00013740128966162354, "log_rms_mae_db": 1.3738835089678214, "rms_envelope_nrmse": 0.10957104393511267, "envelope_correlation": 0.9912752305095779, "rms_gain_error_db": -0.02438480297066592, "reference_mean_power": 0.001344215530062727, "output_mean_power": 0.001336689166754812}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.2437342579426783e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004237422220408671, "output_power": 0.0004138370529196214, "reference_active": true, "delta_db": -0.10272366987131956}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010753440950326828, "output_power": 0.0001120142743327203, "reference_active": true, "delta_db": 0.17725915078984814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.157936749643158e-05, "output_power": 1.6396335598660813e-05, "reference_active": true, "delta_db": 1.5106196140407322}], "active_band_count": 3, "active_band_mae_db": 0.5968674782339666}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00011336523032200158, "log_rms_mae_db": 0.9973334403499221, "rms_envelope_nrmse": 0.05665188328239574, "envelope_correlation": 0.9975563314432107, "rms_gain_error_db": -0.024384940871782706, "reference_mean_power": 0.0013541602801659725, "output_mean_power": 0.0013465781927005339}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.1658618871138855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004110855381392752, "output_power": 0.00040540988078356313, "reference_active": true, "delta_db": -0.060378694503910355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.54169136496196e-05, "output_power": 4.990969191304433e-05, "reference_active": true, "delta_db": 0.40967271038980463}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3416624074278087e-06, "output_power": 1.4945015826916332e-06, "reference_active": true, "delta_db": 0.4685312802160119}], "active_band_count": 3, "active_band_mae_db": 0.31286089503657566}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 4.166087007180048e-05, "log_rms_mae_db": 1.8834156683192511, "rms_envelope_nrmse": 0.2331879005040229, "envelope_correlation": 0.9569456839842425, "rms_gain_error_db": -0.1522746881785842, "reference_mean_power": 7.742789466080943e-05, "output_mean_power": 7.476011860770788e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.015275941663665e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8994572701205476e-05, "output_power": 3.5506941485253753e-05, "reference_active": true, "delta_db": -0.40690901309357985}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.4522009847535795e-06, "output_power": 6.582665830312829e-06, "reference_active": true, "delta_db": -0.5388275059082419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.182412963771591e-06, "output_power": 2.5349032440936856e-06, "reference_active": true, "delta_db": 3.311922045901423}], "active_band_count": 3, "active_band_mae_db": 1.419219521634415}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 3.308768412789438e-05, "log_rms_mae_db": 1.1832301872499682, "rms_envelope_nrmse": 0.1436950762161321, "envelope_correlation": 0.9826406796775472, "rms_gain_error_db": -0.15227594163310942, "reference_mean_power": 7.800066882867515e-05, "output_mean_power": 7.53131361200669e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 6.505338847209246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.845568227980419e-05, "output_power": 3.53802934358953e-05, "reference_active": true, "delta_db": -0.3619909047605626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.058614755593344e-06, "output_power": 2.7721648658927768e-06, "reference_active": true, "delta_db": -0.42705724329844674}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905753913870012e-08, "output_power": 2.1282487556102798e-07, "reference_active": true, "delta_db": 3.3213485341907583}], "active_band_count": 3, "active_band_mae_db": 1.3701322274165892}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 1.92244611767394e-05, "log_rms_mae_db": 1.947681477661774, "rms_envelope_nrmse": 0.32973447331592043, "envelope_correlation": 0.9243498271258311, "rms_gain_error_db": 0.10112708468845741, "reference_mean_power": 1.4177032937223143e-05, "output_mean_power": 1.4511023880479571e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.931109844663442e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.831691262181727e-06, "output_power": 4.225006684661362e-06, "reference_active": true, "delta_db": 0.42436891793896264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0928531053800205e-06, "output_power": 1.490921145357321e-06, "reference_active": true, "delta_db": 1.348928836080785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4875100105629565e-07, "output_power": 1.1645377917006009e-06, "reference_active": true, "delta_db": 8.936936897672274}], "active_band_count": 3, "active_band_mae_db": 3.5700782172306735}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 1.4099069206239137e-05, "log_rms_mae_db": 1.4068648710279774, "rms_envelope_nrmse": 0.26106311900055124, "envelope_correlation": 0.9491797278968142, "rms_gain_error_db": 0.10112638677285685, "reference_mean_power": 1.4281884811465133e-05, "output_mean_power": 1.4618343568127559e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 7.397985811629507e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.714499411475195e-06, "output_power": 4.769231590374202e-06, "reference_active": true, "delta_db": 1.0854811809100688}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.071953545728327e-07, "output_power": 8.385441774101978e-07, "reference_active": true, "delta_db": 2.1835068051420747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3628863245634334e-08, "output_power": 1.5976536818205527e-07, "reference_active": true, "delta_db": 10.690230108105752}], "active_band_count": 3, "active_band_mae_db": 4.653072698052632}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8516054824060347, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f4daf3f0bb50f432e83045f25519157829dd4ee7b2f7f142d7bd375bc3aa23b3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9598259695282501, "standard_deviation": 0.00534195638793646, "minimum": 0.9490464267166304, "maximum": 0.9657547275973924, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8376756773267393, "standard_deviation": 0.045935164985163215, "minimum": 1.779104897424582, "maximum": 1.8859284066531499, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1159802203837947, "standard_deviation": 0.2896777058904828, "minimum": 0.6171804645331798, "maximum": 1.5202470883598067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22614550626825541, "standard_deviation": 0.01387978224910637, "minimum": 0.208820149303122, "maximum": 0.2538527933890296, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.983489399659004, "standard_deviation": 0.002144324138628329, "minimum": 0.9800838554612297, "maximum": 0.9868555898400561, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1527141055860894, "standard_deviation": 0.03397703648855651, "minimum": 1.1055572879011468, "maximum": 1.192770697138043, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.945469913047413, "standard_deviation": 0.43047454756347936, "minimum": 0.25478889940883925, "maximum": 1.5142543203562238, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14054894653583494, "standard_deviation": 0.008887486412867046, "minimum": 0.12498276052886398, "maximum": 0.15364415761527803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9266097219167594, "standard_deviation": 0.002751535301048371, "minimum": 0.9230428924015813, "maximum": 0.9310880823603345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.947021558802919, "standard_deviation": 0.05370183223006829, "minimum": 1.8651219702628317, "maximum": 2.057169024836575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.396078718398745, "standard_deviation": 0.14819323110446833, "minimum": 3.1991579996745174, "maximum": 3.5700782172306735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.32348724398498585, "standard_deviation": 0.00638651106483695, "minimum": 0.31361752110787616, "maximum": 0.3300936838392486, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9493797903147461, "standard_deviation": 0.00199971292339249, "minimum": 0.9463778761996557, "maximum": 0.9524187849754341, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3852884479066199, "standard_deviation": 0.04445528399728427, "minimum": 1.2986301392483905, "maximum": 1.4577169182121321, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.426546836487409, "standard_deviation": 0.16034702794317507, "minimum": 4.22922127505512, "maximum": 4.653072698052632, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.25971443019940577, "standard_deviation": 0.005218986136506985, "minimum": 0.25175315206706356, "maximum": 0.2678353577928016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9912287738731225, "standard_deviation": 0.0012829786701735644, "minimum": 0.9886473135736256, "maximum": 0.9926185594311341, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3957090814637731, "standard_deviation": 0.02791641292487036, "minimum": 1.348773111666501, "maximum": 1.4398736517442479, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7470679365868176, "standard_deviation": 0.3358409299804879, "minimum": 0.3708376710526335, "maximum": 1.2326000073227916, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11000441814263999, "standard_deviation": 0.007454008046429923, "minimum": 0.10254182178193902, "maximum": 0.12498969253266035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972609441086442, "standard_deviation": 0.0007346307647701051, "minimum": 0.9960563102817407, "maximum": 0.9985731815678635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0199443441794624, "standard_deviation": 0.05589747370889891, "minimum": 0.9488523964800556, "maximum": 1.1426886208416123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45210068708316103, "standard_deviation": 0.25970870579004723, "minimum": 0.22427528757610363, "maximum": 1.0332170498053157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06077944238890204, "standard_deviation": 0.008561534483132195, "minimum": 0.044531720817287755, "maximum": 0.07280748281233025, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1889041770919378, "standard_deviation": 0.05633781510489021, "minimum": 0.11205105200677877, "maximum": 0.27517737499520284, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.841318791717154, "standard_deviation": 0.00990530368137846, "minimum": 5.830145948163332, "maximum": 5.859929941182103, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42452607776085516, "standard_deviation": 0.021320375559408123, "minimum": 0.40177190466534013, "maximum": 0.4490155345232677, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06150482204976568, "standard_deviation": 0.05300665906248093, "minimum": -0.13945207322578151, "maximum": 0.012758365136396497, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933729431385825, "standard_deviation": 0.0009681230821861744, "minimum": 0.9915401802566689, "maximum": 0.9947633868772644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.24837816577096, "standard_deviation": 0.035144755788966475, "minimum": 1.1960210769472452, "maximum": 1.3148985331429452, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5199493780269324, "standard_deviation": 0.17405445512980247, "minimum": 0.34034928847913176, "maximum": 0.8344142951565608, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08678566087530504, "standard_deviation": 0.0060323213365670965, "minimum": 0.07776122872696008, "maximum": 0.09799680927812103, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981084410566712, "standard_deviation": 0.00031652758884485904, "minimum": 0.9974986438233115, "maximum": 0.9984590358866469, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6702029441990015, "standard_deviation": 0.02740244290200362, "minimum": 0.6405743945947615, "maximum": 0.7087268607441067, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3629403515192219, "standard_deviation": 0.2509834793272736, "minimum": 0.08430724902190616, "maximum": 0.8142695469352267, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04560823256300059, "standard_deviation": 0.003445769041385452, "minimum": 0.04156825210888844, "maximum": 0.05199852525357148, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1089-134686-0032", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.5845799446105957, "predicted_mos": 2.5845799446105957, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "27061f9f765fd4ad074309bc2383f5eb52c85a04a1c55a5f03f9f25756168fe8", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0448473999276757, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2589876149149877, "全帯域_10ms_rms_envelope_nrmse": 0.08665658156057936, "全帯域_10ms_envelope_correlation": 0.9935315493265785, "全帯域_10ms_modulation_mae_db": 0.5784239904242598, "全帯域_40ms_log_rms_mae_db": 0.7025153892843985, "全帯域_40ms_rms_envelope_nrmse": 0.05933586793721353, "全帯域_40ms_envelope_correlation": 0.9969247726020856, "全帯域_40ms_modulation_mae_db": 0.4408751444274115, "300–1000 Hz_10ms_log_rms_mae_db": 1.3812101254632363, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10704110556057524, "300–1000 Hz_10ms_envelope_correlation": 0.9922639810140719, "300–1000 Hz_10ms_modulation_mae_db": 0.37511106370734454, "300–1000 Hz_40ms_log_rms_mae_db": 1.0039474945363, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06750689506201422, "300–1000 Hz_40ms_envelope_correlation": 0.9973171130128491, "300–1000 Hz_40ms_modulation_mae_db": 0.7186425407245741, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8257484146957703, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23808707047864647, "1000–3000 Hz_10ms_envelope_correlation": 0.9557380619269997, "1000–3000 Hz_10ms_modulation_mae_db": 1.3528151535661719, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1442790200544415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14355530704827632, "1000–3000 Hz_40ms_envelope_correlation": 0.9828690301149055, "1000–3000 Hz_40ms_modulation_mae_db": 1.708985751557704, "3000–6000 Hz_10ms_log_rms_mae_db": 1.998453830395558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32148774278699255, "3000–6000 Hz_10ms_envelope_correlation": 0.9264503716631379, "3000–6000 Hz_10ms_modulation_mae_db": 3.237038600937266, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3653283447753184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2598535193894976, "3000–6000 Hz_40ms_envelope_correlation": 0.9486343228825104, "3000–6000 Hz_40ms_modulation_mae_db": 4.297257802586134, "mr_stft_db": 5.820891611744121, "band_mae_db": 0.21569121329439422, "rms_gain_error_db": -0.14954722061878956, "output_peak": 0.4046717305846427}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2841864799916676, "全帯域_10ms_rms_envelope_nrmse": 0.0901476089175738, "全帯域_10ms_envelope_correlation": 0.9928688197743254, "全帯域_10ms_modulation_mae_db": 0.4073696896715699, "全帯域_40ms_log_rms_mae_db": 0.7861220961994885, "全帯域_40ms_rms_envelope_nrmse": 0.04993138173987952, "全帯域_40ms_envelope_correlation": 0.9976737322229552, "全帯域_40ms_modulation_mae_db": 0.052786437566658066, "300–1000 Hz_10ms_log_rms_mae_db": 1.4385923678327448, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12067545395356305, "300–1000 Hz_10ms_envelope_correlation": 0.9894007428703151, "300–1000 Hz_10ms_modulation_mae_db": 0.8595299607058807, "300–1000 Hz_40ms_log_rms_mae_db": 0.9791809553309736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07006264248013377, "300–1000 Hz_40ms_envelope_correlation": 0.9962686754428176, "300–1000 Hz_40ms_modulation_mae_db": 0.17312834701997415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8007777065619743, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20690781213717, "1000–3000 Hz_10ms_envelope_correlation": 0.967744789993986, "1000–3000 Hz_10ms_modulation_mae_db": 1.3125945643364973, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0624936143310275, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1333859496935669, "1000–3000 Hz_40ms_envelope_correlation": 0.9858319304838215, "1000–3000 Hz_40ms_modulation_mae_db": 1.4265216839573622, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9299157076351054, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3271212428483318, "3000–6000 Hz_10ms_envelope_correlation": 0.9224285405945056, "3000–6000 Hz_10ms_modulation_mae_db": 3.0172447521982644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3110005638927271, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2594097351154503, "3000–6000 Hz_40ms_envelope_correlation": 0.9482249913758777, "3000–6000 Hz_40ms_modulation_mae_db": 3.9851333131622724, "mr_stft_db": 5.818598981288702, "band_mae_db": 0.15591995344925524, "rms_gain_error_db": 0.014207361354279106, "output_peak": 0.3950718171429987}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3252170409402997, "全帯域_10ms_rms_envelope_nrmse": 0.09362427814839763, "全帯域_10ms_envelope_correlation": 0.9922980493813841, "全帯域_10ms_modulation_mae_db": 0.4873472054366709, "全帯域_40ms_log_rms_mae_db": 0.7179381445320318, "全帯域_40ms_rms_envelope_nrmse": 0.050113041221872684, "全帯域_40ms_envelope_correlation": 0.9977414776966202, "全帯域_40ms_modulation_mae_db": 0.15331418517967985, "300–1000 Hz_10ms_log_rms_mae_db": 1.4416822255579544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1228926290489929, "300–1000 Hz_10ms_envelope_correlation": 0.9891955491060532, "300–1000 Hz_10ms_modulation_mae_db": 0.711762279812847, "300–1000 Hz_40ms_log_rms_mae_db": 0.9553606268101155, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0692530992406093, "300–1000 Hz_40ms_envelope_correlation": 0.9967207343568208, "300–1000 Hz_40ms_modulation_mae_db": 0.321146453788919, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7293171469134436, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2022026648669622, "1000–3000 Hz_10ms_envelope_correlation": 0.9680408334899996, "1000–3000 Hz_10ms_modulation_mae_db": 0.9581621757565464, "1000–3000 Hz_40ms_log_rms_mae_db": 1.042804277505211, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11020424435956294, "1000–3000 Hz_40ms_envelope_correlation": 0.9898707713587233, "1000–3000 Hz_40ms_modulation_mae_db": 0.8740150709995977, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8738899677720249, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32606798007157994, "3000–6000 Hz_10ms_envelope_correlation": 0.9251451891987837, "3000–6000 Hz_10ms_modulation_mae_db": 3.2627467485691555, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2833556254346463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25632213020229244, "3000–6000 Hz_40ms_envelope_correlation": 0.9505293405901297, "3000–6000 Hz_40ms_modulation_mae_db": 4.3011696724803405, "mr_stft_db": 5.8100589142476124, "band_mae_db": 0.1335930796686838, "rms_gain_error_db": -0.09583935273220616, "output_peak": 0.42685056104421937}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1639302946012648, "全帯域_10ms_rms_envelope_nrmse": 0.08574198811206839, "全帯域_10ms_envelope_correlation": 0.9935312014343249, "全帯域_10ms_modulation_mae_db": 0.6340876712643763, "全帯域_40ms_log_rms_mae_db": 0.5987903862234621, "全帯域_40ms_rms_envelope_nrmse": 0.03775217362180328, "全帯域_40ms_envelope_correlation": 0.9986983819997933, "全帯域_40ms_modulation_mae_db": 0.1313931691218668, "300–1000 Hz_10ms_log_rms_mae_db": 1.3157366894919271, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11545272195293357, "300–1000 Hz_10ms_envelope_correlation": 0.9903013000974095, "300–1000 Hz_10ms_modulation_mae_db": 1.194279427484214, "300–1000 Hz_40ms_log_rms_mae_db": 0.8369865938843127, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04905760112266988, "300–1000 Hz_40ms_envelope_correlation": 0.9981929229030382, "300–1000 Hz_40ms_modulation_mae_db": 0.37694415411551635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7790773519584275, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20750451914108545, "1000–3000 Hz_10ms_envelope_correlation": 0.9660989364148077, "1000–3000 Hz_10ms_modulation_mae_db": 1.0534767186774776, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0898701690457082, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1249645019307416, "1000–3000 Hz_40ms_envelope_correlation": 0.9876648689195555, "1000–3000 Hz_40ms_modulation_mae_db": 1.0278616814265942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9419934410732402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.33805234466349543, "3000–6000 Hz_10ms_envelope_correlation": 0.9201948766398228, "3000–6000 Hz_10ms_modulation_mae_db": 3.7215826434873436, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3884731714858916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2663313989094991, "3000–6000 Hz_40ms_envelope_correlation": 0.9469411358262072, "3000–6000 Hz_40ms_modulation_mae_db": 4.759146152014012, "mr_stft_db": 5.810010458298533, "band_mae_db": 0.2278549773177505, "rms_gain_error_db": -0.058328457872289095, "output_peak": 0.45752210944030497}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2415486412286176, "全帯域_10ms_rms_envelope_nrmse": 0.08952264658417351, "全帯域_10ms_envelope_correlation": 0.9929759937481829, "全帯域_10ms_modulation_mae_db": 0.2386591565807716, "全帯域_40ms_log_rms_mae_db": 0.6552318259970668, "全帯域_40ms_rms_envelope_nrmse": 0.0448422230179105, "全帯域_40ms_envelope_correlation": 0.9982008925151914, "全帯域_40ms_modulation_mae_db": 0.4950628951596281, "300–1000 Hz_10ms_log_rms_mae_db": 1.384577389998095, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11257461292098529, "300–1000 Hz_10ms_envelope_correlation": 0.9908107897801759, "300–1000 Hz_10ms_modulation_mae_db": 0.4429733595901458, "300–1000 Hz_40ms_log_rms_mae_db": 1.0291109520682171, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06201497620317941, "300–1000 Hz_40ms_envelope_correlation": 0.997160408322664, "300–1000 Hz_40ms_modulation_mae_db": 0.4259727923417815, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8580892305898749, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20577913850414523, "1000–3000 Hz_10ms_envelope_correlation": 0.9665000895011139, "1000–3000 Hz_10ms_modulation_mae_db": 1.3111969715240288, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0935274348374786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11902965048163436, "1000–3000 Hz_40ms_envelope_correlation": 0.9882660861699035, "1000–3000 Hz_40ms_modulation_mae_db": 0.9720195975039244, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9176033499829417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31478442266101625, "3000–6000 Hz_10ms_envelope_correlation": 0.929881634763672, "3000–6000 Hz_10ms_modulation_mae_db": 3.2016367934372503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.295514463308457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25088346615257057, "3000–6000 Hz_40ms_envelope_correlation": 0.9523015477962294, "3000–6000 Hz_40ms_modulation_mae_db": 4.1463218032593145, "mr_stft_db": 5.8140299792037, "band_mae_db": 0.1477598004224617, "rms_gain_error_db": -0.09032862492854882, "output_peak": 0.41601647526380214}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3010821048621135, "全帯域_10ms_rms_envelope_nrmse": 0.09129822641961924, "全帯域_10ms_envelope_correlation": 0.9928055122264797, "全帯域_10ms_modulation_mae_db": 0.6043093672308214, "全帯域_40ms_log_rms_mae_db": 0.702165983393551, "全帯域_40ms_rms_envelope_nrmse": 0.053449381959601416, "全帯域_40ms_envelope_correlation": 0.9975747982604939, "全帯域_40ms_modulation_mae_db": 0.8513818295176101, "300–1000 Hz_10ms_log_rms_mae_db": 1.4334966353851795, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11602179024375872, "300–1000 Hz_10ms_envelope_correlation": 0.9904982474234066, "300–1000 Hz_10ms_modulation_mae_db": 0.3556610973129428, "300–1000 Hz_40ms_log_rms_mae_db": 1.0366613642636402, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07119633558176179, "300–1000 Hz_40ms_envelope_correlation": 0.9966056069873328, "300–1000 Hz_40ms_modulation_mae_db": 0.6045876967782758, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8431772353732023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23875747890879467, "1000–3000 Hz_10ms_envelope_correlation": 0.9573677936126656, "1000–3000 Hz_10ms_modulation_mae_db": 1.4120173518127748, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0997645513813965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1644943790155114, "1000–3000 Hz_40ms_envelope_correlation": 0.9786770319615693, "1000–3000 Hz_40ms_modulation_mae_db": 1.190724105233973, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9655116938318544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.33087087799744713, "3000–6000 Hz_10ms_envelope_correlation": 0.924034871745205, "3000–6000 Hz_10ms_modulation_mae_db": 3.2480897088670093, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3736381703810088, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26407875413686865, "3000–6000 Hz_40ms_envelope_correlation": 0.9481453363579185, "3000–6000 Hz_40ms_modulation_mae_db": 4.282152293795904, "mr_stft_db": 5.815051239113593, "band_mae_db": 0.2406002135888877, "rms_gain_error_db": -0.15529331729057422, "output_peak": 0.4096520328985335}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3241982408825135, "全帯域_10ms_rms_envelope_nrmse": 0.08632857101903815, "全帯域_10ms_envelope_correlation": 0.993636500134822, "全帯域_10ms_modulation_mae_db": 0.33575924270248986, "全帯域_40ms_log_rms_mae_db": 0.7232714452246432, "全帯域_40ms_rms_envelope_nrmse": 0.04424017906488065, "全帯域_40ms_envelope_correlation": 0.9984893235341398, "全帯域_40ms_modulation_mae_db": 0.648851355572019, "300–1000 Hz_10ms_log_rms_mae_db": 1.4030754460397263, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10192928042682048, "300–1000 Hz_10ms_envelope_correlation": 0.9927770489746808, "300–1000 Hz_10ms_modulation_mae_db": 0.42330401882008273, "300–1000 Hz_40ms_log_rms_mae_db": 0.9524544867731695, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05635310646782719, "300–1000 Hz_40ms_envelope_correlation": 0.998052573351351, "300–1000 Hz_40ms_modulation_mae_db": 0.9666595755792287, "1000–3000 Hz_10ms_log_rms_mae_db": 1.728022319074265, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20528852549850982, "1000–3000 Hz_10ms_envelope_correlation": 0.9666433362913849, "1000–3000 Hz_10ms_modulation_mae_db": 1.375119071104382, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0047716894877354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12194840141655458, "1000–3000 Hz_40ms_envelope_correlation": 0.9877023483728072, "1000–3000 Hz_40ms_modulation_mae_db": 1.0340660801636692, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9186502078328818, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3335419324257002, "3000–6000 Hz_10ms_envelope_correlation": 0.9208474606904715, "3000–6000 Hz_10ms_modulation_mae_db": 3.3313183599803353, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3648270124977957, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2677903911985023, "3000–6000 Hz_40ms_envelope_correlation": 0.9454848497467829, "3000–6000 Hz_40ms_modulation_mae_db": 4.354326202256379, "mr_stft_db": 5.825985346699348, "band_mae_db": 0.21248372012031017, "rms_gain_error_db": -0.1669094453440017, "output_peak": 0.3597012582514054}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2777493831808358, "全帯域_10ms_rms_envelope_nrmse": 0.10421223610035858, "全帯域_10ms_envelope_correlation": 0.9905134455520512, "全帯域_10ms_modulation_mae_db": 0.36801741157471146, "全帯域_40ms_log_rms_mae_db": 0.7242254291141867, "全帯域_40ms_rms_envelope_nrmse": 0.06456182595022963, "全帯域_40ms_envelope_correlation": 0.9962813292926167, "全帯域_40ms_modulation_mae_db": 0.6928888542087672, "300–1000 Hz_10ms_log_rms_mae_db": 1.3339199333893206, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11628654994006138, "300–1000 Hz_10ms_envelope_correlation": 0.9902183948275104, "300–1000 Hz_10ms_modulation_mae_db": 0.5821117924864083, "300–1000 Hz_40ms_log_rms_mae_db": 0.9526267086817801, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06970205136227775, "300–1000 Hz_40ms_envelope_correlation": 0.9964441271711222, "300–1000 Hz_40ms_modulation_mae_db": 0.5564935700670708, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7711341264713003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20290623303561273, "1000–3000 Hz_10ms_envelope_correlation": 0.9674824839351449, "1000–3000 Hz_10ms_modulation_mae_db": 1.1992684866812606, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0198754399690508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1105266548825064, "1000–3000 Hz_40ms_envelope_correlation": 0.9902021639211264, "1000–3000 Hz_40ms_modulation_mae_db": 1.113769821663538, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9152077858262524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32813634663190727, "3000–6000 Hz_10ms_envelope_correlation": 0.9235973103835279, "3000–6000 Hz_10ms_modulation_mae_db": 3.2850724144424306, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3143751477673375, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26797884902948427, "3000–6000 Hz_40ms_envelope_correlation": 0.9454485492539723, "3000–6000 Hz_40ms_modulation_mae_db": 4.419302525031601, "mr_stft_db": 5.812064752793816, "band_mae_db": 0.28371234653620814, "rms_gain_error_db": -0.14157583537995228, "output_peak": 0.40217292903551205}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65440, "output_samples": 65440, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.303168019124212, "spectral_convergence": 0.27986063976708436, "spectral_convergence_status": "ok", "amplitude_floor": 1.239800719019687e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.365892181937257, "spectral_convergence": 0.2596863958704033, "spectral_convergence_status": "ok", "amplitude_floor": 7.959069981155994e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.793614634170892, "spectral_convergence": 0.2898911532214947, "spectral_convergence_status": "ok", "amplitude_floor": 5.70263552352201e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002917609391844852, "output_power": 0.00031435635163299985, "delta_db": 0.3239509147090807, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009442146735758444, "output_power": 0.000902007979420232, "delta_db": -0.1986036585264923, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008095018592590047, "output_power": 0.0007533806820122374, "delta_db": -0.3120337062946736, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.140726670679603e-05, "output_power": 7.232617788551391e-05, "delta_db": 0.055531055093005766, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.8558923829027427e-05, "output_power": 3.0106367998539874e-05, "delta_db": 0.2291652770954478, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.629692823024492e-06, "output_power": 8.984245358321379e-06, "delta_db": 0.17486266804766526, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21569121329439422, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14954722061878956, "pitch": null, "output_rms_dbfs": -26.816941469685695, "output_peak": 0.4046717305846427, "reference_rms_dbfs": -26.667394249066906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.820891611744121}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65440, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00016079151656388168, "log_rms_mae_db": 1.2589876149149877, "rms_envelope_nrmse": 0.08665658156057936, "envelope_correlation": 0.9935315493265785, "rms_gain_error_db": -0.14954623762197242, "reference_mean_power": 0.0021567095381351967, "output_mean_power": 0.0020837088450874237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.6620838184249172e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000592954618765589, "output_power": 0.0005746868110814992, "reference_active": true, "delta_db": -0.135902260104536}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014246995030046006, "output_power": 0.0001506217971106077, "reference_active": true, "delta_db": 0.2416455218320932}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.93963265073239e-05, "output_power": 2.6515016708005997e-05, "reference_active": true, "delta_db": 1.3577241893361502}], "active_band_count": 3, "active_band_mae_db": 0.5784239904242598}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00012458719833187296, "log_rms_mae_db": 0.7025153892843985, "rms_envelope_nrmse": 0.05933586793721353, "envelope_correlation": 0.9969247726020856, "rms_gain_error_db": -0.1495428075015685, "reference_mean_power": 0.0021726617564687806, "output_mean_power": 0.002099122767777302}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.5403145620848259e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005607371759141368, "output_power": 0.0005412555568276637, "reference_active": true, "delta_db": -0.15356981942473036}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.842815402518841e-05, "output_power": 7.787830166388907e-05, "reference_active": true, "delta_db": 0.5618164778111685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9268319788025257e-06, "output_power": 2.2159898105258467e-06, "reference_active": true, "delta_db": 0.6072391360463357}], "active_band_count": 3, "active_band_mae_db": 0.4408751444274115}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00013740128966162354, "log_rms_mae_db": 1.3812101254632363, "rms_envelope_nrmse": 0.10704110556057524, "envelope_correlation": 0.9922639810140719, "rms_gain_error_db": -0.29956310882019566, "reference_mean_power": 0.001344215530062727, "output_mean_power": 0.0012546211305468325}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.2437342579426783e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004237422220408671, "output_power": 0.00039560871000816233, "reference_active": true, "delta_db": -0.29835894861622153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010753440950326828, "output_power": 9.48351476271364e-05, "reference_active": true, "delta_db": -0.5457813004461568}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.157936749643158e-05, "output_power": 1.2353901424131838e-05, "reference_active": true, "delta_db": 0.28119294205965517}], "active_band_count": 3, "active_band_mae_db": 0.37511106370734454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00011336523032200158, "log_rms_mae_db": 1.0039474945363, "rms_envelope_nrmse": 0.06750689506201422, "envelope_correlation": 0.9973171130128491, "rms_gain_error_db": -0.2995632688635413, "reference_mean_power": 0.0013541602801659725, "output_mean_power": 0.0012639029984687663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.1658618871138855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004110855381392752, "output_power": 0.00038063681896516587, "reference_active": true, "delta_db": -0.3342140345600495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.54169136496196e-05, "output_power": 4.1241756770309125e-05, "reference_active": true, "delta_db": -0.41880461387109497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3416624074278087e-06, "output_power": 9.712976855954679e-07, "reference_active": true, "delta_db": -1.4029089737425777}], "active_band_count": 3, "active_band_mae_db": 0.7186425407245741}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 4.166087007180048e-05, "log_rms_mae_db": 1.8257484146957703, "rms_envelope_nrmse": 0.23808707047864647, "envelope_correlation": 0.9557380619269997, "rms_gain_error_db": -0.0637494241462399, "reference_mean_power": 7.742789466080943e-05, "output_mean_power": 7.629964338532015e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.015275941663665e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8994572701205476e-05, "output_power": 3.736068063296708e-05, "reference_active": true, "delta_db": -0.1858938618451392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.4522009847535795e-06, "output_power": 8.78548480373592e-06, "reference_active": true, "delta_db": 0.714811726447611}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.182412963771591e-06, "output_power": 2.4464884504130717e-06, "reference_active": true, "delta_db": 3.1577398724057657}], "active_band_count": 3, "active_band_mae_db": 1.3528151535661719}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 3.308768412789438e-05, "log_rms_mae_db": 1.1442790200544415, "rms_envelope_nrmse": 0.14355530704827632, "envelope_correlation": 0.9828690301149055, "rms_gain_error_db": -0.06374919966312075, "reference_mean_power": 7.800066882867515e-05, "output_mean_power": 7.686407526822145e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 6.505338847209246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.845568227980419e-05, "output_power": 3.781741235246546e-05, "reference_active": true, "delta_db": -0.07268711973762443}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.058614755593344e-06, "output_power": 3.851347949878049e-06, "reference_active": true, "delta_db": 1.000879773603033}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905753913870012e-08, "output_power": 2.5189907390972355e-07, "reference_active": true, "delta_db": 4.053390361332454}], "active_band_count": 3, "active_band_mae_db": 1.708985751557704}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 1.92244611767394e-05, "log_rms_mae_db": 1.998453830395558, "rms_envelope_nrmse": 0.32148774278699255, "envelope_correlation": 0.9264503716631379, "rms_gain_error_db": -0.050397009271274856, "reference_mean_power": 1.4177032937223143e-05, "output_mean_power": 1.4013468686672071e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.931109844663442e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.831691262181727e-06, "output_power": 4.232487548850482e-06, "reference_active": true, "delta_db": 0.4320518055527085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0928531053800205e-06, "output_power": 1.3828832166767717e-06, "reference_active": true, "delta_db": 1.0222371518588065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4875100105629565e-07, "output_power": 9.957322689339445e-07, "reference_active": true, "delta_db": 8.256826845400282}], "active_band_count": 3, "active_band_mae_db": 3.237038600937266}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 1.4099069206239137e-05, "log_rms_mae_db": 1.3653283447753184, "rms_envelope_nrmse": 0.2598535193894976, "envelope_correlation": 0.9486343228825104, "rms_gain_error_db": -0.050398457231097135, "reference_mean_power": 1.4281884811465133e-05, "output_mean_power": 1.4117106149861559e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 7.397985811629507e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.714499411475195e-06, "output_power": 4.736163671459649e-06, "reference_active": true, "delta_db": 1.0552640836663545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.071953545728327e-07, "output_power": 7.677480583290909e-07, "reference_active": true, "delta_db": 1.8004345959794827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3628863245634334e-08, "output_power": 1.3742543064789312e-07, "reference_active": true, "delta_db": 10.036074728112563}], "active_band_count": 3, "active_band_mae_db": 4.297257802586134}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8513987065010943, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "948d077f20d5b4677267a0f15712b927c10dd5eb06c4349a8e864c038ecdf7ba", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9644520406457627, "standard_deviation": 0.004938677474945797, "minimum": 0.9557380619269997, "maximum": 0.9680408334899996, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7919179414547823, "standard_deviation": 0.04902988711965224, "minimum": 1.728022319074265, "maximum": 1.8580892305898749, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2468313116823926, "standard_deviation": 0.163192518374015, "minimum": 0.9581621757565464, "maximum": 1.4120173518127748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21342918032136582, "standard_deviation": 0.015532107662213988, "minimum": 0.2022026648669622, "maximum": 0.23875747890879467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9863855289128015, "standard_deviation": 0.0038861291408827285, "minimum": 0.9786770319615693, "maximum": 0.9902021639211264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0696732745765063, "standard_deviation": 0.04615806431554734, "minimum": 1.0047716894877354, "maximum": 1.1442790200544415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1684954740632953, "standard_deviation": 0.27395695114548707, "minimum": 0.8740150709995977, "maximum": 1.708985751557704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1285136361035443, "standard_deviation": 0.01831866779176419, "minimum": 0.11020424435956294, "maximum": 0.1644943790155114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9240725319598908, "standard_deviation": 0.003137734567517147, "minimum": 0.9201948766398228, "maximum": 0.929881634763672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9326532480437324, "standard_deviation": 0.03719481388163701, "minimum": 1.8738899677720249, "maximum": 1.998453830395558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.288091252739882, "standard_deviation": 0.1983768142724186, "minimum": 3.0172447521982644, "maximum": 3.7215826434873436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3275078612608088, "standard_deviation": 0.00716789935717908, "minimum": 0.31478442266101625, "maximum": 0.33805234466349543, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9482137592287035, "standard_deviation": 0.002359373934540102, "minimum": 0.9454485492539723, "maximum": 0.9523015477962294, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3370640624428978, "standard_deviation": 0.040276315471134076, "minimum": 1.2833556254346463, "maximum": 1.3884731714858916, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.318101220573245, "standard_deviation": 0.2231155469964301, "minimum": 3.9851333131622724, "maximum": 4.759146152014012, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2615810305167706, "standard_deviation": 0.006072015424469333, "minimum": 0.25088346615257057, "maximum": 0.26797884902948427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990683256761703, "standard_deviation": 0.0012622684557546486, "minimum": 0.9891955491060532, "maximum": 0.9927770489746808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.391536351644773, "standard_deviation": 0.04757399044864658, "minimum": 1.3157366894919271, "maximum": 1.4416822255579544, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6180916249899833, "standard_deviation": 0.29200575997999145, "minimum": 0.3556610973129428, "maximum": 1.194279427484214, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11410926800596133, "standard_deviation": 0.006877827517893222, "minimum": 0.10192928042682048, "maximum": 0.1228926290489929, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970952701934994, "standard_deviation": 0.0007232882536457703, "minimum": 0.9962686754428176, "maximum": 0.9981929229030382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9682911477935636, "standard_deviation": 0.06286920299727135, "minimum": 0.8369865938843127, "maximum": 1.0366613642636402, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5179468913019176, "standard_deviation": 0.24973858898729637, "minimum": 0.17312834701997415, "maximum": 0.9666595755792287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06439333844005916, "standard_deviation": 0.007976774297395427, "minimum": 0.04905760112266988, "maximum": 0.07119633558176179, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20220191304974394, "standard_deviation": 0.05193027215851749, "minimum": 0.1335930796686838, "maximum": 0.28371234653620814, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.815836410423678, "standard_deviation": 0.005626646234615782, "minimum": 5.810010458298533, "maximum": 5.825985346699348, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40895736420767737, "standard_deviation": 0.027806428325250657, "minimum": 0.3597012582514054, "maximum": 0.45752210944030497, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10545186160151034, "standard_deviation": 0.061377758773863374, "minimum": -0.1669094453440017, "maximum": 0.014207361354279106, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9927701339472685, "standard_deviation": 0.0010196948138556697, "minimum": 0.9905134455520512, "maximum": 0.993636500134822, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2721124750752875, "standard_deviation": 0.052589531493419876, "minimum": 1.1639302946012648, "maximum": 1.3252170409402997, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45674671686070895, "standard_deviation": 0.14216561367373928, "minimum": 0.2386591565807716, "maximum": 0.6340876712643763, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09094151710772608, "standard_deviation": 0.006008980906347937, "minimum": 0.08574198811206839, "maximum": 0.10421223610035858, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997698088515487, "standard_deviation": 0.0008019932584430236, "minimum": 0.9962813292926167, "maximum": 0.9986983819997933, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7012825874961035, "standard_deviation": 0.054918976433913004, "minimum": 0.5987903862234621, "maximum": 0.7861220961994885, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43331923384420507, "standard_deviation": 0.29443494881610727, "minimum": 0.052786437566658066, "maximum": 0.8513818295176101, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0505282593141739, "standard_deviation": 0.008607935841839137, "minimum": 0.03775217362180328, "maximum": 0.06456182595022963, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1188-133604-0003", "language": "en", "style": "normal", "speaker": "librispeech_1188", "source_utterance_id": "1188-133604-0003", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1188-133604-0003.wav", "source_file_sha256": "798fc336114800a819f5c76bb6cc5239c283b0b7fc7455e1d0ca23a2f3aaa6bd", "decoded_pcm_sha256": "81283c3eb08ea201c5ec2d52e3a8e0ed4d0258d76164b522d1ee5ec43edebe81", "original_source_sha256": "a013400c991a40062cb4f458d247cfd6dac24fa0016bf6e2eac7a3e8e08b8ae8", "corpus_entry_id": "librispeech_1188-133604-0003"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.435666084289551, "predicted_mos": 3.435666084289551, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7cc851975e60146506e013525ba5cbeee0496db0d607e9aeb5884216aa9c765e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1974268001504242, "cpu_threads": 4}, "waveform_sha256": "d03996160873a22360bde5b073fc057837439fa39b35136ceaeb8ffda73d15c2"}, "rows": [{"case": "librispeech_1188-133604-0003", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.590787410736084, "predicted_mos": 2.590787410736084, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c8ecbbeed9b4d46842d7abb29cb034209d4d02af28be218cc810058114c82317", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.120109200011939, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.554532395088962, "全帯域_10ms_rms_envelope_nrmse": 0.09869793506641528, "全帯域_10ms_envelope_correlation": 0.9914847573238628, "全帯域_10ms_modulation_mae_db": 0.42537120004936274, "全帯域_40ms_log_rms_mae_db": 0.8028751202558868, "全帯域_40ms_rms_envelope_nrmse": 0.05059332699617219, "全帯域_40ms_envelope_correlation": 0.9977310229341074, "全帯域_40ms_modulation_mae_db": 0.21866030686454238, "300–1000 Hz_10ms_log_rms_mae_db": 1.7772481546794257, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1286502043488684, "300–1000 Hz_10ms_envelope_correlation": 0.989445257816107, "300–1000 Hz_10ms_modulation_mae_db": 0.9290302903455164, "300–1000 Hz_40ms_log_rms_mae_db": 1.1802490639750736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08884439535147232, "300–1000 Hz_40ms_envelope_correlation": 0.9952117584465646, "300–1000 Hz_40ms_modulation_mae_db": 0.9467295579397268, "1000–3000 Hz_10ms_log_rms_mae_db": 1.703728368283872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2090360322790498, "1000–3000 Hz_10ms_envelope_correlation": 0.9689313005276778, "1000–3000 Hz_10ms_modulation_mae_db": 0.9526387368837098, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0441788586918421, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12908003237063542, "1000–3000 Hz_40ms_envelope_correlation": 0.9877910889037925, "1000–3000 Hz_40ms_modulation_mae_db": 0.9904190482107021, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5121493648557571, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25446718222172027, "3000–6000 Hz_10ms_envelope_correlation": 0.9612127018697662, "3000–6000 Hz_10ms_modulation_mae_db": 1.3397898103751713, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5031078125547528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16022944611612353, "3000–6000 Hz_40ms_envelope_correlation": 0.9844775553109052, "3000–6000 Hz_40ms_modulation_mae_db": 1.2135808667096823, "mr_stft_db": 5.809832433425584, "band_mae_db": 0.37913773824770997, "rms_gain_error_db": -0.10239481139990664, "output_peak": 0.24235998564702885}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4839453894553873, "全帯域_10ms_rms_envelope_nrmse": 0.1032948536936459, "全帯域_10ms_envelope_correlation": 0.9906452696330684, "全帯域_10ms_modulation_mae_db": 0.2842968138480287, "全帯域_40ms_log_rms_mae_db": 0.7622996332763219, "全帯域_40ms_rms_envelope_nrmse": 0.052220870911489546, "全帯域_40ms_envelope_correlation": 0.9975109672823499, "全帯域_40ms_modulation_mae_db": 0.32607791942150016, "300–1000 Hz_10ms_log_rms_mae_db": 1.8620269641245255, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13610371051562803, "300–1000 Hz_10ms_envelope_correlation": 0.9878957694422038, "300–1000 Hz_10ms_modulation_mae_db": 0.5105661049308369, "300–1000 Hz_40ms_log_rms_mae_db": 1.2650594476287087, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0824902574737583, "300–1000 Hz_40ms_envelope_correlation": 0.9957218469763223, "300–1000 Hz_40ms_modulation_mae_db": 0.480921863510586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.670182760167976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2070964932518707, "1000–3000 Hz_10ms_envelope_correlation": 0.9695854885821942, "1000–3000 Hz_10ms_modulation_mae_db": 0.7211541570842456, "1000–3000 Hz_40ms_log_rms_mae_db": 1.114679453365703, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12798152710135174, "1000–3000 Hz_40ms_envelope_correlation": 0.9879032270141319, "1000–3000 Hz_40ms_modulation_mae_db": 0.7720083823224845, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5399878330561678, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19263776856836162, "3000–6000 Hz_10ms_envelope_correlation": 0.979337554756885, "3000–6000 Hz_10ms_modulation_mae_db": 0.7604051279960835, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4495340045531393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12670522611035603, "3000–6000 Hz_40ms_envelope_correlation": 0.9910794445781799, "3000–6000 Hz_40ms_modulation_mae_db": 0.7295768876684051, "mr_stft_db": 5.8085432647492015, "band_mae_db": 0.2199346659725652, "rms_gain_error_db": -0.016423973738427085, "output_peak": 0.2562074231053636}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.600406613062993, "全帯域_10ms_rms_envelope_nrmse": 0.10335087287624957, "全帯域_10ms_envelope_correlation": 0.9906281180900823, "全帯域_10ms_modulation_mae_db": 0.2624340167386522, "全帯域_40ms_log_rms_mae_db": 0.828776338614287, "全帯域_40ms_rms_envelope_nrmse": 0.05610979100753884, "全帯域_40ms_envelope_correlation": 0.9970833798215818, "全帯域_40ms_modulation_mae_db": 0.3477722545913049, "300–1000 Hz_10ms_log_rms_mae_db": 1.7211129409427166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12248530388119426, "300–1000 Hz_10ms_envelope_correlation": 0.9900421518317237, "300–1000 Hz_10ms_modulation_mae_db": 0.21932183462008936, "300–1000 Hz_40ms_log_rms_mae_db": 1.1257241272222376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06548606827238032, "300–1000 Hz_40ms_envelope_correlation": 0.9971326031975941, "300–1000 Hz_40ms_modulation_mae_db": 0.5123872364659644, "1000–3000 Hz_10ms_log_rms_mae_db": 1.695730278749033, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2054327554554117, "1000–3000 Hz_10ms_envelope_correlation": 0.9699995432985364, "1000–3000 Hz_10ms_modulation_mae_db": 1.0096739875348621, "1000–3000 Hz_40ms_log_rms_mae_db": 1.080732100858897, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12821473452139476, "1000–3000 Hz_40ms_envelope_correlation": 0.9879734670399906, "1000–3000 Hz_40ms_modulation_mae_db": 1.2690913094339467, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5424947617546736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1911498302519893, "3000–6000 Hz_10ms_envelope_correlation": 0.9783650950810677, "3000–6000 Hz_10ms_modulation_mae_db": 0.42089377103338904, "3000–6000 Hz_40ms_log_rms_mae_db": 1.442417182022107, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1296121692162446, "3000–6000 Hz_40ms_envelope_correlation": 0.99007689508837, "3000–6000 Hz_40ms_modulation_mae_db": 0.33064176448145743, "mr_stft_db": 5.803344351379899, "band_mae_db": 0.1955382259066442, "rms_gain_error_db": -0.013413258533786594, "output_peak": 0.25387262903675695}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5681278981459243, "全帯域_10ms_rms_envelope_nrmse": 0.10212597442460333, "全帯域_10ms_envelope_correlation": 0.9909774065881958, "全帯域_10ms_modulation_mae_db": 0.6584429966486615, "全帯域_40ms_log_rms_mae_db": 0.9049288266248436, "全帯域_40ms_rms_envelope_nrmse": 0.04836622234988286, "全帯域_40ms_envelope_correlation": 0.9980690376318045, "全帯域_40ms_modulation_mae_db": 0.5344766641715898, "300–1000 Hz_10ms_log_rms_mae_db": 1.7749125243389876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13871288364764983, "300–1000 Hz_10ms_envelope_correlation": 0.9875518334691662, "300–1000 Hz_10ms_modulation_mae_db": 0.7508787408941794, "300–1000 Hz_40ms_log_rms_mae_db": 1.184999587320021, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07719258631071621, "300–1000 Hz_40ms_envelope_correlation": 0.9965472062051777, "300–1000 Hz_40ms_modulation_mae_db": 0.9015072340670969, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7204008324245799, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22464184694634165, "1000–3000 Hz_10ms_envelope_correlation": 0.964065050671337, "1000–3000 Hz_10ms_modulation_mae_db": 0.9048823457734549, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0778769341006296, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1569047073259681, "1000–3000 Hz_40ms_envelope_correlation": 0.9817481616264243, "1000–3000 Hz_40ms_modulation_mae_db": 1.06453927976943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5216587775817414, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22606749667575787, "3000–6000 Hz_10ms_envelope_correlation": 0.9697935710443322, "3000–6000 Hz_10ms_modulation_mae_db": 1.881818967016083, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4860880570850632, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14422877219678176, "3000–6000 Hz_40ms_envelope_correlation": 0.9882662785649954, "3000–6000 Hz_40ms_modulation_mae_db": 1.5319168894545505, "mr_stft_db": 5.818375771133894, "band_mae_db": 0.2377714751916019, "rms_gain_error_db": -0.1474747569772579, "output_peak": 0.25134429531421304}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5926252006458974, "全帯域_10ms_rms_envelope_nrmse": 0.0988377655304632, "全帯域_10ms_envelope_correlation": 0.9915801229361356, "全帯域_10ms_modulation_mae_db": 0.4427435043912182, "全帯域_40ms_log_rms_mae_db": 0.872377173075587, "全帯域_40ms_rms_envelope_nrmse": 0.0506391598830672, "全帯域_40ms_envelope_correlation": 0.9978991646041752, "全帯域_40ms_modulation_mae_db": 0.32583344121073327, "300–1000 Hz_10ms_log_rms_mae_db": 1.9050384451660791, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1280365448951502, "300–1000 Hz_10ms_envelope_correlation": 0.9896522752480603, "300–1000 Hz_10ms_modulation_mae_db": 0.554488027613902, "300–1000 Hz_40ms_log_rms_mae_db": 1.2776716701252073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07767322563781083, "300–1000 Hz_40ms_envelope_correlation": 0.9966687267550861, "300–1000 Hz_40ms_modulation_mae_db": 0.7440542861254836, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6842149500119152, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2042561543169133, "1000–3000 Hz_10ms_envelope_correlation": 0.9705581996942632, "1000–3000 Hz_10ms_modulation_mae_db": 0.5719358360158967, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0592103562289241, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12111248987307308, "1000–3000 Hz_40ms_envelope_correlation": 0.9899400406996416, "1000–3000 Hz_40ms_modulation_mae_db": 0.8690694588676035, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5790731181240447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20282766138462066, "3000–6000 Hz_10ms_envelope_correlation": 0.9755863914286191, "3000–6000 Hz_10ms_modulation_mae_db": 0.8637382840673614, "3000–6000 Hz_40ms_log_rms_mae_db": 1.500937519543041, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13459633845732494, "3000–6000 Hz_40ms_envelope_correlation": 0.9889722235620557, "3000–6000 Hz_40ms_modulation_mae_db": 0.6732838828432738, "mr_stft_db": 5.786439435637711, "band_mae_db": 0.2547678116624135, "rms_gain_error_db": -0.1604200392539309, "output_peak": 0.2444203669683282}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6055683971077266, "全帯域_10ms_rms_envelope_nrmse": 0.11727932657976055, "全帯域_10ms_envelope_correlation": 0.9881259522314774, "全帯域_10ms_modulation_mae_db": 0.5029155207400385, "全帯域_40ms_log_rms_mae_db": 0.7884527706908171, "全帯域_40ms_rms_envelope_nrmse": 0.06921612494826795, "全帯域_40ms_envelope_correlation": 0.9956568986390125, "全帯域_40ms_modulation_mae_db": 0.5564608258956266, "300–1000 Hz_10ms_log_rms_mae_db": 1.7945789406137242, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15206778952368868, "300–1000 Hz_10ms_envelope_correlation": 0.9849105391127466, "300–1000 Hz_10ms_modulation_mae_db": 0.9879677753143827, "300–1000 Hz_40ms_log_rms_mae_db": 1.176525450691519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0994850550953301, "300–1000 Hz_40ms_envelope_correlation": 0.993346728030323, "300–1000 Hz_40ms_modulation_mae_db": 1.0700147184722593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7740271596197275, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21764469685870463, "1000–3000 Hz_10ms_envelope_correlation": 0.9668633184615537, "1000–3000 Hz_10ms_modulation_mae_db": 0.6212911713413578, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1149537605223696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14245550228707357, "1000–3000 Hz_40ms_envelope_correlation": 0.9851211811536661, "1000–3000 Hz_40ms_modulation_mae_db": 0.6399442245096948, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5861175811796742, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22796283520783722, "3000–6000 Hz_10ms_envelope_correlation": 0.9692339317229951, "3000–6000 Hz_10ms_modulation_mae_db": 1.09292777533142, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4804703544974758, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1422328587030385, "3000–6000 Hz_40ms_envelope_correlation": 0.9877018238122411, "3000–6000 Hz_40ms_modulation_mae_db": 0.5827139080172198, "mr_stft_db": 5.80582554069498, "band_mae_db": 0.1877695001525398, "rms_gain_error_db": 0.04807538439004945, "output_peak": 0.3586965593401172}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.538531761618464, "全帯域_10ms_rms_envelope_nrmse": 0.11019166665285217, "全帯域_10ms_envelope_correlation": 0.9894051372251798, "全帯域_10ms_modulation_mae_db": 0.23388383948302752, "全帯域_40ms_log_rms_mae_db": 0.7876638842173674, "全帯域_40ms_rms_envelope_nrmse": 0.056321124564190486, "全帯域_40ms_envelope_correlation": 0.9970756836673335, "全帯域_40ms_modulation_mae_db": 0.12308899301132543, "300–1000 Hz_10ms_log_rms_mae_db": 1.879214975747619, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1409994584878352, "300–1000 Hz_10ms_envelope_correlation": 0.9868084307831937, "300–1000 Hz_10ms_modulation_mae_db": 0.6059350322013363, "300–1000 Hz_40ms_log_rms_mae_db": 1.236969906443642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07158226236681314, "300–1000 Hz_40ms_envelope_correlation": 0.9965710726759124, "300–1000 Hz_40ms_modulation_mae_db": 0.4212389346651609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.725428492210084, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21729758936072296, "1000–3000 Hz_10ms_envelope_correlation": 0.9665012815771611, "1000–3000 Hz_10ms_modulation_mae_db": 0.6836050157873483, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0779090184361602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1409822708591153, "1000–3000 Hz_40ms_envelope_correlation": 0.9859408082098999, "1000–3000 Hz_40ms_modulation_mae_db": 0.961055175622349, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5140437150509776, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18879457210284295, "3000–6000 Hz_10ms_envelope_correlation": 0.9788318077527931, "3000–6000 Hz_10ms_modulation_mae_db": 0.22262271271946552, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4439300352782025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.131657501880564, "3000–6000 Hz_40ms_envelope_correlation": 0.989495714214223, "3000–6000 Hz_40ms_modulation_mae_db": 0.05899222419034367, "mr_stft_db": 5.794501615116194, "band_mae_db": 0.2370296594390203, "rms_gain_error_db": 0.01753866702118356, "output_peak": 0.2675976978467967}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5134890243510413, "全帯域_10ms_rms_envelope_nrmse": 0.10626131179644929, "全帯域_10ms_envelope_correlation": 0.9901375286251801, "全帯域_10ms_modulation_mae_db": 0.4450879061421195, "全帯域_40ms_log_rms_mae_db": 0.76902617292814, "全帯域_40ms_rms_envelope_nrmse": 0.058541595825914526, "全帯域_40ms_envelope_correlation": 0.9969667697149134, "全帯域_40ms_modulation_mae_db": 0.3109122749834071, "300–1000 Hz_10ms_log_rms_mae_db": 1.831231217562549, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1394451288209072, "300–1000 Hz_10ms_envelope_correlation": 0.9875984154131509, "300–1000 Hz_10ms_modulation_mae_db": 0.7246172562615136, "300–1000 Hz_40ms_log_rms_mae_db": 1.1840528897187605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0902834970604797, "300–1000 Hz_40ms_envelope_correlation": 0.9952298128442361, "300–1000 Hz_40ms_modulation_mae_db": 0.6117405551195779, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7210378634431325, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2013828757854129, "1000–3000 Hz_10ms_envelope_correlation": 0.9712619682902632, "1000–3000 Hz_10ms_modulation_mae_db": 1.1673426697764484, "1000–3000 Hz_40ms_log_rms_mae_db": 1.031458392309991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12434973573239072, "1000–3000 Hz_40ms_envelope_correlation": 0.9889954914488162, "1000–3000 Hz_40ms_modulation_mae_db": 1.1547560778010588, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5367005472630542, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20189316717394507, "3000–6000 Hz_10ms_envelope_correlation": 0.9758083627399944, "3000–6000 Hz_10ms_modulation_mae_db": 0.777613182052416, "3000–6000 Hz_40ms_log_rms_mae_db": 1.446200859091463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13597435460775203, "3000–6000 Hz_40ms_envelope_correlation": 0.9888340206627424, "3000–6000 Hz_40ms_modulation_mae_db": 0.7303758557260207, "mr_stft_db": 5.805716876537908, "band_mae_db": 0.31613760024336535, "rms_gain_error_db": -0.12279853462007148, "output_peak": 0.24853916109561802}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.011752041770961, "spectral_convergence": 0.3210674673555751, "spectral_convergence_status": "ok", "amplitude_floor": 8.79928282883349e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.625405037424589, "spectral_convergence": 0.28247737702665116, "spectral_convergence_status": "ok", "amplitude_floor": 4.444740709272264e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.792340221081203, "spectral_convergence": 0.30633899119133295, "spectral_convergence_status": "ok", "amplitude_floor": 3.930299196146418e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002477612487613689, "output_power": 0.0002575699299753091, "delta_db": 0.16861778621954163, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004515434578601184, "output_power": 0.00044011327960362357, "delta_db": -0.11135081636541323, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015370997116088465, "output_power": 0.00013711163144528992, "delta_db": -0.4962774268818605, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.4065683577917706e-05, "output_power": 4.13124661436287e-05, "delta_db": -0.28019390675189093, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.662327944441642e-06, "output_power": 1.0355275379429551e-05, "delta_db": 0.7752703111504236, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.476431620775013e-05, "output_power": 2.236221005077418e-05, "delta_db": -0.4431161821171299, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.37913773824770997, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10239481139990664, "pitch": null, "output_rms_dbfs": -30.41519833985982, "output_peak": 0.24235998564702885, "reference_rms_dbfs": -30.312803528459913, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.809832433425584}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.467120737686064e-05, "log_rms_mae_db": 1.554532395088962, "rms_envelope_nrmse": 0.09869793506641528, "envelope_correlation": 0.9914847573238628, "rms_gain_error_db": -0.1023977364359313, "reference_mean_power": 0.0009312818437003785, "output_mean_power": 0.0009095809632745334}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.494393820063344e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023666657525761142, "output_power": 0.00023353160929158955, "reference_active": true, "delta_db": -0.057912541546536564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5441929621447868e-05, "output_power": 1.4838539802968578e-05, "reference_active": true, "delta_db": -0.17310402781571144}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.985747798516416e-06, "output_power": 5.49165441282357e-06, "reference_active": true, "delta_db": -1.04509703078584}], "active_band_count": 3, "active_band_mae_db": 0.42537120004936274}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.334380857742655e-05, "log_rms_mae_db": 0.8028751202558868, "rms_envelope_nrmse": 0.05059332699617219, "envelope_correlation": 0.9977310229341074, "rms_gain_error_db": -0.10238964275158416, "reference_mean_power": 0.0009359486205868301, "output_mean_power": 0.0009141406977970787}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.189856446091556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021037225251172397, "output_power": 0.00020906868023184134, "reference_active": true, "delta_db": -0.026994793615930053}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.644252267790675e-06, "output_power": 7.3246811246470596e-06, "reference_active": true, "delta_db": -0.18546288059013055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4342900916585017e-07, "output_power": 3.100877803003501e-07, "reference_active": true, "delta_db": -0.44352324638756657}], "active_band_count": 3, "active_band_mae_db": 0.21866030686454238}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.589555428061273e-05, "log_rms_mae_db": 1.7772481546794257, "rms_envelope_nrmse": 0.1286502043488684, "envelope_correlation": 0.989445257816107, "rms_gain_error_db": -0.2811871056865553, "reference_mean_power": 0.00038327408961782034, "output_mean_power": 0.0003592450147846601}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.028222657916465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015396875485141256, "output_power": 0.0001423519022922007, "reference_active": true, "delta_db": -0.3406932250689088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.75189127488202e-06, "output_power": 6.956496591102554e-06, "reference_active": true, "delta_db": -0.9971133711568224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1315883357233815e-06, "output_power": 2.9592974177422254e-06, "reference_active": true, "delta_db": -1.449284274810818}], "active_band_count": 3, "active_band_mae_db": 0.9290302903455164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.149058869949865e-05, "log_rms_mae_db": 1.1802490639750736, "rms_envelope_nrmse": 0.08884439535147232, "envelope_correlation": 0.9952117584465646, "rms_gain_error_db": -0.281191614987801, "reference_mean_power": 0.00038520090394369153, "output_mean_power": 0.00036105065408457727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9052955697898775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014058163319658512, "output_power": 0.00013145155485828845, "reference_active": true, "delta_db": -0.29162856922734853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.901478078483036e-06, "output_power": 3.012593952952825e-06, "reference_active": true, "delta_db": -1.1228857103792649}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3393802456709067e-07, "output_power": 1.6847419071724561e-07, "reference_active": true, "delta_db": -1.425674394212567}], "active_band_count": 3, "active_band_mae_db": 0.9467295579397268}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.043858753489399e-05, "log_rms_mae_db": 1.703728368283872, "rms_envelope_nrmse": 0.2090360322790498, "envelope_correlation": 0.9689313005276778, "rms_gain_error_db": -0.1613095110352304, "reference_mean_power": 4.084610004445957e-05, "output_mean_power": 3.935678783827196e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0558567323423266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3562756220605441e-05, "output_power": 1.2248838565816209e-05, "reference_active": true, "delta_db": -0.4425304486336934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5216123885214767e-06, "output_power": 1.0721291314548526e-06, "reference_active": true, "delta_db": -1.5205693886109717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602773162915538e-07, "output_power": 5.373345177971829e-07, "reference_active": true, "delta_db": -0.8948163734064641}], "active_band_count": 3, "active_band_mae_db": 0.9526387368837098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.7804457921821952e-05, "log_rms_mae_db": 1.0441788586918421, "rms_envelope_nrmse": 0.12908003237063542, "envelope_correlation": 0.9877910889037925, "rms_gain_error_db": -0.16134117711206444, "reference_mean_power": 4.105094002568101e-05, "output_mean_power": 3.9553870632342295e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8992373996258767e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2457002317192863e-05, "output_power": 1.136754747397731e-05, "reference_active": true, "delta_db": -0.3974676862077823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.119435793981677e-07, "output_power": 5.625056398258603e-07, "reference_active": true, "delta_db": -1.593989707390791}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.742155092447501e-08, "output_power": 2.986357577578075e-08, "reference_active": true, "delta_db": -0.9797997510335331}], "active_band_count": 3, "active_band_mae_db": 0.9904190482107021}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.7924378350629476e-05, "log_rms_mae_db": 1.5121493648557571, "rms_envelope_nrmse": 0.25446718222172027, "envelope_correlation": 0.9612127018697662, "rms_gain_error_db": -0.2359919190149098, "reference_mean_power": 2.620464364014996e-05, "output_mean_power": 2.4818702096368736e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.347390510898905e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.954495563916905e-05, "output_power": 1.886689444598383e-05, "reference_active": true, "delta_db": -0.15334269337645584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.79596482520383e-06, "output_power": 3.2121770414978597e-06, "reference_active": true, "delta_db": 0.6026777028446013}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.14743402003256e-06, "output_power": 5.41243655494848e-07, "reference_active": true, "delta_db": -3.263349034904457}], "active_band_count": 3, "active_band_mae_db": 1.3397898103751713}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.7050501746546627e-05, "log_rms_mae_db": 1.5031078125547528, "rms_envelope_nrmse": 0.16022944611612353, "envelope_correlation": 0.9844775553109052, "rms_gain_error_db": -0.23599333228715147, "reference_mean_power": 2.633641752064706e-05, "output_mean_power": 2.4943498450186272e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2809655135480154e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9754827323518705e-05, "output_power": 1.9124894924750416e-05, "reference_active": true, "delta_db": -0.14074180127742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4198245590923022e-06, "output_power": 1.5886807938857547e-06, "reference_active": true, "delta_db": 0.48801961224379015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.51727649987637e-08, "output_power": 3.257377018163373e-08, "reference_active": true, "delta_db": -3.011981186607837}], "active_band_count": 3, "active_band_mae_db": 1.2135808667096823}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7202051493575338, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bd75afd4d6d0242af113ca0b529cef61f985e26ffbdc52d00b322be9ab54c973", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9684707688878733, "standard_deviation": 0.0024440344925636265, "minimum": 0.964065050671337, "maximum": 0.9712619682902632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.71184383811379, "standard_deviation": 0.03169851072016113, "minimum": 1.670182760167976, "maximum": 1.7740271596197275, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8290654900246655, "standard_deviation": 0.2104956241438406, "minimum": 0.5719358360158967, "maximum": 1.1673426697764484, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21084855553180346, "standard_deviation": 0.008086897793955364, "minimum": 0.2013828757854129, "maximum": 0.22464184694634165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9869266832620454, "standard_deviation": 0.0025968595116783234, "minimum": 0.9817481616264243, "maximum": 0.9899400406996416, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0751248593143146, "standard_deviation": 0.030036550593797524, "minimum": 1.031458392309991, "maximum": 1.1149537605223696, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9651103695671588, "standard_deviation": 0.20394615688174408, "minimum": 0.6399442245096948, "maximum": 1.2690913094339467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13388512500887534, "standard_deviation": 0.011939613446281413, "minimum": 0.12111248987307308, "maximum": 0.1569047073259681, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9735211770495567, "standard_deviation": 0.00631061463026859, "minimum": 0.9612127018697662, "maximum": 0.979337554756885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5415282123582612, "standard_deviation": 0.02787152619841595, "minimum": 1.5121493648557571, "maximum": 1.5861175811796742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9199762038239236, "standard_deviation": 0.523119112295335, "minimum": 0.22262271271946552, "maximum": 1.881818967016083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21072506419838435, "standard_deviation": 0.023230404276459487, "minimum": 0.18879457210284295, "maximum": 0.25446718222172027, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9886129944742141, "standard_deviation": 0.001972517403393144, "minimum": 0.9844775553109052, "maximum": 0.9910794445781799, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4690857280781557, "standard_deviation": 0.02629702620978771, "minimum": 1.442417182022107, "maximum": 1.5031078125547528, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7313852848863691, "standard_deviation": 0.4646895225736967, "minimum": 0.05899222419034367, "maximum": 1.5319168894545505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13815458341102316, "standard_deviation": 0.010720573400733409, "minimum": 0.12670522611035603, "maximum": 0.16022944611612353, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9879880841395441, "standard_deviation": 0.0017040131225845466, "minimum": 0.9849105391127466, "maximum": 0.9900421518317237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8181705203969534, "standard_deviation": 0.061978178135175734, "minimum": 1.7211129409427166, "maximum": 1.9050384451660791, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6603506327727195, "standard_deviation": 0.24580975282076126, "minimum": 0.21932183462008936, "maximum": 0.9879677753143827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13581262801511523, "standard_deviation": 0.009275236692526946, "minimum": 0.12248530388119426, "maximum": 0.15206778952368868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.995803719391402, "standard_deviation": 0.0012176716506135312, "minimum": 0.993346728030323, "maximum": 0.9971326031975941, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2039065178906463, "standard_deviation": 0.05133599254615325, "minimum": 1.1257241272222376, "maximum": 1.2776716701252073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.711074298295732, "standard_deviation": 0.24143689846371874, "minimum": 0.4212389346651609, "maximum": 1.0700147184722593, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08162966844609511, "standard_deviation": 0.010976623296394204, "minimum": 0.06548606827238032, "maximum": 0.0994850550953301, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25351083460198254, "standard_deviation": 0.06440416220912254, "minimum": 0.1877695001525398, "maximum": 0.37913773824770997, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.804072411084421, "standard_deviation": 0.009758382902697602, "minimum": 5.786439435637711, "maximum": 5.818375771133894, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.26537976479427783, "standard_deviation": 0.038506578885787714, "minimum": 0.24235998564702885, "maximum": 0.3586965593401172, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06216391538901844, "standard_deviation": 0.08035747034100163, "minimum": -0.1604200392539309, "maximum": 0.04807538439004945, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9903730365816478, "standard_deviation": 0.0011488567130287914, "minimum": 0.9881259522314774, "maximum": 0.9915801229361356, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5571533349345494, "standard_deviation": 0.0434464633235418, "minimum": 1.4839453894553873, "maximum": 1.6055683971077266, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40689697475513864, "standard_deviation": 0.14213357444967864, "minimum": 0.23388383948302752, "maximum": 0.6584429966486615, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1050049633275549, "standard_deviation": 0.006220958786309301, "minimum": 0.09869793506641528, "maximum": 0.11727932657976055, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972491155369098, "standard_deviation": 0.0007623326852947996, "minimum": 0.9956568986390125, "maximum": 0.9980690376318045, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8145499899604064, "standard_deviation": 0.05079012000228441, "minimum": 0.7622996332763219, "maximum": 0.9049288266248436, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3429103350187537, "standard_deviation": 0.1452502407876243, "minimum": 0.12308899301132543, "maximum": 0.5564608258956266, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055251027060815444, "standard_deviation": 0.006625494665535162, "minimum": 0.04836622234988286, "maximum": 0.06921612494826795, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0003", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.6685423851013184, "predicted_mos": 2.6685423851013184, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0c2c6d2d0a1d3d5ec9ca26952a695f84c7d12c64e4d671e2518a0c4b21f28c2f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6935716001316905, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.630888791676155, "全帯域_10ms_rms_envelope_nrmse": 0.10039751266480001, "全帯域_10ms_envelope_correlation": 0.9911815337705833, "全帯域_10ms_modulation_mae_db": 0.38190578075903336, "全帯域_40ms_log_rms_mae_db": 0.8507849221698053, "全帯域_40ms_rms_envelope_nrmse": 0.05050796948819988, "全帯域_40ms_envelope_correlation": 0.9977367920995489, "全帯域_40ms_modulation_mae_db": 0.13450744590428015, "300–1000 Hz_10ms_log_rms_mae_db": 1.6687058361924307, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13234733718662886, "300–1000 Hz_10ms_envelope_correlation": 0.9887610726753073, "300–1000 Hz_10ms_modulation_mae_db": 0.636278480578364, "300–1000 Hz_40ms_log_rms_mae_db": 1.001831647386233, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08573564165642578, "300–1000 Hz_40ms_envelope_correlation": 0.9955876314584611, "300–1000 Hz_40ms_modulation_mae_db": 0.7139189118433947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.602204728605218, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20231288511094228, "1000–3000 Hz_10ms_envelope_correlation": 0.970924631320873, "1000–3000 Hz_10ms_modulation_mae_db": 0.9576369296783177, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8708724024321426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1252657108522922, "1000–3000 Hz_40ms_envelope_correlation": 0.988589006533616, "1000–3000 Hz_40ms_modulation_mae_db": 0.993739035966127, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3553176516117902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25459398759074126, "3000–6000 Hz_10ms_envelope_correlation": 0.96125520299578, "3000–6000 Hz_10ms_modulation_mae_db": 1.4441336389125763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1988654650586015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15995871912516116, "3000–6000 Hz_40ms_envelope_correlation": 0.9844268124518933, "3000–6000 Hz_40ms_modulation_mae_db": 1.261249833119639, "mr_stft_db": 5.719773358609501, "band_mae_db": 0.3487202238604589, "rms_gain_error_db": -0.1004145830063758, "output_peak": 0.2447177851972809}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5182572909034566, "全帯域_10ms_rms_envelope_nrmse": 0.09921792186178678, "全帯域_10ms_envelope_correlation": 0.9913663628694961, "全帯域_10ms_modulation_mae_db": 0.24113073159846463, "全帯域_40ms_log_rms_mae_db": 0.7418869164970808, "全帯域_40ms_rms_envelope_nrmse": 0.04955159799146264, "全帯域_40ms_envelope_correlation": 0.9977637317537997, "全帯域_40ms_modulation_mae_db": 0.35880758488648085, "300–1000 Hz_10ms_log_rms_mae_db": 1.7511210217926783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13047406759570734, "300–1000 Hz_10ms_envelope_correlation": 0.9889248281575208, "300–1000 Hz_10ms_modulation_mae_db": 0.39356138316705264, "300–1000 Hz_40ms_log_rms_mae_db": 1.0371398105084446, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07958751987593296, "300–1000 Hz_40ms_envelope_correlation": 0.9960661990024214, "300–1000 Hz_40ms_modulation_mae_db": 0.3768818503331408, "1000–3000 Hz_10ms_log_rms_mae_db": 1.573431082541548, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21904181735932027, "1000–3000 Hz_10ms_envelope_correlation": 0.9660359968077534, "1000–3000 Hz_10ms_modulation_mae_db": 0.6756007198636854, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9247745176762, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14622021764818188, "1000–3000 Hz_40ms_envelope_correlation": 0.9841729727976019, "1000–3000 Hz_40ms_modulation_mae_db": 0.8227598098632197, "3000–6000 Hz_10ms_log_rms_mae_db": 1.37144937426332, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20754481641500905, "3000–6000 Hz_10ms_envelope_correlation": 0.9773890763518014, "3000–6000 Hz_10ms_modulation_mae_db": 1.1223956759165272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1766612335254423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14077143967368622, "3000–6000 Hz_40ms_envelope_correlation": 0.9900494717126962, "3000–6000 Hz_40ms_modulation_mae_db": 1.052464815681961, "mr_stft_db": 5.715283606044913, "band_mae_db": 0.2295530716307673, "rms_gain_error_db": -0.024354902933897898, "output_peak": 0.2559914252254575}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.628952995985804, "全帯域_10ms_rms_envelope_nrmse": 0.10875670627536793, "全帯域_10ms_envelope_correlation": 0.989612008885392, "全帯域_10ms_modulation_mae_db": 0.20767810773959347, "全帯域_40ms_log_rms_mae_db": 0.8692170742031263, "全帯域_40ms_rms_envelope_nrmse": 0.05929581049192207, "全帯域_40ms_envelope_correlation": 0.996746559987948, "全帯域_40ms_modulation_mae_db": 0.28628201450562435, "300–1000 Hz_10ms_log_rms_mae_db": 1.6966003878588178, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13655536962056763, "300–1000 Hz_10ms_envelope_correlation": 0.9876023271523212, "300–1000 Hz_10ms_modulation_mae_db": 0.3129937034557485, "300–1000 Hz_40ms_log_rms_mae_db": 0.9900860967970394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07006236774124237, "300–1000 Hz_40ms_envelope_correlation": 0.9967442916059062, "300–1000 Hz_40ms_modulation_mae_db": 0.559027842773909, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6015410174257154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2066054292178721, "1000–3000 Hz_10ms_envelope_correlation": 0.9696591329386364, "1000–3000 Hz_10ms_modulation_mae_db": 1.1285761413418618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9245177180342719, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1267193814854701, "1000–3000 Hz_40ms_envelope_correlation": 0.9883183096983559, "1000–3000 Hz_40ms_modulation_mae_db": 1.4297535923210292, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3755596236898844, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19648297134369158, "3000–6000 Hz_10ms_envelope_correlation": 0.9770553960789489, "3000–6000 Hz_10ms_modulation_mae_db": 0.5949687327680504, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1561503974123293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12855840203964977, "3000–6000 Hz_40ms_envelope_correlation": 0.9900881077182662, "3000–6000 Hz_40ms_modulation_mae_db": 0.5371590143403108, "mr_stft_db": 5.719409464401973, "band_mae_db": 0.1668162356736238, "rms_gain_error_db": -0.020692926167739073, "output_peak": 0.2525446222808656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.57711228054956, "全帯域_10ms_rms_envelope_nrmse": 0.09943946462154145, "全帯域_10ms_envelope_correlation": 0.9915297670336717, "全帯域_10ms_modulation_mae_db": 0.7462351363190879, "全帯域_40ms_log_rms_mae_db": 0.8941375558071483, "全帯域_40ms_rms_envelope_nrmse": 0.04906542189863939, "全帯域_40ms_envelope_correlation": 0.998100157512418, "全帯域_40ms_modulation_mae_db": 0.5722333658605662, "300–1000 Hz_10ms_log_rms_mae_db": 1.6920229149802444, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1326631589130445, "300–1000 Hz_10ms_envelope_correlation": 0.9887701677794832, "300–1000 Hz_10ms_modulation_mae_db": 1.0447521532885282, "300–1000 Hz_40ms_log_rms_mae_db": 0.9786109788820644, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07780906932806429, "300–1000 Hz_40ms_envelope_correlation": 0.9965847596392242, "300–1000 Hz_40ms_modulation_mae_db": 1.1282994486666535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.688008385653719, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21892706733127912, "1000–3000 Hz_10ms_envelope_correlation": 0.9659383767865789, "1000–3000 Hz_10ms_modulation_mae_db": 0.8039596678426101, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9556073882195787, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1509559983335827, "1000–3000 Hz_40ms_envelope_correlation": 0.9832515868658146, "1000–3000 Hz_40ms_modulation_mae_db": 1.0837133696106738, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3413452091774063, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22213524280926988, "3000–6000 Hz_10ms_envelope_correlation": 0.9707023767234025, "3000–6000 Hz_10ms_modulation_mae_db": 1.6026757320466312, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1967783725912025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14060908290168916, "3000–6000 Hz_40ms_envelope_correlation": 0.9885140174953811, "3000–6000 Hz_40ms_modulation_mae_db": 1.2488858558749263, "mr_stft_db": 5.723101586193729, "band_mae_db": 0.20611213339432796, "rms_gain_error_db": -0.172980493083605, "output_peak": 0.2520025403498914}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6242395587757055, "全帯域_10ms_rms_envelope_nrmse": 0.10000536379574304, "全帯域_10ms_envelope_correlation": 0.9913670782650481, "全帯域_10ms_modulation_mae_db": 0.31138984893291916, "全帯域_40ms_log_rms_mae_db": 0.9448779866929683, "全帯域_40ms_rms_envelope_nrmse": 0.051050138335118414, "全帯域_40ms_envelope_correlation": 0.9978458329294415, "全帯域_40ms_modulation_mae_db": 0.2303074510973717, "300–1000 Hz_10ms_log_rms_mae_db": 1.7766902304318977, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13108148877654882, "300–1000 Hz_10ms_envelope_correlation": 0.988972176833905, "300–1000 Hz_10ms_modulation_mae_db": 0.38178708776707876, "300–1000 Hz_40ms_log_rms_mae_db": 1.045335971426468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07712601851454037, "300–1000 Hz_40ms_envelope_correlation": 0.9965688995323045, "300–1000 Hz_40ms_modulation_mae_db": 0.5132935646271415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.570512896354153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20371930780932732, "1000–3000 Hz_10ms_envelope_correlation": 0.9709204803392235, "1000–3000 Hz_10ms_modulation_mae_db": 0.8876379140514613, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8676855335394258, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12128139517632044, "1000–3000 Hz_40ms_envelope_correlation": 0.9902556309231579, "1000–3000 Hz_40ms_modulation_mae_db": 0.8215592195123348, "3000–6000 Hz_10ms_log_rms_mae_db": 1.395743058461275, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1976338192392536, "3000–6000 Hz_10ms_envelope_correlation": 0.9769305017093751, "3000–6000 Hz_10ms_modulation_mae_db": 1.1078295547044386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2149231912242877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12654233995324568, "3000–6000 Hz_40ms_envelope_correlation": 0.9902579702960015, "3000–6000 Hz_40ms_modulation_mae_db": 0.9355296953941054, "mr_stft_db": 5.699725967160366, "band_mae_db": 0.21689019232744156, "rms_gain_error_db": -0.15710761621792974, "output_peak": 0.2501200024212413}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.652888369308605, "全帯域_10ms_rms_envelope_nrmse": 0.11657557733135573, "全帯域_10ms_envelope_correlation": 0.9881737962095833, "全帯域_10ms_modulation_mae_db": 0.569760062591832, "全帯域_40ms_log_rms_mae_db": 0.8251054127332297, "全帯域_40ms_rms_envelope_nrmse": 0.06692722316827579, "全帯域_40ms_envelope_correlation": 0.9958832082286098, "全帯域_40ms_modulation_mae_db": 0.6179814905242028, "300–1000 Hz_10ms_log_rms_mae_db": 1.6972861626128328, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1551188706622074, "300–1000 Hz_10ms_envelope_correlation": 0.9841370255391055, "300–1000 Hz_10ms_modulation_mae_db": 1.0916967674294165, "300–1000 Hz_40ms_log_rms_mae_db": 0.9610547838090048, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09751650836394213, "300–1000 Hz_40ms_envelope_correlation": 0.9935083085075783, "300–1000 Hz_40ms_modulation_mae_db": 1.1058790780456642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6823338096165787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2129928406192435, "1000–3000 Hz_10ms_envelope_correlation": 0.968154199520658, "1000–3000 Hz_10ms_modulation_mae_db": 0.5827481024838205, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9428132593793539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13491739383695409, "1000–3000 Hz_40ms_envelope_correlation": 0.9865976211551761, "1000–3000 Hz_40ms_modulation_mae_db": 0.6494214044512409, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4183336789037946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2303655804374794, "3000–6000 Hz_10ms_envelope_correlation": 0.968681251615468, "3000–6000 Hz_10ms_modulation_mae_db": 1.2453048777775664, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1950087170836712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13926827760286603, "3000–6000 Hz_40ms_envelope_correlation": 0.9882067859057214, "3000–6000 Hz_40ms_modulation_mae_db": 0.7712542980063145, "mr_stft_db": 5.716812567174614, "band_mae_db": 0.11668024136583839, "rms_gain_error_db": 0.011357824164567815, "output_peak": 0.34691207641327193}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6010099607700783, "全帯域_10ms_rms_envelope_nrmse": 0.10611605515398485, "全帯域_10ms_envelope_correlation": 0.9901658223830759, "全帯域_10ms_modulation_mae_db": 0.219725322777247, "全帯域_40ms_log_rms_mae_db": 0.8201260579833366, "全帯域_40ms_rms_envelope_nrmse": 0.05519622598287223, "全帯域_40ms_envelope_correlation": 0.9971881687855378, "全帯域_40ms_modulation_mae_db": 0.15089671628210777, "300–1000 Hz_10ms_log_rms_mae_db": 1.7514580341236194, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13244576419268814, "300–1000 Hz_10ms_envelope_correlation": 0.988379724980049, "300–1000 Hz_10ms_modulation_mae_db": 0.3556141289957595, "300–1000 Hz_40ms_log_rms_mae_db": 1.0195507713166778, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06757795935534089, "300–1000 Hz_40ms_envelope_correlation": 0.9969704008609416, "300–1000 Hz_40ms_modulation_mae_db": 0.365423651529561, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6200930082995457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20723683638354112, "1000–3000 Hz_10ms_envelope_correlation": 0.9698253427763242, "1000–3000 Hz_10ms_modulation_mae_db": 0.6266443420310774, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8927776453440796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1296978337423718, "1000–3000 Hz_40ms_envelope_correlation": 0.9885740584105333, "1000–3000 Hz_40ms_modulation_mae_db": 0.8649794790579325, "3000–6000 Hz_10ms_log_rms_mae_db": 1.333941127213539, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18735310239910408, "3000–6000 Hz_10ms_envelope_correlation": 0.9791663362430254, "3000–6000 Hz_10ms_modulation_mae_db": 0.38599110650655266, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1688427083444426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13186459629321864, "3000–6000 Hz_40ms_envelope_correlation": 0.9895175243302219, "3000–6000 Hz_40ms_modulation_mae_db": 0.1923433827668394, "mr_stft_db": 5.708132847799495, "band_mae_db": 0.2292101602295906, "rms_gain_error_db": 0.014065306161396833, "output_peak": 0.25465169185968045}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5211107586894879, "全帯域_10ms_rms_envelope_nrmse": 0.10506164549829954, "全帯域_10ms_envelope_correlation": 0.9903772890496895, "全帯域_10ms_modulation_mae_db": 0.4403846507402765, "全帯域_40ms_log_rms_mae_db": 0.7822009673249587, "全帯域_40ms_rms_envelope_nrmse": 0.05678173685995584, "全帯域_40ms_envelope_correlation": 0.9971756508397716, "全帯域_40ms_modulation_mae_db": 0.3607378768294191, "300–1000 Hz_10ms_log_rms_mae_db": 1.763544352750395, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13653131063070933, "300–1000 Hz_10ms_envelope_correlation": 0.9880983878829155, "300–1000 Hz_10ms_modulation_mae_db": 0.7686788704018593, "300–1000 Hz_40ms_log_rms_mae_db": 0.9971941802529055, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08662736543019979, "300–1000 Hz_40ms_envelope_correlation": 0.9956180717581823, "300–1000 Hz_40ms_modulation_mae_db": 0.8023114456115628, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6226582464960368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19999684991006322, "1000–3000 Hz_10ms_envelope_correlation": 0.9718122993116363, "1000–3000 Hz_10ms_modulation_mae_db": 1.0903951768353786, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9325627018079962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12550087739959395, "1000–3000 Hz_40ms_envelope_correlation": 0.9890321353379519, "1000–3000 Hz_40ms_modulation_mae_db": 1.2467011714728533, "3000–6000 Hz_10ms_log_rms_mae_db": 1.380789605245455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20981226077005188, "3000–6000 Hz_10ms_envelope_correlation": 0.9738312258622871, "3000–6000 Hz_10ms_modulation_mae_db": 0.7825138837409464, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1890690815752074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13766106273571285, "3000–6000 Hz_40ms_envelope_correlation": 0.9884947519371275, "3000–6000 Hz_40ms_modulation_mae_db": 0.7204509051081298, "mr_stft_db": 5.71173252165545, "band_mae_db": 0.27600180384398626, "rms_gain_error_db": -0.1297755685485664, "output_peak": 0.24694848665479513}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.926654871099551, "spectral_convergence": 0.32304844649314024, "spectral_convergence_status": "ok", "amplitude_floor": 8.79928282883349e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.534460746974376, "spectral_convergence": 0.28360031323448337, "spectral_convergence_status": "ok", "amplitude_floor": 4.444740709272264e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.698204457754576, "spectral_convergence": 0.30501046264008297, "spectral_convergence_status": "ok", "amplitude_floor": 3.930299196146418e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002477612487613689, "output_power": 0.00025736642230359626, "delta_db": 0.1651850410633597, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004515434578601184, "output_power": 0.00044174755935693353, "delta_db": -0.0952539570894731, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015370997116088465, "output_power": 0.0001367148628806251, "delta_db": -0.5088630986405006, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.4065683577917706e-05, "output_power": 4.084150470826632e-05, "delta_db": -0.3299877733363523, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.662327944441642e-06, "output_power": 9.946187993701314e-06, "delta_db": 0.6002204208091001, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.476431620775013e-05, "output_power": 2.2622741629019096e-05, "delta_db": -0.3928110522239676, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3487202238604589, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1004145830063758, "pitch": null, "output_rms_dbfs": -30.41321811146629, "output_peak": 0.2447177851972809, "reference_rms_dbfs": -30.312803528459913, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.719773358609501}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.467120737686064e-05, "log_rms_mae_db": 1.630888791676155, "rms_envelope_nrmse": 0.10039751266480001, "envelope_correlation": 0.9911815337705833, "rms_gain_error_db": -0.10042099902768024, "reference_mean_power": 0.0009312818437003785, "output_mean_power": 0.0009099950629333988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.494393820063344e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023666657525761142, "output_power": 0.00023264107839490766, "reference_active": true, "delta_db": -0.0745052392769352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5441929621447868e-05, "output_power": 1.531200415641546e-05, "reference_active": true, "delta_db": -0.036695304495483166}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.985747798516416e-06, "output_power": 5.505049427988082e-06, "reference_active": true, "delta_db": -1.0345167985046817}], "active_band_count": 3, "active_band_mae_db": 0.38190578075903336}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.334380857742655e-05, "log_rms_mae_db": 0.8507849221698053, "rms_envelope_nrmse": 0.05050796948819988, "envelope_correlation": 0.9977367920995489, "rms_gain_error_db": -0.10041700677279985, "reference_mean_power": 0.0009359486205868301, "output_mean_power": 0.0009145560096428051}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.189856446091556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021037225251172397, "output_power": 0.0002083305491217327, "reference_active": true, "delta_db": -0.04235498499447568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.644252267790675e-06, "output_power": 7.560423635696468e-06, "reference_active": true, "delta_db": -0.047888796134126416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4342900916585017e-07, "output_power": 3.195281472587919e-07, "reference_active": true, "delta_db": -0.31327855658423837}], "active_band_count": 3, "active_band_mae_db": 0.13450744590428015}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.589555428061273e-05, "log_rms_mae_db": 1.6687058361924307, "rms_envelope_nrmse": 0.13234733718662886, "envelope_correlation": 0.9887610726753073, "rms_gain_error_db": -0.2750207161956132, "reference_mean_power": 0.00038327408961782034, "output_mean_power": 0.00035975545601601565}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.028222657916465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015396875485141256, "output_power": 0.00014196597416617885, "reference_active": true, "delta_db": -0.35248330739282463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.75189127488202e-06, "output_power": 7.5805140532214616e-06, "reference_active": true, "delta_db": -0.6240325656103172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1315883357233815e-06, "output_power": 3.3333819988459562e-06, "reference_active": true, "delta_db": -0.9323195687319502}], "active_band_count": 3, "active_band_mae_db": 0.636278480578364}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.149058869949865e-05, "log_rms_mae_db": 1.001831647386233, "rms_envelope_nrmse": 0.08573564165642578, "envelope_correlation": 0.9955876314584611, "rms_gain_error_db": -0.27502375436019544, "reference_mean_power": 0.00038520090394369153, "output_mean_power": 0.00036156378337522354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9052955697898775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014058163319658512, "output_power": 0.00013099254458435577, "reference_active": true, "delta_db": -0.30682005821809366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.901478078483036e-06, "output_power": 3.2707615093571003e-06, "reference_active": true, "delta_db": -0.7658029276660991}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3393802456709067e-07, "output_power": 1.8288894021936736e-07, "reference_active": true, "delta_db": -1.0691337496459914}], "active_band_count": 3, "active_band_mae_db": 0.7139189118433947}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.043858753489399e-05, "log_rms_mae_db": 1.602204728605218, "rms_envelope_nrmse": 0.20231288511094228, "envelope_correlation": 0.970924631320873, "rms_gain_error_db": -0.21179346426463025, "reference_mean_power": 4.084610004445957e-05, "output_mean_power": 3.890193915946494e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0558567323423266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3562756220605441e-05, "output_power": 1.2527492740359883e-05, "reference_active": true, "delta_db": -0.3448379599539675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5216123885214767e-06, "output_power": 1.0752515103699245e-06, "reference_active": true, "delta_db": -1.5079397424041203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602773162915538e-07, "output_power": 5.220511643864008e-07, "reference_active": true, "delta_db": -1.0201330866768654}], "active_band_count": 3, "active_band_mae_db": 0.9576369296783177}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.7804457921821952e-05, "log_rms_mae_db": 0.8708724024321426, "rms_envelope_nrmse": 0.1252657108522922, "envelope_correlation": 0.988589006533616, "rms_gain_error_db": -0.21180942437240174, "reference_mean_power": 4.105094002568101e-05, "output_mean_power": 3.909688564727078e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8992373996258767e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2457002317192863e-05, "output_power": 1.1593228391331889e-05, "reference_active": true, "delta_db": -0.31209153506365833}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.119435793981677e-07, "output_power": 5.678991260855082e-07, "reference_active": true, "delta_db": -1.5525465152452993}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.742155092447501e-08, "output_power": 2.89376911394783e-08, "reference_active": true, "delta_db": -1.116579057589423}], "active_band_count": 3, "active_band_mae_db": 0.993739035966127}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.7924378350629476e-05, "log_rms_mae_db": 1.3553176516117902, "rms_envelope_nrmse": 0.25459398759074126, "envelope_correlation": 0.96125520299578, "rms_gain_error_db": -0.19138724926597384, "reference_mean_power": 2.620464364014996e-05, "output_mean_power": 2.5074918673707067e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.347390510898905e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.954495563916905e-05, "output_power": 1.925909729008098e-05, "reference_active": true, "delta_db": -0.06398761988341857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.79596482520383e-06, "output_power": 3.1029910092736633e-06, "reference_active": true, "delta_db": 0.4524881380301292}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.14743402003256e-06, "output_power": 4.765793341501639e-07, "reference_active": true, "delta_db": -3.815925158824181}], "active_band_count": 3, "active_band_mae_db": 1.4441336389125763}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.7050501746546627e-05, "log_rms_mae_db": 1.1988654650586015, "rms_envelope_nrmse": 0.15995871912516116, "envelope_correlation": 0.9844268124518933, "rms_gain_error_db": -0.19138778471746168, "reference_mean_power": 2.633641752064706e-05, "output_mean_power": 2.5201008460005386e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2809655135480154e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9754827323518705e-05, "output_power": 1.938807837904737e-05, "reference_active": true, "delta_db": -0.08138471125096322}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4198245590923022e-06, "output_power": 1.5330797004977685e-06, "reference_active": true, "delta_db": 0.3333004917566785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.51727649987637e-08, "output_power": 3.000265699711659e-08, "reference_active": true, "delta_db": -3.369064296351275}], "active_band_count": 3, "active_band_mae_db": 1.261249833119639}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7267720437535482, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "51df9cfe2d914477d0b15fa5a5e74788a26c7d77b24472ca93bdb9397cb3cf42", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9691588074752104, "standard_deviation": 0.0022389281990683944, "minimum": 0.9659383767865789, "maximum": 0.9718122993116363, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6200978968740642, "standard_deviation": 0.04438900413605235, "minimum": 1.570512896354153, "maximum": 1.688008385653719, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8441498742660266, "standard_deviation": 0.20769798980266796, "minimum": 0.5827481024838205, "maximum": 1.1285761413418618, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20885412921769864, "standard_deviation": 0.00734516316915312, "minimum": 0.19999684991006322, "maximum": 0.21904181735932027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9873489152152759, "standard_deviation": 0.0024698151112636834, "minimum": 0.9832515868658146, "maximum": 0.9902556309231579, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.913951395804131, "standard_deviation": 0.032924272649377806, "minimum": 0.8676855335394258, "maximum": 0.9556073882195787, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9890783852819264, "standard_deviation": 0.2552403860145207, "minimum": 0.6494214044512409, "maximum": 1.4297535923210292, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13256985105934588, "standard_deviation": 0.010711278724421135, "minimum": 0.12128139517632044, "maximum": 0.1509559983335827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.973126420947511, "standard_deviation": 0.005995436006012259, "minimum": 0.96125520299578, "maximum": 0.9791663362430254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.371559916070808, "standard_deviation": 0.027943103546626945, "minimum": 1.333941127213539, "maximum": 1.4183336789037946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0357266502966611, "standard_deviation": 0.41839855158561967, "minimum": 0.38599110650655266, "maximum": 1.6026757320466312, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21324022262557507, "standard_deviation": 0.021782135557886842, "minimum": 0.18735310239910408, "maximum": 0.25459398759074126, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9886944302309136, "standard_deviation": 0.0019057579841630936, "minimum": 0.9844268124518933, "maximum": 0.9902579702960015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.187037395851898, "standard_deviation": 0.018795117953980495, "minimum": 1.1561503974123293, "maximum": 1.2149231912242877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8399172250365283, "standard_deviation": 0.36431853113577506, "minimum": 0.1923433827668394, "maximum": 1.261249833119639, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1381542400406537, "standard_deviation": 0.010394171087224667, "minimum": 0.12654233995324568, "maximum": 0.15995871912516116, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.987955713875076, "standard_deviation": 0.0016120847846975697, "minimum": 0.9841370255391055, "maximum": 0.988972176833905, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7246786175928646, "standard_deviation": 0.04030273275864335, "minimum": 1.6687058361924307, "maximum": 1.7766902304318977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6231703218854759, "standard_deviation": 0.31563423987852296, "minimum": 0.3129937034557485, "maximum": 1.0916967674294165, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13590217094726276, "standard_deviation": 0.008087751810566715, "minimum": 0.13047406759570734, "maximum": 0.1551188706622074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959560702956275, "standard_deviation": 0.0011132080786709255, "minimum": 0.9935083085075783, "maximum": 0.9969704008609416, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0038505300473548, "standard_deviation": 0.02874038710734994, "minimum": 0.9610547838090048, "maximum": 1.045335971426468, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6956294741788784, "standard_deviation": 0.29994425993801144, "minimum": 0.365423651529561, "maximum": 1.1282994486666535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08025530628321108, "standard_deviation": 0.009631940080222817, "minimum": 0.06757795935534089, "maximum": 0.09751650836394213, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22374800779075432, "standard_deviation": 0.0691528688063833, "minimum": 0.11668024136583839, "maximum": 0.3487202238604589, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.714246489880004, "standard_deviation": 0.007538642479049731, "minimum": 5.699725967160366, "maximum": 5.723101586193729, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.26298607880031055, "standard_deviation": 0.034117639394899694, "minimum": 0.2447177851972809, "maximum": 0.34691207641327193, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07248786995401865, "standard_deviation": 0.07638446168399865, "minimum": -0.172980493083605, "maximum": 0.014065306161396833, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9904717073083175, "standard_deviation": 0.0011555702923153395, "minimum": 0.9881737962095833, "maximum": 0.9915297670336717, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5943075008323566, "standard_deviation": 0.05119729830864634, "minimum": 1.5182572909034566, "maximum": 1.652888369308605, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38977620518230677, "standard_deviation": 0.18988614416448957, "minimum": 0.20767810773959347, "maximum": 0.7462351363190879, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10444628090035993, "standard_deviation": 0.006063763976187628, "minimum": 0.09921792186178678, "maximum": 0.11657557733135573, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973050127671343, "standard_deviation": 0.0007255916591871941, "minimum": 0.9958832082286098, "maximum": 0.998100157512418, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8410421116764568, "standard_deviation": 0.06378004465406652, "minimum": 0.7418869164970808, "maximum": 0.9448779866929683, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33896924323625666, "standard_deviation": 0.17923912573255427, "minimum": 0.13450744590428015, "maximum": 0.6179814905242028, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054797015527055785, "standard_deviation": 0.006144453440429224, "minimum": 0.04906542189863939, "maximum": 0.06692722316827579, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0003", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.383514881134033, "predicted_mos": 2.383514881134033, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f792fd9349ae96cb18d0fb90bb539a584c114d455fdc2e0293abc2646b92ea3f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9544604998081923, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4327427464533393, "全帯域_10ms_rms_envelope_nrmse": 0.10537311857210577, "全帯域_10ms_envelope_correlation": 0.990259760908149, "全帯域_10ms_modulation_mae_db": 0.8036182823977401, "全帯域_40ms_log_rms_mae_db": 0.7673983959440569, "全帯域_40ms_rms_envelope_nrmse": 0.0494563583126405, "全帯域_40ms_envelope_correlation": 0.9977410552413642, "全帯域_40ms_modulation_mae_db": 0.6481256753603617, "300–1000 Hz_10ms_log_rms_mae_db": 2.147461071611267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14615430008150024, "300–1000 Hz_10ms_envelope_correlation": 0.9858134071384063, "300–1000 Hz_10ms_modulation_mae_db": 0.7119571549908, "300–1000 Hz_40ms_log_rms_mae_db": 1.4787023985002892, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06839719265142048, "300–1000 Hz_40ms_envelope_correlation": 0.9969980033453367, "300–1000 Hz_40ms_modulation_mae_db": 0.38243179951909906, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0922911184575033, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22472896272034676, "1000–3000 Hz_10ms_envelope_correlation": 0.9641030976435796, "1000–3000 Hz_10ms_modulation_mae_db": 1.0661239145060597, "1000–3000 Hz_40ms_log_rms_mae_db": 1.435951199785803, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1423435802468083, "1000–3000 Hz_40ms_envelope_correlation": 0.9853389306205738, "1000–3000 Hz_40ms_modulation_mae_db": 1.0972745196316198, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0957790738422313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21387584773013565, "3000–6000 Hz_10ms_envelope_correlation": 0.9738529039301652, "3000–6000 Hz_10ms_modulation_mae_db": 1.9938384253599883, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9423786743383806, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14893495786082434, "3000–6000 Hz_40ms_envelope_correlation": 0.9884267146459624, "3000–6000 Hz_40ms_modulation_mae_db": 1.8440561363724235, "mr_stft_db": 6.134544729120322, "band_mae_db": 0.2905742979186096, "rms_gain_error_db": -0.03157299604610943, "output_peak": 0.26136001057687763}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4715325622115074, "全帯域_10ms_rms_envelope_nrmse": 0.11222963892000558, "全帯域_10ms_envelope_correlation": 0.9889468561573461, "全帯域_10ms_modulation_mae_db": 0.2650261507364629, "全帯域_40ms_log_rms_mae_db": 0.7681171310098336, "全帯域_40ms_rms_envelope_nrmse": 0.05606112774340939, "全帯域_40ms_envelope_correlation": 0.9971139655970788, "全帯域_40ms_modulation_mae_db": 0.3132500065994361, "300–1000 Hz_10ms_log_rms_mae_db": 2.1133682675678154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.136360355393428, "300–1000 Hz_10ms_envelope_correlation": 0.9877072232260096, "300–1000 Hz_10ms_modulation_mae_db": 0.38931200977255626, "300–1000 Hz_40ms_log_rms_mae_db": 1.4310849989164052, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07057498696529793, "300–1000 Hz_40ms_envelope_correlation": 0.9968226480428308, "300–1000 Hz_40ms_modulation_mae_db": 0.49152063929580364, "1000–3000 Hz_10ms_log_rms_mae_db": 2.157489288690897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2220181611148012, "1000–3000 Hz_10ms_envelope_correlation": 0.9655539429805668, "1000–3000 Hz_10ms_modulation_mae_db": 0.4405152721534451, "1000–3000 Hz_40ms_log_rms_mae_db": 1.47784913247611, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14738272747760886, "1000–3000 Hz_40ms_envelope_correlation": 0.9841706992412504, "1000–3000 Hz_40ms_modulation_mae_db": 0.5459279681936838, "3000–6000 Hz_10ms_log_rms_mae_db": 2.133696016174026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2245275634400537, "3000–6000 Hz_10ms_envelope_correlation": 0.9704270773677035, "3000–6000 Hz_10ms_modulation_mae_db": 1.3518960039388208, "3000–6000 Hz_40ms_log_rms_mae_db": 2.025961202293412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.156736218912799, "3000–6000 Hz_40ms_envelope_correlation": 0.9861541071091169, "3000–6000 Hz_40ms_modulation_mae_db": 1.2314718588997695, "mr_stft_db": 6.129813432358154, "band_mae_db": 0.2616283712454714, "rms_gain_error_db": -0.058684131228298034, "output_peak": 0.260710920730748}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5205024303563601, "全帯域_10ms_rms_envelope_nrmse": 0.10818932286319366, "全帯域_10ms_envelope_correlation": 0.9897107330664773, "全帯域_10ms_modulation_mae_db": 0.7033749797890015, "全帯域_40ms_log_rms_mae_db": 0.7522037021312262, "全帯域_40ms_rms_envelope_nrmse": 0.048404893704173585, "全帯域_40ms_envelope_correlation": 0.9978754139706187, "全帯域_40ms_modulation_mae_db": 0.4660091384408296, "300–1000 Hz_10ms_log_rms_mae_db": 2.215062667826945, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13754269300652186, "300–1000 Hz_10ms_envelope_correlation": 0.9876020537532794, "300–1000 Hz_10ms_modulation_mae_db": 0.919792595197987, "300–1000 Hz_40ms_log_rms_mae_db": 1.5578225432299253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07310982830064625, "300–1000 Hz_40ms_envelope_correlation": 0.9967227411447566, "300–1000 Hz_40ms_modulation_mae_db": 1.0402895649940511, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1367303071035377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2124281346803228, "1000–3000 Hz_10ms_envelope_correlation": 0.968012535746429, "1000–3000 Hz_10ms_modulation_mae_db": 0.7527957851370702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.41156855684637, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1428766581321708, "1000–3000 Hz_40ms_envelope_correlation": 0.9852317638877816, "1000–3000 Hz_40ms_modulation_mae_db": 1.3133168120876013, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0824251878231697, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21150915363598, "3000–6000 Hz_10ms_envelope_correlation": 0.9744441118317456, "3000–6000 Hz_10ms_modulation_mae_db": 1.376764658883082, "3000–6000 Hz_40ms_log_rms_mae_db": 1.993187343741036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1357592076355603, "3000–6000 Hz_40ms_envelope_correlation": 0.98935747728922, "3000–6000 Hz_40ms_modulation_mae_db": 1.360821553077728, "mr_stft_db": 6.135558274637845, "band_mae_db": 0.26267555566275563, "rms_gain_error_db": -0.0608829793535386, "output_peak": 0.2607906336269766}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5251781648424882, "全帯域_10ms_rms_envelope_nrmse": 0.10232306644538723, "全帯域_10ms_envelope_correlation": 0.9908456575957526, "全帯域_10ms_modulation_mae_db": 0.2473970391665968, "全帯域_40ms_log_rms_mae_db": 0.7726282378791387, "全帯域_40ms_rms_envelope_nrmse": 0.050228364791808304, "全帯域_40ms_envelope_correlation": 0.9977986448945334, "全帯域_40ms_modulation_mae_db": 0.3857356778952825, "300–1000 Hz_10ms_log_rms_mae_db": 2.2179333676282975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14888501862462702, "300–1000 Hz_10ms_envelope_correlation": 0.9854860466451801, "300–1000 Hz_10ms_modulation_mae_db": 0.3749459624632654, "300–1000 Hz_40ms_log_rms_mae_db": 1.579402162578561, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08132131267282558, "300–1000 Hz_40ms_envelope_correlation": 0.9960128009163164, "300–1000 Hz_40ms_modulation_mae_db": 0.45045012318097316, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1684688706557482, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22151524689405974, "1000–3000 Hz_10ms_envelope_correlation": 0.9650951653927862, "1000–3000 Hz_10ms_modulation_mae_db": 0.9933105556123412, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5050224756793422, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1394264727156534, "1000–3000 Hz_40ms_envelope_correlation": 0.9857555167419623, "1000–3000 Hz_40ms_modulation_mae_db": 1.4440396454113635, "3000–6000 Hz_10ms_log_rms_mae_db": 2.110627692811531, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21723746157423043, "3000–6000 Hz_10ms_envelope_correlation": 0.9720675843278116, "3000–6000 Hz_10ms_modulation_mae_db": 1.560020765552095, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0025857990805314, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15232092728009705, "3000–6000 Hz_40ms_envelope_correlation": 0.9860539300574063, "3000–6000 Hz_40ms_modulation_mae_db": 1.3785847246496539, "mr_stft_db": 6.134668280091105, "band_mae_db": 0.35045355152294944, "rms_gain_error_db": -0.10212437558605345, "output_peak": 0.25839200837757015}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4956477225448568, "全帯域_10ms_rms_envelope_nrmse": 0.10071662962003014, "全帯域_10ms_envelope_correlation": 0.991126911298606, "全帯域_10ms_modulation_mae_db": 0.470553858018442, "全帯域_40ms_log_rms_mae_db": 0.8142081650335828, "全帯域_40ms_rms_envelope_nrmse": 0.043999020639457785, "全帯域_40ms_envelope_correlation": 0.9983163368806923, "全帯域_40ms_modulation_mae_db": 0.5964728650899735, "300–1000 Hz_10ms_log_rms_mae_db": 2.098534125005955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13047596638314968, "300–1000 Hz_10ms_envelope_correlation": 0.9890892086646846, "300–1000 Hz_10ms_modulation_mae_db": 0.5160636381895238, "300–1000 Hz_40ms_log_rms_mae_db": 1.4410185056432556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07005481701952643, "300–1000 Hz_40ms_envelope_correlation": 0.9972837286383827, "300–1000 Hz_40ms_modulation_mae_db": 0.5589550755407336, "1000–3000 Hz_10ms_log_rms_mae_db": 2.142483073378598, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22057614225966965, "1000–3000 Hz_10ms_envelope_correlation": 0.9654806468555073, "1000–3000 Hz_10ms_modulation_mae_db": 1.072504458835567, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4775174561783282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14739621601705194, "1000–3000 Hz_40ms_envelope_correlation": 0.9842641208262713, "1000–3000 Hz_40ms_modulation_mae_db": 1.4186025802607645, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1328049676241676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23929474322135716, "3000–6000 Hz_10ms_envelope_correlation": 0.9662566488894524, "3000–6000 Hz_10ms_modulation_mae_db": 1.5409556688079586, "3000–6000 Hz_40ms_log_rms_mae_db": 2.00073797637654, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15325630161895776, "3000–6000 Hz_40ms_envelope_correlation": 0.9857959498855576, "3000–6000 Hz_40ms_modulation_mae_db": 2.0528352524849747, "mr_stft_db": 6.1290330241753965, "band_mae_db": 0.25801911717013964, "rms_gain_error_db": -0.10165058771050817, "output_peak": 0.2503053529088782}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4428443065985377, "全帯域_10ms_rms_envelope_nrmse": 0.09909894067117206, "全帯域_10ms_envelope_correlation": 0.9914268929938524, "全帯域_10ms_modulation_mae_db": 0.2914020128587909, "全帯域_40ms_log_rms_mae_db": 0.7630894065661838, "全帯域_40ms_rms_envelope_nrmse": 0.050951393121767934, "全帯域_40ms_envelope_correlation": 0.9976059766477269, "全帯域_40ms_modulation_mae_db": 0.4382160864911009, "300–1000 Hz_10ms_log_rms_mae_db": 2.186396838827962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12792132075920662, "300–1000 Hz_10ms_envelope_correlation": 0.9891509757904406, "300–1000 Hz_10ms_modulation_mae_db": 0.5935204287873778, "300–1000 Hz_40ms_log_rms_mae_db": 1.4991061870423352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07716569662882487, "300–1000 Hz_40ms_envelope_correlation": 0.9959781748955892, "300–1000 Hz_40ms_modulation_mae_db": 0.5767163796674499, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1663425740572353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22007618252490468, "1000–3000 Hz_10ms_envelope_correlation": 0.9657187211732293, "1000–3000 Hz_10ms_modulation_mae_db": 0.4732360308807937, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4822582665856174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14611957299902473, "1000–3000 Hz_40ms_envelope_correlation": 0.9843917688967329, "1000–3000 Hz_40ms_modulation_mae_db": 0.8492084084821879, "3000–6000 Hz_10ms_log_rms_mae_db": 2.126315219965843, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20665968595727255, "3000–6000 Hz_10ms_envelope_correlation": 0.9746272024768693, "3000–6000 Hz_10ms_modulation_mae_db": 0.7315695296188723, "3000–6000 Hz_40ms_log_rms_mae_db": 1.995794514096459, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12586649425008228, "3000–6000 Hz_40ms_envelope_correlation": 0.9905830336805216, "3000–6000 Hz_40ms_modulation_mae_db": 0.4908201812866513, "mr_stft_db": 6.127874579569955, "band_mae_db": 0.22456860696024336, "rms_gain_error_db": 0.018923639951496103, "output_peak": 0.26733087585901155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5110948419732693, "全帯域_10ms_rms_envelope_nrmse": 0.10910943063388077, "全帯域_10ms_envelope_correlation": 0.9896307132845151, "全帯域_10ms_modulation_mae_db": 0.841251152741156, "全帯域_40ms_log_rms_mae_db": 0.7946165287341426, "全帯域_40ms_rms_envelope_nrmse": 0.05241579137322022, "全帯域_40ms_envelope_correlation": 0.9974757744445775, "全帯域_40ms_modulation_mae_db": 1.0546078384776056, "300–1000 Hz_10ms_log_rms_mae_db": 2.0955591876892528, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1422824189212504, "300–1000 Hz_10ms_envelope_correlation": 0.9866799724776347, "300–1000 Hz_10ms_modulation_mae_db": 0.8337127939137355, "300–1000 Hz_40ms_log_rms_mae_db": 1.4617152926323902, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07091530435774913, "300–1000 Hz_40ms_envelope_correlation": 0.9966209752433602, "300–1000 Hz_40ms_modulation_mae_db": 0.8003972564480223, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1253919625411317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22046391947176772, "1000–3000 Hz_10ms_envelope_correlation": 0.9656250071848199, "1000–3000 Hz_10ms_modulation_mae_db": 0.7444880588883201, "1000–3000 Hz_40ms_log_rms_mae_db": 1.405253024209205, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13783779481586902, "1000–3000 Hz_40ms_envelope_correlation": 0.9867779740490934, "1000–3000 Hz_40ms_modulation_mae_db": 0.9941327623401787, "3000–6000 Hz_10ms_log_rms_mae_db": 2.200176365284599, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27206356232691564, "3000–6000 Hz_10ms_envelope_correlation": 0.957277048008784, "3000–6000 Hz_10ms_modulation_mae_db": 2.160239824829429, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0900789593392792, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18723711257710865, "3000–6000 Hz_40ms_envelope_correlation": 0.979224225828304, "3000–6000 Hz_40ms_modulation_mae_db": 2.912942666079244, "mr_stft_db": 6.145600355030737, "band_mae_db": 0.21020339476085867, "rms_gain_error_db": 0.029453847604196042, "output_peak": 0.24421710743324773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4345372246978976, "全帯域_10ms_rms_envelope_nrmse": 0.09778418580452784, "全帯域_10ms_envelope_correlation": 0.9916431710828876, "全帯域_10ms_modulation_mae_db": 0.16429564655802675, "全帯域_40ms_log_rms_mae_db": 0.7606383392751564, "全帯域_40ms_rms_envelope_nrmse": 0.04808079493214412, "全帯域_40ms_envelope_correlation": 0.9979576654117783, "全帯域_40ms_modulation_mae_db": 0.37346332764469553, "300–1000 Hz_10ms_log_rms_mae_db": 2.218733727621946, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12375280271342565, "300–1000 Hz_10ms_envelope_correlation": 0.990160465765298, "300–1000 Hz_10ms_modulation_mae_db": 0.5565854931897203, "300–1000 Hz_40ms_log_rms_mae_db": 1.536813773832802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06058643903204477, "300–1000 Hz_40ms_envelope_correlation": 0.9980563220541039, "300–1000 Hz_40ms_modulation_mae_db": 0.34232177305457673, "1000–3000 Hz_10ms_log_rms_mae_db": 2.079527893563246, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21004474041855098, "1000–3000 Hz_10ms_envelope_correlation": 0.9687108767929251, "1000–3000 Hz_10ms_modulation_mae_db": 0.7277498618818189, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3802781035246585, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13847793573332612, "1000–3000 Hz_40ms_envelope_correlation": 0.9859164564223311, "1000–3000 Hz_40ms_modulation_mae_db": 0.9329606671747922, "3000–6000 Hz_10ms_log_rms_mae_db": 2.077999880169983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18454077772540303, "3000–6000 Hz_10ms_envelope_correlation": 0.979980608961101, "3000–6000 Hz_10ms_modulation_mae_db": 0.7862545792657265, "3000–6000 Hz_40ms_log_rms_mae_db": 2.006951626755953, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1298708357356186, "3000–6000 Hz_40ms_envelope_correlation": 0.9899205705348013, "3000–6000 Hz_40ms_modulation_mae_db": 0.9200526025822509, "mr_stft_db": 6.106857241911047, "band_mae_db": 0.17822093389854166, "rms_gain_error_db": -0.09980683733454813, "output_peak": 0.25584529406579015}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.373364740754384, "spectral_convergence": 0.3308160643988214, "spectral_convergence_status": "ok", "amplitude_floor": 8.79928282883349e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.029833585940234, "spectral_convergence": 0.29159278332927685, "spectral_convergence_status": "ok", "amplitude_floor": 4.444740709272264e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.000435860666349, "spectral_convergence": 0.31220533198542816, "spectral_convergence_status": "ok", "amplitude_floor": 3.930299196146418e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002477612487613689, "output_power": 0.0002622596239674711, "delta_db": 0.24698052897819045, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004515434578601184, "output_power": 0.0004436060837863577, "delta_db": -0.07702060827965182, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015370997116088465, "output_power": 0.00014469149136978916, "delta_db": -0.26259048040980176, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.4065683577917706e-05, "output_power": 4.239286872241925e-05, "delta_db": -0.16807705352843041, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.662327944441642e-06, "output_power": 9.404682524965954e-06, "delta_db": 0.35709518073272245, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.476431620775013e-05, "output_power": 2.1412044987806677e-05, "delta_db": -0.6316819355828608, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2905742979186096, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03157299604610943, "pitch": null, "output_rms_dbfs": -30.344376524506025, "output_peak": 0.26136001057687763, "reference_rms_dbfs": -30.312803528459913, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.134544729120322}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.467120737686064e-05, "log_rms_mae_db": 1.4327427464533393, "rms_envelope_nrmse": 0.10537311857210577, "envelope_correlation": 0.990259760908149, "rms_gain_error_db": -0.03157450125736898, "reference_mean_power": 0.0009312818437003785, "output_mean_power": 0.0009245357005047038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.494393820063344e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023666657525761142, "output_power": 0.00023797128921952023, "reference_active": true, "delta_db": 0.023876371597440674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5441929621447868e-05, "output_power": 1.3785974812889797e-05, "reference_active": true, "delta_db": -0.4926408807212765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.985747798516416e-06, "output_power": 4.516262151268374e-06, "reference_active": true, "delta_db": -1.8943375948745034}], "active_band_count": 3, "active_band_mae_db": 0.8036182823977401}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.334380857742655e-05, "log_rms_mae_db": 0.7673983959440569, "rms_envelope_nrmse": 0.0494563583126405, "envelope_correlation": 0.9977410552413642, "rms_gain_error_db": -0.03154784880975228, "reference_mean_power": 0.0009359486205868301, "output_mean_power": 0.0009291743738543172}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.189856446091556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021037225251172397, "output_power": 0.00021395064088707932, "reference_active": true, "delta_db": 0.07325134666877069}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.644252267790675e-06, "output_power": 6.4033643515795835e-06, "reference_active": true, "delta_db": -0.7692679686823927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4342900916585017e-07, "output_power": 2.6647176946332855e-07, "reference_active": true, "delta_db": -1.101857710729922}], "active_band_count": 3, "active_band_mae_db": 0.6481256753603617}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.589555428061273e-05, "log_rms_mae_db": 2.147461071611267, "rms_envelope_nrmse": 0.14615430008150024, "envelope_correlation": 0.9858134071384063, "rms_gain_error_db": -0.15056992498380717, "reference_mean_power": 0.00038327408961782034, "output_mean_power": 0.0003702136848109575}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.028222657916465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015396875485141256, "output_power": 0.00014800618121103476, "reference_active": true, "delta_db": -0.17152744340294732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.75189127488202e-06, "output_power": 7.956599691656447e-06, "reference_active": true, "delta_db": -0.4137440515348844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1315883357233815e-06, "output_power": 2.8910596052459822e-06, "reference_active": true, "delta_db": -1.5505999700345685}], "active_band_count": 3, "active_band_mae_db": 0.7119571549908}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.149058869949865e-05, "log_rms_mae_db": 1.4787023985002892, "rms_envelope_nrmse": 0.06839719265142048, "envelope_correlation": 0.9969980033453367, "rms_gain_error_db": -0.15057450784901177, "reference_mean_power": 0.00038520090394369153, "output_mean_power": 0.0003720744485989599}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9052955697898775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014058163319658512, "output_power": 0.00013703309111030374, "reference_active": true, "delta_db": -0.11103130001639569}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.901478078483036e-06, "output_power": 3.4525036891566987e-06, "reference_active": true, "delta_db": -0.5309501974234572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3393802456709067e-07, "output_power": 2.0824252840605476e-07, "reference_active": true, "delta_db": -0.5053139011174443}], "active_band_count": 3, "active_band_mae_db": 0.38243179951909906}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.043858753489399e-05, "log_rms_mae_db": 2.0922911184575033, "rms_envelope_nrmse": 0.22472896272034676, "envelope_correlation": 0.9641030976435796, "rms_gain_error_db": -0.1822228037141775, "reference_mean_power": 4.084610004445957e-05, "output_mean_power": 3.9167722243742324e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0558567323423266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3562756220605441e-05, "output_power": 1.099112067782646e-05, "reference_active": true, "delta_db": -0.9130597947603378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5216123885214767e-06, "output_power": 1.0640075991689183e-06, "reference_active": true, "delta_db": -1.5535930579376864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602773162915538e-07, "output_power": 7.81444729691379e-07, "reference_active": true, "delta_db": 0.7317188908201548}], "active_band_count": 3, "active_band_mae_db": 1.0661239145060597}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.7804457921821952e-05, "log_rms_mae_db": 1.435951199785803, "rms_envelope_nrmse": 0.1423435802468083, "envelope_correlation": 0.9853389306205738, "rms_gain_error_db": -0.18225686947631728, "reference_mean_power": 4.105094002568101e-05, "output_mean_power": 3.9363836523761064e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8992373996258767e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2457002317192863e-05, "output_power": 1.0142857475727464e-05, "reference_active": true, "delta_db": -0.8925322227628494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.119435793981677e-07, "output_power": 4.955319060807671e-07, "reference_active": true, "delta_db": -2.1445422901700755}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.742155092447501e-08, "output_power": 3.5289611569915735e-08, "reference_active": true, "delta_db": -0.25474904596193415}], "active_band_count": 3, "active_band_mae_db": 1.0972745196316198}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.7924378350629476e-05, "log_rms_mae_db": 2.0957790738422313, "rms_envelope_nrmse": 0.21387584773013565, "envelope_correlation": 0.9738529039301652, "rms_gain_error_db": -0.530771083374363, "reference_mean_power": 2.620464364014996e-05, "output_mean_power": 2.3190021416703826e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.347390510898905e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.954495563916905e-05, "output_power": 1.6083758864109465e-05, "reference_active": true, "delta_db": -0.8464713574308768}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.79596482520383e-06, "output_power": 1.8730033581436373e-06, "reference_active": true, "delta_db": -1.7399314741048268}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.14743402003256e-06, "output_power": 5.2506911997176e-07, "reference_active": true, "delta_db": -3.3951124445442615}], "active_band_count": 3, "active_band_mae_db": 1.9938384253599883}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.7050501746546627e-05, "log_rms_mae_db": 1.9423786743383806, "rms_envelope_nrmse": 0.14893495786082434, "envelope_correlation": 0.9884267146459624, "rms_gain_error_db": -0.5307731384991378, "reference_mean_power": 2.633641752064706e-05, "output_mean_power": 2.3306624799816966e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2809655135480154e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9754827323518705e-05, "output_power": 1.5932737764335744e-05, "reference_active": true, "delta_db": -0.9338282969692702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4198245590923022e-06, "output_power": 9.366433398280731e-07, "reference_active": true, "delta_db": -1.80660434674774}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.51727649987637e-08, "output_power": 3.426830077573392e-08, "reference_active": true, "delta_db": -2.7917357654002597}], "active_band_count": 3, "active_band_mae_db": 1.8440561363724235}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6736699278088337, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1bbdeccfea14846bf4581c47a2cf64d5c584e9b70a92efa795d7312c79cd7cd9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9660374992212304, "standard_deviation": 0.0015349871913678086, "minimum": 0.9641030976435796, "maximum": 0.9687108767929251, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1335906360559873, "standard_deviation": 0.033047385443001454, "minimum": 2.079527893563246, "maximum": 2.1684688706557482, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7838404922369271, "standard_deviation": 0.24695124110226557, "minimum": 0.4405152721534451, "maximum": 1.072504458835567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21898143626055294, "standard_deviation": 0.005034531639803505, "minimum": 0.21004474041855098, "maximum": 0.22472896272034676, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9852309038357496, "standard_deviation": 0.000919094930334773, "minimum": 0.9841706992412504, "maximum": 0.9867779740490934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4469622769106794, "standard_deviation": 0.044824754601385346, "minimum": 1.3802781035246585, "maximum": 1.5050224756793422, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.074432920447774, "standard_deviation": 0.30918215098127105, "minimum": 0.5459279681936838, "maximum": 1.4440396454113635, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14273261976718915, "standard_deviation": 0.003928075474841137, "minimum": 0.13783779481586902, "maximum": 0.14739621601705194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9711166482242041, "standard_deviation": 0.006824188235866751, "minimum": 0.957277048008784, "maximum": 0.979980608961101, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1199780504619437, "standard_deviation": 0.039042185491993904, "minimum": 2.077999880169983, "maximum": 2.200176365284599, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.4376924320319966, "standard_deviation": 0.5058309525534473, "minimum": 0.7315695296188723, "maximum": 2.160239824829429, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2212135994514185, "standard_deviation": 0.025757577621397375, "minimum": 0.18454077772540303, "maximum": 0.27206356232691564, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9869395011288613, "standard_deviation": 0.0036342572303380235, "minimum": 0.979224225828304, "maximum": 0.9905830336805216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.0072095120026985, "standard_deviation": 0.041078771163620804, "minimum": 1.9423786743383806, "maximum": 2.0900789593392792, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.523948121929087, "standard_deviation": 0.744198790709089, "minimum": 0.4908201812866513, "maximum": 2.912942666079244, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14874775698388099, "standard_deviation": 0.019372175691337208, "minimum": 0.12586649425008228, "maximum": 0.18723711257710865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9877111691826167, "standard_deviation": 0.0016732794620538546, "minimum": 0.9854860466451801, "maximum": 0.990160465765298, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1616311567224304, "standard_deviation": 0.05447902116574842, "minimum": 2.0955591876892528, "maximum": 2.218733727621946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6119862595631208, "standard_deviation": 0.19722535381272052, "minimum": 0.3749459624632654, "maximum": 0.919792595197987, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13667185948538868, "standard_deviation": 0.00889040209626819, "minimum": 0.12375280271342565, "maximum": 0.14888501862462702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968119242850846, "standard_deviation": 0.0006744327715119885, "minimum": 0.9959781748955892, "maximum": 0.9980563220541039, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4982082327969954, "standard_deviation": 0.054936222807175306, "minimum": 1.4310849989164052, "maximum": 1.579402162578561, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5803853264625887, "standard_deviation": 0.23330844658886973, "minimum": 0.34232177305457673, "maximum": 1.0402895649940511, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07151569720354194, "standard_deviation": 0.006131501508561138, "minimum": 0.06058643903204477, "maximum": 0.08132131267282558, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2545429786424461, "standard_deviation": 0.05252895156260686, "minimum": 0.17822093389854166, "maximum": 0.35045355152294944, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.13049373961182, "standard_deviation": 0.011057235375763311, "minimum": 6.106857241911047, "maximum": 6.145600355030737, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2573690254473875, "standard_deviation": 0.007213863251977483, "minimum": 0.24421710743324773, "maximum": 0.26733087585901155, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05079305246292046, "standard_deviation": 0.05272192986586224, "minimum": -0.10212437558605345, "maximum": 0.029453847604196042, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9904488370484482, "standard_deviation": 0.0009640760802099146, "minimum": 0.9889468561573461, "maximum": 0.9916431710828876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.479259999959782, "standard_deviation": 0.03898942726066418, "minimum": 1.4327427464533393, "maximum": 1.5251781648424882, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.47336489028327716, "standard_deviation": 0.2726643500708476, "minimum": 0.16429564655802675, "maximum": 0.841251152741156, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10435304169128788, "standard_deviation": 0.005189568757075839, "minimum": 0.09778418580452784, "maximum": 0.11222963892000558, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977356041360463, "standard_deviation": 0.0003547356359587294, "minimum": 0.9971139655970788, "maximum": 0.9983163368806923, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7741124883216652, "standard_deviation": 0.020334212006948776, "minimum": 0.7522037021312262, "maximum": 0.8142081650335828, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5344850769999107, "standard_deviation": 0.23851935382535075, "minimum": 0.3132500065994361, "maximum": 1.0546078384776056, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04994971807732773, "standard_deviation": 0.00350602343759002, "minimum": 0.043999020639457785, "maximum": 0.05606112774340939, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0003", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.004856824874878, "predicted_mos": 3.004856824874878, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "aaeef230fb592eca0445701ff607a8675bf093411f9964bbaad382ead27ed5a2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3765412997454405, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5814722083548336, "全帯域_10ms_rms_envelope_nrmse": 0.09901476775089901, "全帯域_10ms_envelope_correlation": 0.9914178902362892, "全帯域_10ms_modulation_mae_db": 0.30428249754802533, "全帯域_40ms_log_rms_mae_db": 0.8570387893385399, "全帯域_40ms_rms_envelope_nrmse": 0.050926341698436045, "全帯域_40ms_envelope_correlation": 0.9976020448918848, "全帯域_40ms_modulation_mae_db": 0.41042015895637435, "300–1000 Hz_10ms_log_rms_mae_db": 1.7388433148749505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11357552612776331, "300–1000 Hz_10ms_envelope_correlation": 0.9914684029161338, "300–1000 Hz_10ms_modulation_mae_db": 0.420450093398014, "300–1000 Hz_40ms_log_rms_mae_db": 1.0628445866561105, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06308615127587745, "300–1000 Hz_40ms_envelope_correlation": 0.9973400878334102, "300–1000 Hz_40ms_modulation_mae_db": 0.18630583763629108, "1000–3000 Hz_10ms_log_rms_mae_db": 1.501957682769212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19950909561442107, "1000–3000 Hz_10ms_envelope_correlation": 0.9717516027023306, "1000–3000 Hz_10ms_modulation_mae_db": 0.29081635209383633, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9027189944170037, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12118712077158753, "1000–3000 Hz_40ms_envelope_correlation": 0.9891831712261786, "1000–3000 Hz_40ms_modulation_mae_db": 0.46850632973072237, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2863459879391614, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2213382670607542, "3000–6000 Hz_10ms_envelope_correlation": 0.9729573114383842, "3000–6000 Hz_10ms_modulation_mae_db": 1.7331785590374575, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1844532325951485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17334402968221793, "3000–6000 Hz_40ms_envelope_correlation": 0.9847979143499086, "3000–6000 Hz_40ms_modulation_mae_db": 1.6271011018531718, "mr_stft_db": 5.666507618216606, "band_mae_db": 0.33987855451806603, "rms_gain_error_db": 0.006007041331120481, "output_peak": 0.2590777824751809}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5677538897972232, "全帯域_10ms_rms_envelope_nrmse": 0.10228643221549243, "全帯域_10ms_envelope_correlation": 0.9908710085854152, "全帯域_10ms_modulation_mae_db": 0.3679959508672195, "全帯域_40ms_log_rms_mae_db": 0.840946553381921, "全帯域_40ms_rms_envelope_nrmse": 0.053147396291699664, "全帯域_40ms_envelope_correlation": 0.9973976601962526, "全帯域_40ms_modulation_mae_db": 0.14685743490033473, "300–1000 Hz_10ms_log_rms_mae_db": 1.7147470365070259, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14573647387479585, "300–1000 Hz_10ms_envelope_correlation": 0.9859167602916716, "300–1000 Hz_10ms_modulation_mae_db": 0.638963316323797, "300–1000 Hz_40ms_log_rms_mae_db": 1.0450798806857444, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08625529669520852, "300–1000 Hz_40ms_envelope_correlation": 0.9948987428387626, "300–1000 Hz_40ms_modulation_mae_db": 0.5714228855046541, "1000–3000 Hz_10ms_log_rms_mae_db": 1.553173409922483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2005014146341385, "1000–3000 Hz_10ms_envelope_correlation": 0.9715737586811249, "1000–3000 Hz_10ms_modulation_mae_db": 0.7038670910127349, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9183794861268325, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12967420419057044, "1000–3000 Hz_40ms_envelope_correlation": 0.9880519752358343, "1000–3000 Hz_40ms_modulation_mae_db": 0.874356050355002, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2829223573121518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22343072858603996, "3000–6000 Hz_10ms_envelope_correlation": 0.9705596640360324, "3000–6000 Hz_10ms_modulation_mae_db": 0.4661123757345438, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1858342958886474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14693295872478868, "3000–6000 Hz_40ms_envelope_correlation": 0.9868734132150206, "3000–6000 Hz_40ms_modulation_mae_db": 0.5336647978803678, "mr_stft_db": 5.680133848259818, "band_mae_db": 0.22866652319434444, "rms_gain_error_db": 0.024509850180339805, "output_peak": 0.310574236329677}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.544432679197751, "全帯域_10ms_rms_envelope_nrmse": 0.10542253667888504, "全帯域_10ms_envelope_correlation": 0.9902633439485174, "全帯域_10ms_modulation_mae_db": 0.33044721152331574, "全帯域_40ms_log_rms_mae_db": 0.786142662882415, "全帯域_40ms_rms_envelope_nrmse": 0.046667506238757914, "全帯域_40ms_envelope_correlation": 0.9979919389390616, "全帯域_40ms_modulation_mae_db": 0.5799309680796645, "300–1000 Hz_10ms_log_rms_mae_db": 1.6839600586112737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1531494572136904, "300–1000 Hz_10ms_envelope_correlation": 0.9843666891224075, "300–1000 Hz_10ms_modulation_mae_db": 0.4758549785076666, "300–1000 Hz_40ms_log_rms_mae_db": 1.0171581832410042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08039781097362501, "300–1000 Hz_40ms_envelope_correlation": 0.9956821500538688, "300–1000 Hz_40ms_modulation_mae_db": 0.5526588777504483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5220479204741981, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21567612130691732, "1000–3000 Hz_10ms_envelope_correlation": 0.9671467295764665, "1000–3000 Hz_10ms_modulation_mae_db": 0.5454460910443476, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9234284629507634, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1553892103363228, "1000–3000 Hz_40ms_envelope_correlation": 0.9821689337269015, "1000–3000 Hz_40ms_modulation_mae_db": 0.748543867897502, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2880696193016512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2527701544084585, "3000–6000 Hz_10ms_envelope_correlation": 0.9663305047272336, "3000–6000 Hz_10ms_modulation_mae_db": 1.6874173522606686, "3000–6000 Hz_40ms_log_rms_mae_db": 1.196285451852767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1762221935769576, "3000–6000 Hz_40ms_envelope_correlation": 0.9840161524581301, "3000–6000 Hz_40ms_modulation_mae_db": 2.117203593765089, "mr_stft_db": 5.6673395402652575, "band_mae_db": 0.21347208709295584, "rms_gain_error_db": -0.033121720339649656, "output_peak": 0.27280858075924397}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6105299124350334, "全帯域_10ms_rms_envelope_nrmse": 0.10145777890193294, "全帯域_10ms_envelope_correlation": 0.990967492526572, "全帯域_10ms_modulation_mae_db": 0.29767206827939696, "全帯域_40ms_log_rms_mae_db": 0.8358349918267916, "全帯域_40ms_rms_envelope_nrmse": 0.05056141129203192, "全帯域_40ms_envelope_correlation": 0.9976405214774795, "全帯域_40ms_modulation_mae_db": 0.1545816072454572, "300–1000 Hz_10ms_log_rms_mae_db": 1.5778664390787656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12935667808440943, "300–1000 Hz_10ms_envelope_correlation": 0.9889196379127664, "300–1000 Hz_10ms_modulation_mae_db": 0.24226542586007027, "300–1000 Hz_40ms_log_rms_mae_db": 0.9189153246056879, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06465363120052217, "300–1000 Hz_40ms_envelope_correlation": 0.9973123222010278, "300–1000 Hz_40ms_modulation_mae_db": 0.2808750417795947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.579106364609268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21721231900920449, "1000–3000 Hz_10ms_envelope_correlation": 0.9664596064811168, "1000–3000 Hz_10ms_modulation_mae_db": 0.6384728803698002, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9376358272399781, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14704320738008939, "1000–3000 Hz_40ms_envelope_correlation": 0.9839905204855278, "1000–3000 Hz_40ms_modulation_mae_db": 0.8189949869694967, "3000–6000 Hz_10ms_log_rms_mae_db": 1.305114728297625, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2158568237043157, "3000–6000 Hz_10ms_envelope_correlation": 0.9722528975613528, "3000–6000 Hz_10ms_modulation_mae_db": 1.1519351931641055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1707473086165605, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14849728521081032, "3000–6000 Hz_40ms_envelope_correlation": 0.9868225997866432, "3000–6000 Hz_40ms_modulation_mae_db": 1.1968672241064127, "mr_stft_db": 5.682407184514665, "band_mae_db": 0.24565059121536081, "rms_gain_error_db": -0.03778532745770187, "output_peak": 0.25488372304467116}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6326197759681085, "全帯域_10ms_rms_envelope_nrmse": 0.10641098293792774, "全帯域_10ms_envelope_correlation": 0.9901188527928801, "全帯域_10ms_modulation_mae_db": 0.26623188346625626, "全帯域_40ms_log_rms_mae_db": 0.8970586481346359, "全帯域_40ms_rms_envelope_nrmse": 0.05354584508169215, "全帯域_40ms_envelope_correlation": 0.9975031361101747, "全帯域_40ms_modulation_mae_db": 0.13765908626849493, "300–1000 Hz_10ms_log_rms_mae_db": 1.7326876577832533, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14048575706785335, "300–1000 Hz_10ms_envelope_correlation": 0.9872087000712909, "300–1000 Hz_10ms_modulation_mae_db": 0.44658672758937695, "300–1000 Hz_40ms_log_rms_mae_db": 1.1272275311638198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07308624467282589, "300–1000 Hz_40ms_envelope_correlation": 0.996995037946065, "300–1000 Hz_40ms_modulation_mae_db": 0.42864234445183963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.515714856954341, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19917517871693294, "1000–3000 Hz_10ms_envelope_correlation": 0.9718967938768558, "1000–3000 Hz_10ms_modulation_mae_db": 0.7017367984397013, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9127001675933315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11457057178384149, "1000–3000 Hz_40ms_envelope_correlation": 0.9906657360883754, "1000–3000 Hz_40ms_modulation_mae_db": 0.6473197133024909, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2811394924492603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23869309042390566, "3000–6000 Hz_10ms_envelope_correlation": 0.9669669228704016, "3000–6000 Hz_10ms_modulation_mae_db": 0.4518243230654832, "3000–6000 Hz_40ms_log_rms_mae_db": 1.149228056551677, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14820412125131208, "3000–6000 Hz_40ms_envelope_correlation": 0.9869145740822853, "3000–6000 Hz_40ms_modulation_mae_db": 0.22234177525168922, "mr_stft_db": 5.667950980021145, "band_mae_db": 0.22609424641522202, "rms_gain_error_db": -0.12642945694924104, "output_peak": 0.24232760618422788}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5526226056116927, "全帯域_10ms_rms_envelope_nrmse": 0.09281073471046634, "全帯域_10ms_envelope_correlation": 0.9925122064977868, "全帯域_10ms_modulation_mae_db": 0.4804297101948259, "全帯域_40ms_log_rms_mae_db": 0.7731446320992742, "全帯域_40ms_rms_envelope_nrmse": 0.04440653408292103, "全帯域_40ms_envelope_correlation": 0.9983199319670255, "全帯域_40ms_modulation_mae_db": 0.36474071876868264, "300–1000 Hz_10ms_log_rms_mae_db": 1.6530420821469451, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11151366560878204, "300–1000 Hz_10ms_envelope_correlation": 0.9920577677430231, "300–1000 Hz_10ms_modulation_mae_db": 0.9462186601052874, "300–1000 Hz_40ms_log_rms_mae_db": 0.9982336662473822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06477134169768069, "300–1000 Hz_40ms_envelope_correlation": 0.9975836215655454, "300–1000 Hz_40ms_modulation_mae_db": 1.3773585912991555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5315205276172057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1995672075526427, "1000–3000 Hz_10ms_envelope_correlation": 0.9717032138068681, "1000–3000 Hz_10ms_modulation_mae_db": 0.46568233608665643, "1000–3000 Hz_40ms_log_rms_mae_db": 0.97281731417605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14002360731481697, "1000–3000 Hz_40ms_envelope_correlation": 0.9855679014555202, "1000–3000 Hz_40ms_modulation_mae_db": 0.6216358276129823, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3003821983387334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18560816489591742, "3000–6000 Hz_10ms_envelope_correlation": 0.9800557807399333, "3000–6000 Hz_10ms_modulation_mae_db": 0.7867123521083569, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1466425141797942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1342416840984132, "3000–6000 Hz_40ms_envelope_correlation": 0.9892751504112661, "3000–6000 Hz_40ms_modulation_mae_db": 0.4796199113278902, "mr_stft_db": 5.678934272796192, "band_mae_db": 0.2226301085789226, "rms_gain_error_db": -0.11772952012261506, "output_peak": 0.25255049038346644}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5947247150407666, "全帯域_10ms_rms_envelope_nrmse": 0.1143074529884532, "全帯域_10ms_envelope_correlation": 0.9885700450554635, "全帯域_10ms_modulation_mae_db": 0.4823296637600349, "全帯域_40ms_log_rms_mae_db": 0.8496948582937514, "全帯域_40ms_rms_envelope_nrmse": 0.05785611417639842, "全帯域_40ms_envelope_correlation": 0.996993642777665, "全帯域_40ms_modulation_mae_db": 0.31200030723285527, "300–1000 Hz_10ms_log_rms_mae_db": 1.628880655884207, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14577394414865832, "300–1000 Hz_10ms_envelope_correlation": 0.9859043560009182, "300–1000 Hz_10ms_modulation_mae_db": 0.37833416894287336, "300–1000 Hz_40ms_log_rms_mae_db": 0.9911406940784852, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0827017544588392, "300–1000 Hz_40ms_envelope_correlation": 0.9955356633521671, "300–1000 Hz_40ms_modulation_mae_db": 0.5583754066564138, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5545244339235547, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21324884717417125, "1000–3000 Hz_10ms_envelope_correlation": 0.9682090930718973, "1000–3000 Hz_10ms_modulation_mae_db": 0.6350055029920673, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9563621898436662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15165181611876624, "1000–3000 Hz_40ms_envelope_correlation": 0.9841767046240054, "1000–3000 Hz_40ms_modulation_mae_db": 0.632598316020989, "3000–6000 Hz_10ms_log_rms_mae_db": 1.251951345974347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24866088568080477, "3000–6000 Hz_10ms_envelope_correlation": 0.9646363750660024, "3000–6000 Hz_10ms_modulation_mae_db": 1.7249490918584875, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1741716069433483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18030282910716236, "3000–6000 Hz_40ms_envelope_correlation": 0.9831461145148354, "3000–6000 Hz_40ms_modulation_mae_db": 1.3901741649864612, "mr_stft_db": 5.6567151345150934, "band_mae_db": 0.3997144478733448, "rms_gain_error_db": -0.09815600989505747, "output_peak": 0.25277563809668974}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5880371326326672, "全帯域_10ms_rms_envelope_nrmse": 0.10550613233891216, "全帯域_10ms_envelope_correlation": 0.9902351490908002, "全帯域_10ms_modulation_mae_db": 0.6575848369629763, "全帯域_40ms_log_rms_mae_db": 0.8795061688154329, "全帯域_40ms_rms_envelope_nrmse": 0.04817561495511844, "全帯域_40ms_envelope_correlation": 0.9978655840309931, "全帯域_40ms_modulation_mae_db": 0.8239679137745873, "300–1000 Hz_10ms_log_rms_mae_db": 1.6971747220702038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13863861826000556, "300–1000 Hz_10ms_envelope_correlation": 0.9872134229135031, "300–1000 Hz_10ms_modulation_mae_db": 0.5810086331179245, "300–1000 Hz_40ms_log_rms_mae_db": 1.1030798780972737, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07293391349055577, "300–1000 Hz_40ms_envelope_correlation": 0.9964368211340162, "300–1000 Hz_40ms_modulation_mae_db": 0.16907584486323513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5630593369234462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.213948812501283, "1000–3000 Hz_10ms_envelope_correlation": 0.9674893764187493, "1000–3000 Hz_10ms_modulation_mae_db": 0.7054404533501953, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9707274691791091, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14736721843384568, "1000–3000 Hz_40ms_envelope_correlation": 0.9842217617656668, "1000–3000 Hz_40ms_modulation_mae_db": 0.613043940006807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.258347624182233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2268864562348191, "3000–6000 Hz_10ms_envelope_correlation": 0.9708187919604223, "3000–6000 Hz_10ms_modulation_mae_db": 1.7120590859848168, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1806760516751569, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14133640579442952, "3000–6000 Hz_40ms_envelope_correlation": 0.9884975121238538, "3000–6000 Hz_40ms_modulation_mae_db": 2.357808348939672, "mr_stft_db": 5.671928895141889, "band_mae_db": 0.16718972667447551, "rms_gain_error_db": -0.04594230533033882, "output_peak": 0.24967524564821073}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.808118642507652, "spectral_convergence": 0.30219614428196623, "spectral_convergence_status": "ok", "amplitude_floor": 8.79928282883349e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.451992627253154, "spectral_convergence": 0.27967229487935713, "spectral_convergence_status": "ok", "amplitude_floor": 4.444740709272264e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7394115848890115, "spectral_convergence": 0.3105732175727368, "spectral_convergence_status": "ok", "amplitude_floor": 3.930299196146418e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002477612487613689, "output_power": 0.0002619999058462476, "delta_db": 0.24267753972532582, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004515434578601184, "output_power": 0.00044518754825889406, "delta_db": -0.06156545662639206, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015370997116088465, "output_power": 0.00015205684510063319, "delta_db": -0.046960656487228254, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.4065683577917706e-05, "output_power": 4.2305040569805025e-05, "delta_db": -0.17708395724694054, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.662327944441642e-06, "output_power": 1.0019082234542899e-05, "delta_db": 0.6319331969110719, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.476431620775013e-05, "output_power": 2.02265256114711e-05, "delta_db": -0.8790505201114379, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.33987855451806603, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.006007041331120481, "pitch": null, "output_rms_dbfs": -30.306796487128796, "output_peak": 0.2590777824751809, "reference_rms_dbfs": -30.312803528459913, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.666507618216606}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.467120737686064e-05, "log_rms_mae_db": 1.5814722083548336, "rms_envelope_nrmse": 0.09901476775089901, "envelope_correlation": 0.9914178902362892, "rms_gain_error_db": 0.006003475524580453, "reference_mean_power": 0.0009312818437003785, "output_mean_power": 0.0009325700925920167}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.494393820063344e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023666657525761142, "output_power": 0.00023418555823417068, "reference_active": true, "delta_db": -0.04576816789531307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5441929621447868e-05, "output_power": 1.4395304295812045e-05, "reference_active": true, "delta_db": -0.3048071910545677}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.985747798516416e-06, "output_power": 7.951336184557122e-06, "reference_active": true, "delta_db": 0.5622721336941953}], "active_band_count": 3, "active_band_mae_db": 0.30428249754802533}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.334380857742655e-05, "log_rms_mae_db": 0.8570387893385399, "rms_envelope_nrmse": 0.050926341698436045, "envelope_correlation": 0.9976020448918848, "rms_gain_error_db": 0.005998983831648155, "reference_mean_power": 0.0009359486205868301, "output_mean_power": 0.0009372423557207024}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.189856446091556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021037225251172397, "output_power": 0.0002101685217551756, "reference_active": true, "delta_db": -0.004207874988528204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.644252267790675e-06, "output_power": 6.7836903360085724e-06, "reference_active": true, "delta_db": -0.5186899586903138}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4342900916585017e-07, "output_power": 4.0427160734196325e-07, "reference_active": true, "delta_db": 0.7083626431902813}], "active_band_count": 3, "active_band_mae_db": 0.41042015895637435}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.589555428061273e-05, "log_rms_mae_db": 1.7388433148749505, "rms_envelope_nrmse": 0.11357552612776331, "envelope_correlation": 0.9914684029161338, "rms_gain_error_db": -0.03916333199054012, "reference_mean_power": 0.00038327408961782034, "output_mean_power": 0.0003798333794683638}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.028222657916465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015396875485141256, "output_power": 0.0001488921896176618, "reference_active": true, "delta_db": -0.14560680914014903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.75189127488202e-06, "output_power": 8.741048622114071e-06, "reference_active": true, "delta_db": -0.005383777566353715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1315883357233815e-06, "output_power": 5.335228348301224e-06, "reference_active": true, "delta_db": 1.1103596934875393}], "active_band_count": 3, "active_band_mae_db": 0.420450093398014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.149058869949865e-05, "log_rms_mae_db": 1.0628445866561105, "rms_envelope_nrmse": 0.06308615127587745, "envelope_correlation": 0.9973400878334102, "rms_gain_error_db": -0.03916265927911646, "reference_mean_power": 0.00038520090394369153, "output_mean_power": 0.00038174295561823495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9052955697898775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014058163319658512, "output_power": 0.00013707367005477616, "reference_active": true, "delta_db": -0.10974543521542021}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.901478078483036e-06, "output_power": 3.917048016112116e-06, "reference_active": true, "delta_db": 0.01729724209598455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3393802456709067e-07, "output_power": 2.5839749661765617e-07, "reference_active": true, "delta_db": 0.43187483559746853}], "active_band_count": 3, "active_band_mae_db": 0.18630583763629108}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.043858753489399e-05, "log_rms_mae_db": 1.501957682769212, "rms_envelope_nrmse": 0.19950909561442107, "envelope_correlation": 0.9717516027023306, "rms_gain_error_db": -0.13015287120893565, "reference_mean_power": 4.084610004445957e-05, "output_mean_power": 3.964015186244633e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0558567323423266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3562756220605441e-05, "output_power": 1.1977583143170214e-05, "reference_active": true, "delta_db": -0.5397876151382582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5216123885214767e-06, "output_power": 1.4796217941827752e-06, "reference_active": true, "delta_db": -0.12153315847438645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602773162915538e-07, "output_power": 6.931691177734321e-07, "reference_active": true, "delta_db": 0.21112828266886416}], "active_band_count": 3, "active_band_mae_db": 0.29081635209383633}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.7804457921821952e-05, "log_rms_mae_db": 0.9027189944170037, "rms_envelope_nrmse": 0.12118712077158753, "envelope_correlation": 0.9891831712261786, "rms_gain_error_db": -0.1301593089124966, "reference_mean_power": 4.105094002568101e-05, "output_mean_power": 3.983888505366415e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8992373996258767e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2457002317192863e-05, "output_power": 1.1188458607295704e-05, "reference_active": true, "delta_db": -0.4664328573460089}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.119435793981677e-07, "output_power": 6.984294792770085e-07, "reference_active": true, "delta_db": -0.6540329012776229}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.742155092447501e-08, "output_power": 3.504422497080547e-08, "reference_active": true, "delta_db": -0.2850532305685355}], "active_band_count": 3, "active_band_mae_db": 0.46850632973072237}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.7924378350629476e-05, "log_rms_mae_db": 1.2863459879391614, "rms_envelope_nrmse": 0.2213382670607542, "envelope_correlation": 0.9729573114383842, "rms_gain_error_db": -0.7413597161919094, "reference_mean_power": 2.620464364014996e-05, "output_mean_power": 2.209236889732761e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.347390510898905e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.954495563916905e-05, "output_power": 1.4468541110072178e-05, "reference_active": true, "delta_db": -1.3060994645332964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.79596482520383e-06, "output_power": 1.2763062098246198e-06, "reference_active": true, "delta_db": -3.4057682115411545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.14743402003256e-06, "output_power": 1.0255596416709376e-06, "reference_active": true, "delta_db": -0.4876680010379212}], "active_band_count": 3, "active_band_mae_db": 1.7331785590374575}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.7050501746546627e-05, "log_rms_mae_db": 1.1844532325951485, "rms_envelope_nrmse": 0.17334402968221793, "envelope_correlation": 0.9847979143499086, "rms_gain_error_db": -0.7413602559465546, "reference_mean_power": 2.633641752064706e-05, "output_mean_power": 2.2203460843733773e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2809655135480154e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9754827323518705e-05, "output_power": 1.4401954313757227e-05, "reference_active": true, "delta_db": -1.3725180896904101}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4198245590923022e-06, "output_power": 6.966952338351262e-07, "reference_active": true, "delta_db": -3.091918445222979}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.51727649987637e-08, "output_power": 5.9207874351403294e-08, "reference_active": true, "delta_db": -0.4168667706461266}], "active_band_count": 3, "active_band_mae_db": 1.6271011018531718}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.735713194918816, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "de569f8f998a5115ab9f88c084757155ccffe58e44f343e4189f33e7900acdc8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9695287718269262, "standard_deviation": 0.0024039628455009418, "minimum": 0.9664596064811168, "maximum": 0.9718967938768558, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5401380666492135, "standard_deviation": 0.026394130481491005, "minimum": 1.501957682769212, "maximum": 1.579106364609268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5858084381736675, "standard_deviation": 0.14636679671099134, "minimum": 0.29081635209383633, "maximum": 0.7054404533501953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20735487456371393, "standard_deviation": 0.008286931458310692, "minimum": 0.19917517871693294, "maximum": 0.21721231900920449, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9860033380760013, "standard_deviation": 0.0029640828172898728, "minimum": 0.9821689337269015, "maximum": 0.9906657360883754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9368462389408417, "standard_deviation": 0.026971177890697678, "minimum": 0.9027189944170037, "maximum": 0.97281731417605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6781248789869991, "standard_deviation": 0.12966880770583547, "minimum": 0.46850632973072237, "maximum": 0.874356050355002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13836336954123007, "standard_deviation": 0.014939160800155967, "minimum": 0.11457057178384149, "maximum": 0.1553892103363228, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9705722810499704, "standard_deviation": 0.0048511675888409655, "minimum": 0.9646363750660024, "maximum": 0.9800557807399333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2817841692243954, "standard_deviation": 0.018483578863091005, "minimum": 1.251951345974347, "maximum": 1.305114728297625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2142735416517398, "standard_deviation": 0.576872854265236, "minimum": 0.4518243230654832, "maximum": 1.7331785590374575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22665557137437692, "standard_deviation": 0.021217702661012174, "minimum": 0.18560816489591742, "maximum": 0.2527701544084585, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9862929288677429, "standard_deviation": 0.002140727800134302, "minimum": 0.9831461145148354, "maximum": 0.9892751504112661, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1735048147878875, "standard_deviation": 0.017571326711247955, "minimum": 1.1466425141797942, "maximum": 1.196285451852767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2405976147638444, "standard_deviation": 0.7843062750307481, "minimum": 0.22234177525168922, "maximum": 2.357808348939672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15613518843076146, "standard_deviation": 0.01767951507114736, "minimum": 0.1342416840984132, "maximum": 0.18030282910716236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9878819671214643, "standard_deviation": 0.0027361859828415663, "minimum": 0.9843666891224075, "maximum": 0.9920577677430231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6784002458695781, "standard_deviation": 0.05542980363177742, "minimum": 1.5778664390787656, "maximum": 1.7388433148749505, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5162102504806263, "standard_deviation": 0.21175320854749644, "minimum": 0.24226542586007027, "maximum": 0.9462186601052874, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1347787650482448, "standard_deviation": 0.015335100359417885, "minimum": 0.11151366560878204, "maximum": 0.1531494572136904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964730558656079, "standard_deviation": 0.0009961314507865025, "minimum": 0.9948987428387626, "maximum": 0.9975836215655454, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0329599680969386, "standard_deviation": 0.06660523656576187, "minimum": 0.9189153246056879, "maximum": 1.1272275311638198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5155893537427041, "standard_deviation": 0.38547155434833347, "minimum": 0.16907584486323513, "maximum": 1.3773585912991555, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07348576805814185, "standard_deviation": 0.008930970704695824, "minimum": 0.06308615127587745, "maximum": 0.08625529669520852, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2554120356953365, "standard_deviation": 0.07582377106646737, "minimum": 0.16718972667447551, "maximum": 0.3997144478733448, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.671489684216333, "standard_deviation": 0.008638639218492762, "minimum": 5.6567151345150934, "maximum": 5.682407184514665, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.26183416286517097, "standard_deviation": 0.021542407754925733, "minimum": 0.24232760618422788, "maximum": 0.310574236329677, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05358093107289295, "standard_deviation": 0.05578557471505295, "minimum": -0.12642945694924104, "maximum": 0.024509850180339805, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9906194985917156, "standard_deviation": 0.001142028656966585, "minimum": 0.9885700450554635, "maximum": 0.9925122064977868, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5840241148797594, "standard_deviation": 0.029324469915484753, "minimum": 1.544432679197751, "maximum": 1.6326197759681085, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3983717278252563, "standard_deviation": 0.13253811005600727, "minimum": 0.26623188346625626, "maximum": 0.6575848369629763, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10340210231537111, "standard_deviation": 0.006244311913271419, "minimum": 0.09281073471046634, "maximum": 0.1143074529884532, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976643075488171, "standard_deviation": 0.00040160546063232886, "minimum": 0.996993642777665, "maximum": 0.9983199319670255, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8399209130965952, "standard_deviation": 0.04240819246645751, "minimum": 0.7731446320992742, "maximum": 0.8970586481346359, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3662697744033064, "standard_deviation": 0.24060992841755593, "minimum": 0.13765908626849493, "maximum": 0.8239679137745873, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05066084547713194, "standard_deviation": 0.004269459061806955, "minimum": 0.04440653408292103, "maximum": 0.05785611417639842, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0003", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.604463577270508, "predicted_mos": 2.604463577270508, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7c444d2fc6c66128529232fe6feeec192f1cf9df2d92d6899847e17cb774d18b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8343528998084366, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5242717907123147, "全帯域_10ms_rms_envelope_nrmse": 0.09835176618256397, "全帯域_10ms_envelope_correlation": 0.9915464128120096, "全帯域_10ms_modulation_mae_db": 0.436702321205469, "全帯域_40ms_log_rms_mae_db": 0.7663147809892167, "全帯域_40ms_rms_envelope_nrmse": 0.05044570774115428, "全帯域_40ms_envelope_correlation": 0.9977460018647478, "全帯域_40ms_modulation_mae_db": 0.22280735465638568, "300–1000 Hz_10ms_log_rms_mae_db": 1.8594841885226177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12891385635191366, "300–1000 Hz_10ms_envelope_correlation": 0.9893971860660898, "300–1000 Hz_10ms_modulation_mae_db": 0.9238000074751994, "300–1000 Hz_40ms_log_rms_mae_db": 1.3013298880070945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08900908421937219, "300–1000 Hz_40ms_envelope_correlation": 0.9951876762114152, "300–1000 Hz_40ms_modulation_mae_db": 0.9415439781129136, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7808563188168935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20939755915668334, "1000–3000 Hz_10ms_envelope_correlation": 0.9688454899362473, "1000–3000 Hz_10ms_modulation_mae_db": 0.9391313755649634, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1625338188105363, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12918098106434422, "1000–3000 Hz_40ms_envelope_correlation": 0.9877819906657738, "1000–3000 Hz_40ms_modulation_mae_db": 0.9761630677895918, "3000–6000 Hz_10ms_log_rms_mae_db": 1.597098866452411, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2546119796804401, "3000–6000 Hz_10ms_envelope_correlation": 0.9611816460405975, "3000–6000 Hz_10ms_modulation_mae_db": 1.3350218507175544, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6339956525330357, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16010363948070427, "3000–6000 Hz_40ms_envelope_correlation": 0.9844857278246187, "3000–6000 Hz_40ms_modulation_mae_db": 1.210893344862135, "mr_stft_db": 5.8417733105930765, "band_mae_db": 0.3738619510799753, "rms_gain_error_db": -0.1031278972296274, "output_peak": 0.24227797964759473}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4592713842425362, "全帯域_10ms_rms_envelope_nrmse": 0.10321539784446379, "全帯域_10ms_envelope_correlation": 0.9906602657110126, "全帯域_10ms_modulation_mae_db": 0.2925180978818674, "全帯域_40ms_log_rms_mae_db": 0.7263680664133693, "全帯域_40ms_rms_envelope_nrmse": 0.05256517892486072, "全帯域_40ms_envelope_correlation": 0.9974801232139113, "全帯域_40ms_modulation_mae_db": 0.32456609039214285, "300–1000 Hz_10ms_log_rms_mae_db": 1.9571397798210146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13636598599785793, "300–1000 Hz_10ms_envelope_correlation": 0.9878528370221183, "300–1000 Hz_10ms_modulation_mae_db": 0.5008339604527713, "300–1000 Hz_40ms_log_rms_mae_db": 1.403064058439114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08259842154874174, "300–1000 Hz_40ms_envelope_correlation": 0.9957131486249814, "300–1000 Hz_40ms_modulation_mae_db": 0.47347618044394385, "1000–3000 Hz_10ms_log_rms_mae_db": 1.759399806066103, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20757516322116154, "1000–3000 Hz_10ms_envelope_correlation": 0.9694713581876927, "1000–3000 Hz_10ms_modulation_mae_db": 0.725599423739537, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2431070794752248, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12830743580118714, "1000–3000 Hz_40ms_envelope_correlation": 0.9878593077071512, "1000–3000 Hz_40ms_modulation_mae_db": 0.7731096832213183, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6193646428335222, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19295249408933088, "3000–6000 Hz_10ms_envelope_correlation": 0.9793524913731626, "3000–6000 Hz_10ms_modulation_mae_db": 0.7603449071512566, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5844248734973387, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1269453620999099, "3000–6000 Hz_40ms_envelope_correlation": 0.9911176844911214, "3000–6000 Hz_40ms_modulation_mae_db": 0.7276296182619836, "mr_stft_db": 5.842719508833632, "band_mae_db": 0.2187423715119076, "rms_gain_error_db": -0.01753261418356165, "output_peak": 0.2560062948573231}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5358781784301085, "全帯域_10ms_rms_envelope_nrmse": 0.1035129058150958, "全帯域_10ms_envelope_correlation": 0.9905988259480893, "全帯域_10ms_modulation_mae_db": 0.2525226469262181, "全帯域_40ms_log_rms_mae_db": 0.7884448160070014, "全帯域_40ms_rms_envelope_nrmse": 0.05658577027010334, "全帯域_40ms_envelope_correlation": 0.9970324596679678, "全帯域_40ms_modulation_mae_db": 0.36730807628163076, "300–1000 Hz_10ms_log_rms_mae_db": 1.7992052119620525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12286215042792112, "300–1000 Hz_10ms_envelope_correlation": 0.9899857118252436, "300–1000 Hz_10ms_modulation_mae_db": 0.21360814422356297, "300–1000 Hz_40ms_log_rms_mae_db": 1.2530023091360658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0658557973667674, "300–1000 Hz_40ms_envelope_correlation": 0.9971034824835758, "300–1000 Hz_40ms_modulation_mae_db": 0.5113912776652012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7671717066600487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2057186580720154, "1000–3000 Hz_10ms_envelope_correlation": 0.9699324591189304, "1000–3000 Hz_10ms_modulation_mae_db": 1.015539744495819, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1829197338830315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1282148286078688, "1000–3000 Hz_40ms_envelope_correlation": 0.9879803719042586, "1000–3000 Hz_40ms_modulation_mae_db": 1.2576339067522038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6168552141354962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1912761869513718, "3000–6000 Hz_10ms_envelope_correlation": 0.9783251787829194, "3000–6000 Hz_10ms_modulation_mae_db": 0.42166973552046966, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5736620886109849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12940173882163802, "3000–6000 Hz_40ms_envelope_correlation": 0.9900846406026451, "3000–6000 Hz_40ms_modulation_mae_db": 0.3325401326963643, "mr_stft_db": 5.833810231062444, "band_mae_db": 0.19324810669196715, "rms_gain_error_db": -0.01338686534045403, "output_peak": 0.2538995911559821}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5321995958368282, "全帯域_10ms_rms_envelope_nrmse": 0.10150386945868557, "全帯域_10ms_envelope_correlation": 0.9910967765531269, "全帯域_10ms_modulation_mae_db": 0.6814436879986842, "全帯域_40ms_log_rms_mae_db": 0.865079298922193, "全帯域_40ms_rms_envelope_nrmse": 0.047728384730095746, "全帯域_40ms_envelope_correlation": 0.9981372497897437, "全帯域_40ms_modulation_mae_db": 0.5759726121129218, "300–1000 Hz_10ms_log_rms_mae_db": 1.8527983053334771, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13893439995634002, "300–1000 Hz_10ms_envelope_correlation": 0.9875083505910549, "300–1000 Hz_10ms_modulation_mae_db": 0.7469603295966897, "300–1000 Hz_40ms_log_rms_mae_db": 1.315574655947474, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07728576625151859, "300–1000 Hz_40ms_envelope_correlation": 0.9965342116534875, "300–1000 Hz_40ms_modulation_mae_db": 0.8988494805205026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7928441179985595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22505029324102707, "1000–3000 Hz_10ms_envelope_correlation": 0.9639380589585252, "1000–3000 Hz_10ms_modulation_mae_db": 0.9121064185926747, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1917609184547358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15699816485553603, "1000–3000 Hz_40ms_envelope_correlation": 0.9817135605542676, "1000–3000 Hz_40ms_modulation_mae_db": 1.055064278760627, "3000–6000 Hz_10ms_log_rms_mae_db": 1.602165062208614, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22605133436842134, "3000–6000 Hz_10ms_envelope_correlation": 0.969760590603065, "3000–6000 Hz_10ms_modulation_mae_db": 1.8805314094386096, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6162405716997343, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14383795183767836, "3000–6000 Hz_40ms_envelope_correlation": 0.9882746714594707, "3000–6000 Hz_40ms_modulation_mae_db": 1.5309900524127187, "mr_stft_db": 5.850071276130639, "band_mae_db": 0.23192227964478515, "rms_gain_error_db": -0.15042674917502444, "output_peak": 0.25121455364216816}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5435827852039226, "全帯域_10ms_rms_envelope_nrmse": 0.09834508947109846, "全帯域_10ms_envelope_correlation": 0.9916672432807326, "全帯域_10ms_modulation_mae_db": 0.45155707505291093, "全帯域_40ms_log_rms_mae_db": 0.8437485887278603, "全帯域_40ms_rms_envelope_nrmse": 0.05023132793716096, "全帯域_40ms_envelope_correlation": 0.9979383270726966, "全帯域_40ms_modulation_mae_db": 0.3317411552483143, "300–1000 Hz_10ms_log_rms_mae_db": 1.986927953926347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1282908831642503, "300–1000 Hz_10ms_envelope_correlation": 0.9896063201530669, "300–1000 Hz_10ms_modulation_mae_db": 0.5461157701153281, "300–1000 Hz_40ms_log_rms_mae_db": 1.39620750916132, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07781935011990081, "300–1000 Hz_40ms_envelope_correlation": 0.9966501094118274, "300–1000 Hz_40ms_modulation_mae_db": 0.7369920637872571, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7597730568921985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2046961980934635, "1000–3000 Hz_10ms_envelope_correlation": 0.9704291637964579, "1000–3000 Hz_10ms_modulation_mae_db": 0.5735457142233805, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1743249267013198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12118725508484013, "1000–3000 Hz_40ms_envelope_correlation": 0.9899213732256664, "1000–3000 Hz_40ms_modulation_mae_db": 0.857342687874235, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6654384476705824, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20294415671492191, "3000–6000 Hz_10ms_envelope_correlation": 0.9755753187497678, "3000–6000 Hz_10ms_modulation_mae_db": 0.8599723436852651, "3000–6000 Hz_40ms_log_rms_mae_db": 1.631713308901969, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1344573183893426, "3000–6000 Hz_40ms_envelope_correlation": 0.9889990216059702, "3000–6000 Hz_40ms_modulation_mae_db": 0.6730734321713752, "mr_stft_db": 5.81909920720102, "band_mae_db": 0.2504822649723267, "rms_gain_error_db": -0.16060180552932102, "output_peak": 0.24436522003900835}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5587016513760386, "全帯域_10ms_rms_envelope_nrmse": 0.1170868467509914, "全帯域_10ms_envelope_correlation": 0.9881617479593452, "全帯域_10ms_modulation_mae_db": 0.5108785990442327, "全帯域_40ms_log_rms_mae_db": 0.7717330955576786, "全帯域_40ms_rms_envelope_nrmse": 0.0693271916223851, "全帯域_40ms_envelope_correlation": 0.9956406154016328, "全帯域_40ms_modulation_mae_db": 0.572989626718479, "300–1000 Hz_10ms_log_rms_mae_db": 1.8836110432016497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1525800673611818, "300–1000 Hz_10ms_envelope_correlation": 0.9848223218706538, "300–1000 Hz_10ms_modulation_mae_db": 0.9947688231266266, "300–1000 Hz_40ms_log_rms_mae_db": 1.3097071936598583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10002862331140461, "300–1000 Hz_40ms_envelope_correlation": 0.9932858671101331, "300–1000 Hz_40ms_modulation_mae_db": 1.0775797162547054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8409805652655702, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21805701068976974, "1000–3000 Hz_10ms_envelope_correlation": 0.9667737667009222, "1000–3000 Hz_10ms_modulation_mae_db": 0.6238433731252211, "1000–3000 Hz_40ms_log_rms_mae_db": 1.227595456796037, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14274420867335644, "1000–3000 Hz_40ms_envelope_correlation": 0.9850900293789039, "1000–3000 Hz_40ms_modulation_mae_db": 0.6392288589524964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6675817820246432, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2281093886869445, "3000–6000 Hz_10ms_envelope_correlation": 0.9692358804398324, "3000–6000 Hz_10ms_modulation_mae_db": 1.098872251483711, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6174116493513937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14224768589557488, "3000–6000 Hz_40ms_envelope_correlation": 0.9877201439328605, "3000–6000 Hz_40ms_modulation_mae_db": 0.5901934161305523, "mr_stft_db": 5.839153526639645, "band_mae_db": 0.18861682574968672, "rms_gain_error_db": 0.047776372542730795, "output_peak": 0.3574956494569643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4889083315223506, "全帯域_10ms_rms_envelope_nrmse": 0.11050416505059897, "全帯域_10ms_envelope_correlation": 0.9893428736796704, "全帯域_10ms_modulation_mae_db": 0.24707962401164427, "全帯域_40ms_log_rms_mae_db": 0.7570324909955294, "全帯域_40ms_rms_envelope_nrmse": 0.056359811646106865, "全帯域_40ms_envelope_correlation": 0.9970712277206046, "全帯域_40ms_modulation_mae_db": 0.09989523415069305, "300–1000 Hz_10ms_log_rms_mae_db": 1.9648022831899403, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1414408701911186, "300–1000 Hz_10ms_envelope_correlation": 0.9867387719799806, "300–1000 Hz_10ms_modulation_mae_db": 0.6022126765515522, "300–1000 Hz_40ms_log_rms_mae_db": 1.362955995014044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07213021252234757, "300–1000 Hz_40ms_envelope_correlation": 0.996528829639577, "300–1000 Hz_40ms_modulation_mae_db": 0.4210370049791261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7927385684859676, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21771162738619987, "1000–3000 Hz_10ms_envelope_correlation": 0.9663655194860036, "1000–3000 Hz_10ms_modulation_mae_db": 0.6873977132424597, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1913398141618758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1411166674761718, "1000–3000 Hz_40ms_envelope_correlation": 0.9859044109232487, "1000–3000 Hz_40ms_modulation_mae_db": 0.9620631689411093, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5957471518295503, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18887793126576355, "3000–6000 Hz_10ms_envelope_correlation": 0.9788231348170825, "3000–6000 Hz_10ms_modulation_mae_db": 0.2216778995860644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5787237886047065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13147378090989034, "3000–6000 Hz_40ms_envelope_correlation": 0.9895217850926737, "3000–6000 Hz_40ms_modulation_mae_db": 0.05878101326998062, "mr_stft_db": 5.825791109534836, "band_mae_db": 0.23459733774759872, "rms_gain_error_db": 0.01915162686580903, "output_peak": 0.26773931582306426}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4603061520706866, "全帯域_10ms_rms_envelope_nrmse": 0.10615649262756247, "全帯域_10ms_envelope_correlation": 0.9901551960019009, "全帯域_10ms_modulation_mae_db": 0.4496284474868428, "全帯域_40ms_log_rms_mae_db": 0.7304111928375489, "全帯域_40ms_rms_envelope_nrmse": 0.05873123290311062, "全帯域_40ms_envelope_correlation": 0.9969437258880739, "全帯域_40ms_modulation_mae_db": 0.3176217479755332, "300–1000 Hz_10ms_log_rms_mae_db": 1.9118855705813438, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13966291064324368, "300–1000 Hz_10ms_envelope_correlation": 0.9875513037079848, "300–1000 Hz_10ms_modulation_mae_db": 0.7314724374007242, "300–1000 Hz_40ms_log_rms_mae_db": 1.3018528657591513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09037277639746621, "300–1000 Hz_40ms_envelope_correlation": 0.9952117799654026, "300–1000 Hz_40ms_modulation_mae_db": 0.6028866596844301, "1000–3000 Hz_10ms_log_rms_mae_db": 1.790230549555057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20167935128297101, "1000–3000 Hz_10ms_envelope_correlation": 0.9711731783893738, "1000–3000 Hz_10ms_modulation_mae_db": 1.1531950135579472, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1469683493768972, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12426816519290551, "1000–3000 Hz_40ms_envelope_correlation": 0.9889964546574135, "1000–3000 Hz_40ms_modulation_mae_db": 1.1399665968205834, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6093540749785264, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2021836228923237, "3000–6000 Hz_10ms_envelope_correlation": 0.9757661129277612, "3000–6000 Hz_10ms_modulation_mae_db": 0.778849480416667, "3000–6000 Hz_40ms_log_rms_mae_db": 1.577065294513228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13614915145515777, "3000–6000 Hz_40ms_envelope_correlation": 0.9888172281363063, "3000–6000 Hz_40ms_modulation_mae_db": 0.7339057531386288, "mr_stft_db": 5.838672281607213, "band_mae_db": 0.3126206711686041, "rms_gain_error_db": -0.12167631872037221, "output_peak": 0.24844003348517538}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.040952605461121, "spectral_convergence": 0.32043179931760607, "spectral_convergence_status": "ok", "amplitude_floor": 8.79928282883349e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.656071790185498, "spectral_convergence": 0.28152515132473915, "spectral_convergence_status": "ok", "amplitude_floor": 4.444740709272264e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.828295536132612, "spectral_convergence": 0.30540239027720334, "spectral_convergence_status": "ok", "amplitude_floor": 3.930299196146418e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002477612487613689, "output_power": 0.0002566172616976438, "delta_db": 0.15252485134311525, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004515434578601184, "output_power": 0.0004405402444811658, "delta_db": -0.1071396595428799, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015370997116088465, "output_power": 0.00013726987381618074, "delta_db": -0.4912680662082907, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.4065683577917706e-05, "output_power": 4.140744212717651e-05, "delta_db": -0.2702210804865641, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.662327944441642e-06, "output_power": 1.0398908620526029e-05, "delta_db": 0.7935314037138751, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.476431620775013e-05, "output_power": 2.243766585861002e-05, "delta_db": -0.4284866451851265, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3738619510799753, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1031278972296274, "pitch": null, "output_rms_dbfs": -30.415931425689543, "output_peak": 0.24227797964759473, "reference_rms_dbfs": -30.312803528459913, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8417733105930765}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.467120737686064e-05, "log_rms_mae_db": 1.5242717907123147, "rms_envelope_nrmse": 0.09835176618256397, "envelope_correlation": 0.9915464128120096, "rms_gain_error_db": -0.1031321375691385, "reference_mean_power": 0.0009312818437003785, "output_mean_power": 0.0009094271642185211}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.494393820063344e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023666657525761142, "output_power": 0.00023348250702447752, "reference_active": true, "delta_db": -0.05882578348031329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5441929621447868e-05, "output_power": 1.4865875921471635e-05, "reference_active": true, "delta_db": -0.16511065142069528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.985747798516416e-06, "output_power": 5.439961795519199e-06, "reference_active": true, "delta_db": -1.0861705287153982}], "active_band_count": 3, "active_band_mae_db": 0.436702321205469}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.334380857742655e-05, "log_rms_mae_db": 0.7663147809892167, "rms_envelope_nrmse": 0.05044570774115428, "envelope_correlation": 0.9977460018647478, "rms_gain_error_db": -0.10313334514699446, "reference_mean_power": 0.0009359486205868301, "output_mean_power": 0.0009139841702683143}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.189856446091556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021037225251172397, "output_power": 0.000209150123179046, "reference_active": true, "delta_db": -0.025303323853233178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.644252267790675e-06, "output_power": 7.324542907386669e-06, "reference_active": true, "delta_db": -0.1855448330448981}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4342900916585017e-07, "output_power": 3.0908617931476785e-07, "reference_active": true, "delta_db": -0.4575739070710258}], "active_band_count": 3, "active_band_mae_db": 0.22280735465638568}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.589555428061273e-05, "log_rms_mae_db": 1.8594841885226177, "rms_envelope_nrmse": 0.12891385635191366, "envelope_correlation": 0.9893971860660898, "rms_gain_error_db": -0.2771439859334708, "reference_mean_power": 0.00038327408961782034, "output_mean_power": 0.00035957961422901954}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.028222657916465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015396875485141256, "output_power": 0.00014233811262601208, "reference_active": true, "delta_db": -0.3411139476530439}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.75189127488202e-06, "output_power": 6.958297970981166e-06, "reference_active": true, "delta_db": -0.99598891424591}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1315883357233815e-06, "output_power": 2.969527329654786e-06, "reference_active": true, "delta_db": -1.4342971605266444}], "active_band_count": 3, "active_band_mae_db": 0.9238000074751994}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.149058869949865e-05, "log_rms_mae_db": 1.3013298880070945, "rms_envelope_nrmse": 0.08900908421937219, "envelope_correlation": 0.9951876762114152, "rms_gain_error_db": -0.2771498192832122, "reference_mean_power": 0.00038520090394369153, "output_mean_power": 0.0003613868251172741}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9052955697898775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014058163319658512, "output_power": 0.00013144689197416783, "reference_active": true, "delta_db": -0.29178262604038446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.901478078483036e-06, "output_power": 3.0116726057247306e-06, "reference_active": true, "delta_db": -1.1242141244402981}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3393802456709067e-07, "output_power": 1.691364846417365e-07, "reference_active": true, "delta_db": -1.408635183858058}], "active_band_count": 3, "active_band_mae_db": 0.9415439781129136}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.043858753489399e-05, "log_rms_mae_db": 1.7808563188168935, "rms_envelope_nrmse": 0.20939755915668334, "envelope_correlation": 0.9688454899362473, "rms_gain_error_db": -0.14979932306354093, "reference_mean_power": 4.084610004445957e-05, "output_mean_power": 3.94612342169598e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0558567323423266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3562756220605441e-05, "output_power": 1.22489439672258e-05, "reference_active": true, "delta_db": -0.4424930776998073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5216123885214767e-06, "output_power": 1.0758484177234747e-06, "reference_active": true, "delta_db": -1.5055295003403049}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602773162915538e-07, "output_power": 5.404919407027821e-07, "reference_active": true, "delta_db": -0.8693715486547781}], "active_band_count": 3, "active_band_mae_db": 0.9391313755649634}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.7804457921821952e-05, "log_rms_mae_db": 1.1625338188105363, "rms_envelope_nrmse": 0.12918098106434422, "envelope_correlation": 0.9877819906657738, "rms_gain_error_db": -0.1498384449243972, "reference_mean_power": 4.105094002568101e-05, "output_mean_power": 3.965877195147365e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8992373996258767e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2457002317192863e-05, "output_power": 1.1368087338983163e-05, "reference_active": true, "delta_db": -0.3972614369475463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.119435793981677e-07, "output_power": 5.652119196159984e-07, "reference_active": true, "delta_db": -1.5731453988895094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.742155092447501e-08, "output_power": 3.001328595095352e-08, "reference_active": true, "delta_db": -0.9580823675317198}], "active_band_count": 3, "active_band_mae_db": 0.9761630677895918}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.7924378350629476e-05, "log_rms_mae_db": 1.597098866452411, "rms_envelope_nrmse": 0.2546119796804401, "envelope_correlation": 0.9611816460405975, "rms_gain_error_db": -0.22017478330515344, "reference_mean_power": 2.620464364014996e-05, "output_mean_power": 2.490925735859464e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.347390510898905e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.954495563916905e-05, "output_power": 1.8894153463727294e-05, "reference_active": true, "delta_db": -0.1470725056037083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.79596482520383e-06, "output_power": 3.2072887786934417e-06, "reference_active": true, "delta_db": 0.596063613660731}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.14743402003256e-06, "output_power": 5.414206036735122e-07, "reference_active": true, "delta_db": -3.2619294328882242}], "active_band_count": 3, "active_band_mae_db": 1.3350218507175544}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.7050501746546627e-05, "log_rms_mae_db": 1.6339956525330357, "rms_envelope_nrmse": 0.16010363948070427, "envelope_correlation": 0.9844857278246187, "rms_gain_error_db": -0.22017662912186428, "reference_mean_power": 2.633641752064706e-05, "output_mean_power": 2.5034506559803945e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2809655135480154e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9754827323518705e-05, "output_power": 1.9148768420517155e-05, "reference_active": true, "delta_db": -0.13532390922650214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4198245590923022e-06, "output_power": 1.5864385557872038e-06, "reference_active": true, "delta_db": 0.48188572127498447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.51727649987637e-08, "output_power": 3.254761020771834e-08, "reference_active": true, "delta_db": -3.0154704040849185}], "active_band_count": 3, "active_band_mae_db": 1.210893344862135}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7204590496563738, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "26f849f4fba525f29d855a5248e50461467f3514cf7c59d85a62cb79d9a905d9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9683661243217692, "standard_deviation": 0.0024549136662976537, "minimum": 0.9639380589585252, "maximum": 0.9711731783893738, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7854993362175497, "standard_deviation": 0.026477456711600435, "minimum": 1.759399806066103, "maximum": 1.8409805652655702, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8287948470677503, "standard_deviation": 0.2059902090405445, "minimum": 0.5735457142233805, "maximum": 1.1531950135579472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21123573264291143, "standard_deviation": 0.008111859310694441, "minimum": 0.20167935128297101, "maximum": 0.22505029324102707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9869059373770854, "standard_deviation": 0.002606455281084172, "minimum": 0.9817135605542676, "maximum": 0.9899213732256664, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1900687622074573, "standard_deviation": 0.03193875954215471, "minimum": 1.1469683493768972, "maximum": 1.2431070794752248, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9575715311390206, "standard_deviation": 0.19949582753311654, "minimum": 0.6392288589524964, "maximum": 1.2576339067522038, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13400221334452628, "standard_deviation": 0.011976137992396994, "minimum": 0.12118725508484013, "maximum": 0.15699816485553603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9735025442167735, "standard_deviation": 0.006315736313837976, "minimum": 0.9611816460405975, "maximum": 0.9793524913731626, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6217006552666684, "standard_deviation": 0.028933217008572275, "minimum": 1.5957471518295503, "maximum": 1.6675817820246432, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9196174847499498, "standard_deviation": 0.5226117355162655, "minimum": 0.2216778995860644, "maximum": 1.8805314094386096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2108758868311897, "standard_deviation": 0.02320070103227289, "minimum": 0.18887793126576355, "maximum": 0.2546119796804401, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9886276128932083, "standard_deviation": 0.001978459288525731, "minimum": 0.9844857278246187, "maximum": 0.9911176844911214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6016546534640488, "standard_deviation": 0.025694371421582634, "minimum": 1.5736620886109849, "maximum": 1.6339956525330357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7322508453679674, "standard_deviation": 0.46354528298706565, "minimum": 0.05878101326998062, "maximum": 1.5309900524127187, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13807707861123703, "standard_deviation": 0.010659340302938657, "minimum": 0.1269453620999099, "maximum": 0.16010363948070427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.987932850402024, "standard_deviation": 0.0017151597635006405, "minimum": 0.9848223218706538, "maximum": 0.9899857118252436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9019817920673054, "standard_deviation": 0.06483861775617496, "minimum": 1.7992052119620525, "maximum": 1.986927953926347, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6574715186178068, "standard_deviation": 0.24933866269012744, "minimum": 0.21360814422356297, "maximum": 0.9947688231266266, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1361313905117284, "standard_deviation": 0.009325634127862388, "minimum": 0.12286215042792112, "maximum": 0.1525800673611818, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99577688813755, "standard_deviation": 0.0012269060569255281, "minimum": 0.9932858671101331, "maximum": 0.9971034824835758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3304618093905152, "standard_deviation": 0.052056222072498846, "minimum": 1.2530023091360658, "maximum": 1.403064058439114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.70796954518101, "standard_deviation": 0.24361826092658811, "minimum": 0.4210370049791261, "maximum": 1.0775797162547054, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08188750396718988, "standard_deviation": 0.010969233512509622, "minimum": 0.0658557973667674, "maximum": 0.10002862331140461, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25051147607085644, "standard_deviation": 0.06300910126070548, "minimum": 0.18861682574968672, "maximum": 0.3738619510799753, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.836386306450313, "standard_deviation": 0.009906916874764788, "minimum": 5.81909920720102, "maximum": 5.850071276130639, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.26517982976341004, "standard_deviation": 0.03812226850839255, "minimum": 0.24227797964759473, "maximum": 0.3574956494569643, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06247803134622761, "standard_deviation": 0.08086105245816096, "minimum": -0.16060180552932102, "maximum": 0.047776372542730795, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.990403667743236, "standard_deviation": 0.001177166395838654, "minimum": 0.9881617479593452, "maximum": 0.9916672432807326, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5128899836743481, "standard_deviation": 0.038334531068846246, "minimum": 1.4592713842425362, "maximum": 1.5587016513760386, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41529131245098366, "standard_deviation": 0.14773952216333777, "minimum": 0.24707962401164427, "maximum": 0.6814436879986842, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10483456665013256, "standard_deviation": 0.006365220865958366, "minimum": 0.09834508947109846, "maximum": 0.1170868467509914, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972487163274224, "standard_deviation": 0.000785872913238852, "minimum": 0.9956406154016328, "maximum": 0.9981372497897437, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7811415413062998, "standard_deviation": 0.04997403204119903, "minimum": 0.7263680664133693, "maximum": 0.865079298922193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3516127371920126, "standard_deviation": 0.16116794279240884, "minimum": 0.09989523415069305, "maximum": 0.5759726121129218, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0552468257218722, "standard_deviation": 0.006817555530202232, "minimum": 0.047728384730095746, "maximum": 0.0693271916223851, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0003", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.769108533859253, "predicted_mos": 2.769108533859253, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "dab1e62161c67eb5c18d93d92dfbc5cb4538c9be3b00bba377458f12fd88db1f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0150036001577973, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5620323032513272, "全帯域_10ms_rms_envelope_nrmse": 0.10705278586744102, "全帯域_10ms_envelope_correlation": 0.9899356375166593, "全帯域_10ms_modulation_mae_db": 0.4882567042479464, "全帯域_40ms_log_rms_mae_db": 0.8193897588229875, "全帯域_40ms_rms_envelope_nrmse": 0.05706060869252445, "全帯域_40ms_envelope_correlation": 0.9969888636146098, "全帯域_40ms_modulation_mae_db": 0.8207548247173942, "300–1000 Hz_10ms_log_rms_mae_db": 1.8035731046285146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14336653793399437, "300–1000 Hz_10ms_envelope_correlation": 0.9863355223665482, "300–1000 Hz_10ms_modulation_mae_db": 0.32084022736510837, "300–1000 Hz_40ms_log_rms_mae_db": 1.229332436050827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08869417959215793, "300–1000 Hz_40ms_envelope_correlation": 0.9947133086346545, "300–1000 Hz_40ms_modulation_mae_db": 0.4008320566996304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7077580562087515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20874110051174247, "1000–3000 Hz_10ms_envelope_correlation": 0.9690220048010807, "1000–3000 Hz_10ms_modulation_mae_db": 0.22761016689922398, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0325283410832036, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13173607794885128, "1000–3000 Hz_40ms_envelope_correlation": 0.9872514148462764, "1000–3000 Hz_40ms_modulation_mae_db": 0.33900897795031915, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4909827357160605, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21753962737762653, "3000–6000 Hz_10ms_envelope_correlation": 0.9767036926531907, "3000–6000 Hz_10ms_modulation_mae_db": 2.13681225808735, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4325094008793346, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15605839530648666, "3000–6000 Hz_40ms_envelope_correlation": 0.9889093884133553, "3000–6000 Hz_40ms_modulation_mae_db": 2.632851182615582, "mr_stft_db": 5.787224890408109, "band_mae_db": 0.25925078578707555, "rms_gain_error_db": -0.04006036385618553, "output_peak": 0.25568011050609685}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5408348261973093, "全帯域_10ms_rms_envelope_nrmse": 0.11464094956185912, "全帯域_10ms_envelope_correlation": 0.9884385885730572, "全帯域_10ms_modulation_mae_db": 0.38009884930607635, "全帯域_40ms_log_rms_mae_db": 0.7990725003100412, "全帯域_40ms_rms_envelope_nrmse": 0.052467943050473705, "全帯域_40ms_envelope_correlation": 0.9974963901092246, "全帯域_40ms_modulation_mae_db": 0.5682721690702937, "300–1000 Hz_10ms_log_rms_mae_db": 1.9123057559027525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14278625080932644, "300–1000 Hz_10ms_envelope_correlation": 0.986671854673768, "300–1000 Hz_10ms_modulation_mae_db": 0.44679813699216225, "300–1000 Hz_40ms_log_rms_mae_db": 1.26609748653761, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0751691954617758, "300–1000 Hz_40ms_envelope_correlation": 0.9966595731320796, "300–1000 Hz_40ms_modulation_mae_db": 0.410378519580966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7316353987372046, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20103922666880802, "1000–3000 Hz_10ms_envelope_correlation": 0.9713426818604914, "1000–3000 Hz_10ms_modulation_mae_db": 0.4627140792423376, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0973290611903994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12939330095190904, "1000–3000 Hz_40ms_envelope_correlation": 0.9876130343246496, "1000–3000 Hz_40ms_modulation_mae_db": 0.6120026103316377, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5181314849356253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21439037375207304, "3000–6000 Hz_10ms_envelope_correlation": 0.9728873823857899, "3000–6000 Hz_10ms_modulation_mae_db": 2.0356668291495863, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4395616494263828, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1598817678720887, "3000–6000 Hz_40ms_envelope_correlation": 0.9850913371764373, "3000–6000 Hz_40ms_modulation_mae_db": 1.9839209914123421, "mr_stft_db": 5.791006810718467, "band_mae_db": 0.30242183482280066, "rms_gain_error_db": -0.06337651888932835, "output_peak": 0.24544597908902213}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5185261040210651, "全帯域_10ms_rms_envelope_nrmse": 0.10951855715853458, "全帯域_10ms_envelope_correlation": 0.9894921455941847, "全帯域_10ms_modulation_mae_db": 0.08239370620626822, "全帯域_40ms_log_rms_mae_db": 0.8051021024448825, "全帯域_40ms_rms_envelope_nrmse": 0.051562330396199535, "全帯域_40ms_envelope_correlation": 0.9975386571623509, "全帯域_40ms_modulation_mae_db": 0.31017140771776636, "300–1000 Hz_10ms_log_rms_mae_db": 1.8521486672535146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1396825524469865, "300–1000 Hz_10ms_envelope_correlation": 0.9870460508466692, "300–1000 Hz_10ms_modulation_mae_db": 0.16258147189613495, "300–1000 Hz_40ms_log_rms_mae_db": 1.2253893202638613, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07296166950942862, "300–1000 Hz_40ms_envelope_correlation": 0.9963609491628325, "300–1000 Hz_40ms_modulation_mae_db": 0.23934417995332036, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6935947374871512, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2084420198967166, "1000–3000 Hz_10ms_envelope_correlation": 0.969135983914243, "1000–3000 Hz_10ms_modulation_mae_db": 0.8844496472392391, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0443171133547504, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13434807562140577, "1000–3000 Hz_40ms_envelope_correlation": 0.9869682242427028, "1000–3000 Hz_40ms_modulation_mae_db": 0.9852358367097231, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5342410491908296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20710589681643776, "3000–6000 Hz_10ms_envelope_correlation": 0.9750716205772563, "3000–6000 Hz_10ms_modulation_mae_db": 1.4470287472474632, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4541034919541187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14523872189980877, "3000–6000 Hz_40ms_envelope_correlation": 0.9883415739345726, "3000–6000 Hz_40ms_modulation_mae_db": 1.1779887801609592, "mr_stft_db": 5.778974961499341, "band_mae_db": 0.21882956271468448, "rms_gain_error_db": -0.008128955671538287, "output_peak": 0.32973573243575055}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.433047723310298, "全帯域_10ms_rms_envelope_nrmse": 0.10508324074589129, "全帯域_10ms_envelope_correlation": 0.9903111696859714, "全帯域_10ms_modulation_mae_db": 0.5935481418765362, "全帯域_40ms_log_rms_mae_db": 0.7424271303799976, "全帯域_40ms_rms_envelope_nrmse": 0.049903730186247025, "全帯域_40ms_envelope_correlation": 0.9977692498403167, "全帯域_40ms_modulation_mae_db": 0.34917649973174353, "300–1000 Hz_10ms_log_rms_mae_db": 1.7882764235411304, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1533800512563417, "300–1000 Hz_10ms_envelope_correlation": 0.984419459147881, "300–1000 Hz_10ms_modulation_mae_db": 0.6950097022303421, "300–1000 Hz_40ms_log_rms_mae_db": 1.1754568476605198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0852279630915942, "300–1000 Hz_40ms_envelope_correlation": 0.9953639744502994, "300–1000 Hz_40ms_modulation_mae_db": 0.7837359341285483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6805258016199056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20825175125394138, "1000–3000 Hz_10ms_envelope_correlation": 0.969153980917628, "1000–3000 Hz_10ms_modulation_mae_db": 0.7157608739025391, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0304734552359645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1276248480992024, "1000–3000 Hz_40ms_envelope_correlation": 0.9880611170899549, "1000–3000 Hz_40ms_modulation_mae_db": 0.7672673528349918, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5274582776986665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2416438036255834, "3000–6000 Hz_10ms_envelope_correlation": 0.9657242133816338, "3000–6000 Hz_10ms_modulation_mae_db": 0.8812647531319379, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4346645397300473, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14275492024227285, "3000–6000 Hz_40ms_envelope_correlation": 0.9877139627700136, "3000–6000 Hz_40ms_modulation_mae_db": 0.4083495578202867, "mr_stft_db": 5.809003065181848, "band_mae_db": 0.19301411820210723, "rms_gain_error_db": -0.08590938608778026, "output_peak": 0.27810611666135415}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5123526833875052, "全帯域_10ms_rms_envelope_nrmse": 0.10656973687829957, "全帯域_10ms_envelope_correlation": 0.9900497426775241, "全帯域_10ms_modulation_mae_db": 0.26888163759076167, "全帯域_40ms_log_rms_mae_db": 0.8163975594974163, "全帯域_40ms_rms_envelope_nrmse": 0.048096142826183894, "全帯域_40ms_envelope_correlation": 0.9978876776429277, "全帯域_40ms_modulation_mae_db": 0.5584901263954481, "300–1000 Hz_10ms_log_rms_mae_db": 1.814248125110612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13434075652706615, "300–1000 Hz_10ms_envelope_correlation": 0.9879817076262123, "300–1000 Hz_10ms_modulation_mae_db": 0.5707236291939745, "300–1000 Hz_40ms_log_rms_mae_db": 1.2337706464812515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06482470595579969, "300–1000 Hz_40ms_envelope_correlation": 0.9971598361876234, "300–1000 Hz_40ms_modulation_mae_db": 0.5037063528363357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7190412586253958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21993714813239457, "1000–3000 Hz_10ms_envelope_correlation": 0.9657656531405897, "1000–3000 Hz_10ms_modulation_mae_db": 0.8270718636437765, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0715439816640184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1395428808494053, "1000–3000 Hz_40ms_envelope_correlation": 0.9856147459971722, "1000–3000 Hz_40ms_modulation_mae_db": 0.7894481990872123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.566053836127211, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21813345229274222, "3000–6000 Hz_10ms_envelope_correlation": 0.9719424904202708, "3000–6000 Hz_10ms_modulation_mae_db": 1.2372425553680508, "3000–6000 Hz_40ms_log_rms_mae_db": 1.466062590037463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1369624771972049, "3000–6000 Hz_40ms_envelope_correlation": 0.988610730231545, "3000–6000 Hz_40ms_modulation_mae_db": 1.7226845979211927, "mr_stft_db": 5.7774154016928705, "band_mae_db": 0.19390395370512348, "rms_gain_error_db": -0.05172229575914086, "output_peak": 0.24929828229961756}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6012788117430454, "全帯域_10ms_rms_envelope_nrmse": 0.10003986959347462, "全帯域_10ms_envelope_correlation": 0.9912443376505544, "全帯域_10ms_modulation_mae_db": 0.23233205806194937, "全帯域_40ms_log_rms_mae_db": 0.8397850502892155, "全帯域_40ms_rms_envelope_nrmse": 0.04218639996130088, "全帯域_40ms_envelope_correlation": 0.9984495482660388, "全帯域_40ms_modulation_mae_db": 0.47643969943954895, "300–1000 Hz_10ms_log_rms_mae_db": 1.809796889653694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12884201280294072, "300–1000 Hz_10ms_envelope_correlation": 0.9891460708633961, "300–1000 Hz_10ms_modulation_mae_db": 0.6872599281465278, "300–1000 Hz_40ms_log_rms_mae_db": 1.1686295146610242, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060808396499412116, "300–1000 Hz_40ms_envelope_correlation": 0.9978364756799949, "300–1000 Hz_40ms_modulation_mae_db": 0.5024465605440595, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6641495869394636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1852715096757162, "1000–3000 Hz_10ms_envelope_correlation": 0.975682444331377, "1000–3000 Hz_10ms_modulation_mae_db": 0.6343652761986079, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0709204127759473, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10879127342702948, "1000–3000 Hz_40ms_envelope_correlation": 0.9914737957072166, "1000–3000 Hz_40ms_modulation_mae_db": 0.6763215573735936, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4881701437562147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22491687333074722, "3000–6000 Hz_10ms_envelope_correlation": 0.9709583092637661, "3000–6000 Hz_10ms_modulation_mae_db": 1.8339644225275051, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4295371408817823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14748272392582348, "3000–6000 Hz_40ms_envelope_correlation": 0.9872150040815016, "3000–6000 Hz_40ms_modulation_mae_db": 2.251787822700768, "mr_stft_db": 5.770986832069174, "band_mae_db": 0.1854586831893513, "rms_gain_error_db": -0.09728409356641694, "output_peak": 0.2760191822162912}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5014230228837449, "全帯域_10ms_rms_envelope_nrmse": 0.09517252988560418, "全帯域_10ms_envelope_correlation": 0.992072002792582, "全帯域_10ms_modulation_mae_db": 0.10883185729456431, "全帯域_40ms_log_rms_mae_db": 0.7905695611300567, "全帯域_40ms_rms_envelope_nrmse": 0.039645894327326486, "全帯域_40ms_envelope_correlation": 0.9986175849032612, "全帯域_40ms_modulation_mae_db": 0.39592903410126007, "300–1000 Hz_10ms_log_rms_mae_db": 1.8639789140693745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13654872305294738, "300–1000 Hz_10ms_envelope_correlation": 0.9876323573001952, "300–1000 Hz_10ms_modulation_mae_db": 0.44294229651530875, "300–1000 Hz_40ms_log_rms_mae_db": 1.2481167560902788, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06328372286019472, "300–1000 Hz_40ms_envelope_correlation": 0.9974454267736157, "300–1000 Hz_40ms_modulation_mae_db": 0.6068344004750689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6967876294578956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1920609431230442, "1000–3000 Hz_10ms_envelope_correlation": 0.9738278483845013, "1000–3000 Hz_10ms_modulation_mae_db": 0.7874746176037356, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0406962557188812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11855785912712087, "1000–3000 Hz_40ms_envelope_correlation": 0.9897430576593497, "1000–3000 Hz_40ms_modulation_mae_db": 1.1752330266416988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5426859277145983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19159585803283904, "3000–6000 Hz_10ms_envelope_correlation": 0.9782031521578544, "3000–6000 Hz_10ms_modulation_mae_db": 0.8181355341645236, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4718943304279772, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13280659912758827, "3000–6000 Hz_40ms_envelope_correlation": 0.9893843843021134, "3000–6000 Hz_40ms_modulation_mae_db": 0.7550219854488537, "mr_stft_db": 5.789099915764578, "band_mae_db": 0.2341037370945024, "rms_gain_error_db": -0.08438016412006331, "output_peak": 0.26694414341498096}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5495935489637125, "全帯域_10ms_rms_envelope_nrmse": 0.11360971379335333, "全帯域_10ms_envelope_correlation": 0.9886864236100551, "全帯域_10ms_modulation_mae_db": 0.2633778957646586, "全帯域_40ms_log_rms_mae_db": 0.804573462530384, "全帯域_40ms_rms_envelope_nrmse": 0.049679619850462466, "全帯域_40ms_envelope_correlation": 0.9977666183384797, "全帯域_40ms_modulation_mae_db": 0.3011863666823444, "300–1000 Hz_10ms_log_rms_mae_db": 1.7399247196579963, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14357091183549717, "300–1000 Hz_10ms_envelope_correlation": 0.9863428716660407, "300–1000 Hz_10ms_modulation_mae_db": 0.5250172942161481, "300–1000 Hz_40ms_log_rms_mae_db": 1.0591160062627358, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07201184694777592, "300–1000 Hz_40ms_envelope_correlation": 0.9967044638220467, "300–1000 Hz_40ms_modulation_mae_db": 0.4278681954027142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6776683572039879, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1972723664696944, "1000–3000 Hz_10ms_envelope_correlation": 0.9723868197461554, "1000–3000 Hz_10ms_modulation_mae_db": 0.7515184346797387, "1000–3000 Hz_40ms_log_rms_mae_db": 1.057064608469234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13236409371591168, "1000–3000 Hz_40ms_envelope_correlation": 0.9872255816516926, "1000–3000 Hz_40ms_modulation_mae_db": 1.0238319351729424, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5529074941720316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2242565946804252, "3000–6000 Hz_10ms_envelope_correlation": 0.9700589187615474, "3000–6000 Hz_10ms_modulation_mae_db": 0.6298246237771751, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4966093020288502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15794819542124433, "3000–6000 Hz_40ms_envelope_correlation": 0.9848180301209889, "3000–6000 Hz_40ms_modulation_mae_db": 0.7980090692127861, "mr_stft_db": 5.783482935197081, "band_mae_db": 0.263472395982544, "rms_gain_error_db": -0.07401353695446611, "output_peak": 0.25694506060265704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.998866809293899, "spectral_convergence": 0.3121531016405122, "spectral_convergence_status": "ok", "amplitude_floor": 8.79928282883349e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.599812056737388, "spectral_convergence": 0.28893354794471715, "spectral_convergence_status": "ok", "amplitude_floor": 4.444740709272264e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7629958051930394, "spectral_convergence": 0.3125786287451029, "spectral_convergence_status": "ok", "amplitude_floor": 3.930299196146418e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002477612487613689, "output_power": 0.00025524109829067596, "delta_db": 0.12917223426757898, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004515434578601184, "output_power": 0.0004373019985263257, "delta_db": -0.1391809229837746, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015370997116088465, "output_power": 0.0001495067659779031, "delta_db": -0.12041193837723617, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.4065683577917706e-05, "output_power": 4.233003834011938e-05, "delta_db": -0.17451849759051227, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.662327944441642e-06, "output_power": 9.651111412264713e-06, "delta_db": 0.4694270761072864, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.476431620775013e-05, "output_power": 2.7932238735911068e-05, "delta_db": 0.5227940453960651, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25925078578707555, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04006036385618553, "pitch": null, "output_rms_dbfs": -30.3528638923161, "output_peak": 0.25568011050609685, "reference_rms_dbfs": -30.312803528459913, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.787224890408109}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.467120737686064e-05, "log_rms_mae_db": 1.5620323032513272, "rms_envelope_nrmse": 0.10705278586744102, "envelope_correlation": 0.9899356375166593, "rms_gain_error_db": -0.04005852631040026, "reference_mean_power": 0.0009312818437003785, "output_mean_power": 0.0009227313654582967}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.494393820063344e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023666657525761142, "output_power": 0.00022787138046126708, "reference_active": true, "delta_db": -0.16447142931842493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5441929621447868e-05, "output_power": 1.8024996573855864e-05, "reference_active": true, "delta_db": 0.6717362207114319}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.985747798516416e-06, "output_power": 8.073635693053768e-06, "reference_active": true, "delta_db": 0.6285624627139823}], "active_band_count": 3, "active_band_mae_db": 0.4882567042479464}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.334380857742655e-05, "log_rms_mae_db": 0.8193897588229875, "rms_envelope_nrmse": 0.05706060869252445, "envelope_correlation": 0.9969888636146098, "rms_gain_error_db": -0.04002331120352229, "reference_mean_power": 0.0009359486205868301, "output_mean_power": 0.0009273628143269934}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.189856446091556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021037225251172397, "output_power": 0.0002045428328540636, "reference_active": true, "delta_db": -0.12204190583441175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.644252267790675e-06, "output_power": 9.319735069307318e-06, "reference_active": true, "delta_db": 0.8606855614467674}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4342900916585017e-07, "output_power": 4.82826032138573e-07, "reference_active": true, "delta_db": 1.4795370068710032}], "active_band_count": 3, "active_band_mae_db": 0.8207548247173942}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.589555428061273e-05, "log_rms_mae_db": 1.8035731046285146, "rms_envelope_nrmse": 0.14336653793399437, "envelope_correlation": 0.9863355223665482, "rms_gain_error_db": -0.1352235944479081, "reference_mean_power": 0.00038327408961782034, "output_mean_power": 0.00037152419453389033}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.028222657916465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015396875485141256, "output_power": 0.00013995759556508923, "reference_active": true, "delta_db": -0.4143612482994826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.75189127488202e-06, "output_power": 9.098743304240515e-06, "reference_active": true, "delta_db": 0.16879499027105396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1315883357233815e-06, "output_power": 3.785999901282643e-06, "reference_active": true, "delta_db": -0.37936444352478865}], "active_band_count": 3, "active_band_mae_db": 0.32084022736510837}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.149058869949865e-05, "log_rms_mae_db": 1.229332436050827, "rms_envelope_nrmse": 0.08869417959215793, "envelope_correlation": 0.9947133086346545, "rms_gain_error_db": -0.13522761584648624, "reference_mean_power": 0.00038520090394369153, "output_mean_power": 0.000373391593463856}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9052955697898775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014058163319658512, "output_power": 0.00012897426863108544, "reference_active": true, "delta_db": -0.3742551057646988}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.901478078483036e-06, "output_power": 4.326576396649497e-06, "reference_active": true, "delta_db": 0.4491520562969211}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3393802456709067e-07, "output_power": 2.1438377673462628e-07, "reference_active": true, "delta_db": -0.3790890080372713}], "active_band_count": 3, "active_band_mae_db": 0.4008320566996304}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.043858753489399e-05, "log_rms_mae_db": 1.7077580562087515, "rms_envelope_nrmse": 0.20874110051174247, "envelope_correlation": 0.9690220048010807, "rms_gain_error_db": -0.15111779940866749, "reference_mean_power": 4.084610004445957e-05, "output_mean_power": 3.9449255983497496e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0558567323423266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3562756220605441e-05, "output_power": 1.2318845634592779e-05, "reference_active": true, "delta_db": -0.4177794255345868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5216123885214767e-06, "output_power": 1.4490104406828487e-06, "reference_active": true, "delta_db": -0.2123252076312093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602773162915538e-07, "output_power": 6.683423235179566e-07, "reference_active": true, "delta_db": 0.05272586753187581}], "active_band_count": 3, "active_band_mae_db": 0.22761016689922398}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.7804457921821952e-05, "log_rms_mae_db": 1.0325283410832036, "rms_envelope_nrmse": 0.13173607794885128, "envelope_correlation": 0.9872514148462764, "rms_gain_error_db": -0.15113193615181336, "reference_mean_power": 4.105094002568101e-05, "output_mean_power": 3.964696184629532e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8992373996258767e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2457002317192863e-05, "output_power": 1.1553769370614797e-05, "reference_active": true, "delta_db": -0.32689851066816994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.119435793981677e-07, "output_power": 6.929835347813265e-07, "reference_active": true, "delta_db": -0.688029359507641}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.742155092447501e-08, "output_power": 3.740346843855541e-08, "reference_active": true, "delta_db": -0.002099063675146552}], "active_band_count": 3, "active_band_mae_db": 0.33900897795031915}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.7924378350629476e-05, "log_rms_mae_db": 1.4909827357160605, "rms_envelope_nrmse": 0.21753962737762653, "envelope_correlation": 0.9767036926531907, "rms_gain_error_db": 0.5759651102905397, "reference_mean_power": 2.620464364014996e-05, "output_mean_power": 2.9920908020958718e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.347390510898905e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.954495563916905e-05, "output_power": 2.2263896618822506e-05, "reference_active": true, "delta_db": 0.5656648764908172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.79596482520383e-06, "output_power": 4.557412425147737e-06, "reference_active": true, "delta_db": 2.1218662859978648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.14743402003256e-06, "output_power": 2.7040721995640563e-06, "reference_active": true, "delta_db": 3.7229056117733683}], "active_band_count": 3, "active_band_mae_db": 2.13681225808735}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.7050501746546627e-05, "log_rms_mae_db": 1.4325094008793346, "rms_envelope_nrmse": 0.15605839530648666, "envelope_correlation": 0.9889093884133553, "rms_gain_error_db": 0.5759646773847181, "reference_mean_power": 2.633641752064706e-05, "output_mean_power": 3.0071366682525693e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2809655135480154e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9754827323518705e-05, "output_power": 2.386114551947856e-05, "reference_active": true, "delta_db": 0.8201805145169228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4198245590923022e-06, "output_power": 2.545365474123935e-06, "reference_active": true, "delta_db": 2.5351546493851758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.51727649987637e-08, "output_power": 1.855188250098477e-07, "reference_active": true, "delta_db": 4.543218383944647}], "active_band_count": 3, "active_band_mae_db": 2.632851182615582}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.71106311285335, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "03ce9140ee76e61869c2e66c45de7dcee015dcb49269efa4f1c2908e9f47453e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9707896771370084, "standard_deviation": 0.003157802583317334, "minimum": 0.9657656531405897, "maximum": 0.975682444331377, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6963951032849693, "standard_deviation": 0.022469467150895813, "minimum": 1.6641495869394636, "maximum": 1.7316353987372046, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6613706199261499, "standard_deviation": 0.21796137974279498, "minimum": 0.22761016689922398, "maximum": 0.8844496472392391, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20262700821650723, "standard_deviation": 0.010996905169955827, "minimum": 0.1852715096757162, "maximum": 0.21993714813239457, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9879938714398768, "standard_deviation": 0.0018194498878117387, "minimum": 0.9856147459971722, "maximum": 0.9914737957072166, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0556091536865497, "standard_deviation": 0.023131470140441635, "minimum": 1.0304734552359645, "maximum": 1.0973290611903994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7960436870127648, "standard_deviation": 0.2644191802487197, "minimum": 0.33900897795031915, "maximum": 1.1752330266416988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1277948012176045, "standard_deviation": 0.009757152173714383, "minimum": 0.10879127342702948, "maximum": 0.1395428808494053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9726937224501637, "standard_deviation": 0.00398600831535889, "minimum": 0.9657242133816338, "maximum": 0.9782031521578544, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5275788686639045, "standard_deviation": 0.027712973808165363, "minimum": 1.4881701437562147, "maximum": 1.566053836127211, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.377492465431699, "standard_deviation": 0.5807336911683094, "minimum": 0.6298246237771751, "maximum": 2.13681225808735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21744780998855928, "standard_deviation": 0.014504667250803314, "minimum": 0.19159585803283904, "maximum": 0.2416438036255834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.987510551378816, "standard_deviation": 0.001715474917569092, "minimum": 0.9848180301209889, "maximum": 0.9893843843021134, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4531178056707446, "standard_deviation": 0.02365723583995821, "minimum": 1.4295371408817823, "maximum": 1.4966093020288502, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4663267484115963, "standard_deviation": 0.7987855968460302, "minimum": 0.4083495578202867, "maximum": 2.632851182615582, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14739172512406473, "standard_deviation": 0.009929682636735269, "minimum": 0.13280659912758827, "maximum": 0.1598817678720887, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9869469868113389, "standard_deviation": 0.0013933055188254852, "minimum": 0.984419459147881, "maximum": 0.9891460708633961, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8230315749771986, "standard_deviation": 0.05242678831224436, "minimum": 1.7399247196579963, "maximum": 1.9123057559027525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4813965858194633, "standard_deviation": 0.1804714351078108, "minimum": 0.16258147189613495, "maximum": 0.6950097022303421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14031472458313754, "standard_deviation": 0.00735450631708911, "minimum": 0.12884201280294072, "maximum": 0.1533800512563417, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965305009803933, "standard_deviation": 0.0010472661418080287, "minimum": 0.9947133086346545, "maximum": 0.9978364756799949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2007386267510136, "standard_deviation": 0.06628938151107308, "minimum": 1.0591160062627358, "maximum": 1.26609748653761, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48439327495258044, "standard_deviation": 0.16080226442366904, "minimum": 0.23934417995332036, "maximum": 0.7837359341285483, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07287270998976737, "standard_deviation": 0.010080112862047857, "minimum": 0.060808396499412116, "maximum": 0.08869417959215793, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23130688393727364, "standard_deviation": 0.041410310305307, "minimum": 0.1854586831893513, "maximum": 0.30242183482280066, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.785899351566433, "standard_deviation": 0.01146432525772642, "minimum": 5.770986832069174, "maximum": 5.809003065181848, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2697718259032213, "standard_deviation": 0.026955441804328066, "minimum": 0.24544597908902213, "maximum": 0.32973573243575055, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06310941436311496, "standard_deviation": 0.02913745619875391, "minimum": -0.09728409356641694, "maximum": -0.008128955671538287, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900287560125736, "standard_deviation": 0.0012165410336331592, "minimum": 0.9884385885730572, "maximum": 0.992072002792582, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.527386127969751, "standard_deviation": 0.049588081960800105, "minimum": 1.433047723310298, "maximum": 1.6012788117430454, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30221510629359516, "standard_deviation": 0.17678635598386058, "minimum": 0.08239370620626822, "maximum": 0.5935481418765362, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1064609229355572, "standard_deviation": 0.006527798441709927, "minimum": 0.09517252988560418, "maximum": 0.11464094956185912, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978143237346512, "standard_deviation": 0.0005232979519965685, "minimum": 0.9969888636146098, "maximum": 0.9986175849032612, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8021646406756227, "standard_deviation": 0.028389093316360463, "minimum": 0.7424271303799976, "maximum": 0.8397850502892155, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4725525159819749, "standard_deviation": 0.17510578993569761, "minimum": 0.3011863666823444, "maximum": 0.8207548247173942, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0488253336613398, "standard_deviation": 0.005597723000342818, "minimum": 0.039645894327326486, "maximum": 0.05706060869252445, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1188-133604-0031", "language": "en", "style": "normal", "speaker": "librispeech_1188", "source_utterance_id": "1188-133604-0031", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 68000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1188-133604-0031.wav", "source_file_sha256": "032da1d4dbc01ad18940bf0550a57975fd8fa01447b51693a41c50b7a2df53e5", "decoded_pcm_sha256": "b739cb4e3f2719b98dd98461d6859b487da84f7d1b9f40d280b7f0eb538a6458", "original_source_sha256": "08584f8b87cd859e9212f1730979ed141623f35e38995a88ba50747d04cdc222", "corpus_entry_id": "librispeech_1188-133604-0031"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.388342380523682, "predicted_mos": 4.388342380523682, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "78197fab1270c63f807e907dcfb4198a85410ce5134c8067a179838fa8706354", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.847176800016314, "cpu_threads": 4}, "waveform_sha256": "1e870b501de47bddfbee99af436e8dcdce2f2634d090d067334b6e9fd8bb0809"}, "rows": [{"case": "librispeech_1188-133604-0031", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.927166700363159, "predicted_mos": 2.927166700363159, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "379b54ae396628b06e86c8a8fba90e302d64a6188166a272cc8b4d04a6e51bb8", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.981976200360805, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6578626662390594, "全帯域_10ms_rms_envelope_nrmse": 0.06488238232145116, "全帯域_10ms_envelope_correlation": 0.9965994688996058, "全帯域_10ms_modulation_mae_db": 0.27902417312737915, "全帯域_40ms_log_rms_mae_db": 0.9858317468225857, "全帯域_40ms_rms_envelope_nrmse": 0.03920121494671944, "全帯域_40ms_envelope_correlation": 0.9988680717023187, "全帯域_40ms_modulation_mae_db": 0.9839228824991427, "300–1000 Hz_10ms_log_rms_mae_db": 1.9126425212271063, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09874528102581823, "300–1000 Hz_10ms_envelope_correlation": 0.993960821675097, "300–1000 Hz_10ms_modulation_mae_db": 1.2201496182917018, "300–1000 Hz_40ms_log_rms_mae_db": 1.3939504662548545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059184917727556165, "300–1000 Hz_40ms_envelope_correlation": 0.9982443108065707, "300–1000 Hz_40ms_modulation_mae_db": 1.3439910201051462, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9577411050799507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16207560571854168, "1000–3000 Hz_10ms_envelope_correlation": 0.9841187162636609, "1000–3000 Hz_10ms_modulation_mae_db": 0.9130188277404861, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3509443123748632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10134596431949233, "1000–3000 Hz_40ms_envelope_correlation": 0.9947163048978087, "1000–3000 Hz_40ms_modulation_mae_db": 1.021193260993626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.759159900260816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1665690063048975, "3000–6000 Hz_10ms_envelope_correlation": 0.986249995326335, "3000–6000 Hz_10ms_modulation_mae_db": 2.241329509442563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1970172028858432, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.108894235166483, "3000–6000 Hz_40ms_envelope_correlation": 0.9963329473587311, "3000–6000 Hz_40ms_modulation_mae_db": 2.650697250254778, "mr_stft_db": 5.752517342586425, "band_mae_db": 0.36435767582397527, "rms_gain_error_db": -0.14590630967127718, "output_peak": 0.3023451607946429}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6652940723874932, "全帯域_10ms_rms_envelope_nrmse": 0.05822912342607812, "全帯域_10ms_envelope_correlation": 0.9971456919956213, "全帯域_10ms_modulation_mae_db": 0.42663071624437193, "全帯域_40ms_log_rms_mae_db": 0.8699460646967292, "全帯域_40ms_rms_envelope_nrmse": 0.030052039596324274, "全帯域_40ms_envelope_correlation": 0.9992368656985637, "全帯域_40ms_modulation_mae_db": 0.2223637180642942, "300–1000 Hz_10ms_log_rms_mae_db": 1.9190803179948905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10484001785579936, "300–1000 Hz_10ms_envelope_correlation": 0.9926511760204199, "300–1000 Hz_10ms_modulation_mae_db": 0.5888933684078009, "300–1000 Hz_40ms_log_rms_mae_db": 1.3594871961052752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06091769757334301, "300–1000 Hz_40ms_envelope_correlation": 0.9975399376997484, "300–1000 Hz_40ms_modulation_mae_db": 0.6455989621627292, "1000–3000 Hz_10ms_log_rms_mae_db": 2.044991982549169, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1715138250256797, "1000–3000 Hz_10ms_envelope_correlation": 0.9820826543759454, "1000–3000 Hz_10ms_modulation_mae_db": 1.1958764996621138, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3995352281458049, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1093748031699136, "1000–3000 Hz_40ms_envelope_correlation": 0.9936467018784967, "1000–3000 Hz_40ms_modulation_mae_db": 0.9558717629962317, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7369150527431636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2252039447426566, "3000–6000 Hz_10ms_envelope_correlation": 0.9732325035376165, "3000–6000 Hz_10ms_modulation_mae_db": 2.0143010419972973, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1714475367920958, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14730184690741657, "3000–6000 Hz_40ms_envelope_correlation": 0.9922838602947311, "3000–6000 Hz_40ms_modulation_mae_db": 2.7228362684677205, "mr_stft_db": 5.7558405946830336, "band_mae_db": 0.3797368458231634, "rms_gain_error_db": -0.04267108170404987, "output_peak": 0.30617200050901844}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.650920324864282, "全帯域_10ms_rms_envelope_nrmse": 0.060966170240992164, "全帯域_10ms_envelope_correlation": 0.9968688840688061, "全帯域_10ms_modulation_mae_db": 0.8452820288948683, "全帯域_40ms_log_rms_mae_db": 1.026584507299513, "全帯域_40ms_rms_envelope_nrmse": 0.040707997402063774, "全帯域_40ms_envelope_correlation": 0.9985800212601473, "全帯域_40ms_modulation_mae_db": 0.6957466892763625, "300–1000 Hz_10ms_log_rms_mae_db": 1.9426414754326577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1229971511812328, "300–1000 Hz_10ms_envelope_correlation": 0.9902119837877542, "300–1000 Hz_10ms_modulation_mae_db": 2.0221929828099383, "300–1000 Hz_40ms_log_rms_mae_db": 1.3685039756062367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08510702992891343, "300–1000 Hz_40ms_envelope_correlation": 0.9955246865168538, "300–1000 Hz_40ms_modulation_mae_db": 2.2248429040516107, "1000–3000 Hz_10ms_log_rms_mae_db": 2.005361615718706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1566639297706424, "1000–3000 Hz_10ms_envelope_correlation": 0.985189438194772, "1000–3000 Hz_10ms_modulation_mae_db": 1.0167878960769865, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3117363417076586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09278847048828667, "1000–3000 Hz_40ms_envelope_correlation": 0.9957631427853494, "1000–3000 Hz_40ms_modulation_mae_db": 0.5097782282603359, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7750117547550737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16944972780674664, "3000–6000 Hz_10ms_envelope_correlation": 0.9856840988244779, "3000–6000 Hz_10ms_modulation_mae_db": 2.632414111954968, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1983103535237236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10945041309840153, "3000–6000 Hz_40ms_envelope_correlation": 0.9963355015442896, "3000–6000 Hz_40ms_modulation_mae_db": 3.4710755529939927, "mr_stft_db": 5.7740726119856545, "band_mae_db": 0.34538164938598287, "rms_gain_error_db": -0.06822546323942638, "output_peak": 0.29419105024383346}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.653769456715347, "全帯域_10ms_rms_envelope_nrmse": 0.05744757082143634, "全帯域_10ms_envelope_correlation": 0.9972224511522845, "全帯域_10ms_modulation_mae_db": 0.6406122451482669, "全帯域_40ms_log_rms_mae_db": 0.9159364533951073, "全帯域_40ms_rms_envelope_nrmse": 0.033889111774384986, "全帯域_40ms_envelope_correlation": 0.9990257147870177, "全帯域_40ms_modulation_mae_db": 0.3905291327542344, "300–1000 Hz_10ms_log_rms_mae_db": 1.9637964577451268, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09222645731077413, "300–1000 Hz_10ms_envelope_correlation": 0.9943302152820674, "300–1000 Hz_10ms_modulation_mae_db": 0.4020615459996026, "300–1000 Hz_40ms_log_rms_mae_db": 1.4362477089832208, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058913962331245885, "300–1000 Hz_40ms_envelope_correlation": 0.9976808952565771, "300–1000 Hz_40ms_modulation_mae_db": 0.566194810279664, "1000–3000 Hz_10ms_log_rms_mae_db": 2.083574874525938, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1676981890480817, "1000–3000 Hz_10ms_envelope_correlation": 0.983537323178592, "1000–3000 Hz_10ms_modulation_mae_db": 0.43664248880094725, "1000–3000 Hz_40ms_log_rms_mae_db": 1.458835592496426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11191733779420115, "1000–3000 Hz_40ms_envelope_correlation": 0.9939763507787769, "1000–3000 Hz_40ms_modulation_mae_db": 0.5279197430651171, "3000–6000 Hz_10ms_log_rms_mae_db": 1.744829326961204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17483052955818643, "3000–6000 Hz_10ms_envelope_correlation": 0.9841086810185506, "3000–6000 Hz_10ms_modulation_mae_db": 1.2637233085543484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1763547981726967, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11220564167504983, "3000–6000 Hz_40ms_envelope_correlation": 0.9955245385931983, "3000–6000 Hz_40ms_modulation_mae_db": 1.804332243031631, "mr_stft_db": 5.74564709549949, "band_mae_db": 0.33098341357284533, "rms_gain_error_db": -0.06409580511162548, "output_peak": 0.32123349373839705}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5973542539761036, "全帯域_10ms_rms_envelope_nrmse": 0.057474435608273255, "全帯域_10ms_envelope_correlation": 0.997190600691077, "全帯域_10ms_modulation_mae_db": 0.4128801523175496, "全帯域_40ms_log_rms_mae_db": 0.8677937968682261, "全帯域_40ms_rms_envelope_nrmse": 0.031673327050964296, "全帯域_40ms_envelope_correlation": 0.9991140793467611, "全帯域_40ms_modulation_mae_db": 0.2518854709817738, "300–1000 Hz_10ms_log_rms_mae_db": 1.9397606594076158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10593243581942499, "300–1000 Hz_10ms_envelope_correlation": 0.9924120772018503, "300–1000 Hz_10ms_modulation_mae_db": 0.25317478596115656, "300–1000 Hz_40ms_log_rms_mae_db": 1.3518148905894194, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06309629526083904, "300–1000 Hz_40ms_envelope_correlation": 0.9972090674433014, "300–1000 Hz_40ms_modulation_mae_db": 0.2729121486221728, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0291083215413708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16626022818389624, "1000–3000 Hz_10ms_envelope_correlation": 0.9838236518146325, "1000–3000 Hz_10ms_modulation_mae_db": 0.7667591324912791, "1000–3000 Hz_40ms_log_rms_mae_db": 1.415164705855288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11336159641869836, "1000–3000 Hz_40ms_envelope_correlation": 0.993685907992282, "1000–3000 Hz_40ms_modulation_mae_db": 0.7543403022480364, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7924442993971814, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20117594729491595, "3000–6000 Hz_10ms_envelope_correlation": 0.9764929765998894, "3000–6000 Hz_10ms_modulation_mae_db": 1.1102317421402494, "3000–6000 Hz_40ms_log_rms_mae_db": 1.230143645410553, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12191528412028704, "3000–6000 Hz_40ms_envelope_correlation": 0.9924583499640975, "3000–6000 Hz_40ms_modulation_mae_db": 1.7765599575859092, "mr_stft_db": 5.733591678972187, "band_mae_db": 0.3269240465248597, "rms_gain_error_db": -0.013329167109629874, "output_peak": 0.32095821818023573}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5601783795924073, "全帯域_10ms_rms_envelope_nrmse": 0.05733841992824329, "全帯域_10ms_envelope_correlation": 0.9972121250406846, "全帯域_10ms_modulation_mae_db": 0.8036014376982252, "全帯域_40ms_log_rms_mae_db": 0.90832241500417, "全帯域_40ms_rms_envelope_nrmse": 0.029014993709228918, "全帯域_40ms_envelope_correlation": 0.9992808617296056, "全帯域_40ms_modulation_mae_db": 0.6203077473703201, "300–1000 Hz_10ms_log_rms_mae_db": 1.9990181493822479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10097087686024073, "300–1000 Hz_10ms_envelope_correlation": 0.9931200257190936, "300–1000 Hz_10ms_modulation_mae_db": 0.6173181834312222, "300–1000 Hz_40ms_log_rms_mae_db": 1.4701426984315142, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06447327339768061, "300–1000 Hz_40ms_envelope_correlation": 0.997121427875777, "300–1000 Hz_40ms_modulation_mae_db": 0.49102206987424973, "1000–3000 Hz_10ms_log_rms_mae_db": 2.028252144795424, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1762451329489974, "1000–3000 Hz_10ms_envelope_correlation": 0.9817138875966643, "1000–3000 Hz_10ms_modulation_mae_db": 0.9547056115297111, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3970355734941127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10969037073425417, "1000–3000 Hz_40ms_envelope_correlation": 0.9945765610182131, "1000–3000 Hz_40ms_modulation_mae_db": 1.0255217882888694, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7603480188386846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16887058026635796, "3000–6000 Hz_10ms_envelope_correlation": 0.9840532249975809, "3000–6000 Hz_10ms_modulation_mae_db": 1.4461917919292597, "3000–6000 Hz_40ms_log_rms_mae_db": 1.157213506540643, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11320111201503692, "3000–6000 Hz_40ms_envelope_correlation": 0.9937014905386637, "3000–6000 Hz_40ms_modulation_mae_db": 1.8107525099111628, "mr_stft_db": 5.755171252556319, "band_mae_db": 0.2533296524481318, "rms_gain_error_db": -0.04564912914608009, "output_peak": 0.3148199573748226}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6014011039525047, "全帯域_10ms_rms_envelope_nrmse": 0.05800023199032753, "全帯域_10ms_envelope_correlation": 0.9971545355848024, "全帯域_10ms_modulation_mae_db": 0.4739315520603707, "全帯域_40ms_log_rms_mae_db": 0.8035953968273989, "全帯域_40ms_rms_envelope_nrmse": 0.02783855357289688, "全帯域_40ms_envelope_correlation": 0.9993442988302266, "全帯域_40ms_modulation_mae_db": 0.2693305908350106, "300–1000 Hz_10ms_log_rms_mae_db": 2.0555061585897216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10786494767213858, "300–1000 Hz_10ms_envelope_correlation": 0.9923483515820823, "300–1000 Hz_10ms_modulation_mae_db": 1.6675398531455174, "300–1000 Hz_40ms_log_rms_mae_db": 1.5649613862785425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06533632572046823, "300–1000 Hz_40ms_envelope_correlation": 0.9973333379470726, "300–1000 Hz_40ms_modulation_mae_db": 1.6772600838275356, "1000–3000 Hz_10ms_log_rms_mae_db": 2.089849888863089, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16434106576341115, "1000–3000 Hz_10ms_envelope_correlation": 0.9837165073444899, "1000–3000 Hz_10ms_modulation_mae_db": 1.0953460393417729, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4701413226754634, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10955780482279154, "1000–3000 Hz_40ms_envelope_correlation": 0.9936382562367388, "1000–3000 Hz_40ms_modulation_mae_db": 0.8565083319447457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7378921647776688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1588552696627403, "3000–6000 Hz_10ms_envelope_correlation": 0.9863825485273615, "3000–6000 Hz_10ms_modulation_mae_db": 1.193936042309821, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2082434495953218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09251325654187169, "3000–6000 Hz_40ms_envelope_correlation": 0.9968515028395922, "3000–6000 Hz_40ms_modulation_mae_db": 1.6381359442954775, "mr_stft_db": 5.75519229475383, "band_mae_db": 0.21540261328007385, "rms_gain_error_db": -0.05344539626596438, "output_peak": 0.30138862938882843}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6415878836827804, "全帯域_10ms_rms_envelope_nrmse": 0.07395319412332665, "全帯域_10ms_envelope_correlation": 0.9953405897228149, "全帯域_10ms_modulation_mae_db": 0.40609478575118496, "全帯域_40ms_log_rms_mae_db": 0.9135694381608773, "全帯域_40ms_rms_envelope_nrmse": 0.04231759273617438, "全帯域_40ms_envelope_correlation": 0.9984278478770704, "全帯域_40ms_modulation_mae_db": 0.06299560754218168, "300–1000 Hz_10ms_log_rms_mae_db": 2.1135920350134407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12827436970950262, "300–1000 Hz_10ms_envelope_correlation": 0.9891460801505034, "300–1000 Hz_10ms_modulation_mae_db": 1.7416045570851495, "300–1000 Hz_40ms_log_rms_mae_db": 1.5624294592174583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0780683448231503, "300–1000 Hz_40ms_envelope_correlation": 0.9961779183177428, "300–1000 Hz_40ms_modulation_mae_db": 1.815108459723666, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9802308396415205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13870115403590924, "1000–3000 Hz_10ms_envelope_correlation": 0.989179868147228, "1000–3000 Hz_10ms_modulation_mae_db": 0.7774654776926532, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3705098487197216, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09215013870600561, "1000–3000 Hz_40ms_envelope_correlation": 0.9963163305333467, "1000–3000 Hz_40ms_modulation_mae_db": 0.9961763314618626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8121515950485534, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2031551692698717, "3000–6000 Hz_10ms_envelope_correlation": 0.9750263187044044, "3000–6000 Hz_10ms_modulation_mae_db": 0.462523514963088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.238121190626205, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1093636081394314, "3000–6000 Hz_40ms_envelope_correlation": 0.9927383771525284, "3000–6000 Hz_40ms_modulation_mae_db": 0.8308794777219327, "mr_stft_db": 5.741852031445384, "band_mae_db": 0.2403401994390426, "rms_gain_error_db": -0.032950158724639196, "output_peak": 0.29865163408103007}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68000, "output_samples": 68000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0319521703818495, "spectral_convergence": 0.18367278771216927, "spectral_convergence_status": "ok", "amplitude_floor": 1.0380731662877042e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.575399775201739, "spectral_convergence": 0.18275459996671567, "spectral_convergence_status": "ok", "amplitude_floor": 5.867181533968612e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6502000821756875, "spectral_convergence": 0.17852912401811252, "spectral_convergence_status": "ok", "amplitude_floor": 4.6154090777751923e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006114723941341076, "output_power": 0.0006188552444448836, "delta_db": 0.05212220771398761, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011102815674838913, "output_power": 0.0010639899468516789, "delta_db": -0.18495605334719004, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00031219826577364904, "output_power": 0.00029142809870457456, "delta_db": -0.2989906333055922, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.006894391321554e-05, "output_power": 8.125897963341303e-05, "delta_db": -0.4470370593203147, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.882819827254369e-06, "output_power": 8.119363594907278e-06, "delta_db": -0.8535888778353685, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.595718179316582e-06, "output_power": 5.1631020774710735e-06, "delta_db": -0.34945122342139834, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.36435767582397527, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14590630967127718, "pitch": null, "output_rms_dbfs": -26.842783991536958, "output_peak": 0.3023451607946429, "reference_rms_dbfs": -26.69687768186568, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.752517342586425}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.0001318685794969069, "log_rms_mae_db": 1.6578626662390594, "rms_envelope_nrmse": 0.06488238232145116, "envelope_correlation": 0.9965994688996058, "rms_gain_error_db": -0.145928920246172, "reference_mean_power": 0.0021420152585098387, "output_mean_power": 0.0020712363921171926}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 1.3134511495070254e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003582591828573965, "output_power": 0.0003287191533430899, "reference_active": true, "delta_db": -0.3737232132131605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.463795601086152e-05, "output_power": 4.422395446126898e-05, "reference_active": true, "delta_db": -0.0404672572152381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.556588167851325e-06, "output_power": 6.855474962112994e-06, "reference_active": true, "delta_db": -0.4228820489537389}], "active_band_count": 3, "active_band_mae_db": 0.27902417312737915}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 0.00012588557284836115, "log_rms_mae_db": 0.9858317468225857, "rms_envelope_nrmse": 0.03920121494671944, "envelope_correlation": 0.9988680717023187, "rms_gain_error_db": -0.14590652427311235, "reference_mean_power": 0.0021571542863483557, "output_mean_power": 0.0020858859358491113}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 1.2592546923162062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031562831458090785, "output_power": 0.0002897954930451447, "reference_active": true, "delta_db": -0.37084329287219714}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9876750850691092e-05, "output_power": 1.88430031886078e-05, "reference_active": true, "delta_db": -0.23195272382927776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602361817361658e-07, "output_power": 3.8441653288985816e-07, "reference_active": true, "delta_db": -2.348972630795953}], "active_band_count": 3, "active_band_mae_db": 0.9839228824991427}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.00010444605382463925, "log_rms_mae_db": 1.9126425212271063, "rms_envelope_nrmse": 0.09874528102581823, "envelope_correlation": 0.993960821675097, "rms_gain_error_db": -0.28352813551985667, "reference_mean_power": 0.0005604503758559807, "output_mean_power": 0.0005250302800794047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 5.643164593552106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025020135526255573, "output_power": 0.00023004280416892223, "reference_active": true, "delta_db": -0.3648100489337429}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.833230380435073e-05, "output_power": 3.3899929071202155e-05, "reference_active": true, "delta_db": -0.5336613197977209}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.791716397073579e-06, "output_power": 3.59568628934248e-06, "reference_active": true, "delta_db": -2.7619774861436412}], "active_band_count": 3, "active_band_mae_db": 1.2201496182917018}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 9.32306023645191e-05, "log_rms_mae_db": 1.3939504662548545, "rms_envelope_nrmse": 0.059184917727556165, "envelope_correlation": 0.9982443108065707, "rms_gain_error_db": -0.2835313113515613, "reference_mean_power": 0.0005644391167624198, "output_mean_power": 0.000528766548512293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 5.412619397247975e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022898404327685652, "output_power": 0.00021138125751877544, "reference_active": true, "delta_db": -0.34738742425632546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5949246143041614e-05, "output_power": 1.4138445253828508e-05, "reference_active": true, "delta_db": -0.5233850599429003}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.408378605310095e-07, "output_power": 3.0947680472842667e-07, "reference_active": true, "delta_db": -3.1612005761162134}], "active_band_count": 3, "active_band_mae_db": 1.3439910201051462}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 4.754025004833877e-05, "log_rms_mae_db": 1.9577411050799507, "rms_envelope_nrmse": 0.16207560571854168, "envelope_correlation": 0.9841187162636609, "rms_gain_error_db": -0.4659134099913608, "reference_mean_power": 8.438545283737632e-05, "output_mean_power": 7.580122850843831e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 6.948180578959798e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5558932008941216e-05, "output_power": 2.895806857588547e-05, "reference_active": true, "delta_db": -0.8917911648664647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9367090695871985e-06, "output_power": 2.997601007605016e-06, "reference_active": true, "delta_db": -1.1835949470582685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.57982090332121e-07, "output_power": 8.831327315770549e-07, "reference_active": true, "delta_db": 0.6636703712967252}], "active_band_count": 3, "active_band_mae_db": 0.9130188277404861}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 4.260586736236329e-05, "log_rms_mae_db": 1.3509443123748632, "rms_envelope_nrmse": 0.10134596431949233, "envelope_correlation": 0.9947163048978087, "rms_gain_error_db": -0.46592731390612263, "reference_mean_power": 8.498589632737057e-05, "output_mean_power": 7.634034666865262e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 6.685624503092077e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2789374619144395e-05, "output_power": 2.7008062414315317e-05, "reference_active": true, "delta_db": -0.8423970501380128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.67867676799424e-06, "output_power": 1.258130310679967e-06, "reference_active": true, "delta_db": -1.252414546304009}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.924150120657105e-08, "output_power": 4.739682349993497e-08, "reference_active": true, "delta_db": -0.968768186538856}], "active_band_count": 3, "active_band_mae_db": 1.021193260993626}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 2.7803794970768208e-05, "log_rms_mae_db": 1.759159900260816, "rms_envelope_nrmse": 0.1665690063048975, "envelope_correlation": 0.986249995326335, "rms_gain_error_db": -0.7045899307673744, "reference_mean_power": 9.160796502339738e-06, "output_mean_power": 7.788866188045117e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 3.7361073674013695e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6569760918264083e-06, "output_power": 1.475264768121018e-06, "reference_active": true, "delta_db": -0.5044627128440061}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.026752210402781e-07, "output_power": 4.0011372581782515e-07, "reference_active": true, "delta_db": -1.7789988582433334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8917938491313036e-07, "output_power": 6.804890063887142e-08, "reference_active": true, "delta_db": -4.44052695724035}], "active_band_count": 3, "active_band_mae_db": 2.241329509442563}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 2.058581204447328e-05, "log_rms_mae_db": 1.1970172028858432, "rms_envelope_nrmse": 0.108894235166483, "envelope_correlation": 0.9963329473587311, "rms_gain_error_db": -0.7046179859598487, "reference_mean_power": 9.225978949789671e-06, "output_mean_power": 7.84423617155427e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 3.5247411287487305e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7104686599805127e-06, "output_power": 1.4577961176304287e-06, "reference_active": true, "delta_db": -0.6941833195324408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2748792658937344e-07, "output_power": 1.9714301147238339e-07, "reference_active": true, "delta_db": -2.2041390723732723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.947253788526217e-08, "output_power": 6.081988877363084e-09, "reference_active": true, "delta_db": -5.053769358858621}], "active_band_count": 3, "active_band_mae_db": 2.650697250254778}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8221830074249047, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ed908ae793605a41a7ddea22bf11e1ece65195efcaeaf164a781b39998add4f7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9841702558644981, "standard_deviation": 0.0023070096628067084, "minimum": 0.9817138875966643, "maximum": 0.989179868147228, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0273888465893957, "standard_deviation": 0.0461855082825404, "minimum": 1.9577411050799507, "maximum": 2.089849888863089, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8945752466669937, "standard_deviation": 0.23595919203596027, "minimum": 0.43664248880094725, "maximum": 1.1958764996621138, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16293739131189494, "standard_deviation": 0.01142567369737264, "minimum": 0.13870115403590924, "maximum": 0.1762451329489974, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945399445151265, "standard_deviation": 0.001024147414623673, "minimum": 0.9936382562367388, "maximum": 0.9963163305333467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3967378656836673, "standard_deviation": 0.05291755711320499, "minimum": 1.3117363417076586, "maximum": 1.4701413226754634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8309137186573532, "standard_deviation": 0.21329281476111123, "minimum": 0.5097782282603359, "maximum": 1.0255217882888694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10502331080670543, "standard_deviation": 0.00851272945926325, "minimum": 0.09215013870600561, "maximum": 0.11336159641869836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.981403793442027, "standard_deviation": 0.005509587464818647, "minimum": 0.9732325035376165, "maximum": 0.9863825485273615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7648440140977932, "standard_deviation": 0.026893190329986983, "minimum": 1.7369150527431636, "maximum": 1.8121515950485534, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5455813829114493, "standard_deviation": 0.7031042719654531, "minimum": 0.462523514963088, "maximum": 2.632414111954968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18351377186329662, "standard_deviation": 0.02335622691725966, "minimum": 0.1588552696627403, "maximum": 0.2252039447426566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994528321035729, "standard_deviation": 0.0019317694591436479, "minimum": 0.9922838602947311, "maximum": 0.9968515028395922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1971064604433852, "standard_deviation": 0.028232739749601037, "minimum": 1.157213506540643, "maximum": 1.238121190626205, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.0881586505328253, "standard_deviation": 0.8167532596818535, "minimum": 0.8308794777219327, "maximum": 3.4710755529939927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11435567470799723, "standard_deviation": 0.01560427820186313, "minimum": 0.09251325654187169, "maximum": 0.14730184690741657, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922725914273585, "standard_deviation": 0.0017724793353937666, "minimum": 0.9891460801505034, "maximum": 0.9943302152820674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9807547218491006, "standard_deviation": 0.07114452216879841, "minimum": 1.9126425212271063, "maximum": 2.1135920350134407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0641168618915111, "standard_deviation": 0.6852294378740549, "minimum": 0.25317478596115656, "maximum": 2.0221929828099383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10773144217936644, "standard_deviation": 0.012151009461555607, "minimum": 0.09222645731077413, "maximum": 0.12827436970950262, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971039477329555, "standard_deviation": 0.0008651140041779354, "minimum": 0.9955246865168538, "maximum": 0.9982443108065707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4384422226833153, "standard_deviation": 0.08701291801894423, "minimum": 1.3518148905894194, "maximum": 1.5649613862785425, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1296163073308467, "standard_deviation": 0.7278885513194805, "minimum": 0.2729121486221728, "maximum": 2.2248429040516107, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06688723084539958, "standard_deviation": 0.009547745013873142, "minimum": 0.058913962331245885, "maximum": 0.08510702992891343, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3070570120372593, "standard_deviation": 0.06180610805512882, "minimum": 0.21540261328007385, "maximum": 0.3797368458231634, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7517356128102906, "standard_deviation": 0.011976842769421325, "minimum": 5.733591678972187, "maximum": 5.7740726119856545, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3074700180388511, "standard_deviation": 0.010312361480635494, "minimum": 0.29419105024383346, "maximum": 0.32123349373839705, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05828406387158655, "standard_deviation": 0.039460350265199284, "minimum": -0.14590630967127718, "maximum": -0.013329167109629874, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996841793394462, "standard_deviation": 0.0006444202215471466, "minimum": 0.9953405897228149, "maximum": 0.9972224511522845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6285460176762472, "standard_deviation": 0.03760834553471608, "minimum": 1.5601783795924073, "maximum": 1.6652940723874932, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5360071364052771, "standard_deviation": 0.204273693824563, "minimum": 0.27902417312737915, "maximum": 0.8452820288948683, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06103644105751606, "standard_deviation": 0.0058307257867792, "minimum": 0.05733841992824329, "maximum": 0.07395319412332665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989847201539639, "standard_deviation": 0.0003351935863494011, "minimum": 0.9984278478770704, "maximum": 0.9993442988302266, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.911447477384326, "standard_deviation": 0.06977930228710125, "minimum": 0.8035953968273989, "maximum": 1.026584507299513, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.437135229915415, "standard_deviation": 0.3047631522506959, "minimum": 0.06299560754218168, "maximum": 0.9839228824991427, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03433685384859462, "standard_deviation": 0.005658751190133533, "minimum": 0.02783855357289688, "maximum": 0.04231759273617438, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0031", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.0383942127227783, "predicted_mos": 3.0383942127227783, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "34aacbf01b2374b228e9cf4ac0f09114babafae734cbb2e73889e8117b808831", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8252542996779084, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6492586096084416, "全帯域_10ms_rms_envelope_nrmse": 0.06576908592113369, "全帯域_10ms_envelope_correlation": 0.9965070626864615, "全帯域_10ms_modulation_mae_db": 0.22345872755304774, "全帯域_40ms_log_rms_mae_db": 0.9781649334796915, "全帯域_40ms_rms_envelope_nrmse": 0.04057038040830691, "全帯域_40ms_envelope_correlation": 0.9987786065632299, "全帯域_40ms_modulation_mae_db": 0.7994043826893525, "300–1000 Hz_10ms_log_rms_mae_db": 1.657155094271792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10361336439738007, "300–1000 Hz_10ms_envelope_correlation": 0.9932830394654223, "300–1000 Hz_10ms_modulation_mae_db": 1.2498705406573005, "300–1000 Hz_40ms_log_rms_mae_db": 1.0007220336452083, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06162804404439201, "300–1000 Hz_40ms_envelope_correlation": 0.9980483795988057, "300–1000 Hz_40ms_modulation_mae_db": 1.2765667974414079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7336070033855004, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16345506370457566, "1000–3000 Hz_10ms_envelope_correlation": 0.984228409751401, "1000–3000 Hz_10ms_modulation_mae_db": 0.8867540299219961, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9910473164454441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10518471230219305, "1000–3000 Hz_40ms_envelope_correlation": 0.9946507771655602, "1000–3000 Hz_40ms_modulation_mae_db": 0.9695679159153731, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5357470062326866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17925455776746543, "3000–6000 Hz_10ms_envelope_correlation": 0.9825592998586186, "3000–6000 Hz_10ms_modulation_mae_db": 2.370335928941088, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8986299725731944, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10714289667144071, "3000–6000 Hz_40ms_envelope_correlation": 0.9956949188837286, "3000–6000 Hz_40ms_modulation_mae_db": 2.7007211896519423, "mr_stft_db": 5.627681706263959, "band_mae_db": 0.3766958025191896, "rms_gain_error_db": -0.14845096159464227, "output_peak": 0.3000141114248319}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6993324576871924, "全帯域_10ms_rms_envelope_nrmse": 0.05905526213999022, "全帯域_10ms_envelope_correlation": 0.9970572179941061, "全帯域_10ms_modulation_mae_db": 0.5347047496218911, "全帯域_40ms_log_rms_mae_db": 0.8265964292423534, "全帯域_40ms_rms_envelope_nrmse": 0.02955665436150805, "全帯域_40ms_envelope_correlation": 0.999253084267586, "全帯域_40ms_modulation_mae_db": 0.4232719706864258, "300–1000 Hz_10ms_log_rms_mae_db": 1.7573112574109762, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10762107187386345, "300–1000 Hz_10ms_envelope_correlation": 0.9922184473468685, "300–1000 Hz_10ms_modulation_mae_db": 0.7663685249636109, "300–1000 Hz_40ms_log_rms_mae_db": 1.033645413642748, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05971489349386068, "300–1000 Hz_40ms_envelope_correlation": 0.9976102568139447, "300–1000 Hz_40ms_modulation_mae_db": 0.9506686167207435, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8213147246971833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1689718770836698, "1000–3000 Hz_10ms_envelope_correlation": 0.9830620397983053, "1000–3000 Hz_10ms_modulation_mae_db": 1.0083518604620094, "1000–3000 Hz_40ms_log_rms_mae_db": 1.03969986041855, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11246783926769963, "1000–3000 Hz_40ms_envelope_correlation": 0.9936408039407154, "1000–3000 Hz_40ms_modulation_mae_db": 0.7997326961148978, "3000–6000 Hz_10ms_log_rms_mae_db": 1.533701014573269, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2548969380105038, "3000–6000 Hz_10ms_envelope_correlation": 0.9637268527670512, "3000–6000 Hz_10ms_modulation_mae_db": 2.462769660154153, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8674205824387684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15443546961520363, "3000–6000 Hz_40ms_envelope_correlation": 0.9909430072105506, "3000–6000 Hz_40ms_modulation_mae_db": 3.122710586562746, "mr_stft_db": 5.647096435653201, "band_mae_db": 0.39761695372788136, "rms_gain_error_db": -0.03229950903708762, "output_peak": 0.30275953239747655}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7406622590089538, "全帯域_10ms_rms_envelope_nrmse": 0.0641211477513309, "全帯域_10ms_envelope_correlation": 0.9965239521234834, "全帯域_10ms_modulation_mae_db": 0.7514748741008451, "全帯域_40ms_log_rms_mae_db": 1.077272780428351, "全帯域_40ms_rms_envelope_nrmse": 0.040681137637244305, "全帯域_40ms_envelope_correlation": 0.9985720804304391, "全帯域_40ms_modulation_mae_db": 0.8237862364576903, "300–1000 Hz_10ms_log_rms_mae_db": 1.805494398384062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12657155203562706, "300–1000 Hz_10ms_envelope_correlation": 0.9895773577654017, "300–1000 Hz_10ms_modulation_mae_db": 1.9523473825335114, "300–1000 Hz_40ms_log_rms_mae_db": 1.0775182453378098, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08683480492341981, "300–1000 Hz_40ms_envelope_correlation": 0.9953038465376751, "300–1000 Hz_40ms_modulation_mae_db": 2.1137310398737594, "1000–3000 Hz_10ms_log_rms_mae_db": 1.77616511616958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15730344063322776, "1000–3000 Hz_10ms_envelope_correlation": 0.985313986368866, "1000–3000 Hz_10ms_modulation_mae_db": 0.870663440136041, "1000–3000 Hz_40ms_log_rms_mae_db": 1.015186866218481, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09561224213628719, "1000–3000 Hz_40ms_envelope_correlation": 0.9957145185129338, "1000–3000 Hz_40ms_modulation_mae_db": 0.42617227283398607, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5578194505367728, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17686385366205803, "3000–6000 Hz_10ms_envelope_correlation": 0.9838251813635376, "3000–6000 Hz_10ms_modulation_mae_db": 2.985441474490944, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9168779041008527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11064027895761216, "3000–6000 Hz_40ms_envelope_correlation": 0.9960534425768245, "3000–6000 Hz_40ms_modulation_mae_db": 3.6931035297711685, "mr_stft_db": 5.657711597477033, "band_mae_db": 0.36880077501871233, "rms_gain_error_db": -0.06115361058928012, "output_peak": 0.2930962888521098}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6798241423299083, "全帯域_10ms_rms_envelope_nrmse": 0.059226319901225864, "全帯域_10ms_envelope_correlation": 0.9970354058488015, "全帯域_10ms_modulation_mae_db": 0.7396407034133273, "全帯域_40ms_log_rms_mae_db": 0.943598791074668, "全帯域_40ms_rms_envelope_nrmse": 0.034009385822772645, "全帯域_40ms_envelope_correlation": 0.9990052722161755, "全帯域_40ms_modulation_mae_db": 0.5137112860291843, "300–1000 Hz_10ms_log_rms_mae_db": 1.7327485389276773, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09159360563780576, "300–1000 Hz_10ms_envelope_correlation": 0.9943848035544365, "300–1000 Hz_10ms_modulation_mae_db": 0.40305850399305315, "300–1000 Hz_40ms_log_rms_mae_db": 1.0784150118859765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05629705663144382, "300–1000 Hz_40ms_envelope_correlation": 0.9978674740105861, "300–1000 Hz_40ms_modulation_mae_db": 0.6199421504270431, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8461436875364559, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16704462648361057, "1000–3000 Hz_10ms_envelope_correlation": 0.9837618095221353, "1000–3000 Hz_10ms_modulation_mae_db": 0.4405729657927437, "1000–3000 Hz_40ms_log_rms_mae_db": 1.085259692010935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11138182615136806, "1000–3000 Hz_40ms_envelope_correlation": 0.9941382812900972, "1000–3000 Hz_40ms_modulation_mae_db": 0.5945534579172587, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5442808350346946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17572644641957336, "3000–6000 Hz_10ms_envelope_correlation": 0.9833442971158877, "3000–6000 Hz_10ms_modulation_mae_db": 1.9797588655574536, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9041494479955782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10523446403992368, "3000–6000 Hz_40ms_envelope_correlation": 0.9959362436337389, "3000–6000 Hz_40ms_modulation_mae_db": 2.5528115323688056, "mr_stft_db": 5.639373836476953, "band_mae_db": 0.3226483860302526, "rms_gain_error_db": -0.05375546244069326, "output_peak": 0.32740454403583946}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.555123463758537, "全帯域_10ms_rms_envelope_nrmse": 0.0568354993661089, "全帯域_10ms_envelope_correlation": 0.9972572226040045, "全帯域_10ms_modulation_mae_db": 0.39961396265526966, "全帯域_40ms_log_rms_mae_db": 0.855249176642566, "全帯域_40ms_rms_envelope_nrmse": 0.03137553673957093, "全帯域_40ms_envelope_correlation": 0.9991340476186534, "全帯域_40ms_modulation_mae_db": 0.26673659462073146, "300–1000 Hz_10ms_log_rms_mae_db": 1.7146933484981401, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1027473028128842, "300–1000 Hz_10ms_envelope_correlation": 0.9928520620822444, "300–1000 Hz_10ms_modulation_mae_db": 0.4287167191893108, "300–1000 Hz_40ms_log_rms_mae_db": 1.0497855887564447, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06301079464508848, "300–1000 Hz_40ms_envelope_correlation": 0.9972114547326015, "300–1000 Hz_40ms_modulation_mae_db": 0.44262966479383575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8288794078441122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16277373849252963, "1000–3000 Hz_10ms_envelope_correlation": 0.9846364469394715, "1000–3000 Hz_10ms_modulation_mae_db": 0.7222391791015651, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0747496052040661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11118628453750652, "1000–3000 Hz_40ms_envelope_correlation": 0.9940615084113592, "1000–3000 Hz_40ms_modulation_mae_db": 0.7301367790296042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5537639558389527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20871209399285834, "3000–6000 Hz_10ms_envelope_correlation": 0.9749581874473097, "3000–6000 Hz_10ms_modulation_mae_db": 1.656484863818168, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9124211592201025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12857622923880938, "3000–6000 Hz_40ms_envelope_correlation": 0.9920214986698432, "3000–6000 Hz_40ms_modulation_mae_db": 2.3518053291492054, "mr_stft_db": 5.622958687311923, "band_mae_db": 0.37826872178791576, "rms_gain_error_db": -0.019907194768091788, "output_peak": 0.31455986826706034}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5806356559487058, "全帯域_10ms_rms_envelope_nrmse": 0.05865767064113028, "全帯域_10ms_envelope_correlation": 0.997085664816569, "全帯域_10ms_modulation_mae_db": 0.7120239758950317, "全帯域_40ms_log_rms_mae_db": 0.9043384756671123, "全帯域_40ms_rms_envelope_nrmse": 0.030512593499757247, "全帯域_40ms_envelope_correlation": 0.999205884573461, "全帯域_40ms_modulation_mae_db": 0.48259770914486194, "300–1000 Hz_10ms_log_rms_mae_db": 1.8085870934400277, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10096389163865381, "300–1000 Hz_10ms_envelope_correlation": 0.993128059866877, "300–1000 Hz_10ms_modulation_mae_db": 0.4462009718037531, "300–1000 Hz_40ms_log_rms_mae_db": 1.1442888773928404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06479949539718818, "300–1000 Hz_40ms_envelope_correlation": 0.9971094287533301, "300–1000 Hz_40ms_modulation_mae_db": 0.7006566516340588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8369990860186693, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17744842653971962, "1000–3000 Hz_10ms_envelope_correlation": 0.9817587981308151, "1000–3000 Hz_10ms_modulation_mae_db": 0.700967091124287, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0799015212888807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11183666029962211, "1000–3000 Hz_40ms_envelope_correlation": 0.99463562326236, "1000–3000 Hz_40ms_modulation_mae_db": 0.7485777311528118, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4667561022868876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1499755899617673, "3000–6000 Hz_10ms_envelope_correlation": 0.9872734616974246, "3000–6000 Hz_10ms_modulation_mae_db": 1.6004731034165367, "3000–6000 Hz_40ms_log_rms_mae_db": 0.826066713939931, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09563665374374429, "3000–6000 Hz_40ms_envelope_correlation": 0.9954812075214581, "3000–6000 Hz_40ms_modulation_mae_db": 1.8328793887635069, "mr_stft_db": 5.637067254675995, "band_mae_db": 0.26987832088805574, "rms_gain_error_db": -0.05150695745406565, "output_peak": 0.31597854371464}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6551411000330494, "全帯域_10ms_rms_envelope_nrmse": 0.059629269544736546, "全帯域_10ms_envelope_correlation": 0.9970002152466466, "全帯域_10ms_modulation_mae_db": 0.4673019355130286, "全帯域_40ms_log_rms_mae_db": 0.8160449490079844, "全帯域_40ms_rms_envelope_nrmse": 0.03042909922583196, "全帯域_40ms_envelope_correlation": 0.9992259401240241, "全帯域_40ms_modulation_mae_db": 0.36345186203561636, "300–1000 Hz_10ms_log_rms_mae_db": 1.817592068837436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10501459570038235, "300–1000 Hz_10ms_envelope_correlation": 0.9928350072094018, "300–1000 Hz_10ms_modulation_mae_db": 1.9689055623567893, "300–1000 Hz_40ms_log_rms_mae_db": 1.143019983829202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06370969198751679, "300–1000 Hz_40ms_envelope_correlation": 0.9975646684721597, "300–1000 Hz_40ms_modulation_mae_db": 1.985707690393718, "1000–3000 Hz_10ms_log_rms_mae_db": 1.777307513860278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16585127532986021, "1000–3000 Hz_10ms_envelope_correlation": 0.9836636032574149, "1000–3000 Hz_10ms_modulation_mae_db": 1.0802778810463483, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0479980368146133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1115932624396203, "1000–3000 Hz_40ms_envelope_correlation": 0.9936615036064761, "1000–3000 Hz_40ms_modulation_mae_db": 0.8906315131182972, "3000–6000 Hz_10ms_log_rms_mae_db": 1.488982644978398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15516117478150043, "3000–6000 Hz_10ms_envelope_correlation": 0.9870806719017204, "3000–6000 Hz_10ms_modulation_mae_db": 1.7247972801169247, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8599166769840656, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08968289256081984, "3000–6000 Hz_40ms_envelope_correlation": 0.9971906956704084, "3000–6000 Hz_40ms_modulation_mae_db": 2.1196493925404076, "mr_stft_db": 5.6263199036294935, "band_mae_db": 0.2652489923486679, "rms_gain_error_db": -0.06492934157641507, "output_peak": 0.300808584879202}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6896729036464033, "全帯域_10ms_rms_envelope_nrmse": 0.0737564647648013, "全帯域_10ms_envelope_correlation": 0.995368700544476, "全帯域_10ms_modulation_mae_db": 0.2753387077730797, "全帯域_40ms_log_rms_mae_db": 0.9378580680570995, "全帯域_40ms_rms_envelope_nrmse": 0.043254064485193616, "全帯域_40ms_envelope_correlation": 0.9983656201247363, "全帯域_40ms_modulation_mae_db": 0.1355332495959736, "300–1000 Hz_10ms_log_rms_mae_db": 1.9052997271349077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12921511107262768, "300–1000 Hz_10ms_envelope_correlation": 0.9890222634824696, "300–1000 Hz_10ms_modulation_mae_db": 1.8904880010746783, "300–1000 Hz_40ms_log_rms_mae_db": 1.1931839217385516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07872561047250555, "300–1000 Hz_40ms_envelope_correlation": 0.9961692125850293, "300–1000 Hz_40ms_modulation_mae_db": 1.943359893272313, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7308986300344333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13981869406558867, "1000–3000 Hz_10ms_envelope_correlation": 0.9895912844743706, "1000–3000 Hz_10ms_modulation_mae_db": 0.7645886357813113, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0229526198667291, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09766747746223968, "1000–3000 Hz_40ms_envelope_correlation": 0.9962984787722916, "1000–3000 Hz_40ms_modulation_mae_db": 1.0226364329325845, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5574136174822109, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2150792606716064, "3000–6000 Hz_10ms_envelope_correlation": 0.9721109767673972, "3000–6000 Hz_10ms_modulation_mae_db": 1.0850789869011785, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9274546517345662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1181926040771775, "3000–6000 Hz_40ms_envelope_correlation": 0.9914868288940679, "3000–6000 Hz_40ms_modulation_mae_db": 1.4356140111843831, "mr_stft_db": 5.615450753807717, "band_mae_db": 0.27633531777119463, "rms_gain_error_db": -0.043145556182387974, "output_peak": 0.29662344342698066}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68000, "output_samples": 68000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.901229059383825, "spectral_convergence": 0.18178303536149476, "spectral_convergence_status": "ok", "amplitude_floor": 1.0380731662877042e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.449276386656471, "spectral_convergence": 0.1834736844288911, "spectral_convergence_status": "ok", "amplitude_floor": 5.867181533968612e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.532539672751584, "spectral_convergence": 0.17791006510739954, "spectral_convergence_status": "ok", "amplitude_floor": 4.6154090777751923e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006114723941341076, "output_power": 0.0006175475888934324, "delta_db": 0.0429357551006094, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011102815674838913, "output_power": 0.0010652612945608905, "delta_db": -0.17976982309340794, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00031219826577364904, "output_power": 0.00029118663775386765, "delta_db": -0.30259044491106524, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.006894391321554e-05, "output_power": 8.024279349010661e-05, "delta_db": -0.5016903117670181, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.882819827254369e-06, "output_power": 8.42069948988744e-06, "delta_db": -0.6953270892378881, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.595718179316582e-06, "output_power": 4.943900192439629e-06, "delta_db": -0.5378613910051492, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3766958025191896, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14845096159464227, "pitch": null, "output_rms_dbfs": -26.845328643460324, "output_peak": 0.3000141114248319, "reference_rms_dbfs": -26.69687768186568, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.627681706263959}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.0001318685794969069, "log_rms_mae_db": 1.6492586096084416, "rms_envelope_nrmse": 0.06576908592113369, "envelope_correlation": 0.9965070626864615, "rms_gain_error_db": -0.1484901243112915, "reference_mean_power": 0.0021420152585098387, "output_mean_power": 0.0020700152632853676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 1.3134511495070254e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003582591828573965, "output_power": 0.00032643992353591184, "reference_active": true, "delta_db": -0.40394063540958713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.463795601086152e-05, "output_power": 4.457403964691417e-05, "reference_active": true, "delta_db": -0.006223048528517053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.556588167851325e-06, "output_power": 7.117123533703703e-06, "reference_active": true, "delta_db": -0.2602124987210391}], "active_band_count": 3, "active_band_mae_db": 0.22345872755304774}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 0.00012588557284836115, "log_rms_mae_db": 0.9781649334796915, "rms_envelope_nrmse": 0.04057038040830691, "envelope_correlation": 0.9987786065632299, "rms_gain_error_db": -0.14848939600481204, "reference_mean_power": 0.0021571542863483557, "output_mean_power": 0.0020846457694897362}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 1.2592546923162062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031562831458090785, "output_power": 0.00028822919270369386, "reference_active": true, "delta_db": -0.39437990911028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9876750850691092e-05, "output_power": 1.8982724588753018e-05, "reference_active": true, "delta_db": -0.19986847184816053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602361817361658e-07, "output_power": 4.358156723072492e-07, "reference_active": true, "delta_db": -1.803964767109617}], "active_band_count": 3, "active_band_mae_db": 0.7994043826893525}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.00010444605382463925, "log_rms_mae_db": 1.657155094271792, "rms_envelope_nrmse": 0.10361336439738007, "envelope_correlation": 0.9932830394654223, "rms_gain_error_db": -0.2915081435154272, "reference_mean_power": 0.0005604503758559807, "output_mean_power": 0.0005240664412307006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 5.643164593552106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025020135526255573, "output_power": 0.0002290453578878189, "reference_active": true, "delta_db": -0.383681635408757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.833230380435073e-05, "output_power": 3.343118688039873e-05, "reference_active": true, "delta_db": -0.5941312623763177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.791716397073579e-06, "output_power": 3.5875641067063295e-06, "reference_active": true, "delta_db": -2.7717987241868265}], "active_band_count": 3, "active_band_mae_db": 1.2498705406573005}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 9.32306023645191e-05, "log_rms_mae_db": 1.0007220336452083, "rms_envelope_nrmse": 0.06162804404439201, "envelope_correlation": 0.9980483795988057, "rms_gain_error_db": -0.2915088965757149, "reference_mean_power": 0.0005644391167624198, "output_mean_power": 0.000527796145143788}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 5.412619397247975e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022898404327685652, "output_power": 0.00021055937135756495, "reference_active": true, "delta_db": -0.3643064432639571}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5949246143041614e-05, "output_power": 1.4011761103271585e-05, "reference_active": true, "delta_db": -0.5624743641272401}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.408378605310095e-07, "output_power": 3.2844012091558984e-07, "reference_active": true, "delta_db": -2.9029195849330263}], "active_band_count": 3, "active_band_mae_db": 1.2765667974414079}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 4.754025004833877e-05, "log_rms_mae_db": 1.7336070033855004, "rms_envelope_nrmse": 0.16345506370457566, "envelope_correlation": 0.984228409751401, "rms_gain_error_db": -0.5266366867078088, "reference_mean_power": 8.438545283737632e-05, "output_mean_power": 7.474874692430364e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 6.948180578959798e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5558932008941216e-05, "output_power": 2.8667107197929917e-05, "reference_active": true, "delta_db": -0.9356483828497342}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9367090695871985e-06, "output_power": 3.107562223601465e-06, "reference_active": true, "delta_db": -1.0271348737152857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.57982090332121e-07, "output_power": 8.900344717093404e-07, "reference_active": true, "delta_db": 0.6974788332009684}], "active_band_count": 3, "active_band_mae_db": 0.8867540299219961}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 4.260586736236329e-05, "log_rms_mae_db": 0.9910473164454441, "rms_envelope_nrmse": 0.10518471230219305, "envelope_correlation": 0.9946507771655602, "rms_gain_error_db": -0.5266410824050964, "reference_mean_power": 8.498589632737057e-05, "output_mean_power": 7.528054437538067e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 6.685624503092077e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2789374619144395e-05, "output_power": 2.6809000800293165e-05, "reference_active": true, "delta_db": -0.8745250592230533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.67867676799424e-06, "output_power": 1.316906886440736e-06, "reference_active": true, "delta_db": -1.0541201135338796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.924150120657105e-08, "output_power": 4.727376562462413e-08, "reference_active": true, "delta_db": -0.9800585749891861}], "active_band_count": 3, "active_band_mae_db": 0.9695679159153731}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 2.7803794970768208e-05, "log_rms_mae_db": 1.5357470062326866, "rms_envelope_nrmse": 0.17925455776746543, "envelope_correlation": 0.9825592998586186, "rms_gain_error_db": -0.6395939111208887, "reference_mean_power": 9.160796502339738e-06, "output_mean_power": 7.906310110262582e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 3.7361073674013695e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6569760918264083e-06, "output_power": 1.4969113055461608e-06, "reference_active": true, "delta_db": -0.44120173673218677}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.026752210402781e-07, "output_power": 4.0669855602243855e-07, "reference_active": true, "delta_db": -1.708107053895926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8917938491313036e-07, "output_power": 6.035370134572112e-08, "reference_active": true, "delta_db": -4.9616989961951505}], "active_band_count": 3, "active_band_mae_db": 2.370335928941088}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 2.058581204447328e-05, "log_rms_mae_db": 0.8986299725731944, "rms_envelope_nrmse": 0.10714289667144071, "envelope_correlation": 0.9956949188837286, "rms_gain_error_db": -0.63960337705452, "reference_mean_power": 9.225978949789671e-06, "output_mean_power": 7.96254906884259e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 3.5247411287487305e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7104686599805127e-06, "output_power": 1.4791691172983363e-06, "reference_active": true, "delta_db": -0.6309729037731402}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2748792658937344e-07, "output_power": 1.9945029976805989e-07, "reference_active": true, "delta_db": -2.15360600224872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.947253788526217e-08, "output_power": 5.723531682240623e-09, "reference_active": true, "delta_db": -5.3175846629339665}], "active_band_count": 3, "active_band_mae_db": 2.7007211896519423}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8191473756213026, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a570ba6a3942f23b9d2e275b0bc6c13a1890be694ae6208ffdf8d9c331374220", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9845020472803474, "standard_deviation": 0.0023146273636880724, "minimum": 0.9817587981308151, "maximum": 0.9895912844743706, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7939143961932766, "standard_deviation": 0.0459134485549638, "minimum": 1.7308986300344333, "maximum": 1.8461436875364559, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8093018854207877, "standard_deviation": 0.20003199429961302, "minimum": 0.4405729657927437, "maximum": 1.0802778810463483, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16283339279159775, "standard_deviation": 0.010946421899209904, "minimum": 0.13981869406558867, "maximum": 0.17744842653971962, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946001868702241, "standard_deviation": 0.0009587803735187064, "minimum": 0.9936408039407154, "maximum": 0.9962984787722916, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0445994397834624, "standard_deviation": 0.033902372640680904, "minimum": 0.9910473164454441, "maximum": 1.085259692010935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7727510998768516, "standard_deviation": 0.19642335915951364, "minimum": 0.42617227283398607, "maximum": 1.0226364329325845, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10711628807456706, "standard_deviation": 0.006877271882037224, "minimum": 0.09561224213628719, "maximum": 0.11246783926769963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793598661148684, "standard_deviation": 0.008320364193543654, "minimum": 0.9637268527670512, "maximum": 0.9872734616974246, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.529808078370484, "standard_deviation": 0.03385418528074825, "minimum": 1.4667561022868876, "maximum": 1.5578194505367728, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.9831425204245556, "standard_deviation": 0.5989248762781542, "minimum": 1.0850789869011785, "maximum": 2.985441474490944, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18945873940841662, "standard_deviation": 0.034838064115190207, "minimum": 0.1499755899617673, "maximum": 0.2548969380105038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943509803825775, "standard_deviation": 0.00244374227174378, "minimum": 0.9909430072105506, "maximum": 0.9971906956704084, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8891171386233824, "standard_deviation": 0.03463839794468529, "minimum": 0.826066713939931, "maximum": 0.9274546517345662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.4761618699990207, "standard_deviation": 0.7165783974603229, "minimum": 1.4356140111843831, "maximum": 3.6931035297711685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11369268611309138, "standard_deviation": 0.020452008173547632, "minimum": 0.08968289256081984, "maximum": 0.15443546961520363, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921626300966402, "standard_deviation": 0.0018746994113143353, "minimum": 0.9890222634824696, "maximum": 0.9943848035544365, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7748601908631274, "standard_deviation": 0.07603790372821122, "minimum": 1.657155094271792, "maximum": 1.9052997271349077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1382445258215008, "standard_deviation": 0.7163634420680842, "minimum": 0.40305850399305315, "maximum": 1.9689055623567893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10841756189615304, "standard_deviation": 0.012918175684924853, "minimum": 0.09159360563780576, "maximum": 0.12921511107262768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971105901880165, "standard_deviation": 0.0009311180176688012, "minimum": 0.9953038465376751, "maximum": 0.9980483795988057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0900723845285976, "standard_deviation": 0.06487227052741039, "minimum": 1.0007220336452083, "maximum": 1.1931839217385516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.25415781306961, "standard_deviation": 0.6770534580480508, "minimum": 0.44262966479383575, "maximum": 2.1137310398737594, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06684004894942691, "standard_deviation": 0.01041132997972733, "minimum": 0.05629705663144382, "maximum": 0.08683480492341981, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3319366587614837, "standard_deviation": 0.05516102209499981, "minimum": 0.2652489923486679, "maximum": 0.39761695372788136, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.634207521912034, "standard_deviation": 0.013818145238784055, "minimum": 5.615450753807717, "maximum": 5.657711597477033, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3064056146247676, "standard_deviation": 0.011694863688419533, "minimum": 0.2930962888521098, "maximum": 0.32740454403583946, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05939357420533297, "standard_deviation": 0.03894356419083143, "minimum": -0.14845096159464227, "maximum": -0.019907194768091788, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967294302330686, "standard_deviation": 0.0006121521924801055, "minimum": 0.995368700544476, "maximum": 0.9972572226040045, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6562063240026488, "standard_deviation": 0.061700379256320284, "minimum": 1.555123463758537, "maximum": 1.7406622590089538, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5129447045656902, "standard_deviation": 0.20825109621848412, "minimum": 0.22345872755304774, "maximum": 0.7514748741008451, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06213134000380721, "standard_deviation": 0.005562760314855599, "minimum": 0.0568354993661089, "maximum": 0.0737564647648013, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989425669897882, "standard_deviation": 0.0003345925532355025, "minimum": 0.9983656201247363, "maximum": 0.999253084267586, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9173904504499782, "standard_deviation": 0.08693844671570262, "minimum": 0.8160449490079844, "maximum": 1.077272780428351, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4760616614074795, "standard_deviation": 0.23969248478925667, "minimum": 0.1355332495959736, "maximum": 0.8237862364576903, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03504860652252321, "standard_deviation": 0.005557403756896453, "minimum": 0.02955665436150805, "maximum": 0.043254064485193616, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0031", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.9197678565979004, "predicted_mos": 2.9197678565979004, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "a82c241c60b98a671da40ef0745ab6330ba7e74f2b8d43b784a8db0beb3862f7", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1821008999831975, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5925090121202121, "全帯域_10ms_rms_envelope_nrmse": 0.06373630205489832, "全帯域_10ms_envelope_correlation": 0.9966299134059392, "全帯域_10ms_modulation_mae_db": 0.7544709784248944, "全帯域_40ms_log_rms_mae_db": 0.8971492145836822, "全帯域_40ms_rms_envelope_nrmse": 0.035298334915592026, "全帯域_40ms_envelope_correlation": 0.999020270589317, "全帯域_40ms_modulation_mae_db": 1.1565235961465274, "300–1000 Hz_10ms_log_rms_mae_db": 2.538930148133204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13551698327634173, "300–1000 Hz_10ms_envelope_correlation": 0.9879960823743006, "300–1000 Hz_10ms_modulation_mae_db": 1.413037955511698, "300–1000 Hz_40ms_log_rms_mae_db": 1.9965534793214157, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07628086919654452, "300–1000 Hz_40ms_envelope_correlation": 0.9966525654010467, "300–1000 Hz_40ms_modulation_mae_db": 1.590109480040984, "1000–3000 Hz_10ms_log_rms_mae_db": 2.610529825990723, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15394910216673777, "1000–3000 Hz_10ms_envelope_correlation": 0.984793575648695, "1000–3000 Hz_10ms_modulation_mae_db": 0.5673489789114047, "1000–3000 Hz_40ms_log_rms_mae_db": 1.976082951822139, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09458220947710841, "1000–3000 Hz_40ms_envelope_correlation": 0.9941215920192756, "1000–3000 Hz_40ms_modulation_mae_db": 0.33949554608732996, "3000–6000 Hz_10ms_log_rms_mae_db": 2.550707638901269, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20366655146494597, "3000–6000 Hz_10ms_envelope_correlation": 0.9782448887916448, "3000–6000 Hz_10ms_modulation_mae_db": 1.5208987837226908, "3000–6000 Hz_40ms_log_rms_mae_db": 1.869907600336696, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1359362296471043, "3000–6000 Hz_40ms_envelope_correlation": 0.9930603360670249, "3000–6000 Hz_40ms_modulation_mae_db": 2.2800720296682893, "mr_stft_db": 6.147341059338803, "band_mae_db": 0.2176661320267352, "rms_gain_error_db": -0.10680111205271742, "output_peak": 0.29562825725983793}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4664554086306925, "全帯域_10ms_rms_envelope_nrmse": 0.061353757567810974, "全帯域_10ms_envelope_correlation": 0.9968360930526222, "全帯域_10ms_modulation_mae_db": 0.6235995958440826, "全帯域_40ms_log_rms_mae_db": 0.8046475733082279, "全帯域_40ms_rms_envelope_nrmse": 0.03289245510122936, "全帯域_40ms_envelope_correlation": 0.9990797369040328, "全帯域_40ms_modulation_mae_db": 0.5896903912169683, "300–1000 Hz_10ms_log_rms_mae_db": 2.3116059303955643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09289216862078213, "300–1000 Hz_10ms_envelope_correlation": 0.9944763590666305, "300–1000 Hz_10ms_modulation_mae_db": 1.7950294722773625, "300–1000 Hz_40ms_log_rms_mae_db": 1.7333806373584242, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053207492993459254, "300–1000 Hz_40ms_envelope_correlation": 0.9984046939453167, "300–1000 Hz_40ms_modulation_mae_db": 1.804924019533208, "1000–3000 Hz_10ms_log_rms_mae_db": 2.574052269518561, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16932013506811983, "1000–3000 Hz_10ms_envelope_correlation": 0.9839940415535542, "1000–3000 Hz_10ms_modulation_mae_db": 0.8387782391680476, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9396955444944564, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11160289566658757, "1000–3000 Hz_40ms_envelope_correlation": 0.9950305153999255, "1000–3000 Hz_40ms_modulation_mae_db": 0.5844092718351113, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5474896393886772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20021203178313215, "3000–6000 Hz_10ms_envelope_correlation": 0.9781116291104791, "3000–6000 Hz_10ms_modulation_mae_db": 1.4961898442408774, "3000–6000 Hz_40ms_log_rms_mae_db": 1.873639579104684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1267296913568956, "3000–6000 Hz_40ms_envelope_correlation": 0.9933707176994243, "3000–6000 Hz_40ms_modulation_mae_db": 2.243219384938019, "mr_stft_db": 6.12689828485076, "band_mae_db": 0.32493999578663424, "rms_gain_error_db": -0.04511513627284916, "output_peak": 0.32151794406665507}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5416898757673219, "全帯域_10ms_rms_envelope_nrmse": 0.05987092343916328, "全帯域_10ms_envelope_correlation": 0.9969642278124371, "全帯域_10ms_modulation_mae_db": 0.8677789512027961, "全帯域_40ms_log_rms_mae_db": 0.8343278817150311, "全帯域_40ms_rms_envelope_nrmse": 0.03328514011921868, "全帯域_40ms_envelope_correlation": 0.9990444406791128, "全帯域_40ms_modulation_mae_db": 0.15999472920995178, "300–1000 Hz_10ms_log_rms_mae_db": 2.374119146665995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11115576650404364, "300–1000 Hz_10ms_envelope_correlation": 0.9919319581628181, "300–1000 Hz_10ms_modulation_mae_db": 1.0488805454303092, "300–1000 Hz_40ms_log_rms_mae_db": 1.8482950250668215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07263384265706158, "300–1000 Hz_40ms_envelope_correlation": 0.996694786015753, "300–1000 Hz_40ms_modulation_mae_db": 1.5706448403618836, "1000–3000 Hz_10ms_log_rms_mae_db": 2.573041836938646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15808384971002185, "1000–3000 Hz_10ms_envelope_correlation": 0.9839737300435466, "1000–3000 Hz_10ms_modulation_mae_db": 0.5394152983225605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.865771843803244, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10506477228796766, "1000–3000 Hz_40ms_envelope_correlation": 0.9928191308609097, "1000–3000 Hz_40ms_modulation_mae_db": 1.0047806786186309, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4287306476849, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19579394787178375, "3000–6000 Hz_10ms_envelope_correlation": 0.977051006549791, "3000–6000 Hz_10ms_modulation_mae_db": 1.256805503465313, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7567256647360248, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10747419544575545, "3000–6000 Hz_40ms_envelope_correlation": 0.9931975515197251, "3000–6000 Hz_40ms_modulation_mae_db": 1.6949503300796744, "mr_stft_db": 6.1065494747006275, "band_mae_db": 0.22031806568723522, "rms_gain_error_db": -0.04927314457979491, "output_peak": 0.3056625170529366}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6932711139079581, "全帯域_10ms_rms_envelope_nrmse": 0.06311057062877921, "全帯域_10ms_envelope_correlation": 0.9966168388574685, "全帯域_10ms_modulation_mae_db": 0.21138715472858194, "全帯域_40ms_log_rms_mae_db": 0.8977307750165565, "全帯域_40ms_rms_envelope_nrmse": 0.03374896137966767, "全帯域_40ms_envelope_correlation": 0.9990051548765145, "全帯域_40ms_modulation_mae_db": 0.48447779747394754, "300–1000 Hz_10ms_log_rms_mae_db": 2.4390128275523306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08637333823098778, "300–1000 Hz_10ms_envelope_correlation": 0.9950279040582466, "300–1000 Hz_10ms_modulation_mae_db": 1.0390306652984602, "300–1000 Hz_40ms_log_rms_mae_db": 1.9107630229711254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04968028203706483, "300–1000 Hz_40ms_envelope_correlation": 0.9983576108051584, "300–1000 Hz_40ms_modulation_mae_db": 1.0907000833495675, "1000–3000 Hz_10ms_log_rms_mae_db": 2.593980813637608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18737263267709772, "1000–3000 Hz_10ms_envelope_correlation": 0.9777185796424862, "1000–3000 Hz_10ms_modulation_mae_db": 1.47352567500028, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9095997190936391, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11200607086547694, "1000–3000 Hz_40ms_envelope_correlation": 0.9925569928807281, "1000–3000 Hz_40ms_modulation_mae_db": 1.0774886306051428, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4618460755713873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23679393424706258, "3000–6000 Hz_10ms_envelope_correlation": 0.9714306177515963, "3000–6000 Hz_10ms_modulation_mae_db": 1.4136660549235505, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8051844730590747, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17415364101170328, "3000–6000 Hz_40ms_envelope_correlation": 0.9884096421575316, "3000–6000 Hz_40ms_modulation_mae_db": 1.9032512799686483, "mr_stft_db": 6.131924244152361, "band_mae_db": 0.27619823657341774, "rms_gain_error_db": -0.030290912253981672, "output_peak": 0.30428768531403017}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6099175318591612, "全帯域_10ms_rms_envelope_nrmse": 0.05452428120218982, "全帯域_10ms_envelope_correlation": 0.9975085538176857, "全帯域_10ms_modulation_mae_db": 0.24521396806729967, "全帯域_40ms_log_rms_mae_db": 0.8736497056431499, "全帯域_40ms_rms_envelope_nrmse": 0.030577324791589037, "全帯域_40ms_envelope_correlation": 0.9992239651639533, "全帯域_40ms_modulation_mae_db": 0.39746833667482756, "300–1000 Hz_10ms_log_rms_mae_db": 2.4990580481084668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10267025727730003, "300–1000 Hz_10ms_envelope_correlation": 0.993016624599345, "300–1000 Hz_10ms_modulation_mae_db": 0.8634778783695215, "300–1000 Hz_40ms_log_rms_mae_db": 1.8975563109956202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06316456696483734, "300–1000 Hz_40ms_envelope_correlation": 0.9974203595377061, "300–1000 Hz_40ms_modulation_mae_db": 0.8598931675796497, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5543044103277057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19534196759313777, "1000–3000 Hz_10ms_envelope_correlation": 0.9779475226957901, "1000–3000 Hz_10ms_modulation_mae_db": 1.3943685782734834, "1000–3000 Hz_40ms_log_rms_mae_db": 1.8824204389147847, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1308458877725941, "1000–3000 Hz_40ms_envelope_correlation": 0.9923770773824794, "1000–3000 Hz_40ms_modulation_mae_db": 1.186001245594642, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4917026277617227, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21352770725647666, "3000–6000 Hz_10ms_envelope_correlation": 0.9738148409363937, "3000–6000 Hz_10ms_modulation_mae_db": 1.870767994563317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.818871340931728, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12862352140039918, "3000–6000 Hz_40ms_envelope_correlation": 0.9920537617730394, "3000–6000 Hz_40ms_modulation_mae_db": 2.4456308870003443, "mr_stft_db": 6.119938892698397, "band_mae_db": 0.3121722584653789, "rms_gain_error_db": -0.06371369522194495, "output_peak": 0.30605363752918946}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5031177007842342, "全帯域_10ms_rms_envelope_nrmse": 0.058089769341340344, "全帯域_10ms_envelope_correlation": 0.9971331434896639, "全帯域_10ms_modulation_mae_db": 0.3512319947791718, "全帯域_40ms_log_rms_mae_db": 0.8504806242072995, "全帯域_40ms_rms_envelope_nrmse": 0.03496644775145433, "全帯域_40ms_envelope_correlation": 0.9989200639181166, "全帯域_40ms_modulation_mae_db": 0.2729122016403462, "300–1000 Hz_10ms_log_rms_mae_db": 2.43540611354779, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09646240889226501, "300–1000 Hz_10ms_envelope_correlation": 0.9937489883049319, "300–1000 Hz_10ms_modulation_mae_db": 0.38313165914195757, "300–1000 Hz_40ms_log_rms_mae_db": 1.9287640043711096, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05450188627857039, "300–1000 Hz_40ms_envelope_correlation": 0.997953872777089, "300–1000 Hz_40ms_modulation_mae_db": 0.5378419231437038, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5362316218300442, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16461413247383702, "1000–3000 Hz_10ms_envelope_correlation": 0.9829770876749844, "1000–3000 Hz_10ms_modulation_mae_db": 0.30553525381639735, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9510927283638815, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10686746954925565, "1000–3000 Hz_40ms_envelope_correlation": 0.9931494937252303, "1000–3000 Hz_40ms_modulation_mae_db": 0.4613892261266168, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4865984401934784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23073009018368554, "3000–6000 Hz_10ms_envelope_correlation": 0.9707130211358674, "3000–6000 Hz_10ms_modulation_mae_db": 2.2412712435643107, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8261122310397149, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1524396883482706, "3000–6000 Hz_40ms_envelope_correlation": 0.990080781071221, "3000–6000 Hz_40ms_modulation_mae_db": 2.974668114769654, "mr_stft_db": 6.125733178679002, "band_mae_db": 0.2380216384922734, "rms_gain_error_db": 0.004285040032140565, "output_peak": 0.31254717484393624}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5346569328741055, "全帯域_10ms_rms_envelope_nrmse": 0.06061889072507704, "全帯域_10ms_envelope_correlation": 0.9968863838025805, "全帯域_10ms_modulation_mae_db": 1.0084995226448106, "全帯域_40ms_log_rms_mae_db": 0.8858342588695431, "全帯域_40ms_rms_envelope_nrmse": 0.029666863495655876, "全帯域_40ms_envelope_correlation": 0.9992300047717635, "全帯域_40ms_modulation_mae_db": 0.6282200624354055, "300–1000 Hz_10ms_log_rms_mae_db": 2.5406292231676324, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1147359480200263, "300–1000 Hz_10ms_envelope_correlation": 0.9911181748938265, "300–1000 Hz_10ms_modulation_mae_db": 0.3305564551592973, "300–1000 Hz_40ms_log_rms_mae_db": 1.98956987126412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06282642624826901, "300–1000 Hz_40ms_envelope_correlation": 0.9972343145365767, "300–1000 Hz_40ms_modulation_mae_db": 0.31718770021599596, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5302786951290366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16140952964263441, "1000–3000 Hz_10ms_envelope_correlation": 0.9833592700764134, "1000–3000 Hz_10ms_modulation_mae_db": 1.2595434776094765, "1000–3000 Hz_40ms_log_rms_mae_db": 1.8681837922622826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09374621895029261, "1000–3000 Hz_40ms_envelope_correlation": 0.9945177663172906, "1000–3000 Hz_40ms_modulation_mae_db": 0.5010815631879999, "3000–6000 Hz_10ms_log_rms_mae_db": 2.487033505823349, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1990153602481717, "3000–6000 Hz_10ms_envelope_correlation": 0.9762556132941093, "3000–6000 Hz_10ms_modulation_mae_db": 1.1446010021807107, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7946143832578934, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10500235210188304, "3000–6000 Hz_40ms_envelope_correlation": 0.9934483037994113, "3000–6000 Hz_40ms_modulation_mae_db": 1.583298769460573, "mr_stft_db": 6.13364309270078, "band_mae_db": 0.13719587258782795, "rms_gain_error_db": -0.01138882549441397, "output_peak": 0.33559274049853627}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5391352594699246, "全帯域_10ms_rms_envelope_nrmse": 0.06011050227986304, "全帯域_10ms_envelope_correlation": 0.9969405951786068, "全帯域_10ms_modulation_mae_db": 0.7910988061620259, "全帯域_40ms_log_rms_mae_db": 0.9339684837697423, "全帯域_40ms_rms_envelope_nrmse": 0.028986997196621934, "全帯域_40ms_envelope_correlation": 0.9992867824801919, "全帯域_40ms_modulation_mae_db": 0.7144153801419358, "300–1000 Hz_10ms_log_rms_mae_db": 2.2864145164156584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11565243659068637, "300–1000 Hz_10ms_envelope_correlation": 0.9911179582604275, "300–1000 Hz_10ms_modulation_mae_db": 0.8082187687191998, "300–1000 Hz_40ms_log_rms_mae_db": 1.7622530310389104, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06471001260798914, "300–1000 Hz_40ms_envelope_correlation": 0.9973669486635106, "300–1000 Hz_40ms_modulation_mae_db": 0.8790188999582189, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5466827634649527, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16714021049086167, "1000–3000 Hz_10ms_envelope_correlation": 0.982221564605267, "1000–3000 Hz_10ms_modulation_mae_db": 0.5186430390281528, "1000–3000 Hz_40ms_log_rms_mae_db": 1.8528112252907236, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09126674771212245, "1000–3000 Hz_40ms_envelope_correlation": 0.9949637147328767, "1000–3000 Hz_40ms_modulation_mae_db": 0.7692996573545136, "3000–6000 Hz_10ms_log_rms_mae_db": 2.463522609549349, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19570819531269482, "3000–6000 Hz_10ms_envelope_correlation": 0.9778666424607431, "3000–6000 Hz_10ms_modulation_mae_db": 1.3077971995010798, "3000–6000 Hz_40ms_log_rms_mae_db": 1.80552490553302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11736994244279737, "3000–6000 Hz_40ms_envelope_correlation": 0.9930868399327307, "3000–6000 Hz_40ms_modulation_mae_db": 1.911501742151545, "mr_stft_db": 6.117016529674497, "band_mae_db": 0.23858796356737155, "rms_gain_error_db": -0.05347444582297295, "output_peak": 0.30138198294358054}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68000, "output_samples": 68000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.462154707841728, "spectral_convergence": 0.19453853236363705, "spectral_convergence_status": "ok", "amplitude_floor": 1.0380731662877042e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.058691306369727, "spectral_convergence": 0.18732021858886017, "spectral_convergence_status": "ok", "amplitude_floor": 5.867181533968612e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.921177163804955, "spectral_convergence": 0.18722162293800346, "spectral_convergence_status": "ok", "amplitude_floor": 4.6154090777751923e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006114723941341076, "output_power": 0.0006179264509385701, "delta_db": 0.04559931087691333, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011102815674838913, "output_power": 0.0010723108876537282, "delta_db": -0.151124142679059, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00031219826577364904, "output_power": 0.0002958902984119592, "delta_db": -0.23299760436802794, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.006894391321554e-05, "output_power": 8.739515659228847e-05, "delta_db": -0.1308770567430953, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.882819827254369e-06, "output_power": 8.567518761631958e-06, "delta_db": -0.6202581371050407, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.595718179316582e-06, "output_power": 5.436780251744006e-06, "delta_db": -0.1251405403882749, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2176661320267352, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10680111205271742, "pitch": null, "output_rms_dbfs": -26.803678793918397, "output_peak": 0.29562825725983793, "reference_rms_dbfs": -26.69687768186568, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.147341059338803}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.0001318685794969069, "log_rms_mae_db": 1.5925090121202121, "rms_envelope_nrmse": 0.06373630205489832, "envelope_correlation": 0.9966299134059392, "rms_gain_error_db": -0.10681123932004294, "reference_mean_power": 0.0021420152585098387, "output_mean_power": 0.002089976660616546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 1.3134511495070254e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003582591828573965, "output_power": 0.000348613819357738, "reference_active": true, "delta_db": -0.11852732067719236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.463795601086152e-05, "output_power": 3.684512587915738e-05, "reference_active": true, "delta_db": -0.8332425527146693}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.556588167851325e-06, "output_power": 5.586778430325356e-06, "reference_active": true, "delta_db": -1.3116430618828214}], "active_band_count": 3, "active_band_mae_db": 0.7544709784248944}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 0.00012588557284836115, "log_rms_mae_db": 0.8971492145836822, "rms_envelope_nrmse": 0.035298334915592026, "envelope_correlation": 0.999020270589317, "rms_gain_error_db": -0.10674905856636162, "reference_mean_power": 0.0021571542863483557, "output_mean_power": 0.0021047780327804327}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 1.2592546923162062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031562831458090785, "output_power": 0.0003066645190058612, "reference_active": true, "delta_db": -0.12512425098176694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9876750850691092e-05, "output_power": 1.6042269071189994e-05, "reference_active": true, "delta_db": -0.9307959766809131}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602361817361658e-07, "output_power": 3.7873397361035703e-07, "reference_active": true, "delta_db": -2.413650560776902}], "active_band_count": 3, "active_band_mae_db": 1.1565235961465274}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.00010444605382463925, "log_rms_mae_db": 2.538930148133204, "rms_envelope_nrmse": 0.13551698327634173, "envelope_correlation": 0.9879960823743006, "rms_gain_error_db": -0.28485654814223527, "reference_mean_power": 0.0005604503758559807, "output_mean_power": 0.0005248697092633662}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 5.643164593552106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025020135526255573, "output_power": 0.0002284579676679585, "reference_active": true, "delta_db": -0.3948334875212817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.833230380435073e-05, "output_power": 2.684370137118564e-05, "reference_active": true, "delta_db": -1.547225227456178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.791716397073579e-06, "output_power": 4.001971832273418e-06, "reference_active": true, "delta_db": -2.297055151557634}], "active_band_count": 3, "active_band_mae_db": 1.413037955511698}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 9.32306023645191e-05, "log_rms_mae_db": 1.9965534793214157, "rms_envelope_nrmse": 0.07628086919654452, "envelope_correlation": 0.9966525654010467, "rms_gain_error_db": -0.2848658549530001, "reference_mean_power": 0.0005644391167624198, "output_mean_power": 0.0005286040887912905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 5.412619397247975e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022898404327685652, "output_power": 0.00020974126002368438, "reference_active": true, "delta_db": -0.3812134693345648}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5949246143041614e-05, "output_power": 1.0199654093581177e-05, "reference_active": true, "delta_db": -1.9415471697171784}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.408378605310095e-07, "output_power": 3.6474670871062343e-07, "reference_active": true, "delta_db": -2.4475678010712096}], "active_band_count": 3, "active_band_mae_db": 1.590109480040984}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 4.754025004833877e-05, "log_rms_mae_db": 2.610529825990723, "rms_envelope_nrmse": 0.15394910216673777, "envelope_correlation": 0.984793575648695, "rms_gain_error_db": -0.09476212588681493, "reference_mean_power": 8.438545283737632e-05, "output_mean_power": 8.256412310159447e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 6.948180578959798e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5558932008941216e-05, "output_power": 3.4128800262414443e-05, "reference_active": true, "delta_db": -0.1782768702853454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9367090695871985e-06, "output_power": 3.5436858586467742e-06, "reference_active": true, "delta_db": -0.4567810527443137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.57982090332121e-07, "output_power": 9.690760168423188e-07, "reference_active": true, "delta_db": 1.0669890137045548}], "active_band_count": 3, "active_band_mae_db": 0.5673489789114047}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 4.260586736236329e-05, "log_rms_mae_db": 1.976082951822139, "rms_envelope_nrmse": 0.09458220947710841, "envelope_correlation": 0.9941215920192756, "rms_gain_error_db": -0.09477902679742917, "reference_mean_power": 8.498589632737057e-05, "output_mean_power": 8.315128335537195e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 6.685624503092077e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2789374619144395e-05, "output_power": 3.191601895106512e-05, "reference_active": true, "delta_db": -0.1172441931778223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.67867676799424e-06, "output_power": 1.506294564149868e-06, "reference_active": true, "delta_db": -0.47057171228792427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.924150120657105e-08, "output_power": 6.541737729919462e-08, "reference_active": true, "delta_db": 0.43067073279624335}], "active_band_count": 3, "active_band_mae_db": 0.33949554608732996}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 2.7803794970768208e-05, "log_rms_mae_db": 2.550707638901269, "rms_envelope_nrmse": 0.20366655146494597, "envelope_correlation": 0.9782448887916448, "rms_gain_error_db": -0.7335308992841596, "reference_mean_power": 9.160796502339738e-06, "output_mean_power": 7.737134488395998e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 3.7361073674013695e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6569760918264083e-06, "output_power": 1.4516425985607802e-06, "reference_active": true, "delta_db": -0.5745653799750643}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.026752210402781e-07, "output_power": 4.905730754793253e-07, "reference_active": true, "delta_db": -0.8937962705286342}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8917938491313036e-07, "output_power": 9.277700294604105e-08, "reference_active": true, "delta_db": -3.094334700664374}], "active_band_count": 3, "active_band_mae_db": 1.5208987837226908}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 2.058581204447328e-05, "log_rms_mae_db": 1.869907600336696, "rms_envelope_nrmse": 0.1359362296471043, "envelope_correlation": 0.9930603360670249, "rms_gain_error_db": -0.7335809571183176, "reference_mean_power": 9.225978949789671e-06, "output_mean_power": 7.792097241166929e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 3.5247411287487305e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7104686599805127e-06, "output_power": 1.473410266302178e-06, "reference_active": true, "delta_db": -0.647914296747385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2748792658937344e-07, "output_power": 2.3859078348510906e-07, "reference_active": true, "delta_db": -1.375416302910838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.947253788526217e-08, "output_power": 6.422942121767956e-09, "reference_active": true, "delta_db": -4.816885489346644}], "active_band_count": 3, "active_band_mae_db": 2.2800720296682893}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7859842982159758, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c072ccdb6d56dc84acc5ebbda0007ae5fa504b2c60b8470fe34a34f696cfd998", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9821231714925921, "standard_deviation": 0.0027557484358751443, "minimum": 0.9777185796424862, "maximum": 0.984793575648695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.5648877796046596, "standard_deviation": 0.028138616014503265, "minimum": 2.5302786951290366, "maximum": 2.610529825990723, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8621448175162254, "standard_deviation": 0.4526725375374921, "minimum": 0.30553525381639735, "maximum": 1.47352567500028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.169653944977806, "standard_deviation": 0.014412970231790846, "minimum": 0.15394910216673777, "maximum": 0.19534196759313777, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9936920354148395, "standard_deviation": 0.0010921013417406888, "minimum": 0.9923770773824794, "maximum": 0.9950305153999255, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.905707280505644, "standard_deviation": 0.045553811320574324, "minimum": 1.8528112252907236, "maximum": 1.976082951822139, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7404932274262485, "standard_deviation": 0.3169270518647945, "minimum": 0.33949554608732996, "maximum": 1.186001245594642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10574778403517568, "standard_deviation": 0.01300270043856895, "minimum": 0.09126674771212245, "maximum": 0.1308458877725941, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.975436032503828, "standard_deviation": 0.0030522744509701963, "minimum": 0.9707130211358674, "maximum": 0.9782448887916448, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.4897038981092665, "standard_deviation": 0.04185122200708114, "minimum": 2.4287306476849, "maximum": 2.550707638901269, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5314997032702313, "standard_deviation": 0.3604189755815516, "minimum": 1.1446010021807107, "maximum": 2.2412712435643107, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20943097729599414, "standard_deviation": 0.016125192252113587, "minimum": 0.19570819531269482, "maximum": 0.23679393424706258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920884917525136, "standard_deviation": 0.0018605079530450565, "minimum": 0.9884096421575316, "maximum": 0.9934483037994113, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8188225222498544, "standard_deviation": 0.03867929284605083, "minimum": 1.7567256647360248, "maximum": 1.873639579104684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.1295740672545937, "standard_deviation": 0.45288484728331085, "minimum": 1.583298769460573, "maximum": 2.974668114769654, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1309661577193511, "standard_deviation": 0.02325405341434531, "minimum": 0.10500235210188304, "maximum": 0.17415364101170328, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9923042562150658, "standard_deviation": 0.002275461250496606, "minimum": 0.9879960823743006, "maximum": 0.9950279040582466, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.4281469942483302, "standard_deviation": 0.09762407553188517, "minimum": 2.2864145164156584, "maximum": 2.5406292231676324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9601704249884758, "standard_deviation": 0.4896052775627712, "minimum": 0.3305564551592973, "maximum": 1.7950294722773625, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10693241342655413, "standard_deviation": 0.015672721151007502, "minimum": 0.08637333823098778, "maximum": 0.13551698327634173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975106439602697, "standard_deviation": 0.0006782551733527408, "minimum": 0.9966525654010467, "maximum": 0.9984046939453167, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.8833919227984435, "standard_deviation": 0.096730506715056, "minimum": 1.7333806373584242, "maximum": 1.9965534793214157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0812900142729016, "standard_deviation": 0.5333477100944697, "minimum": 0.31718770021599596, "maximum": 1.804924019533208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.062125672372974505, "standard_deviation": 0.009347147698798124, "minimum": 0.04968028203706483, "maximum": 0.07628086919654452, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2456375203983593, "standard_deviation": 0.05969024039438502, "minimum": 0.13719587258782795, "maximum": 0.32493999578663424, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.126130594599404, "standard_deviation": 0.012220934693265259, "minimum": 6.1065494747006275, "maximum": 6.147341059338803, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3103339924385878, "standard_deviation": 0.012767606193320971, "minimum": 0.29562825725983793, "maximum": 0.33559274049853627, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04447152895831681, "standard_deviation": 0.033874808011221, "minimum": -0.10680111205271742, "maximum": 0.004285040032140565, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9969394686771256, "standard_deviation": 0.00028670569880805634, "minimum": 0.9966168388574685, "maximum": 0.9975085538176857, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5600941044267014, "standard_deviation": 0.07041587731850085, "minimum": 1.4664554086306925, "maximum": 1.6932711139079581, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6066601214817079, "standard_deviation": 0.30188411664204784, "minimum": 0.21138715472858194, "maximum": 1.0084995226448106, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06017687465489026, "standard_deviation": 0.0029096479868116645, "minimum": 0.05452428120218982, "maximum": 0.06373630205489832, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991013024228754, "standard_deviation": 0.0001300381966715411, "minimum": 0.9989200639181166, "maximum": 0.9992867824801919, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8722235646391541, "standard_deviation": 0.04096510460606297, "minimum": 0.8046475733082279, "maximum": 0.9339684837697423, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5504628118674888, "standard_deviation": 0.30683548599439553, "minimum": 0.15999472920995178, "maximum": 1.1565235961465274, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03242781559387861, "standard_deviation": 0.002398429836062256, "minimum": 0.028986997196621934, "maximum": 0.035298334915592026, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0031", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.207749843597412, "predicted_mos": 3.207749843597412, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "8e4a9e60796d4cf435c8aceb541ed6e30f0543249663ab9d9d8c536d68695f2b", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.033046300057322, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5237835031661024, "全帯域_10ms_rms_envelope_nrmse": 0.051243455709341336, "全帯域_10ms_envelope_correlation": 0.9978313979883425, "全帯域_10ms_modulation_mae_db": 0.5450943379511499, "全帯域_40ms_log_rms_mae_db": 0.7863802746530035, "全帯域_40ms_rms_envelope_nrmse": 0.026990525785909894, "全帯域_40ms_envelope_correlation": 0.9994411072695784, "全帯域_40ms_modulation_mae_db": 0.5388767953297631, "300–1000 Hz_10ms_log_rms_mae_db": 1.8273855349683927, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10580757761942103, "300–1000 Hz_10ms_envelope_correlation": 0.9927871930927612, "300–1000 Hz_10ms_modulation_mae_db": 1.0705215199750682, "300–1000 Hz_40ms_log_rms_mae_db": 1.3110817113759077, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06144907028623848, "300–1000 Hz_40ms_envelope_correlation": 0.997849780847162, "300–1000 Hz_40ms_modulation_mae_db": 1.0891804044359537, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6714548617981406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16882773081202138, "1000–3000 Hz_10ms_envelope_correlation": 0.9839691281117947, "1000–3000 Hz_10ms_modulation_mae_db": 0.47385191642214625, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0637252873431882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12127853323236347, "1000–3000 Hz_40ms_envelope_correlation": 0.9932358003878656, "1000–3000 Hz_40ms_modulation_mae_db": 0.5138570508576855, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4355159855803477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1799663741079177, "3000–6000 Hz_10ms_envelope_correlation": 0.9808469335108535, "3000–6000 Hz_10ms_modulation_mae_db": 1.188889447166585, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8446032104139811, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09412404043786876, "3000–6000 Hz_40ms_envelope_correlation": 0.99467832599745, "3000–6000 Hz_40ms_modulation_mae_db": 1.7074475718445565, "mr_stft_db": 5.573041779205646, "band_mae_db": 0.2946236121052287, "rms_gain_error_db": -0.08811934719034785, "output_peak": 0.29140258557243137}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6755100855578047, "全帯域_10ms_rms_envelope_nrmse": 0.06395680288387477, "全帯域_10ms_envelope_correlation": 0.9965215589529116, "全帯域_10ms_modulation_mae_db": 0.9324965585019506, "全帯域_40ms_log_rms_mae_db": 0.9383665357753816, "全帯域_40ms_rms_envelope_nrmse": 0.03818562420342699, "全帯域_40ms_envelope_correlation": 0.9987106729850646, "全帯域_40ms_modulation_mae_db": 1.2003892059532697, "300–1000 Hz_10ms_log_rms_mae_db": 1.6811193117939018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10482215955260116, "300–1000 Hz_10ms_envelope_correlation": 0.99256019894602, "300–1000 Hz_10ms_modulation_mae_db": 0.45803138775238295, "300–1000 Hz_40ms_log_rms_mae_db": 1.1043975393631469, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05698432867859038, "300–1000 Hz_40ms_envelope_correlation": 0.9977559163843321, "300–1000 Hz_40ms_modulation_mae_db": 0.5836768174978254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6732332743723324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15537054585705007, "1000–3000 Hz_10ms_envelope_correlation": 0.9884844465593922, "1000–3000 Hz_10ms_modulation_mae_db": 1.0944675652985152, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1004047471180223, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11758713618133623, "1000–3000 Hz_40ms_envelope_correlation": 0.9955375122795093, "1000–3000 Hz_40ms_modulation_mae_db": 1.4404917882931096, "3000–6000 Hz_10ms_log_rms_mae_db": 1.37950774238309, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16948512314704595, "3000–6000 Hz_10ms_envelope_correlation": 0.9873588319608008, "3000–6000 Hz_10ms_modulation_mae_db": 1.4791964036816465, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8215111808686305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11961577995798842, "3000–6000 Hz_40ms_envelope_correlation": 0.9967405267161601, "3000–6000 Hz_40ms_modulation_mae_db": 1.9984314247904862, "mr_stft_db": 5.595576933071887, "band_mae_db": 0.43662406579712226, "rms_gain_error_db": -0.004494324198612006, "output_peak": 0.31478359807413814}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.74453832963812, "全帯域_10ms_rms_envelope_nrmse": 0.07556192342779519, "全帯域_10ms_envelope_correlation": 0.9952408336655227, "全帯域_10ms_modulation_mae_db": 0.2979959437522957, "全帯域_40ms_log_rms_mae_db": 1.010132384556266, "全帯域_40ms_rms_envelope_nrmse": 0.04726613280562312, "全帯域_40ms_envelope_correlation": 0.9981596872525018, "全帯域_40ms_modulation_mae_db": 0.21925356788937544, "300–1000 Hz_10ms_log_rms_mae_db": 1.7161554917925086, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11128575129167963, "300–1000 Hz_10ms_envelope_correlation": 0.9921341232270345, "300–1000 Hz_10ms_modulation_mae_db": 0.8657890065193018, "300–1000 Hz_40ms_log_rms_mae_db": 1.1341984556561437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07170447544297257, "300–1000 Hz_40ms_envelope_correlation": 0.9970764602581034, "300–1000 Hz_40ms_modulation_mae_db": 0.8766108570093284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6006389222825559, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13538154888150514, "1000–3000 Hz_10ms_envelope_correlation": 0.9905692545740757, "1000–3000 Hz_10ms_modulation_mae_db": 0.8566245838008145, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9854822792381883, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09635506729725826, "1000–3000 Hz_40ms_envelope_correlation": 0.996641536867837, "1000–3000 Hz_40ms_modulation_mae_db": 1.152275931545941, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4736134520807092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18159047097463052, "3000–6000 Hz_10ms_envelope_correlation": 0.9848052496211472, "3000–6000 Hz_10ms_modulation_mae_db": 2.0277366594772555, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9491719652650297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12221010275588129, "3000–6000 Hz_40ms_envelope_correlation": 0.9966567928075627, "3000–6000 Hz_40ms_modulation_mae_db": 2.728633774717119, "mr_stft_db": 5.602155309855866, "band_mae_db": 0.4468021417685529, "rms_gain_error_db": -0.11747747881971789, "output_peak": 0.2869528581339257}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6382021492584182, "全帯域_10ms_rms_envelope_nrmse": 0.05768085602448263, "全帯域_10ms_envelope_correlation": 0.9972157243830033, "全帯域_10ms_modulation_mae_db": 0.45867886593114354, "全帯域_40ms_log_rms_mae_db": 0.8848941521248297, "全帯域_40ms_rms_envelope_nrmse": 0.029173587373855333, "全帯域_40ms_envelope_correlation": 0.9993182398490682, "全帯域_40ms_modulation_mae_db": 0.6753679403303289, "300–1000 Hz_10ms_log_rms_mae_db": 1.734503516645043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10939901348749512, "300–1000 Hz_10ms_envelope_correlation": 0.9925956865766199, "300–1000 Hz_10ms_modulation_mae_db": 0.8730100808029375, "300–1000 Hz_40ms_log_rms_mae_db": 1.1111442419581583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06697186738441582, "300–1000 Hz_40ms_envelope_correlation": 0.997746357760837, "300–1000 Hz_40ms_modulation_mae_db": 1.3865904056655924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7239596531806738, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16425550625452695, "1000–3000 Hz_10ms_envelope_correlation": 0.9869464403919758, "1000–3000 Hz_10ms_modulation_mae_db": 0.8166210514102795, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1330609609102922, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12275307578581493, "1000–3000 Hz_40ms_envelope_correlation": 0.9951643297505259, "1000–3000 Hz_40ms_modulation_mae_db": 0.7998524951785302, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4012352400979047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15790295480259406, "3000–6000 Hz_10ms_envelope_correlation": 0.9886158394576103, "3000–6000 Hz_10ms_modulation_mae_db": 1.8816731227823578, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8531900941550278, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11378465306321912, "3000–6000 Hz_40ms_envelope_correlation": 0.9962846555169286, "3000–6000 Hz_40ms_modulation_mae_db": 2.4960441975657557, "mr_stft_db": 5.577192995039826, "band_mae_db": 0.4206037912105181, "rms_gain_error_db": -0.07924305804407437, "output_peak": 0.30378143329603}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6860359824592814, "全帯域_10ms_rms_envelope_nrmse": 0.0659523174632048, "全帯域_10ms_envelope_correlation": 0.9964203817100338, "全帯域_10ms_modulation_mae_db": 0.6239638126420901, "全帯域_40ms_log_rms_mae_db": 0.9973697607413432, "全帯域_40ms_rms_envelope_nrmse": 0.038702166601247825, "全帯域_40ms_envelope_correlation": 0.9988428543145275, "全帯域_40ms_modulation_mae_db": 0.6195661615793411, "300–1000 Hz_10ms_log_rms_mae_db": 1.6994811689019338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1396201024097811, "300–1000 Hz_10ms_envelope_correlation": 0.9873966642947106, "300–1000 Hz_10ms_modulation_mae_db": 1.4356365822601498, "300–1000 Hz_40ms_log_rms_mae_db": 1.0791903738349735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08331414893443086, "300–1000 Hz_40ms_envelope_correlation": 0.9960833526571974, "300–1000 Hz_40ms_modulation_mae_db": 1.5531854285732063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7363169263213218, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15889182643864808, "1000–3000 Hz_10ms_envelope_correlation": 0.9865261512516993, "1000–3000 Hz_10ms_modulation_mae_db": 1.0135251362409863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1460454541075684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11692127460284439, "1000–3000 Hz_40ms_envelope_correlation": 0.9943229659305988, "1000–3000 Hz_40ms_modulation_mae_db": 0.9733397599226418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4110250992051743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16966948073931662, "3000–6000 Hz_10ms_envelope_correlation": 0.9828889612871411, "3000–6000 Hz_10ms_modulation_mae_db": 1.435788911850338, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8392132948137692, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08913117411790783, "3000–6000 Hz_40ms_envelope_correlation": 0.9956415549436761, "3000–6000 Hz_40ms_modulation_mae_db": 1.8716172215542193, "mr_stft_db": 5.610408881485426, "band_mae_db": 0.35821577534197707, "rms_gain_error_db": -0.12357559731707926, "output_peak": 0.3117126096597492}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6361262770764189, "全帯域_10ms_rms_envelope_nrmse": 0.06434892611617508, "全帯域_10ms_envelope_correlation": 0.9964939217384423, "全帯域_10ms_modulation_mae_db": 0.4182960711879303, "全帯域_40ms_log_rms_mae_db": 0.926276406364319, "全帯域_40ms_rms_envelope_nrmse": 0.034519364983014604, "全帯域_40ms_envelope_correlation": 0.9989732885441986, "全帯域_40ms_modulation_mae_db": 0.3533018685040666, "300–1000 Hz_10ms_log_rms_mae_db": 1.7001498422879975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.093132342512308, "300–1000 Hz_10ms_envelope_correlation": 0.9942563913754607, "300–1000 Hz_10ms_modulation_mae_db": 0.4401600972360444, "300–1000 Hz_40ms_log_rms_mae_db": 1.1400133691712475, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05109466993672017, "300–1000 Hz_40ms_envelope_correlation": 0.9983616003248629, "300–1000 Hz_40ms_modulation_mae_db": 0.6986291578972549, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6911103090291972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14856933091672736, "1000–3000 Hz_10ms_envelope_correlation": 0.9881115579169502, "1000–3000 Hz_10ms_modulation_mae_db": 1.2872737277336717, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0638284830899503, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10125241131919117, "1000–3000 Hz_40ms_envelope_correlation": 0.9960949945977017, "1000–3000 Hz_40ms_modulation_mae_db": 0.8504453192787619, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3763261054340978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21966341224648747, "3000–6000 Hz_10ms_envelope_correlation": 0.9801844946410933, "3000–6000 Hz_10ms_modulation_mae_db": 1.1183341858036016, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8846214654575995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17934332807034734, "3000–6000 Hz_40ms_envelope_correlation": 0.9909952483672824, "3000–6000 Hz_40ms_modulation_mae_db": 1.4490005355052833, "mr_stft_db": 5.608682123319311, "band_mae_db": 0.504973559282406, "rms_gain_error_db": -0.051501940811834554, "output_peak": 0.3127597957658908}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6899984545764883, "全帯域_10ms_rms_envelope_nrmse": 0.07143643449647531, "全帯域_10ms_envelope_correlation": 0.9956945129180879, "全帯域_10ms_modulation_mae_db": 0.8599095580923152, "全帯域_40ms_log_rms_mae_db": 0.8990292597551967, "全帯域_40ms_rms_envelope_nrmse": 0.04142762327164461, "全帯域_40ms_envelope_correlation": 0.9985615672835407, "全帯域_40ms_modulation_mae_db": 0.7778578551051046, "300–1000 Hz_10ms_log_rms_mae_db": 1.7197685295176683, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11436044620182233, "300–1000 Hz_10ms_envelope_correlation": 0.9919349663265695, "300–1000 Hz_10ms_modulation_mae_db": 2.333031069743125, "300–1000 Hz_40ms_log_rms_mae_db": 1.1447314895203529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07819345038652983, "300–1000 Hz_40ms_envelope_correlation": 0.9966729721757569, "300–1000 Hz_40ms_modulation_mae_db": 2.932047190375366, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7676014665626099, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17578249026048787, "1000–3000 Hz_10ms_envelope_correlation": 0.9835793708608481, "1000–3000 Hz_10ms_modulation_mae_db": 0.4632741944198979, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1310504025144907, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12511735258557313, "1000–3000 Hz_40ms_envelope_correlation": 0.9940529842778187, "1000–3000 Hz_40ms_modulation_mae_db": 0.6131510139748669, "3000–6000 Hz_10ms_log_rms_mae_db": 1.392572674302233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19142040448955702, "3000–6000 Hz_10ms_envelope_correlation": 0.9819868256871042, "3000–6000 Hz_10ms_modulation_mae_db": 1.344879726578115, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8806760818507839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12362686836445713, "3000–6000 Hz_40ms_envelope_correlation": 0.9959698072332551, "3000–6000 Hz_40ms_modulation_mae_db": 1.704542523445656, "mr_stft_db": 5.596718997636312, "band_mae_db": 0.45853848536802455, "rms_gain_error_db": -0.08221884318326458, "output_peak": 0.29182078936134576}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6834156222798553, "全帯域_10ms_rms_envelope_nrmse": 0.060192559069168955, "全帯域_10ms_envelope_correlation": 0.9969821486839848, "全帯域_10ms_modulation_mae_db": 0.6491026882058112, "全帯域_40ms_log_rms_mae_db": 0.9504917153508686, "全帯域_40ms_rms_envelope_nrmse": 0.03653196298204553, "全帯域_40ms_envelope_correlation": 0.9989115656777143, "全帯域_40ms_modulation_mae_db": 0.9257937761515876, "300–1000 Hz_10ms_log_rms_mae_db": 1.8006400299502117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.129549754364226, "300–1000 Hz_10ms_envelope_correlation": 0.9888557570889994, "300–1000 Hz_10ms_modulation_mae_db": 1.786052230791669, "300–1000 Hz_40ms_log_rms_mae_db": 1.1513255137275886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08443948804632279, "300–1000 Hz_40ms_envelope_correlation": 0.9953588883810287, "300–1000 Hz_40ms_modulation_mae_db": 1.89049786116665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.650515161919254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18104245965781854, "1000–3000 Hz_10ms_envelope_correlation": 0.9820181891289056, "1000–3000 Hz_10ms_modulation_mae_db": 1.1575654994586537, "1000–3000 Hz_40ms_log_rms_mae_db": 1.028255171906353, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1370067781086086, "1000–3000 Hz_40ms_envelope_correlation": 0.9914852214974791, "1000–3000 Hz_40ms_modulation_mae_db": 1.2376125505959128, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4355740307750762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14548570807457675, "3000–6000 Hz_10ms_envelope_correlation": 0.9876130588882719, "3000–6000 Hz_10ms_modulation_mae_db": 1.8214811242265956, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9005845750634673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07633116707803513, "3000–6000 Hz_40ms_envelope_correlation": 0.9970151667187863, "3000–6000 Hz_40ms_modulation_mae_db": 2.2099009491215003, "mr_stft_db": 5.569352774023131, "band_mae_db": 0.2953644205607454, "rms_gain_error_db": -0.09292303119062195, "output_peak": 0.30856975870237335}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68000, "output_samples": 68000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.748415243330735, "spectral_convergence": 0.18701963530067112, "spectral_convergence_status": "ok", "amplitude_floor": 1.0380731662877042e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.382930423224444, "spectral_convergence": 0.18511755041374575, "spectral_convergence_status": "ok", "amplitude_floor": 5.867181533968612e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.587779671061758, "spectral_convergence": 0.17805375552957212, "spectral_convergence_status": "ok", "amplitude_floor": 4.6154090777751923e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006114723941341076, "output_power": 0.0006166680524796972, "delta_db": 0.03674594737212868, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011102815674838913, "output_power": 0.001087930687395506, "delta_db": -0.08831902739312934, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00031219826577364904, "output_power": 0.0002976801174206792, "delta_db": -0.20680657953887588, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.006894391321554e-05, "output_power": 7.870336036081391e-05, "delta_db": -0.5858179490912284, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.882819827254369e-06, "output_power": 1.0606228439890425e-05, "delta_db": 0.3068009898250256, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.595718179316582e-06, "output_power": 4.93776839728926e-06, "delta_db": -0.5432511794109842, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2946236121052287, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08811934719034785, "pitch": null, "output_rms_dbfs": -26.78499702905603, "output_peak": 0.29140258557243137, "reference_rms_dbfs": -26.69687768186568, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.573041779205646}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.0001318685794969069, "log_rms_mae_db": 1.5237835031661024, "rms_envelope_nrmse": 0.051243455709341336, "envelope_correlation": 0.9978313979883425, "rms_gain_error_db": -0.08813194292543332, "reference_mean_power": 0.0021420152585098387, "output_mean_power": 0.0020989851493112967}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 1.3134511495070254e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003582591828573965, "output_power": 0.000337617713949149, "reference_active": true, "delta_db": -0.2577210611080494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.463795601086152e-05, "output_power": 3.9290749451767225e-05, "reference_active": true, "delta_db": -0.5541398702883576}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.556588167851325e-06, "output_power": 6.251490070076809e-06, "reference_active": true, "delta_db": -0.8234220824570428}], "active_band_count": 3, "active_band_mae_db": 0.5450943379511499}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 0.00012588557284836115, "log_rms_mae_db": 0.7863802746530035, "rms_envelope_nrmse": 0.026990525785909894, "envelope_correlation": 0.9994411072695784, "rms_gain_error_db": -0.08813105452008138, "reference_mean_power": 0.0021571542863483557, "output_mean_power": 0.0021138204875325766}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 1.2592546923162062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031562831458090785, "output_power": 0.0002975336992220755, "reference_active": true, "delta_db": -0.2563979439020579}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9876750850691092e-05, "output_power": 1.7405812466726207e-05, "reference_active": true, "delta_db": -0.5765109395120314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602361817361658e-07, "output_power": 5.512227967442182e-07, "reference_active": true, "delta_db": -0.7837215025751999}], "active_band_count": 3, "active_band_mae_db": 0.5388767953297631}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.00010444605382463925, "log_rms_mae_db": 1.8273855349683927, "rms_envelope_nrmse": 0.10580757761942103, "envelope_correlation": 0.9927871930927612, "rms_gain_error_db": -0.2410769180683052, "reference_mean_power": 0.0005604503758559807, "output_mean_power": 0.000530187486082852}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 5.643164593552106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025020135526255573, "output_power": 0.0002334520926182099, "reference_active": true, "delta_db": -0.3009188658748193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.833230380435073e-05, "output_power": 2.9304102679163606e-05, "reference_active": true, "delta_db": -1.1663649411849475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.791716397073579e-06, "output_power": 4.545183981308761e-06, "reference_active": true, "delta_db": -1.7442807528654378}], "active_band_count": 3, "active_band_mae_db": 1.0705215199750682}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 9.32306023645191e-05, "log_rms_mae_db": 1.3110817113759077, "rms_envelope_nrmse": 0.06144907028623848, "envelope_correlation": 0.997849780847162, "rms_gain_error_db": -0.2410794262534402, "reference_mean_power": 0.0005644391167624198, "output_mean_power": 0.0005339605367825802}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 5.412619397247975e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022898404327685652, "output_power": 0.0002131736685476982, "reference_active": true, "delta_db": -0.3107166051758923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5949246143041614e-05, "output_power": 1.1662516617701517e-05, "reference_active": true, "delta_db": -1.3594788494524501}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.408378605310095e-07, "output_power": 4.436225214911468e-07, "reference_active": true, "delta_db": -1.5973457586795181}], "active_band_count": 3, "active_band_mae_db": 1.0891804044359537}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 4.754025004833877e-05, "log_rms_mae_db": 1.6714548617981406, "rms_envelope_nrmse": 0.16882773081202138, "envelope_correlation": 0.9839691281117947, "rms_gain_error_db": -0.644040040089977, "reference_mean_power": 8.438545283737632e-05, "output_mean_power": 7.275512333634845e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 6.948180578959798e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5558932008941216e-05, "output_power": 2.8422846715365868e-05, "reference_active": true, "delta_db": -0.9728113580342577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9367090695871985e-06, "output_power": 3.854279100705337e-06, "reference_active": true, "delta_db": -0.09190160660306332}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.57982090332121e-07, "output_power": 6.981916743771232e-07, "reference_active": true, "delta_db": -0.3568427846291179}], "active_band_count": 3, "active_band_mae_db": 0.47385191642214625}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 4.260586736236329e-05, "log_rms_mae_db": 1.0637252873431882, "rms_envelope_nrmse": 0.12127853323236347, "envelope_correlation": 0.9932358003878656, "rms_gain_error_db": -0.6440440621672322, "reference_mean_power": 8.498589632737057e-05, "output_mean_power": 7.327274352291888e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 6.685624503092077e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2789374619144395e-05, "output_power": 2.6666866548903205e-05, "reference_active": true, "delta_db": -0.8976114604309876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.67867676799424e-06, "output_power": 1.6862631632368833e-06, "reference_active": true, "delta_db": 0.019582728019455514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.924150120657105e-08, "output_power": 5.130838664509678e-08, "reference_active": true, "delta_db": -0.6243769641226133}], "active_band_count": 3, "active_band_mae_db": 0.5138570508576855}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 2.7803794970768208e-05, "log_rms_mae_db": 1.4355159855803477, "rms_envelope_nrmse": 0.1799663741079177, "envelope_correlation": 0.9808469335108535, "rms_gain_error_db": 0.020940173471665283, "reference_mean_power": 9.160796502339738e-06, "output_mean_power": 9.205073343455249e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 3.7361073674013695e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6569760918264083e-06, "output_power": 1.589203266898535e-06, "reference_active": true, "delta_db": -0.18136792960986625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.026752210402781e-07, "output_power": 5.662823099123951e-07, "reference_active": true, "delta_db": -0.27050340923093963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8917938491313036e-07, "output_power": 9.234090122128423e-08, "reference_active": true, "delta_db": -3.1147970026589493}], "active_band_count": 3, "active_band_mae_db": 1.188889447166585}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 2.058581204447328e-05, "log_rms_mae_db": 0.8446032104139811, "rms_envelope_nrmse": 0.09412404043786876, "envelope_correlation": 0.99467832599745, "rms_gain_error_db": 0.02093174994556922, "reference_mean_power": 9.225978949789671e-06, "output_mean_power": 9.270552855892162e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 3.5247411287487305e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7104686599805127e-06, "output_power": 1.6119899300667808e-06, "reference_active": true, "delta_db": -0.25752796758989777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2748792658937344e-07, "output_power": 2.837447484048656e-07, "reference_active": true, "delta_db": -0.6226746132247953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.947253788526217e-08, "output_power": 7.331765128646626e-09, "reference_active": true, "delta_db": -4.242140134718976}], "active_band_count": 3, "active_band_mae_db": 1.7074475718445565}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8285971014936885, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "863629526f798b8f6e7840f3a72259868ecbdddba28ebf785b3a44b050a87e43", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9862755673494552, "standard_deviation": 0.0028763326952697364, "minimum": 0.9820181891289056, "maximum": 0.9905692545740757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6893538219332607, "standard_deviation": 0.05276348253743636, "minimum": 1.6006389222825559, "maximum": 1.7676014665626099, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8954004593481206, "standard_deviation": 0.3042607822287525, "minimum": 0.4632741944198979, "maximum": 1.2872737277336717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16101517988484818, "standard_deviation": 0.014827839530159398, "minimum": 0.13538154888150514, "maximum": 0.18104245965781854, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945669181986669, "standard_deviation": 0.00167034570862506, "minimum": 0.9914852214974791, "maximum": 0.996641536867837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0814815982785067, "standard_deviation": 0.056467329136731984, "minimum": 0.9854822792381883, "maximum": 1.1460454541075684, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9476282387059312, "standard_deviation": 0.31642236412666, "minimum": 0.5138570508576855, "maximum": 1.4404917882931096, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11728395363912378, "standard_deviation": 0.013046466383629768, "minimum": 0.09635506729725826, "maximum": 0.1370067781086086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9842875243817528, "standard_deviation": 0.0032823374211808365, "minimum": 0.9801844946410933, "maximum": 0.9886158394576103, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.413171291232329, "standard_deviation": 0.03320198708879837, "minimum": 1.3763261054340978, "maximum": 1.4736134520807092, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5372474476958118, "standard_deviation": 0.3354119452817815, "minimum": 1.1183341858036016, "maximum": 2.0277366594772555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17689799107276577, "standard_deviation": 0.022443455871787837, "minimum": 0.14548570807457675, "maximum": 0.21966341224648747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954977597876378, "standard_deviation": 0.0019638570841268017, "minimum": 0.9909952483672824, "maximum": 0.9970151667187863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8716964834860361, "standard_deviation": 0.04094406375004999, "minimum": 0.8215111808686305, "maximum": 0.9491719652650297, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.0207022748180723, "standard_deviation": 0.4326931955542688, "minimum": 1.4490005355052833, "maximum": 2.728633774717119, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11477088923071313, "standard_deviation": 0.03137131388760267, "minimum": 0.07633116707803513, "maximum": 0.17934332807034734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991565122616022, "standard_deviation": 0.0022667117519862036, "minimum": 0.9873966642947106, "maximum": 0.9942563913754607, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7349004282322071, "standard_deviation": 0.05183797612618232, "minimum": 1.6811193117939018, "maximum": 1.8273855349683927, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.157778996885085, "standard_deviation": 0.6581075126981875, "minimum": 0.4401600972360444, "maximum": 2.333031069743125, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11349714342991679, "standard_deviation": 0.01469516637538372, "minimum": 0.093132342512308, "maximum": 0.1396201024097811, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971131660986601, "standard_deviation": 0.0010175862868947749, "minimum": 0.9953588883810287, "maximum": 0.9983616003248629, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.14701033682594, "standard_deviation": 0.07056719831650445, "minimum": 1.0791903738349735, "maximum": 1.3110817113759077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.376302265327647, "standard_deviation": 0.7686499565094793, "minimum": 0.5836768174978254, "maximum": 2.932047190375366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06926893738702761, "standard_deviation": 0.012309315944574026, "minimum": 0.05109466993672017, "maximum": 0.08443948804632279, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.40196823142932187, "standard_deviation": 0.07761645523287344, "minimum": 0.2946236121052287, "maximum": 0.504973559282406, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.591641224204675, "standard_deviation": 0.016238853225701205, "minimum": 5.569352774023131, "maximum": 5.610408881485426, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3027229285707355, "standard_deviation": 0.01107133105566851, "minimum": 0.2869528581339257, "maximum": 0.31478359807413814, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07994420259444406, "standard_deviation": 0.037900321162007436, "minimum": -0.12357559731707926, "maximum": -0.004494324198612006, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9965500600050411, "standard_deviation": 0.0008227524726913546, "minimum": 0.9952408336655227, "maximum": 0.9978313979883425, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.659701300501561, "standard_deviation": 0.06448995950658283, "minimum": 1.5237835031661024, "maximum": 1.74453832963812, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5981922295330857, "standard_deviation": 0.21655589718249782, "minimum": 0.2979959437522957, "maximum": 0.9324965585019506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06379665939881476, "standard_deviation": 0.007650776060887928, "minimum": 0.051243455709341336, "maximum": 0.07556192342779519, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988648728970242, "standard_deviation": 0.00040796801920772876, "minimum": 0.9981596872525018, "maximum": 0.9994411072695784, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9241175611651511, "standard_deviation": 0.07051563480626573, "minimum": 0.7863802746530035, "maximum": 1.010132384556266, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6638008963553546, "standard_deviation": 0.3119763694639211, "minimum": 0.21925356788937544, "maximum": 1.2003892059532697, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03659962350084599, "standard_deviation": 0.006500911069718704, "minimum": 0.026990525785909894, "maximum": 0.04726613280562312, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0031", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.892361879348755, "predicted_mos": 2.892361879348755, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "5cb7f3f761b25ff2c2d4b6c0054d8e4fefd21fd55b465c9ec5e6e7157c34a64d", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8971604998223484, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6213527695111456, "全帯域_10ms_rms_envelope_nrmse": 0.06472329733932543, "全帯域_10ms_envelope_correlation": 0.9966133443014594, "全帯域_10ms_modulation_mae_db": 0.254813343335853, "全帯域_40ms_log_rms_mae_db": 0.9464890719113327, "全帯域_40ms_rms_envelope_nrmse": 0.03887924294458006, "全帯域_40ms_envelope_correlation": 0.9988866378726771, "全帯域_40ms_modulation_mae_db": 0.9326358917330272, "300–1000 Hz_10ms_log_rms_mae_db": 2.0477269620851057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0990028847156784, "300–1000 Hz_10ms_envelope_correlation": 0.9939225631793063, "300–1000 Hz_10ms_modulation_mae_db": 1.2131865438182672, "300–1000 Hz_40ms_log_rms_mae_db": 1.5828672320390589, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05929437351811997, "300–1000 Hz_40ms_envelope_correlation": 0.9982300870938126, "300–1000 Hz_40ms_modulation_mae_db": 1.339658822841166, "1000–3000 Hz_10ms_log_rms_mae_db": 2.088180571515656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16218261540734039, "1000–3000 Hz_10ms_envelope_correlation": 0.9840624334054344, "1000–3000 Hz_10ms_modulation_mae_db": 0.9225470517838487, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5322871045140223, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10112891960548277, "1000–3000 Hz_40ms_envelope_correlation": 0.9947087849389035, "1000–3000 Hz_40ms_modulation_mae_db": 1.0072733373759177, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9131713311958758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16648974617226606, "3000–6000 Hz_10ms_envelope_correlation": 0.9861235997707941, "3000–6000 Hz_10ms_modulation_mae_db": 2.2217051987538636, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3909841423980414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10807853347166572, "3000–6000 Hz_40ms_envelope_correlation": 0.9962633198126932, "3000–6000 Hz_40ms_modulation_mae_db": 2.6429365847707844, "mr_stft_db": 5.8023483688200015, "band_mae_db": 0.360445453206619, "rms_gain_error_db": -0.14463646261622287, "output_peak": 0.3022201239703325}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.60257430690886, "全帯域_10ms_rms_envelope_nrmse": 0.05797735196963572, "全帯域_10ms_envelope_correlation": 0.9971683099690837, "全帯域_10ms_modulation_mae_db": 0.422810221550904, "全帯域_40ms_log_rms_mae_db": 0.8266459830936668, "全帯域_40ms_rms_envelope_nrmse": 0.029891859374575024, "全帯域_40ms_envelope_correlation": 0.9992436398105007, "全帯域_40ms_modulation_mae_db": 0.19852392009023342, "300–1000 Hz_10ms_log_rms_mae_db": 2.045942723400177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1053655120066448, "300–1000 Hz_10ms_envelope_correlation": 0.9925810805849512, "300–1000 Hz_10ms_modulation_mae_db": 0.5789670644034857, "300–1000 Hz_40ms_log_rms_mae_db": 1.5518499115728133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06141025332296055, "300–1000 Hz_40ms_envelope_correlation": 0.9975013340527814, "300–1000 Hz_40ms_modulation_mae_db": 0.6341044071807871, "1000–3000 Hz_10ms_log_rms_mae_db": 2.172158422585417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17161688071886927, "1000–3000 Hz_10ms_envelope_correlation": 0.982025356257921, "1000–3000 Hz_10ms_modulation_mae_db": 1.2051884165364262, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5670229347793962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10918247934614679, "1000–3000 Hz_40ms_envelope_correlation": 0.9936361647653718, "1000–3000 Hz_40ms_modulation_mae_db": 0.9627046010498365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8813177672037937, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22513683859944605, "3000–6000 Hz_10ms_envelope_correlation": 0.9730521457673118, "3000–6000 Hz_10ms_modulation_mae_db": 1.9878855137127578, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3649644086156025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14633137437524385, "3000–6000 Hz_40ms_envelope_correlation": 0.992217059548486, "3000–6000 Hz_40ms_modulation_mae_db": 2.7043147970391685, "mr_stft_db": 5.805179343746094, "band_mae_db": 0.37275501192077204, "rms_gain_error_db": -0.04242796297832977, "output_peak": 0.30606372814192895}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5867027890113037, "全帯域_10ms_rms_envelope_nrmse": 0.060884257491007655, "全帯域_10ms_envelope_correlation": 0.9968776092471059, "全帯域_10ms_modulation_mae_db": 0.8465690990054516, "全帯域_40ms_log_rms_mae_db": 0.9728491135208592, "全帯域_40ms_rms_envelope_nrmse": 0.04057114863530257, "全帯域_40ms_envelope_correlation": 0.9985903305050228, "全帯域_40ms_modulation_mae_db": 0.7052911661439057, "300–1000 Hz_10ms_log_rms_mae_db": 2.0672519300383643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1232721201921447, "300–1000 Hz_10ms_envelope_correlation": 0.9901659934309072, "300–1000 Hz_10ms_modulation_mae_db": 2.0126685892476677, "300–1000 Hz_40ms_log_rms_mae_db": 1.5399476259874845, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0851290602724496, "300–1000 Hz_40ms_envelope_correlation": 0.995521011458808, "300–1000 Hz_40ms_modulation_mae_db": 2.2145740385567163, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1351795880207827, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15683023958802647, "1000–3000 Hz_10ms_envelope_correlation": 0.9851376586619852, "1000–3000 Hz_10ms_modulation_mae_db": 1.0196517805927072, "1000–3000 Hz_40ms_log_rms_mae_db": 1.492728328504215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09267061483933557, "1000–3000 Hz_40ms_envelope_correlation": 0.9957558667913449, "1000–3000 Hz_40ms_modulation_mae_db": 0.5045765352981189, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9184104606259544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16957824434180868, "3000–6000 Hz_10ms_envelope_correlation": 0.9854984401169463, "3000–6000 Hz_10ms_modulation_mae_db": 2.6069079763978817, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3747834770057008, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10856734848399036, "3000–6000 Hz_40ms_envelope_correlation": 0.9962600003550804, "3000–6000 Hz_40ms_modulation_mae_db": 3.457706986531223, "mr_stft_db": 5.822397099705692, "band_mae_db": 0.3367825008786543, "rms_gain_error_db": -0.06825869270653787, "output_peak": 0.29434929574408264}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5999627421169873, "全帯域_10ms_rms_envelope_nrmse": 0.05735787336159863, "全帯域_10ms_envelope_correlation": 0.9972328399823716, "全帯域_10ms_modulation_mae_db": 0.6374355147260573, "全帯域_40ms_log_rms_mae_db": 0.8828680376483555, "全帯域_40ms_rms_envelope_nrmse": 0.03393013510502369, "全帯域_40ms_envelope_correlation": 0.999024643928935, "全帯域_40ms_modulation_mae_db": 0.38071647861025576, "300–1000 Hz_10ms_log_rms_mae_db": 2.1071889925114373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09262158100782654, "300–1000 Hz_10ms_envelope_correlation": 0.9942834681561564, "300–1000 Hz_10ms_modulation_mae_db": 0.39531714837125564, "300–1000 Hz_40ms_log_rms_mae_db": 1.6266688586872726, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059290454845876576, "300–1000 Hz_40ms_envelope_correlation": 0.9976513557964742, "300–1000 Hz_40ms_modulation_mae_db": 0.5593389710424338, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2214002546963125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16779209910135656, "1000–3000 Hz_10ms_envelope_correlation": 0.9834831608748476, "1000–3000 Hz_10ms_modulation_mae_db": 0.4254751904765572, "1000–3000 Hz_40ms_log_rms_mae_db": 1.643909363415319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11180800950882067, "1000–3000 Hz_40ms_envelope_correlation": 0.9939569066204524, "1000–3000 Hz_40ms_modulation_mae_db": 0.5185223296701237, "3000–6000 Hz_10ms_log_rms_mae_db": 1.892927314638385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17520530400101947, "3000–6000 Hz_10ms_envelope_correlation": 0.9838965720875774, "3000–6000 Hz_10ms_modulation_mae_db": 1.2339951218399123, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3828901325676024, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11194304653364812, "3000–6000 Hz_40ms_envelope_correlation": 0.9953955528167989, "3000–6000 Hz_40ms_modulation_mae_db": 1.7811813188749719, "mr_stft_db": 5.796776013758183, "band_mae_db": 0.3227484776249239, "rms_gain_error_db": -0.06509045965811938, "output_peak": 0.3207386368731431}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5683481728587207, "全帯域_10ms_rms_envelope_nrmse": 0.05727707253485809, "全帯域_10ms_envelope_correlation": 0.9972087709405227, "全帯域_10ms_modulation_mae_db": 0.38838653092499437, "全帯域_40ms_log_rms_mae_db": 0.8476729862396981, "全帯域_40ms_rms_envelope_nrmse": 0.03109820488959368, "全帯域_40ms_envelope_correlation": 0.99914610376209, "全帯域_40ms_modulation_mae_db": 0.22477982142870637, "300–1000 Hz_10ms_log_rms_mae_db": 2.0517225660448557, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10646331549895806, "300–1000 Hz_10ms_envelope_correlation": 0.9923414529632804, "300–1000 Hz_10ms_modulation_mae_db": 0.24879291762414682, "300–1000 Hz_40ms_log_rms_mae_db": 1.5259956973896773, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06359016546255082, "300–1000 Hz_40ms_envelope_correlation": 0.9971682715969401, "300–1000 Hz_40ms_modulation_mae_db": 0.2826525353858434, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1681491542199307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16622975632709436, "1000–3000 Hz_10ms_envelope_correlation": 0.983803721898476, "1000–3000 Hz_10ms_modulation_mae_db": 0.7744134379575188, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5797475736882856, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11308184762029007, "1000–3000 Hz_40ms_envelope_correlation": 0.9936988826237851, "1000–3000 Hz_40ms_modulation_mae_db": 0.7506811796154601, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9399711898244354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2015542851270221, "3000–6000 Hz_10ms_envelope_correlation": 0.9763186357173131, "3000–6000 Hz_10ms_modulation_mae_db": 1.0933581523054576, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4259547098816263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12163532154002048, "3000–6000 Hz_40ms_envelope_correlation": 0.9923723685048513, "3000–6000 Hz_40ms_modulation_mae_db": 1.750038995100442, "mr_stft_db": 5.784322874171907, "band_mae_db": 0.316185355634278, "rms_gain_error_db": -0.014073153962085132, "output_peak": 0.32355908376676235}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4920714886166444, "全帯域_10ms_rms_envelope_nrmse": 0.05718735233289607, "全帯域_10ms_envelope_correlation": 0.9972263040613546, "全帯域_10ms_modulation_mae_db": 0.8161798473634471, "全帯域_40ms_log_rms_mae_db": 0.8920865979474218, "全帯域_40ms_rms_envelope_nrmse": 0.02889190971230222, "全帯域_40ms_envelope_correlation": 0.9992860440037279, "全帯域_40ms_modulation_mae_db": 0.6465603585961519, "300–1000 Hz_10ms_log_rms_mae_db": 2.1433583378069394, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10127551635118218, "300–1000 Hz_10ms_envelope_correlation": 0.993081529471522, "300–1000 Hz_10ms_modulation_mae_db": 0.620850185667011, "300–1000 Hz_40ms_log_rms_mae_db": 1.6626062685922158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06466119530214452, "300–1000 Hz_40ms_envelope_correlation": 0.997106078224216, "300–1000 Hz_40ms_modulation_mae_db": 0.48054892059683135, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1509172781417774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1764014516215948, "1000–3000 Hz_10ms_envelope_correlation": 0.9816375104097838, "1000–3000 Hz_10ms_modulation_mae_db": 0.9629766533589225, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5717437059143191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10956052273433942, "1000–3000 Hz_40ms_envelope_correlation": 0.9945533180762478, "1000–3000 Hz_40ms_modulation_mae_db": 1.0311183914964996, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9186223677096004, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1690199280960518, "3000–6000 Hz_10ms_envelope_correlation": 0.9839405118827248, "3000–6000 Hz_10ms_modulation_mae_db": 1.4331667415439782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3579477603619052, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11263040878228175, "3000–6000 Hz_40ms_envelope_correlation": 0.9936652762001175, "3000–6000 Hz_40ms_modulation_mae_db": 1.7952059731963936, "mr_stft_db": 5.804934652394433, "band_mae_db": 0.24297041253707471, "rms_gain_error_db": -0.04473930852419905, "output_peak": 0.31479176100427964}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5499189173052232, "全帯域_10ms_rms_envelope_nrmse": 0.058008876397696606, "全帯域_10ms_envelope_correlation": 0.9971534350165042, "全帯域_10ms_modulation_mae_db": 0.47468972845076074, "全帯域_40ms_log_rms_mae_db": 0.7693690906727078, "全帯域_40ms_rms_envelope_nrmse": 0.02788427756884577, "全帯域_40ms_envelope_correlation": 0.9993413905798484, "全帯域_40ms_modulation_mae_db": 0.2765945814064138, "300–1000 Hz_10ms_log_rms_mae_db": 2.2039310149206153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10832486287791131, "300–1000 Hz_10ms_envelope_correlation": 0.9922803041349771, "300–1000 Hz_10ms_modulation_mae_db": 1.6550254762739975, "300–1000 Hz_40ms_log_rms_mae_db": 1.749963822285834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06567099399147841, "300–1000 Hz_40ms_envelope_correlation": 0.9973008006229905, "300–1000 Hz_40ms_modulation_mae_db": 1.664983918854599, "1000–3000 Hz_10ms_log_rms_mae_db": 2.236369063702007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16437930296650355, "1000–3000 Hz_10ms_envelope_correlation": 0.9836835868824536, "1000–3000 Hz_10ms_modulation_mae_db": 1.0985942699177185, "1000–3000 Hz_40ms_log_rms_mae_db": 1.654783520100354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10937489768912811, "1000–3000 Hz_40ms_envelope_correlation": 0.9936380283332314, "1000–3000 Hz_40ms_modulation_mae_db": 0.8597299160117218, "3000–6000 Hz_10ms_log_rms_mae_db": 1.895253262117545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15926624148524202, "3000–6000 Hz_10ms_envelope_correlation": 0.9861981513954251, "3000–6000 Hz_10ms_modulation_mae_db": 1.1669808561089383, "3000–6000 Hz_40ms_log_rms_mae_db": 1.41670674493594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09220562682922108, "3000–6000 Hz_40ms_envelope_correlation": 0.9967511377695466, "3000–6000 Hz_40ms_modulation_mae_db": 1.6178815784860785, "mr_stft_db": 5.807644850521527, "band_mae_db": 0.20804666254865767, "rms_gain_error_db": -0.052789980802738945, "output_peak": 0.30137493680659444}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5712301932959207, "全帯域_10ms_rms_envelope_nrmse": 0.07430366697541373, "全帯域_10ms_envelope_correlation": 0.9952958513940627, "全帯域_10ms_modulation_mae_db": 0.41709018069628784, "全帯域_40ms_log_rms_mae_db": 0.8824995622283096, "全帯域_40ms_rms_envelope_nrmse": 0.042926821688686514, "全帯域_40ms_envelope_correlation": 0.9983823901638847, "全帯域_40ms_modulation_mae_db": 0.06377319335586337, "300–1000 Hz_10ms_log_rms_mae_db": 2.233252830616494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12844189970899478, "300–1000 Hz_10ms_envelope_correlation": 0.9891188920948984, "300–1000 Hz_10ms_modulation_mae_db": 1.729479533565286, "300–1000 Hz_40ms_log_rms_mae_db": 1.7310500481645876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07810125033093501, "300–1000 Hz_40ms_envelope_correlation": 0.9961740556026969, "300–1000 Hz_40ms_modulation_mae_db": 1.8043413590986976, "1000–3000 Hz_10ms_log_rms_mae_db": 2.102704708968907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13891182448663042, "1000–3000 Hz_10ms_envelope_correlation": 0.989120913106859, "1000–3000 Hz_10ms_modulation_mae_db": 0.7691454697132829, "1000–3000 Hz_40ms_log_rms_mae_db": 1.530694448596845, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09209903900782805, "1000–3000 Hz_40ms_envelope_correlation": 0.996297847417903, "1000–3000 Hz_40ms_modulation_mae_db": 0.9887153952163518, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9695778351065742, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20403957048284904, "3000–6000 Hz_10ms_envelope_correlation": 0.9748114354282025, "3000–6000 Hz_10ms_modulation_mae_db": 0.426440019268772, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4388818731995163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11016499615025757, "3000–6000 Hz_40ms_envelope_correlation": 0.9925897336617538, "3000–6000 Hz_40ms_modulation_mae_db": 0.7968842223931816, "mr_stft_db": 5.792425848779193, "band_mae_db": 0.2308845599470896, "rms_gain_error_db": -0.03139887379912342, "output_peak": 0.29868602709087133}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68000, "output_samples": 68000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.077227732978933, "spectral_convergence": 0.18325281924990622, "spectral_convergence_status": "ok", "amplitude_floor": 1.0380731662877042e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.626158176994287, "spectral_convergence": 0.18294103080166468, "spectral_convergence_status": "ok", "amplitude_floor": 5.867181533968612e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.703659196486784, "spectral_convergence": 0.1787935621202677, "spectral_convergence_status": "ok", "amplitude_floor": 4.6154090777751923e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006114723941341076, "output_power": 0.0006189619175620865, "delta_db": 0.05287074389049685, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011102815674838913, "output_power": 0.001064252089002352, "delta_db": -0.18388618546403201, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00031219826577364904, "output_power": 0.00029154039129246313, "delta_db": -0.29731753938307026, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.006894391321554e-05, "output_power": 8.134591512908132e-05, "delta_db": -0.4423932126879244, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.882819827254369e-06, "output_power": 8.165630994147915e-06, "delta_db": -0.8289112764555051, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.595718179316582e-06, "output_power": 5.153786903710635e-06, "delta_db": -0.35729376135868546, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.360445453206619, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14463646261622287, "pitch": null, "output_rms_dbfs": -26.841514144481906, "output_peak": 0.3022201239703325, "reference_rms_dbfs": -26.69687768186568, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8023483688200015}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.0001318685794969069, "log_rms_mae_db": 1.6213527695111456, "rms_envelope_nrmse": 0.06472329733932543, "envelope_correlation": 0.9966133443014594, "rms_gain_error_db": -0.1446643786431851, "reference_mean_power": 0.0021420152585098387, "output_mean_power": 0.0020718395648597933}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 1.3134511495070254e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003582591828573965, "output_power": 0.00032909853871416556, "reference_active": true, "delta_db": -0.36871377082720697}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.463795601086152e-05, "output_power": 4.443767882949853e-05, "reference_active": true, "delta_db": -0.01952934075441954}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.556588167851325e-06, "output_power": 6.929566423798886e-06, "reference_active": true, "delta_db": -0.3761969184259324}], "active_band_count": 3, "active_band_mae_db": 0.254813343335853}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 0.00012588557284836115, "log_rms_mae_db": 0.9464890719113327, "rms_envelope_nrmse": 0.03887924294458006, "envelope_correlation": 0.9988866378726771, "rms_gain_error_db": -0.14467514812574317, "reference_mean_power": 0.0021571542863483557, "output_mean_power": 0.0020864774410282133}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 1.2592546923162062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031562831458090785, "output_power": 0.00029023897158171223, "reference_active": true, "delta_db": -0.364202297188354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9876750850691092e-05, "output_power": 1.894299231270259e-05, "reference_active": true, "delta_db": -0.20896810975341173}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602361817361658e-07, "output_power": 3.9557204607276627e-07, "reference_active": true, "delta_db": -2.2247372682573157}], "active_band_count": 3, "active_band_mae_db": 0.9326358917330272}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.00010444605382463925, "log_rms_mae_db": 2.0477269620851057, "rms_envelope_nrmse": 0.0990028847156784, "envelope_correlation": 0.9939225631793063, "rms_gain_error_db": -0.28108365253472734, "reference_mean_power": 0.0005604503758559807, "output_mean_power": 0.0005253258833865062}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 5.643164593552106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025020135526255573, "output_power": 0.00023008801566223015, "reference_active": true, "delta_db": -0.36395659155687604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.833230380435073e-05, "output_power": 3.397288591924335e-05, "reference_active": true, "delta_db": -0.5243248074609639}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.791716397073579e-06, "output_power": 3.604555521871819e-06, "reference_active": true, "delta_db": -2.751278232436962}], "active_band_count": 3, "active_band_mae_db": 1.2131865438182672}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 9.32306023645191e-05, "log_rms_mae_db": 1.5828672320390589, "rms_envelope_nrmse": 0.05929437351811997, "envelope_correlation": 0.9982300870938126, "rms_gain_error_db": -0.28108733845844813, "reference_mean_power": 0.0005644391167624198, "output_mean_power": 0.0005290641932784528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 5.412619397247975e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022898404327685652, "output_power": 0.00021144911142839378, "reference_active": true, "delta_db": -0.3459935518696686}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5949246143041614e-05, "output_power": 1.417214313189675e-05, "reference_active": true, "delta_db": -0.5130463066757435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.408378605310095e-07, "output_power": 3.0956688764362134e-07, "reference_active": true, "delta_db": -3.1599366099780863}], "active_band_count": 3, "active_band_mae_db": 1.339658822841166}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 4.754025004833877e-05, "log_rms_mae_db": 2.088180571515656, "rms_envelope_nrmse": 0.16218261540734039, "envelope_correlation": 0.9840624334054344, "rms_gain_error_db": -0.4601242007411711, "reference_mean_power": 8.438545283737632e-05, "output_mean_power": 7.590234003631328e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 6.948180578959798e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5558932008941216e-05, "output_power": 2.896286558399468e-05, "reference_active": true, "delta_db": -0.89107180003354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9367090695871985e-06, "output_power": 2.999777884716727e-06, "reference_active": true, "delta_db": -1.180442217254499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.57982090332121e-07, "output_power": 8.89757501101128e-07, "reference_active": true, "delta_db": 0.6961271380635072}], "active_band_count": 3, "active_band_mae_db": 0.9225470517838487}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 4.260586736236329e-05, "log_rms_mae_db": 1.5322871045140223, "rms_envelope_nrmse": 0.10112891960548277, "envelope_correlation": 0.9947087849389035, "rms_gain_error_db": -0.4601403694398862, "reference_mean_power": 8.498589632737057e-05, "output_mean_power": 7.644213746465805e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 6.685624503092077e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2789374619144395e-05, "output_power": 2.7011710867858626e-05, "reference_active": true, "delta_db": -0.841810411895208}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.67867676799424e-06, "output_power": 1.2585154801551505e-06, "reference_active": true, "delta_db": -1.2510851818015825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.924150120657105e-08, "output_power": 4.783366012240593e-08, "reference_active": true, "delta_db": -0.9289244184309631}], "active_band_count": 3, "active_band_mae_db": 1.0072733373759177}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 2.7803794970768208e-05, "log_rms_mae_db": 1.9131713311958758, "rms_envelope_nrmse": 0.16648974617226606, "envelope_correlation": 0.9861235997707941, "rms_gain_error_db": -0.680861324828182, "reference_mean_power": 9.160796502339738e-06, "output_mean_power": 7.831538790130267e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 3.7361073674013695e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6569760918264083e-06, "output_power": 1.4817118022178662e-06, "reference_active": true, "delta_db": -0.48552501954220495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.026752210402781e-07, "output_power": 4.001775852266314e-07, "reference_active": true, "delta_db": -1.778305765902125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8917938491313036e-07, "output_power": 6.866656584937916e-08, "reference_active": true, "delta_db": -4.401284810817262}], "active_band_count": 3, "active_band_mae_db": 2.2217051987538636}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 2.058581204447328e-05, "log_rms_mae_db": 1.3909841423980414, "rms_envelope_nrmse": 0.10807853347166572, "envelope_correlation": 0.9962633198126932, "rms_gain_error_db": -0.680892738840299, "reference_mean_power": 9.225978949789671e-06, "output_mean_power": 7.887206027383877e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 3.5247411287487305e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7104686599805127e-06, "output_power": 1.4639252826808962e-06, "reference_active": true, "delta_db": -0.6759620992303641}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2748792658937344e-07, "output_power": 1.9717370350636987e-07, "reference_active": true, "delta_db": -2.2034629975065827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.947253788526217e-08, "output_power": 6.088132444028742e-09, "reference_active": true, "delta_db": -5.049384657575406}], "active_band_count": 3, "active_band_mae_db": 2.6429365847707844}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8209130890088207, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "28539380ea9348874c2d91571cc2fcb1f37b52ca528143ab96ed370734c10201", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.98411929268722, "standard_deviation": 0.0023081968922857975, "minimum": 0.9816375104097838, "maximum": 0.989120913106859, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.159382380231349, "standard_deviation": 0.05203397085099776, "minimum": 2.088180571515656, "maximum": 2.236369063702007, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8972490337921228, "standard_deviation": 0.24183192972406434, "minimum": 0.4254751904765572, "maximum": 1.2051884165364262, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16304302127717696, "standard_deviation": 0.011389889058322436, "minimum": 0.13891182448663042, "maximum": 0.1764014516215948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994530724945905, "standard_deviation": 0.0010194034993565525, "minimum": 0.9936361647653718, "maximum": 0.996297847417903, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.5716146224390946, "standard_deviation": 0.055645564486752064, "minimum": 1.492728328504215, "maximum": 1.654783520100354, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8279152107167538, "standard_deviation": 0.21538499558303648, "minimum": 0.5045765352981189, "maximum": 1.0311183914964996, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10486329129392143, "standard_deviation": 0.00847163981032614, "minimum": 0.09209903900782805, "maximum": 0.11308184762029007, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9812299365207868, "standard_deviation": 0.005522728097851931, "minimum": 0.9730521457673118, "maximum": 0.9861981513954251, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9161564410527705, "standard_deviation": 0.02836327676065131, "minimum": 1.8813177672037937, "maximum": 1.9695778351065742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.521304947491445, "standard_deviation": 0.7055413711342798, "minimum": 0.426440019268772, "maximum": 2.6069079763978817, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18378626978821316, "standard_deviation": 0.023390215977112692, "minimum": 0.15926624148524202, "maximum": 0.22513683859944605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944393060836659, "standard_deviation": 0.0019321186851473922, "minimum": 0.992217059548486, "maximum": 0.9967511377695466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.394139156120742, "standard_deviation": 0.02975028507207996, "minimum": 1.3579477603619052, "maximum": 1.4388818731995163, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.06826880704903, "standard_deviation": 0.8231347901077999, "minimum": 0.7968842223931816, "maximum": 3.457706986531223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11394458202079112, "standard_deviation": 0.015417881998482886, "minimum": 0.09220562682922108, "maximum": 0.14633137437524385, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922219105019998, "standard_deviation": 0.00176810656861806, "minimum": 0.9891188920948984, "maximum": 0.9942834681561564, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1125469196779987, "standard_deviation": 0.07395752121990169, "minimum": 2.045942723400177, "maximum": 2.233252830616494, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0567859323713895, "standard_deviation": 0.682152532993026, "minimum": 0.24879291762414682, "maximum": 2.0126685892476677, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1080959615449176, "standard_deviation": 0.012089577840584391, "minimum": 0.09262158100782654, "maximum": 0.12844189970899478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970816243060899, "standard_deviation": 0.0008564923844945735, "minimum": 0.995521011458808, "maximum": 0.9982300870938126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6213686830898681, "standard_deviation": 0.08646713517384778, "minimum": 1.5259956973896773, "maximum": 1.749963822285834, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1225253716946344, "standard_deviation": 0.7242834535750737, "minimum": 0.2826525353858434, "maximum": 2.2145740385567163, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06714346838081443, "standard_deviation": 0.00941623362447775, "minimum": 0.059290454845876576, "maximum": 0.0851290602724496, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2988523042872586, "standard_deviation": 0.0627393160396256, "minimum": 0.20804666254865767, "maximum": 0.37275501192077204, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.802003631487128, "standard_deviation": 0.011327481512267078, "minimum": 5.784322874171907, "maximum": 5.822397099705692, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3077229491747494, "standard_deviation": 0.010717217877827704, "minimum": 0.29434929574408264, "maximum": 0.32355908376676235, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.057926861880919556, "standard_deviation": 0.03917728122419419, "minimum": -0.14463646261622287, "maximum": -0.014073153962085132, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996847058114058, "standard_deviation": 0.0006635788976761665, "minimum": 0.9952958513940627, "maximum": 0.9972328399823716, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5740201724531009, "standard_deviation": 0.04004638715224338, "minimum": 1.4920714886166444, "maximum": 1.6213527695111456, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5322468082567196, "standard_deviation": 0.21270250075609123, "minimum": 0.254813343335853, "maximum": 0.8465690990054516, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06096496855030399, "standard_deviation": 0.00598329646232436, "minimum": 0.05718735233289607, "maximum": 0.07430366697541373, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989876475783358, "standard_deviation": 0.00034653056100921825, "minimum": 0.9983823901638847, "maximum": 0.9993413905798484, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.877560055407794, "standard_deviation": 0.06468116558147499, "minimum": 0.7693690906727078, "maximum": 0.9728491135208592, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42860942642056965, "standard_deviation": 0.30025980280745856, "minimum": 0.06377319335586337, "maximum": 0.9326358917330272, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.034259199989863695, "standard_deviation": 0.005793277348911839, "minimum": 0.02788427756884577, "maximum": 0.042926821688686514, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0031", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.297877311706543, "predicted_mos": 3.297877311706543, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "d635df890ac747a6745505c54d3c0a4bbc5f11a4b52d02f1c073e1a80100e936", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0561999999918044, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5966766719490937, "全帯域_10ms_rms_envelope_nrmse": 0.06111573371670486, "全帯域_10ms_envelope_correlation": 0.9968626463974923, "全帯域_10ms_modulation_mae_db": 0.9201905230655644, "全帯域_40ms_log_rms_mae_db": 0.8951865379041314, "全帯域_40ms_rms_envelope_nrmse": 0.03240225718046062, "全帯域_40ms_envelope_correlation": 0.999136879057691, "全帯域_40ms_modulation_mae_db": 0.24444295083899537, "300–1000 Hz_10ms_log_rms_mae_db": 1.907344956794267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1213067897572695, "300–1000 Hz_10ms_envelope_correlation": 0.9904007129507674, "300–1000 Hz_10ms_modulation_mae_db": 1.1783032247501997, "300–1000 Hz_40ms_log_rms_mae_db": 1.3296357080831647, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07014670658620235, "300–1000 Hz_40ms_envelope_correlation": 0.9971182359648436, "300–1000 Hz_40ms_modulation_mae_db": 0.5648335589013156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9792256902459724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1658884494423637, "1000–3000 Hz_10ms_envelope_correlation": 0.9832334357292036, "1000–3000 Hz_10ms_modulation_mae_db": 0.8087117397712381, "1000–3000 Hz_40ms_log_rms_mae_db": 1.357896374296084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10812072113943746, "1000–3000 Hz_40ms_envelope_correlation": 0.9936865295174683, "1000–3000 Hz_40ms_modulation_mae_db": 0.8436154175112849, "3000–6000 Hz_10ms_log_rms_mae_db": 1.765818852192894, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17294568511268438, "3000–6000 Hz_10ms_envelope_correlation": 0.9820036888854156, "3000–6000 Hz_10ms_modulation_mae_db": 1.1802554591045367, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2184236046221637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.096709374209627, "3000–6000 Hz_40ms_envelope_correlation": 0.9941751892331114, "3000–6000 Hz_40ms_modulation_mae_db": 1.656337873666241, "mr_stft_db": 5.723118342747284, "band_mae_db": 0.16094727491074015, "rms_gain_error_db": -0.07810734885208132, "output_peak": 0.3448369042589772}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5074954041881523, "全帯域_10ms_rms_envelope_nrmse": 0.061571347441086784, "全帯域_10ms_envelope_correlation": 0.996782214479766, "全帯域_10ms_modulation_mae_db": 0.23287331658489271, "全帯域_40ms_log_rms_mae_db": 0.8422020553587722, "全帯域_40ms_rms_envelope_nrmse": 0.028530247979705152, "全帯域_40ms_envelope_correlation": 0.9993052991442004, "全帯域_40ms_modulation_mae_db": 0.30478493136934853, "300–1000 Hz_10ms_log_rms_mae_db": 1.9639612928155017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11680895897718917, "300–1000 Hz_10ms_envelope_correlation": 0.9907451629834307, "300–1000 Hz_10ms_modulation_mae_db": 0.447087504407837, "300–1000 Hz_40ms_log_rms_mae_db": 1.3827721082084319, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06022681356241281, "300–1000 Hz_40ms_envelope_correlation": 0.9975047315681564, "300–1000 Hz_40ms_modulation_mae_db": 0.5681317868044958, "1000–3000 Hz_10ms_log_rms_mae_db": 1.939122660081551, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16105338904866087, "1000–3000 Hz_10ms_envelope_correlation": 0.9856962492228964, "1000–3000 Hz_10ms_modulation_mae_db": 1.0815815952363736, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3243071751382116, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11484433398084985, "1000–3000 Hz_40ms_envelope_correlation": 0.9942736414587358, "1000–3000 Hz_40ms_modulation_mae_db": 1.037136220707013, "3000–6000 Hz_10ms_log_rms_mae_db": 1.767754520417978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16013675690325838, "3000–6000 Hz_10ms_envelope_correlation": 0.9845556747945219, "3000–6000 Hz_10ms_modulation_mae_db": 1.315877865162902, "3000–6000 Hz_40ms_log_rms_mae_db": 1.194658072304619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07867596507730557, "3000–6000 Hz_40ms_envelope_correlation": 0.9961985265437807, "3000–6000 Hz_40ms_modulation_mae_db": 1.8334764913989474, "mr_stft_db": 5.7363769252034125, "band_mae_db": 0.22477883224235773, "rms_gain_error_db": -0.04655760451866217, "output_peak": 0.3232812003876383}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6261076149259097, "全帯域_10ms_rms_envelope_nrmse": 0.056524571237710985, "全帯域_10ms_envelope_correlation": 0.9972976300896281, "全帯域_10ms_modulation_mae_db": 0.33639806589769655, "全帯域_40ms_log_rms_mae_db": 0.8313436735335344, "全帯域_40ms_rms_envelope_nrmse": 0.02812599446900819, "全帯域_40ms_envelope_correlation": 0.9993315568497514, "全帯域_40ms_modulation_mae_db": 0.0911513995207181, "300–1000 Hz_10ms_log_rms_mae_db": 1.9284882416621663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09373522851595424, "300–1000 Hz_10ms_envelope_correlation": 0.9941054161646766, "300–1000 Hz_10ms_modulation_mae_db": 0.5034812251160933, "300–1000 Hz_40ms_log_rms_mae_db": 1.3332577748231056, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04762441022991721, "300–1000 Hz_40ms_envelope_correlation": 0.9984995859446925, "300–1000 Hz_40ms_modulation_mae_db": 0.6097636369032897, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9697145645031977, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1737217499350324, "1000–3000 Hz_10ms_envelope_correlation": 0.9831745743140063, "1000–3000 Hz_10ms_modulation_mae_db": 1.1676190012143224, "1000–3000 Hz_40ms_log_rms_mae_db": 1.357050405366796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12391621494154001, "1000–3000 Hz_40ms_envelope_correlation": 0.9932022130560101, "1000–3000 Hz_40ms_modulation_mae_db": 1.286829899104748, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7429607107739526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22170073596296835, "3000–6000 Hz_10ms_envelope_correlation": 0.9757682572697491, "3000–6000 Hz_10ms_modulation_mae_db": 1.1331273023495354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2262640033922347, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1555088522669473, "3000–6000 Hz_40ms_envelope_correlation": 0.9923665413391176, "3000–6000 Hz_40ms_modulation_mae_db": 1.7320335532920588, "mr_stft_db": 5.728158879419329, "band_mae_db": 0.41152111562554766, "rms_gain_error_db": -0.05320679554128692, "output_peak": 0.28878134895828766}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5662376121580976, "全帯域_10ms_rms_envelope_nrmse": 0.053901674355691605, "全帯域_10ms_envelope_correlation": 0.9975768650966544, "全帯域_10ms_modulation_mae_db": 0.2802261236236219, "全帯域_40ms_log_rms_mae_db": 0.8448447993601631, "全帯域_40ms_rms_envelope_nrmse": 0.025622108062445914, "全帯域_40ms_envelope_correlation": 0.9994839513086444, "全帯域_40ms_modulation_mae_db": 0.18135023983484053, "300–1000 Hz_10ms_log_rms_mae_db": 2.006052781804691, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11239187678723608, "300–1000 Hz_10ms_envelope_correlation": 0.9919879775261269, "300–1000 Hz_10ms_modulation_mae_db": 1.7915024327534745, "300–1000 Hz_40ms_log_rms_mae_db": 1.4298298519525472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0638896834082962, "300–1000 Hz_40ms_envelope_correlation": 0.9978791461441383, "300–1000 Hz_40ms_modulation_mae_db": 1.5858766476754138, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9229712513945232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16843752896859707, "1000–3000 Hz_10ms_envelope_correlation": 0.9845723238774032, "1000–3000 Hz_10ms_modulation_mae_db": 1.2032529901444804, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3155737134927066, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11627441661313688, "1000–3000 Hz_40ms_envelope_correlation": 0.9946816666748433, "1000–3000 Hz_40ms_modulation_mae_db": 1.2583318012284122, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7782381659564077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17075270555994146, "3000–6000 Hz_10ms_envelope_correlation": 0.9877519233666139, "3000–6000 Hz_10ms_modulation_mae_db": 1.3164482598801317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1964006209804932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.132227301343438, "3000–6000 Hz_40ms_envelope_correlation": 0.9952324343024749, "3000–6000 Hz_40ms_modulation_mae_db": 1.751827441489142, "mr_stft_db": 5.720013294265613, "band_mae_db": 0.40860926336940206, "rms_gain_error_db": -0.07632532321232186, "output_peak": 0.2976823972943172}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5432609663154864, "全帯域_10ms_rms_envelope_nrmse": 0.05865672378826253, "全帯域_10ms_envelope_correlation": 0.997116044423553, "全帯域_10ms_modulation_mae_db": 0.23374647991901923, "全帯域_40ms_log_rms_mae_db": 0.8129352184999888, "全帯域_40ms_rms_envelope_nrmse": 0.030484976881744914, "全帯域_40ms_envelope_correlation": 0.9992374070657153, "全帯域_40ms_modulation_mae_db": 0.44697220662885995, "300–1000 Hz_10ms_log_rms_mae_db": 1.9110765663428289, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1064199919173249, "300–1000 Hz_10ms_envelope_correlation": 0.9924621316913639, "300–1000 Hz_10ms_modulation_mae_db": 1.0371735554479773, "300–1000 Hz_40ms_log_rms_mae_db": 1.361965001926133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058355599542417556, "300–1000 Hz_40ms_envelope_correlation": 0.9978419292576582, "300–1000 Hz_40ms_modulation_mae_db": 1.2261083473573267, "1000–3000 Hz_10ms_log_rms_mae_db": 2.010727446596486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14374788795283966, "1000–3000 Hz_10ms_envelope_correlation": 0.9874366399950066, "1000–3000 Hz_10ms_modulation_mae_db": 0.5563557976648363, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3977958648481983, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09614880864996432, "1000–3000 Hz_40ms_envelope_correlation": 0.9948715601693442, "1000–3000 Hz_40ms_modulation_mae_db": 0.684013530857111, "3000–6000 Hz_10ms_log_rms_mae_db": 1.78599882560569, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2141701328788102, "3000–6000 Hz_10ms_envelope_correlation": 0.9771409847489785, "3000–6000 Hz_10ms_modulation_mae_db": 2.5921887388811506, "3000–6000 Hz_40ms_log_rms_mae_db": 1.218177744863199, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14644097678219586, "3000–6000 Hz_40ms_envelope_correlation": 0.993219065735577, "3000–6000 Hz_40ms_modulation_mae_db": 3.1634673707610026, "mr_stft_db": 5.724343174187138, "band_mae_db": 0.3296125485504293, "rms_gain_error_db": -0.0725505676359156, "output_peak": 0.3191315821708386}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6376982824223467, "全帯域_10ms_rms_envelope_nrmse": 0.06376518795376239, "全帯域_10ms_envelope_correlation": 0.9966706396090852, "全帯域_10ms_modulation_mae_db": 1.0596998864189793, "全帯域_40ms_log_rms_mae_db": 0.9292243320404611, "全帯域_40ms_rms_envelope_nrmse": 0.0359751558981266, "全帯域_40ms_envelope_correlation": 0.9990254758435468, "全帯域_40ms_modulation_mae_db": 0.8263027545605937, "300–1000 Hz_10ms_log_rms_mae_db": 2.003215174001506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12326407618371676, "300–1000 Hz_10ms_envelope_correlation": 0.9904850917218897, "300–1000 Hz_10ms_modulation_mae_db": 0.9644587332521456, "300–1000 Hz_40ms_log_rms_mae_db": 1.4469255741181395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07169078408962391, "300–1000 Hz_40ms_envelope_correlation": 0.9974598591666627, "300–1000 Hz_40ms_modulation_mae_db": 0.8380322907099411, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0923358299467063, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16879749091278012, "1000–3000 Hz_10ms_envelope_correlation": 0.9847737977497754, "1000–3000 Hz_10ms_modulation_mae_db": 1.1380591609511377, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4458900085430868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11622164802527334, "1000–3000 Hz_40ms_envelope_correlation": 0.9950586430899729, "1000–3000 Hz_40ms_modulation_mae_db": 1.1386199164150135, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7995545613255781, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23566677153017404, "3000–6000 Hz_10ms_envelope_correlation": 0.9680084836898183, "3000–6000 Hz_10ms_modulation_mae_db": 0.9667666451227905, "3000–6000 Hz_40ms_log_rms_mae_db": 1.245540909821216, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13539051555678897, "3000–6000 Hz_40ms_envelope_correlation": 0.9920122099673517, "3000–6000 Hz_40ms_modulation_mae_db": 1.6692073725164664, "mr_stft_db": 5.765485755198532, "band_mae_db": 0.3894807600145092, "rms_gain_error_db": -0.12604361017164814, "output_peak": 0.28178313223528484}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6068230940708046, "全帯域_10ms_rms_envelope_nrmse": 0.055313909010181295, "全帯域_10ms_envelope_correlation": 0.9975485449221454, "全帯域_10ms_modulation_mae_db": 0.3452319578625918, "全帯域_40ms_log_rms_mae_db": 0.8711056129368878, "全帯域_40ms_rms_envelope_nrmse": 0.030683715370861987, "全帯域_40ms_envelope_correlation": 0.9993465513260575, "全帯域_40ms_modulation_mae_db": 0.40865434922205823, "300–1000 Hz_10ms_log_rms_mae_db": 2.0517427764830587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12019182043739626, "300–1000 Hz_10ms_envelope_correlation": 0.9904258870565742, "300–1000 Hz_10ms_modulation_mae_db": 1.2287652858592257, "300–1000 Hz_40ms_log_rms_mae_db": 1.5136362195547675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0681662898414775, "300–1000 Hz_40ms_envelope_correlation": 0.9971037324421385, "300–1000 Hz_40ms_modulation_mae_db": 1.531060849699055, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9607914459187081, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.169911318686449, "1000–3000 Hz_10ms_envelope_correlation": 0.9837069257173903, "1000–3000 Hz_10ms_modulation_mae_db": 1.1065018631320285, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3657967840333554, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10926563412363038, "1000–3000 Hz_40ms_envelope_correlation": 0.9952271005151335, "1000–3000 Hz_40ms_modulation_mae_db": 0.8372481389970957, "3000–6000 Hz_10ms_log_rms_mae_db": 1.743907191022985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21312008655113426, "3000–6000 Hz_10ms_envelope_correlation": 0.9800960045409866, "3000–6000 Hz_10ms_modulation_mae_db": 1.6923264762496393, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1569993858871597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1617393004297591, "3000–6000 Hz_40ms_envelope_correlation": 0.9929513350098087, "3000–6000 Hz_40ms_modulation_mae_db": 2.2140055811522434, "mr_stft_db": 5.733126225689094, "band_mae_db": 0.4984862151987, "rms_gain_error_db": -0.1264751341419632, "output_peak": 0.31761829806141917}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6448935920212908, "全帯域_10ms_rms_envelope_nrmse": 0.0674490600542352, "全帯域_10ms_envelope_correlation": 0.9962702468885614, "全帯域_10ms_modulation_mae_db": 1.2979467355049588, "全帯域_40ms_log_rms_mae_db": 0.8832070233214766, "全帯域_40ms_rms_envelope_nrmse": 0.03822420241996642, "全帯域_40ms_envelope_correlation": 0.9988988495089166, "全帯域_40ms_modulation_mae_db": 0.9582589592966612, "300–1000 Hz_10ms_log_rms_mae_db": 1.9230490893884669, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1433409603899735, "300–1000 Hz_10ms_envelope_correlation": 0.9865617047533098, "300–1000 Hz_10ms_modulation_mae_db": 1.1499418314714482, "300–1000 Hz_40ms_log_rms_mae_db": 1.3768854142246607, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08850721314793865, "300–1000 Hz_40ms_envelope_correlation": 0.9953185289237009, "300–1000 Hz_40ms_modulation_mae_db": 1.2956563166682655, "1000–3000 Hz_10ms_log_rms_mae_db": 1.932841489612744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15377476389907416, "1000–3000 Hz_10ms_envelope_correlation": 0.9865189615302149, "1000–3000 Hz_10ms_modulation_mae_db": 0.49945033057811866, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2778489875127919, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10326065600804357, "1000–3000 Hz_40ms_envelope_correlation": 0.9951818971208894, "1000–3000 Hz_40ms_modulation_mae_db": 1.0170665088102975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.750301098183118, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17823891587341256, "3000–6000 Hz_10ms_envelope_correlation": 0.9838192314551217, "3000–6000 Hz_10ms_modulation_mae_db": 1.78394653645591, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1934480387274493, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11648976906598471, "3000–6000 Hz_40ms_envelope_correlation": 0.9954913877622882, "3000–6000 Hz_40ms_modulation_mae_db": 2.4007146743919567, "mr_stft_db": 5.731363901445655, "band_mae_db": 0.38994253180579347, "rms_gain_error_db": -0.13343968397304087, "output_peak": 0.31351573447401826}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68000, "output_samples": 68000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.998750798233352, "spectral_convergence": 0.18947096613796605, "spectral_convergence_status": "ok", "amplitude_floor": 1.0380731662877042e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.547476817453635, "spectral_convergence": 0.1791193074619397, "spectral_convergence_status": "ok", "amplitude_floor": 5.867181533968612e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.623127412554863, "spectral_convergence": 0.1773170341932115, "spectral_convergence_status": "ok", "amplitude_floor": 4.6154090777751923e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006114723941341076, "output_power": 0.0006253461067069299, "delta_db": 0.09743595431029066, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011102815674838913, "output_power": 0.0010877710121799302, "delta_db": -0.08895648670254896, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00031219826577364904, "output_power": 0.000291467266250921, "delta_db": -0.2984069865553781, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.006894391321554e-05, "output_power": 8.124156867457156e-05, "delta_db": -0.4479677002882576, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.882819827254369e-06, "output_power": 9.948617574605225e-06, "delta_db": 0.02881859027406298, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.595718179316582e-06, "output_power": 5.590440641925646e-06, "delta_db": -0.004097931333902589, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16094727491074015, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07810734885208132, "pitch": null, "output_rms_dbfs": -26.77498503071776, "output_peak": 0.3448369042589772, "reference_rms_dbfs": -26.69687768186568, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.723118342747284}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.0001318685794969069, "log_rms_mae_db": 1.5966766719490937, "rms_envelope_nrmse": 0.06111573371670486, "envelope_correlation": 0.9968626463974923, "rms_gain_error_db": -0.07810740155878389, "reference_mean_power": 0.0021420152585098387, "output_mean_power": 0.002103835698246685}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 1.3134511495070254e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003582591828573965, "output_power": 0.0003354399343844885, "reference_active": true, "delta_db": -0.2858256655428176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.463795601086152e-05, "output_power": 4.045936792867074e-05, "reference_active": true, "delta_db": -0.4268520609632031}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.556588167851325e-06, "output_power": 1.2109191244094578e-05, "reference_active": true, "delta_db": 2.0478938426906725}], "active_band_count": 3, "active_band_mae_db": 0.9201905230655644}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 0.00012588557284836115, "log_rms_mae_db": 0.8951865379041314, "rms_envelope_nrmse": 0.03240225718046062, "envelope_correlation": 0.999136879057691, "rms_gain_error_db": -0.0779761150081727, "reference_mean_power": 0.0021571542863483557, "output_mean_power": 0.002118768935161296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 1.2592546923162062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031562831458090785, "output_power": 0.00029590351500447676, "reference_active": true, "delta_db": -0.2802583211773245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9876750850691092e-05, "output_power": 1.962461562989803e-05, "reference_active": true, "delta_db": -0.05544234641669254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602361817361658e-07, "output_power": 7.23539322690666e-07, "reference_active": true, "delta_db": 0.397628184922969}], "active_band_count": 3, "active_band_mae_db": 0.24444295083899537}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.00010444605382463925, "log_rms_mae_db": 1.907344956794267, "rms_envelope_nrmse": 0.1213067897572695, "envelope_correlation": 0.9904007129507674, "rms_gain_error_db": -0.27135195933589384, "reference_mean_power": 0.0005604503758559807, "output_mean_power": 0.0005265043562122542}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 5.643164593552106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025020135526255573, "output_power": 0.00022169555882814302, "reference_active": true, "delta_db": -0.525326647127149}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.833230380435073e-05, "output_power": 2.7288570590034236e-05, "reference_active": true, "delta_db": -1.475841341403069}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.791716397073579e-06, "output_power": 9.668379999154645e-06, "reference_active": true, "delta_db": 1.5337416857203805}], "active_band_count": 3, "active_band_mae_db": 1.1783032247501997}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 9.32306023645191e-05, "log_rms_mae_db": 1.3296357080831647, "rms_envelope_nrmse": 0.07014670658620235, "envelope_correlation": 0.9971182359648436, "rms_gain_error_db": -0.27135751635865857, "reference_mean_power": 0.0005644391167624198, "output_mean_power": 0.0005302508238693134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 5.412619397247975e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022898404327685652, "output_power": 0.0002037702829137237, "reference_active": true, "delta_db": -0.5066437121411486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5949246143041614e-05, "output_power": 1.2552964180184133e-05, "reference_active": true, "delta_db": -1.0399387094870163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.408378605310095e-07, "output_power": 6.630403938162951e-07, "reference_active": true, "delta_db": 0.1479182550757821}], "active_band_count": 3, "active_band_mae_db": 0.5648335589013156}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 4.754025004833877e-05, "log_rms_mae_db": 1.9792256902459724, "rms_envelope_nrmse": 0.1658884494423637, "envelope_correlation": 0.9832334357292036, "rms_gain_error_db": -0.4476599875208207, "reference_mean_power": 8.438545283737632e-05, "output_mean_power": 7.612049197824663e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 6.948180578959798e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5558932008941216e-05, "output_power": 2.9486457426073485e-05, "reference_active": true, "delta_db": -0.8132611021932088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9367090695871985e-06, "output_power": 3.2737532643639914e-06, "reference_active": true, "delta_db": -0.8008737629168594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.57982090332121e-07, "output_power": 6.287223640957697e-07, "reference_active": true, "delta_db": -0.812000354203646}], "active_band_count": 3, "active_band_mae_db": 0.8087117397712381}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 4.260586736236329e-05, "log_rms_mae_db": 1.357896374296084, "rms_envelope_nrmse": 0.10812072113943746, "envelope_correlation": 0.9936865295174683, "rms_gain_error_db": -0.4476738693641523, "reference_mean_power": 8.498589632737057e-05, "output_mean_power": 7.6661881213356e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 6.685624503092077e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2789374619144395e-05, "output_power": 2.728954481624563e-05, "reference_active": true, "delta_db": -0.7973684180239277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.67867676799424e-06, "output_power": 1.3832120023572997e-06, "reference_active": true, "delta_db": -0.8407833133601244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.924150120657105e-08, "output_power": 4.823436917970659e-08, "reference_active": true, "delta_db": -0.8926945211498026}], "active_band_count": 3, "active_band_mae_db": 0.8436154175112849}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 2.7803794970768208e-05, "log_rms_mae_db": 1.765818852192894, "rms_envelope_nrmse": 0.17294568511268438, "envelope_correlation": 0.9820036888854156, "rms_gain_error_db": -0.05855799544481274, "reference_mean_power": 9.160796502339738e-06, "output_mean_power": 9.038106121039806e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 3.7361073674013695e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6569760918264083e-06, "output_power": 1.4869153286311984e-06, "reference_active": true, "delta_db": -0.47030003491073213}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.026752210402781e-07, "output_power": 5.237722929569063e-07, "reference_active": true, "delta_db": -0.6094081489778853}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8917938491313036e-07, "output_power": 1.0734158153587053e-07, "reference_active": true, "delta_db": -2.4610581934249924}], "active_band_count": 3, "active_band_mae_db": 1.1802554591045367}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 2.058581204447328e-05, "log_rms_mae_db": 1.2184236046221637, "rms_envelope_nrmse": 0.096709374209627, "envelope_correlation": 0.9941751892331114, "rms_gain_error_db": -0.05857902490210002, "reference_mean_power": 9.225978949789671e-06, "output_mean_power": 9.102371505439446e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 3.5247411287487305e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7104686599805127e-06, "output_power": 1.5110235524399355e-06, "reference_active": true, "delta_db": -0.5384388747097232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2748792658937344e-07, "output_power": 2.667333038363109e-07, "reference_active": true, "delta_db": -0.8911804930698486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.947253788526217e-08, "output_power": 8.619521001142355e-09, "reference_active": true, "delta_db": -3.5393942532191507}], "active_band_count": 3, "active_band_mae_db": 1.656337873666241}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8236360999283469, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "300bdaef071478f1ab7befa0ef402cc43d33473fa3e427fa33620c86b4e6e95c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9848891135169872, "standard_deviation": 0.0015579243866272556, "minimum": 0.9831745743140063, "maximum": 0.9874366399950066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.975966297287486, "standard_deviation": 0.05484506369333226, "minimum": 1.9229712513945232, "maximum": 2.0923358299467063, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.945191559836567, "standard_deviation": 0.284453378164736, "minimum": 0.49945033057811866, "maximum": 1.2032529901444804, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1631665723557246, "standard_deviation": 0.009953887487500977, "minimum": 0.14374788795283966, "maximum": 0.1737217499350324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945229064502997, "standard_deviation": 0.0007435269625404747, "minimum": 0.9932022130560101, "maximum": 0.9952271005151335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.355269914153904, "standard_deviation": 0.05164414066766487, "minimum": 1.2778489875127919, "maximum": 1.4458900085430868, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.012857679203872, "standard_deviation": 0.2135956632953367, "minimum": 0.684013530857111, "maximum": 1.286829899104748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11100655418523447, "standard_deviation": 0.008688628437821186, "minimum": 0.09614880864996432, "maximum": 0.12391621494154001, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9798930310939007, "standard_deviation": 0.0062018838500606, "minimum": 0.9680084836898183, "maximum": 0.9877519233666139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7668167406848254, "standard_deviation": 0.02048959849632046, "minimum": 1.7429607107739526, "maximum": 1.7995545613255781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.4976171604008246, "standard_deviation": 0.520879488019974, "minimum": 0.9667666451227905, "maximum": 2.5921887388811506, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19584147379654795, "standard_deviation": 0.028353851506563964, "minimum": 0.16013675690325838, "maximum": 0.23566677153017404, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9939558362366887, "standard_deviation": 0.0015547526901324275, "minimum": 0.9920122099673517, "maximum": 0.9961985265437807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2062390475748168, "standard_deviation": 0.026874573674804184, "minimum": 1.1569993858871597, "maximum": 1.245540909821216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.052633794833507, "standard_deviation": 0.5246110892493717, "minimum": 1.656337873666241, "maximum": 3.1634673707610026, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1278977568415058, "standard_deviation": 0.028913670119152392, "minimum": 0.07867596507730557, "maximum": 0.1617393004297591, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9908967606060174, "standard_deviation": 0.0021840185447589825, "minimum": 0.9865617047533098, "maximum": 0.9941054161646766, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9618663599115607, "standard_deviation": 0.053341740695972355, "minimum": 1.907344956794267, "maximum": 2.0517427764830587, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0375892241323001, "standard_deviation": 0.4270611913572244, "minimum": 0.447087504407837, "maximum": 1.7915024327534745, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11718246287075756, "standard_deviation": 0.014322869416608682, "minimum": 0.09373522851595424, "maximum": 0.1433409603899735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973407186764989, "standard_deviation": 0.0009350398496836213, "minimum": 0.9953185289237009, "maximum": 0.9984995859446925, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3968634566113687, "standard_deviation": 0.06280323051432095, "minimum": 1.3296357080831647, "maximum": 1.5136362195547675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0274329293398878, "standard_deviation": 0.432961913991523, "minimum": 0.5648335589013156, "maximum": 1.5858766476754138, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06607593755103577, "standard_deviation": 0.011920473371785692, "minimum": 0.04762441022991721, "maximum": 0.08850721314793865, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.35167231771468493, "standard_deviation": 0.10969876578644898, "minimum": 0.16094727491074015, "maximum": 0.4984862151987, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.732748312269507, "standard_deviation": 0.014310798369790685, "minimum": 5.720013294265613, "maximum": 5.765485755198532, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31082882473009765, "standard_deviation": 0.020481473652002122, "minimum": 0.28178313223528484, "maximum": 0.3448369042589772, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.089088258505865, "standard_deviation": 0.03460765698520762, "minimum": -0.13343968397304087, "maximum": -0.04655760451866217, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9970156039883606, "standard_deviation": 0.0004537292128132122, "minimum": 0.9962702468885614, "maximum": 0.9975768650966544, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5911491547563976, "standard_deviation": 0.04849951268041602, "minimum": 1.5074954041881523, "maximum": 1.6448935920212908, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5882891361096656, "standard_deviation": 0.4318482136103332, "minimum": 0.23287331658489271, "maximum": 1.2979467355049588, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.059787275944704454, "standard_deviation": 0.004567730584186762, "minimum": 0.053901674355691605, "maximum": 0.0674490600542352, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992207462630655, "standard_deviation": 0.0001903347927338902, "minimum": 0.9988988495089166, "maximum": 0.9994839513086444, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.863756156619427, "standard_deviation": 0.038101100677688446, "minimum": 0.8129352184999888, "maximum": 0.9292243320404611, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4327397239090094, "standard_deviation": 0.3078850329123771, "minimum": 0.0911513995207181, "maximum": 0.9582589592966612, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031256082282789976, "standard_deviation": 0.004175259893294787, "minimum": 0.025622108062445914, "maximum": 0.03822420241996642, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1188-133604-0037", "language": "en", "style": "normal", "speaker": "librispeech_1188", "source_utterance_id": "1188-133604-0037", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1188-133604-0037.wav", "source_file_sha256": "49b13756dfd2a24f87acc0f1dce00bceea9cd6177c6d66d1fdffa8ac7a7bbbdd", "decoded_pcm_sha256": "55ff59cb08cac135da664331bc418f7f430ca702cefbb6986e7121d9fe94ca8b", "original_source_sha256": "94c0626f16e39947864688eb2721a57f838b43caeb1857767a9b4381807368c8", "corpus_entry_id": "librispeech_1188-133604-0037"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.968539237976074, "predicted_mos": 3.968539237976074, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fbb58b0c21de840cb4e115a0c73146db05aa36a4ff3b9fdd97c318dbb5e3b5df", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.172091200016439, "cpu_threads": 4}, "waveform_sha256": "9f907edaf27e98e2faf5d93890f689b570be9341564ab553dac6f0a537616da3"}, "rows": [{"case": "librispeech_1188-133604-0037", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.9106764793395996, "predicted_mos": 2.9106764793395996, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "453b8cec7f105ddb345db5c9c65eb13d8260d3e01d914b0be35486177cdb2ef6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1538794999942183, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.177233570594949, "全帯域_10ms_rms_envelope_nrmse": 0.05600768045906749, "全帯域_10ms_envelope_correlation": 0.9968521899073424, "全帯域_10ms_modulation_mae_db": 0.1380338753443702, "全帯域_40ms_log_rms_mae_db": 0.6583020479055843, "全帯域_40ms_rms_envelope_nrmse": 0.03210691669818344, "全帯域_40ms_envelope_correlation": 0.9989473989200476, "全帯域_40ms_modulation_mae_db": 0.21375987664788196, "300–1000 Hz_10ms_log_rms_mae_db": 1.7137685625477117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09180089528280892, "300–1000 Hz_10ms_envelope_correlation": 0.9938371111289728, "300–1000 Hz_10ms_modulation_mae_db": 0.3127854857701066, "300–1000 Hz_40ms_log_rms_mae_db": 1.2154473251175233, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05687311259766186, "300–1000 Hz_40ms_envelope_correlation": 0.997653632704787, "300–1000 Hz_40ms_modulation_mae_db": 0.253335986437716, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6330250134647648, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11155772466197313, "1000–3000 Hz_10ms_envelope_correlation": 0.9918872344550212, "1000–3000 Hz_10ms_modulation_mae_db": 1.01008000049008, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9808152428304021, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05825715223735855, "1000–3000 Hz_40ms_envelope_correlation": 0.9979212615921282, "1000–3000 Hz_40ms_modulation_mae_db": 0.9372975047109501, "3000–6000 Hz_10ms_log_rms_mae_db": 1.823044451603997, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1721291586159966, "3000–6000 Hz_10ms_envelope_correlation": 0.9811917413999027, "3000–6000 Hz_10ms_modulation_mae_db": 1.5974881935484522, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2151245277181661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0910616078074702, "3000–6000 Hz_40ms_envelope_correlation": 0.9947034061341553, "3000–6000 Hz_40ms_modulation_mae_db": 1.8627389322618442, "mr_stft_db": 5.5359999341052095, "band_mae_db": 0.11460959659494287, "rms_gain_error_db": -0.06029006980304885, "output_peak": 0.4380494371772787}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.227827868641085, "全帯域_10ms_rms_envelope_nrmse": 0.05523175871252692, "全帯域_10ms_envelope_correlation": 0.996953164182809, "全帯域_10ms_modulation_mae_db": 0.122844831818668, "全帯域_40ms_log_rms_mae_db": 0.6157133602457738, "全帯域_40ms_rms_envelope_nrmse": 0.026328200669421098, "全帯域_40ms_envelope_correlation": 0.9993272443265366, "全帯域_40ms_modulation_mae_db": 0.17057758368804724, "300–1000 Hz_10ms_log_rms_mae_db": 1.6277348281815096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08191454115569739, "300–1000 Hz_10ms_envelope_correlation": 0.9951825489973689, "300–1000 Hz_10ms_modulation_mae_db": 0.42887631385511743, "300–1000 Hz_40ms_log_rms_mae_db": 1.13105336523002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04914007236914171, "300–1000 Hz_40ms_envelope_correlation": 0.9983607774100357, "300–1000 Hz_40ms_modulation_mae_db": 0.6397268906596625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7413176009041722, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11011598950251218, "1000–3000 Hz_10ms_envelope_correlation": 0.9920191909878204, "1000–3000 Hz_10ms_modulation_mae_db": 0.8394895358515607, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0965916478328321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06282670487081127, "1000–3000 Hz_40ms_envelope_correlation": 0.9974215469917197, "1000–3000 Hz_40ms_modulation_mae_db": 0.7709977236271772, "3000–6000 Hz_10ms_log_rms_mae_db": 1.814429047691562, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1533846620877619, "3000–6000 Hz_10ms_envelope_correlation": 0.9850941097445882, "3000–6000 Hz_10ms_modulation_mae_db": 0.4970201467714462, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2066813989101555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08826781990466172, "3000–6000 Hz_40ms_envelope_correlation": 0.9948765379028633, "3000–6000 Hz_40ms_modulation_mae_db": 0.765117971207466, "mr_stft_db": 5.556298614253303, "band_mae_db": 0.14606550548837846, "rms_gain_error_db": -0.0681579224511475, "output_peak": 0.4507915604024877}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.254210137585313, "全帯域_10ms_rms_envelope_nrmse": 0.06532646006212217, "全帯域_10ms_envelope_correlation": 0.9957174420809561, "全帯域_10ms_modulation_mae_db": 0.21789632023419755, "全帯域_40ms_log_rms_mae_db": 0.6571514815411028, "全帯域_40ms_rms_envelope_nrmse": 0.03752293330017644, "全帯域_40ms_envelope_correlation": 0.9985669667493436, "全帯域_40ms_modulation_mae_db": 0.2120737745614016, "300–1000 Hz_10ms_log_rms_mae_db": 1.6394792862444922, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08252776840510341, "300–1000 Hz_10ms_envelope_correlation": 0.9949784809423735, "300–1000 Hz_10ms_modulation_mae_db": 0.2017425889283717, "300–1000 Hz_40ms_log_rms_mae_db": 1.073390305104027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04945133930547846, "300–1000 Hz_40ms_envelope_correlation": 0.9981730297988322, "300–1000 Hz_40ms_modulation_mae_db": 0.31899629857682704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.646184613277427, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11292751946818783, "1000–3000 Hz_10ms_envelope_correlation": 0.9920617970895923, "1000–3000 Hz_10ms_modulation_mae_db": 1.0394534714587094, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0203535446815413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06922491966224731, "1000–3000 Hz_40ms_envelope_correlation": 0.9974145891160041, "1000–3000 Hz_40ms_modulation_mae_db": 1.0619710449168935, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8015369017998852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16998871129385673, "3000–6000 Hz_10ms_envelope_correlation": 0.9816796428530626, "3000–6000 Hz_10ms_modulation_mae_db": 0.5842699739299452, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1892525846732878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08356111351512646, "3000–6000 Hz_40ms_envelope_correlation": 0.9956279657522951, "3000–6000 Hz_40ms_modulation_mae_db": 0.6360551384882843, "mr_stft_db": 5.545436927748976, "band_mae_db": 0.15801980338497068, "rms_gain_error_db": -0.06976599263917835, "output_peak": 0.4347525222548472}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2003516206442804, "全帯域_10ms_rms_envelope_nrmse": 0.05944585814685546, "全帯域_10ms_envelope_correlation": 0.9964529361420943, "全帯域_10ms_modulation_mae_db": 0.30738441117216175, "全帯域_40ms_log_rms_mae_db": 0.6315398841852903, "全帯域_40ms_rms_envelope_nrmse": 0.02992437320032993, "全帯域_40ms_envelope_correlation": 0.9990953639843778, "全帯域_40ms_modulation_mae_db": 0.3477224144121989, "300–1000 Hz_10ms_log_rms_mae_db": 1.7102815487261138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08801755759959069, "300–1000 Hz_10ms_envelope_correlation": 0.9942886102135812, "300–1000 Hz_10ms_modulation_mae_db": 0.4671406807116261, "300–1000 Hz_40ms_log_rms_mae_db": 1.1921134355484453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050153173100528116, "300–1000 Hz_40ms_envelope_correlation": 0.9981453980351083, "300–1000 Hz_40ms_modulation_mae_db": 0.46319533277747754, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6849164924569129, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10811273203425432, "1000–3000 Hz_10ms_envelope_correlation": 0.9925059600461815, "1000–3000 Hz_10ms_modulation_mae_db": 0.6521042582293848, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0429600903597593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06508767968232798, "1000–3000 Hz_40ms_envelope_correlation": 0.997470294842804, "1000–3000 Hz_40ms_modulation_mae_db": 0.6653694784205509, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7777349763838455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1558926636441548, "3000–6000 Hz_10ms_envelope_correlation": 0.9845977809423797, "3000–6000 Hz_10ms_modulation_mae_db": 0.7880467094748509, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1824592779169543, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08004047548265365, "3000–6000 Hz_40ms_envelope_correlation": 0.9958126822296328, "3000–6000 Hz_40ms_modulation_mae_db": 1.0919597277439979, "mr_stft_db": 5.545422104253109, "band_mae_db": 0.15051946288020693, "rms_gain_error_db": -0.05917350731361047, "output_peak": 0.43269856907542686}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.23319305830915, "全帯域_10ms_rms_envelope_nrmse": 0.05978596245696582, "全帯域_10ms_envelope_correlation": 0.9963991712371337, "全帯域_10ms_modulation_mae_db": 0.08114572239004403, "全帯域_40ms_log_rms_mae_db": 0.6217804806037721, "全帯域_40ms_rms_envelope_nrmse": 0.030910276411046152, "全帯域_40ms_envelope_correlation": 0.9990230757884215, "全帯域_40ms_modulation_mae_db": 0.18184612857183094, "300–1000 Hz_10ms_log_rms_mae_db": 1.7096706321793043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09400386215636772, "300–1000 Hz_10ms_envelope_correlation": 0.9934609410869859, "300–1000 Hz_10ms_modulation_mae_db": 0.0797738762873581, "300–1000 Hz_40ms_log_rms_mae_db": 1.132768425674397, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05317493663527154, "300–1000 Hz_40ms_envelope_correlation": 0.9978939456239132, "300–1000 Hz_40ms_modulation_mae_db": 0.022288972732275873, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6501304225467956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11067974807314034, "1000–3000 Hz_10ms_envelope_correlation": 0.9922743033107461, "1000–3000 Hz_10ms_modulation_mae_db": 0.6193712183388485, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0305006796263347, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05856858221896568, "1000–3000 Hz_40ms_envelope_correlation": 0.9982215696798729, "1000–3000 Hz_40ms_modulation_mae_db": 0.6974442981644161, "3000–6000 Hz_10ms_log_rms_mae_db": 1.817239854899426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15920656840803488, "3000–6000 Hz_10ms_envelope_correlation": 0.9839225877942368, "3000–6000 Hz_10ms_modulation_mae_db": 1.1144707317246374, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2065601709009464, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08449444907745302, "3000–6000 Hz_40ms_envelope_correlation": 0.9953256874174143, "3000–6000 Hz_40ms_modulation_mae_db": 1.3932366522548467, "mr_stft_db": 5.548782834110715, "band_mae_db": 0.1287431945198146, "rms_gain_error_db": -0.053740114560842725, "output_peak": 0.4465953410832331}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2413540744382738, "全帯域_10ms_rms_envelope_nrmse": 0.05602835805247261, "全帯域_10ms_envelope_correlation": 0.996894561985935, "全帯域_10ms_modulation_mae_db": 0.1410927783352562, "全帯域_40ms_log_rms_mae_db": 0.6441865576231337, "全帯域_40ms_rms_envelope_nrmse": 0.032086080667694666, "全帯域_40ms_envelope_correlation": 0.9989887346860162, "全帯域_40ms_modulation_mae_db": 0.1792346528747447, "300–1000 Hz_10ms_log_rms_mae_db": 1.7634936415788836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09388203956484985, "300–1000 Hz_10ms_envelope_correlation": 0.9934727581117543, "300–1000 Hz_10ms_modulation_mae_db": 0.4249120018821279, "300–1000 Hz_40ms_log_rms_mae_db": 1.2795876590801172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06136552981033001, "300–1000 Hz_40ms_envelope_correlation": 0.997145189506807, "300–1000 Hz_40ms_modulation_mae_db": 0.6782331406514205, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6604608255033997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10614909703210279, "1000–3000 Hz_10ms_envelope_correlation": 0.9926611732213548, "1000–3000 Hz_10ms_modulation_mae_db": 1.304906480131948, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0305146152767282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05966468488911716, "1000–3000 Hz_40ms_envelope_correlation": 0.997780507251499, "1000–3000 Hz_40ms_modulation_mae_db": 0.7651502847782066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8190858532174763, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16549016916500914, "3000–6000 Hz_10ms_envelope_correlation": 0.9834274157264229, "3000–6000 Hz_10ms_modulation_mae_db": 1.3687544998546926, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1828149638213594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1039655699306394, "3000–6000 Hz_40ms_envelope_correlation": 0.9941662341077692, "3000–6000 Hz_40ms_modulation_mae_db": 1.4935394163012257, "mr_stft_db": 5.550520933232896, "band_mae_db": 0.1423769409140148, "rms_gain_error_db": -0.07530448136731717, "output_peak": 0.43509173366310167}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2249990533903763, "全帯域_10ms_rms_envelope_nrmse": 0.059704702755340164, "全帯域_10ms_envelope_correlation": 0.9963922882021654, "全帯域_10ms_modulation_mae_db": 0.190093896502974, "全帯域_40ms_log_rms_mae_db": 0.7019147534932696, "全帯域_40ms_rms_envelope_nrmse": 0.03133674306622266, "全帯域_40ms_envelope_correlation": 0.9989638133829607, "全帯域_40ms_modulation_mae_db": 0.3633669769098528, "300–1000 Hz_10ms_log_rms_mae_db": 1.7781449352514862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10031253397644159, "300–1000 Hz_10ms_envelope_correlation": 0.9925671373177299, "300–1000 Hz_10ms_modulation_mae_db": 0.3848599763866261, "300–1000 Hz_40ms_log_rms_mae_db": 1.2633930250164203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05968284328102559, "300–1000 Hz_40ms_envelope_correlation": 0.9973415885330553, "300–1000 Hz_40ms_modulation_mae_db": 0.4617903272746879, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6434211404197094, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12929224995287963, "1000–3000 Hz_10ms_envelope_correlation": 0.9896286365241616, "1000–3000 Hz_10ms_modulation_mae_db": 1.9792112148125127, "1000–3000 Hz_40ms_log_rms_mae_db": 1.084584488359537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08583423726153389, "1000–3000 Hz_40ms_envelope_correlation": 0.9959353010434322, "1000–3000 Hz_40ms_modulation_mae_db": 1.8367292593740352, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8144055948739175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1686376708673674, "3000–6000 Hz_10ms_envelope_correlation": 0.9821238808891475, "3000–6000 Hz_10ms_modulation_mae_db": 1.038683159147296, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2510292151518672, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09875916114907461, "3000–6000 Hz_40ms_envelope_correlation": 0.9936253116182271, "3000–6000 Hz_40ms_modulation_mae_db": 1.3483149628060493, "mr_stft_db": 5.558392879227308, "band_mae_db": 0.16135883572836832, "rms_gain_error_db": -0.026300932258349834, "output_peak": 0.43444078757348537}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2726240192178626, "全帯域_10ms_rms_envelope_nrmse": 0.06344655612722137, "全帯域_10ms_envelope_correlation": 0.9959990264824957, "全帯域_10ms_modulation_mae_db": 0.1156161072106437, "全帯域_40ms_log_rms_mae_db": 0.739455289381774, "全帯域_40ms_rms_envelope_nrmse": 0.03426838500243791, "全帯域_40ms_envelope_correlation": 0.9988452518464317, "全帯域_40ms_modulation_mae_db": 0.3669865053643178, "300–1000 Hz_10ms_log_rms_mae_db": 1.6445004115019004, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0858545881617189, "300–1000 Hz_10ms_envelope_correlation": 0.9948642121766954, "300–1000 Hz_10ms_modulation_mae_db": 0.7367594140822935, "300–1000 Hz_40ms_log_rms_mae_db": 1.0942219694171784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055149712560079785, "300–1000 Hz_40ms_envelope_correlation": 0.9980784695211465, "300–1000 Hz_40ms_modulation_mae_db": 0.7543745394617244, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6583500553074633, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11084362050786399, "1000–3000 Hz_10ms_envelope_correlation": 0.992109278744866, "1000–3000 Hz_10ms_modulation_mae_db": 0.8742747477240967, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0344147283168055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06731513914880023, "1000–3000 Hz_40ms_envelope_correlation": 0.9972731408575575, "1000–3000 Hz_40ms_modulation_mae_db": 1.0113511934555248, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8001196916712336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16554302548093056, "3000–6000 Hz_10ms_envelope_correlation": 0.9826789521533267, "3000–6000 Hz_10ms_modulation_mae_db": 0.5840618862450015, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2035324047207792, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09364747789905008, "3000–6000 Hz_40ms_envelope_correlation": 0.9942211037953056, "3000–6000 Hz_40ms_modulation_mae_db": 0.6866129886697174, "mr_stft_db": 5.541643419542129, "band_mae_db": 0.1326443738218921, "rms_gain_error_db": -0.08988381424568316, "output_peak": 0.40700550135643343}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.789750083268836, "spectral_convergence": 0.181370771984219, "spectral_convergence_status": "ok", "amplitude_floor": 1.0507333742302167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.347473114455953, "spectral_convergence": 0.17112896393958843, "spectral_convergence_status": "ok", "amplitude_floor": 5.595595188050952e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.470776604590838, "spectral_convergence": 0.17799467158879165, "spectral_convergence_status": "ok", "amplitude_floor": 5.269142735102521e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007416076491485813, "output_power": 0.0007500912743386426, "delta_db": 0.04939912837127517, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001352027771357195, "output_power": 0.0013288690829865307, "delta_db": -0.07503414939572085, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000371703582440024, "output_power": 0.00035586918834811714, "delta_db": -0.18906358992312397, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018903159721465413, "output_power": 0.0001833136360567357, "delta_db": -0.13339632100462542, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5427403014296353e-05, "output_power": 1.4681584518303934e-05, "delta_db": -0.21519895278299392, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1014091243658427e-05, "output_power": 2.089075182510638e-05, "delta_db": -0.02556543809191791, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11460959659494287, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06029006980304885, "pitch": null, "output_rms_dbfs": -25.761456359376638, "output_peak": 0.4380494371772787, "reference_rms_dbfs": -25.701166289573592, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5359999341052095}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018114550052554645, "log_rms_mae_db": 1.177233570594949, "rms_envelope_nrmse": 0.05600768045906749, "envelope_correlation": 0.9968521899073424, "rms_gain_error_db": -0.060015260222981884, "reference_mean_power": 0.002692702060931978, "output_mean_power": 0.0026557474718423343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0110233554726448e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033858489572855915, "output_power": 0.0003329422796291282, "reference_active": true, "delta_db": -0.0729863154164108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.201391758025471e-05, "output_power": 3.909834130767662e-05, "reference_active": true, "delta_db": -0.31234845685605894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.1566777501091025e-06, "output_power": 7.204239590723325e-06, "reference_active": true, "delta_db": 0.028766853760640816}], "active_band_count": 3, "active_band_mae_db": 0.1380338753443702}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015936923488846764, "log_rms_mae_db": 0.6583020479055843, "rms_envelope_nrmse": 0.03210691669818344, "envelope_correlation": 0.9989473989200476, "rms_gain_error_db": -0.05741403005457128, "reference_mean_power": 0.002702794865208283, "output_mean_power": 0.0026672988773422893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.583684100944911e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031173336044584235, "output_power": 0.0003065167815318437, "reference_active": true, "delta_db": -0.07329024449097286}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.830633609453067e-05, "output_power": 1.738421572045577e-05, "reference_active": true, "delta_db": -0.22446329014231384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.642076702713698e-07, "output_power": 5.024176892107419e-07, "reference_active": true, "delta_db": 0.3435260953103592}], "active_band_count": 3, "active_band_mae_db": 0.21375987664788196}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001259260583850441, "log_rms_mae_db": 1.7137685625477117, "rms_envelope_nrmse": 0.09180089528280892, "envelope_correlation": 0.9938371111289728, "rms_gain_error_db": -0.12123096876429956, "reference_mean_power": 0.0007105481356967998, "output_mean_power": 0.0006909878442916212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8068878391559104e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016083630318364556, "output_power": 0.00015410892874985493, "reference_active": true, "delta_db": -0.1855628073963262}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8008354908350678e-05, "output_power": 1.6881641256201537e-05, "reference_active": true, "delta_db": -0.28059374100568696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.325041103743567e-06, "output_power": 2.0855021608734553e-06, "reference_active": true, "delta_db": -0.4721999089083067}], "active_band_count": 3, "active_band_mae_db": 0.3127854857701066}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010786919997015104, "log_rms_mae_db": 1.2154473251175233, "rms_envelope_nrmse": 0.05687311259766186, "envelope_correlation": 0.997653632704787, "rms_gain_error_db": -0.12295949220574465, "reference_mean_power": 0.0007140165693307253, "output_mean_power": 0.0006940844916823596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6527252040517484e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000151264236566862, "output_power": 0.00014578834765227076, "reference_active": true, "delta_db": -0.16013446085066554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.688501064647705e-06, "output_power": 7.494309937913352e-06, "reference_active": true, "delta_db": -0.1111002883944867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2694181944274492e-07, "output_power": 2.0278564635710135e-07, "reference_active": true, "delta_db": -0.4887732100679956}], "active_band_count": 3, "active_band_mae_db": 0.253335986437716}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.418520257385562e-05, "log_rms_mae_db": 1.6330250134647648, "rms_envelope_nrmse": 0.11155772466197313, "envelope_correlation": 0.9918872344550212, "rms_gain_error_db": -0.17734562664204703, "reference_mean_power": 0.000153818152796868, "output_mean_power": 0.00014766345608809245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.418336470516266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4748514150479256e-05, "output_power": 2.2901930513173722e-05, "reference_active": true, "delta_db": -0.3367703727090534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.277056498060538e-06, "output_power": 2.7134060265345373e-06, "reference_active": true, "delta_db": -0.8196914365347014}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.913659986768058e-07, "output_power": 5.789995745391545e-07, "reference_active": true, "delta_db": -1.8737781922264851}], "active_band_count": 3, "active_band_mae_db": 1.01008000049008}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.421085502718712e-05, "log_rms_mae_db": 0.9808152428304021, "rms_envelope_nrmse": 0.05825715223735855, "envelope_correlation": 0.9979212615921282, "rms_gain_error_db": -0.17739711619357443, "reference_mean_power": 0.00015458728634615244, "output_mean_power": 0.00014840005501413577}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1974511373265086e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3402826997507283e-05, "output_power": 2.1406524577832427e-05, "reference_active": true, "delta_db": -0.38722158335514645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3826120793884328e-06, "output_power": 1.1602167432459485e-06, "reference_active": true, "delta_db": -0.7616121829926502}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6109211314667126e-08, "output_power": 3.1439947996320614e-08, "reference_active": true, "delta_db": -1.6630587477850536}], "active_band_count": 3, "active_band_mae_db": 0.9372975047109501}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.007885661618194e-05, "log_rms_mae_db": 1.823044451603997, "rms_envelope_nrmse": 0.1721291586159966, "envelope_correlation": 0.9811917413999027, "rms_gain_error_db": -0.1749447397847203, "reference_mean_power": 2.0928545188569684e-05, "output_mean_power": 2.0102245206041153e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4882901242606597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1348414996445462e-05, "output_power": 1.1033554598441706e-05, "reference_active": true, "delta_db": -0.12219760571640613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8308296968161944e-06, "output_power": 2.148463796680659e-06, "reference_active": true, "delta_db": -1.1978570304792355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.670421987321005e-07, "output_power": 2.998557041845737e-07, "reference_active": true, "delta_db": -3.4724099444497147}], "active_band_count": 3, "active_band_mae_db": 1.5974881935484522}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3258654185194873e-05, "log_rms_mae_db": 1.2151245277181661, "rms_envelope_nrmse": 0.0910616078074702, "envelope_correlation": 0.9947034061341553, "rms_gain_error_db": -0.17498787981496194, "reference_mean_power": 2.103354073620274e-05, "output_mean_power": 2.02028946400666e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.390796824652454e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1287290435083056e-05, "output_power": 1.0788225971756434e-05, "reference_active": true, "delta_db": -0.1963966541774141}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.514784832162418e-06, "output_power": 1.10084613650538e-06, "reference_active": true, "delta_db": -1.386243252228747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.433812491322156e-08, "output_power": 1.7628673469804495e-08, "reference_active": true, "delta_db": -4.005576890379372}], "active_band_count": 3, "active_band_mae_db": 1.8627389322618442}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8928639105966172, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "74492030f7efbb347359dd0174bf30fd2539cb397365e779586e9aba604897e4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.991893446797468, "standard_deviation": 0.0009509220371037192, "minimum": 0.9896286365241616, "maximum": 0.9926611732213548, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6647257704850806, "standard_deviation": 0.03453453176653256, "minimum": 1.6330250134647648, "maximum": 1.7413176009041722, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0398613658796427, "standard_deviation": 0.43887021909748386, "minimum": 0.6193712183388485, "maximum": 1.9792112148125127, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11245983515411428, "standard_deviation": 0.00711497327905634, "minimum": 0.10614909703210279, "maximum": 0.12929224995287963, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974297764218771, "standard_deviation": 0.00068153585251403, "minimum": 0.9959353010434322, "maximum": 0.9982215696798729, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0400918796604923, "standard_deviation": 0.03644822034685134, "minimum": 0.9808152428304021, "maximum": 1.0965916478328321, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9682888484309693, "standard_deviation": 0.3800537385588344, "minimum": 0.6653694784205509, "maximum": 1.8367292593740352, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06584738749639527, "standard_deviation": 0.009036089155241322, "minimum": 0.05825715223735855, "maximum": 0.08583423726153389, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9830895139378834, "standard_deviation": 0.001403655870494984, "minimum": 0.9811917413999027, "maximum": 0.9850941097445882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.808449546517668, "standard_deviation": 0.014809247592109389, "minimum": 1.7777349763838455, "maximum": 1.823044451603997, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9465994125870402, "standard_deviation": 0.40151638911510074, "minimum": 0.4970201467714462, "maximum": 1.5974881935484522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.163784078695389, "standard_deviation": 0.006856443452702517, "minimum": 0.1533846620877619, "maximum": 0.1721291586159966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947948661197079, "standard_deviation": 0.0007670956574668114, "minimum": 0.9936253116182271, "maximum": 0.9958126822296328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2046818179766894, "standard_deviation": 0.022284013917081003, "minimum": 1.1824592779169543, "maximum": 1.2510292151518672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1596969737166791, "standard_deviation": 0.4398256881064313, "minimum": 0.6360551384882843, "maximum": 1.8627389322618442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09047470934576614, "standard_deviation": 0.008091894432971362, "minimum": 0.08004047548265365, "maximum": 0.1039655699306394, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9940814749969328, "standard_deviation": 0.0009087744492654732, "minimum": 0.9925671373177299, "maximum": 0.9951825489973689, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6983842307764252, "standard_deviation": 0.05665838559221324, "minimum": 1.6277348281815096, "maximum": 1.7781449352514862, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.37960629223795345, "standard_deviation": 0.19483632689628533, "minimum": 0.0797738762873581, "maximum": 0.7367594140822935, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08978922328782231, "standard_deviation": 0.006354534000185234, "minimum": 0.08191454115569739, "maximum": 0.10031253397644159, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978490038917106, "standard_deviation": 0.0004310160057431586, "minimum": 0.997145189506807, "maximum": 0.9983607774100357, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.172746938773516, "standard_deviation": 0.07672267657799409, "minimum": 1.073390305104027, "maximum": 1.2795876590801172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44899268607147397, "standard_deviation": 0.24489384226070843, "minimum": 0.022288972732275873, "maximum": 0.7543745394617244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054373839957439635, "standard_deviation": 0.004702466207737035, "minimum": 0.04914007236914171, "maximum": 0.06136552981033001, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1417922141665736, "standard_deviation": 0.015738078118926992, "minimum": 0.11460959659494287, "maximum": 0.16135883572836832, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.547812205809206, "standard_deviation": 0.007379385572040817, "minimum": 5.5359999341052095, "maximum": 5.558392879227308, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4349281815732867, "standard_deviation": 0.012996999852885107, "minimum": 0.40700550135643343, "maximum": 0.4507915604024877, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06282710432989726, "standard_deviation": 0.018519627528738578, "minimum": -0.08988381424568316, "maximum": -0.026300932258349834, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964575975276164, "standard_deviation": 0.0004404616977477067, "minimum": 0.9957174420809561, "maximum": 0.996953164182809, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2289741753526613, "standard_deviation": 0.02982491759407817, "minimum": 1.177233570594949, "maximum": 1.2726240192178626, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.16426349287603942, "standard_deviation": 0.07195514417280051, "minimum": 0.08114572239004403, "maximum": 0.30738441117216175, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.059372167096571504, "standard_deviation": 0.0036250181964513147, "minimum": 0.05523175871252692, "maximum": 0.06532646006212217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998969731210517, "standard_deviation": 0.0002154908261934793, "minimum": 0.9985669667493436, "maximum": 0.9993272443265366, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6587554818724626, "standard_deviation": 0.04234328643447314, "minimum": 0.6157133602457738, "maximum": 0.739455289381774, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.25444598912878447, "standard_deviation": 0.08835140833928745, "minimum": 0.17057758368804724, "maximum": 0.3669865053643178, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03181048862693904, "standard_deviation": 0.00324003590926791, "minimum": 0.026328200669421098, "maximum": 0.03752293330017644, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0037", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.8838088512420654, "predicted_mos": 2.8838088512420654, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "af32293f9776587eee81ad1b8ec8b2f239720fa9dc5077f278113d06daf189ce", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0261987997218966, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2296511918820605, "全帯域_10ms_rms_envelope_nrmse": 0.05668806589474913, "全帯域_10ms_envelope_correlation": 0.9967772010385464, "全帯域_10ms_modulation_mae_db": 0.15152889089180877, "全帯域_40ms_log_rms_mae_db": 0.7009635176788671, "全帯域_40ms_rms_envelope_nrmse": 0.0333384470684582, "全帯域_40ms_envelope_correlation": 0.9988628306283159, "全帯域_40ms_modulation_mae_db": 0.26821342655041497, "300–1000 Hz_10ms_log_rms_mae_db": 1.594946310073317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09279927293303566, "300–1000 Hz_10ms_envelope_correlation": 0.9937109109628364, "300–1000 Hz_10ms_modulation_mae_db": 0.20361422465832968, "300–1000 Hz_40ms_log_rms_mae_db": 1.0635393964575262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05710758639752852, "300–1000 Hz_40ms_envelope_correlation": 0.9976500789158048, "300–1000 Hz_40ms_modulation_mae_db": 0.16147960831088318, "1000–3000 Hz_10ms_log_rms_mae_db": 1.510876395789798, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1101236536285143, "1000–3000 Hz_10ms_envelope_correlation": 0.9921101410251871, "1000–3000 Hz_10ms_modulation_mae_db": 1.1586820016547839, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8247507818249019, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05641419330518687, "1000–3000 Hz_40ms_envelope_correlation": 0.9980836864739997, "1000–3000 Hz_40ms_modulation_mae_db": 1.1254177456183727, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6764067179477053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1685517876752373, "3000–6000 Hz_10ms_envelope_correlation": 0.9820232581993492, "3000–6000 Hz_10ms_modulation_mae_db": 1.4076372903103884, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0585580216646564, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08890648051941825, "3000–6000 Hz_40ms_envelope_correlation": 0.9949551833996921, "3000–6000 Hz_40ms_modulation_mae_db": 1.721747187030264, "mr_stft_db": 5.438171243051333, "band_mae_db": 0.12531641988596978, "rms_gain_error_db": -0.062041893259638346, "output_peak": 0.44072800437636034}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2637664668992732, "全帯域_10ms_rms_envelope_nrmse": 0.055105438960454445, "全帯域_10ms_envelope_correlation": 0.9969651111242712, "全帯域_10ms_modulation_mae_db": 0.11379348135846314, "全帯域_40ms_log_rms_mae_db": 0.6727049789622294, "全帯域_40ms_rms_envelope_nrmse": 0.0260893375126243, "全帯域_40ms_envelope_correlation": 0.9993380844622084, "全帯域_40ms_modulation_mae_db": 0.1971551130630157, "300–1000 Hz_10ms_log_rms_mae_db": 1.5386363179335778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08309755610580667, "300–1000 Hz_10ms_envelope_correlation": 0.995014602040381, "300–1000 Hz_10ms_modulation_mae_db": 0.30069524980550294, "300–1000 Hz_40ms_log_rms_mae_db": 0.976092771346653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05097136544587473, "300–1000 Hz_40ms_envelope_correlation": 0.9981924431901, "300–1000 Hz_40ms_modulation_mae_db": 0.4828366896735916, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5774025775250229, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10859580387137167, "1000–3000 Hz_10ms_envelope_correlation": 0.9922760755111102, "1000–3000 Hz_10ms_modulation_mae_db": 0.8529266867120432, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9042469940885142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06424124846315957, "1000–3000 Hz_40ms_envelope_correlation": 0.9973501539722663, "1000–3000 Hz_40ms_modulation_mae_db": 0.8297390828624444, "3000–6000 Hz_10ms_log_rms_mae_db": 1.635683552160534, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15256383613408908, "3000–6000 Hz_10ms_envelope_correlation": 0.9853948633216699, "3000–6000 Hz_10ms_modulation_mae_db": 0.4107466833530589, "3000–6000 Hz_40ms_log_rms_mae_db": 1.035703188949508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09001299022827136, "3000–6000 Hz_40ms_envelope_correlation": 0.9947656000523727, "3000–6000 Hz_40ms_modulation_mae_db": 0.639119869748797, "mr_stft_db": 5.454567540379283, "band_mae_db": 0.14979779218743824, "rms_gain_error_db": -0.06596571074966538, "output_peak": 0.45095012150643193}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2592238991439515, "全帯域_10ms_rms_envelope_nrmse": 0.06511356893629126, "全帯域_10ms_envelope_correlation": 0.9957467803076822, "全帯域_10ms_modulation_mae_db": 0.20227775182012642, "全帯域_40ms_log_rms_mae_db": 0.6829936757142697, "全帯域_40ms_rms_envelope_nrmse": 0.03739053280122483, "全帯域_40ms_envelope_correlation": 0.9985766703324158, "全帯域_40ms_modulation_mae_db": 0.19617811549427766, "300–1000 Hz_10ms_log_rms_mae_db": 1.4781276140858766, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0843400356368816, "300–1000 Hz_10ms_envelope_correlation": 0.9947312718457322, "300–1000 Hz_10ms_modulation_mae_db": 0.20720619366737114, "300–1000 Hz_40ms_log_rms_mae_db": 0.8920879472122513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049228565035057424, "300–1000 Hz_40ms_envelope_correlation": 0.9981682461419777, "300–1000 Hz_40ms_modulation_mae_db": 0.3070898889170079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5452366526960513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1134305651980303, "1000–3000 Hz_10ms_envelope_correlation": 0.9920257342385371, "1000–3000 Hz_10ms_modulation_mae_db": 1.1199466372400673, "1000–3000 Hz_40ms_log_rms_mae_db": 0.878739601010016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06985158253190603, "1000–3000 Hz_40ms_envelope_correlation": 0.9974091384532084, "1000–3000 Hz_40ms_modulation_mae_db": 1.1386549257259382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6007590388306594, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16567988442282272, "3000–6000 Hz_10ms_envelope_correlation": 0.9827219099479533, "3000–6000 Hz_10ms_modulation_mae_db": 0.703650574126634, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0005816613114786, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08357211245929637, "3000–6000 Hz_40ms_envelope_correlation": 0.9958018241098752, "3000–6000 Hz_40ms_modulation_mae_db": 0.7430393793297595, "mr_stft_db": 5.44541187270557, "band_mae_db": 0.17498337060043026, "rms_gain_error_db": -0.06975612280727463, "output_peak": 0.43472035213238336}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2509231982576174, "全帯域_10ms_rms_envelope_nrmse": 0.05942556918618782, "全帯域_10ms_envelope_correlation": 0.9964482726712658, "全帯域_10ms_modulation_mae_db": 0.24598475357022717, "全帯域_40ms_log_rms_mae_db": 0.6737061561600458, "全帯域_40ms_rms_envelope_nrmse": 0.029498570293633362, "全帯域_40ms_envelope_correlation": 0.9991098642363249, "全帯域_40ms_modulation_mae_db": 0.24701818157197097, "300–1000 Hz_10ms_log_rms_mae_db": 1.5742753233760185, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08956859066972027, "300–1000 Hz_10ms_envelope_correlation": 0.9940620147921311, "300–1000 Hz_10ms_modulation_mae_db": 0.4332144648609153, "300–1000 Hz_40ms_log_rms_mae_db": 1.0088174656933804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0505412695802882, "300–1000 Hz_40ms_envelope_correlation": 0.998089534346447, "300–1000 Hz_40ms_modulation_mae_db": 0.43661930022591644, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5849409902443397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10801924596719527, "1000–3000 Hz_10ms_envelope_correlation": 0.9924789823308912, "1000–3000 Hz_10ms_modulation_mae_db": 0.5138319849532599, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9015622693517541, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06331616728581001, "1000–3000 Hz_40ms_envelope_correlation": 0.9975894916028123, "1000–3000 Hz_40ms_modulation_mae_db": 0.4141272736571872, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6054373730105398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1496601994817567, "3000–6000 Hz_10ms_envelope_correlation": 0.9858254163897441, "3000–6000 Hz_10ms_modulation_mae_db": 0.6968522859207357, "3000–6000 Hz_40ms_log_rms_mae_db": 1.052104380817859, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08302228513045823, "3000–6000 Hz_40ms_envelope_correlation": 0.9955290086500101, "3000–6000 Hz_40ms_modulation_mae_db": 1.012635481279437, "mr_stft_db": 5.450461039650766, "band_mae_db": 0.12181852076254156, "rms_gain_error_db": -0.04971034122794188, "output_peak": 0.4327011058629279}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2659699947005343, "全帯域_10ms_rms_envelope_nrmse": 0.05904368262849822, "全帯域_10ms_envelope_correlation": 0.9964927128828667, "全帯域_10ms_modulation_mae_db": 0.07328220694565593, "全帯域_40ms_log_rms_mae_db": 0.6501058275646727, "全帯域_40ms_rms_envelope_nrmse": 0.029875684913767762, "全帯域_40ms_envelope_correlation": 0.9990951648671216, "全帯域_40ms_modulation_mae_db": 0.12545348950136706, "300–1000 Hz_10ms_log_rms_mae_db": 1.6424163985725746, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09767987745391066, "300–1000 Hz_10ms_envelope_correlation": 0.9929208267325524, "300–1000 Hz_10ms_modulation_mae_db": 0.2741577310039987, "300–1000 Hz_40ms_log_rms_mae_db": 1.0115994462929663, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055700723446656907, "300–1000 Hz_40ms_envelope_correlation": 0.997665090333562, "300–1000 Hz_40ms_modulation_mae_db": 0.18578119471114848, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5198717471541132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10972857667609213, "1000–3000 Hz_10ms_envelope_correlation": 0.9924737610670957, "1000–3000 Hz_10ms_modulation_mae_db": 0.7212808883408313, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8489158539239983, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0574627468911845, "1000–3000 Hz_40ms_envelope_correlation": 0.998383892735007, "1000–3000 Hz_40ms_modulation_mae_db": 0.6522649641722477, "3000–6000 Hz_10ms_log_rms_mae_db": 1.653419536330639, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15793728490668563, "3000–6000 Hz_10ms_envelope_correlation": 0.9842398713714551, "3000–6000 Hz_10ms_modulation_mae_db": 1.05774327315581, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0502145411923196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0870889072002486, "3000–6000 Hz_40ms_envelope_correlation": 0.995121283363458, "3000–6000 Hz_40ms_modulation_mae_db": 1.3648458136761406, "mr_stft_db": 5.4560105861951085, "band_mae_db": 0.1343876503871004, "rms_gain_error_db": -0.05722226107079248, "output_peak": 0.4489208042186646}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2814778976410197, "全帯域_10ms_rms_envelope_nrmse": 0.054956211060167445, "全帯域_10ms_envelope_correlation": 0.9970130500024371, "全帯域_10ms_modulation_mae_db": 0.1890693105209947, "全帯域_40ms_log_rms_mae_db": 0.6591863174979409, "全帯域_40ms_rms_envelope_nrmse": 0.03168538187720482, "全帯域_40ms_envelope_correlation": 0.9990144223157692, "全帯域_40ms_modulation_mae_db": 0.2800482356459555, "300–1000 Hz_10ms_log_rms_mae_db": 1.6487770881217994, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09401608211832457, "300–1000 Hz_10ms_envelope_correlation": 0.9934412347671494, "300–1000 Hz_10ms_modulation_mae_db": 0.3988189203647716, "300–1000 Hz_40ms_log_rms_mae_db": 1.0664573451790758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06013116433355889, "300–1000 Hz_40ms_envelope_correlation": 0.9972455810814418, "300–1000 Hz_40ms_modulation_mae_db": 0.6249578839377734, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5548582450122759, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10505170851090555, "1000–3000 Hz_10ms_envelope_correlation": 0.9929684957670659, "1000–3000 Hz_10ms_modulation_mae_db": 1.3104642679414489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8772646683280801, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06155956283190152, "1000–3000 Hz_40ms_envelope_correlation": 0.9978129676092541, "1000–3000 Hz_40ms_modulation_mae_db": 0.8185521739315352, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6447614808409405, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16518721664328262, "3000–6000 Hz_10ms_envelope_correlation": 0.983392024995077, "3000–6000 Hz_10ms_modulation_mae_db": 1.1435772138596325, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0293393010291418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1023052382832312, "3000–6000 Hz_40ms_envelope_correlation": 0.9942720036773398, "3000–6000 Hz_40ms_modulation_mae_db": 1.2824956916458317, "mr_stft_db": 5.457378751871487, "band_mae_db": 0.14458942604501143, "rms_gain_error_db": -0.07493026101056686, "output_peak": 0.4314742802888299}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2504390908904743, "全帯域_10ms_rms_envelope_nrmse": 0.06090487576588941, "全帯域_10ms_envelope_correlation": 0.9962437023065714, "全帯域_10ms_modulation_mae_db": 0.14152431114139805, "全帯域_40ms_log_rms_mae_db": 0.7539357079755662, "全帯域_40ms_rms_envelope_nrmse": 0.03367991174768178, "全帯域_40ms_envelope_correlation": 0.9988002420553407, "全帯域_40ms_modulation_mae_db": 0.30233185212626573, "300–1000 Hz_10ms_log_rms_mae_db": 1.6698700627484588, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10184597780988865, "300–1000 Hz_10ms_envelope_correlation": 0.9923166966114672, "300–1000 Hz_10ms_modulation_mae_db": 0.4001721819417719, "300–1000 Hz_40ms_log_rms_mae_db": 1.1325522192093993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06041785223665319, "300–1000 Hz_40ms_envelope_correlation": 0.9972510990881838, "300–1000 Hz_40ms_modulation_mae_db": 0.4118314594392732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5031002827387334, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12957024775276585, "1000–3000 Hz_10ms_envelope_correlation": 0.9897275785662251, "1000–3000 Hz_10ms_modulation_mae_db": 1.9895210108170527, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8956900258465759, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08806868266557825, "1000–3000 Hz_40ms_envelope_correlation": 0.9958478432097747, "1000–3000 Hz_40ms_modulation_mae_db": 1.8618664255349444, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6130779685605805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1635579249473054, "3000–6000 Hz_10ms_envelope_correlation": 0.9833202582596839, "3000–6000 Hz_10ms_modulation_mae_db": 1.037528345802334, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0597014238664817, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09811867022967379, "3000–6000 Hz_40ms_envelope_correlation": 0.9938537535682858, "3000–6000 Hz_40ms_modulation_mae_db": 1.3479944572015399, "mr_stft_db": 5.460151588804291, "band_mae_db": 0.1752831789383268, "rms_gain_error_db": -0.02254991702561354, "output_peak": 0.42917959277142786}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.311284470131398, "全帯域_10ms_rms_envelope_nrmse": 0.06415722806764625, "全帯域_10ms_envelope_correlation": 0.9959176814636478, "全帯域_10ms_modulation_mae_db": 0.12174758286760623, "全帯域_40ms_log_rms_mae_db": 0.772860785881102, "全帯域_40ms_rms_envelope_nrmse": 0.03586712127906506, "全帯域_40ms_envelope_correlation": 0.9987360215053018, "全帯域_40ms_modulation_mae_db": 0.3379725525623976, "300–1000 Hz_10ms_log_rms_mae_db": 1.5604281813772116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0903247523820881, "300–1000 Hz_10ms_envelope_correlation": 0.9942493023750902, "300–1000 Hz_10ms_modulation_mae_db": 0.7903843475761917, "300–1000 Hz_40ms_log_rms_mae_db": 0.9943768794752628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058020896379890004, "300–1000 Hz_40ms_envelope_correlation": 0.9978072288628063, "300–1000 Hz_40ms_modulation_mae_db": 0.8371951440423889, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5422962075497304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1131695865641735, "1000–3000 Hz_10ms_envelope_correlation": 0.9918918518146571, "1000–3000 Hz_10ms_modulation_mae_db": 0.9333901520059378, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8653713953486198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0706501852287482, "1000–3000 Hz_40ms_envelope_correlation": 0.9971217771893954, "1000–3000 Hz_40ms_modulation_mae_db": 1.1492162890446134, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6348002761219151, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16443737867165753, "3000–6000 Hz_10ms_envelope_correlation": 0.9830615795212555, "3000–6000 Hz_10ms_modulation_mae_db": 0.4177007277787192, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0593064823333007, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09457633051039992, "3000–6000 Hz_40ms_envelope_correlation": 0.9942130190155368, "3000–6000 Hz_40ms_modulation_mae_db": 0.4328837447709882, "mr_stft_db": 5.4480649856725805, "band_mae_db": 0.13265231011226017, "rms_gain_error_db": -0.09422242261787117, "output_peak": 0.4101325656498884}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.691190504325745, "spectral_convergence": 0.18142008112656477, "spectral_convergence_status": "ok", "amplitude_floor": 1.0507333742302167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.251896297594756, "spectral_convergence": 0.17039978896384028, "spectral_convergence_status": "ok", "amplitude_floor": 5.595595188050952e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.371426927233499, "spectral_convergence": 0.17599374706714097, "spectral_convergence_status": "ok", "amplitude_floor": 5.269142735102521e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007416076491485813, "output_power": 0.0007507505750106892, "delta_db": 0.05321472909332551, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001352027771357195, "output_power": 0.0013268393150447548, "delta_db": -0.08167280888830677, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000371703582440024, "output_power": 0.0003562186952100506, "delta_db": -0.18480038260788575, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018903159721465413, "output_power": 0.0001834931836466775, "delta_db": -0.1291446806539098, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5427403014296353e-05, "output_power": 1.4517337389188529e-05, "delta_db": -0.2640585472834373, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1014091243658427e-05, "output_power": 2.0826192415246012e-05, "delta_db": -0.03900737078895348, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12531641988596978, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.062041893259638346, "pitch": null, "output_rms_dbfs": -25.763208182833228, "output_peak": 0.44072800437636034, "reference_rms_dbfs": -25.701166289573592, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.438171243051333}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018114550052554645, "log_rms_mae_db": 1.2296511918820605, "rms_envelope_nrmse": 0.05668806589474913, "envelope_correlation": 0.9967772010385464, "rms_gain_error_db": -0.06193200663298108, "reference_mean_power": 0.002692702060931978, "output_mean_power": 0.0026545756238234826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0110233554726448e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033858489572855915, "output_power": 0.00033132410094009957, "reference_active": true, "delta_db": -0.09414552160472933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.201391758025471e-05, "output_power": 3.963713795569215e-05, "reference_active": true, "delta_db": -0.25290890915838254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.1566777501091025e-06, "output_power": 7.336090605143454e-06, "reference_active": true, "delta_db": 0.10753224191231445}], "active_band_count": 3, "active_band_mae_db": 0.15152889089180877}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015936923488846764, "log_rms_mae_db": 0.7009635176788671, "rms_envelope_nrmse": 0.0333384470684582, "envelope_correlation": 0.9988628306283159, "rms_gain_error_db": -0.05949893470105636, "reference_mean_power": 0.002702794865208283, "output_mean_power": 0.002666018702385786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.583684100944911e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031173336044584235, "output_power": 0.0003052723614741314, "reference_active": true, "delta_db": -0.09095794963954977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.830633609453067e-05, "output_power": 1.7776791519486825e-05, "reference_active": true, "delta_db": -0.12748052332687942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.642076702713698e-07, "output_power": 5.31291041315011e-07, "reference_active": true, "delta_db": 0.5862018066848156}], "active_band_count": 3, "active_band_mae_db": 0.26821342655041497}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001259260583850441, "log_rms_mae_db": 1.594946310073317, "rms_envelope_nrmse": 0.09279927293303566, "envelope_correlation": 0.9937109109628364, "rms_gain_error_db": -0.12939180520123064, "reference_mean_power": 0.0007105481356967998, "output_mean_power": 0.0006896906268126178}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8068878391559104e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016083630318364556, "output_power": 0.0001530100108795214, "reference_active": true, "delta_db": -0.21664236273271048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8008354908350678e-05, "output_power": 1.655707740652142e-05, "reference_active": true, "delta_db": -0.36490361971354723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.325041103743567e-06, "output_power": 2.3094095952492526e-06, "reference_active": true, "delta_db": -0.02929669152873139}], "active_band_count": 3, "active_band_mae_db": 0.20361422465832968}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010786919997015104, "log_rms_mae_db": 1.0635393964575262, "rms_envelope_nrmse": 0.05710758639752852, "envelope_correlation": 0.9976500789158048, "rms_gain_error_db": -0.1314041645781281, "reference_mean_power": 0.0007140165693307253, "output_mean_power": 0.0006927361850549032}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6527252040517484e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000151264236566862, "output_power": 0.00014476521797660442, "reference_active": true, "delta_db": -0.19072031224195554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.688501064647705e-06, "output_power": 7.343001559306245e-06, "reference_active": true, "delta_db": -0.1996805840706634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2694181944274492e-07, "output_power": 2.22080660030735e-07, "reference_active": true, "delta_db": -0.09403792862003058}], "active_band_count": 3, "active_band_mae_db": 0.16147960831088318}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.418520257385562e-05, "log_rms_mae_db": 1.510876395789798, "rms_envelope_nrmse": 0.1101236536285143, "envelope_correlation": 0.9921101410251871, "rms_gain_error_db": -0.1848904205989595, "reference_mean_power": 0.000153818152796868, "output_mean_power": 0.0001474071500033323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.418336470516266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4748514150479256e-05, "output_power": 2.278601717217218e-05, "reference_active": true, "delta_db": -0.3588070957118351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.277056498060538e-06, "output_power": 2.673272820255666e-06, "reference_active": true, "delta_db": -0.8844064605282427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.913659986768058e-07, "output_power": 5.330559789874393e-07, "reference_active": true, "delta_db": -2.232832448724274}], "active_band_count": 3, "active_band_mae_db": 1.1586820016547839}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.421085502718712e-05, "log_rms_mae_db": 0.8247507818249019, "rms_envelope_nrmse": 0.05641419330518687, "envelope_correlation": 0.9980836864739997, "rms_gain_error_db": -0.1849387043828858, "reference_mean_power": 0.00015458728634615244, "output_mean_power": 0.00014814257973376206}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1974511373265086e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3402826997507283e-05, "output_power": 2.1315221796327288e-05, "reference_active": true, "delta_db": -0.4057846609093362}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3826120793884328e-06, "output_power": 1.1429455006204446e-06, "reference_active": true, "delta_db": -0.8267482453122099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6109211314667126e-08, "output_power": 2.8145933726944088e-08, "reference_active": true, "delta_db": -2.1437203306335717}], "active_band_count": 3, "active_band_mae_db": 1.1254177456183727}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.007885661618194e-05, "log_rms_mae_db": 1.6764067179477053, "rms_envelope_nrmse": 0.1685517876752373, "envelope_correlation": 0.9820232581993492, "rms_gain_error_db": -0.13514900337775757, "reference_mean_power": 2.0928545188569684e-05, "output_mean_power": 2.028729478731054e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4882901242606597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1348414996445462e-05, "output_power": 1.1459635095269762e-05, "reference_active": true, "delta_db": 0.04235579794356527}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8308296968161944e-06, "output_power": 2.2838036080124984e-06, "reference_active": true, "delta_db": -0.9325498819311968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.670421987321005e-07, "output_power": 3.1575678408818767e-07, "reference_active": true, "delta_db": -3.2480061910564033}], "active_band_count": 3, "active_band_mae_db": 1.4076372903103884}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3258654185194873e-05, "log_rms_mae_db": 1.0585580216646564, "rms_envelope_nrmse": 0.08890648051941825, "envelope_correlation": 0.9949551833996921, "rms_gain_error_db": -0.13517858014185832, "reference_mean_power": 2.103354073620274e-05, "output_mean_power": 2.038893441721873e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.390796824652454e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1287290435083056e-05, "output_power": 1.1197998724231908e-05, "reference_active": true, "delta_db": -0.03449286408929346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.514784832162418e-06, "output_power": 1.1567741104747706e-06, "reference_active": true, "delta_db": -1.1710238759166787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.433812491322156e-08, "output_power": 1.7815779973202645e-08, "reference_active": true, "delta_db": -3.95972482108482}], "active_band_count": 3, "active_band_mae_db": 1.721747187030264}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8963186108245259, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d4c9b409e80ca2cd5fc99701826f69e0b181443f246f141bcd945218b26b7a24", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9919940775400962, "standard_deviation": 0.0009751324639354329, "minimum": 0.9897275785662251, "maximum": 0.9929684957670659, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.542322887338758, "standard_deviation": 0.029866037876648606, "minimum": 1.5031002827387334, "maximum": 1.5849409902443397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0750054537081781, "standard_deviation": 0.4488416256882482, "minimum": 0.5138319849532599, "maximum": 1.9895210108170527, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11221117352113107, "standard_deviation": 0.0075229653134702016, "minimum": 0.10505170851090555, "maximum": 0.12957024775276585, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974498689057147, "standard_deviation": 0.0007663636708595021, "minimum": 0.9958478432097747, "maximum": 0.998383892735007, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8745676987153075, "standard_deviation": 0.027528460286610466, "minimum": 0.8247507818249019, "maximum": 0.9042469940885142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9987298600684105, "standard_deviation": 0.43525029413054855, "minimum": 0.4141272736571872, "maximum": 1.8618664255349444, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06644554615043437, "standard_deviation": 0.01011846686694778, "minimum": 0.05641419330518687, "maximum": 0.08806868266557825, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9837473977507736, "standard_deviation": 0.0013135666576736117, "minimum": 0.9820232581993492, "maximum": 0.9858254163897441, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6330432429754391, "standard_deviation": 0.02576244738138955, "minimum": 1.6007590388306594, "maximum": 1.6764067179477053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8594295492884141, "standard_deviation": 0.3584606490844746, "minimum": 0.4107466833530589, "maximum": 1.4076372903103884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16094693911035463, "standard_deviation": 0.006802728311219055, "minimum": 0.1496601994817567, "maximum": 0.1685517876752373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9948139594795713, "standard_deviation": 0.0006740860021778885, "minimum": 0.9938537535682858, "maximum": 0.9958018241098752, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0431886251455933, "standard_deviation": 0.020565988870230173, "minimum": 1.0005816613114786, "maximum": 1.0597014238664817, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0680952030853448, "standard_deviation": 0.43685295100888216, "minimum": 0.4328837447709882, "maximum": 1.721747187030264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09095037682012472, "standard_deviation": 0.0068736081767212884, "minimum": 0.08302228513045823, "maximum": 0.1023052382832312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938058575159174, "standard_deviation": 0.0009041617415547269, "minimum": 0.9923166966114672, "maximum": 0.995014602040381, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5884346620361043, "standard_deviation": 0.06421140535959342, "minimum": 1.4781276140858766, "maximum": 1.6698700627484588, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.37603291423485663, "standard_deviation": 0.18911210926615057, "minimum": 0.20361422465832968, "maximum": 0.7903843475761917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09170901813870702, "standard_deviation": 0.006323167322868641, "minimum": 0.08309755610580667, "maximum": 0.10184597780988865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977586627450404, "standard_deviation": 0.0003796951072748487, "minimum": 0.9972455810814418, "maximum": 0.9981924431901, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0181904338583143, "standard_deviation": 0.07155145045438766, "minimum": 0.8920879472122513, "maximum": 1.1325522192093993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4309738961572479, "standard_deviation": 0.2248260984610965, "minimum": 0.16147960831088318, "maximum": 0.8371951440423889, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.055264927856938484, "standard_deviation": 0.00444970078788339, "minimum": 0.049228565035057424, "maximum": 0.06041785223665319, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14485358361488482, "standard_deviation": 0.02080178918668003, "minimum": 0.12181852076254156, "maximum": 0.1752831789383268, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.451277201041302, "standard_deviation": 0.007234516540128804, "minimum": 5.438171243051333, "maximum": 5.460151588804291, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4348508533508643, "standard_deviation": 0.012822343727398738, "minimum": 0.4101325656498884, "maximum": 0.45095012150643193, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.062049866221170535, "standard_deviation": 0.020741655404860934, "minimum": -0.09422242261787117, "maximum": -0.02254991702561354, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996450563974661, "standard_deviation": 0.0004647103280274922, "minimum": 0.9957467803076822, "maximum": 0.9970130500024371, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.264092026193291, "standard_deviation": 0.024209556690402978, "minimum": 1.2296511918820605, "maximum": 1.311284470131398, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.15490103613953504, "standard_deviation": 0.05522575962361926, "minimum": 0.07328220694565593, "maximum": 0.24598475357022717, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0594243300624855, "standard_deviation": 0.003839805081371605, "minimum": 0.054956211060167445, "maximum": 0.06511356893629126, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989416625503498, "standard_deviation": 0.00024378587530443196, "minimum": 0.9985766703324158, "maximum": 0.9993380844622084, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6958071209293367, "standard_deviation": 0.044655548026091424, "minimum": 0.6501058275646727, "maximum": 0.772860785881102, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24429637081445812, "standard_deviation": 0.06830840659623491, "minimum": 0.12545348950136706, "maximum": 0.3379725525623976, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03217812343670751, "standard_deviation": 0.003667949446564212, "minimum": 0.0260893375126243, "maximum": 0.03739053280122483, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0037", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.662917375564575, "predicted_mos": 2.662917375564575, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "82aa1ae5807e7e8d78311c68eeb9695946f2a9cfeb16da782eb7dc89a6697bd5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.121854899916798, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1585283468437115, "全帯域_10ms_rms_envelope_nrmse": 0.06052215802766955, "全帯域_10ms_envelope_correlation": 0.9963659679682538, "全帯域_10ms_modulation_mae_db": 0.382106310461183, "全帯域_40ms_log_rms_mae_db": 0.6159330746276324, "全帯域_40ms_rms_envelope_nrmse": 0.034896036224309014, "全帯域_40ms_envelope_correlation": 0.998800937512759, "全帯域_40ms_modulation_mae_db": 0.4744556029095399, "300–1000 Hz_10ms_log_rms_mae_db": 2.024779867689258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08481418588685419, "300–1000 Hz_10ms_envelope_correlation": 0.9947125852934917, "300–1000 Hz_10ms_modulation_mae_db": 0.5518135564846851, "300–1000 Hz_40ms_log_rms_mae_db": 1.4181384160729074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04899237744071564, "300–1000 Hz_40ms_envelope_correlation": 0.9982364264230653, "300–1000 Hz_40ms_modulation_mae_db": 0.38427756560808796, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0858876047662984, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12806780842270707, "1000–3000 Hz_10ms_envelope_correlation": 0.9898642009364856, "1000–3000 Hz_10ms_modulation_mae_db": 0.7000965192487184, "1000–3000 Hz_40ms_log_rms_mae_db": 1.417052396074846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08142135099418167, "1000–3000 Hz_40ms_envelope_correlation": 0.9964651624007925, "1000–3000 Hz_40ms_modulation_mae_db": 0.8961884457708865, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5215139732264866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20315506578996434, "3000–6000 Hz_10ms_envelope_correlation": 0.9739257261611389, "3000–6000 Hz_10ms_modulation_mae_db": 1.333112256226736, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7689241274446852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11600936584822048, "3000–6000 Hz_40ms_envelope_correlation": 0.9913788853988093, "3000–6000 Hz_40ms_modulation_mae_db": 1.4838477697820398, "mr_stft_db": 5.9287050813901985, "band_mae_db": 0.19956522244493, "rms_gain_error_db": -0.08610772785346306, "output_peak": 0.42155381073129095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1924062666487083, "全帯域_10ms_rms_envelope_nrmse": 0.054807338519977086, "全帯域_10ms_envelope_correlation": 0.9970460800523089, "全帯域_10ms_modulation_mae_db": 0.3432447543773498, "全帯域_40ms_log_rms_mae_db": 0.6234230653773779, "全帯域_40ms_rms_envelope_nrmse": 0.028744455807267207, "全帯域_40ms_envelope_correlation": 0.9992283892729584, "全帯域_40ms_modulation_mae_db": 0.3855579183584252, "300–1000 Hz_10ms_log_rms_mae_db": 1.8665874063902563, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08828263989092355, "300–1000 Hz_10ms_envelope_correlation": 0.9944445625437719, "300–1000 Hz_10ms_modulation_mae_db": 0.32030328063793717, "300–1000 Hz_40ms_log_rms_mae_db": 1.3265174649675184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05502287283923932, "300–1000 Hz_40ms_envelope_correlation": 0.9979677673603486, "300–1000 Hz_40ms_modulation_mae_db": 0.5661210025006174, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0959654881376797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12239807513729799, "1000–3000 Hz_10ms_envelope_correlation": 0.9905025026786393, "1000–3000 Hz_10ms_modulation_mae_db": 1.1501650488509183, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3844326464518284, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07258332360287165, "1000–3000 Hz_40ms_envelope_correlation": 0.9970075814888213, "1000–3000 Hz_40ms_modulation_mae_db": 1.2055043943953596, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5019610635536327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1921302319883408, "3000–6000 Hz_10ms_envelope_correlation": 0.9767948188750059, "3000–6000 Hz_10ms_modulation_mae_db": 0.9122686079894845, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7662077377365646, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11237854516347101, "3000–6000 Hz_40ms_envelope_correlation": 0.9918907133603009, "3000–6000 Hz_40ms_modulation_mae_db": 1.1716575996343355, "mr_stft_db": 5.911018144901923, "band_mae_db": 0.1276106826201158, "rms_gain_error_db": -0.09340979953347806, "output_peak": 0.42643579946498483}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2456755437379785, "全帯域_10ms_rms_envelope_nrmse": 0.056729824193488425, "全帯域_10ms_envelope_correlation": 0.9967599498500663, "全帯域_10ms_modulation_mae_db": 0.46482634974914266, "全帯域_40ms_log_rms_mae_db": 0.6502009241240472, "全帯域_40ms_rms_envelope_nrmse": 0.02792977855778913, "全帯域_40ms_envelope_correlation": 0.9992018169482387, "全帯域_40ms_modulation_mae_db": 0.3794454902377278, "300–1000 Hz_10ms_log_rms_mae_db": 1.9793400985147964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09561906891923906, "300–1000 Hz_10ms_envelope_correlation": 0.9933199599850993, "300–1000 Hz_10ms_modulation_mae_db": 0.7982785979180212, "300–1000 Hz_40ms_log_rms_mae_db": 1.3615658485777444, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06016502566827587, "300–1000 Hz_40ms_envelope_correlation": 0.9973683645437432, "300–1000 Hz_40ms_modulation_mae_db": 1.0622191111496875, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0787800661349674, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12283888856611339, "1000–3000 Hz_10ms_envelope_correlation": 0.990167521434886, "1000–3000 Hz_10ms_modulation_mae_db": 0.4953723507635712, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2909710113159198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06515604017998848, "1000–3000 Hz_40ms_envelope_correlation": 0.9974124225452037, "1000–3000 Hz_40ms_modulation_mae_db": 0.48952055454212856, "3000–6000 Hz_10ms_log_rms_mae_db": 2.463689648658262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19071308021404978, "3000–6000 Hz_10ms_envelope_correlation": 0.9771231760604355, "3000–6000 Hz_10ms_modulation_mae_db": 2.0873891821327724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7385096428833806, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11067654347143704, "3000–6000 Hz_40ms_envelope_correlation": 0.9924020505070168, "3000–6000 Hz_40ms_modulation_mae_db": 2.1886534980780454, "mr_stft_db": 5.898448219071619, "band_mae_db": 0.12123407274151422, "rms_gain_error_db": -0.05026392454937409, "output_peak": 0.4272166902336266}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.209547383274089, "全帯域_10ms_rms_envelope_nrmse": 0.05919859416470148, "全帯域_10ms_envelope_correlation": 0.9964758820023607, "全帯域_10ms_modulation_mae_db": 0.32025498145428194, "全帯域_40ms_log_rms_mae_db": 0.6019011465064598, "全帯域_40ms_rms_envelope_nrmse": 0.027810075697185306, "全帯域_40ms_envelope_correlation": 0.9992280046998151, "全帯域_40ms_modulation_mae_db": 0.6268339787279089, "300–1000 Hz_10ms_log_rms_mae_db": 2.0183406788385465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09589188493579771, "300–1000 Hz_10ms_envelope_correlation": 0.9932319851597343, "300–1000 Hz_10ms_modulation_mae_db": 0.6386676255451182, "300–1000 Hz_40ms_log_rms_mae_db": 1.4160256454209943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05400582613090938, "300–1000 Hz_40ms_envelope_correlation": 0.997851920947303, "300–1000 Hz_40ms_modulation_mae_db": 0.610731166418362, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1500277956060194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13986855622305142, "1000–3000 Hz_10ms_envelope_correlation": 0.9889030390228143, "1000–3000 Hz_10ms_modulation_mae_db": 0.8373296921172907, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3789087500074975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09200055575711782, "1000–3000 Hz_40ms_envelope_correlation": 0.9965936295158434, "1000–3000 Hz_40ms_modulation_mae_db": 1.0212012773279537, "3000–6000 Hz_10ms_log_rms_mae_db": 2.51371165667609, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19497915644149855, "3000–6000 Hz_10ms_envelope_correlation": 0.9760472145131502, "3000–6000 Hz_10ms_modulation_mae_db": 0.9142371471981617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7536702553726293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10699587222061537, "3000–6000 Hz_40ms_envelope_correlation": 0.9930096027506699, "3000–6000 Hz_40ms_modulation_mae_db": 1.1256107997011027, "mr_stft_db": 5.91847672863522, "band_mae_db": 0.18520764844896764, "rms_gain_error_db": -0.05820108762595885, "output_peak": 0.40525324275134256}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1570026414976597, "全帯域_10ms_rms_envelope_nrmse": 0.054013452445058965, "全帯域_10ms_envelope_correlation": 0.9970775815766365, "全帯域_10ms_modulation_mae_db": 0.3293927983210833, "全帯域_40ms_log_rms_mae_db": 0.6219767771801936, "全帯域_40ms_rms_envelope_nrmse": 0.024859433370547822, "全帯域_40ms_envelope_correlation": 0.9993854882467753, "全帯域_40ms_modulation_mae_db": 0.27610264496352993, "300–1000 Hz_10ms_log_rms_mae_db": 2.1939844479979955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0913619899511704, "300–1000 Hz_10ms_envelope_correlation": 0.9939136726939758, "300–1000 Hz_10ms_modulation_mae_db": 1.0357512330734793, "300–1000 Hz_40ms_log_rms_mae_db": 1.5780896543720062, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05658431113380073, "300–1000 Hz_40ms_envelope_correlation": 0.9976603597616351, "300–1000 Hz_40ms_modulation_mae_db": 1.2608997686819836, "1000–3000 Hz_10ms_log_rms_mae_db": 2.089761020942217, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11454063583303767, "1000–3000 Hz_10ms_envelope_correlation": 0.9914075384480622, "1000–3000 Hz_10ms_modulation_mae_db": 1.101812622811806, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3688035861216394, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06573701115417079, "1000–3000 Hz_40ms_envelope_correlation": 0.9972369622276762, "1000–3000 Hz_40ms_modulation_mae_db": 1.0624007709821714, "3000–6000 Hz_10ms_log_rms_mae_db": 2.478905364908809, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1960334920897132, "3000–6000 Hz_10ms_envelope_correlation": 0.975752706760672, "3000–6000 Hz_10ms_modulation_mae_db": 2.095977324803261, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7736830879745549, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11348846056788385, "3000–6000 Hz_40ms_envelope_correlation": 0.9918033834007555, "3000–6000 Hz_40ms_modulation_mae_db": 2.2473806209460214, "mr_stft_db": 5.910234729219301, "band_mae_db": 0.1736955774666162, "rms_gain_error_db": -0.05301067630762741, "output_peak": 0.4410484542282304}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1590387565913032, "全帯域_10ms_rms_envelope_nrmse": 0.056355395730870846, "全帯域_10ms_envelope_correlation": 0.9968275866808981, "全帯域_10ms_modulation_mae_db": 0.24779660763780878, "全帯域_40ms_log_rms_mae_db": 0.6244436401999996, "全帯域_40ms_rms_envelope_nrmse": 0.03197221674775392, "全帯域_40ms_envelope_correlation": 0.9989791171733985, "全帯域_40ms_modulation_mae_db": 0.1279248293302363, "300–1000 Hz_10ms_log_rms_mae_db": 2.0421240559604708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10387838576516677, "300–1000 Hz_10ms_envelope_correlation": 0.9920654650383954, "300–1000 Hz_10ms_modulation_mae_db": 0.45595579187995594, "300–1000 Hz_40ms_log_rms_mae_db": 1.4355746768151507, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06470360348760029, "300–1000 Hz_40ms_envelope_correlation": 0.9969214530563328, "300–1000 Hz_40ms_modulation_mae_db": 0.5590423447665819, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0227773799038222, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12366777924508006, "1000–3000 Hz_10ms_envelope_correlation": 0.9902368833626191, "1000–3000 Hz_10ms_modulation_mae_db": 0.7499224845478998, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3262603668133544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07315182085351304, "1000–3000 Hz_40ms_envelope_correlation": 0.9969053282034758, "1000–3000 Hz_40ms_modulation_mae_db": 0.7498462736431608, "3000–6000 Hz_10ms_log_rms_mae_db": 2.465779426459975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1879392137982891, "3000–6000 Hz_10ms_envelope_correlation": 0.9777073320837694, "3000–6000 Hz_10ms_modulation_mae_db": 0.9873264230852609, "3000–6000 Hz_40ms_log_rms_mae_db": 1.753507333248817, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11017090181316821, "3000–6000 Hz_40ms_envelope_correlation": 0.9922872881222682, "3000–6000 Hz_40ms_modulation_mae_db": 1.2195975597288014, "mr_stft_db": 5.916824680636462, "band_mae_db": 0.16137842373425615, "rms_gain_error_db": -0.06992201482487728, "output_peak": 0.43468886671975954}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2196966147084318, "全帯域_10ms_rms_envelope_nrmse": 0.05598204865133241, "全帯域_10ms_envelope_correlation": 0.9968511342113887, "全帯域_10ms_modulation_mae_db": 0.2546944070555133, "全帯域_40ms_log_rms_mae_db": 0.6606179368487918, "全帯域_40ms_rms_envelope_nrmse": 0.029352630003411862, "全帯域_40ms_envelope_correlation": 0.9991165394643599, "全帯域_40ms_modulation_mae_db": 0.29082478496784264, "300–1000 Hz_10ms_log_rms_mae_db": 2.030053397426354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08546505139695548, "300–1000 Hz_10ms_envelope_correlation": 0.9945928055799853, "300–1000 Hz_10ms_modulation_mae_db": 0.40797522149936044, "300–1000 Hz_40ms_log_rms_mae_db": 1.4071301968646723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0522882227445789, "300–1000 Hz_40ms_envelope_correlation": 0.9979246873718662, "300–1000 Hz_40ms_modulation_mae_db": 0.48768988844588274, "1000–3000 Hz_10ms_log_rms_mae_db": 2.029325376119832, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12106342150999885, "1000–3000 Hz_10ms_envelope_correlation": 0.9905874344512472, "1000–3000 Hz_10ms_modulation_mae_db": 1.1510239060792131, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3827438368966432, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06871703343499297, "1000–3000 Hz_40ms_envelope_correlation": 0.9972311552515948, "1000–3000 Hz_40ms_modulation_mae_db": 1.1480518849858605, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4911330668601104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1777268537377745, "3000–6000 Hz_10ms_envelope_correlation": 0.980230651901758, "3000–6000 Hz_10ms_modulation_mae_db": 1.5248189921858806, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7874994934765498, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10160695366512078, "3000–6000 Hz_40ms_envelope_correlation": 0.9936561001654635, "3000–6000 Hz_40ms_modulation_mae_db": 1.9830642591455365, "mr_stft_db": 5.915068741207481, "band_mae_db": 0.1475364841625684, "rms_gain_error_db": -0.04721279032611865, "output_peak": 0.41571852756400957}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2104798502329122, "全帯域_10ms_rms_envelope_nrmse": 0.05753547297485237, "全帯域_10ms_envelope_correlation": 0.9966761625695819, "全帯域_10ms_modulation_mae_db": 0.299065300187432, "全帯域_40ms_log_rms_mae_db": 0.6506285670784204, "全帯域_40ms_rms_envelope_nrmse": 0.03066748658841914, "全帯域_40ms_envelope_correlation": 0.9990451241702126, "全帯域_40ms_modulation_mae_db": 0.2158357894255881, "300–1000 Hz_10ms_log_rms_mae_db": 2.021264855012453, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08590210648676266, "300–1000 Hz_10ms_envelope_correlation": 0.9945739052900848, "300–1000 Hz_10ms_modulation_mae_db": 0.40243872959284666, "300–1000 Hz_40ms_log_rms_mae_db": 1.420560781738401, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050827429462720196, "300–1000 Hz_40ms_envelope_correlation": 0.9980693890117509, "300–1000 Hz_40ms_modulation_mae_db": 0.630980408551716, "1000–3000 Hz_10ms_log_rms_mae_db": 2.070310469340611, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12167851489958424, "1000–3000 Hz_10ms_envelope_correlation": 0.990444672591956, "1000–3000 Hz_10ms_modulation_mae_db": 0.600992260496484, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3702286188290744, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06624974720137826, "1000–3000 Hz_40ms_envelope_correlation": 0.9974238137335332, "1000–3000 Hz_40ms_modulation_mae_db": 0.5285376912584235, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5030191115046896, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1951740479473103, "3000–6000 Hz_10ms_envelope_correlation": 0.9759892980656913, "3000–6000 Hz_10ms_modulation_mae_db": 0.9495362112065182, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8087088788646146, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.121070046459443, "3000–6000 Hz_40ms_envelope_correlation": 0.9908273097800764, "3000–6000 Hz_40ms_modulation_mae_db": 1.423023275945641, "mr_stft_db": 5.918514985499004, "band_mae_db": 0.17444408278833234, "rms_gain_error_db": -0.06007493917984348, "output_peak": 0.4307164414594583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.233720051206364, "spectral_convergence": 0.18948773853121514, "spectral_convergence_status": "ok", "amplitude_floor": 1.0507333742302167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.813208354378459, "spectral_convergence": 0.18045546880495217, "spectral_convergence_status": "ok", "amplitude_floor": 5.595595188050952e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7391868385857725, "spectral_convergence": 0.18756545770163174, "spectral_convergence_status": "ok", "amplitude_floor": 5.269142735102521e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007416076491485813, "output_power": 0.0007567320714680786, "delta_db": 0.08767939830932994, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001352027771357195, "output_power": 0.0013062755303189983, "delta_db": -0.14950820787218938, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000371703582440024, "output_power": 0.0003668978036260358, "delta_db": -0.05651634869900748, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018903159721465413, "output_power": 0.00017254340911912862, "delta_db": -0.39636029310291765, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5427403014296353e-05, "output_power": 1.6066298115511394e-05, "delta_db": 0.17622996315701173, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1014091243658427e-05, "output_power": 1.9471566907740936e-05, "delta_db": -0.33109712352912374, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19956522244493, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08610772785346306, "pitch": null, "output_rms_dbfs": -25.787274017427055, "output_peak": 0.42155381073129095, "reference_rms_dbfs": -25.701166289573592, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9287050813901985}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018114550052554645, "log_rms_mae_db": 1.1585283468437115, "rms_envelope_nrmse": 0.06052215802766955, "envelope_correlation": 0.9963659679682538, "rms_gain_error_db": -0.086262499125299, "reference_mean_power": 0.002692702060931978, "output_mean_power": 0.0026397454673260643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0110233554726448e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033858489572855915, "output_power": 0.0003272523088960106, "reference_active": true, "delta_db": -0.147848609741881}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.201391758025471e-05, "output_power": 3.651718828572493e-05, "reference_active": true, "delta_db": -0.6089584836422823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.1566777501091025e-06, "output_power": 6.542749282601167e-06, "reference_active": true, "delta_db": -0.3895118379993855}], "active_band_count": 3, "active_band_mae_db": 0.382106310461183}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015936923488846764, "log_rms_mae_db": 0.6159330746276324, "rms_envelope_nrmse": 0.034896036224309014, "envelope_correlation": 0.998800937512759, "rms_gain_error_db": -0.08460996704091997, "reference_mean_power": 0.002702794865208283, "output_mean_power": 0.0026506481844823936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.583684100944911e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031173336044584235, "output_power": 0.00030075929447824935, "reference_active": true, "delta_db": -0.15564223820262157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.830633609453067e-05, "output_power": 1.622924279445774e-05, "reference_active": true, "delta_db": -0.5230317403225493}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.642076702713698e-07, "output_power": 3.9105966973905867e-07, "reference_active": true, "delta_db": -0.7446928302034488}], "active_band_count": 3, "active_band_mae_db": 0.4744556029095399}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001259260583850441, "log_rms_mae_db": 2.024779867689258, "rms_envelope_nrmse": 0.08481418588685419, "envelope_correlation": 0.9947125852934917, "rms_gain_error_db": -0.0936806976287867, "reference_mean_power": 0.0007105481356967998, "output_mean_power": 0.0006953851860014258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8068878391559104e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016083630318364556, "output_power": 0.0001496932531193371, "reference_active": true, "delta_db": -0.31181855768381805}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8008354908350678e-05, "output_power": 1.4797696705624762e-05, "reference_active": true, "delta_db": -0.8527991939084423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.325041103743567e-06, "output_power": 2.6032323158743594e-06, "reference_active": true, "delta_db": 0.4908229178617951}], "active_band_count": 3, "active_band_mae_db": 0.5518135564846851}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010786919997015104, "log_rms_mae_db": 1.4181384160729074, "rms_envelope_nrmse": 0.04899237744071564, "envelope_correlation": 0.9982364264230653, "rms_gain_error_db": -0.09398431295149157, "reference_mean_power": 0.0007140165693307253, "output_mean_power": 0.0006987307539990485}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6527252040517484e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000151264236566862, "output_power": 0.00014050497533886694, "reference_active": true, "delta_db": -0.3204455677074244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.688501064647705e-06, "output_power": 6.391632592455899e-06, "reference_active": true, "delta_db": -0.8022987618321686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2694181944274492e-07, "output_power": 2.2537497712317337e-07, "reference_active": true, "delta_db": -0.030088367284671003}], "active_band_count": 3, "active_band_mae_db": 0.38427756560808796}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.418520257385562e-05, "log_rms_mae_db": 2.0858876047662984, "rms_envelope_nrmse": 0.12806780842270707, "envelope_correlation": 0.9898642009364856, "rms_gain_error_db": -0.35349155018453843, "reference_mean_power": 0.000153818152796868, "output_mean_power": 0.00014179419017619328}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.418336470516266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4748514150479256e-05, "output_power": 2.1993323536004856e-05, "reference_active": true, "delta_db": -0.5125826688598227}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.277056498060538e-06, "output_power": 2.6702932747451205e-06, "reference_active": true, "delta_db": -0.889249669295238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.913659986768058e-07, "output_power": 7.589450659205047e-07, "reference_active": true, "delta_db": -0.6984572195910943}], "active_band_count": 3, "active_band_mae_db": 0.7000965192487184}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.421085502718712e-05, "log_rms_mae_db": 1.417052396074846, "rms_envelope_nrmse": 0.08142135099418167, "envelope_correlation": 0.9964651624007925, "rms_gain_error_db": -0.35351402267933146, "reference_mean_power": 0.00015458728634615244, "output_mean_power": 0.00014250246318934555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1974511373265086e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3402826997507283e-05, "output_power": 2.0606896493893944e-05, "reference_active": true, "delta_db": -0.5525573247525725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3826120793884328e-06, "output_power": 1.1785298614755065e-06, "reference_active": true, "delta_db": -0.6935975572764208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6109211314667126e-08, "output_power": 3.307856403676752e-08, "reference_active": true, "delta_db": -1.4424104552836665}], "active_band_count": 3, "active_band_mae_db": 0.8961884457708865}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.007885661618194e-05, "log_rms_mae_db": 2.5215139732264866, "rms_envelope_nrmse": 0.20315506578996434, "envelope_correlation": 0.9739257261611389, "rms_gain_error_db": -0.10298366761837131, "reference_mean_power": 2.0928545188569684e-05, "output_mean_power": 2.043810723099792e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4882901242606597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1348414996445462e-05, "output_power": 1.040166041769762e-05, "reference_active": true, "delta_db": -0.3783253769814553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8308296968161944e-06, "output_power": 2.1232762205614064e-06, "reference_active": true, "delta_db": -1.2490724698479851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.670421987321005e-07, "output_power": 3.86330874489073e-07, "reference_active": true, "delta_db": -2.3719389218507674}], "active_band_count": 3, "active_band_mae_db": 1.333112256226736}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3258654185194873e-05, "log_rms_mae_db": 1.7689241274446852, "rms_envelope_nrmse": 0.11600936584822048, "envelope_correlation": 0.9913788853988093, "rms_gain_error_db": -0.10301742226512364, "reference_mean_power": 2.103354073620274e-05, "output_mean_power": 2.0540482673542238e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.390796824652454e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1287290435083056e-05, "output_power": 1.0203461952475152e-05, "reference_active": true, "delta_db": -0.43842150703675886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.514784832162418e-06, "output_power": 1.092270618832372e-06, "reference_active": true, "delta_db": -1.4202069614296047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.433812491322156e-08, "output_power": 2.4405394854674242e-08, "reference_active": true, "delta_db": -2.592914840879756}], "active_band_count": 3, "active_band_mae_db": 1.4838477697820398}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8775001704422238, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1581c8c95ed404603aefe0f40ad0685e81c433e71db88f75ce109c486c309970", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9902642241158387, "standard_deviation": 0.0007095182707792978, "minimum": 0.9889030390228143, "maximum": 0.9914075384480622, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0778544001189307, "standard_deviation": 0.03998146462860093, "minimum": 2.0227773799038222, "maximum": 2.1500277956060194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8483393606144877, "standard_deviation": 0.25760288016433813, "minimum": 0.4953723507635712, "maximum": 1.1510239060792131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12426545997960883, "standard_deviation": 0.0073222894581451425, "minimum": 0.11454063583303767, "maximum": 0.13986855622305142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970345069208677, "standard_deviation": 0.0003605505550951958, "minimum": 0.9964651624007925, "maximum": 0.9974238137335332, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3649251515638503, "standard_deviation": 0.03894719503709201, "minimum": 1.2909710113159198, "maximum": 1.417052396074846, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.887656411613243, "standard_deviation": 0.2735339651750927, "minimum": 0.48952055454212856, "maximum": 1.2055043943953596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07312711039727683, "standard_deviation": 0.009336534322512134, "minimum": 0.06515604017998848, "maximum": 0.09200055575711782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9766963655527027, "standard_deviation": 0.0018202578171959667, "minimum": 0.9739257261611389, "maximum": 0.980230651901758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.492464163981007, "standard_deviation": 0.021456288396236707, "minimum": 2.463689648658262, "maximum": 2.5215139732264866, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3505832681035095, "standard_deviation": 0.5077218775095397, "minimum": 0.9122686079894845, "maximum": 2.095977324803261, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19223139275086756, "standard_deviation": 0.007378656241918996, "minimum": 0.1777268537377745, "maximum": 0.20315506578996434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921569166856701, "standard_deviation": 0.0008960826758807943, "minimum": 0.9908273097800764, "maximum": 0.9936561001654635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7688388196252247, "standard_deviation": 0.021886845218533558, "minimum": 1.7385096428833806, "maximum": 1.8087088788646146, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6053544228701904, "standard_deviation": 0.46447572908659385, "minimum": 1.1256107997011027, "maximum": 2.2473806209460214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11154958615116997, "standard_deviation": 0.005821517129383558, "minimum": 0.10160695366512078, "maximum": 0.121070046459443, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938568676980672, "standard_deviation": 0.0009277071339592795, "minimum": 0.9920654650383954, "maximum": 0.9947125852934917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0220593509787665, "standard_deviation": 0.08947709778197238, "minimum": 1.8665874063902563, "maximum": 2.1939844479979955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5763980045789254, "standard_deviation": 0.24007058023376432, "minimum": 0.32030328063793717, "maximum": 1.0357512330734793, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09140191415410873, "standard_deviation": 0.006678648592601708, "minimum": 0.08481418588685419, "maximum": 0.10387838576516677, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977500460595057, "standard_deviation": 0.00042546062453366356, "minimum": 0.9969214530563328, "maximum": 0.9982364264230653, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4204503356036744, "standard_deviation": 0.07326276453052397, "minimum": 1.3265174649675184, "maximum": 1.5780896543720062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6952451570153648, "standard_deviation": 0.3025869133641341, "minimum": 0.38427756560808796, "maximum": 1.2608997686819836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05532370861348004, "standard_deviation": 0.005131884489163901, "minimum": 0.04899237744071564, "maximum": 0.06470360348760029, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1613340243009126, "standard_deviation": 0.027495194504979444, "minimum": 0.12123407274151422, "maximum": 0.19956522244493, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.914661413820151, "standard_deviation": 0.008678491545020381, "minimum": 5.898448219071619, "maximum": 5.9287050813901985, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42532897914408785, "standard_deviation": 0.0112005772476183, "minimum": 0.40525324275134256, "maximum": 0.4410484542282304, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06477537002509261, "standard_deviation": 0.017005198824482815, "minimum": -0.09340979953347806, "maximum": -0.04721279032611865, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996760043113937, "standard_deviation": 0.0002503101616213818, "minimum": 0.9963659679682538, "maximum": 0.9970775815766365, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1940469254418493, "standard_deviation": 0.033157283282972703, "minimum": 1.1570026414976597, "maximum": 1.2456755437379785, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33017268865547433, "standard_deviation": 0.07025363935819559, "minimum": 0.24779660763780878, "maximum": 0.46482634974914266, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05689303558849389, "standard_deviation": 0.002160905198537387, "minimum": 0.054013452445058965, "maximum": 0.06052215802766955, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991231771860647, "standard_deviation": 0.00018038756789920166, "minimum": 0.998800937512759, "maximum": 0.9993854882467753, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6311406414928653, "standard_deviation": 0.02030689514385904, "minimum": 0.6019011465064598, "maximum": 0.6606179368487918, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.34712262986509984, "standard_deviation": 0.15605464209629918, "minimum": 0.1279248293302363, "maximum": 0.6268339787279089, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.029529014124585425, "standard_deviation": 0.0030245239548195076, "minimum": 0.024859433370547822, "maximum": 0.034896036224309014, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0037", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8547215461730957, "predicted_mos": 2.8547215461730957, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "333790c661a5044bbf70c717360f5770edd037a747534ba4be2ccab4aa04e121", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.857612400315702, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.22829544858235, "全帯域_10ms_rms_envelope_nrmse": 0.054570824070309866, "全帯域_10ms_envelope_correlation": 0.9970138055490344, "全帯域_10ms_modulation_mae_db": 0.44636726134098614, "全帯域_40ms_log_rms_mae_db": 0.6139651223032014, "全帯域_40ms_rms_envelope_nrmse": 0.027937453774529223, "全帯域_40ms_envelope_correlation": 0.9992162284792611, "全帯域_40ms_modulation_mae_db": 0.40469232098198055, "300–1000 Hz_10ms_log_rms_mae_db": 1.4695896688155512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09100517926501053, "300–1000 Hz_10ms_envelope_correlation": 0.9938721187922461, "300–1000 Hz_10ms_modulation_mae_db": 0.6031520480138287, "300–1000 Hz_40ms_log_rms_mae_db": 0.9674877589354913, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054728302004603706, "300–1000 Hz_40ms_envelope_correlation": 0.9977343453426093, "300–1000 Hz_40ms_modulation_mae_db": 0.824934190568776, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4574045091246217, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11502645923140015, "1000–3000 Hz_10ms_envelope_correlation": 0.9913925260119756, "1000–3000 Hz_10ms_modulation_mae_db": 0.667599980661119, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8392055644100872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06927804511881937, "1000–3000 Hz_40ms_envelope_correlation": 0.9969911785206965, "1000–3000 Hz_40ms_modulation_mae_db": 0.7593311734177045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5468489378505532, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18112240166604138, "3000–6000 Hz_10ms_envelope_correlation": 0.9796048756815129, "3000–6000 Hz_10ms_modulation_mae_db": 0.7818085225895258, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0147269415836837, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11104938248517722, "3000–6000 Hz_40ms_envelope_correlation": 0.9928036324473631, "3000–6000 Hz_40ms_modulation_mae_db": 0.939270807778311, "mr_stft_db": 5.408748156749098, "band_mae_db": 0.14115289912257845, "rms_gain_error_db": -0.055106247198640664, "output_peak": 0.43915065291765226}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2711534516376861, "全帯域_10ms_rms_envelope_nrmse": 0.0582857881318744, "全帯域_10ms_envelope_correlation": 0.9965732786945903, "全帯域_10ms_modulation_mae_db": 0.5026405432686577, "全帯域_40ms_log_rms_mae_db": 0.6440965394003426, "全帯域_40ms_rms_envelope_nrmse": 0.02708074298857173, "全帯域_40ms_envelope_correlation": 0.999252868857936, "全帯域_40ms_modulation_mae_db": 0.640922792927396, "300–1000 Hz_10ms_log_rms_mae_db": 1.434249174447375, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08072381924778299, "300–1000 Hz_10ms_envelope_correlation": 0.9952371310655421, "300–1000 Hz_10ms_modulation_mae_db": 0.36007094866990136, "300–1000 Hz_40ms_log_rms_mae_db": 0.9369618174734468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04692958845541336, "300–1000 Hz_40ms_envelope_correlation": 0.9984065495584449, "300–1000 Hz_40ms_modulation_mae_db": 0.35807601799408145, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4025442928922693, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10491024073357451, "1000–3000 Hz_10ms_envelope_correlation": 0.9929766718567502, "1000–3000 Hz_10ms_modulation_mae_db": 0.5560506299382326, "1000–3000 Hz_40ms_log_rms_mae_db": 0.821125405779692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059503149484851095, "1000–3000 Hz_40ms_envelope_correlation": 0.9979695315113, "1000–3000 Hz_40ms_modulation_mae_db": 0.7004135904576262, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5228536831621677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17546009936340146, "3000–6000 Hz_10ms_envelope_correlation": 0.9815636209538343, "3000–6000 Hz_10ms_modulation_mae_db": 1.5065400052819067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9900620632725623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11502345788968742, "3000–6000 Hz_40ms_envelope_correlation": 0.9930155709023049, "3000–6000 Hz_40ms_modulation_mae_db": 1.4652151525555908, "mr_stft_db": 5.404256609674545, "band_mae_db": 0.19467113174460396, "rms_gain_error_db": -0.04697023430502104, "output_peak": 0.4390861147938102}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2018204956562153, "全帯域_10ms_rms_envelope_nrmse": 0.05234596121749672, "全帯域_10ms_envelope_correlation": 0.9972396995597037, "全帯域_10ms_modulation_mae_db": 0.15576007202631273, "全帯域_40ms_log_rms_mae_db": 0.6139547299404348, "全帯域_40ms_rms_envelope_nrmse": 0.024439638930923477, "全帯域_40ms_envelope_correlation": 0.9993876595392134, "全帯域_40ms_modulation_mae_db": 0.18949619423552325, "300–1000 Hz_10ms_log_rms_mae_db": 1.4857912376721627, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08507240629519214, "300–1000 Hz_10ms_envelope_correlation": 0.9946647261091806, "300–1000 Hz_10ms_modulation_mae_db": 0.3063941884205905, "300–1000 Hz_40ms_log_rms_mae_db": 1.0475394103317364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05642444508831602, "300–1000 Hz_40ms_envelope_correlation": 0.9976031463132544, "300–1000 Hz_40ms_modulation_mae_db": 0.4129844256737743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.444479034082573, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11338323688985293, "1000–3000 Hz_10ms_envelope_correlation": 0.9918201503749198, "1000–3000 Hz_10ms_modulation_mae_db": 0.7537523203647584, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8327868261957577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06020024714880477, "1000–3000 Hz_40ms_envelope_correlation": 0.9980361225210596, "1000–3000 Hz_40ms_modulation_mae_db": 0.871031560791621, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4796100196107569, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16630210255108563, "3000–6000 Hz_10ms_envelope_correlation": 0.9825386905318858, "3000–6000 Hz_10ms_modulation_mae_db": 0.4896457674442594, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9640823081759211, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08522512822985129, "3000–6000 Hz_40ms_envelope_correlation": 0.9955322910020757, "3000–6000 Hz_40ms_modulation_mae_db": 0.5830628149523751, "mr_stft_db": 5.403804152093163, "band_mae_db": 0.09786421599548463, "rms_gain_error_db": -0.042422288932453985, "output_peak": 0.44192439389655586}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2520355528953286, "全帯域_10ms_rms_envelope_nrmse": 0.06033486423928493, "全帯域_10ms_envelope_correlation": 0.9963574911362455, "全帯域_10ms_modulation_mae_db": 0.6498187591082993, "全帯域_40ms_log_rms_mae_db": 0.7209488862479771, "全帯域_40ms_rms_envelope_nrmse": 0.032409454820269135, "全帯域_40ms_envelope_correlation": 0.9989489902899981, "全帯域_40ms_modulation_mae_db": 0.7801648585121755, "300–1000 Hz_10ms_log_rms_mae_db": 1.5305682093382198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09323431649978536, "300–1000 Hz_10ms_envelope_correlation": 0.9936963353003836, "300–1000 Hz_10ms_modulation_mae_db": 1.0346682896849415, "300–1000 Hz_40ms_log_rms_mae_db": 1.02587335684766, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058190874026498196, "300–1000 Hz_40ms_envelope_correlation": 0.9976277148062624, "300–1000 Hz_40ms_modulation_mae_db": 0.6930331453188353, "1000–3000 Hz_10ms_log_rms_mae_db": 1.397621710722475, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10892390097752082, "1000–3000 Hz_10ms_envelope_correlation": 0.9927510898735799, "1000–3000 Hz_10ms_modulation_mae_db": 0.542060691552647, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8177701712965716, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06674109356616756, "1000–3000 Hz_40ms_envelope_correlation": 0.997750503674695, "1000–3000 Hz_40ms_modulation_mae_db": 0.7314312224117366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4943929164413694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1421854412082863, "3000–6000 Hz_10ms_envelope_correlation": 0.9872013909407898, "3000–6000 Hz_10ms_modulation_mae_db": 0.10379155831459895, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9401583626187562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06939604997014365, "3000–6000 Hz_40ms_envelope_correlation": 0.9968612751889828, "3000–6000 Hz_40ms_modulation_mae_db": 0.49334812708861775, "mr_stft_db": 5.401794048183387, "band_mae_db": 0.1762827697696168, "rms_gain_error_db": -0.0736520150642135, "output_peak": 0.4345938193102651}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2440176753144518, "全帯域_10ms_rms_envelope_nrmse": 0.0540602476201049, "全帯域_10ms_envelope_correlation": 0.9970610128170201, "全帯域_10ms_modulation_mae_db": 0.24893101758959077, "全帯域_40ms_log_rms_mae_db": 0.6274502708729716, "全帯域_40ms_rms_envelope_nrmse": 0.025252082166464462, "全帯域_40ms_envelope_correlation": 0.9993519980504, "全帯域_40ms_modulation_mae_db": 0.14367077161755182, "300–1000 Hz_10ms_log_rms_mae_db": 1.571054353517317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08437442661822589, "300–1000 Hz_10ms_envelope_correlation": 0.9947644501657713, "300–1000 Hz_10ms_modulation_mae_db": 0.2846763551815921, "300–1000 Hz_40ms_log_rms_mae_db": 1.109332086016203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049396457589287554, "300–1000 Hz_40ms_envelope_correlation": 0.9981961771770053, "300–1000 Hz_40ms_modulation_mae_db": 0.6461548597355021, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4398275629456914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11548576319258455, "1000–3000 Hz_10ms_envelope_correlation": 0.9915780043320818, "1000–3000 Hz_10ms_modulation_mae_db": 0.6409960797011022, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8784267887737487, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06758079064507357, "1000–3000 Hz_40ms_envelope_correlation": 0.9974761905208199, "1000–3000 Hz_40ms_modulation_mae_db": 0.5661818181737083, "3000–6000 Hz_10ms_log_rms_mae_db": 1.510617798087771, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16894091377375264, "3000–6000 Hz_10ms_envelope_correlation": 0.9820881651533483, "3000–6000 Hz_10ms_modulation_mae_db": 0.3309740684496821, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0011095939739503, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0915765480013438, "3000–6000 Hz_40ms_envelope_correlation": 0.9945425227444783, "3000–6000 Hz_40ms_modulation_mae_db": 0.5879985590267348, "mr_stft_db": 5.4003249245276095, "band_mae_db": 0.10018854065424238, "rms_gain_error_db": -0.05093116564322809, "output_peak": 0.436138490455722}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3106272436605007, "全帯域_10ms_rms_envelope_nrmse": 0.0648817823997289, "全帯域_10ms_envelope_correlation": 0.9958029895403073, "全帯域_10ms_modulation_mae_db": 0.15043657618436532, "全帯域_40ms_log_rms_mae_db": 0.7097558954991312, "全帯域_40ms_rms_envelope_nrmse": 0.036503956571946265, "全帯域_40ms_envelope_correlation": 0.9986834413431738, "全帯域_40ms_modulation_mae_db": 0.36183006455635, "300–1000 Hz_10ms_log_rms_mae_db": 1.5962897373104836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0915483133860335, "300–1000 Hz_10ms_envelope_correlation": 0.9940669331597394, "300–1000 Hz_10ms_modulation_mae_db": 1.0075863159668097, "300–1000 Hz_40ms_log_rms_mae_db": 1.0582327401423495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06155536602666903, "300–1000 Hz_40ms_envelope_correlation": 0.9974531788178057, "300–1000 Hz_40ms_modulation_mae_db": 0.7137810546716579, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4649471794249964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1132378205031195, "1000–3000 Hz_10ms_envelope_correlation": 0.992296731924266, "1000–3000 Hz_10ms_modulation_mae_db": 1.2961535183343147, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9141208539481731, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06959415852486203, "1000–3000 Hz_40ms_envelope_correlation": 0.997708543783566, "1000–3000 Hz_40ms_modulation_mae_db": 1.3194536851182566, "3000–6000 Hz_10ms_log_rms_mae_db": 1.500849356360973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16522463560557393, "3000–6000 Hz_10ms_envelope_correlation": 0.9827091650859896, "3000–6000 Hz_10ms_modulation_mae_db": 1.0571207590373792, "3000–6000 Hz_40ms_log_rms_mae_db": 0.962629628459589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0893755511831972, "3000–6000 Hz_40ms_envelope_correlation": 0.9948533639524881, "3000–6000 Hz_40ms_modulation_mae_db": 1.3094489807334952, "mr_stft_db": 5.418662140046776, "band_mae_db": 0.21625011609681974, "rms_gain_error_db": -0.08971435152222941, "output_peak": 0.44543995362254396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.270216787126708, "全帯域_10ms_rms_envelope_nrmse": 0.062111560852730265, "全帯域_10ms_envelope_correlation": 0.9962173800839421, "全帯域_10ms_modulation_mae_db": 0.3268443474053479, "全帯域_40ms_log_rms_mae_db": 0.7228875698900217, "全帯域_40ms_rms_envelope_nrmse": 0.03191951980005645, "全帯域_40ms_envelope_correlation": 0.9990754841086377, "全帯域_40ms_modulation_mae_db": 0.2560735767211119, "300–1000 Hz_10ms_log_rms_mae_db": 1.5543673321190168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1013508009008967, "300–1000 Hz_10ms_envelope_correlation": 0.9925351814836811, "300–1000 Hz_10ms_modulation_mae_db": 0.3873562803534609, "300–1000 Hz_40ms_log_rms_mae_db": 1.052055696627147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06263693780742007, "300–1000 Hz_40ms_envelope_correlation": 0.9972491304852428, "300–1000 Hz_40ms_modulation_mae_db": 0.723403461939192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.473395537444616, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11543327581885407, "1000–3000 Hz_10ms_envelope_correlation": 0.9918828080052874, "1000–3000 Hz_10ms_modulation_mae_db": 0.6649645215514114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.855955557180497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07103503139841112, "1000–3000 Hz_40ms_envelope_correlation": 0.9974879180973583, "1000–3000 Hz_40ms_modulation_mae_db": 0.40636097303098384, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5288456886783777, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17933760888196598, "3000–6000 Hz_10ms_envelope_correlation": 0.9796333128689256, "3000–6000 Hz_10ms_modulation_mae_db": 0.5041525052757835, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0204395310949708, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11116234556837373, "3000–6000 Hz_40ms_envelope_correlation": 0.9919179069284848, "3000–6000 Hz_40ms_modulation_mae_db": 0.8139058436051968, "mr_stft_db": 5.412998154692546, "band_mae_db": 0.18599387181476904, "rms_gain_error_db": -0.11027299443969492, "output_peak": 0.42587754566220354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.218679834338585, "全帯域_10ms_rms_envelope_nrmse": 0.051767967264930353, "全帯域_10ms_envelope_correlation": 0.9973661624908097, "全帯域_10ms_modulation_mae_db": 0.5977833970196068, "全帯域_40ms_log_rms_mae_db": 0.6543700056722346, "全帯域_40ms_rms_envelope_nrmse": 0.02588228681783995, "全帯域_40ms_envelope_correlation": 0.9993883349259423, "全帯域_40ms_modulation_mae_db": 0.3995736490353156, "300–1000 Hz_10ms_log_rms_mae_db": 1.538400307089212, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08997948751644705, "300–1000 Hz_10ms_envelope_correlation": 0.99407332788263, "300–1000 Hz_10ms_modulation_mae_db": 0.3623417035342311, "300–1000 Hz_40ms_log_rms_mae_db": 1.011675411490707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058597195369146204, "300–1000 Hz_40ms_envelope_correlation": 0.997471869525673, "300–1000 Hz_40ms_modulation_mae_db": 0.2795101108995149, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3998334288230077, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10819912731583409, "1000–3000 Hz_10ms_envelope_correlation": 0.9930006634319973, "1000–3000 Hz_10ms_modulation_mae_db": 0.7168294974693289, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8338430309247263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0682968462603383, "1000–3000 Hz_40ms_envelope_correlation": 0.9977613472107737, "1000–3000 Hz_40ms_modulation_mae_db": 0.7414026202922321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4707884984104769, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14468279504210088, "3000–6000 Hz_10ms_envelope_correlation": 0.9867752405296553, "3000–6000 Hz_10ms_modulation_mae_db": 0.839287202255477, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9618095237079024, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07955432657212955, "3000–6000 Hz_40ms_envelope_correlation": 0.9959260102979975, "3000–6000 Hz_40ms_modulation_mae_db": 0.749324594129482, "mr_stft_db": 5.407940182189033, "band_mae_db": 0.19075884325962766, "rms_gain_error_db": -0.0890573125892946, "output_peak": 0.4165847775745973}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5929270937070195, "spectral_convergence": 0.18729607355674424, "spectral_convergence_status": "ok", "amplitude_floor": 1.0507333742302167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.20545875817002, "spectral_convergence": 0.17176438401431132, "spectral_convergence_status": "ok", "amplitude_floor": 5.595595188050952e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.427858618370253, "spectral_convergence": 0.17919252869469823, "spectral_convergence_status": "ok", "amplitude_floor": 5.269142735102521e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007416076491485813, "output_power": 0.0007539237000718778, "delta_db": 0.07153195256360484, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001352027771357195, "output_power": 0.0013237326875674177, "delta_db": -0.0918531909885654, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000371703582440024, "output_power": 0.0003642085279623275, "delta_db": -0.0884663619374834, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018903159721465413, "output_power": 0.00018009833721416605, "delta_db": -0.21024700756661496, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5427403014296353e-05, "output_power": 1.4601079014792147e-05, "delta_db": -0.2390787367973636, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1014091243658427e-05, "output_power": 2.0320603372398935e-05, "delta_db": -0.14574014488183853, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14115289912257845, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.055106247198640664, "pitch": null, "output_rms_dbfs": -25.75627253677223, "output_peak": 0.43915065291765226, "reference_rms_dbfs": -25.701166289573592, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.408748156749098}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018114550052554645, "log_rms_mae_db": 1.22829544858235, "rms_envelope_nrmse": 0.054570824070309866, "envelope_correlation": 0.9970138055490344, "rms_gain_error_db": -0.055248918762166435, "reference_mean_power": 0.002692702060931978, "output_mean_power": 0.002658663729930317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0110233554726448e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033858489572855915, "output_power": 0.00033137486078722275, "reference_active": true, "delta_db": -0.09348022037976547}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.201391758025471e-05, "output_power": 3.7456802766488566e-05, "reference_active": true, "delta_db": -0.49862474975405313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.1566777501091025e-06, "output_power": 6.025759316626033e-06, "reference_active": true, "delta_db": -0.7469968138891397}], "active_band_count": 3, "active_band_mae_db": 0.44636726134098614}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015936923488846764, "log_rms_mae_db": 0.6139651223032014, "rms_envelope_nrmse": 0.027937453774529223, "envelope_correlation": 0.9992162284792611, "rms_gain_error_db": -0.05638747889245786, "reference_mean_power": 0.002702794865208283, "output_mean_power": 0.002667929427006018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.583684100944911e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031173336044584235, "output_power": 0.0003041734581901686, "reference_active": true, "delta_db": -0.10661965849660479}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.830633609453067e-05, "output_power": 1.6149911527984333e-05, "reference_active": true, "delta_db": -0.5443128398027807}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.642076702713698e-07, "output_power": 4.0775365789598045e-07, "reference_active": true, "delta_db": -0.563144464646556}], "active_band_count": 3, "active_band_mae_db": 0.40469232098198055}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001259260583850441, "log_rms_mae_db": 1.4695896688155512, "rms_envelope_nrmse": 0.09100517926501053, "envelope_correlation": 0.9938721187922461, "rms_gain_error_db": -0.060898058832856046, "reference_mean_power": 0.0007105481356967998, "output_mean_power": 0.000700654149559676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8068878391559104e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016083630318364556, "output_power": 0.00015052144457726745, "reference_active": true, "delta_db": -0.28785704645100835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8008354908350678e-05, "output_power": 1.4555799466545211e-05, "reference_active": true, "delta_db": -0.924379774111549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.325041103743567e-06, "output_power": 2.026323134144525e-06, "reference_active": true, "delta_db": -0.5972193234789287}], "active_band_count": 3, "active_band_mae_db": 0.6031520480138287}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010786919997015104, "log_rms_mae_db": 0.9674877589354913, "rms_envelope_nrmse": 0.054728302004603706, "envelope_correlation": 0.9977343453426093, "rms_gain_error_db": -0.061502535279064345, "reference_mean_power": 0.0007140165693307253, "output_mean_power": 0.0007039762968361472}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6527252040517484e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000151264236566862, "output_power": 0.00014090581384048888, "reference_active": true, "delta_db": -0.30807347155380455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.688501064647705e-06, "output_power": 6.136299270657458e-06, "reference_active": true, "delta_db": -0.9793514660264523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2694181944274492e-07, "output_power": 1.7265404951366248e-07, "reference_active": true, "delta_db": -1.187377634126071}], "active_band_count": 3, "active_band_mae_db": 0.824934190568776}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.418520257385562e-05, "log_rms_mae_db": 1.4574045091246217, "rms_envelope_nrmse": 0.11502645923140015, "envelope_correlation": 0.9913925260119756, "rms_gain_error_db": -0.1974530670676884, "reference_mean_power": 0.000153818152796868, "output_mean_power": 0.0001469813679072787}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.418336470516266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4748514150479256e-05, "output_power": 2.1982595292920868e-05, "reference_active": true, "delta_db": -0.514701654426651}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.277056498060538e-06, "output_power": 3.0360627967941613e-06, "reference_active": true, "delta_db": -0.331731787947456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.913659986768058e-07, "output_power": 6.829979715461695e-07, "reference_active": true, "delta_db": -1.1563664996092502}], "active_band_count": 3, "active_band_mae_db": 0.667599980661119}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.421085502718712e-05, "log_rms_mae_db": 0.8392055644100872, "rms_envelope_nrmse": 0.06927804511881937, "envelope_correlation": 0.9969911785206965, "rms_gain_error_db": -0.1974910917072711, "reference_mean_power": 0.00015458728634615244, "output_mean_power": 0.00014771502230635783}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1974511373265086e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3402826997507283e-05, "output_power": 2.081179553654614e-05, "reference_active": true, "delta_db": -0.5095877160364999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3826120793884328e-06, "output_power": 1.2627954654895615e-06, "reference_active": true, "delta_db": -0.39367333102746827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6109211314667126e-08, "output_power": 3.3598078965165475e-08, "reference_active": true, "delta_db": -1.3747324731891453}], "active_band_count": 3, "active_band_mae_db": 0.7593311734177045}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.007885661618194e-05, "log_rms_mae_db": 1.5468489378505532, "rms_envelope_nrmse": 0.18112240166604138, "envelope_correlation": 0.9796048756815129, "rms_gain_error_db": -0.3833783313456982, "reference_mean_power": 2.0928545188569684e-05, "output_mean_power": 1.9160251572384035e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4882901242606597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1348414996445462e-05, "output_power": 1.003648745121451e-05, "reference_active": true, "delta_db": -0.5335346305597934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8308296968161944e-06, "output_power": 2.652586917003652e-06, "reference_active": true, "delta_db": -0.282441196402024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.670421987321005e-07, "output_power": 4.690383303879339e-07, "reference_active": true, "delta_db": -1.52944974080676}], "active_band_count": 3, "active_band_mae_db": 0.7818085225895258}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3258654185194873e-05, "log_rms_mae_db": 1.0147269415836837, "rms_envelope_nrmse": 0.11104938248517722, "envelope_correlation": 0.9928036324473631, "rms_gain_error_db": -0.3834186439913101, "reference_mean_power": 2.103354073620274e-05, "output_mean_power": 1.9256197098143663e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.390796824652454e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1287290435083056e-05, "output_power": 1.0004334739331852e-05, "reference_active": true, "delta_db": -0.5240148554237098}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.514784832162418e-06, "output_power": 1.4624322599477092e-06, "reference_active": true, "delta_db": -0.15275189151469437}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.433812491322156e-08, "output_power": 2.708150245712849e-08, "reference_active": true, "delta_db": -2.1410456763965287}], "active_band_count": 3, "active_band_mae_db": 0.939270807778311}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9052833168687172, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "97a3ebbadf788b0c93fc27f45ba1c6355e7eb3695522af71789f97b3bd8f486c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9922123307263573, "standard_deviation": 0.0006369470930336936, "minimum": 0.9913925260119756, "maximum": 0.9930006634319973, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4350066569325313, "standard_deviation": 0.030876174676156925, "minimum": 1.397621710722475, "maximum": 1.473395537444616, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7298009049466143, "standard_deviation": 0.2399082457866157, "minimum": 0.542060691552647, "maximum": 1.2961535183343147, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11182497808284259, "standard_deviation": 0.003972612862644575, "minimum": 0.10491024073357451, "maximum": 0.11548576319258455, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9976476669800336, "standard_deviation": 0.0003311452052030212, "minimum": 0.9969911785206965, "maximum": 0.9980361225210596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8491542748136567, "standard_deviation": 0.032723250308115084, "minimum": 0.8177701712965716, "maximum": 0.9141208539481731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7619508304617337, "standard_deviation": 0.26528288445311965, "minimum": 0.40636097303098384, "maximum": 1.3194536851182566, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06652867026841597, "standard_deviation": 0.004325158621307901, "minimum": 0.059503149484851095, "maximum": 0.07103503139841112, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9827643077182426, "standard_deviation": 0.002866386236101848, "minimum": 0.9796048756815129, "maximum": 0.9872013909407898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5068508623253056, "standard_deviation": 0.025600431369724393, "minimum": 1.4707884984104769, "maximum": 1.5468489378505532, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7016650485810766, "standard_deviation": 0.4437443255297828, "minimum": 0.10379155831459895, "maximum": 1.5065400052819067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.165406999761526, "standard_deviation": 0.014755904367508559, "minimum": 0.1421854412082863, "maximum": 0.18112240166604138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944315716830219, "standard_deviation": 0.001711822864707919, "minimum": 0.9919179069284848, "maximum": 0.9968612751889828, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9818772441109169, "standard_deviation": 0.028867520245271578, "minimum": 0.9401583626187562, "maximum": 1.0204395310949708, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8676968599837254, "standard_deviation": 0.35326760311688243, "minimum": 0.49334812708861775, "maximum": 1.4652151525555908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09404534873748799, "standard_deviation": 0.016671128751731216, "minimum": 0.06939604997014365, "maximum": 0.11502345788968742, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9941137754948968, "standard_deviation": 0.0008209137233834911, "minimum": 0.9925351814836811, "maximum": 0.9952371310655421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5225387525386673, "standard_deviation": 0.05486493256301668, "minimum": 1.434249174447375, "maximum": 1.5962897373104836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5432807662281696, "standard_deviation": 0.3103534002256733, "minimum": 0.2846763551815921, "maximum": 1.0346682896849415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08966109371617177, "standard_deviation": 0.00636695138932489, "minimum": 0.08072381924778299, "maximum": 0.1013508009008967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977177640032873, "standard_deviation": 0.0003918659588758728, "minimum": 0.9972491304852428, "maximum": 0.9984065495584449, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0261447847330927, "standard_deviation": 0.054356697703464844, "minimum": 0.9369618174734468, "maximum": 1.109332086016203, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5814846583501667, "standard_deviation": 0.20107874584964083, "minimum": 0.2795101108995149, "maximum": 0.824934190568776, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056057395795919264, "standard_deviation": 0.005530519752104867, "minimum": 0.04692958845541336, "maximum": 0.06263693780742007, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16289529855721782, "standard_deviation": 0.04468860008651296, "minimum": 0.09786421599548463, "maximum": 0.21625011609681974, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.4073160460195195, "standard_deviation": 0.006147269838404853, "minimum": 5.4003249245276095, "maximum": 5.418662140046776, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4348494685291687, "standard_deviation": 0.009369277848597772, "minimum": 0.4165847775745973, "maximum": 0.44543995362254396, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06976582621184703, "standard_deviation": 0.024677907955026366, "minimum": -0.11027299443969492, "maximum": -0.042422288932453985, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967039774839566, "standard_deviation": 0.0005522222704289059, "minimum": 0.9958029895403073, "maximum": 0.9973661624908097, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2496058111514783, "standard_deviation": 0.03452502869626104, "minimum": 1.2018204956562153, "maximum": 1.3106272436605007, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38482274674289585, "standard_deviation": 0.19364124720382578, "minimum": 0.15043657618436532, "maximum": 0.6498187591082993, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05729487447455754, "standard_deviation": 0.004838931322377117, "minimum": 0.051767967264930353, "maximum": 0.0648817823997289, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991631256993203, "standard_deviation": 0.00024822606322394425, "minimum": 0.9986834413431738, "maximum": 0.9993883349259423, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6634286274782895, "standard_deviation": 0.04726021741111135, "minimum": 0.6139547299404348, "maximum": 0.7228875698900217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.39705302857342556, "standard_deviation": 0.21866808480311822, "minimum": 0.14367077161755182, "maximum": 0.7801648585121755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028928141983825085, "standard_deviation": 0.004239145841908665, "minimum": 0.024439638930923477, "maximum": 0.036503956571946265, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0037", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.8871660232543945, "predicted_mos": 2.8871660232543945, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bf9fcac5dfab5fb90021f0ab842f0f0e35de9d61e0f3c6b13ffb9df85f5bd888", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.7102648997679353, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1493980534728734, "全帯域_10ms_rms_envelope_nrmse": 0.055783975806309796, "全帯域_10ms_envelope_correlation": 0.9968796272861126, "全帯域_10ms_modulation_mae_db": 0.12791950360930565, "全帯域_40ms_log_rms_mae_db": 0.626367908160084, "全帯域_40ms_rms_envelope_nrmse": 0.03213359962575846, "全帯域_40ms_envelope_correlation": 0.9989474058472372, "全帯域_40ms_modulation_mae_db": 0.2066560929997466, "300–1000 Hz_10ms_log_rms_mae_db": 1.7751692029014254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09223808773385155, "300–1000 Hz_10ms_envelope_correlation": 0.9937802385679603, "300–1000 Hz_10ms_modulation_mae_db": 0.3106542689345884, "300–1000 Hz_40ms_log_rms_mae_db": 1.2951382315113478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057295463326870806, "300–1000 Hz_40ms_envelope_correlation": 0.9976189354930853, "300–1000 Hz_40ms_modulation_mae_db": 0.2543296742054252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6975973456527553, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11169205857630024, "1000–3000 Hz_10ms_envelope_correlation": 0.9918638175899407, "1000–3000 Hz_10ms_modulation_mae_db": 1.0035801905986164, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0711870926154092, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05828542215773519, "1000–3000 Hz_40ms_envelope_correlation": 0.9979131410246764, "1000–3000 Hz_40ms_modulation_mae_db": 0.936881922095731, "3000–6000 Hz_10ms_log_rms_mae_db": 1.890157633542355, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17195809792088876, "3000–6000 Hz_10ms_envelope_correlation": 0.9812158815882758, "3000–6000 Hz_10ms_modulation_mae_db": 1.5735916511241825, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3040290372538208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09045607401929817, "3000–6000 Hz_40ms_envelope_correlation": 0.9947185491734853, "3000–6000 Hz_40ms_modulation_mae_db": 1.8371526641624707, "mr_stft_db": 5.5636424678416985, "band_mae_db": 0.11164250240732392, "rms_gain_error_db": -0.06177562974120168, "output_peak": 0.4379455572529207}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1848683104837507, "全帯域_10ms_rms_envelope_nrmse": 0.055145287590303786, "全帯域_10ms_envelope_correlation": 0.9969627292277891, "全帯域_10ms_modulation_mae_db": 0.12356915594986036, "全帯域_40ms_log_rms_mae_db": 0.5946631423799533, "全帯域_40ms_rms_envelope_nrmse": 0.026288675200111626, "全帯域_40ms_envelope_correlation": 0.9993292137920017, "全帯域_40ms_modulation_mae_db": 0.15968026662704013, "300–1000 Hz_10ms_log_rms_mae_db": 1.6879739442618371, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08227098200761321, "300–1000 Hz_10ms_envelope_correlation": 0.9951403108494509, "300–1000 Hz_10ms_modulation_mae_db": 0.42338670640110904, "300–1000 Hz_40ms_log_rms_mae_db": 1.2030946893860675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04947965076445404, "300–1000 Hz_40ms_envelope_correlation": 0.9983359789853199, "300–1000 Hz_40ms_modulation_mae_db": 0.6357665930276394, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8076133847820495, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11025218074907657, "1000–3000 Hz_10ms_envelope_correlation": 0.9919950421152451, "1000–3000 Hz_10ms_modulation_mae_db": 0.8387176360015638, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1858809437944058, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06283957826001814, "1000–3000 Hz_40ms_envelope_correlation": 0.9974132581586527, "1000–3000 Hz_40ms_modulation_mae_db": 0.7766042360728248, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8899531522992512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15361896973850545, "3000–6000 Hz_10ms_envelope_correlation": 0.9850693454803522, "3000–6000 Hz_10ms_modulation_mae_db": 0.4748582774405234, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3040378830324761, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08815606329527985, "3000–6000 Hz_40ms_envelope_correlation": 0.9948803749547805, "3000–6000 Hz_40ms_modulation_mae_db": 0.7422452945062318, "mr_stft_db": 5.58397913127119, "band_mae_db": 0.14267343151557613, "rms_gain_error_db": -0.06828847046286352, "output_peak": 0.45072564103186685}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2156362727438816, "全帯域_10ms_rms_envelope_nrmse": 0.06540449765710794, "全帯域_10ms_envelope_correlation": 0.995705551254011, "全帯域_10ms_modulation_mae_db": 0.21731146519621758, "全帯域_40ms_log_rms_mae_db": 0.6334204166050845, "全帯域_40ms_rms_envelope_nrmse": 0.03770947771682896, "全帯域_40ms_envelope_correlation": 0.9985510173736896, "全帯域_40ms_modulation_mae_db": 0.21932980556715517, "300–1000 Hz_10ms_log_rms_mae_db": 1.6862308027090434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08282729678275619, "300–1000 Hz_10ms_envelope_correlation": 0.9949447462247517, "300–1000 Hz_10ms_modulation_mae_db": 0.19884922700985105, "300–1000 Hz_40ms_log_rms_mae_db": 1.1384236762234283, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04975582740719429, "300–1000 Hz_40ms_envelope_correlation": 0.9981522009101897, "300–1000 Hz_40ms_modulation_mae_db": 0.32041052619499233, "1000–3000 Hz_10ms_log_rms_mae_db": 1.70277914216753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11304429151326167, "1000–3000 Hz_10ms_envelope_correlation": 0.9920341137127013, "1000–3000 Hz_10ms_modulation_mae_db": 1.0389707798878962, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1069300540589662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06914034434314481, "1000–3000 Hz_40ms_envelope_correlation": 0.9974097926427817, "1000–3000 Hz_40ms_modulation_mae_db": 1.0634310810437753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8815262569844058, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16997448962737663, "3000–6000 Hz_10ms_envelope_correlation": 0.9816609470091456, "3000–6000 Hz_10ms_modulation_mae_db": 0.560428568734077, "3000–6000 Hz_40ms_log_rms_mae_db": 1.280812955743755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08299194660455086, "3000–6000 Hz_40ms_envelope_correlation": 0.9956240935377663, "3000–6000 Hz_40ms_modulation_mae_db": 0.6268045931210436, "mr_stft_db": 5.574236748842442, "band_mae_db": 0.1510953530139039, "rms_gain_error_db": -0.06918847412696898, "output_peak": 0.43447056697043124}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1675176959999902, "全帯域_10ms_rms_envelope_nrmse": 0.059533412987953026, "全帯域_10ms_envelope_correlation": 0.9964422649883143, "全帯域_10ms_modulation_mae_db": 0.30060906414983607, "全帯域_40ms_log_rms_mae_db": 0.6016224361577254, "全帯域_40ms_rms_envelope_nrmse": 0.02992759953659892, "全帯域_40ms_envelope_correlation": 0.9990949198975742, "全帯域_40ms_modulation_mae_db": 0.3528327488139213, "300–1000 Hz_10ms_log_rms_mae_db": 1.7717156662038922, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08836706837143483, "300–1000 Hz_10ms_envelope_correlation": 0.9942446001325366, "300–1000 Hz_10ms_modulation_mae_db": 0.4768060291653055, "300–1000 Hz_40ms_log_rms_mae_db": 1.274050575922994, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05042768258454204, "300–1000 Hz_40ms_envelope_correlation": 0.9981257841356606, "300–1000 Hz_40ms_modulation_mae_db": 0.4600198831705999, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7450212560556955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10821681564900765, "1000–3000 Hz_10ms_envelope_correlation": 0.9924822250908759, "1000–3000 Hz_10ms_modulation_mae_db": 0.652251977276081, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1226974614186256, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06509994361246133, "1000–3000 Hz_40ms_envelope_correlation": 0.9974565544679709, "1000–3000 Hz_40ms_modulation_mae_db": 0.6673375209510265, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8547510832365444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15652713894798498, "3000–6000 Hz_10ms_envelope_correlation": 0.9844910938269956, "3000–6000 Hz_10ms_modulation_mae_db": 0.7605885442712629, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2737544683685507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08064826451872929, "3000–6000 Hz_40ms_envelope_correlation": 0.9957294555146028, "3000–6000 Hz_40ms_modulation_mae_db": 1.0621571960994907, "mr_stft_db": 5.5715614741933495, "band_mae_db": 0.14986589000219644, "rms_gain_error_db": -0.0589124676613977, "output_peak": 0.4326996245681748}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1961091615126815, "全帯域_10ms_rms_envelope_nrmse": 0.05988877147831846, "全帯域_10ms_envelope_correlation": 0.9963873075770268, "全帯域_10ms_modulation_mae_db": 0.07975404658656975, "全帯域_40ms_log_rms_mae_db": 0.5972234088735996, "全帯域_40ms_rms_envelope_nrmse": 0.031175206199793495, "全帯域_40ms_envelope_correlation": 0.9990062900189106, "全帯域_40ms_modulation_mae_db": 0.16756297977616105, "300–1000 Hz_10ms_log_rms_mae_db": 1.7646006384207384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0945169095061699, "300–1000 Hz_10ms_envelope_correlation": 0.9933917986356934, "300–1000 Hz_10ms_modulation_mae_db": 0.08877871140843425, "300–1000 Hz_40ms_log_rms_mae_db": 1.208662321398758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053651226871370694, "300–1000 Hz_40ms_envelope_correlation": 0.9978577839754784, "300–1000 Hz_40ms_modulation_mae_db": 0.014989022207770063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7151071265399807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11082397680095354, "1000–3000 Hz_10ms_envelope_correlation": 0.9922459238647193, "1000–3000 Hz_10ms_modulation_mae_db": 0.6197538035867818, "1000–3000 Hz_40ms_log_rms_mae_db": 1.115571765714507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058631234382923135, "1000–3000 Hz_40ms_envelope_correlation": 0.9982071858259072, "1000–3000 Hz_40ms_modulation_mae_db": 0.6944055126843068, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8924739764448955, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15964912155380193, "3000–6000 Hz_10ms_envelope_correlation": 0.983849893521592, "3000–6000 Hz_10ms_modulation_mae_db": 1.0858379413346129, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2931417319358678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08464520915769143, "3000–6000 Hz_40ms_envelope_correlation": 0.9952874826118027, "3000–6000 Hz_40ms_modulation_mae_db": 1.3618064906789027, "mr_stft_db": 5.573971573155739, "band_mae_db": 0.12330254166368432, "rms_gain_error_db": -0.05433347688543036, "output_peak": 0.44641075768136}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2288428862678893, "全帯域_10ms_rms_envelope_nrmse": 0.056081919967136705, "全帯域_10ms_envelope_correlation": 0.9968889889376966, "全帯域_10ms_modulation_mae_db": 0.13502528743238537, "全帯域_40ms_log_rms_mae_db": 0.6368602417816951, "全帯域_40ms_rms_envelope_nrmse": 0.03212327424750868, "全帯域_40ms_envelope_correlation": 0.9989873989732785, "全帯域_40ms_modulation_mae_db": 0.17571610377255187, "300–1000 Hz_10ms_log_rms_mae_db": 1.8272813509858665, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09415604702424077, "300–1000 Hz_10ms_envelope_correlation": 0.9934348817871385, "300–1000 Hz_10ms_modulation_mae_db": 0.419764809258246, "300–1000 Hz_40ms_log_rms_mae_db": 1.3649347354992947, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0615400340321199, "300–1000 Hz_40ms_envelope_correlation": 0.9971278241147675, "300–1000 Hz_40ms_modulation_mae_db": 0.6753788045894353, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7285415326094542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10628055562985368, "1000–3000 Hz_10ms_envelope_correlation": 0.9926363029610336, "1000–3000 Hz_10ms_modulation_mae_db": 1.305696525744152, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1092930018772573, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0596408736855183, "1000–3000 Hz_40ms_envelope_correlation": 0.9977729840216155, "1000–3000 Hz_40ms_modulation_mae_db": 0.7688409437503078, "3000–6000 Hz_10ms_log_rms_mae_db": 1.899040409481009, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16501806905851704, "3000–6000 Hz_10ms_envelope_correlation": 0.9834033307799415, "3000–6000 Hz_10ms_modulation_mae_db": 1.3470478811122486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2722615497577776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10288503001121535, "3000–6000 Hz_40ms_envelope_correlation": 0.994152127465131, "3000–6000 Hz_40ms_modulation_mae_db": 1.4703416545874521, "mr_stft_db": 5.577837259347523, "band_mae_db": 0.1382496288954079, "rms_gain_error_db": -0.07577013924366645, "output_peak": 0.4350953704381471}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.208399746172422, "全帯域_10ms_rms_envelope_nrmse": 0.06020584114475362, "全帯域_10ms_envelope_correlation": 0.9963304576811293, "全帯域_10ms_modulation_mae_db": 0.20257340304195823, "全帯域_40ms_log_rms_mae_db": 0.6808049418126264, "全帯域_40ms_rms_envelope_nrmse": 0.03176751433556742, "全帯域_40ms_envelope_correlation": 0.9989344059222464, "全帯域_40ms_modulation_mae_db": 0.36631163788268006, "300–1000 Hz_10ms_log_rms_mae_db": 1.8296675709367474, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10075386953900432, "300–1000 Hz_10ms_envelope_correlation": 0.9925058975309714, "300–1000 Hz_10ms_modulation_mae_db": 0.37484709735231725, "300–1000 Hz_40ms_log_rms_mae_db": 1.3287912250066631, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06007264038961542, "300–1000 Hz_40ms_envelope_correlation": 0.99731035873103, "300–1000 Hz_40ms_modulation_mae_db": 0.4535008229853706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7143524837985085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12940393563035468, "1000–3000 Hz_10ms_envelope_correlation": 0.9895939846686601, "1000–3000 Hz_10ms_modulation_mae_db": 1.976320705434759, "1000–3000 Hz_40ms_log_rms_mae_db": 1.175863997115699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08575785870431463, "1000–3000 Hz_40ms_envelope_correlation": 0.9959255262688378, "1000–3000 Hz_40ms_modulation_mae_db": 1.8442862571285403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8879336893270042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.169209205246674, "3000–6000 Hz_10ms_envelope_correlation": 0.9820705352742662, "3000–6000 Hz_10ms_modulation_mae_db": 1.0321887923142776, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3424868097515146, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09920209132497873, "3000–6000 Hz_40ms_envelope_correlation": 0.993600294140758, "3000–6000 Hz_40ms_modulation_mae_db": 1.3397903961928253, "mr_stft_db": 5.5837003922746264, "band_mae_db": 0.15626505445533503, "rms_gain_error_db": -0.025101434173461435, "output_peak": 0.4346485411664244}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2351477467819194, "全帯域_10ms_rms_envelope_nrmse": 0.06336744724028984, "全帯域_10ms_envelope_correlation": 0.9960092629204832, "全帯域_10ms_modulation_mae_db": 0.12979416561080426, "全帯域_40ms_log_rms_mae_db": 0.7009737036149218, "全帯域_40ms_rms_envelope_nrmse": 0.03433618095482962, "全帯域_40ms_envelope_correlation": 0.9988402692344364, "全帯域_40ms_modulation_mae_db": 0.36499666444810447, "300–1000 Hz_10ms_log_rms_mae_db": 1.6985617591449942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08611117194468607, "300–1000 Hz_10ms_envelope_correlation": 0.9948328822913344, "300–1000 Hz_10ms_modulation_mae_db": 0.7301100584281252, "300–1000 Hz_40ms_log_rms_mae_db": 1.169698805721377, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05542749323678071, "300–1000 Hz_40ms_envelope_correlation": 0.9980559989808917, "300–1000 Hz_40ms_modulation_mae_db": 0.751593538184412, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7304778741610964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11103894377959682, "1000–3000 Hz_10ms_envelope_correlation": 0.9920717813381589, "1000–3000 Hz_10ms_modulation_mae_db": 0.8736522804217691, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1232530085097299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06740283779633635, "1000–3000 Hz_40ms_envelope_correlation": 0.9972538485327204, "1000–3000 Hz_40ms_modulation_mae_db": 1.0170471419277085, "3000–6000 Hz_10ms_log_rms_mae_db": 1.875975837926837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16622299603708246, "3000–6000 Hz_10ms_envelope_correlation": 0.9825891528975664, "3000–6000 Hz_10ms_modulation_mae_db": 0.5553092859796745, "3000–6000 Hz_40ms_log_rms_mae_db": 1.294156686769429, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09432120242618393, "3000–6000 Hz_40ms_envelope_correlation": 0.9941521217740711, "3000–6000 Hz_40ms_modulation_mae_db": 0.6551777702096243, "mr_stft_db": 5.568042864918518, "band_mae_db": 0.13680334833716892, "rms_gain_error_db": -0.08988306633990058, "output_peak": 0.40696600231603264}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.816741386029479, "spectral_convergence": 0.1799962035530116, "spectral_convergence_status": "ok", "amplitude_floor": 1.0507333742302167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.37511915428396, "spectral_convergence": 0.16941598293718704, "spectral_convergence_status": "ok", "amplitude_floor": 5.595595188050952e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.499066863211657, "spectral_convergence": 0.17653192479327273, "spectral_convergence_status": "ok", "amplitude_floor": 5.269142735102521e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007416076491485813, "output_power": 0.0007484888253859053, "delta_db": 0.04011120304671298, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001352027771357195, "output_power": 0.0013294106507062038, "delta_db": -0.07326458487243999, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000371703582440024, "output_power": 0.00035593387187984416, "delta_db": -0.1882742788749982, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018903159721465413, "output_power": 0.00018341056673468518, "delta_db": -0.13110051073519222, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5427403014296353e-05, "output_power": 1.4743593489625579e-05, "delta_db": -0.19689476856971175, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1014091243658427e-05, "output_power": 2.082042770588966e-05, "delta_db": -0.0402096683448883, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11164250240732392, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06177562974120168, "pitch": null, "output_rms_dbfs": -25.762941919314795, "output_peak": 0.4379455572529207, "reference_rms_dbfs": -25.701166289573592, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5636424678416985}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018114550052554645, "log_rms_mae_db": 1.1493980534728734, "rms_envelope_nrmse": 0.055783975806309796, "envelope_correlation": 0.9968796272861126, "rms_gain_error_db": -0.06150705499604569, "reference_mean_power": 0.002692702060931978, "output_mean_power": 0.0026548353833865087}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0110233554726448e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033858489572855915, "output_power": 0.00033289500895860853, "reference_active": true, "delta_db": -0.07360296431093585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.201391758025471e-05, "output_power": 3.9234471027264526e-05, "reference_active": true, "delta_db": -0.2972537746165186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.1566777501091025e-06, "output_power": 7.135448650170971e-06, "reference_active": true, "delta_db": -0.012901771900462532}], "active_band_count": 3, "active_band_mae_db": 0.12791950360930565}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015936923488846764, "log_rms_mae_db": 0.626367908160084, "rms_envelope_nrmse": 0.03213359962575846, "envelope_correlation": 0.9989474058472372, "rms_gain_error_db": -0.058962363420661586, "reference_mean_power": 0.002702794865208283, "output_mean_power": 0.002666348109620213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.583684100944911e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031173336044584235, "output_power": 0.0003064896174364601, "reference_active": true, "delta_db": -0.07367514150864073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.830633609453067e-05, "output_power": 1.7437421670017065e-05, "reference_active": true, "delta_db": -0.21119161769215944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.642076702713698e-07, "output_power": 5.014440288779492e-07, "reference_active": true, "delta_db": 0.3351015197984396}], "active_band_count": 3, "active_band_mae_db": 0.2066560929997466}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001259260583850441, "log_rms_mae_db": 1.7751692029014254, "rms_envelope_nrmse": 0.09223808773385155, "envelope_correlation": 0.9937802385679603, "rms_gain_error_db": -0.11970654647540742, "reference_mean_power": 0.0007105481356967998, "output_mean_power": 0.0006912304313395406}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8068878391559104e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016083630318364556, "output_power": 0.00015405020260126633, "reference_active": true, "delta_db": -0.18721808481565097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8008354908350678e-05, "output_power": 1.687132787654418e-05, "reference_active": true, "delta_db": -0.28324775582538547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.325041103743567e-06, "output_power": 2.090648101749755e-06, "reference_active": true, "delta_db": -0.46149696616272873}], "active_band_count": 3, "active_band_mae_db": 0.3106542689345884}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010786919997015104, "log_rms_mae_db": 1.2951382315113478, "rms_envelope_nrmse": 0.057295463326870806, "envelope_correlation": 0.9976189354930853, "rms_gain_error_db": -0.12147725647119575, "reference_mean_power": 0.0007140165693307253, "output_mean_power": 0.0006943214213377103}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6527252040517484e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000151264236566862, "output_power": 0.00014573048829131442, "reference_active": true, "delta_db": -0.16185839758623927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.688501064647705e-06, "output_power": 7.492389710064972e-06, "reference_active": true, "delta_db": -0.11221320102061544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2694181944274492e-07, "output_power": 2.0277891266993844e-07, "reference_active": true, "delta_db": -0.48891742400942084}], "active_band_count": 3, "active_band_mae_db": 0.2543296742054252}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.418520257385562e-05, "log_rms_mae_db": 1.6975973456527553, "rms_envelope_nrmse": 0.11169205857630024, "envelope_correlation": 0.9918638175899407, "rms_gain_error_db": -0.1738514582774285, "reference_mean_power": 0.000153818152796868, "output_mean_power": 0.00014778230829905886}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.418336470516266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4748514150479256e-05, "output_power": 2.292281443910067e-05, "reference_active": true, "delta_db": -0.3328119112149818}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.277056498060538e-06, "output_power": 2.707268112335313e-06, "reference_active": true, "delta_db": -0.8295266095706694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.913659986768058e-07, "output_power": 5.823926142354503e-07, "reference_active": true, "delta_db": -1.8484020510101982}], "active_band_count": 3, "active_band_mae_db": 1.0035801905986164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.421085502718712e-05, "log_rms_mae_db": 1.0711870926154092, "rms_envelope_nrmse": 0.05828542215773519, "envelope_correlation": 0.9979131410246764, "rms_gain_error_db": -0.17390654200809375, "reference_mean_power": 0.00015458728634615244, "output_mean_power": 0.00014851937718982677}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1974511373265086e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3402826997507283e-05, "output_power": 2.142429277829265e-05, "reference_active": true, "delta_db": -0.3836182751148509}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3826120793884328e-06, "output_power": 1.1586385888514161e-06, "reference_active": true, "delta_db": -0.7675235810990254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6109211314667126e-08, "output_power": 3.146569312343669e-08, "reference_active": true, "delta_db": -1.6595039100733169}], "active_band_count": 3, "active_band_mae_db": 0.936881922095731}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.007885661618194e-05, "log_rms_mae_db": 1.890157633542355, "rms_envelope_nrmse": 0.17195809792088876, "envelope_correlation": 0.9812158815882758, "rms_gain_error_db": -0.1484235318232083, "reference_mean_power": 2.0928545188569684e-05, "output_mean_power": 2.0225379859109835e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4882901242606597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1348414996445462e-05, "output_power": 1.1094693379133475e-05, "reference_active": true, "delta_db": -0.09819904724348709}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8308296968161944e-06, "output_power": 2.1612186022171073e-06, "reference_active": true, "delta_db": -1.1721504594481116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.670421987321005e-07, "output_power": 3.0137745744481856e-07, "reference_active": true, "delta_db": -3.450425446680949}], "active_band_count": 3, "active_band_mae_db": 1.5735916511241825}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3258654185194873e-05, "log_rms_mae_db": 1.3040290372538208, "rms_envelope_nrmse": 0.09045607401929817, "envelope_correlation": 0.9947185491734853, "rms_gain_error_db": -0.14846855853125598, "reference_mean_power": 2.103354073620274e-05, "output_mean_power": 2.0326636982609235e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.390796824652454e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1287290435083056e-05, "output_power": 1.084775834555421e-05, "reference_active": true, "delta_db": -0.17249698194211155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.514784832162418e-06, "output_power": 1.1075880899253046e-06, "reference_active": true, "delta_db": -1.3597267073522974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.433812491322156e-08, "output_power": 1.7735926987111322e-08, "reference_active": true, "delta_db": -3.979234303193003}], "active_band_count": 3, "active_band_mae_db": 1.8371526641624707}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8923662772085076, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5e6ecc939d09872f95d85ddfd63d39b5882164376d6b1eb6155e3830287368b4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9918653989176669, "standard_deviation": 0.0009536963984400305, "minimum": 0.9895939846686601, "maximum": 0.9926363029610336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7301862682208837, "standard_deviation": 0.03486525718065831, "minimum": 1.6975973456527553, "maximum": 1.8076133847820495, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0386179873689523, "standard_deviation": 0.43813644938366814, "minimum": 0.6197538035867818, "maximum": 1.976320705434759, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11259409479105062, "standard_deviation": 0.007107790681848069, "minimum": 0.10628055562985368, "maximum": 0.12940393563035468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974190363678953, "standard_deviation": 0.000681383445693614, "minimum": 0.9959255262688378, "maximum": 0.9982071858259072, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.126334665638075, "standard_deviation": 0.03750369860505185, "minimum": 1.0711870926154092, "maximum": 1.1858809437944058, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9711043269567776, "standard_deviation": 0.3820677007568484, "minimum": 0.6673375209510265, "maximum": 1.8442862571285403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06584976161780648, "standard_deviation": 0.009000574561396066, "minimum": 0.05828542215773519, "maximum": 0.08575785870431463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9830437725472669, "standard_deviation": 0.0013827539489949378, "minimum": 0.9812158815882758, "maximum": 0.9850693454803522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.883976504905288, "standard_deviation": 0.01369490977475005, "minimum": 1.8547510832365444, "maximum": 1.899040409481009, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9237313677888574, "standard_deviation": 0.40273763148520364, "minimum": 0.4748582774405234, "maximum": 1.5735916511241825, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1640222610163539, "standard_deviation": 0.006704747406986461, "minimum": 0.15361896973850545, "maximum": 0.17195809792088876, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947680623965498, "standard_deviation": 0.0007619871072752763, "minimum": 0.993600294140758, "maximum": 0.9957294555146028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.295585140326649, "standard_deviation": 0.022644176564702006, "minimum": 1.2722615497577776, "maximum": 1.3424868097515146, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1369345074447552, "standard_deviation": 0.4386467856077092, "minimum": 0.6268045931210436, "maximum": 1.8371526641624707, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09041323516974095, "standard_deviation": 0.00789989897818979, "minimum": 0.08064826451872929, "maximum": 0.10288503001121535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9940344195024796, "standard_deviation": 0.0009186016974905372, "minimum": 0.9925058975309714, "maximum": 0.9951403108494509, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.755150116945568, "standard_deviation": 0.058535717721536124, "minimum": 1.6862308027090434, "maximum": 1.8296675709367474, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3778996134947471, "standard_deviation": 0.19192172435722765, "minimum": 0.08877871140843425, "maximum": 0.7301100584281252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09015517911371959, "standard_deviation": 0.006404375073549952, "minimum": 0.08227098200761321, "maximum": 0.10075386953900432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978231081658029, "standard_deviation": 0.0004320069729839772, "minimum": 0.9971278241147675, "maximum": 0.9983359789853199, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2478492825837413, "standard_deviation": 0.08000304073801817, "minimum": 1.1384236762234283, "maximum": 1.3649347354992947, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4457486080707056, "standard_deviation": 0.24510825975149758, "minimum": 0.014989022207770063, "maximum": 0.751593538184412, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05470625232661849, "standard_deviation": 0.004689901062266663, "minimum": 0.04947965076445404, "maximum": 0.0615400340321199, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13873721878632456, "standard_deviation": 0.015002952618268573, "minimum": 0.11164250240732392, "maximum": 0.15626505445533503, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.574621488980636, "standard_deviation": 0.0071122072029499205, "minimum": 5.5636424678416985, "maximum": 5.58397913127119, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4348702576781697, "standard_deviation": 0.012970584804155324, "minimum": 0.40696600231603264, "maximum": 0.45072564103186685, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06290664482936134, "standard_deviation": 0.018828504379740638, "minimum": -0.08988306633990058, "maximum": -0.025101434173461435, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964507737340704, "standard_deviation": 0.00044831605705239263, "minimum": 0.995705551254011, "maximum": 0.9969627292277891, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.198239984179426, "standard_deviation": 0.029802462565832497, "minimum": 1.1493980534728734, "maximum": 1.2351477467819194, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.16456951144711715, "standard_deviation": 0.07074387456246872, "minimum": 0.07975404658656975, "maximum": 0.30060906414983607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05942639423402164, "standard_deviation": 0.0036816726938210622, "minimum": 0.055145287590303786, "maximum": 0.06540449765710794, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989613651324218, "standard_deviation": 0.0002203298430623039, "minimum": 0.9985510173736896, "maximum": 0.9993292137920017, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6339920249232113, "standard_deviation": 0.03908079073926052, "minimum": 0.5946631423799533, "maximum": 0.7009737036149218, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2516357874859201, "standard_deviation": 0.09304128626080047, "minimum": 0.15968026662704013, "maximum": 0.36631163788268006, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031932690977124645, "standard_deviation": 0.00328853175366072, "minimum": 0.026288675200111626, "maximum": 0.03770947771682896, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1188-133604-0037", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7691495418548584, "predicted_mos": 2.7691495418548584, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "04c1f163734340cae6ac2325bd7ea31b130db17e8e9fa05b4a918606ab66d787", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.732269000262022, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2413042078509213, "全帯域_10ms_rms_envelope_nrmse": 0.057025966528694, "全帯域_10ms_envelope_correlation": 0.9967278047234465, "全帯域_10ms_modulation_mae_db": 0.2505451012160463, "全帯域_40ms_log_rms_mae_db": 0.6638391947371858, "全帯域_40ms_rms_envelope_nrmse": 0.032942640764494976, "全帯域_40ms_envelope_correlation": 0.9988856447698189, "全帯域_40ms_modulation_mae_db": 0.3674376046355892, "300–1000 Hz_10ms_log_rms_mae_db": 1.6620637767893116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09153574967522073, "300–1000 Hz_10ms_envelope_correlation": 0.993786524028679, "300–1000 Hz_10ms_modulation_mae_db": 0.7542407929985, "300–1000 Hz_40ms_log_rms_mae_db": 1.1265242097695896, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05860880976641068, "300–1000 Hz_40ms_envelope_correlation": 0.9973881690518117, "300–1000 Hz_40ms_modulation_mae_db": 0.8302237552030611, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6264634326533263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11129162959532626, "1000–3000 Hz_10ms_envelope_correlation": 0.9921316688465173, "1000–3000 Hz_10ms_modulation_mae_db": 0.9390260942418109, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9409503324272903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06178534449021999, "1000–3000 Hz_40ms_envelope_correlation": 0.9978859522022939, "1000–3000 Hz_40ms_modulation_mae_db": 0.7077344750346541, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8157578787102207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15491301718881467, "3000–6000 Hz_10ms_envelope_correlation": 0.9853235213436988, "3000–6000 Hz_10ms_modulation_mae_db": 0.3084651702186246, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2375705950446538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08834247908630964, "3000–6000 Hz_40ms_envelope_correlation": 0.9951696561502228, "3000–6000 Hz_40ms_modulation_mae_db": 0.22678312753435648, "mr_stft_db": 5.5232004554334395, "band_mae_db": 0.09592456546624062, "rms_gain_error_db": -0.053984732834862076, "output_peak": 0.4379002502111504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2004481254342716, "全帯域_10ms_rms_envelope_nrmse": 0.05652699822212455, "全帯域_10ms_envelope_correlation": 0.9968003481166808, "全帯域_10ms_modulation_mae_db": 0.03102010000318954, "全帯域_40ms_log_rms_mae_db": 0.620121073463881, "全帯域_40ms_rms_envelope_nrmse": 0.027020382459606695, "全帯域_40ms_envelope_correlation": 0.999273753163067, "全帯域_40ms_modulation_mae_db": 0.053312002167945995, "300–1000 Hz_10ms_log_rms_mae_db": 1.679787814882857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0907332097403946, "300–1000 Hz_10ms_envelope_correlation": 0.9938988714418585, "300–1000 Hz_10ms_modulation_mae_db": 0.14951997690597932, "300–1000 Hz_40ms_log_rms_mae_db": 1.0808674933215685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05220326276101139, "300–1000 Hz_40ms_envelope_correlation": 0.9979427814247357, "300–1000 Hz_40ms_modulation_mae_db": 0.20916253668529117, "1000–3000 Hz_10ms_log_rms_mae_db": 1.607866261795669, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11150379940788635, "1000–3000 Hz_10ms_envelope_correlation": 0.992184802399519, "1000–3000 Hz_10ms_modulation_mae_db": 0.41461500861908535, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0213693405307605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06508165604511304, "1000–3000 Hz_40ms_envelope_correlation": 0.9976977034601758, "1000–3000 Hz_40ms_modulation_mae_db": 0.4962494970441889, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7843957293724846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17400687476643895, "3000–6000 Hz_10ms_envelope_correlation": 0.9807664133948408, "3000–6000 Hz_10ms_modulation_mae_db": 0.4775786834213875, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1791628860231103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09607915783311766, "3000–6000 Hz_40ms_envelope_correlation": 0.9940463147377956, "3000–6000 Hz_40ms_modulation_mae_db": 0.8954474725480286, "mr_stft_db": 5.531092905384135, "band_mae_db": 0.1302484464986424, "rms_gain_error_db": -0.06246322698004797, "output_peak": 0.4196921657068441}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.187450304085477, "全帯域_10ms_rms_envelope_nrmse": 0.05418582394003402, "全帯域_10ms_envelope_correlation": 0.9970471792127026, "全帯域_10ms_modulation_mae_db": 0.12257147582575932, "全帯域_40ms_log_rms_mae_db": 0.5749802566987063, "全帯域_40ms_rms_envelope_nrmse": 0.02489236053814709, "全帯域_40ms_envelope_correlation": 0.999380966204621, "全帯域_40ms_modulation_mae_db": 0.17198726458197267, "300–1000 Hz_10ms_log_rms_mae_db": 1.6541292077800018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09338643567690189, "300–1000 Hz_10ms_envelope_correlation": 0.9935840307973133, "300–1000 Hz_10ms_modulation_mae_db": 0.2281340247091638, "300–1000 Hz_40ms_log_rms_mae_db": 1.1193278703612788, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05535839732685789, "300–1000 Hz_40ms_envelope_correlation": 0.9977233467645453, "300–1000 Hz_40ms_modulation_mae_db": 0.21382825035778186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6581075179639555, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1087274944263421, "1000–3000 Hz_10ms_envelope_correlation": 0.9928239913616501, "1000–3000 Hz_10ms_modulation_mae_db": 1.199411184139536, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0148528850112746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06588166724136882, "1000–3000 Hz_40ms_envelope_correlation": 0.997876077968347, "1000–3000 Hz_40ms_modulation_mae_db": 0.954142593111237, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7948022870444897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1766415786245459, "3000–6000 Hz_10ms_envelope_correlation": 0.9801792690208574, "3000–6000 Hz_10ms_modulation_mae_db": 0.9621802426008487, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2001075791541644, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09760909535765065, "3000–6000 Hz_40ms_envelope_correlation": 0.9937477980077548, "3000–6000 Hz_40ms_modulation_mae_db": 1.4130525701195833, "mr_stft_db": 5.548542516594456, "band_mae_db": 0.16505482321893125, "rms_gain_error_db": -0.047865867964002135, "output_peak": 0.4299445791881081}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1888526369944228, "全帯域_10ms_rms_envelope_nrmse": 0.052408270528020606, "全帯域_10ms_envelope_correlation": 0.9972370429037096, "全帯域_10ms_modulation_mae_db": 0.26996174736708906, "全帯域_40ms_log_rms_mae_db": 0.5902401738536573, "全帯域_40ms_rms_envelope_nrmse": 0.025564166817079542, "全帯域_40ms_envelope_correlation": 0.9993325923579034, "全帯域_40ms_modulation_mae_db": 0.46648403130044586, "300–1000 Hz_10ms_log_rms_mae_db": 1.5720498003640426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08526792332283314, "300–1000 Hz_10ms_envelope_correlation": 0.9946294106175486, "300–1000 Hz_10ms_modulation_mae_db": 0.3998793965426793, "300–1000 Hz_40ms_log_rms_mae_db": 1.0566152666914244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04676327535155797, "300–1000 Hz_40ms_envelope_correlation": 0.9983772199738717, "300–1000 Hz_40ms_modulation_mae_db": 0.7625471558448872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5535145061559543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10087874737552703, "1000–3000 Hz_10ms_envelope_correlation": 0.9933189868084487, "1000–3000 Hz_10ms_modulation_mae_db": 0.9115501565815892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9221457605645699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05321918908957575, "1000–3000 Hz_40ms_envelope_correlation": 0.9981813674996227, "1000–3000 Hz_40ms_modulation_mae_db": 1.09408313311935, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7633706521689378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1568901131605713, "3000–6000 Hz_10ms_envelope_correlation": 0.9843961003951389, "3000–6000 Hz_10ms_modulation_mae_db": 0.8394988783463271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1734567944021501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08333518370864071, "3000–6000 Hz_40ms_envelope_correlation": 0.9954450088251435, "3000–6000 Hz_40ms_modulation_mae_db": 1.2032880511991486, "mr_stft_db": 5.51854458557424, "band_mae_db": 0.09638366735054092, "rms_gain_error_db": -0.04681689410232413, "output_peak": 0.4421901100160487}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1886054960384709, "全帯域_10ms_rms_envelope_nrmse": 0.05612700069819567, "全帯域_10ms_envelope_correlation": 0.9968314363224637, "全帯域_10ms_modulation_mae_db": 0.43757881849813235, "全帯域_40ms_log_rms_mae_db": 0.6663122975474903, "全帯域_40ms_rms_envelope_nrmse": 0.031684082776356495, "全帯域_40ms_envelope_correlation": 0.9989560663586987, "全帯域_40ms_modulation_mae_db": 0.5296705132630967, "300–1000 Hz_10ms_log_rms_mae_db": 1.6603487447539202, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09098834260327407, "300–1000 Hz_10ms_envelope_correlation": 0.9938764353356883, "300–1000 Hz_10ms_modulation_mae_db": 0.6241690284288332, "300–1000 Hz_40ms_log_rms_mae_db": 1.1351617566970846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05807191401203063, "300–1000 Hz_40ms_envelope_correlation": 0.9974208927454201, "300–1000 Hz_40ms_modulation_mae_db": 0.4360033613145811, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6315058797479807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12212449176642955, "1000–3000 Hz_10ms_envelope_correlation": 0.990374079580942, "1000–3000 Hz_10ms_modulation_mae_db": 1.179296624973388, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9641804542887041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07086551630864668, "1000–3000 Hz_40ms_envelope_correlation": 0.9969859154678221, "1000–3000 Hz_40ms_modulation_mae_db": 1.162548874860928, "3000–6000 Hz_10ms_log_rms_mae_db": 1.840058401395313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1644896644518695, "3000–6000 Hz_10ms_envelope_correlation": 0.9834429939935175, "3000–6000 Hz_10ms_modulation_mae_db": 1.7375008588714131, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2546327383906397, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09897938466232893, "3000–6000 Hz_40ms_envelope_correlation": 0.9945705291351078, "3000–6000 Hz_40ms_modulation_mae_db": 2.445845191595457, "mr_stft_db": 5.529705534162205, "band_mae_db": 0.12281097860380397, "rms_gain_error_db": -0.033314085939055424, "output_peak": 0.44037233475158055}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2230789449635182, "全帯域_10ms_rms_envelope_nrmse": 0.05918162366561135, "全帯域_10ms_envelope_correlation": 0.9964993015199477, "全帯域_10ms_modulation_mae_db": 0.29982022879243875, "全帯域_40ms_log_rms_mae_db": 0.6466087921679213, "全帯域_40ms_rms_envelope_nrmse": 0.0297116159084034, "全帯域_40ms_envelope_correlation": 0.9991305751772749, "全帯域_40ms_modulation_mae_db": 0.44711980961329895, "300–1000 Hz_10ms_log_rms_mae_db": 1.6441634303025354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09062485901567616, "300–1000 Hz_10ms_envelope_correlation": 0.9939901400960558, "300–1000 Hz_10ms_modulation_mae_db": 0.626811632422115, "300–1000 Hz_40ms_log_rms_mae_db": 1.1700287724824088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056278975739635195, "300–1000 Hz_40ms_envelope_correlation": 0.9976954352453224, "300–1000 Hz_40ms_modulation_mae_db": 0.531871790762308, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6214536227997847, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12557159314036792, "1000–3000 Hz_10ms_envelope_correlation": 0.9904005277324871, "1000–3000 Hz_10ms_modulation_mae_db": 0.9089013836546274, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0468926669176062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07588618277752147, "1000–3000 Hz_40ms_envelope_correlation": 0.9972136405660602, "1000–3000 Hz_40ms_modulation_mae_db": 1.0176779644409302, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8376136688269962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16036369169106568, "3000–6000 Hz_10ms_envelope_correlation": 0.9845102743614893, "3000–6000 Hz_10ms_modulation_mae_db": 1.1535351029538028, "3000–6000 Hz_40ms_log_rms_mae_db": 1.224294671754168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0978840246790665, "3000–6000 Hz_40ms_envelope_correlation": 0.9949610785924968, "3000–6000 Hz_40ms_modulation_mae_db": 1.3628393687924294, "mr_stft_db": 5.529608955242568, "band_mae_db": 0.16370961098781678, "rms_gain_error_db": -0.07308273997547735, "output_peak": 0.4189180005271863}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1761068726761263, "全帯域_10ms_rms_envelope_nrmse": 0.0594547193988739, "全帯域_10ms_envelope_correlation": 0.996437764115776, "全帯域_10ms_modulation_mae_db": 0.29000504400654237, "全帯域_40ms_log_rms_mae_db": 0.6469424529582871, "全帯域_40ms_rms_envelope_nrmse": 0.029066974442039536, "全帯域_40ms_envelope_correlation": 0.9991100526091766, "全帯域_40ms_modulation_mae_db": 0.4498545293606957, "300–1000 Hz_10ms_log_rms_mae_db": 1.6595807788058474, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.086833802108468, "300–1000 Hz_10ms_envelope_correlation": 0.9943901605301984, "300–1000 Hz_10ms_modulation_mae_db": 0.46718051353427104, "300–1000 Hz_40ms_log_rms_mae_db": 1.0439792188645995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05197183192074749, "300–1000 Hz_40ms_envelope_correlation": 0.9979083806963089, "300–1000 Hz_40ms_modulation_mae_db": 0.27795372666054363, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6534331103798168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11332761252173465, "1000–3000 Hz_10ms_envelope_correlation": 0.991604649102794, "1000–3000 Hz_10ms_modulation_mae_db": 0.9341562737974574, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9966833994502634, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06128382744566277, "1000–3000 Hz_40ms_envelope_correlation": 0.9976532655654031, "1000–3000 Hz_40ms_modulation_mae_db": 0.6952242622373817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7794494121626296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1591472802410373, "3000–6000 Hz_10ms_envelope_correlation": 0.9840031203428198, "3000–6000 Hz_10ms_modulation_mae_db": 1.506635127641716, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2018577024767514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0956439255371625, "3000–6000 Hz_40ms_envelope_correlation": 0.9941988203754775, "3000–6000 Hz_40ms_modulation_mae_db": 1.574634336778235, "mr_stft_db": 5.545817915129625, "band_mae_db": 0.08471104980667725, "rms_gain_error_db": -0.015678824593205274, "output_peak": 0.44973431332149616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1721413275914032, "全帯域_10ms_rms_envelope_nrmse": 0.054927566127146464, "全帯域_10ms_envelope_correlation": 0.9969994884747051, "全帯域_10ms_modulation_mae_db": 0.2399492328594205, "全帯域_40ms_log_rms_mae_db": 0.6386902735106378, "全帯域_40ms_rms_envelope_nrmse": 0.027501453439894057, "全帯域_40ms_envelope_correlation": 0.9992727206429, "全帯域_40ms_modulation_mae_db": 0.29582149142332814, "300–1000 Hz_10ms_log_rms_mae_db": 1.6183284929870583, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07266632244139029, "300–1000 Hz_10ms_envelope_correlation": 0.9961438372339377, "300–1000 Hz_10ms_modulation_mae_db": 0.28594762707090776, "300–1000 Hz_40ms_log_rms_mae_db": 1.0457481587775155, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038787579820980224, "300–1000 Hz_40ms_envelope_correlation": 0.9989203464678954, "300–1000 Hz_40ms_modulation_mae_db": 0.3545412456494857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.63315682114186, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10564847865325708, "1000–3000 Hz_10ms_envelope_correlation": 0.9930684465083274, "1000–3000 Hz_10ms_modulation_mae_db": 0.7576022285165633, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0021500990597028, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06380903795166874, "1000–3000 Hz_40ms_envelope_correlation": 0.9978278661807978, "1000–3000 Hz_40ms_modulation_mae_db": 1.0035455862755545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8450399018447763, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18388863158797963, "3000–6000 Hz_10ms_envelope_correlation": 0.9784879538966645, "3000–6000 Hz_10ms_modulation_mae_db": 1.8057980890523426, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2195506772441016, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10199546690438147, "3000–6000 Hz_40ms_envelope_correlation": 0.9932556385847607, "3000–6000 Hz_40ms_modulation_mae_db": 2.039491198574209, "mr_stft_db": 5.5282113954323835, "band_mae_db": 0.16918910715957994, "rms_gain_error_db": -0.07677880935592413, "output_peak": 0.4264072468076537}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.791933050600004, "spectral_convergence": 0.1865401891487129, "spectral_convergence_status": "ok", "amplitude_floor": 1.0507333742302167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3437246978964525, "spectral_convergence": 0.17325234299457293, "spectral_convergence_status": "ok", "amplitude_floor": 5.595595188050952e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.433943617803863, "spectral_convergence": 0.17941129663426844, "spectral_convergence_status": "ok", "amplitude_floor": 5.269142735102521e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007416076491485813, "output_power": 0.0007586596809460111, "delta_db": 0.09872803454098947, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001352027771357195, "output_power": 0.001321811295884493, "delta_db": -0.09816153405453422, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000371703582440024, "output_power": 0.00035880320017362376, "delta_db": -0.1534043892907287, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018903159721465413, "output_power": 0.00018177226146661035, "delta_db": -0.1700679350551854, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5427403014296353e-05, "output_power": 1.5349435206255865e-05, "delta_db": -0.022004249424710985, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1014091243658427e-05, "output_power": 2.117525937461796e-05, "delta_db": 0.03318125043129492, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09592456546624062, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.053984732834862076, "pitch": null, "output_rms_dbfs": -25.755151022408455, "output_peak": 0.4379002502111504, "reference_rms_dbfs": -25.701166289573592, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5232004554334395}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018114550052554645, "log_rms_mae_db": 1.2413042078509213, "rms_envelope_nrmse": 0.057025966528694, "envelope_correlation": 0.9967278047234465, "rms_gain_error_db": -0.053862036744615246, "reference_mean_power": 0.002692702060931978, "output_mean_power": 0.00265951288686883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0110233554726448e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033858489572855915, "output_power": 0.0003273820760104934, "reference_active": true, "delta_db": -0.14612681987596615}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.201391758025471e-05, "output_power": 3.885405413699802e-05, "reference_active": true, "delta_db": -0.33956838176622395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.1566777501091025e-06, "output_power": 6.731586830321342e-06, "reference_active": true, "delta_db": -0.26594010200594886}], "active_band_count": 3, "active_band_mae_db": 0.2505451012160463}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015936923488846764, "log_rms_mae_db": 0.6638391947371858, "rms_envelope_nrmse": 0.032942640764494976, "envelope_correlation": 0.9988856447698189, "rms_gain_error_db": -0.054223513482159404, "reference_mean_power": 0.002702794865208283, "output_mean_power": 0.002669259111313265}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.583684100944911e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031173336044584235, "output_power": 0.00030261467142920497, "reference_active": true, "delta_db": -0.12893301450662445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.830633609453067e-05, "output_power": 1.72903240860607e-05, "reference_active": true, "delta_db": -0.2479829784225116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.642076702713698e-07, "output_power": 3.9280104100374234e-07, "reference_active": true, "delta_db": -0.7253968209776317}], "active_band_count": 3, "active_band_mae_db": 0.3674376046355892}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001259260583850441, "log_rms_mae_db": 1.6620637767893116, "rms_envelope_nrmse": 0.09153574967522073, "envelope_correlation": 0.993786524028679, "rms_gain_error_db": -0.08736808206270567, "reference_mean_power": 0.0007105481356967998, "output_mean_power": 0.0006963966865768038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8068878391559104e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016083630318364556, "output_power": 0.0001460766969710871, "reference_active": true, "delta_db": -0.4180314209837477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8008354908350678e-05, "output_power": 1.4852782000867885e-05, "reference_active": true, "delta_db": -0.8366623432905065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.325041103743567e-06, "output_power": 1.843434765337015e-06, "reference_active": true, "delta_db": -1.0080286147212458}], "active_band_count": 3, "active_band_mae_db": 0.7542407929985}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010786919997015104, "log_rms_mae_db": 1.1265242097695896, "rms_envelope_nrmse": 0.05860880976641068, "envelope_correlation": 0.9973881690518117, "rms_gain_error_db": -0.08822253241893183, "reference_mean_power": 0.0007140165693307253, "output_mean_power": 0.0006996583745896814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6527252040517484e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000151264236566862, "output_power": 0.00013734551691278716, "reference_active": true, "delta_db": -0.4192177162362357}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.688501064647705e-06, "output_power": 6.574943137833348e-06, "reference_active": true, "delta_db": -0.6794967757510504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2694181944274492e-07, "output_power": 1.647095781976852e-07, "reference_active": true, "delta_db": -1.3919567736218972}], "active_band_count": 3, "active_band_mae_db": 0.8302237552030611}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.418520257385562e-05, "log_rms_mae_db": 1.6264634326533263, "rms_envelope_nrmse": 0.11129162959532626, "envelope_correlation": 0.9921316688465173, "rms_gain_error_db": -0.25253246136388785, "reference_mean_power": 0.000153818152796868, "output_mean_power": 0.00014512904767592817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.418336470516266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4748514150479256e-05, "output_power": 2.2342734833366106e-05, "reference_active": true, "delta_db": -0.4441279862737485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.277056498060538e-06, "output_power": 2.510328032400391e-06, "reference_active": true, "delta_db": -1.1575345309176999}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.913659986768058e-07, "output_power": 6.737743709743088e-07, "reference_active": true, "delta_db": -1.2154157655339843}], "active_band_count": 3, "active_band_mae_db": 0.9390260942418109}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.421085502718712e-05, "log_rms_mae_db": 0.9409503324272903, "rms_envelope_nrmse": 0.06178534449021999, "envelope_correlation": 0.9978859522022939, "rms_gain_error_db": -0.2525459786162907, "reference_mean_power": 0.00015458728634615244, "output_mean_power": 0.00014585427931602422}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1974511373265086e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3402826997507283e-05, "output_power": 2.098739354398628e-05, "reference_active": true, "delta_db": -0.47309815919366704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3826120793884328e-06, "output_power": 1.073191452192358e-06, "reference_active": true, "delta_db": -1.1002314191512768}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6109211314667126e-08, "output_power": 4.062564873498008e-08, "reference_active": true, "delta_db": -0.5498738467590187}], "active_band_count": 3, "active_band_mae_db": 0.7077344750346541}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.007885661618194e-05, "log_rms_mae_db": 1.8157578787102207, "rms_envelope_nrmse": 0.15491301718881467, "envelope_correlation": 0.9853235213436988, "rms_gain_error_db": 0.1449335867236286, "reference_mean_power": 2.0928545188569684e-05, "output_mean_power": 2.1638761432406833e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4882901242606597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1348414996445462e-05, "output_power": 1.1690250983715436e-05, "reference_active": true, "delta_db": 0.12888626356664512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8308296968161944e-06, "output_power": 2.6963893635851644e-06, "reference_active": true, "delta_db": -0.2113113760906947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.670421987321005e-07, "output_power": 7.63261042922563e-07, "reference_active": true, "delta_db": 0.5851978709985338}], "active_band_count": 3, "active_band_mae_db": 0.3084651702186246}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3258654185194873e-05, "log_rms_mae_db": 1.2375705950446538, "rms_envelope_nrmse": 0.08834247908630964, "envelope_correlation": 0.9951696561502228, "rms_gain_error_db": 0.1449141774586949, "reference_mean_power": 2.103354073620274e-05, "output_mean_power": 2.1747222842581766e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.390796824652454e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1287290435083056e-05, "output_power": 1.169895471009604e-05, "reference_active": true, "delta_db": 0.15557359612451557}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.514784832162418e-06, "output_power": 1.5018047815997299e-06, "reference_active": true, "delta_db": -0.03737465049468719}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.433812491322156e-08, "output_power": 4.960408421896567e-08, "reference_active": true, "delta_db": 0.4874011359838667}], "active_band_count": 3, "active_band_mae_db": 0.22678312753435648}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8955777765201445, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "df922e2107538504713a0a59fa7fc09b1bb66509f11fad76283a02c7148cb213", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9919883940425858, "standard_deviation": 0.0011319331130982305, "minimum": 0.990374079580942, "maximum": 0.9933189868084487, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6231876440797934, "standard_deviation": 0.03252601570511688, "minimum": 1.5535145061559543, "maximum": 1.6581075179639555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9055698693155072, "standard_deviation": 0.246745536175184, "minimum": 0.41461500861908535, "maximum": 1.199411184139536, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11238423086085886, "standard_deviation": 0.008130685631003862, "minimum": 0.10087874737552703, "maximum": 0.12557159314036792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9976652236138154, "standard_deviation": 0.0003877816995648223, "minimum": 0.9969859154678221, "maximum": 0.9981813674996227, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9886531172812715, "standard_deviation": 0.04258264837384912, "minimum": 0.9221457605645699, "maximum": 1.0468926669176062, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.891400798265528, "standard_deviation": 0.23158565824482266, "minimum": 0.4962494970441889, "maximum": 1.162548874860928, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06472655266872215, "standard_deviation": 0.006734229398388639, "minimum": 0.05321918908957575, "maximum": 0.07588618277752147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9826387058436283, "standard_deviation": 0.0024810998475334697, "minimum": 0.9784879538966645, "maximum": 0.9853235213436988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.807560991440731, "standard_deviation": 0.03133588457993404, "minimum": 1.7633706521689378, "maximum": 1.8450399018447763, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0988990191383077, "standard_deviation": 0.5572326426934908, "minimum": 0.3084651702186246, "maximum": 1.8057980890523426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16629260646404037, "standard_deviation": 0.010580247835976567, "minimum": 0.15491301718881467, "maximum": 0.18388863158797963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994424355551095, "standard_deviation": 0.0007487184152680012, "minimum": 0.9932556385847607, "maximum": 0.9954450088251435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2113292055612175, "standard_deviation": 0.027990121464777157, "minimum": 1.1734567944021501, "maximum": 1.2546327383906397, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.395172664642681, "standard_deviation": 0.6765785605088627, "minimum": 0.22678312753435648, "maximum": 2.445845191595457, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09498358972108226, "standard_deviation": 0.006116308732880776, "minimum": 0.08333518370864071, "maximum": 0.10199546690438147, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.99428742626016, "standard_deviation": 0.0008218189761871, "minimum": 0.9935840307973133, "maximum": 0.9961438372339377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6438065058331968, "standard_deviation": 0.033922801832043124, "minimum": 1.5720498003640426, "maximum": 1.679787814882857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.44198537407655614, "standard_deviation": 0.21476863677277774, "minimum": 0.14951997690597932, "maximum": 0.7542407929985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08775458057301985, "standard_deviation": 0.006636154322253663, "minimum": 0.07266632244139029, "maximum": 0.09338643567690189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979220715462389, "standard_deviation": 0.0005114987591832569, "minimum": 0.9973881690518117, "maximum": 0.9989203464678954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0972815933706839, "standard_deviation": 0.0470383408968678, "minimum": 1.0439792188645995, "maximum": 1.1700287724824088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45201647780974247, "standard_deviation": 0.23952759107984153, "minimum": 0.20916253668529117, "maximum": 0.8302237552030611, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05225550583740393, "standard_deviation": 0.006672375328185222, "minimum": 0.038787579820980224, "maximum": 0.05860880976641068, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12850403113652914, "standard_deviation": 0.03438663998140689, "minimum": 0.08471104980667725, "maximum": 0.16918910715957994, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.531840532869131, "standard_deviation": 0.010344248721966527, "minimum": 5.51854458557424, "maximum": 5.548542516594456, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.43314487506625854, "standard_deviation": 0.011154066393095604, "minimum": 0.4189180005271863, "maximum": 0.44973431332149616, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.051248147718112314, "standard_deviation": 0.02028269546775601, "minimum": -0.07677880935592413, "maximum": -0.015678824593205274, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996822545673679, "standard_deviation": 0.00027141644029568486, "minimum": 0.996437764115776, "maximum": 0.9972370429037096, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1972484894543265, "standard_deviation": 0.023715245466339616, "minimum": 1.1721413275914032, "maximum": 1.2413042078509213, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24268146857107728, "standard_deviation": 0.12161402436930073, "minimum": 0.03102010000318954, "maximum": 0.43757881849813235, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05622974613858757, "standard_deviation": 0.0024001155441982502, "minimum": 0.052408270528020606, "maximum": 0.0594547193988739, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991677964104326, "standard_deviation": 0.0001788830117839723, "minimum": 0.9988856447698189, "maximum": 0.999380966204621, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6309668143672209, "standard_deviation": 0.03338139959900185, "minimum": 0.5749802566987063, "maximum": 0.6663122975474903, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.34771090579329667, "standard_deviation": 0.16395250644828352, "minimum": 0.053312002167945995, "maximum": 0.5296705132630967, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028547959643252727, "standard_deviation": 0.002840197539925991, "minimum": 0.02489236053814709, "maximum": 0.032942640764494976, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_121-121726-0012", "language": "en", "style": "normal", "speaker": "librispeech_121", "source_utterance_id": "121-121726-0012", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 64720, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_121-121726-0012.wav", "source_file_sha256": "de25b20e46f4a5aa94bc0bf5071e0e1ce164039e4dfa9b3b2fffeb30bf9bbd8b", "decoded_pcm_sha256": "9c9a8032f3e173a5a354dda74e6cb24f7034610d2cd1b6d15dff6a794afd035e", "original_source_sha256": "6d1fab44fd2e6930a69e1f6f7e52a141ed5290b9a45f56df5e0c3177fa05c6ba", "corpus_entry_id": "librispeech_121-121726-0012"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.6820976734161377, "predicted_mos": 3.6820976734161377, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "26fa69114e5ce7a8b5b9b707fdc1678969c8e4bee09c2438054694023a7a465b", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.77547109965235, "cpu_threads": 4}, "waveform_sha256": "d7b6a60f190aeff3e64279914fcc2517ac55145c4993566d15cb7fa6c01169ef"}, "rows": [{"case": "librispeech_121-121726-0012", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.7367422580718994, "predicted_mos": 2.7367422580718994, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "9865dfae22f3fbe7fad2ea48c34cab5fbde734ce9632c38156f5100e7d634b4c", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.225236200261861, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6152536547466612, "全帯域_10ms_rms_envelope_nrmse": 0.06733245536413261, "全帯域_10ms_envelope_correlation": 0.9959659264377441, "全帯域_10ms_modulation_mae_db": 1.5378804042407124, "全帯域_40ms_log_rms_mae_db": 0.4388204989403843, "全帯域_40ms_rms_envelope_nrmse": 0.03629087326045861, "全帯域_40ms_envelope_correlation": 0.9988419048978696, "全帯域_40ms_modulation_mae_db": 2.117884809281593, "300–1000 Hz_10ms_log_rms_mae_db": 0.7996913922635195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07000904847149247, "300–1000 Hz_10ms_envelope_correlation": 0.9961128524041005, "300–1000 Hz_10ms_modulation_mae_db": 0.7094784125087633, "300–1000 Hz_40ms_log_rms_mae_db": 0.575635192181237, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04149855555494797, "300–1000 Hz_40ms_envelope_correlation": 0.9986186199080418, "300–1000 Hz_40ms_modulation_mae_db": 0.9429318262004927, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7052305324122597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11061395801201931, "1000–3000 Hz_10ms_envelope_correlation": 0.9909144240924868, "1000–3000 Hz_10ms_modulation_mae_db": 1.719881033766816, "1000–3000 Hz_40ms_log_rms_mae_db": 0.48605544430882586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06110223087752498, "1000–3000 Hz_40ms_envelope_correlation": 0.9972355438858459, "1000–3000 Hz_40ms_modulation_mae_db": 2.56916758053646, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9393513487334597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13712501777938874, "3000–6000 Hz_10ms_envelope_correlation": 0.9866511029090056, "3000–6000 Hz_10ms_modulation_mae_db": 1.5708267999557537, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6851530594751848, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08794296687784318, "3000–6000 Hz_40ms_envelope_correlation": 0.9951386523089536, "3000–6000 Hz_40ms_modulation_mae_db": 2.02420234370168, "mr_stft_db": 4.663363187314674, "band_mae_db": 0.1551012543094301, "rms_gain_error_db": -0.09094880042144013, "output_peak": 0.502015764418237}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5957455167532432, "全帯域_10ms_rms_envelope_nrmse": 0.05127587909551006, "全帯域_10ms_envelope_correlation": 0.997643596720999, "全帯域_10ms_modulation_mae_db": 0.6647841397122144, "全帯域_40ms_log_rms_mae_db": 0.437181293006885, "全帯域_40ms_rms_envelope_nrmse": 0.0258001419621886, "全帯域_40ms_envelope_correlation": 0.99939159139, "全帯域_40ms_modulation_mae_db": 0.8647328993441139, "300–1000 Hz_10ms_log_rms_mae_db": 0.8025531869385598, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05604710544449034, "300–1000 Hz_10ms_envelope_correlation": 0.9974856568132402, "300–1000 Hz_10ms_modulation_mae_db": 0.32502862797061094, "300–1000 Hz_40ms_log_rms_mae_db": 0.6194848362603771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.025970134640195743, "300–1000 Hz_40ms_envelope_correlation": 0.9994367002484705, "300–1000 Hz_40ms_modulation_mae_db": 0.5492085197434415, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7411889515548363, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08651587202377874, "1000–3000 Hz_10ms_envelope_correlation": 0.9944403151283694, "1000–3000 Hz_10ms_modulation_mae_db": 0.965689318388463, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5401596831660004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04525270317271915, "1000–3000 Hz_40ms_envelope_correlation": 0.9984598852570972, "1000–3000 Hz_40ms_modulation_mae_db": 1.1105284597625464, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9419840481549936, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.126095979416701, "3000–6000 Hz_10ms_envelope_correlation": 0.9891259167193047, "3000–6000 Hz_10ms_modulation_mae_db": 0.6933895855717349, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6995776057596255, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08020703251913437, "3000–6000 Hz_40ms_envelope_correlation": 0.9964289617964123, "3000–6000 Hz_40ms_modulation_mae_db": 1.1046463898083452, "mr_stft_db": 4.643654651173159, "band_mae_db": 0.10008816143950394, "rms_gain_error_db": -0.03746265194043583, "output_peak": 0.5171731322616128}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6089270801151087, "全帯域_10ms_rms_envelope_nrmse": 0.05205072854877552, "全帯域_10ms_envelope_correlation": 0.9975588890886029, "全帯域_10ms_modulation_mae_db": 0.6707557245996396, "全帯域_40ms_log_rms_mae_db": 0.44604003475640003, "全帯域_40ms_rms_envelope_nrmse": 0.025636048014989456, "全帯域_40ms_envelope_correlation": 0.9993869037120547, "全帯域_40ms_modulation_mae_db": 1.0653532635255305, "300–1000 Hz_10ms_log_rms_mae_db": 0.850524397416402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06724770204726051, "300–1000 Hz_10ms_envelope_correlation": 0.996432174648079, "300–1000 Hz_10ms_modulation_mae_db": 0.1290573154473695, "300–1000 Hz_40ms_log_rms_mae_db": 0.6090205092341318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04101096204688534, "300–1000 Hz_40ms_envelope_correlation": 0.9986277110363647, "300–1000 Hz_40ms_modulation_mae_db": 0.3991833168386256, "1000–3000 Hz_10ms_log_rms_mae_db": 0.785555210172366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09581982207090274, "1000–3000 Hz_10ms_envelope_correlation": 0.9931968874601775, "1000–3000 Hz_10ms_modulation_mae_db": 1.4658375186084693, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5626864235804826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05493663612668297, "1000–3000 Hz_40ms_envelope_correlation": 0.9977555710920236, "1000–3000 Hz_40ms_modulation_mae_db": 1.9453182381985465, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9313866720617209, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14486409926976823, "3000–6000 Hz_10ms_envelope_correlation": 0.9846013257423315, "3000–6000 Hz_10ms_modulation_mae_db": 1.236086297058996, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6312554411803882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07395364426426652, "3000–6000 Hz_40ms_envelope_correlation": 0.996575161542946, "3000–6000 Hz_40ms_modulation_mae_db": 1.4220415421466512, "mr_stft_db": 4.667016074409005, "band_mae_db": 0.16836347353491363, "rms_gain_error_db": -0.030854429355916486, "output_peak": 0.5025064268154957}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5875034390400098, "全帯域_10ms_rms_envelope_nrmse": 0.06761024286625758, "全帯域_10ms_envelope_correlation": 0.9959347335587672, "全帯域_10ms_modulation_mae_db": 1.9412082583707366, "全帯域_40ms_log_rms_mae_db": 0.4028234628388091, "全帯域_40ms_rms_envelope_nrmse": 0.04106045480887391, "全帯域_40ms_envelope_correlation": 0.998489037994853, "全帯域_40ms_modulation_mae_db": 2.4942830014816746, "300–1000 Hz_10ms_log_rms_mae_db": 0.8336963210061177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0636930695678145, "300–1000 Hz_10ms_envelope_correlation": 0.9967626571864795, "300–1000 Hz_10ms_modulation_mae_db": 1.115204448455514, "300–1000 Hz_40ms_log_rms_mae_db": 0.558791371155214, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03524942117462753, "300–1000 Hz_40ms_envelope_correlation": 0.9989754325080556, "300–1000 Hz_40ms_modulation_mae_db": 1.6860257011469564, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7682942265742031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1245820416843486, "1000–3000 Hz_10ms_envelope_correlation": 0.9886077946081199, "1000–3000 Hz_10ms_modulation_mae_db": 2.713768737759288, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5794532454275975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08122817013061377, "1000–3000 Hz_40ms_envelope_correlation": 0.9952299182675686, "1000–3000 Hz_40ms_modulation_mae_db": 3.25099982042658, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9070238891107625, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12026821355530039, "3000–6000 Hz_10ms_envelope_correlation": 0.9892330113017248, "3000–6000 Hz_10ms_modulation_mae_db": 0.9170580011636896, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6377611741684873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06117967883741709, "3000–6000 Hz_40ms_envelope_correlation": 0.9973611551995228, "3000–6000 Hz_40ms_modulation_mae_db": 1.164351632873786, "mr_stft_db": 4.657007046971544, "band_mae_db": 0.1478173434831108, "rms_gain_error_db": -0.09115340446503567, "output_peak": 0.5189518568166795}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5770835729781393, "全帯域_10ms_rms_envelope_nrmse": 0.05599558832011591, "全帯域_10ms_envelope_correlation": 0.9971710473710713, "全帯域_10ms_modulation_mae_db": 0.13277137289460225, "全帯域_40ms_log_rms_mae_db": 0.38565673432113706, "全帯域_40ms_rms_envelope_nrmse": 0.03142033158337121, "全帯域_40ms_envelope_correlation": 0.9990650281517474, "全帯域_40ms_modulation_mae_db": 0.27751776923934834, "300–1000 Hz_10ms_log_rms_mae_db": 0.8142856736355684, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07121114488117142, "300–1000 Hz_10ms_envelope_correlation": 0.995943245939061, "300–1000 Hz_10ms_modulation_mae_db": 0.4846218974326411, "300–1000 Hz_40ms_log_rms_mae_db": 0.609961414649715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04062019566259049, "300–1000 Hz_40ms_envelope_correlation": 0.9986211516984166, "300–1000 Hz_40ms_modulation_mae_db": 0.9110957925317745, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7111061325049557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0987618380817373, "1000–3000 Hz_10ms_envelope_correlation": 0.9927276915055563, "1000–3000 Hz_10ms_modulation_mae_db": 0.23526369367123615, "1000–3000 Hz_40ms_log_rms_mae_db": 0.499363652241361, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057834773211527656, "1000–3000 Hz_40ms_envelope_correlation": 0.9973966750830335, "1000–3000 Hz_40ms_modulation_mae_db": 0.35264810575080013, "3000–6000 Hz_10ms_log_rms_mae_db": 0.911482657377219, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12027815463716686, "3000–6000 Hz_10ms_envelope_correlation": 0.9891723390766057, "3000–6000 Hz_10ms_modulation_mae_db": 0.50273104728005, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6544157817759944, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08045516654185901, "3000–6000 Hz_40ms_envelope_correlation": 0.9950249436461042, "3000–6000 Hz_40ms_modulation_mae_db": 0.6430449120600578, "mr_stft_db": 4.6486918924889835, "band_mae_db": 0.07948203582332407, "rms_gain_error_db": -0.0019547972511062333, "output_peak": 0.5490853519375334}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5860455815889115, "全帯域_10ms_rms_envelope_nrmse": 0.05474785626704355, "全帯域_10ms_envelope_correlation": 0.9973052180624915, "全帯域_10ms_modulation_mae_db": 0.20305485572353996, "全帯域_40ms_log_rms_mae_db": 0.4005125406171705, "全帯域_40ms_rms_envelope_nrmse": 0.03338165098549888, "全帯域_40ms_envelope_correlation": 0.9989617842631896, "全帯域_40ms_modulation_mae_db": 0.35642238939355103, "300–1000 Hz_10ms_log_rms_mae_db": 0.8540007430422785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06864901740088067, "300–1000 Hz_10ms_envelope_correlation": 0.9962202779693922, "300–1000 Hz_10ms_modulation_mae_db": 0.8535076852135605, "300–1000 Hz_40ms_log_rms_mae_db": 0.6447606394215797, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03726389217288894, "300–1000 Hz_40ms_envelope_correlation": 0.9988367605294248, "300–1000 Hz_40ms_modulation_mae_db": 1.1490924198033745, "1000–3000 Hz_10ms_log_rms_mae_db": 0.737513289906137, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10062473865505234, "1000–3000 Hz_10ms_envelope_correlation": 0.9924628167903362, "1000–3000 Hz_10ms_modulation_mae_db": 0.8921455872753355, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5030337255232928, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0637712067877255, "1000–3000 Hz_40ms_envelope_correlation": 0.99683372917889, "1000–3000 Hz_40ms_modulation_mae_db": 0.8519960257299012, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9543543261587181, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13194110970124207, "3000–6000 Hz_10ms_envelope_correlation": 0.9877167702647711, "3000–6000 Hz_10ms_modulation_mae_db": 1.0526355101656018, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6863289505210606, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07838211977758819, "3000–6000 Hz_40ms_envelope_correlation": 0.996413006253191, "3000–6000 Hz_40ms_modulation_mae_db": 1.356395891692306, "mr_stft_db": 4.649492447239548, "band_mae_db": 0.13956911246488976, "rms_gain_error_db": -0.043540134495584, "output_peak": 0.5418057644915204}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6120886145001011, "全帯域_10ms_rms_envelope_nrmse": 0.06133695543681744, "全帯域_10ms_envelope_correlation": 0.9966574511967689, "全帯域_10ms_modulation_mae_db": 1.3509621268580139, "全帯域_40ms_log_rms_mae_db": 0.44255565050172546, "全帯域_40ms_rms_envelope_nrmse": 0.03698984114705743, "全帯域_40ms_envelope_correlation": 0.9987739618761996, "全帯域_40ms_modulation_mae_db": 1.7467376835218775, "300–1000 Hz_10ms_log_rms_mae_db": 0.8579143327962024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0703131697836317, "300–1000 Hz_10ms_envelope_correlation": 0.9960436012127001, "300–1000 Hz_10ms_modulation_mae_db": 0.7362688372514553, "300–1000 Hz_40ms_log_rms_mae_db": 0.6469027546859827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041553740759688575, "300–1000 Hz_40ms_envelope_correlation": 0.9985623905760245, "300–1000 Hz_40ms_modulation_mae_db": 1.107976177625371, "1000–3000 Hz_10ms_log_rms_mae_db": 0.737110804516649, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09415472306135055, "1000–3000 Hz_10ms_envelope_correlation": 0.993574552960806, "1000–3000 Hz_10ms_modulation_mae_db": 1.9626125985986842, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5285420679221488, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0637245274261676, "1000–3000 Hz_40ms_envelope_correlation": 0.9970983940360154, "1000–3000 Hz_40ms_modulation_mae_db": 2.3967627418728186, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8951218272842564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13306350879573114, "3000–6000 Hz_10ms_envelope_correlation": 0.9869019230836982, "3000–6000 Hz_10ms_modulation_mae_db": 1.4922585584001353, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6060016242544604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07008230258028235, "3000–6000 Hz_40ms_envelope_correlation": 0.9966763614600573, "3000–6000 Hz_40ms_modulation_mae_db": 1.7650636880480874, "mr_stft_db": 4.652127937488738, "band_mae_db": 0.1044146985155896, "rms_gain_error_db": -0.08105714961035779, "output_peak": 0.5155949778734775}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5962762939262612, "全帯域_10ms_rms_envelope_nrmse": 0.06082291344140017, "全帯域_10ms_envelope_correlation": 0.996662907913807, "全帯域_10ms_modulation_mae_db": 0.13112777771578565, "全帯域_40ms_log_rms_mae_db": 0.4186165631436991, "全帯域_40ms_rms_envelope_nrmse": 0.03157734653677256, "全帯域_40ms_envelope_correlation": 0.999064775971013, "全帯域_40ms_modulation_mae_db": 0.21234876390601198, "300–1000 Hz_10ms_log_rms_mae_db": 0.7776458214106711, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06676721417079547, "300–1000 Hz_10ms_envelope_correlation": 0.9965665162979022, "300–1000 Hz_10ms_modulation_mae_db": 0.24918757669599004, "300–1000 Hz_40ms_log_rms_mae_db": 0.5338616974377864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038843137222097555, "300–1000 Hz_40ms_envelope_correlation": 0.9988472809554183, "300–1000 Hz_40ms_modulation_mae_db": 0.16032435860230002, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7401846831846476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11002656997734428, "1000–3000 Hz_10ms_envelope_correlation": 0.9910537382170812, "1000–3000 Hz_10ms_modulation_mae_db": 0.22985371235171445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5337924896116024, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06499071184804875, "1000–3000 Hz_40ms_envelope_correlation": 0.9969043718373771, "1000–3000 Hz_40ms_modulation_mae_db": 0.2720513444440534, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9322367593249234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12248068238315106, "3000–6000 Hz_10ms_envelope_correlation": 0.9894447909839028, "3000–6000 Hz_10ms_modulation_mae_db": 1.0861710039205767, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6513323438455534, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06900131038886691, "3000–6000 Hz_40ms_envelope_correlation": 0.9973574947873007, "3000–6000 Hz_40ms_modulation_mae_db": 1.655864312231035, "mr_stft_db": 4.667611602234226, "band_mae_db": 0.18019844038905405, "rms_gain_error_db": -0.03335722376501267, "output_peak": 0.5139923513596478}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64720, "output_samples": 64720, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.8570069202122115, "spectral_convergence": 0.2375693597508049, "spectral_convergence_status": "ok", "amplitude_floor": 1.1807658352589429e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.4652931576810255, "spectral_convergence": 0.2299894675487755, "spectral_convergence_status": "ok", "amplitude_floor": 7.24953185483845e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.667789484050786, "spectral_convergence": 0.2351329421015957, "spectral_convergence_status": "ok", "amplitude_floor": 4.6531510695198173e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031402968364977683, "output_power": 0.0003229072201615356, "delta_db": 0.12107054200242517, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005546801738700932, "output_power": 0.0005458967192469998, "delta_db": -0.06932158742160915, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010899709362407707, "output_power": 0.0010743992974415434, "delta_db": -0.062492020864268485, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008204698714335104, "output_power": 0.0007920150456224078, "delta_db": -0.15329206102838003, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0004732551323003661, "output_power": 0.00045598172654938516, "delta_db": -0.16147893767894056, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00010423546715106559, "output_power": 9.587828993080039e-05, "delta_db": -0.36295237686095705, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1551012543094301, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09094880042144013, "pitch": null, "output_rms_dbfs": -24.831899508018317, "output_peak": 0.502015764418237, "reference_rms_dbfs": -24.740950707596877, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.663363187314674}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64720, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00017261805365043103, "log_rms_mae_db": 0.6152536547466612, "rms_envelope_nrmse": 0.06733245536413261, "envelope_correlation": 0.9959659264377441, "rms_gain_error_db": -0.09094880783201571, "reference_mean_power": 0.003360795516302892, "output_mean_power": 0.003291146452366798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.9828935159721717e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046042689591210996, "output_power": 0.00045105321338200573, "reference_active": true, "delta_db": -0.08932903924560488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011507848708673394, "output_power": 9.301537007345665e-05, "reference_active": true, "delta_db": -0.9243942537355142}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.734946402574847e-05, "output_power": 1.630394050923888e-05, "reference_active": true, "delta_db": -3.599917919741018}], "active_band_count": 3, "active_band_mae_db": 1.5378804042407124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00013379439656886598, "log_rms_mae_db": 0.4388204989403843, "rms_envelope_nrmse": 0.03629087326045861, "envelope_correlation": 0.9988419048978696, "rms_gain_error_db": -0.0909488156374072, "reference_mean_power": 0.0033859385941668236, "output_mean_power": 0.0033157684595921987}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.884361456130501e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004231625539622664, "output_power": 0.00041137643945616045, "reference_active": true, "delta_db": -0.12267814661044588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.583961694519604e-05, "output_power": 4.279871179951631e-05, "reference_active": true, "delta_db": -1.1551173304441156}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3554104634271355e-06, "output_power": 1.0427009264365785e-06, "reference_active": true, "delta_db": -5.075858950790217}], "active_band_count": 3, "active_band_mae_db": 2.117884809281593}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00011786303332041996, "log_rms_mae_db": 0.7996913922635195, "rms_envelope_nrmse": 0.07000904847149247, "envelope_correlation": 0.9961128524041005, "rms_gain_error_db": -0.08767985620906435, "reference_mean_power": 0.0013229704572187174, "output_mean_power": 0.0012965287771170794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 9.694412540428036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002357826301518165, "output_power": 0.00022849121230332216, "reference_active": true, "delta_db": -0.13642306034068058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.91091063998743e-05, "output_power": 5.2673254072237325e-05, "reference_active": true, "delta_db": -0.500642445896686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.351021856959089e-05, "output_power": 9.5835276899161e-06, "reference_active": true, "delta_db": -1.491369731288923}], "active_band_count": 3, "active_band_mae_db": 0.7094784125087633}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010411406065660645, "log_rms_mae_db": 0.575635192181237, "rms_envelope_nrmse": 0.04149855555494797, "envelope_correlation": 0.9986186199080418, "rms_gain_error_db": -0.08767986194789788, "reference_mean_power": 0.0013328679887454193, "output_mean_power": 0.0013062284889233658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 9.215200160073525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002182221456522984, "output_power": 0.0002106596292662306, "reference_active": true, "delta_db": -0.15317506078702134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.729502442376393e-05, "output_power": 2.4578182362775303e-05, "reference_active": true, "delta_db": -0.45533725015504617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1926400570958608e-06, "output_power": 7.152882238571367e-07, "reference_active": true, "delta_db": -2.2202831676594106}], "active_band_count": 3, "active_band_mae_db": 0.9429318262004927}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00014037287410750346, "log_rms_mae_db": 0.7052305324122597, "rms_envelope_nrmse": 0.11061395801201931, "envelope_correlation": 0.9909144240924868, "rms_gain_error_db": -0.13567557682942782, "reference_mean_power": 0.001043588584426717, "output_mean_power": 0.0010114903926948628}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 7.131098310180307e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021758763787569653, "output_power": 0.0002119365674802314, "reference_active": true, "delta_db": -0.11428321287304964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8027856042619025e-05, "output_power": 3.964994862072366e-05, "reference_active": true, "delta_db": -1.653938960807442}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3022954991601597e-05, "output_power": 1.0544329572290993e-05, "reference_active": true, "delta_db": -3.3914209276199565}], "active_band_count": 3, "active_band_mae_db": 1.719881033766816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010015064346906524, "log_rms_mae_db": 0.48605544430882586, "rms_envelope_nrmse": 0.06110223087752498, "envelope_correlation": 0.9972355438858459, "rms_gain_error_db": -0.1356755807075014, "reference_mean_power": 0.0010513959774729573, "output_mean_power": 0.0010190576488209552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 6.799409328431789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020213626575120905, "output_power": 0.000193786227955183, "reference_active": true, "delta_db": -0.18321329048270202}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0065691390851235e-05, "output_power": 1.8183174994451793e-05, "reference_active": true, "delta_db": -2.184014769455756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1255469357025214e-06, "output_power": 6.215030026894041e-07, "reference_active": true, "delta_db": -5.3402746816709215}], "active_band_count": 3, "active_band_mae_db": 2.56916758053646}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 5.7697319934351994e-05, "log_rms_mae_db": 0.9393513487334597, "rms_envelope_nrmse": 0.13712501777938874, "envelope_correlation": 0.9866511029090056, "rms_gain_error_db": -0.38096714845928425, "reference_mean_power": 0.0001838054244420426, "output_mean_power": 0.0001683687971458572}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.2570064040799192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.4129018402509514e-05, "output_power": 4.034355776603535e-05, "reference_active": true, "delta_db": -0.389500726193135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0833298446779821e-05, "output_power": 8.484609265054436e-06, "reference_active": true, "delta_db": -1.0612886080293296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.606324451113381e-06, "output_power": 2.1736289920794262e-06, "reference_active": true, "delta_db": -3.2616910656447966}], "active_band_count": 3, "active_band_mae_db": 1.5708267999557537}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 3.902816175045304e-05, "log_rms_mae_db": 0.6851530594751848, "rms_envelope_nrmse": 0.08794296687784318, "envelope_correlation": 0.9951386523089536, "rms_gain_error_db": -0.38096715107403806, "reference_mean_power": 0.00018518052639832005, "output_mean_power": 0.00016962841309170031}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.1869756093173816e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2060123324004214e-05, "output_power": 3.7957202808563986e-05, "reference_active": true, "delta_db": -0.44576341399994396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.372206568913873e-06, "output_power": 3.747209557853796e-06, "reference_active": true, "delta_db": -1.564447224218288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0554113039123233e-07, "output_power": 1.5914583165697623e-07, "reference_active": true, "delta_db": -4.062396392886808}], "active_band_count": 3, "active_band_mae_db": 2.02420234370168}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9512042694623771, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d95848ecb9a2aac772227f86502719b5f9478b185c27ae1f979d10f8a3c152a4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9921222775953666, "standard_deviation": 0.0018544139699644352, "minimum": 0.9886077946081199, "maximum": 0.9944403151283694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.7407729788532569, "standard_deviation": 0.026540564121700288, "minimum": 0.7052305324122597, "maximum": 0.785555210172366, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2731315250525008, "standard_deviation": 0.861295509103669, "minimum": 0.22985371235171445, "maximum": 2.713768737759288, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10263744544581672, "standard_deviation": 0.011935554546550355, "minimum": 0.08651587202377874, "maximum": 0.1245820416843486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971142610797313, "standard_deviation": 0.0009250638113255782, "minimum": 0.9952299182675686, "maximum": 0.9984598852570972, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.5291358414726639, "standard_deviation": 0.03212624439744221, "minimum": 0.48605544430882586, "maximum": 0.5794532454275975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5936840395902134, "standard_deviation": 1.104359500340081, "minimum": 0.2720513444440534, "maximum": 3.25099982042658, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0616051199476263, "standard_deviation": 0.010226357919170604, "minimum": 0.04525270317271915, "maximum": 0.08122817013061377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9878558975101681, "standard_deviation": 0.0017213027904531648, "minimum": 0.9846013257423315, "maximum": 0.9894447909839028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 0.9266176910257566, "standard_deviation": 0.020095787607857598, "minimum": 0.8951218272842564, "maximum": 0.9543543261587181, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0688946004395672, "standard_deviation": 0.36728509894712913, "minimum": 0.50273104728005, "maximum": 1.5708267999557537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1295145956923062, "standard_deviation": 0.008815399176914736, "minimum": 0.12026821355530039, "maximum": 0.14486409926976823, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.996371967124311, "standard_deviation": 0.0008802736582146204, "minimum": 0.9950249436461042, "maximum": 0.9973611551995228, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6564782476225943, "standard_deviation": 0.031928505784340984, "minimum": 0.6060016242544604, "maximum": 0.6995776057596255, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3919513390702436, "standard_deviation": 0.43193591157013705, "minimum": 0.6430449120600578, "maximum": 2.02420234370168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0751505277234072, "standard_deviation": 0.00834877278412633, "minimum": 0.06117967883741709, "maximum": 0.08794296687784318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9964458728088693, "standard_deviation": 0.0005032036562477007, "minimum": 0.995943245939061, "maximum": 0.9974856568132402, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.8237889835636649, "standard_deviation": 0.029619937392503064, "minimum": 0.7776458214106711, "maximum": 0.8579143327962024, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.575294350121988, "standard_deviation": 0.3357616642715208, "minimum": 0.1290573154473695, "maximum": 1.115204448455514, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06674218397094213, "standard_deviation": 0.004942092487034135, "minimum": 0.05604710544449034, "maximum": 0.07121114488117142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998815755932527, "standard_deviation": 0.0002900226048742075, "minimum": 0.9985623905760245, "maximum": 0.9994367002484705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.599802301878253, "standard_deviation": 0.04040756434146459, "minimum": 0.5338616974377864, "maximum": 0.6469027546859827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.863229764061542, "standard_deviation": 0.4833335805280168, "minimum": 0.16032435860230002, "maximum": 1.6860257011469564, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.037751254904240264, "standard_deviation": 0.005262283164082208, "minimum": 0.025970134640195743, "maximum": 0.041553740759688575, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13437931499497702, "standard_deviation": 0.03581440526878295, "minimum": 0.07948203582332407, "maximum": 0.18019844038905405, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.656120604914985, "standard_deviation": 0.009058971027743633, "minimum": 4.643654651173159, "maximum": 4.667611602234226, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5201407032467755, "standard_deviation": 0.01696776087107848, "minimum": 0.502015764418237, "maximum": 0.5490853519375334, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0512910739131111, "standard_deviation": 0.03267242145061124, "minimum": -0.09115340446503567, "maximum": -0.0019547972511062333, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968624712937815, "standard_deviation": 0.0006689808177189092, "minimum": 0.9959347335587672, "maximum": 0.997643596720999, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5973654692060545, "standard_deviation": 0.013685207833367607, "minimum": 0.5770835729781393, "maximum": 0.6152536547466612, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8290680825144057, "standard_deviation": 0.699748773820612, "minimum": 0.13112777771578565, "maximum": 1.9412082583707366, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.058896577417506604, "standard_deviation": 0.006406765756527345, "minimum": 0.05127587909551006, "maximum": 0.06761024286625758, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989968735321157, "standard_deviation": 0.0003047281213581799, "minimum": 0.998489037994853, "maximum": 0.99939159139, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42152584726577635, "standard_deviation": 0.022909110500496373, "minimum": 0.38565673432113706, "maximum": 0.44604003475640003, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1419100724617126, "standard_deviation": 0.8830299279780955, "minimum": 0.21234876390601198, "maximum": 2.4942830014816746, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.032769586037401335, "standard_deviation": 0.0053722246446435305, "minimum": 0.025636048014989456, "maximum": 0.04106045480887391, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-121726-0012", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.6418535709381104, "predicted_mos": 2.6418535709381104, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "01af08c18613edd659fa076f5ff570a6395440f765dcdf64c8d689fa5a55fc2d", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8757416997104883, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6175033224835521, "全帯域_10ms_rms_envelope_nrmse": 0.065960710978044, "全帯域_10ms_envelope_correlation": 0.9961357491695041, "全帯域_10ms_modulation_mae_db": 1.353835997043613, "全帯域_40ms_log_rms_mae_db": 0.437527669677432, "全帯域_40ms_rms_envelope_nrmse": 0.03477259064275162, "全帯域_40ms_envelope_correlation": 0.9989493877764276, "全帯域_40ms_modulation_mae_db": 1.8752306140758446, "300–1000 Hz_10ms_log_rms_mae_db": 0.8131854267426019, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07486142022280226, "300–1000 Hz_10ms_envelope_correlation": 0.9955673395857613, "300–1000 Hz_10ms_modulation_mae_db": 0.8320565059283017, "300–1000 Hz_40ms_log_rms_mae_db": 0.5660368107577419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04380401523093498, "300–1000 Hz_40ms_envelope_correlation": 0.9984834877104226, "300–1000 Hz_40ms_modulation_mae_db": 1.183195172829227, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7051618137448336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10536489832323237, "1000–3000 Hz_10ms_envelope_correlation": 0.9917380590228289, "1000–3000 Hz_10ms_modulation_mae_db": 1.5092713830410698, "1000–3000 Hz_40ms_log_rms_mae_db": 0.48827610746585004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055987743174837214, "1000–3000 Hz_40ms_envelope_correlation": 0.9976504323014889, "1000–3000 Hz_40ms_modulation_mae_db": 2.274678405120606, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9365741893783669, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.132101000047681, "3000–6000 Hz_10ms_envelope_correlation": 0.9876391159943705, "3000–6000 Hz_10ms_modulation_mae_db": 1.0408516407951593, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6691861398589577, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08406397739471476, "3000–6000 Hz_40ms_envelope_correlation": 0.9955948675007786, "3000–6000 Hz_40ms_modulation_mae_db": 1.3145386280435025, "mr_stft_db": 4.607721813006239, "band_mae_db": 0.1473191705556661, "rms_gain_error_db": -0.09280728220740983, "output_peak": 0.5042626313720509}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5931542285219438, "全帯域_10ms_rms_envelope_nrmse": 0.050553052343714544, "全帯域_10ms_envelope_correlation": 0.997703581071154, "全帯域_10ms_modulation_mae_db": 0.5449310658402461, "全帯域_40ms_log_rms_mae_db": 0.4420338593708832, "全帯域_40ms_rms_envelope_nrmse": 0.024455649401486597, "全帯域_40ms_envelope_correlation": 0.9994478555986414, "全帯域_40ms_modulation_mae_db": 0.7310961913104282, "300–1000 Hz_10ms_log_rms_mae_db": 0.7907654821349933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05859636925769668, "300–1000 Hz_10ms_envelope_correlation": 0.9972533460298478, "300–1000 Hz_10ms_modulation_mae_db": 0.2506225567574993, "300–1000 Hz_40ms_log_rms_mae_db": 0.6080319903925079, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.026901108389772117, "300–1000 Hz_40ms_envelope_correlation": 0.9993944477680725, "300–1000 Hz_40ms_modulation_mae_db": 0.4784509085067541, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7327535077324197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08711780206721229, "1000–3000 Hz_10ms_envelope_correlation": 0.9943432590387368, "1000–3000 Hz_10ms_modulation_mae_db": 0.7307712380811803, "1000–3000 Hz_40ms_log_rms_mae_db": 0.50251443154811, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04181389858288084, "1000–3000 Hz_40ms_envelope_correlation": 0.9986676310483182, "1000–3000 Hz_40ms_modulation_mae_db": 0.8586118144687057, "3000–6000 Hz_10ms_log_rms_mae_db": 0.957407122448203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12309670701270706, "3000–6000 Hz_10ms_envelope_correlation": 0.9899358797235368, "3000–6000 Hz_10ms_modulation_mae_db": 0.6611581743207035, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6832770437892258, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07967056696459696, "3000–6000 Hz_40ms_envelope_correlation": 0.9967523602722144, "3000–6000 Hz_40ms_modulation_mae_db": 1.1240699420602096, "mr_stft_db": 4.587961870954234, "band_mae_db": 0.10746927688537501, "rms_gain_error_db": -0.03081082813206179, "output_peak": 0.5286422447847252}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6163356872905592, "全帯域_10ms_rms_envelope_nrmse": 0.05341589146272673, "全帯域_10ms_envelope_correlation": 0.9974362539948554, "全帯域_10ms_modulation_mae_db": 0.9011794676953294, "全帯域_40ms_log_rms_mae_db": 0.45049013921915093, "全帯域_40ms_rms_envelope_nrmse": 0.026138760474400047, "全帯域_40ms_envelope_correlation": 0.9993734765383275, "全帯域_40ms_modulation_mae_db": 1.3433182546759204, "300–1000 Hz_10ms_log_rms_mae_db": 0.8473332598200017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.069838323459262, "300–1000 Hz_10ms_envelope_correlation": 0.9961518655557462, "300–1000 Hz_10ms_modulation_mae_db": 0.2528249681780497, "300–1000 Hz_40ms_log_rms_mae_db": 0.6157170683590267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04318235147497819, "300–1000 Hz_40ms_envelope_correlation": 0.9984762491978525, "300–1000 Hz_40ms_modulation_mae_db": 0.5437251104389522, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7744889400665288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10070565062398458, "1000–3000 Hz_10ms_envelope_correlation": 0.9925107978787995, "1000–3000 Hz_10ms_modulation_mae_db": 1.9589769424761794, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5493440241642215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057203586446538175, "1000–3000 Hz_40ms_envelope_correlation": 0.9976122417411064, "1000–3000 Hz_40ms_modulation_mae_db": 2.5330931889794197, "3000–6000 Hz_10ms_log_rms_mae_db": 0.914186723219898, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13484576617212257, "3000–6000 Hz_10ms_envelope_correlation": 0.9870969158064176, "3000–6000 Hz_10ms_modulation_mae_db": 1.3397502443129026, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6113822891022374, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07324912495268532, "3000–6000 Hz_40ms_envelope_correlation": 0.9970395931460209, "3000–6000 Hz_40ms_modulation_mae_db": 1.451579835580147, "mr_stft_db": 4.602381105895696, "band_mae_db": 0.18596656590634364, "rms_gain_error_db": -0.04399094364177161, "output_peak": 0.5047755845010017}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5740687307606003, "全帯域_10ms_rms_envelope_nrmse": 0.06631570465317445, "全帯域_10ms_envelope_correlation": 0.9960753122970846, "全帯域_10ms_modulation_mae_db": 1.8686590615616245, "全帯域_40ms_log_rms_mae_db": 0.4001124421594973, "全帯域_40ms_rms_envelope_nrmse": 0.03832126228640086, "全帯域_40ms_envelope_correlation": 0.99867654874196, "全帯域_40ms_modulation_mae_db": 2.4023985480526715, "300–1000 Hz_10ms_log_rms_mae_db": 0.810341113666467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06564087475540853, "300–1000 Hz_10ms_envelope_correlation": 0.9965513326324996, "300–1000 Hz_10ms_modulation_mae_db": 1.1096315075525345, "300–1000 Hz_40ms_log_rms_mae_db": 0.5436546854820102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03590721038616179, "300–1000 Hz_40ms_envelope_correlation": 0.9989234675206232, "300–1000 Hz_40ms_modulation_mae_db": 1.7385116761596835, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7571101662927623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11862232356922858, "1000–3000 Hz_10ms_envelope_correlation": 0.9896200387159705, "1000–3000 Hz_10ms_modulation_mae_db": 2.4914717362468553, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5359970216765071, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07084310971599936, "1000–3000 Hz_40ms_envelope_correlation": 0.9963628300990974, "1000–3000 Hz_40ms_modulation_mae_db": 3.0553959874681964, "3000–6000 Hz_10ms_log_rms_mae_db": 0.916119324856397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11614764355828373, "3000–6000 Hz_10ms_envelope_correlation": 0.9900451312092403, "3000–6000 Hz_10ms_modulation_mae_db": 0.5863801836155823, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6238258920966415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05912232599491685, "3000–6000 Hz_40ms_envelope_correlation": 0.9976310164112304, "3000–6000 Hz_40ms_modulation_mae_db": 0.8093011603719775, "mr_stft_db": 4.594344666728006, "band_mae_db": 0.1403391002399525, "rms_gain_error_db": -0.08078413011610089, "output_peak": 0.5178442999054319}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5791020243214559, "全帯域_10ms_rms_envelope_nrmse": 0.058611587552757714, "全帯域_10ms_envelope_correlation": 0.9969019497586762, "全帯域_10ms_modulation_mae_db": 0.23091354079916412, "全帯域_40ms_log_rms_mae_db": 0.3763004877052665, "全帯域_40ms_rms_envelope_nrmse": 0.034014642475857795, "全帯域_40ms_envelope_correlation": 0.9989035774993319, "全帯域_40ms_modulation_mae_db": 0.24702290989592876, "300–1000 Hz_10ms_log_rms_mae_db": 0.8226100074699559, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0761383156603259, "300–1000 Hz_10ms_envelope_correlation": 0.9953580814168822, "300–1000 Hz_10ms_modulation_mae_db": 0.6254997549359399, "300–1000 Hz_40ms_log_rms_mae_db": 0.6162448931949633, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04519104424189813, "300–1000 Hz_40ms_envelope_correlation": 0.9982896055365841, "300–1000 Hz_40ms_modulation_mae_db": 1.1889942915702283, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7340493649998364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10055159893784, "1000–3000 Hz_10ms_envelope_correlation": 0.9925083921170178, "1000–3000 Hz_10ms_modulation_mae_db": 0.4609973351023651, "1000–3000 Hz_40ms_log_rms_mae_db": 0.49537588320019466, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06194859147757557, "1000–3000 Hz_40ms_envelope_correlation": 0.9970333253988445, "1000–3000 Hz_40ms_modulation_mae_db": 0.34335588525763666, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8965811902993938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12178697457953905, "3000–6000 Hz_10ms_envelope_correlation": 0.9889692563029384, "3000–6000 Hz_10ms_modulation_mae_db": 0.750626797177282, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6357891801852138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08193914492563684, "3000–6000 Hz_40ms_envelope_correlation": 0.9949084862221718, "3000–6000 Hz_40ms_modulation_mae_db": 0.53756839773523, "mr_stft_db": 4.593663014745279, "band_mae_db": 0.09267741011417939, "rms_gain_error_db": 0.0028789202822430628, "output_peak": 0.5495502491400005}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5945498058372971, "全帯域_10ms_rms_envelope_nrmse": 0.05331261732729459, "全帯域_10ms_envelope_correlation": 0.9974667707810526, "全帯域_10ms_modulation_mae_db": 0.2027499019090372, "全帯域_40ms_log_rms_mae_db": 0.3888051417433148, "全帯域_40ms_rms_envelope_nrmse": 0.030991993296110024, "全帯域_40ms_envelope_correlation": 0.9991345345261707, "全帯域_40ms_modulation_mae_db": 0.3824913161979771, "300–1000 Hz_10ms_log_rms_mae_db": 0.8617538607525491, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06906676745835694, "300–1000 Hz_10ms_envelope_correlation": 0.9961753064878767, "300–1000 Hz_10ms_modulation_mae_db": 0.9492222404512688, "300–1000 Hz_40ms_log_rms_mae_db": 0.6546047121258826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034448000740738796, "300–1000 Hz_40ms_envelope_correlation": 0.9990032610246233, "300–1000 Hz_40ms_modulation_mae_db": 1.3336690741380821, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7305697299961469, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09431235209893077, "1000–3000 Hz_10ms_envelope_correlation": 0.9933930313345827, "1000–3000 Hz_10ms_modulation_mae_db": 0.6436218860691095, "1000–3000 Hz_40ms_log_rms_mae_db": 0.4912539056810172, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05594365740411561, "1000–3000 Hz_40ms_envelope_correlation": 0.9976215753000328, "1000–3000 Hz_40ms_modulation_mae_db": 0.4805627779351487, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9444363339778502, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13142941539362385, "3000–6000 Hz_10ms_envelope_correlation": 0.9878429729947467, "3000–6000 Hz_10ms_modulation_mae_db": 0.923687412378416, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6852860064417311, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07930758928114696, "3000–6000 Hz_40ms_envelope_correlation": 0.9963049479080883, "3000–6000 Hz_40ms_modulation_mae_db": 1.1964542839987093, "mr_stft_db": 4.585354325651563, "band_mae_db": 0.15443518749503818, "rms_gain_error_db": -0.06405758074391697, "output_peak": 0.5356711511982699}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6268560983438324, "全帯域_10ms_rms_envelope_nrmse": 0.06060877836112864, "全帯域_10ms_envelope_correlation": 0.9967216637493638, "全帯域_10ms_modulation_mae_db": 1.1666430789588313, "全帯域_40ms_log_rms_mae_db": 0.45241951531055663, "全帯域_40ms_rms_envelope_nrmse": 0.034001924665833266, "全帯域_40ms_envelope_correlation": 0.9989583276586865, "全帯域_40ms_modulation_mae_db": 1.4441481363746813, "300–1000 Hz_10ms_log_rms_mae_db": 0.8417477287920903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07083424531741665, "300–1000 Hz_10ms_envelope_correlation": 0.9959812467244231, "300–1000 Hz_10ms_modulation_mae_db": 0.6348193392945326, "300–1000 Hz_40ms_log_rms_mae_db": 0.636415261399386, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03953967625752539, "300–1000 Hz_40ms_envelope_correlation": 0.9986992476845946, "300–1000 Hz_40ms_modulation_mae_db": 1.0173563129249292, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7305766950047726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09312863850637437, "1000–3000 Hz_10ms_envelope_correlation": 0.9936345154598293, "1000–3000 Hz_10ms_modulation_mae_db": 1.6944647176597059, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5011505370765955, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0581967484764646, "1000–3000 Hz_40ms_envelope_correlation": 0.9975318713724038, "1000–3000 Hz_40ms_modulation_mae_db": 2.0315730276130854, "3000–6000 Hz_10ms_log_rms_mae_db": 0.883543986582329, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1313732628249342, "3000–6000 Hz_10ms_envelope_correlation": 0.987317314597087, "3000–6000 Hz_10ms_modulation_mae_db": 0.9221027665209682, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6135326335315469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07241636654329142, "3000–6000 Hz_40ms_envelope_correlation": 0.9964741192956325, "3000–6000 Hz_40ms_modulation_mae_db": 0.9958642725346674, "mr_stft_db": 4.591924277801947, "band_mae_db": 0.09920520069463477, "rms_gain_error_db": -0.07092724035736087, "output_peak": 0.5063346317548599}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5981936961916121, "全帯域_10ms_rms_envelope_nrmse": 0.058812013074217684, "全帯域_10ms_envelope_correlation": 0.9968826987493279, "全帯域_10ms_modulation_mae_db": 0.37563627888857015, "全帯域_40ms_log_rms_mae_db": 0.4088863792928368, "全帯域_40ms_rms_envelope_nrmse": 0.02964137630340619, "全帯域_40ms_envelope_correlation": 0.9991807100171348, "全帯域_40ms_modulation_mae_db": 0.5146052111824758, "300–1000 Hz_10ms_log_rms_mae_db": 0.8180064210815797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07423460379209036, "300–1000 Hz_10ms_envelope_correlation": 0.9957785433926056, "300–1000 Hz_10ms_modulation_mae_db": 0.24829672091114752, "300–1000 Hz_40ms_log_rms_mae_db": 0.582224335597626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04360246908054208, "300–1000 Hz_40ms_envelope_correlation": 0.9985633312038258, "300–1000 Hz_40ms_modulation_mae_db": 0.11813270295338914, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7486379653959289, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10484331275090862, "1000–3000 Hz_10ms_envelope_correlation": 0.9919695029738289, "1000–3000 Hz_10ms_modulation_mae_db": 0.8060849179968481, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5223266454095399, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06035392423215655, "1000–3000 Hz_40ms_envelope_correlation": 0.997457645892024, "1000–3000 Hz_40ms_modulation_mae_db": 0.925909529514295, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9193859729156922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11807700917071672, "3000–6000 Hz_10ms_envelope_correlation": 0.9903357631295671, "3000–6000 Hz_10ms_modulation_mae_db": 1.1216054592652107, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6460364539530992, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06818707477067501, "3000–6000 Hz_40ms_envelope_correlation": 0.9975131239785527, "3000–6000 Hz_40ms_modulation_mae_db": 1.6771571380824184, "mr_stft_db": 4.608743335190209, "band_mae_db": 0.1931380928778874, "rms_gain_error_db": -0.03726147986080634, "output_peak": 0.5109439131848725}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64720, "output_samples": 64720, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.801959538625293, "spectral_convergence": 0.23815602846527711, "spectral_convergence_status": "ok", "amplitude_floor": 1.1807658352589429e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.415745758089505, "spectral_convergence": 0.23106359569431184, "spectral_convergence_status": "ok", "amplitude_floor": 7.24953185483845e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.605460142303919, "spectral_convergence": 0.23498270209719635, "spectral_convergence_status": "ok", "amplitude_floor": 4.6531510695198173e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031402968364977683, "output_power": 0.0003212519812361376, "delta_db": 0.09875113052251264, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005546801738700932, "output_power": 0.0005457584232876508, "delta_db": -0.07042195639338208, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010899709362407707, "output_power": 0.001069352838247137, "delta_db": -0.0829389129570822, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008204698714335104, "output_power": 0.0007962132713516458, "delta_db": -0.13033225738985446, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0004732551323003661, "output_power": 0.0004569033332828666, "delta_db": -0.15271006035965962, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00010423546715106559, "output_power": 9.619210896681492e-05, "delta_db": -0.34876070571150564, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1473191705556661, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09280728220740983, "pitch": null, "output_rms_dbfs": -24.833757989804287, "output_peak": 0.5042626313720509, "reference_rms_dbfs": -24.740950707596877, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.607721813006239}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64720, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00017261805365043103, "log_rms_mae_db": 0.6175033224835521, "rms_envelope_nrmse": 0.065960710978044, "envelope_correlation": 0.9961357491695041, "rms_gain_error_db": -0.09280728897255122, "reference_mean_power": 0.003360795516302892, "output_mean_power": 0.003289738369757768}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.9828935159721717e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046042689591210996, "output_power": 0.0004482758197697373, "reference_active": true, "delta_db": -0.11615371653720061}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011507848708673394, "output_power": 9.756042178620759e-05, "reference_active": true, "delta_db": -0.717204744298528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.734946402574847e-05, "output_power": 1.7761082064933667e-05, "reference_active": true, "delta_db": -3.2281495302951098}], "active_band_count": 3, "active_band_mae_db": 1.353835997043613}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00013379439656886598, "log_rms_mae_db": 0.437527669677432, "rms_envelope_nrmse": 0.03477259064275162, "envelope_correlation": 0.9989493877764276, "rms_gain_error_db": -0.09280729834806778, "reference_mean_power": 0.0033859385941668236, "output_mean_power": 0.0033143498415164857}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.884361456130501e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004231625539622664, "output_power": 0.0004089891952061302, "reference_active": true, "delta_db": -0.14795394772610027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.583961694519604e-05, "output_power": 4.518222429784049e-05, "reference_active": true, "delta_db": -0.919748231975177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3554104634271355e-06, "output_power": 1.1747533393932158e-06, "reference_active": true, "delta_db": -4.557989662526257}], "active_band_count": 3, "active_band_mae_db": 1.8752306140758446}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00011786303332041996, "log_rms_mae_db": 0.8131854267426019, "rms_envelope_nrmse": 0.07486142022280226, "envelope_correlation": 0.9955673395857613, "rms_gain_error_db": -0.10433142142666707, "reference_mean_power": 0.0013229704572187174, "output_mean_power": 0.0012915671902456762}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 9.694412540428036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002357826301518165, "output_power": 0.00022736062205806507, "reference_active": true, "delta_db": -0.157965591249781}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.91091063998743e-05, "output_power": 5.1483158285392634e-05, "reference_active": true, "delta_db": -0.5998921254270042}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.351021856959089e-05, "output_power": 9.053806220786184e-06, "reference_active": true, "delta_db": -1.73831180110812}], "active_band_count": 3, "active_band_mae_db": 0.8320565059283017}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010411406065660645, "log_rms_mae_db": 0.5660368107577419, "rms_envelope_nrmse": 0.04380401523093498, "envelope_correlation": 0.9984834877104226, "rms_gain_error_db": -0.1043314231187065, "reference_mean_power": 0.0013328679887454193, "output_mean_power": 0.0013012297841787875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 9.215200160073525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002182221456522984, "output_power": 0.0002092159496659779, "reference_active": true, "delta_db": -0.18304031546171523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.729502442376393e-05, "output_power": 2.4139610578096066e-05, "reference_active": true, "delta_db": -0.5335322750309308}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1926400570958608e-06, "output_power": 6.211666047283519e-07, "reference_active": true, "delta_db": -2.8330129279950347}], "active_band_count": 3, "active_band_mae_db": 1.183195172829227}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00014037287410750346, "log_rms_mae_db": 0.7051618137448336, "rms_envelope_nrmse": 0.10536489832323237, "envelope_correlation": 0.9917380590228289, "rms_gain_error_db": -0.10955812818706823, "reference_mean_power": 0.001043588584426717, "output_mean_power": 0.0010175915851856285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 7.131098310180307e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021758763787569653, "output_power": 0.00021303688514893846, "reference_active": true, "delta_db": -0.0917941394159959}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8027856042619025e-05, "output_power": 4.4714530479263157e-05, "reference_active": true, "delta_db": -1.1318785012263333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3022954991601597e-05, "output_power": 1.0758380831551691e-05, "reference_active": true, "delta_db": -3.3041415084808796}], "active_band_count": 3, "active_band_mae_db": 1.5092713830410698}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010015064346906524, "log_rms_mae_db": 0.48827610746585004, "rms_envelope_nrmse": 0.055987743174837214, "envelope_correlation": 0.9976504323014889, "rms_gain_error_db": -0.10955812922007592, "reference_mean_power": 0.0010513959774729573, "output_mean_power": 0.0010252044867928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 6.799409328431789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020213626575120905, "output_power": 0.00019511400845331287, "reference_active": true, "delta_db": -0.1535578709096192}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0065691390851235e-05, "output_power": 2.074230229089404e-05, "reference_active": true, "delta_db": -1.6121423611071641}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1255469357025214e-06, "output_power": 6.631888272469494e-07, "reference_active": true, "delta_db": -5.058334983345035}], "active_band_count": 3, "active_band_mae_db": 2.274678405120606}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 5.7697319934351994e-05, "log_rms_mae_db": 0.9365741893783669, "rms_envelope_nrmse": 0.132101000047681, "envelope_correlation": 0.9876391159943705, "rms_gain_error_db": -0.3716403894290493, "reference_mean_power": 0.0001838054244420426, "output_mean_power": 0.0001687307687276684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.2570064040799192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.4129018402509514e-05, "output_power": 4.0727595189929506e-05, "reference_active": true, "delta_db": -0.3483549979020942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0833298446779821e-05, "output_power": 9.00005317350732e-06, "reference_active": true, "delta_db": -0.8051563238335567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.606324451113381e-06, "output_power": 2.9271849569670907e-06, "reference_active": true, "delta_db": -1.969043600649827}], "active_band_count": 3, "active_band_mae_db": 1.0408516407951593}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 3.902816175045304e-05, "log_rms_mae_db": 0.6691861398589577, "rms_envelope_nrmse": 0.08406397739471476, "envelope_correlation": 0.9955948675007786, "rms_gain_error_db": -0.3716403884781557, "reference_mean_power": 0.00018518052639832005, "output_mean_power": 0.00016999309282777163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.1869756093173816e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2060123324004214e-05, "output_power": 3.845888935412085e-05, "reference_active": true, "delta_db": -0.38873803554112235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.372206568913873e-06, "output_power": 4.166511412583887e-06, "reference_active": true, "delta_db": -1.1038012780368411}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0554113039123233e-07, "output_power": 2.3063608124407236e-07, "reference_active": true, "delta_db": -2.4510765705525444}], "active_band_count": 3, "active_band_mae_db": 1.3145386280435025}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9509172278093344, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d424c2e71637658c223b513f8b5a1a44284c6c80f3c1b8d26fa1fff6daaa524f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9924646995676993, "standard_deviation": 0.0014456145692428562, "minimum": 0.9896200387159705, "maximum": 0.9943432590387368, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.7391685229041536, "standard_deviation": 0.020807546668010737, "minimum": 0.7051618137448336, "maximum": 0.7744889400665288, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2869575195841643, "standard_deviation": 0.7323760524722706, "minimum": 0.4609973351023651, "maximum": 2.4914717362468553, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10058082210971395, "standard_deviation": 0.009581605056022829, "minimum": 0.08711780206721229, "maximum": 0.11862232356922858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974921941441646, "standard_deviation": 0.0006465991996585813, "minimum": 0.9963628300990974, "maximum": 0.9986676310483182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.5107798195277544, "standard_deviation": 0.02249582085910848, "minimum": 0.48827610746585004, "maximum": 0.5493440241642215, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.562897577044637, "standard_deviation": 1.032031451566013, "minimum": 0.34335588525763666, "maximum": 3.0553959874681964, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05778640743882099, "standard_deviation": 0.008076893080755067, "minimum": 0.04181389858288084, "maximum": 0.07084310971599936, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.988647793719738, "standard_deviation": 0.0013312110857910506, "minimum": 0.9870969158064176, "maximum": 0.9903357631295671, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 0.9210293554597663, "standard_deviation": 0.024463870790081402, "minimum": 0.883543986582329, "maximum": 0.957407122448203, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.918270334798278, "standard_deviation": 0.25026351455797247, "minimum": 0.5863801836155823, "maximum": 1.3397502443129026, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12610722234495103, "standard_deviation": 0.007168878345270381, "minimum": 0.11614764355828373, "maximum": 0.13484576617212257, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9965273143418362, "standard_deviation": 0.0009293155913311246, "minimum": 0.9949084862221718, "maximum": 0.9976310164112304, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6460394548698316, "standard_deviation": 0.030045256218285913, "minimum": 0.6113822891022374, "maximum": 0.6852860064417311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1383167073008578, "standard_deviation": 0.36106239441329196, "minimum": 0.53756839773523, "maximum": 1.6771571380824184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07474452135345802, "standard_deviation": 0.008267077495774742, "minimum": 0.05912232599491685, "maximum": 0.08406397739471476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9961021327282054, "standard_deviation": 0.0005968775797248575, "minimum": 0.9953580814168822, "maximum": 0.9972533460298478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.8257179125575299, "standard_deviation": 0.023012653414409025, "minimum": 0.7907654821349933, "maximum": 0.8617538607525491, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6128716992511591, "standard_deviation": 0.3385595842276126, "minimum": 0.24829672091114752, "maximum": 1.1096315075525345, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06990136499041992, "standard_deviation": 0.0057296741196114715, "minimum": 0.05859636925769668, "maximum": 0.0761383156603259, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9987291372058249, "standard_deviation": 0.00035885042740316995, "minimum": 0.9982896055365841, "maximum": 0.9993944477680725, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6028662196636431, "standard_deviation": 0.036765136293621506, "minimum": 0.5436546854820102, "maximum": 0.6546047121258826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9502544061901557, "standard_deviation": 0.529744988313984, "minimum": 0.11813270295338914, "maximum": 1.7385116761596835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03907198447531894, "standard_deviation": 0.0062905983821102554, "minimum": 0.026901108389772117, "maximum": 0.04519104424189813, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14006875059613463, "standard_deviation": 0.0380985475493771, "minimum": 0.09267741011417939, "maximum": 0.1931380928778874, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.596511801246647, "standard_deviation": 0.008792622889136759, "minimum": 4.585354325651563, "maximum": 4.608743335190209, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5197530882301515, "standard_deviation": 0.016672700162034586, "minimum": 0.5042626313720509, "maximum": 0.5495502491400005, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05222007059714816, "standard_deviation": 0.0310277399285862, "minimum": -0.09280728220740983, "maximum": 0.0028789202822430628, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9969154974463773, "standard_deviation": 0.0006034324117431252, "minimum": 0.9960753122970846, "maximum": 0.997703581071154, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5999704492188566, "standard_deviation": 0.018835574466186552, "minimum": 0.5740687307606003, "maximum": 0.6268560983438324, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8305685490870519, "standard_deviation": 0.5990071846276289, "minimum": 0.2027499019090372, "maximum": 1.8686590615616245, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.058448794469132294, "standard_deviation": 0.0058210737503955065, "minimum": 0.050553052343714544, "maximum": 0.06631570465317445, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999078052294585, "standard_deviation": 0.00025642986275523354, "minimum": 0.99867654874196, "maximum": 0.9994478555986414, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41957195430986727, "standard_deviation": 0.029697792172154076, "minimum": 0.3763004877052665, "maximum": 0.45241951531055663, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.117538897720741, "standard_deviation": 0.7740035989900506, "minimum": 0.24702290989592876, "maximum": 2.4023985480526715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0315422749432808, "standard_deviation": 0.004658119525632123, "minimum": 0.024455649401486597, "maximum": 0.03832126228640086, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-121726-0012", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.669675350189209, "predicted_mos": 2.669675350189209, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "f5f93a092730a1d1a1a9d6a200de27247bdc4e6bcecafa08ada568e2c469eb59", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.818899099715054, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5268640954135154, "全帯域_10ms_rms_envelope_nrmse": 0.050318464184998794, "全帯域_10ms_envelope_correlation": 0.9977228457662087, "全帯域_10ms_modulation_mae_db": 0.5559009947231924, "全帯域_40ms_log_rms_mae_db": 0.3715052672596427, "全帯域_40ms_rms_envelope_nrmse": 0.027182683741308774, "全帯域_40ms_envelope_correlation": 0.9993104722064021, "全帯域_40ms_modulation_mae_db": 0.818487221022053, "300–1000 Hz_10ms_log_rms_mae_db": 0.9581013928675391, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0675346893358664, "300–1000 Hz_10ms_envelope_correlation": 0.9963904909993683, "300–1000 Hz_10ms_modulation_mae_db": 0.20822969532777633, "300–1000 Hz_40ms_log_rms_mae_db": 0.725278987010766, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03240445991793372, "300–1000 Hz_40ms_envelope_correlation": 0.9991541679736204, "300–1000 Hz_40ms_modulation_mae_db": 0.14458916211256134, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8392264319071425, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08757224199998721, "1000–3000 Hz_10ms_envelope_correlation": 0.9943248596725619, "1000–3000 Hz_10ms_modulation_mae_db": 1.078196997039539, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6386909853784812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05068160785631433, "1000–3000 Hz_40ms_envelope_correlation": 0.9980901233397438, "1000–3000 Hz_40ms_modulation_mae_db": 1.5793568910157145, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1749865973833828, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15596498907826845, "3000–6000 Hz_10ms_envelope_correlation": 0.9822302480571378, "3000–6000 Hz_10ms_modulation_mae_db": 1.292347982619906, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9131057644059589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09992485216177299, "3000–6000 Hz_40ms_envelope_correlation": 0.9932026892787768, "3000–6000 Hz_40ms_modulation_mae_db": 1.409944841174201, "mr_stft_db": 4.867979788081488, "band_mae_db": 0.12182458214950172, "rms_gain_error_db": -0.022055568310576607, "output_peak": 0.5000962248843637}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5932842181871494, "全帯域_10ms_rms_envelope_nrmse": 0.05613026593301858, "全帯域_10ms_envelope_correlation": 0.9971754039532917, "全帯域_10ms_modulation_mae_db": 0.807639280512797, "全帯域_40ms_log_rms_mae_db": 0.4179822196274599, "全帯域_40ms_rms_envelope_nrmse": 0.029735082662946315, "全帯域_40ms_envelope_correlation": 0.999193195924008, "全帯域_40ms_modulation_mae_db": 1.1282008584407597, "300–1000 Hz_10ms_log_rms_mae_db": 1.0050103511166026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08580519006925953, "300–1000 Hz_10ms_envelope_correlation": 0.9940848030592457, "300–1000 Hz_10ms_modulation_mae_db": 1.0501547470203534, "300–1000 Hz_40ms_log_rms_mae_db": 0.7200697232801219, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055265500530933924, "300–1000 Hz_40ms_envelope_correlation": 0.9974325441721995, "300–1000 Hz_40ms_modulation_mae_db": 1.3924112013137433, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8722449162035795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10517385380609971, "1000–3000 Hz_10ms_envelope_correlation": 0.9917495106080897, "1000–3000 Hz_10ms_modulation_mae_db": 0.8718002890313098, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6540023098577824, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06217620481653575, "1000–3000 Hz_40ms_envelope_correlation": 0.9969951546328755, "1000–3000 Hz_40ms_modulation_mae_db": 1.1836046894043417, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1246810855579035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15327140803403083, "3000–6000 Hz_10ms_envelope_correlation": 0.9826021484283327, "3000–6000 Hz_10ms_modulation_mae_db": 1.2208729742157132, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8080248550479057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08120253017499947, "3000–6000 Hz_40ms_envelope_correlation": 0.9956255914316883, "3000–6000 Hz_40ms_modulation_mae_db": 1.1400562378494576, "mr_stft_db": 4.887020046335109, "band_mae_db": 0.17616142936368095, "rms_gain_error_db": -0.05415754652895537, "output_peak": 0.5088040692249014}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.549905549008644, "全帯域_10ms_rms_envelope_nrmse": 0.056132157405717385, "全帯域_10ms_envelope_correlation": 0.9973105986420547, "全帯域_10ms_modulation_mae_db": 0.5905952030234114, "全帯域_40ms_log_rms_mae_db": 0.39352320731664575, "全帯域_40ms_rms_envelope_nrmse": 0.03162414875361218, "全帯域_40ms_envelope_correlation": 0.9992381344872966, "全帯域_40ms_modulation_mae_db": 0.7810751500798584, "300–1000 Hz_10ms_log_rms_mae_db": 0.982931414281452, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07187355380720616, "300–1000 Hz_10ms_envelope_correlation": 0.9959019684649042, "300–1000 Hz_10ms_modulation_mae_db": 0.6915565894196543, "300–1000 Hz_40ms_log_rms_mae_db": 0.7337687499958977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04034838497939223, "300–1000 Hz_40ms_envelope_correlation": 0.9987002560435974, "300–1000 Hz_40ms_modulation_mae_db": 0.904776393810848, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8835212860015373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10120879864113892, "1000–3000 Hz_10ms_envelope_correlation": 0.9929157300401716, "1000–3000 Hz_10ms_modulation_mae_db": 0.8176235563583835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6631108764058664, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0619321182278525, "1000–3000 Hz_40ms_envelope_correlation": 0.9977572355254463, "1000–3000 Hz_40ms_modulation_mae_db": 1.197847517126052, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1332237242847245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14354343547485673, "3000–6000 Hz_10ms_envelope_correlation": 0.9849138693375783, "3000–6000 Hz_10ms_modulation_mae_db": 1.0618072122007287, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8534053105336591, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09187227221559963, "3000–6000 Hz_40ms_envelope_correlation": 0.9941513916844014, "3000–6000 Hz_40ms_modulation_mae_db": 0.8487278491652898, "mr_stft_db": 4.874092795300913, "band_mae_db": 0.17362186832462856, "rms_gain_error_db": -0.1250419488224085, "output_peak": 0.5051119867414418}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5337143386331071, "全帯域_10ms_rms_envelope_nrmse": 0.049307438643777866, "全帯域_10ms_envelope_correlation": 0.9978110693420178, "全帯域_10ms_modulation_mae_db": 0.9988752491921037, "全帯域_40ms_log_rms_mae_db": 0.4053053179735653, "全帯域_40ms_rms_envelope_nrmse": 0.025783954856681172, "全帯域_40ms_envelope_correlation": 0.9993745741734154, "全帯域_40ms_modulation_mae_db": 1.3859949131424194, "300–1000 Hz_10ms_log_rms_mae_db": 0.9561671124699136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07078830173119276, "300–1000 Hz_10ms_envelope_correlation": 0.995976313556291, "300–1000 Hz_10ms_modulation_mae_db": 0.45742910856626334, "300–1000 Hz_40ms_log_rms_mae_db": 0.6906826439914703, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046803220195115895, "300–1000 Hz_40ms_envelope_correlation": 0.9981519630016333, "300–1000 Hz_40ms_modulation_mae_db": 0.5145146805635911, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8463054746277087, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09773177742695854, "1000–3000 Hz_10ms_envelope_correlation": 0.9928687009305294, "1000–3000 Hz_10ms_modulation_mae_db": 1.1173862202631615, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5971264661438376, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056934367155401525, "1000–3000 Hz_40ms_envelope_correlation": 0.9974981952982158, "1000–3000 Hz_40ms_modulation_mae_db": 1.6280310170480112, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1470496074915586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14868503088214, "3000–6000 Hz_10ms_envelope_correlation": 0.9832827129174067, "3000–6000 Hz_10ms_modulation_mae_db": 1.0263078809013235, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8963301447104728, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09676089127279883, "3000–6000 Hz_40ms_envelope_correlation": 0.9927151502928764, "3000–6000 Hz_40ms_modulation_mae_db": 1.3893246463351205, "mr_stft_db": 4.87674688698585, "band_mae_db": 0.1239486941377941, "rms_gain_error_db": -0.016515373619471287, "output_peak": 0.5186303964487347}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5897864567831094, "全帯域_10ms_rms_envelope_nrmse": 0.05660845031603702, "全帯域_10ms_envelope_correlation": 0.9971503311130897, "全帯域_10ms_modulation_mae_db": 0.562360696747792, "全帯域_40ms_log_rms_mae_db": 0.40609364358687816, "全帯域_40ms_rms_envelope_nrmse": 0.030964810942050048, "全帯域_40ms_envelope_correlation": 0.9991515375328079, "全帯域_40ms_modulation_mae_db": 1.0004175491117266, "300–1000 Hz_10ms_log_rms_mae_db": 0.8871744144534723, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.058725327201736885, "300–1000 Hz_10ms_envelope_correlation": 0.9972473691404122, "300–1000 Hz_10ms_modulation_mae_db": 0.15861289586580365, "300–1000 Hz_40ms_log_rms_mae_db": 0.6347132134898075, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0337401007675363, "300–1000 Hz_40ms_envelope_correlation": 0.9990617190435193, "300–1000 Hz_40ms_modulation_mae_db": 0.18471560843710635, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8336453078236165, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1032596155451772, "1000–3000 Hz_10ms_envelope_correlation": 0.9920735825681728, "1000–3000 Hz_10ms_modulation_mae_db": 1.0027926655727477, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6283577803656416, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05839194033885251, "1000–3000 Hz_40ms_envelope_correlation": 0.997443298496466, "1000–3000 Hz_40ms_modulation_mae_db": 1.6777303754961228, "3000–6000 Hz_10ms_log_rms_mae_db": 1.096233991021094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1434881388743475, "3000–6000 Hz_10ms_envelope_correlation": 0.9854753680784731, "3000–6000 Hz_10ms_modulation_mae_db": 1.2268759379493894, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8691731419051758, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08519464000247476, "3000–6000 Hz_40ms_envelope_correlation": 0.9958424238458532, "3000–6000 Hz_40ms_modulation_mae_db": 1.7960156046606277, "mr_stft_db": 4.876357856997004, "band_mae_db": 0.15160960597773063, "rms_gain_error_db": -0.07261562191958558, "output_peak": 0.5079255320124345}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5504862911755216, "全帯域_10ms_rms_envelope_nrmse": 0.05224916359169607, "全帯域_10ms_envelope_correlation": 0.9975997180628073, "全帯域_10ms_modulation_mae_db": 0.7253206138888059, "全帯域_40ms_log_rms_mae_db": 0.39574301312685867, "全帯域_40ms_rms_envelope_nrmse": 0.027283003283149275, "全帯域_40ms_envelope_correlation": 0.9993741310047092, "全帯域_40ms_modulation_mae_db": 1.0042559581753208, "300–1000 Hz_10ms_log_rms_mae_db": 0.935450925956894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06666220926699204, "300–1000 Hz_10ms_envelope_correlation": 0.9964583257990423, "300–1000 Hz_10ms_modulation_mae_db": 0.3609330333952398, "300–1000 Hz_40ms_log_rms_mae_db": 0.711941526186614, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03333433287491844, "300–1000 Hz_40ms_envelope_correlation": 0.9990864510778962, "300–1000 Hz_40ms_modulation_mae_db": 0.3265791126359708, "1000–3000 Hz_10ms_log_rms_mae_db": 0.889537326258387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10262798629260128, "1000–3000 Hz_10ms_envelope_correlation": 0.9922860899214667, "1000–3000 Hz_10ms_modulation_mae_db": 0.9646408586645562, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6854331567573348, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058268741223215445, "1000–3000 Hz_40ms_envelope_correlation": 0.9976173119052204, "1000–3000 Hz_40ms_modulation_mae_db": 1.356771645552442, "3000–6000 Hz_10ms_log_rms_mae_db": 1.135080736400349, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14502076741048875, "3000–6000 Hz_10ms_envelope_correlation": 0.9844617270126187, "3000–6000 Hz_10ms_modulation_mae_db": 1.481491827415442, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8547672303529426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08763709363800595, "3000–6000 Hz_40ms_envelope_correlation": 0.9946205611987556, "3000–6000 Hz_40ms_modulation_mae_db": 1.7077733021374935, "mr_stft_db": 4.8721024294800825, "band_mae_db": 0.1253828089875225, "rms_gain_error_db": -0.07936188085995216, "output_peak": 0.5092734050619256}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5702532971655533, "全帯域_10ms_rms_envelope_nrmse": 0.0647276316525869, "全帯域_10ms_envelope_correlation": 0.9963621771732986, "全帯域_10ms_modulation_mae_db": 1.4568093122553767, "全帯域_40ms_log_rms_mae_db": 0.4224817303170811, "全帯域_40ms_rms_envelope_nrmse": 0.040494201186358124, "全帯域_40ms_envelope_correlation": 0.9986282047706503, "全帯域_40ms_modulation_mae_db": 1.6658772280370024, "300–1000 Hz_10ms_log_rms_mae_db": 0.9132078926379985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08286240850784328, "300–1000 Hz_10ms_envelope_correlation": 0.9944896658764929, "300–1000 Hz_10ms_modulation_mae_db": 0.15572445521668984, "300–1000 Hz_40ms_log_rms_mae_db": 0.6763133599358383, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0402321106010142, "300–1000 Hz_40ms_envelope_correlation": 0.9986345980093649, "300–1000 Hz_40ms_modulation_mae_db": 0.18038037730312487, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8824171524530532, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1421792498581885, "1000–3000 Hz_10ms_envelope_correlation": 0.9855664878033215, "1000–3000 Hz_10ms_modulation_mae_db": 3.1541872926186927, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6723512427247214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1041045249783208, "1000–3000 Hz_40ms_envelope_correlation": 0.9925432984961066, "1000–3000 Hz_40ms_modulation_mae_db": 3.5070060013107134, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0987642148625363, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13816862873203142, "3000–6000 Hz_10ms_envelope_correlation": 0.9861531426827306, "3000–6000 Hz_10ms_modulation_mae_db": 1.1821552798984831, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7891107847037961, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07501171464375252, "3000–6000 Hz_40ms_envelope_correlation": 0.9965483489022885, "3000–6000 Hz_40ms_modulation_mae_db": 1.7025259129328025, "mr_stft_db": 4.869805434719775, "band_mae_db": 0.20099126265395717, "rms_gain_error_db": -0.1278662491991399, "output_peak": 0.5167193328013442}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5480828197980602, "全帯域_10ms_rms_envelope_nrmse": 0.05634552351276099, "全帯域_10ms_envelope_correlation": 0.9972971146323213, "全帯域_10ms_modulation_mae_db": 0.33644395673691907, "全帯域_40ms_log_rms_mae_db": 0.36579920971925106, "全帯域_40ms_rms_envelope_nrmse": 0.03215010192657472, "全帯域_40ms_envelope_correlation": 0.9992193915698132, "全帯域_40ms_modulation_mae_db": 0.4463688932212498, "300–1000 Hz_10ms_log_rms_mae_db": 0.9760489238896786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0681056074919965, "300–1000 Hz_10ms_envelope_correlation": 0.9964909525654525, "300–1000 Hz_10ms_modulation_mae_db": 0.4180922554354704, "300–1000 Hz_40ms_log_rms_mae_db": 0.7260021725444781, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038849039934994695, "300–1000 Hz_40ms_envelope_correlation": 0.9989869094748428, "300–1000 Hz_40ms_modulation_mae_db": 0.6645025422762244, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8473630576339656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09462575906342818, "1000–3000 Hz_10ms_envelope_correlation": 0.9935437146062456, "1000–3000 Hz_10ms_modulation_mae_db": 0.6429522636886332, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6615332239900318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05642058362553454, "1000–3000 Hz_40ms_envelope_correlation": 0.9978479916700402, "1000–3000 Hz_40ms_modulation_mae_db": 0.7605302613761239, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1322459244589391, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14507773035766136, "3000–6000 Hz_10ms_envelope_correlation": 0.9841101904668154, "3000–6000 Hz_10ms_modulation_mae_db": 0.921614629366872, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8485016705196947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08598478841756056, "3000–6000 Hz_40ms_envelope_correlation": 0.9943401028293967, "3000–6000 Hz_40ms_modulation_mae_db": 1.0535798703560386, "mr_stft_db": 4.883210797623274, "band_mae_db": 0.16085522773039326, "rms_gain_error_db": -0.13011320427913353, "output_peak": 0.5127813116813034}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64720, "output_samples": 64720, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.067128817687506, "spectral_convergence": 0.23596381541796996, "spectral_convergence_status": "ok", "amplitude_floor": 1.1807658352589429e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.752559323014237, "spectral_convergence": 0.22884788167097295, "spectral_convergence_status": "ok", "amplitude_floor": 7.24953185483845e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.7842512235427215, "spectral_convergence": 0.23751075076404649, "spectral_convergence_status": "ok", "amplitude_floor": 4.6531510695198173e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031402968364977683, "output_power": 0.00032163137116506565, "delta_db": 0.10387700510411844, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005546801738700932, "output_power": 0.0005650737939441447, "delta_db": 0.08062523760468017, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010899709362407707, "output_power": 0.0010925506880030927, "delta_db": 0.010266770916562598, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008204698714335104, "output_power": 0.0008131756103515227, "delta_db": -0.038782935770876464, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0004732551323003661, "output_power": 0.0004492931029506581, "delta_db": -0.22565580608914002, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00010423546715106559, "output_power": 9.791326740078497e-05, "delta_db": -0.2717397374116327, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12182458214950172, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.022055568310576607, "pitch": null, "output_rms_dbfs": -24.763006275907454, "output_peak": 0.5000962248843637, "reference_rms_dbfs": -24.740950707596877, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.867979788081488}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64720, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00017261805365043103, "log_rms_mae_db": 0.5268640954135154, "rms_envelope_nrmse": 0.050318464184998794, "envelope_correlation": 0.9977228457662087, "rms_gain_error_db": -0.022055569428870117, "reference_mean_power": 0.003360795516302892, "output_mean_power": 0.0033437710407986806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.9828935159721717e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046042689591210996, "output_power": 0.00045892545994886567, "reference_active": true, "delta_db": -0.014185332301237252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011507848708673394, "output_power": 0.00010837669620581391, "reference_active": true, "delta_db": -0.26058235986367034}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.734946402574847e-05, "output_power": 2.7101344997047423e-05, "reference_active": true, "delta_db": -1.3929352920046698}], "active_band_count": 3, "active_band_mae_db": 0.5559009947231924}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00013379439656886598, "log_rms_mae_db": 0.3715052672596427, "rms_envelope_nrmse": 0.027182683741308774, "envelope_correlation": 0.9993104722064021, "rms_gain_error_db": -0.022055562111366882, "reference_mean_power": 0.0033859385941668236, "output_mean_power": 0.0033687867593410422}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.884361456130501e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004231625539622664, "output_power": 0.00042065680532486907, "reference_active": true, "delta_db": -0.02579310353621788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.583961694519604e-05, "output_power": 5.021863343352602e-05, "reference_active": true, "delta_db": -0.4607754003475795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3554104634271355e-06, "output_power": 2.132339434226466e-06, "reference_active": true, "delta_db": -1.9688931591823615}], "active_band_count": 3, "active_band_mae_db": 0.818487221022053}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00011786303332041996, "log_rms_mae_db": 0.9581013928675391, "rms_envelope_nrmse": 0.0675346893358664, "envelope_correlation": 0.9963904909993683, "rms_gain_error_db": 0.039768896316044335, "reference_mean_power": 0.0013229704572187174, "output_mean_power": 0.0013351407026221754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 9.694412540428036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002357826301518165, "output_power": 0.00023735008312351601, "reference_active": true, "delta_db": 0.028775802122076505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.91091063998743e-05, "output_power": 5.523573450144469e-05, "reference_active": true, "delta_db": -0.29434260370584153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.351021856959089e-05, "output_power": 1.4481696554471951e-05, "reference_active": true, "delta_db": 0.30157068015541083}], "active_band_count": 3, "active_band_mae_db": 0.20822969532777633}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010411406065660645, "log_rms_mae_db": 0.725278987010766, "rms_envelope_nrmse": 0.03240445991793372, "envelope_correlation": 0.9991541679736204, "rms_gain_error_db": 0.039768892223204774, "reference_mean_power": 0.0013328679887454193, "output_mean_power": 0.0013451292820704903}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 9.215200160073525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002182221456522984, "output_power": 0.00021893928554306836, "reference_active": true, "delta_db": 0.014248749064626625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.729502442376393e-05, "output_power": 2.5890591274076156e-05, "reference_active": true, "delta_db": -0.2294151854191646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1926400570958608e-06, "output_power": 1.1415607724889307e-06, "reference_active": true, "delta_db": -0.1901035518538928}], "active_band_count": 3, "active_band_mae_db": 0.14458916211256134}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00014037287410750346, "log_rms_mae_db": 0.8392264319071425, "rms_envelope_nrmse": 0.08757224199998721, "envelope_correlation": 0.9943248596725619, "rms_gain_error_db": -0.09581862222036508, "reference_mean_power": 0.001043588584426717, "output_mean_power": 0.0010208159744694535}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 7.131098310180307e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021758763787569653, "output_power": 0.00021594600083733685, "reference_active": true, "delta_db": -0.032890518797826354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8027856042619025e-05, "output_power": 4.784370519019684e-05, "reference_active": true, "delta_db": -0.8381171934249936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3022954991601597e-05, "output_power": 1.3359885516196584e-05, "reference_active": true, "delta_db": -2.3635832788957973}], "active_band_count": 3, "active_band_mae_db": 1.078196997039539}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010015064346906524, "log_rms_mae_db": 0.6386909853784812, "rms_envelope_nrmse": 0.05068160785631433, "envelope_correlation": 0.9980901233397438, "rms_gain_error_db": -0.09581862552119667, "reference_mean_power": 0.0010513959774729573, "output_mean_power": 0.0010284529981431075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 6.799409328431789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020213626575120905, "output_power": 0.00019902071472925606, "reference_active": true, "delta_db": -0.0674595676073605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0065691390851235e-05, "output_power": 2.3060304619626523e-05, "reference_active": true, "delta_db": -1.1520615544042507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1255469357025214e-06, "output_power": 9.454003184788248e-07, "reference_active": true, "delta_db": -3.518549551035533}], "active_band_count": 3, "active_band_mae_db": 1.5793568910157145}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 5.7697319934351994e-05, "log_rms_mae_db": 1.1749865973833828, "rms_envelope_nrmse": 0.15596498907826845, "envelope_correlation": 0.9822302480571378, "rms_gain_error_db": -0.37628395270886783, "reference_mean_power": 0.0001838054244420426, "output_mean_power": 0.00016855045483689049}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.2570064040799192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.4129018402509514e-05, "output_power": 4.0626852471032716e-05, "reference_active": true, "delta_db": -0.3591109013665849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0833298446779821e-05, "output_power": 1.2892852686706072e-05, "reference_active": true, "delta_db": 0.7558831275881608}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.606324451113381e-06, "output_power": 2.4386503303146754e-06, "reference_active": true, "delta_db": -2.7620499189049723}], "active_band_count": 3, "active_band_mae_db": 1.292347982619906}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 3.902816175045304e-05, "log_rms_mae_db": 0.9131057644059589, "rms_envelope_nrmse": 0.09992485216177299, "envelope_correlation": 0.9932026892787768, "rms_gain_error_db": -0.3762839651097146, "reference_mean_power": 0.00018518052639832005, "output_mean_power": 0.00016981142943411863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.1869756093173816e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2060123324004214e-05, "output_power": 3.893612692565796e-05, "reference_active": true, "delta_db": -0.33517792772475763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.372206568913873e-06, "output_power": 5.978547133982372e-06, "reference_active": true, "delta_db": 0.4644295395492978}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0554113039123233e-07, "output_power": 1.8408237196928835e-07, "reference_active": true, "delta_db": -3.4302270562485484}], "active_band_count": 3, "active_band_mae_db": 1.409944841174201}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9503714820776905, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3397e1f8e809919c68e27032020cbe1a86e2d2cefbf4730699e7b23289d7c6c5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9919160845188199, "standard_deviation": 0.0026956308985821935, "minimum": 0.9855664878033215, "maximum": 0.9943248596725619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.8617826191136237, "standard_deviation": 0.022443874938922317, "minimum": 0.8336453078236165, "maximum": 0.889537326258387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.206197517904628, "standard_deviation": 0.801669271187418, "minimum": 0.6429522636886332, "maximum": 3.1541872926186927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10429741032919744, "standard_deviation": 0.01632519991262319, "minimum": 0.08757224199998721, "maximum": 0.1421792498581885, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969740761705143, "standard_deviation": 0.0018190847746944766, "minimum": 0.9925432984961066, "maximum": 0.9980901233397438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6500757552029622, "standard_deviation": 0.027948574262397427, "minimum": 0.5971264661438376, "maximum": 0.6854331567573348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.6113597997911904, "standard_deviation": 0.8230496048585756, "minimum": 0.7605302613761239, "maximum": 3.5070060013107134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06361376102775343, "standard_deviation": 0.016749540688699237, "minimum": 0.05068160785631433, "maximum": 0.1041045249783208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9841536758726366, "standard_deviation": 0.0013776173448792897, "minimum": 0.9822302480571378, "maximum": 0.9861531426827306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.130283235182561, "standard_deviation": 0.025352374433938667, "minimum": 1.096233991021094, "maximum": 1.1749865973833828, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1766842155709822, "standard_deviation": 0.17388741715561912, "minimum": 0.921614629366872, "maximum": 1.481491827415442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14665251610547814, "standard_deviation": 0.005750298925114576, "minimum": 0.13816862873203142, "maximum": 0.15596498907826845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9946307824330046, "standard_deviation": 0.001317914595660667, "minimum": 0.9927151502928764, "maximum": 0.9965483489022885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8540523627724508, "standard_deviation": 0.04118084328871931, "minimum": 0.7891107847037961, "maximum": 0.9131057644059589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.380993533076379, "standard_deviation": 0.3447310005934462, "minimum": 0.8487278491652898, "maximum": 1.7960156046606277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08794859781587058, "standard_deviation": 0.008114296427485939, "minimum": 0.07501171464375252, "maximum": 0.09992485216177299, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9958799861826511, "standard_deviation": 0.0010692816772947956, "minimum": 0.9940848030592457, "maximum": 0.9972473691404122, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9517615534591939, "standard_deviation": 0.03850861389567957, "minimum": 0.8871744144534723, "maximum": 1.0050103511166026, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4375915975309064, "standard_deviation": 0.30619023030519177, "minimum": 0.15572445521668984, "maximum": 1.0501547470203534, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0715446609265117, "standard_deviation": 0.008849656680533193, "minimum": 0.058725327201736885, "maximum": 0.08580519006925953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986510760995843, "standard_deviation": 0.0005928970792105811, "minimum": 0.9974325441721995, "maximum": 0.9991541679736204, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7023462970543742, "standard_deviation": 0.03352409822420359, "minimum": 0.6347132134898075, "maximum": 0.7337687499958977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5390586348066462, "standard_deviation": 0.4364240150846653, "minimum": 0.14458916211256134, "maximum": 1.3924112013137433, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.040122143725229924, "standard_deviation": 0.007764475890910813, "minimum": 0.03240445991793372, "maximum": 0.055265500530933924, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1542994349156511, "standard_deviation": 0.02901624764997775, "minimum": 0.12182458214950172, "maximum": 0.20099126265395717, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.875914504440438, "standard_deviation": 0.006498234632979852, "minimum": 4.867979788081488, "maximum": 4.887020046335109, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5099177823570562, "standard_deviation": 0.006042159454448291, "minimum": 0.5000962248843637, "maximum": 0.5186303964487347, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07846592419240286, "standard_deviation": 0.046172169552245644, "minimum": -0.13011320427913353, "maximum": -0.016515373619471287, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973036573356362, "standard_deviation": 0.00045470109241427, "minimum": 0.9963621771732986, "maximum": 0.9978110693420178, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5577971332705826, "standard_deviation": 0.024450338188737018, "minimum": 0.5268640954135154, "maximum": 0.5932842181871494, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7542431633850497, "standard_deviation": 0.34530049832761484, "minimum": 0.33644395673691907, "maximum": 1.4568093122553767, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0552273869050742, "standard_deviation": 0.004826402680961923, "minimum": 0.049307438643777866, "maximum": 0.0647276316525869, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991862052086378, "standard_deviation": 0.00023983757886386446, "minimum": 0.9986282047706503, "maximum": 0.9993745741734154, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.39730420111592285, "standard_deviation": 0.020275401314439334, "minimum": 0.36579920971925106, "maximum": 0.4224817303170811, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0288347214037987, "standard_deviation": 0.3759880612452446, "minimum": 0.4463688932212498, "maximum": 1.6658772280370024, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.030652248419085076, "standard_deviation": 0.004599395178676457, "minimum": 0.025783954856681172, "maximum": 0.040494201186358124, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-121726-0012", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.0406477451324463, "predicted_mos": 3.0406477451324463, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "87f28ad7853e00ef202838f89948c1dc03e4789eb05a7e2036e1f964b1fcf2d8", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9528869003988802, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.584620176784253, "全帯域_10ms_rms_envelope_nrmse": 0.055588899068444296, "全帯域_10ms_envelope_correlation": 0.997214475815558, "全帯域_10ms_modulation_mae_db": 0.29307322775007655, "全帯域_40ms_log_rms_mae_db": 0.40895802912052437, "全帯域_40ms_rms_envelope_nrmse": 0.026237211772671246, "全帯域_40ms_envelope_correlation": 0.9993494864788424, "全帯域_40ms_modulation_mae_db": 0.5602086861001335, "300–1000 Hz_10ms_log_rms_mae_db": 0.8009463434053173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05431979385709911, "300–1000 Hz_10ms_envelope_correlation": 0.9976564026452757, "300–1000 Hz_10ms_modulation_mae_db": 0.6331327568496956, "300–1000 Hz_40ms_log_rms_mae_db": 0.6034657400648985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.029916512380132165, "300–1000 Hz_40ms_envelope_correlation": 0.9992584220928681, "300–1000 Hz_40ms_modulation_mae_db": 0.8557079671243882, "1000–3000 Hz_10ms_log_rms_mae_db": 0.6826049223142533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09334756108196146, "1000–3000 Hz_10ms_envelope_correlation": 0.9935031507155917, "1000–3000 Hz_10ms_modulation_mae_db": 0.3895259390098846, "1000–3000 Hz_40ms_log_rms_mae_db": 0.45141486219845706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04184613811665808, "1000–3000 Hz_40ms_envelope_correlation": 0.9986356508481063, "1000–3000 Hz_40ms_modulation_mae_db": 0.46394939789362377, "3000–6000 Hz_10ms_log_rms_mae_db": 0.834294158513231, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1181957609305471, "3000–6000 Hz_10ms_envelope_correlation": 0.9899050129407128, "3000–6000 Hz_10ms_modulation_mae_db": 1.1880578405190698, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5463964772889764, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06842205083142551, "3000–6000 Hz_40ms_envelope_correlation": 0.9969848763704561, "3000–6000 Hz_40ms_modulation_mae_db": 1.1972852790123012, "mr_stft_db": 4.546318347905953, "band_mae_db": 0.1106284788973232, "rms_gain_error_db": -0.0012897173853025326, "output_peak": 0.5316845059933387}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.592664363285657, "全帯域_10ms_rms_envelope_nrmse": 0.05616615680750354, "全帯域_10ms_envelope_correlation": 0.9973163638994917, "全帯域_10ms_modulation_mae_db": 0.8124637159644078, "全帯域_40ms_log_rms_mae_db": 0.4019503939537834, "全帯域_40ms_rms_envelope_nrmse": 0.03231451494564946, "全帯域_40ms_envelope_correlation": 0.9992082358610316, "全帯域_40ms_modulation_mae_db": 1.2681349588653743, "300–1000 Hz_10ms_log_rms_mae_db": 0.7500529480142853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0692315651854953, "300–1000 Hz_10ms_envelope_correlation": 0.9961904595777449, "300–1000 Hz_10ms_modulation_mae_db": 0.41472872139273503, "300–1000 Hz_40ms_log_rms_mae_db": 0.5327717678226511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04242870530319483, "300–1000 Hz_40ms_envelope_correlation": 0.9985406854189545, "300–1000 Hz_40ms_modulation_mae_db": 0.2545954454095478, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7179204949667096, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10419253410154908, "1000–3000 Hz_10ms_envelope_correlation": 0.9923541874088742, "1000–3000 Hz_10ms_modulation_mae_db": 1.191542917615455, "1000–3000 Hz_40ms_log_rms_mae_db": 0.4906590321167931, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058366519949091894, "1000–3000 Hz_40ms_envelope_correlation": 0.9979975659616662, "1000–3000 Hz_40ms_modulation_mae_db": 1.874751800712616, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8899372975357133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12496389573699454, "3000–6000 Hz_10ms_envelope_correlation": 0.9888618964501766, "3000–6000 Hz_10ms_modulation_mae_db": 1.0914253013429658, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6129066938038771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08266499565548409, "3000–6000 Hz_40ms_envelope_correlation": 0.995507201128012, "3000–6000 Hz_40ms_modulation_mae_db": 1.4141320484997009, "mr_stft_db": 4.568363997388702, "band_mae_db": 0.14252291134532527, "rms_gain_error_db": -0.12726398000789574, "output_peak": 0.5208821323766814}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5935574058355904, "全帯域_10ms_rms_envelope_nrmse": 0.06121990921732311, "全帯域_10ms_envelope_correlation": 0.9967387711452842, "全帯域_10ms_modulation_mae_db": 0.90712665985762, "全帯域_40ms_log_rms_mae_db": 0.4302484689752165, "全帯域_40ms_rms_envelope_nrmse": 0.0389398657505712, "全帯域_40ms_envelope_correlation": 0.9987099311518111, "全帯域_40ms_modulation_mae_db": 1.2365430686384353, "300–1000 Hz_10ms_log_rms_mae_db": 0.7906491544563923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.058088804102833014, "300–1000 Hz_10ms_envelope_correlation": 0.9973052178154729, "300–1000 Hz_10ms_modulation_mae_db": 0.3607936473820252, "300–1000 Hz_40ms_log_rms_mae_db": 0.5987227204614605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.029735511441379633, "300–1000 Hz_40ms_envelope_correlation": 0.9992747085970578, "300–1000 Hz_40ms_modulation_mae_db": 0.5739109629929473, "1000–3000 Hz_10ms_log_rms_mae_db": 0.72702432893913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11152768338091987, "1000–3000 Hz_10ms_envelope_correlation": 0.9908342560150304, "1000–3000 Hz_10ms_modulation_mae_db": 0.9327020475892812, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5221704560996293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07409347526814335, "1000–3000 Hz_40ms_envelope_correlation": 0.9959471432551771, "1000–3000 Hz_40ms_modulation_mae_db": 1.4788743783245153, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8731579000111406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13015633790236464, "3000–6000 Hz_10ms_envelope_correlation": 0.9890747239194835, "3000–6000 Hz_10ms_modulation_mae_db": 1.1461337758712142, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6226267070702971, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08142693591682566, "3000–6000 Hz_40ms_envelope_correlation": 0.9971907958871145, "3000–6000 Hz_40ms_modulation_mae_db": 1.4557438052969085, "mr_stft_db": 4.555409383068192, "band_mae_db": 0.17688346054825135, "rms_gain_error_db": -0.11023577423817874, "output_peak": 0.5536485292828371}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5466159122654686, "全帯域_10ms_rms_envelope_nrmse": 0.06516241128542721, "全帯域_10ms_envelope_correlation": 0.9962455495000418, "全帯域_10ms_modulation_mae_db": 0.9531457037942052, "全帯域_40ms_log_rms_mae_db": 0.37516142274139386, "全帯域_40ms_rms_envelope_nrmse": 0.035104278989238395, "全帯域_40ms_envelope_correlation": 0.9989461506139556, "全帯域_40ms_modulation_mae_db": 1.3632048844268796, "300–1000 Hz_10ms_log_rms_mae_db": 0.763229982800355, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06766310321052921, "300–1000 Hz_10ms_envelope_correlation": 0.9963319583545894, "300–1000 Hz_10ms_modulation_mae_db": 0.12002145581361684, "300–1000 Hz_40ms_log_rms_mae_db": 0.6052491172310137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04327493707525792, "300–1000 Hz_40ms_envelope_correlation": 0.9984218507619931, "300–1000 Hz_40ms_modulation_mae_db": 0.14122637763594945, "1000–3000 Hz_10ms_log_rms_mae_db": 0.696902333360621, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11578909858877508, "1000–3000 Hz_10ms_envelope_correlation": 0.9905638492133411, "1000–3000 Hz_10ms_modulation_mae_db": 2.181623980651182, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5020717576606709, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06802962371227336, "1000–3000 Hz_40ms_envelope_correlation": 0.9972302028824723, "1000–3000 Hz_40ms_modulation_mae_db": 2.8444149422561207, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8905990652362833, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1240956461309656, "3000–6000 Hz_10ms_envelope_correlation": 0.9890887041167574, "3000–6000 Hz_10ms_modulation_mae_db": 0.40097045256421193, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6289382195128964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0705198446207498, "3000–6000 Hz_40ms_envelope_correlation": 0.9970772762834264, "3000–6000 Hz_40ms_modulation_mae_db": 0.6167336990562203, "mr_stft_db": 4.563975356416282, "band_mae_db": 0.1741504517901806, "rms_gain_error_db": -0.10015280050036844, "output_peak": 0.5020650814888182}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5720078172928983, "全帯域_10ms_rms_envelope_nrmse": 0.0667986149124771, "全帯域_10ms_envelope_correlation": 0.9959777319859345, "全帯域_10ms_modulation_mae_db": 0.4858593049570244, "全帯域_40ms_log_rms_mae_db": 0.378232561802046, "全帯域_40ms_rms_envelope_nrmse": 0.03478987559013961, "全帯域_40ms_envelope_correlation": 0.9988721377401074, "全帯域_40ms_modulation_mae_db": 0.5157255827789292, "300–1000 Hz_10ms_log_rms_mae_db": 0.7761076006874437, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07740249436818775, "300–1000 Hz_10ms_envelope_correlation": 0.9953057771523487, "300–1000 Hz_10ms_modulation_mae_db": 0.4542686562175084, "300–1000 Hz_40ms_log_rms_mae_db": 0.5713969470403958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04278423923365305, "300–1000 Hz_40ms_envelope_correlation": 0.9985418933676875, "300–1000 Hz_40ms_modulation_mae_db": 0.4079936112435607, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7239319120348553, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10674339930090791, "1000–3000 Hz_10ms_envelope_correlation": 0.9916162388014839, "1000–3000 Hz_10ms_modulation_mae_db": 0.5263832074413551, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5176242870529066, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060136918722151524, "1000–3000 Hz_40ms_envelope_correlation": 0.9974121346239554, "1000–3000 Hz_40ms_modulation_mae_db": 0.6570360917588532, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8918426252960417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13454775231807892, "3000–6000 Hz_10ms_envelope_correlation": 0.9867337416695435, "3000–6000 Hz_10ms_modulation_mae_db": 1.5923792219308923, "3000–6000 Hz_40ms_log_rms_mae_db": 0.608617822031469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08550624942716593, "3000–6000 Hz_40ms_envelope_correlation": 0.994891920240596, "3000–6000 Hz_40ms_modulation_mae_db": 2.03404492869101, "mr_stft_db": 4.549785664386339, "band_mae_db": 0.148846126802055, "rms_gain_error_db": -0.0443738467769564, "output_peak": 0.5117401364325064}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.584225523710014, "全帯域_10ms_rms_envelope_nrmse": 0.05861787752468585, "全帯域_10ms_envelope_correlation": 0.996904767637265, "全帯域_10ms_modulation_mae_db": 1.0113843621803245, "全帯域_40ms_log_rms_mae_db": 0.4061929378921562, "全帯域_40ms_rms_envelope_nrmse": 0.033840248031420106, "全帯域_40ms_envelope_correlation": 0.9989258505045939, "全帯域_40ms_modulation_mae_db": 1.4416717892719768, "300–1000 Hz_10ms_log_rms_mae_db": 0.7903362894840866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06512309207432455, "300–1000 Hz_10ms_envelope_correlation": 0.9967403573600216, "300–1000 Hz_10ms_modulation_mae_db": 0.2939058147500146, "300–1000 Hz_40ms_log_rms_mae_db": 0.6084179226569039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03887442901249961, "300–1000 Hz_40ms_envelope_correlation": 0.9988484101288437, "300–1000 Hz_40ms_modulation_mae_db": 0.44241709686248476, "1000–3000 Hz_10ms_log_rms_mae_db": 0.712630637837074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09503969531182326, "1000–3000 Hz_10ms_envelope_correlation": 0.9935621304863244, "1000–3000 Hz_10ms_modulation_mae_db": 1.4597008210481743, "1000–3000 Hz_40ms_log_rms_mae_db": 0.4877045287343424, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05819883109514762, "1000–3000 Hz_40ms_envelope_correlation": 0.9977804770337827, "1000–3000 Hz_40ms_modulation_mae_db": 2.0446498234853165, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8482969325696992, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12409292687519081, "3000–6000 Hz_10ms_envelope_correlation": 0.9884558933109425, "3000–6000 Hz_10ms_modulation_mae_db": 0.7745924004665196, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5786287255949273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07169055021719513, "3000–6000 Hz_40ms_envelope_correlation": 0.9960799660455708, "3000–6000 Hz_40ms_modulation_mae_db": 1.322388985447914, "mr_stft_db": 4.567502487971058, "band_mae_db": 0.13166962030575008, "rms_gain_error_db": -0.03395606157374462, "output_peak": 0.5122057896916642}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5810997471151013, "全帯域_10ms_rms_envelope_nrmse": 0.05898861274098882, "全帯域_10ms_envelope_correlation": 0.996935638687297, "全帯域_10ms_modulation_mae_db": 1.194582531509594, "全帯域_40ms_log_rms_mae_db": 0.4286340515899962, "全帯域_40ms_rms_envelope_nrmse": 0.034832298763705805, "全帯域_40ms_envelope_correlation": 0.9989503958628229, "全帯域_40ms_modulation_mae_db": 1.5160672529290748, "300–1000 Hz_10ms_log_rms_mae_db": 0.7687407960599286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06592877319040227, "300–1000 Hz_10ms_envelope_correlation": 0.9965137140381644, "300–1000 Hz_10ms_modulation_mae_db": 0.5193556496787052, "300–1000 Hz_40ms_log_rms_mae_db": 0.5743561499551455, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04190123930682616, "300–1000 Hz_40ms_envelope_correlation": 0.9985276322993352, "300–1000 Hz_40ms_modulation_mae_db": 0.6544738751504573, "1000–3000 Hz_10ms_log_rms_mae_db": 0.695863865974513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09180773469945558, "1000–3000 Hz_10ms_envelope_correlation": 0.9940268972622489, "1000–3000 Hz_10ms_modulation_mae_db": 1.2748603109716135, "1000–3000 Hz_40ms_log_rms_mae_db": 0.48363870439387535, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05765045797307529, "1000–3000 Hz_40ms_envelope_correlation": 0.9978321241556912, "1000–3000 Hz_40ms_modulation_mae_db": 1.8687524857671056, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8540651808416877, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11736212050135038, "3000–6000 Hz_10ms_envelope_correlation": 0.9908460975670778, "3000–6000 Hz_10ms_modulation_mae_db": 1.4406856300390853, "3000–6000 Hz_40ms_log_rms_mae_db": 0.649939725695184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07784295277639909, "3000–6000 Hz_40ms_envelope_correlation": 0.9967858827689944, "3000–6000 Hz_40ms_modulation_mae_db": 1.8289842093233017, "mr_stft_db": 4.5545835510141215, "band_mae_db": 0.1663940058478796, "rms_gain_error_db": -0.09476541487984423, "output_peak": 0.5245927348364009}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5819116697308078, "全帯域_10ms_rms_envelope_nrmse": 0.05468525503196416, "全帯域_10ms_envelope_correlation": 0.997401607088933, "全帯域_10ms_modulation_mae_db": 0.423808128402388, "全帯域_40ms_log_rms_mae_db": 0.4372381920110589, "全帯域_40ms_rms_envelope_nrmse": 0.03222421119132951, "全帯域_40ms_envelope_correlation": 0.9991414706906497, "全帯域_40ms_modulation_mae_db": 0.6198959586491721, "300–1000 Hz_10ms_log_rms_mae_db": 0.8355266379159867, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08237581487376489, "300–1000 Hz_10ms_envelope_correlation": 0.9945712253272703, "300–1000 Hz_10ms_modulation_mae_db": 0.8466017191522704, "300–1000 Hz_40ms_log_rms_mae_db": 0.6573664206524635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048189738717428786, "300–1000 Hz_40ms_envelope_correlation": 0.9980938229340005, "300–1000 Hz_40ms_modulation_mae_db": 0.5009705513777961, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7206719740034817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09550553033745891, "1000–3000 Hz_10ms_envelope_correlation": 0.9933200681302826, "1000–3000 Hz_10ms_modulation_mae_db": 0.8543560252154258, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5170471408955492, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054761927974890864, "1000–3000 Hz_40ms_envelope_correlation": 0.9978752021476929, "1000–3000 Hz_40ms_modulation_mae_db": 1.1551336374748142, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8767206605892628, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14193000310661424, "3000–6000 Hz_10ms_envelope_correlation": 0.9852894263645807, "3000–6000 Hz_10ms_modulation_mae_db": 1.8844438531994376, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6250200872915359, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09350120107655821, "3000–6000 Hz_40ms_envelope_correlation": 0.9939002446554186, "3000–6000 Hz_40ms_modulation_mae_db": 2.334159249979875, "mr_stft_db": 4.558762815462973, "band_mae_db": 0.13553383683415232, "rms_gain_error_db": -0.10464861740426237, "output_peak": 0.5058271527318818}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64720, "output_samples": 64720, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.678115444260225, "spectral_convergence": 0.23040398520351776, "spectral_convergence_status": "ok", "amplitude_floor": 1.1807658352589429e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.353838710670609, "spectral_convergence": 0.2190898276338432, "spectral_convergence_status": "ok", "amplitude_floor": 7.24953185483845e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.6070008887870255, "spectral_convergence": 0.22347059882157327, "spectral_convergence_status": "ok", "amplitude_floor": 4.6531510695198173e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031402968364977683, "output_power": 0.0003216561056047493, "delta_db": 0.10421097804967652, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005546801738700932, "output_power": 0.0005725677514867935, "delta_db": 0.13784240758588678, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010899709362407707, "output_power": 0.001086030609247441, "delta_db": -0.01572851953234406, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008204698714335104, "output_power": 0.0008128392325466711, "delta_db": -0.0405798077638205, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0004732551323003661, "output_power": 0.00046503247943338344, "delta_db": -0.07612045908502366, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00010423546715106559, "output_power": 9.751841792906069e-05, "delta_db": -0.28928870136718776, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1106284788973232, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0012897173853025326, "pitch": null, "output_rms_dbfs": -24.742240424982178, "output_peak": 0.5316845059933387, "reference_rms_dbfs": -24.740950707596877, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.546318347905953}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64720, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00017261805365043103, "log_rms_mae_db": 0.584620176784253, "rms_envelope_nrmse": 0.055588899068444296, "envelope_correlation": 0.997214475815558, "rms_gain_error_db": -0.0012897221468675991, "reference_mean_power": 0.003360795516302892, "output_mean_power": 0.003359797610723442}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.9828935159721717e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046042689591210996, "output_power": 0.00045548184550076064, "reference_active": true, "delta_db": -0.046896134454477156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011507848708673394, "output_power": 0.00012218135803527833, "reference_active": true, "delta_db": 0.26010804293260215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.734946402574847e-05, "output_power": 3.273880501534203e-05, "reference_active": true, "delta_db": -0.5722155058631503}], "active_band_count": 3, "active_band_mae_db": 0.29307322775007655}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00013379439656886598, "log_rms_mae_db": 0.40895802912052437, "rms_envelope_nrmse": 0.026237211772671246, "envelope_correlation": 0.9993494864788424, "rms_gain_error_db": -0.0012897150175206883, "reference_mean_power": 0.0033859385941668236, "output_mean_power": 0.0033849332285254058}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.884361456130501e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004231625539622664, "output_power": 0.00042098478553196983, "reference_active": true, "delta_db": -0.022408289789487282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.583961694519604e-05, "output_power": 5.876285185711605e-05, "reference_active": true, "delta_db": 0.22160434245950347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3554104634271355e-06, "output_power": 2.4103733217866233e-06, "reference_active": true, "delta_db": -1.4366134260514098}], "active_band_count": 3, "active_band_mae_db": 0.5602086861001335}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00011786303332041996, "log_rms_mae_db": 0.8009463434053173, "rms_envelope_nrmse": 0.05431979385709911, "envelope_correlation": 0.9976564026452757, "rms_gain_error_db": 0.016473759770265915, "reference_mean_power": 0.0013229704572187174, "output_mean_power": 0.0013279983095260397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 9.694412540428036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002357826301518165, "output_power": 0.00023383964080037512, "reference_active": true, "delta_db": -0.035936727057336734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.91091063998743e-05, "output_power": 5.442715941862736e-05, "reference_active": true, "delta_db": -0.35838724961178114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.351021856959089e-05, "output_power": 9.553333650234277e-06, "reference_active": true, "delta_db": -1.5050742938799688}], "active_band_count": 3, "active_band_mae_db": 0.6331327568496956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010411406065660645, "log_rms_mae_db": 0.6034657400648985, "rms_envelope_nrmse": 0.029916512380132165, "envelope_correlation": 0.9992584220928681, "rms_gain_error_db": 0.01647375680822267, "reference_mean_power": 0.0013328679887454193, "output_mean_power": 0.0013379334549837383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 9.215200160073525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002182221456522984, "output_power": 0.00021454381546469995, "reference_active": true, "delta_db": -0.07382821742581386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.729502442376393e-05, "output_power": 2.488937337794508e-05, "reference_active": true, "delta_db": -0.4006952442877701}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1926400570958608e-06, "output_power": 7.366299083041772e-07, "reference_active": true, "delta_db": -2.0926004396595808}], "active_band_count": 3, "active_band_mae_db": 0.8557079671243882}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00014037287410750346, "log_rms_mae_db": 0.6826049223142533, "rms_envelope_nrmse": 0.09334756108196146, "envelope_correlation": 0.9935031507155917, "rms_gain_error_db": -0.009263382679261927, "reference_mean_power": 0.001043588584426717, "output_mean_power": 0.0010413650107338845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 7.131098310180307e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021758763787569653, "output_power": 0.00021489036799308904, "reference_active": true, "delta_db": -0.0541726792551319}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8027856042619025e-05, "output_power": 6.813287493303285e-05, "reference_active": true, "delta_db": 0.697201898384123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3022954991601597e-05, "output_power": 2.534436347797373e-05, "reference_active": true, "delta_db": 0.4172032393903987}], "active_band_count": 3, "active_band_mae_db": 0.3895259390098846}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010015064346906524, "log_rms_mae_db": 0.45141486219845706, "rms_envelope_nrmse": 0.04184613811665808, "envelope_correlation": 0.9986356508481063, "rms_gain_error_db": -0.009263384068226115, "reference_mean_power": 0.0010513959774729573, "output_mean_power": 0.00104915576823599}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 6.799409328431789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020213626575120905, "output_power": 0.00020394186547868063, "reference_active": true, "delta_db": 0.03862149206036372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0065691390851235e-05, "output_power": 3.50363971900036e-05, "reference_active": true, "delta_db": 0.6644824574101922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1255469357025214e-06, "output_power": 1.8138287231557014e-06, "reference_active": true, "delta_db": -0.6887442442103155}], "active_band_count": 3, "active_band_mae_db": 0.46394939789362377}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 5.7697319934351994e-05, "log_rms_mae_db": 0.834294158513231, "rms_envelope_nrmse": 0.1181957609305471, "envelope_correlation": 0.9899050129407128, "rms_gain_error_db": -0.2807692889809087, "reference_mean_power": 0.0001838054244420426, "output_mean_power": 0.00017229845977527317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.2570064040799192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.4129018402509514e-05, "output_power": 3.9600809299283554e-05, "reference_active": true, "delta_db": -0.4702020574453537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0833298446779821e-05, "output_power": 1.328087934452825e-05, "reference_active": true, "delta_db": 0.8846612349449905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.606324451113381e-06, "output_power": 2.7696411413319685e-06, "reference_active": true, "delta_db": -2.209310229166865}], "active_band_count": 3, "active_band_mae_db": 1.1880578405190698}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 3.902816175045304e-05, "log_rms_mae_db": 0.5463964772889764, "rms_envelope_nrmse": 0.06842205083142551, "envelope_correlation": 0.9969848763704561, "rms_gain_error_db": -0.28076929363799263, "reference_mean_power": 0.00018518052639832005, "output_mean_power": 0.00017358747458865847}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.1869756093173816e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2060123324004214e-05, "output_power": 3.836221397152537e-05, "reference_active": true, "delta_db": -0.39966878387552807}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.372206568913873e-06, "output_power": 5.927065974505319e-06, "reference_active": true, "delta_db": 0.42687057869630113}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0554113039123233e-07, "output_power": 2.145375219461637e-07, "reference_active": true, "delta_db": -2.7653164744650742}], "active_band_count": 3, "active_band_mae_db": 1.1972852790123012}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9550916446596927, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6f9fa162a70f8d7e7ad3dee169bb968ed0503a509c64a9a8cac70d24964dbf4f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.992472597254147, "standard_deviation": 0.0013340609480601462, "minimum": 0.9905638492133411, "maximum": 0.9940268972622489, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.7096938086788298, "standard_deviation": 0.015984773499636656, "minimum": 0.6826049223142533, "maximum": 0.72702432893913, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1013369061927964, "standard_deviation": 0.5682524885282157, "minimum": 0.3895259390098846, "maximum": 2.181623980651182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10174415460035639, "standard_deviation": 0.009079571473534625, "minimum": 0.09180773469945558, "maximum": 0.11578909858877508, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997588812613568, "standard_deviation": 0.0007833808273088301, "minimum": 0.9959471432551771, "maximum": 0.9986356508481063, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.496541346144028, "standard_deviation": 0.023506787536758948, "minimum": 0.45141486219845706, "maximum": 0.5221704560996293, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5484453197091206, "standard_deviation": 0.7805771476177861, "minimum": 0.46394939789362377, "maximum": 2.8444149422561207, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05913548660142899, "standard_deviation": 0.009453446050114432, "minimum": 0.04184613811665808, "maximum": 0.07409347526814335, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9885319370424094, "standard_deviation": 0.0017609786840656657, "minimum": 0.9852894263645807, "maximum": 0.9908460975670778, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 0.8698642275741324, "standard_deviation": 0.02188854381782364, "minimum": 0.834294158513231, "maximum": 0.8918426252960417, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1898360594916746, "standard_deviation": 0.4650315600471722, "minimum": 0.40097045256421193, "maximum": 1.8844438531994376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12691805543776327, "standard_deviation": 0.008291283895273935, "minimum": 0.11736212050135038, "maximum": 0.14193000310661424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9960522704224486, "standard_deviation": 0.0011970387992135737, "minimum": 0.9939002446554186, "maximum": 0.9971907958871145, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6091343072861455, "standard_deviation": 0.03245393098238821, "minimum": 0.5463964772889764, "maximum": 0.649939725695184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5254340256634038, "standard_deviation": 0.5340885803899044, "minimum": 0.6167336990562203, "maximum": 2.334159249979875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07894684756522544, "standard_deviation": 0.008539727180268755, "minimum": 0.06842205083142551, "maximum": 0.09350120107655821, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.996326889033861, "standard_deviation": 0.00100529004390008, "minimum": 0.9945712253272703, "maximum": 0.9976564026452757, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.7844487191029744, "standard_deviation": 0.02646790839398898, "minimum": 0.7500529480142853, "maximum": 0.8355266379159867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.45535105265457143, "standard_deviation": 0.2197020574883449, "minimum": 0.12002145581361684, "maximum": 0.8466017191522704, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06751668010782952, "standard_deviation": 0.009202319931641414, "minimum": 0.05431979385709911, "maximum": 0.08237581487376489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986884282000925, "standard_deviation": 0.0004119616175513933, "minimum": 0.9980938229340005, "maximum": 0.9992747085970578, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5939683482356166, "standard_deviation": 0.03606995974219518, "minimum": 0.5327717678226511, "maximum": 0.6573664206524635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.47891198597464146, "standard_deviation": 0.22456922176206032, "minimum": 0.14122637763594945, "maximum": 0.8557079671243882, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03963816405879652, "standard_deviation": 0.006571782728123728, "minimum": 0.029735511441379633, "maximum": 0.048189738717428786, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14832861154636467, "standard_deviation": 0.023006233733110545, "minimum": 0.1106284788973232, "maximum": 0.17688346054825135, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.558087700451702, "standard_deviation": 0.008073366135210517, "minimum": 4.546318347905953, "maximum": 4.568363997388702, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5203307578542661, "standard_deviation": 0.01665298484665668, "minimum": 0.5020650814888182, "maximum": 0.5536485292828371, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07708577659581914, "standard_deviation": 0.04455562670680724, "minimum": -0.12726398000789574, "maximum": -0.0012897173853025326, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968418632199756, "standard_deviation": 0.0005075143716698256, "minimum": 0.9959777319859345, "maximum": 0.997401607088933, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5795878270024738, "standard_deviation": 0.01495358325472511, "minimum": 0.5466159122654686, "maximum": 0.5935574058355904, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.760180454301955, "standard_deviation": 0.3207042424795225, "minimum": 0.29307322775007655, "maximum": 1.194582531509594, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.059653467073601765, "standard_deviation": 0.004451885616658745, "minimum": 0.05468525503196416, "maximum": 0.0667986149124771, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990129573629769, "standard_deviation": 0.00020546843749606552, "minimum": 0.9987099311518111, "maximum": 0.9993494864788424, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4083270072607719, "standard_deviation": 0.023237987430242676, "minimum": 0.37516142274139386, "maximum": 0.4372381920110589, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.065181522707497, "standard_deviation": 0.4241957588072996, "minimum": 0.5157255827789292, "maximum": 1.5160672529290748, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03353531312934066, "standard_deviation": 0.003615276057684283, "minimum": 0.026237211772671246, "maximum": 0.0389398657505712, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-121726-0012", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.757395029067993, "predicted_mos": 2.757395029067993, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "24333d0434a474a5abc3874a8d296bfad4d1306d8af7d54c9f3c73d763a45e01", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9725114996545017, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6070888848669657, "全帯域_10ms_rms_envelope_nrmse": 0.06738607364848416, "全帯域_10ms_envelope_correlation": 0.995959228092371, "全帯域_10ms_modulation_mae_db": 1.5330614401876312, "全帯域_40ms_log_rms_mae_db": 0.4338823171188842, "全帯域_40ms_rms_envelope_nrmse": 0.0363171021971086, "全帯域_40ms_envelope_correlation": 0.9988398980652817, "全帯域_40ms_modulation_mae_db": 2.1119734568465716, "300–1000 Hz_10ms_log_rms_mae_db": 0.8116513972729258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07002476228945112, "300–1000 Hz_10ms_envelope_correlation": 0.9961108982651271, "300–1000 Hz_10ms_modulation_mae_db": 0.7094195703408844, "300–1000 Hz_40ms_log_rms_mae_db": 0.585206072382715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041485840234583744, "300–1000 Hz_40ms_envelope_correlation": 0.9986191812419719, "300–1000 Hz_40ms_modulation_mae_db": 0.9449779366004173, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7076883450675112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11062543985154813, "1000–3000 Hz_10ms_envelope_correlation": 0.9909121189631809, "1000–3000 Hz_10ms_modulation_mae_db": 1.7194662123948437, "1000–3000 Hz_40ms_log_rms_mae_db": 0.4920442999443714, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06109939707857381, "1000–3000 Hz_40ms_envelope_correlation": 0.9972353715160952, "1000–3000 Hz_40ms_modulation_mae_db": 2.56884517590646, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9448368891377675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1372124602788989, "3000–6000 Hz_10ms_envelope_correlation": 0.9866250467137456, "3000–6000 Hz_10ms_modulation_mae_db": 1.5699820431458005, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6955923515001258, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08803675786785899, "3000–6000 Hz_40ms_envelope_correlation": 0.9951160911540715, "3000–6000 Hz_40ms_modulation_mae_db": 2.0238282910904437, "mr_stft_db": 4.677718184984834, "band_mae_db": 0.15426358966355833, "rms_gain_error_db": -0.09085807192494723, "output_peak": 0.502101428464692}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5847655438166446, "全帯域_10ms_rms_envelope_nrmse": 0.05119320188002822, "全帯域_10ms_envelope_correlation": 0.9976510486393206, "全帯域_10ms_modulation_mae_db": 0.6647650915556387, "全帯域_40ms_log_rms_mae_db": 0.4211640126223313, "全帯域_40ms_rms_envelope_nrmse": 0.025591338364267507, "全帯域_40ms_envelope_correlation": 0.9994015733024747, "全帯域_40ms_modulation_mae_db": 0.8639264551195992, "300–1000 Hz_10ms_log_rms_mae_db": 0.8119651159063191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.056075801455504365, "300–1000 Hz_10ms_envelope_correlation": 0.9974833293903339, "300–1000 Hz_10ms_modulation_mae_db": 0.3223007181729331, "300–1000 Hz_40ms_log_rms_mae_db": 0.6294678008247306, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.025983539202296412, "300–1000 Hz_40ms_envelope_correlation": 0.9994362383497714, "300–1000 Hz_40ms_modulation_mae_db": 0.5446973023088364, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7455499306130887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08653884683525553, "1000–3000 Hz_10ms_envelope_correlation": 0.994437475281282, "1000–3000 Hz_10ms_modulation_mae_db": 0.9657406354640687, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5482829484173654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0452719602472532, "1000–3000 Hz_40ms_envelope_correlation": 0.9984585059935229, "1000–3000 Hz_40ms_modulation_mae_db": 1.11038937355553, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9445257959692935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12615880329832246, "3000–6000 Hz_10ms_envelope_correlation": 0.989103152271886, "3000–6000 Hz_10ms_modulation_mae_db": 0.6929748298786036, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7076151830617766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08025918848339611, "3000–6000 Hz_40ms_envelope_correlation": 0.9964103194801294, "3000–6000 Hz_40ms_modulation_mae_db": 1.1045583422316885, "mr_stft_db": 4.659820920747381, "band_mae_db": 0.10203813192259521, "rms_gain_error_db": -0.037685021511624575, "output_peak": 0.517683118181582}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6066599263442338, "全帯域_10ms_rms_envelope_nrmse": 0.05204710203512268, "全帯域_10ms_envelope_correlation": 0.9975591954143168, "全帯域_10ms_modulation_mae_db": 0.668868372476061, "全帯域_40ms_log_rms_mae_db": 0.4412178826450348, "全帯域_40ms_rms_envelope_nrmse": 0.02568345935708335, "全帯域_40ms_envelope_correlation": 0.9993845345659126, "全帯域_40ms_modulation_mae_db": 1.0633927151528721, "300–1000 Hz_10ms_log_rms_mae_db": 0.860953548321306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06729840178880006, "300–1000 Hz_10ms_envelope_correlation": 0.9964277599645085, "300–1000 Hz_10ms_modulation_mae_db": 0.12948180768908923, "300–1000 Hz_40ms_log_rms_mae_db": 0.6173448217816868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04104671796599697, "300–1000 Hz_40ms_envelope_correlation": 0.998626175207677, "300–1000 Hz_40ms_modulation_mae_db": 0.40131964745491455, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7895713987611459, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09583094990336177, "1000–3000 Hz_10ms_envelope_correlation": 0.9931948260335981, "1000–3000 Hz_10ms_modulation_mae_db": 1.4658619315028032, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5678775111322069, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0549383961890313, "1000–3000 Hz_40ms_envelope_correlation": 0.9977547980679892, "1000–3000 Hz_40ms_modulation_mae_db": 1.9452183805256, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9326130331174988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14487829518620357, "3000–6000 Hz_10ms_envelope_correlation": 0.9845914505969886, "3000–6000 Hz_10ms_modulation_mae_db": 1.2355560508870214, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6385896416947416, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07395125460765523, "3000–6000 Hz_40ms_envelope_correlation": 0.9965675153149935, "3000–6000 Hz_40ms_modulation_mae_db": 1.4225352798061097, "mr_stft_db": 4.679504640004628, "band_mae_db": 0.16868196373797603, "rms_gain_error_db": -0.03075147331438025, "output_peak": 0.502508797927986}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5737922349110198, "全帯域_10ms_rms_envelope_nrmse": 0.06761669462325236, "全帯域_10ms_envelope_correlation": 0.9959336733483022, "全帯域_10ms_modulation_mae_db": 1.9408742335269489, "全帯域_40ms_log_rms_mae_db": 0.3955557116095783, "全帯域_40ms_rms_envelope_nrmse": 0.04105563503405043, "全帯域_40ms_envelope_correlation": 0.9984889644324157, "全帯域_40ms_modulation_mae_db": 2.494224978554914, "300–1000 Hz_10ms_log_rms_mae_db": 0.8428112806441598, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06373572695828435, "300–1000 Hz_10ms_envelope_correlation": 0.9967585909502966, "300–1000 Hz_10ms_modulation_mae_db": 1.1162467846017083, "300–1000 Hz_40ms_log_rms_mae_db": 0.5699176230421126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035287406400861955, "300–1000 Hz_40ms_envelope_correlation": 0.9989733560842763, "300–1000 Hz_40ms_modulation_mae_db": 1.6899474941155443, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7712670656485063, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12459882252581089, "1000–3000 Hz_10ms_envelope_correlation": 0.9886043380949311, "1000–3000 Hz_10ms_modulation_mae_db": 2.7134421374684528, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5848402450528805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08122928238458274, "1000–3000 Hz_40ms_envelope_correlation": 0.9952293763707056, "1000–3000 Hz_40ms_modulation_mae_db": 3.251264177139356, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9112522947861448, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12031261973154787, "3000–6000 Hz_10ms_envelope_correlation": 0.9892203318792819, "3000–6000 Hz_10ms_modulation_mae_db": 0.9176205495718476, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6473749379768009, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06121683260734159, "3000–6000 Hz_40ms_envelope_correlation": 0.9973519155124525, "3000–6000 Hz_40ms_modulation_mae_db": 1.164990547155263, "mr_stft_db": 4.671633608019317, "band_mae_db": 0.14684307626634127, "rms_gain_error_db": -0.09091392695152381, "output_peak": 0.5189181998922562}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5679162098205048, "全帯域_10ms_rms_envelope_nrmse": 0.055967986539993776, "全帯域_10ms_envelope_correlation": 0.9971738508587403, "全帯域_10ms_modulation_mae_db": 0.133794720259063, "全帯域_40ms_log_rms_mae_db": 0.37800288658170533, "全帯域_40ms_rms_envelope_nrmse": 0.031404640820470016, "全帯域_40ms_envelope_correlation": 0.9990659846938746, "全帯域_40ms_modulation_mae_db": 0.2789091214337005, "300–1000 Hz_10ms_log_rms_mae_db": 0.8247864343169012, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07124514620997346, "300–1000 Hz_10ms_envelope_correlation": 0.9959396734531929, "300–1000 Hz_10ms_modulation_mae_db": 0.4853441997301364, "300–1000 Hz_40ms_log_rms_mae_db": 0.6227161735407579, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04061846250654068, "300–1000 Hz_40ms_envelope_correlation": 0.9986215045393361, "300–1000 Hz_40ms_modulation_mae_db": 0.9119199697756399, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7133622918582966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09878026213620143, "1000–3000 Hz_10ms_envelope_correlation": 0.9927246382788147, "1000–3000 Hz_10ms_modulation_mae_db": 0.23543739365789704, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5047747600916577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05784947764762054, "1000–3000 Hz_40ms_envelope_correlation": 0.9973950769914328, "1000–3000 Hz_40ms_modulation_mae_db": 0.35280720005440175, "3000–6000 Hz_10ms_log_rms_mae_db": 0.919597299093702, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1203167078587867, "3000–6000 Hz_10ms_envelope_correlation": 0.9891609174151847, "3000–6000 Hz_10ms_modulation_mae_db": 0.5027686509817342, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6636675188422937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08047039285875648, "3000–6000 Hz_40ms_envelope_correlation": 0.9950179327801556, "3000–6000 Hz_40ms_modulation_mae_db": 0.6433000290311183, "mr_stft_db": 4.662441069294322, "band_mae_db": 0.07890015813928554, "rms_gain_error_db": -0.0019260326334690344, "output_peak": 0.5478145739102702}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5851007965124463, "全帯域_10ms_rms_envelope_nrmse": 0.054772270059490265, "全帯域_10ms_envelope_correlation": 0.9973026806669774, "全帯域_10ms_modulation_mae_db": 0.20165934388365828, "全帯域_40ms_log_rms_mae_db": 0.396566765623106, "全帯域_40ms_rms_envelope_nrmse": 0.03340576561981443, "全帯域_40ms_envelope_correlation": 0.9989601012115228, "全帯域_40ms_modulation_mae_db": 0.3532936268583186, "300–1000 Hz_10ms_log_rms_mae_db": 0.8630635005062697, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0686578909889032, "300–1000 Hz_10ms_envelope_correlation": 0.9962193687646971, "300–1000 Hz_10ms_modulation_mae_db": 0.8533418284958673, "300–1000 Hz_40ms_log_rms_mae_db": 0.6543724679756747, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03725424775069739, "300–1000 Hz_40ms_envelope_correlation": 0.9988372672071033, "300–1000 Hz_40ms_modulation_mae_db": 1.148235713506095, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7405712305189761, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10064698223053782, "1000–3000 Hz_10ms_envelope_correlation": 0.9924591016100449, "1000–3000 Hz_10ms_modulation_mae_db": 0.8924669102512173, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5078349683524874, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06379130346735666, "1000–3000 Hz_40ms_envelope_correlation": 0.9968314822332011, "1000–3000 Hz_40ms_modulation_mae_db": 0.8522813038035926, "3000–6000 Hz_10ms_log_rms_mae_db": 0.960022608290583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13196682788960665, "3000–6000 Hz_10ms_envelope_correlation": 0.9877032160483277, "3000–6000 Hz_10ms_modulation_mae_db": 1.0523867521973334, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6963206947476595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07837901957567991, "3000–6000 Hz_40ms_envelope_correlation": 0.9964037487037385, "3000–6000 Hz_40ms_modulation_mae_db": 1.3563207854354913, "mr_stft_db": 4.665901928869929, "band_mae_db": 0.1376749844387151, "rms_gain_error_db": -0.043386431726114665, "output_peak": 0.5426554278991247}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6022306238809403, "全帯域_10ms_rms_envelope_nrmse": 0.06135184349856776, "全帯域_10ms_envelope_correlation": 0.9966561749162673, "全帯域_10ms_modulation_mae_db": 1.3510719281013033, "全帯域_40ms_log_rms_mae_db": 0.43353751725036227, "全帯域_40ms_rms_envelope_nrmse": 0.03702365765420331, "全帯域_40ms_envelope_correlation": 0.9987722572747545, "全帯域_40ms_modulation_mae_db": 1.7468676623907493, "300–1000 Hz_10ms_log_rms_mae_db": 0.8663871527398367, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07034389882735917, "300–1000 Hz_10ms_envelope_correlation": 0.9960402877575978, "300–1000 Hz_10ms_modulation_mae_db": 0.7360916080111419, "300–1000 Hz_40ms_log_rms_mae_db": 0.6571907421281409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04154473080748965, "300–1000 Hz_40ms_envelope_correlation": 0.9985630750039658, "300–1000 Hz_40ms_modulation_mae_db": 1.1080267474652339, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7421197551993258, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09415418244329903, "1000–3000 Hz_10ms_envelope_correlation": 0.9935743785328578, "1000–3000 Hz_10ms_modulation_mae_db": 1.9623134871440329, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5342727787278282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06371281539282224, "1000–3000 Hz_40ms_envelope_correlation": 0.9970990247924435, "1000–3000 Hz_40ms_modulation_mae_db": 2.396555084246142, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8983160294285693, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13307566871995807, "3000–6000 Hz_10ms_envelope_correlation": 0.986894405234551, "3000–6000 Hz_10ms_modulation_mae_db": 1.4921264391614697, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6140361389805298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07008439175727983, "3000–6000 Hz_40ms_envelope_correlation": 0.9966700540247324, "3000–6000 Hz_40ms_modulation_mae_db": 1.7654382926078862, "mr_stft_db": 4.666804168230009, "band_mae_db": 0.10508079583535707, "rms_gain_error_db": -0.08149191106502554, "output_peak": 0.515699165328269}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5918942968123342, "全帯域_10ms_rms_envelope_nrmse": 0.0608601568725809, "全帯域_10ms_envelope_correlation": 0.9966587845598458, "全帯域_10ms_modulation_mae_db": 0.13333273544736635, "全帯域_40ms_log_rms_mae_db": 0.4126733392284605, "全帯域_40ms_rms_envelope_nrmse": 0.03156900337496611, "全帯域_40ms_envelope_correlation": 0.9990652727150579, "全帯域_40ms_modulation_mae_db": 0.21361170925390036, "300–1000 Hz_10ms_log_rms_mae_db": 0.7848529394716552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06683103993466644, "300–1000 Hz_10ms_envelope_correlation": 0.9965609154645145, "300–1000 Hz_10ms_modulation_mae_db": 0.24934745640334366, "300–1000 Hz_40ms_log_rms_mae_db": 0.5412864581890047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03888142906113449, "300–1000 Hz_40ms_envelope_correlation": 0.9988459084054191, "300–1000 Hz_40ms_modulation_mae_db": 0.16145912880343338, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7441485601525438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11003599158291041, "1000–3000 Hz_10ms_envelope_correlation": 0.9910516937576643, "1000–3000 Hz_10ms_modulation_mae_db": 0.22955402675207268, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5375965672089785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06499383688208497, "1000–3000 Hz_40ms_envelope_correlation": 0.996903496739241, "1000–3000 Hz_40ms_modulation_mae_db": 0.2716299474930589, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9368138473678778, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12245041904753205, "3000–6000 Hz_10ms_envelope_correlation": 0.9894441819088248, "3000–6000 Hz_10ms_modulation_mae_db": 1.0856314177374988, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6586156817897983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06891628946147063, "3000–6000 Hz_40ms_envelope_correlation": 0.9973596880766904, "3000–6000 Hz_40ms_modulation_mae_db": 1.655858277811829, "mr_stft_db": 4.682315671020753, "band_mae_db": 0.17941099662819726, "rms_gain_error_db": -0.033350129964347996, "output_peak": 0.5140236159505561}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64720, "output_samples": 64720, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.870150559122728, "spectral_convergence": 0.23750375418101574, "spectral_convergence_status": "ok", "amplitude_floor": 1.1807658352589429e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.480930129224578, "spectral_convergence": 0.22998495895303414, "spectral_convergence_status": "ok", "amplitude_floor": 7.24953185483845e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.682073866607198, "spectral_convergence": 0.23509682739140664, "spectral_convergence_status": "ok", "amplitude_floor": 4.6531510695198173e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031402968364977683, "output_power": 0.00032276994149940766, "delta_db": 0.11922381828711778, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005546801738700932, "output_power": 0.000545913186813234, "delta_db": -0.06919057975980364, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010899709362407707, "output_power": 0.0010744613743228712, "delta_db": -0.06224110048235472, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008204698714335104, "output_power": 0.0007920505949892988, "delta_db": -0.15309713357736188, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0004732551323003661, "output_power": 0.0004560252735999752, "delta_db": -0.16106419867396177, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00010423546715106559, "output_power": 9.592659882538957e-05, "delta_db": -0.36076470720075005, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15426358966355833, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09085807192494723, "pitch": null, "output_rms_dbfs": -24.831808779521825, "output_peak": 0.502101428464692, "reference_rms_dbfs": -24.740950707596877, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.677718184984834}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64720, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00017261805365043103, "log_rms_mae_db": 0.6070888848669657, "rms_envelope_nrmse": 0.06738607364848416, "envelope_correlation": 0.995959228092371, "rms_gain_error_db": -0.09085808089281737, "reference_mean_power": 0.003360795516302892, "output_mean_power": 0.0032912152072728513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.9828935159721717e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046042689591210996, "output_power": 0.0004510205182079092, "reference_active": true, "delta_db": -0.08964385461388283}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011507848708673394, "output_power": 9.307278021832295e-05, "reference_active": true, "delta_db": -0.9217145656551468}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.734946402574847e-05, "output_power": 1.6349398668820682e-05, "reference_active": true, "delta_db": -3.587825900293864}], "active_band_count": 3, "active_band_mae_db": 1.5330614401876312}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00013379439656886598, "log_rms_mae_db": 0.4338823171188842, "rms_envelope_nrmse": 0.0363171021971086, "envelope_correlation": 0.9988398980652817, "rms_gain_error_db": -0.09085809154180487, "reference_mean_power": 0.0033859385941668236, "output_mean_power": 0.003315837726702269}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.884361456130501e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004231625539622664, "output_power": 0.00041137430756257534, "reference_active": true, "delta_db": -0.12270065329605825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.583961694519604e-05, "output_power": 4.284524569241247e-05, "reference_active": true, "delta_db": -1.150397928051501}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3554104634271355e-06, "output_power": 1.0458357313440806e-06, "reference_active": true, "delta_db": -5.062821789192156}], "active_band_count": 3, "active_band_mae_db": 2.1119734568465716}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00011786303332041996, "log_rms_mae_db": 0.8116513972729258, "rms_envelope_nrmse": 0.07002476228945112, "envelope_correlation": 0.9961108982651271, "rms_gain_error_db": -0.08747433839898254, "reference_mean_power": 0.0013229704572187174, "output_mean_power": 0.0012965901331947912}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 9.694412540428036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002357826301518165, "output_power": 0.00022849955444850779, "reference_active": true, "delta_db": -0.13626450363077142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.91091063998743e-05, "output_power": 5.267660764897826e-05, "reference_active": true, "delta_db": -0.5003659500665205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.351021856959089e-05, "output_power": 9.58295722038663e-06, "reference_active": true, "delta_db": -1.4916282573253614}], "active_band_count": 3, "active_band_mae_db": 0.7094195703408844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010411406065660645, "log_rms_mae_db": 0.585206072382715, "rms_envelope_nrmse": 0.041485840234583744, "envelope_correlation": 0.9986191812419719, "rms_gain_error_db": -0.08747434468791808, "reference_mean_power": 0.0013328679887454193, "output_mean_power": 0.0013062903038584168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 9.215200160073525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002182221456522984, "output_power": 0.00021066546607346245, "reference_active": true, "delta_db": -0.15305473122673963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.729502442376393e-05, "output_power": 2.4582641716484827e-05, "reference_active": true, "delta_db": -0.4545493574577675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1926400570958608e-06, "output_power": 7.141285890707083e-07, "reference_active": true, "delta_db": -2.2273297211167447}], "active_band_count": 3, "active_band_mae_db": 0.9449779366004173}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00014037287410750346, "log_rms_mae_db": 0.7076883450675112, "rms_envelope_nrmse": 0.11062543985154813, "envelope_correlation": 0.9909121189631809, "rms_gain_error_db": -0.13545583227082159, "reference_mean_power": 0.001043588584426717, "output_mean_power": 0.0010115415734356819}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 7.131098310180307e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021758763787569653, "output_power": 0.00021194805197060848, "reference_active": true, "delta_db": -0.11404788228681988}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8027856042619025e-05, "output_power": 3.9650742301948595e-05, "reference_active": true, "delta_db": -1.6538520280525362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3022954991601597e-05, "output_power": 1.0546568841219702e-05, "reference_active": true, "delta_db": -3.3904987268451747}], "active_band_count": 3, "active_band_mae_db": 1.7194662123948437}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010015064346906524, "log_rms_mae_db": 0.4920442999443714, "rms_envelope_nrmse": 0.06109939707857381, "envelope_correlation": 0.9972353715160952, "rms_gain_error_db": -0.13545583668111424, "reference_mean_power": 0.0010513959774729573, "output_mean_power": 0.0010191092123350092}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 6.799409328431789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020213626575120905, "output_power": 0.00019379869107430582, "reference_active": true, "delta_db": -0.1829339883943869}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0065691390851235e-05, "output_power": 1.8183500228410037e-05, "reference_active": true, "delta_db": -2.1839370899249593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1255469357025214e-06, "output_power": 6.215903369409233e-07, "reference_active": true, "delta_db": -5.3396644494000345}], "active_band_count": 3, "active_band_mae_db": 2.56884517590646}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 5.7697319934351994e-05, "log_rms_mae_db": 0.9448368891377675, "rms_envelope_nrmse": 0.1372124602788989, "envelope_correlation": 0.9866250467137456, "rms_gain_error_db": -0.3795303835099857, "reference_mean_power": 0.0001838054244420426, "output_mean_power": 0.00016842450736447948}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.2570064040799192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.4129018402509514e-05, "output_power": 4.0345648923078264e-05, "reference_active": true, "delta_db": -0.389275621001995}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0833298446779821e-05, "output_power": 8.484711458510092e-06, "reference_active": true, "delta_db": -1.0612362994489186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.606324451113381e-06, "output_power": 2.1747588346410666e-06, "reference_active": true, "delta_db": -3.2594342089864883}], "active_band_count": 3, "active_band_mae_db": 1.5699820431458005}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 3.902816175045304e-05, "log_rms_mae_db": 0.6955923515001258, "rms_envelope_nrmse": 0.08803675786785899, "envelope_correlation": 0.9951160911540715, "rms_gain_error_db": -0.3795303869956246, "reference_mean_power": 0.00018518052639832005, "output_mean_power": 0.00016968454006064624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.1869756093173816e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2060123324004214e-05, "output_power": 3.7959855263069025e-05, "reference_active": true, "delta_db": -0.44545993902744974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.372206568913873e-06, "output_power": 3.7465869344029934e-06, "reference_active": true, "delta_db": -1.565168892947586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0554113039123233e-07, "output_power": 1.5920228736235138e-07, "reference_active": true, "delta_db": -4.0608560412962955}], "active_band_count": 3, "active_band_mae_db": 2.0238282910904437}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9510658055147053, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0ed6f9fd4a00dabb09b8fde306de8211e9301c547820bb425c62c5221f12a995", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9921198213190467, "standard_deviation": 0.0018547957203920386, "minimum": 0.9886043380949311, "maximum": 0.994437475281282, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.7442848222274243, "standard_deviation": 0.026977551102242023, "minimum": 0.7076883450675112, "maximum": 0.7895713987611459, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2730353418294236, "standard_deviation": 0.8611523707094626, "minimum": 0.22955402675207268, "maximum": 2.7134421374684528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10265143468861562, "standard_deviation": 0.011935217240740837, "minimum": 0.08653884683525553, "maximum": 0.12459882252581089, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997113391588079, "standard_deviation": 0.0009249099787510335, "minimum": 0.9952293763707056, "maximum": 0.9984585059935229, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.534690509865972, "standard_deviation": 0.03217056876343267, "minimum": 0.4920442999443714, "maximum": 0.5848402450528805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5936238303405177, "standard_deviation": 1.104377234757752, "minimum": 0.2716299474930589, "maximum": 3.251264177139356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06161080866116568, "standard_deviation": 0.010221759630255643, "minimum": 0.0452719602472532, "maximum": 0.08122928238458274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9878428377585988, "standard_deviation": 0.0017221693865482049, "minimum": 0.9845914505969886, "maximum": 0.9894441819088248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 0.9309972246489295, "standard_deviation": 0.020147160262581294, "minimum": 0.8983160294285693, "maximum": 0.960022608290583, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0686308416951635, "standard_deviation": 0.3670811568854583, "minimum": 0.5027686509817342, "maximum": 1.5699820431458005, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12954647525135704, "standard_deviation": 0.008819040854717517, "minimum": 0.12031261973154787, "maximum": 0.14487829518620357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9963621581308705, "standard_deviation": 0.0008844151511607596, "minimum": 0.9950179327801556, "maximum": 0.9973596880766904, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6652265185742158, "standard_deviation": 0.03247091275070373, "minimum": 0.6140361389805298, "maximum": 0.7076151830617766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3921037306462287, "standard_deviation": 0.43180638964912044, "minimum": 0.6433000290311183, "maximum": 2.0238282910904437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07516426590242986, "standard_deviation": 0.00837507810472698, "minimum": 0.06121683260734159, "maximum": 0.08803675786785899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9964426030012835, "standard_deviation": 0.000503109644042689, "minimum": 0.9959396734531929, "maximum": 0.9974833293903339, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.8333089211474216, "standard_deviation": 0.02971109929729201, "minimum": 0.7848529394716552, "maximum": 0.8663871527398367, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.575196746680638, "standard_deviation": 0.3361275695559654, "minimum": 0.12948180768908923, "maximum": 1.1162467846017083, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06677658355661777, "standard_deviation": 0.0049398175118797695, "minimum": 0.056075801455504365, "maximum": 0.07124514620997346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9988153382549401, "standard_deviation": 0.00028967234309432724, "minimum": 0.9985630750039658, "maximum": 0.9994362383497714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6096877699831029, "standard_deviation": 0.04094008081358642, "minimum": 0.5412864581890047, "maximum": 0.6571907421281409, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8638229925037644, "standard_deviation": 0.48417422388144343, "minimum": 0.16145912880343338, "maximum": 1.6899474941155443, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03776279674120016, "standard_deviation": 0.005257531598399701, "minimum": 0.025983539202296412, "maximum": 0.04154473080748965, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13411171207900324, "standard_deviation": 0.03535096386851235, "minimum": 0.07890015813928554, "maximum": 0.17941099662819726, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.670767523896396, "standard_deviation": 0.008343507835532046, "minimum": 4.659820920747381, "maximum": 4.682315671020753, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.520175540944342, "standard_deviation": 0.016791690160226816, "minimum": 0.502101428464692, "maximum": 0.5478145739102702, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05129537488642914, "standard_deviation": 0.03267999874188455, "minimum": -0.09091392695152381, "maximum": -0.0019260326334690344, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968618295620177, "standard_deviation": 0.000671945719170097, "minimum": 0.9959336733483022, "maximum": 0.9976510486393206, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5899310646206362, "standard_deviation": 0.01475760724047666, "minimum": 0.5679162098205048, "maximum": 0.6070888848669657, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8284284831797087, "standard_deviation": 0.6987701618712817, "minimum": 0.13333273544736635, "maximum": 1.9408742335269489, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05889941614469002, "standard_deviation": 0.006434724603636244, "minimum": 0.05119320188002822, "maximum": 0.06761669462325236, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989973232826618, "standard_deviation": 0.0003065787666988217, "minimum": 0.9984889644324157, "maximum": 0.9994015733024747, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41407505408493284, "standard_deviation": 0.02239506929231025, "minimum": 0.37800288658170533, "maximum": 0.4412178826450348, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1407749657013282, "standard_deviation": 0.8821731455117784, "minimum": 0.21361170925390036, "maximum": 2.494224978554914, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03275632530274547, "standard_deviation": 0.005408839683392277, "minimum": 0.025591338364267507, "maximum": 0.04105563503405043, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-121726-0012", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7921760082244873, "predicted_mos": 2.7921760082244873, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "662f9eea44f388bea4a5b07def62cabca31b28cdc4e773a1b2aecfad1622ac4d", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9162212000228465, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5757574712303629, "全帯域_10ms_rms_envelope_nrmse": 0.055557191627919236, "全帯域_10ms_envelope_correlation": 0.9972162207839168, "全帯域_10ms_modulation_mae_db": 0.3591032429309591, "全帯域_40ms_log_rms_mae_db": 0.39479121675344414, "全帯域_40ms_rms_envelope_nrmse": 0.030212023296151174, "全帯域_40ms_envelope_correlation": 0.9991419785860847, "全帯域_40ms_modulation_mae_db": 0.5364054947456407, "300–1000 Hz_10ms_log_rms_mae_db": 0.8096574612766231, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06820204773988031, "300–1000 Hz_10ms_envelope_correlation": 0.9963003128581207, "300–1000 Hz_10ms_modulation_mae_db": 0.5909457025317765, "300–1000 Hz_40ms_log_rms_mae_db": 0.6049847078972975, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03909916630655077, "300–1000 Hz_40ms_envelope_correlation": 0.9987653843533594, "300–1000 Hz_40ms_modulation_mae_db": 1.0100498002985392, "1000–3000 Hz_10ms_log_rms_mae_db": 0.718558585033649, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10314014268789853, "1000–3000 Hz_10ms_envelope_correlation": 0.9922687527516062, "1000–3000 Hz_10ms_modulation_mae_db": 0.7298874073486051, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5153594838487323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06109772406705248, "1000–3000 Hz_40ms_envelope_correlation": 0.9972248988493945, "1000–3000 Hz_40ms_modulation_mae_db": 0.7887219776231161, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9689243629967214, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1461836688749034, "3000–6000 Hz_10ms_envelope_correlation": 0.9850715842450547, "3000–6000 Hz_10ms_modulation_mae_db": 1.2300392680458623, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6926316340315226, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08863868310523008, "3000–6000 Hz_40ms_envelope_correlation": 0.9955989807924449, "3000–6000 Hz_40ms_modulation_mae_db": 1.6668917768183267, "mr_stft_db": 4.63662884370936, "band_mae_db": 0.1473235424234561, "rms_gain_error_db": -0.028762765052632956, "output_peak": 0.5420700208940737}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5766926841120223, "全帯域_10ms_rms_envelope_nrmse": 0.050780356533712816, "全帯域_10ms_envelope_correlation": 0.9976737582428287, "全帯域_10ms_modulation_mae_db": 0.4376721131029711, "全帯域_40ms_log_rms_mae_db": 0.4052416813208693, "全帯域_40ms_rms_envelope_nrmse": 0.0270725828359558, "全帯域_40ms_envelope_correlation": 0.9993053786829624, "全帯域_40ms_modulation_mae_db": 0.5725056203620552, "300–1000 Hz_10ms_log_rms_mae_db": 0.8639453179046509, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07270122613213631, "300–1000 Hz_10ms_envelope_correlation": 0.9957967650921595, "300–1000 Hz_10ms_modulation_mae_db": 0.18908324205205304, "300–1000 Hz_40ms_log_rms_mae_db": 0.6729206230630123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04653033900511973, "300–1000 Hz_40ms_envelope_correlation": 0.9981996072870923, "300–1000 Hz_40ms_modulation_mae_db": 0.1998688060585164, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7383811610445634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09021328905467124, "1000–3000 Hz_10ms_envelope_correlation": 0.9939499491139432, "1000–3000 Hz_10ms_modulation_mae_db": 0.6208178313004019, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5027257363003841, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048612308448074855, "1000–3000 Hz_40ms_envelope_correlation": 0.9982231418487292, "1000–3000 Hz_40ms_modulation_mae_db": 0.9275765565150467, "3000–6000 Hz_10ms_log_rms_mae_db": 0.942137912876835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13261712011446927, "3000–6000 Hz_10ms_envelope_correlation": 0.9871704239157392, "3000–6000 Hz_10ms_modulation_mae_db": 0.9625894630377988, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6715394089099566, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08104251648293154, "3000–6000 Hz_40ms_envelope_correlation": 0.9955436130956331, "3000–6000 Hz_40ms_modulation_mae_db": 1.0443942244061262, "mr_stft_db": 4.63337206936673, "band_mae_db": 0.12525797648314638, "rms_gain_error_db": -0.003257859570086669, "output_peak": 0.5192503260624792}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5763613658824374, "全帯域_10ms_rms_envelope_nrmse": 0.05438516595119293, "全帯域_10ms_envelope_correlation": 0.9973771550548006, "全帯域_10ms_modulation_mae_db": 0.6792442349852955, "全帯域_40ms_log_rms_mae_db": 0.38675092655178767, "全帯域_40ms_rms_envelope_nrmse": 0.025784313054741808, "全帯域_40ms_envelope_correlation": 0.9994349097854331, "全帯域_40ms_modulation_mae_db": 1.0298059919664968, "300–1000 Hz_10ms_log_rms_mae_db": 0.8192771603901171, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06882092718553537, "300–1000 Hz_10ms_envelope_correlation": 0.9962522308756621, "300–1000 Hz_10ms_modulation_mae_db": 0.6662604209160278, "300–1000 Hz_40ms_log_rms_mae_db": 0.5992788880481371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04403758991599826, "300–1000 Hz_40ms_envelope_correlation": 0.9984352874721939, "300–1000 Hz_40ms_modulation_mae_db": 0.8831440993517125, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7373970076759094, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08766025794942993, "1000–3000 Hz_10ms_envelope_correlation": 0.9943086874680783, "1000–3000 Hz_10ms_modulation_mae_db": 0.38668240507080726, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5067646208511396, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04407635441575239, "1000–3000 Hz_40ms_envelope_correlation": 0.9985771267924572, "1000–3000 Hz_40ms_modulation_mae_db": 0.973415064348475, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9102268159304859, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1307916923698112, "3000–6000 Hz_10ms_envelope_correlation": 0.9873845168393229, "3000–6000 Hz_10ms_modulation_mae_db": 1.1841546479260494, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6175155659345906, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07853523468081922, "3000–6000 Hz_40ms_envelope_correlation": 0.9956558432590721, "3000–6000 Hz_40ms_modulation_mae_db": 1.6643548441406526, "mr_stft_db": 4.637019278270846, "band_mae_db": 0.0974410627984249, "rms_gain_error_db": -0.07070865935497865, "output_peak": 0.502871712208161}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5497114262889145, "全帯域_10ms_rms_envelope_nrmse": 0.06386607028756011, "全帯域_10ms_envelope_correlation": 0.9964601335990348, "全帯域_10ms_modulation_mae_db": 1.4142229224204954, "全帯域_40ms_log_rms_mae_db": 0.3920932514409549, "全帯域_40ms_rms_envelope_nrmse": 0.038073543145459944, "全帯域_40ms_envelope_correlation": 0.9988080443394268, "全帯域_40ms_modulation_mae_db": 1.8158557989006994, "300–1000 Hz_10ms_log_rms_mae_db": 0.794277337179311, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06693202052810399, "300–1000 Hz_10ms_envelope_correlation": 0.9964467323449006, "300–1000 Hz_10ms_modulation_mae_db": 0.5296326812069488, "300–1000 Hz_40ms_log_rms_mae_db": 0.6091142634099326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037820981229196936, "300–1000 Hz_40ms_envelope_correlation": 0.9988556431303249, "300–1000 Hz_40ms_modulation_mae_db": 0.8029553569447628, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7352904723605531, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10140152905226192, "1000–3000 Hz_10ms_envelope_correlation": 0.9926736096501565, "1000–3000 Hz_10ms_modulation_mae_db": 2.4150819604823326, "1000–3000 Hz_40ms_log_rms_mae_db": 0.541644196827567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06683021919233026, "1000–3000 Hz_40ms_envelope_correlation": 0.9969929330036134, "1000–3000 Hz_40ms_modulation_mae_db": 3.079960477426937, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9743916492594452, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1445109985541641, "3000–6000 Hz_10ms_envelope_correlation": 0.9851786275059066, "3000–6000 Hz_10ms_modulation_mae_db": 1.5831560211688067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6880299661771462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08043787716048, "3000–6000 Hz_40ms_envelope_correlation": 0.9964219409508963, "3000–6000 Hz_40ms_modulation_mae_db": 2.084050267011965, "mr_stft_db": 4.64142836292903, "band_mae_db": 0.1705093248606577, "rms_gain_error_db": -0.12702570785092088, "output_peak": 0.5131800525009563}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5545745263834675, "全帯域_10ms_rms_envelope_nrmse": 0.04934738879610052, "全帯域_10ms_envelope_correlation": 0.9978265658598968, "全帯域_10ms_modulation_mae_db": 0.5039761668864285, "全帯域_40ms_log_rms_mae_db": 0.359646719206422, "全帯域_40ms_rms_envelope_nrmse": 0.02408495535600874, "全帯域_40ms_envelope_correlation": 0.9994853545948292, "全帯域_40ms_modulation_mae_db": 0.6359927022950317, "300–1000 Hz_10ms_log_rms_mae_db": 0.8276950281778531, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06341935068248283, "300–1000 Hz_10ms_envelope_correlation": 0.9967754840878787, "300–1000 Hz_10ms_modulation_mae_db": 0.47788726684474003, "300–1000 Hz_40ms_log_rms_mae_db": 0.6188663944748617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04022113633345403, "300–1000 Hz_40ms_envelope_correlation": 0.9986407991484196, "300–1000 Hz_40ms_modulation_mae_db": 0.6914113407693204, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7191619381369003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08402287026655327, "1000–3000 Hz_10ms_envelope_correlation": 0.994766783707327, "1000–3000 Hz_10ms_modulation_mae_db": 0.5332237420217895, "1000–3000 Hz_40ms_log_rms_mae_db": 0.4838407486663248, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04334365803232032, "1000–3000 Hz_40ms_envelope_correlation": 0.9986058374346649, "1000–3000 Hz_40ms_modulation_mae_db": 0.5032998780886326, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9011835379568544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12328327834190944, "3000–6000 Hz_10ms_envelope_correlation": 0.9888699768410716, "3000–6000 Hz_10ms_modulation_mae_db": 0.6894216366494478, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6436796698831693, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07022257169035222, "3000–6000 Hz_40ms_envelope_correlation": 0.996676861410908, "3000–6000 Hz_40ms_modulation_mae_db": 1.2276143333523801, "mr_stft_db": 4.631668314691862, "band_mae_db": 0.12773271642101394, "rms_gain_error_db": -0.055176058776057475, "output_peak": 0.5108702718044219}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.59241510166588, "全帯域_10ms_rms_envelope_nrmse": 0.060912940473530436, "全帯域_10ms_envelope_correlation": 0.9967082473681488, "全帯域_10ms_modulation_mae_db": 0.6333408395571826, "全帯域_40ms_log_rms_mae_db": 0.3883226323472651, "全帯域_40ms_rms_envelope_nrmse": 0.03195273163068822, "全帯域_40ms_envelope_correlation": 0.999116405948833, "全帯域_40ms_modulation_mae_db": 0.9710553971703617, "300–1000 Hz_10ms_log_rms_mae_db": 0.8126380751851975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07619024077297899, "300–1000 Hz_10ms_envelope_correlation": 0.9953446202630287, "300–1000 Hz_10ms_modulation_mae_db": 0.3851590003232646, "300–1000 Hz_40ms_log_rms_mae_db": 0.5818446180110567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04450219466564458, "300–1000 Hz_40ms_envelope_correlation": 0.9983490371411686, "300–1000 Hz_40ms_modulation_mae_db": 0.4708876842410385, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7769653984360594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10272268408165033, "1000–3000 Hz_10ms_envelope_correlation": 0.9922140565338494, "1000–3000 Hz_10ms_modulation_mae_db": 0.3335658788670299, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5140451071170666, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054320753849516, "1000–3000 Hz_40ms_envelope_correlation": 0.9979004544288461, "1000–3000 Hz_40ms_modulation_mae_db": 0.6185358762573862, "3000–6000 Hz_10ms_log_rms_mae_db": 0.8783270459846096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12660690135046831, "3000–6000 Hz_10ms_envelope_correlation": 0.9882184547488544, "3000–6000 Hz_10ms_modulation_mae_db": 1.2188892367347715, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5995906113937559, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0686273204132001, "3000–6000 Hz_40ms_envelope_correlation": 0.9968472476498927, "3000–6000 Hz_40ms_modulation_mae_db": 1.7142401443312538, "mr_stft_db": 4.642244962501122, "band_mae_db": 0.11985081843863483, "rms_gain_error_db": -0.08227480657436115, "output_peak": 0.5087125516267811}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5418020975522643, "全帯域_10ms_rms_envelope_nrmse": 0.057126047256616214, "全帯域_10ms_envelope_correlation": 0.9970668914884292, "全帯域_10ms_modulation_mae_db": 0.47679023306872553, "全帯域_40ms_log_rms_mae_db": 0.38083756572666727, "全帯域_40ms_rms_envelope_nrmse": 0.02835797566031891, "全帯域_40ms_envelope_correlation": 0.9992436911409432, "全帯域_40ms_modulation_mae_db": 0.7872874397310486, "300–1000 Hz_10ms_log_rms_mae_db": 0.8615826281279277, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07230261953244696, "300–1000 Hz_10ms_envelope_correlation": 0.9958337009621748, "300–1000 Hz_10ms_modulation_mae_db": 0.23062173014171441, "300–1000 Hz_40ms_log_rms_mae_db": 0.6552626943672664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03798378236750098, "300–1000 Hz_40ms_envelope_correlation": 0.9988012142628048, "300–1000 Hz_40ms_modulation_mae_db": 0.3553575574893893, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7279376668985387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10280239180207516, "1000–3000 Hz_10ms_envelope_correlation": 0.9922525936850058, "1000–3000 Hz_10ms_modulation_mae_db": 0.6915324559370712, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5119456314146211, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05589666212276243, "1000–3000 Hz_40ms_envelope_correlation": 0.9976434185107829, "1000–3000 Hz_40ms_modulation_mae_db": 1.1131658528482662, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9491019626391852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13157060375061913, "3000–6000 Hz_10ms_envelope_correlation": 0.9873342005013304, "3000–6000 Hz_10ms_modulation_mae_db": 1.198725762117882, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6582668079538235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07580748126611332, "3000–6000 Hz_40ms_envelope_correlation": 0.996155913496652, "3000–6000 Hz_40ms_modulation_mae_db": 1.7523160677853236, "mr_stft_db": 4.654015673552112, "band_mae_db": 0.08543468541927336, "rms_gain_error_db": 0.019087512945457217, "output_peak": 0.50674830824937}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.574203902277889, "全帯域_10ms_rms_envelope_nrmse": 0.06932815322337267, "全帯域_10ms_envelope_correlation": 0.9956832897470286, "全帯域_10ms_modulation_mae_db": 0.7147679346078849, "全帯域_40ms_log_rms_mae_db": 0.4104742356970825, "全帯域_40ms_rms_envelope_nrmse": 0.03563983111669156, "全帯域_40ms_envelope_correlation": 0.9988097655855199, "全帯域_40ms_modulation_mae_db": 1.280109431461842, "300–1000 Hz_10ms_log_rms_mae_db": 0.8057992636519627, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06358662394024402, "300–1000 Hz_10ms_envelope_correlation": 0.9967604228987115, "300–1000 Hz_10ms_modulation_mae_db": 1.0262548886430538, "300–1000 Hz_40ms_log_rms_mae_db": 0.5937077155871558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03763875402564041, "300–1000 Hz_40ms_envelope_correlation": 0.9988082162602465, "300–1000 Hz_40ms_modulation_mae_db": 1.3313606699243694, "1000–3000 Hz_10ms_log_rms_mae_db": 0.7840718866109948, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1376726324650774, "1000–3000 Hz_10ms_envelope_correlation": 0.986267095213669, "1000–3000 Hz_10ms_modulation_mae_db": 0.7497976514175573, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5602374167055714, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08178460332817075, "1000–3000 Hz_40ms_envelope_correlation": 0.995041579358694, "1000–3000 Hz_40ms_modulation_mae_db": 1.3454968937993819, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9468777952888227, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1258414849636019, "3000–6000 Hz_10ms_envelope_correlation": 0.9888455714554527, "3000–6000 Hz_10ms_modulation_mae_db": 1.1662568288841013, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6716970805398396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07155933466637109, "3000–6000 Hz_40ms_envelope_correlation": 0.9971220875142129, "3000–6000 Hz_40ms_modulation_mae_db": 1.4048285245380308, "mr_stft_db": 4.6517915039756925, "band_mae_db": 0.1484109904934893, "rms_gain_error_db": 0.015128166460406801, "output_peak": 0.5702600632145507}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64720, "output_samples": 64720, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.836827889988321, "spectral_convergence": 0.23953327957255324, "spectral_convergence_status": "ok", "amplitude_floor": 1.1807658352589429e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.444617983380161, "spectral_convergence": 0.2292324789128836, "spectral_convergence_status": "ok", "amplitude_floor": 7.24953185483845e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.628440657759598, "spectral_convergence": 0.2331336158700404, "spectral_convergence_status": "ok", "amplitude_floor": 4.6531510695198173e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031402968364977683, "output_power": 0.00032144543394169384, "delta_db": 0.10136559397251418, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005546801738700932, "output_power": 0.0005634403482795084, "delta_db": 0.06805300820131135, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010899709362407707, "output_power": 0.0010524882647661282, "delta_db": -0.15197655603006793, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008204698714335104, "output_power": 0.0008380102615530727, "delta_db": 0.09186698754253025, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0004732551323003661, "output_power": 0.0004636285133058248, "delta_db": -0.0892519507979473, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00010423546715106559, "output_power": 9.547129232103485e-05, "delta_db": -0.38142715799636545, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1473235424234561, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.028762765052632956, "pitch": null, "output_rms_dbfs": -24.769713472649507, "output_peak": 0.5420700208940737, "reference_rms_dbfs": -24.740950707596877, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.63662884370936}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64720, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00017261805365043103, "log_rms_mae_db": 0.5757574712303629, "rms_envelope_nrmse": 0.055557191627919236, "envelope_correlation": 0.9972162207839168, "rms_gain_error_db": -0.028762765994811184, "reference_mean_power": 0.003360795516302892, "output_mean_power": 0.0033386109429327493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.9828935159721717e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046042689591210996, "output_power": 0.0004436946034130747, "reference_active": true, "delta_db": -0.1607653858771455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011507848708673394, "output_power": 0.00012232436289904498, "reference_active": true, "delta_db": 0.26518818830208946}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.734946402574847e-05, "output_power": 3.2147614994455103e-05, "reference_active": true, "delta_db": -0.6513561546136422}], "active_band_count": 3, "active_band_mae_db": 0.3591032429309591}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00013379439656886598, "log_rms_mae_db": 0.39479121675344414, "rms_envelope_nrmse": 0.030212023296151174, "envelope_correlation": 0.9991419785860847, "rms_gain_error_db": -0.028762754351235646, "reference_mean_power": 0.0033859385941668236, "output_mean_power": 0.0033635880606406286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.884361456130501e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004231625539622664, "output_power": 0.0004075913172727962, "reference_active": true, "delta_db": -0.16282305754612394}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.583961694519604e-05, "output_power": 5.958313094466741e-05, "reference_active": true, "delta_db": 0.2818089016479677}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3554104634271355e-06, "output_power": 2.5661805086698507e-06, "reference_active": true, "delta_db": -1.1645845250428306}], "active_band_count": 3, "active_band_mae_db": 0.5364054947456407}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00011786303332041996, "log_rms_mae_db": 0.8096574612766231, "rms_envelope_nrmse": 0.06820204773988031, "envelope_correlation": 0.9963003128581207, "rms_gain_error_db": -0.07649630772372876, "reference_mean_power": 0.0013229704572187174, "output_mean_power": 0.001299871780158083}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 9.694412540428036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002357826301518165, "output_power": 0.0002249393538096891, "reference_active": true, "delta_db": -0.2044636474290375}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.91091063998743e-05, "output_power": 4.755069512994315e-05, "reference_active": true, "delta_db": -0.9449752367186515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.351021856959089e-05, "output_power": 1.1703683069902708e-05, "reference_active": true, "delta_db": -0.6233982234476406}], "active_band_count": 3, "active_band_mae_db": 0.5909457025317765}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010411406065660645, "log_rms_mae_db": 0.6049847078972975, "rms_envelope_nrmse": 0.03909916630655077, "envelope_correlation": 0.9987653843533594, "rms_gain_error_db": -0.07649631108719447, "reference_mean_power": 0.0013328679887454193, "output_mean_power": 0.0013095965026662666}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 9.215200160073525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002182221456522984, "output_power": 0.00020559944912377362, "reference_active": true, "delta_db": -0.2587687493529483}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.729502442376393e-05, "output_power": 2.2348811836468493e-05, "reference_active": true, "delta_db": -0.8682904822137034}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1926400570958608e-06, "output_power": 7.694853838198371e-07, "reference_active": true, "delta_db": -1.9030901693289657}], "active_band_count": 3, "active_band_mae_db": 1.0100498002985392}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00014037287410750346, "log_rms_mae_db": 0.718558585033649, "rms_envelope_nrmse": 0.10314014268789853, "envelope_correlation": 0.9922687527516062, "rms_gain_error_db": 0.0969233050044282, "reference_mean_power": 0.001043588584426717, "output_mean_power": 0.0010671406174360947}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 7.131098310180307e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021758763787569653, "output_power": 0.0002290843630549066, "reference_active": true, "delta_db": 0.22361228409704806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8027856042619025e-05, "output_power": 8.194462844049711e-05, "reference_active": true, "delta_db": 1.4988396539761508}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3022954991601597e-05, "output_power": 2.0674739019090385e-05, "reference_active": true, "delta_db": -0.4672102839726163}], "active_band_count": 3, "active_band_mae_db": 0.7298874073486051}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010015064346906524, "log_rms_mae_db": 0.5153594838487323, "rms_envelope_nrmse": 0.06109772406705248, "envelope_correlation": 0.9972248988493945, "rms_gain_error_db": 0.09692330474239103, "reference_mean_power": 0.0010513959774729573, "output_mean_power": 0.0010751242101014087}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 6.799409328431789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020213626575120905, "output_power": 0.00021718629266735657, "reference_active": true, "delta_db": 0.31188173727564755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0065691390851235e-05, "output_power": 4.079405077779223e-05, "reference_active": true, "delta_db": 1.3252563680441443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1255469357025214e-06, "output_power": 1.7970820927089456e-06, "reference_active": true, "delta_db": -0.7290278275495565}], "active_band_count": 3, "active_band_mae_db": 0.7887219776231161}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 5.7697319934351994e-05, "log_rms_mae_db": 0.9689243629967214, "rms_envelope_nrmse": 0.1461836688749034, "envelope_correlation": 0.9850715842450547, "rms_gain_error_db": -0.4550613990634582, "reference_mean_power": 0.0001838054244420426, "output_mean_power": 0.0001655206503960783}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.2570064040799192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.4129018402509514e-05, "output_power": 3.775875300167714e-05, "reference_active": true, "delta_db": -0.677066240815315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0833298446779821e-05, "output_power": 9.964310316316007e-06, "reference_active": true, "delta_db": -0.36313463487401676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.606324451113381e-06, "output_power": 2.502435150979832e-06, "reference_active": true, "delta_db": -2.649916928448255}], "active_band_count": 3, "active_band_mae_db": 1.2300392680458623}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 3.902816175045304e-05, "log_rms_mae_db": 0.6926316340315226, "rms_envelope_nrmse": 0.08863868310523008, "envelope_correlation": 0.9955989807924449, "rms_gain_error_db": -0.45506140938907597, "reference_mean_power": 0.00018518052639832005, "output_mean_power": 0.00016675895823155037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.1869756093173816e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2060123324004214e-05, "output_power": 3.603068786427941e-05, "reference_active": true, "delta_db": -0.6719798839522345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.372206568913873e-06, "output_power": 4.440254072443041e-06, "reference_active": true, "delta_db": -0.8274488228868543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0554113039123233e-07, "output_power": 1.8109657843335617e-07, "reference_active": true, "delta_db": -3.5012466236158915}], "active_band_count": 3, "active_band_mae_db": 1.6668917768183267}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9567286053304751, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7fad8a545aa66369ccf497075cf90c5b3a8406aa7b49b4d5c7c7d225473e7fba", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9923376910154544, "standard_deviation": 0.0026559084735955432, "minimum": 0.986267095213669, "maximum": 0.994766783707327, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.7422205145246461, "standard_deviation": 0.024900835413310823, "minimum": 0.718558585033649, "maximum": 0.7840718866109948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8075736665556994, "standard_deviation": 0.667500685957604, "minimum": 0.3335658788670299, "maximum": 2.4150819604823326, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10120447466995222, "standard_deviation": 0.01663915335557016, "minimum": 0.08402287026655327, "maximum": 0.1376726324650774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975261737783978, "standard_deviation": 0.0011623925086537363, "minimum": 0.995041579358694, "maximum": 0.9986058374346649, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.5170703677164259, "standard_deviation": 0.023669733217788466, "minimum": 0.4838407486663248, "maximum": 0.5602374167055714, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.168771572113405, "standard_deviation": 0.8170815362901354, "minimum": 0.5032998780886326, "maximum": 3.079960477426937, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05699528543199744, "standard_deviation": 0.01287281915687047, "minimum": 0.04334365803232032, "maximum": 0.08178460332817075, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9872591695065915, "standard_deviation": 0.001472364534888697, "minimum": 0.9850715842450547, "maximum": 0.9888699768410716, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 0.93389638536662, "standard_deviation": 0.03391755836134332, "minimum": 0.8783270459846096, "maximum": 0.9743916492594452, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.15415410807059, "standard_deviation": 0.25346125644663076, "minimum": 0.6894216366494478, "maximum": 1.5831560211688067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13267571853999335, "standard_deviation": 0.008443662592893968, "minimum": 0.12328327834190944, "maximum": 0.1461836688749034, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.996252811021214, "standard_deviation": 0.0006110454154167694, "minimum": 0.9955436130956331, "maximum": 0.9971220875142129, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6553688431029755, "standard_deviation": 0.03310692586237292, "minimum": 0.5995906113937559, "maximum": 0.6926316340315226, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5698362727980073, "standard_deviation": 0.32892762464216924, "minimum": 1.0443942244061262, "maximum": 2.084050267011965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0768588774331872, "standard_deviation": 0.006685546695910769, "minimum": 0.0686273204132001, "maximum": 0.08863868310523008, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9961887836728296, "standard_deviation": 0.0004994804110517478, "minimum": 0.9953446202630287, "maximum": 0.9967754840878787, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.8243590339867054, "standard_deviation": 0.025613932605920837, "minimum": 0.794277337179311, "maximum": 0.8639453179046509, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5119806165824474, "standard_deviation": 0.2660555745967069, "minimum": 0.18908324205205304, "maximum": 1.0262548886430538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06901938206422609, "standard_deviation": 0.004499151489262755, "minimum": 0.06341935068248283, "maximum": 0.07619024077297899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986068986319513, "standard_deviation": 0.0002473451389725832, "minimum": 0.9981996072870923, "maximum": 0.9988556431303249, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.61699748810734, "standard_deviation": 0.03138164734933526, "minimum": 0.5818446180110567, "maximum": 0.6729206230630123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7181294143847061, "standard_deviation": 0.36989094496097386, "minimum": 0.1998688060585164, "maximum": 1.3313606699243694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04097924298113821, "standard_deviation": 0.00352126242880951, "minimum": 0.03763875402564041, "maximum": 0.04653033900511973, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12774513966726206, "standard_deviation": 0.027816194759673007, "minimum": 0.08543468541927336, "maximum": 0.1705093248606577, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.641021126124594, "standard_deviation": 0.008174987996791882, "minimum": 4.631668314691862, "maximum": 4.654015673552112, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5217454133200993, "standard_deviation": 0.023039114984462362, "minimum": 0.502871712208161, "maximum": 0.5702600632145507, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04162377222164672, "standard_deviation": 0.0514359293315276, "minimum": -0.12702570785092088, "maximum": 0.019087512945457217, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9970015327680106, "standard_deviation": 0.000701112946765063, "minimum": 0.9956832897470286, "maximum": 0.9978265658598968, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5676898219241546, "standard_deviation": 0.017073433116040544, "minimum": 0.5418020975522643, "maximum": 0.59241510166588, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6523897109449928, "standard_deviation": 0.3317372595660247, "minimum": 0.3591032429309591, "maximum": 1.4142229224204954, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.057662914268750615, "standard_deviation": 0.006730560449157913, "minimum": 0.04934738879610052, "maximum": 0.06932815322337267, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999168191083004, "standard_deviation": 0.00025573299833497924, "minimum": 0.9988080443394268, "maximum": 0.9994853545948292, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3897697786305616, "standard_deviation": 0.015570386354825957, "minimum": 0.359646719206422, "maximum": 0.4104742356970825, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.953627234579147, "standard_deviation": 0.43161225230926537, "minimum": 0.5364054947456407, "maximum": 1.8158557989006994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03014724451200202, "standard_deviation": 0.004850629499953954, "minimum": 0.02408495535600874, "maximum": 0.038073543145459944, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_121-123852-0000", "language": "en", "style": "normal", "speaker": "librispeech_121", "source_utterance_id": "121-123852-0000", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_121-123852-0000.wav", "source_file_sha256": "0fdfff29dbb40611c6ea098d5f4030890a4313798d90c45790aac1c61d1bd760", "decoded_pcm_sha256": "085897f5f5cc183ef4daa8868cb997e49f924fa58877dcfe167b6419325ce812", "original_source_sha256": "58d6350b52675d6d19616bc0347afc52a1535f8ea8be82aabb7275e6919a2598", "corpus_entry_id": "librispeech_121-123852-0000"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.144017219543457, "predicted_mos": 4.144017219543457, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "24f639a3432ec89f4bec813c254fcc2928b4405206f72116e3136e6265f94a31", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.934273200109601, "cpu_threads": 4}, "waveform_sha256": "dca47153f802b44bf64244e16fb092bb228bd570415ed247abbb4483db6e82a6"}, "rows": [{"case": "librispeech_121-123852-0000", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.4834401607513428, "predicted_mos": 2.4834401607513428, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2ccdc3816de8ddf25cf673957155df9e3e162ffa46484a18410749611bfd8a4d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0565730002708733, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6505781764907593, "全帯域_10ms_rms_envelope_nrmse": 0.07796336989279112, "全帯域_10ms_envelope_correlation": 0.9945800834372918, "全帯域_10ms_modulation_mae_db": 0.628837511769763, "全帯域_40ms_log_rms_mae_db": 0.34254002354987634, "全帯域_40ms_rms_envelope_nrmse": 0.04492086782530604, "全帯域_40ms_envelope_correlation": 0.9981131106661152, "全帯域_40ms_modulation_mae_db": 1.0447068028958364, "300–1000 Hz_10ms_log_rms_mae_db": 1.0154879980817193, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0631876387425875, "300–1000 Hz_10ms_envelope_correlation": 0.9972414817242357, "300–1000 Hz_10ms_modulation_mae_db": 0.1702677836765496, "300–1000 Hz_40ms_log_rms_mae_db": 0.6564894900569019, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03260618447762523, "300–1000 Hz_40ms_envelope_correlation": 0.999282327025277, "300–1000 Hz_40ms_modulation_mae_db": 0.06424384980244736, "1000–3000 Hz_10ms_log_rms_mae_db": 1.003057941850511, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12032380978437263, "1000–3000 Hz_10ms_envelope_correlation": 0.989158994411295, "1000–3000 Hz_10ms_modulation_mae_db": 0.5800871890571037, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6434632059729368, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06798127519833115, "1000–3000 Hz_40ms_envelope_correlation": 0.9965829216340947, "1000–3000 Hz_40ms_modulation_mae_db": 0.9744847685453344, "3000–6000 Hz_10ms_log_rms_mae_db": 1.098917165509492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1912815930270887, "3000–6000 Hz_10ms_envelope_correlation": 0.9730349160156251, "3000–6000 Hz_10ms_modulation_mae_db": 1.8815477424157088, "3000–6000 Hz_40ms_log_rms_mae_db": 0.664783656315377, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12335578263807105, "3000–6000 Hz_40ms_envelope_correlation": 0.9888378641875696, "3000–6000 Hz_40ms_modulation_mae_db": 2.0281611953167, "mr_stft_db": 4.2192659574367735, "band_mae_db": 0.14540099624173955, "rms_gain_error_db": -0.04391365631665398, "output_peak": 0.5128438398434072}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6569663048985325, "全帯域_10ms_rms_envelope_nrmse": 0.07948159941363699, "全帯域_10ms_envelope_correlation": 0.9943727829988619, "全帯域_10ms_modulation_mae_db": 0.5985884175172033, "全帯域_40ms_log_rms_mae_db": 0.3438655119078978, "全帯域_40ms_rms_envelope_nrmse": 0.043171541413897126, "全帯域_40ms_envelope_correlation": 0.9982431739319861, "全帯域_40ms_modulation_mae_db": 0.9399209733118393, "300–1000 Hz_10ms_log_rms_mae_db": 1.027418498985315, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06236726584747916, "300–1000 Hz_10ms_envelope_correlation": 0.9972284992787588, "300–1000 Hz_10ms_modulation_mae_db": 0.2101644303644055, "300–1000 Hz_40ms_log_rms_mae_db": 0.6933380993868244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03191158538372198, "300–1000 Hz_40ms_envelope_correlation": 0.9992503315152979, "300–1000 Hz_40ms_modulation_mae_db": 0.2474818479639063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0695582202837464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1392429912065716, "1000–3000 Hz_10ms_envelope_correlation": 0.9852585925765962, "1000–3000 Hz_10ms_modulation_mae_db": 0.8776249623019378, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6722364619410783, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07371932793725595, "1000–3000 Hz_40ms_envelope_correlation": 0.9957275942830373, "1000–3000 Hz_40ms_modulation_mae_db": 1.2354315064626542, "3000–6000 Hz_10ms_log_rms_mae_db": 1.035250886231164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1823655891797939, "3000–6000 Hz_10ms_envelope_correlation": 0.9757100205843311, "3000–6000 Hz_10ms_modulation_mae_db": 0.6862660352185884, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6244610061998374, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1119385098375276, "3000–6000 Hz_40ms_envelope_correlation": 0.9902184635826418, "3000–6000 Hz_40ms_modulation_mae_db": 1.1275337439636095, "mr_stft_db": 4.22646597708889, "band_mae_db": 0.10187903203883038, "rms_gain_error_db": 0.0012677512678853442, "output_peak": 0.4730380077426702}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6680073416580468, "全帯域_10ms_rms_envelope_nrmse": 0.0776014620453375, "全帯域_10ms_envelope_correlation": 0.9947575020493613, "全帯域_10ms_modulation_mae_db": 1.3794603651465964, "全帯域_40ms_log_rms_mae_db": 0.3541294250424557, "全帯域_40ms_rms_envelope_nrmse": 0.039229497899871034, "全帯域_40ms_envelope_correlation": 0.9987399730544877, "全帯域_40ms_modulation_mae_db": 1.7550598622680023, "300–1000 Hz_10ms_log_rms_mae_db": 1.0483212292024473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061059363740543784, "300–1000 Hz_10ms_envelope_correlation": 0.9973724415871144, "300–1000 Hz_10ms_modulation_mae_db": 0.40974862272759394, "300–1000 Hz_40ms_log_rms_mae_db": 0.7143627052078421, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03375667156527072, "300–1000 Hz_40ms_envelope_correlation": 0.9991903570661451, "300–1000 Hz_40ms_modulation_mae_db": 0.6512932558600949, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0472078847786832, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15009150272007124, "1000–3000 Hz_10ms_envelope_correlation": 0.9833787490617601, "1000–3000 Hz_10ms_modulation_mae_db": 1.6847067948767522, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6970924579998249, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08660861371795613, "1000–3000 Hz_40ms_envelope_correlation": 0.9949125880997315, "1000–3000 Hz_40ms_modulation_mae_db": 2.173966573347878, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0898122642128885, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.167902996484832, "3000–6000 Hz_10ms_envelope_correlation": 0.9797307632706169, "3000–6000 Hz_10ms_modulation_mae_db": 0.9967222746384575, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6333837377541748, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0878527123412516, "3000–6000 Hz_40ms_envelope_correlation": 0.9953512585463369, "3000–6000 Hz_40ms_modulation_mae_db": 1.1766963880955357, "mr_stft_db": 4.226237980537886, "band_mae_db": 0.20119930913941997, "rms_gain_error_db": -0.13029804363737027, "output_peak": 0.513140417999884}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6309988901259358, "全帯域_10ms_rms_envelope_nrmse": 0.0681004529554509, "全帯域_10ms_envelope_correlation": 0.9958997318111703, "全帯域_10ms_modulation_mae_db": 0.7988924203046782, "全帯域_40ms_log_rms_mae_db": 0.33632873300406774, "全帯域_40ms_rms_envelope_nrmse": 0.03382883939715687, "全帯域_40ms_envelope_correlation": 0.9989753460961843, "全帯域_40ms_modulation_mae_db": 1.1269257833807702, "300–1000 Hz_10ms_log_rms_mae_db": 1.0828189326863873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06524933470677209, "300–1000 Hz_10ms_envelope_correlation": 0.9969888937954743, "300–1000 Hz_10ms_modulation_mae_db": 0.6291186110845262, "300–1000 Hz_40ms_log_rms_mae_db": 0.7293765936407006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04085581875958497, "300–1000 Hz_40ms_envelope_correlation": 0.9987809582360143, "300–1000 Hz_40ms_modulation_mae_db": 0.9587441775683558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0294792204930157, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13992497317748315, "1000–3000 Hz_10ms_envelope_correlation": 0.9857707362305975, "1000–3000 Hz_10ms_modulation_mae_db": 1.5042213160449702, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6341833116262428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07896411824665789, "1000–3000 Hz_40ms_envelope_correlation": 0.9960541113710852, "1000–3000 Hz_40ms_modulation_mae_db": 2.1970041920407444, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0706833993638156, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14148159833718205, "3000–6000 Hz_10ms_envelope_correlation": 0.9851918900089843, "3000–6000 Hz_10ms_modulation_mae_db": 0.3063756768835846, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6274920516387879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07758985460121803, "3000–6000 Hz_40ms_envelope_correlation": 0.99547638610116, "3000–6000 Hz_40ms_modulation_mae_db": 0.5926645180260041, "mr_stft_db": 4.214508552388232, "band_mae_db": 0.1642261172821239, "rms_gain_error_db": -0.0658353596243327, "output_peak": 0.49444883089347436}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6402806842397646, "全帯域_10ms_rms_envelope_nrmse": 0.06966850502072038, "全帯域_10ms_envelope_correlation": 0.9956716339850379, "全帯域_10ms_modulation_mae_db": 0.7177195225525193, "全帯域_40ms_log_rms_mae_db": 0.3721671429030092, "全帯域_40ms_rms_envelope_nrmse": 0.03779740633497798, "全帯域_40ms_envelope_correlation": 0.9986574197948836, "全帯域_40ms_modulation_mae_db": 0.5850777909835095, "300–1000 Hz_10ms_log_rms_mae_db": 1.0767558495156064, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07015513846883964, "300–1000 Hz_10ms_envelope_correlation": 0.9965035769843494, "300–1000 Hz_10ms_modulation_mae_db": 0.329822924702923, "300–1000 Hz_40ms_log_rms_mae_db": 0.7323993131478834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04273398396906882, "300–1000 Hz_40ms_envelope_correlation": 0.9986554086006744, "300–1000 Hz_40ms_modulation_mae_db": 0.5134843215651674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0317695173520023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12221937812021981, "1000–3000 Hz_10ms_envelope_correlation": 0.9888416432394864, "1000–3000 Hz_10ms_modulation_mae_db": 0.9194463282115578, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6685512856702817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07125428689638774, "1000–3000 Hz_40ms_envelope_correlation": 0.9962609479712466, "1000–3000 Hz_40ms_modulation_mae_db": 1.3315270536033526, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0006902287484623, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1559971578437412, "3000–6000 Hz_10ms_envelope_correlation": 0.9818514016687876, "3000–6000 Hz_10ms_modulation_mae_db": 0.5005688773261067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6037913792757655, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08906652346594526, "3000–6000 Hz_40ms_envelope_correlation": 0.993727959020139, "3000–6000 Hz_40ms_modulation_mae_db": 0.5878598228369994, "mr_stft_db": 4.215941852939057, "band_mae_db": 0.12560786834166313, "rms_gain_error_db": -0.02390764044011862, "output_peak": 0.5025976491814874}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6401103897063694, "全帯域_10ms_rms_envelope_nrmse": 0.0678173976727928, "全帯域_10ms_envelope_correlation": 0.9959074518993634, "全帯域_10ms_modulation_mae_db": 0.431838790805714, "全帯域_40ms_log_rms_mae_db": 0.3367917936311433, "全帯域_40ms_rms_envelope_nrmse": 0.034995353950485246, "全帯域_40ms_envelope_correlation": 0.9988554530590501, "全帯域_40ms_modulation_mae_db": 0.7586325893557228, "300–1000 Hz_10ms_log_rms_mae_db": 1.0879396918918072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07441512691800815, "300–1000 Hz_10ms_envelope_correlation": 0.9960670607698874, "300–1000 Hz_10ms_modulation_mae_db": 0.10410783755278215, "300–1000 Hz_40ms_log_rms_mae_db": 0.7120693234656277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04363076050338217, "300–1000 Hz_40ms_envelope_correlation": 0.9986037309467145, "300–1000 Hz_40ms_modulation_mae_db": 0.3070422962786411, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9992606021352936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12379289975055023, "1000–3000 Hz_10ms_envelope_correlation": 0.9883772645548303, "1000–3000 Hz_10ms_modulation_mae_db": 0.5673701295789728, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6101683447515746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06444056479809325, "1000–3000 Hz_40ms_envelope_correlation": 0.9967538938798576, "1000–3000 Hz_40ms_modulation_mae_db": 1.0072390516444985, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0518715022442093, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15490711539191568, "3000–6000 Hz_10ms_envelope_correlation": 0.9821131255296324, "3000–6000 Hz_10ms_modulation_mae_db": 0.5681537805186422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6138697605810541, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09040130968871296, "3000–6000 Hz_40ms_envelope_correlation": 0.993548131990426, "3000–6000 Hz_40ms_modulation_mae_db": 0.7746761203998437, "mr_stft_db": 4.231287184298844, "band_mae_db": 0.08270996328333506, "rms_gain_error_db": -0.02559978159335995, "output_peak": 0.5267398092652436}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6669421807130632, "全帯域_10ms_rms_envelope_nrmse": 0.07117421195851117, "全帯域_10ms_envelope_correlation": 0.9955246757400618, "全帯域_10ms_modulation_mae_db": 0.4901558542317677, "全帯域_40ms_log_rms_mae_db": 0.3449777567501103, "全帯域_40ms_rms_envelope_nrmse": 0.03878272536671159, "全帯域_40ms_envelope_correlation": 0.9986536858497282, "全帯域_40ms_modulation_mae_db": 0.5525595433021436, "300–1000 Hz_10ms_log_rms_mae_db": 1.0527591935811285, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06421877654833094, "300–1000 Hz_10ms_envelope_correlation": 0.9970634553964365, "300–1000 Hz_10ms_modulation_mae_db": 0.23149453171796364, "300–1000 Hz_40ms_log_rms_mae_db": 0.6848500844957064, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03568697136520509, "300–1000 Hz_40ms_envelope_correlation": 0.9990665716461623, "300–1000 Hz_40ms_modulation_mae_db": 0.5077637731146037, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0322258595497191, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12662591216088473, "1000–3000 Hz_10ms_envelope_correlation": 0.9883196522071789, "1000–3000 Hz_10ms_modulation_mae_db": 1.0500104772049277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6794427198708797, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07915033062529432, "1000–3000 Hz_40ms_envelope_correlation": 0.9957173906993466, "1000–3000 Hz_40ms_modulation_mae_db": 1.6023498978331705, "3000–6000 Hz_10ms_log_rms_mae_db": 1.043299736227028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15268004356246423, "3000–6000 Hz_10ms_envelope_correlation": 0.982794953705329, "3000–6000 Hz_10ms_modulation_mae_db": 0.4841031463290331, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6350004753833902, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08902503924430315, "3000–6000 Hz_40ms_envelope_correlation": 0.9940487668037983, "3000–6000 Hz_40ms_modulation_mae_db": 0.38264405017726827, "mr_stft_db": 4.2248760442141355, "band_mae_db": 0.1517524434062054, "rms_gain_error_db": -0.0824992171287043, "output_peak": 0.5013171804262502}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6701828818641163, "全帯域_10ms_rms_envelope_nrmse": 0.07199867657928775, "全帯域_10ms_envelope_correlation": 0.9954300982417884, "全帯域_10ms_modulation_mae_db": 0.5990788813479788, "全帯域_40ms_log_rms_mae_db": 0.34323877313889034, "全帯域_40ms_rms_envelope_nrmse": 0.03730053151780756, "全帯域_40ms_envelope_correlation": 0.9987702927846575, "全帯域_40ms_modulation_mae_db": 0.989997652912462, "300–1000 Hz_10ms_log_rms_mae_db": 1.036513791719435, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06403750155560756, "300–1000 Hz_10ms_envelope_correlation": 0.9970800432853619, "300–1000 Hz_10ms_modulation_mae_db": 0.16489554675256415, "300–1000 Hz_40ms_log_rms_mae_db": 0.6395158700013512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03550269169669359, "300–1000 Hz_40ms_envelope_correlation": 0.9990731120364645, "300–1000 Hz_40ms_modulation_mae_db": 0.2671846237711107, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0355188484996056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12985478121590557, "1000–3000 Hz_10ms_envelope_correlation": 0.9873622956462587, "1000–3000 Hz_10ms_modulation_mae_db": 0.5712993819655694, "1000–3000 Hz_40ms_log_rms_mae_db": 0.656120992339463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0744281000603287, "1000–3000 Hz_40ms_envelope_correlation": 0.9959037405618864, "1000–3000 Hz_40ms_modulation_mae_db": 0.8776673553875008, "3000–6000 Hz_10ms_log_rms_mae_db": 1.058648790163468, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15941725925449038, "3000–6000 Hz_10ms_envelope_correlation": 0.9819137920279093, "3000–6000 Hz_10ms_modulation_mae_db": 1.7006549696850686, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6460481234062279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09221212413738736, "3000–6000 Hz_40ms_envelope_correlation": 0.9947129490135894, "3000–6000 Hz_40ms_modulation_mae_db": 1.9304548104784118, "mr_stft_db": 4.2161844827451, "band_mae_db": 0.165505801575509, "rms_gain_error_db": -0.08693277858279551, "output_peak": 0.490375329691442}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.237007747649607, "spectral_convergence": 0.30890241317896927, "spectral_convergence_status": "ok", "amplitude_floor": 1.0172154037614098e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.13640004552722, "spectral_convergence": 0.2803797159625959, "spectral_convergence_status": "ok", "amplitude_floor": 7.045944280558953e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.284390079133493, "spectral_convergence": 0.28834258072023444, "spectral_convergence_status": "ok", "amplitude_floor": 6.386940858702166e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00026428484162924415, "output_power": 0.00026746022206490717, "delta_db": 0.051869464827939846, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008055137994793597, "output_power": 0.0008249360211905925, "delta_db": 0.10347282837910571, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006109316451161528, "output_power": 0.0006108906291968196, "delta_db": -0.00029158066304326384, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027595006700553216, "output_power": 0.000264478805782961, "delta_db": -0.1843962861980033, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003628265208503686, "output_power": 0.000334147968358861, "delta_db": -0.3576019942896129, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001954243236049976, "output_power": 0.00018771597571219604, "delta_db": -0.17477382309273218, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14540099624173955, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04391365631665398, "pitch": null, "output_rms_dbfs": -26.038652573030138, "output_peak": 0.5128438398434072, "reference_rms_dbfs": -25.994738916713484, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.2192659574367735}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017615675305240148, "log_rms_mae_db": 0.6505781764907593, "rms_envelope_nrmse": 0.07796336989279112, "envelope_correlation": 0.9945800834372918, "rms_gain_error_db": -0.043846365712470844, "reference_mean_power": 0.0025168568603362385, "output_mean_power": 0.0024915745164835806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0162973895242792e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002469993456433397, "output_power": 0.00024709325436592523, "reference_active": true, "delta_db": 0.0016508662753416284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010173002443957726, "output_power": 8.841072143176757e-05, "reference_active": true, "delta_db": -0.6094421440408289}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1084315450169838e-05, "output_power": 1.5718733567863935e-05, "reference_active": true, "delta_db": -1.2754195249931186}], "active_band_count": 3, "active_band_mae_db": 0.628837511769763}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016525538510559444, "log_rms_mae_db": 0.34254002354987634, "rms_envelope_nrmse": 0.04492086782530604, "envelope_correlation": 0.9981131106661152, "rms_gain_error_db": -0.04379508896563162, "reference_mean_power": 0.002527970677484516, "output_mean_power": 0.0025026062409224583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.625293642327984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023064746753792456, "output_power": 0.00022906467061913362, "reference_active": true, "delta_db": -0.029905787861527512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.045652222183457e-05, "output_power": 4.868152052871478e-05, "reference_active": true, "delta_db": -0.9407902603628129}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2874374178901008e-06, "output_power": 1.3899743465760632e-06, "reference_active": true, "delta_db": -2.1634243604631687}], "active_band_count": 3, "active_band_mae_db": 1.0447068028958364}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001279910087199623, "log_rms_mae_db": 1.0154879980817193, "rms_envelope_nrmse": 0.0631876387425875, "envelope_correlation": 0.9972414817242357, "rms_gain_error_db": 0.07689731282398238, "reference_mean_power": 0.0009939686093891967, "output_mean_power": 0.0010117248107035714}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.147046828130101e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013887130907316313, "output_power": 0.0001399784452701852, "reference_active": true, "delta_db": 0.03448636121450175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4048550736300918e-05, "output_power": 2.4564213247066344e-05, "reference_active": true, "delta_db": 0.09213949785007289}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.939409383024551e-06, "output_power": 5.396258821142189e-06, "reference_active": true, "delta_db": 0.3841774919650741}], "active_band_count": 3, "active_band_mae_db": 0.1702677836765496}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011859895721274409, "log_rms_mae_db": 0.6564894900569019, "rms_envelope_nrmse": 0.03260618447762523, "envelope_correlation": 0.999282327025277, "rms_gain_error_db": 0.07682257829525974, "reference_mean_power": 0.0009986896907851761, "output_mean_power": 0.0010165127366743153}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.009742136856967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012678804378904424, "output_power": 0.00012807831572009512, "reference_active": true, "delta_db": 0.04397306636820443}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.298881565077983e-05, "output_power": 1.3295380188145408e-05, "reference_active": true, "delta_db": 0.10131207778802032}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8039990598306815e-07, "output_power": 2.7735322693572077e-07, "reference_active": true, "delta_db": -0.04744640525111737}], "active_band_count": 3, "active_band_mae_db": 0.06424384980244736}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.31629258809413e-05, "log_rms_mae_db": 1.003057941850511, "rms_envelope_nrmse": 0.12032380978437263, "envelope_correlation": 0.989158994411295, "rms_gain_error_db": -0.1885146713027479, "reference_mean_power": 0.0004146029386445929, "output_mean_power": 0.0003969912263958187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1875762004578613e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604229953230037e-05, "output_power": 8.410100870521325e-05, "reference_active": true, "delta_db": -0.09910803906877082}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2897806527831664e-05, "output_power": 2.2756823648886762e-05, "reference_active": true, "delta_db": -0.026822375191731303}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2516459896674564e-05, "output_power": 8.630747217333567e-06, "reference_active": true, "delta_db": -1.6143311529108089}], "active_band_count": 3, "active_band_mae_db": 0.5800871890571037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.786085823827924e-05, "log_rms_mae_db": 0.6434632059729368, "rms_envelope_nrmse": 0.06798127519833115, "envelope_correlation": 0.9965829216340947, "rms_gain_error_db": -0.18851774783948944, "reference_mean_power": 0.0004166837249389974, "output_mean_power": 0.00039898334136358784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.035015228514449e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.632173919031572e-05, "output_power": 8.322553533583967e-05, "reference_active": true, "delta_db": -0.15863584455019428}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4669499526204761e-05, "output_power": 1.2983209119622696e-05, "reference_active": true, "delta_db": -0.5303324513314593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2709360758910173e-06, "output_power": 7.597576933655363e-07, "reference_active": true, "delta_db": -2.2344860097543497}], "active_band_count": 3, "active_band_mae_db": 0.9744847685453344}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011544071276927888, "log_rms_mae_db": 1.098917165509492, "rms_envelope_nrmse": 0.1912815930270887, "envelope_correlation": 0.9730349160156251, "rms_gain_error_db": -0.3846656297764179, "reference_mean_power": 0.00033862428581899314, "output_mean_power": 0.0003099213813759141}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7742895295757852e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.9949103091747194e-05, "output_power": 5.128945129494583e-05, "reference_active": true, "delta_db": -0.6775463701400861}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5523159384577664e-05, "output_power": 3.2474845241187833e-05, "reference_active": true, "delta_db": -1.466853050871383}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.055143212029728e-06, "output_power": 3.1512398023724274e-06, "reference_active": true, "delta_db": -3.500243806235657}], "active_band_count": 3, "active_band_mae_db": 1.8815477424157088}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.669712818495073e-05, "log_rms_mae_db": 0.664783656315377, "rms_envelope_nrmse": 0.12335578263807105, "envelope_correlation": 0.9888378641875696, "rms_gain_error_db": -0.3846817930279346, "reference_mean_power": 0.00034032151520785135, "output_mean_power": 0.00031147358881684873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6870888960608118e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.86684133644439e-05, "output_power": 5.536010276227632e-05, "reference_active": true, "delta_db": -0.9356012519885291}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.573388989930172e-05, "output_power": 1.731901894476959e-05, "reference_active": true, "delta_db": -1.7198215124666096}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.935502106463895e-07, "output_power": 4.057073534671688e-07, "reference_active": true, "delta_db": -3.4290608214949616}], "active_band_count": 3, "active_band_mae_db": 2.0281611953167}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9000879831313928, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c193e3fc0c413824716b396affbb96d56956cebea09136093ee49f486eb10085", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9870584909910003, "standard_deviation": 0.002051208091598695, "minimum": 0.9833787490617601, "maximum": 0.989158994411295, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0310097618678222, "standard_deviation": 0.022559150057450402, "minimum": 0.9992606021352936, "maximum": 1.0695582202837464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9693458224052239, "standard_deviation": 0.4287507245925521, "minimum": 0.5673701295789728, "maximum": 1.6847067948767522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13150953101700735, "standard_deviation": 0.010511744276297056, "minimum": 0.12032380978437263, "maximum": 0.15009150272007124, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9959891485625357, "standard_deviation": 0.000576198983477282, "minimum": 0.9949125880997315, "maximum": 0.9967538938798576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6576573475215353, "standard_deviation": 0.027699996491302704, "minimum": 0.6101683447515746, "maximum": 0.6970924579998249, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4249587998581417, "standard_deviation": 0.5223749318879944, "minimum": 0.8776673553875008, "maximum": 2.1970041920407444, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07456832718503814, "standard_deviation": 0.00699969780597435, "minimum": 0.06444056479809325, "maximum": 0.08660861371795613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9802926078514019, "standard_deviation": 0.004011179511684054, "minimum": 0.9730349160156251, "maximum": 0.9851918900089843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.056146746587566, "standard_deviation": 0.03134060818081406, "minimum": 1.0006902287484623, "maximum": 1.098917165509492, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8905490628768987, "standard_deviation": 0.5920707439111577, "minimum": 0.3063756768835846, "maximum": 1.8815477424157088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16325416913518853, "standard_deviation": 0.016454746727903768, "minimum": 0.14148159833718205, "maximum": 0.1912815930270887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932402224057076, "standard_deviation": 0.0024238119018294623, "minimum": 0.9888378641875696, "maximum": 0.99547638610116, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6311037738193268, "standard_deviation": 0.018815645446211903, "minimum": 0.6037913792757655, "maximum": 0.664783656315377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0750863311617966, "standard_deviation": 0.6202448076941284, "minimum": 0.38264405017726827, "maximum": 2.0281611953167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09518023199430212, "standard_deviation": 0.014855143620307676, "minimum": 0.07758985460121803, "maximum": 0.12335578263807105, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9969431816027022, "standard_deviation": 0.00043928183556014985, "minimum": 0.9960670607698874, "maximum": 0.9973724415871144, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0535018982079807, "standard_deviation": 0.026799742581992553, "minimum": 1.0154879980817193, "maximum": 1.0879396918918072, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2812025360724135, "standard_deviation": 0.1709834157763342, "minimum": 0.10410783755278215, "maximum": 0.6291186110845262, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0655862683160211, "standard_deviation": 0.0044691613293708615, "minimum": 0.061059363740543784, "maximum": 0.07441512691800815, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989878496340937, "standard_deviation": 0.00027017402371986736, "minimum": 0.9986037309467145, "maximum": 0.999282327025277, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6953001849253547, "standard_deviation": 0.033622579231124744, "minimum": 0.6395158700013512, "maximum": 0.7323993131478834, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4396547682405409, "standard_deviation": 0.2801861909594545, "minimum": 0.06424384980244736, "maximum": 0.9587441775683558, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03708558346506907, "standard_deviation": 0.0046496855948985755, "minimum": 0.03191158538372198, "maximum": 0.04363076050338217, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1422851914136033, "standard_deviation": 0.03789234743459402, "minimum": 0.08270996328333506, "maximum": 0.20119930913941997, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.221846003956115, "standard_deviation": 0.006168552706840973, "minimum": 4.214508552388232, "maximum": 4.231287184298844, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5018126331304823, "standard_deviation": 0.016411009478735264, "minimum": 0.4730380077426702, "maximum": 0.5267398092652436, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05721484075693125, "standard_deviation": 0.042500319397545804, "minimum": -0.13029804363737027, "maximum": 0.0012677512678853442, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952679950203671, "standard_deviation": 0.0006092516857391051, "minimum": 0.9943727829988619, "maximum": 0.9959074518993634, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6530083562120735, "standard_deviation": 0.014872910482301684, "minimum": 0.6309988901259358, "maximum": 0.6701828818641163, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7055714704595275, "standard_deviation": 0.2959223427151602, "minimum": 0.431838790805714, "maximum": 1.3794603651465964, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07297570944231607, "standard_deviation": 0.004692164316978201, "minimum": 0.0678173976727928, "maximum": 0.07948159941363699, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986260569046366, "standard_deviation": 0.00029757750902063867, "minimum": 0.9981131106661152, "maximum": 0.9989753460961843, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3467548949909314, "standard_deviation": 0.01164823278428351, "minimum": 0.33632873300406774, "maximum": 0.3721671429030092, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9691101248012857, "standard_deviation": 0.38077401215777706, "minimum": 0.5525595433021436, "maximum": 1.7550598622680023, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038753345463276676, "standard_deviation": 0.003764194469859332, "minimum": 0.03382883939715687, "maximum": 0.04492086782530604, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-123852-0000", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5477240085601807, "predicted_mos": 2.5477240085601807, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "052ca9c02d7692371415cc071ecf89463bacc2adf0b314ce244751c92afaa540", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9838427999056876, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6466225035588662, "全帯域_10ms_rms_envelope_nrmse": 0.07723134617405271, "全帯域_10ms_envelope_correlation": 0.994680630941055, "全帯域_10ms_modulation_mae_db": 0.7683791835230389, "全帯域_40ms_log_rms_mae_db": 0.3349473766646003, "全帯域_40ms_rms_envelope_nrmse": 0.04459842990712507, "全帯域_40ms_envelope_correlation": 0.9981378779635253, "全帯域_40ms_modulation_mae_db": 1.148119268717669, "300–1000 Hz_10ms_log_rms_mae_db": 0.9969063424812113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06109114652367597, "300–1000 Hz_10ms_envelope_correlation": 0.9974489130146832, "300–1000 Hz_10ms_modulation_mae_db": 0.14643935812513742, "300–1000 Hz_40ms_log_rms_mae_db": 0.6414448864383904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03040945855879888, "300–1000 Hz_40ms_envelope_correlation": 0.9994045710073234, "300–1000 Hz_40ms_modulation_mae_db": 0.16400036803576995, "1000–3000 Hz_10ms_log_rms_mae_db": 1.008219556740845, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1206899214182664, "1000–3000 Hz_10ms_envelope_correlation": 0.9891106893853593, "1000–3000 Hz_10ms_modulation_mae_db": 0.851072667070588, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6488112832952483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06889967037903115, "1000–3000 Hz_40ms_envelope_correlation": 0.9965121452981077, "1000–3000 Hz_40ms_modulation_mae_db": 1.2436268773804973, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0908965164966846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19469977465365915, "3000–6000 Hz_10ms_envelope_correlation": 0.9718261087098781, "3000–6000 Hz_10ms_modulation_mae_db": 1.8941121850589813, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6522704104509318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12213222760987182, "3000–6000 Hz_40ms_envelope_correlation": 0.9887715064160494, "3000–6000 Hz_40ms_modulation_mae_db": 1.971965823689733, "mr_stft_db": 4.179752126102147, "band_mae_db": 0.13412222446449315, "rms_gain_error_db": -0.04053496293013748, "output_peak": 0.5071409852799277}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6631931031833198, "全帯域_10ms_rms_envelope_nrmse": 0.08206575694602472, "全帯域_10ms_envelope_correlation": 0.994011744020976, "全帯域_10ms_modulation_mae_db": 0.44889503978217765, "全帯域_40ms_log_rms_mae_db": 0.34790587328123285, "全帯域_40ms_rms_envelope_nrmse": 0.044116054767155966, "全帯域_40ms_envelope_correlation": 0.9981719335392736, "全帯域_40ms_modulation_mae_db": 0.697810899799112, "300–1000 Hz_10ms_log_rms_mae_db": 1.0031736057915015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06387675370272294, "300–1000 Hz_10ms_envelope_correlation": 0.9970937446157174, "300–1000 Hz_10ms_modulation_mae_db": 0.19449871720323864, "300–1000 Hz_40ms_log_rms_mae_db": 0.643027231352993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0318673010173098, "300–1000 Hz_40ms_envelope_correlation": 0.999252166484739, "300–1000 Hz_40ms_modulation_mae_db": 0.24986148213541093, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0538932315673868, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1420137888858595, "1000–3000 Hz_10ms_envelope_correlation": 0.9846571151244481, "1000–3000 Hz_10ms_modulation_mae_db": 0.6781276004852623, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6635122872296122, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07906158984144625, "1000–3000 Hz_40ms_envelope_correlation": 0.9949981415186302, "1000–3000 Hz_40ms_modulation_mae_db": 0.709238636179006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0684911543959934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18587088756958217, "3000–6000 Hz_10ms_envelope_correlation": 0.9750339608615073, "3000–6000 Hz_10ms_modulation_mae_db": 0.5308449616114609, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6295995502710664, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10867586219224037, "3000–6000 Hz_40ms_envelope_correlation": 0.9909749310444127, "3000–6000 Hz_40ms_modulation_mae_db": 1.0350751576319541, "mr_stft_db": 4.1941482504662595, "band_mae_db": 0.09663179658510619, "rms_gain_error_db": 0.014614385924508581, "output_peak": 0.47620764020564593}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6828758182719469, "全帯域_10ms_rms_envelope_nrmse": 0.08035596880875759, "全帯域_10ms_envelope_correlation": 0.9943577241815814, "全帯域_10ms_modulation_mae_db": 1.4635991881288788, "全帯域_40ms_log_rms_mae_db": 0.36216347334570115, "全帯域_40ms_rms_envelope_nrmse": 0.04100825505866106, "全帯域_40ms_envelope_correlation": 0.9985951409307305, "全帯域_40ms_modulation_mae_db": 1.9113073459011245, "300–1000 Hz_10ms_log_rms_mae_db": 1.0530406217534602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06517057963525383, "300–1000 Hz_10ms_envelope_correlation": 0.9969960905952796, "300–1000 Hz_10ms_modulation_mae_db": 0.3155908471406369, "300–1000 Hz_40ms_log_rms_mae_db": 0.6864391657282968, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03473645748991319, "300–1000 Hz_40ms_envelope_correlation": 0.9991375265996574, "300–1000 Hz_40ms_modulation_mae_db": 0.5403303665160027, "1000–3000 Hz_10ms_log_rms_mae_db": 1.062597634105646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14936860413639325, "1000–3000 Hz_10ms_envelope_correlation": 0.983437972128168, "1000–3000 Hz_10ms_modulation_mae_db": 1.6077486242529337, "1000–3000 Hz_40ms_log_rms_mae_db": 0.687562094987613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08333474640328792, "1000–3000 Hz_40ms_envelope_correlation": 0.9952225250715673, "1000–3000 Hz_40ms_modulation_mae_db": 2.083085883308048, "3000–6000 Hz_10ms_log_rms_mae_db": 1.108305973001847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1753142007500979, "3000–6000 Hz_10ms_envelope_correlation": 0.9776371823489876, "3000–6000 Hz_10ms_modulation_mae_db": 1.5078963102378067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6529294539585939, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09292808025097753, "3000–6000 Hz_40ms_envelope_correlation": 0.994409447947937, "3000–6000 Hz_40ms_modulation_mae_db": 1.7018180886131613, "mr_stft_db": 4.192467569581958, "band_mae_db": 0.19027323634150325, "rms_gain_error_db": -0.1275903810746608, "output_peak": 0.5027564478013269}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6090264531780892, "全帯域_10ms_rms_envelope_nrmse": 0.07006110606579306, "全帯域_10ms_envelope_correlation": 0.9956569473809274, "全帯域_10ms_modulation_mae_db": 0.9461164659615137, "全帯域_40ms_log_rms_mae_db": 0.3388230511081638, "全帯域_40ms_rms_envelope_nrmse": 0.038329384285857596, "全帯域_40ms_envelope_correlation": 0.9986696029009581, "全帯域_40ms_modulation_mae_db": 1.1491887488181856, "300–1000 Hz_10ms_log_rms_mae_db": 1.0446125335015226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0633899708723333, "300–1000 Hz_10ms_envelope_correlation": 0.9971589051507668, "300–1000 Hz_10ms_modulation_mae_db": 0.8267123218715744, "300–1000 Hz_40ms_log_rms_mae_db": 0.6795228752168107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0373288058862565, "300–1000 Hz_40ms_envelope_correlation": 0.9989840815853189, "300–1000 Hz_40ms_modulation_mae_db": 1.2148001812493876, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0283269457771387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1435306130779499, "1000–3000 Hz_10ms_envelope_correlation": 0.9850383807942584, "1000–3000 Hz_10ms_modulation_mae_db": 2.00040230100838, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6289918100786751, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0841349775739732, "1000–3000 Hz_40ms_envelope_correlation": 0.995448168105249, "1000–3000 Hz_40ms_modulation_mae_db": 2.8696788095801598, "3000–6000 Hz_10ms_log_rms_mae_db": 1.084022617461149, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14812039493961277, "3000–6000 Hz_10ms_envelope_correlation": 0.9836994522457658, "3000–6000 Hz_10ms_modulation_mae_db": 0.3246495779525535, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6253183135750501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08486445795396765, "3000–6000 Hz_40ms_envelope_correlation": 0.9943948050697784, "3000–6000 Hz_40ms_modulation_mae_db": 0.5302206578696796, "mr_stft_db": 4.178414349610511, "band_mae_db": 0.1489920766103329, "rms_gain_error_db": -0.07188352344129097, "output_peak": 0.4940053477973987}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6345110507736895, "全帯域_10ms_rms_envelope_nrmse": 0.06974648628510058, "全帯域_10ms_envelope_correlation": 0.9956620235311404, "全帯域_10ms_modulation_mae_db": 0.604591232685017, "全帯域_40ms_log_rms_mae_db": 0.374213722240089, "全帯域_40ms_rms_envelope_nrmse": 0.0382303007055667, "全帯域_40ms_envelope_correlation": 0.9986246618926756, "全帯域_40ms_modulation_mae_db": 0.48609863509861034, "300–1000 Hz_10ms_log_rms_mae_db": 1.0641395803656404, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06997104307201331, "300–1000 Hz_10ms_envelope_correlation": 0.9965343390138596, "300–1000 Hz_10ms_modulation_mae_db": 0.321262058865666, "300–1000 Hz_40ms_log_rms_mae_db": 0.7052426884683088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04156125916655405, "300–1000 Hz_40ms_envelope_correlation": 0.998736754318751, "300–1000 Hz_40ms_modulation_mae_db": 0.5321973684563134, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0248057759508982, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11956668735743185, "1000–3000 Hz_10ms_envelope_correlation": 0.9893194712538214, "1000–3000 Hz_10ms_modulation_mae_db": 0.7558062248386205, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6533443156103927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06675570139297557, "1000–3000 Hz_40ms_envelope_correlation": 0.9967411205432525, "1000–3000 Hz_40ms_modulation_mae_db": 1.0394813595124945, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0349745846371134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1629221742637969, "3000–6000 Hz_10ms_envelope_correlation": 0.9802258282616599, "3000–6000 Hz_10ms_modulation_mae_db": 0.4084298438195222, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6399044540844254, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09447904282188913, "3000–6000 Hz_40ms_envelope_correlation": 0.9928999711514682, "3000–6000 Hz_40ms_modulation_mae_db": 0.5668862273246239, "mr_stft_db": 4.185721722400872, "band_mae_db": 0.11381411091191057, "rms_gain_error_db": -0.017061523515553564, "output_peak": 0.49677941827091165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.653411888876793, "全帯域_10ms_rms_envelope_nrmse": 0.0688447073894567, "全帯域_10ms_envelope_correlation": 0.9957799721192141, "全帯域_10ms_modulation_mae_db": 0.38109212078841487, "全帯域_40ms_log_rms_mae_db": 0.34301525324761467, "全帯域_40ms_rms_envelope_nrmse": 0.03763209444128393, "全帯域_40ms_envelope_correlation": 0.9986742211737338, "全帯域_40ms_modulation_mae_db": 0.7092548282949771, "300–1000 Hz_10ms_log_rms_mae_db": 1.0823586878192946, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07331569742747686, "300–1000 Hz_10ms_envelope_correlation": 0.9961880639300394, "300–1000 Hz_10ms_modulation_mae_db": 0.11837467347255799, "300–1000 Hz_40ms_log_rms_mae_db": 0.6662782401504133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04158914637164017, "300–1000 Hz_40ms_envelope_correlation": 0.99873526277822, "300–1000 Hz_40ms_modulation_mae_db": 0.22951765935605128, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9973175962917703, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12355524004701929, "1000–3000 Hz_10ms_envelope_correlation": 0.9884357330718422, "1000–3000 Hz_10ms_modulation_mae_db": 0.6795976486373791, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6332595413300661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06572602551785504, "1000–3000 Hz_40ms_envelope_correlation": 0.9966460167947144, "1000–3000 Hz_40ms_modulation_mae_db": 1.1657185223543423, "3000–6000 Hz_10ms_log_rms_mae_db": 1.079174670791792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15965670772989687, "3000–6000 Hz_10ms_envelope_correlation": 0.9810452489268588, "3000–6000 Hz_10ms_modulation_mae_db": 0.5177734703007149, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6327640109401581, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09665353560679771, "3000–6000 Hz_40ms_envelope_correlation": 0.9926213359294941, "3000–6000 Hz_40ms_modulation_mae_db": 0.7296484042166087, "mr_stft_db": 4.196984307596246, "band_mae_db": 0.09027738721086503, "rms_gain_error_db": -0.027931415593438383, "output_peak": 0.5225001523597383}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.664774156502588, "全帯域_10ms_rms_envelope_nrmse": 0.07057403422778281, "全帯域_10ms_envelope_correlation": 0.9956035558561863, "全帯域_10ms_modulation_mae_db": 0.44167827904335005, "全帯域_40ms_log_rms_mae_db": 0.3422596895725105, "全帯域_40ms_rms_envelope_nrmse": 0.03814852473071078, "全帯域_40ms_envelope_correlation": 0.9987035255073764, "全帯域_40ms_modulation_mae_db": 0.5062215614406348, "300–1000 Hz_10ms_log_rms_mae_db": 1.020263424769328, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06363982088602069, "300–1000 Hz_10ms_envelope_correlation": 0.9971153784572587, "300–1000 Hz_10ms_modulation_mae_db": 0.10734071072185698, "300–1000 Hz_40ms_log_rms_mae_db": 0.6593878310849386, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03364223180780007, "300–1000 Hz_40ms_envelope_correlation": 0.9991666189575616, "300–1000 Hz_40ms_modulation_mae_db": 0.44655743825470157, "1000–3000 Hz_10ms_log_rms_mae_db": 1.046175700096002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1276817304559661, "1000–3000 Hz_10ms_envelope_correlation": 0.9880990560338253, "1000–3000 Hz_10ms_modulation_mae_db": 1.3006771926292393, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6603174210182441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07899353091637228, "1000–3000 Hz_40ms_envelope_correlation": 0.9957266231977723, "1000–3000 Hz_40ms_modulation_mae_db": 1.9408255553881977, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0704433314102706, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1552903890566629, "3000–6000 Hz_10ms_envelope_correlation": 0.9822682732386601, "3000–6000 Hz_10ms_modulation_mae_db": 0.4913003841478871, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6636691271917213, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09100967675519787, "3000–6000 Hz_40ms_envelope_correlation": 0.9938971869598148, "3000–6000 Hz_40ms_modulation_mae_db": 0.3223655978561081, "mr_stft_db": 4.189630590568406, "band_mae_db": 0.15286322810493258, "rms_gain_error_db": -0.0840846666323234, "output_peak": 0.4958535325629959}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6719504000430785, "全帯域_10ms_rms_envelope_nrmse": 0.07444425554135671, "全帯域_10ms_envelope_correlation": 0.9951163278564662, "全帯域_10ms_modulation_mae_db": 0.6104960171389938, "全帯域_40ms_log_rms_mae_db": 0.3579188969945939, "全帯域_40ms_rms_envelope_nrmse": 0.040698036378929936, "全帯域_40ms_envelope_correlation": 0.998530924005987, "全帯域_40ms_modulation_mae_db": 1.0651340809041974, "300–1000 Hz_10ms_log_rms_mae_db": 1.0501721599664564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06634325102089675, "300–1000 Hz_10ms_envelope_correlation": 0.9968661689183461, "300–1000 Hz_10ms_modulation_mae_db": 0.21257521426434078, "300–1000 Hz_40ms_log_rms_mae_db": 0.6394036103717279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03703466203805833, "300–1000 Hz_40ms_envelope_correlation": 0.9989898523993443, "300–1000 Hz_40ms_modulation_mae_db": 0.397905600347828, "1000–3000 Hz_10ms_log_rms_mae_db": 1.048878560169419, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13088065785516056, "1000–3000 Hz_10ms_envelope_correlation": 0.9872351158362478, "1000–3000 Hz_10ms_modulation_mae_db": 0.6808577821257064, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6634174527816796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07572651540249269, "1000–3000 Hz_40ms_envelope_correlation": 0.9958595148863939, "1000–3000 Hz_40ms_modulation_mae_db": 1.02740789642967, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0426563932286304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16532912903081498, "3000–6000 Hz_10ms_envelope_correlation": 0.9805677620395098, "3000–6000 Hz_10ms_modulation_mae_db": 1.8455706520930055, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6559386484203317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09839543853919833, "3000–6000 Hz_40ms_envelope_correlation": 0.9939297161508497, "3000–6000 Hz_40ms_modulation_mae_db": 2.1599663563035367, "mr_stft_db": 4.185223745212288, "band_mae_db": 0.17640691042356638, "rms_gain_error_db": -0.09342414684827328, "output_peak": 0.4854668614159547}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.208091919316343, "spectral_convergence": 0.3089634527993491, "spectral_convergence_status": "ok", "amplitude_floor": 1.0172154037614098e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.097066367798939, "spectral_convergence": 0.2796327275050435, "spectral_convergence_status": "ok", "amplitude_floor": 7.045944280558953e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.234098091191158, "spectral_convergence": 0.2846042573421067, "spectral_convergence_status": "ok", "amplitude_floor": 6.386940858702166e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00026428484162924415, "output_power": 0.00026523217703780945, "delta_db": 0.015539556422093708, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008055137994793597, "output_power": 0.0008251371088688134, "delta_db": 0.10453134230641584, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006109316451161528, "output_power": 0.0006128519708953326, "delta_db": 0.013629657448454896, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027595006700553216, "output_power": 0.0002649534741904206, "delta_db": -0.17660885301244966, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003628265208503686, "output_power": 0.00033387297444505116, "delta_db": -0.3611775811737423, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001954243236049976, "output_power": 0.00018951953486189287, "delta_db": -0.1332463564238026, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13412222446449315, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04053496293013748, "pitch": null, "output_rms_dbfs": -26.035273879643622, "output_peak": 0.5071409852799277, "reference_rms_dbfs": -25.994738916713484, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.179752126102147}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017615675305240148, "log_rms_mae_db": 0.6466225035588662, "rms_envelope_nrmse": 0.07723134617405271, "envelope_correlation": 0.994680630941055, "rms_gain_error_db": -0.0404395381141535, "reference_mean_power": 0.0025168568603362385, "output_mean_power": 0.002493529801527239}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0162973895242792e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002469993456433397, "output_power": 0.00024795990234147483, "reference_active": true, "delta_db": 0.016856539158233003}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010173002443957726, "output_power": 8.993503675947755e-05, "reference_active": true, "delta_db": -0.5352023232779921}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1084315450169838e-05, "output_power": 1.4081588092884145e-05, "reference_active": true, "delta_db": -1.7530786881328915}], "active_band_count": 3, "active_band_mae_db": 0.7683791835230389}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016525538510559444, "log_rms_mae_db": 0.3349473766646003, "rms_envelope_nrmse": 0.04459842990712507, "envelope_correlation": 0.9981378779635253, "rms_gain_error_db": -0.0403603596451333, "reference_mean_power": 0.002527970677484516, "output_mean_power": 0.00250458627414474}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.625293642327984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023064746753792456, "output_power": 0.00022982096473682487, "reference_active": true, "delta_db": -0.015590471185952436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.045652222183457e-05, "output_power": 4.88843001860488e-05, "reference_active": true, "delta_db": -0.9227375832298069}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2874374178901008e-06, "output_power": 1.2845358984630836e-06, "reference_active": true, "delta_db": -2.5060297517372474}], "active_band_count": 3, "active_band_mae_db": 1.148119268717669}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001279910087199623, "log_rms_mae_db": 0.9969063424812113, "rms_envelope_nrmse": 0.06109114652367597, "envelope_correlation": 0.9974489130146832, "rms_gain_error_db": 0.08814235462571653, "reference_mean_power": 0.0009939686093891967, "output_mean_power": 0.0010143478303091713}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.147046828130101e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013887130907316313, "output_power": 0.0001430580274616577, "reference_active": true, "delta_db": 0.12899703125298162}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4048550736300918e-05, "output_power": 2.4512970347570082e-05, "reference_active": true, "delta_db": 0.0830703072628152}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.939409383024551e-06, "output_power": 5.204752662035016e-06, "reference_active": true, "delta_db": 0.22725073585961544}], "active_band_count": 3, "active_band_mae_db": 0.14643935812513742}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011859895721274409, "log_rms_mae_db": 0.6414448864383904, "rms_envelope_nrmse": 0.03040945855879888, "envelope_correlation": 0.9994045710073234, "rms_gain_error_db": 0.08811337296099946, "reference_mean_power": 0.0009986896907851761, "output_mean_power": 0.0010191589063287656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.009742136856967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012678804378904424, "output_power": 0.00013094089137164744, "reference_active": true, "delta_db": 0.13996991857817093}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.298881565077983e-05, "output_power": 1.3185485378615942e-05, "reference_active": true, "delta_db": 0.06526568554442543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8039990598306815e-07, "output_power": 2.6248308013341156e-07, "reference_active": true, "delta_db": -0.2867654999847135}], "active_band_count": 3, "active_band_mae_db": 0.16400036803576995}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.31629258809413e-05, "log_rms_mae_db": 1.008219556740845, "rms_envelope_nrmse": 0.1206899214182664, "envelope_correlation": 0.9891106893853593, "rms_gain_error_db": -0.1959602486708416, "reference_mean_power": 0.0004146029386445929, "output_mean_power": 0.0003963112047254297}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1875762004578613e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604229953230037e-05, "output_power": 8.484647830140882e-05, "reference_active": true, "delta_db": -0.06078187761729642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2897806527831664e-05, "output_power": 2.181647013216834e-05, "reference_active": true, "delta_db": -0.2100939770626512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2516459896674564e-05, "output_power": 7.400265437214711e-06, "reference_active": true, "delta_db": -2.2823421465318168}], "active_band_count": 3, "active_band_mae_db": 0.851072667070588}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.786085823827924e-05, "log_rms_mae_db": 0.6488112832952483, "rms_envelope_nrmse": 0.06889967037903115, "envelope_correlation": 0.9965121452981077, "rms_gain_error_db": -0.19596072427818695, "reference_mean_power": 0.0004166837249389974, "output_mean_power": 0.0003983001458585591}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.035015228514449e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.632173919031572e-05, "output_power": 8.336125552176358e-05, "reference_active": true, "delta_db": -0.1515593482004731}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4669499526204761e-05, "output_power": 1.2139156726294615e-05, "reference_active": true, "delta_db": -0.8222677888623936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2709360758910173e-06, "output_power": 6.736252502471778e-07, "reference_active": true, "delta_db": -2.757053495078625}], "active_band_count": 3, "active_band_mae_db": 1.2436268773804973}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011544071276927888, "log_rms_mae_db": 1.0908965164966846, "rms_envelope_nrmse": 0.19469977465365915, "envelope_correlation": 0.9718261087098781, "rms_gain_error_db": -0.33352443918996666, "reference_mean_power": 0.00033862428581899314, "output_mean_power": 0.00031359249334701864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7742895295757852e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.9949103091747194e-05, "output_power": 5.4385508867112586e-05, "reference_active": true, "delta_db": -0.42299493496564466}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5523159384577664e-05, "output_power": 3.364424127752712e-05, "reference_active": true, "delta_db": -1.3132165637782045}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.055143212029728e-06, "output_power": 2.8437625345515574e-06, "reference_active": true, "delta_db": -3.9461250564330945}], "active_band_count": 3, "active_band_mae_db": 1.8941121850589813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.669712818495073e-05, "log_rms_mae_db": 0.6522704104509318, "rms_envelope_nrmse": 0.12213222760987182, "envelope_correlation": 0.9887715064160494, "rms_gain_error_db": -0.3335313761674533, "reference_mean_power": 0.00034032151520785135, "output_mean_power": 0.0003151637566939478}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6870888960608118e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.86684133644439e-05, "output_power": 5.896103185103149e-05, "reference_active": true, "delta_db": -0.6619193749429088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.573388989930172e-05, "output_power": 1.7676813808392523e-05, "reference_active": true, "delta_db": -1.63101451047841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.935502106463895e-07, "output_power": 3.8799184990628836e-07, "reference_active": true, "delta_db": -3.62296358564788}], "active_band_count": 3, "active_band_mae_db": 1.971965823689733}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8994195861735838, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4c1238491a02344cc0af17ff8da9e30c5f2f480b5b35cd230dab005b3a40612f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9869166917034963, "standard_deviation": 0.002240187156859164, "minimum": 0.983437972128168, "maximum": 0.9893194712538214, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0337768750873884, "standard_deviation": 0.023020902250211424, "minimum": 0.9973175962917703, "maximum": 1.062597634105646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0692862551310136, "standard_deviation": 0.5087921917906288, "minimum": 0.6781276004852623, "maximum": 2.00040230100838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13216090540425585, "standard_deviation": 0.011391800191404497, "minimum": 0.11956668735743185, "maximum": 0.14936860413639325, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995894281926961, "standard_deviation": 0.0006704643104979771, "minimum": 0.9949981415186302, "maximum": 0.9967411205432525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6549020257914413, "standard_deviation": 0.01860075167961519, "minimum": 0.6289918100786751, "maximum": 0.687562094987613, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5098829425165519, "standard_deviation": 0.7218775630248868, "minimum": 0.709238636179006, "maximum": 2.8696788095801598, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07532909467842926, "standard_deviation": 0.007331060106317707, "minimum": 0.06572602551785504, "maximum": 0.0841349775739732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9790379770791034, "standard_deviation": 0.003963222964373477, "minimum": 0.9718261087098781, "maximum": 0.9836994522457658, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.0723706551779348, "standard_deviation": 0.02425102653656024, "minimum": 1.0349745846371134, "maximum": 1.108305973001847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9400721731527415, "standard_deviation": 0.6825510814021121, "minimum": 0.3246495779525535, "maximum": 1.8941121850589813, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16840045724926545, "standard_deviation": 0.015783003938048844, "minimum": 0.14812039493961277, "maximum": 0.19469977465365915, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927373625837255, "standard_deviation": 0.001970640522282853, "minimum": 0.9887715064160494, "maximum": 0.994409447947937, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6440492461115349, "standard_deviation": 0.014025082799618525, "minimum": 0.6253183135750501, "maximum": 0.6636691271917213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1272432891881756, "standard_deviation": 0.7166439749576919, "minimum": 0.3223655978561081, "maximum": 2.1599663563035367, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09864229021626755, "standard_deviation": 0.01167742682325928, "minimum": 0.08486445795396765, "maximum": 0.12213222760987182, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.996925200461994, "standard_deviation": 0.00039607369327983386, "minimum": 0.9961880639300394, "maximum": 0.9974489130146832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.039333369556052, "standard_deviation": 0.029930891244418126, "minimum": 0.9969063424812113, "maximum": 1.0823586878192946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.28034923770812614, "standard_deviation": 0.23527845070134237, "minimum": 0.10734071072185698, "maximum": 0.8267123218715744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0658497828925492, "standard_deviation": 0.0039803367533101965, "minimum": 0.06109114652367597, "maximum": 0.07331569742747686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9990508542663645, "standard_deviation": 0.00023694251333600654, "minimum": 0.99873526277822, "maximum": 0.9994045710073234, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6650933161014849, "standard_deviation": 0.023952057711837203, "minimum": 0.6394036103717279, "maximum": 0.7052426884683088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4718963080439332, "standard_deviation": 0.33148778674906393, "minimum": 0.16400036803576995, "maximum": 1.2148001812493876, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03602116529204137, "standard_deviation": 0.004148800042544371, "minimum": 0.03040945855879888, "maximum": 0.04158914637164017, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13792262133158875, "standard_deviation": 0.036130805620335896, "minimum": 0.09027738721086503, "maximum": 0.19027323634150325, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.187792832692336, "standard_deviation": 0.006693319491357336, "minimum": 4.178414349610511, "maximum": 4.196984307596246, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.49758879821173746, "standard_deviation": 0.013934167023893605, "minimum": 0.47620764020564593, "maximum": 0.5225001523597383, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05598702926389616, "standard_deviation": 0.04643950312451184, "minimum": -0.1275903810746608, "maximum": 0.014614385924508581, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951086157359433, "standard_deviation": 0.0006816795480519623, "minimum": 0.994011744020976, "maximum": 0.9957799721192141, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6532956717985463, "standard_deviation": 0.023329530934786322, "minimum": 0.6090264531780892, "maximum": 0.6828758182719469, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7081059408814231, "standard_deviation": 0.35785211256346394, "minimum": 0.38109212078841487, "maximum": 1.4635991881288788, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0741654576797906, "standard_deviation": 0.005179579043332698, "minimum": 0.0688447073894567, "maximum": 0.08206575694602472, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985134859892826, "standard_deviation": 0.000227845332431007, "minimum": 0.9981378779635253, "maximum": 0.9987035255073764, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.35015591705681326, "standard_deviation": 0.013423641151582214, "minimum": 0.3349473766646003, "maximum": 0.374213722240089, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9591419211218138, "standard_deviation": 0.4704599319867272, "minimum": 0.48609863509861034, "maximum": 1.9113073459011245, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04034513503441138, "standard_deviation": 0.002767053147189909, "minimum": 0.03763209444128393, "maximum": 0.04459842990712507, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-123852-0000", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.6640822887420654, "predicted_mos": 2.6640822887420654, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8f982262e02f515f561dae4865c43da7791ee12ebf21f5c0ea42ce0b2113089c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.784906200133264, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6289412106791312, "全帯域_10ms_rms_envelope_nrmse": 0.0711921530589569, "全帯域_10ms_envelope_correlation": 0.9954850338090805, "全帯域_10ms_modulation_mae_db": 1.013504136875952, "全帯域_40ms_log_rms_mae_db": 0.3413488004011963, "全帯域_40ms_rms_envelope_nrmse": 0.038612229832562635, "全帯域_40ms_envelope_correlation": 0.9986123824977522, "全帯域_40ms_modulation_mae_db": 1.3494113231109204, "300–1000 Hz_10ms_log_rms_mae_db": 1.3178911945139378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06535201095005795, "300–1000 Hz_10ms_envelope_correlation": 0.9969673841743909, "300–1000 Hz_10ms_modulation_mae_db": 0.27570462074874197, "300–1000 Hz_40ms_log_rms_mae_db": 0.9349425580505859, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03881257764562292, "300–1000 Hz_40ms_envelope_correlation": 0.9988914147469561, "300–1000 Hz_40ms_modulation_mae_db": 0.6500126176050038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.206774482776524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1417887783500988, "1000–3000 Hz_10ms_envelope_correlation": 0.9848930448399501, "1000–3000 Hz_10ms_modulation_mae_db": 1.6634080114131387, "1000–3000 Hz_40ms_log_rms_mae_db": 0.813351399883142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08653455312985248, "1000–3000 Hz_40ms_envelope_correlation": 0.9943760247889739, "1000–3000 Hz_40ms_modulation_mae_db": 2.113004109160864, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3015906229673397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16360688855894812, "3000–6000 Hz_10ms_envelope_correlation": 0.9801646751931736, "3000–6000 Hz_10ms_modulation_mae_db": 0.9415031875514717, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7739801628962564, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0955466279041648, "3000–6000 Hz_40ms_envelope_correlation": 0.9931429647422816, "3000–6000 Hz_40ms_modulation_mae_db": 1.3992432917734436, "mr_stft_db": 4.441440938675264, "band_mae_db": 0.13515767248494334, "rms_gain_error_db": -0.04338349496720162, "output_peak": 0.4796792258142698}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6399708053870075, "全帯域_10ms_rms_envelope_nrmse": 0.06764857405187444, "全帯域_10ms_envelope_correlation": 0.9959299976046759, "全帯域_10ms_modulation_mae_db": 0.5378197515014783, "全帯域_40ms_log_rms_mae_db": 0.34582216933878784, "全帯域_40ms_rms_envelope_nrmse": 0.038090728776600954, "全帯域_40ms_envelope_correlation": 0.9986552522742586, "全帯域_40ms_modulation_mae_db": 0.6178122732856991, "300–1000 Hz_10ms_log_rms_mae_db": 1.2291860344475476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06738106291693341, "300–1000 Hz_10ms_envelope_correlation": 0.9967706432175097, "300–1000 Hz_10ms_modulation_mae_db": 0.17926328814322498, "300–1000 Hz_40ms_log_rms_mae_db": 0.878777440403469, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03575357154975265, "300–1000 Hz_40ms_envelope_correlation": 0.9990633480500788, "300–1000 Hz_40ms_modulation_mae_db": 0.25841220518822944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2061277890380355, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12636827226636252, "1000–3000 Hz_10ms_envelope_correlation": 0.9879689911115872, "1000–3000 Hz_10ms_modulation_mae_db": 1.0189910692685957, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8192172311952307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06935327384071711, "1000–3000 Hz_40ms_envelope_correlation": 0.996373595385027, "1000–3000 Hz_40ms_modulation_mae_db": 1.0681717858227584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.301971049048132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1576475367448064, "3000–6000 Hz_10ms_envelope_correlation": 0.9815119033513746, "3000–6000 Hz_10ms_modulation_mae_db": 0.4342387793906912, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8192155789990555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09815913247062479, "3000–6000 Hz_40ms_envelope_correlation": 0.9923901851971424, "3000–6000 Hz_40ms_modulation_mae_db": 0.7742619067787887, "mr_stft_db": 4.440808350041833, "band_mae_db": 0.08393333956578948, "rms_gain_error_db": -0.04922735037247819, "output_peak": 0.4743646170672833}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.655729845849821, "全帯域_10ms_rms_envelope_nrmse": 0.07547699028595845, "全帯域_10ms_envelope_correlation": 0.9949570890358755, "全帯域_10ms_modulation_mae_db": 0.6731859733665772, "全帯域_40ms_log_rms_mae_db": 0.3653725040701534, "全帯域_40ms_rms_envelope_nrmse": 0.04527761286251018, "全帯域_40ms_envelope_correlation": 0.9981301916985124, "全帯域_40ms_modulation_mae_db": 0.6964356251279403, "300–1000 Hz_10ms_log_rms_mae_db": 1.2769895063840992, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.060616914897697445, "300–1000 Hz_10ms_envelope_correlation": 0.9974282502464077, "300–1000 Hz_10ms_modulation_mae_db": 0.207978586068596, "300–1000 Hz_40ms_log_rms_mae_db": 0.8632061928186137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035464141330866644, "300–1000 Hz_40ms_envelope_correlation": 0.9991236633638226, "300–1000 Hz_40ms_modulation_mae_db": 0.7603703744927489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1908044444914148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13986209972634353, "1000–3000 Hz_10ms_envelope_correlation": 0.9852092804672529, "1000–3000 Hz_10ms_modulation_mae_db": 1.0778751559227393, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8102043024788151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08083513663379839, "1000–3000 Hz_40ms_envelope_correlation": 0.9950014175354308, "1000–3000 Hz_40ms_modulation_mae_db": 1.1286094180629027, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2629122724832569, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17652875899224393, "3000–6000 Hz_10ms_envelope_correlation": 0.9771476005838401, "3000–6000 Hz_10ms_modulation_mae_db": 0.42281102754129013, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7822362071793643, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11959545820379017, "3000–6000 Hz_40ms_envelope_correlation": 0.9894311998932684, "3000–6000 Hz_40ms_modulation_mae_db": 0.2658765989930078, "mr_stft_db": 4.446633736343668, "band_mae_db": 0.17180600731710435, "rms_gain_error_db": -0.07772385958749085, "output_peak": 0.5005391950388067}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6764835618523974, "全帯域_10ms_rms_envelope_nrmse": 0.07823900188187119, "全帯域_10ms_envelope_correlation": 0.9945440078372658, "全帯域_10ms_modulation_mae_db": 0.7045177546376368, "全帯域_40ms_log_rms_mae_db": 0.3875897762259138, "全帯域_40ms_rms_envelope_nrmse": 0.046401447318890295, "全帯域_40ms_envelope_correlation": 0.997984670467062, "全帯域_40ms_modulation_mae_db": 1.0610445191893456, "300–1000 Hz_10ms_log_rms_mae_db": 1.2898876491635463, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06805590753732584, "300–1000 Hz_10ms_envelope_correlation": 0.9967155719930252, "300–1000 Hz_10ms_modulation_mae_db": 0.29529077731958936, "300–1000 Hz_40ms_log_rms_mae_db": 0.9158484339920404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04015664831263381, "300–1000 Hz_40ms_envelope_correlation": 0.9988161234075996, "300–1000 Hz_40ms_modulation_mae_db": 0.41851029579383364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2220888241453023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14798289910686685, "1000–3000 Hz_10ms_envelope_correlation": 0.9835530214646411, "1000–3000 Hz_10ms_modulation_mae_db": 1.353192093103002, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8192094338266857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08545783709910527, "1000–3000 Hz_40ms_envelope_correlation": 0.9946110228603608, "1000–3000 Hz_40ms_modulation_mae_db": 2.0127140421553125, "3000–6000 Hz_10ms_log_rms_mae_db": 1.263444566329756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17249860179713347, "3000–6000 Hz_10ms_envelope_correlation": 0.9781670033431001, "3000–6000 Hz_10ms_modulation_mae_db": 1.0891953454398708, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7812268118915484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11780632846206547, "3000–6000 Hz_40ms_envelope_correlation": 0.9896803308903381, "3000–6000 Hz_40ms_modulation_mae_db": 1.3266919549858793, "mr_stft_db": 4.442462866192464, "band_mae_db": 0.17862498957146258, "rms_gain_error_db": -0.044323835847938864, "output_peak": 0.5050583724941751}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6461358235727591, "全帯域_10ms_rms_envelope_nrmse": 0.07617168626433339, "全帯域_10ms_envelope_correlation": 0.9948294278605924, "全帯域_10ms_modulation_mae_db": 0.6345229692614544, "全帯域_40ms_log_rms_mae_db": 0.3370951810217448, "全帯域_40ms_rms_envelope_nrmse": 0.03424473750071487, "全帯域_40ms_envelope_correlation": 0.9989061152006257, "全帯域_40ms_modulation_mae_db": 0.8574788590930528, "300–1000 Hz_10ms_log_rms_mae_db": 1.223760424765824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06581850807735476, "300–1000 Hz_10ms_envelope_correlation": 0.9969335569647636, "300–1000 Hz_10ms_modulation_mae_db": 0.5280703110653723, "300–1000 Hz_40ms_log_rms_mae_db": 0.8504812813030025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03904068388468123, "300–1000 Hz_40ms_envelope_correlation": 0.9988938454196125, "300–1000 Hz_40ms_modulation_mae_db": 0.4104643835167808, "1000–3000 Hz_10ms_log_rms_mae_db": 1.187087315983532, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1534033799631474, "1000–3000 Hz_10ms_envelope_correlation": 0.9820799416932484, "1000–3000 Hz_10ms_modulation_mae_db": 1.2216473798773986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7246499877794667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07059843080042513, "1000–3000 Hz_40ms_envelope_correlation": 0.9960139216297175, "1000–3000 Hz_40ms_modulation_mae_db": 1.7081510438591037, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2916319357337398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1736928384596098, "3000–6000 Hz_10ms_envelope_correlation": 0.9775286333367152, "3000–6000 Hz_10ms_modulation_mae_db": 0.4988501229821812, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7804308224301727, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09131363677592023, "3000–6000 Hz_40ms_envelope_correlation": 0.9937403465632576, "3000–6000 Hz_40ms_modulation_mae_db": 0.5149416916300225, "mr_stft_db": 4.438378451645484, "band_mae_db": 0.10261250227990071, "rms_gain_error_db": -0.033835621219028895, "output_peak": 0.506505133756906}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6231095088193401, "全帯域_10ms_rms_envelope_nrmse": 0.06314992402887111, "全帯域_10ms_envelope_correlation": 0.9964533940456936, "全帯域_10ms_modulation_mae_db": 0.3798742273104308, "全帯域_40ms_log_rms_mae_db": 0.3304770389453587, "全帯域_40ms_rms_envelope_nrmse": 0.030319562690438676, "全帯域_40ms_envelope_correlation": 0.9991530863346074, "全帯域_40ms_modulation_mae_db": 0.7164716042347891, "300–1000 Hz_10ms_log_rms_mae_db": 1.2329110225868884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.062092950122970655, "300–1000 Hz_10ms_envelope_correlation": 0.9972617210848458, "300–1000 Hz_10ms_modulation_mae_db": 0.12852282423668235, "300–1000 Hz_40ms_log_rms_mae_db": 0.86148542762787, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03246103509555053, "300–1000 Hz_40ms_envelope_correlation": 0.9992247573954828, "300–1000 Hz_40ms_modulation_mae_db": 0.12355340530480378, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2073362035519803, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1309260497733584, "1000–3000 Hz_10ms_envelope_correlation": 0.98709996142924, "1000–3000 Hz_10ms_modulation_mae_db": 0.9035946577332772, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7886329511887128, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08040932252817193, "1000–3000 Hz_40ms_envelope_correlation": 0.9950693372612627, "1000–3000 Hz_40ms_modulation_mae_db": 1.4152944785629462, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2604311598671505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15171609112366236, "3000–6000 Hz_10ms_envelope_correlation": 0.983069219038968, "3000–6000 Hz_10ms_modulation_mae_db": 0.6390001258598536, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7517025303594039, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08453492953687955, "3000–6000 Hz_40ms_envelope_correlation": 0.9948312939514252, "3000–6000 Hz_40ms_modulation_mae_db": 0.9292369975887441, "mr_stft_db": 4.445799727466022, "band_mae_db": 0.1562155228320504, "rms_gain_error_db": -0.0440702137219369, "output_peak": 0.49653127018625237}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6261148293059788, "全帯域_10ms_rms_envelope_nrmse": 0.06723317864324832, "全帯域_10ms_envelope_correlation": 0.9959740409839952, "全帯域_10ms_modulation_mae_db": 0.9841844869097125, "全帯域_40ms_log_rms_mae_db": 0.3049826048275635, "全帯域_40ms_rms_envelope_nrmse": 0.0333840987736415, "全帯域_40ms_envelope_correlation": 0.9989681430535985, "全帯域_40ms_modulation_mae_db": 1.3497896239532603, "300–1000 Hz_10ms_log_rms_mae_db": 1.2669138288651485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06841081894544875, "300–1000 Hz_10ms_envelope_correlation": 0.9967053784611944, "300–1000 Hz_10ms_modulation_mae_db": 0.360742008100219, "300–1000 Hz_40ms_log_rms_mae_db": 0.9247604001211867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04041004553417333, "300–1000 Hz_40ms_envelope_correlation": 0.998820798385756, "300–1000 Hz_40ms_modulation_mae_db": 0.34322163125221294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2085930985369824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13234788061790673, "1000–3000 Hz_10ms_envelope_correlation": 0.9869751760682405, "1000–3000 Hz_10ms_modulation_mae_db": 1.7470908999239272, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7781858962042223, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07558736288094021, "1000–3000 Hz_40ms_envelope_correlation": 0.9959208646604917, "1000–3000 Hz_40ms_modulation_mae_db": 2.1319112335238994, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2511013608674402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16040395999254103, "3000–6000 Hz_10ms_envelope_correlation": 0.9812025899144969, "3000–6000 Hz_10ms_modulation_mae_db": 1.3447222639825978, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7279715363167406, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09340308327804457, "3000–6000 Hz_40ms_envelope_correlation": 0.9938493334613344, "3000–6000 Hz_40ms_modulation_mae_db": 1.6552827339947649, "mr_stft_db": 4.43739942016122, "band_mae_db": 0.1603495515888327, "rms_gain_error_db": -0.03763905399971796, "output_peak": 0.5078878888781168}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6500558011380554, "全帯域_10ms_rms_envelope_nrmse": 0.06650880479047373, "全帯域_10ms_envelope_correlation": 0.9961274321667491, "全帯域_10ms_modulation_mae_db": 0.7425189891448182, "全帯域_40ms_log_rms_mae_db": 0.3546581508871762, "全帯域_40ms_rms_envelope_nrmse": 0.03471218920125947, "全帯域_40ms_envelope_correlation": 0.9989598570093582, "全帯域_40ms_modulation_mae_db": 1.1459246788741309, "300–1000 Hz_10ms_log_rms_mae_db": 1.2952444652205344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07188460415975169, "300–1000 Hz_10ms_envelope_correlation": 0.9963563633378099, "300–1000 Hz_10ms_modulation_mae_db": 0.3289548003813078, "300–1000 Hz_40ms_log_rms_mae_db": 0.9231805366331651, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04276008143528894, "300–1000 Hz_40ms_envelope_correlation": 0.9986897012482326, "300–1000 Hz_40ms_modulation_mae_db": 0.2615382590562492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2436649010465688, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13994944895045153, "1000–3000 Hz_10ms_envelope_correlation": 0.9855767700710841, "1000–3000 Hz_10ms_modulation_mae_db": 1.4550539108682077, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7987326008005722, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08313290994686094, "1000–3000 Hz_40ms_envelope_correlation": 0.9952259127391541, "1000–3000 Hz_40ms_modulation_mae_db": 1.8478994877692303, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2793273432248289, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14651188889429637, "3000–6000 Hz_10ms_envelope_correlation": 0.9846553284985254, "3000–6000 Hz_10ms_modulation_mae_db": 0.9471728244723544, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7730297217362494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08777556449692395, "3000–6000 Hz_40ms_envelope_correlation": 0.9949103646159877, "3000–6000 Hz_40ms_modulation_mae_db": 1.333690011317174, "mr_stft_db": 4.449327730363296, "band_mae_db": 0.17381809405297063, "rms_gain_error_db": -0.0891213359311045, "output_peak": 0.4797861619274673}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.4778606302154325, "spectral_convergence": 0.31236453281985066, "spectral_convergence_status": "ok", "amplitude_floor": 1.0172154037614098e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.424652085907798, "spectral_convergence": 0.2893291238938013, "spectral_convergence_status": "ok", "amplitude_floor": 7.045944280558953e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.421810099902562, "spectral_convergence": 0.28103562157632717, "spectral_convergence_status": "ok", "amplitude_floor": 6.386940858702166e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00026428484162924415, "output_power": 0.0002751608413980215, "delta_db": 0.17514374412348122, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008055137994793597, "output_power": 0.0008119580369892364, "delta_db": 0.034606000993887234, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006109316451161528, "output_power": 0.0006118369971670437, "delta_db": 0.00643113436402195, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027595006700553216, "output_power": 0.0002644496214131416, "delta_db": -0.1848755423745124, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003628265208503686, "output_power": 0.00033246514390177477, "delta_db": -0.37952904357169315, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001954243236049976, "output_power": 0.00019406291943809643, "delta_db": -0.030360569482064226, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13515767248494334, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04338349496720162, "pitch": null, "output_rms_dbfs": -26.038122411680686, "output_peak": 0.4796792258142698, "reference_rms_dbfs": -25.994738916713484, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.441440938675264}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017615675305240148, "log_rms_mae_db": 0.6289412106791312, "rms_envelope_nrmse": 0.0711921530589569, "envelope_correlation": 0.9954850338090805, "rms_gain_error_db": -0.04316612140734559, "reference_mean_power": 0.0025168568603362385, "output_mean_power": 0.002491964807447293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0162973895242792e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002469993456433397, "output_power": 0.00024426473415836204, "reference_active": true, "delta_db": -0.04835032698755265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010173002443957726, "output_power": 0.00010276038853335157, "reference_active": true, "delta_db": 0.04376588816826299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1084315450169838e-05, "output_power": 1.0693500293368166e-05, "reference_active": true, "delta_db": -2.9483961954720406}], "active_band_count": 3, "active_band_mae_db": 1.013504136875952}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016525538510559444, "log_rms_mae_db": 0.3413488004011963, "rms_envelope_nrmse": 0.038612229832562635, "envelope_correlation": 0.9986123824977522, "rms_gain_error_db": -0.04322021141990654, "reference_mean_power": 0.002527970677484516, "output_mean_power": 0.002502937533954724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.625293642327984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023064746753792456, "output_power": 0.00022720122128888946, "reference_active": true, "delta_db": -0.06538028970312881}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.045652222183457e-05, "output_power": 5.830061351435344e-05, "reference_active": true, "delta_db": -0.1577003565961911}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2874374178901008e-06, "output_power": 9.480557764791047e-07, "reference_active": true, "delta_db": -3.825153323033441}], "active_band_count": 3, "active_band_mae_db": 1.3494113231109204}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001279910087199623, "log_rms_mae_db": 1.3178911945139378, "rms_envelope_nrmse": 0.06535201095005795, "envelope_correlation": 0.9969673841743909, "rms_gain_error_db": -0.002947204175870578, "reference_mean_power": 0.0009939686093891967, "output_mean_power": 0.0009932943123855632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.147046828130101e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013887130907316313, "output_power": 0.00013555255916702208, "reference_active": true, "delta_db": -0.10504808142834021}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4048550736300918e-05, "output_power": 2.4191287186281808e-05, "reference_active": true, "delta_db": 0.025700680670144842}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.939409383024551e-06, "output_power": 4.207639392530269e-06, "reference_active": true, "delta_db": -0.6963651001477409}], "active_band_count": 3, "active_band_mae_db": 0.27570462074874197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011859895721274409, "log_rms_mae_db": 0.9349425580505859, "rms_envelope_nrmse": 0.03881257764562292, "envelope_correlation": 0.9988914147469561, "rms_gain_error_db": -0.0029951000437432687, "reference_mean_power": 0.0009986896907851761, "output_mean_power": 0.0009980011846055683}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.009742136856967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012678804378904424, "output_power": 0.0001236088259947658, "reference_active": true, "delta_db": -0.1102881954343494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.298881565077983e-05, "output_power": 1.358611556399589e-05, "reference_active": true, "delta_db": 0.19525751457653132}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8039990598306815e-07, "output_power": 1.9201212297090197e-07, "reference_active": true, "delta_db": -1.6444921428041308}], "active_band_count": 3, "active_band_mae_db": 0.6500126176050038}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.31629258809413e-05, "log_rms_mae_db": 1.206774482776524, "rms_envelope_nrmse": 0.1417887783500988, "envelope_correlation": 0.9848930448399501, "rms_gain_error_db": -0.2287933351762627, "reference_mean_power": 0.0004146029386445929, "output_mean_power": 0.0003933263504670957}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1875762004578613e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604229953230037e-05, "output_power": 8.133426522624859e-05, "reference_active": true, "delta_db": -0.24438461045893667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2897806527831664e-05, "output_power": 2.1305213983656766e-05, "reference_active": true, "delta_db": -0.31307981041062743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2516459896674564e-05, "output_power": 4.510301258490484e-06, "reference_active": true, "delta_db": -4.432759613369852}], "active_band_count": 3, "active_band_mae_db": 1.6634080114131387}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.786085823827924e-05, "log_rms_mae_db": 0.813351399883142, "rms_envelope_nrmse": 0.08653455312985248, "envelope_correlation": 0.9943760247889739, "rms_gain_error_db": -0.2287903460844777, "reference_mean_power": 0.0004166837249389974, "output_mean_power": 0.00039530062706871045}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.035015228514449e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.632173919031572e-05, "output_power": 7.832608099947436e-05, "reference_active": true, "delta_db": -0.4221378441620588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4669499526204761e-05, "output_power": 1.2130132054671782e-05, "reference_active": true, "delta_db": -0.8254976859581099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2709360758910173e-06, "output_power": 3.935374142171163e-07, "reference_active": true, "delta_db": -5.091376797362423}], "active_band_count": 3, "active_band_mae_db": 2.113004109160864}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011544071276927888, "log_rms_mae_db": 1.3015906229673397, "rms_envelope_nrmse": 0.16360688855894812, "envelope_correlation": 0.9801646751931736, "rms_gain_error_db": -0.248945566313172, "reference_mean_power": 0.00033862428581899314, "output_mean_power": 0.00031975956574777537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7742895295757852e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.9949103091747194e-05, "output_power": 5.76198276607841e-05, "reference_active": true, "delta_db": -0.1721073494903379}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5523159384577664e-05, "output_power": 4.041805277329202e-05, "reference_active": true, "delta_db": -0.5165700859599989}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.055143212029728e-06, "output_power": 4.314421993574875e-06, "reference_active": true, "delta_db": -2.135832127204078}], "active_band_count": 3, "active_band_mae_db": 0.9415031875514717}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.669712818495073e-05, "log_rms_mae_db": 0.7739801628962564, "rms_envelope_nrmse": 0.0955466279041648, "envelope_correlation": 0.9931429647422816, "rms_gain_error_db": -0.24892883337623548, "reference_mean_power": 0.00034032151520785135, "output_mean_power": 0.0003213634808616394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6870888960608118e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.86684133644439e-05, "output_power": 6.429344051873747e-05, "reference_active": true, "delta_db": -0.2859034613321412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.573388989930172e-05, "output_power": 2.2166951410482533e-05, "reference_active": true, "delta_db": -0.647994692126906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.935502106463895e-07, "output_power": 4.2144003500017006e-07, "reference_active": true, "delta_db": -3.2638317218612833}], "active_band_count": 3, "active_band_mae_db": 1.3992432917734436}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8842701531269358, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "de850ec78b0baeb2476a1e85c5e2584bbb962109d0dcbad65d1fa8b9af7084ae", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9854195233931555, "standard_deviation": 0.0019539870060265078, "minimum": 0.9820799416932484, "maximum": 0.9879689911115872, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2090596324462926, "standard_deviation": 0.017729412628496716, "minimum": 1.187087315983532, "maximum": 1.2436649010465688, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3051066472637856, "standard_deviation": 0.3047715633512274, "minimum": 0.9035946577332772, "maximum": 1.7470908999239272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13907860109431697, "standard_deviation": 0.008996207690521845, "minimum": 0.12636827226636252, "maximum": 0.1534033799631474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953240121075523, "standard_deviation": 0.0007090654633540122, "minimum": 0.9943760247889739, "maximum": 0.996373595385027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7940229754196059, "standard_deviation": 0.03166609041106181, "minimum": 0.7246499877794667, "maximum": 0.8192172311952307, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.6782194498646272, "standard_deviation": 0.42758569105541694, "minimum": 1.0681717858227584, "maximum": 2.1319112335238994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07898860335748392, "standard_deviation": 0.006504929510909086, "minimum": 0.06935327384071711, "maximum": 0.08653455312985248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9804308691575243, "standard_deviation": 0.0026984097860620125, "minimum": 0.9771476005838401, "maximum": 0.9846553284985254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2765512888152055, "standard_deviation": 0.01990560289147079, "minimum": 1.2511013608674402, "maximum": 1.301971049048132, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7896867096525388, "standard_deviation": 0.3409761978994054, "minimum": 0.42281102754129013, "maximum": 1.3447222639825978, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16282582057040518, "standard_deviation": 0.010838330460582371, "minimum": 0.14651188889429637, "maximum": 0.17652875899224393, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927470024143794, "standard_deviation": 0.002135061017828152, "minimum": 0.9894311998932684, "maximum": 0.9949103646159877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.773724171476099, "standard_deviation": 0.026246076670082445, "minimum": 0.7279715363167406, "maximum": 0.8192155789990555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.024903148382728, "standard_deviation": 0.4828619752319562, "minimum": 0.2658765989930078, "maximum": 1.6552827339947649, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09851684514105169, "standard_deviation": 0.013170573743931439, "minimum": 0.08453492953687955, "maximum": 0.11959545820379017, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9968923586849934, "standard_deviation": 0.0003379043136137081, "minimum": 0.9963563633378099, "maximum": 0.9974282502464077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2665980157434409, "standard_deviation": 0.034788397387862616, "minimum": 1.223760424765824, "maximum": 1.3178911945139378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2880659020079667, "standard_deviation": 0.12447995564086217, "minimum": 0.12852282423668235, "maximum": 0.5280703110653723, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06620159720094256, "standard_deviation": 0.0036044624390745997, "minimum": 0.060616914897697445, "maximum": 0.07188460415975169, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989404565021927, "standard_deviation": 0.00017997093099422656, "minimum": 0.9986897012482326, "maximum": 0.9992247573954828, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8940852838687416, "standard_deviation": 0.03397973356009267, "minimum": 0.8504812813030025, "maximum": 0.9349425580505859, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4032603965262328, "standard_deviation": 0.21098594025043113, "minimum": 0.12355340530480378, "maximum": 0.7603703744927489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.038107348098571255, "standard_deviation": 0.0033150327587316173, "minimum": 0.03246103509555053, "maximum": 0.04276008143528894, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1453147099616318, "standard_deviation": 0.03517912781095915, "minimum": 0.08393333956578948, "maximum": 0.17862498957146258, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.442781402611157, "standard_deviation": 0.004156717882912984, "minimum": 4.43739942016122, "maximum": 4.449327730363296, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4937939831454097, "standard_deviation": 0.01369585086226898, "minimum": 0.4743646170672833, "maximum": 0.5078878888781168, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.052415595705862225, "standard_deviation": 0.01992215906624365, "minimum": -0.0891213359311045, "maximum": -0.033835621219028895, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955375529179911, "standard_deviation": 0.0006927715210626942, "minimum": 0.9945440078372658, "maximum": 0.9964533940456936, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6433176733255613, "standard_deviation": 0.017846866407918768, "minimum": 0.6231095088193401, "maximum": 0.6764835618523974, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7087660361260075, "standard_deviation": 0.2118113617585927, "minimum": 0.3798742273104308, "maximum": 1.013504136875952, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07070253912569845, "standard_deviation": 0.00542135347213955, "minimum": 0.06314992402887111, "maximum": 0.07823900188187119, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986712123169719, "standard_deviation": 0.00041860725648723275, "minimum": 0.997984670467062, "maximum": 0.9991530863346074, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3459182782147368, "standard_deviation": 0.024507327355526354, "minimum": 0.3049826048275635, "maximum": 0.3875897762259138, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9742960633586423, "standard_deviation": 0.2935242445701357, "minimum": 0.6178122732856991, "maximum": 1.3497896239532603, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037630325869577326, "standard_deviation": 0.00570770292912841, "minimum": 0.030319562690438676, "maximum": 0.046401447318890295, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-123852-0000", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.7069318294525146, "predicted_mos": 2.7069318294525146, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e5e4a056b4e70779f88ac17e12f6cfab36019f3be53d3bce8ebaec52ae53e514", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.743422799743712, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6837921581778765, "全帯域_10ms_rms_envelope_nrmse": 0.07141848773992496, "全帯域_10ms_envelope_correlation": 0.9954629202829214, "全帯域_10ms_modulation_mae_db": 0.7104865040537572, "全帯域_40ms_log_rms_mae_db": 0.36793295778130203, "全帯域_40ms_rms_envelope_nrmse": 0.03918416365060695, "全帯域_40ms_envelope_correlation": 0.9985739843774077, "全帯域_40ms_modulation_mae_db": 0.7791303010415381, "300–1000 Hz_10ms_log_rms_mae_db": 1.0121707658241836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06487580208605231, "300–1000 Hz_10ms_envelope_correlation": 0.9970060722012076, "300–1000 Hz_10ms_modulation_mae_db": 0.6547484057365246, "300–1000 Hz_40ms_log_rms_mae_db": 0.6735963363451364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03363838611131392, "300–1000 Hz_40ms_envelope_correlation": 0.9991692486771308, "300–1000 Hz_40ms_modulation_mae_db": 0.3541857308499365, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9197175599892354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13152567001071036, "1000–3000 Hz_10ms_envelope_correlation": 0.9873570687937483, "1000–3000 Hz_10ms_modulation_mae_db": 1.866797568155265, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6075590236123549, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07372223317305754, "1000–3000 Hz_40ms_envelope_correlation": 0.9964444150022137, "1000–3000 Hz_40ms_modulation_mae_db": 2.475153888901763, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9592984693934616, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14617595675584866, "3000–6000 Hz_10ms_envelope_correlation": 0.9842502056930968, "3000–6000 Hz_10ms_modulation_mae_db": 0.30368212446381027, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5973398905777869, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09161875845786328, "3000–6000 Hz_40ms_envelope_correlation": 0.993640553718129, "3000–6000 Hz_40ms_modulation_mae_db": 0.2401860791080653, "mr_stft_db": 4.149961469902602, "band_mae_db": 0.17321443299219322, "rms_gain_error_db": -0.04805343494607803, "output_peak": 0.4903423815409364}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7323420995563569, "全帯域_10ms_rms_envelope_nrmse": 0.08074860566320102, "全帯域_10ms_envelope_correlation": 0.9942583305682423, "全帯域_10ms_modulation_mae_db": 1.0745356656851737, "全帯域_40ms_log_rms_mae_db": 0.4114062533175461, "全帯域_40ms_rms_envelope_nrmse": 0.045390773317170056, "全帯域_40ms_envelope_correlation": 0.99818002786857, "全帯域_40ms_modulation_mae_db": 1.3661119465485665, "300–1000 Hz_10ms_log_rms_mae_db": 0.9541981357943745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05218360479298882, "300–1000 Hz_10ms_envelope_correlation": 0.9980805580428219, "300–1000 Hz_10ms_modulation_mae_db": 0.38436865391093283, "300–1000 Hz_40ms_log_rms_mae_db": 0.5813954841746267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.027616106205672755, "300–1000 Hz_40ms_envelope_correlation": 0.999448780731285, "300–1000 Hz_40ms_modulation_mae_db": 0.3398755605107491, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0104799230963941, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1390053896931046, "1000–3000 Hz_10ms_envelope_correlation": 0.9862126658141048, "1000–3000 Hz_10ms_modulation_mae_db": 1.2860731112481234, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6489759282757003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08200356197731434, "1000–3000 Hz_40ms_envelope_correlation": 0.9959499535860261, "1000–3000 Hz_40ms_modulation_mae_db": 1.7539547401648967, "3000–6000 Hz_10ms_log_rms_mae_db": 1.012536664598299, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17599224257402815, "3000–6000 Hz_10ms_envelope_correlation": 0.97823301078546, "3000–6000 Hz_10ms_modulation_mae_db": 1.5424056142099982, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6180705322526273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10981224649391613, "3000–6000 Hz_40ms_envelope_correlation": 0.9926218340325415, "3000–6000 Hz_40ms_modulation_mae_db": 1.8571276973229196, "mr_stft_db": 4.162152584688478, "band_mae_db": 0.25551813400766404, "rms_gain_error_db": -0.10843266294095306, "output_peak": 0.5012032887136794}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6337845808361842, "全帯域_10ms_rms_envelope_nrmse": 0.06457729875246669, "全帯域_10ms_envelope_correlation": 0.9963012128686253, "全帯域_10ms_modulation_mae_db": 0.7293428863543188, "全帯域_40ms_log_rms_mae_db": 0.3497889500458767, "全帯域_40ms_rms_envelope_nrmse": 0.03370192325696523, "全帯域_40ms_envelope_correlation": 0.9989620802047866, "全帯域_40ms_modulation_mae_db": 1.034596800116461, "300–1000 Hz_10ms_log_rms_mae_db": 1.0426412686915303, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0647137056856535, "300–1000 Hz_10ms_envelope_correlation": 0.997016526503048, "300–1000 Hz_10ms_modulation_mae_db": 0.12664885677672102, "300–1000 Hz_40ms_log_rms_mae_db": 0.6717593218175204, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0319514198350538, "300–1000 Hz_40ms_envelope_correlation": 0.9992491697338388, "300–1000 Hz_40ms_modulation_mae_db": 0.2773277422488275, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9343243527116254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13339879782631572, "1000–3000 Hz_10ms_envelope_correlation": 0.9867523365217912, "1000–3000 Hz_10ms_modulation_mae_db": 1.5905262401907254, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6072414513539477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07969612621812529, "1000–3000 Hz_40ms_envelope_correlation": 0.9954046803770286, "1000–3000 Hz_40ms_modulation_mae_db": 2.283156605802379, "3000–6000 Hz_10ms_log_rms_mae_db": 0.956456854331207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1310475598447238, "3000–6000 Hz_10ms_envelope_correlation": 0.9873845923257293, "3000–6000 Hz_10ms_modulation_mae_db": 0.317445166876614, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6005667342204662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0771898847137308, "3000–6000 Hz_40ms_envelope_correlation": 0.9955702057710143, "3000–6000 Hz_40ms_modulation_mae_db": 0.40371727312622, "mr_stft_db": 4.134586026873606, "band_mae_db": 0.12472353926160402, "rms_gain_error_db": -0.05592556165892341, "output_peak": 0.5238817074805387}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6828259257788726, "全帯域_10ms_rms_envelope_nrmse": 0.07774183507394071, "全帯域_10ms_envelope_correlation": 0.9946283103733492, "全帯域_10ms_modulation_mae_db": 1.0672448622265351, "全帯域_40ms_log_rms_mae_db": 0.340681679402256, "全帯域_40ms_rms_envelope_nrmse": 0.03903596906261379, "全帯域_40ms_envelope_correlation": 0.9986109313803415, "全帯域_40ms_modulation_mae_db": 1.5860475167737234, "300–1000 Hz_10ms_log_rms_mae_db": 0.9752455788634964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06796356676909637, "300–1000 Hz_10ms_envelope_correlation": 0.9967256166910209, "300–1000 Hz_10ms_modulation_mae_db": 0.17422485030828574, "300–1000 Hz_40ms_log_rms_mae_db": 0.6559944242152048, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033784253821081744, "300–1000 Hz_40ms_envelope_correlation": 0.9991683753117818, "300–1000 Hz_40ms_modulation_mae_db": 0.15023982132133487, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9664307575657362, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13689061721937051, "1000–3000 Hz_10ms_envelope_correlation": 0.986037527324668, "1000–3000 Hz_10ms_modulation_mae_db": 1.1450414125362194, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5811231819236388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07005931595530242, "1000–3000 Hz_40ms_envelope_correlation": 0.9966210093003701, "1000–3000 Hz_40ms_modulation_mae_db": 1.7670601508407422, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9676166189481717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1584160427206505, "3000–6000 Hz_10ms_envelope_correlation": 0.981859935117682, "3000–6000 Hz_10ms_modulation_mae_db": 1.7269150247828577, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5980605898142742, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09544098054119914, "3000–6000 Hz_40ms_envelope_correlation": 0.9937741157914998, "3000–6000 Hz_40ms_modulation_mae_db": 2.0196190403473433, "mr_stft_db": 4.142002521487797, "band_mae_db": 0.17851323790917847, "rms_gain_error_db": -0.0688135619857427, "output_peak": 0.5077648288586714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.693689218388972, "全帯域_10ms_rms_envelope_nrmse": 0.0702631697205328, "全帯域_10ms_envelope_correlation": 0.9956151626374302, "全帯域_10ms_modulation_mae_db": 0.3472917009255883, "全帯域_40ms_log_rms_mae_db": 0.37605451476316476, "全帯域_40ms_rms_envelope_nrmse": 0.041232748242993285, "全帯域_40ms_envelope_correlation": 0.9984318290274765, "全帯域_40ms_modulation_mae_db": 0.5941245236102772, "300–1000 Hz_10ms_log_rms_mae_db": 1.0746486142707448, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06389572386103745, "300–1000 Hz_10ms_envelope_correlation": 0.9971427673175683, "300–1000 Hz_10ms_modulation_mae_db": 0.29693528569723776, "300–1000 Hz_40ms_log_rms_mae_db": 0.7151584700024415, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03702240606638671, "300–1000 Hz_40ms_envelope_correlation": 0.9990280602348255, "300–1000 Hz_40ms_modulation_mae_db": 0.30239092685191227, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9883341637845314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13525185206837118, "1000–3000 Hz_10ms_envelope_correlation": 0.9863152584152584, "1000–3000 Hz_10ms_modulation_mae_db": 0.30633804587332364, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6721833408695533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07598888863630505, "1000–3000 Hz_40ms_envelope_correlation": 0.9957911109180451, "1000–3000 Hz_40ms_modulation_mae_db": 0.44420015337684315, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9774980106581017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1525624465549676, "3000–6000 Hz_10ms_envelope_correlation": 0.9830376415457216, "3000–6000 Hz_10ms_modulation_mae_db": 0.7099425232870109, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6272544755673075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10592625446222032, "3000–6000 Hz_40ms_envelope_correlation": 0.9916901963964231, "3000–6000 Hz_40ms_modulation_mae_db": 1.022501266850661, "mr_stft_db": 4.144960025431714, "band_mae_db": 0.16518356393285247, "rms_gain_error_db": -0.058223151854861425, "output_peak": 0.496069999305719}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6678914537373365, "全帯域_10ms_rms_envelope_nrmse": 0.07837035960748329, "全帯域_10ms_envelope_correlation": 0.9946383447461906, "全帯域_10ms_modulation_mae_db": 1.2147113720893272, "全帯域_40ms_log_rms_mae_db": 0.36489604202548986, "全帯域_40ms_rms_envelope_nrmse": 0.042898572931517394, "全帯域_40ms_envelope_correlation": 0.9984305101827744, "全帯域_40ms_modulation_mae_db": 1.69012644702092, "300–1000 Hz_10ms_log_rms_mae_db": 0.9677952132172912, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06599537558748812, "300–1000 Hz_10ms_envelope_correlation": 0.9969199649263856, "300–1000 Hz_10ms_modulation_mae_db": 0.3497583106239439, "300–1000 Hz_40ms_log_rms_mae_db": 0.641981503868529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037444159019552316, "300–1000 Hz_40ms_envelope_correlation": 0.9990016391575446, "300–1000 Hz_40ms_modulation_mae_db": 0.3352049302643863, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9268079081107963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13611800545055552, "1000–3000 Hz_10ms_envelope_correlation": 0.9860029385495761, "1000–3000 Hz_10ms_modulation_mae_db": 0.8425408210147064, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5915591447786188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07169874903524696, "1000–3000 Hz_40ms_envelope_correlation": 0.996134735572866, "1000–3000 Hz_40ms_modulation_mae_db": 1.496349025483033, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9706308052114031, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.172813858463189, "3000–6000 Hz_10ms_envelope_correlation": 0.9799290304090504, "3000–6000 Hz_10ms_modulation_mae_db": 2.1560953229814115, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5803385319862302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1073065035267463, "3000–6000 Hz_40ms_envelope_correlation": 0.9941413981098445, "3000–6000 Hz_40ms_modulation_mae_db": 2.3076290833460553, "mr_stft_db": 4.140736325605713, "band_mae_db": 0.21666928330229843, "rms_gain_error_db": -0.12278441317633904, "output_peak": 0.48501577312127825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6934191100207726, "全帯域_10ms_rms_envelope_nrmse": 0.0759306675301967, "全帯域_10ms_envelope_correlation": 0.9948643511638345, "全帯域_10ms_modulation_mae_db": 1.004848049998943, "全帯域_40ms_log_rms_mae_db": 0.4124074491653145, "全帯域_40ms_rms_envelope_nrmse": 0.04333351417921469, "全帯域_40ms_envelope_correlation": 0.998254923812352, "全帯域_40ms_modulation_mae_db": 1.3301581146596961, "300–1000 Hz_10ms_log_rms_mae_db": 0.9664760717841288, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.060721667620559173, "300–1000 Hz_10ms_envelope_correlation": 0.9974151491776416, "300–1000 Hz_10ms_modulation_mae_db": 0.5614284372201933, "300–1000 Hz_40ms_log_rms_mae_db": 0.6394818167874049, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03472752465089981, "300–1000 Hz_40ms_envelope_correlation": 0.9991419842941996, "300–1000 Hz_40ms_modulation_mae_db": 0.6314484916754376, "1000–3000 Hz_10ms_log_rms_mae_db": 0.8979204735865597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12305755023749271, "1000–3000 Hz_10ms_envelope_correlation": 0.9885702287728385, "1000–3000 Hz_10ms_modulation_mae_db": 1.6201733763996398, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5878016132193101, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07069570493352109, "1000–3000 Hz_40ms_envelope_correlation": 0.9961590469224205, "1000–3000 Hz_40ms_modulation_mae_db": 2.430910429736946, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9528026301798331, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17282954603288786, "3000–6000 Hz_10ms_envelope_correlation": 0.9785550708304876, "3000–6000 Hz_10ms_modulation_mae_db": 0.7389790747317967, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5920207849136578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10919468452466975, "3000–6000 Hz_40ms_envelope_correlation": 0.9920177395601762, "3000–6000 Hz_40ms_modulation_mae_db": 1.1170982718047042, "mr_stft_db": 4.1418183853772845, "band_mae_db": 0.18640073770656276, "rms_gain_error_db": -0.05114937622986207, "output_peak": 0.48555898937984776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6676570928460428, "全帯域_10ms_rms_envelope_nrmse": 0.06946199043795812, "全帯域_10ms_envelope_correlation": 0.9957112592562273, "全帯域_10ms_modulation_mae_db": 0.5896176283163596, "全帯域_40ms_log_rms_mae_db": 0.37255769108397485, "全帯域_40ms_rms_envelope_nrmse": 0.03832314041323108, "全帯域_40ms_envelope_correlation": 0.9986430654598558, "全帯域_40ms_modulation_mae_db": 0.8416027745920647, "300–1000 Hz_10ms_log_rms_mae_db": 1.0095983136913709, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.059626282833792565, "300–1000 Hz_10ms_envelope_correlation": 0.997479503822355, "300–1000 Hz_10ms_modulation_mae_db": 0.34722345632062596, "300–1000 Hz_40ms_log_rms_mae_db": 0.6282096742297714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031201584025788768, "300–1000 Hz_40ms_envelope_correlation": 0.999297712929342, "300–1000 Hz_40ms_modulation_mae_db": 0.4566503363975041, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9453876600190376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12436931712928179, "1000–3000 Hz_10ms_envelope_correlation": 0.988357560152661, "1000–3000 Hz_10ms_modulation_mae_db": 0.3495880345954668, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6169134975710168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06889153832921449, "1000–3000 Hz_40ms_envelope_correlation": 0.9964077896567314, "1000–3000 Hz_40ms_modulation_mae_db": 0.46826689235286717, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0103187598266703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1626240963227638, "3000–6000 Hz_10ms_envelope_correlation": 0.9803062470109717, "3000–6000 Hz_10ms_modulation_mae_db": 1.278175089530803, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6381414899912016, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09826039543055036, "3000–6000 Hz_40ms_envelope_correlation": 0.9924590372827181, "3000–6000 Hz_40ms_modulation_mae_db": 1.5923837582121367, "mr_stft_db": 4.133050242771371, "band_mae_db": 0.11596202815106448, "rms_gain_error_db": -0.05248358714551273, "output_peak": 0.5210644974956021}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.123774520835289, "spectral_convergence": 0.30395103046246524, "spectral_convergence_status": "ok", "amplitude_floor": 1.0172154037614098e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.053733756139142, "spectral_convergence": 0.27528952579834887, "spectral_convergence_status": "ok", "amplitude_floor": 7.045944280558953e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.272376132733377, "spectral_convergence": 0.27892338660283017, "spectral_convergence_status": "ok", "amplitude_floor": 6.386940858702166e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00026428484162924415, "output_power": 0.00027751103790005807, "delta_db": 0.2120800733585878, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008055137994793597, "output_power": 0.0008140674201014788, "delta_db": 0.04587389296652271, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006109316451161528, "output_power": 0.000607902734904617, "delta_db": -0.02158524112093913, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027595006700553216, "output_power": 0.00026406549552561, "delta_db": -0.19118846616189006, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003628265208503686, "output_power": 0.0003416287853495802, "delta_db": -0.26144567517602707, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001954243236049976, "output_power": 0.0001820821163415863, "delta_db": -0.3071132491691925, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17321443299219322, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04805343494607803, "pitch": null, "output_rms_dbfs": -26.042792351659564, "output_peak": 0.4903423815409364, "reference_rms_dbfs": -25.994738916713484, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.149961469902602}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017615675305240148, "log_rms_mae_db": 0.6837921581778765, "rms_envelope_nrmse": 0.07141848773992496, "envelope_correlation": 0.9954629202829214, "rms_gain_error_db": -0.048059834158621736, "reference_mean_power": 0.0025168568603362385, "output_mean_power": 0.0024891583956033786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0162973895242792e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002469993456433397, "output_power": 0.0002331146077603565, "reference_active": true, "delta_db": -0.251263139635804}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010173002443957726, "output_power": 0.00010535146109710868, "reference_active": true, "delta_db": 0.15191414291625177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1084315450169838e-05, "output_power": 1.4162218229650689e-05, "reference_active": true, "delta_db": -1.7282822296092157}], "active_band_count": 3, "active_band_mae_db": 0.7104865040537572}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016525538510559444, "log_rms_mae_db": 0.36793295778130203, "rms_envelope_nrmse": 0.03918416365060695, "envelope_correlation": 0.9985739843774077, "rms_gain_error_db": -0.04805415101590335, "reference_mean_power": 0.002527970677484516, "output_mean_power": 0.002500153174863068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.625293642327984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023064746753792456, "output_power": 0.0002175063463521773, "reference_active": true, "delta_db": -0.25476757281687973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.045652222183457e-05, "output_power": 6.0085186331759163e-05, "reference_active": true, "delta_db": -0.02675748337569242}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2874374178901008e-06, "output_power": 1.4248286405722786e-06, "reference_active": true, "delta_db": -2.055865846932042}], "active_band_count": 3, "active_band_mae_db": 0.7791303010415381}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001279910087199623, "log_rms_mae_db": 1.0121707658241836, "rms_envelope_nrmse": 0.06487580208605231, "envelope_correlation": 0.9970060722012076, "rms_gain_error_db": 0.0008063977290353327, "reference_mean_power": 0.0009939686093891967, "output_mean_power": 0.0009941531865555544}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.147046828130101e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013887130907316313, "output_power": 0.00013430990051885896, "reference_active": true, "delta_db": -0.14504501978934353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4048550736300918e-05, "output_power": 2.659003572367362e-05, "reference_active": true, "delta_db": 0.4363001166334812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.939409383024551e-06, "output_power": 6.791489451393148e-06, "reference_active": true, "delta_db": 1.382900080786749}], "active_band_count": 3, "active_band_mae_db": 0.6547484057365246}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011859895721274409, "log_rms_mae_db": 0.6735963363451364, "rms_envelope_nrmse": 0.03363838611131392, "envelope_correlation": 0.9991692486771308, "rms_gain_error_db": 0.0007458806550166111, "reference_mean_power": 0.0009986896907851761, "output_mean_power": 0.0009988612258431317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.009742136856967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012678804378904424, "output_power": 0.0001240219668626473, "reference_active": true, "delta_db": -0.09579686634120634}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.298881565077983e-05, "output_power": 1.4401582593508478e-05, "reference_active": true, "delta_db": 0.4484066650926552}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8039990598306815e-07, "output_power": 3.1594626800282134e-07, "reference_active": true, "delta_db": 0.5183536611159479}], "active_band_count": 3, "active_band_mae_db": 0.3541857308499365}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.31629258809413e-05, "log_rms_mae_db": 0.9197175599892354, "rms_envelope_nrmse": 0.13152567001071036, "envelope_correlation": 0.9873570687937483, "rms_gain_error_db": -0.3010499607033968, "reference_mean_power": 0.0004146029386445929, "output_mean_power": 0.0003868364418058022}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1875762004578613e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604229953230037e-05, "output_power": 7.863483103453925e-05, "reference_active": true, "delta_db": -0.39097050712147413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2897806527831664e-05, "output_power": 1.927185938077012e-05, "reference_active": true, "delta_db": -0.7487026340803492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2516459896674564e-05, "output_power": 4.481357136436426e-06, "reference_active": true, "delta_db": -4.4607195632639725}], "active_band_count": 3, "active_band_mae_db": 1.866797568155265}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.786085823827924e-05, "log_rms_mae_db": 0.6075590236123549, "rms_envelope_nrmse": 0.07372223317305754, "envelope_correlation": 0.9964444150022137, "rms_gain_error_db": -0.3010395043950965, "reference_mean_power": 0.0004166837249389974, "output_mean_power": 0.00038877881118551454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.035015228514449e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.632173919031572e-05, "output_power": 7.559868734576492e-05, "reference_active": true, "delta_db": -0.5760592711221081}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4669499526204761e-05, "output_power": 1.0480112103933725e-05, "reference_active": true, "delta_db": -1.4604936918962808}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2709360758910173e-06, "output_power": 3.6747926332005625e-07, "reference_active": true, "delta_db": -5.3889087036869}], "active_band_count": 3, "active_band_mae_db": 2.475153888901763}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011544071276927888, "log_rms_mae_db": 0.9592984693934616, "rms_envelope_nrmse": 0.14617595675584866, "envelope_correlation": 0.9842502056930968, "rms_gain_error_db": -0.20649826106731964, "reference_mean_power": 0.00033862428581899314, "output_mean_power": 0.00032290017172177194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7742895295757852e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.9949103091747194e-05, "output_power": 6.269529557897199e-05, "reference_active": true, "delta_db": 0.19452264282668086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5523159384577664e-05, "output_power": 4.907864401644013e-05, "reference_active": true, "delta_db": 0.3266015998196626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.055143212029728e-06, "output_power": 7.717880448472292e-06, "reference_active": true, "delta_db": 0.3899221307450874}], "active_band_count": 3, "active_band_mae_db": 0.30368212446381027}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.669712818495073e-05, "log_rms_mae_db": 0.5973398905777869, "rms_envelope_nrmse": 0.09161875845786328, "envelope_correlation": 0.993640553718129, "rms_gain_error_db": -0.20650079888792403, "reference_mean_power": 0.00034032151520785135, "output_mean_power": 0.00032451840015193853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6870888960608118e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.86684133644439e-05, "output_power": 7.376516642993977e-05, "reference_active": true, "delta_db": 0.3109431380358975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.573388989930172e-05, "output_power": 2.8080974662320045e-05, "reference_active": true, "delta_db": 0.3790673911571823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.935502106463895e-07, "output_power": 8.872871474689337e-07, "reference_active": true, "delta_db": -0.030547708131116118}], "active_band_count": 3, "active_band_mae_db": 0.2401860791080653}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8953767258630387, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d78d7fd6cc87295bad166ff197be744ab1339a085c2e8dec71583439bffc4faa", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9869506980430808, "standard_deviation": 0.0010342166925634048, "minimum": 0.9860029385495761, "maximum": 0.9885702287728385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.9486753498579895, "standard_deviation": 0.03744984720483756, "minimum": 0.8979204735865597, "maximum": 1.0104799230963941, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1258848262516838, "standard_deviation": 0.5842906842838105, "minimum": 0.30633804587332364, "maximum": 1.866797568155265, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1324521499544003, "standard_deviation": 0.00584645153405044, "minimum": 0.12305755023749271, "maximum": 0.1390053896931046, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961140926669627, "standard_deviation": 0.0003944810152904468, "minimum": 0.9954046803770286, "maximum": 0.9966210093003701, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6141696477005176, "standard_deviation": 0.03156596656273648, "minimum": 0.5811231819236388, "maximum": 0.6721833408695533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.639881485832434, "standard_deviation": 0.8097030274996383, "minimum": 0.44420015337684315, "maximum": 2.475153888901763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0740945147822609, "standard_deviation": 0.004755503872265499, "minimum": 0.06889153832921449, "maximum": 0.08200356197731434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9816944667147749, "standard_deviation": 0.003108105361724613, "minimum": 0.97823301078546, "maximum": 0.9873845923257293, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 0.9758948516433935, "standard_deviation": 0.023339003794599533, "minimum": 0.9528026301798331, "maximum": 1.012536664598299, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0967049926080379, "standard_deviation": 0.6829282729865025, "minimum": 0.30368212446381027, "maximum": 2.1560953229814115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15905771865863244, "standard_deviation": 0.01545269673942082, "minimum": 0.1310475598447238, "maximum": 0.17599224257402815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932393850827932, "standard_deviation": 0.001285964959757954, "minimum": 0.9916901963964231, "maximum": 0.9955702057710143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.606474128665444, "standard_deviation": 0.019457679265565207, "minimum": 0.5803385319862302, "maximum": 0.6381414899912016, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3200328087647633, "standard_deviation": 0.7524079270766756, "minimum": 0.2401860791080653, "maximum": 2.3076290833460553, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09934371351886201, "standard_deviation": 0.011213764797812567, "minimum": 0.0771898847137308, "maximum": 0.10981224649391613, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9972232698352561, "standard_deviation": 0.0004265253972885845, "minimum": 0.9967256166910209, "maximum": 0.9980805580428219, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.00034674526714, "standard_deviation": 0.04226038486344439, "minimum": 0.9541981357943745, "maximum": 1.0746486142707448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.36191703207430814, "standard_deviation": 0.17777390212979496, "minimum": 0.12664885677672102, "maximum": 0.6547484057365246, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.062496966154583534, "standard_deviation": 0.004962859755735003, "minimum": 0.05218360479298882, "maximum": 0.06796356676909637, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9991881213837435, "standard_deviation": 0.00014488932372292592, "minimum": 0.9990016391575446, "maximum": 0.999448780731285, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6509471289300794, "standard_deviation": 0.03902556750027802, "minimum": 0.5813954841746267, "maximum": 0.7151584700024415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.35591544251501106, "standard_deviation": 0.14045768891301585, "minimum": 0.15023982132133487, "maximum": 0.6314484916754376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.033423229966968726, "standard_deviation": 0.0032032210615505785, "minimum": 0.027616106205672755, "maximum": 0.037444159019552316, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17702311965792725, "standard_deviation": 0.0454215649250383, "minimum": 0.11596202815106448, "maximum": 0.25551813400766404, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.14365844776732, "standard_deviation": 0.00920376356553136, "minimum": 4.133050242771371, "maximum": 4.162152584688478, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5013626832370341, "standard_deviation": 0.015131417531874696, "minimum": 0.48501577312127825, "maximum": 0.5238817074805387, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07073321874228405, "standard_deviation": 0.028638079127526663, "minimum": -0.12278441317633904, "maximum": -0.04805343494607803, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951849864871026, "standard_deviation": 0.0006923486702640904, "minimum": 0.9942583305682423, "maximum": 0.9963012128686253, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6819252049178017, "standard_deviation": 0.028118897626891213, "minimum": 0.6337845808361842, "maximum": 0.7323420995563569, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8422598337062503, "standard_deviation": 0.2948756119161726, "minimum": 0.3472917009255883, "maximum": 1.2147113720893272, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07356405181571304, "standard_deviation": 0.005488443149612313, "minimum": 0.06457729875246669, "maximum": 0.08074860566320102, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985109190391956, "standard_deviation": 0.0002457443840966881, "minimum": 0.99818002786857, "maximum": 0.9989620802047866, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37446569219811565, "standard_deviation": 0.02589788394609293, "minimum": 0.340681679402256, "maximum": 0.4124074491653145, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1527373030454058, "standard_deviation": 0.3992750607200031, "minimum": 0.5941245236102772, "maximum": 1.69012644702092, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04038760063178906, "standard_deviation": 0.0036442684224503863, "minimum": 0.03370192325696523, "maximum": 0.045390773317170056, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-123852-0000", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.4902849197387695, "predicted_mos": 2.4902849197387695, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e25b76bd555ec5acdad68dda3a5d40e35f881187d0f167664846c1f37b5a3af1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8256733999587595, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6370608911367943, "全帯域_10ms_rms_envelope_nrmse": 0.07795064602643043, "全帯域_10ms_envelope_correlation": 0.9945817421033637, "全帯域_10ms_modulation_mae_db": 0.6275072395091451, "全帯域_40ms_log_rms_mae_db": 0.3322045070530202, "全帯域_40ms_rms_envelope_nrmse": 0.044889539302204025, "全帯域_40ms_envelope_correlation": 0.9981157180403559, "全帯域_40ms_modulation_mae_db": 1.0375424503836894, "300–1000 Hz_10ms_log_rms_mae_db": 1.0299969421425923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06351340906581693, "300–1000 Hz_10ms_envelope_correlation": 0.9972167268150433, "300–1000 Hz_10ms_modulation_mae_db": 0.17230569209529958, "300–1000 Hz_40ms_log_rms_mae_db": 0.6690765369091958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03281249806089155, "300–1000 Hz_40ms_envelope_correlation": 0.9992763466308733, "300–1000 Hz_40ms_modulation_mae_db": 0.06220624849140118, "1000–3000 Hz_10ms_log_rms_mae_db": 1.019622362296123, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12022709377067149, "1000–3000 Hz_10ms_envelope_correlation": 0.9891577024571214, "1000–3000 Hz_10ms_modulation_mae_db": 0.5785725588921962, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6514633866963969, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06749803519142081, "1000–3000 Hz_40ms_envelope_correlation": 0.9966079025778981, "1000–3000 Hz_40ms_modulation_mae_db": 0.9746600636213469, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1163756746182325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1913453467469897, "3000–6000 Hz_10ms_envelope_correlation": 0.9729594814232801, "3000–6000 Hz_10ms_modulation_mae_db": 1.8758990376796445, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6747004061696842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12314568370753601, "3000–6000 Hz_40ms_envelope_correlation": 0.9887944131896825, "3000–6000 Hz_40ms_modulation_mae_db": 2.027444192090386, "mr_stft_db": 4.240260253070528, "band_mae_db": 0.1339619836719039, "rms_gain_error_db": -0.04371753285534543, "output_peak": 0.5120446648119915}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6537920047737439, "全帯域_10ms_rms_envelope_nrmse": 0.07964283635305805, "全帯域_10ms_envelope_correlation": 0.9943497763854778, "全帯域_10ms_modulation_mae_db": 0.5919970399587932, "全帯域_40ms_log_rms_mae_db": 0.34186770814311385, "全帯域_40ms_rms_envelope_nrmse": 0.04327431104437108, "全帯域_40ms_envelope_correlation": 0.9982347632531464, "全帯域_40ms_modulation_mae_db": 0.9259593872819792, "300–1000 Hz_10ms_log_rms_mae_db": 1.0450214147923147, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06255280249998277, "300–1000 Hz_10ms_envelope_correlation": 0.9972124249941747, "300–1000 Hz_10ms_modulation_mae_db": 0.20930179499268112, "300–1000 Hz_40ms_log_rms_mae_db": 0.7095750385492053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031938442586194604, "300–1000 Hz_40ms_envelope_correlation": 0.9992489690179227, "300–1000 Hz_40ms_modulation_mae_db": 0.24715337757239209, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0884953578106975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13934592121084932, "1000–3000 Hz_10ms_envelope_correlation": 0.9852302472451222, "1000–3000 Hz_10ms_modulation_mae_db": 0.872791263687618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.681439737171964, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07359735393457612, "1000–3000 Hz_40ms_envelope_correlation": 0.9957252256265352, "1000–3000 Hz_40ms_modulation_mae_db": 1.2367037790839215, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0576963229271337, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18254434774718417, "3000–6000 Hz_10ms_envelope_correlation": 0.9756827896612865, "3000–6000 Hz_10ms_modulation_mae_db": 0.6845723611808895, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6407979249557673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11183932927490539, "3000–6000 Hz_40ms_envelope_correlation": 0.9902432868246804, "3000–6000 Hz_40ms_modulation_mae_db": 1.127871829217652, "mr_stft_db": 4.249326068802566, "band_mae_db": 0.0935885901904749, "rms_gain_error_db": 0.0011851509983961772, "output_peak": 0.46934865936240855}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6651000885717011, "全帯域_10ms_rms_envelope_nrmse": 0.07775841701845002, "全帯域_10ms_envelope_correlation": 0.9947327749816417, "全帯域_10ms_modulation_mae_db": 1.3808524611935127, "全帯域_40ms_log_rms_mae_db": 0.3539450254279041, "全帯域_40ms_rms_envelope_nrmse": 0.03930476312992051, "全帯域_40ms_envelope_correlation": 0.9987308828177804, "全帯域_40ms_modulation_mae_db": 1.7532260364428474, "300–1000 Hz_10ms_log_rms_mae_db": 1.0624609689978624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06129629590143404, "300–1000 Hz_10ms_envelope_correlation": 0.9973529222769891, "300–1000 Hz_10ms_modulation_mae_db": 0.40797517902404196, "300–1000 Hz_40ms_log_rms_mae_db": 0.7284291747881696, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03382123059635143, "300–1000 Hz_40ms_envelope_correlation": 0.9991875698598763, "300–1000 Hz_40ms_modulation_mae_db": 0.647041579484543, "1000–3000 Hz_10ms_log_rms_mae_db": 1.060108400999732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15002130814862938, "1000–3000 Hz_10ms_envelope_correlation": 0.9833583854251473, "1000–3000 Hz_10ms_modulation_mae_db": 1.6837923753413289, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7057143356250625, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0861478339059659, "1000–3000 Hz_40ms_envelope_correlation": 0.9949278085755738, "1000–3000 Hz_40ms_modulation_mae_db": 2.1750118524960094, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1080251239021086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16772841900835112, "3000–6000 Hz_10ms_envelope_correlation": 0.979706853330662, "3000–6000 Hz_10ms_modulation_mae_db": 0.9923746636859829, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6411284208270868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08717835232353155, "3000–6000 Hz_40ms_envelope_correlation": 0.9953531138354514, "3000–6000 Hz_40ms_modulation_mae_db": 1.1789406636524613, "mr_stft_db": 4.24596146231102, "band_mae_db": 0.19092608697498917, "rms_gain_error_db": -0.12914370234806127, "output_peak": 0.5123072947476341}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6319064022561239, "全帯域_10ms_rms_envelope_nrmse": 0.06791433667472685, "全帯域_10ms_envelope_correlation": 0.9959217967376301, "全帯域_10ms_modulation_mae_db": 0.8020422259226091, "全帯域_40ms_log_rms_mae_db": 0.3374267450508623, "全帯域_40ms_rms_envelope_nrmse": 0.033450479135955107, "全帯域_40ms_envelope_correlation": 0.9989983621573897, "全帯域_40ms_modulation_mae_db": 1.1249935331010563, "300–1000 Hz_10ms_log_rms_mae_db": 1.095917638665537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06556700593092447, "300–1000 Hz_10ms_envelope_correlation": 0.9969619187592219, "300–1000 Hz_10ms_modulation_mae_db": 0.6312480164365281, "300–1000 Hz_40ms_log_rms_mae_db": 0.7421731892141641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04103941620131012, "300–1000 Hz_40ms_envelope_correlation": 0.9987717880299688, "300–1000 Hz_40ms_modulation_mae_db": 0.9531956172004743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.047054067695971, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13986969537654506, "1000–3000 Hz_10ms_envelope_correlation": 0.985741808659264, "1000–3000 Hz_10ms_modulation_mae_db": 1.4985619159960526, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6430615296029665, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07845692502237496, "1000–3000 Hz_40ms_envelope_correlation": 0.996065522027465, "1000–3000 Hz_40ms_modulation_mae_db": 2.1920826617004825, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0904833958698477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14153167091516017, "3000–6000 Hz_10ms_envelope_correlation": 0.9851534777403311, "3000–6000 Hz_10ms_modulation_mae_db": 0.30243192152377757, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6398693387994859, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07726169138617085, "3000–6000 Hz_40ms_envelope_correlation": 0.9954749708682475, "3000–6000 Hz_40ms_modulation_mae_db": 0.5927785405126257, "mr_stft_db": 4.235538650977403, "band_mae_db": 0.15731970270622905, "rms_gain_error_db": -0.065022094276724, "output_peak": 0.49502802544876845}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6354960047000857, "全帯域_10ms_rms_envelope_nrmse": 0.06962058787568752, "全帯域_10ms_envelope_correlation": 0.9956776574448689, "全帯域_10ms_modulation_mae_db": 0.7133563577488621, "全帯域_40ms_log_rms_mae_db": 0.36976619752325307, "全帯域_40ms_rms_envelope_nrmse": 0.037817719778614846, "全帯域_40ms_envelope_correlation": 0.9986564323039518, "全帯域_40ms_modulation_mae_db": 0.6151082343100717, "300–1000 Hz_10ms_log_rms_mae_db": 1.0938105656738597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07042462981153583, "300–1000 Hz_10ms_envelope_correlation": 0.9964786175380237, "300–1000 Hz_10ms_modulation_mae_db": 0.3280868859269491, "300–1000 Hz_40ms_log_rms_mae_db": 0.7481807449513984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042855481182625374, "300–1000 Hz_40ms_envelope_correlation": 0.998649080402283, "300–1000 Hz_40ms_modulation_mae_db": 0.510174881412167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0465262507864248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12229570549870986, "1000–3000 Hz_10ms_envelope_correlation": 0.988801514319104, "1000–3000 Hz_10ms_modulation_mae_db": 0.914666809078068, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6763729774028973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07089607513338761, "1000–3000 Hz_40ms_envelope_correlation": 0.9962673171692664, "1000–3000 Hz_40ms_modulation_mae_db": 1.3248240819542392, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0158696651241321, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.156125742293522, "3000–6000 Hz_10ms_envelope_correlation": 0.9818117766114626, "3000–6000 Hz_10ms_modulation_mae_db": 0.4977351812401824, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6155465961235506, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08878907772336057, "3000–6000 Hz_40ms_envelope_correlation": 0.9937412086792093, "3000–6000 Hz_40ms_modulation_mae_db": 0.5879511867802494, "mr_stft_db": 4.235573963242483, "band_mae_db": 0.11092872226487523, "rms_gain_error_db": -0.02533753262962648, "output_peak": 0.5049648720456746}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6378020521428056, "全帯域_10ms_rms_envelope_nrmse": 0.06803904553000908, "全帯域_10ms_envelope_correlation": 0.9958806549735238, "全帯域_10ms_modulation_mae_db": 0.42757710843753616, "全帯域_40ms_log_rms_mae_db": 0.33496913920135196, "全帯域_40ms_rms_envelope_nrmse": 0.03508169382637381, "全帯域_40ms_envelope_correlation": 0.9988494436593063, "全帯域_40ms_modulation_mae_db": 0.7602456469373028, "300–1000 Hz_10ms_log_rms_mae_db": 1.1005754401094183, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07468691247507947, "300–1000 Hz_10ms_envelope_correlation": 0.9960401394110612, "300–1000 Hz_10ms_modulation_mae_db": 0.1052278197480353, "300–1000 Hz_40ms_log_rms_mae_db": 0.7252523267717347, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04379515814327522, "300–1000 Hz_40ms_envelope_correlation": 0.9985944882847337, "300–1000 Hz_40ms_modulation_mae_db": 0.3112178084400274, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0216247660403555, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12395101412578523, "1000–3000 Hz_10ms_envelope_correlation": 0.9883424401007781, "1000–3000 Hz_10ms_modulation_mae_db": 0.571933739460518, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6247399014192513, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06436948030063429, "1000–3000 Hz_40ms_envelope_correlation": 0.9967473859590203, "1000–3000 Hz_40ms_modulation_mae_db": 1.0113785866629976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0704619854316564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15516484876843303, "3000–6000 Hz_10ms_envelope_correlation": 0.982042010091338, "3000–6000 Hz_10ms_modulation_mae_db": 0.5625167696602769, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6264144812218642, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09013748869478998, "3000–6000 Hz_40ms_envelope_correlation": 0.9935562351148772, "3000–6000 Hz_40ms_modulation_mae_db": 0.7713217473688938, "mr_stft_db": 4.252149082294978, "band_mae_db": 0.07667877572560235, "rms_gain_error_db": -0.024337743412725276, "output_peak": 0.5270377010831746}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6637211547920304, "全帯域_10ms_rms_envelope_nrmse": 0.07121816916014449, "全帯域_10ms_envelope_correlation": 0.9955209520670745, "全帯域_10ms_modulation_mae_db": 0.4795132865929479, "全帯域_40ms_log_rms_mae_db": 0.348328492198541, "全帯域_40ms_rms_envelope_nrmse": 0.03896908635872822, "全帯域_40ms_envelope_correlation": 0.9986425204761393, "全帯域_40ms_modulation_mae_db": 0.5455360995588994, "300–1000 Hz_10ms_log_rms_mae_db": 1.068174155286632, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06440773502475647, "300–1000 Hz_10ms_envelope_correlation": 0.9970463285144652, "300–1000 Hz_10ms_modulation_mae_db": 0.2381536428451777, "300–1000 Hz_40ms_log_rms_mae_db": 0.6997812776681258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03572838991161947, "300–1000 Hz_40ms_envelope_correlation": 0.9990641766001159, "300–1000 Hz_40ms_modulation_mae_db": 0.5161497318718737, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0531130034032627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12656588413642605, "1000–3000 Hz_10ms_envelope_correlation": 0.9883001772976722, "1000–3000 Hz_10ms_modulation_mae_db": 1.0517181827919246, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6938134154648941, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07875261622373202, "1000–3000 Hz_40ms_envelope_correlation": 0.9957279886774995, "1000–3000 Hz_40ms_modulation_mae_db": 1.6070175628559091, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0605852720558657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15255678359495495, "3000–6000 Hz_10ms_envelope_correlation": 0.9827817662640541, "3000–6000 Hz_10ms_modulation_mae_db": 0.4872286457617417, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6462572359317738, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08840156789373874, "3000–6000 Hz_40ms_envelope_correlation": 0.9940799289974879, "3000–6000 Hz_40ms_modulation_mae_db": 0.381007575284676, "mr_stft_db": 4.246298467869656, "band_mae_db": 0.13977365547856174, "rms_gain_error_db": -0.08381804598797336, "output_peak": 0.5012621035110171}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6662933149322029, "全帯域_10ms_rms_envelope_nrmse": 0.07220447689153185, "全帯域_10ms_envelope_correlation": 0.9954028383060348, "全帯域_10ms_modulation_mae_db": 0.5954006116172663, "全帯域_40ms_log_rms_mae_db": 0.343926502554829, "全帯域_40ms_rms_envelope_nrmse": 0.03741172154075806, "全帯域_40ms_envelope_correlation": 0.9987608211413024, "全帯域_40ms_modulation_mae_db": 0.9879061268755351, "300–1000 Hz_10ms_log_rms_mae_db": 1.0477577744331965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06415361664708155, "300–1000 Hz_10ms_envelope_correlation": 0.9970701241816559, "300–1000 Hz_10ms_modulation_mae_db": 0.16953880023860615, "300–1000 Hz_40ms_log_rms_mae_db": 0.6531849072344298, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035504340191335096, "300–1000 Hz_40ms_envelope_correlation": 0.9990733603852915, "300–1000 Hz_40ms_modulation_mae_db": 0.27021597931468305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0476438179823353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12989388106913585, "1000–3000 Hz_10ms_envelope_correlation": 0.9873346185200097, "1000–3000 Hz_10ms_modulation_mae_db": 0.5687510339620768, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6603783148983172, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07410755778454209, "1000–3000 Hz_40ms_envelope_correlation": 0.9959130132572813, "1000–3000 Hz_40ms_modulation_mae_db": 0.8767182752266894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0837219385131518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15941083687877794, "3000–6000 Hz_10ms_envelope_correlation": 0.9818568288622175, "3000–6000 Hz_10ms_modulation_mae_db": 1.6950087980984083, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6623633576189355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09182270047726108, "3000–6000 Hz_40ms_envelope_correlation": 0.994690666258037, "3000–6000 Hz_40ms_modulation_mae_db": 1.9302975740282085, "mr_stft_db": 4.237823466033501, "band_mae_db": 0.15582210972015534, "rms_gain_error_db": -0.08599209041438088, "output_peak": 0.4902808402675533}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.2567909163010675, "spectral_convergence": 0.30793755931485245, "spectral_convergence_status": "ok", "amplitude_floor": 1.0172154037614098e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.159696525120253, "spectral_convergence": 0.2810407499398818, "spectral_convergence_status": "ok", "amplitude_floor": 7.045944280558953e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.304293317790264, "spectral_convergence": 0.28907386284514475, "spectral_convergence_status": "ok", "amplitude_floor": 6.386940858702166e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00026428484162924415, "output_power": 0.0002650230311925606, "delta_db": 0.012113625292896108, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008055137994793597, "output_power": 0.0008253634642818528, "delta_db": 0.10572255555465648, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006109316451161528, "output_power": 0.0006110914509005156, "delta_db": 0.00113586675975048, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027595006700553216, "output_power": 0.0002648241445958804, "delta_db": -0.17872925706047402, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003628265208503686, "output_power": 0.0003351458566097611, "delta_db": -0.34465169754756464, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001954243236049976, "output_power": 0.00018829410668712998, "delta_db": -0.1614188998160817, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1339619836719039, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04371753285534543, "pitch": null, "output_rms_dbfs": -26.038456449568834, "output_peak": 0.5120446648119915, "reference_rms_dbfs": -25.994738916713484, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.240260253070528}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017615675305240148, "log_rms_mae_db": 0.6370608911367943, "rms_envelope_nrmse": 0.07795064602643043, "envelope_correlation": 0.9945817421033637, "rms_gain_error_db": -0.04365204955138283, "reference_mean_power": 0.0025168568603362385, "output_mean_power": 0.0024916859993705767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0162973895242792e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002469993456433397, "output_power": 0.0002472734817780538, "reference_active": true, "delta_db": 0.0048174132260226914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010173002443957726, "output_power": 8.831409432581362e-05, "reference_active": true, "delta_db": -0.6141912929274759}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1084315450169838e-05, "output_power": 1.5761886795046995e-05, "reference_active": true, "delta_db": -1.2635130123739367}], "active_band_count": 3, "active_band_mae_db": 0.6275072395091451}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016525538510559444, "log_rms_mae_db": 0.3322045070530202, "rms_envelope_nrmse": 0.044889539302204025, "envelope_correlation": 0.9981157180403559, "rms_gain_error_db": -0.043604578552067993, "reference_mean_power": 0.002527970677484516, "output_mean_power": 0.0025027160242669794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.625293642327984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023064746753792456, "output_power": 0.00022925319987494367, "reference_active": true, "delta_db": -0.026332843171175423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.045652222183457e-05, "output_power": 4.877515201653561e-05, "reference_active": true, "delta_db": -0.9324452902458618}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2874374178901008e-06, "output_power": 1.393042284550931e-06, "reference_active": true, "delta_db": -2.153849217734031}], "active_band_count": 3, "active_band_mae_db": 1.0375424503836894}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001279910087199623, "log_rms_mae_db": 1.0299969421425923, "rms_envelope_nrmse": 0.06351340906581693, "envelope_correlation": 0.9972167268150433, "rms_gain_error_db": 0.07890348486836463, "reference_mean_power": 0.0009939686093891967, "output_mean_power": 0.0010121922729867039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.147046828130101e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013887130907316313, "output_power": 0.0001400269552567683, "reference_active": true, "delta_db": 0.03599116216915329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4048550736300918e-05, "output_power": 2.4562211151072392e-05, "reference_active": true, "delta_db": 0.09178551351127828}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.939409383024551e-06, "output_power": 5.402428931188709e-06, "reference_active": true, "delta_db": 0.38914040060546723}], "active_band_count": 3, "active_band_mae_db": 0.17230569209529958}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011859895721274409, "log_rms_mae_db": 0.6690765369091958, "rms_envelope_nrmse": 0.03281249806089155, "envelope_correlation": 0.9992763466308733, "rms_gain_error_db": 0.07882471724347585, "reference_mean_power": 0.0009986896907851761, "output_mean_power": 0.0010169814667691276}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.009742136856967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012678804378904424, "output_power": 0.00012812753282933847, "reference_active": true, "delta_db": 0.0456416246266171}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.298881565077983e-05, "output_power": 1.3295034490981462e-05, "reference_active": true, "delta_db": 0.10119915411042334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8039990598306815e-07, "output_power": 2.778433885864595e-07, "reference_active": true, "delta_db": -0.03977796673716311}], "active_band_count": 3, "active_band_mae_db": 0.06220624849140118}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.31629258809413e-05, "log_rms_mae_db": 1.019622362296123, "rms_envelope_nrmse": 0.12022709377067149, "envelope_correlation": 0.9891577024571214, "rms_gain_error_db": -0.1804421496658075, "reference_mean_power": 0.0004146029386445929, "output_mean_power": 0.00039772982673694245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1875762004578613e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604229953230037e-05, "output_power": 8.424423971922052e-05, "reference_active": true, "delta_db": -0.09171793395886482}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2897806527831664e-05, "output_power": 2.2748037709360485e-05, "reference_active": true, "delta_db": -0.028499420231808237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2516459896674564e-05, "output_power": 8.628424035870325e-06, "reference_active": true, "delta_db": -1.6155003224859155}], "active_band_count": 3, "active_band_mae_db": 0.5785725588921962}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.786085823827924e-05, "log_rms_mae_db": 0.6514633866963969, "rms_envelope_nrmse": 0.06749803519142081, "envelope_correlation": 0.9966079025778981, "rms_gain_error_db": -0.18044714068410697, "reference_mean_power": 0.0004166837249389974, "output_mean_power": 0.0003997254718161105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.035015228514449e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.632173919031572e-05, "output_power": 8.334770424702516e-05, "reference_active": true, "delta_db": -0.152265398296734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4669499526204761e-05, "output_power": 1.2981067332282787e-05, "reference_active": true, "delta_db": -0.5310489484208093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2709360758910173e-06, "output_power": 7.586773577915139e-07, "reference_active": true, "delta_db": -2.2406658441464975}], "active_band_count": 3, "active_band_mae_db": 0.9746600636213469}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011544071276927888, "log_rms_mae_db": 1.1163756746182325, "rms_envelope_nrmse": 0.1913453467469897, "envelope_correlation": 0.9729594814232801, "rms_gain_error_db": -0.37082149620481514, "reference_mean_power": 0.00033862428581899314, "output_mean_power": 0.0003109109032503468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7742895295757852e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.9949103091747194e-05, "output_power": 5.1283155300407066e-05, "reference_active": true, "delta_db": -0.6780795174940375}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5523159384577664e-05, "output_power": 3.247928667000238e-05, "reference_active": true, "delta_db": -1.466259127752969}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.055143212029728e-06, "output_power": 3.16351564895923e-06, "reference_active": true, "delta_db": -3.483358467791927}], "active_band_count": 3, "active_band_mae_db": 1.8758990376796445}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.669712818495073e-05, "log_rms_mae_db": 0.6747004061696842, "rms_envelope_nrmse": 0.12314568370753601, "envelope_correlation": 0.9887944131896825, "rms_gain_error_db": -0.370840210811701, "reference_mean_power": 0.00034032151520785135, "output_mean_power": 0.0003124678830377174}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6870888960608118e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.86684133644439e-05, "output_power": 5.533076988951984e-05, "reference_active": true, "delta_db": -0.9379029964358393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.573388989930172e-05, "output_power": 1.7333987234453676e-05, "reference_active": true, "delta_db": -1.7160696613377335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.935502106463895e-07, "output_power": 4.0577283540617085e-07, "reference_active": true, "delta_db": -3.428359918497585}], "active_band_count": 3, "active_band_mae_db": 2.027444192090386}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8993249132349539, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bf2bfedddc00b7f2b5836fb0fed8d3f4662b344877064f0d7cafcfa5d869c089", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9870333617530274, "standard_deviation": 0.0020519409849999192, "minimum": 0.9833583854251473, "maximum": 0.9891577024571214, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0480235033768628, "standard_deviation": 0.02176748112567143, "minimum": 1.019622362296123, "maximum": 1.0884953578106975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9675984849012229, "standard_deviation": 0.4277337623116807, "minimum": 0.5687510339620768, "maximum": 1.6837923753413289, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.131521312917094, "standard_deviation": 0.010490507676548626, "minimum": 0.12022709377067149, "maximum": 0.15002130814862938, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9959977704838174, "standard_deviation": 0.0005745178618369225, "minimum": 0.9949278085755738, "maximum": 0.9967473859590203, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6671229497852187, "standard_deviation": 0.027145628895990745, "minimum": 0.6247399014192513, "maximum": 0.7057143356250625, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4247996079501994, "standard_deviation": 0.5215438515537474, "minimum": 0.8767182752266894, "maximum": 2.1920826617004825, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07422823468707923, "standard_deviation": 0.0069120513785030716, "minimum": 0.06436948030063429, "maximum": 0.0861478339059659, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.980249372998079, "standard_deviation": 0.0040176815678529855, "minimum": 0.9729594814232801, "maximum": 0.9851534777403311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.075402422305266, "standard_deviation": 0.031940366687579144, "minimum": 1.0158696651241321, "maximum": 1.1163756746182325, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.887220922353863, "standard_deviation": 0.5905485758075372, "minimum": 0.30243192152377757, "maximum": 1.8758990376796445, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16330099949417165, "standard_deviation": 0.016468803322600433, "minimum": 0.14153167091516017, "maximum": 0.1913453467469897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932417279709591, "standard_deviation": 0.002430901685167439, "minimum": 0.9887944131896825, "maximum": 0.9954749708682475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6433847202060186, "standard_deviation": 0.018655534141061927, "minimum": 0.6155465961235506, "maximum": 0.6747004061696842, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.074701663616894, "standard_deviation": 0.6205852431565867, "minimum": 0.381007575284676, "maximum": 2.027444192090386, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09482198643516176, "standard_deviation": 0.014962662512550315, "minimum": 0.07726169138617085, "maximum": 0.12314568370753601, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9969224003113293, "standard_deviation": 0.00044242211025687035, "minimum": 0.9960401394110612, "maximum": 0.9973529222769891, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0679643625126767, "standard_deviation": 0.02650993253805765, "minimum": 1.0299969421425923, "maximum": 1.1005754401094183, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.28272972891341486, "standard_deviation": 0.17033578691667767, "minimum": 0.1052278197480353, "maximum": 0.6312480164365281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06582530091957643, "standard_deviation": 0.0044899681767142235, "minimum": 0.06129629590143404, "maximum": 0.07468691247507947, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989832224013832, "standard_deviation": 0.00027267169761711874, "minimum": 0.9985944882847337, "maximum": 0.9992763466308733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.709456649510803, "standard_deviation": 0.033953912003061075, "minimum": 0.6531849072344298, "maximum": 0.7481807449513984, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4396694029734452, "standard_deviation": 0.2783337625029108, "minimum": 0.06220624849140118, "maximum": 0.9531956172004743, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.037186869609200354, "standard_deviation": 0.004684495043039764, "minimum": 0.031938442586194604, "maximum": 0.04379515814327522, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13237495334159896, "standard_deviation": 0.03727099861891924, "minimum": 0.07667877572560235, "maximum": 0.19092608697498917, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.242866426825267, "standard_deviation": 0.006418196638160265, "minimum": 4.235538650977403, "maximum": 4.252149082294978, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5015342701597778, "standard_deviation": 0.017294421042401465, "minimum": 0.46934865936240855, "maximum": 0.5270377010831746, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05702294886580507, "standard_deviation": 0.0421901868153237, "minimum": -0.12914370234806127, "maximum": 0.0011851509983961772, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952585241249519, "standard_deviation": 0.0006155849273475608, "minimum": 0.9943497763854778, "maximum": 0.9959217967376301, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6488964891631861, "standard_deviation": 0.014834067523904206, "minimum": 0.6319064022561239, "maximum": 0.6662933149322029, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7022807913725841, "standard_deviation": 0.298759090809713, "minimum": 0.42757710843753616, "maximum": 1.3808524611935127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07304356444125479, "standard_deviation": 0.004735379992369436, "minimum": 0.06791433667472685, "maximum": 0.07964283635305805, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986236179811716, "standard_deviation": 0.0003005626438913212, "minimum": 0.9981157180403559, "maximum": 0.9989983621573897, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3453042896441094, "standard_deviation": 0.012167737142404178, "minimum": 0.3322045070530202, "maximum": 0.36976619752325307, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9688146893614227, "standard_deviation": 0.37690309934814703, "minimum": 0.5455360995588994, "maximum": 1.7532260364428474, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03877491426461571, "standard_deviation": 0.003830622024933574, "minimum": 0.033450479135955107, "maximum": 0.044889539302204025, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-123852-0000", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6061856746673584, "predicted_mos": 2.6061856746673584, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a1a8845ec5d5ad85e59c0c0c4b0f7e6cdf632ccbe3d04145ba031682e145ac34", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.7176021998748183, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6636904638215709, "全帯域_10ms_rms_envelope_nrmse": 0.07552218406585766, "全帯域_10ms_envelope_correlation": 0.9949670861016543, "全帯域_10ms_modulation_mae_db": 0.4756894369220353, "全帯域_40ms_log_rms_mae_db": 0.3491420604737422, "全帯域_40ms_rms_envelope_nrmse": 0.045388598772396094, "全帯域_40ms_envelope_correlation": 0.998135676998545, "全帯域_40ms_modulation_mae_db": 0.8241039610802353, "300–1000 Hz_10ms_log_rms_mae_db": 1.0693614463344778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06569887722058358, "300–1000 Hz_10ms_envelope_correlation": 0.9969427063729766, "300–1000 Hz_10ms_modulation_mae_db": 0.37125461225895157, "300–1000 Hz_40ms_log_rms_mae_db": 0.7224653478882339, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03911916574668919, "300–1000 Hz_40ms_envelope_correlation": 0.9988984452819337, "300–1000 Hz_40ms_modulation_mae_db": 0.21904249147177543, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0385656200189248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13862500184465226, "1000–3000 Hz_10ms_envelope_correlation": 0.9854653208217867, "1000–3000 Hz_10ms_modulation_mae_db": 0.19334599327251087, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6633998071978133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0821699531562241, "1000–3000 Hz_40ms_envelope_correlation": 0.9947857760030427, "1000–3000 Hz_40ms_modulation_mae_db": 0.303771672508127, "3000–6000 Hz_10ms_log_rms_mae_db": 1.038579439402059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16227575470841682, "3000–6000 Hz_10ms_envelope_correlation": 0.9814913240607088, "3000–6000 Hz_10ms_modulation_mae_db": 1.6625043027234516, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6460161817909896, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11462264836336071, "3000–6000 Hz_40ms_envelope_correlation": 0.9911734680592219, "3000–6000 Hz_40ms_modulation_mae_db": 2.003979692242961, "mr_stft_db": 4.202082848864051, "band_mae_db": 0.1827381443668609, "rms_gain_error_db": -0.08321687489820657, "output_peak": 0.4749984240989598}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.642055591373551, "全帯域_10ms_rms_envelope_nrmse": 0.0679201404757471, "全帯域_10ms_envelope_correlation": 0.9958884725592062, "全帯域_10ms_modulation_mae_db": 0.5798075725110855, "全帯域_40ms_log_rms_mae_db": 0.3171899215408448, "全帯域_40ms_rms_envelope_nrmse": 0.032388515120648234, "全帯域_40ms_envelope_correlation": 0.9990207406522174, "全帯域_40ms_modulation_mae_db": 0.8465291402619263, "300–1000 Hz_10ms_log_rms_mae_db": 1.0541184125438547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0656559323759818, "300–1000 Hz_10ms_envelope_correlation": 0.996931343698988, "300–1000 Hz_10ms_modulation_mae_db": 0.2631058014373915, "300–1000 Hz_40ms_log_rms_mae_db": 0.7054453095383239, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03310375313984801, "300–1000 Hz_40ms_envelope_correlation": 0.9991908188507345, "300–1000 Hz_40ms_modulation_mae_db": 0.1881546855786588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.009856511580352, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12954665851727906, "1000–3000 Hz_10ms_envelope_correlation": 0.9872605330081431, "1000–3000 Hz_10ms_modulation_mae_db": 0.6019073418808683, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6394245801029422, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06520244518481846, "1000–3000 Hz_40ms_envelope_correlation": 0.9966824658880834, "1000–3000 Hz_40ms_modulation_mae_db": 0.7746948380964366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0413814278581444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1435641344835441, "3000–6000 Hz_10ms_envelope_correlation": 0.984761821439452, "3000–6000 Hz_10ms_modulation_mae_db": 1.314038387807061, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5998929127422868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08453940451678221, "3000–6000 Hz_40ms_envelope_correlation": 0.9945926970989066, "3000–6000 Hz_40ms_modulation_mae_db": 1.574968966597514, "mr_stft_db": 4.202198501380948, "band_mae_db": 0.11415269972637314, "rms_gain_error_db": -0.03288349116677932, "output_peak": 0.49117405602640396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6626003605992675, "全帯域_10ms_rms_envelope_nrmse": 0.07398756342679075, "全帯域_10ms_envelope_correlation": 0.9951175381664658, "全帯域_10ms_modulation_mae_db": 1.003435962513423, "全帯域_40ms_log_rms_mae_db": 0.3692536510511571, "全帯域_40ms_rms_envelope_nrmse": 0.041414437615460674, "全帯域_40ms_envelope_correlation": 0.9983888105370183, "全帯域_40ms_modulation_mae_db": 1.367338905379755, "300–1000 Hz_10ms_log_rms_mae_db": 1.0916253662676707, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06841600198153472, "300–1000 Hz_10ms_envelope_correlation": 0.9966934695357694, "300–1000 Hz_10ms_modulation_mae_db": 0.18877729774978072, "300–1000 Hz_40ms_log_rms_mae_db": 0.7046488860635935, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03600720284669217, "300–1000 Hz_40ms_envelope_correlation": 0.9990648209470199, "300–1000 Hz_40ms_modulation_mae_db": 0.12510211458180273, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0417790291318083, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1340701830895033, "1000–3000 Hz_10ms_envelope_correlation": 0.9864765428912737, "1000–3000 Hz_10ms_modulation_mae_db": 1.1194824159585313, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6346285039207924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07283519515510704, "1000–3000 Hz_40ms_envelope_correlation": 0.9960601734466129, "1000–3000 Hz_40ms_modulation_mae_db": 1.8515284351674552, "3000–6000 Hz_10ms_log_rms_mae_db": 1.046350675438989, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1576134940708126, "3000–6000 Hz_10ms_envelope_correlation": 0.9815816922637668, "3000–6000 Hz_10ms_modulation_mae_db": 1.451603835908981, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6267698144718891, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1066197792670032, "3000–6000 Hz_40ms_envelope_correlation": 0.9912036714726414, "3000–6000 Hz_40ms_modulation_mae_db": 1.7759230384159241, "mr_stft_db": 4.21457638909096, "band_mae_db": 0.13665379476267667, "rms_gain_error_db": -0.030399201664253847, "output_peak": 0.47949554575952974}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6566265573640246, "全帯域_10ms_rms_envelope_nrmse": 0.07899037516944465, "全帯域_10ms_envelope_correlation": 0.9945155910209711, "全帯域_10ms_modulation_mae_db": 0.8389419804646426, "全帯域_40ms_log_rms_mae_db": 0.36412941188362363, "全帯域_40ms_rms_envelope_nrmse": 0.039906886559776554, "全帯域_40ms_envelope_correlation": 0.9986337995983462, "全帯域_40ms_modulation_mae_db": 1.286248640366903, "300–1000 Hz_10ms_log_rms_mae_db": 1.0963937114365383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06331832928646532, "300–1000 Hz_10ms_envelope_correlation": 0.997148707633552, "300–1000 Hz_10ms_modulation_mae_db": 0.4187239989330924, "300–1000 Hz_40ms_log_rms_mae_db": 0.740642510236693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035237405102750745, "300–1000 Hz_40ms_envelope_correlation": 0.9990855824228912, "300–1000 Hz_40ms_modulation_mae_db": 0.9261514000949784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0400431210989385, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13634932941042635, "1000–3000 Hz_10ms_envelope_correlation": 0.985929996447944, "1000–3000 Hz_10ms_modulation_mae_db": 0.9989797549953656, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6310606242394363, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0689824847036376, "1000–3000 Hz_40ms_envelope_correlation": 0.9963973037784916, "1000–3000 Hz_40ms_modulation_mae_db": 1.1575441219435785, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0926353596231413, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18391136880805192, "3000–6000 Hz_10ms_envelope_correlation": 0.9775751416699096, "3000–6000 Hz_10ms_modulation_mae_db": 1.8314396684445766, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6633579431460053, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11425656732971692, "3000–6000 Hz_40ms_envelope_correlation": 0.9938246873034857, "3000–6000 Hz_40ms_modulation_mae_db": 2.051866824102079, "mr_stft_db": 4.210202695649553, "band_mae_db": 0.2497565500608988, "rms_gain_error_db": -0.11095137089210227, "output_peak": 0.48226661463081005}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6576811924969239, "全帯域_10ms_rms_envelope_nrmse": 0.07226672445163798, "全帯域_10ms_envelope_correlation": 0.9953629922109337, "全帯域_10ms_modulation_mae_db": 0.4547128819749242, "全帯域_40ms_log_rms_mae_db": 0.3282036512830412, "全帯域_40ms_rms_envelope_nrmse": 0.03401768977948946, "全帯域_40ms_envelope_correlation": 0.998950103201076, "全帯域_40ms_modulation_mae_db": 0.7309301077931059, "300–1000 Hz_10ms_log_rms_mae_db": 1.0982723525781661, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06453192796393126, "300–1000 Hz_10ms_envelope_correlation": 0.9970388297645054, "300–1000 Hz_10ms_modulation_mae_db": 0.2502816549068431, "300–1000 Hz_40ms_log_rms_mae_db": 0.7662326565473975, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03689930761714247, "300–1000 Hz_40ms_envelope_correlation": 0.9989961486814437, "300–1000 Hz_40ms_modulation_mae_db": 0.10176593166481228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0702389554978073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14618050203884633, "1000–3000 Hz_10ms_envelope_correlation": 0.9837295921961094, "1000–3000 Hz_10ms_modulation_mae_db": 0.7696149352671254, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6959644717780443, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07401859666582487, "1000–3000 Hz_40ms_envelope_correlation": 0.9956566147086877, "1000–3000 Hz_40ms_modulation_mae_db": 1.3305689077054215, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0881566927145372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16571921557037575, "3000–6000 Hz_10ms_envelope_correlation": 0.9805974974370011, "3000–6000 Hz_10ms_modulation_mae_db": 0.2992645090965923, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6647899537983913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10395254847591161, "3000–6000 Hz_40ms_envelope_correlation": 0.9931435290907448, "3000–6000 Hz_40ms_modulation_mae_db": 0.5230617344492099, "mr_stft_db": 4.203867829353716, "band_mae_db": 0.16007876776831845, "rms_gain_error_db": -0.061885907972946054, "output_peak": 0.4717582102277697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7153553836442574, "全帯域_10ms_rms_envelope_nrmse": 0.07733618777240382, "全帯域_10ms_envelope_correlation": 0.9946689734419205, "全帯域_10ms_modulation_mae_db": 0.578287245415379, "全帯域_40ms_log_rms_mae_db": 0.4081044027122606, "全帯域_40ms_rms_envelope_nrmse": 0.03747892379010365, "全帯域_40ms_envelope_correlation": 0.9986949425338154, "全帯域_40ms_modulation_mae_db": 0.8621288135177602, "300–1000 Hz_10ms_log_rms_mae_db": 1.0596189851194653, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06894253943720458, "300–1000 Hz_10ms_envelope_correlation": 0.9966606807079783, "300–1000 Hz_10ms_modulation_mae_db": 0.36172983513894996, "300–1000 Hz_40ms_log_rms_mae_db": 0.7542435222032132, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03863879565773507, "300–1000 Hz_40ms_envelope_correlation": 0.9989305787322954, "300–1000 Hz_40ms_modulation_mae_db": 0.6030093218462053, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0557663063025524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12762988316618123, "1000–3000 Hz_10ms_envelope_correlation": 0.9879224840397878, "1000–3000 Hz_10ms_modulation_mae_db": 0.9220481848027942, "1000–3000 Hz_40ms_log_rms_mae_db": 0.670357276320735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06795229851606098, "1000–3000 Hz_40ms_envelope_correlation": 0.9968236691361425, "1000–3000 Hz_40ms_modulation_mae_db": 1.3230568811455943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0377906194887088, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17073430324413347, "3000–6000 Hz_10ms_envelope_correlation": 0.9783341805155983, "3000–6000 Hz_10ms_modulation_mae_db": 0.4665408627676415, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5946198938892286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0825135844563978, "3000–6000 Hz_40ms_envelope_correlation": 0.9949874679407946, "3000–6000 Hz_40ms_modulation_mae_db": 0.5535556398903599, "mr_stft_db": 4.216776008972789, "band_mae_db": 0.16402701280600232, "rms_gain_error_db": -0.04384559604824524, "output_peak": 0.4774502874716284}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6356728203183621, "全帯域_10ms_rms_envelope_nrmse": 0.07562341488313666, "全帯域_10ms_envelope_correlation": 0.9949010997249577, "全帯域_10ms_modulation_mae_db": 0.6429808778593388, "全帯域_40ms_log_rms_mae_db": 0.33943352962859596, "全帯域_40ms_rms_envelope_nrmse": 0.03785753419924663, "全帯域_40ms_envelope_correlation": 0.9986606866502341, "全帯域_40ms_modulation_mae_db": 0.9501898420386025, "300–1000 Hz_10ms_log_rms_mae_db": 1.0667055916613397, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061762501456424, "300–1000 Hz_10ms_envelope_correlation": 0.9973253654004741, "300–1000 Hz_10ms_modulation_mae_db": 0.37088745181153465, "300–1000 Hz_40ms_log_rms_mae_db": 0.6948076134316712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03187479102315695, "300–1000 Hz_40ms_envelope_correlation": 0.9992802587975019, "300–1000 Hz_40ms_modulation_mae_db": 0.8897655114882963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0282490026896556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12109978754381547, "1000–3000 Hz_10ms_envelope_correlation": 0.9888948143825244, "1000–3000 Hz_10ms_modulation_mae_db": 0.27158132409362373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6559024118581105, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06253668699144345, "1000–3000 Hz_40ms_envelope_correlation": 0.9969753713834423, "1000–3000 Hz_40ms_modulation_mae_db": 0.57713541218658, "3000–6000 Hz_10ms_log_rms_mae_db": 1.053294167458372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1793658524709451, "3000–6000 Hz_10ms_envelope_correlation": 0.9766958712003531, "3000–6000 Hz_10ms_modulation_mae_db": 1.7860431711989004, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6283083029131395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10031823629863387, "3000–6000 Hz_40ms_envelope_correlation": 0.9934691623703824, "3000–6000 Hz_40ms_modulation_mae_db": 1.6225525740362627, "mr_stft_db": 4.210658468638823, "band_mae_db": 0.16346525412845297, "rms_gain_error_db": -0.036964256860393355, "output_peak": 0.5002737739126992}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6673975624836208, "全帯域_10ms_rms_envelope_nrmse": 0.06902461037422564, "全帯域_10ms_envelope_correlation": 0.995851692177636, "全帯域_10ms_modulation_mae_db": 0.9079432073396658, "全帯域_40ms_log_rms_mae_db": 0.34309653201874835, "全帯域_40ms_rms_envelope_nrmse": 0.036670600527063474, "全帯域_40ms_envelope_correlation": 0.9988705258625084, "全帯域_40ms_modulation_mae_db": 1.2763459850747347, "300–1000 Hz_10ms_log_rms_mae_db": 1.0302590230438429, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06713600248182501, "300–1000 Hz_10ms_envelope_correlation": 0.9967958243710165, "300–1000 Hz_10ms_modulation_mae_db": 0.1292134876345293, "300–1000 Hz_40ms_log_rms_mae_db": 0.681179551699119, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03644752635819971, "300–1000 Hz_40ms_envelope_correlation": 0.999034403408309, "300–1000 Hz_40ms_modulation_mae_db": 0.1355076041066056, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9891993343609432, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12954236367126887, "1000–3000 Hz_10ms_envelope_correlation": 0.987590937883154, "1000–3000 Hz_10ms_modulation_mae_db": 1.3271362536576972, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6538870556234612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07263052693906953, "1000–3000 Hz_40ms_envelope_correlation": 0.9963509830725115, "1000–3000 Hz_40ms_modulation_mae_db": 1.884279305940769, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0220431633587277, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15508028952384914, "3000–6000 Hz_10ms_envelope_correlation": 0.9838624215666358, "3000–6000 Hz_10ms_modulation_mae_db": 1.0727842765654578, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6084342768841979, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09207358785370875, "3000–6000 Hz_40ms_envelope_correlation": 0.9958513320746166, "3000–6000 Hz_40ms_modulation_mae_db": 1.3203058872919355, "mr_stft_db": 4.2098698886976775, "band_mae_db": 0.22102258912735673, "rms_gain_error_db": -0.11099050115264547, "output_peak": 0.501863189066651}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.223552635424914, "spectral_convergence": 0.30524492761467636, "spectral_convergence_status": "ok", "amplitude_floor": 1.0172154037614098e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.119963137446145, "spectral_convergence": 0.28003765073895553, "spectral_convergence_status": "ok", "amplitude_floor": 7.045944280558953e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.262732773721091, "spectral_convergence": 0.2768354007115116, "spectral_convergence_status": "ok", "amplitude_floor": 6.386940858702166e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00026428484162924415, "output_power": 0.0002722132419220999, "delta_db": 0.12836993439481098, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008055137994793597, "output_power": 0.0008151278593892365, "delta_db": 0.05152751843307445, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006109316451161528, "output_power": 0.0005960424528721026, "delta_db": -0.10715428073223744, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027595006700553216, "output_power": 0.00026507336098645763, "delta_db": -0.17464419117254465, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003628265208503686, "output_power": 0.0003368948398858065, "delta_db": -0.32204665280678474, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001954243236049976, "output_power": 0.00018184861118284852, "delta_db": -0.3126862886617132, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1827381443668609, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08321687489820657, "pitch": null, "output_rms_dbfs": -26.077955791611693, "output_peak": 0.4749984240989598, "reference_rms_dbfs": -25.994738916713484, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.202082848864051}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017615675305240148, "log_rms_mae_db": 0.6636904638215709, "rms_envelope_nrmse": 0.07552218406585766, "envelope_correlation": 0.9949670861016543, "rms_gain_error_db": -0.08311883364419698, "reference_mean_power": 0.0025168568603362385, "output_mean_power": 0.00246914522168612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0162973895242792e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002469993456433397, "output_power": 0.000240973835540177, "reference_active": true, "delta_db": -0.10725912414173609}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010173002443957726, "output_power": 8.550459044638655e-05, "reference_active": true, "delta_db": -0.7545971770642147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1084315450169838e-05, "output_power": 1.8511359037920342e-05, "reference_active": true, "delta_db": -0.5652120095601552}], "active_band_count": 3, "active_band_mae_db": 0.4756894369220353}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016525538510559444, "log_rms_mae_db": 0.3491420604737422, "rms_envelope_nrmse": 0.045388598772396094, "envelope_correlation": 0.998135676998545, "rms_gain_error_db": -0.08297484076630458, "reference_mean_power": 0.002527970677484516, "output_mean_power": 0.002480130584861692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.625293642327984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023064746753792456, "output_power": 0.0002234949645968669, "reference_active": true, "delta_db": -0.13680947701933086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.045652222183457e-05, "output_power": 4.85818733112378e-05, "reference_active": true, "delta_db": -0.9496890351474285}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2874374178901008e-06, "output_power": 1.6625238486593115e-06, "reference_active": true, "delta_db": -1.3858133710739462}], "active_band_count": 3, "active_band_mae_db": 0.8241039610802353}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001279910087199623, "log_rms_mae_db": 1.0693614463344778, "rms_envelope_nrmse": 0.06569887722058358, "envelope_correlation": 0.9969427063729766, "rms_gain_error_db": -0.05848537795365506, "reference_mean_power": 0.0009939686093891967, "output_mean_power": 0.00098067280333489}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.147046828130101e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013887130907316313, "output_power": 0.00013135194889704084, "reference_active": true, "delta_db": -0.24176008517082018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4048550736300918e-05, "output_power": 2.2721864028253236e-05, "reference_active": true, "delta_db": -0.24644952527140648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.939409383024551e-06, "output_power": 5.704668927278176e-06, "reference_active": true, "delta_db": 0.6255542263346281}], "active_band_count": 3, "active_band_mae_db": 0.37125461225895157}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011859895721274409, "log_rms_mae_db": 0.7224653478882339, "rms_envelope_nrmse": 0.03911916574668919, "envelope_correlation": 0.9988984452819337, "rms_gain_error_db": -0.05820724781169322, "reference_mean_power": 0.0009986896907851761, "output_mean_power": 0.0009853938376628599}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.009742136856967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012678804378904424, "output_power": 0.0001192560991442993, "reference_active": true, "delta_db": -0.2659770165910871}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.298881565077983e-05, "output_power": 1.244947910871813e-05, "reference_active": true, "delta_db": -0.18418372183822043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8039990598306815e-07, "output_power": 2.6735061660797434e-07, "reference_active": true, "delta_db": -0.20696673598601878}], "active_band_count": 3, "active_band_mae_db": 0.21904249147177543}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.31629258809413e-05, "log_rms_mae_db": 1.0385656200189248, "rms_envelope_nrmse": 0.13862500184465226, "envelope_correlation": 0.9854653208217867, "rms_gain_error_db": -0.17366777283448892, "reference_mean_power": 0.0004146029386445929, "output_mean_power": 0.0003983507128746483}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1875762004578613e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604229953230037e-05, "output_power": 8.276061412681676e-05, "reference_active": true, "delta_db": -0.16888303955571737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2897806527831664e-05, "output_power": 2.4465858602051684e-05, "reference_active": true, "delta_db": 0.28766579911724893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2516459896674564e-05, "output_power": 1.2165573812766269e-05, "reference_active": true, "delta_db": -0.12348914114456622}], "active_band_count": 3, "active_band_mae_db": 0.19334599327251087}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.786085823827924e-05, "log_rms_mae_db": 0.6633998071978133, "rms_envelope_nrmse": 0.0821699531562241, "envelope_correlation": 0.9947857760030427, "rms_gain_error_db": -0.1736573572932311, "reference_mean_power": 0.0004166837249389974, "output_mean_power": 0.0004003508935445489}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.035015228514449e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.632173919031572e-05, "output_power": 8.399371976187173e-05, "reference_active": true, "delta_db": -0.11873366893011619}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4669499526204761e-05, "output_power": 1.4407503902630291e-05, "reference_active": true, "delta_db": -0.07826551541751771}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2709360758910173e-06, "output_power": 1.0781821167442369e-06, "reference_active": true, "delta_db": -0.714315833176747}], "active_band_count": 3, "active_band_mae_db": 0.303771672508127}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011544071276927888, "log_rms_mae_db": 1.038579439402059, "rms_envelope_nrmse": 0.16227575470841682, "envelope_correlation": 0.9814913240607088, "rms_gain_error_db": -0.4485609810258324, "reference_mean_power": 0.00033862428581899314, "output_mean_power": 0.0003053950573657556}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7742895295757852e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.9949103091747194e-05, "output_power": 4.9153648445891514e-05, "reference_active": true, "delta_db": -0.8622693091648636}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5523159384577664e-05, "output_power": 3.0985680595534245e-05, "reference_active": true, "delta_db": -1.670713554081036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.055143212029728e-06, "output_power": 4.009154837104044e-06, "reference_active": true, "delta_db": -2.454530044924455}], "active_band_count": 3, "active_band_mae_db": 1.6625043027234516}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.669712818495073e-05, "log_rms_mae_db": 0.6460161817909896, "rms_envelope_nrmse": 0.11462264836336071, "envelope_correlation": 0.9911734680592219, "rms_gain_error_db": -0.44855721802746107, "reference_mean_power": 0.00034032151520785135, "output_mean_power": 0.00030692600343765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6870888960608118e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.86684133644439e-05, "output_power": 5.3006316368018494e-05, "reference_active": true, "delta_db": -1.1242938848413924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.573388989930172e-05, "output_power": 1.7285310762502592e-05, "reference_active": true, "delta_db": -1.728282466519282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.935502106463895e-07, "output_power": 4.317006365175037e-07, "reference_active": true, "delta_db": -3.1593627253682084}], "active_band_count": 3, "active_band_mae_db": 2.003979692242961}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8913332882670383, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "65d1b8e1b0f36f5f71f0959b761d8872590aeb92773fa7a39c448ca2a2d4dfd7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9866587777088405, "standard_deviation": 0.0016211391273672508, "minimum": 0.9837295921961094, "maximum": 0.9888948143825244, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0342122350851228, "standard_deviation": 0.02543235502162195, "minimum": 0.9891993343609432, "maximum": 1.0702389554978073, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7755120254910646, "standard_deviation": 0.3994732949162883, "minimum": 0.19334599327251087, "maximum": 1.3271362536576972, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13288046366024658, "standard_deviation": 0.0076713688040107204, "minimum": 0.12109978754381547, "maximum": 0.14618050203884633, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962165446771268, "standard_deviation": 0.0007168702531222242, "minimum": 0.9947857760030427, "maximum": 0.9969753713834423, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.655578091380167, "standard_deviation": 0.021429470327040866, "minimum": 0.6310606242394363, "maximum": 0.6959644717780443, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1503224468367452, "standard_deviation": 0.5705947177907722, "minimum": 0.303771672508127, "maximum": 1.884279305940769, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07079102341402325, "standard_deviation": 0.006072833302194936, "minimum": 0.06253668699144345, "maximum": 0.0821699531562241, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9806124937691783, "standard_deviation": 0.0029092712759676014, "minimum": 0.9766958712003531, "maximum": 0.984761821439452, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.0525289431678349, "standard_deviation": 0.025020204542170597, "minimum": 1.0220431633587277, "maximum": 1.0926353596231413, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2355273768140829, "standard_deviation": 0.5840444139660965, "minimum": 0.2992645090965923, "maximum": 1.8314396684445766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1647830516100161, "standard_deviation": 0.013185053919024105, "minimum": 0.1435641344835441, "maximum": 0.18391136880805192, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935307519263492, "standard_deviation": 0.001683833415275569, "minimum": 0.9911734680592219, "maximum": 0.9958513320746166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.629023659954516, "standard_deviation": 0.02727780463242485, "minimum": 0.5946198938892286, "maximum": 0.6647899537983913, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4282767946282808, "standard_deviation": 0.5972110244813486, "minimum": 0.5230617344492099, "maximum": 2.051866824102079, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09986204457018938, "standard_deviation": 0.01244472031306265, "minimum": 0.0825135844563978, "maximum": 0.11462264836336071, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9969421159356575, "standard_deviation": 0.0002272387912533276, "minimum": 0.9966606807079783, "maximum": 0.9973253654004741, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0707943611231694, "standard_deviation": 0.023631573098447118, "minimum": 1.0302590230438429, "maximum": 1.0982723525781661, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2942467674838841, "standard_deviation": 0.10220080487141725, "minimum": 0.1292134876345293, "maximum": 0.4187239989330924, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06568276402549378, "standard_deviation": 0.0024638251598664215, "minimum": 0.061762501456424, "maximum": 0.06894253943720458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9990601321402661, "standard_deviation": 0.00012755225239572093, "minimum": 0.9988984452819337, "maximum": 0.9992802587975019, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7212081747010306, "standard_deviation": 0.03004703523606449, "minimum": 0.681179551699119, "maximum": 0.7662326565473975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3985623826041419, "standard_deviation": 0.35254067534625216, "minimum": 0.10176593166481228, "maximum": 0.9261514000949784, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.035915993436526794, "standard_deviation": 0.0024966323753740113, "minimum": 0.03187479102315695, "maximum": 0.03911916574668919, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1739868515933675, "standard_deviation": 0.0437643249951521, "minimum": 0.11415269972637314, "maximum": 0.2497565500608988, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.208779078831065, "standard_deviation": 0.005562273710048402, "minimum": 4.202082848864051, "maximum": 4.216776008972789, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4849100126493065, "standard_deviation": 0.011500649604410296, "minimum": 0.4717582102277697, "maximum": 0.501863189066651, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06389215008194651, "standard_deviation": 0.03384963645482513, "minimum": -0.11099050115264547, "maximum": -0.030399201664253847, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951591806754682, "standard_deviation": 0.0005088823819156552, "minimum": 0.9945155910209711, "maximum": 0.9958884725592062, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6626349915126972, "standard_deviation": 0.023947409168130013, "minimum": 0.6356728203183621, "maximum": 0.7153553836442574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6852248956250617, "standard_deviation": 0.20558786741843363, "minimum": 0.4547128819749242, "maximum": 1.003435962513423, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07383390007740553, "standard_deviation": 0.003881998610458549, "minimum": 0.0679201404757471, "maximum": 0.07899037516944465, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986694107542201, "standard_deviation": 0.0002947814138727174, "minimum": 0.998135676998545, "maximum": 0.9990207406522174, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.35231914507400175, "standard_deviation": 0.028322287525972666, "minimum": 0.3171899215408448, "maximum": 0.4081044027122606, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.017976924439128, "standard_deviation": 0.25043925541590156, "minimum": 0.7309301077931059, "maximum": 1.367338905379755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0381403982955231, "standard_deviation": 0.0041254432848088374, "minimum": 0.032388515120648234, "maximum": 0.045388598772396094, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_121-127105-0035", "language": "en", "style": "normal", "speaker": "librispeech_121", "source_utterance_id": "121-127105-0035", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_121-127105-0035.wav", "source_file_sha256": "410d21641211b5289dc8b6fe6dac37e5820a2dece01f00e7e43c1143e5e19cb9", "decoded_pcm_sha256": "48306d9610578521a036cf3d52224eb0076f9d14ddb6850ac09278e898fc7303", "original_source_sha256": "7e293a1a5f64d336563f850f69929dd6b112e3ce37913f4422f2c29e9c55042d", "corpus_entry_id": "librispeech_121-127105-0035"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.8925719261169434, "predicted_mos": 3.8925719261169434, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cddb4be4134f4baa46869e07aa6333d20588e11279af60c4637072acc842238d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2092229002155364, "cpu_threads": 4}, "waveform_sha256": "502b0e88139f1eaaba84c2464122ef75f0d24e733969e87e335a26d172b18038"}, "rows": [{"case": "librispeech_121-127105-0035", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.2551629543304443, "predicted_mos": 2.2551629543304443, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5b32bbe72d2a0574ac3d06fd370e5768a988a2e5ef09867e5ee0e1213bbce370", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0198037000373006, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6372705335461756, "全帯域_10ms_rms_envelope_nrmse": 0.08825130794523928, "全帯域_10ms_envelope_correlation": 0.9926839762491118, "全帯域_10ms_modulation_mae_db": 0.25949806540259107, "全帯域_40ms_log_rms_mae_db": 0.3662120983626934, "全帯域_40ms_rms_envelope_nrmse": 0.050671526113528306, "全帯域_40ms_envelope_correlation": 0.9973786113944672, "全帯域_40ms_modulation_mae_db": 0.3471252462582574, "300–1000 Hz_10ms_log_rms_mae_db": 1.3301368532897542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10433056691113124, "300–1000 Hz_10ms_envelope_correlation": 0.992254722705086, "300–1000 Hz_10ms_modulation_mae_db": 0.44127888491989004, "300–1000 Hz_40ms_log_rms_mae_db": 0.8383535760916417, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0677387989189024, "300–1000 Hz_40ms_envelope_correlation": 0.9966114529859244, "300–1000 Hz_40ms_modulation_mae_db": 0.6670401646685017, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1353266282738494, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1589640923441597, "1000–3000 Hz_10ms_envelope_correlation": 0.9810815866720939, "1000–3000 Hz_10ms_modulation_mae_db": 0.6872414181138646, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6927675070021382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08291638966010204, "1000–3000 Hz_40ms_envelope_correlation": 0.9944340520075444, "1000–3000 Hz_40ms_modulation_mae_db": 0.6738194682979336, "3000–6000 Hz_10ms_log_rms_mae_db": 1.167649699805804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17027726545927319, "3000–6000 Hz_10ms_envelope_correlation": 0.9838963817762867, "3000–6000 Hz_10ms_modulation_mae_db": 0.5490931946708107, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8593332419088194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10585581620933392, "3000–6000 Hz_40ms_envelope_correlation": 0.9939299699676681, "3000–6000 Hz_40ms_modulation_mae_db": 0.586296182509877, "mr_stft_db": 4.394819865764792, "band_mae_db": 0.16978324596057773, "rms_gain_error_db": -0.007982442129467417, "output_peak": 0.435300159878377}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6671763741242306, "全帯域_10ms_rms_envelope_nrmse": 0.09273030220299991, "全帯域_10ms_envelope_correlation": 0.9920166423406533, "全帯域_10ms_modulation_mae_db": 0.3818553824369093, "全帯域_40ms_log_rms_mae_db": 0.3610979828878998, "全帯域_40ms_rms_envelope_nrmse": 0.05240389580039904, "全帯域_40ms_envelope_correlation": 0.9973778445758332, "全帯域_40ms_modulation_mae_db": 0.671606099674201, "300–1000 Hz_10ms_log_rms_mae_db": 1.3100315367150266, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10075458724930679, "300–1000 Hz_10ms_envelope_correlation": 0.9927360868297938, "300–1000 Hz_10ms_modulation_mae_db": 0.6072218091962869, "300–1000 Hz_40ms_log_rms_mae_db": 0.9200641281221653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06846553946494438, "300–1000 Hz_40ms_envelope_correlation": 0.9964496357638083, "300–1000 Hz_40ms_modulation_mae_db": 0.7644435188997588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1585115663369823, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16709416718870373, "1000–3000 Hz_10ms_envelope_correlation": 0.9790643280824816, "1000–3000 Hz_10ms_modulation_mae_db": 0.2971343284624826, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6931847878119622, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08634050472247348, "1000–3000 Hz_40ms_envelope_correlation": 0.9939832800482395, "1000–3000 Hz_40ms_modulation_mae_db": 0.4776981776697446, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1917433344523585, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1914704848456402, "3000–6000 Hz_10ms_envelope_correlation": 0.9788022425703585, "3000–6000 Hz_10ms_modulation_mae_db": 0.7085357824599517, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9277370250980516, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13106194981306776, "3000–6000 Hz_40ms_envelope_correlation": 0.9907697627313283, "3000–6000 Hz_40ms_modulation_mae_db": 0.7126553267915235, "mr_stft_db": 4.40241214473465, "band_mae_db": 0.14208660237206636, "rms_gain_error_db": -0.13119187979848632, "output_peak": 0.46800923613119444}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6638115072477407, "全帯域_10ms_rms_envelope_nrmse": 0.09859235524741358, "全帯域_10ms_envelope_correlation": 0.9908650420908661, "全帯域_10ms_modulation_mae_db": 0.5038911842509489, "全帯域_40ms_log_rms_mae_db": 0.3564747914334871, "全帯域_40ms_rms_envelope_nrmse": 0.04944000840211199, "全帯域_40ms_envelope_correlation": 0.9974997472992391, "全帯域_40ms_modulation_mae_db": 0.5047282493274001, "300–1000 Hz_10ms_log_rms_mae_db": 1.2863990929923983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09580862858912999, "300–1000 Hz_10ms_envelope_correlation": 0.9934882553205305, "300–1000 Hz_10ms_modulation_mae_db": 0.402282590861667, "300–1000 Hz_40ms_log_rms_mae_db": 0.8101618116747141, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05027256808452329, "300–1000 Hz_40ms_envelope_correlation": 0.9981387276572368, "300–1000 Hz_40ms_modulation_mae_db": 0.6329263069024652, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1283909197894426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.153767312410394, "1000–3000 Hz_10ms_envelope_correlation": 0.9822769773412721, "1000–3000 Hz_10ms_modulation_mae_db": 0.6050895468550102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.669379274838148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07687774766701162, "1000–3000 Hz_40ms_envelope_correlation": 0.9953193596841038, "1000–3000 Hz_40ms_modulation_mae_db": 0.7152283253002177, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2030509122135615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21184425230292828, "3000–6000 Hz_10ms_envelope_correlation": 0.9740768548806702, "3000–6000 Hz_10ms_modulation_mae_db": 0.7049731620404164, "3000–6000 Hz_40ms_log_rms_mae_db": 0.92713905470803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12212131542604268, "3000–6000 Hz_40ms_envelope_correlation": 0.9911574698399661, "3000–6000 Hz_40ms_modulation_mae_db": 0.791667231753757, "mr_stft_db": 4.414393088325988, "band_mae_db": 0.11194231538512893, "rms_gain_error_db": -0.004730431832790509, "output_peak": 0.39520230450211963}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6484674050151129, "全帯域_10ms_rms_envelope_nrmse": 0.10207302345076157, "全帯域_10ms_envelope_correlation": 0.9902828788395708, "全帯域_10ms_modulation_mae_db": 0.8918520647833005, "全帯域_40ms_log_rms_mae_db": 0.35178444294916306, "全帯域_40ms_rms_envelope_nrmse": 0.06690698697968345, "全帯域_40ms_envelope_correlation": 0.9954767319040763, "全帯域_40ms_modulation_mae_db": 1.1234100735742538, "300–1000 Hz_10ms_log_rms_mae_db": 1.3008346458125515, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11236762592503453, "300–1000 Hz_10ms_envelope_correlation": 0.9909644387799891, "300–1000 Hz_10ms_modulation_mae_db": 0.631398378572306, "300–1000 Hz_40ms_log_rms_mae_db": 0.8271817331554374, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06588193461836463, "300–1000 Hz_40ms_envelope_correlation": 0.9967371560351473, "300–1000 Hz_40ms_modulation_mae_db": 0.4022010244380316, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1466393123012255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1500493971432861, "1000–3000 Hz_10ms_envelope_correlation": 0.9834212701314995, "1000–3000 Hz_10ms_modulation_mae_db": 1.0495215205748627, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7262540181301622, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0823082393875062, "1000–3000 Hz_40ms_envelope_correlation": 0.9951050496085824, "1000–3000 Hz_40ms_modulation_mae_db": 1.1751123531065124, "3000–6000 Hz_10ms_log_rms_mae_db": 1.168629207952041, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22903534917839255, "3000–6000 Hz_10ms_envelope_correlation": 0.9740743758832578, "3000–6000 Hz_10ms_modulation_mae_db": 1.9069832504181665, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8669855359816434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17096469606719844, "3000–6000 Hz_40ms_envelope_correlation": 0.9863389411226398, "3000–6000 Hz_40ms_modulation_mae_db": 2.017174832609306, "mr_stft_db": 4.40269965873502, "band_mae_db": 0.1882003300666095, "rms_gain_error_db": 0.011452998829279174, "output_peak": 0.5041658471678924}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6391224088516704, "全帯域_10ms_rms_envelope_nrmse": 0.08073049593533696, "全帯域_10ms_envelope_correlation": 0.9938778445843574, "全帯域_10ms_modulation_mae_db": 0.590586919438019, "全帯域_40ms_log_rms_mae_db": 0.37313232513076827, "全帯域_40ms_rms_envelope_nrmse": 0.04379142546133314, "全帯域_40ms_envelope_correlation": 0.9980552748193151, "全帯域_40ms_modulation_mae_db": 0.7193125275290155, "300–1000 Hz_10ms_log_rms_mae_db": 1.2727978502081747, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10783664739920369, "300–1000 Hz_10ms_envelope_correlation": 0.991751075134577, "300–1000 Hz_10ms_modulation_mae_db": 0.39134991131727737, "300–1000 Hz_40ms_log_rms_mae_db": 0.8155859707127425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07051023973451787, "300–1000 Hz_40ms_envelope_correlation": 0.9963750940667121, "300–1000 Hz_40ms_modulation_mae_db": 0.3564935896426895, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1493333801592631, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15235986599386092, "1000–3000 Hz_10ms_envelope_correlation": 0.9827379035404616, "1000–3000 Hz_10ms_modulation_mae_db": 0.9350894011327818, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7090744130095956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07597809278750906, "1000–3000 Hz_40ms_envelope_correlation": 0.995734227501182, "1000–3000 Hz_40ms_modulation_mae_db": 1.0177552892872341, "3000–6000 Hz_10ms_log_rms_mae_db": 1.147141113625797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.148245057613783, "3000–6000 Hz_10ms_envelope_correlation": 0.9869042513844871, "3000–6000 Hz_10ms_modulation_mae_db": 0.767161988522823, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8279379710556466, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07709179698302737, "3000–6000 Hz_40ms_envelope_correlation": 0.9962684414107722, "3000–6000 Hz_40ms_modulation_mae_db": 0.9141523464481268, "mr_stft_db": 4.384498130368912, "band_mae_db": 0.12973158011982885, "rms_gain_error_db": -0.047331720832391455, "output_peak": 0.4084242121685332}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6668113741467748, "全帯域_10ms_rms_envelope_nrmse": 0.09456508969490825, "全帯域_10ms_envelope_correlation": 0.9916428047663286, "全帯域_10ms_modulation_mae_db": 0.6496065903211646, "全帯域_40ms_log_rms_mae_db": 0.3815926438499595, "全帯域_40ms_rms_envelope_nrmse": 0.055040544476282725, "全帯域_40ms_envelope_correlation": 0.997010231059968, "全帯域_40ms_modulation_mae_db": 0.4786207895332149, "300–1000 Hz_10ms_log_rms_mae_db": 1.3253515376650344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11283089944974375, "300–1000 Hz_10ms_envelope_correlation": 0.9909615607040482, "300–1000 Hz_10ms_modulation_mae_db": 0.5459593403375343, "300–1000 Hz_40ms_log_rms_mae_db": 0.8744038387195799, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06844955573924652, "300–1000 Hz_40ms_envelope_correlation": 0.9965716676046843, "300–1000 Hz_40ms_modulation_mae_db": 0.6369886639210645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.14274528083787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16976148003151376, "1000–3000 Hz_10ms_envelope_correlation": 0.978359340324465, "1000–3000 Hz_10ms_modulation_mae_db": 0.8389041834336531, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7346998422718894, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11599271640726379, "1000–3000 Hz_40ms_envelope_correlation": 0.9891648942422137, "1000–3000 Hz_40ms_modulation_mae_db": 1.0083413296178387, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1926506506568322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18783908232371802, "3000–6000 Hz_10ms_envelope_correlation": 0.9784868266069795, "3000–6000 Hz_10ms_modulation_mae_db": 0.4987653479074136, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8980042762901879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10113482491193856, "3000–6000 Hz_40ms_envelope_correlation": 0.9934794631143228, "3000–6000 Hz_40ms_modulation_mae_db": 0.6306846079020535, "mr_stft_db": 4.402339662513293, "band_mae_db": 0.11935781173995641, "rms_gain_error_db": -0.10077463734167402, "output_peak": 0.4335724521289224}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6420493721327555, "全帯域_10ms_rms_envelope_nrmse": 0.08838751777065584, "全帯域_10ms_envelope_correlation": 0.9928634610363816, "全帯域_10ms_modulation_mae_db": 0.7169762074785941, "全帯域_40ms_log_rms_mae_db": 0.3505061115409527, "全帯域_40ms_rms_envelope_nrmse": 0.050950762516659184, "全帯域_40ms_envelope_correlation": 0.9976314904124569, "全帯域_40ms_modulation_mae_db": 0.9070536798608225, "300–1000 Hz_10ms_log_rms_mae_db": 1.2281447327239199, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1031767594327376, "300–1000 Hz_10ms_envelope_correlation": 0.992490945493254, "300–1000 Hz_10ms_modulation_mae_db": 0.5479201309997717, "300–1000 Hz_40ms_log_rms_mae_db": 0.8366244477484294, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06529990012309339, "300–1000 Hz_40ms_envelope_correlation": 0.9969297045130674, "300–1000 Hz_40ms_modulation_mae_db": 0.7802396433918188, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1865023779380008, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16552575365233085, "1000–3000 Hz_10ms_envelope_correlation": 0.9794933891708403, "1000–3000 Hz_10ms_modulation_mae_db": 0.7329597331289119, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7287501976518271, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08585808854950018, "1000–3000 Hz_40ms_envelope_correlation": 0.9943402895702832, "1000–3000 Hz_40ms_modulation_mae_db": 0.9373997895181988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1880297412965473, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17398012019327613, "3000–6000 Hz_10ms_envelope_correlation": 0.9834614349526186, "3000–6000 Hz_10ms_modulation_mae_db": 1.0858881008165406, "3000–6000 Hz_40ms_log_rms_mae_db": 0.91018357576479, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12747092773423346, "3000–6000 Hz_40ms_envelope_correlation": 0.9920652008288271, "3000–6000 Hz_40ms_modulation_mae_db": 1.3524426687588587, "mr_stft_db": 4.404072482019504, "band_mae_db": 0.17853364028226015, "rms_gain_error_db": -0.15338025028486532, "output_peak": 0.42638473892118617}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6774135438459166, "全帯域_10ms_rms_envelope_nrmse": 0.09341839980599716, "全帯域_10ms_envelope_correlation": 0.9917957647137209, "全帯域_10ms_modulation_mae_db": 0.11686293603762193, "全帯域_40ms_log_rms_mae_db": 0.38845332525039333, "全帯域_40ms_rms_envelope_nrmse": 0.04707853792195731, "全帯域_40ms_envelope_correlation": 0.9977760751146735, "全帯域_40ms_modulation_mae_db": 0.1235067833468304, "300–1000 Hz_10ms_log_rms_mae_db": 1.2686918418679878, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10134335180856606, "300–1000 Hz_10ms_envelope_correlation": 0.9926732743480972, "300–1000 Hz_10ms_modulation_mae_db": 0.27062312784115805, "300–1000 Hz_40ms_log_rms_mae_db": 0.7937928866763316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056217730415284134, "300–1000 Hz_40ms_envelope_correlation": 0.9976684585286145, "300–1000 Hz_40ms_modulation_mae_db": 0.19508047315652766, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1810536761765216, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17636942308553097, "1000–3000 Hz_10ms_envelope_correlation": 0.9768414652889329, "1000–3000 Hz_10ms_modulation_mae_db": 0.4862490629459157, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7378324581872016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10448088131182398, "1000–3000 Hz_40ms_envelope_correlation": 0.9917664632742019, "1000–3000 Hz_40ms_modulation_mae_db": 0.5021901102307473, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1955154300386135, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1782190958211607, "3000–6000 Hz_10ms_envelope_correlation": 0.9808300208465873, "3000–6000 Hz_10ms_modulation_mae_db": 1.0902433283471793, "3000–6000 Hz_40ms_log_rms_mae_db": 0.913801430500688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10191217841661616, "3000–6000 Hz_40ms_envelope_correlation": 0.9933324846856771, "3000–6000 Hz_40ms_modulation_mae_db": 1.093645238750314, "mr_stft_db": 4.408779650351803, "band_mae_db": 0.13038806728968386, "rms_gain_error_db": -0.06773324065780326, "output_peak": 0.4205079998653818}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.396644573717405, "spectral_convergence": 0.37256858507227264, "spectral_convergence_status": "ok", "amplitude_floor": 1.0971770400792783e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.348202352761786, "spectral_convergence": 0.36575460802857956, "spectral_convergence_status": "ok", "amplitude_floor": 6.752711905169626e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.439612670815187, "spectral_convergence": 0.3877909438836425, "spectral_convergence_status": "ok", "amplitude_floor": 5.858543156261471e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003996059347809583, "output_power": 0.00039536029879735175, "delta_db": -0.04638873282294374, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00030506165058519967, "output_power": 0.000297367159324179, "delta_db": -0.11094611591478203, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002850207180759584, "output_power": 0.00027940609656586506, "delta_db": -0.08640551829544081, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017376514501848305, "output_power": 0.0001669510148545685, "delta_db": -0.17373603771376564, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002541837569284722, "output_power": 0.0002827730340090238, "delta_db": 0.4629019717801602, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023122095798090296, "output_power": 0.00022397270041517894, "delta_db": -0.13832109923637398, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16978324596057773, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.007982442129467417, "pitch": null, "output_rms_dbfs": -27.83614945460531, "output_peak": 0.435300159878377, "reference_rms_dbfs": -27.828167012475838, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.394819865764792}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015167567220720444, "log_rms_mae_db": 0.6372705335461756, "rms_envelope_nrmse": 0.08825130794523928, "envelope_correlation": 0.9926839762491118, "rms_gain_error_db": -0.007982421265291183, "reference_mean_power": 0.0016502333338478435, "output_mean_power": 0.001647202957080898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103350035549285e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021122178949431673, "output_power": 0.00021405028339985203, "reference_active": true, "delta_db": 0.05777089478681867}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018786794884032712, "output_power": 0.0001918737036727536, "reference_active": true, "delta_db": 0.09162764943527313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.617925051494854e-05, "output_power": 2.2648927614857296e-05, "reference_active": true, "delta_db": -0.6290956519856814}], "active_band_count": 3, "active_band_mae_db": 0.25949806540259107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014379702931337134, "log_rms_mae_db": 0.3662120983626934, "rms_envelope_nrmse": 0.050671526113528306, "envelope_correlation": 0.9973786113944672, "rms_gain_error_db": -0.007982202821972624, "reference_mean_power": 0.0016585326982667058, "output_mean_power": 0.0016554871643772492}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0955155560805602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019170493341265238, "output_power": 0.00019747557624907758, "reference_active": true, "delta_db": 0.1288010047212682}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.87936190430084e-05, "output_power": 8.107318564685018e-05, "reference_active": true, "delta_db": 0.12386189920053808}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1061472825588806e-06, "output_power": 1.7563758403235146e-06, "reference_active": true, "delta_db": -0.7887128348529658}], "active_band_count": 3, "active_band_mae_db": 0.3471252462582574}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010769717002798511, "log_rms_mae_db": 1.3301368532897542, "rms_envelope_nrmse": 0.10433056691113124, "envelope_correlation": 0.992254722705086, "rms_gain_error_db": -0.11679586045216922, "reference_mean_power": 0.0005226470255969395, "output_mean_power": 0.0005087786716593647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.88003979644585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021484485335499559, "output_power": 0.00021003344908724627, "reference_active": true, "delta_db": -0.09836490301867154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0907744988019035e-05, "output_power": 3.2677202697558254e-05, "reference_active": true, "delta_db": -0.9756066790198531}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.643789476493196e-06, "output_power": 8.160516995870366e-06, "reference_active": true, "delta_db": -0.24986507272114544}], "active_band_count": 3, "active_band_mae_db": 0.44127888491989004}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.974444405364767e-05, "log_rms_mae_db": 0.8383535760916417, "rms_envelope_nrmse": 0.0677387989189024, "envelope_correlation": 0.9966114529859244, "rms_gain_error_db": -0.11679584972923922, "reference_mean_power": 0.0005252755682472565, "output_mean_power": 0.0005113374676198907}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.5461080823927677e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019511319942898067, "output_power": 0.00019112928069399242, "reference_active": true, "delta_db": -0.08959425098837889}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8168801206395614e-05, "output_power": 1.4741663871945547e-05, "reference_active": true, "delta_db": -0.907797685898984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.358572422696516e-07, "output_power": 5.840105771979642e-07, "reference_active": true, "delta_db": -1.0037285571181422}], "active_band_count": 3, "active_band_mae_db": 0.6670401646685017}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.224188331064728e-05, "log_rms_mae_db": 1.1353266282738494, "rms_envelope_nrmse": 0.1589640923441597, "envelope_correlation": 0.9810815866720939, "rms_gain_error_db": -0.049814961264662595, "reference_mean_power": 0.00028515402410464236, "output_mean_power": 0.0002819019036487159}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.768616554938329e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001096694429350374, "output_power": 0.00011075513376878291, "reference_active": true, "delta_db": 0.0427822848258385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.898846044442003e-05, "output_power": 4.728066728329194e-05, "reference_active": true, "delta_db": -0.1541019392269506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.522948535574358e-06, "output_power": 5.547609242122606e-06, "reference_active": true, "delta_db": -1.8648400302888046}], "active_band_count": 3, "active_band_mae_db": 0.6872414181138646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.641451530936745e-05, "log_rms_mae_db": 0.6927675070021382, "rms_envelope_nrmse": 0.08291638966010204, "envelope_correlation": 0.9944340520075444, "rms_gain_error_db": -0.04981514247096969, "reference_mean_power": 0.0002865881294452572, "output_mean_power": 0.00028331964150434224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.5110883907146213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010245851234810692, "output_power": 0.00010381436672414842, "reference_active": true, "delta_db": 0.057094132498535885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5140429681693116e-05, "output_power": 2.3889654344815227e-05, "reference_active": true, "delta_db": -0.22162829989271465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.446669785409335e-07, "output_power": 4.985267867391718e-07, "reference_active": true, "delta_db": -1.7427359725025502}], "active_band_count": 3, "active_band_mae_db": 0.6738194682979336}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001420862345348913, "log_rms_mae_db": 1.167649699805804, "rms_envelope_nrmse": 0.17027726545927319, "envelope_correlation": 0.9838963817762867, "rms_gain_error_db": 0.3061994584579802, "reference_mean_power": 0.00023492613523189528, "output_mean_power": 0.0002520874818397592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4459776587443134e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000158142036380199, "output_power": 0.00018183248923532753, "reference_active": true, "delta_db": 0.6062415727206126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.639730090455687e-05, "output_power": 0.00011176456137368313, "reference_active": true, "delta_db": 0.6423924408530673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.267646990584481e-05, "output_power": 1.389513871794962e-05, "reference_active": true, "delta_db": 0.3986455704387523}], "active_band_count": 3, "active_band_mae_db": 0.5490931946708107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010454414213188805, "log_rms_mae_db": 0.8593332419088194, "rms_envelope_nrmse": 0.10585581620933392, "envelope_correlation": 0.9939299699676681, "rms_gain_error_db": 0.3061994436091413, "reference_mean_power": 0.00023610764979113905, "output_mean_power": 0.00025335530513323983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2342047370765003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017362277943143474, "output_power": 0.00020300312362260645, "reference_active": true, "delta_db": 0.6789601616097403}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.8688155273395004e-05, "output_power": 5.661541174501362e-05, "reference_active": true, "delta_db": 0.6551135009831024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.341220903531162e-06, "output_power": 1.2162387187612558e-06, "reference_active": true, "delta_db": -0.42481488493678854}], "active_band_count": 3, "active_band_mae_db": 0.586296182509877}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9055189532603056, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "97b2db2273cfa230a2bb6b8429dfb964771cd8dcf84f135efcfa3a53fa55710f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9804095325690059, "standard_deviation": 0.002330363779457894, "minimum": 0.9768414652889329, "maximum": 0.9834212701314995, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1535628927266444, "standard_deviation": 0.020751169548325257, "minimum": 1.1283909197894426, "maximum": 1.1865023779380008, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7040236493309353, "standard_deviation": 0.2434409596785615, "minimum": 0.2971343284624826, "maximum": 1.0495215205748627, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1617364364812225, "standard_deviation": 0.009390645930655705, "minimum": 0.1500493971432861, "maximum": 0.17636942308553097, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937309519920439, "standard_deviation": 0.002205891678560343, "minimum": 0.9891648942422137, "maximum": 0.995734227501182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7114928123628655, "standard_deviation": 0.024533572461752973, "minimum": 0.669379274838148, "maximum": 0.7378324581872016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8134431053785534, "standard_deviation": 0.25763836147600794, "minimum": 0.4776981776697446, "maximum": 1.1751123531065124, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08884408256164879, "standard_deviation": 0.014057654294699906, "minimum": 0.07597809278750906, "maximum": 0.11599271640726379, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9800665486126557, "standard_deviation": 0.004619933530383662, "minimum": 0.9740743758832578, "maximum": 0.9869042513844871, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1818012612551945, "standard_deviation": 0.018784756883147957, "minimum": 1.147141113625797, "maximum": 1.2030509122135615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9139555193979128, "standard_deviation": 0.4566444237098877, "minimum": 0.4987653479074136, "maximum": 1.9069832504181665, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1863638384672715, "standard_deviation": 0.025156300242496314, "minimum": 0.148245057613783, "maximum": 0.22903534917839255, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921677167126501, "standard_deviation": 0.002926696116313181, "minimum": 0.9863389411226398, "maximum": 0.9962684414107722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8913902639134821, "standard_deviation": 0.03614987469626716, "minimum": 0.8279379710556466, "maximum": 0.9277370250980516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.012339804440477, "standard_deviation": 0.4791802181136861, "minimum": 0.586296182509877, "maximum": 2.017174832609306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11720168819518229, "standard_deviation": 0.0278815691031221, "minimum": 0.07709179698302737, "maximum": 0.17096469606719844, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921650449144219, "standard_deviation": 0.000887269117807354, "minimum": 0.9909615607040482, "maximum": 0.9934882553205305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.290298511409356, "standard_deviation": 0.03373770575939861, "minimum": 1.2281447327239199, "maximum": 1.3301368532897542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4797542717557365, "standard_deviation": 0.12381263882982053, "minimum": 0.27062312784115805, "maximum": 0.631398378572306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1048061333456067, "standard_deviation": 0.005891871553880333, "minimum": 0.09580862858912999, "maximum": 0.11283089944974375, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969352371443994, "standard_deviation": 0.0006337933088513176, "minimum": 0.9963750940667121, "maximum": 0.9981387276572368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8395210491126303, "standard_deviation": 0.04034126961465039, "minimum": 0.7937928866763316, "maximum": 0.9200641281221653, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5544266731276072, "standard_deviation": 0.21120027443221176, "minimum": 0.19508047315652766, "maximum": 0.7802396433918188, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06410453338735958, "standard_deviation": 0.007073608243070002, "minimum": 0.05027256808452329, "maximum": 0.07051023973451787, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14625294915201398, "standard_deviation": 0.028779648602869454, "minimum": 0.11194231538512893, "maximum": 0.1882003300666095, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.401751835351745, "standard_deviation": 0.008960224150563445, "minimum": 4.384498130368912, "maximum": 4.414393088325988, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4364458688454509, "standard_deviation": 0.03470960681489262, "minimum": 0.39520230450211963, "maximum": 0.5041658471678924, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06270895050602489, "standard_deviation": 0.061503065524224625, "minimum": -0.15338025028486532, "maximum": 0.011452998829279174, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920035518276238, "standard_deviation": 0.0011430410453562294, "minimum": 0.9902828788395708, "maximum": 0.9938778445843574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6552653148637971, "standard_deviation": 0.015325257204766603, "minimum": 0.6372705335461756, "maximum": 0.6774135438459166, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5138911687686437, "standard_deviation": 0.2530611453131055, "minimum": 0.11686293603762193, "maximum": 0.8918520647833005, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09234356150666406, "standard_deviation": 0.006624737184860324, "minimum": 0.08073049593533696, "maximum": 0.10207302345076157, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972757508225036, "standard_deviation": 0.0007894555430602937, "minimum": 0.9954767319040763, "maximum": 0.9980552748193151, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3661567151756646, "standard_deviation": 0.013912496639443576, "minimum": 0.3505061115409527, "maximum": 0.38845332525039333, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6094204311379995, "standard_deviation": 0.3166101861448146, "minimum": 0.1235067833468304, "maximum": 1.1234100735742538, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05203546095899439, "standard_deviation": 0.006893444376389028, "minimum": 0.04379142546133314, "maximum": 0.06690698697968345, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-127105-0035", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.602020025253296, "predicted_mos": 2.602020025253296, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ddeaa34407811130ddb96d8a49570487b408d6743b2df3f28f5976f97479c32a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2441376000642776, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6538809732814929, "全帯域_10ms_rms_envelope_nrmse": 0.09260734046125005, "全帯域_10ms_envelope_correlation": 0.9919444377508451, "全帯域_10ms_modulation_mae_db": 0.24421578114805972, "全帯域_40ms_log_rms_mae_db": 0.38356299569138624, "全帯域_40ms_rms_envelope_nrmse": 0.05587240842763272, "全帯域_40ms_envelope_correlation": 0.9968155886703266, "全帯域_40ms_modulation_mae_db": 0.35448012377216337, "300–1000 Hz_10ms_log_rms_mae_db": 1.2570932952043432, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10657300524024065, "300–1000 Hz_10ms_envelope_correlation": 0.9919120663358987, "300–1000 Hz_10ms_modulation_mae_db": 0.6192256540738664, "300–1000 Hz_40ms_log_rms_mae_db": 0.759008416512528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0714083023977861, "300–1000 Hz_40ms_envelope_correlation": 0.996234294256201, "300–1000 Hz_40ms_modulation_mae_db": 0.8179507687418374, "1000–3000 Hz_10ms_log_rms_mae_db": 1.200167329284388, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1582283490782197, "1000–3000 Hz_10ms_envelope_correlation": 0.981231301083322, "1000–3000 Hz_10ms_modulation_mae_db": 0.8850911301155083, "1000–3000 Hz_40ms_log_rms_mae_db": 0.709709618911829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08156267330038608, "1000–3000 Hz_40ms_envelope_correlation": 0.9947471672781322, "1000–3000 Hz_40ms_modulation_mae_db": 0.8854134285736334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2083666313331058, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1903443369906782, "3000–6000 Hz_10ms_envelope_correlation": 0.9807610883678105, "3000–6000 Hz_10ms_modulation_mae_db": 0.6407427531121898, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9059114675751783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12716616313367884, "3000–6000 Hz_40ms_envelope_correlation": 0.9918346431654655, "3000–6000 Hz_40ms_modulation_mae_db": 0.7054045446319983, "mr_stft_db": 4.369467267327221, "band_mae_db": 0.19898196053772088, "rms_gain_error_db": -0.011355546618473189, "output_peak": 0.41766819916771203}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6699407941230988, "全帯域_10ms_rms_envelope_nrmse": 0.09380576309185483, "全帯域_10ms_envelope_correlation": 0.9918009430101202, "全帯域_10ms_modulation_mae_db": 0.3071899889167075, "全帯域_40ms_log_rms_mae_db": 0.3615859948135283, "全帯域_40ms_rms_envelope_nrmse": 0.050805428595069896, "全帯域_40ms_envelope_correlation": 0.9975170755960778, "全帯域_40ms_modulation_mae_db": 0.43675903802441457, "300–1000 Hz_10ms_log_rms_mae_db": 1.2057633223400945, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09870335969166831, "300–1000 Hz_10ms_envelope_correlation": 0.9930176365358561, "300–1000 Hz_10ms_modulation_mae_db": 0.5574469258289381, "300–1000 Hz_40ms_log_rms_mae_db": 0.7684382808400277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06323837051839488, "300–1000 Hz_40ms_envelope_correlation": 0.9969750347129389, "300–1000 Hz_40ms_modulation_mae_db": 0.7536086175529805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2092408397946601, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.170404457929033, "1000–3000 Hz_10ms_envelope_correlation": 0.9782422867902227, "1000–3000 Hz_10ms_modulation_mae_db": 0.24837204781723923, "1000–3000 Hz_40ms_log_rms_mae_db": 0.711457577846285, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0873878037359618, "1000–3000 Hz_40ms_envelope_correlation": 0.9938597812865831, "1000–3000 Hz_40ms_modulation_mae_db": 0.43492610370544976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2195482279320136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19047545852507258, "3000–6000 Hz_10ms_envelope_correlation": 0.9783968544112344, "3000–6000 Hz_10ms_modulation_mae_db": 0.7335916217942152, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9382933814713864, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12019969867013257, "3000–6000 Hz_40ms_envelope_correlation": 0.9917474344188014, "3000–6000 Hz_40ms_modulation_mae_db": 0.4713356708773912, "mr_stft_db": 4.367133644946048, "band_mae_db": 0.13084793227259076, "rms_gain_error_db": -0.12092950611398146, "output_peak": 0.4576205723245004}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6621156462193787, "全帯域_10ms_rms_envelope_nrmse": 0.09818610270060026, "全帯域_10ms_envelope_correlation": 0.9909424698531228, "全帯域_10ms_modulation_mae_db": 0.5474437976472982, "全帯域_40ms_log_rms_mae_db": 0.36448834628594734, "全帯域_40ms_rms_envelope_nrmse": 0.04842432961028974, "全帯域_40ms_envelope_correlation": 0.9976106942246518, "全帯域_40ms_modulation_mae_db": 0.5633919410704133, "300–1000 Hz_10ms_log_rms_mae_db": 1.22972368643194, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09710444065208375, "300–1000 Hz_10ms_envelope_correlation": 0.9933063991823158, "300–1000 Hz_10ms_modulation_mae_db": 0.45272611579674776, "300–1000 Hz_40ms_log_rms_mae_db": 0.7418788255341348, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05329790816798684, "300–1000 Hz_40ms_envelope_correlation": 0.9978800008493516, "300–1000 Hz_40ms_modulation_mae_db": 0.739966954475721, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1892004918745704, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1559454913719135, "1000–3000 Hz_10ms_envelope_correlation": 0.9818098037219138, "1000–3000 Hz_10ms_modulation_mae_db": 0.7364515088360989, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6939472630298493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07986993335457511, "1000–3000 Hz_40ms_envelope_correlation": 0.9950630843116873, "1000–3000 Hz_40ms_modulation_mae_db": 0.8630902075744986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2286270386759317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2115549896702299, "3000–6000 Hz_10ms_envelope_correlation": 0.973938583742768, "3000–6000 Hz_10ms_modulation_mae_db": 0.6629397741478096, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9560504338120527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11671950336038271, "3000–6000 Hz_40ms_envelope_correlation": 0.9918150609787652, "3000–6000 Hz_40ms_modulation_mae_db": 0.7961072303467468, "mr_stft_db": 4.377341441736172, "band_mae_db": 0.10189149067952019, "rms_gain_error_db": -0.010196025847596612, "output_peak": 0.3921347068519494}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6465712633830805, "全帯域_10ms_rms_envelope_nrmse": 0.1048033597431499, "全帯域_10ms_envelope_correlation": 0.9897715622500997, "全帯域_10ms_modulation_mae_db": 1.0483651057416479, "全帯域_40ms_log_rms_mae_db": 0.361895273617411, "全帯域_40ms_rms_envelope_nrmse": 0.06844906205743391, "全帯域_40ms_envelope_correlation": 0.9952672996427362, "全帯域_40ms_modulation_mae_db": 1.3255370715857504, "300–1000 Hz_10ms_log_rms_mae_db": 1.2194010486509603, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1113877573104905, "300–1000 Hz_10ms_envelope_correlation": 0.9911088490810144, "300–1000 Hz_10ms_modulation_mae_db": 0.559935587354903, "300–1000 Hz_40ms_log_rms_mae_db": 0.7474076496863213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06666773165511111, "300–1000 Hz_40ms_envelope_correlation": 0.9966582457390065, "300–1000 Hz_40ms_modulation_mae_db": 0.3524421604034093, "1000–3000 Hz_10ms_log_rms_mae_db": 1.19226943819801, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1514819036981685, "1000–3000 Hz_10ms_envelope_correlation": 0.9830632968202974, "1000–3000 Hz_10ms_modulation_mae_db": 0.9462483322677407, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7378945572226471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07877034754197466, "1000–3000 Hz_40ms_envelope_correlation": 0.9955730549403564, "1000–3000 Hz_40ms_modulation_mae_db": 1.0379975317640626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.195087079856715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23879921328871903, "3000–6000 Hz_10ms_envelope_correlation": 0.9726909212337551, "3000–6000 Hz_10ms_modulation_mae_db": 2.1006744076054855, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9007857276202731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1781294521810949, "3000–6000 Hz_40ms_envelope_correlation": 0.9859474779567196, "3000–6000 Hz_40ms_modulation_mae_db": 2.282905546342926, "mr_stft_db": 4.378384062891916, "band_mae_db": 0.20001827330423097, "rms_gain_error_db": 0.012693983286103572, "output_peak": 0.5054142153022005}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6388021890613905, "全帯域_10ms_rms_envelope_nrmse": 0.08288760383472246, "全帯域_10ms_envelope_correlation": 0.9935394269113276, "全帯域_10ms_modulation_mae_db": 0.5088645477686524, "全帯域_40ms_log_rms_mae_db": 0.3790763346940431, "全帯域_40ms_rms_envelope_nrmse": 0.043126311482948564, "全帯域_40ms_envelope_correlation": 0.9980975628419595, "全帯域_40ms_modulation_mae_db": 0.6753902803747049, "300–1000 Hz_10ms_log_rms_mae_db": 1.225759574980855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10922819046623955, "300–1000 Hz_10ms_envelope_correlation": 0.9914919708679081, "300–1000 Hz_10ms_modulation_mae_db": 0.4096758850866385, "300–1000 Hz_40ms_log_rms_mae_db": 0.7275024319721942, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0682380377736933, "300–1000 Hz_40ms_envelope_correlation": 0.9965743191658935, "300–1000 Hz_40ms_modulation_mae_db": 0.3653753989947803, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1754187838961077, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14821008522031107, "1000–3000 Hz_10ms_envelope_correlation": 0.9836986279876181, "1000–3000 Hz_10ms_modulation_mae_db": 0.8537499236000515, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7152183578278655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07472869139455238, "1000–3000 Hz_40ms_envelope_correlation": 0.9959031353534415, "1000–3000 Hz_40ms_modulation_mae_db": 0.9287398937344635, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1565530018422714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17748209809416446, "3000–6000 Hz_10ms_envelope_correlation": 0.9817707472972659, "3000–6000 Hz_10ms_modulation_mae_db": 0.9198719733588253, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8378399132825268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09405419022271215, "3000–6000 Hz_40ms_envelope_correlation": 0.9948174579101957, "3000–6000 Hz_40ms_modulation_mae_db": 1.0698076893447623, "mr_stft_db": 4.354615576282964, "band_mae_db": 0.14352112811216225, "rms_gain_error_db": -0.024321909031929028, "output_peak": 0.4069884518640391}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6637934640238025, "全帯域_10ms_rms_envelope_nrmse": 0.09528015518857365, "全帯域_10ms_envelope_correlation": 0.9915014870628098, "全帯域_10ms_modulation_mae_db": 0.6708956863614947, "全帯域_40ms_log_rms_mae_db": 0.3811545191711726, "全帯域_40ms_rms_envelope_nrmse": 0.05491389837020743, "全帯域_40ms_envelope_correlation": 0.997005175323235, "全帯域_40ms_modulation_mae_db": 0.49919950904306626, "300–1000 Hz_10ms_log_rms_mae_db": 1.2315832127356818, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11396681064594787, "300–1000 Hz_10ms_envelope_correlation": 0.9907476836799005, "300–1000 Hz_10ms_modulation_mae_db": 0.4526748322607632, "300–1000 Hz_40ms_log_rms_mae_db": 0.7800663745491007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0712249252439797, "300–1000 Hz_40ms_envelope_correlation": 0.9962413385993372, "300–1000 Hz_40ms_modulation_mae_db": 0.5507422362443841, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1990211831558375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1708613209104899, "1000–3000 Hz_10ms_envelope_correlation": 0.9781043253240475, "1000–3000 Hz_10ms_modulation_mae_db": 0.9166902620628882, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7840327936461655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.118577636243806, "1000–3000 Hz_40ms_envelope_correlation": 0.9887296042889229, "1000–3000 Hz_40ms_modulation_mae_db": 1.1215920994818387, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2374293267204861, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1954424411532657, "3000–6000 Hz_10ms_envelope_correlation": 0.976809429961754, "3000–6000 Hz_10ms_modulation_mae_db": 0.6006639332114775, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9420306558811533, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10304688207091461, "3000–6000 Hz_40ms_envelope_correlation": 0.9931403294642167, "3000–6000 Hz_40ms_modulation_mae_db": 0.7346577660525403, "mr_stft_db": 4.378935302149422, "band_mae_db": 0.1111227998205912, "rms_gain_error_db": -0.0901539206223804, "output_peak": 0.43576396138114204}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6615659608938097, "全帯域_10ms_rms_envelope_nrmse": 0.09492545229638634, "全帯域_10ms_envelope_correlation": 0.9917895721346427, "全帯域_10ms_modulation_mae_db": 0.7157151164111933, "全帯域_40ms_log_rms_mae_db": 0.3709021071104603, "全帯域_40ms_rms_envelope_nrmse": 0.05960069984893802, "全帯域_40ms_envelope_correlation": 0.996733470452404, "全帯域_40ms_modulation_mae_db": 0.9034137413573973, "300–1000 Hz_10ms_log_rms_mae_db": 1.1761711870221097, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10286511976912828, "300–1000 Hz_10ms_envelope_correlation": 0.9924661316252773, "300–1000 Hz_10ms_modulation_mae_db": 0.38376541866391567, "300–1000 Hz_40ms_log_rms_mae_db": 0.7179262151438635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0638051164502994, "300–1000 Hz_40ms_envelope_correlation": 0.9969979318655631, "300–1000 Hz_40ms_modulation_mae_db": 0.6053603381096789, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2150975865885398, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16460904100067053, "1000–3000 Hz_10ms_envelope_correlation": 0.9799074419446498, "1000–3000 Hz_10ms_modulation_mae_db": 0.7330550712457601, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7184627835560283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0808579297682249, "1000–3000 Hz_40ms_envelope_correlation": 0.9953754136718016, "1000–3000 Hz_40ms_modulation_mae_db": 0.9281177195934337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2080365518073755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19391167821856153, "3000–6000 Hz_10ms_envelope_correlation": 0.9794980360901135, "3000–6000 Hz_10ms_modulation_mae_db": 1.4031500163641109, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9337686370985357, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14956882058459825, "3000–6000 Hz_40ms_envelope_correlation": 0.9888125024228527, "3000–6000 Hz_40ms_modulation_mae_db": 1.7509416270208298, "mr_stft_db": 4.370659619006942, "band_mae_db": 0.2121686667314255, "rms_gain_error_db": -0.1787950715779046, "output_peak": 0.43117648006649695}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6842504972880776, "全帯域_10ms_rms_envelope_nrmse": 0.09758615229971633, "全帯域_10ms_envelope_correlation": 0.9910354615919987, "全帯域_10ms_modulation_mae_db": 0.2342536955023443, "全帯域_40ms_log_rms_mae_db": 0.38788515386746325, "全帯域_40ms_rms_envelope_nrmse": 0.049462970507203406, "全帯域_40ms_envelope_correlation": 0.9975231331600044, "全帯域_40ms_modulation_mae_db": 0.1287211377484049, "300–1000 Hz_10ms_log_rms_mae_db": 1.2324175071194983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1051016002891762, "300–1000 Hz_10ms_envelope_correlation": 0.9920898060583323, "300–1000 Hz_10ms_modulation_mae_db": 0.2545743635895177, "300–1000 Hz_40ms_log_rms_mae_db": 0.7545284034403538, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058112858371983295, "300–1000 Hz_40ms_envelope_correlation": 0.9974753470653206, "300–1000 Hz_40ms_modulation_mae_db": 0.13127195219190627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2350993104215684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18268648732608028, "1000–3000 Hz_10ms_envelope_correlation": 0.9751184503859088, "1000–3000 Hz_10ms_modulation_mae_db": 0.674591858558339, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7428731294263078, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10836689765276472, "1000–3000 Hz_40ms_envelope_correlation": 0.9910934171443458, "1000–3000 Hz_40ms_modulation_mae_db": 0.5776758060815333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2286088252104734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18467165329928306, "3000–6000 Hz_10ms_envelope_correlation": 0.979472332835317, "3000–6000 Hz_10ms_modulation_mae_db": 1.1131841721979583, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9421691586753789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1054824614873042, "3000–6000 Hz_40ms_envelope_correlation": 0.9928903557507754, "3000–6000 Hz_40ms_modulation_mae_db": 1.1724418911360737, "mr_stft_db": 4.385068565202551, "band_mae_db": 0.121585005427499, "rms_gain_error_db": -0.057842181578467676, "output_peak": 0.4236944690930464}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.3897329837223875, "spectral_convergence": 0.37333787016303555, "spectral_convergence_status": "ok", "amplitude_floor": 1.0971770400792783e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.331555091050245, "spectral_convergence": 0.3668115590947206, "spectral_convergence_status": "ok", "amplitude_floor": 6.752711905169626e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.387113727209031, "spectral_convergence": 0.3878111026012104, "spectral_convergence_status": "ok", "amplitude_floor": 5.858543156261471e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003996059347809583, "output_power": 0.0003942084840089896, "delta_db": -0.05905962793664214, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00030506165058519967, "output_power": 0.00029675613271147377, "delta_db": -0.11987912976262344, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002850207180759584, "output_power": 0.00027823720428908497, "delta_db": -0.10461228984477625, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017376514501848305, "output_power": 0.0001635303819667407, "delta_db": -0.26364215986590506, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002541837569284722, "output_power": 0.00028692571135896245, "delta_db": 0.5262167259556528, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023122095798090296, "output_power": 0.00022489459254456637, "delta_db": -0.12048182986072561, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19898196053772088, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.011355546618473189, "pitch": null, "output_rms_dbfs": -27.839522559094313, "output_peak": 0.41766819916771203, "reference_rms_dbfs": -27.828167012475838, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.369467267327221}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015167567220720444, "log_rms_mae_db": 0.6538809732814929, "rms_envelope_nrmse": 0.09260734046125005, "envelope_correlation": 0.9919444377508451, "rms_gain_error_db": -0.011355528952850724, "reference_mean_power": 0.0016502333338478435, "output_mean_power": 0.0016459240930757754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103350035549285e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021122178949431673, "output_power": 0.00021336008250725263, "reference_active": true, "delta_db": 0.04374452932982172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018786794884032712, "output_power": 0.00019371203144766074, "reference_active": true, "delta_db": 0.13303901842660731}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.617925051494854e-05, "output_power": 2.303407780215263e-05, "reference_active": true, "delta_db": -0.5558637956877501}], "active_band_count": 3, "active_band_mae_db": 0.24421578114805972}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014379702931337134, "log_rms_mae_db": 0.38356299569138624, "rms_envelope_nrmse": 0.05587240842763272, "envelope_correlation": 0.9968155886703266, "rms_gain_error_db": -0.011355320501198458, "reference_mean_power": 0.0016585326982667058, "output_mean_power": 0.0016542018648303823}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0955155560805602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019170493341265238, "output_power": 0.0001971045711537535, "reference_active": true, "delta_db": 0.12063407010987678}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.87936190430084e-05, "output_power": 8.174889176505492e-05, "reference_active": true, "delta_db": 0.15990825456682395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1061472825588806e-06, "output_power": 1.7587290347795906e-06, "reference_active": true, "delta_db": -0.7828980466397895}], "active_band_count": 3, "active_band_mae_db": 0.35448012377216337}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010769717002798511, "log_rms_mae_db": 1.2570932952043432, "rms_envelope_nrmse": 0.10657300524024065, "envelope_correlation": 0.9919120663358987, "rms_gain_error_db": -0.1269883691894239, "reference_mean_power": 0.0005226470255969395, "output_mean_power": 0.0005075860130365006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.88003979644585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021484485335499559, "output_power": 0.00021057772249490582, "reference_active": true, "delta_db": -0.087125302860786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0907744988019035e-05, "output_power": 3.2073874976792825e-05, "reference_active": true, "delta_db": -1.0565410796512023}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.643789476493196e-06, "output_power": 7.333361985299717e-06, "reference_active": true, "delta_db": -0.7140105797096112}], "active_band_count": 3, "active_band_mae_db": 0.6192256540738664}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.974444405364767e-05, "log_rms_mae_db": 0.759008416512528, "rms_envelope_nrmse": 0.0714083023977861, "envelope_correlation": 0.996234294256201, "rms_gain_error_db": -0.1269883394821647, "reference_mean_power": 0.0005252755682472565, "output_mean_power": 0.0005101388129997993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.5461080823927677e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019511319942898067, "output_power": 0.00019104698284098026, "reference_active": true, "delta_db": -0.0914646708651934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8168801206395614e-05, "output_power": 1.4515601696632525e-05, "reference_active": true, "delta_db": -0.974912303252209}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.358572422696516e-07, "output_power": 5.3462102636091e-07, "reference_active": true, "delta_db": -1.3874753321081097}], "active_band_count": 3, "active_band_mae_db": 0.8179507687418374}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.224188331064728e-05, "log_rms_mae_db": 1.200167329284388, "rms_envelope_nrmse": 0.1582283490782197, "envelope_correlation": 0.981231301083322, "rms_gain_error_db": -0.13501555494330336, "reference_mean_power": 0.00028515402410464236, "output_mean_power": 0.00027642540161788317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.768616554938329e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001096694429350374, "output_power": 0.00010781680753200281, "reference_active": true, "delta_db": -0.07399169306775308}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.898846044442003e-05, "output_power": 4.6063404234297846e-05, "reference_active": true, "delta_db": -0.26737760561683804}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.522948535574358e-06, "output_power": 5.0026423678677855e-06, "reference_active": true, "delta_db": -2.313904091661934}], "active_band_count": 3, "active_band_mae_db": 0.8850911301155083}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.641451530936745e-05, "log_rms_mae_db": 0.709709618911829, "rms_envelope_nrmse": 0.08156267330038608, "envelope_correlation": 0.9947471672781322, "rms_gain_error_db": -0.13501569088719137, "reference_mean_power": 0.0002865881294452572, "output_mean_power": 0.00027781560000703754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.5110883907146213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010245851234810692, "output_power": 0.00010036526425114219, "reference_active": true, "delta_db": -0.08964613435832869}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5140429681693116e-05, "output_power": 2.359033696489243e-05, "reference_active": true, "delta_db": -0.27638551652697696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.446669785409335e-07, "output_power": 4.394821418975614e-07, "reference_active": true, "delta_db": -2.2902086348355946}], "active_band_count": 3, "active_band_mae_db": 0.8854134285736334}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001420862345348913, "log_rms_mae_db": 1.2083666313331058, "rms_envelope_nrmse": 0.1903443369906782, "envelope_correlation": 0.9807610883678105, "rms_gain_error_db": 0.42801528052131155, "reference_mean_power": 0.00023492613523189528, "output_mean_power": 0.0002592584153693551}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4459776587443134e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000158142036380199, "output_power": 0.00018955634726856727, "reference_active": true, "delta_db": 0.7869100443084999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.639730090455687e-05, "output_power": 0.00011506916784635599, "reference_active": true, "delta_db": 0.7689409842965972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.267646990584481e-05, "output_power": 1.379227960424981e-05, "reference_active": true, "delta_db": 0.3663772307314722}], "active_band_count": 3, "active_band_mae_db": 0.6407427531121898}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010454414213188805, "log_rms_mae_db": 0.9059114675751783, "rms_envelope_nrmse": 0.12716616313367884, "envelope_correlation": 0.9918346431654655, "rms_gain_error_db": 0.4280152797531118, "reference_mean_power": 0.00023610764979113905, "output_mean_power": 0.00026056230427569323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2342047370765003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017362277943143474, "output_power": 0.00021133985622982682, "reference_active": true, "delta_db": 0.8537470332401532}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.8688155273395004e-05, "output_power": 5.820654326949556e-05, "reference_active": true, "delta_db": 0.775484884234588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.341220903531162e-06, "output_power": 1.1989529534299755e-06, "reference_active": true, "delta_db": -0.48698171642125376}], "active_band_count": 3, "active_band_mae_db": 0.7054045446319983}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9037590390984367, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ce34d8f16c682eb811d2b14d076319a093cc5d2f30b0adf91bf04267dc4327f2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9801469417572476, "standard_deviation": 0.0028835401125009057, "minimum": 0.9751184503859088, "maximum": 0.9836986279876181, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.20193937040171, "standard_deviation": 0.01811398749472789, "minimum": 1.1754187838961077, "maximum": 1.2350993104215684, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7492812668129533, "standard_deviation": 0.2245936843273943, "minimum": 0.24837204781723923, "maximum": 0.9462483322677407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1628033920668608, "standard_deviation": 0.011528887869416776, "minimum": 0.14821008522031107, "maximum": 0.18268648732608028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937930822844088, "standard_deviation": 0.002551723019850824, "minimum": 0.9887296042889229, "maximum": 0.9959031353534415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7266995101833722, "standard_deviation": 0.027947306426785813, "minimum": 0.6939472630298493, "maximum": 0.7840327936461655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8471940988136142, "standard_deviation": 0.22966341061106563, "minimum": 0.43492610370544976, "maximum": 1.1215920994818387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0887652391240307, "standard_deviation": 0.015880593987073918, "minimum": 0.07472869139455238, "maximum": 0.118577636243806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9779172492425023, "standard_deviation": 0.003217266109988998, "minimum": 0.9726909212337551, "maximum": 0.9817707472972659, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2102820854222966, "standard_deviation": 0.02569506570287388, "minimum": 1.1565530018422714, "maximum": 1.2374293267204861, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0218523314740091, "standard_deviation": 0.5158270935558761, "minimum": 0.6006639332114775, "maximum": 2.1006744076054855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19783523365499683, "standard_deviation": 0.019217584482916224, "minimum": 0.17748209809416446, "maximum": 0.23879921328871903, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991375657758474, "standard_deviation": 0.002773116514862427, "minimum": 0.9859474779567196, "maximum": 0.9948174579101957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9196061719270606, "standard_deviation": 0.03801765078816487, "minimum": 0.8378399132825268, "maximum": 0.9560504338120527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1229502457191585, "standard_deviation": 0.6096470175208417, "minimum": 0.4713356708773912, "maximum": 2.282905546342926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12429589646385228, "standard_deviation": 0.027617846608456, "minimum": 0.09405419022271215, "maximum": 0.1781294521810949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992017567920813, "standard_deviation": 0.0008939007425292653, "minimum": 0.9907476836799005, "maximum": 0.9933063991823158, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2222391043106855, "standard_deviation": 0.0235308263071918, "minimum": 1.1761711870221097, "maximum": 1.2570932952043432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.46125309783191126, "standard_deviation": 0.11678659374938245, "minimum": 0.2545743635895177, "maximum": 0.6192256540738664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10561628550812188, "standard_deviation": 0.00591699956433479, "minimum": 0.09710444065208375, "maximum": 0.11396681064594787, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968795640317015, "standard_deviation": 0.0005787693649709169, "minimum": 0.996234294256201, "maximum": 0.9978800008493516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7495945747098155, "standard_deviation": 0.020535134088867243, "minimum": 0.7179262151438635, "maximum": 0.7800663745491007, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5395898033393373, "standard_deviation": 0.23904225242224297, "minimum": 0.13127195219190627, "maximum": 0.8179507687418374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06449915632240433, "standard_deviation": 0.006324309724884334, "minimum": 0.05329790816798684, "maximum": 0.0714083023977861, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1525171571107176, "standard_deviation": 0.04433213823726597, "minimum": 0.10189149067952019, "maximum": 0.2121686667314255, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.372700684942904, "standard_deviation": 0.009398756388519312, "minimum": 4.354615576282964, "maximum": 4.385068565202551, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.43380763200638583, "standard_deviation": 0.03489569673418315, "minimum": 0.3921347068519494, "maximum": 0.5054142153022005, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06011252226307867, "standard_deviation": 0.06562071497079022, "minimum": -0.1787950715779046, "maximum": 0.012693983286103572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9915406700706209, "standard_deviation": 0.0010711881510195146, "minimum": 0.9897715622500997, "maximum": 0.9935394269113276, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6601150985342664, "standard_deviation": 0.014016516943439071, "minimum": 0.6388021890613905, "maximum": 0.6842504972880776, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5346179649371747, "standard_deviation": 0.27836442398238526, "minimum": 0.2342536955023443, "maximum": 1.0483651057416479, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09501024120203172, "standard_deviation": 0.006174714151250081, "minimum": 0.08288760383472246, "maximum": 0.1048033597431499, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970712499889245, "standard_deviation": 0.0008606756039083632, "minimum": 0.9952672996427362, "maximum": 0.9980975628419595, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37381884065642657, "standard_deviation": 0.01043101539811042, "minimum": 0.3615859948135283, "maximum": 0.38788515386746325, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6108616053720395, "standard_deviation": 0.3671776750809942, "minimum": 0.1287211377484049, "maximum": 1.3255370715857504, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05383188861246546, "standard_deviation": 0.007786190986242516, "minimum": 0.043126311482948564, "maximum": 0.06844906205743391, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-127105-0035", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.356356620788574, "predicted_mos": 2.356356620788574, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "152b96cdf2ced7ef34952bb35deba0040a1c3b8d881f384e96b2afcdcc575295", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3108890000730753, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6317274454898891, "全帯域_10ms_rms_envelope_nrmse": 0.08416622635009208, "全帯域_10ms_envelope_correlation": 0.9935862031561922, "全帯域_10ms_modulation_mae_db": 0.717425993344842, "全帯域_40ms_log_rms_mae_db": 0.34870549450973476, "全帯域_40ms_rms_envelope_nrmse": 0.04579078623682213, "全帯域_40ms_envelope_correlation": 0.9982000018708606, "全帯域_40ms_modulation_mae_db": 0.8064482030098906, "300–1000 Hz_10ms_log_rms_mae_db": 1.721422339356513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11475150257310036, "300–1000 Hz_10ms_envelope_correlation": 0.9906390546595316, "300–1000 Hz_10ms_modulation_mae_db": 0.1660190425724657, "300–1000 Hz_40ms_log_rms_mae_db": 1.2424965816313298, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06945617059434565, "300–1000 Hz_40ms_envelope_correlation": 0.9964370630962052, "300–1000 Hz_40ms_modulation_mae_db": 0.3597711301415445, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3406621025148946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16189496141305196, "1000–3000 Hz_10ms_envelope_correlation": 0.9807304529597608, "1000–3000 Hz_10ms_modulation_mae_db": 1.0076135312379488, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8359193098091117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09243972579933847, "1000–3000 Hz_40ms_envelope_correlation": 0.9938920967843696, "1000–3000 Hz_40ms_modulation_mae_db": 1.1446310085074327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4353335221023926, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18795838139366652, "3000–6000 Hz_10ms_envelope_correlation": 0.9795916373704867, "3000–6000 Hz_10ms_modulation_mae_db": 1.3301686904543646, "3000–6000 Hz_40ms_log_rms_mae_db": 1.051838565510741, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11668955801763606, "3000–6000 Hz_40ms_envelope_correlation": 0.9930183398462186, "3000–6000 Hz_40ms_modulation_mae_db": 1.7524230622408823, "mr_stft_db": 4.6517152875395, "band_mae_db": 0.20027473226010642, "rms_gain_error_db": -0.16888519615743738, "output_peak": 0.4176669478163845}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6433848021301664, "全帯域_10ms_rms_envelope_nrmse": 0.08505721048492682, "全帯域_10ms_envelope_correlation": 0.9932000725998167, "全帯域_10ms_modulation_mae_db": 0.42130126446956034, "全帯域_40ms_log_rms_mae_db": 0.365064454642669, "全帯域_40ms_rms_envelope_nrmse": 0.045885179783356586, "全帯域_40ms_envelope_correlation": 0.9978693485458814, "全帯域_40ms_modulation_mae_db": 0.2696291662312886, "300–1000 Hz_10ms_log_rms_mae_db": 1.6658485059193011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11431473125667134, "300–1000 Hz_10ms_envelope_correlation": 0.990715770281428, "300–1000 Hz_10ms_modulation_mae_db": 0.3212143067142382, "300–1000 Hz_40ms_log_rms_mae_db": 1.2284076444038814, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06840345759276253, "300–1000 Hz_40ms_envelope_correlation": 0.9965535900113175, "300–1000 Hz_40ms_modulation_mae_db": 0.4348028424852046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2854536836978248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15512359711664175, "1000–3000 Hz_10ms_envelope_correlation": 0.9820406076191788, "1000–3000 Hz_10ms_modulation_mae_db": 0.8985733348947296, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7882419824260326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08689482267685342, "1000–3000 Hz_40ms_envelope_correlation": 0.9941527502166899, "1000–3000 Hz_40ms_modulation_mae_db": 1.1570117817563845, "3000–6000 Hz_10ms_log_rms_mae_db": 1.510369965999441, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.182304441793217, "3000–6000 Hz_10ms_envelope_correlation": 0.9798234914984321, "3000–6000 Hz_10ms_modulation_mae_db": 0.44448587048897653, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0841719707311706, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09715425211629555, "3000–6000 Hz_40ms_envelope_correlation": 0.9939714087512773, "3000–6000 Hz_40ms_modulation_mae_db": 0.4065814383826656, "mr_stft_db": 4.665833885835334, "band_mae_db": 0.0925530738580904, "rms_gain_error_db": -0.052701932781623086, "output_peak": 0.4257166746795837}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6366203426043431, "全帯域_10ms_rms_envelope_nrmse": 0.08366488604708819, "全帯域_10ms_envelope_correlation": 0.9934906665850597, "全帯域_10ms_modulation_mae_db": 0.20806529301169877, "全帯域_40ms_log_rms_mae_db": 0.3624007654661687, "全帯域_40ms_rms_envelope_nrmse": 0.044788739063194936, "全帯域_40ms_envelope_correlation": 0.9980775901969776, "全帯域_40ms_modulation_mae_db": 0.032731331425501935, "300–1000 Hz_10ms_log_rms_mae_db": 1.7402193558353383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10638277248929026, "300–1000 Hz_10ms_envelope_correlation": 0.9922282248994643, "300–1000 Hz_10ms_modulation_mae_db": 0.8706791323388, "300–1000 Hz_40ms_log_rms_mae_db": 1.2389011708522708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06615286329961971, "300–1000 Hz_40ms_envelope_correlation": 0.9970569931669557, "300–1000 Hz_40ms_modulation_mae_db": 1.0174306521516157, "1000–3000 Hz_10ms_log_rms_mae_db": 1.299503378306781, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18358376975951587, "1000–3000 Hz_10ms_envelope_correlation": 0.9747684479673783, "1000–3000 Hz_10ms_modulation_mae_db": 0.3731882165421698, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8322222985376656, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10972394482694972, "1000–3000 Hz_40ms_envelope_correlation": 0.9902243135187776, "1000–3000 Hz_40ms_modulation_mae_db": 0.41013414861469827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.527754492590976, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17286504427751379, "3000–6000 Hz_10ms_envelope_correlation": 0.9818730734275024, "3000–6000 Hz_10ms_modulation_mae_db": 0.6126825596921641, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0880060581087925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09916157654844572, "3000–6000 Hz_40ms_envelope_correlation": 0.9938410289069748, "3000–6000 Hz_40ms_modulation_mae_db": 0.5741235011335108, "mr_stft_db": 4.665285835052351, "band_mae_db": 0.1320420025486486, "rms_gain_error_db": -0.10779190313103312, "output_peak": 0.4284686696285146}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6674039096473607, "全帯域_10ms_rms_envelope_nrmse": 0.09292319582641086, "全帯域_10ms_envelope_correlation": 0.9919325130027925, "全帯域_10ms_modulation_mae_db": 0.7742939693101866, "全帯域_40ms_log_rms_mae_db": 0.35376153189988363, "全帯域_40ms_rms_envelope_nrmse": 0.04255568316116473, "全帯域_40ms_envelope_correlation": 0.9982653253423686, "全帯域_40ms_modulation_mae_db": 0.8807045264733468, "300–1000 Hz_10ms_log_rms_mae_db": 1.7531645916678027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12000346782293181, "300–1000 Hz_10ms_envelope_correlation": 0.9898955182829506, "300–1000 Hz_10ms_modulation_mae_db": 0.8284961291382302, "300–1000 Hz_40ms_log_rms_mae_db": 1.246321038077962, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07521147369892217, "300–1000 Hz_40ms_envelope_correlation": 0.9959716525272678, "300–1000 Hz_40ms_modulation_mae_db": 1.0715784415711784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3506536894923995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16099925639302828, "1000–3000 Hz_10ms_envelope_correlation": 0.9806844382609202, "1000–3000 Hz_10ms_modulation_mae_db": 0.7911158167787194, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8457876602442661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08506403367844463, "1000–3000 Hz_40ms_envelope_correlation": 0.9945282072931246, "1000–3000 Hz_40ms_modulation_mae_db": 0.8306137122255016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4945531271348427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19951736789625932, "3000–6000 Hz_10ms_envelope_correlation": 0.9758006835477774, "3000–6000 Hz_10ms_modulation_mae_db": 0.767179010041084, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0627944120997945, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09997629018798504, "3000–6000 Hz_40ms_envelope_correlation": 0.994071333722876, "3000–6000 Hz_40ms_modulation_mae_db": 1.0878489194519056, "mr_stft_db": 4.665021522647193, "band_mae_db": 0.14387984483867408, "rms_gain_error_db": -0.10117708466906175, "output_peak": 0.38977023125104215}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6521210880380951, "全帯域_10ms_rms_envelope_nrmse": 0.08122046205081689, "全帯域_10ms_envelope_correlation": 0.993936808590288, "全帯域_10ms_modulation_mae_db": 0.4880943279620889, "全帯域_40ms_log_rms_mae_db": 0.3590852907329775, "全帯域_40ms_rms_envelope_nrmse": 0.042329354401434156, "全帯域_40ms_envelope_correlation": 0.9983740643914697, "全帯域_40ms_modulation_mae_db": 0.6625510387375587, "300–1000 Hz_10ms_log_rms_mae_db": 1.7677073550213696, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11944410873016659, "300–1000 Hz_10ms_envelope_correlation": 0.9898528328048677, "300–1000 Hz_10ms_modulation_mae_db": 0.3149465758993876, "300–1000 Hz_40ms_log_rms_mae_db": 1.2818394112557325, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07159812605752362, "300–1000 Hz_40ms_envelope_correlation": 0.9961841125127561, "300–1000 Hz_40ms_modulation_mae_db": 0.40680477916305446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3613609836468705, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16172358936012438, "1000–3000 Hz_10ms_envelope_correlation": 0.9807418838374922, "1000–3000 Hz_10ms_modulation_mae_db": 0.9624336848689014, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8570980233765139, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09355131488516953, "1000–3000 Hz_40ms_envelope_correlation": 0.9936416631011161, "1000–3000 Hz_40ms_modulation_mae_db": 1.0445607736373355, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5135165894192841, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17675354118492356, "3000–6000 Hz_10ms_envelope_correlation": 0.9815013448460826, "3000–6000 Hz_10ms_modulation_mae_db": 0.6597247880301297, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1046157823348788, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10772450304767142, "3000–6000 Hz_40ms_envelope_correlation": 0.9934349200731026, "3000–6000 Hz_40ms_modulation_mae_db": 0.8646848937034882, "mr_stft_db": 4.673658425636714, "band_mae_db": 0.16598549238491464, "rms_gain_error_db": -0.13209956826216449, "output_peak": 0.4081109553411775}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6353516881938734, "全帯域_10ms_rms_envelope_nrmse": 0.0841738104707625, "全帯域_10ms_envelope_correlation": 0.993485896016782, "全帯域_10ms_modulation_mae_db": 0.6335467511206694, "全帯域_40ms_log_rms_mae_db": 0.34770805545988537, "全帯域_40ms_rms_envelope_nrmse": 0.04405034817264354, "全帯域_40ms_envelope_correlation": 0.9982449441786964, "全帯域_40ms_modulation_mae_db": 0.507879648360326, "300–1000 Hz_10ms_log_rms_mae_db": 1.7494076322855934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1146512881925946, "300–1000 Hz_10ms_envelope_correlation": 0.9909581173104619, "300–1000 Hz_10ms_modulation_mae_db": 0.648056871735982, "300–1000 Hz_40ms_log_rms_mae_db": 1.2830501409368, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08040042244330406, "300–1000 Hz_40ms_envelope_correlation": 0.9955449610220396, "300–1000 Hz_40ms_modulation_mae_db": 0.8987001176137492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.360111820671402, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1738408298566121, "1000–3000 Hz_10ms_envelope_correlation": 0.9774026080681603, "1000–3000 Hz_10ms_modulation_mae_db": 0.9082599197442113, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8571582324254747, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10138064325032518, "1000–3000 Hz_40ms_envelope_correlation": 0.9920500781865671, "1000–3000 Hz_40ms_modulation_mae_db": 1.1198843558552831, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4724374650712855, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18444025833358027, "3000–6000 Hz_10ms_envelope_correlation": 0.9793746117075717, "3000–6000 Hz_10ms_modulation_mae_db": 0.671464643665237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0622339540980694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0988388114933692, "3000–6000 Hz_40ms_envelope_correlation": 0.9940087789673229, "3000–6000 Hz_40ms_modulation_mae_db": 0.9362910139529171, "mr_stft_db": 4.665240837875415, "band_mae_db": 0.152681479892368, "rms_gain_error_db": -0.13956263544854114, "output_peak": 0.40771241987329004}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6552453953420445, "全帯域_10ms_rms_envelope_nrmse": 0.08691434483882765, "全帯域_10ms_envelope_correlation": 0.9930342749581317, "全帯域_10ms_modulation_mae_db": 0.2831498795063087, "全帯域_40ms_log_rms_mae_db": 0.37259170717234064, "全帯域_40ms_rms_envelope_nrmse": 0.04730003870890845, "全帯域_40ms_envelope_correlation": 0.9979298853967452, "全帯域_40ms_modulation_mae_db": 0.3259594171996163, "300–1000 Hz_10ms_log_rms_mae_db": 1.7320010334649363, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11687208366469762, "300–1000 Hz_10ms_envelope_correlation": 0.9905305718003758, "300–1000 Hz_10ms_modulation_mae_db": 0.7503413313516539, "300–1000 Hz_40ms_log_rms_mae_db": 1.2866308899976344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08372313075951521, "300–1000 Hz_40ms_envelope_correlation": 0.9950307802400545, "300–1000 Hz_40ms_modulation_mae_db": 0.9504714348711737, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3051855465472184, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16042725512178407, "1000–3000 Hz_10ms_envelope_correlation": 0.9808631703743027, "1000–3000 Hz_10ms_modulation_mae_db": 0.4618680964799178, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8380346830450696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0921259958385216, "1000–3000 Hz_40ms_envelope_correlation": 0.9935644451124588, "1000–3000 Hz_40ms_modulation_mae_db": 0.49999697199163146, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5163734399982542, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16891708231836375, "3000–6000 Hz_10ms_envelope_correlation": 0.9826781744024398, "3000–6000 Hz_10ms_modulation_mae_db": 0.1482465574385328, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1169312735758803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09097474632922833, "3000–6000 Hz_40ms_envelope_correlation": 0.9946728966312283, "3000–6000 Hz_40ms_modulation_mae_db": 0.1339301509849255, "mr_stft_db": 4.674640201464867, "band_mae_db": 0.1562793013659171, "rms_gain_error_db": -0.13775027195404482, "output_peak": 0.45021139101090096}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6250312431151102, "全帯域_10ms_rms_envelope_nrmse": 0.0866651333794139, "全帯域_10ms_envelope_correlation": 0.9929399587742326, "全帯域_10ms_modulation_mae_db": 0.5303453660189396, "全帯域_40ms_log_rms_mae_db": 0.34497184166567285, "全帯域_40ms_rms_envelope_nrmse": 0.049180944326208606, "全帯域_40ms_envelope_correlation": 0.997550083065389, "全帯域_40ms_modulation_mae_db": 0.41964306973713666, "300–1000 Hz_10ms_log_rms_mae_db": 1.7341621333519077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11179215176790219, "300–1000 Hz_10ms_envelope_correlation": 0.9913000705484537, "300–1000 Hz_10ms_modulation_mae_db": 0.7342992769868539, "300–1000 Hz_40ms_log_rms_mae_db": 1.2627930763718307, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07093289201130624, "300–1000 Hz_40ms_envelope_correlation": 0.996472139631289, "300–1000 Hz_40ms_modulation_mae_db": 0.9430350251626289, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3494588889695174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16076023857153743, "1000–3000 Hz_10ms_envelope_correlation": 0.9806371334658338, "1000–3000 Hz_10ms_modulation_mae_db": 0.7423833440662283, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8401944852860765, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09421863404229083, "1000–3000 Hz_40ms_envelope_correlation": 0.9929630751677558, "1000–3000 Hz_40ms_modulation_mae_db": 0.7982911563904086, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4577737557613089, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19101716221908308, "3000–6000 Hz_10ms_envelope_correlation": 0.9777566431298101, "3000–6000 Hz_10ms_modulation_mae_db": 0.38173990798701, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1020255056182255, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11552691992574642, "3000–6000 Hz_40ms_envelope_correlation": 0.9913935355167064, "3000–6000 Hz_40ms_modulation_mae_db": 0.8344746133705486, "mr_stft_db": 4.670160012402654, "band_mae_db": 0.10764638704767633, "rms_gain_error_db": -0.05548357053164994, "output_peak": 0.4288414802193522}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.672358061312308, "spectral_convergence": 0.3756199956950558, "spectral_convergence_status": "ok", "amplitude_floor": 1.0971770400792783e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.702530538586602, "spectral_convergence": 0.3772942177704748, "spectral_convergence_status": "ok", "amplitude_floor": 6.752711905169626e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.58025726271959, "spectral_convergence": 0.37826541242511386, "spectral_convergence_status": "ok", "amplitude_floor": 5.858543156261471e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003996059347809583, "output_power": 0.00039697538453091014, "delta_db": -0.028683514993663905, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00030506165058519967, "output_power": 0.000298901837344364, "delta_db": -0.08859031103034268, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002850207180759584, "output_power": 0.00028293297043713403, "delta_db": -0.031928707617631974, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017376514501848305, "output_power": 0.00016131982003754714, "delta_db": -0.32274938287285315, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002541837569284722, "output_power": 0.00022615265752811892, "delta_db": -0.5074609891874704, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023122095798090296, "output_power": 0.00021968664041797833, "delta_db": -0.22223548785867647, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20027473226010642, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16888519615743738, "pitch": null, "output_rms_dbfs": -27.997052208633278, "output_peak": 0.4176669478163845, "reference_rms_dbfs": -27.828167012475838, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.6517152875395}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015167567220720444, "log_rms_mae_db": 0.6317274454898891, "rms_envelope_nrmse": 0.08416622635009208, "envelope_correlation": 0.9935862031561922, "rms_gain_error_db": -0.16888518172097383, "reference_mean_power": 0.0016502333338478435, "output_mean_power": 0.0015872920358783523}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103350035549285e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021122178949431673, "output_power": 0.0002075261499040943, "reference_active": true, "delta_db": -0.07665888723799813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018786794884032712, "output_power": 0.0001601535167013531, "reference_active": true, "delta_db": -0.6931621425006922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.617925051494854e-05, "output_power": 1.9041954099495494e-05, "reference_active": true, "delta_db": -1.382456950295836}], "active_band_count": 3, "active_band_mae_db": 0.717425993344842}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014379702931337134, "log_rms_mae_db": 0.34870549450973476, "rms_envelope_nrmse": 0.04579078623682213, "envelope_correlation": 0.9982000018708606, "rms_gain_error_db": -0.16888514757424622, "reference_mean_power": 0.0016585326982667058, "output_mean_power": 0.0015952748680473153}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0955155560805602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019170493341265238, "output_power": 0.0001869108285583331, "reference_active": true, "delta_db": -0.10998826654069717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.87936190430084e-05, "output_power": 6.942686827391984e-05, "reference_active": true, "delta_db": -0.54963472945404}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1061472825588806e-06, "output_power": 1.4044832908236318e-06, "reference_active": true, "delta_db": -1.7597216130349345}], "active_band_count": 3, "active_band_mae_db": 0.8064482030098906}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010769717002798511, "log_rms_mae_db": 1.721422339356513, "rms_envelope_nrmse": 0.11475150257310036, "envelope_correlation": 0.9906390546595316, "rms_gain_error_db": -0.08473450619080022, "reference_mean_power": 0.0005226470255969395, "output_mean_power": 0.0005125485774951739}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.88003979644585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021484485335499559, "output_power": 0.0002068135427136485, "reference_active": true, "delta_db": -0.1654598040861072}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0907744988019035e-05, "output_power": 3.885549390616098e-05, "reference_active": true, "delta_db": -0.22353106161751896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.643789476493196e-06, "output_power": 8.429417309513888e-06, "reference_active": true, "delta_db": -0.10906626201377091}], "active_band_count": 3, "active_band_mae_db": 0.1660190425724657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.974444405364767e-05, "log_rms_mae_db": 1.2424965816313298, "rms_envelope_nrmse": 0.06945617059434565, "envelope_correlation": 0.9964370630962052, "rms_gain_error_db": -0.08473459026725136, "reference_mean_power": 0.0005252755682472565, "output_mean_power": 0.0005151263221645748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.5461080823927677e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019511319942898067, "output_power": 0.00018922020961437228, "reference_active": true, "delta_db": -0.13319131206757662}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8168801206395614e-05, "output_power": 1.7295757201740023e-05, "reference_active": true, "delta_db": -0.2138669310302579}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.358572422696516e-07, "output_power": 6.216816574362499e-07, "reference_active": true, "delta_db": -0.732255147326799}], "active_band_count": 3, "active_band_mae_db": 0.3597711301415445}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.224188331064728e-05, "log_rms_mae_db": 1.3406621025148946, "rms_envelope_nrmse": 0.16189496141305196, "envelope_correlation": 0.9807304529597608, "rms_gain_error_db": -0.2956253707022965, "reference_mean_power": 0.00028515402410464236, "output_mean_power": 0.00026638941717867644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.768616554938329e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001096694429350374, "output_power": 0.0001023600210921775, "reference_active": true, "delta_db": -0.29955270775181797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.898846044442003e-05, "output_power": 4.527984208483297e-05, "reference_active": true, "delta_db": -0.34188887575989363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.522948535574358e-06, "output_power": 4.925495935344753e-06, "reference_active": true, "delta_db": -2.381399010202135}], "active_band_count": 3, "active_band_mae_db": 1.0076135312379488}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.641451530936745e-05, "log_rms_mae_db": 0.8359193098091117, "rms_envelope_nrmse": 0.09243972579933847, "envelope_correlation": 0.9938920967843696, "rms_gain_error_db": -0.2956255134788051, "reference_mean_power": 0.0002865881294452572, "output_mean_power": 0.00026772914218241674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.5110883907146213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010245851234810692, "output_power": 9.519649079182436e-05, "reference_active": true, "delta_db": -0.31927106450369425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5140429681693116e-05, "output_power": 2.3126610300290602e-05, "reference_active": true, "delta_db": -0.362607137707208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.446669785409335e-07, "output_power": 3.9514875106519195e-07, "reference_active": true, "delta_db": -2.752014823311396}], "active_band_count": 3, "active_band_mae_db": 1.1446310085074327}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001420862345348913, "log_rms_mae_db": 1.4353335221023926, "rms_envelope_nrmse": 0.18795838139366652, "envelope_correlation": 0.9795916373704867, "rms_gain_error_db": -0.4895653629122133, "reference_mean_power": 0.00023492613523189528, "output_mean_power": 0.0002098818085618869}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4459776587443134e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000158142036380199, "output_power": 0.0001382529616569986, "reference_active": true, "delta_db": -0.5837288332764874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.639730090455687e-05, "output_power": 7.434949280222031e-05, "reference_active": true, "delta_db": -1.127868637625577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.267646990584481e-05, "output_power": 7.500797997017546e-06, "reference_active": true, "delta_db": -2.2789086004610297}], "active_band_count": 3, "active_band_mae_db": 1.3301686904543646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010454414213188805, "log_rms_mae_db": 1.051838565510741, "rms_envelope_nrmse": 0.11668955801763606, "envelope_correlation": 0.9930183398462186, "rms_gain_error_db": -0.4895654115583811, "reference_mean_power": 0.00023610764979113905, "output_mean_power": 0.0002109373652679115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2342047370765003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017362277943143474, "output_power": 0.00014633990843818782, "reference_active": true, "delta_db": -0.7424392535330998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.8688155273395004e-05, "output_power": 3.7824606540877456e-05, "reference_active": true, "delta_db": -1.0964890094200712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.341220903531162e-06, "output_power": 6.104726631719568e-07, "reference_active": true, "delta_db": -3.4183409237694766}], "active_band_count": 3, "active_band_mae_db": 1.7524230622408823}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8810874454966935, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "104d71afc65deec0b2443c4435a1cfa2984ea657a6fbd603d0074952076ef8ea", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9797335928191284, "standard_deviation": 0.0024028343299716346, "minimum": 0.9747684479673783, "maximum": 0.9820406076191788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3315487617308635, "standard_deviation": 0.03004787310472039, "minimum": 1.2854536836978248, "maximum": 1.3613609836468705, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7681794930766033, "standard_deviation": 0.23382677533673885, "minimum": 0.3731882165421698, "maximum": 1.0076135312379488, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.164794187199037, "standard_deviation": 0.009227202695432103, "minimum": 0.15512359711664175, "maximum": 0.18358376975951587, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9931270786726074, "standard_deviation": 0.0013974147308763985, "minimum": 0.9902243135187776, "maximum": 0.9945282072931246, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8368320843937763, "standard_deviation": 0.021712064424667575, "minimum": 0.7882419824260326, "maximum": 0.8571582324254747, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8756404886223345, "standard_deviation": 0.29411595568556675, "minimum": 0.41013414861469827, "maximum": 1.1570117817563845, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09442488937473667, "standard_deviation": 0.007899390510954916, "minimum": 0.08506403367844463, "maximum": 0.10972394482694972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9797999574912628, "standard_deviation": 0.0022637155619118785, "minimum": 0.9758006835477774, "maximum": 0.9826781744024398, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4910140447597233, "standard_deviation": 0.03259453825844601, "minimum": 1.4353335221023926, "maximum": 1.527754492590976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6269615034746874, "standard_deviation": 0.3464139007830805, "minimum": 0.1482465574385328, "maximum": 1.3301686904543646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18297165992707592, "standard_deviation": 0.010030665438501778, "minimum": 0.16891708231836375, "maximum": 0.19951736789625932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935515303019633, "standard_deviation": 0.00099654210992992, "minimum": 0.9913935355167064, "maximum": 0.9946728966312283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0840771902596942, "standard_deviation": 0.023324493879121735, "minimum": 1.051838565510741, "maximum": 1.1169312735758803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8237946991526055, "standard_deviation": 0.4868018758493056, "minimum": 0.1339301509849255, "maximum": 1.7524230622408823, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10325583220829722, "standard_deviation": 0.009150303267542565, "minimum": 0.09097474632922833, "maximum": 0.11668955801763606, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907650200734417, "standard_deviation": 0.0007677343023957229, "minimum": 0.9898528328048677, "maximum": 0.9922282248994643, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.732991618362845, "standard_deviation": 0.030655351560073797, "minimum": 1.6658485059193011, "maximum": 1.7677073550213696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5792565833422014, "standard_deviation": 0.2705285890497084, "minimum": 0.1660190425724657, "maximum": 0.8706791323388, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11477651331216934, "standard_deviation": 0.004360369642081033, "minimum": 0.10638277248929026, "maximum": 0.12000346782293181, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961564115259858, "standard_deviation": 0.0006350554896481841, "minimum": 0.9950307802400545, "maximum": 0.9970569931669557, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.25880499419093, "standard_deviation": 0.022836597586169828, "minimum": 1.2284076444038814, "maximum": 1.2866308899976344, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7603243028950186, "standard_deviation": 0.3030919093632826, "minimum": 0.3597711301415445, "maximum": 1.0715784415711784, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0732348170571624, "standard_deviation": 0.006106316227393358, "minimum": 0.06615286329961971, "maximum": 0.08372313075951521, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14391778927454943, "standard_deviation": 0.03378013396981627, "minimum": 0.0925530738580904, "maximum": 0.20027473226010642, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.666444501056754, "standard_deviation": 0.007122940794060533, "minimum": 4.6517152875395, "maximum": 4.674640201464867, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41956234622753075, "standard_deviation": 0.0181477365244742, "minimum": 0.38977023125104215, "maximum": 0.45021139101090096, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11193152036694445, "standard_deviation": 0.04124468312105944, "minimum": -0.16888519615743738, "maximum": -0.052701932781623086, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9932007992104119, "standard_deviation": 0.0006047275401511121, "minimum": 0.9919325130027925, "maximum": 0.993936808590288, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6433607393201103, "standard_deviation": 0.01402520782184898, "minimum": 0.6250312431151102, "maximum": 0.6674039096473607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5070278555930368, "standard_deviation": 0.19980589796277778, "minimum": 0.20806529301169877, "maximum": 0.7742939693101866, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08559815868104237, "standard_deviation": 0.0034599864580422703, "minimum": 0.08122046205081689, "maximum": 0.09292319582641086, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980639053735485, "standard_deviation": 0.0002694121319561828, "minimum": 0.997550083065389, "maximum": 0.9983740643914697, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.35678614269366654, "standard_deviation": 0.009648026518629378, "minimum": 0.34497184166567285, "maximum": 0.37259170717234064, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.48819330014683315, "standard_deviation": 0.2859506130566428, "minimum": 0.032731331425501935, "maximum": 0.8807045264733468, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04523513423171664, "standard_deviation": 0.0023234997397865715, "minimum": 0.042329354401434156, "maximum": 0.049180944326208606, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-127105-0035", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.5325937271118164, "predicted_mos": 2.5325937271118164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "08252d07340ed8bb84e3dacad347b753921aa3f735fe40939a0378744710bd65", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1148272999562323, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6617259997352102, "全帯域_10ms_rms_envelope_nrmse": 0.08793877060274559, "全帯域_10ms_envelope_correlation": 0.9928171014219933, "全帯域_10ms_modulation_mae_db": 0.4667514438146287, "全帯域_40ms_log_rms_mae_db": 0.37689242496852804, "全帯域_40ms_rms_envelope_nrmse": 0.04588642166130236, "全帯域_40ms_envelope_correlation": 0.9980007904905764, "全帯域_40ms_modulation_mae_db": 0.4347817392669382, "300–1000 Hz_10ms_log_rms_mae_db": 1.209194211676607, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10306547024249349, "300–1000 Hz_10ms_envelope_correlation": 0.9924550797919316, "300–1000 Hz_10ms_modulation_mae_db": 0.30212050097121207, "300–1000 Hz_40ms_log_rms_mae_db": 0.7916727513091386, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0620269378826432, "300–1000 Hz_40ms_envelope_correlation": 0.997193926589986, "300–1000 Hz_40ms_modulation_mae_db": 0.5074590561829536, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0484617513886416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13773837726150656, "1000–3000 Hz_10ms_envelope_correlation": 0.9858311792988118, "1000–3000 Hz_10ms_modulation_mae_db": 0.34918180431547263, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6530062692695028, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08132024220426232, "1000–3000 Hz_40ms_envelope_correlation": 0.9947765136466745, "1000–3000 Hz_40ms_modulation_mae_db": 0.13427495692220695, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1047005957763283, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17780753717341363, "3000–6000 Hz_10ms_envelope_correlation": 0.9808887574321828, "3000–6000 Hz_10ms_modulation_mae_db": 0.807752591512493, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8415071508325133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10527044162569875, "3000–6000 Hz_40ms_envelope_correlation": 0.993260628255889, "3000–6000 Hz_40ms_modulation_mae_db": 0.9841488943882384, "mr_stft_db": 4.302109601411888, "band_mae_db": 0.1271086690224512, "rms_gain_error_db": -0.10942428786276827, "output_peak": 0.4214168494513621}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6751639470760931, "全帯域_10ms_rms_envelope_nrmse": 0.08710561671625694, "全帯域_10ms_envelope_correlation": 0.9928679654554531, "全帯域_10ms_modulation_mae_db": 0.09016532776847386, "全帯域_40ms_log_rms_mae_db": 0.36831192226538984, "全帯域_40ms_rms_envelope_nrmse": 0.047783754410188756, "全帯域_40ms_envelope_correlation": 0.99768468873916, "全帯域_40ms_modulation_mae_db": 0.25447026556200775, "300–1000 Hz_10ms_log_rms_mae_db": 1.1500743264880078, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10382224957820543, "300–1000 Hz_10ms_envelope_correlation": 0.9923006196809704, "300–1000 Hz_10ms_modulation_mae_db": 0.348570772763964, "300–1000 Hz_40ms_log_rms_mae_db": 0.7384891022242029, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05822158587923836, "300–1000 Hz_40ms_envelope_correlation": 0.9975039282925048, "300–1000 Hz_40ms_modulation_mae_db": 0.45583360745802026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0900575412983322, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1456382258749604, "1000–3000 Hz_10ms_envelope_correlation": 0.9842095173177334, "1000–3000 Hz_10ms_modulation_mae_db": 0.5251753640046111, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6669868184390686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08437017092451396, "1000–3000 Hz_40ms_envelope_correlation": 0.9945096620932491, "1000–3000 Hz_40ms_modulation_mae_db": 0.5259397011200478, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0922744837533296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17055424180197257, "3000–6000 Hz_10ms_envelope_correlation": 0.9834707760337016, "3000–6000 Hz_10ms_modulation_mae_db": 0.6422585666835425, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8353673159494678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10824114946828886, "3000–6000 Hz_40ms_envelope_correlation": 0.9932672189589649, "3000–6000 Hz_40ms_modulation_mae_db": 0.4715869555290066, "mr_stft_db": 4.300703179325928, "band_mae_db": 0.12395493702380339, "rms_gain_error_db": -0.03799746441378738, "output_peak": 0.44130984266654966}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6853218167513306, "全帯域_10ms_rms_envelope_nrmse": 0.09174353298394217, "全帯域_10ms_envelope_correlation": 0.9921159812330972, "全帯域_10ms_modulation_mae_db": 0.1762497138735485, "全帯域_40ms_log_rms_mae_db": 0.3984342511463212, "全帯域_40ms_rms_envelope_nrmse": 0.05056766950511494, "全帯域_40ms_envelope_correlation": 0.9974660284869759, "全帯域_40ms_modulation_mae_db": 0.27206789052530517, "300–1000 Hz_10ms_log_rms_mae_db": 1.2105542289259654, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10626341094464858, "300–1000 Hz_10ms_envelope_correlation": 0.9922322722797152, "300–1000 Hz_10ms_modulation_mae_db": 0.7021593225856658, "300–1000 Hz_40ms_log_rms_mae_db": 0.7827714884941745, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07232803881837628, "300–1000 Hz_40ms_envelope_correlation": 0.9964612310289507, "300–1000 Hz_40ms_modulation_mae_db": 0.6900006920924703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0782928265755893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15674395909009559, "1000–3000 Hz_10ms_envelope_correlation": 0.9815837434148803, "1000–3000 Hz_10ms_modulation_mae_db": 0.44016671768994425, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7158905827289979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10349505737052576, "1000–3000 Hz_40ms_envelope_correlation": 0.9914030106992596, "1000–3000 Hz_40ms_modulation_mae_db": 0.41777045682697395, "3000–6000 Hz_10ms_log_rms_mae_db": 1.113915784736873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1731376187770262, "3000–6000 Hz_10ms_envelope_correlation": 0.9820076132006742, "3000–6000 Hz_10ms_modulation_mae_db": 0.46553223457412196, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8473977302472782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09518455019082683, "3000–6000 Hz_40ms_envelope_correlation": 0.9942353305728924, "3000–6000 Hz_40ms_modulation_mae_db": 0.6518994848698473, "mr_stft_db": 4.311066294400329, "band_mae_db": 0.1486542829618382, "rms_gain_error_db": -0.0898967539802633, "output_peak": 0.4100546536683474}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.663404881968605, "全帯域_10ms_rms_envelope_nrmse": 0.09216982150184778, "全帯域_10ms_envelope_correlation": 0.9920068058858419, "全帯域_10ms_modulation_mae_db": 0.37671766188811495, "全帯域_40ms_log_rms_mae_db": 0.3872104319613824, "全帯域_40ms_rms_envelope_nrmse": 0.04992120933034241, "全帯域_40ms_envelope_correlation": 0.9974623997880044, "全帯域_40ms_modulation_mae_db": 0.4047505328554856, "300–1000 Hz_10ms_log_rms_mae_db": 1.156722724762074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09812216296341934, "300–1000 Hz_10ms_envelope_correlation": 0.9931401856390139, "300–1000 Hz_10ms_modulation_mae_db": 0.49087696879477954, "300–1000 Hz_40ms_log_rms_mae_db": 0.7577719480417952, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059519163552319836, "300–1000 Hz_40ms_envelope_correlation": 0.997353807054363, "300–1000 Hz_40ms_modulation_mae_db": 0.5813139944250804, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0478659419465333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14406921284559387, "1000–3000 Hz_10ms_envelope_correlation": 0.9851206639815276, "1000–3000 Hz_10ms_modulation_mae_db": 0.9425888064249297, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6969747321085371, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08597866570707231, "1000–3000 Hz_40ms_envelope_correlation": 0.995089315912034, "1000–3000 Hz_40ms_modulation_mae_db": 1.0900638119370387, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0767264077298393, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18134103423116327, "3000–6000 Hz_10ms_envelope_correlation": 0.981423073454232, "3000–6000 Hz_10ms_modulation_mae_db": 0.5219190633048539, "3000–6000 Hz_40ms_log_rms_mae_db": 0.853023698102889, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11005530169333244, "3000–6000 Hz_40ms_envelope_correlation": 0.9931433264160021, "3000–6000 Hz_40ms_modulation_mae_db": 0.5808366867619782, "mr_stft_db": 4.297263318449808, "band_mae_db": 0.11547619246884118, "rms_gain_error_db": -0.045829854641858375, "output_peak": 0.4437816292437913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6789019621890099, "全帯域_10ms_rms_envelope_nrmse": 0.09074379637502344, "全帯域_10ms_envelope_correlation": 0.9923879163443521, "全帯域_10ms_modulation_mae_db": 0.8303320722786368, "全帯域_40ms_log_rms_mae_db": 0.3836356514766254, "全帯域_40ms_rms_envelope_nrmse": 0.04982570491790161, "全帯域_40ms_envelope_correlation": 0.9976628872922133, "全帯域_40ms_modulation_mae_db": 0.9242503668160476, "300–1000 Hz_10ms_log_rms_mae_db": 1.2080251659747252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1040016901544131, "300–1000 Hz_10ms_envelope_correlation": 0.9923129490799762, "300–1000 Hz_10ms_modulation_mae_db": 0.510455628319253, "300–1000 Hz_40ms_log_rms_mae_db": 0.7568652353124272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05903746813440332, "300–1000 Hz_40ms_envelope_correlation": 0.9974739207308969, "300–1000 Hz_40ms_modulation_mae_db": 0.5800929656493459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0804174666185937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1579117699905809, "1000–3000 Hz_10ms_envelope_correlation": 0.9820018958988852, "1000–3000 Hz_10ms_modulation_mae_db": 0.9060485787346996, "1000–3000 Hz_40ms_log_rms_mae_db": 0.692013376694595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08819949174563482, "1000–3000 Hz_40ms_envelope_correlation": 0.9949604727225206, "1000–3000 Hz_40ms_modulation_mae_db": 0.9080735356793364, "3000–6000 Hz_10ms_log_rms_mae_db": 1.093564953260316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18474254663642564, "3000–6000 Hz_10ms_envelope_correlation": 0.9793508875164573, "3000–6000 Hz_10ms_modulation_mae_db": 1.0875435276497294, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8271885253311273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10407350283579174, "3000–6000 Hz_40ms_envelope_correlation": 0.9935141967696349, "3000–6000 Hz_40ms_modulation_mae_db": 1.2197108946876922, "mr_stft_db": 4.319229945353242, "band_mae_db": 0.19243803758234077, "rms_gain_error_db": -0.13130684479859603, "output_peak": 0.4130627505418257}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6840396279298383, "全帯域_10ms_rms_envelope_nrmse": 0.08459526580044556, "全帯域_10ms_envelope_correlation": 0.9933954770825162, "全帯域_10ms_modulation_mae_db": 0.38667080606044185, "全帯域_40ms_log_rms_mae_db": 0.3769823956190271, "全帯域_40ms_rms_envelope_nrmse": 0.043516303495576454, "全帯域_40ms_envelope_correlation": 0.9982598772337422, "全帯域_40ms_modulation_mae_db": 0.4920391427750241, "300–1000 Hz_10ms_log_rms_mae_db": 1.155799365080829, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1004326357480519, "300–1000 Hz_10ms_envelope_correlation": 0.9929095594852193, "300–1000 Hz_10ms_modulation_mae_db": 0.5467951472305105, "300–1000 Hz_40ms_log_rms_mae_db": 0.709920737054815, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06593775430419035, "300–1000 Hz_40ms_envelope_correlation": 0.9969009574989642, "300–1000 Hz_40ms_modulation_mae_db": 0.5575191390198245, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0829602861853196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15269481813870278, "1000–3000 Hz_10ms_envelope_correlation": 0.9827767131701601, "1000–3000 Hz_10ms_modulation_mae_db": 0.325151153363935, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6730113831055403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0897868810120074, "1000–3000 Hz_40ms_envelope_correlation": 0.9940175389015319, "1000–3000 Hz_40ms_modulation_mae_db": 0.5416454609095033, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0531548752344972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16345204134862565, "3000–6000 Hz_10ms_envelope_correlation": 0.9838084049492564, "3000–6000 Hz_10ms_modulation_mae_db": 0.8308493493338241, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8151757756730478, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09508862943584828, "3000–6000 Hz_40ms_envelope_correlation": 0.9943793677080824, "3000–6000 Hz_40ms_modulation_mae_db": 0.9850675135581316, "mr_stft_db": 4.315426441965663, "band_mae_db": 0.16211144410355552, "rms_gain_error_db": -0.1296047118378371, "output_peak": 0.44057248770032803}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6630707327146752, "全帯域_10ms_rms_envelope_nrmse": 0.09641893384892505, "全帯域_10ms_envelope_correlation": 0.9913071082725068, "全帯域_10ms_modulation_mae_db": 0.5169668514544548, "全帯域_40ms_log_rms_mae_db": 0.36820420950752475, "全帯域_40ms_rms_envelope_nrmse": 0.05385289999469238, "全帯域_40ms_envelope_correlation": 0.9970521830281086, "全帯域_40ms_modulation_mae_db": 0.6181864648606948, "300–1000 Hz_10ms_log_rms_mae_db": 1.1575133166332556, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11366287171883437, "300–1000 Hz_10ms_envelope_correlation": 0.99081337639915, "300–1000 Hz_10ms_modulation_mae_db": 0.40639493949261607, "300–1000 Hz_40ms_log_rms_mae_db": 0.6805058882066731, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06294060823404878, "300–1000 Hz_40ms_envelope_correlation": 0.997030201494929, "300–1000 Hz_40ms_modulation_mae_db": 0.21217076480363847, "1000–3000 Hz_10ms_log_rms_mae_db": 1.056689326812859, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15373269437189427, "1000–3000 Hz_10ms_envelope_correlation": 0.982353554293513, "1000–3000 Hz_10ms_modulation_mae_db": 0.5874924085820425, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6617160979949057, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09280058391294906, "1000–3000 Hz_40ms_envelope_correlation": 0.9932653099976924, "1000–3000 Hz_40ms_modulation_mae_db": 0.78305541468066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0758360966403182, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19727157499457276, "3000–6000 Hz_10ms_envelope_correlation": 0.9779420173874503, "3000–6000 Hz_10ms_modulation_mae_db": 1.2254663206076561, "3000–6000 Hz_40ms_log_rms_mae_db": 0.824383536795854, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12138684971546035, "3000–6000 Hz_40ms_envelope_correlation": 0.9915883098016518, "3000–6000 Hz_40ms_modulation_mae_db": 1.1553248237512654, "mr_stft_db": 4.303714140139586, "band_mae_db": 0.16933363427050205, "rms_gain_error_db": 0.01976936276364321, "output_peak": 0.4372023495210147}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6637495941625896, "全帯域_10ms_rms_envelope_nrmse": 0.09021988242543084, "全帯域_10ms_envelope_correlation": 0.9923669836193184, "全帯域_10ms_modulation_mae_db": 0.43312671426663757, "全帯域_40ms_log_rms_mae_db": 0.4072274194544842, "全帯域_40ms_rms_envelope_nrmse": 0.05021485169422536, "全帯域_40ms_envelope_correlation": 0.9974731283861642, "全帯域_40ms_modulation_mae_db": 0.5541717307036756, "300–1000 Hz_10ms_log_rms_mae_db": 1.189302600138316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11016774892613672, "300–1000 Hz_10ms_envelope_correlation": 0.9914064171366779, "300–1000 Hz_10ms_modulation_mae_db": 0.5929272937116447, "300–1000 Hz_40ms_log_rms_mae_db": 0.8081535087846203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07455925363839855, "300–1000 Hz_40ms_envelope_correlation": 0.9959350794592403, "300–1000 Hz_40ms_modulation_mae_db": 0.7310690239472638, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0773645113726382, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.148468255292346, "1000–3000 Hz_10ms_envelope_correlation": 0.9838694466836678, "1000–3000 Hz_10ms_modulation_mae_db": 0.27364202827728396, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6969791381047901, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09094993505539908, "1000–3000 Hz_40ms_envelope_correlation": 0.9940114085729079, "1000–3000 Hz_40ms_modulation_mae_db": 0.3722100629453431, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0931006051471637, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1497218837792518, "3000–6000 Hz_10ms_envelope_correlation": 0.9866185539275288, "3000–6000 Hz_10ms_modulation_mae_db": 0.20594318144699372, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8619194636499865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0910574779347178, "3000–6000 Hz_40ms_envelope_correlation": 0.9947568565708937, "3000–6000 Hz_40ms_modulation_mae_db": 0.3359908318230453, "mr_stft_db": 4.313743085748623, "band_mae_db": 0.11862142000915361, "rms_gain_error_db": -0.07339234596301322, "output_peak": 0.42323284006951173}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.25645805972517, "spectral_convergence": 0.3573005519213731, "spectral_convergence_status": "ok", "amplitude_floor": 1.0971770400792783e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.243096341043141, "spectral_convergence": 0.3521209286226189, "spectral_convergence_status": "ok", "amplitude_floor": 6.752711905169626e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.4067744034673515, "spectral_convergence": 0.37064278851885724, "spectral_convergence_status": "ok", "amplitude_floor": 5.858543156261471e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003996059347809583, "output_power": 0.00039853385443247075, "delta_db": -0.011667100571015825, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00030506165058519967, "output_power": 0.00029819080045991747, "delta_db": -0.09893374927630551, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002850207180759584, "output_power": 0.00027714439233818234, "delta_db": -0.1217033427508526, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017376514501848305, "output_power": 0.0001646081673169787, "delta_db": -0.23511287462620467, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002541837569284722, "output_power": 0.0002474702460641304, "delta_db": -0.1162480429035964, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023122095798090296, "output_power": 0.00022188528713245018, "delta_db": -0.17898690400673217, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1271086690224512, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10942428786276827, "pitch": null, "output_rms_dbfs": -27.93759130033861, "output_peak": 0.4214168494513621, "reference_rms_dbfs": -27.828167012475838, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.302109601411888}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015167567220720444, "log_rms_mae_db": 0.6617259997352102, "rms_envelope_nrmse": 0.08793877060274559, "envelope_correlation": 0.9928171014219933, "rms_gain_error_db": -0.10942427926911558, "reference_mean_power": 0.0016502333338478435, "output_mean_power": 0.001609173705643005}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103350035549285e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021122178949431673, "output_power": 0.00020082462843806792, "reference_active": true, "delta_db": -0.2192174557965282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018786794884032712, "output_power": 0.00014929573570918967, "reference_active": true, "delta_db": -0.9980529045479315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.617925051494854e-05, "output_power": 2.7305844085020467e-05, "reference_active": true, "delta_db": 0.18298397109942638}], "active_band_count": 3, "active_band_mae_db": 0.4667514438146287}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014379702931337134, "log_rms_mae_db": 0.37689242496852804, "rms_envelope_nrmse": 0.04588642166130236, "envelope_correlation": 0.9980007904905764, "rms_gain_error_db": -0.10942444879221815, "reference_mean_power": 0.0016585326982667058, "output_mean_power": 0.0016172665095789893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0955155560805602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019170493341265238, "output_power": 0.00018136711269472984, "reference_active": true, "delta_db": -0.2407475010390449}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.87936190430084e-05, "output_power": 6.433921998862512e-05, "reference_active": true, "delta_db": -0.880152569064942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1061472825588806e-06, "output_power": 2.0190367469969067e-06, "reference_active": true, "delta_db": -0.1834451476968277}], "active_band_count": 3, "active_band_mae_db": 0.4347817392669382}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010769717002798511, "log_rms_mae_db": 1.209194211676607, "rms_envelope_nrmse": 0.10306547024249349, "envelope_correlation": 0.9924550797919316, "rms_gain_error_db": -0.11386528786617672, "reference_mean_power": 0.0005226470255969395, "output_mean_power": 0.0005091221059102147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.88003979644585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021484485335499559, "output_power": 0.00020542961811285832, "reference_active": true, "delta_db": -0.1946189573462834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0907744988019035e-05, "output_power": 3.472758245131826e-05, "reference_active": true, "delta_db": -0.7113098886402519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.643789476493196e-06, "output_power": 8.64465063910063e-06, "reference_active": true, "delta_db": 0.00043265692710096806}], "active_band_count": 3, "active_band_mae_db": 0.30212050097121207}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.974444405364767e-05, "log_rms_mae_db": 0.7916727513091386, "rms_envelope_nrmse": 0.0620269378826432, "envelope_correlation": 0.997193926589986, "rms_gain_error_db": -0.11386527086472722, "reference_mean_power": 0.0005252755682472565, "output_mean_power": 0.0005116826298412353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.5461080823927677e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019511319942898067, "output_power": 0.0001866970754345736, "reference_active": true, "delta_db": -0.19149135574615028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8168801206395614e-05, "output_power": 1.5879370791650888e-05, "reference_active": true, "delta_db": -0.5849298328490169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.358572422696516e-07, "output_power": 6.197235079552058e-07, "reference_active": true, "delta_db": -0.7459559799536937}], "active_band_count": 3, "active_band_mae_db": 0.5074590561829536}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.224188331064728e-05, "log_rms_mae_db": 1.0484617513886416, "rms_envelope_nrmse": 0.13773837726150656, "envelope_correlation": 0.9858311792988118, "rms_gain_error_db": -0.13952544611258055, "reference_mean_power": 0.00028515402410464236, "output_mean_power": 0.00027613849918916124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.768616554938329e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001096694429350374, "output_power": 0.00010018436331885128, "reference_active": true, "delta_db": -0.3928569493402984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.898846044442003e-05, "output_power": 4.6817955771609584e-05, "reference_active": true, "delta_db": -0.19681344114050078}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.522948535574358e-06, "output_power": 9.47059725520659e-06, "reference_active": true, "delta_db": 0.45787502246561884}], "active_band_count": 3, "active_band_mae_db": 0.34918180431547263}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.641451530936745e-05, "log_rms_mae_db": 0.6530062692695028, "rms_envelope_nrmse": 0.08132024220426232, "envelope_correlation": 0.9947765136466745, "rms_gain_error_db": -0.13952553101764423, "reference_mean_power": 0.0002865881294452572, "output_mean_power": 0.00027752725795040546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.5110883907146213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010245851234810692, "output_power": 9.419670983385479e-05, "reference_active": true, "delta_db": -0.36512312096431776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5140429681693116e-05, "output_power": 2.513075540344809e-05, "reference_active": true, "delta_db": -0.0016715284125331537}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.446669785409335e-07, "output_power": 7.508706311727616e-07, "reference_active": true, "delta_db": 0.03603022138976998}], "active_band_count": 3, "active_band_mae_db": 0.13427495692220695}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001420862345348913, "log_rms_mae_db": 1.1047005957763283, "rms_envelope_nrmse": 0.17780753717341363, "envelope_correlation": 0.9808887574321828, "rms_gain_error_db": -0.2411815901777539, "reference_mean_power": 0.00023492613523189528, "output_mean_power": 0.0002222353672650007}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4459776587443134e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000158142036380199, "output_power": 0.00014938634715505006, "reference_active": true, "delta_db": -0.2473641898700471}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.639730090455687e-05, "output_power": 7.348625084609156e-05, "reference_active": true, "delta_db": -1.1785878304044743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.267646990584481e-05, "output_power": 1.0075526609139639e-05, "reference_active": true, "delta_db": -0.9973057542629578}], "active_band_count": 3, "active_band_mae_db": 0.807752591512493}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010454414213188805, "log_rms_mae_db": 0.8415071508325133, "rms_envelope_nrmse": 0.10527044162569875, "envelope_correlation": 0.993260628255889, "rms_gain_error_db": -0.2411816256998463, "reference_mean_power": 0.00023610764979113905, "output_mean_power": 0.00022335305428847847}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2342047370765003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017362277943143474, "output_power": 0.00015731679050281364, "reference_active": true, "delta_db": -0.42831626748971385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.8688155273395004e-05, "output_power": 3.690968025141215e-05, "reference_active": true, "delta_db": -1.2028303712413455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.341220903531162e-06, "output_power": 9.893963617862588e-07, "reference_active": true, "delta_db": -1.3213000444336558}], "active_band_count": 3, "active_band_mae_db": 0.9841488943882384}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9095093846078987, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dc83305fdac0eb979dc4a9794686924996bf00f4af97f63214de0f5f855353a0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9834683392573974, "standard_deviation": 0.001532815892095479, "minimum": 0.9815837434148803, "maximum": 0.9858311792988118, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0702637065248133, "standard_deviation": 0.016614257916722698, "minimum": 1.0478659419465333, "maximum": 1.0900575412983322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5436808576741148, "standard_deviation": 0.2568827086751879, "minimum": 0.27364202827728396, "maximum": 0.9425888064249297, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14962466410821004, "standard_deviation": 0.006918278772656897, "minimum": 0.13773837726150656, "maximum": 0.1579117699905809, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9940041540682338, "standard_deviation": 0.0012104466971018734, "minimum": 0.9914030106992596, "maximum": 0.995089315912034, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6820722998057421, "standard_deviation": 0.021566106440464635, "minimum": 0.6530062692695028, "maximum": 0.7158905827289979, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5966291751276388, "standard_deviation": 0.31153447091968456, "minimum": 0.13427495692220695, "maximum": 1.0900638119370387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08961262849154558, "standard_deviation": 0.0067204731635784, "minimum": 0.08132024220426232, "maximum": 0.10349505737052576, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9819387604876855, "standard_deviation": 0.002719961418821236, "minimum": 0.9779420173874503, "maximum": 0.9866185539275288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.0879092252848332, "standard_deviation": 0.01895470207014667, "minimum": 1.0531548752344972, "maximum": 1.113915784736873, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7234081043891518, "standard_deviation": 0.3348278159686435, "minimum": 0.20594318144699372, "maximum": 1.2254663206076561, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17475355984280644, "standard_deviation": 0.014313415905828594, "minimum": 0.1497218837792518, "maximum": 0.19727157499457276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935181543817514, "standard_deviation": 0.0009844877752870592, "minimum": 0.9915883098016518, "maximum": 0.9947568565708937, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8382453995727704, "standard_deviation": 0.015712416451839965, "minimum": 0.8151757756730478, "maximum": 0.8619194636499865, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7980707606711506, "standard_deviation": 0.3302963920134172, "minimum": 0.3359908318230453, "maximum": 1.2197108946876922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10379473786249563, "standard_deviation": 0.009876903019151567, "minimum": 0.0910574779347178, "maximum": 0.12138684971546035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921963074365818, "standard_deviation": 0.0007586456358184308, "minimum": 0.99081337639915, "maximum": 0.9931401856390139, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1796482424599726, "standard_deviation": 0.02721743261735847, "minimum": 1.1500743264880078, "maximum": 1.2105542289259654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48753757173370565, "standard_deviation": 0.13174256644496948, "minimum": 0.30212050097121207, "maximum": 0.7021593225856658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10494228003452537, "standard_deviation": 0.005037177672616671, "minimum": 0.09812216296341934, "maximum": 0.11366287171883437, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969816315187294, "standard_deviation": 0.000544279555002048, "minimum": 0.9959350794592403, "maximum": 0.9975039282925048, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7532688324284809, "standard_deviation": 0.0427329772494817, "minimum": 0.6805058882066731, "maximum": 0.8081535087846203, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5394324054473247, "standard_deviation": 0.15957135287170687, "minimum": 0.21217076480363847, "maximum": 0.7310690239472638, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06432135130545233, "standard_deviation": 0.006174750416972898, "minimum": 0.05822158587923836, "maximum": 0.07455925363839855, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14471232718031074, "standard_deviation": 0.02797566676807742, "minimum": 0.11547619246884118, "maximum": 0.19243803758234077, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.307907000849384, "standard_deviation": 0.007974647599605868, "minimum": 4.297263318449808, "maximum": 4.319229945353242, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4288291753578413, "standard_deviation": 0.013494351856418474, "minimum": 0.4100546536683474, "maximum": 0.4437816292437913, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07471036259181006, "standard_deviation": 0.05177190464169529, "minimum": -0.13130684479859603, "maximum": 0.01976936276364321, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924081674143849, "standard_deviation": 0.0006340710784617041, "minimum": 0.9913071082725068, "maximum": 0.9933954770825162, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.671922320315919, "standard_deviation": 0.010051687551895552, "minimum": 0.6617259997352102, "maximum": 0.6853218167513306, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4096225739256171, "standard_deviation": 0.2237539366377157, "minimum": 0.09016532776847386, "maximum": 0.8303320722786368, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09011695253182717, "standard_deviation": 0.003611097489147795, "minimum": 0.08459526580044556, "maximum": 0.09641893384892505, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976327479306182, "standard_deviation": 0.0003687237806784776, "minimum": 0.9970521830281086, "maximum": 0.9982598772337422, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.38336233829991034, "standard_deviation": 0.01388953623986614, "minimum": 0.36820420950752475, "maximum": 0.4072274194544842, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4943397666706474, "standard_deviation": 0.21456329052666165, "minimum": 0.25447026556200775, "maximum": 0.9242503668160476, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04894610187616803, "standard_deviation": 0.003167599736010585, "minimum": 0.043516303495576454, "maximum": 0.05385289999469238, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-127105-0035", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.244995355606079, "predicted_mos": 2.244995355606079, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e5ed77401447955a1ffe6b63fa05bc75cde6f7be2f4b2b3bdfa1693f3506b486", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9305218001827598, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6396242287791588, "全帯域_10ms_rms_envelope_nrmse": 0.08834335345509489, "全帯域_10ms_envelope_correlation": 0.9926693839565395, "全帯域_10ms_modulation_mae_db": 0.25788575034055344, "全帯域_40ms_log_rms_mae_db": 0.36777846384048707, "全帯域_40ms_rms_envelope_nrmse": 0.05073003190551097, "全帯域_40ms_envelope_correlation": 0.9973726208959893, "全帯域_40ms_modulation_mae_db": 0.3476054076466277, "300–1000 Hz_10ms_log_rms_mae_db": 1.335865263109056, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10437770594136482, "300–1000 Hz_10ms_envelope_correlation": 0.9922474064585094, "300–1000 Hz_10ms_modulation_mae_db": 0.44103525814500855, "300–1000 Hz_40ms_log_rms_mae_db": 0.8439096588527141, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06773888206026756, "300–1000 Hz_40ms_envelope_correlation": 0.9966106091631367, "300–1000 Hz_40ms_modulation_mae_db": 0.6667850313106093, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1399544001167663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1590247575822906, "1000–3000 Hz_10ms_envelope_correlation": 0.9810685516396511, "1000–3000 Hz_10ms_modulation_mae_db": 0.686857118965067, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6962561533414685, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08294138301459868, "1000–3000 Hz_40ms_envelope_correlation": 0.9944297334977598, "1000–3000 Hz_40ms_modulation_mae_db": 0.6736851941960206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.171285342968153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17030491566504533, "3000–6000 Hz_10ms_envelope_correlation": 0.9839100924920962, "3000–6000 Hz_10ms_modulation_mae_db": 0.5499206540610723, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8623826301994952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10586224256065718, "3000–6000 Hz_40ms_envelope_correlation": 0.9939457041927015, "3000–6000 Hz_40ms_modulation_mae_db": 0.586573227086803, "mr_stft_db": 4.396301390875708, "band_mae_db": 0.168933171250193, "rms_gain_error_db": -0.006984250750376292, "output_peak": 0.43437158437384044}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.663483263923103, "全帯域_10ms_rms_envelope_nrmse": 0.09275784420593572, "全帯域_10ms_envelope_correlation": 0.9920100171838832, "全帯域_10ms_modulation_mae_db": 0.38186471421428586, "全帯域_40ms_log_rms_mae_db": 0.3614186148131361, "全帯域_40ms_rms_envelope_nrmse": 0.05240066929451617, "全帯域_40ms_envelope_correlation": 0.997375712091009, "全帯域_40ms_modulation_mae_db": 0.6705210957891659, "300–1000 Hz_10ms_log_rms_mae_db": 1.3151991407481252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10080145004174315, "300–1000 Hz_10ms_envelope_correlation": 0.9927296224780021, "300–1000 Hz_10ms_modulation_mae_db": 0.6070753075964325, "300–1000 Hz_40ms_log_rms_mae_db": 0.92379264146846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06847842367985243, "300–1000 Hz_40ms_envelope_correlation": 0.9964478641038109, "300–1000 Hz_40ms_modulation_mae_db": 0.7641332062960635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.164648791406319, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16716960204069164, "1000–3000 Hz_10ms_envelope_correlation": 0.979045747740764, "1000–3000 Hz_10ms_modulation_mae_db": 0.2961957610889125, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6980700148521075, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08639428550151594, "1000–3000 Hz_40ms_envelope_correlation": 0.9939738180852027, "1000–3000 Hz_40ms_modulation_mae_db": 0.4770785364550651, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1941742473984056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19146736922233548, "3000–6000 Hz_10ms_envelope_correlation": 0.9787834678036954, "3000–6000 Hz_10ms_modulation_mae_db": 0.7080302573836205, "3000–6000 Hz_40ms_log_rms_mae_db": 0.929760203056042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13100364166497636, "3000–6000 Hz_40ms_envelope_correlation": 0.9907536844260875, "3000–6000 Hz_40ms_modulation_mae_db": 0.7117699819057884, "mr_stft_db": 4.404666699026985, "band_mae_db": 0.14106489058295948, "rms_gain_error_db": -0.1304216304756921, "output_peak": 0.4685530926977903}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6617293464731749, "全帯域_10ms_rms_envelope_nrmse": 0.09863061122824805, "全帯域_10ms_envelope_correlation": 0.9908586790207914, "全帯域_10ms_modulation_mae_db": 0.5029720187925665, "全帯域_40ms_log_rms_mae_db": 0.3542811310383359, "全帯域_40ms_rms_envelope_nrmse": 0.04944117913832545, "全帯域_40ms_envelope_correlation": 0.997499777146463, "全帯域_40ms_modulation_mae_db": 0.5052404245096052, "300–1000 Hz_10ms_log_rms_mae_db": 1.2914643171066988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09586835429738853, "300–1000 Hz_10ms_envelope_correlation": 0.9934810233888649, "300–1000 Hz_10ms_modulation_mae_db": 0.40193171303190334, "300–1000 Hz_40ms_log_rms_mae_db": 0.8154291998423985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05029545805007016, "300–1000 Hz_40ms_envelope_correlation": 0.9981374438247836, "300–1000 Hz_40ms_modulation_mae_db": 0.6323264843896006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1324759001076141, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15382059116166452, "1000–3000 Hz_10ms_envelope_correlation": 0.9822637019895333, "1000–3000 Hz_10ms_modulation_mae_db": 0.6043343627258854, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6728373024128973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07688404475366038, "1000–3000 Hz_40ms_envelope_correlation": 0.9953153055067496, "1000–3000 Hz_40ms_modulation_mae_db": 0.7146506245897454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2067993716084566, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21197201666606263, "3000–6000 Hz_10ms_envelope_correlation": 0.9740694618089341, "3000–6000 Hz_10ms_modulation_mae_db": 0.705677781546903, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9307278704753443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12224619003723741, "3000–6000 Hz_40ms_envelope_correlation": 0.9911570596332451, "3000–6000 Hz_40ms_modulation_mae_db": 0.7922499316573733, "mr_stft_db": 4.415732386355387, "band_mae_db": 0.11102985770324424, "rms_gain_error_db": -0.0039418290608754115, "output_peak": 0.3953023842012907}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6483918563521732, "全帯域_10ms_rms_envelope_nrmse": 0.10206165718637132, "全帯域_10ms_envelope_correlation": 0.990285356766179, "全帯域_10ms_modulation_mae_db": 0.8920058033198998, "全帯域_40ms_log_rms_mae_db": 0.35164692706523143, "全帯域_40ms_rms_envelope_nrmse": 0.06689801872002367, "全帯域_40ms_envelope_correlation": 0.9954781786911001, "全帯域_40ms_modulation_mae_db": 1.1232106074097425, "300–1000 Hz_10ms_log_rms_mae_db": 1.3061645686575332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1124183449263438, "300–1000 Hz_10ms_envelope_correlation": 0.9909569828381359, "300–1000 Hz_10ms_modulation_mae_db": 0.6323121958304571, "300–1000 Hz_40ms_log_rms_mae_db": 0.8321424072489084, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06590206411792653, "300–1000 Hz_40ms_envelope_correlation": 0.9967354206470159, "300–1000 Hz_40ms_modulation_mae_db": 0.4031332359539576, "1000–3000 Hz_10ms_log_rms_mae_db": 1.152632981515773, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15011294591651905, "1000–3000 Hz_10ms_envelope_correlation": 0.9834020899137601, "1000–3000 Hz_10ms_modulation_mae_db": 1.048883826049923, "1000–3000 Hz_40ms_log_rms_mae_db": 0.731075968497684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08230996261887864, "1000–3000 Hz_40ms_envelope_correlation": 0.9950969608748437, "1000–3000 Hz_40ms_modulation_mae_db": 1.1747717930750767, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1725292742959412, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2291073644296812, "3000–6000 Hz_10ms_envelope_correlation": 0.9740918621249116, "3000–6000 Hz_10ms_modulation_mae_db": 1.907604026887875, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8703249872398434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17097291587205865, "3000–6000 Hz_40ms_envelope_correlation": 0.9863710716448474, "3000–6000 Hz_40ms_modulation_mae_db": 2.017680805777563, "mr_stft_db": 4.404924634213631, "band_mae_db": 0.18801658272533184, "rms_gain_error_db": 0.011820906949263119, "output_peak": 0.5048209740170552}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6361846531122302, "全帯域_10ms_rms_envelope_nrmse": 0.08086376848937156, "全帯域_10ms_envelope_correlation": 0.9938571196861892, "全帯域_10ms_modulation_mae_db": 0.5842799961184572, "全帯域_40ms_log_rms_mae_db": 0.3696964699000373, "全帯域_40ms_rms_envelope_nrmse": 0.043903415732042594, "全帯域_40ms_envelope_correlation": 0.9980446062919719, "全帯域_40ms_modulation_mae_db": 0.7150985454978133, "300–1000 Hz_10ms_log_rms_mae_db": 1.2786672845989258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10787502581367579, "300–1000 Hz_10ms_envelope_correlation": 0.9917449059868989, "300–1000 Hz_10ms_modulation_mae_db": 0.3923306695069682, "300–1000 Hz_40ms_log_rms_mae_db": 0.8215522537226704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07050824226669122, "300–1000 Hz_40ms_envelope_correlation": 0.9963746273313479, "300–1000 Hz_40ms_modulation_mae_db": 0.35598097905233655, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1542473348963658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15237292042070466, "1000–3000 Hz_10ms_envelope_correlation": 0.982731937054608, "1000–3000 Hz_10ms_modulation_mae_db": 0.9348060983025128, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7133033265299794, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07592714133671612, "1000–3000 Hz_40ms_envelope_correlation": 0.9957350625441324, "1000–3000 Hz_40ms_modulation_mae_db": 1.017845746455146, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1519827539599838, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14844957101526543, "3000–6000 Hz_10ms_envelope_correlation": 0.9868788220425209, "3000–6000 Hz_10ms_modulation_mae_db": 0.7672341411221413, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8318031204237366, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07733077207154382, "3000–6000 Hz_40ms_envelope_correlation": 0.9962519572609703, "3000–6000 Hz_40ms_modulation_mae_db": 0.9131680038262905, "mr_stft_db": 4.386730658185916, "band_mae_db": 0.12923854579956517, "rms_gain_error_db": -0.04681029249081561, "output_peak": 0.40881462497642057}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6644697608665086, "全帯域_10ms_rms_envelope_nrmse": 0.0945612738518728, "全帯域_10ms_envelope_correlation": 0.9916430463712809, "全帯域_10ms_modulation_mae_db": 0.6482257208299685, "全帯域_40ms_log_rms_mae_db": 0.3802081096211404, "全帯域_40ms_rms_envelope_nrmse": 0.05505216461050554, "全帯域_40ms_envelope_correlation": 0.9970079637500375, "全帯域_40ms_modulation_mae_db": 0.47750426464719903, "300–1000 Hz_10ms_log_rms_mae_db": 1.3296016599330254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11290026609172694, "300–1000 Hz_10ms_envelope_correlation": 0.9909509349319964, "300–1000 Hz_10ms_modulation_mae_db": 0.5455770288806704, "300–1000 Hz_40ms_log_rms_mae_db": 0.8804814780855114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06846781538385344, "300–1000 Hz_40ms_envelope_correlation": 0.9965699208256492, "300–1000 Hz_40ms_modulation_mae_db": 0.636503492399297, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1488910071598528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16980453657357983, "1000–3000 Hz_10ms_envelope_correlation": 0.9783469936112125, "1000–3000 Hz_10ms_modulation_mae_db": 0.8377505305097209, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7393306563311588, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11598305719413513, "1000–3000 Hz_40ms_envelope_correlation": 0.9891631372051525, "1000–3000 Hz_40ms_modulation_mae_db": 1.0074498871175879, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1963487123294696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18793539512036492, "3000–6000 Hz_10ms_envelope_correlation": 0.9784664960830104, "3000–6000 Hz_10ms_modulation_mae_db": 0.49751497780153103, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9014396262715341, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10120436346038761, "3000–6000 Hz_40ms_envelope_correlation": 0.9934652070291377, "3000–6000 Hz_40ms_modulation_mae_db": 0.6293617545244299, "mr_stft_db": 4.404528264459262, "band_mae_db": 0.11871336933600636, "rms_gain_error_db": -0.1003807625411307, "output_peak": 0.43324162408596056}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6433186883593043, "全帯域_10ms_rms_envelope_nrmse": 0.08839091027943935, "全帯域_10ms_envelope_correlation": 0.9928633305098469, "全帯域_10ms_modulation_mae_db": 0.7124462677239656, "全帯域_40ms_log_rms_mae_db": 0.3526421847125383, "全帯域_40ms_rms_envelope_nrmse": 0.05092694041652672, "全帯域_40ms_envelope_correlation": 0.9976345467809679, "全帯域_40ms_modulation_mae_db": 0.9040832374410975, "300–1000 Hz_10ms_log_rms_mae_db": 1.235742042646319, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10321461143784637, "300–1000 Hz_10ms_envelope_correlation": 0.9924853037404898, "300–1000 Hz_10ms_modulation_mae_db": 0.5476790300357789, "300–1000 Hz_40ms_log_rms_mae_db": 0.8435200712425809, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06530018498978603, "300–1000 Hz_40ms_envelope_correlation": 0.9969290862621939, "300–1000 Hz_40ms_modulation_mae_db": 0.7800601860072361, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1901276929794173, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16558183888704806, "1000–3000 Hz_10ms_envelope_correlation": 0.9794767836665557, "1000–3000 Hz_10ms_modulation_mae_db": 0.7323342049307296, "1000–3000 Hz_40ms_log_rms_mae_db": 0.732028076035657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08588036628002226, "1000–3000 Hz_40ms_envelope_correlation": 0.9943317033589512, "1000–3000 Hz_40ms_modulation_mae_db": 0.9369648083571763, "3000–6000 Hz_10ms_log_rms_mae_db": 1.190185912925229, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17390595933405104, "3000–6000 Hz_10ms_envelope_correlation": 0.9834541697948861, "3000–6000 Hz_10ms_modulation_mae_db": 1.08485011358149, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9116331558777487, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1273415633325713, "3000–6000 Hz_40ms_envelope_correlation": 0.992057770555064, "3000–6000 Hz_40ms_modulation_mae_db": 1.351366178317378, "mr_stft_db": 4.405760863118073, "band_mae_db": 0.17787033575941769, "rms_gain_error_db": -0.1534288353013093, "output_peak": 0.42691280105214313}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6750657664975149, "全帯域_10ms_rms_envelope_nrmse": 0.09340193184518557, "全帯域_10ms_envelope_correlation": 0.9917985724148128, "全帯域_10ms_modulation_mae_db": 0.11602419400156865, "全帯域_40ms_log_rms_mae_db": 0.38589127429020287, "全帯域_40ms_rms_envelope_nrmse": 0.04711165595127852, "全帯域_40ms_envelope_correlation": 0.9977726593109055, "全帯域_40ms_modulation_mae_db": 0.12300739505847098, "300–1000 Hz_10ms_log_rms_mae_db": 1.273630520490114, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10139187319256338, "300–1000 Hz_10ms_envelope_correlation": 0.9926662922945773, "300–1000 Hz_10ms_modulation_mae_db": 0.27171393176932496, "300–1000 Hz_40ms_log_rms_mae_db": 0.7983277596808304, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05622792495505173, "300–1000 Hz_40ms_envelope_correlation": 0.9976671968717181, "300–1000 Hz_40ms_modulation_mae_db": 0.19424530658286085, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1871118320322116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17642365378481933, "1000–3000 Hz_10ms_envelope_correlation": 0.9768218831073346, "1000–3000 Hz_10ms_modulation_mae_db": 0.4870693249330351, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7422690686257006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10448814129794952, "1000–3000 Hz_40ms_envelope_correlation": 0.9917575657533899, "1000–3000 Hz_40ms_modulation_mae_db": 0.5027570025700399, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1996264296513814, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1783376596350538, "3000–6000 Hz_10ms_envelope_correlation": 0.9808131522193958, "3000–6000 Hz_10ms_modulation_mae_db": 1.0901166733166823, "3000–6000 Hz_40ms_log_rms_mae_db": 0.917371279797881, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10202310881047755, "3000–6000 Hz_40ms_envelope_correlation": 0.9933212390389418, "3000–6000 Hz_40ms_modulation_mae_db": 1.0934317468792807, "mr_stft_db": 4.410828634706898, "band_mae_db": 0.12964884534073795, "rms_gain_error_db": -0.06756558411906209, "output_peak": 0.4208539691949261}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.395208882865693, "spectral_convergence": 0.37251369948600815, "spectral_convergence_status": "ok", "amplitude_floor": 1.0971770400792783e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.353356744312125, "spectral_convergence": 0.3657705079030013, "spectral_convergence_status": "ok", "amplitude_floor": 6.752711905169626e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.440338545449306, "spectral_convergence": 0.38787920250448316, "spectral_convergence_status": "ok", "amplitude_floor": 5.858543156261471e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003996059347809583, "output_power": 0.0003952206636987841, "delta_db": -0.047922864209742885, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00030506165058519967, "output_power": 0.00029743521455661664, "delta_db": -0.10995230642984472, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002850207180759584, "output_power": 0.00027943894914588143, "delta_db": -0.08589490468820546, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017376514501848305, "output_power": 0.0001670058401625163, "delta_db": -0.17231008531503134, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002541837569284722, "output_power": 0.0002828684821757718, "delta_db": 0.46436765671783864, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023122095798090296, "output_power": 0.0002242394787281107, "delta_db": -0.13315121014049497, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.168933171250193, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.006984250750376292, "pitch": null, "output_rms_dbfs": -27.835151263226216, "output_peak": 0.43437158437384044, "reference_rms_dbfs": -27.828167012475838, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.396301390875708}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015167567220720444, "log_rms_mae_db": 0.6396242287791588, "rms_envelope_nrmse": 0.08834335345509489, "envelope_correlation": 0.9926693839565395, "rms_gain_error_db": -0.006984230136486097, "reference_mean_power": 0.0016502333338478435, "output_mean_power": 0.0016475815970171382}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103350035549285e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021122178949431673, "output_power": 0.0002140380847018898, "reference_active": true, "delta_db": 0.057523383870493675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018786794884032712, "output_power": 0.00019194071224814507, "reference_active": true, "delta_db": 0.09314408308764525}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.617925051494854e-05, "output_power": 2.268079277220725e-05, "reference_active": true, "delta_db": -0.6229897840635215}], "active_band_count": 3, "active_band_mae_db": 0.25788575034055344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014379702931337134, "log_rms_mae_db": 0.36777846384048707, "rms_envelope_nrmse": 0.05073003190551097, "envelope_correlation": 0.9973726208959893, "rms_gain_error_db": -0.0069840133783616995, "reference_mean_power": 0.0016585326982667058, "output_mean_power": 0.00165586770794858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0955155560805602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019170493341265238, "output_power": 0.00019747609024645452, "reference_active": true, "delta_db": 0.12881230869810756}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.87936190430084e-05, "output_power": 8.109591912718386e-05, "reference_active": true, "delta_db": 0.12507952016993346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1061472825588806e-06, "output_power": 1.756290283525978e-06, "reference_active": true, "delta_db": -0.788924394071842}], "active_band_count": 3, "active_band_mae_db": 0.3476054076466277}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010769717002798511, "log_rms_mae_db": 1.335865263109056, "rms_envelope_nrmse": 0.10437770594136482, "envelope_correlation": 0.9922474064585094, "rms_gain_error_db": -0.11624149769545304, "reference_mean_power": 0.0005226470255969395, "output_mean_power": 0.0005088436197443047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.88003979644585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021484485335499559, "output_power": 0.00021002303631415144, "reference_active": true, "delta_db": -0.09858021738969022}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0907744988019035e-05, "output_power": 3.2669000020525094e-05, "reference_active": true, "delta_db": -0.9766969879415253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.643789476493196e-06, "output_power": 8.164344541504664e-06, "reference_active": true, "delta_db": -0.24782856910381026}], "active_band_count": 3, "active_band_mae_db": 0.44103525814500855}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.974444405364767e-05, "log_rms_mae_db": 0.8439096588527141, "rms_envelope_nrmse": 0.06773888206026756, "envelope_correlation": 0.9966106091631367, "rms_gain_error_db": -0.11624149445141121, "reference_mean_power": 0.0005252755682472565, "output_mean_power": 0.0005114027414669702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.5461080823927677e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019511319942898067, "output_power": 0.00019111893511133786, "reference_active": true, "delta_db": -0.08982933537984286}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8168801206395614e-05, "output_power": 1.4739789470411757e-05, "reference_active": true, "delta_db": -0.9083499261394089}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.358572422696516e-07, "output_power": 5.842194147064378e-07, "reference_active": true, "delta_db": -1.0021758324125762}], "active_band_count": 3, "active_band_mae_db": 0.6667850313106093}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.224188331064728e-05, "log_rms_mae_db": 1.1399544001167663, "rms_envelope_nrmse": 0.1590247575822906, "envelope_correlation": 0.9810685516396511, "rms_gain_error_db": -0.048330422497979655, "reference_mean_power": 0.00028515402410464236, "output_mean_power": 0.00028199828199484495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.768616554938329e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001096694429350374, "output_power": 0.00011076823154292233, "reference_active": true, "delta_db": 0.043295846095783304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.898846044442003e-05, "output_power": 4.728371160880366e-05, "reference_active": true, "delta_db": -0.15382231305337324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.522948535574358e-06, "output_power": 5.5493810430278264e-06, "reference_active": true, "delta_db": -1.8634531977460442}], "active_band_count": 3, "active_band_mae_db": 0.686857118965067}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.641451530936745e-05, "log_rms_mae_db": 0.6962561533414685, "rms_envelope_nrmse": 0.08294138301459868, "envelope_correlation": 0.9944297334977598, "rms_gain_error_db": -0.048330632356149504, "reference_mean_power": 0.0002865881294452572, "output_mean_power": 0.00028341650268556254}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.5110883907146213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010245851234810692, "output_power": 0.00010382821367597546, "reference_active": true, "delta_db": 0.0576733638649534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5140429681693116e-05, "output_power": 2.3891515062749716e-05, "reference_active": true, "delta_db": -0.22129004968169827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.446669785409335e-07, "output_power": 4.986006944388036e-07, "reference_active": true, "delta_db": -1.7420921690414104}], "active_band_count": 3, "active_band_mae_db": 0.6736851941960206}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001420862345348913, "log_rms_mae_db": 1.171285342968153, "rms_envelope_nrmse": 0.17030491566504533, "envelope_correlation": 0.9839100924920962, "rms_gain_error_db": 0.3092053884986463, "reference_mean_power": 0.00023492613523189528, "output_mean_power": 0.000252262022310358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4459776587443134e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000158142036380199, "output_power": 0.00018188719035090972, "reference_active": true, "delta_db": 0.6075478750277865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.639730090455687e-05, "output_power": 0.00011177645829466588, "reference_active": true, "delta_db": 0.6428547065389802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.267646990584481e-05, "output_power": 1.3897422722697852e-05, "reference_active": true, "delta_db": 0.3993593806164502}], "active_band_count": 3, "active_band_mae_db": 0.5499206540610723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010454414213188805, "log_rms_mae_db": 0.8623826301994952, "rms_envelope_nrmse": 0.10586224256065718, "envelope_correlation": 0.9939457041927015, "rms_gain_error_db": 0.30920536478467003, "reference_mean_power": 0.00023610764979113905, "output_mean_power": 0.0002535307229025088}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2342047370765003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017362277943143474, "output_power": 0.000203056228673644, "reference_active": true, "delta_db": 0.6800961152888203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.8688155273395004e-05, "output_power": 5.662127132300235e-05, "reference_active": true, "delta_db": 0.6555629635282968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.341220903531162e-06, "output_power": 1.2164499734011917e-06, "reference_active": true, "delta_db": -0.4240606024432922}], "active_band_count": 3, "active_band_mae_db": 0.586573227086803}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9054321861817344, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6fabff4506ec628657ff5c0138050508ead2049faf53b6301170335be06c1dd0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9803947110904274, "standard_deviation": 0.0023322178349334806, "minimum": 0.9768218831073346, "maximum": 0.9834020899137601, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.15876124252679, "standard_deviation": 0.02077856866359311, "minimum": 1.1324759001076141, "maximum": 1.1901276929794173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7035289034382233, "standard_deviation": 0.24333894565038308, "minimum": 0.2961957610889125, "maximum": 1.048883826049923, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16178885579591473, "standard_deviation": 0.009395194760420918, "minimum": 0.15011294591651905, "maximum": 0.17642365378481933, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937254108532727, "standard_deviation": 0.0022058278883034257, "minimum": 0.9891631372051525, "maximum": 0.9957350625441324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7156463208283317, "standard_deviation": 0.024782370422423393, "minimum": 0.6728373024128973, "maximum": 0.7422690686257006, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8131504491019821, "standard_deviation": 0.25751383408647693, "minimum": 0.4770785364550651, "maximum": 1.1747717930750767, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08885104774968458, "standard_deviation": 0.014058407296077076, "minimum": 0.07592714133671612, "maximum": 0.11598305719413513, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9800584405461814, "standard_deviation": 0.0046149007162906465, "minimum": 0.9740694618089341, "maximum": 0.9868788220425209, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1853665056421274, "standard_deviation": 0.01836795955955835, "minimum": 1.1519827539599838, "maximum": 1.2067993716084566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9138685782126644, "standard_deviation": 0.4568261799426661, "minimum": 0.49751497780153103, "maximum": 1.907604026887875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18643503138598247, "standard_deviation": 0.02514603646281167, "minimum": 0.14844957101526543, "maximum": 0.2291073644296812, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921654617226243, "standard_deviation": 0.0029152410283541782, "minimum": 0.9863710716448474, "maximum": 0.9962519572609703, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8944303591677032, "standard_deviation": 0.03579177079428974, "minimum": 0.8318031204237366, "maximum": 0.9307278704753443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0119502037468635, "standard_deviation": 0.47940294436170533, "minimum": 0.586573227086803, "maximum": 2.017680805777563, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11724809972623873, "standard_deviation": 0.027812284876857343, "minimum": 0.07733077207154382, "maximum": 0.17097291587205865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921578090146844, "standard_deviation": 0.0008880733663605516, "minimum": 0.9909509349319964, "maximum": 0.9934810233888649, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2957918496612248, "standard_deviation": 0.03304174866183943, "minimum": 1.235742042646319, "maximum": 1.335865263109056, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.479956891849568, "standard_deviation": 0.12358330216750364, "minimum": 0.27171393176932496, "maximum": 0.6323121958304571, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1048559539678316, "standard_deviation": 0.005893750217480514, "minimum": 0.09586835429738853, "maximum": 0.11290026609172694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996934021128707, "standard_deviation": 0.0006337725778844629, "minimum": 0.9963746273313479, "maximum": 0.9981374438247836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8448944337680093, "standard_deviation": 0.04006842406796116, "minimum": 0.7983277596808304, "maximum": 0.92379264146846, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5541459902489951, "standard_deviation": 0.2112267170160919, "minimum": 0.19424530658286085, "maximum": 0.7800601860072361, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06411487443793738, "standard_deviation": 0.007068809657412815, "minimum": 0.05029545805007016, "maximum": 0.07050824226669122, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14556444981218197, "standard_deviation": 0.028897991064643653, "minimum": 0.11102985770324424, "maximum": 0.18801658272533184, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.4036841913677325, "standard_deviation": 0.008829003731501186, "minimum": 4.386730658185916, "maximum": 4.415732386355387, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4366088818249284, "standard_deviation": 0.03486841544211229, "minimum": 0.3953023842012907, "maximum": 0.5048209740170552, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0622140347237498, "standard_deviation": 0.061669836351965886, "minimum": -0.1534288353013093, "maximum": 0.011820906949263119, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9919981882386903, "standard_deviation": 0.001137226326605769, "minimum": 0.990285356766179, "maximum": 0.9938571196861892, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.654033445545396, "standard_deviation": 0.014003263037577594, "minimum": 0.6361846531122302, "maximum": 0.6750657664975149, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5119630581676582, "standard_deviation": 0.2526297512751782, "minimum": 0.11602419400156865, "maximum": 0.8920058033198998, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0923764188176899, "standard_deviation": 0.006585487625863165, "minimum": 0.08086376848937156, "maximum": 0.10206165718637132, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972732581198056, "standard_deviation": 0.0007873355568894609, "minimum": 0.9954781786911001, "maximum": 0.9980446062919719, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.36544539691013866, "standard_deviation": 0.01284335420388838, "minimum": 0.35164692706523143, "maximum": 0.38589127429020287, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6082838722499653, "standard_deviation": 0.31602383105324544, "minimum": 0.12300739505847098, "maximum": 1.1232106074097425, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05205800947109121, "standard_deviation": 0.006867758108573371, "minimum": 0.043903415732042594, "maximum": 0.06689801872002367, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_121-127105-0035", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.3039052486419678, "predicted_mos": 2.3039052486419678, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "27a3a61997e03b53c0a586c4b894434229da8aab09e4a3184d4256ece093f5db", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.976044800132513, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6349389517573747, "全帯域_10ms_rms_envelope_nrmse": 0.08662929866240299, "全帯域_10ms_envelope_correlation": 0.993026721369907, "全帯域_10ms_modulation_mae_db": 0.42910338904594864, "全帯域_40ms_log_rms_mae_db": 0.3540246623394189, "全帯域_40ms_rms_envelope_nrmse": 0.04774958309490192, "全帯域_40ms_envelope_correlation": 0.997813159808006, "全帯域_40ms_modulation_mae_db": 0.2908489086812405, "300–1000 Hz_10ms_log_rms_mae_db": 1.269309471293182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1106389727709063, "300–1000 Hz_10ms_envelope_correlation": 0.9913047677527196, "300–1000 Hz_10ms_modulation_mae_db": 0.5663299043916471, "300–1000 Hz_40ms_log_rms_mae_db": 0.8221569317615823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06643467178280278, "300–1000 Hz_40ms_envelope_correlation": 0.9968031423152748, "300–1000 Hz_40ms_modulation_mae_db": 0.28531558784522054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1528515546811657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16325965210843674, "1000–3000 Hz_10ms_envelope_correlation": 0.9799968148884898, "1000–3000 Hz_10ms_modulation_mae_db": 0.20572641274860068, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7124439533035464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08787869264427639, "1000–3000 Hz_40ms_envelope_correlation": 0.9937797694382361, "1000–3000 Hz_40ms_modulation_mae_db": 0.23045041184852647, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1588856437882102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15754848204860736, "3000–6000 Hz_10ms_envelope_correlation": 0.9850649527684805, "3000–6000 Hz_10ms_modulation_mae_db": 0.8613265048305179, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8473414137012496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09211508736124213, "3000–6000 Hz_40ms_envelope_correlation": 0.9948664172876246, "3000–6000 Hz_40ms_modulation_mae_db": 1.0104224985104306, "mr_stft_db": 4.3926465609077985, "band_mae_db": 0.13375377338249306, "rms_gain_error_db": -0.11514685953175882, "output_peak": 0.39290630671789556}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6493468342413575, "全帯域_10ms_rms_envelope_nrmse": 0.09300335611130026, "全帯域_10ms_envelope_correlation": 0.9918722994868067, "全帯域_10ms_modulation_mae_db": 0.033315073803911355, "全帯域_40ms_log_rms_mae_db": 0.35481817951937633, "全帯域_40ms_rms_envelope_nrmse": 0.046966273743289384, "全帯域_40ms_envelope_correlation": 0.9977394995228164, "全帯域_40ms_modulation_mae_db": 0.12246755080065162, "300–1000 Hz_10ms_log_rms_mae_db": 1.311362306434768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10578347975748294, "300–1000 Hz_10ms_envelope_correlation": 0.9920116965764226, "300–1000 Hz_10ms_modulation_mae_db": 0.5830732379249114, "300–1000 Hz_40ms_log_rms_mae_db": 0.857440616671002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06631252427193059, "300–1000 Hz_40ms_envelope_correlation": 0.996673685672036, "300–1000 Hz_40ms_modulation_mae_db": 0.6891530712894053, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1610032663627572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16400993577838344, "1000–3000 Hz_10ms_envelope_correlation": 0.9799216338649723, "1000–3000 Hz_10ms_modulation_mae_db": 0.4504767435680818, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7316248848047758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10085449084147893, "1000–3000 Hz_40ms_envelope_correlation": 0.9917581842289671, "1000–3000 Hz_40ms_modulation_mae_db": 0.4554222347705333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.128044036873574, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18932654365016638, "3000–6000 Hz_10ms_envelope_correlation": 0.9783092628246726, "3000–6000 Hz_10ms_modulation_mae_db": 0.7508093807517779, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8260609556021398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09222364114444111, "3000–6000 Hz_40ms_envelope_correlation": 0.994559892098948, "3000–6000 Hz_40ms_modulation_mae_db": 0.9958924540767429, "mr_stft_db": 4.379494241890156, "band_mae_db": 0.07084508379399927, "rms_gain_error_db": -0.003447499555393723, "output_peak": 0.4139064502004348}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6706989267052624, "全帯域_10ms_rms_envelope_nrmse": 0.09260502073925705, "全帯域_10ms_envelope_correlation": 0.9919376900087646, "全帯域_10ms_modulation_mae_db": 0.5008860235434599, "全帯域_40ms_log_rms_mae_db": 0.3640123837831411, "全帯域_40ms_rms_envelope_nrmse": 0.047399570121710775, "全帯域_40ms_envelope_correlation": 0.9977347256377287, "全帯域_40ms_modulation_mae_db": 0.44999012672734656, "300–1000 Hz_10ms_log_rms_mae_db": 1.231814970569568, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1048034642538149, "300–1000 Hz_10ms_envelope_correlation": 0.9922292960028997, "300–1000 Hz_10ms_modulation_mae_db": 0.34829375387130795, "300–1000 Hz_40ms_log_rms_mae_db": 0.817213580443447, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06208018152395862, "300–1000 Hz_40ms_envelope_correlation": 0.9972373089798635, "300–1000 Hz_40ms_modulation_mae_db": 0.3222702502610701, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1796508440098472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16364260932194846, "1000–3000 Hz_10ms_envelope_correlation": 0.9798987561900414, "1000–3000 Hz_10ms_modulation_mae_db": 0.4604562889072832, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6989788722278734, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08623248992801773, "1000–3000 Hz_40ms_envelope_correlation": 0.9940193248996534, "1000–3000 Hz_40ms_modulation_mae_db": 0.4091145954028339, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1812979840143056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18395646173832514, "3000–6000 Hz_10ms_envelope_correlation": 0.9795826350431773, "3000–6000 Hz_10ms_modulation_mae_db": 0.7554550695350937, "3000–6000 Hz_40ms_log_rms_mae_db": 0.888560531117689, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1028744853296382, "3000–6000 Hz_40ms_envelope_correlation": 0.9932086571875843, "3000–6000 Hz_40ms_modulation_mae_db": 0.6797922972522588, "mr_stft_db": 4.379809353475955, "band_mae_db": 0.074655373412045, "rms_gain_error_db": -0.06235839573765173, "output_peak": 0.4245246082545641}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6524037206302397, "全帯域_10ms_rms_envelope_nrmse": 0.0877026316523017, "全帯域_10ms_envelope_correlation": 0.9927660716668093, "全帯域_10ms_modulation_mae_db": 0.27230179639665325, "全帯域_40ms_log_rms_mae_db": 0.36281579811711023, "全帯域_40ms_rms_envelope_nrmse": 0.050279864384409304, "全帯域_40ms_envelope_correlation": 0.997424251333834, "全帯域_40ms_modulation_mae_db": 0.2164546854423243, "300–1000 Hz_10ms_log_rms_mae_db": 1.2743206458211291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10509034338165238, "300–1000 Hz_10ms_envelope_correlation": 0.9923145724777135, "300–1000 Hz_10ms_modulation_mae_db": 0.6330831074278759, "300–1000 Hz_40ms_log_rms_mae_db": 0.8533879759110877, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06729685003913527, "300–1000 Hz_40ms_envelope_correlation": 0.99685246735432, "300–1000 Hz_40ms_modulation_mae_db": 0.8050773294705276, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1139629726470697, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1522146836289693, "1000–3000 Hz_10ms_envelope_correlation": 0.9827522495150923, "1000–3000 Hz_10ms_modulation_mae_db": 1.2605151705607407, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6621370743550525, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08377920078932713, "1000–3000 Hz_40ms_envelope_correlation": 0.9946560659918724, "1000–3000 Hz_40ms_modulation_mae_db": 1.3363884187967432, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1912512272152025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16563819942200375, "3000–6000 Hz_10ms_envelope_correlation": 0.9839506255786015, "3000–6000 Hz_10ms_modulation_mae_db": 0.42970940713631606, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8780664475304837, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10949848722255333, "3000–6000 Hz_40ms_envelope_correlation": 0.9926918503092337, "3000–6000 Hz_40ms_modulation_mae_db": 0.41381979095095794, "mr_stft_db": 4.375687610791977, "band_mae_db": 0.18134967941478775, "rms_gain_error_db": -0.04510807701190584, "output_peak": 0.4752091238057389}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.657034650206123, "全帯域_10ms_rms_envelope_nrmse": 0.10216108171810036, "全帯域_10ms_envelope_correlation": 0.9903480525960163, "全帯域_10ms_modulation_mae_db": 0.8377887077933476, "全帯域_40ms_log_rms_mae_db": 0.3505613995030272, "全帯域_40ms_rms_envelope_nrmse": 0.05239995889464328, "全帯域_40ms_envelope_correlation": 0.9974824212439745, "全帯域_40ms_modulation_mae_db": 0.6177227177586667, "300–1000 Hz_10ms_log_rms_mae_db": 1.2793195398855453, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09567224169266285, "300–1000 Hz_10ms_envelope_correlation": 0.9934841616098806, "300–1000 Hz_10ms_modulation_mae_db": 0.28960807655607596, "300–1000 Hz_40ms_log_rms_mae_db": 0.8126571020612835, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05138648087650045, "300–1000 Hz_40ms_envelope_correlation": 0.9980511010551836, "300–1000 Hz_40ms_modulation_mae_db": 0.3231026299763724, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1393025751688486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1383745772078966, "1000–3000 Hz_10ms_envelope_correlation": 0.9860133624459301, "1000–3000 Hz_10ms_modulation_mae_db": 0.3732076192098272, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6855652692245893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07531499138873145, "1000–3000 Hz_40ms_envelope_correlation": 0.9960458734209272, "1000–3000 Hz_40ms_modulation_mae_db": 0.6790429789418675, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1710243201863906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2416064868779038, "3000–6000 Hz_10ms_envelope_correlation": 0.9649406514182833, "3000–6000 Hz_10ms_modulation_mae_db": 1.0458047120416494, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8878433614360286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14293143160239247, "3000–6000 Hz_40ms_envelope_correlation": 0.9888846052835438, "3000–6000 Hz_40ms_modulation_mae_db": 0.9827514165818183, "mr_stft_db": 4.384382471973307, "band_mae_db": 0.18604700519853737, "rms_gain_error_db": -0.1553647190779591, "output_peak": 0.40775711889252186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6427220203993882, "全帯域_10ms_rms_envelope_nrmse": 0.08193244040628044, "全帯域_10ms_envelope_correlation": 0.9937191072769298, "全帯域_10ms_modulation_mae_db": 0.4224288183199329, "全帯域_40ms_log_rms_mae_db": 0.3448514689259006, "全帯域_40ms_rms_envelope_nrmse": 0.03936363365618335, "全帯域_40ms_envelope_correlation": 0.9984760892390061, "全帯域_40ms_modulation_mae_db": 0.5468653117009502, "300–1000 Hz_10ms_log_rms_mae_db": 1.2314121805371547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10551955827045394, "300–1000 Hz_10ms_envelope_correlation": 0.9920540331549987, "300–1000 Hz_10ms_modulation_mae_db": 0.44721437075997544, "300–1000 Hz_40ms_log_rms_mae_db": 0.810303739942524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057694407234292935, "300–1000 Hz_40ms_envelope_correlation": 0.9975078435449277, "300–1000 Hz_40ms_modulation_mae_db": 0.4048509132015982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1484331405595556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15718575648007735, "1000–3000 Hz_10ms_envelope_correlation": 0.9815220022402126, "1000–3000 Hz_10ms_modulation_mae_db": 0.6189087492285621, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6683773147682369, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08535657307445683, "1000–3000 Hz_40ms_envelope_correlation": 0.9943392978695055, "1000–3000 Hz_40ms_modulation_mae_db": 0.6937022146528893, "3000–6000 Hz_10ms_log_rms_mae_db": 1.176897645542587, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1740640578021947, "3000–6000 Hz_10ms_envelope_correlation": 0.9817898857622477, "3000–6000 Hz_10ms_modulation_mae_db": 1.0771098483080308, "3000–6000 Hz_40ms_log_rms_mae_db": 0.844138814027979, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09521923164348894, "3000–6000 Hz_40ms_envelope_correlation": 0.9947409113008574, "3000–6000 Hz_40ms_modulation_mae_db": 1.1626212134917069, "mr_stft_db": 4.390094962704985, "band_mae_db": 0.11061822155360208, "rms_gain_error_db": -0.07006009903765956, "output_peak": 0.42236251068112984}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6431232593794936, "全帯域_10ms_rms_envelope_nrmse": 0.08814960150605504, "全帯域_10ms_envelope_correlation": 0.9927453608960402, "全帯域_10ms_modulation_mae_db": 0.4179605108960773, "全帯域_40ms_log_rms_mae_db": 0.3519696111978867, "全帯域_40ms_rms_envelope_nrmse": 0.04219495623488719, "全帯域_40ms_envelope_correlation": 0.9982958150797389, "全帯域_40ms_modulation_mae_db": 0.4037255824187682, "300–1000 Hz_10ms_log_rms_mae_db": 1.2868203444278885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10058989435281214, "300–1000 Hz_10ms_envelope_correlation": 0.9930228059829335, "300–1000 Hz_10ms_modulation_mae_db": 0.738910550495524, "300–1000 Hz_40ms_log_rms_mae_db": 0.8569561174666536, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06594925687121597, "300–1000 Hz_40ms_envelope_correlation": 0.9970364218341875, "300–1000 Hz_40ms_modulation_mae_db": 0.9088678439614389, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1238417971189922, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15012850512321746, "1000–3000 Hz_10ms_envelope_correlation": 0.9831250513179514, "1000–3000 Hz_10ms_modulation_mae_db": 0.6515142941723902, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6667714266150024, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07774958362567126, "1000–3000 Hz_40ms_envelope_correlation": 0.9952450973600164, "1000–3000 Hz_40ms_modulation_mae_db": 0.8206194041258517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2131532157299665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1866860500442019, "3000–6000 Hz_10ms_envelope_correlation": 0.9789053805617633, "3000–6000 Hz_10ms_modulation_mae_db": 0.08103465525387389, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8751757734828742, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08722279473641055, "3000–6000 Hz_40ms_envelope_correlation": 0.9951006223281429, "3000–6000 Hz_40ms_modulation_mae_db": 0.31325025179834237, "mr_stft_db": 4.382779897986685, "band_mae_db": 0.11926885671788222, "rms_gain_error_db": -0.1012222392383093, "output_peak": 0.41652914668157365}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.637395599562474, "全帯域_10ms_rms_envelope_nrmse": 0.08351880148517185, "全帯域_10ms_envelope_correlation": 0.9934611008608648, "全帯域_10ms_modulation_mae_db": 0.7997244565555857, "全帯域_40ms_log_rms_mae_db": 0.3505135304553358, "全帯域_40ms_rms_envelope_nrmse": 0.046667714271663364, "全帯域_40ms_envelope_correlation": 0.9977981273794829, "全帯域_40ms_modulation_mae_db": 0.8910276158815323, "300–1000 Hz_10ms_log_rms_mae_db": 1.2371090200636097, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11233817863229391, "300–1000 Hz_10ms_envelope_correlation": 0.9909941566357665, "300–1000 Hz_10ms_modulation_mae_db": 0.43288039211378004, "300–1000 Hz_40ms_log_rms_mae_db": 0.7439928965752881, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06841527022583939, "300–1000 Hz_40ms_envelope_correlation": 0.9964783289614366, "300–1000 Hz_40ms_modulation_mae_db": 0.6082685668041656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.151450462192435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15372084592290927, "1000–3000 Hz_10ms_envelope_correlation": 0.982821434231687, "1000–3000 Hz_10ms_modulation_mae_db": 0.913424712533681, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7254785913294478, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08460710234528761, "1000–3000 Hz_40ms_envelope_correlation": 0.9951825474719351, "1000–3000 Hz_40ms_modulation_mae_db": 1.0057706957614563, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1663449142406603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14445757704283843, "3000–6000 Hz_10ms_envelope_correlation": 0.9873813500626183, "3000–6000 Hz_10ms_modulation_mae_db": 1.0295640048810517, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8311806214211604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07887376801344197, "3000–6000 Hz_40ms_envelope_correlation": 0.9960525050931852, "3000–6000 Hz_40ms_modulation_mae_db": 1.297368184231967, "mr_stft_db": 4.392850954867238, "band_mae_db": 0.17536796848699462, "rms_gain_error_db": -0.05145812757153354, "output_peak": 0.4282722739952118}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.405998761493727, "spectral_convergence": 0.36755273307233866, "spectral_convergence_status": "ok", "amplitude_floor": 1.0971770400792783e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.348924949785469, "spectral_convergence": 0.36102883558493465, "spectral_convergence_status": "ok", "amplitude_floor": 6.752711905169626e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.423015971444198, "spectral_convergence": 0.36941854495567494, "spectral_convergence_status": "ok", "amplitude_floor": 5.858543156261471e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003996059347809583, "output_power": 0.00039572747376827083, "delta_db": -0.042357269292754234, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00030506165058519967, "output_power": 0.0003058864058668213, "delta_db": 0.011725608471351519, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002850207180759584, "output_power": 0.0002732953732735373, "delta_db": -0.18244150489601502, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017376514501848305, "output_power": 0.00016724950996027665, "delta_db": -0.16597813177824783, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002541837569284722, "output_power": 0.00024274255281856762, "delta_db": -0.2000187952481776, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023122095798090296, "output_power": 0.000220814232950598, "delta_db": -0.2000013306084122, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13375377338249306, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11514685953175882, "pitch": null, "output_rms_dbfs": -27.9433138720076, "output_peak": 0.39290630671789556, "reference_rms_dbfs": -27.828167012475838, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.3926465609077985}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015167567220720444, "log_rms_mae_db": 0.6349389517573747, "rms_envelope_nrmse": 0.08662929866240299, "envelope_correlation": 0.993026721369907, "rms_gain_error_db": -0.11514684790708613, "reference_mean_power": 0.0016502333338478435, "output_mean_power": 0.001607054741881055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103350035549285e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021122178949431673, "output_power": 0.00020411453499214195, "reference_active": true, "delta_db": -0.14864785773380254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018786794884032712, "output_power": 0.00017139850052769981, "reference_active": true, "delta_db": -0.3984567553212731}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.617925051494854e-05, "output_power": 2.207682695779479e-05, "reference_active": true, "delta_db": -0.7402055540827702}], "active_band_count": 3, "active_band_mae_db": 0.42910338904594864}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014379702931337134, "log_rms_mae_db": 0.3540246623394189, "rms_envelope_nrmse": 0.04774958309490192, "envelope_correlation": 0.997813159808006, "rms_gain_error_db": -0.11514671532223943, "reference_mean_power": 0.0016585326982667058, "output_mean_power": 0.0016151370015472248}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0955155560805602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019170493341265238, "output_power": 0.0001857573111882664, "reference_active": true, "delta_db": -0.13687373245910758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.87936190430084e-05, "output_power": 7.530035347044772e-05, "reference_active": true, "delta_db": -0.19694033525177787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1061472825588806e-06, "output_power": 1.860439159633779e-06, "reference_active": true, "delta_db": -0.5387326583328362}], "active_band_count": 3, "active_band_mae_db": 0.2908489086812405}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010769717002798511, "log_rms_mae_db": 1.269309471293182, "rms_envelope_nrmse": 0.1106389727709063, "envelope_correlation": 0.9913047677527196, "rms_gain_error_db": -0.12496101114926908, "reference_mean_power": 0.0005226470255969395, "output_mean_power": 0.0005078230178467392}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.88003979644585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021484485335499559, "output_power": 0.0002020167162137216, "reference_active": true, "delta_db": -0.2673764718383925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0907744988019035e-05, "output_power": 3.561450507840947e-05, "reference_active": true, "delta_db": -0.6017862647351317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.643789476493196e-06, "output_power": 1.0463738522249675e-05, "reference_active": true, "delta_db": 0.8298269766014171}], "active_band_count": 3, "active_band_mae_db": 0.5663299043916471}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.974444405364767e-05, "log_rms_mae_db": 0.8221569317615823, "rms_envelope_nrmse": 0.06643467178280278, "envelope_correlation": 0.9968031423152748, "rms_gain_error_db": -0.12496102787786073, "reference_mean_power": 0.0005252755682472565, "output_mean_power": 0.000510377004320144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.5461080823927677e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019511319942898067, "output_power": 0.0001851353711830817, "reference_active": true, "delta_db": -0.22797249304811504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8168801206395614e-05, "output_power": 1.6048995336347925e-05, "reference_active": true, "delta_db": -0.5387842226780922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.358572422696516e-07, "output_power": 7.20899233499408e-07, "reference_active": true, "delta_db": -0.08919004780945443}], "active_band_count": 3, "active_band_mae_db": 0.28531558784522054}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.224188331064728e-05, "log_rms_mae_db": 1.1528515546811657, "rms_envelope_nrmse": 0.16325965210843674, "envelope_correlation": 0.9799968148884898, "rms_gain_error_db": -0.09815083137463805, "reference_mean_power": 0.00028515402410464236, "output_mean_power": 0.0002787818024078275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.768616554938329e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001096694429350374, "output_power": 0.00010807534335946624, "reference_active": true, "delta_db": -0.06359013502587557}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.898846044442003e-05, "output_power": 4.9801547948043784e-05, "reference_active": true, "delta_db": 0.07149050712642299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.522948535574358e-06, "output_power": 7.627461187885606e-06, "reference_active": true, "delta_db": -0.4820985960935034}], "active_band_count": 3, "active_band_mae_db": 0.20572641274860068}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.641451530936745e-05, "log_rms_mae_db": 0.7124439533035464, "rms_envelope_nrmse": 0.08787869264427639, "envelope_correlation": 0.9937797694382361, "rms_gain_error_db": -0.0981508834692086, "reference_mean_power": 0.0002865881294452572, "output_mean_power": 0.00028018385701529807}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.5110883907146213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010245851234810692, "output_power": 0.00010344479716523711, "reference_active": true, "delta_db": 0.04160606555753863}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5140429681693116e-05, "output_power": 2.5581114939973325e-05, "reference_active": true, "delta_db": 0.07546773006230928}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.446669785409335e-07, "output_power": 6.524306451110013e-07, "reference_active": true, "delta_db": -0.5742774399257315}], "active_band_count": 3, "active_band_mae_db": 0.23045041184852647}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001420862345348913, "log_rms_mae_db": 1.1588856437882102, "rms_envelope_nrmse": 0.15754848204860736, "envelope_correlation": 0.9850649527684805, "rms_gain_error_db": -0.23076635030480508, "reference_mean_power": 0.00023492613523189528, "output_mean_power": 0.00022276897117906735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4459776587443134e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000158142036380199, "output_power": 0.0001521476023629592, "reference_active": true, "delta_db": -0.16782214002735601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.639730090455687e-05, "output_power": 8.53736976641034e-05, "reference_active": true, "delta_db": -0.5274078217757534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.267646990584481e-05, "output_power": 8.205846230801685e-06, "reference_active": true, "delta_db": -1.8887495526884441}], "active_band_count": 3, "active_band_mae_db": 0.8613265048305179}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010454414213188805, "log_rms_mae_db": 0.8473414137012496, "rms_envelope_nrmse": 0.09211508736124213, "envelope_correlation": 0.9948664172876246, "rms_gain_error_db": -0.2307663682785192, "reference_mean_power": 0.00023610764979113905, "output_mean_power": 0.00022388934275822836}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2342047370765003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017362277943143474, "output_power": 0.00016357950635112342, "reference_active": true, "delta_db": -0.25877811075282137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.8688155273395004e-05, "output_power": 4.328694267903986e-05, "reference_active": true, "delta_db": -0.510664070063888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.341220903531162e-06, "output_power": 7.967421294269569e-07, "reference_active": true, "delta_db": -2.2618253147145824}], "active_band_count": 3, "active_band_mae_db": 1.0104224985104306}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9038979113582254, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1462213e8ab6703a2f08dc96de36bc8245ff92a307cd9a60a9c304fc99282b2b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9820064130867971, "standard_deviation": 0.0021243212996278804, "minimum": 0.9798987561900414, "maximum": 0.9860133624459301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1463120765925838, "standard_deviation": 0.02070708042561341, "minimum": 1.1139629726470697, "maximum": 1.1796508440098472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6167787488661458, "standard_deviation": 0.3346619808679374, "minimum": 0.20572641274860068, "maximum": 1.2605151705607407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15531707069647982, "standard_deviation": 0.008763553748686969, "minimum": 0.1383745772078966, "maximum": 0.16400993577838344, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9943782700851391, "standard_deviation": 0.001287840744842905, "minimum": 0.9917581842289671, "maximum": 0.9960458734209272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6939221733285655, "standard_deviation": 0.027401151476938633, "minimum": 0.6621370743550525, "maximum": 0.7316248848047758, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7038138692875877, "standard_deviation": 0.3542208532706723, "minimum": 0.23045041184852647, "maximum": 1.3363884187967432, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08522164057965592, "standard_deviation": 0.007643088187298749, "minimum": 0.07531499138873145, "maximum": 0.10085449084147893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9799905930024806, "standard_deviation": 0.006865685461446936, "minimum": 0.9649406514182833, "maximum": 0.9873813500626183, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.173362373448862, "standard_deviation": 0.024775658678722504, "minimum": 1.128044036873574, "maximum": 1.2131532157299665, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7538516978422889, "standard_deviation": 0.34539073704346984, "minimum": 0.08103465525387389, "maximum": 1.0771098483080308, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18041048232828016, "standard_deviation": 0.029155209640441968, "minimum": 0.14445757704283843, "maximum": 0.2416064868779038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937631826111399, "standard_deviation": 0.002239047522665616, "minimum": 0.9888846052835438, "maximum": 0.9960525050931852, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8597959897899505, "standard_deviation": 0.02547143140142627, "minimum": 0.8260609556021398, "maximum": 0.888560531117689, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.856989763361778, "standard_deviation": 0.35242963124934007, "minimum": 0.31325025179834237, "maximum": 1.297368184231967, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10011986588170108, "standard_deviation": 0.019627061511265892, "minimum": 0.07887376801344197, "maximum": 0.14293143160239247, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921769362741668, "standard_deviation": 0.0008148684245666738, "minimum": 0.9909941566357665, "maximum": 0.9934841616098806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2651835598791057, "standard_deviation": 0.0291378769089605, "minimum": 1.2314121805371547, "maximum": 1.311362306434768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5049241741926371, "standard_deviation": 0.15140631933725834, "minimum": 0.28960807655607596, "maximum": 0.738910550495524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10505451663900991, "standard_deviation": 0.005250983301812842, "minimum": 0.09567224169266285, "maximum": 0.11233817863229391, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970800374646538, "standard_deviation": 0.0005082537573417174, "minimum": 0.9964783289614366, "maximum": 0.9980511010551836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8217636201041085, "standard_deviation": 0.03740676094773163, "minimum": 0.7439928965752881, "maximum": 0.857440616671002, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5433632741012249, "standard_deviation": 0.24222963083956853, "minimum": 0.28531558784522054, "maximum": 0.9088678439614389, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06319620535320951, "standard_deviation": 0.005889666616917906, "minimum": 0.05138648087650045, "maximum": 0.06841527022583939, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13148824524504266, "standard_deviation": 0.046114490023792185, "minimum": 0.07084508379399927, "maximum": 0.18604700519853737, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.384718256824763, "standard_deviation": 0.006488649073041629, "minimum": 4.375687610791977, "maximum": 4.392850954867238, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42268344240363376, "standard_deviation": 0.02395847307724177, "minimum": 0.39290630671789556, "maximum": 0.4752091238057389, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07552075209527145, "standard_deviation": 0.04706745112607562, "minimum": -0.1553647190779591, "maximum": -0.003447499555393723, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924845505202673, "standard_deviation": 0.0010795195703079859, "minimum": 0.9903480525960163, "maximum": 0.9937191072769298, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6484579953602141, "standard_deviation": 0.011656259969651512, "minimum": 0.6349389517573747, "maximum": 0.6706989267052624, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.46418859704436455, "standard_deviation": 0.2621592975550873, "minimum": 0.033315073803911355, "maximum": 0.8377887077933476, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08946277903510871, "standard_deviation": 0.00642032814050268, "minimum": 0.08193244040628044, "maximum": 0.10216108171810036, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978455111555735, "standard_deviation": 0.00036576733313312164, "minimum": 0.997424251333834, "maximum": 0.9984760892390061, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3541958792301496, "standard_deviation": 0.00643790943000015, "minimum": 0.3448514689259006, "maximum": 0.3640123837831411, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4423878124264351, "standard_deviation": 0.24519581223249373, "minimum": 0.12246755080065162, "maximum": 0.8910276158815323, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.046627694300211076, "standard_deviation": 0.004158361508657515, "minimum": 0.03936363365618335, "maximum": 0.05239995889464328, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1221-135766-0005", "language": "en", "style": "normal", "speaker": "librispeech_1221", "source_utterance_id": "1221-135766-0005", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1221-135766-0005.wav", "source_file_sha256": "60387e4db1efc35f3ae13d39e0b62d47791f680301d7bbbb42ff42a0641d2b22", "decoded_pcm_sha256": "ae61e68356f4a720a3d9f729ddcb9fa3e695cdbc220194eb938e92185005daf4", "original_source_sha256": "57ad0fad287d98afb900ccaa898a6e75348b80a3388201cd3e1feee0a0afc6b1", "corpus_entry_id": "librispeech_1221-135766-0005"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.516181945800781, "predicted_mos": 4.516181945800781, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7ccd5e4ed85dd1100d19d3f4fd9a2c974dca00abcc6817ea21e805aa15923f99", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.911769099999219, "cpu_threads": 4}, "waveform_sha256": "bc7fa6d71f6aee37b8d76a9a811d92b77e57fed28d8e0624cf7dff4190039e39"}, "rows": [{"case": "librispeech_1221-135766-0005", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.2393245697021484, "predicted_mos": 2.2393245697021484, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0d5ae44ffb20405f6936ad3b1d9db5d74405d4f7854d03388a21bf1f9c57a25e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.831661199685186, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8668413332253561, "全帯域_10ms_rms_envelope_nrmse": 0.06327684740587114, "全帯域_10ms_envelope_correlation": 0.9949816965573561, "全帯域_10ms_modulation_mae_db": 0.2030124234101063, "全帯域_40ms_log_rms_mae_db": 0.4261080880327152, "全帯域_40ms_rms_envelope_nrmse": 0.033300550450813746, "全帯域_40ms_envelope_correlation": 0.9984840738961662, "全帯域_40ms_modulation_mae_db": 0.1659057701745755, "300–1000 Hz_10ms_log_rms_mae_db": 1.304406308106801, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061322511806752245, "300–1000 Hz_10ms_envelope_correlation": 0.9967593791769878, "300–1000 Hz_10ms_modulation_mae_db": 0.40398916185244627, "300–1000 Hz_40ms_log_rms_mae_db": 0.7190617905973719, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03719963865336173, "300–1000 Hz_40ms_envelope_correlation": 0.9987099190737779, "300–1000 Hz_40ms_modulation_mae_db": 0.4162429049906196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2169500660258117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0858889578320674, "1000–3000 Hz_10ms_envelope_correlation": 0.9942911522545248, "1000–3000 Hz_10ms_modulation_mae_db": 0.21266000849043856, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6928431512578567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05055744575716196, "1000–3000 Hz_40ms_envelope_correlation": 0.9980817971688818, "1000–3000 Hz_40ms_modulation_mae_db": 0.16317675754488412, "3000–6000 Hz_10ms_log_rms_mae_db": 1.24858965426988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1471487368092421, "3000–6000 Hz_10ms_envelope_correlation": 0.9823861785732546, "3000–6000 Hz_10ms_modulation_mae_db": 0.20528956968421622, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7008496437792803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09151103044362255, "3000–6000 Hz_40ms_envelope_correlation": 0.9928306165745835, "3000–6000 Hz_40ms_modulation_mae_db": 0.16511098310296526, "mr_stft_db": 5.346937873088461, "band_mae_db": 0.10366935070189336, "rms_gain_error_db": -0.042436936998511386, "output_peak": 0.3599799233858057}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8521189323007733, "全帯域_10ms_rms_envelope_nrmse": 0.06575713623302949, "全帯域_10ms_envelope_correlation": 0.9946287331240777, "全帯域_10ms_modulation_mae_db": 0.7391132745549928, "全帯域_40ms_log_rms_mae_db": 0.4047339589127459, "全帯域_40ms_rms_envelope_nrmse": 0.03551828788026171, "全帯域_40ms_envelope_correlation": 0.9983486969095959, "全帯域_40ms_modulation_mae_db": 0.5272470088827644, "300–1000 Hz_10ms_log_rms_mae_db": 1.311219999975951, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06434956601539785, "300–1000 Hz_10ms_envelope_correlation": 0.9964448963177769, "300–1000 Hz_10ms_modulation_mae_db": 0.3175230663294341, "300–1000 Hz_40ms_log_rms_mae_db": 0.7536237051291781, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03666085570649707, "300–1000 Hz_40ms_envelope_correlation": 0.9987915746040662, "300–1000 Hz_40ms_modulation_mae_db": 0.476217535143634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1760263247176208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08638664369030506, "1000–3000 Hz_10ms_envelope_correlation": 0.9940836795279246, "1000–3000 Hz_10ms_modulation_mae_db": 0.4848516878444909, "1000–3000 Hz_40ms_log_rms_mae_db": 0.66389691467786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04553307116324269, "1000–3000 Hz_40ms_envelope_correlation": 0.9983212164226685, "1000–3000 Hz_40ms_modulation_mae_db": 0.5981587460394321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2468679185186455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14750422788980913, "3000–6000 Hz_10ms_envelope_correlation": 0.9823521037983881, "3000–6000 Hz_10ms_modulation_mae_db": 0.8287992504056935, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7175237332156106, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08136238847595587, "3000–6000 Hz_40ms_envelope_correlation": 0.9945459242731381, "3000–6000 Hz_40ms_modulation_mae_db": 0.9317198575974223, "mr_stft_db": 5.349320027360865, "band_mae_db": 0.14297738383327135, "rms_gain_error_db": -0.08626265678433861, "output_peak": 0.3661721203165848}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8544700383832018, "全帯域_10ms_rms_envelope_nrmse": 0.06208257174290738, "全帯域_10ms_envelope_correlation": 0.9951910011553253, "全帯域_10ms_modulation_mae_db": 0.5638839189895558, "全帯域_40ms_log_rms_mae_db": 0.4369448786242736, "全帯域_40ms_rms_envelope_nrmse": 0.034417193644338026, "全帯域_40ms_envelope_correlation": 0.9984102929524272, "全帯域_40ms_modulation_mae_db": 0.7923433115636521, "300–1000 Hz_10ms_log_rms_mae_db": 1.309386836712583, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06972569049981993, "300–1000 Hz_10ms_envelope_correlation": 0.9957868565504702, "300–1000 Hz_10ms_modulation_mae_db": 0.27917060212593364, "300–1000 Hz_40ms_log_rms_mae_db": 0.7334079846794357, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04049259662120589, "300–1000 Hz_40ms_envelope_correlation": 0.9984837378285814, "300–1000 Hz_40ms_modulation_mae_db": 0.4760770040901823, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2385392210395334, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09256721073218367, "1000–3000 Hz_10ms_envelope_correlation": 0.9934117627280026, "1000–3000 Hz_10ms_modulation_mae_db": 0.5135082620305438, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7082173839260686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05343426469397889, "1000–3000 Hz_40ms_envelope_correlation": 0.9979348921443131, "1000–3000 Hz_40ms_modulation_mae_db": 0.5512563304337612, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2452037278795882, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15344524824613573, "3000–6000 Hz_10ms_envelope_correlation": 0.981389127130793, "3000–6000 Hz_10ms_modulation_mae_db": 0.8666523724155848, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7231955733985178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09014327086694601, "3000–6000 Hz_40ms_envelope_correlation": 0.9939387903148035, "3000–6000 Hz_40ms_modulation_mae_db": 1.2041357671649575, "mr_stft_db": 5.36062008296646, "band_mae_db": 0.15087688238105038, "rms_gain_error_db": -0.06848316751166203, "output_peak": 0.33402520935952995}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9292372152280047, "全帯域_10ms_rms_envelope_nrmse": 0.07473996012991664, "全帯域_10ms_envelope_correlation": 0.993098678901809, "全帯域_10ms_modulation_mae_db": 0.7562001347556725, "全帯域_40ms_log_rms_mae_db": 0.44410088292651956, "全帯域_40ms_rms_envelope_nrmse": 0.041124326099942, "全帯域_40ms_envelope_correlation": 0.9978405731386543, "全帯域_40ms_modulation_mae_db": 0.7723513751206408, "300–1000 Hz_10ms_log_rms_mae_db": 1.350307920364694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07762240554775529, "300–1000 Hz_10ms_envelope_correlation": 0.9948517129473161, "300–1000 Hz_10ms_modulation_mae_db": 0.4955653344437227, "300–1000 Hz_40ms_log_rms_mae_db": 0.7850233144100704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05090488612610975, "300–1000 Hz_40ms_envelope_correlation": 0.9976781580583997, "300–1000 Hz_40ms_modulation_mae_db": 0.47334477494039257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2177277697073812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10071053868575852, "1000–3000 Hz_10ms_envelope_correlation": 0.9925355324346707, "1000–3000 Hz_10ms_modulation_mae_db": 0.450236370864376, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6977875024473809, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05948872073029628, "1000–3000 Hz_40ms_envelope_correlation": 0.9978419472608283, "1000–3000 Hz_40ms_modulation_mae_db": 0.4678759207501308, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3148394557531724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20567935989207592, "3000–6000 Hz_10ms_envelope_correlation": 0.965114503433384, "3000–6000 Hz_10ms_modulation_mae_db": 0.8151592563045226, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7615987170537002, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10836040759984303, "3000–6000 Hz_40ms_envelope_correlation": 0.9894283923039473, "3000–6000 Hz_40ms_modulation_mae_db": 0.8123479194235795, "mr_stft_db": 5.347670877384623, "band_mae_db": 0.12367428984630137, "rms_gain_error_db": -0.11001410814576099, "output_peak": 0.3454904690598328}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8752043201622774, "全帯域_10ms_rms_envelope_nrmse": 0.06797837138602832, "全帯域_10ms_envelope_correlation": 0.9941927505341798, "全帯域_10ms_modulation_mae_db": 0.15148865793287225, "全帯域_40ms_log_rms_mae_db": 0.41559080000928145, "全帯域_40ms_rms_envelope_nrmse": 0.03404961470448396, "全帯域_40ms_envelope_correlation": 0.9984077015612937, "全帯域_40ms_modulation_mae_db": 0.17426085115400816, "300–1000 Hz_10ms_log_rms_mae_db": 1.4044504646256575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07095397059038093, "300–1000 Hz_10ms_envelope_correlation": 0.9956413727774642, "300–1000 Hz_10ms_modulation_mae_db": 0.2417123177711281, "300–1000 Hz_40ms_log_rms_mae_db": 0.779777791640567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04052746635777647, "300–1000 Hz_40ms_envelope_correlation": 0.9984568409121958, "300–1000 Hz_40ms_modulation_mae_db": 0.3177691540168339, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1556918299454706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09837191071034128, "1000–3000 Hz_10ms_envelope_correlation": 0.9923422059652364, "1000–3000 Hz_10ms_modulation_mae_db": 0.3337932405780395, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6777202899232748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05210001906962915, "1000–3000 Hz_40ms_envelope_correlation": 0.9978391841890508, "1000–3000 Hz_40ms_modulation_mae_db": 0.39876025304309737, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2451463930170255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14503848216237128, "3000–6000 Hz_10ms_envelope_correlation": 0.9829043354900957, "3000–6000 Hz_10ms_modulation_mae_db": 0.663803223152603, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7503239412456147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08047871724381697, "3000–6000 Hz_40ms_envelope_correlation": 0.9945937175569841, "3000–6000 Hz_40ms_modulation_mae_db": 0.6800006511286457, "mr_stft_db": 5.341835116834878, "band_mae_db": 0.14253670894976303, "rms_gain_error_db": -0.04864108481909821, "output_peak": 0.3306437437748273}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8369256399520757, "全帯域_10ms_rms_envelope_nrmse": 0.0606231388347518, "全帯域_10ms_envelope_correlation": 0.995419853310957, "全帯域_10ms_modulation_mae_db": 0.32253843471482735, "全帯域_40ms_log_rms_mae_db": 0.37973852336783237, "全帯域_40ms_rms_envelope_nrmse": 0.03374242425806888, "全帯域_40ms_envelope_correlation": 0.9984769051077462, "全帯域_40ms_modulation_mae_db": 0.5124270736302187, "300–1000 Hz_10ms_log_rms_mae_db": 1.3615181573028554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06572286094680335, "300–1000 Hz_10ms_envelope_correlation": 0.996269076051319, "300–1000 Hz_10ms_modulation_mae_db": 0.3030004684229044, "300–1000 Hz_40ms_log_rms_mae_db": 0.7963384860060665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037853424239050644, "300–1000 Hz_40ms_envelope_correlation": 0.9986730661933837, "300–1000 Hz_40ms_modulation_mae_db": 0.3863687630902601, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1435978375605038, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08492642179290422, "1000–3000 Hz_10ms_envelope_correlation": 0.9947184817002444, "1000–3000 Hz_10ms_modulation_mae_db": 0.5122885987113158, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6475661054935324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05225776538414957, "1000–3000 Hz_40ms_envelope_correlation": 0.9982761548175572, "1000–3000 Hz_40ms_modulation_mae_db": 0.5355664296796742, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2897534055383748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15369262447211057, "3000–6000 Hz_10ms_envelope_correlation": 0.9810880315774555, "3000–6000 Hz_10ms_modulation_mae_db": 1.0918924052982713, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7425056826965898, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09785666986845037, "3000–6000 Hz_40ms_envelope_correlation": 0.9922539405708631, "3000–6000 Hz_40ms_modulation_mae_db": 1.3048538249709114, "mr_stft_db": 5.331580452294771, "band_mae_db": 0.1287795482267981, "rms_gain_error_db": -0.06859776019696336, "output_peak": 0.3333982347659171}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8690454380455479, "全帯域_10ms_rms_envelope_nrmse": 0.06245923552711302, "全帯域_10ms_envelope_correlation": 0.9950987094509417, "全帯域_10ms_modulation_mae_db": 0.5520760261350989, "全帯域_40ms_log_rms_mae_db": 0.4127386562853645, "全帯域_40ms_rms_envelope_nrmse": 0.033066502802542634, "全帯域_40ms_envelope_correlation": 0.9984822210536553, "全帯域_40ms_modulation_mae_db": 0.6769288217423091, "300–1000 Hz_10ms_log_rms_mae_db": 1.3096723619658581, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06407995282791243, "300–1000 Hz_10ms_envelope_correlation": 0.9964424373994117, "300–1000 Hz_10ms_modulation_mae_db": 0.2751672934515036, "300–1000 Hz_40ms_log_rms_mae_db": 0.7787840154205756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037797480473577484, "300–1000 Hz_40ms_envelope_correlation": 0.9986731340661069, "300–1000 Hz_40ms_modulation_mae_db": 0.3832654706597541, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1975912268034106, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08966393590176992, "1000–3000 Hz_10ms_envelope_correlation": 0.9936178243609591, "1000–3000 Hz_10ms_modulation_mae_db": 0.30163490059529646, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7085953070815281, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04914770939063631, "1000–3000 Hz_40ms_envelope_correlation": 0.9980269903567581, "1000–3000 Hz_40ms_modulation_mae_db": 0.3396102891209343, "3000–6000 Hz_10ms_log_rms_mae_db": 1.298708588758393, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16742743695310333, "3000–6000 Hz_10ms_envelope_correlation": 0.9776174067155714, "3000–6000 Hz_10ms_modulation_mae_db": 0.7913633848606002, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7467965603125184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10150372615950916, "3000–6000 Hz_40ms_envelope_correlation": 0.9912021015525759, "3000–6000 Hz_40ms_modulation_mae_db": 0.8247759500467371, "mr_stft_db": 5.344339167768214, "band_mae_db": 0.07673726644232685, "rms_gain_error_db": -0.02468109183825247, "output_peak": 0.30816959479199724}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8653891312094204, "全帯域_10ms_rms_envelope_nrmse": 0.06271942578622451, "全帯域_10ms_envelope_correlation": 0.9951402169098591, "全帯域_10ms_modulation_mae_db": 0.5038943554612697, "全帯域_40ms_log_rms_mae_db": 0.4414937673094301, "全帯域_40ms_rms_envelope_nrmse": 0.032212241407837176, "全帯域_40ms_envelope_correlation": 0.9986856063518238, "全帯域_40ms_modulation_mae_db": 0.4266159214819821, "300–1000 Hz_10ms_log_rms_mae_db": 1.3265639840803702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06068216196778924, "300–1000 Hz_10ms_envelope_correlation": 0.9968415934744815, "300–1000 Hz_10ms_modulation_mae_db": 0.33109260020675196, "300–1000 Hz_40ms_log_rms_mae_db": 0.709441983811733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03248395122944044, "300–1000 Hz_40ms_envelope_correlation": 0.9990648513304977, "300–1000 Hz_40ms_modulation_mae_db": 0.6988547283067194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.206765116539526, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08877718609988383, "1000–3000 Hz_10ms_envelope_correlation": 0.9939323708070946, "1000–3000 Hz_10ms_modulation_mae_db": 0.3256016225830584, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6809122493009531, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05061139046962392, "1000–3000 Hz_40ms_envelope_correlation": 0.998137479946656, "1000–3000 Hz_40ms_modulation_mae_db": 0.45174782621578463, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2662633527471592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1430461517438166, "3000–6000 Hz_10ms_envelope_correlation": 0.9832630249644536, "3000–6000 Hz_10ms_modulation_mae_db": 0.925562214907742, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7298944704963262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07402279311010154, "3000–6000 Hz_40ms_envelope_correlation": 0.9952934926171482, "3000–6000 Hz_40ms_modulation_mae_db": 0.9442394159770101, "mr_stft_db": 5.343642849496255, "band_mae_db": 0.1172340038993771, "rms_gain_error_db": -0.09231358547630825, "output_peak": 0.35783967716142506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.348186098022581, "spectral_convergence": 0.29979755004840625, "spectral_convergence_status": "ok", "amplitude_floor": 6.643709912602353e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.184014571544879, "spectral_convergence": 0.27363977565047387, "spectral_convergence_status": "ok", "amplitude_floor": 5.274693968404162e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.508612949697922, "spectral_convergence": 0.2798589673471376, "spectral_convergence_status": "ok", "amplitude_floor": 3.8895875894681354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00010232532114960487, "output_power": 0.0001032971088543776, "delta_db": 0.0410505001828783, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003592345857428397, "output_power": 0.0003650050636762232, "delta_db": 0.06920747186187998, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003905970862424533, "output_power": 0.0003865906887247884, "delta_db": -0.044776096666150546, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022506467673995083, "output_power": 0.0002185214906577473, "delta_db": -0.12813184647632325, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011237312839533447, "output_power": 0.00010657588595009523, "delta_db": -0.23003519948080414, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00013068329846048222, "output_power": 0.00012744963282408325, "delta_db": -0.10881498954332396, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10366935070189336, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.042436936998511386, "pitch": null, "output_rms_dbfs": -28.835782752554717, "output_peak": 0.3599799233858057, "reference_rms_dbfs": -28.793345815556208, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.346937873088461}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001036839322212551, "log_rms_mae_db": 0.8668413332253561, "rms_envelope_nrmse": 0.06327684740587114, "envelope_correlation": 0.9949816965573561, "rms_gain_error_db": -0.04244909893515203, "reference_mean_power": 0.0013213586135692834, "output_mean_power": 0.0013085062163775698}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.530367535346178e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002613603148630271, "output_power": 0.00025364302946328073, "reference_active": true, "delta_db": -0.13016712990821108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.246118337274924e-05, "output_power": 7.580530058289805e-05, "reference_active": true, "delta_db": -0.3654998877750207}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.028948119628933e-05, "output_power": 1.0561619155917523e-05, "reference_active": true, "delta_db": 0.11337025254708721}], "active_band_count": 3, "active_band_mae_db": 0.2030124234101063}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.673473491549852e-05, "log_rms_mae_db": 0.4261080880327152, "rms_envelope_nrmse": 0.033300550450813746, "envelope_correlation": 0.9984840738961662, "rms_gain_error_db": -0.042403712517649864, "reference_mean_power": 0.0013278763848253465, "output_mean_power": 0.001314974333742563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.825313121587468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002333069672738863, "output_power": 0.0002269936730664003, "reference_active": true, "delta_db": -0.11913955986972113}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6274669036315644e-05, "output_power": 4.2620611785320336e-05, "reference_active": true, "delta_db": -0.3572364191563398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.136183723659665e-07, "output_power": 6.166411281329314e-07, "reference_active": true, "delta_db": 0.021341331497665535}], "active_band_count": 3, "active_band_mae_db": 0.1659057701745755}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.601318795186366e-05, "log_rms_mae_db": 1.304406308106801, "rms_envelope_nrmse": 0.061322511806752245, "envelope_correlation": 0.9967593791769878, "rms_gain_error_db": -0.032032142346859555, "reference_mean_power": 0.0005515675704872272, "output_mean_power": 0.0005475143542331144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3413004486315466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020691279435977171, "output_power": 0.00020245105124972598, "reference_active": true, "delta_db": -0.09467309667443849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313846735404118e-05, "output_power": 2.883066490158762e-05, "reference_active": true, "delta_db": -0.6047776019938447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.722807599466502e-06, "output_power": 2.4197208984444573e-06, "reference_active": true, "delta_db": -0.5125167868890557}], "active_band_count": 3, "active_band_mae_db": 0.40398916185244627}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.977903967560393e-05, "log_rms_mae_db": 0.7190617905973719, "rms_envelope_nrmse": 0.03719963865336173, "envelope_correlation": 0.9987099190737779, "rms_gain_error_db": -0.03203306788556605, "reference_mean_power": 0.0005543343450028261, "output_mean_power": 0.0005502606797282113}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0568367997569596e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001948094737446988, "output_power": 0.0001904197406766446, "reference_active": true, "delta_db": -0.09898103776623716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.636770712067895e-05, "output_power": 1.4384219639115877e-05, "reference_active": true, "delta_db": -0.5610153941224281}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1494053604425814e-07, "output_power": 1.8769171357766715e-07, "reference_active": true, "delta_db": -0.5887322830831937}], "active_band_count": 3, "active_band_mae_db": 0.4162429049906196}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.251077329795781e-05, "log_rms_mae_db": 1.2169500660258117, "rms_envelope_nrmse": 0.0858889578320674, "envelope_correlation": 0.9942911522545248, "rms_gain_error_db": -0.1614763889600206, "reference_mean_power": 0.0002789441343631021, "output_mean_power": 0.0002687630708297895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.041685297045403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011680629747545264, "output_power": 0.00011161434046589355, "reference_active": true, "delta_db": -0.19746260553247463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1432581274087105e-05, "output_power": 1.0748607714127576e-05, "reference_active": true, "delta_db": -0.26792084483822093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6649698962790864e-06, "output_power": 1.7324712770634718e-06, "reference_active": true, "delta_db": 0.1725965751006201}], "active_band_count": 3, "active_band_mae_db": 0.21266000849043856}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.285244854844391e-05, "log_rms_mae_db": 0.6928431512578567, "rms_envelope_nrmse": 0.05055744575716196, "envelope_correlation": 0.9980817971688818, "rms_gain_error_db": -0.16142969870128338, "reference_mean_power": 0.00028031132153199156, "output_mean_power": 0.0002700832612180967}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9116341431814158e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010971300365261617, "output_power": 0.00010513373730188651, "reference_active": true, "delta_db": -0.18516002037088267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3756201199309846e-06, "output_power": 5.0385472712262715e-06, "reference_active": true, "delta_db": -0.28123233320598523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905227736918092e-08, "output_power": 9.852595955595963e-08, "reference_active": true, "delta_db": -0.02313791905778439}], "active_band_count": 3, "active_band_mae_db": 0.16317675754488412}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.076518340019586e-05, "log_rms_mae_db": 1.24858965426988, "rms_envelope_nrmse": 0.1471487368092421, "envelope_correlation": 0.9823861785732546, "rms_gain_error_db": -0.2213459542498905, "reference_mean_power": 9.153664281853568e-05, "output_mean_power": 8.698820811955626e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.939589252350226e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.46860955530461e-05, "output_power": 3.3108493446038034e-05, "reference_active": true, "delta_db": -0.20215996890887566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1944580647265704e-05, "output_power": 2.0456793929525115e-05, "reference_active": true, "delta_db": -0.3048971571213877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8366553780699585e-06, "output_power": 4.959367622596774e-06, "reference_active": true, "delta_db": 0.10881158302238525}], "active_band_count": 3, "active_band_mae_db": 0.20528956968421622}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4119030940985276e-05, "log_rms_mae_db": 0.7008496437792803, "rms_envelope_nrmse": 0.09151103044362255, "envelope_correlation": 0.9928306165745835, "rms_gain_error_db": -0.22146438672297591, "reference_mean_power": 9.196260234115854e-05, "output_mean_power": 8.739061862953966e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.164079440499503e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.035826325529754e-05, "output_power": 3.8585634159949146e-05, "reference_active": true, "delta_db": -0.19506826559899523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2217507839367818e-05, "output_power": 1.1526809388052394e-05, "reference_active": true, "delta_db": -0.25273514814121234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.5236785056789705e-07, "output_power": 4.47444092762327e-07, "reference_active": true, "delta_db": -0.04752953556868815}], "active_band_count": 3, "active_band_mae_db": 0.16511098310296526}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8575808025737137, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1d999dedb45fbc45df604164bbcb1a98be2921c5e362d9f3c4115aea3ccce90d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9936166262223322, "standard_deviation": 0.000829955726155115, "minimum": 0.9923422059652364, "maximum": 0.9947184817002444, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.194111174042407, "standard_deviation": 0.03290711582670257, "minimum": 1.1435978375605038, "maximum": 1.2385392210395334, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.39182183646219493, "standard_deviation": 0.11299286748776269, "minimum": 0.21266000849043856, "maximum": 0.5135082620305438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09091160068065174, "standard_deviation": 0.005880395913299717, "minimum": 0.08492642179290422, "maximum": 0.10071053868575852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9980574577883392, "standard_deviation": 0.00018300340048983178, "minimum": 0.9978391841890508, "maximum": 0.9983212164226685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6846923630135568, "standard_deviation": 0.021502172985297614, "minimum": 0.6475661054935324, "maximum": 0.7085953070815281, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4382690691034623, "standard_deviation": 0.13929728242616463, "minimum": 0.16317675754488412, "maximum": 0.5981587460394321, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05164129833233985, "standard_deviation": 0.003984812140698197, "minimum": 0.04553307116324269, "maximum": 0.05948872073029628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9795143389604245, "standard_deviation": 0.006080296627508301, "minimum": 0.965114503433384, "maximum": 0.9832630249644536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.26942156206028, "standard_deviation": 0.027934350817768096, "minimum": 1.2451463930170255, "maximum": 1.3148394557531724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7735652096286543, "standard_deviation": 0.25999534299204685, "minimum": 0.20528956968421622, "maximum": 1.0918924052982713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15787278352108308, "standard_deviation": 0.02077394442836179, "minimum": 0.1430461517438166, "maximum": 0.20567935989207592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9930108719705055, "standard_deviation": 0.0019872602579653767, "minimum": 0.9894283923039473, "maximum": 0.9952934926171482, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7340860402747698, "standard_deviation": 0.019886119002028056, "minimum": 0.7008496437792803, "maximum": 0.7615987170537002, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8583980461765286, "standard_deviation": 0.3479174378333369, "minimum": 0.16511098310296526, "maximum": 1.3048538249709114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0906548754710307, "standard_deviation": 0.011653170548675831, "minimum": 0.07402279311010154, "maximum": 0.10836040759984303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9961296655869034, "standard_deviation": 0.0006664536499650479, "minimum": 0.9948517129473161, "maximum": 0.9968415934744815, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3346907541418462, "standard_deviation": 0.03511267413976372, "minimum": 1.304406308106801, "maximum": 1.4044504646256575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3309026055754781, "standard_deviation": 0.0820718927438657, "minimum": 0.2417123177711281, "maximum": 0.4955653344437227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06680739002532642, "standard_deviation": 0.005668629117136136, "minimum": 0.06068216196778924, "maximum": 0.07762240554775529, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9985664102583762, "standard_deviation": 0.00040538075006948493, "minimum": 0.9976781580583997, "maximum": 0.9990648513304977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7569323839618747, "standard_deviation": 0.03293979869526665, "minimum": 0.709441983811733, "maximum": 0.7963384860060665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45351754190479954, "standard_deviation": 0.11393545077954279, "minimum": 0.3177691540168339, "maximum": 0.6988547283067194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.039240037425877436, "standard_deviation": 0.005342001308611795, "minimum": 0.03248395122944044, "maximum": 0.05090488612610975, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1233106792850977, "standard_deviation": 0.02431385946269201, "minimum": 0.07673726644232685, "maximum": 0.15087688238105038, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.345743305899315, "standard_deviation": 0.008119622087431161, "minimum": 5.331580452294771, "maximum": 5.36062008296646, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.34196487157699, "standard_deviation": 0.0192044877316175, "minimum": 0.30816959479199724, "maximum": 0.3661721203165848, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06767879897136192, "standard_deviation": 0.028270103727730145, "minimum": -0.11001410814576099, "maximum": -0.02468109183825247, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947189549930633, "standard_deviation": 0.0007573562986602863, "minimum": 0.993098678901809, "maximum": 0.995419853310957, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8686540060633321, "standard_deviation": 0.027277968809712207, "minimum": 0.8369256399520757, "maximum": 0.9292372152280047, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.47402590324429944, "standard_deviation": 0.22846664478606643, "minimum": 0.15148865793287225, "maximum": 0.7562001347556725, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06495458588073029, "standard_deviation": 0.004576154564572864, "minimum": 0.0606231388347518, "maximum": 0.07473996012991664, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983920088714203, "standard_deviation": 0.00024394347108314975, "minimum": 0.9978405731386543, "maximum": 0.9986856063518238, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4201811944335203, "standard_deviation": 0.021690077582668072, "minimum": 0.37973852336783237, "maximum": 0.44410088292651956, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5060100167187689, "standard_deviation": 0.2433873253617167, "minimum": 0.1659057701745755, "maximum": 0.7923433115636521, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03467889265603602, "standard_deviation": 0.0027819223428178826, "minimum": 0.032212241407837176, "maximum": 0.041124326099942, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135766-0005", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.22229266166687, "predicted_mos": 2.22229266166687, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "febe37349ecb7936ce14610838dd06f8715d5c1c7a81316d5245994d796a7081", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0872077997773886, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8814506200820587, "全帯域_10ms_rms_envelope_nrmse": 0.06462423600380894, "全帯域_10ms_envelope_correlation": 0.9947564991445956, "全帯域_10ms_modulation_mae_db": 0.18046996423783424, "全帯域_40ms_log_rms_mae_db": 0.4356718135546509, "全帯域_40ms_rms_envelope_nrmse": 0.034181726597080714, "全帯域_40ms_envelope_correlation": 0.9983895526318791, "全帯域_40ms_modulation_mae_db": 0.19071966674703145, "300–1000 Hz_10ms_log_rms_mae_db": 1.2931917786904084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06045779813427129, "300–1000 Hz_10ms_envelope_correlation": 0.9968357563556012, "300–1000 Hz_10ms_modulation_mae_db": 0.4205829255077595, "300–1000 Hz_40ms_log_rms_mae_db": 0.7002800119381619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03617390621448034, "300–1000 Hz_40ms_envelope_correlation": 0.9987696601312885, "300–1000 Hz_40ms_modulation_mae_db": 0.4600067249698408, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1856361789112073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08601964475322779, "1000–3000 Hz_10ms_envelope_correlation": 0.9942361287277982, "1000–3000 Hz_10ms_modulation_mae_db": 0.24574384965198848, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6530906079297767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049302107986482734, "1000–3000 Hz_40ms_envelope_correlation": 0.9981470138842135, "1000–3000 Hz_40ms_modulation_mae_db": 0.18643402774717566, "3000–6000 Hz_10ms_log_rms_mae_db": 1.259388002402592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15582415587569537, "3000–6000 Hz_10ms_envelope_correlation": 0.9801079954116818, "3000–6000 Hz_10ms_modulation_mae_db": 0.26652802170700823, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7172300651506583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09750549124448452, "3000–6000 Hz_40ms_envelope_correlation": 0.9916463387158051, "3000–6000 Hz_40ms_modulation_mae_db": 0.3101756103669355, "mr_stft_db": 5.3332919077728915, "band_mae_db": 0.08498505754402959, "rms_gain_error_db": -0.03129365701942786, "output_peak": 0.36465805514147837}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8666117011771589, "全帯域_10ms_rms_envelope_nrmse": 0.06627526601938666, "全帯域_10ms_envelope_correlation": 0.9945333485389241, "全帯域_10ms_modulation_mae_db": 0.7002734050639633, "全帯域_40ms_log_rms_mae_db": 0.41788223176164807, "全帯域_40ms_rms_envelope_nrmse": 0.03482379413581052, "全帯域_40ms_envelope_correlation": 0.9984068107935606, "全帯域_40ms_modulation_mae_db": 0.5179710387252169, "300–1000 Hz_10ms_log_rms_mae_db": 1.2898671207068788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06520435781821719, "300–1000 Hz_10ms_envelope_correlation": 0.9963395615703521, "300–1000 Hz_10ms_modulation_mae_db": 0.3809816753301785, "300–1000 Hz_40ms_log_rms_mae_db": 0.7049894336083379, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03640584242721368, "300–1000 Hz_40ms_envelope_correlation": 0.9988019473067782, "300–1000 Hz_40ms_modulation_mae_db": 0.5381831796239146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1675026024768753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08706907281909883, "1000–3000 Hz_10ms_envelope_correlation": 0.9939983185272688, "1000–3000 Hz_10ms_modulation_mae_db": 0.47211593178236605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6253372439833538, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04470692204185608, "1000–3000 Hz_40ms_envelope_correlation": 0.9983974676362877, "1000–3000 Hz_40ms_modulation_mae_db": 0.6128299617546321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.278913495506689, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15358446689949104, "3000–6000 Hz_10ms_envelope_correlation": 0.9807378723702512, "3000–6000 Hz_10ms_modulation_mae_db": 0.958125678145671, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7399553084698878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07869816372071635, "3000–6000 Hz_40ms_envelope_correlation": 0.9948133358081356, "3000–6000 Hz_40ms_modulation_mae_db": 1.1079736158935294, "mr_stft_db": 5.331402556464904, "band_mae_db": 0.13723037772983349, "rms_gain_error_db": -0.08247077521025463, "output_peak": 0.3659210238645526}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8856504810701847, "全帯域_10ms_rms_envelope_nrmse": 0.06106441320967743, "全帯域_10ms_envelope_correlation": 0.9953472031330496, "全帯域_10ms_modulation_mae_db": 0.5231479964239718, "全帯域_40ms_log_rms_mae_db": 0.4444114381144247, "全帯域_40ms_rms_envelope_nrmse": 0.03247519332837111, "全帯域_40ms_envelope_correlation": 0.9985886263639208, "全帯域_40ms_modulation_mae_db": 0.7503099952808796, "300–1000 Hz_10ms_log_rms_mae_db": 1.282537899931885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07174975510404882, "300–1000 Hz_10ms_envelope_correlation": 0.9955237906585332, "300–1000 Hz_10ms_modulation_mae_db": 0.33156928389133455, "300–1000 Hz_40ms_log_rms_mae_db": 0.6626724989120139, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04066158195492155, "300–1000 Hz_40ms_envelope_correlation": 0.998454139069074, "300–1000 Hz_40ms_modulation_mae_db": 0.49064552670301015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2517997174258384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0928796323627145, "1000–3000 Hz_10ms_envelope_correlation": 0.9934009856741887, "1000–3000 Hz_10ms_modulation_mae_db": 0.46144848678813255, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6939906095191264, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05417290960720697, "1000–3000 Hz_40ms_envelope_correlation": 0.9979119417046497, "1000–3000 Hz_40ms_modulation_mae_db": 0.6147647237366648, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2470363143414178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15272642275777487, "3000–6000 Hz_10ms_envelope_correlation": 0.9814925975878683, "3000–6000 Hz_10ms_modulation_mae_db": 0.7778729866899182, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7260789536495441, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08342809225675739, "3000–6000 Hz_40ms_envelope_correlation": 0.9949308384049462, "3000–6000 Hz_40ms_modulation_mae_db": 1.1198244417374967, "mr_stft_db": 5.348518401776288, "band_mae_db": 0.1473022904046011, "rms_gain_error_db": -0.06589475184018635, "output_peak": 0.33603032858159604}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9346038663459957, "全帯域_10ms_rms_envelope_nrmse": 0.07451157899830373, "全帯域_10ms_envelope_correlation": 0.9931444948263979, "全帯域_10ms_modulation_mae_db": 0.7480568257604676, "全帯域_40ms_log_rms_mae_db": 0.4419512148158622, "全帯域_40ms_rms_envelope_nrmse": 0.04122226226231404, "全帯域_40ms_envelope_correlation": 0.9978358188578476, "全帯域_40ms_modulation_mae_db": 0.7917798480164193, "300–1000 Hz_10ms_log_rms_mae_db": 1.3385213555765971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07828108986098553, "300–1000 Hz_10ms_envelope_correlation": 0.9947301241062521, "300–1000 Hz_10ms_modulation_mae_db": 0.5383355599788598, "300–1000 Hz_40ms_log_rms_mae_db": 0.7279916282496798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050027373969977175, "300–1000 Hz_40ms_envelope_correlation": 0.9977311786113023, "300–1000 Hz_40ms_modulation_mae_db": 0.5390591468907666, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2174938674574487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10172185112414445, "1000–3000 Hz_10ms_envelope_correlation": 0.9924247290369881, "1000–3000 Hz_10ms_modulation_mae_db": 0.38012109404950384, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6867854301456782, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06121830746333978, "1000–3000 Hz_40ms_envelope_correlation": 0.9977277375851453, "1000–3000 Hz_40ms_modulation_mae_db": 0.4364455662082056, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3330671692965508, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2046909007937542, "3000–6000 Hz_10ms_envelope_correlation": 0.9654876252272496, "3000–6000 Hz_10ms_modulation_mae_db": 0.9770014996828552, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7666086665140478, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10627340651171081, "3000–6000 Hz_40ms_envelope_correlation": 0.989926564996759, "3000–6000 Hz_40ms_modulation_mae_db": 0.9901673601646265, "mr_stft_db": 5.336431018100977, "band_mae_db": 0.1349250185702883, "rms_gain_error_db": -0.11334998953227673, "output_peak": 0.35405988436721775}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9075644078383597, "全帯域_10ms_rms_envelope_nrmse": 0.068418527156083, "全帯域_10ms_envelope_correlation": 0.9941110455880604, "全帯域_10ms_modulation_mae_db": 0.2044066575336231, "全帯域_40ms_log_rms_mae_db": 0.4309106916178398, "全帯域_40ms_rms_envelope_nrmse": 0.03374419110232034, "全帯域_40ms_envelope_correlation": 0.9984267151277936, "全帯域_40ms_modulation_mae_db": 0.25879183204401246, "300–1000 Hz_10ms_log_rms_mae_db": 1.4042775928571563, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07240342444870869, "300–1000 Hz_10ms_envelope_correlation": 0.9954608420098223, "300–1000 Hz_10ms_modulation_mae_db": 0.3012611576624709, "300–1000 Hz_40ms_log_rms_mae_db": 0.7470160121050125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041258842799300054, "300–1000 Hz_40ms_envelope_correlation": 0.9984011844321997, "300–1000 Hz_40ms_modulation_mae_db": 0.4008240801066276, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1794376347162685, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09703189204587069, "1000–3000 Hz_10ms_envelope_correlation": 0.992577153050734, "1000–3000 Hz_10ms_modulation_mae_db": 0.2990644917014189, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6598288330013129, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0504003585211783, "1000–3000 Hz_40ms_envelope_correlation": 0.9980223299972405, "1000–3000 Hz_40ms_modulation_mae_db": 0.36124959440146287, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2767636966608327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15408127518866876, "3000–6000 Hz_10ms_envelope_correlation": 0.9806154573984844, "3000–6000 Hz_10ms_modulation_mae_db": 0.9876461792745065, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7452478370265738, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08284765179149364, "3000–6000 Hz_40ms_envelope_correlation": 0.994204319722981, "3000–6000 Hz_40ms_modulation_mae_db": 1.027276621930375, "mr_stft_db": 5.332140112327663, "band_mae_db": 0.13095494172792574, "rms_gain_error_db": -0.042089176549162624, "output_peak": 0.33026312846238376}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8494660026788775, "全帯域_10ms_rms_envelope_nrmse": 0.06032326036865744, "全帯域_10ms_envelope_correlation": 0.9954480045909421, "全帯域_10ms_modulation_mae_db": 0.288558341976273, "全帯域_40ms_log_rms_mae_db": 0.3832075548695413, "全帯域_40ms_rms_envelope_nrmse": 0.03310080126727899, "全帯域_40ms_envelope_correlation": 0.9985125633407003, "全帯域_40ms_modulation_mae_db": 0.447957299792714, "300–1000 Hz_10ms_log_rms_mae_db": 1.349206381010911, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06746524365488146, "300–1000 Hz_10ms_envelope_correlation": 0.9960568740930221, "300–1000 Hz_10ms_modulation_mae_db": 0.3853962368667707, "300–1000 Hz_40ms_log_rms_mae_db": 0.7682341937008055, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038730122725123314, "300–1000 Hz_40ms_envelope_correlation": 0.998601481788297, "300–1000 Hz_40ms_modulation_mae_db": 0.47625372482137224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1615372369238406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08481074053585165, "1000–3000 Hz_10ms_envelope_correlation": 0.9946205624796232, "1000–3000 Hz_10ms_modulation_mae_db": 0.5310026536210821, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6405271121077993, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04949212215692084, "1000–3000 Hz_40ms_envelope_correlation": 0.9983914533889557, "1000–3000 Hz_40ms_modulation_mae_db": 0.5821350711868348, "3000–6000 Hz_10ms_log_rms_mae_db": 1.290842235613746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14940540885880144, "3000–6000 Hz_10ms_envelope_correlation": 0.9821628972164629, "3000–6000 Hz_10ms_modulation_mae_db": 1.229413034345139, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7369699918581851, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09059718896880943, "3000–6000 Hz_40ms_envelope_correlation": 0.993500574015865, "3000–6000 Hz_40ms_modulation_mae_db": 1.4271205815004475, "mr_stft_db": 5.31626177301546, "band_mae_db": 0.1410937087301617, "rms_gain_error_db": -0.05581454523771144, "output_peak": 0.3363936378804541}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.904842876062745, "全帯域_10ms_rms_envelope_nrmse": 0.06249445840967608, "全帯域_10ms_envelope_correlation": 0.9950927313088965, "全帯域_10ms_modulation_mae_db": 0.6073417417219082, "全帯域_40ms_log_rms_mae_db": 0.42565776282756124, "全帯域_40ms_rms_envelope_nrmse": 0.03270338542645514, "全帯域_40ms_envelope_correlation": 0.9985223021384911, "全帯域_40ms_modulation_mae_db": 0.7206475332615853, "300–1000 Hz_10ms_log_rms_mae_db": 1.2792425101810465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06556145313330083, "300–1000 Hz_10ms_envelope_correlation": 0.9962601356843691, "300–1000 Hz_10ms_modulation_mae_db": 0.3019865297505454, "300–1000 Hz_40ms_log_rms_mae_db": 0.6756493422135166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03784692624878285, "300–1000 Hz_40ms_envelope_correlation": 0.9986572183744951, "300–1000 Hz_40ms_modulation_mae_db": 0.4289088496613795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1783018355874557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09090322782654064, "1000–3000 Hz_10ms_envelope_correlation": 0.9934537509322208, "1000–3000 Hz_10ms_modulation_mae_db": 0.28119174491657245, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6575733678184817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049624025076309794, "1000–3000 Hz_40ms_envelope_correlation": 0.9980099923572122, "1000–3000 Hz_40ms_modulation_mae_db": 0.32732360512363995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.323916027699273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16238630321260988, "3000–6000 Hz_10ms_envelope_correlation": 0.9787818607085256, "3000–6000 Hz_10ms_modulation_mae_db": 0.8421606758585485, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7382327090697456, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09600956338695092, "3000–6000 Hz_40ms_envelope_correlation": 0.9920820514280531, "3000–6000 Hz_40ms_modulation_mae_db": 0.9654764245660538, "mr_stft_db": 5.329161214680204, "band_mae_db": 0.08403893379846532, "rms_gain_error_db": -0.03365027138343037, "output_peak": 0.3101348535159543}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8903997246568981, "全帯域_10ms_rms_envelope_nrmse": 0.06299227256483049, "全帯域_10ms_envelope_correlation": 0.9951212702485407, "全帯域_10ms_modulation_mae_db": 0.487326301809403, "全帯域_40ms_log_rms_mae_db": 0.45681359626543344, "全帯域_40ms_rms_envelope_nrmse": 0.03288501317102046, "全帯域_40ms_envelope_correlation": 0.9986555400763559, "全帯域_40ms_modulation_mae_db": 0.398979718641534, "300–1000 Hz_10ms_log_rms_mae_db": 1.3014805221993182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06279252038404673, "300–1000 Hz_10ms_envelope_correlation": 0.9966016267268641, "300–1000 Hz_10ms_modulation_mae_db": 0.47820855778344074, "300–1000 Hz_40ms_log_rms_mae_db": 0.6601683943445489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0332738049212547, "300–1000 Hz_40ms_envelope_correlation": 0.9990070781895339, "300–1000 Hz_40ms_modulation_mae_db": 0.8641680205979144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2033539268723636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08993898209649992, "1000–3000 Hz_10ms_envelope_correlation": 0.9938638363996722, "1000–3000 Hz_10ms_modulation_mae_db": 0.3632939175605534, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6382492380397555, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049691198345619296, "1000–3000 Hz_40ms_envelope_correlation": 0.9983431490925588, "1000–3000 Hz_40ms_modulation_mae_db": 0.47784699063778735, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2860560567910055, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1486299516253527, "3000–6000 Hz_10ms_envelope_correlation": 0.9819752713289205, "3000–6000 Hz_10ms_modulation_mae_db": 0.9615039113490883, "3000–6000 Hz_40ms_log_rms_mae_db": 0.723130159558163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07638888224009176, "3000–6000 Hz_40ms_envelope_correlation": 0.9951044967246527, "3000–6000 Hz_40ms_modulation_mae_db": 0.9134011969931212, "mr_stft_db": 5.325634295869096, "band_mae_db": 0.1310653125800386, "rms_gain_error_db": -0.10170905937593204, "output_peak": 0.353744375324284}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.341334749699073, "spectral_convergence": 0.3019614813598405, "spectral_convergence_status": "ok", "amplitude_floor": 6.643709912602353e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.167379036577325, "spectral_convergence": 0.27401864527540354, "spectral_convergence_status": "ok", "amplitude_floor": 5.274693968404162e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.491161937042274, "spectral_convergence": 0.2795627643326028, "spectral_convergence_status": "ok", "amplitude_floor": 3.8895875894681354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00010232532114960487, "output_power": 0.00010313065137896357, "delta_db": 0.034046444271892004, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003592345857428397, "output_power": 0.00036479495763902493, "delta_db": 0.0667068445002547, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003905970862424533, "output_power": 0.0003872766181015553, "delta_db": -0.037077220678220904, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022506467673995083, "output_power": 0.00021917871940967907, "delta_db": -0.11508953905845755, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011237312839533447, "output_power": 0.00010672658519870228, "delta_db": -0.2238985745618392, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00013068329846048222, "output_power": 0.00012969132144261206, "delta_db": -0.033091722193513216, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.08498505754402959, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03129365701942786, "pitch": null, "output_rms_dbfs": -28.82463947257563, "output_peak": 0.36465805514147837, "reference_rms_dbfs": -28.793345815556208, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.3332919077728915}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001036839322212551, "log_rms_mae_db": 0.8814506200820587, "rms_envelope_nrmse": 0.06462423600380894, "envelope_correlation": 0.9947564991445956, "rms_gain_error_db": -0.03130356429565875, "reference_mean_power": 0.0013213586135692834, "output_mean_power": 0.0013118686195037193}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.530367535346178e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002613603148630271, "output_power": 0.00025394149336310464, "reference_active": true, "delta_db": -0.12505975433426397}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.246118337274924e-05, "output_power": 7.544789817804138e-05, "reference_active": true, "delta_db": -0.3860241718769903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.028948119628933e-05, "output_power": 1.021788195809914e-05, "reference_active": true, "delta_db": -0.030325966502248336}], "active_band_count": 3, "active_band_mae_db": 0.18046996423783424}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.673473491549852e-05, "log_rms_mae_db": 0.4356718135546509, "rms_envelope_nrmse": 0.034181726597080714, "envelope_correlation": 0.9983895526318791, "rms_gain_error_db": -0.03125786984998576, "reference_mean_power": 0.0013278763848253465, "output_mean_power": 0.0013183534511728037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.825313121587468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002333069672738863, "output_power": 0.00022697194059573055, "reference_active": true, "delta_db": -0.11955537515674512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6274669036315644e-05, "output_power": 4.19103349295492e-05, "reference_active": true, "delta_db": -0.43022189749592177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.136183723659665e-07, "output_power": 6.104641737062825e-07, "reference_active": true, "delta_db": -0.022381727588427477}], "active_band_count": 3, "active_band_mae_db": 0.19071966674703145}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.601318795186366e-05, "log_rms_mae_db": 1.2931917786904084, "rms_envelope_nrmse": 0.06045779813427129, "envelope_correlation": 0.9968357563556012, "rms_gain_error_db": -0.026583551648935153, "reference_mean_power": 0.0005515675704872272, "output_mean_power": 0.0005482016882544254}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3413004486315466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020691279435977171, "output_power": 0.0002034982642223741, "reference_active": true, "delta_db": -0.07226636721764632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313846735404118e-05, "output_power": 2.8769744608660248e-05, "reference_active": true, "delta_db": -0.6139641189454665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.722807599466502e-06, "output_power": 2.384872276373957e-06, "reference_active": true, "delta_db": -0.5755182903601654}], "active_band_count": 3, "active_band_mae_db": 0.4205829255077595}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.977903967560393e-05, "log_rms_mae_db": 0.7002800119381619, "rms_envelope_nrmse": 0.03617390621448034, "envelope_correlation": 0.9987696601312885, "rms_gain_error_db": -0.026583171815022934, "reference_mean_power": 0.0005543343450028261, "output_mean_power": 0.000550951627010049}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0568367997569596e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001948094737446988, "output_power": 0.0001913977477067342, "reference_active": true, "delta_db": -0.07673250288898595}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.636770712067895e-05, "output_power": 1.434483889436471e-05, "reference_active": true, "delta_db": -0.5729217020376625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1494053604425814e-07, "output_power": 1.8166937857940005e-07, "reference_active": true, "delta_db": -0.7303659699828741}], "active_band_count": 3, "active_band_mae_db": 0.4600067249698408}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.251077329795781e-05, "log_rms_mae_db": 1.1856361789112073, "rms_envelope_nrmse": 0.08601964475322779, "envelope_correlation": 0.9942361287277982, "rms_gain_error_db": -0.1472683968411987, "reference_mean_power": 0.0002789441343631021, "output_mean_power": 0.00026964377202172006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.041685297045403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011680629747545264, "output_power": 0.00011280057185420984, "reference_active": true, "delta_db": -0.1515495653476938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1432581274087105e-05, "output_power": 1.0566008863198851e-05, "reference_active": true, "delta_db": -0.3423332680312225}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6649698962790864e-06, "output_power": 1.7609266134931697e-06, "reference_active": true, "delta_db": 0.24334871557704915}], "active_band_count": 3, "active_band_mae_db": 0.24574384965198848}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.285244854844391e-05, "log_rms_mae_db": 0.6530906079297767, "rms_envelope_nrmse": 0.049302107986482734, "envelope_correlation": 0.9981470138842135, "rms_gain_error_db": -0.14721203016248016, "reference_mean_power": 0.00028031132153199156, "output_mean_power": 0.00027096889223925065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9116341431814158e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010971300365261617, "output_power": 0.00010624246751872465, "reference_active": true, "delta_db": -0.13959956274338683}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3756201199309846e-06, "output_power": 4.934720591022783e-06, "reference_active": true, "delta_db": -0.3716600325170304}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905227736918092e-08, "output_power": 1.001540956730177e-07, "reference_active": true, "delta_db": 0.048042487981109765}], "active_band_count": 3, "active_band_mae_db": 0.18643402774717566}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.076518340019586e-05, "log_rms_mae_db": 1.259388002402592, "rms_envelope_nrmse": 0.15582415587569537, "envelope_correlation": 0.9801079954116818, "rms_gain_error_db": -0.19126272189270527, "reference_mean_power": 9.153664281853568e-05, "output_mean_power": 8.759286026969788e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.939589252350226e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.46860955530461e-05, "output_power": 3.403810912437253e-05, "reference_active": true, "delta_db": -0.08189989876564803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1944580647265704e-05, "output_power": 2.0423481305991845e-05, "reference_active": true, "delta_db": -0.31197513871381594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8366553780699585e-06, "output_power": 4.405287342944467e-06, "reference_active": true, "delta_db": -0.40570902764156064}], "active_band_count": 3, "active_band_mae_db": 0.26652802170700823}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4119030940985276e-05, "log_rms_mae_db": 0.7172300651506583, "rms_envelope_nrmse": 0.09750549124448452, "envelope_correlation": 0.9916463387158051, "rms_gain_error_db": -0.19128481388884902, "reference_mean_power": 9.196260234115854e-05, "output_mean_power": 8.800002002346986e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.164079440499503e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.035826325529754e-05, "output_power": 3.931808899423133e-05, "reference_active": true, "delta_db": -0.11340067549739105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2217507839367818e-05, "output_power": 1.1191264890873301e-05, "reference_active": true, "delta_db": -0.3810345100058645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.5236785056789705e-07, "output_power": 4.091500115939425e-07, "reference_active": true, "delta_db": -0.43609164559755076}], "active_band_count": 3, "active_band_mae_db": 0.3101756103669355}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8618613422360136, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "202e59bba5a7459d317c7462a3524c7fa7b62b8ef77772f9dcb5ab81aefef992", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9935719331035617, "standard_deviation": 0.000770588460901618, "minimum": 0.9924247290369881, "maximum": 0.9946205624796232, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1931328750464123, "standard_deviation": 0.02990136465803662, "minimum": 1.1615372369238406, "maximum": 1.2517997174258384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.37924777125895226, "standard_deviation": 0.10179385684585392, "minimum": 0.24574384965198848, "maximum": 0.5310026536210821, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09129688044549356, "standard_deviation": 0.005780112584843216, "minimum": 0.08481074053585165, "maximum": 0.10172185112414445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.998118885705783, "standard_deviation": 0.00024494598036605, "minimum": 0.9977277375851453, "maximum": 0.9983974676362877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6569228053181605, "standard_deviation": 0.023617550951842283, "minimum": 0.6253372439833538, "maximum": 0.6939906095191264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.44987869259955043, "standard_deviation": 0.1532844701820956, "minimum": 0.18643402774717566, "maximum": 0.6147647237366648, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.051075993899864225, "standard_deviation": 0.004827897559853391, "minimum": 0.04470692204185608, "maximum": 0.06121830746333978, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9789201971561805, "standard_deviation": 0.0055356980025783944, "minimum": 0.9654876252272496, "maximum": 0.9821628972164629, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2869978747890134, "standard_deviation": 0.029350889494821945, "minimum": 1.2470363143414178, "maximum": 1.3330671692965508, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8750314983815919, "standard_deviation": 0.27873312632016717, "minimum": 0.26652802170700823, "maximum": 1.229413034345139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16016611065151853, "standard_deviation": 0.01848090383027608, "minimum": 0.1486299516253527, "maximum": 0.2046909007937542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932760649771497, "standard_deviation": 0.0018768334541768327, "minimum": 0.989926564996759, "maximum": 0.9951044967246527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7366817114121007, "standard_deviation": 0.015383831302771652, "minimum": 0.7172300651506583, "maximum": 0.7666086665140478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9826769816440732, "standard_deviation": 0.31428765426734484, "minimum": 0.3101756103669355, "maximum": 1.4271205815004475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08896855501512685, "standard_deviation": 0.010395443378624754, "minimum": 0.07638888224009176, "maximum": 0.10627340651171081, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9959760889006021, "standard_deviation": 0.0006940695968194383, "minimum": 0.9947301241062521, "maximum": 0.9968357563556012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3172906451442752, "standard_deviation": 0.043546797901330536, "minimum": 1.2792425101810465, "maximum": 1.4042775928571563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.39229024084642006, "standard_deviation": 0.08435906946971883, "minimum": 0.3012611576624709, "maximum": 0.5383355599788598, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06798945531730757, "standard_deviation": 0.005821161797823611, "minimum": 0.06045779813427129, "maximum": 0.07828108986098553, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998552985987871, "standard_deviation": 0.00038480554890409584, "minimum": 0.9977311786113023, "maximum": 0.9990070781895339, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7058751893840096, "standard_deviation": 0.039590088170722515, "minimum": 0.6601683943445489, "maximum": 0.7682341937008055, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5247561566718533, "standard_deviation": 0.1453270447691123, "minimum": 0.4008240801066276, "maximum": 0.8641680205979144, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03929730015763171, "standard_deviation": 0.0050379599823293444, "minimum": 0.0332738049212547, "maximum": 0.050027373969977175, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12394945513566798, "standard_deviation": 0.0249196208028915, "minimum": 0.08403893379846532, "maximum": 0.1473022904046011, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.331605160000936, "standard_deviation": 0.009177321468721555, "minimum": 5.31626177301546, "maximum": 5.348518401776288, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.34390066089224014, "standard_deviation": 0.019124718114121083, "minimum": 0.3101348535159543, "maximum": 0.3659210238645526, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06578402826854776, "standard_deviation": 0.030971389895871643, "minimum": -0.11334998953227673, "maximum": -0.03129365701942786, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946943246724258, "standard_deviation": 0.0007659970252938311, "minimum": 0.9931444948263979, "maximum": 0.9954480045909421, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8900737099890348, "standard_deviation": 0.02618742411539376, "minimum": 0.8494660026788775, "maximum": 0.9346038663459957, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.46744765431593055, "standard_deviation": 0.22024840793930642, "minimum": 0.18046996423783424, "maximum": 0.7480568257604676, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06508800159130297, "standard_deviation": 0.004652671364720291, "minimum": 0.06032326036865744, "maximum": 0.07451157899830373, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984172411663186, "standard_deviation": 0.00025221886347090964, "minimum": 0.9978358188578476, "maximum": 0.9986555400763559, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4295632879783702, "standard_deviation": 0.022220297573924838, "minimum": 0.3832075548695413, "maximum": 0.45681359626543344, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5096446165636741, "standard_deviation": 0.227660135571916, "minimum": 0.19071966674703145, "maximum": 0.7917798480164193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03439204591133141, "standard_deviation": 0.002872693372012314, "minimum": 0.03247519332837111, "maximum": 0.04122226226231404, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135766-0005", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.6164815425872803, "predicted_mos": 2.6164815425872803, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "15432fbee7f2a7e5ebb9ba84387dd2ab5cb462a70f599662a3adc96a0a383a4c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0134062999859452, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8237989977485676, "全帯域_10ms_rms_envelope_nrmse": 0.057248212177766426, "全帯域_10ms_envelope_correlation": 0.9958858471912474, "全帯域_10ms_modulation_mae_db": 0.3400482427868167, "全帯域_40ms_log_rms_mae_db": 0.3540106790113106, "全帯域_40ms_rms_envelope_nrmse": 0.028177669730268817, "全帯域_40ms_envelope_correlation": 0.9988945017774368, "全帯域_40ms_modulation_mae_db": 0.5384713412035311, "300–1000 Hz_10ms_log_rms_mae_db": 1.5929284832150064, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06331795350986844, "300–1000 Hz_10ms_envelope_correlation": 0.9965918774042497, "300–1000 Hz_10ms_modulation_mae_db": 0.2613386214021258, "300–1000 Hz_40ms_log_rms_mae_db": 0.9784908467067293, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03523459373113759, "300–1000 Hz_40ms_envelope_correlation": 0.9988975283118634, "300–1000 Hz_40ms_modulation_mae_db": 0.48989128781000896, "1000–3000 Hz_10ms_log_rms_mae_db": 1.437265586410058, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0952436338606137, "1000–3000 Hz_10ms_envelope_correlation": 0.9927707495297099, "1000–3000 Hz_10ms_modulation_mae_db": 0.318394895735764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8145664165165583, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050397191049156374, "1000–3000 Hz_40ms_envelope_correlation": 0.9978437425888552, "1000–3000 Hz_40ms_modulation_mae_db": 0.3270914402671214, "3000–6000 Hz_10ms_log_rms_mae_db": 1.597645989923624, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14974581718546517, "3000–6000 Hz_10ms_envelope_correlation": 0.981593109466827, "3000–6000 Hz_10ms_modulation_mae_db": 0.515535530501343, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8897559559098742, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08314249477605491, "3000–6000 Hz_40ms_envelope_correlation": 0.9939104372993849, "3000–6000 Hz_40ms_modulation_mae_db": 0.7670072806258809, "mr_stft_db": 5.544858055990168, "band_mae_db": 0.07195268337255963, "rms_gain_error_db": 0.003017624414075021, "output_peak": 0.357581750423346}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8652121215923343, "全帯域_10ms_rms_envelope_nrmse": 0.0656722199808386, "全帯域_10ms_envelope_correlation": 0.9946156310626699, "全帯域_10ms_modulation_mae_db": 0.29011120561538295, "全帯域_40ms_log_rms_mae_db": 0.3829616193015592, "全帯域_40ms_rms_envelope_nrmse": 0.03603350845901783, "全帯域_40ms_envelope_correlation": 0.9982591573708298, "全帯域_40ms_modulation_mae_db": 0.30087217218183987, "300–1000 Hz_10ms_log_rms_mae_db": 1.6003982052421915, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07388485115643433, "300–1000 Hz_10ms_envelope_correlation": 0.9953353518441428, "300–1000 Hz_10ms_modulation_mae_db": 0.5148927067545315, "300–1000 Hz_40ms_log_rms_mae_db": 0.9534607511480817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046252727091539186, "300–1000 Hz_40ms_envelope_correlation": 0.9980553491928763, "300–1000 Hz_40ms_modulation_mae_db": 0.5650738141993591, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4622963197289034, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08522758601787901, "1000–3000 Hz_10ms_envelope_correlation": 0.9945657567314506, "1000–3000 Hz_10ms_modulation_mae_db": 0.3004308913777803, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8402741840918994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048079265562489225, "1000–3000 Hz_40ms_envelope_correlation": 0.998494919612959, "1000–3000 Hz_40ms_modulation_mae_db": 0.4061024327084759, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5765538955452292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18060502943513254, "3000–6000 Hz_10ms_envelope_correlation": 0.9731369612648433, "3000–6000 Hz_10ms_modulation_mae_db": 0.3640110537918559, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8788918927094005, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09988321209983811, "3000–6000 Hz_40ms_envelope_correlation": 0.9912217395817982, "3000–6000 Hz_40ms_modulation_mae_db": 0.5118825943000479, "mr_stft_db": 5.531415029038006, "band_mae_db": 0.12210458050498994, "rms_gain_error_db": -0.07275987321114463, "output_peak": 0.3414040177685822}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8665872888904119, "全帯域_10ms_rms_envelope_nrmse": 0.06498958627994832, "全帯域_10ms_envelope_correlation": 0.9946815130834735, "全帯域_10ms_modulation_mae_db": 0.5519405348800238, "全帯域_40ms_log_rms_mae_db": 0.4076769641859012, "全帯域_40ms_rms_envelope_nrmse": 0.030864410007976267, "全帯域_40ms_envelope_correlation": 0.9986647845577781, "全帯域_40ms_modulation_mae_db": 0.36883144841931376, "300–1000 Hz_10ms_log_rms_mae_db": 1.5379639319538898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07402592212887163, "300–1000 Hz_10ms_envelope_correlation": 0.995269142717936, "300–1000 Hz_10ms_modulation_mae_db": 0.23772861772729303, "300–1000 Hz_40ms_log_rms_mae_db": 0.9485934167697904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0428220361132492, "300–1000 Hz_40ms_envelope_correlation": 0.9983141158662594, "300–1000 Hz_40ms_modulation_mae_db": 0.3353361384158735, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4127456605084987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08804742286173803, "1000–3000 Hz_10ms_envelope_correlation": 0.9938156012576487, "1000–3000 Hz_10ms_modulation_mae_db": 0.40094556109147655, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8081736968022855, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04963646723903349, "1000–3000 Hz_40ms_envelope_correlation": 0.9979119345467504, "1000–3000 Hz_40ms_modulation_mae_db": 0.48855953177974065, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5562210541614931, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17051019304465442, "3000–6000 Hz_10ms_envelope_correlation": 0.9760496224223076, "3000–6000 Hz_10ms_modulation_mae_db": 1.01638935537769, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8309969109151625, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08669041914014249, "3000–6000 Hz_40ms_envelope_correlation": 0.9932935732439578, "3000–6000 Hz_40ms_modulation_mae_db": 1.2273530360912097, "mr_stft_db": 5.533392993802771, "band_mae_db": 0.09430379573718252, "rms_gain_error_db": -0.016571062979849713, "output_peak": 0.34646687667631787}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8655951752332155, "全帯域_10ms_rms_envelope_nrmse": 0.060727426321770264, "全帯域_10ms_envelope_correlation": 0.9954988382018248, "全帯域_10ms_modulation_mae_db": 0.38588606431358957, "全帯域_40ms_log_rms_mae_db": 0.40701411861780573, "全帯域_40ms_rms_envelope_nrmse": 0.03195703826126427, "全帯域_40ms_envelope_correlation": 0.9987646893540939, "全帯域_40ms_modulation_mae_db": 0.5734078357887517, "300–1000 Hz_10ms_log_rms_mae_db": 1.6782769286091614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07269652081851592, "300–1000 Hz_10ms_envelope_correlation": 0.9955822636403961, "300–1000 Hz_10ms_modulation_mae_db": 0.23173214672153694, "300–1000 Hz_40ms_log_rms_mae_db": 0.9846811293186608, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043530670012545965, "300–1000 Hz_40ms_envelope_correlation": 0.9984155042363091, "300–1000 Hz_40ms_modulation_mae_db": 0.436202971033323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4020026627185385, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08928269676537098, "1000–3000 Hz_10ms_envelope_correlation": 0.9938177831652949, "1000–3000 Hz_10ms_modulation_mae_db": 0.2843092580575886, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7741912243478906, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.046969421443159624, "1000–3000 Hz_40ms_envelope_correlation": 0.9983768154337129, "1000–3000 Hz_40ms_modulation_mae_db": 0.40938888193650363, "3000–6000 Hz_10ms_log_rms_mae_db": 1.593351797752556, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18696035788214985, "3000–6000 Hz_10ms_envelope_correlation": 0.9714298314638965, "3000–6000 Hz_10ms_modulation_mae_db": 1.2321784495906807, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8745302092059022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10428342901684855, "3000–6000 Hz_40ms_envelope_correlation": 0.9910385175578104, "3000–6000 Hz_40ms_modulation_mae_db": 1.527402724060395, "mr_stft_db": 5.5389886321116775, "band_mae_db": 0.17484886823903442, "rms_gain_error_db": -0.10329882444773766, "output_peak": 0.328635571123437}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8048098348750303, "全帯域_10ms_rms_envelope_nrmse": 0.06364065072467741, "全帯域_10ms_envelope_correlation": 0.9949094860949197, "全帯域_10ms_modulation_mae_db": 0.2212604360944097, "全帯域_40ms_log_rms_mae_db": 0.4014468899933977, "全帯域_40ms_rms_envelope_nrmse": 0.034721115059270995, "全帯域_40ms_envelope_correlation": 0.998332387781399, "全帯域_40ms_modulation_mae_db": 0.14383263151266068, "300–1000 Hz_10ms_log_rms_mae_db": 1.6052374077329008, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06939746146457791, "300–1000 Hz_10ms_envelope_correlation": 0.9958685737788987, "300–1000 Hz_10ms_modulation_mae_db": 0.34505304408828036, "300–1000 Hz_40ms_log_rms_mae_db": 0.9833934164917837, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040923353997682835, "300–1000 Hz_40ms_envelope_correlation": 0.9984588742032898, "300–1000 Hz_40ms_modulation_mae_db": 0.6120472611419986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.386214714774451, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09515757715939159, "1000–3000 Hz_10ms_envelope_correlation": 0.9930641460472647, "1000–3000 Hz_10ms_modulation_mae_db": 0.8669515549611582, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8111574309157913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05328221137631048, "1000–3000 Hz_40ms_envelope_correlation": 0.9980087032095084, "1000–3000 Hz_40ms_modulation_mae_db": 0.9933230151774665, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5783087392773032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17444062286684017, "3000–6000 Hz_10ms_envelope_correlation": 0.9768341180997184, "3000–6000 Hz_10ms_modulation_mae_db": 0.9529721200651965, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9000125897175909, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10599024937646044, "3000–6000 Hz_40ms_envelope_correlation": 0.9913962112456045, "3000–6000 Hz_40ms_modulation_mae_db": 1.0095477241464603, "mr_stft_db": 5.52757778207296, "band_mae_db": 0.04951512406320161, "rms_gain_error_db": -0.04250753525703529, "output_peak": 0.39784064243582457}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8260297233920401, "全帯域_10ms_rms_envelope_nrmse": 0.05927082886387931, "全帯域_10ms_envelope_correlation": 0.9956326492171264, "全帯域_10ms_modulation_mae_db": 0.38287464173567604, "全帯域_40ms_log_rms_mae_db": 0.39619982182801144, "全帯域_40ms_rms_envelope_nrmse": 0.02899206056035181, "全帯域_40ms_envelope_correlation": 0.9989031510940514, "全帯域_40ms_modulation_mae_db": 0.4263454767913501, "300–1000 Hz_10ms_log_rms_mae_db": 1.6224638548406218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07203667900813496, "300–1000 Hz_10ms_envelope_correlation": 0.9955774940454116, "300–1000 Hz_10ms_modulation_mae_db": 0.25657806560440616, "300–1000 Hz_40ms_log_rms_mae_db": 0.9538319133659943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041627112501861484, "300–1000 Hz_40ms_envelope_correlation": 0.9984686430092761, "300–1000 Hz_40ms_modulation_mae_db": 0.3173074831362623, "1000–3000 Hz_10ms_log_rms_mae_db": 1.468858484571965, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09019685264629057, "1000–3000 Hz_10ms_envelope_correlation": 0.9936543422141483, "1000–3000 Hz_10ms_modulation_mae_db": 0.5416552878964852, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8468372864564662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0509376669567713, "1000–3000 Hz_40ms_envelope_correlation": 0.9979979393237554, "1000–3000 Hz_40ms_modulation_mae_db": 0.38322048368080947, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5725202552474686, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1773292688270709, "3000–6000 Hz_10ms_envelope_correlation": 0.9741041512449048, "3000–6000 Hz_10ms_modulation_mae_db": 0.9163144621757829, "3000–6000 Hz_40ms_log_rms_mae_db": 0.879285534116012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10393093332467122, "3000–6000 Hz_40ms_envelope_correlation": 0.9903439910313871, "3000–6000 Hz_40ms_modulation_mae_db": 1.0003555561369823, "mr_stft_db": 5.5279990399247305, "band_mae_db": 0.11286463011481786, "rms_gain_error_db": -0.06584898773327157, "output_peak": 0.36236807881109123}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.875286901421498, "全帯域_10ms_rms_envelope_nrmse": 0.06730788151798744, "全帯域_10ms_envelope_correlation": 0.9942958800151527, "全帯域_10ms_modulation_mae_db": 0.3638828796723905, "全帯域_40ms_log_rms_mae_db": 0.42870375974122765, "全帯域_40ms_rms_envelope_nrmse": 0.03776267101059532, "全帯域_40ms_envelope_correlation": 0.9980061576742281, "全帯域_40ms_modulation_mae_db": 0.5350219307985212, "300–1000 Hz_10ms_log_rms_mae_db": 1.681051530314691, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0712581115282239, "300–1000 Hz_10ms_envelope_correlation": 0.9956727563730046, "300–1000 Hz_10ms_modulation_mae_db": 0.3901799428022705, "300–1000 Hz_40ms_log_rms_mae_db": 1.0527244506617688, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0404007135968624, "300–1000 Hz_40ms_envelope_correlation": 0.9985253467508931, "300–1000 Hz_40ms_modulation_mae_db": 0.8073813528632744, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4669685478798082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09839177732276524, "1000–3000 Hz_10ms_envelope_correlation": 0.9922815499220974, "1000–3000 Hz_10ms_modulation_mae_db": 0.2500192075473997, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8130745935411547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05536927724497455, "1000–3000 Hz_40ms_envelope_correlation": 0.9974325471415, "1000–3000 Hz_40ms_modulation_mae_db": 0.16749364608906345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.579608540627051, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17044466510729767, "3000–6000 Hz_10ms_envelope_correlation": 0.9765810762663789, "3000–6000 Hz_10ms_modulation_mae_db": 1.474857514748927, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8643320782244536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1069089344986377, "3000–6000 Hz_40ms_envelope_correlation": 0.9905971294279368, "3000–6000 Hz_40ms_modulation_mae_db": 1.7542902358949195, "mr_stft_db": 5.5239587996486454, "band_mae_db": 0.13373266490352753, "rms_gain_error_db": -0.02437533586438705, "output_peak": 0.3545876269896396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8238260993090328, "全帯域_10ms_rms_envelope_nrmse": 0.06318760848484915, "全帯域_10ms_envelope_correlation": 0.9950301110512532, "全帯域_10ms_modulation_mae_db": 0.6154344677224847, "全帯域_40ms_log_rms_mae_db": 0.4072194349680607, "全帯域_40ms_rms_envelope_nrmse": 0.03451635546921059, "全帯域_40ms_envelope_correlation": 0.9984196281253637, "全帯域_40ms_modulation_mae_db": 0.7109880815441932, "300–1000 Hz_10ms_log_rms_mae_db": 1.6081010339648287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07187530093824135, "300–1000 Hz_10ms_envelope_correlation": 0.9956398967836912, "300–1000 Hz_10ms_modulation_mae_db": 0.6481504773547726, "300–1000 Hz_40ms_log_rms_mae_db": 0.9701359642456154, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04383538888013, "300–1000 Hz_40ms_envelope_correlation": 0.9983099415211224, "300–1000 Hz_40ms_modulation_mae_db": 0.7200863019804942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4310804300036613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09212703872805575, "1000–3000 Hz_10ms_envelope_correlation": 0.9933212558861769, "1000–3000 Hz_10ms_modulation_mae_db": 0.4391704977540387, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8377404727887692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052017194214458884, "1000–3000 Hz_40ms_envelope_correlation": 0.9978382384808394, "1000–3000 Hz_40ms_modulation_mae_db": 0.32296684331459385, "3000–6000 Hz_10ms_log_rms_mae_db": 1.563247259033951, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16710694662716888, "3000–6000 Hz_10ms_envelope_correlation": 0.9770708341960119, "3000–6000 Hz_10ms_modulation_mae_db": 1.029207978208411, "3000–6000 Hz_40ms_log_rms_mae_db": 0.860982395760611, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08323072998559539, "3000–6000 Hz_40ms_envelope_correlation": 0.9937912855117008, "3000–6000 Hz_40ms_modulation_mae_db": 1.1269833214352731, "mr_stft_db": 5.52379481201008, "band_mae_db": 0.12776206531202267, "rms_gain_error_db": -0.07422789886682203, "output_peak": 0.35498756094996275}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.588478480676504, "spectral_convergence": 0.30924429376136553, "spectral_convergence_status": "ok", "amplitude_floor": 6.643709912602353e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.44644626696123, "spectral_convergence": 0.2822225685931095, "spectral_convergence_status": "ok", "amplitude_floor": 5.274693968404162e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.599649420332766, "spectral_convergence": 0.2860033491465031, "spectral_convergence_status": "ok", "amplitude_floor": 3.8895875894681354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00010232532114960487, "output_power": 0.00010642121173416646, "delta_db": 0.17045083225223312, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003592345857428397, "output_power": 0.00036429961877423314, "delta_db": 0.06080574536418207, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003905970862424533, "output_power": 0.0003844622457333552, "delta_db": -0.0687530080514005, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022506467673995083, "output_power": 0.00022596190741051156, "delta_db": 0.017278930197142033, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011237312839533447, "output_power": 0.00010989227398759239, "delta_db": -0.09695311229555992, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00013068329846048222, "output_power": 0.00013015853158965997, "delta_db": -0.01747447207484008, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.07195268337255963, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.003017624414075021, "pitch": null, "output_rms_dbfs": -28.79032819114213, "output_peak": 0.357581750423346, "reference_rms_dbfs": -28.793345815556208, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.544858055990168}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001036839322212551, "log_rms_mae_db": 0.8237989977485676, "rms_envelope_nrmse": 0.057248212177766426, "envelope_correlation": 0.9958858471912474, "rms_gain_error_db": 0.0030357468977427766, "reference_mean_power": 0.0013213586135692834, "output_mean_power": 0.0013222825747919043}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.530367535346178e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002613603148630271, "output_power": 0.0002617784579850431, "reference_active": true, "delta_db": 0.006942604854974665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.246118337274924e-05, "output_power": 8.573249628473143e-05, "reference_active": true, "delta_db": 0.16895906397898205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.028948119628933e-05, "output_power": 8.471672763974508e-06, "reference_active": true, "delta_db": -0.8442430595264934}], "active_band_count": 3, "active_band_mae_db": 0.3400482427868167}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.673473491549852e-05, "log_rms_mae_db": 0.3540106790113106, "rms_envelope_nrmse": 0.028177669730268817, "envelope_correlation": 0.9988945017774368, "rms_gain_error_db": 0.003145768453313403, "reference_mean_power": 0.0013278763848253465, "output_mean_power": 0.001328838567177218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.825313121587468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002333069672738863, "output_power": 0.00023475222414344796, "reference_active": true, "delta_db": 0.02682007274685117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6274669036315644e-05, "output_power": 4.735528648023701e-05, "reference_active": true, "delta_db": 0.10025146967095314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.136183723659665e-07, "output_power": 4.3557616437005307e-07, "reference_active": true, "delta_db": -1.488342481192789}], "active_band_count": 3, "active_band_mae_db": 0.5384713412035311}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.601318795186366e-05, "log_rms_mae_db": 1.5929284832150064, "rms_envelope_nrmse": 0.06331795350986844, "envelope_correlation": 0.9965918774042497, "rms_gain_error_db": -0.034853966893938174, "reference_mean_power": 0.0005515675704872272, "output_mean_power": 0.0005471587228148163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3413004486315466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020691279435977171, "output_power": 0.00019916209797649328, "reference_active": true, "delta_db": -0.16580653408370294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313846735404118e-05, "output_power": 2.9826116826784607e-05, "reference_active": true, "delta_db": -0.45735703908230535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.722807599466502e-06, "output_power": 2.6238060248263086e-06, "reference_active": true, "delta_db": -0.1608522910403691}], "active_band_count": 3, "active_band_mae_db": 0.2613386214021258}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.977903967560393e-05, "log_rms_mae_db": 0.9784908467067293, "rms_envelope_nrmse": 0.03523459373113759, "envelope_correlation": 0.9988975283118634, "rms_gain_error_db": -0.034871424907830226, "reference_mean_power": 0.0005543343450028261, "output_mean_power": 0.0005499011711304762}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0568367997569596e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001948094737446988, "output_power": 0.00018733431044892776, "reference_active": true, "delta_db": -0.16992747064917244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.636770712067895e-05, "output_power": 1.5126303604974297e-05, "reference_active": true, "delta_db": -0.3425503227590465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1494053604425814e-07, "output_power": 1.724243988976957e-07, "reference_active": true, "delta_db": -0.9571960700218078}], "active_band_count": 3, "active_band_mae_db": 0.48989128781000896}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.251077329795781e-05, "log_rms_mae_db": 1.437265586410058, "rms_envelope_nrmse": 0.0952436338606137, "envelope_correlation": 0.9927707495297099, "rms_gain_error_db": -0.0047708095033662905, "reference_mean_power": 0.0002789441343631021, "output_mean_power": 0.0002786378770420612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.041685297045403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011680629747545264, "output_power": 0.00011624567888135349, "reference_active": true, "delta_db": -0.020894397491224504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1432581274087105e-05, "output_power": 1.043744442729669e-05, "reference_active": true, "delta_db": -0.3955012130011456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6649698962790864e-06, "output_power": 1.470711464740619e-06, "reference_active": true, "delta_db": -0.5387890767149218}], "active_band_count": 3, "active_band_mae_db": 0.318394895735764}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.285244854844391e-05, "log_rms_mae_db": 0.8145664165165583, "rms_envelope_nrmse": 0.050397191049156374, "envelope_correlation": 0.9978437425888552, "rms_gain_error_db": -0.004647549269830518, "reference_mean_power": 0.00028031132153199156, "output_mean_power": 0.00028001151024866834}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9116341431814158e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010971300365261617, "output_power": 0.00010926513303752841, "reference_active": true, "delta_db": -0.017765061927949695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3756201199309846e-06, "output_power": 4.966413782893211e-06, "reference_active": true, "delta_db": -0.34385670244634936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905227736918092e-08, "output_power": 8.588141858271951e-08, "reference_active": true, "delta_db": -0.6196525564270651}], "active_band_count": 3, "active_band_mae_db": 0.3270914402671214}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.076518340019586e-05, "log_rms_mae_db": 1.597645989923624, "rms_envelope_nrmse": 0.14974581718546517, "envelope_correlation": 0.981593109466827, "rms_gain_error_db": -0.14525232676101515, "reference_mean_power": 9.153664281853568e-05, "output_mean_power": 8.852577706235977e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.939589252350226e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.46860955530461e-05, "output_power": 3.555560111728971e-05, "reference_active": true, "delta_db": 0.10752609243280918}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1944580647265704e-05, "output_power": 2.236806579510317e-05, "reference_active": true, "delta_db": 0.08301145478707439}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8366553780699585e-06, "output_power": 3.539470032550562e-06, "reference_active": true, "delta_db": -1.3560690442841454}], "active_band_count": 3, "active_band_mae_db": 0.515535530501343}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4119030940985276e-05, "log_rms_mae_db": 0.8897559559098742, "rms_envelope_nrmse": 0.08314249477605491, "envelope_correlation": 0.9939104372993849, "rms_gain_error_db": -0.14497141200007907, "reference_mean_power": 9.196260234115854e-05, "output_mean_power": 8.894347867281449e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.164079440499503e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.035826325529754e-05, "output_power": 4.1153746104254155e-05, "reference_active": true, "delta_db": 0.08476905003567506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2217507839367818e-05, "output_power": 1.2392104437585044e-05, "reference_active": true, "delta_db": 0.061624385273422955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.5236785056789705e-07, "output_power": 2.754411936279876e-07, "reference_active": true, "delta_db": -2.154628406568545}], "active_band_count": 3, "active_band_mae_db": 0.7670072806258809}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8506803656105089, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "73365c4e325979acd2523faa76152a3e62f7b638644ac17fa5bb7c421c4f6fd2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9934113980942239, "standard_deviation": 0.0007107131035715605, "minimum": 0.9922815499220974, "maximum": 0.9945657567314506, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4334290508244854, "standard_deviation": 0.031346703883331835, "minimum": 1.386214714774451, "maximum": 1.468858484571965, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4252346443027114, "standard_deviation": 0.20248558515705944, "minimum": 0.2500192075473997, "maximum": 0.8669515549611582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09170932317026312, "standard_deviation": 0.0043557843412763845, "minimum": 0.08522758601787901, "maximum": 0.09839177732276524, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997988105042235, "standard_deviation": 0.00033084843417038136, "minimum": 0.9974325471415, "maximum": 0.998494919612959, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8182519131826019, "standard_deviation": 0.023346892989091827, "minimum": 0.7741912243478906, "maximum": 0.8468372864564662, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4372682843692218, "standard_deviation": 0.24333383997146488, "minimum": 0.16749364608906345, "maximum": 0.9933230151774665, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05083608688579424, "standard_deviation": 0.0027287243276593568, "minimum": 0.046969421443159624, "maximum": 0.05536927724497455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9758499630531111, "standard_deviation": 0.0030697372960150557, "minimum": 0.9714298314638965, "maximum": 0.981593109466827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5771821914460844, "standard_deviation": 0.013851557514553265, "minimum": 1.5562210541614931, "maximum": 1.597645989923624, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9376833080574859, "standard_deviation": 0.35814081440703993, "minimum": 0.3640110537918559, "maximum": 1.474857514748927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17214286262197243, "standard_deviation": 0.011057528291140081, "minimum": 0.14974581718546517, "maximum": 0.18696035788214985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9919491106124476, "standard_deviation": 0.0014695090788554685, "minimum": 0.9903439910313871, "maximum": 0.9939104372993849, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8723484458198758, "standard_deviation": 0.020933312242773932, "minimum": 0.8309969109151625, "maximum": 0.9000125897175909, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.115602809086396, "standard_deviation": 0.3965286837023772, "minimum": 0.5118825943000479, "maximum": 1.7542902358949195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0967575502772811, "standard_deviation": 0.010527861701756767, "minimum": 0.08314249477605491, "maximum": 0.1069089344986377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9956921695734663, "standard_deviation": 0.00040973900136695755, "minimum": 0.995269142717936, "maximum": 0.9965918774042497, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6158026719841614, "standard_deviation": 0.046591159807049955, "minimum": 1.5379639319538898, "maximum": 1.681051530314691, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3607067028069021, "standard_deviation": 0.1511790756788352, "minimum": 0.23173214672153694, "maximum": 0.6481504773547726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07106160006910855, "standard_deviation": 0.0034603022567765746, "minimum": 0.06331795350986844, "maximum": 0.07402592212887163, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984306628864863, "standard_deviation": 0.0002385712642796484, "minimum": 0.9980553491928763, "maximum": 0.9988975283118634, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.978163986088553, "standard_deviation": 0.03331742808095635, "minimum": 0.9485934167697904, "maximum": 1.0527244506617688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5354158263225742, "standard_deviation": 0.17498539138838118, "minimum": 0.3173074831362623, "maximum": 0.8073813528632744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04182832449062608, "standard_deviation": 0.0032483821237580565, "minimum": 0.03523459373113759, "maximum": 0.046252727091539186, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11088555153091703, "standard_deviation": 0.038886186989615226, "minimum": 0.04951512406320161, "maximum": 0.17484886823903442, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.53149814307488, "standard_deviation": 0.007387131942699356, "minimum": 5.52379481201008, "maximum": 5.544858055990168, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.35548401564727516, "standard_deviation": 0.02016828747764243, "minimum": 0.328635571123437, "maximum": 0.39784064243582457, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.049571486743271616, "standard_deviation": 0.03552989041773859, "minimum": -0.10329882444773766, "maximum": 0.003017624414075021, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9950687444897084, "standard_deviation": 0.0005542125923085796, "minimum": 0.9942958800151527, "maximum": 0.9958858471912474, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8438932678077663, "standard_deviation": 0.026936796465144137, "minimum": 0.8048098348750303, "maximum": 0.875286901421498, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3939298091025968, "standard_deviation": 0.1301744312005269, "minimum": 0.2212604360944097, "maximum": 0.6154344677224847, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06275555179396461, "standard_deviation": 0.0034179989737724756, "minimum": 0.057248212177766426, "maximum": 0.06730788151798744, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985305572168975, "standard_deviation": 0.0003260906583974312, "minimum": 0.9980061576742281, "maximum": 0.9989031510940514, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3981541609559093, "standard_deviation": 0.02196998483659168, "minimum": 0.3540106790113106, "maximum": 0.42870375974122765, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4497213647800202, "standard_deviation": 0.17789442907149644, "minimum": 0.14383263151266068, "maximum": 0.7109880815441932, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03287810356974449, "standard_deviation": 0.003422506837088682, "minimum": 0.028177669730268817, "maximum": 0.03776267101059532, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135766-0005", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.3081142902374268, "predicted_mos": 2.3081142902374268, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "386c88c2f01257043685543716fb21f74afba44bc5b31d5571d31c6336e0de11", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9130357997491956, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8422648623817953, "全帯域_10ms_rms_envelope_nrmse": 0.0689659449282972, "全帯域_10ms_envelope_correlation": 0.9940130110634301, "全帯域_10ms_modulation_mae_db": 0.849405968722571, "全帯域_40ms_log_rms_mae_db": 0.4270059442457008, "全帯域_40ms_rms_envelope_nrmse": 0.03962145749373768, "全帯域_40ms_envelope_correlation": 0.9978145116361702, "全帯域_40ms_modulation_mae_db": 0.8339597494865784, "300–1000 Hz_10ms_log_rms_mae_db": 1.2208907343050857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06818055130044423, "300–1000 Hz_10ms_envelope_correlation": 0.9959414983144401, "300–1000 Hz_10ms_modulation_mae_db": 0.26952798892666485, "300–1000 Hz_40ms_log_rms_mae_db": 0.6710394760426535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03728589188633603, "300–1000 Hz_40ms_envelope_correlation": 0.998678936294835, "300–1000 Hz_40ms_modulation_mae_db": 0.3141011772521669, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1141290662896024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09181048803555934, "1000–3000 Hz_10ms_envelope_correlation": 0.9936110086060926, "1000–3000 Hz_10ms_modulation_mae_db": 0.32747526160609014, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6116607036949706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054278916281864695, "1000–3000 Hz_40ms_envelope_correlation": 0.9979648514951339, "1000–3000 Hz_40ms_modulation_mae_db": 0.28300748797591635, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1509332401428238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1689753625619916, "3000–6000 Hz_10ms_envelope_correlation": 0.976577371883264, "3000–6000 Hz_10ms_modulation_mae_db": 1.1742292737709705, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6592995623256603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09400039332371032, "3000–6000 Hz_40ms_envelope_correlation": 0.992321706895002, "3000–6000 Hz_40ms_modulation_mae_db": 1.287342959806061, "mr_stft_db": 5.273544078334264, "band_mae_db": 0.11765978063985112, "rms_gain_error_db": -0.03998269300931331, "output_peak": 0.3825321116791075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8988843989028101, "全帯域_10ms_rms_envelope_nrmse": 0.0629432231361113, "全帯域_10ms_envelope_correlation": 0.9950104176522911, "全帯域_10ms_modulation_mae_db": 0.2969820612545069, "全帯域_40ms_log_rms_mae_db": 0.45671968697127274, "全帯域_40ms_rms_envelope_nrmse": 0.031174633404213885, "全帯域_40ms_envelope_correlation": 0.9986377570260235, "全帯域_40ms_modulation_mae_db": 0.2967419778288446, "300–1000 Hz_10ms_log_rms_mae_db": 1.2207799534562267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07141641289866676, "300–1000 Hz_10ms_envelope_correlation": 0.9955577257152217, "300–1000 Hz_10ms_modulation_mae_db": 0.1913750925554356, "300–1000 Hz_40ms_log_rms_mae_db": 0.6617174135868176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03976999613884207, "300–1000 Hz_40ms_envelope_correlation": 0.9985126974716001, "300–1000 Hz_40ms_modulation_mae_db": 0.30449717015221583, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1053084197122438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0793072221650351, "1000–3000 Hz_10ms_envelope_correlation": 0.9949919165744385, "1000–3000 Hz_10ms_modulation_mae_db": 0.6276668070412513, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6122445022890401, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.041241528121262395, "1000–3000 Hz_40ms_envelope_correlation": 0.9985599103405963, "1000–3000 Hz_40ms_modulation_mae_db": 0.6241347183754563, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1213517375707533, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15095798923871304, "3000–6000 Hz_10ms_envelope_correlation": 0.9813661906655128, "3000–6000 Hz_10ms_modulation_mae_db": 0.9413680074446953, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6372587822536849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07738194151356097, "3000–6000 Hz_40ms_envelope_correlation": 0.9949005378287304, "3000–6000 Hz_40ms_modulation_mae_db": 1.1414611151957652, "mr_stft_db": 5.275407894770269, "band_mae_db": 0.05936370833178648, "rms_gain_error_db": -0.017090887084306977, "output_peak": 0.3357857279376851}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.882825502660519, "全帯域_10ms_rms_envelope_nrmse": 0.06993576578272195, "全帯域_10ms_envelope_correlation": 0.9938369906944025, "全帯域_10ms_modulation_mae_db": 0.7082053622490724, "全帯域_40ms_log_rms_mae_db": 0.4351959800004273, "全帯域_40ms_rms_envelope_nrmse": 0.039115230441130654, "全帯域_40ms_envelope_correlation": 0.9978561854024145, "全帯域_40ms_modulation_mae_db": 0.6833145590695504, "300–1000 Hz_10ms_log_rms_mae_db": 1.2241582930798798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06436641018770808, "300–1000 Hz_10ms_envelope_correlation": 0.9963989118274514, "300–1000 Hz_10ms_modulation_mae_db": 0.20207955184851176, "300–1000 Hz_40ms_log_rms_mae_db": 0.6416902682523292, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03592457756767977, "300–1000 Hz_40ms_envelope_correlation": 0.9987833887558819, "300–1000 Hz_40ms_modulation_mae_db": 0.44308133518044795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1308609880789513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09024053485967447, "1000–3000 Hz_10ms_envelope_correlation": 0.9938075286213861, "1000–3000 Hz_10ms_modulation_mae_db": 0.3755421467556039, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6672768997095322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05141875663620127, "1000–3000 Hz_40ms_envelope_correlation": 0.9981809278796959, "1000–3000 Hz_40ms_modulation_mae_db": 0.48270901311782016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.153680859281619, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17834584815219076, "3000–6000 Hz_10ms_envelope_correlation": 0.9738959501475531, "3000–6000 Hz_10ms_modulation_mae_db": 0.811731910297988, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7088421357740269, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10945170187733458, "3000–6000 Hz_40ms_envelope_correlation": 0.9892741860874873, "3000–6000 Hz_40ms_modulation_mae_db": 0.9859696396687524, "mr_stft_db": 5.273920375770335, "band_mae_db": 0.1302832573023391, "rms_gain_error_db": -0.025097555017854857, "output_peak": 0.33222462408125236}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8971249250406486, "全帯域_10ms_rms_envelope_nrmse": 0.06545032703004837, "全帯域_10ms_envelope_correlation": 0.9946190371577295, "全帯域_10ms_modulation_mae_db": 0.43469485707083605, "全帯域_40ms_log_rms_mae_db": 0.4296292461649874, "全帯域_40ms_rms_envelope_nrmse": 0.033896567212197116, "全帯域_40ms_envelope_correlation": 0.9984191216385252, "全帯域_40ms_modulation_mae_db": 0.5238929041487506, "300–1000 Hz_10ms_log_rms_mae_db": 1.2247449242197133, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06621769449382593, "300–1000 Hz_10ms_envelope_correlation": 0.9961759353946582, "300–1000 Hz_10ms_modulation_mae_db": 0.43084559259617966, "300–1000 Hz_40ms_log_rms_mae_db": 0.7272050835730577, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03450850866941172, "300–1000 Hz_40ms_envelope_correlation": 0.9988860633314889, "300–1000 Hz_40ms_modulation_mae_db": 0.18535287873920991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.111822813614356, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08203773859015093, "1000–3000 Hz_10ms_envelope_correlation": 0.9947472716162421, "1000–3000 Hz_10ms_modulation_mae_db": 0.16975205197122342, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6276978219081939, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04396491658394052, "1000–3000 Hz_40ms_envelope_correlation": 0.9985391005822817, "1000–3000 Hz_40ms_modulation_mae_db": 0.20806067402866313, "3000–6000 Hz_10ms_log_rms_mae_db": 1.149279033032625, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.144215120600916, "3000–6000 Hz_10ms_envelope_correlation": 0.9829940854894436, "3000–6000 Hz_10ms_modulation_mae_db": 0.6424843267288273, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6702406944829238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07523802867902442, "3000–6000 Hz_40ms_envelope_correlation": 0.9951001274861936, "3000–6000 Hz_40ms_modulation_mae_db": 0.731598103410624, "mr_stft_db": 5.273282718590352, "band_mae_db": 0.09057399429539509, "rms_gain_error_db": -0.04713029645529502, "output_peak": 0.36968054343053175}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9269934380077623, "全帯域_10ms_rms_envelope_nrmse": 0.07191934814828893, "全帯域_10ms_envelope_correlation": 0.9935363831400292, "全帯域_10ms_modulation_mae_db": 0.2679677732452675, "全帯域_40ms_log_rms_mae_db": 0.4900658565640334, "全帯域_40ms_rms_envelope_nrmse": 0.03953378527508143, "全帯域_40ms_envelope_correlation": 0.9978741157772507, "全帯域_40ms_modulation_mae_db": 0.3491215749603825, "300–1000 Hz_10ms_log_rms_mae_db": 1.2220740379755295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06022412255998135, "300–1000 Hz_10ms_envelope_correlation": 0.9968908793315264, "300–1000 Hz_10ms_modulation_mae_db": 0.5830535454898124, "300–1000 Hz_40ms_log_rms_mae_db": 0.6391105750069384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03092126805702094, "300–1000 Hz_40ms_envelope_correlation": 0.9991498365061526, "300–1000 Hz_40ms_modulation_mae_db": 0.6986618317283769, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1032397071881932, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09090422365466298, "1000–3000 Hz_10ms_envelope_correlation": 0.993504168263562, "1000–3000 Hz_10ms_modulation_mae_db": 0.22747029803850896, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6562979406120334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055317385459856004, "1000–3000 Hz_40ms_envelope_correlation": 0.9975411385241888, "1000–3000 Hz_40ms_modulation_mae_db": 0.0761433097232412, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2159046575864327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16009222466240491, "3000–6000 Hz_10ms_envelope_correlation": 0.9791832066750789, "3000–6000 Hz_10ms_modulation_mae_db": 0.5178774401967352, "3000–6000 Hz_40ms_log_rms_mae_db": 0.724610217382445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08912518734517696, "3000–6000 Hz_40ms_envelope_correlation": 0.9933818138366112, "3000–6000 Hz_40ms_modulation_mae_db": 0.7387644013617645, "mr_stft_db": 5.280420021452064, "band_mae_db": 0.15241375257287132, "rms_gain_error_db": -0.05711365916434191, "output_peak": 0.35917337187339377}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8642657994604147, "全帯域_10ms_rms_envelope_nrmse": 0.07018814895229854, "全帯域_10ms_envelope_correlation": 0.9938327030267659, "全帯域_10ms_modulation_mae_db": 0.5045893837488434, "全帯域_40ms_log_rms_mae_db": 0.4314441109039298, "全帯域_40ms_rms_envelope_nrmse": 0.03781655870783662, "全帯域_40ms_envelope_correlation": 0.9980644834121601, "全帯域_40ms_modulation_mae_db": 0.4169302746516088, "300–1000 Hz_10ms_log_rms_mae_db": 1.2124822524344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06863686487281186, "300–1000 Hz_10ms_envelope_correlation": 0.995906897917956, "300–1000 Hz_10ms_modulation_mae_db": 0.5409592375607503, "300–1000 Hz_40ms_log_rms_mae_db": 0.6839032268837184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04038905913019737, "300–1000 Hz_40ms_envelope_correlation": 0.9984836045468335, "300–1000 Hz_40ms_modulation_mae_db": 0.47773130338281034, "1000–3000 Hz_10ms_log_rms_mae_db": 1.088924946067642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08824454263814543, "1000–3000 Hz_10ms_envelope_correlation": 0.9943220147582587, "1000–3000 Hz_10ms_modulation_mae_db": 0.3421304257497249, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6430759831825529, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057294447393060124, "1000–3000 Hz_40ms_envelope_correlation": 0.997883812738943, "1000–3000 Hz_40ms_modulation_mae_db": 0.24633095228482102, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1544742802991568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15816525337981607, "3000–6000 Hz_10ms_envelope_correlation": 0.9797128395512865, "3000–6000 Hz_10ms_modulation_mae_db": 0.7172445299703109, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6710149363719156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.078880553337284, "3000–6000 Hz_40ms_envelope_correlation": 0.9950469299999817, "3000–6000 Hz_40ms_modulation_mae_db": 0.693628720690473, "mr_stft_db": 5.267751315247881, "band_mae_db": 0.13511532087259262, "rms_gain_error_db": -0.06616699900149735, "output_peak": 0.3469357774191718}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8664328041600482, "全帯域_10ms_rms_envelope_nrmse": 0.06793231891286604, "全帯域_10ms_envelope_correlation": 0.9942021280578568, "全帯域_10ms_modulation_mae_db": 0.4003333928085664, "全帯域_40ms_log_rms_mae_db": 0.42797561120411676, "全帯域_40ms_rms_envelope_nrmse": 0.037112101772919764, "全帯域_40ms_envelope_correlation": 0.9980752422929531, "全帯域_40ms_modulation_mae_db": 0.3543058673427151, "300–1000 Hz_10ms_log_rms_mae_db": 1.2392094342519793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0684180366857296, "300–1000 Hz_10ms_envelope_correlation": 0.9959002721090889, "300–1000 Hz_10ms_modulation_mae_db": 0.23786684369341551, "300–1000 Hz_40ms_log_rms_mae_db": 0.6887442262692772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03549054414926609, "300–1000 Hz_40ms_envelope_correlation": 0.9988025563960651, "300–1000 Hz_40ms_modulation_mae_db": 0.22990919656871264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0942351539351673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08481526313669689, "1000–3000 Hz_10ms_envelope_correlation": 0.9946340345125361, "1000–3000 Hz_10ms_modulation_mae_db": 0.41131863749575653, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6155451545795017, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0518051548136576, "1000–3000 Hz_40ms_envelope_correlation": 0.9982082985337835, "1000–3000 Hz_40ms_modulation_mae_db": 0.45981863677243595, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1307372605308184, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1385612404660023, "3000–6000 Hz_10ms_envelope_correlation": 0.9844024725475938, "3000–6000 Hz_10ms_modulation_mae_db": 0.42270376578933005, "3000–6000 Hz_40ms_log_rms_mae_db": 0.664152601670033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08156515965509183, "3000–6000 Hz_40ms_envelope_correlation": 0.9942192416328002, "3000–6000 Hz_40ms_modulation_mae_db": 0.46171242625187797, "mr_stft_db": 5.271398898582522, "band_mae_db": 0.1479009706710837, "rms_gain_error_db": -0.025643795817258196, "output_peak": 0.38376308367838535}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9102584108438738, "全帯域_10ms_rms_envelope_nrmse": 0.06852493011634005, "全帯域_10ms_envelope_correlation": 0.9941736376841779, "全帯域_10ms_modulation_mae_db": 0.23321901108176668, "全帯域_40ms_log_rms_mae_db": 0.45738663832761706, "全帯域_40ms_rms_envelope_nrmse": 0.035025150430683605, "全帯域_40ms_envelope_correlation": 0.9984219869684204, "全帯域_40ms_modulation_mae_db": 0.21072715837635383, "300–1000 Hz_10ms_log_rms_mae_db": 1.261751826110526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06441884357861422, "300–1000 Hz_10ms_envelope_correlation": 0.9963944697135638, "300–1000 Hz_10ms_modulation_mae_db": 0.41778452257382703, "300–1000 Hz_40ms_log_rms_mae_db": 0.7181400100543245, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03816635816246652, "300–1000 Hz_40ms_envelope_correlation": 0.9986375533270797, "300–1000 Hz_40ms_modulation_mae_db": 0.4050807731322556, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0701440258483115, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08464287108033584, "1000–3000 Hz_10ms_envelope_correlation": 0.994804778170063, "1000–3000 Hz_10ms_modulation_mae_db": 0.37060187903887165, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6223484585966108, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04970335324741682, "1000–3000 Hz_40ms_envelope_correlation": 0.9985604553025129, "1000–3000 Hz_40ms_modulation_mae_db": 0.34706652836761204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1932141460726304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16633982865145727, "3000–6000 Hz_10ms_envelope_correlation": 0.9784909294472526, "3000–6000 Hz_10ms_modulation_mae_db": 0.8905973598408911, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7326769484360789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09961121151346283, "3000–6000 Hz_40ms_envelope_correlation": 0.993157145074947, "3000–6000 Hz_40ms_modulation_mae_db": 1.1171438759949792, "mr_stft_db": 5.283012307729628, "band_mae_db": 0.1611172359498881, "rms_gain_error_db": -0.09311814290879912, "output_peak": 0.36100033974590656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.2410879081355715, "spectral_convergence": 0.2998600690090546, "spectral_convergence_status": "ok", "amplitude_floor": 6.643709912602353e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.07958969181311, "spectral_convergence": 0.2696200439107354, "spectral_convergence_status": "ok", "amplitude_floor": 5.274693968404162e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.499954635054109, "spectral_convergence": 0.2798468273762917, "spectral_convergence_status": "ok", "amplitude_floor": 3.8895875894681354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00010232532114960487, "output_power": 0.00010415416262784708, "delta_db": 0.07693515295253443, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003592345857428397, "output_power": 0.00036536670493890163, "delta_db": 0.0735082628944512, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003905970862424533, "output_power": 0.0003914066699184834, "delta_db": 0.008992229154418626, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022506467673995083, "output_power": 0.00021247323074018225, "delta_db": -0.2500311749991687, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011237312839533447, "output_power": 0.00010938846669580432, "delta_db": -0.1169093666461056, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00013068329846048222, "output_power": 0.00012538969265149554, "delta_db": -0.17958249719242808, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11765978063985112, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03998269300931331, "pitch": null, "output_rms_dbfs": -28.833328508565515, "output_peak": 0.3825321116791075, "reference_rms_dbfs": -28.793345815556208, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.273544078334264}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001036839322212551, "log_rms_mae_db": 0.8422648623817953, "rms_envelope_nrmse": 0.0689659449282972, "envelope_correlation": 0.9940130110634301, "rms_gain_error_db": -0.03998685555632497, "reference_mean_power": 0.0013213586135692834, "output_mean_power": 0.0013092482875747275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.530367535346178e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002613603148630271, "output_power": 0.0002585624445544336, "reference_active": true, "delta_db": -0.046741995555704546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.246118337274924e-05, "output_power": 7.209210391258129e-05, "reference_active": true, "delta_db": -0.5836186290649151}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.028948119628933e-05, "output_power": 6.616186665884567e-06, "reference_active": true, "delta_db": -1.9178572815470933}], "active_band_count": 3, "active_band_mae_db": 0.849405968722571}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.673473491549852e-05, "log_rms_mae_db": 0.4270059442457008, "rms_envelope_nrmse": 0.03962145749373768, "envelope_correlation": 0.9978145116361702, "rms_gain_error_db": -0.03998893099004559, "reference_mean_power": 0.0013278763848253465, "output_mean_power": 0.0013157056943328856}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.825313121587468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002333069672738863, "output_power": 0.0002303162785680501, "reference_active": true, "delta_db": -0.05603073744175845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6274669036315644e-05, "output_power": 3.911856530853146e-05, "reference_active": true, "delta_db": -0.7296040283713323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.136183723659665e-07, "output_power": 4.1330808994966395e-07, "reference_active": true, "delta_db": -1.7162444826466445}], "active_band_count": 3, "active_band_mae_db": 0.8339597494865784}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.601318795186366e-05, "log_rms_mae_db": 1.2208907343050857, "rms_envelope_nrmse": 0.06818055130044423, "envelope_correlation": 0.9959414983144401, "rms_gain_error_db": 0.006655651750753095, "reference_mean_power": 0.0005515675704872272, "output_mean_power": 0.0005524135071108785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3413004486315466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020691279435977171, "output_power": 0.0002041581088307127, "reference_active": true, "delta_db": -0.05820711822644534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313846735404118e-05, "output_power": 2.9806118721436374e-05, "reference_active": true, "delta_db": -0.46026991566558306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.722807599466502e-06, "output_power": 2.5468669678943633e-06, "reference_active": true, "delta_db": -0.29010693288796624}], "active_band_count": 3, "active_band_mae_db": 0.26952798892666485}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.977903967560393e-05, "log_rms_mae_db": 0.6710394760426535, "rms_envelope_nrmse": 0.03728589188633603, "envelope_correlation": 0.998678936294835, "rms_gain_error_db": 0.006664283585881357, "reference_mean_power": 0.0005543343450028261, "output_mean_power": 0.000555185628475751}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0568367997569596e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001948094737446988, "output_power": 0.000191844082840495, "reference_active": true, "delta_db": -0.06661664611861354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.636770712067895e-05, "output_power": 1.4589431169752268e-05, "reference_active": true, "delta_db": -0.49949485860512977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1494053604425814e-07, "output_power": 1.9710567337285326e-07, "reference_active": true, "delta_db": -0.3761920270327574}], "active_band_count": 3, "active_band_mae_db": 0.3141011772521669}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.251077329795781e-05, "log_rms_mae_db": 1.1141290662896024, "rms_envelope_nrmse": 0.09181048803555934, "envelope_correlation": 0.9936110086060926, "rms_gain_error_db": -0.21148487898857682, "reference_mean_power": 0.0002789441343631021, "output_mean_power": 0.0002656860460077809}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.041685297045403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011680629747545264, "output_power": 0.00011129157173491767, "reference_active": true, "delta_db": -0.2100398203205542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1432581274087105e-05, "output_power": 1.043832090494031e-05, "reference_active": true, "delta_db": -0.39513653233119056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6649698962790864e-06, "output_power": 1.8160647500199691e-06, "reference_active": true, "delta_db": 0.37724943216652573}], "active_band_count": 3, "active_band_mae_db": 0.32747526160609014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.285244854844391e-05, "log_rms_mae_db": 0.6116607036949706, "rms_envelope_nrmse": 0.054278916281864695, "envelope_correlation": 0.9979648514951339, "rms_gain_error_db": -0.21157533359493452, "reference_mean_power": 0.00028031132153199156, "output_mean_power": 0.00026698269063169265}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9116341431814158e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010971300365261617, "output_power": 0.00010471980568017992, "reference_active": true, "delta_db": -0.20229277398353532}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3756201199309846e-06, "output_power": 4.8841036802064424e-06, "reference_active": true, "delta_db": -0.41643696137032316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905227736918092e-08, "output_power": 1.0444646310906957e-07, "reference_active": true, "delta_db": 0.2302927285738906}], "active_band_count": 3, "active_band_mae_db": 0.28300748797591635}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.076518340019586e-05, "log_rms_mae_db": 1.1509332401428238, "rms_envelope_nrmse": 0.1689753625619916, "envelope_correlation": 0.976577371883264, "rms_gain_error_db": -0.19987130771190784, "reference_mean_power": 9.153664281853568e-05, "output_mean_power": 8.741940565766331e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.939589252350226e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.46860955530461e-05, "output_power": 3.3764271792355566e-05, "reference_active": true, "delta_db": -0.11698028598964294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1944580647265704e-05, "output_power": 1.930703438265823e-05, "reference_active": true, "delta_db": -0.5561171606429434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8366553780699585e-06, "output_power": 2.5094936819966085e-06, "reference_active": true, "delta_db": -2.849590374680325}], "active_band_count": 3, "active_band_mae_db": 1.1742292737709705}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4119030940985276e-05, "log_rms_mae_db": 0.6592995623256603, "rms_envelope_nrmse": 0.09400039332371032, "envelope_correlation": 0.992321706895002, "rms_gain_error_db": -0.2001613759752961, "reference_mean_power": 9.196260234115854e-05, "output_mean_power": 8.78203401213985e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.164079440499503e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.035826325529754e-05, "output_power": 3.7591740513219e-05, "reference_active": true, "delta_db": -0.3084003469674281}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2217507839367818e-05, "output_power": 1.0140086953045102e-05, "reference_active": true, "delta_db": -0.8094094719995428}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.5236785056789705e-07, "output_power": 2.404749407118757e-07, "reference_active": true, "delta_db": -2.7442190604512122}], "active_band_count": 3, "active_band_mae_db": 1.287342959806061}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8642545092394602, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "daed19e023b114b729927562a9867e3836c03498b6d95a7c319cfdc878a4fb88", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9943028401403224, "standard_deviation": 0.000585074594996818, "minimum": 0.993504168263562, "maximum": 0.9949919165744385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1023331400918084, "standard_deviation": 0.018260833045031656, "minimum": 1.0701440258483115, "maximum": 1.1308609880789513, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.35649468846212884, "standard_deviation": 0.13610414424352024, "minimum": 0.16975205197122342, "maximum": 0.6276668070412513, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08650036052003261, "standard_deviation": 0.004512974853288139, "minimum": 0.0793072221650351, "maximum": 0.09181048803555934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.998179811924642, "standard_deviation": 0.0003705194182934814, "minimum": 0.9975411385241888, "maximum": 0.9985604553025129, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6320184330715545, "standard_deviation": 0.021192700380543574, "minimum": 0.6116607036949706, "maximum": 0.6672768997095322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3409089150807458, "standard_deviation": 0.17507084639727388, "minimum": 0.0761433097232412, "maximum": 0.6241347183754563, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05062805731715743, "standard_deviation": 0.005543716424620177, "minimum": 0.041241528121262395, "maximum": 0.057294447393060124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9795778808008732, "standard_deviation": 0.0033972263194199004, "minimum": 0.9738959501475531, "maximum": 0.9844024725475938, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1586969018146074, "standard_deviation": 0.031242215521492, "minimum": 1.1213517375707533, "maximum": 1.2159046575864327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7647795767549685, "standard_deviation": 0.2426223987841257, "minimum": 0.42270376578933005, "maximum": 1.1742292737709705, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1582066084641865, "standard_deviation": 0.013231489854685542, "minimum": 0.1385612404660023, "maximum": 0.17834584815219076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9934252111052192, "standard_deviation": 0.001955390477483598, "minimum": 0.9892741860874873, "maximum": 0.9951001274861936, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6835119848370961, "standard_deviation": 0.03418555281367913, "minimum": 0.6372587822536849, "maximum": 0.7326769484360789, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8947026552975372, "standard_deviation": 0.28073805912060296, "minimum": 0.46171242625187797, "maximum": 1.287342959806061, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08815677215558074, "standard_deviation": 0.01214350178074492, "minimum": 0.07523802867902442, "maximum": 0.10945170187733458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9961458237904883, "standard_deviation": 0.00041177009267502753, "minimum": 0.9955577257152217, "maximum": 0.9968908793315264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2282614319791674, "standard_deviation": 0.015437583476894388, "minimum": 1.2124822524344, "maximum": 1.261751826110526, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3591865469055746, "standard_deviation": 0.15460867893683589, "minimum": 0.1913750925554356, "maximum": 0.5830535454898124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06648486707222276, "standard_deviation": 0.003455602990804128, "minimum": 0.06022412255998135, "maximum": 0.07141641289866676, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9987418295787421, "standard_deviation": 0.0002160352918397627, "minimum": 0.9984836045468335, "maximum": 0.9991498365061526, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6789437849586396, "standard_deviation": 0.03233723376259288, "minimum": 0.6391105750069384, "maximum": 0.7272050835730577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3823019582670245, "standard_deviation": 0.16306035772939936, "minimum": 0.18535287873920991, "maximum": 0.6986618317283769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03655702547015256, "standard_deviation": 0.003061848552671914, "minimum": 0.03092126805702094, "maximum": 0.04038905913019737, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12430350257947595, "standard_deviation": 0.03433849599059254, "minimum": 0.05936370833178648, "maximum": 0.1611172359498881, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.274842201309664, "standard_deviation": 0.004859256208425541, "minimum": 5.267751315247881, "maximum": 5.283012307729628, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3588869474806793, "standard_deviation": 0.01958987501124682, "minimum": 0.33222462408125236, "maximum": 0.38376308367838535, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04641800355733334, "standard_deviation": 0.02527972779623008, "minimum": -0.09311814290879912, "maximum": -0.017090887084306977, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941530385595854, "standard_deviation": 0.000471693062682062, "minimum": 0.9935363831400292, "maximum": 0.9950104176522911, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8861312676822339, "standard_deviation": 0.027611220858178118, "minimum": 0.8422648623817953, "maximum": 0.9269934380077623, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.46192472627267883, "standard_deviation": 0.2185284494625701, "minimum": 0.23321901108176668, "maximum": 0.849405968722571, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06823250087587154, "standard_deviation": 0.002847280108610207, "minimum": 0.0629432231361113, "maximum": 0.07191934814828893, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981454255192397, "standard_deviation": 0.00030984702278353366, "minimum": 0.9978145116361702, "maximum": 0.9986377570260235, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4444278842977607, "standard_deviation": 0.02219668455215652, "minimum": 0.4270059442457008, "maximum": 0.4900658565640334, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.458624258233098, "standard_deviation": 0.20972528147867714, "minimum": 0.21072715837635383, "maximum": 0.8339597494865784, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.036661935592225095, "standard_deviation": 0.0030471898875690266, "minimum": 0.031174633404213885, "maximum": 0.03962145749373768, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135766-0005", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2227303981781006, "predicted_mos": 2.2227303981781006, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "faefd8b4ed00dbd7d2b377faefb8de491a4c2e369a233186e0489bcdaf00ee3d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.7745067002251744, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8562333918908535, "全帯域_10ms_rms_envelope_nrmse": 0.06339826943711128, "全帯域_10ms_envelope_correlation": 0.9949629785148003, "全帯域_10ms_modulation_mae_db": 0.20364389255902007, "全帯域_40ms_log_rms_mae_db": 0.4098775876820877, "全帯域_40ms_rms_envelope_nrmse": 0.033438080007532865, "全帯域_40ms_envelope_correlation": 0.9984717032831618, "全帯域_40ms_modulation_mae_db": 0.18355974345270218, "300–1000 Hz_10ms_log_rms_mae_db": 1.3262305824185339, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061625881681086396, "300–1000 Hz_10ms_envelope_correlation": 0.9967337747868038, "300–1000 Hz_10ms_modulation_mae_db": 0.4057898170835057, "300–1000 Hz_40ms_log_rms_mae_db": 0.7457495663792298, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03732239668442336, "300–1000 Hz_40ms_envelope_correlation": 0.9987048135371404, "300–1000 Hz_40ms_modulation_mae_db": 0.4176575377862635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2444146882160483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08610279180756174, "1000–3000 Hz_10ms_envelope_correlation": 0.9942508135338237, "1000–3000 Hz_10ms_modulation_mae_db": 0.22077590564504487, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7188405642365627, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050468927344354236, "1000–3000 Hz_40ms_envelope_correlation": 0.9980738878937068, "1000–3000 Hz_40ms_modulation_mae_db": 0.16107876746632893, "3000–6000 Hz_10ms_log_rms_mae_db": 1.283465214707314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14756516969586958, "3000–6000 Hz_10ms_envelope_correlation": 0.9822511992658369, "3000–6000 Hz_10ms_modulation_mae_db": 0.2045978999219409, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7353630155052334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09163221445267106, "3000–6000 Hz_40ms_envelope_correlation": 0.9927537711182342, "3000–6000 Hz_40ms_modulation_mae_db": 0.16712667205841467, "mr_stft_db": 5.359950761033301, "band_mae_db": 0.0952540075614118, "rms_gain_error_db": -0.043076950790658104, "output_peak": 0.3596662270079259}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8383641260214049, "全帯域_10ms_rms_envelope_nrmse": 0.06561991502787023, "全帯域_10ms_envelope_correlation": 0.9946529416825768, "全帯域_10ms_modulation_mae_db": 0.7462158692682007, "全帯域_40ms_log_rms_mae_db": 0.3904247378414145, "全帯域_40ms_rms_envelope_nrmse": 0.035453650605009705, "全帯域_40ms_envelope_correlation": 0.9983569322696271, "全帯域_40ms_modulation_mae_db": 0.5360712266519934, "300–1000 Hz_10ms_log_rms_mae_db": 1.332739425691853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06478592107650187, "300–1000 Hz_10ms_envelope_correlation": 0.9964016884603019, "300–1000 Hz_10ms_modulation_mae_db": 0.31479457325199156, "300–1000 Hz_40ms_log_rms_mae_db": 0.7874575580921063, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036856439717585045, "300–1000 Hz_40ms_envelope_correlation": 0.9987818761046257, "300–1000 Hz_40ms_modulation_mae_db": 0.47329022580803154, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2036879933835398, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08668963015083518, "1000–3000 Hz_10ms_envelope_correlation": 0.9940324169409395, "1000–3000 Hz_10ms_modulation_mae_db": 0.48232377353548617, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6908182409761024, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04548591682406774, "1000–3000 Hz_40ms_envelope_correlation": 0.9983105599411514, "1000–3000 Hz_40ms_modulation_mae_db": 0.5989078537635474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2752456927216902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14767523384643857, "3000–6000 Hz_10ms_envelope_correlation": 0.9822668862149923, "3000–6000 Hz_10ms_modulation_mae_db": 0.8262768920398674, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7406451721527626, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08104751516865492, "3000–6000 Hz_40ms_envelope_correlation": 0.9945165410053007, "3000–6000 Hz_40ms_modulation_mae_db": 0.9311173341814042, "mr_stft_db": 5.36354924474372, "band_mae_db": 0.1472476719993185, "rms_gain_error_db": -0.08684756831209112, "output_peak": 0.3666057190377536}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8374031902813194, "全帯域_10ms_rms_envelope_nrmse": 0.06206734769070806, "全帯域_10ms_envelope_correlation": 0.9951944620355345, "全帯域_10ms_modulation_mae_db": 0.5628091141570696, "全帯域_40ms_log_rms_mae_db": 0.43254366297003394, "全帯域_40ms_rms_envelope_nrmse": 0.034392476158709856, "全帯域_40ms_envelope_correlation": 0.9984132936258644, "全帯域_40ms_modulation_mae_db": 0.799152460696437, "300–1000 Hz_10ms_log_rms_mae_db": 1.337291566817669, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07010733644376213, "300–1000 Hz_10ms_envelope_correlation": 0.995743907444833, "300–1000 Hz_10ms_modulation_mae_db": 0.2745678272048703, "300–1000 Hz_40ms_log_rms_mae_db": 0.7689565095854068, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04066654866134291, "300–1000 Hz_40ms_envelope_correlation": 0.9984715509015668, "300–1000 Hz_40ms_modulation_mae_db": 0.47508699628767664, "1000–3000 Hz_10ms_log_rms_mae_db": 1.259852175446493, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09289956458029536, "1000–3000 Hz_10ms_envelope_correlation": 0.9933451939367033, "1000–3000 Hz_10ms_modulation_mae_db": 0.5209817034539556, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7302925355902771, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05344396680428155, "1000–3000 Hz_40ms_envelope_correlation": 0.9979122843176719, "1000–3000 Hz_40ms_modulation_mae_db": 0.5508758266880481, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2750551791215794, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15359299953122216, "3000–6000 Hz_10ms_envelope_correlation": 0.9812854997733855, "3000–6000 Hz_10ms_modulation_mae_db": 0.8675467350328855, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7509307985055719, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08989300754899318, "3000–6000 Hz_40ms_envelope_correlation": 0.9938784380977214, "3000–6000 Hz_40ms_modulation_mae_db": 1.208071146263528, "mr_stft_db": 5.3724372139107155, "band_mae_db": 0.13599016541783618, "rms_gain_error_db": -0.0687367062253297, "output_peak": 0.334192494637263}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9135116799197637, "全帯域_10ms_rms_envelope_nrmse": 0.07477635531715819, "全帯域_10ms_envelope_correlation": 0.9930946697681957, "全帯域_10ms_modulation_mae_db": 0.7467150941442656, "全帯域_40ms_log_rms_mae_db": 0.4273947887568132, "全帯域_40ms_rms_envelope_nrmse": 0.04112359406370437, "全帯域_40ms_envelope_correlation": 0.9978440724950746, "全帯域_40ms_modulation_mae_db": 0.7891036372429866, "300–1000 Hz_10ms_log_rms_mae_db": 1.374370232580238, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07795612502260253, "300–1000 Hz_10ms_envelope_correlation": 0.994812357900273, "300–1000 Hz_10ms_modulation_mae_db": 0.4918885776481758, "300–1000 Hz_40ms_log_rms_mae_db": 0.8189258961460051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0510610427528308, "300–1000 Hz_40ms_envelope_correlation": 0.9976667689635959, "300–1000 Hz_40ms_modulation_mae_db": 0.46662147535630777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.24547000644924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10087929525191823, "1000–3000 Hz_10ms_envelope_correlation": 0.9924877558779714, "1000–3000 Hz_10ms_modulation_mae_db": 0.45909652603862733, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7191775708115983, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05935692501885927, "1000–3000 Hz_40ms_envelope_correlation": 0.9978314130736046, "1000–3000 Hz_40ms_modulation_mae_db": 0.47477873689535005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3477497481875038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20618142863681144, "3000–6000 Hz_10ms_envelope_correlation": 0.964941537997124, "3000–6000 Hz_10ms_modulation_mae_db": 0.8119099313143966, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7886587530542956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10862175101740283, "3000–6000 Hz_40ms_envelope_correlation": 0.9893354923547244, "3000–6000 Hz_40ms_modulation_mae_db": 0.8106102680256715, "mr_stft_db": 5.361377614351281, "band_mae_db": 0.12576965977107832, "rms_gain_error_db": -0.10981249423291657, "output_peak": 0.3450829573203862}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8630329134355484, "全帯域_10ms_rms_envelope_nrmse": 0.06774629975585378, "全帯域_10ms_envelope_correlation": 0.9942323188060859, "全帯域_10ms_modulation_mae_db": 0.16458897421300905, "全帯域_40ms_log_rms_mae_db": 0.4199062999039955, "全帯域_40ms_rms_envelope_nrmse": 0.03398337387127754, "全帯域_40ms_envelope_correlation": 0.9984137689361434, "全帯域_40ms_modulation_mae_db": 0.18388707976561328, "300–1000 Hz_10ms_log_rms_mae_db": 1.4353968617233321, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07151357072536725, "300–1000 Hz_10ms_envelope_correlation": 0.9955805290570686, "300–1000 Hz_10ms_modulation_mae_db": 0.23271678217706745, "300–1000 Hz_40ms_log_rms_mae_db": 0.8144148485173006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04084703405983249, "300–1000 Hz_40ms_envelope_correlation": 0.9984377641282574, "300–1000 Hz_40ms_modulation_mae_db": 0.30646499701103525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1840354443290775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09863254026329318, "1000–3000 Hz_10ms_envelope_correlation": 0.992287708143141, "1000–3000 Hz_10ms_modulation_mae_db": 0.34209649526098856, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7038593548340312, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052084894565149566, "1000–3000 Hz_40ms_envelope_correlation": 0.9978213405003099, "1000–3000 Hz_40ms_modulation_mae_db": 0.40609533795795444, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2785677853232544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14529114889797454, "3000–6000 Hz_10ms_envelope_correlation": 0.9827969458196709, "3000–6000 Hz_10ms_modulation_mae_db": 0.6623798832548942, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7810013246515449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0801682829236755, "3000–6000 Hz_40ms_envelope_correlation": 0.9945561671390213, "3000–6000 Hz_40ms_modulation_mae_db": 0.6792470168599513, "mr_stft_db": 5.353800690184069, "band_mae_db": 0.1327779001699856, "rms_gain_error_db": -0.048508544711957535, "output_peak": 0.3300790322733967}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8300383023991923, "全帯域_10ms_rms_envelope_nrmse": 0.06078084428523813, "全帯域_10ms_envelope_correlation": 0.9953956111459894, "全帯域_10ms_modulation_mae_db": 0.3226451039847785, "全帯域_40ms_log_rms_mae_db": 0.3777894467385743, "全帯域_40ms_rms_envelope_nrmse": 0.03393674949279919, "全帯域_40ms_envelope_correlation": 0.9984576509308077, "全帯域_40ms_modulation_mae_db": 0.5072334725269017, "300–1000 Hz_10ms_log_rms_mae_db": 1.394220571239011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06617342112098647, "300–1000 Hz_10ms_envelope_correlation": 0.9962248190588303, "300–1000 Hz_10ms_modulation_mae_db": 0.30179958482389097, "300–1000 Hz_40ms_log_rms_mae_db": 0.828058064664775, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03810597845896836, "300–1000 Hz_40ms_envelope_correlation": 0.9986602879029979, "300–1000 Hz_40ms_modulation_mae_db": 0.38058159658840357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1726785230480987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08520391515219111, "1000–3000 Hz_10ms_envelope_correlation": 0.9946681902191753, "1000–3000 Hz_10ms_modulation_mae_db": 0.5249696620177488, "1000–3000 Hz_40ms_log_rms_mae_db": 0.671177292105322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05228362838374988, "1000–3000 Hz_40ms_envelope_correlation": 0.9982578546304821, "1000–3000 Hz_40ms_modulation_mae_db": 0.5465054479755025, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3206250421119599, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1538073735133909, "3000–6000 Hz_10ms_envelope_correlation": 0.9809877906562474, "3000–6000 Hz_10ms_modulation_mae_db": 1.0901134729812691, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7660905720129084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09755888562432612, "3000–6000 Hz_40ms_envelope_correlation": 0.9922030434209419, "3000–6000 Hz_40ms_modulation_mae_db": 1.3060103134309067, "mr_stft_db": 5.34739447189131, "band_mae_db": 0.1323733394800273, "rms_gain_error_db": -0.06779601630356405, "output_peak": 0.33281452786384585}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8519601181402231, "全帯域_10ms_rms_envelope_nrmse": 0.062256182537261257, "全帯域_10ms_envelope_correlation": 0.9951327443136677, "全帯域_10ms_modulation_mae_db": 0.5635614300386204, "全帯域_40ms_log_rms_mae_db": 0.4127611419421384, "全帯域_40ms_rms_envelope_nrmse": 0.03308786757177523, "全帯域_40ms_envelope_correlation": 0.9984834800133814, "全帯域_40ms_modulation_mae_db": 0.6736634611336597, "300–1000 Hz_10ms_log_rms_mae_db": 1.3367505846719114, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06444930982661913, "300–1000 Hz_10ms_envelope_correlation": 0.9964068037281376, "300–1000 Hz_10ms_modulation_mae_db": 0.27483412988307365, "300–1000 Hz_40ms_log_rms_mae_db": 0.8170913799235658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037987330972567124, "300–1000 Hz_40ms_envelope_correlation": 0.9986634445249507, "300–1000 Hz_40ms_modulation_mae_db": 0.38143885645124365, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2275267222217234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08996677128114174, "1000–3000 Hz_10ms_envelope_correlation": 0.9935633582558682, "1000–3000 Hz_10ms_modulation_mae_db": 0.2973421830768004, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7391076800613059, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049152784410625414, "1000–3000 Hz_40ms_envelope_correlation": 0.9980099430295217, "1000–3000 Hz_40ms_modulation_mae_db": 0.33983679313007364, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3274037006204258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16761608945941714, "3000–6000 Hz_10ms_envelope_correlation": 0.9775582388260796, "3000–6000 Hz_10ms_modulation_mae_db": 0.7923023636972077, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7720660120450332, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10140591571546687, "3000–6000 Hz_40ms_envelope_correlation": 0.9911857981267191, "3000–6000 Hz_40ms_modulation_mae_db": 0.8276289692128861, "mr_stft_db": 5.356564762096518, "band_mae_db": 0.0890963251961824, "rms_gain_error_db": -0.027074961624467105, "output_peak": 0.30831467694596915}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8562511330225115, "全帯域_10ms_rms_envelope_nrmse": 0.06255886224668294, "全帯域_10ms_envelope_correlation": 0.9951662254246062, "全帯域_10ms_modulation_mae_db": 0.5051274198440209, "全帯域_40ms_log_rms_mae_db": 0.43466502219847897, "全帯域_40ms_rms_envelope_nrmse": 0.03225110607171362, "全帯域_40ms_envelope_correlation": 0.9986826475475211, "全帯域_40ms_modulation_mae_db": 0.44215937451796433, "300–1000 Hz_10ms_log_rms_mae_db": 1.349102795910748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06105895863701039, "300–1000 Hz_10ms_envelope_correlation": 0.9968038161926313, "300–1000 Hz_10ms_modulation_mae_db": 0.32860040541904284, "300–1000 Hz_40ms_log_rms_mae_db": 0.7422005019989989, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03268256736505066, "300–1000 Hz_40ms_envelope_correlation": 0.9990518349098011, "300–1000 Hz_40ms_modulation_mae_db": 0.6969261908181941, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2364761338818717, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08897217918561078, "1000–3000 Hz_10ms_envelope_correlation": 0.9938959725207879, "1000–3000 Hz_10ms_modulation_mae_db": 0.32498763468538633, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7151651851405131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05051956859366323, "1000–3000 Hz_40ms_envelope_correlation": 0.9981316009009278, "1000–3000 Hz_40ms_modulation_mae_db": 0.45238103665959667, "3000–6000 Hz_10ms_log_rms_mae_db": 1.296462923272218, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.143337082510057, "3000–6000 Hz_10ms_envelope_correlation": 0.9831639253169373, "3000–6000 Hz_10ms_modulation_mae_db": 0.9262844425847044, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7583692953727579, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07382586157202174, "3000–6000 Hz_40ms_envelope_correlation": 0.9952563511616168, "3000–6000 Hz_40ms_modulation_mae_db": 0.9455619566461007, "mr_stft_db": 5.3565985653417485, "band_mae_db": 0.12018244856551684, "rms_gain_error_db": -0.09246374930699683, "output_peak": 0.35796387545160985}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.362127419541406, "spectral_convergence": 0.29897215635726293, "spectral_convergence_status": "ok", "amplitude_floor": 6.643709912602353e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.196573298012586, "spectral_convergence": 0.27382726022964854, "spectral_convergence_status": "ok", "amplitude_floor": 5.274693968404162e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.52115156554591, "spectral_convergence": 0.2792046778773485, "spectral_convergence_status": "ok", "amplitude_floor": 3.8895875894681354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00010232532114960487, "output_power": 0.00010188746130600289, "delta_db": -0.01862375146611951, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003592345857428397, "output_power": 0.0003652275736687115, "delta_db": 0.0718541589573288, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003905970862424533, "output_power": 0.00038669748523203996, "delta_db": -0.04357651452736124, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022506467673995083, "output_power": 0.00021870814055052895, "delta_db": -0.12442390823878907, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011237312839533447, "output_power": 0.00010686880372228749, "delta_db": -0.21811523537489177, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00013068329846048222, "output_power": 0.00012785774479250384, "delta_db": -0.09493047680398041, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.0952540075614118, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.043076950790658104, "pitch": null, "output_rms_dbfs": -28.836422766346864, "output_peak": 0.3596662270079259, "reference_rms_dbfs": -28.793345815556208, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.359950761033301}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001036839322212551, "log_rms_mae_db": 0.8562333918908535, "rms_envelope_nrmse": 0.06339826943711128, "envelope_correlation": 0.9949629785148003, "rms_gain_error_db": -0.04309021717748305, "reference_mean_power": 0.0013213586135692834, "output_mean_power": 0.0013083130651121142}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.530367535346178e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002613603148630271, "output_power": 0.00025304540006894247, "reference_active": true, "delta_db": -0.14041197677722517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.246118337274924e-05, "output_power": 7.58619423678098e-05, "reference_active": true, "delta_db": -0.36225604747076123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.028948119628933e-05, "output_power": 1.0549207702182165e-05, "reference_active": true, "delta_db": 0.10826365342907388}], "active_band_count": 3, "active_band_mae_db": 0.20364389255902007}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.673473491549852e-05, "log_rms_mae_db": 0.4098775876820877, "rms_envelope_nrmse": 0.033438080007532865, "envelope_correlation": 0.9984717032831618, "rms_gain_error_db": -0.0430563235510887, "reference_mean_power": 0.0013278763848253465, "output_mean_power": 0.0013147767483898036}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.825313121587468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002333069672738863, "output_power": 0.00022646761751539233, "reference_active": true, "delta_db": -0.1292159712508708}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6274669036315644e-05, "output_power": 4.268278337345944e-05, "reference_active": true, "delta_db": -0.35090588978161036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.136183723659665e-07, "output_power": 6.236689047646892e-07, "reference_active": true, "delta_db": 0.07055736932562535}], "active_band_count": 3, "active_band_mae_db": 0.18355974345270218}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.601318795186366e-05, "log_rms_mae_db": 1.3262305824185339, "rms_envelope_nrmse": 0.061625881681086396, "envelope_correlation": 0.9967337747868038, "rms_gain_error_db": -0.030570967512459087, "reference_mean_power": 0.0005515675704872272, "output_mean_power": 0.0005476985953013016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3413004486315466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020691279435977171, "output_power": 0.00020231469621083742, "reference_active": true, "delta_db": -0.09759914679506444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313846735404118e-05, "output_power": 2.879999306506362e-05, "reference_active": true, "delta_db": -0.6094003535222469}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.722807599466502e-06, "output_power": 2.420917328103989e-06, "reference_active": true, "delta_db": -0.5103699509332058}], "active_band_count": 3, "active_band_mae_db": 0.4057898170835057}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.977903967560393e-05, "log_rms_mae_db": 0.7457495663792298, "rms_envelope_nrmse": 0.03732239668442336, "envelope_correlation": 0.9987048135371404, "rms_gain_error_db": -0.030573682963798317, "reference_mean_power": 0.0005543343450028261, "output_mean_power": 0.0005504456180852644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0568367997569596e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001948094737446988, "output_power": 0.00019030791148723402, "reference_active": true, "delta_db": -0.10153229985594309}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.636770712067895e-05, "output_power": 1.4368192471391244e-05, "reference_active": true, "delta_db": -0.5658570826719038}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1494053604425814e-07, "output_power": 1.8782785745565668e-07, "reference_active": true, "delta_db": -0.5855832308309438}], "active_band_count": 3, "active_band_mae_db": 0.4176575377862635}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.251077329795781e-05, "log_rms_mae_db": 1.2444146882160483, "rms_envelope_nrmse": 0.08610279180756174, "envelope_correlation": 0.9942508135338237, "rms_gain_error_db": -0.15646695971067753, "reference_mean_power": 0.0002789441343631021, "output_mean_power": 0.0002690732581394659}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.041685297045403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011680629747545264, "output_power": 0.00011151229631922046, "reference_active": true, "delta_db": -0.20143498763251835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1432581274087105e-05, "output_power": 1.0740921378881523e-05, "reference_active": true, "delta_db": -0.2710275980028199}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6649698962790864e-06, "output_power": 1.7393736985289781e-06, "reference_active": true, "delta_db": 0.18986513129979632}], "active_band_count": 3, "active_band_mae_db": 0.22077590564504487}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.285244854844391e-05, "log_rms_mae_db": 0.7188405642365627, "rms_envelope_nrmse": 0.050468927344354236, "envelope_correlation": 0.9980738878937068, "rms_gain_error_db": -0.15643217526077197, "reference_mean_power": 0.00028031132153199156, "output_mean_power": 0.00027039423093456794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9116341431814158e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010971300365261617, "output_power": 0.00010504992743998797, "reference_active": true, "delta_db": -0.18862348291212178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3756201199309846e-06, "output_power": 5.034064138143783e-06, "reference_active": true, "delta_db": -0.2850982623256513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905227736918092e-08, "output_power": 9.883551040647653e-08, "reference_active": true, "delta_db": -0.00951455716121372}], "active_band_count": 3, "active_band_mae_db": 0.16107876746632893}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.076518340019586e-05, "log_rms_mae_db": 1.283465214707314, "rms_envelope_nrmse": 0.14756516969586958, "envelope_correlation": 0.9822511992658369, "rms_gain_error_db": -0.20440254578200368, "reference_mean_power": 9.153664281853568e-05, "output_mean_power": 8.732824365096385e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.939589252350226e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.46860955530461e-05, "output_power": 3.313638716147998e-05, "reference_active": true, "delta_db": -0.19850260329780872}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1944580647265704e-05, "output_power": 2.04520639325617e-05, "reference_active": true, "delta_db": -0.3059014440641863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8366553780699585e-06, "output_power": 4.960027785113003e-06, "reference_active": true, "delta_db": 0.10938965240382767}], "active_band_count": 3, "active_band_mae_db": 0.2045978999219409}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4119030940985276e-05, "log_rms_mae_db": 0.7353630155052334, "rms_envelope_nrmse": 0.09163221445267106, "envelope_correlation": 0.9927537711182342, "rms_gain_error_db": -0.20455344052946517, "reference_mean_power": 9.196260234115854e-05, "output_mean_power": 8.773157140633862e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.164079440499503e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.035826325529754e-05, "output_power": 3.860326076081753e-05, "reference_active": true, "delta_db": -0.1930847845211095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2217507839367818e-05, "output_power": 1.1526966146435918e-05, "reference_active": true, "delta_db": -0.25267608683834164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.5236785056789705e-07, "output_power": 4.466114137963891e-07, "reference_active": true, "delta_db": -0.055619144815792845}], "active_band_count": 3, "active_band_mae_db": 0.16712667205841467}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8560155388468386, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cf8ad06f9f98a83bc567038609a54f961b7c29451f33993dd179fe26881fcfff", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9935664261785513, "standard_deviation": 0.0008328626009133187, "minimum": 0.992287708143141, "maximum": 0.9946681902191753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2217677108720115, "standard_deviation": 0.03147553073121364, "minimum": 1.1726785230480987, "maximum": 1.259852175446493, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3965717354642547, "standard_deviation": 0.1146853722622623, "minimum": 0.22077590564504487, "maximum": 0.5249696620177488, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09116833595910591, "standard_deviation": 0.005862397728617173, "minimum": 0.08520391515219111, "maximum": 0.10087929525191823, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9980436105359219, "standard_deviation": 0.00018463606986506663, "minimum": 0.9978213405003099, "maximum": 0.9983105599411514, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7110548029694641, "standard_deviation": 0.02188298665610231, "minimum": 0.671177292105322, "maximum": 0.7391076800613059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4413074750670502, "standard_deviation": 0.14102532099748577, "minimum": 0.16107876746632893, "maximum": 0.5989078537635474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.051599576493093866, "standard_deviation": 0.003965748046788666, "minimum": 0.04548591682406774, "maximum": 0.05935692501885927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9794065029837842, "standard_deviation": 0.006101168582211713, "minimum": 0.964941537997124, "maximum": 0.9831639253169373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3005719107582432, "standard_deviation": 0.02786202394030251, "minimum": 1.2750551791215794, "maximum": 1.3477497481875038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7726764526033957, "standard_deviation": 0.2599555103565796, "minimum": 0.2045978999219409, "maximum": 1.0901134729812691, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15813331576139766, "standard_deviation": 0.020849047091886307, "minimum": 0.143337082510057, "maximum": 0.20618142863681144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992960700303035, "standard_deviation": 0.001999560104564164, "minimum": 0.9893354923547244, "maximum": 0.9952563511616168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7616406179125135, "standard_deviation": 0.018860504614149662, "minimum": 0.7353630155052334, "maximum": 0.7886587530542956, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8594217095848579, "standard_deviation": 0.3481967607917925, "minimum": 0.16712667205841467, "maximum": 1.3060103134309067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09051917925290154, "standard_deviation": 0.011789419408864717, "minimum": 0.07382586157202174, "maximum": 0.10862175101740283, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9960884620786099, "standard_deviation": 0.0006709653090743434, "minimum": 0.994812357900273, "maximum": 0.9968038161926313, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3607628276316621, "standard_deviation": 0.03801513302649581, "minimum": 1.3262305824185339, "maximum": 1.4353968617233321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3281239621864523, "standard_deviation": 0.08326575574945362, "minimum": 0.23271678217706745, "maximum": 0.4918885776481758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06720881556674202, "standard_deviation": 0.005681853609983018, "minimum": 0.06105895863701039, "maximum": 0.07795612502260253, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998554792621617, "standard_deviation": 0.0004058975204064906, "minimum": 0.9976667689635959, "maximum": 0.9990518349098011, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7903567906634236, "standard_deviation": 0.034447306417610665, "minimum": 0.7422005019989989, "maximum": 0.828058064664775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4497584845133945, "standard_deviation": 0.11562749277628309, "minimum": 0.30646499701103525, "maximum": 0.6969261908181941, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.039441167334075086, "standard_deviation": 0.005335120488022634, "minimum": 0.03268256736505066, "maximum": 0.0510610427528308, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12233643977016961, "standard_deviation": 0.020253913890622026, "minimum": 0.0890963251961824, "maximum": 0.1472476719993185, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.358959165444083, "standard_deviation": 0.007379666848721641, "minimum": 5.34739447189131, "maximum": 5.3724372139107155, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.34183993881726876, "standard_deviation": 0.019286644060529504, "minimum": 0.30831467694596915, "maximum": 0.3666057190377536, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06803962393849762, "standard_deviation": 0.027719246272856598, "minimum": -0.10981249423291657, "maximum": -0.027074961624467105, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947289939614321, "standard_deviation": 0.0007553106453587198, "minimum": 0.9930946697681957, "maximum": 0.9953956111459894, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8558493568888521, "standard_deviation": 0.025923477946332547, "minimum": 0.8300383023991923, "maximum": 0.9135116799197637, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4769133622761231, "standard_deviation": 0.22583819670874022, "minimum": 0.16458897421300905, "maximum": 0.7467150941442656, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06490050953723549, "standard_deviation": 0.004565074660144518, "minimum": 0.06078084428523813, "maximum": 0.07477635531715819, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983904436376977, "standard_deviation": 0.00024079317477651048, "minimum": 0.9978440724950746, "maximum": 0.9986826475475211, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41317033600419206, "standard_deviation": 0.020231639967383787, "minimum": 0.3777894467385743, "maximum": 0.43466502219847897, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5143538069985323, "standard_deviation": 0.24093409297800342, "minimum": 0.18355974345270218, "maximum": 0.799152460696437, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0347083622303153, "standard_deviation": 0.0027570470817813815, "minimum": 0.03225110607171362, "maximum": 0.04112359406370437, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135766-0005", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.374060869216919, "predicted_mos": 2.374060869216919, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e95e8348d7fc5ce1658633d9dcb377b7e9bfe5b33c1c17872c2765a1df679b25", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9417432001791894, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.807853070530199, "全帯域_10ms_rms_envelope_nrmse": 0.06170015797915739, "全帯域_10ms_envelope_correlation": 0.9952081283370322, "全帯域_10ms_modulation_mae_db": 0.4715953296689997, "全帯域_40ms_log_rms_mae_db": 0.4097603378709243, "全帯域_40ms_rms_envelope_nrmse": 0.03268452146002816, "全帯域_40ms_envelope_correlation": 0.9985093795306405, "全帯域_40ms_modulation_mae_db": 0.40520372065570887, "300–1000 Hz_10ms_log_rms_mae_db": 1.3083805523299954, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06466853974845962, "300–1000 Hz_10ms_envelope_correlation": 0.9963844997701343, "300–1000 Hz_10ms_modulation_mae_db": 0.43628289140345533, "300–1000 Hz_40ms_log_rms_mae_db": 0.7260202176724375, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03669820678780367, "300–1000 Hz_40ms_envelope_correlation": 0.9987556605488749, "300–1000 Hz_40ms_modulation_mae_db": 0.37816760624013757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2207182074585254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09814415092416735, "1000–3000 Hz_10ms_envelope_correlation": 0.992427051640315, "1000–3000 Hz_10ms_modulation_mae_db": 0.5704385665484552, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7162449638863674, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056340559247269245, "1000–3000 Hz_40ms_envelope_correlation": 0.9975094503436995, "1000–3000 Hz_40ms_modulation_mae_db": 0.7740084642425603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2733757571437563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.158438036364026, "3000–6000 Hz_10ms_envelope_correlation": 0.9795788515346506, "3000–6000 Hz_10ms_modulation_mae_db": 1.0379313977764608, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7185250446027845, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08729995278174477, "3000–6000 Hz_40ms_envelope_correlation": 0.993688760562331, "3000–6000 Hz_40ms_modulation_mae_db": 1.0969373153422861, "mr_stft_db": 5.3222040871848, "band_mae_db": 0.10781606818289081, "rms_gain_error_db": -0.028325410175968714, "output_peak": 0.35284802568290874}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8507918214021246, "全帯域_10ms_rms_envelope_nrmse": 0.05446761263585592, "全帯域_10ms_envelope_correlation": 0.9964244776124164, "全帯域_10ms_modulation_mae_db": 0.5600896802041203, "全帯域_40ms_log_rms_mae_db": 0.41349450587238745, "全帯域_40ms_rms_envelope_nrmse": 0.028796642419902582, "全帯域_40ms_envelope_correlation": 0.9990342044959759, "全帯域_40ms_modulation_mae_db": 0.5720364702830464, "300–1000 Hz_10ms_log_rms_mae_db": 1.3369389246100336, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06723669279893796, "300–1000 Hz_10ms_envelope_correlation": 0.9961142443044019, "300–1000 Hz_10ms_modulation_mae_db": 0.33076168161262026, "300–1000 Hz_40ms_log_rms_mae_db": 0.7819389371756993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037301720033574085, "300–1000 Hz_40ms_envelope_correlation": 0.9987569132393183, "300–1000 Hz_40ms_modulation_mae_db": 0.17770568402764367, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2394923693246025, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0946156333065597, "1000–3000 Hz_10ms_envelope_correlation": 0.9933872108695119, "1000–3000 Hz_10ms_modulation_mae_db": 0.6492377365466859, "1000–3000 Hz_40ms_log_rms_mae_db": 0.709919096766533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0578251325974698, "1000–3000 Hz_40ms_envelope_correlation": 0.9978534387855775, "1000–3000 Hz_40ms_modulation_mae_db": 0.5400526882888906, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2276901866885204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12725535464752402, "3000–6000 Hz_10ms_envelope_correlation": 0.9875250624982063, "3000–6000 Hz_10ms_modulation_mae_db": 0.7911396559484682, "3000–6000 Hz_40ms_log_rms_mae_db": 0.663422359432279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07169799334694334, "3000–6000 Hz_40ms_envelope_correlation": 0.9965255930447436, "3000–6000 Hz_40ms_modulation_mae_db": 0.9287023715287236, "mr_stft_db": 5.31442019801453, "band_mae_db": 0.16481971240830598, "rms_gain_error_db": -0.10090416802494467, "output_peak": 0.3241517823646509}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8488312608005741, "全帯域_10ms_rms_envelope_nrmse": 0.06952045853004278, "全帯域_10ms_envelope_correlation": 0.9940469807730014, "全帯域_10ms_modulation_mae_db": 0.5897774015589637, "全帯域_40ms_log_rms_mae_db": 0.37988359516408393, "全帯域_40ms_rms_envelope_nrmse": 0.03736509815875252, "全帯域_40ms_envelope_correlation": 0.9982409146940263, "全帯域_40ms_modulation_mae_db": 0.8018297069190895, "300–1000 Hz_10ms_log_rms_mae_db": 1.3345701284342806, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07059941799833527, "300–1000 Hz_10ms_envelope_correlation": 0.9957998597854474, "300–1000 Hz_10ms_modulation_mae_db": 0.28338607606740424, "300–1000 Hz_40ms_log_rms_mae_db": 0.7685818756023753, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0422012878892526, "300–1000 Hz_40ms_envelope_correlation": 0.9984920089898116, "300–1000 Hz_40ms_modulation_mae_db": 0.5563230184374702, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2120302496592776, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08904018728439171, "1000–3000 Hz_10ms_envelope_correlation": 0.9939347852548261, "1000–3000 Hz_10ms_modulation_mae_db": 0.48352204215469846, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6748774927009304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049747376386752404, "1000–3000 Hz_40ms_envelope_correlation": 0.998266624686525, "1000–3000 Hz_40ms_modulation_mae_db": 0.2701922381880865, "3000–6000 Hz_10ms_log_rms_mae_db": 1.236559821378228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16181600229938314, "3000–6000 Hz_10ms_envelope_correlation": 0.9785020624447862, "3000–6000 Hz_10ms_modulation_mae_db": 0.9592783816739222, "3000–6000 Hz_40ms_log_rms_mae_db": 0.713627564746363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09353948733916885, "3000–6000 Hz_40ms_envelope_correlation": 0.9923187347537624, "3000–6000 Hz_40ms_modulation_mae_db": 1.1695305001916658, "mr_stft_db": 5.338218406449539, "band_mae_db": 0.1359904565649698, "rms_gain_error_db": -0.10204230066394693, "output_peak": 0.3495181200199856}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8450821611319124, "全帯域_10ms_rms_envelope_nrmse": 0.058282836450437615, "全帯域_10ms_envelope_correlation": 0.9957342663835983, "全帯域_10ms_modulation_mae_db": 0.3814576701781816, "全帯域_40ms_log_rms_mae_db": 0.41836833514167704, "全帯域_40ms_rms_envelope_nrmse": 0.02985551287127328, "全帯域_40ms_envelope_correlation": 0.9987705046171756, "全帯域_40ms_modulation_mae_db": 0.372556564570708, "300–1000 Hz_10ms_log_rms_mae_db": 1.3426703567141007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06329116951206154, "300–1000 Hz_10ms_envelope_correlation": 0.9965365649469214, "300–1000 Hz_10ms_modulation_mae_db": 0.18725591162056962, "300–1000 Hz_40ms_log_rms_mae_db": 0.7370436845563659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034360559510746395, "300–1000 Hz_40ms_envelope_correlation": 0.9989183798498575, "300–1000 Hz_40ms_modulation_mae_db": 0.26132767493168146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1721768340142826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08233807995138812, "1000–3000 Hz_10ms_envelope_correlation": 0.9947471697558602, "1000–3000 Hz_10ms_modulation_mae_db": 0.5264719309260678, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6632691989440447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0476742508752899, "1000–3000 Hz_40ms_envelope_correlation": 0.9982867190206486, "1000–3000 Hz_40ms_modulation_mae_db": 0.6947082546482534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2506163056459805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14268181150774026, "3000–6000 Hz_10ms_envelope_correlation": 0.9833320984233223, "3000–6000 Hz_10ms_modulation_mae_db": 0.19932869852839832, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6742751751951731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06849878920442075, "3000–6000 Hz_40ms_envelope_correlation": 0.9958088218733361, "3000–6000 Hz_40ms_modulation_mae_db": 0.31458515906057855, "mr_stft_db": 5.32877806844879, "band_mae_db": 0.08179536754325983, "rms_gain_error_db": -0.037012445551353336, "output_peak": 0.34426090751080696}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8642089881294426, "全帯域_10ms_rms_envelope_nrmse": 0.06379395015709212, "全帯域_10ms_envelope_correlation": 0.9949108749225553, "全帯域_10ms_modulation_mae_db": 0.2840263558161684, "全帯域_40ms_log_rms_mae_db": 0.4348600746572103, "全帯域_40ms_rms_envelope_nrmse": 0.03335919700052772, "全帯域_40ms_envelope_correlation": 0.9985010163137609, "全帯域_40ms_modulation_mae_db": 0.24667389026116804, "300–1000 Hz_10ms_log_rms_mae_db": 1.335975651772831, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06866920661034694, "300–1000 Hz_10ms_envelope_correlation": 0.9958986592423656, "300–1000 Hz_10ms_modulation_mae_db": 0.3804928096599583, "300–1000 Hz_40ms_log_rms_mae_db": 0.7777045953530337, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04254409868176895, "300–1000 Hz_40ms_envelope_correlation": 0.9982936259400793, "300–1000 Hz_40ms_modulation_mae_db": 0.4705155044941363, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1716792818888642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0888612372941455, "1000–3000 Hz_10ms_envelope_correlation": 0.9939884498575463, "1000–3000 Hz_10ms_modulation_mae_db": 0.26758091352815966, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6623319033277987, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054004651675818534, "1000–3000 Hz_40ms_envelope_correlation": 0.9979199657174305, "1000–3000 Hz_40ms_modulation_mae_db": 0.28567506306740953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2770379691414944, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16940029657996783, "3000–6000 Hz_10ms_envelope_correlation": 0.9769693479029655, "3000–6000 Hz_10ms_modulation_mae_db": 1.2766169470445536, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7203351815257083, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09853226155526967, "3000–6000 Hz_40ms_envelope_correlation": 0.9924469164824874, "3000–6000 Hz_40ms_modulation_mae_db": 1.402496431162268, "mr_stft_db": 5.322303970172739, "band_mae_db": 0.08999203744621888, "rms_gain_error_db": -0.06367320560373421, "output_peak": 0.34497785929649233}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8067541946742647, "全帯域_10ms_rms_envelope_nrmse": 0.06657187739111836, "全帯域_10ms_envelope_correlation": 0.9944547621943861, "全帯域_10ms_modulation_mae_db": 0.5050383685492424, "全帯域_40ms_log_rms_mae_db": 0.3812888465324456, "全帯域_40ms_rms_envelope_nrmse": 0.03520065399173848, "全帯域_40ms_envelope_correlation": 0.9983239521128633, "全帯域_40ms_modulation_mae_db": 0.6621277904181958, "300–1000 Hz_10ms_log_rms_mae_db": 1.361264668442505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07149010409922193, "300–1000 Hz_10ms_envelope_correlation": 0.9955744591163272, "300–1000 Hz_10ms_modulation_mae_db": 0.24791671586047212, "300–1000 Hz_40ms_log_rms_mae_db": 0.7508262211832859, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0376501145488623, "300–1000 Hz_40ms_envelope_correlation": 0.9986788137691566, "300–1000 Hz_40ms_modulation_mae_db": 0.3327474784293066, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1925846955287591, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08828804053344882, "1000–3000 Hz_10ms_envelope_correlation": 0.99416942396381, "1000–3000 Hz_10ms_modulation_mae_db": 0.579364260444232, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7236309583579005, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05603296166724362, "1000–3000 Hz_40ms_envelope_correlation": 0.9978364144020666, "1000–3000 Hz_40ms_modulation_mae_db": 0.32552628208192197, "3000–6000 Hz_10ms_log_rms_mae_db": 1.241666602994227, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14011215485016995, "3000–6000 Hz_10ms_envelope_correlation": 0.9840138718983275, "3000–6000 Hz_10ms_modulation_mae_db": 1.078310405335764, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6781547146675835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07614983208754084, "3000–6000 Hz_40ms_envelope_correlation": 0.9950746286394385, "3000–6000 Hz_40ms_modulation_mae_db": 1.3009160799866848, "mr_stft_db": 5.3194626283172335, "band_mae_db": 0.10886931212291702, "rms_gain_error_db": -0.057190268328378806, "output_peak": 0.33825802433999147}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8573331285782333, "全帯域_10ms_rms_envelope_nrmse": 0.05904907696609342, "全帯域_10ms_envelope_correlation": 0.9956185025704087, "全帯域_10ms_modulation_mae_db": 0.3331310348741939, "全帯域_40ms_log_rms_mae_db": 0.40505387771439044, "全帯域_40ms_rms_envelope_nrmse": 0.02923246371033506, "全帯域_40ms_envelope_correlation": 0.9988154381859227, "全帯域_40ms_modulation_mae_db": 0.4573436283078683, "300–1000 Hz_10ms_log_rms_mae_db": 1.3660204943435414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06714097381139364, "300–1000 Hz_10ms_envelope_correlation": 0.9960917350829518, "300–1000 Hz_10ms_modulation_mae_db": 0.3473722579822021, "300–1000 Hz_40ms_log_rms_mae_db": 0.7395973772055269, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03585678711687875, "300–1000 Hz_40ms_envelope_correlation": 0.998810034645386, "300–1000 Hz_40ms_modulation_mae_db": 0.46131497767875285, "1000–3000 Hz_10ms_log_rms_mae_db": 1.130823376131129, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08177419088370087, "1000–3000 Hz_10ms_envelope_correlation": 0.9946992464203952, "1000–3000 Hz_10ms_modulation_mae_db": 0.21946504823222632, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6254847726449001, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04465050624969491, "1000–3000 Hz_40ms_envelope_correlation": 0.9983378801822022, "1000–3000 Hz_40ms_modulation_mae_db": 0.4612315057020617, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2557099338759254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1480725644682693, "3000–6000 Hz_10ms_envelope_correlation": 0.9823645215733563, "3000–6000 Hz_10ms_modulation_mae_db": 0.869860491912208, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7105476561803401, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08349858064197463, "3000–6000 Hz_40ms_envelope_correlation": 0.9944748849876743, "3000–6000 Hz_40ms_modulation_mae_db": 1.1144209083833738, "mr_stft_db": 5.330387746310945, "band_mae_db": 0.07446928773727819, "rms_gain_error_db": -0.02557772995782269, "output_peak": 0.35794827524634865}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8610442727816998, "全帯域_10ms_rms_envelope_nrmse": 0.06216408667776952, "全帯域_10ms_envelope_correlation": 0.9951393229107821, "全帯域_10ms_modulation_mae_db": 0.2316005826226859, "全帯域_40ms_log_rms_mae_db": 0.38719284425638106, "全帯域_40ms_rms_envelope_nrmse": 0.0329354208952768, "全帯域_40ms_envelope_correlation": 0.998478297309255, "全帯域_40ms_modulation_mae_db": 0.19758085200184786, "300–1000 Hz_10ms_log_rms_mae_db": 1.3559234305891557, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06519149249418603, "300–1000 Hz_10ms_envelope_correlation": 0.9962955602566643, "300–1000 Hz_10ms_modulation_mae_db": 0.29296190573973624, "300–1000 Hz_40ms_log_rms_mae_db": 0.7117614139683656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034530754887059055, "300–1000 Hz_40ms_envelope_correlation": 0.9988691574403419, "300–1000 Hz_40ms_modulation_mae_db": 0.40324651662718924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.207034450846406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09055959066250636, "1000–3000 Hz_10ms_envelope_correlation": 0.9934559284956771, "1000–3000 Hz_10ms_modulation_mae_db": 0.1360489984716885, "1000–3000 Hz_40ms_log_rms_mae_db": 0.672019163950872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0523299666113539, "1000–3000 Hz_40ms_envelope_correlation": 0.9976879270085343, "1000–3000 Hz_40ms_modulation_mae_db": 0.1183053705711904, "3000–6000 Hz_10ms_log_rms_mae_db": 1.265153095960655, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1374542112662648, "3000–6000 Hz_10ms_envelope_correlation": 0.9846754345132601, "3000–6000 Hz_10ms_modulation_mae_db": 1.1410143825988202, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7199070604860247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07566268348756432, "3000–6000 Hz_40ms_envelope_correlation": 0.9952188751637163, "3000–6000 Hz_40ms_modulation_mae_db": 1.3087146264948668, "mr_stft_db": 5.3381378240151705, "band_mae_db": 0.0873312469230253, "rms_gain_error_db": -0.0081442417166241, "output_peak": 0.36251058334854663}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.33265385758622, "spectral_convergence": 0.3087061750657474, "spectral_convergence_status": "ok", "amplitude_floor": 6.643709912602353e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.148443700785856, "spectral_convergence": 0.27654012780631143, "spectral_convergence_status": "ok", "amplitude_floor": 5.274693968404162e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.485514703182322, "spectral_convergence": 0.2834934554048058, "spectral_convergence_status": "ok", "amplitude_floor": 3.8895875894681354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00010232532114960487, "output_power": 0.00010602206819559729, "delta_db": 0.15413155490773284, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003592345857428397, "output_power": 0.0003640773464825892, "delta_db": 0.05815514993511409, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003905970862424533, "output_power": 0.00038610942398972293, "delta_db": -0.050185974116951304, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022506467673995083, "output_power": 0.00021976456381454768, "delta_db": -0.1034967363849651, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011237312839533447, "output_power": 0.00010664741156690558, "delta_db": -0.2271215233279321, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00013068329846048222, "output_power": 0.0001290742292930122, "delta_db": -0.05380547042464951, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10781606818289081, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.028325410175968714, "pitch": null, "output_rms_dbfs": -28.821671225732175, "output_peak": 0.35284802568290874, "reference_rms_dbfs": -28.793345815556208, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.3222040871848}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001036839322212551, "log_rms_mae_db": 0.807853070530199, "rms_envelope_nrmse": 0.06170015797915739, "envelope_correlation": 0.9952081283370322, "rms_gain_error_db": -0.028311947560524867, "reference_mean_power": 0.0013213586135692834, "output_mean_power": 0.0013127726052355618}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.530367535346178e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002613603148630271, "output_power": 0.00025447194217526536, "reference_active": true, "delta_db": -0.11599740233718765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.246118337274924e-05, "output_power": 8.156719535767856e-05, "reference_active": true, "delta_db": -0.04734033332915207}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.028948119628933e-05, "output_power": 7.713449712644344e-06, "reference_active": true, "delta_db": -1.2514482533406595}], "active_band_count": 3, "active_band_mae_db": 0.4715953296689997}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.673473491549852e-05, "log_rms_mae_db": 0.4097603378709243, "rms_envelope_nrmse": 0.03268452146002816, "envelope_correlation": 0.9985093795306405, "rms_gain_error_db": -0.02822257693452042, "reference_mean_power": 0.0013278763848253465, "output_mean_power": 0.0013192751731246908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.825313121587468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002333069672738863, "output_power": 0.00022749767092373926, "reference_active": true, "delta_db": -0.10950753580811547}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6274669036315644e-05, "output_power": 4.4214492595999656e-05, "reference_active": true, "delta_db": -0.19778675862687067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.136183723659665e-07, "output_power": 4.978134973868887e-07, "reference_active": true, "delta_db": -0.9083168675321405}], "active_band_count": 3, "active_band_mae_db": 0.40520372065570887}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.601318795186366e-05, "log_rms_mae_db": 1.3083805523299954, "rms_envelope_nrmse": 0.06466853974845962, "envelope_correlation": 0.9963844997701343, "rms_gain_error_db": -0.0414164330527133, "reference_mean_power": 0.0005515675704872272, "output_mean_power": 0.000546332555497197}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3413004486315466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020691279435977171, "output_power": 0.00020356720729892774, "reference_active": true, "delta_db": -0.07079527226640579}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313846735404118e-05, "output_power": 2.889990135015026e-05, "reference_active": true, "delta_db": -0.5943605824057837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.722807599466502e-06, "output_power": 2.347727419120308e-06, "reference_active": true, "delta_db": -0.6436928195381764}], "active_band_count": 3, "active_band_mae_db": 0.43628289140345533}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.977903967560393e-05, "log_rms_mae_db": 0.7260202176724375, "rms_envelope_nrmse": 0.03669820678780367, "envelope_correlation": 0.9987556605488749, "rms_gain_error_db": -0.04141527102959322, "reference_mean_power": 0.0005543343450028261, "output_mean_power": 0.0005490732170317094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0568367997569596e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001948094737446988, "output_power": 0.000191144263996744, "reference_active": true, "delta_db": -0.08248803243843805}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.636770712067895e-05, "output_power": 1.409080309329206e-05, "reference_active": true, "delta_db": -0.6505209923174061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1494053604425814e-07, "output_power": 1.9596068469981304e-07, "reference_active": true, "delta_db": -0.40149379396456847}], "active_band_count": 3, "active_band_mae_db": 0.37816760624013757}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.251077329795781e-05, "log_rms_mae_db": 1.2207182074585254, "rms_envelope_nrmse": 0.09814415092416735, "envelope_correlation": 0.992427051640315, "rms_gain_error_db": -0.1481399961683375, "reference_mean_power": 0.0002789441343631021, "output_mean_power": 0.00026958966179051885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.041685297045403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011680629747545264, "output_power": 0.00011141493545357306, "reference_active": true, "delta_db": -0.2052284487593732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1432581274087105e-05, "output_power": 1.025181280018804e-05, "reference_active": true, "delta_db": -0.47343630163186406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6649698962790864e-06, "output_power": 1.3126268714241908e-06, "reference_active": true, "delta_db": -1.0326509492541283}], "active_band_count": 3, "active_band_mae_db": 0.5704385665484552}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.285244854844391e-05, "log_rms_mae_db": 0.7162449638863674, "rms_envelope_nrmse": 0.056340559247269245, "envelope_correlation": 0.9975094503436995, "rms_gain_error_db": -0.14805394760684712, "reference_mean_power": 0.00028031132153199156, "output_mean_power": 0.00027091636766540263}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9116341431814158e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010971300365261617, "output_power": 0.0001045541918144154, "reference_active": true, "delta_db": -0.2091665570530217}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3756201199309846e-06, "output_power": 4.750321261993232e-06, "reference_active": true, "delta_db": -0.5370558900163628}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905227736918092e-08, "output_power": 6.89103047490268e-08, "reference_active": true, "delta_db": -1.5758029456582965}], "active_band_count": 3, "active_band_mae_db": 0.7740084642425603}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.076518340019586e-05, "log_rms_mae_db": 1.2733757571437563, "rms_envelope_nrmse": 0.158438036364026, "envelope_correlation": 0.9795788515346506, "rms_gain_error_db": -0.2562337973911085, "reference_mean_power": 9.153664281853568e-05, "output_mean_power": 8.629221175650994e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.939589252350226e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.46860955530461e-05, "output_power": 3.314018487772692e-05, "reference_active": true, "delta_db": -0.19800489282156056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1944580647265704e-05, "output_power": 1.968041093927351e-05, "reference_active": true, "delta_db": -0.47293123524999986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8366553780699585e-06, "output_power": 2.7558738733760917e-06, "reference_active": true, "delta_db": -2.4428580652578216}], "active_band_count": 3, "active_band_mae_db": 1.0379313977764608}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4119030940985276e-05, "log_rms_mae_db": 0.7185250446027845, "rms_envelope_nrmse": 0.08729995278174477, "envelope_correlation": 0.993688760562331, "rms_gain_error_db": -0.2561222966603389, "reference_mean_power": 9.196260234115854e-05, "output_mean_power": 8.669599248061589e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.164079440499503e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.035826325529754e-05, "output_power": 3.729892675202493e-05, "reference_active": true, "delta_db": -0.34236133392941176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2217507839367818e-05, "output_power": 1.0644838534244025e-05, "reference_active": true, "delta_db": -0.5984354811756399}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.5236785056789705e-07, "output_power": 2.6332386400353725e-07, "reference_active": true, "delta_db": -2.3500151309218067}], "active_band_count": 3, "active_band_mae_db": 1.0969373153422861}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8596975802777784, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1ceedf53f0a3ca827eb2d089cc6e28be1b652f099971501999ebbc831f0ba347", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9938511582822427, "standard_deviation": 0.0007605222881806285, "minimum": 0.992427051640315, "maximum": 0.9947471697558602, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1933174331064809, "standard_deviation": 0.03431457796943937, "minimum": 1.130823376131129, "maximum": 1.2394923693246025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4290161871065268, "standard_deviation": 0.19249255073326604, "minimum": 0.1360489984716885, "maximum": 0.6492377365466859, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08920263885503854, "standard_deviation": 0.005533997638692289, "minimum": 0.08177419088370087, "maximum": 0.09814415092416735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9979623025183355, "standard_deviation": 0.0003044351157495265, "minimum": 0.9975094503436995, "maximum": 0.9983378801822022, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6809721938224184, "standard_deviation": 0.033292800984369444, "minimum": 0.6254847726449001, "maximum": 0.7236309583579005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4337124833487968, "standard_deviation": 0.22546001620270784, "minimum": 0.1183053705711904, "maximum": 0.7740084642425603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05232567566386154, "standard_deviation": 0.004630918921260166, "minimum": 0.04465050624969491, "maximum": 0.0578251325974698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9821201563486093, "standard_deviation": 0.0035241370741442993, "minimum": 0.9769693479029655, "maximum": 0.9875250624982063, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2534762091035985, "standard_deviation": 0.017692681017981562, "minimum": 1.2276901866885204, "maximum": 1.2770379691414944, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9191850451023245, "standard_deviation": 0.32841971504554457, "minimum": 0.19932869852839832, "maximum": 1.2766169470445536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14815380399791817, "standard_deviation": 0.014089907418286545, "minimum": 0.12725535464752402, "maximum": 0.16940029657996783, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944446519384362, "standard_deviation": 0.0015245760245934596, "minimum": 0.9923187347537624, "maximum": 0.9965255930447436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.699849344604532, "standard_deviation": 0.023684463954962855, "minimum": 0.663422359432279, "maximum": 0.7203351815257083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0795379240188059, "standard_deviation": 0.34296079907787314, "minimum": 0.31458515906057855, "maximum": 1.402496431162268, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0818599475555784, "standard_deviation": 0.010686799904043504, "minimum": 0.06849878920442075, "maximum": 0.09853226155526967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9960869478131518, "standard_deviation": 0.00031970894977668155, "minimum": 0.9955744591163272, "maximum": 0.9965365649469214, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3427180259045555, "standard_deviation": 0.018461062944905765, "minimum": 1.3083805523299954, "maximum": 1.3660204943435414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.31330378124330227, "standard_deviation": 0.07798906627193118, "minimum": 0.18725591162056962, "maximum": 0.43628289140345533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06728594963411787, "standard_deviation": 0.0028731934671701914, "minimum": 0.06329116951206154, "maximum": 0.07149010409922193, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986968243028533, "standard_deviation": 0.00020822966304293104, "minimum": 0.9982936259400793, "maximum": 0.9989183798498575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7491842903396363, "standard_deviation": 0.025189527554661646, "minimum": 0.7117614139683656, "maximum": 0.7819389371756993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3801685576082897, "standard_deviation": 0.12170070530087236, "minimum": 0.17770568402764367, "maximum": 0.5563230184374702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.037642941181993225, "standard_deviation": 0.0031500329989317755, "minimum": 0.034360559510746395, "maximum": 0.04254409868176895, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.10638543611610822, "standard_deviation": 0.030599616886515642, "minimum": 0.07446928773727819, "maximum": 0.16481971240830598, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.326739116114219, "standard_deviation": 0.008652947491778345, "minimum": 5.31442019801453, "maximum": 5.338218406449539, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3468091972262164, "standard_deviation": 0.012016339365088103, "minimum": 0.3241517823646509, "maximum": 0.36251058334854663, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.052858721252846685, "standard_deviation": 0.03476395872430997, "minimum": -0.10204230066394693, "maximum": -0.0081442417166241, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951921644630225, "standard_deviation": 0.0007494391276571028, "minimum": 0.9940469807730014, "maximum": 0.9964244776124164, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8427373622535563, "standard_deviation": 0.022767207737009385, "minimum": 0.8067541946742647, "maximum": 0.8642089881294426, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4195895529340695, "standard_deviation": 0.13166876171747333, "minimum": 0.2316005826226859, "maximum": 0.5897774015589637, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06194375709844589, "standard_deviation": 0.004781381317462842, "minimum": 0.05446761263585592, "maximum": 0.06952045853004278, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985842134074525, "standard_deviation": 0.0002672589237385836, "minimum": 0.9982409146940263, "maximum": 0.9990342044959759, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4037378021511875, "standard_deviation": 0.019502963260227667, "minimum": 0.37988359516408393, "maximum": 0.4348600746572103, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46441907792720405, "standard_deviation": 0.2053314456955612, "minimum": 0.19758085200184786, "maximum": 0.8018297069190895, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03242868881347932, "standard_deviation": 0.0030063944173834415, "minimum": 0.028796642419902582, "maximum": 0.03736509815875252, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1221-135767-0006", "language": "en", "style": "normal", "speaker": "librispeech_1221", "source_utterance_id": "1221-135767-0006", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1221-135767-0006.wav", "source_file_sha256": "225d1fdaced09705b0911093e8eb40cc30c0d4f2183589421c9692ddaad70476", "decoded_pcm_sha256": "1bf36f00ad8a7e2e9cd6b41efd7f023aab063334b9ad526bda961251493c65cb", "original_source_sha256": "d70493003bb4a6833163a79d2df440c45bad786d8b50b5c194b896f5a4afd905", "corpus_entry_id": "librispeech_1221-135767-0006"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.328368663787842, "predicted_mos": 4.328368663787842, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "56179cfd36d6aead071e45cb2b7effc63e172033c952d5287bd0ef4531eda47f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.012870999984443, "cpu_threads": 4}, "waveform_sha256": "84fceb0c0dfc381187a05e6508e65bd777fc11fbe297cc91b853b9885470b700"}, "rows": [{"case": "librispeech_1221-135767-0006", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.9785609245300293, "predicted_mos": 1.9785609245300293, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fd552f26644b7de6d1c15e21e2b5772d2d5bce92f8d30bb739acda523ac7e797", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9141962002031505, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8803563901400239, "全帯域_10ms_rms_envelope_nrmse": 0.08632403443803635, "全帯域_10ms_envelope_correlation": 0.9899416579358108, "全帯域_10ms_modulation_mae_db": 0.4313026625224546, "全帯域_40ms_log_rms_mae_db": 0.4553167821411913, "全帯域_40ms_rms_envelope_nrmse": 0.047171587046624754, "全帯域_40ms_envelope_correlation": 0.9967770794992178, "全帯域_40ms_modulation_mae_db": 0.46453115055450905, "300–1000 Hz_10ms_log_rms_mae_db": 1.376745476782376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1075428263295487, "300–1000 Hz_10ms_envelope_correlation": 0.9900739436785474, "300–1000 Hz_10ms_modulation_mae_db": 0.8900212249391638, "300–1000 Hz_40ms_log_rms_mae_db": 0.7581676716672091, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05982225539177356, "300–1000 Hz_40ms_envelope_correlation": 0.9969950671301281, "300–1000 Hz_40ms_modulation_mae_db": 1.0371400400742685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2842658269672924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12202299445911771, "1000–3000 Hz_10ms_envelope_correlation": 0.9889820445603761, "1000–3000 Hz_10ms_modulation_mae_db": 0.8842398578872729, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7607209020766068, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07871014879390889, "1000–3000 Hz_40ms_envelope_correlation": 0.9957711315488016, "1000–3000 Hz_40ms_modulation_mae_db": 0.9272538310126658, "3000–6000 Hz_10ms_log_rms_mae_db": 1.276470864225047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15312049085147655, "3000–6000 Hz_10ms_envelope_correlation": 0.9864304084503189, "3000–6000 Hz_10ms_modulation_mae_db": 0.7974204259700503, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7749369959239061, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10337609782029264, "3000–6000 Hz_40ms_envelope_correlation": 0.9945308710909734, "3000–6000 Hz_40ms_modulation_mae_db": 1.0602422565298923, "mr_stft_db": 5.448912497085609, "band_mae_db": 0.15871643897982216, "rms_gain_error_db": -0.08024015736170112, "output_peak": 0.2950827748031276}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8814253598253459, "全帯域_10ms_rms_envelope_nrmse": 0.08059179838555997, "全帯域_10ms_envelope_correlation": 0.9912542938150716, "全帯域_10ms_modulation_mae_db": 0.3367363202409739, "全帯域_40ms_log_rms_mae_db": 0.4626268779532239, "全帯域_40ms_rms_envelope_nrmse": 0.042267914084041174, "全帯域_40ms_envelope_correlation": 0.9973704457611954, "全帯域_40ms_modulation_mae_db": 0.41774383096421924, "300–1000 Hz_10ms_log_rms_mae_db": 1.332071966163132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08626482576389101, "300–1000 Hz_10ms_envelope_correlation": 0.9934716447476261, "300–1000 Hz_10ms_modulation_mae_db": 0.4849248718171972, "300–1000 Hz_40ms_log_rms_mae_db": 0.7029955551860495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04235001669555565, "300–1000 Hz_40ms_envelope_correlation": 0.9982995670696841, "300–1000 Hz_40ms_modulation_mae_db": 0.48518984403717375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.253281394353584, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.116810229753379, "1000–3000 Hz_10ms_envelope_correlation": 0.9894035746684638, "1000–3000 Hz_10ms_modulation_mae_db": 0.20264340398708405, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7340745564570776, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05898491136547543, "1000–3000 Hz_40ms_envelope_correlation": 0.997197130224817, "1000–3000 Hz_40ms_modulation_mae_db": 0.6672687606766953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3072713880414863, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15182135723521004, "3000–6000 Hz_10ms_envelope_correlation": 0.9843060892524484, "3000–6000 Hz_10ms_modulation_mae_db": 0.4692706140660208, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7688753515455156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09012126649910229, "3000–6000 Hz_40ms_envelope_correlation": 0.9944165662034382, "3000–6000 Hz_40ms_modulation_mae_db": 0.6163451112122554, "mr_stft_db": 5.435671938220402, "band_mae_db": 0.05385946995229074, "rms_gain_error_db": -0.03903214523776952, "output_peak": 0.30168826901755164}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8951205446675794, "全帯域_10ms_rms_envelope_nrmse": 0.09153276721611964, "全帯域_10ms_envelope_correlation": 0.9887351038754623, "全帯域_10ms_modulation_mae_db": 0.22232449891675513, "全帯域_40ms_log_rms_mae_db": 0.4767112518873481, "全帯域_40ms_rms_envelope_nrmse": 0.0489585963017608, "全帯域_40ms_envelope_correlation": 0.9964539636506857, "全帯域_40ms_modulation_mae_db": 0.1918541463256084, "300–1000 Hz_10ms_log_rms_mae_db": 1.3144234282986385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09289993578776506, "300–1000 Hz_10ms_envelope_correlation": 0.992423308709367, "300–1000 Hz_10ms_modulation_mae_db": 0.2105790906620777, "300–1000 Hz_40ms_log_rms_mae_db": 0.6958801082501811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05267644584428778, "300–1000 Hz_40ms_envelope_correlation": 0.9973728950163824, "300–1000 Hz_40ms_modulation_mae_db": 0.2149883527556514, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2369579040607344, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11504844072700568, "1000–3000 Hz_10ms_envelope_correlation": 0.9899557804805404, "1000–3000 Hz_10ms_modulation_mae_db": 0.42871131814306024, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7316255696983508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06657492501814714, "1000–3000 Hz_40ms_envelope_correlation": 0.9967988227287252, "1000–3000 Hz_40ms_modulation_mae_db": 0.7807787770705197, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3038964897023517, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1901625147578121, "3000–6000 Hz_10ms_envelope_correlation": 0.9762773190954098, "3000–6000 Hz_10ms_modulation_mae_db": 0.3397190411033817, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7697115235078975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10019151405235711, "3000–6000 Hz_40ms_envelope_correlation": 0.9933392417900198, "3000–6000 Hz_40ms_modulation_mae_db": 0.4876015630917921, "mr_stft_db": 5.453850236230234, "band_mae_db": 0.17205154058542851, "rms_gain_error_db": 0.012549073126321773, "output_peak": 0.288324165400333}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8477244870802343, "全帯域_10ms_rms_envelope_nrmse": 0.0889441363787597, "全帯域_10ms_envelope_correlation": 0.9893897321740674, "全帯域_10ms_modulation_mae_db": 0.34681092505842764, "全帯域_40ms_log_rms_mae_db": 0.42552947752941944, "全帯域_40ms_rms_envelope_nrmse": 0.04971088715002501, "全帯域_40ms_envelope_correlation": 0.9964014847185224, "全帯域_40ms_modulation_mae_db": 0.44735835808369356, "300–1000 Hz_10ms_log_rms_mae_db": 1.393852933125597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1178649858871829, "300–1000 Hz_10ms_envelope_correlation": 0.9877763043686104, "300–1000 Hz_10ms_modulation_mae_db": 0.24639932178338167, "300–1000 Hz_40ms_log_rms_mae_db": 0.7762165886067628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07086167029975145, "300–1000 Hz_40ms_envelope_correlation": 0.9951737374986867, "300–1000 Hz_40ms_modulation_mae_db": 0.45131662707877745, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2510793900688009, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10075649420756158, "1000–3000 Hz_10ms_envelope_correlation": 0.9922320185238593, "1000–3000 Hz_10ms_modulation_mae_db": 0.48101170060675713, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7529654593079272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058069378556817944, "1000–3000 Hz_40ms_envelope_correlation": 0.9974643930803155, "1000–3000 Hz_40ms_modulation_mae_db": 0.3667682643433042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3317642428922334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1427934113177815, "3000–6000 Hz_10ms_envelope_correlation": 0.9860459388476444, "3000–6000 Hz_10ms_modulation_mae_db": 0.8730399615938854, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7940428882768265, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08274170636485537, "3000–6000 Hz_40ms_envelope_correlation": 0.9951207722121352, "3000–6000 Hz_40ms_modulation_mae_db": 1.0282989222082808, "mr_stft_db": 5.438527953325743, "band_mae_db": 0.09401551648432387, "rms_gain_error_db": -0.042666274647430585, "output_peak": 0.29293604116636923}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8528029910808006, "全帯域_10ms_rms_envelope_nrmse": 0.086897491589368, "全帯域_10ms_envelope_correlation": 0.989763315667837, "全帯域_10ms_modulation_mae_db": 0.33453201154171014, "全帯域_40ms_log_rms_mae_db": 0.39651119066179363, "全帯域_40ms_rms_envelope_nrmse": 0.03989398054863048, "全帯域_40ms_envelope_correlation": 0.9976472949582413, "全帯域_40ms_modulation_mae_db": 0.7558810014688375, "300–1000 Hz_10ms_log_rms_mae_db": 1.3814757150476744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1024840180998879, "300–1000 Hz_10ms_envelope_correlation": 0.9907845635427036, "300–1000 Hz_10ms_modulation_mae_db": 0.47523335594598365, "300–1000 Hz_40ms_log_rms_mae_db": 0.7315831007307908, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05417683007396353, "300–1000 Hz_40ms_envelope_correlation": 0.997187717071321, "300–1000 Hz_40ms_modulation_mae_db": 0.9278414079738194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2888085822735778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12130038347836179, "1000–3000 Hz_10ms_envelope_correlation": 0.9887305740891814, "1000–3000 Hz_10ms_modulation_mae_db": 0.6672110033270764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6934805067464315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06388156043112582, "1000–3000 Hz_40ms_envelope_correlation": 0.9970496604300005, "1000–3000 Hz_40ms_modulation_mae_db": 0.3353941608644673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3071978873712717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15472106444645684, "3000–6000 Hz_10ms_envelope_correlation": 0.9840006278275488, "3000–6000 Hz_10ms_modulation_mae_db": 0.9644778586579457, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7963249261304977, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0925458442375459, "3000–6000 Hz_40ms_envelope_correlation": 0.9940422867702027, "3000–6000 Hz_40ms_modulation_mae_db": 1.2473721035503997, "mr_stft_db": 5.446748560134959, "band_mae_db": 0.12369058475688278, "rms_gain_error_db": -0.04158800457010128, "output_peak": 0.290642109950153}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8247944150577843, "全帯域_10ms_rms_envelope_nrmse": 0.07823282333677457, "全帯域_10ms_envelope_correlation": 0.9917534022235143, "全帯域_10ms_modulation_mae_db": 0.1667961413172846, "全帯域_40ms_log_rms_mae_db": 0.40225806752975163, "全帯域_40ms_rms_envelope_nrmse": 0.0391490293846711, "全帯域_40ms_envelope_correlation": 0.9977291328348827, "全帯域_40ms_modulation_mae_db": 0.09047689381076233, "300–1000 Hz_10ms_log_rms_mae_db": 1.3773948674216074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09692390258949915, "300–1000 Hz_10ms_envelope_correlation": 0.991836110341454, "300–1000 Hz_10ms_modulation_mae_db": 0.19010580840899127, "300–1000 Hz_40ms_log_rms_mae_db": 0.73436666271511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05326612286270029, "300–1000 Hz_40ms_envelope_correlation": 0.9973379942424659, "300–1000 Hz_40ms_modulation_mae_db": 0.22679489879555123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2137498986999546, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11220367138125856, "1000–3000 Hz_10ms_envelope_correlation": 0.9902272123667657, "1000–3000 Hz_10ms_modulation_mae_db": 0.36152784001681343, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6840931565465931, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06247115494521909, "1000–3000 Hz_40ms_envelope_correlation": 0.9967957357351596, "1000–3000 Hz_40ms_modulation_mae_db": 0.18520895453250366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2796160803534016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14627224185971374, "3000–6000 Hz_10ms_envelope_correlation": 0.985851889593106, "3000–6000 Hz_10ms_modulation_mae_db": 0.7090039075463942, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7459011012167788, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08247783344420327, "3000–6000 Hz_40ms_envelope_correlation": 0.9954208436742742, "3000–6000 Hz_40ms_modulation_mae_db": 0.7399673005010726, "mr_stft_db": 5.438021692932554, "band_mae_db": 0.09214332512845275, "rms_gain_error_db": 0.011554886565086528, "output_peak": 0.29835473173112514}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9122815681760887, "全帯域_10ms_rms_envelope_nrmse": 0.09341943776035988, "全帯域_10ms_envelope_correlation": 0.9883489911175198, "全帯域_10ms_modulation_mae_db": 0.10715769689201649, "全帯域_40ms_log_rms_mae_db": 0.45369681352522423, "全帯域_40ms_rms_envelope_nrmse": 0.04457421649458673, "全帯域_40ms_envelope_correlation": 0.997118790422676, "全帯域_40ms_modulation_mae_db": 0.22221871775095944, "300–1000 Hz_10ms_log_rms_mae_db": 1.340143780947641, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09688329734813209, "300–1000 Hz_10ms_envelope_correlation": 0.9917515130367588, "300–1000 Hz_10ms_modulation_mae_db": 0.4395118342901994, "300–1000 Hz_40ms_log_rms_mae_db": 0.71392153010441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05417862359385247, "300–1000 Hz_40ms_envelope_correlation": 0.9971822433735993, "300–1000 Hz_40ms_modulation_mae_db": 0.7579064926083396, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2486998983934543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12115496945596112, "1000–3000 Hz_10ms_envelope_correlation": 0.9886127215351198, "1000–3000 Hz_10ms_modulation_mae_db": 0.18845715441211444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.715420904495002, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0565347339137265, "1000–3000 Hz_40ms_envelope_correlation": 0.9973956588929129, "1000–3000 Hz_40ms_modulation_mae_db": 0.23005566536921754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3072952266242082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1604054371040788, "3000–6000 Hz_10ms_envelope_correlation": 0.9830211644521365, "3000–6000 Hz_10ms_modulation_mae_db": 0.5222425005148159, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7848166382569781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0848792409667878, "3000–6000 Hz_40ms_envelope_correlation": 0.9951577292082967, "3000–6000 Hz_40ms_modulation_mae_db": 0.7840019298309366, "mr_stft_db": 5.448814776750713, "band_mae_db": 0.08094055723520616, "rms_gain_error_db": 0.0326174101112604, "output_peak": 0.32881667785511937}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8492618279693238, "全帯域_10ms_rms_envelope_nrmse": 0.07762553198498304, "全帯域_10ms_envelope_correlation": 0.9920413711281839, "全帯域_10ms_modulation_mae_db": 0.2177770729400734, "全帯域_40ms_log_rms_mae_db": 0.41110270859012377, "全帯域_40ms_rms_envelope_nrmse": 0.04394921667558982, "全帯域_40ms_envelope_correlation": 0.9974434930614547, "全帯域_40ms_modulation_mae_db": 0.1795793361049197, "300–1000 Hz_10ms_log_rms_mae_db": 1.332424287094441, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08282602564889958, "300–1000 Hz_10ms_envelope_correlation": 0.9941248033086156, "300–1000 Hz_10ms_modulation_mae_db": 0.3636925020259402, "300–1000 Hz_40ms_log_rms_mae_db": 0.6782177564571669, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049732279488358226, "300–1000 Hz_40ms_envelope_correlation": 0.9978279171675299, "300–1000 Hz_40ms_modulation_mae_db": 0.26670286100241686, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2439425787719476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11901904491964152, "1000–3000 Hz_10ms_envelope_correlation": 0.9890150258642403, "1000–3000 Hz_10ms_modulation_mae_db": 0.2973276859664131, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7079105380276618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060374017783441244, "1000–3000 Hz_40ms_envelope_correlation": 0.9971754707069951, "1000–3000 Hz_40ms_modulation_mae_db": 0.4466200309775867, "3000–6000 Hz_10ms_log_rms_mae_db": 1.291394679182128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16188306043371523, "3000–6000 Hz_10ms_envelope_correlation": 0.9820367427208773, "3000–6000 Hz_10ms_modulation_mae_db": 0.5192125879087789, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7305624751196987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08587513636466695, "3000–6000 Hz_40ms_envelope_correlation": 0.9948180972558284, "3000–6000 Hz_40ms_modulation_mae_db": 0.6459966537822471, "mr_stft_db": 5.441496959706413, "band_mae_db": 0.14777987389001007, "rms_gain_error_db": -0.13484589050846593, "output_peak": 0.29061290771878495}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.463191548240701, "spectral_convergence": 0.3753116276146516, "spectral_convergence_status": "ok", "amplitude_floor": 6.641091871383268e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.27991637146556, "spectral_convergence": 0.3633028730236156, "spectral_convergence_status": "ok", "amplitude_floor": 5.299135910844449e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.603629571550567, "spectral_convergence": 0.38314492116052606, "spectral_convergence_status": "ok", "amplitude_floor": 3.9942669914893874e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.486389753340235e-05, "output_power": 7.424893742148629e-05, "delta_db": -0.035821905935761754, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003919851371758182, "output_power": 0.0003844922193342627, "delta_db": -0.08382044521731555, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038865006012116416, "output_power": 0.00037357072263314217, "delta_db": -0.17185906721952754, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001490171424082218, "output_power": 0.0001428935956081897, "delta_db": -0.18223466546802664, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.76347124356667e-05, "output_power": 9.12202164296238e-05, "delta_db": -0.2951315315249387, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002530434948901305, "output_power": 0.0002639600968417364, "delta_db": 0.18343101851336266, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15871643897982216, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08024015736170112, "pitch": null, "output_rms_dbfs": -28.760224030771425, "output_peak": 0.2950827748031276, "reference_rms_dbfs": -28.67998387340972, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.448912497085609}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.947854524209456e-05, "log_rms_mae_db": 0.8803563901400239, "rms_envelope_nrmse": 0.08632403443803635, "envelope_correlation": 0.9899416579358108, "rms_gain_error_db": -0.07911552894539321, "reference_mean_power": 0.0013552494150092414, "output_mean_power": 0.001330784321392962}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.097356675930481e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024105831474109917, "output_power": 0.00023360316059905606, "reference_active": true, "delta_db": -0.13643401658455215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7795131277493674e-05, "output_power": 2.379879838144523e-05, "reference_active": true, "delta_db": -0.674136998019151}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0822198872325408e-05, "output_power": 9.682373086797552e-06, "reference_active": true, "delta_db": -0.4833369729636608}], "active_band_count": 3, "active_band_mae_db": 0.4313026625224546}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.267346213866395e-05, "log_rms_mae_db": 0.4553167821411913, "rms_envelope_nrmse": 0.047171587046624754, "envelope_correlation": 0.9967770794992178, "rms_gain_error_db": -0.07889355218975386, "reference_mean_power": 0.0013563897957452433, "output_mean_power": 0.0013319721939375824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.582417446963071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021668841148527672, "output_power": 0.00021086389476000845, "reference_active": true, "delta_db": -0.11833461945866064}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1881573874210085e-05, "output_power": 1.0193000431017608e-05, "reference_active": true, "delta_db": -0.6657193002259031}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.720949484930854e-07, "output_power": 5.840858403633255e-07, "reference_active": true, "delta_db": -0.6095395319789634}], "active_band_count": 3, "active_band_mae_db": 0.46453115055450905}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.299288573172057e-05, "log_rms_mae_db": 1.376745476782376, "rms_envelope_nrmse": 0.1075428263295487, "envelope_correlation": 0.9900739436785474, "rms_gain_error_db": -0.18480320855135268, "reference_mean_power": 0.0005726237314993749, "output_mean_power": 0.0005487683072980904}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0939920051603367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018116410089920085, "output_power": 0.00016912630920517274, "reference_active": true, "delta_db": -0.2986097148836171}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.90482495868649e-05, "output_power": 2.3843249113650283e-05, "reference_active": true, "delta_db": -0.8575453114107552}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.620984223411989e-06, "output_power": 3.966628258490229e-06, "reference_active": true, "delta_db": -1.5139086485231195}], "active_band_count": 3, "active_band_mae_db": 0.8900212249391638}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.559252308295269e-05, "log_rms_mae_db": 0.7581676716672091, "rms_envelope_nrmse": 0.05982225539177356, "envelope_correlation": 0.9969950671301281, "rms_gain_error_db": -0.1843249620245754, "reference_mean_power": 0.0005714831913152958, "output_mean_power": 0.0005477355953626951}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7768006225472703e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016765591938032763, "output_power": 0.00015592100622614409, "reference_active": true, "delta_db": -0.3151426273849268}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3133776830395815e-05, "output_power": 1.0939918574589688e-05, "reference_active": true, "delta_db": -0.7937554289531391}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.763861134744573e-07, "output_power": 3.0040481335344806e-07, "reference_active": true, "delta_db": -2.00252206388474}], "active_band_count": 3, "active_band_mae_db": 1.0371400400742685}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.686331455816738e-05, "log_rms_mae_db": 1.2842658269672924, "rms_envelope_nrmse": 0.12202299445911771, "envelope_correlation": 0.9889820445603761, "rms_gain_error_db": -0.29758944953429173, "reference_mean_power": 0.00020459070686192204, "output_mean_power": 0.00019104116501160912}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3670238086467597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.952541720107036e-05, "output_power": 6.384807348789493e-05, "reference_active": true, "delta_db": -0.36995805949057203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.449912227458803e-06, "output_power": 5.877820583673949e-06, "reference_active": true, "delta_db": -1.0293483065008402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5578562674598188e-06, "output_power": 1.1673096212335576e-06, "reference_active": true, "delta_db": -1.2534132076704068}], "active_band_count": 3, "active_band_mae_db": 0.8842398578872729}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4935905625855925e-05, "log_rms_mae_db": 0.7607209020766068, "rms_envelope_nrmse": 0.07871014879390889, "envelope_correlation": 0.9957711315488016, "rms_gain_error_db": -0.30046064948592976, "reference_mean_power": 0.00020499017149025147, "output_mean_power": 0.00019128766849914078}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2857171527439997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.293887842389662e-05, "output_power": 5.778964777533624e-05, "reference_active": true, "delta_db": -0.3706895200806095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.713520804403163e-06, "output_power": 2.980002554983929e-06, "reference_active": true, "delta_db": -0.9556922489316694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3334126499972412e-07, "output_power": 9.537325358959374e-08, "reference_active": true, "delta_db": -1.4553797240257187}], "active_band_count": 3, "active_band_mae_db": 0.9272538310126658}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.417692344469317e-05, "log_rms_mae_db": 1.276470864225047, "rms_envelope_nrmse": 0.15312049085147655, "envelope_correlation": 0.9864304084503189, "rms_gain_error_db": 0.397996595953686, "reference_mean_power": 0.00010545678906217645, "output_mean_power": 0.00011557774143470983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.569551408613973e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4047256300828896e-05, "output_power": 3.620003818986472e-05, "reference_active": true, "delta_db": 0.26626908674973254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.550080432854394e-06, "output_power": 6.95172628431055e-06, "reference_active": true, "delta_db": -0.35858914334123315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.516411722896473e-06, "output_power": 2.3407736201786076e-06, "reference_active": true, "delta_db": -1.7674030478191851}], "active_band_count": 3, "active_band_mae_db": 0.7974204259700503}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.848412221971719e-05, "log_rms_mae_db": 0.7749369959239061, "rms_envelope_nrmse": 0.10337609782029264, "envelope_correlation": 0.9945308710909734, "rms_gain_error_db": 0.3980353344482718, "reference_mean_power": 0.0001059674543662489, "output_mean_power": 0.00011613845250303854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.064372154195953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3923869911506206e-05, "output_power": 3.5473111876749385e-05, "reference_active": true, "delta_db": 0.193938994589701}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.204493449095079e-06, "output_power": 3.6486312825930487e-06, "reference_active": true, "delta_db": -0.6158370279295167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4577951814618017e-07, "output_power": 1.4238053279085566e-07, "reference_active": true, "delta_db": -2.3709507470704594}], "active_band_count": 3, "active_band_mae_db": 1.0602422565298923}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8668629130364772, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d06bfbd41eff832b2e1fd54d14649cbb7268b485fda80de1c8a0321520531927", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9896448690110684, "standard_deviation": 0.001190717523501733, "minimum": 0.9886127215351198, "maximum": 0.9922320185238593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2525981841986682, "standard_deviation": 0.0243801143367005, "minimum": 1.2137498986999546, "maximum": 1.2888085822735778, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4388912455433239, "standard_deviation": 0.2381172209897529, "minimum": 0.18845715441211444, "maximum": 0.8842398578872729, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11603952854778588, "standard_deviation": 0.00705473383795893, "minimum": 0.10075649420756158, "maximum": 0.12202299445911771, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969560004184659, "standard_deviation": 0.0005373412715479551, "minimum": 0.9957711315488016, "maximum": 0.9974643930803155, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7225364491694564, "standard_deviation": 0.027219740536482608, "minimum": 0.6840931565465931, "maximum": 0.7607209020766068, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.49241855560587, "standard_deviation": 0.26953931982867707, "minimum": 0.18520895453250366, "maximum": 0.9272538310126658, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06320010385098276, "standard_deviation": 0.007065537793862094, "minimum": 0.0565347339137265, "maximum": 0.07871014879390889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9834962725299363, "standard_deviation": 0.003295848139456347, "minimum": 0.9762773190954098, "maximum": 0.9864304084503189, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.300613357299016, "standard_deviation": 0.017817403434839743, "minimum": 1.276470864225047, "maximum": 1.3317642428922334, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6492983621701591, "standard_deviation": 0.2191601381305372, "minimum": 0.3397190411033817, "maximum": 0.9644778586579457, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15764744725078061, "standard_deviation": 0.014618064160820819, "minimum": 0.1427934113177815, "maximum": 0.1901625147578121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994605801025646, "standard_deviation": 0.0006805805480443867, "minimum": 0.9933392417900198, "maximum": 0.9954208436742742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7706464874972623, "standard_deviation": 0.02283183456075245, "minimum": 0.7305624751196987, "maximum": 0.7963249261304977, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8262282300883597, "standard_deviation": 0.26005045225732015, "minimum": 0.4876015630917921, "maximum": 1.2473721035503997, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09027607996872641, "standard_deviation": 0.007938835000026043, "minimum": 0.08247783344420327, "maximum": 0.10337609782029264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915302739667105, "standard_deviation": 0.0020074776771210946, "minimum": 0.9877763043686104, "maximum": 0.9941248033086156, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3560665568601384, "standard_deviation": 0.029467493865931393, "minimum": 1.3144234282986385, "maximum": 1.393852933125597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.41255850123411686, "standard_deviation": 0.22619201872041897, "minimum": 0.19010580840899127, "maximum": 0.8900212249391638, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0979612271818508, "standard_deviation": 0.011352514307417245, "minimum": 0.08282602564889958, "maximum": 0.1178649858871829, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971721423212248, "standard_deviation": 0.0009098082827106933, "minimum": 0.9951737374986867, "maximum": 0.9982995670696841, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.72391862171471, "standard_deviation": 0.03270741383679848, "minimum": 0.6782177564571669, "maximum": 0.7762165886067628, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5459850655407499, "standard_deviation": 0.3238288954960637, "minimum": 0.2149883527556514, "maximum": 1.0371400400742685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054633030531280366, "standard_deviation": 0.00820538027145869, "minimum": 0.04235001669555565, "maximum": 0.07086167029975145, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11539966337655214, "standard_deviation": 0.0417060462556669, "minimum": 0.05385946995229074, "maximum": 0.17205154058542851, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.444005576798328, "standard_deviation": 0.006471933520879926, "minimum": 5.435671938220402, "maximum": 5.453850236230234, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2983072097053205, "standard_deviation": 0.01308829962776825, "minimum": 0.288324165400333, "maximum": 0.32881667785511937, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03520638781534997, "standard_deviation": 0.05494718328529703, "minimum": -0.13484589050846593, "maximum": 0.0326174101112604, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901534834921835, "standard_deviation": 0.001382753649206884, "minimum": 0.9883489911175198, "maximum": 0.9920413711281839, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8679709479996476, "standard_deviation": 0.02900537462580366, "minimum": 0.8247944150577843, "maximum": 0.9122815681760887, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.270429666178712, "standard_deviation": 0.10869613400711416, "minimum": 0.10715769689201649, "maximum": 0.4313026625224546, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08544600263624515, "standard_deviation": 0.006005855311476428, "minimum": 0.07762553198498304, "maximum": 0.09341943776035988, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971177106133595, "standard_deviation": 0.0005201478161553587, "minimum": 0.9964014847185224, "maximum": 0.9977291328348827, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4354691462272595, "standard_deviation": 0.030420501839841903, "minimum": 0.39651119066179363, "maximum": 0.4767112518873481, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.34620542938293863, "standard_deviation": 0.21697368854638505, "minimum": 0.09047689381076233, "maximum": 0.7558810014688375, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04445942846074123, "standard_deviation": 0.003951419535684575, "minimum": 0.0391490293846711, "maximum": 0.04971088715002501, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0006", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.984668493270874, "predicted_mos": 1.984668493270874, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ca429c51d7ff085ec889158424e8fcd49283b052ddeddd92e64eaa656eb6f633", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9254512996412814, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8925823617314199, "全帯域_10ms_rms_envelope_nrmse": 0.08682980010154777, "全帯域_10ms_envelope_correlation": 0.9898255716583284, "全帯域_10ms_modulation_mae_db": 0.4371615568793788, "全帯域_40ms_log_rms_mae_db": 0.46314862266974294, "全帯域_40ms_rms_envelope_nrmse": 0.0492017331556606, "全帯域_40ms_envelope_correlation": 0.9964784805150247, "全帯域_40ms_modulation_mae_db": 0.48240792708685815, "300–1000 Hz_10ms_log_rms_mae_db": 1.3798763183810236, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10946638846824232, "300–1000 Hz_10ms_envelope_correlation": 0.9896346096326967, "300–1000 Hz_10ms_modulation_mae_db": 0.9245897979555971, "300–1000 Hz_40ms_log_rms_mae_db": 0.7376641808274147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06362103139072818, "300–1000 Hz_40ms_envelope_correlation": 0.9964420151660356, "300–1000 Hz_40ms_modulation_mae_db": 1.078589684571484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2674028940258388, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12148517839243266, "1000–3000 Hz_10ms_envelope_correlation": 0.9891707739835437, "1000–3000 Hz_10ms_modulation_mae_db": 0.9143439064332752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7175100594712678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07851931192785921, "1000–3000 Hz_40ms_envelope_correlation": 0.9958927988711948, "1000–3000 Hz_40ms_modulation_mae_db": 0.9666252621493613, "3000–6000 Hz_10ms_log_rms_mae_db": 1.286199075143695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15004499138348973, "3000–6000 Hz_10ms_envelope_correlation": 0.9864499366509657, "3000–6000 Hz_10ms_modulation_mae_db": 0.7503952435214668, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7470696698329455, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0943627938641635, "3000–6000 Hz_40ms_envelope_correlation": 0.995196237873111, "3000–6000 Hz_40ms_modulation_mae_db": 1.0011414859402958, "mr_stft_db": 5.437969437087449, "band_mae_db": 0.17482139820984577, "rms_gain_error_db": -0.0779648525353272, "output_peak": 0.2963287565181182}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8911521951366982, "全帯域_10ms_rms_envelope_nrmse": 0.0815204269348409, "全帯域_10ms_envelope_correlation": 0.9910417894498039, "全帯域_10ms_modulation_mae_db": 0.28900648423612313, "全帯域_40ms_log_rms_mae_db": 0.46378563946375595, "全帯域_40ms_rms_envelope_nrmse": 0.042783620685268454, "全帯域_40ms_envelope_correlation": 0.9973100439074969, "全帯域_40ms_modulation_mae_db": 0.38112821702248384, "300–1000 Hz_10ms_log_rms_mae_db": 1.3142440735116632, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0856830402432234, "300–1000 Hz_10ms_envelope_correlation": 0.9935617876578433, "300–1000 Hz_10ms_modulation_mae_db": 0.5071193413623937, "300–1000 Hz_40ms_log_rms_mae_db": 0.670713983750307, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04281915079782713, "300–1000 Hz_40ms_envelope_correlation": 0.998264166934312, "300–1000 Hz_40ms_modulation_mae_db": 0.4809615376697649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2511784312169774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12027468652669787, "1000–3000 Hz_10ms_envelope_correlation": 0.988786727627462, "1000–3000 Hz_10ms_modulation_mae_db": 0.24870536711193722, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7138339062848096, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058583340842101296, "1000–3000 Hz_40ms_envelope_correlation": 0.9972846420487289, "1000–3000 Hz_40ms_modulation_mae_db": 0.6903232880702124, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3272767134846808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15140262867885648, "3000–6000 Hz_10ms_envelope_correlation": 0.9845923357867243, "3000–6000 Hz_10ms_modulation_mae_db": 0.5554719366717015, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7878874123879887, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0910524845560977, "3000–6000 Hz_40ms_envelope_correlation": 0.9945961563378384, "3000–6000 Hz_40ms_modulation_mae_db": 0.6936865902747161, "mr_stft_db": 5.420777052730177, "band_mae_db": 0.07717960684695252, "rms_gain_error_db": -0.047969108069618974, "output_peak": 0.3018098096460492}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9078452078730005, "全帯域_10ms_rms_envelope_nrmse": 0.0907438154810237, "全帯域_10ms_envelope_correlation": 0.9889345496849992, "全帯域_10ms_modulation_mae_db": 0.2187017132261706, "全帯域_40ms_log_rms_mae_db": 0.49068341781119457, "全帯域_40ms_rms_envelope_nrmse": 0.049829122596567174, "全帯域_40ms_envelope_correlation": 0.9963313125336898, "全帯域_40ms_modulation_mae_db": 0.21006073660998778, "300–1000 Hz_10ms_log_rms_mae_db": 1.3075624449443128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09272129658118065, "300–1000 Hz_10ms_envelope_correlation": 0.9924403597522601, "300–1000 Hz_10ms_modulation_mae_db": 0.21468902230170614, "300–1000 Hz_40ms_log_rms_mae_db": 0.6802472763268106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05214074385923907, "300–1000 Hz_40ms_envelope_correlation": 0.9974026919580127, "300–1000 Hz_40ms_modulation_mae_db": 0.23304017516440476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2240238152646015, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1130123817818433, "1000–3000 Hz_10ms_envelope_correlation": 0.9903225061979997, "1000–3000 Hz_10ms_modulation_mae_db": 0.4531349859433412, "1000–3000 Hz_40ms_log_rms_mae_db": 0.729512728933606, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06634732477300537, "1000–3000 Hz_40ms_envelope_correlation": 0.9968154921179371, "1000–3000 Hz_40ms_modulation_mae_db": 0.9113612988613066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.304762180096979, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18626989409922587, "3000–6000 Hz_10ms_envelope_correlation": 0.9768153546644313, "3000–6000 Hz_10ms_modulation_mae_db": 0.4391220251122188, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7504794597158334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09905343479261472, "3000–6000 Hz_40ms_envelope_correlation": 0.9932155217661509, "3000–6000 Hz_40ms_modulation_mae_db": 0.6031660990175792, "mr_stft_db": 5.4325315545814945, "band_mae_db": 0.18986512224486185, "rms_gain_error_db": 0.016819236372759193, "output_peak": 0.28778822273887217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8529208518961813, "全帯域_10ms_rms_envelope_nrmse": 0.08923225345081305, "全帯域_10ms_envelope_correlation": 0.9893109733998825, "全帯域_10ms_modulation_mae_db": 0.39345093605713427, "全帯域_40ms_log_rms_mae_db": 0.4289197999189669, "全帯域_40ms_rms_envelope_nrmse": 0.049445748912848024, "全帯域_40ms_envelope_correlation": 0.9964500445547595, "全帯域_40ms_modulation_mae_db": 0.5098650937081249, "300–1000 Hz_10ms_log_rms_mae_db": 1.3907808786548674, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11240612497422448, "300–1000 Hz_10ms_envelope_correlation": 0.988895777895606, "300–1000 Hz_10ms_modulation_mae_db": 0.2764543056709272, "300–1000 Hz_40ms_log_rms_mae_db": 0.7735184666820395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06669963714611063, "300–1000 Hz_40ms_envelope_correlation": 0.995729283121032, "300–1000 Hz_40ms_modulation_mae_db": 0.4377671056105014, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2412892265335678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10626127338117261, "1000–3000 Hz_10ms_envelope_correlation": 0.9914021081894119, "1000–3000 Hz_10ms_modulation_mae_db": 0.5141490695952263, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7160410801993592, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0637103102315254, "1000–3000 Hz_40ms_envelope_correlation": 0.9969857860941946, "1000–3000 Hz_40ms_modulation_mae_db": 0.33670599251009875, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3288424619732278, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14439204974416894, "3000–6000 Hz_10ms_envelope_correlation": 0.9858953268580263, "3000–6000 Hz_10ms_modulation_mae_db": 1.0139516272348725, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7834367460113568, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08413300770848008, "3000–6000 Hz_40ms_envelope_correlation": 0.9952749901843086, "3000–6000 Hz_40ms_modulation_mae_db": 1.2094469179210958, "mr_stft_db": 5.422606598136954, "band_mae_db": 0.12325642355005206, "rms_gain_error_db": -0.05510745478367124, "output_peak": 0.29649944101965964}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8855936159162074, "全帯域_10ms_rms_envelope_nrmse": 0.08745110842704004, "全帯域_10ms_envelope_correlation": 0.9896392212973825, "全帯域_10ms_modulation_mae_db": 0.2928651577944048, "全帯域_40ms_log_rms_mae_db": 0.4072459471480773, "全帯域_40ms_rms_envelope_nrmse": 0.04008849989542764, "全帯域_40ms_envelope_correlation": 0.9976355676817897, "全帯域_40ms_modulation_mae_db": 0.7148598798121353, "300–1000 Hz_10ms_log_rms_mae_db": 1.3774750829241755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10426833199674888, "300–1000 Hz_10ms_envelope_correlation": 0.9904734290482043, "300–1000 Hz_10ms_modulation_mae_db": 0.4431369494541282, "300–1000 Hz_40ms_log_rms_mae_db": 0.7143123248090665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05487157474500766, "300–1000 Hz_40ms_envelope_correlation": 0.9971225394771711, "300–1000 Hz_40ms_modulation_mae_db": 0.901519135304912, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2937329902769428, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12219947206645423, "1000–3000 Hz_10ms_envelope_correlation": 0.9885229943259004, "1000–3000 Hz_10ms_modulation_mae_db": 0.6602310701481144, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6629017263351272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06352756795756166, "1000–3000 Hz_40ms_envelope_correlation": 0.9970189860817308, "1000–3000 Hz_40ms_modulation_mae_db": 0.4174872274712893, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3199901835982075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14862138695767008, "3000–6000 Hz_10ms_envelope_correlation": 0.9849014608049361, "3000–6000 Hz_10ms_modulation_mae_db": 1.165827999193842, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7898373418928663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08367126115347055, "3000–6000 Hz_40ms_envelope_correlation": 0.9949690716789812, "3000–6000 Hz_40ms_modulation_mae_db": 1.443722666303996, "mr_stft_db": 5.431571608094544, "band_mae_db": 0.1230332657857705, "rms_gain_error_db": -0.049934878579903696, "output_peak": 0.2888811140507918}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8363964246021884, "全帯域_10ms_rms_envelope_nrmse": 0.07939125226141165, "全帯域_10ms_envelope_correlation": 0.9915115629476127, "全帯域_10ms_modulation_mae_db": 0.21190409414481046, "全帯域_40ms_log_rms_mae_db": 0.41207660911491606, "全帯域_40ms_rms_envelope_nrmse": 0.03996546034474609, "全帯域_40ms_envelope_correlation": 0.9976358888756242, "全帯域_40ms_modulation_mae_db": 0.12156687300472914, "300–1000 Hz_10ms_log_rms_mae_db": 1.382449128542062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09995163223910611, "300–1000 Hz_10ms_envelope_correlation": 0.9913727256899315, "300–1000 Hz_10ms_modulation_mae_db": 0.19057866104497836, "300–1000 Hz_40ms_log_rms_mae_db": 0.7218121916007918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05748764767341691, "300–1000 Hz_40ms_envelope_correlation": 0.9969394545521175, "300–1000 Hz_40ms_modulation_mae_db": 0.36051193478016447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1877829389990726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10805817283597713, "1000–3000 Hz_10ms_envelope_correlation": 0.990923316938541, "1000–3000 Hz_10ms_modulation_mae_db": 0.31851326949877995, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6460523126693289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05609404961425531, "1000–3000 Hz_40ms_envelope_correlation": 0.9974411014238527, "1000–3000 Hz_40ms_modulation_mae_db": 0.3421299544115955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2957391116901404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1455836735985283, "3000–6000 Hz_10ms_envelope_correlation": 0.9858208578093183, "3000–6000 Hz_10ms_modulation_mae_db": 0.8127422060761198, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7387473211279619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07857712955675479, "3000–6000 Hz_40ms_envelope_correlation": 0.9957493538634421, "3000–6000 Hz_40ms_modulation_mae_db": 0.8128366831292677, "mr_stft_db": 5.421812682349454, "band_mae_db": 0.09689855328836482, "rms_gain_error_db": 0.013743563143836372, "output_peak": 0.2962974481520127}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9113822161409836, "全帯域_10ms_rms_envelope_nrmse": 0.0956995419080209, "全帯域_10ms_envelope_correlation": 0.9877919029023208, "全帯域_10ms_modulation_mae_db": 0.10813977731948336, "全帯域_40ms_log_rms_mae_db": 0.4589320004345368, "全帯域_40ms_rms_envelope_nrmse": 0.045741476553474214, "全帯域_40ms_envelope_correlation": 0.9969781877483606, "全帯域_40ms_modulation_mae_db": 0.21931639414895254, "300–1000 Hz_10ms_log_rms_mae_db": 1.3455479936895738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09879652209524795, "300–1000 Hz_10ms_envelope_correlation": 0.9914333571312551, "300–1000 Hz_10ms_modulation_mae_db": 0.39722922530584653, "300–1000 Hz_40ms_log_rms_mae_db": 0.7046388290344833, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054162851044817516, "300–1000 Hz_40ms_envelope_correlation": 0.9971896931472778, "300–1000 Hz_40ms_modulation_mae_db": 0.7019987960383639, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2512300328257082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12328394087534697, "1000–3000 Hz_10ms_envelope_correlation": 0.9882454533099828, "1000–3000 Hz_10ms_modulation_mae_db": 0.10942872908712416, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7008415440434289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05630024124304515, "1000–3000 Hz_40ms_envelope_correlation": 0.9974581470801512, "1000–3000 Hz_40ms_modulation_mae_db": 0.2645062281847799, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3250983364378994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1608302984556494, "3000–6000 Hz_10ms_envelope_correlation": 0.9826544844162639, "3000–6000 Hz_10ms_modulation_mae_db": 0.48509439643870017, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8042863746266065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0856535486095625, "3000–6000 Hz_40ms_envelope_correlation": 0.9948757682456928, "3000–6000 Hz_40ms_modulation_mae_db": 0.7836683577843152, "mr_stft_db": 5.432288182393605, "band_mae_db": 0.09713061844235955, "rms_gain_error_db": 0.04506024216924591, "output_peak": 0.33028421698974947}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8702090999144441, "全帯域_10ms_rms_envelope_nrmse": 0.08035152654203462, "全帯域_10ms_envelope_correlation": 0.9914967815840651, "全帯域_10ms_modulation_mae_db": 0.21894787201086938, "全帯域_40ms_log_rms_mae_db": 0.4185461273160113, "全帯域_40ms_rms_envelope_nrmse": 0.046129984488208375, "全帯域_40ms_envelope_correlation": 0.9972137889633661, "全帯域_40ms_modulation_mae_db": 0.1769278099178988, "300–1000 Hz_10ms_log_rms_mae_db": 1.3143294253556088, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08571465536224479, "300–1000 Hz_10ms_envelope_correlation": 0.9936853026284573, "300–1000 Hz_10ms_modulation_mae_db": 0.3479341329317734, "300–1000 Hz_40ms_log_rms_mae_db": 0.651831086869671, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05292090604403092, "300–1000 Hz_40ms_envelope_correlation": 0.9975032868939142, "300–1000 Hz_40ms_modulation_mae_db": 0.2626138194977647, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2419792473533906, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1218207013315014, "1000–3000 Hz_10ms_envelope_correlation": 0.9884828102341364, "1000–3000 Hz_10ms_modulation_mae_db": 0.18386774381066848, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6942674882015218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05550408032820806, "1000–3000 Hz_40ms_envelope_correlation": 0.997642577742174, "1000–3000 Hz_40ms_modulation_mae_db": 0.3947632188094712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2806289049517583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1563311717190911, "3000–6000 Hz_10ms_envelope_correlation": 0.9834391980937928, "3000–6000 Hz_10ms_modulation_mae_db": 0.6690969742907337, "3000–6000 Hz_40ms_log_rms_mae_db": 0.723648495314619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08374455783777933, "3000–6000 Hz_40ms_envelope_correlation": 0.9954213482228426, "3000–6000 Hz_40ms_modulation_mae_db": 0.9318376147497189, "mr_stft_db": 5.426325617794233, "band_mae_db": 0.1616687318620266, "rms_gain_error_db": -0.14697038133926232, "output_peak": 0.28650423255311963}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.455842631517663, "spectral_convergence": 0.37500325600983364, "spectral_convergence_status": "ok", "amplitude_floor": 6.641091871383268e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.267786884508974, "spectral_convergence": 0.36404057441319804, "spectral_convergence_status": "ok", "amplitude_floor": 5.299135910844449e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.59027879523571, "spectral_convergence": 0.3828171020705783, "spectral_convergence_status": "ok", "amplitude_floor": 3.9942669914893874e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.486389753340235e-05, "output_power": 7.332972350655325e-05, "delta_db": -0.08992385958146867, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003919851371758182, "output_power": 0.0003835872203476272, "delta_db": -0.09405470557119244, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038865006012116416, "output_power": 0.00037686631777781703, "delta_db": -0.13371414872604112, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001490171424082218, "output_power": 0.00014225760085163355, "delta_db": -0.2016075086985988, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.76347124356667e-05, "output_power": 9.039019078113132e-05, "delta_db": -0.33482948435094745, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002530434948901305, "output_power": 0.0002646519177710302, "delta_db": 0.19479868233082617, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17482139820984577, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0779648525353272, "pitch": null, "output_rms_dbfs": -28.75794872594505, "output_peak": 0.2963287565181182, "reference_rms_dbfs": -28.67998387340972, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.437969437087449}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.947854524209456e-05, "log_rms_mae_db": 0.8925823617314199, "rms_envelope_nrmse": 0.08682980010154777, "envelope_correlation": 0.9898255716583284, "rms_gain_error_db": -0.07692978427747099, "reference_mean_power": 0.0013552494150092414, "output_mean_power": 0.0013314542554922666}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.097356675930481e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024105831474109917, "output_power": 0.00023544029967991284, "reference_active": true, "delta_db": -0.10241314163643844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7795131277493674e-05, "output_power": 2.3551620579312976e-05, "reference_active": true, "delta_db": -0.7194793341483838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0822198872325408e-05, "output_power": 9.668437428202499e-06, "reference_active": true, "delta_db": -0.4895921948533142}], "active_band_count": 3, "active_band_mae_db": 0.4371615568793788}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.267346213866395e-05, "log_rms_mae_db": 0.46314862266974294, "rms_envelope_nrmse": 0.0492017331556606, "envelope_correlation": 0.9964784805150247, "rms_gain_error_db": -0.07645642122103377, "reference_mean_power": 0.0013563897957452433, "output_mean_power": 0.0013327198667314288}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.582417446963071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021668841148527672, "output_power": 0.00021261536457735288, "reference_active": true, "delta_db": -0.08241040369893379}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1881573874210085e-05, "output_power": 1.0119564637802823e-05, "reference_active": true, "delta_db": -0.6971214383703194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.720949484930854e-07, "output_power": 5.763170091177381e-07, "reference_active": true, "delta_db": -0.6676919391913212}], "active_band_count": 3, "active_band_mae_db": 0.48240792708685815}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.299288573172057e-05, "log_rms_mae_db": 1.3798763183810236, "rms_envelope_nrmse": 0.10946638846824232, "envelope_correlation": 0.9896346096326967, "rms_gain_error_db": -0.16946928257753435, "reference_mean_power": 0.0005726237314993749, "output_mean_power": 0.0005507093048818868}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0939920051603367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018116410089920085, "output_power": 0.00016999837168509083, "reference_active": true, "delta_db": -0.2762738143103382}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.90482495868649e-05, "output_power": 2.4057313181203034e-05, "reference_active": true, "delta_db": -0.8187284550921871}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.620984223411989e-06, "output_power": 3.818876852640162e-06, "reference_active": true, "delta_db": -1.6787671244642663}], "active_band_count": 3, "active_band_mae_db": 0.9245897979555971}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.559252308295269e-05, "log_rms_mae_db": 0.7376641808274147, "rms_envelope_nrmse": 0.06362103139072818, "envelope_correlation": 0.9964420151660356, "rms_gain_error_db": -0.16879611289962834, "reference_mean_power": 0.0005714831913152958, "output_mean_power": 0.0005496976116038349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7768006225472703e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016765591938032763, "output_power": 0.0001569234031027026, "reference_active": true, "delta_db": -0.2873117370774644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3133776830395815e-05, "output_power": 1.1066389623678461e-05, "reference_active": true, "delta_db": -0.7438367574887839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.763861134744573e-07, "output_power": 2.867457815509611e-07, "reference_active": true, "delta_db": -2.2046205591482035}], "active_band_count": 3, "active_band_mae_db": 1.078589684571484}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.686331455816738e-05, "log_rms_mae_db": 1.2674028940258388, "rms_envelope_nrmse": 0.12148517839243266, "envelope_correlation": 0.9891707739835437, "rms_gain_error_db": -0.31800248408473464, "reference_mean_power": 0.00020459070686192204, "output_mean_power": 0.00019014532601178624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3670238086467597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.952541720107036e-05, "output_power": 6.380313672655041e-05, "reference_active": true, "delta_db": -0.37301573332850857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.449912227458803e-06, "output_power": 5.788505229233808e-06, "reference_active": true, "delta_db": -1.0958472609511456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5578562674598188e-06, "output_power": 1.1617442011069332e-06, "reference_active": true, "delta_db": -1.2741687250201714}], "active_band_count": 3, "active_band_mae_db": 0.9143439064332752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4935905625855925e-05, "log_rms_mae_db": 0.7175100594712678, "rms_envelope_nrmse": 0.07851931192785921, "envelope_correlation": 0.9958927988711948, "rms_gain_error_db": -0.3201268693051585, "reference_mean_power": 0.00020499017149025147, "output_mean_power": 0.0001904234160642578}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2857171527439997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.293887842389662e-05, "output_power": 5.776644885408358e-05, "reference_active": true, "delta_db": -0.37243329023473665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.713520804403163e-06, "output_power": 2.9433836849226833e-06, "reference_active": true, "delta_db": -1.0093898305543343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3334126499972412e-07, "output_power": 9.400680754468215e-08, "reference_active": true, "delta_db": -1.5180526656590128}], "active_band_count": 3, "active_band_mae_db": 0.9666252621493613}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.417692344469317e-05, "log_rms_mae_db": 1.286199075143695, "rms_envelope_nrmse": 0.15004499138348973, "envelope_correlation": 0.9864499366509657, "rms_gain_error_db": 0.32456720268814854, "reference_mean_power": 0.00010545678906217645, "output_mean_power": 0.0001136400102205304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.569551408613973e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4047256300828896e-05, "output_power": 3.593850959348995e-05, "reference_active": true, "delta_db": 0.23477942487942982}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.550080432854394e-06, "output_power": 6.77785131602e-06, "reference_active": true, "delta_db": -0.4685954069965018}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.516411722896473e-06, "output_power": 2.462173367330387e-06, "reference_active": true, "delta_db": -1.547810898688469}], "active_band_count": 3, "active_band_mae_db": 0.7503952435214668}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.848412221971719e-05, "log_rms_mae_db": 0.7470696698329455, "rms_envelope_nrmse": 0.0943627938641635, "envelope_correlation": 0.995196237873111, "rms_gain_error_db": 0.3246509908719025, "reference_mean_power": 0.0001059674543662489, "output_mean_power": 0.00011419250514755712}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.064372154195953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3923869911506206e-05, "output_power": 3.527167534156024e-05, "reference_active": true, "delta_db": 0.1692069922613615}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.204493449095079e-06, "output_power": 3.5640998887134376e-06, "reference_active": true, "delta_db": -0.7176381292941298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4577951814618017e-07, "output_power": 1.509689918232645e-07, "reference_active": true, "delta_db": -2.1165793362653966}], "active_band_count": 3, "active_band_mae_db": 1.0011414859402958}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8675282445878024, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1cba1a4465c2d66dedf912527b7368bc7a42003724b21fc86b26d3a1dcfa4949", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9894820863508722, "standard_deviation": 0.0012246308877537518, "minimum": 0.9882454533099828, "maximum": 0.9914021081894119, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2448274470620124, "standard_deviation": 0.030915158808665855, "minimum": 1.1877829389990726, "maximum": 1.2937329902769428, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4252967677035584, "standard_deviation": 0.2682426294980173, "minimum": 0.10942872908712416, "maximum": 0.9143439064332752, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11704947589892828, "standard_deviation": 0.006884350987614296, "minimum": 0.10626127338117261, "maximum": 0.12328394087534697, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970674414324955, "standard_deviation": 0.000550650225236406, "minimum": 0.9958927988711948, "maximum": 0.997642577742174, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6976201057673062, "standard_deviation": 0.029034930100343647, "minimum": 0.6460523126693289, "maximum": 0.729512728933606, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5404878088085143, "standard_deviation": 0.2764557429441035, "minimum": 0.2645062281847799, "maximum": 0.9666252621493613, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06232327836469519, "standard_deviation": 0.00772239308788109, "minimum": 0.05550408032820806, "maximum": 0.07851931192785921, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9838211193855573, "standard_deviation": 0.0031067098648278673, "minimum": 0.9768153546644313, "maximum": 0.9864499366509657, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3085671209220735, "standard_deviation": 0.019364232621059765, "minimum": 1.2806289049517583, "maximum": 1.3288424619732278, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.736462801067457, "standard_deviation": 0.2553039203141435, "minimum": 0.4391220251122188, "maximum": 1.165827999193842, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15543451182958498, "standard_deviation": 0.013582741032275162, "minimum": 0.14439204974416894, "maximum": 0.18626989409922587, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9949123060215459, "standard_deviation": 0.0007704836079992841, "minimum": 0.9932155217661509, "maximum": 0.9957493538634421, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7656741026137723, "standard_deviation": 0.029159457265148896, "minimum": 0.723648495314619, "maximum": 0.8042863746266065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9349383018901232, "standard_deviation": 0.27894586516190134, "minimum": 0.6031660990175792, "maximum": 1.443722666303996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08753102725986539, "standard_deviation": 0.006722228200824546, "minimum": 0.07857712955675479, "maximum": 0.09905343479261472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9914371686795318, "standard_deviation": 0.0017409091270886871, "minimum": 0.988895777895606, "maximum": 0.9936853026284573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3515331682504108, "standard_deviation": 0.03528726675125044, "minimum": 1.3075624449443128, "maximum": 1.3907808786548674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4127164295034188, "standard_deviation": 0.23403774761018123, "minimum": 0.19057866104497836, "maximum": 0.9245897979555971, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09862599899502733, "standard_deviation": 0.010072002960910613, "minimum": 0.0856830402432234, "maximum": 0.11240612497422448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970741414062341, "standard_deviation": 0.0007512213474233455, "minimum": 0.995729283121032, "maximum": 0.998264166934312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.706842292487573, "standard_deviation": 0.03913208008037695, "minimum": 0.651831086869671, "maximum": 0.7735184666820395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.55712527357967, "standard_deviation": 0.30752301678260574, "minimum": 0.23304017516440476, "maximum": 1.078589684571484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05559044283764725, "standard_deviation": 0.007330868947159402, "minimum": 0.04281915079782713, "maximum": 0.06669963714611063, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1304817150287792, "standard_deviation": 0.04080257334645632, "minimum": 0.07717960684695252, "maximum": 0.18986512224486185, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.428235341645989, "standard_deviation": 0.006244685469186702, "minimum": 5.420777052730177, "maximum": 5.437969437087449, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2980491552085466, "standard_deviation": 0.01406259753544205, "minimum": 0.28650423255311963, "maximum": 0.33028421698974947, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03779045420274274, "standard_deviation": 0.06162956208090407, "minimum": -0.14697038133926232, "maximum": 0.04506024216924591, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899440441155494, "standard_deviation": 0.001321885869587319, "minimum": 0.9877919029023208, "maximum": 0.9915115629476127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8810102466513905, "standard_deviation": 0.02617639040146326, "minimum": 0.8363964246021884, "maximum": 0.9113822161409836, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2712721989585469, "standard_deviation": 0.106144062235954, "minimum": 0.10813977731948336, "maximum": 0.4371615568793788, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08640246563834159, "standard_deviation": 0.00565910379064024, "minimum": 0.07939125226141165, "maximum": 0.0956995419080209, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970041643475139, "standard_deviation": 0.0005307740194780214, "minimum": 0.9963313125336898, "maximum": 0.9976358888756242, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44291727048465024, "standard_deviation": 0.03022852093343707, "minimum": 0.4072459471480773, "maximum": 0.49068341781119457, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3520166164138963, "standard_deviation": 0.20561571297801715, "minimum": 0.12156687300472914, "maximum": 0.7148598798121353, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04539820582902507, "standard_deviation": 0.004065175086885438, "minimum": 0.03996546034474609, "maximum": 0.049829122596567174, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0006", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.037497043609619, "predicted_mos": 2.037497043609619, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d20c4e997284b832db42f46ed72d1c7fdb17770fad229fff1dca085f50494c89", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.948435800150037, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.834947694005919, "全帯域_10ms_rms_envelope_nrmse": 0.07660192471019905, "全帯域_10ms_envelope_correlation": 0.9920761960609149, "全帯域_10ms_modulation_mae_db": 0.37745079636359974, "全帯域_40ms_log_rms_mae_db": 0.42584687199800586, "全帯域_40ms_rms_envelope_nrmse": 0.040852111210426, "全帯域_40ms_envelope_correlation": 0.9975394590573506, "全帯域_40ms_modulation_mae_db": 0.522372938354119, "300–1000 Hz_10ms_log_rms_mae_db": 1.6277469773346098, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09969113314035342, "300–1000 Hz_10ms_envelope_correlation": 0.9912847029713261, "300–1000 Hz_10ms_modulation_mae_db": 0.3649272443131404, "300–1000 Hz_40ms_log_rms_mae_db": 0.9320214105411326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05426910074443945, "300–1000 Hz_40ms_envelope_correlation": 0.997235153740811, "300–1000 Hz_40ms_modulation_mae_db": 0.5662879614665024, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4965948571116476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11235539407231904, "1000–3000 Hz_10ms_envelope_correlation": 0.9902665415790632, "1000–3000 Hz_10ms_modulation_mae_db": 0.7914775420606771, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8537325757672063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06136188017925685, "1000–3000 Hz_40ms_envelope_correlation": 0.9970850442187477, "1000–3000 Hz_40ms_modulation_mae_db": 0.8328508124364223, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6289900080114286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17355549622712738, "3000–6000 Hz_10ms_envelope_correlation": 0.9803512649946299, "3000–6000 Hz_10ms_modulation_mae_db": 1.126816878749026, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9840774753496352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09760723968759252, "3000–6000 Hz_40ms_envelope_correlation": 0.9937602040506281, "3000–6000 Hz_40ms_modulation_mae_db": 1.329063058842693, "mr_stft_db": 5.661080873982331, "band_mae_db": 0.060599378967268636, "rms_gain_error_db": -0.03793518354445511, "output_peak": 0.29611675471725346}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8469230857383169, "全帯域_10ms_rms_envelope_nrmse": 0.07884905066659864, "全帯域_10ms_envelope_correlation": 0.9916055602420656, "全帯域_10ms_modulation_mae_db": 0.2160541658477075, "全帯域_40ms_log_rms_mae_db": 0.40856702887927465, "全帯域_40ms_rms_envelope_nrmse": 0.04065397384297084, "全帯域_40ms_envelope_correlation": 0.997604868892794, "全帯域_40ms_modulation_mae_db": 0.22936160398171337, "300–1000 Hz_10ms_log_rms_mae_db": 1.6049036009234816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09017140044979532, "300–1000 Hz_10ms_envelope_correlation": 0.9929132452902665, "300–1000 Hz_10ms_modulation_mae_db": 0.22190310337285915, "300–1000 Hz_40ms_log_rms_mae_db": 0.9038250137847456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049599043208738765, "300–1000 Hz_40ms_envelope_correlation": 0.9976838451295883, "300–1000 Hz_40ms_modulation_mae_db": 0.44443212212202693, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6183631761426405, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12913080626277176, "1000–3000 Hz_10ms_envelope_correlation": 0.9870149454042041, "1000–3000 Hz_10ms_modulation_mae_db": 0.37645520103824986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9973438994053816, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07737562333926222, "1000–3000 Hz_40ms_envelope_correlation": 0.9951139655687223, "1000–3000 Hz_40ms_modulation_mae_db": 0.707516189945045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7053665275651377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1743424771079924, "3000–6000 Hz_10ms_envelope_correlation": 0.9792737467049492, "3000–6000 Hz_10ms_modulation_mae_db": 0.5645004975829951, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9776217372639441, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08713575793892914, "3000–6000 Hz_40ms_envelope_correlation": 0.9945626273060268, "3000–6000 Hz_40ms_modulation_mae_db": 0.8012853098583784, "mr_stft_db": 5.662501096628664, "band_mae_db": 0.13052435291040365, "rms_gain_error_db": -0.06919030370691326, "output_peak": 0.2879929076899897}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8113502996326734, "全帯域_10ms_rms_envelope_nrmse": 0.08737710422531009, "全帯域_10ms_envelope_correlation": 0.9897121125060024, "全帯域_10ms_modulation_mae_db": 0.49234386558766285, "全帯域_40ms_log_rms_mae_db": 0.3920955912643989, "全帯域_40ms_rms_envelope_nrmse": 0.044873542622574335, "全帯域_40ms_envelope_correlation": 0.9970394924105432, "全帯域_40ms_modulation_mae_db": 0.6475143074119177, "300–1000 Hz_10ms_log_rms_mae_db": 1.6139981495695424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11285083003688416, "300–1000 Hz_10ms_envelope_correlation": 0.9888637408573753, "300–1000 Hz_10ms_modulation_mae_db": 0.7982871658235989, "300–1000 Hz_40ms_log_rms_mae_db": 0.9499748992737954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060426787720022725, "300–1000 Hz_40ms_envelope_correlation": 0.9965209298969561, "300–1000 Hz_40ms_modulation_mae_db": 1.0107087241706165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.56522303042739, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12184717183648423, "1000–3000 Hz_10ms_envelope_correlation": 0.9888715466681016, "1000–3000 Hz_10ms_modulation_mae_db": 0.601213333731388, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9173935476809012, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0688412543888419, "1000–3000 Hz_40ms_envelope_correlation": 0.996810718705101, "1000–3000 Hz_40ms_modulation_mae_db": 0.8645992777198659, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6805556440539637, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17925022100638524, "3000–6000 Hz_10ms_envelope_correlation": 0.9801479795029483, "3000–6000 Hz_10ms_modulation_mae_db": 0.6680188013959848, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0309096987902109, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11650186927163825, "3000–6000 Hz_40ms_envelope_correlation": 0.9919417923459685, "3000–6000 Hz_40ms_modulation_mae_db": 0.7426074390080516, "mr_stft_db": 5.650900653175678, "band_mae_db": 0.12890707089788642, "rms_gain_error_db": -0.026103108020459592, "output_peak": 0.3289882835654262}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9116187325379492, "全帯域_10ms_rms_envelope_nrmse": 0.09306889067838182, "全帯域_10ms_envelope_correlation": 0.9882832636344245, "全帯域_10ms_modulation_mae_db": 0.5288337122510048, "全帯域_40ms_log_rms_mae_db": 0.42015110170087516, "全帯域_40ms_rms_envelope_nrmse": 0.056867121047823974, "全帯域_40ms_envelope_correlation": 0.9951890260782628, "全帯域_40ms_modulation_mae_db": 0.6602793156652744, "300–1000 Hz_10ms_log_rms_mae_db": 1.6672712341907179, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12608574524253893, "300–1000 Hz_10ms_envelope_correlation": 0.9860902964360979, "300–1000 Hz_10ms_modulation_mae_db": 0.44830846129223456, "300–1000 Hz_40ms_log_rms_mae_db": 0.9687759498513786, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0809216795306812, "300–1000 Hz_40ms_envelope_correlation": 0.9937520708209374, "300–1000 Hz_40ms_modulation_mae_db": 0.6691437148856577, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5114367163151243, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10811694554946795, "1000–3000 Hz_10ms_envelope_correlation": 0.9913439512816717, "1000–3000 Hz_10ms_modulation_mae_db": 0.6122164429832979, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9098426437291567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06297260557482975, "1000–3000 Hz_40ms_envelope_correlation": 0.9973541833304976, "1000–3000 Hz_40ms_modulation_mae_db": 0.8706213941663404, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6619796836074965, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16301597157128486, "3000–6000 Hz_10ms_envelope_correlation": 0.9819038960811435, "3000–6000 Hz_10ms_modulation_mae_db": 1.2158334042934063, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9733040797974138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08586253056697755, "3000–6000 Hz_40ms_envelope_correlation": 0.9947034223045533, "3000–6000 Hz_40ms_modulation_mae_db": 1.1697696010034961, "mr_stft_db": 5.637760895383107, "band_mae_db": 0.15685027146351702, "rms_gain_error_db": -0.023116417006280046, "output_peak": 0.3466452240098816}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8317037147490277, "全帯域_10ms_rms_envelope_nrmse": 0.07720031846456886, "全帯域_10ms_envelope_correlation": 0.9920444708839337, "全帯域_10ms_modulation_mae_db": 0.2375378150788304, "全帯域_40ms_log_rms_mae_db": 0.44583701384378854, "全帯域_40ms_rms_envelope_nrmse": 0.039283959163191856, "全帯域_40ms_envelope_correlation": 0.9979198411587406, "全帯域_40ms_modulation_mae_db": 0.2710893507154338, "300–1000 Hz_10ms_log_rms_mae_db": 1.630256051756251, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1007915035963064, "300–1000 Hz_10ms_envelope_correlation": 0.9911506693444853, "300–1000 Hz_10ms_modulation_mae_db": 0.27145336104878565, "300–1000 Hz_40ms_log_rms_mae_db": 0.9223676744996928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05470518377625749, "300–1000 Hz_40ms_envelope_correlation": 0.9972490596595238, "300–1000 Hz_40ms_modulation_mae_db": 0.4643603058681382, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5132634782420817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1143108448801476, "1000–3000 Hz_10ms_envelope_correlation": 0.9907613215691864, "1000–3000 Hz_10ms_modulation_mae_db": 0.5043558229336266, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9327499407410984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06751747882811118, "1000–3000 Hz_40ms_envelope_correlation": 0.9974895963403734, "1000–3000 Hz_40ms_modulation_mae_db": 1.0774080455266757, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6826571941016284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16912977522911715, "3000–6000 Hz_10ms_envelope_correlation": 0.9807107572034522, "3000–6000 Hz_10ms_modulation_mae_db": 0.6579328021073602, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9974687790126526, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09173568570437046, "3000–6000 Hz_40ms_envelope_correlation": 0.9940523221464413, "3000–6000 Hz_40ms_modulation_mae_db": 0.8668051525543935, "mr_stft_db": 5.637066971134042, "band_mae_db": 0.192028083495942, "rms_gain_error_db": -0.10205600149521546, "output_peak": 0.306428417247416}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8307828987404435, "全帯域_10ms_rms_envelope_nrmse": 0.08583122342902243, "全帯域_10ms_envelope_correlation": 0.9901301351161335, "全帯域_10ms_modulation_mae_db": 0.21355029569421582, "全帯域_40ms_log_rms_mae_db": 0.39273995469571277, "全帯域_40ms_rms_envelope_nrmse": 0.04212716222420461, "全帯域_40ms_envelope_correlation": 0.997403706665333, "全帯域_40ms_modulation_mae_db": 0.2553819547548602, "300–1000 Hz_10ms_log_rms_mae_db": 1.571746938914612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09944389550513928, "300–1000 Hz_10ms_envelope_correlation": 0.9913544411578219, "300–1000 Hz_10ms_modulation_mae_db": 0.5040467027447136, "300–1000 Hz_40ms_log_rms_mae_db": 0.9167088511177436, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05109947418566539, "300–1000 Hz_40ms_envelope_correlation": 0.9975336669904371, "300–1000 Hz_40ms_modulation_mae_db": 0.5461553681057152, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5478141954930487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13554410430405567, "1000–3000 Hz_10ms_envelope_correlation": 0.9857031490688886, "1000–3000 Hz_10ms_modulation_mae_db": 0.5404862988260555, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9599093640084609, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07048929258087522, "1000–3000 Hz_40ms_envelope_correlation": 0.9959083752778172, "1000–3000 Hz_40ms_modulation_mae_db": 0.7293296732455676, "3000–6000 Hz_10ms_log_rms_mae_db": 1.688850852251529, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19473056691688376, "3000–6000 Hz_10ms_envelope_correlation": 0.9744955931145776, "3000–6000 Hz_10ms_modulation_mae_db": 0.8041251070303203, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0015683289897144, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10167978015048272, "3000–6000 Hz_40ms_envelope_correlation": 0.9927325887653217, "3000–6000 Hz_40ms_modulation_mae_db": 1.0747333224338567, "mr_stft_db": 5.634085840486261, "band_mae_db": 0.07712737786316333, "rms_gain_error_db": 0.026605322197599833, "output_peak": 0.29802517234758047}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8006357954436054, "全帯域_10ms_rms_envelope_nrmse": 0.08698124561032808, "全帯域_10ms_envelope_correlation": 0.9898109789780362, "全帯域_10ms_modulation_mae_db": 0.3017918688958749, "全帯域_40ms_log_rms_mae_db": 0.4120251551391292, "全帯域_40ms_rms_envelope_nrmse": 0.04049280202249697, "全帯域_40ms_envelope_correlation": 0.9975664059640893, "全帯域_40ms_modulation_mae_db": 0.3533700942247158, "300–1000 Hz_10ms_log_rms_mae_db": 1.5704458964687906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1155769289509419, "300–1000 Hz_10ms_envelope_correlation": 0.9883041037672212, "300–1000 Hz_10ms_modulation_mae_db": 0.697617266561752, "300–1000 Hz_40ms_log_rms_mae_db": 0.8973851651433827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0566758980691652, "300–1000 Hz_40ms_envelope_correlation": 0.9970800331176922, "300–1000 Hz_40ms_modulation_mae_db": 0.9039969711211074, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5611296079517814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11344363709731753, "1000–3000 Hz_10ms_envelope_correlation": 0.9899948026371883, "1000–3000 Hz_10ms_modulation_mae_db": 0.34191660334602475, "1000–3000 Hz_40ms_log_rms_mae_db": 0.931420180397795, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05843485125137199, "1000–3000 Hz_40ms_envelope_correlation": 0.9972717382014984, "1000–3000 Hz_40ms_modulation_mae_db": 0.5941446057767213, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6853884194251743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17758017413063906, "3000–6000 Hz_10ms_envelope_correlation": 0.9804717323290513, "3000–6000 Hz_10ms_modulation_mae_db": 1.2042510954354544, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9879905567425924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1053413517587977, "3000–6000 Hz_40ms_envelope_correlation": 0.9935606847478184, "3000–6000 Hz_40ms_modulation_mae_db": 1.3900954250984203, "mr_stft_db": 5.651142086702909, "band_mae_db": 0.1792575075884724, "rms_gain_error_db": 0.005246152052818836, "output_peak": 0.2978825895521492}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8372530876956261, "全帯域_10ms_rms_envelope_nrmse": 0.07657456705735485, "全帯域_10ms_envelope_correlation": 0.9920975815864017, "全帯域_10ms_modulation_mae_db": 0.08845844462296726, "全帯域_40ms_log_rms_mae_db": 0.3948433909797796, "全帯域_40ms_rms_envelope_nrmse": 0.04148867148173964, "全帯域_40ms_envelope_correlation": 0.997526642790097, "全帯域_40ms_modulation_mae_db": 0.3350396242554241, "300–1000 Hz_10ms_log_rms_mae_db": 1.6719613646528189, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10266218636113753, "300–1000 Hz_10ms_envelope_correlation": 0.9909819424154637, "300–1000 Hz_10ms_modulation_mae_db": 0.29280985833838846, "300–1000 Hz_40ms_log_rms_mae_db": 0.9561015922056602, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06275716306865453, "300–1000 Hz_40ms_envelope_correlation": 0.9965789351071032, "300–1000 Hz_40ms_modulation_mae_db": 0.6808477273260936, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4978892567758815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11259230802321218, "1000–3000 Hz_10ms_envelope_correlation": 0.990164666952091, "1000–3000 Hz_10ms_modulation_mae_db": 0.4414957352060171, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8959714827238596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058688842582375236, "1000–3000 Hz_40ms_envelope_correlation": 0.9972691969380987, "1000–3000 Hz_40ms_modulation_mae_db": 0.7151862385024271, "3000–6000 Hz_10ms_log_rms_mae_db": 1.649930880367207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17353756196563488, "3000–6000 Hz_10ms_envelope_correlation": 0.9795430481526415, "3000–6000 Hz_10ms_modulation_mae_db": 0.809874364096959, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9961131792301751, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09707120046953119, "3000–6000 Hz_40ms_envelope_correlation": 0.9932535939240867, "3000–6000 Hz_40ms_modulation_mae_db": 0.7661989001122662, "mr_stft_db": 5.655908661378134, "band_mae_db": 0.09893234416644009, "rms_gain_error_db": -0.0765542593188284, "output_peak": 0.29210728004155645}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7094911967163195, "spectral_convergence": 0.39196293680132727, "spectral_convergence_status": "ok", "amplitude_floor": 6.641091871383268e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.567491544047614, "spectral_convergence": 0.38089608777753386, "spectral_convergence_status": "ok", "amplitude_floor": 5.299135910844449e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7062598811830565, "spectral_convergence": 0.37652522628401064, "spectral_convergence_status": "ok", "amplitude_floor": 3.9942669914893874e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.486389753340235e-05, "output_power": 7.664220161271605e-05, "delta_db": 0.10195538302238635, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003919851371758182, "output_power": 0.0003915835888711562, "delta_db": -0.0044511788794326145, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038865006012116416, "output_power": 0.0003847118938368052, "delta_db": -0.04423125966232845, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001490171424082218, "output_power": 0.000145362923460564, "delta_db": -0.10782582312868907, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.76347124356667e-05, "output_power": 9.777119249400118e-05, "delta_db": 0.006066607443006226, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002530434948901305, "output_power": 0.00024733670736455687, "delta_db": -0.09906602166776911, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.060599378967268636, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03793518354445511, "pitch": null, "output_rms_dbfs": -28.717919056954177, "output_peak": 0.29611675471725346, "reference_rms_dbfs": -28.67998387340972, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.661080873982331}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.947854524209456e-05, "log_rms_mae_db": 0.834947694005919, "rms_envelope_nrmse": 0.07660192471019905, "envelope_correlation": 0.9920761960609149, "rms_gain_error_db": -0.03716782618315412, "reference_mean_power": 0.0013552494150092414, "output_mean_power": 0.001343700398219395}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.097356675930481e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024105831474109917, "output_power": 0.00023733601045972198, "reference_active": true, "delta_db": -0.06758478303767712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7795131277493674e-05, "output_power": 2.4558925526979205e-05, "reference_active": true, "delta_db": -0.5375936742227534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0822198872325408e-05, "output_power": 9.58513242286149e-06, "reference_active": true, "delta_db": -0.5271739318303688}], "active_band_count": 3, "active_band_mae_db": 0.37745079636359974}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.267346213866395e-05, "log_rms_mae_db": 0.42584687199800586, "rms_envelope_nrmse": 0.040852111210426, "envelope_correlation": 0.9975394590573506, "rms_gain_error_db": -0.035527271103271946, "reference_mean_power": 0.0013563897957452433, "output_mean_power": 0.00134533916925622}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.582417446963071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021668841148527672, "output_power": 0.00021302825875954838, "reference_active": true, "delta_db": -0.07398468268015668}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1881573874210085e-05, "output_power": 1.0441420131558702e-05, "reference_active": true, "delta_db": -0.5611440177229713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.720949484930854e-07, "output_power": 5.4229002969602e-07, "reference_active": true, "delta_db": -0.9319901146592289}], "active_band_count": 3, "active_band_mae_db": 0.522372938354119}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.299288573172057e-05, "log_rms_mae_db": 1.6277469773346098, "rms_envelope_nrmse": 0.09969113314035342, "envelope_correlation": 0.9912847029713261, "rms_gain_error_db": -0.06685578487795284, "reference_mean_power": 0.0005726237314993749, "output_mean_power": 0.0005638761998561798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0939920051603367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018116410089920085, "output_power": 0.00017161268104318224, "reference_active": true, "delta_db": -0.2352276678557204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.90482495868649e-05, "output_power": 2.5198906388822323e-05, "reference_active": true, "delta_db": -0.6173827426527349}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.620984223411989e-06, "output_power": 5.943325261010946e-06, "reference_active": true, "delta_db": 0.24217132243096587}], "active_band_count": 3, "active_band_mae_db": 0.3649272443131404}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.559252308295269e-05, "log_rms_mae_db": 0.9320214105411326, "rms_envelope_nrmse": 0.05426910074443945, "envelope_correlation": 0.997235153740811, "rms_gain_error_db": -0.06446252270921453, "reference_mean_power": 0.0005714831913152958, "output_mean_power": 0.000563063284056172}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7768006225472703e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016765591938032763, "output_power": 0.00015897396514433432, "reference_active": true, "delta_db": -0.2309288496356341}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3133776830395815e-05, "output_power": 1.1633347447259532e-05, "reference_active": true, "delta_db": -0.52684933328016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.763861134744573e-07, "output_power": 3.8357520275790646e-07, "reference_active": true, "delta_db": -0.9410857014837131}], "active_band_count": 3, "active_band_mae_db": 0.5662879614665024}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.686331455816738e-05, "log_rms_mae_db": 1.4965948571116476, "rms_envelope_nrmse": 0.11235539407231904, "envelope_correlation": 0.9902665415790632, "rms_gain_error_db": -0.12419982656959343, "reference_mean_power": 0.00020459070686192204, "output_mean_power": 0.00019882267847606486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3670238086467597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.952541720107036e-05, "output_power": 6.361568409437352e-05, "reference_active": true, "delta_db": -0.38579401866809354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.449912227458803e-06, "output_power": 6.587573362373502e-06, "reference_active": true, "delta_db": -0.5342569129738213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5578562674598188e-06, "output_power": 1.1145234648109623e-06, "reference_active": true, "delta_db": -1.4543816945401165}], "active_band_count": 3, "active_band_mae_db": 0.7914775420606771}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4935905625855925e-05, "log_rms_mae_db": 0.8537325757672063, "rms_envelope_nrmse": 0.06136188017925685, "envelope_correlation": 0.9970850442187477, "rms_gain_error_db": -0.1235776076542465, "reference_mean_power": 0.00020499017149025147, "output_mean_power": 0.00019923942421969256}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2857171527439997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.293887842389662e-05, "output_power": 5.754924824030153e-05, "reference_active": true, "delta_db": -0.388793446216713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.713520804403163e-06, "output_power": 3.253241506127838e-06, "reference_active": true, "delta_db": -0.5746955676107299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3334126499972412e-07, "output_power": 9.363931511718645e-08, "reference_active": true, "delta_db": -1.5350634234818243}], "active_band_count": 3, "active_band_mae_db": 0.8328508124364223}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.417692344469317e-05, "log_rms_mae_db": 1.6289900080114286, "rms_envelope_nrmse": 0.17355549622712738, "envelope_correlation": 0.9803512649946299, "rms_gain_error_db": 0.2058456321402989, "reference_mean_power": 0.00010545678906217645, "output_mean_power": 0.00011057554970434008}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.569551408613973e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4047256300828896e-05, "output_power": 3.617247884457367e-05, "reference_active": true, "delta_db": 0.2629615129209548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.550080432854394e-06, "output_power": 6.430412051659802e-06, "reference_active": true, "delta_db": -0.697127755105368}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.516411722896473e-06, "output_power": 2.014019088497362e-06, "reference_active": true, "delta_db": -2.420361368220755}], "active_band_count": 3, "active_band_mae_db": 1.126816878749026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.848412221971719e-05, "log_rms_mae_db": 0.9840774753496352, "rms_envelope_nrmse": 0.09760723968759252, "envelope_correlation": 0.9937602040506281, "rms_gain_error_db": 0.20574379590013225, "reference_mean_power": 0.0001059674543662489, "output_mean_power": 0.0001111083967867511}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.064372154195953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3923869911506206e-05, "output_power": 3.536814956439717e-05, "reference_active": true, "delta_db": 0.1810694929765572}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.204493449095079e-06, "output_power": 3.3950532654975807e-06, "reference_active": true, "delta_db": -0.9286708777333691}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4577951814618017e-07, "output_power": 1.2670705819819766e-07, "reference_active": true, "delta_db": -2.877448805818153}], "active_band_count": 3, "active_band_mae_db": 1.329063058842693}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.864466837770488, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8d047d563cb9f2b6e0a1fb1da47e7f33f36847d6e8d8bd8771009bbf458fe570", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9892651156450494, "standard_deviation": 0.001957054925698843, "minimum": 0.9857031490688886, "maximum": 0.9913439512816717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5389642898074494, "standard_deviation": 0.04219673811411718, "minimum": 1.4965948571116476, "maximum": 1.6183631761426405, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.526202122515667, "standard_deviation": 0.1452278785541016, "minimum": 0.34191660334602475, "maximum": 0.7914775420606771, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11841765150322199, "standard_deviation": 0.009548713414242288, "minimum": 0.10811694554946795, "maximum": 0.13554410430405567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996787852322607, "standard_deviation": 0.000840343151961955, "minimum": 0.9951139655687223, "maximum": 0.9974895963403734, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9247954543067325, "standard_deviation": 0.0426992024030377, "minimum": 0.8537325757672063, "maximum": 0.9973438994053816, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7989570296648831, "standard_deviation": 0.1465330352688321, "minimum": 0.5941446057767213, "maximum": 1.0774080455266757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06571022859061554, "standard_deviation": 0.00655334215521313, "minimum": 0.05843485125137199, "maximum": 0.07737562333926222, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9796122522604241, "standard_deviation": 0.0022145547773557176, "minimum": 0.9744955931145776, "maximum": 0.9819038960811435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6729649011729455, "standard_deviation": 0.024474957779041993, "minimum": 1.6289900080114286, "maximum": 1.7053665275651377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8814191188364382, "standard_deviation": 0.2627572801602417, "minimum": 0.5645004975829951, "maximum": 1.2158334042934063, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1756427805193831, "standard_deviation": 0.009207494946967562, "minimum": 0.16301597157128486, "maximum": 0.19473056691688376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935709044488557, "standard_deviation": 0.0009251469457134736, "minimum": 0.9919417923459685, "maximum": 0.9947034223045533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9936317293970423, "standard_deviation": 0.01799268479873448, "minimum": 0.9733040797974138, "maximum": 1.0309096987902109, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0175697761139446, "standard_deviation": 0.25926721688245286, "minimum": 0.7426074390080516, "maximum": 1.3900954250984203, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09786692694353993, "standard_deviation": 0.010108145295923913, "minimum": 0.08586253056697755, "maximum": 0.11650186927163825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9901178927800072, "standard_deviation": 0.0021909542184673178, "minimum": 0.9860902964360979, "maximum": 0.9929132452902665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6197912767263531, "standard_deviation": 0.03807379542277296, "minimum": 1.5704458964687906, "maximum": 1.6719613646528189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4499191454369341, "standard_deviation": 0.20750054271374904, "minimum": 0.22190310337285915, "maximum": 0.7982871658235989, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10590920291038712, "standard_deviation": 0.011424415927415322, "minimum": 0.09017140044979532, "maximum": 0.12608574524253893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996704211807881, "standard_deviation": 0.001261070998158888, "minimum": 0.9937520708209374, "maximum": 0.9976838451295883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9308950695521914, "standard_deviation": 0.025537941787917174, "minimum": 0.8973851651433827, "maximum": 0.9687759498513786, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6607416118832322, "standard_deviation": 0.2033757650777836, "minimum": 0.44443212212202693, "maximum": 1.0107087241706165, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.058806791287953096, "standard_deviation": 0.00995301816965296, "minimum": 0.049599043208738765, "maximum": 0.0809216795306812, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12802829841913668, "standard_deviation": 0.047120287512801624, "minimum": 0.060599378967268636, "maximum": 0.192028083495942, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.648805884858891, "standard_deviation": 0.011180359726710979, "minimum": 5.634085840486261, "maximum": 5.662501096628664, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3067733286464066, "standard_deviation": 0.020425217141568384, "minimum": 0.2879929076899897, "maximum": 0.3466452240098816, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.037887974855216645, "standard_deviation": 0.043043956758098564, "minimum": -0.10205600149521546, "maximum": 0.026605322197599833, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.990720037375989, "standard_deviation": 0.0014344346796779455, "minimum": 0.9882832636344245, "maximum": 0.9920975815864017, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8381519135679452, "standard_deviation": 0.0331814258239134, "minimum": 0.8006357954436054, "maximum": 0.9116187325379492, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.307002620542733, "standard_deviation": 0.15030558502601798, "minimum": 0.08845844462296726, "maximum": 0.5288337122510048, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08281054060522047, "standard_deviation": 0.006292953280164695, "minimum": 0.07657456705735485, "maximum": 0.09306889067838182, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972236803771513, "standard_deviation": 0.0008573264532599003, "minimum": 0.9951890260782628, "maximum": 0.9979198411587406, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4115132635626206, "standard_deviation": 0.01880157485372914, "minimum": 0.3920955912643989, "maximum": 0.44583701384378854, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4093011486704323, "standard_deviation": 0.17579846207687685, "minimum": 0.22936160398171337, "maximum": 0.6602793156652744, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04332991795192853, "standard_deviation": 0.005709574253925142, "minimum": 0.039283959163191856, "maximum": 0.056867121047823974, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0006", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.3042941093444824, "predicted_mos": 2.3042941093444824, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b9a4f7a84ac765b750f686abf412e9918352d6bca78398c3604ed204ad6a819d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.943236299790442, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8719788762472717, "全帯域_10ms_rms_envelope_nrmse": 0.08016391911349957, "全帯域_10ms_envelope_correlation": 0.991319957242797, "全帯域_10ms_modulation_mae_db": 0.4267354482068484, "全帯域_40ms_log_rms_mae_db": 0.44765823935986626, "全帯域_40ms_rms_envelope_nrmse": 0.03770058734434898, "全帯域_40ms_envelope_correlation": 0.997893814064011, "全帯域_40ms_modulation_mae_db": 0.4010170877799141, "300–1000 Hz_10ms_log_rms_mae_db": 1.331493224509837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09321125084301331, "300–1000 Hz_10ms_envelope_correlation": 0.9924114554104133, "300–1000 Hz_10ms_modulation_mae_db": 0.9007937578150771, "300–1000 Hz_40ms_log_rms_mae_db": 0.6933980416594806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05093217246267216, "300–1000 Hz_40ms_envelope_correlation": 0.9975323938439642, "300–1000 Hz_40ms_modulation_mae_db": 1.1680630857346344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.135747782426169, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.106975029285937, "1000–3000 Hz_10ms_envelope_correlation": 0.9915269382640379, "1000–3000 Hz_10ms_modulation_mae_db": 0.5332519041591156, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6639574601611461, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06026083612940739, "1000–3000 Hz_40ms_envelope_correlation": 0.997644226513871, "1000–3000 Hz_40ms_modulation_mae_db": 0.662057840831782, "3000–6000 Hz_10ms_log_rms_mae_db": 1.147301688915032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14436012286488772, "3000–6000 Hz_10ms_envelope_correlation": 0.9858020917139919, "3000–6000 Hz_10ms_modulation_mae_db": 0.4693854315295196, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6873618958695619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08278892386358781, "3000–6000 Hz_40ms_envelope_correlation": 0.9950897838639877, "3000–6000 Hz_40ms_modulation_mae_db": 0.5726355777183594, "mr_stft_db": 5.360642725903887, "band_mae_db": 0.10545899943516336, "rms_gain_error_db": -0.021330012669965925, "output_peak": 0.3039534811006465}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9285406367102959, "全帯域_10ms_rms_envelope_nrmse": 0.09318851865698606, "全帯域_10ms_envelope_correlation": 0.9882798490741499, "全帯域_10ms_modulation_mae_db": 0.23021057643554366, "全帯域_40ms_log_rms_mae_db": 0.4617284767235766, "全帯域_40ms_rms_envelope_nrmse": 0.04711390747618348, "全帯域_40ms_envelope_correlation": 0.9966975551790895, "全帯域_40ms_modulation_mae_db": 0.5016718876027458, "300–1000 Hz_10ms_log_rms_mae_db": 1.2142703923090394, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11525794998883437, "300–1000 Hz_10ms_envelope_correlation": 0.9882860600943499, "300–1000 Hz_10ms_modulation_mae_db": 0.7058389869027284, "300–1000 Hz_40ms_log_rms_mae_db": 0.6561604394626199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06134099279623376, "300–1000 Hz_40ms_envelope_correlation": 0.9964156761461864, "300–1000 Hz_40ms_modulation_mae_db": 1.1163512602225805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1508142588607608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10763130969916332, "1000–3000 Hz_10ms_envelope_correlation": 0.9910881921782523, "1000–3000 Hz_10ms_modulation_mae_db": 0.33382827057977993, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6544109141042934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059593297641787765, "1000–3000 Hz_40ms_envelope_correlation": 0.9973012139423917, "1000–3000 Hz_40ms_modulation_mae_db": 0.5030478577231332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1564487766222287, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1487266796389024, "3000–6000 Hz_10ms_envelope_correlation": 0.9857244566499619, "3000–6000 Hz_10ms_modulation_mae_db": 0.2343917824775462, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7128162161329775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08301331097868414, "3000–6000 Hz_40ms_envelope_correlation": 0.9956030417760641, "3000–6000 Hz_40ms_modulation_mae_db": 0.5296849880147211, "mr_stft_db": 5.3533613340302395, "band_mae_db": 0.07832276479171911, "rms_gain_error_db": -0.013381191483432982, "output_peak": 0.3172876832693471}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8879092648279181, "全帯域_10ms_rms_envelope_nrmse": 0.08622465465369579, "全帯域_10ms_envelope_correlation": 0.9901443702383328, "全帯域_10ms_modulation_mae_db": 0.5335125892968589, "全帯域_40ms_log_rms_mae_db": 0.4354071360117261, "全帯域_40ms_rms_envelope_nrmse": 0.040920187747447344, "全帯域_40ms_envelope_correlation": 0.9978324762387679, "全帯域_40ms_modulation_mae_db": 0.7003595004440509, "300–1000 Hz_10ms_log_rms_mae_db": 1.2367262626706943, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09157258328309213, "300–1000 Hz_10ms_envelope_correlation": 0.992741904574443, "300–1000 Hz_10ms_modulation_mae_db": 0.5708477312076494, "300–1000 Hz_40ms_log_rms_mae_db": 0.6915006092390503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0447390801324795, "300–1000 Hz_40ms_envelope_correlation": 0.9982803011916124, "300–1000 Hz_40ms_modulation_mae_db": 0.7455473016060955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.15270123579745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12690967692786448, "1000–3000 Hz_10ms_envelope_correlation": 0.9875265346392879, "1000–3000 Hz_10ms_modulation_mae_db": 0.2845724750714346, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6795724926915888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08060554394071176, "1000–3000 Hz_40ms_envelope_correlation": 0.9947653659928972, "1000–3000 Hz_40ms_modulation_mae_db": 0.44590209386520474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1659768823067838, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15534124196285548, "3000–6000 Hz_10ms_envelope_correlation": 0.9835795094305279, "3000–6000 Hz_10ms_modulation_mae_db": 1.0705658660033286, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7374137358201924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09119640250684367, "3000–6000 Hz_40ms_envelope_correlation": 0.9943242561798196, "3000–6000 Hz_40ms_modulation_mae_db": 1.4173422883128115, "mr_stft_db": 5.342609226606029, "band_mae_db": 0.16042310929746514, "rms_gain_error_db": -0.13845796445082795, "output_peak": 0.32694970750251534}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.89858031344649, "全帯域_10ms_rms_envelope_nrmse": 0.07956267995531449, "全帯域_10ms_envelope_correlation": 0.9914477075192404, "全帯域_10ms_modulation_mae_db": 0.2861816697620569, "全帯域_40ms_log_rms_mae_db": 0.42375883093572414, "全帯域_40ms_rms_envelope_nrmse": 0.03570945734930326, "全帯域_40ms_envelope_correlation": 0.9981222411457692, "全帯域_40ms_modulation_mae_db": 0.48269367331848695, "300–1000 Hz_10ms_log_rms_mae_db": 1.2225700332669227, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09736196155937665, "300–1000 Hz_10ms_envelope_correlation": 0.9917194106434976, "300–1000 Hz_10ms_modulation_mae_db": 0.5775299541297046, "300–1000 Hz_40ms_log_rms_mae_db": 0.6801025292820887, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056483070945525814, "300–1000 Hz_40ms_envelope_correlation": 0.9970475769791108, "300–1000 Hz_40ms_modulation_mae_db": 0.57741908573162, "1000–3000 Hz_10ms_log_rms_mae_db": 1.118499098948014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10618561215705977, "1000–3000 Hz_10ms_envelope_correlation": 0.9912612293242326, "1000–3000 Hz_10ms_modulation_mae_db": 0.20083923696597972, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6539450661061909, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05917863846150452, "1000–3000 Hz_40ms_envelope_correlation": 0.9971523615881341, "1000–3000 Hz_40ms_modulation_mae_db": 0.2222460056405909, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1401265374670133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14594098703508357, "3000–6000 Hz_10ms_envelope_correlation": 0.98571878697272, "3000–6000 Hz_10ms_modulation_mae_db": 0.7423786389879448, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7189243833202162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08494260467165117, "3000–6000 Hz_40ms_envelope_correlation": 0.9949895331634379, "3000–6000 Hz_40ms_modulation_mae_db": 0.9060576584071941, "mr_stft_db": 5.356640434376498, "band_mae_db": 0.0630011698774617, "rms_gain_error_db": -0.037652319872026045, "output_peak": 0.3194143112164707}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9326752301866428, "全帯域_10ms_rms_envelope_nrmse": 0.09311166687060247, "全帯域_10ms_envelope_correlation": 0.9884769488800765, "全帯域_10ms_modulation_mae_db": 0.5330718404092014, "全帯域_40ms_log_rms_mae_db": 0.4997615464121176, "全帯域_40ms_rms_envelope_nrmse": 0.05406752258552925, "全帯域_40ms_envelope_correlation": 0.9960707000288005, "全帯域_40ms_modulation_mae_db": 0.8062791359774977, "300–1000 Hz_10ms_log_rms_mae_db": 1.2297312749295597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10560180045979387, "300–1000 Hz_10ms_envelope_correlation": 0.9902664680895562, "300–1000 Hz_10ms_modulation_mae_db": 0.6954465481867098, "300–1000 Hz_40ms_log_rms_mae_db": 0.6757421316331843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05719289054396873, "300–1000 Hz_40ms_envelope_correlation": 0.9970519698270365, "300–1000 Hz_40ms_modulation_mae_db": 0.8054480354545607, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1692746627614024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10927793958183278, "1000–3000 Hz_10ms_envelope_correlation": 0.9910648528046192, "1000–3000 Hz_10ms_modulation_mae_db": 0.2023287276049047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.655197446899679, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06687200387305198, "1000–3000 Hz_40ms_envelope_correlation": 0.9968643159049141, "1000–3000 Hz_40ms_modulation_mae_db": 0.5973743118456009, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1963267256207222, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16096437886111814, "3000–6000 Hz_10ms_envelope_correlation": 0.9824788796881299, "3000–6000 Hz_10ms_modulation_mae_db": 1.0956447745783053, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7412644738719603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08709842399836817, "3000–6000 Hz_40ms_envelope_correlation": 0.9951179544956392, "3000–6000 Hz_40ms_modulation_mae_db": 1.3957119308269075, "mr_stft_db": 5.368507261525893, "band_mae_db": 0.16238073299390368, "rms_gain_error_db": -0.15578894213953473, "output_peak": 0.3147079463842533}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8532391634766715, "全帯域_10ms_rms_envelope_nrmse": 0.08416913947885835, "全帯域_10ms_envelope_correlation": 0.9905436163520472, "全帯域_10ms_modulation_mae_db": 0.336603418010373, "全帯域_40ms_log_rms_mae_db": 0.4557596288607981, "全帯域_40ms_rms_envelope_nrmse": 0.04764314561280652, "全帯域_40ms_envelope_correlation": 0.9968879044957942, "全帯域_40ms_modulation_mae_db": 0.46543668048221304, "300–1000 Hz_10ms_log_rms_mae_db": 1.2834695589264347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09793327767470181, "300–1000 Hz_10ms_envelope_correlation": 0.9915589877184192, "300–1000 Hz_10ms_modulation_mae_db": 0.5921092444186102, "300–1000 Hz_40ms_log_rms_mae_db": 0.6895158316751525, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04680221778800438, "300–1000 Hz_40ms_envelope_correlation": 0.9979208389265168, "300–1000 Hz_40ms_modulation_mae_db": 0.6936661460090582, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1536252281526536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09491192155904044, "1000–3000 Hz_10ms_envelope_correlation": 0.9932273600888113, "1000–3000 Hz_10ms_modulation_mae_db": 0.3443446065973608, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7114160669676167, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054745152330653246, "1000–3000 Hz_40ms_envelope_correlation": 0.9978791773678717, "1000–3000 Hz_40ms_modulation_mae_db": 0.4323675159972174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1769318146208498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14787985357747338, "3000–6000 Hz_10ms_envelope_correlation": 0.9850880768123911, "3000–6000 Hz_10ms_modulation_mae_db": 0.715002495109036, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7375541194403381, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08401394288891824, "3000–6000 Hz_40ms_envelope_correlation": 0.9951130216055364, "3000–6000 Hz_40ms_modulation_mae_db": 0.81345284275955, "mr_stft_db": 5.343648685231588, "band_mae_db": 0.15567690642509224, "rms_gain_error_db": -0.12311253968386349, "output_peak": 0.307438040779384}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8842933310955274, "全帯域_10ms_rms_envelope_nrmse": 0.08359524162394308, "全帯域_10ms_envelope_correlation": 0.9905814684872626, "全帯域_10ms_modulation_mae_db": 0.22853556178928047, "全帯域_40ms_log_rms_mae_db": 0.4194894995772396, "全帯域_40ms_rms_envelope_nrmse": 0.03872048564235015, "全帯域_40ms_envelope_correlation": 0.9978101554500652, "全帯域_40ms_modulation_mae_db": 0.32155184331030884, "300–1000 Hz_10ms_log_rms_mae_db": 1.1829971723861386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09257798977968104, "300–1000 Hz_10ms_envelope_correlation": 0.9924650893626549, "300–1000 Hz_10ms_modulation_mae_db": 0.32988371055641447, "300–1000 Hz_40ms_log_rms_mae_db": 0.6090629801387236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042109038539923, "300–1000 Hz_40ms_envelope_correlation": 0.9983253903961847, "300–1000 Hz_40ms_modulation_mae_db": 0.44577778217307057, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0994396732733087, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13097463136448076, "1000–3000 Hz_10ms_envelope_correlation": 0.9866751976898553, "1000–3000 Hz_10ms_modulation_mae_db": 0.26669156452109344, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6744814833271687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06264216474524291, "1000–3000 Hz_40ms_envelope_correlation": 0.9968979826945064, "1000–3000 Hz_40ms_modulation_mae_db": 0.6277241769493193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1561202008459952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14516619672919978, "3000–6000 Hz_10ms_envelope_correlation": 0.9867588765317102, "3000–6000 Hz_10ms_modulation_mae_db": 0.4405267130227261, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7372656709424102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08528964288883355, "3000–6000 Hz_40ms_envelope_correlation": 0.995665119755829, "3000–6000 Hz_40ms_modulation_mae_db": 0.5618075764958841, "mr_stft_db": 5.348931040613421, "band_mae_db": 0.05867200235518951, "rms_gain_error_db": -0.04173289170739715, "output_peak": 0.33035345302073676}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9012126400612348, "全帯域_10ms_rms_envelope_nrmse": 0.09368497038748061, "全帯域_10ms_envelope_correlation": 0.9882175472644035, "全帯域_10ms_modulation_mae_db": 0.4653129881032177, "全帯域_40ms_log_rms_mae_db": 0.4472805498574794, "全帯域_40ms_rms_envelope_nrmse": 0.05050226790703411, "全帯域_40ms_envelope_correlation": 0.9964506388446239, "全帯域_40ms_modulation_mae_db": 0.5965942625435009, "300–1000 Hz_10ms_log_rms_mae_db": 1.275137103809349, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10066813340753626, "300–1000 Hz_10ms_envelope_correlation": 0.991144034236869, "300–1000 Hz_10ms_modulation_mae_db": 0.667613305965619, "300–1000 Hz_40ms_log_rms_mae_db": 0.6919036062648305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05512360567197969, "300–1000 Hz_40ms_envelope_correlation": 0.9972160298909276, "300–1000 Hz_40ms_modulation_mae_db": 0.7896109035307987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1461853264624458, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11535602935045057, "1000–3000 Hz_10ms_envelope_correlation": 0.990846236440862, "1000–3000 Hz_10ms_modulation_mae_db": 0.42839592376047336, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6766363013116665, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07230587706201233, "1000–3000 Hz_40ms_envelope_correlation": 0.997306746101805, "1000–3000 Hz_40ms_modulation_mae_db": 0.5755695200522658, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1844279342138218, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.155236238458099, "3000–6000 Hz_10ms_envelope_correlation": 0.9835655369538687, "3000–6000 Hz_10ms_modulation_mae_db": 0.44156975704265733, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7246907100030618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09266091953480575, "3000–6000 Hz_40ms_envelope_correlation": 0.9938638293477077, "3000–6000 Hz_40ms_modulation_mae_db": 0.4520697056593754, "mr_stft_db": 5.373561810741897, "band_mae_db": 0.1755633060977362, "rms_gain_error_db": -0.12487677394663069, "output_peak": 0.30286765050675885}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.327499608230112, "spectral_convergence": 0.36818690218812344, "spectral_convergence_status": "ok", "amplitude_floor": 6.641091871383268e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.170931494245931, "spectral_convergence": 0.34753906819865993, "spectral_convergence_status": "ok", "amplitude_floor": 5.299135910844449e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5834970752356154, "spectral_convergence": 0.3735138363360647, "spectral_convergence_status": "ok", "amplitude_floor": 3.9942669914893874e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.486389753340235e-05, "output_power": 7.528705745453725e-05, "delta_db": 0.024478899287813518, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003919851371758182, "output_power": 0.0003898389732515042, "delta_db": -0.02384345732863332, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038865006012116416, "output_power": 0.00039484123164347377, "delta_db": 0.06863758726079258, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001490171424082218, "output_power": 0.00014087101857340127, "delta_db": -0.24414576200026408, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.76347124356667e-05, "output_power": 9.250177560196384e-05, "delta_db": -0.23454182224023257, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002530434948901305, "output_power": 0.0002552147813985401, "delta_db": 0.03710646849324422, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10545899943516336, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.021330012669965925, "pitch": null, "output_rms_dbfs": -28.701313886079692, "output_peak": 0.3039534811006465, "reference_rms_dbfs": -28.67998387340972, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.360642725903887}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.947854524209456e-05, "log_rms_mae_db": 0.8719788762472717, "rms_envelope_nrmse": 0.08016391911349957, "envelope_correlation": 0.991319957242797, "rms_gain_error_db": -0.020559653708011993, "reference_mean_power": 0.0013552494150092414, "output_mean_power": 0.0013488487789262023}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.097356675930481e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024105831474109917, "output_power": 0.00023723874013319136, "reference_active": true, "delta_db": -0.06936507019387095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7795131277493674e-05, "output_power": 2.7959718818357147e-05, "reference_active": true, "delta_db": 0.025640699703980425}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0822198872325408e-05, "output_power": 8.237500480549984e-06, "reference_active": true, "delta_db": -1.1852005747226937}], "active_band_count": 3, "active_band_mae_db": 0.4267354482068484}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.267346213866395e-05, "log_rms_mae_db": 0.44765823935986626, "rms_envelope_nrmse": 0.03770058734434898, "envelope_correlation": 0.997893814064011, "rms_gain_error_db": -0.020238940235179144, "reference_mean_power": 0.0013563897957452433, "output_mean_power": 0.0013500834697592367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.582417446963071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021668841148527672, "output_power": 0.00021379976058400938, "reference_active": true, "delta_db": -0.058284712909996456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1881573874210085e-05, "output_power": 1.2086579070830928e-05, "reference_active": true, "delta_db": 0.07429424962107183}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.720949484930854e-07, "output_power": 5.2527097206866e-07, "reference_active": true, "delta_db": -1.070472300808674}], "active_band_count": 3, "active_band_mae_db": 0.4010170877799141}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.299288573172057e-05, "log_rms_mae_db": 1.331493224509837, "rms_envelope_nrmse": 0.09321125084301331, "envelope_correlation": 0.9924114554104133, "rms_gain_error_db": 0.03246665371074281, "reference_mean_power": 0.0005726237314993749, "output_mean_power": 0.0005769205489607445}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0939920051603367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018116410089920085, "output_power": 0.0001765921219657397, "reference_active": true, "delta_db": -0.11100817818381845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.90482495868649e-05, "output_power": 2.5500786335576188e-05, "reference_active": true, "delta_db": -0.5656639501194975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.620984223411989e-06, "output_power": 3.5256683081504443e-06, "reference_active": true, "delta_db": -2.0257091451419154}], "active_band_count": 3, "active_band_mae_db": 0.9007937578150771}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.559252308295269e-05, "log_rms_mae_db": 0.6933980416594806, "rms_envelope_nrmse": 0.05093217246267216, "envelope_correlation": 0.9975323938439642, "rms_gain_error_db": 0.03268193135511888, "reference_mean_power": 0.0005714831913152958, "output_mean_power": 0.0005757999918796801}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7768006225472703e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016765591938032763, "output_power": 0.0001625567769881025, "reference_active": true, "delta_db": -0.13413811612472623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3133776830395815e-05, "output_power": 1.1642595203694948e-05, "reference_active": true, "delta_db": -0.5233983453984294}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.763861134744573e-07, "output_power": 2.4733969098295455e-07, "reference_active": true, "delta_db": -2.8466527956807477}], "active_band_count": 3, "active_band_mae_db": 1.1680630857346344}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.686331455816738e-05, "log_rms_mae_db": 1.135747782426169, "rms_envelope_nrmse": 0.106975029285937, "envelope_correlation": 0.9915269382640379, "rms_gain_error_db": -0.25441310740767675, "reference_mean_power": 0.00020459070686192204, "output_mean_power": 0.00019294991522496416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3670238086467597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.952541720107036e-05, "output_power": 6.520190903782056e-05, "reference_active": true, "delta_db": -0.2788329204718445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.449912227458803e-06, "output_power": 5.708785370053431e-06, "reference_active": true, "delta_db": -1.1560744066461617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5578562674598188e-06, "output_power": 1.4998317650740808e-06, "reference_active": true, "delta_db": -0.1648483853593406}], "active_band_count": 3, "active_band_mae_db": 0.5332519041591156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4935905625855925e-05, "log_rms_mae_db": 0.6639574601611461, "rms_envelope_nrmse": 0.06026083612940739, "envelope_correlation": 0.997644226513871, "rms_gain_error_db": -0.2532362305679673, "reference_mean_power": 0.00020499017149025147, "output_mean_power": 0.00019337904703199015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2857171527439997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.293887842389662e-05, "output_power": 5.910003327077776e-05, "reference_active": true, "delta_db": -0.2733127411154273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.713520804403163e-06, "output_power": 2.8643203410937685e-06, "reference_active": true, "delta_db": -1.1276427415777828}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3334126499972412e-07, "output_power": 1.1653134710896158e-07, "reference_active": true, "delta_db": -0.5852180398021359}], "active_band_count": 3, "active_band_mae_db": 0.662057840831782}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.417692344469317e-05, "log_rms_mae_db": 1.147301688915032, "rms_envelope_nrmse": 0.14436012286488772, "envelope_correlation": 0.9858020917139919, "rms_gain_error_db": -0.03277841963718791, "reference_mean_power": 0.00010545678906217645, "output_mean_power": 0.00010466384901327709}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.569551408613973e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4047256300828896e-05, "output_power": 3.3642074435845916e-05, "reference_active": true, "delta_db": -0.051993526296859285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.550080432854394e-06, "output_power": 8.395286293353089e-06, "reference_active": true, "delta_db": 0.46083932598120964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.516411722896473e-06, "output_power": 2.8613262227562353e-06, "reference_active": true, "delta_db": -0.89532344231049}], "active_band_count": 3, "active_band_mae_db": 0.4693854315295196}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.848412221971719e-05, "log_rms_mae_db": 0.6873618958695619, "rms_envelope_nrmse": 0.08278892386358781, "envelope_correlation": 0.9950897838639877, "rms_gain_error_db": -0.03286895051369177, "reference_mean_power": 0.0001059674543662489, "output_mean_power": 0.0001051684822613777}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.064372154195953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3923869911506206e-05, "output_power": 3.341806892270001e-05, "reference_active": true, "delta_db": -0.06524038751803542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.204493449095079e-06, "output_power": 4.568541712924837e-06, "reference_active": true, "delta_db": 0.36063914423570853}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4577951814618017e-07, "output_power": 1.825336626658791e-07, "reference_active": true, "delta_db": -1.2920272014013343}], "active_band_count": 3, "active_band_mae_db": 0.5726355777183594}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8800571346246051, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "03a2996c893f1c961adabdc204a59d8e6c358eec0e7c1d22830831dca709bfdd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9904020676787448, "standard_deviation": 0.0021794939381606492, "minimum": 0.9866751976898553, "maximum": 0.9932273600888113, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1407859083352754, "standard_deviation": 0.022295898278028945, "minimum": 1.0994396732733087, "maximum": 1.1692746627614024, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.32428158865751777, "standard_deviation": 0.11339311546954621, "minimum": 0.20083923696597972, "maximum": 0.5332519041591156, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11227776874072865, "standard_deviation": 0.01177488683633942, "minimum": 0.09491192155904044, "maximum": 0.13097463136448076, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969764237632989, "standard_deviation": 0.0009572264881594413, "minimum": 0.9947653659928972, "maximum": 0.9978791773678717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6712021539461688, "standard_deviation": 0.01933130097541845, "minimum": 0.6539450661061909, "maximum": 0.7114160669676167, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5082861653631393, "standard_deviation": 0.14236609798545785, "minimum": 0.2222460056405909, "maximum": 0.662057840831782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06452543927304649, "standard_deviation": 0.008408441500892306, "minimum": 0.054745152330653246, "maximum": 0.08060554394071176, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9848395268441628, "standard_deviation": 0.0014644238112068455, "minimum": 0.9824788796881299, "maximum": 0.9867588765317102, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1654575700765557, "standard_deviation": 0.019214390106051672, "minimum": 1.1401265374670133, "maximum": 1.1963267256207222, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.651183182343883, "standard_deviation": 0.31188296515024483, "minimum": 0.2343917824775462, "maximum": 1.0956447745783053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15045196239095243, "standard_deviation": 0.006003526614624739, "minimum": 0.14436012286488772, "maximum": 0.16096437886111814, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9949708175235027, "standard_deviation": 0.00060728929828714, "minimum": 0.9938638293477077, "maximum": 0.995665119755829, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7246614006750899, "standard_deviation": 0.018235951932167347, "minimum": 0.6873618958695619, "maximum": 0.7412644738719603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8310953210243504, "standard_deviation": 0.38586340403158126, "minimum": 0.4520697056593754, "maximum": 1.4173422883128115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08637552141646157, "standard_deviation": 0.0037080470194058145, "minimum": 0.08278892386358781, "maximum": 0.09266091953480575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913241762662754, "standard_deviation": 0.0014680651226074805, "minimum": 0.9882860600943499, "maximum": 0.992741904574443, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.247049377850997, "standard_deviation": 0.046948272146484, "minimum": 1.1829971723861386, "maximum": 1.331493224509837, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6300079048978141, "standard_deviation": 0.1614826020377561, "minimum": 0.32988371055641447, "maximum": 0.9007937578150771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09927311837450367, "standard_deviation": 0.00796268121260517, "minimum": 0.09157258328309213, "maximum": 0.11525794998883437, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974737721501925, "standard_deviation": 0.0006681109495769163, "minimum": 0.9964156761461864, "maximum": 0.9983253903961847, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6734232711693913, "standard_deviation": 0.028825873594897154, "minimum": 0.6090629801387236, "maximum": 0.6933980416594806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7927354500578023, "standard_deviation": 0.246272673421583, "minimum": 0.44577778217307057, "maximum": 1.1680630857346344, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05184038361009838, "standard_deviation": 0.006788441315756268, "minimum": 0.042109038539923, "maximum": 0.06134099279623376, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11993737390921637, "standard_deviation": 0.048914894192764936, "minimum": 0.05867200235518951, "maximum": 0.1755633060977362, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.355987814878682, "standard_deviation": 0.011184256748465638, "minimum": 5.342609226606029, "maximum": 5.373561810741897, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31537153422251407, "standard_deviation": 0.010202432290341434, "minimum": 0.30286765050675885, "maximum": 0.33035345302073676, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08204157949420987, "standard_deviation": 0.05872271348384061, "minimum": -0.15578894213953473, "maximum": -0.013381191483432982, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9898764331322887, "standard_deviation": 0.0013536301879209876, "minimum": 0.9882175472644035, "maximum": 0.9914477075192404, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8948036820065066, "standard_deviation": 0.026817785638039295, "minimum": 0.8532391634766715, "maximum": 0.9326752301866428, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3800205115016726, "standard_deviation": 0.12678615176891161, "minimum": 0.22853556178928047, "maximum": 0.5335125892968589, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08671259884254756, "standard_deviation": 0.005878912817538364, "minimum": 0.07956267995531449, "maximum": 0.09368497038748061, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972206856808652, "standard_deviation": 0.0007826251010011663, "minimum": 0.9960707000288005, "maximum": 0.9981222411457692, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.448855488467316, "standard_deviation": 0.025303255748359302, "minimum": 0.4194894995772396, "maximum": 0.4997615464121176, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5344505089323397, "standard_deviation": 0.15898107629033223, "minimum": 0.32155184331030884, "maximum": 0.8062791359774977, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04404719520812539, "standard_deviation": 0.006680032315819741, "minimum": 0.03570945734930326, "maximum": 0.05406752258552925, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0006", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.0289130210876465, "predicted_mos": 2.0289130210876465, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e72017c55e1a046a015a8b4a6bc703ff279b6dee1ed742f9af9b65c33e6077bb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0031865001656115, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8678347411611994, "全帯域_10ms_rms_envelope_nrmse": 0.08643714572712016, "全帯域_10ms_envelope_correlation": 0.9899150961591262, "全帯域_10ms_modulation_mae_db": 0.4343846093279959, "全帯域_40ms_log_rms_mae_db": 0.44906277372615533, "全帯域_40ms_rms_envelope_nrmse": 0.047258152595198395, "全帯域_40ms_envelope_correlation": 0.9967653192757001, "全帯域_40ms_modulation_mae_db": 0.46747179860501736, "300–1000 Hz_10ms_log_rms_mae_db": 1.3985817626475407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10785383965275917, "300–1000 Hz_10ms_envelope_correlation": 0.990015542699496, "300–1000 Hz_10ms_modulation_mae_db": 0.8928642395797391, "300–1000 Hz_40ms_log_rms_mae_db": 0.7881498841881854, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060004153762528746, "300–1000 Hz_40ms_envelope_correlation": 0.9969738036444342, "300–1000 Hz_40ms_modulation_mae_db": 1.0376843966048204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3152340536403833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12226936345847685, "1000–3000 Hz_10ms_envelope_correlation": 0.9889148739717155, "1000–3000 Hz_10ms_modulation_mae_db": 0.88253129560562, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7962208202430375, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07865923339900473, "1000–3000 Hz_40ms_envelope_correlation": 0.9957540888083154, "1000–3000 Hz_40ms_modulation_mae_db": 0.928472604586914, "3000–6000 Hz_10ms_log_rms_mae_db": 1.310571065977729, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15340506297308767, "3000–6000 Hz_10ms_envelope_correlation": 0.986417726469633, "3000–6000 Hz_10ms_modulation_mae_db": 0.7970180115417019, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8092865519623621, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10359844130554524, "3000–6000 Hz_40ms_envelope_correlation": 0.9945304427088273, "3000–6000 Hz_40ms_modulation_mae_db": 1.062435948170091, "mr_stft_db": 5.463193165679956, "band_mae_db": 0.16673375572190582, "rms_gain_error_db": -0.08041599966046328, "output_peak": 0.29451199140071366}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.87516727347211, "全帯域_10ms_rms_envelope_nrmse": 0.08075422273970718, "全帯域_10ms_envelope_correlation": 0.9912202668346005, "全帯域_10ms_modulation_mae_db": 0.331040322547164, "全帯域_40ms_log_rms_mae_db": 0.45291932519127454, "全帯域_40ms_rms_envelope_nrmse": 0.042077808331397855, "全帯域_40ms_envelope_correlation": 0.9973968681949638, "全帯域_40ms_modulation_mae_db": 0.4216086240281101, "300–1000 Hz_10ms_log_rms_mae_db": 1.3595065183893753, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08658308357666068, "300–1000 Hz_10ms_envelope_correlation": 0.9934278019906537, "300–1000 Hz_10ms_modulation_mae_db": 0.48701093147389146, "300–1000 Hz_40ms_log_rms_mae_db": 0.7348684842818781, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04251599549765786, "300–1000 Hz_40ms_envelope_correlation": 0.9982883594877316, "300–1000 Hz_40ms_modulation_mae_db": 0.48619950196798145, "1000–3000 Hz_10ms_log_rms_mae_db": 1.280408485822423, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11695601008807141, "1000–3000 Hz_10ms_envelope_correlation": 0.9893621759450695, "1000–3000 Hz_10ms_modulation_mae_db": 0.19879932935267797, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7702387338986405, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058856884493384005, "1000–3000 Hz_40ms_envelope_correlation": 0.9971887871813706, "1000–3000 Hz_40ms_modulation_mae_db": 0.6665896592262631, "3000–6000 Hz_10ms_log_rms_mae_db": 1.332150040986938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15192784621148211, "3000–6000 Hz_10ms_envelope_correlation": 0.9842591362653269, "3000–6000 Hz_10ms_modulation_mae_db": 0.4678577893659694, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7961416912432276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09002263289485143, "3000–6000 Hz_40ms_envelope_correlation": 0.9943914251674363, "3000–6000 Hz_40ms_modulation_mae_db": 0.6183038899486605, "mr_stft_db": 5.449973218840429, "band_mae_db": 0.06860920793884318, "rms_gain_error_db": -0.03992089464670209, "output_peak": 0.30081788164453316}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8724660316382258, "全帯域_10ms_rms_envelope_nrmse": 0.09168626862694591, "全帯域_10ms_envelope_correlation": 0.988705436027425, "全帯域_10ms_modulation_mae_db": 0.22122321242370732, "全帯域_40ms_log_rms_mae_db": 0.45509637597062363, "全帯域_40ms_rms_envelope_nrmse": 0.04895434681294677, "全帯域_40ms_envelope_correlation": 0.9964599879659428, "全帯域_40ms_modulation_mae_db": 0.19102299261773778, "300–1000 Hz_10ms_log_rms_mae_db": 1.3405038869415946, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09318570466432569, "300–1000 Hz_10ms_envelope_correlation": 0.9923781962771014, "300–1000 Hz_10ms_modulation_mae_db": 0.21146249748013182, "300–1000 Hz_40ms_log_rms_mae_db": 0.7322200938235689, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05276561066604743, "300–1000 Hz_40ms_envelope_correlation": 0.9973634861312662, "300–1000 Hz_40ms_modulation_mae_db": 0.2167641285373059, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2536426489090695, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11521194040337342, "1000–3000 Hz_10ms_envelope_correlation": 0.9899159041190481, "1000–3000 Hz_10ms_modulation_mae_db": 0.4354533385773253, "1000–3000 Hz_40ms_log_rms_mae_db": 0.744002653562283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06653646207769397, "1000–3000 Hz_40ms_envelope_correlation": 0.9967897962306915, "1000–3000 Hz_40ms_modulation_mae_db": 0.7806666983288831, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3319513312705247, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19034305538010887, "3000–6000 Hz_10ms_envelope_correlation": 0.9762582855268871, "3000–6000 Hz_10ms_modulation_mae_db": 0.339654698492845, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8077011821764621, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10032917237723166, "3000–6000 Hz_40ms_envelope_correlation": 0.9933348048231818, "3000–6000 Hz_40ms_modulation_mae_db": 0.48999617836529635, "mr_stft_db": 5.4678153485849705, "band_mae_db": 0.1778705736827307, "rms_gain_error_db": 0.013048267713111369, "output_peak": 0.288436965997707}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8339270003444572, "全帯域_10ms_rms_envelope_nrmse": 0.0889435257848823, "全帯域_10ms_envelope_correlation": 0.9893815745433413, "全帯域_10ms_modulation_mae_db": 0.34258123995775164, "全帯域_40ms_log_rms_mae_db": 0.40804702822212496, "全帯域_40ms_rms_envelope_nrmse": 0.049655760007127896, "全帯域_40ms_envelope_correlation": 0.9964002616391415, "全帯域_40ms_modulation_mae_db": 0.4378822581291466, "300–1000 Hz_10ms_log_rms_mae_db": 1.4089803826241671, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11817531684650327, "300–1000 Hz_10ms_envelope_correlation": 0.9877104323942236, "300–1000 Hz_10ms_modulation_mae_db": 0.24500809001370294, "300–1000 Hz_40ms_log_rms_mae_db": 0.7997594938036331, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07111323126184795, "300–1000 Hz_40ms_envelope_correlation": 0.9951375720731984, "300–1000 Hz_40ms_modulation_mae_db": 0.4491405640287238, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2787329997458468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10112256572382536, "1000–3000 Hz_10ms_envelope_correlation": 0.9921686834434665, "1000–3000 Hz_10ms_modulation_mae_db": 0.4862077814825896, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7803859351809515, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05814772057015747, "1000–3000 Hz_40ms_envelope_correlation": 0.9974486944579809, "1000–3000 Hz_40ms_modulation_mae_db": 0.37149585640385835, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3616805470668354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14297695263022336, "3000–6000 Hz_10ms_envelope_correlation": 0.9860097138059936, "3000–6000 Hz_10ms_modulation_mae_db": 0.871067997345787, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8267862824118494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08278364889946924, "3000–6000 Hz_40ms_envelope_correlation": 0.995103657400514, "3000–6000 Hz_40ms_modulation_mae_db": 1.0302007222519656, "mr_stft_db": 5.4537673700477, "band_mae_db": 0.09737778882114125, "rms_gain_error_db": -0.0407771073872051, "output_peak": 0.2933015143747868}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8422113163977649, "全帯域_10ms_rms_envelope_nrmse": 0.08702087116414742, "全帯域_10ms_envelope_correlation": 0.9897310276166585, "全帯域_10ms_modulation_mae_db": 0.32944718942845824, "全帯域_40ms_log_rms_mae_db": 0.3950159773646487, "全帯域_40ms_rms_envelope_nrmse": 0.040048008390609575, "全帯域_40ms_envelope_correlation": 0.9976296040578145, "全帯域_40ms_modulation_mae_db": 0.749161044894854, "300–1000 Hz_10ms_log_rms_mae_db": 1.3994030389422711, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10287362227656845, "300–1000 Hz_10ms_envelope_correlation": 0.9907190610567739, "300–1000 Hz_10ms_modulation_mae_db": 0.4754810267886264, "300–1000 Hz_40ms_log_rms_mae_db": 0.757212766170136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05444195452868138, "300–1000 Hz_40ms_envelope_correlation": 0.9971629721003246, "300–1000 Hz_40ms_modulation_mae_db": 0.9240204109483129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3149039921569672, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12145046845098532, "1000–3000 Hz_10ms_envelope_correlation": 0.9886867227015056, "1000–3000 Hz_10ms_modulation_mae_db": 0.6727289785649567, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7221324616049534, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06379432114001934, "1000–3000 Hz_40ms_envelope_correlation": 0.9970441478586165, "1000–3000 Hz_40ms_modulation_mae_db": 0.33493860669015646, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3307803577992428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1550173721439385, "3000–6000 Hz_10ms_envelope_correlation": 0.9839714152177556, "3000–6000 Hz_10ms_modulation_mae_db": 0.9601224979497932, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8199125547773548, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09279721010778696, "3000–6000 Hz_40ms_envelope_correlation": 0.9940282932394313, "3000–6000 Hz_40ms_modulation_mae_db": 1.248024249459765, "mr_stft_db": 5.4585342235299885, "band_mae_db": 0.10909386928752592, "rms_gain_error_db": -0.04116752647986687, "output_peak": 0.2906622764579036}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8091706722166625, "全帯域_10ms_rms_envelope_nrmse": 0.07825787573523947, "全帯域_10ms_envelope_correlation": 0.9917460211165453, "全帯域_10ms_modulation_mae_db": 0.15650283399531623, "全帯域_40ms_log_rms_mae_db": 0.3961594366547138, "全帯域_40ms_rms_envelope_nrmse": 0.03916653860559867, "全帯域_40ms_envelope_correlation": 0.9977266141220894, "全帯域_40ms_modulation_mae_db": 0.10088598799639233, "300–1000 Hz_10ms_log_rms_mae_db": 1.4063808060716905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09733584868280946, "300–1000 Hz_10ms_envelope_correlation": 0.9917741925970547, "300–1000 Hz_10ms_modulation_mae_db": 0.1915843295211803, "300–1000 Hz_40ms_log_rms_mae_db": 0.7711820814782926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053552280437137244, "300–1000 Hz_40ms_envelope_correlation": 0.997315688011492, "300–1000 Hz_40ms_modulation_mae_db": 0.22465336140939804, "1000–3000 Hz_10ms_log_rms_mae_db": 1.241251145922338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11236462201925354, "1000–3000 Hz_10ms_envelope_correlation": 0.9901974745120888, "1000–3000 Hz_10ms_modulation_mae_db": 0.36737273966297695, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7148167764888457, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06238134416784224, "1000–3000 Hz_40ms_envelope_correlation": 0.9968009048650296, "1000–3000 Hz_40ms_modulation_mae_db": 0.18807815255995172, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3088309588445743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14635371010480433, "3000–6000 Hz_10ms_envelope_correlation": 0.9858604011242754, "3000–6000 Hz_10ms_modulation_mae_db": 0.7076710889385023, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7777093217122649, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08244693340986334, "3000–6000 Hz_40ms_envelope_correlation": 0.9954346790056013, "3000–6000 Hz_40ms_modulation_mae_db": 0.7420394976054063, "mr_stft_db": 5.453123979395758, "band_mae_db": 0.08780424914726982, "rms_gain_error_db": 0.0125514750932782, "output_peak": 0.2989126231742527}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.905367721728025, "全帯域_10ms_rms_envelope_nrmse": 0.09367291783998839, "全帯域_10ms_envelope_correlation": 0.9882790074556894, "全帯域_10ms_modulation_mae_db": 0.09763847266347746, "全帯域_40ms_log_rms_mae_db": 0.4505274940012691, "全帯域_40ms_rms_envelope_nrmse": 0.044737960990773595, "全帯域_40ms_envelope_correlation": 0.9970948200435824, "全帯域_40ms_modulation_mae_db": 0.21353822177023707, "300–1000 Hz_10ms_log_rms_mae_db": 1.3629303188449526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09726774991958131, "300–1000 Hz_10ms_envelope_correlation": 0.9916879650320712, "300–1000 Hz_10ms_modulation_mae_db": 0.4383374031399759, "300–1000 Hz_40ms_log_rms_mae_db": 0.7473939906057537, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05447217445934556, "300–1000 Hz_40ms_envelope_correlation": 0.997151034215229, "300–1000 Hz_40ms_modulation_mae_db": 0.7579522085130249, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2708898618127562, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12130231028199255, "1000–3000 Hz_10ms_envelope_correlation": 0.98857742732299, "1000–3000 Hz_10ms_modulation_mae_db": 0.18611893279733271, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7299187831723647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056420006420687265, "1000–3000 Hz_40ms_envelope_correlation": 0.997394682782508, "1000–3000 Hz_40ms_modulation_mae_db": 0.23063966338652606, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3272510763359104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16060227592252857, "3000–6000 Hz_10ms_envelope_correlation": 0.9830046336099014, "3000–6000 Hz_10ms_modulation_mae_db": 0.5214630389963925, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8103847009092894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0849566637475596, "3000–6000 Hz_40ms_envelope_correlation": 0.9951657628284254, "3000–6000 Hz_40ms_modulation_mae_db": 0.7861384786930542, "mr_stft_db": 5.462793543458058, "band_mae_db": 0.08867191100449821, "rms_gain_error_db": 0.03319758671458508, "output_peak": 0.3265190409888072}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8362398254135425, "全帯域_10ms_rms_envelope_nrmse": 0.07764042624271882, "全帯域_10ms_envelope_correlation": 0.9920375813363318, "全帯域_10ms_modulation_mae_db": 0.20761713133373416, "全帯域_40ms_log_rms_mae_db": 0.407323181075088, "全帯域_40ms_rms_envelope_nrmse": 0.04388259870647547, "全帯域_40ms_envelope_correlation": 0.997452845927914, "全帯域_40ms_modulation_mae_db": 0.17942810209806884, "300–1000 Hz_10ms_log_rms_mae_db": 1.3577196360861354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08317802118027112, "300–1000 Hz_10ms_envelope_correlation": 0.9940767618530131, "300–1000 Hz_10ms_modulation_mae_db": 0.3635892788281157, "300–1000 Hz_40ms_log_rms_mae_db": 0.7090664622347594, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04985133927512849, "300–1000 Hz_40ms_envelope_correlation": 0.9978178357094892, "300–1000 Hz_40ms_modulation_mae_db": 0.2665879584598126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.270706519988641, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11921939136219524, "1000–3000 Hz_10ms_envelope_correlation": 0.9889722481602432, "1000–3000 Hz_10ms_modulation_mae_db": 0.2957684692077416, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7365165358952148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06034789557720321, "1000–3000 Hz_40ms_envelope_correlation": 0.99716992947759, "1000–3000 Hz_40ms_modulation_mae_db": 0.44746276157461756, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3180694459828426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1619896833995913, "3000–6000 Hz_10ms_envelope_correlation": 0.9820110312334125, "3000–6000 Hz_10ms_modulation_mae_db": 0.5179965936483418, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7601674234603227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08582290934825994, "3000–6000 Hz_40ms_envelope_correlation": 0.9948079246256695, "3000–6000 Hz_40ms_modulation_mae_db": 0.6460172518738084, "mr_stft_db": 5.45383009137603, "band_mae_db": 0.14868950291004743, "rms_gain_error_db": -0.13461870912706012, "output_peak": 0.2905735206868811}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.478735409582556, "spectral_convergence": 0.37469992579414846, "spectral_convergence_status": "ok", "amplitude_floor": 6.641091871383268e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2947699678961175, "spectral_convergence": 0.3634413388798525, "spectral_convergence_status": "ok", "amplitude_floor": 5.299135910844449e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.616074119561195, "spectral_convergence": 0.3827448829668065, "spectral_convergence_status": "ok", "amplitude_floor": 3.9942669914893874e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.486389753340235e-05, "output_power": 7.31165560560787e-05, "delta_db": -0.10256706405517524, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003919851371758182, "output_power": 0.00038466791824178274, "delta_db": -0.08183633128286107, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038865006012116416, "output_power": 0.0003738190849693119, "delta_db": -0.16897269132881415, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001490171424082218, "output_power": 0.00014311935785805022, "delta_db": -0.17537852007241223, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.76347124356667e-05, "output_power": 9.143842437855187e-05, "delta_db": -0.2847551756308283, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002530434948901305, "output_power": 0.0002641705816180704, "delta_db": 0.18689275196134386, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16673375572190582, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08041599966046328, "pitch": null, "output_rms_dbfs": -28.760399873070188, "output_peak": 0.29451199140071366, "reference_rms_dbfs": -28.67998387340972, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.463193165679956}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.947854524209456e-05, "log_rms_mae_db": 0.8678347411611994, "rms_envelope_nrmse": 0.08643714572712016, "envelope_correlation": 0.9899150961591262, "rms_gain_error_db": -0.07929869051241129, "reference_mean_power": 0.0013552494150092414, "output_mean_power": 0.0013307281974006225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.097356675930481e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024105831474109917, "output_power": 0.0002332079660215933, "reference_active": true, "delta_db": -0.1437873488204596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7795131277493674e-05, "output_power": 2.3814336257323326e-05, "reference_active": true, "delta_db": -0.6713024801723024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0822198872325408e-05, "output_power": 9.671840157583696e-06, "reference_active": true, "delta_db": -0.48806399899122554}], "active_band_count": 3, "active_band_mae_db": 0.4343846093279959}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.267346213866395e-05, "log_rms_mae_db": 0.44906277372615533, "rms_envelope_nrmse": 0.047258152595198395, "envelope_correlation": 0.9967653192757001, "rms_gain_error_db": -0.07909923338072186, "reference_mean_power": 0.0013563897957452433, "output_mean_power": 0.0013319091134354547}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.582417446963071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021668841148527672, "output_power": 0.00021050675146144866, "reference_active": true, "delta_db": -0.12569656563815126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1881573874210085e-05, "output_power": 1.0205281254135007e-05, "reference_active": true, "delta_db": -0.6604899437145628}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.720949484930854e-07, "output_power": 5.831868766688022e-07, "reference_active": true, "delta_db": -0.616228886462338}], "active_band_count": 3, "active_band_mae_db": 0.46747179860501736}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.299288573172057e-05, "log_rms_mae_db": 1.3985817626475407, "rms_envelope_nrmse": 0.10785383965275917, "envelope_correlation": 0.990015542699496, "rms_gain_error_db": -0.1822365623031439, "reference_mean_power": 0.0005726237314993749, "output_mean_power": 0.0005490927209073463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0939920051603367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018116410089920085, "output_power": 0.00016889327757984267, "reference_active": true, "delta_db": -0.30459779184600344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.90482495868649e-05, "output_power": 2.3813726524694122e-05, "reference_active": true, "delta_db": -0.8629260553390186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.620984223411989e-06, "output_power": 3.969222816704556e-06, "reference_active": true, "delta_db": -1.5110688715541956}], "active_band_count": 3, "active_band_mae_db": 0.8928642395797391}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.559252308295269e-05, "log_rms_mae_db": 0.7881498841881854, "rms_envelope_nrmse": 0.060004153762528746, "envelope_correlation": 0.9969738036444342, "rms_gain_error_db": -0.1817559972712229, "reference_mean_power": 0.0005714831913152958, "output_mean_power": 0.0005480596910515544}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7768006225472703e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016765591938032763, "output_power": 0.0001557096876877483, "reference_active": true, "delta_db": -0.32103257924193573}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3133776830395815e-05, "output_power": 1.0929603786250706e-05, "reference_active": true, "delta_db": -0.7978521402720975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.763861134744573e-07, "output_power": 3.0098319388317677e-07, "reference_active": true, "delta_db": -1.994168470300428}], "active_band_count": 3, "active_band_mae_db": 1.0376843966048204}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.686331455816738e-05, "log_rms_mae_db": 1.3152340536403833, "rms_envelope_nrmse": 0.12226936345847685, "envelope_correlation": 0.9889148739717155, "rms_gain_error_db": -0.28973268487315845, "reference_mean_power": 0.00020459070686192204, "output_mean_power": 0.00019138708789211622}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3670238086467597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.952541720107036e-05, "output_power": 6.380130721270892e-05, "reference_active": true, "delta_db": -0.3731402662587815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.449912227458803e-06, "output_power": 5.875859256471069e-06, "reference_active": true, "delta_db": -1.0307977140070985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5578562674598188e-06, "output_power": 1.1699351659429444e-06, "reference_active": true, "delta_db": -1.24365590655098}], "active_band_count": 3, "active_band_mae_db": 0.88253129560562}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4935905625855925e-05, "log_rms_mae_db": 0.7962208202430375, "rms_envelope_nrmse": 0.07865923339900473, "envelope_correlation": 0.9957540888083154, "rms_gain_error_db": -0.29259945150745503, "reference_mean_power": 0.00020499017149025147, "output_mean_power": 0.00019163423335136212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2857171527439997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.293887842389662e-05, "output_power": 5.774550206079211e-05, "reference_active": true, "delta_db": -0.3740083786593925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.713520804403163e-06, "output_power": 2.979053144350519e-06, "reference_active": true, "delta_db": -0.9570761051110476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3334126499972412e-07, "output_power": 9.539623569771186e-08, "reference_active": true, "delta_db": -1.4543333299903023}], "active_band_count": 3, "active_band_mae_db": 0.928472604586914}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.417692344469317e-05, "log_rms_mae_db": 1.310571065977729, "rms_envelope_nrmse": 0.15340506297308767, "envelope_correlation": 0.986417726469633, "rms_gain_error_db": 0.407433671872279, "reference_mean_power": 0.00010545678906217645, "output_mean_power": 0.0001158291611210304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.569551408613973e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4047256300828896e-05, "output_power": 3.6212919921612727e-05, "reference_active": true, "delta_db": 0.2678142425396441}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.550080432854394e-06, "output_power": 6.949011527638349e-06, "reference_active": true, "delta_db": -0.360285461728625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.516411722896473e-06, "output_power": 2.3431726319702426e-06, "reference_active": true, "delta_db": -1.7629543303568367}], "active_band_count": 3, "active_band_mae_db": 0.7970180115417019}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.848412221971719e-05, "log_rms_mae_db": 0.8092865519623621, "rms_envelope_nrmse": 0.10359844130554524, "envelope_correlation": 0.9945304427088273, "rms_gain_error_db": 0.40746778126947447, "reference_mean_power": 0.0001059674543662489, "output_mean_power": 0.00011639096786083788}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.064372154195953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3923869911506206e-05, "output_power": 3.548409373355766e-05, "reference_active": true, "delta_db": 0.1952832864935165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.204493449095079e-06, "output_power": 3.646712881649011e-06, "reference_active": true, "delta_db": -0.6181210900505641}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4577951814618017e-07, "output_power": 1.4228376271706222e-07, "reference_active": true, "delta_db": -2.3739034679661923}], "active_band_count": 3, "active_band_mae_db": 1.062435948170091}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8652336567227503, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "46eabc29fe7473e78ca2a1721c00c5711e3e4585c61eba451b425d5ee22bfc12", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895994387720158, "standard_deviation": 0.001186511329770644, "minimum": 0.98857742732299, "maximum": 0.9921686834434665, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2782212134998032, "standard_deviation": 0.026190875567120245, "minimum": 1.241251145922338, "maximum": 1.3152340536403833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.44062260815640264, "standard_deviation": 0.2393026381413901, "minimum": 0.18611893279733271, "maximum": 0.88253129560562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11623708397352171, "standard_deviation": 0.00700133503612283, "minimum": 0.10112256572382536, "maximum": 0.12226936345847685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969488789577629, "standard_deviation": 0.0005396796911588643, "minimum": 0.9957540888083154, "maximum": 0.9974486944579809, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7492790875057864, "standard_deviation": 0.02952143062410864, "minimum": 0.7148167764888457, "maximum": 0.7962208202430375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.49354300034464627, "standard_deviation": 0.2688999043702018, "minimum": 0.18807815255995172, "maximum": 0.928472604586914, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06314298348074904, "standard_deviation": 0.00706710970135731, "minimum": 0.056420006420687265, "maximum": 0.07865923339900473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9834740429066482, "standard_deviation": 0.0032967789253144295, "minimum": 0.9762582855268871, "maximum": 0.986417726469633, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3276606030330749, "standard_deviation": 0.016666564675249598, "minimum": 1.3088309588445743, "maximum": 1.3616805470668354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6478564645349166, "standard_deviation": 0.21823550717467288, "minimum": 0.339654698492845, "maximum": 0.9601224979497932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1578269948457206, "standard_deviation": 0.014622545147317761, "minimum": 0.14297695263022336, "maximum": 0.19034305538010887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945996237248859, "standard_deviation": 0.0006855225377003976, "minimum": 0.9933348048231818, "maximum": 0.9954346790056013, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8010112135816416, "standard_deviation": 0.02222488838463433, "minimum": 0.7601674234603227, "maximum": 0.8267862824118494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8278945270460059, "standard_deviation": 0.2598746908403204, "minimum": 0.48999617836529635, "maximum": 1.248024249459765, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09034470151132093, "standard_deviation": 0.008022215729797812, "minimum": 0.08244693340986334, "maximum": 0.10359844130554524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9914737442375484, "standard_deviation": 0.0020144708672612133, "minimum": 0.9877104323942236, "maximum": 0.9940767618530131, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.379250793818466, "standard_deviation": 0.026781638742715073, "minimum": 1.3405038869415946, "maximum": 1.4089803826241671, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.41316722460317046, "standard_deviation": 0.22696692982587033, "minimum": 0.1915843295211803, "maximum": 0.8928642395797391, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0983066483499349, "standard_deviation": 0.011347329815871162, "minimum": 0.08317802118027112, "maximum": 0.11817531684650327, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971513439216457, "standard_deviation": 0.0009177582871823572, "minimum": 0.9951375720731984, "maximum": 0.9982883594877316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7549816570732759, "standard_deviation": 0.03038097109256249, "minimum": 0.7090664622347594, "maximum": 0.7997594938036331, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5453753163086725, "standard_deviation": 0.3234330254946804, "minimum": 0.2167641285373059, "maximum": 1.0376843966048204, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054839592486046834, "standard_deviation": 0.008233289655544734, "minimum": 0.04251599549765786, "maximum": 0.07111323126184795, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11810635731424529, "standard_deviation": 0.04073816200705413, "minimum": 0.06860920793884318, "maximum": 0.1778705736827307, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.457878867614111, "standard_deviation": 0.006210178210129239, "minimum": 5.449973218840429, "maximum": 5.4678153485849705, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2979669768406982, "standard_deviation": 0.012286977339362792, "minimum": 0.288436965997707, "maximum": 0.3265190409888072, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03476286347254035, "standard_deviation": 0.05516516572682658, "minimum": -0.13461870912706012, "maximum": 0.03319758671458508, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901270013862147, "standard_deviation": 0.0013969384810931977, "minimum": 0.9882790074556894, "maximum": 0.9920375813363318, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8552980727964985, "standard_deviation": 0.03040817923906228, "minimum": 0.8091706722166625, "maximum": 0.905367721728025, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2650543764597006, "standard_deviation": 0.11225955387524125, "minimum": 0.09763847266347746, "maximum": 0.4343846093279959, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08555165673259371, "standard_deviation": 0.006057321243861966, "minimum": 0.07764042624271882, "maximum": 0.09367291783998839, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971157901533936, "standard_deviation": 0.0005203088718786938, "minimum": 0.9964002616391415, "maximum": 0.9977266141220894, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42676894902573725, "standard_deviation": 0.02731198072200255, "minimum": 0.3950159773646487, "maximum": 0.45509637597062363, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3451248787674455, "standard_deviation": 0.2140888281018349, "minimum": 0.10088598799639233, "maximum": 0.749161044894854, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04447264680501603, "standard_deviation": 0.003938733461166255, "minimum": 0.03916653860559867, "maximum": 0.049655760007127896, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0006", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.1629719734191895, "predicted_mos": 2.1629719734191895, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b08eb0c2c83c1e9b87159e6311275b56d3596f2411ef2d1560050a14eb4a3047", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.853032400365919, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.85095095318651, "全帯域_10ms_rms_envelope_nrmse": 0.08152672132393801, "全帯域_10ms_envelope_correlation": 0.9910267875629217, "全帯域_10ms_modulation_mae_db": 0.2643948954803281, "全帯域_40ms_log_rms_mae_db": 0.39863656354943006, "全帯域_40ms_rms_envelope_nrmse": 0.041256487925681117, "全帯域_40ms_envelope_correlation": 0.9975476983411341, "全帯域_40ms_modulation_mae_db": 0.38525610499581253, "300–1000 Hz_10ms_log_rms_mae_db": 1.2777412522212226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09341839232763004, "300–1000 Hz_10ms_envelope_correlation": 0.992371804722504, "300–1000 Hz_10ms_modulation_mae_db": 0.4120780918607061, "300–1000 Hz_40ms_log_rms_mae_db": 0.6519133267199078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05500885780696406, "300–1000 Hz_40ms_envelope_correlation": 0.997191468732745, "300–1000 Hz_40ms_modulation_mae_db": 0.3642041058568906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.232828267919375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12018453643912437, "1000–3000 Hz_10ms_envelope_correlation": 0.9887853136329252, "1000–3000 Hz_10ms_modulation_mae_db": 0.22331010204438587, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7043808431860927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06398364667931182, "1000–3000 Hz_40ms_envelope_correlation": 0.9966505820879512, "1000–3000 Hz_40ms_modulation_mae_db": 0.10073706416571233, "3000–6000 Hz_10ms_log_rms_mae_db": 1.304375614079387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1499654304895519, "3000–6000 Hz_10ms_envelope_correlation": 0.9847187626608322, "3000–6000 Hz_10ms_modulation_mae_db": 0.5864620043622125, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7504557740267863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08361684931493411, "3000–6000 Hz_40ms_envelope_correlation": 0.9949945701863536, "3000–6000 Hz_40ms_modulation_mae_db": 0.7641990728428723, "mr_stft_db": 5.427767095290864, "band_mae_db": 0.09641018068618445, "rms_gain_error_db": -0.0739664902890855, "output_peak": 0.3002542027922044}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8298249772137537, "全帯域_10ms_rms_envelope_nrmse": 0.08413644661739172, "全帯域_10ms_envelope_correlation": 0.9904210221612778, "全帯域_10ms_modulation_mae_db": 0.35883845175588447, "全帯域_40ms_log_rms_mae_db": 0.42848917478509346, "全帯域_40ms_rms_envelope_nrmse": 0.04262396652513536, "全帯域_40ms_envelope_correlation": 0.9973370262772332, "全帯域_40ms_modulation_mae_db": 0.5164845137685546, "300–1000 Hz_10ms_log_rms_mae_db": 1.323903179798288, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0977983506596134, "300–1000 Hz_10ms_envelope_correlation": 0.9916099662883938, "300–1000 Hz_10ms_modulation_mae_db": 0.34208957804352796, "300–1000 Hz_40ms_log_rms_mae_db": 0.6904737082630983, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047247603249683225, "300–1000 Hz_40ms_envelope_correlation": 0.9978645054225161, "300–1000 Hz_40ms_modulation_mae_db": 0.5245554465061374, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2804597271454359, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11114409974560258, "1000–3000 Hz_10ms_envelope_correlation": 0.9906055958430346, "1000–3000 Hz_10ms_modulation_mae_db": 0.38314102288525936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7941789586693655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07863753926048037, "1000–3000 Hz_40ms_envelope_correlation": 0.9952426135026005, "1000–3000 Hz_40ms_modulation_mae_db": 0.6494945033192839, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2753540984602951, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16843025585659607, "3000–6000 Hz_10ms_envelope_correlation": 0.9806191861878093, "3000–6000 Hz_10ms_modulation_mae_db": 0.8625319625884754, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7142215854629309, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08541571192760264, "3000–6000 Hz_40ms_envelope_correlation": 0.9950944236650249, "3000–6000 Hz_40ms_modulation_mae_db": 1.0166635818655483, "mr_stft_db": 5.421061291819579, "band_mae_db": 0.129542993531407, "rms_gain_error_db": -0.05839463005561506, "output_peak": 0.33443533701017125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8774461748322511, "全帯域_10ms_rms_envelope_nrmse": 0.07882820088267939, "全帯域_10ms_envelope_correlation": 0.9916074459149373, "全帯域_10ms_modulation_mae_db": 0.4849017690075342, "全帯域_40ms_log_rms_mae_db": 0.43999034594119085, "全帯域_40ms_rms_envelope_nrmse": 0.040512809238630626, "全帯域_40ms_envelope_correlation": 0.9976279517470608, "全帯域_40ms_modulation_mae_db": 0.4356253836654536, "300–1000 Hz_10ms_log_rms_mae_db": 1.2934147952827262, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08377887040862886, "300–1000 Hz_10ms_envelope_correlation": 0.9938863391523635, "300–1000 Hz_10ms_modulation_mae_db": 0.5358538960212723, "300–1000 Hz_40ms_log_rms_mae_db": 0.6622711479210044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04623229854049558, "300–1000 Hz_40ms_envelope_correlation": 0.9980339649067583, "300–1000 Hz_40ms_modulation_mae_db": 0.8646276744731662, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2515822592405328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10513900894185339, "1000–3000 Hz_10ms_envelope_correlation": 0.9920042986846932, "1000–3000 Hz_10ms_modulation_mae_db": 0.42787454430342936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7658241599108166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06713631962989543, "1000–3000 Hz_40ms_envelope_correlation": 0.9971096286270676, "1000–3000 Hz_40ms_modulation_mae_db": 0.7200119812594439, "3000–6000 Hz_10ms_log_rms_mae_db": 1.242385259768782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14528309848010348, "3000–6000 Hz_10ms_envelope_correlation": 0.9855558231269294, "3000–6000 Hz_10ms_modulation_mae_db": 1.0764237665915366, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7404593672779647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07952808118655454, "3000–6000 Hz_40ms_envelope_correlation": 0.9954944435266109, "3000–6000 Hz_40ms_modulation_mae_db": 1.1605195865060287, "mr_stft_db": 5.426284343637417, "band_mae_db": 0.15841895263277503, "rms_gain_error_db": -0.06844172553545459, "output_peak": 0.29783793754028554}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8146636485292849, "全帯域_10ms_rms_envelope_nrmse": 0.07681208644939959, "全帯域_10ms_envelope_correlation": 0.9920218354227339, "全帯域_10ms_modulation_mae_db": 0.05920894139756868, "全帯域_40ms_log_rms_mae_db": 0.3797909414269076, "全帯域_40ms_rms_envelope_nrmse": 0.035743210121454976, "全帯域_40ms_envelope_correlation": 0.9981185413536957, "全帯域_40ms_modulation_mae_db": 0.18895323036539136, "300–1000 Hz_10ms_log_rms_mae_db": 1.3198994264483772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08957791442456833, "300–1000 Hz_10ms_envelope_correlation": 0.9929687479660599, "300–1000 Hz_10ms_modulation_mae_db": 0.18085357194105198, "300–1000 Hz_40ms_log_rms_mae_db": 0.6922320584097629, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047926993653746426, "300–1000 Hz_40ms_envelope_correlation": 0.9978025156089898, "300–1000 Hz_40ms_modulation_mae_db": 0.2078041470383948, "1000–3000 Hz_10ms_log_rms_mae_db": 1.231057414813472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10854131350466263, "1000–3000 Hz_10ms_envelope_correlation": 0.9912433372446439, "1000–3000 Hz_10ms_modulation_mae_db": 0.5702788289711579, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7146638086117669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06044843754094656, "1000–3000 Hz_40ms_envelope_correlation": 0.9976119296952842, "1000–3000 Hz_40ms_modulation_mae_db": 0.6816413740408942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2845846303908337, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14856273817028184, "3000–6000 Hz_10ms_envelope_correlation": 0.9848946028779906, "3000–6000 Hz_10ms_modulation_mae_db": 1.0955190342634005, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7885817037047771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09026443874925007, "3000–6000 Hz_40ms_envelope_correlation": 0.9941887747099699, "3000–6000 Hz_40ms_modulation_mae_db": 1.3686441192776482, "mr_stft_db": 5.418235403583961, "band_mae_db": 0.08291320268375037, "rms_gain_error_db": -0.04208297081027633, "output_peak": 0.3528880809435878}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8515633328140032, "全帯域_10ms_rms_envelope_nrmse": 0.08535466481816417, "全帯域_10ms_envelope_correlation": 0.9901690665307756, "全帯域_10ms_modulation_mae_db": 0.2720196805440942, "全帯域_40ms_log_rms_mae_db": 0.4050265389004262, "全帯域_40ms_rms_envelope_nrmse": 0.038030250520090444, "全帯域_40ms_envelope_correlation": 0.9978684213595308, "全帯域_40ms_modulation_mae_db": 0.17857068319647973, "300–1000 Hz_10ms_log_rms_mae_db": 1.31647898330733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09590242093679548, "300–1000 Hz_10ms_envelope_correlation": 0.9919175876197603, "300–1000 Hz_10ms_modulation_mae_db": 0.5672594457518392, "300–1000 Hz_40ms_log_rms_mae_db": 0.6937591315021302, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04735901197701164, "300–1000 Hz_40ms_envelope_correlation": 0.9978569328576874, "300–1000 Hz_40ms_modulation_mae_db": 0.6149186348203729, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2591665522002804, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10829935137284186, "1000–3000 Hz_10ms_envelope_correlation": 0.9910305179678924, "1000–3000 Hz_10ms_modulation_mae_db": 0.6932897324115039, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7440805672617427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06110360461022909, "1000–3000 Hz_40ms_envelope_correlation": 0.997207900273024, "1000–3000 Hz_40ms_modulation_mae_db": 0.9438462320826385, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3247420249673076, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16217678769223798, "3000–6000 Hz_10ms_envelope_correlation": 0.9831128344304251, "3000–6000 Hz_10ms_modulation_mae_db": 0.48841270366853734, "3000–6000 Hz_40ms_log_rms_mae_db": 0.782564285611394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08795403277548362, "3000–6000 Hz_40ms_envelope_correlation": 0.9952008378554368, "3000–6000 Hz_40ms_modulation_mae_db": 0.6010895690061245, "mr_stft_db": 5.4196021131651095, "band_mae_db": 0.07256328442396419, "rms_gain_error_db": -0.03593312246893804, "output_peak": 0.29935975867167386}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8270125432317585, "全帯域_10ms_rms_envelope_nrmse": 0.07669943951448635, "全帯域_10ms_envelope_correlation": 0.9920736851499853, "全帯域_10ms_modulation_mae_db": 0.20011932114773442, "全帯域_40ms_log_rms_mae_db": 0.3860009402001596, "全帯域_40ms_rms_envelope_nrmse": 0.042297611211292024, "全帯域_40ms_envelope_correlation": 0.9973497870386189, "全帯域_40ms_modulation_mae_db": 0.13246094784440424, "300–1000 Hz_10ms_log_rms_mae_db": 1.3807955759330528, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08992769497588575, "300–1000 Hz_10ms_envelope_correlation": 0.9928856614708248, "300–1000 Hz_10ms_modulation_mae_db": 0.27219746564826836, "300–1000 Hz_40ms_log_rms_mae_db": 0.7734587504902911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05447882854468748, "300–1000 Hz_40ms_envelope_correlation": 0.9971560456662103, "300–1000 Hz_40ms_modulation_mae_db": 0.577500348281299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2104561330505403, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11373887222827476, "1000–3000 Hz_10ms_envelope_correlation": 0.989923613926283, "1000–3000 Hz_10ms_modulation_mae_db": 0.4661239929318591, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6963331843814157, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06688911545098812, "1000–3000 Hz_40ms_envelope_correlation": 0.9963184501799134, "1000–3000 Hz_40ms_modulation_mae_db": 0.2328920435298877, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2548885001221353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1442572813493201, "3000–6000 Hz_10ms_envelope_correlation": 0.9860825463129628, "3000–6000 Hz_10ms_modulation_mae_db": 0.6319310082292172, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7466929203103932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07582067244680321, "3000–6000 Hz_40ms_envelope_correlation": 0.9960262266700528, "3000–6000 Hz_40ms_modulation_mae_db": 0.7650463018815271, "mr_stft_db": 5.410483133267134, "band_mae_db": 0.05359828203768994, "rms_gain_error_db": -0.007790235694291004, "output_peak": 0.3123576995365391}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8970461308794466, "全帯域_10ms_rms_envelope_nrmse": 0.0885164393912975, "全帯域_10ms_envelope_correlation": 0.9894261072903207, "全帯域_10ms_modulation_mae_db": 0.38150154383454865, "全帯域_40ms_log_rms_mae_db": 0.41643323769666707, "全帯域_40ms_rms_envelope_nrmse": 0.04408304514962882, "全帯域_40ms_envelope_correlation": 0.9971428834916256, "全帯域_40ms_modulation_mae_db": 0.4257836881405735, "300–1000 Hz_10ms_log_rms_mae_db": 1.3586072754672496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1137296790997191, "300–1000 Hz_10ms_envelope_correlation": 0.9886157943909815, "300–1000 Hz_10ms_modulation_mae_db": 0.4099242055654861, "300–1000 Hz_40ms_log_rms_mae_db": 0.7368323835365889, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06079558928430206, "300–1000 Hz_40ms_envelope_correlation": 0.996447915161478, "300–1000 Hz_40ms_modulation_mae_db": 0.6622801178129223, "1000–3000 Hz_10ms_log_rms_mae_db": 1.213752370329659, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11816183577637851, "1000–3000 Hz_10ms_envelope_correlation": 0.9894135181810615, "1000–3000 Hz_10ms_modulation_mae_db": 0.5370355298787993, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7097706239764554, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07158724820602647, "1000–3000 Hz_40ms_envelope_correlation": 0.9962701865112632, "1000–3000 Hz_40ms_modulation_mae_db": 0.7212280267218474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2715497870010466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15968497070995696, "3000–6000 Hz_10ms_envelope_correlation": 0.9826404329398546, "3000–6000 Hz_10ms_modulation_mae_db": 0.7949969427501152, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7388134270922055, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07695711079829022, "3000–6000 Hz_40ms_envelope_correlation": 0.9957515845821585, "3000–6000 Hz_40ms_modulation_mae_db": 0.8431662143950119, "mr_stft_db": 5.418704646313494, "band_mae_db": 0.11664087941921447, "rms_gain_error_db": -0.05172883211590873, "output_peak": 0.3310277727251628}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8589976660121693, "全帯域_10ms_rms_envelope_nrmse": 0.08516269623998757, "全帯域_10ms_envelope_correlation": 0.9902471885828765, "全帯域_10ms_modulation_mae_db": 0.7909130095408363, "全帯域_40ms_log_rms_mae_db": 0.4177884579525709, "全帯域_40ms_rms_envelope_nrmse": 0.04087335078901277, "全帯域_40ms_envelope_correlation": 0.9976887298562346, "全帯域_40ms_modulation_mae_db": 0.7295371642410217, "300–1000 Hz_10ms_log_rms_mae_db": 1.322555931454265, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09032408310850652, "300–1000 Hz_10ms_envelope_correlation": 0.9929147316367982, "300–1000 Hz_10ms_modulation_mae_db": 0.9253337368926892, "300–1000 Hz_40ms_log_rms_mae_db": 0.686087271405362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04847314102276214, "300–1000 Hz_40ms_envelope_correlation": 0.9978811581195725, "300–1000 Hz_40ms_modulation_mae_db": 1.162317328262615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.252442469611223, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12075502373461838, "1000–3000 Hz_10ms_envelope_correlation": 0.9888065023313068, "1000–3000 Hz_10ms_modulation_mae_db": 0.39127039843414796, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7488092453836785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06580084838910943, "1000–3000 Hz_40ms_envelope_correlation": 0.9967706320989651, "1000–3000 Hz_40ms_modulation_mae_db": 0.6516154427213738, "3000–6000 Hz_10ms_log_rms_mae_db": 1.275551462610558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1658373002045282, "3000–6000 Hz_10ms_envelope_correlation": 0.9820070024628454, "3000–6000 Hz_10ms_modulation_mae_db": 0.7732636655841306, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7258671278059663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09477233812619332, "3000–6000 Hz_40ms_envelope_correlation": 0.9940641799640534, "3000–6000 Hz_40ms_modulation_mae_db": 0.9392008173622841, "mr_stft_db": 5.400335939419308, "band_mae_db": 0.14876009109933314, "rms_gain_error_db": -0.10197224880037185, "output_peak": 0.29492835440111936}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.447497146798281, "spectral_convergence": 0.3799678659466307, "spectral_convergence_status": "ok", "amplitude_floor": 6.641091871383268e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.235496396062249, "spectral_convergence": 0.3602911108166974, "spectral_convergence_status": "ok", "amplitude_floor": 5.299135910844449e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.600307743012063, "spectral_convergence": 0.3839365162422586, "spectral_convergence_status": "ok", "amplitude_floor": 3.9942669914893874e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.486389753340235e-05, "output_power": 7.68405712830825e-05, "delta_db": 0.11318151603545626, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003919851371758182, "output_power": 0.00039044535512834536, "delta_db": -0.017093395934994513, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038865006012116416, "output_power": 0.0003836288704061651, "delta_db": -0.056474561933826745, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001490171424082218, "output_power": 0.00014542898682246951, "delta_db": -0.10585252517846563, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.76347124356667e-05, "output_power": 9.804514215633458e-05, "delta_db": 0.018218290754196367, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002530434948901305, "output_power": 0.00023792008029393022, "delta_db": -0.26764079428016724, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09641018068618445, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0739664902890855, "pitch": null, "output_rms_dbfs": -28.75395036369881, "output_peak": 0.3002542027922044, "reference_rms_dbfs": -28.67998387340972, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.427767095290864}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.947854524209456e-05, "log_rms_mae_db": 0.85095095318651, "rms_envelope_nrmse": 0.08152672132393801, "envelope_correlation": 0.9910267875629217, "rms_gain_error_db": -0.07309431886888604, "reference_mean_power": 0.0013552494150092414, "output_mean_power": 0.001332630646771665}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.097356675930481e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024105831474109917, "output_power": 0.0002386326047172495, "reference_active": true, "delta_db": -0.043923343763807035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7795131277493674e-05, "output_power": 2.764778550929117e-05, "reference_active": true, "delta_db": -0.02308378001251306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0822198872325408e-05, "output_power": 9.155830712305306e-06, "reference_active": true, "delta_db": -0.726177562664664}], "active_band_count": 3, "active_band_mae_db": 0.2643948954803281}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.267346213866395e-05, "log_rms_mae_db": 0.39863656354943006, "rms_envelope_nrmse": 0.041256487925681117, "envelope_correlation": 0.9975476983411341, "rms_gain_error_db": -0.07258649486639357, "reference_mean_power": 0.0013563897957452433, "output_mean_power": 0.0013339079606463134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.582417446963071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021668841148527672, "output_power": 0.00021418272689431228, "reference_active": true, "delta_db": -0.05051242294130351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1881573874210085e-05, "output_power": 1.263798906098619e-05, "reference_active": true, "delta_db": 0.2680400251857375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.720949484930854e-07, "output_power": 5.542542660682694e-07, "reference_active": true, "delta_db": -0.8372158668603967}], "active_band_count": 3, "active_band_mae_db": 0.38525610499581253}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.299288573172057e-05, "log_rms_mae_db": 1.2777412522212226, "rms_envelope_nrmse": 0.09341839232763004, "envelope_correlation": 0.992371804722504, "rms_gain_error_db": -0.09099535221777535, "reference_mean_power": 0.0005726237314993749, "output_mean_power": 0.0005607506781778281}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0939920051603367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018116410089920085, "output_power": 0.00018232498610841016, "reference_active": true, "delta_db": 0.027740461689840658}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.90482495868649e-05, "output_power": 2.868371339409517e-05, "reference_active": true, "delta_db": -0.05484593035874893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.620984223411989e-06, "output_power": 4.309705340798498e-06, "reference_active": true, "delta_db": -1.1536478835335287}], "active_band_count": 3, "active_band_mae_db": 0.4120780918607061}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.559252308295269e-05, "log_rms_mae_db": 0.6519133267199078, "rms_envelope_nrmse": 0.05500885780696406, "envelope_correlation": 0.997191468732745, "rms_gain_error_db": -0.09024281272430046, "reference_mean_power": 0.0005714831913152958, "output_mean_power": 0.0005597307674708743}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7768006225472703e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016765591938032763, "output_power": 0.00016865882961908014, "reference_active": true, "delta_db": 0.025901907360883325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3133776830395815e-05, "output_power": 1.3537895835450596e-05, "reference_active": true, "delta_db": 0.13161535717760253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.763861134744573e-07, "output_power": 3.84104670631344e-07, "reference_active": true, "delta_db": -0.9350950530321861}], "active_band_count": 3, "active_band_mae_db": 0.3642041058568906}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.686331455816738e-05, "log_rms_mae_db": 1.232828267919375, "rms_envelope_nrmse": 0.12018453643912437, "envelope_correlation": 0.9887853136329252, "rms_gain_error_db": -0.03005342547860709, "reference_mean_power": 0.00020459070686192204, "output_mean_power": 0.00020317981487573468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3670238086467597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.952541720107036e-05, "output_power": 6.827717727665115e-05, "reference_active": true, "delta_db": -0.07868045455632981}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.449912227458803e-06, "output_power": 7.701546068175649e-06, "reference_active": true, "delta_db": 0.1442676151571214}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5578562674598188e-06, "output_power": 1.726734671930148e-06, "reference_active": true, "delta_db": 0.44698223641970636}], "active_band_count": 3, "active_band_mae_db": 0.22331010204438587}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4935905625855925e-05, "log_rms_mae_db": 0.7043808431860927, "rms_envelope_nrmse": 0.06398364667931182, "envelope_correlation": 0.9966505820879512, "rms_gain_error_db": -0.02901070410327288, "reference_mean_power": 0.00020499017149025147, "output_mean_power": 0.0002036254083981037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2857171527439997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.293887842389662e-05, "output_power": 6.224402492934427e-05, "reference_active": true, "delta_db": -0.0482133150153128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.713520804403163e-06, "output_power": 3.7692967532695645e-06, "reference_active": true, "delta_db": 0.06474469077643898}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3334126499972412e-07, "output_power": 1.2765541806988757e-07, "reference_active": true, "delta_db": -0.18925318670538527}], "active_band_count": 3, "active_band_mae_db": 0.10073706416571233}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.417692344469317e-05, "log_rms_mae_db": 1.304375614079387, "rms_envelope_nrmse": 0.1499654304895519, "envelope_correlation": 0.9847187626608322, "rms_gain_error_db": 0.01626313138232711, "reference_mean_power": 0.00010545678906217645, "output_mean_power": 0.00010585243600415616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.569551408613973e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4047256300828896e-05, "output_power": 3.453332662798087e-05, "reference_active": true, "delta_db": 0.061562963373845675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.550080432854394e-06, "output_power": 7.796465340998444e-06, "reference_active": true, "delta_db": 0.13946174285480137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.516411722896473e-06, "output_power": 2.456199216732809e-06, "reference_active": true, "delta_db": -1.5583613068579905}], "active_band_count": 3, "active_band_mae_db": 0.5864620043622125}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.848412221971719e-05, "log_rms_mae_db": 0.7504557740267863, "rms_envelope_nrmse": 0.08361684931493411, "envelope_correlation": 0.9949945701863536, "rms_gain_error_db": 0.016112669997974182, "reference_mean_power": 0.0001059674543662489, "output_mean_power": 0.00010636133224035645}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.064372154195953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3923869911506206e-05, "output_power": 3.4054109679188735e-05, "reference_active": true, "delta_db": 0.016641412992201204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.204493449095079e-06, "output_power": 4.259246785220038e-06, "reference_active": true, "delta_db": 0.05619124125677229}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4577951814618017e-07, "output_power": 1.4742435352696238e-07, "reference_active": true, "delta_db": -2.219764564279643}], "active_band_count": 3, "active_band_mae_db": 0.7641990728428723}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.873500027162502, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bc187595127ff09b5fb01735f392a521f6fe943a3bd1ac0333fcd4566a6f4110", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9902265872264802, "standard_deviation": 0.001185072364923438, "minimum": 0.9887853136329252, "maximum": 0.9920042986846932, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2414681492888149, "standard_deviation": 0.023812505816691126, "minimum": 1.2104561330505403, "maximum": 1.2804597271454359, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46154051898256787, "standard_deviation": 0.14153430837160952, "minimum": 0.22331010204438587, "maximum": 0.6932897324115039, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11324550521791957, "standard_deviation": 0.005924305675720902, "minimum": 0.10513900894185339, "maximum": 0.12075502373461838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966477403720087, "standard_deviation": 0.0007267591313885215, "minimum": 0.9952426135026005, "maximum": 0.9976119296952842, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7347551739226668, "standard_deviation": 0.03424217932314117, "minimum": 0.6963331843814157, "maximum": 0.7941789586693655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5876833334801352, "standard_deviation": 0.2782410680830026, "minimum": 0.10073706416571233, "maximum": 0.9438462320826385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0669483449708734, "standard_deviation": 0.005912595512002502, "minimum": 0.06044843754094656, "maximum": 0.07863753926048037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9837038988749561, "standard_deviation": 0.0019052760716530644, "minimum": 0.9806191861878093, "maximum": 0.9860825463129628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2791789221750431, "standard_deviation": 0.026103861071588224, "minimum": 1.242385259768782, "maximum": 1.3247420249673076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7886926360047032, "standard_deviation": 0.21983797924200557, "minimum": 0.48841270366853734, "maximum": 1.0955190342634005, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1555247328690721, "standard_deviation": 0.009604171212351652, "minimum": 0.1442572813493201, "maximum": 0.16843025585659607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9951018801449576, "standard_deviation": 0.000693086803305508, "minimum": 0.9940641799640534, "maximum": 0.9960262266700528, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7484570239115522, "standard_deviation": 0.025689721167875903, "minimum": 0.7142215854629309, "maximum": 0.7885817037047771, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9323161578921306, "standard_deviation": 0.24597389693415436, "minimum": 0.6010895690061245, "maximum": 1.3686441192776482, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08429115441563896, "standard_deviation": 0.006642885127297463, "minimum": 0.07582067244680321, "maximum": 0.09477233812619332, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921463291559607, "standard_deviation": 0.0015899948295095192, "minimum": 0.9886157943909815, "maximum": 0.9938863391523635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.324174552489064, "standard_deviation": 0.032893678961986805, "minimum": 1.2777412522212226, "maximum": 1.3807955759330528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.45569874896560514, "standard_deviation": 0.22851863479260745, "minimum": 0.18085357194105198, "maximum": 0.9253337368926892, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09430717574266843, "standard_deviation": 0.00895235552948545, "minimum": 0.08377887040862886, "maximum": 0.1137296790997191, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975293133094947, "standard_deviation": 0.0005471596095858206, "minimum": 0.996447915161478, "maximum": 0.9980339649067583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6983784722810181, "standard_deviation": 0.039372172401221484, "minimum": 0.6519133267199078, "maximum": 0.7734587504902911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6222759753814748, "standard_deviation": 0.2932516266236676, "minimum": 0.2078041470383948, "maximum": 1.162317328262615, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.050940290509956576, "standard_deviation": 0.005209810433323544, "minimum": 0.04623229854049558, "maximum": 0.06079558928430206, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.10735598331428982, "standard_deviation": 0.03725948040028179, "minimum": 0.05359828203768994, "maximum": 0.15841895263277503, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.4178092458121085, "standard_deviation": 0.008821628775693294, "minimum": 5.400335939419308, "maximum": 5.427767095290864, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31538614295259304, "standard_deviation": 0.021498772159413777, "minimum": 0.29492835440111936, "maximum": 0.3528880809435878, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05503878197124264, "standard_deviation": 0.02810228568361308, "minimum": -0.10197224880037185, "maximum": -0.007790235694291004, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9908741423269787, "standard_deviation": 0.0009643253284794579, "minimum": 0.9894261072903207, "maximum": 0.9920736851499853, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8509381783373972, "standard_deviation": 0.027328708814783248, "minimum": 0.8146636485292849, "maximum": 0.8970461308794466, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35148720158856617, "standard_deviation": 0.21846489172248126, "minimum": 0.05920894139756868, "maximum": 0.7909130095408363, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08212958690466804, "standard_deviation": 0.004366187593740423, "minimum": 0.07669943951448635, "maximum": 0.0885164393912975, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975851299331417, "standard_deviation": 0.00031425039353233733, "minimum": 0.9971428834916256, "maximum": 0.9981185413536957, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4090195250565557, "standard_deviation": 0.02062602899235056, "minimum": 0.3797909414269076, "maximum": 0.43999034594119085, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3740839645272114, "standard_deviation": 0.20145557189182062, "minimum": 0.13246094784440424, "maximum": 0.7295371642410217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04067759143511576, "standard_deviation": 0.002666380210285609, "minimum": 0.035743210121454976, "maximum": 0.04408304514962882, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1221-135767-0019", "language": "en", "style": "normal", "speaker": "librispeech_1221", "source_utterance_id": "1221-135767-0019", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 60480, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1221-135767-0019.wav", "source_file_sha256": "d55c6fc6e5881dc060a18f66dac7425f74673ca5455efd1f8624ac3ea23d0363", "decoded_pcm_sha256": "0ed57e185be961c721604990dce623ab138f9937e79722c597c3874498c02937", "original_source_sha256": "245fd9a32160c5fa9dd1a76624a71c7e1398de763f8e9ab88a7dd25aae1002f4", "corpus_entry_id": "librispeech_1221-135767-0019"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.184542179107666, "predicted_mos": 4.184542179107666, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "70239a0fa36ebe9bbdd465391645ea523ac00560058f3ddec57c33181feb5125", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.900996900163591, "cpu_threads": 4}, "waveform_sha256": "68d777578f1606bd7012b7443ee9a0a11c0ee94589b0f4eeaaf9ab2f9397aa6b"}, "rows": [{"case": "librispeech_1221-135767-0019", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.4922969341278076, "predicted_mos": 2.4922969341278076, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "9eceece5a28c6b3d395b24abffb78fd30710377d953d2bb03ec147d2696392ae", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.801988999824971, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3220531763801835, "全帯域_10ms_rms_envelope_nrmse": 0.1285262192109336, "全帯域_10ms_envelope_correlation": 0.9856131276195729, "全帯域_10ms_modulation_mae_db": 0.5444076307975868, "全帯域_40ms_log_rms_mae_db": 0.6603248214599696, "全帯域_40ms_rms_envelope_nrmse": 0.06698716460460559, "全帯域_40ms_envelope_correlation": 0.9963698727266226, "全帯域_40ms_modulation_mae_db": 0.8487005878433939, "300–1000 Hz_10ms_log_rms_mae_db": 1.6773443304996811, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16149831707623682, "300–1000 Hz_10ms_envelope_correlation": 0.981684720544353, "300–1000 Hz_10ms_modulation_mae_db": 0.5425445563661625, "300–1000 Hz_40ms_log_rms_mae_db": 1.0009786852631837, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0897775033541427, "300–1000 Hz_40ms_envelope_correlation": 0.9946689771285516, "300–1000 Hz_40ms_modulation_mae_db": 0.506570899380748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.524278418715901, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16656552807583477, "1000–3000 Hz_10ms_envelope_correlation": 0.9800804198849119, "1000–3000 Hz_10ms_modulation_mae_db": 0.7192102496133935, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9479436960424329, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0985791320558167, "1000–3000 Hz_40ms_envelope_correlation": 0.9939753310066319, "1000–3000 Hz_40ms_modulation_mae_db": 0.8796161136269983, "3000–6000 Hz_10ms_log_rms_mae_db": 1.367298140609427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19996952229011306, "3000–6000 Hz_10ms_envelope_correlation": 0.9751295346014964, "3000–6000 Hz_10ms_modulation_mae_db": 0.9633212115796704, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8430996651420843, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1121093624064054, "3000–6000 Hz_40ms_envelope_correlation": 0.992079963683742, "3000–6000 Hz_40ms_modulation_mae_db": 0.30912962531823757, "mr_stft_db": 6.000564108044915, "band_mae_db": 0.2759229098475637, "rms_gain_error_db": -0.238942006870536, "output_peak": 0.35210558716312174}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3744949331497052, "全帯域_10ms_rms_envelope_nrmse": 0.136418106984883, "全帯域_10ms_envelope_correlation": 0.9835066015366685, "全帯域_10ms_modulation_mae_db": 0.7100652817409429, "全帯域_40ms_log_rms_mae_db": 0.7419034873225148, "全帯域_40ms_rms_envelope_nrmse": 0.07974532179680606, "全帯域_40ms_envelope_correlation": 0.9940844541113658, "全帯域_40ms_modulation_mae_db": 0.5428571226714008, "300–1000 Hz_10ms_log_rms_mae_db": 1.7191571305433095, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17701732515992322, "300–1000 Hz_10ms_envelope_correlation": 0.9776103874987867, "300–1000 Hz_10ms_modulation_mae_db": 0.9955863768972435, "300–1000 Hz_40ms_log_rms_mae_db": 1.037607624744725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11103405801112538, "300–1000 Hz_40ms_envelope_correlation": 0.9908049640422373, "300–1000 Hz_40ms_modulation_mae_db": 0.6472806891799837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6642393192345197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21297116741527086, "1000–3000 Hz_10ms_envelope_correlation": 0.965521406755027, "1000–3000 Hz_10ms_modulation_mae_db": 0.8372284814509444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9958003746178109, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11747437951752573, "1000–3000 Hz_40ms_envelope_correlation": 0.9889644386699726, "1000–3000 Hz_40ms_modulation_mae_db": 0.7008467550183801, "3000–6000 Hz_10ms_log_rms_mae_db": 1.340539055546752, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1554644786713932, "3000–6000 Hz_10ms_envelope_correlation": 0.9843685150506161, "3000–6000 Hz_10ms_modulation_mae_db": 0.7796582059698576, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8003782837049941, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08805228074691507, "3000–6000 Hz_40ms_envelope_correlation": 0.9949020461947227, "3000–6000 Hz_40ms_modulation_mae_db": 0.5129163751102099, "mr_stft_db": 6.024770772574925, "band_mae_db": 0.2341745104733335, "rms_gain_error_db": -0.11924566123672123, "output_peak": 0.3689502215537107}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.370257163408021, "全帯域_10ms_rms_envelope_nrmse": 0.12846415934822714, "全帯域_10ms_envelope_correlation": 0.985344336641704, "全帯域_10ms_modulation_mae_db": 0.4981436083716149, "全帯域_40ms_log_rms_mae_db": 0.7490015503714696, "全帯域_40ms_rms_envelope_nrmse": 0.0805252187495414, "全帯域_40ms_envelope_correlation": 0.9939117564467268, "全帯域_40ms_modulation_mae_db": 0.7023335262878003, "300–1000 Hz_10ms_log_rms_mae_db": 1.6235479924447402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16805480981291918, "300–1000 Hz_10ms_envelope_correlation": 0.9799453442305843, "300–1000 Hz_10ms_modulation_mae_db": 1.329241356425655, "300–1000 Hz_40ms_log_rms_mae_db": 0.9679973922774712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09901339515205172, "300–1000 Hz_40ms_envelope_correlation": 0.9926004828949951, "300–1000 Hz_40ms_modulation_mae_db": 1.0939921453707377, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6073144161160002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17991361232770803, "1000–3000 Hz_10ms_envelope_correlation": 0.9754880481502212, "1000–3000 Hz_10ms_modulation_mae_db": 0.7950263804768175, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9880023247718973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08940040013419844, "1000–3000 Hz_40ms_envelope_correlation": 0.9936266827778478, "1000–3000 Hz_40ms_modulation_mae_db": 0.5946041649641317, "3000–6000 Hz_10ms_log_rms_mae_db": 1.311193685937006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17707450005630926, "3000–6000 Hz_10ms_envelope_correlation": 0.9816626220529103, "3000–6000 Hz_10ms_modulation_mae_db": 1.1455944262784337, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7624953301626358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1347896439860812, "3000–6000 Hz_40ms_envelope_correlation": 0.9905064248231483, "3000–6000 Hz_40ms_modulation_mae_db": 1.123440848456479, "mr_stft_db": 6.007653568728945, "band_mae_db": 0.26723795076856544, "rms_gain_error_db": -0.10294407697929325, "output_peak": 0.4575451528352336}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2989340124062136, "全帯域_10ms_rms_envelope_nrmse": 0.12467379862198877, "全帯域_10ms_envelope_correlation": 0.9864416332764084, "全帯域_10ms_modulation_mae_db": 0.36895618423292537, "全帯域_40ms_log_rms_mae_db": 0.7056145028245477, "全帯域_40ms_rms_envelope_nrmse": 0.06602456410730959, "全帯域_40ms_envelope_correlation": 0.9959718076687809, "全帯域_40ms_modulation_mae_db": 0.7418482584534963, "300–1000 Hz_10ms_log_rms_mae_db": 1.6874672138462958, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15378163030347342, "300–1000 Hz_10ms_envelope_correlation": 0.983117358742778, "300–1000 Hz_10ms_modulation_mae_db": 0.5377177509014056, "300–1000 Hz_40ms_log_rms_mae_db": 0.9137984816394005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08736773339032161, "300–1000 Hz_40ms_envelope_correlation": 0.994241417277598, "300–1000 Hz_40ms_modulation_mae_db": 0.2585966392149837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5531760279839117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20445337570493163, "1000–3000 Hz_10ms_envelope_correlation": 0.9691768528587729, "1000–3000 Hz_10ms_modulation_mae_db": 0.5566809658605932, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9747571417443719, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10776449019712865, "1000–3000 Hz_40ms_envelope_correlation": 0.990858444443498, "1000–3000 Hz_40ms_modulation_mae_db": 0.543228645850305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3147081560108735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16882453685727633, "3000–6000 Hz_10ms_envelope_correlation": 0.9825826043641361, "3000–6000 Hz_10ms_modulation_mae_db": 1.0973537775221482, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8163880453271118, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0873720358522454, "3000–6000 Hz_40ms_envelope_correlation": 0.995454757430214, "3000–6000 Hz_40ms_modulation_mae_db": 1.1433133591907814, "mr_stft_db": 6.007013012171682, "band_mae_db": 0.21123048068282957, "rms_gain_error_db": 0.06268772996444733, "output_peak": 0.43812201208424406}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3848638036838676, "全帯域_10ms_rms_envelope_nrmse": 0.13027889186925107, "全帯域_10ms_envelope_correlation": 0.9849413460662225, "全帯域_10ms_modulation_mae_db": 0.9670991919549733, "全帯域_40ms_log_rms_mae_db": 0.7278994017223432, "全帯域_40ms_rms_envelope_nrmse": 0.07325482957573838, "全帯域_40ms_envelope_correlation": 0.9949144752351395, "全帯域_40ms_modulation_mae_db": 1.3338237645660225, "300–1000 Hz_10ms_log_rms_mae_db": 1.7160152957032286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16375994309552544, "300–1000 Hz_10ms_envelope_correlation": 0.9808196860656062, "300–1000 Hz_10ms_modulation_mae_db": 1.0433768594501949, "300–1000 Hz_40ms_log_rms_mae_db": 0.9951484027348102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09335097967512242, "300–1000 Hz_40ms_envelope_correlation": 0.9934675156110059, "300–1000 Hz_40ms_modulation_mae_db": 1.011499223367319, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5331335669401742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17369559955589747, "1000–3000 Hz_10ms_envelope_correlation": 0.9771886356789179, "1000–3000 Hz_10ms_modulation_mae_db": 0.227487839929678, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8799615816492518, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10668368287875904, "1000–3000 Hz_40ms_envelope_correlation": 0.990834264924591, "1000–3000 Hz_40ms_modulation_mae_db": 0.2415772842788507, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3082495948336572, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1807473690762788, "3000–6000 Hz_10ms_envelope_correlation": 0.9805935499849278, "3000–6000 Hz_10ms_modulation_mae_db": 1.5891299799549072, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8470383015317647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10711128496723135, "3000–6000 Hz_40ms_envelope_correlation": 0.9934065787885409, "3000–6000 Hz_40ms_modulation_mae_db": 1.7926071189157977, "mr_stft_db": 6.003803981466263, "band_mae_db": 0.17418781618607307, "rms_gain_error_db": -0.025719762526192375, "output_peak": 0.4938374794238893}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.276478926726376, "全帯域_10ms_rms_envelope_nrmse": 0.11855873977099346, "全帯域_10ms_envelope_correlation": 0.9876405169741275, "全帯域_10ms_modulation_mae_db": 0.8186275308210372, "全帯域_40ms_log_rms_mae_db": 0.7043625445752312, "全帯域_40ms_rms_envelope_nrmse": 0.059686282441961475, "全帯域_40ms_envelope_correlation": 0.9968901426096423, "全帯域_40ms_modulation_mae_db": 1.4585154135241536, "300–1000 Hz_10ms_log_rms_mae_db": 1.6982863180248091, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14599779404938282, "300–1000 Hz_10ms_envelope_correlation": 0.9851407715334256, "300–1000 Hz_10ms_modulation_mae_db": 1.0862383082878602, "300–1000 Hz_40ms_log_rms_mae_db": 1.0259566817584311, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08596938745072616, "300–1000 Hz_40ms_envelope_correlation": 0.9950975319626525, "300–1000 Hz_40ms_modulation_mae_db": 0.9556851198583648, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5473009037564944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1782173607188396, "1000–3000 Hz_10ms_envelope_correlation": 0.9760221660728267, "1000–3000 Hz_10ms_modulation_mae_db": 0.600716277681674, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8730659440896587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08431849217534582, "1000–3000 Hz_40ms_envelope_correlation": 0.9946542327671578, "1000–3000 Hz_40ms_modulation_mae_db": 0.7180691147278662, "3000–6000 Hz_10ms_log_rms_mae_db": 1.329278030625784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21560217210526136, "3000–6000 Hz_10ms_envelope_correlation": 0.9712343186054413, "3000–6000 Hz_10ms_modulation_mae_db": 1.871907168399649, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7765222055688901, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12600112654071988, "3000–6000 Hz_40ms_envelope_correlation": 0.9899933541572473, "3000–6000 Hz_40ms_modulation_mae_db": 2.0135810532189793, "mr_stft_db": 6.009078250017448, "band_mae_db": 0.3057222486841158, "rms_gain_error_db": -0.16167150418956208, "output_peak": 0.5529387515938328}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4081445482419435, "全帯域_10ms_rms_envelope_nrmse": 0.1424069859070339, "全帯域_10ms_envelope_correlation": 0.9819910895047076, "全帯域_10ms_modulation_mae_db": 1.1136469222786571, "全帯域_40ms_log_rms_mae_db": 0.7131382697777949, "全帯域_40ms_rms_envelope_nrmse": 0.0729830886873985, "全帯域_40ms_envelope_correlation": 0.9951271684858496, "全帯域_40ms_modulation_mae_db": 1.441562666455854, "300–1000 Hz_10ms_log_rms_mae_db": 1.7381362588349658, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.156930621916414, "300–1000 Hz_10ms_envelope_correlation": 0.9826575718859015, "300–1000 Hz_10ms_modulation_mae_db": 0.9323992849833322, "300–1000 Hz_40ms_log_rms_mae_db": 1.0224065134859917, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09687261512627729, "300–1000 Hz_40ms_envelope_correlation": 0.9934625234410095, "300–1000 Hz_40ms_modulation_mae_db": 0.9826424580335447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5304731124278086, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1991556589711796, "1000–3000 Hz_10ms_envelope_correlation": 0.9707016184875941, "1000–3000 Hz_10ms_modulation_mae_db": 0.26718443830820776, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9561611342919071, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1026749832258923, "1000–3000 Hz_40ms_envelope_correlation": 0.991692675852184, "1000–3000 Hz_40ms_modulation_mae_db": 0.2702383607631743, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3820370465050467, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24918404934774863, "3000–6000 Hz_10ms_envelope_correlation": 0.9598698007779384, "3000–6000 Hz_10ms_modulation_mae_db": 2.3065656159052907, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8374619173770523, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11274283888458422, "3000–6000 Hz_40ms_envelope_correlation": 0.9914684460118202, "3000–6000 Hz_40ms_modulation_mae_db": 2.3059304119777484, "mr_stft_db": 6.035175499692173, "band_mae_db": 0.27391464563723456, "rms_gain_error_db": -0.14323856526093062, "output_peak": 0.42587848892582175}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3184565661113823, "全帯域_10ms_rms_envelope_nrmse": 0.12320606060867613, "全帯域_10ms_envelope_correlation": 0.9865597159865367, "全帯域_10ms_modulation_mae_db": 0.45623621750343085, "全帯域_40ms_log_rms_mae_db": 0.6390200478992402, "全帯域_40ms_rms_envelope_nrmse": 0.06390546336101767, "全帯域_40ms_envelope_correlation": 0.9961270056137154, "全帯域_40ms_modulation_mae_db": 0.8358696081385472, "300–1000 Hz_10ms_log_rms_mae_db": 1.6513404435647534, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15321170892120875, "300–1000 Hz_10ms_envelope_correlation": 0.9832699083911278, "300–1000 Hz_10ms_modulation_mae_db": 0.768617811317908, "300–1000 Hz_40ms_log_rms_mae_db": 0.9436773278984191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08446028943001944, "300–1000 Hz_40ms_envelope_correlation": 0.9946174250559526, "300–1000 Hz_40ms_modulation_mae_db": 0.5160149872976078, "1000–3000 Hz_10ms_log_rms_mae_db": 1.583082706471689, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18353329700197393, "1000–3000 Hz_10ms_envelope_correlation": 0.9744784073250925, "1000–3000 Hz_10ms_modulation_mae_db": 0.3736376370899695, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9348222613553521, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10721564296982498, "1000–3000 Hz_40ms_envelope_correlation": 0.9907856112548514, "1000–3000 Hz_40ms_modulation_mae_db": 0.5713640950822034, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3204529554045095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.173372015128577, "3000–6000 Hz_10ms_envelope_correlation": 0.9812589444033457, "3000–6000 Hz_10ms_modulation_mae_db": 1.310374689577564, "3000–6000 Hz_40ms_log_rms_mae_db": 0.722946922599776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10188838125033975, "3000–6000 Hz_40ms_envelope_correlation": 0.9944188191888602, "3000–6000 Hz_40ms_modulation_mae_db": 1.266792138096788, "mr_stft_db": 6.012967539553741, "band_mae_db": 0.22932650298321997, "rms_gain_error_db": -0.026919973750930665, "output_peak": 0.42712035272682475}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60480, "output_samples": 60480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.991841821219805, "spectral_convergence": 0.4478331442794471, "spectral_convergence_status": "ok", "amplitude_floor": 9.224568988143785e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.877808727141924, "spectral_convergence": 0.4807916683326631, "spectral_convergence_status": "ok", "amplitude_floor": 8.440100575569802e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.132041775773015, "spectral_convergence": 0.4826295643837019, "spectral_convergence_status": "ok", "amplitude_floor": 6.303303867538867e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.6236728988818796e-05, "output_power": 5.188495588928688e-05, "delta_db": 0.5005434945431158, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005640699493062965, "output_power": 0.0005411780116452365, "delta_db": -0.17992820740636478, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007824000374604723, "output_power": 0.0007196034563865775, "delta_db": -0.36335621899010656, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028236131111099825, "output_power": 0.0002539544624478327, "delta_db": -0.46049341236916275, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018611455363091946, "output_power": 0.00018375042598162977, "delta_db": -0.055519804911884706, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00035218826443925665, "output_power": 0.00034451270957547146, "delta_db": -0.09569632086474722, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2759229098475637, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.238942006870536, "pitch": null, "output_rms_dbfs": -26.788400156722858, "output_peak": 0.35210558716312174, "reference_rms_dbfs": -26.54945814985232, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.000564108044915}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00014780031156712295, "log_rms_mae_db": 1.3220531763801835, "rms_envelope_nrmse": 0.1285262192109336, "envelope_correlation": 0.9856131276195729, "rms_gain_error_db": -0.23898105914463358, "reference_mean_power": 0.0022162793576100603, "output_mean_power": 0.0020976179731242574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 1.2692207138635437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044041649987677594, "output_power": 0.00040991266375086765, "reference_active": true, "delta_db": -0.31172245603322923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888390879465172, "output_power": 9.018850496230026e-05, "reference_active": true, "delta_db": -1.199718880898887}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.435323106110239e-05, "output_power": 2.5045790577548384e-05, "reference_active": true, "delta_db": 0.12178155546064424}], "active_band_count": 3, "active_band_mae_db": 0.5444076307975868}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00013486753957748745, "log_rms_mae_db": 0.6603248214599696, "rms_envelope_nrmse": 0.06698716460460559, "envelope_correlation": 0.9963698727266226, "rms_gain_error_db": -0.23909623188763493, "reference_mean_power": 0.0022339157834068563, "output_mean_power": 0.0021142540607453934}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.176190391785113e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004109863651179866, "output_power": 0.0003807832143366839, "reference_active": true, "delta_db": -0.33149618372548295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3981406650238314e-05, "output_power": 3.983497213296035e-05, "reference_active": true, "delta_db": -1.3197967954993133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0096166790138946e-06, "output_power": 1.2406160488918836e-06, "reference_active": true, "delta_db": 0.8948087843053852}], "active_band_count": 3, "active_band_mae_db": 0.8487005878433939}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00012936155891805085, "log_rms_mae_db": 1.6773443304996811, "rms_envelope_nrmse": 0.16149831707623682, "envelope_correlation": 0.981684720544353, "rms_gain_error_db": -0.308950446501539, "reference_mean_power": 0.0012072233727869219, "output_mean_power": 0.0011243268715489033}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 8.078028680738071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004104434555454037, "output_power": 0.00037543181848785113, "reference_active": true, "delta_db": -0.3872225825755099}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.258149083667088e-05, "output_power": 7.557293790822766e-05, "reference_active": true, "delta_db": -0.8815786412977676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4118213823838e-05, "output_power": 1.533426624845882e-05, "reference_active": true, "delta_db": 0.3588324452252098}], "active_band_count": 3, "active_band_mae_db": 0.5425445563661625}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00011150674016193665, "log_rms_mae_db": 1.0009786852631837, "rms_envelope_nrmse": 0.0897775033541427, "envelope_correlation": 0.9946689771285516, "rms_gain_error_db": -0.30895197755108755, "reference_mean_power": 0.0012168889554241998, "output_mean_power": 0.0011333283473309724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.50453906071102e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038524435492848886, "output_power": 0.00035322079291391607, "reference_active": true, "delta_db": -0.3769002251718242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.764413089442773e-05, "output_power": 3.713925341546657e-05, "reference_active": true, "delta_db": -1.0817623995427663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0419619324101079e-06, "output_power": 1.0274172312011102e-06, "reference_active": true, "delta_db": -0.06105007342765324}], "active_band_count": 3, "active_band_mae_db": 0.506570899380748}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 7.756656945350889e-05, "log_rms_mae_db": 1.524278418715901, "rms_envelope_nrmse": 0.16656552807583477, "envelope_correlation": 0.9800804198849119, "rms_gain_error_db": -0.4493556929606774, "reference_mean_power": 0.0003662329710023159, "output_mean_power": 0.00033023406525548027}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 2.3479287649550626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015936395121522886, "output_power": 0.00013523055325225856, "reference_active": true, "delta_db": -0.7131526414777111}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.480344306075996e-05, "output_power": 1.8703424164515306e-05, "reference_active": true, "delta_db": -1.2259084799461886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.158977700048883e-06, "output_power": 6.807600270469022e-06, "reference_active": true, "delta_db": -0.218569627416281}], "active_band_count": 3, "active_band_mae_db": 0.7192102496133935}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 6.756499548428438e-05, "log_rms_mae_db": 0.9479436960424329, "rms_envelope_nrmse": 0.0985791320558167, "envelope_correlation": 0.9939753310066319, "rms_gain_error_db": -0.44937900935683583, "reference_mean_power": 0.00036915132536056565, "output_mean_power": 0.0003328637726029548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 2.14579734589437e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015227225639161801, "output_power": 0.00012926362767691518, "reference_active": true, "delta_db": -0.7114444334696901}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5037342052746137e-05, "output_power": 1.159105812895044e-05, "reference_active": true, "delta_db": -1.130479948724633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0212382710455496e-07, "output_power": 3.3470901621717876e-07, "reference_active": true, "delta_db": -0.7969239586866719}], "active_band_count": 3, "active_band_mae_db": 0.8796161136269983}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00011110909618382604, "log_rms_mae_db": 1.367298140609427, "rms_envelope_nrmse": 0.19996952229011306, "envelope_correlation": 0.9751295346014964, "rms_gain_error_db": 0.18769895719780416, "reference_mean_power": 0.00024441441718797436, "output_mean_power": 0.0002552094288697365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 3.469643399885664e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011046778972231385, "output_power": 0.00011467738436340397, "reference_active": true, "delta_db": 0.16242114134104257}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.9627372395634935e-05, "output_power": 5.600794502098609e-05, "reference_active": true, "delta_db": -0.2719603334833915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.408768678752184e-06, "output_power": 7.760246715380707e-06, "reference_active": true, "delta_db": 2.4555821599145773}], "active_band_count": 3, "active_band_mae_db": 0.9633212115796704}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.166729119583493e-05, "log_rms_mae_db": 0.8430996651420843, "rms_envelope_nrmse": 0.1121093624064054, "envelope_correlation": 0.992079963683742, "rms_gain_error_db": 0.18771756458801248, "reference_mean_power": 0.00024636420339127556, "output_mean_power": 0.0002572464331346601}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.3341266023063243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010489106418058692, "output_power": 0.00010985782150269443, "reference_active": true, "delta_db": 0.20092491042233637}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0862352851100365e-05, "output_power": 3.08754444457613e-05, "reference_active": true, "delta_db": 0.001841856254908978}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.747611403563906e-07, "output_power": 4.4280932981871386e-07, "reference_active": true, "delta_db": 0.7246221092774674}], "active_band_count": 3, "active_band_mae_db": 0.30912962531823757}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6743799325276517, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "80c0153be39a89eb374ec0f9d52feeff3b7e9d5b86f1f44fa33ebd19d96a15cf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9735821944016705, "standard_deviation": 0.004755498173977415, "minimum": 0.965521406755027, "maximum": 0.9800804198849119, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5678748089558123, "standard_deviation": 0.04813479196141593, "minimum": 1.524278418715901, "maximum": 1.6642393192345197, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5471465338014097, "standard_deviation": 0.2356826839544566, "minimum": 0.227487839929678, "maximum": 0.8372284814509444, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1873131999714545, "standard_deviation": 0.016300926960120306, "minimum": 0.16656552807583477, "maximum": 0.21297116741527086, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919239602120918, "standard_deviation": 0.001963693048902379, "minimum": 0.9889644386699726, "maximum": 0.9946542327671578, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9438143073203353, "standard_deviation": 0.04620674878833402, "minimum": 0.8730659440896587, "maximum": 0.9958003746178109, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5649430667889888, "standard_deviation": 0.21835271103560955, "minimum": 0.2415772842788507, "maximum": 0.8796161136269983, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10176390039431146, "standard_deviation": 0.010723293850109797, "minimum": 0.08431849217534582, "maximum": 0.11747437951752573, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9770874862301016, "standard_deviation": 0.008179914043255137, "minimum": 0.9598698007779384, "maximum": 0.9843685150506161, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.334219583184132, "standard_deviation": 0.02730751728562624, "minimum": 1.3082495948336572, "maximum": 1.3820370465050467, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.38298813439844, "standard_deviation": 0.5096041105102086, "minimum": 0.7796582059698576, "maximum": 2.3065656159052907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1900298304416197, "standard_deviation": 0.030302944814447084, "minimum": 0.1554644786713932, "maximum": 0.24918404934774863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927787987847869, "standard_deviation": 0.0020660690410750577, "minimum": 0.9899933541572473, "maximum": 0.995454757430214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8007913339267887, "standard_deviation": 0.044130105501245846, "minimum": 0.722946922599776, "maximum": 0.8470383015317647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3084638662856278, "standard_deviation": 0.699115147074909, "minimum": 0.30912962531823757, "maximum": 2.3059304119777484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10875836932931529, "standard_deviation": 0.016639334103360246, "minimum": 0.0873720358522454, "maximum": 0.1347896439860812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9817807186115703, "standard_deviation": 0.002322442686797347, "minimum": 0.9776103874987867, "maximum": 0.9851407715334256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6889118729327228, "standard_deviation": 0.037759077956850906, "minimum": 1.6235479924447402, "maximum": 1.7381362588349658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9044652880787202, "standard_deviation": 0.2740700728434565, "minimum": 0.5377177509014056, "maximum": 1.329241356425655, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16003151879188546, "standard_deviation": 0.009717990431198255, "minimum": 0.14599779404938282, "maximum": 0.17701732515992322, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936201046767503, "standard_deviation": 0.00139777183560013, "minimum": 0.9908049640422373, "maximum": 0.9950975319626525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9884463887253041, "standard_deviation": 0.043381785229348456, "minimum": 0.9137984816394005, "maximum": 1.037607624744725, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7465352702129111, "standard_deviation": 0.30450909437353135, "minimum": 0.2585966392149837, "maximum": 1.0939921453707377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09348074519872335, "standard_deviation": 0.008775962771077072, "minimum": 0.08446028943001944, "maximum": 0.11103405801112538, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24646463315786693, "standard_deviation": 0.04221447850263877, "minimum": 0.17418781618607307, "maximum": 0.3057222486841158, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.012628341531261, "standard_deviation": 0.011638768505419488, "minimum": 6.000564108044915, "maximum": 6.035175499692173, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4395622557883348, "standard_deviation": 0.06447841641428965, "minimum": 0.35210558716312174, "maximum": 0.5529387515938328, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09449922760621486, "standard_deviation": 0.09444602821190423, "minimum": -0.238942006870536, "maximum": 0.06268772996444733, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9852547959507435, "standard_deviation": 0.0018026051530434524, "minimum": 0.9819910895047076, "maximum": 0.9876405169741275, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3442103912634615, "standard_deviation": 0.046487429389174155, "minimum": 1.276478926726376, "maximum": 1.4081445482419435, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6846478209626461, "standard_deviation": 0.2642238818608129, "minimum": 0.36895618423292537, "maximum": 1.1136469222786571, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12906662029024837, "standard_deviation": 0.007542194653156193, "minimum": 0.11855873977099346, "maximum": 0.1424069859070339, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9954245853622303, "standard_deviation": 0.001086774056321083, "minimum": 0.9939117564467268, "maximum": 0.9968901426096423, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7051580782441389, "standard_deviation": 0.03824437877643001, "minimum": 0.6390200478992402, "maximum": 0.7490015503714696, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9881888684925836, "standard_deviation": 0.36441613089502983, "minimum": 0.5428571226714008, "maximum": 1.4585154135241536, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07038899166554732, "standard_deviation": 0.007487429070244413, "minimum": 0.059686282441961475, "maximum": 0.0805252187495414, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0019", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1741349697113037, "predicted_mos": 2.1741349697113037, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "bd2a06755eff59fff0f71e8595bea7184eefe759b75b59082a7dff6bb3325284", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2553261001594365, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3089922240180534, "全帯域_10ms_rms_envelope_nrmse": 0.12714270461702182, "全帯域_10ms_envelope_correlation": 0.9861010617025493, "全帯域_10ms_modulation_mae_db": 0.5380642503464353, "全帯域_40ms_log_rms_mae_db": 0.6508725037951991, "全帯域_40ms_rms_envelope_nrmse": 0.06758487895157685, "全帯域_40ms_envelope_correlation": 0.9965264240793553, "全帯域_40ms_modulation_mae_db": 0.9648043742857784, "300–1000 Hz_10ms_log_rms_mae_db": 1.6812683871814051, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1589065422231595, "300–1000 Hz_10ms_envelope_correlation": 0.9822819554174298, "300–1000 Hz_10ms_modulation_mae_db": 0.5464295926055643, "300–1000 Hz_40ms_log_rms_mae_db": 1.0004090223965276, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08736335744078602, "300–1000 Hz_40ms_envelope_correlation": 0.994975514850087, "300–1000 Hz_40ms_modulation_mae_db": 0.5701787262999912, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5127417927673414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1736086348547424, "1000–3000 Hz_10ms_envelope_correlation": 0.9790174036853657, "1000–3000 Hz_10ms_modulation_mae_db": 0.8972970952428058, "1000–3000 Hz_40ms_log_rms_mae_db": 0.898423822334382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10731495177779851, "1000–3000 Hz_40ms_envelope_correlation": 0.9935721432832607, "1000–3000 Hz_40ms_modulation_mae_db": 1.0406886484413562, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3571753583093824, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1893533570576756, "3000–6000 Hz_10ms_envelope_correlation": 0.9774454856157405, "3000–6000 Hz_10ms_modulation_mae_db": 1.0036415134862375, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7857269287490624, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10080301306010331, "3000–6000 Hz_40ms_envelope_correlation": 0.993447614160723, "3000–6000 Hz_40ms_modulation_mae_db": 0.3564544081252348, "mr_stft_db": 5.98817298956833, "band_mae_db": 0.26228144241717505, "rms_gain_error_db": -0.276295226866844, "output_peak": 0.34975913442966233}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.379859919839073, "全帯域_10ms_rms_envelope_nrmse": 0.1360793490876426, "全帯域_10ms_envelope_correlation": 0.9835963825588977, "全帯域_10ms_modulation_mae_db": 0.7168927106243191, "全帯域_40ms_log_rms_mae_db": 0.7352674688219433, "全帯域_40ms_rms_envelope_nrmse": 0.07717056735234445, "全帯域_40ms_envelope_correlation": 0.9944683253438755, "全帯域_40ms_modulation_mae_db": 0.5591197013374573, "300–1000 Hz_10ms_log_rms_mae_db": 1.678449052983742, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1753287089687993, "300–1000 Hz_10ms_envelope_correlation": 0.9780362565091791, "300–1000 Hz_10ms_modulation_mae_db": 0.8298220479153681, "300–1000 Hz_40ms_log_rms_mae_db": 0.9356273164843832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10593009306888389, "300–1000 Hz_40ms_envelope_correlation": 0.9916079485206541, "300–1000 Hz_40ms_modulation_mae_db": 0.4897340981050892, "1000–3000 Hz_10ms_log_rms_mae_db": 1.663066722106331, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2120834752754132, "1000–3000 Hz_10ms_envelope_correlation": 0.965791923759896, "1000–3000 Hz_10ms_modulation_mae_db": 0.9452238295644518, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9701265777557736, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11563875485575996, "1000–3000 Hz_40ms_envelope_correlation": 0.9893815607218617, "1000–3000 Hz_40ms_modulation_mae_db": 0.8505871247054295, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3363222199626874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15788693410752844, "3000–6000 Hz_10ms_envelope_correlation": 0.9838865165562546, "3000–6000 Hz_10ms_modulation_mae_db": 1.2253767424468942, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7745898465050806, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08231500004537809, "3000–6000 Hz_40ms_envelope_correlation": 0.9956073406316747, "3000–6000 Hz_40ms_modulation_mae_db": 0.987605480173043, "mr_stft_db": 6.003826587729073, "band_mae_db": 0.13744605010620864, "rms_gain_error_db": -0.11800207872668424, "output_peak": 0.3594586875418457}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4078183916263234, "全帯域_10ms_rms_envelope_nrmse": 0.12871471401230392, "全帯域_10ms_envelope_correlation": 0.9852911304310148, "全帯域_10ms_modulation_mae_db": 0.5088394528543407, "全帯域_40ms_log_rms_mae_db": 0.7632929071524757, "全帯域_40ms_rms_envelope_nrmse": 0.07954933509600244, "全帯域_40ms_envelope_correlation": 0.994026868590552, "全帯域_40ms_modulation_mae_db": 0.8637430341322871, "300–1000 Hz_10ms_log_rms_mae_db": 1.626046445567632, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17063375195251165, "300–1000 Hz_10ms_envelope_correlation": 0.9794621203469758, "300–1000 Hz_10ms_modulation_mae_db": 1.456042721143887, "300–1000 Hz_40ms_log_rms_mae_db": 0.9254235641196914, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10160824073794059, "300–1000 Hz_40ms_envelope_correlation": 0.9922541821072148, "300–1000 Hz_40ms_modulation_mae_db": 1.2407631701262993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6076396643294573, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17798817698820343, "1000–3000 Hz_10ms_envelope_correlation": 0.9760080609508952, "1000–3000 Hz_10ms_modulation_mae_db": 0.8186490739415881, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9410855104416133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08775030345990133, "1000–3000 Hz_40ms_envelope_correlation": 0.99388650308939, "1000–3000 Hz_40ms_modulation_mae_db": 0.637661392155603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3268409330612143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16617972211502083, "3000–6000 Hz_10ms_envelope_correlation": 0.9838750816554185, "3000–6000 Hz_10ms_modulation_mae_db": 1.0008601341635985, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7305279871549081, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12191037521774822, "3000–6000 Hz_40ms_envelope_correlation": 0.9924155796288275, "3000–6000 Hz_40ms_modulation_mae_db": 1.0078516709882426, "mr_stft_db": 5.986184284763056, "band_mae_db": 0.17010555891362847, "rms_gain_error_db": -0.07991560490861474, "output_peak": 0.4632053295211114}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3220933204735033, "全帯域_10ms_rms_envelope_nrmse": 0.12785603512387916, "全帯域_10ms_envelope_correlation": 0.9858454622411165, "全帯域_10ms_modulation_mae_db": 0.45737337815853857, "全帯域_40ms_log_rms_mae_db": 0.7057382632888085, "全帯域_40ms_rms_envelope_nrmse": 0.06794898445966818, "全帯域_40ms_envelope_correlation": 0.9958039541312078, "全帯域_40ms_modulation_mae_db": 0.8509370218389529, "300–1000 Hz_10ms_log_rms_mae_db": 1.6610570565997311, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1548996262582897, "300–1000 Hz_10ms_envelope_correlation": 0.9829599495768538, "300–1000 Hz_10ms_modulation_mae_db": 0.5236436094944826, "300–1000 Hz_40ms_log_rms_mae_db": 0.8364655408152352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08984195496679942, "300–1000 Hz_40ms_envelope_correlation": 0.9939135663725097, "300–1000 Hz_40ms_modulation_mae_db": 0.3667851477947656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5357136066033354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.198041800355661, "1000–3000 Hz_10ms_envelope_correlation": 0.9709052405540095, "1000–3000 Hz_10ms_modulation_mae_db": 0.5114673538357924, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9168725110773852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10477725427101574, "1000–3000 Hz_40ms_envelope_correlation": 0.9912715558026113, "1000–3000 Hz_40ms_modulation_mae_db": 0.49152301754124733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.295388670822073, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17719976153235584, "3000–6000 Hz_10ms_envelope_correlation": 0.9811051938211623, "3000–6000 Hz_10ms_modulation_mae_db": 1.1635344060329154, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7382174595983511, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0921966693138425, "3000–6000 Hz_40ms_envelope_correlation": 0.9951147183922187, "3000–6000 Hz_40ms_modulation_mae_db": 1.0844117061776501, "mr_stft_db": 5.994789911584064, "band_mae_db": 0.1754277000458487, "rms_gain_error_db": 0.09282976982189507, "output_peak": 0.432795321189376}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4329491587776892, "全帯域_10ms_rms_envelope_nrmse": 0.13131584257558426, "全帯域_10ms_envelope_correlation": 0.9847162705265954, "全帯域_10ms_modulation_mae_db": 0.9672985576075782, "全帯域_40ms_log_rms_mae_db": 0.7613866919635195, "全帯域_40ms_rms_envelope_nrmse": 0.07336113230551998, "全帯域_40ms_envelope_correlation": 0.9948945563196643, "全帯域_40ms_modulation_mae_db": 1.3158467679705208, "300–1000 Hz_10ms_log_rms_mae_db": 1.6788901254080055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16223452825004242, "300–1000 Hz_10ms_envelope_correlation": 0.981215601929654, "300–1000 Hz_10ms_modulation_mae_db": 0.9031385532588289, "300–1000 Hz_40ms_log_rms_mae_db": 0.9295951184010457, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09186367166838422, "300–1000 Hz_40ms_envelope_correlation": 0.9936357908310857, "300–1000 Hz_40ms_modulation_mae_db": 0.853332198265263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.553312911704916, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1821100745794849, "1000–3000 Hz_10ms_envelope_correlation": 0.9748683341774811, "1000–3000 Hz_10ms_modulation_mae_db": 0.22700906648982497, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8760522262039555, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11497531733891443, "1000–3000 Hz_40ms_envelope_correlation": 0.9893901780827286, "1000–3000 Hz_40ms_modulation_mae_db": 0.296117099199082, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2844951290433844, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17652762308864564, "3000–6000 Hz_10ms_envelope_correlation": 0.9821554941890135, "3000–6000 Hz_10ms_modulation_mae_db": 1.6451318098567478, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7980526777603415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11080871834235416, "3000–6000 Hz_40ms_envelope_correlation": 0.9934104992918383, "3000–6000 Hz_40ms_modulation_mae_db": 1.8340401663050283, "mr_stft_db": 5.984303743365854, "band_mae_db": 0.1628542497502539, "rms_gain_error_db": -0.025579295698033317, "output_peak": 0.5014623729697194}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2911948544696505, "全帯域_10ms_rms_envelope_nrmse": 0.12269314656562069, "全帯域_10ms_envelope_correlation": 0.986728318352004, "全帯域_10ms_modulation_mae_db": 0.855708422869319, "全帯域_40ms_log_rms_mae_db": 0.7245848601936993, "全帯域_40ms_rms_envelope_nrmse": 0.06257097382026497, "全帯域_40ms_envelope_correlation": 0.9965210920034556, "全帯域_40ms_modulation_mae_db": 1.512750237963519, "300–1000 Hz_10ms_log_rms_mae_db": 1.6656355400453013, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15052538655124112, "300–1000 Hz_10ms_envelope_correlation": 0.9840278310008771, "300–1000 Hz_10ms_modulation_mae_db": 0.8459679649981965, "300–1000 Hz_40ms_log_rms_mae_db": 0.9638719962049739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08961016979536036, "300–1000 Hz_40ms_envelope_correlation": 0.9944090764138285, "300–1000 Hz_40ms_modulation_mae_db": 0.7444139518472274, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5648277679849594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18106720977128563, "1000–3000 Hz_10ms_envelope_correlation": 0.9752439632201556, "1000–3000 Hz_10ms_modulation_mae_db": 0.6994279658120403, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8524645109535948, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08743423843961956, "1000–3000 Hz_40ms_envelope_correlation": 0.9942158130160691, "1000–3000 Hz_40ms_modulation_mae_db": 0.7992213525038524, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3076546796327322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20639185293645676, "3000–6000 Hz_10ms_envelope_correlation": 0.9735498302152576, "3000–6000 Hz_10ms_modulation_mae_db": 1.6516564213968665, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7367120946614617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12127106417938689, "3000–6000 Hz_40ms_envelope_correlation": 0.9906717222220786, "3000–6000 Hz_40ms_modulation_mae_db": 1.7873116151992512, "mr_stft_db": 5.989232577878634, "band_mae_db": 0.23471838042720425, "rms_gain_error_db": -0.15341883005835658, "output_peak": 0.5579537498005938}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4290466970092803, "全帯域_10ms_rms_envelope_nrmse": 0.14578533896805454, "全帯域_10ms_envelope_correlation": 0.9811159488742531, "全帯域_10ms_modulation_mae_db": 1.1864488087197607, "全帯域_40ms_log_rms_mae_db": 0.7315123113886804, "全帯域_40ms_rms_envelope_nrmse": 0.07338711111155909, "全帯域_40ms_envelope_correlation": 0.9950183982205953, "全帯域_40ms_modulation_mae_db": 1.5667378242283458, "300–1000 Hz_10ms_log_rms_mae_db": 1.68244245411025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15970662080102058, "300–1000 Hz_10ms_envelope_correlation": 0.9818975936131511, "300–1000 Hz_10ms_modulation_mae_db": 1.0110031659293066, "300–1000 Hz_40ms_log_rms_mae_db": 0.9143061003583702, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09282760277070329, "300–1000 Hz_40ms_envelope_correlation": 0.993840685149688, "300–1000 Hz_40ms_modulation_mae_db": 1.1165083415010315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5355382520801277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19549367386772, "1000–3000 Hz_10ms_envelope_correlation": 0.9717159420720654, "1000–3000 Hz_10ms_modulation_mae_db": 0.19590403826297298, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9220763306517115, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10580632354860466, "1000–3000 Hz_40ms_envelope_correlation": 0.9911417694537732, "1000–3000 Hz_40ms_modulation_mae_db": 0.13253163471770182, "3000–6000 Hz_10ms_log_rms_mae_db": 1.377663651651843, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2580617009728121, "3000–6000 Hz_10ms_envelope_correlation": 0.9569667501932946, "3000–6000 Hz_10ms_modulation_mae_db": 2.469483854992308, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7835448354722703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1132458421740372, "3000–6000 Hz_40ms_envelope_correlation": 0.9913961725036387, "3000–6000 Hz_40ms_modulation_mae_db": 2.5329816198118, "mr_stft_db": 6.017370800296586, "band_mae_db": 0.20501593604206111, "rms_gain_error_db": -0.12321995468619348, "output_peak": 0.4201612228922957}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3716289963398303, "全帯域_10ms_rms_envelope_nrmse": 0.12453234732598516, "全帯域_10ms_envelope_correlation": 0.9862547081076413, "全帯域_10ms_modulation_mae_db": 0.40536663947828805, "全帯域_40ms_log_rms_mae_db": 0.657171700536096, "全帯域_40ms_rms_envelope_nrmse": 0.06349701667525522, "全帯域_40ms_envelope_correlation": 0.9961765252124776, "全帯域_40ms_modulation_mae_db": 0.7756523546886843, "300–1000 Hz_10ms_log_rms_mae_db": 1.610856916267603, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15251809722972415, "300–1000 Hz_10ms_envelope_correlation": 0.9834521878878415, "300–1000 Hz_10ms_modulation_mae_db": 0.5721836840427331, "300–1000 Hz_40ms_log_rms_mae_db": 0.8867277290517863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0821832529769957, "300–1000 Hz_40ms_envelope_correlation": 0.9948992405844466, "300–1000 Hz_40ms_modulation_mae_db": 0.35058642594716066, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5948792600359478, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18999284133068453, "1000–3000 Hz_10ms_envelope_correlation": 0.9726316627991642, "1000–3000 Hz_10ms_modulation_mae_db": 0.5324203876119065, "1000–3000 Hz_40ms_log_rms_mae_db": 0.88566258411347, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11230404988910787, "1000–3000 Hz_40ms_envelope_correlation": 0.9898784504662178, "1000–3000 Hz_40ms_modulation_mae_db": 0.7077989427593349, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3289364937428911, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17350794587373286, "3000–6000 Hz_10ms_envelope_correlation": 0.981015761358655, "3000–6000 Hz_10ms_modulation_mae_db": 1.3854078780864316, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7350904129665182, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0964618741882852, "3000–6000 Hz_40ms_envelope_correlation": 0.9948701066544618, "3000–6000 Hz_40ms_modulation_mae_db": 1.3900546468307784, "mr_stft_db": 5.988736920263583, "band_mae_db": 0.23471228319059403, "rms_gain_error_db": -0.02327611862156098, "output_peak": 0.42630922469263666}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60480, "output_samples": 60480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.988421229804062, "spectral_convergence": 0.4433869900692745, "spectral_convergence_status": "ok", "amplitude_floor": 9.224568988143785e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.871432289352447, "spectral_convergence": 0.47908839762670113, "spectral_convergence_status": "ok", "amplitude_floor": 8.440100575569802e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.104665449548482, "spectral_convergence": 0.48109159891753805, "spectral_convergence_status": "ok", "amplitude_floor": 6.303303867538867e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.6236728988818796e-05, "output_power": 4.8526765521317536e-05, "delta_db": 0.20994242639029964, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005640699493062965, "output_power": 0.0005424968228547062, "delta_db": -0.169357643331836, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007824000374604723, "output_power": 0.0007137394762154391, "delta_db": -0.3988914475913466, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028236131111099825, "output_power": 0.00024857890551410505, "delta_db": -0.5534091832302629, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018611455363091946, "output_power": 0.00018089288207008463, "delta_db": -0.12358856856665644, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00035218826443925665, "output_power": 0.00034270855186444027, "delta_db": -0.11849938539264858, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.26228144241717505, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.276295226866844, "pitch": null, "output_rms_dbfs": -26.825753376719167, "output_peak": 0.34975913442966233, "reference_rms_dbfs": -26.54945814985232, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.98817298956833}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00014780031156712295, "log_rms_mae_db": 1.3089922240180534, "rms_envelope_nrmse": 0.12714270461702182, "envelope_correlation": 0.9861010617025493, "rms_gain_error_db": -0.2763470230121143, "reference_mean_power": 0.0022162793576100603, "output_mean_power": 0.002079647839143719}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 1.2692207138635437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044041649987677594, "output_power": 0.0004091067708329389, "reference_active": true, "delta_db": -0.3202691386988144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888390879465172, "output_power": 9.60593222847612e-05, "reference_active": true, "delta_db": -0.9258355756770625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.435323106110239e-05, "output_power": 2.6507293877652277e-05, "reference_active": true, "delta_db": 0.368088036663429}], "active_band_count": 3, "active_band_mae_db": 0.5380642503464353}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00013486753957748745, "log_rms_mae_db": 0.6508725037951991, "rms_envelope_nrmse": 0.06758487895157685, "envelope_correlation": 0.9965264240793553, "rms_gain_error_db": -0.2764716362199397, "reference_mean_power": 0.0022339157834068563, "output_mean_power": 0.0020961368501878816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.176190391785113e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004109863651179866, "output_power": 0.00038257300785616037, "reference_active": true, "delta_db": -0.31113088503495323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3981406650238314e-05, "output_power": 4.328756072418246e-05, "reference_active": true, "delta_db": -0.9588108344545296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0096166790138946e-06, "output_power": 1.46758632691328e-06, "reference_active": true, "delta_db": 1.624471403367852}], "active_band_count": 3, "active_band_mae_db": 0.9648043742857784}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00012936155891805085, "log_rms_mae_db": 1.6812683871814051, "rms_envelope_nrmse": 0.1589065422231595, "envelope_correlation": 0.9822819554174298, "rms_gain_error_db": -0.30425014440983517, "reference_mean_power": 0.0012072233727869219, "output_mean_power": 0.0011255443718764394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 8.078028680738071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004104434555454037, "output_power": 0.00038435388924504027, "reference_active": true, "delta_db": -0.2852205528455101}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.258149083667088e-05, "output_power": 8.014122794208204e-05, "reference_active": true, "delta_db": -0.6266817742643921}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4118213823838e-05, "output_power": 1.6692385279678748e-05, "reference_active": true, "delta_db": 0.7273864507067908}], "active_band_count": 3, "active_band_mae_db": 0.5464295926055643}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00011150674016193665, "log_rms_mae_db": 1.0004090223965276, "rms_envelope_nrmse": 0.08736335744078602, "envelope_correlation": 0.994975514850087, "rms_gain_error_db": -0.30425087310492127, "reference_mean_power": 0.0012168889554241998, "output_mean_power": 0.0011345558046984624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.50453906071102e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038524435492848886, "output_power": 0.0003627310002640656, "reference_active": true, "delta_db": -0.2615161012622753}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.764413089442773e-05, "output_power": 3.993845539210023e-05, "reference_active": true, "delta_db": -0.7661814447704972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0419619324101079e-06, "output_power": 1.2193705892468712e-06, "reference_active": true, "delta_db": 0.682838632867201}], "active_band_count": 3, "active_band_mae_db": 0.5701787262999912}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 7.756656945350889e-05, "log_rms_mae_db": 1.5127417927673414, "rms_envelope_nrmse": 0.1736086348547424, "envelope_correlation": 0.9790174036853657, "rms_gain_error_db": -0.5623996953870127, "reference_mean_power": 0.0003662329710023159, "output_mean_power": 0.00032174919618167266}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 2.3479287649550626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015936395121522886, "output_power": 0.00012904993965196686, "reference_active": true, "delta_db": -0.9163228317233528}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.480344306075996e-05, "output_power": 1.832377911445463e-05, "reference_active": true, "delta_db": -1.3149692319227426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.158977700048883e-06, "output_power": 6.438594694696834e-06, "reference_active": true, "delta_db": -0.4605992220823221}], "active_band_count": 3, "active_band_mae_db": 0.8972970952428058}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 6.756499548428438e-05, "log_rms_mae_db": 0.898423822334382, "rms_envelope_nrmse": 0.10731495177779851, "envelope_correlation": 0.9935721432832607, "rms_gain_error_db": -0.5624083544406727, "reference_mean_power": 0.00036915132536056565, "output_mean_power": 0.00032431243168789506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 2.14579734589437e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015227225639161801, "output_power": 0.0001232880303756209, "reference_active": true, "delta_db": -0.916998687833433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5037342052746137e-05, "output_power": 1.1287098616403013e-05, "reference_active": true, "delta_db": -1.2458875915519005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0212382710455496e-07, "output_power": 3.224347495480494e-07, "reference_active": true, "delta_db": -0.9591796659387353}], "active_band_count": 3, "active_band_mae_db": 1.0406886484413562}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00011110909618382604, "log_rms_mae_db": 1.3571753583093824, "rms_envelope_nrmse": 0.1893533570576756, "envelope_correlation": 0.9774454856157405, "rms_gain_error_db": 0.12861131667209932, "reference_mean_power": 0.00024441441718797436, "output_mean_power": 0.00025176070819891}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 3.469643399885664e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011046778972231385, "output_power": 0.00011156167555532697, "reference_active": true, "delta_db": 0.04279363771339689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.9627372395634935e-05, "output_power": 5.5041162088160375e-05, "reference_active": true, "delta_db": -0.34758077060410564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.408768678752184e-06, "output_power": 8.060691978747403e-06, "reference_active": true, "delta_db": 2.62055013214121}], "active_band_count": 3, "active_band_mae_db": 1.0036415134862375}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.166729119583493e-05, "log_rms_mae_db": 0.7857269287490624, "rms_envelope_nrmse": 0.10080301306010331, "envelope_correlation": 0.993447614160723, "rms_gain_error_db": 0.1286337914918639, "reference_mean_power": 0.00024636420339127556, "output_mean_power": 0.0002537704118059429}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.3341266023063243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010489106418058692, "output_power": 0.00010674991613660403, "reference_active": true, "delta_db": 0.07629050909878299}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0862352851100365e-05, "output_power": 3.069577800477141e-05, "reference_active": true, "delta_db": -0.023503869318627475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.747611403563906e-07, "output_power": 4.6850198617773266e-07, "reference_active": true, "delta_db": 0.9695688459582941}], "active_band_count": 3, "active_band_mae_db": 0.3564544081252348}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6672641401032823, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0bff03ec440e0e4446689bdb140cc70d3d6b3616fb14fa80681d5139dfd9368c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9732728164023791, "standard_deviation": 0.003987785785359051, "minimum": 0.965791923759896, "maximum": 0.9790174036853657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.570964997201552, "standard_deviation": 0.04875092519960539, "minimum": 1.5127417927673414, "maximum": 1.663066722106331, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6034248513451728, "standard_deviation": 0.2878455371068766, "minimum": 0.19590403826297298, "maximum": 0.9452238295644518, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1887982358778994, "standard_deviation": 0.012682853716582215, "minimum": 0.1736086348547424, "maximum": 0.2120834752754132, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9915922467394891, "standard_deviation": 0.0020370080193596184, "minimum": 0.9893815607218617, "maximum": 0.9942158130160691, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9078455091914857, "standard_deviation": 0.037657384751579606, "minimum": 0.8524645109535948, "maximum": 0.9701265777557736, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6195161515029509, "standard_deviation": 0.29983810775915837, "minimum": 0.13253163471770182, "maximum": 1.0406886484413562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10450014919759025, "standard_deviation": 0.011188049645289268, "minimum": 0.08743423843961956, "maximum": 0.11563875485575996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9775000142005996, "standard_deviation": 0.008989533971466103, "minimum": 0.9569667501932946, "maximum": 0.9838865165562546, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.326809642028276, "standard_deviation": 0.03104818527346014, "minimum": 1.2844951290433844, "maximum": 1.377663651651843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.44313659505775, "standard_deviation": 0.4862928305954442, "minimum": 1.0008601341635985, "maximum": 2.469483854992308, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18813861221052852, "standard_deviation": 0.03182492631880406, "minimum": 0.15788693410752844, "maximum": 0.2580617009728121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933667191856826, "standard_deviation": 0.0017886646888039298, "minimum": 0.9906717222220786, "maximum": 0.9956073406316747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7603077803584992, "standard_deviation": 0.027728888712942476, "minimum": 0.7305279871549081, "maximum": 0.7980526777603415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3725889142013785, "standard_deviation": 0.6685345806385657, "minimum": 0.3564544081252348, "maximum": 2.5329816198118, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10487656956514195, "standard_deviation": 0.01424966341581107, "minimum": 0.08231500004537809, "maximum": 0.12191037521774822, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9816666870352453, "standard_deviation": 0.002039293750735854, "minimum": 0.9780362565091791, "maximum": 0.9840278310008771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6605807472704588, "standard_deviation": 0.027389459683183556, "minimum": 1.610856916267603, "maximum": 1.68244245411025, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8360289174235459, "standard_deviation": 0.3095482363252225, "minimum": 0.5236436094944826, "maximum": 1.456042721143887, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16059415777934855, "standard_deviation": 0.008645871274283519, "minimum": 0.15052538655124112, "maximum": 0.1753287089687993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936920006036893, "standard_deviation": 0.0012014270113735895, "minimum": 0.9916079485206541, "maximum": 0.994975514850087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9240532984790018, "standard_deviation": 0.04890529309530018, "minimum": 0.8364655408152352, "maximum": 1.0004090223965276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7165377574858536, "standard_deviation": 0.3345330826122863, "minimum": 0.35058642594716066, "maximum": 1.2407631701262993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09265354292823169, "standard_deviation": 0.007670795926123585, "minimum": 0.0821832529769957, "maximum": 0.10593009306888389, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19782020011162177, "standard_deviation": 0.043205014505828275, "minimum": 0.13744605010620864, "maximum": 0.26228144241717505, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.9940772269311475, "standard_deviation": 0.011230173330081554, "minimum": 5.984303743365854, "maximum": 6.017370800296586, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.43888813037965513, "standard_deviation": 0.06916083357201128, "minimum": 0.34975913442966233, "maximum": 0.5579537498005938, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08835966746804902, "standard_deviation": 0.10870966515161674, "minimum": -0.276295226866844, "maximum": 0.09282976982189507, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.984956160349259, "standard_deviation": 0.0018398136212180289, "minimum": 0.9811159488742531, "maximum": 0.986728318352004, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3679479453191754, "standard_deviation": 0.05499710852179572, "minimum": 1.2911948544696505, "maximum": 1.4329491587776892, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7044990275823225, "standard_deviation": 0.2779752161600455, "minimum": 0.40536663947828805, "maximum": 1.1864488087197607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.13051493478451154, "standard_deviation": 0.007401193011285232, "minimum": 0.12269314656562069, "maximum": 0.14578533896805454, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9954295179876479, "standard_deviation": 0.0009591275130668023, "minimum": 0.994026868590552, "maximum": 0.9965264240793553, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7162283383925527, "standard_deviation": 0.04274406152734172, "minimum": 0.6508725037951991, "maximum": 0.7632929071524757, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0511989145556933, "standard_deviation": 0.36832146013024614, "minimum": 0.5591197013374573, "maximum": 1.5667378242283458, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07063374997152388, "standard_deviation": 0.006206933373282463, "minimum": 0.06257097382026497, "maximum": 0.07954933509600244, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0019", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.10176682472229, "predicted_mos": 2.10176682472229, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "a8a03739ba75fe1df17ec49cabe9b840bd4a473b58aaf9465fb587b95fbba03a", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1250547003000975, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2719042539655019, "全帯域_10ms_rms_envelope_nrmse": 0.10899696651631645, "全帯域_10ms_envelope_correlation": 0.9895534772124215, "全帯域_10ms_modulation_mae_db": 0.18153308067514418, "全帯域_40ms_log_rms_mae_db": 0.6902710835148629, "全帯域_40ms_rms_envelope_nrmse": 0.059264535858297517, "全帯域_40ms_envelope_correlation": 0.9967060327957248, "全帯域_40ms_modulation_mae_db": 0.5156158754110077, "300–1000 Hz_10ms_log_rms_mae_db": 2.0708086901578477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15813728274242655, "300–1000 Hz_10ms_envelope_correlation": 0.9823272811065642, "300–1000 Hz_10ms_modulation_mae_db": 0.595818050178586, "300–1000 Hz_40ms_log_rms_mae_db": 1.2872675870601615, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08827887517552031, "300–1000 Hz_40ms_envelope_correlation": 0.9941600803750477, "300–1000 Hz_40ms_modulation_mae_db": 0.1341948882653556, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8266091536900886, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1895450030131074, "1000–3000 Hz_10ms_envelope_correlation": 0.9727504605052447, "1000–3000 Hz_10ms_modulation_mae_db": 0.5326818662080325, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1303445814474333, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11008660659671533, "1000–3000 Hz_40ms_envelope_correlation": 0.9904238100155708, "1000–3000 Hz_40ms_modulation_mae_db": 0.6126245328902721, "3000–6000 Hz_10ms_log_rms_mae_db": 1.710089942445485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14824207340092369, "3000–6000 Hz_10ms_envelope_correlation": 0.9859295736398308, "3000–6000 Hz_10ms_modulation_mae_db": 0.706077269200224, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0558170199515138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07407850133660188, "3000–6000 Hz_40ms_envelope_correlation": 0.9965551419286498, "3000–6000 Hz_40ms_modulation_mae_db": 0.6642225189691229, "mr_stft_db": 6.251707941625184, "band_mae_db": 0.25149674762028223, "rms_gain_error_db": -0.014422814213645385, "output_peak": 0.4083824087143868}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.189304210288296, "全帯域_10ms_rms_envelope_nrmse": 0.11666181932015568, "全帯域_10ms_envelope_correlation": 0.988181353544374, "全帯域_10ms_modulation_mae_db": 0.9541500550621905, "全帯域_40ms_log_rms_mae_db": 0.6405223460813325, "全帯域_40ms_rms_envelope_nrmse": 0.05913341832848494, "全帯域_40ms_envelope_correlation": 0.9968005871164668, "全帯域_40ms_modulation_mae_db": 1.2809514452389246, "300–1000 Hz_10ms_log_rms_mae_db": 1.8634655107004818, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15162313490076396, "300–1000 Hz_10ms_envelope_correlation": 0.9836027099809769, "300–1000 Hz_10ms_modulation_mae_db": 0.9266835622422319, "300–1000 Hz_40ms_log_rms_mae_db": 1.09551889269927, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08540853186879077, "300–1000 Hz_40ms_envelope_correlation": 0.994680564090784, "300–1000 Hz_40ms_modulation_mae_db": 0.670117165327083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8934644261646159, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19226690084111914, "1000–3000 Hz_10ms_envelope_correlation": 0.9732536570801633, "1000–3000 Hz_10ms_modulation_mae_db": 0.32237373406401276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0834578243499131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10730478193595401, "1000–3000 Hz_40ms_envelope_correlation": 0.9913428878383626, "1000–3000 Hz_40ms_modulation_mae_db": 0.32181742384665907, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7404369159115483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22096481455750713, "3000–6000 Hz_10ms_envelope_correlation": 0.9762991970934131, "3000–6000 Hz_10ms_modulation_mae_db": 2.7956089678633584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0887146409077775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.158173693714155, "3000–6000 Hz_40ms_envelope_correlation": 0.9898033022424588, "3000–6000 Hz_40ms_modulation_mae_db": 2.9064839904675517, "mr_stft_db": 6.226081097179723, "band_mae_db": 0.3876311645251466, "rms_gain_error_db": 0.06564417647327375, "output_peak": 0.43093068084695013}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2910979802736524, "全帯域_10ms_rms_envelope_nrmse": 0.12742930318543097, "全帯域_10ms_envelope_correlation": 0.9857002756600031, "全帯域_10ms_modulation_mae_db": 0.8242789239147729, "全帯域_40ms_log_rms_mae_db": 0.6530202983236388, "全帯域_40ms_rms_envelope_nrmse": 0.05762193143571448, "全帯域_40ms_envelope_correlation": 0.9968713186549564, "全帯域_40ms_modulation_mae_db": 1.352096456311262, "300–1000 Hz_10ms_log_rms_mae_db": 1.9453614365119483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17499555884953683, "300–1000 Hz_10ms_envelope_correlation": 0.9781560652288489, "300–1000 Hz_10ms_modulation_mae_db": 1.4505681812185172, "300–1000 Hz_40ms_log_rms_mae_db": 1.1277359036588799, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08888966206126658, "300–1000 Hz_40ms_envelope_correlation": 0.9940555250853521, "300–1000 Hz_40ms_modulation_mae_db": 1.335002273727045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8333109526106242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19657789340714749, "1000–3000 Hz_10ms_envelope_correlation": 0.971632079898938, "1000–3000 Hz_10ms_modulation_mae_db": 0.6837253438347326, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0176673836842607, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09362622314133787, "1000–3000 Hz_40ms_envelope_correlation": 0.9932138614591293, "1000–3000 Hz_40ms_modulation_mae_db": 0.7836947126320952, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7570188538990894, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19716211892323757, "3000–6000 Hz_10ms_envelope_correlation": 0.9749265142683563, "3000–6000 Hz_10ms_modulation_mae_db": 0.9646420803714145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0781184683581344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08692962065441007, "3000–6000 Hz_40ms_envelope_correlation": 0.9950503598390469, "3000–6000 Hz_40ms_modulation_mae_db": 1.317860397047527, "mr_stft_db": 6.244575423785824, "band_mae_db": 0.24537572775720876, "rms_gain_error_db": -0.03610049949357624, "output_peak": 0.40127757015420706}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3494429610261967, "全帯域_10ms_rms_envelope_nrmse": 0.11278572644350462, "全帯域_10ms_envelope_correlation": 0.9890681894516288, "全帯域_10ms_modulation_mae_db": 0.5477305710528708, "全帯域_40ms_log_rms_mae_db": 0.6794010370783412, "全帯域_40ms_rms_envelope_nrmse": 0.059642788759405596, "全帯域_40ms_envelope_correlation": 0.9972284055850498, "全帯域_40ms_modulation_mae_db": 0.889388009975252, "300–1000 Hz_10ms_log_rms_mae_db": 1.9742417888292443, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16526637758250134, "300–1000 Hz_10ms_envelope_correlation": 0.9806514009488775, "300–1000 Hz_10ms_modulation_mae_db": 1.2662142065750341, "300–1000 Hz_40ms_log_rms_mae_db": 1.2057145684567685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09049341276428094, "300–1000 Hz_40ms_envelope_correlation": 0.9943470815063216, "300–1000 Hz_40ms_modulation_mae_db": 1.011171385405093, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8967429705479901, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20823049930067764, "1000–3000 Hz_10ms_envelope_correlation": 0.9672204055846646, "1000–3000 Hz_10ms_modulation_mae_db": 0.5598070462044649, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1356711660402152, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11412890514390918, "1000–3000 Hz_40ms_envelope_correlation": 0.9903771712825685, "1000–3000 Hz_40ms_modulation_mae_db": 0.756759261115901, "3000–6000 Hz_10ms_log_rms_mae_db": 1.66355592144687, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15173326077477614, "3000–6000 Hz_10ms_envelope_correlation": 0.9854098271171645, "3000–6000 Hz_10ms_modulation_mae_db": 0.6998080285733598, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0275840796507425, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08610633484527663, "3000–6000 Hz_40ms_envelope_correlation": 0.995577336105759, "3000–6000 Hz_40ms_modulation_mae_db": 0.8410681084907322, "mr_stft_db": 6.24237273951601, "band_mae_db": 0.3580312306906101, "rms_gain_error_db": -0.23396011243628612, "output_peak": 0.42150361794433244}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.351260132965468, "全帯域_10ms_rms_envelope_nrmse": 0.13377280155329566, "全帯域_10ms_envelope_correlation": 0.9843367880728061, "全帯域_10ms_modulation_mae_db": 0.6908781375139288, "全帯域_40ms_log_rms_mae_db": 0.733848890974639, "全帯域_40ms_rms_envelope_nrmse": 0.07179089953106045, "全帯域_40ms_envelope_correlation": 0.9955981805645614, "全帯域_40ms_modulation_mae_db": 0.9342613728790689, "300–1000 Hz_10ms_log_rms_mae_db": 2.0207396872126293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18035700997555923, "300–1000 Hz_10ms_envelope_correlation": 0.976751405743714, "300–1000 Hz_10ms_modulation_mae_db": 1.2381982626090202, "300–1000 Hz_40ms_log_rms_mae_db": 1.1969091419930373, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10533339142997215, "300–1000 Hz_40ms_envelope_correlation": 0.991928907892228, "300–1000 Hz_40ms_modulation_mae_db": 0.8322642725813201, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8289730176121628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1797387521235054, "1000–3000 Hz_10ms_envelope_correlation": 0.9755471950202942, "1000–3000 Hz_10ms_modulation_mae_db": 0.5999300160882448, "1000–3000 Hz_40ms_log_rms_mae_db": 1.16312691087799, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09000868703769078, "1000–3000 Hz_40ms_envelope_correlation": 0.9937301444032487, "1000–3000 Hz_40ms_modulation_mae_db": 0.42477947196581417, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7476288320925577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.198529657978357, "3000–6000 Hz_10ms_envelope_correlation": 0.9752207972154769, "3000–6000 Hz_10ms_modulation_mae_db": 0.8952071838161908, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1257028502684583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11675815708799298, "3000–6000 Hz_40ms_envelope_correlation": 0.9926035298785625, "3000–6000 Hz_40ms_modulation_mae_db": 1.0573099610201604, "mr_stft_db": 6.251213332020058, "band_mae_db": 0.3085052296307713, "rms_gain_error_db": -0.21917648373027881, "output_peak": 0.4499891987835806}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2375155925925865, "全帯域_10ms_rms_envelope_nrmse": 0.13407377079040642, "全帯域_10ms_envelope_correlation": 0.9841153444968886, "全帯域_10ms_modulation_mae_db": 0.5894916460550251, "全帯域_40ms_log_rms_mae_db": 0.6705072184823546, "全帯域_40ms_rms_envelope_nrmse": 0.07787491105758046, "全帯域_40ms_envelope_correlation": 0.9942706867962581, "全帯域_40ms_modulation_mae_db": 0.3943830634257948, "300–1000 Hz_10ms_log_rms_mae_db": 1.9055116280960966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16315094815660694, "300–1000 Hz_10ms_envelope_correlation": 0.9814611530331995, "300–1000 Hz_10ms_modulation_mae_db": 0.4899344969132429, "300–1000 Hz_40ms_log_rms_mae_db": 1.1520684927446694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09624190531413958, "300–1000 Hz_40ms_envelope_correlation": 0.9932379716198674, "300–1000 Hz_40ms_modulation_mae_db": 0.7279850099372679, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8918503373156346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19336099402199908, "1000–3000 Hz_10ms_envelope_correlation": 0.9716556739347243, "1000–3000 Hz_10ms_modulation_mae_db": 0.33318861050762216, "1000–3000 Hz_40ms_log_rms_mae_db": 1.105168810514633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08968403859279064, "1000–3000 Hz_40ms_envelope_correlation": 0.9935766905127271, "1000–3000 Hz_40ms_modulation_mae_db": 0.31224286400718687, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7560101531245516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17587270448358325, "3000–6000 Hz_10ms_envelope_correlation": 0.9801734116915816, "3000–6000 Hz_10ms_modulation_mae_db": 1.385530494077818, "3000–6000 Hz_40ms_log_rms_mae_db": 1.081794544284341, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08907972148046306, "3000–6000 Hz_40ms_envelope_correlation": 0.9953141953485443, "3000–6000 Hz_40ms_modulation_mae_db": 1.3152887542196796, "mr_stft_db": 6.256138119045257, "band_mae_db": 0.3511753413267207, "rms_gain_error_db": -0.012694082492350746, "output_peak": 0.3986325642476312}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2789474537053154, "全帯域_10ms_rms_envelope_nrmse": 0.11860762919836174, "全帯域_10ms_envelope_correlation": 0.9875394456858101, "全帯域_10ms_modulation_mae_db": 0.5116368315926055, "全帯域_40ms_log_rms_mae_db": 0.7079074542174854, "全帯域_40ms_rms_envelope_nrmse": 0.06660256677334722, "全帯域_40ms_envelope_correlation": 0.9958426650536959, "全帯域_40ms_modulation_mae_db": 0.7779715773902488, "300–1000 Hz_10ms_log_rms_mae_db": 1.9803561183783283, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15721350020639635, "300–1000 Hz_10ms_envelope_correlation": 0.9825661942467439, "300–1000 Hz_10ms_modulation_mae_db": 0.7731078912848655, "300–1000 Hz_40ms_log_rms_mae_db": 1.2195763126103945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.094801391148293, "300–1000 Hz_40ms_envelope_correlation": 0.9937495863507273, "300–1000 Hz_40ms_modulation_mae_db": 0.8604596113910935, "1000–3000 Hz_10ms_log_rms_mae_db": 1.911427516365042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19235899744796844, "1000–3000 Hz_10ms_envelope_correlation": 0.9719422044922744, "1000–3000 Hz_10ms_modulation_mae_db": 0.38812981821319276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1669858582302366, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10882315956793188, "1000–3000 Hz_40ms_envelope_correlation": 0.9905925055894151, "1000–3000 Hz_40ms_modulation_mae_db": 0.165599617427728, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6985931914273091, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19148579585442244, "3000–6000 Hz_10ms_envelope_correlation": 0.9801675110795859, "3000–6000 Hz_10ms_modulation_mae_db": 1.7613911847108892, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0403778721737966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11594400370467862, "3000–6000 Hz_40ms_envelope_correlation": 0.9939387441407095, "3000–6000 Hz_40ms_modulation_mae_db": 1.695005057519029, "mr_stft_db": 6.240410920016952, "band_mae_db": 0.4252843654251472, "rms_gain_error_db": -0.08130055027275568, "output_peak": 0.47334419919539383}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2999945850830739, "全帯域_10ms_rms_envelope_nrmse": 0.12060724800347349, "全帯域_10ms_envelope_correlation": 0.9873831774203967, "全帯域_10ms_modulation_mae_db": 0.9519079650222496, "全帯域_40ms_log_rms_mae_db": 0.7028281268144912, "全帯域_40ms_rms_envelope_nrmse": 0.07023556418164595, "全帯域_40ms_envelope_correlation": 0.9954476987460142, "全帯域_40ms_modulation_mae_db": 1.2026194457588808, "300–1000 Hz_10ms_log_rms_mae_db": 1.9761038100187083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16226039904564526, "300–1000 Hz_10ms_envelope_correlation": 0.9813669530625828, "300–1000 Hz_10ms_modulation_mae_db": 1.0941429576636508, "300–1000 Hz_40ms_log_rms_mae_db": 1.2181494152038987, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09220914940970248, "300–1000 Hz_40ms_envelope_correlation": 0.993606737076607, "300–1000 Hz_40ms_modulation_mae_db": 0.6352915921295543, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9181742412844487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21697412508560962, "1000–3000 Hz_10ms_envelope_correlation": 0.9648572767082999, "1000–3000 Hz_10ms_modulation_mae_db": 0.39195260902486057, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1821462198322477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12005815109853439, "1000–3000 Hz_40ms_envelope_correlation": 0.9884279546149685, "1000–3000 Hz_40ms_modulation_mae_db": 0.035812508940135546, "3000–6000 Hz_10ms_log_rms_mae_db": 1.737964604031525, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1872439156386873, "3000–6000 Hz_10ms_envelope_correlation": 0.9782387625527587, "3000–6000 Hz_10ms_modulation_mae_db": 1.4128436822805372, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0600211348127107, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08146693077636653, "3000–6000 Hz_40ms_envelope_correlation": 0.9960380716028945, "3000–6000 Hz_40ms_modulation_mae_db": 1.6223836005792667, "mr_stft_db": 6.224815256985452, "band_mae_db": 0.28106187614116335, "rms_gain_error_db": 0.054050553214689226, "output_peak": 0.47643842959979377}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60480, "output_samples": 60480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2823484953468345, "spectral_convergence": 0.45054655932792226, "spectral_convergence_status": "ok", "amplitude_floor": 9.224568988143785e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.214450187148987, "spectral_convergence": 0.4873225479244195, "spectral_convergence_status": "ok", "amplitude_floor": 8.440100575569802e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.2583251423797295, "spectral_convergence": 0.49118655294763547, "spectral_convergence_status": "ok", "amplitude_floor": 6.303303867538867e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.6236728988818796e-05, "output_power": 5.655014574496923e-05, "delta_db": 0.8744662616243916, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005640699493062965, "output_power": 0.0005667678778609095, "delta_db": 0.020722647912361367, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007824000374604723, "output_power": 0.0007823325647538152, "delta_db": -0.0003745435374321129, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028236131111099825, "output_power": 0.0002897788357431257, "delta_db": 0.11261473384011061, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018611455363091946, "output_power": 0.0001730183898147401, "delta_db": -0.3168806909229386, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00035218826443925665, "output_power": 0.000337584672226554, "delta_db": -0.18392160788445894, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25149674762028223, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.014422814213645385, "pitch": null, "output_rms_dbfs": -26.563880964065966, "output_peak": 0.4083824087143868, "reference_rms_dbfs": -26.54945814985232, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.251707941625184}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00014780031156712295, "log_rms_mae_db": 1.2719042539655019, "rms_envelope_nrmse": 0.10899696651631645, "envelope_correlation": 0.9895534772124215, "rms_gain_error_db": -0.014420687392592333, "reference_mean_power": 0.0022162793576100603, "output_mean_power": 0.002208932437493713}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 1.2692207138635437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044041649987677594, "output_power": 0.00044079702176903784, "reference_active": true, "delta_db": 0.003750704324548261}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888390879465172, "output_power": 0.00013246341054733808, "reference_active": true, "delta_db": 0.4697285681558936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.435323106110239e-05, "output_power": 2.4755322374802813e-05, "reference_active": true, "delta_db": 0.07111996954499067}], "active_band_count": 3, "active_band_mae_db": 0.18153308067514418}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00013486753957748745, "log_rms_mae_db": 0.6902710835148629, "rms_envelope_nrmse": 0.059264535858297517, "envelope_correlation": 0.9967060327957248, "rms_gain_error_db": -0.01439383269720947, "reference_mean_power": 0.0022339157834068563, "output_mean_power": 0.0022265241666235636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.176190391785113e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004109863651179866, "output_power": 0.00041760532473928585, "reference_active": true, "delta_db": 0.06938613603332162}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3981406650238314e-05, "output_power": 6.071337011178974e-05, "reference_active": true, "delta_db": 0.5104014336802292}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0096166790138946e-06, "output_power": 1.2614281608876907e-06, "reference_active": true, "delta_db": 0.9670600565194722}], "active_band_count": 3, "active_band_mae_db": 0.5156158754110077}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00012936155891805085, "log_rms_mae_db": 2.0708086901578477, "rms_envelope_nrmse": 0.15813728274242655, "envelope_correlation": 0.9823272811065642, "rms_gain_error_db": 0.034236130222501805, "reference_mean_power": 0.0012072233727869219, "output_mean_power": 0.0012167777179018363}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 8.078028680738071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004104434555454037, "output_power": 0.0003978617360359804, "reference_active": true, "delta_db": -0.13521161994848196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.258149083667088e-05, "output_power": 9.657670621280225e-05, "reference_active": true, "delta_db": 0.1834821959715968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4118213823838e-05, "output_power": 1.979957137030638e-05, "reference_active": true, "delta_db": 1.4687603346156795}], "active_band_count": 3, "active_band_mae_db": 0.595818050178586}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00011150674016193665, "log_rms_mae_db": 1.2872675870601615, "rms_envelope_nrmse": 0.08827887517552031, "envelope_correlation": 0.9941600803750477, "rms_gain_error_db": 0.03423342219603002, "reference_mean_power": 0.0012168889554241998, "output_mean_power": 0.0012265190322058516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.50453906071102e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038524435492848886, "output_power": 0.0003819816719082623, "reference_active": true, "delta_db": -0.036937583555512826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.764413089442773e-05, "output_power": 4.59204477460957e-05, "reference_active": true, "delta_db": -0.16003295139151474}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0419619324101079e-06, "output_power": 9.937804285460858e-07, "reference_active": true, "delta_db": -0.20561412984903926}], "active_band_count": 3, "active_band_mae_db": 0.1341948882653556}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 7.756656945350889e-05, "log_rms_mae_db": 1.8266091536900886, "rms_envelope_nrmse": 0.1895450030131074, "envelope_correlation": 0.9727504605052447, "rms_gain_error_db": -0.18796170665083134, "reference_mean_power": 0.0003662329710023159, "output_mean_power": 0.00035072059517063354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 2.3479287649550626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015936395121522886, "output_power": 0.00014334913447615496, "reference_active": true, "delta_db": -0.4599501388619788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.480344306075996e-05, "output_power": 2.2948217557749818e-05, "reference_active": true, "delta_db": -0.33763012583044766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.158977700048883e-06, "output_power": 5.953940307432026e-06, "reference_active": true, "delta_db": -0.8004653339316709}], "active_band_count": 3, "active_band_mae_db": 0.5326818662080325}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 6.756499548428438e-05, "log_rms_mae_db": 1.1303445814474333, "rms_envelope_nrmse": 0.11008660659671533, "envelope_correlation": 0.9904238100155708, "rms_gain_error_db": -0.1879872433331347, "reference_mean_power": 0.00036915132536056565, "output_mean_power": 0.00035351325934380486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 2.14579734589437e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015227225639161801, "output_power": 0.00013803160611245728, "reference_active": true, "delta_db": -0.4264224182937354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5037342052746137e-05, "output_power": 1.309798692520534e-05, "reference_active": true, "delta_db": -0.5996652611914061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0212382710455496e-07, "output_power": 3.3356556848737953e-07, "reference_active": true, "delta_db": -0.8117859191856746}], "active_band_count": 3, "active_band_mae_db": 0.6126245328902721}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00011110909618382604, "log_rms_mae_db": 1.710089942445485, "rms_envelope_nrmse": 0.14824207340092369, "envelope_correlation": 0.9859295736398308, "rms_gain_error_db": -0.1038673735527938, "reference_mean_power": 0.00024441441718797436, "output_mean_power": 0.00023863826482690333}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 3.469643399885664e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011046778972231385, "output_power": 0.00010612126563749808, "reference_active": true, "delta_db": -0.17433243699871967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.9627372395634935e-05, "output_power": 5.116501580568424e-05, "reference_active": true, "delta_db": -0.6647255905290352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.408768678752184e-06, "output_power": 3.2839756984245587e-06, "reference_active": true, "delta_db": -1.279173780072917}], "active_band_count": 3, "active_band_mae_db": 0.706077269200224}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.166729119583493e-05, "log_rms_mae_db": 1.0558170199515138, "rms_envelope_nrmse": 0.07407850133660188, "envelope_correlation": 0.9965551419286498, "rms_gain_error_db": -0.10387646432566155, "reference_mean_power": 0.00024636420339127556, "output_mean_power": 0.00024054146897060902}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.3341266023063243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010489106418058692, "output_power": 9.85678258944267e-05, "reference_active": true, "delta_db": -0.2700331415429724}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0862352851100365e-05, "output_power": 2.932290118601805e-05, "reference_active": true, "delta_db": -0.222220954005494}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.747611403563906e-07, "output_power": 2.6528531230747217e-07, "reference_active": true, "delta_db": -1.5004134613589022}], "active_band_count": 3, "active_band_mae_db": 0.6642225189691229}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6326748145088862, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c0566f427c731d152503ccc044e935d1e7a0e34422147b17201026d7d2f106f0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9711073691530754, "standard_deviation": 0.0034322713363255284, "minimum": 0.9648572767082999, "maximum": 0.9755471950202942, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8750690769488259, "standard_deviation": 0.03870529742639091, "minimum": 1.8266091536900886, "maximum": 1.9181742412844487, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4764736305181454, "standard_deviation": 0.1349732158792171, "minimum": 0.32237373406401276, "maximum": 0.6837253438347326, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19613164565514177, "standard_deviation": 0.011526037619729963, "minimum": 0.1797387521235054, "maximum": 0.21697412508560962, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9914606282144989, "standard_deviation": 0.0018866345676013337, "minimum": 0.9884279546149685, "maximum": 0.9937301444032487, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1230710943721163, "standard_deviation": 0.05376424189393654, "minimum": 1.0176673836842607, "maximum": 1.1821462198322477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.426666299103224, "standard_deviation": 0.2715098480652064, "minimum": 0.035812508940135546, "maximum": 0.7836947126320952, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10421506913935802, "standard_deviation": 0.011585945708366177, "minimum": 0.08968403859279064, "maximum": 0.12005815109853439, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9795456993322709, "standard_deviation": 0.004279720084546691, "minimum": 0.9749265142683563, "maximum": 0.9859295736398308, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.726412301797367, "standard_deviation": 0.03291618652584408, "minimum": 1.66355592144687, "maximum": 1.7570188538990894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.327638611361724, "standard_deviation": 0.7023392343186651, "minimum": 0.6998080285733598, "maximum": 2.7956089678633584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18390429270143682, "standard_deviation": 0.024503823046449862, "minimum": 0.14824207340092369, "maximum": 0.22096481455750713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943600851358282, "standard_deviation": 0.002220994090118186, "minimum": 0.9898033022424588, "maximum": 0.9965551419286498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0697663263009343, "standard_deviation": 0.030787160398966684, "minimum": 1.0275840796507425, "maximum": 1.1257028502684583, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4274527985391336, "standard_deviation": 0.6954012148610448, "minimum": 0.6642225189691229, "maximum": 2.9064839904675517, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1010671204499931, "standard_deviation": 0.027796802831180377, "minimum": 0.07407850133660188, "maximum": 0.158173693714155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9808603954189384, "standard_deviation": 0.0023127034085905877, "minimum": 0.976751405743714, "maximum": 0.9836027099809769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9670735837381605, "standard_deviation": 0.06430541119381383, "minimum": 1.8634655107004818, "maximum": 2.0708086901578477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9793334510856436, "standard_deviation": 0.34151999389025384, "minimum": 0.4899344969132429, "maximum": 1.4505681812185172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16412552643242956, "standard_deviation": 0.009466434463350767, "minimum": 0.15162313490076396, "maximum": 0.18035700997555923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9937208067496168, "standard_deviation": 0.0008515719364344161, "minimum": 0.991928907892228, "maximum": 0.994680564090784, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.187867539303385, "standard_deviation": 0.06050824348200884, "minimum": 1.09551889269927, "maximum": 1.2872675870601615, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7758107748454766, "standard_deviation": 0.3429529750906665, "minimum": 0.1341948882653556, "maximum": 1.335002273727045, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09270703989649573, "standard_deviation": 0.006197967518405439, "minimum": 0.08540853186879077, "maximum": 0.10533339142997215, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3260702103896313, "standard_deviation": 0.06512850333778726, "minimum": 0.24537572775720876, "maximum": 0.4252843654251472, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.242164353771807, "standard_deviation": 0.011568576325809718, "minimum": 6.224815256985452, "maximum": 6.256138119045257, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4325623336857845, "standard_deviation": 0.031010504934498948, "minimum": 0.3986325642476312, "maximum": 0.47643842959979377, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05974497661886625, "standard_deviation": 0.11322087795116524, "minimum": -0.23396011243628612, "maximum": 0.06564417647327375, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9869847564430412, "standard_deviation": 0.0020609441607102276, "minimum": 0.9841153444968886, "maximum": 0.9895534772124215, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2836833962375114, "standard_deviation": 0.05400005400016729, "minimum": 1.189304210288296, "maximum": 1.351260132965468, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6564509013610984, "standard_deviation": 0.2585931292940154, "minimum": 0.18153308067514418, "maximum": 0.9541500550621905, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12161690812636812, "standard_deviation": 0.009324345118519168, "minimum": 0.10899696651631645, "maximum": 0.13407377079040642, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.996095696914091, "standard_deviation": 0.0009867775332168664, "minimum": 0.9942706867962581, "maximum": 0.9972284055850498, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6847883069358932, "standard_deviation": 0.03046915452673783, "minimum": 0.6405223460813325, "maximum": 0.733848890974639, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9184109057988049, "standard_deviation": 0.3500936773589068, "minimum": 0.3943830634257948, "maximum": 1.352096456311262, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06527082699069209, "standard_deviation": 0.007480643241236461, "minimum": 0.05762193143571448, "maximum": 0.07787491105758046, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0019", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.334247589111328, "predicted_mos": 2.334247589111328, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "0c46e592a9ff22f384e801d85d32014d8b74f60fde65c5186989eb77d62d1b6b", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.151231899857521, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3556737364705194, "全帯域_10ms_rms_envelope_nrmse": 0.1309863510291073, "全帯域_10ms_envelope_correlation": 0.9852929200723698, "全帯域_10ms_modulation_mae_db": 1.451674597283126, "全帯域_40ms_log_rms_mae_db": 0.7669620303561198, "全帯域_40ms_rms_envelope_nrmse": 0.08544505593238866, "全帯域_40ms_envelope_correlation": 0.9939565969286471, "全帯域_40ms_modulation_mae_db": 1.380252784950833, "300–1000 Hz_10ms_log_rms_mae_db": 1.5620732146497305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17376019129853226, "300–1000 Hz_10ms_envelope_correlation": 0.979397764138133, "300–1000 Hz_10ms_modulation_mae_db": 1.4123073517961613, "300–1000 Hz_40ms_log_rms_mae_db": 0.9403669408002561, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11608239949553695, "300–1000 Hz_40ms_envelope_correlation": 0.9914942626665835, "300–1000 Hz_40ms_modulation_mae_db": 1.720010365125476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4780734694514643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18492921561066622, "1000–3000 Hz_10ms_envelope_correlation": 0.9743409992324904, "1000–3000 Hz_10ms_modulation_mae_db": 0.6839791209357452, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9134929985549499, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10455214001044962, "1000–3000 Hz_40ms_envelope_correlation": 0.9911871076274417, "1000–3000 Hz_40ms_modulation_mae_db": 0.625289245474507, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2182494387661893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14230203168618008, "3000–6000 Hz_10ms_envelope_correlation": 0.9877688366558396, "3000–6000 Hz_10ms_modulation_mae_db": 0.722906661664883, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7295060698786525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08778988989475728, "3000–6000 Hz_40ms_envelope_correlation": 0.9960953838007501, "3000–6000 Hz_40ms_modulation_mae_db": 0.5779038289606641, "mr_stft_db": 5.935921225621617, "band_mae_db": 0.38128530298949054, "rms_gain_error_db": -0.29648392216375374, "output_peak": 0.32441084522809055}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4301040240828429, "全帯域_10ms_rms_envelope_nrmse": 0.13817613955532598, "全帯域_10ms_envelope_correlation": 0.9836486030869934, "全帯域_10ms_modulation_mae_db": 1.3469936973692536, "全帯域_40ms_log_rms_mae_db": 0.774512075253831, "全帯域_40ms_rms_envelope_nrmse": 0.05998489008707448, "全帯域_40ms_envelope_correlation": 0.9968396857307725, "全帯域_40ms_modulation_mae_db": 1.8963379794775932, "300–1000 Hz_10ms_log_rms_mae_db": 1.5301815989166039, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1680851692115233, "300–1000 Hz_10ms_envelope_correlation": 0.9807842697627956, "300–1000 Hz_10ms_modulation_mae_db": 1.6054107824904917, "300–1000 Hz_40ms_log_rms_mae_db": 0.8312675343239639, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08123790540936103, "300–1000 Hz_40ms_envelope_correlation": 0.9954428491633427, "300–1000 Hz_40ms_modulation_mae_db": 1.4450033429932552, "1000–3000 Hz_10ms_log_rms_mae_db": 1.437129753100144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1751281604543032, "1000–3000 Hz_10ms_envelope_correlation": 0.9768932927485183, "1000–3000 Hz_10ms_modulation_mae_db": 0.45375186665879513, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8520877432045345, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0945016338100334, "1000–3000 Hz_40ms_envelope_correlation": 0.9928008450037757, "1000–3000 Hz_40ms_modulation_mae_db": 0.3337210083753345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2320237930123235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20594893507961284, "3000–6000 Hz_10ms_envelope_correlation": 0.9742074815150046, "3000–6000 Hz_10ms_modulation_mae_db": 2.319054421860105, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7454594474257956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09064881953441246, "3000–6000 Hz_40ms_envelope_correlation": 0.9952644567403085, "3000–6000 Hz_40ms_modulation_mae_db": 2.495307105642615, "mr_stft_db": 5.928203100085642, "band_mae_db": 0.22941272791923392, "rms_gain_error_db": 0.12228941175238191, "output_peak": 0.49569729012146635}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3974114694612827, "全帯域_10ms_rms_envelope_nrmse": 0.12992005383679067, "全帯域_10ms_envelope_correlation": 0.9850772133085903, "全帯域_10ms_modulation_mae_db": 0.9718796663113691, "全帯域_40ms_log_rms_mae_db": 0.7705263037348256, "全帯域_40ms_rms_envelope_nrmse": 0.07531180409877894, "全帯域_40ms_envelope_correlation": 0.994846732471614, "全帯域_40ms_modulation_mae_db": 1.216543691307818, "300–1000 Hz_10ms_log_rms_mae_db": 1.5200145677832537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1612602723755555, "300–1000 Hz_10ms_envelope_correlation": 0.981466362270264, "300–1000 Hz_10ms_modulation_mae_db": 1.1555910849039226, "300–1000 Hz_40ms_log_rms_mae_db": 0.9077604120850098, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0945714033069481, "300–1000 Hz_40ms_envelope_correlation": 0.9933527089472267, "300–1000 Hz_40ms_modulation_mae_db": 1.2546391660596143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4918440448359618, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20133433233943965, "1000–3000 Hz_10ms_envelope_correlation": 0.9703347911251248, "1000–3000 Hz_10ms_modulation_mae_db": 1.2630768672625836, "1000–3000 Hz_40ms_log_rms_mae_db": 0.88757250130239, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1270355021884234, "1000–3000 Hz_40ms_envelope_correlation": 0.9891121975750391, "1000–3000 Hz_40ms_modulation_mae_db": 1.5673206990092725, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2034178821000139, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16668951008742866, "3000–6000 Hz_10ms_envelope_correlation": 0.9820917991015482, "3000–6000 Hz_10ms_modulation_mae_db": 1.4658193437697864, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7269270864605609, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08873199141696483, "3000–6000 Hz_40ms_envelope_correlation": 0.9947320670219592, "3000–6000 Hz_40ms_modulation_mae_db": 1.1050805348048354, "mr_stft_db": 5.912441063501583, "band_mae_db": 0.3026992101882076, "rms_gain_error_db": -0.16017116221020175, "output_peak": 0.35923332596538454}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3065736052108166, "全帯域_10ms_rms_envelope_nrmse": 0.11110373226489131, "全帯域_10ms_envelope_correlation": 0.9890737463243916, "全帯域_10ms_modulation_mae_db": 0.15811390498889502, "全帯域_40ms_log_rms_mae_db": 0.6551422128036704, "全帯域_40ms_rms_envelope_nrmse": 0.060941289511008474, "全帯域_40ms_envelope_correlation": 0.9965536607535695, "全帯域_40ms_modulation_mae_db": 0.2870844342740951, "300–1000 Hz_10ms_log_rms_mae_db": 1.595776001431446, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13940224558654715, "300–1000 Hz_10ms_envelope_correlation": 0.986251760430268, "300–1000 Hz_10ms_modulation_mae_db": 0.7488500340634401, "300–1000 Hz_40ms_log_rms_mae_db": 0.9810535530035114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08471666370750368, "300–1000 Hz_40ms_envelope_correlation": 0.994872187069833, "300–1000 Hz_40ms_modulation_mae_db": 0.6383501924992206, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4047622416344128, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.177227082569126, "1000–3000 Hz_10ms_envelope_correlation": 0.9767940141758241, "1000–3000 Hz_10ms_modulation_mae_db": 0.4136872651096191, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8529773149330738, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0979993548217257, "1000–3000 Hz_40ms_envelope_correlation": 0.9924063598395172, "1000–3000 Hz_40ms_modulation_mae_db": 0.7069509295216365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.138513446744298, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15964327950658086, "3000–6000 Hz_10ms_envelope_correlation": 0.9836283313562801, "3000–6000 Hz_10ms_modulation_mae_db": 0.7510981647658506, "3000–6000 Hz_40ms_log_rms_mae_db": 0.685966552461424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07722640970414177, "3000–6000 Hz_40ms_envelope_correlation": 0.9960149817269591, "3000–6000 Hz_40ms_modulation_mae_db": 0.5869421182942931, "mr_stft_db": 5.914965564664908, "band_mae_db": 0.2557609387163664, "rms_gain_error_db": -0.09391230030457672, "output_peak": 0.43670593532229685}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3829837899170645, "全帯域_10ms_rms_envelope_nrmse": 0.12559585880795235, "全帯域_10ms_envelope_correlation": 0.9860152088009099, "全帯域_10ms_modulation_mae_db": 0.4675495623655426, "全帯域_40ms_log_rms_mae_db": 0.7876172424719009, "全帯域_40ms_rms_envelope_nrmse": 0.07257725503389921, "全帯域_40ms_envelope_correlation": 0.9950811793063591, "全帯域_40ms_modulation_mae_db": 0.5862539774159208, "300–1000 Hz_10ms_log_rms_mae_db": 1.6140312006848763, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1675341097778942, "300–1000 Hz_10ms_envelope_correlation": 0.98001356729629, "300–1000 Hz_10ms_modulation_mae_db": 0.6020773394208833, "300–1000 Hz_40ms_log_rms_mae_db": 0.9743186162032953, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10777793906875253, "300–1000 Hz_40ms_envelope_correlation": 0.9912475408450823, "300–1000 Hz_40ms_modulation_mae_db": 0.5694168244397848, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4058988275296973, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16327464472932957, "1000–3000 Hz_10ms_envelope_correlation": 0.9807187075220603, "1000–3000 Hz_10ms_modulation_mae_db": 0.7630422886393958, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8410949235904551, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09268469509597135, "1000–3000 Hz_40ms_envelope_correlation": 0.9945996815707517, "1000–3000 Hz_40ms_modulation_mae_db": 0.6463870230790335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.221335481049804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14992232536030117, "3000–6000 Hz_10ms_envelope_correlation": 0.9856419597180085, "3000–6000 Hz_10ms_modulation_mae_db": 0.7034569460460328, "3000–6000 Hz_40ms_log_rms_mae_db": 0.771850419544385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08606297511169969, "3000–6000 Hz_40ms_envelope_correlation": 0.9951136202472424, "3000–6000 Hz_40ms_modulation_mae_db": 0.48157414066945003, "mr_stft_db": 5.911402232990572, "band_mae_db": 0.3565371091163844, "rms_gain_error_db": -0.10036272225211262, "output_peak": 0.3549237362959353}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.335435294047688, "全帯域_10ms_rms_envelope_nrmse": 0.12692279452627095, "全帯域_10ms_envelope_correlation": 0.9859043208326012, "全帯域_10ms_modulation_mae_db": 0.5262855622486365, "全帯域_40ms_log_rms_mae_db": 0.7078817474523211, "全帯域_40ms_rms_envelope_nrmse": 0.07577655283718085, "全帯域_40ms_envelope_correlation": 0.9946380044636629, "全帯域_40ms_modulation_mae_db": 0.7378893517573593, "300–1000 Hz_10ms_log_rms_mae_db": 1.5851187942876896, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14704229668027202, "300–1000 Hz_10ms_envelope_correlation": 0.9845726175946268, "300–1000 Hz_10ms_modulation_mae_db": 0.4849582169872666, "300–1000 Hz_40ms_log_rms_mae_db": 0.9584886784610283, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08532410554494542, "300–1000 Hz_40ms_envelope_correlation": 0.9945991775484122, "300–1000 Hz_40ms_modulation_mae_db": 0.5789379335677164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.520272742486192, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20007463104266934, "1000–3000 Hz_10ms_envelope_correlation": 0.9704530359000562, "1000–3000 Hz_10ms_modulation_mae_db": 0.9267212877666143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.893771481134512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11507405813840473, "1000–3000 Hz_40ms_envelope_correlation": 0.9895051813457673, "1000–3000 Hz_40ms_modulation_mae_db": 1.0305278860977842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2224785582240922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20629108074372113, "3000–6000 Hz_10ms_envelope_correlation": 0.9807616113628008, "3000–6000 Hz_10ms_modulation_mae_db": 1.8815017651116033, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7398961720002489, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1666969771702814, "3000–6000 Hz_40ms_envelope_correlation": 0.9889142393009636, "3000–6000 Hz_40ms_modulation_mae_db": 2.1965464950370106, "mr_stft_db": 5.923684463298024, "band_mae_db": 0.2957421637968202, "rms_gain_error_db": 0.02811117020330101, "output_peak": 0.41829984646979673}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3944868952651819, "全帯域_10ms_rms_envelope_nrmse": 0.12436065401180262, "全帯域_10ms_envelope_correlation": 0.9863671257105946, "全帯域_10ms_modulation_mae_db": 0.4083939749529426, "全帯域_40ms_log_rms_mae_db": 0.7363904459654428, "全帯域_40ms_rms_envelope_nrmse": 0.07754576600069123, "全帯域_40ms_envelope_correlation": 0.9944550730790042, "全帯域_40ms_modulation_mae_db": 0.1650230438777157, "300–1000 Hz_10ms_log_rms_mae_db": 1.5337920152526452, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1453844130434753, "300–1000 Hz_10ms_envelope_correlation": 0.9850320516070313, "300–1000 Hz_10ms_modulation_mae_db": 0.37639650377287687, "300–1000 Hz_40ms_log_rms_mae_db": 0.8024849912948638, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09655786387104476, "300–1000 Hz_40ms_envelope_correlation": 0.9931393545078663, "300–1000 Hz_40ms_modulation_mae_db": 0.47774336653921984, "1000–3000 Hz_10ms_log_rms_mae_db": 1.446337699512742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1719508851261953, "1000–3000 Hz_10ms_envelope_correlation": 0.9778224092505804, "1000–3000 Hz_10ms_modulation_mae_db": 0.6734653346827796, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8376722047644559, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10396709261699985, "1000–3000 Hz_40ms_envelope_correlation": 0.9912901546984022, "1000–3000 Hz_40ms_modulation_mae_db": 0.9273770942886724, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1731123441889808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1602584759237213, "3000–6000 Hz_10ms_envelope_correlation": 0.9841469355716022, "3000–6000 Hz_10ms_modulation_mae_db": 0.5088925038494251, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7172563217710918, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09710305707639205, "3000–6000 Hz_40ms_envelope_correlation": 0.9949579687157248, "3000–6000 Hz_40ms_modulation_mae_db": 0.8204351587923128, "mr_stft_db": 5.932882894990963, "band_mae_db": 0.17792269310660178, "rms_gain_error_db": -0.1334527032148181, "output_peak": 0.3702439566437306}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3474532703830717, "全帯域_10ms_rms_envelope_nrmse": 0.1418256788328344, "全帯域_10ms_envelope_correlation": 0.9833102849977267, "全帯域_10ms_modulation_mae_db": 0.8670180169212465, "全帯域_40ms_log_rms_mae_db": 0.735751480622535, "全帯域_40ms_rms_envelope_nrmse": 0.07875957951499844, "全帯域_40ms_envelope_correlation": 0.9948404379834713, "全帯域_40ms_modulation_mae_db": 1.2932179910442356, "300–1000 Hz_10ms_log_rms_mae_db": 1.5686792379033754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17015335891935432, "300–1000 Hz_10ms_envelope_correlation": 0.9812428426183804, "300–1000 Hz_10ms_modulation_mae_db": 1.0599486203602047, "300–1000 Hz_40ms_log_rms_mae_db": 0.9165896202221857, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1006142267091745, "300–1000 Hz_40ms_envelope_correlation": 0.9936440212350547, "300–1000 Hz_40ms_modulation_mae_db": 0.9702257931858664, "1000–3000 Hz_10ms_log_rms_mae_db": 1.412899669693094, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17215532380031412, "1000–3000 Hz_10ms_envelope_correlation": 0.9777043146056507, "1000–3000 Hz_10ms_modulation_mae_db": 0.6675152921078915, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8516604655404987, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08489928533261894, "1000–3000 Hz_40ms_envelope_correlation": 0.9946621382892522, "1000–3000 Hz_40ms_modulation_mae_db": 0.8082720987026885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.229225431597447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1409125785459265, "3000–6000 Hz_10ms_envelope_correlation": 0.9871804809155397, "3000–6000 Hz_10ms_modulation_mae_db": 0.3277234719792724, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7374136017587402, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07499689125648074, "3000–6000 Hz_40ms_envelope_correlation": 0.9962417562354456, "3000–6000 Hz_40ms_modulation_mae_db": 0.43711140429555795, "mr_stft_db": 5.9306068898252065, "band_mae_db": 0.23820558973731698, "rms_gain_error_db": 0.19563396324392335, "output_peak": 0.46426932594119463}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60480, "output_samples": 60480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.872813068132651, "spectral_convergence": 0.43542383294949466, "spectral_convergence_status": "ok", "amplitude_floor": 9.224568988143785e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7951110235304535, "spectral_convergence": 0.4736752843080856, "spectral_convergence_status": "ok", "amplitude_floor": 8.440100575569802e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.139839585201748, "spectral_convergence": 0.4894315083441511, "spectral_convergence_status": "ok", "amplitude_floor": 6.303303867538867e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.6236728988818796e-05, "output_power": 5.7108880476674815e-05, "delta_db": 0.9171654424262679, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005640699493062965, "output_power": 0.0005650961926103916, "delta_db": 0.007894178017371202, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007824000374604723, "output_power": 0.0006512190696998023, "delta_db": -0.79701753025898, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028236131111099825, "output_power": 0.00028877768742541763, "delta_db": 0.09758444357412377, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018611455363091946, "output_power": 0.0001823171748278301, "delta_db": -0.08952752624626097, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00035218826443925665, "output_power": 0.00032279188695640134, "delta_db": -0.3785226974139394, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.38128530298949054, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.29648392216375374, "pitch": null, "output_rms_dbfs": -26.845942072016076, "output_peak": 0.32441084522809055, "reference_rms_dbfs": -26.54945814985232, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.935921225621617}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00014780031156712295, "log_rms_mae_db": 1.3556737364705194, "rms_envelope_nrmse": 0.1309863510291073, "envelope_correlation": 0.9852929200723698, "rms_gain_error_db": -0.2965176890433437, "reference_mean_power": 0.0022162793576100603, "output_mean_power": 0.0020700113777982867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 1.2692207138635437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044041649987677594, "output_power": 0.00039566742302226445, "reference_active": true, "delta_db": -0.4653328669384037}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888390879465172, "output_power": 7.620130788432104e-05, "reference_active": true, "delta_db": -1.9316065042330002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.435323106110239e-05, "output_power": 1.551487022509919e-05, "reference_active": true, "delta_db": -1.9580844206779742}], "active_band_count": 3, "active_band_mae_db": 1.451674597283126}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00013486753957748745, "log_rms_mae_db": 0.7669620303561198, "rms_envelope_nrmse": 0.08544505593238866, "envelope_correlation": 0.9939565969286471, "rms_gain_error_db": -0.2965425770865864, "reference_mean_power": 0.0022339157834068563, "output_mean_power": 0.0020864718939366676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.176190391785113e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004109863651179866, "output_power": 0.00036359242800466446, "reference_active": true, "delta_db": -0.53212583720155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3981406650238314e-05, "output_power": 3.348121898140277e-05, "reference_active": true, "delta_db": -2.074429358696212}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0096166790138946e-06, "output_power": 7.091468868272701e-07, "reference_active": true, "delta_db": -1.5342031589547371}], "active_band_count": 3, "active_band_mae_db": 1.380252784950833}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00012936155891805085, "log_rms_mae_db": 1.5620732146497305, "rms_envelope_nrmse": 0.17376019129853226, "envelope_correlation": 0.979397764138133, "rms_gain_error_db": -0.4653594227041747, "reference_mean_power": 0.0012072233727869219, "output_mean_power": 0.0010845552796692505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 8.078028680738071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004104434555454037, "output_power": 0.00036651458618204206, "reference_active": true, "delta_db": -0.4916207236199433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.258149083667088e-05, "output_power": 5.358939023705489e-05, "reference_active": true, "delta_db": -2.374453544488864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4118213823838e-05, "output_power": 1.029662690201924e-05, "reference_active": true, "delta_db": -1.3708477872796767}], "active_band_count": 3, "active_band_mae_db": 1.4123073517961613}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00011150674016193665, "log_rms_mae_db": 0.9403669408002561, "rms_envelope_nrmse": 0.11608239949553695, "envelope_correlation": 0.9914942626665835, "rms_gain_error_db": -0.46536372616090615, "reference_mean_power": 0.0012168889554241998, "output_mean_power": 0.001093237642133404}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.50453906071102e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038524435492848886, "output_power": 0.0003370114608286419, "reference_active": true, "delta_db": -0.5809161334523765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.764413089442773e-05, "output_power": 2.559581519118082e-05, "reference_active": true, "delta_db": -2.6984044334355155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0419619324101079e-06, "output_power": 6.757418238665756e-07, "reference_active": true, "delta_db": -1.8807105284885357}], "active_band_count": 3, "active_band_mae_db": 1.720010365125476}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 7.756656945350889e-05, "log_rms_mae_db": 1.4780734694514643, "rms_envelope_nrmse": 0.18492921561066622, "envelope_correlation": 0.9743409992324904, "rms_gain_error_db": -0.017106151306548904, "reference_mean_power": 0.0003662329710023159, "output_mean_power": 0.00036479327628492243}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 2.3479287649550626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015936395121522886, "output_power": 0.00015794386252932554, "reference_active": true, "delta_db": -0.03887334449839776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.480344306075996e-05, "output_power": 2.0612314986889133e-05, "reference_active": true, "delta_db": -0.8038520058169596}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.158977700048883e-06, "output_power": 9.45740986434593e-06, "reference_active": true, "delta_db": 1.209212012491878}], "active_band_count": 3, "active_band_mae_db": 0.6839791209357452}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 6.756499548428438e-05, "log_rms_mae_db": 0.9134929985549499, "rms_envelope_nrmse": 0.10455214001044962, "envelope_correlation": 0.9911871076274417, "rms_gain_error_db": -0.017085319338710377, "reference_mean_power": 0.00036915132536056565, "output_mean_power": 0.0003677019220954579}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 2.14579734589437e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015227225639161801, "output_power": 0.00015186048637429324, "reference_active": true, "delta_db": -0.011759967307768512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5037342052746137e-05, "output_power": 1.2774697535910565e-05, "reference_active": true, "delta_db": -0.7082045246620904}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0212382710455496e-07, "output_power": 5.247472020787799e-07, "reference_active": true, "delta_db": 1.1559032444536623}], "active_band_count": 3, "active_band_mae_db": 0.625289245474507}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00011110909618382604, "log_rms_mae_db": 1.2182494387661893, "rms_envelope_nrmse": 0.14230203168618008, "envelope_correlation": 0.9877688366558396, "rms_gain_error_db": -0.40025036095357763, "reference_mean_power": 0.00024441441718797436, "output_mean_power": 0.00022289574796379612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 3.469643399885664e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011046778972231385, "output_power": 9.583939084410987e-05, "reference_active": true, "delta_db": -0.61691619906429}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.9627372395634935e-05, "output_power": 5.358505852383355e-05, "reference_active": true, "delta_db": -0.4640196230554613}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.408768678752184e-06, "output_power": 3.4319337558182756e-06, "reference_active": true, "delta_db": -1.087784162874898}], "active_band_count": 3, "active_band_mae_db": 0.722906661664883}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.166729119583493e-05, "log_rms_mae_db": 0.7295060698786525, "rms_envelope_nrmse": 0.08778988989475728, "envelope_correlation": 0.9960953838007501, "rms_gain_error_db": -0.40022781624737747, "reference_mean_power": 0.00024636420339127556, "output_mean_power": 0.00022467503791777756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.3341266023063243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010489106418058692, "output_power": 9.0600939075969e-05, "reference_active": true, "delta_db": -0.6360579245368437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0862352851100365e-05, "output_power": 2.9581462602304875e-05, "reference_active": true, "delta_db": -0.18409389165601284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.747611403563906e-07, "output_power": 3.0366768925234594e-07, "reference_active": true, "delta_db": -0.9135596706891357}], "active_band_count": 3, "active_band_mae_db": 0.5779038289606641}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6395391026018442, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3a3ed663452d7d9e8c49ef28c3b0db208fa75d46d21bd1c29126f9ae5f5137f1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9756326955700381, "standard_deviation": 0.0036733962563044262, "minimum": 0.9703347911251248, "maximum": 0.9807187075220603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4496523060304636, "standard_deviation": 0.04313300590240396, "minimum": 1.4047622416344128, "maximum": 1.520272742486192, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.730654915395428, "standard_deviation": 0.2697998673189456, "minimum": 0.4136872651096191, "maximum": 1.2630768672625836, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18075928445900544, "standard_deviation": 0.013707756963460651, "minimum": 0.16327464472932957, "maximum": 0.20133433233943965, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919454582437434, "standard_deviation": 0.0020824506332317483, "minimum": 0.9891121975750391, "maximum": 0.9946621382892522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8662912041281088, "standard_deviation": 0.02798730520308794, "minimum": 0.8376722047644559, "maximum": 0.9134929985549499, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8307307480686161, "standard_deviation": 0.36466374024006215, "minimum": 0.3337210083753345, "maximum": 1.5673206990092725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10258922025182837, "standard_deviation": 0.01339910870671149, "minimum": 0.08489928533261894, "maximum": 0.1270355021884234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.983178429524578, "standard_deviation": 0.004339146286930673, "minimum": 0.9742074815150046, "maximum": 0.9877688366558396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2047945469603936, "standard_deviation": 0.03274250679289378, "minimum": 1.138513446744298, "maximum": 1.2320237930123235, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.08505665988087, "standard_deviation": 0.7167260320489898, "minimum": 0.3277234719792724, "maximum": 2.319054421860105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16649602711668407, "standard_deviation": 0.026023705445601578, "minimum": 0.1409125785459265, "maximum": 0.20629108074372113, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9946668092236691, "standard_deviation": 0.0023927500108635756, "minimum": 0.9889142393009636, "maximum": 0.9962417562354456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7317844589126123, "standard_deviation": 0.024579044775933458, "minimum": 0.685966552461424, "maximum": 0.771850419544385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0876125983120923, "standard_deviation": 0.809226747398978, "minimum": 0.43711140429555795, "maximum": 2.495307105642615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09615712639564128, "standard_deviation": 0.029375085479495116, "minimum": 0.07499689125648074, "maximum": 0.1666969771702814, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9823451544647237, "standard_deviation": 0.0025631931831792323, "minimum": 0.979397764138133, "maximum": 0.986251760430268, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5637083288637026, "standard_deviation": 0.03375265241527297, "minimum": 1.5200145677832537, "maximum": 1.6140312006848763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9306924917244059, "standard_deviation": 0.4474470204882261, "minimum": 0.37639650377287687, "maximum": 1.6054107824904917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15907775711164424, "standard_deviation": 0.013175384561609618, "minimum": 0.13940224558654715, "maximum": 0.17376019129853226, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9934740127479252, "standard_deviation": 0.0015179594031599671, "minimum": 0.9912475408450823, "maximum": 0.9954428491633427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9140412932992643, "standard_deviation": 0.06561234150234535, "minimum": 0.8024849912948638, "maximum": 0.9810535530035114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9567908730512692, "standard_deviation": 0.4680463941527516, "minimum": 0.47774336653921984, "maximum": 1.720010365125476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09586031338915837, "standard_deviation": 0.012099099996949478, "minimum": 0.08123790540936103, "maximum": 0.11608239949553695, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27969571694630274, "standard_deviation": 0.06781500501113948, "minimum": 0.17792269310660178, "maximum": 0.38128530298949054, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.923763429372315, "standard_deviation": 0.009680053820480235, "minimum": 5.911402232990572, "maximum": 5.935921225621617, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40297303274848695, "standard_deviation": 0.05999160651905114, "minimum": 0.32441084522809055, "maximum": 0.49569729012146635, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05479353311823209, "standard_deviation": 0.16047174784094753, "minimum": -0.29648392216375374, "maximum": 0.19563396324392335, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9855861778917722, "standard_deviation": 0.001786346039288671, "minimum": 0.9833102849977267, "maximum": 0.9890737463243916, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3687652606048084, "standard_deviation": 0.03973321002993787, "minimum": 1.3065736052108166, "maximum": 1.4301040240828429, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7747386228051265, "standard_deviation": 0.4634693312870367, "minimum": 0.15811390498889502, "maximum": 1.451674597283126, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12861140785812195, "standard_deviation": 0.009334930738084236, "minimum": 0.11110373226489131, "maximum": 0.1418256788328344, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9951514213396376, "standard_deviation": 0.0010132758112343903, "minimum": 0.9939565969286471, "maximum": 0.9968396857307725, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7418479423325808, "standard_deviation": 0.04367653388340214, "minimum": 0.6551422128036704, "maximum": 0.7876172424719009, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9453254067631962, "standard_deviation": 0.5978879863444366, "minimum": 0.1650230438777157, "maximum": 1.8963379794775932, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07329277412700254, "standard_deviation": 0.008751068589650503, "minimum": 0.05998489008707448, "maximum": 0.08544505593238866, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0019", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.5388219356536865, "predicted_mos": 2.5388219356536865, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "7a03a131119169ac12e9a4c00b5865b9bfddc1ae42df2b0aff3ce8ce31c0d6d3", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.843492599669844, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.276936397866459, "全帯域_10ms_rms_envelope_nrmse": 0.12823792631873923, "全帯域_10ms_envelope_correlation": 0.9856795392760268, "全帯域_10ms_modulation_mae_db": 0.5518298094343184, "全帯域_40ms_log_rms_mae_db": 0.6461630488085509, "全帯域_40ms_rms_envelope_nrmse": 0.066988949315034, "全帯域_40ms_envelope_correlation": 0.996370623288993, "全帯域_40ms_modulation_mae_db": 0.8563768570082578, "300–1000 Hz_10ms_log_rms_mae_db": 1.704046250809024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16152682325914144, "300–1000 Hz_10ms_envelope_correlation": 0.9816680956262486, "300–1000 Hz_10ms_modulation_mae_db": 0.5433549154958909, "300–1000 Hz_40ms_log_rms_mae_db": 1.0483044114502207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08970449378442216, "300–1000 Hz_40ms_envelope_correlation": 0.9946703525398152, "300–1000 Hz_40ms_modulation_mae_db": 0.5085520410168575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5808503877875997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16661491799359635, "1000–3000 Hz_10ms_envelope_correlation": 0.9800421516147847, "1000–3000 Hz_10ms_modulation_mae_db": 0.7194448823408068, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0167401634185218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09847397799834684, "1000–3000 Hz_40ms_envelope_correlation": 0.993968335452833, "1000–3000 Hz_40ms_modulation_mae_db": 0.88162763546859, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4228515701834974, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20030667519786746, "3000–6000 Hz_10ms_envelope_correlation": 0.9750749028717416, "3000–6000 Hz_10ms_modulation_mae_db": 0.9658615511507089, "3000–6000 Hz_40ms_log_rms_mae_db": 0.920194321904171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11242347006680702, "3000–6000 Hz_40ms_envelope_correlation": 0.9920633443337312, "3000–6000 Hz_40ms_modulation_mae_db": 0.30917069175320855, "mr_stft_db": 6.013893417956962, "band_mae_db": 0.2487567308786406, "rms_gain_error_db": -0.23856593915862967, "output_peak": 0.35237655929870776}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.33418843716527, "全帯域_10ms_rms_envelope_nrmse": 0.1363256553603149, "全帯域_10ms_envelope_correlation": 0.9835192008598181, "全帯域_10ms_modulation_mae_db": 0.7190100373292446, "全帯域_40ms_log_rms_mae_db": 0.7144202297511224, "全帯域_40ms_rms_envelope_nrmse": 0.07955679217666409, "全帯域_40ms_envelope_correlation": 0.9941071652684164, "全帯域_40ms_modulation_mae_db": 0.5439153543325577, "300–1000 Hz_10ms_log_rms_mae_db": 1.7538767671714526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1770436552464381, "300–1000 Hz_10ms_envelope_correlation": 0.9775935332632154, "300–1000 Hz_10ms_modulation_mae_db": 0.9968006090717362, "300–1000 Hz_40ms_log_rms_mae_db": 1.097952729375911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11086609727764489, "300–1000 Hz_40ms_envelope_correlation": 0.9908241943349282, "300–1000 Hz_40ms_modulation_mae_db": 0.647921755054213, "1000–3000 Hz_10ms_log_rms_mae_db": 1.697835618939684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21309905797775724, "1000–3000 Hz_10ms_envelope_correlation": 0.9654681085818292, "1000–3000 Hz_10ms_modulation_mae_db": 0.8393406990793754, "1000–3000 Hz_40ms_log_rms_mae_db": 1.058703560835511, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11757619374753774, "1000–3000 Hz_40ms_envelope_correlation": 0.9889414629648092, "1000–3000 Hz_40ms_modulation_mae_db": 0.7014269853335428, "3000–6000 Hz_10ms_log_rms_mae_db": 1.386151381300217, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1556432371507655, "3000–6000 Hz_10ms_envelope_correlation": 0.9843360420258375, "3000–6000 Hz_10ms_modulation_mae_db": 0.7812551330561317, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8653455980687051, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08819985046088116, "3000–6000 Hz_40ms_envelope_correlation": 0.9948869723169523, "3000–6000 Hz_40ms_modulation_mae_db": 0.516019690794304, "mr_stft_db": 6.037787606782712, "band_mae_db": 0.21883779804984918, "rms_gain_error_db": -0.11785475243994835, "output_peak": 0.36893004309891125}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3330559886727382, "全帯域_10ms_rms_envelope_nrmse": 0.12833828938230057, "全帯域_10ms_envelope_correlation": 0.9853698916948895, "全帯域_10ms_modulation_mae_db": 0.5064845471011835, "全帯域_40ms_log_rms_mae_db": 0.7311467986321161, "全帯域_40ms_rms_envelope_nrmse": 0.08038484130166987, "全帯域_40ms_envelope_correlation": 0.9939316759723553, "全帯域_40ms_modulation_mae_db": 0.7126194171518954, "300–1000 Hz_10ms_log_rms_mae_db": 1.6503817198178357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16811935314131046, "300–1000 Hz_10ms_envelope_correlation": 0.9799249156266391, "300–1000 Hz_10ms_modulation_mae_db": 1.3318826892914006, "300–1000 Hz_40ms_log_rms_mae_db": 1.0058160947166404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09897643367064113, "300–1000 Hz_40ms_envelope_correlation": 0.992603700474462, "300–1000 Hz_40ms_modulation_mae_db": 1.098160465279222, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6471410517513438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18005834826202619, "1000–3000 Hz_10ms_envelope_correlation": 0.9754512346116521, "1000–3000 Hz_10ms_modulation_mae_db": 0.7967656384432066, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0501119755531882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08946854967292175, "1000–3000 Hz_40ms_envelope_correlation": 0.993620949958956, "1000–3000 Hz_40ms_modulation_mae_db": 0.5972560341745827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3620537449929404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17714432392816956, "3000–6000 Hz_10ms_envelope_correlation": 0.9816426260549516, "3000–6000 Hz_10ms_modulation_mae_db": 1.1481353761024249, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8281189409062671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13477581648079207, "3000–6000 Hz_40ms_envelope_correlation": 0.9905038451257996, "3000–6000 Hz_40ms_modulation_mae_db": 1.1291171744212434, "mr_stft_db": 6.017670193279297, "band_mae_db": 0.24564787501770624, "rms_gain_error_db": -0.10201634538849355, "output_peak": 0.45730597783433996}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2561129799776647, "全帯域_10ms_rms_envelope_nrmse": 0.12478101497817286, "全帯域_10ms_envelope_correlation": 0.986422572154264, "全帯域_10ms_modulation_mae_db": 0.36818878338008004, "全帯域_40ms_log_rms_mae_db": 0.6554309599315371, "全帯域_40ms_rms_envelope_nrmse": 0.06606961464578878, "全帯域_40ms_envelope_correlation": 0.9959705028107235, "全帯域_40ms_modulation_mae_db": 0.7388432554581442, "300–1000 Hz_10ms_log_rms_mae_db": 1.7204636053276519, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15402689286900803, "300–1000 Hz_10ms_envelope_correlation": 0.983063046900987, "300–1000 Hz_10ms_modulation_mae_db": 0.5382585702844357, "300–1000 Hz_40ms_log_rms_mae_db": 0.9693220034280463, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08748431030246921, "300–1000 Hz_40ms_envelope_correlation": 0.9942245423287948, "300–1000 Hz_40ms_modulation_mae_db": 0.2581813192784748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5971101335242435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20454484622120772, "1000–3000 Hz_10ms_envelope_correlation": 0.9691413162229473, "1000–3000 Hz_10ms_modulation_mae_db": 0.5560863696728136, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0475225191305053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10782215624185396, "1000–3000 Hz_40ms_envelope_correlation": 0.9908475493587802, "1000–3000 Hz_40ms_modulation_mae_db": 0.5414035636647655, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3598184878304853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16907824066384672, "3000–6000 Hz_10ms_envelope_correlation": 0.9825499816019205, "3000–6000 Hz_10ms_modulation_mae_db": 1.0960308783829016, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8867381726467544, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08767813570345477, "3000–6000 Hz_40ms_envelope_correlation": 0.995443681525742, "3000–6000 Hz_40ms_modulation_mae_db": 1.1412016296882654, "mr_stft_db": 6.017444849617864, "band_mae_db": 0.20059887049614397, "rms_gain_error_db": 0.06505556900622296, "output_peak": 0.43784327003174633}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3543529955094242, "全帯域_10ms_rms_envelope_nrmse": 0.13054853913929187, "全帯域_10ms_envelope_correlation": 0.9848807585140961, "全帯域_10ms_modulation_mae_db": 0.9637766441795422, "全帯域_40ms_log_rms_mae_db": 0.7061561711951676, "全帯域_40ms_rms_envelope_nrmse": 0.07319117502607395, "全帯域_40ms_envelope_correlation": 0.9949219160088862, "全帯域_40ms_modulation_mae_db": 1.3315236256440022, "300–1000 Hz_10ms_log_rms_mae_db": 1.7352142533282717, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16389782659582058, "300–1000 Hz_10ms_envelope_correlation": 0.9807840328283219, "300–1000 Hz_10ms_modulation_mae_db": 1.0457707346138732, "300–1000 Hz_40ms_log_rms_mae_db": 1.043247801117429, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0933467098416668, "300–1000 Hz_40ms_envelope_correlation": 0.9934679932039234, "300–1000 Hz_40ms_modulation_mae_db": 1.0129459655977762, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5832022178135987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17394827132856938, "1000–3000 Hz_10ms_envelope_correlation": 0.9771277451227374, "1000–3000 Hz_10ms_modulation_mae_db": 0.2252085539873929, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9483547483613883, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10687036915449905, "1000–3000 Hz_40ms_envelope_correlation": 0.9908121737001854, "1000–3000 Hz_40ms_modulation_mae_db": 0.24586876366746604, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3555127870838475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1809432396107169, "3000–6000 Hz_10ms_envelope_correlation": 0.9805698931012953, "3000–6000 Hz_10ms_modulation_mae_db": 1.5880589164035523, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9173350046192917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1072332580131952, "3000–6000 Hz_40ms_envelope_correlation": 0.9934132831985432, "3000–6000 Hz_40ms_modulation_mae_db": 1.7897790048017423, "mr_stft_db": 6.0157412076884995, "band_mae_db": 0.15265608082510165, "rms_gain_error_db": -0.024818262933279503, "output_peak": 0.49469049348921373}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2536424650706723, "全帯域_10ms_rms_envelope_nrmse": 0.11849419986347057, "全帯域_10ms_envelope_correlation": 0.9876524362881743, "全帯域_10ms_modulation_mae_db": 0.8080517245838665, "全帯域_40ms_log_rms_mae_db": 0.6836435162355258, "全帯域_40ms_rms_envelope_nrmse": 0.05963212868597295, "全帯域_40ms_envelope_correlation": 0.9968947588635123, "全帯域_40ms_modulation_mae_db": 1.4549537221971895, "300–1000 Hz_10ms_log_rms_mae_db": 1.7339195451588274, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1460623851537947, "300–1000 Hz_10ms_envelope_correlation": 0.9851151515710793, "300–1000 Hz_10ms_modulation_mae_db": 1.0881359510402642, "300–1000 Hz_40ms_log_rms_mae_db": 1.0879021436086898, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08595004729846308, "300–1000 Hz_40ms_envelope_correlation": 0.9950853293703279, "300–1000 Hz_40ms_modulation_mae_db": 0.957440655880507, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5907625494400974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1783360112050872, "1000–3000 Hz_10ms_envelope_correlation": 0.975972877839035, "1000–3000 Hz_10ms_modulation_mae_db": 0.603010375392183, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9390295313089588, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0842941403409014, "1000–3000 Hz_40ms_envelope_correlation": 0.9946499419239323, "1000–3000 Hz_40ms_modulation_mae_db": 0.7204873696646263, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3885086251089187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21575236716602306, "3000–6000 Hz_10ms_envelope_correlation": 0.9712048300841485, "3000–6000 Hz_10ms_modulation_mae_db": 1.8704104351171962, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8539282801538499, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12609241551126846, "3000–6000 Hz_40ms_envelope_correlation": 0.9899894960830075, "3000–6000 Hz_40ms_modulation_mae_db": 2.011007830554112, "mr_stft_db": 6.022793214221598, "band_mae_db": 0.28663803631662643, "rms_gain_error_db": -0.16120709776191597, "output_peak": 0.5533309010913424}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.357217929817281, "全帯域_10ms_rms_envelope_nrmse": 0.14235201243905876, "全帯域_10ms_envelope_correlation": 0.9820052277763666, "全帯域_10ms_modulation_mae_db": 1.1170563514446934, "全帯域_40ms_log_rms_mae_db": 0.6892816816239825, "全帯域_40ms_rms_envelope_nrmse": 0.07291678690522829, "全帯域_40ms_envelope_correlation": 0.9951365828052977, "全帯域_40ms_modulation_mae_db": 1.4473436923673137, "300–1000 Hz_10ms_log_rms_mae_db": 1.7734352047892017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15690223327708164, "300–1000 Hz_10ms_envelope_correlation": 0.9826544349861519, "300–1000 Hz_10ms_modulation_mae_db": 0.9348301780508023, "300–1000 Hz_40ms_log_rms_mae_db": 1.0753435737636683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0967761481588025, "300–1000 Hz_40ms_envelope_correlation": 0.9934661979914733, "300–1000 Hz_40ms_modulation_mae_db": 0.9834304112205651, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5810054463664107, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19930548176080146, "1000–3000 Hz_10ms_envelope_correlation": 0.9706590458634211, "1000–3000 Hz_10ms_modulation_mae_db": 0.268597569125223, "1000–3000 Hz_40ms_log_rms_mae_db": 1.017336668519586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1028056525924894, "1000–3000 Hz_40ms_envelope_correlation": 0.9916802335084511, "1000–3000 Hz_40ms_modulation_mae_db": 0.2682383807265792, "3000–6000 Hz_10ms_log_rms_mae_db": 1.426321658323661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24929826319519555, "3000–6000 Hz_10ms_envelope_correlation": 0.9598408825772358, "3000–6000 Hz_10ms_modulation_mae_db": 2.3094323925749003, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9031193983219804, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11272874368473357, "3000–6000 Hz_40ms_envelope_correlation": 0.9914763485313919, "3000–6000 Hz_40ms_modulation_mae_db": 2.30731239108617, "mr_stft_db": 6.045963215387893, "band_mae_db": 0.2527307717872894, "rms_gain_error_db": -0.14302856364202196, "output_peak": 0.42518756912423133}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.298509774613979, "全帯域_10ms_rms_envelope_nrmse": 0.12342442794137445, "全帯域_10ms_envelope_correlation": 0.9865035404604696, "全帯域_10ms_modulation_mae_db": 0.4609870941418388, "全帯域_40ms_log_rms_mae_db": 0.628852634692707, "全帯域_40ms_rms_envelope_nrmse": 0.0644712157971489, "全帯域_40ms_envelope_correlation": 0.9960587008291835, "全帯域_40ms_modulation_mae_db": 0.8410321052328231, "300–1000 Hz_10ms_log_rms_mae_db": 1.687346555502877, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15329932131583493, "300–1000 Hz_10ms_envelope_correlation": 0.9832475400550831, "300–1000 Hz_10ms_modulation_mae_db": 0.7703561447497181, "300–1000 Hz_40ms_log_rms_mae_db": 1.0081964332582454, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08443552379444977, "300–1000 Hz_40ms_envelope_correlation": 0.994620263884937, "300–1000 Hz_40ms_modulation_mae_db": 0.5169664853196779, "1000–3000 Hz_10ms_log_rms_mae_db": 1.619721338306258, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18369469225817184, "1000–3000 Hz_10ms_envelope_correlation": 0.9744301275560193, "1000–3000 Hz_10ms_modulation_mae_db": 0.3775226740233588, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9999654482467301, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10732670693128324, "1000–3000 Hz_40ms_envelope_correlation": 0.9907697986910958, "1000–3000 Hz_40ms_modulation_mae_db": 0.5751779480817188, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3619811294123165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17353072488117222, "3000–6000 Hz_10ms_envelope_correlation": 0.981216451859376, "3000–6000 Hz_10ms_modulation_mae_db": 1.3152547476169543, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7809452728957368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1019147030466279, "3000–6000 Hz_40ms_envelope_correlation": 0.9944057450399884, "3000–6000 Hz_40ms_modulation_mae_db": 1.2704455035288273, "mr_stft_db": 6.025142483389899, "band_mae_db": 0.18456883453253534, "rms_gain_error_db": -0.029848504033296912, "output_peak": 0.42641357183971024}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60480, "output_samples": 60480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.003972272858592, "spectral_convergence": 0.44741130528928885, "spectral_convergence_status": "ok", "amplitude_floor": 9.224568988143785e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.8963798469816435, "spectral_convergence": 0.4807609411413532, "spectral_convergence_status": "ok", "amplitude_floor": 8.440100575569802e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.141328134030651, "spectral_convergence": 0.482237746636202, "spectral_convergence_status": "ok", "amplitude_floor": 6.303303867538867e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.6236728988818796e-05, "output_power": 5.023392016209281e-05, "delta_db": 0.3600996862463378, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005640699493062965, "output_power": 0.0005414816689055986, "delta_db": -0.17749204619352604, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007824000374604723, "output_power": 0.0007200792720005807, "delta_db": -0.3604855295108983, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028236131111099825, "output_power": 0.00025421215029347084, "delta_db": -0.45608885621766804, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018611455363091946, "output_power": 0.00018400665922083928, "delta_db": -0.049467945989037435, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00035218826443925665, "output_power": 0.00034505176105146, "delta_db": -0.08890632111437599, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2487567308786406, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.23856593915862967, "pitch": null, "output_rms_dbfs": -26.78802408901095, "output_peak": 0.35237655929870776, "reference_rms_dbfs": -26.54945814985232, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.013893417956962}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00014780031156712295, "log_rms_mae_db": 1.276936397866459, "rms_envelope_nrmse": 0.12823792631873923, "envelope_correlation": 0.9856795392760268, "rms_gain_error_db": -0.23862075002753116, "reference_mean_power": 0.0022162793576100603, "output_mean_power": 0.002097792007624829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 1.2692207138635437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044041649987677594, "output_power": 0.0004092942872651365, "reference_active": true, "delta_db": -0.31827898112536307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888390879465172, "output_power": 9.001341221764688e-05, "reference_active": true, "delta_db": -1.2081585066192582}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.435323106110239e-05, "output_power": 2.508775405000991e-05, "reference_active": true, "delta_db": 0.129051940558334}], "active_band_count": 3, "active_band_mae_db": 0.5518298094343184}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00013486753957748745, "log_rms_mae_db": 0.6461630488085509, "rms_envelope_nrmse": 0.066988949315034, "envelope_correlation": 0.996370623288993, "rms_gain_error_db": -0.2387583556523275, "reference_mean_power": 0.0022339157834068563, "output_mean_power": 0.0021144185537382986}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.176190391785113e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004109863651179866, "output_power": 0.00038006971462076547, "reference_active": true, "delta_db": -0.3396414919316414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3981406650238314e-05, "output_power": 3.974965716264706e-05, "reference_active": true, "delta_db": -1.3291080997931084}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0096166790138946e-06, "output_power": 1.2422088370898472e-06, "reference_active": true, "delta_db": 0.9003809793000239}], "active_band_count": 3, "active_band_mae_db": 0.8563768570082578}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00012936155891805085, "log_rms_mae_db": 1.704046250809024, "rms_envelope_nrmse": 0.16152682325914144, "envelope_correlation": 0.9816680956262486, "rms_gain_error_db": -0.3066067344941117, "reference_mean_power": 0.0012072233727869219, "output_mean_power": 0.0011249337891255688}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 8.078028680738071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004104434555454037, "output_power": 0.00037524838680081036, "reference_active": true, "delta_db": -0.3893450142184999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.258149083667088e-05, "output_power": 7.556949371171079e-05, "reference_active": true, "delta_db": -0.8817765732122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4118213823838e-05, "output_power": 1.5334657166852748e-05, "reference_active": true, "delta_db": 0.35894315905697244}], "active_band_count": 3, "active_band_mae_db": 0.5433549154958909}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00011150674016193665, "log_rms_mae_db": 1.0483044114502207, "rms_envelope_nrmse": 0.08970449378442216, "envelope_correlation": 0.9946703525398152, "rms_gain_error_db": -0.3066095027670576, "reference_mean_power": 0.0012168889554241998, "output_mean_power": 0.001133939800913558}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.50453906071102e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038524435492848886, "output_power": 0.00035306190094103683, "reference_active": true, "delta_db": -0.37885428432702434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.764413089442773e-05, "output_power": 3.713785691793512e-05, "reference_active": true, "delta_db": -1.0819257045419095}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0419619324101079e-06, "output_power": 1.0265124926277486e-06, "reference_active": true, "delta_db": -0.0648761341816385}], "active_band_count": 3, "active_band_mae_db": 0.5085520410168575}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 7.756656945350889e-05, "log_rms_mae_db": 1.5808503877875997, "rms_envelope_nrmse": 0.16661491799359635, "envelope_correlation": 0.9800421516147847, "rms_gain_error_db": -0.44413461721189756, "reference_mean_power": 0.0003662329710023159, "output_mean_power": 0.0003306313104329374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 2.3479287649550626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015936395121522886, "output_power": 0.00013512972851561188, "reference_active": true, "delta_db": -0.7163918472152252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.480344306075996e-05, "output_power": 1.8684930212846163e-05, "reference_active": true, "delta_db": -1.230204909648596}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.158977700048883e-06, "output_power": 6.818317499766657e-06, "reference_active": true, "delta_db": -0.21173789015859923}], "active_band_count": 3, "active_band_mae_db": 0.7194448823408068}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 6.756499548428438e-05, "log_rms_mae_db": 1.0167401634185218, "rms_envelope_nrmse": 0.09847397799834684, "envelope_correlation": 0.993968335452833, "rms_gain_error_db": -0.44416761187289366, "reference_mean_power": 0.00036915132536056565, "output_mean_power": 0.0003332634384287546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 2.14579734589437e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015227225639161801, "output_power": 0.00012916440303020586, "reference_active": true, "delta_db": -0.7147794214378553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5037342052746137e-05, "output_power": 1.1575705716346296e-05, "reference_active": true, "delta_db": -1.1362360129711837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0212382710455496e-07, "output_power": 3.349446613327751e-07, "reference_active": true, "delta_db": -0.7938674719967308}], "active_band_count": 3, "active_band_mae_db": 0.88162763546859}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00011110909618382604, "log_rms_mae_db": 1.4228515701834974, "rms_envelope_nrmse": 0.20030667519786746, "envelope_correlation": 0.9750749028717416, "rms_gain_error_db": 0.19391212562814564, "reference_mean_power": 0.00024441441718797436, "output_mean_power": 0.0002555748016809198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 3.469643399885664e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011046778972231385, "output_power": 0.00011468809508780106, "reference_active": true, "delta_db": 0.16282674801885239}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.9627372395634935e-05, "output_power": 5.5894350396480315e-05, "reference_active": true, "delta_db": -0.2807775852400871}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.408768678752184e-06, "output_power": 7.757384975296732e-06, "reference_active": true, "delta_db": 2.453980320193187}], "active_band_count": 3, "active_band_mae_db": 0.9658615511507089}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.166729119583493e-05, "log_rms_mae_db": 0.920194321904171, "rms_envelope_nrmse": 0.11242347006680702, "envelope_correlation": 0.9920633443337312, "rms_gain_error_db": 0.19392295209976843, "reference_mean_power": 0.00024636420339127556, "output_mean_power": 0.00025761426069284267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.3341266023063243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010489106418058692, "output_power": 0.0001098307243100276, "reference_active": true, "delta_db": 0.19985356086549766}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0862352851100365e-05, "output_power": 3.0829488665674324e-05, "reference_active": true, "delta_db": -0.004627106399158679}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.747611403563906e-07, "output_power": 4.426471706348176e-07, "reference_active": true, "delta_db": 0.7230314079949693}], "active_band_count": 3, "active_band_mae_db": 0.30917069175320855}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6736589003855598, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4a87ec9af3971582e53c378b7ae38c256bfebe128adca7e0dce0bfd1bd3d87f4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9735365759265533, "standard_deviation": 0.004755712100109384, "minimum": 0.9654681085818292, "maximum": 0.9800421516147847, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6122035929911545, "standard_deviation": 0.041567406266508004, "minimum": 1.5808503877875997, "maximum": 1.697835618939684, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5482470952580449, "standard_deviation": 0.23621137340692513, "minimum": 0.2252085539873929, "maximum": 0.8393406990793754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18745020337590218, "standard_deviation": 0.01629575471735995, "minimum": 0.16661491799359635, "maximum": 0.21309905797775724, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919113056948804, "standard_deviation": 0.001970150481946841, "minimum": 0.9889414629648092, "maximum": 0.9946499419239323, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0097205769217987, "standard_deviation": 0.04539543000482618, "minimum": 0.9390295313089588, "maximum": 1.058703560835511, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.566435835097734, "standard_deviation": 0.21864460722103782, "minimum": 0.24586876366746604, "maximum": 0.88162763546859, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10182971833497917, "standard_deviation": 0.010770291622649402, "minimum": 0.0842941403409014, "maximum": 0.11757619374753774, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9770544512720634, "standard_deviation": 0.008180097559814062, "minimum": 0.9598408825772358, "maximum": 0.9843360420258375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3828999230294854, "standard_deviation": 0.02848201853164257, "minimum": 1.3555127870838475, "maximum": 1.426321658323661, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3843049288005962, "standard_deviation": 0.509353217984584, "minimum": 0.7812551330561317, "maximum": 2.3094323925749003, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19021213397421963, "standard_deviation": 0.030289020844008224, "minimum": 0.1556432371507655, "maximum": 0.24929826319519555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927728395193944, "standard_deviation": 0.002061868486779577, "minimum": 0.9899894960830075, "maximum": 0.995443681525742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8694656236895946, "standard_deviation": 0.0479576901098997, "minimum": 0.7809452728957368, "maximum": 0.920194321904171, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3092567395784842, "standard_deviation": 0.6980654429129506, "minimum": 0.30917069175320855, "maximum": 2.30731239108617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10888079912097001, "standard_deviation": 0.016572974202102967, "minimum": 0.08767813570345477, "maximum": 0.13477581648079207, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9817563438572159, "standard_deviation": 0.0023193470070003476, "minimum": 0.9775935332632154, "maximum": 0.9851151515710793, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7198354877381428, "standard_deviation": 0.03890585084266376, "minimum": 1.6503817198178357, "maximum": 1.7734352047892017, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9061737240747652, "standard_deviation": 0.2747220902759535, "minimum": 0.5382585702844357, "maximum": 1.3318826892914006, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16010981135730373, "standard_deviation": 0.009697335588798586, "minimum": 0.1460623851537947, "maximum": 0.1770436552464381, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936203217660826, "standard_deviation": 0.0013893772616497204, "minimum": 0.9908241943349282, "maximum": 0.9950853293703279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0420106488398564, "standard_deviation": 0.04492033844124544, "minimum": 0.9693220034280463, "maximum": 1.097952729375911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7479498873309116, "standard_deviation": 0.3053689496718784, "minimum": 0.2581813192784748, "maximum": 1.098160465279222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09344247051606994, "standard_deviation": 0.008718320058936264, "minimum": 0.08443552379444977, "maximum": 0.11086609727764489, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2238043747379866, "standard_deviation": 0.04317610732820753, "minimum": 0.15265608082510165, "maximum": 0.28663803631662643, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.024554523540591, "standard_deviation": 0.011502606294819067, "minimum": 6.013893417956962, "maximum": 6.045963215387893, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.43950979822602537, "standard_deviation": 0.06466393835367368, "minimum": 0.35237655929870776, "maximum": 0.5533309010913424, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09403548704392037, "standard_deviation": 0.09460701193443842, "minimum": -0.23856593915862967, "maximum": 0.06505556900622296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9852541458780131, "standard_deviation": 0.001795919242746637, "minimum": 0.9820052277763666, "maximum": 0.9876524362881743, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3080021210866861, "standard_deviation": 0.042403938138389975, "minimum": 1.2536424650706723, "maximum": 1.357217929817281, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6869231239493458, "standard_deviation": 0.26208262765403156, "minimum": 0.36818878338008004, "maximum": 1.1170563514446934, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1290627581778404, "standard_deviation": 0.007507950257912308, "minimum": 0.11849419986347057, "maximum": 0.14235201243905876, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.995423990730921, "standard_deviation": 0.0010728146852093104, "minimum": 0.9939316759723553, "maximum": 0.9968947588635123, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6818868801088387, "standard_deviation": 0.03569806874265566, "minimum": 0.628852634692707, "maximum": 0.7311467986321161, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9908260036740231, "standard_deviation": 0.362734836748018, "minimum": 0.5439153543325577, "maximum": 1.4549537221971895, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0704014379816976, "standard_deviation": 0.007359675802711528, "minimum": 0.05963212868597295, "maximum": 0.08038484130166987, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1221-135767-0019", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.2385027408599854, "predicted_mos": 2.2385027408599854, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "2ff4c340d275d0a925ff1162f9c06516cc8cf8ddb027dbc9ce3229f61116ab59", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.979998900089413, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2690893856237855, "全帯域_10ms_rms_envelope_nrmse": 0.11240926680045885, "全帯域_10ms_envelope_correlation": 0.9888254068074638, "全帯域_10ms_modulation_mae_db": 0.2044414228627808, "全帯域_40ms_log_rms_mae_db": 0.5959643213282626, "全帯域_40ms_rms_envelope_nrmse": 0.057699605246583845, "全帯域_40ms_envelope_correlation": 0.9968451742303649, "全帯域_40ms_modulation_mae_db": 0.7153585753644235, "300–1000 Hz_10ms_log_rms_mae_db": 1.6332181768712128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14639193669833953, "300–1000 Hz_10ms_envelope_correlation": 0.984807750688354, "300–1000 Hz_10ms_modulation_mae_db": 0.6168399404787573, "300–1000 Hz_40ms_log_rms_mae_db": 0.9287245959114976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07217889223586892, "300–1000 Hz_40ms_envelope_correlation": 0.9960863892732559, "300–1000 Hz_40ms_modulation_mae_db": 0.24776695175515084, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5509446622128877, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18062668507327534, "1000–3000 Hz_10ms_envelope_correlation": 0.9756203530415829, "1000–3000 Hz_10ms_modulation_mae_db": 0.3482771239028182, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9185131255305558, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1005893036624681, "1000–3000 Hz_40ms_envelope_correlation": 0.992648425027901, "1000–3000 Hz_40ms_modulation_mae_db": 0.3810180455567238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3057373607571636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14238196389410412, "3000–6000 Hz_10ms_envelope_correlation": 0.9872043656438627, "3000–6000 Hz_10ms_modulation_mae_db": 0.3461980387773666, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8069319456623342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0670783391906714, "3000–6000 Hz_40ms_envelope_correlation": 0.9971118937275958, "3000–6000 Hz_40ms_modulation_mae_db": 0.4276139469058129, "mr_stft_db": 5.980086797083762, "band_mae_db": 0.17629020275171392, "rms_gain_error_db": -0.015710593747125204, "output_peak": 0.3674078103278427}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3132518058206517, "全帯域_10ms_rms_envelope_nrmse": 0.122431440327305, "全帯域_10ms_envelope_correlation": 0.9868286610357327, "全帯域_10ms_modulation_mae_db": 1.0252391735061916, "全帯域_40ms_log_rms_mae_db": 0.6995708893144145, "全帯域_40ms_rms_envelope_nrmse": 0.0605058630596822, "全帯域_40ms_envelope_correlation": 0.9965527089632525, "全帯域_40ms_modulation_mae_db": 1.3332914690370916, "300–1000 Hz_10ms_log_rms_mae_db": 1.6275862122668212, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16303399058418858, "300–1000 Hz_10ms_envelope_correlation": 0.9811573160881238, "300–1000 Hz_10ms_modulation_mae_db": 1.2762402220494713, "300–1000 Hz_40ms_log_rms_mae_db": 0.9656098808874074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0867833450898848, "300–1000 Hz_40ms_envelope_correlation": 0.9943348856267717, "300–1000 Hz_40ms_modulation_mae_db": 0.9085352279879872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5159901688959816, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16271123978401322, "1000–3000 Hz_10ms_envelope_correlation": 0.9801239078734746, "1000–3000 Hz_10ms_modulation_mae_db": 1.0453457230346952, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8549216419603471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07769679837454868, "1000–3000 Hz_40ms_envelope_correlation": 0.9951363220047709, "1000–3000 Hz_40ms_modulation_mae_db": 1.039854986574584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3539295396440112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17520041547434115, "3000–6000 Hz_10ms_envelope_correlation": 0.980363933745369, "3000–6000 Hz_10ms_modulation_mae_db": 1.8122582418685205, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7510494366708991, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08464277292895409, "3000–6000 Hz_40ms_envelope_correlation": 0.9952527408977039, "3000–6000 Hz_40ms_modulation_mae_db": 1.2366131414295853, "mr_stft_db": 5.974945504769869, "band_mae_db": 0.1547110492441521, "rms_gain_error_db": 0.017728132762175615, "output_peak": 0.45854943872645315}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2731353819493456, "全帯域_10ms_rms_envelope_nrmse": 0.12939266532474822, "全帯域_10ms_envelope_correlation": 0.9852724754468873, "全帯域_10ms_modulation_mae_db": 0.5024205415199874, "全帯域_40ms_log_rms_mae_db": 0.6661172319055143, "全帯域_40ms_rms_envelope_nrmse": 0.06947083005179358, "全帯域_40ms_envelope_correlation": 0.9954658078321011, "全帯域_40ms_modulation_mae_db": 1.0670901553338072, "300–1000 Hz_10ms_log_rms_mae_db": 1.6264346974271238, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16085663661224553, "300–1000 Hz_10ms_envelope_correlation": 0.981567791835641, "300–1000 Hz_10ms_modulation_mae_db": 0.4518507968701719, "300–1000 Hz_40ms_log_rms_mae_db": 0.9524892713045342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09355335432067209, "300–1000 Hz_40ms_envelope_correlation": 0.9934192949448908, "300–1000 Hz_40ms_modulation_mae_db": 0.6997575202276222, "1000–3000 Hz_10ms_log_rms_mae_db": 1.533653550055669, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.186923372758684, "1000–3000 Hz_10ms_envelope_correlation": 0.973614135729559, "1000–3000 Hz_10ms_modulation_mae_db": 0.36141281159501504, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9168929330916786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08725205421142605, "1000–3000 Hz_40ms_envelope_correlation": 0.9938683754554711, "1000–3000 Hz_40ms_modulation_mae_db": 0.44481065109641565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.308548563693492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17047524065142902, "3000–6000 Hz_10ms_envelope_correlation": 0.9813232339926635, "3000–6000 Hz_10ms_modulation_mae_db": 1.405474238447141, "3000–6000 Hz_40ms_log_rms_mae_db": 0.799725289484462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09128477505050545, "3000–6000 Hz_40ms_envelope_correlation": 0.9944235626826414, "3000–6000 Hz_40ms_modulation_mae_db": 1.5117361299452652, "mr_stft_db": 5.975694772944962, "band_mae_db": 0.19188311012019932, "rms_gain_error_db": -0.01525959223724522, "output_peak": 0.4272471303190007}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3217156324859254, "全帯域_10ms_rms_envelope_nrmse": 0.12683506431575908, "全帯域_10ms_envelope_correlation": 0.9859903043510376, "全帯域_10ms_modulation_mae_db": 0.5052548016919359, "全帯域_40ms_log_rms_mae_db": 0.6810203034062657, "全帯域_40ms_rms_envelope_nrmse": 0.06813117443265167, "全帯域_40ms_envelope_correlation": 0.9957107478728408, "全帯域_40ms_modulation_mae_db": 0.7816192485690046, "300–1000 Hz_10ms_log_rms_mae_db": 1.6741349743095246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16441655580187559, "300–1000 Hz_10ms_envelope_correlation": 0.9808942411361598, "300–1000 Hz_10ms_modulation_mae_db": 0.9393197672363508, "300–1000 Hz_40ms_log_rms_mae_db": 0.9659932767561702, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0920250425321192, "300–1000 Hz_40ms_envelope_correlation": 0.9936569523480197, "300–1000 Hz_40ms_modulation_mae_db": 0.9361896686636807, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5449004247935791, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1696629432992214, "1000–3000 Hz_10ms_envelope_correlation": 0.9783932351008402, "1000–3000 Hz_10ms_modulation_mae_db": 0.5413998904878824, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8889465320059984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08807751375525595, "1000–3000 Hz_40ms_envelope_correlation": 0.9937419089857411, "1000–3000 Hz_40ms_modulation_mae_db": 0.4138238732969653, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3280161262674826, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14838614228660305, "3000–6000 Hz_10ms_envelope_correlation": 0.9860322547118062, "3000–6000 Hz_10ms_modulation_mae_db": 0.2952857072327672, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7863285147276353, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07591745642908511, "3000–6000 Hz_40ms_envelope_correlation": 0.9962356755647245, "3000–6000 Hz_40ms_modulation_mae_db": 0.24124362680569397, "mr_stft_db": 5.987694162993449, "band_mae_db": 0.16801885029394134, "rms_gain_error_db": 0.07338326372921494, "output_peak": 0.47447913839188355}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2784252828919105, "全帯域_10ms_rms_envelope_nrmse": 0.11800985869107805, "全帯域_10ms_envelope_correlation": 0.9876797031734957, "全帯域_10ms_modulation_mae_db": 0.6024409908363696, "全帯域_40ms_log_rms_mae_db": 0.6975575758434186, "全帯域_40ms_rms_envelope_nrmse": 0.06810167916086132, "全帯域_40ms_envelope_correlation": 0.9957386728717996, "全帯域_40ms_modulation_mae_db": 0.4736191145477974, "300–1000 Hz_10ms_log_rms_mae_db": 1.5094341648221858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14514213979375698, "300–1000 Hz_10ms_envelope_correlation": 0.9849614691614446, "300–1000 Hz_10ms_modulation_mae_db": 0.3676804567226797, "300–1000 Hz_40ms_log_rms_mae_db": 0.8649003579063339, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09014517247902416, "300–1000 Hz_40ms_envelope_correlation": 0.993884980709958, "300–1000 Hz_40ms_modulation_mae_db": 0.5057519812795986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.583290005518686, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1834688707844436, "1000–3000 Hz_10ms_envelope_correlation": 0.9745800377101758, "1000–3000 Hz_10ms_modulation_mae_db": 0.4069944736601598, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8992378318594095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09595616941394096, "1000–3000 Hz_40ms_envelope_correlation": 0.992582995667081, "1000–3000 Hz_40ms_modulation_mae_db": 0.6983129338101747, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3073814993771196, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1789127208772901, "3000–6000 Hz_10ms_envelope_correlation": 0.9809798621067456, "3000–6000 Hz_10ms_modulation_mae_db": 0.947337205114, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7646716773485195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1208903085526117, "3000–6000 Hz_40ms_envelope_correlation": 0.9927753096386781, "3000–6000 Hz_40ms_modulation_mae_db": 0.881684691470123, "mr_stft_db": 5.971916206350104, "band_mae_db": 0.3067814522727419, "rms_gain_error_db": -0.12553586503487774, "output_peak": 0.4263039092626827}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2733469317917574, "全帯域_10ms_rms_envelope_nrmse": 0.11019682033336402, "全帯域_10ms_envelope_correlation": 0.9892447209820697, "全帯域_10ms_modulation_mae_db": 0.922458283799351, "全帯域_40ms_log_rms_mae_db": 0.6985131394684132, "全帯域_40ms_rms_envelope_nrmse": 0.05363931209431537, "全帯域_40ms_envelope_correlation": 0.9972862059139221, "全帯域_40ms_modulation_mae_db": 0.8908354753163817, "300–1000 Hz_10ms_log_rms_mae_db": 1.5846778647915498, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1399514544718812, "300–1000 Hz_10ms_envelope_correlation": 0.9860899049110748, "300–1000 Hz_10ms_modulation_mae_db": 0.8663050638600881, "300–1000 Hz_40ms_log_rms_mae_db": 0.9260339612637793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08145637759510707, "300–1000 Hz_40ms_envelope_correlation": 0.9949953682544199, "300–1000 Hz_40ms_modulation_mae_db": 0.58675514462917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.513156973280376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.175360963624885, "1000–3000 Hz_10ms_envelope_correlation": 0.9767184347321393, "1000–3000 Hz_10ms_modulation_mae_db": 0.23854056332209672, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9033794843463283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09135728655667041, "1000–3000 Hz_40ms_envelope_correlation": 0.9933812052531038, "1000–3000 Hz_40ms_modulation_mae_db": 0.26759482630003956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3442256226401332, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2275336051671659, "3000–6000 Hz_10ms_envelope_correlation": 0.9663064500206863, "3000–6000 Hz_10ms_modulation_mae_db": 2.2346249279164208, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8062457516103771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09623196092947289, "3000–6000 Hz_40ms_envelope_correlation": 0.9939174555493124, "3000–6000 Hz_40ms_modulation_mae_db": 2.0738161787225127, "mr_stft_db": 5.989134144277899, "band_mae_db": 0.25759796232238963, "rms_gain_error_db": -0.04157703239606386, "output_peak": 0.38545461725884705}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.337393878142992, "全帯域_10ms_rms_envelope_nrmse": 0.13449625869173296, "全帯域_10ms_envelope_correlation": 0.9841670419808527, "全帯域_10ms_modulation_mae_db": 0.711025997139413, "全帯域_40ms_log_rms_mae_db": 0.7524209905605043, "全帯域_40ms_rms_envelope_nrmse": 0.07924695336275146, "全帯域_40ms_envelope_correlation": 0.9945623943981152, "全帯域_40ms_modulation_mae_db": 0.666495395326162, "300–1000 Hz_10ms_log_rms_mae_db": 1.6235602707215664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16574998267365945, "300–1000 Hz_10ms_envelope_correlation": 0.9809696331215172, "300–1000 Hz_10ms_modulation_mae_db": 0.4967604631059888, "300–1000 Hz_40ms_log_rms_mae_db": 0.9622656678638983, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10847494779407821, "300–1000 Hz_40ms_envelope_correlation": 0.9922005861791057, "300–1000 Hz_40ms_modulation_mae_db": 0.5525335154481853, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5514942552380224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19583069558717192, "1000–3000 Hz_10ms_envelope_correlation": 0.9718864723764636, "1000–3000 Hz_10ms_modulation_mae_db": 1.0358956272944382, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9233723824377397, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1068949971522815, "1000–3000 Hz_40ms_envelope_correlation": 0.9910910780504414, "1000–3000 Hz_40ms_modulation_mae_db": 1.2055911895198685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3112230946931926, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1561617570350674, "3000–6000 Hz_10ms_envelope_correlation": 0.9845454751977649, "3000–6000 Hz_10ms_modulation_mae_db": 1.1184482407053864, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7811703051502916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10346265530251911, "3000–6000 Hz_40ms_envelope_correlation": 0.993369390690836, "3000–6000 Hz_40ms_modulation_mae_db": 0.7470288152745748, "mr_stft_db": 5.9975004012474455, "band_mae_db": 0.40524945741985413, "rms_gain_error_db": -0.23196606290954516, "output_peak": 0.3390502621841816}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2574560539243558, "全帯域_10ms_rms_envelope_nrmse": 0.09839123534351235, "全帯域_10ms_envelope_correlation": 0.9914806113097726, "全帯域_10ms_modulation_mae_db": 0.36226634126776797, "全帯域_40ms_log_rms_mae_db": 0.6569203691274907, "全帯域_40ms_rms_envelope_nrmse": 0.04999646381112975, "全帯域_40ms_envelope_correlation": 0.9976453868674409, "全帯域_40ms_modulation_mae_db": 0.09951178302249035, "300–1000 Hz_10ms_log_rms_mae_db": 1.5094152471084124, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12627918415608294, "300–1000 Hz_10ms_envelope_correlation": 0.9886485370789287, "300–1000 Hz_10ms_modulation_mae_db": 0.11846223095215348, "300–1000 Hz_40ms_log_rms_mae_db": 0.8583508663749679, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0658528826472017, "300–1000 Hz_40ms_envelope_correlation": 0.9967700685774867, "300–1000 Hz_40ms_modulation_mae_db": 0.0901386816612872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4909583875851866, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17445812391872084, "1000–3000 Hz_10ms_envelope_correlation": 0.9783055472024919, "1000–3000 Hz_10ms_modulation_mae_db": 0.46252109594367613, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8302046417424633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08108767566777901, "1000–3000 Hz_40ms_envelope_correlation": 0.9953811983995423, "1000–3000 Hz_40ms_modulation_mae_db": 0.5878148095032335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3212329668471625, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14350047672688807, "3000–6000 Hz_10ms_envelope_correlation": 0.9867600570801321, "3000–6000 Hz_10ms_modulation_mae_db": 0.5815463543757188, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7858062292408766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09041342436910618, "3000–6000 Hz_40ms_envelope_correlation": 0.9945472810653755, "3000–6000 Hz_40ms_modulation_mae_db": 0.3959441188087281, "mr_stft_db": 5.986776003466026, "band_mae_db": 0.2883856038002461, "rms_gain_error_db": 0.011355310462605447, "output_peak": 0.3792434718867608}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60480, "output_samples": 60480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.996124518024049, "spectral_convergence": 0.4424403740940435, "spectral_convergence_status": "ok", "amplitude_floor": 9.224568988143785e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.831494539875058, "spectral_convergence": 0.48353949259621126, "spectral_convergence_status": "ok", "amplitude_floor": 8.440100575569802e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.112641333352178, "spectral_convergence": 0.4887978242867085, "spectral_convergence_status": "ok", "amplitude_floor": 6.303303867538867e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.6236728988818796e-05, "output_power": 5.162448304748647e-05, "delta_db": 0.4786861339355517, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005640699493062965, "output_power": 0.0005812039634712894, "delta_db": 0.12995603865145167, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007824000374604723, "output_power": 0.0007667027587157983, "delta_db": -0.08801836654945319, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028236131111099825, "output_power": 0.0002640719911616462, "delta_db": -0.29082849687528123, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018611455363091946, "output_power": 0.00018864519945718057, "delta_db": 0.05865422925389451, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00035218826443925665, "output_power": 0.0003531300496677413, "delta_db": 0.011597951244651333, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17629020275171392, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.015710593747125204, "pitch": null, "output_rms_dbfs": -26.565168743599447, "output_peak": 0.3674078103278427, "reference_rms_dbfs": -26.54945814985232, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.980086797083762}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00014780031156712295, "log_rms_mae_db": 1.2690893856237855, "rms_envelope_nrmse": 0.11240926680045885, "envelope_correlation": 0.9888254068074638, "rms_gain_error_db": -0.015690794986430695, "reference_mean_power": 0.0022162793576100603, "output_mean_power": 0.002208286522850657}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 1.2692207138635437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044041649987677594, "output_power": 0.0004256339179520499, "reference_active": true, "delta_db": -0.14827352362624616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888390879465172, "output_power": 0.00011905052753978838, "reference_active": true, "delta_db": 0.006082483438506809}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.435323106110239e-05, "output_power": 2.706782845757141e-05, "reference_active": true, "delta_db": 0.4589682615235895}], "active_band_count": 3, "active_band_mae_db": 0.2044414228627808}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00013486753957748745, "log_rms_mae_db": 0.5959643213282626, "rms_envelope_nrmse": 0.057699605246583845, "envelope_correlation": 0.9968451742303649, "rms_gain_error_db": -0.015639166158973008, "reference_mean_power": 0.0022339157834068563, "output_mean_power": 0.0022258858054059207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.176190391785113e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004109863651179866, "output_power": 0.0003994746613301262, "reference_active": true, "delta_db": -0.1233817672983252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3981406650238314e-05, "output_power": 5.9658968942477614e-05, "reference_active": true, "delta_db": 0.4343154615075544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0096166790138946e-06, "output_power": 1.4554402003681766e-06, "reference_active": true, "delta_db": 1.588378497287391}], "active_band_count": 3, "active_band_mae_db": 0.7153585753644235}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00012936155891805085, "log_rms_mae_db": 1.6332181768712128, "rms_envelope_nrmse": 0.14639193669833953, "envelope_correlation": 0.984807750688354, "rms_gain_error_db": 0.014968190137004468, "reference_mean_power": 0.0012072233727869219, "output_mean_power": 0.0012113913106864946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 8.078028680738071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004104434555454037, "output_power": 0.00039808932664562766, "reference_active": true, "delta_db": -0.13272801628914319}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.258149083667088e-05, "output_power": 8.491207826595851e-05, "reference_active": true, "delta_db": -0.3755469934522766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4118213823838e-05, "output_power": 1.9231104875816577e-05, "reference_active": true, "delta_db": 1.3422448116948524}], "active_band_count": 3, "active_band_mae_db": 0.6168399404787573}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00011150674016193665, "log_rms_mae_db": 0.9287245959114976, "rms_envelope_nrmse": 0.07217889223586892, "envelope_correlation": 0.9960863892732559, "rms_gain_error_db": 0.014970320845950598, "reference_mean_power": 0.0012168889554241998, "output_mean_power": 0.0012210908628254244}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.50453906071102e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038524435492848886, "output_power": 0.00037512748721220624, "reference_active": true, "delta_db": -0.11557395672828419}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.764413089442773e-05, "output_power": 4.4750443757572053e-05, "reference_active": true, "delta_db": -0.2721206287093525}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0419619324101079e-06, "output_power": 1.1308694696869763e-06, "reference_active": true, "delta_db": 0.35560626982781585}], "active_band_count": 3, "active_band_mae_db": 0.24776695175515084}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 7.756656945350889e-05, "log_rms_mae_db": 1.5509446622128877, "rms_envelope_nrmse": 0.18062668507327534, "envelope_correlation": 0.9756203530415829, "rms_gain_error_db": -0.3226069089416257, "reference_mean_power": 0.0003662329710023159, "output_mean_power": 0.00034001396073445565}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 2.3479287649550626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015936395121522886, "output_power": 0.00014512625153639588, "reference_active": true, "delta_db": -0.4064411094032121}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.480344306075996e-05, "output_power": 2.7269814910612297e-05, "reference_active": true, "delta_db": 0.41170219160896443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.158977700048883e-06, "output_power": 7.542578115355963e-06, "reference_active": true, "delta_db": 0.22668807069627792}], "active_band_count": 3, "active_band_mae_db": 0.3482771239028182}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 6.756499548428438e-05, "log_rms_mae_db": 0.9185131255305558, "rms_envelope_nrmse": 0.1005893036624681, "envelope_correlation": 0.992648425027901, "rms_gain_error_db": -0.32260114798460104, "reference_mean_power": 0.00036915132536056565, "output_mean_power": 0.00034272384160761104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 2.14579734589437e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015227225639161801, "output_power": 0.00014109417546288104, "reference_active": true, "delta_db": -0.3311169729420559}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5037342052746137e-05, "output_power": 1.6302239451905225e-05, "reference_active": true, "delta_db": 0.3507618920195039}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0212382710455496e-07, "output_power": 4.4717481331412957e-07, "reference_active": true, "delta_db": 0.4611752717086116}], "active_band_count": 3, "active_band_mae_db": 0.3810180455567238}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00011110909618382604, "log_rms_mae_db": 1.3057373607571636, "rms_envelope_nrmse": 0.14238196389410412, "envelope_correlation": 0.9872043656438627, "rms_gain_error_db": 0.09940426674665914, "reference_mean_power": 0.00024441441718797436, "output_mean_power": 0.0002500732547567373}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 3.469643399885664e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011046778972231385, "output_power": 0.00011512097383290335, "reference_active": true, "delta_db": 0.17918790152420425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.9627372395634935e-05, "output_power": 5.576779602239655e-05, "reference_active": true, "delta_db": -0.29062190430814944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.408768678752184e-06, "output_power": 5.025690546804986e-06, "reference_active": true, "delta_db": 0.5687843104997461}], "active_band_count": 3, "active_band_mae_db": 0.3461980387773666}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.166729119583493e-05, "log_rms_mae_db": 0.8069319456623342, "rms_envelope_nrmse": 0.0670783391906714, "envelope_correlation": 0.9971118937275958, "rms_gain_error_db": 0.09938174809349426, "reference_mean_power": 0.00024636420339127556, "output_mean_power": 0.0002520668766483159}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.3341266023063243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010489106418058692, "output_power": 0.0001083831452017232, "reference_active": true, "delta_db": 0.14223258173080192}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0862352851100365e-05, "output_power": 3.113922004779555e-05, "reference_active": true, "delta_db": 0.03878698212416976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.747611403563906e-07, "output_power": 4.829884879803981e-07, "reference_active": true, "delta_db": 1.101822276862467}], "active_band_count": 3, "active_band_mae_db": 0.4276139469058129}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6399909132847883, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4637083b902cc806fc94dfd54b36a3f3e8a7abb460cbefc990bd810d5ec83f66", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.976155265470841, "standard_deviation": 0.0027554338620442223, "minimum": 0.9718864723764636, "maximum": 0.9801239078734746, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5355485534475486, "standard_deviation": 0.028620164001667406, "minimum": 1.4909583875851866, "maximum": 1.583290005518686, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5550484136550977, "standard_deviation": 0.3122955061838845, "minimum": 0.23854056332209672, "maximum": 1.0453457230346952, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17863036185380193, "standard_deviation": 0.01037821593977604, "minimum": 0.16271123978401322, "maximum": 0.19583069558717192, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934789386055065, "standard_deviation": 0.0014035782296515698, "minimum": 0.9910910780504414, "maximum": 0.9953811983995423, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8919335716218151, "standard_deviation": 0.033147868923502605, "minimum": 0.8302046417424633, "maximum": 0.9233723824377397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6298526644572506, "standard_deviation": 0.33382825723510806, "minimum": 0.26759482630003956, "maximum": 1.2055911895198685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09111397484929633, "standard_deviation": 0.00976151625118106, "minimum": 0.07769679837454868, "maximum": 0.1068949971522815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9816894540623788, "standard_deviation": 0.006784156903742407, "minimum": 0.9663064500206863, "maximum": 0.9872043656438627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3225368467399696, "standard_deviation": 0.018217758626719684, "minimum": 1.3057373607571636, "maximum": 1.3539295396440112, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0926466193046651, "standard_deviation": 0.6965004770553415, "minimum": 0.2952857072327672, "maximum": 2.2346249279164208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1678190402641111, "standard_deviation": 0.02803672661363523, "minimum": 0.14238196389410412, "maximum": 0.2275336051671659, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947041637271083, "standard_deviation": 0.0014475587923731322, "minimum": 0.9927753096386781, "maximum": 0.9971118937275958, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7852411437369244, "standard_deviation": 0.019754652695396037, "minimum": 0.7510494366708991, "maximum": 0.8069319456623342, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.939460081170287, "standard_deviation": 0.6310799864036791, "minimum": 0.24124362680569397, "maximum": 2.0738161787225127, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09124021159411574, "standard_deviation": 0.016545299743111298, "minimum": 0.0670783391906714, "maximum": 0.1208903085526117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9836370805026555, "standard_deviation": 0.0029111889828771008, "minimum": 0.9808942411361598, "maximum": 0.9886485370789287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5985577010397996, "standard_deviation": 0.06004785515046017, "minimum": 1.5094152471084124, "maximum": 1.6741349743095246, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6416823676594576, "standard_deviation": 0.3680448218768179, "minimum": 0.11846223095215348, "maximum": 1.2762402220494713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15147773509900372, "standard_deviation": 0.014275094167362676, "minimum": 0.12627918415608294, "maximum": 0.16574998267365945, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9944185657392386, "standard_deviation": 0.0014848591964229784, "minimum": 0.9922005861791057, "maximum": 0.9967700685774867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9280459847835736, "standard_deviation": 0.04386471523744184, "minimum": 0.8583508663749679, "maximum": 0.9659932767561702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5659285864565853, "standard_deviation": 0.2937977973021483, "minimum": 0.0901386816612872, "maximum": 0.9361896686636807, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08630875183674452, "standard_deviation": 0.013272339131789193, "minimum": 0.0658528826472017, "maximum": 0.10847494779407821, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24361471102815482, "standard_deviation": 0.08713137517798944, "minimum": 0.1547110492441521, "maximum": 0.40524945741985413, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.982968499141689, "standard_deviation": 0.008732151272767068, "minimum": 5.971916206350104, "maximum": 5.9975004012474455, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40721697229470655, "standard_deviation": 0.0469229247306151, "minimum": 0.3390502621841816, "maximum": 0.47447913839188355, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04094780492135765, "standard_deviation": 0.09576040534679257, "minimum": -0.23196606290954516, "maximum": 0.07338326372921494, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.987436115635914, "standard_deviation": 0.002373846071383489, "minimum": 0.9841670419808527, "maximum": 0.9914806113097726, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2904767940788404, "standard_deviation": 0.029235411896308544, "minimum": 1.2574560539243558, "maximum": 1.337393878142992, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6044434440779747, "standard_deviation": 0.27476347234631904, "minimum": 0.2044414228627808, "maximum": 1.0252391735061916, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11902032622849482, "standard_deviation": 0.01175910913249497, "minimum": 0.09839123534351235, "maximum": 0.13449625869173296, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9962258873687296, "standard_deviation": 0.001033527886643571, "minimum": 0.9945623943981152, "maximum": 0.9976453868674409, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6810106026192855, "standard_deviation": 0.04481563943153273, "minimum": 0.5959643213282626, "maximum": 0.7524209905605043, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7534776520646448, "standard_deviation": 0.3720996643585279, "minimum": 0.09951178302249035, "maximum": 1.3332914690370916, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06334898515247114, "standard_deviation": 0.009623232501593708, "minimum": 0.04999646381112975, "maximum": 0.07924695336275146, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1284-1180-0020", "language": "en", "style": "normal", "speaker": "librispeech_1284", "source_utterance_id": "1284-1180-0020", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 93920, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1284-1180-0020.wav", "source_file_sha256": "99a0025e0c01f3b117b86a2b5c166124459eed0ed6a7829cdd0014f1aa83a7e5", "decoded_pcm_sha256": "0bce114d9020d75b12c3d76b3c232bd8cd315dc052825ca64d389fcb87293422", "original_source_sha256": "b67810620f4f39a1bca2214b4f6475b98f19452b1cd0a4cc67efee2a8a805016", "corpus_entry_id": "librispeech_1284-1180-0020"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.140270709991455, "predicted_mos": 4.140270709991455, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "c397b5490b9d7fa2f0b21a1024965580556c040b06869432306f99625585032f", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.686292628997762, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1284-1180-0020", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c397b5490b9d7fa2f0b21a1024965580556c040b06869432306f99625585032f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "44e354e53fa0fa7d23c7248eb1c344287bbb4c5a6c63a146c8dad456478e78b7"}, "rows": [{"case": "librispeech_1284-1180-0020", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.378013849258423, "predicted_mos": 2.378013849258423, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "9a732fea90114a6893d214d74cb378fe679a77d3a11c3bf0b76c7febae4bc785", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.356130227999529, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1284-1180-0020", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9a732fea90114a6893d214d74cb378fe679a77d3a11c3bf0b76c7febae4bc785", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9042010578886195, "全帯域_10ms_rms_envelope_nrmse": 0.08372270474480947, "全帯域_10ms_envelope_correlation": 0.9921156696095735, "全帯域_10ms_modulation_mae_db": 0.26483796714710683, "全帯域_40ms_log_rms_mae_db": 0.47611483515060155, "全帯域_40ms_rms_envelope_nrmse": 0.049657942477446974, "全帯域_40ms_envelope_correlation": 0.9971368025688211, "全帯域_40ms_modulation_mae_db": 0.44598877458562214, "300–1000 Hz_10ms_log_rms_mae_db": 1.3884864207488592, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0972609121468172, "300–1000 Hz_10ms_envelope_correlation": 0.991910633268549, "300–1000 Hz_10ms_modulation_mae_db": 0.5834932263256932, "300–1000 Hz_40ms_log_rms_mae_db": 0.8101564659786546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05495485611328132, "300–1000 Hz_40ms_envelope_correlation": 0.9974706108852845, "300–1000 Hz_40ms_modulation_mae_db": 0.7402417727130438, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6365563311249935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1988390367411773, "1000–3000 Hz_10ms_envelope_correlation": 0.9659947718783907, "1000–3000 Hz_10ms_modulation_mae_db": 0.4332778352161153, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1228575179231848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13234585957497666, "1000–3000 Hz_40ms_envelope_correlation": 0.9834237034520573, "1000–3000 Hz_40ms_modulation_mae_db": 0.2156030904128493, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7108576977179604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18096547848637584, "3000–6000 Hz_10ms_envelope_correlation": 0.9716637127370457, "3000–6000 Hz_10ms_modulation_mae_db": 0.48951484025754727, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0309220093382598, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10266716788548252, "3000–6000 Hz_40ms_envelope_correlation": 0.9899685203174009, "3000–6000 Hz_40ms_modulation_mae_db": 0.43989712889033195, "mr_stft_db": 5.740404053655694, "band_mae_db": 0.22569181718122802, "rms_gain_error_db": -0.15198522264160336, "output_peak": 0.419495965061527}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8463446956186151, "全帯域_10ms_rms_envelope_nrmse": 0.08651761029415231, "全帯域_10ms_envelope_correlation": 0.9914750704538164, "全帯域_10ms_modulation_mae_db": 0.2681086212211165, "全帯域_40ms_log_rms_mae_db": 0.4484678733217263, "全帯域_40ms_rms_envelope_nrmse": 0.053680487711332776, "全帯域_40ms_envelope_correlation": 0.9964739190343498, "全帯域_40ms_modulation_mae_db": 0.4636111695288309, "300–1000 Hz_10ms_log_rms_mae_db": 1.3441064455798712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1009203074962268, "300–1000 Hz_10ms_envelope_correlation": 0.9910935385611659, "300–1000 Hz_10ms_modulation_mae_db": 0.37363740503623166, "300–1000 Hz_40ms_log_rms_mae_db": 0.7655768950733277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05990241588273545, "300–1000 Hz_40ms_envelope_correlation": 0.9966849612171469, "300–1000 Hz_40ms_modulation_mae_db": 0.5011325654773553, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6484360581176463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2046958138565919, "1000–3000 Hz_10ms_envelope_correlation": 0.9638017814440819, "1000–3000 Hz_10ms_modulation_mae_db": 0.3739827235271975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1726468960548542, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13521595338476078, "1000–3000 Hz_40ms_envelope_correlation": 0.9825533838825098, "1000–3000 Hz_40ms_modulation_mae_db": 0.16063890236839465, "3000–6000 Hz_10ms_log_rms_mae_db": 1.698283771227528, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16403874856745473, "3000–6000 Hz_10ms_envelope_correlation": 0.9757210191447956, "3000–6000 Hz_10ms_modulation_mae_db": 1.0419283005449522, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0331619731301884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09957071610863806, "3000–6000 Hz_40ms_envelope_correlation": 0.9899290779212481, "3000–6000 Hz_40ms_modulation_mae_db": 0.6380006445379737, "mr_stft_db": 5.730043582127409, "band_mae_db": 0.17826284596263584, "rms_gain_error_db": -0.12597169246937345, "output_peak": 0.41802946919040823}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.903851793534401, "全帯域_10ms_rms_envelope_nrmse": 0.08809985547278497, "全帯域_10ms_envelope_correlation": 0.991059076685996, "全帯域_10ms_modulation_mae_db": 0.1549575619368518, "全帯域_40ms_log_rms_mae_db": 0.46804457978777214, "全帯域_40ms_rms_envelope_nrmse": 0.047181446724603324, "全帯域_40ms_envelope_correlation": 0.9972049716303127, "全帯域_40ms_modulation_mae_db": 0.08905366276260895, "300–1000 Hz_10ms_log_rms_mae_db": 1.3418349308738355, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1057223398409955, "300–1000 Hz_10ms_envelope_correlation": 0.9901318515451399, "300–1000 Hz_10ms_modulation_mae_db": 0.08507722759395915, "300–1000 Hz_40ms_log_rms_mae_db": 0.786729690048149, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0595712201878296, "300–1000 Hz_40ms_envelope_correlation": 0.9966336862496882, "300–1000 Hz_40ms_modulation_mae_db": 0.11332471114594089, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6421449676324311, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1894634824373215, "1000–3000 Hz_10ms_envelope_correlation": 0.9682410030881976, "1000–3000 Hz_10ms_modulation_mae_db": 0.774012149828822, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1550882485293468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12588226448034004, "1000–3000 Hz_40ms_envelope_correlation": 0.9850432409246794, "1000–3000 Hz_40ms_modulation_mae_db": 0.4877130443991038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7312999857271862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17878579399701106, "3000–6000 Hz_10ms_envelope_correlation": 0.9719028838909245, "3000–6000 Hz_10ms_modulation_mae_db": 0.6765659742567479, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0770199603515749, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11199492378232755, "3000–6000 Hz_40ms_envelope_correlation": 0.9888566015154211, "3000–6000 Hz_40ms_modulation_mae_db": 0.9122266049448143, "mr_stft_db": 5.752184054596351, "band_mae_db": 0.22031268057493159, "rms_gain_error_db": -0.09397662777215879, "output_peak": 0.40954558514273703}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9003327087749149, "全帯域_10ms_rms_envelope_nrmse": 0.07862538193801108, "全帯域_10ms_envelope_correlation": 0.9929411505774781, "全帯域_10ms_modulation_mae_db": 0.16354999968168507, "全帯域_40ms_log_rms_mae_db": 0.46656404078396124, "全帯域_40ms_rms_envelope_nrmse": 0.04164780958741401, "全帯域_40ms_envelope_correlation": 0.9978950431121792, "全帯域_40ms_modulation_mae_db": 0.22626558452343848, "300–1000 Hz_10ms_log_rms_mae_db": 1.4066910184041108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1005255998652168, "300–1000 Hz_10ms_envelope_correlation": 0.9911665212970366, "300–1000 Hz_10ms_modulation_mae_db": 0.29203926285248677, "300–1000 Hz_40ms_log_rms_mae_db": 0.8242973702625279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05464990329216009, "300–1000 Hz_40ms_envelope_correlation": 0.997296444003037, "300–1000 Hz_40ms_modulation_mae_db": 0.38345619527822894, "1000–3000 Hz_10ms_log_rms_mae_db": 1.656562404025621, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1914053974570616, "1000–3000 Hz_10ms_envelope_correlation": 0.9678982281691888, "1000–3000 Hz_10ms_modulation_mae_db": 0.31171063282114014, "1000–3000 Hz_40ms_log_rms_mae_db": 1.148046071622314, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12639688697060655, "1000–3000 Hz_40ms_envelope_correlation": 0.9844887678726829, "1000–3000 Hz_40ms_modulation_mae_db": 0.1006016871148417, "3000–6000 Hz_10ms_log_rms_mae_db": 1.695782107649975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1746900420301959, "3000–6000 Hz_10ms_envelope_correlation": 0.9733551459210328, "3000–6000 Hz_10ms_modulation_mae_db": 0.6490324127254291, "3000–6000 Hz_40ms_log_rms_mae_db": 1.027424645340092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1129703085040656, "3000–6000 Hz_40ms_envelope_correlation": 0.9887229225376567, "3000–6000 Hz_40ms_modulation_mae_db": 0.6828279124057706, "mr_stft_db": 5.740076298820114, "band_mae_db": 0.18780180502549634, "rms_gain_error_db": -0.10585376465677065, "output_peak": 0.40959235674593514}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9006525978579398, "全帯域_10ms_rms_envelope_nrmse": 0.07739252650328464, "全帯域_10ms_envelope_correlation": 0.9931235641539571, "全帯域_10ms_modulation_mae_db": 0.2963942849376268, "全帯域_40ms_log_rms_mae_db": 0.45779489184201844, "全帯域_40ms_rms_envelope_nrmse": 0.040003864164881005, "全帯域_40ms_envelope_correlation": 0.9980152834692382, "全帯域_40ms_modulation_mae_db": 0.5931042454547035, "300–1000 Hz_10ms_log_rms_mae_db": 1.4546531341102147, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10545121757260739, "300–1000 Hz_10ms_envelope_correlation": 0.9902236936352539, "300–1000 Hz_10ms_modulation_mae_db": 0.2186411020756707, "300–1000 Hz_40ms_log_rms_mae_db": 0.8724196248174325, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057750146748617566, "300–1000 Hz_40ms_envelope_correlation": 0.9968994915647786, "300–1000 Hz_40ms_modulation_mae_db": 0.5541774299797063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6594285036937186, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22334858663809942, "1000–3000 Hz_10ms_envelope_correlation": 0.9571535284546278, "1000–3000 Hz_10ms_modulation_mae_db": 0.25279163241938335, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1230179178745456, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15437703457447488, "1000–3000 Hz_40ms_envelope_correlation": 0.977379575582299, "1000–3000 Hz_40ms_modulation_mae_db": 0.2875548539553594, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7229821060117274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1793992328008048, "3000–6000 Hz_10ms_envelope_correlation": 0.9709276793244641, "3000–6000 Hz_10ms_modulation_mae_db": 0.4113942130153581, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0434902175563043, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10978753711676657, "3000–6000 Hz_40ms_envelope_correlation": 0.9876583524357749, "3000–6000 Hz_40ms_modulation_mae_db": 0.450452456280259, "mr_stft_db": 5.741032141826565, "band_mae_db": 0.12432455232362229, "rms_gain_error_db": -0.08788548464676986, "output_peak": 0.39974503937855843}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9441880238346018, "全帯域_10ms_rms_envelope_nrmse": 0.08714305015526759, "全帯域_10ms_envelope_correlation": 0.991230064093993, "全帯域_10ms_modulation_mae_db": 0.45310390305711307, "全帯域_40ms_log_rms_mae_db": 0.4595163631724958, "全帯域_40ms_rms_envelope_nrmse": 0.0464386293801596, "全帯域_40ms_envelope_correlation": 0.9972240638627274, "全帯域_40ms_modulation_mae_db": 0.5097102802475828, "300–1000 Hz_10ms_log_rms_mae_db": 1.4104128852763136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10591187493391402, "300–1000 Hz_10ms_envelope_correlation": 0.9900423963659031, "300–1000 Hz_10ms_modulation_mae_db": 0.8077645330222424, "300–1000 Hz_40ms_log_rms_mae_db": 0.8348002555697287, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06591060732995888, "300–1000 Hz_40ms_envelope_correlation": 0.9957215859377204, "300–1000 Hz_40ms_modulation_mae_db": 0.9796570042116022, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6879059973584378, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22080906297746747, "1000–3000 Hz_10ms_envelope_correlation": 0.957334279505963, "1000–3000 Hz_10ms_modulation_mae_db": 0.5418236659340565, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1934264970954251, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1527483162058212, "1000–3000 Hz_40ms_envelope_correlation": 0.9773565102652901, "1000–3000 Hz_40ms_modulation_mae_db": 0.302503928605554, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6802029294421241, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17422230788493068, "3000–6000 Hz_10ms_envelope_correlation": 0.9730303346918447, "3000–6000 Hz_10ms_modulation_mae_db": 1.0444239705804657, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9481425371164525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10767048537867167, "3000–6000 Hz_40ms_envelope_correlation": 0.989210669916478, "3000–6000 Hz_40ms_modulation_mae_db": 1.2970491952257508, "mr_stft_db": 5.7384731303537535, "band_mae_db": 0.13320606699709034, "rms_gain_error_db": -0.057214107003001696, "output_peak": 0.45965607467976854}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.911762177642551, "全帯域_10ms_rms_envelope_nrmse": 0.08365981868817586, "全帯域_10ms_envelope_correlation": 0.9920202090282838, "全帯域_10ms_modulation_mae_db": 0.26110722561122657, "全帯域_40ms_log_rms_mae_db": 0.4300167930802085, "全帯域_40ms_rms_envelope_nrmse": 0.043128103598392836, "全帯域_40ms_envelope_correlation": 0.9977946194630544, "全帯域_40ms_modulation_mae_db": 0.3343254114162871, "300–1000 Hz_10ms_log_rms_mae_db": 1.3616718034595052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0975665762594878, "300–1000 Hz_10ms_envelope_correlation": 0.9917014992678328, "300–1000 Hz_10ms_modulation_mae_db": 0.37435539195629436, "300–1000 Hz_40ms_log_rms_mae_db": 0.7815267129596019, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053907740425008496, "300–1000 Hz_40ms_envelope_correlation": 0.9973758892599855, "300–1000 Hz_40ms_modulation_mae_db": 0.7180308112965476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6698800010874224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21212157830729964, "1000–3000 Hz_10ms_envelope_correlation": 0.9606642108788979, "1000–3000 Hz_10ms_modulation_mae_db": 0.29609216028884644, "1000–3000 Hz_40ms_log_rms_mae_db": 1.17554637767136, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13684069509994856, "1000–3000 Hz_40ms_envelope_correlation": 0.9818739139711501, "1000–3000 Hz_40ms_modulation_mae_db": 0.3642541111833975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7186653510942669, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17352602443391857, "3000–6000 Hz_10ms_envelope_correlation": 0.9728058664369975, "3000–6000 Hz_10ms_modulation_mae_db": 0.8011139427563777, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0450156910144932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09983957170879934, "3000–6000 Hz_40ms_envelope_correlation": 0.9900751217844392, "3000–6000 Hz_40ms_modulation_mae_db": 1.1314890777739184, "mr_stft_db": 5.7388240369261245, "band_mae_db": 0.14894589719786142, "rms_gain_error_db": -0.12200771361856622, "output_peak": 0.535234734942967}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.892245793878679, "全帯域_10ms_rms_envelope_nrmse": 0.08767714308762323, "全帯域_10ms_envelope_correlation": 0.9910945451017357, "全帯域_10ms_modulation_mae_db": 0.4719010493747011, "全帯域_40ms_log_rms_mae_db": 0.4757481739369863, "全帯域_40ms_rms_envelope_nrmse": 0.050443060311198294, "全帯域_40ms_envelope_correlation": 0.9966896009710862, "全帯域_40ms_modulation_mae_db": 0.4978659146721666, "300–1000 Hz_10ms_log_rms_mae_db": 1.3748909575917647, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10624984813986478, "300–1000 Hz_10ms_envelope_correlation": 0.9899788244763178, "300–1000 Hz_10ms_modulation_mae_db": 0.3740325121904162, "300–1000 Hz_40ms_log_rms_mae_db": 0.8314690807787354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06363611200073135, "300–1000 Hz_40ms_envelope_correlation": 0.9960158674046349, "300–1000 Hz_40ms_modulation_mae_db": 0.6609057110750548, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6292773382450763, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19811272812216849, "1000–3000 Hz_10ms_envelope_correlation": 0.9656262327094628, "1000–3000 Hz_10ms_modulation_mae_db": 0.36531880743656525, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1094873296702903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12370491689806266, "1000–3000 Hz_40ms_envelope_correlation": 0.985212141955443, "1000–3000 Hz_40ms_modulation_mae_db": 0.41062675014317734, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7132600618894205, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17781969145858345, "3000–6000 Hz_10ms_envelope_correlation": 0.9714305497822907, "3000–6000 Hz_10ms_modulation_mae_db": 0.7377513194783059, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0267075741360632, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10327926092566335, "3000–6000 Hz_40ms_envelope_correlation": 0.989437368107763, "3000–6000 Hz_40ms_modulation_mae_db": 0.6656063062664631, "mr_stft_db": 5.750551706254946, "band_mae_db": 0.09759287458454043, "rms_gain_error_db": -0.05132232190584293, "output_peak": 0.43988986674913005}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 93920, "output_samples": 93920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.641761260660288, "spectral_convergence": 0.2752986376285834, "spectral_convergence_status": "ok", "amplitude_floor": 2.214924872151779e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.662855949101305, "spectral_convergence": 0.2756324216046703, "spectral_convergence_status": "ok", "amplitude_floor": 1.3113821592368984e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9165949512054885, "spectral_convergence": 0.2888155945760999, "spectral_convergence_status": "ok", "amplitude_floor": 7.380112905156557e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00039502697710642294, "output_power": 0.0003658390503325081, "delta_db": -0.33336694388767124, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014539054117940992, "output_power": 0.0014702642977974859, "delta_db": 0.04859258350606073, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0016903939251448888, "output_power": 0.001569399833043814, "delta_db": -0.3225432122875094, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.480236057553453e-05, "output_power": 6.856340624681429e-05, "delta_db": 0.24501558969821452, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.753736279208405e-05, "output_power": 9.512062526871164e-05, "delta_db": -0.1089631256846898, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.937191536111839e-05, "output_power": 5.2850234739177156e-05, "delta_db": 0.2956694480232228, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22569181718122802, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15198522264160336, "pitch": null, "output_rms_dbfs": -24.410470639260208, "output_peak": 0.419495965061527, "reference_rms_dbfs": -24.258485416618605, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.740404053655694}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 93920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002156956510733609, "log_rms_mae_db": 0.9042010578886195, "rms_envelope_nrmse": 0.08372270474480947, "envelope_correlation": 0.9921156696095735, "rms_gain_error_db": -0.15198711468024512, "reference_mean_power": 0.003754233895192234, "output_mean_power": 0.0036251219001048655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.680528929094005e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006742966926403748, "output_power": 0.0006447001065753353, "reference_active": true, "delta_db": -0.1949328750540909}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002358850904334285, "output_power": 0.00022420513206910316, "reference_active": true, "delta_db": -0.22054941943074646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0118677140216256e-05, "output_power": 2.7601503048643887e-05, "reference_active": true, "delta_db": -0.37903160695648314}], "active_band_count": 3, "active_band_mae_db": 0.26483796714710683}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00016458512240675186, "log_rms_mae_db": 0.47611483515060155, "rms_envelope_nrmse": 0.049657942477446974, "envelope_correlation": 0.9971368025688211, "rms_gain_error_db": -0.15198678231658913, "reference_mean_power": 0.003773515049488898, "output_mean_power": 0.0036437402343733123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.4929366354790985e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006129787155244603, "output_power": 0.000583615519096496, "reference_active": true, "delta_db": -0.2131856293120464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073825851363778, "output_power": 0.00010598501663126739, "reference_active": true, "delta_db": -0.1905321715025396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.608951147124256e-06, "output_power": 2.103977517080805e-06, "reference_active": true, "delta_db": -0.9342485229422804}], "active_band_count": 3, "active_band_mae_db": 0.44598877458562214}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002116840322083979, "log_rms_mae_db": 1.3884864207488592, "rms_envelope_nrmse": 0.0972609121468172, "envelope_correlation": 0.991910633268549, "rms_gain_error_db": -0.2019774141178007, "reference_mean_power": 0.002333704816993391, "output_mean_power": 0.0022276562787475478}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.3033890876287954e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006736793990758906, "output_power": 0.0006407700112124164, "reference_active": true, "delta_db": -0.21751088893198828}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002058984330594163, "output_power": 0.00018627748598167444, "reference_active": true, "delta_db": -0.4349267351532792}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6276513462408744e-05, "output_power": 2.0406265565339515e-05, "reference_active": true, "delta_db": -1.0980420548918124}], "active_band_count": 3, "active_band_mae_db": 0.5834932263256932}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00015464611695479593, "log_rms_mae_db": 0.8101564659786546, "rms_envelope_nrmse": 0.05495485611328132, "envelope_correlation": 0.9974706108852845, "rms_gain_error_db": -0.201977627874267, "reference_mean_power": 0.002345702993864749, "output_mean_power": 0.002239109122592733}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.1648344391379404e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377113304270822, "output_power": 0.0006053190113549928, "reference_active": true, "delta_db": -0.22639819222924798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.690162436721034e-05, "output_power": 8.866165387330184e-05, "reference_active": true, "delta_db": -0.3859522894950541}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0235059299775393e-06, "output_power": 2.0877252368141363e-06, "reference_active": true, "delta_db": -1.6083748364148294}], "active_band_count": 3, "active_band_mae_db": 0.7402417727130438}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 4.7944659079237455e-05, "log_rms_mae_db": 1.6365563311249935, "rms_envelope_nrmse": 0.1988390367411773, "envelope_correlation": 0.9659947718783907, "rms_gain_error_db": 0.12543892749268148, "reference_mean_power": 0.0001071814340711088, "output_mean_power": 0.00011032233781258061}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 8.167234143423064e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.040560289520861e-05, "output_power": 3.790892019201982e-05, "reference_active": true, "delta_db": -0.27700177252042113}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.346030743633636e-06, "output_power": 7.816291433426278e-06, "reference_active": true, "delta_db": -0.28479236306867795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1788672389317654e-06, "output_power": 2.582466556138583e-06, "reference_active": true, "delta_db": 0.7380393700592469}], "active_band_count": 3, "active_band_mae_db": 0.4332778352161153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 3.945206980281183e-05, "log_rms_mae_db": 1.1228575179231848, "rms_envelope_nrmse": 0.13234585957497666, "envelope_correlation": 0.9834237034520573, "rms_gain_error_db": 0.12543292746532328, "reference_mean_power": 0.00010773182378082642, "output_mean_power": 0.00011088870324610149}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 7.380645247762918e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7294287822712847e-05, "output_power": 3.507944086426608e-05, "reference_active": true, "delta_db": -0.2658965574149008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.878816663714596e-06, "output_power": 3.8707781465593784e-06, "reference_active": true, "delta_db": -0.009009722393873798}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8761762452219972e-07, "output_power": 1.7221990042710937e-07, "reference_active": true, "delta_db": -0.3719029914297733}], "active_band_count": 3, "active_band_mae_db": 0.2156030904128493}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 5.718269674480757e-05, "log_rms_mae_db": 1.7108576977179604, "rms_envelope_nrmse": 0.18096547848637584, "envelope_correlation": 0.9716637127370457, "rms_gain_error_db": 0.1603246233062611, "reference_mean_power": 7.041944137047507e-05, "output_mean_power": 7.30676328504915e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 5.931095881530669e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9173785649337023e-05, "output_power": 3.0506366309282985e-05, "reference_active": true, "delta_db": 0.19397693097649238}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1741151998348071e-05, "output_power": 1.2090997144192392e-05, "reference_active": true, "delta_db": 0.12751408223147048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.929169683988829e-06, "output_power": 2.512324764640619e-06, "reference_active": true, "delta_db": 1.1470535075646788}], "active_band_count": 3, "active_band_mae_db": 0.48951484025754727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 4.673628190489769e-05, "log_rms_mae_db": 1.0309220093382598, "rms_envelope_nrmse": 0.10266716788548252, "envelope_correlation": 0.9899685203174009, "rms_gain_error_db": 0.16027667788643868, "reference_mean_power": 7.077979037597618e-05, "output_mean_power": 7.344072235192907e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 5.253277550211025e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7931175609497952e-05, "output_power": 2.9916189987145468e-05, "reference_active": true, "delta_db": 0.2981706728318085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.1737648370747885e-06, "output_power": 5.817929256420174e-06, "reference_active": true, "delta_db": 0.5096175124488276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7085501560216023e-07, "output_power": 1.9222862194299677e-07, "reference_active": true, "delta_db": 0.5119032013903596}], "active_band_count": 3, "active_band_mae_db": 0.43989712889033195}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8737982713886107, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "deada37ac30ab8339ee8f31ec01b84d28cf9da7ddfa5e9fe8515cc34c98956f4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9633392545161013, "standard_deviation": 0.004452706126646939, "minimum": 0.9571535284546278, "maximum": 0.9682410030881976, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6537739501606685, "standard_deviation": 0.018990252009487184, "minimum": 1.6292773382450763, "maximum": 1.6879059973584378, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4186262009340158, "standard_deviation": 0.16923053196920673, "minimum": 0.25279163241938335, "maximum": 0.774012149828822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20484946081714842, "standard_deviation": 0.012810786709116007, "minimum": 0.1894634824373215, "maximum": 0.22334858663809942, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.982166404738264, "standard_deviation": 0.0031805433978637316, "minimum": 0.9773565102652901, "maximum": 0.985212141955443, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1500146070551651, "standard_deviation": 0.029729951015427133, "minimum": 1.1094873296702903, "maximum": 1.1934264970954251, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.29118704602283474, "standard_deviation": 0.12953041717090977, "minimum": 0.1006016871148417, "maximum": 0.4877130443991038, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13593899089862393, "standard_deviation": 0.011823846235313662, "minimum": 0.12370491689806266, "maximum": 0.15437703457447488, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9726046489911744, "standard_deviation": 0.0015149387945202317, "minimum": 0.9709276793244641, "maximum": 0.9757210191447956, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7089167513450236, "standard_deviation": 0.016589897524577978, "minimum": 1.6802029294421241, "maximum": 1.7312999857271862, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.731465621701898, "standard_deviation": 0.22991162719660255, "minimum": 0.4113942130153581, "maximum": 1.0444239705804657, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17543091495740937, "standard_deviation": 0.005328630127073897, "minimum": 0.16403874856745473, "maximum": 0.18096547848637584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9892323293170228, "standard_deviation": 0.0008157656224558053, "minimum": 0.9876583524357749, "maximum": 0.9900751217844392, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0289855759979285, "standard_deviation": 0.036526049596918915, "minimum": 0.9481425371164525, "maximum": 1.0770199603515749, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7771936657906602, "standard_deviation": 0.3105913733217972, "minimum": 0.43989712889033195, "maximum": 1.2970491952257508, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10597249642630183, "standard_deviation": 0.005337883183132878, "minimum": 0.09957071610863806, "maximum": 0.1129703085040656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907811198021499, "standard_deviation": 0.0007829159899364487, "minimum": 0.9899788244763178, "maximum": 0.991910633268549, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3853434495055594, "standard_deviation": 0.038099473292210945, "minimum": 1.3418349308738355, "maximum": 1.4546531341102147, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38863008263162435, "standard_deviation": 0.22189358591460603, "minimum": 0.08507722759395915, "maximum": 0.8077645330222424, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10245108453189128, "standard_deviation": 0.003835113421675622, "minimum": 0.0972609121468172, "maximum": 0.10624984813986478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967623170652845, "standard_deviation": 0.0006375284410721955, "minimum": 0.9957215859377204, "maximum": 0.9974706108852845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8133720119360197, "standard_deviation": 0.03466432270809373, "minimum": 0.7655768950733277, "maximum": 0.8724196248174325, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.581365775147185, "standard_deviation": 0.26029890310605264, "minimum": 0.11332471114594089, "maximum": 0.9796570042116022, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05878537524754034, "standard_deviation": 0.004355652091028358, "minimum": 0.053907740425008496, "maximum": 0.06591060732995888, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1645173174809258, "standard_deviation": 0.04614225184037479, "minimum": 0.09759287458454043, "maximum": 0.22569181718122802, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.74144862557012, "standard_deviation": 0.007041514016342887, "minimum": 5.730043582127409, "maximum": 5.752184054596351, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4363986364863789, "standard_deviation": 0.044284707562964105, "minimum": 0.39974503937855843, "maximum": 0.535234734942967, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09952711683926087, "standard_deviation": 0.03438646065314603, "minimum": -0.15198522264160336, "maximum": -0.05132232190584293, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918824187131042, "standard_deviation": 0.0008128330870801441, "minimum": 0.991059076685996, "maximum": 0.9931235641539571, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9004473561287902, "standard_deviation": 0.026889643343524866, "minimum": 0.8463446956186151, "maximum": 0.9441880238346018, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29174507662092847, "standard_deviation": 0.11703516866382765, "minimum": 0.1549575619368518, "maximum": 0.4719010493747011, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08410476136051365, "standard_deviation": 0.004126523334662436, "minimum": 0.07739252650328464, "maximum": 0.08809985547278497, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973042880139711, "standard_deviation": 0.0005614566373385087, "minimum": 0.9964739190343498, "maximum": 0.9980152834692382, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4602834438844713, "standard_deviation": 0.015361971207493072, "minimum": 0.4300167930802085, "maximum": 0.47611483515060155, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3949906303989051, "standard_deviation": 0.16735264469869562, "minimum": 0.08905366276260895, "maximum": 0.5931042454547035, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0465226679944286, "standard_deviation": 0.0047004098955056636, "minimum": 0.040003864164881005, "maximum": 0.053680487711332776, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0020", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.399773359298706, "predicted_mos": 2.399773359298706, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "257f1c6ff6a3c321cb879ee7e5711d8532f7c42f8a5d148d87960de1d2f4e238", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.757655099965632, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9153216535084484, "全帯域_10ms_rms_envelope_nrmse": 0.0816200252418175, "全帯域_10ms_envelope_correlation": 0.9924783328195946, "全帯域_10ms_modulation_mae_db": 0.24236374480920472, "全帯域_40ms_log_rms_mae_db": 0.4620496030462119, "全帯域_40ms_rms_envelope_nrmse": 0.045958551273617296, "全帯域_40ms_envelope_correlation": 0.9975448239809909, "全帯域_40ms_modulation_mae_db": 0.4423798629647808, "300–1000 Hz_10ms_log_rms_mae_db": 1.4053028192590906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09536212380417604, "300–1000 Hz_10ms_envelope_correlation": 0.9921713488552252, "300–1000 Hz_10ms_modulation_mae_db": 0.6019955036887599, "300–1000 Hz_40ms_log_rms_mae_db": 0.8058494074284558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05143408994365475, "300–1000 Hz_40ms_envelope_correlation": 0.9977542799296545, "300–1000 Hz_40ms_modulation_mae_db": 0.8459420928750925, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5538665400395693, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18003015687589896, "1000–3000 Hz_10ms_envelope_correlation": 0.971552268413788, "1000–3000 Hz_10ms_modulation_mae_db": 0.3720432019242699, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0163466104126004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11358899051230488, "1000–3000 Hz_40ms_envelope_correlation": 0.9874575910142579, "1000–3000 Hz_40ms_modulation_mae_db": 0.330975186136283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6542493773622087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17137401579288716, "3000–6000 Hz_10ms_envelope_correlation": 0.9746400302308923, "3000–6000 Hz_10ms_modulation_mae_db": 0.37991644991328816, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9764139674312985, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09775096224971995, "3000–6000 Hz_40ms_envelope_correlation": 0.9911014383376121, "3000–6000 Hz_40ms_modulation_mae_db": 0.2044228968844055, "mr_stft_db": 5.674785295759546, "band_mae_db": 0.20627142697971687, "rms_gain_error_db": -0.13833608167311545, "output_peak": 0.41682197273504246}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8676538660736458, "全帯域_10ms_rms_envelope_nrmse": 0.08510456322940717, "全帯域_10ms_envelope_correlation": 0.9917442426696007, "全帯域_10ms_modulation_mae_db": 0.299976385231513, "全帯域_40ms_log_rms_mae_db": 0.45209695741337025, "全帯域_40ms_rms_envelope_nrmse": 0.05014424774764114, "全帯域_40ms_envelope_correlation": 0.9969416313516306, "全帯域_40ms_modulation_mae_db": 0.5111487989877849, "300–1000 Hz_10ms_log_rms_mae_db": 1.3813755604494702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10313201659061447, "300–1000 Hz_10ms_envelope_correlation": 0.9906905003712138, "300–1000 Hz_10ms_modulation_mae_db": 0.412140513675884, "300–1000 Hz_40ms_log_rms_mae_db": 0.7821515003440574, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0596282605846923, "300–1000 Hz_40ms_envelope_correlation": 0.9967245284936187, "300–1000 Hz_40ms_modulation_mae_db": 0.5736946088147885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5936398734052706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1890631405758392, "1000–3000 Hz_10ms_envelope_correlation": 0.9687965953730433, "1000–3000 Hz_10ms_modulation_mae_db": 0.2953673400547789, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0837499870906724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11946079416208077, "1000–3000 Hz_40ms_envelope_correlation": 0.9861927486398375, "1000–3000 Hz_40ms_modulation_mae_db": 0.2725229400498081, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6394769851637647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1601187059102333, "3000–6000 Hz_10ms_envelope_correlation": 0.9769175294005827, "3000–6000 Hz_10ms_modulation_mae_db": 1.1427137449602458, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9724043988420078, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09688362212577395, "3000–6000 Hz_40ms_envelope_correlation": 0.9905868212620322, "3000–6000 Hz_40ms_modulation_mae_db": 0.5866636298052258, "mr_stft_db": 5.662018852091989, "band_mae_db": 0.17903099210531712, "rms_gain_error_db": -0.12010051804640078, "output_peak": 0.41483296690347987}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.912280763462204, "全帯域_10ms_rms_envelope_nrmse": 0.08740036884941649, "全帯域_10ms_envelope_correlation": 0.991225533639362, "全帯域_10ms_modulation_mae_db": 0.20483039330386185, "全帯域_40ms_log_rms_mae_db": 0.47169274891146007, "全帯域_40ms_rms_envelope_nrmse": 0.047286051124738104, "全帯域_40ms_envelope_correlation": 0.997227703044617, "全帯域_40ms_modulation_mae_db": 0.16729363630676494, "300–1000 Hz_10ms_log_rms_mae_db": 1.3448796248838502, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10577039582353846, "300–1000 Hz_10ms_envelope_correlation": 0.9901180434579699, "300–1000 Hz_10ms_modulation_mae_db": 0.2154122019428993, "300–1000 Hz_40ms_log_rms_mae_db": 0.7576478653685828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05940405015857012, "300–1000 Hz_40ms_envelope_correlation": 0.9966403682216505, "300–1000 Hz_40ms_modulation_mae_db": 0.2863310729440001, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5784622954889846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1863334526842664, "1000–3000 Hz_10ms_envelope_correlation": 0.9694910320386995, "1000–3000 Hz_10ms_modulation_mae_db": 0.7686396485745289, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1008630362826173, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12159913441594997, "1000–3000 Hz_40ms_envelope_correlation": 0.9864111807693791, "1000–3000 Hz_40ms_modulation_mae_db": 0.4295810749467323, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6463916179162579, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17254989683805613, "3000–6000 Hz_10ms_envelope_correlation": 0.974259122735799, "3000–6000 Hz_10ms_modulation_mae_db": 0.7007709894489084, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0076690129431092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10796790705466532, "3000–6000 Hz_40ms_envelope_correlation": 0.9902140047779804, "3000–6000 Hz_40ms_modulation_mae_db": 0.9708507137502602, "mr_stft_db": 5.683537216830866, "band_mae_db": 0.2607520796564614, "rms_gain_error_db": -0.10523480732482948, "output_peak": 0.4065091240288959}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9159754976778188, "全帯域_10ms_rms_envelope_nrmse": 0.08320091520245759, "全帯域_10ms_envelope_correlation": 0.9920979125662135, "全帯域_10ms_modulation_mae_db": 0.1833609191064922, "全帯域_40ms_log_rms_mae_db": 0.4693113751636469, "全帯域_40ms_rms_envelope_nrmse": 0.04388786466657109, "全帯域_40ms_envelope_correlation": 0.9976737410463304, "全帯域_40ms_modulation_mae_db": 0.2733689959186045, "300–1000 Hz_10ms_log_rms_mae_db": 1.4296812245628994, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10326642501805042, "300–1000 Hz_10ms_envelope_correlation": 0.990658382002192, "300–1000 Hz_10ms_modulation_mae_db": 0.36994430924552946, "300–1000 Hz_40ms_log_rms_mae_db": 0.8473987320016005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05541160148958647, "300–1000 Hz_40ms_envelope_correlation": 0.9972077929235907, "300–1000 Hz_40ms_modulation_mae_db": 0.4953833460714716, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5919637687699353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17851546979082347, "1000–3000 Hz_10ms_envelope_correlation": 0.9719630256805905, "1000–3000 Hz_10ms_modulation_mae_db": 0.20531840506860302, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0920667952829146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11433983057163591, "1000–3000 Hz_40ms_envelope_correlation": 0.9872961654086847, "1000–3000 Hz_40ms_modulation_mae_db": 0.2334147487324072, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6385265969645122, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16799262823358946, "3000–6000 Hz_10ms_envelope_correlation": 0.9753062360615831, "3000–6000 Hz_10ms_modulation_mae_db": 0.4603177242690995, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9647943424971102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10730398118258043, "3000–6000 Hz_40ms_envelope_correlation": 0.9897495673903683, "3000–6000 Hz_40ms_modulation_mae_db": 0.5622221467556049, "mr_stft_db": 5.665521413295405, "band_mae_db": 0.20757158405517573, "rms_gain_error_db": -0.11537112577524036, "output_peak": 0.40759553478987337}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9206570091653399, "全帯域_10ms_rms_envelope_nrmse": 0.08007517962120936, "全帯域_10ms_envelope_correlation": 0.9926660213212074, "全帯域_10ms_modulation_mae_db": 0.1975548373409108, "全帯域_40ms_log_rms_mae_db": 0.4826042184433319, "全帯域_40ms_rms_envelope_nrmse": 0.04247801225656316, "全帯域_40ms_envelope_correlation": 0.997800519635333, "全帯域_40ms_modulation_mae_db": 0.44014617943868545, "300–1000 Hz_10ms_log_rms_mae_db": 1.433150698067765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1071418453222776, "300–1000 Hz_10ms_envelope_correlation": 0.9898887141466262, "300–1000 Hz_10ms_modulation_mae_db": 0.11439527860341438, "300–1000 Hz_40ms_log_rms_mae_db": 0.8833406207045177, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05915710509902037, "300–1000 Hz_40ms_envelope_correlation": 0.9967154379126228, "300–1000 Hz_40ms_modulation_mae_db": 0.3859190207562098, "1000–3000 Hz_10ms_log_rms_mae_db": 1.613152088152188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19237188021637888, "1000–3000 Hz_10ms_envelope_correlation": 0.9675499203548402, "1000–3000 Hz_10ms_modulation_mae_db": 0.1231368358463924, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0718781140374813, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12573564514967805, "1000–3000 Hz_40ms_envelope_correlation": 0.9846315699063468, "1000–3000 Hz_40ms_modulation_mae_db": 0.4601298934842457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6622619984863563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1739428559969739, "3000–6000 Hz_10ms_envelope_correlation": 0.9727740062414254, "3000–6000 Hz_10ms_modulation_mae_db": 0.3266652744552989, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0115753023858398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1054501689981324, "3000–6000 Hz_40ms_envelope_correlation": 0.9889813940005414, "3000–6000 Hz_40ms_modulation_mae_db": 0.5905828163219379, "mr_stft_db": 5.676563566201383, "band_mae_db": 0.12256621576585175, "rms_gain_error_db": -0.10544653775306168, "output_peak": 0.39837203805861415}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9595758838240409, "全帯域_10ms_rms_envelope_nrmse": 0.08924687614321929, "全帯域_10ms_envelope_correlation": 0.990823927781939, "全帯域_10ms_modulation_mae_db": 0.47561218365565083, "全帯域_40ms_log_rms_mae_db": 0.4622485956738963, "全帯域_40ms_rms_envelope_nrmse": 0.04821448491882754, "全帯域_40ms_envelope_correlation": 0.9970415123731935, "全帯域_40ms_modulation_mae_db": 0.5357846192946275, "300–1000 Hz_10ms_log_rms_mae_db": 1.4258690488217083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11007900994676156, "300–1000 Hz_10ms_envelope_correlation": 0.9892541537989744, "300–1000 Hz_10ms_modulation_mae_db": 0.8285999660955591, "300–1000 Hz_40ms_log_rms_mae_db": 0.8350708988081493, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06867122861567662, "300–1000 Hz_40ms_envelope_correlation": 0.9953845849610236, "300–1000 Hz_40ms_modulation_mae_db": 0.9919991508106673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.619865311797409, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19800381726063332, "1000–3000 Hz_10ms_envelope_correlation": 0.9653984519527751, "1000–3000 Hz_10ms_modulation_mae_db": 0.4953416483268993, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1313740271669035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13029742549367487, "1000–3000 Hz_40ms_envelope_correlation": 0.9834520207209207, "1000–3000 Hz_40ms_modulation_mae_db": 0.294006963094494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6225424036198521, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17429543418575288, "3000–6000 Hz_10ms_envelope_correlation": 0.9727620797724225, "3000–6000 Hz_10ms_modulation_mae_db": 0.8198812203820091, "3000–6000 Hz_40ms_log_rms_mae_db": 0.938752944681241, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10499651131800992, "3000–6000 Hz_40ms_envelope_correlation": 0.9894460514683763, "3000–6000 Hz_40ms_modulation_mae_db": 1.2593313991169321, "mr_stft_db": 5.671912929607513, "band_mae_db": 0.1539098883877397, "rms_gain_error_db": -0.07562856007087405, "output_peak": 0.4492982040538823}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9281357034924758, "全帯域_10ms_rms_envelope_nrmse": 0.08550383976526708, "全帯域_10ms_envelope_correlation": 0.9916065652943861, "全帯域_10ms_modulation_mae_db": 0.2582323112687191, "全帯域_40ms_log_rms_mae_db": 0.44052853609488035, "全帯域_40ms_rms_envelope_nrmse": 0.04323277610500444, "全帯域_40ms_envelope_correlation": 0.997714950212436, "全帯域_40ms_modulation_mae_db": 0.32984897708264393, "300–1000 Hz_10ms_log_rms_mae_db": 1.3829619415815309, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09830218269381416, "300–1000 Hz_10ms_envelope_correlation": 0.9915414251246112, "300–1000 Hz_10ms_modulation_mae_db": 0.43367530786582237, "300–1000 Hz_40ms_log_rms_mae_db": 0.7795862498091827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05422404944963052, "300–1000 Hz_40ms_envelope_correlation": 0.9972969952397637, "300–1000 Hz_40ms_modulation_mae_db": 0.7324881263260609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6028511528140916, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18976837072034855, "1000–3000 Hz_10ms_envelope_correlation": 0.9681673752412907, "1000–3000 Hz_10ms_modulation_mae_db": 0.427790283084966, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0940530803449313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11655133570811704, "1000–3000 Hz_40ms_envelope_correlation": 0.9867739336380872, "1000–3000 Hz_40ms_modulation_mae_db": 0.5886805242150407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6482786998114674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16781763198352687, "3000–6000 Hz_10ms_envelope_correlation": 0.9746851914750349, "3000–6000 Hz_10ms_modulation_mae_db": 0.7039227978734446, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9827657490321139, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09671475249872083, "3000–6000 Hz_40ms_envelope_correlation": 0.9908986967736195, "3000–6000 Hz_40ms_modulation_mae_db": 1.1559864039012224, "mr_stft_db": 5.670555148340658, "band_mae_db": 0.15844892338558303, "rms_gain_error_db": -0.10318157997193259, "output_peak": 0.49144966118738376}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9168876048652401, "全帯域_10ms_rms_envelope_nrmse": 0.08887919678482106, "全帯域_10ms_envelope_correlation": 0.9908537942480923, "全帯域_10ms_modulation_mae_db": 0.46069308840379986, "全帯域_40ms_log_rms_mae_db": 0.4791784381258799, "全帯域_40ms_rms_envelope_nrmse": 0.05253091725949685, "全帯域_40ms_envelope_correlation": 0.9964212620398917, "全帯域_40ms_modulation_mae_db": 0.48837599654959485, "300–1000 Hz_10ms_log_rms_mae_db": 1.3961360769740168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10983835228028403, "300–1000 Hz_10ms_envelope_correlation": 0.9892815195769253, "300–1000 Hz_10ms_modulation_mae_db": 0.39174520466975604, "300–1000 Hz_40ms_log_rms_mae_db": 0.8105814891382862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0659259713752738, "300–1000 Hz_40ms_envelope_correlation": 0.9957139099475568, "300–1000 Hz_40ms_modulation_mae_db": 0.7129603306602577, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5741183254433222, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18307682146699839, "1000–3000 Hz_10ms_envelope_correlation": 0.9704060058872493, "1000–3000 Hz_10ms_modulation_mae_db": 0.34546218860183303, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0633332250947674, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11030061318999268, "1000–3000 Hz_40ms_envelope_correlation": 0.9881575100379021, "1000–3000 Hz_40ms_modulation_mae_db": 0.5615995907472082, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6671751680849711, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1717567691764644, "3000–6000 Hz_10ms_envelope_correlation": 0.9736115078365103, "3000–6000 Hz_10ms_modulation_mae_db": 0.6521123416165803, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9924540566201218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10000839539074158, "3000–6000 Hz_40ms_envelope_correlation": 0.9905366101032644, "3000–6000 Hz_40ms_modulation_mae_db": 0.7336794452503486, "mr_stft_db": 5.683035781032263, "band_mae_db": 0.13607771955644007, "rms_gain_error_db": -0.06150164506237511, "output_peak": 0.43921243132536175}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 93920, "output_samples": 93920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.581318943656075, "spectral_convergence": 0.2750918091198837, "spectral_convergence_status": "ok", "amplitude_floor": 2.214924872151779e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.592255834489881, "spectral_convergence": 0.2712451498002044, "spectral_convergence_status": "ok", "amplitude_floor": 1.3113821592368984e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.850781109132683, "spectral_convergence": 0.2872001723857188, "spectral_convergence_status": "ok", "amplitude_floor": 7.380112905156557e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00039502697710642294, "output_power": 0.0003721662149717278, "delta_db": -0.2588980980591598, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014539054117940992, "output_power": 0.0014672607656718256, "delta_db": 0.0397115162485849, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0016903939251448888, "output_power": 0.0015827314766699678, "delta_db": -0.2858068379694093, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.480236057553453e-05, "output_power": 6.676502023296834e-05, "delta_db": 0.1295815861254435, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.753736279208405e-05, "output_power": 9.198662284896869e-05, "delta_db": -0.25446334313208563, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.937191536111839e-05, "output_power": 5.252870495727255e-05, "delta_db": 0.2691671803436182, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20627142697971687, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13833608167311545, "pitch": null, "output_rms_dbfs": -24.39682149829172, "output_peak": 0.41682197273504246, "reference_rms_dbfs": -24.258485416618605, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.674785295759546}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 93920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002156956510733609, "log_rms_mae_db": 0.9153216535084484, "rms_envelope_nrmse": 0.0816200252418175, "envelope_correlation": 0.9924783328195946, "rms_gain_error_db": -0.13833933220689304, "reference_mean_power": 0.003754233895192234, "output_mean_power": 0.0036365318296655014}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.680528929094005e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006742966926403748, "output_power": 0.0006468872445682066, "reference_active": true, "delta_db": -0.18022441776642625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002358850904334285, "output_power": 0.00022404899974702165, "reference_active": true, "delta_db": -0.2235748197051034}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0118677140216256e-05, "output_power": 2.7958038162226067e-05, "reference_active": true, "delta_db": -0.3232919969560844}], "active_band_count": 3, "active_band_mae_db": 0.24236374480920472}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00016458512240675186, "log_rms_mae_db": 0.4620496030462119, "rms_envelope_nrmse": 0.045958551273617296, "envelope_correlation": 0.9975448239809909, "rms_gain_error_db": -0.13833925248937676, "reference_mean_power": 0.003773515049488898, "output_mean_power": 0.003655208551775881}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.4929366354790985e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006129787155244603, "output_power": 0.0005870437079036966, "reference_active": true, "delta_db": -0.18774957237044923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073825851363778, "output_power": 0.00010452586069289035, "reference_active": true, "delta_db": -0.25073937119921824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.608951147124256e-06, "output_power": 2.1261841800153485e-06, "reference_active": true, "delta_db": -0.8886506453246751}], "active_band_count": 3, "active_band_mae_db": 0.4423798629647808}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002116840322083979, "log_rms_mae_db": 1.4053028192590906, "rms_envelope_nrmse": 0.09536212380417604, "envelope_correlation": 0.9921713488552252, "rms_gain_error_db": -0.1812978656242808, "reference_mean_power": 0.002333704816993391, "output_mean_power": 0.0022382888746844084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.3033890876287954e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006736793990758906, "output_power": 0.0006450983565360742, "reference_active": true, "delta_db": -0.18827331475001527}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002058984330594163, "output_power": 0.0001840881714127108, "reference_active": true, "delta_db": -0.4862715773248796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6276513462408744e-05, "output_power": 2.024993240270076e-05, "reference_active": true, "delta_db": -1.1314416189913847}], "active_band_count": 3, "active_band_mae_db": 0.6019955036887599}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00015464611695479593, "log_rms_mae_db": 0.8058494074284558, "rms_envelope_nrmse": 0.05143408994365475, "envelope_correlation": 0.9977542799296545, "rms_gain_error_db": -0.18129796105099824, "reference_mean_power": 0.002345702993864749, "output_mean_power": 0.0022497964442137262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.1648344391379404e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377113304270822, "output_power": 0.0006091989867209874, "reference_active": true, "delta_db": -0.19864961005014506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.690162436721034e-05, "output_power": 8.725409504125956e-05, "reference_active": true, "delta_db": -0.45545238636068297}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0235059299775393e-06, "output_power": 1.959468997454739e-06, "reference_active": true, "delta_db": -1.8837242822144493}], "active_band_count": 3, "active_band_mae_db": 0.8459420928750925}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 4.7944659079237455e-05, "log_rms_mae_db": 1.5538665400395693, "rms_envelope_nrmse": 0.18003015687589896, "envelope_correlation": 0.971552268413788, "rms_gain_error_db": -0.010548348972861282, "reference_mean_power": 0.0001071814340711088, "output_mean_power": 0.00010692142264653531}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 8.167234143423064e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.040560289520861e-05, "output_power": 3.76359012579149e-05, "reference_active": true, "delta_db": -0.30839270851888484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.346030743633636e-06, "output_power": 7.817443246528931e-06, "reference_active": true, "delta_db": -0.2841524314262247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1788672389317654e-06, "output_power": 2.4580415258895495e-06, "reference_active": true, "delta_db": 0.5235844658276999}], "active_band_count": 3, "active_band_mae_db": 0.3720432019242699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 3.945206980281183e-05, "log_rms_mae_db": 1.0163466104126004, "rms_envelope_nrmse": 0.11358899051230488, "envelope_correlation": 0.9874575910142579, "rms_gain_error_db": -0.01055051591111644, "reference_mean_power": 0.00010773182378082642, "output_mean_power": 0.00010747042354291698}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 7.380645247762918e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7294287822712847e-05, "output_power": 3.4787495513113035e-05, "reference_active": true, "delta_db": -0.3021915488802654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.878816663714596e-06, "output_power": 3.851436326967865e-06, "reference_active": true, "delta_db": -0.030765303291976073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8761762452219972e-07, "output_power": 1.6116723764866947e-07, "reference_active": true, "delta_db": -0.6599687062366076}], "active_band_count": 3, "active_band_mae_db": 0.330975186136283}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 5.718269674480757e-05, "log_rms_mae_db": 1.6542493773622087, "rms_envelope_nrmse": 0.17137401579288716, "envelope_correlation": 0.9746400302308923, "rms_gain_error_db": 0.10654206358038523, "reference_mean_power": 7.041944137047507e-05, "output_mean_power": 7.216835097216594e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 5.931095881530669e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9173785649337023e-05, "output_power": 3.141550165195208e-05, "reference_active": true, "delta_db": 0.32151211341733604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1741151998348071e-05, "output_power": 1.1468758182215934e-05, "reference_active": true, "delta_db": -0.10194314594953845}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.929169683988829e-06, "output_power": 2.2750965800406965e-06, "reference_active": true, "delta_db": 0.71629409037299}], "active_band_count": 3, "active_band_mae_db": 0.37991644991328816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 4.673628190489769e-05, "log_rms_mae_db": 0.9764139674312985, "rms_envelope_nrmse": 0.09775096224971995, "envelope_correlation": 0.9911014383376121, "rms_gain_error_db": 0.10651265093866386, "reference_mean_power": 7.077979037597618e-05, "output_mean_power": 7.25371582037393e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 5.253277550211025e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7931175609497952e-05, "output_power": 3.05602367973605e-05, "reference_active": true, "delta_db": 0.3906749973920969}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.1737648370747885e-06, "output_power": 5.421488345784701e-06, "reference_active": true, "delta_db": 0.2031184367683223}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7085501560216023e-07, "output_power": 1.7162290864693066e-07, "reference_active": true, "delta_db": 0.019475256492797403}], "active_band_count": 3, "active_band_mae_db": 0.2044228968844055}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.87518949192797, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9051f535b29eb0b95b33c6bd9ea1d0c3694ef00e5eb52ca07c930a36d2c58892", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9691655843677847, "standard_deviation": 0.002174551892751697, "minimum": 0.9653984519527751, "maximum": 0.9719630256805905, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5909899194888464, "standard_deviation": 0.02169050761925606, "minimum": 1.5538665400395693, "maximum": 1.619865311797409, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3791374439352839, "standard_deviation": 0.19695308879618262, "minimum": 0.1231368358463924, "maximum": 0.7686396485745289, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18714538869889838, "standard_deviation": 0.006525707351109554, "minimum": 0.17851546979082347, "maximum": 0.19800381726063332, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.986296590016927, "standard_deviation": 0.001554979791200812, "minimum": 0.9834520207209207, "maximum": 0.9881575100379021, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.081708109464111, "standard_deviation": 0.03338920973864942, "minimum": 1.0163466104126004, "maximum": 1.1313740271669035, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3963638651757774, "standard_deviation": 0.13421711396959132, "minimum": 0.2334147487324072, "maximum": 0.5886805242150407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11898422115042927, "standard_deviation": 0.006683006894079592, "minimum": 0.11030061318999268, "maximum": 0.13029742549367487, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9743694629692813, "standard_deviation": 0.0013762740258502588, "minimum": 0.9727620797724225, "maximum": 0.9769175294005827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6473628559261737, "standard_deviation": 0.01424299173879021, "minimum": 1.6225424036198521, "maximum": 1.6671751680849711, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6482875678648593, "standard_deviation": 0.26472699201384264, "minimum": 0.3266652744552989, "maximum": 1.1427137449602458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16998099226468552, "standard_deviation": 0.0046583619681564645, "minimum": 0.1601187059102333, "maximum": 0.17429543418575288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9901893230142244, "standard_deviation": 0.0007386588197577109, "minimum": 0.9889813940005414, "maximum": 0.9911014383376121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9808537218041053, "standard_deviation": 0.02367210698497766, "minimum": 0.938752944681241, "maximum": 1.0115753023858398, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7579674314732421, "standard_deviation": 0.3500580561501752, "minimum": 0.2044228968844055, "maximum": 1.2593313991169321, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10213453760229305, "standard_deviation": 0.004790139405756262, "minimum": 0.09671475249872083, "maximum": 0.10796790705466532, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904505109167172, "standard_deviation": 0.0010334507849040912, "minimum": 0.9892541537989744, "maximum": 0.9921713488552252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3999196243250414, "standard_deviation": 0.0301604533129849, "minimum": 1.3448796248838502, "maximum": 1.433150698067765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4209885357234531, "standard_deviation": 0.2200232397119865, "minimum": 0.11439527860341438, "maximum": 0.8285999660955591, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1041115439349396, "standard_deviation": 0.0052415405446631464, "minimum": 0.09536212380417604, "maximum": 0.11007900994676156, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966797372036851, "standard_deviation": 0.0007954841054111901, "minimum": 0.9953845849610236, "maximum": 0.9977542799296545, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.812703345450354, "standard_deviation": 0.04106865999630331, "minimum": 0.7576478653685828, "maximum": 0.8833406207045177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6280897186573184, "standard_deviation": 0.2372482750831175, "minimum": 0.2863310729440001, "maximum": 0.9919991508106673, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059232044589513116, "standard_deviation": 0.005788941888394895, "minimum": 0.05143408994365475, "maximum": 0.06867122861567662, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17807860373653572, "standard_deviation": 0.04522267521813385, "minimum": 0.12256621576585175, "maximum": 0.2607520796564614, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.673491275394953, "standard_deviation": 0.007647401076085595, "minimum": 5.662018852091989, "maximum": 5.683537216830866, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4280114916353167, "standard_deviation": 0.030844137423728504, "minimum": 0.39837203805861415, "maximum": 0.49144966118738376, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10310010695972868, "standard_deviation": 0.024425844145522273, "minimum": -0.13833608167311545, "maximum": -0.06150164506237511, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9916870412925495, "standard_deviation": 0.0006985047957570735, "minimum": 0.990823927781939, "maximum": 0.9926660213212074, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9170609977586517, "standard_deviation": 0.02510334336722897, "minimum": 0.8676538660736458, "maximum": 0.9595758838240409, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29032798289001904, "standard_deviation": 0.11596609311340439, "minimum": 0.1833609191064922, "maximum": 0.47561218365565083, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08512887060470195, "standard_deviation": 0.0033357931898568938, "minimum": 0.08007517962120936, "maximum": 0.08924687614321929, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997295767960553, "standard_deviation": 0.0004773187985150573, "minimum": 0.9964212620398917, "maximum": 0.997800519635333, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46496380910908475, "standard_deviation": 0.013917733301908704, "minimum": 0.44052853609488035, "maximum": 0.4826042184433319, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3985433833179359, "standard_deviation": 0.129339784336264, "minimum": 0.16729363630676494, "maximum": 0.5357846192946275, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04671661316905745, "standard_deviation": 0.0035214044385004774, "minimum": 0.04247801225656316, "maximum": 0.05253091725949685, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0020", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.553428888320923, "predicted_mos": 2.553428888320923, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "3d88949e4811a48cb511a621e2dd7c736fb8bdc2ae8b5910dc4097035f53cb0b", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1680431999266148, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.888044814450776, "全帯域_10ms_rms_envelope_nrmse": 0.09310300619708971, "全帯域_10ms_envelope_correlation": 0.9900064307788392, "全帯域_10ms_modulation_mae_db": 0.4001167352543827, "全帯域_40ms_log_rms_mae_db": 0.4476871924267882, "全帯域_40ms_rms_envelope_nrmse": 0.04792191095794252, "全帯域_40ms_envelope_correlation": 0.9971277642972354, "全帯域_40ms_modulation_mae_db": 0.5216326581220899, "300–1000 Hz_10ms_log_rms_mae_db": 1.6267186487296, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11011317187845918, "300–1000 Hz_10ms_envelope_correlation": 0.989487496501855, "300–1000 Hz_10ms_modulation_mae_db": 0.8027589669832755, "300–1000 Hz_40ms_log_rms_mae_db": 1.0413093635611361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06320315666364494, "300–1000 Hz_40ms_envelope_correlation": 0.996460217269619, "300–1000 Hz_40ms_modulation_mae_db": 1.0831682198625467, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0440335224594133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22680301955399426, "1000–3000 Hz_10ms_envelope_correlation": 0.9547638896105776, "1000–3000 Hz_10ms_modulation_mae_db": 0.4673767827371358, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4099393663919622, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15127376456183309, "1000–3000 Hz_40ms_envelope_correlation": 0.9778856267625095, "1000–3000 Hz_40ms_modulation_mae_db": 0.3917715950261666, "3000–6000 Hz_10ms_log_rms_mae_db": 2.414343924427781, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23303350528639777, "3000–6000 Hz_10ms_envelope_correlation": 0.95094049013377, "3000–6000 Hz_10ms_modulation_mae_db": 1.3850029713230985, "3000–6000 Hz_40ms_log_rms_mae_db": 1.472588148062385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14108905223534426, "3000–6000 Hz_40ms_envelope_correlation": 0.9800607036704351, "3000–6000 Hz_40ms_modulation_mae_db": 1.0955385046896442, "mr_stft_db": 6.15247189323465, "band_mae_db": 0.08707589377835656, "rms_gain_error_db": -0.09980998510666125, "output_peak": 0.5596610448801249}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.905216012519439, "全帯域_10ms_rms_envelope_nrmse": 0.08387199779596852, "全帯域_10ms_envelope_correlation": 0.9919375815916708, "全帯域_10ms_modulation_mae_db": 0.4737287970646135, "全帯域_40ms_log_rms_mae_db": 0.45706100488021334, "全帯域_40ms_rms_envelope_nrmse": 0.04164079220616581, "全帯域_40ms_envelope_correlation": 0.9978943079816504, "全帯域_40ms_modulation_mae_db": 0.5150463137729777, "300–1000 Hz_10ms_log_rms_mae_db": 1.5705754891101387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10932548799432783, "300–1000 Hz_10ms_envelope_correlation": 0.9895153417926693, "300–1000 Hz_10ms_modulation_mae_db": 0.39676551935273197, "300–1000 Hz_40ms_log_rms_mae_db": 0.9355434252169424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056878317116152856, "300–1000 Hz_40ms_envelope_correlation": 0.9970613527015894, "300–1000 Hz_40ms_modulation_mae_db": 0.640258836573949, "1000–3000 Hz_10ms_log_rms_mae_db": 2.063733170261746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21485079529319054, "1000–3000 Hz_10ms_envelope_correlation": 0.9611825862396445, "1000–3000 Hz_10ms_modulation_mae_db": 0.3444409879456825, "1000–3000 Hz_40ms_log_rms_mae_db": 1.429883662928735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14665725455929352, "1000–3000 Hz_40ms_envelope_correlation": 0.9803802642895786, "1000–3000 Hz_40ms_modulation_mae_db": 0.43949019854034627, "3000–6000 Hz_10ms_log_rms_mae_db": 2.444502612901991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21789716984703397, "3000–6000 Hz_10ms_envelope_correlation": 0.9577045789375064, "3000–6000 Hz_10ms_modulation_mae_db": 0.27788631478234654, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4617720028556394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1267177709872629, "3000–6000 Hz_40ms_envelope_correlation": 0.9838641214784369, "3000–6000 Hz_40ms_modulation_mae_db": 0.3115136997477016, "mr_stft_db": 6.146246948652344, "band_mae_db": 0.18816082233286888, "rms_gain_error_db": -0.10409341871456855, "output_peak": 0.39563669553770225}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9022675170619805, "全帯域_10ms_rms_envelope_nrmse": 0.09680448730823411, "全帯域_10ms_envelope_correlation": 0.9891479355744116, "全帯域_10ms_modulation_mae_db": 0.14252687351677668, "全帯域_40ms_log_rms_mae_db": 0.47376544648002034, "全帯域_40ms_rms_envelope_nrmse": 0.05017355693793892, "全帯域_40ms_envelope_correlation": 0.9967778235252476, "全帯域_40ms_modulation_mae_db": 0.1021038022332124, "300–1000 Hz_10ms_log_rms_mae_db": 1.6031562861094926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10573987712853618, "300–1000 Hz_10ms_envelope_correlation": 0.9901876370867302, "300–1000 Hz_10ms_modulation_mae_db": 0.2687158324940656, "300–1000 Hz_40ms_log_rms_mae_db": 0.9908602752023135, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06526193912511218, "300–1000 Hz_40ms_envelope_correlation": 0.9959967454368787, "300–1000 Hz_40ms_modulation_mae_db": 0.5316093532867745, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0358472408908463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22009281957028345, "1000–3000 Hz_10ms_envelope_correlation": 0.9573231862132008, "1000–3000 Hz_10ms_modulation_mae_db": 0.5262598997447135, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3641346779235448, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1463335335254353, "1000–3000 Hz_40ms_envelope_correlation": 0.9792618510976194, "1000–3000 Hz_40ms_modulation_mae_db": 0.6906009908305921, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3800641059146246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20130549462152178, "3000–6000 Hz_10ms_envelope_correlation": 0.9641082225408184, "3000–6000 Hz_10ms_modulation_mae_db": 1.073514594157397, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4031447299646784, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12425358077037019, "3000–6000 Hz_40ms_envelope_correlation": 0.9852011490805198, "3000–6000 Hz_40ms_modulation_mae_db": 1.341468726469544, "mr_stft_db": 6.134633788604085, "band_mae_db": 0.11665373737628142, "rms_gain_error_db": -0.07619291851586445, "output_peak": 0.5327078685038497}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9204482612681825, "全帯域_10ms_rms_envelope_nrmse": 0.09596073938268289, "全帯域_10ms_envelope_correlation": 0.9893190247878796, "全帯域_10ms_modulation_mae_db": 0.2462576891787872, "全帯域_40ms_log_rms_mae_db": 0.46497147920656934, "全帯域_40ms_rms_envelope_nrmse": 0.06076500906779805, "全帯域_40ms_envelope_correlation": 0.9951590524706228, "全帯域_40ms_modulation_mae_db": 0.11636623043310339, "300–1000 Hz_10ms_log_rms_mae_db": 1.632622621024891, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10445380795801704, "300–1000 Hz_10ms_envelope_correlation": 0.9905393214937833, "300–1000 Hz_10ms_modulation_mae_db": 0.2111728565810084, "300–1000 Hz_40ms_log_rms_mae_db": 0.9739297561554346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06067241519948669, "300–1000 Hz_40ms_envelope_correlation": 0.9967308631892073, "300–1000 Hz_40ms_modulation_mae_db": 0.4939390962780732, "1000–3000 Hz_10ms_log_rms_mae_db": 2.065078998543571, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23390836426693382, "1000–3000 Hz_10ms_envelope_correlation": 0.9532794154549081, "1000–3000 Hz_10ms_modulation_mae_db": 0.5788509995513715, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4083173796385902, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16090193878189363, "1000–3000 Hz_40ms_envelope_correlation": 0.975757092286729, "1000–3000 Hz_40ms_modulation_mae_db": 0.15469732638803183, "3000–6000 Hz_10ms_log_rms_mae_db": 2.344199528459577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2132730719355721, "3000–6000 Hz_10ms_envelope_correlation": 0.959266001278663, "3000–6000 Hz_10ms_modulation_mae_db": 1.1312620822983617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4518896219821713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13318293749508792, "3000–6000 Hz_40ms_envelope_correlation": 0.9821775722933405, "3000–6000 Hz_40ms_modulation_mae_db": 1.0878708495229654, "mr_stft_db": 6.130643413751149, "band_mae_db": 0.18543785762375653, "rms_gain_error_db": -0.03691071342421152, "output_peak": 0.4829564820473059}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.882691006458677, "全帯域_10ms_rms_envelope_nrmse": 0.08162845301396844, "全帯域_10ms_envelope_correlation": 0.9923024147172838, "全帯域_10ms_modulation_mae_db": 0.4665007910737263, "全帯域_40ms_log_rms_mae_db": 0.4827849258736602, "全帯域_40ms_rms_envelope_nrmse": 0.045969168306083164, "全帯域_40ms_envelope_correlation": 0.9972807887470025, "全帯域_40ms_modulation_mae_db": 0.47930629583840484, "300–1000 Hz_10ms_log_rms_mae_db": 1.6758609716302209, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1050965763147181, "300–1000 Hz_10ms_envelope_correlation": 0.9901885560714766, "300–1000 Hz_10ms_modulation_mae_db": 0.39814760938003585, "300–1000 Hz_40ms_log_rms_mae_db": 1.0237061570988923, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06302954895695811, "300–1000 Hz_40ms_envelope_correlation": 0.9960741607310031, "300–1000 Hz_40ms_modulation_mae_db": 0.5688194347347009, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0588814721091757, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24364582220532421, "1000–3000 Hz_10ms_envelope_correlation": 0.9496171652693004, "1000–3000 Hz_10ms_modulation_mae_db": 0.4235823403977809, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4170154379788136, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1740719393326302, "1000–3000 Hz_40ms_envelope_correlation": 0.9718690872393121, "1000–3000 Hz_40ms_modulation_mae_db": 0.5623181043527875, "3000–6000 Hz_10ms_log_rms_mae_db": 2.370537455710897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1992236735743593, "3000–6000 Hz_10ms_envelope_correlation": 0.9649373508970469, "3000–6000 Hz_10ms_modulation_mae_db": 0.9771694559347566, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3894305230411592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12638164337206625, "3000–6000 Hz_40ms_envelope_correlation": 0.984147728365306, "3000–6000 Hz_40ms_modulation_mae_db": 1.0758604173361324, "mr_stft_db": 6.112038162482321, "band_mae_db": 0.14943253345277277, "rms_gain_error_db": -0.06295721498113567, "output_peak": 0.4629490398300641}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.898384761448849, "全帯域_10ms_rms_envelope_nrmse": 0.09011549512233903, "全帯域_10ms_envelope_correlation": 0.9906027092741551, "全帯域_10ms_modulation_mae_db": 0.15262576403730713, "全帯域_40ms_log_rms_mae_db": 0.5033075391360291, "全帯域_40ms_rms_envelope_nrmse": 0.05031632077325476, "全帯域_40ms_envelope_correlation": 0.9967284857828629, "全帯域_40ms_modulation_mae_db": 0.2816083241096348, "300–1000 Hz_10ms_log_rms_mae_db": 1.6722305015557042, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11137342342870854, "300–1000 Hz_10ms_envelope_correlation": 0.9889982244330664, "300–1000 Hz_10ms_modulation_mae_db": 0.24214497753618183, "300–1000 Hz_40ms_log_rms_mae_db": 1.0619037262532893, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06619183958455994, "300–1000 Hz_40ms_envelope_correlation": 0.9957302960041021, "300–1000 Hz_40ms_modulation_mae_db": 0.47919250956327447, "1000–3000 Hz_10ms_log_rms_mae_db": 2.080624155575799, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2279979043893039, "1000–3000 Hz_10ms_envelope_correlation": 0.9539703079335258, "1000–3000 Hz_10ms_modulation_mae_db": 0.8943196077341634, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4069245210217478, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14817307456084353, "1000–3000 Hz_40ms_envelope_correlation": 0.9788737868607382, "1000–3000 Hz_40ms_modulation_mae_db": 0.5276400274240499, "3000–6000 Hz_10ms_log_rms_mae_db": 2.356874292151608, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2050746515180014, "3000–6000 Hz_10ms_envelope_correlation": 0.9627596353890403, "3000–6000 Hz_10ms_modulation_mae_db": 1.0087738133965596, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3965898043692297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12370669680968442, "3000–6000 Hz_40ms_envelope_correlation": 0.9848659440861316, "3000–6000 Hz_40ms_modulation_mae_db": 1.166663789329185, "mr_stft_db": 6.127990012236579, "band_mae_db": 0.08241309743676427, "rms_gain_error_db": -0.06483361248772668, "output_peak": 0.4036257088779038}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9885862149991832, "全帯域_10ms_rms_envelope_nrmse": 0.09246896868071422, "全帯域_10ms_envelope_correlation": 0.9903594893301514, "全帯域_10ms_modulation_mae_db": 0.27397101279074865, "全帯域_40ms_log_rms_mae_db": 0.5226386368082675, "全帯域_40ms_rms_envelope_nrmse": 0.047163189692184505, "全帯域_40ms_envelope_correlation": 0.997549047299959, "全帯域_40ms_modulation_mae_db": 0.37418031491372167, "300–1000 Hz_10ms_log_rms_mae_db": 1.6566800466303144, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09978124491819103, "300–1000 Hz_10ms_envelope_correlation": 0.9917438753214278, "300–1000 Hz_10ms_modulation_mae_db": 0.5779459972593544, "300–1000 Hz_40ms_log_rms_mae_db": 0.9666075128132382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057444297425575275, "300–1000 Hz_40ms_envelope_correlation": 0.9975421844640477, "300–1000 Hz_40ms_modulation_mae_db": 0.5702919288218619, "1000–3000 Hz_10ms_log_rms_mae_db": 2.002543025167031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22118557989862353, "1000–3000 Hz_10ms_envelope_correlation": 0.9577419278825601, "1000–3000 Hz_10ms_modulation_mae_db": 0.3549261694869858, "1000–3000 Hz_40ms_log_rms_mae_db": 1.376285595498704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1446337781008132, "1000–3000 Hz_40ms_envelope_correlation": 0.9801795369765623, "1000–3000 Hz_40ms_modulation_mae_db": 0.5470427723165375, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3628394956868246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20126235042616125, "3000–6000 Hz_10ms_envelope_correlation": 0.964345206338731, "3000–6000 Hz_10ms_modulation_mae_db": 0.7967138312406791, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3807142613762913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12488067019033965, "3000–6000 Hz_40ms_envelope_correlation": 0.9846716222876241, "3000–6000 Hz_40ms_modulation_mae_db": 1.0114204903191597, "mr_stft_db": 6.144435534145619, "band_mae_db": 0.16196857846318688, "rms_gain_error_db": -0.17043893352027742, "output_peak": 0.4491744653117364}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9263439454806384, "全帯域_10ms_rms_envelope_nrmse": 0.10380365633215122, "全帯域_10ms_envelope_correlation": 0.9875814699490413, "全帯域_10ms_modulation_mae_db": 0.27993334175794543, "全帯域_40ms_log_rms_mae_db": 0.5106466719331578, "全帯域_40ms_rms_envelope_nrmse": 0.06740926624628352, "全帯域_40ms_envelope_correlation": 0.9940777500070179, "全帯域_40ms_modulation_mae_db": 0.4420668183227387, "300–1000 Hz_10ms_log_rms_mae_db": 1.6179849288346853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11196408554950027, "300–1000 Hz_10ms_envelope_correlation": 0.9890035593206827, "300–1000 Hz_10ms_modulation_mae_db": 0.5027903881084569, "300–1000 Hz_40ms_log_rms_mae_db": 0.9420145263235349, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06261409131479809, "300–1000 Hz_40ms_envelope_correlation": 0.9964056345065017, "300–1000 Hz_40ms_modulation_mae_db": 0.7970601784537105, "1000–3000 Hz_10ms_log_rms_mae_db": 2.037286093244661, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2137596077605246, "1000–3000 Hz_10ms_envelope_correlation": 0.9601214025885692, "1000–3000 Hz_10ms_modulation_mae_db": 0.608374336268364, "1000–3000 Hz_40ms_log_rms_mae_db": 1.369673379694033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13837113079170002, "1000–3000 Hz_40ms_envelope_correlation": 0.9817206038225768, "1000–3000 Hz_40ms_modulation_mae_db": 0.5857859377601994, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3648501203274015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21686204521163846, "3000–6000 Hz_10ms_envelope_correlation": 0.9586105752745998, "3000–6000 Hz_10ms_modulation_mae_db": 0.33390329046884454, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4243015593698358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1379518505461165, "3000–6000 Hz_40ms_envelope_correlation": 0.9811491425531779, "3000–6000 Hz_40ms_modulation_mae_db": 0.6089976390762595, "mr_stft_db": 6.1410931702374185, "band_mae_db": 0.18898758771334387, "rms_gain_error_db": -0.02761615953126883, "output_peak": 0.5226766018529986}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 93920, "output_samples": 93920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.112992620483987, "spectral_convergence": 0.2987702479878601, "spectral_convergence_status": "ok", "amplitude_floor": 2.214924872151779e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.183010312232472, "spectral_convergence": 0.2898496141284918, "spectral_convergence_status": "ok", "amplitude_floor": 1.3113821592368984e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.161412746987491, "spectral_convergence": 0.28856272981189485, "spectral_convergence_status": "ok", "amplitude_floor": 7.380112905156557e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00039502697710642294, "output_power": 0.0004032338859045754, "delta_db": 0.08930265669415943, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014539054117940992, "output_power": 0.0014495721962591028, "delta_db": -0.01296302742424039, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0016903939251448888, "output_power": 0.001600317922389526, "delta_db": -0.23781654255023654, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.480236057553453e-05, "output_power": 6.409076079875394e-05, "delta_db": -0.04795399366649565, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.753736279208405e-05, "output_power": 9.782568611307084e-05, "delta_db": 0.012818935711913462, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.937191536111839e-05, "output_power": 5.0773838403753564e-05, "delta_db": 0.12160020662309387, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.08707589377835656, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09980998510666125, "pitch": null, "output_rms_dbfs": -24.358295401725268, "output_peak": 0.5596610448801249, "reference_rms_dbfs": -24.258485416618605, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.15247189323465}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 93920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002156956510733609, "log_rms_mae_db": 0.888044814450776, "rms_envelope_nrmse": 0.09310300619708971, "envelope_correlation": 0.9900064307788392, "rms_gain_error_db": -0.0998104272858597, "reference_mean_power": 0.003754233895192234, "output_mean_power": 0.0036689372468516573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.680528929094005e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006742966926403748, "output_power": 0.0006453629368188031, "reference_active": true, "delta_db": -0.19047009249502994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002358850904334285, "output_power": 0.00021119401033541807, "reference_active": true, "delta_db": -0.4801889431428433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0118677140216256e-05, "output_power": 2.6660407057238464e-05, "reference_active": true, "delta_db": -0.5296911701252749}], "active_band_count": 3, "active_band_mae_db": 0.4001167352543827}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00016458512240675186, "log_rms_mae_db": 0.4476871924267882, "rms_envelope_nrmse": 0.04792191095794252, "envelope_correlation": 0.9971277642972354, "rms_gain_error_db": -0.09980917781722512, "reference_mean_power": 0.003773515049488898, "output_mean_power": 0.003687781391967236}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.4929366354790985e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006129787155244603, "output_power": 0.0005892238612543114, "reference_active": true, "delta_db": -0.1716506901278444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073825851363778, "output_power": 9.59254412027633e-05, "reference_active": true, "delta_db": -0.6236388406597158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.608951147124256e-06, "output_power": 2.1852699946635252e-06, "reference_active": true, "delta_db": -0.7696084435787097}], "active_band_count": 3, "active_band_mae_db": 0.5216326581220899}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002116840322083979, "log_rms_mae_db": 1.6267186487296, "rms_envelope_nrmse": 0.11011317187845918, "envelope_correlation": 0.989487496501855, "rms_gain_error_db": -0.19483370470455522, "reference_mean_power": 0.002333704816993391, "output_mean_power": 0.002231323565626942}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.3033890876287954e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006736793990758906, "output_power": 0.0006373446445982843, "reference_active": true, "delta_db": -0.240789260302835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002058984330594163, "output_power": 0.00017884784060492195, "reference_active": true, "delta_db": -0.6116934065828068}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6276513462408744e-05, "output_power": 1.8364889361734074e-05, "reference_active": true, "delta_db": -1.5557942340641846}], "active_band_count": 3, "active_band_mae_db": 0.8027589669832755}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00015464611695479593, "log_rms_mae_db": 1.0413093635611361, "rms_envelope_nrmse": 0.06320315666364494, "envelope_correlation": 0.996460217269619, "rms_gain_error_db": -0.19483428060237531, "reference_mean_power": 0.002345702993864749, "output_mean_power": 0.0022427950767380453}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.1648344391379404e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377113304270822, "output_power": 0.000603929953440767, "reference_active": true, "delta_db": -0.23637563227821676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.690162436721034e-05, "output_power": 8.053856152556097e-05, "reference_active": true, "delta_db": -0.8032718867313551}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0235059299775393e-06, "output_power": 1.8177121521537073e-06, "reference_active": true, "delta_db": -2.2098571405780683}], "active_band_count": 3, "active_band_mae_db": 1.0831682198625467}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 4.7944659079237455e-05, "log_rms_mae_db": 2.0440335224594133, "rms_envelope_nrmse": 0.22680301955399426, "envelope_correlation": 0.9547638896105776, "rms_gain_error_db": 0.022946497753783407, "reference_mean_power": 0.0001071814340711088, "output_mean_power": 0.000107749239436922}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 8.167234143423064e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.040560289520861e-05, "output_power": 3.516046245802582e-05, "reference_active": true, "delta_db": -0.6038701267113742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.346030743633636e-06, "output_power": 7.223295373436796e-06, "reference_active": true, "delta_db": -0.6274460585622065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1788672389317654e-06, "output_power": 2.2662727941108487e-06, "reference_active": true, "delta_db": 0.17081416293782686}], "active_band_count": 3, "active_band_mae_db": 0.4673767827371358}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 3.945206980281183e-05, "log_rms_mae_db": 1.4099393663919622, "rms_envelope_nrmse": 0.15127376456183309, "envelope_correlation": 0.9778856267625095, "rms_gain_error_db": 0.02293420415327984, "reference_mean_power": 0.00010773182378082642, "output_mean_power": 0.00010830223832406104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 7.380645247762918e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7294287822712847e-05, "output_power": 3.2616541721459514e-05, "reference_active": true, "delta_db": -0.582044066027994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.878816663714596e-06, "output_power": 3.6591269660424825e-06, "reference_active": true, "delta_db": -0.25321773612428217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8761762452219972e-07, "output_power": 1.7348755827315344e-07, "reference_active": true, "delta_db": -0.34005298292622343}], "active_band_count": 3, "active_band_mae_db": 0.3917715950261666}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 5.718269674480757e-05, "log_rms_mae_db": 2.414343924427781, "rms_envelope_nrmse": 0.23303350528639777, "envelope_correlation": 0.95094049013377, "rms_gain_error_db": -0.055777454811446706, "reference_mean_power": 7.041944137047507e-05, "output_mean_power": 6.952081104702418e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 5.931095881530669e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9173785649337023e-05, "output_power": 2.630495414716744e-05, "reference_active": true, "delta_db": -0.4495523860325001}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1741151998348071e-05, "output_power": 8.012323165539934e-06, "reference_active": true, "delta_db": -1.6595225274964218}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.929169683988829e-06, "output_power": 3.0900380596235405e-06, "reference_active": true, "delta_db": 2.045934000440374}], "active_band_count": 3, "active_band_mae_db": 1.3850029713230985}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 4.673628190489769e-05, "log_rms_mae_db": 1.472588148062385, "rms_envelope_nrmse": 0.14108905223534426, "envelope_correlation": 0.9800607036704351, "rms_gain_error_db": -0.05580401460039561, "reference_mean_power": 7.077979037597618e-05, "output_mean_power": 6.987613426177771e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 5.253277550211025e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7931175609497952e-05, "output_power": 2.5425465088888942e-05, "reference_active": true, "delta_db": -0.4082030943828059}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.1737648370747885e-06, "output_power": 3.649285010496107e-06, "reference_active": true, "delta_db": -1.5159890161213059}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7085501560216023e-07, "output_power": 2.3381376369446756e-07, "reference_active": true, "delta_db": 1.3624234035648206}], "active_band_count": 3, "active_band_mae_db": 1.0955385046896442}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8602803905628056, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e1ff788a6c4baa6cca3067ec1ae0216fe3077b5cd495cd8f882438525db50b7a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9559999851490358, "standard_deviation": 0.0038251063883516733, "minimum": 0.9496171652693004, "maximum": 0.9611825862396445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0485034597815304, "standard_deviation": 0.024096490209727792, "minimum": 2.002543025167031, "maximum": 2.080624155575799, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5247663904832747, "standard_deviation": 0.17783699399177733, "minimum": 0.3444409879456825, "maximum": 0.8943196077341634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22528048911727228, "standard_deviation": 0.010047157032590523, "minimum": 0.2137596077605246, "maximum": 0.24364582220532421, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9782409811669532, "standard_deviation": 0.0031346436603097707, "minimum": 0.9718690872393121, "maximum": 0.9817206038225768, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3977717526345164, "standard_deviation": 0.02428154724554309, "minimum": 1.3641346779235448, "maximum": 1.429883662928735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.48741836907983893, "standard_deviation": 0.16208061712039828, "minimum": 0.15469732638803183, "maximum": 0.6906009908305921, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1513020517768053, "standard_deviation": 0.011201733377007072, "minimum": 0.13837113079170002, "maximum": 0.1740719393326302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.960334007598772, "standard_deviation": 0.004730928259634919, "minimum": 0.95094049013377, "maximum": 0.9649373508970469, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3797764419475884, "standard_deviation": 0.03331604103782202, "minimum": 2.344199528459577, "maximum": 2.444502612901991, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8730282942002554, "standard_deviation": 0.3871034165847299, "minimum": 0.27788631478234654, "maximum": 1.3850029713230985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21099149530258576, "standard_deviation": 0.01155837995955253, "minimum": 0.1992236735743593, "maximum": 0.23303350528639777, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9832672479768714, "standard_deviation": 0.0019032027027672891, "minimum": 0.9800607036704351, "maximum": 0.9852011490805198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.422553831377674, "standard_deviation": 0.0354636394734465, "minimum": 1.3807142613762913, "maximum": 1.472588148062385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.962416764561324, "standard_deviation": 0.3343881550398714, "minimum": 0.3115136997477016, "maximum": 1.341468726469544, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.129770525300784, "standard_deviation": 0.006746521826307446, "minimum": 0.12370669680968442, "maximum": 0.14108905223534426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9899580015027114, "standard_deviation": 0.0009167193538451783, "minimum": 0.9889982244330664, "maximum": 0.9917438753214278, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.631978686703131, "standard_deviation": 0.035819013451356164, "minimum": 1.5705754891101387, "maximum": 1.6758609716302209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4250552684618888, "standard_deviation": 0.1990142204375613, "minimum": 0.2111728565810084, "maximum": 0.8027589669832755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10723095939630727, "standard_deviation": 0.004178968151296676, "minimum": 0.09978124491819103, "maximum": 0.11196408554950027, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965001817878686, "standard_deviation": 0.0005966691895207806, "minimum": 0.9957302960041021, "maximum": 0.9975421844640477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9919843428280977, "standard_deviation": 0.04626279934298404, "minimum": 0.9355434252169424, "maximum": 1.0619037262532893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6455424446968614, "standard_deviation": 0.20343041812622617, "minimum": 0.47919250956327447, "maximum": 1.0831682198625467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06191195067328601, "standard_deviation": 0.0033760279398379863, "minimum": 0.056878317116152856, "maximum": 0.06619183958455994, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1450162635221664, "standard_deviation": 0.04442714057405976, "minimum": 0.08241309743676427, "maximum": 0.18898758771334387, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.1361941154180215, "standard_deviation": 0.012772791995130612, "minimum": 6.112038162482321, "maximum": 6.15247189323465, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4761734883552107, "standard_deviation": 0.059794941723743604, "minimum": 0.39563669553770225, "maximum": 0.5596610448801249, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0803566195352143, "standard_deviation": 0.04517716659724415, "minimum": -0.17043893352027742, "maximum": -0.02761615953126883, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.990157132000429, "standard_deviation": 0.0015301434697020753, "minimum": 0.9875814699490413, "maximum": 0.9923024147172838, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9139978167109657, "standard_deviation": 0.033529574352047266, "minimum": 0.882691006458677, "maximum": 0.9885862149991832, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30445762558428596, "standard_deviation": 0.12997470717656043, "minimum": 0.14252687351677668, "maximum": 0.4737287970646135, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09221960047914352, "standard_deviation": 0.00713474779048655, "minimum": 0.08162845301396844, "maximum": 0.10380365633215122, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9965743775139498, "standard_deviation": 0.0012982892544817562, "minimum": 0.9940777500070179, "maximum": 0.9978943079816504, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48285786209308823, "standard_deviation": 0.02693370475127361, "minimum": 0.4476871924267882, "maximum": 0.5226386368082675, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35403884471823543, "standard_deviation": 0.1703204316994659, "minimum": 0.1021038022332124, "maximum": 0.5216326581220899, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0514199017734564, "standard_deviation": 0.008463865123574058, "minimum": 0.04164079220616581, "maximum": 0.06740926624628352, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0020", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.672551155090332, "predicted_mos": 2.672551155090332, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "070618eb799381907bffc961a5f7d24285b8bb8db11db86e90d9c498cff8e7cb", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.743998599704355, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9735061490972299, "全帯域_10ms_rms_envelope_nrmse": 0.0921210311060086, "全帯域_10ms_envelope_correlation": 0.9902358885913183, "全帯域_10ms_modulation_mae_db": 0.2769364766220676, "全帯域_40ms_log_rms_mae_db": 0.5173849116427434, "全帯域_40ms_rms_envelope_nrmse": 0.05995138461321964, "全帯域_40ms_envelope_correlation": 0.9953173523751073, "全帯域_40ms_modulation_mae_db": 0.5044151194605474, "300–1000 Hz_10ms_log_rms_mae_db": 1.3408721564317778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10298995115075901, "300–1000 Hz_10ms_envelope_correlation": 0.9906462085427422, "300–1000 Hz_10ms_modulation_mae_db": 0.34848508071177386, "300–1000 Hz_40ms_log_rms_mae_db": 0.7826161669684952, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05705235818466204, "300–1000 Hz_40ms_envelope_correlation": 0.9969125460307312, "300–1000 Hz_40ms_modulation_mae_db": 0.6470464301265341, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4628010377193776, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19732634758309145, "1000–3000 Hz_10ms_envelope_correlation": 0.9659616528520465, "1000–3000 Hz_10ms_modulation_mae_db": 0.783973476811382, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0954560320086097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13088245188427192, "1000–3000 Hz_40ms_envelope_correlation": 0.9833890132100347, "1000–3000 Hz_40ms_modulation_mae_db": 0.3374597952981679, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5415888812711518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16867131052629955, "3000–6000 Hz_10ms_envelope_correlation": 0.9746776308335592, "3000–6000 Hz_10ms_modulation_mae_db": 0.4382880826871814, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9933536280826331, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10590898455537438, "3000–6000 Hz_40ms_envelope_correlation": 0.9894479737308731, "3000–6000 Hz_40ms_modulation_mae_db": 0.5669188701660798, "mr_stft_db": 5.624965993584688, "band_mae_db": 0.22215566101754214, "rms_gain_error_db": 0.005603266112590684, "output_peak": 0.5785012140736885}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0204750195214274, "全帯域_10ms_rms_envelope_nrmse": 0.09120252562492023, "全帯域_10ms_envelope_correlation": 0.9903615505551231, "全帯域_10ms_modulation_mae_db": 0.06427888702177913, "全帯域_40ms_log_rms_mae_db": 0.44502707946773196, "全帯域_40ms_rms_envelope_nrmse": 0.04065862944843446, "全帯域_40ms_envelope_correlation": 0.9978402483402036, "全帯域_40ms_modulation_mae_db": 0.13999451753179584, "300–1000 Hz_10ms_log_rms_mae_db": 1.3287563013053498, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10065989147934068, "300–1000 Hz_10ms_envelope_correlation": 0.9910067599989596, "300–1000 Hz_10ms_modulation_mae_db": 0.1707356259120794, "300–1000 Hz_40ms_log_rms_mae_db": 0.7736024175312586, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04970546986810134, "300–1000 Hz_40ms_envelope_correlation": 0.9975783818847913, "300–1000 Hz_40ms_modulation_mae_db": 0.2803246545800641, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4636337881050245, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18754378530182841, "1000–3000 Hz_10ms_envelope_correlation": 0.9689725896075274, "1000–3000 Hz_10ms_modulation_mae_db": 0.7530950497073104, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0556921440701486, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12142478267355179, "1000–3000 Hz_40ms_envelope_correlation": 0.9856618403206912, "1000–3000 Hz_40ms_modulation_mae_db": 0.4635348197995393, "3000–6000 Hz_10ms_log_rms_mae_db": 1.483064004301535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1786712346439256, "3000–6000 Hz_10ms_envelope_correlation": 0.9711374803223959, "3000–6000 Hz_10ms_modulation_mae_db": 0.8121156493358136, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8745754601008356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10116244682429425, "3000–6000 Hz_40ms_envelope_correlation": 0.989927517684834, "3000–6000 Hz_40ms_modulation_mae_db": 0.6298136383467653, "mr_stft_db": 5.611915390624216, "band_mae_db": 0.09667987375398913, "rms_gain_error_db": -0.02765031288401387, "output_peak": 0.44677254508006436}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.86426215998745, "全帯域_10ms_rms_envelope_nrmse": 0.08417112740840807, "全帯域_10ms_envelope_correlation": 0.9919450518902765, "全帯域_10ms_modulation_mae_db": 0.2138403702239643, "全帯域_40ms_log_rms_mae_db": 0.438706178407415, "全帯域_40ms_rms_envelope_nrmse": 0.05077799441040008, "全帯域_40ms_envelope_correlation": 0.9968666152460945, "全帯域_40ms_modulation_mae_db": 0.5649408525207954, "300–1000 Hz_10ms_log_rms_mae_db": 1.3624447219181706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09826311736908087, "300–1000 Hz_10ms_envelope_correlation": 0.991484888935624, "300–1000 Hz_10ms_modulation_mae_db": 0.5424892627833672, "300–1000 Hz_40ms_log_rms_mae_db": 0.8216938247623554, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05268309810390334, "300–1000 Hz_40ms_envelope_correlation": 0.997373378697274, "300–1000 Hz_40ms_modulation_mae_db": 0.9730860639795317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4691403993444652, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18536190006009065, "1000–3000 Hz_10ms_envelope_correlation": 0.9697800964704406, "1000–3000 Hz_10ms_modulation_mae_db": 0.4455499402065861, "1000–3000 Hz_40ms_log_rms_mae_db": 1.044579432696601, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11907958477928032, "1000–3000 Hz_40ms_envelope_correlation": 0.9861883785376859, "1000–3000 Hz_40ms_modulation_mae_db": 0.2829237480295775, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4793536018770692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16510989408907054, "3000–6000 Hz_10ms_envelope_correlation": 0.976018898984613, "3000–6000 Hz_10ms_modulation_mae_db": 0.30818530846191494, "3000–6000 Hz_40ms_log_rms_mae_db": 0.915431168689871, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10439474195651267, "3000–6000 Hz_40ms_envelope_correlation": 0.9892329348722851, "3000–6000 Hz_40ms_modulation_mae_db": 0.33471673172083555, "mr_stft_db": 5.608218340299942, "band_mae_db": 0.1789569546445252, "rms_gain_error_db": -0.13109136604581492, "output_peak": 0.4042945504052082}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9507512053234053, "全帯域_10ms_rms_envelope_nrmse": 0.09105141355532437, "全帯域_10ms_envelope_correlation": 0.9904516321323608, "全帯域_10ms_modulation_mae_db": 0.0954476142792743, "全帯域_40ms_log_rms_mae_db": 0.4880348521904394, "全帯域_40ms_rms_envelope_nrmse": 0.056438518037401485, "全帯域_40ms_envelope_correlation": 0.99585192850085, "全帯域_40ms_modulation_mae_db": 0.2919363542720548, "300–1000 Hz_10ms_log_rms_mae_db": 1.2990557220953847, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10301006313044582, "300–1000 Hz_10ms_envelope_correlation": 0.9906070701178793, "300–1000 Hz_10ms_modulation_mae_db": 0.18973288629762405, "300–1000 Hz_40ms_log_rms_mae_db": 0.7771549243793077, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06188084762343938, "300–1000 Hz_40ms_envelope_correlation": 0.9962790245769713, "300–1000 Hz_40ms_modulation_mae_db": 0.5452534596641239, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5104615251234104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19564363317263078, "1000–3000 Hz_10ms_envelope_correlation": 0.96607829831874, "1000–3000 Hz_10ms_modulation_mae_db": 0.4063097197208491, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0518594813128597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12059741158327866, "1000–3000 Hz_40ms_envelope_correlation": 0.985896017014963, "1000–3000 Hz_40ms_modulation_mae_db": 0.30469837572372105, "3000–6000 Hz_10ms_log_rms_mae_db": 1.454793238104375, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16387822176837044, "3000–6000 Hz_10ms_envelope_correlation": 0.9763394600759265, "3000–6000 Hz_10ms_modulation_mae_db": 0.5683374871117342, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8763563840779354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10221018873394223, "3000–6000 Hz_40ms_envelope_correlation": 0.9906252949363709, "3000–6000 Hz_40ms_modulation_mae_db": 0.7406772949479858, "mr_stft_db": 5.608214172044008, "band_mae_db": 0.23043378543791468, "rms_gain_error_db": -0.0031053974180694586, "output_peak": 0.43016191892444056}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9431636941035663, "全帯域_10ms_rms_envelope_nrmse": 0.10412942867248352, "全帯域_10ms_envelope_correlation": 0.9874558928165144, "全帯域_10ms_modulation_mae_db": 0.12159598452981506, "全帯域_40ms_log_rms_mae_db": 0.5022918280678492, "全帯域_40ms_rms_envelope_nrmse": 0.0634126763944262, "全帯域_40ms_envelope_correlation": 0.9947779672247921, "全帯域_40ms_modulation_mae_db": 0.2290613184699736, "300–1000 Hz_10ms_log_rms_mae_db": 1.3697833555815144, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10756548636943088, "300–1000 Hz_10ms_envelope_correlation": 0.9897228552606167, "300–1000 Hz_10ms_modulation_mae_db": 0.14439205175374664, "300–1000 Hz_40ms_log_rms_mae_db": 0.7914584840440161, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062153349939374235, "300–1000 Hz_40ms_envelope_correlation": 0.9961759528292577, "300–1000 Hz_40ms_modulation_mae_db": 0.20066192569404626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4665967448920132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18116373002117137, "1000–3000 Hz_10ms_envelope_correlation": 0.9711563369086627, "1000–3000 Hz_10ms_modulation_mae_db": 0.4724024443321184, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0547653605670744, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11710735472882644, "1000–3000 Hz_40ms_envelope_correlation": 0.9866468127273921, "1000–3000 Hz_40ms_modulation_mae_db": 0.3130729320747955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4449771331183032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1530200289429258, "3000–6000 Hz_10ms_envelope_correlation": 0.9796437634397688, "3000–6000 Hz_10ms_modulation_mae_db": 0.6464677270833639, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8936361344225814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09405493805129811, "3000–6000 Hz_40ms_envelope_correlation": 0.9923695488384608, "3000–6000 Hz_40ms_modulation_mae_db": 0.8077249735272914, "mr_stft_db": 5.630579476690084, "band_mae_db": 0.1645228386231988, "rms_gain_error_db": -0.06600540866029349, "output_peak": 0.43940496316193967}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9270261680676974, "全帯域_10ms_rms_envelope_nrmse": 0.09467764775320785, "全帯域_10ms_envelope_correlation": 0.9897711629429565, "全帯域_10ms_modulation_mae_db": 0.20182815852200875, "全帯域_40ms_log_rms_mae_db": 0.46307474389218706, "全帯域_40ms_rms_envelope_nrmse": 0.051214267904859195, "全帯域_40ms_envelope_correlation": 0.9968861812358029, "全帯域_40ms_modulation_mae_db": 0.2148106208364328, "300–1000 Hz_10ms_log_rms_mae_db": 1.342227389972773, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10278830275510248, "300–1000 Hz_10ms_envelope_correlation": 0.9910195101482824, "300–1000 Hz_10ms_modulation_mae_db": 0.4821957995861161, "300–1000 Hz_40ms_log_rms_mae_db": 0.799684412059107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0646732293684709, "300–1000 Hz_40ms_envelope_correlation": 0.9964420106330224, "300–1000 Hz_40ms_modulation_mae_db": 0.45739171456143835, "1000–3000 Hz_10ms_log_rms_mae_db": 1.458215082027663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18460724708065285, "1000–3000 Hz_10ms_envelope_correlation": 0.9703378348896953, "1000–3000 Hz_10ms_modulation_mae_db": 0.34264621651489513, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0319831171200575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11719703008577294, "1000–3000 Hz_40ms_envelope_correlation": 0.9867266864774242, "1000–3000 Hz_40ms_modulation_mae_db": 0.3731442791556397, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4306034782193575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15704299347151077, "3000–6000 Hz_10ms_envelope_correlation": 0.977800548095421, "3000–6000 Hz_10ms_modulation_mae_db": 0.9150831340871749, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8621280034438311, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08755246430209065, "3000–6000 Hz_40ms_envelope_correlation": 0.992370073550236, "3000–6000 Hz_40ms_modulation_mae_db": 0.5483391542271395, "mr_stft_db": 5.607062146405483, "band_mae_db": 0.18377407353193098, "rms_gain_error_db": -0.1440263474434332, "output_peak": 0.4002272616720864}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9099336057314807, "全帯域_10ms_rms_envelope_nrmse": 0.09566381122281416, "全帯域_10ms_envelope_correlation": 0.9894219523862375, "全帯域_10ms_modulation_mae_db": 0.2566090952557665, "全帯域_40ms_log_rms_mae_db": 0.47820571550089475, "全帯域_40ms_rms_envelope_nrmse": 0.05122198173283194, "全帯域_40ms_envelope_correlation": 0.9966666671997464, "全帯域_40ms_modulation_mae_db": 0.29339213322354357, "300–1000 Hz_10ms_log_rms_mae_db": 1.284880778340341, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11099142206322551, "300–1000 Hz_10ms_envelope_correlation": 0.9891236817904623, "300–1000 Hz_10ms_modulation_mae_db": 0.3310720701236053, "300–1000 Hz_40ms_log_rms_mae_db": 0.7625401166799076, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058773521963718815, "300–1000 Hz_40ms_envelope_correlation": 0.9967473933679731, "300–1000 Hz_40ms_modulation_mae_db": 0.6520035583065771, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4476560539000367, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19353516468234017, "1000–3000 Hz_10ms_envelope_correlation": 0.9668112860883, "1000–3000 Hz_10ms_modulation_mae_db": 0.8458895678198655, "1000–3000 Hz_40ms_log_rms_mae_db": 1.048116838453991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.125500297489657, "1000–3000 Hz_40ms_envelope_correlation": 0.9847078560668789, "1000–3000 Hz_40ms_modulation_mae_db": 0.5395597988530324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5190091940648494, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1809037874275549, "3000–6000 Hz_10ms_envelope_correlation": 0.9721888816787763, "3000–6000 Hz_10ms_modulation_mae_db": 0.27957133871738205, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9511713823003872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10997639432805419, "3000–6000 Hz_40ms_envelope_correlation": 0.9888188350403542, "3000–6000 Hz_40ms_modulation_mae_db": 0.324791411924252, "mr_stft_db": 5.6123383249421, "band_mae_db": 0.19135284417866327, "rms_gain_error_db": -0.08911272498752836, "output_peak": 0.44344665665607036}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9503733406685514, "全帯域_10ms_rms_envelope_nrmse": 0.09108651186046884, "全帯域_10ms_envelope_correlation": 0.9903987785731343, "全帯域_10ms_modulation_mae_db": 0.0766507039028959, "全帯域_40ms_log_rms_mae_db": 0.4808060850455415, "全帯域_40ms_rms_envelope_nrmse": 0.04855416959297112, "全帯域_40ms_envelope_correlation": 0.9969556820562165, "全帯域_40ms_modulation_mae_db": 0.10172883317415309, "300–1000 Hz_10ms_log_rms_mae_db": 1.3289856916658256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09693890417372122, "300–1000 Hz_10ms_envelope_correlation": 0.991859967719337, "300–1000 Hz_10ms_modulation_mae_db": 0.5124073255394642, "300–1000 Hz_40ms_log_rms_mae_db": 0.7550958093981708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0574577426730607, "300–1000 Hz_40ms_envelope_correlation": 0.9970451348521905, "300–1000 Hz_40ms_modulation_mae_db": 0.5137371310452831, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4609547533991853, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20301060064621773, "1000–3000 Hz_10ms_envelope_correlation": 0.96339187645193, "1000–3000 Hz_10ms_modulation_mae_db": 0.7432606922897035, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1016963159817812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13906928166602375, "1000–3000 Hz_40ms_envelope_correlation": 0.9814132865549304, "1000–3000 Hz_40ms_modulation_mae_db": 0.5184018383719642, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4772818796765845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18570315280532085, "3000–6000 Hz_10ms_envelope_correlation": 0.9697818120940594, "3000–6000 Hz_10ms_modulation_mae_db": 0.7633569187561914, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9053876912366294, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11063029676308556, "3000–6000 Hz_40ms_envelope_correlation": 0.9879533903952081, "3000–6000 Hz_40ms_modulation_mae_db": 0.780311621199188, "mr_stft_db": 5.617852762168053, "band_mae_db": 0.22179111516797878, "rms_gain_error_db": -0.06315296365738873, "output_peak": 0.4702105818512839}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 93920, "output_samples": 93920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.472890547867673, "spectral_convergence": 0.29701111083199105, "spectral_convergence_status": "ok", "amplitude_floor": 2.214924872151779e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.512831498863262, "spectral_convergence": 0.276125745877813, "spectral_convergence_status": "ok", "amplitude_floor": 1.3113821592368984e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.889175934023128, "spectral_convergence": 0.3008304752801322, "spectral_convergence_status": "ok", "amplitude_floor": 7.380112905156557e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00039502697710642294, "output_power": 0.0004600068092708314, "delta_db": 0.6613750499759307, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014539054117940992, "output_power": 0.001473634483304064, "delta_db": 0.058536224349302855, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0016903939251448888, "output_power": 0.0016181863645576074, "delta_db": -0.18959385976245813, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.480236057553453e-05, "output_power": 6.241858147877206e-05, "delta_db": -0.16276931648406023, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.753736279208405e-05, "output_power": 9.272604169339249e-05, "delta_db": -0.21969288057985603, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.937191536111839e-05, "output_power": 4.890838388007928e-05, "delta_db": -0.04096663495364475, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22215566101754214, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.005603266112590684, "pitch": null, "output_rms_dbfs": -24.252882150506014, "output_peak": 0.5785012140736885, "reference_rms_dbfs": -24.258485416618605, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.624965993584688}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 93920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002156956510733609, "log_rms_mae_db": 0.9735061490972299, "rms_envelope_nrmse": 0.0921210311060086, "envelope_correlation": 0.9902358885913183, "rms_gain_error_db": 0.0056017610091661655, "reference_mean_power": 0.003754233895192234, "output_mean_power": 0.003759079429910356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.680528929094005e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006742966926403748, "output_power": 0.0006798317997273829, "reference_active": true, "delta_db": 0.0355044571691617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002358850904334285, "output_power": 0.00022744090787591204, "reference_active": true, "delta_db": -0.158319110836933}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0118677140216256e-05, "output_power": 2.6009817614940403e-05, "reference_active": true, "delta_db": -0.6369858618601081}], "active_band_count": 3, "active_band_mae_db": 0.2769364766220676}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00016458512240675186, "log_rms_mae_db": 0.5173849116427434, "rms_envelope_nrmse": 0.05995138461321964, "envelope_correlation": 0.9953173523751073, "rms_gain_error_db": 0.005601908260683391, "reference_mean_power": 0.003773515049488898, "output_mean_power": 0.003778385598220169}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.4929366354790985e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006129787155244603, "output_power": 0.0006236691937391792, "reference_active": true, "delta_db": 0.07508897758865946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073825851363778, "output_power": 0.00010647022828635583, "reference_active": true, "delta_db": -0.17069504196360832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.608951147124256e-06, "output_power": 1.948587054997378e-06, "reference_active": true, "delta_db": -1.2674613388293743}], "active_band_count": 3, "active_band_mae_db": 0.5044151194605474}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002116840322083979, "log_rms_mae_db": 1.3408721564317778, "rms_envelope_nrmse": 0.10298995115075901, "envelope_correlation": 0.9906462085427422, "rms_gain_error_db": -0.12177395416994569, "reference_mean_power": 0.002333704816993391, "output_mean_power": 0.0022691778059006945}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.3033890876287954e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006736793990758906, "output_power": 0.000676980867715113, "reference_active": true, "delta_db": 0.021231282255376318}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002058984330594163, "output_power": 0.00018832755169210442, "reference_active": true, "delta_db": -0.38739181039214976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6276513462408744e-05, "output_power": 2.2692613942127973e-05, "reference_active": true, "delta_db": -0.6368321494877955}], "active_band_count": 3, "active_band_mae_db": 0.34848508071177386}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00015464611695479593, "log_rms_mae_db": 0.7826161669684952, "rms_envelope_nrmse": 0.05705235818466204, "envelope_correlation": 0.9969125460307312, "rms_gain_error_db": -0.1217740687671866, "reference_mean_power": 0.002345702993864749, "output_mean_power": 0.0022808441726221342}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.1648344391379404e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377113304270822, "output_power": 0.0006416900224727399, "reference_active": true, "delta_db": 0.02701153265052705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.690162436721034e-05, "output_power": 8.89852241204525e-05, "reference_active": true, "delta_db": -0.37013158610483815}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0235059299775393e-06, "output_power": 2.1189036425657062e-06, "reference_active": true, "delta_db": -1.543996171624237}], "active_band_count": 3, "active_band_mae_db": 0.6470464301265341}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 4.7944659079237455e-05, "log_rms_mae_db": 1.4628010377193776, "rms_envelope_nrmse": 0.19732634758309145, "envelope_correlation": 0.9659616528520465, "rms_gain_error_db": 0.015626752144309055, "reference_mean_power": 0.0001071814340711088, "output_mean_power": 0.00010756778819125718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 8.167234143423064e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.040560289520861e-05, "output_power": 3.7417985764722885e-05, "reference_active": true, "delta_db": -0.3336118586278424}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.346030743633636e-06, "output_power": 7.459248179289393e-06, "reference_active": true, "delta_db": -0.487849231980111}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1788672389317654e-06, "output_power": 3.09939339315854e-06, "reference_active": true, "delta_db": 1.5304593398261923}], "active_band_count": 3, "active_band_mae_db": 0.783973476811382}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 3.945206980281183e-05, "log_rms_mae_db": 1.0954560320086097, "rms_envelope_nrmse": 0.13088245188427192, "envelope_correlation": 0.9833890132100347, "rms_gain_error_db": 0.015635349496311447, "reference_mean_power": 0.00010773182378082642, "output_mean_power": 0.00010812037591274665}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 7.380645247762918e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7294287822712847e-05, "output_power": 3.475682736751604e-05, "reference_active": true, "delta_db": -0.30602191395239975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.878816663714596e-06, "output_power": 3.5773612977943805e-06, "reference_active": true, "delta_db": -0.3513644850572062}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8761762452219972e-07, "output_power": 2.0359773003580312e-07, "reference_active": true, "delta_db": 0.35499298688489755}], "active_band_count": 3, "active_band_mae_db": 0.3374597952981679}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 5.718269674480757e-05, "log_rms_mae_db": 1.5415888812711518, "rms_envelope_nrmse": 0.16867131052629955, "envelope_correlation": 0.9746776308335592, "rms_gain_error_db": -0.3072851176174138, "reference_mean_power": 7.041944137047507e-05, "output_mean_power": 6.560909750137925e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 5.931095881530669e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9173785649337023e-05, "output_power": 2.6186252030194636e-05, "reference_active": true, "delta_db": -0.46919444268199134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1741151998348071e-05, "output_power": 1.0603891973635151e-05, "reference_active": true, "delta_db": -0.442454156275307}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.929169683988829e-06, "output_power": 1.7581214138625356e-06, "reference_active": true, "delta_db": -0.4032156491042459}], "active_band_count": 3, "active_band_mae_db": 0.4382880826871814}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 4.673628190489769e-05, "log_rms_mae_db": 0.9933536280826331, "rms_envelope_nrmse": 0.10590898455537438, "envelope_correlation": 0.9894479737308731, "rms_gain_error_db": -0.30728174123020624, "reference_mean_power": 7.077979037597618e-05, "output_mean_power": 6.59448823764204e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 5.253277550211025e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7931175609497952e-05, "output_power": 2.5320791010096994e-05, "reference_active": true, "delta_db": -0.4261194660660056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.1737648370747885e-06, "output_power": 4.70904349628517e-06, "reference_active": true, "delta_db": -0.4087398292614355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7085501560216023e-07, "output_power": 1.3997097922774885e-07, "reference_active": true, "delta_db": -0.8658973151707984}], "active_band_count": 3, "active_band_mae_db": 0.5669188701660798}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8713669043832394, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5e11d8e06f1927245749da8da58d1076e18057d67e5621fec6d8539d7f9dd87e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9678112464484179, "standard_deviation": 0.002666008928845903, "minimum": 0.96339187645193, "maximum": 0.9711563369086627, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4674324230638969, "standard_deviation": 0.01855117585860728, "minimum": 1.4476560539000367, "maximum": 1.5104615251234104, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5991408884253389, "standard_deviation": 0.20076517451371498, "minimum": 0.34264621651489513, "maximum": 0.8458895678198655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19102405106850293, "standard_deviation": 0.007498724186333069, "minimum": 0.18116373002117137, "maximum": 0.20301060064621773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9850787363637501, "standard_deviation": 0.001845539237218096, "minimum": 0.9814132865549304, "maximum": 0.9867266864774242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0605185902763903, "standard_deviation": 0.024697419625326875, "minimum": 1.0319831171200575, "maximum": 1.1016963159817812, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3915994484133047, "standard_deviation": 0.10138055774898219, "minimum": 0.2829237480295775, "maximum": 0.5395597988530324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12385727436133284, "standard_deviation": 0.007678803089467693, "minimum": 0.11710735472882644, "maximum": 0.13906928166602375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.974698559440565, "standard_deviation": 0.0034156747303558626, "minimum": 0.9697818120940594, "maximum": 0.9796437634397688, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4788339263291532, "standard_deviation": 0.037077567804847326, "minimum": 1.4306034782193575, "maximum": 1.5415888812711518, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5914257057800946, "standard_deviation": 0.23528420946543754, "minimum": 0.27957133871738205, "maximum": 0.9150831340871749, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1691250779593723, "standard_deviation": 0.011673572996048076, "minimum": 0.1530200289429258, "maximum": 0.18570315280532085, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900931961310777, "standard_deviation": 0.0016051549579109666, "minimum": 0.9879533903952081, "maximum": 0.992370073550236, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9090049815443381, "standard_deviation": 0.04412976902111827, "minimum": 0.8621280034438311, "maximum": 0.9933536280826331, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5916617120074421, "standard_deviation": 0.18733578212669771, "minimum": 0.324791411924252, "maximum": 0.8077249735272914, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1019863069393315, "standard_deviation": 0.007853291482154658, "minimum": 0.08755246430209065, "maximum": 0.11063029676308556, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990683867814238, "standard_deviation": 0.0008949702809859156, "minimum": 0.9891236817904623, "maximum": 0.991859967719337, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3321257646638922, "standard_deviation": 0.028940345825501316, "minimum": 1.284880778340341, "maximum": 1.3697833555815144, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3401887628384721, "standard_deviation": 0.1606194951476377, "minimum": 0.14439205175374664, "maximum": 0.5424892627833672, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1029008923113883, "standard_deviation": 0.0046186574926810765, "minimum": 0.09693890417372122, "maximum": 0.11099142206322551, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968192278590264, "standard_deviation": 0.0005065592077455284, "minimum": 0.9961759528292577, "maximum": 0.9975783818847913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7829807694778272, "standard_deviation": 0.021270375596107657, "minimum": 0.7550958093981708, "maximum": 0.8216938247623554, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5336881172446998, "standard_deviation": 0.2394552254055113, "minimum": 0.20066192569404626, "maximum": 0.9730860639795317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05804745221559134, "standard_deviation": 0.005009766898286988, "minimum": 0.04970546986810134, "maximum": 0.0646732293684709, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18620839329446787, "standard_deviation": 0.043234931653424805, "minimum": 0.09667987375398913, "maximum": 0.23043378543791468, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.615143325844821, "standard_deviation": 0.008632680876519253, "minimum": 5.607062146405483, "maximum": 5.630579476690084, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45162746147809774, "standard_deviation": 0.05609344402003654, "minimum": 0.4002272616720864, "maximum": 0.5785012140736885, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06481765687299391, "standard_deviation": 0.055303802894655266, "minimum": -0.1440263474434332, "maximum": 0.005603266112590684, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900052387359901, "standard_deviation": 0.0012644226389565921, "minimum": 0.9874558928165144, "maximum": 0.9919450518902765, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.942436417812601, "standard_deviation": 0.0456588813176215, "minimum": 0.86426215998745, "maximum": 1.0204750195214274, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.16339841129469646, "standard_deviation": 0.08393446346800854, "minimum": 0.06427888702177913, "maximum": 0.2769364766220676, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09301293715045446, "standard_deviation": 0.005646543234687591, "minimum": 0.08417112740840807, "maximum": 0.10412942867248352, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963953302723517, "standard_deviation": 0.0010008307497331543, "minimum": 0.9947779672247921, "maximum": 0.9978402483402036, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4766914242768503, "standard_deviation": 0.02696601320417444, "minimum": 0.438706178407415, "maximum": 0.5173849116427434, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.292534968686162, "standard_deviation": 0.16424929050423082, "minimum": 0.10172883317415309, "maximum": 0.5649408525207954, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.052778702766818016, "standard_deviation": 0.007086602412999326, "minimum": 0.04065862944843446, "maximum": 0.0634126763944262, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0020", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.409708261489868, "predicted_mos": 2.409708261489868, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "1ce75f7082bb3a1b3a947b33a2ebd1915f02fa7d220da1ad6db01f7a42f95903", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.88927310006693, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8855576389296078, "全帯域_10ms_rms_envelope_nrmse": 0.08315659193044429, "全帯域_10ms_envelope_correlation": 0.9922411464246534, "全帯域_10ms_modulation_mae_db": 0.2847892767861273, "全帯域_40ms_log_rms_mae_db": 0.4641376214236186, "全帯域_40ms_rms_envelope_nrmse": 0.049586714265995896, "全帯域_40ms_envelope_correlation": 0.9971623920322136, "全帯域_40ms_modulation_mae_db": 0.4554298837160018, "300–1000 Hz_10ms_log_rms_mae_db": 1.4129666607894271, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09794882694478332, "300–1000 Hz_10ms_envelope_correlation": 0.9917668819181605, "300–1000 Hz_10ms_modulation_mae_db": 0.5803404392833668, "300–1000 Hz_40ms_log_rms_mae_db": 0.8332639234671907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05553389007287119, "300–1000 Hz_40ms_envelope_correlation": 0.9973761585542487, "300–1000 Hz_40ms_modulation_mae_db": 0.7396225452147288, "1000–3000 Hz_10ms_log_rms_mae_db": 1.665273887372956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19928754112359903, "1000–3000 Hz_10ms_envelope_correlation": 0.9659455115301111, "1000–3000 Hz_10ms_modulation_mae_db": 0.4389476645800329, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1569060881024582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.132916867198244, "1000–3000 Hz_40ms_envelope_correlation": 0.9833665969318681, "1000–3000 Hz_40ms_modulation_mae_db": 0.2148873069843357, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7373219642242776, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18130711295404056, "3000–6000 Hz_10ms_envelope_correlation": 0.9716234781610579, "3000–6000 Hz_10ms_modulation_mae_db": 0.49019977388176234, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0567426948832066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10301436794005414, "3000–6000 Hz_40ms_envelope_correlation": 0.9899465868441275, "3000–6000 Hz_40ms_modulation_mae_db": 0.43875558862623704, "mr_stft_db": 5.758701485879574, "band_mae_db": 0.24034992790339552, "rms_gain_error_db": -0.15543597272188547, "output_peak": 0.4195045917015632}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8289946680452983, "全帯域_10ms_rms_envelope_nrmse": 0.08469905386378618, "全帯域_10ms_envelope_correlation": 0.991789779573346, "全帯域_10ms_modulation_mae_db": 0.24821780147814024, "全帯域_40ms_log_rms_mae_db": 0.4256228697954521, "全帯域_40ms_rms_envelope_nrmse": 0.050294888350641644, "全帯域_40ms_envelope_correlation": 0.9968713544290881, "全帯域_40ms_modulation_mae_db": 0.4565670915766436, "300–1000 Hz_10ms_log_rms_mae_db": 1.3670874096841514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10093263284734647, "300–1000 Hz_10ms_envelope_correlation": 0.9910805737228309, "300–1000 Hz_10ms_modulation_mae_db": 0.37283378972863906, "300–1000 Hz_40ms_log_rms_mae_db": 0.7867637714275827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0598305284525829, "300–1000 Hz_40ms_envelope_correlation": 0.9966782837851692, "300–1000 Hz_40ms_modulation_mae_db": 0.5012093175763717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6754853862934171, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20534720162060824, "1000–3000 Hz_10ms_envelope_correlation": 0.9636703655227971, "1000–3000 Hz_10ms_modulation_mae_db": 0.3794696092821614, "1000–3000 Hz_40ms_log_rms_mae_db": 1.196724210540632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13607695708398582, "1000–3000 Hz_40ms_envelope_correlation": 0.9824067101492003, "1000–3000 Hz_40ms_modulation_mae_db": 0.16484021497541726, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7251083264271512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16452084260066097, "3000–6000 Hz_10ms_envelope_correlation": 0.9756014347494589, "3000–6000 Hz_10ms_modulation_mae_db": 1.0458084933630198, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0543518970984578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09997836320804404, "3000–6000 Hz_40ms_envelope_correlation": 0.9898505699296419, "3000–6000 Hz_40ms_modulation_mae_db": 0.6416509330677892, "mr_stft_db": 5.745438918664926, "band_mae_db": 0.15637524289507362, "rms_gain_error_db": -0.107463511113456, "output_peak": 0.41808852224908427}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8837863207781008, "全帯域_10ms_rms_envelope_nrmse": 0.08358026651812958, "全帯域_10ms_envelope_correlation": 0.991960364796225, "全帯域_10ms_modulation_mae_db": 0.15998492222117505, "全帯域_40ms_log_rms_mae_db": 0.45002556630404406, "全帯域_40ms_rms_envelope_nrmse": 0.04523635284714505, "全帯域_40ms_envelope_correlation": 0.997432503813554, "全帯域_40ms_modulation_mae_db": 0.09730456372050182, "300–1000 Hz_10ms_log_rms_mae_db": 1.3663127158479578, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10644338132325774, "300–1000 Hz_10ms_envelope_correlation": 0.9899866924829117, "300–1000 Hz_10ms_modulation_mae_db": 0.08218854433772449, "300–1000 Hz_40ms_log_rms_mae_db": 0.8202831450758554, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06044497376688659, "300–1000 Hz_40ms_envelope_correlation": 0.9965149585677031, "300–1000 Hz_40ms_modulation_mae_db": 0.11146658788216772, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6731941253740057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18979996678213287, "1000–3000 Hz_10ms_envelope_correlation": 0.9680959875082399, "1000–3000 Hz_10ms_modulation_mae_db": 0.7803566456659086, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1822201189671657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12612825262536853, "1000–3000 Hz_40ms_envelope_correlation": 0.9849165096361149, "1000–3000 Hz_40ms_modulation_mae_db": 0.49209101341085715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7564177209217353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1789207263918424, "3000–6000 Hz_10ms_envelope_correlation": 0.9718063403884041, "3000–6000 Hz_10ms_modulation_mae_db": 0.674268690200514, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1031913994514355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1118929031481525, "3000–6000 Hz_40ms_envelope_correlation": 0.9887912289057398, "3000–6000 Hz_40ms_modulation_mae_db": 0.9124715458149346, "mr_stft_db": 5.7699197943932825, "band_mae_db": 0.21444313422505687, "rms_gain_error_db": -0.08989301074571285, "output_peak": 0.4093394921146829}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8887873571074173, "全帯域_10ms_rms_envelope_nrmse": 0.07919997407353144, "全帯域_10ms_envelope_correlation": 0.9928192267506422, "全帯域_10ms_modulation_mae_db": 0.16931255464375847, "全帯域_40ms_log_rms_mae_db": 0.4555114483285723, "全帯域_40ms_rms_envelope_nrmse": 0.04174636988462798, "全帯域_40ms_envelope_correlation": 0.9978612860490866, "全帯域_40ms_modulation_mae_db": 0.22266820910253027, "300–1000 Hz_10ms_log_rms_mae_db": 1.4286643570652786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10116020527986545, "300–1000 Hz_10ms_envelope_correlation": 0.9910376537734815, "300–1000 Hz_10ms_modulation_mae_db": 0.2898374131378201, "300–1000 Hz_40ms_log_rms_mae_db": 0.8545136799163277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0553945451208031, "300–1000 Hz_40ms_envelope_correlation": 0.9971938666686845, "300–1000 Hz_40ms_modulation_mae_db": 0.38373323884938837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.680691272985994, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19190919873327839, "1000–3000 Hz_10ms_envelope_correlation": 0.9677815283108023, "1000–3000 Hz_10ms_modulation_mae_db": 0.3168724196177699, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1742917235602888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12703285155969185, "1000–3000 Hz_40ms_envelope_correlation": 0.9843636769045327, "1000–3000 Hz_40ms_modulation_mae_db": 0.10424971795002218, "3000–6000 Hz_10ms_log_rms_mae_db": 1.726244998697554, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17498261072090782, "3000–6000 Hz_10ms_envelope_correlation": 0.9732261708387991, "3000–6000 Hz_10ms_modulation_mae_db": 0.6465970155196644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0548653759623887, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11305129851515662, "3000–6000 Hz_40ms_envelope_correlation": 0.9886390438284384, "3000–6000 Hz_40ms_modulation_mae_db": 0.6816283847522194, "mr_stft_db": 5.756529083649201, "band_mae_db": 0.18129361075916095, "rms_gain_error_db": -0.09891485176700644, "output_peak": 0.4097535488141505}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8874245360111199, "全帯域_10ms_rms_envelope_nrmse": 0.07830699347744839, "全帯域_10ms_envelope_correlation": 0.9929534198912428, "全帯域_10ms_modulation_mae_db": 0.28253364743876636, "全帯域_40ms_log_rms_mae_db": 0.45276732759945365, "全帯域_40ms_rms_envelope_nrmse": 0.04003033698070453, "全帯域_40ms_envelope_correlation": 0.998007290312868, "全帯域_40ms_modulation_mae_db": 0.590917370364688, "300–1000 Hz_10ms_log_rms_mae_db": 1.472655788270505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10584417808458425, "300–1000 Hz_10ms_envelope_correlation": 0.9901358008239454, "300–1000 Hz_10ms_modulation_mae_db": 0.21477486640590163, "300–1000 Hz_40ms_log_rms_mae_db": 0.8937312288618904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058258925142678865, "300–1000 Hz_40ms_envelope_correlation": 0.996820126931903, "300–1000 Hz_40ms_modulation_mae_db": 0.5530365275532975, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6815306699058652, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2243694904149324, "1000–3000 Hz_10ms_envelope_correlation": 0.9568863844937602, "1000–3000 Hz_10ms_modulation_mae_db": 0.257911707973876, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1447918657640077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15568903423210848, "1000–3000 Hz_40ms_envelope_correlation": 0.9770922562506296, "1000–3000 Hz_40ms_modulation_mae_db": 0.28876098390043575, "3000–6000 Hz_10ms_log_rms_mae_db": 1.754446948201932, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1800008372070723, "3000–6000 Hz_10ms_envelope_correlation": 0.9707604331267593, "3000–6000 Hz_10ms_modulation_mae_db": 0.41505602252306556, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0684455756763949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11033272675991086, "3000–6000 Hz_40ms_envelope_correlation": 0.9875327984187263, "3000–6000 Hz_40ms_modulation_mae_db": 0.44848483030364283, "mr_stft_db": 5.754936840331298, "band_mae_db": 0.12933417370371117, "rms_gain_error_db": -0.08596333481779031, "output_peak": 0.39965627439000123}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9258624614015543, "全帯域_10ms_rms_envelope_nrmse": 0.08622322096051972, "全帯域_10ms_envelope_correlation": 0.9914176191697119, "全帯域_10ms_modulation_mae_db": 0.45094728682915824, "全帯域_40ms_log_rms_mae_db": 0.45606443129884106, "全帯域_40ms_rms_envelope_nrmse": 0.04657603425655672, "全帯域_40ms_envelope_correlation": 0.9972042166443024, "全帯域_40ms_modulation_mae_db": 0.5072950945011975, "300–1000 Hz_10ms_log_rms_mae_db": 1.4376579980286697, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1061821611402694, "300–1000 Hz_10ms_envelope_correlation": 0.9899867092058423, "300–1000 Hz_10ms_modulation_mae_db": 0.807890064560928, "300–1000 Hz_40ms_log_rms_mae_db": 0.8651938827704609, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0660278248226848, "300–1000 Hz_40ms_envelope_correlation": 0.9956964497315323, "300–1000 Hz_40ms_modulation_mae_db": 0.9813004597252583, "1000–3000 Hz_10ms_log_rms_mae_db": 1.71840214302831, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22116668548747556, "1000–3000 Hz_10ms_envelope_correlation": 0.9572603513844921, "1000–3000 Hz_10ms_modulation_mae_db": 0.5488843385559009, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2214334973069556, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15315589446119268, "1000–3000 Hz_40ms_envelope_correlation": 0.9772779217634094, "1000–3000 Hz_40ms_modulation_mae_db": 0.3070867014633439, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7107010459104088, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17462130396123587, "3000–6000 Hz_10ms_envelope_correlation": 0.9728921373362943, "3000–6000 Hz_10ms_modulation_mae_db": 1.0415909709499067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9744720216181196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10788578705528673, "3000–6000 Hz_40ms_envelope_correlation": 0.9891222454738191, "3000–6000 Hz_40ms_modulation_mae_db": 1.296375203775046, "mr_stft_db": 5.7534565850003645, "band_mae_db": 0.12748922215676098, "rms_gain_error_db": -0.05156967195420155, "output_peak": 0.4598745789906534}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8993802780662985, "全帯域_10ms_rms_envelope_nrmse": 0.08193281878656099, "全帯域_10ms_envelope_correlation": 0.992334507428973, "全帯域_10ms_modulation_mae_db": 0.25604320236050865, "全帯域_40ms_log_rms_mae_db": 0.4188260963698904, "全帯域_40ms_rms_envelope_nrmse": 0.041993938192864774, "全帯域_40ms_envelope_correlation": 0.9978908700105141, "全帯域_40ms_modulation_mae_db": 0.3295794478185733, "300–1000 Hz_10ms_log_rms_mae_db": 1.3858490626810291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09758295994846324, "300–1000 Hz_10ms_envelope_correlation": 0.9916901002510186, "300–1000 Hz_10ms_modulation_mae_db": 0.37321895334036864, "300–1000 Hz_40ms_log_rms_mae_db": 0.8128648051860309, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053630267477084555, "300–1000 Hz_40ms_envelope_correlation": 0.9973929067953448, "300–1000 Hz_40ms_modulation_mae_db": 0.7187020003282886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7017742459740033, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21282283343965006, "1000–3000 Hz_10ms_envelope_correlation": 0.9604834444330319, "1000–3000 Hz_10ms_modulation_mae_db": 0.31426228204110185, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2046973516086186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13765800959490535, "1000–3000 Hz_40ms_envelope_correlation": 0.9817099415795302, "1000–3000 Hz_40ms_modulation_mae_db": 0.3638179113088316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.743595078783924, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17390229328202064, "3000–6000 Hz_10ms_envelope_correlation": 0.9726901105859324, "3000–6000 Hz_10ms_modulation_mae_db": 0.7985727190711897, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0663157186714418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09999178254563282, "3000–6000 Hz_40ms_envelope_correlation": 0.9900170642747715, "3000–6000 Hz_40ms_modulation_mae_db": 1.1309745545246, "mr_stft_db": 5.755236022300167, "band_mae_db": 0.140683096091538, "rms_gain_error_db": -0.114078744333089, "output_peak": 0.5338388673058017}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8863511583328478, "全帯域_10ms_rms_envelope_nrmse": 0.0902909732808649, "全帯域_10ms_envelope_correlation": 0.9905564540120515, "全帯域_10ms_modulation_mae_db": 0.47274166308231685, "全帯域_40ms_log_rms_mae_db": 0.4739851954626901, "全帯域_40ms_rms_envelope_nrmse": 0.05297143853100997, "全帯域_40ms_envelope_correlation": 0.996355753732124, "全帯域_40ms_modulation_mae_db": 0.4943429912503324, "300–1000 Hz_10ms_log_rms_mae_db": 1.390445165055816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10682209695775442, "300–1000 Hz_10ms_envelope_correlation": 0.9898652988128828, "300–1000 Hz_10ms_modulation_mae_db": 0.3689511014944755, "300–1000 Hz_40ms_log_rms_mae_db": 0.8550113874096633, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06390984590250039, "300–1000 Hz_40ms_envelope_correlation": 0.9959715538715902, "300–1000 Hz_40ms_modulation_mae_db": 0.6579390870211936, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6547730414590307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1987088017762904, "1000–3000 Hz_10ms_envelope_correlation": 0.9654975644060936, "1000–3000 Hz_10ms_modulation_mae_db": 0.37139292925003803, "1000–3000 Hz_40ms_log_rms_mae_db": 1.134765515279155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12442905428527115, "1000–3000 Hz_40ms_envelope_correlation": 0.9850899358434037, "1000–3000 Hz_40ms_modulation_mae_db": 0.4113265397567954, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7420666412716936, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1781769165085139, "3000–6000 Hz_10ms_envelope_correlation": 0.9713112746646234, "3000–6000 Hz_10ms_modulation_mae_db": 0.740685345190856, "3000–6000 Hz_40ms_log_rms_mae_db": 1.049492314657328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1035191028274377, "3000–6000 Hz_40ms_envelope_correlation": 0.9893535329318357, "3000–6000 Hz_40ms_modulation_mae_db": 0.6652458782332077, "mr_stft_db": 5.766577609949447, "band_mae_db": 0.12001561370625662, "rms_gain_error_db": -0.05911704817784824, "output_peak": 0.4396643015212284}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 93920, "output_samples": 93920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.659176132989016, "spectral_convergence": 0.2769921064064206, "spectral_convergence_status": "ok", "amplitude_floor": 2.214924872151779e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.683725806540738, "spectral_convergence": 0.27872335467335957, "spectral_convergence_status": "ok", "amplitude_floor": 1.3113821592368984e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.933202518108972, "spectral_convergence": 0.28987934359995415, "spectral_convergence_status": "ok", "amplitude_floor": 7.380112905156557e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00039502697710642294, "output_power": 0.00035827132734213215, "delta_db": -0.4241470267182387, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014539054117940992, "output_power": 0.001471698153024178, "delta_db": 0.05282591736069978, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0016903939251448888, "output_power": 0.0015718572667708032, "delta_db": -0.3157481611911884, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.480236057553453e-05, "output_power": 6.870071924872537e-05, "delta_db": 0.25370457506864963, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.753736279208405e-05, "output_power": 9.561676245130782e-05, "delta_db": -0.08636974436581034, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.937191536111839e-05, "output_power": 5.301641901927381e-05, "delta_db": 0.30930414271578627, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24034992790339552, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15543597272188547, "pitch": null, "output_rms_dbfs": -24.41392138934049, "output_peak": 0.4195045917015632, "reference_rms_dbfs": -24.258485416618605, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.758701485879574}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 93920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002156956510733609, "log_rms_mae_db": 0.8855576389296078, "rms_envelope_nrmse": 0.08315659193044429, "envelope_correlation": 0.9922411464246534, "rms_gain_error_db": -0.15543864506708296, "reference_mean_power": 0.003754233895192234, "output_mean_power": 0.0036222419998930256}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.680528929094005e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006742966926403748, "output_power": 0.0006452238690503899, "reference_active": true, "delta_db": -0.19140604448820892}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002358850904334285, "output_power": 0.0002238762678500128, "reference_active": true, "delta_db": -0.22692432994106615}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0118677140216256e-05, "output_power": 2.72415708790481e-05, "reference_active": true, "delta_db": -0.43603745592910687}], "active_band_count": 3, "active_band_mae_db": 0.2847892767861273}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00016458512240675186, "log_rms_mae_db": 0.4641376214236186, "rms_envelope_nrmse": 0.049586714265995896, "envelope_correlation": 0.9971623920322136, "rms_gain_error_db": -0.15544123727262096, "reference_mean_power": 0.003773515049488898, "output_mean_power": 0.0036408430914559656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.4929366354790985e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006129787155244603, "output_power": 0.0005839561236799013, "reference_active": true, "delta_db": -0.21065177717209105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073825851363778, "output_power": 0.00010595169852984469, "reference_active": true, "delta_db": -0.1918976609792994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.608951147124256e-06, "output_power": 2.089738412022554e-06, "reference_active": true, "delta_db": -0.9637402129966148}], "active_band_count": 3, "active_band_mae_db": 0.4554298837160018}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002116840322083979, "log_rms_mae_db": 1.4129666607894271, "rms_envelope_nrmse": 0.09794882694478332, "envelope_correlation": 0.9917668819181605, "rms_gain_error_db": -0.19561493113103612, "reference_mean_power": 0.002333704816993391, "output_mean_power": 0.0022309222222451123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.3033890876287954e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006736793990758906, "output_power": 0.0006408430431493142, "reference_active": true, "delta_db": -0.21701592881709555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002058984330594163, "output_power": 0.00018627920051052105, "reference_active": true, "delta_db": -0.4348867621535099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6276513462408744e-05, "output_power": 2.044823732672139e-05, "reference_active": true, "delta_db": -1.0891186268794948}], "active_band_count": 3, "active_band_mae_db": 0.5803404392833668}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00015464611695479593, "log_rms_mae_db": 0.8332639234671907, "rms_envelope_nrmse": 0.05553389007287119, "envelope_correlation": 0.9973761585542487, "rms_gain_error_db": -0.195615226578861, "reference_mean_power": 0.002345702993864749, "output_mean_power": 0.002242391814804812}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.1648344391379404e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377113304270822, "output_power": 0.0006054225799719701, "reference_active": true, "delta_db": -0.22565518844902788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.690162436721034e-05, "output_power": 8.865597906571144e-05, "reference_active": true, "delta_db": -0.3862302694773457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0235059299775393e-06, "output_power": 2.08839481818195e-06, "reference_active": true, "delta_db": -1.606982177717813}], "active_band_count": 3, "active_band_mae_db": 0.7396225452147288}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 4.7944659079237455e-05, "log_rms_mae_db": 1.665273887372956, "rms_envelope_nrmse": 0.19928754112359903, "envelope_correlation": 0.9659455115301111, "rms_gain_error_db": 0.14212974481291246, "reference_mean_power": 0.0001071814340711088, "output_mean_power": 0.00011074714470743523}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 8.167234143423064e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.040560289520861e-05, "output_power": 3.7862123296845184e-05, "reference_active": true, "delta_db": -0.2823662587391246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.346030743633636e-06, "output_power": 7.819645655517465e-06, "reference_active": true, "delta_db": -0.2829290655133866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1788672389317654e-06, "output_power": 2.590511567531282e-06, "reference_active": true, "delta_db": 0.7515476694875876}], "active_band_count": 3, "active_band_mae_db": 0.4389476645800329}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 3.945206980281183e-05, "log_rms_mae_db": 1.1569060881024582, "rms_envelope_nrmse": 0.132916867198244, "envelope_correlation": 0.9833665969318681, "rms_gain_error_db": 0.14212120287271518, "reference_mean_power": 0.00010773182378082642, "output_mean_power": 0.00011131562583343107}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 7.380645247762918e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7294287822712847e-05, "output_power": 3.503653957145162e-05, "reference_active": true, "delta_db": -0.27121112249554163}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.878816663714596e-06, "output_power": 3.869301982746298e-06, "reference_active": true, "delta_db": -0.01066626805094411}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8761762452219972e-07, "output_power": 1.7258187375565013e-07, "reference_active": true, "delta_db": -0.3627845304065213}], "active_band_count": 3, "active_band_mae_db": 0.2148873069843357}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 5.718269674480757e-05, "log_rms_mae_db": 1.7373219642242776, "rms_envelope_nrmse": 0.18130711295404056, "envelope_correlation": 0.9716234781610579, "rms_gain_error_db": 0.1746055624565163, "reference_mean_power": 7.041944137047507e-05, "output_mean_power": 7.330829718695298e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 5.931095881530669e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9173785649337023e-05, "output_power": 3.0511554583994846e-05, "reference_active": true, "delta_db": 0.1947154809140863}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1741151998348071e-05, "output_power": 1.2086587457605797e-05, "reference_active": true, "delta_db": 0.12592988546057107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.929169683988829e-06, "output_power": 2.5140031880889314e-06, "reference_active": true, "delta_db": 1.1499539552706297}], "active_band_count": 3, "active_band_mae_db": 0.49019977388176234}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 4.673628190489769e-05, "log_rms_mae_db": 1.0567426948832066, "rms_envelope_nrmse": 0.10301436794005414, "envelope_correlation": 0.9899465868441275, "rms_gain_error_db": 0.17454616878391696, "reference_mean_power": 7.077979037597618e-05, "output_mean_power": 7.368242130944449e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 5.253277550211025e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7931175609497952e-05, "output_power": 2.9915369699080415e-05, "reference_active": true, "delta_db": 0.2980515896649604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.1737648370747885e-06, "output_power": 5.815217305525874e-06, "reference_active": true, "delta_db": 0.5075926341003055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7085501560216023e-07, "output_power": 1.9217194540558126e-07, "reference_active": true, "delta_db": 0.5106225421134452}], "active_band_count": 3, "active_band_mae_db": 0.43875558862623704}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8733607000562423, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "70e5ed2f3fec6ac8b4bfc9f4f5f7b5b68397a1b586aeed969b6ebf235a2160bf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.963202642198666, "standard_deviation": 0.00448042981939606, "minimum": 0.9568863844937602, "maximum": 0.9680959875082399, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6813905965491978, "standard_deviation": 0.020199549542999115, "minimum": 1.6547730414590307, "maximum": 1.71840214302831, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4260121996208487, "standard_deviation": 0.16845163086374784, "minimum": 0.257911707973876, "maximum": 0.7803566456659086, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20542646492224587, "standard_deviation": 0.0129340228029431, "minimum": 0.18979996678213287, "maximum": 0.2243694904149324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9820279436323361, "standard_deviation": 0.0032101526332008433, "minimum": 0.9770922562506296, "maximum": 0.9850899358434037, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1769787963911602, "standard_deviation": 0.030213490749795992, "minimum": 1.134765515279155, "maximum": 1.2214334973069556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.29338254871875485, "standard_deviation": 0.12929589770478797, "minimum": 0.10424971795002218, "maximum": 0.49209101341085715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13663586513009598, "standard_deviation": 0.011968749464706883, "minimum": 0.12442905428527115, "maximum": 0.15568903423210848, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9724889224814162, "standard_deviation": 0.0015129804382964416, "minimum": 0.9707604331267593, "maximum": 0.9756014347494589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7369878405548345, "standard_deviation": 0.015581258077884928, "minimum": 1.7107010459104088, "maximum": 1.7564177209217353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7315973788374973, "standard_deviation": 0.22940164814912026, "minimum": 0.41505602252306556, "maximum": 1.0458084933630198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1758040804532868, "standard_deviation": 0.005294768178402704, "minimum": 0.16452084260066097, "maximum": 0.18130711295404056, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9891566338258876, "standard_deviation": 0.0008387173073761603, "minimum": 0.9875327984187263, "maximum": 0.9900170642747715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0534846247523464, "standard_deviation": 0.03611773233833252, "minimum": 0.9744720216181196, "maximum": 1.1031913994514355, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7769483648872095, "standard_deviation": 0.31067592281892537, "minimum": 0.43875558862623704, "maximum": 1.296375203775046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10620829149995943, "standard_deviation": 0.005262486995939846, "minimum": 0.09997836320804404, "maximum": 0.11305129851515662, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9906937138738843, "standard_deviation": 0.0007936987547974036, "minimum": 0.9898652988128828, "maximum": 0.9917668819181605, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4077048946778543, "standard_deviation": 0.03726633954083636, "minimum": 1.3663127158479578, "maximum": 1.472655788270505, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.386254396536153, "standard_deviation": 0.2227275640264665, "minimum": 0.08218854433772449, "maximum": 0.807890064560928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10286455531579053, "standard_deviation": 0.003910636780468426, "minimum": 0.09758295994846324, "maximum": 0.10682209695775442, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967055381132719, "standard_deviation": 0.0006288150360713139, "minimum": 0.9956964497315323, "maximum": 0.9973929067953448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8402032280143753, "standard_deviation": 0.03372574017633778, "minimum": 0.7867637714275827, "maximum": 0.8937312288618904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5808762205188368, "standard_deviation": 0.2609891131513265, "minimum": 0.11146658788216772, "maximum": 0.9813004597252583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05912885009476154, "standard_deviation": 0.004318370371720913, "minimum": 0.053630267477084555, "maximum": 0.0660278248226848, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16374800268011924, "standard_deviation": 0.044309914742062584, "minimum": 0.12001561370625662, "maximum": 0.24034992790339552, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7575995425210325, "standard_deviation": 0.007676267160912943, "minimum": 5.745438918664926, "maximum": 5.7699197943932825, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4362150221358957, "standard_deviation": 0.04386495755493394, "minimum": 0.39965627439000123, "maximum": 0.5338388673058017, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09530451820387373, "standard_deviation": 0.03266267176895211, "minimum": -0.15543597272188547, "maximum": -0.05156967195420155, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920090647558557, "standard_deviation": 0.0007762811940570624, "minimum": 0.9905564540120515, "maximum": 0.9929534198912428, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8857680523340306, "standard_deviation": 0.02681215922809665, "minimum": 0.8289946680452983, "maximum": 0.9258624614015543, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29057129435499385, "standard_deviation": 0.1157256779374133, "minimum": 0.15998492222117505, "maximum": 0.47274166308231685, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08342373661141068, "standard_deviation": 0.0038352206179319563, "minimum": 0.07830699347744839, "maximum": 0.0902909732808649, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973482083779688, "standard_deviation": 0.0005690646847168709, "minimum": 0.996355753732124, "maximum": 0.998007290312868, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4496175695728203, "standard_deviation": 0.018573640332979865, "minimum": 0.4188260963698904, "maximum": 0.4739851954626901, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3942630815063086, "standard_deviation": 0.16514125337925936, "minimum": 0.09730456372050182, "maximum": 0.590917370364688, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04605450916369332, "standard_deviation": 0.004638064863316772, "minimum": 0.04003033698070453, "maximum": 0.05297143853100997, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0020", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7567331790924072, "predicted_mos": 2.7567331790924072, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "bd493e7f10d105f38c131fa28c85ca41a229bd89cbf6b43fb1909110fdc2caf0", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.7273369999602437, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.88769784902211, "全帯域_10ms_rms_envelope_nrmse": 0.08680876510161221, "全帯域_10ms_envelope_correlation": 0.9913288143700802, "全帯域_10ms_modulation_mae_db": 0.20777892960050356, "全帯域_40ms_log_rms_mae_db": 0.4411892577901476, "全帯域_40ms_rms_envelope_nrmse": 0.052542280089737096, "全帯域_40ms_envelope_correlation": 0.9964048368990995, "全帯域_40ms_modulation_mae_db": 0.05386879124444521, "300–1000 Hz_10ms_log_rms_mae_db": 1.3743168249352824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11005622872072338, "300–1000 Hz_10ms_envelope_correlation": 0.9892365914062234, "300–1000 Hz_10ms_modulation_mae_db": 0.3011361361557824, "300–1000 Hz_40ms_log_rms_mae_db": 0.7697274806949711, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06461185839199811, "300–1000 Hz_40ms_envelope_correlation": 0.9958450452166685, "300–1000 Hz_40ms_modulation_mae_db": 0.4987985921079419, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6013137006034925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1962163184137155, "1000–3000 Hz_10ms_envelope_correlation": 0.9672914268303865, "1000–3000 Hz_10ms_modulation_mae_db": 0.24243577925569326, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0740936352198533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12808328083495524, "1000–3000 Hz_40ms_envelope_correlation": 0.9847658937850097, "1000–3000 Hz_40ms_modulation_mae_db": 0.1124020661264853, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7241885345611696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1661131145590117, "3000–6000 Hz_10ms_envelope_correlation": 0.9751303143759319, "3000–6000 Hz_10ms_modulation_mae_db": 0.3795915551747116, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0065611772979095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10335518244872229, "3000–6000 Hz_40ms_envelope_correlation": 0.9890718000622174, "3000–6000 Hz_40ms_modulation_mae_db": 0.4481377210675586, "mr_stft_db": 5.745265585151961, "band_mae_db": 0.13271830044310765, "rms_gain_error_db": 0.023932216735199713, "output_peak": 0.43841465594295514}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9357102427089742, "全帯域_10ms_rms_envelope_nrmse": 0.09956275013545536, "全帯域_10ms_envelope_correlation": 0.9887383507634988, "全帯域_10ms_modulation_mae_db": 0.5053510969425715, "全帯域_40ms_log_rms_mae_db": 0.5046172582117173, "全帯域_40ms_rms_envelope_nrmse": 0.06477032200469741, "全帯域_40ms_envelope_correlation": 0.994893473371156, "全帯域_40ms_modulation_mae_db": 0.4960845670360996, "300–1000 Hz_10ms_log_rms_mae_db": 1.4214444425422204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10667370955672471, "300–1000 Hz_10ms_envelope_correlation": 0.9901521646474888, "300–1000 Hz_10ms_modulation_mae_db": 0.5004006130032687, "300–1000 Hz_40ms_log_rms_mae_db": 0.8560101238834646, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06529421111920847, "300–1000 Hz_40ms_envelope_correlation": 0.9961931356233491, "300–1000 Hz_40ms_modulation_mae_db": 0.5541760146735745, "1000–3000 Hz_10ms_log_rms_mae_db": 1.598002591013047, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19957609735031107, "1000–3000 Hz_10ms_envelope_correlation": 0.9655995970268983, "1000–3000 Hz_10ms_modulation_mae_db": 0.45400364168802315, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1070056807715238, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13182164274666192, "1000–3000 Hz_40ms_envelope_correlation": 0.9834724311551933, "1000–3000 Hz_40ms_modulation_mae_db": 0.47282716443566675, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7032122660318927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1650600732895276, "3000–6000 Hz_10ms_envelope_correlation": 0.9756088601199887, "3000–6000 Hz_10ms_modulation_mae_db": 0.8673385655711984, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0037001807379657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10165443460078874, "3000–6000 Hz_40ms_envelope_correlation": 0.9899087999947125, "3000–6000 Hz_40ms_modulation_mae_db": 1.2070077829127361, "mr_stft_db": 5.754577963410604, "band_mae_db": 0.2452837184477652, "rms_gain_error_db": -0.1674528745227412, "output_peak": 0.40981230177249184}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9499246111556783, "全帯域_10ms_rms_envelope_nrmse": 0.08711793977360037, "全帯域_10ms_envelope_correlation": 0.9912085603266341, "全帯域_10ms_modulation_mae_db": 0.15862148872347845, "全帯域_40ms_log_rms_mae_db": 0.4852771969683983, "全帯域_40ms_rms_envelope_nrmse": 0.04474400561545278, "全帯域_40ms_envelope_correlation": 0.9973931257092769, "全帯域_40ms_modulation_mae_db": 0.17108625819869852, "300–1000 Hz_10ms_log_rms_mae_db": 1.4229948150160334, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10874715689306111, "300–1000 Hz_10ms_envelope_correlation": 0.9894877947914656, "300–1000 Hz_10ms_modulation_mae_db": 0.21212064545381146, "300–1000 Hz_40ms_log_rms_mae_db": 0.8490932609479606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06049795189374545, "300–1000 Hz_40ms_envelope_correlation": 0.9963800442053741, "300–1000 Hz_40ms_modulation_mae_db": 0.39285890467023493, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6656699345836314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2274054371909611, "1000–3000 Hz_10ms_envelope_correlation": 0.9554143428227095, "1000–3000 Hz_10ms_modulation_mae_db": 0.9169609771592123, "1000–3000 Hz_40ms_log_rms_mae_db": 1.17021222175952, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15674726846425416, "1000–3000 Hz_40ms_envelope_correlation": 0.9765534912026091, "1000–3000 Hz_40ms_modulation_mae_db": 0.5235123457498404, "3000–6000 Hz_10ms_log_rms_mae_db": 1.676293743218118, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17330173744967858, "3000–6000 Hz_10ms_envelope_correlation": 0.9730486516014091, "3000–6000 Hz_10ms_modulation_mae_db": 0.8192110835948802, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9982805232726818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09789298294263549, "3000–6000 Hz_40ms_envelope_correlation": 0.99076966285373, "3000–6000 Hz_40ms_modulation_mae_db": 0.9815204754473873, "mr_stft_db": 5.742779941048476, "band_mae_db": 0.11219263879849121, "rms_gain_error_db": -0.04114527477415829, "output_peak": 0.4977021042484224}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9011861378200489, "全帯域_10ms_rms_envelope_nrmse": 0.07317973737118547, "全帯域_10ms_envelope_correlation": 0.9939063796259032, "全帯域_10ms_modulation_mae_db": 0.057768208340369116, "全帯域_40ms_log_rms_mae_db": 0.47215042518053696, "全帯域_40ms_rms_envelope_nrmse": 0.04111170505011067, "全帯域_40ms_envelope_correlation": 0.997945788059813, "全帯域_40ms_modulation_mae_db": 0.0879422241757301, "300–1000 Hz_10ms_log_rms_mae_db": 1.3988130152206062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08836226700476149, "300–1000 Hz_10ms_envelope_correlation": 0.99331695760597, "300–1000 Hz_10ms_modulation_mae_db": 0.2777268660643051, "300–1000 Hz_40ms_log_rms_mae_db": 0.8505926177023593, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05227266570578394, "300–1000 Hz_40ms_envelope_correlation": 0.9976388727403146, "300–1000 Hz_40ms_modulation_mae_db": 0.5012315341824732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6434674196526462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19767902247883537, "1000–3000 Hz_10ms_envelope_correlation": 0.9658613337717948, "1000–3000 Hz_10ms_modulation_mae_db": 0.4029782346283448, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1478502346972774, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13402191883468684, "1000–3000 Hz_40ms_envelope_correlation": 0.9826570794234615, "1000–3000 Hz_40ms_modulation_mae_db": 0.29651146484266505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7039914554147189, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16677781208529852, "3000–6000 Hz_10ms_envelope_correlation": 0.9749036867924542, "3000–6000 Hz_10ms_modulation_mae_db": 0.18747271568259613, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9847470070874271, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09530281651826063, "3000–6000 Hz_40ms_envelope_correlation": 0.9908741714780678, "3000–6000 Hz_40ms_modulation_mae_db": 0.3916007449980075, "mr_stft_db": 5.737055154084232, "band_mae_db": 0.0904527195416822, "rms_gain_error_db": -0.10395217727335096, "output_peak": 0.4272129927228425}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8852500909168297, "全帯域_10ms_rms_envelope_nrmse": 0.0850923092310045, "全帯域_10ms_envelope_correlation": 0.9916648808871413, "全帯域_10ms_modulation_mae_db": 0.27512551304862237, "全帯域_40ms_log_rms_mae_db": 0.461242682487212, "全帯域_40ms_rms_envelope_nrmse": 0.052893425172049875, "全帯域_40ms_envelope_correlation": 0.9963917923300999, "全帯域_40ms_modulation_mae_db": 0.3166953008817401, "300–1000 Hz_10ms_log_rms_mae_db": 1.3471813096166798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09703222712448958, "300–1000 Hz_10ms_envelope_correlation": 0.9917871871465106, "300–1000 Hz_10ms_modulation_mae_db": 0.5728540763129081, "300–1000 Hz_40ms_log_rms_mae_db": 0.7628674315930023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05628966969675008, "300–1000 Hz_40ms_envelope_correlation": 0.9971050396707758, "300–1000 Hz_40ms_modulation_mae_db": 0.6985528580595018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6444001052366293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21316128731837514, "1000–3000 Hz_10ms_envelope_correlation": 0.9601639730139754, "1000–3000 Hz_10ms_modulation_mae_db": 0.5697115827554439, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1385062082849893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13332695173968515, "1000–3000 Hz_40ms_envelope_correlation": 0.9827759216542195, "1000–3000 Hz_40ms_modulation_mae_db": 0.27115791045216275, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6467544859917882, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17022955415575747, "3000–6000 Hz_10ms_envelope_correlation": 0.9738796438980669, "3000–6000 Hz_10ms_modulation_mae_db": 0.6950845270502994, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9743813439091918, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10197613228092912, "3000–6000 Hz_40ms_envelope_correlation": 0.9893681605087955, "3000–6000 Hz_40ms_modulation_mae_db": 0.518554005202271, "mr_stft_db": 5.727536066794658, "band_mae_db": 0.20199937801622816, "rms_gain_error_db": -0.03965699455409235, "output_peak": 0.41958932249266345}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9067092638791601, "全帯域_10ms_rms_envelope_nrmse": 0.0846820805626407, "全帯域_10ms_envelope_correlation": 0.9917864594721972, "全帯域_10ms_modulation_mae_db": 0.2600720296106209, "全帯域_40ms_log_rms_mae_db": 0.44126540710682643, "全帯域_40ms_rms_envelope_nrmse": 0.044635166154614554, "全帯域_40ms_envelope_correlation": 0.9975607858601624, "全帯域_40ms_modulation_mae_db": 0.41527830519197634, "300–1000 Hz_10ms_log_rms_mae_db": 1.4372156265861968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11049692645981327, "300–1000 Hz_10ms_envelope_correlation": 0.9891496031483725, "300–1000 Hz_10ms_modulation_mae_db": 0.24336214884729554, "300–1000 Hz_40ms_log_rms_mae_db": 0.8734452073437865, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062481461977588686, "300–1000 Hz_40ms_envelope_correlation": 0.9961605650980739, "300–1000 Hz_40ms_modulation_mae_db": 0.6282170234396943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.642698698173408, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1947216229135529, "1000–3000 Hz_10ms_envelope_correlation": 0.9665215722687304, "1000–3000 Hz_10ms_modulation_mae_db": 0.5078884470481922, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1595880858180805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12767415927055004, "1000–3000 Hz_40ms_envelope_correlation": 0.9841511609108781, "1000–3000 Hz_40ms_modulation_mae_db": 0.45603541461883523, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7200587090571664, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1786216066436004, "3000–6000 Hz_10ms_envelope_correlation": 0.9711472081991153, "3000–6000 Hz_10ms_modulation_mae_db": 0.5925121415213932, "3000–6000 Hz_40ms_log_rms_mae_db": 1.038432221481016, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1050905014689221, "3000–6000 Hz_40ms_envelope_correlation": 0.9888658764687432, "3000–6000 Hz_40ms_modulation_mae_db": 0.8384300559804257, "mr_stft_db": 5.749689963128058, "band_mae_db": 0.14959044416084363, "rms_gain_error_db": -0.10862959968035109, "output_peak": 0.39598278982883667}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8923986220979854, "全帯域_10ms_rms_envelope_nrmse": 0.09765226150784245, "全帯域_10ms_envelope_correlation": 0.9890418774040574, "全帯域_10ms_modulation_mae_db": 0.19769120387582773, "全帯域_40ms_log_rms_mae_db": 0.4836872889432594, "全帯域_40ms_rms_envelope_nrmse": 0.07036125621590451, "全帯域_40ms_envelope_correlation": 0.9935599282805978, "全帯域_40ms_modulation_mae_db": 0.4081387755128874, "300–1000 Hz_10ms_log_rms_mae_db": 1.3608452378875882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10055531605615083, "300–1000 Hz_10ms_envelope_correlation": 0.9912586394955777, "300–1000 Hz_10ms_modulation_mae_db": 0.5995608194777358, "300–1000 Hz_40ms_log_rms_mae_db": 0.8018196449959376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05814075037275183, "300–1000 Hz_40ms_envelope_correlation": 0.9970314572264669, "300–1000 Hz_40ms_modulation_mae_db": 0.8532095071610769, "1000–3000 Hz_10ms_log_rms_mae_db": 1.671225185829347, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19716922071050477, "1000–3000 Hz_10ms_envelope_correlation": 0.9656474991700082, "1000–3000 Hz_10ms_modulation_mae_db": 0.7775231547110376, "1000–3000 Hz_40ms_log_rms_mae_db": 1.129094757212158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1268422995028299, "1000–3000 Hz_40ms_envelope_correlation": 0.9843905515005973, "1000–3000 Hz_40ms_modulation_mae_db": 0.5669683801757152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6994816083412136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16942754390760978, "3000–6000 Hz_10ms_envelope_correlation": 0.9741033949878787, "3000–6000 Hz_10ms_modulation_mae_db": 0.4615267169883559, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9822831033750109, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09738689359898571, "3000–6000 Hz_40ms_envelope_correlation": 0.9904212789733843, "3000–6000 Hz_40ms_modulation_mae_db": 0.4459163217294224, "mr_stft_db": 5.724893100801881, "band_mae_db": 0.23950806355964074, "rms_gain_error_db": -0.0009738411266140431, "output_peak": 0.5302917950881699}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9013982876540281, "全帯域_10ms_rms_envelope_nrmse": 0.08274184412069739, "全帯域_10ms_envelope_correlation": 0.9921685971781316, "全帯域_10ms_modulation_mae_db": 0.13031080543629822, "全帯域_40ms_log_rms_mae_db": 0.462973211300499, "全帯域_40ms_rms_envelope_nrmse": 0.0497082044463466, "全帯域_40ms_envelope_correlation": 0.99693627624139, "全帯域_40ms_modulation_mae_db": 0.19704342464416413, "300–1000 Hz_10ms_log_rms_mae_db": 1.3893493946153581, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09216008525975643, "300–1000 Hz_10ms_envelope_correlation": 0.9925575441654992, "300–1000 Hz_10ms_modulation_mae_db": 0.21297311362803126, "300–1000 Hz_40ms_log_rms_mae_db": 0.7525814314479694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05250906124681156, "300–1000 Hz_40ms_envelope_correlation": 0.9974220024511586, "300–1000 Hz_40ms_modulation_mae_db": 0.3155999318798403, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5889336422793523, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19186691982001816, "1000–3000 Hz_10ms_envelope_correlation": 0.9673719201884406, "1000–3000 Hz_10ms_modulation_mae_db": 0.5046466289744326, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1020253049319215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12946915683877086, "1000–3000 Hz_40ms_envelope_correlation": 0.983692653790557, "1000–3000 Hz_40ms_modulation_mae_db": 0.4822367609268074, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7959740518445115, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16467093533542887, "3000–6000 Hz_10ms_envelope_correlation": 0.9755733488007297, "3000–6000 Hz_10ms_modulation_mae_db": 0.718950373691885, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0750502242488744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09457991304051717, "3000–6000 Hz_40ms_envelope_correlation": 0.9910003935938628, "3000–6000 Hz_40ms_modulation_mae_db": 0.3467609616638407, "mr_stft_db": 5.733021417311863, "band_mae_db": 0.1388643982959253, "rms_gain_error_db": -0.10958096222783849, "output_peak": 0.42719532792515835}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 93920, "output_samples": 93920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6781685122862795, "spectral_convergence": 0.28701100031301713, "spectral_convergence_status": "ok", "amplitude_floor": 2.214924872151779e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.651852264437044, "spectral_convergence": 0.2721618357054635, "spectral_convergence_status": "ok", "amplitude_floor": 1.3113821592368984e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.905775978732559, "spectral_convergence": 0.29554827712317133, "spectral_convergence_status": "ok", "amplitude_floor": 7.380112905156557e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00039502697710642294, "output_power": 0.00041432682783963356, "delta_db": 0.20716299567991034, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014539054117940992, "output_power": 0.0014791774557267232, "delta_db": 0.07484125977875045, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0016903939251448888, "output_power": 0.0016641140241189897, "delta_db": -0.06804842722805547, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.480236057553453e-05, "output_power": 6.917989516333221e-05, "delta_db": 0.28389073202341913, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.753736279208405e-05, "output_power": 9.503138810565602e-05, "delta_db": -0.1130393602354717, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.937191536111839e-05, "output_power": 4.9935876653828295e-05, "delta_db": 0.04932702771303868, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13271830044310765, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.023932216735199713, "pitch": null, "output_rms_dbfs": -24.234553199883404, "output_peak": 0.43841465594295514, "reference_rms_dbfs": -24.258485416618605, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.745265585151961}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 93920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002156956510733609, "log_rms_mae_db": 0.88769784902211, "rms_envelope_nrmse": 0.08680876510161221, "envelope_correlation": 0.9913288143700802, "rms_gain_error_db": 0.02393302209558737, "reference_mean_power": 0.003754233895192234, "output_mean_power": 0.0037749797702261094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.680528929094005e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006742966926403748, "output_power": 0.0006619803649417572, "reference_active": true, "delta_db": -0.08005921526074625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002358850904334285, "output_power": 0.00022454115733420779, "reference_active": true, "delta_db": -0.21404534578660148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0118677140216256e-05, "output_power": 3.249070356604764e-05, "reference_active": true, "delta_db": 0.32923222775416294}], "active_band_count": 3, "active_band_mae_db": 0.20777892960050356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00016458512240675186, "log_rms_mae_db": 0.4411892577901476, "rms_envelope_nrmse": 0.052542280089737096, "envelope_correlation": 0.9964048368990995, "rms_gain_error_db": 0.023943298409696865, "reference_mean_power": 0.003773515049488898, "output_mean_power": 0.0037943764503473487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.4929366354790985e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006129787155244603, "output_power": 0.0006038828514581899, "reference_active": true, "delta_db": -0.06492697696731749}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073825851363778, "output_power": 0.00010832858758522835, "reference_active": true, "delta_db": -0.09554608704730269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.608951147124256e-06, "output_power": 2.6096320527391343e-06, "reference_active": true, "delta_db": 0.0011333097187154338}], "active_band_count": 3, "active_band_mae_db": 0.05386879124444521}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002116840322083979, "log_rms_mae_db": 1.3743168249352824, "rms_envelope_nrmse": 0.11005622872072338, "envelope_correlation": 0.9892365914062234, "rms_gain_error_db": -0.03244772681169743, "reference_mean_power": 0.002333704816993391, "output_mean_power": 0.002316333829325337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.3033890876287954e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006736793990758906, "output_power": 0.0006504022832192119, "reference_active": true, "delta_db": -0.15271209786750592}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002058984330594163, "output_power": 0.00018900540901306792, "reference_active": true, "delta_db": -0.371788084168588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6276513462408744e-05, "output_power": 2.4081133042115782e-05, "reference_active": true, "delta_db": -0.3789082264312532}], "active_band_count": 3, "active_band_mae_db": 0.3011361361557824}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00015464611695479593, "log_rms_mae_db": 0.7697274806949711, "rms_envelope_nrmse": 0.06461185839199811, "envelope_correlation": 0.9958450452166685, "rms_gain_error_db": -0.032447811773902005, "reference_mean_power": 0.002345702993864749, "output_mean_power": 0.002328242651931907}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.1648344391379404e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377113304270822, "output_power": 0.0006158774568625007, "reference_active": true, "delta_db": -0.15129825531898655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.690162436721034e-05, "output_power": 9.301524062669571e-05, "reference_active": true, "delta_db": -0.17776943307367438}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0235059299775393e-06, "output_power": 2.31088330979221e-06, "reference_active": true, "delta_db": -1.167328087931165}], "active_band_count": 3, "active_band_mae_db": 0.4987985921079419}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 4.7944659079237455e-05, "log_rms_mae_db": 1.6013137006034925, "rms_envelope_nrmse": 0.1962163184137155, "envelope_correlation": 0.9672914268303865, "rms_gain_error_db": 0.16819837425520265, "reference_mean_power": 0.0001071814340711088, "output_mean_power": 0.00011141390619820179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 8.167234143423064e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.040560289520861e-05, "output_power": 3.865589004275025e-05, "reference_active": true, "delta_db": -0.19225914042951553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.346030743633636e-06, "output_power": 8.17307299650565e-06, "reference_active": true, "delta_db": -0.09094602387495222}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1788672389317654e-06, "output_power": 2.413464895668645e-06, "reference_active": true, "delta_db": 0.44410217346261205}], "active_band_count": 3, "active_band_mae_db": 0.24243577925569326}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 3.945206980281183e-05, "log_rms_mae_db": 1.0740936352198533, "rms_envelope_nrmse": 0.12808328083495524, "envelope_correlation": 0.9847658937850097, "rms_gain_error_db": 0.16819513906768857, "reference_mean_power": 0.00010773182378082642, "output_mean_power": 0.00011198594674579297}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 7.380645247762918e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7294287822712847e-05, "output_power": 3.6004451666326664e-05, "reference_active": true, "delta_db": -0.15286117042168473}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.878816663714596e-06, "output_power": 3.954581618537716e-06, "reference_active": true, "delta_db": 0.08401290658649689}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8761762452219972e-07, "output_power": 1.8333290529553106e-07, "reference_active": true, "delta_db": -0.10033212137127427}], "active_band_count": 3, "active_band_mae_db": 0.1124020661264853}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 5.718269674480757e-05, "log_rms_mae_db": 1.7241885345611696, "rms_envelope_nrmse": 0.1661131145590117, "envelope_correlation": 0.9751303143759319, "rms_gain_error_db": -0.05222686254592695, "reference_mean_power": 7.041944137047507e-05, "output_mean_power": 6.957767130989565e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 5.931095881530669e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9173785649337023e-05, "output_power": 2.7300982426339346e-05, "reference_active": true, "delta_db": -0.28814512290738853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1741151998348071e-05, "output_power": 1.1308142043695324e-05, "reference_active": true, "delta_db": -0.16319455261953023}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.929169683988829e-06, "output_power": 1.6467471844255332e-06, "reference_active": true, "delta_db": -0.6874349899972161}], "active_band_count": 3, "active_band_mae_db": 0.3795915551747116}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 4.673628190489769e-05, "log_rms_mae_db": 1.0065611772979095, "rms_envelope_nrmse": 0.10335518244872229, "envelope_correlation": 0.9890718000622174, "rms_gain_error_db": -0.05226346547667585, "reference_mean_power": 7.077979037597618e-05, "output_mean_power": 6.993312341752054e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 5.253277550211025e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7931175609497952e-05, "output_power": 2.6394460994969237e-05, "reference_active": true, "delta_db": -0.24576417689348257}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.1737648370747885e-06, "output_power": 5.001199622067361e-06, "reference_active": true, "delta_db": -0.14732495172440158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7085501560216023e-07, "output_power": 1.372446199651379e-07, "reference_active": true, "delta_db": -0.9513240345847918}], "active_band_count": 3, "active_band_mae_db": 0.4481377210675586}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8793794754402354, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6a1747580589c37eb76faa5adf28aaafdb56ef878878f8ad31271471a789b195", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9642339581366179, "standard_deviation": 0.004230668835050044, "minimum": 0.9554143428227095, "maximum": 0.9673719201884406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6319639096714442, "standard_deviation": 0.031677345893272944, "minimum": 1.5889336422793523, "maximum": 1.671225185829347, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5470185557775474, "standard_deviation": 0.2124689356939458, "minimum": 0.24243577925569326, "maximum": 0.9169609771592123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20222449077453425, "standard_deviation": 0.011991634248090046, "minimum": 0.19186691982001816, "maximum": 0.2274054371909611, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9828073979278157, "standard_deviation": 0.0026324530408096142, "minimum": 0.9765534912026091, "maximum": 0.9847658937850097, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1285470160869155, "standard_deviation": 0.03230267156069243, "minimum": 1.0740936352198533, "maximum": 1.17021222175952, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.39770643841602227, "standard_deviation": 0.1550672835847947, "minimum": 0.1124020661264853, "maximum": 0.5669683801757152, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13349833477904927, "standard_deviation": 0.009764429879662924, "minimum": 0.1268422995028299, "maximum": 0.15674726846425416, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9741743885969467, "standard_deviation": 0.0015110775544656427, "minimum": 0.9711472081991153, "maximum": 0.9756088601199887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7087443568075724, "standard_deviation": 0.043111708560090535, "minimum": 1.6467544859917882, "maximum": 1.7959740518445115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.590210959909415, "standard_deviation": 0.23297351935371105, "minimum": 0.18747271568259613, "maximum": 0.8673385655711984, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1692752971782391, "standard_deviation": 0.004776498993521957, "minimum": 0.16467093533542887, "maximum": 0.1786216066436004, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900350179916892, "standard_deviation": 0.0008519171498836285, "minimum": 0.9888658764687432, "maximum": 0.9910003935938628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0079294726762598, "standard_deviation": 0.03352848310730631, "minimum": 0.9743813439091918, "maximum": 1.0750502242488744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6472410086252062, "standard_deviation": 0.3193764352817614, "minimum": 0.3467609616638407, "maximum": 1.2070077829127361, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09965485711247016, "standard_deviation": 0.003883452798354975, "minimum": 0.09457991304051717, "maximum": 0.1050905014689221, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9908683103008885, "standard_deviation": 0.0015982680227418892, "minimum": 0.9891496031483725, "maximum": 0.99331695760597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3940200833024956, "standard_deviation": 0.032037773521084466, "minimum": 1.3471813096166798, "maximum": 1.4372156265861968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3650168023678923, "standard_deviation": 0.16454727930580415, "minimum": 0.21212064545381146, "maximum": 0.5995608194777358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1017604896344351, "standard_deviation": 0.008565603208577343, "minimum": 0.08836226700476149, "maximum": 0.11049692645981327, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967220202790227, "standard_deviation": 0.000660522774577416, "minimum": 0.9958450452166685, "maximum": 0.9976388727403146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8145171498261814, "standard_deviation": 0.04834725194745769, "minimum": 0.7525814314479694, "maximum": 0.8734452073437865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5553305457717923, "standard_deviation": 0.1709421974919032, "minimum": 0.3155999318798403, "maximum": 0.8532095071610769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05901220380057977, "standard_deviation": 0.005084244669548264, "minimum": 0.05227266570578394, "maximum": 0.06529421111920847, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16382620765796052, "standard_deviation": 0.05814195102600271, "minimum": 0.0904527195416822, "maximum": 0.2452837184477652, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.739352398966467, "standard_deviation": 0.010552268091963135, "minimum": 5.724893100801881, "maximum": 5.754577963410604, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44327516125269256, "standard_deviation": 0.046278220447877196, "minimum": 0.39598278982883667, "maximum": 0.5302917950881699, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06843243842799335, "standard_deviation": 0.06439855670453155, "minimum": -0.1674528745227412, "maximum": 0.023932216735199713, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9912304900034554, "standard_deviation": 0.001671148430018723, "minimum": 0.9887383507634988, "maximum": 0.9939063796259032, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9075343881568518, "standard_deviation": 0.023264357256603724, "minimum": 0.8852500909168297, "maximum": 0.9499246111556783, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2240899094472865, "standard_deviation": 0.133477490792237, "minimum": 0.057768208340369116, "maximum": 0.5053510969425715, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08710471097550482, "standard_deviation": 0.0083614738693676, "minimum": 0.07317973737118547, "maximum": 0.09956275013545536, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963857508439494, "standard_deviation": 0.0014805824977804588, "minimum": 0.9935599282805978, "maximum": 0.997945788059813, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4690503409985747, "standard_deviation": 0.022033418668324232, "minimum": 0.4411892577901476, "maximum": 0.5046172582117173, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2682672058607177, "standard_deviation": 0.16406579367208593, "minimum": 0.05386879124444521, "maximum": 0.4960845670360996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.052595795593614186, "standard_deviation": 0.010206208732943006, "minimum": 0.04111170505011067, "maximum": 0.07036125621590451, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1284-1180-0027", "language": "en", "style": "normal", "speaker": "librispeech_1284", "source_utterance_id": "1284-1180-0027", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 52320, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1284-1180-0027.wav", "source_file_sha256": "077e7fbcbc3b10543faad90a9556ad0989007af37ec02cf4a6bc100d3d035acb", "decoded_pcm_sha256": "96e7e523ecd8732e2ed1542dd9b6cbde8714adfeef42d0c1bfb37fb1a7c1cf83", "original_source_sha256": "9c4d5811c87e0273494850f3401459de8eafd6731f43a615c07f9ff944262a95", "corpus_entry_id": "librispeech_1284-1180-0027"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.256382942199707, "predicted_mos": 4.256382942199707, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "30486f3b2de234fe157905dd9aa6db4ce32c3a1f10728b8d678fb3b2da754d3e", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.5111223890016845, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1284-1180-0027", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "30486f3b2de234fe157905dd9aa6db4ce32c3a1f10728b8d678fb3b2da754d3e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "8a672729a28394fa02b7f210ccee664bc7753f7685217fb2e49e5b4bebd022ab"}, "rows": [{"case": "librispeech_1284-1180-0027", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.6718060970306396, "predicted_mos": 2.6718060970306396, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "dcc115f170fc7bc7408f70880458ddef789b991f78dd0d2d5d002933d2dbef6a", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.3554405180002504, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1284-1180-0027", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "dcc115f170fc7bc7408f70880458ddef789b991f78dd0d2d5d002933d2dbef6a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9254297053664768, "全帯域_10ms_rms_envelope_nrmse": 0.06114487803857682, "全帯域_10ms_envelope_correlation": 0.9955397788435723, "全帯域_10ms_modulation_mae_db": 0.4269890857388075, "全帯域_40ms_log_rms_mae_db": 0.5062060085554234, "全帯域_40ms_rms_envelope_nrmse": 0.03739729387640607, "全帯域_40ms_envelope_correlation": 0.9982859455155793, "全帯域_40ms_modulation_mae_db": 0.39788540865410177, "300–1000 Hz_10ms_log_rms_mae_db": 1.4331161446437226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08402705525597681, "300–1000 Hz_10ms_envelope_correlation": 0.993686337957775, "300–1000 Hz_10ms_modulation_mae_db": 0.4886590739124246, "300–1000 Hz_40ms_log_rms_mae_db": 0.8845880707272026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0519165297755269, "300–1000 Hz_40ms_envelope_correlation": 0.9975524452018124, "300–1000 Hz_40ms_modulation_mae_db": 0.5437752331232014, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4598920080413404, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13832566918305458, "1000–3000 Hz_10ms_envelope_correlation": 0.9846837290731181, "1000–3000 Hz_10ms_modulation_mae_db": 0.48525977981706386, "1000–3000 Hz_40ms_log_rms_mae_db": 1.019835974457442, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08607184232002121, "1000–3000 Hz_40ms_envelope_correlation": 0.9939216899576905, "1000–3000 Hz_40ms_modulation_mae_db": 0.5895928837480694, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5067798733480056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18069918647780425, "3000–6000 Hz_10ms_envelope_correlation": 0.9684001993233838, "3000–6000 Hz_10ms_modulation_mae_db": 0.5729825075224659, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9392787407583782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10787355208989102, "3000–6000 Hz_40ms_envelope_correlation": 0.9874086671456372, "3000–6000 Hz_40ms_modulation_mae_db": 0.7245549506042762, "mr_stft_db": 5.60720508243461, "band_mae_db": 0.1600282162332363, "rms_gain_error_db": -0.11432853888502045, "output_peak": 0.3770372669200526}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8398622654046971, "全帯域_10ms_rms_envelope_nrmse": 0.061959939113721496, "全帯域_10ms_envelope_correlation": 0.995323453618027, "全帯域_10ms_modulation_mae_db": 0.8190381391941827, "全帯域_40ms_log_rms_mae_db": 0.4258824152410732, "全帯域_40ms_rms_envelope_nrmse": 0.036809349489426796, "全帯域_40ms_envelope_correlation": 0.9982306734781965, "全帯域_40ms_modulation_mae_db": 0.7992200673870803, "300–1000 Hz_10ms_log_rms_mae_db": 1.3323042802038583, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08069636562021928, "300–1000 Hz_10ms_envelope_correlation": 0.9940805229154037, "300–1000 Hz_10ms_modulation_mae_db": 0.6920566288236039, "300–1000 Hz_40ms_log_rms_mae_db": 0.8008299354664886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05040076692824127, "300–1000 Hz_40ms_envelope_correlation": 0.9975536199066583, "300–1000 Hz_40ms_modulation_mae_db": 0.8253653321006658, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4625075508078058, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15471102347065438, "1000–3000 Hz_10ms_envelope_correlation": 0.9806249616542597, "1000–3000 Hz_10ms_modulation_mae_db": 0.5601701256841801, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0136198157461493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0924857837225791, "1000–3000 Hz_40ms_envelope_correlation": 0.992807020833991, "1000–3000 Hz_40ms_modulation_mae_db": 0.6800210636432601, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5410779786951272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20080748731698408, "3000–6000 Hz_10ms_envelope_correlation": 0.9642874545626382, "3000–6000 Hz_10ms_modulation_mae_db": 1.6616915252881743, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9684691506940585, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13273720230993954, "3000–6000 Hz_40ms_envelope_correlation": 0.9859735244039671, "3000–6000 Hz_40ms_modulation_mae_db": 2.3424899023381793, "mr_stft_db": 5.595042571115304, "band_mae_db": 0.1863107391530097, "rms_gain_error_db": -0.0803372519975819, "output_peak": 0.40879387938805556}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9336367932519374, "全帯域_10ms_rms_envelope_nrmse": 0.06393749094250253, "全帯域_10ms_envelope_correlation": 0.9950238725110795, "全帯域_10ms_modulation_mae_db": 0.34239940582557665, "全帯域_40ms_log_rms_mae_db": 0.42883521247704587, "全帯域_40ms_rms_envelope_nrmse": 0.03190169788062236, "全帯域_40ms_envelope_correlation": 0.9987173727091763, "全帯域_40ms_modulation_mae_db": 0.299211690687663, "300–1000 Hz_10ms_log_rms_mae_db": 1.4346656768419765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08401350559351721, "300–1000 Hz_10ms_envelope_correlation": 0.9936132469125574, "300–1000 Hz_10ms_modulation_mae_db": 0.7187633694892508, "300–1000 Hz_40ms_log_rms_mae_db": 0.8231681870360982, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043213322322646686, "300–1000 Hz_40ms_envelope_correlation": 0.9983017656823528, "300–1000 Hz_40ms_modulation_mae_db": 0.9581107439280908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5021512858083417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14295712673665742, "1000–3000 Hz_10ms_envelope_correlation": 0.9836133075282657, "1000–3000 Hz_10ms_modulation_mae_db": 0.9561599204587422, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0333636242975057, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08925211734731159, "1000–3000 Hz_40ms_envelope_correlation": 0.993463300537514, "1000–3000 Hz_40ms_modulation_mae_db": 0.8265849825496528, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4946845183797546, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1922655557292374, "3000–6000 Hz_10ms_envelope_correlation": 0.9645582076111531, "3000–6000 Hz_10ms_modulation_mae_db": 0.38233086095373486, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9683181261409805, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12782335225240404, "3000–6000 Hz_40ms_envelope_correlation": 0.9815251808237425, "3000–6000 Hz_40ms_modulation_mae_db": 0.5352953216223688, "mr_stft_db": 5.598348945637876, "band_mae_db": 0.14853004057602912, "rms_gain_error_db": -0.08380149970177733, "output_peak": 0.3752556743023631}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9067588332771749, "全帯域_10ms_rms_envelope_nrmse": 0.05686556288830639, "全帯域_10ms_envelope_correlation": 0.9960338463633742, "全帯域_10ms_modulation_mae_db": 0.06364260968579712, "全帯域_40ms_log_rms_mae_db": 0.4938802946014668, "全帯域_40ms_rms_envelope_nrmse": 0.02947116416556286, "全帯域_40ms_envelope_correlation": 0.9988479705495028, "全帯域_40ms_modulation_mae_db": 0.06507441211900113, "300–1000 Hz_10ms_log_rms_mae_db": 1.3691870787787377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0757020627280222, "300–1000 Hz_10ms_envelope_correlation": 0.9947515317669646, "300–1000 Hz_10ms_modulation_mae_db": 0.30271289804188906, "300–1000 Hz_40ms_log_rms_mae_db": 0.8402310023184436, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04100625395044998, "300–1000 Hz_40ms_envelope_correlation": 0.9983602077578497, "300–1000 Hz_40ms_modulation_mae_db": 0.2386581460048062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.478273654216207, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14634639293882415, "1000–3000 Hz_10ms_envelope_correlation": 0.9827093059062546, "1000–3000 Hz_10ms_modulation_mae_db": 0.9568705863483432, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9173679443889547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07902576274017009, "1000–3000 Hz_40ms_envelope_correlation": 0.9948771333256519, "1000–3000 Hz_40ms_modulation_mae_db": 1.0235192278553917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4849871373427903, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16792909654889812, "3000–6000 Hz_10ms_envelope_correlation": 0.9745749587371099, "3000–6000 Hz_10ms_modulation_mae_db": 1.124168454028296, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9546855064587465, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1057260990292872, "3000–6000 Hz_40ms_envelope_correlation": 0.9904279227806354, "3000–6000 Hz_40ms_modulation_mae_db": 1.431086545227121, "mr_stft_db": 5.613524345439312, "band_mae_db": 0.15730977001418653, "rms_gain_error_db": -0.053256184325033255, "output_peak": 0.36528319008941446}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9284872772783599, "全帯域_10ms_rms_envelope_nrmse": 0.06254167139487134, "全帯域_10ms_envelope_correlation": 0.9952446929414631, "全帯域_10ms_modulation_mae_db": 0.43193657220091514, "全帯域_40ms_log_rms_mae_db": 0.5394990278202741, "全帯域_40ms_rms_envelope_nrmse": 0.0337140468471804, "全帯域_40ms_envelope_correlation": 0.9985514971317965, "全帯域_40ms_modulation_mae_db": 0.4296374274538754, "300–1000 Hz_10ms_log_rms_mae_db": 1.4356594495128698, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07738847890363336, "300–1000 Hz_10ms_envelope_correlation": 0.9945474983515961, "300–1000 Hz_10ms_modulation_mae_db": 0.32614681213690133, "300–1000 Hz_40ms_log_rms_mae_db": 0.873122428026143, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041384780136831986, "300–1000 Hz_40ms_envelope_correlation": 0.998381358910514, "300–1000 Hz_40ms_modulation_mae_db": 0.3687945706320184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4799622100266558, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13585749838526517, "1000–3000 Hz_10ms_envelope_correlation": 0.985853122096162, "1000–3000 Hz_10ms_modulation_mae_db": 0.9313156499556411, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9412739905315218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08356675303901065, "1000–3000 Hz_40ms_envelope_correlation": 0.9951991194809757, "1000–3000 Hz_40ms_modulation_mae_db": 0.9236558935165426, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5973490728590163, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.204739756557199, "3000–6000 Hz_10ms_envelope_correlation": 0.9619063019636802, "3000–6000 Hz_10ms_modulation_mae_db": 2.0179750359482487, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0147180107113323, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12629928977755836, "3000–6000 Hz_40ms_envelope_correlation": 0.9869368576439801, "3000–6000 Hz_40ms_modulation_mae_db": 2.688652405032412, "mr_stft_db": 5.594352116463525, "band_mae_db": 0.25778616887293065, "rms_gain_error_db": -0.08542553572515393, "output_peak": 0.3831993764401483}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8586965862891148, "全帯域_10ms_rms_envelope_nrmse": 0.053851270153961704, "全帯域_10ms_envelope_correlation": 0.9965944886622327, "全帯域_10ms_modulation_mae_db": 0.5107060300568053, "全帯域_40ms_log_rms_mae_db": 0.46717592554401716, "全帯域_40ms_rms_envelope_nrmse": 0.02797561855606156, "全帯域_40ms_envelope_correlation": 0.9991536912678195, "全帯域_40ms_modulation_mae_db": 0.48524299721523817, "300–1000 Hz_10ms_log_rms_mae_db": 1.3882915577589574, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07699775164852038, "300–1000 Hz_10ms_envelope_correlation": 0.9948618624507296, "300–1000 Hz_10ms_modulation_mae_db": 0.47310456749480817, "300–1000 Hz_40ms_log_rms_mae_db": 0.9207193363651993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04679058806552035, "300–1000 Hz_40ms_envelope_correlation": 0.9982103924130872, "300–1000 Hz_40ms_modulation_mae_db": 0.5648331060723689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4481367640392082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16588525011796412, "1000–3000 Hz_10ms_envelope_correlation": 0.97827397715505, "1000–3000 Hz_10ms_modulation_mae_db": 0.5813144870742676, "1000–3000 Hz_40ms_log_rms_mae_db": 1.021327141296384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11849429373555254, "1000–3000 Hz_40ms_envelope_correlation": 0.9881936121653707, "1000–3000 Hz_40ms_modulation_mae_db": 0.5013577746954577, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5584743319183243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20974380636847464, "3000–6000 Hz_10ms_envelope_correlation": 0.9581990056476849, "3000–6000 Hz_10ms_modulation_mae_db": 0.9380333568584871, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9747064475277023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12591368013945956, "3000–6000 Hz_40ms_envelope_correlation": 0.9848182000009968, "3000–6000 Hz_40ms_modulation_mae_db": 1.1343105717366175, "mr_stft_db": 5.618308995611598, "band_mae_db": 0.22945141751072518, "rms_gain_error_db": -0.11623975740990047, "output_peak": 0.3765580586310963}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8705376905612651, "全帯域_10ms_rms_envelope_nrmse": 0.060745856003482444, "全帯域_10ms_envelope_correlation": 0.9955779529161537, "全帯域_10ms_modulation_mae_db": 0.6489805001022172, "全帯域_40ms_log_rms_mae_db": 0.4079266581943736, "全帯域_40ms_rms_envelope_nrmse": 0.030864735896496607, "全帯域_40ms_envelope_correlation": 0.9988778936716227, "全帯域_40ms_modulation_mae_db": 0.8066313222116618, "300–1000 Hz_10ms_log_rms_mae_db": 1.4014592807997646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0762186814596296, "300–1000 Hz_10ms_envelope_correlation": 0.994865234944714, "300–1000 Hz_10ms_modulation_mae_db": 0.6782497950190279, "300–1000 Hz_40ms_log_rms_mae_db": 0.8259973627862179, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03821940588568458, "300–1000 Hz_40ms_envelope_correlation": 0.998828894746343, "300–1000 Hz_40ms_modulation_mae_db": 0.7048864869060637, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4090843709019103, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14642603520544598, "1000–3000 Hz_10ms_envelope_correlation": 0.9831391254351289, "1000–3000 Hz_10ms_modulation_mae_db": 1.1951557885808206, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9205539879583707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08346684347981675, "1000–3000 Hz_40ms_envelope_correlation": 0.9949893840759908, "1000–3000 Hz_40ms_modulation_mae_db": 1.0941867497613165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5414096385576301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20916330885980505, "3000–6000 Hz_10ms_envelope_correlation": 0.9573621792412299, "3000–6000 Hz_10ms_modulation_mae_db": 0.4803374526982315, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9872578783916724, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1143052594383643, "3000–6000 Hz_40ms_envelope_correlation": 0.9851487900715311, "3000–6000 Hz_40ms_modulation_mae_db": 0.8488642637516411, "mr_stft_db": 5.618135170419261, "band_mae_db": 0.16155831435436815, "rms_gain_error_db": -0.10550167031081754, "output_peak": 0.37461189136093287}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9319542693937722, "全帯域_10ms_rms_envelope_nrmse": 0.061237444854111485, "全帯域_10ms_envelope_correlation": 0.9956724892122315, "全帯域_10ms_modulation_mae_db": 0.5950931770803857, "全帯域_40ms_log_rms_mae_db": 0.4766702161724041, "全帯域_40ms_rms_envelope_nrmse": 0.037467553254317415, "全帯域_40ms_envelope_correlation": 0.9984519039777678, "全帯域_40ms_modulation_mae_db": 0.711258294492208, "300–1000 Hz_10ms_log_rms_mae_db": 1.4548927012922035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07850740596995338, "300–1000 Hz_10ms_envelope_correlation": 0.994817288292501, "300–1000 Hz_10ms_modulation_mae_db": 0.6023280528599656, "300–1000 Hz_40ms_log_rms_mae_db": 0.8684388154088557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050849995179742494, "300–1000 Hz_40ms_envelope_correlation": 0.9980057217824986, "300–1000 Hz_40ms_modulation_mae_db": 0.5741642980033657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5096222171323628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14455120423555892, "1000–3000 Hz_10ms_envelope_correlation": 0.9831565679446014, "1000–3000 Hz_10ms_modulation_mae_db": 0.49204960844325724, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9978797529601084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08974211477060566, "1000–3000 Hz_40ms_envelope_correlation": 0.9932768352383334, "1000–3000 Hz_40ms_modulation_mae_db": 0.2982598745276613, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5260251877413378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2034389989530914, "3000–6000 Hz_10ms_envelope_correlation": 0.9622988886502435, "3000–6000 Hz_10ms_modulation_mae_db": 1.9545262913975392, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9876816838674434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14333201596767076, "3000–6000 Hz_40ms_envelope_correlation": 0.980921039876165, "3000–6000 Hz_40ms_modulation_mae_db": 2.3971869910281622, "mr_stft_db": 5.614709036998453, "band_mae_db": 0.220487710546905, "rms_gain_error_db": -0.15210981594087525, "output_peak": 0.3914053402814468}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 52320, "output_samples": 52320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.593195469647668, "spectral_convergence": 0.20815423666103314, "spectral_convergence_status": "ok", "amplitude_floor": 1.2746778473169441e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.477577765956582, "spectral_convergence": 0.1977656148745445, "spectral_convergence_status": "ok", "amplitude_floor": 1.090624511245042e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7508420116995795, "spectral_convergence": 0.20427944293463177, "spectral_convergence_status": "ok", "amplitude_floor": 7.935262595002381e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004730173956506593, "output_power": 0.0004807343203169454, "delta_db": 0.07028015533976506, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018195471619226037, "output_power": 0.0018068245430083423, "delta_db": -0.03047335634287619, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018173989836957952, "output_power": 0.0017212676921938789, "delta_db": -0.23601863635519196, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013671881984837986, "output_power": 0.00013433138912653531, "delta_db": -0.07650795085092249, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012324536624073353, "output_power": 0.0001175140888349487, "delta_db": -0.20680662166354727, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.679288082514536e-05, "output_power": 8.025569241186344e-05, "delta_db": -0.34008257684711474, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1600282162332363, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11432853888502045, "pitch": null, "output_rms_dbfs": -23.624174448702068, "output_peak": 0.3770372669200526, "reference_rms_dbfs": -23.509845909817045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.60720508243461}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 52320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00017439460044090568, "log_rms_mae_db": 0.9254297053664768, "rms_envelope_nrmse": 0.06114487803857682, "envelope_correlation": 0.9955397788435723, "rms_gain_error_db": -0.11433045939004183, "reference_mean_power": 0.004463526965610992, "output_mean_power": 0.004347555320490487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 1.3056377024951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008270090313661458, "output_power": 0.0007935611722742347, "reference_active": true, "delta_db": -0.1792984199275667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010887958720618104, "output_power": 9.63849528753164e-05, "reference_active": true, "delta_db": -0.5293722629650981}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8937894537156724e-05, "output_power": 2.5365140491222116e-05, "reference_active": true, "delta_db": -0.5722965743237578}], "active_band_count": 3, "active_band_mae_db": 0.4269890857388075}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015806536675822195, "log_rms_mae_db": 0.5062060085554234, "rms_envelope_nrmse": 0.03739729387640607, "envelope_correlation": 0.9982859455155793, "rms_gain_error_db": -0.11433543159494158, "reference_mean_power": 0.004504823508022685, "output_mean_power": 0.004387773869945071}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.0922276511075418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000711888936919089, "output_power": 0.0006778871537544728, "reference_active": true, "delta_db": -0.21254839928740815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.550528203435593e-05, "output_power": 6.204246629076114e-05, "reference_active": true, "delta_db": -0.23587267435537773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5248696894099073e-06, "output_power": 1.2844263797369487e-06, "reference_active": true, "delta_db": -0.7452351523195195}], "active_band_count": 3, "active_band_mae_db": 0.39788540865410177}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00016816220894332027, "log_rms_mae_db": 1.4331161446437226, "rms_envelope_nrmse": 0.08402705525597681, "envelope_correlation": 0.993686337957775, "rms_gain_error_db": -0.15316979001919723, "reference_mean_power": 0.002573247092376177, "output_mean_power": 0.002484073899828285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 9.221294414644421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007324784888809201, "output_power": 0.0006806564346901484, "reference_active": true, "delta_db": -0.318669204934241}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.576574656453582e-05, "output_power": 7.286439387765659e-05, "reference_active": true, "delta_db": -0.7079851579310181}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.804896765293062e-05, "output_power": 1.6312483324259727e-05, "reference_active": true, "delta_db": -0.43932285887201467}], "active_band_count": 3, "active_band_mae_db": 0.4886590739124246}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.0001473388036867024, "log_rms_mae_db": 0.8845880707272026, "rms_envelope_nrmse": 0.0519165297755269, "envelope_correlation": 0.9975524452018124, "rms_gain_error_db": -0.15317189122486266, "reference_mean_power": 0.002597106693874505, "output_mean_power": 0.002507105458747277}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.911179205548874e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006603877113347142, "output_power": 0.0006087885680331465, "reference_active": true, "delta_db": -0.35332494641021644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.506710493994746e-05, "output_power": 4.082089489410913e-05, "reference_active": true, "delta_db": -0.4297713909439957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1610111255482175e-06, "output_power": 9.550221519197475e-07, "reference_active": true, "delta_db": -0.848229362015392}], "active_band_count": 3, "active_band_mae_db": 0.5437752331232014}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 5.380259383983556e-05, "log_rms_mae_db": 1.4598920080413404, "rms_envelope_nrmse": 0.13832566918305458, "envelope_correlation": 0.9846837290731181, "rms_gain_error_db": -0.20153349990398617, "reference_mean_power": 0.00019859577428883629, "output_mean_power": 0.00018959053555661054}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 8.760825576371796e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.452705657171872e-05, "output_power": 4.594797845587183e-05, "reference_active": true, "delta_db": -0.7434564588033337}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.928893294666125e-06, "output_power": 5.177146606968398e-06, "reference_active": true, "delta_db": -0.5888317064307812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9788817769833428e-06, "output_power": 1.9234049606233543e-06, "reference_active": true, "delta_db": -0.12349117421707656}], "active_band_count": 3, "active_band_mae_db": 0.48525977981706386}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 5.051409218298903e-05, "log_rms_mae_db": 1.019835974457442, "rms_envelope_nrmse": 0.08607184232002121, "envelope_correlation": 0.9939216899576905, "rms_gain_error_db": -0.20146715058641382, "reference_mean_power": 0.00020041602526910178, "output_mean_power": 0.00019133117108907549}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.85711379260727e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.022793901144941e-05, "output_power": 4.2053111005054526e-05, "reference_active": true, "delta_db": -0.7714722868514061}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.619667417561746e-06, "output_power": 2.1732028173556625e-06, "reference_active": true, "delta_db": -0.8114589915431021}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1855266901360492e-07, "output_power": 1.1358646886064843e-07, "reference_active": true, "delta_db": -0.1858473728497001}], "active_band_count": 3, "active_band_mae_db": 0.5895928837480694}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 6.112392927022402e-05, "log_rms_mae_db": 1.5067798733480056, "rms_envelope_nrmse": 0.18069918647780425, "envelope_correlation": 0.9684001993233838, "rms_gain_error_db": -0.28299027329861065, "reference_mean_power": 8.711996692518435e-05, "output_mean_power": 8.162415052532555e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 4.203401537603793e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8248175978150172e-05, "output_power": 1.72423233636989e-05, "reference_active": true, "delta_db": -0.24623674830545675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.243431842943793e-05, "output_power": 1.1822420675045821e-05, "reference_active": true, "delta_db": -0.21915576276796783}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.767669752037568e-06, "output_power": 5.070883776681152e-06, "reference_active": true, "delta_db": -1.2535550114939733}], "active_band_count": 3, "active_band_mae_db": 0.5729825075224659}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 3.626522691368819e-05, "log_rms_mae_db": 0.9392787407583782, "rms_envelope_nrmse": 0.10787355208989102, "envelope_correlation": 0.9874086671456372, "rms_gain_error_db": -0.2829758869786889, "reference_mean_power": 8.789733624317315e-05, "output_mean_power": 8.235275360803263e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 3.223393789064049e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.051966991658498e-05, "output_power": 1.8817800887585064e-05, "reference_active": true, "delta_db": -0.3760150142846943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.880534452438109e-06, "output_power": 5.579408251709851e-06, "reference_active": true, "delta_db": -0.22828658329572193}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.871896444237529e-07, "output_power": 4.0911185311703363e-07, "reference_active": true, "delta_db": -1.5693632542324125}], "active_band_count": 3, "active_band_mae_db": 0.7245549506042762}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8953767766103765, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "616b2cf222496905bf046279b51ed6b6101743b3d236a452a94c7fba2c5cfbc6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9827567620991051, "standard_deviation": 0.0023598592433473936, "minimum": 0.97827397715505, "maximum": 0.985853122096162, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.468703757621729, "standard_deviation": 0.031858852883393025, "minimum": 1.4090843709019103, "maximum": 1.5096222171323628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7697869932952894, "standard_deviation": 0.2710614491841636, "minimum": 0.48525977981706386, "maximum": 1.1951557885808206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1468825250341781, "standard_deviation": 0.009550613832880557, "minimum": 0.13585749838526517, "maximum": 0.16588525011796412, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9933410119519397, "standard_deviation": 0.002257257953975409, "minimum": 0.9881936121653707, "maximum": 0.9951991194809757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9831527789545547, "standard_deviation": 0.04850519540592847, "minimum": 0.9173679443889547, "maximum": 1.0333636242975057, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.742147306287169, "standard_deviation": 0.27394608147242827, "minimum": 0.2982598745276613, "maximum": 1.0941867497613165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09026318889438345, "standard_deviation": 0.01217789970793822, "minimum": 0.07902576274017009, "maximum": 0.11849429373555254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9639483994671405, "standard_deviation": 0.005562039483506465, "minimum": 0.9573621792412299, "maximum": 0.9745749587371099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5313484673552482, "standard_deviation": 0.03667103606853208, "minimum": 1.4849871373427903, "maximum": 1.5973490728590163, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1415056855868972, "standard_deviation": 0.6631315700212514, "minimum": 0.38233086095373486, "maximum": 2.0179750359482487, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19609839960143674, "standard_deviation": 0.014894668399895538, "minimum": 0.16792909654889812, "maximum": 0.20974380636847464, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9853950228433319, "standard_deviation": 0.0031048883067026347, "minimum": 0.980921039876165, "maximum": 0.9904279227806354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9743894430687893, "standard_deviation": 0.0228713527575994, "minimum": 0.9392787407583782, "maximum": 1.0147180107113323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5128051189175973, "standard_deviation": 0.8469054774518313, "minimum": 0.5352953216223688, "maximum": 2.688652405032412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12300130637557184, "standard_deviation": 0.012837662977020348, "minimum": 0.1057260990292872, "maximum": 0.14333201596767076, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9944029404490302, "standard_deviation": 0.0005317774389341802, "minimum": 0.9936132469125574, "maximum": 0.994865234944714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4061970212290114, "standard_deviation": 0.04131297326136464, "minimum": 1.3323042802038583, "maximum": 1.4548927012922035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5352526497222339, "standard_deviation": 0.16346441214075766, "minimum": 0.30271289804188906, "maximum": 0.7187633694892508, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07919391339743403, "standard_deviation": 0.003347581869501004, "minimum": 0.0757020627280222, "maximum": 0.08402705525597681, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981493008001394, "standard_deviation": 0.0004340581980429736, "minimum": 0.9975524452018124, "maximum": 0.998828894746343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8546368922668311, "standard_deviation": 0.03911410221937639, "minimum": 0.8008299354664886, "maximum": 0.9207193363651993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5973234895963226, "standard_deviation": 0.23270643912018824, "minimum": 0.2386581460048062, "maximum": 0.9581107439280908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04547270528058053, "standard_deviation": 0.0052213327574371264, "minimum": 0.03821940588568458, "maximum": 0.0519165297755269, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19018279715767383, "standard_deviation": 0.04069023896874487, "minimum": 0.14853004057602912, "maximum": 0.25778616887293065, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.607453283014992, "standard_deviation": 0.010212526459464796, "minimum": 5.594352116463525, "maximum": 5.618308995611598, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3815180846766888, "standard_deviation": 0.01330799019245003, "minimum": 0.36528319008941446, "maximum": 0.40879387938805556, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09887503178702, "standard_deviation": 0.029883172514556494, "minimum": -0.15210981594087525, "maximum": -0.053256184325033255, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956263218835167, "standard_deviation": 0.000495075589048683, "minimum": 0.9950238725110795, "maximum": 0.9965944886622327, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8994204276028498, "standard_deviation": 0.037500943706935214, "minimum": 0.8398622654046971, "maximum": 0.9336367932519374, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.47984818998558587, "standard_deviation": 0.22522941785595454, "minimum": 0.06364260968579712, "maximum": 0.8190381391941827, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.060285514173691775, "standard_deviation": 0.0032989832714922195, "minimum": 0.053851270153961704, "maximum": 0.06393749094250253, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986396185376827, "standard_deviation": 0.0003175407487371677, "minimum": 0.9982306734781965, "maximum": 0.9991536912678195, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46825946982575983, "standard_deviation": 0.04509274694835819, "minimum": 0.4079266581943736, "maximum": 0.5394990278202741, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4992702025276037, "standard_deviation": 0.2600146933506777, "minimum": 0.06507441211900113, "maximum": 0.8066313222116618, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033200182495759255, "standard_deviation": 0.003731303090516346, "minimum": 0.02797561855606156, "maximum": 0.037467553254317415, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0027", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.8366286754608154, "predicted_mos": 2.8366286754608154, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "4639443e5495eed4a10b5c2ba3d0ea1f5a32abf77563bc38dbdff3811dc467c7", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.936265300028026, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9503081269712876, "全帯域_10ms_rms_envelope_nrmse": 0.0619745073562062, "全帯域_10ms_envelope_correlation": 0.9954137206702114, "全帯域_10ms_modulation_mae_db": 0.3725013360390516, "全帯域_40ms_log_rms_mae_db": 0.5207497577881234, "全帯域_40ms_rms_envelope_nrmse": 0.03784053606820438, "全帯域_40ms_envelope_correlation": 0.998242453478394, "全帯域_40ms_modulation_mae_db": 0.3535980721687477, "300–1000 Hz_10ms_log_rms_mae_db": 1.3801482505814464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08447895154648317, "300–1000 Hz_10ms_envelope_correlation": 0.9936159481249827, "300–1000 Hz_10ms_modulation_mae_db": 0.4953471988783759, "300–1000 Hz_40ms_log_rms_mae_db": 0.7761030164084637, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05189376006140664, "300–1000 Hz_40ms_envelope_correlation": 0.9975544077664855, "300–1000 Hz_40ms_modulation_mae_db": 0.6018358442683632, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3811099278292984, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1345141051514268, "1000–3000 Hz_10ms_envelope_correlation": 0.9857575509163125, "1000–3000 Hz_10ms_modulation_mae_db": 0.5180328921612256, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9175480962846763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08536373145760898, "1000–3000 Hz_40ms_envelope_correlation": 0.9943247677275446, "1000–3000 Hz_40ms_modulation_mae_db": 0.5927307593414831, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4997388664376377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18715647704304886, "3000–6000 Hz_10ms_envelope_correlation": 0.9659953240378483, "3000–6000 Hz_10ms_modulation_mae_db": 0.46938372522929456, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9255119622929167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11184348216008852, "3000–6000 Hz_40ms_envelope_correlation": 0.9860653931240038, "3000–6000 Hz_40ms_modulation_mae_db": 0.5279521460899642, "mr_stft_db": 5.54996136719661, "band_mae_db": 0.16835875327295935, "rms_gain_error_db": -0.11460250288261781, "output_peak": 0.377093210675066}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8773390947204877, "全帯域_10ms_rms_envelope_nrmse": 0.06259863300453168, "全帯域_10ms_envelope_correlation": 0.995233267375337, "全帯域_10ms_modulation_mae_db": 0.872897991263501, "全帯域_40ms_log_rms_mae_db": 0.4373118508441379, "全帯域_40ms_rms_envelope_nrmse": 0.03667734715412225, "全帯域_40ms_envelope_correlation": 0.9982569847511659, "全帯域_40ms_modulation_mae_db": 0.8464533643827883, "300–1000 Hz_10ms_log_rms_mae_db": 1.2953865426075137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08178726775326922, "300–1000 Hz_10ms_envelope_correlation": 0.9939192063618095, "300–1000 Hz_10ms_modulation_mae_db": 0.8054799085897493, "300–1000 Hz_40ms_log_rms_mae_db": 0.6947409568607084, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05079331206937008, "300–1000 Hz_40ms_envelope_correlation": 0.9975178528944545, "300–1000 Hz_40ms_modulation_mae_db": 0.9931515076789516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3719371146699764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14250555010876811, "1000–3000 Hz_10ms_envelope_correlation": 0.9838266778278877, "1000–3000 Hz_10ms_modulation_mae_db": 0.8309210475116114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8886741055274048, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08692182847977112, "1000–3000 Hz_40ms_envelope_correlation": 0.9940208832698252, "1000–3000 Hz_40ms_modulation_mae_db": 0.938210211217795, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5330475676028579, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19387670956251277, "3000–6000 Hz_10ms_envelope_correlation": 0.9662801171486253, "3000–6000 Hz_10ms_modulation_mae_db": 1.4294420646701311, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9764380584238654, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12367655700610453, "3000–6000 Hz_40ms_envelope_correlation": 0.9875444127290711, "3000–6000 Hz_40ms_modulation_mae_db": 2.057689437818615, "mr_stft_db": 5.5435139407717715, "band_mae_db": 0.201447227143413, "rms_gain_error_db": -0.08473801811498516, "output_peak": 0.40657728786899755}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.979154152222988, "全帯域_10ms_rms_envelope_nrmse": 0.0645736832511192, "全帯域_10ms_envelope_correlation": 0.9949175174079403, "全帯域_10ms_modulation_mae_db": 0.3561107275921313, "全帯域_40ms_log_rms_mae_db": 0.46440064545422155, "全帯域_40ms_rms_envelope_nrmse": 0.032013788531096245, "全帯域_40ms_envelope_correlation": 0.9987013626024925, "全帯域_40ms_modulation_mae_db": 0.34616090651490783, "300–1000 Hz_10ms_log_rms_mae_db": 1.3909692155883229, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08492845703109575, "300–1000 Hz_10ms_envelope_correlation": 0.9934369286897007, "300–1000 Hz_10ms_modulation_mae_db": 0.762508669734328, "300–1000 Hz_40ms_log_rms_mae_db": 0.7084131806253074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043393884942291745, "300–1000 Hz_40ms_envelope_correlation": 0.9982399703907515, "300–1000 Hz_40ms_modulation_mae_db": 1.0060904803587325, "1000–3000 Hz_10ms_log_rms_mae_db": 1.45127595951694, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13584069714744568, "1000–3000 Hz_10ms_envelope_correlation": 0.9853729643823645, "1000–3000 Hz_10ms_modulation_mae_db": 0.8131972793466254, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9700914164210858, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08284557555172657, "1000–3000 Hz_40ms_envelope_correlation": 0.9946111245829231, "1000–3000 Hz_40ms_modulation_mae_db": 0.703275244479786, "3000–6000 Hz_10ms_log_rms_mae_db": 1.494307004508546, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19392045717116052, "3000–6000 Hz_10ms_envelope_correlation": 0.9638896289895106, "3000–6000 Hz_10ms_modulation_mae_db": 0.5832624496516914, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9413546397523603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12432681854123535, "3000–6000 Hz_40ms_envelope_correlation": 0.9827017342501895, "3000–6000 Hz_40ms_modulation_mae_db": 0.7186585141639346, "mr_stft_db": 5.547892739549478, "band_mae_db": 0.1608397861554994, "rms_gain_error_db": -0.08368587034144076, "output_peak": 0.3768444714798556}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9196606443258438, "全帯域_10ms_rms_envelope_nrmse": 0.059512832134168324, "全帯域_10ms_envelope_correlation": 0.9956787555774654, "全帯域_10ms_modulation_mae_db": 0.0910688317100911, "全帯域_40ms_log_rms_mae_db": 0.4942285071342268, "全帯域_40ms_rms_envelope_nrmse": 0.03169430071034543, "全帯域_40ms_envelope_correlation": 0.9986979681625776, "全帯域_40ms_modulation_mae_db": 0.07868168033026939, "300–1000 Hz_10ms_log_rms_mae_db": 1.3208542720606833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07925597541380508, "300–1000 Hz_10ms_envelope_correlation": 0.9942583394883201, "300–1000 Hz_10ms_modulation_mae_db": 0.31724490379362513, "300–1000 Hz_40ms_log_rms_mae_db": 0.7600168053829648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04462304909488542, "300–1000 Hz_40ms_envelope_correlation": 0.9980709415873734, "300–1000 Hz_40ms_modulation_mae_db": 0.23884947132828982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4513262815023036, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14276779399777761, "1000–3000 Hz_10ms_envelope_correlation": 0.9836320542711626, "1000–3000 Hz_10ms_modulation_mae_db": 0.8856753236310945, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9271687060176999, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08469091921886658, "1000–3000 Hz_40ms_envelope_correlation": 0.9941560910929502, "1000–3000 Hz_40ms_modulation_mae_db": 0.9177436060016051, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4908115175017638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17205587225253452, "3000–6000 Hz_10ms_envelope_correlation": 0.9742538630332351, "3000–6000 Hz_10ms_modulation_mae_db": 1.2752403942119301, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9577838428570458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11092584801715151, "3000–6000 Hz_40ms_envelope_correlation": 0.9906661937795788, "3000–6000 Hz_40ms_modulation_mae_db": 1.6839513535501072, "mr_stft_db": 5.561879584502343, "band_mae_db": 0.1942399396922725, "rms_gain_error_db": -0.07125349985294963, "output_peak": 0.36437672751227024}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9443613741977762, "全帯域_10ms_rms_envelope_nrmse": 0.06370060409568694, "全帯域_10ms_envelope_correlation": 0.9950767034246386, "全帯域_10ms_modulation_mae_db": 0.4560935107872209, "全帯域_40ms_log_rms_mae_db": 0.5540115845295702, "全帯域_40ms_rms_envelope_nrmse": 0.03568073302915666, "全帯域_40ms_envelope_correlation": 0.9983789118379991, "全帯域_40ms_modulation_mae_db": 0.5005008940547371, "300–1000 Hz_10ms_log_rms_mae_db": 1.3436136509674512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07950972331982611, "300–1000 Hz_10ms_envelope_correlation": 0.9942448478564532, "300–1000 Hz_10ms_modulation_mae_db": 0.30119295557042974, "300–1000 Hz_40ms_log_rms_mae_db": 0.7725875012818996, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04495179216996169, "300–1000 Hz_40ms_envelope_correlation": 0.9980702487673164, "300–1000 Hz_40ms_modulation_mae_db": 0.4768894636812054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4326951150190486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13623245946195955, "1000–3000 Hz_10ms_envelope_correlation": 0.9858064857256138, "1000–3000 Hz_10ms_modulation_mae_db": 0.9237800485438846, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9155367605489226, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08575927587473484, "1000–3000 Hz_40ms_envelope_correlation": 0.9949059357381133, "1000–3000 Hz_40ms_modulation_mae_db": 0.857346672787591, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6019936015695901, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21425876697847587, "3000–6000 Hz_10ms_envelope_correlation": 0.9575109936785517, "3000–6000 Hz_10ms_modulation_mae_db": 1.9075079433173832, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0759842702182674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12972883278079334, "3000–6000 Hz_40ms_envelope_correlation": 0.9856700062325128, "3000–6000 Hz_40ms_modulation_mae_db": 2.70907805596388, "mr_stft_db": 5.544474680044437, "band_mae_db": 0.2761168709564416, "rms_gain_error_db": -0.0914819351755535, "output_peak": 0.38266576329274454}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8363055474368416, "全帯域_10ms_rms_envelope_nrmse": 0.055216418598594615, "全帯域_10ms_envelope_correlation": 0.9963965501765357, "全帯域_10ms_modulation_mae_db": 0.3685474867588687, "全帯域_40ms_log_rms_mae_db": 0.4340193908990214, "全帯域_40ms_rms_envelope_nrmse": 0.028364663274899633, "全帯域_40ms_envelope_correlation": 0.9991084415157, "全帯域_40ms_modulation_mae_db": 0.3713726905788863, "300–1000 Hz_10ms_log_rms_mae_db": 1.2790231319219607, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07631062767030866, "300–1000 Hz_10ms_envelope_correlation": 0.9949261570381832, "300–1000 Hz_10ms_modulation_mae_db": 0.3899745757527218, "300–1000 Hz_40ms_log_rms_mae_db": 0.7307291886126055, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04569743235131266, "300–1000 Hz_40ms_envelope_correlation": 0.9982719014895649, "300–1000 Hz_40ms_modulation_mae_db": 0.5529806642418446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.397718060369816, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15916186170715663, "1000–3000 Hz_10ms_envelope_correlation": 0.9798375460321083, "1000–3000 Hz_10ms_modulation_mae_db": 0.4616283382280224, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9579072951640747, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11201603876864198, "1000–3000 Hz_40ms_envelope_correlation": 0.9893329758399377, "1000–3000 Hz_40ms_modulation_mae_db": 0.6803181353445735, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5683484975561435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22293201897813886, "3000–6000 Hz_10ms_envelope_correlation": 0.9520252975618752, "3000–6000 Hz_10ms_modulation_mae_db": 0.6387134932729827, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9933505105897688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12999189590176222, "3000–6000 Hz_40ms_envelope_correlation": 0.9827767672944204, "3000–6000 Hz_40ms_modulation_mae_db": 0.8458288686270449, "mr_stft_db": 5.559672850424779, "band_mae_db": 0.21299983881116788, "rms_gain_error_db": -0.11276289192060583, "output_peak": 0.376893284323713}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8730030574913362, "全帯域_10ms_rms_envelope_nrmse": 0.06235794909189895, "全帯域_10ms_envelope_correlation": 0.9953626372289464, "全帯域_10ms_modulation_mae_db": 0.563055132095461, "全帯域_40ms_log_rms_mae_db": 0.3948836698237954, "全帯域_40ms_rms_envelope_nrmse": 0.03217949279953348, "全帯域_40ms_envelope_correlation": 0.9988042355228277, "全帯域_40ms_modulation_mae_db": 0.7239683224573842, "300–1000 Hz_10ms_log_rms_mae_db": 1.3543914952332783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07737922736778646, "300–1000 Hz_10ms_envelope_correlation": 0.9947246046627692, "300–1000 Hz_10ms_modulation_mae_db": 0.6720768364338553, "300–1000 Hz_40ms_log_rms_mae_db": 0.7397396118847317, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038867133733161655, "300–1000 Hz_40ms_envelope_correlation": 0.9988101583295065, "300–1000 Hz_40ms_modulation_mae_db": 0.7909592211253499, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3897844596181095, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14355913498205816, "1000–3000 Hz_10ms_envelope_correlation": 0.9840116810921822, "1000–3000 Hz_10ms_modulation_mae_db": 1.1396998607780235, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8651069489716979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08351727018161063, "1000–3000 Hz_40ms_envelope_correlation": 0.9951920293192738, "1000–3000 Hz_40ms_modulation_mae_db": 1.0536505607660271, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5232252850774388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21484859988630245, "3000–6000 Hz_10ms_envelope_correlation": 0.9553526339228307, "3000–6000 Hz_10ms_modulation_mae_db": 0.6561617211010883, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0022676649516766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11971043027788374, "3000–6000 Hz_40ms_envelope_correlation": 0.9840022583399554, "3000–6000 Hz_40ms_modulation_mae_db": 0.9782374094185945, "mr_stft_db": 5.5626802682364955, "band_mae_db": 0.17938263321544431, "rms_gain_error_db": -0.11538930113034475, "output_peak": 0.3738472128841773}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9626220920065534, "全帯域_10ms_rms_envelope_nrmse": 0.059300580916014034, "全帯域_10ms_envelope_correlation": 0.9959024951043414, "全帯域_10ms_modulation_mae_db": 0.5332459238980669, "全帯域_40ms_log_rms_mae_db": 0.4779036592482399, "全帯域_40ms_rms_envelope_nrmse": 0.034803785471539464, "全帯域_40ms_envelope_correlation": 0.9986468090119385, "全帯域_40ms_modulation_mae_db": 0.6451178885164435, "300–1000 Hz_10ms_log_rms_mae_db": 1.4032366189284828, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07557323733829159, "300–1000 Hz_10ms_envelope_correlation": 0.9951232911750932, "300–1000 Hz_10ms_modulation_mae_db": 0.545530213345675, "300–1000 Hz_40ms_log_rms_mae_db": 0.7893702427937247, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046533864955803854, "300–1000 Hz_40ms_envelope_correlation": 0.9983005869531277, "300–1000 Hz_40ms_modulation_mae_db": 0.5998555763151724, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4349738102562093, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14241705580046898, "1000–3000 Hz_10ms_envelope_correlation": 0.9836680241673771, "1000–3000 Hz_10ms_modulation_mae_db": 0.451262867377002, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9262846927282244, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08791447357636802, "1000–3000 Hz_40ms_envelope_correlation": 0.9935547423756029, "1000–3000 Hz_40ms_modulation_mae_db": 0.4377002996752976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5545187275511598, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18893650312195231, "3000–6000 Hz_10ms_envelope_correlation": 0.9673381044617365, "3000–6000 Hz_10ms_modulation_mae_db": 1.6727283630088323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0067130436245522, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12390618360977579, "3000–6000 Hz_40ms_envelope_correlation": 0.9860967060750883, "3000–6000 Hz_40ms_modulation_mae_db": 2.08515751715554, "mr_stft_db": 5.575084046110394, "band_mae_db": 0.21362210346501911, "rms_gain_error_db": -0.1371686588965419, "output_peak": 0.39195302566456136}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 52320, "output_samples": 52320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5442728646496775, "spectral_convergence": 0.20834111993380583, "spectral_convergence_status": "ok", "amplitude_floor": 1.2746778473169441e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.430821433534925, "spectral_convergence": 0.19769714812991024, "spectral_convergence_status": "ok", "amplitude_floor": 1.090624511245042e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.674789803405227, "spectral_convergence": 0.2047732058251833, "spectral_convergence_status": "ok", "amplitude_floor": 7.935262595002381e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004730173956506593, "output_power": 0.0004797283856390837, "delta_db": 0.0611830391939619, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018195471619226037, "output_power": 0.00181244141454893, "delta_db": -0.016993394921613118, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018173989836957952, "output_power": 0.0017188391554303658, "delta_db": -0.24215042433398154, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013671881984837986, "output_power": 0.00013278987674514815, "delta_db": -0.12663333122218398, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012324536624073353, "output_power": 0.00011551970260218818, "delta_db": -0.2811453755814529, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.679288082514536e-05, "output_power": 8.133536280954213e-05, "delta_db": -0.2820469543845626, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16835875327295935, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11460250288261781, "pitch": null, "output_rms_dbfs": -23.62444841269966, "output_peak": 0.377093210675066, "reference_rms_dbfs": -23.509845909817045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.54996136719661}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 52320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00017439460044090568, "log_rms_mae_db": 0.9503081269712876, "rms_envelope_nrmse": 0.0619745073562062, "envelope_correlation": 0.9954137206702114, "rms_gain_error_db": -0.11460149563463962, "reference_mean_power": 0.004463526965610992, "output_mean_power": 0.004347284004978134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 1.3056377024951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008270090313661458, "output_power": 0.0007955947187791145, "reference_active": true, "delta_db": -0.16818360717967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010887958720618104, "output_power": 9.799035553857982e-05, "reference_active": true, "delta_db": -0.457631322409886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8937894537156724e-05, "output_power": 2.5840327981241534e-05, "reference_active": true, "delta_db": -0.4916890785275986}], "active_band_count": 3, "active_band_mae_db": 0.3725013360390516}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015806536675822195, "log_rms_mae_db": 0.5207497577881234, "rms_envelope_nrmse": 0.03784053606820438, "envelope_correlation": 0.998242453478394, "rms_gain_error_db": -0.11460275345223503, "reference_mean_power": 0.004504823508022685, "output_mean_power": 0.004387503797031275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.0922276511075418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000711888936919089, "output_power": 0.0006788903208383988, "reference_active": true, "delta_db": -0.20612627013912144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.550528203435593e-05, "output_power": 6.330698416088416e-05, "reference_active": true, "delta_db": -0.1482469589535279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5248696894099073e-06, "output_power": 1.295957123082368e-06, "reference_active": true, "delta_db": -0.7064209874135938}], "active_band_count": 3, "active_band_mae_db": 0.3535980721687477}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00016816220894332027, "log_rms_mae_db": 1.3801482505814464, "rms_envelope_nrmse": 0.08447895154648317, "envelope_correlation": 0.9936159481249827, "rms_gain_error_db": -0.15048697473204525, "reference_mean_power": 0.002573247092376177, "output_mean_power": 0.0024856088883082094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 9.221294414644421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007324784888809201, "output_power": 0.000685252564312583, "reference_active": true, "delta_db": -0.2894420563008264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.576574656453582e-05, "output_power": 7.270795266162765e-05, "reference_active": true, "delta_db": -0.7173195658229095}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.804896765293062e-05, "output_power": 1.61630891796646e-05, "reference_active": true, "delta_db": -0.4792799745113919}], "active_band_count": 3, "active_band_mae_db": 0.4953471988783759}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.0001473388036867024, "log_rms_mae_db": 0.7761030164084637, "rms_envelope_nrmse": 0.05189376006140664, "envelope_correlation": 0.9975544077664855, "rms_gain_error_db": -0.15048858666594397, "reference_mean_power": 0.002597106693874505, "output_mean_power": 0.002508654961784447}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.911179205548874e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006603877113347142, "output_power": 0.0006131224849346232, "reference_active": true, "delta_db": -0.3225173997638891}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.506710493994746e-05, "output_power": 4.102428285504478e-05, "reference_active": true, "delta_db": -0.4081866248345298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1610111255482175e-06, "output_power": 9.06475416594075e-07, "reference_active": true, "delta_db": -1.0748035082066707}], "active_band_count": 3, "active_band_mae_db": 0.6018358442683632}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 5.380259383983556e-05, "log_rms_mae_db": 1.3811099278292984, "rms_envelope_nrmse": 0.1345141051514268, "envelope_correlation": 0.9857575509163125, "rms_gain_error_db": -0.26467674768940697, "reference_mean_power": 0.00019859577428883629, "output_mean_power": 0.00018685396959300892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 8.760825576371796e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.452705657171872e-05, "output_power": 4.5663150579225724e-05, "reference_active": true, "delta_db": -0.7704618183057197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.928893294666125e-06, "output_power": 5.183710111008557e-06, "reference_active": true, "delta_db": -0.5833292773450565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9788817769833428e-06, "output_power": 1.88968358763527e-06, "reference_active": true, "delta_db": -0.20030758083290073}], "active_band_count": 3, "active_band_mae_db": 0.5180328921612256}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 5.051409218298903e-05, "log_rms_mae_db": 0.9175480962846763, "rms_envelope_nrmse": 0.08536373145760898, "envelope_correlation": 0.9943247677275446, "rms_gain_error_db": -0.2646031427369415, "reference_mean_power": 0.00020041602526910178, "output_mean_power": 0.00018856979568166322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.85711379260727e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.022793901144941e-05, "output_power": 4.184313440092942e-05, "reference_active": true, "delta_db": -0.793211488759709}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.619667417561746e-06, "output_power": 2.193657315014021e-06, "reference_active": true, "delta_db": -0.7707737387004209}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1855266901360492e-07, "output_power": 1.128471592257062e-07, "reference_active": true, "delta_db": -0.21420705056431938}], "active_band_count": 3, "active_band_mae_db": 0.5927307593414831}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 6.112392927022402e-05, "log_rms_mae_db": 1.4997388664376377, "rms_envelope_nrmse": 0.18715647704304886, "envelope_correlation": 0.9659953240378483, "rms_gain_error_db": -0.22451332184381903, "reference_mean_power": 8.711996692518435e-05, "output_mean_power": 8.273063726918586e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 4.203401537603793e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8248175978150172e-05, "output_power": 1.8658790924919847e-05, "reference_active": true, "delta_db": 0.09664037926957472}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.243431842943793e-05, "output_power": 1.2640364125610122e-05, "reference_active": true, "delta_db": 0.07137599677356857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.767669752037568e-06, "output_power": 5.086577641246362e-06, "reference_active": true, "delta_db": -1.2401347996447405}], "active_band_count": 3, "active_band_mae_db": 0.46938372522929456}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 3.626522691368819e-05, "log_rms_mae_db": 0.9255119622929167, "rms_envelope_nrmse": 0.11184348216008852, "envelope_correlation": 0.9860653931240038, "rms_gain_error_db": -0.22445237397082718, "reference_mean_power": 8.789733624317315e-05, "output_mean_power": 8.347001209724977e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 3.223393789064049e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.051966991658498e-05, "output_power": 2.049507154254664e-05, "reference_active": true, "delta_db": -0.005209316947736886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.880534452438109e-06, "output_power": 5.897877368713593e-06, "reference_active": true, "delta_db": 0.012789395372056873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.871896444237529e-07, "output_power": 4.0944221239985257e-07, "reference_active": true, "delta_db": -1.565857725950099}], "active_band_count": 3, "active_band_mae_db": 0.5279521460899642}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8920667735524852, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3b492624b86f608da08586ed39d0f3c27ac52ecdb49bb5a273e73833a5ae6aca", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.983989123051876, "standard_deviation": 0.001921244117098836, "minimum": 0.9798375460321083, "maximum": 0.9858064857256138, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4138525910977127, "standard_deviation": 0.032236890535200506, "minimum": 1.3719371146699764, "maximum": 1.4513262815023036, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7530247071971862, "standard_deviation": 0.24990084612642396, "minimum": 0.451262867377002, "maximum": 1.1396998607780235, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14212483229463269, "standard_deviation": 0.007792807265871176, "minimum": 0.1345141051514268, "maximum": 0.15916186170715663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937623187432714, "standard_deviation": 0.0018621351476521802, "minimum": 0.9893329758399377, "maximum": 0.9951920293192738, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9210397527079732, "standard_deviation": 0.03390921303961974, "minimum": 0.8651069489716979, "maximum": 0.9700914164210858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7726219362017698, "standard_deviation": 0.2044146479632264, "minimum": 0.4377002996752976, "maximum": 1.0536505607660271, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08862863913866609, "standard_deviation": 0.009594359350804828, "minimum": 0.08284557555172657, "maximum": 0.11201603876864198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9628307453542766, "standard_deviation": 0.007317469045967824, "minimum": 0.9520252975618752, "maximum": 0.9742538630332351, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5332488834756424, "standard_deviation": 0.03955296576295539, "minimum": 1.4908115175017638, "maximum": 1.6019936015695901, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0790550193079167, "standard_deviation": 0.5593734798062152, "minimum": 0.46938372522929456, "maximum": 1.9075079433173832, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19849817562426578, "standard_deviation": 0.017214254411354178, "minimum": 0.17205587225253452, "maximum": 0.22293201897813886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9856904339781025, "standard_deviation": 0.002639621403431185, "minimum": 0.9827017342501895, "maximum": 0.9906661937795788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9849254990888066, "standard_deviation": 0.04686641002751207, "minimum": 0.9255119622929167, "maximum": 1.0759842702182674, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4508191628484601, "standard_deviation": 0.7915023919025846, "minimum": 0.5279521460899642, "maximum": 2.70907805596388, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12176375603684939, "standard_deviation": 0.007231390468287744, "minimum": 0.11092584801715151, "maximum": 0.12999189590176222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9942811654246639, "standard_deviation": 0.0006106123848479081, "minimum": 0.9934369286897007, "maximum": 0.9951232911750932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3459528972361425, "standard_deviation": 0.04507198778418061, "minimum": 1.2790231319219607, "maximum": 1.4032366189284828, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.536169407762345, "standard_deviation": 0.1958488171970329, "minimum": 0.30119295557042974, "maximum": 0.8054799085897493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07990293343010826, "standard_deviation": 0.0035524586090075884, "minimum": 0.07557323733829159, "maximum": 0.08492845703109575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981045085223226, "standard_deviation": 0.00041949542171908433, "minimum": 0.9975178528944545, "maximum": 0.9988101583295065, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7464625629813006, "standard_deviation": 0.0338092813385089, "minimum": 0.6947409568607084, "maximum": 0.7893702427937247, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6575765286247386, "standard_deviation": 0.26110279169139766, "minimum": 0.23884947132828982, "maximum": 1.0060904803587325, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04584427867227422, "standard_deviation": 0.004115461517224017, "minimum": 0.038867133733161655, "maximum": 0.05189376006140664, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20087589408902715, "standard_deviation": 0.03612920751484819, "minimum": 0.1608397861554994, "maximum": 0.2761168709564416, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.555644934604539, "standard_deviation": 0.010998733729801109, "minimum": 5.5435139407717715, "maximum": 5.575084046110394, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3812813729626732, "standard_deviation": 0.012835218918602615, "minimum": 0.36437672751227024, "maximum": 0.40657728786899755, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10138533478937992, "standard_deviation": 0.021964610142113195, "minimum": -0.1371686588965419, "maximum": -0.07125349985294963, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954977058706771, "standard_deviation": 0.000480125115799496, "minimum": 0.9949175174079403, "maximum": 0.9963965501765357, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9178442611716393, "standard_deviation": 0.050461130561663875, "minimum": 0.8363055474368416, "maximum": 0.979154152222988, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4516901175180491, "standard_deviation": 0.2232409558307701, "minimum": 0.0910688317100911, "maximum": 0.872897991263501, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0611544010560275, "standard_deviation": 0.0030183387599728062, "minimum": 0.055216418598594615, "maximum": 0.0645736832511192, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998604645860387, "standard_deviation": 0.0002967669659624502, "minimum": 0.998242453478394, "maximum": 0.9991084415157, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4721886332151671, "standard_deviation": 0.051104356638949855, "minimum": 0.3948836698237954, "maximum": 0.5540115845295702, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.48323172737552056, "standard_deviation": 0.24731584729411565, "minimum": 0.07868168033026939, "maximum": 0.8464533643827883, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033656830879862196, "standard_deviation": 0.0031345829296512214, "minimum": 0.028364663274899633, "maximum": 0.03784053606820438, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0027", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5626752376556396, "predicted_mos": 2.5626752376556396, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "4261f68c883aa6e187c9b399d8ab25445ee4a58f4ba9a6fa06d0515a93ce0596", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.202333299908787, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.835376285526767, "全帯域_10ms_rms_envelope_nrmse": 0.059570135986125095, "全帯域_10ms_envelope_correlation": 0.9956968592178179, "全帯域_10ms_modulation_mae_db": 0.28000849795260346, "全帯域_40ms_log_rms_mae_db": 0.4103824596163088, "全帯域_40ms_rms_envelope_nrmse": 0.03157870784758466, "全帯域_40ms_envelope_correlation": 0.9987411432860397, "全帯域_40ms_modulation_mae_db": 0.25506213109314185, "300–1000 Hz_10ms_log_rms_mae_db": 1.623219869826778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08217438818182304, "300–1000 Hz_10ms_envelope_correlation": 0.993891445992195, "300–1000 Hz_10ms_modulation_mae_db": 0.34667568425649975, "300–1000 Hz_40ms_log_rms_mae_db": 1.12663025487867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04639147839452062, "300–1000 Hz_40ms_envelope_correlation": 0.9979994441981263, "300–1000 Hz_40ms_modulation_mae_db": 0.34240838905546017, "1000–3000 Hz_10ms_log_rms_mae_db": 1.859868284155472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1422007951195526, "1000–3000 Hz_10ms_envelope_correlation": 0.9836650556301658, "1000–3000 Hz_10ms_modulation_mae_db": 0.961651893560397, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2203510083591598, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0824392686674026, "1000–3000 Hz_40ms_envelope_correlation": 0.9942741250352013, "1000–3000 Hz_40ms_modulation_mae_db": 0.7983377644456936, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9810063995510632, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20660399738032972, "3000–6000 Hz_10ms_envelope_correlation": 0.958680610112849, "3000–6000 Hz_10ms_modulation_mae_db": 1.2831329182983222, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2236589597771594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11825823657282009, "3000–6000 Hz_40ms_envelope_correlation": 0.9853397336171076, "3000–6000 Hz_40ms_modulation_mae_db": 1.626594154036816, "mr_stft_db": 5.926707236959184, "band_mae_db": 0.1619889815558931, "rms_gain_error_db": -0.08379593512035055, "output_peak": 0.37227892787430544}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8530449607069529, "全帯域_10ms_rms_envelope_nrmse": 0.057394748183321645, "全帯域_10ms_envelope_correlation": 0.995952981089339, "全帯域_10ms_modulation_mae_db": 0.3021751182685108, "全帯域_40ms_log_rms_mae_db": 0.4518740300726467, "全帯域_40ms_rms_envelope_nrmse": 0.027266285764288713, "全帯域_40ms_envelope_correlation": 0.9990123037240998, "全帯域_40ms_modulation_mae_db": 0.5011827225827077, "300–1000 Hz_10ms_log_rms_mae_db": 1.605803803708744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08248967287202698, "300–1000 Hz_10ms_envelope_correlation": 0.9937864247231283, "300–1000 Hz_10ms_modulation_mae_db": 0.6230104521797176, "300–1000 Hz_40ms_log_rms_mae_db": 1.1227831778185355, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04299639054785317, "300–1000 Hz_40ms_envelope_correlation": 0.9982315212321177, "300–1000 Hz_40ms_modulation_mae_db": 0.8713484995895625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8946853119505236, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15813629088275355, "1000–3000 Hz_10ms_envelope_correlation": 0.9797813870903933, "1000–3000 Hz_10ms_modulation_mae_db": 0.907069737930855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.311699154225425, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10114258774041404, "1000–3000 Hz_40ms_envelope_correlation": 0.9911397683207629, "1000–3000 Hz_40ms_modulation_mae_db": 0.6552518771805937, "3000–6000 Hz_10ms_log_rms_mae_db": 1.967527636948046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1974691993317382, "3000–6000 Hz_10ms_envelope_correlation": 0.9620652557413137, "3000–6000 Hz_10ms_modulation_mae_db": 0.9815541260650278, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2848350084525229, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12275808215538088, "3000–6000 Hz_40ms_envelope_correlation": 0.9831214382227861, "3000–6000 Hz_40ms_modulation_mae_db": 1.163246970324778, "mr_stft_db": 5.910201653088348, "band_mae_db": 0.15856670232766248, "rms_gain_error_db": -0.048218522070211225, "output_peak": 0.3674872515101474}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.893345483813465, "全帯域_10ms_rms_envelope_nrmse": 0.06117281599095404, "全帯域_10ms_envelope_correlation": 0.9954122387690088, "全帯域_10ms_modulation_mae_db": 0.16255019133309115, "全帯域_40ms_log_rms_mae_db": 0.46899018089564726, "全帯域_40ms_rms_envelope_nrmse": 0.03593650912953303, "全帯域_40ms_envelope_correlation": 0.9982729685952144, "全帯域_40ms_modulation_mae_db": 0.2084247251413273, "300–1000 Hz_10ms_log_rms_mae_db": 1.5643409490882458, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08323147689749542, "300–1000 Hz_10ms_envelope_correlation": 0.9936393389360721, "300–1000 Hz_10ms_modulation_mae_db": 0.2965716243815206, "300–1000 Hz_40ms_log_rms_mae_db": 1.0351905987607082, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04936466752461592, "300–1000 Hz_40ms_envelope_correlation": 0.9975830515709826, "300–1000 Hz_40ms_modulation_mae_db": 0.38884829473854204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8158694929056123, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14107525699398446, "1000–3000 Hz_10ms_envelope_correlation": 0.9839619596132254, "1000–3000 Hz_10ms_modulation_mae_db": 0.37899715045110055, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2233021363513366, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08442971026698465, "1000–3000 Hz_40ms_envelope_correlation": 0.9940465951534682, "1000–3000 Hz_40ms_modulation_mae_db": 0.3282788352136614, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9404670942320927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20066393843389277, "3000–6000 Hz_10ms_envelope_correlation": 0.963106423219534, "3000–6000 Hz_10ms_modulation_mae_db": 1.1608231622327938, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3049824488823838, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13129864913385828, "3000–6000 Hz_40ms_envelope_correlation": 0.9844379766207896, "3000–6000 Hz_40ms_modulation_mae_db": 1.416759697743804, "mr_stft_db": 5.904433048879704, "band_mae_db": 0.17910574998657736, "rms_gain_error_db": -0.05898759154403722, "output_peak": 0.3907119728673631}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9612491333671107, "全帯域_10ms_rms_envelope_nrmse": 0.06613889520577068, "全帯域_10ms_envelope_correlation": 0.9949553045413642, "全帯域_10ms_modulation_mae_db": 0.4631592136567442, "全帯域_40ms_log_rms_mae_db": 0.49819514281644384, "全帯域_40ms_rms_envelope_nrmse": 0.03776104824761334, "全帯域_40ms_envelope_correlation": 0.9985094843146161, "全帯域_40ms_modulation_mae_db": 0.4713018612026801, "300–1000 Hz_10ms_log_rms_mae_db": 1.5966351531614345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0857326677551902, "300–1000 Hz_10ms_envelope_correlation": 0.9937903739528724, "300–1000 Hz_10ms_modulation_mae_db": 0.991658068306375, "300–1000 Hz_40ms_log_rms_mae_db": 1.0418663595925202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053297718828412735, "300–1000 Hz_40ms_envelope_correlation": 0.9978492644584227, "300–1000 Hz_40ms_modulation_mae_db": 0.9313563244705568, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8856674094290904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14728395880850462, "1000–3000 Hz_10ms_envelope_correlation": 0.982496220453778, "1000–3000 Hz_10ms_modulation_mae_db": 0.4159997116670788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2027095401031154, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08653490739724316, "1000–3000 Hz_40ms_envelope_correlation": 0.9937699503879505, "1000–3000 Hz_40ms_modulation_mae_db": 0.6096006701302303, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9775460234862705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2265557652935587, "3000–6000 Hz_10ms_envelope_correlation": 0.9497172868704985, "3000–6000 Hz_10ms_modulation_mae_db": 0.6352125697354521, "3000–6000 Hz_40ms_log_rms_mae_db": 1.224428834393065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13229317497738613, "3000–6000 Hz_40ms_envelope_correlation": 0.9797268423375964, "3000–6000 Hz_40ms_modulation_mae_db": 0.701869613051784, "mr_stft_db": 5.910604149290866, "band_mae_db": 0.13173606221235923, "rms_gain_error_db": -0.16610964245435317, "output_peak": 0.3600570437638728}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.94004982448159, "全帯域_10ms_rms_envelope_nrmse": 0.05368752226253446, "全帯域_10ms_envelope_correlation": 0.9964758741416104, "全帯域_10ms_modulation_mae_db": 0.38242433478735655, "全帯域_40ms_log_rms_mae_db": 0.5385598338734353, "全帯域_40ms_rms_envelope_nrmse": 0.029448936341955004, "全帯域_40ms_envelope_correlation": 0.9988591363627822, "全帯域_40ms_modulation_mae_db": 0.4513713548481435, "300–1000 Hz_10ms_log_rms_mae_db": 1.5291074178488089, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06955699945197945, "300–1000 Hz_10ms_envelope_correlation": 0.9955697063482626, "300–1000 Hz_10ms_modulation_mae_db": 0.24417104990864372, "300–1000 Hz_40ms_log_rms_mae_db": 1.089224100165117, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03974820210381392, "300–1000 Hz_40ms_envelope_correlation": 0.9984450583199146, "300–1000 Hz_40ms_modulation_mae_db": 0.22101568903091517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8125725000719541, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12943820566615652, "1000–3000 Hz_10ms_envelope_correlation": 0.986756367578333, "1000–3000 Hz_10ms_modulation_mae_db": 0.5031904519123795, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1900138096423656, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07609732744356881, "1000–3000 Hz_40ms_envelope_correlation": 0.9955183182076233, "1000–3000 Hz_40ms_modulation_mae_db": 0.5680037177754846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9589399333400692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2106830718701034, "3000–6000 Hz_10ms_envelope_correlation": 0.957695347914285, "3000–6000 Hz_10ms_modulation_mae_db": 1.8764307870649601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1652646031485394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12565844778013363, "3000–6000 Hz_40ms_envelope_correlation": 0.9842771371535083, "3000–6000 Hz_40ms_modulation_mae_db": 2.3889265789791474, "mr_stft_db": 5.926532720801068, "band_mae_db": 0.1533848066597187, "rms_gain_error_db": -0.058606128382022615, "output_peak": 0.35970288898243424}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9119295942749824, "全帯域_10ms_rms_envelope_nrmse": 0.06524778226061452, "全帯域_10ms_envelope_correlation": 0.9948026490203771, "全帯域_10ms_modulation_mae_db": 0.28538791935433516, "全帯域_40ms_log_rms_mae_db": 0.5075682979469796, "全帯域_40ms_rms_envelope_nrmse": 0.030996510011689578, "全帯域_40ms_envelope_correlation": 0.9987816365964436, "全帯域_40ms_modulation_mae_db": 0.44040620639296474, "300–1000 Hz_10ms_log_rms_mae_db": 1.7447423997125062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08681660933195638, "300–1000 Hz_10ms_envelope_correlation": 0.9932721759588303, "300–1000 Hz_10ms_modulation_mae_db": 0.5295059171323208, "300–1000 Hz_40ms_log_rms_mae_db": 1.0671612589224622, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04487850230517835, "300–1000 Hz_40ms_envelope_correlation": 0.9982719753725895, "300–1000 Hz_40ms_modulation_mae_db": 0.5934870804781002, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8502422983672338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1679924920769119, "1000–3000 Hz_10ms_envelope_correlation": 0.977069230630835, "1000–3000 Hz_10ms_modulation_mae_db": 0.7104396823261943, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2259714297087356, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10185840693290381, "1000–3000 Hz_40ms_envelope_correlation": 0.9910459249332451, "1000–3000 Hz_40ms_modulation_mae_db": 0.6984042112258323, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8832601836583334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20201929864464474, "3000–6000 Hz_10ms_envelope_correlation": 0.9610736587192685, "3000–6000 Hz_10ms_modulation_mae_db": 1.133236701581027, "3000–6000 Hz_40ms_log_rms_mae_db": 1.207030051881587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12233964785607952, "3000–6000 Hz_40ms_envelope_correlation": 0.9849394487078871, "3000–6000 Hz_40ms_modulation_mae_db": 1.4363648145493026, "mr_stft_db": 5.916567103535285, "band_mae_db": 0.19080318335755606, "rms_gain_error_db": -0.07901854926974754, "output_peak": 0.3734099383643709}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8991835979911136, "全帯域_10ms_rms_envelope_nrmse": 0.06550951176977964, "全帯域_10ms_envelope_correlation": 0.9947468117942113, "全帯域_10ms_modulation_mae_db": 0.2271090795226588, "全帯域_40ms_log_rms_mae_db": 0.530141763791692, "全帯域_40ms_rms_envelope_nrmse": 0.035869743634039775, "全帯域_40ms_envelope_correlation": 0.9983113667745014, "全帯域_40ms_modulation_mae_db": 0.32399605551309074, "300–1000 Hz_10ms_log_rms_mae_db": 1.6803867184068355, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08868024062752856, "300–1000 Hz_10ms_envelope_correlation": 0.9928241367203392, "300–1000 Hz_10ms_modulation_mae_db": 0.23789190984017772, "300–1000 Hz_40ms_log_rms_mae_db": 1.1311016560992537, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04994610885737856, "300–1000 Hz_40ms_envelope_correlation": 0.9976092278427626, "300–1000 Hz_40ms_modulation_mae_db": 0.23266325806186502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8590992615607362, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15967273380109603, "1000–3000 Hz_10ms_envelope_correlation": 0.9801585999534987, "1000–3000 Hz_10ms_modulation_mae_db": 0.8348475875760579, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3355884839920864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0969745489186227, "1000–3000 Hz_40ms_envelope_correlation": 0.9933864555585006, "1000–3000 Hz_40ms_modulation_mae_db": 0.7365448427314717, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8497408937498634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1905635942271505, "3000–6000 Hz_10ms_envelope_correlation": 0.9666517225616469, "3000–6000 Hz_10ms_modulation_mae_db": 1.3863169828516213, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2245623556991976, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12045326671075032, "3000–6000 Hz_40ms_envelope_correlation": 0.98686312278788, "3000–6000 Hz_40ms_modulation_mae_db": 1.7518791202803141, "mr_stft_db": 5.917676640535294, "band_mae_db": 0.2780092795822121, "rms_gain_error_db": -0.07177442869359668, "output_peak": 0.3863323944461895}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8901262329013759, "全帯域_10ms_rms_envelope_nrmse": 0.05500877357234794, "全帯域_10ms_envelope_correlation": 0.9963366519003685, "全帯域_10ms_modulation_mae_db": 0.1841326979493434, "全帯域_40ms_log_rms_mae_db": 0.4454083745432449, "全帯域_40ms_rms_envelope_nrmse": 0.028108203230538156, "全帯域_40ms_envelope_correlation": 0.9990179232470234, "全帯域_40ms_modulation_mae_db": 0.18061366086308797, "300–1000 Hz_10ms_log_rms_mae_db": 1.6043680674326481, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07708671666563394, "300–1000 Hz_10ms_envelope_correlation": 0.994717008628688, "300–1000 Hz_10ms_modulation_mae_db": 0.31968170091922987, "300–1000 Hz_40ms_log_rms_mae_db": 1.0049671207286224, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042614821988678955, "300–1000 Hz_40ms_envelope_correlation": 0.9984386563290859, "300–1000 Hz_40ms_modulation_mae_db": 0.3151619291894186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8066235909690898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15537762403486882, "1000–3000 Hz_10ms_envelope_correlation": 0.9804500031364654, "1000–3000 Hz_10ms_modulation_mae_db": 1.127726121566137, "1000–3000 Hz_40ms_log_rms_mae_db": 1.246854351628528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09039303397705215, "1000–3000 Hz_40ms_envelope_correlation": 0.9930512441039667, "1000–3000 Hz_40ms_modulation_mae_db": 1.011693365783798, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9835596819609331, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18996946348803453, "3000–6000 Hz_10ms_envelope_correlation": 0.9649972789716864, "3000–6000 Hz_10ms_modulation_mae_db": 0.568978249713591, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2085874299536532, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1135622504653389, "3000–6000 Hz_40ms_envelope_correlation": 0.9858385348335069, "3000–6000 Hz_40ms_modulation_mae_db": 0.5600643913951829, "mr_stft_db": 5.915843905827174, "band_mae_db": 0.15216023482955798, "rms_gain_error_db": -0.07992702877687254, "output_peak": 0.37881096600018493}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 52320, "output_samples": 52320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.943543795715719, "spectral_convergence": 0.21421544444141433, "spectral_convergence_status": "ok", "amplitude_floor": 1.2746778473169441e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.883386812479317, "spectral_convergence": 0.20592156695633115, "spectral_convergence_status": "ok", "amplitude_floor": 1.090624511245042e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.953191102682514, "spectral_convergence": 0.20911094296704438, "spectral_convergence_status": "ok", "amplitude_floor": 7.935262595002381e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004730173956506593, "output_power": 0.0004813849033427809, "delta_db": 0.07615353681382861, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018195471619226037, "output_power": 0.0017915112344813206, "delta_db": -0.0674378085392095, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018173989836957952, "output_power": 0.0017640947217133695, "delta_db": -0.1292838032452013, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013671881984837986, "output_power": 0.0001391343768612491, "delta_db": 0.0760614637779432, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012324536624073353, "output_power": 0.0001144692683366614, "delta_db": -0.32081692813131063, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.679288082514536e-05, "output_power": 8.095917404699913e-05, "delta_db": -0.3021803488278654, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1619889815558931, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08379593512035055, "pitch": null, "output_rms_dbfs": -23.593641844937395, "output_peak": 0.37227892787430544, "reference_rms_dbfs": -23.509845909817045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.926707236959184}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 52320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00017439460044090568, "log_rms_mae_db": 0.835376285526767, "rms_envelope_nrmse": 0.059570135986125095, "envelope_correlation": 0.9956968592178179, "rms_gain_error_db": -0.0837938808067171, "reference_mean_power": 0.004463526965610992, "output_mean_power": 0.004378232039253812}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 1.3056377024951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008270090313661458, "output_power": 0.0008063708392134486, "reference_active": true, "delta_db": -0.10975438303488773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010887958720618104, "output_power": 9.838950321252445e-05, "reference_active": true, "delta_db": -0.4399769792332143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8937894537156724e-05, "output_power": 2.7066837549065294e-05, "reference_active": true, "delta_db": -0.29029413158970835}], "active_band_count": 3, "active_band_mae_db": 0.28000849795260346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015806536675822195, "log_rms_mae_db": 0.4103824596163088, "rms_envelope_nrmse": 0.03157870784758466, "envelope_correlation": 0.9987411432860397, "rms_gain_error_db": -0.08380909395188835, "reference_mean_power": 0.004504823508022685, "output_mean_power": 0.004418723954556981}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.0922276511075418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000711888936919089, "output_power": 0.0006899131003908856, "reference_active": true, "delta_db": -0.13617852243268894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.550528203435593e-05, "output_power": 6.22607023802293e-05, "reference_active": true, "delta_db": -0.22062305074503544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5248696894099073e-06, "output_power": 1.388015288708969e-06, "reference_active": true, "delta_db": -0.4083848201017013}], "active_band_count": 3, "active_band_mae_db": 0.25506213109314185}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00016816220894332027, "log_rms_mae_db": 1.623219869826778, "rms_envelope_nrmse": 0.08217438818182304, "envelope_correlation": 0.993891445992195, "rms_gain_error_db": -0.12100730619652394, "reference_mean_power": 0.002573247092376177, "output_mean_power": 0.0025025384570520398}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 9.221294414644421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007324784888809201, "output_power": 0.0006840050312825507, "reference_active": true, "delta_db": -0.29735578798424983}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.576574656453582e-05, "output_power": 7.422679495490819e-05, "reference_active": true, "delta_db": -0.6275316384884448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.804896765293062e-05, "output_power": 1.8533879250414064e-05, "reference_active": true, "delta_db": 0.11513962629680462}], "active_band_count": 3, "active_band_mae_db": 0.34667568425649975}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.0001473388036867024, "log_rms_mae_db": 1.12663025487867, "rms_envelope_nrmse": 0.04639147839452062, "envelope_correlation": 0.9979994441981263, "rms_gain_error_db": -0.12101139978809913, "reference_mean_power": 0.002597106693874505, "output_mean_power": 0.0025257400548718543}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.911179205548874e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006603877113347142, "output_power": 0.0006171922009214396, "reference_active": true, "delta_db": -0.2937855368863052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.506710493994746e-05, "output_power": 4.200687330613227e-05, "reference_active": true, "delta_db": -0.3053930344034073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1610111255482175e-06, "output_power": 1.052038714318594e-06, "reference_active": true, "delta_db": -0.42804659587666805}], "active_band_count": 3, "active_band_mae_db": 0.34240838905546017}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 5.380259383983556e-05, "log_rms_mae_db": 1.859868284155472, "rms_envelope_nrmse": 0.1422007951195526, "envelope_correlation": 0.9836650556301658, "rms_gain_error_db": -0.13027730753658578, "reference_mean_power": 0.00019859577428883629, "output_mean_power": 0.0001927268719300862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 8.760825576371796e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.452705657171872e-05, "output_power": 5.020139474453369e-05, "reference_active": true, "delta_db": -0.35896271334154234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.928893294666125e-06, "output_power": 3.982843127384624e-06, "reference_active": true, "delta_db": -1.7278043304599013}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9788817769833428e-06, "output_power": 1.6466488133914535e-06, "reference_active": true, "delta_db": -0.7981886368797471}], "active_band_count": 3, "active_band_mae_db": 0.961651893560397}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 5.051409218298903e-05, "log_rms_mae_db": 1.2203510083591598, "rms_envelope_nrmse": 0.0824392686674026, "envelope_correlation": 0.9942741250352013, "rms_gain_error_db": -0.13021021976155017, "reference_mean_power": 0.00020041602526910178, "output_mean_power": 0.00019449633531802145}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.85711379260727e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.022793901144941e-05, "output_power": 4.558116778198766e-05, "reference_active": true, "delta_db": -0.42159910705030834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.619667417561746e-06, "output_power": 1.7010491666068345e-06, "reference_active": true, "delta_db": -1.8752929202946214}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1855266901360492e-07, "output_power": 1.1590420948289624e-07, "reference_active": true, "delta_db": -0.0981212659921514}], "active_band_count": 3, "active_band_mae_db": 0.7983377644456936}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 6.112392927022402e-05, "log_rms_mae_db": 1.9810063995510632, "rms_envelope_nrmse": 0.20660399738032972, "envelope_correlation": 0.958680610112849, "rms_gain_error_db": -0.30601072097533855, "reference_mean_power": 8.711996692518435e-05, "output_mean_power": 8.119263381889519e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 4.203401537603793e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8248175978150172e-05, "output_power": 1.583645483949809e-05, "reference_active": true, "delta_db": -0.6156149380229127}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.243431842943793e-05, "output_power": 9.83985432294468e-06, "reference_active": true, "delta_db": -1.0163331614934321}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.767669752037568e-06, "output_power": 4.061571400779461e-06, "reference_active": true, "delta_db": -2.2174506553786215}], "active_band_count": 3, "active_band_mae_db": 1.2831329182983222}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 3.626522691368819e-05, "log_rms_mae_db": 1.2236589597771594, "rms_envelope_nrmse": 0.11825823657282009, "envelope_correlation": 0.9853397336171076, "rms_gain_error_db": -0.3066306488821139, "reference_mean_power": 8.789733624317315e-05, "output_mean_power": 8.190542137425512e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 3.223393789064049e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.051966991658498e-05, "output_power": 1.6551620989551886e-05, "reference_active": true, "delta_db": -0.9332983735360951}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.880534452438109e-06, "output_power": 4.6379471834484225e-06, "reference_active": true, "delta_db": -1.0309100010746857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.871896444237529e-07, "output_power": 3.0006889505891694e-07, "reference_active": true, "delta_db": -2.915574087499668}], "active_band_count": 3, "active_band_mae_db": 1.626594154036816}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8766902971073217, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "197380598912476b079bbfb2cc0583fb286f5efac961eb6489bd70b4a38ab9f9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9817923530108368, "standard_deviation": 0.003027823336523232, "minimum": 0.977069230630835, "maximum": 0.986756367578333, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.848078518676214, "standard_deviation": 0.03352369077682241, "minimum": 1.8066235909690898, "maximum": 1.8946853119505236, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7299902921237751, "standard_deviation": 0.274595737694382, "minimum": 0.37899715045110055, "maximum": 1.127726121566137, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15014716967297856, "standard_deviation": 0.01243772630280504, "minimum": 0.12943820566615652, "maximum": 0.1679924920769119, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9932790477125899, "standard_deviation": 0.0015335025423597552, "minimum": 0.9910459249332451, "maximum": 0.9955183182076233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.244561239251344, "standard_deviation": 0.051974402806806284, "minimum": 1.1900138096423656, "maximum": 1.3355884839920864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6757644105608457, "standard_deviation": 0.19611815247307038, "minimum": 0.3282788352136614, "maximum": 1.011693365783798, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08998372391802399, "standard_deviation": 0.009312508657930934, "minimum": 0.07609732744356881, "maximum": 0.10185840693290381, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9604984480138852, "standard_deviation": 0.005279135434263727, "minimum": 0.9497172868704985, "maximum": 0.9666517225616469, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.942755980865834, "standard_deviation": 0.04987231439170974, "minimum": 1.8497408937498634, "maximum": 1.9835596819609331, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1282106871928494, "standard_deviation": 0.4189848479391677, "minimum": 0.568978249713591, "maximum": 1.8764307870649601, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20306604108368156, "standard_deviation": 0.011882311961204383, "minimum": 0.18996946348803453, "maximum": 0.2265557652935587, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9843180292851328, "standard_deviation": 0.0021628805920673464, "minimum": 0.9797268423375964, "maximum": 0.98686312278788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2304187115235137, "standard_deviation": 0.04457008312535307, "minimum": 1.1652646031485394, "maximum": 1.3049824488823838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.380713167545141, "standard_deviation": 0.5856250503697553, "minimum": 0.5600643913951829, "maximum": 2.3889265789791474, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12332771945646846, "standard_deviation": 0.006324269570383499, "minimum": 0.1135622504653389, "maximum": 0.13229317497738613, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9939363264075485, "standard_deviation": 0.0008526802217471894, "minimum": 0.9928241367203392, "maximum": 0.9955697063482626, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6185755473982502, "standard_deviation": 0.06713797509308025, "minimum": 1.5291074178488089, "maximum": 1.7447423997125062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4486458008655606, "standard_deviation": 0.25861914576473255, "minimum": 0.23789190984017772, "maximum": 0.991658068306375, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08197109647295424, "standard_deviation": 0.006123803759900803, "minimum": 0.06955699945197945, "maximum": 0.08868024062752856, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980535249155003, "standard_deviation": 0.00034732035819084214, "minimum": 0.9975830515709826, "maximum": 0.9984450583199146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0773655658707364, "standard_deviation": 0.04766332151238853, "minimum": 1.0049671207286224, "maximum": 1.1311016560992537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4870361830768026, "standard_deviation": 0.2808200862350269, "minimum": 0.22101568903091517, "maximum": 0.9313563244705568, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.046154736318806525, "standard_deviation": 0.004487044929479513, "minimum": 0.03974820210381392, "maximum": 0.053297718828412735, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17571937506394214, "standard_deviation": 0.04499737945668496, "minimum": 0.13173606221235923, "maximum": 0.2780092795822121, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.9160708073646155, "standard_deviation": 0.0077940499889142265, "minimum": 5.904433048879704, "maximum": 5.926707236959184, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3735989229761085, "standard_deviation": 0.011324915947067407, "minimum": 0.35970288898243424, "maximum": 0.3907119728673631, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08080472828889894, "standard_deviation": 0.036648930857692004, "minimum": -0.16610964245435317, "maximum": -0.048218522070211225, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955474213092621, "standard_deviation": 0.000679943060164335, "minimum": 0.9947468117942113, "maximum": 0.9964758741416104, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8980381391329197, "standard_deviation": 0.04138738485554006, "minimum": 0.835376285526767, "maximum": 0.9612491333671107, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2858683816030805, "standard_deviation": 0.1000469539249208, "minimum": 0.16255019133309115, "maximum": 0.4631592136567442, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06046627315393101, "standard_deviation": 0.00488361724246689, "minimum": 0.05368752226253446, "maximum": 0.06613889520577068, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.99868824536259, "standard_deviation": 0.0002929829229338194, "minimum": 0.9982729685952144, "maximum": 0.9990179232470234, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4813900104445498, "standard_deviation": 0.044680671185045345, "minimum": 0.4103824596163088, "maximum": 0.5385598338734353, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35404483970464296, "standard_deviation": 0.12779090034111384, "minimum": 0.18061366086308797, "maximum": 0.5011827225827077, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.032120743025905285, "standard_deviation": 0.0039422727923817185, "minimum": 0.027266285764288713, "maximum": 0.03776104824761334, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0027", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.4564902782440186, "predicted_mos": 2.4564902782440186, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "3946d8a9626ed98dd4c289df0081f150c5ec4ec978ead69eadc6f1c6d3252f8a", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.672906399704516, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9120359963422582, "全帯域_10ms_rms_envelope_nrmse": 0.06713237512281262, "全帯域_10ms_envelope_correlation": 0.9944890155605223, "全帯域_10ms_modulation_mae_db": 0.3688839671383605, "全帯域_40ms_log_rms_mae_db": 0.46529720490495285, "全帯域_40ms_rms_envelope_nrmse": 0.033266527319645946, "全帯域_40ms_envelope_correlation": 0.9985783985713316, "全帯域_40ms_modulation_mae_db": 0.4495210726525533, "300–1000 Hz_10ms_log_rms_mae_db": 1.273625265648671, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08697986597148735, "300–1000 Hz_10ms_envelope_correlation": 0.9931449257031938, "300–1000 Hz_10ms_modulation_mae_db": 0.2834538756027156, "300–1000 Hz_40ms_log_rms_mae_db": 0.7423485502897516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04418638356222429, "300–1000 Hz_40ms_envelope_correlation": 0.9982346466343894, "300–1000 Hz_40ms_modulation_mae_db": 0.5567027619935164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3782377070061347, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15954765934176762, "1000–3000 Hz_10ms_envelope_correlation": 0.9793686139603869, "1000–3000 Hz_10ms_modulation_mae_db": 0.9308270228338565, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9292625439862514, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08473120136521572, "1000–3000 Hz_40ms_envelope_correlation": 0.9939848963166389, "1000–3000 Hz_40ms_modulation_mae_db": 0.8635052253731071, "3000–6000 Hz_10ms_log_rms_mae_db": 1.405278259154483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17785889255521364, "3000–6000 Hz_10ms_envelope_correlation": 0.9717741605212215, "3000–6000 Hz_10ms_modulation_mae_db": 1.188760916910472, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9095799394848095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10663113915890687, "3000–6000 Hz_40ms_envelope_correlation": 0.9915649870836722, "3000–6000 Hz_40ms_modulation_mae_db": 1.636843835277953, "mr_stft_db": 5.466734426260527, "band_mae_db": 0.2527175464553064, "rms_gain_error_db": -0.06433749275619532, "output_peak": 0.37144090166037014}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0521785994234991, "全帯域_10ms_rms_envelope_nrmse": 0.06963840256761804, "全帯域_10ms_envelope_correlation": 0.9940596102138513, "全帯域_10ms_modulation_mae_db": 0.05749135595075968, "全帯域_40ms_log_rms_mae_db": 0.4714399895322557, "全帯域_40ms_rms_envelope_nrmse": 0.03251787362307008, "全帯域_40ms_envelope_correlation": 0.9986552618306215, "全帯域_40ms_modulation_mae_db": 0.08676108821144614, "300–1000 Hz_10ms_log_rms_mae_db": 1.2940209291719513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08195303709073781, "300–1000 Hz_10ms_envelope_correlation": 0.9939138051702697, "300–1000 Hz_10ms_modulation_mae_db": 0.24608913349579772, "300–1000 Hz_40ms_log_rms_mae_db": 0.7119502900393512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04037650520539047, "300–1000 Hz_40ms_envelope_correlation": 0.9985296967738413, "300–1000 Hz_40ms_modulation_mae_db": 0.23548249558301046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3554858837823862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13732731454472588, "1000–3000 Hz_10ms_envelope_correlation": 0.9847960933691307, "1000–3000 Hz_10ms_modulation_mae_db": 0.1572045762520768, "1000–3000 Hz_40ms_log_rms_mae_db": 0.885720230526199, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07135071337837377, "1000–3000 Hz_40ms_envelope_correlation": 0.9956837815686268, "1000–3000 Hz_40ms_modulation_mae_db": 0.3570513483979254, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3555419277898766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19654359562441626, "3000–6000 Hz_10ms_envelope_correlation": 0.9624152383448468, "3000–6000 Hz_10ms_modulation_mae_db": 0.9121032039407382, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8680972158654162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10099220292718118, "3000–6000 Hz_40ms_envelope_correlation": 0.9890002114785975, "3000–6000 Hz_40ms_modulation_mae_db": 1.2125204991427285, "mr_stft_db": 5.518723774875174, "band_mae_db": 0.22220359090939146, "rms_gain_error_db": -0.0782282913391868, "output_peak": 0.3878805861474167}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.916606406423613, "全帯域_10ms_rms_envelope_nrmse": 0.06276975135314669, "全帯域_10ms_envelope_correlation": 0.9952153361367694, "全帯域_10ms_modulation_mae_db": 0.5568669827575196, "全帯域_40ms_log_rms_mae_db": 0.4988922015524287, "全帯域_40ms_rms_envelope_nrmse": 0.033355935076849134, "全帯域_40ms_envelope_correlation": 0.9985923731131054, "全帯域_40ms_modulation_mae_db": 0.6398426279840382, "300–1000 Hz_10ms_log_rms_mae_db": 1.1813451535632737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08285267155838599, "300–1000 Hz_10ms_envelope_correlation": 0.993785987418353, "300–1000 Hz_10ms_modulation_mae_db": 1.067037874264629, "300–1000 Hz_40ms_log_rms_mae_db": 0.7264363593818935, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04726081380656373, "300–1000 Hz_40ms_envelope_correlation": 0.9979097444452529, "300–1000 Hz_40ms_modulation_mae_db": 1.380526937504026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.308950561488519, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14382168277595897, "1000–3000 Hz_10ms_envelope_correlation": 0.9843150254611442, "1000–3000 Hz_10ms_modulation_mae_db": 0.720771348436794, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8935428649309038, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09395741951481001, "1000–3000 Hz_40ms_envelope_correlation": 0.9939369221345252, "1000–3000 Hz_40ms_modulation_mae_db": 1.0014610308343037, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3634866556922705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16921244157822016, "3000–6000 Hz_10ms_envelope_correlation": 0.9723264668600354, "3000–6000 Hz_10ms_modulation_mae_db": 0.29037851386507835, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8809558590875511, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10193043539272391, "3000–6000 Hz_40ms_envelope_correlation": 0.9885129337850344, "3000–6000 Hz_40ms_modulation_mae_db": 0.40114755882933223, "mr_stft_db": 5.473168567047036, "band_mae_db": 0.215996096961982, "rms_gain_error_db": -0.08594687980219536, "output_peak": 0.35756807258597584}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9439325986148717, "全帯域_10ms_rms_envelope_nrmse": 0.06681706359096576, "全帯域_10ms_envelope_correlation": 0.9945412077202521, "全帯域_10ms_modulation_mae_db": 0.6000797356534416, "全帯域_40ms_log_rms_mae_db": 0.5067299471153113, "全帯域_40ms_rms_envelope_nrmse": 0.036810263005197665, "全帯域_40ms_envelope_correlation": 0.9982284137954573, "全帯域_40ms_modulation_mae_db": 0.7105988079039682, "300–1000 Hz_10ms_log_rms_mae_db": 1.2965209151663675, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08388326192372532, "300–1000 Hz_10ms_envelope_correlation": 0.9935649230810308, "300–1000 Hz_10ms_modulation_mae_db": 0.7708150339740678, "300–1000 Hz_40ms_log_rms_mae_db": 0.8237150536250534, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04637404932222149, "300–1000 Hz_40ms_envelope_correlation": 0.9979137016588093, "300–1000 Hz_40ms_modulation_mae_db": 0.678870981648683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3203916950402326, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1356966768356623, "1000–3000 Hz_10ms_envelope_correlation": 0.9854887725884656, "1000–3000 Hz_10ms_modulation_mae_db": 0.6963232131462883, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9486305361876216, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07887010688329255, "1000–3000 Hz_40ms_envelope_correlation": 0.9953091336243679, "1000–3000 Hz_40ms_modulation_mae_db": 0.7385507801277337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4301118063331397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20356706244828918, "3000–6000 Hz_10ms_envelope_correlation": 0.9615027302396107, "3000–6000 Hz_10ms_modulation_mae_db": 1.5439466629419296, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9420822873450572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12480796452072805, "3000–6000 Hz_40ms_envelope_correlation": 0.9862531482056166, "3000–6000 Hz_40ms_modulation_mae_db": 1.9392896256916903, "mr_stft_db": 5.492012838820478, "band_mae_db": 0.24444247707576758, "rms_gain_error_db": -0.07973031994698662, "output_peak": 0.39431228981630706}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9489625096852974, "全帯域_10ms_rms_envelope_nrmse": 0.0634107968254106, "全帯域_10ms_envelope_correlation": 0.9951875673757118, "全帯域_10ms_modulation_mae_db": 0.3057630260354737, "全帯域_40ms_log_rms_mae_db": 0.5381532768897258, "全帯域_40ms_rms_envelope_nrmse": 0.03521842622317051, "全帯域_40ms_envelope_correlation": 0.9985075193913588, "全帯域_40ms_modulation_mae_db": 0.39745719681017744, "300–1000 Hz_10ms_log_rms_mae_db": 1.2795965271690506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07967546552731834, "300–1000 Hz_10ms_envelope_correlation": 0.9942441135356823, "300–1000 Hz_10ms_modulation_mae_db": 0.4754399024947736, "300–1000 Hz_40ms_log_rms_mae_db": 0.7329161023176258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045730620973206494, "300–1000 Hz_40ms_envelope_correlation": 0.9980262060340744, "300–1000 Hz_40ms_modulation_mae_db": 0.5297993342716706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3483789917465134, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15039260240383753, "1000–3000 Hz_10ms_envelope_correlation": 0.98286753051906, "1000–3000 Hz_10ms_modulation_mae_db": 0.7760496310998324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9232672367867952, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09706533258112325, "1000–3000 Hz_40ms_envelope_correlation": 0.9936583399836199, "1000–3000 Hz_40ms_modulation_mae_db": 0.8153526927960374, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4031408606720077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21264209978532458, "3000–6000 Hz_10ms_envelope_correlation": 0.9599394228898857, "3000–6000 Hz_10ms_modulation_mae_db": 1.9608402363331254, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9622279721999338, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1479448902427345, "3000–6000 Hz_40ms_envelope_correlation": 0.9819070840644496, "3000–6000 Hz_40ms_modulation_mae_db": 2.7159215108084083, "mr_stft_db": 5.502520017215836, "band_mae_db": 0.29804323381152825, "rms_gain_error_db": -0.11430345668482932, "output_peak": 0.41113061187522315}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8759965071069786, "全帯域_10ms_rms_envelope_nrmse": 0.059822598957898564, "全帯域_10ms_envelope_correlation": 0.9956808255717882, "全帯域_10ms_modulation_mae_db": 0.3979869292618113, "全帯域_40ms_log_rms_mae_db": 0.4879196044234745, "全帯域_40ms_rms_envelope_nrmse": 0.02974499195635705, "全帯域_40ms_envelope_correlation": 0.9989193796992635, "全帯域_40ms_modulation_mae_db": 0.3371095098472951, "300–1000 Hz_10ms_log_rms_mae_db": 1.377596655998641, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08261450562859429, "300–1000 Hz_10ms_envelope_correlation": 0.9937793076948577, "300–1000 Hz_10ms_modulation_mae_db": 0.44142559662370456, "300–1000 Hz_40ms_log_rms_mae_db": 0.805251704230002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04377104061956287, "300–1000 Hz_40ms_envelope_correlation": 0.9981737085955901, "300–1000 Hz_40ms_modulation_mae_db": 0.38541395624423513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3137951838182296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13533238445797985, "1000–3000 Hz_10ms_envelope_correlation": 0.9860530523520169, "1000–3000 Hz_10ms_modulation_mae_db": 0.6821680423252966, "1000–3000 Hz_40ms_log_rms_mae_db": 0.933016215128201, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08489228963809861, "1000–3000 Hz_40ms_envelope_correlation": 0.9950635315624969, "1000–3000 Hz_40ms_modulation_mae_db": 0.7900209830258881, "3000–6000 Hz_10ms_log_rms_mae_db": 1.33965163100589, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17509204314715118, "3000–6000 Hz_10ms_envelope_correlation": 0.9736566052197592, "3000–6000 Hz_10ms_modulation_mae_db": 1.1940425400663688, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9152132054691264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11473682108462284, "3000–6000 Hz_40ms_envelope_correlation": 0.990425364617156, "3000–6000 Hz_40ms_modulation_mae_db": 1.360936667147609, "mr_stft_db": 5.492420407360551, "band_mae_db": 0.25294196484374964, "rms_gain_error_db": -0.08673451719330734, "output_peak": 0.39639414030161524}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8792500915946394, "全帯域_10ms_rms_envelope_nrmse": 0.063099530806297, "全帯域_10ms_envelope_correlation": 0.9952069508807281, "全帯域_10ms_modulation_mae_db": 0.20589522191853304, "全帯域_40ms_log_rms_mae_db": 0.4425685513394286, "全帯域_40ms_rms_envelope_nrmse": 0.030801060735474704, "全帯域_40ms_envelope_correlation": 0.9988824295990827, "全帯域_40ms_modulation_mae_db": 0.13226326422298226, "300–1000 Hz_10ms_log_rms_mae_db": 1.2168786689922846, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07664834142854399, "300–1000 Hz_10ms_envelope_correlation": 0.9947420913977234, "300–1000 Hz_10ms_modulation_mae_db": 0.33483274854477446, "300–1000 Hz_40ms_log_rms_mae_db": 0.7106892875151586, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037346408340206264, "300–1000 Hz_40ms_envelope_correlation": 0.9988202116428135, "300–1000 Hz_40ms_modulation_mae_db": 0.21391341508486109, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3787190517727619, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16213280599685956, "1000–3000 Hz_10ms_envelope_correlation": 0.9788276526859584, "1000–3000 Hz_10ms_modulation_mae_db": 0.3322519999250004, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9740173224986965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10712970399219969, "1000–3000 Hz_40ms_envelope_correlation": 0.9904220851145962, "1000–3000 Hz_40ms_modulation_mae_db": 0.386086061893534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4018508112435122, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2103850658454383, "3000–6000 Hz_10ms_envelope_correlation": 0.9573607789859147, "3000–6000 Hz_10ms_modulation_mae_db": 0.20039846316388876, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9593227204640037, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12282916014614928, "3000–6000 Hz_40ms_envelope_correlation": 0.9834309685007401, "3000–6000 Hz_40ms_modulation_mae_db": 0.31836915337931687, "mr_stft_db": 5.4885487502955534, "band_mae_db": 0.23062864526165086, "rms_gain_error_db": -0.1008559484012315, "output_peak": 0.35986136702500654}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9864912961108633, "全帯域_10ms_rms_envelope_nrmse": 0.0671378609752966, "全帯域_10ms_envelope_correlation": 0.9945102346246414, "全帯域_10ms_modulation_mae_db": 0.28286080844179906, "全帯域_40ms_log_rms_mae_db": 0.5578762477308801, "全帯域_40ms_rms_envelope_nrmse": 0.03715639219511744, "全帯域_40ms_envelope_correlation": 0.9982259594213663, "全帯域_40ms_modulation_mae_db": 0.34501883559624286, "300–1000 Hz_10ms_log_rms_mae_db": 1.3703961489282541, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0867593529474286, "300–1000 Hz_10ms_envelope_correlation": 0.9931175777897268, "300–1000 Hz_10ms_modulation_mae_db": 0.33301318166784927, "300–1000 Hz_40ms_log_rms_mae_db": 0.8339056660553297, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04983721635361891, "300–1000 Hz_40ms_envelope_correlation": 0.9975878483826833, "300–1000 Hz_40ms_modulation_mae_db": 0.3913339669067007, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3490656993753531, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1380670101036851, "1000–3000 Hz_10ms_envelope_correlation": 0.9857154957153065, "1000–3000 Hz_10ms_modulation_mae_db": 0.7928445926115933, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8890673237807574, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08728773473407879, "1000–3000 Hz_40ms_envelope_correlation": 0.9950930740274305, "1000–3000 Hz_40ms_modulation_mae_db": 0.549486426548469, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4125251531556886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20563874224142592, "3000–6000 Hz_10ms_envelope_correlation": 0.9591027723527423, "3000–6000 Hz_10ms_modulation_mae_db": 0.4603458353052344, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9629834406603587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1285327892861911, "3000–6000 Hz_40ms_envelope_correlation": 0.9821056424403503, "3000–6000 Hz_40ms_modulation_mae_db": 0.71920560096272, "mr_stft_db": 5.493562794422846, "band_mae_db": 0.19724759164102149, "rms_gain_error_db": -0.0902320059009152, "output_peak": 0.3846074336434092}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 52320, "output_samples": 52320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.396910810530969, "spectral_convergence": 0.21470721938876614, "spectral_convergence_status": "ok", "amplitude_floor": 1.2746778473169441e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.328998463966321, "spectral_convergence": 0.2032481166245994, "spectral_convergence_status": "ok", "amplitude_floor": 1.090624511245042e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.674294004284293, "spectral_convergence": 0.20589373682584772, "spectral_convergence_status": "ok", "amplitude_floor": 7.935262595002381e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004730173956506593, "output_power": 0.0004961158018284441, "delta_db": 0.2070594739725703, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018195471619226037, "output_power": 0.0018369812573528878, "delta_db": 0.04141408434464493, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018173989836957952, "output_power": 0.0017296638462190018, "delta_db": -0.21488573054833932, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013671881984837986, "output_power": 0.00013958540629428176, "delta_db": 0.09011714096765668, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012324536624073353, "output_power": 0.00011299634829504134, "delta_db": -0.37706191114957077, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.679288082514536e-05, "output_power": 7.584158661437824e-05, "delta_db": -0.5857669377490566, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2527175464553064, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06433749275619532, "pitch": null, "output_rms_dbfs": -23.57418340257324, "output_peak": 0.37144090166037014, "reference_rms_dbfs": -23.509845909817045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.466734426260527}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 52320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00017439460044090568, "log_rms_mae_db": 0.9120359963422582, "rms_envelope_nrmse": 0.06713237512281262, "envelope_correlation": 0.9944890155605223, "rms_gain_error_db": -0.06433572340725938, "reference_mean_power": 0.004463526965610992, "output_mean_power": 0.004397892307845023}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 1.3056377024951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008270090313661458, "output_power": 0.0007981637409677109, "reference_active": true, "delta_db": -0.15418257559431164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010887958720618104, "output_power": 9.950447622288979e-05, "reference_active": true, "delta_db": -0.3910384767949376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8937894537156724e-05, "output_power": 2.542868164281868e-05, "reference_active": true, "delta_db": -0.5614308490258322}], "active_band_count": 3, "active_band_mae_db": 0.3688839671383605}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015806536675822195, "log_rms_mae_db": 0.46529720490495285, "rms_envelope_nrmse": 0.033266527319645946, "envelope_correlation": 0.9985783985713316, "rms_gain_error_db": -0.06432964914703902, "reference_mean_power": 0.004504823508022685, "output_mean_power": 0.004438587806563451}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.0922276511075418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000711888936919089, "output_power": 0.0006799194149065918, "reference_active": true, "delta_db": -0.19954801346538112}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.550528203435593e-05, "output_power": 6.109507838611315e-05, "reference_active": true, "delta_db": -0.30270094495110567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5248696894099073e-06, "output_power": 1.2548774231794486e-06, "reference_active": true, "delta_db": -0.8463142595411732}], "active_band_count": 3, "active_band_mae_db": 0.4495210726525533}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00016816220894332027, "log_rms_mae_db": 1.273625265648671, "rms_envelope_nrmse": 0.08697986597148735, "envelope_correlation": 0.9931449257031938, "rms_gain_error_db": -0.10701758828091877, "reference_mean_power": 0.002573247092376177, "output_mean_power": 0.002510612760742699}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 9.221294414644421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007324784888809201, "output_power": 0.0006935926122457388, "reference_active": true, "delta_db": -0.2369041640157012}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.576574656453582e-05, "output_power": 7.46092525945254e-05, "reference_active": true, "delta_db": -0.6052118294400495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.804896765293062e-05, "output_power": 1.801473190196868e-05, "reference_active": true, "delta_db": -0.008245633352396123}], "active_band_count": 3, "active_band_mae_db": 0.2834538756027156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.0001473388036867024, "log_rms_mae_db": 0.7423485502897516, "rms_envelope_nrmse": 0.04418638356222429, "envelope_correlation": 0.9982346466343894, "rms_gain_error_db": -0.10701930114441434, "reference_mean_power": 0.002597106693874505, "output_mean_power": 0.0025338906062861663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.911179205548874e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006603877113347142, "output_power": 0.0006239319018471253, "reference_active": true, "delta_db": -0.24661791381535159}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.506710493994746e-05, "output_power": 3.82535974935935e-05, "reference_active": true, "delta_db": -0.7118737649306559}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1610111255482175e-06, "output_power": 9.855410554188e-07, "reference_active": true, "delta_db": -0.711616607234542}], "active_band_count": 3, "active_band_mae_db": 0.5567027619935164}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 5.380259383983556e-05, "log_rms_mae_db": 1.3782377070061347, "rms_envelope_nrmse": 0.15954765934176762, "envelope_correlation": 0.9793686139603869, "rms_gain_error_db": -0.16181298567799712, "reference_mean_power": 0.00019859577428883629, "output_mean_power": 0.0001913324817808648}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 8.760825576371796e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.452705657171872e-05, "output_power": 4.927740402503441e-05, "reference_active": true, "delta_db": -0.43964233881812365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.928893294666125e-06, "output_power": 4.684476978249317e-06, "reference_active": true, "delta_db": -1.0231252523351178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9788817769833428e-06, "output_power": 2.687764397715906e-06, "reference_active": true, "delta_db": 1.3297134773483283}], "active_band_count": 3, "active_band_mae_db": 0.9308270228338565}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 5.051409218298903e-05, "log_rms_mae_db": 0.9292625439862514, "rms_envelope_nrmse": 0.08473120136521572, "envelope_correlation": 0.9939848963166389, "rms_gain_error_db": -0.16179953225080204, "reference_mean_power": 0.00020041602526910178, "output_mean_power": 0.00019308675840726424}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.85711379260727e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.022793901144941e-05, "output_power": 4.513658594525867e-05, "reference_active": true, "delta_db": -0.4641665155817364}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.619667417561746e-06, "output_power": 2.1332906284466784e-06, "reference_active": true, "delta_db": -0.8919613303191201}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1855266901360492e-07, "output_power": 1.5752522871628217e-07, "reference_active": true, "delta_db": 1.234387830218465}], "active_band_count": 3, "active_band_mae_db": 0.8635052253731071}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 6.112392927022402e-05, "log_rms_mae_db": 1.405278259154483, "rms_envelope_nrmse": 0.17785889255521364, "envelope_correlation": 0.9717741605212215, "rms_gain_error_db": -0.5788480063692331, "reference_mean_power": 8.711996692518435e-05, "output_mean_power": 7.624878034776742e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 4.203401537603793e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8248175978150172e-05, "output_power": 1.5744831377818355e-05, "reference_active": true, "delta_db": -0.6408144661156451}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.243431842943793e-05, "output_power": 1.0029212552541413e-05, "reference_active": true, "delta_db": -0.9335514942454073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.767669752037568e-06, "output_power": 4.278065985028189e-06, "reference_active": true, "delta_db": -1.991916790370364}], "active_band_count": 3, "active_band_mae_db": 1.188760916910472}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 3.626522691368819e-05, "log_rms_mae_db": 0.9095799394848095, "rms_envelope_nrmse": 0.10663113915890687, "envelope_correlation": 0.9915649870836722, "rms_gain_error_db": -0.5789237356908568, "reference_mean_power": 8.789733624317315e-05, "output_mean_power": 7.692780490932207e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 3.223393789064049e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.051966991658498e-05, "output_power": 1.6637778825256386e-05, "reference_active": true, "delta_db": -0.9107502365523447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.880534452438109e-06, "output_power": 4.670075534168392e-06, "reference_active": true, "delta_db": -1.0009289379267523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.871896444237529e-07, "output_power": 2.943697335103289e-07, "reference_active": true, "delta_db": -2.998852331354762}], "active_band_count": 3, "active_band_mae_db": 1.636843835277953}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8926756596941946, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "04784d1088e9551fe11f512689609e2d0fdd08fd8196af874f92527dc7d92a5c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9834290295814336, "standard_deviation": 0.002853187234913188, "minimum": 0.9788276526859584, "maximum": 0.9860530523520169, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3441280967537663, "standard_deviation": 0.02739241395394251, "minimum": 1.308950561488519, "maximum": 1.3787190517727619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6360550533288423, "standard_deviation": 0.2579040628917687, "minimum": 0.1572045762520768, "maximum": 0.9308270228338565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1452897670575596, "standard_deviation": 0.0108358885811502, "minimum": 0.13533238445797985, "maximum": 0.16213280599685956, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941439705415378, "standard_deviation": 0.0016746622145066387, "minimum": 0.9904220851145962, "maximum": 0.9956837815686268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9220655342281783, "standard_deviation": 0.031170617442081956, "minimum": 0.885720230526199, "maximum": 0.9740173224986965, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6876893186246249, "standard_deviation": 0.23241174870832318, "minimum": 0.3570513483979254, "maximum": 1.0014610308343037, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08816056276089905, "standard_deviation": 0.011115075715045332, "minimum": 0.07135071337837377, "maximum": 0.10712970399219969, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.964759771926752, "standard_deviation": 0.0066729499370405156, "minimum": 0.9573607789859147, "maximum": 0.9736566052197592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3889483881308586, "standard_deviation": 0.03180142855756946, "minimum": 1.33965163100589, "maximum": 1.4301118063331397, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9688520465658544, "standard_deviation": 0.6240983551710484, "minimum": 0.20039846316388876, "maximum": 1.9608402363331254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1938674929031849, "standard_deviation": 0.01724523089386306, "minimum": 0.16921244157822016, "maximum": 0.21264209978532458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9866500425219521, "standard_deviation": 0.0037998230153108875, "minimum": 0.9819070840644496, "maximum": 0.9915649870836722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.925057830072032, "standard_deviation": 0.03743539846843547, "minimum": 0.8680972158654162, "maximum": 0.9629834406603587, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.28802930640497, "standard_deviation": 0.8146923384297252, "minimum": 0.31836915337931687, "maximum": 2.7159215108084083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11855067534490471, "standard_deviation": 0.01586674296013514, "minimum": 0.10099220292718118, "maximum": 0.1479448902427345, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937865914738546, "standard_deviation": 0.0005407659160345243, "minimum": 0.9931175777897268, "maximum": 0.9947420913977234, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2862475330798118, "standard_deviation": 0.06720089914503011, "minimum": 1.1813451535632737, "maximum": 1.377596655998641, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.49401341833353896, "standard_deviation": 0.28413979202579986, "minimum": 0.24608913349579772, "maximum": 1.067037874264629, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0826708127595277, "standard_deviation": 0.0034340281309229665, "minimum": 0.07664834142854399, "maximum": 0.08697986597148735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981494705209318, "standard_deviation": 0.00038655570992873566, "minimum": 0.9975878483826833, "maximum": 0.9988202116428135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7609016266817707, "standard_deviation": 0.051373727275174405, "minimum": 0.7106892875151586, "maximum": 0.8339056660553297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.546505481154588, "standard_deviation": 0.372420068787688, "minimum": 0.21391341508486109, "maximum": 1.380526937504026, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04436037977287431, "standard_deviation": 0.003960254855862839, "minimum": 0.037346408340206264, "maximum": 0.04983721635361891, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2392776433700497, "standard_deviation": 0.030524694664730578, "minimum": 0.19724759164102149, "maximum": 0.29804323381152825, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.49096144703725, "standard_deviation": 0.016140917876672267, "minimum": 5.466734426260527, "maximum": 5.518723774875174, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.38289942538191546, "standard_deviation": 0.01867319424882594, "minimum": 0.35756807258597584, "maximum": 0.41113061187522315, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08754611400310594, "standard_deviation": 0.015071666368657841, "minimum": -0.11430345668482932, "maximum": -0.06433749275619532, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994861343510533, "standard_deviation": 0.0005385027985114017, "minimum": 0.9940596102138513, "maximum": 0.9956808255717882, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9394317506627526, "standard_deviation": 0.0584990007615398, "minimum": 0.8759965071069786, "maximum": 1.0521785994234991, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3469785033947123, "standard_deviation": 0.17756777845525062, "minimum": 0.05749135595075968, "maximum": 0.6000797356534416, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06497854752493074, "standard_deviation": 0.00320372841905391, "minimum": 0.059822598957898564, "maximum": 0.06963840256761804, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985737169276985, "standard_deviation": 0.0002581133354379448, "minimum": 0.9982259594213663, "maximum": 0.9989193796992635, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4961096279360572, "standard_deviation": 0.03816325359351779, "minimum": 0.4425685513394286, "maximum": 0.5578762477308801, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38732155040358796, "standard_deviation": 0.21776418478150625, "minimum": 0.08676108821144614, "maximum": 0.7105988079039682, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03360893376686031, "standard_deviation": 0.0026621499416908533, "minimum": 0.02974499195635705, "maximum": 0.03715639219511744, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0027", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.701666831970215, "predicted_mos": 2.701666831970215, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "63909735762770252680fd40bc920a759aa6a2970104e8867808eb3d7fe1e9b6", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.5879950001835823, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.911308630874922, "全帯域_10ms_rms_envelope_nrmse": 0.06106893948252381, "全帯域_10ms_envelope_correlation": 0.9955507838776806, "全帯域_10ms_modulation_mae_db": 0.4259009240390341, "全帯域_40ms_log_rms_mae_db": 0.4968170378809172, "全帯域_40ms_rms_envelope_nrmse": 0.03728931533101154, "全帯域_40ms_envelope_correlation": 0.9982965079801702, "全帯域_40ms_modulation_mae_db": 0.4168516823272159, "300–1000 Hz_10ms_log_rms_mae_db": 1.4603193485745491, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08419256306282046, "300–1000 Hz_10ms_envelope_correlation": 0.993657657596624, "300–1000 Hz_10ms_modulation_mae_db": 0.4875244238528816, "300–1000 Hz_40ms_log_rms_mae_db": 0.9238741729350439, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05200523992590207, "300–1000 Hz_40ms_envelope_correlation": 0.9975398088955205, "300–1000 Hz_40ms_modulation_mae_db": 0.54253581813479, "1000–3000 Hz_10ms_log_rms_mae_db": 1.479294563107964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13827788680863146, "1000–3000 Hz_10ms_envelope_correlation": 0.984670951628406, "1000–3000 Hz_10ms_modulation_mae_db": 0.47707264549563605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0309695485613823, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08561247095025558, "1000–3000 Hz_40ms_envelope_correlation": 0.9939612050490899, "1000–3000 Hz_40ms_modulation_mae_db": 0.5793626415665742, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5168085041254393, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18075172005877524, "3000–6000 Hz_10ms_envelope_correlation": 0.9683232782409494, "3000–6000 Hz_10ms_modulation_mae_db": 0.5656802633215161, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9481315406549973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10745264296481728, "3000–6000 Hz_40ms_envelope_correlation": 0.9874022920230914, "3000–6000 Hz_40ms_modulation_mae_db": 0.720271381253799, "mr_stft_db": 5.626763287166177, "band_mae_db": 0.15103146478149224, "rms_gain_error_db": -0.11418665090378392, "output_peak": 0.3770600684996338}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8237176517728776, "全帯域_10ms_rms_envelope_nrmse": 0.06209182039161235, "全帯域_10ms_envelope_correlation": 0.9953029793314069, "全帯域_10ms_modulation_mae_db": 0.8229328276850035, "全帯域_40ms_log_rms_mae_db": 0.4066588738120912, "全帯域_40ms_rms_envelope_nrmse": 0.036961426416364705, "全帯域_40ms_envelope_correlation": 0.9982149201705682, "全帯域_40ms_modulation_mae_db": 0.8132868010607108, "300–1000 Hz_10ms_log_rms_mae_db": 1.3636138451134951, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08085047219465753, "300–1000 Hz_10ms_envelope_correlation": 0.9940551463032887, "300–1000 Hz_10ms_modulation_mae_db": 0.689856471423785, "300–1000 Hz_40ms_log_rms_mae_db": 0.8382413708847869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050450607918323405, "300–1000 Hz_40ms_envelope_correlation": 0.9975451909978317, "300–1000 Hz_40ms_modulation_mae_db": 0.8250592755996546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4830149263883792, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15479994212035428, "1000–3000 Hz_10ms_envelope_correlation": 0.9805930402995304, "1000–3000 Hz_10ms_modulation_mae_db": 0.5512412626963962, "1000–3000 Hz_40ms_log_rms_mae_db": 1.030010351499327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09230792345595103, "1000–3000 Hz_40ms_envelope_correlation": 0.9928206506378752, "1000–3000 Hz_40ms_modulation_mae_db": 0.6738820782612956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5521061957274165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2005228328999503, "3000–6000 Hz_10ms_envelope_correlation": 0.9642252977178051, "3000–6000 Hz_10ms_modulation_mae_db": 1.6567946313848612, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9768740078062917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13185273617758525, "3000–6000 Hz_40ms_envelope_correlation": 0.9859904404057218, "3000–6000 Hz_40ms_modulation_mae_db": 2.34259067369411, "mr_stft_db": 5.613805492909667, "band_mae_db": 0.1784582002547815, "rms_gain_error_db": -0.08030755857267509, "output_peak": 0.4092322585345144}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9104850085854092, "全帯域_10ms_rms_envelope_nrmse": 0.06398356045724876, "全帯域_10ms_envelope_correlation": 0.995015019961159, "全帯域_10ms_modulation_mae_db": 0.3468432049722751, "全帯域_40ms_log_rms_mae_db": 0.43157438049290586, "全帯域_40ms_rms_envelope_nrmse": 0.03194423929897383, "全帯域_40ms_envelope_correlation": 0.9987114928189023, "全帯域_40ms_modulation_mae_db": 0.27252147016440986, "300–1000 Hz_10ms_log_rms_mae_db": 1.4554204200980705, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0841817391149559, "300–1000 Hz_10ms_envelope_correlation": 0.9935849963663723, "300–1000 Hz_10ms_modulation_mae_db": 0.7185894785641883, "300–1000 Hz_40ms_log_rms_mae_db": 0.8566037282903672, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04325812131527657, "300–1000 Hz_40ms_envelope_correlation": 0.9982949843966724, "300–1000 Hz_40ms_modulation_mae_db": 0.9607155826258283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.519440831777031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14285933248308555, "1000–3000 Hz_10ms_envelope_correlation": 0.9836168264379958, "1000–3000 Hz_10ms_modulation_mae_db": 0.9589641415008586, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0452415080698365, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.088868490959477, "1000–3000 Hz_40ms_envelope_correlation": 0.9934968757129908, "1000–3000 Hz_40ms_modulation_mae_db": 0.8305014601538915, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5031867748557153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1924764204657501, "3000–6000 Hz_10ms_envelope_correlation": 0.9645237580939217, "3000–6000 Hz_10ms_modulation_mae_db": 0.38468660169617036, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9803241272189406, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12761531606625043, "3000–6000 Hz_40ms_envelope_correlation": 0.9815867897590265, "3000–6000 Hz_40ms_modulation_mae_db": 0.5395044942984526, "mr_stft_db": 5.617871400483789, "band_mae_db": 0.1406347422125599, "rms_gain_error_db": -0.08352369476889608, "output_peak": 0.3748054937073717}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9065610355730302, "全帯域_10ms_rms_envelope_nrmse": 0.05693764066119189, "全帯域_10ms_envelope_correlation": 0.9960223133679849, "全帯域_10ms_modulation_mae_db": 0.06977827461890808, "全帯域_40ms_log_rms_mae_db": 0.49625847304995746, "全帯域_40ms_rms_envelope_nrmse": 0.02954805551848574, "全帯域_40ms_envelope_correlation": 0.9988399731571074, "全帯域_40ms_modulation_mae_db": 0.06629286129414234, "300–1000 Hz_10ms_log_rms_mae_db": 1.3977522504676965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07592942372233291, "300–1000 Hz_10ms_envelope_correlation": 0.9947197849921607, "300–1000 Hz_10ms_modulation_mae_db": 0.30572049546811103, "300–1000 Hz_40ms_log_rms_mae_db": 0.8715697725179286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04115392476085973, "300–1000 Hz_40ms_envelope_correlation": 0.9983481011659451, "300–1000 Hz_40ms_modulation_mae_db": 0.24122543843840227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4945244279004806, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1464249263492513, "1000–3000 Hz_10ms_envelope_correlation": 0.9826762680321212, "1000–3000 Hz_10ms_modulation_mae_db": 0.9506141627073359, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9326292857535503, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0788146733539632, "1000–3000 Hz_40ms_envelope_correlation": 0.9948873620031798, "1000–3000 Hz_40ms_modulation_mae_db": 1.0190888297782934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.493683825866398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1676101519764823, "3000–6000 Hz_10ms_envelope_correlation": 0.9745481252162456, "3000–6000 Hz_10ms_modulation_mae_db": 1.1179995315219322, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9577473741957517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1047839025773496, "3000–6000 Hz_40ms_envelope_correlation": 0.9904562157872399, "3000–6000 Hz_40ms_modulation_mae_db": 1.429044549472106, "mr_stft_db": 5.635192200505699, "band_mae_db": 0.15111098207096182, "rms_gain_error_db": -0.05243195018062974, "output_peak": 0.3650214820912704}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9150481135786912, "全帯域_10ms_rms_envelope_nrmse": 0.06258114684274982, "全帯域_10ms_envelope_correlation": 0.9952391106497139, "全帯域_10ms_modulation_mae_db": 0.4228687925041417, "全帯域_40ms_log_rms_mae_db": 0.5252478566274844, "全帯域_40ms_rms_envelope_nrmse": 0.0338094004813638, "全帯域_40ms_envelope_correlation": 0.9985432096011817, "全帯域_40ms_modulation_mae_db": 0.4181050096873756, "300–1000 Hz_10ms_log_rms_mae_db": 1.4605728890873666, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07760747034946683, "300–1000 Hz_10ms_envelope_correlation": 0.9945158467661207, "300–1000 Hz_10ms_modulation_mae_db": 0.33187809653260636, "300–1000 Hz_40ms_log_rms_mae_db": 0.9060872361280824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041496448662720356, "300–1000 Hz_40ms_envelope_correlation": 0.9983717482930451, "300–1000 Hz_40ms_modulation_mae_db": 0.3671515770621676, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5017890717761895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13587290401768837, "1000–3000 Hz_10ms_envelope_correlation": 0.9858174695513082, "1000–3000 Hz_10ms_modulation_mae_db": 0.9278899144238598, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9549928164261661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0833685118730441, "1000–3000 Hz_40ms_envelope_correlation": 0.9951891672286843, "1000–3000 Hz_40ms_modulation_mae_db": 0.9211665954808099, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6040892159568645, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20464125191886293, "3000–6000 Hz_10ms_envelope_correlation": 0.9617559183292621, "3000–6000 Hz_10ms_modulation_mae_db": 2.0065912067532907, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0202680465092275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12556742214953798, "3000–6000 Hz_40ms_envelope_correlation": 0.9868603852728557, "3000–6000 Hz_40ms_modulation_mae_db": 2.6818835433417925, "mr_stft_db": 5.615200624314838, "band_mae_db": 0.24768655328149122, "rms_gain_error_db": -0.08556396131536823, "output_peak": 0.38265203477308607}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.836075847060398, "全帯域_10ms_rms_envelope_nrmse": 0.053964897242367256, "全帯域_10ms_envelope_correlation": 0.9965807203794392, "全帯域_10ms_modulation_mae_db": 0.5054452968439058, "全帯域_40ms_log_rms_mae_db": 0.4605665770316051, "全帯域_40ms_rms_envelope_nrmse": 0.028006093993294214, "全帯域_40ms_envelope_correlation": 0.999153715672511, "全帯域_40ms_modulation_mae_db": 0.4867609946226623, "300–1000 Hz_10ms_log_rms_mae_db": 1.4242109674348715, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07716549393588533, "300–1000 Hz_10ms_envelope_correlation": 0.9948348836104366, "300–1000 Hz_10ms_modulation_mae_db": 0.47260162509875303, "300–1000 Hz_40ms_log_rms_mae_db": 0.9616596354290989, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0468154801980819, "300–1000 Hz_40ms_envelope_correlation": 0.9982045755993758, "300–1000 Hz_40ms_modulation_mae_db": 0.5667514178197974, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4716737633821955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16605795193867673, "1000–3000 Hz_10ms_envelope_correlation": 0.9782511915197639, "1000–3000 Hz_10ms_modulation_mae_db": 0.5795999975258358, "1000–3000 Hz_40ms_log_rms_mae_db": 1.038333791998038, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11855643439771267, "1000–3000 Hz_40ms_envelope_correlation": 0.9881983266554834, "1000–3000 Hz_40ms_modulation_mae_db": 0.499280756503387, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5657797399895375, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20947000231481114, "3000–6000 Hz_10ms_envelope_correlation": 0.9581710665293683, "3000–6000 Hz_10ms_modulation_mae_db": 0.932524932811281, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9761262713398562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12495597934591748, "3000–6000 Hz_40ms_envelope_correlation": 0.9848771992854687, "3000–6000 Hz_40ms_modulation_mae_db": 1.1319672606517888, "mr_stft_db": 5.639045074974042, "band_mae_db": 0.22102444284966735, "rms_gain_error_db": -0.11671486557964782, "output_peak": 0.3759746331989187}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8456791844933205, "全帯域_10ms_rms_envelope_nrmse": 0.060838538619240976, "全帯域_10ms_envelope_correlation": 0.9955654483100904, "全帯域_10ms_modulation_mae_db": 0.6517952169792656, "全帯域_40ms_log_rms_mae_db": 0.40251493592949283, "全帯域_40ms_rms_envelope_nrmse": 0.03093350832873159, "全帯域_40ms_envelope_correlation": 0.9988738043268576, "全帯域_40ms_modulation_mae_db": 0.8024364754966843, "300–1000 Hz_10ms_log_rms_mae_db": 1.4273824404501645, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07635369051070137, "300–1000 Hz_10ms_envelope_correlation": 0.9948443381233648, "300–1000 Hz_10ms_modulation_mae_db": 0.678214110987712, "300–1000 Hz_40ms_log_rms_mae_db": 0.8611546809213656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03824037476440402, "300–1000 Hz_40ms_envelope_correlation": 0.9988250511769652, "300–1000 Hz_40ms_modulation_mae_db": 0.7059734951107405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4277583508931913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14659278299595266, "1000–3000 Hz_10ms_envelope_correlation": 0.9830715350679802, "1000–3000 Hz_10ms_modulation_mae_db": 1.1957966276304608, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9330167942558384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08346626754501787, "1000–3000 Hz_40ms_envelope_correlation": 0.9949554250399283, "1000–3000 Hz_40ms_modulation_mae_db": 1.0960243002343464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5533090192133314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2092620345657371, "3000–6000 Hz_10ms_envelope_correlation": 0.9573123459595322, "3000–6000 Hz_10ms_modulation_mae_db": 0.4768420775106934, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9936024664892434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11408719425944834, "3000–6000 Hz_40ms_envelope_correlation": 0.9851343595344384, "3000–6000 Hz_40ms_modulation_mae_db": 0.8450770350602103, "mr_stft_db": 5.639410658321801, "band_mae_db": 0.1514584935117935, "rms_gain_error_db": -0.10590929436971172, "output_peak": 0.3746943130950178}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9238846379754524, "全帯域_10ms_rms_envelope_nrmse": 0.06136840656415703, "全帯域_10ms_envelope_correlation": 0.9956509062722477, "全帯域_10ms_modulation_mae_db": 0.589707259647975, "全帯域_40ms_log_rms_mae_db": 0.4762330288842016, "全帯域_40ms_rms_envelope_nrmse": 0.037566711240336975, "全帯域_40ms_envelope_correlation": 0.998439572155677, "全帯域_40ms_modulation_mae_db": 0.7072363351543037, "300–1000 Hz_10ms_log_rms_mae_db": 1.4830815652681233, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07864787332837729, "300–1000 Hz_10ms_envelope_correlation": 0.9947941442462711, "300–1000 Hz_10ms_modulation_mae_db": 0.6028199514231573, "300–1000 Hz_40ms_log_rms_mae_db": 0.9000438313726206, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05087151389113576, "300–1000 Hz_40ms_envelope_correlation": 0.9980000369520325, "300–1000 Hz_40ms_modulation_mae_db": 0.5774842151013382, "1000–3000 Hz_10ms_log_rms_mae_db": 1.526782136830732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14453180962954093, "1000–3000 Hz_10ms_envelope_correlation": 0.9831449890357775, "1000–3000 Hz_10ms_modulation_mae_db": 0.4922745877544368, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0109927250054183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0893661434337161, "1000–3000 Hz_40ms_envelope_correlation": 0.9933121505636825, "1000–3000 Hz_40ms_modulation_mae_db": 0.28354568134281033, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5314770940366091, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2033894440957799, "3000–6000 Hz_10ms_envelope_correlation": 0.9621543211935795, "3000–6000 Hz_10ms_modulation_mae_db": 1.9444136347982106, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9884785456819298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14290180823937812, "3000–6000 Hz_40ms_envelope_correlation": 0.9808137927127684, "3000–6000 Hz_40ms_modulation_mae_db": 2.3896251547030376, "mr_stft_db": 5.6343773449561185, "band_mae_db": 0.21595133676086067, "rms_gain_error_db": -0.1517311182635971, "output_peak": 0.3915146824337042}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 52320, "output_samples": 52320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.608381828956334, "spectral_convergence": 0.20792414118289393, "spectral_convergence_status": "ok", "amplitude_floor": 1.2746778473169441e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.498807151324647, "spectral_convergence": 0.1980163810882508, "spectral_convergence_status": "ok", "amplitude_floor": 1.090624511245042e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.773100881217547, "spectral_convergence": 0.2046928106538778, "spectral_convergence_status": "ok", "amplitude_floor": 7.935262595002381e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004730173956506593, "output_power": 0.0004791629210624787, "delta_db": 0.05606091175194776, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018195471619226037, "output_power": 0.0018071753637050984, "delta_db": -0.02963019365803924, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018173989836957952, "output_power": 0.0017217627698491455, "delta_db": -0.23476968141478077, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013671881984837986, "output_power": 0.0001344378938430915, "delta_db": -0.07306600892905969, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012324536624073353, "output_power": 0.00011791479563568708, "delta_db": -0.1920229708929429, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.679288082514536e-05, "output_power": 8.061580619472488e-05, "delta_db": -0.3206390220421831, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15103146478149224, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11418665090378392, "pitch": null, "output_rms_dbfs": -23.624032560720828, "output_peak": 0.3770600684996338, "reference_rms_dbfs": -23.509845909817045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.626763287166177}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 52320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00017439460044090568, "log_rms_mae_db": 0.911308630874922, "rms_envelope_nrmse": 0.06106893948252381, "envelope_correlation": 0.9955507838776806, "rms_gain_error_db": -0.11419096056188208, "reference_mean_power": 0.004463526965610992, "output_mean_power": 0.004347694969654415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 1.3056377024951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008270090313661458, "output_power": 0.0007932977734929189, "reference_active": true, "delta_db": -0.18074016922256478}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010887958720618104, "output_power": 9.688688300068809e-05, "reference_active": true, "delta_db": -0.5068148145059201}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8937894537156724e-05, "output_power": 2.5261093754495928e-05, "reference_active": true, "delta_db": -0.5901477883886175}], "active_band_count": 3, "active_band_mae_db": 0.4259009240390341}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015806536675822195, "log_rms_mae_db": 0.4968170378809172, "rms_envelope_nrmse": 0.03728931533101154, "envelope_correlation": 0.9982965079801702, "rms_gain_error_db": -0.1141973716270274, "reference_mean_power": 0.004504823508022685, "output_mean_power": 0.004387913357222413}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.0922276511075418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000711888936919089, "output_power": 0.0006777009220035846, "reference_active": true, "delta_db": -0.21374167361197713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.550528203435593e-05, "output_power": 6.227702314109344e-05, "reference_active": true, "delta_db": -0.21948475848946986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5248696894099073e-06, "output_power": 1.263280726632253e-06, "reference_active": true, "delta_db": -0.8173286148802007}], "active_band_count": 3, "active_band_mae_db": 0.4168516823272159}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00016816220894332027, "log_rms_mae_db": 1.4603193485745491, "rms_envelope_nrmse": 0.08419256306282046, "envelope_correlation": 0.993657657596624, "rms_gain_error_db": -0.15184749248207247, "reference_mean_power": 0.002573247092376177, "output_mean_power": 0.002484830341604969}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 9.221294414644421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007324784888809201, "output_power": 0.0006802296135852621, "reference_active": true, "delta_db": -0.3213934015254959}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.576574656453582e-05, "output_power": 7.280307511960963e-05, "reference_active": true, "delta_db": -0.7116414855947353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.804896765293062e-05, "output_power": 1.6349276102917007e-05, "reference_active": true, "delta_db": -0.4295383844384134}], "active_band_count": 3, "active_band_mae_db": 0.4875244238528816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.0001473388036867024, "log_rms_mae_db": 0.9238741729350439, "rms_envelope_nrmse": 0.05200523992590207, "envelope_correlation": 0.9975398088955205, "rms_gain_error_db": -0.15184988968098734, "reference_mean_power": 0.002597106693874505, "output_mean_power": 0.0025078687430926225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.911179205548874e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006603877113347142, "output_power": 0.0006083501392438107, "reference_active": true, "delta_db": -0.35645371408497667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.506710493994746e-05, "output_power": 4.0797754268387016e-05, "reference_active": true, "delta_db": -0.4322340257538854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1610111255482175e-06, "output_power": 9.570715575940085e-07, "reference_active": true, "delta_db": -0.8389197145655081}], "active_band_count": 3, "active_band_mae_db": 0.54253581813479}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 5.380259383983556e-05, "log_rms_mae_db": 1.479294563107964, "rms_envelope_nrmse": 0.13827788680863146, "envelope_correlation": 0.984670951628406, "rms_gain_error_db": -0.19438089920330703, "reference_mean_power": 0.00019859577428883629, "output_mean_power": 0.00018990303842076085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 8.760825576371796e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.452705657171872e-05, "output_power": 4.601647935896862e-05, "reference_active": true, "delta_db": -0.7369866615731662}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.928893294666125e-06, "output_power": 5.183799015397344e-06, "reference_active": true, "delta_db": -0.5832547933296627}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9788817769833428e-06, "output_power": 1.928955465556438e-06, "reference_active": true, "delta_db": -0.11097648158407925}], "active_band_count": 3, "active_band_mae_db": 0.47707264549563605}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 5.051409218298903e-05, "log_rms_mae_db": 1.0309695485613823, "rms_envelope_nrmse": 0.08561247095025558, "envelope_correlation": 0.9939612050490899, "rms_gain_error_db": -0.19432499773921294, "reference_mean_power": 0.00020041602526910178, "output_mean_power": 0.00019164608200506784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.85711379260727e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.022793901144941e-05, "output_power": 4.211741424152688e-05, "reference_active": true, "delta_db": -0.7648365800835852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.619667417561746e-06, "output_power": 2.1727113630157067e-06, "reference_active": true, "delta_db": -0.8124412286513449}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1855266901360492e-07, "output_power": 1.1424319075707118e-07, "reference_active": true, "delta_db": -0.16081011596479244}], "active_band_count": 3, "active_band_mae_db": 0.5793626415665742}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 6.112392927022402e-05, "log_rms_mae_db": 1.5168085041254393, "rms_envelope_nrmse": 0.18075172005877524, "envelope_correlation": 0.9683232782409494, "rms_gain_error_db": -0.26388103923688605, "reference_mean_power": 8.711996692518435e-05, "output_mean_power": 8.198409329593129e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 4.203401537603793e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8248175978150172e-05, "output_power": 1.7264062877352335e-05, "reference_active": true, "delta_db": -0.2407645136865435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.243431842943793e-05, "output_power": 1.1851019065061232e-05, "reference_active": true, "delta_db": -0.20866288200949015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.767669752037568e-06, "output_power": 5.077826040764028e-06, "reference_active": true, "delta_db": -1.2476133942685146}], "active_band_count": 3, "active_band_mae_db": 0.5656802633215161}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 3.626522691368819e-05, "log_rms_mae_db": 0.9481315406549973, "rms_envelope_nrmse": 0.10745264296481728, "envelope_correlation": 0.9874022920230914, "rms_gain_error_db": -0.263897354408183, "reference_mean_power": 8.789733624317315e-05, "output_mean_power": 8.271532460226443e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 3.223393789064049e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.051966991658498e-05, "output_power": 1.8837849413798236e-05, "reference_active": true, "delta_db": -0.371390494427673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.880534452438109e-06, "output_power": 5.589067110815069e-06, "reference_active": true, "delta_db": -0.22077474283312276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.871896444237529e-07, "output_power": 4.0917915126909226e-07, "reference_active": true, "delta_db": -1.5686489065006013}], "active_band_count": 3, "active_band_mae_db": 0.720271381253799}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8950688219202783, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "41c29b4816b771e632bbc72d4159b9ccdb6d98227d297b0f2bd6b35ac49970a2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9827302839466103, "standard_deviation": 0.0023605172959424354, "minimum": 0.9782511915197639, "maximum": 0.9858174695513082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4880347590070202, "standard_deviation": 0.030987358975558793, "minimum": 1.4277583508931913, "maximum": 1.526782136830732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7666816674668525, "standard_deviation": 0.27295164689221146, "minimum": 0.47707264549563605, "maximum": 1.1957966276304608, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14692719204289767, "standard_deviation": 0.00961860823979153, "minimum": 0.13587290401768837, "maximum": 0.16605795193867673, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9933526453613643, "standard_deviation": 0.0022532477278661184, "minimum": 0.9881983266554834, "maximum": 0.9951891672286843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9970233526961946, "standard_deviation": 0.048516821721007075, "minimum": 0.9326292857535503, "maximum": 1.0452415080698365, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.737856542915176, "standard_deviation": 0.2782833117824265, "minimum": 0.28354568134281033, "maximum": 1.0960243002343464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09004511449614219, "standard_deviation": 0.012267788824017621, "minimum": 0.0788146733539632, "maximum": 0.11855643439771267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9638767639100829, "standard_deviation": 0.005571622995519886, "minimum": 0.9573123459595322, "maximum": 0.9745481252162456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.540055046221414, "standard_deviation": 0.036266031453042906, "minimum": 1.493683825866398, "maximum": 1.6040892159568645, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1356916099747445, "standard_deviation": 0.6599393535531455, "minimum": 0.38468660169617036, "maximum": 2.0065912067532907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19601548228701862, "standard_deviation": 0.014918640624033738, "minimum": 0.1676101519764823, "maximum": 0.20947000231481114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9853901843475763, "standard_deviation": 0.003116151948016103, "minimum": 0.9808137927127684, "maximum": 0.9904562157872399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9801940474870298, "standard_deviation": 0.022083568168316213, "minimum": 0.9481315406549973, "maximum": 1.0202680465092275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5099955115594121, "standard_deviation": 0.8449325066747887, "minimum": 0.5395044942984526, "maximum": 2.6818835433417925, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12240212522253557, "standard_deviation": 0.012852440298454778, "minimum": 0.1047839025773496, "maximum": 0.14290180823937812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943758497505799, "standard_deviation": 0.0005328031110802446, "minimum": 0.9935849963663723, "maximum": 0.9948443381233648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.434044215811792, "standard_deviation": 0.03905010105840463, "minimum": 1.3636138451134951, "maximum": 1.4830815652681233, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5359005816688993, "standard_deviation": 0.16158325973539833, "minimum": 0.30572049546811103, "maximum": 0.7185894785641883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0793660907773997, "standard_deviation": 0.0033386296635325107, "minimum": 0.07592942372233291, "maximum": 0.08419256306282046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981411871846735, "standard_deviation": 0.00043557911261061887, "minimum": 0.9975398088955205, "maximum": 0.9988250511769652, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8899043035599118, "standard_deviation": 0.0407223346300576, "minimum": 0.8382413708847869, "maximum": 0.9616596354290989, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5983621024865899, "standard_deviation": 0.2329403553808784, "minimum": 0.24122543843840227, "maximum": 0.9607155826258283, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.045536463929587975, "standard_deviation": 0.005210492845297292, "minimum": 0.03824037476440402, "maximum": 0.05200523992590207, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18216952696545102, "standard_deviation": 0.04063722412565066, "minimum": 0.1406347422125599, "maximum": 0.24768655328149122, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.627708260454017, "standard_deviation": 0.010779214269547118, "minimum": 5.613805492909667, "maximum": 5.639410658321801, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.38136937079168964, "standard_deviation": 0.013540841454674515, "minimum": 0.3650214820912704, "maximum": 0.4092322585345144, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09879613674428872, "standard_deviation": 0.030024897568550125, "minimum": -0.1517311182635971, "maximum": -0.05243195018062974, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956159102687153, "standard_deviation": 0.0004935306462204022, "minimum": 0.995015019961159, "maximum": 0.9965807203794392, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8840950137392626, "standard_deviation": 0.041248524912766586, "minimum": 0.8237176517728776, "maximum": 0.9238846379754524, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4794089746613136, "standard_deviation": 0.22424481305381433, "minimum": 0.06977827461890808, "maximum": 0.8229328276850035, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06035436878263649, "standard_deviation": 0.0032824423135298184, "minimum": 0.053964897242367256, "maximum": 0.06398356045724876, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986341494853719, "standard_deviation": 0.0003188403642548946, "minimum": 0.9982149201705682, "maximum": 0.999153715672511, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.461983895463582, "standard_deviation": 0.04490883404325392, "minimum": 0.40251493592949283, "maximum": 0.5252478566274844, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49793645372593814, "standard_deviation": 0.2635400054711616, "minimum": 0.06629286129414234, "maximum": 0.8132868010607108, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033257343826070304, "standard_deviation": 0.0037284875445100225, "minimum": 0.028006093993294214, "maximum": 0.037566711240336975, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0027", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.878957748413086, "predicted_mos": 2.878957748413086, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "77d48f60974d8e5b21f693f8cc6a88c5f2b84682baf52fd4c1ad1cddd03ef712", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.5456471000798047, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8669629877038829, "全帯域_10ms_rms_envelope_nrmse": 0.06509032650609449, "全帯域_10ms_envelope_correlation": 0.9950318654731803, "全帯域_10ms_modulation_mae_db": 0.26348262349308543, "全帯域_40ms_log_rms_mae_db": 0.4633180987375004, "全帯域_40ms_rms_envelope_nrmse": 0.038093083236616226, "全帯域_40ms_envelope_correlation": 0.9983549860555876, "全帯域_40ms_modulation_mae_db": 0.266202555490349, "300–1000 Hz_10ms_log_rms_mae_db": 1.3762399787859214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08833479042888094, "300–1000 Hz_10ms_envelope_correlation": 0.9930837387967069, "300–1000 Hz_10ms_modulation_mae_db": 0.6982315809445409, "300–1000 Hz_40ms_log_rms_mae_db": 0.7995676776238065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05286252788322125, "300–1000 Hz_40ms_envelope_correlation": 0.9975545442185617, "300–1000 Hz_40ms_modulation_mae_db": 0.5715602669228051, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5069351665351136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14774899721287515, "1000–3000 Hz_10ms_envelope_correlation": 0.9825411092961067, "1000–3000 Hz_10ms_modulation_mae_db": 0.9405790340965842, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0655624723221209, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09736356818531933, "1000–3000 Hz_40ms_envelope_correlation": 0.9922548169934553, "1000–3000 Hz_40ms_modulation_mae_db": 0.8192411634863431, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4675942366345256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1830712584517919, "3000–6000 Hz_10ms_envelope_correlation": 0.967852563868846, "3000–6000 Hz_10ms_modulation_mae_db": 0.9331452243991508, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8868725394611736, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09823302236963541, "3000–6000 Hz_40ms_envelope_correlation": 0.9904759978613888, "3000–6000 Hz_40ms_modulation_mae_db": 1.3239270285313334, "mr_stft_db": 5.5819664797595765, "band_mae_db": 0.25487133145603075, "rms_gain_error_db": -0.14642176012489885, "output_peak": 0.3778397736978986}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.844589770832838, "全帯域_10ms_rms_envelope_nrmse": 0.06450526331444464, "全帯域_10ms_envelope_correlation": 0.9950126410827161, "全帯域_10ms_modulation_mae_db": 0.31772236556023753, "全帯域_40ms_log_rms_mae_db": 0.457153225483839, "全帯域_40ms_rms_envelope_nrmse": 0.03513372200695746, "全帯域_40ms_envelope_correlation": 0.9985096318485371, "全帯域_40ms_modulation_mae_db": 0.33202234579477613, "300–1000 Hz_10ms_log_rms_mae_db": 1.3827289823670261, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08528044591247609, "300–1000 Hz_10ms_envelope_correlation": 0.9935315315534873, "300–1000 Hz_10ms_modulation_mae_db": 0.36088082489764567, "300–1000 Hz_40ms_log_rms_mae_db": 0.8625384240470526, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04750002271803143, "300–1000 Hz_40ms_envelope_correlation": 0.9980536836879307, "300–1000 Hz_40ms_modulation_mae_db": 0.49999609806009077, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4997466969219098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15091432417464, "1000–3000 Hz_10ms_envelope_correlation": 0.9815493497394485, "1000–3000 Hz_10ms_modulation_mae_db": 0.8026747082061115, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9471069790088135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07596442775260656, "1000–3000 Hz_40ms_envelope_correlation": 0.9951230619442516, "1000–3000 Hz_40ms_modulation_mae_db": 0.967915434627482, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5786242233728252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19351073853743933, "3000–6000 Hz_10ms_envelope_correlation": 0.9662993359459091, "3000–6000 Hz_10ms_modulation_mae_db": 1.7450410488523715, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0374462095343862, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.125696226448387, "3000–6000 Hz_40ms_envelope_correlation": 0.9865782028251564, "3000–6000 Hz_40ms_modulation_mae_db": 2.1385167425176466, "mr_stft_db": 5.596655248273303, "band_mae_db": 0.2128885021094277, "rms_gain_error_db": -0.10605288224548848, "output_peak": 0.37414928795224023}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9453583295246496, "全帯域_10ms_rms_envelope_nrmse": 0.06147827862225194, "全帯域_10ms_envelope_correlation": 0.9954047203435757, "全帯域_10ms_modulation_mae_db": 0.3393612929284678, "全帯域_40ms_log_rms_mae_db": 0.5283667464449413, "全帯域_40ms_rms_envelope_nrmse": 0.03529720758325607, "全帯域_40ms_envelope_correlation": 0.9983913028663274, "全帯域_40ms_modulation_mae_db": 0.23356965301760627, "300–1000 Hz_10ms_log_rms_mae_db": 1.3758451021805007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07812605853213757, "300–1000 Hz_10ms_envelope_correlation": 0.9944607580189827, "300–1000 Hz_10ms_modulation_mae_db": 0.5214196459386399, "300–1000 Hz_40ms_log_rms_mae_db": 0.7955433779904671, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04619991507949316, "300–1000 Hz_40ms_envelope_correlation": 0.9979694002327187, "300–1000 Hz_40ms_modulation_mae_db": 0.46155402888377745, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4348022406503944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14426299583469127, "1000–3000 Hz_10ms_envelope_correlation": 0.9832098048735136, "1000–3000 Hz_10ms_modulation_mae_db": 0.705712637414754, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9739579703025278, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09003799163946531, "1000–3000 Hz_40ms_envelope_correlation": 0.9931835630766895, "1000–3000 Hz_40ms_modulation_mae_db": 0.8935577032778123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5446641403352135, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19491730941024468, "3000–6000 Hz_10ms_envelope_correlation": 0.9633696007780218, "3000–6000 Hz_10ms_modulation_mae_db": 1.14428252405171, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9543100021338021, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11456836017961858, "3000–6000 Hz_40ms_envelope_correlation": 0.9863598318381483, "3000–6000 Hz_40ms_modulation_mae_db": 1.2547059929497253, "mr_stft_db": 5.566839477191699, "band_mae_db": 0.17088723710153383, "rms_gain_error_db": -0.08294140538453806, "output_peak": 0.3887130279582719}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9122282796451916, "全帯域_10ms_rms_envelope_nrmse": 0.06199231128178806, "全帯域_10ms_envelope_correlation": 0.9954274072859877, "全帯域_10ms_modulation_mae_db": 0.2590508623265034, "全帯域_40ms_log_rms_mae_db": 0.4720188518677298, "全帯域_40ms_rms_envelope_nrmse": 0.03728495134054024, "全帯域_40ms_envelope_correlation": 0.9983193428779612, "全帯域_40ms_modulation_mae_db": 0.1595800779988158, "300–1000 Hz_10ms_log_rms_mae_db": 1.3579947043843859, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07914947345519514, "300–1000 Hz_10ms_envelope_correlation": 0.994453356346624, "300–1000 Hz_10ms_modulation_mae_db": 0.6434040561137983, "300–1000 Hz_40ms_log_rms_mae_db": 0.8991936009490951, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04674857199506487, "300–1000 Hz_40ms_envelope_correlation": 0.9980990817489576, "300–1000 Hz_40ms_modulation_mae_db": 0.6954007904048735, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5040387652185376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16141299925462294, "1000–3000 Hz_10ms_envelope_correlation": 0.9806307102701661, "1000–3000 Hz_10ms_modulation_mae_db": 0.8897157385407816, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9728423419768888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10539955685912301, "1000–3000 Hz_40ms_envelope_correlation": 0.9930628464776953, "1000–3000 Hz_40ms_modulation_mae_db": 1.16819269852178, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5487496211570062, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20215169725991536, "3000–6000 Hz_10ms_envelope_correlation": 0.9617583727532747, "3000–6000 Hz_10ms_modulation_mae_db": 0.5985446198606674, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0008771384212858, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1489667067264915, "3000–6000 Hz_40ms_envelope_correlation": 0.9760595558720987, "3000–6000 Hz_40ms_modulation_mae_db": 0.888706676794505, "mr_stft_db": 5.5961581927282715, "band_mae_db": 0.22972161506374508, "rms_gain_error_db": -0.12037281612769105, "output_peak": 0.3807603193753109}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8306111273417385, "全帯域_10ms_rms_envelope_nrmse": 0.06206654929123858, "全帯域_10ms_envelope_correlation": 0.9952717339777153, "全帯域_10ms_modulation_mae_db": 0.36695609363644355, "全帯域_40ms_log_rms_mae_db": 0.3973161111786982, "全帯域_40ms_rms_envelope_nrmse": 0.030794213056665798, "全帯域_40ms_envelope_correlation": 0.9987371246681855, "全帯域_40ms_modulation_mae_db": 0.3567963669388999, "300–1000 Hz_10ms_log_rms_mae_db": 1.434938745376035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07795063686436429, "300–1000 Hz_10ms_envelope_correlation": 0.994425108490652, "300–1000 Hz_10ms_modulation_mae_db": 0.3145400689475169, "300–1000 Hz_40ms_log_rms_mae_db": 0.8571574654500551, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0422602735481605, "300–1000 Hz_40ms_envelope_correlation": 0.9982178737189817, "300–1000 Hz_40ms_modulation_mae_db": 0.2886117159606371, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5142964311113745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14000842899938656, "1000–3000 Hz_10ms_envelope_correlation": 0.9841735983613714, "1000–3000 Hz_10ms_modulation_mae_db": 0.6562518876745491, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9403444876446301, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07212564496035426, "1000–3000 Hz_40ms_envelope_correlation": 0.9957236165066843, "1000–3000 Hz_40ms_modulation_mae_db": 0.5149649279687152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.490667454259253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1985622062157847, "3000–6000 Hz_10ms_envelope_correlation": 0.9616290823225401, "3000–6000 Hz_10ms_modulation_mae_db": 0.6542960870356597, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9675080617670111, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11798126760314548, "3000–6000 Hz_40ms_envelope_correlation": 0.9847942229239008, "3000–6000 Hz_40ms_modulation_mae_db": 0.9917734723457347, "mr_stft_db": 5.588876522056878, "band_mae_db": 0.1328131472177889, "rms_gain_error_db": -0.0424256117085644, "output_peak": 0.40785333793670386}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8823878930283695, "全帯域_10ms_rms_envelope_nrmse": 0.06688252126942966, "全帯域_10ms_envelope_correlation": 0.9945210005691691, "全帯域_10ms_modulation_mae_db": 0.504382757247135, "全帯域_40ms_log_rms_mae_db": 0.5008860665643261, "全帯域_40ms_rms_envelope_nrmse": 0.0334125598892899, "全帯域_40ms_envelope_correlation": 0.9985469608453309, "全帯域_40ms_modulation_mae_db": 0.5721033409781265, "300–1000 Hz_10ms_log_rms_mae_db": 1.480390020379608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08982714060208132, "300–1000 Hz_10ms_envelope_correlation": 0.9926014919455085, "300–1000 Hz_10ms_modulation_mae_db": 0.7643928073091627, "300–1000 Hz_40ms_log_rms_mae_db": 0.8804657824479947, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04410817243220728, "300–1000 Hz_40ms_envelope_correlation": 0.9980966406088695, "300–1000 Hz_40ms_modulation_mae_db": 0.8938252412532277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5029406260518874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15006162938193054, "1000–3000 Hz_10ms_envelope_correlation": 0.9822027292989449, "1000–3000 Hz_10ms_modulation_mae_db": 0.9820463230920707, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9832658833477563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08253050354037238, "1000–3000 Hz_40ms_envelope_correlation": 0.995060349239633, "1000–3000 Hz_40ms_modulation_mae_db": 0.9977691220564386, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5594496678408658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18947029019672254, "3000–6000 Hz_10ms_envelope_correlation": 0.9679011568650921, "3000–6000 Hz_10ms_modulation_mae_db": 1.4026915641210937, "3000–6000 Hz_40ms_log_rms_mae_db": 0.983647867431, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12429740691025012, "3000–6000 Hz_40ms_envelope_correlation": 0.9872328776716535, "3000–6000 Hz_40ms_modulation_mae_db": 1.8438912805286662, "mr_stft_db": 5.61351929705857, "band_mae_db": 0.21359433393154692, "rms_gain_error_db": -0.07449519872390552, "output_peak": 0.3985573585186337}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9208457037725336, "全帯域_10ms_rms_envelope_nrmse": 0.066881223613285, "全帯域_10ms_envelope_correlation": 0.9946681277665412, "全帯域_10ms_modulation_mae_db": 0.22031704610332778, "全帯域_40ms_log_rms_mae_db": 0.4790877265541371, "全帯域_40ms_rms_envelope_nrmse": 0.036884703673674545, "全帯域_40ms_envelope_correlation": 0.9984107714080404, "全帯域_40ms_modulation_mae_db": 0.2773145270312362, "300–1000 Hz_10ms_log_rms_mae_db": 1.422998653971986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0858008116845285, "300–1000 Hz_10ms_envelope_correlation": 0.9935702387752737, "300–1000 Hz_10ms_modulation_mae_db": 0.5219595872778264, "300–1000 Hz_40ms_log_rms_mae_db": 0.8857269611955622, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05113125710343325, "300–1000 Hz_40ms_envelope_correlation": 0.9978416430720826, "300–1000 Hz_40ms_modulation_mae_db": 0.7662295075323383, "1000–3000 Hz_10ms_log_rms_mae_db": 1.456458082140185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1396578021514976, "1000–3000 Hz_10ms_envelope_correlation": 0.9847439799009378, "1000–3000 Hz_10ms_modulation_mae_db": 0.8837372445457943, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9353062694813131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08254564589296275, "1000–3000 Hz_40ms_envelope_correlation": 0.9950389007584146, "1000–3000 Hz_40ms_modulation_mae_db": 0.8726382849383448, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5002872571165338, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1848237976295748, "3000–6000 Hz_10ms_envelope_correlation": 0.9673831401561178, "3000–6000 Hz_10ms_modulation_mae_db": 0.34306965800357414, "3000–6000 Hz_40ms_log_rms_mae_db": 1.017861531121448, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1309358341520274, "3000–6000 Hz_40ms_envelope_correlation": 0.9808412058623767, "3000–6000 Hz_40ms_modulation_mae_db": 0.5262728497559817, "mr_stft_db": 5.603754065808441, "band_mae_db": 0.15091535472160864, "rms_gain_error_db": -0.12821932856773713, "output_peak": 0.37881243609755316}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8849966051331819, "全帯域_10ms_rms_envelope_nrmse": 0.05423145241430172, "全帯域_10ms_envelope_correlation": 0.9964433394839151, "全帯域_10ms_modulation_mae_db": 0.34049457198100724, "全帯域_40ms_log_rms_mae_db": 0.4718209580404201, "全帯域_40ms_rms_envelope_nrmse": 0.028342461457758206, "全帯域_40ms_envelope_correlation": 0.9990069094294078, "全帯域_40ms_modulation_mae_db": 0.323945572874265, "300–1000 Hz_10ms_log_rms_mae_db": 1.3734589262275292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07278529695752618, "300–1000 Hz_10ms_envelope_correlation": 0.9952020240430838, "300–1000 Hz_10ms_modulation_mae_db": 0.39681904775098076, "300–1000 Hz_40ms_log_rms_mae_db": 0.8587987871308816, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0400152643412455, "300–1000 Hz_40ms_envelope_correlation": 0.9985023981939248, "300–1000 Hz_40ms_modulation_mae_db": 0.3442566179655455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4657177760213658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13711181081096804, "1000–3000 Hz_10ms_envelope_correlation": 0.9848400115309252, "1000–3000 Hz_10ms_modulation_mae_db": 0.31540379955083325, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9340284049968174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07889821643340925, "1000–3000 Hz_40ms_envelope_correlation": 0.9948410681580715, "1000–3000 Hz_40ms_modulation_mae_db": 0.3927792861407338, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5464331454842848, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17496022356003477, "3000–6000 Hz_10ms_envelope_correlation": 0.9713167290811534, "3000–6000 Hz_10ms_modulation_mae_db": 1.0113276455129714, "3000–6000 Hz_40ms_log_rms_mae_db": 0.908960365516174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09916646245434237, "3000–6000 Hz_40ms_envelope_correlation": 0.9910347754228768, "3000–6000 Hz_40ms_modulation_mae_db": 1.3993601479701934, "mr_stft_db": 5.599106236520832, "band_mae_db": 0.15047894565945194, "rms_gain_error_db": -0.07924032177098626, "output_peak": 0.3794917788084986}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 52320, "output_samples": 52320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.589991757621765, "spectral_convergence": 0.2156992025669246, "spectral_convergence_status": "ok", "amplitude_floor": 1.2746778473169441e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.440241940448331, "spectral_convergence": 0.20185477561343537, "spectral_convergence_status": "ok", "amplitude_floor": 1.090624511245042e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.715665741208636, "spectral_convergence": 0.2037716890637414, "spectral_convergence_status": "ok", "amplitude_floor": 7.935262595002381e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004730173956506593, "output_power": 0.0004834381034042205, "delta_db": 0.09463764833989737, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018195471619226037, "output_power": 0.001772648669586175, "delta_db": -0.11340647745879748, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018173989836957952, "output_power": 0.001723114880601815, "delta_db": -0.23136047894929396, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013671881984837986, "output_power": 0.00014162216260389486, "delta_db": 0.15302920965184902, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012324536624073353, "output_power": 0.0001091773704106994, "delta_db": -0.5263796981767813, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.679288082514536e-05, "output_power": 7.896645757798626e-05, "delta_db": -0.4104144761595652, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25487133145603075, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14642176012489885, "pitch": null, "output_rms_dbfs": -23.65626766994194, "output_peak": 0.3778397736978986, "reference_rms_dbfs": -23.509845909817045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5819664797595765}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 52320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00017439460044090568, "log_rms_mae_db": 0.8669629877038829, "rms_envelope_nrmse": 0.06509032650609449, "envelope_correlation": 0.9950318654731803, "rms_gain_error_db": -0.14641939896181516, "reference_mean_power": 0.004463526965610992, "output_mean_power": 0.0043155506977739155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 1.3056377024951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008270090313661458, "output_power": 0.0008225043436396586, "reference_active": true, "delta_db": -0.023720521645199392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010887958720618104, "output_power": 0.00010258090551412738, "reference_active": true, "delta_db": -0.2587993726162183}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8937894537156724e-05, "output_power": 2.5743887678696923e-05, "reference_active": true, "delta_db": -0.5079279762178387}], "active_band_count": 3, "active_band_mae_db": 0.26348262349308543}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015806536675822195, "log_rms_mae_db": 0.4633180987375004, "rms_envelope_nrmse": 0.038093083236616226, "envelope_correlation": 0.9983549860555876, "rms_gain_error_db": -0.14642514088264638, "reference_mean_power": 0.004504823508022685, "output_mean_power": 0.004355472405568722}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.0922276511075418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000711888936919089, "output_power": 0.0007022292834335724, "reference_active": true, "delta_db": -0.05933307974773181}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.550528203435593e-05, "output_power": 6.491417420519707e-05, "reference_active": true, "delta_db": -0.03936784183487202}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5248696894099073e-06, "output_power": 1.2979024652373535e-06, "reference_active": true, "delta_db": -0.6999067448884432}], "active_band_count": 3, "active_band_mae_db": 0.266202555490349}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00016816220894332027, "log_rms_mae_db": 1.3762399787859214, "rms_envelope_nrmse": 0.08833479042888094, "envelope_correlation": 0.9930837387967069, "rms_gain_error_db": -0.17445070360299553, "reference_mean_power": 0.002573247092376177, "output_mean_power": 0.0024719314349098296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 9.221294414644421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007324784888809201, "output_power": 0.0007092432375649383, "reference_active": true, "delta_db": -0.13999671424667365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.576574656453582e-05, "output_power": 7.505423687082165e-05, "reference_active": true, "delta_db": -0.5793865883093879}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.804896765293062e-05, "output_power": 1.314986282932173e-05, "reference_active": true, "delta_db": -1.3753114402775612}], "active_band_count": 3, "active_band_mae_db": 0.6982315809445409}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.0001473388036867024, "log_rms_mae_db": 0.7995676776238065, "rms_envelope_nrmse": 0.05286252788322125, "envelope_correlation": 0.9975545442185617, "rms_gain_error_db": -0.17445080494412593, "reference_mean_power": 0.002597106693874505, "output_mean_power": 0.0024948515615223956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.911179205548874e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006603877113347142, "output_power": 0.0006359325522017155, "reference_active": true, "delta_db": -0.16387926924495197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.506710493994746e-05, "output_power": 4.180245832438225e-05, "reference_active": true, "delta_db": -0.3265783780657269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1610111255482175e-06, "output_power": 8.758184021951335e-07, "reference_active": true, "delta_db": -1.2242231534577366}], "active_band_count": 3, "active_band_mae_db": 0.5715602669228051}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 5.380259383983556e-05, "log_rms_mae_db": 1.5069351665351136, "rms_envelope_nrmse": 0.14774899721287515, "envelope_correlation": 0.9825411092961067, "rms_gain_error_db": -0.24379803979412976, "reference_mean_power": 0.00019859577428883629, "output_mean_power": 0.00018775443283528155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 8.760825576371796e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.452705657171872e-05, "output_power": 4.837659381391692e-05, "reference_active": true, "delta_db": -0.519767681267005}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.928893294666125e-06, "output_power": 4.98104018187799e-06, "reference_active": true, "delta_db": -0.7565358900973264}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9788817769833428e-06, "output_power": 2.8246406334908526e-06, "reference_active": true, "delta_db": 1.545433530925421}], "active_band_count": 3, "active_band_mae_db": 0.9405790340965842}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 5.051409218298903e-05, "log_rms_mae_db": 1.0655624723221209, "rms_envelope_nrmse": 0.09736356818531933, "envelope_correlation": 0.9922548169934553, "rms_gain_error_db": -0.24381621701752165, "reference_mean_power": 0.00020041602526910178, "output_mean_power": 0.00018947452329186069}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.85711379260727e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.022793901144941e-05, "output_power": 4.473226253871803e-05, "reference_active": true, "delta_db": -0.5032449308733865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.619667417561746e-06, "output_power": 2.1842878793938654e-06, "reference_active": true, "delta_db": -0.7893628265950945}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1855266901360492e-07, "output_power": 1.550325566737473e-07, "reference_active": true, "delta_db": 1.1651157329905484}], "active_band_count": 3, "active_band_mae_db": 0.8192411634863431}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 6.112392927022402e-05, "log_rms_mae_db": 1.4675942366345256, "rms_envelope_nrmse": 0.1830712584517919, "envelope_correlation": 0.967852563868846, "rms_gain_error_db": -0.3428482689794163, "reference_mean_power": 8.711996692518435e-05, "output_mean_power": 8.050685754147193e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 4.203401537603793e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8248175978150172e-05, "output_power": 1.6595502426000655e-05, "reference_active": true, "delta_db": -0.41229055317113683}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.243431842943793e-05, "output_power": 1.122343459431197e-05, "reference_active": true, "delta_db": -0.4449620497333909}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.767669752037568e-06, "output_power": 4.327338323250497e-06, "reference_active": true, "delta_db": -1.9421830702929248}], "active_band_count": 3, "active_band_mae_db": 0.9331452243991508}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 3.626522691368819e-05, "log_rms_mae_db": 0.8868725394611736, "rms_envelope_nrmse": 0.09823302236963541, "envelope_correlation": 0.9904759978613888, "rms_gain_error_db": -0.34282950015298386, "reference_mean_power": 8.789733624317315e-05, "output_mean_power": 8.122556927771756e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 3.223393789064049e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.051966991658498e-05, "output_power": 1.8069430669036413e-05, "reference_active": true, "delta_db": -0.5522590130839482}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.880534452438109e-06, "output_power": 5.0083300956880055e-06, "reference_active": true, "delta_db": -0.6972385352400454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.871896444237529e-07, "output_power": 3.137256545716421e-07, "reference_active": true, "delta_db": -2.7222835372700067}], "active_band_count": 3, "active_band_mae_db": 1.3239270285313334}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8959562918995876, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "105d6da3f2b43dc0c8966fe4ac49c0623a2d67200315f1d2da324cc6c86705b4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9829864116589268, "standard_deviation": 0.0015311584612725676, "minimum": 0.9806307102701661, "maximum": 0.9848400115309252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.485616973081346, "standard_deviation": 0.029141166647908737, "minimum": 1.4348022406503944, "maximum": 1.5142964311113745, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7720151716401848, "standard_deviation": 0.2158349970273104, "minimum": 0.31540379955083325, "maximum": 0.9820463230920707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1463973734775765, "standard_deviation": 0.00791150332737157, "minimum": 0.13711181081096804, "maximum": 0.16141299925462294, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942860278943619, "standard_deviation": 0.0012580692113999673, "minimum": 0.9922548169934553, "maximum": 0.9957236165066843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9690518511351085, "standard_deviation": 0.04346734855287583, "minimum": 0.9340284049968174, "maximum": 1.0655624723221209, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8283823276272062, "standard_deviation": 0.2557629814147684, "minimum": 0.3927792861407338, "maximum": 1.16819269852178, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0856081944079516, "standard_deviation": 0.01126414353882725, "minimum": 0.07212564496035426, "maximum": 0.10539955685912301, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9659387477213693, "standard_deviation": 0.003408015457572675, "minimum": 0.9616290823225401, "maximum": 0.9713167290811534, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5295587182750636, "standard_deviation": 0.03852837418097749, "minimum": 1.4675942366345256, "maximum": 1.5786242233728252, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9790497964796498, "standard_deviation": 0.45534693010721194, "minimum": 0.34306965800357414, "maximum": 1.7450410488523715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1901834401576885, "standard_deviation": 0.008932182059611021, "minimum": 0.17496022356003477, "maximum": 0.20215169725991536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9854220837847, "standard_deviation": 0.004950828109407333, "minimum": 0.9760595558720987, "maximum": 0.9910347754228768, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9696854644232851, "standard_deviation": 0.0519161209563306, "minimum": 0.8868725394611736, "maximum": 1.0374462095343862, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2958942739242234, "standard_deviation": 0.51742620985112, "minimum": 0.5262728497559817, "maximum": 2.1385167425176466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11998066085548723, "standard_deviation": 0.016694694925199115, "minimum": 0.09823302236963541, "maximum": 0.1489667067264915, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9939160309962898, "standard_deviation": 0.0008607736865483704, "minimum": 0.9926014919455085, "maximum": 0.9952020240430838, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.400574389209124, "standard_deviation": 0.041618555630688815, "minimum": 1.3579947043843859, "maximum": 1.480390020379608, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5277059523975139, "standard_deviation": 0.16429935470013696, "minimum": 0.3145400689475169, "maximum": 0.7643928073091627, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08215683180464875, "standard_deviation": 0.005986239493958608, "minimum": 0.07278529695752618, "maximum": 0.08982714060208132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980419081852534, "standard_deviation": 0.00027576003775407407, "minimum": 0.9975545442185617, "maximum": 0.9985023981939248, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8548740096043643, "standard_deviation": 0.0382274118905746, "minimum": 0.7955433779904671, "maximum": 0.8991936009490951, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5651792833729119, "standard_deviation": 0.20898092891829248, "minimum": 0.2886117159606371, "maximum": 0.8938252412532277, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04635325063760715, "standard_deviation": 0.0042871539163478216, "minimum": 0.0400152643412455, "maximum": 0.05286252788322125, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18952130840764173, "standard_deviation": 0.04406163925195494, "minimum": 0.1328131472177889, "maximum": 0.25487133145603075, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.593359439924697, "standard_deviation": 0.014239751428699236, "minimum": 5.566839477191699, "maximum": 5.61351929705857, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3857721650431389, "standard_deviation": 0.011775894402316342, "minimum": 0.37414928795224023, "maximum": 0.40785333793670386, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09752116558172622, "standard_deviation": 0.03389291401598609, "minimum": -0.14642176012489885, "maximum": -0.0424256117085644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.99522260449785, "standard_deviation": 0.0005913515397337485, "minimum": 0.9945210005691691, "maximum": 0.9964433394839151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8859975871227982, "standard_deviation": 0.03888582777303019, "minimum": 0.8306111273417385, "maximum": 0.9453583295246496, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32647095165952594, "standard_deviation": 0.08743378512397014, "minimum": 0.22031704610332778, "maximum": 0.504382757247135, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06289099078910426, "standard_deviation": 0.004101705501024759, "minimum": 0.05423145241430172, "maximum": 0.06688252126942966, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985346287499222, "standard_deviation": 0.00023278838248038351, "minimum": 0.9983193428779612, "maximum": 0.9990069094294078, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.471245973108949, "standard_deviation": 0.037665482769066796, "minimum": 0.3973161111786982, "maximum": 0.5283667464449413, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31519180501550936, "standard_deviation": 0.12118827882772151, "minimum": 0.1595800779988158, "maximum": 0.5721033409781265, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.034405362780594806, "standard_deviation": 0.003383767350478629, "minimum": 0.028342461457758206, "maximum": 0.038093083236616226, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1284-1180-0028", "language": "en", "style": "normal", "speaker": "librispeech_1284", "source_utterance_id": "1284-1180-0028", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1284-1180-0028.wav", "source_file_sha256": "f3781b83f2a93643f38b979ceefa38c09fbc85a1c69befb14dbadb12d1ad4097", "decoded_pcm_sha256": "0bfea5c92c4d2ee444f3984a4d2b53407bcff068a4ef104efccbb0731c86faf4", "original_source_sha256": "1ee65c4ba89a7fe76bef27dade9b0ab41fe2109b77c8e68d5b80e34194031ee4", "corpus_entry_id": "librispeech_1284-1180-0028"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.210789203643799, "predicted_mos": 4.210789203643799, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "abcca764b7cf0ddd8946994bf5b3236d18ad615ea60ec5a099546dcdb7026454", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.305339930000628, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1284-1180-0028", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "abcca764b7cf0ddd8946994bf5b3236d18ad615ea60ec5a099546dcdb7026454", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "de015d84918b419eb8e95556ce706a15f950525e9ec55d79e15a4b5d0839f33e"}, "rows": [{"case": "librispeech_1284-1180-0028", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.955986499786377, "predicted_mos": 2.955986499786377, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "93dcbaac12447c7ced6a0210d35b66842120859f10b34c77f7950e622e03044c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.280375778002053, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1284-1180-0028", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "93dcbaac12447c7ced6a0210d35b66842120859f10b34c77f7950e622e03044c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9079802820763336, "全帯域_10ms_rms_envelope_nrmse": 0.07893064188008075, "全帯域_10ms_envelope_correlation": 0.9936366468597547, "全帯域_10ms_modulation_mae_db": 0.2384599554341229, "全帯域_40ms_log_rms_mae_db": 0.5103019658320179, "全帯域_40ms_rms_envelope_nrmse": 0.045493480827039044, "全帯域_40ms_envelope_correlation": 0.9977459813016465, "全帯域_40ms_modulation_mae_db": 0.3077102429894161, "300–1000 Hz_10ms_log_rms_mae_db": 1.2310699735357535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.098931414451227, "300–1000 Hz_10ms_envelope_correlation": 0.9922430226325406, "300–1000 Hz_10ms_modulation_mae_db": 0.20234100775529404, "300–1000 Hz_40ms_log_rms_mae_db": 0.7215297072788004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05860746320710548, "300–1000 Hz_40ms_envelope_correlation": 0.9971445626378784, "300–1000 Hz_40ms_modulation_mae_db": 0.42822672045918736, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6379091083625197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1770776737810438, "1000–3000 Hz_10ms_envelope_correlation": 0.9768530739430745, "1000–3000 Hz_10ms_modulation_mae_db": 0.2976945702760037, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1012093805599696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11874841792280992, "1000–3000 Hz_40ms_envelope_correlation": 0.9891873769351622, "1000–3000 Hz_40ms_modulation_mae_db": 0.23398221219778415, "3000–6000 Hz_10ms_log_rms_mae_db": 1.904406262100707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22215261249603457, "3000–6000 Hz_10ms_envelope_correlation": 0.9558195925148688, "3000–6000 Hz_10ms_modulation_mae_db": 1.4148671329817128, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3014645737288908, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14296850341641093, "3000–6000 Hz_40ms_envelope_correlation": 0.9796055024444832, "3000–6000 Hz_40ms_modulation_mae_db": 1.752383773794173, "mr_stft_db": 5.603993916517152, "band_mae_db": 0.09584854334599287, "rms_gain_error_db": -0.08903258612703761, "output_peak": 0.550219094502173}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8951297031103593, "全帯域_10ms_rms_envelope_nrmse": 0.07098197029358762, "全帯域_10ms_envelope_correlation": 0.9948614605720836, "全帯域_10ms_modulation_mae_db": 0.4296778065756466, "全帯域_40ms_log_rms_mae_db": 0.48655488914854506, "全帯域_40ms_rms_envelope_nrmse": 0.040008622894996536, "全帯域_40ms_envelope_correlation": 0.9982639836193758, "全帯域_40ms_modulation_mae_db": 0.6143031502075325, "300–1000 Hz_10ms_log_rms_mae_db": 1.2746897024310968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09367936488902551, "300–1000 Hz_10ms_envelope_correlation": 0.9930642978006432, "300–1000 Hz_10ms_modulation_mae_db": 0.5160074404520668, "300–1000 Hz_40ms_log_rms_mae_db": 0.7482875438972371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05250518274823432, "300–1000 Hz_40ms_envelope_correlation": 0.997739565463874, "300–1000 Hz_40ms_modulation_mae_db": 0.8316289907642039, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6068806375828866, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16768955934284413, "1000–3000 Hz_10ms_envelope_correlation": 0.9799408859046425, "1000–3000 Hz_10ms_modulation_mae_db": 0.3705719924124986, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0745948587906236, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1117439234875681, "1000–3000 Hz_40ms_envelope_correlation": 0.9905361283069904, "1000–3000 Hz_40ms_modulation_mae_db": 0.4865472176421079, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9358133766097427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25602591454214646, "3000–6000 Hz_10ms_envelope_correlation": 0.940904108188737, "3000–6000 Hz_10ms_modulation_mae_db": 1.7235549385451432, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3926081044573233, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1804284349338227, "3000–6000 Hz_40ms_envelope_correlation": 0.9686420228987335, "3000–6000 Hz_40ms_modulation_mae_db": 1.7830472798581025, "mr_stft_db": 5.59173459488638, "band_mae_db": 0.16187522899564913, "rms_gain_error_db": -0.07955792479356072, "output_peak": 0.5382310483250473}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8953803001629582, "全帯域_10ms_rms_envelope_nrmse": 0.08532788787200611, "全帯域_10ms_envelope_correlation": 0.9925641656600493, "全帯域_10ms_modulation_mae_db": 0.29244008539384697, "全帯域_40ms_log_rms_mae_db": 0.47383430881791505, "全帯域_40ms_rms_envelope_nrmse": 0.04117048314306274, "全帯域_40ms_envelope_correlation": 0.998212523038257, "全帯域_40ms_modulation_mae_db": 0.5951788773900334, "300–1000 Hz_10ms_log_rms_mae_db": 1.2152443108799167, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10608974820794566, "300–1000 Hz_10ms_envelope_correlation": 0.9911982764952005, "300–1000 Hz_10ms_modulation_mae_db": 0.297870683318386, "300–1000 Hz_40ms_log_rms_mae_db": 0.7249280497787798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054546111438193226, "300–1000 Hz_40ms_envelope_correlation": 0.997769140402775, "300–1000 Hz_40ms_modulation_mae_db": 0.5276926617459269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.692336725351373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17142445441957352, "1000–3000 Hz_10ms_envelope_correlation": 0.9804081473582, "1000–3000 Hz_10ms_modulation_mae_db": 0.589158315463616, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1774393283414184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11907458849318808, "1000–3000 Hz_40ms_envelope_correlation": 0.9904510803466199, "1000–3000 Hz_40ms_modulation_mae_db": 0.6565010419228369, "3000–6000 Hz_10ms_log_rms_mae_db": 1.932254374649648, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23536453166725527, "3000–6000 Hz_10ms_envelope_correlation": 0.9504837959511544, "3000–6000 Hz_10ms_modulation_mae_db": 1.0184293840513892, "3000–6000 Hz_40ms_log_rms_mae_db": 1.331860914866077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15332389091138732, "3000–6000 Hz_40ms_envelope_correlation": 0.9764661452584966, "3000–6000 Hz_40ms_modulation_mae_db": 1.0465253919788182, "mr_stft_db": 5.616047572312406, "band_mae_db": 0.17974739476384413, "rms_gain_error_db": -0.10136195470570786, "output_peak": 0.606021723652486}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.867356225985997, "全帯域_10ms_rms_envelope_nrmse": 0.07105303603824249, "全帯域_10ms_envelope_correlation": 0.9948920480395307, "全帯域_10ms_modulation_mae_db": 0.14843232259764674, "全帯域_40ms_log_rms_mae_db": 0.4821447884925486, "全帯域_40ms_rms_envelope_nrmse": 0.0364453313434501, "全帯域_40ms_envelope_correlation": 0.9986524092692688, "全帯域_40ms_modulation_mae_db": 0.2956290816292162, "300–1000 Hz_10ms_log_rms_mae_db": 1.2004516315928944, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08469187786241733, "300–1000 Hz_10ms_envelope_correlation": 0.9944635275733179, "300–1000 Hz_10ms_modulation_mae_db": 0.24563833742914074, "300–1000 Hz_40ms_log_rms_mae_db": 0.674100219704222, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045255689827654544, "300–1000 Hz_40ms_envelope_correlation": 0.9985088401220864, "300–1000 Hz_40ms_modulation_mae_db": 0.44874378034289836, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6490966060575618, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17663308019102955, "1000–3000 Hz_10ms_envelope_correlation": 0.976913273959821, "1000–3000 Hz_10ms_modulation_mae_db": 0.7474832438549731, "1000–3000 Hz_40ms_log_rms_mae_db": 1.115928331446235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11682883445589791, "1000–3000 Hz_40ms_envelope_correlation": 0.9891644418856061, "1000–3000 Hz_40ms_modulation_mae_db": 0.8355281110168483, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8993497044266208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22863232021516788, "3000–6000 Hz_10ms_envelope_correlation": 0.9544706364350093, "3000–6000 Hz_10ms_modulation_mae_db": 0.7346331909324443, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2986833121833716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15141088260662036, "3000–6000 Hz_40ms_envelope_correlation": 0.9776434787346637, "3000–6000 Hz_40ms_modulation_mae_db": 0.7959823492522974, "mr_stft_db": 5.5829220401023605, "band_mae_db": 0.0748474228093901, "rms_gain_error_db": -0.09262431067549955, "output_peak": 0.6252176880399438}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8833107639769809, "全帯域_10ms_rms_envelope_nrmse": 0.07556808721104988, "全帯域_10ms_envelope_correlation": 0.9941381330905403, "全帯域_10ms_modulation_mae_db": 0.16789003513877368, "全帯域_40ms_log_rms_mae_db": 0.5006259090755741, "全帯域_40ms_rms_envelope_nrmse": 0.04162667125552693, "全帯域_40ms_envelope_correlation": 0.9980751432991617, "全帯域_40ms_modulation_mae_db": 0.13289841193211882, "300–1000 Hz_10ms_log_rms_mae_db": 1.350112634914224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09345142701174508, "300–1000 Hz_10ms_envelope_correlation": 0.9930856377180804, "300–1000 Hz_10ms_modulation_mae_db": 0.35406299282317716, "300–1000 Hz_40ms_log_rms_mae_db": 0.7642517963588156, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05036550331504141, "300–1000 Hz_40ms_envelope_correlation": 0.997926766811481, "300–1000 Hz_40ms_modulation_mae_db": 0.12254172755380953, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6675028615925211, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17030934953711035, "1000–3000 Hz_10ms_envelope_correlation": 0.9789206278333777, "1000–3000 Hz_10ms_modulation_mae_db": 0.26480525687488693, "1000–3000 Hz_40ms_log_rms_mae_db": 1.131152091393746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09981380097383596, "1000–3000 Hz_40ms_envelope_correlation": 0.9922952436288747, "1000–3000 Hz_40ms_modulation_mae_db": 0.369803634574324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9129921912258274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2252680622619496, "3000–6000 Hz_10ms_envelope_correlation": 0.9548154598892734, "3000–6000 Hz_10ms_modulation_mae_db": 0.6241226605122385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2854543854177745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15066293491346594, "3000–6000 Hz_40ms_envelope_correlation": 0.9773088630192949, "3000–6000 Hz_40ms_modulation_mae_db": 0.8791622030285083, "mr_stft_db": 5.599990986221113, "band_mae_db": 0.08254494533614358, "rms_gain_error_db": -0.05072620014152866, "output_peak": 0.5193671434739701}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8843708755096117, "全帯域_10ms_rms_envelope_nrmse": 0.0751884263049923, "全帯域_10ms_envelope_correlation": 0.9942913654447668, "全帯域_10ms_modulation_mae_db": 0.5188266123015994, "全帯域_40ms_log_rms_mae_db": 0.5022010640311861, "全帯域_40ms_rms_envelope_nrmse": 0.042297190285438144, "全帯域_40ms_envelope_correlation": 0.9981389096535984, "全帯域_40ms_modulation_mae_db": 0.7886829935098123, "300–1000 Hz_10ms_log_rms_mae_db": 1.2750729610766158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09235184019556317, "300–1000 Hz_10ms_envelope_correlation": 0.993392646488609, "300–1000 Hz_10ms_modulation_mae_db": 0.397161838859704, "300–1000 Hz_40ms_log_rms_mae_db": 0.7334059796412693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05048536187553163, "300–1000 Hz_40ms_envelope_correlation": 0.9981009283136973, "300–1000 Hz_40ms_modulation_mae_db": 0.7798406630113225, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6719712913231748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17884460524362422, "1000–3000 Hz_10ms_envelope_correlation": 0.976277760709693, "1000–3000 Hz_10ms_modulation_mae_db": 0.681630571333197, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1375612197394769, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10359705455647558, "1000–3000 Hz_40ms_envelope_correlation": 0.991474013846342, "1000–3000 Hz_40ms_modulation_mae_db": 0.8020967142756362, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8719497357367638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23059735279518104, "3000–6000 Hz_10ms_envelope_correlation": 0.9526421291350621, "3000–6000 Hz_10ms_modulation_mae_db": 1.040131024448, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2933629660189794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14899331759002185, "3000–6000 Hz_40ms_envelope_correlation": 0.9778086415450684, "3000–6000 Hz_40ms_modulation_mae_db": 1.2509455723348297, "mr_stft_db": 5.602731334255185, "band_mae_db": 0.11965434808062493, "rms_gain_error_db": -0.11177797867312551, "output_peak": 0.5359357665818084}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8896994341478951, "全帯域_10ms_rms_envelope_nrmse": 0.07700966476342518, "全帯域_10ms_envelope_correlation": 0.9940109044922371, "全帯域_10ms_modulation_mae_db": 0.3930686093128923, "全帯域_40ms_log_rms_mae_db": 0.4739570967239683, "全帯域_40ms_rms_envelope_nrmse": 0.04004968268183171, "全帯域_40ms_envelope_correlation": 0.9983683512501892, "全帯域_40ms_modulation_mae_db": 0.43810916087184326, "300–1000 Hz_10ms_log_rms_mae_db": 1.2732524692138298, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10257251964141122, "300–1000 Hz_10ms_envelope_correlation": 0.991788963708584, "300–1000 Hz_10ms_modulation_mae_db": 0.43572184182024687, "300–1000 Hz_40ms_log_rms_mae_db": 0.7765564142566932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05451162321134194, "300–1000 Hz_40ms_envelope_correlation": 0.9977542236088847, "300–1000 Hz_40ms_modulation_mae_db": 0.7012351303435066, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6596563097321053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17054216251841375, "1000–3000 Hz_10ms_envelope_correlation": 0.9785090198083013, "1000–3000 Hz_10ms_modulation_mae_db": 0.5186919647300469, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1365105593084661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09962180403974609, "1000–3000 Hz_40ms_envelope_correlation": 0.9923309416456865, "1000–3000 Hz_40ms_modulation_mae_db": 0.7710230807087927, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8550631541165132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2250020298683804, "3000–6000 Hz_10ms_envelope_correlation": 0.9545157716914107, "3000–6000 Hz_10ms_modulation_mae_db": 1.095464375566838, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2487884109221699, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14814749887585696, "3000–6000 Hz_40ms_envelope_correlation": 0.9782913537231818, "3000–6000 Hz_40ms_modulation_mae_db": 1.519977183974781, "mr_stft_db": 5.582484366545448, "band_mae_db": 0.1276101702537922, "rms_gain_error_db": -0.11667127155474219, "output_peak": 0.520843778066266}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.851396603956737, "全帯域_10ms_rms_envelope_nrmse": 0.06893038539946149, "全帯域_10ms_envelope_correlation": 0.995169773300391, "全帯域_10ms_modulation_mae_db": 0.3423128392259021, "全帯域_40ms_log_rms_mae_db": 0.45444054726509936, "全帯域_40ms_rms_envelope_nrmse": 0.03829904159181624, "全帯域_40ms_envelope_correlation": 0.9984290312369719, "全帯域_40ms_modulation_mae_db": 0.47233496763125515, "300–1000 Hz_10ms_log_rms_mae_db": 1.1702570897473337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08865331745483498, "300–1000 Hz_10ms_envelope_correlation": 0.9939261303267695, "300–1000 Hz_10ms_modulation_mae_db": 0.3600817506576531, "300–1000 Hz_40ms_log_rms_mae_db": 0.6861639681748978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05367410802919072, "300–1000 Hz_40ms_envelope_correlation": 0.9977862692023006, "300–1000 Hz_40ms_modulation_mae_db": 0.6907517549137833, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6237059364242956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17099586264874464, "1000–3000 Hz_10ms_envelope_correlation": 0.9788578430754542, "1000–3000 Hz_10ms_modulation_mae_db": 0.2577652402787465, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0942966742417444, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10502299079519323, "1000–3000 Hz_40ms_envelope_correlation": 0.9914696396993428, "1000–3000 Hz_40ms_modulation_mae_db": 0.4449504042220271, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8645700308389737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21853307942446223, "3000–6000 Hz_10ms_envelope_correlation": 0.9576056226606888, "3000–6000 Hz_10ms_modulation_mae_db": 1.0327521803249433, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2882893557795099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14085644262445826, "3000–6000 Hz_40ms_envelope_correlation": 0.9802332026047929, "3000–6000 Hz_40ms_modulation_mae_db": 1.207443106746293, "mr_stft_db": 5.590977292392353, "band_mae_db": 0.1088498359609739, "rms_gain_error_db": -0.08064602831843361, "output_peak": 0.5223744761749849}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.590716957528458, "spectral_convergence": 0.21253021225400695, "spectral_convergence_status": "ok", "amplitude_floor": 1.8586249352090494e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5049101828936235, "spectral_convergence": 0.20727142466723505, "spectral_convergence_status": "ok", "amplitude_floor": 1.6686015821705495e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.716354609129377, "spectral_convergence": 0.20805405427340526, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267872572435527e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003360135402914431, "output_power": 0.00034196797508655944, "delta_db": 0.07628658326714706, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011242144422279699, "output_power": 0.0011146637575969835, "delta_db": -0.03705279540391973, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015417458345753522, "output_power": 0.00148443461317456, "delta_db": -0.16451711322698637, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.127065821964785e-05, "output_power": 7.985153756183747e-05, "delta_db": -0.07650494157312611, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.295885949753132e-05, "output_power": 5.099519335754179e-05, "delta_db": -0.1640938089671226, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.948150516097124e-05, "output_power": 1.973722580128926e-05, "delta_db": 0.05663601763765533, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09584854334599287, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08903258612703761, "pitch": null, "output_rms_dbfs": -25.098096351447783, "output_peak": 0.550219094502173, "reference_rms_dbfs": -25.009063765320747, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.603993916517152}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025375764068356454, "log_rms_mae_db": 0.9079802820763336, "rms_envelope_nrmse": 0.07893064188008075, "envelope_correlation": 0.9936366468597547, "rms_gain_error_db": -0.08903044250111766, "reference_mean_power": 0.0031583120304531104, "output_mean_power": 0.00309422570989803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1807904078754825e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008327873716353349, "output_power": 0.0008255409183695526, "reference_active": true, "delta_db": -0.03795526756321406}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013733244323789595, "output_power": 0.00012416806556081334, "reference_active": true, "delta_db": -0.4376323130899051}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.747498790065721e-05, "output_power": 1.653627382395719e-05, "reference_active": true, "delta_db": -0.23979228564924954}], "active_band_count": 3, "active_band_mae_db": 0.2384599554341229}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002402786371599188, "log_rms_mae_db": 0.5103019658320179, "rms_envelope_nrmse": 0.045493480827039044, "envelope_correlation": 0.9977459813016465, "rms_gain_error_db": -0.08902741110077833, "reference_mean_power": 0.0031741738524038612, "output_mean_power": 0.003109767845176354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0384864838590044e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000782161356346486, "output_power": 0.000775288984413327, "reference_active": true, "delta_db": -0.03832741887670628}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.85949543312588e-05, "output_power": 5.308804756259404e-05, "reference_active": true, "delta_db": -0.42863466669050887}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3154862476080315e-06, "output_power": 1.1843210484688414e-06, "reference_active": true, "delta_db": -0.4561686434010332}], "active_band_count": 3, "active_band_mae_db": 0.3077102429894161}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002501327279402343, "log_rms_mae_db": 1.2310699735357535, "rms_envelope_nrmse": 0.098931414451227, "envelope_correlation": 0.9922430226325406, "rms_gain_error_db": -0.12092190349462033, "reference_mean_power": 0.0021283926153372414, "output_mean_power": 0.002069948560378179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0698576130619868e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007463968445819582, "output_power": 0.0007358290535451529, "reference_active": true, "delta_db": -0.061928631888735194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014378870760237378, "output_power": 0.00012809763346142004, "reference_active": true, "delta_db": -0.5018367376622895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4173849116031066e-05, "output_power": 1.4033372069994232e-05, "reference_active": true, "delta_db": -0.04325765371485732}], "active_band_count": 3, "active_band_mae_db": 0.20234100775529404}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00023719519351520602, "log_rms_mae_db": 0.7215297072788004, "rms_envelope_nrmse": 0.05860746320710548, "envelope_correlation": 0.9971445626378784, "rms_gain_error_db": -0.12092189662946665, "reference_mean_power": 0.0021390960955203144, "output_mean_power": 0.002080358134361688}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.624199297765067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007211274534636877, "output_power": 0.000708178340254448, "reference_active": true, "delta_db": -0.07869389891407258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.136224424017926e-05, "output_power": 5.432212961911718e-05, "reference_active": true, "delta_db": -0.5292444747626238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3649293472151981e-06, "output_power": 1.1679815777400235e-06, "reference_active": true, "delta_db": -0.6767417877008657}], "active_band_count": 3, "active_band_mae_db": 0.42822672045918736}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.486004098842285e-05, "log_rms_mae_db": 1.6379091083625197, "rms_envelope_nrmse": 0.1770776737810438, "envelope_correlation": 0.9768530739430745, "rms_gain_error_db": -0.20542071378891003, "reference_mean_power": 9.73047769043704e-05, "output_mean_power": 9.280942668109588e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.517204404381221e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1360183676562794e-05, "output_power": 3.848068493702395e-05, "reference_active": true, "delta_db": -0.31339665089804764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.821587959830417e-06, "output_power": 4.536867384927741e-06, "reference_active": true, "delta_db": -0.2643400917763266}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2148111012630244e-06, "output_power": 2.059690637317448e-06, "reference_active": true, "delta_db": -0.3153469681536369}], "active_band_count": 3, "active_band_mae_db": 0.2976945702760037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5230962699738786e-05, "log_rms_mae_db": 1.1012093805599696, "rms_envelope_nrmse": 0.11874841792280992, "envelope_correlation": 0.9891873769351622, "rms_gain_error_db": -0.20543739316354445, "reference_mean_power": 9.77871242521371e-05, "output_mean_power": 9.32691320210182e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.922472441667658e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.88087102342988e-05, "output_power": 3.608371313410002e-05, "reference_active": true, "delta_db": -0.3161798815416985}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.146701619692133e-06, "output_power": 2.0281632566744733e-06, "reference_active": true, "delta_db": -0.24668773434714314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.612101873452723e-07, "output_power": 1.5612936041023493e-07, "reference_active": true, "delta_db": -0.13907902070451073}], "active_band_count": 3, "active_band_mae_db": 0.23398221219778415}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.5953225979159985e-05, "log_rms_mae_db": 1.904406262100707, "rms_envelope_nrmse": 0.22215261249603457, "envelope_correlation": 0.9558195925148688, "rms_gain_error_db": -0.11145637926665385, "reference_mean_power": 3.136903007980474e-05, "output_mean_power": 3.0574224679537015e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9944851157580815e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0899878382466362e-05, "output_power": 1.0553186232751883e-05, "reference_active": true, "delta_db": -0.14038050012021674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5519027988888207e-06, "output_power": 2.0908529827863075e-06, "reference_active": true, "delta_db": -0.865406315287191}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4088478818809383e-06, "output_power": 6.683172102105848e-07, "reference_active": true, "delta_db": -3.2388145835377307}], "active_band_count": 3, "active_band_mae_db": 1.4148671329817128}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9738091980327894e-05, "log_rms_mae_db": 1.3014645737288908, "rms_envelope_nrmse": 0.14296850341641093, "envelope_correlation": 0.9796055024444832, "rms_gain_error_db": -0.1113204308182824, "reference_mean_power": 3.151908471872331e-05, "output_mean_power": 3.072143901140976e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7532611532165288e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0780259770731776e-05, "output_power": 1.0176006768589489e-05, "reference_active": true, "delta_db": -0.2505183894688183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1761748042676738e-06, "output_power": 9.169547816368621e-07, "reference_active": true, "delta_db": -1.0812395225088713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.367404555815231e-08, "output_power": 3.793848077599021e-08, "reference_active": true, "delta_db": -3.9253934094048293}], "active_band_count": 3, "active_band_mae_db": 1.752383773794173}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9017714806314916, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8d420eb2c0e6dc708e86da227a6ee4269f63be097c4e1935ddde6bdca43a7eb8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9783350790740706, "standard_deviation": 0.0015105845822575058, "minimum": 0.976277760709693, "maximum": 0.9804081473582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.651132434553305, "standard_deviation": 0.027677051218604108, "minimum": 1.6068806375828866, "maximum": 1.692336725351373, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46597514440299603, "standard_deviation": 0.19455004162028755, "minimum": 0.2577652402787465, "maximum": 0.7474832438549731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.172939593460298, "standard_deviation": 0.003998877866955104, "minimum": 0.16768955934284413, "maximum": 0.17884460524362422, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9908636082868281, "standard_deviation": 0.0012488198086108263, "minimum": 0.9891644418856061, "maximum": 0.9923309416456865, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.12108655547771, "standard_deviation": 0.03181901820063269, "minimum": 1.0745948587906236, "maximum": 1.1774393283414184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5750540520700447, "standard_deviation": 0.22285508774831117, "minimum": 0.23398221219778415, "maximum": 0.8355281110168483, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10930642684058936, "standard_deviation": 0.008297528072326222, "minimum": 0.09962180403974609, "maximum": 0.11907458849318808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9526571395582756, "standard_deviation": 0.005192681922224157, "minimum": 0.940904108188737, "maximum": 0.9576056226606888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8970498537130995, "standard_deviation": 0.030480370601760093, "minimum": 1.8550631541165132, "maximum": 1.9358133766097427, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0854943609203387, "standard_deviation": 0.3507159641283554, "minimum": 0.6241226605122385, "maximum": 1.7235549385451432, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23019698790882218, "standard_deviation": 0.011640361252043306, "minimum": 0.21853307942446223, "maximum": 0.25602591454214646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9769999012785894, "standard_deviation": 0.0035899458996647765, "minimum": 0.9686420228987335, "maximum": 0.9802332026047929, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3050640029217622, "standard_deviation": 0.04209766826447095, "minimum": 1.2487884109221699, "maximum": 1.3926081044573233, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2794333576209755, "standard_deviation": 0.37594691048242235, "minimum": 0.7959823492522974, "maximum": 1.7830472798581025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15209898823400553, "standard_deviation": 0.0121984408749892, "minimum": 0.14085644262445826, "maximum": 0.1804284349338227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9928953128429681, "standard_deviation": 0.0010930576802128669, "minimum": 0.9911982764952005, "maximum": 0.9944635275733179, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2487688466739582, "standard_deviation": 0.05627686944881991, "minimum": 1.1702570897473337, "maximum": 1.350112634914224, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3511107366394586, "standard_deviation": 0.10187613170487624, "minimum": 0.20234100775529404, "maximum": 0.5160074404520668, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09505268871427124, "standard_deviation": 0.0071058531899128976, "minimum": 0.08469187786241733, "maximum": 0.10608974820794566, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978412870703721, "standard_deviation": 0.0003839515988970963, "minimum": 0.9971445626378784, "maximum": 0.9985088401220864, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7286529598863394, "standard_deviation": 0.035488544924345854, "minimum": 0.674100219704222, "maximum": 0.7765564142566932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5663326786418298, "standard_deviation": 0.23329638858657656, "minimum": 0.12254172755380953, "maximum": 0.8316289907642039, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05249388045653666, "standard_deviation": 0.003928057252755256, "minimum": 0.045255689827654544, "maximum": 0.05860746320710548, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11887223619330135, "standard_deviation": 0.03687895393784953, "minimum": 0.0748474228093901, "maximum": 0.17974739476384413, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.596360262904049, "standard_deviation": 0.011483746459070507, "minimum": 5.582484366545448, "maximum": 5.616047572312406, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5522763398520849, "standard_deviation": 0.04078281493938678, "minimum": 0.5193671434739701, "maximum": 0.6252176880399438, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09029978187370447, "standard_deviation": 0.020910694854468736, "minimum": -0.11667127155474219, "maximum": -0.05072620014152866, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941955621824192, "standard_deviation": 0.0008364906537106858, "minimum": 0.9925641656600493, "maximum": 0.995169773300391, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.884328023615859, "standard_deviation": 0.017730794225914715, "minimum": 0.851396603956737, "maximum": 0.9079802820763336, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31638853324755384, "standard_deviation": 0.1294356006664591, "minimum": 0.14843232259764674, "maximum": 0.5188266123015994, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07537376247035572, "standard_deviation": 0.005263864890002724, "minimum": 0.06893038539946149, "maximum": 0.08532788787200611, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982357915835587, "standard_deviation": 0.00026857513790326336, "minimum": 0.9977459813016465, "maximum": 0.9986524092692688, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48550757117335686, "standard_deviation": 0.01839736985103402, "minimum": 0.45444054726509936, "maximum": 0.5103019658320179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4556058607701534, "standard_deviation": 0.20969284056159226, "minimum": 0.13289841193211882, "maximum": 0.7886829935098123, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04067381300289518, "standard_deviation": 0.0027090004680367947, "minimum": 0.0364453313434501, "maximum": 0.045493480827039044, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0028", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.171628713607788, "predicted_mos": 3.171628713607788, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9653fa5461167c798804201552a0a8cbb348eb290fa45b039f129ffb1fb3e108", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.4132320997305214, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.908340381534836, "全帯域_10ms_rms_envelope_nrmse": 0.08062660482551941, "全帯域_10ms_envelope_correlation": 0.9933426634853302, "全帯域_10ms_modulation_mae_db": 0.24066090163028828, "全帯域_40ms_log_rms_mae_db": 0.5064347928734465, "全帯域_40ms_rms_envelope_nrmse": 0.045963757910386246, "全帯域_40ms_envelope_correlation": 0.9976781578992729, "全帯域_40ms_modulation_mae_db": 0.30944497522352427, "300–1000 Hz_10ms_log_rms_mae_db": 1.2503890371043136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10221390810279271, "300–1000 Hz_10ms_envelope_correlation": 0.9916795413965173, "300–1000 Hz_10ms_modulation_mae_db": 0.22714675511389892, "300–1000 Hz_40ms_log_rms_mae_db": 0.7148139522986386, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05946439716750877, "300–1000 Hz_40ms_envelope_correlation": 0.9970111306710933, "300–1000 Hz_40ms_modulation_mae_db": 0.4557132569129272, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6243783799437117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17590609876375893, "1000–3000 Hz_10ms_envelope_correlation": 0.9773581401864985, "1000–3000 Hz_10ms_modulation_mae_db": 0.38834712177531355, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0517507147403642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11472614287937885, "1000–3000 Hz_40ms_envelope_correlation": 0.9902972166017886, "1000–3000 Hz_40ms_modulation_mae_db": 0.3819776015744405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8627571867891444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21174684460751741, "3000–6000 Hz_10ms_envelope_correlation": 0.9599124064355008, "3000–6000 Hz_10ms_modulation_mae_db": 1.312641654266199, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2473930724763034, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13663441712191238, "3000–6000 Hz_40ms_envelope_correlation": 0.9817107939351953, "3000–6000 Hz_40ms_modulation_mae_db": 1.6236886136609083, "mr_stft_db": 5.546759530896772, "band_mae_db": 0.12640473730499063, "rms_gain_error_db": -0.08105299585659109, "output_peak": 0.5673681486466735}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9267712526080784, "全帯域_10ms_rms_envelope_nrmse": 0.0735861682960412, "全帯域_10ms_envelope_correlation": 0.9944599198183331, "全帯域_10ms_modulation_mae_db": 0.32812503435780743, "全帯域_40ms_log_rms_mae_db": 0.504213658897563, "全帯域_40ms_rms_envelope_nrmse": 0.04133568783685489, "全帯域_40ms_envelope_correlation": 0.9981257877102822, "全帯域_40ms_modulation_mae_db": 0.4614405591616948, "300–1000 Hz_10ms_log_rms_mae_db": 1.310931946952845, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09721148896520726, "300–1000 Hz_10ms_envelope_correlation": 0.9924962357269347, "300–1000 Hz_10ms_modulation_mae_db": 0.4411925577302096, "300–1000 Hz_40ms_log_rms_mae_db": 0.7417279897146069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053715779281366324, "300–1000 Hz_40ms_envelope_correlation": 0.9975917928357503, "300–1000 Hz_40ms_modulation_mae_db": 0.7134105729279844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5899183942997726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16021666203873244, "1000–3000 Hz_10ms_envelope_correlation": 0.9813770134792884, "1000–3000 Hz_10ms_modulation_mae_db": 0.5552027754027267, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0297634798868998, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10575123773356823, "1000–3000 Hz_40ms_envelope_correlation": 0.9912816359147023, "1000–3000 Hz_40ms_modulation_mae_db": 0.7146120999807563, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9007989568814094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24738900168861208, "3000–6000 Hz_10ms_envelope_correlation": 0.9459524719506673, "3000–6000 Hz_10ms_modulation_mae_db": 1.9220030833669834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.338832755001581, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17488933052896236, "3000–6000 Hz_40ms_envelope_correlation": 0.972368114202346, "3000–6000 Hz_40ms_modulation_mae_db": 1.9736204086655806, "mr_stft_db": 5.535039483529464, "band_mae_db": 0.1957658903719579, "rms_gain_error_db": -0.07272598025051996, "output_peak": 0.5356537506029133}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9225788530478215, "全帯域_10ms_rms_envelope_nrmse": 0.08687627965683228, "全帯域_10ms_envelope_correlation": 0.9922472074239601, "全帯域_10ms_modulation_mae_db": 0.18083591743277208, "全帯域_40ms_log_rms_mae_db": 0.49480935079095073, "全帯域_40ms_rms_envelope_nrmse": 0.040125499310601, "全帯域_40ms_envelope_correlation": 0.9982437711741782, "全帯域_40ms_modulation_mae_db": 0.3912964066240017, "300–1000 Hz_10ms_log_rms_mae_db": 1.2407262297628636, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10836914148096563, "300–1000 Hz_10ms_envelope_correlation": 0.990687674187258, "300–1000 Hz_10ms_modulation_mae_db": 0.36910246685887693, "300–1000 Hz_40ms_log_rms_mae_db": 0.7128441814265424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05348431819585066, "300–1000 Hz_40ms_envelope_correlation": 0.9977080302258075, "300–1000 Hz_40ms_modulation_mae_db": 0.3912480901457142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.678055370900718, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16796068322363122, "1000–3000 Hz_10ms_envelope_correlation": 0.9811278998424244, "1000–3000 Hz_10ms_modulation_mae_db": 0.62258412231451, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1317901469935803, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11659433432924159, "1000–3000 Hz_40ms_envelope_correlation": 0.9907999370472877, "1000–3000 Hz_40ms_modulation_mae_db": 0.7132516685526608, "3000–6000 Hz_10ms_log_rms_mae_db": 1.908015660947723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22850319244726958, "3000–6000 Hz_10ms_envelope_correlation": 0.9534452060900237, "3000–6000 Hz_10ms_modulation_mae_db": 0.8767754023400317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3080813150304516, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1499888610609945, "3000–6000 Hz_40ms_envelope_correlation": 0.9777473976075144, "3000–6000 Hz_40ms_modulation_mae_db": 0.8952431248433363, "mr_stft_db": 5.559683763220787, "band_mae_db": 0.19853611644845404, "rms_gain_error_db": -0.07507255764168985, "output_peak": 0.5960262495073178}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8788183229805313, "全帯域_10ms_rms_envelope_nrmse": 0.0716850891182157, "全帯域_10ms_envelope_correlation": 0.9947768844560831, "全帯域_10ms_modulation_mae_db": 0.15537783426311608, "全帯域_40ms_log_rms_mae_db": 0.48402715289425396, "全帯域_40ms_rms_envelope_nrmse": 0.03744696920147158, "全帯域_40ms_envelope_correlation": 0.9985364548961989, "全帯域_40ms_modulation_mae_db": 0.22344177442207327, "300–1000 Hz_10ms_log_rms_mae_db": 1.2224613331140515, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08732242444942906, "300–1000 Hz_10ms_envelope_correlation": 0.9940484656230402, "300–1000 Hz_10ms_modulation_mae_db": 0.22922203861364443, "300–1000 Hz_40ms_log_rms_mae_db": 0.685469220175008, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04726124506089458, "300–1000 Hz_40ms_envelope_correlation": 0.9982800329640377, "300–1000 Hz_40ms_modulation_mae_db": 0.47020919079112494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.668280098712852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16882084914053894, "1000–3000 Hz_10ms_envelope_correlation": 0.9788627350848982, "1000–3000 Hz_10ms_modulation_mae_db": 0.8161679048844012, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0911242397733862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10809123804212355, "1000–3000 Hz_40ms_envelope_correlation": 0.9906935594148609, "1000–3000 Hz_40ms_modulation_mae_db": 0.9563140621874787, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8767539342870547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2185669354241532, "3000–6000 Hz_10ms_envelope_correlation": 0.9577921117893776, "3000–6000 Hz_10ms_modulation_mae_db": 0.6756005166012543, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2935579662751522, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14525674973969221, "3000–6000 Hz_40ms_envelope_correlation": 0.9791820594166141, "3000–6000 Hz_40ms_modulation_mae_db": 0.7235717957012763, "mr_stft_db": 5.520791635347918, "band_mae_db": 0.09346741087921441, "rms_gain_error_db": -0.0849316273166546, "output_peak": 0.6039346961093155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9044550771673703, "全帯域_10ms_rms_envelope_nrmse": 0.07650573776781708, "全帯域_10ms_envelope_correlation": 0.9940052895455691, "全帯域_10ms_modulation_mae_db": 0.17650651233575873, "全帯域_40ms_log_rms_mae_db": 0.5114698870655784, "全帯域_40ms_rms_envelope_nrmse": 0.042239083796579875, "全帯域_40ms_envelope_correlation": 0.9980409494802617, "全帯域_40ms_modulation_mae_db": 0.16248610077944023, "300–1000 Hz_10ms_log_rms_mae_db": 1.3367490441523548, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09374993821946796, "300–1000 Hz_10ms_envelope_correlation": 0.993062720262746, "300–1000 Hz_10ms_modulation_mae_db": 0.3049758644873343, "300–1000 Hz_40ms_log_rms_mae_db": 0.7364171329142176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05089605115297581, "300–1000 Hz_40ms_envelope_correlation": 0.9979073433510737, "300–1000 Hz_40ms_modulation_mae_db": 0.0988765469288811, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6413011877911514, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1644835276888918, "1000–3000 Hz_10ms_envelope_correlation": 0.9800604482172292, "1000–3000 Hz_10ms_modulation_mae_db": 0.38996286467809704, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0706282869213102, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09077168354412367, "1000–3000 Hz_40ms_envelope_correlation": 0.993485328629944, "1000–3000 Hz_40ms_modulation_mae_db": 0.40280817159241256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.917457866407196, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21670559164471476, "3000–6000 Hz_10ms_envelope_correlation": 0.9581194580631908, "3000–6000 Hz_10ms_modulation_mae_db": 0.5429873912803216, "3000–6000 Hz_40ms_log_rms_mae_db": 1.277361538077179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14153324056606195, "3000–6000 Hz_40ms_envelope_correlation": 0.9800988566055109, "3000–6000 Hz_40ms_modulation_mae_db": 0.7831820286830539, "mr_stft_db": 5.54987072797623, "band_mae_db": 0.1033093880134728, "rms_gain_error_db": -0.06101979850378667, "output_peak": 0.505345739962805}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8925740085439329, "全帯域_10ms_rms_envelope_nrmse": 0.07611731278545365, "全帯域_10ms_envelope_correlation": 0.9940998695473166, "全帯域_10ms_modulation_mae_db": 0.5159564683223764, "全帯域_40ms_log_rms_mae_db": 0.5088031909117762, "全帯域_40ms_rms_envelope_nrmse": 0.042522272619340745, "全帯域_40ms_envelope_correlation": 0.9980573897325705, "全帯域_40ms_modulation_mae_db": 0.8420845277915338, "300–1000 Hz_10ms_log_rms_mae_db": 1.2471167330808453, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09274179130726355, "300–1000 Hz_10ms_envelope_correlation": 0.9932388460478233, "300–1000 Hz_10ms_modulation_mae_db": 0.43264009614007737, "300–1000 Hz_40ms_log_rms_mae_db": 0.6939190161683622, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04927943413699217, "300–1000 Hz_40ms_envelope_correlation": 0.9980788331440451, "300–1000 Hz_40ms_modulation_mae_db": 0.8547859338277849, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6507543526988484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17276032515902212, "1000–3000 Hz_10ms_envelope_correlation": 0.9778381551997386, "1000–3000 Hz_10ms_modulation_mae_db": 0.7269332713485929, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0986090237228578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09601306900686224, "1000–3000 Hz_40ms_envelope_correlation": 0.9927022593102909, "1000–3000 Hz_40ms_modulation_mae_db": 0.8947400003201942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8495701705655943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2219058446949544, "3000–6000 Hz_10ms_envelope_correlation": 0.9561328839656528, "3000–6000 Hz_10ms_modulation_mae_db": 1.040072694693319, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2913564394031525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14657782455238624, "3000–6000 Hz_40ms_envelope_correlation": 0.9786933321092389, "3000–6000 Hz_40ms_modulation_mae_db": 1.2487990664029842, "mr_stft_db": 5.547116781572982, "band_mae_db": 0.12834949208359905, "rms_gain_error_db": -0.09248471700967237, "output_peak": 0.5431510467726174}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9156738965654839, "全帯域_10ms_rms_envelope_nrmse": 0.08021393780854805, "全帯域_10ms_envelope_correlation": 0.9935688301204272, "全帯域_10ms_modulation_mae_db": 0.35819762894105006, "全帯域_40ms_log_rms_mae_db": 0.47785976631545235, "全帯域_40ms_rms_envelope_nrmse": 0.042539861992833286, "全帯域_40ms_envelope_correlation": 0.9982442721382558, "全帯域_40ms_modulation_mae_db": 0.45201305865954106, "300–1000 Hz_10ms_log_rms_mae_db": 1.2984693462044117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10740183469256502, "300–1000 Hz_10ms_envelope_correlation": 0.9910830176519411, "300–1000 Hz_10ms_modulation_mae_db": 0.45446146429484685, "300–1000 Hz_40ms_log_rms_mae_db": 0.782347091782083, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058054061575426634, "300–1000 Hz_40ms_envelope_correlation": 0.9975640979188816, "300–1000 Hz_40ms_modulation_mae_db": 0.7703101864062368, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6324425354380168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16806671209743257, "1000–3000 Hz_10ms_envelope_correlation": 0.9790732471153676, "1000–3000 Hz_10ms_modulation_mae_db": 0.5652964917518563, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0954915681502977, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0941285991416995, "1000–3000 Hz_40ms_envelope_correlation": 0.9931343500607077, "1000–3000 Hz_40ms_modulation_mae_db": 0.9332168468294535, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8477228070369143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21830431935010092, "3000–6000 Hz_10ms_envelope_correlation": 0.9574075878629389, "3000–6000 Hz_10ms_modulation_mae_db": 1.1694343551067783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2454134103312706, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1466286490955157, "3000–6000 Hz_40ms_envelope_correlation": 0.9792812348308827, "3000–6000 Hz_40ms_modulation_mae_db": 1.613658446402124, "mr_stft_db": 5.5307302413314625, "band_mae_db": 0.17885167628759166, "rms_gain_error_db": -0.14365991985125898, "output_peak": 0.5248470129538759}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8658956607746621, "全帯域_10ms_rms_envelope_nrmse": 0.06846295320454314, "全帯域_10ms_envelope_correlation": 0.9952830089815811, "全帯域_10ms_modulation_mae_db": 0.3415530371727134, "全帯域_40ms_log_rms_mae_db": 0.4566792474576204, "全帯域_40ms_rms_envelope_nrmse": 0.03889217591356016, "全帯域_40ms_envelope_correlation": 0.9984365142743863, "全帯域_40ms_modulation_mae_db": 0.4040757282228837, "300–1000 Hz_10ms_log_rms_mae_db": 1.2033983417756775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08810300598044774, "300–1000 Hz_10ms_envelope_correlation": 0.9940807911197677, "300–1000 Hz_10ms_modulation_mae_db": 0.35334003076648585, "300–1000 Hz_40ms_log_rms_mae_db": 0.6830221893449263, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05459238637968134, "300–1000 Hz_40ms_envelope_correlation": 0.9977875603467202, "300–1000 Hz_40ms_modulation_mae_db": 0.6426337106540786, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6270320591744312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16757796974743508, "1000–3000 Hz_10ms_envelope_correlation": 0.9794041222812117, "1000–3000 Hz_10ms_modulation_mae_db": 0.19440127284670217, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0529861207576736, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10172246799150914, "1000–3000 Hz_40ms_envelope_correlation": 0.991810790303082, "1000–3000 Hz_40ms_modulation_mae_db": 0.4346127913535925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8597810253221145, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2145848137924861, "3000–6000 Hz_10ms_envelope_correlation": 0.9587755382993735, "3000–6000 Hz_10ms_modulation_mae_db": 1.2461981402833795, "3000–6000 Hz_40ms_log_rms_mae_db": 1.293973507833602, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13850458872219074, "3000–6000 Hz_40ms_envelope_correlation": 0.9809932510943077, "3000–6000 Hz_40ms_modulation_mae_db": 1.4569732490757532, "mr_stft_db": 5.542519127964336, "band_mae_db": 0.14168827906243722, "rms_gain_error_db": -0.10157543460044945, "output_peak": 0.517975881090333}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.540845225770513, "spectral_convergence": 0.21323069730499716, "spectral_convergence_status": "ok", "amplitude_floor": 1.8586249352090494e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.456931684902496, "spectral_convergence": 0.2075992777414556, "spectral_convergence_status": "ok", "amplitude_floor": 1.6686015821705495e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.642501682017304, "spectral_convergence": 0.2092164455476309, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267872572435527e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003360135402914431, "output_power": 0.0003399491090318446, "delta_db": 0.05057128827414971, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011242144422279699, "output_power": 0.0011203134665681471, "delta_db": -0.015096037810730274, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015417458345753522, "output_power": 0.0014902390069441817, "delta_db": -0.14756856890804748, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.127065821964785e-05, "output_power": 7.800240740907377e-05, "delta_db": -0.17825770520386164, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.295885949753132e-05, "output_power": 4.978913710564015e-05, "delta_db": -0.26804024133274545, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.948150516097124e-05, "output_power": 1.9042898370537614e-05, "delta_db": -0.09889458230040915, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12640473730499063, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08105299585659109, "pitch": null, "output_rms_dbfs": -25.090116761177335, "output_peak": 0.5673681486466735, "reference_rms_dbfs": -25.009063765320747, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.546759530896772}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025375764068356454, "log_rms_mae_db": 0.908340381534836, "rms_envelope_nrmse": 0.08062660482551941, "envelope_correlation": 0.9933426634853302, "rms_gain_error_db": -0.0810511749403986, "reference_mean_power": 0.0031583120304531104, "output_mean_power": 0.0030999159387265656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1807904078754825e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008327873716353349, "output_power": 0.0008282834135405775, "reference_active": true, "delta_db": -0.0235516626797938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013733244323789595, "output_power": 0.00012287326468910629, "reference_active": true, "delta_db": -0.4831574907902533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.747498790065721e-05, "output_power": 1.6629895821573584e-05, "reference_active": true, "delta_db": -0.21527355142081772}], "active_band_count": 3, "active_band_mae_db": 0.24066090163028828}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002402786371599188, "log_rms_mae_db": 0.5064347928734465, "rms_envelope_nrmse": 0.045963757910386246, "envelope_correlation": 0.9976781578992729, "rms_gain_error_db": -0.08104822149658375, "reference_mean_power": 0.0031741738524038612, "output_mean_power": 0.0031154865998057456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0384864838590044e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000782161356346486, "output_power": 0.0007773622179384677, "reference_active": true, "delta_db": -0.02672926452712696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.85949543312588e-05, "output_power": 5.2469220873275766e-05, "reference_active": true, "delta_db": -0.47955604791970663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3154862476080315e-06, "output_power": 1.1936619519150118e-06, "reference_active": true, "delta_db": -0.42204961322373924}], "active_band_count": 3, "active_band_mae_db": 0.30944497522352427}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002501327279402343, "log_rms_mae_db": 1.2503890371043136, "rms_envelope_nrmse": 0.10221390810279271, "envelope_correlation": 0.9916795413965173, "rms_gain_error_db": -0.10174654637566448, "reference_mean_power": 0.0021283926153372414, "output_mean_power": 0.002079108188110596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0698576130619868e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007463968445819582, "output_power": 0.0007393789678650322, "reference_active": true, "delta_db": -0.041027021352028105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014378870760237378, "output_power": 0.00012685575336777178, "reference_active": true, "delta_db": -0.5441461138371974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4173849116031066e-05, "output_power": 1.3863123502815343e-05, "reference_active": true, "delta_db": -0.09626713015247135}], "active_band_count": 3, "active_band_mae_db": 0.22714675511389892}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00023719519351520602, "log_rms_mae_db": 0.7148139522986386, "rms_envelope_nrmse": 0.05946439716750877, "envelope_correlation": 0.9970111306710933, "rms_gain_error_db": -0.10174648559829878, "reference_mean_power": 0.0021390960955203144, "output_mean_power": 0.002089563850928105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.624199297765067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007211274534636877, "output_power": 0.000710858461111666, "reference_active": true, "delta_db": -0.06228892503298846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.136224424017926e-05, "output_power": 5.372471453671761e-05, "reference_active": true, "delta_db": -0.5772711838819778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3649293472151981e-06, "output_power": 1.154389068125916e-06, "reference_active": true, "delta_db": -0.7275796618238153}], "active_band_count": 3, "active_band_mae_db": 0.4557132569129272}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.486004098842285e-05, "log_rms_mae_db": 1.6243783799437117, "rms_envelope_nrmse": 0.17590609876375893, "envelope_correlation": 0.9773581401864985, "rms_gain_error_db": -0.29788540573804123, "reference_mean_power": 9.73047769043704e-05, "output_mean_power": 9.085432804039447e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.517204404381221e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1360183676562794e-05, "output_power": 3.844617058656856e-05, "reference_active": true, "delta_db": -0.3172937016361849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.821587959830417e-06, "output_power": 4.5028876869321235e-06, "reference_active": true, "delta_db": -0.29698979240044865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2148111012630244e-06, "output_power": 1.951016330318057e-06, "reference_active": true, "delta_db": -0.5507578712893071}], "active_band_count": 3, "active_band_mae_db": 0.38834712177531355}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5230962699738786e-05, "log_rms_mae_db": 1.0517507147403642, "rms_envelope_nrmse": 0.11472614287937885, "envelope_correlation": 0.9902972166017886, "rms_gain_error_db": -0.2979083795232217, "reference_mean_power": 9.77871242521371e-05, "output_mean_power": 9.130421701908245e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.922472441667658e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.88087102342988e-05, "output_power": 3.6086940681198836e-05, "reference_active": true, "delta_db": -0.31579143947503197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.146701619692133e-06, "output_power": 1.9771261282876816e-06, "reference_active": true, "delta_db": -0.35737308496874143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.612101873452723e-07, "output_power": 1.4458247493776155e-07, "reference_active": true, "delta_db": -0.4727682802795482}], "active_band_count": 3, "active_band_mae_db": 0.3819776015744405}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.5953225979159985e-05, "log_rms_mae_db": 1.8627571867891444, "rms_envelope_nrmse": 0.21174684460751741, "envelope_correlation": 0.9599124064355008, "rms_gain_error_db": -0.2478800225510787, "reference_mean_power": 3.136903007980474e-05, "output_mean_power": 2.96287335502883e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9944851157580815e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0899878382466362e-05, "output_power": 1.0245000783370308e-05, "reference_active": true, "delta_db": -0.26909656296722895}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5519027988888207e-06, "output_power": 2.0843676747026305e-06, "reference_active": true, "delta_db": -0.8788979891908439}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4088478818809383e-06, "output_power": 7.41090186382067e-07, "reference_active": true, "delta_db": -2.789930410640524}], "active_band_count": 3, "active_band_mae_db": 1.312641654266199}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9738091980327894e-05, "log_rms_mae_db": 1.2473930724763034, "rms_envelope_nrmse": 0.13663441712191238, "envelope_correlation": 0.9817107939351953, "rms_gain_error_db": -0.2477699533710711, "reference_mean_power": 3.151908471872331e-05, "output_mean_power": 2.9771217954803223e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7532611532165288e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0780259770731776e-05, "output_power": 9.925132887054119e-06, "reference_active": true, "delta_db": -0.35892895904257605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1761748042676738e-06, "output_power": 9.22992416218727e-07, "reference_active": true, "delta_db": -1.0527373916092655}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.367404555815231e-08, "output_power": 4.223566414508688e-08, "reference_active": true, "delta_db": -3.459399490330884}], "active_band_count": 3, "active_band_mae_db": 1.6236886136609083}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9014431849142414, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "baf0d9e28e8ea7a6342332929ffeb0129c4061d5f0674bb6332c053e3076303c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9793877201758321, "standard_deviation": 0.001430943463275704, "minimum": 0.9773581401864985, "maximum": 0.9813770134792884, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.639020297369938, "standard_deviation": 0.02760727361157176, "minimum": 1.5899183942997726, "maximum": 1.678055370900718, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.532361978125275, "standard_deviation": 0.20120545164185705, "minimum": 0.19440127284670217, "maximum": 0.8161679048844012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1682241034824304, "standard_deviation": 0.004759488880238248, "minimum": 0.16021666203873244, "maximum": 0.17590609876375893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.991775634660333, "standard_deviation": 0.0012060358651115053, "minimum": 0.9902972166017886, "maximum": 0.993485328629944, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0777679476182962, "standard_deviation": 0.03271457685071807, "minimum": 1.0297634798868998, "maximum": 1.1317901469935803, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6789416552988736, "standard_deviation": 0.24341139551426577, "minimum": 0.3819776015744405, "maximum": 0.9563140621874787, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10347484658356335, "standard_deviation": 0.00950696963583562, "minimum": 0.09077168354412367, "maximum": 0.11659433432924159, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9559422080570906, "standard_deviation": 0.004474791909137303, "minimum": 0.9459524719506673, "maximum": 0.9599124064355008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8778572010296437, "standard_deviation": 0.027433772497464433, "minimum": 1.8477228070369143, "maximum": 1.917457866407196, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0982141547422835, "standard_deviation": 0.4294304408736612, "minimum": 0.5429873912803216, "maximum": 1.9220030833669834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22221331795622606, "standard_deviation": 0.011339719555888734, "minimum": 0.21174684460751741, "maximum": 0.24738900168861208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9787593799752012, "standard_deviation": 0.002874095171804776, "minimum": 0.972368114202346, "maximum": 0.9817107939351953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2869962505535866, "standard_deviation": 0.030782009160791265, "minimum": 1.2454134103312706, "maximum": 1.338832755001581, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2898420916793771, "standard_deviation": 0.4545885870719815, "minimum": 0.7235717957012763, "maximum": 1.9736204086655806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14750170767346452, "standard_deviation": 0.011937888052282533, "minimum": 0.13663441712191238, "maximum": 0.17488933052896236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925471615020035, "standard_deviation": 0.0012932202654290562, "minimum": 0.990687674187258, "maximum": 0.9940807911197677, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2637802515184204, "standard_deviation": 0.04643393740808527, "minimum": 1.2033983417756775, "maximum": 1.3367490441523548, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3515101592506718, "standard_deviation": 0.09112228009682534, "minimum": 0.22714675511389892, "maximum": 0.45446146429484685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09713919164976736, "standard_deviation": 0.00815945838292367, "minimum": 0.08732242444942906, "maximum": 0.10836914148096563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977411026821761, "standard_deviation": 0.00038207684884724566, "minimum": 0.9970111306710933, "maximum": 0.9982800329640377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.718820096728048, "standard_deviation": 0.033670041906722334, "minimum": 0.6830221893449263, "maximum": 0.782347091782083, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5496484360743414, "standard_deviation": 0.24514869439926185, "minimum": 0.0988765469288811, "maximum": 0.8547859338277849, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05334345911883703, "standard_deviation": 0.004155433166979644, "minimum": 0.04726124506089458, "maximum": 0.05946439716750877, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14579662380646471, "standard_deviation": 0.04072881758770786, "minimum": 0.09346741087921441, "maximum": 0.19853611644845404, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.541563911479994, "standard_deviation": 0.012232845236213437, "minimum": 5.520791635347918, "maximum": 5.559683763220787, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5492878157057315, "standard_deviation": 0.03631195187853187, "minimum": 0.505345739962805, "maximum": 0.6039346961093155, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08906537887882787, "standard_deviation": 0.02529645098584985, "minimum": -0.14365991985125898, "maximum": -0.06101979850378667, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993972959172325, "standard_deviation": 0.00093774962941603, "minimum": 0.9922472074239601, "maximum": 0.9952830089815811, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9018884316528395, "standard_deviation": 0.021380801244574633, "minimum": 0.8658956607746621, "maximum": 0.9267712526080784, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.28715166680698534, "standard_deviation": 0.1225378162278037, "minimum": 0.15537783426311608, "maximum": 0.5159564683223764, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07675926043287132, "standard_deviation": 0.005786265002933039, "minimum": 0.06846295320454314, "maximum": 0.08687627965683228, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981704121631758, "standard_deviation": 0.00026443327935622913, "minimum": 0.9976781578992729, "maximum": 0.9985364548961989, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4930371309008302, "standard_deviation": 0.01900870411055153, "minimum": 0.4566792474576204, "maximum": 0.5114698870655784, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40578539136058656, "standard_deviation": 0.20616550224827163, "minimum": 0.16248610077944023, "maximum": 0.8420845277915338, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04138316357270347, "standard_deviation": 0.002608573325432337, "minimum": 0.03744696920147158, "maximum": 0.045963757910386246, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0028", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.3817949295043945, "predicted_mos": 3.3817949295043945, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "48650bb094c007578222dca57ecd659de0c98509c1f153f783350e308c22caae", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.505912400316447, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8228879018065455, "全帯域_10ms_rms_envelope_nrmse": 0.06990737538380108, "全帯域_10ms_envelope_correlation": 0.9952100032056632, "全帯域_10ms_modulation_mae_db": 0.38869587957001395, "全帯域_40ms_log_rms_mae_db": 0.4473377773634654, "全帯域_40ms_rms_envelope_nrmse": 0.04493692078840125, "全帯域_40ms_envelope_correlation": 0.9980066321599933, "全帯域_40ms_modulation_mae_db": 0.5189760099427775, "300–1000 Hz_10ms_log_rms_mae_db": 1.3477307911447143, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08951479641996639, "300–1000 Hz_10ms_envelope_correlation": 0.9940511848399214, "300–1000 Hz_10ms_modulation_mae_db": 0.4987328446112606, "300–1000 Hz_40ms_log_rms_mae_db": 0.7720318363531835, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056362456871885305, "300–1000 Hz_40ms_envelope_correlation": 0.9977786423621894, "300–1000 Hz_40ms_modulation_mae_db": 0.7444607944234335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.929434000610556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1767456078930179, "1000–3000 Hz_10ms_envelope_correlation": 0.9770672445050964, "1000–3000 Hz_10ms_modulation_mae_db": 0.4364445360904277, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3099665624403405, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11226175834278107, "1000–3000 Hz_40ms_envelope_correlation": 0.9903843280610924, "1000–3000 Hz_40ms_modulation_mae_db": 0.5534096194533109, "3000–6000 Hz_10ms_log_rms_mae_db": 2.386118389404549, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28209043596432926, "3000–6000 Hz_10ms_envelope_correlation": 0.929719851809578, "3000–6000 Hz_10ms_modulation_mae_db": 1.2266892229723938, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6516175860787532, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18034260665322155, "3000–6000 Hz_40ms_envelope_correlation": 0.9679323595173035, "3000–6000 Hz_40ms_modulation_mae_db": 1.4022559881962187, "mr_stft_db": 5.906781649265916, "band_mae_db": 0.08772312282621074, "rms_gain_error_db": -0.11568533901189407, "output_peak": 0.484611849862581}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8004539327992027, "全帯域_10ms_rms_envelope_nrmse": 0.07431513078776113, "全帯域_10ms_envelope_correlation": 0.9943196557091121, "全帯域_10ms_modulation_mae_db": 0.3524136712985522, "全帯域_40ms_log_rms_mae_db": 0.42766513960057645, "全帯域_40ms_rms_envelope_nrmse": 0.041583139820653287, "全帯域_40ms_envelope_correlation": 0.9980560501871012, "全帯域_40ms_modulation_mae_db": 0.7421179640464629, "300–1000 Hz_10ms_log_rms_mae_db": 1.4073474835760946, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09366170109862385, "300–1000 Hz_10ms_envelope_correlation": 0.992998350077545, "300–1000 Hz_10ms_modulation_mae_db": 0.4330579382837469, "300–1000 Hz_40ms_log_rms_mae_db": 0.8393062709802428, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056927298956503274, "300–1000 Hz_40ms_envelope_correlation": 0.9972097311232168, "300–1000 Hz_40ms_modulation_mae_db": 0.8234408015882231, "1000–3000 Hz_10ms_log_rms_mae_db": 2.037471401566601, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19061753402501533, "1000–3000 Hz_10ms_envelope_correlation": 0.9732238598866998, "1000–3000 Hz_10ms_modulation_mae_db": 0.9816654255183991, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3757033605170272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11234084953430727, "1000–3000 Hz_40ms_envelope_correlation": 0.9900332086618143, "1000–3000 Hz_40ms_modulation_mae_db": 1.563951453668577, "3000–6000 Hz_10ms_log_rms_mae_db": 2.39360482128943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28601007712644644, "3000–6000 Hz_10ms_envelope_correlation": 0.9278201542130876, "3000–6000 Hz_10ms_modulation_mae_db": 1.3035160598803148, "3000–6000 Hz_40ms_log_rms_mae_db": 1.64453468280813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17723979345200727, "3000–6000 Hz_40ms_envelope_correlation": 0.9690243377114285, "3000–6000 Hz_40ms_modulation_mae_db": 1.5552570168739361, "mr_stft_db": 5.899114995803465, "band_mae_db": 0.16967641724559754, "rms_gain_error_db": -0.04289837650328915, "output_peak": 0.5355126760511938}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.82303299677531, "全帯域_10ms_rms_envelope_nrmse": 0.066757548455494, "全帯域_10ms_envelope_correlation": 0.9954910428241082, "全帯域_10ms_modulation_mae_db": 0.5610946857054606, "全帯域_40ms_log_rms_mae_db": 0.41824458542034915, "全帯域_40ms_rms_envelope_nrmse": 0.03620751989782693, "全帯域_40ms_envelope_correlation": 0.998621489253708, "全帯域_40ms_modulation_mae_db": 0.5485144780938426, "300–1000 Hz_10ms_log_rms_mae_db": 1.3258114951874447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0826629078436214, "300–1000 Hz_10ms_envelope_correlation": 0.9947432017028772, "300–1000 Hz_10ms_modulation_mae_db": 0.5594228593754803, "300–1000 Hz_40ms_log_rms_mae_db": 0.781244482100772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048460441625616024, "300–1000 Hz_40ms_envelope_correlation": 0.9982145422321702, "300–1000 Hz_40ms_modulation_mae_db": 0.6742060700698844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9633921102748801, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21289181225090278, "1000–3000 Hz_10ms_envelope_correlation": 0.9663255502494054, "1000–3000 Hz_10ms_modulation_mae_db": 0.9927234588093256, "1000–3000 Hz_40ms_log_rms_mae_db": 1.296271048284648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12656592723809748, "1000–3000 Hz_40ms_envelope_correlation": 0.9876578594772789, "1000–3000 Hz_40ms_modulation_mae_db": 1.1076733623443842, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4138106534870754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26699791530681133, "3000–6000 Hz_10ms_envelope_correlation": 0.9362138221852921, "3000–6000 Hz_10ms_modulation_mae_db": 1.2703332932954632, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7050221425034748, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16962244885929903, "3000–6000 Hz_40ms_envelope_correlation": 0.9712489246258565, "3000–6000 Hz_40ms_modulation_mae_db": 1.328829432926844, "mr_stft_db": 5.888936995440495, "band_mae_db": 0.10699821490137607, "rms_gain_error_db": -0.08445368806894381, "output_peak": 0.5364989465900862}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8898894499836484, "全帯域_10ms_rms_envelope_nrmse": 0.08761989183564098, "全帯域_10ms_envelope_correlation": 0.9924345205965588, "全帯域_10ms_modulation_mae_db": 0.22924196079756654, "全帯域_40ms_log_rms_mae_db": 0.4666899950629656, "全帯域_40ms_rms_envelope_nrmse": 0.05420005768488613, "全帯域_40ms_envelope_correlation": 0.9971594722397127, "全帯域_40ms_modulation_mae_db": 0.32995810885583493, "300–1000 Hz_10ms_log_rms_mae_db": 1.4016870301294708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10860005001070189, "300–1000 Hz_10ms_envelope_correlation": 0.9911653538067764, "300–1000 Hz_10ms_modulation_mae_db": 0.25200200927177663, "300–1000 Hz_40ms_log_rms_mae_db": 0.8300561713911491, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07031550818203423, "300–1000 Hz_40ms_envelope_correlation": 0.9965178600187634, "300–1000 Hz_40ms_modulation_mae_db": 0.49086610492713456, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9626697855813187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1821249500294843, "1000–3000 Hz_10ms_envelope_correlation": 0.9756720624758076, "1000–3000 Hz_10ms_modulation_mae_db": 1.1058211638795923, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3238831635280723, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.109526178094982, "1000–3000 Hz_40ms_envelope_correlation": 0.9908314515182134, "1000–3000 Hz_40ms_modulation_mae_db": 1.2312416787103457, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3778309053027478, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27878554740917055, "3000–6000 Hz_10ms_envelope_correlation": 0.9343788904639959, "3000–6000 Hz_10ms_modulation_mae_db": 0.8385045775127072, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5925044040864587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17540121046623106, "3000–6000 Hz_40ms_envelope_correlation": 0.9715378650624196, "3000–6000 Hz_40ms_modulation_mae_db": 0.7721124997005999, "mr_stft_db": 5.9030621425305325, "band_mae_db": 0.14418207883460243, "rms_gain_error_db": -0.18342745011402511, "output_peak": 0.4730326286698824}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8708786694469189, "全帯域_10ms_rms_envelope_nrmse": 0.07323190008534398, "全帯域_10ms_envelope_correlation": 0.9945353080513003, "全帯域_10ms_modulation_mae_db": 0.3167060809822721, "全帯域_40ms_log_rms_mae_db": 0.4729126503274834, "全帯域_40ms_rms_envelope_nrmse": 0.03762280873871966, "全帯域_40ms_envelope_correlation": 0.998494331258561, "全帯域_40ms_modulation_mae_db": 0.3738559851191478, "300–1000 Hz_10ms_log_rms_mae_db": 1.4527110048629501, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09851818383137889, "300–1000 Hz_10ms_envelope_correlation": 0.9923526480026102, "300–1000 Hz_10ms_modulation_mae_db": 0.4804697638106834, "300–1000 Hz_40ms_log_rms_mae_db": 0.9011839402379704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05115636839540113, "300–1000 Hz_40ms_envelope_correlation": 0.9979312154650997, "300–1000 Hz_40ms_modulation_mae_db": 0.5232159427964733, "1000–3000 Hz_10ms_log_rms_mae_db": 2.011669775094941, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19653302129705172, "1000–3000 Hz_10ms_envelope_correlation": 0.9713704395610205, "1000–3000 Hz_10ms_modulation_mae_db": 1.4334731259421047, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3674728025168115, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11878376310298074, "1000–3000 Hz_40ms_envelope_correlation": 0.9890241155356143, "1000–3000 Hz_40ms_modulation_mae_db": 1.73547494206864, "3000–6000 Hz_10ms_log_rms_mae_db": 2.374766820362133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25936919766689004, "3000–6000 Hz_10ms_envelope_correlation": 0.9405876872567309, "3000–6000 Hz_10ms_modulation_mae_db": 1.4047707594179244, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6101741989432445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17148251758481675, "3000–6000 Hz_40ms_envelope_correlation": 0.9710941441406197, "3000–6000 Hz_40ms_modulation_mae_db": 1.673777020536143, "mr_stft_db": 5.9099291601481765, "band_mae_db": 0.1504054076691081, "rms_gain_error_db": -0.08305485678679898, "output_peak": 0.5332380875253034}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8583746972679573, "全帯域_10ms_rms_envelope_nrmse": 0.0684658200623994, "全帯域_10ms_envelope_correlation": 0.9952237807178064, "全帯域_10ms_modulation_mae_db": 0.5701129229558219, "全帯域_40ms_log_rms_mae_db": 0.44341078098372744, "全帯域_40ms_rms_envelope_nrmse": 0.03636180018355641, "全帯域_40ms_envelope_correlation": 0.9985774528858009, "全帯域_40ms_modulation_mae_db": 0.691800506044619, "300–1000 Hz_10ms_log_rms_mae_db": 1.3827947146487176, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08679223362776398, "300–1000 Hz_10ms_envelope_correlation": 0.9940474226092554, "300–1000 Hz_10ms_modulation_mae_db": 0.6126753698770572, "300–1000 Hz_40ms_log_rms_mae_db": 0.8273482294391657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049056484448990745, "300–1000 Hz_40ms_envelope_correlation": 0.9980108547307367, "300–1000 Hz_40ms_modulation_mae_db": 0.8180462170969292, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0230633435639986, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18218233200067285, "1000–3000 Hz_10ms_envelope_correlation": 0.9755341199662612, "1000–3000 Hz_10ms_modulation_mae_db": 1.0516972927380082, "1000–3000 Hz_40ms_log_rms_mae_db": 1.376058070514034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11610845805813591, "1000–3000 Hz_40ms_envelope_correlation": 0.9895580260961909, "1000–3000 Hz_40ms_modulation_mae_db": 1.3423669833323377, "3000–6000 Hz_10ms_log_rms_mae_db": 2.332861659857867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25050525711420296, "3000–6000 Hz_10ms_envelope_correlation": 0.9441914473903961, "3000–6000 Hz_10ms_modulation_mae_db": 1.3464059055282298, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5942382624052065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16334257416647183, "3000–6000 Hz_40ms_envelope_correlation": 0.9735081250187855, "3000–6000 Hz_40ms_modulation_mae_db": 1.5557821685577358, "mr_stft_db": 5.889276911918821, "band_mae_db": 0.09564885382379669, "rms_gain_error_db": -0.07095046798634996, "output_peak": 0.5454761097550267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8394812990953857, "全帯域_10ms_rms_envelope_nrmse": 0.07059068846057943, "全帯域_10ms_envelope_correlation": 0.9950896844809138, "全帯域_10ms_modulation_mae_db": 0.4205702456343146, "全帯域_40ms_log_rms_mae_db": 0.45144559071007595, "全帯域_40ms_rms_envelope_nrmse": 0.03701857339293047, "全帯域_40ms_envelope_correlation": 0.9987637501058904, "全帯域_40ms_modulation_mae_db": 0.31340617096086687, "300–1000 Hz_10ms_log_rms_mae_db": 1.4642193420613947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0932209750214819, "300–1000 Hz_10ms_envelope_correlation": 0.993458851384202, "300–1000 Hz_10ms_modulation_mae_db": 0.5780563182727544, "300–1000 Hz_40ms_log_rms_mae_db": 0.8663381397059126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0503572681306556, "300–1000 Hz_40ms_envelope_correlation": 0.9983480204540041, "300–1000 Hz_40ms_modulation_mae_db": 0.24335202008131948, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9367645231827824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1824976048979074, "1000–3000 Hz_10ms_envelope_correlation": 0.9754991683396516, "1000–3000 Hz_10ms_modulation_mae_db": 1.2918992539655119, "1000–3000 Hz_40ms_log_rms_mae_db": 1.342972713126302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1148394071324019, "1000–3000 Hz_40ms_envelope_correlation": 0.9897351037319629, "1000–3000 Hz_40ms_modulation_mae_db": 1.4869595447867472, "3000–6000 Hz_10ms_log_rms_mae_db": 2.390737959230876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2531553214176107, "3000–6000 Hz_10ms_envelope_correlation": 0.944404182209383, "3000–6000 Hz_10ms_modulation_mae_db": 1.0687031681159755, "3000–6000 Hz_40ms_log_rms_mae_db": 1.578704306412171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16015034181894947, "3000–6000 Hz_40ms_envelope_correlation": 0.9753546898206121, "3000–6000 Hz_40ms_modulation_mae_db": 1.3377487590221318, "mr_stft_db": 5.888548734923508, "band_mae_db": 0.07937984033333548, "rms_gain_error_db": -0.12641712029352112, "output_peak": 0.5540919225567829}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8740218811737711, "全帯域_10ms_rms_envelope_nrmse": 0.07652412508906473, "全帯域_10ms_envelope_correlation": 0.994102909251376, "全帯域_10ms_modulation_mae_db": 0.3575418370122325, "全帯域_40ms_log_rms_mae_db": 0.4944343356367747, "全帯域_40ms_rms_envelope_nrmse": 0.04120229357146942, "全帯域_40ms_envelope_correlation": 0.9982800581785161, "全帯域_40ms_modulation_mae_db": 0.5641667037287402, "300–1000 Hz_10ms_log_rms_mae_db": 1.38592119852755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09150837467288434, "300–1000 Hz_10ms_envelope_correlation": 0.9936000278238284, "300–1000 Hz_10ms_modulation_mae_db": 0.36260014982902583, "300–1000 Hz_40ms_log_rms_mae_db": 0.8023506525021066, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05117856057068627, "300–1000 Hz_40ms_envelope_correlation": 0.9981311190177077, "300–1000 Hz_40ms_modulation_mae_db": 0.6298101456171812, "1000–3000 Hz_10ms_log_rms_mae_db": 1.956404123278851, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.190228660094957, "1000–3000 Hz_10ms_envelope_correlation": 0.9732825764980158, "1000–3000 Hz_10ms_modulation_mae_db": 0.7269198180449429, "1000–3000 Hz_40ms_log_rms_mae_db": 1.310628923037237, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11771689504140406, "1000–3000 Hz_40ms_envelope_correlation": 0.9891085467285631, "1000–3000 Hz_40ms_modulation_mae_db": 0.9255383732509527, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3594579678817187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2583166624743489, "3000–6000 Hz_10ms_envelope_correlation": 0.941172425074083, "3000–6000 Hz_10ms_modulation_mae_db": 1.1520690085908907, "3000–6000 Hz_40ms_log_rms_mae_db": 1.652736580787112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17157145033747515, "3000–6000 Hz_40ms_envelope_correlation": 0.9709503264052062, "3000–6000 Hz_40ms_modulation_mae_db": 1.2378908177247794, "mr_stft_db": 5.896410851166858, "band_mae_db": 0.12301710733376513, "rms_gain_error_db": -0.11988460777504897, "output_peak": 0.53867060003673}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.945648936078213, "spectral_convergence": 0.21343726940059254, "spectral_convergence_status": "ok", "amplitude_floor": 1.8586249352090494e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.8834342518450775, "spectral_convergence": 0.21051590375591178, "spectral_convergence_status": "ok", "amplitude_floor": 1.6686015821705495e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.891261759874457, "spectral_convergence": 0.21858962457716483, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267872572435527e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003360135402914431, "output_power": 0.00034516624355143536, "delta_db": 0.11671537691308692, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011242144422279699, "output_power": 0.0011361022145099332, "delta_db": 0.04568246215995947, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015417458345753522, "output_power": 0.001437890302984608, "delta_db": -0.3028702886657276, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.127065821964785e-05, "output_power": 8.152082741606137e-05, "delta_db": 0.013348018374180107, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.295885949753132e-05, "output_power": 5.252127595763341e-05, "delta_db": -0.036033553832040155, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.948150516097124e-05, "output_power": 1.9534010263792643e-05, "delta_db": 0.011689037012270122, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.08772312282621074, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11568533901189407, "pitch": null, "output_rms_dbfs": -25.12474910433264, "output_peak": 0.484611849862581, "reference_rms_dbfs": -25.009063765320747, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.906781649265916}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025375764068356454, "log_rms_mae_db": 0.8228879018065455, "rms_envelope_nrmse": 0.06990737538380108, "envelope_correlation": 0.9952100032056632, "rms_gain_error_db": -0.11568455874451486, "reference_mean_power": 0.0031583120304531104, "output_mean_power": 0.0030752935596567894}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1807904078754825e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008327873716353349, "output_power": 0.0008025407401844621, "reference_active": true, "delta_db": -0.1606704275788509}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013733244323789595, "output_power": 0.00012541075880747194, "reference_active": true, "delta_db": -0.39438350933411254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.747498790065721e-05, "output_power": 1.5181459087690116e-05, "reference_active": true, "delta_db": -0.6110337017970784}], "active_band_count": 3, "active_band_mae_db": 0.38869587957001395}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002402786371599188, "log_rms_mae_db": 0.4473377773634654, "rms_envelope_nrmse": 0.04493692078840125, "envelope_correlation": 0.9980066321599933, "rms_gain_error_db": -0.1156810693710481, "reference_mean_power": 0.0031741738524038612, "output_mean_power": 0.003090740925651691}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0384864838590044e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000782161356346486, "output_power": 0.0007495610674296808, "reference_active": true, "delta_db": -0.1848933429109007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.85949543312588e-05, "output_power": 5.2841270181066116e-05, "reference_active": true, "delta_db": -0.4488697161037847}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3154862476080315e-06, "output_power": 1.0635791503706593e-06, "reference_active": true, "delta_db": -0.9231649708136469}], "active_band_count": 3, "active_band_mae_db": 0.5189760099427775}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002501327279402343, "log_rms_mae_db": 1.3477307911447143, "rms_envelope_nrmse": 0.08951479641996639, "envelope_correlation": 0.9940511848399214, "rms_gain_error_db": -0.19275268448350316, "reference_mean_power": 0.0021283926153372414, "output_mean_power": 0.002035993911695331}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0698576130619868e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007463968445819582, "output_power": 0.0007050186563048031, "reference_active": true, "delta_db": -0.24769185241305752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014378870760237378, "output_power": 0.00012666128426317024, "reference_active": true, "delta_db": -0.5508089303787358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4173849116031066e-05, "output_power": 1.2070299050058466e-05, "reference_active": true, "delta_db": -0.6976977510419884}], "active_band_count": 3, "active_band_mae_db": 0.4987328446112606}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00023719519351520602, "log_rms_mae_db": 0.7720318363531835, "rms_envelope_nrmse": 0.056362456871885305, "envelope_correlation": 0.9977786423621894, "rms_gain_error_db": -0.19275340852892134, "reference_mean_power": 0.0021390960955203144, "output_mean_power": 0.002046232386609863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.624199297765067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007211274534636877, "output_power": 0.0006776185298513549, "reference_active": true, "delta_db": -0.2702675604533695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.136224424017926e-05, "output_power": 5.369172081070858e-05, "reference_active": true, "delta_db": -0.5799391170613715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3649293472151981e-06, "output_power": 9.926418952920428e-07, "reference_active": true, "delta_db": -1.3831757057555598}], "active_band_count": 3, "active_band_mae_db": 0.7444607944234335}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.486004098842285e-05, "log_rms_mae_db": 1.929434000610556, "rms_envelope_nrmse": 0.1767456078930179, "envelope_correlation": 0.9770672445050964, "rms_gain_error_db": -0.14474935751345858, "reference_mean_power": 9.73047769043704e-05, "output_mean_power": 9.411508211777026e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.517204404381221e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1360183676562794e-05, "output_power": 4.010346770206547e-05, "reference_active": true, "delta_db": -0.13400531547129763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.821587959830417e-06, "output_power": 4.466451701777627e-06, "reference_active": true, "delta_db": -0.3322745188521093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2148111012630244e-06, "output_power": 1.8240272167691185e-06, "reference_active": true, "delta_db": -0.8430537739478762}], "active_band_count": 3, "active_band_mae_db": 0.4364445360904277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5230962699738786e-05, "log_rms_mae_db": 1.3099665624403405, "rms_envelope_nrmse": 0.11226175834278107, "envelope_correlation": 0.9903843280610924, "rms_gain_error_db": -0.14472462895933308, "reference_mean_power": 9.77871242521371e-05, "output_mean_power": 9.458215644585917e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.922472441667658e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.88087102342988e-05, "output_power": 3.772387843162624e-05, "reference_active": true, "delta_db": -0.12312873037053415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.146701619692133e-06, "output_power": 2.0017921372506152e-06, "reference_active": true, "delta_db": -0.30352704895941907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.612101873452723e-07, "output_power": 1.2134871303482687e-07, "reference_active": true, "delta_db": -1.2335730790299795}], "active_band_count": 3, "active_band_mae_db": 0.5534096194533109}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.5953225979159985e-05, "log_rms_mae_db": 2.386118389404549, "rms_envelope_nrmse": 0.28209043596432926, "envelope_correlation": 0.929719851809578, "rms_gain_error_db": 0.07357127134375181, "reference_mean_power": 3.136903007980474e-05, "output_mean_power": 3.190496097870255e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9944851157580815e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0899878382466362e-05, "output_power": 1.121708939316205e-05, "reference_active": true, "delta_db": 0.12458528700423163}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5519027988888207e-06, "output_power": 1.9505793865237408e-06, "reference_active": true, "delta_db": -1.167004978999328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4088478818809383e-06, "output_power": 8.12861142088094e-07, "reference_active": true, "delta_db": -2.3884774029136215}], "active_band_count": 3, "active_band_mae_db": 1.2266892229723938}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9738091980327894e-05, "log_rms_mae_db": 1.6516175860787532, "rms_envelope_nrmse": 0.18034260665322155, "envelope_correlation": 0.9679323595173035, "rms_gain_error_db": 0.0737496332892853, "reference_mean_power": 3.151908471872331e-05, "output_mean_power": 3.205889586933214e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7532611532165288e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0780259770731776e-05, "output_power": 1.080193210883781e-05, "reference_active": true, "delta_db": 0.008722172488846757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1761748042676738e-06, "output_power": 8.714154159454033e-07, "reference_active": true, "delta_db": -1.3024663317481733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.367404555815231e-08, "output_power": 4.8090725701638027e-08, "reference_active": true, "delta_db": -2.895579460351636}], "active_band_count": 3, "active_band_mae_db": 1.4022559881962187}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8875298355211437, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f2dd8ba17407e02bef83046adc725c47c567e00c71a92b518080ddb41489e741", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9734968776852448, "standard_deviation": 0.003416527311321217, "minimum": 0.9663255502494054, "maximum": 0.9770672445050964, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.977608632894241, "standard_deviation": 0.04083728968104113, "minimum": 1.929434000610556, "maximum": 2.037471401566601, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.002580509373539, "standard_deviation": 0.3115578903565479, "minimum": 0.4364445360904277, "maximum": 1.4334731259421047, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18922769031112616, "standard_deviation": 0.011448484356608203, "minimum": 0.1767456078930179, "maximum": 0.21289181225090278, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9895415799763412, "standard_deviation": 0.000975199889531611, "minimum": 0.9876578594772789, "maximum": 0.9908314515182134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3378695804955592, "standard_deviation": 0.032157731739333624, "minimum": 1.296271048284648, "maximum": 1.376058070514034, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2433269947019119, "standard_deviation": 0.3797929774461217, "minimum": 0.5534096194533109, "maximum": 1.73547494206864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1160179045681363, "standard_deviation": 0.005250391294804814, "minimum": 0.109526178094982, "maximum": 0.12656592723809748, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9373110575753183, "standard_deviation": 0.006333627071775482, "minimum": 0.9278201542130876, "maximum": 0.944404182209383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3786486471020494, "standard_deviation": 0.02432655811331501, "minimum": 2.332861659857867, "maximum": 2.4138106534870754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2013739994142374, "standard_deviation": 0.18124194168976007, "minimum": 0.8385045775127072, "maximum": 1.4047707594179244, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2669038018099763, "standard_deviation": 0.013759726700162756, "minimum": 0.25050525711420296, "maximum": 0.28601007712644644, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9713313465377789, "standard_deviation": 0.0023315213552019772, "minimum": 0.9679323595173035, "maximum": 0.9753546898206121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.628691520503069, "standard_deviation": 0.04229029982000874, "minimum": 1.578704306412171, "maximum": 1.7050221425034748, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3579567129422987, "standard_deviation": 0.27739946300379903, "minimum": 0.7721124997005999, "maximum": 1.673777020536143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.171144117917309, "standard_deviation": 0.006805796836632348, "minimum": 0.16015034181894947, "maximum": 0.18034260665322155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933021300308771, "standard_deviation": 0.0011254299108558366, "minimum": 0.9911653538067764, "maximum": 0.9947432017028772, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3960278825172923, "standard_deviation": 0.04715237111330421, "minimum": 1.3258114951874447, "maximum": 1.4642193420613947, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47212715666647315, "standard_deviation": 0.12032719590414362, "minimum": 0.25200200927177663, "maximum": 0.6126753698770572, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09305990281580281, "standard_deviation": 0.007874310059129006, "minimum": 0.0826629078436214, "maximum": 0.10860005001070189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977677481754861, "standard_deviation": 0.0006127133663938549, "minimum": 0.9965178600187634, "maximum": 0.9983480204540041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8274824653388129, "standard_deviation": 0.04298246579392684, "minimum": 0.7720318363531835, "maximum": 0.9011839402379704, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6184247620750722, "standard_deviation": 0.19531003066968516, "minimum": 0.24335202008131948, "maximum": 0.8234408015882231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05422679839772157, "standard_deviation": 0.00721482499935071, "minimum": 0.048460441625616024, "maximum": 0.07031550818203423, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11962888037097401, "standard_deviation": 0.03259892239314531, "minimum": 0.07937984033333548, "maximum": 0.16967641724559754, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.897757680149722, "standard_deviation": 0.008416846201945946, "minimum": 5.888548734923508, "maximum": 5.9099291601481765, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5251416026309483, "standard_deviation": 0.029507273908725453, "minimum": 0.4730326286698824, "maximum": 0.5540919225567829, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10334648831748389, "standard_deviation": 0.042833791233906904, "minimum": -0.18342745011402511, "maximum": -0.04289837650328915, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945508631046048, "standard_deviation": 0.000985235088183238, "minimum": 0.9924345205965588, "maximum": 0.9954910428241082, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8473776035435925, "standard_deviation": 0.030815113771122078, "minimum": 0.8004539327992027, "maximum": 0.8898894499836484, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3995471604945293, "standard_deviation": 0.1168653200027977, "minimum": 0.22924196079756654, "maximum": 0.5701129229558219, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07342656002001059, "standard_deviation": 0.006561625583269255, "minimum": 0.066757548455494, "maximum": 0.08761989183564098, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982449045336604, "standard_deviation": 0.0005148727455207397, "minimum": 0.9971594722397127, "maximum": 0.9987637501058904, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45276760688817724, "standard_deviation": 0.024717043841713124, "minimum": 0.41824458542034915, "maximum": 0.4944343356367747, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5103494908490365, "standard_deviation": 0.16081336592522852, "minimum": 0.31340617096086687, "maximum": 0.7421179640464629, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04114163925980545, "standard_deviation": 0.0061143775665971145, "minimum": 0.03620751989782693, "maximum": 0.05420005768488613, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0028", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.916451930999756, "predicted_mos": 2.916451930999756, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4854abdeb9ae5acc88ba3175e0a0d70ef9084db45057a2dd5e800f7bbac3393c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.411318500060588, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8742242958950402, "全帯域_10ms_rms_envelope_nrmse": 0.07049697554669894, "全帯域_10ms_envelope_correlation": 0.9950322707643279, "全帯域_10ms_modulation_mae_db": 0.3249026884081955, "全帯域_40ms_log_rms_mae_db": 0.4824213248439855, "全帯域_40ms_rms_envelope_nrmse": 0.04116655227463171, "全帯域_40ms_envelope_correlation": 0.998287207011334, "全帯域_40ms_modulation_mae_db": 0.37426613168330186, "300–1000 Hz_10ms_log_rms_mae_db": 1.1953872408173498, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08912511481474923, "300–1000 Hz_10ms_envelope_correlation": 0.9939574063544762, "300–1000 Hz_10ms_modulation_mae_db": 0.42382545209770256, "300–1000 Hz_40ms_log_rms_mae_db": 0.6997620681429743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05052270850914484, "300–1000 Hz_40ms_envelope_correlation": 0.9982097785531101, "300–1000 Hz_40ms_modulation_mae_db": 0.4751550913696172, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5618300276084187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17744434724777636, "1000–3000 Hz_10ms_envelope_correlation": 0.9776121052178741, "1000–3000 Hz_10ms_modulation_mae_db": 0.38522300859007436, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0962854009175984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11440216331496349, "1000–3000 Hz_40ms_envelope_correlation": 0.9901373192922035, "1000–3000 Hz_40ms_modulation_mae_db": 0.6941873189011586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.672777246061019, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21460642957847006, "3000–6000 Hz_10ms_envelope_correlation": 0.9588381085341889, "3000–6000 Hz_10ms_modulation_mae_db": 1.1360992673959678, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1578050383132983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14947808088579007, "3000–6000 Hz_40ms_envelope_correlation": 0.9784070962825633, "3000–6000 Hz_40ms_modulation_mae_db": 1.1730546831987776, "mr_stft_db": 5.476755474185942, "band_mae_db": 0.21812934924491298, "rms_gain_error_db": -0.12011904223367201, "output_peak": 0.5116810231602343}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8904354825302807, "全帯域_10ms_rms_envelope_nrmse": 0.06366480407903934, "全帯域_10ms_envelope_correlation": 0.9960117870289188, "全帯域_10ms_modulation_mae_db": 0.3214285416344446, "全帯域_40ms_log_rms_mae_db": 0.4882070539378111, "全帯域_40ms_rms_envelope_nrmse": 0.034565019033353975, "全帯域_40ms_envelope_correlation": 0.9988903109565722, "全帯域_40ms_modulation_mae_db": 0.18621840374501195, "300–1000 Hz_10ms_log_rms_mae_db": 1.235564371402935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08548602012736109, "300–1000 Hz_10ms_envelope_correlation": 0.9945167277820806, "300–1000 Hz_10ms_modulation_mae_db": 0.35376594895880475, "300–1000 Hz_40ms_log_rms_mae_db": 0.720717823754558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04954244369798139, "300–1000 Hz_40ms_envelope_correlation": 0.9983469121553223, "300–1000 Hz_40ms_modulation_mae_db": 0.14753489306463471, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5159091139616707, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16596772155929296, "1000–3000 Hz_10ms_envelope_correlation": 0.9797020606453106, "1000–3000 Hz_10ms_modulation_mae_db": 0.4862651798818323, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0431101074693006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1088826567895723, "1000–3000 Hz_40ms_envelope_correlation": 0.9905602704917645, "1000–3000 Hz_40ms_modulation_mae_db": 0.6027301942004336, "3000–6000 Hz_10ms_log_rms_mae_db": 1.698357707983286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21724413252865343, "3000–6000 Hz_10ms_envelope_correlation": 0.9583694995464929, "3000–6000 Hz_10ms_modulation_mae_db": 0.8356661236278429, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1935560769219848, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14893949367904796, "3000–6000 Hz_40ms_envelope_correlation": 0.978077501976798, "3000–6000 Hz_40ms_modulation_mae_db": 0.8264446280094172, "mr_stft_db": 5.472646348246933, "band_mae_db": 0.1500982124408381, "rms_gain_error_db": -0.12956296569017814, "output_peak": 0.557157618625113}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8835036879800277, "全帯域_10ms_rms_envelope_nrmse": 0.07353420741255358, "全帯域_10ms_envelope_correlation": 0.9945017078326471, "全帯域_10ms_modulation_mae_db": 0.378436105893891, "全帯域_40ms_log_rms_mae_db": 0.4755405986719503, "全帯域_40ms_rms_envelope_nrmse": 0.041193992273228695, "全帯域_40ms_envelope_correlation": 0.9981863292191735, "全帯域_40ms_modulation_mae_db": 0.3462818005314885, "300–1000 Hz_10ms_log_rms_mae_db": 1.1583433933685912, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09154680464444205, "300–1000 Hz_10ms_envelope_correlation": 0.9934468987925675, "300–1000 Hz_10ms_modulation_mae_db": 0.1876831961007601, "300–1000 Hz_40ms_log_rms_mae_db": 0.6476562543991737, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05173402795828672, "300–1000 Hz_40ms_envelope_correlation": 0.9978999792036944, "300–1000 Hz_40ms_modulation_mae_db": 0.3703750762467168, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5023072533896682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1762954372866293, "1000–3000 Hz_10ms_envelope_correlation": 0.9784481654263248, "1000–3000 Hz_10ms_modulation_mae_db": 0.20607165799814944, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0215969030835532, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11402413522701438, "1000–3000 Hz_40ms_envelope_correlation": 0.9906445557332031, "1000–3000 Hz_40ms_modulation_mae_db": 0.29229973765275913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6813479658438115, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22359030397763516, "3000–6000 Hz_10ms_envelope_correlation": 0.9551275197380176, "3000–6000 Hz_10ms_modulation_mae_db": 1.033887665789573, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2055106052725773, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14900670366711932, "3000–6000 Hz_40ms_envelope_correlation": 0.9779288086533087, "3000–6000 Hz_40ms_modulation_mae_db": 1.1490031241769607, "mr_stft_db": 5.475741860816107, "band_mae_db": 0.21585255471239886, "rms_gain_error_db": -0.08211963240814744, "output_peak": 0.5126386528440918}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.898477269972758, "全帯域_10ms_rms_envelope_nrmse": 0.07445720504329716, "全帯域_10ms_envelope_correlation": 0.9944881682626884, "全帯域_10ms_modulation_mae_db": 0.45570047116243345, "全帯域_40ms_log_rms_mae_db": 0.5145245648132494, "全帯域_40ms_rms_envelope_nrmse": 0.040814710459762725, "全帯域_40ms_envelope_correlation": 0.9983921290836378, "全帯域_40ms_modulation_mae_db": 0.7240156553915572, "300–1000 Hz_10ms_log_rms_mae_db": 1.259448244067041, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09190069454717258, "300–1000 Hz_10ms_envelope_correlation": 0.9936145005557404, "300–1000 Hz_10ms_modulation_mae_db": 0.511339936853849, "300–1000 Hz_40ms_log_rms_mae_db": 0.7561722209543913, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05016051428562208, "300–1000 Hz_40ms_envelope_correlation": 0.998316188377778, "300–1000 Hz_40ms_modulation_mae_db": 0.9208195868850728, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5636809251115495, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18522119678943957, "1000–3000 Hz_10ms_envelope_correlation": 0.974894871548163, "1000–3000 Hz_10ms_modulation_mae_db": 0.6512209193049204, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1080851909167906, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11189719887262306, "1000–3000 Hz_40ms_envelope_correlation": 0.9901149407768576, "1000–3000 Hz_40ms_modulation_mae_db": 0.7426066333390668, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6444376230323436, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2071428673804587, "3000–6000 Hz_10ms_envelope_correlation": 0.9615946494427074, "3000–6000 Hz_10ms_modulation_mae_db": 0.926813243297285, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1762594011681826, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13834491931384227, "3000–6000 Hz_40ms_envelope_correlation": 0.981233476425788, "3000–6000 Hz_40ms_modulation_mae_db": 1.147372087065148, "mr_stft_db": 5.482465911850837, "band_mae_db": 0.17977254958736077, "rms_gain_error_db": -0.14076920473943552, "output_peak": 0.5483076783505709}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9258225191324512, "全帯域_10ms_rms_envelope_nrmse": 0.0705477430984418, "全帯域_10ms_envelope_correlation": 0.995357771465423, "全帯域_10ms_modulation_mae_db": 0.5077672510803737, "全帯域_40ms_log_rms_mae_db": 0.5097130788980437, "全帯域_40ms_rms_envelope_nrmse": 0.04268827584291766, "全帯域_40ms_envelope_correlation": 0.9985290325000634, "全帯域_40ms_modulation_mae_db": 0.6368976440658948, "300–1000 Hz_10ms_log_rms_mae_db": 1.1441676120945248, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08908437477148594, "300–1000 Hz_10ms_envelope_correlation": 0.9942744547397349, "300–1000 Hz_10ms_modulation_mae_db": 0.5533836364707417, "300–1000 Hz_40ms_log_rms_mae_db": 0.6542234643388126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05568607191285574, "300–1000 Hz_40ms_envelope_correlation": 0.9980863712539721, "300–1000 Hz_40ms_modulation_mae_db": 0.715613929303183, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5185429791802727, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1842674326323586, "1000–3000 Hz_10ms_envelope_correlation": 0.9754093153756466, "1000–3000 Hz_10ms_modulation_mae_db": 0.2735304753763457, "1000–3000 Hz_40ms_log_rms_mae_db": 1.059789243090628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13327450533894214, "1000–3000 Hz_40ms_envelope_correlation": 0.9870337060314475, "1000–3000 Hz_40ms_modulation_mae_db": 0.4725935972140716, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6963352399994711, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21792065039311437, "3000–6000 Hz_10ms_envelope_correlation": 0.9585508764345722, "3000–6000 Hz_10ms_modulation_mae_db": 0.7786794565179972, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2078853831289107, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14689396340549865, "3000–6000 Hz_40ms_envelope_correlation": 0.9789157574976696, "3000–6000 Hz_40ms_modulation_mae_db": 0.9576343160892714, "mr_stft_db": 5.488867888502154, "band_mae_db": 0.1559762700063449, "rms_gain_error_db": -0.19470092310637832, "output_peak": 0.5207522826140839}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8729587123216732, "全帯域_10ms_rms_envelope_nrmse": 0.07630307245798444, "全帯域_10ms_envelope_correlation": 0.994134088204676, "全帯域_10ms_modulation_mae_db": 0.3077099034140522, "全帯域_40ms_log_rms_mae_db": 0.498241894986021, "全帯域_40ms_rms_envelope_nrmse": 0.04748619054482347, "全帯域_40ms_envelope_correlation": 0.9976275718932557, "全帯域_40ms_modulation_mae_db": 0.5360605073443323, "300–1000 Hz_10ms_log_rms_mae_db": 1.137394667380988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09268240665325421, "300–1000 Hz_10ms_envelope_correlation": 0.9934080028445326, "300–1000 Hz_10ms_modulation_mae_db": 0.604139548641727, "300–1000 Hz_40ms_log_rms_mae_db": 0.6861989347118571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06165289739169436, "300–1000 Hz_40ms_envelope_correlation": 0.9970549940069873, "300–1000 Hz_40ms_modulation_mae_db": 0.8584786432638211, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5101911037792053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15239810968549383, "1000–3000 Hz_10ms_envelope_correlation": 0.9829291706506611, "1000–3000 Hz_10ms_modulation_mae_db": 0.2636848709559921, "1000–3000 Hz_40ms_log_rms_mae_db": 1.03056873539117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09423770872468655, "1000–3000 Hz_40ms_envelope_correlation": 0.9929719260034379, "1000–3000 Hz_40ms_modulation_mae_db": 0.5065006680709411, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6669571337864175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2171605832315759, "3000–6000 Hz_10ms_envelope_correlation": 0.9579984010269513, "3000–6000 Hz_10ms_modulation_mae_db": 1.179851402306592, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1773242606216427, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16233406801299524, "3000–6000 Hz_40ms_envelope_correlation": 0.9745908290110902, "3000–6000 Hz_40ms_modulation_mae_db": 1.426741979909048, "mr_stft_db": 5.47455951439438, "band_mae_db": 0.18537542437622578, "rms_gain_error_db": -0.11388148372911372, "output_peak": 0.5587149944683031}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8494062592066673, "全帯域_10ms_rms_envelope_nrmse": 0.0767219973726459, "全帯域_10ms_envelope_correlation": 0.9940949760513977, "全帯域_10ms_modulation_mae_db": 0.269623667575145, "全帯域_40ms_log_rms_mae_db": 0.44333047009239857, "全帯域_40ms_rms_envelope_nrmse": 0.049720555980450526, "全帯域_40ms_envelope_correlation": 0.9974192594924518, "全帯域_40ms_modulation_mae_db": 0.32003798268707845, "300–1000 Hz_10ms_log_rms_mae_db": 1.1262862684860013, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09306761626689905, "300–1000 Hz_10ms_envelope_correlation": 0.9933172926434323, "300–1000 Hz_10ms_modulation_mae_db": 0.31746222926153145, "300–1000 Hz_40ms_log_rms_mae_db": 0.6621306610194193, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05970059706040663, "300–1000 Hz_40ms_envelope_correlation": 0.9972377227870901, "300–1000 Hz_40ms_modulation_mae_db": 0.5527263016095457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5366474052483363, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20214733550632732, "1000–3000 Hz_10ms_envelope_correlation": 0.9697041957373645, "1000–3000 Hz_10ms_modulation_mae_db": 1.2025920418148002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0576729303373102, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1399068980320635, "1000–3000 Hz_40ms_envelope_correlation": 0.9848776353731794, "1000–3000 Hz_40ms_modulation_mae_db": 1.2742777822791853, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7134665585461901, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22938535769101717, "3000–6000 Hz_10ms_envelope_correlation": 0.9542789385841155, "3000–6000 Hz_10ms_modulation_mae_db": 0.4438902290361842, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1974124096175771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15720149109568737, "3000–6000 Hz_40ms_envelope_correlation": 0.9759263089028948, "3000–6000 Hz_40ms_modulation_mae_db": 0.5480020836639655, "mr_stft_db": 5.489857290931326, "band_mae_db": 0.09547663577925804, "rms_gain_error_db": -0.116038580027769, "output_peak": 0.5098120005416217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.860845090749244, "全帯域_10ms_rms_envelope_nrmse": 0.07546094936902455, "全帯域_10ms_envelope_correlation": 0.9946202516982786, "全帯域_10ms_modulation_mae_db": 0.33964693855136324, "全帯域_40ms_log_rms_mae_db": 0.49016619013326773, "全帯域_40ms_rms_envelope_nrmse": 0.04710340709658276, "全帯域_40ms_envelope_correlation": 0.9980964160785102, "全帯域_40ms_modulation_mae_db": 0.4041778918459742, "300–1000 Hz_10ms_log_rms_mae_db": 1.1865458833187146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09753974188059426, "300–1000 Hz_10ms_envelope_correlation": 0.9931328158476894, "300–1000 Hz_10ms_modulation_mae_db": 0.42739842760899743, "300–1000 Hz_40ms_log_rms_mae_db": 0.7210100374374184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06121736193129735, "300–1000 Hz_40ms_envelope_correlation": 0.9976861023741581, "300–1000 Hz_40ms_modulation_mae_db": 0.5824287629237451, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5028620593333966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16851160609498944, "1000–3000 Hz_10ms_envelope_correlation": 0.9789901381340752, "1000–3000 Hz_10ms_modulation_mae_db": 0.5568998135018534, "1000–3000 Hz_40ms_log_rms_mae_db": 1.03534889458785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09891889704881511, "1000–3000 Hz_40ms_envelope_correlation": 0.9925065831315317, "1000–3000 Hz_40ms_modulation_mae_db": 0.8270545737117786, "3000–6000 Hz_10ms_log_rms_mae_db": 1.702839183900046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21499489902061686, "3000–6000 Hz_10ms_envelope_correlation": 0.9585581854050768, "3000–6000 Hz_10ms_modulation_mae_db": 1.0228621389384924, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1758069583744932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13977505131863846, "3000–6000 Hz_40ms_envelope_correlation": 0.9808000756127101, "3000–6000 Hz_40ms_modulation_mae_db": 1.1924666043706518, "mr_stft_db": 5.463039539359187, "band_mae_db": 0.15939986540706583, "rms_gain_error_db": -0.19290296011043817, "output_peak": 0.5446409073893256}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.391843216287546, "spectral_convergence": 0.2067515227821972, "spectral_convergence_status": "ok", "amplitude_floor": 1.8586249352090494e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.363304541480717, "spectral_convergence": 0.2009492261827808, "spectral_convergence_status": "ok", "amplitude_floor": 1.6686015821705495e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6751186647895615, "spectral_convergence": 0.2078884478588044, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267872572435527e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003360135402914431, "output_power": 0.0003435799865148924, "delta_db": 0.0967107985696581, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011242144422279699, "output_power": 0.0011128804186873202, "delta_db": -0.044006591620786475, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015417458345753522, "output_power": 0.0014558643374259503, "delta_db": -0.24891875970423002, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.127065821964785e-05, "output_power": 8.936900699107667e-05, "delta_db": 0.4125315520632018, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.295885949753132e-05, "output_power": 4.92720550229753e-05, "delta_db": -0.31337947659744575, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.948150516097124e-05, "output_power": 1.8633722184584808e-05, "delta_db": -0.1932289169141557, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21812934924491298, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12011904223367201, "pitch": null, "output_rms_dbfs": -25.129182807554415, "output_peak": 0.5116810231602343, "reference_rms_dbfs": -25.009063765320747, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.476755474185942}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025375764068356454, "log_rms_mae_db": 0.8742242958950402, "rms_envelope_nrmse": 0.07049697554669894, "envelope_correlation": 0.9950322707643279, "rms_gain_error_db": -0.12011794179820089, "reference_mean_power": 0.0031583120304531104, "output_mean_power": 0.003072155827460647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1807904078754825e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008327873716353349, "output_power": 0.0008006113356827404, "reference_active": true, "delta_db": -0.17112396028646548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013733244323789595, "output_power": 0.00012974136666979075, "reference_active": true, "delta_db": -0.24694678131187905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.747498790065721e-05, "output_power": 1.537280616998976e-05, "reference_active": true, "delta_db": -0.5566373236262419}], "active_band_count": 3, "active_band_mae_db": 0.3249026884081955}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002402786371599188, "log_rms_mae_db": 0.4824213248439855, "rms_envelope_nrmse": 0.04116655227463171, "envelope_correlation": 0.998287207011334, "rms_gain_error_db": -0.12012006126149222, "reference_mean_power": 0.0031741738524038612, "output_mean_power": 0.003087583444894679}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0384864838590044e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000782161356346486, "output_power": 0.0007520132581904505, "reference_active": true, "delta_db": -0.17070857903274633}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.85949543312588e-05, "output_power": 5.7671111496966544e-05, "reference_active": true, "delta_db": -0.06901898444781604}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3154862476080315e-06, "output_power": 1.0734435951824111e-06, "reference_active": true, "delta_db": -0.8830708315693432}], "active_band_count": 3, "active_band_mae_db": 0.37426613168330186}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002501327279402343, "log_rms_mae_db": 1.1953872408173498, "rms_envelope_nrmse": 0.08912511481474923, "envelope_correlation": 0.9939574063544762, "rms_gain_error_db": -0.19425768240536187, "reference_mean_power": 0.0021283926153372414, "output_mean_power": 0.002035288483496486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0698576130619868e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007463968445819582, "output_power": 0.0007022442093314515, "reference_active": true, "delta_db": -0.26481628028544085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014378870760237378, "output_power": 0.00013208342710052616, "reference_active": true, "delta_db": -0.3687645139977145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4173849116031066e-05, "output_power": 1.2237656226818694e-05, "reference_active": true, "delta_db": -0.6378955620099523}], "active_band_count": 3, "active_band_mae_db": 0.42382545209770256}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00023719519351520602, "log_rms_mae_db": 0.6997620681429743, "rms_envelope_nrmse": 0.05052270850914484, "envelope_correlation": 0.9982097785531101, "rms_gain_error_db": -0.1942575090086458, "reference_mean_power": 0.0021390960955203144, "output_mean_power": 0.0020455238336937187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.624199297765067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007211274534636877, "output_power": 0.0006773722940390724, "reference_active": true, "delta_db": -0.27184600443709206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.136224424017926e-05, "output_power": 5.8394250568670817e-05, "reference_active": true, "delta_db": -0.21531145880418254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3649293472151981e-06, "output_power": 1.0997130797845483e-06, "reference_active": true, "delta_db": -0.9383078108675771}], "active_band_count": 3, "active_band_mae_db": 0.4751550913696172}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.486004098842285e-05, "log_rms_mae_db": 1.5618300276084187, "rms_envelope_nrmse": 0.17744434724777636, "envelope_correlation": 0.9776121052178741, "rms_gain_error_db": 0.17893652790983522, "reference_mean_power": 9.73047769043704e-05, "output_mean_power": 0.0001013976322974234}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.517204404381221e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1360183676562794e-05, "output_power": 4.152811783138386e-05, "reference_active": true, "delta_db": 0.01759789384673461}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.821587959830417e-06, "output_power": 4.445717372733382e-06, "reference_active": true, "delta_db": -0.35248244098955356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2148111012630244e-06, "output_power": 1.8483228075346142e-06, "reference_active": true, "delta_db": -0.7855886909339349}], "active_band_count": 3, "active_band_mae_db": 0.38522300859007436}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5230962699738786e-05, "log_rms_mae_db": 1.0962854009175984, "rms_envelope_nrmse": 0.11440216331496349, "envelope_correlation": 0.9901373192922035, "rms_gain_error_db": 0.1789341746480475, "reference_mean_power": 9.77871242521371e-05, "output_mean_power": 0.00010190021303221102}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.922472441667658e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.88087102342988e-05, "output_power": 3.9182597890101725e-05, "reference_active": true, "delta_db": 0.04164017596669122}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.146701619692133e-06, "output_power": 1.9279299509526535e-06, "reference_active": true, "delta_db": -0.46680433688656736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.612101873452723e-07, "output_power": 1.121968691963594e-07, "reference_active": true, "delta_db": -1.574117443850217}], "active_band_count": 3, "active_band_mae_db": 0.6941873189011586}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.5953225979159985e-05, "log_rms_mae_db": 1.672777246061019, "rms_envelope_nrmse": 0.21460642957847006, "envelope_correlation": 0.9588381085341889, "rms_gain_error_db": -0.31425912451184934, "reference_mean_power": 3.136903007980474e-05, "output_mean_power": 2.9179320766596437e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9944851157580815e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0899878382466362e-05, "output_power": 9.773923032252695e-06, "reference_active": true, "delta_db": -0.4735273752083957}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5519027988888207e-06, "output_power": 2.2949150944189892e-06, "reference_active": true, "delta_db": -0.4609750577215941}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4088478818809383e-06, "output_power": 7.970481653477902e-07, "reference_active": true, "delta_db": -2.4737953692579135}], "active_band_count": 3, "active_band_mae_db": 1.1360992673959678}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9738091980327894e-05, "log_rms_mae_db": 1.1578050383132983, "rms_envelope_nrmse": 0.14947808088579007, "envelope_correlation": 0.9784070962825633, "rms_gain_error_db": -0.31442241392127096, "reference_mean_power": 3.151908471872331e-05, "output_mean_power": 2.9317798535277545e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7532611532165288e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0780259770731776e-05, "output_power": 9.537067848998662e-06, "reference_active": true, "delta_db": -0.5321435380206997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1761748042676738e-06, "output_power": 1.0637304156723815e-06, "reference_active": true, "delta_db": -0.43640294438384186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.367404555815231e-08, "output_power": 5.2066396348261684e-08, "reference_active": true, "delta_db": -2.5506175671917912}], "active_band_count": 3, "active_band_mae_db": 1.1730546831987776}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8972663850126276, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bceaadbff3b770d8854bb3a8884c046e931f076ba724863c9f6de08145c66bd4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9772112528419274, "standard_deviation": 0.0039410347287745055, "minimum": 0.9697041957373645, "maximum": 0.9829291706506611, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5264963584515647, "standard_deviation": 0.02483978416343409, "minimum": 1.5023072533896682, "maximum": 1.5636809251115495, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.503185995927996, "standard_deviation": 0.32216883039445876, "minimum": 0.20607165799814944, "maximum": 1.2025920418148002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1765316483502884, "standard_deviation": 0.014884067077931898, "minimum": 0.15239810968549383, "maximum": 0.20214733550632732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.989855867104203, "standard_deviation": 0.002692530705417177, "minimum": 0.9848776353731794, "maximum": 0.9929719260034379, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0565571757242753, "standard_deviation": 0.03110671475969049, "minimum": 1.0215969030835532, "maximum": 1.1080851909167906, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6765313131711743, "standard_deviation": 0.29476801939786823, "minimum": 0.29229973765275913, "maximum": 1.2742777822791853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11444302041858506, "standard_deviation": 0.015537341653507062, "minimum": 0.09423770872468655, "maximum": 0.1399068980320635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9579145223390153, "standard_deviation": 0.002280804628298089, "minimum": 0.9542789385841155, "maximum": 0.9615946494427074, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6845648323940732, "standard_deviation": 0.022575482950480445, "minimum": 1.6444376230323436, "maximum": 1.7134665585461901, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9197186908637418, "standard_deviation": 0.23651718893215545, "minimum": 0.4438902290361842, "maximum": 1.179851402306592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2177556529751927, "standard_deviation": 0.006547701427572605, "minimum": 0.2071428673804587, "maximum": 0.22938535769101717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9782349817953528, "standard_deviation": 0.0022294175158466345, "minimum": 0.9745908290110902, "maximum": 0.981233476425788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1864450166773335, "standard_deviation": 0.017379743822065967, "minimum": 1.1578050383132983, "maximum": 1.2078853831289107, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.052589938310405, "standard_deviation": 0.2691556425597643, "minimum": 0.5480020836639655, "maximum": 1.426741979909048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14899672142232742, "standard_deviation": 0.008003212033769405, "minimum": 0.13834491931384227, "maximum": 0.16233406801299524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937085124450318, "standard_deviation": 0.0004909926321769434, "minimum": 0.9931328158476894, "maximum": 0.9945167277820806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1803922101170181, "standard_deviation": 0.04800469612065681, "minimum": 1.1262862684860013, "maximum": 1.259448244067041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4223747969992643, "standard_deviation": 0.13571916738220827, "minimum": 0.1876831961007601, "maximum": 0.604139548641727, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0913040967132448, "standard_deviation": 0.003542254912323445, "minimum": 0.08548602012736109, "maximum": 0.09753974188059426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978547560890141, "standard_deviation": 0.0004907228741555645, "minimum": 0.9970549940069873, "maximum": 0.9983469121553223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6934839330948256, "standard_deviation": 0.03807186259100809, "minimum": 0.6476562543991737, "maximum": 0.7561722209543913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5778915355832921, "standard_deviation": 0.25488787567314486, "minimum": 0.14753489306463471, "maximum": 0.9208195868850728, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05502707784341114, "standard_deviation": 0.005201583962697594, "minimum": 0.04954244369798139, "maximum": 0.06165289739169436, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17001010769430067, "standard_deviation": 0.03965861000807042, "minimum": 0.09547663577925804, "maximum": 0.21812934924491298, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.477991728535859, "standard_deviation": 0.008860926410809352, "minimum": 5.463039539359187, "maximum": 5.489857290931326, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5329631447491681, "standard_deviation": 0.02128530507362649, "minimum": 0.5098120005416217, "maximum": 0.5587149944683031, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13626184900564153, "standard_deviation": 0.03926169221473446, "minimum": -0.19470092310637832, "maximum": -0.08211963240814744, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947801276635447, "standard_deviation": 0.0006534907617373996, "minimum": 0.9940949760513977, "maximum": 0.9960117870289188, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8819591647235179, "standard_deviation": 0.0236564919911504, "minimum": 0.8494062592066673, "maximum": 0.9258225191324512, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36315194596498734, "standard_deviation": 0.0804364114613425, "minimum": 0.269623667575145, "maximum": 0.5077672510803737, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07264836929746071, "standard_deviation": 0.004336952073246435, "minimum": 0.06366480407903934, "maximum": 0.0767219973726459, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981785320293748, "standard_deviation": 0.0004741563119180302, "minimum": 0.9974192594924518, "maximum": 0.9988903109565722, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4877681470470909, "standard_deviation": 0.022251603959772392, "minimum": 0.44333047009239857, "maximum": 0.5145245648132494, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4409945021618299, "standard_deviation": 0.1780194746668617, "minimum": 0.18621840374501195, "maximum": 0.7240156553915572, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04309233793821894, "standard_deviation": 0.004850718333191459, "minimum": 0.034565019033353975, "maximum": 0.049720555980450526, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0028", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.991077184677124, "predicted_mos": 2.991077184677124, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "35e8e087a33d18052dd435e5b9cdb134e91071ed0cd326afd190ca5fa2b21bc8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.340354800224304, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8974960460830801, "全帯域_10ms_rms_envelope_nrmse": 0.07897341509807031, "全帯域_10ms_envelope_correlation": 0.9936293620317979, "全帯域_10ms_modulation_mae_db": 0.23974204322936618, "全帯域_40ms_log_rms_mae_db": 0.5006386954719368, "全帯域_40ms_rms_envelope_nrmse": 0.04542199836848625, "全帯域_40ms_envelope_correlation": 0.9977530593466659, "全帯域_40ms_modulation_mae_db": 0.3070933650685606, "300–1000 Hz_10ms_log_rms_mae_db": 1.2412645956034742, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09901202659649239, "300–1000 Hz_10ms_envelope_correlation": 0.992230008638375, "300–1000 Hz_10ms_modulation_mae_db": 0.20215049418924813, "300–1000 Hz_40ms_log_rms_mae_db": 0.7321627422356418, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05864233932798882, "300–1000 Hz_40ms_envelope_correlation": 0.9971403412510679, "300–1000 Hz_40ms_modulation_mae_db": 0.4289022487088742, "1000–3000 Hz_10ms_log_rms_mae_db": 1.649036038904572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17718163858455885, "1000–3000 Hz_10ms_envelope_correlation": 0.976831869536925, "1000–3000 Hz_10ms_modulation_mae_db": 0.29501555032027044, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1205874433300904, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1187898871643647, "1000–3000 Hz_40ms_envelope_correlation": 0.9891849813278658, "1000–3000 Hz_40ms_modulation_mae_db": 0.23349542689744082, "3000–6000 Hz_10ms_log_rms_mae_db": 1.926555828886782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22236239527355764, "3000–6000 Hz_10ms_envelope_correlation": 0.9557586983296079, "3000–6000 Hz_10ms_modulation_mae_db": 1.4104636519960965, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3255369962273988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14304928590590657, "3000–6000 Hz_40ms_envelope_correlation": 0.979590204213012, "3000–6000 Hz_40ms_modulation_mae_db": 1.7518226185780381, "mr_stft_db": 5.6178250998440875, "band_mae_db": 0.09464594237202241, "rms_gain_error_db": -0.08885313354097144, "output_peak": 0.5508650945923662}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8834666474590775, "全帯域_10ms_rms_envelope_nrmse": 0.07100677311366256, "全帯域_10ms_envelope_correlation": 0.9948579649499542, "全帯域_10ms_modulation_mae_db": 0.434503231392392, "全帯域_40ms_log_rms_mae_db": 0.47066716588002483, "全帯域_40ms_rms_envelope_nrmse": 0.03997535841450657, "全帯域_40ms_envelope_correlation": 0.9982672032591451, "全帯域_40ms_modulation_mae_db": 0.617391410294846, "300–1000 Hz_10ms_log_rms_mae_db": 1.2864968668073322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0938018470518888, "300–1000 Hz_10ms_envelope_correlation": 0.9930458973346488, "300–1000 Hz_10ms_modulation_mae_db": 0.5162057104965599, "300–1000 Hz_40ms_log_rms_mae_db": 0.7618068006433545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052585847594894444, "300–1000 Hz_40ms_envelope_correlation": 0.9977314297795064, "300–1000 Hz_40ms_modulation_mae_db": 0.8325531897634139, "1000–3000 Hz_10ms_log_rms_mae_db": 1.620669592282862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1678478598920509, "1000–3000 Hz_10ms_envelope_correlation": 0.9799263407411863, "1000–3000 Hz_10ms_modulation_mae_db": 0.3685316199719834, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1007545200131015, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11186759520264616, "1000–3000 Hz_40ms_envelope_correlation": 0.9905361591463481, "1000–3000 Hz_40ms_modulation_mae_db": 0.48564460798146875, "3000–6000 Hz_10ms_log_rms_mae_db": 1.954260625412146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25625228440361125, "3000–6000 Hz_10ms_envelope_correlation": 0.9407904286931495, "3000–6000 Hz_10ms_modulation_mae_db": 1.7209867381696544, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4147102268948946, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18054317599421152, "3000–6000 Hz_40ms_envelope_correlation": 0.9685704834252465, "3000–6000 Hz_40ms_modulation_mae_db": 1.7840950436360181, "mr_stft_db": 5.606051875953326, "band_mae_db": 0.15718985251166426, "rms_gain_error_db": -0.07965300321491231, "output_peak": 0.5378756524010668}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8808931572650429, "全帯域_10ms_rms_envelope_nrmse": 0.08533123995298299, "全帯域_10ms_envelope_correlation": 0.9925637347494689, "全帯域_10ms_modulation_mae_db": 0.2954201026283614, "全帯域_40ms_log_rms_mae_db": 0.46873597862399924, "全帯域_40ms_rms_envelope_nrmse": 0.04118817026097948, "全帯域_40ms_envelope_correlation": 0.9982109974648959, "全帯域_40ms_modulation_mae_db": 0.5954009745214978, "300–1000 Hz_10ms_log_rms_mae_db": 1.2247464956312744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10616341926306708, "300–1000 Hz_10ms_envelope_correlation": 0.9911842625762115, "300–1000 Hz_10ms_modulation_mae_db": 0.29950871221424563, "300–1000 Hz_40ms_log_rms_mae_db": 0.7336723088381641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05457115034000915, "300–1000 Hz_40ms_envelope_correlation": 0.9977652115893412, "300–1000 Hz_40ms_modulation_mae_db": 0.5284555335596742, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7054754882328613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17165547867092246, "1000–3000 Hz_10ms_envelope_correlation": 0.9803776753385716, "1000–3000 Hz_10ms_modulation_mae_db": 0.586777494841556, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2008779606613582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1192793818407932, "1000–3000 Hz_40ms_envelope_correlation": 0.9904370881746558, "1000–3000 Hz_40ms_modulation_mae_db": 0.6557699425881895, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9570701030564979, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23568109113348476, "3000–6000 Hz_10ms_envelope_correlation": 0.950371546817913, "3000–6000 Hz_10ms_modulation_mae_db": 1.0155412678503009, "3000–6000 Hz_40ms_log_rms_mae_db": 1.357379036930587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1535171560688036, "3000–6000 Hz_40ms_envelope_correlation": 0.9764059459421863, "3000–6000 Hz_40ms_modulation_mae_db": 1.0477449778834405, "mr_stft_db": 5.62971417663645, "band_mae_db": 0.17847994227799946, "rms_gain_error_db": -0.10144083085289592, "output_peak": 0.6064329087047391}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8568671039873714, "全帯域_10ms_rms_envelope_nrmse": 0.07078079211368295, "全帯域_10ms_envelope_correlation": 0.9949337907928314, "全帯域_10ms_modulation_mae_db": 0.15715741927105153, "全帯域_40ms_log_rms_mae_db": 0.47707338833694557, "全帯域_40ms_rms_envelope_nrmse": 0.036074038995803324, "全帯域_40ms_envelope_correlation": 0.9986840137804555, "全帯域_40ms_modulation_mae_db": 0.30841215053226095, "300–1000 Hz_10ms_log_rms_mae_db": 1.2085336591275886, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08477987413169454, "300–1000 Hz_10ms_envelope_correlation": 0.9944522099683737, "300–1000 Hz_10ms_modulation_mae_db": 0.24707934260407438, "300–1000 Hz_40ms_log_rms_mae_db": 0.6836865953759078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04529752402070213, "300–1000 Hz_40ms_envelope_correlation": 0.998506149723659, "300–1000 Hz_40ms_modulation_mae_db": 0.45006365797247927, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6601645120349198, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17681863440518533, "1000–3000 Hz_10ms_envelope_correlation": 0.9768786011345877, "1000–3000 Hz_10ms_modulation_mae_db": 0.7447516662795465, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1318358447157832, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.116950316122467, "1000–3000 Hz_40ms_envelope_correlation": 0.9891525375822859, "1000–3000 Hz_40ms_modulation_mae_db": 0.8340166110378933, "3000–6000 Hz_10ms_log_rms_mae_db": 1.923971289345991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22900416647100122, "3000–6000 Hz_10ms_envelope_correlation": 0.9543625571331592, "3000–6000 Hz_10ms_modulation_mae_db": 0.7327595772538529, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3266935578127408, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15169026776545066, "3000–6000 Hz_40ms_envelope_correlation": 0.9775843330391824, "3000–6000 Hz_40ms_modulation_mae_db": 0.7940853159196312, "mr_stft_db": 5.596904459395422, "band_mae_db": 0.0733251634125393, "rms_gain_error_db": -0.09301118921403254, "output_peak": 0.6243206481767642}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8787688950962654, "全帯域_10ms_rms_envelope_nrmse": 0.07561545750541378, "全帯域_10ms_envelope_correlation": 0.9941312641544141, "全帯域_10ms_modulation_mae_db": 0.1674936448120369, "全帯域_40ms_log_rms_mae_db": 0.4941303489170185, "全帯域_40ms_rms_envelope_nrmse": 0.04166247817028476, "全帯域_40ms_envelope_correlation": 0.9980725332215166, "全帯域_40ms_modulation_mae_db": 0.13103964977030635, "300–1000 Hz_10ms_log_rms_mae_db": 1.36229082434418, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0935700136198913, "300–1000 Hz_10ms_envelope_correlation": 0.9930675081094246, "300–1000 Hz_10ms_modulation_mae_db": 0.3555168606759356, "300–1000 Hz_40ms_log_rms_mae_db": 0.7784850216904871, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05042513618465176, "300–1000 Hz_40ms_envelope_correlation": 0.9979210817761501, "300–1000 Hz_40ms_modulation_mae_db": 0.12354284909357194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6832233209878855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17056312194674042, "1000–3000 Hz_10ms_envelope_correlation": 0.9788818675137663, "1000–3000 Hz_10ms_modulation_mae_db": 0.26570858555001736, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1594891232420224, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10005402056072883, "1000–3000 Hz_40ms_envelope_correlation": 0.9922797560051266, "1000–3000 Hz_40ms_modulation_mae_db": 0.3695250811114774, "3000–6000 Hz_10ms_log_rms_mae_db": 1.93228014463661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2254484640396058, "3000–6000 Hz_10ms_envelope_correlation": 0.9547714675238511, "3000–6000 Hz_10ms_modulation_mae_db": 0.6230917540025279, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3090857511009015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15074528051005978, "3000–6000 Hz_40ms_envelope_correlation": 0.9773010419963981, "3000–6000 Hz_40ms_modulation_mae_db": 0.8804521577392229, "mr_stft_db": 5.612283068458741, "band_mae_db": 0.08295615305014988, "rms_gain_error_db": -0.05008033264288608, "output_peak": 0.5194490873414165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.881868976089351, "全帯域_10ms_rms_envelope_nrmse": 0.07527725375607171, "全帯域_10ms_envelope_correlation": 0.9942768817134665, "全帯域_10ms_modulation_mae_db": 0.5222721697727452, "全帯域_40ms_log_rms_mae_db": 0.5019325291942147, "全帯域_40ms_rms_envelope_nrmse": 0.04244370111854694, "全帯域_40ms_envelope_correlation": 0.9981237297533367, "全帯域_40ms_modulation_mae_db": 0.7917442303449879, "300–1000 Hz_10ms_log_rms_mae_db": 1.2873240369657004, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09244201944967802, "300–1000 Hz_10ms_envelope_correlation": 0.9933787975579949, "300–1000 Hz_10ms_modulation_mae_db": 0.3972662290342412, "300–1000 Hz_40ms_log_rms_mae_db": 0.7420679422132013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05053240853152042, "300–1000 Hz_40ms_envelope_correlation": 0.9980959646339057, "300–1000 Hz_40ms_modulation_mae_db": 0.7811876286643503, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6878779564836024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17895409682209595, "1000–3000 Hz_10ms_envelope_correlation": 0.9762561983748391, "1000–3000 Hz_10ms_modulation_mae_db": 0.6801057811607177, "1000–3000 Hz_40ms_log_rms_mae_db": 1.162654834396026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10363712645376977, "1000–3000 Hz_40ms_envelope_correlation": 0.9914727064332555, "1000–3000 Hz_40ms_modulation_mae_db": 0.8033933917586987, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8959617022503734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23081202279949464, "3000–6000 Hz_10ms_envelope_correlation": 0.9525765146422407, "3000–6000 Hz_10ms_modulation_mae_db": 1.038098360038451, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3219742196087527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.149059368896305, "3000–6000 Hz_40ms_envelope_correlation": 0.9777943325896035, "3000–6000 Hz_40ms_modulation_mae_db": 1.2497569607199546, "mr_stft_db": 5.616223766070386, "band_mae_db": 0.1191957543676893, "rms_gain_error_db": -0.11168818335425668, "output_peak": 0.5355635027790437}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8751011853298185, "全帯域_10ms_rms_envelope_nrmse": 0.07680552499277608, "全帯域_10ms_envelope_correlation": 0.9940450845980826, "全帯域_10ms_modulation_mae_db": 0.4074078693590146, "全帯域_40ms_log_rms_mae_db": 0.4611470767958113, "全帯域_40ms_rms_envelope_nrmse": 0.03980546382696611, "全帯域_40ms_envelope_correlation": 0.998391893713873, "全帯域_40ms_modulation_mae_db": 0.45729556402433674, "300–1000 Hz_10ms_log_rms_mae_db": 1.2856107615335712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10266079261521929, "300–1000 Hz_10ms_envelope_correlation": 0.9917731345001869, "300–1000 Hz_10ms_modulation_mae_db": 0.43588852170127623, "300–1000 Hz_40ms_log_rms_mae_db": 0.7896246077926192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05453520945698217, "300–1000 Hz_40ms_envelope_correlation": 0.997750102301027, "300–1000 Hz_40ms_modulation_mae_db": 0.7024902021608836, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6731758611371106, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17068948355091595, "1000–3000 Hz_10ms_envelope_correlation": 0.9784833503492585, "1000–3000 Hz_10ms_modulation_mae_db": 0.5163292881765992, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1568977888933585, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09972152339151184, "1000–3000 Hz_40ms_envelope_correlation": 0.9923233108283153, "1000–3000 Hz_40ms_modulation_mae_db": 0.7704529883114795, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8704323631616961, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2252761030927274, "3000–6000 Hz_10ms_envelope_correlation": 0.9544155386017231, "3000–6000 Hz_10ms_modulation_mae_db": 1.090052053382417, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2683417500617842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14828785621087578, "3000–6000 Hz_40ms_envelope_correlation": 0.9782433413596663, "3000–6000 Hz_40ms_modulation_mae_db": 1.5184739550367292, "mr_stft_db": 5.595859649030047, "band_mae_db": 0.12240538631473356, "rms_gain_error_db": -0.11722213522038794, "output_peak": 0.5197617615810072}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8441368705998101, "全帯域_10ms_rms_envelope_nrmse": 0.06905891320792644, "全帯域_10ms_envelope_correlation": 0.9951512285962032, "全帯域_10ms_modulation_mae_db": 0.3394978372414858, "全帯域_40ms_log_rms_mae_db": 0.44681580710676183, "全帯域_40ms_rms_envelope_nrmse": 0.03831343332778858, "全帯域_40ms_envelope_correlation": 0.9984274356792238, "全帯域_40ms_modulation_mae_db": 0.4701723052860372, "300–1000 Hz_10ms_log_rms_mae_db": 1.180771473861984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08874688457124599, "300–1000 Hz_10ms_envelope_correlation": 0.9939120871687933, "300–1000 Hz_10ms_modulation_mae_db": 0.3603880726145207, "300–1000 Hz_40ms_log_rms_mae_db": 0.694889818859757, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053704009629283914, "300–1000 Hz_40ms_envelope_correlation": 0.997782113113994, "300–1000 Hz_40ms_modulation_mae_db": 0.6921814703155907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6345814626748911, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17114689676952394, "1000–3000 Hz_10ms_envelope_correlation": 0.9788369136415218, "1000–3000 Hz_10ms_modulation_mae_db": 0.2560058163659298, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1177114090336842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1051397638425473, "1000–3000 Hz_40ms_envelope_correlation": 0.9914668209256706, "1000–3000 Hz_40ms_modulation_mae_db": 0.4469053643732459, "3000–6000 Hz_10ms_log_rms_mae_db": 1.886306498311675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21881269704413076, "3000–6000 Hz_10ms_envelope_correlation": 0.9575356021731394, "3000–6000 Hz_10ms_modulation_mae_db": 1.0286235769573406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.306895964483794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14096864183268046, "3000–6000 Hz_40ms_envelope_correlation": 0.980226197576408, "3000–6000 Hz_40ms_modulation_mae_db": 1.2076617932988316, "mr_stft_db": 5.603726773355846, "band_mae_db": 0.10845968842774023, "rms_gain_error_db": -0.08047982289233227, "output_peak": 0.5222188413253888}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6024475541756065, "spectral_convergence": 0.21229275486022428, "spectral_convergence_status": "ok", "amplitude_floor": 1.8586249352090494e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.520511082556523, "spectral_convergence": 0.2072470422689199, "spectral_convergence_status": "ok", "amplitude_floor": 1.6686015821705495e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.730516662800133, "spectral_convergence": 0.20817326455633045, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267872572435527e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003360135402914431, "output_power": 0.00034145797713480675, "delta_db": 0.06980484704914576, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011242144422279699, "output_power": 0.0011149154333276708, "delta_db": -0.03607232893764163, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015417458345753522, "output_power": 0.0014846156695277311, "delta_db": -0.16398743694867357, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.127065821964785e-05, "output_power": 7.990025038501288e-05, "delta_db": -0.07385636890541428, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.295885949753132e-05, "output_power": 5.109663397590442e-05, "delta_db": -0.15546332081841874, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.948150516097124e-05, "output_power": 1.9792089356817535e-05, "delta_db": 0.06869135157284059, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09464594237202241, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08885313354097144, "pitch": null, "output_rms_dbfs": -25.097916898861715, "output_peak": 0.5508650945923662, "reference_rms_dbfs": -25.009063765320747, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6178250998440875}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025375764068356454, "log_rms_mae_db": 0.8974960460830801, "rms_envelope_nrmse": 0.07897341509807031, "envelope_correlation": 0.9936293620317979, "rms_gain_error_db": -0.08885128752450355, "reference_mean_power": 0.0031583120304531104, "output_mean_power": 0.0030943533553993846}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1807904078754825e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008327873716353349, "output_power": 0.0008252260533213563, "reference_active": true, "delta_db": -0.03961200236563266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013733244323789595, "output_power": 0.00012416575015937543, "reference_active": true, "delta_db": -0.43771329811926873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.747498790065721e-05, "output_power": 1.652824724472359e-05, "reference_active": true, "delta_db": -0.24190082920319714}], "active_band_count": 3, "active_band_mae_db": 0.23974204322936618}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002402786371599188, "log_rms_mae_db": 0.5006386954719368, "rms_envelope_nrmse": 0.04542199836848625, "envelope_correlation": 0.9977530593466659, "rms_gain_error_db": -0.08884887135718275, "reference_mean_power": 0.0031741738524038612, "output_mean_power": 0.0031098956912783845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0384864838590044e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000782161356346486, "output_power": 0.0007750030524036875, "reference_active": true, "delta_db": -0.03992942278880697}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.85949543312588e-05, "output_power": 5.3070435031330504e-05, "reference_active": true, "delta_db": -0.4300757244417844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3154862476080315e-06, "output_power": 1.1856563113632028e-06, "reference_active": true, "delta_db": -0.4512749479750905}], "active_band_count": 3, "active_band_mae_db": 0.3070933650685606}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002501327279402343, "log_rms_mae_db": 1.2412645956034742, "rms_envelope_nrmse": 0.09901202659649239, "envelope_correlation": 0.992230008638375, "rms_gain_error_db": -0.12025367483190606, "reference_mean_power": 0.0021283926153372414, "output_mean_power": 0.0020702670782122004}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0698576130619868e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007463968445819582, "output_power": 0.0007355760852306816, "reference_active": true, "delta_db": -0.06342193568737593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014378870760237378, "output_power": 0.00012806998678754108, "reference_active": true, "delta_db": -0.5027741549630055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4173849116031066e-05, "output_power": 1.404307664252905e-05, "reference_active": true, "delta_db": -0.04025539191736295}], "active_band_count": 3, "active_band_mae_db": 0.20215049418924813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00023719519351520602, "log_rms_mae_db": 0.7321627422356418, "rms_envelope_nrmse": 0.05864233932798882, "envelope_correlation": 0.9971403412510679, "rms_gain_error_db": -0.12025369329405672, "reference_mean_power": 0.0021390960955203144, "output_mean_power": 0.0020806782418573764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.624199297765067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007211274534636877, "output_power": 0.0007079415981188229, "reference_active": true, "delta_db": -0.08014597654263694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.136224424017926e-05, "output_power": 5.430444234988553e-05, "reference_active": true, "delta_db": -0.5306587665617469}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3649293472151981e-06, "output_power": 1.1682074493348689e-06, "reference_active": true, "delta_db": -0.6759020030222387}], "active_band_count": 3, "active_band_mae_db": 0.4289022487088742}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.486004098842285e-05, "log_rms_mae_db": 1.649036038904572, "rms_envelope_nrmse": 0.17718163858455885, "envelope_correlation": 0.976831869536925, "rms_gain_error_db": -0.2012139137873824, "reference_mean_power": 9.73047769043704e-05, "output_mean_power": 9.289937022615359e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.517204404381221e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1360183676562794e-05, "output_power": 3.8435206966121366e-05, "reference_active": true, "delta_db": -0.3185323474684371}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.821587959830417e-06, "output_power": 4.543650874352252e-06, "reference_active": true, "delta_db": -0.2578514036001208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2148111012630244e-06, "output_power": 2.0628630728558883e-06, "reference_active": true, "delta_db": -0.3086628998922535}], "active_band_count": 3, "active_band_mae_db": 0.29501555032027044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5230962699738786e-05, "log_rms_mae_db": 1.1205874433300904, "rms_envelope_nrmse": 0.1187898871643647, "envelope_correlation": 0.9891849813278658, "rms_gain_error_db": -0.20123447313412574, "reference_mean_power": 9.77871242521371e-05, "output_mean_power": 9.335943766903799e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.922472441667658e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.88087102342988e-05, "output_power": 3.603859407666171e-05, "reference_active": true, "delta_db": -0.3216136955690633}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.146701619692133e-06, "output_power": 2.030526354225764e-06, "reference_active": true, "delta_db": -0.24163053391508266}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.612101873452723e-07, "output_power": 1.561954136411917e-07, "reference_active": true, "delta_db": -0.13724205120817645}], "active_band_count": 3, "active_band_mae_db": 0.23349542689744082}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.5953225979159985e-05, "log_rms_mae_db": 1.926555828886782, "rms_envelope_nrmse": 0.22236239527355764, "envelope_correlation": 0.9557586983296079, "rms_gain_error_db": -0.10067737812414768, "reference_mean_power": 3.136903007980474e-05, "output_mean_power": 3.065020283063109e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9944851157580815e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0899878382466362e-05, "output_power": 1.0562871930431712e-05, "reference_active": true, "delta_db": -0.13639638024497192}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5519027988888207e-06, "output_power": 2.0895367881578095e-06, "reference_active": true, "delta_db": -0.8681410654634956}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4088478818809383e-06, "output_power": 6.701603856509244e-07, "reference_active": true, "delta_db": -3.226853510279822}], "active_band_count": 3, "active_band_mae_db": 1.4104636519960965}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9738091980327894e-05, "log_rms_mae_db": 1.3255369962273988, "rms_envelope_nrmse": 0.14304928590590657, "envelope_correlation": 0.979590204213012, "rms_gain_error_db": -0.1005551932130544, "reference_mean_power": 3.151908471872331e-05, "output_mean_power": 3.0797685392606e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7532611532165288e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0780259770731776e-05, "output_power": 1.0183260251688004e-05, "reference_active": true, "delta_db": -0.24742383030378523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1761748042676738e-06, "output_power": 9.154433288382783e-07, "reference_active": true, "delta_db": -1.0884040764908876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.367404555815231e-08, "output_power": 3.7988774345779976e-08, "reference_active": true, "delta_db": -3.919639948939441}], "active_band_count": 3, "active_band_mae_db": 1.7518226185780381}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9016105143938604, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8bd488d70eb832395f385555b668d5b31ca7ecd3bc50e9d1b7054bcb199fafb2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.978309102078832, "standard_deviation": 0.0015106491756410475, "minimum": 0.9762561983748391, "maximum": 0.9803776753385716, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6642755290923381, "standard_deviation": 0.028598460917703853, "minimum": 1.620669592282862, "maximum": 1.7054754882328613, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46415322533332753, "standard_deviation": 0.19404959155379198, "minimum": 0.2560058163659298, "maximum": 0.7447516662795465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17310715133024923, "standard_deviation": 0.003972955457404932, "minimum": 0.1678478598920509, "maximum": 0.17895409682209595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9908566700529404, "standard_deviation": 0.0012481632542933393, "minimum": 0.9891525375822859, "maximum": 0.9923233108283153, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.143851115535678, "standard_deviation": 0.03217490117135316, "minimum": 1.1007545200131015, "maximum": 1.2008779606613582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5749004267574866, "standard_deviation": 0.22271591727094917, "minimum": 0.23349542689744082, "maximum": 0.8340166110378933, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1094299518223536, "standard_deviation": 0.008291491801179118, "minimum": 0.09972152339151184, "maximum": 0.1192793818407932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.952572794239348, "standard_deviation": 0.005208265568020154, "minimum": 0.9407904286931495, "maximum": 0.9575356021731394, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9183548193827216, "standard_deviation": 0.031409562830350714, "minimum": 1.8704323631616961, "maximum": 1.9570701030564979, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0824521224563302, "standard_deviation": 0.35010497062957074, "minimum": 0.6230917540025279, "maximum": 1.7209867381696544, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23045615303220168, "standard_deviation": 0.011637158026782812, "minimum": 0.21881269704413076, "maximum": 0.25625228440361125, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9769644850177128, "standard_deviation": 0.0036080465915363227, "minimum": 0.9685704834252465, "maximum": 0.980226197576408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3288271878901068, "standard_deviation": 0.042732785106923576, "minimum": 1.2683417500617842, "maximum": 1.4147102268948946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2792616028514834, "standard_deviation": 0.3759626260187632, "minimum": 0.7940853159196312, "maximum": 1.7840950436360181, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15223262914803667, "standard_deviation": 0.012203544283024075, "minimum": 0.14096864183268046, "maximum": 0.18054317599421152, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992880488231751, "standard_deviation": 0.0010935977216499125, "minimum": 0.9911842625762115, "maximum": 0.9944522099683737, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.259629839234388, "standard_deviation": 0.05742906580057236, "minimum": 1.180771473861984, "maximum": 1.36229082434418, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3517504929412627, "standard_deviation": 0.1016651471382078, "minimum": 0.20215049418924813, "maximum": 0.5162057104965599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09514710966239717, "standard_deviation": 0.007099386511570966, "minimum": 0.08477987413169454, "maximum": 0.10616341926306708, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978365492710815, "standard_deviation": 0.00038435251570119835, "minimum": 0.9971403412510679, "maximum": 0.998506149723659, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7395494797061416, "standard_deviation": 0.03728170376944797, "minimum": 0.6836865953759078, "maximum": 0.7896246077926192, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5674220975298547, "standard_deviation": 0.23339311549076922, "minimum": 0.12354284909357194, "maximum": 0.8325531897634139, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.052536703135754106, "standard_deviation": 0.003921705933316515, "minimum": 0.04529752402070213, "maximum": 0.05864233932798882, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1170822353418173, "standard_deviation": 0.035992278268958684, "minimum": 0.0733251634125393, "maximum": 0.17847994227799946, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.609823608593038, "standard_deviation": 0.011464002464722475, "minimum": 5.595859649030047, "maximum": 5.62971417663645, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5520609371127241, "standard_deviation": 0.04079550918151814, "minimum": 0.5194490873414165, "maximum": 0.6243206481767642, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0903035788665844, "standard_deviation": 0.02119040818199113, "minimum": -0.11722213522038794, "maximum": -0.05008033264288608, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941986639482774, "standard_deviation": 0.0008378385645651377, "minimum": 0.9925637347494689, "maximum": 0.9951512285962032, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8748248602387272, "standard_deviation": 0.016711725716832267, "minimum": 0.8441368705998101, "maximum": 0.8974960460830801, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3204367897133067, "standard_deviation": 0.1303101800188421, "minimum": 0.15715741927105153, "maximum": 0.5222721697727452, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07535617121757335, "standard_deviation": 0.005268062120366839, "minimum": 0.06905891320792644, "maximum": 0.08533123995298299, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998241358277389, "standard_deviation": 0.0002766104967612658, "minimum": 0.9977530593466659, "maximum": 0.9986840137804555, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4776426237908391, "standard_deviation": 0.019789495643749808, "minimum": 0.44681580710676183, "maximum": 0.5019325291942147, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4598187062303542, "standard_deviation": 0.20970880091486208, "minimum": 0.13103964977030635, "maximum": 0.7917442303449879, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04061058031042025, "standard_deviation": 0.0027994638683980426, "minimum": 0.036074038995803324, "maximum": 0.04542199836848625, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1284-1180-0028", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7321014404296875, "predicted_mos": 2.7321014404296875, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "edae26595c9b92378e61629d01bf1d8fb228a34ffd7be251fbe35cdcf563c867", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.504788199905306, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8502460666491214, "全帯域_10ms_rms_envelope_nrmse": 0.07146548984863632, "全帯域_10ms_envelope_correlation": 0.9948676246422044, "全帯域_10ms_modulation_mae_db": 0.2275529124668502, "全帯域_40ms_log_rms_mae_db": 0.4686215370833271, "全帯域_40ms_rms_envelope_nrmse": 0.03879200137001191, "全帯域_40ms_envelope_correlation": 0.9984809685688493, "全帯域_40ms_modulation_mae_db": 0.1543445815346136, "300–1000 Hz_10ms_log_rms_mae_db": 1.1711349589063544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0927692199568481, "300–1000 Hz_10ms_envelope_correlation": 0.993369993262371, "300–1000 Hz_10ms_modulation_mae_db": 0.28971432536443037, "300–1000 Hz_40ms_log_rms_mae_db": 0.7259625301884612, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05133880064495477, "300–1000 Hz_40ms_envelope_correlation": 0.9980759632781108, "300–1000 Hz_40ms_modulation_mae_db": 0.21426648977092577, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6402270997000097, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1721786629759003, "1000–3000 Hz_10ms_envelope_correlation": 0.9780833486339511, "1000–3000 Hz_10ms_modulation_mae_db": 0.38599143502617306, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1388081884711991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11016227241963107, "1000–3000 Hz_40ms_envelope_correlation": 0.9905844432873668, "1000–3000 Hz_40ms_modulation_mae_db": 0.6114207091791866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8547142189733645, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22654509873941128, "3000–6000 Hz_10ms_envelope_correlation": 0.9542721977160612, "3000–6000 Hz_10ms_modulation_mae_db": 1.3020166798780493, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2637638595776608, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14984495998735867, "3000–6000 Hz_40ms_envelope_correlation": 0.9775507401030261, "3000–6000 Hz_40ms_modulation_mae_db": 1.5271933831128999, "mr_stft_db": 5.592179315132031, "band_mae_db": 0.10602961610308465, "rms_gain_error_db": -0.10431890372472757, "output_peak": 0.5160164961678803}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8551994051852829, "全帯域_10ms_rms_envelope_nrmse": 0.07197895228920369, "全帯域_10ms_envelope_correlation": 0.9947180820033737, "全帯域_10ms_modulation_mae_db": 0.5544771775822205, "全帯域_40ms_log_rms_mae_db": 0.4498298529774914, "全帯域_40ms_rms_envelope_nrmse": 0.03830841313028859, "全帯域_40ms_envelope_correlation": 0.9984219820695913, "全帯域_40ms_modulation_mae_db": 0.5607226713807205, "300–1000 Hz_10ms_log_rms_mae_db": 1.2437078438437215, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08942756918628517, "300–1000 Hz_10ms_envelope_correlation": 0.9937049396283013, "300–1000 Hz_10ms_modulation_mae_db": 0.5045221572878321, "300–1000 Hz_40ms_log_rms_mae_db": 0.7269766214869966, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05071097546700803, "300–1000 Hz_40ms_envelope_correlation": 0.9979172528824517, "300–1000 Hz_40ms_modulation_mae_db": 0.6499196967356469, "1000–3000 Hz_10ms_log_rms_mae_db": 1.66637754696154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17640602965650745, "1000–3000 Hz_10ms_envelope_correlation": 0.977286459828309, "1000–3000 Hz_10ms_modulation_mae_db": 0.7427651719613394, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1032683552418332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10757478951074002, "1000–3000 Hz_40ms_envelope_correlation": 0.9909007275721188, "1000–3000 Hz_40ms_modulation_mae_db": 0.9488376470455968, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8482843099251456, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23402764084976957, "3000–6000 Hz_10ms_envelope_correlation": 0.9509007761952366, "3000–6000 Hz_10ms_modulation_mae_db": 0.9393587306168586, "3000–6000 Hz_40ms_log_rms_mae_db": 1.229811125395499, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15249400023842113, "3000–6000 Hz_40ms_envelope_correlation": 0.97689940287864, "3000–6000 Hz_40ms_modulation_mae_db": 1.0579208440928942, "mr_stft_db": 5.577620711303165, "band_mae_db": 0.14373190747147743, "rms_gain_error_db": -0.07793605334124365, "output_peak": 0.5564189183868234}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8760189204636497, "全帯域_10ms_rms_envelope_nrmse": 0.07613140655658591, "全帯域_10ms_envelope_correlation": 0.9942513493883863, "全帯域_10ms_modulation_mae_db": 0.35513852637596627, "全帯域_40ms_log_rms_mae_db": 0.4778944780435269, "全帯域_40ms_rms_envelope_nrmse": 0.042725150868872394, "全帯域_40ms_envelope_correlation": 0.99823132008077, "全帯域_40ms_modulation_mae_db": 0.6666274850890078, "300–1000 Hz_10ms_log_rms_mae_db": 1.310823460211046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09947565860599733, "300–1000 Hz_10ms_envelope_correlation": 0.9925401316111991, "300–1000 Hz_10ms_modulation_mae_db": 0.32956003352578317, "300–1000 Hz_40ms_log_rms_mae_db": 0.768746536142107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06025403767379012, "300–1000 Hz_40ms_envelope_correlation": 0.9974533477254218, "300–1000 Hz_40ms_modulation_mae_db": 0.6965511118734086, "1000–3000 Hz_10ms_log_rms_mae_db": 1.598204507993312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16138114830488887, "1000–3000 Hz_10ms_envelope_correlation": 0.9818014251397504, "1000–3000 Hz_10ms_modulation_mae_db": 0.1519374742715235, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0686560355766195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09872499587544914, "1000–3000 Hz_40ms_envelope_correlation": 0.9929656871500973, "1000–3000 Hz_40ms_modulation_mae_db": 0.16771944939467867, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8185199151076936, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22357015628696147, "3000–6000 Hz_10ms_envelope_correlation": 0.9551033652641467, "3000–6000 Hz_10ms_modulation_mae_db": 1.146954452862654, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2558789824665777, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14523315802993686, "3000–6000 Hz_40ms_envelope_correlation": 0.9791505101076358, "3000–6000 Hz_40ms_modulation_mae_db": 1.3361047938311978, "mr_stft_db": 5.579900049198127, "band_mae_db": 0.16330194420124391, "rms_gain_error_db": -0.13405269622649696, "output_peak": 0.49342623920222545}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8488671873309229, "全帯域_10ms_rms_envelope_nrmse": 0.07051868350720628, "全帯域_10ms_envelope_correlation": 0.9948830968756212, "全帯域_10ms_modulation_mae_db": 0.21780175447207686, "全帯域_40ms_log_rms_mae_db": 0.4226160896698999, "全帯域_40ms_rms_envelope_nrmse": 0.03618805350536246, "全帯域_40ms_envelope_correlation": 0.9985145006717773, "全帯域_40ms_modulation_mae_db": 0.18542139467256216, "300–1000 Hz_10ms_log_rms_mae_db": 1.232564174664276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09094818209951068, "300–1000 Hz_10ms_envelope_correlation": 0.9933924907750468, "300–1000 Hz_10ms_modulation_mae_db": 0.19603226508322869, "300–1000 Hz_40ms_log_rms_mae_db": 0.6860781339582459, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04522051731667153, "300–1000 Hz_40ms_envelope_correlation": 0.9982325368591795, "300–1000 Hz_40ms_modulation_mae_db": 0.2838143411221989, "1000–3000 Hz_10ms_log_rms_mae_db": 1.571511085484176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1747613911720004, "1000–3000 Hz_10ms_envelope_correlation": 0.9773607507026374, "1000–3000 Hz_10ms_modulation_mae_db": 0.5262402433040815, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0504533420586384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09807412034197358, "1000–3000 Hz_40ms_envelope_correlation": 0.9925505764115184, "1000–3000 Hz_40ms_modulation_mae_db": 0.8586414603411087, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8740529482306276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22244961556036771, "3000–6000 Hz_10ms_envelope_correlation": 0.9565326349742833, "3000–6000 Hz_10ms_modulation_mae_db": 0.6628946797466685, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2712685808427608, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1458297908410336, "3000–6000 Hz_40ms_envelope_correlation": 0.9790047359921812, "3000–6000 Hz_40ms_modulation_mae_db": 0.8622903770621897, "mr_stft_db": 5.569575361672, "band_mae_db": 0.0673330108331381, "rms_gain_error_db": -0.014068315640823292, "output_peak": 0.5911766842695771}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8769077365204866, "全帯域_10ms_rms_envelope_nrmse": 0.07776857386319153, "全帯域_10ms_envelope_correlation": 0.9938499706626116, "全帯域_10ms_modulation_mae_db": 0.2561689661770556, "全帯域_40ms_log_rms_mae_db": 0.4625625531683551, "全帯域_40ms_rms_envelope_nrmse": 0.039314836494129216, "全帯域_40ms_envelope_correlation": 0.9983824028071107, "全帯域_40ms_modulation_mae_db": 0.31246576843980334, "300–1000 Hz_10ms_log_rms_mae_db": 1.23088017178798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09695943412917761, "300–1000 Hz_10ms_envelope_correlation": 0.9926447033373906, "300–1000 Hz_10ms_modulation_mae_db": 0.2771198868735732, "300–1000 Hz_40ms_log_rms_mae_db": 0.6886879739609317, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051897917502654066, "300–1000 Hz_40ms_envelope_correlation": 0.9979149432907348, "300–1000 Hz_40ms_modulation_mae_db": 0.5578479496272276, "1000–3000 Hz_10ms_log_rms_mae_db": 1.623223233144595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16346715957737387, "1000–3000 Hz_10ms_envelope_correlation": 0.9803036605941544, "1000–3000 Hz_10ms_modulation_mae_db": 0.634269939922666, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0997743479272533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09235899550872549, "1000–3000 Hz_40ms_envelope_correlation": 0.9932548853067027, "1000–3000 Hz_40ms_modulation_mae_db": 0.7842415866107908, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8688229466738866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2323229149029743, "3000–6000 Hz_10ms_envelope_correlation": 0.9519999870497796, "3000–6000 Hz_10ms_modulation_mae_db": 0.9448629499705214, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2580828996977884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1474134250279591, "3000–6000 Hz_40ms_envelope_correlation": 0.9782929351416361, "3000–6000 Hz_40ms_modulation_mae_db": 1.1790894138484895, "mr_stft_db": 5.577003005992915, "band_mae_db": 0.13154303866564712, "rms_gain_error_db": -0.09096888045522847, "output_peak": 0.5534202471999855}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8368719131959117, "全帯域_10ms_rms_envelope_nrmse": 0.07258951040492538, "全帯域_10ms_envelope_correlation": 0.9946399387139453, "全帯域_10ms_modulation_mae_db": 0.1397100995572355, "全帯域_40ms_log_rms_mae_db": 0.4478721863622121, "全帯域_40ms_rms_envelope_nrmse": 0.04059357206708481, "全帯域_40ms_envelope_correlation": 0.9982325997932433, "全帯域_40ms_modulation_mae_db": 0.13790646108548646, "300–1000 Hz_10ms_log_rms_mae_db": 1.2372693302813635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08611918333250408, "300–1000 Hz_10ms_envelope_correlation": 0.9942008959114287, "300–1000 Hz_10ms_modulation_mae_db": 0.22920862949023277, "300–1000 Hz_40ms_log_rms_mae_db": 0.6781958232248362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048103252003873255, "300–1000 Hz_40ms_envelope_correlation": 0.9981539954301729, "300–1000 Hz_40ms_modulation_mae_db": 0.15437017484921986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6056081634328438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16882710630126438, "1000–3000 Hz_10ms_envelope_correlation": 0.9790157179248692, "1000–3000 Hz_10ms_modulation_mae_db": 0.6736563086565566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.069175090081271, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1081489041280961, "1000–3000 Hz_40ms_envelope_correlation": 0.9910340068100986, "1000–3000 Hz_40ms_modulation_mae_db": 0.8092918923801746, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8426950278447811, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22887432637964655, "3000–6000 Hz_10ms_envelope_correlation": 0.952897755078818, "3000–6000 Hz_10ms_modulation_mae_db": 1.5493639371183905, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2489259455873887, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14962639641621883, "3000–6000 Hz_40ms_envelope_correlation": 0.9778051351625149, "3000–6000 Hz_40ms_modulation_mae_db": 1.8141149475355742, "mr_stft_db": 5.578821141597252, "band_mae_db": 0.12337060823605929, "rms_gain_error_db": -0.06961772835315258, "output_peak": 0.5568492395693745}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8764084535847402, "全帯域_10ms_rms_envelope_nrmse": 0.07892798550064417, "全帯域_10ms_envelope_correlation": 0.9938082481936574, "全帯域_10ms_modulation_mae_db": 0.47258831994676315, "全帯域_40ms_log_rms_mae_db": 0.4711256766852875, "全帯域_40ms_rms_envelope_nrmse": 0.04716657351409233, "全帯域_40ms_envelope_correlation": 0.9977953228666822, "全帯域_40ms_modulation_mae_db": 0.7001006085692577, "300–1000 Hz_10ms_log_rms_mae_db": 1.2379560239901364, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09776271870997173, "300–1000 Hz_10ms_envelope_correlation": 0.9929053453852913, "300–1000 Hz_10ms_modulation_mae_db": 0.5961966948298749, "300–1000 Hz_40ms_log_rms_mae_db": 0.69172817309123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06091268883842253, "300–1000 Hz_40ms_envelope_correlation": 0.9974790394451218, "300–1000 Hz_40ms_modulation_mae_db": 1.0372349647064645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6435995954769023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1788916875741854, "1000–3000 Hz_10ms_envelope_correlation": 0.9780577894035277, "1000–3000 Hz_10ms_modulation_mae_db": 0.4012493886542153, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0965851793679577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11778121835909851, "1000–3000 Hz_40ms_envelope_correlation": 0.9902418442095694, "1000–3000 Hz_40ms_modulation_mae_db": 0.7808553781462871, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8730629197076987, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22813478765214318, "3000–6000 Hz_10ms_envelope_correlation": 0.9543489987291894, "3000–6000 Hz_10ms_modulation_mae_db": 0.5135554575328046, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2391343064561193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.146201199211702, "3000–6000 Hz_40ms_envelope_correlation": 0.9789435061742106, "3000–6000 Hz_40ms_modulation_mae_db": 0.7052796956151653, "mr_stft_db": 5.574633481048566, "band_mae_db": 0.1282571699553088, "rms_gain_error_db": -0.14350439755478703, "output_peak": 0.5374036860681071}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8538441933926904, "全帯域_10ms_rms_envelope_nrmse": 0.07134265883843248, "全帯域_10ms_envelope_correlation": 0.9950555990246008, "全帯域_10ms_modulation_mae_db": 0.3632657727216126, "全帯域_40ms_log_rms_mae_db": 0.46760428276971205, "全帯域_40ms_rms_envelope_nrmse": 0.040551407199994505, "全帯域_40ms_envelope_correlation": 0.9985278406831488, "全帯域_40ms_modulation_mae_db": 0.5480465170884153, "300–1000 Hz_10ms_log_rms_mae_db": 1.25283671462146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09004841776139334, "300–1000 Hz_10ms_envelope_correlation": 0.9939264906270492, "300–1000 Hz_10ms_modulation_mae_db": 0.5506159793254104, "300–1000 Hz_40ms_log_rms_mae_db": 0.7517590029343674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05271345466377905, "300–1000 Hz_40ms_envelope_correlation": 0.998129556479516, "300–1000 Hz_40ms_modulation_mae_db": 0.7960642996360047, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6048719306244488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1716279475613773, "1000–3000 Hz_10ms_envelope_correlation": 0.9783804412257354, "1000–3000 Hz_10ms_modulation_mae_db": 0.7471743575544797, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0496897060141543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11133181776088555, "1000–3000 Hz_40ms_envelope_correlation": 0.9907085052987614, "1000–3000 Hz_40ms_modulation_mae_db": 0.9561794605402433, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8668906549528057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21890101731193326, "3000–6000 Hz_10ms_envelope_correlation": 0.9576032493284947, "3000–6000 Hz_10ms_modulation_mae_db": 0.6224250576906031, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2808769306064272, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14420099415902035, "3000–6000 Hz_40ms_envelope_correlation": 0.9793189067253505, "3000–6000 Hz_40ms_modulation_mae_db": 0.7397703901244924, "mr_stft_db": 5.574136263568561, "band_mae_db": 0.14549980217185515, "rms_gain_error_db": -0.16259232382314162, "output_peak": 0.5422638614539652}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.582273264153956, "spectral_convergence": 0.2101059821819543, "spectral_convergence_status": "ok", "amplitude_floor": 1.8586249352090494e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.483220272410426, "spectral_convergence": 0.20503178482647172, "spectral_convergence_status": "ok", "amplitude_floor": 1.6686015821705495e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.711044408831711, "spectral_convergence": 0.21148478119835762, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267872572435527e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003360135402914431, "output_power": 0.0003455674612334611, "delta_db": 0.12176063947188198, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011242144422279699, "output_power": 0.0011137569816310918, "delta_db": -0.040587207134275056, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015417458345753522, "output_power": 0.0014700200290929926, "delta_db": -0.2068953166968136, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.127065821964785e-05, "output_power": 8.115680265769247e-05, "delta_db": -0.006088484068232945, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.295885949753132e-05, "output_power": 5.0542574139644476e-05, "delta_db": -0.20281266906902728, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.948150516097124e-05, "output_power": 1.9743577366556986e-05, "delta_db": 0.05803338017827697, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10602961610308465, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10431890372472757, "pitch": null, "output_rms_dbfs": -25.11338266904547, "output_peak": 0.5160164961678803, "reference_rms_dbfs": -25.009063765320747, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.592179315132031}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025375764068356454, "log_rms_mae_db": 0.8502460666491214, "rms_envelope_nrmse": 0.07146548984863632, "envelope_correlation": 0.9948676246422044, "rms_gain_error_db": -0.10431701767768362, "reference_mean_power": 0.0031583120304531104, "output_mean_power": 0.003083353601571728}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1807904078754825e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008327873716353349, "output_power": 0.000796631261835967, "reference_active": true, "delta_db": -0.1927678569561836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013733244323789595, "output_power": 0.00013298726930055263, "reference_active": true, "delta_db": -0.1396307801399061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.747498790065721e-05, "output_power": 1.612095900890292e-05, "reference_active": true, "delta_db": -0.35026010030446086}], "active_band_count": 3, "active_band_mae_db": 0.2275529124668502}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002402786371599188, "log_rms_mae_db": 0.4686215370833271, "rms_envelope_nrmse": 0.03879200137001191, "envelope_correlation": 0.9984809685688493, "rms_gain_error_db": -0.10431328029658334, "reference_mean_power": 0.0031741738524038612, "output_mean_power": 0.003098841630555439}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0384864838590044e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000782161356346486, "output_power": 0.0007481439940786888, "reference_active": true, "delta_db": -0.19311161413469977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.85949543312588e-05, "output_power": 5.7237608580063976e-05, "reference_active": true, "delta_db": -0.10178739710599684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3154862476080315e-06, "output_power": 1.265531148460155e-06, "reference_active": true, "delta_db": -0.16813473336314422}], "active_band_count": 3, "active_band_mae_db": 0.1543445815346136}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002501327279402343, "log_rms_mae_db": 1.1711349589063544, "rms_envelope_nrmse": 0.0927692199568481, "envelope_correlation": 0.993369993262371, "rms_gain_error_db": -0.18162752884098257, "reference_mean_power": 0.0021283926153372414, "output_mean_power": 0.002041216125347001}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0698576130619868e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007463968445819582, "output_power": 0.0007067706739637324, "reference_active": true, "delta_db": -0.23691273720425365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014378870760237378, "output_power": 0.00013695512847275047, "reference_active": true, "delta_db": -0.2114648056811619}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4173849116031066e-05, "output_power": 1.2865043587463792e-05, "reference_active": true, "delta_db": -0.42076543320787557}], "active_band_count": 3, "active_band_mae_db": 0.28971432536443037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00023719519351520602, "log_rms_mae_db": 0.7259625301884612, "rms_envelope_nrmse": 0.05133880064495477, "envelope_correlation": 0.9980759632781108, "rms_gain_error_db": -0.18162750082930365, "reference_mean_power": 0.0021390960955203144, "output_mean_power": 0.002051481216642785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.624199297765067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007211274534636877, "output_power": 0.0006845362002732005, "reference_active": true, "delta_db": -0.2261560971411504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.136224424017926e-05, "output_power": 5.8944875809893246e-05, "reference_active": true, "delta_db": -0.1745517793110185}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3649293472151981e-06, "output_power": 1.2909250026500597e-06, "reference_active": true, "delta_db": -0.24209159286060847}], "active_band_count": 3, "active_band_mae_db": 0.21426648977092577}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.486004098842285e-05, "log_rms_mae_db": 1.6402270997000097, "rms_envelope_nrmse": 0.1721786629759003, "envelope_correlation": 0.9780833486339511, "rms_gain_error_db": -0.15405047705467664, "reference_mean_power": 9.73047769043704e-05, "output_mean_power": 9.391373511582906e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.517204404381221e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1360183676562794e-05, "output_power": 3.9915390628026683e-05, "reference_active": true, "delta_db": -0.1544207494192049}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.821587959830417e-06, "output_power": 4.5356668253376865e-06, "reference_active": true, "delta_db": -0.26548948707452236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2148111012630244e-06, "output_power": 1.8686599805693255e-06, "reference_active": true, "delta_db": -0.7380640685847918}], "active_band_count": 3, "active_band_mae_db": 0.38599143502617306}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5230962699738786e-05, "log_rms_mae_db": 1.1388081884711991, "rms_envelope_nrmse": 0.11016227241963107, "envelope_correlation": 0.9905844432873668, "rms_gain_error_db": -0.15403211882692186, "reference_mean_power": 9.77871242521371e-05, "output_mean_power": 9.43796717607532e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.922472441667658e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.88087102342988e-05, "output_power": 3.769420217660611e-05, "reference_active": true, "delta_db": -0.12654654062506018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.146701619692133e-06, "output_power": 1.913925145862199e-06, "reference_active": true, "delta_db": -0.4984673557083072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.612101873452723e-07, "output_power": 1.2203029455306105e-07, "reference_active": true, "delta_db": -1.2092482312041923}], "active_band_count": 3, "active_band_mae_db": 0.6114207091791866}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.5953225979159985e-05, "log_rms_mae_db": 1.8547142189733645, "rms_envelope_nrmse": 0.22654509873941128, "envelope_correlation": 0.9542721977160612, "rms_gain_error_db": -0.06049405675774992, "reference_mean_power": 3.136903007980474e-05, "output_mean_power": 3.093511144950124e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9944851157580815e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0899878382466362e-05, "output_power": 1.0571027732842557e-05, "reference_active": true, "delta_db": -0.1330444005206499}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5519027988888207e-06, "output_power": 2.0741086144598274e-06, "reference_active": true, "delta_db": -0.9003263296796118}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4088478818809383e-06, "output_power": 7.271033949610268e-07, "reference_active": true, "delta_db": -2.8726793094338863}], "active_band_count": 3, "active_band_mae_db": 1.3020166798780493}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9738091980327894e-05, "log_rms_mae_db": 1.2637638595776608, "rms_envelope_nrmse": 0.14984495998735867, "envelope_correlation": 0.9775507401030261, "rms_gain_error_db": -0.06033996554624833, "reference_mean_power": 3.151908471872331e-05, "output_mean_power": 3.1084193299303616e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7532611532165288e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0780259770731776e-05, "output_power": 1.0151083156886889e-05, "reference_active": true, "delta_db": -0.26116840626468135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1761748042676738e-06, "output_power": 9.394949334793044e-07, "reference_active": true, "delta_db": -0.975774294404233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.367404555815231e-08, "output_power": 4.336661574710443e-08, "reference_active": true, "delta_db": -3.344637448669785}], "active_band_count": 3, "active_band_mae_db": 1.5271933831128999}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8970064246072279, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "24a502f2f6c95bdce67af34bc3846cc2f8278d41b30e351368265c4f04323ad4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9787861991816168, "standard_deviation": 0.0015545921045481182, "minimum": 0.977286459828309, "maximum": 0.9818014251397504, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6192028953522284, "standard_deviation": 0.030200428809655618, "minimum": 1.571511085484176, "maximum": 1.66637754696154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5329105399188794, "standard_deviation": 0.20844180457611464, "minimum": 0.1519374742715235, "maximum": 0.7471743575544797, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17094264164043724, "standard_deviation": 0.006110841156209818, "minimum": 0.16138114830488887, "maximum": 0.1788916875741854, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9915300845057792, "standard_deviation": 0.0011921332909153115, "minimum": 0.9902418442095694, "maximum": 0.9932548853067027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0845512805923658, "standard_deviation": 0.03056090471243049, "minimum": 1.0496897060141543, "maximum": 1.1388081884711991, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7396484479547583, "standard_deviation": 0.25547911244567634, "minimum": 0.16771944939467867, "maximum": 0.9561794605402433, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10551963923807493, "standard_deviation": 0.008380452890200622, "minimum": 0.09235899550872549, "maximum": 0.11778121835909851, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9542073705420012, "standard_deviation": 0.0022452208220320805, "minimum": 0.9509007761952366, "maximum": 0.9576032493284947, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8558803676770004, "standard_deviation": 0.01905981619644234, "minimum": 1.8185199151076936, "maximum": 1.8740529482306276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9601789931770688, "standard_deviation": 0.35886195228991846, "minimum": 0.5135554575328046, "maximum": 1.5493639371183905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22685319471040094, "standard_deviation": 0.005079649442473154, "minimum": 0.21890101731193326, "maximum": 0.23402764084976957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9783707340356493, "standard_deviation": 0.0008783319216285559, "minimum": 0.97689940287864, "maximum": 0.9793189067253505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2559678288287777, "standard_deviation": 0.016634579332608358, "minimum": 1.229811125395499, "maximum": 1.2808769306064272, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1527204806528628, "standard_deviation": 0.3920760443740123, "minimum": 0.7052796956151653, "maximum": 1.8141149475355742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14760549048895633, "standard_deviation": 0.0028126601770441483, "minimum": 0.14420099415902035, "maximum": 0.15249400023842113, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933356238172597, "standard_deviation": 0.0006016580839897724, "minimum": 0.9925401316111991, "maximum": 0.9942008959114287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2396465847882923, "standard_deviation": 0.03795859340108778, "minimum": 1.1711349589063544, "maximum": 1.310823460211046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3716212464725457, "standard_deviation": 0.15522202470590482, "minimum": 0.19603226508322869, "maximum": 0.5961966948298749, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09293879797271101, "standard_deviation": 0.004680472362863405, "minimum": 0.08611918333250408, "maximum": 0.09947565860599733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979195794238387, "standard_deviation": 0.0003006929782376842, "minimum": 0.9974533477254218, "maximum": 0.9982325368591795, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.714766849373397, "standard_deviation": 0.033653737865097545, "minimum": 0.6781958232248362, "maximum": 0.768746536142107, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5487586285401371, "standard_deviation": 0.3092735128595181, "minimum": 0.15437017484921986, "maximum": 1.0372349647064645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05264395551389417, "standard_deviation": 0.005451237975333922, "minimum": 0.04522051731667153, "maximum": 0.06091268883842253, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1261333872047268, "standard_deviation": 0.029202057601524577, "minimum": 0.0673330108331381, "maximum": 0.16330194420124391, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.577983666189077, "standard_deviation": 0.006586887415049533, "minimum": 5.569575361672, "maximum": 5.592179315132031, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5433719215397423, "standard_deviation": 0.029320554782960064, "minimum": 0.49342623920222545, "maximum": 0.5911766842695771, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09963241238995015, "standard_deviation": 0.04760447962455909, "minimum": -0.16259232382314162, "maximum": -0.014068315640823292, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945092386880501, "standard_deviation": 0.00048106236223364465, "minimum": 0.9938082481936574, "maximum": 0.9950555990246008, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8592954845403507, "standard_deviation": 0.015225311530304207, "minimum": 0.8368719131959117, "maximum": 0.8769077365204866, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32333794116247255, "standard_deviation": 0.1398213375140141, "minimum": 0.1397100995572355, "maximum": 0.5544771775822205, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07384040760110322, "standard_deviation": 0.0032624532645260072, "minimum": 0.07051868350720628, "maximum": 0.07892798550064417, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983233671926466, "standard_deviation": 0.00024258486245795383, "minimum": 0.9977953228666822, "maximum": 0.9985278406831488, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4585158320949765, "standard_deviation": 0.017781074183027514, "minimum": 0.4226160896698999, "maximum": 0.4778944780435269, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40820443598248335, "standard_deviation": 0.2363736341802977, "minimum": 0.13790646108548646, "maximum": 0.7001006085692577, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04045500101872953, "standard_deviation": 0.0033189392756736774, "minimum": 0.03618805350536246, "maximum": 0.04716657351409233, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1320-122617-0019", "language": "en", "style": "normal", "speaker": "librispeech_1320", "source_utterance_id": "1320-122617-0019", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1320-122617-0019.wav", "source_file_sha256": "60a0d09ac62360c0c9b54ec961c07f697010c318bfb94607f03cc09f575c43bb", "decoded_pcm_sha256": "8e4bd7f99207ae6f91c9d166201c4877840bfbd64affa6d81029b74db79e47d6", "original_source_sha256": "3f0ea5a1e33f502dca87255f51487aa60fcb4f12605e9feba2adeb14df1d89f7", "corpus_entry_id": "librispeech_1320-122617-0019"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.940173864364624, "predicted_mos": 3.940173864364624, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1ea5ba5d2e83ffcaa7657ab2853ba086d5b370841572dfec823bddde3e610925", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.329569170000468, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1320-122617-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1ea5ba5d2e83ffcaa7657ab2853ba086d5b370841572dfec823bddde3e610925", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "f3f4aff3fa8a41019096118f74e8dcbcb298b16d0a3ada1d17bd3677fe04aee8"}, "rows": [{"case": "librispeech_1320-122617-0019", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.7263052463531494, "predicted_mos": 2.7263052463531494, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4133bd3dfb9121cf0ba92e3ec3eae90db7bfc64c237ad9fbf15d8479f072f39c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.147984241000813, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1320-122617-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "4133bd3dfb9121cf0ba92e3ec3eae90db7bfc64c237ad9fbf15d8479f072f39c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9699980313846934, "全帯域_10ms_rms_envelope_nrmse": 0.06155660369121225, "全帯域_10ms_envelope_correlation": 0.9956032883151251, "全帯域_10ms_modulation_mae_db": 0.1542280461544481, "全帯域_40ms_log_rms_mae_db": 0.5067071037530378, "全帯域_40ms_rms_envelope_nrmse": 0.03380637970138419, "全帯域_40ms_envelope_correlation": 0.9986368543225532, "全帯域_40ms_modulation_mae_db": 0.12952186734731874, "300–1000 Hz_10ms_log_rms_mae_db": 1.742994621054775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11429002440069011, "300–1000 Hz_10ms_envelope_correlation": 0.9901221687656706, "300–1000 Hz_10ms_modulation_mae_db": 0.34582059067067594, "300–1000 Hz_40ms_log_rms_mae_db": 1.035745062390327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0585945705300743, "300–1000 Hz_40ms_envelope_correlation": 0.9975131633339573, "300–1000 Hz_40ms_modulation_mae_db": 0.3448645349341127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.934785402430476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17988046848925116, "1000–3000 Hz_10ms_envelope_correlation": 0.9758345137120433, "1000–3000 Hz_10ms_modulation_mae_db": 0.8314440396314101, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1268636055705517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1304068871948252, "1000–3000 Hz_40ms_envelope_correlation": 0.9864612576607592, "1000–3000 Hz_40ms_modulation_mae_db": 0.90285898210464, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1622006172413872, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.38269922356494235, "3000–6000 Hz_10ms_envelope_correlation": 0.9038203640461441, "3000–6000 Hz_10ms_modulation_mae_db": 0.634879575756626, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4812351419351162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33857088865546486, "3000–6000 Hz_40ms_envelope_correlation": 0.9215531492506592, "3000–6000 Hz_40ms_modulation_mae_db": 1.0605388667811464, "mr_stft_db": 5.653004696794291, "band_mae_db": 0.204322652695741, "rms_gain_error_db": -0.09240114479901324, "output_peak": 0.60541673597898}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9449378658846225, "全帯域_10ms_rms_envelope_nrmse": 0.06510595224884068, "全帯域_10ms_envelope_correlation": 0.9950100767712176, "全帯域_10ms_modulation_mae_db": 0.1790286351566007, "全帯域_40ms_log_rms_mae_db": 0.4996811111948789, "全帯域_40ms_rms_envelope_nrmse": 0.03491775723886853, "全帯域_40ms_envelope_correlation": 0.9984613774274991, "全帯域_40ms_modulation_mae_db": 0.31082184496991627, "300–1000 Hz_10ms_log_rms_mae_db": 1.669930829343964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11704804127444275, "300–1000 Hz_10ms_envelope_correlation": 0.9894462869139693, "300–1000 Hz_10ms_modulation_mae_db": 0.5061796875547236, "300–1000 Hz_40ms_log_rms_mae_db": 0.9868020914029862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06442833306730593, "300–1000 Hz_40ms_envelope_correlation": 0.9966674045223439, "300–1000 Hz_40ms_modulation_mae_db": 0.44190215774417263, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0078817605739037, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17366574915524768, "1000–3000 Hz_10ms_envelope_correlation": 0.9785791020624864, "1000–3000 Hz_10ms_modulation_mae_db": 0.7982083552978123, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2321905162706182, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1258568433167636, "1000–3000 Hz_40ms_envelope_correlation": 0.9889664778679553, "1000–3000 Hz_40ms_modulation_mae_db": 0.5867848704403594, "3000–6000 Hz_10ms_log_rms_mae_db": 2.106882075937646, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34971937225808253, "3000–6000 Hz_10ms_envelope_correlation": 0.9267525245492054, "3000–6000 Hz_10ms_modulation_mae_db": 0.4504335682989889, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5065002648140609, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3049556525705816, "3000–6000 Hz_40ms_envelope_correlation": 0.9416499477688666, "3000–6000 Hz_40ms_modulation_mae_db": 0.9981412633990754, "mr_stft_db": 5.652984451403243, "band_mae_db": 0.2489696482527229, "rms_gain_error_db": -0.07409237554401263, "output_peak": 0.734263551071569}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0236831074481696, "全帯域_10ms_rms_envelope_nrmse": 0.0625560931451055, "全帯域_10ms_envelope_correlation": 0.9954596738098158, "全帯域_10ms_modulation_mae_db": 0.32073669609580185, "全帯域_40ms_log_rms_mae_db": 0.5603351386038935, "全帯域_40ms_rms_envelope_nrmse": 0.03287349264979121, "全帯域_40ms_envelope_correlation": 0.9987239497744524, "全帯域_40ms_modulation_mae_db": 0.27538508068545775, "300–1000 Hz_10ms_log_rms_mae_db": 1.7472382657290149, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11329201287363136, "300–1000 Hz_10ms_envelope_correlation": 0.9907571594515329, "300–1000 Hz_10ms_modulation_mae_db": 0.477465936836657, "300–1000 Hz_40ms_log_rms_mae_db": 1.0529560217104659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06923865033828054, "300–1000 Hz_40ms_envelope_correlation": 0.9969131433022695, "300–1000 Hz_40ms_modulation_mae_db": 0.5408410412767437, "1000–3000 Hz_10ms_log_rms_mae_db": 2.003534744900557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16691649260719393, "1000–3000 Hz_10ms_envelope_correlation": 0.9807435600585943, "1000–3000 Hz_10ms_modulation_mae_db": 1.3604932935364824, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2212910700539203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10701727419983738, "1000–3000 Hz_40ms_envelope_correlation": 0.9931567353007205, "1000–3000 Hz_40ms_modulation_mae_db": 1.499423041858595, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1003808893458182, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3763078500486068, "3000–6000 Hz_10ms_envelope_correlation": 0.9288168931242098, "3000–6000 Hz_10ms_modulation_mae_db": 0.7830335076269286, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4427760150821207, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3260947906670589, "3000–6000 Hz_40ms_envelope_correlation": 0.945897477318612, "3000–6000 Hz_40ms_modulation_mae_db": 0.87565265603167, "mr_stft_db": 5.65145189210357, "band_mae_db": 0.36844954650231276, "rms_gain_error_db": -0.09616348195538635, "output_peak": 0.6038095523450902}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.962334459702632, "全帯域_10ms_rms_envelope_nrmse": 0.057631357239050066, "全帯域_10ms_envelope_correlation": 0.9961377077777319, "全帯域_10ms_modulation_mae_db": 0.26954636215650546, "全帯域_40ms_log_rms_mae_db": 0.5490754216394812, "全帯域_40ms_rms_envelope_nrmse": 0.030313612539195442, "全帯域_40ms_envelope_correlation": 0.9989003303174732, "全帯域_40ms_modulation_mae_db": 0.35651114254648725, "300–1000 Hz_10ms_log_rms_mae_db": 1.7194071629992285, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11492612453680959, "300–1000 Hz_10ms_envelope_correlation": 0.9903382201075648, "300–1000 Hz_10ms_modulation_mae_db": 0.2904013313327569, "300–1000 Hz_40ms_log_rms_mae_db": 1.03636996078855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07670615722995834, "300–1000 Hz_40ms_envelope_correlation": 0.9958301511563568, "300–1000 Hz_40ms_modulation_mae_db": 0.5480302401688748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9591940086756994, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16244895624604777, "1000–3000 Hz_10ms_envelope_correlation": 0.9815613142817593, "1000–3000 Hz_10ms_modulation_mae_db": 1.0041838733728408, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1898655822991868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10962545884322088, "1000–3000 Hz_40ms_envelope_correlation": 0.9922347871025041, "1000–3000 Hz_40ms_modulation_mae_db": 1.0422548702942767, "3000–6000 Hz_10ms_log_rms_mae_db": 2.105775876786633, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3585912216543883, "3000–6000 Hz_10ms_envelope_correlation": 0.9314288656459796, "3000–6000 Hz_10ms_modulation_mae_db": 0.8321234545910982, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5094704973131536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31645089359376666, "3000–6000 Hz_40ms_envelope_correlation": 0.9449512378257925, "3000–6000 Hz_40ms_modulation_mae_db": 1.470048448940154, "mr_stft_db": 5.668961302282639, "band_mae_db": 0.33505983430002395, "rms_gain_error_db": -0.08763490557968977, "output_peak": 0.5800521522888112}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.00256091118249, "全帯域_10ms_rms_envelope_nrmse": 0.06392140297570117, "全帯域_10ms_envelope_correlation": 0.9951859350303359, "全帯域_10ms_modulation_mae_db": 0.3627246876798111, "全帯域_40ms_log_rms_mae_db": 0.5247989770499838, "全帯域_40ms_rms_envelope_nrmse": 0.034504187898200006, "全帯域_40ms_envelope_correlation": 0.9984885578417164, "全帯域_40ms_modulation_mae_db": 0.4115738042208105, "300–1000 Hz_10ms_log_rms_mae_db": 1.7671623414513367, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12399079247514098, "300–1000 Hz_10ms_envelope_correlation": 0.9884107004392059, "300–1000 Hz_10ms_modulation_mae_db": 0.6888234966667236, "300–1000 Hz_40ms_log_rms_mae_db": 1.135335374025881, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07199791538461216, "300–1000 Hz_40ms_envelope_correlation": 0.9961617621490557, "300–1000 Hz_40ms_modulation_mae_db": 0.6295314836017265, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0045019789889604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15454365354213137, "1000–3000 Hz_10ms_envelope_correlation": 0.9835766429622386, "1000–3000 Hz_10ms_modulation_mae_db": 0.8559432608092928, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2033832091127292, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09593855669967535, "1000–3000 Hz_40ms_envelope_correlation": 0.9946880295672674, "1000–3000 Hz_40ms_modulation_mae_db": 0.8712695810113184, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0889828912993926, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3787460110038196, "3000–6000 Hz_10ms_envelope_correlation": 0.9201024474753107, "3000–6000 Hz_10ms_modulation_mae_db": 0.6422362625253129, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4761180047502773, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.314412168576516, "3000–6000 Hz_40ms_envelope_correlation": 0.9432093361278598, "3000–6000 Hz_40ms_modulation_mae_db": 0.9810896803893675, "mr_stft_db": 5.662633283974, "band_mae_db": 0.29820374406637784, "rms_gain_error_db": -0.06990553597935924, "output_peak": 0.6042441073048832}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9591984886017599, "全帯域_10ms_rms_envelope_nrmse": 0.06251714100930278, "全帯域_10ms_envelope_correlation": 0.9954447130677914, "全帯域_10ms_modulation_mae_db": 0.3095234066853458, "全帯域_40ms_log_rms_mae_db": 0.4530378116640235, "全帯域_40ms_rms_envelope_nrmse": 0.03207928167049498, "全帯域_40ms_envelope_correlation": 0.9987671636189355, "全帯域_40ms_modulation_mae_db": 0.49272579349059004, "300–1000 Hz_10ms_log_rms_mae_db": 1.7815393507690924, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12075410721199006, "300–1000 Hz_10ms_envelope_correlation": 0.9889293945705361, "300–1000 Hz_10ms_modulation_mae_db": 0.6894877878669595, "300–1000 Hz_40ms_log_rms_mae_db": 1.1200498774069416, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0661190147730658, "300–1000 Hz_40ms_envelope_correlation": 0.9967305323953021, "300–1000 Hz_40ms_modulation_mae_db": 0.77158516264302, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0171367178753106, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16593770886089096, "1000–3000 Hz_10ms_envelope_correlation": 0.98064159527362, "1000–3000 Hz_10ms_modulation_mae_db": 0.8608904273436044, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2597567534617407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11025136559769201, "1000–3000 Hz_40ms_envelope_correlation": 0.9920947538717084, "1000–3000 Hz_40ms_modulation_mae_db": 0.9022618618099217, "3000–6000 Hz_10ms_log_rms_mae_db": 2.119270141640057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3440719532730852, "3000–6000 Hz_10ms_envelope_correlation": 0.9262077096446143, "3000–6000 Hz_10ms_modulation_mae_db": 0.538947159410511, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4877094238128161, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30883863473925294, "3000–6000 Hz_40ms_envelope_correlation": 0.9375237130887099, "3000–6000 Hz_40ms_modulation_mae_db": 0.9195818528097043, "mr_stft_db": 5.672806312168185, "band_mae_db": 0.19312621289823872, "rms_gain_error_db": -0.09214808991617895, "output_peak": 0.5899759386466619}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9837689175004365, "全帯域_10ms_rms_envelope_nrmse": 0.0705852934121439, "全帯域_10ms_envelope_correlation": 0.9943263233621829, "全帯域_10ms_modulation_mae_db": 0.25121550365895834, "全帯域_40ms_log_rms_mae_db": 0.48509573146971335, "全帯域_40ms_rms_envelope_nrmse": 0.04328832499043587, "全帯域_40ms_envelope_correlation": 0.9978488167166407, "全帯域_40ms_modulation_mae_db": 0.1958026026921156, "300–1000 Hz_10ms_log_rms_mae_db": 1.769259720208161, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14154560787345258, "300–1000 Hz_10ms_envelope_correlation": 0.9849945070952913, "300–1000 Hz_10ms_modulation_mae_db": 0.4650083055034387, "300–1000 Hz_40ms_log_rms_mae_db": 1.068957320964803, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09291423082349123, "300–1000 Hz_40ms_envelope_correlation": 0.9935955037263131, "300–1000 Hz_40ms_modulation_mae_db": 0.4499770165116875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.983776234093266, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16101954157451742, "1000–3000 Hz_10ms_envelope_correlation": 0.9812969448214764, "1000–3000 Hz_10ms_modulation_mae_db": 0.6507116253681925, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1890798835339893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09719403201679895, "1000–3000 Hz_40ms_envelope_correlation": 0.9935658821236004, "1000–3000 Hz_40ms_modulation_mae_db": 0.7241529142164148, "3000–6000 Hz_10ms_log_rms_mae_db": 2.096505313059206, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3532335395045104, "3000–6000 Hz_10ms_envelope_correlation": 0.9217395822658161, "3000–6000 Hz_10ms_modulation_mae_db": 0.32216745697824933, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4707676426726197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3118471028144287, "3000–6000 Hz_40ms_envelope_correlation": 0.9358988816773918, "3000–6000 Hz_40ms_modulation_mae_db": 0.7295195803916492, "mr_stft_db": 5.658282148393442, "band_mae_db": 0.14441777532678737, "rms_gain_error_db": -0.1479400117628711, "output_peak": 0.6027461954722504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9912094802899868, "全帯域_10ms_rms_envelope_nrmse": 0.06657717050317684, "全帯域_10ms_envelope_correlation": 0.9947455259485634, "全帯域_10ms_modulation_mae_db": 0.34834483203316663, "全帯域_40ms_log_rms_mae_db": 0.5079927491860717, "全帯域_40ms_rms_envelope_nrmse": 0.038228327450218674, "全帯域_40ms_envelope_correlation": 0.9980900820802827, "全帯域_40ms_modulation_mae_db": 0.3297705915152889, "300–1000 Hz_10ms_log_rms_mae_db": 1.7286605980772836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10439279900949129, "300–1000 Hz_10ms_envelope_correlation": 0.991584150730642, "300–1000 Hz_10ms_modulation_mae_db": 0.38882383897406325, "300–1000 Hz_40ms_log_rms_mae_db": 1.0546096832496676, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06678978100547449, "300–1000 Hz_40ms_envelope_correlation": 0.9962870078190751, "300–1000 Hz_40ms_modulation_mae_db": 0.4498200089235127, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0140724156140957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15288582677078108, "1000–3000 Hz_10ms_envelope_correlation": 0.9832752077928142, "1000–3000 Hz_10ms_modulation_mae_db": 1.2167370588132742, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1755142872692317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09217225987988703, "1000–3000 Hz_40ms_envelope_correlation": 0.9943582165469618, "1000–3000 Hz_40ms_modulation_mae_db": 0.9091686206283606, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0971464064780707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40204040633712296, "3000–6000 Hz_10ms_envelope_correlation": 0.9246247492191096, "3000–6000 Hz_10ms_modulation_mae_db": 0.565568199769993, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4424797310116977, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3480428348298606, "3000–6000 Hz_40ms_envelope_correlation": 0.9438979526550331, "3000–6000 Hz_40ms_modulation_mae_db": 1.2650509209165188, "mr_stft_db": 5.66891579639704, "band_mae_db": 0.3594709569141657, "rms_gain_error_db": -0.0476402636408323, "output_peak": 0.7340477204479791}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.895862524805004, "spectral_convergence": 0.1907480660016933, "spectral_convergence_status": "ok", "amplitude_floor": 1.884454455950008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.520975288446642, "spectral_convergence": 0.19902597406456737, "spectral_convergence_status": "ok", "amplitude_floor": 1.301578596614512e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.542176277131226, "spectral_convergence": 0.19739050234938088, "spectral_convergence_status": "ok", "amplitude_floor": 1.1393421242566376e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004137534659655858, "output_power": 0.004053405007921461, "delta_db": -0.08921645109996804, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026137240204796635, "output_power": 0.0025738934406898265, "delta_db": -0.06669165980531205, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008071793462917449, "output_power": 0.0007862982786166411, "delta_db": -0.11382715958107592, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004252507832785635, "output_power": 0.00040880656014168765, "delta_db": -0.17127265891325016, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.026398390024539e-05, "output_power": 6.773390095818792e-05, "delta_db": -0.1592668184325663, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.066875275629503e-05, "output_power": 4.387065283542695e-05, "delta_db": -0.6256611683422736, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.204322652695741, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09240114479901324, "pitch": null, "output_rms_dbfs": -21.005073746493352, "output_peak": 0.60541673597898, "reference_rms_dbfs": -20.91267260169434, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.653004696794291}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025953969251226393, "log_rms_mae_db": 0.9699980313846934, "rms_envelope_nrmse": 0.06155660369121225, "envelope_correlation": 0.9956032883151251, "rms_gain_error_db": -0.09239923425504576, "reference_mean_power": 0.008109529837298932, "output_mean_power": 0.007938816298093844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5635786081597607e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013500334014131179, "output_power": 0.0013437195207190182, "reference_active": true, "delta_db": -0.020358872143521924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020457869490256438, "output_power": 0.00018651572927474533, "reference_active": true, "delta_db": -0.40144941044105936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3880284077345985e-05, "output_power": 3.3562898343867104e-05, "reference_active": true, "delta_db": -0.04087585587876295}], "active_band_count": 3, "active_band_mae_db": 0.1542280461544481}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024049824298809385, "log_rms_mae_db": 0.5067071037530378, "rms_envelope_nrmse": 0.03380637970138419, "envelope_correlation": 0.9986368543225532, "rms_gain_error_db": -0.09273061594822976, "reference_mean_power": 0.008139876536629037, "output_mean_power": 0.007967916166979018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2818780696338306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012006574574251973, "output_power": 0.0012021966909156103, "reference_active": true, "delta_db": 0.005564055589022027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.159020687440259e-05, "output_power": 8.42820954561216e-05, "reference_active": true, "delta_db": -0.36113715280170605}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2353163446294293e-06, "output_power": 2.246598336919638e-06, "reference_active": true, "delta_db": 0.02186439365122815}], "active_band_count": 3, "active_band_mae_db": 0.12952186734731874}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002074995834794702, "log_rms_mae_db": 1.742994621054775, "rms_envelope_nrmse": 0.11429002440069011, "envelope_correlation": 0.9901221687656706, "rms_gain_error_db": -0.19193045669850559, "reference_mean_power": 0.0020359851811281022, "output_mean_power": 0.0019479668729220753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.345568889858965e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045532158312677247, "output_power": 0.0004336200480733782, "reference_active": true, "delta_db": -0.212088839476844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.561332252218219e-05, "output_power": 7.153962865506403e-05, "reference_active": true, "delta_db": -0.7799467018156874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0669601144719856e-05, "output_power": 1.0558581139336061e-05, "reference_active": true, "delta_db": -0.04542623071949649}], "active_band_count": 3, "active_band_mae_db": 0.34582059067067594}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001833599147850085, "log_rms_mae_db": 1.035745062390327, "rms_envelope_nrmse": 0.0585945705300743, "envelope_correlation": 0.9975131633339573, "rms_gain_error_db": -0.19181851254689752, "reference_mean_power": 0.0020457391187242045, "output_mean_power": 0.001957349587185854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.75393473065512e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00043556686984963635, "output_power": 0.00041293763103071113, "reference_active": true, "delta_db": -0.2317037701021627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.69121464262054e-05, "output_power": 3.202769085959868e-05, "reference_active": true, "delta_db": -0.6164367205506136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.183033529622033e-07, "output_power": 6.881174674038584e-07, "reference_active": true, "delta_db": -0.1864531141495618}], "active_band_count": 3, "active_band_mae_db": 0.3448645349341127}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011093565856595723, "log_rms_mae_db": 1.934785402430476, "rms_envelope_nrmse": 0.17988046848925116, "envelope_correlation": 0.9758345137120433, "rms_gain_error_db": -0.18038613424526512, "reference_mean_power": 0.00042557205365836593, "output_mean_power": 0.00040825779693690696}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.99964400169894e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011758808293094384, "output_power": 0.00011007303191549411, "reference_active": true, "delta_db": -0.2868238090928182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1475307548341578e-05, "output_power": 1.841944794202687e-05, "reference_active": true, "delta_db": -0.6666278247756935}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0561557218273926e-06, "output_power": 3.5459482452302297e-06, "reference_active": true, "delta_db": -1.5408804850257185}], "active_band_count": 3, "active_band_mae_db": 0.8314440396314101}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010031642073947198, "log_rms_mae_db": 1.1268636055705517, "rms_envelope_nrmse": 0.1304068871948252, "envelope_correlation": 0.9864612576607592, "rms_gain_error_db": -0.180461218818227, "reference_mean_power": 0.0004276640413209328, "output_mean_power": 0.0004102575798329056}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8247169971067077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011139646210262638, "output_power": 0.00010435001863596478, "reference_active": true, "delta_db": -0.283788671130138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0378224135538677e-05, "output_power": 9.144164327372641e-06, "reference_active": true, "delta_db": -0.5497902374462831}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.220864105203889e-07, "output_power": 2.0915709327518967e-07, "reference_active": true, "delta_db": -1.8749980377374988}], "active_band_count": 3, "active_band_mae_db": 0.90285898210464}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.985234496629778e-05, "log_rms_mae_db": 2.1622006172413872, "rms_envelope_nrmse": 0.38269922356494235, "envelope_correlation": 0.9038203640461441, "rms_gain_error_db": -0.5566596758089923, "reference_mean_power": 6.361685525644655e-05, "output_mean_power": 5.5963675464590496e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.697638884988847e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.1912782580556425e-06, "output_power": 5.097658539016891e-06, "reference_active": true, "delta_db": -0.8440958181447702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.016318108845029e-06, "output_power": 1.668325276515244e-06, "reference_active": true, "delta_db": -0.8227832083066343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.069070181696055e-07, "output_power": 6.410591888906157e-07, "reference_active": true, "delta_db": 0.23775970081847364}], "active_band_count": 3, "active_band_mae_db": 0.634879575756626}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.357134521214995e-05, "log_rms_mae_db": 1.4812351419351162, "rms_envelope_nrmse": 0.33857088865546486, "envelope_correlation": 0.9215531492506592, "rms_gain_error_db": -0.5568392719223408, "reference_mean_power": 6.39232761364752e-05, "output_mean_power": 5.623090817063184e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.609918878818377e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.077093160223491e-06, "output_power": 5.080372507702856e-06, "reference_active": true, "delta_db": -0.7780033683016674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.511523340291838e-07, "output_power": 8.729003813763659e-07, "reference_active": true, "delta_db": -0.37285394718388903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7938163224025654e-08, "output_power": 4.459372782151083e-08, "reference_active": true, "delta_db": 2.0307592848578824}], "active_band_count": 3, "active_band_mae_db": 1.0605388667811464}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8944045737033754, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "64596c75385753bbb7c595b41b2bf1083871d9cc9186b05c6f789d3605a30172", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.980688610120629, "standard_deviation": 0.0025122673582646114, "minimum": 0.9758345137120433, "maximum": 0.9835766429622386, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9906104078940337, "standard_deviation": 0.02942921796884901, "minimum": 1.934785402430476, "maximum": 2.0171367178753106, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9473264917716137, "standard_deviation": 0.23482183215692579, "minimum": 0.6507116253681925, "maximum": 1.3604932935364824, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16466229965575768, "standard_deviation": 0.00909162651099951, "minimum": 0.15288582677078108, "maximum": 0.17988046848925116, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919407675051846, "standard_deviation": 0.0028408416815163744, "minimum": 0.9864612576607592, "maximum": 0.9946880295672674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.199743113446496, "standard_deviation": 0.040074509975127154, "minimum": 1.1268636055705517, "maximum": 1.2597567534617407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9297718427954859, "standard_deviation": 0.26790249707858677, "minimum": 0.5867848704403594, "maximum": 1.499423041858595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10855783471858754, "standard_deviation": 0.013835189294845751, "minimum": 0.09217225987988703, "maximum": 0.1304068871948252, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9229366419962987, "standard_deviation": 0.0085331576662984, "minimum": 0.9038203640461441, "maximum": 0.9314288656459796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1096430264735266, "standard_deviation": 0.02304948956040821, "minimum": 2.0889828912993926, "maximum": 2.1622006172413872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5961736481197135, "standard_deviation": 0.16679326796869082, "minimum": 0.32216745697824933, "maximum": 0.8321234545910982, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.36817619720556977, "standard_deviation": 0.01990108193260079, "minimum": 0.3440719532730852, "maximum": 0.40204040633712296, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9393227119641157, "standard_deviation": 0.007991771785643488, "minimum": 0.9215531492506592, "maximum": 0.945897477318612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4771320901739828, "standard_deviation": 0.025240663708515167, "minimum": 1.4424797310116977, "maximum": 1.5094704973131536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0374529087074107, "standard_deviation": 0.23271404257655395, "minimum": 0.7295195803916492, "maximum": 1.470048448940154, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.32115162080586623, "standard_deviation": 0.015208143180824208, "minimum": 0.3049556525705816, "maximum": 0.3480428348298606, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9893228235093017, "standard_deviation": 0.0020205099959005146, "minimum": 0.9849945070952913, "maximum": 0.991584150730642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7407741112041069, "standard_deviation": 0.035533371821432964, "minimum": 1.669930829343964, "maximum": 1.7815393507690924, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4815013719257498, "standard_deviation": 0.14670464151410567, "minimum": 0.2904013313327569, "maximum": 0.6894877878669595, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11877993870695609, "standard_deviation": 0.010852543035468457, "minimum": 0.10439279900949129, "maximum": 0.14154560787345258, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962123335505841, "standard_deviation": 0.0011747891902107643, "minimum": 0.9935955037263131, "maximum": 0.9975131633339573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0613531739924529, "standard_deviation": 0.047704865034346795, "minimum": 0.9868020914029862, "maximum": 1.135335374025881, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5220689557254813, "standard_deviation": 0.13259939245612967, "minimum": 0.3448645349341127, "maximum": 0.77158516264302, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07084858164403285, "standard_deviation": 0.010387767241814031, "minimum": 0.0585945705300743, "maximum": 0.09291423082349123, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2690025463695463, "standard_deviation": 0.08379729213744255, "minimum": 0.14441777532678737, "maximum": 0.36844954650231276, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.661129985439551, "standard_deviation": 0.008406550123791042, "minimum": 5.65145189210357, "maximum": 5.672806312168185, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6318194941945281, "standard_deviation": 0.0637668263572502, "minimum": 0.5800521522888112, "maximum": 0.734263551071569, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08849072614716794, "standard_deviation": 0.028893822708994946, "minimum": -0.1479400117628711, "maximum": -0.0476402636408323, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952391555103455, "standard_deviation": 0.0005558922390129354, "minimum": 0.9943263233621829, "maximum": 0.9961377077777319, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.979711407749349, "standard_deviation": 0.025719826450946046, "minimum": 0.9449378658846225, "maximum": 1.0236831074481696, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27441852120257976, "standard_deviation": 0.0762963239988688, "minimum": 0.1542280461544481, "maximum": 0.3627246876798111, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06380637677806665, "standard_deviation": 0.003812588534037896, "minimum": 0.057631357239050066, "maximum": 0.0705852934121439, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984896415124441, "standard_deviation": 0.0003572434101879086, "minimum": 0.9978488167166407, "maximum": 0.9989003303174732, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5108405055701355, "standard_deviation": 0.034351415146858064, "minimum": 0.4530378116640235, "maximum": 0.5603351386038935, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31276409093349816, "standard_deviation": 0.11527065317081772, "minimum": 0.12952186734731874, "maximum": 0.49272579349059004, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03500142051732361, "standard_deviation": 0.004067477150043216, "minimum": 0.030313612539195442, "maximum": 0.04328832499043587, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0019", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5160484313964844, "predicted_mos": 2.5160484313964844, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "08904a7933abf80f5dd45c86b73726f73f6075d13193559d3cd6771083ca01c0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.3236332000233233, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9762303941447787, "全帯域_10ms_rms_envelope_nrmse": 0.0613460963229519, "全帯域_10ms_envelope_correlation": 0.9956247246358841, "全帯域_10ms_modulation_mae_db": 0.15368337718961575, "全帯域_40ms_log_rms_mae_db": 0.5119598828770303, "全帯域_40ms_rms_envelope_nrmse": 0.03300748385786052, "全帯域_40ms_envelope_correlation": 0.9986940582361267, "全帯域_40ms_modulation_mae_db": 0.14675634241593327, "300–1000 Hz_10ms_log_rms_mae_db": 1.7234242105125452, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11200119993898285, "300–1000 Hz_10ms_envelope_correlation": 0.9905439871362841, "300–1000 Hz_10ms_modulation_mae_db": 0.3150184069691221, "300–1000 Hz_40ms_log_rms_mae_db": 0.9829146709899571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06006396813832496, "300–1000 Hz_40ms_envelope_correlation": 0.9973837341090152, "300–1000 Hz_40ms_modulation_mae_db": 0.29862665922721304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8926501765479742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16602172055287473, "1000–3000 Hz_10ms_envelope_correlation": 0.979559973958332, "1000–3000 Hz_10ms_modulation_mae_db": 0.9041095933768705, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0695683457301588, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11735799836923402, "1000–3000 Hz_40ms_envelope_correlation": 0.989258511197568, "1000–3000 Hz_40ms_modulation_mae_db": 0.9607949199898901, "3000–6000 Hz_10ms_log_rms_mae_db": 2.129081959715819, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37908221361620725, "3000–6000 Hz_10ms_envelope_correlation": 0.905911010337964, "3000–6000 Hz_10ms_modulation_mae_db": 0.6588776049800721, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5189130716858257, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33613618551507474, "3000–6000 Hz_40ms_envelope_correlation": 0.9228333300848915, "3000–6000 Hz_40ms_modulation_mae_db": 1.0358723273740535, "mr_stft_db": 5.561981581481649, "band_mae_db": 0.23547199054882248, "rms_gain_error_db": -0.08912057245865808, "output_peak": 0.65567952639057}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9568265379861665, "全帯域_10ms_rms_envelope_nrmse": 0.07119138753968468, "全帯域_10ms_envelope_correlation": 0.9939910172828467, "全帯域_10ms_modulation_mae_db": 0.1831658127928474, "全帯域_40ms_log_rms_mae_db": 0.5107509252112326, "全帯域_40ms_rms_envelope_nrmse": 0.038754163836051875, "全帯域_40ms_envelope_correlation": 0.9980383527011876, "全帯域_40ms_modulation_mae_db": 0.34437157195382656, "300–1000 Hz_10ms_log_rms_mae_db": 1.6888224022811373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12249283575509987, "300–1000 Hz_10ms_envelope_correlation": 0.9883902674855768, "300–1000 Hz_10ms_modulation_mae_db": 0.4603748040754721, "300–1000 Hz_40ms_log_rms_mae_db": 0.9592396211680708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06552018167465357, "300–1000 Hz_40ms_envelope_correlation": 0.9964281456219498, "300–1000 Hz_40ms_modulation_mae_db": 0.3795693755396652, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9488805532133229, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16850678078714662, "1000–3000 Hz_10ms_envelope_correlation": 0.9800906915383768, "1000–3000 Hz_10ms_modulation_mae_db": 0.8165749383639055, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1506190670670289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12016828292546221, "1000–3000 Hz_40ms_envelope_correlation": 0.9903182483018222, "1000–3000 Hz_40ms_modulation_mae_db": 0.5730523755186426, "3000–6000 Hz_10ms_log_rms_mae_db": 2.090965575390424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3519640977805129, "3000–6000 Hz_10ms_envelope_correlation": 0.9250161043874614, "3000–6000 Hz_10ms_modulation_mae_db": 0.34223054484236043, "3000–6000 Hz_40ms_log_rms_mae_db": 1.529223679849306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30763058625182604, "3000–6000 Hz_40ms_envelope_correlation": 0.9398997015623055, "3000–6000 Hz_40ms_modulation_mae_db": 0.9524033755915994, "mr_stft_db": 5.566716221350546, "band_mae_db": 0.2505109286125279, "rms_gain_error_db": -0.05203276839727384, "output_peak": 0.7771159982769201}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0151201562144685, "全帯域_10ms_rms_envelope_nrmse": 0.0623093567119182, "全帯域_10ms_envelope_correlation": 0.9954673261013283, "全帯域_10ms_modulation_mae_db": 0.2874748135535407, "全帯域_40ms_log_rms_mae_db": 0.5452619446812628, "全帯域_40ms_rms_envelope_nrmse": 0.03228965687755658, "全帯域_40ms_envelope_correlation": 0.9987362422095636, "全帯域_40ms_modulation_mae_db": 0.29414725551013404, "300–1000 Hz_10ms_log_rms_mae_db": 1.7648939593647415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11745244151678434, "300–1000 Hz_10ms_envelope_correlation": 0.9897354333490006, "300–1000 Hz_10ms_modulation_mae_db": 0.4700217538885676, "300–1000 Hz_40ms_log_rms_mae_db": 1.0245219464027306, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06769202341181489, "300–1000 Hz_40ms_envelope_correlation": 0.9967841908727473, "300–1000 Hz_40ms_modulation_mae_db": 0.6174347798208873, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9644533307899268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1651912714303588, "1000–3000 Hz_10ms_envelope_correlation": 0.9814842428979305, "1000–3000 Hz_10ms_modulation_mae_db": 1.38339401496782, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1785232727982684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10791127041643894, "1000–3000 Hz_40ms_envelope_correlation": 0.9933571746679962, "1000–3000 Hz_40ms_modulation_mae_db": 1.5215170959013136, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0771628201263934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3797707881878971, "3000–6000 Hz_10ms_envelope_correlation": 0.930379799280367, "3000–6000 Hz_10ms_modulation_mae_db": 0.7104905793979848, "3000–6000 Hz_40ms_log_rms_mae_db": 1.459730855301974, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3293440130630198, "3000–6000 Hz_40ms_envelope_correlation": 0.9475407192363379, "3000–6000 Hz_40ms_modulation_mae_db": 0.9992702926689758, "mr_stft_db": 5.557873679433842, "band_mae_db": 0.40486127718448134, "rms_gain_error_db": -0.08558018658797747, "output_peak": 0.6074235203580737}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9879564768464574, "全帯域_10ms_rms_envelope_nrmse": 0.06092039171778375, "全帯域_10ms_envelope_correlation": 0.9956369376084263, "全帯域_10ms_modulation_mae_db": 0.1566326760337464, "全帯域_40ms_log_rms_mae_db": 0.571277118043975, "全帯域_40ms_rms_envelope_nrmse": 0.03133921224377833, "全帯域_40ms_envelope_correlation": 0.9987705662960985, "全帯域_40ms_modulation_mae_db": 0.27547591745944544, "300–1000 Hz_10ms_log_rms_mae_db": 1.6975082197423772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10936527857782313, "300–1000 Hz_10ms_envelope_correlation": 0.9910808801789597, "300–1000 Hz_10ms_modulation_mae_db": 0.3813511114372751, "300–1000 Hz_40ms_log_rms_mae_db": 0.9763806351935983, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06790934974309433, "300–1000 Hz_40ms_envelope_correlation": 0.9966161029479029, "300–1000 Hz_40ms_modulation_mae_db": 0.43197119040785603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8899333256576953, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14943657913352934, "1000–3000 Hz_10ms_envelope_correlation": 0.9848050214585266, "1000–3000 Hz_10ms_modulation_mae_db": 1.006484874578779, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1278964166229177, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0967065445896488, "1000–3000 Hz_40ms_envelope_correlation": 0.9945657230026685, "1000–3000 Hz_40ms_modulation_mae_db": 1.04770802512072, "3000–6000 Hz_10ms_log_rms_mae_db": 2.100701335075389, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3503037505917815, "3000–6000 Hz_10ms_envelope_correlation": 0.931611941979882, "3000–6000 Hz_10ms_modulation_mae_db": 0.7832030762244603, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5246900527218739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3099490794863844, "3000–6000 Hz_40ms_envelope_correlation": 0.9443314653175506, "3000–6000 Hz_40ms_modulation_mae_db": 1.4404252112469305, "mr_stft_db": 5.582884404087775, "band_mae_db": 0.32654372300128554, "rms_gain_error_db": -0.07070084225762514, "output_peak": 0.579974323856633}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0008324431763693, "全帯域_10ms_rms_envelope_nrmse": 0.06654050775570239, "全帯域_10ms_envelope_correlation": 0.9947804354020826, "全帯域_10ms_modulation_mae_db": 0.391378065542668, "全帯域_40ms_log_rms_mae_db": 0.530071029864828, "全帯域_40ms_rms_envelope_nrmse": 0.03596189797788359, "全帯域_40ms_envelope_correlation": 0.9983565601435418, "全帯域_40ms_modulation_mae_db": 0.4777953487128284, "300–1000 Hz_10ms_log_rms_mae_db": 1.7262256923765849, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12913279756526602, "300–1000 Hz_10ms_envelope_correlation": 0.9873675115139641, "300–1000 Hz_10ms_modulation_mae_db": 0.7037137151854312, "300–1000 Hz_40ms_log_rms_mae_db": 1.0572362875143515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07211068348752764, "300–1000 Hz_40ms_envelope_correlation": 0.996143556225223, "300–1000 Hz_40ms_modulation_mae_db": 0.6636958043753386, "1000–3000 Hz_10ms_log_rms_mae_db": 1.936364993134902, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14705419316679752, "1000–3000 Hz_10ms_envelope_correlation": 0.9856570066975723, "1000–3000 Hz_10ms_modulation_mae_db": 0.8568644116852484, "1000–3000 Hz_40ms_log_rms_mae_db": 1.120277966528724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09067125962391862, "1000–3000 Hz_40ms_envelope_correlation": 0.9958995228124989, "1000–3000 Hz_40ms_modulation_mae_db": 0.8862538978938438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.063061795206431, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3807586979724201, "3000–6000 Hz_10ms_envelope_correlation": 0.920993312463831, "3000–6000 Hz_10ms_modulation_mae_db": 0.5759470685352809, "3000–6000 Hz_40ms_log_rms_mae_db": 1.477011690494641, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31364488772311955, "3000–6000 Hz_40ms_envelope_correlation": 0.9450802205672051, "3000–6000 Hz_40ms_modulation_mae_db": 0.9718324153258003, "mr_stft_db": 5.571645955591987, "band_mae_db": 0.3561372555651361, "rms_gain_error_db": -0.07226458907951884, "output_peak": 0.609933577680188}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9503497061515834, "全帯域_10ms_rms_envelope_nrmse": 0.06383023613657454, "全帯域_10ms_envelope_correlation": 0.9952458925691466, "全帯域_10ms_modulation_mae_db": 0.19171033394661982, "全帯域_40ms_log_rms_mae_db": 0.44894419009367037, "全帯域_40ms_rms_envelope_nrmse": 0.03416562737296816, "全帯域_40ms_envelope_correlation": 0.9985826644205467, "全帯域_40ms_modulation_mae_db": 0.3416303796829571, "300–1000 Hz_10ms_log_rms_mae_db": 1.7805655586920321, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12434586883696692, "300–1000 Hz_10ms_envelope_correlation": 0.9882918997158285, "300–1000 Hz_10ms_modulation_mae_db": 0.7171839907138233, "300–1000 Hz_40ms_log_rms_mae_db": 1.0520797205452335, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06744164520308449, "300–1000 Hz_40ms_envelope_correlation": 0.9966636070892632, "300–1000 Hz_40ms_modulation_mae_db": 0.7215632064887227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9614850372814503, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16087859087233655, "1000–3000 Hz_10ms_envelope_correlation": 0.9821127819084515, "1000–3000 Hz_10ms_modulation_mae_db": 0.9070485275263828, "1000–3000 Hz_40ms_log_rms_mae_db": 1.182372404653839, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10445839699552174, "1000–3000 Hz_40ms_envelope_correlation": 0.9933656220880416, "1000–3000 Hz_40ms_modulation_mae_db": 0.9763206102075633, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0724327614597406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34427368103403116, "3000–6000 Hz_10ms_envelope_correlation": 0.9256671736853863, "3000–6000 Hz_10ms_modulation_mae_db": 0.39269689537100216, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4801928754125981, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3088793769168951, "3000–6000 Hz_40ms_envelope_correlation": 0.9370907469340876, "3000–6000 Hz_40ms_modulation_mae_db": 0.8164821670574524, "mr_stft_db": 5.573550790039945, "band_mae_db": 0.21156446870955026, "rms_gain_error_db": -0.09210889459152957, "output_peak": 0.5893460193260659}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.971462626220445, "全帯域_10ms_rms_envelope_nrmse": 0.0720266868221312, "全帯域_10ms_envelope_correlation": 0.9940330560589876, "全帯域_10ms_modulation_mae_db": 0.2548227044014776, "全帯域_40ms_log_rms_mae_db": 0.5008261062175516, "全帯域_40ms_rms_envelope_nrmse": 0.0425769439019841, "全帯域_40ms_envelope_correlation": 0.9978752331979398, "全帯域_40ms_modulation_mae_db": 0.162722766228645, "300–1000 Hz_10ms_log_rms_mae_db": 1.7691357985198346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1424293008707456, "300–1000 Hz_10ms_envelope_correlation": 0.9846093917932288, "300–1000 Hz_10ms_modulation_mae_db": 0.4368741664874148, "300–1000 Hz_40ms_log_rms_mae_db": 1.0115814879520078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09001176381502, "300–1000 Hz_40ms_envelope_correlation": 0.9938137368605228, "300–1000 Hz_40ms_modulation_mae_db": 0.511734409006676, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9274997524348438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15693482090383534, "1000–3000 Hz_10ms_envelope_correlation": 0.9826021596477321, "1000–3000 Hz_10ms_modulation_mae_db": 0.648183147473739, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1355679217076493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08994082294901565, "1000–3000 Hz_40ms_envelope_correlation": 0.9950986461443598, "1000–3000 Hz_40ms_modulation_mae_db": 0.678443377918545, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0545706636730485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3500636441190373, "3000–6000 Hz_10ms_envelope_correlation": 0.9220795095573082, "3000–6000 Hz_10ms_modulation_mae_db": 0.34554511512133135, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4909368582163811, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31050548855660515, "3000–6000 Hz_40ms_envelope_correlation": 0.9355922298423361, "3000–6000 Hz_40ms_modulation_mae_db": 0.6421451378709405, "mr_stft_db": 5.568870962289631, "band_mae_db": 0.1893121990011131, "rms_gain_error_db": -0.13619397622798618, "output_peak": 0.5866387438653509}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0365865523096571, "全帯域_10ms_rms_envelope_nrmse": 0.06757695980783372, "全帯域_10ms_envelope_correlation": 0.9945826122920819, "全帯域_10ms_modulation_mae_db": 0.4664048212413335, "全帯域_40ms_log_rms_mae_db": 0.5314613119575337, "全帯域_40ms_rms_envelope_nrmse": 0.039599738113985435, "全帯域_40ms_envelope_correlation": 0.997944232465546, "全帯域_40ms_modulation_mae_db": 0.48819898312967086, "300–1000 Hz_10ms_log_rms_mae_db": 1.7156338621738845, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10661157897762238, "300–1000 Hz_10ms_envelope_correlation": 0.9912424305159426, "300–1000 Hz_10ms_modulation_mae_db": 0.37974650163213813, "300–1000 Hz_40ms_log_rms_mae_db": 0.9924061855324173, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06981768102665298, "300–1000 Hz_40ms_envelope_correlation": 0.995939054603655, "300–1000 Hz_40ms_modulation_mae_db": 0.5124514267757864, "1000–3000 Hz_10ms_log_rms_mae_db": 1.951383836777351, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1490269111930027, "1000–3000 Hz_10ms_envelope_correlation": 0.9847157922393559, "1000–3000 Hz_10ms_modulation_mae_db": 1.3145289740859731, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1118939077963879, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09241062424042878, "1000–3000 Hz_40ms_envelope_correlation": 0.9950179670588076, "1000–3000 Hz_40ms_modulation_mae_db": 1.0407209565706343, "3000–6000 Hz_10ms_log_rms_mae_db": 2.081092775226164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41303786386839864, "3000–6000 Hz_10ms_envelope_correlation": 0.9237687365783531, "3000–6000 Hz_10ms_modulation_mae_db": 0.6147662436867385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4658751906030907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3610285544317869, "3000–6000 Hz_40ms_envelope_correlation": 0.9425381491296471, "3000–6000 Hz_40ms_modulation_mae_db": 1.4031949216822792, "mr_stft_db": 5.578035267769825, "band_mae_db": 0.4153674439702013, "rms_gain_error_db": -0.041024739717563555, "output_peak": 0.7389234387416272}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7983652609969, "spectral_convergence": 0.19013964013166793, "spectral_convergence_status": "ok", "amplitude_floor": 1.884454455950008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.433065517291865, "spectral_convergence": 0.1957263337057834, "spectral_convergence_status": "ok", "amplitude_floor": 1.301578596614512e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.454513966156183, "spectral_convergence": 0.19727525823007624, "spectral_convergence_status": "ok", "amplitude_floor": 1.1393421242566376e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004137534659655858, "output_power": 0.004060557720742394, "delta_db": -0.08155956477375674, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026137240204796635, "output_power": 0.002588872105149977, "delta_db": -0.04149133057883779, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008071793462917449, "output_power": 0.0007758676855507596, "delta_db": -0.1718237673154649, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004252507832785635, "output_power": 0.00040520150468888893, "delta_db": -0.20974072806724137, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.026398390024539e-05, "output_power": 6.521333647844716e-05, "delta_db": -0.32396349858380524, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.066875275629503e-05, "output_power": 4.429094112623611e-05, "delta_db": -0.5842530539738288, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23547199054882248, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08912057245865808, "pitch": null, "output_rms_dbfs": -21.001793174152997, "output_peak": 0.65567952639057, "reference_rms_dbfs": -20.91267260169434, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.561981581481649}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025953969251226393, "log_rms_mae_db": 0.9762303941447787, "rms_envelope_nrmse": 0.0613460963229519, "envelope_correlation": 0.9956247246358841, "rms_gain_error_db": -0.08925904890323533, "reference_mean_power": 0.008109529837298932, "output_mean_power": 0.007944558569870603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5635786081597607e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013500334014131179, "output_power": 0.001337666722314971, "reference_active": true, "delta_db": -0.03996590502590186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020457869490256438, "output_power": 0.00018878486714056912, "reference_active": true, "delta_db": -0.3489322510260209}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3880284077345985e-05, "output_power": 3.4447859461462825e-05, "reference_active": true, "delta_db": 0.07215197551692448}], "active_band_count": 3, "active_band_mae_db": 0.15368337718961575}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024049824298809385, "log_rms_mae_db": 0.5119598828770303, "rms_envelope_nrmse": 0.03300748385786052, "envelope_correlation": 0.9986940582361267, "rms_gain_error_db": -0.08972839835186858, "reference_mean_power": 0.008139876536629037, "output_mean_power": 0.007973426181335185}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2818780696338306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012006574574251973, "output_power": 0.0011937284660743235, "reference_active": true, "delta_db": -0.025135723355663678}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.159020687440259e-05, "output_power": 8.561847701544779e-05, "reference_active": true, "delta_db": -0.2928154163737712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2353163446294293e-06, "output_power": 2.2991684031593364e-06, "reference_active": true, "delta_db": 0.122317887518365}], "active_band_count": 3, "active_band_mae_db": 0.14675634241593327}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002074995834794702, "log_rms_mae_db": 1.7234242105125452, "rms_envelope_nrmse": 0.11200119993898285, "envelope_correlation": 0.9905439871362841, "rms_gain_error_db": -0.19619198159894438, "reference_mean_power": 0.0020359851811281022, "output_mean_power": 0.0019460563633088915}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.345568889858965e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045532158312677247, "output_power": 0.0004368495862343171, "reference_active": true, "delta_db": -0.17986308495105988}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.561332252218219e-05, "output_power": 7.30547120756467e-05, "reference_active": true, "delta_db": -0.6889311823702488}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0669601144719856e-05, "output_power": 1.0483881160994013e-05, "reference_active": true, "delta_db": -0.07626095358605768}], "active_band_count": 3, "active_band_mae_db": 0.3150184069691221}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001833599147850085, "log_rms_mae_db": 0.9829146709899571, "rms_envelope_nrmse": 0.06006396813832496, "envelope_correlation": 0.9973837341090152, "rms_gain_error_db": -0.19602214053032546, "reference_mean_power": 0.0020457391187242045, "output_mean_power": 0.0019554559437912258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.75393473065512e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00043556686984963635, "output_power": 0.0004165315702802025, "reference_active": true, "delta_db": -0.1940691539091269}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.69121464262054e-05, "output_power": 3.28129087610213e-05, "reference_active": true, "delta_db": -0.511245690456837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.183033529622033e-07, "output_power": 6.874662472741451e-07, "reference_active": true, "delta_db": -0.1905651333156751}], "active_band_count": 3, "active_band_mae_db": 0.29862665922721304}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011093565856595723, "log_rms_mae_db": 1.8926501765479742, "rms_envelope_nrmse": 0.16602172055287473, "envelope_correlation": 0.979559973958332, "rms_gain_error_db": -0.22614114181847345, "reference_mean_power": 0.00042557205365836593, "output_mean_power": 0.000403979183348457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.99964400169894e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011758808293094384, "output_power": 0.00011046217288959526, "reference_active": true, "delta_db": -0.2714972807697165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1475307548341578e-05, "output_power": 1.783226372638442e-05, "reference_active": true, "delta_db": -0.8073291366501457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0561557218273926e-06, "output_power": 3.4711246108863008e-06, "reference_active": true, "delta_db": -1.6335023627107492}], "active_band_count": 3, "active_band_mae_db": 0.9041095933768705}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010031642073947198, "log_rms_mae_db": 1.0695683457301588, "rms_envelope_nrmse": 0.11735799836923402, "envelope_correlation": 0.989258511197568, "rms_gain_error_db": -0.22619810470715287, "reference_mean_power": 0.0004276640413209328, "output_mean_power": 0.00040595970210078187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8247169971067077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011139646210262638, "output_power": 0.00010471563846274058, "reference_active": true, "delta_db": -0.26859853064272815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0378224135538677e-05, "output_power": 8.905913307608108e-06, "reference_active": true, "delta_db": -0.6644458250945474}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.220864105203889e-07, "output_power": 2.0560721998051925e-07, "reference_active": true, "delta_db": -1.949340404232395}], "active_band_count": 3, "active_band_mae_db": 0.9607949199898901}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.985234496629778e-05, "log_rms_mae_db": 2.129081959715819, "rms_envelope_nrmse": 0.37908221361620725, "envelope_correlation": 0.905911010337964, "rms_gain_error_db": -0.56957104609014, "reference_mean_power": 6.361685525644655e-05, "output_mean_power": 5.579754516591686e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.697638884988847e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.1912782580556425e-06, "output_power": 5.019891557569712e-06, "reference_active": true, "delta_db": -0.9108598780900061}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.016318108845029e-06, "output_power": 1.7214197107655298e-06, "reference_active": true, "delta_db": -0.686722791410558}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.069070181696055e-07, "output_power": 6.622579404417814e-07, "reference_active": true, "delta_db": 0.3790501454396522}], "active_band_count": 3, "active_band_mae_db": 0.6588776049800721}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.357134521214995e-05, "log_rms_mae_db": 1.5189130716858257, "rms_envelope_nrmse": 0.33613618551507474, "envelope_correlation": 0.9228333300848915, "rms_gain_error_db": -0.5696829652631659, "reference_mean_power": 6.39232761364752e-05, "output_mean_power": 5.6064858244117786e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.609918878818377e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.077093160223491e-06, "output_power": 5.0643583998852965e-06, "reference_active": true, "delta_db": -0.791714613080797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.511523340291838e-07, "output_power": 8.934846942374393e-07, "reference_active": true, "delta_db": -0.27162960696381844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7938163224025654e-08, "output_power": 4.47327014607789e-08, "reference_active": true, "delta_db": 2.0442727620775454}], "active_band_count": 3, "active_band_mae_db": 1.0358723273740535}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.896668636391765, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "98c1e9037a96fb8d49e368644c8b19fe867d9c676820c3544db05eb6ef732f8c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826284587932846, "standard_deviation": 0.0022582608507934147, "minimum": 0.979559973958332, "maximum": 0.9856570066975723, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9340813757296833, "standard_deviation": 0.029046599008109537, "minimum": 1.8899333256576953, "maximum": 1.9644533307899268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9796485602573398, "standard_deviation": 0.25025434061355945, "minimum": 0.648183147473739, "maximum": 1.38339401496782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1578813585049852, "standard_deviation": 0.008530803850161393, "minimum": 0.14705419316679752, "maximum": 0.16850678078714662, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9933601769092204, "standard_deviation": 0.002382951123160318, "minimum": 0.989258511197568, "maximum": 0.9958995228124989, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1345899128631216, "standard_deviation": 0.03676905620787608, "minimum": 1.0695683457301588, "maximum": 1.182372404653839, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9606014073901441, "standard_deviation": 0.2839052163974506, "minimum": 0.5730523755186426, "maximum": 1.5215170959013136, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10245315001370858, "standard_deviation": 0.01194654522630331, "minimum": 0.08994082294901565, "maximum": 0.12016828292546221, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.923178448533819, "standard_deviation": 0.007901621393628332, "minimum": 0.905911010337964, "maximum": 0.931611941979882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0836337107341762, "standard_deviation": 0.023458954524286955, "minimum": 2.0545706636730485, "maximum": 2.129081959715819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5529696410199039, "standard_deviation": 0.17175847386872325, "minimum": 0.34223054484236043, "maximum": 0.7832030762244603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.36865684214628575, "standard_deviation": 0.023616179118841986, "minimum": 0.34427368103403116, "maximum": 0.41303786386839864, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9393633203342951, "standard_deviation": 0.007812341878222764, "minimum": 0.9228333300848915, "maximum": 0.9475407192363379, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4933217842857114, "standard_deviation": 0.027395436677705484, "minimum": 1.459730855301974, "maximum": 1.529223679849306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.032703231102254, "standard_deviation": 0.2709954153958204, "minimum": 0.6421451378709405, "maximum": 1.4404252112469305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3221397714930889, "standard_deviation": 0.018908143455443192, "minimum": 0.30763058625182604, "maximum": 0.3610285544317869, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9889077252110982, "standard_deviation": 0.0022370528177719806, "minimum": 0.9846093917932288, "maximum": 0.9912424305159426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7332762129578922, "standard_deviation": 0.03429991449525275, "minimum": 1.6888224022811373, "maximum": 1.7805655586920321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4830355562986556, "standard_deviation": 0.14908108982890508, "minimum": 0.3150184069691221, "maximum": 0.7171839907138233, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12047891275491139, "standard_deviation": 0.011787193107020437, "minimum": 0.10661157897762238, "maximum": 0.1424293008707456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962215160412848, "standard_deviation": 0.001065327292973684, "minimum": 0.9938137368605228, "maximum": 0.9973837341090152, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0070450694122959, "standard_deviation": 0.03564816155707098, "minimum": 0.9592396211680708, "maximum": 1.0572362875143515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5171308564552681, "standard_deviation": 0.1450641136128721, "minimum": 0.29862665922721304, "maximum": 0.7215632064887227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07007091206252161, "standard_deviation": 0.008782734221962038, "minimum": 0.06006396813832496, "maximum": 0.09001176381502, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.29872116082413974, "standard_deviation": 0.08850672009364978, "minimum": 0.1893121990011131, "maximum": 0.4153674439702013, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.57019485775565, "standard_deviation": 0.008185243992939085, "minimum": 5.557873679433842, "maximum": 5.582884404087775, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6431293935619287, "standard_deviation": 0.07534273537743498, "minimum": 0.579974323856633, "maximum": 0.7771159982769201, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07987832116476659, "standard_deviation": 0.028948577673891827, "minimum": -0.13619397622798618, "maximum": -0.041024739717563555, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949202502438479, "standard_deviation": 0.0006760594711114694, "minimum": 0.9939910172828467, "maximum": 0.9956369376084263, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9869206116312408, "standard_deviation": 0.029405651815553543, "minimum": 0.9503497061515834, "maximum": 1.0365865523096571, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26065907558773116, "standard_deviation": 0.1153852799630685, "minimum": 0.15368337718961575, "maximum": 0.4664048212413335, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06571770285182255, "standard_deviation": 0.004329348992774774, "minimum": 0.06092039171778375, "maximum": 0.0720266868221312, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983747387088188, "standard_deviation": 0.0003743355464364331, "minimum": 0.9978752331979398, "maximum": 0.9987705662960985, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5188190636183856, "standard_deviation": 0.03596803005929823, "minimum": 0.44894419009367037, "maximum": 0.571277118043975, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31638732063668007, "standard_deviation": 0.1262355828872248, "minimum": 0.14675634241593327, "maximum": 0.48819898312967086, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03596184052275858, "standard_deviation": 0.003992836281642088, "minimum": 0.03133921224377833, "maximum": 0.0425769439019841, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0019", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.723212718963623, "predicted_mos": 2.723212718963623, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "432ead2efbafde678395e13c958fa4a144de813321eea51c015209c38c045e49", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.2511606998741627, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9936545767912022, "全帯域_10ms_rms_envelope_nrmse": 0.06885018294296937, "全帯域_10ms_envelope_correlation": 0.9943646933476089, "全帯域_10ms_modulation_mae_db": 0.30703725407221844, "全帯域_40ms_log_rms_mae_db": 0.5258965394414953, "全帯域_40ms_rms_envelope_nrmse": 0.03952342069151852, "全帯域_40ms_envelope_correlation": 0.9979251355828739, "全帯域_40ms_modulation_mae_db": 0.12672314124340636, "300–1000 Hz_10ms_log_rms_mae_db": 2.0283220015573313, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11414956867359778, "300–1000 Hz_10ms_envelope_correlation": 0.9900144771847736, "300–1000 Hz_10ms_modulation_mae_db": 0.43672126569080216, "300–1000 Hz_40ms_log_rms_mae_db": 1.2897594445926575, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06825484692424137, "300–1000 Hz_40ms_envelope_correlation": 0.9962467608224441, "300–1000 Hz_40ms_modulation_mae_db": 0.37488035375039647, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4437189163780917, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1781112790821271, "1000–3000 Hz_10ms_envelope_correlation": 0.9764278048825069, "1000–3000 Hz_10ms_modulation_mae_db": 0.7341015448467693, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5158082720978545, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10785091464416133, "1000–3000 Hz_40ms_envelope_correlation": 0.9906932254501193, "1000–3000 Hz_40ms_modulation_mae_db": 0.8850381707573692, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5968868477253504, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40539110470392503, "3000–6000 Hz_10ms_envelope_correlation": 0.9184664514754932, "3000–6000 Hz_10ms_modulation_mae_db": 0.7643731311130099, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8165710036413831, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3364239644119745, "3000–6000 Hz_40ms_envelope_correlation": 0.944039384841891, "3000–6000 Hz_40ms_modulation_mae_db": 1.3706845101023195, "mr_stft_db": 6.0326576664321765, "band_mae_db": 0.2401068917526988, "rms_gain_error_db": -0.01724211515769057, "output_peak": 0.6212153593544489}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9179882979446649, "全帯域_10ms_rms_envelope_nrmse": 0.06259899412222557, "全帯域_10ms_envelope_correlation": 0.9953914355895249, "全帯域_10ms_modulation_mae_db": 0.16929866270121033, "全帯域_40ms_log_rms_mae_db": 0.48931700026030844, "全帯域_40ms_rms_envelope_nrmse": 0.03392587932366694, "全帯域_40ms_envelope_correlation": 0.998551386575332, "全帯域_40ms_modulation_mae_db": 0.1986597408652092, "300–1000 Hz_10ms_log_rms_mae_db": 1.9915293012039637, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10692769720015617, "300–1000 Hz_10ms_envelope_correlation": 0.9912917362872943, "300–1000 Hz_10ms_modulation_mae_db": 0.5275556857365956, "300–1000 Hz_40ms_log_rms_mae_db": 1.2302133817362522, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05475257641736446, "300–1000 Hz_40ms_envelope_correlation": 0.9976833774000412, "300–1000 Hz_40ms_modulation_mae_db": 0.552275813796001, "1000–3000 Hz_10ms_log_rms_mae_db": 2.42440619764465, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1707876396780451, "1000–3000 Hz_10ms_envelope_correlation": 0.9795046987205959, "1000–3000 Hz_10ms_modulation_mae_db": 1.1925150801375295, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4613805122224903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10568931791278312, "1000–3000 Hz_40ms_envelope_correlation": 0.9930925845325059, "1000–3000 Hz_40ms_modulation_mae_db": 1.535418678216228, "3000–6000 Hz_10ms_log_rms_mae_db": 2.591266757926595, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37676367925425036, "3000–6000 Hz_10ms_envelope_correlation": 0.9106272693435021, "3000–6000 Hz_10ms_modulation_mae_db": 0.6328958924186104, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8022100367018814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33418958405537535, "3000–6000 Hz_40ms_envelope_correlation": 0.9261596221899652, "3000–6000 Hz_40ms_modulation_mae_db": 0.8743342129489324, "mr_stft_db": 6.029134670076108, "band_mae_db": 0.11759418611900359, "rms_gain_error_db": -0.06817898648090072, "output_peak": 0.6733629008036053}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.934439660556085, "全帯域_10ms_rms_envelope_nrmse": 0.06887529774854004, "全帯域_10ms_envelope_correlation": 0.9944292186624397, "全帯域_10ms_modulation_mae_db": 0.1112098710530424, "全帯域_40ms_log_rms_mae_db": 0.5173057424602985, "全帯域_40ms_rms_envelope_nrmse": 0.04029071687002977, "全帯域_40ms_envelope_correlation": 0.9979523305388321, "全帯域_40ms_modulation_mae_db": 0.24974672717683652, "300–1000 Hz_10ms_log_rms_mae_db": 2.0137181404344213, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12007003722513526, "300–1000 Hz_10ms_envelope_correlation": 0.9890196346494937, "300–1000 Hz_10ms_modulation_mae_db": 0.4923290688956521, "300–1000 Hz_40ms_log_rms_mae_db": 1.2718015085180252, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06775637125687903, "300–1000 Hz_40ms_envelope_correlation": 0.9964766809656316, "300–1000 Hz_40ms_modulation_mae_db": 0.5161298956492302, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3951096685547855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1843435899538586, "1000–3000 Hz_10ms_envelope_correlation": 0.9747973145191025, "1000–3000 Hz_10ms_modulation_mae_db": 0.6787275359082265, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4577283471583984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12060524619320817, "1000–3000 Hz_40ms_envelope_correlation": 0.9889148828594293, "1000–3000 Hz_40ms_modulation_mae_db": 0.9073553890837642, "3000–6000 Hz_10ms_log_rms_mae_db": 2.620451684402739, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3694006970526042, "3000–6000 Hz_10ms_envelope_correlation": 0.914911823750518, "3000–6000 Hz_10ms_modulation_mae_db": 0.2380339795566977, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8275070132778335, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32296096293581394, "3000–6000 Hz_40ms_envelope_correlation": 0.9317298753283397, "3000–6000 Hz_40ms_modulation_mae_db": 0.7267839251771063, "mr_stft_db": 6.041184816276053, "band_mae_db": 0.13077614780125138, "rms_gain_error_db": -0.0868909034331371, "output_peak": 0.692840576820382}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9994376822530753, "全帯域_10ms_rms_envelope_nrmse": 0.07317609607627143, "全帯域_10ms_envelope_correlation": 0.9936814487278497, "全帯域_10ms_modulation_mae_db": 0.21140026346406385, "全帯域_40ms_log_rms_mae_db": 0.5114712089425362, "全帯域_40ms_rms_envelope_nrmse": 0.041204919580582915, "全帯域_40ms_envelope_correlation": 0.9978437730687116, "全帯域_40ms_modulation_mae_db": 0.2155505957674879, "300–1000 Hz_10ms_log_rms_mae_db": 2.053794799833403, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1237930985657778, "300–1000 Hz_10ms_envelope_correlation": 0.9883345386584542, "300–1000 Hz_10ms_modulation_mae_db": 0.6230460231453331, "300–1000 Hz_40ms_log_rms_mae_db": 1.3405490617707838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06869950730262468, "300–1000 Hz_40ms_envelope_correlation": 0.9964012107363507, "300–1000 Hz_40ms_modulation_mae_db": 0.6617152774097019, "1000–3000 Hz_10ms_log_rms_mae_db": 2.408724193984034, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1810459819318834, "1000–3000 Hz_10ms_envelope_correlation": 0.976038363974015, "1000–3000 Hz_10ms_modulation_mae_db": 0.6850759732009681, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4762128289186665, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10902083732514262, "1000–3000 Hz_40ms_envelope_correlation": 0.991644277113728, "1000–3000 Hz_40ms_modulation_mae_db": 0.7169956976185216, "3000–6000 Hz_10ms_log_rms_mae_db": 2.548253424286657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37180833361940424, "3000–6000 Hz_10ms_envelope_correlation": 0.9183475461056795, "3000–6000 Hz_10ms_modulation_mae_db": 0.4332621067099492, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7877142653467695, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31517123449928036, "3000–6000 Hz_40ms_envelope_correlation": 0.9387847940721394, "3000–6000 Hz_40ms_modulation_mae_db": 0.7948606516296763, "mr_stft_db": 6.0209244909187545, "band_mae_db": 0.2184051821303764, "rms_gain_error_db": -0.0735013770860982, "output_peak": 0.5996900243402832}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9744684680541841, "全帯域_10ms_rms_envelope_nrmse": 0.062956308445518, "全帯域_10ms_envelope_correlation": 0.9953721721523912, "全帯域_10ms_modulation_mae_db": 0.15639805665918063, "全帯域_40ms_log_rms_mae_db": 0.5158790885017766, "全帯域_40ms_rms_envelope_nrmse": 0.033222473893093005, "全帯域_40ms_envelope_correlation": 0.9986570018314742, "全帯域_40ms_modulation_mae_db": 0.31541950016459563, "300–1000 Hz_10ms_log_rms_mae_db": 1.9542143973621644, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11424697377830977, "300–1000 Hz_10ms_envelope_correlation": 0.9902218122216476, "300–1000 Hz_10ms_modulation_mae_db": 0.5340393079444571, "300–1000 Hz_40ms_log_rms_mae_db": 1.26672163215958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06766551007997385, "300–1000 Hz_40ms_envelope_correlation": 0.9966174760012082, "300–1000 Hz_40ms_modulation_mae_db": 0.6081650473352054, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4691597224941435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1817148525378689, "1000–3000 Hz_10ms_envelope_correlation": 0.9753528795165171, "1000–3000 Hz_10ms_modulation_mae_db": 0.8705018449556845, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4554656155119672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10531637268939636, "1000–3000 Hz_40ms_envelope_correlation": 0.991377169351602, "1000–3000 Hz_40ms_modulation_mae_db": 0.9764343034576667, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5798422146743962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3614093316711386, "3000–6000 Hz_10ms_envelope_correlation": 0.9250761346886434, "3000–6000 Hz_10ms_modulation_mae_db": 0.6121152096865324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8254434960047472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3122432648281571, "3000–6000 Hz_40ms_envelope_correlation": 0.9418845124161536, "3000–6000 Hz_40ms_modulation_mae_db": 0.8610180096644929, "mr_stft_db": 6.047712261198384, "band_mae_db": 0.19107995695887162, "rms_gain_error_db": -0.08944760432836797, "output_peak": 0.6199524177113452}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9356184276893067, "全帯域_10ms_rms_envelope_nrmse": 0.065009453369722, "全帯域_10ms_envelope_correlation": 0.994977479437806, "全帯域_10ms_modulation_mae_db": 0.30551364059381625, "全帯域_40ms_log_rms_mae_db": 0.5206968448159345, "全帯域_40ms_rms_envelope_nrmse": 0.03670547378472966, "全帯域_40ms_envelope_correlation": 0.9982166553131463, "全帯域_40ms_modulation_mae_db": 0.3172013826363981, "300–1000 Hz_10ms_log_rms_mae_db": 1.9974050542127706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1152335560827578, "300–1000 Hz_10ms_envelope_correlation": 0.9900227737882239, "300–1000 Hz_10ms_modulation_mae_db": 0.6624065726276377, "300–1000 Hz_40ms_log_rms_mae_db": 1.2841491593478038, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.068986495442261, "300–1000 Hz_40ms_envelope_correlation": 0.9964780821171193, "300–1000 Hz_40ms_modulation_mae_db": 0.654826828513173, "1000–3000 Hz_10ms_log_rms_mae_db": 2.472140435416864, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18745768851373687, "1000–3000 Hz_10ms_envelope_correlation": 0.9738346649110213, "1000–3000 Hz_10ms_modulation_mae_db": 0.5735819337494633, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5022412733589812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12124856523027086, "1000–3000 Hz_40ms_envelope_correlation": 0.9886566574960197, "1000–3000 Hz_40ms_modulation_mae_db": 0.5764109393610076, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4867584194483854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3729026153743765, "3000–6000 Hz_10ms_envelope_correlation": 0.913079438104246, "3000–6000 Hz_10ms_modulation_mae_db": 0.6574193618714591, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7492431860245639, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3163778721572722, "3000–6000 Hz_40ms_envelope_correlation": 0.9343397801059902, "3000–6000 Hz_40ms_modulation_mae_db": 0.8086458801907163, "mr_stft_db": 6.032051927764152, "band_mae_db": 0.11808795631333764, "rms_gain_error_db": -0.02443764201972163, "output_peak": 0.6553238241700237}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9716289185421795, "全帯域_10ms_rms_envelope_nrmse": 0.0631677554810506, "全帯域_10ms_envelope_correlation": 0.9952880751028768, "全帯域_10ms_modulation_mae_db": 0.2817183030277965, "全帯域_40ms_log_rms_mae_db": 0.5539879147675282, "全帯域_40ms_rms_envelope_nrmse": 0.032806219244742425, "全帯域_40ms_envelope_correlation": 0.9986275438433202, "全帯域_40ms_modulation_mae_db": 0.33622214296329367, "300–1000 Hz_10ms_log_rms_mae_db": 2.076331007692316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11138776648472153, "300–1000 Hz_10ms_envelope_correlation": 0.990744768261544, "300–1000 Hz_10ms_modulation_mae_db": 0.4169438937469081, "300–1000 Hz_40ms_log_rms_mae_db": 1.3914378780340564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06388423541432502, "300–1000 Hz_40ms_envelope_correlation": 0.9970806393389314, "300–1000 Hz_40ms_modulation_mae_db": 0.3487683150397927, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4510432076441657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1798480298392046, "1000–3000 Hz_10ms_envelope_correlation": 0.9758249058734129, "1000–3000 Hz_10ms_modulation_mae_db": 0.8585321782396683, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4942117996195017, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11752347297279943, "1000–3000 Hz_40ms_envelope_correlation": 0.989024327444231, "1000–3000 Hz_40ms_modulation_mae_db": 0.8977852591696971, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5974977126676646, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36880358820008924, "3000–6000 Hz_10ms_envelope_correlation": 0.9212808996905812, "3000–6000 Hz_10ms_modulation_mae_db": 0.47689629417751406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8228960467447994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3229037496340364, "3000–6000 Hz_40ms_envelope_correlation": 0.937104750298304, "3000–6000 Hz_40ms_modulation_mae_db": 0.7678160648182386, "mr_stft_db": 6.039542124412798, "band_mae_db": 0.1564575932382557, "rms_gain_error_db": -0.056995120318616144, "output_peak": 0.7031369409299161}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9184565882195924, "全帯域_10ms_rms_envelope_nrmse": 0.0634769297158663, "全帯域_10ms_envelope_correlation": 0.9952643751189175, "全帯域_10ms_modulation_mae_db": 0.27454024227194057, "全帯域_40ms_log_rms_mae_db": 0.48911181710217233, "全帯域_40ms_rms_envelope_nrmse": 0.03435829431813131, "全帯域_40ms_envelope_correlation": 0.9985199699092908, "全帯域_40ms_modulation_mae_db": 0.40320550463891025, "300–1000 Hz_10ms_log_rms_mae_db": 1.850908038443618, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1012988069867173, "300–1000 Hz_10ms_envelope_correlation": 0.9922201658324026, "300–1000 Hz_10ms_modulation_mae_db": 0.2931769172078176, "300–1000 Hz_40ms_log_rms_mae_db": 1.2301818044757533, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05440537546156855, "300–1000 Hz_40ms_envelope_correlation": 0.9977669133310603, "300–1000 Hz_40ms_modulation_mae_db": 0.44133429832758697, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4198603415242017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18669570828178778, "1000–3000 Hz_10ms_envelope_correlation": 0.9750294242078218, "1000–3000 Hz_10ms_modulation_mae_db": 0.988972088510531, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5550421279378936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12029745800706018, "1000–3000 Hz_40ms_envelope_correlation": 0.9903449140580973, "1000–3000 Hz_40ms_modulation_mae_db": 1.197681416777287, "3000–6000 Hz_10ms_log_rms_mae_db": 2.608748322708768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3566919541118291, "3000–6000 Hz_10ms_envelope_correlation": 0.92169893288944, "3000–6000 Hz_10ms_modulation_mae_db": 0.6518147118720524, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8410254176816876, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31338982747153254, "3000–6000 Hz_40ms_envelope_correlation": 0.9365638580238357, "3000–6000 Hz_40ms_modulation_mae_db": 0.9891443780863219, "mr_stft_db": 6.0395033074641375, "band_mae_db": 0.1507154939664072, "rms_gain_error_db": -0.0757613920035874, "output_peak": 0.6405581668915209}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.354319728202127, "spectral_convergence": 0.19630373879298638, "spectral_convergence_status": "ok", "amplitude_floor": 1.884454455950008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.974353656699348, "spectral_convergence": 0.19883756651926934, "spectral_convergence_status": "ok", "amplitude_floor": 1.301578596614512e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.769299614395057, "spectral_convergence": 0.19660231343329115, "spectral_convergence_status": "ok", "amplitude_floor": 1.1393421242566376e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004137534659655858, "output_power": 0.004102169812729717, "delta_db": -0.037280104487545124, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026137240204796635, "output_power": 0.0026229294951347925, "delta_db": 0.015268877164325107, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008071793462917449, "output_power": 0.0007842319106024729, "delta_db": -0.12525530977052074, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004252507832785635, "output_power": 0.00042643656839098476, "delta_db": 0.012093176276257783, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.026398390024539e-05, "output_power": 6.97020538502331e-05, "delta_db": -0.03487194966114124, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.066875275629503e-05, "output_power": 6.703898145600856e-05, "delta_db": 1.2158719331564027, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2401068917526988, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.01724211515769057, "pitch": null, "output_rms_dbfs": -20.929914716852025, "output_peak": 0.6212153593544489, "reference_rms_dbfs": -20.91267260169434, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0326576664321765}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025953969251226393, "log_rms_mae_db": 0.9936545767912022, "rms_envelope_nrmse": 0.06885018294296937, "envelope_correlation": 0.9943646933476089, "rms_gain_error_db": -0.01707448253025717, "reference_mean_power": 0.008109529837298932, "output_mean_power": 0.00807770944929991}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5635786081597607e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013500334014131179, "output_power": 0.0013165897803008198, "reference_active": true, "delta_db": -0.1089403391711237}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020457869490256438, "output_power": 0.00019287306118136968, "reference_active": true, "delta_db": -0.25588830292927167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3880284077345985e-05, "output_power": 3.8510162848265996e-05, "reference_active": true, "delta_db": 0.5562831201162599}], "active_band_count": 3, "active_band_mae_db": 0.30703725407221844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024049824298809385, "log_rms_mae_db": 0.5258965394414953, "rms_envelope_nrmse": 0.03952342069151852, "envelope_correlation": 0.9979251355828739, "rms_gain_error_db": -0.016528127205928375, "reference_mean_power": 0.008139876536629037, "output_mean_power": 0.00810895714011348}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2818780696338306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012006574574251973, "output_power": 0.0011800607867439058, "reference_active": true, "delta_db": -0.07514743607805273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.159020687440259e-05, "output_power": 8.641988650237025e-05, "reference_active": true, "delta_db": -0.25235348353874687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2353163446294293e-06, "output_power": 2.2625898997867687e-06, "reference_active": true, "delta_db": 0.05266850411341946}], "active_band_count": 3, "active_band_mae_db": 0.12672314124340636}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002074995834794702, "log_rms_mae_db": 2.0283220015573313, "rms_envelope_nrmse": 0.11414956867359778, "envelope_correlation": 0.9900144771847736, "rms_gain_error_db": -0.09711755027342439, "reference_mean_power": 0.0020359851811281022, "output_mean_power": 0.001990961480102702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.345568889858965e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045532158312677247, "output_power": 0.00042814674813949926, "reference_active": true, "delta_db": -0.2672558738093126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.561332252218219e-05, "output_power": 7.012265693020261e-05, "reference_active": true, "delta_db": -0.8668298863811561}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0669601144719856e-05, "output_power": 1.0245670409967154e-05, "reference_active": true, "delta_db": -0.17607803688193802}], "active_band_count": 3, "active_band_mae_db": 0.43672126569080216}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001833599147850085, "log_rms_mae_db": 1.2897594445926575, "rms_envelope_nrmse": 0.06825484692424137, "envelope_correlation": 0.9962467608224441, "rms_gain_error_db": -0.09691429761700261, "reference_mean_power": 0.0020457391187242045, "output_mean_power": 0.0020005933463688444}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.75393473065512e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00043556686984963635, "output_power": 0.0004067234659818373, "reference_active": true, "delta_db": -0.2975560914627463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.69121464262054e-05, "output_power": 3.086018856510229e-05, "reference_active": true, "delta_db": -0.7777072457018851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.183033529622033e-07, "output_power": 7.101827532392805e-07, "reference_active": true, "delta_db": -0.049377724086558133}], "active_band_count": 3, "active_band_mae_db": 0.37488035375039647}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011093565856595723, "log_rms_mae_db": 2.4437189163780917, "rms_envelope_nrmse": 0.1781112790821271, "envelope_correlation": 0.9764278048825069, "rms_gain_error_db": -0.02840416423170381, "reference_mean_power": 0.00042557205365836593, "output_mean_power": 0.0004227977667614293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.99964400169894e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011758808293094384, "output_power": 0.00010369837310780977, "reference_active": true, "delta_db": -0.545913670803423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1475307548341578e-05, "output_power": 1.6307679905069364e-05, "reference_active": true, "delta_db": -1.1954721376419157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0561557218273926e-06, "output_power": 4.547037736322502e-06, "reference_active": true, "delta_db": -0.460918826094969}], "active_band_count": 3, "active_band_mae_db": 0.7341015448467693}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010031642073947198, "log_rms_mae_db": 1.5158082720978545, "rms_envelope_nrmse": 0.10785091464416133, "envelope_correlation": 0.9906932254501193, "rms_gain_error_db": -0.02847208508236391, "reference_mean_power": 0.0004276640413209328, "output_mean_power": 0.0004248694721056056}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8247169971067077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011139646210262638, "output_power": 9.898376557177191e-05, "reference_active": true, "delta_db": -0.5130742671785298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0378224135538677e-05, "output_power": 7.950367492962532e-06, "reference_active": true, "delta_db": -1.1573584210888324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.220864105203889e-07, "output_power": 2.567463150114486e-07, "reference_active": true, "delta_db": -0.9846818240047456}], "active_band_count": 3, "active_band_mae_db": 0.8850381707573692}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.985234496629778e-05, "log_rms_mae_db": 2.5968868477253504, "rms_envelope_nrmse": 0.40539110470392503, "envelope_correlation": 0.9184664514754932, "rms_gain_error_db": 1.0444924786673697, "reference_mean_power": 6.361685525644655e-05, "output_mean_power": 8.091358522439846e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.697638884988847e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.1912782580556425e-06, "output_power": 6.762769504342275e-06, "reference_active": true, "delta_db": 0.3834426241737454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.016318108845029e-06, "output_power": 1.9617659326231787e-06, "reference_active": true, "delta_db": -0.11911862150138366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.069070181696055e-07, "output_power": 9.165960353090569e-07, "reference_active": true, "delta_db": 1.7905581476639008}], "active_band_count": 3, "active_band_mae_db": 0.7643731311130099}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.357134521214995e-05, "log_rms_mae_db": 1.8165710036413831, "rms_envelope_nrmse": 0.3364239644119745, "envelope_correlation": 0.944039384841891, "rms_gain_error_db": 1.0447395240438166, "reference_mean_power": 6.39232761364752e-05, "output_mean_power": 8.1307943617794e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.609918878818377e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.077093160223491e-06, "output_power": 6.808769101642204e-06, "reference_active": true, "delta_db": 0.49372712655771067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.511523340291838e-07, "output_power": 1.0466763041069139e-06, "reference_active": true, "delta_db": 0.41562314250617705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7938163224025654e-08, "output_power": 5.840744066213628e-08, "reference_active": true, "delta_db": 3.2027032612430713}], "active_band_count": 3, "active_band_mae_db": 1.3706845101023195}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8864855101546065, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3ac3998d3ae6734b1ad285d81105b99904bfb270276e959716a32ee446111e7d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9758512570756241, "standard_deviation": 0.0016825565126966405, "minimum": 0.9738346649110213, "maximum": 0.9795046987205959, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.4355203354551174, "standard_deviation": 0.028032203516598728, "minimum": 2.3951096685547855, "maximum": 2.472140435416864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.822751022443605, "standard_deviation": 0.19912540263577383, "minimum": 0.5735819337494633, "maximum": 1.1925150801375295, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18125059622731404, "standard_deviation": 0.005335825219684959, "minimum": 0.1707876396780451, "maximum": 0.18745768851373687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9904685047882165, "standard_deviation": 0.0015557203342030205, "minimum": 0.9886566574960197, "maximum": 0.9930925845325059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4897613471032192, "standard_deviation": 0.03444855825176836, "minimum": 1.4554656155119672, "maximum": 1.5550421279378936, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9616399818051926, "standard_deviation": 0.2943032314820053, "minimum": 0.5764109393610076, "maximum": 1.535418678216228, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11344402312185276, "standard_deviation": 0.007099769248555836, "minimum": 0.10531637268939636, "maximum": 0.12124856523027086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.917936062006013, "standard_deviation": 0.004824486196493522, "minimum": 0.9106272693435021, "maximum": 0.9250761346886434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.57871317298007, "standard_deviation": 0.042878058254482446, "minimum": 2.4867584194483854, "maximum": 2.620451684402739, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5583513359257282, "standard_deviation": 0.1666166470022865, "minimum": 0.2380339795566977, "maximum": 0.7643731311130099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.37289641299845216, "standard_deviation": 0.014616788109979484, "minimum": 0.3566919541118291, "maximum": 0.40539110470392503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9363258221595774, "standard_deviation": 0.005676099411990622, "minimum": 0.9261596221899652, "maximum": 0.944039384841891, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8090763081779582, "standard_deviation": 0.02917670799160157, "minimum": 1.7492431860245639, "maximum": 1.8410254176816876, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8991609540772254, "standard_deviation": 0.20651516149276974, "minimum": 0.7267839251771063, "maximum": 1.3706845101023195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.32170755749918034, "standard_deviation": 0.00929870999473242, "minimum": 0.3122432648281571, "maximum": 0.3364239644119745, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9902337383604793, "standard_deviation": 0.0012247528234619297, "minimum": 0.9883345386584542, "maximum": 0.9922201658324026, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9957778425924986, "standard_deviation": 0.06966371709824867, "minimum": 1.850908038443618, "maximum": 2.076331007692316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4982773418744004, "standard_deviation": 0.11780590228938724, "minimum": 0.2931769172078176, "maximum": 0.6624065726276377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11338843812464668, "standard_deviation": 0.0070662642206348455, "minimum": 0.1012988069867173, "maximum": 0.1237930985657778, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968438925890983, "standard_deviation": 0.0005958440328094468, "minimum": 0.9962467608224441, "maximum": 0.9977669133310603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.288101733829364, "standard_deviation": 0.05461035665027139, "minimum": 1.2301818044757533, "maximum": 1.3914378780340564, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.519761978727636, "standard_deviation": 0.12168995499233426, "minimum": 0.3487683150397927, "maximum": 0.6617152774097019, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06430061478740474, "standard_deviation": 0.006204406585920959, "minimum": 0.05440537546156855, "maximum": 0.068986495442261, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1654029260350253, "standard_deviation": 0.04639637754455298, "minimum": 0.11759418611900359, "maximum": 0.2401068917526988, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.03533890806782, "standard_deviation": 0.008337051958215911, "minimum": 6.0209244909187545, "maximum": 6.047712261198384, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6507600263776907, "standard_deviation": 0.03699430696067253, "minimum": 0.5996900243402832, "maximum": 0.7031369409299161, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06155689260351497, "standard_deviation": 0.02717783386635714, "minimum": -0.08944760432836797, "maximum": -0.01724211515769057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9948461122674268, "standard_deviation": 0.0006237735895522652, "minimum": 0.9936814487278497, "maximum": 0.9953914355895249, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9557115775062862, "standard_deviation": 0.033002240806347, "minimum": 0.9179882979446649, "maximum": 0.9994376822530753, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.22713953673040863, "standard_deviation": 0.07539339666741562, "minimum": 0.1112098710530424, "maximum": 0.30703725407221844, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06601387723777041, "standard_deviation": 0.003856076258604999, "minimum": 0.06259899412222557, "maximum": 0.07317609607627143, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982867245828726, "standard_deviation": 0.00034255220451776057, "minimum": 0.9978437730687116, "maximum": 0.9986570018314742, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5154582695365063, "standard_deviation": 0.020763711901250825, "minimum": 0.48911181710217233, "maximum": 0.5539879147675282, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2703410919320172, "standard_deviation": 0.08897164312537154, "minimum": 0.12672314124340636, "maximum": 0.40320550463891025, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03650467471331182, "standard_deviation": 0.003407724714579258, "minimum": 0.032806219244742425, "maximum": 0.041204919580582915, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0019", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.6556811332702637, "predicted_mos": 2.6556811332702637, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f8b0955e39281de2be755834d8ad07abf126a3137f7c1017bccc573a53bcba61", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.46343299979344, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0093851828958356, "全帯域_10ms_rms_envelope_nrmse": 0.0649740675091112, "全帯域_10ms_envelope_correlation": 0.9951241306016148, "全帯域_10ms_modulation_mae_db": 0.20702902367232454, "全帯域_40ms_log_rms_mae_db": 0.5502230817680893, "全帯域_40ms_rms_envelope_nrmse": 0.041216663523816165, "全帯域_40ms_envelope_correlation": 0.9979395279125262, "全帯域_40ms_modulation_mae_db": 0.25491523283884954, "300–1000 Hz_10ms_log_rms_mae_db": 1.7066341505804121, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10693602196458853, "300–1000 Hz_10ms_envelope_correlation": 0.9913883430182441, "300–1000 Hz_10ms_modulation_mae_db": 0.4752116416833087, "300–1000 Hz_40ms_log_rms_mae_db": 1.050376159161232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0628547336668817, "300–1000 Hz_40ms_envelope_correlation": 0.9970491789588763, "300–1000 Hz_40ms_modulation_mae_db": 0.3496608551154759, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7920884291789556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14887612253283114, "1000–3000 Hz_10ms_envelope_correlation": 0.9840784434429267, "1000–3000 Hz_10ms_modulation_mae_db": 0.7265414424067731, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0395162003338856, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08957164908820577, "1000–3000 Hz_40ms_envelope_correlation": 0.9945627683545739, "1000–3000 Hz_40ms_modulation_mae_db": 0.7346269028348377, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8819243914062795, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3494240192495265, "3000–6000 Hz_10ms_envelope_correlation": 0.9238838994547404, "3000–6000 Hz_10ms_modulation_mae_db": 0.6240438780697323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3010582901834684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3098004180077721, "3000–6000 Hz_40ms_envelope_correlation": 0.9370206493111047, "3000–6000 Hz_40ms_modulation_mae_db": 1.2680279475118177, "mr_stft_db": 5.522999075287458, "band_mae_db": 0.18995116343861415, "rms_gain_error_db": -0.1118266154525426, "output_peak": 0.609067617981912}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0226302531869578, "全帯域_10ms_rms_envelope_nrmse": 0.06037149134959371, "全帯域_10ms_envelope_correlation": 0.9956808634094217, "全帯域_10ms_modulation_mae_db": 0.4561447203096575, "全帯域_40ms_log_rms_mae_db": 0.49756372115000974, "全帯域_40ms_rms_envelope_nrmse": 0.030895549276190776, "全帯域_40ms_envelope_correlation": 0.9987551243313687, "全帯域_40ms_modulation_mae_db": 0.3967601407777284, "300–1000 Hz_10ms_log_rms_mae_db": 1.567122787279738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0924960330754986, "300–1000 Hz_10ms_envelope_correlation": 0.9934802299930995, "300–1000 Hz_10ms_modulation_mae_db": 0.5294648652239885, "300–1000 Hz_40ms_log_rms_mae_db": 0.9047188300863787, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0532764448137427, "300–1000 Hz_40ms_envelope_correlation": 0.9977648409480613, "300–1000 Hz_40ms_modulation_mae_db": 0.3495873018860343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8076435322857756, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14706107727614426, "1000–3000 Hz_10ms_envelope_correlation": 0.9857097025915414, "1000–3000 Hz_10ms_modulation_mae_db": 1.2673497792020736, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0953657354242328, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09570058547687554, "1000–3000 Hz_40ms_envelope_correlation": 0.9951498041292083, "1000–3000 Hz_40ms_modulation_mae_db": 1.4795676910213853, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9235170880324355, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.343150175364941, "3000–6000 Hz_10ms_envelope_correlation": 0.9358015203035603, "3000–6000 Hz_10ms_modulation_mae_db": 0.23447123541901316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3664507734606304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3096906817952583, "3000–6000 Hz_40ms_envelope_correlation": 0.9458485498864607, "3000–6000 Hz_40ms_modulation_mae_db": 0.8082567076488761, "mr_stft_db": 5.5087769838735765, "band_mae_db": 0.3443176788350056, "rms_gain_error_db": -0.04192580232759161, "output_peak": 0.696132871291892}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9859522381657572, "全帯域_10ms_rms_envelope_nrmse": 0.058688213341415746, "全帯域_10ms_envelope_correlation": 0.99592743624665, "全帯域_10ms_modulation_mae_db": 0.27545327417715, "全帯域_40ms_log_rms_mae_db": 0.5131529888212788, "全帯域_40ms_rms_envelope_nrmse": 0.02960923120041528, "全帯域_40ms_envelope_correlation": 0.9988755345784116, "全帯域_40ms_modulation_mae_db": 0.37349606906884975, "300–1000 Hz_10ms_log_rms_mae_db": 1.7113768864504988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1103363644797801, "300–1000 Hz_10ms_envelope_correlation": 0.9907374530449897, "300–1000 Hz_10ms_modulation_mae_db": 0.3737925091801675, "300–1000 Hz_40ms_log_rms_mae_db": 1.0506437060906124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06386573166167223, "300–1000 Hz_40ms_envelope_correlation": 0.9968438036500147, "300–1000 Hz_40ms_modulation_mae_db": 0.4258277563933271, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8476484872844632, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17888666993726804, "1000–3000 Hz_10ms_envelope_correlation": 0.9766762857133662, "1000–3000 Hz_10ms_modulation_mae_db": 0.9079826977780062, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1880984442583506, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11717500974291724, "1000–3000 Hz_40ms_envelope_correlation": 0.9900234912274783, "1000–3000 Hz_40ms_modulation_mae_db": 0.8873809731724774, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8845240397598566, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34578524982868625, "3000–6000 Hz_10ms_envelope_correlation": 0.9267813807713399, "3000–6000 Hz_10ms_modulation_mae_db": 0.6096056024072946, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3559551367002574, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3061476084112485, "3000–6000 Hz_40ms_envelope_correlation": 0.9397273076330968, "3000–6000 Hz_40ms_modulation_mae_db": 0.9049609922539924, "mr_stft_db": 5.509394621469888, "band_mae_db": 0.19448538937362733, "rms_gain_error_db": -0.04888542477456375, "output_peak": 0.5983319023347217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.050874922377228, "全帯域_10ms_rms_envelope_nrmse": 0.0645984833246095, "全帯域_10ms_envelope_correlation": 0.9950798547891695, "全帯域_10ms_modulation_mae_db": 0.19893622326125052, "全帯域_40ms_log_rms_mae_db": 0.5412532796304255, "全帯域_40ms_rms_envelope_nrmse": 0.03575822212632171, "全帯域_40ms_envelope_correlation": 0.9983682228981001, "全帯域_40ms_modulation_mae_db": 0.17895931692267994, "300–1000 Hz_10ms_log_rms_mae_db": 1.7113645075003294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10968075137329368, "300–1000 Hz_10ms_envelope_correlation": 0.9908842612594768, "300–1000 Hz_10ms_modulation_mae_db": 0.6265030044836645, "300–1000 Hz_40ms_log_rms_mae_db": 1.0108364413901543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0590752200076669, "300–1000 Hz_40ms_envelope_correlation": 0.9974085620956292, "300–1000 Hz_40ms_modulation_mae_db": 0.5493149107219695, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8498473007979603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1601005831362769, "1000–3000 Hz_10ms_envelope_correlation": 0.9826586797575165, "1000–3000 Hz_10ms_modulation_mae_db": 0.9546351013150963, "1000–3000 Hz_40ms_log_rms_mae_db": 1.12870335791707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10196829844067266, "1000–3000 Hz_40ms_envelope_correlation": 0.9942595348001193, "1000–3000 Hz_40ms_modulation_mae_db": 1.1289672256820649, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9123495418766883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3774886992789301, "3000–6000 Hz_10ms_envelope_correlation": 0.9081787964132252, "3000–6000 Hz_10ms_modulation_mae_db": 0.5378152317247141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.39468471979654, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32132235061405956, "3000–6000 Hz_40ms_envelope_correlation": 0.9302047123003242, "3000–6000 Hz_40ms_modulation_mae_db": 0.6928372223156781, "mr_stft_db": 5.517118489204184, "band_mae_db": 0.2260734539886217, "rms_gain_error_db": -0.0684423112233262, "output_peak": 0.5918721441582219}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9910764101540107, "全帯域_10ms_rms_envelope_nrmse": 0.06337174145254346, "全帯域_10ms_envelope_correlation": 0.9953351707828189, "全帯域_10ms_modulation_mae_db": 0.3688247516054914, "全帯域_40ms_log_rms_mae_db": 0.5641180972228396, "全帯域_40ms_rms_envelope_nrmse": 0.036747753996555776, "全帯域_40ms_envelope_correlation": 0.9983596644132346, "全帯域_40ms_modulation_mae_db": 0.23383763518209566, "300–1000 Hz_10ms_log_rms_mae_db": 1.7342135577815505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12671507939495016, "300–1000 Hz_10ms_envelope_correlation": 0.9880206930965716, "300–1000 Hz_10ms_modulation_mae_db": 0.3569464919450797, "300–1000 Hz_40ms_log_rms_mae_db": 1.054617269482422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07833640364682973, "300–1000 Hz_40ms_envelope_correlation": 0.9955653317802342, "300–1000 Hz_40ms_modulation_mae_db": 0.2702836979187279, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8179077405185613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1449484575280211, "1000–3000 Hz_10ms_envelope_correlation": 0.9855912287175495, "1000–3000 Hz_10ms_modulation_mae_db": 0.333600726278595, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1440374809175438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08731790721654431, "1000–3000 Hz_40ms_envelope_correlation": 0.9957271299313355, "1000–3000 Hz_40ms_modulation_mae_db": 0.39375296904953955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9031925481220817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35404955939724303, "3000–6000 Hz_10ms_envelope_correlation": 0.9268257463385501, "3000–6000 Hz_10ms_modulation_mae_db": 0.3263529335526649, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3740840778163046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3133357275817125, "3000–6000 Hz_40ms_envelope_correlation": 0.9400614012025179, "3000–6000 Hz_40ms_modulation_mae_db": 0.9202668083156902, "mr_stft_db": 5.519031984937537, "band_mae_db": 0.284095952634205, "rms_gain_error_db": -0.0999628148223042, "output_peak": 0.5981893826710394}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9394754190553096, "全帯域_10ms_rms_envelope_nrmse": 0.06378173986800763, "全帯域_10ms_envelope_correlation": 0.9953121100439535, "全帯域_10ms_modulation_mae_db": 0.6093843680579843, "全帯域_40ms_log_rms_mae_db": 0.5070858857209642, "全帯域_40ms_rms_envelope_nrmse": 0.0365944694421714, "全帯域_40ms_envelope_correlation": 0.9984359977779427, "全帯域_40ms_modulation_mae_db": 0.7189156031571886, "300–1000 Hz_10ms_log_rms_mae_db": 1.6731423583630647, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11244313222248238, "300–1000 Hz_10ms_envelope_correlation": 0.990401494112106, "300–1000 Hz_10ms_modulation_mae_db": 0.39642776918434497, "300–1000 Hz_40ms_log_rms_mae_db": 1.0061019382188396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0709935841320303, "300–1000 Hz_40ms_envelope_correlation": 0.9960798413291535, "300–1000 Hz_40ms_modulation_mae_db": 0.5885197828042635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8314983968758423, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15949234722828043, "1000–3000 Hz_10ms_envelope_correlation": 0.9830450563858961, "1000–3000 Hz_10ms_modulation_mae_db": 1.0073709280511869, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1073649909488668, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11098956269372116, "1000–3000 Hz_40ms_envelope_correlation": 0.9928950425879204, "1000–3000 Hz_40ms_modulation_mae_db": 1.024578521816219, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8989646274334537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37678746853251727, "3000–6000 Hz_10ms_envelope_correlation": 0.9071889767864767, "3000–6000 Hz_10ms_modulation_mae_db": 0.4105629547349359, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4039330318483236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.331978336886589, "3000–6000 Hz_40ms_envelope_correlation": 0.9247368703512461, "3000–6000 Hz_40ms_modulation_mae_db": 0.7499669204078788, "mr_stft_db": 5.520768134633074, "band_mae_db": 0.3140683822403029, "rms_gain_error_db": -0.11314440996237864, "output_peak": 0.7696013942675056}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.033610326270313, "全帯域_10ms_rms_envelope_nrmse": 0.07574250737308016, "全帯域_10ms_envelope_correlation": 0.9932275744383018, "全帯域_10ms_modulation_mae_db": 0.19747311219741556, "全帯域_40ms_log_rms_mae_db": 0.5867505972122532, "全帯域_40ms_rms_envelope_nrmse": 0.04127328722320521, "全帯域_40ms_envelope_correlation": 0.9978317113034375, "全帯域_40ms_modulation_mae_db": 0.10750164030045861, "300–1000 Hz_10ms_log_rms_mae_db": 1.7048286908095451, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13742635692429864, "300–1000 Hz_10ms_envelope_correlation": 0.9857280075793574, "300–1000 Hz_10ms_modulation_mae_db": 0.541639114326013, "300–1000 Hz_40ms_log_rms_mae_db": 1.0115241348811268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08023701827416352, "300–1000 Hz_40ms_envelope_correlation": 0.9952707643427914, "300–1000 Hz_40ms_modulation_mae_db": 0.5579923925684267, "1000–3000 Hz_10ms_log_rms_mae_db": 1.840562257515572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.179720244276347, "1000–3000 Hz_10ms_envelope_correlation": 0.977435028430974, "1000–3000 Hz_10ms_modulation_mae_db": 0.8676246208004507, "1000–3000 Hz_40ms_log_rms_mae_db": 1.133757606930863, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12829958428056398, "1000–3000 Hz_40ms_envelope_correlation": 0.989150773182319, "1000–3000 Hz_40ms_modulation_mae_db": 0.8870907363005749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.951123791978353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3616199116416566, "3000–6000 Hz_10ms_envelope_correlation": 0.9172897664277835, "3000–6000 Hz_10ms_modulation_mae_db": 0.39115720233168777, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3961632944240352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3132754411785216, "3000–6000 Hz_40ms_envelope_correlation": 0.9346733082986781, "3000–6000 Hz_40ms_modulation_mae_db": 1.126143348177797, "mr_stft_db": 5.51511833994229, "band_mae_db": 0.19845958354887824, "rms_gain_error_db": -0.08175883508617962, "output_peak": 0.6833186940060254}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.021409387777207, "全帯域_10ms_rms_envelope_nrmse": 0.058843650443302455, "全帯域_10ms_envelope_correlation": 0.995961679540935, "全帯域_10ms_modulation_mae_db": 0.08489601875852443, "全帯域_40ms_log_rms_mae_db": 0.5228945014971579, "全帯域_40ms_rms_envelope_nrmse": 0.03204650474516659, "全帯域_40ms_envelope_correlation": 0.9987484415889154, "全帯域_40ms_modulation_mae_db": 0.17296452856053293, "300–1000 Hz_10ms_log_rms_mae_db": 1.7064721311801712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11387736453845221, "300–1000 Hz_10ms_envelope_correlation": 0.9905726222005988, "300–1000 Hz_10ms_modulation_mae_db": 0.49675530110280475, "300–1000 Hz_40ms_log_rms_mae_db": 1.0467903998566972, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07276719627125668, "300–1000 Hz_40ms_envelope_correlation": 0.996391567419304, "300–1000 Hz_40ms_modulation_mae_db": 0.6177336051627492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8450572977433168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15426325767651558, "1000–3000 Hz_10ms_envelope_correlation": 0.9853502900925284, "1000–3000 Hz_10ms_modulation_mae_db": 0.817540628201137, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1757343148290316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10590059810774105, "1000–3000 Hz_40ms_envelope_correlation": 0.995046409605198, "1000–3000 Hz_40ms_modulation_mae_db": 0.8766742178948972, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8538934895893395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35680189481691504, "3000–6000 Hz_10ms_envelope_correlation": 0.9288705086851973, "3000–6000 Hz_10ms_modulation_mae_db": 0.306937166910809, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3216909977592988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3101884400113199, "3000–6000 Hz_40ms_envelope_correlation": 0.9443389275524093, "3000–6000 Hz_40ms_modulation_mae_db": 0.5371378367938143, "mr_stft_db": 5.501222711050286, "band_mae_db": 0.34851180245112645, "rms_gain_error_db": -0.08357090682793451, "output_peak": 0.6947386265982316}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6976694315994125, "spectral_convergence": 0.19801932444361883, "spectral_convergence_status": "ok", "amplitude_floor": 1.884454455950008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3635889883144126, "spectral_convergence": 0.1919511834259188, "spectral_convergence_status": "ok", "amplitude_floor": 1.301578596614512e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.507738805948546, "spectral_convergence": 0.19307833293279625, "spectral_convergence_status": "ok", "amplitude_floor": 1.1393421242566376e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004137534659655858, "output_power": 0.0040490400923013925, "delta_db": -0.09389567799432783, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026137240204796635, "output_power": 0.0025182469309942513, "delta_db": -0.16161415763585235, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008071793462917449, "output_power": 0.0008222350315212534, "delta_db": 0.08025935202007382, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004252507832785635, "output_power": 0.0003931472597143689, "delta_db": -0.3408986929578092, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.026398390024539e-05, "output_power": 6.432499217406355e-05, "delta_db": -0.3835302818809649, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.066875275629503e-05, "output_power": 5.160491841648099e-05, "delta_db": 0.07950881814265673, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18995116343861415, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1118266154525426, "pitch": null, "output_rms_dbfs": -21.024499217146875, "output_peak": 0.609067617981912, "reference_rms_dbfs": -20.91267260169434, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.522999075287458}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025953969251226393, "log_rms_mae_db": 1.0093851828958356, "rms_envelope_nrmse": 0.0649740675091112, "envelope_correlation": 0.9951241306016148, "rms_gain_error_db": -0.11148674840723483, "reference_mean_power": 0.008109529837298932, "output_mean_power": 0.007904001267013717}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5635786081597607e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013500334014131179, "output_power": 0.001351003211512708, "reference_active": true, "delta_db": 0.00311867814623297}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020457869490256438, "output_power": 0.0001860923383850878, "reference_active": true, "delta_db": -0.4113191058214574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3880284077345985e-05, "output_power": 3.230592008419376e-05, "reference_active": true, "delta_db": -0.20664928704928326}], "active_band_count": 3, "active_band_mae_db": 0.20702902367232454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024049824298809385, "log_rms_mae_db": 0.5502230817680893, "rms_envelope_nrmse": 0.041216663523816165, "envelope_correlation": 0.9979395279125262, "rms_gain_error_db": -0.11196399487346095, "reference_mean_power": 0.008139876536629037, "output_mean_power": 0.00793270708362308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2818780696338306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012006574574251973, "output_power": 0.0012127197276101732, "reference_active": true, "delta_db": 0.043413197366071875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.159020687440259e-05, "output_power": 8.539049072029676e-05, "reference_active": true, "delta_db": -0.3043953060024839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2353163446294293e-06, "output_power": 2.0306976871232734e-06, "reference_active": true, "delta_db": -0.4169371951479928}], "active_band_count": 3, "active_band_mae_db": 0.25491523283884954}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002074995834794702, "log_rms_mae_db": 1.7066341505804121, "rms_envelope_nrmse": 0.10693602196458853, "envelope_correlation": 0.9913883430182441, "rms_gain_error_db": -0.18783854935489133, "reference_mean_power": 0.0020359851811281022, "output_mean_power": 0.0019498031053729358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.345568889858965e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045532158312677247, "output_power": 0.0004330296537610302, "reference_active": true, "delta_db": -0.2180059944962626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.561332252218219e-05, "output_power": 7.644761133812712e-05, "reference_active": true, "delta_db": -0.4917743157293742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0669601144719856e-05, "output_power": 9.04821107706216e-06, "reference_active": true, "delta_db": -0.7158546148242894}], "active_band_count": 3, "active_band_mae_db": 0.4752116416833087}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001833599147850085, "log_rms_mae_db": 1.050376159161232, "rms_envelope_nrmse": 0.0628547336668817, "envelope_correlation": 0.9970491789588763, "rms_gain_error_db": -0.18774863817831905, "reference_mean_power": 0.0020457391187242045, "output_mean_power": 0.0019591847246452337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.75393473065512e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00043556686984963635, "output_power": 0.0004140690412413518, "reference_active": true, "delta_db": -0.21982078210997488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.69121464262054e-05, "output_power": 3.4302281230425955e-05, "reference_active": true, "delta_db": -0.31846296760295834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.183033529622033e-07, "output_power": 6.386133779887906e-07, "reference_active": true, "delta_db": -0.5106988156334945}], "active_band_count": 3, "active_band_mae_db": 0.3496608551154759}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011093565856595723, "log_rms_mae_db": 1.7920884291789556, "rms_envelope_nrmse": 0.14887612253283114, "envelope_correlation": 0.9840784434429267, "rms_gain_error_db": -0.3372516768540315, "reference_mean_power": 0.00042557205365836593, "output_mean_power": 0.0003937748125300163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.99964400169894e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011758808293094384, "output_power": 0.00010400255202950605, "reference_active": true, "delta_db": -0.5331931379930291}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1475307548341578e-05, "output_power": 1.777311120603307e-05, "reference_active": true, "delta_db": -0.8217593381929457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0561557218273926e-06, "output_power": 4.181703689432595e-06, "reference_active": true, "delta_db": -0.8246718510343445}], "active_band_count": 3, "active_band_mae_db": 0.7265414424067731}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010031642073947198, "log_rms_mae_db": 1.0395162003338856, "rms_envelope_nrmse": 0.08957164908820577, "envelope_correlation": 0.9945627683545739, "rms_gain_error_db": -0.3372432225673916, "reference_mean_power": 0.0004276640413209328, "output_mean_power": 0.00039571126458043984}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8247169971067077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011139646210262638, "output_power": 9.893401066039071e-05, "reference_active": true, "delta_db": -0.5152578288905704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0378224135538677e-05, "output_power": 8.598755598760968e-06, "reference_active": true, "delta_db": -0.8168744056141369}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.220864105203889e-07, "output_power": 2.6351027488433665e-07, "reference_active": true, "delta_db": -0.8717484739998058}], "active_band_count": 3, "active_band_mae_db": 0.7346269028348377}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.985234496629778e-05, "log_rms_mae_db": 1.8819243914062795, "rms_envelope_nrmse": 0.3494240192495265, "envelope_correlation": 0.9238838994547404, "rms_gain_error_db": 0.05967003169532578, "reference_mean_power": 6.361685525644655e-05, "output_mean_power": 6.449695333294724e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.697638884988847e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.1912782580556425e-06, "output_power": 6.051449380879624e-06, "reference_active": true, "delta_db": -0.09920918333499}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.016318108845029e-06, "output_power": 2.1015361298686905e-06, "reference_active": true, "delta_db": 0.17977810250395043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.069070181696055e-07, "output_power": 8.758638351195637e-07, "reference_active": true, "delta_db": 1.5931443483702565}], "active_band_count": 3, "active_band_mae_db": 0.6240438780697323}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.357134521214995e-05, "log_rms_mae_db": 1.3010582901834684, "rms_envelope_nrmse": 0.3098004180077721, "envelope_correlation": 0.9370206493111047, "rms_gain_error_db": 0.05975318728239202, "reference_mean_power": 6.39232761364752e-05, "output_mean_power": 6.480885424930653e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.609918878818377e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.077093160223491e-06, "output_power": 6.164360848130976e-06, "reference_active": true, "delta_db": 0.06192159417262414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.511523340291838e-07, "output_power": 1.0939128479895506e-06, "reference_active": true, "delta_db": 0.6073264518707886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7938163224025654e-08, "output_power": 5.750179869999684e-08, "reference_active": true, "delta_db": 3.13483579649204}], "active_band_count": 3, "active_band_mae_db": 1.2680279475118177}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9004171505784418, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f2bb27d87794fa5b8557b349831e91c532f8082dad45951e5c35f395b9c578ff", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9825680893915374, "standard_deviation": 0.0035922194595532516, "minimum": 0.9766762857133662, "maximum": 0.9857097025915414, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.829031680275056, "standard_deviation": 0.021117811000511265, "minimum": 1.7920884291789556, "maximum": 1.8498473007979603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8603307405041649, "standard_deviation": 0.26610352837317275, "minimum": 0.333600726278595, "maximum": 1.2673497792020736, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15916859494896057, "standard_deviation": 0.013580173591890848, "minimum": 0.1449484575280211, "maximum": 0.179720244276347, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.993351869227269, "standard_deviation": 0.002477672657145764, "minimum": 0.989150773182319, "maximum": 0.9957271299313355, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1265722664449807, "standard_deviation": 0.04703358617602525, "minimum": 1.0395162003338856, "maximum": 1.1880984442583506, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9265799047214995, "standard_deviation": 0.31229493271141445, "minimum": 0.39375296904953955, "maximum": 1.4795676910213853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10461539938090522, "standard_deviation": 0.014004921731250543, "minimum": 0.08731790721654431, "maximum": 0.12829958428056398, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9218525743976091, "standard_deviation": 0.01014036391500124, "minimum": 0.9071889767864767, "maximum": 0.9358015203035603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.901186189774811, "standard_deviation": 0.02930445598583448, "minimum": 1.8538934895893395, "maximum": 1.951123791978353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4301182756438565, "standard_deviation": 0.14517991105631642, "minimum": 0.23447123541901316, "maximum": 0.6240438780697323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.35813837226380196, "standard_deviation": 0.013125957134781947, "minimum": 0.343150175364941, "maximum": 0.3774886992789301, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9370764658169797, "standard_deviation": 0.00708009791258218, "minimum": 0.9247368703512461, "maximum": 0.9458485498864607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3642525402486072, "standard_deviation": 0.03682835011215745, "minimum": 1.3010582901834684, "maximum": 1.4039330318483236, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8759497229281932, "standard_deviation": 0.23540407055234835, "minimum": 0.5371378367938143, "maximum": 1.2680279475118177, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.31446737556081017, "standard_deviation": 0.008353364550930612, "minimum": 0.3061476084112485, "maximum": 0.331978336886589, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9901516380380555, "standard_deviation": 0.002324956777470172, "minimum": 0.9857280075793574, "maximum": 0.9934802299930995, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6893943837431638, "standard_deviation": 0.0521210453903505, "minimum": 1.567122787279738, "maximum": 1.7342135577815505, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4745925871411715, "standard_deviation": 0.09349220435297083, "minimum": 0.3569464919450797, "maximum": 0.6265030044836645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11373888799666804, "standard_deviation": 0.01339914387867746, "minimum": 0.0924960330754986, "maximum": 0.13742635692429864, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996546736315508, "standard_deviation": 0.0008783546171175187, "minimum": 0.9952707643427914, "maximum": 0.9977648409480613, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.016951109895933, "standard_deviation": 0.04977402958250252, "minimum": 0.9047188300863787, "maximum": 1.054617269482422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4636150378213717, "standard_deviation": 0.13114770700452413, "minimum": 0.2702836979187279, "maximum": 0.6177336051627492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06767579155928047, "standard_deviation": 0.009467488651007551, "minimum": 0.0532764448137427, "maximum": 0.08023701827416352, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2624954258137977, "standard_deviation": 0.06817544061268381, "minimum": 0.18995116343861415, "maximum": 0.34851180245112645, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.514303792549787, "standard_deviation": 0.007310440053747151, "minimum": 5.501222711050286, "maximum": 5.522999075287458, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6551565791636937, "standard_deviation": 0.06518099289336161, "minimum": 0.5918721441582219, "maximum": 0.7696013942675056, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08118964005960265, "standard_deviation": 0.026901576578669515, "minimum": -0.11314440996237864, "maximum": -0.04192580232759161, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952061024816081, "standard_deviation": 0.0008686396876320163, "minimum": 0.9932275744383018, "maximum": 0.995961679540935, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0068017674853273, "standard_deviation": 0.0345306619451685, "minimum": 0.9394754190553096, "maximum": 1.050874922377228, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29976768650497476, "standard_deviation": 0.16948568931638755, "minimum": 0.08489601875852443, "maximum": 0.6093843680579843, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06379648683270799, "standard_deviation": 0.005441553978490827, "minimum": 0.058688213341415746, "maximum": 0.07574250737308016, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984142781004921, "standard_deviation": 0.0003800274846106806, "minimum": 0.9978317113034375, "maximum": 0.9988755345784116, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5353802691278773, "standard_deviation": 0.03070640818890061, "minimum": 0.49756372115000974, "maximum": 0.5867505972122532, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.30466877085104793, "standard_deviation": 0.19432471771655516, "minimum": 0.10750164030045861, "maximum": 0.7189156031571886, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03551771019173036, "standard_deviation": 0.004418447740595602, "minimum": 0.02960923120041528, "maximum": 0.04127328722320521, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0019", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.697422742843628, "predicted_mos": 2.697422742843628, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c4a3c3ab992dbc93cfafc3962996584b1c1123daa429a6e2abdc9c4fe2c5bedb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.338150199968368, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9469349678474412, "全帯域_10ms_rms_envelope_nrmse": 0.061343308282246274, "全帯域_10ms_envelope_correlation": 0.9956327931986654, "全帯域_10ms_modulation_mae_db": 0.15503490853532342, "全帯域_40ms_log_rms_mae_db": 0.4884109585512152, "全帯域_40ms_rms_envelope_nrmse": 0.03333084318737704, "全帯域_40ms_envelope_correlation": 0.998678187921266, "全帯域_40ms_modulation_mae_db": 0.15819843876021342, "300–1000 Hz_10ms_log_rms_mae_db": 1.777955733213029, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11456439954335489, "300–1000 Hz_10ms_envelope_correlation": 0.9900727817834278, "300–1000 Hz_10ms_modulation_mae_db": 0.3486211352715154, "300–1000 Hz_40ms_log_rms_mae_db": 1.0853920861987285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058734244009977336, "300–1000 Hz_40ms_envelope_correlation": 0.9974959930620767, "300–1000 Hz_40ms_modulation_mae_db": 0.3487554752525938, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9744915661169535, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18030543654319792, "1000–3000 Hz_10ms_envelope_correlation": 0.9757079866718532, "1000–3000 Hz_10ms_modulation_mae_db": 0.8214370058264343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1751353095463712, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13080603993369347, "1000–3000 Hz_40ms_envelope_correlation": 0.9863548822384771, "1000–3000 Hz_40ms_modulation_mae_db": 0.8922236912402042, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1893936748705465, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.38265459660277445, "3000–6000 Hz_10ms_envelope_correlation": 0.9038623468605168, "3000–6000 Hz_10ms_modulation_mae_db": 0.620021074495492, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5152552602137486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33837742364675755, "3000–6000 Hz_40ms_envelope_correlation": 0.9216390463914904, "3000–6000 Hz_40ms_modulation_mae_db": 1.0438716938654238, "mr_stft_db": 5.677792776301369, "band_mae_db": 0.19692037402919715, "rms_gain_error_db": -0.09253039130554973, "output_peak": 0.6061180473494444}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9206569150498629, "全帯域_10ms_rms_envelope_nrmse": 0.06427378432629552, "全帯域_10ms_envelope_correlation": 0.995140271155645, "全帯域_10ms_modulation_mae_db": 0.1590703165276768, "全帯域_40ms_log_rms_mae_db": 0.48429754034469047, "全帯域_40ms_rms_envelope_nrmse": 0.03412986409413716, "全帯域_40ms_envelope_correlation": 0.9985361706604798, "全帯域_40ms_modulation_mae_db": 0.30651024424990064, "300–1000 Hz_10ms_log_rms_mae_db": 1.705252020346266, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11746928542697586, "300–1000 Hz_10ms_envelope_correlation": 0.989367497258691, "300–1000 Hz_10ms_modulation_mae_db": 0.5121834917342031, "300–1000 Hz_40ms_log_rms_mae_db": 1.038043356344761, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06470685863840364, "300–1000 Hz_40ms_envelope_correlation": 0.9966339944295769, "300–1000 Hz_40ms_modulation_mae_db": 0.4451637909550396, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0498438806984898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17390599952383906, "1000–3000 Hz_10ms_envelope_correlation": 0.9784820316681415, "1000–3000 Hz_10ms_modulation_mae_db": 0.7907880108582831, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2796889915760343, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1259555972719159, "1000–3000 Hz_40ms_envelope_correlation": 0.9889018882517336, "1000–3000 Hz_40ms_modulation_mae_db": 0.5782460479025557, "3000–6000 Hz_10ms_log_rms_mae_db": 2.134518400582724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.349974483191831, "3000–6000 Hz_10ms_envelope_correlation": 0.9268211126987085, "3000–6000 Hz_10ms_modulation_mae_db": 0.4391327351808876, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5410089082940341, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30500998851972316, "3000–6000 Hz_40ms_envelope_correlation": 0.9417860296332963, "3000–6000 Hz_40ms_modulation_mae_db": 1.01687589745386, "mr_stft_db": 5.67779897242086, "band_mae_db": 0.24680749991280068, "rms_gain_error_db": -0.07477563789924682, "output_peak": 0.737726944428828}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.997604675320386, "全帯域_10ms_rms_envelope_nrmse": 0.06270788640022563, "全帯域_10ms_envelope_correlation": 0.9954377784475507, "全帯域_10ms_modulation_mae_db": 0.32130746234930035, "全帯域_40ms_log_rms_mae_db": 0.5318282966096101, "全帯域_40ms_rms_envelope_nrmse": 0.032643165936761184, "全帯域_40ms_envelope_correlation": 0.9987452939646243, "全帯域_40ms_modulation_mae_db": 0.2686499019196882, "300–1000 Hz_10ms_log_rms_mae_db": 1.778712371583356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11354316664440688, "300–1000 Hz_10ms_envelope_correlation": 0.9907111149107788, "300–1000 Hz_10ms_modulation_mae_db": 0.4863756876774086, "300–1000 Hz_40ms_log_rms_mae_db": 1.1012440374143675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06931628832639163, "300–1000 Hz_40ms_envelope_correlation": 0.9969000988267372, "300–1000 Hz_40ms_modulation_mae_db": 0.5374582893219967, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0457213810713712, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16730079272656628, "1000–3000 Hz_10ms_envelope_correlation": 0.9805968485102443, "1000–3000 Hz_10ms_modulation_mae_db": 1.3433911330458068, "1000–3000 Hz_40ms_log_rms_mae_db": 1.269587716018835, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10718393405158838, "1000–3000 Hz_40ms_envelope_correlation": 0.9930683055430123, "1000–3000 Hz_40ms_modulation_mae_db": 1.4846923193518233, "3000–6000 Hz_10ms_log_rms_mae_db": 2.129775732508071, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3767566863082671, "3000–6000 Hz_10ms_envelope_correlation": 0.9289109166685705, "3000–6000 Hz_10ms_modulation_mae_db": 0.7719599981863275, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4784078498922215, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32650324222150007, "3000–6000 Hz_40ms_envelope_correlation": 0.9460091367254271, "3000–6000 Hz_40ms_modulation_mae_db": 0.8982026284577879, "mr_stft_db": 5.674303339578088, "band_mae_db": 0.36571522275064555, "rms_gain_error_db": -0.09564146889953562, "output_peak": 0.604044269210807}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9402199777287016, "全帯域_10ms_rms_envelope_nrmse": 0.05810333941955748, "全帯域_10ms_envelope_correlation": 0.99607075980598, "全帯域_10ms_modulation_mae_db": 0.25838756291573456, "全帯域_40ms_log_rms_mae_db": 0.515283447362203, "全帯域_40ms_rms_envelope_nrmse": 0.03017045849282319, "全帯域_40ms_envelope_correlation": 0.9989104219238922, "全帯域_40ms_modulation_mae_db": 0.3353978642932152, "300–1000 Hz_10ms_log_rms_mae_db": 1.7611023271664445, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1150869809111368, "300–1000 Hz_10ms_envelope_correlation": 0.9903070430638474, "300–1000 Hz_10ms_modulation_mae_db": 0.29857662746343266, "300–1000 Hz_40ms_log_rms_mae_db": 1.088012358146971, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07671112279802188, "300–1000 Hz_40ms_envelope_correlation": 0.9958242832956863, "300–1000 Hz_40ms_modulation_mae_db": 0.5489462091306079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9932835138924534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.162565493846788, "1000–3000 Hz_10ms_envelope_correlation": 0.9814904547144588, "1000–3000 Hz_10ms_modulation_mae_db": 0.9936243718028441, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2250977458462544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10960171007232963, "1000–3000 Hz_40ms_envelope_correlation": 0.99218488062856, "1000–3000 Hz_40ms_modulation_mae_db": 1.028595508518438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.131832537461501, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3588860977173378, "3000–6000 Hz_10ms_envelope_correlation": 0.9315828203868873, "3000–6000 Hz_10ms_modulation_mae_db": 0.8562547964953223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5485889726135424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3166144543472988, "3000–6000 Hz_40ms_envelope_correlation": 0.9451507629378632, "3000–6000 Hz_40ms_modulation_mae_db": 1.4939030816750467, "mr_stft_db": 5.692191115739672, "band_mae_db": 0.3327180024210956, "rms_gain_error_db": -0.08687922989684443, "output_peak": 0.5801676316012763}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9884139699436757, "全帯域_10ms_rms_envelope_nrmse": 0.06371280800167216, "全帯域_10ms_envelope_correlation": 0.9952156873941258, "全帯域_10ms_modulation_mae_db": 0.36495096220511947, "全帯域_40ms_log_rms_mae_db": 0.5225576197829919, "全帯域_40ms_rms_envelope_nrmse": 0.034141916236323205, "全帯域_40ms_envelope_correlation": 0.9985188126924094, "全帯域_40ms_modulation_mae_db": 0.41851374066837427, "300–1000 Hz_10ms_log_rms_mae_db": 1.8048313168930463, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12435815044419453, "300–1000 Hz_10ms_envelope_correlation": 0.9883425914555403, "300–1000 Hz_10ms_modulation_mae_db": 0.6901384654944995, "300–1000 Hz_40ms_log_rms_mae_db": 1.1838450746274993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0721885358994578, "300–1000 Hz_40ms_envelope_correlation": 0.9961395503001202, "300–1000 Hz_40ms_modulation_mae_db": 0.6287867085266362, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0469501068491027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1548174049108775, "1000–3000 Hz_10ms_envelope_correlation": 0.9834768695422529, "1000–3000 Hz_10ms_modulation_mae_db": 0.8471004340317538, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2437929199411069, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09596796780662611, "1000–3000 Hz_40ms_envelope_correlation": 0.994641855254828, "1000–3000 Hz_40ms_modulation_mae_db": 0.859027522601049, "3000–6000 Hz_10ms_log_rms_mae_db": 2.114447358776923, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.378967401975945, "3000–6000 Hz_10ms_envelope_correlation": 0.920254413242271, "3000–6000 Hz_10ms_modulation_mae_db": 0.6289477329991319, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5153118859018688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31451901247217184, "3000–6000 Hz_40ms_envelope_correlation": 0.943392864332546, "3000–6000 Hz_40ms_modulation_mae_db": 0.967311381270179, "mr_stft_db": 5.68763737056198, "band_mae_db": 0.2954551764959556, "rms_gain_error_db": -0.06851764135671186, "output_peak": 0.6045462780528305}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9589387173955439, "全帯域_10ms_rms_envelope_nrmse": 0.06207396593479966, "全帯域_10ms_envelope_correlation": 0.9955133745113232, "全帯域_10ms_modulation_mae_db": 0.27566740469636625, "全帯域_40ms_log_rms_mae_db": 0.4620332796553727, "全帯域_40ms_rms_envelope_nrmse": 0.031863853424257065, "全帯域_40ms_envelope_correlation": 0.9987887585636157, "全帯域_40ms_modulation_mae_db": 0.4617621032323198, "300–1000 Hz_10ms_log_rms_mae_db": 1.8223483024205591, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12116675450839358, "300–1000 Hz_10ms_envelope_correlation": 0.9888502417414855, "300–1000 Hz_10ms_modulation_mae_db": 0.6856478727909208, "300–1000 Hz_40ms_log_rms_mae_db": 1.1723395277675273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06631993974971102, "300–1000 Hz_40ms_envelope_correlation": 0.9967050437685668, "300–1000 Hz_40ms_modulation_mae_db": 0.7701060257192779, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0605014196828937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16621910576104304, "1000–3000 Hz_10ms_envelope_correlation": 0.9805195605859809, "1000–3000 Hz_10ms_modulation_mae_db": 0.8518969194784863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3023467001533817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11033700953861073, "1000–3000 Hz_40ms_envelope_correlation": 0.9920157400776157, "1000–3000 Hz_40ms_modulation_mae_db": 0.8952016364568002, "3000–6000 Hz_10ms_log_rms_mae_db": 2.153443032433676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34425378866919853, "3000–6000 Hz_10ms_envelope_correlation": 0.926287072968866, "3000–6000 Hz_10ms_modulation_mae_db": 0.5258392009183893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5343996952594463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.308860402751825, "3000–6000 Hz_40ms_envelope_correlation": 0.9376559260866684, "3000–6000 Hz_40ms_modulation_mae_db": 0.9057250104888498, "mr_stft_db": 5.699546791037519, "band_mae_db": 0.19143190303488158, "rms_gain_error_db": -0.09306781917198857, "output_peak": 0.590141869846386}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9753272728941142, "全帯域_10ms_rms_envelope_nrmse": 0.07064155926356157, "全帯域_10ms_envelope_correlation": 0.9943146804871005, "全帯域_10ms_modulation_mae_db": 0.22436347943317395, "全帯域_40ms_log_rms_mae_db": 0.4899908130165108, "全帯域_40ms_rms_envelope_nrmse": 0.04310957850018778, "全帯域_40ms_envelope_correlation": 0.997866948108004, "全帯域_40ms_modulation_mae_db": 0.2142749442910492, "300–1000 Hz_10ms_log_rms_mae_db": 1.8113032450228024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14185567285283537, "300–1000 Hz_10ms_envelope_correlation": 0.9849208251116444, "300–1000 Hz_10ms_modulation_mae_db": 0.4611827581269344, "300–1000 Hz_40ms_log_rms_mae_db": 1.1222760201891733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09301315491447255, "300–1000 Hz_40ms_envelope_correlation": 0.9935716831661704, "300–1000 Hz_40ms_modulation_mae_db": 0.44737961055684927, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0220625923947666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16109054608507234, "1000–3000 Hz_10ms_envelope_correlation": 0.9812414733839986, "1000–3000 Hz_10ms_modulation_mae_db": 0.6434176801319639, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2248633966222877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0970547301687832, "1000–3000 Hz_40ms_envelope_correlation": 0.9935366356804388, "1000–3000 Hz_40ms_modulation_mae_db": 0.7147630454607133, "3000–6000 Hz_10ms_log_rms_mae_db": 2.128334580348799, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35327734178673087, "3000–6000 Hz_10ms_envelope_correlation": 0.9218438132443786, "3000–6000 Hz_10ms_modulation_mae_db": 0.30796246060104826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5092070522444365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31170148136997183, "3000–6000 Hz_40ms_envelope_correlation": 0.9360656482475219, "3000–6000 Hz_40ms_modulation_mae_db": 0.742223543851725, "mr_stft_db": 5.683098087072183, "band_mae_db": 0.141982376946501, "rms_gain_error_db": -0.14736028254115136, "output_peak": 0.6029338855765579}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.976075152540462, "全帯域_10ms_rms_envelope_nrmse": 0.0668431700969175, "全帯域_10ms_envelope_correlation": 0.9947038746581454, "全帯域_10ms_modulation_mae_db": 0.3383373869879076, "全帯域_40ms_log_rms_mae_db": 0.5036374964187966, "全帯域_40ms_rms_envelope_nrmse": 0.0386061150571452, "全帯域_40ms_envelope_correlation": 0.9980525824773125, "全帯域_40ms_modulation_mae_db": 0.3475769312073049, "300–1000 Hz_10ms_log_rms_mae_db": 1.7565261604736304, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10486327589531012, "300–1000 Hz_10ms_envelope_correlation": 0.9915093661483715, "300–1000 Hz_10ms_modulation_mae_db": 0.38779912162024116, "300–1000 Hz_40ms_log_rms_mae_db": 1.0944867917164585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06708263212979301, "300–1000 Hz_40ms_envelope_correlation": 0.9962521916958219, "300–1000 Hz_40ms_modulation_mae_db": 0.4494615577938788, "1000–3000 Hz_10ms_log_rms_mae_db": 2.054687641937974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15297712915733638, "1000–3000 Hz_10ms_envelope_correlation": 0.9832246986463188, "1000–3000 Hz_10ms_modulation_mae_db": 1.211212453360085, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2159442967373348, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0920388903366548, "1000–3000 Hz_40ms_envelope_correlation": 0.9943378352677273, "1000–3000 Hz_40ms_modulation_mae_db": 0.9004627182135018, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1249775973368923, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4021121672637087, "3000–6000 Hz_10ms_envelope_correlation": 0.9249148837442092, "3000–6000 Hz_10ms_modulation_mae_db": 0.5899052225349412, "3000–6000 Hz_40ms_log_rms_mae_db": 1.481473435808829, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3480574867342073, "3000–6000 Hz_40ms_envelope_correlation": 0.9441957560489389, "3000–6000 Hz_40ms_modulation_mae_db": 1.2875282511384931, "mr_stft_db": 5.694329856922047, "band_mae_db": 0.358098138147704, "rms_gain_error_db": -0.047647453618334466, "output_peak": 0.7312340654672531}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.921821767675956, "spectral_convergence": 0.18973594340589214, "spectral_convergence_status": "ok", "amplitude_floor": 1.884454455950008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.546008067205531, "spectral_convergence": 0.19897298338958297, "spectral_convergence_status": "ok", "amplitude_floor": 1.301578596614512e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.565548494022618, "spectral_convergence": 0.19708179078248478, "spectral_convergence_status": "ok", "amplitude_floor": 1.1393421242566376e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004137534659655858, "output_power": 0.004049973856024657, "delta_db": -0.09289425131004392, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026137240204796635, "output_power": 0.0025754676321773098, "delta_db": -0.06403632960765708, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008071793462917449, "output_power": 0.0007867635295516601, "delta_db": -0.11125820877925108, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004252507832785635, "output_power": 0.00040949517367836445, "delta_db": -0.16396334702068627, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.026398390024539e-05, "output_power": 6.803775005882989e-05, "delta_db": -0.13982826744575694, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.066875275629503e-05, "output_power": 4.4033786256157786e-05, "delta_db": -0.6095418400117877, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19692037402919715, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09253039130554973, "pitch": null, "output_rms_dbfs": -21.005202992999887, "output_peak": 0.6061180473494444, "reference_rms_dbfs": -20.91267260169434, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.677792776301369}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025953969251226393, "log_rms_mae_db": 0.9469349678474412, "rms_envelope_nrmse": 0.061343308282246274, "envelope_correlation": 0.9956327931986654, "rms_gain_error_db": -0.09252756580198251, "reference_mean_power": 0.008109529837298932, "output_mean_power": 0.007938581714057783}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5635786081597607e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013500334014131179, "output_power": 0.0013458433528097014, "reference_active": true, "delta_db": -0.01349999735559433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020457869490256438, "output_power": 0.00018486838857943507, "reference_active": true, "delta_db": -0.43997748051959673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3880284077345985e-05, "output_power": 3.378969863178938e-05, "reference_active": true, "delta_db": -0.011627247730779234}], "active_band_count": 3, "active_band_mae_db": 0.15503490853532342}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024049824298809385, "log_rms_mae_db": 0.4884109585512152, "rms_envelope_nrmse": 0.03333084318737704, "envelope_correlation": 0.998678187921266, "rms_gain_error_db": -0.0928650323852705, "reference_mean_power": 0.008139876536629037, "output_mean_power": 0.007967669559579591}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2818780696338306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012006574574251973, "output_power": 0.0012046754833130759, "reference_active": true, "delta_db": 0.014509493175095103}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.159020687440259e-05, "output_power": 8.352190595347804e-05, "reference_active": true, "delta_db": -0.4004864367250743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2353163446294293e-06, "output_power": 2.2662036376381245e-06, "reference_active": true, "delta_db": 0.05959938638047086}], "active_band_count": 3, "active_band_mae_db": 0.15819843876021342}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002074995834794702, "log_rms_mae_db": 1.777955733213029, "rms_envelope_nrmse": 0.11456439954335489, "envelope_correlation": 0.9900727817834278, "rms_gain_error_db": -0.18937214496028032, "reference_mean_power": 0.0020359851811281022, "output_mean_power": 0.0019491147057491861}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.345568889858965e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045532158312677247, "output_power": 0.00043300390837755146, "reference_active": true, "delta_db": -0.21826420799909663}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.561332252218219e-05, "output_power": 7.141840192626279e-05, "reference_active": true, "delta_db": -0.7873122358515086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0669601144719856e-05, "output_power": 1.0571083141469013e-05, "reference_active": true, "delta_db": -0.04028696196394109}], "active_band_count": 3, "active_band_mae_db": 0.3486211352715154}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001833599147850085, "log_rms_mae_db": 1.0853920861987285, "rms_envelope_nrmse": 0.058734244009977336, "envelope_correlation": 0.9974959930620767, "rms_gain_error_db": -0.18926314911527545, "reference_mean_power": 0.0020457391187242045, "output_mean_power": 0.0019585016191715995}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.75393473065512e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00043556686984963635, "output_power": 0.0004122970144714253, "reference_active": true, "delta_db": -0.23844648974235333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.69121464262054e-05, "output_power": 3.2015526802677616e-05, "reference_active": true, "delta_db": -0.6180864761546369}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.183033529622033e-07, "output_power": 6.875979095910966e-07, "reference_active": true, "delta_db": -0.18973345986079115}], "active_band_count": 3, "active_band_mae_db": 0.3487554752525938}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011093565856595723, "log_rms_mae_db": 1.9744915661169535, "rms_envelope_nrmse": 0.18030543654319792, "envelope_correlation": 0.9757079866718532, "rms_gain_error_db": -0.17186618763322228, "reference_mean_power": 0.00042557205365836593, "output_mean_power": 0.0004090594992132587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.99964400169894e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011758808293094384, "output_power": 0.00010985684742975807, "reference_active": true, "delta_db": -0.2953617812634783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1475307548341578e-05, "output_power": 1.85042841839951e-05, "reference_active": true, "delta_db": -0.646671024926929}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0561557218273926e-06, "output_power": 3.5611692933100477e-06, "reference_active": true, "delta_db": -1.5222782112888953}], "active_band_count": 3, "active_band_mae_db": 0.8214370058264343}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010031642073947198, "log_rms_mae_db": 1.1751353095463712, "rms_envelope_nrmse": 0.13080603993369347, "envelope_correlation": 0.9863548822384771, "rms_gain_error_db": -0.17194353558271913, "reference_mean_power": 0.0004276640413209328, "output_mean_power": 0.0004110629948841301}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8247169971067077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011139646210262638, "output_power": 0.00010408790154448487, "reference_active": true, "delta_db": -0.2947114500899252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0378224135538677e-05, "output_power": 9.205442097021968e-06, "reference_active": true, "delta_db": -0.5207839519325694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.220864105203889e-07, "output_power": 2.0982384166815274e-07, "reference_active": true, "delta_db": -1.8611756716981178}], "active_band_count": 3, "active_band_mae_db": 0.8922236912402042}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.985234496629778e-05, "log_rms_mae_db": 2.1893936748705465, "rms_envelope_nrmse": 0.38265459660277445, "envelope_correlation": 0.9038623468605168, "rms_gain_error_db": -0.5385800972895981, "reference_mean_power": 6.361685525644655e-05, "output_mean_power": 5.6197136556506626e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.697638884988847e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.1912782580556425e-06, "output_power": 5.124778775933996e-06, "reference_active": true, "delta_db": -0.8210520047925187}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.016318108845029e-06, "output_power": 1.6853837107966563e-06, "reference_active": true, "delta_db": -0.7786025844248745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.069070181696055e-07, "output_power": 6.444111155907856e-07, "reference_active": true, "delta_db": 0.2604086342690828}], "active_band_count": 3, "active_band_mae_db": 0.620021074495492}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.357134521214995e-05, "log_rms_mae_db": 1.5152552602137486, "rms_envelope_nrmse": 0.33837742364675755, "envelope_correlation": 0.9216390463914904, "rms_gain_error_db": -0.5387632113008922, "reference_mean_power": 6.39232761364752e-05, "output_mean_power": 5.646543832645321e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.609918878818377e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.077093160223491e-06, "output_power": 5.107340183145964e-06, "reference_active": true, "delta_db": -0.7550110827014753}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.511523340291838e-07, "output_power": 8.821213061353206e-07, "reference_active": true, "delta_db": -0.3272176611070855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7938163224025654e-08, "output_power": 4.4785402746367697e-08, "reference_active": true, "delta_db": 2.049386337787711}], "active_band_count": 3, "active_band_mae_db": 1.0438716938654238}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8939476544883853, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bbd28770019ca5437244a610791c1b2dfd1c0080d0b365e846b4803db338beb6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9805924904654061, "standard_deviation": 0.0025296366930526565, "minimum": 0.9757079866718532, "maximum": 0.9834768695422529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0309427628305006, "standard_deviation": 0.03151698504204182, "minimum": 1.9744915661169535, "maximum": 2.0605014196828937, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9378585010669571, "standard_deviation": 0.23292992265951937, "minimum": 0.6434176801319639, "maximum": 1.3433911330458068, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16489773856934006, "standard_deviation": 0.009178505467305248, "minimum": 0.15297712915733638, "maximum": 0.18030543654319792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.991880252867799, "standard_deviation": 0.0028618865758652705, "minimum": 0.9863548822384771, "maximum": 0.994641855254828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2420571345552007, "standard_deviation": 0.0406300218001892, "minimum": 1.1751353095463712, "maximum": 1.3023467001533817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9191515612181357, "standard_deviation": 0.265939961818367, "minimum": 0.5782460479025557, "maximum": 1.4846923193518233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10861823489752528, "standard_deviation": 0.013976919853702127, "minimum": 0.0920388903366548, "maximum": 0.13080603993369347, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.923059672476801, "standard_deviation": 0.008558809730985066, "minimum": 0.9038623468605168, "maximum": 0.9315828203868873, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.138340364289892, "standard_deviation": 0.023343393699381394, "minimum": 2.114447358776923, "maximum": 2.1893936748705465, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5925029026764426, "standard_deviation": 0.17435262077086422, "minimum": 0.30796246060104826, "maximum": 0.8562547964953223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3683603204394742, "standard_deviation": 0.0198672151334775, "minimum": 0.34425378866919853, "maximum": 0.4021121672637087, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.939486896300469, "standard_deviation": 0.00802619018735072, "minimum": 0.9216390463914904, "maximum": 0.9460091367254271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5154566325285157, "standard_deviation": 0.025845969821721774, "minimum": 1.4784078498922215, "maximum": 1.5485889726135424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0444551860251707, "standard_deviation": 0.2389799417318654, "minimum": 0.742223543851725, "maximum": 1.4939030816750467, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.32120543650793193, "standard_deviation": 0.015188244654792902, "minimum": 0.30500998851972316, "maximum": 0.3480574867342073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9892601826842233, "standard_deviation": 0.0020273779805449576, "minimum": 0.9849208251116444, "maximum": 0.9915093661483715, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7772539346398917, "standard_deviation": 0.03747528675653611, "minimum": 1.705252020346266, "maximum": 1.8223483024205591, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48381564502239444, "standard_deviation": 0.14463703950997958, "minimum": 0.29857662746343266, "maximum": 0.6901384654944995, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11911346077832599, "standard_deviation": 0.010840642782998585, "minimum": 0.10486327589531012, "maximum": 0.14185567285283537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961903548180946, "standard_deviation": 0.0011752205313514952, "minimum": 0.9935716831661704, "maximum": 0.9974959930620767, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1107049065506858, "standard_deviation": 0.04791380383985257, "minimum": 1.038043356344761, "maximum": 1.1838450746274993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.52200720840711, "standard_deviation": 0.13129712981818073, "minimum": 0.3487554752525938, "maximum": 0.7701060257192779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07100909705827861, "standard_deviation": 0.010342311751908962, "minimum": 0.058734244009977336, "maximum": 0.09301315491447255, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.26614108671734765, "standard_deviation": 0.08436743737194327, "minimum": 0.141982376946501, "maximum": 0.36571522275064555, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.685837288704215, "standard_deviation": 0.009051830898839848, "minimum": 5.674303339578088, "maximum": 5.699546791037519, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6321141239416729, "standard_deviation": 0.06382355590710727, "minimum": 0.5801676316012763, "maximum": 0.737726944428828, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08830249058617035, "standard_deviation": 0.02881353996895415, "minimum": -0.14736028254115136, "maximum": -0.047647453618334466, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.995253652457317, "standard_deviation": 0.0005491539858874219, "minimum": 0.9943146804871005, "maximum": 0.99607075980598, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9630214560900234, "standard_deviation": 0.02605685142313324, "minimum": 0.9206569150498629, "maximum": 0.997604675320386, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2621399354563253, "standard_deviation": 0.0789176076414221, "minimum": 0.15503490853532342, "maximum": 0.36495096220511947, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06371247771565947, "standard_deviation": 0.003758757239526606, "minimum": 0.05810333941955748, "maximum": 0.07064155926356157, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985121470389504, "standard_deviation": 0.0003674558662011726, "minimum": 0.997866948108004, "maximum": 0.9989104219238922, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49975493146767386, "standard_deviation": 0.0229619254694589, "minimum": 0.4620332796553727, "maximum": 0.5318282966096101, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3138605210777582, "standard_deviation": 0.10048015118049522, "minimum": 0.15819843876021342, "maximum": 0.4617621032323198, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03474947436612648, "standard_deviation": 0.004162096081325046, "minimum": 0.03017045849282319, "maximum": 0.04310957850018778, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0019", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.663985013961792, "predicted_mos": 2.663985013961792, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "20bbe1671774b8e17708def943bb6dc0ec55ed568f0c2bbceaac41afff9044ca", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.347182999830693, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9602805588871965, "全帯域_10ms_rms_envelope_nrmse": 0.059539152776465556, "全帯域_10ms_envelope_correlation": 0.9958868160132353, "全帯域_10ms_modulation_mae_db": 0.14270363568447506, "全帯域_40ms_log_rms_mae_db": 0.49768208406808834, "全帯域_40ms_rms_envelope_nrmse": 0.03399698473000642, "全帯域_40ms_envelope_correlation": 0.9986032352536716, "全帯域_40ms_modulation_mae_db": 0.13579580290986212, "300–1000 Hz_10ms_log_rms_mae_db": 1.7397981813638146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13868393063014478, "300–1000 Hz_10ms_envelope_correlation": 0.9858311361910184, "300–1000 Hz_10ms_modulation_mae_db": 0.47609504965178456, "300–1000 Hz_40ms_log_rms_mae_db": 1.1088926487389548, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08168364892190909, "300–1000 Hz_40ms_envelope_correlation": 0.9955949357009513, "300–1000 Hz_40ms_modulation_mae_db": 0.2797642830584621, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0149823981674175, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17053818339292068, "1000–3000 Hz_10ms_envelope_correlation": 0.978592797733032, "1000–3000 Hz_10ms_modulation_mae_db": 0.6131904074743364, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2103093181330844, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1161613511773316, "1000–3000 Hz_40ms_envelope_correlation": 0.9898187265476779, "1000–3000 Hz_40ms_modulation_mae_db": 0.7603536458325868, "3000–6000 Hz_10ms_log_rms_mae_db": 2.123568097058498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3377853120212675, "3000–6000 Hz_10ms_envelope_correlation": 0.933580909426188, "3000–6000 Hz_10ms_modulation_mae_db": 0.34484342932548157, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4774190738239286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3038538750410972, "3000–6000 Hz_40ms_envelope_correlation": 0.9438667360249815, "3000–6000 Hz_40ms_modulation_mae_db": 0.9862394986864023, "mr_stft_db": 5.665969361552766, "band_mae_db": 0.23667231722391932, "rms_gain_error_db": -0.08482217499913133, "output_peak": 0.6290554203790134}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9535202356767546, "全帯域_10ms_rms_envelope_nrmse": 0.06722534175421561, "全帯域_10ms_envelope_correlation": 0.994656998210137, "全帯域_10ms_modulation_mae_db": 0.1843769571577644, "全帯域_40ms_log_rms_mae_db": 0.5245206409934805, "全帯域_40ms_rms_envelope_nrmse": 0.03679249435885932, "全帯域_40ms_envelope_correlation": 0.9982670808305993, "全帯域_40ms_modulation_mae_db": 0.2953798526524566, "300–1000 Hz_10ms_log_rms_mae_db": 1.7533782500079567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12172833120586155, "300–1000 Hz_10ms_envelope_correlation": 0.988658997616478, "300–1000 Hz_10ms_modulation_mae_db": 0.5972880534352698, "300–1000 Hz_40ms_log_rms_mae_db": 1.060041309113671, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07386486073355507, "300–1000 Hz_40ms_envelope_correlation": 0.9956967799449568, "300–1000 Hz_40ms_modulation_mae_db": 0.4577006945886087, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9394986292638245, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20687328196087468, "1000–3000 Hz_10ms_envelope_correlation": 0.9685124624394149, "1000–3000 Hz_10ms_modulation_mae_db": 1.0495405237754285, "1000–3000 Hz_40ms_log_rms_mae_db": 1.189965487769156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1362470154677637, "1000–3000 Hz_40ms_envelope_correlation": 0.9864927283029269, "1000–3000 Hz_40ms_modulation_mae_db": 0.717217930308674, "3000–6000 Hz_10ms_log_rms_mae_db": 2.163793165184142, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3707475962358548, "3000–6000 Hz_10ms_envelope_correlation": 0.9168810367681014, "3000–6000 Hz_10ms_modulation_mae_db": 0.38859452421863394, "3000–6000 Hz_40ms_log_rms_mae_db": 1.553818331084822, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31017099559075706, "3000–6000 Hz_40ms_envelope_correlation": 0.938966296373412, "3000–6000 Hz_40ms_modulation_mae_db": 0.9420765800894907, "mr_stft_db": 5.662796984835768, "band_mae_db": 0.17690549117739493, "rms_gain_error_db": -0.061590449919474016, "output_peak": 0.6180601866583181}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9624493423867163, "全帯域_10ms_rms_envelope_nrmse": 0.06613322196949409, "全帯域_10ms_envelope_correlation": 0.9948426353049575, "全帯域_10ms_modulation_mae_db": 0.06531621331690103, "全帯域_40ms_log_rms_mae_db": 0.507514489807584, "全帯域_40ms_rms_envelope_nrmse": 0.035152069877279216, "全帯域_40ms_envelope_correlation": 0.9984322559868744, "全帯域_40ms_modulation_mae_db": 0.3041078545381008, "300–1000 Hz_10ms_log_rms_mae_db": 1.7289667172395462, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13236442516907562, "300–1000 Hz_10ms_envelope_correlation": 0.9866120931694918, "300–1000 Hz_10ms_modulation_mae_db": 0.3728617264951997, "300–1000 Hz_40ms_log_rms_mae_db": 1.0365338284238388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07939800424835163, "300–1000 Hz_40ms_envelope_correlation": 0.9950840582178825, "300–1000 Hz_40ms_modulation_mae_db": 0.4272030352058656, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0174003102310714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1513056369821685, "1000–3000 Hz_10ms_envelope_correlation": 0.9840692052051451, "1000–3000 Hz_10ms_modulation_mae_db": 0.9994435030149798, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2007345751491358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09328168664376797, "1000–3000 Hz_40ms_envelope_correlation": 0.9947342385790288, "1000–3000 Hz_40ms_modulation_mae_db": 1.01825875689852, "3000–6000 Hz_10ms_log_rms_mae_db": 2.070051986113004, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3700422273833216, "3000–6000 Hz_10ms_envelope_correlation": 0.9128203201537155, "3000–6000 Hz_10ms_modulation_mae_db": 0.3829311650118202, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4731363716588939, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3196115218448092, "3000–6000 Hz_40ms_envelope_correlation": 0.9316937900149179, "3000–6000 Hz_40ms_modulation_mae_db": 0.901252402089601, "mr_stft_db": 5.632793675714859, "band_mae_db": 0.1913506784268025, "rms_gain_error_db": -0.06878294402141136, "output_peak": 0.5945574085326286}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9540840587073779, "全帯域_10ms_rms_envelope_nrmse": 0.05541948308613692, "全帯域_10ms_envelope_correlation": 0.9964131507510195, "全帯域_10ms_modulation_mae_db": 0.19974162979073404, "全帯域_40ms_log_rms_mae_db": 0.46644812855104273, "全帯域_40ms_rms_envelope_nrmse": 0.028899643545528687, "全帯域_40ms_envelope_correlation": 0.998984056763079, "全帯域_40ms_modulation_mae_db": 0.2674662944961892, "300–1000 Hz_10ms_log_rms_mae_db": 1.7375082531693018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08812190088513532, "300–1000 Hz_10ms_envelope_correlation": 0.9941325332642695, "300–1000 Hz_10ms_modulation_mae_db": 0.2560480704986086, "300–1000 Hz_40ms_log_rms_mae_db": 0.980945433430283, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0467045893174167, "300–1000 Hz_40ms_envelope_correlation": 0.9983427102659479, "300–1000 Hz_40ms_modulation_mae_db": 0.2872405681039961, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9805647305671423, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16345873933719826, "1000–3000 Hz_10ms_envelope_correlation": 0.9815898977785578, "1000–3000 Hz_10ms_modulation_mae_db": 1.5126833065948049, "1000–3000 Hz_40ms_log_rms_mae_db": 1.204005374844799, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09915770301627418, "1000–3000 Hz_40ms_envelope_correlation": 0.9944591768165207, "1000–3000 Hz_40ms_modulation_mae_db": 1.5117495911946597, "3000–6000 Hz_10ms_log_rms_mae_db": 2.095584409777628, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3499887338778662, "3000–6000 Hz_10ms_envelope_correlation": 0.9220965772298659, "3000–6000 Hz_10ms_modulation_mae_db": 0.5473491397279177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.428921420831562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31075688562154846, "3000–6000 Hz_40ms_envelope_correlation": 0.9354698954903149, "3000–6000 Hz_40ms_modulation_mae_db": 0.8207648074821128, "mr_stft_db": 5.654403614955908, "band_mae_db": 0.18553490420850274, "rms_gain_error_db": -0.07611500095286217, "output_peak": 0.6240985256156997}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.955657337736162, "全帯域_10ms_rms_envelope_nrmse": 0.06894338918605482, "全帯域_10ms_envelope_correlation": 0.9943728349697605, "全帯域_10ms_modulation_mae_db": 0.22620285892122574, "全帯域_40ms_log_rms_mae_db": 0.46199916940625313, "全帯域_40ms_rms_envelope_nrmse": 0.035385857163375924, "全帯域_40ms_envelope_correlation": 0.9983904882701992, "全帯域_40ms_modulation_mae_db": 0.19033637798540112, "300–1000 Hz_10ms_log_rms_mae_db": 1.7553704357606117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14004910734267959, "300–1000 Hz_10ms_envelope_correlation": 0.9849499640850461, "300–1000 Hz_10ms_modulation_mae_db": 0.4232140166338291, "300–1000 Hz_40ms_log_rms_mae_db": 1.059597810357831, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07228329751702545, "300–1000 Hz_40ms_envelope_correlation": 0.9960582359202588, "300–1000 Hz_40ms_modulation_mae_db": 0.21347786964429563, "1000–3000 Hz_10ms_log_rms_mae_db": 2.046336811864869, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15884907023269876, "1000–3000 Hz_10ms_envelope_correlation": 0.9820523108325736, "1000–3000 Hz_10ms_modulation_mae_db": 0.5455710953860478, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2081985088250138, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09978859614872938, "1000–3000 Hz_40ms_envelope_correlation": 0.9934311469600995, "1000–3000 Hz_40ms_modulation_mae_db": 0.5684204462699701, "3000–6000 Hz_10ms_log_rms_mae_db": 2.120587284351325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36705890755967735, "3000–6000 Hz_10ms_envelope_correlation": 0.9163035198764117, "3000–6000 Hz_10ms_modulation_mae_db": 0.5708661557186899, "3000–6000 Hz_40ms_log_rms_mae_db": 1.469267694998011, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31341213664401485, "3000–6000 Hz_40ms_envelope_correlation": 0.935913738107055, "3000–6000 Hz_40ms_modulation_mae_db": 0.9692047179743817, "mr_stft_db": 5.659086770349582, "band_mae_db": 0.19227561501815613, "rms_gain_error_db": -0.05936678789096868, "output_peak": 0.5926737634956795}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9894442416461227, "全帯域_10ms_rms_envelope_nrmse": 0.06148114941936142, "全帯域_10ms_envelope_correlation": 0.9956768467701204, "全帯域_10ms_modulation_mae_db": 0.2061368025365992, "全帯域_40ms_log_rms_mae_db": 0.5178497760002504, "全帯域_40ms_rms_envelope_nrmse": 0.036044691833243, "全帯域_40ms_envelope_correlation": 0.998499447619699, "全帯域_40ms_modulation_mae_db": 0.41075359482039575, "300–1000 Hz_10ms_log_rms_mae_db": 1.7804752368021712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12269766606489525, "300–1000 Hz_10ms_envelope_correlation": 0.9890851556659737, "300–1000 Hz_10ms_modulation_mae_db": 0.6513872332046791, "300–1000 Hz_40ms_log_rms_mae_db": 1.0655963047871886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07798533264012976, "300–1000 Hz_40ms_envelope_correlation": 0.9959223637769911, "300–1000 Hz_40ms_modulation_mae_db": 0.7483521236986149, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0313060504362763, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15531218885062512, "1000–3000 Hz_10ms_envelope_correlation": 0.9831604049309053, "1000–3000 Hz_10ms_modulation_mae_db": 0.6439380686914516, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1699308792421703, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09487457254891472, "1000–3000 Hz_40ms_envelope_correlation": 0.994556242573739, "1000–3000 Hz_40ms_modulation_mae_db": 0.6056097709370433, "3000–6000 Hz_10ms_log_rms_mae_db": 2.145519466843303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36041608406042663, "3000–6000 Hz_10ms_envelope_correlation": 0.9237634024525181, "3000–6000 Hz_10ms_modulation_mae_db": 0.5235938389176961, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5338488679258193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3069201409093043, "3000–6000 Hz_40ms_envelope_correlation": 0.9422971185940563, "3000–6000 Hz_40ms_modulation_mae_db": 0.7876756923828067, "mr_stft_db": 5.6564515785026055, "band_mae_db": 0.27285535187854365, "rms_gain_error_db": -0.11508971574781893, "output_peak": 0.6182988894695373}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9463495409483742, "全帯域_10ms_rms_envelope_nrmse": 0.059724348314420694, "全帯域_10ms_envelope_correlation": 0.9958020278208796, "全帯域_10ms_modulation_mae_db": 0.3713268150736038, "全帯域_40ms_log_rms_mae_db": 0.5062438775513713, "全帯域_40ms_rms_envelope_nrmse": 0.034057619425814, "全帯域_40ms_envelope_correlation": 0.9985269040903271, "全帯域_40ms_modulation_mae_db": 0.3351666216621673, "300–1000 Hz_10ms_log_rms_mae_db": 1.7506141083798303, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11271266785239165, "300–1000 Hz_10ms_envelope_correlation": 0.9903279722440126, "300–1000 Hz_10ms_modulation_mae_db": 0.4397952397468646, "300–1000 Hz_40ms_log_rms_mae_db": 1.0600377626310127, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06258560051784282, "300–1000 Hz_40ms_envelope_correlation": 0.9970064177874862, "300–1000 Hz_40ms_modulation_mae_db": 0.4936207437888746, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9635731500154352, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15454673728019824, "1000–3000 Hz_10ms_envelope_correlation": 0.9831379063498781, "1000–3000 Hz_10ms_modulation_mae_db": 1.004452594129417, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1724149536448047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09992469312710617, "1000–3000 Hz_40ms_envelope_correlation": 0.9934576150517807, "1000–3000 Hz_40ms_modulation_mae_db": 1.0836163749758123, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1066500745882784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3561441166926036, "3000–6000 Hz_10ms_envelope_correlation": 0.9232617128280353, "3000–6000 Hz_10ms_modulation_mae_db": 0.40802676038153135, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4539888991725922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31106162423214334, "3000–6000 Hz_40ms_envelope_correlation": 0.9386356340998888, "3000–6000 Hz_40ms_modulation_mae_db": 0.7076922818178785, "mr_stft_db": 5.649027733264379, "band_mae_db": 0.21295573643859025, "rms_gain_error_db": -0.06373636223010433, "output_peak": 0.6085026497174667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0610270882931958, "全帯域_10ms_rms_envelope_nrmse": 0.06737153705381349, "全帯域_10ms_envelope_correlation": 0.994770630908757, "全帯域_10ms_modulation_mae_db": 0.1768656811868652, "全帯域_40ms_log_rms_mae_db": 0.5829456643082459, "全帯域_40ms_rms_envelope_nrmse": 0.0369415012140375, "全帯域_40ms_envelope_correlation": 0.9984255840605388, "全帯域_40ms_modulation_mae_db": 0.14872297217350852, "300–1000 Hz_10ms_log_rms_mae_db": 1.7104349238298497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11131780214827353, "300–1000 Hz_10ms_envelope_correlation": 0.9906480123781529, "300–1000 Hz_10ms_modulation_mae_db": 0.6256053849859629, "300–1000 Hz_40ms_log_rms_mae_db": 1.1098604436870738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06903922410878799, "300–1000 Hz_40ms_envelope_correlation": 0.9963642720330524, "300–1000 Hz_40ms_modulation_mae_db": 0.6811274591532227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.970953889954841, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15685682847034094, "1000–3000 Hz_10ms_envelope_correlation": 0.9841425323029319, "1000–3000 Hz_10ms_modulation_mae_db": 0.7790429801191315, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1908050021664034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10183222494199284, "1000–3000 Hz_40ms_envelope_correlation": 0.9950956588840105, "1000–3000 Hz_40ms_modulation_mae_db": 0.9210241129745573, "3000–6000 Hz_10ms_log_rms_mae_db": 2.100878324364864, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4009479996976573, "3000–6000 Hz_10ms_envelope_correlation": 0.9041693876552921, "3000–6000 Hz_10ms_modulation_mae_db": 0.4941054470992487, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4781851269495652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3118700371942047, "3000–6000 Hz_40ms_envelope_correlation": 0.9395206636919484, "3000–6000 Hz_40ms_modulation_mae_db": 0.7450085796688377, "mr_stft_db": 5.643672200649685, "band_mae_db": 0.2979021149573016, "rms_gain_error_db": -0.12239506999005309, "output_peak": 0.5952661976590605}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.91747001059524, "spectral_convergence": 0.1980356968012186, "spectral_convergence_status": "ok", "amplitude_floor": 1.884454455950008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.514304602481591, "spectral_convergence": 0.1978995016669855, "spectral_convergence_status": "ok", "amplitude_floor": 1.301578596614512e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.566133471581468, "spectral_convergence": 0.19943701034581435, "spectral_convergence_status": "ok", "amplitude_floor": 1.1393421242566376e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004137534659655858, "output_power": 0.004137365878237009, "delta_db": -0.00017716427775404898, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026137240204796635, "output_power": 0.002535595572555523, "delta_db": -0.13179744337006383, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008071793462917449, "output_power": 0.0007530727573818133, "delta_db": -0.3013310369651264, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004252507832785635, "output_power": 0.00039841544694352, "delta_db": -0.2830895413598388, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.026398390024539e-05, "output_power": 6.678532896537949e-05, "delta_db": -0.22051700709128816, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.066875275629503e-05, "output_power": 5.663075934801798e-05, "delta_db": 0.48312171027944467, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23667231722391932, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08482217499913133, "pitch": null, "output_rms_dbfs": -20.99749477669347, "output_peak": 0.6290554203790134, "reference_rms_dbfs": -20.91267260169434, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.665969361552766}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025953969251226393, "log_rms_mae_db": 0.9602805588871965, "rms_envelope_nrmse": 0.059539152776465556, "envelope_correlation": 0.9958868160132353, "rms_gain_error_db": -0.0846116233912636, "reference_mean_power": 0.008109529837298932, "output_mean_power": 0.007953064666060463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5635786081597607e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013500334014131179, "output_power": 0.0012775790060825193, "reference_active": true, "delta_db": -0.23956746973024523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020457869490256438, "output_power": 0.00019962909506379982, "reference_active": true, "delta_db": -0.10636565648312912}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3880284077345985e-05, "output_power": 3.324522411826676e-05, "reference_active": true, "delta_db": -0.08217778084005076}], "active_band_count": 3, "active_band_mae_db": 0.14270363568447506}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024049824298809385, "log_rms_mae_db": 0.49768208406808834, "rms_envelope_nrmse": 0.03399698473000642, "envelope_correlation": 0.9986032352536716, "rms_gain_error_db": -0.08463680213445914, "reference_mean_power": 0.008139876536629037, "output_mean_power": 0.007982779575282654}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2818780696338306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012006574574251973, "output_power": 0.001135139973711772, "reference_active": true, "delta_db": -0.24369705134712955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.159020687440259e-05, "output_power": 8.952189490993271e-05, "reference_active": true, "delta_db": -0.09919773629689889}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2353163446294293e-06, "output_power": 2.268758428493798e-06, "reference_active": true, "delta_db": 0.06449262108555799}], "active_band_count": 3, "active_band_mae_db": 0.13579580290986212}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002074995834794702, "log_rms_mae_db": 1.7397981813638146, "rms_envelope_nrmse": 0.13868393063014478, "envelope_correlation": 0.9858311361910184, "rms_gain_error_db": -0.3447983161588407, "reference_mean_power": 0.0020359851811281022, "output_mean_power": 0.0018805928520202794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.345568889858965e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045532158312677247, "output_power": 0.00042929564066110777, "reference_active": true, "delta_db": -0.25561758795937134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.561332252218219e-05, "output_power": 7.634237010650376e-05, "reference_active": true, "delta_db": -0.4977571286021543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0669601144719856e-05, "output_power": 9.133918696596428e-06, "reference_active": true, "delta_db": -0.674910432393828}], "active_band_count": 3, "active_band_mae_db": 0.47609504965178456}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001833599147850085, "log_rms_mae_db": 1.1088926487389548, "rms_envelope_nrmse": 0.08168364892190909, "envelope_correlation": 0.9955949357009513, "rms_gain_error_db": -0.34441290975105376, "reference_mean_power": 0.0020457391187242045, "output_mean_power": 0.0018897700372611113}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.75393473065512e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00043556686984963635, "output_power": 0.00040895446613850583, "reference_active": true, "delta_db": -0.27379883543862915}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.69121464262054e-05, "output_power": 3.52758673783963e-05, "reference_active": true, "delta_db": -0.1969159879668442}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.183033529622033e-07, "output_power": 6.598573994318757e-07, "reference_active": true, "delta_db": -0.36857802576991305}], "active_band_count": 3, "active_band_mae_db": 0.2797642830584621}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011093565856595723, "log_rms_mae_db": 2.0149823981674175, "rms_envelope_nrmse": 0.17053818339292068, "envelope_correlation": 0.978592797733032, "rms_gain_error_db": -0.3054225343256446, "reference_mean_power": 0.00042557205365836593, "output_mean_power": 0.00039667136226669594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.99964400169894e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011758808293094384, "output_power": 0.0001041224938736725, "reference_active": true, "delta_db": -0.5281874851066671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1475307548341578e-05, "output_power": 1.93390840922089e-05, "reference_active": true, "delta_db": -0.45503490237981364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0561557218273926e-06, "output_power": 4.151313746672395e-06, "reference_active": true, "delta_db": -0.8563488349365285}], "active_band_count": 3, "active_band_mae_db": 0.6131904074743364}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010031642073947198, "log_rms_mae_db": 1.2103093181330844, "rms_envelope_nrmse": 0.1161613511773316, "envelope_correlation": 0.9898187265476779, "rms_gain_error_db": -0.3055548316116762, "reference_mean_power": 0.0004276640413209328, "output_mean_power": 0.00039860913974270717}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8247169971067077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011139646210262638, "output_power": 9.953239371664178e-05, "reference_active": true, "delta_db": -0.4890694926155837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0378224135538677e-05, "output_power": 9.73745894718831e-06, "reference_active": true, "delta_db": -0.27677406080912625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.220864105203889e-07, "output_power": 2.2722214904190114e-07, "reference_active": true, "delta_db": -1.5152173840730505}], "active_band_count": 3, "active_band_mae_db": 0.7603536458325868}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.985234496629778e-05, "log_rms_mae_db": 2.123568097058498, "rms_envelope_nrmse": 0.3377853120212675, "envelope_correlation": 0.933580909426188, "rms_gain_error_db": 0.4265275072645375, "reference_mean_power": 6.361685525644655e-05, "output_mean_power": 7.01818719243739e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.697638884988847e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.1912782580556425e-06, "output_power": 6.3020840523748e-06, "reference_active": true, "delta_db": 0.07703867974565905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.016318108845029e-06, "output_power": 1.985676139290292e-06, "reference_active": true, "delta_db": -0.06650633423373413}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.069070181696055e-07, "output_power": 7.451105061161807e-07, "reference_active": true, "delta_db": 0.8909852739970515}], "active_band_count": 3, "active_band_mae_db": 0.34484342932548157}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.357134521214995e-05, "log_rms_mae_db": 1.4774190738239286, "rms_envelope_nrmse": 0.3038538750410972, "envelope_correlation": 0.9438667360249815, "rms_gain_error_db": 0.42663994127088223, "reference_mean_power": 6.39232761364752e-05, "output_mean_power": 7.052173997546526e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.609918878818377e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.077093160223491e-06, "output_power": 6.324890387332729e-06, "reference_active": true, "delta_db": 0.17357109413251925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.511523340291838e-07, "output_power": 1.055857480048382e-06, "reference_active": true, "delta_db": 0.4535522299997331}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7938163224025654e-08, "output_power": 4.779223794540823e-08, "reference_active": true, "delta_db": 2.3315951719269545}], "active_band_count": 3, "active_band_mae_db": 0.9862394986864023}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9018884172975314, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "71c1b47caee38c3cb1df481e0f1a357c0eb912df67c63a6a30b4b456e2a65a69", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9806571896965548, "standard_deviation": 0.005220590090046227, "minimum": 0.9685124624394149, "maximum": 0.9841425323029319, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9955769963126095, "standard_deviation": 0.03723793333453618, "minimum": 1.9394986292638245, "maximum": 2.046336811864869, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8934828098981997, "standard_deviation": 0.31764325702252655, "minimum": 0.5455710953860478, "maximum": 1.5126833065948049, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16471758331337813, "standard_deviation": 0.01804987444674655, "minimum": 0.1513056369821685, "maximum": 0.20687328196087468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.992755691714473, "standard_deviation": 0.0030315802146521713, "minimum": 0.9864927283029269, "maximum": 0.9950956588840105, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.193295512471821, "standard_deviation": 0.015495645553942833, "minimum": 1.1699308792421703, "maximum": 1.2103093181330844, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8982813286739779, "standard_deviation": 0.30981032178670403, "minimum": 0.5684204462699701, "maximum": 1.5117495911946597, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10515848038398506, "standard_deviation": 0.014331178095780358, "minimum": 0.09328168664376797, "maximum": 0.1362470154677637, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.919109608298766, "standard_deviation": 0.00872471295730756, "minimum": 0.9041693876552921, "maximum": 0.933580909426188, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.11582910103513, "standard_deviation": 0.029466139559295508, "minimum": 2.070051986113004, "maximum": 2.163793165184142, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.45753880755012744, "standard_deviation": 0.08625599401360695, "minimum": 0.34484342932548157, "maximum": 0.5708661557186899, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3641413721910844, "standard_deviation": 0.018572156868465525, "minimum": 0.3377853120212675, "maximum": 0.4009479996976573, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9382954840495719, "standard_deviation": 0.003897942910888878, "minimum": 0.9316937900149179, "maximum": 0.9438667360249815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4835732233056493, "standard_deviation": 0.040891054745856595, "minimum": 1.428921420831562, "maximum": 1.553818331084822, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.857489320023939, "standard_deviation": 0.10653024560063026, "minimum": 0.7076922818178785, "maximum": 0.9862394986864023, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.31095715213473485, "standard_deviation": 0.004616453361333781, "minimum": 0.3038538750410972, "maximum": 0.3196115218448092, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9887807330768055, "standard_deviation": 0.002993035984125275, "minimum": 0.9849499640850461, "maximum": 0.9941325332642695, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7445682633191355, "standard_deviation": 0.020706301014986335, "minimum": 1.7104349238298497, "maximum": 1.7804752368021712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4802868468315248, "standard_deviation": 0.1367060123771053, "minimum": 0.2560480704986086, "maximum": 0.6513872332046791, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12095947891230716, "standard_deviation": 0.017119469634745418, "minimum": 0.08812190088513532, "maximum": 0.14004910734267959, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962587217059409, "standard_deviation": 0.0010143790880643097, "minimum": 0.9950840582178825, "maximum": 0.9983427102659479, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0601881926462315, "standard_deviation": 0.0408971546928776, "minimum": 0.980945433430283, "maximum": 1.1098604436870738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4485608471552426, "standard_deviation": 0.19135331739040462, "minimum": 0.21347786964429563, "maximum": 0.7483521236986149, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07044306975062731, "standard_deviation": 0.01137508776625488, "minimum": 0.0467045893174167, "maximum": 0.08168364892190909, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22080652616615137, "standard_deviation": 0.04443167521987961, "minimum": 0.17690549117739493, "maximum": 0.2979021149573016, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.653025239978194, "standard_deviation": 0.01086406094656156, "minimum": 5.632793675714859, "maximum": 5.665969361552766, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6100641301909255, "standard_deviation": 0.014410719404185287, "minimum": 0.5926737634956795, "maximum": 0.6290554203790134, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08148731321897798, "standard_deviation": 0.024507285009018404, "minimum": -0.12239506999005309, "maximum": -0.05936678789096868, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9953027425936083, "standard_deviation": 0.0007309429593257527, "minimum": 0.9943728349697605, "maximum": 0.9964131507510195, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9728515505352375, "standard_deviation": 0.03787270893056544, "minimum": 0.9463495409483742, "maximum": 1.0610270882931958, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.19658382420852105, "standard_deviation": 0.08634292456479768, "minimum": 0.06531621331690103, "maximum": 0.3713268150736038, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06322970294499533, "standard_deviation": 0.0048426851727093765, "minimum": 0.05541948308613692, "maximum": 0.06894338918605482, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985161316093736, "standard_deviation": 0.00021380295130735664, "minimum": 0.9982670808305993, "maximum": 0.998984056763079, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5081504788357896, "standard_deviation": 0.03768105743371307, "minimum": 0.46199916940625313, "maximum": 0.5829456643082459, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2609661714047602, "standard_deviation": 0.09583055101433136, "minimum": 0.13579580290986212, "maximum": 0.41075359482039575, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03465885776851801, "standard_deviation": 0.002575247574463183, "minimum": 0.028899643545528687, "maximum": 0.0369415012140375, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1320-122617-0023", "language": "en", "style": "normal", "speaker": "librispeech_1320", "source_utterance_id": "1320-122617-0023", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1320-122617-0023.wav", "source_file_sha256": "3c1bade91baf1e4f0fca6c970a7f871aad72072eb56eb9457f5732cb70698c35", "decoded_pcm_sha256": "b3661757a6ed833feab8313fa697ecafb680dc14c418cd468619e460e550745e", "original_source_sha256": "553fc6594343774831ac876fee34c4ec2b47b53d451f313b10e64eea5cadd8c2", "corpus_entry_id": "librispeech_1320-122617-0023"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.9725382328033447, "predicted_mos": 3.9725382328033447, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bd37eebbe4307f494c9d2c5714d3b9fe2029626edf1ed65d9be6cf7f295fe0d8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.365234932000021, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1320-122617-0023", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "bd37eebbe4307f494c9d2c5714d3b9fe2029626edf1ed65d9be6cf7f295fe0d8", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0723c442122070015a9e67ddc186a8789762dfbf2d067e5f42b899767c84db30"}, "rows": [{"case": "librispeech_1320-122617-0023", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.0592904090881348, "predicted_mos": 3.0592904090881348, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "222387c70ed627a716eef22d4a00719fbca5ac32d23f139059d171bb621e3196", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.996787366999342, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1320-122617-0023", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "222387c70ed627a716eef22d4a00719fbca5ac32d23f139059d171bb621e3196", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9080525001731538, "全帯域_10ms_rms_envelope_nrmse": 0.0538404602747757, "全帯域_10ms_envelope_correlation": 0.9965521501030369, "全帯域_10ms_modulation_mae_db": 0.3024570740537489, "全帯域_40ms_log_rms_mae_db": 0.515873270572206, "全帯域_40ms_rms_envelope_nrmse": 0.02976280305682023, "全帯域_40ms_envelope_correlation": 0.9989035132919077, "全帯域_40ms_modulation_mae_db": 0.48712898299186896, "300–1000 Hz_10ms_log_rms_mae_db": 1.5555580545692516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07377002934922665, "300–1000 Hz_10ms_envelope_correlation": 0.9954965052616316, "300–1000 Hz_10ms_modulation_mae_db": 0.8667227999810446, "300–1000 Hz_40ms_log_rms_mae_db": 0.9636409872730838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04661141231361519, "300–1000 Hz_40ms_envelope_correlation": 0.9982676678418103, "300–1000 Hz_40ms_modulation_mae_db": 0.8173855894183463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6571189433335478, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15788362054447888, "1000–3000 Hz_10ms_envelope_correlation": 0.9831178537638243, "1000–3000 Hz_10ms_modulation_mae_db": 1.0763470226297291, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0637587388652108, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09925389237878428, "1000–3000 Hz_40ms_envelope_correlation": 0.9945695784616079, "1000–3000 Hz_40ms_modulation_mae_db": 0.8232314494751353, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0562813663230224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1787786053430546, "3000–6000 Hz_10ms_envelope_correlation": 0.9795387389368905, "3000–6000 Hz_10ms_modulation_mae_db": 0.3013982955408397, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2961739915926882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10709010222897451, "3000–6000 Hz_40ms_envelope_correlation": 0.9923816936008422, "3000–6000 Hz_40ms_modulation_mae_db": 0.5608443996031118, "mr_stft_db": 5.467590349097283, "band_mae_db": 0.15559161866796936, "rms_gain_error_db": -0.06613005928186223, "output_peak": 0.562089091132391}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8464370691029531, "全帯域_10ms_rms_envelope_nrmse": 0.05130559397573385, "全帯域_10ms_envelope_correlation": 0.996852376150853, "全帯域_10ms_modulation_mae_db": 0.18262243526390434, "全帯域_40ms_log_rms_mae_db": 0.427885099073483, "全帯域_40ms_rms_envelope_nrmse": 0.026261556715365286, "全帯域_40ms_envelope_correlation": 0.9991321995047435, "全帯域_40ms_modulation_mae_db": 0.19456090883444302, "300–1000 Hz_10ms_log_rms_mae_db": 1.5890889451433228, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08108338743701327, "300–1000 Hz_10ms_envelope_correlation": 0.9944672519161106, "300–1000 Hz_10ms_modulation_mae_db": 0.8322351036380651, "300–1000 Hz_40ms_log_rms_mae_db": 0.9501565913196828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045944050737333694, "300–1000 Hz_40ms_envelope_correlation": 0.9982920257328796, "300–1000 Hz_40ms_modulation_mae_db": 0.8869154912197682, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7147516224679378, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15176504799580695, "1000–3000 Hz_10ms_envelope_correlation": 0.9840724116987063, "1000–3000 Hz_10ms_modulation_mae_db": 1.3569359320068326, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0907713538062975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08895480902432021, "1000–3000 Hz_40ms_envelope_correlation": 0.9953597395611127, "1000–3000 Hz_40ms_modulation_mae_db": 1.128807353155245, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1062616788796564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17828879868165665, "3000–6000 Hz_10ms_envelope_correlation": 0.980812959620598, "3000–6000 Hz_10ms_modulation_mae_db": 0.7899301993416223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.335913074864943, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10518046971513557, "3000–6000 Hz_40ms_envelope_correlation": 0.9948094251308708, "3000–6000 Hz_40ms_modulation_mae_db": 0.740135336719367, "mr_stft_db": 5.484298482468361, "band_mae_db": 0.3161345894105298, "rms_gain_error_db": -0.0498886893906719, "output_peak": 0.6222086510365292}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8852497842781721, "全帯域_10ms_rms_envelope_nrmse": 0.055508974783226986, "全帯域_10ms_envelope_correlation": 0.9963095985965619, "全帯域_10ms_modulation_mae_db": 0.2231471576540279, "全帯域_40ms_log_rms_mae_db": 0.460383814327909, "全帯域_40ms_rms_envelope_nrmse": 0.029383303658051953, "全帯域_40ms_envelope_correlation": 0.9989032937073312, "全帯域_40ms_modulation_mae_db": 0.1456626587521607, "300–1000 Hz_10ms_log_rms_mae_db": 1.551735198984268, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0787255869831683, "300–1000 Hz_10ms_envelope_correlation": 0.9947570528333258, "300–1000 Hz_10ms_modulation_mae_db": 0.7953795088699319, "300–1000 Hz_40ms_log_rms_mae_db": 0.9203787452098015, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04406367695814218, "300–1000 Hz_40ms_envelope_correlation": 0.9984029583704412, "300–1000 Hz_40ms_modulation_mae_db": 0.7790400450296909, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7068168322922628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1614380009338258, "1000–3000 Hz_10ms_envelope_correlation": 0.9816517071040193, "1000–3000 Hz_10ms_modulation_mae_db": 1.1389511040102844, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1173144144023313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10457792615711121, "1000–3000 Hz_40ms_envelope_correlation": 0.9926890394361441, "1000–3000 Hz_40ms_modulation_mae_db": 1.0162149804568326, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0809882190318008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16291536054388292, "3000–6000 Hz_10ms_envelope_correlation": 0.9829214796765993, "3000–6000 Hz_10ms_modulation_mae_db": 0.6815477907635805, "3000–6000 Hz_40ms_log_rms_mae_db": 1.254122759313664, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08589175990744662, "3000–6000 Hz_40ms_envelope_correlation": 0.9953938943152244, "3000–6000 Hz_40ms_modulation_mae_db": 0.7626523921166638, "mr_stft_db": 5.475243096023718, "band_mae_db": 0.2101995828867522, "rms_gain_error_db": -0.050003700052551676, "output_peak": 0.6043140779506916}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9177724538137695, "全帯域_10ms_rms_envelope_nrmse": 0.054428671064308105, "全帯域_10ms_envelope_correlation": 0.9964836723408282, "全帯域_10ms_modulation_mae_db": 0.2587084961754655, "全帯域_40ms_log_rms_mae_db": 0.4813569542319614, "全帯域_40ms_rms_envelope_nrmse": 0.02869225052581943, "全帯域_40ms_envelope_correlation": 0.9989954940757232, "全帯域_40ms_modulation_mae_db": 0.16344700093050832, "300–1000 Hz_10ms_log_rms_mae_db": 1.650066380797624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08907388773458562, "300–1000 Hz_10ms_envelope_correlation": 0.9933041642517441, "300–1000 Hz_10ms_modulation_mae_db": 0.7412057441026146, "300–1000 Hz_40ms_log_rms_mae_db": 0.9760471363969971, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05067548965598603, "300–1000 Hz_40ms_envelope_correlation": 0.9979133993433771, "300–1000 Hz_40ms_modulation_mae_db": 0.722463817446842, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7352063241159126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1787557984611866, "1000–3000 Hz_10ms_envelope_correlation": 0.9778718489796301, "1000–3000 Hz_10ms_modulation_mae_db": 0.9946413670210124, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1249543518090985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11437866950989964, "1000–3000 Hz_40ms_envelope_correlation": 0.9921598160055475, "1000–3000 Hz_40ms_modulation_mae_db": 0.7031558184500124, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0571793707722614, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17426067000331538, "3000–6000 Hz_10ms_envelope_correlation": 0.9804632434995211, "3000–6000 Hz_10ms_modulation_mae_db": 0.7002160006408223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2760898991362903, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09349415970614985, "3000–6000 Hz_40ms_envelope_correlation": 0.9941733404523657, "3000–6000 Hz_40ms_modulation_mae_db": 0.6056791915940772, "mr_stft_db": 5.467261626262581, "band_mae_db": 0.19687933036781455, "rms_gain_error_db": -0.06684724907856872, "output_peak": 0.6187846142897881}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8874730219529771, "全帯域_10ms_rms_envelope_nrmse": 0.04938743768755371, "全帯域_10ms_envelope_correlation": 0.9971992252520506, "全帯域_10ms_modulation_mae_db": 0.25227239676863883, "全帯域_40ms_log_rms_mae_db": 0.4472023527896046, "全帯域_40ms_rms_envelope_nrmse": 0.02586687669564595, "全帯域_40ms_envelope_correlation": 0.9992940432080359, "全帯域_40ms_modulation_mae_db": 0.2950852127595853, "300–1000 Hz_10ms_log_rms_mae_db": 1.6347667823406906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08174476208676028, "300–1000 Hz_10ms_envelope_correlation": 0.9945460503773659, "300–1000 Hz_10ms_modulation_mae_db": 0.7473914614473864, "300–1000 Hz_40ms_log_rms_mae_db": 0.9672990414597646, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04809928065021784, "300–1000 Hz_40ms_envelope_correlation": 0.9983166465146576, "300–1000 Hz_40ms_modulation_mae_db": 0.6291200563130296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6788094021544706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14705920519765667, "1000–3000 Hz_10ms_envelope_correlation": 0.9869913411366322, "1000–3000 Hz_10ms_modulation_mae_db": 0.7348243391956614, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0906186898026924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09588878025490027, "1000–3000 Hz_40ms_envelope_correlation": 0.9965045648912001, "1000–3000 Hz_40ms_modulation_mae_db": 0.7547510011058972, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1027294453496808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14728753811621711, "3000–6000 Hz_10ms_envelope_correlation": 0.985969092192323, "3000–6000 Hz_10ms_modulation_mae_db": 0.5811479169477618, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2902292518874283, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07974298920866452, "3000–6000 Hz_40ms_envelope_correlation": 0.9957777883101561, "3000–6000 Hz_40ms_modulation_mae_db": 0.6105113212456016, "mr_stft_db": 5.4693414638342075, "band_mae_db": 0.2487557966456051, "rms_gain_error_db": -0.1000470689510604, "output_peak": 0.512417857890608}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9170125774160471, "全帯域_10ms_rms_envelope_nrmse": 0.051401838401112426, "全帯域_10ms_envelope_correlation": 0.9968595142718304, "全帯域_10ms_modulation_mae_db": 0.31954044712518526, "全帯域_40ms_log_rms_mae_db": 0.4853487628646892, "全帯域_40ms_rms_envelope_nrmse": 0.026747291171376444, "全帯域_40ms_envelope_correlation": 0.9991195911927799, "全帯域_40ms_modulation_mae_db": 0.3249498532393439, "300–1000 Hz_10ms_log_rms_mae_db": 1.5996619621546797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08201691448684732, "300–1000 Hz_10ms_envelope_correlation": 0.9942739726384997, "300–1000 Hz_10ms_modulation_mae_db": 0.82570693645189, "300–1000 Hz_40ms_log_rms_mae_db": 0.937387768957318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046408083804395255, "300–1000 Hz_40ms_envelope_correlation": 0.9981830602695593, "300–1000 Hz_40ms_modulation_mae_db": 0.7552228596531224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7117300903435662, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14919316603818608, "1000–3000 Hz_10ms_envelope_correlation": 0.9862893259199319, "1000–3000 Hz_10ms_modulation_mae_db": 0.6132178388911127, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1021695722698635, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09960179924744186, "1000–3000 Hz_40ms_envelope_correlation": 0.9956748375690717, "1000–3000 Hz_40ms_modulation_mae_db": 0.6884112324529141, "3000–6000 Hz_10ms_log_rms_mae_db": 2.06570705688053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14451618918074755, "3000–6000 Hz_10ms_envelope_correlation": 0.9865633291184079, "3000–6000 Hz_10ms_modulation_mae_db": 0.6725621732460708, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2740188339570007, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07744729164446185, "3000–6000 Hz_40ms_envelope_correlation": 0.99618155202189, "3000–6000 Hz_40ms_modulation_mae_db": 0.9688603135076693, "mr_stft_db": 5.485470493437997, "band_mae_db": 0.25164694051287423, "rms_gain_error_db": -0.06427062970840672, "output_peak": 0.5318160995325633}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9301756794518775, "全帯域_10ms_rms_envelope_nrmse": 0.05132251742828616, "全帯域_10ms_envelope_correlation": 0.9969013103824809, "全帯域_10ms_modulation_mae_db": 0.415745964218278, "全帯域_40ms_log_rms_mae_db": 0.47240660360785647, "全帯域_40ms_rms_envelope_nrmse": 0.027422650803496318, "全帯域_40ms_envelope_correlation": 0.9991123382801054, "全帯域_40ms_modulation_mae_db": 0.5365821403281721, "300–1000 Hz_10ms_log_rms_mae_db": 1.6177676808418466, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08081009933903587, "300–1000 Hz_10ms_envelope_correlation": 0.9945650066218952, "300–1000 Hz_10ms_modulation_mae_db": 1.019295532558458, "300–1000 Hz_40ms_log_rms_mae_db": 0.9644838662539769, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04679654480391708, "300–1000 Hz_40ms_envelope_correlation": 0.9982960145727984, "300–1000 Hz_40ms_modulation_mae_db": 1.0938910745953174, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8101770607514203, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17880038947976704, "1000–3000 Hz_10ms_envelope_correlation": 0.9774890184394818, "1000–3000 Hz_10ms_modulation_mae_db": 0.8419561330185692, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1810394443753047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1084824403581081, "1000–3000 Hz_40ms_envelope_correlation": 0.9926037537176494, "1000–3000 Hz_40ms_modulation_mae_db": 0.8898992474869353, "3000–6000 Hz_10ms_log_rms_mae_db": 2.069332379416016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1960070892330658, "3000–6000 Hz_10ms_envelope_correlation": 0.9750800840341448, "3000–6000 Hz_10ms_modulation_mae_db": 1.229575547479311, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2521179228325356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10502269286736278, "3000–6000 Hz_40ms_envelope_correlation": 0.9927041049770924, "3000–6000 Hz_40ms_modulation_mae_db": 1.4135094070965633, "mr_stft_db": 5.483075229270064, "band_mae_db": 0.22757990487969157, "rms_gain_error_db": -0.08054812089158267, "output_peak": 0.5509801154822548}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9076869810255372, "全帯域_10ms_rms_envelope_nrmse": 0.04906361172998878, "全帯域_10ms_envelope_correlation": 0.9971543454888134, "全帯域_10ms_modulation_mae_db": 0.28847716528084866, "全帯域_40ms_log_rms_mae_db": 0.4747443027013056, "全帯域_40ms_rms_envelope_nrmse": 0.025370062771111552, "全帯域_40ms_envelope_correlation": 0.9992307188478522, "全帯域_40ms_modulation_mae_db": 0.23331219895691357, "300–1000 Hz_10ms_log_rms_mae_db": 1.584526632899253, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07354980836545805, "300–1000 Hz_10ms_envelope_correlation": 0.9954415287246636, "300–1000 Hz_10ms_modulation_mae_db": 0.7113395844247087, "300–1000 Hz_40ms_log_rms_mae_db": 0.9424795688548984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041283990534898604, "300–1000 Hz_40ms_envelope_correlation": 0.9986240920762569, "300–1000 Hz_40ms_modulation_mae_db": 0.6482754460507248, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6580722597538236, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18187543107386506, "1000–3000 Hz_10ms_envelope_correlation": 0.978655092840105, "1000–3000 Hz_10ms_modulation_mae_db": 1.3197161431066549, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0700115531915562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12454223155886193, "1000–3000 Hz_40ms_envelope_correlation": 0.9921411269874223, "1000–3000 Hz_40ms_modulation_mae_db": 1.2141771942223223, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1173550958717966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1805244257115774, "3000–6000 Hz_10ms_envelope_correlation": 0.9788584846973382, "3000–6000 Hz_10ms_modulation_mae_db": 0.2530884140572893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3008425804538715, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10893683803700453, "3000–6000 Hz_40ms_envelope_correlation": 0.9920285043437745, "3000–6000 Hz_40ms_modulation_mae_db": 0.16737787489015613, "mr_stft_db": 5.492302204362417, "band_mae_db": 0.24171592234623127, "rms_gain_error_db": -0.06887665427547231, "output_peak": 0.5534499159520366}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.729074303146154, "spectral_convergence": 0.16701614534804513, "spectral_convergence_status": "ok", "amplitude_floor": 1.2959192499433599e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.277522224595532, "spectral_convergence": 0.16405933931505579, "spectral_convergence_status": "ok", "amplitude_floor": 8.646137696849433e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.396174519550165, "spectral_convergence": 0.1656254704139092, "spectral_convergence_status": "ok", "amplitude_floor": 6.578152434885499e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0020500598531285057, "output_power": 0.002079000389686763, "delta_db": 0.060880299216201646, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0025325885847670413, "output_power": 0.0024630909827155715, "delta_db": -0.12084190738389401, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009745321457083826, "output_power": 0.0009546648125170296, "delta_db": -0.0894525380416989, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029510461183058167, "output_power": 0.00026756940185228295, "delta_db": -0.4253954882653257, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.807836063095196e-05, "output_power": 3.6277219377558456e-05, "delta_db": -0.21044251338807898, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.519436277144243e-05, "output_power": 3.5410071432057475e-05, "delta_db": 0.026536965712616906, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15559161866796936, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06613005928186223, "pitch": null, "output_rms_dbfs": -22.33883758475585, "output_peak": 0.562089091132391, "reference_rms_dbfs": -22.27270752547399, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.467590349097283}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019946925088635918, "log_rms_mae_db": 0.9080525001731538, "rms_envelope_nrmse": 0.0538404602747757, "envelope_correlation": 0.9965521501030369, "rms_gain_error_db": -0.06451248578507703, "reference_mean_power": 0.0059224241180442505, "output_mean_power": 0.005835099370516164}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0772791897966612e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010245127053592495, "output_power": 0.0010170406373080283, "reference_active": true, "delta_db": -0.031790424920105476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001329567656774176, "output_power": 0.00012191465428363438, "reference_active": true, "delta_db": -0.37654530408431175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.301896656350168e-05, "output_power": 2.0520232292996128e-05, "reference_active": true, "delta_db": -0.49903549315682943}], "active_band_count": 3, "active_band_mae_db": 0.3024570740537489}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001930447992601615, "log_rms_mae_db": 0.515873270572206, "rms_envelope_nrmse": 0.02976280305682023, "envelope_correlation": 0.9989035132919077, "rms_gain_error_db": -0.06520140307848486, "reference_mean_power": 0.005919423744261807, "output_mean_power": 0.005831218162459675}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.875987153432414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009473592447040891, "output_power": 0.0009413819197082165, "reference_active": true, "delta_db": -0.02748844550661903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.123049684235515e-05, "output_power": 5.5850508437990815e-05, "reference_active": true, "delta_db": -0.3994065235197135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.329909700774956e-06, "output_power": 1.0480281692461365e-06, "reference_active": true, "delta_db": -1.0344919799492742}], "active_band_count": 3, "active_band_mae_db": 0.48712898299186896}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015110036587973006, "log_rms_mae_db": 1.5555580545692516, "rms_envelope_nrmse": 0.07377002934922665, "envelope_correlation": 0.9954965052616316, "rms_gain_error_db": -0.1596782067948972, "reference_mean_power": 0.0025365732398871734, "output_mean_power": 0.002445004066540861}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.31579034521966e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006679887319597139, "output_power": 0.0006439749418285682, "reference_active": true, "delta_db": -0.15900168049703275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.085773760663629e-05, "output_power": 7.567558050698749e-05, "reference_active": true, "delta_db": -0.7940615739789472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6589878271689645e-05, "output_power": 1.1353576350667466e-05, "reference_active": true, "delta_db": -1.6471051454671537}], "active_band_count": 3, "active_band_mae_db": 0.8667227999810446}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013775289901134618, "log_rms_mae_db": 0.9636409872730838, "rms_envelope_nrmse": 0.04661141231361519, "envelope_correlation": 0.9982676678418103, "rms_gain_error_db": -0.16066448546287815, "reference_mean_power": 0.0025400163932975937, "output_mean_power": 0.002447766974940035}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2092894134690622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006291541070562532, "output_power": 0.0006056721145619018, "reference_active": true, "delta_db": -0.16519456985735193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.975083640469782e-05, "output_power": 3.3680379478107835e-05, "reference_active": true, "delta_db": -0.7196929509160238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.7844516380393e-07, "output_power": 6.820395610669006e-07, "reference_active": true, "delta_db": -1.5672692474816632}], "active_band_count": 3, "active_band_mae_db": 0.8173855894183463}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010600794217574433, "log_rms_mae_db": 1.6571189433335478, "rms_envelope_nrmse": 0.15788362054447888, "envelope_correlation": 0.9831178537638243, "rms_gain_error_db": -0.4862455628159157, "reference_mean_power": 0.0002691478602162077, "output_mean_power": 0.00024063920002084888}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.02821083589085e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010735336162950012, "output_power": 9.482432459979633e-05, "reference_active": true, "delta_db": -0.5389589028690847}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.315168535027072e-06, "output_power": 7.11118717926356e-06, "reference_active": true, "delta_db": -0.6792894588773656}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.408982675283079e-06, "output_power": 3.827480152245548e-06, "reference_active": true, "delta_db": 2.010792706142737}], "active_band_count": 3, "active_band_mae_db": 1.0763470226297291}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.461607013685591e-05, "log_rms_mae_db": 1.0637587388652108, "rms_envelope_nrmse": 0.09925389237878428, "envelope_correlation": 0.9945695784616079, "rms_gain_error_db": -0.4984217612858604, "reference_mean_power": 0.00026921330106685756, "output_mean_power": 0.00024002381663637405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9187018796507895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.751052870210862e-05, "output_power": 8.718265836208759e-05, "reference_active": true, "delta_db": -0.4862140385591974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.818200149529557e-06, "output_power": 3.3375091586742317e-06, "reference_active": true, "delta_db": -0.584362241619135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4850622077020166e-07, "output_power": 2.0495402563707795e-07, "reference_active": true, "delta_db": 1.3991180682470734}], "active_band_count": 3, "active_band_mae_db": 0.8232314494751353}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.633594187238404e-05, "log_rms_mae_db": 2.0562813663230224, "rms_envelope_nrmse": 0.1787786053430546, "envelope_correlation": 0.9795387389368905, "rms_gain_error_db": 0.02081239716809216, "reference_mean_power": 4.9508565949953196e-05, "output_mean_power": 4.974639186481615e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.164436282244953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0281496224399232e-05, "output_power": 9.908424163956443e-06, "reference_active": true, "delta_db": -0.1605173059981398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.187880808785332e-06, "output_power": 3.6822782742911926e-06, "reference_active": true, "delta_db": -0.5587770715825701}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.689088055617476e-07, "output_power": 5.451959298016784e-07, "reference_active": true, "delta_db": -0.18490050904180907}], "active_band_count": 3, "active_band_mae_db": 0.3013982955408397}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.3561893402191595e-05, "log_rms_mae_db": 1.2961739915926882, "rms_envelope_nrmse": 0.10709010222897451, "envelope_correlation": 0.9923816936008422, "rms_gain_error_db": 0.020870834979165663, "reference_mean_power": 4.9751584691373346e-05, "output_mean_power": 4.9991250671208105e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0334524009764534e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.035252748280077e-05, "output_power": 1.0079500316064671e-05, "reference_active": true, "delta_db": -0.11607389286806989}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2970706912154704e-06, "output_power": 1.9914435394858926e-06, "reference_active": true, "delta_db": -0.6200636260125606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.66930165214842e-08, "output_power": 3.7550209441729104e-08, "reference_active": true, "delta_db": -0.9463956799287049}], "active_band_count": 3, "active_band_mae_db": 0.5608443996031118}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9208403591029459, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "78b40e2b66b4a36254c949cea9bc5e8fe2353f960512e7e1d74a1d68bd089ec2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9820173249852914, "standard_deviation": 0.003733137067840035, "minimum": 0.9774890184394818, "maximum": 0.9869913411366322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7090853169016178, "standard_deviation": 0.04951992752618916, "minimum": 1.6571189433335478, "maximum": 1.8101770607514203, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0095737349849823, "standard_deviation": 0.266980429392736, "minimum": 0.6132178388911127, "maximum": 1.3569359320068326, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16334633246559666, "standard_deviation": 0.014406341614540687, "minimum": 0.14705920519765667, "maximum": 0.18187543107386506, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939628070787194, "standard_deviation": 0.0017627080731944357, "minimum": 0.9921411269874223, "maximum": 0.9965045648912001, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1050797648152941, "standard_deviation": 0.037175429609161516, "minimum": 1.0637587388652108, "maximum": 1.1810394443753047, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9023310346006618, "standard_deviation": 0.19825406201003912, "minimum": 0.6884112324529141, "maximum": 1.2141771942223223, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10446006856117843, "standard_deviation": 0.011227850503055541, "minimum": 0.08895480902432021, "maximum": 0.12454223155886193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9812759264719778, "standard_deviation": 0.003795022628820639, "minimum": 0.9750800840341448, "maximum": 0.9865633291184079, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0819793265655955, "standard_deviation": 0.023823105268669882, "minimum": 2.0562813663230224, "maximum": 2.1173550958717966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6511832922521622, "standard_deviation": 0.30315144746746314, "minimum": 0.2530884140572893, "maximum": 1.229575547479311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17032233460168966, "standard_deviation": 0.017589343727381353, "minimum": 0.14451618918074755, "maximum": 0.1960070892330658, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994181287894027, "standard_deviation": 0.0016243675713145363, "minimum": 0.9920285043437745, "maximum": 0.99618155202189, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2849385392548027, "standard_deviation": 0.027333685099904356, "minimum": 1.2521179228325356, "maximum": 1.335913074864943, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7286962795966513, "standard_deviation": 0.35862538070220984, "minimum": 0.16737787489015613, "maximum": 1.4135094070965633, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09535078791440002, "standard_deviation": 0.01292627162299213, "minimum": 0.07744729164446185, "maximum": 0.10893683803700453, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9946064415781546, "standard_deviation": 0.0006911137180113735, "minimum": 0.9933041642517441, "maximum": 0.9954965052616316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5978964547163672, "standard_deviation": 0.035170667382220654, "minimum": 1.551735198984268, "maximum": 1.650066380797624, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8174095839342624, "standard_deviation": 0.09704498464042993, "minimum": 0.7113395844247087, "maximum": 1.019295532558458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08009680947276193, "standard_deviation": 0.004974854270104652, "minimum": 0.07354980836545805, "maximum": 0.08907388773458562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982869830902226, "standard_deviation": 0.00019957723035798458, "minimum": 0.9979133993433771, "maximum": 0.9986240920762569, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9527342132156904, "standard_deviation": 0.018542731355118146, "minimum": 0.9203787452098015, "maximum": 0.9760471363969971, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7915392974658553, "standard_deviation": 0.14848392130448898, "minimum": 0.6291200563130296, "maximum": 1.0938910745953174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04623531618231323, "standard_deviation": 0.0027514218837588382, "minimum": 0.041283990534898604, "maximum": 0.05067548965598603, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23106296071468352, "standard_deviation": 0.04690303148055601, "minimum": 0.15559161866796936, "maximum": 0.3161345894105298, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.4780728680945785, "standard_deviation": 0.009501818982365701, "minimum": 5.467261626262581, "maximum": 5.492302204362417, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5695075529083579, "standard_deviation": 0.04097587244769266, "minimum": 0.512417857890608, "maximum": 0.6222086510365292, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06832652145377208, "standard_deviation": 0.01627884542430288, "minimum": -0.1000470689510604, "maximum": -0.0498886893906719, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967890240733069, "standard_deviation": 0.00031679489957722984, "minimum": 0.9963095985965619, "maximum": 0.9971992252520506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8999825084018109, "standard_deviation": 0.026412310801357256, "minimum": 0.8464370691029531, "maximum": 0.9301756794518775, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.28037139206751216, "standard_deviation": 0.07020610140653821, "minimum": 0.18262243526390434, "maximum": 0.415745964218278, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.052032388168123216, "standard_deviation": 0.002340240732795723, "minimum": 0.04906361172998878, "maximum": 0.055508974783226986, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990863990135599, "standard_deviation": 0.00014295604991402042, "minimum": 0.9989032937073312, "maximum": 0.9992940432080359, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4706501450211269, "standard_deviation": 0.02640267613550151, "minimum": 0.427885099073483, "maximum": 0.515873270572206, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2975911195991245, "standard_deviation": 0.14619947004102363, "minimum": 0.1456626587521607, "maximum": 0.5365821403281721, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027438349424710896, "standard_deviation": 0.0016633951890897916, "minimum": 0.025370062771111552, "maximum": 0.02976280305682023, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0023", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.0304317474365234, "predicted_mos": 3.0304317474365234, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f1db5c79fd977545df8931b0e556077d3c6422e5b1f6c8389116d90744724135", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.3641049000434577, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8997967532013342, "全帯域_10ms_rms_envelope_nrmse": 0.05349995079901508, "全帯域_10ms_envelope_correlation": 0.9966044352428887, "全帯域_10ms_modulation_mae_db": 0.31257858182982756, "全帯域_40ms_log_rms_mae_db": 0.4885790246803877, "全帯域_40ms_rms_envelope_nrmse": 0.029507387154170693, "全帯域_40ms_envelope_correlation": 0.9989327436943837, "全帯域_40ms_modulation_mae_db": 0.4806277147617286, "300–1000 Hz_10ms_log_rms_mae_db": 1.5360619789127852, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07625414645867772, "300–1000 Hz_10ms_envelope_correlation": 0.9951861980137285, "300–1000 Hz_10ms_modulation_mae_db": 0.8661373176644078, "300–1000 Hz_40ms_log_rms_mae_db": 0.8844926567606616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04745158983594363, "300–1000 Hz_40ms_envelope_correlation": 0.9982159490628246, "300–1000 Hz_40ms_modulation_mae_db": 0.7980330573764453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6532464770030664, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16053197707428984, "1000–3000 Hz_10ms_envelope_correlation": 0.9825626980143556, "1000–3000 Hz_10ms_modulation_mae_db": 0.9702686701054706, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0372641031053147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.101707506038036, "1000–3000 Hz_40ms_envelope_correlation": 0.9942457754376319, "1000–3000 Hz_40ms_modulation_mae_db": 0.7774476031018825, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9775369836274561, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17127488471030688, "3000–6000 Hz_10ms_envelope_correlation": 0.981201866486939, "3000–6000 Hz_10ms_modulation_mae_db": 0.39388444090690067, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2935838584033348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10404778580080605, "3000–6000 Hz_40ms_envelope_correlation": 0.9929152990347389, "3000–6000 Hz_40ms_modulation_mae_db": 0.6527031620156254, "mr_stft_db": 5.398326339140323, "band_mae_db": 0.18048370814823703, "rms_gain_error_db": -0.07140254347031205, "output_peak": 0.5680634546490505}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8577333603422691, "全帯域_10ms_rms_envelope_nrmse": 0.050933754562488255, "全帯域_10ms_envelope_correlation": 0.9968956499949183, "全帯域_10ms_modulation_mae_db": 0.18943072989813903, "全帯域_40ms_log_rms_mae_db": 0.43639085096541674, "全帯域_40ms_rms_envelope_nrmse": 0.025939981039323826, "全帯域_40ms_envelope_correlation": 0.9991513483121739, "全帯域_40ms_modulation_mae_db": 0.13826466701483256, "300–1000 Hz_10ms_log_rms_mae_db": 1.5805150335780067, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08298420026452737, "300–1000 Hz_10ms_envelope_correlation": 0.9941614551104069, "300–1000 Hz_10ms_modulation_mae_db": 0.7912371774549602, "300–1000 Hz_40ms_log_rms_mae_db": 0.9280748765916965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04533956323138745, "300–1000 Hz_40ms_envelope_correlation": 0.9983091842718159, "300–1000 Hz_40ms_modulation_mae_db": 0.8598187774872191, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6736465897014614, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15195414676474434, "1000–3000 Hz_10ms_envelope_correlation": 0.9844615483678439, "1000–3000 Hz_10ms_modulation_mae_db": 1.4564864834257305, "1000–3000 Hz_40ms_log_rms_mae_db": 1.071136884402208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09007009237692055, "1000–3000 Hz_40ms_envelope_correlation": 0.995734525154827, "1000–3000 Hz_40ms_modulation_mae_db": 1.1880693310954247, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0341456410836263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17276038595904625, "3000–6000 Hz_10ms_envelope_correlation": 0.9821406176061973, "3000–6000 Hz_10ms_modulation_mae_db": 0.9326279657366913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3258676995750982, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10056826486308786, "3000–6000 Hz_40ms_envelope_correlation": 0.9954776894788748, "3000–6000 Hz_40ms_modulation_mae_db": 0.885311994956302, "mr_stft_db": 5.408853738389062, "band_mae_db": 0.33448994337640126, "rms_gain_error_db": -0.0475828581356013, "output_peak": 0.5909908406852765}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8994256539945813, "全帯域_10ms_rms_envelope_nrmse": 0.05555966628437072, "全帯域_10ms_envelope_correlation": 0.9963038430029395, "全帯域_10ms_modulation_mae_db": 0.22306880558638356, "全帯域_40ms_log_rms_mae_db": 0.4512762092276518, "全帯域_40ms_rms_envelope_nrmse": 0.028859160035193204, "全帯域_40ms_envelope_correlation": 0.9989453439884663, "全帯域_40ms_modulation_mae_db": 0.14617632480622686, "300–1000 Hz_10ms_log_rms_mae_db": 1.587983565178319, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0816413727129892, "300–1000 Hz_10ms_envelope_correlation": 0.994318709005542, "300–1000 Hz_10ms_modulation_mae_db": 0.7046837391334616, "300–1000 Hz_40ms_log_rms_mae_db": 0.9223918260510908, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04584896937133885, "300–1000 Hz_40ms_envelope_correlation": 0.9982231790811339, "300–1000 Hz_40ms_modulation_mae_db": 0.6784530047814602, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7094903748924122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16240742306272776, "1000–3000 Hz_10ms_envelope_correlation": 0.9818852998792397, "1000–3000 Hz_10ms_modulation_mae_db": 1.2470032955648922, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1589106474296471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10684041139335847, "1000–3000 Hz_40ms_envelope_correlation": 0.9928981305757082, "1000–3000 Hz_40ms_modulation_mae_db": 1.043963798412762, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0136353657176524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15910461528913275, "3000–6000 Hz_10ms_envelope_correlation": 0.9838071013103724, "3000–6000 Hz_10ms_modulation_mae_db": 0.6526818828659433, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2637808930359922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08337268394225948, "3000–6000 Hz_40ms_envelope_correlation": 0.99580767576785, "3000–6000 Hz_40ms_modulation_mae_db": 0.7419545459073627, "mr_stft_db": 5.409895742801113, "band_mae_db": 0.23208758204906865, "rms_gain_error_db": -0.05059028542445855, "output_peak": 0.5822393278362265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9531856958757071, "全帯域_10ms_rms_envelope_nrmse": 0.05613023987297326, "全帯域_10ms_envelope_correlation": 0.9962711373065176, "全帯域_10ms_modulation_mae_db": 0.3132021792364374, "全帯域_40ms_log_rms_mae_db": 0.5069673860548892, "全帯域_40ms_rms_envelope_nrmse": 0.03068462040428609, "全帯域_40ms_envelope_correlation": 0.998858420034151, "全帯域_40ms_modulation_mae_db": 0.22156405180750882, "300–1000 Hz_10ms_log_rms_mae_db": 1.6564865184335367, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09211299702041204, "300–1000 Hz_10ms_envelope_correlation": 0.9928345613716533, "300–1000 Hz_10ms_modulation_mae_db": 0.8145655384305365, "300–1000 Hz_40ms_log_rms_mae_db": 0.9575495563501518, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05220944354122819, "300–1000 Hz_40ms_envelope_correlation": 0.9977880858666659, "300–1000 Hz_40ms_modulation_mae_db": 0.7444681674837449, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7193628678837323, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18154184187333786, "1000–3000 Hz_10ms_envelope_correlation": 0.977517166297113, "1000–3000 Hz_10ms_modulation_mae_db": 0.9129237442118235, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0951758798802602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12212851988473432, "1000–3000 Hz_40ms_envelope_correlation": 0.9911646829173684, "1000–3000 Hz_40ms_modulation_mae_db": 0.6851529349328046, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9723526629444563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17633999906296394, "3000–6000 Hz_10ms_envelope_correlation": 0.9801779966026706, "3000–6000 Hz_10ms_modulation_mae_db": 0.6914341110464282, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2691015066896754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09525185451200785, "3000–6000 Hz_40ms_envelope_correlation": 0.9941364238233095, "3000–6000 Hz_40ms_modulation_mae_db": 0.6751387576017049, "mr_stft_db": 5.395650505677877, "band_mae_db": 0.23310412844716635, "rms_gain_error_db": -0.07354342122798542, "output_peak": 0.6193271660581299}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8809534747877363, "全帯域_10ms_rms_envelope_nrmse": 0.048625009856257576, "全帯域_10ms_envelope_correlation": 0.9973017745038306, "全帯域_10ms_modulation_mae_db": 0.30257402175379583, "全帯域_40ms_log_rms_mae_db": 0.46069470991279915, "全帯域_40ms_rms_envelope_nrmse": 0.02659338539600049, "全帯域_40ms_envelope_correlation": 0.9992570293624605, "全帯域_40ms_modulation_mae_db": 0.36277647002171065, "300–1000 Hz_10ms_log_rms_mae_db": 1.6244880638836157, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08098146687266823, "300–1000 Hz_10ms_envelope_correlation": 0.9946719401306172, "300–1000 Hz_10ms_modulation_mae_db": 0.8173255733435214, "300–1000 Hz_40ms_log_rms_mae_db": 0.9630203348298879, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047423406474816424, "300–1000 Hz_40ms_envelope_correlation": 0.9983919945136901, "300–1000 Hz_40ms_modulation_mae_db": 0.6998080666249429, "1000–3000 Hz_10ms_log_rms_mae_db": 1.640755083096135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14626790123877087, "1000–3000 Hz_10ms_envelope_correlation": 0.9875444142198212, "1000–3000 Hz_10ms_modulation_mae_db": 0.8048651809516837, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0407135695006826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09716808653989281, "1000–3000 Hz_40ms_envelope_correlation": 0.9967124637943258, "1000–3000 Hz_40ms_modulation_mae_db": 0.7975644360085434, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0336849780343034, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14250886668188523, "3000–6000 Hz_10ms_envelope_correlation": 0.986938945514675, "3000–6000 Hz_10ms_modulation_mae_db": 0.8683184628087309, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3107226296320686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08025908285513537, "3000–6000 Hz_40ms_envelope_correlation": 0.9958414845405897, "3000–6000 Hz_40ms_modulation_mae_db": 0.9375863773043895, "mr_stft_db": 5.402407700002285, "band_mae_db": 0.2715254413933775, "rms_gain_error_db": -0.10474622837546499, "output_peak": 0.5106327996227837}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9138468128037328, "全帯域_10ms_rms_envelope_nrmse": 0.05230439079775523, "全帯域_10ms_envelope_correlation": 0.9967529767881171, "全帯域_10ms_modulation_mae_db": 0.28509330073516553, "全帯域_40ms_log_rms_mae_db": 0.46849851540594456, "全帯域_40ms_rms_envelope_nrmse": 0.02648105443616783, "全帯域_40ms_envelope_correlation": 0.999145060139271, "全帯域_40ms_modulation_mae_db": 0.32823705182226004, "300–1000 Hz_10ms_log_rms_mae_db": 1.5905035036764874, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08188713358155746, "300–1000 Hz_10ms_envelope_correlation": 0.9942944115767581, "300–1000 Hz_10ms_modulation_mae_db": 0.8137030576728161, "300–1000 Hz_40ms_log_rms_mae_db": 0.9107252183529919, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04647219494577552, "300–1000 Hz_40ms_envelope_correlation": 0.9981790321042245, "300–1000 Hz_40ms_modulation_mae_db": 0.7973788922934072, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6851403920429744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15178634593202536, "1000–3000 Hz_10ms_envelope_correlation": 0.9862311745603155, "1000–3000 Hz_10ms_modulation_mae_db": 0.6045672825546746, "1000–3000 Hz_40ms_log_rms_mae_db": 1.082246376228654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10289119534784544, "1000–3000 Hz_40ms_envelope_correlation": 0.9957583701447106, "1000–3000 Hz_40ms_modulation_mae_db": 0.6135539862898781, "3000–6000 Hz_10ms_log_rms_mae_db": 2.006741284379956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13891582619445791, "3000–6000 Hz_10ms_envelope_correlation": 0.9876106270036316, "3000–6000 Hz_10ms_modulation_mae_db": 0.6899029145740978, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3005604029456552, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0719405427000731, "3000–6000 Hz_40ms_envelope_correlation": 0.9967426363773128, "3000–6000 Hz_40ms_modulation_mae_db": 0.9608396079406519, "mr_stft_db": 5.406771680864345, "band_mae_db": 0.2603564370785303, "rms_gain_error_db": -0.06838964878214096, "output_peak": 0.5292384404280375}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9354107059694003, "全帯域_10ms_rms_envelope_nrmse": 0.052888371658126286, "全帯域_10ms_envelope_correlation": 0.9967106586240552, "全帯域_10ms_modulation_mae_db": 0.33980508185130326, "全帯域_40ms_log_rms_mae_db": 0.4807289737281986, "全帯域_40ms_rms_envelope_nrmse": 0.029125288614360933, "全帯域_40ms_envelope_correlation": 0.9989918475367597, "全帯域_40ms_modulation_mae_db": 0.4344906465921792, "300–1000 Hz_10ms_log_rms_mae_db": 1.6117658202915004, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08168019392046469, "300–1000 Hz_10ms_envelope_correlation": 0.9944569567707104, "300–1000 Hz_10ms_modulation_mae_db": 0.9542470101853172, "300–1000 Hz_40ms_log_rms_mae_db": 0.9241453833251704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04644076880461398, "300–1000 Hz_40ms_envelope_correlation": 0.9983499223230998, "300–1000 Hz_40ms_modulation_mae_db": 0.9883600385684922, "1000–3000 Hz_10ms_log_rms_mae_db": 1.778550970388749, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1838805341923755, "1000–3000 Hz_10ms_envelope_correlation": 0.9762849613642846, "1000–3000 Hz_10ms_modulation_mae_db": 0.8971842351536706, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1556631333974507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11213599615295462, "1000–3000 Hz_40ms_envelope_correlation": 0.9921848904904629, "1000–3000 Hz_40ms_modulation_mae_db": 0.8475425530400112, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0022412316319977, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19551731180025178, "3000–6000 Hz_10ms_envelope_correlation": 0.9753858776411618, "3000–6000 Hz_10ms_modulation_mae_db": 1.2560730838854919, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2587412847955584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1076670495036371, "3000–6000 Hz_40ms_envelope_correlation": 0.9924423994266501, "3000–6000 Hz_40ms_modulation_mae_db": 1.4233808540741384, "mr_stft_db": 5.409768872218348, "band_mae_db": 0.23186820583524667, "rms_gain_error_db": -0.08392404456574071, "output_peak": 0.5527447191393802}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9342458406953669, "全帯域_10ms_rms_envelope_nrmse": 0.04901528765977203, "全帯域_10ms_envelope_correlation": 0.9971689670221182, "全帯域_10ms_modulation_mae_db": 0.34077687090766634, "全帯域_40ms_log_rms_mae_db": 0.48462239925518624, "全帯域_40ms_rms_envelope_nrmse": 0.025730167771656653, "全帯域_40ms_envelope_correlation": 0.9992163503953945, "全帯域_40ms_modulation_mae_db": 0.2599875668980474, "300–1000 Hz_10ms_log_rms_mae_db": 1.54724836076011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07527633265946117, "300–1000 Hz_10ms_envelope_correlation": 0.9952257600923752, "300–1000 Hz_10ms_modulation_mae_db": 0.7917754949519393, "300–1000 Hz_40ms_log_rms_mae_db": 0.9001741133454342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04226081724156515, "300–1000 Hz_40ms_envelope_correlation": 0.9985616809899508, "300–1000 Hz_40ms_modulation_mae_db": 0.7129129173188001, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6439185914671053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17858235402846104, "1000–3000 Hz_10ms_envelope_correlation": 0.9797376053639603, "1000–3000 Hz_10ms_modulation_mae_db": 1.2287765730287725, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0477688003573271, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12321275493898719, "1000–3000 Hz_40ms_envelope_correlation": 0.9925338510431941, "1000–3000 Hz_40ms_modulation_mae_db": 1.095943138503883, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0487307038379714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17839682966769, "3000–6000 Hz_10ms_envelope_correlation": 0.9794519317847427, "3000–6000 Hz_10ms_modulation_mae_db": 0.42354119181110045, "3000–6000 Hz_40ms_log_rms_mae_db": 1.315568884518743, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10757028723188519, "3000–6000 Hz_40ms_envelope_correlation": 0.9923905427513211, "3000–6000 Hz_40ms_modulation_mae_db": 0.32401935837049917, "mr_stft_db": 5.415932462456223, "band_mae_db": 0.256820548725598, "rms_gain_error_db": -0.07324282842194041, "output_peak": 0.5492464458386119}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.650094102399637, "spectral_convergence": 0.16776566265760828, "spectral_convergence_status": "ok", "amplitude_floor": 1.2959192499433599e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2177316447143784, "spectral_convergence": 0.16477345887654826, "spectral_convergence_status": "ok", "amplitude_floor": 8.646137696849433e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.327153270306956, "spectral_convergence": 0.16607661037899762, "spectral_convergence_status": "ok", "amplitude_floor": 6.578152434885499e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0020500598531285057, "output_power": 0.002081063370898424, "delta_db": 0.06518764391810633, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0025325885847670413, "output_power": 0.002458673095137817, "delta_db": -0.12863856192338208, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009745321457083826, "output_power": 0.0009527415950819043, "delta_db": -0.09821043042684316, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029510461183058167, "output_power": 0.0002660550156575363, "delta_db": -0.4500454597401805, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.807836063095196e-05, "output_power": 3.529805831173064e-05, "delta_db": -0.3292742642883409, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.519436277144243e-05, "output_power": 3.5100921445804904e-05, "delta_db": -0.011545888592569168, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18048370814823703, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07140254347031205, "pitch": null, "output_rms_dbfs": -22.3441100689443, "output_peak": 0.5680634546490505, "reference_rms_dbfs": -22.27270752547399, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.398326339140323}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019946925088635918, "log_rms_mae_db": 0.8997967532013342, "rms_envelope_nrmse": 0.05349995079901508, "envelope_correlation": 0.9966044352428887, "rms_gain_error_db": -0.0695033043426076, "reference_mean_power": 0.0059224241180442505, "output_mean_power": 0.0058283976515900535}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0772791897966612e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010245127053592495, "output_power": 0.0010162746595894392, "reference_active": true, "delta_db": -0.03506251865478546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001329567656774176, "output_power": 0.00012165108441146365, "reference_active": true, "delta_db": -0.3859445723371844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.301896656350168e-05, "output_power": 2.0436802908125544e-05, "reference_active": true, "delta_db": -0.5167286544975128}], "active_band_count": 3, "active_band_mae_db": 0.31257858182982756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001930447992601615, "log_rms_mae_db": 0.4885790246803877, "rms_envelope_nrmse": 0.029507387154170693, "envelope_correlation": 0.9989327436943837, "rms_gain_error_db": -0.06973644247745493, "reference_mean_power": 0.005919423744261807, "output_mean_power": 0.005825132199416331}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.875987153432414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009473592447040891, "output_power": 0.0009405019125566989, "reference_active": true, "delta_db": -0.031550144440889945}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.123049684235515e-05, "output_power": 5.574740386077765e-05, "reference_active": true, "delta_db": -0.4074313621140453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.329909700774956e-06, "output_power": 1.0556792616986298e-06, "reference_active": true, "delta_db": -1.0029016377302507}], "active_band_count": 3, "active_band_mae_db": 0.4806277147617286}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015110036587973006, "log_rms_mae_db": 1.5360619789127852, "rms_envelope_nrmse": 0.07625414645867772, "envelope_correlation": 0.9951861980137285, "rms_gain_error_db": -0.16620435453437088, "reference_mean_power": 0.0025365732398871734, "output_mean_power": 0.002441332715525289}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.31579034521966e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006679887319597139, "output_power": 0.0006438238631081172, "reference_active": true, "delta_db": -0.1600206697160689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.085773760663629e-05, "output_power": 7.590389199225017e-05, "reference_active": true, "delta_db": -0.7809787347852971}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6589878271689645e-05, "output_power": 1.1326662095666661e-05, "reference_active": true, "delta_db": -1.6574125484918576}], "active_band_count": 3, "active_band_mae_db": 0.8661373176644078}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013775289901134618, "log_rms_mae_db": 0.8844926567606616, "rms_envelope_nrmse": 0.04745158983594363, "envelope_correlation": 0.9982159490628246, "rms_gain_error_db": -0.1668764498752945, "reference_mean_power": 0.0025400163932975937, "output_mean_power": 0.002444268295462265}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2092894134690622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006291541070562532, "output_power": 0.0006053758860943843, "reference_active": true, "delta_db": -0.16731918244614372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.975083640469782e-05, "output_power": 3.3677461438611174e-05, "reference_active": true, "delta_db": -0.720069236270459}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.7844516380393e-07, "output_power": 6.91616611291151e-07, "reference_active": true, "delta_db": -1.506710753412733}], "active_band_count": 3, "active_band_mae_db": 0.7980330573764453}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010600794217574433, "log_rms_mae_db": 1.6532464770030664, "rms_envelope_nrmse": 0.16053197707428984, "envelope_correlation": 0.9825626980143556, "rms_gain_error_db": -0.49951377596415625, "reference_mean_power": 0.0002691478602162077, "output_mean_power": 0.00023990514052390512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.02821083589085e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010735336162950012, "output_power": 9.522732157417163e-05, "reference_active": true, "delta_db": -0.5205407909468529}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.315168535027072e-06, "output_power": 7.379876600448063e-06, "reference_active": true, "delta_db": -0.5182195608364247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.408982675283079e-06, "output_power": 3.707133641424189e-06, "reference_active": true, "delta_db": 1.8720456585331344}], "active_band_count": 3, "active_band_mae_db": 0.9702686701054706}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.461607013685591e-05, "log_rms_mae_db": 1.0372641031053147, "rms_envelope_nrmse": 0.101707506038036, "envelope_correlation": 0.9942457754376319, "rms_gain_error_db": -0.5100212282498263, "reference_mean_power": 0.00026921330106685756, "output_mean_power": 0.00023938359814454314}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9187018796507895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.751052870210862e-05, "output_power": 8.7417669241111e-05, "reference_active": true, "delta_db": -0.47452288184894686}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.818200149529557e-06, "output_power": 3.441585705904976e-06, "reference_active": true, "delta_db": -0.4510010103965504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4850622077020166e-07, "output_power": 2.0531776963700241e-07, "reference_active": true, "delta_db": 1.4068189170601504}], "active_band_count": 3, "active_band_mae_db": 0.7774476031018825}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.633594187238404e-05, "log_rms_mae_db": 1.9775369836274561, "rms_envelope_nrmse": 0.17127488471030688, "envelope_correlation": 0.981201866486939, "rms_gain_error_db": -0.05094132256355136, "reference_mean_power": 4.9508565949953196e-05, "output_mean_power": 4.893123920713758e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.164436282244953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0281496224399232e-05, "output_power": 9.909428283281443e-06, "reference_active": true, "delta_db": -0.16007721443569187}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.187880808785332e-06, "output_power": 3.6813930267099736e-06, "reference_active": true, "delta_db": -0.5598212739261519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.689088055617476e-07, "output_power": 5.115253631762949e-07, "reference_active": true, "delta_db": -0.46175483435885817}], "active_band_count": 3, "active_band_mae_db": 0.39388444090690067}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.3561893402191595e-05, "log_rms_mae_db": 1.2935838584033348, "rms_envelope_nrmse": 0.10404778580080605, "envelope_correlation": 0.9929152990347389, "rms_gain_error_db": -0.05084563407847392, "reference_mean_power": 4.9751584691373346e-05, "output_mean_power": 4.917250748115543e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0334524009764534e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.035252748280077e-05, "output_power": 1.0101226960760361e-05, "reference_active": true, "delta_db": -0.10672262874585375}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2970706912154704e-06, "output_power": 1.9954125644271403e-06, "reference_active": true, "delta_db": -0.6114165810476019}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.66930165214842e-08, "output_power": 3.5095780408126144e-08, "reference_active": true, "delta_db": -1.2399702762534206}], "active_band_count": 3, "active_band_mae_db": 0.6527031620156254}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9193510121538515, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4a618b6c9d437bc9d9bba45837ebec4b780ae28a868383770db9d7e779d70d3d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9820281085083666, "standard_deviation": 0.004016073884846348, "minimum": 0.9762849613642846, "maximum": 0.9875444142198212, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6880139183094545, "standard_deviation": 0.046663231534290764, "minimum": 1.640755083096135, "maximum": 1.778550970388749, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0152594331245899, "standard_deviation": 0.2760301212793272, "minimum": 0.6045672825546746, "maximum": 1.4564864834257305, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16461906552084155, "standard_deviation": 0.014812146100856623, "minimum": 0.14626790123877087, "maximum": 0.1838805341923755, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939040861947787, "standard_deviation": 0.0020050674124028433, "minimum": 0.9911646829173684, "maximum": 0.9967124637943258, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.086109924287693, "standard_deviation": 0.0483671815009975, "minimum": 1.0372641031053147, "maximum": 1.1589106474296471, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8811547226731486, "standard_deviation": 0.20554338127906885, "minimum": 0.6135539862898781, "maximum": 1.1880693310954247, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10701932033409117, "standard_deviation": 0.01162698547000866, "minimum": 0.09007009237692055, "maximum": 0.12321275493898719, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9820893704937987, "standard_deviation": 0.0040229660280245005, "minimum": 0.9753858776411618, "maximum": 0.9876106270036316, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0111336064071774, "standard_deviation": 0.027226510444989128, "minimum": 1.9723526629444563, "maximum": 2.0487307038379714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7385580067044231, "standard_deviation": 0.28089676010940967, "minimum": 0.39388444090690067, "maximum": 1.2560730838854919, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16685233992071685, "standard_deviation": 0.01902072617005284, "minimum": 0.13891582619445791, "maximum": 0.19551731180025178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944692689000808, "standard_deviation": 0.0017243769437685769, "minimum": 0.9923905427513211, "maximum": 0.9967426363773128, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2922408949495159, "standard_deviation": 0.02550959552265438, "minimum": 1.2587412847955584, "maximum": 1.3258676995750982, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8251168322713343, "standard_deviation": 0.31687039003925266, "minimum": 0.32401935837049917, "maximum": 1.4233808540741384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09383469392611152, "standard_deviation": 0.013652639087508994, "minimum": 0.0719405427000731, "maximum": 0.1076670495036371, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943937490089739, "standard_deviation": 0.0007466237205469263, "minimum": 0.9928345613716533, "maximum": 0.9952257600923752, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.591881605589295, "standard_deviation": 0.03943640086072968, "minimum": 1.5360619789127852, "maximum": 1.6564865184335367, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.81920936360462, "standard_deviation": 0.07085600192044755, "minimum": 0.7046837391334616, "maximum": 0.9542470101853172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08160223043634474, "standard_deviation": 0.005091718308521475, "minimum": 0.07527633265946117, "maximum": 0.09211299702041204, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982523785266757, "standard_deviation": 0.00022393538857753476, "minimum": 0.9977880858666659, "maximum": 0.9985616809899508, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9238217457008857, "standard_deviation": 0.026650756934436637, "minimum": 0.8844926567606616, "maximum": 0.9630203348298879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.784904115241814, "standard_deviation": 0.1019281843807176, "minimum": 0.6784530047814602, "maximum": 0.9883600385684922, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.046680844180833646, "standard_deviation": 0.0027733352182759173, "minimum": 0.04226081724156515, "maximum": 0.05220944354122819, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25009199938170323, "standard_deviation": 0.04395323018610372, "minimum": 0.18048370814823703, "maximum": 0.33448994337640126, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.405950880193696, "standard_deviation": 0.006715193384175807, "minimum": 5.395650505677877, "maximum": 5.415932462456223, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5628103992821871, "standard_deviation": 0.03490443511975237, "minimum": 0.5106327996227837, "maximum": 0.6193271660581299, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07167773230045554, "standard_deviation": 0.018069542316479833, "minimum": -0.10474622837546499, "maximum": -0.0475828581356013, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967511803106731, "standard_deviation": 0.0003688403629776516, "minimum": 0.9962711373065176, "maximum": 0.9973017745038306, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.909324787208766, "standard_deviation": 0.031398692088528246, "minimum": 0.8577333603422691, "maximum": 0.9531856958757071, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.28831619647483986, "standard_deviation": 0.05458994323507232, "minimum": 0.18943072989813903, "maximum": 0.34077687090766634, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.052369583936344805, "standard_deviation": 0.0027569398950342233, "minimum": 0.048625009856257576, "maximum": 0.05613023987297326, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990622679328826, "standard_deviation": 0.00014804633360744406, "minimum": 0.998858420034151, "maximum": 0.9992570293624605, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4722197586538093, "standard_deviation": 0.022584042148964627, "minimum": 0.43639085096541674, "maximum": 0.5069673860548892, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.29651556171556176, "standard_deviation": 0.12701429892569346, "minimum": 0.13826466701483256, "maximum": 0.4806277147617286, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027865130606394964, "standard_deviation": 0.0018905740581686534, "minimum": 0.025730167771656653, "maximum": 0.03068462040428609, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0023", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.5034878253936768, "predicted_mos": 3.5034878253936768, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "475d35de181370a596d5d67bc54faa68d695fe6fa43421d4b6de64b0a66cb4bb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 14.04486249992624, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.860328810808378, "全帯域_10ms_rms_envelope_nrmse": 0.050368308695589205, "全帯域_10ms_envelope_correlation": 0.9969509707976757, "全帯域_10ms_modulation_mae_db": 0.13318102701484635, "全帯域_40ms_log_rms_mae_db": 0.4456735854848855, "全帯域_40ms_rms_envelope_nrmse": 0.02680667849942169, "全帯域_40ms_envelope_correlation": 0.9990692479063576, "全帯域_40ms_modulation_mae_db": 0.0701278208627964, "300–1000 Hz_10ms_log_rms_mae_db": 1.9367233795491445, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08154523232167245, "300–1000 Hz_10ms_envelope_correlation": 0.9943902476480814, "300–1000 Hz_10ms_modulation_mae_db": 0.6572353513361378, "300–1000 Hz_40ms_log_rms_mae_db": 1.1657753500828023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05010244055107497, "300–1000 Hz_40ms_envelope_correlation": 0.9978967187912782, "300–1000 Hz_40ms_modulation_mae_db": 0.84843569519533, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0826206040365736, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16654121722455698, "1000–3000 Hz_10ms_envelope_correlation": 0.980114239783992, "1000–3000 Hz_10ms_modulation_mae_db": 1.4738731518121277, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3135574465649407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08678817288389172, "1000–3000 Hz_40ms_envelope_correlation": 0.9951351130796886, "1000–3000 Hz_40ms_modulation_mae_db": 1.251877912092774, "3000–6000 Hz_10ms_log_rms_mae_db": 2.629450537330441, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2044862546790846, "3000–6000 Hz_10ms_envelope_correlation": 0.9729838253487338, "3000–6000 Hz_10ms_modulation_mae_db": 0.3894958422315846, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6563446837898803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1171123564755427, "3000–6000 Hz_40ms_envelope_correlation": 0.9907636848589024, "3000–6000 Hz_40ms_modulation_mae_db": 0.6532471582173466, "mr_stft_db": 5.8202595220990085, "band_mae_db": 0.19542572402031955, "rms_gain_error_db": -0.032293540290158636, "output_peak": 0.5472523477106392}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8667187227595072, "全帯域_10ms_rms_envelope_nrmse": 0.04952491212447185, "全帯域_10ms_envelope_correlation": 0.9970985978531505, "全帯域_10ms_modulation_mae_db": 0.3501638538449566, "全帯域_40ms_log_rms_mae_db": 0.4224330753044301, "全帯域_40ms_rms_envelope_nrmse": 0.025459694092527373, "全帯域_40ms_envelope_correlation": 0.9992214650558242, "全帯域_40ms_modulation_mae_db": 0.3598721640687212, "300–1000 Hz_10ms_log_rms_mae_db": 1.76276787079034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07882709243923049, "300–1000 Hz_10ms_envelope_correlation": 0.9948934216229077, "300–1000 Hz_10ms_modulation_mae_db": 0.8883205646289101, "300–1000 Hz_40ms_log_rms_mae_db": 1.0577833745123133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04620961289489795, "300–1000 Hz_40ms_envelope_correlation": 0.9983860366015774, "300–1000 Hz_40ms_modulation_mae_db": 0.8986957680521951, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9863962061414682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15223689747135855, "1000–3000 Hz_10ms_envelope_correlation": 0.9834114891323125, "1000–3000 Hz_10ms_modulation_mae_db": 0.8119852075908031, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2905699381088145, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08870017940741423, "1000–3000 Hz_40ms_envelope_correlation": 0.9946472609147806, "1000–3000 Hz_40ms_modulation_mae_db": 0.5356886989816051, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7021553718163647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21100900912886594, "3000–6000 Hz_10ms_envelope_correlation": 0.9710288784372786, "3000–6000 Hz_10ms_modulation_mae_db": 0.2999662213622097, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7705331149315142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12551236905345522, "3000–6000 Hz_40ms_envelope_correlation": 0.9896676961294543, "3000–6000 Hz_40ms_modulation_mae_db": 0.7813515241611184, "mr_stft_db": 5.802977228752166, "band_mae_db": 0.1847371244634847, "rms_gain_error_db": -0.06533446696182983, "output_peak": 0.5544720679954194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8987812074846897, "全帯域_10ms_rms_envelope_nrmse": 0.05310837592371483, "全帯域_10ms_envelope_correlation": 0.9966309058491171, "全帯域_10ms_modulation_mae_db": 0.22893511717467555, "全帯域_40ms_log_rms_mae_db": 0.46942583440166236, "全帯域_40ms_rms_envelope_nrmse": 0.027402910350053985, "全帯域_40ms_envelope_correlation": 0.9990611816058659, "全帯域_40ms_modulation_mae_db": 0.2818735696277314, "300–1000 Hz_10ms_log_rms_mae_db": 1.7853520336888558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0741146698748035, "300–1000 Hz_10ms_envelope_correlation": 0.9954293053671467, "300–1000 Hz_10ms_modulation_mae_db": 0.5760433853794613, "300–1000 Hz_40ms_log_rms_mae_db": 1.0547175515017166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04321743757324506, "300–1000 Hz_40ms_envelope_correlation": 0.9985280690214586, "300–1000 Hz_40ms_modulation_mae_db": 0.6495618510525989, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0496912240285297, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1612962711167751, "1000–3000 Hz_10ms_envelope_correlation": 0.9829322332183588, "1000–3000 Hz_10ms_modulation_mae_db": 1.35062742614483, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3473923979134468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09757716355528398, "1000–3000 Hz_40ms_envelope_correlation": 0.9955119976811605, "1000–3000 Hz_40ms_modulation_mae_db": 1.2049048854388191, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6067896891969022, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17880880800376397, "3000–6000 Hz_10ms_envelope_correlation": 0.9793171699471226, "3000–6000 Hz_10ms_modulation_mae_db": 0.5588748373756734, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6721900227615167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10299601764127082, "3000–6000 Hz_40ms_envelope_correlation": 0.9929632111762373, "3000–6000 Hz_40ms_modulation_mae_db": 0.9200425158138601, "mr_stft_db": 5.8178694396581685, "band_mae_db": 0.20505648581931213, "rms_gain_error_db": -0.05256406247843671, "output_peak": 0.5409381756255487}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8981119112675982, "全帯域_10ms_rms_envelope_nrmse": 0.05399821342840262, "全帯域_10ms_envelope_correlation": 0.9965531531376421, "全帯域_10ms_modulation_mae_db": 0.19142251997227663, "全帯域_40ms_log_rms_mae_db": 0.4786953700763545, "全帯域_40ms_rms_envelope_nrmse": 0.029329739304956297, "全帯域_40ms_envelope_correlation": 0.9989701413584329, "全帯域_40ms_modulation_mae_db": 0.2072297574906984, "300–1000 Hz_10ms_log_rms_mae_db": 1.9063976263913567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08433425157070991, "300–1000 Hz_10ms_envelope_correlation": 0.9941575200261565, "300–1000 Hz_10ms_modulation_mae_db": 0.7571421879574548, "300–1000 Hz_40ms_log_rms_mae_db": 1.218738227156685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049957171845736996, "300–1000 Hz_40ms_envelope_correlation": 0.9981146152388075, "300–1000 Hz_40ms_modulation_mae_db": 0.9525963076047432, "1000–3000 Hz_10ms_log_rms_mae_db": 2.121037498181448, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18266120414768813, "1000–3000 Hz_10ms_envelope_correlation": 0.9757498459268638, "1000–3000 Hz_10ms_modulation_mae_db": 1.0324060941564028, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3709349815508698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1025936028905908, "1000–3000 Hz_40ms_envelope_correlation": 0.9925499655033638, "1000–3000 Hz_40ms_modulation_mae_db": 0.7436833280473835, "3000–6000 Hz_10ms_log_rms_mae_db": 2.623699984187618, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17357070457315965, "3000–6000 Hz_10ms_envelope_correlation": 0.9805707748520932, "3000–6000 Hz_10ms_modulation_mae_db": 0.8974086313333017, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6263464715513356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10140241074173734, "3000–6000 Hz_40ms_envelope_correlation": 0.993353263502835, "3000–6000 Hz_40ms_modulation_mae_db": 0.6127912723551119, "mr_stft_db": 5.803612704755564, "band_mae_db": 0.20710940751271442, "rms_gain_error_db": -0.07624200400446672, "output_peak": 0.5729592838054264}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8448940981585851, "全帯域_10ms_rms_envelope_nrmse": 0.051047815452911455, "全帯域_10ms_envelope_correlation": 0.9968888736512511, "全帯域_10ms_modulation_mae_db": 0.3171330018667465, "全帯域_40ms_log_rms_mae_db": 0.41768966406431984, "全帯域_40ms_rms_envelope_nrmse": 0.02762421883962295, "全帯域_40ms_envelope_correlation": 0.9990371643678322, "全帯域_40ms_modulation_mae_db": 0.22641696469309405, "300–1000 Hz_10ms_log_rms_mae_db": 1.8528746908151796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07720001132360284, "300–1000 Hz_10ms_envelope_correlation": 0.995024165281163, "300–1000 Hz_10ms_modulation_mae_db": 0.4748221580147429, "300–1000 Hz_40ms_log_rms_mae_db": 1.1303722455835188, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04510236671092036, "300–1000 Hz_40ms_envelope_correlation": 0.9983738335933126, "300–1000 Hz_40ms_modulation_mae_db": 0.6149808659847804, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0758084533343366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16010045691111632, "1000–3000 Hz_10ms_envelope_correlation": 0.9820960722582315, "1000–3000 Hz_10ms_modulation_mae_db": 1.1167372502753492, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3389063959622813, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10226071538205696, "1000–3000 Hz_40ms_envelope_correlation": 0.9933032573268578, "1000–3000 Hz_40ms_modulation_mae_db": 1.034888111456635, "3000–6000 Hz_10ms_log_rms_mae_db": 2.611165748935469, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1729219945682348, "3000–6000 Hz_10ms_envelope_correlation": 0.9807607745691983, "3000–6000 Hz_10ms_modulation_mae_db": 1.0992089721799354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6484399821642255, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0964290455961891, "3000–6000 Hz_40ms_envelope_correlation": 0.9937381507703497, "3000–6000 Hz_40ms_modulation_mae_db": 1.261252457034985, "mr_stft_db": 5.808285605844053, "band_mae_db": 0.21406977546280945, "rms_gain_error_db": -0.054353451982082274, "output_peak": 0.5775735999320196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.814078697998297, "全帯域_10ms_rms_envelope_nrmse": 0.05181243788253693, "全帯域_10ms_envelope_correlation": 0.9968442725785864, "全帯域_10ms_modulation_mae_db": 0.3597548217490953, "全帯域_40ms_log_rms_mae_db": 0.4330974914912617, "全帯域_40ms_rms_envelope_nrmse": 0.027016163566848484, "全帯域_40ms_envelope_correlation": 0.999144704278804, "全帯域_40ms_modulation_mae_db": 0.4704182272194229, "300–1000 Hz_10ms_log_rms_mae_db": 1.8021001537449415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08279821651033922, "300–1000 Hz_10ms_envelope_correlation": 0.9944910602476192, "300–1000 Hz_10ms_modulation_mae_db": 0.7486686499476324, "300–1000 Hz_40ms_log_rms_mae_db": 1.1058142513263736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04888714727759866, "300–1000 Hz_40ms_envelope_correlation": 0.998352148493242, "300–1000 Hz_40ms_modulation_mae_db": 1.0036382190178326, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1033742447751305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15879329887085564, "1000–3000 Hz_10ms_envelope_correlation": 0.9828736978068192, "1000–3000 Hz_10ms_modulation_mae_db": 0.8004041714616696, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3524712956761198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0964247900769633, "1000–3000 Hz_40ms_envelope_correlation": 0.9947813970405753, "1000–3000 Hz_40ms_modulation_mae_db": 1.0070123721353605, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5697907448449095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16304584076983378, "3000–6000 Hz_10ms_envelope_correlation": 0.9834523176546606, "3000–6000 Hz_10ms_modulation_mae_db": 1.1565369479922334, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6297411203364038, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09201009953557346, "3000–6000 Hz_40ms_envelope_correlation": 0.9953285070908964, "3000–6000 Hz_40ms_modulation_mae_db": 1.1891284342997734, "mr_stft_db": 5.789951337852298, "band_mae_db": 0.2815385352038042, "rms_gain_error_db": -0.08283323564875356, "output_peak": 0.5470499070273255}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8625580161152266, "全帯域_10ms_rms_envelope_nrmse": 0.05221749114414063, "全帯域_10ms_envelope_correlation": 0.9967709892253896, "全帯域_10ms_modulation_mae_db": 0.12811169774710074, "全帯域_40ms_log_rms_mae_db": 0.4497850155512504, "全帯域_40ms_rms_envelope_nrmse": 0.026334499383986812, "全帯域_40ms_envelope_correlation": 0.9991652013109087, "全帯域_40ms_modulation_mae_db": 0.3272495183287036, "300–1000 Hz_10ms_log_rms_mae_db": 1.8940087453402767, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07933281149521637, "300–1000 Hz_10ms_envelope_correlation": 0.9946416632911232, "300–1000 Hz_10ms_modulation_mae_db": 0.6203338416137977, "300–1000 Hz_40ms_log_rms_mae_db": 1.1180918518646537, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04445710964001365, "300–1000 Hz_40ms_envelope_correlation": 0.9983344194787822, "300–1000 Hz_40ms_modulation_mae_db": 0.853529976223399, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9945588586842191, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15844565532044314, "1000–3000 Hz_10ms_envelope_correlation": 0.9830415223487591, "1000–3000 Hz_10ms_modulation_mae_db": 0.7944920190433397, "1000–3000 Hz_40ms_log_rms_mae_db": 1.291618639505529, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10574103859995457, "1000–3000 Hz_40ms_envelope_correlation": 0.9934001850863623, "1000–3000 Hz_40ms_modulation_mae_db": 0.6629154335445482, "3000–6000 Hz_10ms_log_rms_mae_db": 2.633373533618667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18986469721132582, "3000–6000 Hz_10ms_envelope_correlation": 0.9767119471295304, "3000–6000 Hz_10ms_modulation_mae_db": 0.8741985717191149, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6416257732328459, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10548635349266135, "3000–6000 Hz_40ms_envelope_correlation": 0.9929666412426191, "3000–6000 Hz_40ms_modulation_mae_db": 0.351568669222354, "mr_stft_db": 5.80937188033615, "band_mae_db": 0.23676314131106038, "rms_gain_error_db": -0.06506673259360193, "output_peak": 0.5484596278189559}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8687235561445235, "全帯域_10ms_rms_envelope_nrmse": 0.05322398727215402, "全帯域_10ms_envelope_correlation": 0.9966285765260774, "全帯域_10ms_modulation_mae_db": 0.38598104513008447, "全帯域_40ms_log_rms_mae_db": 0.43095902321095947, "全帯域_40ms_rms_envelope_nrmse": 0.02948419982950831, "全帯域_40ms_envelope_correlation": 0.9989176635536222, "全帯域_40ms_modulation_mae_db": 0.38666375294563454, "300–1000 Hz_10ms_log_rms_mae_db": 1.798289175943655, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07949966426006423, "300–1000 Hz_10ms_envelope_correlation": 0.9947711229804634, "300–1000 Hz_10ms_modulation_mae_db": 0.8134756739391884, "300–1000 Hz_40ms_log_rms_mae_db": 1.1130031282155524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044786797623714805, "300–1000 Hz_40ms_envelope_correlation": 0.9984802116160952, "300–1000 Hz_40ms_modulation_mae_db": 0.966765517270228, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0639891164945205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16191666998295132, "1000–3000 Hz_10ms_envelope_correlation": 0.9811176595009176, "1000–3000 Hz_10ms_modulation_mae_db": 0.2616444512684824, "1000–3000 Hz_40ms_log_rms_mae_db": 1.291518661148537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0862725700100784, "1000–3000 Hz_40ms_envelope_correlation": 0.9949903728185884, "1000–3000 Hz_40ms_modulation_mae_db": 0.4837710001708282, "3000–6000 Hz_10ms_log_rms_mae_db": 2.564154722157067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1791791722364978, "3000–6000 Hz_10ms_envelope_correlation": 0.9792188987865978, "3000–6000 Hz_10ms_modulation_mae_db": 0.6564227454835453, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6044653678113308, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10354336101065045, "3000–6000 Hz_40ms_envelope_correlation": 0.9928314743575446, "3000–6000 Hz_40ms_modulation_mae_db": 0.6141477240770218, "mr_stft_db": 5.8015510501006835, "band_mae_db": 0.19442709431829017, "rms_gain_error_db": -0.06289093136552103, "output_peak": 0.5995463633634265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.132421698997253, "spectral_convergence": 0.16678962441142856, "spectral_convergence_status": "ok", "amplitude_floor": 1.2959192499433599e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.699483152129029, "spectral_convergence": 0.1709550213830871, "spectral_convergence_status": "ok", "amplitude_floor": 8.646137696849433e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.628873715170744, "spectral_convergence": 0.16741773708795232, "spectral_convergence_status": "ok", "amplitude_floor": 6.578152434885499e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0020500598531285057, "output_power": 0.0020900599855292316, "delta_db": 0.08392209895585359, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0025325885847670413, "output_power": 0.0024986928356449793, "delta_db": -0.05851773451180104, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009745321457083826, "output_power": 0.0009512448801369377, "delta_db": -0.10503836942675476, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029510461183058167, "output_power": 0.0002713392473701347, "delta_db": -0.36463380574851906, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.807836063095196e-05, "output_power": 3.427749182314846e-05, "delta_db": -0.45669206735790774, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.519436277144243e-05, "output_power": 3.604525755905362e-05, "delta_db": 0.10375026812108121, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19542572402031955, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.032293540290158636, "pitch": null, "output_rms_dbfs": -22.30500106576415, "output_peak": 0.5472523477106392, "reference_rms_dbfs": -22.27270752547399, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8202595220990085}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019946925088635918, "log_rms_mae_db": 0.860328810808378, "rms_envelope_nrmse": 0.050368308695589205, "envelope_correlation": 0.9969509707976757, "rms_gain_error_db": -0.031594542593165124, "reference_mean_power": 0.0059224241180442505, "output_mean_power": 0.00587949534297124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0772791897966612e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010245127053592495, "output_power": 0.001002281666385324, "reference_active": true, "delta_db": -0.09527562247911597}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001329567656774176, "output_power": 0.00012724841753982764, "reference_active": true, "delta_db": -0.1905805170760094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.301896656350168e-05, "output_power": 2.2424208763143915e-05, "reference_active": true, "delta_db": -0.11368694148941372}], "active_band_count": 3, "active_band_mae_db": 0.13318102701484635}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001930447992601615, "log_rms_mae_db": 0.4456735854848855, "rms_envelope_nrmse": 0.02680667849942169, "envelope_correlation": 0.9990692479063576, "rms_gain_error_db": -0.029938573366117473, "reference_mean_power": 0.005919423744261807, "output_mean_power": 0.005878757866423838}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.875987153432414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009473592447040891, "output_power": 0.0009290912995251098, "reference_active": true, "delta_db": -0.08456304345393648}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.123049684235515e-05, "output_power": 5.960236235620108e-05, "reference_active": true, "delta_db": -0.11704309979967129}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.329909700774956e-06, "output_power": 1.3326002359954232e-06, "reference_active": true, "delta_db": 0.008777319334781418}], "active_band_count": 3, "active_band_mae_db": 0.0701278208627964}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015110036587973006, "log_rms_mae_db": 1.9367233795491445, "rms_envelope_nrmse": 0.08154523232167245, "envelope_correlation": 0.9943902476480814, "rms_gain_error_db": -0.14756522046347637, "reference_mean_power": 0.0025365732398871734, "output_mean_power": 0.002451832990734188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.31579034521966e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006679887319597139, "output_power": 0.0006287777950702431, "reference_active": true, "delta_db": -0.2627194015397577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.085773760663629e-05, "output_power": 7.651373785251963e-05, "reference_active": true, "delta_db": -0.7462249985449172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6589878271689645e-05, "output_power": 1.3291286943057728e-05, "reference_active": true, "delta_db": -0.9627616539237385}], "active_band_count": 3, "active_band_mae_db": 0.6572353513361378}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013775289901134618, "log_rms_mae_db": 1.1657753500828023, "rms_envelope_nrmse": 0.05010244055107497, "envelope_correlation": 0.9978967187912782, "rms_gain_error_db": -0.14584610881660373, "reference_mean_power": 0.0025400163932975937, "output_mean_power": 0.0024561331609932015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2092894134690622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006291541070562532, "output_power": 0.0005939918205406909, "reference_active": true, "delta_db": -0.24976571333580594}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.975083640469782e-05, "output_power": 3.492211776389266e-05, "reference_active": true, "delta_db": -0.5624569863175121}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.7844516380393e-07, "output_power": 6.564899052403937e-07, "reference_active": true, "delta_db": -1.733084385932672}], "active_band_count": 3, "active_band_mae_db": 0.84843569519533}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010600794217574433, "log_rms_mae_db": 2.0826206040365736, "rms_envelope_nrmse": 0.16654121722455698, "envelope_correlation": 0.980114239783992, "rms_gain_error_db": -0.3350739455936797, "reference_mean_power": 0.0002691478602162077, "output_mean_power": 0.00024916299173653034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.02821083589085e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010735336162950012, "output_power": 9.653597088167745e-05, "reference_active": true, "delta_db": -0.4612647961855116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.315168535027072e-06, "output_power": 5.9758523581185465e-06, "reference_active": true, "delta_db": -1.4347119701129154}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.408982675283079e-06, "output_power": 4.309212843716888e-06, "reference_active": true, "delta_db": 2.525642689137957}], "active_band_count": 3, "active_band_mae_db": 1.4738731518121277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.461607013685591e-05, "log_rms_mae_db": 1.3135574465649407, "rms_envelope_nrmse": 0.08678817288389172, "envelope_correlation": 0.9951351130796886, "rms_gain_error_db": -0.33119308173816114, "reference_mean_power": 0.00026921330106685756, "output_mean_power": 0.0002494463796489217}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9187018796507895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.751052870210862e-05, "output_power": 8.934470880689602e-05, "reference_active": true, "delta_db": -0.3798267347486982}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.818200149529557e-06, "output_power": 2.5262151866174314e-06, "reference_active": true, "delta_db": -1.793883486924859}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4850622077020166e-07, "output_power": 2.1376519614638088e-07, "reference_active": true, "delta_db": 1.5819235146047652}], "active_band_count": 3, "active_band_mae_db": 1.251877912092774}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.633594187238404e-05, "log_rms_mae_db": 2.629450537330441, "rms_envelope_nrmse": 0.2044862546790846, "envelope_correlation": 0.9729838253487338, "rms_gain_error_db": -0.03011088469526755, "reference_mean_power": 4.9508565949953196e-05, "output_mean_power": 4.916649603709091e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.164436282244953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0281496224399232e-05, "output_power": 9.918391678645733e-06, "reference_active": true, "delta_db": -0.15615065728671956}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.187880808785332e-06, "output_power": 3.6944968088662045e-06, "reference_active": true, "delta_db": -0.5443901696211996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.689088055617476e-07, "output_power": 5.10796585915203e-07, "reference_active": true, "delta_db": -0.46794669978683473}], "active_band_count": 3, "active_band_mae_db": 0.3894958422315846}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.3561893402191595e-05, "log_rms_mae_db": 1.6563446837898803, "rms_envelope_nrmse": 0.1171123564755427, "envelope_correlation": 0.9907636848589024, "rms_gain_error_db": -0.030161087519751807, "reference_mean_power": 4.9751584691373346e-05, "output_mean_power": 4.9407264554388314e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0334524009764534e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.035252748280077e-05, "output_power": 1.0011102063179075e-05, "reference_active": true, "delta_db": -0.1456450306625642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2970706912154704e-06, "output_power": 2.005258307992738e-06, "reference_active": true, "delta_db": -0.5900403625592999}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.66930165214842e-08, "output_power": 3.52246205563893e-08, "reference_active": true, "delta_db": -1.2240560814301755}], "active_band_count": 3, "active_band_mae_db": 0.6532471582173466}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9038492259955145, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "67fc71662647d13343a26dcb92e8e6713486a89c12befd34042d3559614871f3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9814170949970318, "standard_deviation": 0.0025466931844512425, "minimum": 0.9757498459268638, "maximum": 0.9834114891323125, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.059684525709528, "standard_deviation": 0.04809209823001966, "minimum": 1.9863962061414682, "maximum": 2.121037498181448, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9552712214691255, "standard_deviation": 0.37961602611751916, "minimum": 0.2616444512684824, "maximum": 1.4738731518121277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16274895888071816, "standard_deviation": 0.00898701767527449, "minimum": 0.15223689747135855, "maximum": 0.18266120414768813, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942899436814221, "standard_deviation": 0.001059668051500597, "minimum": 0.9925499655033638, "maximum": 0.9955119976811605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3246212195538174, "standard_deviation": 0.03187034139146464, "minimum": 1.2905699381088145, "maximum": 1.3709349815508698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8655927177334942, "standard_deviation": 0.2984660578769116, "minimum": 0.4837710001708282, "maximum": 1.251877912092774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09579477910077924, "standard_deviation": 0.0076779739648961025, "minimum": 0.0862725700100784, "maximum": 0.10574103859995457, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9780055733406519, "standard_deviation": 0.004184948437390911, "minimum": 0.9710288784372786, "maximum": 0.9834523176546606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.6175725415109294, "standard_deviation": 0.042889291511979906, "minimum": 2.564154722157067, "maximum": 2.7021553718163647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7415140962096998, "standard_deviation": 0.3165837469142048, "minimum": 0.2999662213622097, "maximum": 1.1565369479922334, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18411081014634578, "standard_deviation": 0.016493408632925276, "minimum": 0.16304584076983378, "maximum": 0.21100900912886594, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927015786411049, "standard_deviation": 0.001753176141723861, "minimum": 0.9896676961294543, "maximum": 0.9953285070908964, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6562108170723815, "standard_deviation": 0.05052321333440984, "minimum": 1.6044653678113308, "maximum": 1.7705331149315142, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7979412193976965, "standard_deviation": 0.30963436070430705, "minimum": 0.351568669222354, "maximum": 1.261252457034985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10556150169338506, "standard_deviation": 0.010865804121585628, "minimum": 0.09201009953557346, "maximum": 0.12551236905345522, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9947248133080826, "standard_deviation": 0.00039853379168835925, "minimum": 0.9941575200261565, "maximum": 0.9954293053671467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8423142095329688, "standard_deviation": 0.06428875467801452, "minimum": 1.76276787079034, "maximum": 1.9367233795491445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6920052266021657, "standard_deviation": 0.13507125862913685, "minimum": 0.4748221580147429, "maximum": 0.8883205646289101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07970649372445487, "standard_deviation": 0.0032262885254196864, "minimum": 0.0741146698748035, "maximum": 0.08433425157070991, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983082566043193, "standard_deviation": 0.00020622714182449723, "minimum": 0.9978967187912782, "maximum": 0.9985280690214586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.120536997530452, "standard_deviation": 0.053899954947533725, "minimum": 1.0547175515017166, "maximum": 1.218738227156685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8485255250501385, "standard_deviation": 0.14419388168317857, "minimum": 0.6149808659847804, "maximum": 1.0036382190178326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0465900105146503, "standard_deviation": 0.002685534132827808, "minimum": 0.04321743757324506, "maximum": 0.05010244055107497, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21489091101397437, "standard_deviation": 0.031133707937001275, "minimum": 0.1847371244634847, "maximum": 0.2815385352038042, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.806734846174761, "standard_deviation": 0.009626453023538413, "minimum": 5.789951337852298, "maximum": 5.8202595220990085, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5610314216598451, "standard_deviation": 0.020317049768842334, "minimum": 0.5409381756255487, "maximum": 0.5995463633634265, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06144730316560633, "standard_deviation": 0.01551824840422035, "minimum": -0.08283323564875356, "maximum": -0.032293540290158636, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967957924523613, "standard_deviation": 0.00018571239549181935, "minimum": 0.9965531531376421, "maximum": 0.9970985978531505, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8642743775921007, "standard_deviation": 0.027451736563166174, "minimum": 0.814078697998297, "maximum": 0.8987812074846897, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26183538556247277, "standard_deviation": 0.10442664780840473, "minimum": 0.12811169774710074, "maximum": 0.38598104513008447, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05191269274049019, "standard_deviation": 0.0015323246736104093, "minimum": 0.04952491212447185, "maximum": 0.05399821342840262, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990733461797059, "standard_deviation": 0.00010129924291611376, "minimum": 0.9989176635536222, "maximum": 0.9992214650558242, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44346988244814045, "standard_deviation": 0.02181270482139511, "minimum": 0.41768966406431984, "maximum": 0.4786953700763545, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2912314719046003, "standard_deviation": 0.12389838062040535, "minimum": 0.0701278208627964, "maximum": 0.4704182272194229, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02743226298336574, "standard_deviation": 0.001389202107048212, "minimum": 0.025459694092527373, "maximum": 0.02948419982950831, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0023", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.20560622215271, "predicted_mos": 3.20560622215271, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7dd663491cfe9e0941f8603923f41ff8f9eb8711f29620f252732d88645b2c4f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8145188000053167, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9208450370405737, "全帯域_10ms_rms_envelope_nrmse": 0.052885263164931404, "全帯域_10ms_envelope_correlation": 0.9966312460340498, "全帯域_10ms_modulation_mae_db": 0.4215168402045039, "全帯域_40ms_log_rms_mae_db": 0.5173725576115058, "全帯域_40ms_rms_envelope_nrmse": 0.02816054022415803, "全帯域_40ms_envelope_correlation": 0.9989617838282034, "全帯域_40ms_modulation_mae_db": 0.27277711686067646, "300–1000 Hz_10ms_log_rms_mae_db": 1.4435171156258069, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07582953340084365, "300–1000 Hz_10ms_envelope_correlation": 0.9950887954975443, "300–1000 Hz_10ms_modulation_mae_db": 0.6906853286554657, "300–1000 Hz_40ms_log_rms_mae_db": 0.8328848465984876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03847232394220792, "300–1000 Hz_40ms_envelope_correlation": 0.9987502758179756, "300–1000 Hz_40ms_modulation_mae_db": 0.5729782838832443, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6045556954203755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14514630860108793, "1000–3000 Hz_10ms_envelope_correlation": 0.9849474355843234, "1000–3000 Hz_10ms_modulation_mae_db": 0.5670311492486261, "1000–3000 Hz_40ms_log_rms_mae_db": 1.06138912975228, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08777838279688537, "1000–3000 Hz_40ms_envelope_correlation": 0.9945570897971525, "1000–3000 Hz_40ms_modulation_mae_db": 0.33628190143215725, "3000–6000 Hz_10ms_log_rms_mae_db": 1.878392079662793, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17463754231927478, "3000–6000 Hz_10ms_envelope_correlation": 0.9804617251560044, "3000–6000 Hz_10ms_modulation_mae_db": 0.41773385633431154, "3000–6000 Hz_40ms_log_rms_mae_db": 1.183899119447847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10046479058937505, "3000–6000 Hz_40ms_envelope_correlation": 0.9932842665300234, "3000–6000 Hz_40ms_modulation_mae_db": 0.6137999741430734, "mr_stft_db": 5.363582454232284, "band_mae_db": 0.11906386045521179, "rms_gain_error_db": -0.018242055373987036, "output_peak": 0.6669557940442419}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9382054492184687, "全帯域_10ms_rms_envelope_nrmse": 0.05249844558753635, "全帯域_10ms_envelope_correlation": 0.9967078895268664, "全帯域_10ms_modulation_mae_db": 0.25188147787488524, "全帯域_40ms_log_rms_mae_db": 0.5310791558944065, "全帯域_40ms_rms_envelope_nrmse": 0.026730650472678797, "全帯域_40ms_envelope_correlation": 0.9991063094127464, "全帯域_40ms_modulation_mae_db": 0.3817130208431651, "300–1000 Hz_10ms_log_rms_mae_db": 1.4759760350991757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07497967852058719, "300–1000 Hz_10ms_envelope_correlation": 0.995359931225995, "300–1000 Hz_10ms_modulation_mae_db": 0.9145877684210362, "300–1000 Hz_40ms_log_rms_mae_db": 0.84141855204751, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043809870653153206, "300–1000 Hz_40ms_envelope_correlation": 0.9985454029476565, "300–1000 Hz_40ms_modulation_mae_db": 1.1099063104704845, "1000–3000 Hz_10ms_log_rms_mae_db": 1.596514494364516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16122650048183076, "1000–3000 Hz_10ms_envelope_correlation": 0.9816579295499767, "1000–3000 Hz_10ms_modulation_mae_db": 0.6310019458478527, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0469880212717657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.099740224002723, "1000–3000 Hz_40ms_envelope_correlation": 0.9934173728818048, "1000–3000 Hz_40ms_modulation_mae_db": 0.5973543395228909, "3000–6000 Hz_10ms_log_rms_mae_db": 1.833614101143772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17697390951019873, "3000–6000 Hz_10ms_envelope_correlation": 0.9798886115160825, "3000–6000 Hz_10ms_modulation_mae_db": 0.5008355859723895, "3000–6000 Hz_40ms_log_rms_mae_db": 1.214759840184424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10067548009007282, "3000–6000 Hz_40ms_envelope_correlation": 0.9936763876683877, "3000–6000 Hz_40ms_modulation_mae_db": 0.5604513672337502, "mr_stft_db": 5.373265185600739, "band_mae_db": 0.23604204796503145, "rms_gain_error_db": -0.054480431092901106, "output_peak": 0.6297110256209114}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8949876770213201, "全帯域_10ms_rms_envelope_nrmse": 0.0553335244813198, "全帯域_10ms_envelope_correlation": 0.9963633004024326, "全帯域_10ms_modulation_mae_db": 0.17436272668400424, "全帯域_40ms_log_rms_mae_db": 0.5191001330050323, "全帯域_40ms_rms_envelope_nrmse": 0.031175821600913523, "全帯域_40ms_envelope_correlation": 0.9988039439641979, "全帯域_40ms_modulation_mae_db": 0.21669998898537723, "300–1000 Hz_10ms_log_rms_mae_db": 1.531178263107572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08048424719680378, "300–1000 Hz_10ms_envelope_correlation": 0.9946312590894592, "300–1000 Hz_10ms_modulation_mae_db": 0.5715817954499204, "300–1000 Hz_40ms_log_rms_mae_db": 0.9374176280943745, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0471137513350663, "300–1000 Hz_40ms_envelope_correlation": 0.9982920764013989, "300–1000 Hz_40ms_modulation_mae_db": 0.5303000896667761, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5601402700132003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1734675871955883, "1000–3000 Hz_10ms_envelope_correlation": 0.978241832158037, "1000–3000 Hz_10ms_modulation_mae_db": 1.0828942678220674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0068682156422073, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10365203130820913, "1000–3000 Hz_40ms_envelope_correlation": 0.9922386184322498, "1000–3000 Hz_40ms_modulation_mae_db": 0.7509788974781824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7968315446972314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1685608435170101, "3000–6000 Hz_10ms_envelope_correlation": 0.9819422362299371, "3000–6000 Hz_10ms_modulation_mae_db": 0.5269497523705399, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1160934724370977, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08678581352781752, "3000–6000 Hz_40ms_envelope_correlation": 0.995078034428377, "3000–6000 Hz_40ms_modulation_mae_db": 0.45159158337485517, "mr_stft_db": 5.343829022653392, "band_mae_db": 0.1857086577407782, "rms_gain_error_db": -0.06878224291492717, "output_peak": 0.5669085890884311}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8864792024561288, "全帯域_10ms_rms_envelope_nrmse": 0.050484401026292834, "全帯域_10ms_envelope_correlation": 0.9969836107660993, "全帯域_10ms_modulation_mae_db": 0.4523872771295214, "全帯域_40ms_log_rms_mae_db": 0.48576351143511104, "全帯域_40ms_rms_envelope_nrmse": 0.026171541316893414, "全帯域_40ms_envelope_correlation": 0.9991740120871891, "全帯域_40ms_modulation_mae_db": 0.39335278855924466, "300–1000 Hz_10ms_log_rms_mae_db": 1.4336126139917267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07629627317878603, "300–1000 Hz_10ms_envelope_correlation": 0.9952881055208119, "300–1000 Hz_10ms_modulation_mae_db": 1.0308909816974012, "300–1000 Hz_40ms_log_rms_mae_db": 0.853623074000744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04417010591535663, "300–1000 Hz_40ms_envelope_correlation": 0.9986305855690333, "300–1000 Hz_40ms_modulation_mae_db": 1.1288495698856884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5921184429025037, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.146621866964117, "1000–3000 Hz_10ms_envelope_correlation": 0.9852724896845118, "1000–3000 Hz_10ms_modulation_mae_db": 0.6227794005258294, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0278911587073662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09096064028452729, "1000–3000 Hz_40ms_envelope_correlation": 0.9950888249254722, "1000–3000 Hz_40ms_modulation_mae_db": 0.4168243100763928, "3000–6000 Hz_10ms_log_rms_mae_db": 1.860372459356046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16484031590678488, "3000–6000 Hz_10ms_envelope_correlation": 0.9831400047415472, "3000–6000 Hz_10ms_modulation_mae_db": 0.3686880941886718, "3000–6000 Hz_40ms_log_rms_mae_db": 1.158571585584882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09540723403640544, "3000–6000 Hz_40ms_envelope_correlation": 0.9943469096383852, "3000–6000 Hz_40ms_modulation_mae_db": 0.5377354335914221, "mr_stft_db": 5.3485323614909, "band_mae_db": 0.25021415780670836, "rms_gain_error_db": -0.06927350013762103, "output_peak": 0.5816211094598698}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8834086306057111, "全帯域_10ms_rms_envelope_nrmse": 0.04892807704357428, "全帯域_10ms_envelope_correlation": 0.9971856755209318, "全帯域_10ms_modulation_mae_db": 0.208723993370017, "全帯域_40ms_log_rms_mae_db": 0.4446160875407067, "全帯域_40ms_rms_envelope_nrmse": 0.026977095946517884, "全帯域_40ms_envelope_correlation": 0.9991387306099574, "全帯域_40ms_modulation_mae_db": 0.2730951976470541, "300–1000 Hz_10ms_log_rms_mae_db": 1.4648148123382356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07461712541213206, "300–1000 Hz_10ms_envelope_correlation": 0.9954428586932086, "300–1000 Hz_10ms_modulation_mae_db": 0.5452914439020453, "300–1000 Hz_40ms_log_rms_mae_db": 0.8340185287742026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04565461839858915, "300–1000 Hz_40ms_envelope_correlation": 0.9984228756255, "300–1000 Hz_40ms_modulation_mae_db": 0.6455628832383355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6297665378415835, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1547511260309665, "1000–3000 Hz_10ms_envelope_correlation": 0.983540875736341, "1000–3000 Hz_10ms_modulation_mae_db": 0.5325933651392826, "1000–3000 Hz_40ms_log_rms_mae_db": 1.068386804250901, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09498406310595783, "1000–3000 Hz_40ms_envelope_correlation": 0.9947033850492164, "1000–3000 Hz_40ms_modulation_mae_db": 0.7223987778085118, "3000–6000 Hz_10ms_log_rms_mae_db": 1.823499754702756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1836050739342081, "3000–6000 Hz_10ms_envelope_correlation": 0.9786931384684202, "3000–6000 Hz_10ms_modulation_mae_db": 0.48468570020698026, "3000–6000 Hz_40ms_log_rms_mae_db": 1.160888072182725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10669011618823981, "3000–6000 Hz_40ms_envelope_correlation": 0.9925980759646023, "3000–6000 Hz_40ms_modulation_mae_db": 0.6277701778110347, "mr_stft_db": 5.369970421823832, "band_mae_db": 0.26427074867659334, "rms_gain_error_db": -0.07251724805104834, "output_peak": 0.6291063181772688}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.907410667586478, "全帯域_10ms_rms_envelope_nrmse": 0.05124220909911122, "全帯域_10ms_envelope_correlation": 0.9968487655984429, "全帯域_10ms_modulation_mae_db": 0.41617117825322003, "全帯域_40ms_log_rms_mae_db": 0.4580344748462526, "全帯域_40ms_rms_envelope_nrmse": 0.026121088664308352, "全帯域_40ms_envelope_correlation": 0.999123218858809, "全帯域_40ms_modulation_mae_db": 0.3955415984595592, "300–1000 Hz_10ms_log_rms_mae_db": 1.5210856698117876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07587518962200016, "300–1000 Hz_10ms_envelope_correlation": 0.9951094816730975, "300–1000 Hz_10ms_modulation_mae_db": 0.8731770093516321, "300–1000 Hz_40ms_log_rms_mae_db": 0.9185709815420733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04107530932785896, "300–1000 Hz_40ms_envelope_correlation": 0.9985977985669444, "300–1000 Hz_40ms_modulation_mae_db": 0.9304022697200777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5684275873293727, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13404339821853797, "1000–3000 Hz_10ms_envelope_correlation": 0.9873589829867527, "1000–3000 Hz_10ms_modulation_mae_db": 0.30375425207786105, "1000–3000 Hz_40ms_log_rms_mae_db": 0.991464027670377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07394517191027511, "1000–3000 Hz_40ms_envelope_correlation": 0.9965469368221576, "1000–3000 Hz_40ms_modulation_mae_db": 0.37386884980462565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.831430473525575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16679024767013384, "3000–6000 Hz_10ms_envelope_correlation": 0.982100064212488, "3000–6000 Hz_10ms_modulation_mae_db": 0.2257597134962408, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1815727991001652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08985309429035182, "3000–6000 Hz_40ms_envelope_correlation": 0.9949474374392776, "3000–6000 Hz_40ms_modulation_mae_db": 0.19321585714492395, "mr_stft_db": 5.360863860247631, "band_mae_db": 0.20077094030386833, "rms_gain_error_db": -0.035710605161116, "output_peak": 0.5409802682794007}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8670261188420453, "全帯域_10ms_rms_envelope_nrmse": 0.050103532362725994, "全帯域_10ms_envelope_correlation": 0.9969815525502966, "全帯域_10ms_modulation_mae_db": 0.24684298257571213, "全帯域_40ms_log_rms_mae_db": 0.4357799170008516, "全帯域_40ms_rms_envelope_nrmse": 0.025135356395318567, "全帯域_40ms_envelope_correlation": 0.9991837085960176, "全帯域_40ms_modulation_mae_db": 0.5244575942971273, "300–1000 Hz_10ms_log_rms_mae_db": 1.4638137447265203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07532444017112619, "300–1000 Hz_10ms_envelope_correlation": 0.9951649229740998, "300–1000 Hz_10ms_modulation_mae_db": 0.5262874000487697, "300–1000 Hz_40ms_log_rms_mae_db": 0.8720143744673146, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041345141527997044, "300–1000 Hz_40ms_envelope_correlation": 0.9985611871186922, "300–1000 Hz_40ms_modulation_mae_db": 0.9337926232015272, "1000–3000 Hz_10ms_log_rms_mae_db": 1.584419907201153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1500460256413246, "1000–3000 Hz_10ms_envelope_correlation": 0.9847394961200671, "1000–3000 Hz_10ms_modulation_mae_db": 0.6643132133854711, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0544657144927307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09025926454920283, "1000–3000 Hz_40ms_envelope_correlation": 0.9955674592272586, "1000–3000 Hz_40ms_modulation_mae_db": 0.6364813997076366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8389515521591782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17241740709888084, "3000–6000 Hz_10ms_envelope_correlation": 0.9819565556400089, "3000–6000 Hz_10ms_modulation_mae_db": 0.6113648878409729, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1701086657046156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09430870750236765, "3000–6000 Hz_40ms_envelope_correlation": 0.9947734279427259, "3000–6000 Hz_40ms_modulation_mae_db": 0.5784508314184025, "mr_stft_db": 5.3541713603530345, "band_mae_db": 0.13455829028810254, "rms_gain_error_db": -0.02332993274978276, "output_peak": 0.5512135703479184}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.926098922397453, "全帯域_10ms_rms_envelope_nrmse": 0.05177122039807671, "全帯域_10ms_envelope_correlation": 0.9968772776092175, "全帯域_10ms_modulation_mae_db": 0.25138966692674464, "全帯域_40ms_log_rms_mae_db": 0.482008401310357, "全帯域_40ms_rms_envelope_nrmse": 0.02762285561827317, "全帯域_40ms_envelope_correlation": 0.9991290141412206, "全帯域_40ms_modulation_mae_db": 0.2719717748576768, "300–1000 Hz_10ms_log_rms_mae_db": 1.5296056671034908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07872241493624696, "300–1000 Hz_10ms_envelope_correlation": 0.9949693009499616, "300–1000 Hz_10ms_modulation_mae_db": 0.6061338286872066, "300–1000 Hz_40ms_log_rms_mae_db": 0.915178141005911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04573822571374601, "300–1000 Hz_40ms_envelope_correlation": 0.9985082429544825, "300–1000 Hz_40ms_modulation_mae_db": 0.6553130090620628, "1000–3000 Hz_10ms_log_rms_mae_db": 1.644597295005985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15345057682056773, "1000–3000 Hz_10ms_envelope_correlation": 0.984315060108292, "1000–3000 Hz_10ms_modulation_mae_db": 0.8312498879554305, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0384346807673464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09559903310388553, "1000–3000 Hz_40ms_envelope_correlation": 0.9951637959307027, "1000–3000 Hz_40ms_modulation_mae_db": 0.7992555755699601, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8112751853175286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1698190602097379, "3000–6000 Hz_10ms_envelope_correlation": 0.9837468509706798, "3000–6000 Hz_10ms_modulation_mae_db": 0.5497729366177978, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1574458199198756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10162037354133738, "3000–6000 Hz_40ms_envelope_correlation": 0.9949548475343102, "3000–6000 Hz_40ms_modulation_mae_db": 0.7507101715426826, "mr_stft_db": 5.347246631550042, "band_mae_db": 0.2310415869689054, "rms_gain_error_db": -0.08690048477931007, "output_peak": 0.5515997631628988}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.574812083387685, "spectral_convergence": 0.16451597009637844, "spectral_convergence_status": "ok", "amplitude_floor": 1.2959192499433599e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.153322991621885, "spectral_convergence": 0.16151446410177808, "spectral_convergence_status": "ok", "amplitude_floor": 8.646137696849433e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.362612287687282, "spectral_convergence": 0.1600698842361081, "spectral_convergence_status": "ok", "amplitude_floor": 6.578152434885499e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0020500598531285057, "output_power": 0.002080769235793893, "delta_db": 0.06457377370699315, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0025325885847670413, "output_power": 0.002514109658979673, "delta_db": -0.03180428462043715, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009745321457083826, "output_power": 0.0009522458606805594, "delta_db": -0.10047075733033392, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029510461183058167, "output_power": 0.00028131390572089, "delta_db": -0.2078479616228915, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.807836063095196e-05, "output_power": 3.6275572243259994e-05, "delta_db": -0.21063970534194248, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.519436277144243e-05, "output_power": 3.600624019386691e-05, "delta_db": 0.0990466801086726, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11906386045521179, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.018242055373987036, "pitch": null, "output_rms_dbfs": -22.290949580847975, "output_peak": 0.6669557940442419, "reference_rms_dbfs": -22.27270752547399, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.363582454232284}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019946925088635918, "log_rms_mae_db": 0.9208450370405737, "rms_envelope_nrmse": 0.052885263164931404, "envelope_correlation": 0.9966312460340498, "rms_gain_error_db": -0.01795393026332267, "reference_mean_power": 0.0059224241180442505, "output_mean_power": 0.005897991087361335}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0772791897966612e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010245127053592495, "output_power": 0.001028428629283451, "reference_active": true, "delta_db": 0.016568093649137142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001329567656774176, "output_power": 0.00011927970090219908, "reference_active": true, "delta_db": -0.47143900429216423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.301896656350168e-05, "output_power": 1.9250033064929e-05, "reference_active": true, "delta_db": -0.7765434226722103}], "active_band_count": 3, "active_band_mae_db": 0.4215168402045039}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001930447992601615, "log_rms_mae_db": 0.5173725576115058, "rms_envelope_nrmse": 0.02816054022415803, "envelope_correlation": 0.9989617838282034, "rms_gain_error_db": -0.01633978573538515, "reference_mean_power": 0.005919423744261807, "output_mean_power": 0.005897194497693676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.875987153432414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009473592447040891, "output_power": 0.0009501569055571178, "reference_active": true, "delta_db": 0.012806315782192174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.123049684235515e-05, "output_power": 5.514514174483444e-05, "reference_active": true, "delta_db": -0.4546052607560295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.329909700774956e-06, "output_power": 1.2266768703761627e-06, "reference_active": true, "delta_db": -0.35091977404380775}], "active_band_count": 3, "active_band_mae_db": 0.27277711686067646}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015110036587973006, "log_rms_mae_db": 1.4435171156258069, "rms_envelope_nrmse": 0.07582953340084365, "envelope_correlation": 0.9950887954975443, "rms_gain_error_db": -0.10399533512025416, "reference_mean_power": 0.0025365732398871734, "output_mean_power": 0.0024765544046257268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.31579034521966e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006679887319597139, "output_power": 0.000643035061137671, "reference_active": true, "delta_db": -0.1653448335312235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.085773760663629e-05, "output_power": 7.771995855788227e-05, "reference_active": true, "delta_db": -0.6782935815828195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6589878271689645e-05, "output_power": 1.2502631823953814e-05, "reference_active": true, "delta_db": -1.228417570852354}], "active_band_count": 3, "active_band_mae_db": 0.6906853286554657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013775289901134618, "log_rms_mae_db": 0.8328848465984876, "rms_envelope_nrmse": 0.03847232394220792, "envelope_correlation": 0.9987502758179756, "rms_gain_error_db": -0.10318076371894346, "reference_mean_power": 0.0025400163932975937, "output_mean_power": 0.002480381269895413}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2092894134690622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006291541070562532, "output_power": 0.000606836543805364, "reference_active": true, "delta_db": -0.15685309786904428}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.975083640469782e-05, "output_power": 3.544477417295224e-05, "reference_active": true, "delta_db": -0.4979405745521134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.7844516380393e-07, "output_power": 7.658123628150789e-07, "reference_active": true, "delta_db": -1.0641411792285755}], "active_band_count": 3, "active_band_mae_db": 0.5729782838832443}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010600794217574433, "log_rms_mae_db": 1.6045556954203755, "rms_envelope_nrmse": 0.14514630860108793, "envelope_correlation": 0.9849474355843234, "rms_gain_error_db": -0.1634364388214676, "reference_mean_power": 0.0002691478602162077, "output_mean_power": 0.0002592073353537867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.02821083589085e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010735336162950012, "output_power": 0.00010619317711521832, "reference_active": true, "delta_db": -0.04719034015267752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.315168535027072e-06, "output_power": 7.242480704907727e-06, "reference_active": true, "delta_db": -0.5998370924251831}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.408982675283079e-06, "output_power": 3.0707204283968808e-06, "reference_active": true, "delta_db": 1.0540660151680177}], "active_band_count": 3, "active_band_mae_db": 0.5670311492486261}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.461607013685591e-05, "log_rms_mae_db": 1.06138912975228, "rms_envelope_nrmse": 0.08777838279688537, "envelope_correlation": 0.9945570897971525, "rms_gain_error_db": -0.154275283781724, "reference_mean_power": 0.00026921330106685756, "output_mean_power": 0.0002598178501871469}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9187018796507895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.751052870210862e-05, "output_power": 9.726431243395305e-05, "reference_active": true, "delta_db": -0.010979900992160363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.818200149529557e-06, "output_power": 3.2048598214946586e-06, "reference_active": true, "delta_db": -0.7604965180248007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4850622077020166e-07, "output_power": 1.5684893518331924e-07, "reference_active": true, "delta_db": 0.23736928527951073}], "active_band_count": 3, "active_band_mae_db": 0.33628190143215725}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.633594187238404e-05, "log_rms_mae_db": 1.878392079662793, "rms_envelope_nrmse": 0.17463754231927478, "envelope_correlation": 0.9804617251560044, "rms_gain_error_db": -0.005837367701249922, "reference_mean_power": 4.9508565949953196e-05, "output_mean_power": 4.944206601038084e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.164436282244953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0281496224399232e-05, "output_power": 1.088175614419977e-05, "reference_active": true, "delta_db": 0.24642668942427137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.187880808785332e-06, "output_power": 4.186775306267286e-06, "reference_active": true, "delta_db": -0.0011465871302901549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.689088055617476e-07, "output_power": 7.171425399929688e-07, "reference_active": true, "delta_db": 1.005628292448373}], "active_band_count": 3, "active_band_mae_db": 0.41773385633431154}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.3561893402191595e-05, "log_rms_mae_db": 1.183899119447847, "rms_envelope_nrmse": 0.10046479058937505, "envelope_correlation": 0.9932842665300234, "rms_gain_error_db": -0.005857246121386442, "reference_mean_power": 4.9751584691373346e-05, "output_mean_power": 4.9684530913529144e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0334524009764534e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.035252748280077e-05, "output_power": 1.115175889796011e-05, "reference_active": true, "delta_db": 0.322969792740589}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2970706912154704e-06, "output_power": 2.1933433377993088e-06, "reference_active": true, "delta_db": -0.20067740758671823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.66930165214842e-08, "output_power": 6.324514664351544e-08, "reference_active": true, "delta_db": 1.3177527221019127}], "active_band_count": 3, "active_band_mae_db": 0.6137999741430734}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.921745853218186, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "23ad6eadf0d05c96577de9a77ec5107c507ebb9dcbeb9c5b66fd287f1694de82", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9837592627410376, "standard_deviation": 0.0027473285421115495, "minimum": 0.978241832158037, "maximum": 0.9873589829867527, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.597567528759836, "standard_deviation": 0.02866308433935901, "minimum": 1.5601402700132003, "maximum": 1.644597295005985, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6544521852503027, "standard_deviation": 0.22739249095600508, "minimum": 0.30375425207786105, "maximum": 1.0828942678220674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1523441737442526, "standard_deviation": 0.011691668573559555, "minimum": 0.13404339821853797, "maximum": 0.1734675871955883, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946604353832518, "standard_deviation": 0.0013224110061248726, "minimum": 0.9922386184322498, "maximum": 0.9965469368221576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.036985969069372, "standard_deviation": 0.026862745873569797, "minimum": 0.991464027670377, "maximum": 1.068386804250901, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5791805064250447, "standard_deviation": 0.18110179466280424, "minimum": 0.33628190143215725, "maximum": 0.7992555755699601, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09211485138270825, "standard_deviation": 0.008986834371052384, "minimum": 0.07394517191027511, "maximum": 0.10365203130820913, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.981491148366896, "standard_deviation": 0.0016926091513785708, "minimum": 0.9786931384684202, "maximum": 0.9837468509706798, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.83429589382061, "standard_deviation": 0.02592543869394197, "minimum": 1.7968315446972314, "maximum": 1.878392079662793, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.46072381587848804, "standard_deviation": 0.12113647511806581, "minimum": 0.2257597134962408, "maximum": 0.6113648878409729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17220555002077864, "standard_deviation": 0.006108509987296762, "minimum": 0.16484031590678488, "maximum": 0.1836050739342081, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9942074233932612, "standard_deviation": 0.000920268707137005, "minimum": 0.9925980759646023, "maximum": 0.995078034428377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.167917421820204, "standard_deviation": 0.028216080225325884, "minimum": 1.1160934724370977, "maximum": 1.214759840184424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5392156745325181, "standard_deviation": 0.16367259930463277, "minimum": 0.19321585714492395, "maximum": 0.7507101715426826, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09697570122074593, "standard_deviation": 0.006611064674491333, "minimum": 0.08678581352781752, "maximum": 0.10669011618823981, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9951318319530222, "standard_deviation": 0.000254565328497638, "minimum": 0.9946312590894592, "maximum": 0.9954428586932086, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4829504902255395, "standard_deviation": 0.039078480481811104, "minimum": 1.4336126139917267, "maximum": 1.531178263107572, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7198294445266846, "standard_deviation": 0.1934207238569576, "minimum": 0.5262874000487697, "maximum": 1.0308909816974012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07651611280481575, "standard_deviation": 0.0020329861923024627, "minimum": 0.07461712541213206, "maximum": 0.08048424719680378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9985385556252104, "standard_deviation": 0.00013753777411633047, "minimum": 0.9982920764013989, "maximum": 0.9987502758179756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8756407658163272, "standard_deviation": 0.042176765436766045, "minimum": 0.8328848465984876, "maximum": 0.9374176280943745, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8133881298910246, "standard_deviation": 0.2410445401288701, "minimum": 0.5303000896667761, "maximum": 1.1288495698856884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0434224183517469, "standard_deviation": 0.00290340220124226, "minimum": 0.03847232394220792, "maximum": 0.0471137513350663, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20270878627564992, "standard_deviation": 0.05331069686952867, "minimum": 0.11906386045521179, "maximum": 0.26427074867659334, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.357682662243982, "standard_deviation": 0.010925873272007475, "minimum": 5.343829022653392, "maximum": 5.373265185600739, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5897620547726177, "standard_deviation": 0.04631156820555998, "minimum": 0.5409802682794007, "maximum": 0.6669557940442419, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.053654562532586686, "standard_deviation": 0.02515718209411104, "minimum": -0.08690048477931007, "maximum": -0.018242055373987036, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968224147510422, "standard_deviation": 0.00025281816507883464, "minimum": 0.9963633004024326, "maximum": 0.9971856755209318, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9030577131460223, "standard_deviation": 0.02427795152275353, "minimum": 0.8670261188420453, "maximum": 0.9382054492184687, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3029095178773261, "standard_deviation": 0.10891432017468007, "minimum": 0.17436272668400424, "maximum": 0.4523872771295214, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05165583414544607, "standard_deviation": 0.001969548563693205, "minimum": 0.04892807704357428, "maximum": 0.0553335244813198, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990775901872927, "standard_deviation": 0.00012991261434720482, "minimum": 0.9988039439641979, "maximum": 0.9991837085960176, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4842192798305279, "standard_deviation": 0.0360938520458614, "minimum": 0.4357799170008516, "maximum": 0.5310791558944065, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.34120113506373506, "standard_deviation": 0.10036385827319674, "minimum": 0.21669998898537723, "maximum": 0.5244575942971273, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02726186877988272, "standard_deviation": 0.0018374083135289268, "minimum": 0.025135356395318567, "maximum": 0.031175821600913523, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0023", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.063091278076172, "predicted_mos": 3.063091278076172, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ba87e46e5e9a46737a4da4e3ea15dc8754956f15d51df837aabfe3b8a2b39524", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.526504899840802, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9017698426010031, "全帯域_10ms_rms_envelope_nrmse": 0.053888809210109606, "全帯域_10ms_envelope_correlation": 0.9965471913307175, "全帯域_10ms_modulation_mae_db": 0.29628918935910026, "全帯域_40ms_log_rms_mae_db": 0.502672202694853, "全帯域_40ms_rms_envelope_nrmse": 0.029642445024589265, "全帯域_40ms_envelope_correlation": 0.9989150410114203, "全帯域_40ms_modulation_mae_db": 0.479904951489664, "300–1000 Hz_10ms_log_rms_mae_db": 1.5855009235222113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07399731709624595, "300–1000 Hz_10ms_envelope_correlation": 0.9954671478632526, "300–1000 Hz_10ms_modulation_mae_db": 0.866167424658182, "300–1000 Hz_40ms_log_rms_mae_db": 1.0009230131735192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046740959448722155, "300–1000 Hz_40ms_envelope_correlation": 0.9982548153405065, "300–1000 Hz_40ms_modulation_mae_db": 0.8171183672597874, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6857637480066805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15794518156314433, "1000–3000 Hz_10ms_envelope_correlation": 0.9830774104332501, "1000–3000 Hz_10ms_modulation_mae_db": 1.0769011118907421, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0908427448224085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09922540072791544, "1000–3000 Hz_40ms_envelope_correlation": 0.9945459265923013, "1000–3000 Hz_40ms_modulation_mae_db": 0.8216888396907582, "3000–6000 Hz_10ms_log_rms_mae_db": 2.09269906650761, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17927116213424776, "3000–6000 Hz_10ms_envelope_correlation": 0.9794477440964126, "3000–6000 Hz_10ms_modulation_mae_db": 0.3000358549118031, "3000–6000 Hz_40ms_log_rms_mae_db": 1.327095279554849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10748649488461218, "3000–6000 Hz_40ms_envelope_correlation": 0.9923255808889148, "3000–6000 Hz_40ms_modulation_mae_db": 0.5610567730904704, "mr_stft_db": 5.48468344385656, "band_mae_db": 0.15334194060250528, "rms_gain_error_db": -0.06747970968911084, "output_peak": 0.5621440738210801}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8381405435751226, "全帯域_10ms_rms_envelope_nrmse": 0.051391966916152255, "全帯域_10ms_envelope_correlation": 0.9968408729122753, "全帯域_10ms_modulation_mae_db": 0.17758975661542054, "全帯域_40ms_log_rms_mae_db": 0.4215629394619232, "全帯域_40ms_rms_envelope_nrmse": 0.02649243927056662, "全帯域_40ms_envelope_correlation": 0.9991146756323208, "全帯域_40ms_modulation_mae_db": 0.18940615325076124, "300–1000 Hz_10ms_log_rms_mae_db": 1.616138337551173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08129741603034676, "300–1000 Hz_10ms_envelope_correlation": 0.9944363937455888, "300–1000 Hz_10ms_modulation_mae_db": 0.8326894585644995, "300–1000 Hz_40ms_log_rms_mae_db": 0.9821636081556387, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046037989367859154, "300–1000 Hz_40ms_envelope_correlation": 0.9982821003140899, "300–1000 Hz_40ms_modulation_mae_db": 0.8900794406915691, "1000–3000 Hz_10ms_log_rms_mae_db": 1.747916921490306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15184232124765437, "1000–3000 Hz_10ms_envelope_correlation": 0.9840301583944976, "1000–3000 Hz_10ms_modulation_mae_db": 1.3573343426595417, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1195039535182016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0888348403732693, "1000–3000 Hz_40ms_envelope_correlation": 0.995349264839506, "1000–3000 Hz_40ms_modulation_mae_db": 1.1278404860366182, "3000–6000 Hz_10ms_log_rms_mae_db": 2.138974817557384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17865613238191966, "3000–6000 Hz_10ms_envelope_correlation": 0.9806402734023935, "3000–6000 Hz_10ms_modulation_mae_db": 0.7870771685087212, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3704828984659543, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10510160154935613, "3000–6000 Hz_40ms_envelope_correlation": 0.9947139504526532, "3000–6000 Hz_40ms_modulation_mae_db": 0.7300613448273001, "mr_stft_db": 5.5007997549634835, "band_mae_db": 0.31126517022839256, "rms_gain_error_db": -0.04898467317251161, "output_peak": 0.6220660905076162}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8738435246059479, "全帯域_10ms_rms_envelope_nrmse": 0.054787492482433114, "全帯域_10ms_envelope_correlation": 0.9964077707115238, "全帯域_10ms_modulation_mae_db": 0.2110426347133619, "全帯域_40ms_log_rms_mae_db": 0.4578202394879418, "全帯域_40ms_rms_envelope_nrmse": 0.02887049628033464, "全帯域_40ms_envelope_correlation": 0.9989448008964527, "全帯域_40ms_modulation_mae_db": 0.13403068206312924, "300–1000 Hz_10ms_log_rms_mae_db": 1.5762032649358786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07895471251227341, "300–1000 Hz_10ms_envelope_correlation": 0.9947244230799681, "300–1000 Hz_10ms_modulation_mae_db": 0.7966592408163601, "300–1000 Hz_40ms_log_rms_mae_db": 0.9524940469650321, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04419244728752464, "300–1000 Hz_40ms_envelope_correlation": 0.9983903642653736, "300–1000 Hz_40ms_modulation_mae_db": 0.7805799160743944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7282243599572193, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16142308406792627, "1000–3000 Hz_10ms_envelope_correlation": 0.9816188282039032, "1000–3000 Hz_10ms_modulation_mae_db": 1.1391700293201736, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1383337081847167, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10438793906812623, "1000–3000 Hz_40ms_envelope_correlation": 0.9926779506363704, "1000–3000 Hz_40ms_modulation_mae_db": 1.0149639013612402, "3000–6000 Hz_10ms_log_rms_mae_db": 2.115678007411097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16321607746369007, "3000–6000 Hz_10ms_envelope_correlation": 0.9828348575598249, "3000–6000 Hz_10ms_modulation_mae_db": 0.6806921052353436, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2893114312770142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08592582499613152, "3000–6000 Hz_40ms_envelope_correlation": 0.9953471324207219, "3000–6000 Hz_40ms_modulation_mae_db": 0.7661608441814418, "mr_stft_db": 5.489665967860994, "band_mae_db": 0.20434397060952617, "rms_gain_error_db": -0.0514628011277985, "output_peak": 0.6045106136962324}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9032014647352828, "全帯域_10ms_rms_envelope_nrmse": 0.05447795536432247, "全帯域_10ms_envelope_correlation": 0.9964775330519995, "全帯域_10ms_modulation_mae_db": 0.2556101691790856, "全帯域_40ms_log_rms_mae_db": 0.46630694695650965, "全帯域_40ms_rms_envelope_nrmse": 0.02858512942156364, "全帯域_40ms_envelope_correlation": 0.9990035558249488, "全帯域_40ms_modulation_mae_db": 0.16572112633825786, "300–1000 Hz_10ms_log_rms_mae_db": 1.6757525621759917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08924035279349063, "300–1000 Hz_10ms_envelope_correlation": 0.993277962676901, "300–1000 Hz_10ms_modulation_mae_db": 0.7413777010251001, "300–1000 Hz_40ms_log_rms_mae_db": 1.0041895589991137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05072881181654697, "300–1000 Hz_40ms_envelope_correlation": 0.9979066567909461, "300–1000 Hz_40ms_modulation_mae_db": 0.7234754584992619, "1000–3000 Hz_10ms_log_rms_mae_db": 1.759426875804148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1788812397161562, "1000–3000 Hz_10ms_envelope_correlation": 0.9778116841631826, "1000–3000 Hz_10ms_modulation_mae_db": 0.9948318909740014, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1551038274984495, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11437849777942682, "1000–3000 Hz_40ms_envelope_correlation": 0.992133060567265, "1000–3000 Hz_40ms_modulation_mae_db": 0.7034784495789964, "3000–6000 Hz_10ms_log_rms_mae_db": 2.094143982694736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17464946894122377, "3000–6000 Hz_10ms_envelope_correlation": 0.9803875320975084, "3000–6000 Hz_10ms_modulation_mae_db": 0.6961559656195421, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3108205032200726, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09369559575591717, "3000–6000 Hz_40ms_envelope_correlation": 0.9941383784421717, "3000–6000 Hz_40ms_modulation_mae_db": 0.606135537863199, "mr_stft_db": 5.48291420319147, "band_mae_db": 0.19534163100022373, "rms_gain_error_db": -0.0665173501600847, "output_peak": 0.6190067013563916}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8652961326457576, "全帯域_10ms_rms_envelope_nrmse": 0.048946535439866615, "全帯域_10ms_envelope_correlation": 0.9972533998369358, "全帯域_10ms_modulation_mae_db": 0.2592523735094386, "全帯域_40ms_log_rms_mae_db": 0.43617993553367645, "全帯域_40ms_rms_envelope_nrmse": 0.02581278574177738, "全帯域_40ms_envelope_correlation": 0.9992989358325146, "全帯域_40ms_modulation_mae_db": 0.29752143528640257, "300–1000 Hz_10ms_log_rms_mae_db": 1.659505429334541, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08188378231995039, "300–1000 Hz_10ms_envelope_correlation": 0.9945247466207575, "300–1000 Hz_10ms_modulation_mae_db": 0.747514304725989, "300–1000 Hz_40ms_log_rms_mae_db": 0.9892255884869624, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04815006366939221, "300–1000 Hz_40ms_envelope_correlation": 0.9983093566219514, "300–1000 Hz_40ms_modulation_mae_db": 0.6295840198669942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.709330263319752, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1471756170713121, "1000–3000 Hz_10ms_envelope_correlation": 0.9869331135307851, "1000–3000 Hz_10ms_modulation_mae_db": 0.7363989677800683, "1000–3000 Hz_40ms_log_rms_mae_db": 1.12334836538987, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09584746246784082, "1000–3000 Hz_40ms_envelope_correlation": 0.9964763006732208, "1000–3000 Hz_40ms_modulation_mae_db": 0.7521285172533462, "3000–6000 Hz_10ms_log_rms_mae_db": 2.137986341929786, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14750669883800532, "3000–6000 Hz_10ms_envelope_correlation": 0.9859230832602902, "3000–6000 Hz_10ms_modulation_mae_db": 0.5795644474257177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3262857477733823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07977258639842547, "3000–6000 Hz_40ms_envelope_correlation": 0.995753285947255, "3000–6000 Hz_40ms_modulation_mae_db": 0.6123885893242147, "mr_stft_db": 5.485240342716118, "band_mae_db": 0.24678673431380024, "rms_gain_error_db": -0.10045562007603072, "output_peak": 0.512366190950882}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9068250495373552, "全帯域_10ms_rms_envelope_nrmse": 0.051536108029327655, "全帯域_10ms_envelope_correlation": 0.9968429818366605, "全帯域_10ms_modulation_mae_db": 0.30587126476070753, "全帯域_40ms_log_rms_mae_db": 0.479013474108389, "全帯域_40ms_rms_envelope_nrmse": 0.026870951573816502, "全帯域_40ms_envelope_correlation": 0.9991114838156449, "全帯域_40ms_modulation_mae_db": 0.3093899012660785, "300–1000 Hz_10ms_log_rms_mae_db": 1.62307791285586, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0822316959089946, "300–1000 Hz_10ms_envelope_correlation": 0.9942416675826431, "300–1000 Hz_10ms_modulation_mae_db": 0.8252721596318654, "300–1000 Hz_40ms_log_rms_mae_db": 0.964657559674846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046554073347444866, "300–1000 Hz_40ms_envelope_correlation": 0.9981674649063028, "300–1000 Hz_40ms_modulation_mae_db": 0.7560040575314436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.744127420986542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1492015162488298, "1000–3000 Hz_10ms_envelope_correlation": 0.9862600618852136, "1000–3000 Hz_10ms_modulation_mae_db": 0.61103968370462, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1258495502207202, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09943381616254388, "1000–3000 Hz_40ms_envelope_correlation": 0.9956712841886274, "1000–3000 Hz_40ms_modulation_mae_db": 0.6863177210042508, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0991074138516588, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14488201509643261, "3000–6000 Hz_10ms_envelope_correlation": 0.986474168050704, "3000–6000 Hz_10ms_modulation_mae_db": 0.6735351634927135, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3050975885623515, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07775601386345367, "3000–6000 Hz_40ms_envelope_correlation": 0.996109686239718, "3000–6000 Hz_40ms_modulation_mae_db": 0.9783928615584342, "mr_stft_db": 5.499901709370974, "band_mae_db": 0.24579958961152948, "rms_gain_error_db": -0.06455076457755735, "output_peak": 0.5315358876052}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9291854566498118, "全帯域_10ms_rms_envelope_nrmse": 0.051547876795853834, "全帯域_10ms_envelope_correlation": 0.9968730812030332, "全帯域_10ms_modulation_mae_db": 0.418347024299317, "全帯域_40ms_log_rms_mae_db": 0.4711225195294936, "全帯域_40ms_rms_envelope_nrmse": 0.027532374950902407, "全帯域_40ms_envelope_correlation": 0.9991045698164526, "全帯域_40ms_modulation_mae_db": 0.5403764302450987, "300–1000 Hz_10ms_log_rms_mae_db": 1.6409334313884198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08101852409272516, "300–1000 Hz_10ms_envelope_correlation": 0.9945325310461041, "300–1000 Hz_10ms_modulation_mae_db": 1.0192248660176235, "300–1000 Hz_40ms_log_rms_mae_db": 0.9905590059305359, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04686749087931408, "300–1000 Hz_40ms_envelope_correlation": 0.9982852630083806, "300–1000 Hz_40ms_modulation_mae_db": 1.0956647302439257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8367621604817874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17899138933787165, "1000–3000 Hz_10ms_envelope_correlation": 0.9774067188686272, "1000–3000 Hz_10ms_modulation_mae_db": 0.8428118724367462, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2073548428720555, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10858143835327179, "1000–3000 Hz_40ms_envelope_correlation": 0.9925531789720847, "1000–3000 Hz_40ms_modulation_mae_db": 0.8884280468953126, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1044154478370602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1961391765693781, "3000–6000 Hz_10ms_envelope_correlation": 0.9750362648276459, "3000–6000 Hz_10ms_modulation_mae_db": 1.2222749800277144, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2776750115315392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10472378540259676, "3000–6000 Hz_40ms_envelope_correlation": 0.9927112005962381, "3000–6000 Hz_40ms_modulation_mae_db": 1.4124589496796112, "mr_stft_db": 5.4993242841715855, "band_mae_db": 0.22190113648607623, "rms_gain_error_db": -0.0805254421909192, "output_peak": 0.5512867257919504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9023269504174161, "全帯域_10ms_rms_envelope_nrmse": 0.049804471805152614, "全帯域_10ms_envelope_correlation": 0.997065110900172, "全帯域_10ms_modulation_mae_db": 0.2882035076887084, "全帯域_40ms_log_rms_mae_db": 0.4788486518623306, "全帯域_40ms_rms_envelope_nrmse": 0.025927257418153556, "全帯域_40ms_envelope_correlation": 0.9991926478124812, "全帯域_40ms_modulation_mae_db": 0.23430363114858835, "300–1000 Hz_10ms_log_rms_mae_db": 1.6139197597897024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0736907847594912, "300–1000 Hz_10ms_envelope_correlation": 0.9954210826721289, "300–1000 Hz_10ms_modulation_mae_db": 0.7106689416854882, "300–1000 Hz_40ms_log_rms_mae_db": 0.9812200191708458, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04132282102474976, "300–1000 Hz_40ms_envelope_correlation": 0.998617215112186, "300–1000 Hz_40ms_modulation_mae_db": 0.6480222934957739, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6898655453940639, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1819431679109975, "1000–3000 Hz_10ms_envelope_correlation": 0.9785997604291792, "1000–3000 Hz_10ms_modulation_mae_db": 1.321365482847118, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1034760157754149, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12449674308607758, "1000–3000 Hz_40ms_envelope_correlation": 0.9921144673853227, "1000–3000 Hz_40ms_modulation_mae_db": 1.2148673048985075, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1568049070344273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1806689841017927, "3000–6000 Hz_10ms_envelope_correlation": 0.9788247467758747, "3000–6000 Hz_10ms_modulation_mae_db": 0.2539079053035715, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3351899939857366, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10892954496648194, "3000–6000 Hz_40ms_envelope_correlation": 0.9920110671139639, "3000–6000 Hz_40ms_modulation_mae_db": 0.16627230871681278, "mr_stft_db": 5.508478384219202, "band_mae_db": 0.23643970849699716, "rms_gain_error_db": -0.06868810129430059, "output_peak": 0.5533206691662224}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7486298419175546, "spectral_convergence": 0.1666910002435213, "spectral_convergence_status": "ok", "amplitude_floor": 1.2959192499433599e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.29472989993649, "spectral_convergence": 0.16440238747502464, "spectral_convergence_status": "ok", "amplitude_floor": 8.646137696849433e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.410690589715636, "spectral_convergence": 0.1652887112991687, "spectral_convergence_status": "ok", "amplitude_floor": 6.578152434885499e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0020500598531285057, "output_power": 0.0020761524550753203, "delta_db": 0.054927004468896964, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0025325885847670413, "output_power": 0.0024635612924920596, "delta_db": -0.12001273199287955, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009745321457083826, "output_power": 0.00095480656950067, "delta_db": -0.08880770747278724, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029510461183058167, "output_power": 0.0002677911127048142, "delta_db": -0.4217983680923251, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.807836063095196e-05, "output_power": 3.6389063655274e-05, "delta_db": -0.1970736195975767, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.519436277144243e-05, "output_power": 3.549901702186451e-05, "delta_db": 0.03743221199056612, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15334194060250528, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06747970968911084, "pitch": null, "output_rms_dbfs": -22.3401872351631, "output_peak": 0.5621440738210801, "reference_rms_dbfs": -22.27270752547399, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.48468344385656}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019946925088635918, "log_rms_mae_db": 0.9017698426010031, "rms_envelope_nrmse": 0.053888809210109606, "envelope_correlation": 0.9965471913307175, "rms_gain_error_db": -0.06584347397453587, "reference_mean_power": 0.0059224241180442505, "output_mean_power": 0.005833311353699318}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0772791897966612e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010245127053592495, "output_power": 0.0010174338084276413, "reference_active": true, "delta_db": -0.03011183858863014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001329567656774176, "output_power": 0.0001220899493603585, "reference_active": true, "delta_db": -0.3703052826678056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.301896656350168e-05, "output_power": 2.0570307192313614e-05, "reference_active": true, "delta_db": -0.48845044682086514}], "active_band_count": 3, "active_band_mae_db": 0.29628918935910026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001930447992601615, "log_rms_mae_db": 0.502672202694853, "rms_envelope_nrmse": 0.029642445024589265, "envelope_correlation": 0.9989150410114203, "rms_gain_error_db": -0.06652562424635874, "reference_mean_power": 0.005919423744261807, "output_mean_power": 0.0058294404181585616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.875987153432414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009473592447040891, "output_power": 0.0009418918052277462, "reference_active": true, "delta_db": -0.025136790831966346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.123049684235515e-05, "output_power": 5.592130209177419e-05, "reference_active": true, "delta_db": -0.3939050831909779}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.329909700774956e-06, "output_power": 1.051368245478181e-06, "reference_active": true, "delta_db": -1.0206729804460477}], "active_band_count": 3, "active_band_mae_db": 0.479904951489664}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015110036587973006, "log_rms_mae_db": 1.5855009235222113, "rms_envelope_nrmse": 0.07399731709624595, "envelope_correlation": 0.9954671478632526, "rms_gain_error_db": -0.15900657744792687, "reference_mean_power": 0.0025365732398871734, "output_mean_power": 0.0024453822116788713}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.31579034521966e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006679887319597139, "output_power": 0.0006437358198206126, "reference_active": true, "delta_db": -0.16061461055622406}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.085773760663629e-05, "output_power": 7.562459542994479e-05, "reference_active": true, "delta_db": -0.796988541951069}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6589878271689645e-05, "output_power": 1.1369812093455161e-05, "reference_active": true, "delta_db": -1.640899121467253}], "active_band_count": 3, "active_band_mae_db": 0.866167424658182}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013775289901134618, "log_rms_mae_db": 1.0009230131735192, "rms_envelope_nrmse": 0.046740959448722155, "envelope_correlation": 0.9982548153405065, "rms_gain_error_db": -0.15998493637949526, "reference_mean_power": 0.0025400163932975937, "output_mean_power": 0.002448150011800104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2092894134690622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006291541070562532, "output_power": 0.0006054556706996376, "reference_active": true, "delta_db": -0.16674684827368108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.975083640469782e-05, "output_power": 3.3666907224664004e-05, "reference_active": true, "delta_db": -0.7214304894848229}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.7844516380393e-07, "output_power": 6.826824125424895e-07, "reference_active": true, "delta_db": -1.5631777640208582}], "active_band_count": 3, "active_band_mae_db": 0.8171183672597874}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010600794217574433, "log_rms_mae_db": 1.6857637480066805, "rms_envelope_nrmse": 0.15794518156314433, "envelope_correlation": 0.9830774104332501, "rms_gain_error_db": -0.482100371374806, "reference_mean_power": 0.0002691478602162077, "output_mean_power": 0.0002408689915063485}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.02821083589085e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010735336162950012, "output_power": 9.481231257463545e-05, "reference_active": true, "delta_db": -0.5395090872984817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.315168535027072e-06, "output_power": 7.118666697437279e-06, "reference_active": true, "delta_db": -0.6747239674852074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.408982675283079e-06, "output_power": 3.8324871270514544e-06, "reference_active": true, "delta_db": 2.0164702808885373}], "active_band_count": 3, "active_band_mae_db": 1.0769011118907421}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.461607013685591e-05, "log_rms_mae_db": 1.0908427448224085, "rms_envelope_nrmse": 0.09922540072791544, "envelope_correlation": 0.9945459265923013, "rms_gain_error_db": -0.49424235030818775, "reference_mean_power": 0.00026921330106685756, "output_mean_power": 0.0002402549135217703}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9187018796507895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.751052870210862e-05, "output_power": 8.71721852736697e-05, "reference_active": true, "delta_db": -0.48673577967277953}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.818200149529557e-06, "output_power": 3.342895006554436e-06, "reference_active": true, "delta_db": -0.5773595379992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4850622077020166e-07, "output_power": 2.0504149811460967e-07, "reference_active": true, "delta_db": 1.400971201400295}], "active_band_count": 3, "active_band_mae_db": 0.8216888396907582}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.633594187238404e-05, "log_rms_mae_db": 2.09269906650761, "rms_envelope_nrmse": 0.17927116213424776, "envelope_correlation": 0.9794477440964126, "rms_gain_error_db": 0.03344174892465787, "reference_mean_power": 4.9508565949953196e-05, "output_mean_power": 4.9891265699418326e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.164436282244953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0281496224399232e-05, "output_power": 9.932401859646475e-06, "reference_active": true, "delta_db": -0.15002037811412766}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.187880808785332e-06, "output_power": 3.678141252746536e-06, "reference_active": true, "delta_db": -0.5636590917200346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.689088055617476e-07, "output_power": 5.450041965021405e-07, "reference_active": true, "delta_db": -0.18642809490124707}], "active_band_count": 3, "active_band_mae_db": 0.3000358549118031}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.3561893402191595e-05, "log_rms_mae_db": 1.327095279554849, "rms_envelope_nrmse": 0.10748649488461218, "envelope_correlation": 0.9923255808889148, "rms_gain_error_db": 0.03350071755293677, "reference_mean_power": 4.9751584691373346e-05, "output_mean_power": 5.013684372338536e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0334524009764534e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.035252748280077e-05, "output_power": 1.009801333109664e-05, "reference_active": true, "delta_db": -0.10810452392749872}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2970706912154704e-06, "output_power": 1.990290254114213e-06, "reference_active": true, "delta_db": -0.622579442051251}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.66930165214842e-08, "output_power": 3.7497584834952625e-08, "reference_active": true, "delta_db": -0.9524863532926614}], "active_band_count": 3, "active_band_mae_db": 0.5610567730904704}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9205112494248391, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5f389db2c2f5bf55a6dfbf74dbd12e43099a8a62d46aee126e018e18dab1bc77", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9819672169885798, "standard_deviation": 0.003743654385243377, "minimum": 0.9774067188686272, "maximum": 0.9869331135307851, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7376771619300624, "standard_deviation": 0.048271850502091644, "minimum": 1.6857637480066805, "maximum": 1.8367621604817874, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0099816727016264, "standard_deviation": 0.2675174393244248, "minimum": 0.61103968370462, "maximum": 1.3573343426595417, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16342543964548653, "standard_deviation": 0.014435446574209755, "minimum": 0.1471756170713121, "maximum": 0.1819431679109975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939401792318373, "standard_deviation": 0.001768654888949458, "minimum": 0.9921144673853227, "maximum": 0.9964763006732208, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1329766260352296, "standard_deviation": 0.03589865095516544, "minimum": 1.0908427448224085, "maximum": 1.2073548428720555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9012141583398787, "standard_deviation": 0.19880750611651524, "minimum": 0.6863177210042508, "maximum": 1.2148673048985075, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10439826725230898, "standard_deviation": 0.011261778600517225, "minimum": 0.0888348403732693, "maximum": 0.12449674308607758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9811960837588318, "standard_deviation": 0.0037885967722325123, "minimum": 0.9750362648276459, "maximum": 0.986474168050704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.11747624810297, "standard_deviation": 0.024198971148529983, "minimum": 2.09269906650761, "maximum": 2.1568049070344273, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6491554488156409, "standard_deviation": 0.3010065012491793, "minimum": 0.2539079053035715, "maximum": 1.2222749800277144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17062371444083624, "standard_deviation": 0.017563580908014843, "minimum": 0.14488201509643261, "maximum": 0.1961391765693781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9941387852627046, "standard_deviation": 0.0016095639932544026, "minimum": 0.9920110671139639, "maximum": 0.996109686239718, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3177448067963624, "standard_deviation": 0.02895307367760843, "minimum": 1.2776750115315392, "maximum": 1.3704828984659543, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7291159011551855, "standard_deviation": 0.3594154305018107, "minimum": 0.16627230871681278, "maximum": 1.4124589496796112, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09542393097712185, "standard_deviation": 0.012863973499723506, "minimum": 0.07775601386345367, "maximum": 0.10892954496648194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9945782444109179, "standard_deviation": 0.00069189014106261, "minimum": 0.993277962676901, "maximum": 0.9954671478632526, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.623878952694222, "standard_deviation": 0.034135363061294714, "minimum": 1.5762032649358786, "maximum": 1.6757525621759917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8174467621406385, "standard_deviation": 0.09702161573846485, "minimum": 0.7106689416854882, "maximum": 1.0192248660176235, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08028932318918977, "standard_deviation": 0.004969841382267321, "minimum": 0.0736907847594912, "maximum": 0.08924035279349063, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982766545449671, "standard_deviation": 0.00019977925473805495, "minimum": 0.9979066567909461, "maximum": 0.998617215112186, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9831790500695617, "standard_deviation": 0.01747160158561653, "minimum": 0.9524940469650321, "maximum": 1.0041895589991137, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7925660354578937, "standard_deviation": 0.14913583637404618, "minimum": 0.6295840198669942, "maximum": 1.0956647302439257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.046324332105194224, "standard_deviation": 0.002748925926215534, "minimum": 0.04132282102474976, "maximum": 0.05072881181654697, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22690248516863135, "standard_deviation": 0.046138992620377746, "minimum": 0.15334194060250528, "maximum": 0.31126517022839256, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.493876011293798, "standard_deviation": 0.009442487615440857, "minimum": 5.48291420319147, "maximum": 5.508478384219202, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5695296191119468, "standard_deviation": 0.041045259206742145, "minimum": 0.512366190950882, "maximum": 0.6220660905076162, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06858305778603918, "standard_deviation": 0.016289504015003998, "minimum": -0.10045562007603072, "maximum": -0.04898467317251161, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967884927229147, "standard_deviation": 0.00029408349838323117, "minimum": 0.9964077707115238, "maximum": 0.9972533998369358, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8900736205959622, "standard_deviation": 0.028893935343777225, "minimum": 0.8381405435751226, "maximum": 0.9291854566498118, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27652574001564245, "standard_deviation": 0.07202538924522858, "minimum": 0.17758975661542054, "maximum": 0.418347024299317, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05204765200540227, "standard_deviation": 0.0021511829605403857, "minimum": 0.048946535439866615, "maximum": 0.054787492482433114, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990857138302794, "standard_deviation": 0.00012789087367271607, "minimum": 0.9989150410114203, "maximum": 0.9992989358325146, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46419086370438967, "standard_deviation": 0.02566330487372384, "minimum": 0.4215629394619232, "maximum": 0.502672202694853, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2938317888859976, "standard_deviation": 0.14738685777682253, "minimum": 0.13403068206312924, "maximum": 0.5403764302450987, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027466734960213, "standard_deviation": 0.0014327358425500936, "minimum": 0.02581278574177738, "maximum": 0.029642445024589265, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0023", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.094993829727173, "predicted_mos": 3.094993829727173, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f036f0ffda763e6cb6d89aa92c912ec6d797e0f11f11a1c4853e5bffc6872461", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.679598900023848, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9302473803960087, "全帯域_10ms_rms_envelope_nrmse": 0.05600454523185582, "全帯域_10ms_envelope_correlation": 0.9962773661515758, "全帯域_10ms_modulation_mae_db": 0.3431662231778389, "全帯域_40ms_log_rms_mae_db": 0.4957067302627268, "全帯域_40ms_rms_envelope_nrmse": 0.03018645212432958, "全帯域_40ms_envelope_correlation": 0.9988822305742421, "全帯域_40ms_modulation_mae_db": 0.39591199075731875, "300–1000 Hz_10ms_log_rms_mae_db": 1.6044955382958763, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07652904030410312, "300–1000 Hz_10ms_envelope_correlation": 0.9952274970522113, "300–1000 Hz_10ms_modulation_mae_db": 0.8331562546309258, "300–1000 Hz_40ms_log_rms_mae_db": 0.9088588828368384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042901676482112065, "300–1000 Hz_40ms_envelope_correlation": 0.9986908367660423, "300–1000 Hz_40ms_modulation_mae_db": 0.8604515307708729, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7289421698423242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16225441524322035, "1000–3000 Hz_10ms_envelope_correlation": 0.9825370472928818, "1000–3000 Hz_10ms_modulation_mae_db": 1.4870812112943674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1209265179442076, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1029591682650125, "1000–3000 Hz_40ms_envelope_correlation": 0.9944793123757917, "1000–3000 Hz_40ms_modulation_mae_db": 1.4038024291508773, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0260041383494247, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16370014752090906, "3000–6000 Hz_10ms_envelope_correlation": 0.9834983962012024, "3000–6000 Hz_10ms_modulation_mae_db": 0.3336759180573812, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1862329328918382, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08763793411680844, "3000–6000 Hz_40ms_envelope_correlation": 0.9953304523444918, "3000–6000 Hz_40ms_modulation_mae_db": 0.8451626953229399, "mr_stft_db": 5.475328157704798, "band_mae_db": 0.2775002609715384, "rms_gain_error_db": -0.06734060725876022, "output_peak": 0.5470764156672883}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8455921458975816, "全帯域_10ms_rms_envelope_nrmse": 0.0522047097377181, "全帯域_10ms_envelope_correlation": 0.9967781741908236, "全帯域_10ms_modulation_mae_db": 0.24469573616601514, "全帯域_40ms_log_rms_mae_db": 0.4337925371914479, "全帯域_40ms_rms_envelope_nrmse": 0.027440198354850488, "全帯域_40ms_envelope_correlation": 0.999094909031449, "全帯域_40ms_modulation_mae_db": 0.24050084147855724, "300–1000 Hz_10ms_log_rms_mae_db": 1.604381650782709, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07642065184344274, "300–1000 Hz_10ms_envelope_correlation": 0.9950792097469715, "300–1000 Hz_10ms_modulation_mae_db": 0.6723602175462268, "300–1000 Hz_40ms_log_rms_mae_db": 0.9347055628438355, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03939455887411587, "300–1000 Hz_40ms_envelope_correlation": 0.9987708458781314, "300–1000 Hz_40ms_modulation_mae_db": 0.6972303285058823, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6617300738534435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14775414599127684, "1000–3000 Hz_10ms_envelope_correlation": 0.9854498703565333, "1000–3000 Hz_10ms_modulation_mae_db": 1.3474757394608117, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0617894695664014, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08999177052049373, "1000–3000 Hz_40ms_envelope_correlation": 0.9957767661763106, "1000–3000 Hz_40ms_modulation_mae_db": 1.0408074869257458, "3000–6000 Hz_10ms_log_rms_mae_db": 2.042255841058735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1845010682627478, "3000–6000 Hz_10ms_envelope_correlation": 0.9779601564097138, "3000–6000 Hz_10ms_modulation_mae_db": 1.2552669269469277, "3000–6000 Hz_40ms_log_rms_mae_db": 1.26562877270025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10566687899047161, "3000–6000 Hz_40ms_envelope_correlation": 0.9925367551747603, "3000–6000 Hz_40ms_modulation_mae_db": 1.639512550540651, "mr_stft_db": 5.460183447619033, "band_mae_db": 0.25395802181089067, "rms_gain_error_db": -0.07482806655585014, "output_peak": 0.5670446270615586}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9185858841548858, "全帯域_10ms_rms_envelope_nrmse": 0.053069013960375895, "全帯域_10ms_envelope_correlation": 0.9966688579527501, "全帯域_10ms_modulation_mae_db": 0.36075126620893067, "全帯域_40ms_log_rms_mae_db": 0.49506708163390845, "全帯域_40ms_rms_envelope_nrmse": 0.029088948786132635, "全帯域_40ms_envelope_correlation": 0.9989788664932164, "全帯域_40ms_modulation_mae_db": 0.33489582567431375, "300–1000 Hz_10ms_log_rms_mae_db": 1.5530468282946293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0771676430409285, "300–1000 Hz_10ms_envelope_correlation": 0.9950467545693922, "300–1000 Hz_10ms_modulation_mae_db": 0.9926082369278385, "300–1000 Hz_40ms_log_rms_mae_db": 0.9347809354815988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04719676636488233, "300–1000 Hz_40ms_envelope_correlation": 0.9982366631700121, "300–1000 Hz_40ms_modulation_mae_db": 0.8978413501907102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7175401898544924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15888383222504932, "1000–3000 Hz_10ms_envelope_correlation": 0.9819549147958558, "1000–3000 Hz_10ms_modulation_mae_db": 0.6197711341121764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0453654888655044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07898907532373318, "1000–3000 Hz_40ms_envelope_correlation": 0.9961124814459293, "1000–3000 Hz_40ms_modulation_mae_db": 0.8131597499137619, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0272963451706114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18554356589890672, "3000–6000 Hz_10ms_envelope_correlation": 0.9778842818945308, "3000–6000 Hz_10ms_modulation_mae_db": 0.4727292090540403, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2231582155212917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10048198607174179, "3000–6000 Hz_40ms_envelope_correlation": 0.9938928812955097, "3000–6000 Hz_40ms_modulation_mae_db": 0.49765174080547037, "mr_stft_db": 5.464988936138977, "band_mae_db": 0.2333304054656656, "rms_gain_error_db": -0.07555709715975463, "output_peak": 0.570018572883602}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8841061839240596, "全帯域_10ms_rms_envelope_nrmse": 0.05195981663312482, "全帯域_10ms_envelope_correlation": 0.996850717234975, "全帯域_10ms_modulation_mae_db": 0.3535965220855186, "全帯域_40ms_log_rms_mae_db": 0.47299101502579327, "全帯域_40ms_rms_envelope_nrmse": 0.028994229454781507, "全帯域_40ms_envelope_correlation": 0.9990307252557987, "全帯域_40ms_modulation_mae_db": 0.3130399336132465, "300–1000 Hz_10ms_log_rms_mae_db": 1.5364202141330092, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0892271181665516, "300–1000 Hz_10ms_envelope_correlation": 0.9934825735814377, "300–1000 Hz_10ms_modulation_mae_db": 1.1450957860670943, "300–1000 Hz_40ms_log_rms_mae_db": 0.8802324230398998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051126124851544455, "300–1000 Hz_40ms_envelope_correlation": 0.9981271596588508, "300–1000 Hz_40ms_modulation_mae_db": 1.2145708823548498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7524663164206882, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16117429831290192, "1000–3000 Hz_10ms_envelope_correlation": 0.9826521079056567, "1000–3000 Hz_10ms_modulation_mae_db": 1.0394183327461188, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1155844356821243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10394318413696299, "1000–3000 Hz_40ms_envelope_correlation": 0.9941232678670905, "1000–3000 Hz_40ms_modulation_mae_db": 0.9522509436717207, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0171311324689545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1576079031174675, "3000–6000 Hz_10ms_envelope_correlation": 0.9840795964147966, "3000–6000 Hz_10ms_modulation_mae_db": 0.8802659858393965, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2533180920872167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07774497395868944, "3000–6000 Hz_40ms_envelope_correlation": 0.9963658211655295, "3000–6000 Hz_40ms_modulation_mae_db": 0.8477344414207377, "mr_stft_db": 5.478969628406119, "band_mae_db": 0.2699357383165196, "rms_gain_error_db": -0.09142646887616643, "output_peak": 0.5354982267092117}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8878986327591178, "全帯域_10ms_rms_envelope_nrmse": 0.052154030186932245, "全帯域_10ms_envelope_correlation": 0.9968127930828699, "全帯域_10ms_modulation_mae_db": 0.31032301749776675, "全帯域_40ms_log_rms_mae_db": 0.4670583495215033, "全帯域_40ms_rms_envelope_nrmse": 0.028291682827706825, "全帯域_40ms_envelope_correlation": 0.9990661221788665, "全帯域_40ms_modulation_mae_db": 0.38331657326381596, "300–1000 Hz_10ms_log_rms_mae_db": 1.525372367836137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07466243589977294, "300–1000 Hz_10ms_envelope_correlation": 0.9953940795399684, "300–1000 Hz_10ms_modulation_mae_db": 0.4235700680522492, "300–1000 Hz_40ms_log_rms_mae_db": 0.8865258998038278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04472450989470758, "300–1000 Hz_40ms_envelope_correlation": 0.998452486172557, "300–1000 Hz_40ms_modulation_mae_db": 0.41222353809546314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7245614301289796, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17070606238423264, "1000–3000 Hz_10ms_envelope_correlation": 0.982130123072268, "1000–3000 Hz_10ms_modulation_mae_db": 0.7901553564786533, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1372429687710628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1195708903288326, "1000–3000 Hz_40ms_envelope_correlation": 0.9936240833212486, "1000–3000 Hz_40ms_modulation_mae_db": 0.7330947959553873, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0865259855048266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18106157861541422, "3000–6000 Hz_10ms_envelope_correlation": 0.9789042156862121, "3000–6000 Hz_10ms_modulation_mae_db": 0.40076005375567747, "3000–6000 Hz_40ms_log_rms_mae_db": 1.280059203263769, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09593076863749427, "3000–6000 Hz_40ms_envelope_correlation": 0.9938185096222603, "3000–6000 Hz_40ms_modulation_mae_db": 0.745715522550026, "mr_stft_db": 5.467904073623792, "band_mae_db": 0.32092637667063445, "rms_gain_error_db": -0.08552509167263023, "output_peak": 0.6234078488752806}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9306495850773656, "全帯域_10ms_rms_envelope_nrmse": 0.05480043232089468, "全帯域_10ms_envelope_correlation": 0.9964675263929281, "全帯域_10ms_modulation_mae_db": 0.3407680225947949, "全帯域_40ms_log_rms_mae_db": 0.5078408381122707, "全帯域_40ms_rms_envelope_nrmse": 0.02894310260475325, "全帯域_40ms_envelope_correlation": 0.999016040564852, "全帯域_40ms_modulation_mae_db": 0.4193687504543598, "300–1000 Hz_10ms_log_rms_mae_db": 1.5867289822039257, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07922285929028373, "300–1000 Hz_10ms_envelope_correlation": 0.9947369920711957, "300–1000 Hz_10ms_modulation_mae_db": 0.46356237788372345, "300–1000 Hz_40ms_log_rms_mae_db": 0.9447532341646556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0462148730274638, "300–1000 Hz_40ms_envelope_correlation": 0.9982776257629218, "300–1000 Hz_40ms_modulation_mae_db": 0.5399453951018981, "1000–3000 Hz_10ms_log_rms_mae_db": 1.793757722358279, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15115958116876368, "1000–3000 Hz_10ms_envelope_correlation": 0.9868638967012627, "1000–3000 Hz_10ms_modulation_mae_db": 1.0171473620827642, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1726772493390003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10796102917250719, "1000–3000 Hz_40ms_envelope_correlation": 0.9954363487151071, "1000–3000 Hz_40ms_modulation_mae_db": 0.9989443227393116, "3000–6000 Hz_10ms_log_rms_mae_db": 2.058950762119193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17995865036140793, "3000–6000 Hz_10ms_envelope_correlation": 0.980875905671199, "3000–6000 Hz_10ms_modulation_mae_db": 1.5816509402269763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2995975660203447, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11594884543215074, "3000–6000 Hz_40ms_envelope_correlation": 0.9936762632871979, "3000–6000 Hz_40ms_modulation_mae_db": 1.620663978608155, "mr_stft_db": 5.47851098434185, "band_mae_db": 0.3754904146816405, "rms_gain_error_db": -0.08729350563160625, "output_peak": 0.5071068116626506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8773820902106068, "全帯域_10ms_rms_envelope_nrmse": 0.05411538269260151, "全帯域_10ms_envelope_correlation": 0.9965657075663824, "全帯域_10ms_modulation_mae_db": 0.31324224069679746, "全帯域_40ms_log_rms_mae_db": 0.45271119549665867, "全帯域_40ms_rms_envelope_nrmse": 0.027584272872939554, "全帯域_40ms_envelope_correlation": 0.9991239710930564, "全帯域_40ms_modulation_mae_db": 0.44869674887771843, "300–1000 Hz_10ms_log_rms_mae_db": 1.5941500499404926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07985926479904233, "300–1000 Hz_10ms_envelope_correlation": 0.9946727046989872, "300–1000 Hz_10ms_modulation_mae_db": 0.8279453283017794, "300–1000 Hz_40ms_log_rms_mae_db": 0.9722953596753979, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04111783700966804, "300–1000 Hz_40ms_envelope_correlation": 0.9987415458191048, "300–1000 Hz_40ms_modulation_mae_db": 0.9917544545256199, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6245939276884631, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15168179529419237, "1000–3000 Hz_10ms_envelope_correlation": 0.9844348210304384, "1000–3000 Hz_10ms_modulation_mae_db": 0.7178269866782744, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0187095561517114, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09165953135890743, "1000–3000 Hz_40ms_envelope_correlation": 0.9953564020549696, "1000–3000 Hz_40ms_modulation_mae_db": 0.8379354730122631, "3000–6000 Hz_10ms_log_rms_mae_db": 2.040827171739988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16403339550210472, "3000–6000 Hz_10ms_envelope_correlation": 0.9825706312699924, "3000–6000 Hz_10ms_modulation_mae_db": 0.27836613760775086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2209160164433852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08145063981902641, "3000–6000 Hz_40ms_envelope_correlation": 0.9955652985263395, "3000–6000 Hz_40ms_modulation_mae_db": 0.24348709590195652, "mr_stft_db": 5.471090054623239, "band_mae_db": 0.17742080737388635, "rms_gain_error_db": -0.08940151398758715, "output_peak": 0.5359290685233136}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8779403129787368, "全帯域_10ms_rms_envelope_nrmse": 0.048589574364362134, "全帯域_10ms_envelope_correlation": 0.9972319086810205, "全帯域_10ms_modulation_mae_db": 0.3797021878461065, "全帯域_40ms_log_rms_mae_db": 0.4526913388673914, "全帯域_40ms_rms_envelope_nrmse": 0.026237539156367208, "全帯域_40ms_envelope_correlation": 0.9991995684803668, "全帯域_40ms_modulation_mae_db": 0.4947176728221054, "300–1000 Hz_10ms_log_rms_mae_db": 1.5393377550525182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08076452328519314, "300–1000 Hz_10ms_envelope_correlation": 0.9948368038483731, "300–1000 Hz_10ms_modulation_mae_db": 0.8994908032638108, "300–1000 Hz_40ms_log_rms_mae_db": 0.9288364777590521, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049314057812858504, "300–1000 Hz_40ms_envelope_correlation": 0.9983710492815453, "300–1000 Hz_40ms_modulation_mae_db": 0.9398790132682074, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6393707528058301, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1525266106998862, "1000–3000 Hz_10ms_envelope_correlation": 0.983094686832476, "1000–3000 Hz_10ms_modulation_mae_db": 1.3295662229161107, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0509187962073199, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08456820164343216, "1000–3000 Hz_40ms_envelope_correlation": 0.9947307694710574, "1000–3000 Hz_40ms_modulation_mae_db": 1.0804855628829495, "3000–6000 Hz_10ms_log_rms_mae_db": 2.046705506962024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1650248661110715, "3000–6000 Hz_10ms_envelope_correlation": 0.9828282630391972, "3000–6000 Hz_10ms_modulation_mae_db": 0.9580877340611501, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2577334117412973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09418215590403627, "3000–6000 Hz_40ms_envelope_correlation": 0.99491294837851, "3000–6000 Hz_40ms_modulation_mae_db": 0.595862311727883, "mr_stft_db": 5.463138333704219, "band_mae_db": 0.23176919816215658, "rms_gain_error_db": -0.07929324185369163, "output_peak": 0.5519760900931723}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.759265085627747, "spectral_convergence": 0.16693246507742937, "spectral_convergence_status": "ok", "amplitude_floor": 1.2959192499433599e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.282008194081186, "spectral_convergence": 0.1648804294595313, "spectral_convergence_status": "ok", "amplitude_floor": 8.646137696849433e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.384711193405462, "spectral_convergence": 0.15846785908464597, "spectral_convergence_status": "ok", "amplitude_floor": 6.578152434885499e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0020500598531285057, "output_power": 0.0020900218620025527, "delta_db": 0.08384288118992562, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0025325885847670413, "output_power": 0.0024576094211180327, "delta_db": -0.1305178183046694, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009745321457083826, "output_power": 0.0009546352637539262, "delta_db": -0.08958696284923366, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029510461183058167, "output_power": 0.00025900385577478713, "delta_db": -0.566697671673926, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.807836063095196e-05, "output_power": 3.46560043176052e-05, "delta_db": -0.4089975359521611, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.519436277144243e-05, "output_power": 3.845997307546462e-05, "delta_db": 0.38535869585931476, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2775002609715384, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06734060725876022, "pitch": null, "output_rms_dbfs": -22.34004813273275, "output_peak": 0.5470764156672883, "reference_rms_dbfs": -22.27270752547399, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.475328157704798}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019946925088635918, "log_rms_mae_db": 0.9302473803960087, "rms_envelope_nrmse": 0.05600454523185582, "envelope_correlation": 0.9962773661515758, "rms_gain_error_db": -0.06596925549139793, "reference_mean_power": 0.0059224241180442505, "output_mean_power": 0.005833142410239077}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0772791897966612e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010245127053592495, "output_power": 0.0010086456527042323, "reference_active": true, "delta_db": -0.06778727576489078}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001329567656774176, "output_power": 0.00012630415961229426, "reference_active": true, "delta_db": -0.2229278832807562}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.301896656350168e-05, "output_power": 1.941813298528581e-05, "reference_active": true, "delta_db": -0.7387835104878697}], "active_band_count": 3, "active_band_mae_db": 0.3431662231778389}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001930447992601615, "log_rms_mae_db": 0.4957067302627268, "rms_envelope_nrmse": 0.03018645212432958, "envelope_correlation": 0.9988822305742421, "rms_gain_error_db": -0.06403729573139166, "reference_mean_power": 0.005919423744261807, "output_mean_power": 0.005832781404462146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.875987153432414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009473592447040891, "output_power": 0.0009340963125462644, "reference_active": true, "delta_db": -0.06123039835430711}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.123049684235515e-05, "output_power": 5.868810601933643e-05, "reference_active": true, "delta_db": -0.18417689241644677}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.329909700774956e-06, "output_power": 1.0705064436312718e-06, "reference_active": true, "delta_db": -0.9423286815012024}], "active_band_count": 3, "active_band_mae_db": 0.39591199075731875}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015110036587973006, "log_rms_mae_db": 1.6044955382958763, "rms_envelope_nrmse": 0.07652904030410312, "envelope_correlation": 0.9952274970522113, "rms_gain_error_db": -0.18854380901580423, "reference_mean_power": 0.0025365732398871734, "output_mean_power": 0.0024288071100206504}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.31579034521966e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006679887319597139, "output_power": 0.0006421470845699719, "reference_active": true, "delta_db": -0.17134621451938306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.085773760663629e-05, "output_power": 7.910614258164034e-05, "reference_active": true, "delta_db": -0.6015171079317663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6589878271689645e-05, "output_power": 1.1147632813337464e-05, "reference_active": true, "delta_db": -1.7266054414416279}], "active_band_count": 3, "active_band_mae_db": 0.8331562546309258}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013775289901134618, "log_rms_mae_db": 0.9088588828368384, "rms_envelope_nrmse": 0.042901676482112065, "envelope_correlation": 0.9986908367660423, "rms_gain_error_db": -0.18718758481056302, "reference_mean_power": 0.0025400163932975937, "output_mean_power": 0.002432863602604229}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2092894134690622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006291541070562532, "output_power": 0.0006062314016920328, "reference_active": true, "delta_db": -0.16118607696329804}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.975083640469782e-05, "output_power": 3.5847102185410216e-05, "reference_active": true, "delta_db": -0.4489221731295613}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.7844516380393e-07, "output_power": 6.214581054844817e-07, "reference_active": true, "delta_db": -1.9712463422197595}], "active_band_count": 3, "active_band_mae_db": 0.8604515307708729}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010600794217574433, "log_rms_mae_db": 1.7289421698423242, "rms_envelope_nrmse": 0.16225441524322035, "envelope_correlation": 0.9825370472928818, "rms_gain_error_db": -0.5516784332776832, "reference_mean_power": 0.0002691478602162077, "output_mean_power": 0.00023704079121654058}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.02821083589085e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010735336162950012, "output_power": 9.071645144394445e-05, "reference_active": true, "delta_db": -0.7312959476555638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.315168535027072e-06, "output_power": 5.495387825118716e-06, "reference_active": true, "delta_db": -1.7987270885647981}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.408982675283079e-06, "output_power": 3.7579910067350057e-06, "reference_active": true, "delta_db": 1.9312205976627403}], "active_band_count": 3, "active_band_mae_db": 1.4870812112943674}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.461607013685591e-05, "log_rms_mae_db": 1.1209265179442076, "rms_envelope_nrmse": 0.1029591682650125, "envelope_correlation": 0.9944793123757917, "rms_gain_error_db": -0.5538239307000359, "reference_mean_power": 0.00026921330106685756, "output_mean_power": 0.00023698132331742414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9187018796507895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.751052870210862e-05, "output_power": 8.299369922524365e-05, "reference_active": true, "delta_db": -0.7000638863954314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.818200149529557e-06, "output_power": 2.496699195773203e-06, "reference_active": true, "delta_db": -1.844924689395879}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4850622077020166e-07, "output_power": 2.179648772963747e-07, "reference_active": true, "delta_db": 1.6664187116613214}], "active_band_count": 3, "active_band_mae_db": 1.4038024291508773}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.633594187238404e-05, "log_rms_mae_db": 2.0260041383494247, "rms_envelope_nrmse": 0.16370014752090906, "envelope_correlation": 0.9834983962012024, "rms_gain_error_db": 0.19177238122572904, "reference_mean_power": 4.9508565949953196e-05, "output_mean_power": 5.174371247638611e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.164436282244953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0281496224399232e-05, "output_power": 9.882986464697227e-06, "reference_active": true, "delta_db": -0.1716811979624219}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.187880808785332e-06, "output_power": 3.7488450080167054e-06, "reference_active": true, "delta_db": -0.4809682738145459}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.689088055617476e-07, "output_power": 5.250550351131461e-07, "reference_active": true, "delta_db": -0.34837828239517593}], "active_band_count": 3, "active_band_mae_db": 0.3336759180573812}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.3561893402191595e-05, "log_rms_mae_db": 1.1862329328918382, "rms_envelope_nrmse": 0.08763793411680844, "envelope_correlation": 0.9953304523444918, "rms_gain_error_db": 0.1918923286622951, "reference_mean_power": 4.9751584691373346e-05, "output_mean_power": 5.199913884306485e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0334524009764534e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.035252748280077e-05, "output_power": 9.952682670628402e-06, "reference_active": true, "delta_db": -0.17106234764717262}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2970706912154704e-06, "output_power": 1.9705135083968628e-06, "reference_active": true, "delta_db": -0.6659494418216972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.66930165214842e-08, "output_power": 3.157943110400241e-08, "reference_active": true, "delta_db": -1.6984762964999498}], "active_band_count": 3, "active_band_mae_db": 0.8451626953229399}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9175403177766562, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9b3ec2525e8eded3339772812bee170b1c2a994e96d82445539da0a908690c82", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9836396834984216, "standard_deviation": 0.0017691774809585263, "minimum": 0.9819549147958558, "maximum": 0.9868638967012627, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7053703228690624, "standard_deviation": 0.05841678064705844, "minimum": 1.6245939276884631, "maximum": 1.793757722358279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0435552932211596, "standard_deviation": 0.32104808139130225, "minimum": 0.6197711341121764, "maximum": 1.4870812112943674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15701759266494042, "standard_deviation": 0.007600065427059957, "minimum": 0.14775414599127684, "maximum": 0.17070606238423264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949549289284381, "standard_deviation": 0.0008575345635421657, "minimum": 0.9936240833212486, "maximum": 0.9961124814459293, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0904018103159165, "standard_deviation": 0.05355222870003312, "minimum": 1.0187095561517114, "maximum": 1.1726772493390003, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9825600955315021, "standard_deviation": 0.2082100729029732, "minimum": 0.7330947959553873, "maximum": 1.4038024291508773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09745535634373523, "standard_deviation": 0.013460146739517703, "minimum": 0.07898907532373318, "maximum": 0.1195708903288326, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9810751808233555, "standard_deviation": 0.0025311253395586893, "minimum": 0.9778842818945308, "maximum": 0.9840795964147966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0432121104217194, "standard_deviation": 0.02194879245721496, "minimum": 2.0171311324689545, "maximum": 2.0865259855048266, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7701003631936626, "standard_deviation": 0.4778803439746058, "minimum": 0.27836613760775086, "maximum": 1.5816509402269763, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17267889692375368, "standard_deviation": 0.011145846651985026, "minimum": 0.1576079031174675, "maximum": 0.18554356589890672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945123662243249, "standard_deviation": 0.001244473400729052, "minimum": 0.9925367551747603, "maximum": 0.9963658211655295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2483305263336741, "standard_deviation": 0.036427715874611646, "minimum": 1.1862329328918382, "maximum": 1.2995975660203447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8794737921097274, "standard_deviation": 0.5039558296573106, "minimum": 0.24348709590195652, "maximum": 1.639512550540651, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09488052286630236, "standard_deviation": 0.012647025080089124, "minimum": 0.07774497395868944, "maximum": 0.11594884543215074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9948095768885671, "standard_deviation": 0.0005893019332535248, "minimum": 0.9934825735814377, "maximum": 0.9953940795399684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5679916733174122, "standard_deviation": 0.03284088847236273, "minimum": 1.525372367836137, "maximum": 1.6044955382958763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.782223634084206, "standard_deviation": 0.24974885123366608, "minimum": 0.4235700680522492, "maximum": 1.1450957860670943, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07923169207866476, "standard_deviation": 0.004519745789323113, "minimum": 0.07466243589977294, "maximum": 0.0892271181665516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984585265636456, "standard_deviation": 0.00024815102982937557, "minimum": 0.9981271596588508, "maximum": 0.9987708458781314, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9238735969506382, "standard_deviation": 0.030627723768827317, "minimum": 0.8802324230398998, "maximum": 0.9722953596753979, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.819237061601688, "standard_deviation": 0.25830002324590556, "minimum": 0.41222353809546314, "maximum": 1.2145708823548498, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04524880053966908, "standard_deviation": 0.004016359436442559, "minimum": 0.03939455887411587, "maximum": 0.051126124851544455, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.26754140293161655, "standard_deviation": 0.06019507481318136, "minimum": 0.17742080737388635, "maximum": 0.3754904146816405, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.470014202020254, "standard_deviation": 0.007125585606385946, "minimum": 5.460183447619033, "maximum": 5.478969628406119, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5547572076845098, "standard_deviation": 0.03414150610365301, "minimum": 0.5071068116626506, "maximum": 0.6234078488752806, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08133319912450584, "standard_deviation": 0.008415149368288766, "minimum": -0.09142646887616643, "maximum": -0.06734060725876022, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967066314066657, "standard_deviation": 0.00028680302890796524, "minimum": 0.9962773661515758, "maximum": 0.9972319086810205, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8940502769247953, "standard_deviation": 0.029938062824970506, "minimum": 0.8455921458975816, "maximum": 0.9306495850773656, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33078065203422113, "standard_deviation": 0.04175094810347307, "minimum": 0.24469573616601514, "maximum": 0.3797021878461065, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.052862188140983146, "standard_deviation": 0.0022439678012810844, "minimum": 0.048589574364362134, "maximum": 0.05600454523185582, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999049054208981, "standard_deviation": 9.624623000048728e-05, "minimum": 0.9988822305742421, "maximum": 0.9991995684803668, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47223238576396254, "standard_deviation": 0.025657991165941255, "minimum": 0.4337925371914479, "maximum": 0.5078408381122707, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37880604211767943, "standard_deviation": 0.08076431793577243, "minimum": 0.24050084147855724, "maximum": 0.4947176728221054, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02834580327273263, "standard_deviation": 0.0012285433813257808, "minimum": 0.026237539156367208, "maximum": 0.03018645212432958, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1320-122617-0039", "language": "en", "style": "normal", "speaker": "librispeech_1320", "source_utterance_id": "1320-122617-0039", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1320-122617-0039.wav", "source_file_sha256": "70553e4b2ebc329aa1415369181878075ad8973d363084e79aefbba190a88c73", "decoded_pcm_sha256": "9cf3ed8985625af93ca84569280377a536aa4892a8aefbabb265fc0a052a863c", "original_source_sha256": "264a53a8c31bd50758ba7eeeb3b5efa9c5bc7908757d44a635bec4c9a4f2c3c0", "corpus_entry_id": "librispeech_1320-122617-0039"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.4140121936798096, "predicted_mos": 3.4140121936798096, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8700e4d8c918e770f9ed571fbd6d154bae096a64bce40dbf4087cb0c57606c23", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.986023325000133, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1320-122617-0039", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "8700e4d8c918e770f9ed571fbd6d154bae096a64bce40dbf4087cb0c57606c23", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "cfdb8185be9ad3c052e4be2c79b53474d7eff756d56a20dec359086677830e79"}, "rows": [{"case": "librispeech_1320-122617-0039", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.0155856609344482, "predicted_mos": 3.0155856609344482, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "108421909f81322b97cdd4ad555c17e50a29e17f99b1564f356165d5e25449b4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.950666275002732, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1320-122617-0039", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "108421909f81322b97cdd4ad555c17e50a29e17f99b1564f356165d5e25449b4", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.763053590266098, "全帯域_10ms_rms_envelope_nrmse": 0.05747624011365568, "全帯域_10ms_envelope_correlation": 0.9956176902958148, "全帯域_10ms_modulation_mae_db": 0.2921162798191899, "全帯域_40ms_log_rms_mae_db": 0.4182091511232521, "全帯域_40ms_rms_envelope_nrmse": 0.030583147359703026, "全帯域_40ms_envelope_correlation": 0.998640382985547, "全帯域_40ms_modulation_mae_db": 0.47413910699664336, "300–1000 Hz_10ms_log_rms_mae_db": 1.7029522333499512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09839085093866391, "300–1000 Hz_10ms_envelope_correlation": 0.9923684302323444, "300–1000 Hz_10ms_modulation_mae_db": 0.3504404365937028, "300–1000 Hz_40ms_log_rms_mae_db": 1.088139671337801, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06601351451831762, "300–1000 Hz_40ms_envelope_correlation": 0.9963542736643743, "300–1000 Hz_40ms_modulation_mae_db": 0.40195058214347074, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7216996486733613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15207152532689233, "1000–3000 Hz_10ms_envelope_correlation": 0.981063608501306, "1000–3000 Hz_10ms_modulation_mae_db": 0.7223290274091179, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1192597784660199, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09663818170365046, "1000–3000 Hz_40ms_envelope_correlation": 0.9935501828074842, "1000–3000 Hz_40ms_modulation_mae_db": 0.5809213786006501, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8299454157824957, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1673085162732792, "3000–6000 Hz_10ms_envelope_correlation": 0.9806949667334076, "3000–6000 Hz_10ms_modulation_mae_db": 0.5274353706931632, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2127009800602626, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09733490687143494, "3000–6000 Hz_40ms_envelope_correlation": 0.9931241204825552, "3000–6000 Hz_40ms_modulation_mae_db": 0.37849668499565126, "mr_stft_db": 5.460297365457477, "band_mae_db": 0.20440997688905824, "rms_gain_error_db": -0.049110647579318864, "output_peak": 0.5659362190017526}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.754968063341811, "全帯域_10ms_rms_envelope_nrmse": 0.04942731436487262, "全帯域_10ms_envelope_correlation": 0.9967491990048162, "全帯域_10ms_modulation_mae_db": 0.27315645517634723, "全帯域_40ms_log_rms_mae_db": 0.3965607773728473, "全帯域_40ms_rms_envelope_nrmse": 0.024022696214601784, "全帯域_40ms_envelope_correlation": 0.999148324696194, "全帯域_40ms_modulation_mae_db": 0.2509625658455141, "300–1000 Hz_10ms_log_rms_mae_db": 1.727196237159418, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09057097501914742, "300–1000 Hz_10ms_envelope_correlation": 0.9936442103674579, "300–1000 Hz_10ms_modulation_mae_db": 0.4404255919219338, "300–1000 Hz_40ms_log_rms_mae_db": 1.1002201645329441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050634053245140066, "300–1000 Hz_40ms_envelope_correlation": 0.9980273310951956, "300–1000 Hz_40ms_modulation_mae_db": 0.6521475077310356, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7451760548325574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1524069871467188, "1000–3000 Hz_10ms_envelope_correlation": 0.9807025158967426, "1000–3000 Hz_10ms_modulation_mae_db": 0.9651012098018441, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0791604054359192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09010852788597393, "1000–3000 Hz_40ms_envelope_correlation": 0.9945319139800431, "1000–3000 Hz_40ms_modulation_mae_db": 0.8751792731926932, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7846469149393684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15688858772109518, "3000–6000 Hz_10ms_envelope_correlation": 0.9831363100178157, "3000–6000 Hz_10ms_modulation_mae_db": 0.5612399466724253, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1460933308218355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09898948369440748, "3000–6000 Hz_40ms_envelope_correlation": 0.9934170373513117, "3000–6000 Hz_40ms_modulation_mae_db": 0.6812214485982561, "mr_stft_db": 5.451465342494643, "band_mae_db": 0.24729264487536382, "rms_gain_error_db": -0.027382604140720673, "output_peak": 0.5523751641140909}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7672810215089961, "全帯域_10ms_rms_envelope_nrmse": 0.05167342400022374, "全帯域_10ms_envelope_correlation": 0.9964926023324185, "全帯域_10ms_modulation_mae_db": 0.3697190407126223, "全帯域_40ms_log_rms_mae_db": 0.4425763625170737, "全帯域_40ms_rms_envelope_nrmse": 0.02900091814327883, "全帯域_40ms_envelope_correlation": 0.9988099497649792, "全帯域_40ms_modulation_mae_db": 0.36399258555081904, "300–1000 Hz_10ms_log_rms_mae_db": 1.7108886846244862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0905549370702052, "300–1000 Hz_10ms_envelope_correlation": 0.993723352724587, "300–1000 Hz_10ms_modulation_mae_db": 0.9549947835459361, "300–1000 Hz_40ms_log_rms_mae_db": 1.0747241776959524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05409554053845562, "300–1000 Hz_40ms_envelope_correlation": 0.9977840358842676, "300–1000 Hz_40ms_modulation_mae_db": 0.917020330145097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6718875200895402, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1487757282991596, "1000–3000 Hz_10ms_envelope_correlation": 0.9818369903987225, "1000–3000 Hz_10ms_modulation_mae_db": 0.662095255297863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0831647092806524, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08848403807847612, "1000–3000 Hz_40ms_envelope_correlation": 0.994948703748437, "1000–3000 Hz_40ms_modulation_mae_db": 0.8959856702012242, "3000–6000 Hz_10ms_log_rms_mae_db": 1.750924852731757, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17010094894296074, "3000–6000 Hz_10ms_envelope_correlation": 0.9798079938633607, "3000–6000 Hz_10ms_modulation_mae_db": 0.31900108522923715, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1609156062485644, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10753448077322751, "3000–6000 Hz_40ms_envelope_correlation": 0.991592004271886, "3000–6000 Hz_40ms_modulation_mae_db": 0.10450134646146696, "mr_stft_db": 5.458890813398426, "band_mae_db": 0.21959560997271346, "rms_gain_error_db": -0.06244342495502056, "output_peak": 0.5535469253765704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7501425014810489, "全帯域_10ms_rms_envelope_nrmse": 0.051842637969672145, "全帯域_10ms_envelope_correlation": 0.9964328889094888, "全帯域_10ms_modulation_mae_db": 0.4487128027501048, "全帯域_40ms_log_rms_mae_db": 0.37267562581323543, "全帯域_40ms_rms_envelope_nrmse": 0.027130548110177125, "全帯域_40ms_envelope_correlation": 0.9989257381166508, "全帯域_40ms_modulation_mae_db": 0.421464807384538, "300–1000 Hz_10ms_log_rms_mae_db": 1.670384098499602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10176091754494532, "300–1000 Hz_10ms_envelope_correlation": 0.9919423762419961, "300–1000 Hz_10ms_modulation_mae_db": 0.3214894864086122, "300–1000 Hz_40ms_log_rms_mae_db": 1.0643491921173545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05620370260241869, "300–1000 Hz_40ms_envelope_correlation": 0.997580759582787, "300–1000 Hz_40ms_modulation_mae_db": 0.32179108142658985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6858070061180859, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15417199143467694, "1000–3000 Hz_10ms_envelope_correlation": 0.9815158521999435, "1000–3000 Hz_10ms_modulation_mae_db": 1.0395081156234847, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1209972619380304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10228032524321076, "1000–3000 Hz_40ms_envelope_correlation": 0.9937240608451285, "1000–3000 Hz_40ms_modulation_mae_db": 0.9220161490185212, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8028691873378082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18274968012756168, "3000–6000 Hz_10ms_envelope_correlation": 0.9776551350297712, "3000–6000 Hz_10ms_modulation_mae_db": 0.21510101547747862, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2196917854889957, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12512642858894957, "3000–6000 Hz_40ms_envelope_correlation": 0.989119961830804, "3000–6000 Hz_40ms_modulation_mae_db": 0.23528060621262017, "mr_stft_db": 5.458795824190659, "band_mae_db": 0.2397047620849655, "rms_gain_error_db": -0.04112268593527396, "output_peak": 0.5638323026418425}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7477026657830089, "全帯域_10ms_rms_envelope_nrmse": 0.0497266096452902, "全帯域_10ms_envelope_correlation": 0.9967266096267898, "全帯域_10ms_modulation_mae_db": 0.18872672112054698, "全帯域_40ms_log_rms_mae_db": 0.4056176384206223, "全帯域_40ms_rms_envelope_nrmse": 0.025341152060761856, "全帯域_40ms_envelope_correlation": 0.9990679243919734, "全帯域_40ms_modulation_mae_db": 0.2778576981205172, "300–1000 Hz_10ms_log_rms_mae_db": 1.7553629469747356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09878051606175063, "300–1000 Hz_10ms_envelope_correlation": 0.9923206964198925, "300–1000 Hz_10ms_modulation_mae_db": 0.415726143627037, "300–1000 Hz_40ms_log_rms_mae_db": 1.1497801039313924, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05995646711329745, "300–1000 Hz_40ms_envelope_correlation": 0.997044245471078, "300–1000 Hz_40ms_modulation_mae_db": 0.5928210205721088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6794174590694153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15825361571685193, "1000–3000 Hz_10ms_envelope_correlation": 0.9793001979742335, "1000–3000 Hz_10ms_modulation_mae_db": 0.63650212947455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1006628877632016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09934661366709206, "1000–3000 Hz_40ms_envelope_correlation": 0.9931092628903334, "1000–3000 Hz_40ms_modulation_mae_db": 0.6083742170319639, "3000–6000 Hz_10ms_log_rms_mae_db": 1.821065400574152, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17155870603227294, "3000–6000 Hz_10ms_envelope_correlation": 0.9796263406290968, "3000–6000 Hz_10ms_modulation_mae_db": 0.6780236783403545, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1841993697426494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10201118154107437, "3000–6000 Hz_40ms_envelope_correlation": 0.9923809802530777, "3000–6000 Hz_40ms_modulation_mae_db": 0.6488071514447903, "mr_stft_db": 5.443899281975647, "band_mae_db": 0.1844264216475785, "rms_gain_error_db": -0.03528941994987792, "output_peak": 0.5593287600652239}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7179948787907423, "全帯域_10ms_rms_envelope_nrmse": 0.05142161074356839, "全帯域_10ms_envelope_correlation": 0.9964808773386171, "全帯域_10ms_modulation_mae_db": 0.2527919724819809, "全帯域_40ms_log_rms_mae_db": 0.39137495039498077, "全帯域_40ms_rms_envelope_nrmse": 0.02728815928608077, "全帯域_40ms_envelope_correlation": 0.9988970959105774, "全帯域_40ms_modulation_mae_db": 0.25125430894358874, "300–1000 Hz_10ms_log_rms_mae_db": 1.6655641523064426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09767812710533544, "300–1000 Hz_10ms_envelope_correlation": 0.9924387961882696, "300–1000 Hz_10ms_modulation_mae_db": 0.47281046379359815, "300–1000 Hz_40ms_log_rms_mae_db": 1.0420779721280113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05958551339384257, "300–1000 Hz_40ms_envelope_correlation": 0.9970191662837184, "300–1000 Hz_40ms_modulation_mae_db": 0.5372512694791655, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6751032076638819, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1504781802044563, "1000–3000 Hz_10ms_envelope_correlation": 0.9812175342810056, "1000–3000 Hz_10ms_modulation_mae_db": 0.6310467870737143, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0960617470099516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09459685257659159, "1000–3000 Hz_40ms_envelope_correlation": 0.9935953460679664, "1000–3000 Hz_40ms_modulation_mae_db": 0.47105278507835785, "3000–6000 Hz_10ms_log_rms_mae_db": 1.783979561246187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16451093709695364, "3000–6000 Hz_10ms_envelope_correlation": 0.9811611935084207, "3000–6000 Hz_10ms_modulation_mae_db": 0.5363695872134506, "3000–6000 Hz_40ms_log_rms_mae_db": 1.144388491209744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.100046534104896, "3000–6000 Hz_40ms_envelope_correlation": 0.9926540313006038, "3000–6000 Hz_40ms_modulation_mae_db": 0.33379278834161746, "mr_stft_db": 5.461795585142942, "band_mae_db": 0.17819821866237673, "rms_gain_error_db": -0.02889019017798393, "output_peak": 0.5531244239582706}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7635623294551283, "全帯域_10ms_rms_envelope_nrmse": 0.05436347687983302, "全帯域_10ms_envelope_correlation": 0.9961586922827933, "全帯域_10ms_modulation_mae_db": 0.5382973189211469, "全帯域_40ms_log_rms_mae_db": 0.38181837963773424, "全帯域_40ms_rms_envelope_nrmse": 0.027473764617090522, "全帯域_40ms_envelope_correlation": 0.9990034259726619, "全帯域_40ms_modulation_mae_db": 0.5053326266705117, "300–1000 Hz_10ms_log_rms_mae_db": 1.818139935131142, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10147561020023065, "300–1000 Hz_10ms_envelope_correlation": 0.9922038891594133, "300–1000 Hz_10ms_modulation_mae_db": 0.7110570719042068, "300–1000 Hz_40ms_log_rms_mae_db": 1.1744122693749472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06322270692348134, "300–1000 Hz_40ms_envelope_correlation": 0.9971070326785237, "300–1000 Hz_40ms_modulation_mae_db": 0.8452336657431762, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6739819651769887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14629013573633873, "1000–3000 Hz_10ms_envelope_correlation": 0.9826097763008109, "1000–3000 Hz_10ms_modulation_mae_db": 0.8065846747167185, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0201103488991032, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08901001788978381, "1000–3000 Hz_40ms_envelope_correlation": 0.9949470296248837, "1000–3000 Hz_40ms_modulation_mae_db": 0.9250407613707708, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8115195968284479, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1681271922779986, "3000–6000 Hz_10ms_envelope_correlation": 0.9803537069522752, "3000–6000 Hz_10ms_modulation_mae_db": 0.4939492034396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.189648465394942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10048390607935112, "3000–6000 Hz_40ms_envelope_correlation": 0.9926038706895635, "3000–6000 Hz_40ms_modulation_mae_db": 0.6160418998776342, "mr_stft_db": 5.460554272000699, "band_mae_db": 0.1989110283503721, "rms_gain_error_db": -0.08148226817814823, "output_peak": 0.5966564112443061}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7618441984112967, "全帯域_10ms_rms_envelope_nrmse": 0.05018755718255769, "全帯域_10ms_envelope_correlation": 0.9966727950142146, "全帯域_10ms_modulation_mae_db": 0.36230026125609965, "全帯域_40ms_log_rms_mae_db": 0.4099534311728644, "全帯域_40ms_rms_envelope_nrmse": 0.02776413207777687, "全帯域_40ms_envelope_correlation": 0.9988919031966171, "全帯域_40ms_modulation_mae_db": 0.3221214081629829, "300–1000 Hz_10ms_log_rms_mae_db": 1.741990421734224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10042812759354336, "300–1000 Hz_10ms_envelope_correlation": 0.9922739526894604, "300–1000 Hz_10ms_modulation_mae_db": 0.571354687972803, "300–1000 Hz_40ms_log_rms_mae_db": 1.0879740169501508, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06598931476500064, "300–1000 Hz_40ms_envelope_correlation": 0.9966404272606486, "300–1000 Hz_40ms_modulation_mae_db": 0.37217457013549843, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6523287896841892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1493856845825761, "1000–3000 Hz_10ms_envelope_correlation": 0.9817600939588009, "1000–3000 Hz_10ms_modulation_mae_db": 1.1345966371095006, "1000–3000 Hz_40ms_log_rms_mae_db": 1.031115495206334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09347981052665344, "1000–3000 Hz_40ms_envelope_correlation": 0.9941001476876506, "1000–3000 Hz_40ms_modulation_mae_db": 1.3809850344513375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8146337546319444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1786229943367557, "3000–6000 Hz_10ms_envelope_correlation": 0.9781977821125754, "3000–6000 Hz_10ms_modulation_mae_db": 0.07221501428169742, "3000–6000 Hz_40ms_log_rms_mae_db": 1.202087769392356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10718647623627431, "3000–6000 Hz_40ms_envelope_correlation": 0.9917631225159708, "3000–6000 Hz_40ms_modulation_mae_db": 0.17052276714919792, "mr_stft_db": 5.450959302997517, "band_mae_db": 0.21272265479474564, "rms_gain_error_db": -0.05166703143516424, "output_peak": 0.5590665643697971}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6707287024593835, "spectral_convergence": 0.16936077080688527, "spectral_convergence_status": "ok", "amplitude_floor": 2.2281588625700377e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.257552802253829, "spectral_convergence": 0.16185377787058433, "spectral_convergence_status": "ok", "amplitude_floor": 1.3568197926870596e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.45261059165922, "spectral_convergence": 0.16092847718117068, "spectral_convergence_status": "ok", "amplitude_floor": 1.1580179000740219e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0038028140043242973, "output_power": 0.0038232825308519027, "delta_db": 0.023313078791493823, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002493613270814922, "output_power": 0.002444207325447846, "delta_db": -0.08691059182473211, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006858506604602334, "output_power": 0.0006695996606653772, "delta_db": -0.10414336292933585, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002628229680145485, "output_power": 0.00023157861229248193, "delta_db": -0.5496486825707778, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001273968713381539, "output_power": 0.00011844837763258, "delta_db": -0.3162964591636761, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.457810201861998e-05, "output_power": 5.6446003374611105e-05, "delta_db": 0.14614768605433376, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20440997688905824, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.049110647579318864, "pitch": null, "output_rms_dbfs": -21.340932039524215, "output_peak": 0.5659362190017526, "reference_rms_dbfs": -21.2918213919449, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.460297365457477}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025738076998093343, "log_rms_mae_db": 0.763053590266098, "rms_envelope_nrmse": 0.05747624011365568, "envelope_correlation": 0.9956176902958148, "rms_gain_error_db": -0.04883677240612976, "reference_mean_power": 0.007432639167823318, "output_mean_power": 0.007349526706878513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.369590439777949e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014473681837183883, "output_power": 0.001413318726908984, "reference_active": true, "delta_db": -0.10338908131549968}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002579988221497507, "output_power": 0.00023233243211987468, "reference_active": true, "delta_db": -0.45507884508428575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2677922934851034e-05, "output_power": 3.037150778808005e-05, "reference_active": true, "delta_db": -0.31788091305778443}], "active_band_count": 3, "active_band_mae_db": 0.2921162798191899}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022253035095363898, "log_rms_mae_db": 0.4182091511232521, "rms_envelope_nrmse": 0.030583147359703026, "envelope_correlation": 0.998640382985547, "rms_gain_error_db": -0.048831767511823956, "reference_mean_power": 0.0074694638279480755, "output_mean_power": 0.007385948101942051}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0676309697735546e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013241614802002638, "output_power": 0.0012896418700895382, "reference_active": true, "delta_db": -0.11471825432777563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011975787910962369, "output_power": 0.00011108469730785794, "reference_active": true, "delta_db": -0.3264986016024573}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4293821094800366e-06, "output_power": 1.9380982187931454e-06, "reference_active": true, "delta_db": -0.981200465059697}], "active_band_count": 3, "active_band_mae_db": 0.47413910699664336}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017894181312046162, "log_rms_mae_db": 1.7029522333499512, "rms_envelope_nrmse": 0.09839085093866391, "envelope_correlation": 0.9923684302323444, "rms_gain_error_db": -0.14158215452752068, "reference_mean_power": 0.0017811195275832714, "output_mean_power": 0.001723990425953134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1758328243827778e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007391946250824912, "output_power": 0.0007176010621096576, "reference_active": true, "delta_db": -0.12875727502522616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015375544244981539, "output_power": 0.00013407035978078386, "reference_active": true, "delta_db": -0.594977225316454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0532547306567537e-05, "output_power": 9.767304730007505e-06, "reference_active": true, "delta_db": -0.3275868094394281}], "active_band_count": 3, "active_band_mae_db": 0.3504404365937028}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016320284141686103, "log_rms_mae_db": 1.088139671337801, "rms_envelope_nrmse": 0.06601351451831762, "envelope_correlation": 0.9963542736643743, "rms_gain_error_db": -0.1415893443490811, "reference_mean_power": 0.0017900337371132, "output_mean_power": 0.00173261584537157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0791360125207298e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007228891058818755, "output_power": 0.0006992044966018345, "reference_active": true, "delta_db": -0.14467467276579996}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980895739386466e-05, "output_power": 6.182559605792947e-05, "reference_active": true, "delta_db": -0.5274283974047279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.398569514368023e-07, "output_power": 8.311643173830985e-07, "reference_active": true, "delta_db": -0.5337486762598843}], "active_band_count": 3, "active_band_mae_db": 0.40195058214347074}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.442874453255196e-05, "log_rms_mae_db": 1.7216996486733613, "rms_envelope_nrmse": 0.15207152532689233, "envelope_correlation": 0.981063608501306, "rms_gain_error_db": -0.5044397847554376, "reference_mean_power": 0.000318038298123639, "output_mean_power": 0.0002831623072266758}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6089659140343417e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.507437744605689e-05, "output_power": 8.031669039766023e-05, "reference_active": true, "delta_db": -0.24992975478739438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3978123456671057e-05, "output_power": 1.9851961478630382e-05, "reference_active": true, "delta_db": -0.8201176812043022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6559576292684446e-06, "output_power": 3.6167213911992855e-06, "reference_active": true, "delta_db": -1.096939646235657}], "active_band_count": 3, "active_band_mae_db": 0.7223290274091179}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.707405683354541e-05, "log_rms_mae_db": 1.1192597784660199, "rms_envelope_nrmse": 0.09663818170365046, "envelope_correlation": 0.9935501828074842, "rms_gain_error_db": -0.5046975907823206, "reference_mean_power": 0.00031952379334241086, "output_mean_power": 0.0002844680163543249}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4169476048907225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.318339905927379e-05, "output_power": 7.838315249598666e-05, "reference_active": true, "delta_db": -0.25813936227829226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1060276940136043e-05, "output_power": 9.245727690971671e-06, "reference_active": true, "delta_db": -0.7782490320926193}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55972585077424e-07, "output_power": 2.175483933578251e-07, "reference_active": true, "delta_db": -0.7063757414310385}], "active_band_count": 3, "active_band_mae_db": 0.5809213786006501}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.0412968032519945e-05, "log_rms_mae_db": 1.8299454157824957, "rms_envelope_nrmse": 0.1673085162732792, "envelope_correlation": 0.9806949667334076, "rms_gain_error_db": 0.0029123049545257094, "reference_mean_power": 8.301357548516487e-05, "output_mean_power": 8.306926164645284e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.362198065588504e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2275109921127774e-05, "output_power": 3.110097112091524e-05, "reference_active": true, "delta_db": -0.1609378005443354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8777051155174143e-06, "output_power": 2.251752375591233e-06, "reference_active": true, "delta_db": -1.0652565921831965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.623133887825791e-07, "output_power": 7.189103008502492e-07, "reference_active": true, "delta_db": 0.3561117193519576}], "active_band_count": 3, "active_band_mae_db": 0.5274353706931632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.404038681520291e-05, "log_rms_mae_db": 1.2127009800602626, "rms_envelope_nrmse": 0.09733490687143494, "envelope_correlation": 0.9931241204825552, "rms_gain_error_db": 0.002098655806355645, "reference_mean_power": 8.336414632310516e-05, "output_mean_power": 8.34044403943639e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1059391757915914e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.094184470310106e-05, "output_power": 2.9899926526840323e-05, "reference_active": true, "delta_db": -0.1487608094201593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2229001330453808e-06, "output_power": 9.972922864419733e-07, "reference_active": true, "delta_db": -0.8856853224400318}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7483618813517085e-08, "output_power": 4.860133538513484e-08, "reference_active": true, "delta_db": 0.10104392312676251}], "active_band_count": 3, "active_band_mae_db": 0.37849668499565126}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9201279194059577, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3d936d26cb6dbd5e7a5521485ef39e57168f3ba170b23cdeb9d8a7156d595c0a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9812508211889457, "standard_deviation": 0.0009755350002761891, "minimum": 0.9793001979742335, "maximum": 0.9826097763008109, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6881752064135025, "standard_deviation": 0.030184837789706147, "minimum": 1.6523287896841892, "maximum": 1.7451760548325574, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8247204795633492, "standard_deviation": 0.19714361624169466, "minimum": 0.6310467870737143, "maximum": 1.1345966371095006, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15147923105595884, "standard_deviation": 0.003664023294020856, "minimum": 0.14629013573633873, "maximum": 0.15825361571685193, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9940633309564908, "standard_deviation": 0.0006859211534169551, "minimum": 0.9931092628903334, "maximum": 0.994948703748437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0813165792499015, "standard_deviation": 0.03756966617165048, "minimum": 1.0201103488991032, "maximum": 1.1209972619380304, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8324444086181899, "standard_deviation": 0.2847136282894141, "minimum": 0.47105278507835785, "maximum": 1.3809850344513375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09424304594642902, "standard_deviation": 0.0049961933875980104, "minimum": 0.08848403807847612, "maximum": 0.10228032524321076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9800791786058404, "standard_deviation": 0.0017182055827364932, "minimum": 0.9776551350297712, "maximum": 0.9831363100178157, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7999480855090202, "standard_deviation": 0.025624428486063906, "minimum": 1.750924852731757, "maximum": 1.8299454157824957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4254168626684258, "standard_deviation": 0.20353367472829978, "minimum": 0.07221501428169742, "maximum": 0.6780236783403545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1699834453511097, "standard_deviation": 0.008036864832361997, "minimum": 0.15688858772109518, "maximum": 0.18274968012756168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920818910869715, "standard_deviation": 0.0013455082168381746, "minimum": 0.989119961830804, "maximum": 0.9934170373513117, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1824657247949188, "standard_deviation": 0.02922036116788259, "minimum": 1.144388491209744, "maximum": 1.2196917854889957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3960830866351543, "standard_deviation": 0.22662095078517333, "minimum": 0.10450134646146696, "maximum": 0.6812214485982561, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10483917473620191, "standard_deviation": 0.008982645204804624, "minimum": 0.09733490687143494, "maximum": 0.12512642858894957, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9926144630029277, "standard_deviation": 0.0006765378859220354, "minimum": 0.9919423762419961, "maximum": 0.993723352724587, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7240598387225003, "standard_deviation": 0.04940233171391902, "minimum": 1.6655641523064426, "maximum": 1.818139935131142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5297873332209787, "standard_deviation": 0.2121936718798692, "minimum": 0.3214894864086122, "maximum": 0.9549947835459361, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09745500769172774, "standard_deviation": 0.004491722289146453, "minimum": 0.0905549370702052, "maximum": 0.10176091754494532, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971946589900742, "standard_deviation": 0.0005687457652539448, "minimum": 0.9963542736643743, "maximum": 0.9980273310951956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0977096960085693, "standard_deviation": 0.04395359466416396, "minimum": 1.0420779721280113, "maximum": 1.1744122693749472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5800487534220178, "standard_deviation": 0.2180083638268498, "minimum": 0.32179108142658985, "maximum": 0.917020330145097, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05946260163749425, "standard_deviation": 0.005571340957697432, "minimum": 0.050634053245140066, "maximum": 0.06601351451831762, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21065766465964675, "standard_deviation": 0.0244619748834459, "minimum": 0.17819821866237673, "maximum": 0.24729264487536382, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.455832223457252, "standard_deviation": 0.006337472630658359, "minimum": 5.443899281975647, "maximum": 5.461795585142942, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5629833463464817, "standard_deviation": 0.014495273056692747, "minimum": 0.5523751641140909, "maximum": 0.5966564112443061, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04717353404393855, "standard_deviation": 0.018275702972493833, "minimum": -0.08148226817814823, "maximum": -0.027382604140720673, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964164193506191, "standard_deviation": 0.0003755586422428433, "minimum": 0.9956176902958148, "maximum": 0.9967491990048162, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7533186561297663, "standard_deviation": 0.015854027877519945, "minimum": 0.7179948787907423, "maximum": 0.7672810215089961, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3407276065297548, "standard_deviation": 0.1132420561865095, "minimum": 0.18872672112054698, "maximum": 0.5382973189211469, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.052014858862459185, "standard_deviation": 0.0027021519289611427, "minimum": 0.04942731436487262, "maximum": 0.05747624011365568, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989230931294002, "standard_deviation": 0.00015699901210177206, "minimum": 0.998640382985547, "maximum": 0.999148324696194, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4023482895565763, "standard_deviation": 0.02202026744772174, "minimum": 0.37267562581323543, "maximum": 0.4425763625170737, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35839063845938934, "standard_deviation": 0.09987090517158284, "minimum": 0.2509625658455141, "maximum": 0.5053326266705117, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027325564733683846, "standard_deviation": 0.002018458495800382, "minimum": 0.024022696214601784, "maximum": 0.030583147359703026, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0039", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.9800336360931396, "predicted_mos": 2.9800336360931396, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cb45701ed31e01337a8209c73c01a4509223aea837d5d3bdd42abad60bebea0d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 14.012258799746633, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7780437736514727, "全帯域_10ms_rms_envelope_nrmse": 0.05748521746941324, "全帯域_10ms_envelope_correlation": 0.9956166953134198, "全帯域_10ms_modulation_mae_db": 0.31179078854882375, "全帯域_40ms_log_rms_mae_db": 0.4246359701649067, "全帯域_40ms_rms_envelope_nrmse": 0.029887473126233837, "全帯域_40ms_envelope_correlation": 0.9987040733950115, "全帯域_40ms_modulation_mae_db": 0.5082193238052065, "300–1000 Hz_10ms_log_rms_mae_db": 1.6818464504297397, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0983784137072582, "300–1000 Hz_10ms_envelope_correlation": 0.9923547168463339, "300–1000 Hz_10ms_modulation_mae_db": 0.4831741542004098, "300–1000 Hz_40ms_log_rms_mae_db": 1.04760086800771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0652153774519225, "300–1000 Hz_40ms_envelope_correlation": 0.9964176975527774, "300–1000 Hz_40ms_modulation_mae_db": 0.6326378013290379, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6986196595208949, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1503840553778096, "1000–3000 Hz_10ms_envelope_correlation": 0.9817398634409867, "1000–3000 Hz_10ms_modulation_mae_db": 0.629668786089992, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1306528655542796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09639663354657642, "1000–3000 Hz_40ms_envelope_correlation": 0.9938582412148387, "1000–3000 Hz_40ms_modulation_mae_db": 0.5187114187313568, "3000–6000 Hz_10ms_log_rms_mae_db": 1.825959806256586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16903718204108295, "3000–6000 Hz_10ms_envelope_correlation": 0.9803675125111261, "3000–6000 Hz_10ms_modulation_mae_db": 0.5762869675562693, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2929875985169843, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10081206978327426, "3000–6000 Hz_40ms_envelope_correlation": 0.9927612440565544, "3000–6000 Hz_40ms_modulation_mae_db": 0.4927840886134427, "mr_stft_db": 5.398517245695872, "band_mae_db": 0.21723854680804291, "rms_gain_error_db": -0.04999438705033531, "output_peak": 0.5682734239862989}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7515751635146433, "全帯域_10ms_rms_envelope_nrmse": 0.049387443075870945, "全帯域_10ms_envelope_correlation": 0.9967608780161662, "全帯域_10ms_modulation_mae_db": 0.25111130471275017, "全帯域_40ms_log_rms_mae_db": 0.4039672955057328, "全帯域_40ms_rms_envelope_nrmse": 0.023934922798394494, "全帯域_40ms_envelope_correlation": 0.9991635694232878, "全帯域_40ms_modulation_mae_db": 0.25354043905697077, "300–1000 Hz_10ms_log_rms_mae_db": 1.7039900853078294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09268294322881665, "300–1000 Hz_10ms_envelope_correlation": 0.993348393698199, "300–1000 Hz_10ms_modulation_mae_db": 0.5134830137211406, "300–1000 Hz_40ms_log_rms_mae_db": 1.0660981408317476, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050982135488617464, "300–1000 Hz_40ms_envelope_correlation": 0.9980109569192233, "300–1000 Hz_40ms_modulation_mae_db": 0.7262519191972029, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7596193683342967, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1523304800315472, "1000–3000 Hz_10ms_envelope_correlation": 0.980779607578513, "1000–3000 Hz_10ms_modulation_mae_db": 0.9086853625762901, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1231140429781945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0909374048301009, "1000–3000 Hz_40ms_envelope_correlation": 0.9944222960423336, "1000–3000 Hz_40ms_modulation_mae_db": 0.8152255810864326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.772683659361203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1515650581811373, "3000–6000 Hz_10ms_envelope_correlation": 0.9844465855908638, "3000–6000 Hz_10ms_modulation_mae_db": 0.5870411965474228, "3000–6000 Hz_40ms_log_rms_mae_db": 1.237406031596994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09716944767327428, "3000–6000 Hz_40ms_envelope_correlation": 0.9938313649671321, "3000–6000 Hz_40ms_modulation_mae_db": 0.6030401057082995, "mr_stft_db": 5.393921109744023, "band_mae_db": 0.25851744120055314, "rms_gain_error_db": -0.03597006388731143, "output_peak": 0.5493193408052117}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7713112106507172, "全帯域_10ms_rms_envelope_nrmse": 0.05118226428304922, "全帯域_10ms_envelope_correlation": 0.9965662670272425, "全帯域_10ms_modulation_mae_db": 0.3504088451429082, "全帯域_40ms_log_rms_mae_db": 0.4425365886217314, "全帯域_40ms_rms_envelope_nrmse": 0.028248630174544855, "全帯域_40ms_envelope_correlation": 0.9988802483092047, "全帯域_40ms_modulation_mae_db": 0.31251369268789814, "300–1000 Hz_10ms_log_rms_mae_db": 1.7041968403644558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09486250853338628, "300–1000 Hz_10ms_envelope_correlation": 0.9930694265814353, "300–1000 Hz_10ms_modulation_mae_db": 0.9950767882866177, "300–1000 Hz_40ms_log_rms_mae_db": 1.0554866534293426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05531013151381753, "300–1000 Hz_40ms_envelope_correlation": 0.9976596200281896, "300–1000 Hz_40ms_modulation_mae_db": 0.9197771615584763, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6805807812401807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1485148870379841, "1000–3000 Hz_10ms_envelope_correlation": 0.9820659600467272, "1000–3000 Hz_10ms_modulation_mae_db": 0.5454506120374211, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0925418513234337, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08903763975986059, "1000–3000 Hz_40ms_envelope_correlation": 0.9950242235577332, "1000–3000 Hz_40ms_modulation_mae_db": 0.8204131350931542, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7719425390472379, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1697373886094024, "3000–6000 Hz_10ms_envelope_correlation": 0.979985796626975, "3000–6000 Hz_10ms_modulation_mae_db": 0.3386601740207231, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2808168968889606, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10739442980766765, "3000–6000 Hz_40ms_envelope_correlation": 0.9917031084750935, "3000–6000 Hz_40ms_modulation_mae_db": 0.166414076327826, "mr_stft_db": 5.413639384625938, "band_mae_db": 0.22635717644546957, "rms_gain_error_db": -0.06429852474805581, "output_peak": 0.5517504782200433}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7525664303736073, "全帯域_10ms_rms_envelope_nrmse": 0.05194064394795348, "全帯域_10ms_envelope_correlation": 0.9964184037094393, "全帯域_10ms_modulation_mae_db": 0.3913627227806433, "全帯域_40ms_log_rms_mae_db": 0.3807751579267247, "全帯域_40ms_rms_envelope_nrmse": 0.02716491946153445, "全帯域_40ms_envelope_correlation": 0.9989211980051955, "全帯域_40ms_modulation_mae_db": 0.4109934935182764, "300–1000 Hz_10ms_log_rms_mae_db": 1.712824057001264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10649351227736387, "300–1000 Hz_10ms_envelope_correlation": 0.9911062854900705, "300–1000 Hz_10ms_modulation_mae_db": 0.31373103346735626, "300–1000 Hz_40ms_log_rms_mae_db": 1.0524087840639285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05870948322135098, "300–1000 Hz_40ms_envelope_correlation": 0.9972832866576716, "300–1000 Hz_40ms_modulation_mae_db": 0.2783990367937006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6956711593528233, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15777228146381947, "1000–3000 Hz_10ms_envelope_correlation": 0.9805560423441356, "1000–3000 Hz_10ms_modulation_mae_db": 0.9705240399854421, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1440185499509805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10305439144574884, "1000–3000 Hz_40ms_envelope_correlation": 0.9937387191773155, "1000–3000 Hz_40ms_modulation_mae_db": 0.9040201801102512, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8016428278236727, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18550279930300773, "3000–6000 Hz_10ms_envelope_correlation": 0.9770231051274981, "3000–6000 Hz_10ms_modulation_mae_db": 0.2600402370323171, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2994148995933905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12902318652792708, "3000–6000 Hz_40ms_envelope_correlation": 0.9885084666331446, "3000–6000 Hz_40ms_modulation_mae_db": 0.2912053306026002, "mr_stft_db": 5.4067489679815735, "band_mae_db": 0.24948133160619898, "rms_gain_error_db": -0.03809114782093209, "output_peak": 0.5614642087682927}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7596203985534976, "全帯域_10ms_rms_envelope_nrmse": 0.05037154531461528, "全帯域_10ms_envelope_correlation": 0.9966399334267361, "全帯域_10ms_modulation_mae_db": 0.20006891221278664, "全帯域_40ms_log_rms_mae_db": 0.4186504983739074, "全帯域_40ms_rms_envelope_nrmse": 0.02598721582356711, "全帯域_40ms_envelope_correlation": 0.9990185818647745, "全帯域_40ms_modulation_mae_db": 0.28277075919426947, "300–1000 Hz_10ms_log_rms_mae_db": 1.765118780983024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10017855999133204, "300–1000 Hz_10ms_envelope_correlation": 0.9920737360908027, "300–1000 Hz_10ms_modulation_mae_db": 0.3570589223670033, "300–1000 Hz_40ms_log_rms_mae_db": 1.1299677909210175, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05911726873209233, "300–1000 Hz_40ms_envelope_correlation": 0.9971034976816957, "300–1000 Hz_40ms_modulation_mae_db": 0.561169818222819, "1000–3000 Hz_10ms_log_rms_mae_db": 1.675832289560623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15392437298464817, "1000–3000 Hz_10ms_envelope_correlation": 0.980499745646894, "1000–3000 Hz_10ms_modulation_mae_db": 0.5803467208446788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.123594981063385, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09530657575927112, "1000–3000 Hz_40ms_envelope_correlation": 0.993822632026184, "1000–3000 Hz_40ms_modulation_mae_db": 0.5641987389965291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8055771688008713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1696904647670318, "3000–6000 Hz_10ms_envelope_correlation": 0.9801360744286771, "3000–6000 Hz_10ms_modulation_mae_db": 0.771219450356256, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2587731514866654, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10155769034233615, "3000–6000 Hz_40ms_envelope_correlation": 0.9925167505351999, "3000–6000 Hz_40ms_modulation_mae_db": 0.8257911795396957, "mr_stft_db": 5.391569525796214, "band_mae_db": 0.18219854260015608, "rms_gain_error_db": -0.035079615309434466, "output_peak": 0.5591693064986496}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7050919785333878, "全帯域_10ms_rms_envelope_nrmse": 0.05197177439164491, "全帯域_10ms_envelope_correlation": 0.9964084315577271, "全帯域_10ms_modulation_mae_db": 0.3116106507204451, "全帯域_40ms_log_rms_mae_db": 0.3826515744402213, "全帯域_40ms_rms_envelope_nrmse": 0.02740998489924144, "全帯域_40ms_envelope_correlation": 0.9988929653503987, "全帯域_40ms_modulation_mae_db": 0.28715592477671775, "300–1000 Hz_10ms_log_rms_mae_db": 1.666950025865854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0995472176314692, "300–1000 Hz_10ms_envelope_correlation": 0.9921499876343313, "300–1000 Hz_10ms_modulation_mae_db": 0.5424065722823821, "300–1000 Hz_40ms_log_rms_mae_db": 1.0197056959918982, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06105550170524606, "300–1000 Hz_40ms_envelope_correlation": 0.9968750776642887, "300–1000 Hz_40ms_modulation_mae_db": 0.7231011841733787, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6945111496125391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14877362465664268, "1000–3000 Hz_10ms_envelope_correlation": 0.9818742093649823, "1000–3000 Hz_10ms_modulation_mae_db": 0.5902559103795425, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1554827264586445, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09385918227524193, "1000–3000 Hz_40ms_envelope_correlation": 0.9939372092783392, "1000–3000 Hz_40ms_modulation_mae_db": 0.5183741360179378, "3000–6000 Hz_10ms_log_rms_mae_db": 1.769268134616834, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16370115572818586, "3000–6000 Hz_10ms_envelope_correlation": 0.9813843218943936, "3000–6000 Hz_10ms_modulation_mae_db": 0.43447154164959473, "3000–6000 Hz_40ms_log_rms_mae_db": 1.218649444971345, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10134502011171835, "3000–6000 Hz_40ms_envelope_correlation": 0.9925802055245744, "3000–6000 Hz_40ms_modulation_mae_db": 0.3365334319684794, "mr_stft_db": 5.4076904406884205, "band_mae_db": 0.18284627922552263, "rms_gain_error_db": -0.03417852231787778, "output_peak": 0.5518041370390535}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7744228764649174, "全帯域_10ms_rms_envelope_nrmse": 0.05525946749406692, "全帯域_10ms_envelope_correlation": 0.9960255195999329, "全帯域_10ms_modulation_mae_db": 0.4230754347617606, "全帯域_40ms_log_rms_mae_db": 0.3897465550500201, "全帯域_40ms_rms_envelope_nrmse": 0.027813994522194598, "全帯域_40ms_envelope_correlation": 0.9989777333379596, "全帯域_40ms_modulation_mae_db": 0.3817354226591931, "300–1000 Hz_10ms_log_rms_mae_db": 1.7819520631048296, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10407590721570475, "300–1000 Hz_10ms_envelope_correlation": 0.9917313982555552, "300–1000 Hz_10ms_modulation_mae_db": 0.7049987906840675, "300–1000 Hz_40ms_log_rms_mae_db": 1.1516066125626958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06437108985674593, "300–1000 Hz_40ms_envelope_correlation": 0.9969343159742606, "300–1000 Hz_40ms_modulation_mae_db": 0.8320727336376473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7155077806967092, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15051934890586566, "1000–3000 Hz_10ms_envelope_correlation": 0.9815365189219805, "1000–3000 Hz_10ms_modulation_mae_db": 0.7437517903729227, "1000–3000 Hz_40ms_log_rms_mae_db": 1.086954812371246, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09383521244972794, "1000–3000 Hz_40ms_envelope_correlation": 0.9941912815639322, "1000–3000 Hz_40ms_modulation_mae_db": 0.8204018648874088, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8035364679706347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1662550084342474, "3000–6000 Hz_10ms_envelope_correlation": 0.9808681354611289, "3000–6000 Hz_10ms_modulation_mae_db": 0.38183651672825447, "3000–6000 Hz_40ms_log_rms_mae_db": 1.280654460976334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09947928031013983, "3000–6000 Hz_40ms_envelope_correlation": 0.9928670055863799, "3000–6000 Hz_40ms_modulation_mae_db": 0.5482473454585396, "mr_stft_db": 5.409769283003939, "band_mae_db": 0.20784092293348244, "rms_gain_error_db": -0.08374483763541324, "output_peak": 0.593695042300414}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7342367255248178, "全帯域_10ms_rms_envelope_nrmse": 0.051019738887439374, "全帯域_10ms_envelope_correlation": 0.9965608233973648, "全帯域_10ms_modulation_mae_db": 0.2725469177215322, "全帯域_40ms_log_rms_mae_db": 0.3889833249320248, "全帯域_40ms_rms_envelope_nrmse": 0.027720670883255614, "全帯域_40ms_envelope_correlation": 0.9988973601455816, "全帯域_40ms_modulation_mae_db": 0.2253419971817984, "300–1000 Hz_10ms_log_rms_mae_db": 1.7321355882254803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10009569891061464, "300–1000 Hz_10ms_envelope_correlation": 0.9922866777177219, "300–1000 Hz_10ms_modulation_mae_db": 0.639486183834323, "300–1000 Hz_40ms_log_rms_mae_db": 1.0874041195520099, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06369905127264075, "300–1000 Hz_40ms_envelope_correlation": 0.9968449788332153, "300–1000 Hz_40ms_modulation_mae_db": 0.44318343616302425, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6468736933561263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1483314232237251, "1000–3000 Hz_10ms_envelope_correlation": 0.9819922960620737, "1000–3000 Hz_10ms_modulation_mae_db": 1.0505183028326213, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0519200823771737, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09432822131129982, "1000–3000 Hz_40ms_envelope_correlation": 0.9938420071179304, "1000–3000 Hz_40ms_modulation_mae_db": 1.3056220324766346, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8285017598718187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17847741730509917, "3000–6000 Hz_10ms_envelope_correlation": 0.9785010808425607, "3000–6000 Hz_10ms_modulation_mae_db": 0.10740410491505918, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3187090313037328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10958399526107336, "3000–6000 Hz_40ms_envelope_correlation": 0.9916291397375643, "3000–6000 Hz_40ms_modulation_mae_db": 0.1915415560477398, "mr_stft_db": 5.404327612826002, "band_mae_db": 0.21431023874850932, "rms_gain_error_db": -0.052544261580676724, "output_peak": 0.5577969217552295}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.601290223651258, "spectral_convergence": 0.1689108919141667, "spectral_convergence_status": "ok", "amplitude_floor": 2.2281588625700377e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2000764368332115, "spectral_convergence": 0.16153112968402078, "spectral_convergence_status": "ok", "amplitude_floor": 1.3568197926870596e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.394185076603146, "spectral_convergence": 0.1599201625530674, "spectral_convergence_status": "ok", "amplitude_floor": 1.1580179000740219e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0038028140043242973, "output_power": 0.003819436027284251, "delta_db": 0.01894155731679165, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002493613270814922, "output_power": 0.002450806162761587, "delta_db": -0.07520136824334286, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006858506604602334, "output_power": 0.000668343803502825, "delta_db": -0.11229635310481048, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002628229680145485, "output_power": 0.0002313795340161023, "delta_db": -0.5533837330463747, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001273968713381539, "output_power": 0.0001158524102442893, "delta_db": -0.41253689085114537, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.457810201861998e-05, "output_power": 5.625039372130395e-05, "delta_db": 0.13107137828579257, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21723854680804291, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04999438705033531, "pitch": null, "output_rms_dbfs": -21.341815778995233, "output_peak": 0.5682734239862989, "reference_rms_dbfs": -21.2918213919449, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.398517245695872}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025738076998093343, "log_rms_mae_db": 0.7780437736514727, "rms_envelope_nrmse": 0.05748521746941324, "envelope_correlation": 0.9956166953134198, "rms_gain_error_db": -0.049762495316255304, "reference_mean_power": 0.007432639167823318, "output_mean_power": 0.007347960281222468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.369590439777949e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014473681837183883, "output_power": 0.0014131875138191218, "reference_active": true, "delta_db": -0.103792300810298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002579988221497507, "output_power": 0.00023102749248683202, "reference_active": true, "delta_db": -0.47954058853914466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2677922934851034e-05, "output_power": 3.0133563749654985e-05, "reference_active": true, "delta_db": -0.3520394762970286}], "active_band_count": 3, "active_band_mae_db": 0.31179078854882375}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022253035095363898, "log_rms_mae_db": 0.4246359701649067, "rms_envelope_nrmse": 0.029887473126233837, "envelope_correlation": 0.9987040733950115, "rms_gain_error_db": -0.04984992755239826, "reference_mean_power": 0.0074694638279480755, "output_mean_power": 0.007384216743129573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0676309697735546e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013241614802002638, "output_power": 0.0012897594949017649, "reference_active": true, "delta_db": -0.11432216392002659}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011975787910962369, "output_power": 0.00011063975568555002, "reference_active": true, "delta_db": -0.34392888032131835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4293821094800366e-06, "output_power": 1.9004442596842556e-06, "reference_active": true, "delta_db": -1.0664069271742747}], "active_band_count": 3, "active_band_mae_db": 0.5082193238052065}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017894181312046162, "log_rms_mae_db": 1.6818464504297397, "rms_envelope_nrmse": 0.0983784137072582, "envelope_correlation": 0.9923547168463339, "rms_gain_error_db": -0.12933652480294736, "reference_mean_power": 0.0017811195275832714, "output_mean_power": 0.001728858353292041}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1758328243827778e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007391946250824912, "output_power": 0.0007178980838551634, "reference_active": true, "delta_db": -0.12696006173837854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015375544244981539, "output_power": 0.00013329243415272996, "reference_active": true, "delta_db": -0.6202499834313497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0532547306567537e-05, "output_power": 8.95987965957713e-06, "reference_active": true, "delta_db": -0.7023124174315012}], "active_band_count": 3, "active_band_mae_db": 0.4831741542004098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016320284141686103, "log_rms_mae_db": 1.04760086800771, "rms_envelope_nrmse": 0.0652153774519225, "envelope_correlation": 0.9964176975527774, "rms_gain_error_db": -0.1293282453392906, "reference_mean_power": 0.0017900337371132, "output_mean_power": 0.0017375143166860837}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0791360125207298e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007228891058818755, "output_power": 0.0006985637293968658, "reference_active": true, "delta_db": -0.14865647283279287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980895739386466e-05, "output_power": 6.138094890478905e-05, "reference_active": true, "delta_db": -0.5587755378427712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.398569514368023e-07, "output_power": 7.145187315907747e-07, "reference_active": true, "delta_db": -1.1904813933115497}], "active_band_count": 3, "active_band_mae_db": 0.6326378013290379}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.442874453255196e-05, "log_rms_mae_db": 1.6986196595208949, "rms_envelope_nrmse": 0.1503840553778096, "envelope_correlation": 0.9817398634409867, "rms_gain_error_db": -0.5304867044708618, "reference_mean_power": 0.000318038298123639, "output_mean_power": 0.00028146911678807174}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6089659140343417e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.507437744605689e-05, "output_power": 8.053553819391813e-05, "reference_active": true, "delta_db": -0.23811214432972574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3978123456671057e-05, "output_power": 2.0963194725528486e-05, "reference_active": true, "delta_db": -0.583577234472404}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6559576292684446e-06, "output_power": 3.6414749058787295e-06, "reference_active": true, "delta_db": -1.0673169794678463}], "active_band_count": 3, "active_band_mae_db": 0.629668786089992}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.707405683354541e-05, "log_rms_mae_db": 1.1306528655542796, "rms_envelope_nrmse": 0.09639663354657642, "envelope_correlation": 0.9938582412148387, "rms_gain_error_db": -0.530812492697494, "reference_mean_power": 0.00031952379334241086, "output_mean_power": 0.00028276259207599385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4169476048907225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.318339905927379e-05, "output_power": 7.893783216297662e-05, "reference_active": true, "delta_db": -0.22751467259371683}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1060276940136043e-05, "output_power": 9.711815192286115e-06, "reference_active": true, "delta_db": -0.5646559192811441}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55972585077424e-07, "output_power": 2.1468271945252558e-07, "reference_active": true, "delta_db": -0.7639636643192095}], "active_band_count": 3, "active_band_mae_db": 0.5187114187313568}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.0412968032519945e-05, "log_rms_mae_db": 1.825959806256586, "rms_envelope_nrmse": 0.16903718204108295, "envelope_correlation": 0.9803675125111261, "rms_gain_error_db": -0.05049856799284332, "reference_mean_power": 8.301357548516487e-05, "output_mean_power": 8.205390665435592e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.362198065588504e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2275109921127774e-05, "output_power": 3.1183984361885924e-05, "reference_active": true, "delta_db": -0.14936126057697538}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8777051155174143e-06, "output_power": 2.3015001752374426e-06, "reference_active": true, "delta_db": -0.9703527651617747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.623133887825791e-07, "output_power": 7.620407962953539e-07, "reference_active": true, "delta_db": 0.6091468769300576}], "active_band_count": 3, "active_band_mae_db": 0.5762869675562693}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.404038681520291e-05, "log_rms_mae_db": 1.2929875985169843, "rms_envelope_nrmse": 0.10081206978327426, "envelope_correlation": 0.9927612440565544, "rms_gain_error_db": -0.051665142718265084, "reference_mean_power": 8.336414632310516e-05, "output_mean_power": 8.237829384347458e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1059391757915914e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.094184470310106e-05, "output_power": 3.002544612073128e-05, "reference_active": true, "delta_db": -0.13056733166241172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2229001330453808e-06, "output_power": 9.919645618596009e-07, "reference_active": true, "delta_db": -0.9089483507567176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7483618813517085e-08, "output_power": 5.253242731163571e-08, "reference_active": true, "delta_db": 0.4388365834211986}], "active_band_count": 3, "active_band_mae_db": 0.4927840886134427}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9189301856246772, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9acefdd6595c820c494d3c353359d594c1db1b7c0cdfc12af931cf31c5436c0d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9813805304257865, "standard_deviation": 0.0006609423113637841, "minimum": 0.980499745646894, "maximum": 0.9820659600467272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6959019852092743, "standard_deviation": 0.032721707437598925, "minimum": 1.6468736933561263, "maximum": 1.7596193683342967, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7524001906398639, "standard_deviation": 0.19811128261708108, "minimum": 0.5454506120374211, "maximum": 1.0505183028326213, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15131880921025526, "standard_deviation": 0.0032565771921449415, "minimum": 0.1483314232237251, "maximum": 0.15777228146381947, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941045762473257, "standard_deviation": 0.00043510162907543277, "minimum": 0.9937387191773155, "maximum": 0.9950242235577332, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.113534989009667, "standard_deviation": 0.034055538497691475, "minimum": 1.0519200823771737, "maximum": 1.1554827264586445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7833708859249632, "standard_deviation": 0.2616881995586783, "minimum": 0.5183741360179378, "maximum": 1.3056220324766346, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09459440767222844, "standard_deviation": 0.004153581162302435, "minimum": 0.08903763975986059, "maximum": 0.10305439144574884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.980339076560403, "standard_deviation": 0.0021641491745115547, "minimum": 0.9770231051274981, "maximum": 0.9844465855908638, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7973890454686072, "standard_deviation": 0.023772392811497223, "minimum": 1.769268134616834, "maximum": 1.8285017598718187, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4321200236007371, "standard_deviation": 0.20911161877945222, "minimum": 0.10740410491505918, "maximum": 0.771219450356256, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16924580929614935, "standard_deviation": 0.010016359743153366, "minimum": 0.1515650581811373, "maximum": 0.18550279930300773, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920496606894553, "standard_deviation": 0.0015894974992444392, "minimum": 0.9885084666331446, "maximum": 0.9938313649671321, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2734264394168007, "standard_deviation": 0.033226759955859934, "minimum": 1.218649444971345, "maximum": 1.3187090313037328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4319446392833278, "standard_deviation": 0.2264090759612557, "minimum": 0.166414076327826, "maximum": 0.8257911795396957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10579563997717636, "standard_deviation": 0.010235362250030673, "minimum": 0.09716944767327428, "maximum": 0.12902318652792708, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922650777893063, "standard_deviation": 0.0007077653850760692, "minimum": 0.9911062854900705, "maximum": 0.993348393698199, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7186267364103096, "standard_deviation": 0.03934007894556894, "minimum": 1.666950025865854, "maximum": 1.7819520631048296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5686769323554125, "standard_deviation": 0.21588398185382032, "minimum": 0.31373103346735626, "maximum": 0.9950767882866177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0995393451869932, "standard_deviation": 0.0044703760270992455, "minimum": 0.09268294322881665, "maximum": 0.10649351227736387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971411789139153, "standard_deviation": 0.000503086384539132, "minimum": 0.9964176975527774, "maximum": 0.9980109569192233, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0762848331700439, "standard_deviation": 0.04441562829344288, "minimum": 1.0197056959918982, "maximum": 1.1516066125626958, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.639574136384411, "standard_deviation": 0.20883053982836156, "minimum": 0.2783990367937006, "maximum": 0.9197771615584763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05980750490530419, "standard_deviation": 0.00487806478882723, "minimum": 0.050982135488617464, "maximum": 0.0652153774519225, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2173488099459919, "standard_deviation": 0.027604649513990238, "minimum": 0.18219854260015608, "maximum": 0.25851744120055314, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.403272946295248, "standard_deviation": 0.007832335226612372, "minimum": 5.391569525796214, "maximum": 5.413639384625938, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5616591074216492, "standard_deviation": 0.014336641833636874, "minimum": 0.5493193408052117, "maximum": 0.593695042300414, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04923767004375461, "standard_deviation": 0.017552301040519414, "minimum": -0.08374483763541324, "maximum": -0.03417852231787778, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963746190060037, "standard_deviation": 0.0003761397749688258, "minimum": 0.9956166953134198, "maximum": 0.9967608780161662, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7533585696583827, "standard_deviation": 0.02424190262790984, "minimum": 0.7050919785333878, "maximum": 0.7780437736514727, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31399694707520626, "standard_deviation": 0.07355491016871163, "minimum": 0.20006891221278664, "maximum": 0.4230754347617606, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05232726185800667, "standard_deviation": 0.002698311063387894, "minimum": 0.049387443075870945, "maximum": 0.05748521746941324, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989319662289268, "standard_deviation": 0.000131301957021546, "minimum": 0.9987040733950115, "maximum": 0.9991635694232878, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40399337062690865, "standard_deviation": 0.022517801003066214, "minimum": 0.3807751579267247, "maximum": 0.4425365886217314, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33278388161004135, "standard_deviation": 0.09419431001831752, "minimum": 0.2253419971817984, "maximum": 0.5082193238052065, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0272709764611208, "standard_deviation": 0.0017364504895258873, "minimum": 0.023934922798394494, "maximum": 0.029887473126233837, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0039", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5486464500427246, "predicted_mos": 2.5486464500427246, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bd691bc52115049456bb36a38741d6e54f0e446447c0d72825a6ae4c3a26387e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6637951000593603, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7224856531084576, "全帯域_10ms_rms_envelope_nrmse": 0.05181397794940629, "全帯域_10ms_envelope_correlation": 0.996535508434027, "全帯域_10ms_modulation_mae_db": 0.3612658137953348, "全帯域_40ms_log_rms_mae_db": 0.3987507694388034, "全帯域_40ms_rms_envelope_nrmse": 0.028608994918051148, "全帯域_40ms_envelope_correlation": 0.998931119918026, "全帯域_40ms_modulation_mae_db": 0.29650328189922376, "300–1000 Hz_10ms_log_rms_mae_db": 1.8958357032102948, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10170125842998788, "300–1000 Hz_10ms_envelope_correlation": 0.9919113957881522, "300–1000 Hz_10ms_modulation_mae_db": 0.39028955727094833, "300–1000 Hz_40ms_log_rms_mae_db": 1.2245338827307213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05980392357143953, "300–1000 Hz_40ms_envelope_correlation": 0.9971528038245056, "300–1000 Hz_40ms_modulation_mae_db": 0.602347043213555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9457745195579608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15932167099562142, "1000–3000 Hz_10ms_envelope_correlation": 0.979353006702337, "1000–3000 Hz_10ms_modulation_mae_db": 1.111900634796144, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2026979406911613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09624279849116582, "1000–3000 Hz_40ms_envelope_correlation": 0.9942165993397252, "1000–3000 Hz_40ms_modulation_mae_db": 1.184346635457336, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3233069239763404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18380981263531782, "3000–6000 Hz_10ms_envelope_correlation": 0.9764673811581526, "3000–6000 Hz_10ms_modulation_mae_db": 0.8305847817136772, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4937162915251572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10799497808989957, "3000–6000 Hz_40ms_envelope_correlation": 0.9915015759886379, "3000–6000 Hz_40ms_modulation_mae_db": 0.7146143200249591, "mr_stft_db": 5.78594334949132, "band_mae_db": 0.19161880946258603, "rms_gain_error_db": -0.08993206299448643, "output_peak": 0.5692107334757386}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.749341100636896, "全帯域_10ms_rms_envelope_nrmse": 0.05181369357644022, "全帯域_10ms_envelope_correlation": 0.996453143678488, "全帯域_10ms_modulation_mae_db": 0.35269472185818646, "全帯域_40ms_log_rms_mae_db": 0.4000047438941004, "全帯域_40ms_rms_envelope_nrmse": 0.027240323968462262, "全帯域_40ms_envelope_correlation": 0.9989377861900942, "全帯域_40ms_modulation_mae_db": 0.3094018443691942, "300–1000 Hz_10ms_log_rms_mae_db": 1.9060259075529589, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10264040107889898, "300–1000 Hz_10ms_envelope_correlation": 0.991976337269853, "300–1000 Hz_10ms_modulation_mae_db": 0.7219723643188815, "300–1000 Hz_40ms_log_rms_mae_db": 1.231725012311118, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0603131835121788, "300–1000 Hz_40ms_envelope_correlation": 0.9973644231780565, "300–1000 Hz_40ms_modulation_mae_db": 0.9720509962354184, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0082455919518543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15611796104169162, "1000–3000 Hz_10ms_envelope_correlation": 0.9795734029843192, "1000–3000 Hz_10ms_modulation_mae_db": 0.9636066002427174, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2942421467501233, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09742793765706996, "1000–3000 Hz_40ms_envelope_correlation": 0.9929681301434283, "1000–3000 Hz_40ms_modulation_mae_db": 0.8693400705735401, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3939563798992967, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1956308666833078, "3000–6000 Hz_10ms_envelope_correlation": 0.9732472407627746, "3000–6000 Hz_10ms_modulation_mae_db": 0.7778469235688593, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5444555348054863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11972773975035643, "3000–6000 Hz_40ms_envelope_correlation": 0.9897104270377359, "3000–6000 Hz_40ms_modulation_mae_db": 0.42370873411813365, "mr_stft_db": 5.793468053310131, "band_mae_db": 0.21005784156698862, "rms_gain_error_db": -0.04757065184356903, "output_peak": 0.5709295162786523}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7293197853905677, "全帯域_10ms_rms_envelope_nrmse": 0.05318444838129202, "全帯域_10ms_envelope_correlation": 0.9963563938504404, "全帯域_10ms_modulation_mae_db": 0.40896461275168833, "全帯域_40ms_log_rms_mae_db": 0.4075884027113145, "全帯域_40ms_rms_envelope_nrmse": 0.03107346836521644, "全帯域_40ms_envelope_correlation": 0.9987237928923499, "全帯域_40ms_modulation_mae_db": 0.4142380145391746, "300–1000 Hz_10ms_log_rms_mae_db": 1.9441860427269102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10208800813162258, "300–1000 Hz_10ms_envelope_correlation": 0.9923011468039062, "300–1000 Hz_10ms_modulation_mae_db": 1.18107740681791, "300–1000 Hz_40ms_log_rms_mae_db": 1.2942805492466405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06481799897898867, "300–1000 Hz_40ms_envelope_correlation": 0.9971363743862756, "300–1000 Hz_40ms_modulation_mae_db": 1.41986628921985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9488702218997542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16116276086324047, "1000–3000 Hz_10ms_envelope_correlation": 0.979460283032369, "1000–3000 Hz_10ms_modulation_mae_db": 1.2971898178167527, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2541336785683863, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1069284472599788, "1000–3000 Hz_40ms_envelope_correlation": 0.9927662443492796, "1000–3000 Hz_40ms_modulation_mae_db": 1.332038421971384, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2859084355752985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17932696272648155, "3000–6000 Hz_10ms_envelope_correlation": 0.9776288852632155, "3000–6000 Hz_10ms_modulation_mae_db": 1.0356100095290985, "3000–6000 Hz_40ms_log_rms_mae_db": 1.466662529657803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10923008538185268, "3000–6000 Hz_40ms_envelope_correlation": 0.9914086723020048, "3000–6000 Hz_40ms_modulation_mae_db": 0.7400279408671379, "mr_stft_db": 5.771404785732183, "band_mae_db": 0.21687772985870643, "rms_gain_error_db": -0.0972287324451092, "output_peak": 0.5650727745598716}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7940152475839327, "全帯域_10ms_rms_envelope_nrmse": 0.05267291902856322, "全帯域_10ms_envelope_correlation": 0.9963687644427137, "全帯域_10ms_modulation_mae_db": 0.11719867526810308, "全帯域_40ms_log_rms_mae_db": 0.41420063820092595, "全帯域_40ms_rms_envelope_nrmse": 0.026558258486429773, "全帯域_40ms_envelope_correlation": 0.9990419337100668, "全帯域_40ms_modulation_mae_db": 0.05063251598600015, "300–1000 Hz_10ms_log_rms_mae_db": 1.880967610983594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10062391189044576, "300–1000 Hz_10ms_envelope_correlation": 0.99248980018747, "300–1000 Hz_10ms_modulation_mae_db": 0.7068485040434894, "300–1000 Hz_40ms_log_rms_mae_db": 1.2047437675912949, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0600738856299068, "300–1000 Hz_40ms_envelope_correlation": 0.9976008315883761, "300–1000 Hz_40ms_modulation_mae_db": 0.7337340591457595, "1000–3000 Hz_10ms_log_rms_mae_db": 2.037061318738114, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15745962950431988, "1000–3000 Hz_10ms_envelope_correlation": 0.9793613890966865, "1000–3000 Hz_10ms_modulation_mae_db": 0.7604860617893147, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2743973454463242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09153238149878547, "1000–3000 Hz_40ms_envelope_correlation": 0.9944704647965822, "1000–3000 Hz_40ms_modulation_mae_db": 0.6654468679499435, "3000–6000 Hz_10ms_log_rms_mae_db": 2.337687404229179, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18053378296983622, "3000–6000 Hz_10ms_envelope_correlation": 0.9772850099314061, "3000–6000 Hz_10ms_modulation_mae_db": 0.9906723369130405, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4752207826413197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10212810956830262, "3000–6000 Hz_40ms_envelope_correlation": 0.9926354460696709, "3000–6000 Hz_40ms_modulation_mae_db": 0.495015651208528, "mr_stft_db": 5.794556728701217, "band_mae_db": 0.21173077859825098, "rms_gain_error_db": -0.07314568838255936, "output_peak": 0.5597790282379208}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7660995752218035, "全帯域_10ms_rms_envelope_nrmse": 0.05335923410673675, "全帯域_10ms_envelope_correlation": 0.9962308520434564, "全帯域_10ms_modulation_mae_db": 0.29003031254202033, "全帯域_40ms_log_rms_mae_db": 0.397424571500044, "全帯域_40ms_rms_envelope_nrmse": 0.02826553729374918, "全帯域_40ms_envelope_correlation": 0.9988465540281609, "全帯域_40ms_modulation_mae_db": 0.19991093767060644, "300–1000 Hz_10ms_log_rms_mae_db": 1.8855496625022954, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10663495329006455, "300–1000 Hz_10ms_envelope_correlation": 0.9910078366537365, "300–1000 Hz_10ms_modulation_mae_db": 0.29040454859557846, "300–1000 Hz_40ms_log_rms_mae_db": 1.2233429486381917, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057853692856702464, "300–1000 Hz_40ms_envelope_correlation": 0.9972834593767046, "300–1000 Hz_40ms_modulation_mae_db": 0.32933092138038855, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9493804649464481, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1654867582817513, "1000–3000 Hz_10ms_envelope_correlation": 0.9760288104602479, "1000–3000 Hz_10ms_modulation_mae_db": 0.8120050618976036, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1862843018152687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09769992597797096, "1000–3000 Hz_40ms_envelope_correlation": 0.9920504657690693, "1000–3000 Hz_40ms_modulation_mae_db": 0.6249046935116259, "3000–6000 Hz_10ms_log_rms_mae_db": 2.304493451799311, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18471059777481216, "3000–6000 Hz_10ms_envelope_correlation": 0.9765450356548249, "3000–6000 Hz_10ms_modulation_mae_db": 0.689294015025657, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4782837971866003, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11189392752214711, "3000–6000 Hz_40ms_envelope_correlation": 0.9910087057133815, "3000–6000 Hz_40ms_modulation_mae_db": 0.44454181635679224, "mr_stft_db": 5.778864102099305, "band_mae_db": 0.12886897199795003, "rms_gain_error_db": -0.049089133245156874, "output_peak": 0.5770233480016368}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7248951022202933, "全帯域_10ms_rms_envelope_nrmse": 0.05286363577129264, "全帯域_10ms_envelope_correlation": 0.9963190592973092, "全帯域_10ms_modulation_mae_db": 0.33355138843106397, "全帯域_40ms_log_rms_mae_db": 0.38554965133602404, "全帯域_40ms_rms_envelope_nrmse": 0.029226269876133148, "全帯域_40ms_envelope_correlation": 0.9987883289246102, "全帯域_40ms_modulation_mae_db": 0.20292205713344327, "300–1000 Hz_10ms_log_rms_mae_db": 1.8649966360895884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09986671763109777, "300–1000 Hz_10ms_envelope_correlation": 0.9922172514138928, "300–1000 Hz_10ms_modulation_mae_db": 0.9389356372904446, "300–1000 Hz_40ms_log_rms_mae_db": 1.1644085242960192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0632821390244357, "300–1000 Hz_40ms_envelope_correlation": 0.9967793194430334, "300–1000 Hz_40ms_modulation_mae_db": 0.8124774328726164, "1000–3000 Hz_10ms_log_rms_mae_db": 2.010388898578992, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17131809116874566, "1000–3000 Hz_10ms_envelope_correlation": 0.9746043638314378, "1000–3000 Hz_10ms_modulation_mae_db": 1.0925652642800914, "1000–3000 Hz_40ms_log_rms_mae_db": 1.225488731036094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10181226061229337, "1000–3000 Hz_40ms_envelope_correlation": 0.9918362461226375, "1000–3000 Hz_40ms_modulation_mae_db": 0.9508526867073045, "3000–6000 Hz_10ms_log_rms_mae_db": 2.323705911297352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18393576142876733, "3000–6000 Hz_10ms_envelope_correlation": 0.9764529283380201, "3000–6000 Hz_10ms_modulation_mae_db": 0.6564065004033696, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4815331939985505, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10957217160743672, "3000–6000 Hz_40ms_envelope_correlation": 0.9915193973900923, "3000–6000 Hz_40ms_modulation_mae_db": 0.8534397422090967, "mr_stft_db": 5.791913346424295, "band_mae_db": 0.16311436744323327, "rms_gain_error_db": -0.06252660135218523, "output_peak": 0.5661493767899528}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7344296000423642, "全帯域_10ms_rms_envelope_nrmse": 0.05569428408571877, "全帯域_10ms_envelope_correlation": 0.9958773783032843, "全帯域_10ms_modulation_mae_db": 0.0980308081291744, "全帯域_40ms_log_rms_mae_db": 0.4079069320724912, "全帯域_40ms_rms_envelope_nrmse": 0.030982988729730876, "全帯域_40ms_envelope_correlation": 0.998588046259771, "全帯域_40ms_modulation_mae_db": 0.19354971195319573, "300–1000 Hz_10ms_log_rms_mae_db": 1.9478348806825239, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10148852680698621, "300–1000 Hz_10ms_envelope_correlation": 0.991956842623841, "300–1000 Hz_10ms_modulation_mae_db": 0.4694498298462375, "300–1000 Hz_40ms_log_rms_mae_db": 1.2358478039202292, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059487954624715914, "300–1000 Hz_40ms_envelope_correlation": 0.997211249694461, "300–1000 Hz_40ms_modulation_mae_db": 0.44108434614422487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9963403395404786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15831673648881164, "1000–3000 Hz_10ms_envelope_correlation": 0.9780949313375417, "1000–3000 Hz_10ms_modulation_mae_db": 0.5472544352537404, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2385466352258068, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08971967454792919, "1000–3000 Hz_40ms_envelope_correlation": 0.9934217007280552, "1000–3000 Hz_40ms_modulation_mae_db": 0.48451933107254375, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3388270416343664, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17878779072941856, "3000–6000 Hz_10ms_envelope_correlation": 0.9778869760704524, "3000–6000 Hz_10ms_modulation_mae_db": 0.6671356835878214, "3000–6000 Hz_40ms_log_rms_mae_db": 1.475951915300468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10591430992947189, "3000–6000 Hz_40ms_envelope_correlation": 0.9918817514134839, "3000–6000 Hz_40ms_modulation_mae_db": 0.25267095975915704, "mr_stft_db": 5.788400100028388, "band_mae_db": 0.15404574590456047, "rms_gain_error_db": -0.040527225508074416, "output_peak": 0.5644910099676134}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7390124979827086, "全帯域_10ms_rms_envelope_nrmse": 0.052221670606692784, "全帯域_10ms_envelope_correlation": 0.9964140405660306, "全帯域_10ms_modulation_mae_db": 0.5182488798221305, "全帯域_40ms_log_rms_mae_db": 0.3896770799961489, "全帯域_40ms_rms_envelope_nrmse": 0.028000251900962655, "全帯域_40ms_envelope_correlation": 0.9988958922734328, "全帯域_40ms_modulation_mae_db": 0.5551426324674003, "300–1000 Hz_10ms_log_rms_mae_db": 1.8572532052899138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09407791277397311, "300–1000 Hz_10ms_envelope_correlation": 0.9932011164772467, "300–1000 Hz_10ms_modulation_mae_db": 0.4339403493631697, "300–1000 Hz_40ms_log_rms_mae_db": 1.1360405649282062, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056270624110743205, "300–1000 Hz_40ms_envelope_correlation": 0.9975762635674106, "300–1000 Hz_40ms_modulation_mae_db": 0.3419714677572221, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9750072385889397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1570696055800987, "1000–3000 Hz_10ms_envelope_correlation": 0.9779110419607151, "1000–3000 Hz_10ms_modulation_mae_db": 0.9814538375126326, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2160171944902647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08608463453919467, "1000–3000 Hz_40ms_envelope_correlation": 0.9932338884320903, "1000–3000 Hz_40ms_modulation_mae_db": 1.0780979142782041, "3000–6000 Hz_10ms_log_rms_mae_db": 2.355937714888772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18463880635775948, "3000–6000 Hz_10ms_envelope_correlation": 0.976504362144609, "3000–6000 Hz_10ms_modulation_mae_db": 0.6801025830156952, "3000–6000 Hz_40ms_log_rms_mae_db": 1.514698934614646, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11237494974171501, "3000–6000 Hz_40ms_envelope_correlation": 0.9908238471702844, "3000–6000 Hz_40ms_modulation_mae_db": 0.6386087364741684, "mr_stft_db": 5.787368841151683, "band_mae_db": 0.11806936133475475, "rms_gain_error_db": -0.06330335208211704, "output_peak": 0.5694932505608146}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.05706471697979, "spectral_convergence": 0.17020202144088212, "spectral_convergence_status": "ok", "amplitude_floor": 2.2281588625700377e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.673765379738821, "spectral_convergence": 0.16816741878110375, "spectral_convergence_status": "ok", "amplitude_floor": 1.3568197926870596e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.626999951755349, "spectral_convergence": 0.16294940408674294, "spectral_convergence_status": "ok", "amplitude_floor": 1.1580179000740219e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0038028140043242973, "output_power": 0.003773454588816393, "delta_db": -0.03365956688667019, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002493613270814922, "output_power": 0.002426872847198642, "delta_db": -0.11782077914685524, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006858506604602334, "output_power": 0.0006698535511942198, "delta_db": -0.10249697086385601, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002628229680145485, "output_power": 0.0002309198129125749, "delta_db": -0.5620211843261168, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001273968713381539, "output_power": 0.00011879954804630507, "delta_db": -0.303439741126337, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.457810201861998e-05, "output_power": 5.4959894479492875e-05, "delta_db": 0.030274614425680857, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19161880946258603, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08993206299448643, "pitch": null, "output_rms_dbfs": -21.381753454939382, "output_peak": 0.5692107334757386, "reference_rms_dbfs": -21.2918213919449, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.78594334949132}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025738076998093343, "log_rms_mae_db": 0.7224856531084576, "rms_envelope_nrmse": 0.05181397794940629, "envelope_correlation": 0.996535508434027, "rms_gain_error_db": -0.08976265584988177, "reference_mean_power": 0.007432639167823318, "output_mean_power": 0.007280593507316901}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.369590439777949e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014473681837183883, "output_power": 0.0014221794968402662, "reference_active": true, "delta_db": -0.07624608414771412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002579988221497507, "output_power": 0.00024775464334717316, "reference_active": true, "delta_db": -0.17595920611865074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2677922934851034e-05, "output_power": 2.6983310391287456e-05, "reference_active": true, "delta_db": -0.8315921511196394}], "active_band_count": 3, "active_band_mae_db": 0.3612658137953348}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022253035095363898, "log_rms_mae_db": 0.3987507694388034, "rms_envelope_nrmse": 0.028608994918051148, "envelope_correlation": 0.998931119918026, "rms_gain_error_db": -0.08971635621546874, "reference_mean_power": 0.0074694638279480755, "output_mean_power": 0.007316742866926416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0676309697735546e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013241614802002638, "output_power": 0.0012967526776541804, "reference_active": true, "delta_db": -0.09083796683770706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011975787910962369, "output_power": 0.00011786758821151288, "reference_active": true, "delta_db": -0.06909698970987578}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4293821094800366e-06, "output_power": 2.053706321961308e-06, "reference_active": true, "delta_db": -0.7295748891500885}], "active_band_count": 3, "active_band_mae_db": 0.29650328189922376}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017894181312046162, "log_rms_mae_db": 1.8958357032102948, "rms_envelope_nrmse": 0.10170125842998788, "envelope_correlation": 0.9919113957881522, "rms_gain_error_db": -0.1651616779614163, "reference_mean_power": 0.0017811195275832714, "output_mean_power": 0.0017146555807850106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1758328243827778e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007391946250824912, "output_power": 0.0007015916638220217, "reference_active": true, "delta_db": -0.22674380146571316}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015375544244981539, "output_power": 0.000133697346973524, "reference_active": true, "delta_db": -0.6070770793746948}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0532547306567537e-05, "output_power": 9.746050094693797e-06, "reference_active": true, "delta_db": -0.337047790972437}], "active_band_count": 3, "active_band_mae_db": 0.39028955727094833}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016320284141686103, "log_rms_mae_db": 1.2245338827307213, "rms_envelope_nrmse": 0.05980392357143953, "envelope_correlation": 0.9971528038245056, "rms_gain_error_db": -0.16513096740961442, "reference_mean_power": 0.0017900337371132, "output_mean_power": 0.001723249334904543}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0791360125207298e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007228891058818755, "output_power": 0.0006804216265590795, "reference_active": true, "delta_db": -0.2629357107068941}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980895739386466e-05, "output_power": 6.204028100800327e-05, "reference_active": true, "delta_db": -0.5123739554980113}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.398569514368023e-07, "output_power": 7.41120133467215e-07, "reference_active": true, "delta_db": -1.0317314634357597}], "active_band_count": 3, "active_band_mae_db": 0.602347043213555}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.442874453255196e-05, "log_rms_mae_db": 1.9457745195579608, "rms_envelope_nrmse": 0.15932167099562142, "envelope_correlation": 0.979353006702337, "rms_gain_error_db": -0.5400529692675885, "reference_mean_power": 0.000318038298123639, "output_mean_power": 0.00028084980319580615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6089659140343417e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.507437744605689e-05, "output_power": 7.725298122860618e-05, "reference_active": true, "delta_db": -0.418835316693057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3978123456671057e-05, "output_power": 1.8782852130663887e-05, "reference_active": true, "delta_db": -1.0605365242376599}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6559576292684446e-06, "output_power": 3.0365190711314315e-06, "reference_active": true, "delta_db": -1.8563300634577153}], "active_band_count": 3, "active_band_mae_db": 1.111900634796144}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.707405683354541e-05, "log_rms_mae_db": 1.2026979406911613, "rms_envelope_nrmse": 0.09624279849116582, "envelope_correlation": 0.9942165993397252, "rms_gain_error_db": -0.5395492754292722, "reference_mean_power": 0.00031952379334241086, "output_mean_power": 0.00028219432510746394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4169476048907225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.318339905927379e-05, "output_power": 7.508831261868301e-05, "reference_active": true, "delta_db": -0.4446431764226616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1060276940136043e-05, "output_power": 8.974781719848928e-06, "reference_active": true, "delta_db": -0.9074210679177442}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55972585077424e-07, "output_power": 1.5420408837897622e-07, "reference_active": true, "delta_db": -2.2009756620316017}], "active_band_count": 3, "active_band_mae_db": 1.184346635457336}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.0412968032519945e-05, "log_rms_mae_db": 2.3233069239763404, "rms_envelope_nrmse": 0.18380981263531782, "envelope_correlation": 0.9764673811581526, "rms_gain_error_db": -0.05902553379092175, "reference_mean_power": 8.301357548516487e-05, "output_mean_power": 8.189295954013945e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.362198065588504e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2275109921127774e-05, "output_power": 2.956875037499051e-05, "reference_active": true, "delta_db": -0.38034759141367425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8777051155174143e-06, "output_power": 1.9059417467355383e-06, "reference_active": true, "delta_db": -1.7893666609885504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.623133887825791e-07, "output_power": 7.132923130050795e-07, "reference_active": true, "delta_db": 0.322040092738807}], "active_band_count": 3, "active_band_mae_db": 0.8305847817136772}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.404038681520291e-05, "log_rms_mae_db": 1.4937162915251572, "rms_envelope_nrmse": 0.10799497808989957, "envelope_correlation": 0.9915015759886379, "rms_gain_error_db": -0.05759429969499758, "reference_mean_power": 8.336414632310516e-05, "output_mean_power": 8.22659045331639e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1059391757915914e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.094184470310106e-05, "output_power": 2.8342890652766424e-05, "reference_active": true, "delta_db": -0.3810206080265354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2229001330453808e-06, "output_power": 8.797818625409657e-07, "reference_active": true, "delta_db": -1.4301598786701597}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7483618813517085e-08, "output_power": 4.3982262131207356e-08, "reference_active": true, "delta_db": -0.3326624733781819}], "active_band_count": 3, "active_band_mae_db": 0.7146143200249591}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9067730754237778, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "118a066bbe59ef6b28075b52997049576855130f4a3fba284502b4afa2c34905", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9780484036757068, "standard_deviation": 0.001840269632834755, "minimum": 0.9746043638314378, "maximum": 0.9795734029843192, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9838835742253178, "standard_deviation": 0.03428154941768385, "minimum": 1.9457745195579608, "maximum": 2.037061318738114, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9458077141986246, "standard_deviation": 0.234604057553993, "minimum": 0.5472544352537404, "maximum": 1.2971898178167527, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1607816517405351, "standard_deviation": 0.005183384276364659, "minimum": 0.15611796104169162, "maximum": 0.17131809116874566, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9931204674601084, "standard_deviation": 0.0009312366921267522, "minimum": 0.9918362461226375, "maximum": 0.9944704647965822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2364759967529286, "standard_deviation": 0.036430920840461734, "minimum": 1.1862843018152687, "maximum": 1.2942421467501233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8986933276902352, "standard_deviation": 0.2942443658307771, "minimum": 0.48451933107254375, "maximum": 1.332038421971384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09593100757304852, "standard_deviation": 0.006717562629487055, "minimum": 0.08608463453919467, "maximum": 0.1069284472599788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9765022274154318, "standard_deviation": 0.001434277074395427, "minimum": 0.9732472407627746, "maximum": 0.9778869760704524, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3329779079124893, "standard_deviation": 0.03275082511670906, "minimum": 2.2859084355752985, "maximum": 2.3939563798992967, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7909566042196523, "standard_deviation": 0.15002055367507378, "minimum": 0.6564065004033696, "maximum": 1.0356100095290985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18392179766321262, "standard_deviation": 0.005307683364563835, "minimum": 0.17878779072941856, "maximum": 0.1956308666833078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9913112278856615, "standard_deviation": 0.0008501878979697874, "minimum": 0.9897104270377359, "maximum": 0.9926354460696709, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4913153724662538, "standard_deviation": 0.026035084873560144, "minimum": 1.466662529657803, "maximum": 1.5444555348054863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5703284876272465, "standard_deviation": 0.19940212344592187, "minimum": 0.25267095975915704, "maximum": 0.8534397422090967, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10985453394889774, "standard_deviation": 0.0051705543533007384, "minimum": 0.10212810956830262, "maximum": 0.11972773975035643, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921327159022624, "standard_deviation": 0.0006180875804950023, "minimum": 0.9910078366537365, "maximum": 0.9932011164772467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.89783120612976, "standard_deviation": 0.03355203592939692, "minimum": 1.8572532052899138, "maximum": 1.9478348806825239, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6416147746933325, "standard_deviation": 0.30458750360073317, "minimum": 0.29040454859557846, "maximum": 1.18107740681791, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10114021125413461, "standard_deviation": 0.003495744085810665, "minimum": 0.09407791277397311, "maximum": 0.10663495329006455, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997263090632353, "standard_deviation": 0.00026397284618349725, "minimum": 0.9967793194430334, "maximum": 0.9976008315883761, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2143653817078026, "standard_deviation": 0.04789321506796637, "minimum": 1.1360405649282062, "maximum": 1.2942805492466405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7066078194961293, "standard_deviation": 0.36796981317518906, "minimum": 0.32933092138038855, "maximum": 1.41986628921985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06023792528863889, "standard_deviation": 0.0027373538186932756, "minimum": 0.056270624110743205, "maximum": 0.06481799897898867, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1742979507708788, "standard_deviation": 0.03882875746183641, "minimum": 0.11806936133475475, "maximum": 0.21687772985870643, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7864899133673156, "standard_deviation": 0.007870520996375373, "minimum": 5.771404785732183, "maximum": 5.794556728701217, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5677686297340252, "standard_deviation": 0.005143110964825427, "minimum": 0.5597790282379208, "maximum": 0.5770233480016368, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0654154309816572, "standard_deviation": 0.020319313123269973, "minimum": -0.0972287324451092, "maximum": -0.040527225508074416, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963193925769687, "standard_deviation": 0.00020020578460301942, "minimum": 0.9958773783032843, "maximum": 0.996535508434027, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.744949820273378, "standard_deviation": 0.024410381215227678, "minimum": 0.7224856531084576, "maximum": 0.7940152475839327, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3099981515747128, "standard_deviation": 0.14175903038532967, "minimum": 0.0980308081291744, "maximum": 0.5182488798221305, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.052952982938267834, "standard_deviation": 0.0012507082162243049, "minimum": 0.05181369357644022, "maximum": 0.05569428408571877, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988441817745639, "standard_deviation": 0.0001420884455027998, "minimum": 0.998588046259771, "maximum": 0.9990419337100668, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4001378486437316, "standard_deviation": 0.009600433198614148, "minimum": 0.38554965133602404, "maximum": 0.41420063820092595, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2777876245022798, "standard_deviation": 0.15457746318785673, "minimum": 0.05063251598600015, "maximum": 0.5551426324674003, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028744511692341934, "standard_deviation": 0.001626165396411842, "minimum": 0.026558258486429773, "maximum": 0.03107346836521644, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0039", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.9787585735321045, "predicted_mos": 2.9787585735321045, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "52ff6ebc2b6e471116282daf4b6577bb6b39d257aa9d0938b92d9b7c9c3d3452", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5990643999539316, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8045313083586828, "全帯域_10ms_rms_envelope_nrmse": 0.05041876770558046, "全帯域_10ms_envelope_correlation": 0.9966744858343295, "全帯域_10ms_modulation_mae_db": 0.2867001962708296, "全帯域_40ms_log_rms_mae_db": 0.44965640957326697, "全帯域_40ms_rms_envelope_nrmse": 0.026398731742519138, "全帯域_40ms_envelope_correlation": 0.9990439257956258, "全帯域_40ms_modulation_mae_db": 0.42774402188760124, "300–1000 Hz_10ms_log_rms_mae_db": 1.6138477535945044, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10291247893712548, "300–1000 Hz_10ms_envelope_correlation": 0.992056284172034, "300–1000 Hz_10ms_modulation_mae_db": 0.7540508867706871, "300–1000 Hz_40ms_log_rms_mae_db": 1.1090903229316897, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06339218335670038, "300–1000 Hz_40ms_envelope_correlation": 0.9972046718499913, "300–1000 Hz_40ms_modulation_mae_db": 0.4761826351894247, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5707054119170585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14625656944494175, "1000–3000 Hz_10ms_envelope_correlation": 0.9826514448039491, "1000–3000 Hz_10ms_modulation_mae_db": 0.8764382926856983, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0449231746900194, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09400701456372985, "1000–3000 Hz_40ms_envelope_correlation": 0.9940277599716252, "1000–3000 Hz_40ms_modulation_mae_db": 0.6604971329800405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.638499033976351, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1694851227097919, "3000–6000 Hz_10ms_envelope_correlation": 0.980409611118081, "3000–6000 Hz_10ms_modulation_mae_db": 0.4219286508204189, "3000–6000 Hz_40ms_log_rms_mae_db": 1.142984099897074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1105684678049388, "3000–6000 Hz_40ms_envelope_correlation": 0.9918906020900301, "3000–6000 Hz_40ms_modulation_mae_db": 0.7198921797014425, "mr_stft_db": 5.351101791368006, "band_mae_db": 0.2639198555256251, "rms_gain_error_db": -0.069976253948938, "output_peak": 0.5676343385571145}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7839247335155034, "全帯域_10ms_rms_envelope_nrmse": 0.052055455262163645, "全帯域_10ms_envelope_correlation": 0.996438981928974, "全帯域_10ms_modulation_mae_db": 0.2656207098511288, "全帯域_40ms_log_rms_mae_db": 0.44054825683661564, "全帯域_40ms_rms_envelope_nrmse": 0.02741384923973511, "全帯域_40ms_envelope_correlation": 0.9989318432535799, "全帯域_40ms_modulation_mae_db": 0.24049526335338212, "300–1000 Hz_10ms_log_rms_mae_db": 1.5559638571083514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09074469606743368, "300–1000 Hz_10ms_envelope_correlation": 0.9937162719095157, "300–1000 Hz_10ms_modulation_mae_db": 0.4768923523975936, "300–1000 Hz_40ms_log_rms_mae_db": 0.9990408426470649, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05338523353492018, "300–1000 Hz_40ms_envelope_correlation": 0.9979134608678359, "300–1000 Hz_40ms_modulation_mae_db": 0.5680707837335178, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5311746676164804, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1400604391397207, "1000–3000 Hz_10ms_envelope_correlation": 0.9830085702062442, "1000–3000 Hz_10ms_modulation_mae_db": 0.2264327230493767, "1000–3000 Hz_40ms_log_rms_mae_db": 1.056348911766758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08561311091000066, "1000–3000 Hz_40ms_envelope_correlation": 0.9938674686838417, "1000–3000 Hz_40ms_modulation_mae_db": 0.3636462555343876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.645928454990444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15936659235171227, "3000–6000 Hz_10ms_envelope_correlation": 0.9827538474756853, "3000–6000 Hz_10ms_modulation_mae_db": 0.4970532869456257, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1159093548932117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09950967686461637, "3000–6000 Hz_40ms_envelope_correlation": 0.9929769091053872, "3000–6000 Hz_40ms_modulation_mae_db": 0.2942909570363074, "mr_stft_db": 5.344146908484337, "band_mae_db": 0.14417273034881659, "rms_gain_error_db": -0.03457105072902139, "output_peak": 0.5608734310126117}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7611355105146818, "全帯域_10ms_rms_envelope_nrmse": 0.05165832417587837, "全帯域_10ms_envelope_correlation": 0.9964864972764859, "全帯域_10ms_modulation_mae_db": 0.1461671345403205, "全帯域_40ms_log_rms_mae_db": 0.4387235095128015, "全帯域_40ms_rms_envelope_nrmse": 0.030260763324186885, "全帯域_40ms_envelope_correlation": 0.9986865827579917, "全帯域_40ms_modulation_mae_db": 0.1802774983235489, "300–1000 Hz_10ms_log_rms_mae_db": 1.6281570851790226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09650365614589655, "300–1000 Hz_10ms_envelope_correlation": 0.9926727351524597, "300–1000 Hz_10ms_modulation_mae_db": 0.21208572505475495, "300–1000 Hz_40ms_log_rms_mae_db": 1.0949311080555084, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060250764938706146, "300–1000 Hz_40ms_envelope_correlation": 0.9970208168672752, "300–1000 Hz_40ms_modulation_mae_db": 0.4153318466548453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5511272870315362, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14522771672885101, "1000–3000 Hz_10ms_envelope_correlation": 0.9824072489889875, "1000–3000 Hz_10ms_modulation_mae_db": 0.6840234036810519, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0362569995415019, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08636815094983892, "1000–3000 Hz_40ms_envelope_correlation": 0.9947797471749423, "1000–3000 Hz_40ms_modulation_mae_db": 0.688082803977959, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6394322483552914, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16894180489182828, "3000–6000 Hz_10ms_envelope_correlation": 0.9818310153310482, "3000–6000 Hz_10ms_modulation_mae_db": 0.7466375242513535, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1794574729960168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11064673320059089, "3000–6000 Hz_40ms_envelope_correlation": 0.9923129978740575, "3000–6000 Hz_40ms_modulation_mae_db": 0.6604644106159844, "mr_stft_db": 5.343533879040052, "band_mae_db": 0.18020866547667333, "rms_gain_error_db": -0.05589355865518944, "output_peak": 0.5637456620297011}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8151103987806633, "全帯域_10ms_rms_envelope_nrmse": 0.05800168082922513, "全帯域_10ms_envelope_correlation": 0.9956003403678643, "全帯域_10ms_modulation_mae_db": 0.254608935969381, "全帯域_40ms_log_rms_mae_db": 0.4412074569239908, "全帯域_40ms_rms_envelope_nrmse": 0.03453642281950256, "全帯域_40ms_envelope_correlation": 0.9983422555869529, "全帯域_40ms_modulation_mae_db": 0.2651245860659361, "300–1000 Hz_10ms_log_rms_mae_db": 1.60863557510674, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1125710314732435, "300–1000 Hz_10ms_envelope_correlation": 0.9903183325857627, "300–1000 Hz_10ms_modulation_mae_db": 0.4129843657797003, "300–1000 Hz_40ms_log_rms_mae_db": 1.0622934663311563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07201927978072944, "300–1000 Hz_40ms_envelope_correlation": 0.9961396753809898, "300–1000 Hz_40ms_modulation_mae_db": 0.3887861087805484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5771890098823513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1551130729211475, "1000–3000 Hz_10ms_envelope_correlation": 0.9804018251379012, "1000–3000 Hz_10ms_modulation_mae_db": 0.46150569616418474, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0248121041599447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10192097971173657, "1000–3000 Hz_40ms_envelope_correlation": 0.9927734152901447, "1000–3000 Hz_40ms_modulation_mae_db": 0.47284524133375677, "3000–6000 Hz_10ms_log_rms_mae_db": 1.534129091353093, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17881554411901368, "3000–6000 Hz_10ms_envelope_correlation": 0.9781784659658131, "3000–6000 Hz_10ms_modulation_mae_db": 0.5252077391421496, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0745484182035019, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1125234972100129, "3000–6000 Hz_40ms_envelope_correlation": 0.9909118050825777, "3000–6000 Hz_40ms_modulation_mae_db": 0.6099845546813842, "mr_stft_db": 5.344475461165349, "band_mae_db": 0.2038180652704196, "rms_gain_error_db": -0.08238613596660552, "output_peak": 0.5401834609456607}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7541308535274445, "全帯域_10ms_rms_envelope_nrmse": 0.051340429556819896, "全帯域_10ms_envelope_correlation": 0.996508536755228, "全帯域_10ms_modulation_mae_db": 0.11843779018481854, "全帯域_40ms_log_rms_mae_db": 0.4098117327155084, "全帯域_40ms_rms_envelope_nrmse": 0.028335008148611382, "全帯域_40ms_envelope_correlation": 0.9988325080535067, "全帯域_40ms_modulation_mae_db": 0.08409793774503443, "300–1000 Hz_10ms_log_rms_mae_db": 1.5984905570851615, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09403101613333578, "300–1000 Hz_10ms_envelope_correlation": 0.9930133130475034, "300–1000 Hz_10ms_modulation_mae_db": 0.3139260227422879, "300–1000 Hz_40ms_log_rms_mae_db": 1.0552340630895072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05460709081334471, "300–1000 Hz_40ms_envelope_correlation": 0.9975111443805202, "300–1000 Hz_40ms_modulation_mae_db": 0.43174957137964204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5582026938619937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.143275998921183, "1000–3000 Hz_10ms_envelope_correlation": 0.9829745471173901, "1000–3000 Hz_10ms_modulation_mae_db": 0.9068839824791426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0056281390350477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08814296636571256, "1000–3000 Hz_40ms_envelope_correlation": 0.9944925469115787, "1000–3000 Hz_40ms_modulation_mae_db": 1.059848568431442, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6391294581240057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16097939313493528, "3000–6000 Hz_10ms_envelope_correlation": 0.9828454263874785, "3000–6000 Hz_10ms_modulation_mae_db": 0.5075796627736221, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1234631039776375, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10925621700804747, "3000–6000 Hz_40ms_envelope_correlation": 0.9919061011247824, "3000–6000 Hz_40ms_modulation_mae_db": 0.3196390205545719, "mr_stft_db": 5.341465522906749, "band_mae_db": 0.22375624771192762, "rms_gain_error_db": -0.04512806692740176, "output_peak": 0.5582516654158797}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7607850872021654, "全帯域_10ms_rms_envelope_nrmse": 0.04837273151161397, "全帯域_10ms_envelope_correlation": 0.9969109159878863, "全帯域_10ms_modulation_mae_db": 0.21377000132154703, "全帯域_40ms_log_rms_mae_db": 0.4163774442743693, "全帯域_40ms_rms_envelope_nrmse": 0.02526657146046445, "全帯域_40ms_envelope_correlation": 0.9990872564116456, "全帯域_40ms_modulation_mae_db": 0.14955992666266818, "300–1000 Hz_10ms_log_rms_mae_db": 1.6242333868247638, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.098384091618296, "300–1000 Hz_10ms_envelope_correlation": 0.9927275024023485, "300–1000 Hz_10ms_modulation_mae_db": 0.7432873109062422, "300–1000 Hz_40ms_log_rms_mae_db": 1.0936937720256936, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05972342563748485, "300–1000 Hz_40ms_envelope_correlation": 0.99750643906574, "300–1000 Hz_40ms_modulation_mae_db": 0.6684613166452684, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5921659154870134, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13894635197930305, "1000–3000 Hz_10ms_envelope_correlation": 0.9842677733948626, "1000–3000 Hz_10ms_modulation_mae_db": 0.58085458549091, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0156655836821693, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08515537827223897, "1000–3000 Hz_40ms_envelope_correlation": 0.9952248926850895, "1000–3000 Hz_40ms_modulation_mae_db": 0.43259432167015993, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5897753331074342, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.160771570752415, "3000–6000 Hz_10ms_envelope_correlation": 0.9820150131866273, "3000–6000 Hz_10ms_modulation_mae_db": 0.25516230431934084, "3000–6000 Hz_40ms_log_rms_mae_db": 1.081845618335409, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09839815974741724, "3000–6000 Hz_40ms_envelope_correlation": 0.9929762010244518, "3000–6000 Hz_40ms_modulation_mae_db": 0.353979821494256, "mr_stft_db": 5.327236732093361, "band_mae_db": 0.23180833060959585, "rms_gain_error_db": -0.0499355996454041, "output_peak": 0.5605010932911716}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7370269605941099, "全帯域_10ms_rms_envelope_nrmse": 0.04919799743993986, "全帯域_10ms_envelope_correlation": 0.9968395113436552, "全帯域_10ms_modulation_mae_db": 0.45038954646655505, "全帯域_40ms_log_rms_mae_db": 0.3867162704961927, "全帯域_40ms_rms_envelope_nrmse": 0.024509500674771315, "全帯域_40ms_envelope_correlation": 0.9991897199475167, "全帯域_40ms_modulation_mae_db": 0.21482079943848098, "300–1000 Hz_10ms_log_rms_mae_db": 1.6103778223166756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0884571206066544, "300–1000 Hz_10ms_envelope_correlation": 0.994167064484789, "300–1000 Hz_10ms_modulation_mae_db": 1.000658031287735, "300–1000 Hz_40ms_log_rms_mae_db": 1.0220486882479958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052659975803855834, "300–1000 Hz_40ms_envelope_correlation": 0.9981247046461734, "300–1000 Hz_40ms_modulation_mae_db": 1.211104738256961, "1000–3000 Hz_10ms_log_rms_mae_db": 1.530713234115064, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14483579172475397, "1000–3000 Hz_10ms_envelope_correlation": 0.9828512145046902, "1000–3000 Hz_10ms_modulation_mae_db": 0.9395599017165216, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0019226326735917, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08913260202652176, "1000–3000 Hz_40ms_envelope_correlation": 0.9947171435085186, "1000–3000 Hz_40ms_modulation_mae_db": 0.8023180435865503, "3000–6000 Hz_10ms_log_rms_mae_db": 1.610554786032093, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.162613675129912, "3000–6000 Hz_10ms_envelope_correlation": 0.9816955706721863, "3000–6000 Hz_10ms_modulation_mae_db": 0.31986430076029615, "3000–6000 Hz_40ms_log_rms_mae_db": 1.086555966111024, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09637216521339526, "3000–6000 Hz_40ms_envelope_correlation": 0.9932341165575446, "3000–6000 Hz_40ms_modulation_mae_db": 0.7823495400392039, "mr_stft_db": 5.338784902311118, "band_mae_db": 0.21574614469929945, "rms_gain_error_db": -0.06812039371103631, "output_peak": 0.5590192853876806}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7829279659870011, "全帯域_10ms_rms_envelope_nrmse": 0.05424188033151294, "全帯域_10ms_envelope_correlation": 0.9961341987017042, "全帯域_10ms_modulation_mae_db": 0.39863603578047396, "全帯域_40ms_log_rms_mae_db": 0.4343800377681619, "全帯域_40ms_rms_envelope_nrmse": 0.02994994827242576, "全帯域_40ms_envelope_correlation": 0.9987459466382893, "全帯域_40ms_modulation_mae_db": 0.29642446618828694, "300–1000 Hz_10ms_log_rms_mae_db": 1.7474354594931263, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10390370992854704, "300–1000 Hz_10ms_envelope_correlation": 0.9914546945787373, "300–1000 Hz_10ms_modulation_mae_db": 0.2248030223006207, "300–1000 Hz_40ms_log_rms_mae_db": 1.107483251966483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06287189023827842, "300–1000 Hz_40ms_envelope_correlation": 0.99671829985136, "300–1000 Hz_40ms_modulation_mae_db": 0.18708160938463858, "1000–3000 Hz_10ms_log_rms_mae_db": 1.623909047679912, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14540213222841436, "1000–3000 Hz_10ms_envelope_correlation": 0.982820915796392, "1000–3000 Hz_10ms_modulation_mae_db": 0.9169650121169376, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0718023770988787, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09348083509805029, "1000–3000 Hz_40ms_envelope_correlation": 0.9940267848369146, "1000–3000 Hz_40ms_modulation_mae_db": 0.6227189224405475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5936717255218726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15349359544730623, "3000–6000 Hz_10ms_envelope_correlation": 0.9842811499386936, "3000–6000 Hz_10ms_modulation_mae_db": 0.4142246182567682, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1353681806445242, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09723596421669684, "3000–6000 Hz_40ms_envelope_correlation": 0.9935199408625229, "3000–6000 Hz_40ms_modulation_mae_db": 0.3797749314309186, "mr_stft_db": 5.354586306306005, "band_mae_db": 0.22676051066700573, "rms_gain_error_db": -0.06821009324562732, "output_peak": 0.5658096673599535}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.499806956994036, "spectral_convergence": 0.16597139725195093, "spectral_convergence_status": "ok", "amplitude_floor": 2.2281588625700377e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.153428372417587, "spectral_convergence": 0.16265129053583746, "spectral_convergence_status": "ok", "amplitude_floor": 1.3568197926870596e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.400070044692393, "spectral_convergence": 0.16001751041318027, "spectral_convergence_status": "ok", "amplitude_floor": 1.1580179000740219e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0038028140043242973, "output_power": 0.003834314205535765, "delta_db": 0.035826139796579864, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002493613270814922, "output_power": 0.002427618573588695, "delta_db": -0.11648648959648027, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006858506604602334, "output_power": 0.0006459058056212403, "delta_db": -0.2606037304241319, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002628229680145485, "output_power": 0.00023228147004686805, "delta_db": -0.5364874949831904, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001273968713381539, "output_power": 0.0001168044000027046, "delta_db": -0.3769955967088054, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.457810201861998e-05, "output_power": 5.144065153481214e-05, "delta_db": -0.2571196816445629, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2639198555256251, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.069976253948938, "pitch": null, "output_rms_dbfs": -21.361797645893837, "output_peak": 0.5676343385571145, "reference_rms_dbfs": -21.2918213919449, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.351101791368006}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025738076998093343, "log_rms_mae_db": 0.8045313083586828, "rms_envelope_nrmse": 0.05041876770558046, "envelope_correlation": 0.9966744858343295, "rms_gain_error_db": -0.06974560191437745, "reference_mean_power": 0.007432639167823318, "output_mean_power": 0.007314227921603427}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.369590439777949e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014473681837183883, "output_power": 0.0013963882742546157, "reference_active": true, "delta_db": -0.15572828361620422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002579988221497507, "output_power": 0.00022870949625176378, "reference_active": true, "delta_db": -0.5233352593794953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2677922934851034e-05, "output_power": 3.1343734479421335e-05, "reference_active": true, "delta_db": -0.1810370458167894}], "active_band_count": 3, "active_band_mae_db": 0.2867001962708296}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022253035095363898, "log_rms_mae_db": 0.44965640957326697, "rms_envelope_nrmse": 0.026398731742519138, "envelope_correlation": 0.9990439257956258, "rms_gain_error_db": -0.06973027125876757, "reference_mean_power": 0.0074694638279480755, "output_mean_power": 0.007350491866087513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0676309697735546e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013241614802002638, "output_power": 0.0012750213876655252, "reference_active": true, "delta_db": -0.16423480279205718}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011975787910962369, "output_power": 0.0001062627032112205, "reference_active": true, "delta_db": -0.5192323641805474}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4293821094800366e-06, "output_power": 2.116017909024065e-06, "reference_active": true, "delta_db": -0.5997648986901992}], "active_band_count": 3, "active_band_mae_db": 0.42774402188760124}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017894181312046162, "log_rms_mae_db": 1.6138477535945044, "rms_envelope_nrmse": 0.10291247893712548, "envelope_correlation": 0.992056284172034, "rms_gain_error_db": -0.2658544363821465, "reference_mean_power": 0.0017811195275832714, "output_mean_power": 0.0016753579910946718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1758328243827778e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007391946250824912, "output_power": 0.0006862716813691589, "reference_active": true, "delta_db": -0.3226272199841193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015375544244981539, "output_power": 0.00012035983718951344, "reference_active": true, "delta_db": -1.0634890575285303}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0532547306567537e-05, "output_power": 1.2886561677223536e-05, "reference_active": true, "delta_db": 0.8760363827994117}], "active_band_count": 3, "active_band_mae_db": 0.7540508867706871}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016320284141686103, "log_rms_mae_db": 1.1090903229316897, "rms_envelope_nrmse": 0.06339218335670038, "envelope_correlation": 0.9972046718499913, "rms_gain_error_db": -0.2658068608428995, "reference_mean_power": 0.0017900337371132, "output_mean_power": 0.0016837613265277792}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0791360125207298e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007228891058818755, "output_power": 0.0006672394229902004, "reference_active": true, "delta_db": -0.3478998174748381}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980895739386466e-05, "output_power": 5.553049629130142e-05, "reference_active": true, "delta_db": -0.9937959690952047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.398569514368023e-07, "output_power": 9.588418249675358e-07, "reference_active": true, "delta_db": 0.08685211899823117}], "active_band_count": 3, "active_band_mae_db": 0.4761826351894247}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.442874453255196e-05, "log_rms_mae_db": 1.5707054119170585, "rms_envelope_nrmse": 0.14625656944494175, "envelope_correlation": 0.9826514448039491, "rms_gain_error_db": -0.5012134543278741, "reference_mean_power": 0.000318038298123639, "output_mean_power": 0.0002833727438388866}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6089659140343417e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.507437744605689e-05, "output_power": 7.626036341979532e-05, "reference_active": true, "delta_db": -0.47499909137524055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3978123456671057e-05, "output_power": 1.94859937134168e-05, "reference_active": true, "delta_db": -0.9009263382640996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6559576292684446e-06, "output_power": 3.488751808261555e-06, "reference_active": true, "delta_db": -1.2533894484177548}], "active_band_count": 3, "active_band_mae_db": 0.8764382926856983}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.707405683354541e-05, "log_rms_mae_db": 1.0449231746900194, "rms_envelope_nrmse": 0.09400701456372985, "envelope_correlation": 0.9940277599716252, "rms_gain_error_db": -0.5010947858348559, "reference_mean_power": 0.00031952379334241086, "output_mean_power": 0.0002847041022468127}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4169476048907225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.318339905927379e-05, "output_power": 7.37387965739238e-05, "reference_active": true, "delta_db": -0.5234061704349731}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1060276940136043e-05, "output_power": 8.848724411022702e-06, "reference_active": true, "delta_db": -0.9688533204310548}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55972585077424e-07, "output_power": 2.287021585540108e-07, "reference_active": true, "delta_db": -0.4892319080740937}], "active_band_count": 3, "active_band_mae_db": 0.6604971329800405}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.0412968032519945e-05, "log_rms_mae_db": 1.638499033976351, "rms_envelope_nrmse": 0.1694851227097919, "envelope_correlation": 0.980409611118081, "rms_gain_error_db": -0.3495281731961603, "reference_mean_power": 8.301357548516487e-05, "output_mean_power": 7.659427372209992e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.362198065588504e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2275109921127774e-05, "output_power": 2.92677989344287e-05, "reference_active": true, "delta_db": -0.42477667210780223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8777051155174143e-06, "output_power": 2.5699260286280683e-06, "reference_active": true, "delta_db": -0.49125665801738133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.623133887825791e-07, "output_power": 6.110662141493807e-07, "reference_active": true, "delta_db": -0.3497526223360732}], "active_band_count": 3, "active_band_mae_db": 0.4219286508204189}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.404038681520291e-05, "log_rms_mae_db": 1.142984099897074, "rms_envelope_nrmse": 0.1105684678049388, "envelope_correlation": 0.9918906020900301, "rms_gain_error_db": -0.35362376199019724, "reference_mean_power": 8.336414632310516e-05, "output_mean_power": 7.684523286994659e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1059391757915914e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.094184470310106e-05, "output_power": 2.8215866657146557e-05, "reference_active": true, "delta_db": -0.40052807804269874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2229001330453808e-06, "output_power": 1.0260999400476922e-06, "reference_active": true, "delta_db": -0.762013300109313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7483618813517085e-08, "output_power": 3.7742467854188994e-08, "reference_active": true, "delta_db": -0.9971351609523159}], "active_band_count": 3, "active_band_mae_db": 0.7198921797014425}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9188221895512814, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ae8ae1c447e3004a1474b1651bbafbdfbf393059b81c7995d39341857af426a0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826729424938021, "standard_deviation": 0.0010710344235428037, "minimum": 0.9804018251379012, "maximum": 0.9842677733948626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5668984084489264, "standard_deviation": 0.0314391609173032, "minimum": 1.530713234115064, "maximum": 1.623909047679912, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6990829496729779, "standard_deviation": 0.26021525942524376, "minimum": 0.2264327230493767, "maximum": 0.9395599017165216, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14488975913603944, "standard_deviation": 0.004899192520343896, "minimum": 0.13894635197930305, "maximum": 0.1551130729211475, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994238719882832, "standard_deviation": 0.0007495646548673332, "minimum": 0.9927734152901447, "maximum": 0.9952248926850895, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0321699903309889, "standard_deviation": 0.02472881523793921, "minimum": 1.0019226326735917, "maximum": 1.0718023770988787, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6378189112443554, "standard_deviation": 0.2245445725625897, "minimum": 0.3636462555343876, "maximum": 1.059848568431442, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0904776297372287, "standard_deviation": 0.00571418699271246, "minimum": 0.08515537827223897, "maximum": 0.10192097971173657, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9817512625094517, "standard_deviation": 0.001822346670890366, "minimum": 0.9781784659658131, "maximum": 0.9842811499386936, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6113900164325732, "standard_deviation": 0.038237465712431785, "minimum": 1.534129091353093, "maximum": 1.645928454990444, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4609572609086969, "standard_deviation": 0.14889299519347016, "minimum": 0.25516230431934084, "maximum": 0.7466375242513535, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16430841231711432, "standard_deviation": 0.007806308163086253, "minimum": 0.15349359544730623, "maximum": 0.17881554411901368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9924660842151692, "standard_deviation": 0.0008706014774201681, "minimum": 0.9909118050825777, "maximum": 0.9935199408625229, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1175165268823, "standard_deviation": 0.03569540347958059, "minimum": 1.0745484182035019, "maximum": 1.1794574729960168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5150469269442586, "standard_deviation": 0.19810900160557873, "minimum": 0.2942909570363074, "maximum": 0.7823495400392039, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10431386015821448, "standard_deviation": 0.006992784521643779, "minimum": 0.09637216521339526, "maximum": 0.1125234972100129, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925157747916438, "standard_deviation": 0.0012335046197534056, "minimum": 0.9903183325857627, "maximum": 0.994167064484789, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6233926870885431, "standard_deviation": 0.0548439912800824, "minimum": 1.5559638571083514, "maximum": 1.7474354594931263, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5173359646549527, "standard_deviation": 0.2861922338186675, "minimum": 0.21208572505475495, "maximum": 1.000658031287735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09843847511381655, "standard_deviation": 0.007852283576164682, "minimum": 0.0884571206066544, "maximum": 0.1125710314732435, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972674016137357, "standard_deviation": 0.00064441875814786, "minimum": 0.9961396753809898, "maximum": 0.9981247046461734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0679769394118872, "standard_deviation": 0.040874746269954985, "minimum": 0.9990408426470649, "maximum": 1.1090903229316897, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5433460762531057, "standard_deviation": 0.30386375466948873, "minimum": 0.18708160938463858, "maximum": 1.211104738256961, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0598637305130025, "standard_deviation": 0.006446739258221671, "minimum": 0.052659975803855834, "maximum": 0.07201927978072944, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2112738187886704, "standard_deviation": 0.0361190685623416, "minimum": 0.14417273034881659, "maximum": 0.2639198555256251, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.343166437959372, "standard_deviation": 0.008211834661520137, "minimum": 5.327236732093361, "maximum": 5.354586306306005, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5595023254999716, "standard_deviation": 0.008466839316635923, "minimum": 0.5401834609456607, "maximum": 0.5676343385571145, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05927764410365298, "standard_deviation": 0.015656115357096954, "minimum": -0.08238613596660552, "maximum": -0.03457105072902139, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996449183524516, "standard_deviation": 0.00042106720582708006, "minimum": 0.9956003403678643, "maximum": 0.9969109159878863, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7749466023100315, "standard_deviation": 0.026444507334037356, "minimum": 0.7370269605941099, "maximum": 0.8151103987806633, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2667912937981318, "standard_deviation": 0.1140223700889356, "minimum": 0.11843779018481854, "maximum": 0.45038954646655505, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05191090835159179, "standard_deviation": 0.0030483284189612857, "minimum": 0.04837273151161397, "maximum": 0.05800168082922513, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988575048056385, "standard_deviation": 0.00027023742401122733, "minimum": 0.9983422555869529, "maximum": 0.9991897199475167, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4271776397626134, "standard_deviation": 0.02111677485219105, "minimum": 0.3867162704961927, "maximum": 0.44965640957326697, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.23231806245811737, "standard_deviation": 0.10370050293498649, "minimum": 0.08409793774503443, "maximum": 0.42774402188760124, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028333849460277075, "standard_deviation": 0.0032386418885832035, "minimum": 0.024509500674771315, "maximum": 0.03453642281950256, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0039", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.028024435043335, "predicted_mos": 3.028024435043335, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "74e85b5aad7f40e77ec01dbb4f96ec1eecba478f7312e8e2ebb96dbd8421b683", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.637054399587214, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7672573504940376, "全帯域_10ms_rms_envelope_nrmse": 0.057983279880233905, "全帯域_10ms_envelope_correlation": 0.9955385310760527, "全帯域_10ms_modulation_mae_db": 0.27577155930482194, "全帯域_40ms_log_rms_mae_db": 0.4249474842096034, "全帯域_40ms_rms_envelope_nrmse": 0.030890000123393803, "全帯域_40ms_envelope_correlation": 0.9986114226582088, "全帯域_40ms_modulation_mae_db": 0.461407547994981, "300–1000 Hz_10ms_log_rms_mae_db": 1.7270093788865875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09868646154278178, "300–1000 Hz_10ms_envelope_correlation": 0.9923204560081461, "300–1000 Hz_10ms_modulation_mae_db": 0.34863729051280723, "300–1000 Hz_40ms_log_rms_mae_db": 1.110075078702755, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06615100814137545, "300–1000 Hz_40ms_envelope_correlation": 0.9963344426524846, "300–1000 Hz_40ms_modulation_mae_db": 0.4005417175672334, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7414004844089155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15232450187846797, "1000–3000 Hz_10ms_envelope_correlation": 0.980942785329796, "1000–3000 Hz_10ms_modulation_mae_db": 0.7244736593494153, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1356341008072313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09665733966635663, "1000–3000 Hz_40ms_envelope_correlation": 0.9934811307719615, "1000–3000 Hz_40ms_modulation_mae_db": 0.5847303988968605, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8532488111850443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16769433521257127, "3000–6000 Hz_10ms_envelope_correlation": 0.9806170234065099, "3000–6000 Hz_10ms_modulation_mae_db": 0.5284482309897465, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2293411764072177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09759217131243969, "3000–6000 Hz_40ms_envelope_correlation": 0.9930842323369674, "3000–6000 Hz_40ms_modulation_mae_db": 0.3750339737156272, "mr_stft_db": 5.472835018002488, "band_mae_db": 0.20215748970609052, "rms_gain_error_db": -0.048576222172273406, "output_peak": 0.567088163070914}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7447046319417179, "全帯域_10ms_rms_envelope_nrmse": 0.04924004038211335, "全帯域_10ms_envelope_correlation": 0.9967735757606593, "全帯域_10ms_modulation_mae_db": 0.282253213049424, "全帯域_40ms_log_rms_mae_db": 0.39544251315582474, "全帯域_40ms_rms_envelope_nrmse": 0.023916910072933333, "全帯域_40ms_envelope_correlation": 0.9991552207318745, "全帯域_40ms_modulation_mae_db": 0.2507080314994821, "300–1000 Hz_10ms_log_rms_mae_db": 1.7471384617576027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09071254968994633, "300–1000 Hz_10ms_envelope_correlation": 0.9936249113302328, "300–1000 Hz_10ms_modulation_mae_db": 0.4382808749175286, "300–1000 Hz_40ms_log_rms_mae_db": 1.1169712477518334, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050616240692051354, "300–1000 Hz_40ms_envelope_correlation": 0.9980281665031279, "300–1000 Hz_40ms_modulation_mae_db": 0.650168086718169, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7635927892868144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15263142148763015, "1000–3000 Hz_10ms_envelope_correlation": 0.9805768889903101, "1000–3000 Hz_10ms_modulation_mae_db": 0.9605416896544949, "1000–3000 Hz_40ms_log_rms_mae_db": 1.083157293566346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09001601989694233, "1000–3000 Hz_40ms_envelope_correlation": 0.9944733365854005, "1000–3000 Hz_40ms_modulation_mae_db": 0.8735094816290111, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8040700110151937, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1571542619565664, "3000–6000 Hz_10ms_envelope_correlation": 0.9830446708256299, "3000–6000 Hz_10ms_modulation_mae_db": 0.5496892496840974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1567209898245152, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09896718244412055, "3000–6000 Hz_40ms_envelope_correlation": 0.9933770433382126, "3000–6000 Hz_40ms_modulation_mae_db": 0.6692545549557715, "mr_stft_db": 5.463430378579623, "band_mae_db": 0.24139635097286116, "rms_gain_error_db": -0.02722903075965216, "output_peak": 0.5523512392211989}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7524575210593193, "全帯域_10ms_rms_envelope_nrmse": 0.051839675401813085, "全帯域_10ms_envelope_correlation": 0.996469888349498, "全帯域_10ms_modulation_mae_db": 0.3406732963674542, "全帯域_40ms_log_rms_mae_db": 0.43529041432761795, "全帯域_40ms_rms_envelope_nrmse": 0.028998190359575998, "全帯域_40ms_envelope_correlation": 0.9988098312975806, "全帯域_40ms_modulation_mae_db": 0.3352709082306424, "300–1000 Hz_10ms_log_rms_mae_db": 1.7323346292101596, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0908044195231098, "300–1000 Hz_10ms_envelope_correlation": 0.993686636491901, "300–1000 Hz_10ms_modulation_mae_db": 0.9522573748445634, "300–1000 Hz_40ms_log_rms_mae_db": 1.0957868336929693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054149616057392436, "300–1000 Hz_40ms_envelope_correlation": 0.997775137714807, "300–1000 Hz_40ms_modulation_mae_db": 0.9148504409486818, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6904368893010073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14897552289676488, "1000–3000 Hz_10ms_envelope_correlation": 0.9817157885022194, "1000–3000 Hz_10ms_modulation_mae_db": 0.6646172984177081, "1000–3000 Hz_40ms_log_rms_mae_db": 1.093508290554839, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08839935652092337, "1000–3000 Hz_40ms_envelope_correlation": 0.9948845379429246, "1000–3000 Hz_40ms_modulation_mae_db": 0.8966596504746421, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7711663823062278, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17039381209997925, "3000–6000 Hz_10ms_envelope_correlation": 0.9797302524867441, "3000–6000 Hz_10ms_modulation_mae_db": 0.31564839564844255, "3000–6000 Hz_40ms_log_rms_mae_db": 1.17652770396616, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1076049091515891, "3000–6000 Hz_40ms_envelope_correlation": 0.9915604412069344, "3000–6000 Hz_40ms_modulation_mae_db": 0.10649729164333484, "mr_stft_db": 5.471743655903694, "band_mae_db": 0.21411910405549547, "rms_gain_error_db": -0.06206277314934816, "output_peak": 0.5534542457795204}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7518109768078829, "全帯域_10ms_rms_envelope_nrmse": 0.052013757694646626, "全帯域_10ms_envelope_correlation": 0.9964089639413617, "全帯域_10ms_modulation_mae_db": 0.4491483679225303, "全帯域_40ms_log_rms_mae_db": 0.3798836304497232, "全帯域_40ms_rms_envelope_nrmse": 0.02731713275353766, "全帯域_40ms_envelope_correlation": 0.9989102658356549, "全帯域_40ms_modulation_mae_db": 0.41417764505881416, "300–1000 Hz_10ms_log_rms_mae_db": 1.6951768856586535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10209634029652095, "300–1000 Hz_10ms_envelope_correlation": 0.9918870372311849, "300–1000 Hz_10ms_modulation_mae_db": 0.3300606164504587, "300–1000 Hz_40ms_log_rms_mae_db": 1.086812425825581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056341434671645124, "300–1000 Hz_40ms_envelope_correlation": 0.9975651714769205, "300–1000 Hz_40ms_modulation_mae_db": 0.31885280363667656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.712383758545148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15434297201398237, "1000–3000 Hz_10ms_envelope_correlation": 0.9813944234106119, "1000–3000 Hz_10ms_modulation_mae_db": 1.0383983742574134, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1399505568292734, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10221243910331484, "1000–3000 Hz_40ms_envelope_correlation": 0.9936490067441057, "1000–3000 Hz_40ms_modulation_mae_db": 0.9243176049705538, "3000–6000 Hz_10ms_log_rms_mae_db": 1.824464750847684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18300572864425735, "3000–6000 Hz_10ms_envelope_correlation": 0.9776217923806841, "3000–6000 Hz_10ms_modulation_mae_db": 0.22311738795463307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2336199403061754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12521693642698983, "3000–6000 Hz_40ms_envelope_correlation": 0.9891210326870582, "3000–6000 Hz_40ms_modulation_mae_db": 0.2432815674940404, "mr_stft_db": 5.47164446229205, "band_mae_db": 0.23726408497304188, "rms_gain_error_db": -0.04057301655457545, "output_peak": 0.5638952907692562}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7511421702614564, "全帯域_10ms_rms_envelope_nrmse": 0.05009480622564322, "全帯域_10ms_envelope_correlation": 0.9966773052976348, "全帯域_10ms_modulation_mae_db": 0.20268476670314117, "全帯域_40ms_log_rms_mae_db": 0.411019882821851, "全帯域_40ms_rms_envelope_nrmse": 0.025607561497124842, "全帯域_40ms_envelope_correlation": 0.9990472316452467, "全帯域_40ms_modulation_mae_db": 0.2585824833894817, "300–1000 Hz_10ms_log_rms_mae_db": 1.7804433323894022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0990693842439784, "300–1000 Hz_10ms_envelope_correlation": 0.9922742471324408, "300–1000 Hz_10ms_modulation_mae_db": 0.4141645091313164, "300–1000 Hz_40ms_log_rms_mae_db": 1.1672921567967127, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06012460065902077, "300–1000 Hz_40ms_envelope_correlation": 0.9970233054931149, "300–1000 Hz_40ms_modulation_mae_db": 0.5907265915180768, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7042183685270291, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15847117994684773, "1000–3000 Hz_10ms_envelope_correlation": 0.9791811776240016, "1000–3000 Hz_10ms_modulation_mae_db": 0.6358522320103618, "1000–3000 Hz_40ms_log_rms_mae_db": 1.120835857702475, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09929418797312137, "1000–3000 Hz_40ms_envelope_correlation": 0.9930510098118177, "1000–3000 Hz_40ms_modulation_mae_db": 0.6098459957273622, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8457556867688703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17207967331143809, "3000–6000 Hz_10ms_envelope_correlation": 0.9795211961473514, "3000–6000 Hz_10ms_modulation_mae_db": 0.6783483025082585, "3000–6000 Hz_40ms_log_rms_mae_db": 1.200637127276988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10244949119701084, "3000–6000 Hz_40ms_envelope_correlation": 0.9923236949305058, "3000–6000 Hz_40ms_modulation_mae_db": 0.6535562884688354, "mr_stft_db": 5.458875467166702, "band_mae_db": 0.18247353764273297, "rms_gain_error_db": -0.03500000881537929, "output_peak": 0.5589228838563607}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7175556127303006, "全帯域_10ms_rms_envelope_nrmse": 0.051419493037572, "全帯域_10ms_envelope_correlation": 0.9964804735850916, "全帯域_10ms_modulation_mae_db": 0.2422972886009507, "全帯域_40ms_log_rms_mae_db": 0.39276492754759695, "全帯域_40ms_rms_envelope_nrmse": 0.027294816360126383, "全帯域_40ms_envelope_correlation": 0.9988954012887632, "全帯域_40ms_modulation_mae_db": 0.25021567544545653, "300–1000 Hz_10ms_log_rms_mae_db": 1.6888401936893034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09791876595422944, "300–1000 Hz_10ms_envelope_correlation": 0.9923988548460688, "300–1000 Hz_10ms_modulation_mae_db": 0.4684155030754977, "300–1000 Hz_40ms_log_rms_mae_db": 1.0635353344641494, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05967406253466724, "300–1000 Hz_40ms_envelope_correlation": 0.9970050063793617, "300–1000 Hz_40ms_modulation_mae_db": 0.532144508256457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6941878783459294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15065124900537635, "1000–3000 Hz_10ms_envelope_correlation": 0.9811033072023669, "1000–3000 Hz_10ms_modulation_mae_db": 0.6266517299066656, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0978450203197248, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09447899887527911, "1000–3000 Hz_40ms_envelope_correlation": 0.993539572494961, "1000–3000 Hz_40ms_modulation_mae_db": 0.47005469661602267, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8051500154012325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16482403044309377, "3000–6000 Hz_10ms_envelope_correlation": 0.9811017437694672, "3000–6000 Hz_10ms_modulation_mae_db": 0.5361165883266868, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1542051550276269, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1002340156532103, "3000–6000 Hz_40ms_envelope_correlation": 0.9926267005189152, "3000–6000 Hz_40ms_modulation_mae_db": 0.33033336631782434, "mr_stft_db": 5.473575621602017, "band_mae_db": 0.17573926851560187, "rms_gain_error_db": -0.027801906618580327, "output_peak": 0.5537414563061426}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7539077015767579, "全帯域_10ms_rms_envelope_nrmse": 0.054475180758510065, "全帯域_10ms_envelope_correlation": 0.9961406716451626, "全帯域_10ms_modulation_mae_db": 0.5247581280287319, "全帯域_40ms_log_rms_mae_db": 0.3728604476965578, "全帯域_40ms_rms_envelope_nrmse": 0.027498079976323895, "全帯域_40ms_envelope_correlation": 0.9989996457048405, "全帯域_40ms_modulation_mae_db": 0.5019233122137442, "300–1000 Hz_10ms_log_rms_mae_db": 1.839288450335106, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10173951363217512, "300–1000 Hz_10ms_envelope_correlation": 0.9921578285874435, "300–1000 Hz_10ms_modulation_mae_db": 0.7093153929625621, "300–1000 Hz_40ms_log_rms_mae_db": 1.1934413134082744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06331101974814628, "300–1000 Hz_40ms_envelope_correlation": 0.9970928004725818, "300–1000 Hz_40ms_modulation_mae_db": 0.8444886999262922, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6945570159765337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14642776806579022, "1000–3000 Hz_10ms_envelope_correlation": 0.9825205095815371, "1000–3000 Hz_10ms_modulation_mae_db": 0.8031530431948765, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0262584927974103, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08887007803407536, "1000–3000 Hz_40ms_envelope_correlation": 0.9949081309179677, "1000–3000 Hz_40ms_modulation_mae_db": 0.9254130652878301, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8362964905335077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16852518872991334, "3000–6000 Hz_10ms_envelope_correlation": 0.9802664470071221, "3000–6000 Hz_10ms_modulation_mae_db": 0.48825606261558496, "3000–6000 Hz_40ms_log_rms_mae_db": 1.206063543392098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10066601252475338, "3000–6000 Hz_40ms_envelope_correlation": 0.9925692127994385, "3000–6000 Hz_40ms_modulation_mae_db": 0.6099981530253354, "mr_stft_db": 5.473160253097127, "band_mae_db": 0.19699682009153688, "rms_gain_error_db": -0.08137470695353116, "output_peak": 0.5980561039557237}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7485865181891701, "全帯域_10ms_rms_envelope_nrmse": 0.05024025049672359, "全帯域_10ms_envelope_correlation": 0.9966650248467694, "全帯域_10ms_modulation_mae_db": 0.3579781999067962, "全帯域_40ms_log_rms_mae_db": 0.40553460321052304, "全帯域_40ms_rms_envelope_nrmse": 0.027793777159971913, "全帯域_40ms_envelope_correlation": 0.9988883790890707, "全帯域_40ms_modulation_mae_db": 0.316224840496857, "300–1000 Hz_10ms_log_rms_mae_db": 1.7591143050387748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1006436245965294, "300–1000 Hz_10ms_envelope_correlation": 0.9922380759054585, "300–1000 Hz_10ms_modulation_mae_db": 0.5692961937911787, "300–1000 Hz_40ms_log_rms_mae_db": 1.1012563688086447, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06602246277706443, "300–1000 Hz_40ms_envelope_correlation": 0.9966333630872117, "300–1000 Hz_40ms_modulation_mae_db": 0.3709528451304547, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6726607085810727, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14951996197791148, "1000–3000 Hz_10ms_envelope_correlation": 0.9816628531722533, "1000–3000 Hz_10ms_modulation_mae_db": 1.1332811175939674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0420924609530287, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09331043446793144, "1000–3000 Hz_40ms_envelope_correlation": 0.9940578892173662, "1000–3000 Hz_40ms_modulation_mae_db": 1.3794360352803887, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8356677940544885, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.178976194568028, "3000–6000 Hz_10ms_envelope_correlation": 0.9781318621698402, "3000–6000 Hz_10ms_modulation_mae_db": 0.08259386026898467, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2218429038078615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10736626079199767, "3000–6000 Hz_40ms_envelope_correlation": 0.9917409989671275, "3000–6000 Hz_40ms_modulation_mae_db": 0.17685800596467774, "mr_stft_db": 5.462921139176692, "band_mae_db": 0.21062342961686217, "rms_gain_error_db": -0.05119496664023601, "output_peak": 0.5584631438627656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.687130466101864, "spectral_convergence": 0.16885602172786163, "spectral_convergence_status": "ok", "amplitude_floor": 2.2281588625700377e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.269515828495424, "spectral_convergence": 0.16198000983598135, "spectral_convergence_status": "ok", "amplitude_floor": 1.3568197926870596e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.461858759410177, "spectral_convergence": 0.16122821957355976, "spectral_convergence_status": "ok", "amplitude_floor": 1.1580179000740219e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0038028140043242973, "output_power": 0.0038223612627807553, "delta_db": 0.02226646544567571, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002493613270814922, "output_power": 0.0024449057037053794, "delta_db": -0.08566986851241676, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006858506604602334, "output_power": 0.000669963414278338, "delta_db": -0.10178474027514267, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002628229680145485, "output_power": 0.00023185531458213372, "delta_db": -0.5444626016717763, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001273968713381539, "output_power": 0.0001188101354961329, "delta_db": -0.30305271388299004, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.457810201861998e-05, "output_power": 5.657040443564097e-05, "delta_db": 0.15570854844854165, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20215748970609052, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.048576222172273406, "pitch": null, "output_rms_dbfs": -21.34039761411717, "output_peak": 0.567088163070914, "reference_rms_dbfs": -21.2918213919449, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.472835018002488}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025738076998093343, "log_rms_mae_db": 0.7672573504940376, "rms_envelope_nrmse": 0.057983279880233905, "envelope_correlation": 0.9955385310760527, "rms_gain_error_db": -0.048303482578909596, "reference_mean_power": 0.007432639167823318, "output_mean_power": 0.007350429243899583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.369590439777949e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014473681837183883, "output_power": 0.001412911424799907, "reference_active": true, "delta_db": -0.10464084815803601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002579988221497507, "output_power": 0.00023245344230003834, "reference_active": true, "delta_db": -0.4528174142801652}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2677922934851034e-05, "output_power": 3.0709221200640846e-05, "reference_active": true, "delta_db": -0.26985641547626454}], "active_band_count": 3, "active_band_mae_db": 0.27577155930482194}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022253035095363898, "log_rms_mae_db": 0.4249474842096034, "rms_envelope_nrmse": 0.030890000123393803, "envelope_correlation": 0.9986114226582088, "rms_gain_error_db": -0.04830732158545862, "reference_mean_power": 0.0074694638279480755, "output_mean_power": 0.007386840069131457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0676309697735546e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013241614802002638, "output_power": 0.0012890375870807083, "reference_active": true, "delta_db": -0.11675368979314003}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011975787910962369, "output_power": 0.0001110247740744541, "reference_active": true, "delta_db": -0.32884198030436856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4293821094800366e-06, "output_power": 1.9571906462823917e-06, "reference_active": true, "delta_db": -0.9386269738874344}], "active_band_count": 3, "active_band_mae_db": 0.461407547994981}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017894181312046162, "log_rms_mae_db": 1.7270093788865875, "rms_envelope_nrmse": 0.09868646154278178, "envelope_correlation": 0.9923204560081461, "rms_gain_error_db": -0.13981758456056137, "reference_mean_power": 0.0017811195275832714, "output_mean_power": 0.0017246910380846379}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1758328243827778e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007391946250824912, "output_power": 0.0007169447752381917, "reference_active": true, "delta_db": -0.13273096173898963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015375544244981539, "output_power": 0.00013398836827650776, "reference_active": true, "delta_db": -0.5976339908124777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0532547306567537e-05, "output_power": 9.79442007255219e-06, "reference_active": true, "delta_db": -0.3155469189869545}], "active_band_count": 3, "active_band_mae_db": 0.34863729051280723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016320284141686103, "log_rms_mae_db": 1.110075078702755, "rms_envelope_nrmse": 0.06615100814137545, "envelope_correlation": 0.9963344426524846, "rms_gain_error_db": -0.13982755895720667, "reference_mean_power": 0.0017900337371132, "output_mean_power": 0.0017333188514295234}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0791360125207298e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007228891058818755, "output_power": 0.000698522119669983, "reference_active": true, "delta_db": -0.1489151666674789}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980895739386466e-05, "output_power": 6.182499255436413e-05, "reference_active": true, "delta_db": -0.5274707907764338}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.398569514368023e-07, "output_power": 8.327944809982344e-07, "reference_active": true, "delta_db": -0.5252391952577876}], "active_band_count": 3, "active_band_mae_db": 0.4005417175672334}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.442874453255196e-05, "log_rms_mae_db": 1.7414004844089155, "rms_envelope_nrmse": 0.15232450187846797, "envelope_correlation": 0.980942785329796, "rms_gain_error_db": -0.49741261418738814, "reference_mean_power": 0.000318038298123639, "output_mean_power": 0.00028362085335626076}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6089659140343417e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.507437744605689e-05, "output_power": 8.020037402288383e-05, "reference_active": true, "delta_db": -0.2562238604944221}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3978123456671057e-05, "output_power": 1.9839706036303604e-05, "reference_active": true, "delta_db": -0.822799589765056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6559576292684446e-06, "output_power": 3.618839038542654e-06, "reference_active": true, "delta_db": -1.0943975277887683}], "active_band_count": 3, "active_band_mae_db": 0.7244736593494153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.707405683354541e-05, "log_rms_mae_db": 1.1356341008072313, "rms_envelope_nrmse": 0.09665733966635663, "envelope_correlation": 0.9934811307719615, "rms_gain_error_db": -0.49771023617518995, "reference_mean_power": 0.00031952379334241086, "output_mean_power": 0.00028492606471358124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4169476048907225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.318339905927379e-05, "output_power": 7.825984659162257e-05, "reference_active": true, "delta_db": -0.2649767039053412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1060276940136043e-05, "output_power": 9.249165196612599e-06, "reference_active": true, "delta_db": -0.7766346515632149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55972585077424e-07, "output_power": 2.172378371781293e-07, "reference_active": true, "delta_db": -0.7125798412220257}], "active_band_count": 3, "active_band_mae_db": 0.5847303988968605}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.0412968032519945e-05, "log_rms_mae_db": 1.8532488111850443, "rms_envelope_nrmse": 0.16769433521257127, "envelope_correlation": 0.9806170234065099, "rms_gain_error_db": 0.01487309141891542, "reference_mean_power": 8.301357548516487e-05, "output_mean_power": 8.329835577171636e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.362198065588504e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2275109921127774e-05, "output_power": 3.113865653905504e-05, "reference_active": true, "delta_db": -0.15567858812236052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8777051155174143e-06, "output_power": 2.2558286734148655e-06, "reference_active": true, "delta_db": -1.0574017622212448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.623133887825791e-07, "output_power": 7.215891073299309e-07, "reference_active": true, "delta_db": 0.3722643426256341}], "active_band_count": 3, "active_band_mae_db": 0.5284482309897465}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.404038681520291e-05, "log_rms_mae_db": 1.2293411764072177, "rms_envelope_nrmse": 0.09759217131243969, "envelope_correlation": 0.9930842323369674, "rms_gain_error_db": 0.013962644843350885, "reference_mean_power": 8.336414632310516e-05, "output_mean_power": 8.363259483851911e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1059391757915914e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.094184470310106e-05, "output_power": 2.9936785288946523e-05, "reference_active": true, "delta_db": -0.1434103954246713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2229001330453808e-06, "output_power": 9.999736210143507e-07, "reference_active": true, "delta_db": -0.8740244864525503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7483618813517085e-08, "output_power": 4.8675510368864e-08, "reference_active": true, "delta_db": 0.10766703926965998}], "active_band_count": 3, "active_band_mae_db": 0.3750339737156272}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9198096638080558, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3949a084f4753f90fa287288ef9e5595d2845a5694753879659a28a14fbc46e3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.981137216726637, "standard_deviation": 0.0009834126397550227, "minimum": 0.9791811776240016, "maximum": 0.9825205095815371, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7091797366215564, "standard_deviation": 0.029658415636584733, "minimum": 1.6726607085810727, "maximum": 1.7635927892868144, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8233711430481129, "standard_deviation": 0.19652364375125364, "minimum": 0.6266517299066656, "maximum": 1.1332811175939674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1516680721590964, "standard_deviation": 0.003686950383399487, "minimum": 0.14642776806579022, "maximum": 0.15847117994684773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9940055768108131, "standard_deviation": 0.0006906783850274059, "minimum": 0.9930510098118177, "maximum": 0.9949081309179677, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0924102591912912, "standard_deviation": 0.04133606471887091, "minimum": 1.0262584927974103, "maximum": 1.1399505568292734, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8329958661103338, "standard_deviation": 0.28393404340580963, "minimum": 0.47005469661602267, "maximum": 1.3794360352803887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09415485681724305, "standard_deviation": 0.005022841201029433, "minimum": 0.08839935652092337, "maximum": 0.10221243910331484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9800043735241686, "standard_deviation": 0.0017064267563429898, "minimum": 0.9776217923806841, "maximum": 0.9830446708256299, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.821977492764031, "standard_deviation": 0.027064877630760206, "minimum": 1.7711663823062278, "maximum": 1.8532488111850443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.42527725974955427, "standard_deviation": 0.1988316752168462, "minimum": 0.08259386026898467, "maximum": 0.6783483025082585, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17033165312073092, "standard_deviation": 0.008041108935006818, "minimum": 0.1571542619565664, "maximum": 0.18300572864425735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992050419598145, "standard_deviation": 0.0013321171358239604, "minimum": 0.9891210326870582, "maximum": 0.9933770433382126, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1973698175010803, "standard_deviation": 0.031557965498060274, "minimum": 1.1542051550276269, "maximum": 1.2336199403061754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.39560165019818083, "standard_deviation": 0.22255235850458127, "minimum": 0.10649729164333484, "maximum": 0.6692545549557715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10501212243776392, "standard_deviation": 0.008947322405526701, "minimum": 0.09759217131243969, "maximum": 0.12521693642698983, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925735059416095, "standard_deviation": 0.0006850595682539029, "minimum": 0.9918870372311849, "maximum": 0.993686636491901, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7461682046206988, "standard_deviation": 0.04849116461407829, "minimum": 1.6888401936893034, "maximum": 1.839288450335106, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5288034694607391, "standard_deviation": 0.21060427913390975, "minimum": 0.3300606164504587, "maximum": 0.9522573748445634, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09770888243490891, "standard_deviation": 0.004528945683772773, "minimum": 0.09071254968994633, "maximum": 0.10209634029652095, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971821742224513, "standard_deviation": 0.0005730187482590707, "minimum": 0.9963344426524846, "maximum": 0.9980281665031279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.116896344931365, "standard_deviation": 0.04290864927923963, "minimum": 1.0635353344641494, "maximum": 1.1934413134082744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5778407117127552, "standard_deviation": 0.21826406172846896, "minimum": 0.31885280363667656, "maximum": 0.9148504409486818, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05954880566017039, "standard_deviation": 0.005596274168887008, "minimum": 0.050616240692051354, "maximum": 0.06615100814137545, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20759626069677786, "standard_deviation": 0.02348946472851993, "minimum": 0.17573926851560187, "maximum": 0.24139635097286116, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.468523249477549, "standard_deviation": 0.005808039732881478, "minimum": 5.458875467166702, "maximum": 5.473575621602017, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5632465658527352, "standard_deviation": 0.014992392310788916, "minimum": 0.5523512392211989, "maximum": 0.5980561039557237, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.046726578957947, "standard_deviation": 0.018411004786859152, "minimum": -0.08137470695353116, "maximum": -0.02722903075965216, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963943043127788, "standard_deviation": 0.0003977178385520467, "minimum": 0.9955385310760527, "maximum": 0.9967735757606593, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7484278103825803, "standard_deviation": 0.014074357406353178, "minimum": 0.7175556127303006, "maximum": 0.7672573504940376, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3344456024854813, "standard_deviation": 0.1082240186874773, "minimum": 0.20268476670314117, "maximum": 0.5247581280287319, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.052163310484656986, "standard_deviation": 0.0028390022318258916, "minimum": 0.04924004038211335, "maximum": 0.057983279880233905, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989146747814049, "standard_deviation": 0.0001633095703381835, "minimum": 0.9986114226582088, "maximum": 0.9991552207318745, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40221798792741226, "standard_deviation": 0.02136367003376249, "minimum": 0.3728604476965578, "maximum": 0.43529041432761795, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3485638055411824, "standard_deviation": 0.09937323565795285, "minimum": 0.25021567544545653, "maximum": 0.5019233122137442, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02741455853787348, "standard_deviation": 0.0020791474769810783, "minimum": 0.023916910072933333, "maximum": 0.030890000123393803, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1320-122617-0039", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7673544883728027, "predicted_mos": 2.7673544883728027, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "86bf04b96675a4e6ed2869f3ff57c1fcde312fc71c2a067208acb08c2ae1e11e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.510958000086248, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7460995332294494, "全帯域_10ms_rms_envelope_nrmse": 0.05382606194903739, "全帯域_10ms_envelope_correlation": 0.9961615267284727, "全帯域_10ms_modulation_mae_db": 0.24085748249692215, "全帯域_40ms_log_rms_mae_db": 0.41164227919321467, "全帯域_40ms_rms_envelope_nrmse": 0.029879684064921783, "全帯域_40ms_envelope_correlation": 0.9986975292820601, "全帯域_40ms_modulation_mae_db": 0.3426061821188851, "300–1000 Hz_10ms_log_rms_mae_db": 1.6470023881133624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09456204242039377, "300–1000 Hz_10ms_envelope_correlation": 0.9929868522829812, "300–1000 Hz_10ms_modulation_mae_db": 0.20530205778706456, "300–1000 Hz_40ms_log_rms_mae_db": 1.0680484699302368, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06083650019252281, "300–1000 Hz_40ms_envelope_correlation": 0.9969633847002964, "300–1000 Hz_40ms_modulation_mae_db": 0.186539625549467, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6915561042716636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14016514258320986, "1000–3000 Hz_10ms_envelope_correlation": 0.983005541587472, "1000–3000 Hz_10ms_modulation_mae_db": 0.4632069686508035, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0431345014951723, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08104403895902189, "1000–3000 Hz_40ms_envelope_correlation": 0.9947009672692592, "1000–3000 Hz_40ms_modulation_mae_db": 0.3670532707660519, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7965080421464086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17665141594018255, "3000–6000 Hz_10ms_envelope_correlation": 0.9785661812164687, "3000–6000 Hz_10ms_modulation_mae_db": 0.6319522746212044, "3000–6000 Hz_40ms_log_rms_mae_db": 1.183669707506913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11475046264293189, "3000–6000 Hz_40ms_envelope_correlation": 0.9904699688613078, "3000–6000 Hz_40ms_modulation_mae_db": 0.7411948420046132, "mr_stft_db": 5.452602490420618, "band_mae_db": 0.12330761197146133, "rms_gain_error_db": -0.038418266771167386, "output_peak": 0.5672387597395535}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.777134610568236, "全帯域_10ms_rms_envelope_nrmse": 0.05156740522088677, "全帯域_10ms_envelope_correlation": 0.9965931022876268, "全帯域_10ms_modulation_mae_db": 0.29348572275910856, "全帯域_40ms_log_rms_mae_db": 0.44630218218270773, "全帯域_40ms_rms_envelope_nrmse": 0.030027099597635283, "全帯域_40ms_envelope_correlation": 0.9988288949452125, "全帯域_40ms_modulation_mae_db": 0.264499156910267, "300–1000 Hz_10ms_log_rms_mae_db": 1.6420100727458322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09380584091295595, "300–1000 Hz_10ms_envelope_correlation": 0.9934341558276729, "300–1000 Hz_10ms_modulation_mae_db": 1.159896238537062, "300–1000 Hz_40ms_log_rms_mae_db": 1.0410377622821014, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05896052146653912, "300–1000 Hz_40ms_envelope_correlation": 0.9975561366940334, "300–1000 Hz_40ms_modulation_mae_db": 1.1663869710019126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6694932945171723, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15621554038809923, "1000–3000 Hz_10ms_envelope_correlation": 0.9820197701869386, "1000–3000 Hz_10ms_modulation_mae_db": 0.9662705894309913, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1024204682054806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10971235646423819, "1000–3000 Hz_40ms_envelope_correlation": 0.9933889933002472, "1000–3000 Hz_40ms_modulation_mae_db": 1.1286132990804143, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8303360150668224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17375461010166293, "3000–6000 Hz_10ms_envelope_correlation": 0.9794268451069775, "3000–6000 Hz_10ms_modulation_mae_db": 0.3765812128703758, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2368574069115197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11143166635270008, "3000–6000 Hz_40ms_envelope_correlation": 0.9911214434408022, "3000–6000 Hz_40ms_modulation_mae_db": 0.5168317404554837, "mr_stft_db": 5.453543826738843, "band_mae_db": 0.2899479692750631, "rms_gain_error_db": -0.09936339668775049, "output_peak": 0.5591075902049428}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7681272611843585, "全帯域_10ms_rms_envelope_nrmse": 0.05438358341506485, "全帯域_10ms_envelope_correlation": 0.9961682953404875, "全帯域_10ms_modulation_mae_db": 0.3685264874863541, "全帯域_40ms_log_rms_mae_db": 0.41567873509004655, "全帯域_40ms_rms_envelope_nrmse": 0.029760990523593493, "全帯域_40ms_envelope_correlation": 0.9988273798418092, "全帯域_40ms_modulation_mae_db": 0.3381050074391781, "300–1000 Hz_10ms_log_rms_mae_db": 1.7042253476453642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10107819570045018, "300–1000 Hz_10ms_envelope_correlation": 0.992018935350609, "300–1000 Hz_10ms_modulation_mae_db": 0.5253067061237644, "300–1000 Hz_40ms_log_rms_mae_db": 1.042173932996941, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05931654512804452, "300–1000 Hz_40ms_envelope_correlation": 0.9972158996374153, "300–1000 Hz_40ms_modulation_mae_db": 0.5041948953524642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6990361433510153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16328938932779105, "1000–3000 Hz_10ms_envelope_correlation": 0.9788738092899538, "1000–3000 Hz_10ms_modulation_mae_db": 1.1348845911529104, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0751163278356648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10286513849728833, "1000–3000 Hz_40ms_envelope_correlation": 0.9938384396265362, "1000–3000 Hz_40ms_modulation_mae_db": 0.7239006130893971, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7384044713235203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17538666367542619, "3000–6000 Hz_10ms_envelope_correlation": 0.9785622865786211, "3000–6000 Hz_10ms_modulation_mae_db": 0.3511419382986811, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1342462513000642, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11628022551786445, "3000–6000 Hz_40ms_envelope_correlation": 0.990322715185576, "3000–6000 Hz_40ms_modulation_mae_db": 0.4966445292784642, "mr_stft_db": 5.4585429213123575, "band_mae_db": 0.25570300139087526, "rms_gain_error_db": -0.09230816639414573, "output_peak": 0.5572077887893766}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7327086325679635, "全帯域_10ms_rms_envelope_nrmse": 0.054849098375404165, "全帯域_10ms_envelope_correlation": 0.9960096961732301, "全帯域_10ms_modulation_mae_db": 0.18557199168867033, "全帯域_40ms_log_rms_mae_db": 0.389882484062478, "全帯域_40ms_rms_envelope_nrmse": 0.02874813424758488, "全帯域_40ms_envelope_correlation": 0.9987983928562457, "全帯域_40ms_modulation_mae_db": 0.14226999864697162, "300–1000 Hz_10ms_log_rms_mae_db": 1.6920863893131795, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1010004861561592, "300–1000 Hz_10ms_envelope_correlation": 0.9921027435315694, "300–1000 Hz_10ms_modulation_mae_db": 0.9755254439824621, "300–1000 Hz_40ms_log_rms_mae_db": 1.0738773608455199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0576724033866777, "300–1000 Hz_40ms_envelope_correlation": 0.9974553019467716, "300–1000 Hz_40ms_modulation_mae_db": 1.0756641720137372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6628376734179358, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14751555287679632, "1000–3000 Hz_10ms_envelope_correlation": 0.9822063465338033, "1000–3000 Hz_10ms_modulation_mae_db": 0.5061403906016665, "1000–3000 Hz_40ms_log_rms_mae_db": 1.051221448129886, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09098244588479414, "1000–3000 Hz_40ms_envelope_correlation": 0.9945088273739018, "1000–3000 Hz_40ms_modulation_mae_db": 0.49487489648136646, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7255020588183425, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16002663300186637, "3000–6000 Hz_10ms_envelope_correlation": 0.9823941957626577, "3000–6000 Hz_10ms_modulation_mae_db": 0.420559703918779, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1045236227893849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0985187858157871, "3000–6000 Hz_40ms_envelope_correlation": 0.9934435687801358, "3000–6000 Hz_40ms_modulation_mae_db": 0.3045249752990607, "mr_stft_db": 5.441744145713542, "band_mae_db": 0.2225047287780384, "rms_gain_error_db": -0.040362594523462594, "output_peak": 0.554842212083456}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.741310335372341, "全帯域_10ms_rms_envelope_nrmse": 0.04964818442006126, "全帯域_10ms_envelope_correlation": 0.9967208480214876, "全帯域_10ms_modulation_mae_db": 0.3213628318689213, "全帯域_40ms_log_rms_mae_db": 0.3975007774805272, "全帯域_40ms_rms_envelope_nrmse": 0.027032329340280565, "全帯域_40ms_envelope_correlation": 0.9989181165424023, "全帯域_40ms_modulation_mae_db": 0.347211335517402, "300–1000 Hz_10ms_log_rms_mae_db": 1.6960395968557878, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09058721811459038, "300–1000 Hz_10ms_envelope_correlation": 0.9935555250778122, "300–1000 Hz_10ms_modulation_mae_db": 0.9253992177125697, "300–1000 Hz_40ms_log_rms_mae_db": 1.1141045348615488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055002244325910214, "300–1000 Hz_40ms_envelope_correlation": 0.997525893999554, "300–1000 Hz_40ms_modulation_mae_db": 0.8348738866181478, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7084738697143897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.145580945935793, "1000–3000 Hz_10ms_envelope_correlation": 0.9821821941814345, "1000–3000 Hz_10ms_modulation_mae_db": 0.6741898514045742, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0913369130157464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08897525113267027, "1000–3000 Hz_40ms_envelope_correlation": 0.9941359021379205, "1000–3000 Hz_40ms_modulation_mae_db": 0.6140767790295735, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8049357563825077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16310357208528276, "3000–6000 Hz_10ms_envelope_correlation": 0.9814956442987242, "3000–6000 Hz_10ms_modulation_mae_db": 0.3584945251222161, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1963068485220965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10036054670307892, "3000–6000 Hz_40ms_envelope_correlation": 0.99280328852966, "3000–6000 Hz_40ms_modulation_mae_db": 0.7330083317628486, "mr_stft_db": 5.447012871161696, "band_mae_db": 0.17742063780283943, "rms_gain_error_db": -0.029176746640521768, "output_peak": 0.5655987958705513}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7589071694966265, "全帯域_10ms_rms_envelope_nrmse": 0.052443437361717075, "全帯域_10ms_envelope_correlation": 0.9963505063608856, "全帯域_10ms_modulation_mae_db": 0.28477848490612084, "全帯域_40ms_log_rms_mae_db": 0.4225876463542432, "全帯域_40ms_rms_envelope_nrmse": 0.027031856816937882, "全帯域_40ms_envelope_correlation": 0.9989289801841695, "全帯域_40ms_modulation_mae_db": 0.21318784676209923, "300–1000 Hz_10ms_log_rms_mae_db": 1.697851206636303, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09438158284164637, "300–1000 Hz_10ms_envelope_correlation": 0.993112230592175, "300–1000 Hz_10ms_modulation_mae_db": 0.4841724547969341, "300–1000 Hz_40ms_log_rms_mae_db": 1.0933297412677176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05996257898744394, "300–1000 Hz_40ms_envelope_correlation": 0.9971779444609659, "300–1000 Hz_40ms_modulation_mae_db": 0.4517573083976618, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6347775547844121, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14470601195156674, "1000–3000 Hz_10ms_envelope_correlation": 0.9827518463871835, "1000–3000 Hz_10ms_modulation_mae_db": 1.1849306810446134, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0173587910510014, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08883060847059744, "1000–3000 Hz_40ms_envelope_correlation": 0.9945587506129044, "1000–3000 Hz_40ms_modulation_mae_db": 1.0787475676539449, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7447388196260583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15762638110878469, "3000–6000 Hz_10ms_envelope_correlation": 0.9826777545007666, "3000–6000 Hz_10ms_modulation_mae_db": 0.4264858449160515, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0954690106444023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09903945948281404, "3000–6000 Hz_40ms_envelope_correlation": 0.9928353154118419, "3000–6000 Hz_40ms_modulation_mae_db": 0.2640258297635922, "mr_stft_db": 5.433144659012657, "band_mae_db": 0.18687438786985588, "rms_gain_error_db": -0.03417073193201088, "output_peak": 0.5886305578807205}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7421877417018793, "全帯域_10ms_rms_envelope_nrmse": 0.05104318263694938, "全帯域_10ms_envelope_correlation": 0.9965660617110221, "全帯域_10ms_modulation_mae_db": 0.28097432336768896, "全帯域_40ms_log_rms_mae_db": 0.4017650519225068, "全帯域_40ms_rms_envelope_nrmse": 0.02787706279590368, "全帯域_40ms_envelope_correlation": 0.9988938588808565, "全帯域_40ms_modulation_mae_db": 0.22166441989224497, "300–1000 Hz_10ms_log_rms_mae_db": 1.632929921099929, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09355983999989152, "300–1000 Hz_10ms_envelope_correlation": 0.9932048805932457, "300–1000 Hz_10ms_modulation_mae_db": 0.5720862613452885, "300–1000 Hz_40ms_log_rms_mae_db": 1.0692310276988977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05649378899017482, "300–1000 Hz_40ms_envelope_correlation": 0.9974974687574576, "300–1000 Hz_40ms_modulation_mae_db": 0.38271074337852773, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6945991040746962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14953106193341462, "1000–3000 Hz_10ms_envelope_correlation": 0.9808251287027415, "1000–3000 Hz_10ms_modulation_mae_db": 0.7680277377509329, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0869928793594557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09142578330918687, "1000–3000 Hz_40ms_envelope_correlation": 0.9933312924504365, "1000–3000 Hz_40ms_modulation_mae_db": 0.6996264777235305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7775824196292438, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1616333271745078, "3000–6000 Hz_10ms_envelope_correlation": 0.9821022353236869, "3000–6000 Hz_10ms_modulation_mae_db": 0.6030827836077451, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1412711121516936, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09390312185158144, "3000–6000 Hz_40ms_envelope_correlation": 0.9936487704749712, "3000–6000 Hz_40ms_modulation_mae_db": 0.8128891676739839, "mr_stft_db": 5.4649716240096, "band_mae_db": 0.17730243456673608, "rms_gain_error_db": -0.05689801329041267, "output_peak": 0.5709932198903478}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7661799617839009, "全帯域_10ms_rms_envelope_nrmse": 0.05210151599746199, "全帯域_10ms_envelope_correlation": 0.996419294707575, "全帯域_10ms_modulation_mae_db": 0.34023551704220933, "全帯域_40ms_log_rms_mae_db": 0.4110621836688909, "全帯域_40ms_rms_envelope_nrmse": 0.027082285262661123, "全帯域_40ms_envelope_correlation": 0.998958507233491, "全帯域_40ms_modulation_mae_db": 0.4421884704851375, "300–1000 Hz_10ms_log_rms_mae_db": 1.6769894668258882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0963044138141664, "300–1000 Hz_10ms_envelope_correlation": 0.9928796344538544, "300–1000 Hz_10ms_modulation_mae_db": 0.4184213532851117, "300–1000 Hz_40ms_log_rms_mae_db": 1.063392654495302, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055488380402719244, "300–1000 Hz_40ms_envelope_correlation": 0.9977439888852679, "300–1000 Hz_40ms_modulation_mae_db": 0.6709404637681496, "1000–3000 Hz_10ms_log_rms_mae_db": 1.658470030245348, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14110298822816844, "1000–3000 Hz_10ms_envelope_correlation": 0.9830358491103706, "1000–3000 Hz_10ms_modulation_mae_db": 0.3838417282265807, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1012080069816608, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08790218665019377, "1000–3000 Hz_40ms_envelope_correlation": 0.9938582631623829, "1000–3000 Hz_40ms_modulation_mae_db": 0.41006878187124, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7896251168485704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1700308696706566, "3000–6000 Hz_10ms_envelope_correlation": 0.9800857483255924, "3000–6000 Hz_10ms_modulation_mae_db": 0.43186119266047823, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1333337460318125, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10458036417734462, "3000–6000 Hz_40ms_envelope_correlation": 0.992623623709099, "3000–6000 Hz_40ms_modulation_mae_db": 0.7174002985201682, "mr_stft_db": 5.446354202508744, "band_mae_db": 0.222534101700441, "rms_gain_error_db": -0.05638887998985505, "output_peak": 0.5790909311803095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.653612510113159, "spectral_convergence": 0.1713856372076731, "spectral_convergence_status": "ok", "amplitude_floor": 2.2281588625700377e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.264218728774583, "spectral_convergence": 0.16311822239020818, "spectral_convergence_status": "ok", "amplitude_floor": 1.3568197926870596e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.43997623237411, "spectral_convergence": 0.15987491612061377, "spectral_convergence_status": "ok", "amplitude_floor": 1.1580179000740219e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0038028140043242973, "output_power": 0.00382567706086296, "delta_db": 0.026032222988966708, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002493613270814922, "output_power": 0.002468254316048456, "delta_db": -0.044391954517785666, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006858506604602334, "output_power": 0.0006465721232834598, "delta_db": -0.2561258495557372, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002628229680145485, "output_power": 0.0002392629078825189, "delta_db": -0.4078793877068968, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001273968713381539, "output_power": 0.00012738177580624946, "delta_db": -0.0005146354522750532, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.457810201861998e-05, "output_power": 5.451653773344638e-05, "delta_db": -0.004901621607106464, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12330761197146133, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.038418266771167386, "pitch": null, "output_rms_dbfs": -21.330239658716064, "output_peak": 0.5672387597395535, "reference_rms_dbfs": -21.2918213919449, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.452602490420618}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025738076998093343, "log_rms_mae_db": 0.7460995332294494, "rms_envelope_nrmse": 0.05382606194903739, "envelope_correlation": 0.9961615267284727, "rms_gain_error_db": -0.03819855559919192, "reference_mean_power": 0.007432639167823318, "output_mean_power": 0.007367551733611944}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.369590439777949e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014473681837183883, "output_power": 0.0013821280793894818, "reference_active": true, "delta_db": -0.20030731451025793}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002579988221497507, "output_power": 0.00025787776808267114, "reference_active": true, "delta_db": -0.0020382050786937394}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2677922934851034e-05, "output_power": 2.8988901022605008e-05, "reference_active": true, "delta_db": -0.5202269279018148}], "active_band_count": 3, "active_band_mae_db": 0.24085748249692215}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022253035095363898, "log_rms_mae_db": 0.41164227919321467, "rms_envelope_nrmse": 0.029879684064921783, "envelope_correlation": 0.9986975292820601, "rms_gain_error_db": -0.03822069269618042, "reference_mean_power": 0.0074694638279480755, "output_mean_power": 0.007404016180860044}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0676309697735546e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013241614802002638, "output_power": 0.0012619196025267605, "reference_active": true, "delta_db": -0.2090926343147559}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011975787910962369, "output_power": 0.00012298246339402784, "reference_active": true, "delta_db": 0.11539091783076713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4293821094800366e-06, "output_power": 2.0661522927083834e-06, "reference_active": true, "delta_db": -0.7033349942111324}], "active_band_count": 3, "active_band_mae_db": 0.3426061821188851}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017894181312046162, "log_rms_mae_db": 1.6470023881133624, "rms_envelope_nrmse": 0.09456204242039377, "envelope_correlation": 0.9929868522829812, "rms_gain_error_db": -0.14648944290345428, "reference_mean_power": 0.0017811195275832714, "output_mean_power": 0.0017220435119130347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1758328243827778e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007391946250824912, "output_power": 0.0007053589788395984, "reference_active": true, "delta_db": -0.20348601282195494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015375544244981539, "output_power": 0.00014479734870331308, "reference_active": true, "delta_db": -0.26069887521795865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0532547306567537e-05, "output_power": 1.090700587218399e-05, "reference_active": true, "delta_db": 0.15172128532128012}], "active_band_count": 3, "active_band_mae_db": 0.20530205778706456}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016320284141686103, "log_rms_mae_db": 1.0680484699302368, "rms_envelope_nrmse": 0.06083650019252281, "envelope_correlation": 0.9969633847002964, "rms_gain_error_db": -0.14647085308151298, "reference_mean_power": 0.0017900337371132, "output_mean_power": 0.001730669463779808}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0791360125207298e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007228891058818755, "output_power": 0.0006899854340544426, "reference_active": true, "delta_db": -0.2023175721542675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980895739386466e-05, "output_power": 6.753486783616566e-05, "reference_active": true, "delta_db": -0.1438309750973683}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.398569514368023e-07, "output_power": 8.947768204333311e-07, "reference_active": true, "delta_db": -0.21347032939676508}], "active_band_count": 3, "active_band_mae_db": 0.186539625549467}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.442874453255196e-05, "log_rms_mae_db": 1.6915561042716636, "rms_envelope_nrmse": 0.14016514258320986, "envelope_correlation": 0.983005541587472, "rms_gain_error_db": -0.3404322847457269, "reference_mean_power": 0.000318038298123639, "output_mean_power": 0.0002940601621233082}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6089659140343417e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.507437744605689e-05, "output_power": 8.008331922675988e-05, "reference_active": true, "delta_db": -0.2625671459583207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3978123456671057e-05, "output_power": 2.0657978239706706e-05, "reference_active": true, "delta_db": -0.647273763078689}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6559576292684446e-06, "output_power": 4.168991596824589e-06, "reference_active": true, "delta_db": -0.4797799969154009}], "active_band_count": 3, "active_band_mae_db": 0.4632069686508035}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.707405683354541e-05, "log_rms_mae_db": 1.0431345014951723, "rms_envelope_nrmse": 0.08104403895902189, "envelope_correlation": 0.9947009672692592, "rms_gain_error_db": -0.34087453401851675, "reference_mean_power": 0.00031952379334241086, "output_mean_power": 0.0002954035771506953}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4169476048907225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.318339905927379e-05, "output_power": 7.8160148161327e-05, "reference_active": true, "delta_db": -0.27051288667577145}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1060276940136043e-05, "output_power": 9.868423504279667e-06, "reference_active": true, "delta_db": -0.49518222473191464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55972585077424e-07, "output_power": 2.369446921047196e-07, "reference_active": true, "delta_db": -0.33546470089046954}], "active_band_count": 3, "active_band_mae_db": 0.3670532707660519}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.0412968032519945e-05, "log_rms_mae_db": 1.7965080421464086, "rms_envelope_nrmse": 0.17665141594018255, "envelope_correlation": 0.9785661812164687, "rms_gain_error_db": 0.04646570344158861, "reference_mean_power": 8.301357548516487e-05, "output_mean_power": 8.390651631458986e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.362198065588504e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2275109921127774e-05, "output_power": 3.501524431925972e-05, "reference_active": true, "delta_db": 0.3538943094747278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8777051155174143e-06, "output_power": 2.2864491762004943e-06, "reference_active": true, "delta_db": -0.9988473660267329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.623133887825791e-07, "output_power": 5.844562667279947e-07, "reference_active": true, "delta_db": -0.5431151483621526}], "active_band_count": 3, "active_band_mae_db": 0.6319522746212044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.404038681520291e-05, "log_rms_mae_db": 1.183669707506913, "rms_envelope_nrmse": 0.11475046264293189, "envelope_correlation": 0.9904699688613078, "rms_gain_error_db": 0.04697314997292473, "reference_mean_power": 8.336414632310516e-05, "output_mean_power": 8.4270704031106e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1059391757915914e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.094184470310106e-05, "output_power": 3.3579163151053694e-05, "reference_active": true, "delta_db": 0.355236665166327}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2229001330453808e-06, "output_power": 9.819899748409692e-07, "reference_active": true, "delta_db": -0.9528393815936784}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7483618813517085e-08, "output_power": 3.8458554955407945e-08, "reference_active": true, "delta_db": -0.9155084792538342}], "active_band_count": 3, "active_band_mae_db": 0.7411948420046132}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9156816781319583, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9b9b2b035a02eb61e8b64c215a19754e4d16389d13a9002fc05ecde447eae2a4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9818625607474872, "standard_deviation": 0.001400398375015154, "minimum": 0.9788738092899538, "maximum": 0.9830358491103706, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.677405471797079, "standard_deviation": 0.02501287167460898, "minimum": 1.6347775547844121, "maximum": 1.7084738697143897, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7601865672828841, "standard_deviation": 0.3082451469488382, "minimum": 0.3838417282265807, "maximum": 1.1849306810446134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1485133291531049, "standard_deviation": 0.007813790528290033, "minimum": 0.14016514258320986, "maximum": 0.16328938932779105, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9940401794916985, "standard_deviation": 0.0005257249381961069, "minimum": 0.9933312924504365, "maximum": 0.9947009672692592, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0710986670092586, "standard_deviation": 0.03077813328880526, "minimum": 1.0173587910510014, "maximum": 1.1024204682054806, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6896202107119398, "standard_deviation": 0.28566745965012075, "minimum": 0.3670532707660519, "maximum": 1.1286132990804143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09271722617099887, "standard_deviation": 0.009142115613094425, "minimum": 0.08104403895902189, "maximum": 0.10971235646423819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9806638613891869, "standard_deviation": 0.0017109691356112294, "minimum": 0.9785622865786211, "maximum": 0.9826777545007666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.775954087480184, "standard_deviation": 0.036526152195018015, "minimum": 1.7255020588183425, "maximum": 1.8303360150668224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.45001993450194144, "standard_deviation": 0.1081177368505205, "minimum": 0.3511419382986811, "maximum": 0.6319522746212044, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16727668409479624, "standard_deviation": 0.007541886456446161, "minimum": 0.15762638110878469, "maximum": 0.17665141594018255, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921585867991742, "standard_deviation": 0.0013233727911563145, "minimum": 0.990322715185576, "maximum": 0.9936487704749712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1532097132322359, "standard_deviation": 0.04837890962212605, "minimum": 1.0954690106444023, "maximum": 1.2368574069115197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5733149643447768, "standard_deviation": 0.21004031739543427, "minimum": 0.2640258297635922, "maximum": 0.8128891676739839, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10485807906801281, "standard_deviation": 0.00833148688039648, "minimum": 0.09390312185158144, "maximum": 0.11628022551786445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.99291186971374, "standard_deviation": 0.0005696807781338218, "minimum": 0.992018935350609, "maximum": 0.9935555250778122, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6736417986544558, "standard_deviation": 0.028641132576075385, "minimum": 1.632929921099929, "maximum": 1.7042253476453642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6582637166962821, "standard_deviation": 0.32552420979562896, "minimum": 0.20530205778706456, "maximum": 1.159896238537062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09565995249503173, "standard_deviation": 0.003678401466840706, "minimum": 0.09058721811459038, "maximum": 0.10107819570045018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973920023852203, "standard_deviation": 0.0002519743792672479, "minimum": 0.9969633847002964, "maximum": 0.9977439888852679, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.070649435547283, "standard_deviation": 0.024384727770167626, "minimum": 1.0410377622821014, "maximum": 1.1141045348615488, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6591335082600085, "standard_deviation": 0.3440501082803028, "minimum": 0.186539625549467, "maximum": 1.1663869710019126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05796662036000405, "standard_deviation": 0.002145266943537053, "minimum": 0.055002244325910214, "maximum": 0.06083650019252281, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20694935916941382, "standard_deviation": 0.051962036284323, "minimum": 0.12330761197146133, "maximum": 0.2899479692750631, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.449739592609757, "standard_deviation": 0.009931167577378837, "minimum": 5.433144659012657, "maximum": 5.4649716240096, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5678387319549072, "standard_deviation": 0.01154122675337197, "minimum": 0.554842212083456, "maximum": 0.5886305578807205, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05588584952866582, "standard_deviation": 0.026589928787896833, "minimum": -0.09936339668775049, "maximum": -0.029176746640521768, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963736664163485, "standard_deviation": 0.00024726397430985204, "minimum": 0.9960096961732301, "maximum": 0.9967208480214876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7540819057380943, "standard_deviation": 0.015691055525926693, "minimum": 0.7327086325679635, "maximum": 0.777134610568236, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.28947410520199945, "standard_deviation": 0.05746117245296267, "minimum": 0.18557199168867033, "maximum": 0.3685264874863541, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05248280867207286, "standard_deviation": 0.0017772015081138934, "minimum": 0.04964818442006126, "maximum": 0.054849098375404165, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988564574707809, "standard_deviation": 8.54477907168971e-05, "minimum": 0.9986975292820601, "maximum": 0.998958507233491, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4120526674943269, "standard_deviation": 0.01735109074164039, "minimum": 0.389882484062478, "maximum": 0.44630218218270773, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28896655222152323, "standard_deviation": 0.09602955659246003, "minimum": 0.14226999864697162, "maximum": 0.4421884704851375, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02842993033118984, "standard_deviation": 0.0013392651687561966, "minimum": 0.027031856816937882, "maximum": 0.030027099597635283, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1580-141083-0010", "language": "en", "style": "normal", "speaker": "librispeech_1580", "source_utterance_id": "1580-141083-0010", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 85120, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1580-141083-0010.wav", "source_file_sha256": "529f1238a1ae020571bff5a1b8a751809fc517d9b162a6e3fdde4500f628b626", "decoded_pcm_sha256": "d43cef78fec64a97f03dd03591e102c104a281997670664d424ccf19276cb4a0", "original_source_sha256": "15b52d8b6b3ab1e0d95b51380278481420b17cdd350b6fe62c29aca97d46a6a7", "corpus_entry_id": "librispeech_1580-141083-0010"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.066134452819824, "predicted_mos": 4.066134452819824, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "60e50d988c75b342274e7cc9b831c4110606ab222f94130695a684705250abb8", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.242711352999322, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1580-141083-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "60e50d988c75b342274e7cc9b831c4110606ab222f94130695a684705250abb8", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "69e1962ed3ad3de0c128b68c6c38b1999259bf5ef6b5c7aeeab25571bb776795"}, "rows": [{"case": "librispeech_1580-141083-0010", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1630260944366455, "predicted_mos": 2.1630260944366455, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "7592fcdd7b8332273a04a05189ac27604844fc8cc060ccb31a9184f00eb7d0ca", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.530076839000685, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1580-141083-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7592fcdd7b8332273a04a05189ac27604844fc8cc060ccb31a9184f00eb7d0ca", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8241597849474137, "全帯域_10ms_rms_envelope_nrmse": 0.08089452511266484, "全帯域_10ms_envelope_correlation": 0.9939680619451456, "全帯域_10ms_modulation_mae_db": 0.09503696935854965, "全帯域_40ms_log_rms_mae_db": 0.37943156011893786, "全帯域_40ms_rms_envelope_nrmse": 0.043838167734456854, "全帯域_40ms_envelope_correlation": 0.9980983501539856, "全帯域_40ms_modulation_mae_db": 0.08005690573196822, "300–1000 Hz_10ms_log_rms_mae_db": 1.5570809964756103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11375398245641226, "300–1000 Hz_10ms_envelope_correlation": 0.9892644119060378, "300–1000 Hz_10ms_modulation_mae_db": 0.6012157996666865, "300–1000 Hz_40ms_log_rms_mae_db": 0.8321332913607543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06921096320277062, "300–1000 Hz_40ms_envelope_correlation": 0.995673015234016, "300–1000 Hz_40ms_modulation_mae_db": 0.4153090979669642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3602079370771332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12511714824572776, "1000–3000 Hz_10ms_envelope_correlation": 0.9883189928333512, "1000–3000 Hz_10ms_modulation_mae_db": 0.23766312145799043, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6771889823377854, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06699705707822338, "1000–3000 Hz_40ms_envelope_correlation": 0.9965450139105627, "1000–3000 Hz_40ms_modulation_mae_db": 0.3775467544037711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5082767818997689, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16169967084894318, "3000–6000 Hz_10ms_envelope_correlation": 0.9811745558981099, "3000–6000 Hz_10ms_modulation_mae_db": 1.007361559612133, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7814066408516537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09024474272964578, "3000–6000 Hz_40ms_envelope_correlation": 0.9937728821824671, "3000–6000 Hz_40ms_modulation_mae_db": 0.6563564282548954, "mr_stft_db": 5.716295215777005, "band_mae_db": 0.6795934512110953, "rms_gain_error_db": -0.050786954801364734, "output_peak": 0.6172037022044785}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8124278480964942, "全帯域_10ms_rms_envelope_nrmse": 0.08167534656189959, "全帯域_10ms_envelope_correlation": 0.993898175674608, "全帯域_10ms_modulation_mae_db": 0.20680455537519837, "全帯域_40ms_log_rms_mae_db": 0.41354074127054125, "全帯域_40ms_rms_envelope_nrmse": 0.0431843549294943, "全帯域_40ms_envelope_correlation": 0.9981683407699508, "全帯域_40ms_modulation_mae_db": 0.20928882599036536, "300–1000 Hz_10ms_log_rms_mae_db": 1.5431649595312107, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11644801761113727, "300–1000 Hz_10ms_envelope_correlation": 0.9887827775671222, "300–1000 Hz_10ms_modulation_mae_db": 0.9413224577043707, "300–1000 Hz_40ms_log_rms_mae_db": 0.8902924657530411, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06144406681409044, "300–1000 Hz_40ms_envelope_correlation": 0.9965968145325824, "300–1000 Hz_40ms_modulation_mae_db": 0.8719016921990862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3612704795866821, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12887768628407642, "1000–3000 Hz_10ms_envelope_correlation": 0.9877109778953084, "1000–3000 Hz_10ms_modulation_mae_db": 0.3704096659681717, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7435539844691907, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06934062451268583, "1000–3000 Hz_40ms_envelope_correlation": 0.9962373658011059, "1000–3000 Hz_40ms_modulation_mae_db": 0.478185824950821, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5044448910388397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1261913752569675, "3000–6000 Hz_10ms_envelope_correlation": 0.9885839902048366, "3000–6000 Hz_10ms_modulation_mae_db": 0.1965290349423181, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8029046947199567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06799064200857169, "3000–6000 Hz_40ms_envelope_correlation": 0.9964597800142005, "3000–6000 Hz_40ms_modulation_mae_db": 0.19469929448673415, "mr_stft_db": 5.69683274340795, "band_mae_db": 0.7483681969671165, "rms_gain_error_db": 0.0282229602485363, "output_peak": 0.5560603424367108}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8326558222416979, "全帯域_10ms_rms_envelope_nrmse": 0.09332551990187206, "全帯域_10ms_envelope_correlation": 0.9919692072247889, "全帯域_10ms_modulation_mae_db": 0.14988454663528314, "全帯域_40ms_log_rms_mae_db": 0.4442799601787826, "全帯域_40ms_rms_envelope_nrmse": 0.05423601744031444, "全帯域_40ms_envelope_correlation": 0.9970598427997681, "全帯域_40ms_modulation_mae_db": 0.3681292933954805, "300–1000 Hz_10ms_log_rms_mae_db": 1.5699618124642498, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11452739278722922, "300–1000 Hz_10ms_envelope_correlation": 0.9891185485993632, "300–1000 Hz_10ms_modulation_mae_db": 0.34465506167994525, "300–1000 Hz_40ms_log_rms_mae_db": 0.9012642500086024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061452931884604586, "300–1000 Hz_40ms_envelope_correlation": 0.9965895391952029, "300–1000 Hz_40ms_modulation_mae_db": 0.36376166004921195, "1000–3000 Hz_10ms_log_rms_mae_db": 1.354544541907925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14857350892245402, "1000–3000 Hz_10ms_envelope_correlation": 0.9835611803959728, "1000–3000 Hz_10ms_modulation_mae_db": 0.4869086895991608, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7543508911218585, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0888587736333295, "1000–3000 Hz_40ms_envelope_correlation": 0.9937753329375013, "1000–3000 Hz_40ms_modulation_mae_db": 0.6148192309809046, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5235699409743406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13717315321364182, "3000–6000 Hz_10ms_envelope_correlation": 0.986785186572166, "3000–6000 Hz_10ms_modulation_mae_db": 0.7433563839922264, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7898014351216702, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07160908175180797, "3000–6000 Hz_40ms_envelope_correlation": 0.9962384988404718, "3000–6000 Hz_40ms_modulation_mae_db": 0.8690378860926735, "mr_stft_db": 5.706628116767843, "band_mae_db": 0.7094980034653937, "rms_gain_error_db": -0.01863935778592811, "output_peak": 0.5597908486359982}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8174040215858996, "全帯域_10ms_rms_envelope_nrmse": 0.0875402870183732, "全帯域_10ms_envelope_correlation": 0.9930994286793446, "全帯域_10ms_modulation_mae_db": 0.5439705008720359, "全帯域_40ms_log_rms_mae_db": 0.3869017217613393, "全帯域_40ms_rms_envelope_nrmse": 0.04564624686411936, "全帯域_40ms_envelope_correlation": 0.9980272660940299, "全帯域_40ms_modulation_mae_db": 0.4690593342602088, "300–1000 Hz_10ms_log_rms_mae_db": 1.5321646343679722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11056886935695429, "300–1000 Hz_10ms_envelope_correlation": 0.9898536432568973, "300–1000 Hz_10ms_modulation_mae_db": 0.7406534058924034, "300–1000 Hz_40ms_log_rms_mae_db": 0.823988855387626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047462594405781426, "300–1000 Hz_40ms_envelope_correlation": 0.9980410155738303, "300–1000 Hz_40ms_modulation_mae_db": 0.44768720226664405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3144282708999644, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13776149302676163, "1000–3000 Hz_10ms_envelope_correlation": 0.9866752855744583, "1000–3000 Hz_10ms_modulation_mae_db": 1.3676384101083763, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7033733386228084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08246518454548252, "1000–3000 Hz_40ms_envelope_correlation": 0.9952012600155934, "1000–3000 Hz_40ms_modulation_mae_db": 1.4156365698373736, "3000–6000 Hz_10ms_log_rms_mae_db": 1.55609434624987, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14323550997443343, "3000–6000 Hz_10ms_envelope_correlation": 0.9863152482562476, "3000–6000 Hz_10ms_modulation_mae_db": 0.8856784514766044, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8251409091910601, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07973277095274761, "3000–6000 Hz_40ms_envelope_correlation": 0.9959336747886767, "3000–6000 Hz_40ms_modulation_mae_db": 0.6474822197322574, "mr_stft_db": 5.695550395849584, "band_mae_db": 0.7737454034670684, "rms_gain_error_db": 0.07728976335523316, "output_peak": 0.5770328394522529}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8370767745122433, "全帯域_10ms_rms_envelope_nrmse": 0.07391715796893339, "全帯域_10ms_envelope_correlation": 0.99520707053044, "全帯域_10ms_modulation_mae_db": 0.3692849046524635, "全帯域_40ms_log_rms_mae_db": 0.40603310100690887, "全帯域_40ms_rms_envelope_nrmse": 0.03810126810017439, "全帯域_40ms_envelope_correlation": 0.9988805057569424, "全帯域_40ms_modulation_mae_db": 0.2385064004665963, "300–1000 Hz_10ms_log_rms_mae_db": 1.4890914073487127, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11475410860925014, "300–1000 Hz_10ms_envelope_correlation": 0.9891254658525561, "300–1000 Hz_10ms_modulation_mae_db": 0.3325191016852194, "300–1000 Hz_40ms_log_rms_mae_db": 0.880605972814186, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06873664862302431, "300–1000 Hz_40ms_envelope_correlation": 0.9959140933519792, "300–1000 Hz_40ms_modulation_mae_db": 0.3271459094934165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.409253786714553, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11752472311617337, "1000–3000 Hz_10ms_envelope_correlation": 0.9898829081183989, "1000–3000 Hz_10ms_modulation_mae_db": 0.4562566712619705, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7616501769128498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0637883253082533, "1000–3000 Hz_40ms_envelope_correlation": 0.9971512665777674, "1000–3000 Hz_40ms_modulation_mae_db": 0.39144223047930343, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5603048352618683, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12336009392544758, "3000–6000 Hz_10ms_envelope_correlation": 0.9892454460558757, "3000–6000 Hz_10ms_modulation_mae_db": 0.1759588406759703, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7824444101184881, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06410358676442057, "3000–6000 Hz_40ms_envelope_correlation": 0.9971994363680764, "3000–6000 Hz_40ms_modulation_mae_db": 0.1937083404375637, "mr_stft_db": 5.704711462948942, "band_mae_db": 0.8151806333804098, "rms_gain_error_db": -0.15231387663717227, "output_peak": 0.533941509981277}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8207266680731098, "全帯域_10ms_rms_envelope_nrmse": 0.08178633686760337, "全帯域_10ms_envelope_correlation": 0.9938816825805222, "全帯域_10ms_modulation_mae_db": 0.48292337525225754, "全帯域_40ms_log_rms_mae_db": 0.3845776787374403, "全帯域_40ms_rms_envelope_nrmse": 0.03725038920545283, "全帯域_40ms_envelope_correlation": 0.9987166562153021, "全帯域_40ms_modulation_mae_db": 0.3420184186839734, "300–1000 Hz_10ms_log_rms_mae_db": 1.509889310838807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12430389846802777, "300–1000 Hz_10ms_envelope_correlation": 0.9871775299695941, "300–1000 Hz_10ms_modulation_mae_db": 0.6085145658717915, "300–1000 Hz_40ms_log_rms_mae_db": 0.8743964921336272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07204471932854546, "300–1000 Hz_40ms_envelope_correlation": 0.9954279699278984, "300–1000 Hz_40ms_modulation_mae_db": 0.476643696417419, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2954691857734075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1224794233028845, "1000–3000 Hz_10ms_envelope_correlation": 0.988783748519924, "1000–3000 Hz_10ms_modulation_mae_db": 0.3753012977002769, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6762360486336857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055673995488781006, "1000–3000 Hz_40ms_envelope_correlation": 0.9976017886573947, "1000–3000 Hz_40ms_modulation_mae_db": 0.23815169299422156, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5390852101637953, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1292999199429095, "3000–6000 Hz_10ms_envelope_correlation": 0.9880180984420194, "3000–6000 Hz_10ms_modulation_mae_db": 0.3327636701647173, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8123436017084307, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06727945195216661, "3000–6000 Hz_40ms_envelope_correlation": 0.9966611946605195, "3000–6000 Hz_40ms_modulation_mae_db": 0.3845746168745127, "mr_stft_db": 5.704619876020434, "band_mae_db": 0.7211080113634715, "rms_gain_error_db": -0.0911194493129901, "output_peak": 0.6024915334169589}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8790579671441623, "全帯域_10ms_rms_envelope_nrmse": 0.10013762299908066, "全帯域_10ms_envelope_correlation": 0.9909032927803212, "全帯域_10ms_modulation_mae_db": 0.41929497245351044, "全帯域_40ms_log_rms_mae_db": 0.4475378774586686, "全帯域_40ms_rms_envelope_nrmse": 0.047944534234201294, "全帯域_40ms_envelope_correlation": 0.9980159159566607, "全帯域_40ms_modulation_mae_db": 0.19758862258285795, "300–1000 Hz_10ms_log_rms_mae_db": 1.5309767013812918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11310305132321019, "300–1000 Hz_10ms_envelope_correlation": 0.9899397653133791, "300–1000 Hz_10ms_modulation_mae_db": 0.7672607065096871, "300–1000 Hz_40ms_log_rms_mae_db": 0.9117047727739068, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07218899651253047, "300–1000 Hz_40ms_envelope_correlation": 0.9961102224150651, "300–1000 Hz_40ms_modulation_mae_db": 0.7674718074817503, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3727078996005389, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16680980464187278, "1000–3000 Hz_10ms_envelope_correlation": 0.9792609177048902, "1000–3000 Hz_10ms_modulation_mae_db": 0.9843680015947495, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7473375645346898, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08579942363751257, "1000–3000 Hz_40ms_envelope_correlation": 0.9942113766954502, "1000–3000 Hz_40ms_modulation_mae_db": 1.1348328849465446, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5334090363089712, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14015692240028912, "3000–6000 Hz_10ms_envelope_correlation": 0.9860297661131656, "3000–6000 Hz_10ms_modulation_mae_db": 0.15078833706557612, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8322206119993484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07891430574216397, "3000–6000 Hz_40ms_envelope_correlation": 0.9955892478011242, "3000–6000 Hz_40ms_modulation_mae_db": 0.3390183266404507, "mr_stft_db": 5.716163371420861, "band_mae_db": 0.8280699081652934, "rms_gain_error_db": -0.15939495025547118, "output_peak": 0.562772531320049}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8211550596555124, "全帯域_10ms_rms_envelope_nrmse": 0.09649797114009247, "全帯域_10ms_envelope_correlation": 0.9914171323722368, "全帯域_10ms_modulation_mae_db": 0.7290115940097865, "全帯域_40ms_log_rms_mae_db": 0.4278896880572177, "全帯域_40ms_rms_envelope_nrmse": 0.051483409810801045, "全帯域_40ms_envelope_correlation": 0.9973440242605238, "全帯域_40ms_modulation_mae_db": 0.6192073763445504, "300–1000 Hz_10ms_log_rms_mae_db": 1.5026961151165286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.129282344531878, "300–1000 Hz_10ms_envelope_correlation": 0.9863500877679885, "300–1000 Hz_10ms_modulation_mae_db": 1.0257810278424946, "300–1000 Hz_40ms_log_rms_mae_db": 0.8766399875825237, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07594086802664511, "300–1000 Hz_40ms_envelope_correlation": 0.9949088557718971, "300–1000 Hz_40ms_modulation_mae_db": 0.9852769165653702, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3353843643496197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15150277413754223, "1000–3000 Hz_10ms_envelope_correlation": 0.9829131135163188, "1000–3000 Hz_10ms_modulation_mae_db": 0.8660650432620036, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7252925879075416, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09016506621790255, "1000–3000 Hz_40ms_envelope_correlation": 0.9935819892610943, "1000–3000 Hz_40ms_modulation_mae_db": 1.0891481791996416, "3000–6000 Hz_10ms_log_rms_mae_db": 1.531410342124549, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12278485630403309, "3000–6000 Hz_10ms_envelope_correlation": 0.989183936218784, "3000–6000 Hz_10ms_modulation_mae_db": 0.5928386157718568, "3000–6000 Hz_40ms_log_rms_mae_db": 0.84086835847893, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06787601886993855, "3000–6000 Hz_40ms_envelope_correlation": 0.9965303938093134, "3000–6000 Hz_40ms_modulation_mae_db": 0.24264287203904897, "mr_stft_db": 5.690409073565529, "band_mae_db": 0.6541895187712549, "rms_gain_error_db": -0.011074801540506414, "output_peak": 0.6094245499737653}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85120, "output_samples": 85120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.496151961204703, "spectral_convergence": 0.3508235687653108, "spectral_convergence_status": "ok", "amplitude_floor": 1.1374446115550809e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7197756836454445, "spectral_convergence": 0.37512125106048316, "spectral_convergence_status": "ok", "amplitude_floor": 9.193083193108671e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.932958002480866, "spectral_convergence": 0.3779695605037421, "spectral_convergence_status": "ok", "amplitude_floor": 7.3161107874683e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.1849686170678433e-06, "output_power": 5.060909893749273e-06, "delta_db": 3.6478340150896575, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004950757600852583, "output_power": 0.00048827750053521567, "delta_db": -0.06004950134297178, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008267786751374999, "output_power": 0.0008367054943988359, "delta_db": 0.05183354411040495, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007700805582515175, "output_power": 0.0007578049489854149, "delta_db": -0.06978721976119452, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006256626820185013, "output_power": 0.0006012367947702531, "delta_db": -0.1729470114122472, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00016725866300063496, "output_power": 0.00016439086449671856, "delta_db": -0.07510941555009604, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6795934512110953, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.050786954801364734, "pitch": null, "output_rms_dbfs": -25.446256977712874, "output_peak": 0.6172037022044785, "reference_rms_dbfs": -25.39547002291151, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.716295215777005}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.0002036923290800657, "log_rms_mae_db": 0.8241597849474137, "rms_envelope_nrmse": 0.08089452511266484, "envelope_correlation": 0.9939680619451456, "rms_gain_error_db": -0.050785615610477106, "reference_mean_power": 0.0028897508304992817, "output_mean_power": 0.0028561554162256893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 1.5709970652253345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007599193486798772, "output_power": 0.0007550475741622529, "reference_active": true, "delta_db": -0.027931858464077274}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014055064928452022, "output_power": 0.00013259323989113755, "reference_active": true, "delta_db": -0.2531147356000784}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3577241130570316e-05, "output_power": 2.359931605166252e-05, "reference_active": true, "delta_db": 0.004064314011493278}], "active_band_count": 3, "active_band_mae_db": 0.09503696935854965}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.0001938989159680862, "log_rms_mae_db": 0.37943156011893786, "rms_envelope_nrmse": 0.043838167734456854, "envelope_correlation": 0.9980983501539856, "rms_gain_error_db": -0.05078892074202089, "reference_mean_power": 0.002906120097080002, "output_mean_power": 0.0028723321924632294}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 1.4139945205650041e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006810740061577038, "output_power": 0.0006820108948248324, "reference_active": true, "delta_db": 0.005970070420005468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.5174236911387e-05, "output_power": 6.377768485079802e-05, "reference_active": true, "delta_db": -0.09407204787096805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3046113483012314e-06, "output_power": 1.2631888747575692e-06, "reference_active": true, "delta_db": -0.14012859890493115}], "active_band_count": 3, "active_band_mae_db": 0.08005690573196822}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013759847516455276, "log_rms_mae_db": 1.5570809964756103, "rms_envelope_nrmse": 0.11375398245641226, "envelope_correlation": 0.9892644119060378, "rms_gain_error_db": -0.016863566679200245, "reference_mean_power": 0.001146104301837711, "output_mean_power": 0.001141662631110387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 6.878091269334419e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031946504747131925, "output_power": 0.00031427705881288166, "reference_active": true, "delta_db": -0.07110669015522045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.126588772977069e-05, "output_power": 6.318317533114777e-05, "reference_active": true, "delta_db": -0.5228025133201515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2614620971957042e-05, "output_power": 1.6666638876753014e-05, "reference_active": true, "delta_db": 1.2097381955246875}], "active_band_count": 3, "active_band_mae_db": 0.6012157996666865}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00012449215623942515, "log_rms_mae_db": 0.8321332913607543, "rms_envelope_nrmse": 0.06921096320277062, "envelope_correlation": 0.995673015234016, "rms_gain_error_db": -0.016863710579410343, "reference_mean_power": 0.001152608697675057, "output_mean_power": 0.0011481417814394686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 6.232509709606391e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00029975403822873686, "output_power": 0.0002951192595260696, "reference_active": true, "delta_db": -0.06767489995751456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2561293827419036e-05, "output_power": 2.9138804445561286e-05, "reference_active": true, "delta_db": -0.48229924453552175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.743814177835147e-07, "output_power": 9.089714402059558e-07, "reference_active": true, "delta_db": 0.6959531494078562}], "active_band_count": 3, "active_band_mae_db": 0.4153090979669642}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013997823549865974, "log_rms_mae_db": 1.3602079370771332, "rms_envelope_nrmse": 0.12511714824572776, "envelope_correlation": 0.9883189928333512, "rms_gain_error_db": -0.11672777195469114, "reference_mean_power": 0.0010964326399058018, "output_mean_power": 0.0010673557117095787}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 8.778805813988002e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045662233680262583, "output_power": 0.00044854366373672956, "reference_active": true, "delta_db": -0.07752426174608551}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.737497372188522e-05, "output_power": 4.427158002900812e-05, "reference_active": true, "delta_db": -0.294239600870495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.406352772002978e-05, "output_power": 1.5213068603118732e-05, "reference_active": true, "delta_db": 0.34122550175739086}], "active_band_count": 3, "active_band_mae_db": 0.23766312145799043}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013340521153934576, "log_rms_mae_db": 0.6771889823377854, "rms_envelope_nrmse": 0.06699705707822338, "envelope_correlation": 0.9965450139105627, "rms_gain_error_db": -0.1167361724019595, "reference_mean_power": 0.0011026403148293462, "output_mean_power": 0.0010733966854833785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 8.248821386475387e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042363838931095744, "output_power": 0.0004178572605368552, "reference_active": true, "delta_db": -0.059673554518082025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.073594132211124e-05, "output_power": 2.099612558744813e-05, "reference_active": true, "delta_db": 0.054154065782605926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.366449484579509e-07, "output_power": 9.314069608825791e-07, "reference_active": true, "delta_db": 1.0188126429106252}], "active_band_count": 3, "active_band_mae_db": 0.3775467544037711}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00010989278407704884, "log_rms_mae_db": 1.5082767818997689, "rms_envelope_nrmse": 0.16169967084894318, "envelope_correlation": 0.9811745558981099, "rms_gain_error_db": -0.08117660312594852, "reference_mean_power": 0.0003075575386675287, "output_mean_power": 0.0003018621884938937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 2.696003640873154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014789902573029864, "output_power": 0.000136108969842222, "reference_active": true, "delta_db": -0.36078566149884006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.17770472077717e-05, "output_power": 3.6935350978919496e-05, "reference_active": true, "delta_db": 0.6532869110880035}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.120339140119136e-06, "output_power": 4.954536239295655e-06, "reference_active": true, "delta_db": 2.0080121062495553}], "active_band_count": 3, "active_band_mae_db": 1.007361559612133}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010174719078202204, "log_rms_mae_db": 0.7814066408516537, "rms_envelope_nrmse": 0.09024474272964578, "envelope_correlation": 0.9937728821824671, "rms_gain_error_db": -0.08118060316615949, "reference_mean_power": 0.000309290771080255, "output_mean_power": 0.0003035630453158096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.5256223542383704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014158879062551878, "output_power": 0.00013415705140497762, "reference_active": true, "delta_db": -0.23415367781812277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.501151157810192e-05, "output_power": 1.8083233739738234e-05, "reference_active": true, "delta_db": 0.8085167037338327}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.631467090177325e-07, "output_power": 3.257150554402572e-07, "reference_active": true, "delta_db": 0.9263989032127308}], "active_band_count": 3, "active_band_mae_db": 0.6563564282548954}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8632369522292529, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a1c2d25a589ac84b9d025a8d67dd408d7a208894c32b8ebbcc4873fdc89242d2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9858883905698278, "standard_deviation": 0.0036327813250554724, "minimum": 0.9792609177048902, "maximum": 0.9898829081183989, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.350408308238728, "standard_deviation": 0.03531446321839359, "minimum": 1.2954691857734075, "maximum": 1.409253786714553, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6430763626190874, "standard_deviation": 0.38930240601990357, "minimum": 0.23766312145799043, "maximum": 1.3676384101083763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1373308202096866, "standard_deviation": 0.017029021689970638, "minimum": 0.11752472311617337, "maximum": 0.16680980464187278, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955381742320587, "standard_deviation": 0.0015662210372880702, "minimum": 0.9935819892610943, "maximum": 0.9976017886573947, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7236229468175512, "standard_deviation": 0.034151702855341234, "minimum": 0.6762360486336857, "maximum": 0.7616501769128498, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7174704209740728, "standard_deviation": 0.43410911424540455, "minimum": 0.23815169299422156, "maximum": 1.4156365698373736, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07538605630277134, "standard_deviation": 0.013030079385458072, "minimum": 0.055673995488781006, "maximum": 0.09016506621790255, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9869170284701506, "standard_deviation": 0.002637484276049496, "minimum": 0.9811745558981099, "maximum": 0.9892454460558757, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5320744230027503, "standard_deviation": 0.020097374366558774, "minimum": 1.5044448910388397, "maximum": 1.5603048352618683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5106593617126753, "standard_deviation": 0.34228435494733317, "minimum": 0.15078833706557612, "maximum": 1.007361559612133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13548768773333314, "standard_deviation": 0.013119721841399376, "minimum": 0.12278485630403309, "maximum": 0.16169967084894318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9960481385581061, "standard_deviation": 0.0010378586843651481, "minimum": 0.9937728821824671, "maximum": 0.9971994363680764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8083913327736922, "standard_deviation": 0.022977828979302572, "minimum": 0.7814066408516537, "maximum": 0.84086835847893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4409399980697671, "standard_deviation": 0.25269307740689323, "minimum": 0.1937083404375637, "maximum": 0.8690378860926735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07346882509643285, "standard_deviation": 0.008791930839969877, "minimum": 0.06410358676442057, "maximum": 0.09024474272964578, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9887015287791172, "standard_deviation": 0.0012755677524580126, "minimum": 0.9863500877679885, "maximum": 0.9899397653133791, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.529378242190548, "standard_deviation": 0.02759307576528563, "minimum": 1.4890914073487127, "maximum": 1.5699618124642498, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6702402658565748, "standard_deviation": 0.25153855519416235, "minimum": 0.3325191016852194, "maximum": 1.0257810278424946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1170927081430124, "standard_deviation": 0.006354738850302474, "minimum": 0.11056886935695429, "maximum": 0.129282344531878, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961576907503089, "standard_deviation": 0.000949798455296794, "minimum": 0.9949088557718971, "maximum": 0.9980410155738303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8738782609767834, "standard_deviation": 0.031018465907198606, "minimum": 0.823988855387626, "maximum": 0.9117047727739068, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5818997478049828, "standard_deviation": 0.2537094201678763, "minimum": 0.3271459094934165, "maximum": 0.9852769165653702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06606022359974906, "standard_deviation": 0.009075094992348, "minimum": 0.047462594405781426, "maximum": 0.07594086802664511, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.741219140848888, "standard_deviation": 0.06199653648461578, "minimum": 0.6541895187712549, "maximum": 0.8280699081652934, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7039012819697685, "standard_deviation": 0.009368015917281307, "minimum": 5.690409073565529, "maximum": 5.716295215777005, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5773397321776863, "standard_deviation": 0.029538506829398868, "minimum": 0.533941509981277, "maximum": 0.6172037022044785, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04722708334120792, "standard_deviation": 0.08356817293125668, "minimum": -0.15939495025547118, "maximum": 0.07728976335523316, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930430064734259, "standard_deviation": 0.0014809100685030056, "minimum": 0.9909032927803212, "maximum": 0.99520707053044, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8305829932820666, "standard_deviation": 0.021135191308646172, "minimum": 0.8124278480964942, "maximum": 0.8790579671441623, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37452642732613567, "standard_deviation": 0.21542701116522964, "minimum": 0.09503696935854965, "maximum": 0.7290115940097865, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08697184594631495, "standard_deviation": 0.009000620307456617, "minimum": 0.07391715796893339, "maximum": 0.10013762299908066, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980388627508954, "standard_deviation": 0.0006129567531690173, "minimum": 0.9970598427997681, "maximum": 0.9988805057569424, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41127404107372956, "standard_deviation": 0.026816282198323597, "minimum": 0.37943156011893786, "maximum": 0.4475378774586686, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3154818971820001, "standard_deviation": 0.17147972701174863, "minimum": 0.08005690573196822, "maximum": 0.6192073763445504, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04521054853987681, "standard_deviation": 0.005956596377428251, "minimum": 0.03725038920545283, "maximum": 0.05423601744031444, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141083-0010", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.150810956954956, "predicted_mos": 2.150810956954956, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "ab19eab0bbb440affbaf84b58c9d0543fb575dbe7ac0351ec561f4172bb6db3c", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5705549996346235, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8232536111531567, "全帯域_10ms_rms_envelope_nrmse": 0.08001118778828419, "全帯域_10ms_envelope_correlation": 0.994097778079297, "全帯域_10ms_modulation_mae_db": 0.1718720039869672, "全帯域_40ms_log_rms_mae_db": 0.3864628522289545, "全帯域_40ms_rms_envelope_nrmse": 0.043180945617546276, "全帯域_40ms_envelope_correlation": 0.9981515973480446, "全帯域_40ms_modulation_mae_db": 0.125383809021375, "300–1000 Hz_10ms_log_rms_mae_db": 1.5321369369297881, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10957381896955737, "300–1000 Hz_10ms_envelope_correlation": 0.9900422918254502, "300–1000 Hz_10ms_modulation_mae_db": 0.6426947761940406, "300–1000 Hz_40ms_log_rms_mae_db": 0.8051644508140041, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0666314368341558, "300–1000 Hz_40ms_envelope_correlation": 0.9959889295436971, "300–1000 Hz_40ms_modulation_mae_db": 0.4405403645359139, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3760531439536197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12132937778877995, "1000–3000 Hz_10ms_envelope_correlation": 0.9890153411527326, "1000–3000 Hz_10ms_modulation_mae_db": 0.15813944250554457, "1000–3000 Hz_40ms_log_rms_mae_db": 0.681026124890677, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0624597224750308, "1000–3000 Hz_40ms_envelope_correlation": 0.9969971009664848, "1000–3000 Hz_40ms_modulation_mae_db": 0.2820224467898814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5081264050242367, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1524954785801399, "3000–6000 Hz_10ms_envelope_correlation": 0.9832938031818459, "3000–6000 Hz_10ms_modulation_mae_db": 0.8564658820520464, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7746361748781057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08146529899781652, "3000–6000 Hz_40ms_envelope_correlation": 0.9950003772905354, "3000–6000 Hz_40ms_modulation_mae_db": 0.5637055063246031, "mr_stft_db": 5.667604496443151, "band_mae_db": 0.5565047961755282, "rms_gain_error_db": -0.04670146302424408, "output_peak": 0.6012393720951974}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8091844717564369, "全帯域_10ms_rms_envelope_nrmse": 0.08375967728803624, "全帯域_10ms_envelope_correlation": 0.9935859928002434, "全帯域_10ms_modulation_mae_db": 0.27172850433513923, "全帯域_40ms_log_rms_mae_db": 0.4145550570688355, "全帯域_40ms_rms_envelope_nrmse": 0.04532781856028719, "全帯域_40ms_envelope_correlation": 0.9979846126640739, "全帯域_40ms_modulation_mae_db": 0.24513747364391858, "300–1000 Hz_10ms_log_rms_mae_db": 1.4778005509335337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10923083645627886, "300–1000 Hz_10ms_envelope_correlation": 0.990125466529256, "300–1000 Hz_10ms_modulation_mae_db": 0.8347187408973514, "300–1000 Hz_40ms_log_rms_mae_db": 0.7784381933755247, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056634641474362084, "300–1000 Hz_40ms_envelope_correlation": 0.9971081683981382, "300–1000 Hz_40ms_modulation_mae_db": 0.7327062105188139, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3856601101615904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13383468205605992, "1000–3000 Hz_10ms_envelope_correlation": 0.9867733766150251, "1000–3000 Hz_10ms_modulation_mae_db": 0.5152139030081194, "1000–3000 Hz_40ms_log_rms_mae_db": 0.750334821523688, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07394132795150034, "1000–3000 Hz_40ms_envelope_correlation": 0.9957387008997506, "1000–3000 Hz_40ms_modulation_mae_db": 0.5703279811831055, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5275414888955723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1229890255614555, "3000–6000 Hz_10ms_envelope_correlation": 0.989188904664665, "3000–6000 Hz_10ms_modulation_mae_db": 0.06788405673549323, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7984650559058936, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06586696112014112, "3000–6000 Hz_40ms_envelope_correlation": 0.9967182406207898, "3000–6000 Hz_40ms_modulation_mae_db": 0.159770648009971, "mr_stft_db": 5.65172239885166, "band_mae_db": 0.6783509370992694, "rms_gain_error_db": 0.028886452192670713, "output_peak": 0.5451617019563091}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8372684552139436, "全帯域_10ms_rms_envelope_nrmse": 0.09544262109138772, "全帯域_10ms_envelope_correlation": 0.9916172831154072, "全帯域_10ms_modulation_mae_db": 0.16769820331230456, "全帯域_40ms_log_rms_mae_db": 0.4524889409127306, "全帯域_40ms_rms_envelope_nrmse": 0.056062517939954405, "全帯域_40ms_envelope_correlation": 0.9968652343081469, "全帯域_40ms_modulation_mae_db": 0.348879750713593, "300–1000 Hz_10ms_log_rms_mae_db": 1.5272225030056112, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1137845365189512, "300–1000 Hz_10ms_envelope_correlation": 0.989248669488264, "300–1000 Hz_10ms_modulation_mae_db": 0.277480860208509, "300–1000 Hz_40ms_log_rms_mae_db": 0.7985850792194293, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061923410479877386, "300–1000 Hz_40ms_envelope_correlation": 0.9965427651511739, "300–1000 Hz_40ms_modulation_mae_db": 0.26478789288542165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3672001730932857, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1526490576430119, "1000–3000 Hz_10ms_envelope_correlation": 0.9827978781286066, "1000–3000 Hz_10ms_modulation_mae_db": 0.43039171147252536, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7468745967525166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0931804281949483, "1000–3000 Hz_40ms_envelope_correlation": 0.993216342571509, "1000–3000 Hz_40ms_modulation_mae_db": 0.5545139267249449, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5616706264169917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.138615606673332, "3000–6000 Hz_10ms_envelope_correlation": 0.9864618236177838, "3000–6000 Hz_10ms_modulation_mae_db": 0.891742070717223, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8239156540245193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07337272662239981, "3000–6000 Hz_40ms_envelope_correlation": 0.9960341333210674, "3000–6000 Hz_40ms_modulation_mae_db": 0.950107372398112, "mr_stft_db": 5.667384301109808, "band_mae_db": 0.568547056869684, "rms_gain_error_db": -0.0028749827095874113, "output_peak": 0.563297686556644}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.802533759277757, "全帯域_10ms_rms_envelope_nrmse": 0.09013721800419813, "全帯域_10ms_envelope_correlation": 0.9926655193764643, "全帯域_10ms_modulation_mae_db": 0.576102583744888, "全帯域_40ms_log_rms_mae_db": 0.3817529598771013, "全帯域_40ms_rms_envelope_nrmse": 0.04684577877187666, "全帯域_40ms_envelope_correlation": 0.9979046249681561, "全帯域_40ms_modulation_mae_db": 0.5175193707456299, "300–1000 Hz_10ms_log_rms_mae_db": 1.514089022931293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10993640887829954, "300–1000 Hz_10ms_envelope_correlation": 0.9899721688028227, "300–1000 Hz_10ms_modulation_mae_db": 0.7318088635692163, "300–1000 Hz_40ms_log_rms_mae_db": 0.7885253375907555, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04832760671057665, "300–1000 Hz_40ms_envelope_correlation": 0.9979696943833914, "300–1000 Hz_40ms_modulation_mae_db": 0.4685077463785558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3403105255251468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14474026381304347, "1000–3000 Hz_10ms_envelope_correlation": 0.9852698963874756, "1000–3000 Hz_10ms_modulation_mae_db": 1.4056544573850467, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7077770480548494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08536256942168027, "1000–3000 Hz_40ms_envelope_correlation": 0.9948422973588471, "1000–3000 Hz_40ms_modulation_mae_db": 1.4783262268525081, "3000–6000 Hz_10ms_log_rms_mae_db": 1.574493865857512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14251766604815413, "3000–6000 Hz_10ms_envelope_correlation": 0.9864064567309682, "3000–6000 Hz_10ms_modulation_mae_db": 0.7254098836770693, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8314319127862927, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08252127729441396, "3000–6000 Hz_40ms_envelope_correlation": 0.9956116508595069, "3000–6000 Hz_40ms_modulation_mae_db": 0.4388204623642173, "mr_stft_db": 5.650954123222299, "band_mae_db": 0.6966526601824327, "rms_gain_error_db": 0.071487091049672, "output_peak": 0.5836007871784096}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8233108186400074, "全帯域_10ms_rms_envelope_nrmse": 0.07394578445060629, "全帯域_10ms_envelope_correlation": 0.995267153875375, "全帯域_10ms_modulation_mae_db": 0.3695062970726619, "全帯域_40ms_log_rms_mae_db": 0.39769258562710885, "全帯域_40ms_rms_envelope_nrmse": 0.039723690072836146, "全帯域_40ms_envelope_correlation": 0.9988292049179052, "全帯域_40ms_modulation_mae_db": 0.24623389375968396, "300–1000 Hz_10ms_log_rms_mae_db": 1.452991769213025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11386662308723089, "300–1000 Hz_10ms_envelope_correlation": 0.9892944781399876, "300–1000 Hz_10ms_modulation_mae_db": 0.3119618884369157, "300–1000 Hz_40ms_log_rms_mae_db": 0.8269335033685665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06668367368639283, "300–1000 Hz_40ms_envelope_correlation": 0.9961630848609712, "300–1000 Hz_40ms_modulation_mae_db": 0.2749519714650364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.384029190123281, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11663670151084773, "1000–3000 Hz_10ms_envelope_correlation": 0.9901200092327473, "1000–3000 Hz_10ms_modulation_mae_db": 0.3940045576180779, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7317968719735256, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06326061870053105, "1000–3000 Hz_40ms_envelope_correlation": 0.9973209067306837, "1000–3000 Hz_40ms_modulation_mae_db": 0.3447297896251371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5878936449852956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12407537715429134, "3000–6000 Hz_10ms_envelope_correlation": 0.989302947692484, "3000–6000 Hz_10ms_modulation_mae_db": 0.2884776023095548, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8211371382739404, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06856212562206015, "3000–6000 Hz_40ms_envelope_correlation": 0.9969943531137431, "3000–6000 Hz_40ms_modulation_mae_db": 0.23997952226919236, "mr_stft_db": 5.6689874753094385, "band_mae_db": 0.7290510810803147, "rms_gain_error_db": -0.1671835992493601, "output_peak": 0.5420066383168408}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8314686608629682, "全帯域_10ms_rms_envelope_nrmse": 0.0804363551036058, "全帯域_10ms_envelope_correlation": 0.9940564358513507, "全帯域_10ms_modulation_mae_db": 0.4736793865007063, "全帯域_40ms_log_rms_mae_db": 0.3937243832103015, "全帯域_40ms_rms_envelope_nrmse": 0.037232064203986746, "全帯域_40ms_envelope_correlation": 0.9986721742201456, "全帯域_40ms_modulation_mae_db": 0.299746402858741, "300–1000 Hz_10ms_log_rms_mae_db": 1.455378137791409, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11985401643963257, "300–1000 Hz_10ms_envelope_correlation": 0.9880653171094373, "300–1000 Hz_10ms_modulation_mae_db": 0.46755185806534083, "300–1000 Hz_40ms_log_rms_mae_db": 0.7894108662083986, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06902519080654845, "300–1000 Hz_40ms_envelope_correlation": 0.9957572283416313, "300–1000 Hz_40ms_modulation_mae_db": 0.30460191392074637, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3201832705326246, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1250724762510278, "1000–3000 Hz_10ms_envelope_correlation": 0.9883077385169599, "1000–3000 Hz_10ms_modulation_mae_db": 0.4158540509956114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7194182340141163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059194842623641775, "1000–3000 Hz_40ms_envelope_correlation": 0.9972451758497536, "1000–3000 Hz_40ms_modulation_mae_db": 0.2729248806907736, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5494587999250182, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12582434505180856, "3000–6000 Hz_10ms_envelope_correlation": 0.9886707943595899, "3000–6000 Hz_10ms_modulation_mae_db": 0.38466511357921324, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8170726644229772, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06502244547030375, "3000–6000 Hz_40ms_envelope_correlation": 0.9969063452758792, "3000–6000 Hz_40ms_modulation_mae_db": 0.47856866967527445, "mr_stft_db": 5.6613932551885595, "band_mae_db": 0.6126667562402434, "rms_gain_error_db": -0.07084662682800764, "output_peak": 0.6058625866462582}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8848377674688844, "全帯域_10ms_rms_envelope_nrmse": 0.09989587457359052, "全帯域_10ms_envelope_correlation": 0.9909353651057577, "全帯域_10ms_modulation_mae_db": 0.4038289745699715, "全帯域_40ms_log_rms_mae_db": 0.4548060511415375, "全帯域_40ms_rms_envelope_nrmse": 0.04774584347522013, "全帯域_40ms_envelope_correlation": 0.9980154247056839, "全帯域_40ms_modulation_mae_db": 0.20280890568488874, "300–1000 Hz_10ms_log_rms_mae_db": 1.5337239580253776, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11058411267884517, "300–1000 Hz_10ms_envelope_correlation": 0.9904067465123492, "300–1000 Hz_10ms_modulation_mae_db": 0.7305545477422108, "300–1000 Hz_40ms_log_rms_mae_db": 0.8636800287051453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07288710112615239, "300–1000 Hz_40ms_envelope_correlation": 0.9959870913094108, "300–1000 Hz_40ms_modulation_mae_db": 0.7704264979534642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.375298611501498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16850005969757026, "1000–3000 Hz_10ms_envelope_correlation": 0.9788680825933753, "1000–3000 Hz_10ms_modulation_mae_db": 0.8298530061384487, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7502352744451225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08746011933869915, "1000–3000 Hz_40ms_envelope_correlation": 0.9939840598608709, "1000–3000 Hz_40ms_modulation_mae_db": 1.0159529529086464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.536074904656162, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14036215786041484, "3000–6000 Hz_10ms_envelope_correlation": 0.9860007269017969, "3000–6000 Hz_10ms_modulation_mae_db": 0.18441419329003558, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8309075040252079, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0752071896585875, "3000–6000 Hz_40ms_envelope_correlation": 0.9960600942936849, "3000–6000 Hz_40ms_modulation_mae_db": 0.4700866105417334, "mr_stft_db": 5.678587934767559, "band_mae_db": 0.7168604416357138, "rms_gain_error_db": -0.15584278038956226, "output_peak": 0.564231826439066}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8437057686482647, "全帯域_10ms_rms_envelope_nrmse": 0.10016398060524319, "全帯域_10ms_envelope_correlation": 0.9907471074788958, "全帯域_10ms_modulation_mae_db": 0.7153569474885663, "全帯域_40ms_log_rms_mae_db": 0.44329639631425666, "全帯域_40ms_rms_envelope_nrmse": 0.05474602763707359, "全帯域_40ms_envelope_correlation": 0.9969963964845833, "全帯域_40ms_modulation_mae_db": 0.5339495582721328, "300–1000 Hz_10ms_log_rms_mae_db": 1.4851143269147002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13280821080146682, "300–1000 Hz_10ms_envelope_correlation": 0.9856362140860606, "300–1000 Hz_10ms_modulation_mae_db": 1.0595008839074602, "300–1000 Hz_40ms_log_rms_mae_db": 0.8355888250680996, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0774046916643837, "300–1000 Hz_40ms_envelope_correlation": 0.9947416626716808, "300–1000 Hz_40ms_modulation_mae_db": 0.9437326254969305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.361166703069959, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15746320373299677, "1000–3000 Hz_10ms_envelope_correlation": 0.9815384945242279, "1000–3000 Hz_10ms_modulation_mae_db": 0.9512393381677583, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7155309353331423, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09441622222629006, "1000–3000 Hz_40ms_envelope_correlation": 0.9929796871722333, "1000–3000 Hz_40ms_modulation_mae_db": 1.1265477840109803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5679491382495925, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12063734336183025, "3000–6000 Hz_10ms_envelope_correlation": 0.9895983416354086, "3000–6000 Hz_10ms_modulation_mae_db": 0.49455460687312364, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8529394919605776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06703684378035976, "3000–6000 Hz_40ms_envelope_correlation": 0.996692144252418, "3000–6000 Hz_40ms_modulation_mae_db": 0.3790053814706616, "mr_stft_db": 5.659615779371766, "band_mae_db": 0.5456083323586993, "rms_gain_error_db": -0.016957179697039557, "output_peak": 0.6106022121957072}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85120, "output_samples": 85120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.465945219098326, "spectral_convergence": 0.3485159406737208, "spectral_convergence_status": "ok", "amplitude_floor": 1.1374446115550809e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.672968121500871, "spectral_convergence": 0.372874604496647, "spectral_convergence_status": "ok", "amplitude_floor": 9.193083193108671e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8639001487302576, "spectral_convergence": 0.3766022516785383, "spectral_convergence_status": "ok", "amplitude_floor": 7.3161107874683e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.1849686170678433e-06, "output_power": 4.3101931393003295e-06, "delta_db": 2.9505152775243157, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004950757600852583, "output_power": 0.0004918470716442954, "delta_db": -0.028415728788670408, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008267786751374999, "output_power": 0.0008346237509423288, "delta_db": 0.0410147282970414, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007700805582515175, "output_power": 0.0007569268709249225, "delta_db": -0.07482236187751709, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006256626820185013, "output_power": 0.0006049219378705573, "delta_db": -0.14640917273374282, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00016725866300063496, "output_power": 0.00016353227182554106, "delta_db": -0.0978515078318812, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5565047961755282, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04670146302424408, "pitch": null, "output_rms_dbfs": -25.442171485935752, "output_peak": 0.6012393720951974, "reference_rms_dbfs": -25.39547002291151, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.667604496443151}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.0002036923290800657, "log_rms_mae_db": 0.8232536111531567, "rms_envelope_nrmse": 0.08001118778828419, "envelope_correlation": 0.994097778079297, "rms_gain_error_db": -0.04670012890116329, "reference_mean_power": 0.0028897508304992817, "output_mean_power": 0.002858843517437918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 1.5709970652253345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007599193486798772, "output_power": 0.0007538927341657024, "reference_active": true, "delta_db": -0.034579447280386914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014055064928452022, "output_power": 0.0001309376527011727, "reference_active": true, "delta_db": -0.3076830507832259}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3577241130570316e-05, "output_power": 2.4537388353487175e-05, "reference_active": true, "delta_db": 0.17335351389728876}], "active_band_count": 3, "active_band_mae_db": 0.1718720039869672}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.0001938989159680862, "log_rms_mae_db": 0.3864628522289545, "rms_envelope_nrmse": 0.043180945617546276, "envelope_correlation": 0.9981515973480446, "rms_gain_error_db": -0.04670312421435878, "reference_mean_power": 0.002906120097080002, "output_mean_power": 0.0028750357237215854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 1.4139945205650041e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006810740061577038, "output_power": 0.0006798740883427808, "reference_active": true, "delta_db": -0.0076581586453687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.5174236911387e-05, "output_power": 6.243527536486166e-05, "reference_active": true, "delta_db": -0.1864592348711973}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3046113483012314e-06, "output_power": 1.3604561721846e-06, "reference_active": true, "delta_db": 0.182034033547559}], "active_band_count": 3, "active_band_mae_db": 0.125383809021375}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013759847516455276, "log_rms_mae_db": 1.5321369369297881, "rms_envelope_nrmse": 0.10957381896955737, "envelope_correlation": 0.9900422918254502, "rms_gain_error_db": -0.01526306034623971, "reference_mean_power": 0.001146104301837711, "output_mean_power": 0.0011420834458075475}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 6.878091269334419e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031946504747131925, "output_power": 0.00031726829019577263, "reference_active": true, "delta_db": -0.029966810018748083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.126588772977069e-05, "output_power": 6.405572522766598e-05, "reference_active": true, "delta_db": -0.46323746774831537}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2614620971957042e-05, "output_power": 1.7553438863304695e-05, "reference_active": true, "delta_db": 1.4348800508150583}], "active_band_count": 3, "active_band_mae_db": 0.6426947761940406}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00012449215623942515, "log_rms_mae_db": 0.8051644508140041, "rms_envelope_nrmse": 0.0666314368341558, "envelope_correlation": 0.9959889295436971, "rms_gain_error_db": -0.015263171724069014, "reference_mean_power": 0.001152608697675057, "output_mean_power": 0.0011485649929403384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 6.232509709606391e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00029975403822873686, "output_power": 0.0002983353788351841, "reference_active": true, "delta_db": -0.02060284245911311}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2561293827419036e-05, "output_power": 2.9515127729320977e-05, "reference_active": true, "delta_db": -0.42656986301072086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.743814177835147e-07, "output_power": 9.471085600454707e-07, "reference_active": true, "delta_db": 0.8744483881379078}], "active_band_count": 3, "active_band_mae_db": 0.4405403645359139}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013997823549865974, "log_rms_mae_db": 1.3760531439536197, "rms_envelope_nrmse": 0.12132937778877995, "envelope_correlation": 0.9890153411527326, "rms_gain_error_db": -0.10570100511923795, "reference_mean_power": 0.0010964326399058018, "output_mean_power": 0.0010700691785176346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 8.778805813988002e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045662233680262583, "output_power": 0.0004472776694824455, "reference_active": true, "delta_db": -0.08979935867440769}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.737497372188522e-05, "output_power": 4.373424597245302e-05, "reference_active": true, "delta_db": -0.34727338879192626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.406352772002978e-05, "output_power": 1.4184983388494894e-05, "reference_active": true, "delta_db": 0.03734558005029976}], "active_band_count": 3, "active_band_mae_db": 0.15813944250554457}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013340521153934576, "log_rms_mae_db": 0.681026124890677, "rms_envelope_nrmse": 0.0624597224750308, "envelope_correlation": 0.9969971009664848, "rms_gain_error_db": -0.10570842923352837, "reference_mean_power": 0.0011026403148293462, "output_mean_power": 0.0010761257517764213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 8.248821386475387e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042363838931095744, "output_power": 0.00041636895241765805, "reference_active": true, "delta_db": -0.07516970362646594}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.073594132211124e-05, "output_power": 2.04187663009909e-05, "reference_active": true, "delta_db": -0.0669425670830641}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.366449484579509e-07, "output_power": 8.662708925797946e-07, "reference_active": true, "delta_db": 0.7039550696601142}], "active_band_count": 3, "active_band_mae_db": 0.2820224467898814}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00010989278407704884, "log_rms_mae_db": 1.5081264050242367, "rms_envelope_nrmse": 0.1524954785801399, "envelope_correlation": 0.9832938031818459, "rms_gain_error_db": -0.10697219424044703, "reference_mean_power": 0.0003075575386675287, "output_mean_power": 0.0003000745456767516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 2.696003640873154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014789902573029864, "output_power": 0.00013833026258562472, "reference_active": true, "delta_db": -0.29048111795853215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.17770472077717e-05, "output_power": 3.47696410174462e-05, "reference_active": true, "delta_db": 0.39086667630711247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.120339140119136e-06, "output_power": 4.819553259661623e-06, "reference_active": true, "delta_db": 1.8880498518904945}], "active_band_count": 3, "active_band_mae_db": 0.8564658820520464}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010174719078202204, "log_rms_mae_db": 0.7746361748781057, "rms_envelope_nrmse": 0.08146529899781652, "envelope_correlation": 0.9950003772905354, "rms_gain_error_db": -0.10697728607572536, "reference_mean_power": 0.000309290771080255, "output_mean_power": 0.00030176525407808844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.5256223542383704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014158879062551878, "output_power": 0.0001357651427516912, "reference_active": true, "delta_db": -0.1824059167081883}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.501151157810192e-05, "output_power": 1.6903835615605634e-05, "reference_active": true, "delta_db": 0.5156083513707352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.631467090177325e-07, "output_power": 3.3075633276274175e-07, "reference_active": true, "delta_db": 0.993102250894886}], "active_band_count": 3, "active_band_mae_db": 0.5637055063246031}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8674267088357227, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e2b6b1d992b3cba7c47a4c73bb528851c5aa79610d18b51e67f2221aafcc9893", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9853363521438938, "standard_deviation": 0.003964539457145344, "minimum": 0.9788680825933753, "maximum": 0.9901200092327473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3637377159951256, "standard_deviation": 0.022810026142886564, "minimum": 1.3201832705326246, "maximum": 1.3856601101615904, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6375438084113916, "standard_deviation": 0.4002757157399755, "minimum": 0.15813944250554457, "maximum": 1.4056544573850467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1400282278116672, "standard_deviation": 0.018727204176419862, "minimum": 0.11663670151084773, "maximum": 0.16850005969757026, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9952905339262667, "standard_deviation": 0.0017970744667594473, "minimum": 0.9929796871722333, "maximum": 0.9973209067306837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7253742383734547, "standard_deviation": 0.024358625160029502, "minimum": 0.681026124890677, "maximum": 0.750334821523688, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7056682485982472, "standard_deviation": 0.44858512624963576, "minimum": 0.2729248806907736, "maximum": 1.4783262268525081, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07740948136654022, "standard_deviation": 0.01448992600836761, "minimum": 0.059194842623641775, "maximum": 0.09441622222629006, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9873654748480678, "standard_deviation": 0.002202128802686814, "minimum": 0.9832938031818459, "maximum": 0.9895983416354086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5516511092512975, "standard_deviation": 0.026523379015464112, "minimum": 1.5081264050242367, "maximum": 1.5878936449852956, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4867016761542199, "standard_deviation": 0.31037229493356233, "minimum": 0.06788405673549323, "maximum": 0.891742070717223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1334396250364283, "standard_deviation": 0.011582275267463055, "minimum": 0.12063734336183025, "maximum": 0.1524954785801399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.996252167378453, "standard_deviation": 0.000702166347743029, "minimum": 0.9950003772905354, "maximum": 0.9969943531137431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8188131995346892, "standard_deviation": 0.0235136119002802, "minimum": 0.7746361748781057, "maximum": 0.8529394919605776, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.46000552163172065, "standard_deviation": 0.23813620920607326, "minimum": 0.159770648009971, "maximum": 0.950107372398112, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07238185857076032, "standard_deviation": 0.006908022356871983, "minimum": 0.06502244547030375, "maximum": 0.08252127729441396, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9890989190617034, "standard_deviation": 0.0015827668772017793, "minimum": 0.9856362140860606, "maximum": 0.9904067465123492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4973071507180924, "standard_deviation": 0.0337362140095267, "minimum": 1.452991769213025, "maximum": 1.5337239580253776, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6320340523776306, "standard_deviation": 0.26693832231937253, "minimum": 0.277480860208509, "maximum": 1.0595008839074602, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1149548204787828, "standard_deviation": 0.008029464643095107, "minimum": 0.10923083645627886, "maximum": 0.13280821080146682, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962823280825118, "standard_deviation": 0.0009587112460894197, "minimum": 0.9947416626716808, "maximum": 0.9979696943833914, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8107907855437405, "standard_deviation": 0.028916081531394827, "minimum": 0.7784381933755247, "maximum": 0.8636800287051453, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5250319028943604, "standard_deviation": 0.25852535594117193, "minimum": 0.26478789288542165, "maximum": 0.9437326254969305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06493971909780616, "standard_deviation": 0.00922082656471165, "minimum": 0.04832760671057665, "maximum": 0.0774046916643837, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.6380302577052357, "standard_deviation": 0.07580121278024901, "minimum": 0.5456083323586993, "maximum": 0.7290510810803147, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6632812205330305, "standard_deviation": 0.009296284879025223, "minimum": 5.650954123222299, "maximum": 5.678587934767559, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5770003514230541, "standard_deviation": 0.02720889391436872, "minimum": 0.5420066383168408, "maximum": 0.6106022121957072, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04500413608193229, "standard_deviation": 0.08407077282101992, "minimum": -0.1671835992493601, "maximum": 0.071487091049672, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9928715794603489, "standard_deviation": 0.0016490100157611287, "minimum": 0.9907471074788958, "maximum": 0.995267153875375, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8319454141276773, "standard_deviation": 0.02534808715613139, "minimum": 0.802533759277757, "maximum": 0.8848377674688844, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3937216126264006, "standard_deviation": 0.19234642928683973, "minimum": 0.16769820331230456, "maximum": 0.7153569474885663, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08797408736311901, "standard_deviation": 0.009899208842182376, "minimum": 0.07394578445060629, "maximum": 0.10016398060524319, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979274087020924, "standard_deviation": 0.0006994598979315297, "minimum": 0.9968652343081469, "maximum": 0.9988292049179052, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4155974032976033, "standard_deviation": 0.03037833869510986, "minimum": 0.3817529598771013, "maximum": 0.4548060511415375, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31495739558749536, "standard_deviation": 0.1456434939398336, "minimum": 0.125383809021375, "maximum": 0.5339495582721328, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04635808578484764, "standard_deviation": 0.006602515044400642, "minimum": 0.037232064203986746, "maximum": 0.056062517939954405, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141083-0010", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.0962932109832764, "predicted_mos": 2.0962932109832764, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "b98052bc6ed40e9bc2e110fd55d6d7a42de0939eb4036cbf2a3844468cd1b6b7", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6058416003361344, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7757918731859916, "全帯域_10ms_rms_envelope_nrmse": 0.07431443206417887, "全帯域_10ms_envelope_correlation": 0.9949524711304688, "全帯域_10ms_modulation_mae_db": 0.16045811894094011, "全帯域_40ms_log_rms_mae_db": 0.39266046179122016, "全帯域_40ms_rms_envelope_nrmse": 0.03589342924962537, "全帯域_40ms_envelope_correlation": 0.9987938989634922, "全帯域_40ms_modulation_mae_db": 0.10433573348797769, "300–1000 Hz_10ms_log_rms_mae_db": 1.786181278630355, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11585963543924115, "300–1000 Hz_10ms_envelope_correlation": 0.9894010940204166, "300–1000 Hz_10ms_modulation_mae_db": 0.8914891389852682, "300–1000 Hz_40ms_log_rms_mae_db": 1.1425532695247083, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07123426560413583, "300–1000 Hz_40ms_envelope_correlation": 0.995826205572138, "300–1000 Hz_40ms_modulation_mae_db": 0.7602292973219525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4591054623545936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1547920604918512, "1000–3000 Hz_10ms_envelope_correlation": 0.9826534385105411, "1000–3000 Hz_10ms_modulation_mae_db": 0.7279265900480704, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8441101570446493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09556017622299896, "1000–3000 Hz_40ms_envelope_correlation": 0.9938269371757146, "1000–3000 Hz_40ms_modulation_mae_db": 0.6529119358386328, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8229964225071074, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13905723622174934, "3000–6000 Hz_10ms_envelope_correlation": 0.9863667471561468, "3000–6000 Hz_10ms_modulation_mae_db": 0.5151708834811345, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9405444416918781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08301471766329012, "3000–6000 Hz_40ms_envelope_correlation": 0.9951734254613943, "3000–6000 Hz_40ms_modulation_mae_db": 0.9627737477092264, "mr_stft_db": 5.993328950575317, "band_mae_db": 0.9074060953462921, "rms_gain_error_db": -0.08529313717857144, "output_peak": 0.5610607867684839}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8030183043539805, "全帯域_10ms_rms_envelope_nrmse": 0.07820102494895909, "全帯域_10ms_envelope_correlation": 0.9945201501597596, "全帯域_10ms_modulation_mae_db": 0.34621770356336085, "全帯域_40ms_log_rms_mae_db": 0.4388627071527473, "全帯域_40ms_rms_envelope_nrmse": 0.04988848873845576, "全帯域_40ms_envelope_correlation": 0.997731691413319, "全帯域_40ms_modulation_mae_db": 0.5229531587817452, "300–1000 Hz_10ms_log_rms_mae_db": 1.7544719993668925, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10530213633429188, "300–1000 Hz_10ms_envelope_correlation": 0.9910943980677601, "300–1000 Hz_10ms_modulation_mae_db": 0.51302037989202, "300–1000 Hz_40ms_log_rms_mae_db": 1.137658111287793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06286110227328925, "300–1000 Hz_40ms_envelope_correlation": 0.9969074081961024, "300–1000 Hz_40ms_modulation_mae_db": 0.4110152924813879, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4203968943196907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1232325668874551, "1000–3000 Hz_10ms_envelope_correlation": 0.9888015055379502, "1000–3000 Hz_10ms_modulation_mae_db": 0.6172637960642199, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7980572903823965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0737320860850328, "1000–3000 Hz_40ms_envelope_correlation": 0.9959943771306567, "1000–3000 Hz_40ms_modulation_mae_db": 0.8403230312160691, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8399192947219933, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14976271485084633, "3000–6000 Hz_10ms_envelope_correlation": 0.9844493578724823, "3000–6000 Hz_10ms_modulation_mae_db": 0.5561733005958434, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9392584506265818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09219975066825642, "3000–6000 Hz_40ms_envelope_correlation": 0.9938643537255072, "3000–6000 Hz_40ms_modulation_mae_db": 0.6252881994788272, "mr_stft_db": 6.003353941836792, "band_mae_db": 0.9887682767204039, "rms_gain_error_db": -0.13692952437395592, "output_peak": 0.6066134521363757}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8149962818895548, "全帯域_10ms_rms_envelope_nrmse": 0.08061497461382348, "全帯域_10ms_envelope_correlation": 0.9942786718832608, "全帯域_10ms_modulation_mae_db": 0.6282183154567976, "全帯域_40ms_log_rms_mae_db": 0.43220984841740384, "全帯域_40ms_rms_envelope_nrmse": 0.04736419633385353, "全帯域_40ms_envelope_correlation": 0.9981286075494846, "全帯域_40ms_modulation_mae_db": 0.6578693529521337, "300–1000 Hz_10ms_log_rms_mae_db": 1.849619303813306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1110775403323232, "300–1000 Hz_10ms_envelope_correlation": 0.9898352128363849, "300–1000 Hz_10ms_modulation_mae_db": 1.1907748429938925, "300–1000 Hz_40ms_log_rms_mae_db": 1.1972939101431843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055287518169628465, "300–1000 Hz_40ms_envelope_correlation": 0.9974374734950323, "300–1000 Hz_40ms_modulation_mae_db": 1.0244027360766104, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4995639496420345, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13335246433245204, "1000–3000 Hz_10ms_envelope_correlation": 0.9873367883945827, "1000–3000 Hz_10ms_modulation_mae_db": 1.4535665207458892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8159753608058761, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08844728808661834, "1000–3000 Hz_40ms_envelope_correlation": 0.9947464457688849, "1000–3000 Hz_40ms_modulation_mae_db": 1.178723668676708, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8824607850983466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14590945644219847, "3000–6000 Hz_10ms_envelope_correlation": 0.9847247141138984, "3000–6000 Hz_10ms_modulation_mae_db": 0.22569291881227752, "3000–6000 Hz_40ms_log_rms_mae_db": 0.962960870135504, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08467256822061038, "3000–6000 Hz_40ms_envelope_correlation": 0.9945055841904665, "3000–6000 Hz_40ms_modulation_mae_db": 0.6191004754257549, "mr_stft_db": 6.000183423896215, "band_mae_db": 0.936721799207716, "rms_gain_error_db": -0.17737029802149856, "output_peak": 0.5533969279903677}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.815690045611656, "全帯域_10ms_rms_envelope_nrmse": 0.08400844596360345, "全帯域_10ms_envelope_correlation": 0.993853216302455, "全帯域_10ms_modulation_mae_db": 0.7600855550314954, "全帯域_40ms_log_rms_mae_db": 0.4296647412936526, "全帯域_40ms_rms_envelope_nrmse": 0.05570990769105419, "全帯域_40ms_envelope_correlation": 0.9973667481421636, "全帯域_40ms_modulation_mae_db": 0.6794123443743801, "300–1000 Hz_10ms_log_rms_mae_db": 1.8474441439637224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1317870279212061, "300–1000 Hz_10ms_envelope_correlation": 0.9855784957864618, "300–1000 Hz_10ms_modulation_mae_db": 0.748733286393865, "300–1000 Hz_40ms_log_rms_mae_db": 1.191594306214742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08383987768480897, "300–1000 Hz_40ms_envelope_correlation": 0.99376427585449, "300–1000 Hz_40ms_modulation_mae_db": 0.41753967574878414, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4913575620588195, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13565212830104925, "1000–3000 Hz_10ms_envelope_correlation": 0.9869779228055334, "1000–3000 Hz_10ms_modulation_mae_db": 0.40670296592669736, "1000–3000 Hz_40ms_log_rms_mae_db": 0.816526117541067, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07948144667755204, "1000–3000 Hz_40ms_envelope_correlation": 0.9961692512188776, "1000–3000 Hz_40ms_modulation_mae_db": 0.6816664459546226, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7446479044922165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13422206429389072, "3000–6000 Hz_10ms_envelope_correlation": 0.9871597465779033, "3000–6000 Hz_10ms_modulation_mae_db": 0.6206836072565606, "3000–6000 Hz_40ms_log_rms_mae_db": 0.861743537992836, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06702301986564743, "3000–6000 Hz_40ms_envelope_correlation": 0.9968158124201544, "3000–6000 Hz_40ms_modulation_mae_db": 0.3808262953234935, "mr_stft_db": 5.9970061709682305, "band_mae_db": 0.9526139868053778, "rms_gain_error_db": -0.20089305501971408, "output_peak": 0.5715366612980859}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7810548402346115, "全帯域_10ms_rms_envelope_nrmse": 0.0698110643076859, "全帯域_10ms_envelope_correlation": 0.9955148958403021, "全帯域_10ms_modulation_mae_db": 0.5398935172725823, "全帯域_40ms_log_rms_mae_db": 0.3788309874846798, "全帯域_40ms_rms_envelope_nrmse": 0.030920630236450163, "全帯域_40ms_envelope_correlation": 0.9990674338802835, "全帯域_40ms_modulation_mae_db": 0.3465304224505008, "300–1000 Hz_10ms_log_rms_mae_db": 1.878791944463123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10955195969209214, "300–1000 Hz_10ms_envelope_correlation": 0.9900328889532803, "300–1000 Hz_10ms_modulation_mae_db": 0.6320735018449694, "300–1000 Hz_40ms_log_rms_mae_db": 1.2044798975927407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061367374048170144, "300–1000 Hz_40ms_envelope_correlation": 0.9966076263343182, "300–1000 Hz_40ms_modulation_mae_db": 0.3107802067165616, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5258235296045253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12705989936601791, "1000–3000 Hz_10ms_envelope_correlation": 0.9879315212969101, "1000–3000 Hz_10ms_modulation_mae_db": 0.8300616966345121, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8031759913706676, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058176589019780824, "1000–3000 Hz_40ms_envelope_correlation": 0.9973721771844027, "1000–3000 Hz_40ms_modulation_mae_db": 0.9283219524344144, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7892111519605098, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13727871023355998, "3000–6000 Hz_10ms_envelope_correlation": 0.986493021993024, "3000–6000 Hz_10ms_modulation_mae_db": 0.27298009324557254, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9210863121825704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07023139631845549, "3000–6000 Hz_40ms_envelope_correlation": 0.99636147333921, "3000–6000 Hz_40ms_modulation_mae_db": 0.686296762837472, "mr_stft_db": 6.009616425423883, "band_mae_db": 0.8655164565655394, "rms_gain_error_db": -0.04718012041480128, "output_peak": 0.5855071325870955}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7831325000835477, "全帯域_10ms_rms_envelope_nrmse": 0.08130596942114544, "全帯域_10ms_envelope_correlation": 0.9940204361377265, "全帯域_10ms_modulation_mae_db": 0.6410142763050025, "全帯域_40ms_log_rms_mae_db": 0.37851150453586174, "全帯域_40ms_rms_envelope_nrmse": 0.044819463930258895, "全帯域_40ms_envelope_correlation": 0.9980702045279963, "全帯域_40ms_modulation_mae_db": 0.4920987745558632, "300–1000 Hz_10ms_log_rms_mae_db": 1.8411475282990175, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11911158602313376, "300–1000 Hz_10ms_envelope_correlation": 0.9882050017262964, "300–1000 Hz_10ms_modulation_mae_db": 0.8833286021688727, "300–1000 Hz_40ms_log_rms_mae_db": 1.155696020737121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05661669798370394, "300–1000 Hz_40ms_envelope_correlation": 0.9971508486656513, "300–1000 Hz_40ms_modulation_mae_db": 0.5069058746109111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4583939109900337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13798814954290436, "1000–3000 Hz_10ms_envelope_correlation": 0.9862909322994373, "1000–3000 Hz_10ms_modulation_mae_db": 0.4767942730292741, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8020133286259055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08156083360109667, "1000–3000 Hz_40ms_envelope_correlation": 0.995057573280418, "1000–3000 Hz_40ms_modulation_mae_db": 0.4188698487351283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.744828631696613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1353133117960746, "3000–6000 Hz_10ms_envelope_correlation": 0.987414129683652, "3000–6000 Hz_10ms_modulation_mae_db": 0.23000373800981785, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8743452127600326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07444835459347088, "3000–6000 Hz_40ms_envelope_correlation": 0.9961487808547288, "3000–6000 Hz_40ms_modulation_mae_db": 0.6578516047066663, "mr_stft_db": 5.968541072042228, "band_mae_db": 0.8533450838931763, "rms_gain_error_db": 0.06187949273895258, "output_peak": 0.5847658652562692}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.822686433554806, "全帯域_10ms_rms_envelope_nrmse": 0.07591926975567127, "全帯域_10ms_envelope_correlation": 0.99476348711433, "全帯域_10ms_modulation_mae_db": 0.29464856889929375, "全帯域_40ms_log_rms_mae_db": 0.4021654504289266, "全帯域_40ms_rms_envelope_nrmse": 0.038690297006193496, "全帯域_40ms_envelope_correlation": 0.9986262475205907, "全帯域_40ms_modulation_mae_db": 0.33311706124542956, "300–1000 Hz_10ms_log_rms_mae_db": 1.7869106680384708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09843782231529191, "300–1000 Hz_10ms_envelope_correlation": 0.9919755995361638, "300–1000 Hz_10ms_modulation_mae_db": 0.41187991448165606, "300–1000 Hz_40ms_log_rms_mae_db": 1.0675666269044823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05454348670976743, "300–1000 Hz_40ms_envelope_correlation": 0.997316068340161, "300–1000 Hz_40ms_modulation_mae_db": 0.270243533716313, "1000–3000 Hz_10ms_log_rms_mae_db": 1.439328936921524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11867918301370511, "1000–3000 Hz_10ms_envelope_correlation": 0.9899231820873869, "1000–3000 Hz_10ms_modulation_mae_db": 0.6789753184761356, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7725738081390001, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07091913351094158, "1000–3000 Hz_40ms_envelope_correlation": 0.9966959450032219, "1000–3000 Hz_40ms_modulation_mae_db": 0.4540146683426527, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8333527128634293, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13889799408926617, "3000–6000 Hz_10ms_envelope_correlation": 0.9862250718997263, "3000–6000 Hz_10ms_modulation_mae_db": 0.6479389077956578, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9205750566407579, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07931042329928036, "3000–6000 Hz_40ms_envelope_correlation": 0.9953416844003213, "3000–6000 Hz_40ms_modulation_mae_db": 0.7006015158105253, "mr_stft_db": 6.0168811059675305, "band_mae_db": 0.9621759418331514, "rms_gain_error_db": -0.10096416157107435, "output_peak": 0.5939531920254465}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8334401956523304, "全帯域_10ms_rms_envelope_nrmse": 0.08498256417779752, "全帯域_10ms_envelope_correlation": 0.9933631992614343, "全帯域_10ms_modulation_mae_db": 0.29143528752834685, "全帯域_40ms_log_rms_mae_db": 0.43801995129184684, "全帯域_40ms_rms_envelope_nrmse": 0.04940145867868859, "全帯域_40ms_envelope_correlation": 0.9975645420624174, "全帯域_40ms_modulation_mae_db": 0.2845619035766778, "300–1000 Hz_10ms_log_rms_mae_db": 1.7737056208789013, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11203156771716045, "300–1000 Hz_10ms_envelope_correlation": 0.9899558900685619, "300–1000 Hz_10ms_modulation_mae_db": 0.43837311530075196, "300–1000 Hz_40ms_log_rms_mae_db": 1.110626212940592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07037592796961566, "300–1000 Hz_40ms_envelope_correlation": 0.9958067571975114, "300–1000 Hz_40ms_modulation_mae_db": 0.4656235845426037, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4529124336738846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13627939456700508, "1000–3000 Hz_10ms_envelope_correlation": 0.9861028202016375, "1000–3000 Hz_10ms_modulation_mae_db": 0.6412670709348666, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7804894363884557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07239392998356904, "1000–3000 Hz_40ms_envelope_correlation": 0.9958473216898824, "1000–3000 Hz_40ms_modulation_mae_db": 0.6812430904029062, "3000–6000 Hz_10ms_log_rms_mae_db": 1.789676277443662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14142630115590063, "3000–6000 Hz_10ms_envelope_correlation": 0.9857007922986829, "3000–6000 Hz_10ms_modulation_mae_db": 0.21015802050299245, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9175619292599674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07674030130232827, "3000–6000 Hz_40ms_envelope_correlation": 0.99572016471739, "3000–6000 Hz_40ms_modulation_mae_db": 0.16322717908723047, "mr_stft_db": 5.999568130269254, "band_mae_db": 0.875120098103443, "rms_gain_error_db": 0.017379090229028244, "output_peak": 0.5655431165894973}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85120, "output_samples": 85120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.805339047945468, "spectral_convergence": 0.3518638403630249, "spectral_convergence_status": "ok", "amplitude_floor": 1.1374446115550809e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.099734057304352, "spectral_convergence": 0.38196274866156593, "spectral_convergence_status": "ok", "amplitude_floor": 9.193083193108671e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.07491374647613, "spectral_convergence": 0.37382139902682526, "spectral_convergence_status": "ok", "amplitude_floor": 7.3161107874683e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.1849686170678433e-06, "output_power": 6.051264514244918e-06, "delta_db": 4.424009337918131, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004950757600852583, "output_power": 0.0004964137173365777, "delta_db": 0.011721108840981147, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008267786751374999, "output_power": 0.0008700708399787563, "delta_db": 0.22165347253240555, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007700805582515175, "output_power": 0.0006997658299529717, "delta_db": -0.4158342739858615, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006256626820185013, "output_power": 0.0005979091356419721, "delta_db": -0.1970506283299886, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00016725866300063496, "output_power": 0.00016068370826400474, "delta_db": -0.1741677504703844, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9074060953462921, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08529313717857144, "pitch": null, "output_rms_dbfs": -25.480763160090078, "output_peak": 0.5610607867684839, "reference_rms_dbfs": -25.39547002291151, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.993328950575317}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.0002036923290800657, "log_rms_mae_db": 0.7757918731859916, "rms_envelope_nrmse": 0.07431443206417887, "envelope_correlation": 0.9949524711304688, "rms_gain_error_db": -0.08529177441351937, "reference_mean_power": 0.0028897508304992817, "output_mean_power": 0.0028335522137470093}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 1.5709970652253345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007599193486798772, "output_power": 0.000736630853552156, "reference_active": true, "delta_db": -0.1351759724820411}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014055064928452022, "output_power": 0.00013866554044896263, "reference_active": true, "delta_db": -0.05864307548600112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3577241130570316e-05, "output_power": 2.5191180286403343e-05, "reference_active": true, "delta_db": 0.28755530885477815}], "active_band_count": 3, "active_band_mae_db": 0.16045811894094011}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.0001938989159680862, "log_rms_mae_db": 0.39266046179122016, "rms_envelope_nrmse": 0.03589342924962537, "envelope_correlation": 0.9987938989634922, "rms_gain_error_db": -0.08528576464198068, "reference_mean_power": 0.002906120097080002, "output_mean_power": 0.002849607081240427}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 1.4139945205650041e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006810740061577038, "output_power": 0.0006642002533767963, "reference_active": true, "delta_db": -0.1089526839816401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.5174236911387e-05, "output_power": 6.354560270439718e-05, "reference_active": true, "delta_db": -0.10990451797028537}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3046113483012314e-06, "output_power": 1.333202599583918e-06, "reference_active": true, "delta_db": 0.09414999851200755}], "active_band_count": 3, "active_band_mae_db": 0.10433573348797769}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013759847516455276, "log_rms_mae_db": 1.786181278630355, "rms_envelope_nrmse": 0.11585963543924115, "envelope_correlation": 0.9894010940204166, "rms_gain_error_db": 0.15835006321385808, "reference_mean_power": 0.001146104301837711, "output_mean_power": 0.0011886641071150155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 6.878091269334419e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031946504747131925, "output_power": 0.00034747912187930595, "reference_active": true, "delta_db": 0.3650536620141782}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.126588772977069e-05, "output_power": 8.664853099197797e-05, "reference_active": true, "delta_db": 0.8487950490412056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2614620971957042e-05, "output_power": 1.7657778980008296e-05, "reference_active": true, "delta_db": 1.4606187059004205}], "active_band_count": 3, "active_band_mae_db": 0.8914891389852682}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00012449215623942515, "log_rms_mae_db": 1.1425532695247083, "rms_envelope_nrmse": 0.07123426560413583, "envelope_correlation": 0.995826205572138, "rms_gain_error_db": 0.15834957898358912, "reference_mean_power": 0.001152608697675057, "output_mean_power": 0.0011954099059906325}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 6.232509709606391e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00029975403822873686, "output_power": 0.0003307356690388893, "reference_active": true, "delta_db": 0.4271599250295648}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2561293827419036e-05, "output_power": 4.070951968898714e-05, "reference_active": true, "delta_db": 0.9699432507884052}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.743814177835147e-07, "output_power": 9.491031059226638e-07, "reference_active": true, "delta_db": 0.8835847161478876}], "active_band_count": 3, "active_band_mae_db": 0.7602292973219525}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013997823549865974, "log_rms_mae_db": 1.4591054623545936, "rms_envelope_nrmse": 0.1547920604918512, "envelope_correlation": 0.9826534385105411, "rms_gain_error_db": -0.36210244196971475, "reference_mean_power": 0.0010964326399058018, "output_mean_power": 0.0010087225078725737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 8.778805813988002e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045662233680262583, "output_power": 0.0003955609317831422, "reference_active": true, "delta_db": -0.6234376101014119}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.737497372188522e-05, "output_power": 3.961456156632429e-05, "reference_active": true, "delta_db": -0.7769412833989008}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.406352772002978e-05, "output_power": 1.68435783552817e-05, "reference_active": true, "delta_db": 0.783400876643898}], "active_band_count": 3, "active_band_mae_db": 0.7279265900480704}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013340521153934576, "log_rms_mae_db": 0.8441101570446493, "rms_envelope_nrmse": 0.09556017622299896, "envelope_correlation": 0.9938269371757146, "rms_gain_error_db": -0.36209442319463303, "reference_mean_power": 0.0011026403148293462, "output_mean_power": 0.0010144354672048867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 8.248821386475387e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042363838931095744, "output_power": 0.00036773033674515995, "reference_active": true, "delta_db": -0.6146584871309236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.073594132211124e-05, "output_power": 1.6849737854385773e-05, "reference_active": true, "delta_db": -0.9013060666634369}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.366449484579509e-07, "output_power": 8.157091621682691e-07, "reference_active": true, "delta_db": 0.44277125372153814}], "active_band_count": 3, "active_band_mae_db": 0.6529119358386328}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00010989278407704884, "log_rms_mae_db": 1.8229964225071074, "rms_envelope_nrmse": 0.13905723622174934, "envelope_correlation": 0.9863667471561468, "rms_gain_error_db": -0.21607032031486006, "reference_mean_power": 0.0003075575386675287, "output_mean_power": 0.0002926303350968073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 2.696003640873154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014789902573029864, "output_power": 0.00013381036504432677, "reference_active": true, "delta_db": -0.43475557647185337}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.17770472077717e-05, "output_power": 2.491707486300333e-05, "reference_active": true, "delta_db": -1.056164821286782}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.120339140119136e-06, "output_power": 3.1598103987170693e-06, "reference_active": true, "delta_db": 0.05459225268476789}], "active_band_count": 3, "active_band_mae_db": 0.5151708834811345}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010174719078202204, "log_rms_mae_db": 0.9405444416918781, "rms_envelope_nrmse": 0.08301471766329012, "envelope_correlation": 0.9951734254613943, "rms_gain_error_db": -0.21605874071164866, "reference_mean_power": 0.000309290771080255, "output_mean_power": 0.0002942802302857402}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.5256223542383704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014158879062551878, "output_power": 0.00012705992769529068, "reference_active": true, "delta_db": -0.4702026840516399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.501151157810192e-05, "output_power": 1.14097006504973e-05, "reference_active": true, "delta_db": -1.191501753100982}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.631467090177325e-07, "output_power": 1.9839753063722732e-07, "reference_active": true, "delta_db": -1.2266168059750575}], "active_band_count": 3, "active_band_mae_db": 0.9627737477092264}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8430495566092443, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "faa5e505437e18cc9c133490c5d5c82473bf02f5f09d52552dc1684338e2fba3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9870022638917475, "standard_deviation": 0.002170329989580081, "minimum": 0.9826534385105411, "maximum": 0.9899231820873869, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4683603349456384, "standard_deviation": 0.0345821803827513, "minimum": 1.4203968943196907, "maximum": 1.5258235296045253, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7290697789824581, "standard_deviation": 0.3219092251222591, "minimum": 0.40670296592669736, "maximum": 1.4535665207458892, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.133379480812805, "standard_deviation": 0.01103442198309057, "minimum": 0.11867918301370511, "maximum": 0.1547920604918512, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957137535565073, "standard_deviation": 0.0011304531386104142, "minimum": 0.9938269371757146, "maximum": 0.9973721771844027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8041151862872522, "standard_deviation": 0.02232655443114981, "minimum": 0.7725738081390001, "maximum": 0.8441101570446493, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7295093302001417, "standard_deviation": 0.249708282785661, "minimum": 0.4188698487351283, "maximum": 1.178723668676708, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07753393539844877, "standard_deviation": 0.01148534165907256, "minimum": 0.058176589019780824, "maximum": 0.09556017622299896, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9860666976994396, "standard_deviation": 0.0010595004275535808, "minimum": 0.9844493578724823, "maximum": 0.987414129683652, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8058866475979847, "standard_deviation": 0.047940531555966205, "minimum": 1.7446479044922165, "maximum": 1.8824607850983466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.40985018371248205, "standard_deviation": 0.1921713897660117, "minimum": 0.21015802050299245, "maximum": 0.6479389077956578, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14023347363543576, "standard_deviation": 0.005300791828821456, "minimum": 0.13422206429389072, "maximum": 0.14976271485084633, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954914098886466, "standard_deviation": 0.000981122805421212, "minimum": 0.9938643537255072, "maximum": 0.9968158124201544, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.917259476411266, "standard_deviation": 0.03394142965293125, "minimum": 0.861743537992836, "maximum": 0.962960870135504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5994957225473995, "standard_deviation": 0.2367425645813359, "minimum": 0.16322717908723047, "maximum": 0.9627737477092264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07845506649141742, "standard_deviation": 0.008162722987597638, "minimum": 0.06702301986564743, "maximum": 0.09219975066825642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9895098226244157, "standard_deviation": 0.0019397753421332442, "minimum": 0.9855784957864618, "maximum": 0.9919755995361638, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8147840609317236, "standard_deviation": 0.04470897150334093, "minimum": 1.7544719993668925, "maximum": 1.878791944463123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7137090977576619, "standard_deviation": 0.26807409144119804, "minimum": 0.41187991448165606, "maximum": 1.1907748429938925, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11289490947184258, "standard_deviation": 0.009901153487526631, "minimum": 0.09843782231529191, "maximum": 0.1317870279212061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9963520829569256, "standard_deviation": 0.001218296729212128, "minimum": 0.99376427585449, "maximum": 0.9974374734950323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1509335444181705, "standard_deviation": 0.04707693432439762, "minimum": 1.0675666269044823, "maximum": 1.2044798975927407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5208425251518906, "standard_deviation": 0.25178840499023286, "minimum": 0.270243533716313, "maximum": 1.0244027360766104, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06451578130538996, "standard_deviation": 0.010089269401300066, "minimum": 0.05454348670976743, "maximum": 0.08383987768480897, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.9177084673093875, "standard_deviation": 0.049847358513148476, "minimum": 0.8533450838931763, "maximum": 0.9887682767204039, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.998559902622431, "standard_deviation": 0.01421998330092438, "minimum": 5.968541072042228, "maximum": 6.0168811059675305, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5777971418314527, "standard_deviation": 0.017985055417158072, "minimum": 0.5533969279903677, "maximum": 0.6066134521363757, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08367146420145435, "standard_deviation": 0.09132858634030622, "minimum": -0.20089305501971408, "maximum": 0.06187949273895258, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944083159787171, "standard_deviation": 0.0006789635172956154, "minimum": 0.9933631992614343, "maximum": 0.9955148958403021, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8037263093208098, "standard_deviation": 0.021485806285807606, "minimum": 0.7757918731859916, "maximum": 0.8334401956523304, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4577464178747275, "standard_deviation": 0.21242953884302662, "minimum": 0.16045811894094011, "maximum": 0.7600855550314954, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07864471815660813, "standard_deviation": 0.005131301527871982, "minimum": 0.0698110643076859, "maximum": 0.08498256417779752, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981686717574685, "standard_deviation": 0.0006115980148053103, "minimum": 0.9973667481421636, "maximum": 0.9990674338802835, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4113657065495424, "standard_deviation": 0.026208788483658197, "minimum": 0.37851150453586174, "maximum": 0.4388627071527473, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42760984392808854, "standard_deviation": 0.1964487329972867, "minimum": 0.10433573348797769, "maximum": 0.6794123443743801, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0440859839830725, "standard_deviation": 0.008262493003077983, "minimum": 0.030920630236450163, "maximum": 0.05570990769105419, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141083-0010", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.8770332336425781, "predicted_mos": 1.8770332336425781, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "52b2e0f3ba8f34ac20a801d211b54b746203493334e52933e2fd629ee3b94f91", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7593700997531414, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8561591710719134, "全帯域_10ms_rms_envelope_nrmse": 0.08836185912555664, "全帯域_10ms_envelope_correlation": 0.9927959177582699, "全帯域_10ms_modulation_mae_db": 0.3321695775774582, "全帯域_40ms_log_rms_mae_db": 0.45595763606544615, "全帯域_40ms_rms_envelope_nrmse": 0.05414291702223283, "全帯域_40ms_envelope_correlation": 0.9970804493150898, "全帯域_40ms_modulation_mae_db": 0.29775949055413414, "300–1000 Hz_10ms_log_rms_mae_db": 1.4087614934479888, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1200584406947387, "300–1000 Hz_10ms_envelope_correlation": 0.9881297762049229, "300–1000 Hz_10ms_modulation_mae_db": 0.5783682336842797, "300–1000 Hz_40ms_log_rms_mae_db": 0.7749203418887399, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07813011613525037, "300–1000 Hz_40ms_envelope_correlation": 0.9945287916029529, "300–1000 Hz_40ms_modulation_mae_db": 0.5912485253332564, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3212227048947853, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12689735207335548, "1000–3000 Hz_10ms_envelope_correlation": 0.9879993010688972, "1000–3000 Hz_10ms_modulation_mae_db": 0.40122062065453745, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7555987834573772, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.068383801775426, "1000–3000 Hz_40ms_envelope_correlation": 0.9964209478236287, "1000–3000 Hz_40ms_modulation_mae_db": 0.48298340371502935, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3788902284942635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14019461254656782, "3000–6000 Hz_10ms_envelope_correlation": 0.986013282236506, "3000–6000 Hz_10ms_modulation_mae_db": 0.5396045664253285, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7382304616419004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08602517344676391, "3000–6000 Hz_40ms_envelope_correlation": 0.9946604777227946, "3000–6000 Hz_40ms_modulation_mae_db": 0.5232508311783153, "mr_stft_db": 5.617899318823386, "band_mae_db": 0.7489643866914758, "rms_gain_error_db": -0.04861258908006879, "output_peak": 0.5996823750254727}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8199389080635863, "全帯域_10ms_rms_envelope_nrmse": 0.08681791984232734, "全帯域_10ms_envelope_correlation": 0.9930845516199822, "全帯域_10ms_modulation_mae_db": 0.13704341174763707, "全帯域_40ms_log_rms_mae_db": 0.40183369415628656, "全帯域_40ms_rms_envelope_nrmse": 0.04457570054828126, "全帯域_40ms_envelope_correlation": 0.9980259982261329, "全帯域_40ms_modulation_mae_db": 0.6545086755517154, "300–1000 Hz_10ms_log_rms_mae_db": 1.364983146601229, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10435208100371825, "300–1000 Hz_10ms_envelope_correlation": 0.9909594087577412, "300–1000 Hz_10ms_modulation_mae_db": 0.5492828432146145, "300–1000 Hz_40ms_log_rms_mae_db": 0.7634769140428626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05062299794742226, "300–1000 Hz_40ms_envelope_correlation": 0.9977042876162917, "300–1000 Hz_40ms_modulation_mae_db": 0.15271402425912825, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2185308291442642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12789871642978135, "1000–3000 Hz_10ms_envelope_correlation": 0.988140525943644, "1000–3000 Hz_10ms_modulation_mae_db": 0.0996481644486325, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6715962176216322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06841812113215025, "1000–3000 Hz_40ms_envelope_correlation": 0.9965076930019369, "1000–3000 Hz_40ms_modulation_mae_db": 0.40334934221855584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3917551155948067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11816112241086339, "3000–6000 Hz_10ms_envelope_correlation": 0.990026596884972, "3000–6000 Hz_10ms_modulation_mae_db": 0.13113819286788045, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7925854040842774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06404654393234618, "3000–6000 Hz_40ms_envelope_correlation": 0.9969869280646132, "3000–6000 Hz_40ms_modulation_mae_db": 0.5125845817577247, "mr_stft_db": 5.60917994588985, "band_mae_db": 0.7151921007212771, "rms_gain_error_db": 0.016763726936550545, "output_peak": 0.5576525943777787}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.807941769484754, "全帯域_10ms_rms_envelope_nrmse": 0.0912260841705861, "全帯域_10ms_envelope_correlation": 0.9924871503664048, "全帯域_10ms_modulation_mae_db": 0.31568817458876736, "全帯域_40ms_log_rms_mae_db": 0.4056910691679348, "全帯域_40ms_rms_envelope_nrmse": 0.04615863878245694, "全帯域_40ms_envelope_correlation": 0.9981589686957238, "全帯域_40ms_modulation_mae_db": 0.38659645617679805, "300–1000 Hz_10ms_log_rms_mae_db": 1.3942399339018305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11182642440443477, "300–1000 Hz_10ms_envelope_correlation": 0.9896403965777519, "300–1000 Hz_10ms_modulation_mae_db": 0.893967230091909, "300–1000 Hz_40ms_log_rms_mae_db": 0.8147073251008967, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06379397537743571, "300–1000 Hz_40ms_envelope_correlation": 0.996435568439923, "300–1000 Hz_40ms_modulation_mae_db": 0.777282942495813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2486744023494782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13778397568540887, "1000–3000 Hz_10ms_envelope_correlation": 0.9859501663986133, "1000–3000 Hz_10ms_modulation_mae_db": 0.25986357973565927, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6999269962360112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06768972296811969, "1000–3000 Hz_40ms_envelope_correlation": 0.9967563860154774, "1000–3000 Hz_40ms_modulation_mae_db": 0.28896139901457013, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4177566340041285, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13517221772768298, "3000–6000 Hz_10ms_envelope_correlation": 0.9868811856103408, "3000–6000 Hz_10ms_modulation_mae_db": 0.5361258998988522, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7963225509826156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.076186865473162, "3000–6000 Hz_40ms_envelope_correlation": 0.9955937439077973, "3000–6000 Hz_40ms_modulation_mae_db": 0.34703873894297654, "mr_stft_db": 5.592366959538187, "band_mae_db": 0.725029027865239, "rms_gain_error_db": -0.15085201668386317, "output_peak": 0.5655298400415785}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7809808625502566, "全帯域_10ms_rms_envelope_nrmse": 0.08034231902212689, "全帯域_10ms_envelope_correlation": 0.9940983881017542, "全帯域_10ms_modulation_mae_db": 0.2894526977271952, "全帯域_40ms_log_rms_mae_db": 0.408804692251766, "全帯域_40ms_rms_envelope_nrmse": 0.050172999924244, "全帯域_40ms_envelope_correlation": 0.997571685296666, "全帯域_40ms_modulation_mae_db": 0.3931412266024064, "300–1000 Hz_10ms_log_rms_mae_db": 1.3715902826851187, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.107936468454986, "300–1000 Hz_10ms_envelope_correlation": 0.9903293788300561, "300–1000 Hz_10ms_modulation_mae_db": 0.3718653129301391, "300–1000 Hz_40ms_log_rms_mae_db": 0.7617143763166163, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0687406900361057, "300–1000 Hz_40ms_envelope_correlation": 0.9957534225770963, "300–1000 Hz_40ms_modulation_mae_db": 0.6727959656153452, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2488611337101876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11874743646644127, "1000–3000 Hz_10ms_envelope_correlation": 0.9895083886270797, "1000–3000 Hz_10ms_modulation_mae_db": 0.48830793468927963, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6698033897562344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0672939535907808, "1000–3000 Hz_40ms_envelope_correlation": 0.9965344152496386, "1000–3000 Hz_40ms_modulation_mae_db": 0.30433587269708123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3963052434032321, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13034140047689585, "3000–6000 Hz_10ms_envelope_correlation": 0.9883887720268439, "3000–6000 Hz_10ms_modulation_mae_db": 0.5053564207160172, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7881066396730759, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08052401755988771, "3000–6000 Hz_40ms_envelope_correlation": 0.9959331330460357, "3000–6000 Hz_40ms_modulation_mae_db": 0.6425057669046179, "mr_stft_db": 5.597320564063444, "band_mae_db": 0.8652803831305297, "rms_gain_error_db": -0.09023979859706036, "output_peak": 0.5756808900990494}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8495602548302869, "全帯域_10ms_rms_envelope_nrmse": 0.09078603720302669, "全帯域_10ms_envelope_correlation": 0.9923990639747939, "全帯域_10ms_modulation_mae_db": 0.3846861592645568, "全帯域_40ms_log_rms_mae_db": 0.4038464855698336, "全帯域_40ms_rms_envelope_nrmse": 0.04294734589417863, "全帯域_40ms_envelope_correlation": 0.9981810830193036, "全帯域_40ms_modulation_mae_db": 0.24542588049470537, "300–1000 Hz_10ms_log_rms_mae_db": 1.4165860356232862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10768737212217709, "300–1000 Hz_10ms_envelope_correlation": 0.9904173625100724, "300–1000 Hz_10ms_modulation_mae_db": 0.5894773969011371, "300–1000 Hz_40ms_log_rms_mae_db": 0.8157270448034984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05793752875384058, "300–1000 Hz_40ms_envelope_correlation": 0.9969796352495484, "300–1000 Hz_40ms_modulation_mae_db": 0.40274774347115194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.254692924402519, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13495401237808663, "1000–3000 Hz_10ms_envelope_correlation": 0.9864984664689831, "1000–3000 Hz_10ms_modulation_mae_db": 0.2538931971512835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.682711182637209, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06998496908266916, "1000–3000 Hz_40ms_envelope_correlation": 0.9963518516051499, "1000–3000 Hz_40ms_modulation_mae_db": 0.18980011382377646, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3854050830410092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1226599093230848, "3000–6000 Hz_10ms_envelope_correlation": 0.989216024462149, "3000–6000 Hz_10ms_modulation_mae_db": 0.27594352418966245, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7644067757411863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06520391918575977, "3000–6000 Hz_40ms_envelope_correlation": 0.9968096207290404, "3000–6000 Hz_40ms_modulation_mae_db": 0.43586483750919275, "mr_stft_db": 5.606339301322364, "band_mae_db": 0.6020451718937936, "rms_gain_error_db": -0.05342004065818595, "output_peak": 0.564022978541989}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8353407804366902, "全帯域_10ms_rms_envelope_nrmse": 0.0913545334186052, "全帯域_10ms_envelope_correlation": 0.9923513567170287, "全帯域_10ms_modulation_mae_db": 0.3055446579841738, "全帯域_40ms_log_rms_mae_db": 0.4054340787755013, "全帯域_40ms_rms_envelope_nrmse": 0.0440955766973306, "全帯域_40ms_envelope_correlation": 0.9981790617114753, "全帯域_40ms_modulation_mae_db": 0.3839981855892136, "300–1000 Hz_10ms_log_rms_mae_db": 1.4218291431637835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11457087751290806, "300–1000 Hz_10ms_envelope_correlation": 0.9891654035921134, "300–1000 Hz_10ms_modulation_mae_db": 1.0456061314678948, "300–1000 Hz_40ms_log_rms_mae_db": 0.8179378895237647, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058591658203344665, "300–1000 Hz_40ms_envelope_correlation": 0.9971091533467425, "300–1000 Hz_40ms_modulation_mae_db": 0.7650450777220689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2707338289352232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14456945032336796, "1000–3000 Hz_10ms_envelope_correlation": 0.9843719098146565, "1000–3000 Hz_10ms_modulation_mae_db": 0.7012270831883765, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6744160326248471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0702114949922812, "1000–3000 Hz_40ms_envelope_correlation": 0.9961059142031163, "1000–3000 Hz_40ms_modulation_mae_db": 0.5507852834681407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3778859239344414, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11667598421618257, "3000–6000 Hz_10ms_envelope_correlation": 0.9903806198665793, "3000–6000 Hz_10ms_modulation_mae_db": 0.27762197127786176, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7518200605198554, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06658921615251617, "3000–6000 Hz_40ms_envelope_correlation": 0.9968869758146452, "3000–6000 Hz_40ms_modulation_mae_db": 0.4990927030013113, "mr_stft_db": 5.603719120770731, "band_mae_db": 0.8351824818267327, "rms_gain_error_db": -0.10228640853406419, "output_peak": 0.5658379872850227}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7981284985030029, "全帯域_10ms_rms_envelope_nrmse": 0.09381156797243081, "全帯域_10ms_envelope_correlation": 0.9918853216235541, "全帯域_10ms_modulation_mae_db": 0.5941234689703974, "全帯域_40ms_log_rms_mae_db": 0.3911381955335379, "全帯域_40ms_rms_envelope_nrmse": 0.04413783696151415, "全帯域_40ms_envelope_correlation": 0.998087299031981, "全帯域_40ms_modulation_mae_db": 0.5177124893341172, "300–1000 Hz_10ms_log_rms_mae_db": 1.342616851709013, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11295278870026082, "300–1000 Hz_10ms_envelope_correlation": 0.9894085824660386, "300–1000 Hz_10ms_modulation_mae_db": 0.5875317672003934, "300–1000 Hz_40ms_log_rms_mae_db": 0.7273524503351793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06810146786017753, "300–1000 Hz_40ms_envelope_correlation": 0.9958639432982775, "300–1000 Hz_40ms_modulation_mae_db": 0.8580836375022582, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3051376170559212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15649177195398162, "1000–3000 Hz_10ms_envelope_correlation": 0.9817487492499267, "1000–3000 Hz_10ms_modulation_mae_db": 1.1790165864027993, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6946850405701162, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08425980785125457, "1000–3000 Hz_40ms_envelope_correlation": 0.9944330480011081, "1000–3000 Hz_40ms_modulation_mae_db": 1.495695630877522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3756217718877386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12467561680141907, "3000–6000 Hz_10ms_envelope_correlation": 0.9889778102476491, "3000–6000 Hz_10ms_modulation_mae_db": 0.22980521119525069, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7572079197246278, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06758599581970771, "3000–6000 Hz_40ms_envelope_correlation": 0.9966097655731349, "3000–6000 Hz_40ms_modulation_mae_db": 0.4218901880091644, "mr_stft_db": 5.613589400132219, "band_mae_db": 0.7591009814266417, "rms_gain_error_db": -0.060676542747203945, "output_peak": 0.5526503980069762}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8345431450568952, "全帯域_10ms_rms_envelope_nrmse": 0.07791561686822142, "全帯域_10ms_envelope_correlation": 0.9944281833574851, "全帯域_10ms_modulation_mae_db": 0.41224928179036696, "全帯域_40ms_log_rms_mae_db": 0.42436548586534184, "全帯域_40ms_rms_envelope_nrmse": 0.04138914436780285, "全帯域_40ms_envelope_correlation": 0.998344425638315, "全帯域_40ms_modulation_mae_db": 0.28437316882826946, "300–1000 Hz_10ms_log_rms_mae_db": 1.4115496582493121, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10471744639241792, "300–1000 Hz_10ms_envelope_correlation": 0.9909498336850803, "300–1000 Hz_10ms_modulation_mae_db": 0.30265371440099775, "300–1000 Hz_40ms_log_rms_mae_db": 0.8004162176703956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060242756229339556, "300–1000 Hz_40ms_envelope_correlation": 0.9968442625918424, "300–1000 Hz_40ms_modulation_mae_db": 0.12195958314828292, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3007495381794945, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11983095721023178, "1000–3000 Hz_10ms_envelope_correlation": 0.9893193064511732, "1000–3000 Hz_10ms_modulation_mae_db": 0.6511153776006874, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7135987673993911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05622434428669891, "1000–3000 Hz_40ms_envelope_correlation": 0.9975303016076194, "1000–3000 Hz_40ms_modulation_mae_db": 0.8143865758913581, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4274839089447051, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13080517344421824, "3000–6000 Hz_10ms_envelope_correlation": 0.9878968233816857, "3000–6000 Hz_10ms_modulation_mae_db": 0.7518791740884226, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7709906434187181, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08935366888317933, "3000–6000 Hz_40ms_envelope_correlation": 0.9942015746422218, "3000–6000 Hz_40ms_modulation_mae_db": 1.259874421772089, "mr_stft_db": 5.601133076528959, "band_mae_db": 0.7073138910002936, "rms_gain_error_db": -0.07438208914690933, "output_peak": 0.575487914143462}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85120, "output_samples": 85120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.3559344392638994, "spectral_convergence": 0.34701348904578294, "spectral_convergence_status": "ok", "amplitude_floor": 1.1374446115550809e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.602427763930224, "spectral_convergence": 0.3767337171749748, "spectral_convergence_status": "ok", "amplitude_floor": 9.193083193108671e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.895335753276032, "spectral_convergence": 0.3830875848702499, "spectral_convergence_status": "ok", "amplitude_floor": 7.3161107874683e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.1849686170678433e-06, "output_power": 5.215323489130862e-06, "delta_db": 3.7783604790272407, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004950757600852583, "output_power": 0.0005188521381079171, "delta_db": 0.20371947820749578, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008267786751374999, "output_power": 0.0008161183068873578, "delta_db": -0.0563614655732421, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007700805582515175, "output_power": 0.0007493645109681123, "delta_db": -0.11843037601690859, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006256626820185013, "output_power": 0.0006054017229796131, "delta_db": -0.14296599397062815, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00016725866300063496, "output_power": 0.00015995350761360252, "delta_db": -0.19394852735334056, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7489643866914758, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04861258908006879, "pitch": null, "output_rms_dbfs": -25.444082611991583, "output_peak": 0.5996823750254727, "reference_rms_dbfs": -25.39547002291151, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.617899318823386}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.0002036923290800657, "log_rms_mae_db": 0.8561591710719134, "rms_envelope_nrmse": 0.08836185912555664, "envelope_correlation": 0.9927959177582699, "rms_gain_error_db": -0.04861060470535957, "reference_mean_power": 0.0028897508304992817, "output_mean_power": 0.0028575861792836314}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 1.5709970652253345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007599193486798772, "output_power": 0.0007358382763925334, "reference_active": true, "delta_db": -0.13985127484525192}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014055064928452022, "output_power": 0.00012118505109707113, "reference_active": true, "delta_db": -0.6438380589982089}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3577241130570316e-05, "output_power": 2.476138455813966e-05, "reference_active": true, "delta_db": 0.21281939888891363}], "active_band_count": 3, "active_band_mae_db": 0.3321695775774582}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.0001938989159680862, "log_rms_mae_db": 0.45595763606544615, "rms_envelope_nrmse": 0.05414291702223283, "envelope_correlation": 0.9970804493150898, "rms_gain_error_db": -0.04860851650058556, "reference_mean_power": 0.002906120097080002, "output_mean_power": 0.002873774627946739}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 1.4139945205650041e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006810740061577038, "output_power": 0.0006634083836181291, "reference_active": true, "delta_db": -0.11413349853459645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.5174236911387e-05, "output_power": 5.4808400842845794e-05, "reference_active": true, "delta_db": -0.7522882421108417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3046113483012314e-06, "output_power": 1.2965685376270408e-06, "reference_active": true, "delta_db": -0.026856731016964234}], "active_band_count": 3, "active_band_mae_db": 0.29775949055413414}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013759847516455276, "log_rms_mae_db": 1.4087614934479888, "rms_envelope_nrmse": 0.1200584406947387, "envelope_correlation": 0.9881297762049229, "rms_gain_error_db": 0.03413388603590939, "reference_mean_power": 0.001146104301837711, "output_mean_power": 0.0011551477359255762}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 6.878091269334419e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031946504747131925, "output_power": 0.0003118335696567821, "reference_active": true, "delta_db": -0.10500482880486724}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.126588772977069e-05, "output_power": 6.0992943005808485e-05, "reference_active": true, "delta_db": -0.6760211007803273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2614620971957042e-05, "output_power": 1.571383118129528e-05, "reference_active": true, "delta_db": 0.9540787714676444}], "active_band_count": 3, "active_band_mae_db": 0.5783682336842797}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00012449215623942515, "log_rms_mae_db": 0.7749203418887399, "rms_envelope_nrmse": 0.07813011613525037, "envelope_correlation": 0.9945287916029529, "rms_gain_error_db": 0.03413434725008444, "reference_mean_power": 0.001152608697675057, "output_mean_power": 0.0011617035786261405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 6.232509709606391e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00029975403822873686, "output_power": 0.0002926697831160523, "reference_active": true, "delta_db": -0.10387156812634203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2561293827419036e-05, "output_power": 2.715682060458662e-05, "reference_active": true, "delta_db": -0.7882272992211412}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.743814177835147e-07, "output_power": 9.486796700586247e-07, "reference_active": true, "delta_db": 0.8816467086522861}], "active_band_count": 3, "active_band_mae_db": 0.5912485253332564}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013997823549865974, "log_rms_mae_db": 1.3212227048947853, "rms_envelope_nrmse": 0.12689735207335548, "envelope_correlation": 0.9879993010688972, "rms_gain_error_db": -0.13187323936287326, "reference_mean_power": 0.0010964326399058018, "output_mean_power": 0.0010636399274164023}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 8.778805813988002e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045662233680262583, "output_power": 0.0004432322821491812, "reference_active": true, "delta_db": -0.12925768204209762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.737497372188522e-05, "output_power": 4.4533876888867354e-05, "reference_active": true, "delta_db": -0.26858477873046843}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.406352772002978e-05, "output_power": 1.693075065128165e-05, "reference_active": true, "delta_db": 0.8058194011910463}], "active_band_count": 3, "active_band_mae_db": 0.40122062065453745}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013340521153934576, "log_rms_mae_db": 0.7555987834573772, "rms_envelope_nrmse": 0.068383801775426, "envelope_correlation": 0.9964209478236287, "rms_gain_error_db": -0.13186709852413345, "reference_mean_power": 0.0011026403148293462, "output_mean_power": 0.0010696634522545242}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 8.248821386475387e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042363838931095744, "output_power": 0.00041373995873680624, "reference_active": true, "delta_db": -0.10267841364220583}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.073594132211124e-05, "output_power": 1.9233275427420055e-05, "reference_active": true, "delta_db": -0.32670504338842243}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.366449484579509e-07, "output_power": 9.315687049153795e-07, "reference_active": true, "delta_db": 1.0195667541144597}], "active_band_count": 3, "active_band_mae_db": 0.48298340371502935}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00010989278407704884, "log_rms_mae_db": 1.3788902284942635, "rms_envelope_nrmse": 0.14019461254656782, "envelope_correlation": 0.986013282236506, "rms_gain_error_db": -0.21152653404294505, "reference_mean_power": 0.0003075575386675287, "output_mean_power": 0.0002929366584712484}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 2.696003640873154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014789902573029864, "output_power": 0.00012586820663559084, "reference_active": true, "delta_db": -0.7004926870261176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.17770472077717e-05, "output_power": 2.6203618993637847e-05, "reference_active": true, "delta_db": -0.837522630910318}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.120339140119136e-06, "output_power": 3.0628234291998412e-06, "reference_active": true, "delta_db": -0.08079838133954995}], "active_band_count": 3, "active_band_mae_db": 0.5396045664253285}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010174719078202204, "log_rms_mae_db": 0.7382304616419004, "rms_envelope_nrmse": 0.08602517344676391, "envelope_correlation": 0.9946604777227946, "rms_gain_error_db": -0.211544282917453, "reference_mean_power": 0.000309290771080255, "output_mean_power": 0.00029458629137215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.5256223542383704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014158879062551878, "output_power": 0.00011914180864285136, "reference_active": true, "delta_db": -0.7496468360592267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.501151157810192e-05, "output_power": 1.2521156322694302e-05, "reference_active": true, "delta_db": -0.7877998794224851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.631467090177325e-07, "output_power": 2.6119650723480703e-07, "reference_active": true, "delta_db": -0.032305778053234105}], "active_band_count": 3, "active_band_mae_db": 0.5232508311783153}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8563230104402481, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7fbe8fc5dc4238a1208d72d7f522fd598b0a4428952a2c6797089d8e4de9e634", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9866921017528718, "standard_deviation": 0.0026470567554075634, "minimum": 0.9817487492499267, "maximum": 0.9895083886270797, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2710753723339843, "standard_deviation": 0.03501072024008407, "minimum": 1.2185308291442642, "maximum": 1.3212227048947853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5042865679839069, "standard_deviation": 0.3407249508234733, "minimum": 0.0996481644486325, "maximum": 1.1790165864027993, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1333967090650819, "standard_deviation": 0.012822891107333603, "minimum": 0.11874743646644127, "maximum": 0.15649177195398162, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963300696884594, "standard_deviation": 0.0008745022806107655, "minimum": 0.9944330480011081, "maximum": 0.9975303016076194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6952920512878522, "standard_deviation": 0.02878278202445895, "minimum": 0.6698033897562344, "maximum": 0.7555987834573772, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5662872027132543, "standard_deviation": 0.42196014248460345, "minimum": 0.18980011382377646, "maximum": 1.495695630877522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06905827695992257, "standard_deviation": 0.007595445370739236, "minimum": 0.05622434428669891, "maximum": 0.08425980785125457, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9884726393395907, "standard_deviation": 0.0015015588522487864, "minimum": 0.986013282236506, "maximum": 0.9903806198665793, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3938879886630406, "standard_deviation": 0.0192554393306998, "minimum": 1.3756217718877386, "maximum": 1.4274839089447051, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4059343700824095, "standard_deviation": 0.20848988283852624, "minimum": 0.13113819286788045, "maximum": 0.7518791740884226, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12733575461836433, "standard_deviation": 0.008233863863843935, "minimum": 0.11667598421618257, "maximum": 0.14019461254656782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9959602774375355, "standard_deviation": 0.001067279500440421, "minimum": 0.9942015746422218, "maximum": 0.9969869280646132, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7699588069732821, "standard_deviation": 0.02093026042415955, "minimum": 0.7382304616419004, "maximum": 0.7963225509826156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.580262758634424, "standard_deviation": 0.2879711176720782, "minimum": 0.34703873894297654, "maximum": 1.259874421772089, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07443942505666534, "standard_deviation": 0.009990294343693121, "minimum": 0.06404654393234618, "maximum": 0.08935366888317933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.989875017827972, "standard_deviation": 0.0009749537904176101, "minimum": 0.9881297762049229, "maximum": 0.9909594087577412, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3915195681726953, "standard_deviation": 0.02864990845001479, "minimum": 1.342616851709013, "maximum": 1.4218291431637835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6148440787364207, "standard_deviation": 0.24697755181301892, "minimum": 0.30265371440099775, "maximum": 1.0456061314678948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1105127374107052, "standard_deviation": 0.005364266854266612, "minimum": 0.10435208100371825, "maximum": 0.1200584406947387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964023830903344, "standard_deviation": 0.0009960974296616357, "minimum": 0.9945287916029529, "maximum": 0.9977042876162917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7845315699602442, "standard_deviation": 0.03290199376583757, "minimum": 0.7273524503351793, "maximum": 0.8179378895237647, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5427346874434131, "standard_deviation": 0.28583547540791676, "minimum": 0.12195958314828292, "maximum": 0.8580836375022582, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06327014881786455, "standard_deviation": 0.008398997644698399, "minimum": 0.05062299794742226, "maximum": 0.07813011613525037, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.7447635530694978, "standard_deviation": 0.08102736191657124, "minimum": 0.6020451718937936, "maximum": 0.8652803831305297, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.605193460883642, "standard_deviation": 0.00840321091564089, "minimum": 5.592366959538187, "maximum": 5.617899318823386, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5695681221901662, "standard_deviation": 0.014483673241147417, "minimum": 0.5526503980069762, "maximum": 0.5996823750254727, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07046321981385065, "standard_deviation": 0.048334820211799516, "minimum": -0.15085201668386317, "maximum": 0.016763726936550545, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9929412416899092, "standard_deviation": 0.0008908043872971413, "minimum": 0.9918853216235541, "maximum": 0.9944281833574851, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8228241737496731, "standard_deviation": 0.02594111351361526, "minimum": 0.7809808625502566, "maximum": 0.8561591710719134, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3463696787063191, "standard_deviation": 0.12931518749699597, "minimum": 0.13704341174763707, "maximum": 0.5941234689703974, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08757699220286014, "standard_deviation": 0.005652161040431992, "minimum": 0.07791561686822142, "maximum": 0.09381156797243081, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997953621366836, "standard_deviation": 0.000418990347312954, "minimum": 0.9970804493150898, "maximum": 0.998344425638315, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.412133917173206, "standard_deviation": 0.01993993809174272, "minimum": 0.3911381955335379, "maximum": 0.45595763606544615, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.39543944664141994, "standard_deviation": 0.13451665165335577, "minimum": 0.24542588049470537, "maximum": 0.6545086755517154, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04595252002475515, "standard_deviation": 0.00419914876499965, "minimum": 0.04138914436780285, "maximum": 0.05414291702223283, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141083-0010", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.1818771362304688, "predicted_mos": 2.1818771362304688, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "59a9c2dc846ced58e35de2a21c0815d1adcb7811f41b3be9a9115087919511a9", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.086341000162065, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8224810849743631, "全帯域_10ms_rms_envelope_nrmse": 0.08090252224902479, "全帯域_10ms_envelope_correlation": 0.9939668706408568, "全帯域_10ms_modulation_mae_db": 0.09534986756745124, "全帯域_40ms_log_rms_mae_db": 0.3787510433710172, "全帯域_40ms_rms_envelope_nrmse": 0.04384117701461851, "全帯域_40ms_envelope_correlation": 0.9980980907985432, "全帯域_40ms_modulation_mae_db": 0.07897519156493303, "300–1000 Hz_10ms_log_rms_mae_db": 1.5662142706711164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11377632232238115, "300–1000 Hz_10ms_envelope_correlation": 0.9892602550069806, "300–1000 Hz_10ms_modulation_mae_db": 0.601517835182028, "300–1000 Hz_40ms_log_rms_mae_db": 0.8380273498002625, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06922578370594637, "300–1000 Hz_40ms_envelope_correlation": 0.9956713027834754, "300–1000 Hz_40ms_modulation_mae_db": 0.41560408265526766, "1000–3000 Hz_10ms_log_rms_mae_db": 1.367188073544547, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12514066774108745, "1000–3000 Hz_10ms_envelope_correlation": 0.9883135509957535, "1000–3000 Hz_10ms_modulation_mae_db": 0.23823018051911818, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6796405020588935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0669971436174954, "1000–3000 Hz_40ms_envelope_correlation": 0.996543958640188, "1000–3000 Hz_40ms_modulation_mae_db": 0.37755095505930186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5151924108871813, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.161715814758011, "3000–6000 Hz_10ms_envelope_correlation": 0.9811700003657503, "3000–6000 Hz_10ms_modulation_mae_db": 1.0077142713537612, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7819574996213938, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09023679818276724, "3000–6000 Hz_40ms_envelope_correlation": 0.9937728618938331, "3000–6000 Hz_40ms_modulation_mae_db": 0.6559451591641033, "mr_stft_db": 5.717909728562397, "band_mae_db": 0.6576836406248517, "rms_gain_error_db": -0.050793409802089955, "output_peak": 0.6173433862923521}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8108791109432053, "全帯域_10ms_rms_envelope_nrmse": 0.08164861267197966, "全帯域_10ms_envelope_correlation": 0.9939021855521045, "全帯域_10ms_modulation_mae_db": 0.20701330633229306, "全帯域_40ms_log_rms_mae_db": 0.41381125728358525, "全帯域_40ms_rms_envelope_nrmse": 0.043176073818538295, "全帯域_40ms_envelope_correlation": 0.9981691530722205, "全帯域_40ms_modulation_mae_db": 0.2087424592821521, "300–1000 Hz_10ms_log_rms_mae_db": 1.5517355048742918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11646764789653874, "300–1000 Hz_10ms_envelope_correlation": 0.9887788697471697, "300–1000 Hz_10ms_modulation_mae_db": 0.9410216499859855, "300–1000 Hz_40ms_log_rms_mae_db": 0.8966697043612974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06145592478491953, "300–1000 Hz_40ms_envelope_correlation": 0.996595449962992, "300–1000 Hz_40ms_modulation_mae_db": 0.8715681619325036, "1000–3000 Hz_10ms_log_rms_mae_db": 1.36426784941567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1288987247727267, "1000–3000 Hz_10ms_envelope_correlation": 0.9877062205287882, "1000–3000 Hz_10ms_modulation_mae_db": 0.37047151788161253, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7463527580634457, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0693522410658072, "1000–3000 Hz_40ms_envelope_correlation": 0.9962355034935476, "1000–3000 Hz_40ms_modulation_mae_db": 0.47870711212523004, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5075466031460074, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12621607134730536, "3000–6000 Hz_10ms_envelope_correlation": 0.9885797165678191, "3000–6000 Hz_10ms_modulation_mae_db": 0.19715965804362967, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8025486129452009, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06799110478584772, "3000–6000 Hz_40ms_envelope_correlation": 0.9964592985802326, "3000–6000 Hz_40ms_modulation_mae_db": 0.19412137140414054, "mr_stft_db": 5.70111756568199, "band_mae_db": 0.7240878476395732, "rms_gain_error_db": 0.028203931002087654, "output_peak": 0.5557290100266886}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8294352277301912, "全帯域_10ms_rms_envelope_nrmse": 0.09331011407090407, "全帯域_10ms_envelope_correlation": 0.9919720151692895, "全帯域_10ms_modulation_mae_db": 0.15063932692637652, "全帯域_40ms_log_rms_mae_db": 0.4433452133646333, "全帯域_40ms_rms_envelope_nrmse": 0.05423830413938996, "全帯域_40ms_envelope_correlation": 0.9970597922904368, "全帯域_40ms_modulation_mae_db": 0.36763059620769306, "300–1000 Hz_10ms_log_rms_mae_db": 1.5780590109932908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11454415040388485, "300–1000 Hz_10ms_envelope_correlation": 0.9891151978776055, "300–1000 Hz_10ms_modulation_mae_db": 0.3447384919211052, "300–1000 Hz_40ms_log_rms_mae_db": 0.9071570558115798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06146191971089568, "300–1000 Hz_40ms_envelope_correlation": 0.9965884555678218, "300–1000 Hz_40ms_modulation_mae_db": 0.36383158430426144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3574214458224658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14858458278873918, "1000–3000 Hz_10ms_envelope_correlation": 0.9835570946769494, "1000–3000 Hz_10ms_modulation_mae_db": 0.48667919356651135, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7548942475760979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08885602477995144, "1000–3000 Hz_40ms_envelope_correlation": 0.9937744760902363, "1000–3000 Hz_40ms_modulation_mae_db": 0.6143217496578964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5316279467059641, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13720418981967597, "3000–6000 Hz_10ms_envelope_correlation": 0.9867797944756924, "3000–6000 Hz_10ms_modulation_mae_db": 0.7437201582798943, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7930966989134849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0716133144086953, "3000–6000 Hz_40ms_envelope_correlation": 0.996238098689197, "3000–6000 Hz_40ms_modulation_mae_db": 0.8689936428530309, "mr_stft_db": 5.709878662374895, "band_mae_db": 0.6887466142039965, "rms_gain_error_db": -0.01867028771173409, "output_peak": 0.5595641481370512}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8162674454924642, "全帯域_10ms_rms_envelope_nrmse": 0.08755865550540215, "全帯域_10ms_envelope_correlation": 0.9930963362635178, "全帯域_10ms_modulation_mae_db": 0.5435286791469841, "全帯域_40ms_log_rms_mae_db": 0.386310808177666, "全帯域_40ms_rms_envelope_nrmse": 0.04565235179797488, "全帯域_40ms_envelope_correlation": 0.9980266486556191, "全帯域_40ms_modulation_mae_db": 0.4683470304088481, "300–1000 Hz_10ms_log_rms_mae_db": 1.5369926780593566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11058539960566323, "300–1000 Hz_10ms_envelope_correlation": 0.9898507375633372, "300–1000 Hz_10ms_modulation_mae_db": 0.7408720486080002, "300–1000 Hz_40ms_log_rms_mae_db": 0.8262675941067122, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047475966322793085, "300–1000 Hz_40ms_envelope_correlation": 0.9980401439007652, "300–1000 Hz_40ms_modulation_mae_db": 0.4478729723458163, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3196538440940093, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1377708760939564, "1000–3000 Hz_10ms_envelope_correlation": 0.9866721274208436, "1000–3000 Hz_10ms_modulation_mae_db": 1.3674434996745604, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7065514280746564, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08246566515419618, "1000–3000 Hz_40ms_envelope_correlation": 0.9952005126514527, "1000–3000 Hz_40ms_modulation_mae_db": 1.4153977087379925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5624644598068762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1432756491680062, "3000–6000 Hz_10ms_envelope_correlation": 0.9863083411389932, "3000–6000 Hz_10ms_modulation_mae_db": 0.8865211265914096, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8255181654444625, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0797633929691971, "3000–6000 Hz_40ms_envelope_correlation": 0.9959307585115811, "3000–6000 Hz_40ms_modulation_mae_db": 0.6479456117816361, "mr_stft_db": 5.700019920688891, "band_mae_db": 0.7597001437414117, "rms_gain_error_db": 0.07736506439545712, "output_peak": 0.5769508961732742}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8338627474631597, "全帯域_10ms_rms_envelope_nrmse": 0.07395578638251865, "全帯域_10ms_envelope_correlation": 0.9952014992570681, "全帯域_10ms_modulation_mae_db": 0.3700996399672323, "全帯域_40ms_log_rms_mae_db": 0.406253175130316, "全帯域_40ms_rms_envelope_nrmse": 0.03811147255379077, "全帯域_40ms_envelope_correlation": 0.9988795184422706, "全帯域_40ms_modulation_mae_db": 0.2382082320139085, "300–1000 Hz_10ms_log_rms_mae_db": 1.4969713916819467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11477090936118173, "300–1000 Hz_10ms_envelope_correlation": 0.9891224764527041, "300–1000 Hz_10ms_modulation_mae_db": 0.33285417540641987, "300–1000 Hz_40ms_log_rms_mae_db": 0.8868342140296619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06874775361651615, "300–1000 Hz_40ms_envelope_correlation": 0.9959131239768816, "300–1000 Hz_40ms_modulation_mae_db": 0.3271379255892074, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4121833713782264, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11754151276988883, "1000–3000 Hz_10ms_envelope_correlation": 0.98987890232721, "1000–3000 Hz_10ms_modulation_mae_db": 0.45671542173040486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7610519574665111, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06379207367437635, "1000–3000 Hz_40ms_envelope_correlation": 0.9971500391756721, "1000–3000 Hz_40ms_modulation_mae_db": 0.3912165341460203, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5670511122467172, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12338792005384594, "3000–6000 Hz_10ms_envelope_correlation": 0.9892389722086621, "3000–6000 Hz_10ms_modulation_mae_db": 0.17567423759971024, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7827699198966336, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06409040608327791, "3000–6000 Hz_40ms_envelope_correlation": 0.9971984152233977, "3000–6000 Hz_40ms_modulation_mae_db": 0.19445001383357619, "mr_stft_db": 5.70833192503994, "band_mae_db": 0.8023814474653235, "rms_gain_error_db": -0.15224091918282806, "output_peak": 0.5339435435186544}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8172630309984356, "全帯域_10ms_rms_envelope_nrmse": 0.08179896224699415, "全帯域_10ms_envelope_correlation": 0.9938798670180774, "全帯域_10ms_modulation_mae_db": 0.48392842657043306, "全帯域_40ms_log_rms_mae_db": 0.3832271493015747, "全帯域_40ms_rms_envelope_nrmse": 0.03725555235477382, "全帯域_40ms_envelope_correlation": 0.9987163684955943, "全帯域_40ms_modulation_mae_db": 0.3426114975768848, "300–1000 Hz_10ms_log_rms_mae_db": 1.5151234651789744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12431750795531185, "300–1000 Hz_10ms_envelope_correlation": 0.9871744816666135, "300–1000 Hz_10ms_modulation_mae_db": 0.6089979484372884, "300–1000 Hz_40ms_log_rms_mae_db": 0.8809592768488402, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07205055700221115, "300–1000 Hz_40ms_envelope_correlation": 0.995427177482948, "300–1000 Hz_40ms_modulation_mae_db": 0.4773143261564197, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3018275749915464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12249901120130512, "1000–3000 Hz_10ms_envelope_correlation": 0.9887795133714615, "1000–3000 Hz_10ms_modulation_mae_db": 0.3754371672009526, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6791514064426332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055680660180643056, "1000–3000 Hz_40ms_envelope_correlation": 0.9976006327273675, "1000–3000 Hz_40ms_modulation_mae_db": 0.23825499298248576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.545730323585041, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12934052182509406, "3000–6000 Hz_10ms_envelope_correlation": 0.9880100503065496, "3000–6000 Hz_10ms_modulation_mae_db": 0.33157816267970025, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8133521910245682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06729568395549962, "3000–6000 Hz_40ms_envelope_correlation": 0.9966582647757443, "3000–6000 Hz_40ms_modulation_mae_db": 0.3851323944065852, "mr_stft_db": 5.708606022790817, "band_mae_db": 0.6951095676621329, "rms_gain_error_db": -0.0911597409233898, "output_peak": 0.6024800360137387}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8726211739902057, "全帯域_10ms_rms_envelope_nrmse": 0.10013841967359781, "全帯域_10ms_envelope_correlation": 0.9909033385874652, "全帯域_10ms_modulation_mae_db": 0.4178735442200649, "全帯域_40ms_log_rms_mae_db": 0.44677686599452887, "全帯域_40ms_rms_envelope_nrmse": 0.04794885923186221, "全帯域_40ms_envelope_correlation": 0.9980158008358222, "全帯域_40ms_modulation_mae_db": 0.19827582976900984, "300–1000 Hz_10ms_log_rms_mae_db": 1.5385263084174283, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1131197531879278, "300–1000 Hz_10ms_envelope_correlation": 0.9899373394816238, "300–1000 Hz_10ms_modulation_mae_db": 0.7676881137270661, "300–1000 Hz_40ms_log_rms_mae_db": 0.9180600594292005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07219965668363541, "300–1000 Hz_40ms_envelope_correlation": 0.996109513143574, "300–1000 Hz_40ms_modulation_mae_db": 0.7678631198621492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3748571131744771, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16682602157653914, "1000–3000 Hz_10ms_envelope_correlation": 0.9792548607295917, "1000–3000 Hz_10ms_modulation_mae_db": 0.9841135557394748, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7481371707363321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08580160230486991, "1000–3000 Hz_40ms_envelope_correlation": 0.9942096846856515, "1000–3000 Hz_40ms_modulation_mae_db": 1.1343091804853176, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5403556309788753, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14018229264729873, "3000–6000 Hz_10ms_envelope_correlation": 0.9860234391511555, "3000–6000 Hz_10ms_modulation_mae_db": 0.15156422715062667, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8320150687234159, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07890971406514667, "3000–6000 Hz_40ms_envelope_correlation": 0.9955876221596426, "3000–6000 Hz_40ms_modulation_mae_db": 0.33867753425800523, "mr_stft_db": 5.71949519312232, "band_mae_db": 0.8050105597989067, "rms_gain_error_db": -0.15947636262311604, "output_peak": 0.5634238655218788}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8219854874415395, "全帯域_10ms_rms_envelope_nrmse": 0.09650649444950046, "全帯域_10ms_envelope_correlation": 0.9914155627720188, "全帯域_10ms_modulation_mae_db": 0.729149014684842, "全帯域_40ms_log_rms_mae_db": 0.4284670111697127, "全帯域_40ms_rms_envelope_nrmse": 0.05149040730208563, "全帯域_40ms_envelope_correlation": 0.9973432809788241, "全帯域_40ms_modulation_mae_db": 0.619030572935149, "300–1000 Hz_10ms_log_rms_mae_db": 1.5106734551842864, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12929400385540177, "300–1000 Hz_10ms_envelope_correlation": 0.9863471062689174, "300–1000 Hz_10ms_modulation_mae_db": 1.0255596851861257, "300–1000 Hz_40ms_log_rms_mae_db": 0.8823711858532899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0759477615647461, "300–1000 Hz_40ms_envelope_correlation": 0.9949076311390482, "300–1000 Hz_40ms_modulation_mae_db": 0.9850709932348097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3406006551608642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1515107833824686, "1000–3000 Hz_10ms_envelope_correlation": 0.9829094951612984, "1000–3000 Hz_10ms_modulation_mae_db": 0.8659649111075423, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7266956367276216, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0901651716382329, "1000–3000 Hz_40ms_envelope_correlation": 0.9935807750854633, "1000–3000 Hz_40ms_modulation_mae_db": 1.088839108712402, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5357922334404575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12281670235433871, "3000–6000 Hz_10ms_envelope_correlation": 0.9891785691861023, "3000–6000 Hz_10ms_modulation_mae_db": 0.5928972239877012, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8408731250037759, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06788101234986557, "3000–6000 Hz_40ms_envelope_correlation": 0.9965294374866028, "3000–6000 Hz_40ms_modulation_mae_db": 0.24230309615855852, "mr_stft_db": 5.694118382908947, "band_mae_db": 0.6331635443534718, "rms_gain_error_db": -0.011083331165926337, "output_peak": 0.6097209679539782}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85120, "output_samples": 85120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.497387885965725, "spectral_convergence": 0.35077592385568035, "spectral_convergence_status": "ok", "amplitude_floor": 1.1374446115550809e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.722003109263566, "spectral_convergence": 0.375116371772414, "spectral_convergence_status": "ok", "amplitude_floor": 9.193083193108671e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.934338190457901, "spectral_convergence": 0.377975308867465, "spectral_convergence_status": "ok", "amplitude_floor": 7.3161107874683e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.1849686170678433e-06, "output_power": 4.911030756341793e-06, "delta_db": 3.5172745046448566, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004950757600852583, "output_power": 0.0004882968396926196, "delta_db": -0.05987749417464248, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008267786751374999, "output_power": 0.0008367256963574641, "delta_db": 0.05193840171321234, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007700805582515175, "output_power": 0.0007578437033175815, "delta_db": -0.06956512616994054, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006256626820185013, "output_power": 0.0006012976930829824, "delta_db": -0.17250714375829926, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00016725866300063496, "output_power": 0.00016439730870043614, "delta_db": -0.07493917328815906, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6576836406248517, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.050793409802089955, "pitch": null, "output_rms_dbfs": -25.4462634327136, "output_peak": 0.6173433862923521, "reference_rms_dbfs": -25.39547002291151, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.717909728562397}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.0002036923290800657, "log_rms_mae_db": 0.8224810849743631, "rms_envelope_nrmse": 0.08090252224902479, "envelope_correlation": 0.9939668706408568, "rms_gain_error_db": -0.05079208073000972, "reference_mean_power": 0.0028897508304992817, "output_mean_power": 0.002856151164416561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 1.5709970652253345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007599193486798772, "output_power": 0.0007550020064765108, "reference_active": true, "delta_db": -0.028193966340939555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014055064928452022, "output_power": 0.00013258935384740498, "reference_active": true, "delta_db": -0.253242020526726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3577241130570316e-05, "output_power": 2.3602301115416894e-05, "reference_active": true, "delta_db": 0.00461361583468819}], "active_band_count": 3, "active_band_mae_db": 0.09534986756745124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.0001938989159680862, "log_rms_mae_db": 0.3787510433710172, "rms_envelope_nrmse": 0.04384117701461851, "envelope_correlation": 0.9980980907985432, "rms_gain_error_db": -0.050795506986892405, "reference_mean_power": 0.002906120097080002, "output_mean_power": 0.002872327836462954}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 1.4139945205650041e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006810740061577038, "output_power": 0.0006819646944204855, "reference_active": true, "delta_db": 0.0056758630920228405}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.5174236911387e-05, "output_power": 6.378440947350159e-05, "reference_active": true, "delta_db": -0.09361415847189282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3046113483012314e-06, "output_power": 1.263914210187332e-06, "reference_active": true, "delta_db": -0.13763555313088344}], "active_band_count": 3, "active_band_mae_db": 0.07897519156493303}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013759847516455276, "log_rms_mae_db": 1.5662142706711164, "rms_envelope_nrmse": 0.11377632232238115, "envelope_correlation": 0.9892602550069806, "rms_gain_error_db": -0.016763539877145305, "reference_mean_power": 0.001146104301837711, "output_mean_power": 0.0011416889262124148}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 6.878091269334419e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031946504747131925, "output_power": 0.00031424764579958625, "reference_active": true, "delta_db": -0.07151316294096559}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.126588772977069e-05, "output_power": 6.317406543520237e-05, "reference_active": true, "delta_db": -0.523428734266836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2614620971957042e-05, "output_power": 1.6666153088362255e-05, "reference_active": true, "delta_db": 1.2096116083382824}], "active_band_count": 3, "active_band_mae_db": 0.601517835182028}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00012449215623942515, "log_rms_mae_db": 0.8380273498002625, "rms_envelope_nrmse": 0.06922578370594637, "envelope_correlation": 0.9956713027834754, "rms_gain_error_db": -0.01676370225301302, "reference_mean_power": 0.001152608697675057, "output_mean_power": 0.0011481682208866388}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 6.232509709606391e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00029975403822873686, "output_power": 0.0002950943205707822, "reference_active": true, "delta_db": -0.06804191457857522}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2561293827419036e-05, "output_power": 2.9134858448641957e-05, "reference_active": true, "delta_db": -0.4828874089326704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.743814177835147e-07, "output_power": 9.089567423538028e-07, "reference_active": true, "delta_db": 0.6958829244545575}], "active_band_count": 3, "active_band_mae_db": 0.41560408265526766}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013997823549865974, "log_rms_mae_db": 1.367188073544547, "rms_envelope_nrmse": 0.12514066774108745, "envelope_correlation": 0.9883135509957535, "rms_gain_error_db": -0.11647074731456965, "reference_mean_power": 0.0010964326399058018, "output_mean_power": 0.0010674188819424942}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 8.778805813988002e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045662233680262583, "output_power": 0.0004484736707108839, "reference_active": true, "delta_db": -0.07820200974026247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.737497372188522e-05, "output_power": 4.426619533030418e-05, "reference_active": true, "delta_db": -0.2947678601196635}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.406352772002978e-05, "output_power": 1.5214803251852874e-05, "reference_active": true, "delta_db": 0.34172067169742854}], "active_band_count": 3, "active_band_mae_db": 0.23823018051911818}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013340521153934576, "log_rms_mae_db": 0.6796405020588935, "rms_envelope_nrmse": 0.0669971436174954, "envelope_correlation": 0.996543958640188, "rms_gain_error_db": -0.11647937304664178, "reference_mean_power": 0.0011026403148293462, "output_mean_power": 0.001073460157560053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 8.248821386475387e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042363838931095744, "output_power": 0.0004178001463126014, "reference_active": true, "delta_db": -0.06026720431444441}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.073594132211124e-05, "output_power": 2.099647353231308e-05, "reference_active": true, "delta_db": 0.05422603586210901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.366449484579509e-07, "output_power": 9.312669223311736e-07, "reference_active": true, "delta_db": 1.018159625001352}], "active_band_count": 3, "active_band_mae_db": 0.37755095505930186}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00010989278407704884, "log_rms_mae_db": 1.5151924108871813, "rms_envelope_nrmse": 0.161715814758011, "envelope_correlation": 0.9811700003657503, "rms_gain_error_db": -0.08042089740666042, "reference_mean_power": 0.0003075575386675287, "output_mean_power": 0.00030191471940095463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 2.696003640873154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014789902573029864, "output_power": 0.00013611354179307808, "reference_active": true, "delta_db": -0.3606397828191834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.17770472077717e-05, "output_power": 3.693836711803395e-05, "reference_active": true, "delta_db": 0.6536415412889812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.120339140119136e-06, "output_power": 4.955505331482612e-06, "reference_active": true, "delta_db": 2.008861489953119}], "active_band_count": 3, "active_band_mae_db": 1.0077142713537612}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010174719078202204, "log_rms_mae_db": 0.7819574996213938, "rms_envelope_nrmse": 0.09023679818276724, "envelope_correlation": 0.9937728618938331, "rms_gain_error_db": -0.08042536413322394, "reference_mean_power": 0.000309290771080255, "output_mean_power": 0.00030361583958465297}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.5256223542383704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014158879062551878, "output_power": 0.00013416672988037707, "reference_active": true, "delta_db": -0.2338403765936906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.501151157810192e-05, "output_power": 1.8085870088253783e-05, "reference_active": true, "delta_db": 0.8091498140371177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.631467090177325e-07, "output_power": 3.255985571298204e-07, "reference_active": true, "delta_db": 0.9248452868615019}], "active_band_count": 3, "active_band_mae_db": 0.6559451591641033}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8630141708895033, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cb2e99240f8de005ccaac2c232fbc4ba23808768fd91aa13fc2bb173212ea9fa", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.985883970651487, "standard_deviation": 0.003633086993054826, "minimum": 0.9792548607295917, "maximum": 0.98987890232721, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3547499909477259, "standard_deviation": 0.03420118437671612, "minimum": 1.3018275749915464, "maximum": 1.4121833713782264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6431319309275221, "standard_deviation": 0.3890883929902003, "minimum": 0.23823018051911818, "maximum": 1.3674434996745604, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13734652254083893, "standard_deviation": 0.01702593150876293, "minimum": 0.11754151276988883, "maximum": 0.16682602157653914, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955369478186974, "standard_deviation": 0.0015661893862702531, "minimum": 0.9935807750854633, "maximum": 0.9976006327273675, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.725309388393274, "standard_deviation": 0.033137625199715526, "minimum": 0.6791514064426332, "maximum": 0.7610519574665111, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7173246677383308, "standard_deviation": 0.4339278111731938, "minimum": 0.23825499298248576, "maximum": 1.4153977087379925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07538882280194656, "standard_deviation": 0.013027282095282528, "minimum": 0.055680660180643056, "maximum": 0.0901651716382329, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9869111104250906, "standard_deviation": 0.0026371278947741068, "minimum": 0.9811700003657503, "maximum": 0.9892389722086621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5382200900996401, "standard_deviation": 0.020703585254337777, "minimum": 1.5075466031460074, "maximum": 1.5670511122467172, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5108536332108041, "standard_deviation": 0.3424558861149939, "minimum": 0.15156422715062667, "maximum": 1.0077142713537612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13551739524669698, "standard_deviation": 0.013116262492666815, "minimum": 0.12281670235433871, "maximum": 0.161715814758011, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9960468446650289, "standard_deviation": 0.0010375039012816242, "minimum": 0.9937728618938331, "maximum": 0.9971984152233977, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.809016410196617, "standard_deviation": 0.022523727163002195, "minimum": 0.7819574996213938, "maximum": 0.8408731250037759, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4409461029824545, "standard_deviation": 0.2527036467738817, "minimum": 0.19412137140414054, "maximum": 0.8689936428530309, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07347267835003714, "standard_deviation": 0.008792237370499278, "minimum": 0.06409040608327791, "maximum": 0.09023679818276724, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.988698308008119, "standard_deviation": 0.0012755659732928107, "minimum": 0.9863471062689174, "maximum": 0.9899373394816238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5367870106325865, "standard_deviation": 0.028112978892098493, "minimum": 1.4969713916819467, "maximum": 1.5780590109932908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6704062435567524, "standard_deviation": 0.2513714758023239, "minimum": 0.33285417540641987, "maximum": 1.0255596851861257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11710946182353639, "standard_deviation": 0.006352407313472994, "minimum": 0.11058539960566323, "maximum": 0.12929400385540177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961565997446882, "standard_deviation": 0.0009498735032957997, "minimum": 0.9949076311390482, "maximum": 0.9980401439007652, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8795433050301056, "standard_deviation": 0.03199198182789315, "minimum": 0.8262675941067122, "maximum": 0.9180600594292005, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5820328957600545, "standard_deviation": 0.2535603400727596, "minimum": 0.3271379255892074, "maximum": 0.9850709932348097, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06607066542395792, "standard_deviation": 0.009073520664654885, "minimum": 0.047475966322793085, "maximum": 0.0759477615647461, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.7207354206862084, "standard_deviation": 0.0639443840349193, "minimum": 0.6331635443534718, "maximum": 0.8050105597989067, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.707434675146274, "standard_deviation": 0.008746763636322152, "minimum": 5.694118382908947, "maximum": 5.71949519312232, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.577394481704702, "standard_deviation": 0.02961870571599046, "minimum": 0.5339435435186544, "maximum": 0.6173433862923521, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04723188200144243, "standard_deviation": 0.08358529084572014, "minimum": -0.15947636262311604, "maximum": 0.07736506439545712, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930422094075497, "standard_deviation": 0.0014797564171331974, "minimum": 0.9909033385874652, "maximum": 0.9952014992570681, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8280994136291955, "standard_deviation": 0.019420514761623847, "minimum": 0.8108791109432053, "maximum": 0.8726211739902057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37469772567695964, "standard_deviation": 0.2152446127474808, "minimum": 0.09534986756745124, "maximum": 0.729149014684842, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08697744590624021, "standard_deviation": 0.0089931399106458, "minimum": 0.07395578638251865, "maximum": 0.10013841967359781, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980385816961663, "standard_deviation": 0.0006128729683451491, "minimum": 0.9970597922904368, "maximum": 0.9988795184422706, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41086781547412926, "standard_deviation": 0.026945637896324728, "minimum": 0.3787510433710172, "maximum": 0.44677686599452887, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3152276762198223, "standard_deviation": 0.17154830368531862, "minimum": 0.07897519156493303, "maximum": 0.619030572935149, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04521427477662926, "standard_deviation": 0.005956071711779228, "minimum": 0.03725555235477382, "maximum": 0.05423830413938996, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141083-0010", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.9899771213531494, "predicted_mos": 1.9899771213531494, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "4a26d34186a741a4f0fc2ef5e04148a0211c12aabf5586f374173ab018c767c1", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.052381599787623, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8168236574690277, "全帯域_10ms_rms_envelope_nrmse": 0.07938813693738996, "全帯域_10ms_envelope_correlation": 0.9941858627906776, "全帯域_10ms_modulation_mae_db": 0.10505351598032686, "全帯域_40ms_log_rms_mae_db": 0.3944909238590969, "全帯域_40ms_rms_envelope_nrmse": 0.047365416213206085, "全帯域_40ms_envelope_correlation": 0.9977602569233098, "全帯域_40ms_modulation_mae_db": 0.11061153316058224, "300–1000 Hz_10ms_log_rms_mae_db": 1.5019458075045047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11626387613801517, "300–1000 Hz_10ms_envelope_correlation": 0.9887992339460059, "300–1000 Hz_10ms_modulation_mae_db": 0.35273094108129865, "300–1000 Hz_40ms_log_rms_mae_db": 0.8741224708454666, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060069584485932845, "300–1000 Hz_40ms_envelope_correlation": 0.9967420340265171, "300–1000 Hz_40ms_modulation_mae_db": 0.27727452702137007, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2959682799938708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12587754215192257, "1000–3000 Hz_10ms_envelope_correlation": 0.9881756085926016, "1000–3000 Hz_10ms_modulation_mae_db": 0.9013355069594206, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6497602630321222, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07783543114952068, "1000–3000 Hz_40ms_envelope_correlation": 0.9952556681897263, "1000–3000 Hz_40ms_modulation_mae_db": 1.124711036656282, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5007955034005598, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11807409297133042, "3000–6000 Hz_10ms_envelope_correlation": 0.9900119996846795, "3000–6000 Hz_10ms_modulation_mae_db": 0.1897958907994324, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7658307579878925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05966731789991296, "3000–6000 Hz_40ms_envelope_correlation": 0.9972762747676943, "3000–6000 Hz_40ms_modulation_mae_db": 0.3617272588901052, "mr_stft_db": 5.687963304671974, "band_mae_db": 0.731894870650443, "rms_gain_error_db": -0.03416984591840243, "output_peak": 0.6226966086235206}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8096379638264362, "全帯域_10ms_rms_envelope_nrmse": 0.07967833776766733, "全帯域_10ms_envelope_correlation": 0.9941523609810985, "全帯域_10ms_modulation_mae_db": 0.3005043116438917, "全帯域_40ms_log_rms_mae_db": 0.40188382345501455, "全帯域_40ms_rms_envelope_nrmse": 0.042944312849532866, "全帯域_40ms_envelope_correlation": 0.9981776871257548, "全帯域_40ms_modulation_mae_db": 0.543096975097413, "300–1000 Hz_10ms_log_rms_mae_db": 1.566741847945142, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12212552005796143, "300–1000 Hz_10ms_envelope_correlation": 0.9876137122378386, "300–1000 Hz_10ms_modulation_mae_db": 0.6120089212576215, "300–1000 Hz_40ms_log_rms_mae_db": 0.9048405452922667, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0679909352231519, "300–1000 Hz_40ms_envelope_correlation": 0.9958458044429201, "300–1000 Hz_40ms_modulation_mae_db": 0.4132664975712265, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3181455373708482, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.114906567050753, "1000–3000 Hz_10ms_envelope_correlation": 0.9901312491284908, "1000–3000 Hz_10ms_modulation_mae_db": 0.7032250139704955, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7163458119171596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061342219114182986, "1000–3000 Hz_40ms_envelope_correlation": 0.9970246287215707, "1000–3000 Hz_40ms_modulation_mae_db": 0.9138146249039395, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4805482285859397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13462373538540828, "3000–6000 Hz_10ms_envelope_correlation": 0.9870295235046223, "3000–6000 Hz_10ms_modulation_mae_db": 0.10572449158742492, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7917291931625321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07376223534863949, "3000–6000 Hz_40ms_envelope_correlation": 0.9960181299617602, "3000–6000 Hz_40ms_modulation_mae_db": 0.25390386864012787, "mr_stft_db": 5.693749702573315, "band_mae_db": 0.7629709588110606, "rms_gain_error_db": -0.037642135615845965, "output_peak": 0.5779945781420834}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8045481822289932, "全帯域_10ms_rms_envelope_nrmse": 0.07825615838767964, "全帯域_10ms_envelope_correlation": 0.9945788505099012, "全帯域_10ms_modulation_mae_db": 0.6121126085800241, "全帯域_40ms_log_rms_mae_db": 0.39824255809030135, "全帯域_40ms_rms_envelope_nrmse": 0.041832626220787934, "全帯域_40ms_envelope_correlation": 0.99856769944526, "全帯域_40ms_modulation_mae_db": 0.49103329687155584, "300–1000 Hz_10ms_log_rms_mae_db": 1.5245168759309453, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1037442602318224, "300–1000 Hz_10ms_envelope_correlation": 0.9910906674342823, "300–1000 Hz_10ms_modulation_mae_db": 0.4255607243044546, "300–1000 Hz_40ms_log_rms_mae_db": 0.8811956412642986, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06644413206217041, "300–1000 Hz_40ms_envelope_correlation": 0.9960813786310775, "300–1000 Hz_40ms_modulation_mae_db": 0.5361526022323476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3631841459115566, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13017241694668275, "1000–3000 Hz_10ms_envelope_correlation": 0.9876634109433744, "1000–3000 Hz_10ms_modulation_mae_db": 0.21281905183664662, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7727686991685819, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07782720878580736, "1000–3000 Hz_40ms_envelope_correlation": 0.9957636116821756, "1000–3000 Hz_40ms_modulation_mae_db": 0.23401606034384959, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4599277439767773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11985182317175004, "3000–6000 Hz_10ms_envelope_correlation": 0.9899312028332302, "3000–6000 Hz_10ms_modulation_mae_db": 0.6856736656137298, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7896439811872591, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06696874092318374, "3000–6000 Hz_40ms_envelope_correlation": 0.99698538717637, "3000–6000 Hz_40ms_modulation_mae_db": 0.536822339218557, "mr_stft_db": 5.685647348348592, "band_mae_db": 0.7794512275355192, "rms_gain_error_db": -0.16264360303630565, "output_peak": 0.5685566753508314}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7832344435096374, "全帯域_10ms_rms_envelope_nrmse": 0.09172628681048715, "全帯域_10ms_envelope_correlation": 0.992362817393631, "全帯域_10ms_modulation_mae_db": 0.5251939761646067, "全帯域_40ms_log_rms_mae_db": 0.40490018838396646, "全帯域_40ms_rms_envelope_nrmse": 0.051120130254359554, "全帯域_40ms_envelope_correlation": 0.9976039562585879, "全帯域_40ms_modulation_mae_db": 0.3489803350386385, "300–1000 Hz_10ms_log_rms_mae_db": 1.4114847625179512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10852300577503197, "300–1000 Hz_10ms_envelope_correlation": 0.9902906202488337, "300–1000 Hz_10ms_modulation_mae_db": 0.8821303332415719, "300–1000 Hz_40ms_log_rms_mae_db": 0.7754621488437986, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057029049555885565, "300–1000 Hz_40ms_envelope_correlation": 0.997245788069158, "300–1000 Hz_40ms_modulation_mae_db": 0.7276901490244692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3753539184085242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14665662048154682, "1000–3000 Hz_10ms_envelope_correlation": 0.9840732681057703, "1000–3000 Hz_10ms_modulation_mae_db": 0.4677472794467373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7096427739920443, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07879137094616731, "1000–3000 Hz_40ms_envelope_correlation": 0.995500020660785, "1000–3000 Hz_40ms_modulation_mae_db": 0.40833025100275067, "3000–6000 Hz_10ms_log_rms_mae_db": 1.513900174869761, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12865640161495984, "3000–6000 Hz_10ms_envelope_correlation": 0.9881394494632502, "3000–6000 Hz_10ms_modulation_mae_db": 0.5141608857698908, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8080068541036839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.067640225287776, "3000–6000 Hz_40ms_envelope_correlation": 0.9964984800566301, "3000–6000 Hz_40ms_modulation_mae_db": 0.436984743713718, "mr_stft_db": 5.678373607098013, "band_mae_db": 0.776866602221366, "rms_gain_error_db": -0.14089751148775534, "output_peak": 0.5608070947666595}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8567693159311296, "全帯域_10ms_rms_envelope_nrmse": 0.08209744428135815, "全帯域_10ms_envelope_correlation": 0.994304231633882, "全帯域_10ms_modulation_mae_db": 0.47600932035374144, "全帯域_40ms_log_rms_mae_db": 0.4406583905454286, "全帯域_40ms_rms_envelope_nrmse": 0.05022515165918765, "全帯域_40ms_envelope_correlation": 0.9981255761611189, "全帯域_40ms_modulation_mae_db": 0.5929941189790767, "300–1000 Hz_10ms_log_rms_mae_db": 1.4576177664447614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10607128746732396, "300–1000 Hz_10ms_envelope_correlation": 0.9909759635683401, "300–1000 Hz_10ms_modulation_mae_db": 0.5387866431753499, "300–1000 Hz_40ms_log_rms_mae_db": 0.8930287264719133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06568684397756101, "300–1000 Hz_40ms_envelope_correlation": 0.9965784402435555, "300–1000 Hz_40ms_modulation_mae_db": 0.5808983865989767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4022477810579612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1235493630187985, "1000–3000 Hz_10ms_envelope_correlation": 0.9891479976534095, "1000–3000 Hz_10ms_modulation_mae_db": 0.4078144456429826, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7411529541917837, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07018934152472771, "1000–3000 Hz_40ms_envelope_correlation": 0.9969590578118982, "1000–3000 Hz_40ms_modulation_mae_db": 0.5459804690299567, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4888850134314928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12851139501829142, "3000–6000 Hz_10ms_envelope_correlation": 0.9882401835394091, "3000–6000 Hz_10ms_modulation_mae_db": 0.362975280264781, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7783694083389984, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06649482211128117, "3000–6000 Hz_40ms_envelope_correlation": 0.9968750768410404, "3000–6000 Hz_40ms_modulation_mae_db": 0.3514237660282274, "mr_stft_db": 5.6853389783660715, "band_mae_db": 0.8173009885129735, "rms_gain_error_db": -0.22033606872100314, "output_peak": 0.5510789556236113}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8085707575808656, "全帯域_10ms_rms_envelope_nrmse": 0.09006852427808283, "全帯域_10ms_envelope_correlation": 0.9925561548749605, "全帯域_10ms_modulation_mae_db": 0.4648910360047231, "全帯域_40ms_log_rms_mae_db": 0.3977347350368258, "全帯域_40ms_rms_envelope_nrmse": 0.04316859432553646, "全帯域_40ms_envelope_correlation": 0.9982362066134997, "全帯域_40ms_modulation_mae_db": 0.42788286735205167, "300–1000 Hz_10ms_log_rms_mae_db": 1.5143997012724544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12640151245475295, "300–1000 Hz_10ms_envelope_correlation": 0.9869483516402093, "300–1000 Hz_10ms_modulation_mae_db": 0.4333125982838444, "300–1000 Hz_40ms_log_rms_mae_db": 0.8510873425240396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06631551928416328, "300–1000 Hz_40ms_envelope_correlation": 0.9961312508882305, "300–1000 Hz_40ms_modulation_mae_db": 0.3720406935811908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3394633918601786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13546227794334903, "1000–3000 Hz_10ms_envelope_correlation": 0.986713459616182, "1000–3000 Hz_10ms_modulation_mae_db": 0.4250515187464255, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7356748574407069, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.070506689111295, "1000–3000 Hz_40ms_envelope_correlation": 0.9968562464189612, "1000–3000 Hz_40ms_modulation_mae_db": 0.45859991022775937, "3000–6000 Hz_10ms_log_rms_mae_db": 1.525054284493174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1393566321685965, "3000–6000 Hz_10ms_envelope_correlation": 0.98614641391388, "3000–6000 Hz_10ms_modulation_mae_db": 0.8434058854828258, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8390028065264753, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07679910018233516, "3000–6000 Hz_40ms_envelope_correlation": 0.9957672573430566, "3000–6000 Hz_40ms_modulation_mae_db": 0.8117135586104679, "mr_stft_db": 5.700487521565843, "band_mae_db": 0.7996724030904914, "rms_gain_error_db": -0.09633924723344445, "output_peak": 0.5998137317964151}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8225004287831655, "全帯域_10ms_rms_envelope_nrmse": 0.08918113980731854, "全帯域_10ms_envelope_correlation": 0.9926824856743184, "全帯域_10ms_modulation_mae_db": 0.42516223443521994, "全帯域_40ms_log_rms_mae_db": 0.41666818060991634, "全帯域_40ms_rms_envelope_nrmse": 0.04495557067696756, "全帯域_40ms_envelope_correlation": 0.9980413204643177, "全帯域_40ms_modulation_mae_db": 0.35694206801311507, "300–1000 Hz_10ms_log_rms_mae_db": 1.5319653072996136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13014447881780797, "300–1000 Hz_10ms_envelope_correlation": 0.9859066560115956, "300–1000 Hz_10ms_modulation_mae_db": 0.8727034356695754, "300–1000 Hz_40ms_log_rms_mae_db": 0.9008446131009523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06458519845597507, "300–1000 Hz_40ms_envelope_correlation": 0.9962623844393302, "300–1000 Hz_40ms_modulation_mae_db": 0.5518588967150729, "1000–3000 Hz_10ms_log_rms_mae_db": 1.345089276350755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12766006177232894, "1000–3000 Hz_10ms_envelope_correlation": 0.9878847001780873, "1000–3000 Hz_10ms_modulation_mae_db": 0.6066050253985807, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6831444597521137, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.067465373663204, "1000–3000 Hz_40ms_envelope_correlation": 0.9966138525170831, "1000–3000 Hz_40ms_modulation_mae_db": 0.7155428545262437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4510778058163643, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1336898301512564, "3000–6000 Hz_10ms_envelope_correlation": 0.9872871314523638, "3000–6000 Hz_10ms_modulation_mae_db": 0.41780307288958657, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7624126632668876, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06782852500392429, "3000–6000 Hz_40ms_envelope_correlation": 0.9965115208321554, "3000–6000 Hz_40ms_modulation_mae_db": 0.08494320229216436, "mr_stft_db": 5.671400313874682, "band_mae_db": 0.7536076395642981, "rms_gain_error_db": -0.07605728062606304, "output_peak": 0.5879250350287754}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8137795532381077, "全帯域_10ms_rms_envelope_nrmse": 0.07625022054494092, "全帯域_10ms_envelope_correlation": 0.9946453006577541, "全帯域_10ms_modulation_mae_db": 0.09925122544384914, "全帯域_40ms_log_rms_mae_db": 0.3930197823689001, "全帯域_40ms_rms_envelope_nrmse": 0.03520290468215658, "全帯域_40ms_envelope_correlation": 0.998786483325199, "全帯域_40ms_modulation_mae_db": 0.07864278015749934, "300–1000 Hz_10ms_log_rms_mae_db": 1.4913609375211563, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10516651008190048, "300–1000 Hz_10ms_envelope_correlation": 0.9908171840883921, "300–1000 Hz_10ms_modulation_mae_db": 0.4269057455650056, "300–1000 Hz_40ms_log_rms_mae_db": 0.8810255818912943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05837412995800538, "300–1000 Hz_40ms_envelope_correlation": 0.9969325055189816, "300–1000 Hz_40ms_modulation_mae_db": 0.3307035358545884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3022030286885753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1262674429652396, "1000–3000 Hz_10ms_envelope_correlation": 0.9881156474834929, "1000–3000 Hz_10ms_modulation_mae_db": 0.2996109900304244, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6751783762480298, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0647628002858608, "1000–3000 Hz_40ms_envelope_correlation": 0.9967219065758162, "1000–3000 Hz_40ms_modulation_mae_db": 0.130701238380826, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4926714458521364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13206967867861125, "3000–6000 Hz_10ms_envelope_correlation": 0.9874726794518782, "3000–6000 Hz_10ms_modulation_mae_db": 0.6098170991441371, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7799168921936166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06999867514478922, "3000–6000 Hz_40ms_envelope_correlation": 0.9963269679433644, "3000–6000 Hz_40ms_modulation_mae_db": 0.3808910326362611, "mr_stft_db": 5.687478955366924, "band_mae_db": 0.7451168557796012, "rms_gain_error_db": -0.04920549156920021, "output_peak": 0.5984828470786072}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85120, "output_samples": 85120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.4775621585721215, "spectral_convergence": 0.3535135980233847, "spectral_convergence_status": "ok", "amplitude_floor": 1.1374446115550809e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.70756582858282, "spectral_convergence": 0.3769890808306926, "spectral_convergence_status": "ok", "amplitude_floor": 9.193083193108671e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.878761926860978, "spectral_convergence": 0.3745252886295634, "spectral_convergence_status": "ok", "amplitude_floor": 7.3161107874683e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.1849686170678433e-06, "output_power": 5.387572141562753e-06, "delta_db": 3.9194789503548835, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004950757600852583, "output_power": 0.0004997691452633913, "delta_db": 0.040977772426846554, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008267786751374999, "output_power": 0.0008252828463610659, "delta_db": -0.007864481468121671, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007700805582515175, "output_power": 0.0007732742992915495, "delta_db": 0.017974169496821544, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006256626820185013, "output_power": 0.0005881116580394317, "delta_db": -0.2688046359146043, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00016725866300063496, "output_power": 0.00017258996565002843, "delta_db": 0.1362692142413806, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.731894870650443, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03416984591840243, "pitch": null, "output_rms_dbfs": -25.429639868829913, "output_peak": 0.6226966086235206, "reference_rms_dbfs": -25.39547002291151, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.687963304671974}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.0002036923290800657, "log_rms_mae_db": 0.8168236574690277, "rms_envelope_nrmse": 0.07938813693738996, "envelope_correlation": 0.9941858627906776, "rms_gain_error_db": -0.03416616523301763, "reference_mean_power": 0.0028897508304992817, "output_mean_power": 0.0028671062054258243}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 1.5709970652253345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007599193486798772, "output_power": 0.0007847940488385649, "reference_active": true, "delta_db": 0.13988198655447365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014055064928452022, "output_power": 0.0001352645061164819, "reference_active": true, "delta_db": -0.1664900506191861}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3577241130570316e-05, "output_power": 2.362500103649211e-05, "reference_active": true, "delta_db": 0.008788510767320873}], "active_band_count": 3, "active_band_mae_db": 0.10505351598032686}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.0001938989159680862, "log_rms_mae_db": 0.3944909238590969, "rms_envelope_nrmse": 0.047365416213206085, "envelope_correlation": 0.9977602569233098, "rms_gain_error_db": -0.03415963292643875, "reference_mean_power": 0.002906120097080002, "output_mean_power": 0.0028833515362895992}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 1.4139945205650041e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006810740061577038, "output_power": 0.0007091179053392137, "reference_active": true, "delta_db": 0.17524146183605138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.5174236911387e-05, "output_power": 6.288768476328352e-05, "reference_active": true, "delta_db": -0.1551034867165116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3046113483012314e-06, "output_power": 1.3041639370800748e-06, "reference_active": true, "delta_db": -0.0014896509291836867}], "active_band_count": 3, "active_band_mae_db": 0.11061153316058224}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013759847516455276, "log_rms_mae_db": 1.5019458075045047, "rms_envelope_nrmse": 0.11626387613801517, "envelope_correlation": 0.9887992339460059, "rms_gain_error_db": -0.004556581317417455, "reference_mean_power": 0.001146104301837711, "output_mean_power": 0.0011449024494036883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 6.878091269334419e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031946504747131925, "output_power": 0.0003235285532368231, "reference_active": true, "delta_db": 0.05489266529286301}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.126588772977069e-05, "output_power": 7.31508016726417e-05, "reference_active": true, "delta_db": 0.11337390662031974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2614620971957042e-05, "output_power": 1.5483417737019815e-05, "reference_active": true, "delta_db": 0.8899262513307132}], "active_band_count": 3, "active_band_mae_db": 0.35273094108129865}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00012449215623942515, "log_rms_mae_db": 0.8741224708454666, "rms_envelope_nrmse": 0.060069584485932845, "envelope_correlation": 0.9967420340265171, "rms_gain_error_db": -0.00455664414082074, "reference_mean_power": 0.001152608697675057, "output_mean_power": 0.0011514000078064765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 6.232509709606391e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00029975403822873686, "output_power": 0.00030770090717478803, "reference_active": true, "delta_db": 0.11363734109817669}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2561293827419036e-05, "output_power": 3.317952740887984e-05, "reference_active": true, "delta_db": 0.08168542568230472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.743814177835147e-07, "output_power": 8.966129442961096e-07, "reference_active": true, "delta_db": 0.6365008142836287}], "active_band_count": 3, "active_band_mae_db": 0.27727452702137007}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013997823549865974, "log_rms_mae_db": 1.2959682799938708, "rms_envelope_nrmse": 0.12587754215192257, "envelope_correlation": 0.9881756085926016, "rms_gain_error_db": -0.09481834538170694, "reference_mean_power": 0.0010964326399058018, "output_mean_power": 0.0010727539470010202}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 8.778805813988002e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045662233680262583, "output_power": 0.0004758758637541375, "reference_active": true, "delta_db": 0.17936525848985088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.737497372188522e-05, "output_power": 5.185208398035423e-05, "reference_active": true, "delta_db": 0.39217233668564055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.406352772002978e-05, "output_power": 2.2979531285086706e-05, "reference_active": true, "delta_db": 2.1324689257027702}], "active_band_count": 3, "active_band_mae_db": 0.9013355069594206}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013340521153934576, "log_rms_mae_db": 0.6497602630321222, "rms_envelope_nrmse": 0.07783543114952068, "envelope_correlation": 0.9952556681897263, "rms_gain_error_db": -0.09480470788233211, "reference_mean_power": 0.0011026403148293462, "output_mean_power": 0.0010788309479064172}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 8.248821386475387e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042363838931095744, "output_power": 0.00045119475557803875, "reference_active": true, "delta_db": 0.27368734856845356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.073594132211124e-05, "output_power": 2.367935685403838e-05, "reference_active": true, "delta_db": 0.5764614726471328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.366449484579509e-07, "output_power": 1.3172149279109241e-06, "reference_active": true, "delta_db": 2.52398428875326}], "active_band_count": 3, "active_band_mae_db": 1.124711036656282}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00010989278407704884, "log_rms_mae_db": 1.5007955034005598, "rms_envelope_nrmse": 0.11807409297133042, "envelope_correlation": 0.9900119996846795, "rms_gain_error_db": -0.017055275532895205, "reference_mean_power": 0.0003075575386675287, "output_mean_power": 0.00030635209111200555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 2.696003640873154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014789902573029864, "output_power": 0.00014897714829813935, "reference_active": true, "delta_db": 0.03154343671502774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.17770472077717e-05, "output_power": 2.970930175251263e-05, "reference_active": true, "delta_db": -0.29221094309939516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.120339140119136e-06, "output_power": 2.948753500598337e-06, "reference_active": true, "delta_db": -0.2456332925838743}], "active_band_count": 3, "active_band_mae_db": 0.1897958907994324}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010174719078202204, "log_rms_mae_db": 0.7658307579878925, "rms_envelope_nrmse": 0.05966731789991296, "envelope_correlation": 0.9972762747676943, "rms_gain_error_db": -0.017053696167984488, "reference_mean_power": 0.000309290771080255, "output_mean_power": 0.00030807864229330555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.5256223542383704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014158879062551878, "output_power": 0.00014235758514931856, "reference_active": true, "delta_db": 0.02351740073221108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.501151157810192e-05, "output_power": 1.3863765869975516e-05, "reference_active": true, "delta_db": -0.3454321011095113}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.631467090177325e-07, "output_power": 2.2313860594492515e-07, "reference_active": true, "delta_db": -0.7162322748285933}], "active_band_count": 3, "active_band_mae_db": 0.3617272588901052}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8599930119281608, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9ccf73ae3f084fe01bc8b9eaab776dcce8060fbc749747c19bf5d7139393c637", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9877381677126761, "standard_deviation": 0.0017951076678814203, "minimum": 0.9840732681057703, "maximum": 0.9901312491284908, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3427069199552837, "standard_deviation": 0.03679048929093378, "minimum": 1.2959682799938708, "maximum": 1.4022477810579612, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5030261040039641, "standard_deviation": 0.22406832822932032, "minimum": 0.21281905183664662, "maximum": 0.9013355069594206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12881903654132765, "standard_deviation": 0.009273435205446648, "minimum": 0.114906567050753, "maximum": 0.14665662048154682, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996336874072252, "standard_deviation": 0.0007124197545123345, "minimum": 0.9952556681897263, "maximum": 0.9970246287215707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7104585244678178, "standard_deviation": 0.03994803923185482, "minimum": 0.6497602630321222, "maximum": 0.7727686991685819, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.566462055633951, "standard_deviation": 0.3363646091168243, "minimum": 0.130701238380826, "maximum": 1.124711036656282, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07109005432259573, "standard_deviation": 0.006541559226833633, "minimum": 0.061342219114182986, "maximum": 0.07879137094616731, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9880323229804142, "standard_deviation": 0.0013633109766062865, "minimum": 0.98614641391388, "maximum": 0.9900119996846795, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4891075250532757, "standard_deviation": 0.025148659733694942, "minimum": 1.4510778058163643, "maximum": 1.525054284493174, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.466169533943976, "standard_deviation": 0.24863010658755816, "minimum": 0.10572449158742492, "maximum": 0.8434058854828258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1293541986450255, "standard_deviation": 0.007296979693344416, "minimum": 0.11807409297133042, "maximum": 0.1393566321685965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9965323868652589, "standard_deviation": 0.0005028036051951036, "minimum": 0.9957672573430566, "maximum": 0.9972762747676943, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7893640695959182, "standard_deviation": 0.024806556085110936, "minimum": 0.7624126632668876, "maximum": 0.8390028065264753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4023012212537036, "standard_deviation": 0.2119125103835224, "minimum": 0.08494320229216436, "maximum": 0.8117135586104679, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06864495523773026, "standard_deviation": 0.005129967596055048, "minimum": 0.05966731789991296, "maximum": 0.07679910018233516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9890552986469372, "standard_deviation": 0.0020339566033416943, "minimum": 0.9859066560115956, "maximum": 0.9910906674342823, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5000041258045662, "standard_deviation": 0.04780063382552671, "minimum": 1.4114847625179512, "maximum": 1.566741847945142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5680174178223403, "standard_deviation": 0.2066709782361169, "minimum": 0.35273094108129865, "maximum": 0.8821303332415719, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11480505637807704, "standard_deviation": 0.010395594418251575, "minimum": 0.1037442602318224, "maximum": 0.13014447881780797, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964774482824714, "standard_deviation": 0.0004780284375381057, "minimum": 0.9958458044429201, "maximum": 0.997245788069158, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8702008837792538, "standard_deviation": 0.041840253484368206, "minimum": 0.7754621488437986, "maximum": 0.9048405452922667, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4737356610749053, "standard_deviation": 0.15072652547333024, "minimum": 0.27727452702137007, "maximum": 0.7276901490244692, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06331192412535569, "standard_deviation": 0.004180840953484917, "minimum": 0.057029049555885565, "maximum": 0.0679909352231519, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.7708601932707192, "standard_deviation": 0.028379268432857218, "minimum": 0.731894870650443, "maximum": 0.8173009885129735, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.686304966483177, "standard_deviation": 0.008840798487710213, "minimum": 5.671400313874682, "maximum": 5.700487521565843, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.583419440801313, "standard_deviation": 0.023501463127129828, "minimum": 0.5510789556236113, "maximum": 0.6226966086235206, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10216139802600252, "standard_deviation": 0.06700647431042532, "minimum": -0.22033606872100314, "maximum": -0.03416984591840243, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936835080645279, "standard_deviation": 0.0009711082808985443, "minimum": 0.992362817393631, "maximum": 0.9946453006577541, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8144830378209204, "standard_deviation": 0.020674301048345615, "minimum": 0.7832344435096374, "maximum": 0.8567693159311296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3760222785757979, "standard_deviation": 0.19049930206065624, "minimum": 0.09925122544384914, "maximum": 0.6121126085800241, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08333078110186556, "standard_deviation": 0.006052159410565315, "minimum": 0.07625022054494092, "maximum": 0.09172628681048715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998162398289631, "standard_deviation": 0.0003871181925513337, "minimum": 0.9976039562585879, "maximum": 0.998786483325199, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40594982279368125, "standard_deviation": 0.015868134551046784, "minimum": 0.3930197823689001, "maximum": 0.4406583905454286, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36877299683374154, "standard_deviation": 0.1891560667294826, "minimum": 0.07864278015749934, "maximum": 0.5929941189790767, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04460183836021683, "standard_deviation": 0.005108568994714747, "minimum": 0.03520290468215658, "maximum": 0.051120130254359554, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1580-141084-0016", "language": "en", "style": "normal", "speaker": "librispeech_1580", "source_utterance_id": "1580-141084-0016", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 95360, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1580-141084-0016.wav", "source_file_sha256": "42e036b26920687901e64e1918f792405eb196a6b3db84384a9cbea61b4b5d13", "decoded_pcm_sha256": "0698365674d63379a8ffa4e1a7bae8ef5f6ed226e14107811f5311d9de011ca9", "original_source_sha256": "86f57be846b5ad8817f2752d90d5f2aba1e97ec394b2f4ccf28bc5dc39a181ef", "corpus_entry_id": "librispeech_1580-141084-0016"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.4727299213409424, "predicted_mos": 3.4727299213409424, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "7d1d126930807769a7c7787f7748c65428a500e0cf7bb3728b161e47fa96e2b6", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.5562430520003545, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1580-141084-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7d1d126930807769a7c7787f7748c65428a500e0cf7bb3728b161e47fa96e2b6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "93e1881b9bee5339d8df1b2c0d3aaebe7308511149012f5220080d4003f1ce2c"}, "rows": [{"case": "librispeech_1580-141084-0016", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.18715763092041, "predicted_mos": 2.18715763092041, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "2340c05c24961f963e4b5180a516a2a12c6b335af775fb1f1eeafcacaab8d9be", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.040852326001186, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1580-141084-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "2340c05c24961f963e4b5180a516a2a12c6b335af775fb1f1eeafcacaab8d9be", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9794216925215307, "全帯域_10ms_rms_envelope_nrmse": 0.08653119285375423, "全帯域_10ms_envelope_correlation": 0.9931173660692947, "全帯域_10ms_modulation_mae_db": 0.3181054961958859, "全帯域_40ms_log_rms_mae_db": 0.501252606239812, "全帯域_40ms_rms_envelope_nrmse": 0.04759110267702032, "全帯域_40ms_envelope_correlation": 0.9977756172039551, "全帯域_40ms_modulation_mae_db": 0.25883334271827624, "300–1000 Hz_10ms_log_rms_mae_db": 1.861482346417773, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11747576409383771, "300–1000 Hz_10ms_envelope_correlation": 0.9898956400960336, "300–1000 Hz_10ms_modulation_mae_db": 0.47154596170984325, "300–1000 Hz_40ms_log_rms_mae_db": 1.125715269185103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07465706249411112, "300–1000 Hz_40ms_envelope_correlation": 0.9957852936856041, "300–1000 Hz_40ms_modulation_mae_db": 0.5617586456551348, "1000–3000 Hz_10ms_log_rms_mae_db": 1.471455735638841, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10893753023380369, "1000–3000 Hz_10ms_envelope_correlation": 0.9915900303891632, "1000–3000 Hz_10ms_modulation_mae_db": 0.4205592006101789, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8200506815234273, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0628176482576366, "1000–3000 Hz_40ms_envelope_correlation": 0.9970988995001606, "1000–3000 Hz_40ms_modulation_mae_db": 0.38976891900369476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5651952612496556, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15031462336803694, "3000–6000 Hz_10ms_envelope_correlation": 0.9829227284477204, "3000–6000 Hz_10ms_modulation_mae_db": 1.1536702072628076, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8437096354308957, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09000529915722236, "3000–6000 Hz_40ms_envelope_correlation": 0.994106569375576, "3000–6000 Hz_40ms_modulation_mae_db": 1.3906077692521512, "mr_stft_db": 5.831450772527685, "band_mae_db": 0.4672284087604617, "rms_gain_error_db": -0.10213142329419267, "output_peak": 0.4987029364397825}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9332519568166423, "全帯域_10ms_rms_envelope_nrmse": 0.08210577156850424, "全帯域_10ms_envelope_correlation": 0.9937542763373378, "全帯域_10ms_modulation_mae_db": 0.40442137031655795, "全帯域_40ms_log_rms_mae_db": 0.4701226733135792, "全帯域_40ms_rms_envelope_nrmse": 0.04540383120817839, "全帯域_40ms_envelope_correlation": 0.9978932484478282, "全帯域_40ms_modulation_mae_db": 0.2843821150417716, "300–1000 Hz_10ms_log_rms_mae_db": 1.88931046487671, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12445048071234925, "300–1000 Hz_10ms_envelope_correlation": 0.9886753915569377, "300–1000 Hz_10ms_modulation_mae_db": 0.7502286350235449, "300–1000 Hz_40ms_log_rms_mae_db": 1.139451002190486, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08104996334620453, "300–1000 Hz_40ms_envelope_correlation": 0.9950429812605498, "300–1000 Hz_40ms_modulation_mae_db": 0.9706728670214674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4877901667958804, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10452050360372629, "1000–3000 Hz_10ms_envelope_correlation": 0.9922446325438228, "1000–3000 Hz_10ms_modulation_mae_db": 0.21597947654596292, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7868422343489196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05277044281557214, "1000–3000 Hz_40ms_envelope_correlation": 0.9979271739554795, "1000–3000 Hz_40ms_modulation_mae_db": 0.3740415141696427, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5504845928545197, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1398515580899845, "3000–6000 Hz_10ms_envelope_correlation": 0.9845851202219729, "3000–6000 Hz_10ms_modulation_mae_db": 0.26358118229727767, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8126207519098048, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08055563154953167, "3000–6000 Hz_40ms_envelope_correlation": 0.9943769309946353, "3000–6000 Hz_40ms_modulation_mae_db": 0.17208380040445034, "mr_stft_db": 5.835292122867859, "band_mae_db": 0.4212050059316393, "rms_gain_error_db": -0.04482175860979865, "output_peak": 0.5257181349798173}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9533850140178134, "全帯域_10ms_rms_envelope_nrmse": 0.0880202733077022, "全帯域_10ms_envelope_correlation": 0.9928299877650131, "全帯域_10ms_modulation_mae_db": 0.4402611709709518, "全帯域_40ms_log_rms_mae_db": 0.47774916714847504, "全帯域_40ms_rms_envelope_nrmse": 0.04178619465502359, "全帯域_40ms_envelope_correlation": 0.9982388710839516, "全帯域_40ms_modulation_mae_db": 0.3670646213234281, "300–1000 Hz_10ms_log_rms_mae_db": 1.8730170891588103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12393370255166837, "300–1000 Hz_10ms_envelope_correlation": 0.9886803287769124, "300–1000 Hz_10ms_modulation_mae_db": 0.21432827749042863, "300–1000 Hz_40ms_log_rms_mae_db": 1.0260777713176001, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06796561424975685, "300–1000 Hz_40ms_envelope_correlation": 0.9965153659831202, "300–1000 Hz_40ms_modulation_mae_db": 0.5999817738336611, "1000–3000 Hz_10ms_log_rms_mae_db": 1.466186570426464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11127996231350025, "1000–3000 Hz_10ms_envelope_correlation": 0.9911862301718671, "1000–3000 Hz_10ms_modulation_mae_db": 0.6614579076296593, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7695038422260355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05699293897251483, "1000–3000 Hz_40ms_envelope_correlation": 0.9975615787031613, "1000–3000 Hz_40ms_modulation_mae_db": 0.8089813469372983, "3000–6000 Hz_10ms_log_rms_mae_db": 1.487835445504299, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1395233684502766, "3000–6000 Hz_10ms_envelope_correlation": 0.9846796287093744, "3000–6000 Hz_10ms_modulation_mae_db": 0.8030341152027652, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7847100159586384, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07375905065405232, "3000–6000 Hz_40ms_envelope_correlation": 0.995238765530331, "3000–6000 Hz_40ms_modulation_mae_db": 0.9183662852656617, "mr_stft_db": 5.844196922058422, "band_mae_db": 0.4147698308598975, "rms_gain_error_db": -0.05800055305753605, "output_peak": 0.5078225826541514}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9556943805286437, "全帯域_10ms_rms_envelope_nrmse": 0.07926602365299269, "全帯域_10ms_envelope_correlation": 0.9943011944792877, "全帯域_10ms_modulation_mae_db": 0.5212280048565486, "全帯域_40ms_log_rms_mae_db": 0.4956028896014821, "全帯域_40ms_rms_envelope_nrmse": 0.043161391589003006, "全帯域_40ms_envelope_correlation": 0.998260569541462, "全帯域_40ms_modulation_mae_db": 0.3823729507256663, "300–1000 Hz_10ms_log_rms_mae_db": 1.9617243378056166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10541767752570275, "300–1000 Hz_10ms_envelope_correlation": 0.9918801049726376, "300–1000 Hz_10ms_modulation_mae_db": 0.468061656094853, "300–1000 Hz_40ms_log_rms_mae_db": 1.1147007218239229, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0646903557059558, "300–1000 Hz_40ms_envelope_correlation": 0.9968121644707877, "300–1000 Hz_40ms_modulation_mae_db": 0.47721041799850844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4968308815796414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10106488270464598, "1000–3000 Hz_10ms_envelope_correlation": 0.9928705720586823, "1000–3000 Hz_10ms_modulation_mae_db": 0.5631106216144824, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8051564407603533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057158233356282256, "1000–3000 Hz_40ms_envelope_correlation": 0.9977482224634047, "1000–3000 Hz_40ms_modulation_mae_db": 0.3911636730370958, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5926984129720458, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1418131596342065, "3000–6000 Hz_10ms_envelope_correlation": 0.984166626795822, "3000–6000 Hz_10ms_modulation_mae_db": 0.2279340846627199, "3000–6000 Hz_40ms_log_rms_mae_db": 0.890478830963032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07559685871520311, "3000–6000 Hz_40ms_envelope_correlation": 0.995088881377512, "3000–6000 Hz_40ms_modulation_mae_db": 0.3598759136575101, "mr_stft_db": 5.840645948582629, "band_mae_db": 0.4533014528845589, "rms_gain_error_db": -0.12368673469569073, "output_peak": 0.5039996792229088}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9824894415678986, "全帯域_10ms_rms_envelope_nrmse": 0.08355105876914512, "全帯域_10ms_envelope_correlation": 0.9935458192235851, "全帯域_10ms_modulation_mae_db": 0.24609832524867356, "全帯域_40ms_log_rms_mae_db": 0.4872257071600737, "全帯域_40ms_rms_envelope_nrmse": 0.04341256050767889, "全帯域_40ms_envelope_correlation": 0.9980969861717037, "全帯域_40ms_modulation_mae_db": 0.5551713055832684, "300–1000 Hz_10ms_log_rms_mae_db": 1.9331496946786755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12465804946013288, "300–1000 Hz_10ms_envelope_correlation": 0.9884643422906234, "300–1000 Hz_10ms_modulation_mae_db": 0.19264396520006033, "300–1000 Hz_40ms_log_rms_mae_db": 1.1018714164526926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0864549702749926, "300–1000 Hz_40ms_envelope_correlation": 0.9940049159432224, "300–1000 Hz_40ms_modulation_mae_db": 0.44559325850825576, "1000–3000 Hz_10ms_log_rms_mae_db": 1.548339522161346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11670104546249241, "1000–3000 Hz_10ms_envelope_correlation": 0.9903727499296933, "1000–3000 Hz_10ms_modulation_mae_db": 0.32879370327880947, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8544670323987885, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06915931045557658, "1000–3000 Hz_40ms_envelope_correlation": 0.9965279754018362, "1000–3000 Hz_40ms_modulation_mae_db": 0.4726025931100876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5623975285665062, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14440879099096354, "3000–6000 Hz_10ms_envelope_correlation": 0.983695954490581, "3000–6000 Hz_10ms_modulation_mae_db": 0.14331614107568041, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8256626708488449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07700428847479952, "3000–6000 Hz_40ms_envelope_correlation": 0.9948882907981907, "3000–6000 Hz_40ms_modulation_mae_db": 0.4084561964110273, "mr_stft_db": 5.834414793425974, "band_mae_db": 0.4941635713818426, "rms_gain_error_db": -0.0640025200542231, "output_peak": 0.5050242353271894}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9564743920802631, "全帯域_10ms_rms_envelope_nrmse": 0.08671505870730115, "全帯域_10ms_envelope_correlation": 0.9930301824053668, "全帯域_10ms_modulation_mae_db": 0.368294029730113, "全帯域_40ms_log_rms_mae_db": 0.4825697530019327, "全帯域_40ms_rms_envelope_nrmse": 0.04489505098958716, "全帯域_40ms_envelope_correlation": 0.9979360931813651, "全帯域_40ms_modulation_mae_db": 0.3423163811854262, "300–1000 Hz_10ms_log_rms_mae_db": 1.8621971095034504, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12112383953556007, "300–1000 Hz_10ms_envelope_correlation": 0.9892108238848462, "300–1000 Hz_10ms_modulation_mae_db": 0.3880028597080711, "300–1000 Hz_40ms_log_rms_mae_db": 1.10944930229316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0724357282513724, "300–1000 Hz_40ms_envelope_correlation": 0.9960157639681753, "300–1000 Hz_40ms_modulation_mae_db": 0.4458695505615902, "1000–3000 Hz_10ms_log_rms_mae_db": 1.500604386772965, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11311412630424043, "1000–3000 Hz_10ms_envelope_correlation": 0.9909044509699978, "1000–3000 Hz_10ms_modulation_mae_db": 0.16922204130882068, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8473408828422225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06039090316328642, "1000–3000 Hz_40ms_envelope_correlation": 0.9972770318273494, "1000–3000 Hz_40ms_modulation_mae_db": 0.3205903844781524, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5327656205461444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13603761731807357, "3000–6000 Hz_10ms_envelope_correlation": 0.9854962775867793, "3000–6000 Hz_10ms_modulation_mae_db": 0.37942867659467794, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8300616478220059, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06774615819729808, "3000–6000 Hz_40ms_envelope_correlation": 0.9959937062651354, "3000–6000 Hz_40ms_modulation_mae_db": 0.44241597105245106, "mr_stft_db": 5.8380204055270335, "band_mae_db": 0.4150408465381575, "rms_gain_error_db": -0.045937587494159805, "output_peak": 0.5199563100489094}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.946267545280829, "全帯域_10ms_rms_envelope_nrmse": 0.08429362315294002, "全帯域_10ms_envelope_correlation": 0.9934883873622041, "全帯域_10ms_modulation_mae_db": 0.4532681229129883, "全帯域_40ms_log_rms_mae_db": 0.45174882906644837, "全帯域_40ms_rms_envelope_nrmse": 0.04188915178506664, "全帯域_40ms_envelope_correlation": 0.9983201628906655, "全帯域_40ms_modulation_mae_db": 0.5635487504925439, "300–1000 Hz_10ms_log_rms_mae_db": 1.913413122277371, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12905322762927582, "300–1000 Hz_10ms_envelope_correlation": 0.9876339450234157, "300–1000 Hz_10ms_modulation_mae_db": 0.13730470414514784, "300–1000 Hz_40ms_log_rms_mae_db": 1.0737404625714897, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08272000708444252, "300–1000 Hz_40ms_envelope_correlation": 0.9945590752731855, "300–1000 Hz_40ms_modulation_mae_db": 0.29130848739308307, "1000–3000 Hz_10ms_log_rms_mae_db": 1.490354161282691, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10782650746363781, "1000–3000 Hz_10ms_envelope_correlation": 0.9917948164885073, "1000–3000 Hz_10ms_modulation_mae_db": 0.11892088807220123, "1000–3000 Hz_40ms_log_rms_mae_db": 0.840982110287144, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05582289703994403, "1000–3000 Hz_40ms_envelope_correlation": 0.9977876501669377, "1000–3000 Hz_40ms_modulation_mae_db": 0.16898992945215277, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5089002702517018, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13684191344980412, "3000–6000 Hz_10ms_envelope_correlation": 0.9856178479463911, "3000–6000 Hz_10ms_modulation_mae_db": 0.7906679525428914, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7630447406969724, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06585601763203795, "3000–6000 Hz_40ms_envelope_correlation": 0.9969427972001069, "3000–6000 Hz_40ms_modulation_mae_db": 0.8852198017844923, "mr_stft_db": 5.843414221658257, "band_mae_db": 0.5113934478765358, "rms_gain_error_db": -0.10535343814631112, "output_peak": 0.5239449333819548}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9315378982349638, "全帯域_10ms_rms_envelope_nrmse": 0.08190693871543467, "全帯域_10ms_envelope_correlation": 0.9939602323918519, "全帯域_10ms_modulation_mae_db": 0.5688491015825292, "全帯域_40ms_log_rms_mae_db": 0.43331135358895806, "全帯域_40ms_rms_envelope_nrmse": 0.043720365128835495, "全帯域_40ms_envelope_correlation": 0.99829027616193, "全帯域_40ms_modulation_mae_db": 0.7524361005893002, "300–1000 Hz_10ms_log_rms_mae_db": 1.9110989601520378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10983964413593343, "300–1000 Hz_10ms_envelope_correlation": 0.9911272025602819, "300–1000 Hz_10ms_modulation_mae_db": 0.6877728348077078, "300–1000 Hz_40ms_log_rms_mae_db": 1.0990435348950527, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06291293861365434, "300–1000 Hz_40ms_envelope_correlation": 0.9969835031855256, "300–1000 Hz_40ms_modulation_mae_db": 1.0381466318539931, "1000–3000 Hz_10ms_log_rms_mae_db": 1.477647926029139, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11170760297291162, "1000–3000 Hz_10ms_envelope_correlation": 0.9914205672898633, "1000–3000 Hz_10ms_modulation_mae_db": 0.43124940973608955, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8233233927787883, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06888881002574974, "1000–3000 Hz_40ms_envelope_correlation": 0.9968563394736494, "1000–3000 Hz_40ms_modulation_mae_db": 0.47012102026657815, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5679461184590622, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1409051741622351, "3000–6000 Hz_10ms_envelope_correlation": 0.9843541925708188, "3000–6000 Hz_10ms_modulation_mae_db": 0.06114090127215327, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8811283614661855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06991458606388452, "3000–6000 Hz_40ms_envelope_correlation": 0.9957881124409936, "3000–6000 Hz_40ms_modulation_mae_db": 0.20376068480619827, "mr_stft_db": 5.830795040961346, "band_mae_db": 0.550175795504924, "rms_gain_error_db": -0.14584007741256477, "output_peak": 0.5052701482599592}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 95360, "output_samples": 95360, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.888344773782408, "spectral_convergence": 0.40593823062353457, "spectral_convergence_status": "ok", "amplitude_floor": 1.2569032294341943e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.730300741782244, "spectral_convergence": 0.4068116334870667, "spectral_convergence_status": "ok", "amplitude_floor": 9.073914887165124e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.875706802018404, "spectral_convergence": 0.42607985861819425, "spectral_convergence_status": "ok", "amplitude_floor": 7.969434026985316e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.036863249031963e-06, "output_power": 8.018313467559049e-06, "delta_db": 2.0192287062255603, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00014485054714414723, "output_power": 0.00015139145969919648, "delta_db": 0.19181236414216807, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004849180298355377, "output_power": 0.00046238128451067857, "delta_db": -0.20668084877770634, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006903353830986323, "output_power": 0.0006783844998764496, "delta_db": -0.07584217347249528, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00029452981369653373, "output_power": 0.0002783324776214251, "delta_db": -0.24565377122703158, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004100383435072048, "output_power": 0.00040402590477228435, "delta_db": -0.06415258871780874, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4672284087604617, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10213142329419267, "pitch": null, "output_rms_dbfs": -27.027793690515768, "output_peak": 0.4987029364397825, "reference_rms_dbfs": -26.925662267221572, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.831450772527685}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 95360, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.0001898595271537964, "log_rms_mae_db": 0.9794216925215307, "rms_envelope_nrmse": 0.08653119285375423, "envelope_correlation": 0.9931173660692947, "rms_gain_error_db": -0.1021303729927691, "reference_mean_power": 0.002031411160407025, "output_mean_power": 0.0019841970371620734}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 9.927152563814217e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006381525961065371, "output_power": 0.0006097762866749452, "reference_active": true, "delta_db": -0.19754009087457702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010674365906282643, "output_power": 9.254880746184463e-05, "reference_active": true, "delta_db": -0.6197125886971874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4567838102832594e-05, "output_power": 3.567620084635659e-05, "reference_active": true, "delta_db": 0.1370638090158933}], "active_band_count": 3, "active_band_mae_db": 0.3181054961958859}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00017019756464998765, "log_rms_mae_db": 0.501252606239812, "rms_envelope_nrmse": 0.04759110267702032, "envelope_correlation": 0.9977756172039551, "rms_gain_error_db": -0.10212990207228742, "reference_mean_power": 0.0020416854120747243, "output_mean_power": 0.0019942327105850026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 8.390142156345837e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005509511464645055, "output_power": 0.0005317465471616696, "reference_active": true, "delta_db": -0.15408412577657404}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5897699076585876e-05, "output_power": 4.125726548447723e-05, "reference_active": true, "delta_db": -0.4629047450005791}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848237969563024e-06, "output_power": 1.917383579440178e-06, "reference_active": true, "delta_db": 0.15951115737767563}], "active_band_count": 3, "active_band_mae_db": 0.25883334271827624}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 9.246932648014587e-05, "log_rms_mae_db": 1.861482346417773, "rms_envelope_nrmse": 0.11747576409383771, "envelope_correlation": 0.9898956400960336, "rms_gain_error_db": -0.15865916142448194, "reference_mean_power": 0.0005236182881812637, "output_mean_power": 0.0005048343416894553}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.7855833380403224e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028934531457073826, "output_power": 0.0002702938597856939, "reference_active": true, "delta_db": -0.29580274109552984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6197786234688146e-05, "output_power": 3.0030351423407605e-05, "reference_active": true, "delta_db": -0.8112159652829125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0243815325354375e-05, "output_power": 9.543327384120434e-06, "reference_active": true, "delta_db": -0.3076191787510873}], "active_band_count": 3, "active_band_mae_db": 0.47154596170984325}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 8.334521268786246e-05, "log_rms_mae_db": 1.125715269185103, "rms_envelope_nrmse": 0.07465706249411112, "envelope_correlation": 0.9957852936856041, "rms_gain_error_db": -0.15866059687624545, "reference_mean_power": 0.000526268698480611, "output_mean_power": 0.0005073895051654178}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.428386331422997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002786455084699618, "output_power": 0.0002597051698752411, "reference_active": true, "delta_db": -0.30571451899506563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6309295594140896e-05, "output_power": 1.3424314162083511e-05, "reference_active": true, "delta_db": -0.8454309690377942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.788822138965803e-07, "output_power": 6.887454483112602e-07, "reference_active": true, "delta_db": -0.5341304489325441}], "active_band_count": 3, "active_band_mae_db": 0.5617586456551348}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00017019739791537767, "log_rms_mae_db": 1.471455735638841, "rms_envelope_nrmse": 0.10893753023380369, "envelope_correlation": 0.9915900303891632, "rms_gain_error_db": -0.11229129693907625, "reference_mean_power": 0.0008446325483556021, "output_mean_power": 0.0008230736218329218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 4.004634586327471e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028725854625925005, "output_power": 0.0002687056800446792, "reference_active": true, "delta_db": -0.28996111413659315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3436307686794414e-05, "output_power": 3.9478176668303864e-05, "reference_active": true, "delta_db": -0.41495814631718675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1340506745101254e-05, "output_power": 1.2891643742112509e-05, "reference_active": true, "delta_db": 0.5567583413767568}], "active_band_count": 3, "active_band_mae_db": 0.4205592006101789}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00015333717053704127, "log_rms_mae_db": 0.8200506815234273, "rms_envelope_nrmse": 0.0628176482576366, "envelope_correlation": 0.9970988995001606, "rms_gain_error_db": -0.11229167369888661, "reference_mean_power": 0.0008489024890862695, "output_mean_power": 0.000827234502176115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.579273675799131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002706567111490358, "output_power": 0.00025623954599368, "reference_active": true, "delta_db": -0.23772644067042184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.939557003916555e-05, "output_power": 1.793211850102751e-05, "reference_active": true, "delta_db": -0.34070947957232983}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.660068008029229e-07, "output_power": 8.776466694951757e-07, "reference_active": true, "delta_db": 0.5908708367683326}], "active_band_count": 3, "active_band_mae_db": 0.38976891900369476}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00010107776753147987, "log_rms_mae_db": 1.5651952612496556, "rms_envelope_nrmse": 0.15031462336803694, "envelope_correlation": 0.9829227284477204, "rms_gain_error_db": -0.33645469332199285, "reference_mean_power": 0.00032790731816785513, "output_mean_power": 0.00030346292526686085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.172510420573334e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020494142716961824, "output_power": 0.0001864476230431553, "reference_active": true, "delta_db": -0.410729051566727}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6580284356723e-05, "output_power": 2.5058969336847517e-05, "reference_active": true, "delta_db": -1.642838724344524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.565019545039446e-06, "output_power": 6.917379096900744e-06, "reference_active": true, "delta_db": -1.4074428458771717}], "active_band_count": 3, "active_band_mae_db": 1.1536702072628076}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 7.865643818167715e-05, "log_rms_mae_db": 0.8437096354308957, "rms_envelope_nrmse": 0.09000529915722236, "envelope_correlation": 0.994106569375576, "rms_gain_error_db": -0.3364502219044597, "reference_mean_power": 0.0003295653902210026, "output_mean_power": 0.0003049977076188042}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 2.853826866715288e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001952923309195638, "output_power": 0.0001740907915754317, "reference_active": true, "delta_db": -0.49909389000981297}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5170390000310775e-05, "output_power": 1.03975284847909e-05, "reference_active": true, "delta_db": -1.6406662695048326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.363045655471935e-07, "output_power": 4.611595200876204e-07, "reference_active": true, "delta_db": -2.032063148241808}], "active_band_count": 3, "active_band_mae_db": 1.3906077692521512}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8232507735195648, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "46defb958dd4dc7138dd468e1b0f8d7d5ab8e97baaa9403b7028730807632f7e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9915480062301996, "standard_deviation": 0.0007782455791031202, "minimum": 0.9903727499296933, "maximum": 0.9928705720586823, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.492401168835871, "standard_deviation": 0.025593081115548075, "minimum": 1.466186570426464, "maximum": 1.548339522161346, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.36366165609952555, "standard_deviation": 0.19160917809419273, "minimum": 0.11892088807220123, "maximum": 0.6614579076296593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1093940201323698, "standard_deviation": 0.004955983760129329, "minimum": 0.10106488270464598, "maximum": 0.11670104546249241, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973481089364973, "standard_deviation": 0.0004950882292263214, "minimum": 0.9965279754018362, "maximum": 0.9979271739554795, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8184583271457099, "standard_deviation": 0.029850611957213124, "minimum": 0.7695038422260355, "maximum": 0.8544670323987885, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4245324225568378, "standard_deviation": 0.18250919996816042, "minimum": 0.16898992945215277, "maximum": 0.8089813469372983, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06050014801082033, "standard_deviation": 0.006045392747760197, "minimum": 0.05277044281557214, "maximum": 0.06915931045557658, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9844397970961825, "standard_deviation": 0.0008879625233469477, "minimum": 0.9829227284477204, "maximum": 0.9856178479463911, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5460279063004918, "standard_deviation": 0.034339628038266644, "minimum": 1.487835445504299, "maximum": 1.5926984129720458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4778466576138717, "standard_deviation": 0.38986223872350506, "minimum": 0.06114090127215327, "maximum": 1.1536702072628076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1412120256829476, "standard_deviation": 0.004537257370324553, "minimum": 0.13603761731807357, "maximum": 0.15031462336803694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9953030067478101, "standard_deviation": 0.0009190329309839223, "minimum": 0.994106569375576, "maximum": 0.9969427972001069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8289270818870474, "standard_deviation": 0.04359814398065523, "minimum": 0.7630447406969724, "maximum": 0.890478830963032, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5975983028292429, "standard_deviation": 0.4254444947895829, "minimum": 0.17208380040445034, "maximum": 1.3906077692521512, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07505473630550369, "standard_deviation": 0.007782815442168253, "minimum": 0.06585601763203795, "maximum": 0.09000529915722236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.989445972395211, "standard_deviation": 0.001436069933257662, "minimum": 0.9876339450234157, "maximum": 0.9918801049726376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9006741406088055, "standard_deviation": 0.035783617524588356, "minimum": 1.861482346417773, "maximum": 1.9617243378056166, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.41373611177245706, "standard_deviation": 0.22688522344542222, "minimum": 0.13730470414514784, "maximum": 0.7502286350235449, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11949404820555753, "standard_deviation": 0.008109984975418877, "minimum": 0.10541767752570275, "maximum": 0.12905322762927582, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9957148829712713, "standard_deviation": 0.0010863447633025648, "minimum": 0.9940049159432224, "maximum": 0.9969835031855256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0987561850911884, "standard_deviation": 0.035191632120673236, "minimum": 1.0260777713176001, "maximum": 1.139451002190486, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6038177041032118, "standard_deviation": 0.2642190694464701, "minimum": 0.29130848739308307, "maximum": 1.0381466318539931, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07411083000256127, "standard_deviation": 0.008698550130784924, "minimum": 0.06291293861365434, "maximum": 0.0864549702749926, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.46590979496725216, "standard_deviation": 0.049756507756106925, "minimum": 0.4147698308598975, "maximum": 0.550175795504924, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.83727877845115, "standard_deviation": 0.0051484524060866664, "minimum": 5.830795040961346, "maximum": 5.844196922058422, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5113048700393341, "standard_deviation": 0.010299278267890299, "minimum": 0.4987029364397825, "maximum": 0.5257181349798173, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08622176159555961, "standard_deviation": 0.038180047911234284, "minimum": -0.14584007741256477, "maximum": -0.04482175860979865, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9935034307542427, "standard_deviation": 0.0004980786928755009, "minimum": 0.9928299877650131, "maximum": 0.9943011944792877, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9548152901310731, "standard_deviation": 0.018734166893877944, "minimum": 0.9315378982349638, "maximum": 0.9824894415678986, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.415065702726781, "standard_deviation": 0.1050619899970315, "minimum": 0.24609832524867356, "maximum": 0.5688491015825292, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08404874259097178, "standard_deviation": 0.002942044321424319, "minimum": 0.07926602365299269, "maximum": 0.0880202733077022, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981014780853577, "standard_deviation": 0.00020859116711505676, "minimum": 0.9977756172039551, "maximum": 0.9983201628906655, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47494787239009517, "standard_deviation": 0.022773205298797818, "minimum": 0.43331135358895806, "maximum": 0.501252606239812, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4382656959574601, "standard_deviation": 0.16959691596351686, "minimum": 0.25883334271827624, "maximum": 0.7524361005893002, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.043982456067549185, "standard_deviation": 0.0019327468843454894, "minimum": 0.04178619465502359, "maximum": 0.04759110267702032, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0016", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.332294225692749, "predicted_mos": 2.332294225692749, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "87505e0764f10588b930c818a4cf33759f4b2b68eac9c098017da626a8cde449", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.585216599982232, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9958800149789752, "全帯域_10ms_rms_envelope_nrmse": 0.08705559122528014, "全帯域_10ms_envelope_correlation": 0.9930541400130056, "全帯域_10ms_modulation_mae_db": 0.29782202138674313, "全帯域_40ms_log_rms_mae_db": 0.5128584672497246, "全帯域_40ms_rms_envelope_nrmse": 0.048474469249048135, "全帯域_40ms_envelope_correlation": 0.9977162649950632, "全帯域_40ms_modulation_mae_db": 0.24876944805270232, "300–1000 Hz_10ms_log_rms_mae_db": 1.7468276816014796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1185982428912726, "300–1000 Hz_10ms_envelope_correlation": 0.9896535414975415, "300–1000 Hz_10ms_modulation_mae_db": 0.4567238513105953, "300–1000 Hz_40ms_log_rms_mae_db": 1.0217527021349286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07430437653908345, "300–1000 Hz_40ms_envelope_correlation": 0.9957700535049764, "300–1000 Hz_40ms_modulation_mae_db": 0.525285667453809, "1000–3000 Hz_10ms_log_rms_mae_db": 1.500526935365583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11127052992586889, "1000–3000 Hz_10ms_envelope_correlation": 0.9912548841029725, "1000–3000 Hz_10ms_modulation_mae_db": 0.38602051273638, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8156111292337657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06393546970650536, "1000–3000 Hz_40ms_envelope_correlation": 0.9970434644430589, "1000–3000 Hz_40ms_modulation_mae_db": 0.3388964485333441, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6340078188447384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15046272355402068, "3000–6000 Hz_10ms_envelope_correlation": 0.9830549005564585, "3000–6000 Hz_10ms_modulation_mae_db": 1.3379834608573455, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8446349165737279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09252675604730719, "3000–6000 Hz_40ms_envelope_correlation": 0.9939078887665339, "3000–6000 Hz_40ms_modulation_mae_db": 1.5986396486179217, "mr_stft_db": 5.821320884514473, "band_mae_db": 0.4292452686222032, "rms_gain_error_db": -0.11258365978652621, "output_peak": 0.4981321278417582}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9310543797604347, "全帯域_10ms_rms_envelope_nrmse": 0.08209488101734844, "全帯域_10ms_envelope_correlation": 0.9937571712798284, "全帯域_10ms_modulation_mae_db": 0.4320719747259509, "全帯域_40ms_log_rms_mae_db": 0.48055473324825565, "全帯域_40ms_rms_envelope_nrmse": 0.04534881474212995, "全帯域_40ms_envelope_correlation": 0.9979024639608236, "全帯域_40ms_modulation_mae_db": 0.30953637174691934, "300–1000 Hz_10ms_log_rms_mae_db": 1.7683866885639938, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11912812655249323, "300–1000 Hz_10ms_envelope_correlation": 0.9895928039696666, "300–1000 Hz_10ms_modulation_mae_db": 0.8810681387285381, "300–1000 Hz_40ms_log_rms_mae_db": 1.0345119780922214, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07653879338113509, "300–1000 Hz_40ms_envelope_correlation": 0.9955293882684171, "300–1000 Hz_40ms_modulation_mae_db": 1.0383034011871641, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5294629988443436, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10463725921667402, "1000–3000 Hz_10ms_envelope_correlation": 0.9922393100194552, "1000–3000 Hz_10ms_modulation_mae_db": 0.11666500805447155, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7845889051596651, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0525633038809765, "1000–3000 Hz_40ms_envelope_correlation": 0.9979533351829073, "1000–3000 Hz_40ms_modulation_mae_db": 0.2816548934016569, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6040133338172713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13927017057707944, "3000–6000 Hz_10ms_envelope_correlation": 0.9847793190532189, "3000–6000 Hz_10ms_modulation_mae_db": 0.41248674857849393, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8309947708020687, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08042869644371066, "3000–6000 Hz_40ms_envelope_correlation": 0.9945715164074549, "3000–6000 Hz_40ms_modulation_mae_db": 0.30743018150802554, "mr_stft_db": 5.825640796984186, "band_mae_db": 0.3792131624825567, "rms_gain_error_db": -0.04751887930005006, "output_peak": 0.5197704029806934}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9765553344732676, "全帯域_10ms_rms_envelope_nrmse": 0.08781005363107267, "全帯域_10ms_envelope_correlation": 0.9928582133490508, "全帯域_10ms_modulation_mae_db": 0.44607531509918524, "全帯域_40ms_log_rms_mae_db": 0.4886682569665662, "全帯域_40ms_rms_envelope_nrmse": 0.04080517355947527, "全帯域_40ms_envelope_correlation": 0.9983008784375761, "全帯域_40ms_modulation_mae_db": 0.37849716276545026, "300–1000 Hz_10ms_log_rms_mae_db": 1.830571643973366, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12172420137102047, "300–1000 Hz_10ms_envelope_correlation": 0.9890594366829248, "300–1000 Hz_10ms_modulation_mae_db": 0.2865608418026069, "300–1000 Hz_40ms_log_rms_mae_db": 0.9826425781372196, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06675021123357047, "300–1000 Hz_40ms_envelope_correlation": 0.996594240079095, "300–1000 Hz_40ms_modulation_mae_db": 0.7028911429532488, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5111176343748987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11152968792289972, "1000–3000 Hz_10ms_envelope_correlation": 0.9911515143192821, "1000–3000 Hz_10ms_modulation_mae_db": 0.5635051675063433, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7939919769743796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05683356413112092, "1000–3000 Hz_40ms_envelope_correlation": 0.9975539668774479, "1000–3000 Hz_40ms_modulation_mae_db": 0.6815147125131663, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5602060988431625, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1369096837255326, "3000–6000 Hz_10ms_envelope_correlation": 0.9852407315012772, "3000–6000 Hz_10ms_modulation_mae_db": 0.8678143869414497, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8043565943437099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06831570277642492, "3000–6000 Hz_40ms_envelope_correlation": 0.9959265092225209, "3000–6000 Hz_40ms_modulation_mae_db": 1.0151662515643822, "mr_stft_db": 5.833851852229421, "band_mae_db": 0.32225742111030886, "rms_gain_error_db": -0.038959105674021076, "output_peak": 0.5088587234329519}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9713916955830747, "全帯域_10ms_rms_envelope_nrmse": 0.0791342262468137, "全帯域_10ms_envelope_correlation": 0.9943189846391576, "全帯域_10ms_modulation_mae_db": 0.49112475478646506, "全帯域_40ms_log_rms_mae_db": 0.5010514561637693, "全帯域_40ms_rms_envelope_nrmse": 0.04211590279506978, "全帯域_40ms_envelope_correlation": 0.9983524438055656, "全帯域_40ms_modulation_mae_db": 0.38502779664205616, "300–1000 Hz_10ms_log_rms_mae_db": 1.8751821973296876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10045485394637595, "300–1000 Hz_10ms_envelope_correlation": 0.9925791298520352, "300–1000 Hz_10ms_modulation_mae_db": 0.4800097647736203, "300–1000 Hz_40ms_log_rms_mae_db": 1.031090797181589, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060314438055315424, "300–1000 Hz_40ms_envelope_correlation": 0.9971672815222479, "300–1000 Hz_40ms_modulation_mae_db": 0.5369769699635674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5132037157781377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09873731182368033, "1000–3000 Hz_10ms_envelope_correlation": 0.9932634264994654, "1000–3000 Hz_10ms_modulation_mae_db": 0.5303213612619996, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8031681625931469, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05626822751473362, "1000–3000 Hz_40ms_envelope_correlation": 0.9979038480338343, "1000–3000 Hz_40ms_modulation_mae_db": 0.3650405689829966, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6264449822067193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1420421365330743, "3000–6000 Hz_10ms_envelope_correlation": 0.9841289999667214, "3000–6000 Hz_10ms_modulation_mae_db": 0.2631098876267903, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8694320284265279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07324017547639715, "3000–6000 Hz_40ms_envelope_correlation": 0.995472925871996, "3000–6000 Hz_40ms_modulation_mae_db": 0.4197348744023995, "mr_stft_db": 5.829379344090801, "band_mae_db": 0.40229108497192184, "rms_gain_error_db": -0.12316080692692459, "output_peak": 0.5035268632562955}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9977489439627336, "全帯域_10ms_rms_envelope_nrmse": 0.08478589223687166, "全帯域_10ms_envelope_correlation": 0.9933424092821618, "全帯域_10ms_modulation_mae_db": 0.24170612079399842, "全帯域_40ms_log_rms_mae_db": 0.49789511758753724, "全帯域_40ms_rms_envelope_nrmse": 0.04445070708172245, "全帯域_40ms_envelope_correlation": 0.9979821337192691, "全帯域_40ms_modulation_mae_db": 0.6134345767080339, "300–1000 Hz_10ms_log_rms_mae_db": 1.8616483764153844, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12462298915761766, "300–1000 Hz_10ms_envelope_correlation": 0.9884857773526649, "300–1000 Hz_10ms_modulation_mae_db": 0.22534405022173365, "300–1000 Hz_40ms_log_rms_mae_db": 1.002530585305583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08476704663497775, "300–1000 Hz_40ms_envelope_correlation": 0.9942312389874838, "300–1000 Hz_40ms_modulation_mae_db": 0.5449435317550981, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5864037729059681, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1164781109335141, "1000–3000 Hz_10ms_envelope_correlation": 0.9903938365591392, "1000–3000 Hz_10ms_modulation_mae_db": 0.40260258331337456, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8465190968057262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06912995631845102, "1000–3000 Hz_40ms_envelope_correlation": 0.99650527861249, "1000–3000 Hz_40ms_modulation_mae_db": 0.5761852900207561, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6417964813594947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14818956513790837, "3000–6000 Hz_10ms_envelope_correlation": 0.9828722222450508, "3000–6000 Hz_10ms_modulation_mae_db": 0.16552412958261023, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8313383098713457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0801817402725651, "3000–6000 Hz_40ms_envelope_correlation": 0.994504444465483, "3000–6000 Hz_40ms_modulation_mae_db": 0.5091251315216861, "mr_stft_db": 5.82954711179211, "band_mae_db": 0.45051065380290717, "rms_gain_error_db": -0.049281243472264785, "output_peak": 0.5039814738830322}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9713124809475095, "全帯域_10ms_rms_envelope_nrmse": 0.08862373315291369, "全帯域_10ms_envelope_correlation": 0.9927199965154176, "全帯域_10ms_modulation_mae_db": 0.39731029062105816, "全帯域_40ms_log_rms_mae_db": 0.4942516378778943, "全帯域_40ms_rms_envelope_nrmse": 0.04672936572131423, "全帯域_40ms_envelope_correlation": 0.9977696610689798, "全帯域_40ms_modulation_mae_db": 0.372575256054034, "300–1000 Hz_10ms_log_rms_mae_db": 1.8345800095219436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12573647876705074, "300–1000 Hz_10ms_envelope_correlation": 0.9883504280082009, "300–1000 Hz_10ms_modulation_mae_db": 0.4525500043654403, "300–1000 Hz_40ms_log_rms_mae_db": 1.0325841018730886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07714741095864218, "300–1000 Hz_40ms_envelope_correlation": 0.9954405057969615, "300–1000 Hz_40ms_modulation_mae_db": 0.5036748814335117, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5463324941549168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1144409039108693, "1000–3000 Hz_10ms_envelope_correlation": 0.9906967615698028, "1000–3000 Hz_10ms_modulation_mae_db": 0.23820397940661167, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8675666073841065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062185288270082746, "1000–3000 Hz_40ms_envelope_correlation": 0.997112928098224, "1000–3000 Hz_40ms_modulation_mae_db": 0.3977237252537343, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5948955696780938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13606874448210293, "3000–6000 Hz_10ms_envelope_correlation": 0.9854224758219108, "3000–6000 Hz_10ms_modulation_mae_db": 0.45458180136229304, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8375305237038835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06673435284922928, "3000–6000 Hz_40ms_envelope_correlation": 0.9961110456268046, "3000–6000 Hz_40ms_modulation_mae_db": 0.6002346682329681, "mr_stft_db": 5.824026026047268, "band_mae_db": 0.3640490558526986, "rms_gain_error_db": -0.053499852682929415, "output_peak": 0.5214086988844837}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9602881356437899, "全帯域_10ms_rms_envelope_nrmse": 0.08656733157582265, "全帯域_10ms_envelope_correlation": 0.99312937625956, "全帯域_10ms_modulation_mae_db": 0.4153177878616689, "全帯域_40ms_log_rms_mae_db": 0.46462299810514646, "全帯域_40ms_rms_envelope_nrmse": 0.043218873834015084, "全帯域_40ms_envelope_correlation": 0.9982131793616953, "全帯域_40ms_modulation_mae_db": 0.5206257644405874, "300–1000 Hz_10ms_log_rms_mae_db": 1.887863238551953, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1266838221692289, "300–1000 Hz_10ms_envelope_correlation": 0.988079587268961, "300–1000 Hz_10ms_modulation_mae_db": 0.2081222198721118, "300–1000 Hz_40ms_log_rms_mae_db": 1.033657099015391, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08009564474520467, "300–1000 Hz_40ms_envelope_correlation": 0.9948792075636014, "300–1000 Hz_40ms_modulation_mae_db": 0.27193280719369617, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5201782906526526, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10830841636949479, "1000–3000 Hz_10ms_envelope_correlation": 0.9917614809666295, "1000–3000 Hz_10ms_modulation_mae_db": 0.11132590133686758, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8268248483030727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05695396715486525, "1000–3000 Hz_40ms_envelope_correlation": 0.9977497018755256, "1000–3000 Hz_40ms_modulation_mae_db": 0.15120151667815068, "3000–6000 Hz_10ms_log_rms_mae_db": 1.568679827327268, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.138985691494983, "3000–6000 Hz_10ms_envelope_correlation": 0.9850947363585802, "3000–6000 Hz_10ms_modulation_mae_db": 0.706646543676856, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7820604821103279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06637035552931368, "3000–6000 Hz_40ms_envelope_correlation": 0.996826335821908, "3000–6000 Hz_40ms_modulation_mae_db": 0.8552789924383796, "mr_stft_db": 5.832665896922758, "band_mae_db": 0.44719760600390984, "rms_gain_error_db": -0.10999025671570357, "output_peak": 0.5238933856690584}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9464535468682713, "全帯域_10ms_rms_envelope_nrmse": 0.08342868699387751, "全帯域_10ms_envelope_correlation": 0.993705770749475, "全帯域_10ms_modulation_mae_db": 0.5231402395941185, "全帯域_40ms_log_rms_mae_db": 0.4553426947567303, "全帯域_40ms_rms_envelope_nrmse": 0.044682547628295766, "全帯域_40ms_envelope_correlation": 0.9981802733739662, "全帯域_40ms_modulation_mae_db": 0.6652642862749684, "300–1000 Hz_10ms_log_rms_mae_db": 1.8864567817973028, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10814122545913582, "300–1000 Hz_10ms_envelope_correlation": 0.9913915915704724, "300–1000 Hz_10ms_modulation_mae_db": 0.7022380610111648, "300–1000 Hz_40ms_log_rms_mae_db": 1.045139106619611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06184603065242506, "300–1000 Hz_40ms_envelope_correlation": 0.9970729085267745, "300–1000 Hz_40ms_modulation_mae_db": 1.139949110255296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5096957006052072, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1129966372813369, "1000–3000 Hz_10ms_envelope_correlation": 0.9911927662794469, "1000–3000 Hz_10ms_modulation_mae_db": 0.3905798200062922, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8037439145423085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06891611383507375, "1000–3000 Hz_40ms_envelope_correlation": 0.9968336663458036, "1000–3000 Hz_40ms_modulation_mae_db": 0.4334197043044587, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6202953012377106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14423967207088115, "3000–6000 Hz_10ms_envelope_correlation": 0.98362480012671, "3000–6000 Hz_10ms_modulation_mae_db": 0.09668721792114156, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8884002141359949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07245777870504254, "3000–6000 Hz_40ms_envelope_correlation": 0.9954975551781741, "3000–6000 Hz_40ms_modulation_mae_db": 0.20847682104592155, "mr_stft_db": 5.818812582526822, "band_mae_db": 0.4881144306127682, "rms_gain_error_db": -0.13976155372507, "output_peak": 0.505579828828907}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 95360, "output_samples": 95360, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.888704856669968, "spectral_convergence": 0.40618059666466605, "spectral_convergence_status": "ok", "amplitude_floor": 1.2569032294341943e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.737185933993003, "spectral_convergence": 0.4066070137200333, "spectral_convergence_status": "ok", "amplitude_floor": 9.073914887165124e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.838071862880449, "spectral_convergence": 0.42484297559046774, "spectral_convergence_status": "ok", "amplitude_floor": 7.969434026985316e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.036863249031963e-06, "output_power": 7.582880838126926e-06, "delta_db": 1.7767407136860034, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00014485054714414723, "output_power": 0.000150365424138846, "delta_db": 0.16227843645173773, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004849180298355377, "output_power": 0.0004639780520536665, "delta_db": -0.19170894765404, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006903353830986323, "output_power": 0.0006762504764099792, "delta_db": -0.08952549626643905, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00029452981369653373, "output_power": 0.0002751633967527336, "delta_db": -0.2953860072751962, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004100383435072048, "output_power": 0.00040442804991632744, "delta_db": -0.05983201039980257, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4292452686222032, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11258365978652621, "pitch": null, "output_rms_dbfs": -27.038245927008102, "output_peak": 0.4981321278417582, "reference_rms_dbfs": -26.925662267221572, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.821320884514473}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 95360, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.0001898595271537964, "log_rms_mae_db": 0.9958800149789752, "rms_envelope_nrmse": 0.08705559122528014, "envelope_correlation": 0.9930541400130056, "rms_gain_error_db": -0.11258250897356539, "reference_mean_power": 0.002031411160407025, "output_mean_power": 0.0019794274253446824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 9.927152563814217e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006381525961065371, "output_power": 0.0006112076690383296, "reference_active": true, "delta_db": -0.18735745507853163}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010674365906282643, "output_power": 9.24732095250805e-05, "reference_active": true, "delta_db": -0.6232615467230083}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4567838102832594e-05, "output_power": 3.523359237146946e-05, "reference_active": true, "delta_db": 0.08284706235868954}], "active_band_count": 3, "active_band_mae_db": 0.29782202138674313}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00017019756464998765, "log_rms_mae_db": 0.5128584672497246, "rms_envelope_nrmse": 0.048474469249048135, "envelope_correlation": 0.9977162649950632, "rms_gain_error_db": -0.11258209482586112, "reference_mean_power": 0.0020416854120747243, "output_mean_power": 0.0019894389490142275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 8.390142156345837e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005509511464645055, "output_power": 0.0005330079070975134, "reference_active": true, "delta_db": -0.14379439345566317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5897699076585876e-05, "output_power": 4.123935693786092e-05, "reference_active": true, "delta_db": -0.4647902968857771}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848237969563024e-06, "output_power": 1.90778861777836e-06, "reference_active": true, "delta_db": 0.13772365381666674}], "active_band_count": 3, "active_band_mae_db": 0.24876944805270232}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 9.246932648014587e-05, "log_rms_mae_db": 1.7468276816014796, "rms_envelope_nrmse": 0.1185982428912726, "envelope_correlation": 0.9896535414975415, "rms_gain_error_db": -0.14530104872617935, "reference_mean_power": 0.0005236182881812637, "output_mean_power": 0.000506389511292644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.7855833380403224e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028934531457073826, "output_power": 0.000272449392838719, "reference_active": true, "delta_db": -0.2613060973152262}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6197786234688146e-05, "output_power": 2.9797395887680002e-05, "reference_active": true, "delta_db": -0.8450369996852796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0243815325354375e-05, "output_power": 9.640041305148217e-06, "reference_active": true, "delta_db": -0.2638284569312802}], "active_band_count": 3, "active_band_mae_db": 0.4567238513105953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 8.334521268786246e-05, "log_rms_mae_db": 1.0217527021349286, "rms_envelope_nrmse": 0.07430437653908345, "envelope_correlation": 0.9957700535049764, "rms_gain_error_db": -0.14530252102859426, "reference_mean_power": 0.000526268698480611, "output_mean_power": 0.0005089525417698932}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.428386331422997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002786455084699618, "output_power": 0.00026167399704544965, "reference_active": true, "delta_db": -0.27291478726970453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6309295594140896e-05, "output_power": 1.3342949083306517e-05, "reference_active": true, "delta_db": -0.871833753098531}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.788822138965803e-07, "output_power": 7.052789800784558e-07, "reference_active": true, "delta_db": -0.43110846199319175}], "active_band_count": 3, "active_band_mae_db": 0.525285667453809}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00017019739791537767, "log_rms_mae_db": 1.500526935365583, "rms_envelope_nrmse": 0.11127052992586889, "envelope_correlation": 0.9912548841029725, "rms_gain_error_db": -0.13643307321288012, "reference_mean_power": 0.0008446325483556021, "output_mean_power": 0.0008185109727989218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 4.004634586327471e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028725854625925005, "output_power": 0.00027119930826403357, "reference_active": true, "delta_db": -0.2498438083248502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3436307686794414e-05, "output_power": 4.0475559368318776e-05, "reference_active": true, "delta_db": -0.30660041474005256}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1340506745101254e-05, "output_power": 1.3025493687236057e-05, "reference_active": true, "delta_db": 0.6016173151442373}], "active_band_count": 3, "active_band_mae_db": 0.38602051273638}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00015333717053704127, "log_rms_mae_db": 0.8156111292337657, "rms_envelope_nrmse": 0.06393546970650536, "envelope_correlation": 0.9970434644430589, "rms_gain_error_db": -0.13643372378393356, "reference_mean_power": 0.0008489024890862695, "output_mean_power": 0.0008226487357373309}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.579273675799131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002706567111490358, "output_power": 0.0002591666446454453, "reference_active": true, "delta_db": -0.18839694161180387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.939557003916555e-05, "output_power": 1.8280255550127376e-05, "reference_active": true, "delta_db": -0.257202854230287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.660068008029229e-07, "output_power": 8.736582461178133e-07, "reference_active": true, "delta_db": 0.5710895497579415}], "active_band_count": 3, "active_band_mae_db": 0.3388964485333441}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00010107776753147987, "log_rms_mae_db": 1.6340078188447384, "rms_envelope_nrmse": 0.15046272355402068, "envelope_correlation": 0.9830549005564585, "rms_gain_error_db": -0.3669844728493956, "reference_mean_power": 0.00032790731816785513, "output_mean_power": 0.0003013371399551932}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.172510420573334e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020494142716961824, "output_power": 0.00018560463764116897, "reference_active": true, "delta_db": -0.4304093262801722}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6580284356723e-05, "output_power": 2.4343396388385278e-05, "reference_active": true, "delta_db": -1.7686590629343535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.565019545039446e-06, "output_power": 6.297927581338849e-06, "reference_active": true, "delta_db": -1.8148819933575113}], "active_band_count": 3, "active_band_mae_db": 1.3379834608573455}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 7.865643818167715e-05, "log_rms_mae_db": 0.8446349165737279, "rms_envelope_nrmse": 0.09252675604730719, "envelope_correlation": 0.9939078887665339, "rms_gain_error_db": -0.3669796848214601, "reference_mean_power": 0.0003295653902210026, "output_mean_power": 0.0003028611930968346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 2.853826866715288e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001952923309195638, "output_power": 0.00017294295182089728, "reference_active": true, "delta_db": -0.5278232167034801}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5170390000310775e-05, "output_power": 1.003279403700623e-05, "reference_active": true, "delta_db": -1.7957484905396384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.363045655471935e-07, "output_power": 4.1669929867604306e-07, "reference_active": true, "delta_db": -2.472347238610647}], "active_band_count": 3, "active_band_mae_db": 1.5986396486179217}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8292179682281434, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e649b616a5f43f59a1ba2c734f881f904ed302d6feafaca18248ed984ac9989a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9914942475395242, "standard_deviation": 0.0009156626061967974, "minimum": 0.9903938365591392, "maximum": 0.9932634264994654, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5271151928352136, "standard_deviation": 0.027769197779422605, "minimum": 1.500526935365583, "maximum": 1.5864037729059681, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.34240304170279257, "standard_deviation": 0.172070859422385, "minimum": 0.11132590133686758, "maximum": 0.5635051675063433, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10979985717304225, "standard_deviation": 0.005766525406888966, "minimum": 0.09873731182368033, "maximum": 0.1164781109335141, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973320236836615, "standard_deviation": 0.0005345039996604209, "minimum": 0.99650527861249, "maximum": 0.9979533351829073, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8177518301245215, "standard_deviation": 0.027951961843917406, "minimum": 0.7845889051596651, "maximum": 0.8675666073841065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.403204607461033, "standard_deviation": 0.16569927456934433, "minimum": 0.15120151667815068, "maximum": 0.6815147125131663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06084823635147614, "standard_deviation": 0.006163206064114097, "minimum": 0.0525633038809765, "maximum": 0.06912995631845102, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.984277273203741, "standard_deviation": 0.0010054214558126713, "minimum": 0.9828722222450508, "maximum": 0.9854224758219108, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6062924266643073, "standard_deviation": 0.030011013952068424, "minimum": 1.5602060988431625, "maximum": 1.6417964813594947, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5381042720683726, "standard_deviation": 0.4155830033799307, "minimum": 0.09668721792114156, "maximum": 1.3379834608573455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1420210484469478, "standard_deviation": 0.005245023559229304, "minimum": 0.13606874448210293, "maximum": 0.15046272355402068, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9953522776701094, "standard_deviation": 0.0009657453274651421, "minimum": 0.9939078887665339, "maximum": 0.996826335821908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8360934799959483, "standard_deviation": 0.033589704483569606, "minimum": 0.7820604821103279, "maximum": 0.8884002141359949, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6892608211664606, "standard_deviation": 0.4552413622456925, "minimum": 0.20847682104592155, "maximum": 1.5986396486179217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0750319447624988, "standard_deviation": 0.008948516065968574, "minimum": 0.06637035552931368, "maximum": 0.09252675604730719, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9896490370253084, "standard_deviation": 0.0015800308455250444, "minimum": 0.988079587268961, "maximum": 0.9925791298520352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8364395772193889, "standard_deviation": 0.05342814399028971, "minimum": 1.7468276816014796, "maximum": 1.887863238551953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4615771165107264, "standard_deviation": 0.23452424670232652, "minimum": 0.2081222198721118, "maximum": 0.8810681387285381, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11813624253927442, "standard_deviation": 0.00925438263545268, "minimum": 0.10045485394637595, "maximum": 0.1266838221692289, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958356030311947, "standard_deviation": 0.00104352869659955, "minimum": 0.9942312389874838, "maximum": 0.9971672815222479, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0229886185449542, "standard_deviation": 0.020503757849609585, "minimum": 0.9826425781372196, "maximum": 1.045139106619611, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6579946890244239, "standard_deviation": 0.29199471344937666, "minimum": 0.27193280719369617, "maximum": 1.139949110255296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07272049402504427, "standard_deviation": 0.00881562266039757, "minimum": 0.060314438055315424, "maximum": 0.08476704663497775, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4103598354324093, "standard_deviation": 0.053869538038578746, "minimum": 0.32225742111030886, "maximum": 0.4881144306127682, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.82690556188848, "standard_deviation": 0.005359985164286412, "minimum": 5.818812582526822, "maximum": 5.833851852229421, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5106439380971475, "standard_deviation": 0.00967425944511254, "minimum": 0.4981321278417582, "maximum": 0.5238933856690584, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08434441978543622, "standard_deviation": 0.04076009094066061, "minimum": -0.13976155372507, "maximum": -0.038959105674021076, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933607577609571, "standard_deviation": 0.0005352423695647575, "minimum": 0.9927199965154176, "maximum": 0.9943189846391576, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.968835566527257, "standard_deviation": 0.022806893559848614, "minimum": 0.9310543797604347, "maximum": 0.9977489439627336, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40557106310864854, "standard_deviation": 0.09419558109313289, "minimum": 0.24170612079399842, "maximum": 0.5231402395941185, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08493754951000004, "standard_deviation": 0.003230225802408365, "minimum": 0.0791342262468137, "maximum": 0.08862373315291369, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980521623403673, "standard_deviation": 0.00024328559964676335, "minimum": 0.9977162649950632, "maximum": 0.9983524438055656, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48690567024445297, "standard_deviation": 0.01921518052046723, "minimum": 0.4553426947567303, "maximum": 0.5128584672497246, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43671633283559397, "standard_deviation": 0.14744409387059895, "minimum": 0.24876944805270232, "maximum": 0.6652642862749684, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04447823182638383, "standard_deviation": 0.0024639072501129265, "minimum": 0.04080517355947527, "maximum": 0.048474469249048135, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0016", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.9577364921569824, "predicted_mos": 1.9577364921569824, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "7a84b766c77cf2fe1eddcf8be5bd7c56c7f102dcb1ac799c3764b5de2efe391c", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6497621000744402, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9652883516047819, "全帯域_10ms_rms_envelope_nrmse": 0.09194635996484313, "全帯域_10ms_envelope_correlation": 0.9921691490189193, "全帯域_10ms_modulation_mae_db": 0.5368718200699555, "全帯域_40ms_log_rms_mae_db": 0.5006605258692524, "全帯域_40ms_rms_envelope_nrmse": 0.05123035520732577, "全帯域_40ms_envelope_correlation": 0.9973256685641699, "全帯域_40ms_modulation_mae_db": 0.5605243768453999, "300–1000 Hz_10ms_log_rms_mae_db": 2.481157769345635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13702160916452374, "300–1000 Hz_10ms_envelope_correlation": 0.9862971740908073, "300–1000 Hz_10ms_modulation_mae_db": 0.5245895902942675, "300–1000 Hz_40ms_log_rms_mae_db": 1.6240607960342102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0938003498471896, "300–1000 Hz_40ms_envelope_correlation": 0.9933024383332205, "300–1000 Hz_40ms_modulation_mae_db": 0.525185855603838, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8022028368723075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11761924506158934, "1000–3000 Hz_10ms_envelope_correlation": 0.9901975113261798, "1000–3000 Hz_10ms_modulation_mae_db": 0.2014371132937962, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0207521173804628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06425418899591402, "1000–3000 Hz_40ms_envelope_correlation": 0.9968756912382902, "1000–3000 Hz_40ms_modulation_mae_db": 0.2005742500299681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8771747698990446, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15241427259103893, "3000–6000 Hz_10ms_envelope_correlation": 0.9821777075790157, "3000–6000 Hz_10ms_modulation_mae_db": 0.8649606471777046, "3000–6000 Hz_40ms_log_rms_mae_db": 0.964565783685839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08891525963881547, "3000–6000 Hz_40ms_envelope_correlation": 0.9939818871207566, "3000–6000 Hz_40ms_modulation_mae_db": 1.0921976593258356, "mr_stft_db": 6.169110687954929, "band_mae_db": 0.6117808405129632, "rms_gain_error_db": -0.0642011619985787, "output_peak": 0.5464727919802213}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9501116529361743, "全帯域_10ms_rms_envelope_nrmse": 0.08710243116513249, "全帯域_10ms_envelope_correlation": 0.9930430450321389, "全帯域_10ms_modulation_mae_db": 0.5795208932604597, "全帯域_40ms_log_rms_mae_db": 0.4707098650611207, "全帯域_40ms_rms_envelope_nrmse": 0.0463939198971536, "全帯域_40ms_envelope_correlation": 0.9979189925368509, "全帯域_40ms_modulation_mae_db": 0.6098630120157961, "300–1000 Hz_10ms_log_rms_mae_db": 2.4685586182458383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13201585951500597, "300–1000 Hz_10ms_envelope_correlation": 0.9871720851183752, "300–1000 Hz_10ms_modulation_mae_db": 0.45897770601941823, "300–1000 Hz_40ms_log_rms_mae_db": 1.5519738908371108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07644749863717337, "300–1000 Hz_40ms_envelope_correlation": 0.9955842144722314, "300–1000 Hz_40ms_modulation_mae_db": 0.7929559433941503, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7960102634153496, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11581174579530086, "1000–3000 Hz_10ms_envelope_correlation": 0.9906516351295281, "1000–3000 Hz_10ms_modulation_mae_db": 0.911544724829402, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0390182928463796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06246975197801713, "1000–3000 Hz_40ms_envelope_correlation": 0.997371773068141, "1000–3000 Hz_40ms_modulation_mae_db": 1.118007894667698, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9068039201236207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1619070919866245, "3000–6000 Hz_10ms_envelope_correlation": 0.9798824731237817, "3000–6000 Hz_10ms_modulation_mae_db": 0.48277162719316885, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9893273302176842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08588536314754981, "3000–6000 Hz_40ms_envelope_correlation": 0.9938931322133269, "3000–6000 Hz_40ms_modulation_mae_db": 0.3259406739978208, "mr_stft_db": 6.168264265189895, "band_mae_db": 0.6081263811378187, "rms_gain_error_db": -0.11177647089551948, "output_peak": 0.5426568556965847}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9674446131599721, "全帯域_10ms_rms_envelope_nrmse": 0.08591746894208119, "全帯域_10ms_envelope_correlation": 0.9934857588714714, "全帯域_10ms_modulation_mae_db": 0.7638812621631578, "全帯域_40ms_log_rms_mae_db": 0.4837756858451238, "全帯域_40ms_rms_envelope_nrmse": 0.05093888874306477, "全帯域_40ms_envelope_correlation": 0.9977749638929992, "全帯域_40ms_modulation_mae_db": 0.7693688430280918, "300–1000 Hz_10ms_log_rms_mae_db": 2.43006408942373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13424863547407295, "300–1000 Hz_10ms_envelope_correlation": 0.9866519722563621, "300–1000 Hz_10ms_modulation_mae_db": 0.292662192032277, "300–1000 Hz_40ms_log_rms_mae_db": 1.5085462130139984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07347294199439404, "300–1000 Hz_40ms_envelope_correlation": 0.9957927926031837, "300–1000 Hz_40ms_modulation_mae_db": 0.6561842542881312, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7662740404924366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12531894758925494, "1000–3000 Hz_10ms_envelope_correlation": 0.9894259860512935, "1000–3000 Hz_10ms_modulation_mae_db": 0.8414641300688087, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0104421691604595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07543499016792603, "1000–3000 Hz_40ms_envelope_correlation": 0.9965824308447891, "1000–3000 Hz_40ms_modulation_mae_db": 0.8114649665812047, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8779803044126657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1461460520761616, "3000–6000 Hz_10ms_envelope_correlation": 0.9831571671905933, "3000–6000 Hz_10ms_modulation_mae_db": 0.6110184421221903, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9510465317486713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07135907414232334, "3000–6000 Hz_40ms_envelope_correlation": 0.9956281411111322, "3000–6000 Hz_40ms_modulation_mae_db": 0.6904162948376841, "mr_stft_db": 6.159137048839881, "band_mae_db": 0.6240129007093347, "rms_gain_error_db": -0.18937872137135264, "output_peak": 0.5020473298951719}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.985407478182937, "全帯域_10ms_rms_envelope_nrmse": 0.07950110109341163, "全帯域_10ms_envelope_correlation": 0.9941449294843565, "全帯域_10ms_modulation_mae_db": 0.1501716942752809, "全帯域_40ms_log_rms_mae_db": 0.5019938739627474, "全帯域_40ms_rms_envelope_nrmse": 0.037349478646104245, "全帯域_40ms_envelope_correlation": 0.9985603733711517, "全帯域_40ms_modulation_mae_db": 0.3910235636669516, "300–1000 Hz_10ms_log_rms_mae_db": 2.3973790534868837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12352014047282467, "300–1000 Hz_10ms_envelope_correlation": 0.9887400017081248, "300–1000 Hz_10ms_modulation_mae_db": 0.4102311867857013, "300–1000 Hz_40ms_log_rms_mae_db": 1.5209919354154768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07417878401733065, "300–1000 Hz_40ms_envelope_correlation": 0.9956957755545935, "300–1000 Hz_40ms_modulation_mae_db": 0.7596397137043032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7886034260489505, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11627984132937004, "1000–3000 Hz_10ms_envelope_correlation": 0.9903804521566516, "1000–3000 Hz_10ms_modulation_mae_db": 0.426692006915715, "1000–3000 Hz_40ms_log_rms_mae_db": 1.011772692159098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06135110686509744, "1000–3000 Hz_40ms_envelope_correlation": 0.9971445119239764, "1000–3000 Hz_40ms_modulation_mae_db": 0.47932367324140274, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8855708943419807, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14294254986864216, "3000–6000 Hz_10ms_envelope_correlation": 0.9839592021874768, "3000–6000 Hz_10ms_modulation_mae_db": 0.21118136539348673, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9783688680685975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07055124622815996, "3000–6000 Hz_40ms_envelope_correlation": 0.995646986344242, "3000–6000 Hz_40ms_modulation_mae_db": 0.21795301105361845, "mr_stft_db": 6.171334527752077, "band_mae_db": 0.5175304404569574, "rms_gain_error_db": -0.011445021462511005, "output_peak": 0.5328362473347101}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9320134318826284, "全帯域_10ms_rms_envelope_nrmse": 0.0793347770404186, "全帯域_10ms_envelope_correlation": 0.9941862231314013, "全帯域_10ms_modulation_mae_db": 0.3943340461908654, "全帯域_40ms_log_rms_mae_db": 0.4481276450456377, "全帯域_40ms_rms_envelope_nrmse": 0.038965988391881666, "全帯域_40ms_envelope_correlation": 0.9984718058988258, "全帯域_40ms_modulation_mae_db": 0.41832674312113055, "300–1000 Hz_10ms_log_rms_mae_db": 2.4840159995016484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11794629487508536, "300–1000 Hz_10ms_envelope_correlation": 0.9897694884256883, "300–1000 Hz_10ms_modulation_mae_db": 0.4143183046067757, "300–1000 Hz_40ms_log_rms_mae_db": 1.606145590072349, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06979289745588799, "300–1000 Hz_40ms_envelope_correlation": 0.9961569033171782, "300–1000 Hz_40ms_modulation_mae_db": 0.19257708122985015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8102601912283767, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11288468060228204, "1000–3000 Hz_10ms_envelope_correlation": 0.9909784257258323, "1000–3000 Hz_10ms_modulation_mae_db": 0.4257103844892945, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0252711379651631, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06142154769527582, "1000–3000 Hz_40ms_envelope_correlation": 0.9972499175268811, "1000–3000 Hz_40ms_modulation_mae_db": 0.5701117414737175, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8655889167895963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13646975120463392, "3000–6000 Hz_10ms_envelope_correlation": 0.9853829818087158, "3000–6000 Hz_10ms_modulation_mae_db": 0.4652762504312733, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9509633663593735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06756224417411154, "3000–6000 Hz_40ms_envelope_correlation": 0.9961203202425789, "3000–6000 Hz_40ms_modulation_mae_db": 0.5516569960407318, "mr_stft_db": 6.1593998596863315, "band_mae_db": 0.5607820992378011, "rms_gain_error_db": -0.060129865534993596, "output_peak": 0.5042655444664232}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.960274405857894, "全帯域_10ms_rms_envelope_nrmse": 0.08784112739509359, "全帯域_10ms_envelope_correlation": 0.9928754838870779, "全帯域_10ms_modulation_mae_db": 0.22163844135225422, "全帯域_40ms_log_rms_mae_db": 0.47120310868210813, "全帯域_40ms_rms_envelope_nrmse": 0.04560860412667215, "全帯域_40ms_envelope_correlation": 0.9978942887782327, "全帯域_40ms_modulation_mae_db": 0.5375234838259811, "300–1000 Hz_10ms_log_rms_mae_db": 2.3731337604177627, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11011940910640701, "300–1000 Hz_10ms_envelope_correlation": 0.9911134204707748, "300–1000 Hz_10ms_modulation_mae_db": 0.25879198490444927, "300–1000 Hz_40ms_log_rms_mae_db": 1.4750395137770878, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0674688048946055, "300–1000 Hz_40ms_envelope_correlation": 0.9964269196406615, "300–1000 Hz_40ms_modulation_mae_db": 0.45307964436700215, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7394193139943048, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11308250634540447, "1000–3000 Hz_10ms_envelope_correlation": 0.9910918868734956, "1000–3000 Hz_10ms_modulation_mae_db": 0.22621835891732697, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9773676748313127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061188189930455154, "1000–3000 Hz_40ms_envelope_correlation": 0.9974816801042903, "1000–3000 Hz_40ms_modulation_mae_db": 0.2890211991105425, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8910130792612787, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15726292934064928, "3000–6000 Hz_10ms_envelope_correlation": 0.9808663641766479, "3000–6000 Hz_10ms_modulation_mae_db": 0.2393033066819815, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0011411689561154, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08632376739810003, "3000–6000 Hz_40ms_envelope_correlation": 0.9936726746757403, "3000–6000 Hz_40ms_modulation_mae_db": 0.6230584536983114, "mr_stft_db": 6.174748640453228, "band_mae_db": 0.5776262619343423, "rms_gain_error_db": -0.051803737804486244, "output_peak": 0.5075337722689444}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9378417576854335, "全帯域_10ms_rms_envelope_nrmse": 0.08372537479031568, "全帯域_10ms_envelope_correlation": 0.9935601246588858, "全帯域_10ms_modulation_mae_db": 0.35478108777584527, "全帯域_40ms_log_rms_mae_db": 0.453503047976976, "全帯域_40ms_rms_envelope_nrmse": 0.04182907514482284, "全帯域_40ms_envelope_correlation": 0.9983082720183544, "全帯域_40ms_modulation_mae_db": 0.21340062269422433, "300–1000 Hz_10ms_log_rms_mae_db": 2.4020670623944445, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11895922363984585, "300–1000 Hz_10ms_envelope_correlation": 0.9896010265654973, "300–1000 Hz_10ms_modulation_mae_db": 0.2939422306223034, "300–1000 Hz_40ms_log_rms_mae_db": 1.5086421199743756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06454747260753582, "300–1000 Hz_40ms_envelope_correlation": 0.9968217307073292, "300–1000 Hz_40ms_modulation_mae_db": 0.40892511357865696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.766841068585367, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11135688682648837, "1000–3000 Hz_10ms_envelope_correlation": 0.9913871784663154, "1000–3000 Hz_10ms_modulation_mae_db": 0.824618243891225, "1000–3000 Hz_40ms_log_rms_mae_db": 1.022251638405492, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06431562554512017, "1000–3000 Hz_40ms_envelope_correlation": 0.9972241856504567, "1000–3000 Hz_40ms_modulation_mae_db": 0.9191593101731049, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9310641836397013, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16675232233753032, "3000–6000 Hz_10ms_envelope_correlation": 0.9780529893614981, "3000–6000 Hz_10ms_modulation_mae_db": 0.30124103704499045, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0123619541094875, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0872824724487706, "3000–6000 Hz_40ms_envelope_correlation": 0.993347222382181, "3000–6000 Hz_40ms_modulation_mae_db": 0.3221849577757467, "mr_stft_db": 6.164918278086058, "band_mae_db": 0.6836621295323898, "rms_gain_error_db": -0.09973274086434182, "output_peak": 0.5425199394074268}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9315817976923204, "全帯域_10ms_rms_envelope_nrmse": 0.08085258169573303, "全帯域_10ms_envelope_correlation": 0.9940726178740538, "全帯域_10ms_modulation_mae_db": 0.4873245690933336, "全帯域_40ms_log_rms_mae_db": 0.45487151559178407, "全帯域_40ms_rms_envelope_nrmse": 0.046964490129855305, "全帯域_40ms_envelope_correlation": 0.9978992402103751, "全帯域_40ms_modulation_mae_db": 0.46033938519937, "300–1000 Hz_10ms_log_rms_mae_db": 2.476344030958428, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12146809914123781, "300–1000 Hz_10ms_envelope_correlation": 0.9891984098964512, "300–1000 Hz_10ms_modulation_mae_db": 0.47559597708290663, "300–1000 Hz_40ms_log_rms_mae_db": 1.556031198474013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07712581253131541, "300–1000 Hz_40ms_envelope_correlation": 0.9954374124988873, "300–1000 Hz_40ms_modulation_mae_db": 0.5097476462950388, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7597569674300848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11530516022154011, "1000–3000 Hz_10ms_envelope_correlation": 0.9907449860867424, "1000–3000 Hz_10ms_modulation_mae_db": 0.45878196386253417, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0144267767222175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06781520773384514, "1000–3000 Hz_40ms_envelope_correlation": 0.9968476175350482, "1000–3000 Hz_40ms_modulation_mae_db": 0.6627753928363538, "3000–6000 Hz_10ms_log_rms_mae_db": 1.844075590886971, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13673084564935623, "3000–6000 Hz_10ms_envelope_correlation": 0.9852951665837362, "3000–6000 Hz_10ms_modulation_mae_db": 0.5197216644771671, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9529673401280996, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06849710050539434, "3000–6000 Hz_40ms_envelope_correlation": 0.9959088644092514, "3000–6000 Hz_40ms_modulation_mae_db": 0.7842481970065558, "mr_stft_db": 6.161946958914164, "band_mae_db": 0.6603431704670076, "rms_gain_error_db": -0.11801281152020462, "output_peak": 0.522957969906134}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 95360, "output_samples": 95360, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.269960821363042, "spectral_convergence": 0.4155997553017588, "spectral_convergence_status": "ok", "amplitude_floor": 1.2569032294341943e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.147259802454069, "spectral_convergence": 0.42351849944053255, "spectral_convergence_status": "ok", "amplitude_floor": 9.073914887165124e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.090111440047677, "spectral_convergence": 0.43582421774131236, "spectral_convergence_status": "ok", "amplitude_floor": 7.969434026985316e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.036863249031963e-06, "output_power": 9.682430461680655e-06, "delta_db": 2.8382422669524225, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00014485054714414723, "output_power": 0.00015257496538685804, "delta_db": 0.2256314010068591, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004849180298355377, "output_power": 0.00045913227073279925, "delta_db": -0.2373051319284379, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006903353830986323, "output_power": 0.00069157033842888, "delta_db": 0.007762243396870488, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00029452981369653373, "output_power": 0.0003011792695469036, "delta_db": 0.09695812975183549, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004100383435072048, "output_power": 0.00038578548488908654, "delta_db": -0.26478587004135345, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6117808405129632, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0642011619985787, "pitch": null, "output_rms_dbfs": -26.989863429220154, "output_peak": 0.5464727919802213, "reference_rms_dbfs": -26.925662267221572, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.169110687954929}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 95360, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.0001898595271537964, "log_rms_mae_db": 0.9652883516047819, "rms_envelope_nrmse": 0.09194635996484313, "envelope_correlation": 0.9921691490189193, "rms_gain_error_db": -0.06420184200703735, "reference_mean_power": 0.002031411160407025, "output_mean_power": 0.002001601647900678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 9.927152563814217e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006381525961065371, "output_power": 0.0006105646581746694, "reference_active": true, "delta_db": -0.19192878296761157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010674365906282643, "output_power": 9.798564782919784e-05, "reference_active": true, "delta_db": -0.3717961759675115}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4567838102832594e-05, "output_power": 2.716333996909274e-05, "reference_active": true, "delta_db": -1.0468905012747434}], "active_band_count": 3, "active_band_mae_db": 0.5368718200699555}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00017019756464998765, "log_rms_mae_db": 0.5006605258692524, "rms_envelope_nrmse": 0.05123035520732577, "envelope_correlation": 0.9973256685641699, "rms_gain_error_db": -0.0641983470301624, "reference_mean_power": 0.0020416854120747243, "output_mean_power": 0.0020117267511727157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 8.390142156345837e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005509511464645055, "output_power": 0.0005300734962221453, "reference_active": true, "delta_db": -0.16777001156867816}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5897699076585876e-05, "output_power": 4.1431799193619605e-05, "reference_active": true, "delta_db": -0.44457121089969054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848237969563024e-06, "output_power": 1.4448897066748943e-06, "reference_active": true, "delta_db": -1.0692319080678312}], "active_band_count": 3, "active_band_mae_db": 0.5605243768453999}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 9.246932648014587e-05, "log_rms_mae_db": 2.481157769345635, "rms_envelope_nrmse": 0.13702160916452374, "envelope_correlation": 0.9862971740908073, "rms_gain_error_db": -0.12921771849390107, "reference_mean_power": 0.0005236182881812637, "output_mean_power": 0.0005082683122996588}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.7855833380403224e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028934531457073826, "output_power": 0.00027089553459875296, "reference_active": true, "delta_db": -0.28614607887775195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6197786234688146e-05, "output_power": 3.401762957207754e-05, "reference_active": true, "delta_db": -0.26977963313623377}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0243815325354375e-05, "output_power": 8.103589528218048e-06, "reference_active": true, "delta_db": -1.017843058868817}], "active_band_count": 3, "active_band_mae_db": 0.5245895902942675}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 8.334521268786246e-05, "log_rms_mae_db": 1.6240607960342102, "rms_envelope_nrmse": 0.0938003498471896, "envelope_correlation": 0.9933024383332205, "rms_gain_error_db": -0.12921809543949717, "reference_mean_power": 0.000526268698480611, "output_mean_power": 0.0005108409809473252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.428386331422997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002786455084699618, "output_power": 0.00026016415045905045, "reference_active": true, "delta_db": -0.29804594618863917}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6309295594140896e-05, "output_power": 1.5337550911196157e-05, "reference_active": true, "delta_db": -0.26679186743175703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.788822138965803e-07, "output_power": 6.1716290035107e-07, "reference_active": true, "delta_db": -1.0107197531911178}], "active_band_count": 3, "active_band_mae_db": 0.525185855603838}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00017019739791537767, "log_rms_mae_db": 1.8022028368723075, "rms_envelope_nrmse": 0.11761924506158934, "envelope_correlation": 0.9901975113261798, "rms_gain_error_db": 0.0133474443564422, "reference_mean_power": 0.0008446325483556021, "output_mean_power": 0.0008472324035771752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 4.004634586327471e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028725854625925005, "output_power": 0.00027952001128480773, "reference_active": true, "delta_db": -0.11860053121146892}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3436307686794414e-05, "output_power": 4.004380650818063e-05, "reference_active": true, "delta_db": -0.3531754665369402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1340506745101254e-05, "output_power": 1.1691924316503533e-05, "reference_active": true, "delta_db": 0.13253534213297957}], "active_band_count": 3, "active_band_mae_db": 0.2014371132937962}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00015333717053704127, "log_rms_mae_db": 1.0207521173804628, "rms_envelope_nrmse": 0.06425418899591402, "envelope_correlation": 0.9968756912382902, "rms_gain_error_db": 0.013354293241535576, "reference_mean_power": 0.0008489024890862695, "output_mean_power": 0.0008515168304237879}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.579273675799131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002706567111490358, "output_power": 0.00026494595339247363, "reference_active": true, "delta_db": -0.09261509461509512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.939557003916555e-05, "output_power": 1.798245268824992e-05, "reference_active": true, "delta_db": -0.32853621766203955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.660068008029229e-07, "output_power": 7.348107168423681e-07, "reference_active": true, "delta_db": -0.18057143781276958}], "active_band_count": 3, "active_band_mae_db": 0.2005742500299681}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00010107776753147987, "log_rms_mae_db": 1.8771747698990446, "rms_envelope_nrmse": 0.15241427259103893, "envelope_correlation": 0.9821777075790157, "rms_gain_error_db": -0.28821985783608606, "reference_mean_power": 0.00032790731816785513, "output_mean_power": 0.00030685211673296415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.172510420573334e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020494142716961824, "output_power": 0.00018221988773732794, "reference_active": true, "delta_db": -0.5103398147747473}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6580284356723e-05, "output_power": 2.899490679088968e-05, "reference_active": true, "delta_db": -1.0092536016804279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.565019545039446e-06, "output_power": 7.467186687614549e-06, "reference_active": true, "delta_db": -1.0752885250779385}], "active_band_count": 3, "active_band_mae_db": 0.8649606471777046}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 7.865643818167715e-05, "log_rms_mae_db": 0.964565783685839, "rms_envelope_nrmse": 0.08891525963881547, "envelope_correlation": 0.9939818871207566, "rms_gain_error_db": -0.2882186375124094, "reference_mean_power": 0.0003295653902210026, "output_mean_power": 0.00030840380925984355}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 2.853826866715288e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001952923309195638, "output_power": 0.00017207914973270746, "reference_active": true, "delta_db": -0.549569375136993}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5170390000310775e-05, "output_power": 1.241284666269415e-05, "reference_active": true, "delta_db": -0.8712535523655176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.363045655471935e-07, "output_power": 4.802645027813926e-07, "reference_active": true, "delta_db": -1.8557700504749963}], "active_band_count": 3, "active_band_mae_db": 1.0921976593258356}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7976415082971156, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "201b5ab35a224bfc4f709fb27146ece6a5e9428d8cbcaf4794930ed0bc54656f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9906072577270049, "standard_deviation": 0.0006113897002601716, "minimum": 0.9894259860512935, "maximum": 0.9913871784663154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7786710135083972, "standard_deviation": 0.02432809914688222, "minimum": 1.7394193139943048, "maximum": 1.8102601912283767, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5395583657835128, "standard_deviation": 0.28170730051292303, "minimum": 0.2014371132937962, "maximum": 0.911544724829402, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11595737672140377, "standard_deviation": 0.0043065107194823786, "minimum": 0.11135688682648837, "maximum": 0.12531894758925494, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970972259864841, "standard_deviation": 0.00030266590562911107, "minimum": 0.9965824308447891, "maximum": 0.9974816801042903, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.015162812433823, "standard_deviation": 0.017790629706824622, "minimum": 0.9773676748313127, "maximum": 1.0390182928463796, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.631304803514249, "standard_deviation": 0.31229974809927735, "minimum": 0.2005742500299681, "maximum": 1.118007894667698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06478132611395636, "standard_deviation": 0.004847801618942921, "minimum": 0.061188189930455154, "maximum": 0.07543499016792603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823467565014332, "standard_deviation": 0.0026158254773688576, "minimum": 0.9780529893614981, "maximum": 0.9853829818087158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8849089574193574, "standard_deviation": 0.026174809167302953, "minimum": 1.844075590886971, "maximum": 1.9310641836397013, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4619342925652454, "standard_deviation": 0.21603735736502466, "minimum": 0.21118136539348673, "maximum": 0.8649606471777046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15007822688182965, "standard_deviation": 0.011368802230359623, "minimum": 0.13646975120463392, "maximum": 0.16675232233753032, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947749035624012, "standard_deviation": 0.0011491695012454497, "minimum": 0.993347222382181, "maximum": 0.9961203202425789, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9750927929092335, "standard_deviation": 0.024028758000934847, "minimum": 0.9509633663593735, "maximum": 1.0123619541094875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5759570304670381, "standard_deviation": 0.2880661821732393, "minimum": 0.21795301105361845, "maximum": 1.0921976593258356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07829706596040313, "standard_deviation": 0.009523676754120391, "minimum": 0.06756224417411154, "maximum": 0.08891525963881547, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9885679473165102, "standard_deviation": 0.0016984562687946579, "minimum": 0.9862971740908073, "maximum": 0.9911134204707748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.439090047971796, "standard_deviation": 0.0440559684661242, "minimum": 2.3731337604177627, "maximum": 2.4840159995016484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3911386465435124, "standard_deviation": 0.09791143878106622, "minimum": 0.25879198490444927, "maximum": 0.5245895902942675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12441240892362541, "standard_deviation": 0.009249390341671831, "minimum": 0.11011940910640701, "maximum": 0.13702160916452374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956522733909107, "standard_deviation": 0.0010567144968254925, "minimum": 0.9933024383332205, "maximum": 0.9968217307073292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.543928907199828, "standard_deviation": 0.05109119025675908, "minimum": 1.4750395137770878, "maximum": 1.6240607960342102, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5372869065576213, "standard_deviation": 0.19720958011001924, "minimum": 0.19257708122985015, "maximum": 0.7929559433941503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07460432024817905, "standard_deviation": 0.008903526052918076, "minimum": 0.06454747260753582, "maximum": 0.0938003498471896, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.6054830279985768, "standard_deviation": 0.053480895376830484, "minimum": 0.5175304404569574, "maximum": 0.6836621295323898, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.16610753335957, "standard_deviation": 0.005706711340445073, "minimum": 6.159137048839881, "maximum": 6.174748640453228, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.525161306369452, "standard_deviation": 0.018543285072500804, "minimum": 0.5020473298951719, "maximum": 0.5464727919802213, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08831006643149851, "standard_deviation": 0.05397696052186742, "minimum": -0.18937872137135264, "maximum": -0.011445021462511005, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9934421664947881, "standard_deviation": 0.0007133202123161439, "minimum": 0.9921691490189193, "maximum": 0.9941862231314013, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9537454361252677, "standard_deviation": 0.01925637908181539, "minimum": 0.9315817976923204, "maximum": 0.985407478182937, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.43606547677264407, "standard_deviation": 0.19888795540929205, "minimum": 0.1501716942752809, "maximum": 0.7638812621631578, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08452765276087867, "standard_deviation": 0.004487432940494802, "minimum": 0.0793347770404186, "maximum": 0.09194635996484313, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.99801920065887, "standard_deviation": 0.0004074231176651556, "minimum": 0.9973256685641699, "maximum": 0.9985603733711517, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4731056585043438, "standard_deviation": 0.020899952757831613, "minimum": 0.4481276450456377, "maximum": 0.5019938739627474, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49504625379961814, "standard_deviation": 0.1655134445236446, "minimum": 0.21340062269422433, "maximum": 0.7693688430280918, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.044910100035860046, "standard_deviation": 0.0051435151977586285, "minimum": 0.037349478646104245, "maximum": 0.05123035520732577, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0016", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.9674909114837646, "predicted_mos": 1.9674909114837646, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "e9897ff644c3637b8798fb51dce54b224fff4643043291f80645a1fa89b4ad2f", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.990374700166285, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.011434058378268, "全帯域_10ms_rms_envelope_nrmse": 0.09788980292749953, "全帯域_10ms_envelope_correlation": 0.9911146333207217, "全帯域_10ms_modulation_mae_db": 0.2054675135225267, "全帯域_40ms_log_rms_mae_db": 0.5204360020719034, "全帯域_40ms_rms_envelope_nrmse": 0.05524951021669179, "全帯域_40ms_envelope_correlation": 0.9968438713184906, "全帯域_40ms_modulation_mae_db": 0.23084832828028745, "300–1000 Hz_10ms_log_rms_mae_db": 1.749284913717138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1218946366917546, "300–1000 Hz_10ms_envelope_correlation": 0.9889992316340761, "300–1000 Hz_10ms_modulation_mae_db": 0.17946627491422085, "300–1000 Hz_40ms_log_rms_mae_db": 0.9434467371636925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06085558245139891, "300–1000 Hz_40ms_envelope_correlation": 0.9970307765214856, "300–1000 Hz_40ms_modulation_mae_db": 0.3910751853898167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3861423680967546, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12825488045800634, "1000–3000 Hz_10ms_envelope_correlation": 0.9883924029120633, "1000–3000 Hz_10ms_modulation_mae_db": 0.2826068002571444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.765148036142468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08341663910448698, "1000–3000 Hz_40ms_envelope_correlation": 0.9949526284218249, "1000–3000 Hz_40ms_modulation_mae_db": 0.24994702693765233, "3000–6000 Hz_10ms_log_rms_mae_db": 1.391446514051296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14689659856211607, "3000–6000 Hz_10ms_envelope_correlation": 0.9839033743291477, "3000–6000 Hz_10ms_modulation_mae_db": 0.7600524817900695, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7894875714436845, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08360023571769212, "3000–6000 Hz_40ms_envelope_correlation": 0.9946486509711691, "3000–6000 Hz_40ms_modulation_mae_db": 0.6598230626756989, "mr_stft_db": 5.74197019652662, "band_mae_db": 0.41497495669281265, "rms_gain_error_db": -0.014076089895672633, "output_peak": 0.4917664196682542}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0065210892507601, "全帯域_10ms_rms_envelope_nrmse": 0.08733095459267358, "全帯域_10ms_envelope_correlation": 0.9930395556237969, "全帯域_10ms_modulation_mae_db": 0.7050674850197155, "全帯域_40ms_log_rms_mae_db": 0.5092393085025204, "全帯域_40ms_rms_envelope_nrmse": 0.047548216319924376, "全帯域_40ms_envelope_correlation": 0.9978415047174561, "全帯域_40ms_modulation_mae_db": 0.7035217502385195, "300–1000 Hz_10ms_log_rms_mae_db": 1.6988418754316241, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10361676583818, "300–1000 Hz_10ms_envelope_correlation": 0.9921270593179428, "300–1000 Hz_10ms_modulation_mae_db": 0.5451459295423338, "300–1000 Hz_40ms_log_rms_mae_db": 0.977968980179848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05782880590613671, "300–1000 Hz_40ms_envelope_correlation": 0.9974836147622421, "300–1000 Hz_40ms_modulation_mae_db": 0.8347987097738594, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4128562459538514, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11215166583986605, "1000–3000 Hz_10ms_envelope_correlation": 0.9911169537145863, "1000–3000 Hz_10ms_modulation_mae_db": 0.24103335129031003, "1000–3000 Hz_40ms_log_rms_mae_db": 0.763816514468652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060422451245748694, "1000–3000 Hz_40ms_envelope_correlation": 0.997386564813335, "1000–3000 Hz_40ms_modulation_mae_db": 0.3261915596053198, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4374712168937407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15604144095207786, "3000–6000 Hz_10ms_envelope_correlation": 0.9809126343076001, "3000–6000 Hz_10ms_modulation_mae_db": 0.6085603526253663, "3000–6000 Hz_40ms_log_rms_mae_db": 0.786513270164315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09140118097020648, "3000–6000 Hz_40ms_envelope_correlation": 0.9927892124697082, "3000–6000 Hz_40ms_modulation_mae_db": 0.5187771688495556, "mr_stft_db": 5.710693698165071, "band_mae_db": 0.44155406451237306, "rms_gain_error_db": -0.1217781458123278, "output_peak": 0.48995765846763856}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.945361713144853, "全帯域_10ms_rms_envelope_nrmse": 0.08201129900486283, "全帯域_10ms_envelope_correlation": 0.9937921221335547, "全帯域_10ms_modulation_mae_db": 0.04263816078673677, "全帯域_40ms_log_rms_mae_db": 0.5018852993764806, "全帯域_40ms_rms_envelope_nrmse": 0.047903650565520414, "全帯域_40ms_envelope_correlation": 0.9976477527710512, "全帯域_40ms_modulation_mae_db": 0.08103691153940921, "300–1000 Hz_10ms_log_rms_mae_db": 1.7385031750112567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1312351193684153, "300–1000 Hz_10ms_envelope_correlation": 0.987361893033966, "300–1000 Hz_10ms_modulation_mae_db": 0.3824821513216632, "300–1000 Hz_40ms_log_rms_mae_db": 1.011301580160771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08655822143398949, "300–1000 Hz_40ms_envelope_correlation": 0.9940593333489843, "300–1000 Hz_40ms_modulation_mae_db": 0.11394934400151403, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3079804246888431, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10083287933791502, "1000–3000 Hz_10ms_envelope_correlation": 0.9927895588894324, "1000–3000 Hz_10ms_modulation_mae_db": 0.28769135957465, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7161176507111585, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059257589080963034, "1000–3000 Hz_40ms_envelope_correlation": 0.997375681167044, "1000–3000 Hz_40ms_modulation_mae_db": 0.2109877485208942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4050476370422904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1305890093349201, "3000–6000 Hz_10ms_envelope_correlation": 0.9869701335929896, "3000–6000 Hz_10ms_modulation_mae_db": 0.2511593271505431, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7539410175941086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06987107175597437, "3000–6000 Hz_40ms_envelope_correlation": 0.9960336825812116, "3000–6000 Hz_40ms_modulation_mae_db": 0.1852963726844368, "mr_stft_db": 5.727973475611265, "band_mae_db": 0.3226125565066265, "rms_gain_error_db": 0.011801534609115571, "output_peak": 0.5352827434081906}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9908491576286121, "全帯域_10ms_rms_envelope_nrmse": 0.09006202241071001, "全帯域_10ms_envelope_correlation": 0.9925095423776469, "全帯域_10ms_modulation_mae_db": 0.3897997753633852, "全帯域_40ms_log_rms_mae_db": 0.5114262345833748, "全帯域_40ms_rms_envelope_nrmse": 0.04771225009268662, "全帯域_40ms_envelope_correlation": 0.997730536989443, "全帯域_40ms_modulation_mae_db": 0.41253967654286616, "300–1000 Hz_10ms_log_rms_mae_db": 1.7776564393331875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14385526968973128, "300–1000 Hz_10ms_envelope_correlation": 0.9847363600026638, "300–1000 Hz_10ms_modulation_mae_db": 0.2776920056827217, "300–1000 Hz_40ms_log_rms_mae_db": 1.0107862091807072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07677213555645443, "300–1000 Hz_40ms_envelope_correlation": 0.9952920079843461, "300–1000 Hz_40ms_modulation_mae_db": 0.2380536498709758, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3804387551609112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10739555141188799, "1000–3000 Hz_10ms_envelope_correlation": 0.9918277838476185, "1000–3000 Hz_10ms_modulation_mae_db": 0.35595657107716594, "1000–3000 Hz_40ms_log_rms_mae_db": 0.757905146342486, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055892301896091436, "1000–3000 Hz_40ms_envelope_correlation": 0.9977250032357609, "1000–3000 Hz_40ms_modulation_mae_db": 0.49365804494237353, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4066178199330608, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13583639345387383, "3000–6000 Hz_10ms_envelope_correlation": 0.9858575224316691, "3000–6000 Hz_10ms_modulation_mae_db": 0.8588086960156686, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7906695526990949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08117575348746141, "3000–6000 Hz_40ms_envelope_correlation": 0.9949293196716251, "3000–6000 Hz_40ms_modulation_mae_db": 0.8001884139050043, "mr_stft_db": 5.710488993028034, "band_mae_db": 0.4851896180801882, "rms_gain_error_db": -0.08795372560621426, "output_peak": 0.5196182295752088}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9616230803199822, "全帯域_10ms_rms_envelope_nrmse": 0.08783468782329197, "全帯域_10ms_envelope_correlation": 0.9928509423250971, "全帯域_10ms_modulation_mae_db": 0.1047831106798877, "全帯域_40ms_log_rms_mae_db": 0.4962754282208235, "全帯域_40ms_rms_envelope_nrmse": 0.04529698049501676, "全帯域_40ms_envelope_correlation": 0.9978797478181816, "全帯域_40ms_modulation_mae_db": 0.1994054309228079, "300–1000 Hz_10ms_log_rms_mae_db": 1.697428875112361, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12301929175317848, "300–1000 Hz_10ms_envelope_correlation": 0.9889980221140384, "300–1000 Hz_10ms_modulation_mae_db": 0.38663525829840645, "300–1000 Hz_40ms_log_rms_mae_db": 0.978077827273095, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059090224245394736, "300–1000 Hz_40ms_envelope_correlation": 0.9973180809028883, "300–1000 Hz_40ms_modulation_mae_db": 0.2958632078588262, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4235609250915753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1105790718265143, "1000–3000 Hz_10ms_envelope_correlation": 0.991477649667235, "1000–3000 Hz_10ms_modulation_mae_db": 0.23669189818551742, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8166565827150121, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06673490187625804, "1000–3000 Hz_40ms_envelope_correlation": 0.9969217453973303, "1000–3000 Hz_40ms_modulation_mae_db": 0.1590790731081842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3798645158175782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14321400823072586, "3000–6000 Hz_10ms_envelope_correlation": 0.9838486669889575, "3000–6000 Hz_10ms_modulation_mae_db": 0.2551475052466849, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7973463358810878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07703962948132195, "3000–6000 Hz_40ms_envelope_correlation": 0.9948331845207804, "3000–6000 Hz_40ms_modulation_mae_db": 0.3963034206323315, "mr_stft_db": 5.730681546248216, "band_mae_db": 0.4990169865010117, "rms_gain_error_db": -0.012536200306321733, "output_peak": 0.4961343953973021}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0000446153022189, "全帯域_10ms_rms_envelope_nrmse": 0.0899289026793944, "全帯域_10ms_envelope_correlation": 0.9925344270868841, "全帯域_10ms_modulation_mae_db": 0.5516105298325091, "全帯域_40ms_log_rms_mae_db": 0.5386705622022572, "全帯域_40ms_rms_envelope_nrmse": 0.05298658373769512, "全帯域_40ms_envelope_correlation": 0.9971809652244429, "全帯域_40ms_modulation_mae_db": 0.41563198976394683, "300–1000 Hz_10ms_log_rms_mae_db": 1.7004366578737133, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12090637829399077, "300–1000 Hz_10ms_envelope_correlation": 0.9893118177009025, "300–1000 Hz_10ms_modulation_mae_db": 0.8461616690692851, "300–1000 Hz_40ms_log_rms_mae_db": 0.9879208470782366, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07381391829500546, "300–1000 Hz_40ms_envelope_correlation": 0.9959458858991109, "300–1000 Hz_40ms_modulation_mae_db": 0.8960242512416707, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4039051721142102, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1015653402299435, "1000–3000 Hz_10ms_envelope_correlation": 0.9928575334217796, "1000–3000 Hz_10ms_modulation_mae_db": 0.829705298054558, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7808173323517416, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06389104429913314, "1000–3000 Hz_40ms_envelope_correlation": 0.997210893605069, "1000–3000 Hz_40ms_modulation_mae_db": 1.043839622660848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.405920760534995, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14314543725991158, "3000–6000 Hz_10ms_envelope_correlation": 0.9839843570069392, "3000–6000 Hz_10ms_modulation_mae_db": 0.3192674996084692, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7482747888651613, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07559270425100877, "3000–6000 Hz_40ms_envelope_correlation": 0.9950616197960965, "3000–6000 Hz_40ms_modulation_mae_db": 0.23233076809089223, "mr_stft_db": 5.719803056678011, "band_mae_db": 0.494601388985241, "rms_gain_error_db": -0.08984195799793217, "output_peak": 0.49529198900692273}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.984195991662268, "全帯域_10ms_rms_envelope_nrmse": 0.09355947559795731, "全帯域_10ms_envelope_correlation": 0.9918993557812644, "全帯域_10ms_modulation_mae_db": 0.7408280574855265, "全帯域_40ms_log_rms_mae_db": 0.5026601855163907, "全帯域_40ms_rms_envelope_nrmse": 0.05160648556192606, "全帯域_40ms_envelope_correlation": 0.997307759594388, "全帯域_40ms_modulation_mae_db": 0.7592192609488402, "300–1000 Hz_10ms_log_rms_mae_db": 1.72194327766342, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12534720953089026, "300–1000 Hz_10ms_envelope_correlation": 0.9883549702324409, "300–1000 Hz_10ms_modulation_mae_db": 0.4887673738398872, "300–1000 Hz_40ms_log_rms_mae_db": 1.018819028345341, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07468757474481783, "300–1000 Hz_40ms_envelope_correlation": 0.9956241239959343, "300–1000 Hz_40ms_modulation_mae_db": 0.42133920521127116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3420631998626302, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1029349751653266, "1000–3000 Hz_10ms_envelope_correlation": 0.9925402172697096, "1000–3000 Hz_10ms_modulation_mae_db": 0.587899447032707, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7395901546610076, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05820979644011621, "1000–3000 Hz_40ms_envelope_correlation": 0.9975780532328, "1000–3000 Hz_40ms_modulation_mae_db": 0.5869777903680892, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4317362328585963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14844023355734698, "3000–6000 Hz_10ms_envelope_correlation": 0.9828094790612636, "3000–6000 Hz_10ms_modulation_mae_db": 0.5010708153755997, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8173164873768196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09111357933187764, "3000–6000 Hz_40ms_envelope_correlation": 0.9928670694707915, "3000–6000 Hz_40ms_modulation_mae_db": 0.5997546767571359, "mr_stft_db": 5.721692809789004, "band_mae_db": 0.4837080452938014, "rms_gain_error_db": -0.07834280400071986, "output_peak": 0.5092254960109392}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9641030997011764, "全帯域_10ms_rms_envelope_nrmse": 0.08644616126227384, "全帯域_10ms_envelope_correlation": 0.9931102870356179, "全帯域_10ms_modulation_mae_db": 0.3859439607632826, "全帯域_40ms_log_rms_mae_db": 0.49412923553950294, "全帯域_40ms_rms_envelope_nrmse": 0.04619444136560395, "全帯域_40ms_envelope_correlation": 0.997878601395938, "全帯域_40ms_modulation_mae_db": 0.48736818343646665, "300–1000 Hz_10ms_log_rms_mae_db": 1.7667104121889623, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12849167623580965, "300–1000 Hz_10ms_envelope_correlation": 0.9877574113472228, "300–1000 Hz_10ms_modulation_mae_db": 0.5219363971670424, "300–1000 Hz_40ms_log_rms_mae_db": 0.9863688660340714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07520391153485394, "300–1000 Hz_40ms_envelope_correlation": 0.995565176322208, "300–1000 Hz_40ms_modulation_mae_db": 0.6547418659540563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3898741786344742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11070481940481366, "1000–3000 Hz_10ms_envelope_correlation": 0.9913091470306886, "1000–3000 Hz_10ms_modulation_mae_db": 0.4020355674806219, "1000–3000 Hz_40ms_log_rms_mae_db": 0.777450016000468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06435028291256384, "1000–3000 Hz_40ms_envelope_correlation": 0.9969360620604903, "1000–3000 Hz_40ms_modulation_mae_db": 0.322627107942599, "3000–6000 Hz_10ms_log_rms_mae_db": 1.389868000058758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13773576232235163, "3000–6000 Hz_10ms_envelope_correlation": 0.9850623865475945, "3000–6000 Hz_10ms_modulation_mae_db": 0.5488449462425529, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7494738422212484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08136568176856908, "3000–6000 Hz_40ms_envelope_correlation": 0.9942034983233047, "3000–6000 Hz_40ms_modulation_mae_db": 0.6573434172100204, "mr_stft_db": 5.728969655619946, "band_mae_db": 0.48202029396282864, "rms_gain_error_db": -0.08445762590706467, "output_peak": 0.5072941879406365}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 95360, "output_samples": 95360, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7235020236419585, "spectral_convergence": 0.4099676062859286, "spectral_convergence_status": "ok", "amplitude_floor": 1.2569032294341943e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.602743156631457, "spectral_convergence": 0.41512954410584124, "spectral_convergence_status": "ok", "amplitude_floor": 9.073914887165124e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.899665409306446, "spectral_convergence": 0.43459170518699247, "spectral_convergence_status": "ok", "amplitude_floor": 7.969434026985316e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.036863249031963e-06, "output_power": 8.032885626551741e-06, "delta_db": 2.027114235451781, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00014485054714414723, "output_power": 0.00014570595237060346, "delta_db": 0.025571538999013512, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004849180298355377, "output_power": 0.0004831830362943862, "delta_db": -0.01556653440698208, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006903353830986323, "output_power": 0.0006677978456532981, "delta_db": -0.14415120029844028, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00029452981369653373, "output_power": 0.0002892171819285779, "delta_db": -0.07905172556506303, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004100383435072048, "output_power": 0.00042920415827845786, "delta_db": 0.19839450543559578, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.41497495669281265, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.014076089895672633, "pitch": null, "output_rms_dbfs": -26.93973835711725, "output_peak": 0.4917664196682542, "reference_rms_dbfs": -26.925662267221572, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.74197019652662}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 95360, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.0001898595271537964, "log_rms_mae_db": 1.011434058378268, "rms_envelope_nrmse": 0.09788980292749953, "envelope_correlation": 0.9911146333207217, "rms_gain_error_db": -0.01407506215521975, "reference_mean_power": 0.002031411160407025, "output_mean_power": 0.0020248382111262317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 9.927152563814217e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006381525961065371, "output_power": 0.0006343634833865906, "reference_active": true, "delta_db": -0.025863651322429702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010674365906282643, "output_power": 0.00010981261453553166, "reference_active": true, "delta_db": 0.12310145921788104}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4567838102832594e-05, "output_power": 3.8496031613499076e-05, "reference_active": true, "delta_db": 0.4674374300272694}], "active_band_count": 3, "active_band_mae_db": 0.2054675135225267}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00017019756464998765, "log_rms_mae_db": 0.5204360020719034, "rms_envelope_nrmse": 0.05524951021669179, "envelope_correlation": 0.9968438713184906, "rms_gain_error_db": -0.014071715723020324, "reference_mean_power": 0.0020416854120747243, "output_mean_power": 0.002035080786961613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 8.390142156345837e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005509511464645055, "output_power": 0.000551701962625202, "reference_active": true, "delta_db": 0.005914378025529951}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5897699076585876e-05, "output_power": 4.920120935905631e-05, "reference_active": true, "delta_db": 0.30184863576458265}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848237969563024e-06, "output_power": 2.0194638580909005e-06, "reference_active": true, "delta_db": 0.38478197105074974}], "active_band_count": 3, "active_band_mae_db": 0.23084832828028745}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 9.246932648014587e-05, "log_rms_mae_db": 1.749284913717138, "rms_envelope_nrmse": 0.1218946366917546, "envelope_correlation": 0.9889992316340761, "rms_gain_error_db": -0.012868282613580145, "reference_mean_power": 0.0005236182881812637, "output_mean_power": 0.0005220690869612908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.7855833380403224e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028934531457073826, "output_power": 0.0002765482153069431, "reference_active": true, "delta_db": -0.19645594023198942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6197786234688146e-05, "output_power": 3.373760175718175e-05, "reference_active": true, "delta_db": -0.3056780353253694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0243815325354375e-05, "output_power": 1.032971228835604e-05, "reference_active": true, "delta_db": 0.03626484918530366}], "active_band_count": 3, "active_band_mae_db": 0.17946627491422085}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 8.334521268786246e-05, "log_rms_mae_db": 0.9434467371636925, "rms_envelope_nrmse": 0.06085558245139891, "envelope_correlation": 0.9970307765214856, "rms_gain_error_db": -0.012868140061920814, "reference_mean_power": 0.000526268698480611, "output_mean_power": 0.0005247116728574683}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.428386331422997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002786455084699618, "output_power": 0.00026764263691846123, "reference_active": true, "delta_db": -0.17496746923559786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6309295594140896e-05, "output_power": 1.4846313951780004e-05, "reference_active": true, "delta_db": -0.40816563825122665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.788822138965803e-07, "output_power": 6.799273982447109e-07, "reference_active": true, "delta_db": -0.5900924486826258}], "active_band_count": 3, "active_band_mae_db": 0.3910751853898167}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00017019739791537767, "log_rms_mae_db": 1.3861423680967546, "rms_envelope_nrmse": 0.12825488045800634, "envelope_correlation": 0.9883924029120633, "rms_gain_error_db": -0.18430519995555927, "reference_mean_power": 0.0008446325483556021, "output_mean_power": 0.0008095381000530384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 4.004634586327471e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028725854625925005, "output_power": 0.0002685904835894476, "reference_active": true, "delta_db": -0.29182337145367876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3436307686794414e-05, "output_power": 3.9620665955850575e-05, "reference_active": true, "delta_db": -0.39931129834912593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1340506745101254e-05, "output_power": 1.0938654068773629e-05, "reference_active": true, "delta_db": -0.15668573096862856}], "active_band_count": 3, "active_band_mae_db": 0.2826068002571444}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00015333717053704127, "log_rms_mae_db": 0.765148036142468, "rms_envelope_nrmse": 0.08341663910448698, "envelope_correlation": 0.9949526284218249, "rms_gain_error_db": -0.18430149023558573, "reference_mean_power": 0.0008489024890862695, "output_mean_power": 0.0008136313199178179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.579273675799131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002706567111490358, "output_power": 0.00025349142588582054, "reference_active": true, "delta_db": -0.2845552593204119}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.939557003916555e-05, "output_power": 1.7764694129064385e-05, "reference_active": true, "delta_db": -0.38144813868090366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.660068008029229e-07, "output_power": 7.809377084104038e-07, "reference_active": true, "delta_db": 0.0838376828116414}], "active_band_count": 3, "active_band_mae_db": 0.24994702693765233}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00010107776753147987, "log_rms_mae_db": 1.391446514051296, "rms_envelope_nrmse": 0.14689659856211607, "envelope_correlation": 0.9839033743291477, "rms_gain_error_db": 0.15377316211399378, "reference_mean_power": 0.00032790731816785513, "output_mean_power": 0.00033972571823557814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.172510420573334e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020494142716961824, "output_power": 0.00022517238210778993, "reference_active": true, "delta_db": 0.40885366046235894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6580284356723e-05, "output_power": 4.0704323088062647e-05, "reference_active": true, "delta_db": 0.4639345969145805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.565019545039446e-06, "output_power": 1.322582526308279e-05, "reference_active": true, "delta_db": 1.407369187993269}], "active_band_count": 3, "active_band_mae_db": 0.7600524817900695}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 7.865643818167715e-05, "log_rms_mae_db": 0.7894875714436845, "rms_envelope_nrmse": 0.08360023571769212, "envelope_correlation": 0.9946486509711691, "rms_gain_error_db": 0.15377743997505092, "reference_mean_power": 0.0003295653902210026, "output_mean_power": 0.00034144388666900263}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 2.853826866715288e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001952923309195638, "output_power": 0.00021832618974947129, "reference_active": true, "delta_db": 0.4842064649848391}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5170390000310775e-05, "output_power": 1.781469408975235e-05, "reference_active": true, "delta_db": 0.6978162336528598}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.363045655471935e-07, "output_power": 8.847124570358323e-07, "reference_active": true, "delta_db": 0.7974464893893978}], "active_band_count": 3, "active_band_mae_db": 0.6598230626756989}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8336021466349767, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a0344fdc9ec2223e210c8a4569f227141d58c43ebee52ace12ba7616758ee3ef", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9915389058441392, "standard_deviation": 0.0014404390468013906, "minimum": 0.9883924029120633, "maximum": 0.9928575334217796, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3808526587004062, "standard_deviation": 0.03836696637220933, "minimum": 1.3079804246888431, "maximum": 1.4235609250915753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.40295253661908437, "standard_deviation": 0.2070753480319947, "minimum": 0.23669189818551742, "maximum": 0.829705298054558, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10930239795928418, "standard_deviation": 0.008837311704771434, "minimum": 0.10083287933791502, "maximum": 0.12825488045800634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970108289917068, "standard_deviation": 0.0008779703968512133, "minimum": 0.9949526284218249, "maximum": 0.9977250032357609, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7646876791741243, "standard_deviation": 0.02966411227107303, "minimum": 0.7161176507111585, "maximum": 0.8166565827150121, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.424163496760745, "standard_deviation": 0.2881739692375956, "minimum": 0.1590790731081842, "maximum": 1.043839622660848, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06402187585692018, "standard_deviation": 0.008612008606328804, "minimum": 0.055892301896091436, "maximum": 0.08341663910448698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9841685692832701, "standard_deviation": 0.001857612654637266, "minimum": 0.9809126343076001, "maximum": 0.9869701335929896, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4059965871487896, "standard_deviation": 0.020032812247207377, "minimum": 1.3798645158175782, "maximum": 1.4374712168937407, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5128639530068693, "standard_deviation": 0.22795930409237306, "minimum": 0.2511593271505431, "maximum": 0.8588086960156686, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1427373604591655, "standard_deviation": 0.007994963282990722, "minimum": 0.1305890093349201, "maximum": 0.15604144095207786, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994420779725586, "standard_deviation": 0.0011094391110647573, "minimum": 0.9927892124697082, "maximum": 0.9960336825812116, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.77912785828069, "standard_deviation": 0.02549793408567053, "minimum": 0.7482747888651613, "maximum": 0.8173164873768196, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5062271626006345, "standard_deviation": 0.21773016545603818, "minimum": 0.1852963726844368, "maximum": 0.8001884139050043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08139497959551398, "standard_deviation": 0.007415459194768276, "minimum": 0.06987107175597437, "maximum": 0.09140118097020648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9884558456729067, "standard_deviation": 0.002081517774725735, "minimum": 0.9847363600026638, "maximum": 0.9921270593179428, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.731350703291458, "standard_deviation": 0.03166570054231189, "minimum": 1.697428875112361, "maximum": 1.7776564393331875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4535358824794451, "standard_deviation": 0.20143302411765152, "minimum": 0.17946627491422085, "maximum": 0.8461616690692851, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12479579342524379, "standard_deviation": 0.011300718003294657, "minimum": 0.10361676583818, "maximum": 0.14385526968973128, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99603987496715, "standard_deviation": 0.0011704060738671619, "minimum": 0.9940593333489843, "maximum": 0.9974836147622421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9893362594269703, "standard_deviation": 0.02444946310755288, "minimum": 0.9434467371636925, "maximum": 1.018819028345341, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4807306774127488, "standard_deviation": 0.28467398383214804, "minimum": 0.11394934400151403, "maximum": 0.8960242512416707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07060129677100643, "standard_deviation": 0.010222703197146795, "minimum": 0.05782880590613671, "maximum": 0.08655822143398949, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.45295973881686036, "standard_deviation": 0.06005635144532919, "minimum": 0.3226125565066265, "maximum": 0.4990169865010117, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.724034178958271, "standard_deviation": 0.010627391166990088, "minimum": 5.710488993028034, "maximum": 5.74197019652662, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5055713899343866, "standard_deviation": 0.015700261676683525, "minimum": 0.48995765846763856, "maximum": 0.5352827434081906, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.059648126864642194, "standard_deviation": 0.04771854103756246, "minimum": -0.1217781458123278, "maximum": 0.011801534609115571, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9926063582105729, "standard_deviation": 0.0008147834684983957, "minimum": 0.9911146333207217, "maximum": 0.9937921221335547, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9830166006735173, "standard_deviation": 0.023742667639197215, "minimum": 0.945361713144853, "maximum": 1.011434058378268, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.39076732418169624, "standard_deviation": 0.2630966010808518, "minimum": 0.04263816078673677, "maximum": 0.7408280574855265, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08938291328733293, "standard_deviation": 0.004783048696931076, "minimum": 0.08201129900486283, "maximum": 0.09788980292749953, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997538842478674, "standard_deviation": 0.0003847681272787223, "minimum": 0.9968438713184906, "maximum": 0.9978797478181816, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5093402820016566, "standard_deviation": 0.01458966624569894, "minimum": 0.49412923553950294, "maximum": 0.5386705622022572, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.41119644145914297, "standard_deviation": 0.23868202725791166, "minimum": 0.08103691153940921, "maximum": 0.7592192609488402, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04931226479438314, "standard_deviation": 0.0035358246336918333, "minimum": 0.04529698049501676, "maximum": 0.05524951021669179, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0016", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2352099418640137, "predicted_mos": 2.2352099418640137, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "9b923c8abb11d688a505a53498bb6019289871cbb429467174eb0063d7304681", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.1080165999010205, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9694074063881527, "全帯域_10ms_rms_envelope_nrmse": 0.08653623344103412, "全帯域_10ms_envelope_correlation": 0.9931159994425264, "全帯域_10ms_modulation_mae_db": 0.31747997029264285, "全帯域_40ms_log_rms_mae_db": 0.4919303455534433, "全帯域_40ms_rms_envelope_nrmse": 0.04758852482751619, "全帯域_40ms_envelope_correlation": 0.9977750123489627, "全帯域_40ms_modulation_mae_db": 0.25681361349114956, "300–1000 Hz_10ms_log_rms_mae_db": 1.8730578840695575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11749702373492972, "300–1000 Hz_10ms_envelope_correlation": 0.9898929435378685, "300–1000 Hz_10ms_modulation_mae_db": 0.47171580868560176, "300–1000 Hz_40ms_log_rms_mae_db": 1.136868156945053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07467244545397095, "300–1000 Hz_40ms_envelope_correlation": 0.995784468701514, "300–1000 Hz_40ms_modulation_mae_db": 0.5619785108264012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4861424634164135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10895195129923367, "1000–3000 Hz_10ms_envelope_correlation": 0.9915871110572064, "1000–3000 Hz_10ms_modulation_mae_db": 0.42124711694591, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8351731608495151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06281967604189279, "1000–3000 Hz_40ms_envelope_correlation": 0.9970977670218499, "1000–3000 Hz_40ms_modulation_mae_db": 0.39029908731433455, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5817693998791738, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15037388196766846, "3000–6000 Hz_10ms_envelope_correlation": 0.9829004007820794, "3000–6000 Hz_10ms_modulation_mae_db": 1.1521404044361623, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8634158388860466, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0900248856887964, "3000–6000 Hz_40ms_envelope_correlation": 0.9940901956503023, "3000–6000 Hz_40ms_modulation_mae_db": 1.3890025902234744, "mr_stft_db": 5.835610264838368, "band_mae_db": 0.4551976869191228, "rms_gain_error_db": -0.10178928490972834, "output_peak": 0.49873233179501486}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9262587527162888, "全帯域_10ms_rms_envelope_nrmse": 0.08210034184447533, "全帯域_10ms_envelope_correlation": 0.9937550449204743, "全帯域_10ms_modulation_mae_db": 0.4041007732967456, "全帯域_40ms_log_rms_mae_db": 0.4651857434761376, "全帯域_40ms_rms_envelope_nrmse": 0.04540782691354565, "全帯域_40ms_envelope_correlation": 0.9978927386693918, "全帯域_40ms_modulation_mae_db": 0.2847524372597654, "300–1000 Hz_10ms_log_rms_mae_db": 1.9051550051060606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1244742751525971, "300–1000 Hz_10ms_envelope_correlation": 0.988672057493952, "300–1000 Hz_10ms_modulation_mae_db": 0.7500372523842463, "300–1000 Hz_40ms_log_rms_mae_db": 1.1504012184731118, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08105999553069725, "300–1000 Hz_40ms_envelope_correlation": 0.9950426792609407, "300–1000 Hz_40ms_modulation_mae_db": 0.9702415612992424, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5024593280285543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10453212741200671, "1000–3000 Hz_10ms_envelope_correlation": 0.9922419224807211, "1000–3000 Hz_10ms_modulation_mae_db": 0.21660189237989091, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8040393510647041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05276931630861809, "1000–3000 Hz_40ms_envelope_correlation": 0.9979260633179314, "1000–3000 Hz_40ms_modulation_mae_db": 0.3745800818841947, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5701222867821607, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13987009737912495, "3000–6000 Hz_10ms_envelope_correlation": 0.984580081927989, "3000–6000 Hz_10ms_modulation_mae_db": 0.2630066513139757, "3000–6000 Hz_40ms_log_rms_mae_db": 0.830652544224249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08050626829413034, "3000–6000 Hz_40ms_envelope_correlation": 0.9943800847951779, "3000–6000 Hz_40ms_modulation_mae_db": 0.17106021677987723, "mr_stft_db": 5.840608810586027, "band_mae_db": 0.4133506649756209, "rms_gain_error_db": -0.04487097969661381, "output_peak": 0.5256766736461678}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9390627097510617, "全帯域_10ms_rms_envelope_nrmse": 0.08803314354815922, "全帯域_10ms_envelope_correlation": 0.9928280830295657, "全帯域_10ms_modulation_mae_db": 0.43887244779708867, "全帯域_40ms_log_rms_mae_db": 0.4687401220885732, "全帯域_40ms_rms_envelope_nrmse": 0.04177871043289413, "全帯域_40ms_envelope_correlation": 0.998239667910912, "全帯域_40ms_modulation_mae_db": 0.36536487311447347, "300–1000 Hz_10ms_log_rms_mae_db": 1.8856303535065433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12395690312150813, "300–1000 Hz_10ms_envelope_correlation": 0.9886769058795402, "300–1000 Hz_10ms_modulation_mae_db": 0.214757551584252, "300–1000 Hz_40ms_log_rms_mae_db": 1.0454317348125224, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06798206655491086, "300–1000 Hz_40ms_envelope_correlation": 0.9965144765679714, "300–1000 Hz_40ms_modulation_mae_db": 0.6002847052224615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4720468494960646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11129657712499609, "1000–3000 Hz_10ms_envelope_correlation": 0.9911836527865998, "1000–3000 Hz_10ms_modulation_mae_db": 0.6621938236562882, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7774118545807859, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057000735425113955, "1000–3000 Hz_40ms_envelope_correlation": 0.9975606807768888, "1000–3000 Hz_40ms_modulation_mae_db": 0.8098345839684719, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5001094293188786, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13961217171154328, "3000–6000 Hz_10ms_envelope_correlation": 0.984662728796987, "3000–6000 Hz_10ms_modulation_mae_db": 0.8016807146836564, "3000–6000 Hz_40ms_log_rms_mae_db": 0.802292792222597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07381824518793823, "3000–6000 Hz_40ms_envelope_correlation": 0.9952306142279036, "3000–6000 Hz_40ms_modulation_mae_db": 0.9166372558766892, "mr_stft_db": 5.849046715109661, "band_mae_db": 0.40574434550330674, "rms_gain_error_db": -0.05808658550140544, "output_peak": 0.5078104573419533}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9545520192203285, "全帯域_10ms_rms_envelope_nrmse": 0.07924872072293891, "全帯域_10ms_envelope_correlation": 0.9943033016262368, "全帯域_10ms_modulation_mae_db": 0.5209422804622018, "全帯域_40ms_log_rms_mae_db": 0.49379850194397357, "全帯域_40ms_rms_envelope_nrmse": 0.043138107161376275, "全帯域_40ms_envelope_correlation": 0.9982617881559152, "全帯域_40ms_modulation_mae_db": 0.3812207641652667, "300–1000 Hz_10ms_log_rms_mae_db": 1.9719241790712994, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10545195279173993, "300–1000 Hz_10ms_envelope_correlation": 0.991876043199218, "300–1000 Hz_10ms_modulation_mae_db": 0.46846798113410265, "300–1000 Hz_40ms_log_rms_mae_db": 1.1272179198448822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0647090630933266, "300–1000 Hz_40ms_envelope_correlation": 0.9968111041340503, "300–1000 Hz_40ms_modulation_mae_db": 0.4773286096589076, "1000–3000 Hz_10ms_log_rms_mae_db": 1.512656716361892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1010812438972723, "1000–3000 Hz_10ms_envelope_correlation": 0.992867726390272, "1000–3000 Hz_10ms_modulation_mae_db": 0.5634745517216834, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8177900114502215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05715756780931011, "1000–3000 Hz_40ms_envelope_correlation": 0.9977473028714016, "1000–3000 Hz_40ms_modulation_mae_db": 0.3914069476586826, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6109347732622317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14184666652011466, "3000–6000 Hz_10ms_envelope_correlation": 0.9841577521741031, "3000–6000 Hz_10ms_modulation_mae_db": 0.22799202628397922, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9107597705598759, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0755904775330747, "3000–6000 Hz_40ms_envelope_correlation": 0.9950851167278688, "3000–6000 Hz_40ms_modulation_mae_db": 0.3600570755586359, "mr_stft_db": 5.847154935305665, "band_mae_db": 0.44563331906673026, "rms_gain_error_db": -0.12334157830010574, "output_peak": 0.5041356661516057}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9774972721238528, "全帯域_10ms_rms_envelope_nrmse": 0.08346145240715507, "全帯域_10ms_envelope_correlation": 0.9935611038485151, "全帯域_10ms_modulation_mae_db": 0.25200942260713205, "全帯域_40ms_log_rms_mae_db": 0.48720034317927613, "全帯域_40ms_rms_envelope_nrmse": 0.043272879684545774, "全帯域_40ms_envelope_correlation": 0.9981119109281222, "全帯域_40ms_modulation_mae_db": 0.5622208909368763, "300–1000 Hz_10ms_log_rms_mae_db": 1.9433887048796092, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12467965503418409, "300–1000 Hz_10ms_envelope_correlation": 0.9884608052323999, "300–1000 Hz_10ms_modulation_mae_db": 0.19303771280849258, "300–1000 Hz_40ms_log_rms_mae_db": 1.111494669124492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08646794046578347, "300–1000 Hz_40ms_envelope_correlation": 0.9940033780376349, "300–1000 Hz_40ms_modulation_mae_db": 0.44602403543119706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5608973335347138, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11671759700380194, "1000–3000 Hz_10ms_envelope_correlation": 0.990369794149363, "1000–3000 Hz_10ms_modulation_mae_db": 0.329234992884143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8720875169107862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0691690008303486, "1000–3000 Hz_40ms_envelope_correlation": 0.9965265392881937, "1000–3000 Hz_40ms_modulation_mae_db": 0.4730096412156777, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5752575127330648, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1445171613859099, "3000–6000 Hz_10ms_envelope_correlation": 0.9836740104770508, "3000–6000 Hz_10ms_modulation_mae_db": 0.1444379527325124, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8414914707569414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07710758228435746, "3000–6000 Hz_40ms_envelope_correlation": 0.9948741366138553, "3000–6000 Hz_40ms_modulation_mae_db": 0.40920963061592913, "mr_stft_db": 5.838301123208324, "band_mae_db": 0.48363386826628213, "rms_gain_error_db": -0.06547983719097158, "output_peak": 0.5049870970138969}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9430920116465163, "全帯域_10ms_rms_envelope_nrmse": 0.08650419777170014, "全帯域_10ms_envelope_correlation": 0.9930645820178615, "全帯域_10ms_modulation_mae_db": 0.3747851796111312, "全帯域_40ms_log_rms_mae_db": 0.4770028469869225, "全帯域_40ms_rms_envelope_nrmse": 0.04474157240269607, "全帯域_40ms_envelope_correlation": 0.9979514397704599, "全帯域_40ms_modulation_mae_db": 0.3515660834359311, "300–1000 Hz_10ms_log_rms_mae_db": 1.8788531920493183, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12114184544476748, "300–1000 Hz_10ms_envelope_correlation": 0.9892082046676447, "300–1000 Hz_10ms_modulation_mae_db": 0.3886205285998286, "300–1000 Hz_40ms_log_rms_mae_db": 1.12797926253152, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07244477283644156, "300–1000 Hz_40ms_envelope_correlation": 0.9960153201183273, "300–1000 Hz_40ms_modulation_mae_db": 0.4464549156011586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5159936837114631, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11312326292607484, "1000–3000 Hz_10ms_envelope_correlation": 0.9909021344710837, "1000–3000 Hz_10ms_modulation_mae_db": 0.16982634087130277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8651464618467904, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06038776644752991, "1000–3000 Hz_40ms_envelope_correlation": 0.997276433596014, "1000–3000 Hz_40ms_modulation_mae_db": 0.321097169878699, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5498475231258015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13615788639737422, "3000–6000 Hz_10ms_envelope_correlation": 0.9854748815828428, "3000–6000 Hz_10ms_modulation_mae_db": 0.3791801624198457, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8530061755795595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0678612321523415, "3000–6000 Hz_40ms_envelope_correlation": 0.9959815187302667, "3000–6000 Hz_40ms_modulation_mae_db": 0.44200622513291704, "mr_stft_db": 5.8425369454142535, "band_mae_db": 0.4021344536316415, "rms_gain_error_db": -0.046971988854506506, "output_peak": 0.5200231236620666}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9402476099255849, "全帯域_10ms_rms_envelope_nrmse": 0.08424685582975829, "全帯域_10ms_envelope_correlation": 0.993497464284324, "全帯域_10ms_modulation_mae_db": 0.45865390132236183, "全帯域_40ms_log_rms_mae_db": 0.44643242060513616, "全帯域_40ms_rms_envelope_nrmse": 0.04192949606824975, "全帯域_40ms_envelope_correlation": 0.9983188803104052, "全帯域_40ms_modulation_mae_db": 0.5686171533268122, "300–1000 Hz_10ms_log_rms_mae_db": 1.9265350144493154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12908142033848974, "300–1000 Hz_10ms_envelope_correlation": 0.9876290277787563, "300–1000 Hz_10ms_modulation_mae_db": 0.13758081935545788, "300–1000 Hz_40ms_log_rms_mae_db": 1.0887855456294582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08273836290126624, "300–1000 Hz_40ms_envelope_correlation": 0.9945571791976962, "300–1000 Hz_40ms_modulation_mae_db": 0.2924751104419469, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5017259703179022, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10784725673048988, "1000–3000 Hz_10ms_envelope_correlation": 0.9917912135770584, "1000–3000 Hz_10ms_modulation_mae_db": 0.11933390987447201, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8547001791525725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0558366282391066, "1000–3000 Hz_40ms_envelope_correlation": 0.9977859220689478, "1000–3000 Hz_40ms_modulation_mae_db": 0.16927989004416064, "3000–6000 Hz_10ms_log_rms_mae_db": 1.519359683093006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1368833227036307, "3000–6000 Hz_10ms_envelope_correlation": 0.9856029635984821, "3000–6000 Hz_10ms_modulation_mae_db": 0.7900684121494121, "3000–6000 Hz_40ms_log_rms_mae_db": 0.77255121366049, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06587137832094175, "3000–6000 Hz_40ms_envelope_correlation": 0.9969311133591616, "3000–6000 Hz_40ms_modulation_mae_db": 0.8837989776271055, "mr_stft_db": 5.845582356819375, "band_mae_db": 0.5015280558076941, "rms_gain_error_db": -0.10620568146109943, "output_peak": 0.5238817792290245}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.923940940258681, "全帯域_10ms_rms_envelope_nrmse": 0.08191892731804515, "全帯域_10ms_envelope_correlation": 0.9939596005115996, "全帯域_10ms_modulation_mae_db": 0.5701155430309609, "全帯域_40ms_log_rms_mae_db": 0.4281851296129837, "全帯域_40ms_rms_envelope_nrmse": 0.04372134119767964, "全帯域_40ms_envelope_correlation": 0.9982914220940142, "全帯域_40ms_modulation_mae_db": 0.7540198379042757, "300–1000 Hz_10ms_log_rms_mae_db": 1.924041102198518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10986049763784361, "300–1000 Hz_10ms_envelope_correlation": 0.9911245880354631, "300–1000 Hz_10ms_modulation_mae_db": 0.688033123149325, "300–1000 Hz_40ms_log_rms_mae_db": 1.112850061907592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06292287416209526, "300–1000 Hz_40ms_envelope_correlation": 0.996983188024506, "300–1000 Hz_40ms_modulation_mae_db": 1.0382221414083477, "1000–3000 Hz_10ms_log_rms_mae_db": 1.494865356809758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11172116019934357, "1000–3000 Hz_10ms_envelope_correlation": 0.991418183164486, "1000–3000 Hz_10ms_modulation_mae_db": 0.43149932245398664, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8419434807583459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06889180840148437, "1000–3000 Hz_40ms_envelope_correlation": 0.9968555805619191, "1000–3000 Hz_40ms_modulation_mae_db": 0.4704085214048845, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5836811402491597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14092765926056644, "3000–6000 Hz_10ms_envelope_correlation": 0.9843483011382981, "3000–6000 Hz_10ms_modulation_mae_db": 0.06013523394793041, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8984853176218247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06990679782530126, "3000–6000 Hz_40ms_envelope_correlation": 0.995784908002167, "3000–6000 Hz_40ms_modulation_mae_db": 0.20453846498051162, "mr_stft_db": 5.835689284623903, "band_mae_db": 0.5403786798236466, "rms_gain_error_db": -0.1461348154151874, "output_peak": 0.5053327148795377}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 95360, "output_samples": 95360, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.89120181784666, "spectral_convergence": 0.405669778695054, "spectral_convergence_status": "ok", "amplitude_floor": 1.2569032294341943e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.737513115992016, "spectral_convergence": 0.4068711650959494, "spectral_convergence_status": "ok", "amplitude_floor": 9.073914887165124e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.878115860676427, "spectral_convergence": 0.4260668586963454, "spectral_convergence_status": "ok", "amplitude_floor": 7.969434026985316e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.036863249031963e-06, "output_power": 7.890930401129225e-06, "delta_db": 1.9496805296276067, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00014485054714414723, "output_power": 0.00015140613181467146, "delta_db": 0.1922332405937778, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004849180298355377, "output_power": 0.00046238756612864345, "delta_db": -0.20662184869313965, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006903353830986323, "output_power": 0.0006784130915850874, "delta_db": -0.07565913627712316, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00029452981369653373, "output_power": 0.00027839425494927, "delta_db": -0.24468993920208143, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004100383435072048, "output_power": 0.0004041981557888425, "delta_db": -0.062301427121008134, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4551976869191228, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10178928490972834, "pitch": null, "output_rms_dbfs": -27.0274515521313, "output_peak": 0.49873233179501486, "reference_rms_dbfs": -26.925662267221572, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.835610264838368}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 95360, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.0001898595271537964, "log_rms_mae_db": 0.9694074063881527, "rms_envelope_nrmse": 0.08653623344103412, "envelope_correlation": 0.9931159994425264, "rms_gain_error_db": -0.1017882709322629, "reference_mean_power": 0.002031411160407025, "output_mean_power": 0.0019843533423096068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 9.927152563814217e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006381525961065371, "output_power": 0.0006098111397354531, "reference_active": true, "delta_db": -0.19729186772087184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010674365906282643, "output_power": 9.260133808259881e-05, "reference_active": true, "delta_db": -0.617248236398727}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4567838102832594e-05, "output_power": 3.568306901885369e-05, "reference_active": true, "delta_db": 0.13789980675832983}], "active_band_count": 3, "active_band_mae_db": 0.31747997029264285}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00017019756464998765, "log_rms_mae_db": 0.4919303455534433, "rms_envelope_nrmse": 0.04758852482751619, "envelope_correlation": 0.9977750123489627, "rms_gain_error_db": -0.10178784877914493, "reference_mean_power": 0.0020416854120747243, "output_mean_power": 0.001994389783897638}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 8.390142156345837e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005509511464645055, "output_power": 0.0005317672241235223, "reference_active": true, "delta_db": -0.15391525367207057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5897699076585876e-05, "output_power": 4.130557378391468e-05, "reference_active": true, "delta_db": -0.45782254810524087}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848237969563024e-06, "output_power": 1.917026833176145e-06, "reference_active": true, "delta_db": 0.1587030386961373}], "active_band_count": 3, "active_band_mae_db": 0.25681361349114956}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 9.246932648014587e-05, "log_rms_mae_db": 1.8730578840695575, "rms_envelope_nrmse": 0.11749702373492972, "envelope_correlation": 0.9898929435378685, "rms_gain_error_db": -0.15855948541796797, "reference_mean_power": 0.0005236182881812637, "output_mean_power": 0.0005048459284009342}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.7855833380403224e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028934531457073826, "output_power": 0.0002702698264348386, "reference_active": true, "delta_db": -0.29618891397140146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6197786234688146e-05, "output_power": 3.0029235032575616e-05, "reference_active": true, "delta_db": -0.8113774190679433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0243815325354375e-05, "output_power": 9.54341107531006e-06, "reference_active": true, "delta_db": -0.30758109301746045}], "active_band_count": 3, "active_band_mae_db": 0.47171580868560176}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 8.334521268786246e-05, "log_rms_mae_db": 1.136868156945053, "rms_envelope_nrmse": 0.07467244545397095, "envelope_correlation": 0.995784468701514, "rms_gain_error_db": -0.15856096783041346, "reference_mean_power": 0.000526268698480611, "output_mean_power": 0.0005074011450351844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.428386331422997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002786455084699618, "output_power": 0.0002596890880687592, "reference_active": true, "delta_db": -0.3059834568860618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6309295594140896e-05, "output_power": 1.3424550105021425e-05, "reference_active": true, "delta_db": -0.8453546390187938}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.788822138965803e-07, "output_power": 6.88671392956119e-07, "reference_active": true, "delta_db": -0.534597436574348}], "active_band_count": 3, "active_band_mae_db": 0.5619785108264012}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00017019739791537767, "log_rms_mae_db": 1.4861424634164135, "rms_envelope_nrmse": 0.10895195129923367, "envelope_correlation": 0.9915871110572064, "rms_gain_error_db": -0.11201118672682893, "reference_mean_power": 0.0008446325483556021, "output_mean_power": 0.0008231267099497862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 4.004634586327471e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028725854625925005, "output_power": 0.00026866133850658085, "reference_active": true, "delta_db": -0.29067784161312665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3436307686794414e-05, "output_power": 3.947153941158165e-05, "reference_active": true, "delta_db": -0.415688364010005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1340506745101254e-05, "output_power": 1.2893474799230602e-05, "reference_active": true, "delta_db": 0.5573751452145982}], "active_band_count": 3, "active_band_mae_db": 0.42124711694591}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00015333717053704127, "log_rms_mae_db": 0.8351731608495151, "rms_envelope_nrmse": 0.06281967604189279, "envelope_correlation": 0.9970977670218499, "rms_gain_error_db": -0.11201167545387768, "reference_mean_power": 0.0008489024890862695, "output_mean_power": 0.0008272878373404488}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.579273675799131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002706567111490358, "output_power": 0.00025620453741064577, "reference_active": true, "delta_db": -0.23831983358672496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.939557003916555e-05, "output_power": 1.7932045110933266e-05, "reference_active": true, "delta_db": -0.34072725381254465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.660068008029229e-07, "output_power": 8.778446018867677e-07, "reference_active": true, "delta_db": 0.591850174543734}], "active_band_count": 3, "active_band_mae_db": 0.39029908731433455}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00010107776753147987, "log_rms_mae_db": 1.5817693998791738, "rms_envelope_nrmse": 0.15037388196766846, "envelope_correlation": 0.9829004007820794, "rms_gain_error_db": -0.33413581485048915, "reference_mean_power": 0.00032790731816785513, "output_mean_power": 0.0003036249999817587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.172510420573334e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020494142716961824, "output_power": 0.0001865257507502975, "reference_active": true, "delta_db": -0.40890959554246553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6580284356723e-05, "output_power": 2.506863797660582e-05, "reference_active": true, "delta_db": -1.6411633852762308}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.565019545039446e-06, "output_power": 6.919122800857706e-06, "reference_active": true, "delta_db": -1.4063482324897902}], "active_band_count": 3, "active_band_mae_db": 1.1521404044361623}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 7.865643818167715e-05, "log_rms_mae_db": 0.8634158388860466, "rms_envelope_nrmse": 0.0900248856887964, "envelope_correlation": 0.9940901956503023, "rms_gain_error_db": -0.33413141400678575, "reference_mean_power": 0.0003295653902210026, "output_mean_power": 0.0003051605970775822}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 2.853826866715288e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001952923309195638, "output_power": 0.00017417196505417756, "reference_active": true, "delta_db": -0.49706937295402515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5170390000310775e-05, "output_power": 1.0402253342484725e-05, "reference_active": true, "delta_db": -1.6386931914441671}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.363045655471935e-07, "output_power": 4.612463821742522e-07, "reference_active": true, "delta_db": -2.0312452062722306}], "active_band_count": 3, "active_band_mae_db": 1.3890025902234744}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.822990142061189, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9b99946171f18c4a05598443847299b851d7de9081cd19674873967ac4ffd6f1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9915452172595989, "standard_deviation": 0.0007781607743126107, "minimum": 0.990369794149363, "maximum": 0.992867726390272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5058484627095954, "standard_deviation": 0.026348329474517776, "minimum": 1.4720468494960646, "maximum": 1.5608973335347138, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3641764938484596, "standard_deviation": 0.19162556807764347, "minimum": 0.11933390987447201, "maximum": 0.6621938236562882, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10940889707415237, "standard_deviation": 0.004955569952145552, "minimum": 0.1010812438972723, "maximum": 0.11671759700380194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973470361878933, "standard_deviation": 0.0004950633870066214, "minimum": 0.9965265392881937, "maximum": 0.9979260633179314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8335365020767151, "standard_deviation": 0.03223715117837902, "minimum": 0.7774118545807859, "maximum": 0.8720875169107862, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4249894904211382, "standard_deviation": 0.1826502915382619, "minimum": 0.16927989004416064, "maximum": 0.8098345839684719, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060504062437925554, "standard_deviation": 0.006046186483551114, "minimum": 0.05276931630861809, "maximum": 0.0691690008303486, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.984425140059729, "standard_deviation": 0.0008893100825447094, "minimum": 0.9829004007820794, "maximum": 0.9856029635984821, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5613852185554347, "standard_deviation": 0.03643103983515882, "minimum": 1.5001094293188786, "maximum": 1.6109347732622317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.47733019474593424, "standard_deviation": 0.38931876105861796, "minimum": 0.06013523394793041, "maximum": 1.1521404044361623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14127360591574156, "standard_deviation": 0.004534929934160694, "minimum": 0.13615788639737422, "maximum": 0.15037388196766846, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9952947110133379, "standard_deviation": 0.0009182382146531925, "minimum": 0.9940901956503023, "maximum": 0.9969311133591616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8465818904389479, "standard_deviation": 0.046073142282512754, "minimum": 0.77255121366049, "maximum": 0.9107597705598759, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5970388045993925, "standard_deviation": 0.42469665101943865, "minimum": 0.17106021677987723, "maximum": 1.3890025902234744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0750858584108602, "standard_deviation": 0.007768331011479012, "minimum": 0.06587137832094175, "maximum": 0.0900248856887964, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9894425719781053, "standard_deviation": 0.0014363380237851176, "minimum": 0.9876290277787563, "maximum": 0.991876043199218, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9135731794162778, "standard_deviation": 0.034371967715663386, "minimum": 1.8730578840695575, "maximum": 1.9719241790712994, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4140313472126633, "standard_deviation": 0.22674297023633205, "minimum": 0.13758081935545788, "maximum": 0.7500372523842463, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11951794665700748, "standard_deviation": 0.008108306768658442, "minimum": 0.10545195279173993, "maximum": 0.12908142033848974, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99571397425533, "standard_deviation": 0.0010866811535345522, "minimum": 0.9940033780376349, "maximum": 0.996983188024506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.112628571158579, "standard_deviation": 0.03283739744829614, "minimum": 1.0454317348125224, "maximum": 1.1504012184731118, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6041261987362079, "standard_deviation": 0.26385371699800003, "minimum": 0.2924751104419469, "maximum": 1.0382221414083477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07412469012481153, "standard_deviation": 0.008698426330000565, "minimum": 0.06292287416209526, "maximum": 0.08646794046578347, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4559501342492556, "standard_deviation": 0.04972837894768211, "minimum": 0.4021344536316415, "maximum": 0.5403786798236466, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.841816304488197, "standard_deviation": 0.00514576291403945, "minimum": 5.835610264838368, "maximum": 5.849046715109661, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5113224804649084, "standard_deviation": 0.010267778063538827, "minimum": 0.49873233179501486, "maximum": 0.5256766736461678, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08661009391620228, "standard_deviation": 0.03794770953996224, "minimum": -0.1461348154151874, "maximum": -0.04487097969661381, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993510647460138, "standard_deviation": 0.0004946751993267166, "minimum": 0.9928280830295657, "maximum": 0.9943033016262368, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9467573402538083, "standard_deviation": 0.019177832026090837, "minimum": 0.923940940258681, "maximum": 0.9774972721238528, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41711993980253315, "standard_deviation": 0.1038792648224028, "minimum": 0.25200942260713205, "maximum": 0.5701155430309609, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08400623411040828, "standard_deviation": 0.0029236624098822744, "minimum": 0.07924872072293891, "maximum": 0.08803314354815922, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981053575235228, "standard_deviation": 0.00020729212001024798, "minimum": 0.9977750123489627, "maximum": 0.9983188803104052, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4698094316808058, "standard_deviation": 0.0230545577590235, "minimum": 0.4281851296129837, "maximum": 0.49379850194397357, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4405719567043188, "standard_deviation": 0.1709599686231081, "minimum": 0.25681361349114956, "maximum": 0.7540198379042757, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04394730733606293, "standard_deviation": 0.0019256736249514672, "minimum": 0.04177871043289413, "maximum": 0.04758852482751619, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0016", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.1464736461639404, "predicted_mos": 2.1464736461639404, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "8a17c57c18b790d6b348863068ef3a44da84f5f7c326275cdc455a1b23811fc0", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9076777999289334, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.967061824878659, "全帯域_10ms_rms_envelope_nrmse": 0.07984996451100092, "全帯域_10ms_envelope_correlation": 0.9942053925038299, "全帯域_10ms_modulation_mae_db": 0.43260792551954436, "全帯域_40ms_log_rms_mae_db": 0.4720989002344965, "全帯域_40ms_rms_envelope_nrmse": 0.04428786334405229, "全帯域_40ms_envelope_correlation": 0.9981465856133636, "全帯域_40ms_modulation_mae_db": 0.6396944379632772, "300–1000 Hz_10ms_log_rms_mae_db": 1.9380708937069113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12458235560123672, "300–1000 Hz_10ms_envelope_correlation": 0.9885019828251201, "300–1000 Hz_10ms_modulation_mae_db": 0.23910103556179352, "300–1000 Hz_40ms_log_rms_mae_db": 1.122817794004088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06407568643868794, "300–1000 Hz_40ms_envelope_correlation": 0.9967073391710801, "300–1000 Hz_40ms_modulation_mae_db": 0.5893665514251023, "1000–3000 Hz_10ms_log_rms_mae_db": 1.465698218258268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10809848349169866, "1000–3000 Hz_10ms_envelope_correlation": 0.9923744799167614, "1000–3000 Hz_10ms_modulation_mae_db": 0.7123747950837808, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7646864022615082, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06385491937281185, "1000–3000 Hz_40ms_envelope_correlation": 0.9978304600993649, "1000–3000 Hz_40ms_modulation_mae_db": 0.8090527215141181, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5233132538637035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12856044397313823, "3000–6000 Hz_10ms_envelope_correlation": 0.9871442736337251, "3000–6000 Hz_10ms_modulation_mae_db": 0.19410466774391613, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8074933640406476, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06725454402675403, "3000–6000 Hz_40ms_envelope_correlation": 0.996126283843809, "3000–6000 Hz_40ms_modulation_mae_db": 0.2828774477386628, "mr_stft_db": 5.8178739993833775, "band_mae_db": 0.49084925276974695, "rms_gain_error_db": -0.11357282232118227, "output_peak": 0.5722446223318489}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9883925140236194, "全帯域_10ms_rms_envelope_nrmse": 0.08273645295751059, "全帯域_10ms_envelope_correlation": 0.9936737927206107, "全帯域_10ms_modulation_mae_db": 0.36327188289656775, "全帯域_40ms_log_rms_mae_db": 0.4947482817542093, "全帯域_40ms_rms_envelope_nrmse": 0.042561046995970325, "全帯域_40ms_envelope_correlation": 0.9981818880421985, "全帯域_40ms_modulation_mae_db": 0.32491808173703124, "300–1000 Hz_10ms_log_rms_mae_db": 1.825427737050585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11646260414010104, "300–1000 Hz_10ms_envelope_correlation": 0.9899759566783654, "300–1000 Hz_10ms_modulation_mae_db": 0.4342071608007732, "300–1000 Hz_40ms_log_rms_mae_db": 1.1304288020055793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0676567552973962, "300–1000 Hz_40ms_envelope_correlation": 0.9964160904179274, "300–1000 Hz_40ms_modulation_mae_db": 0.6514237296828128, "1000–3000 Hz_10ms_log_rms_mae_db": 1.472018493230663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10157859113398544, "1000–3000 Hz_10ms_envelope_correlation": 0.992683811984976, "1000–3000 Hz_10ms_modulation_mae_db": 0.5802995427013605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7880382984387633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05385056997559569, "1000–3000 Hz_40ms_envelope_correlation": 0.9978676003238346, "1000–3000 Hz_40ms_modulation_mae_db": 0.5371276326850626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5012443576858776, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1386786851720857, "3000–6000 Hz_10ms_envelope_correlation": 0.9849394644806954, "3000–6000 Hz_10ms_modulation_mae_db": 0.49402972709871146, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7907944104858573, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07907460113261436, "3000–6000 Hz_40ms_envelope_correlation": 0.9945660993671761, "3000–6000 Hz_40ms_modulation_mae_db": 0.4615006787015765, "mr_stft_db": 5.818385386297794, "band_mae_db": 0.47612115684312367, "rms_gain_error_db": -0.07030037481609511, "output_peak": 0.5235080629251856}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9598006793169319, "全帯域_10ms_rms_envelope_nrmse": 0.07925670488911467, "全帯域_10ms_envelope_correlation": 0.994245212644458, "全帯域_10ms_modulation_mae_db": 0.5217105926349619, "全帯域_40ms_log_rms_mae_db": 0.49197269776142527, "全帯域_40ms_rms_envelope_nrmse": 0.041172162353309213, "全帯域_40ms_envelope_correlation": 0.9983665932289504, "全帯域_40ms_modulation_mae_db": 0.7009884088025587, "300–1000 Hz_10ms_log_rms_mae_db": 1.8793938459920783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10680148952224965, "300–1000 Hz_10ms_envelope_correlation": 0.9915848536345635, "300–1000 Hz_10ms_modulation_mae_db": 0.12722004776949633, "300–1000 Hz_40ms_log_rms_mae_db": 1.0750003915428585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06352359946184181, "300–1000 Hz_40ms_envelope_correlation": 0.9967780759202766, "300–1000 Hz_40ms_modulation_mae_db": 0.5611259674031256, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5340756504861746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11211849607217245, "1000–3000 Hz_10ms_envelope_correlation": 0.9911685470913457, "1000–3000 Hz_10ms_modulation_mae_db": 0.7404982356905679, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8777239905315617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06754926164025096, "1000–3000 Hz_40ms_envelope_correlation": 0.9967655789526351, "1000–3000 Hz_40ms_modulation_mae_db": 0.9422663494038254, "3000–6000 Hz_10ms_log_rms_mae_db": 1.510590194619296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.139773081131912, "3000–6000 Hz_10ms_envelope_correlation": 0.9846258784970225, "3000–6000 Hz_10ms_modulation_mae_db": 0.3154115383065813, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7911478083365859, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07432065739724088, "3000–6000 Hz_40ms_envelope_correlation": 0.9953283494436641, "3000–6000 Hz_40ms_modulation_mae_db": 0.5654860315492951, "mr_stft_db": 5.815267449909143, "band_mae_db": 0.4966328434995717, "rms_gain_error_db": -0.09999053545774361, "output_peak": 0.5087404962254962}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.971289704366185, "全帯域_10ms_rms_envelope_nrmse": 0.09096871605371588, "全帯域_10ms_envelope_correlation": 0.9923608011062002, "全帯域_10ms_modulation_mae_db": 0.4463079550481483, "全帯域_40ms_log_rms_mae_db": 0.49157786813114557, "全帯域_40ms_rms_envelope_nrmse": 0.04701585268304527, "全帯域_40ms_envelope_correlation": 0.9978115824733981, "全帯域_40ms_modulation_mae_db": 0.5745210191040183, "300–1000 Hz_10ms_log_rms_mae_db": 1.8928564087623467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12662481756629376, "300–1000 Hz_10ms_envelope_correlation": 0.9881127449878351, "300–1000 Hz_10ms_modulation_mae_db": 0.15635783812170273, "300–1000 Hz_40ms_log_rms_mae_db": 1.113633711438204, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07104479159734021, "300–1000 Hz_40ms_envelope_correlation": 0.9959944121883878, "300–1000 Hz_40ms_modulation_mae_db": 0.47116473160027145, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5008686785957692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10820480556692062, "1000–3000 Hz_10ms_envelope_correlation": 0.991709154490321, "1000–3000 Hz_10ms_modulation_mae_db": 0.8568290528531791, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8094964388530227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060603640803124026, "1000–3000 Hz_40ms_envelope_correlation": 0.9973188219372795, "1000–3000 Hz_40ms_modulation_mae_db": 1.1058283319016649, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5104985266307198, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13863955633800598, "3000–6000 Hz_10ms_envelope_correlation": 0.9849231895594374, "3000–6000 Hz_10ms_modulation_mae_db": 0.45848636910112933, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8290526938033923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07065246853662939, "3000–6000 Hz_40ms_envelope_correlation": 0.9958839294329321, "3000–6000 Hz_40ms_modulation_mae_db": 0.7799390255138002, "mr_stft_db": 5.809544920937, "band_mae_db": 0.46594194801352934, "rms_gain_error_db": -0.08992135287120992, "output_peak": 0.5030310014921837}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9588590503980184, "全帯域_10ms_rms_envelope_nrmse": 0.0873539868865234, "全帯域_10ms_envelope_correlation": 0.9929942922968922, "全帯域_10ms_modulation_mae_db": 0.11775192275240866, "全帯域_40ms_log_rms_mae_db": 0.4934144063450774, "全帯域_40ms_rms_envelope_nrmse": 0.044960735419923674, "全帯域_40ms_envelope_correlation": 0.9979484844272719, "全帯域_40ms_modulation_mae_db": 0.07849622544811417, "300–1000 Hz_10ms_log_rms_mae_db": 1.9261411347814341, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13049599116338712, "300–1000 Hz_10ms_envelope_correlation": 0.9874826110349021, "300–1000 Hz_10ms_modulation_mae_db": 0.05181687687040776, "300–1000 Hz_40ms_log_rms_mae_db": 1.092906013818129, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07108585914004566, "300–1000 Hz_40ms_envelope_correlation": 0.9960029259082881, "300–1000 Hz_40ms_modulation_mae_db": 0.26838449167492595, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5175141181478908, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10404911882226082, "1000–3000 Hz_10ms_envelope_correlation": 0.9923558184921166, "1000–3000 Hz_10ms_modulation_mae_db": 0.23162054284637698, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8569579995947203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05786664257766064, "1000–3000 Hz_40ms_envelope_correlation": 0.9974817036432883, "1000–3000 Hz_40ms_modulation_mae_db": 0.2622891097976809, "3000–6000 Hz_10ms_log_rms_mae_db": 1.513568657934155, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14127406549518834, "3000–6000 Hz_10ms_envelope_correlation": 0.9851932696697677, "3000–6000 Hz_10ms_modulation_mae_db": 0.4098509239425616, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8137999504592165, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08215989955253655, "3000–6000 Hz_40ms_envelope_correlation": 0.9948843177087243, "3000–6000 Hz_40ms_modulation_mae_db": 0.36792607334068, "mr_stft_db": 5.821417534005895, "band_mae_db": 0.4112500565406152, "rms_gain_error_db": 0.04175809875277871, "output_peak": 0.51542441344434}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9544091838536682, "全帯域_10ms_rms_envelope_nrmse": 0.08032626585066362, "全帯域_10ms_envelope_correlation": 0.9940315793043667, "全帯域_10ms_modulation_mae_db": 0.5374762403732868, "全帯域_40ms_log_rms_mae_db": 0.49574453553172354, "全帯域_40ms_rms_envelope_nrmse": 0.04506581933530882, "全帯域_40ms_envelope_correlation": 0.9979476393063516, "全帯域_40ms_modulation_mae_db": 0.4840088166572361, "300–1000 Hz_10ms_log_rms_mae_db": 1.8736881742153564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12188426839539063, "300–1000 Hz_10ms_envelope_correlation": 0.9891758047983026, "300–1000 Hz_10ms_modulation_mae_db": 0.2505928105682924, "300–1000 Hz_40ms_log_rms_mae_db": 1.113121393637123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06871303572127295, "300–1000 Hz_40ms_envelope_correlation": 0.9963214407929339, "300–1000 Hz_40ms_modulation_mae_db": 0.35672525653282167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5272227020588751, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12535691970382773, "1000–3000 Hz_10ms_envelope_correlation": 0.9888227519520301, "1000–3000 Hz_10ms_modulation_mae_db": 0.30470987418953005, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8621285645693847, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0736196558587071, "1000–3000 Hz_40ms_envelope_correlation": 0.9959458895945007, "1000–3000 Hz_40ms_modulation_mae_db": 0.1950476708102359, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5175324455116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13947675240535787, "3000–6000 Hz_10ms_envelope_correlation": 0.9848340733146241, "3000–6000 Hz_10ms_modulation_mae_db": 0.8630651917812875, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8070075344216442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07635496974884647, "3000–6000 Hz_40ms_envelope_correlation": 0.9952580150678563, "3000–6000 Hz_40ms_modulation_mae_db": 0.9753756455812578, "mr_stft_db": 5.829365782274931, "band_mae_db": 0.46931781916494897, "rms_gain_error_db": -0.05940636215782725, "output_peak": 0.4883467344390898}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9550748092517962, "全帯域_10ms_rms_envelope_nrmse": 0.08004496294434281, "全帯域_10ms_envelope_correlation": 0.9940910002690636, "全帯域_10ms_modulation_mae_db": 0.4416952864873118, "全帯域_40ms_log_rms_mae_db": 0.433584146967034, "全帯域_40ms_rms_envelope_nrmse": 0.03831524182076991, "全帯域_40ms_envelope_correlation": 0.9985523732890629, "全帯域_40ms_modulation_mae_db": 0.4036280423231278, "300–1000 Hz_10ms_log_rms_mae_db": 1.895152787906994, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11592361557334867, "300–1000 Hz_10ms_envelope_correlation": 0.9901576956105946, "300–1000 Hz_10ms_modulation_mae_db": 0.4316913856817913, "300–1000 Hz_40ms_log_rms_mae_db": 1.1086774984459369, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07621012472525841, "300–1000 Hz_40ms_envelope_correlation": 0.9955698630282548, "300–1000 Hz_40ms_modulation_mae_db": 0.8491806086205779, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4951305093326415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11046204907797684, "1000–3000 Hz_10ms_envelope_correlation": 0.991323540028571, "1000–3000 Hz_10ms_modulation_mae_db": 0.5353774896112985, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7989179535302642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05778074959226695, "1000–3000 Hz_40ms_envelope_correlation": 0.9974979094197287, "1000–3000 Hz_40ms_modulation_mae_db": 0.4632390415891838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5458864976660422, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13591601217531865, "3000–6000 Hz_10ms_envelope_correlation": 0.9854661622803977, "3000–6000 Hz_10ms_modulation_mae_db": 0.18397208451817693, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8485176286977533, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.071292152527244, "3000–6000 Hz_40ms_envelope_correlation": 0.9955666249418474, "3000–6000 Hz_40ms_modulation_mae_db": 0.21040751496958385, "mr_stft_db": 5.80997268409675, "band_mae_db": 0.489226931123614, "rms_gain_error_db": -0.07729259384890891, "output_peak": 0.5476769593198367}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.90439555663659, "全帯域_10ms_rms_envelope_nrmse": 0.08007816359651748, "全帯域_10ms_envelope_correlation": 0.9940592303021221, "全帯域_10ms_modulation_mae_db": 0.3450122248385701, "全帯域_40ms_log_rms_mae_db": 0.44382609687210717, "全帯域_40ms_rms_envelope_nrmse": 0.04416058798064128, "全帯域_40ms_envelope_correlation": 0.9979856227932237, "全帯域_40ms_modulation_mae_db": 0.35371793184677913, "300–1000 Hz_10ms_log_rms_mae_db": 1.8674082121535793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12797220664650372, "300–1000 Hz_10ms_envelope_correlation": 0.9879710839674815, "300–1000 Hz_10ms_modulation_mae_db": 0.10490146571181601, "300–1000 Hz_40ms_log_rms_mae_db": 1.0789010049667496, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08088747888344147, "300–1000 Hz_40ms_envelope_correlation": 0.9948165033399825, "300–1000 Hz_40ms_modulation_mae_db": 0.34358070523236, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4833307223663914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10744437150796199, "1000–3000 Hz_10ms_envelope_correlation": 0.9918021208155063, "1000–3000 Hz_10ms_modulation_mae_db": 0.5948172137730141, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7842543830180798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06176156881535827, "1000–3000 Hz_40ms_envelope_correlation": 0.9971611470803664, "1000–3000 Hz_40ms_modulation_mae_db": 0.5206156825228314, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5119387868232075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13226284624451573, "3000–6000 Hz_10ms_envelope_correlation": 0.9862374654644572, "3000–6000 Hz_10ms_modulation_mae_db": 0.3474668050568546, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8106641667007082, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06669329198995548, "3000–6000 Hz_40ms_envelope_correlation": 0.9961159734384953, "3000–6000 Hz_40ms_modulation_mae_db": 0.542277563239901, "mr_stft_db": 5.814123239485842, "band_mae_db": 0.3905610405085986, "rms_gain_error_db": -0.015977985721056912, "output_peak": 0.5191292921495174}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 95360, "output_samples": 95360, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.894327733152248, "spectral_convergence": 0.4075041878392246, "spectral_convergence_status": "ok", "amplitude_floor": 1.2569032294341943e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.695708539881491, "spectral_convergence": 0.3960582200791238, "spectral_convergence_status": "ok", "amplitude_floor": 9.073914887165124e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.863585725116392, "spectral_convergence": 0.41007370247255726, "spectral_convergence_status": "ok", "amplitude_floor": 7.969434026985316e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.036863249031963e-06, "output_power": 8.544084382740605e-06, "delta_db": 2.2950536895594933, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00014485054714414723, "output_power": 0.00015047069492520095, "delta_db": 0.16531786693155343, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004849180298355377, "output_power": 0.0004793242820283753, "delta_db": -0.05038901588384263, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006903353830986323, "output_power": 0.0006445664345315768, "delta_db": -0.29792447594953, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00029452981369653373, "output_power": 0.0002878937585682671, "delta_db": -0.09897013049439601, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004100383435072048, "output_power": 0.00040651861435286657, "delta_db": -0.03744033779966558, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.49084925276974695, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11357282232118227, "pitch": null, "output_rms_dbfs": -27.039235089542757, "output_peak": 0.5722446223318489, "reference_rms_dbfs": -26.925662267221572, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8178739993833775}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 95360, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.0001898595271537964, "log_rms_mae_db": 0.967061824878659, "rms_envelope_nrmse": 0.07984996451100092, "envelope_correlation": 0.9942053925038299, "rms_gain_error_db": -0.11357111021155482, "reference_mean_power": 0.002031411160407025, "output_mean_power": 0.00197897689194457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 9.927152563814217e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006381525961065371, "output_power": 0.0006164571970050288, "reference_active": true, "delta_db": -0.15021613114028134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010674365906282643, "output_power": 9.725484458382456e-05, "reference_active": true, "delta_db": -0.40430841717049026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4567838102832594e-05, "output_power": 2.9130121529074883e-05, "reference_active": true, "delta_db": -0.7432992282478617}], "active_band_count": 3, "active_band_mae_db": 0.43260792551954436}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00017019756464998765, "log_rms_mae_db": 0.4720989002344965, "rms_envelope_nrmse": 0.04428786334405229, "envelope_correlation": 0.9981465856133636, "rms_gain_error_db": -0.11357012847520019, "reference_mean_power": 0.0020416854120747243, "output_mean_power": 0.001988986396856654}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 8.390142156345837e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005509511464645055, "output_power": 0.0005345833068707853, "reference_active": true, "delta_db": -0.13097697952430412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5897699076585876e-05, "output_power": 4.116344026819927e-05, "reference_active": true, "delta_db": -0.47279250097701125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848237969563024e-06, "output_power": 1.3652950154247276e-06, "reference_active": true, "delta_db": -1.3153138333885162}], "active_band_count": 3, "active_band_mae_db": 0.6396944379632772}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 9.246932648014587e-05, "log_rms_mae_db": 1.9380708937069113, "rms_envelope_nrmse": 0.12458235560123672, "envelope_correlation": 0.9885019828251201, "rms_gain_error_db": -0.014792039712546952, "reference_mean_power": 0.0005236182881812637, "output_mean_power": 0.0005218378816970678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.7855833380403224e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028934531457073826, "output_power": 0.00028320051559290604, "reference_active": true, "delta_db": -0.09322414443037313}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6197786234688146e-05, "output_power": 3.312583319946479e-05, "reference_active": true, "delta_db": -0.3851520029042945}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0243815325354375e-05, "output_power": 9.695473965206918e-06, "reference_active": true, "delta_db": -0.23892695935071284}], "active_band_count": 3, "active_band_mae_db": 0.23910103556179352}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 8.334521268786246e-05, "log_rms_mae_db": 1.122817794004088, "rms_envelope_nrmse": 0.06407568643868794, "envelope_correlation": 0.9967073391710801, "rms_gain_error_db": -0.014792968095552717, "reference_mean_power": 0.000526268698480611, "output_mean_power": 0.0005244791679565172}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.428386331422997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002786455084699618, "output_power": 0.0002744700651969691, "reference_active": true, "delta_db": -0.06557061512971823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6309295594140896e-05, "output_power": 1.435205806979396e-05, "reference_active": true, "delta_db": -0.5552102116966601}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.788822138965803e-07, "output_power": 5.980533176140829e-07, "reference_active": true, "delta_db": -1.1473188274489285}], "active_band_count": 3, "active_band_mae_db": 0.5893665514251023}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00017019739791537767, "log_rms_mae_db": 1.465698218258268, "rms_envelope_nrmse": 0.10809848349169866, "envelope_correlation": 0.9923744799167614, "rms_gain_error_db": -0.3171820161284683, "reference_mean_power": 0.0008446325483556021, "output_mean_power": 0.0007851445339793631}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 4.004634586327471e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028725854625925005, "output_power": 0.0002585806188145454, "reference_active": true, "delta_db": -0.4567698785468697}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3436307686794414e-05, "output_power": 3.75155591925279e-05, "reference_active": true, "delta_db": -0.6364147658735154}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1340506745101254e-05, "output_power": 8.917404913605794e-06, "reference_active": true, "delta_db": -1.0439397408309574}], "active_band_count": 3, "active_band_mae_db": 0.7123747950837808}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00015333717053704127, "log_rms_mae_db": 0.7646864022615082, "rms_envelope_nrmse": 0.06385491937281185, "envelope_correlation": 0.9978304600993649, "rms_gain_error_db": -0.31717972426051483, "reference_mean_power": 0.0008489024890862695, "output_mean_power": 0.0007891141564671326}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.579273675799131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002706567111490358, "output_power": 0.00024424450195251315, "reference_active": true, "delta_db": -0.4459400389109622}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.939557003916555e-05, "output_power": 1.718364906194555e-05, "reference_active": true, "delta_db": -0.5258715350867905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.660068008029229e-07, "output_power": 5.478957938080277e-07, "reference_active": true, "delta_db": -1.4553465905446017}], "active_band_count": 3, "active_band_mae_db": 0.8090527215141181}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00010107776753147987, "log_rms_mae_db": 1.5233132538637035, "rms_envelope_nrmse": 0.12856044397313823, "envelope_correlation": 0.9871442736337251, "rms_gain_error_db": 0.048101855651213894, "reference_mean_power": 0.00032790731816785513, "output_mean_power": 0.00033155936166364086}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.172510420573334e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020494142716961824, "output_power": 0.00020670563136721402, "reference_active": true, "delta_db": 0.037225522352969186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6580284356723e-05, "output_power": 3.788072080179578e-05, "reference_active": true, "delta_db": 0.1517115711028322}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.565019545039446e-06, "output_power": 8.736715039771978e-06, "reference_active": true, "delta_db": -0.39337690977594697}], "active_band_count": 3, "active_band_mae_db": 0.19410466774391613}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 7.865643818167715e-05, "log_rms_mae_db": 0.8074933640406476, "rms_envelope_nrmse": 0.06725454402675403, "envelope_correlation": 0.996126283843809, "rms_gain_error_db": 0.048097539045178876, "reference_mean_power": 0.0003295653902210026, "output_mean_power": 0.00033323556915685715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 2.853826866715288e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001952923309195638, "output_power": 0.00019893936914853824, "reference_active": true, "delta_db": 0.08035547439688705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5170390000310775e-05, "output_power": 1.622091369844192e-05, "reference_active": true, "delta_db": 0.29078567915560916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.363045655471935e-07, "output_power": 6.596421133718969e-07, "reference_active": true, "delta_db": -0.47749118966349213}], "active_band_count": 3, "active_band_mae_db": 0.2828774477386628}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8233381557138045, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8599ee13755ca94dbe1c8bf6aa3f500ddfcb10e81649d43782391ab22f85e069", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9915300280964536, "standard_deviation": 0.0012150847466001408, "minimum": 0.9888227519520301, "maximum": 0.992683811984976, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4994823865595843, "standard_deviation": 0.02526639416576698, "minimum": 1.465698218258268, "maximum": 1.5340756504861746, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5695658433436385, "standard_deviation": 0.2132770930886533, "minimum": 0.23162054284637698, "maximum": 0.8568290528531791, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10966410442210056, "standard_deviation": 0.00716476228483581, "minimum": 0.10157859113398544, "maximum": 0.12535691970382773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972336388813747, "standard_deviation": 0.0006301209926080268, "minimum": 0.9959458895945007, "maximum": 0.9978676003238346, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8177755038496631, "standard_deviation": 0.041998048899628077, "minimum": 0.7646864022615082, "maximum": 0.8777239905315617, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6044333175280754, "standard_deviation": 0.32150266821435936, "minimum": 0.1950476708102359, "maximum": 1.1058283319016649, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06211087607947193, "standard_deviation": 0.006238814912757559, "minimum": 0.05385056997559569, "maximum": 0.0736196558587071, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9854204721125159, "standard_deviation": 0.000857452973011937, "minimum": 0.9846258784970225, "maximum": 0.9871442736337251, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5168215900918252, "standard_deviation": 0.013327205216249943, "minimum": 1.5012443576858776, "maximum": 1.5458864976660422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.40829841344365236, "standard_deviation": 0.21554371141048942, "minimum": 0.18397208451817693, "maximum": 0.8630651917812875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1368226803669403, "standard_deviation": 0.004345576887739588, "minimum": 0.12856044397313823, "maximum": 0.14127406549518834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954661991555631, "standard_deviation": 0.0005674014882745044, "minimum": 0.9945660993671761, "maximum": 0.996126283843809, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8123096946182257, "standard_deviation": 0.0191173592090583, "minimum": 0.7907944104858573, "maximum": 0.8485176286977533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5232237475793446, "standard_deviation": 0.2552423681288544, "minimum": 0.21040751496958385, "maximum": 0.9753756455812578, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07347532311397764, "standard_deviation": 0.005518922221829332, "minimum": 0.06669329198995548, "maximum": 0.08215989955253655, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9891203416921457, "standard_deviation": 0.001377661022871339, "minimum": 0.9874826110349021, "maximum": 0.9915848536345635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8872673993211606, "standard_deviation": 0.035153740039427395, "minimum": 1.825427737050585, "maximum": 1.9380708937069113, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.22448607763575915, "standard_deviation": 0.14443456293576729, "minimum": 0.05181687687040776, "maximum": 0.4342071608007732, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12134341857606391, "standard_deviation": 0.007847211860416588, "minimum": 0.10680148952224965, "maximum": 0.13049599116338712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960758313458914, "standard_deviation": 0.0006460766273733197, "minimum": 0.9948165033399825, "maximum": 0.9967780759202766, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1044358262323335, "standard_deviation": 0.020152886967989534, "minimum": 1.0750003915428585, "maximum": 1.1304288020055793, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5113690052714998, "standard_deviation": 0.19086900436721574, "minimum": 0.26838449167492595, "maximum": 0.8491806086205779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07039966640816059, "standard_deviation": 0.005883196253330924, "minimum": 0.06352359946184181, "maximum": 0.08088747888344147, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.46123763105796856, "standard_deviation": 0.039126073801464656, "minimum": 0.3905610405085986, "maximum": 0.4966328434995717, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.816993874548841, "standard_deviation": 0.006451455001161435, "minimum": 5.809544920937, "maximum": 5.829365782274931, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5222626977909373, "standard_deviation": 0.026471573475001044, "minimum": 0.4883467344390898, "maximum": 0.5722446223318489, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06058799105515566, "standard_deviation": 0.05079362170260117, "minimum": -0.11357282232118227, "maximum": 0.04175809875277871, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9937076626434429, "standard_deviation": 0.0006799416989814268, "minimum": 0.9923608011062002, "maximum": 0.994245212644458, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9574104153406835, "standard_deviation": 0.024138068344374607, "minimum": 0.90439555663659, "maximum": 0.9883925140236194, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40072925381884994, "standard_deviation": 0.13239104311173186, "minimum": 0.11775192275240866, "maximum": 0.5374762403732868, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08257690221117367, "standard_deviation": 0.004300302458209458, "minimum": 0.07925670488911467, "maximum": 0.09096871605371588, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981175961467275, "standard_deviation": 0.0002462722021896406, "minimum": 0.9978115824733981, "maximum": 0.9985523732890629, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47712086669965237, "standard_deviation": 0.025019523299243965, "minimum": 0.433584146967034, "maximum": 0.49574453553172354, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44499662048526784, "standard_deviation": 0.20018638115277712, "minimum": 0.07849622544811417, "maximum": 0.7009884088025587, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0434424137416276, "standard_deviation": 0.0027034240772130353, "minimum": 0.03831524182076991, "maximum": 0.04701585268304527, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1580-141084-0047", "language": "en", "style": "normal", "speaker": "librispeech_1580", "source_utterance_id": "1580-141084-0047", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 84000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1580-141084-0047.wav", "source_file_sha256": "9f263b905dc84fc1def8bb73feb7cbf89f606a16c230405bc9aef0adc4d14907", "decoded_pcm_sha256": "ca185084384947ca6386941bb27f596abb096a46a774b4716da42a2e461e4cab", "original_source_sha256": "0e78a5bcc7c122a563c24dc2d7a65f1be8d851e0d45c8caa4226571676cc5d4c", "corpus_entry_id": "librispeech_1580-141084-0047"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.5208258628845215, "predicted_mos": 4.5208258628845215, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "c351b90320dd37e220a70447d155c8859fc05d3b743f250c74e7393cf8c02148", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.019626001998404, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1580-141084-0047", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c351b90320dd37e220a70447d155c8859fc05d3b743f250c74e7393cf8c02148", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "47811c228da75f3becc352fe4eec0fd525c13a036f7e91e4f7d68d6b945e3c79"}, "rows": [{"case": "librispeech_1580-141084-0047", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.188936710357666, "predicted_mos": 3.188936710357666, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "7b88dbea7575f895c749ea5f1be3abe4d0a97eaec8b489c5ddad6b259a302e5f", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.014035434000107, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1580-141084-0047", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7b88dbea7575f895c749ea5f1be3abe4d0a97eaec8b489c5ddad6b259a302e5f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8004498229028746, "全帯域_10ms_rms_envelope_nrmse": 0.07281115342323378, "全帯域_10ms_envelope_correlation": 0.9949537321220259, "全帯域_10ms_modulation_mae_db": 0.3848965565441876, "全帯域_40ms_log_rms_mae_db": 0.4317720003607835, "全帯域_40ms_rms_envelope_nrmse": 0.03826302352244726, "全帯域_40ms_envelope_correlation": 0.9985070982318829, "全帯域_40ms_modulation_mae_db": 0.3622508126170858, "300–1000 Hz_10ms_log_rms_mae_db": 1.5232807392214052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09430252110864372, "300–1000 Hz_10ms_envelope_correlation": 0.992987713944975, "300–1000 Hz_10ms_modulation_mae_db": 0.2378359873984154, "300–1000 Hz_40ms_log_rms_mae_db": 0.8909285672668311, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043056126010745495, "300–1000 Hz_40ms_envelope_correlation": 0.998524078084921, "300–1000 Hz_40ms_modulation_mae_db": 0.09565582173530218, "1000–3000 Hz_10ms_log_rms_mae_db": 1.247410011871001, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08965649574816899, "1000–3000 Hz_10ms_envelope_correlation": 0.994127178422255, "1000–3000 Hz_10ms_modulation_mae_db": 0.32877255883845397, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7444068586256354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0454699054945729, "1000–3000 Hz_40ms_envelope_correlation": 0.9984763532874393, "1000–3000 Hz_40ms_modulation_mae_db": 0.44184032699007697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4380593601821605, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1514029490699294, "3000–6000 Hz_10ms_envelope_correlation": 0.9822094882535213, "3000–6000 Hz_10ms_modulation_mae_db": 0.5900923362387752, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8740837459923035, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08364644228099032, "3000–6000 Hz_40ms_envelope_correlation": 0.9940492149565491, "3000–6000 Hz_40ms_modulation_mae_db": 0.539153375835849, "mr_stft_db": 5.629486341027671, "band_mae_db": 0.17015235876454138, "rms_gain_error_db": -0.058441543216020464, "output_peak": 0.6323025494071863}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.815174509732012, "全帯域_10ms_rms_envelope_nrmse": 0.07587625186931059, "全帯域_10ms_envelope_correlation": 0.9945604656993771, "全帯域_10ms_modulation_mae_db": 0.5224698899876644, "全帯域_40ms_log_rms_mae_db": 0.43046486457044025, "全帯域_40ms_rms_envelope_nrmse": 0.039411415631466634, "全帯域_40ms_envelope_correlation": 0.9984598854306503, "全帯域_40ms_modulation_mae_db": 0.5476369038432303, "300–1000 Hz_10ms_log_rms_mae_db": 1.541766107644251, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10680086413245554, "300–1000 Hz_10ms_envelope_correlation": 0.9909716012336054, "300–1000 Hz_10ms_modulation_mae_db": 0.327599273237298, "300–1000 Hz_40ms_log_rms_mae_db": 0.9779930655416396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05467329315816751, "300–1000 Hz_40ms_envelope_correlation": 0.9975354162328139, "300–1000 Hz_40ms_modulation_mae_db": 0.28522295827669847, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2626160701239464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10725316243958592, "1000–3000 Hz_10ms_envelope_correlation": 0.9915680229250765, "1000–3000 Hz_10ms_modulation_mae_db": 0.9373016783234047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7990414702810634, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.063640630905653, "1000–3000 Hz_40ms_envelope_correlation": 0.9969718044399136, "1000–3000 Hz_40ms_modulation_mae_db": 1.0143380377444104, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3852323180916848, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1387805480968202, "3000–6000 Hz_10ms_envelope_correlation": 0.9855008259930954, "3000–6000 Hz_10ms_modulation_mae_db": 1.1441889469260418, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8716323812753061, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0802837434814529, "3000–6000 Hz_40ms_envelope_correlation": 0.9953576417472456, "3000–6000 Hz_40ms_modulation_mae_db": 1.244517040748573, "mr_stft_db": 5.6152916810068945, "band_mae_db": 0.19827831312254254, "rms_gain_error_db": -0.08150992539565463, "output_peak": 0.6221312094884716}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8710307941079809, "全帯域_10ms_rms_envelope_nrmse": 0.07085671863517327, "全帯域_10ms_envelope_correlation": 0.9952171077743718, "全帯域_10ms_modulation_mae_db": 0.39327871339713755, "全帯域_40ms_log_rms_mae_db": 0.4348917116895858, "全帯域_40ms_rms_envelope_nrmse": 0.036600488815203516, "全帯域_40ms_envelope_correlation": 0.998628807582947, "全帯域_40ms_modulation_mae_db": 0.2752267740551198, "300–1000 Hz_10ms_log_rms_mae_db": 1.4347950331372294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07868430699055469, "300–1000 Hz_10ms_envelope_correlation": 0.9951482586335815, "300–1000 Hz_10ms_modulation_mae_db": 0.2665117656108445, "300–1000 Hz_40ms_log_rms_mae_db": 0.8892420241384319, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04637342078964627, "300–1000 Hz_40ms_envelope_correlation": 0.9982091550598349, "300–1000 Hz_40ms_modulation_mae_db": 0.08973273396502914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2764358733829353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10922181259690283, "1000–3000 Hz_10ms_envelope_correlation": 0.9915100708792914, "1000–3000 Hz_10ms_modulation_mae_db": 0.5749764185520602, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7430779986175985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058834761501633845, "1000–3000 Hz_40ms_envelope_correlation": 0.9977859766047733, "1000–3000 Hz_40ms_modulation_mae_db": 0.41607328068554494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3894591631407271, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.142597721785041, "3000–6000 Hz_10ms_envelope_correlation": 0.9841522596488796, "3000–6000 Hz_10ms_modulation_mae_db": 0.21247116012636225, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8640929550477308, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08766309002797931, "3000–6000 Hz_40ms_envelope_correlation": 0.9934379309990246, "3000–6000 Hz_40ms_modulation_mae_db": 0.11748060413328083, "mr_stft_db": 5.616792282430723, "band_mae_db": 0.24870146555998382, "rms_gain_error_db": -0.05066114347913517, "output_peak": 0.6160823646366861}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8084896220354737, "全帯域_10ms_rms_envelope_nrmse": 0.07297120045595877, "全帯域_10ms_envelope_correlation": 0.9951317772119979, "全帯域_10ms_modulation_mae_db": 0.7974262661388788, "全帯域_40ms_log_rms_mae_db": 0.4651265350935128, "全帯域_40ms_rms_envelope_nrmse": 0.04481100399983944, "全帯域_40ms_envelope_correlation": 0.9981834064101721, "全帯域_40ms_modulation_mae_db": 0.7960450589672726, "300–1000 Hz_10ms_log_rms_mae_db": 1.4726177304122712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09997952631143062, "300–1000 Hz_10ms_envelope_correlation": 0.9921447019707286, "300–1000 Hz_10ms_modulation_mae_db": 0.4201131627726961, "300–1000 Hz_40ms_log_rms_mae_db": 0.9053639657170541, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06443462718457821, "300–1000 Hz_40ms_envelope_correlation": 0.9966108797998892, "300–1000 Hz_40ms_modulation_mae_db": 0.5048106327806782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3043597774161324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10134208232339315, "1000–3000 Hz_10ms_envelope_correlation": 0.9928266455514478, "1000–3000 Hz_10ms_modulation_mae_db": 0.18641590102231043, "1000–3000 Hz_40ms_log_rms_mae_db": 0.816560973253303, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06129264327041486, "1000–3000 Hz_40ms_envelope_correlation": 0.9976399430843207, "1000–3000 Hz_40ms_modulation_mae_db": 0.3342908501506033, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4156553080374292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1354845163372546, "3000–6000 Hz_10ms_envelope_correlation": 0.9858516219432782, "3000–6000 Hz_10ms_modulation_mae_db": 1.0204991682176374, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8446078673735202, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08036448171741975, "3000–6000 Hz_40ms_envelope_correlation": 0.9948125053602371, "3000–6000 Hz_40ms_modulation_mae_db": 1.0630554109594175, "mr_stft_db": 5.616512988650556, "band_mae_db": 0.26563930982687434, "rms_gain_error_db": -0.14928511256391896, "output_peak": 0.6208087008589793}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8155866270133552, "全帯域_10ms_rms_envelope_nrmse": 0.07216038324175987, "全帯域_10ms_envelope_correlation": 0.9952631013763308, "全帯域_10ms_modulation_mae_db": 0.23790400289143007, "全帯域_40ms_log_rms_mae_db": 0.4431507716496641, "全帯域_40ms_rms_envelope_nrmse": 0.04422872278968537, "全帯域_40ms_envelope_correlation": 0.9982570633114716, "全帯域_40ms_modulation_mae_db": 0.3333251501942471, "300–1000 Hz_10ms_log_rms_mae_db": 1.4715916947518433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09466374634316563, "300–1000 Hz_10ms_envelope_correlation": 0.9930397224397379, "300–1000 Hz_10ms_modulation_mae_db": 0.19924123698309945, "300–1000 Hz_40ms_log_rms_mae_db": 0.93134116294461, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058873926756709306, "300–1000 Hz_40ms_envelope_correlation": 0.9972849237971254, "300–1000 Hz_40ms_modulation_mae_db": 0.23866564856624614, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2554460241938417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10347301736416858, "1000–3000 Hz_10ms_envelope_correlation": 0.992536912821639, "1000–3000 Hz_10ms_modulation_mae_db": 0.0787203695009962, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7726203490219826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06591950116612352, "1000–3000 Hz_40ms_envelope_correlation": 0.9972086182986196, "1000–3000 Hz_40ms_modulation_mae_db": 0.3028897948462555, "3000–6000 Hz_10ms_log_rms_mae_db": 1.415406524031293, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13231839637276332, "3000–6000 Hz_10ms_envelope_correlation": 0.9864427307429816, "3000–6000 Hz_10ms_modulation_mae_db": 0.20802721411604566, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8460760040763501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07405854026420157, "3000–6000 Hz_40ms_envelope_correlation": 0.9953566317598862, "3000–6000 Hz_40ms_modulation_mae_db": 0.28958074026679653, "mr_stft_db": 5.61783669039722, "band_mae_db": 0.24630723285081824, "rms_gain_error_db": -0.15559802101717513, "output_peak": 0.6287148114816983}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8151431885987744, "全帯域_10ms_rms_envelope_nrmse": 0.07792172881449781, "全帯域_10ms_envelope_correlation": 0.9943323999265935, "全帯域_10ms_modulation_mae_db": 0.5229880051477541, "全帯域_40ms_log_rms_mae_db": 0.4187517934249733, "全帯域_40ms_rms_envelope_nrmse": 0.043559384704929864, "全帯域_40ms_envelope_correlation": 0.9981986033155354, "全帯域_40ms_modulation_mae_db": 0.5638530508499126, "300–1000 Hz_10ms_log_rms_mae_db": 1.5110345209852265, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09663242925755039, "300–1000 Hz_10ms_envelope_correlation": 0.992645219108724, "300–1000 Hz_10ms_modulation_mae_db": 0.789439660166703, "300–1000 Hz_40ms_log_rms_mae_db": 0.9457913085506382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06110247502804745, "300–1000 Hz_40ms_envelope_correlation": 0.9969248107157372, "300–1000 Hz_40ms_modulation_mae_db": 0.40569111700754457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.282508002846719, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10223657194481017, "1000–3000 Hz_10ms_envelope_correlation": 0.9924988913423519, "1000–3000 Hz_10ms_modulation_mae_db": 0.6637394881621962, "1000–3000 Hz_40ms_log_rms_mae_db": 0.754342861276833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05647086360723084, "1000–3000 Hz_40ms_envelope_correlation": 0.9978496471073913, "1000–3000 Hz_40ms_modulation_mae_db": 0.3576961296235535, "3000–6000 Hz_10ms_log_rms_mae_db": 1.478911267061897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1585089700376432, "3000–6000 Hz_10ms_envelope_correlation": 0.9804549814908501, "3000–6000 Hz_10ms_modulation_mae_db": 0.8291080248235313, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8896624378370515, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08233161432381571, "3000–6000 Hz_40ms_envelope_correlation": 0.9945444501661497, "3000–6000 Hz_40ms_modulation_mae_db": 0.702691587346917, "mr_stft_db": 5.622188071853789, "band_mae_db": 0.22521026553824017, "rms_gain_error_db": -0.10601572331098354, "output_peak": 0.6140982737681769}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8096432618626975, "全帯域_10ms_rms_envelope_nrmse": 0.0704201031231679, "全帯域_10ms_envelope_correlation": 0.9953810240929754, "全帯域_10ms_modulation_mae_db": 0.43713351190824506, "全帯域_40ms_log_rms_mae_db": 0.3993224879012774, "全帯域_40ms_rms_envelope_nrmse": 0.039584225205697675, "全帯域_40ms_envelope_correlation": 0.9985210145612486, "全帯域_40ms_modulation_mae_db": 0.3924674906044314, "300–1000 Hz_10ms_log_rms_mae_db": 1.5522839990871415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09284362052202672, "300–1000 Hz_10ms_envelope_correlation": 0.9932009903559976, "300–1000 Hz_10ms_modulation_mae_db": 0.12028789762233205, "300–1000 Hz_40ms_log_rms_mae_db": 0.9353020482653768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05419136656321218, "300–1000 Hz_40ms_envelope_correlation": 0.9975821615150464, "300–1000 Hz_40ms_modulation_mae_db": 0.15466185237980515, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2798097607020282, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10609590829196132, "1000–3000 Hz_10ms_envelope_correlation": 0.9918999545932629, "1000–3000 Hz_10ms_modulation_mae_db": 0.5264408389482172, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7889341635042624, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0633461984162509, "1000–3000 Hz_40ms_envelope_correlation": 0.9971987093263093, "1000–3000 Hz_40ms_modulation_mae_db": 0.5550157308108824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4061266809330186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12543114898722726, "3000–6000 Hz_10ms_envelope_correlation": 0.9878216153458066, "3000–6000 Hz_10ms_modulation_mae_db": 0.39753616654931384, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8496533262828043, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06925901824293616, "3000–6000 Hz_40ms_envelope_correlation": 0.9960869672502733, "3000–6000 Hz_40ms_modulation_mae_db": 0.2784381522361014, "mr_stft_db": 5.636148705166495, "band_mae_db": 0.22902810745347416, "rms_gain_error_db": -0.11260686087070232, "output_peak": 0.6194146337005424}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8662331396214279, "全帯域_10ms_rms_envelope_nrmse": 0.07399364898883938, "全帯域_10ms_envelope_correlation": 0.9948284219986817, "全帯域_10ms_modulation_mae_db": 0.11514851600269153, "全帯域_40ms_log_rms_mae_db": 0.4497143890940237, "全帯域_40ms_rms_envelope_nrmse": 0.04299999094980203, "全帯域_40ms_envelope_correlation": 0.9981558105957742, "全帯域_40ms_modulation_mae_db": 0.23190981382561623, "300–1000 Hz_10ms_log_rms_mae_db": 1.5284886531392405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10462803572749801, "300–1000 Hz_10ms_envelope_correlation": 0.9913566535579958, "300–1000 Hz_10ms_modulation_mae_db": 0.34875421327341766, "300–1000 Hz_40ms_log_rms_mae_db": 0.9595862414353074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059613522388254535, "300–1000 Hz_40ms_envelope_correlation": 0.9970876948746541, "300–1000 Hz_40ms_modulation_mae_db": 0.3286297532608849, "1000–3000 Hz_10ms_log_rms_mae_db": 1.333356356126464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10057239423216768, "1000–3000 Hz_10ms_envelope_correlation": 0.9927944720367691, "1000–3000 Hz_10ms_modulation_mae_db": 0.2774547431283763, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8273318361045156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06283803612278674, "1000–3000 Hz_40ms_envelope_correlation": 0.9972915005295169, "1000–3000 Hz_40ms_modulation_mae_db": 0.47390360821176314, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4338964274384958, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12267623159875501, "3000–6000 Hz_10ms_envelope_correlation": 0.9883650679858096, "3000–6000 Hz_10ms_modulation_mae_db": 0.49546030123768275, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8710296784049275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07080468285888306, "3000–6000 Hz_40ms_envelope_correlation": 0.9957395229900786, "3000–6000 Hz_40ms_modulation_mae_db": 0.5592625516214396, "mr_stft_db": 5.624958176324653, "band_mae_db": 0.19760586551672865, "rms_gain_error_db": -0.08822373872273598, "output_peak": 0.6105594532767861}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84000, "output_samples": 84000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.50838951049656, "spectral_convergence": 0.342290272554515, "spectral_convergence_status": "ok", "amplitude_floor": 1.02822042353332e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.457869916769944, "spectral_convergence": 0.3437013825988664, "spectral_convergence_status": "ok", "amplitude_floor": 8.928413196720056e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.922199595816507, "spectral_convergence": 0.3545623398694279, "spectral_convergence_status": "ok", "amplitude_floor": 7.736392334055093e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 1.4784881078912855e-05, "output_power": 1.73242026077484e-05, "delta_db": 0.6883541853058242, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002602315552138397, "output_power": 0.00025404963595819583, "delta_db": -0.10441380270141742, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005931268593866254, "output_power": 0.0005845485103950103, "delta_db": -0.0632703310118094, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005283196568297616, "output_power": 0.0005207184827780319, "delta_db": -0.06293776358577201, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003079318069923152, "output_power": 0.0003055622521424767, "delta_db": -0.03354848177893477, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00026143236338356946, "output_power": 0.0002573477586690261, "delta_db": -0.06838958820349048, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17015235876454138, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.058441543216020464, "pitch": null, "output_rms_dbfs": -27.12298831506196, "output_peak": 0.6323025494071863, "reference_rms_dbfs": -27.06454677184594, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.629486341027671}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00019312594202620684, "log_rms_mae_db": 0.8004498229028746, "rms_envelope_nrmse": 0.07281115342323378, "envelope_correlation": 0.9949537321220259, "rms_gain_error_db": -0.058441975569212226, "reference_mean_power": 0.0019676998400409485, "output_mean_power": 0.0019413983347108803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 4.0378624939923077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018862042103601667, "output_power": 0.00020072299515465613, "reference_active": true, "delta_db": 0.2700841887880565}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.862312285918816e-05, "output_power": 9.804326090485272e-05, "reference_active": true, "delta_db": -0.02561002769987662}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3789377405749525e-05, "output_power": 1.1314753911869108e-05, "reference_active": true, "delta_db": -0.8589954531446297}], "active_band_count": 3, "active_band_mae_db": 0.3848965565441876}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00018244538675777041, "log_rms_mae_db": 0.4317720003607835, "rms_envelope_nrmse": 0.03826302352244726, "envelope_correlation": 0.9985070982318829, "rms_gain_error_db": -0.05844340279099594, "reference_mean_power": 0.001979010526094432, "output_mean_power": 0.0019525571933996246}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 3.3102005537920435e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017836230285954771, "output_power": 0.00019192467856336668, "reference_active": true, "delta_db": 0.3182775097251295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3320761072042273e-05, "output_power": 4.3467959199301464e-05, "reference_active": true, "delta_db": 0.014731729080226085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.132775238046075e-07, "output_power": 6.836982214696769e-07, "reference_active": true, "delta_db": -0.7537431990459018}], "active_band_count": 3, "active_band_mae_db": 0.3622508126170858}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00014606965842361196, "log_rms_mae_db": 1.5232807392214052, "rms_envelope_nrmse": 0.09430252110864372, "envelope_correlation": 0.992987713944975, "rms_gain_error_db": -0.0932241748762644, "reference_mean_power": 0.0007714288246159527, "output_mean_power": 0.0007550460586065433}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 2.155891354220712e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.355627855804233e-05, "output_power": 8.318533044712876e-05, "reference_active": true, "delta_db": -0.019323429152379015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.117824128471733e-05, "output_power": 1.9499707463209893e-05, "reference_active": true, "delta_db": -0.3586179585797178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3214088476749303e-06, "output_power": 3.588219447484683e-06, "reference_active": true, "delta_db": 0.3355665744631495}], "active_band_count": 3, "active_band_mae_db": 0.2378359873984154}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.0001346093108760191, "log_rms_mae_db": 0.8909285672668311, "rms_envelope_nrmse": 0.043056126010745495, "envelope_correlation": 0.998524078084921, "rms_gain_error_db": -0.0932260672684841, "reference_mean_power": 0.0007758659430163304, "output_mean_power": 0.0007593886154186043}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.945223621465206e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.346976557991944e-05, "output_power": 7.290358252689117e-05, "reference_active": true, "delta_db": -0.03359783802244552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.57892144723787e-06, "output_power": 9.069853417017838e-06, "reference_active": true, "delta_db": -0.2371634356438758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.32297115664878e-07, "output_power": 2.331655780942327e-07, "reference_active": true, "delta_db": 0.016206191539585243}], "active_band_count": 3, "active_band_mae_db": 0.09565582173530218}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00012444218003818797, "log_rms_mae_db": 1.247410011871001, "rms_envelope_nrmse": 0.08965649574816899, "envelope_correlation": 0.994127178422255, "rms_gain_error_db": -0.09003493653077317, "reference_mean_power": 0.0006875503513213645, "output_mean_power": 0.0006734432660921514}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.3390858183888e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.254326674690693e-05, "output_power": 3.9866371160702325e-05, "reference_active": true, "delta_db": -0.28224129008480375}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4859458498211302e-05, "output_power": 1.5385452590877623e-05, "reference_active": true, "delta_db": 0.15107292981765708}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.141405389451205e-06, "output_power": 3.56799576437788e-06, "reference_active": true, "delta_db": 0.5530034566129011}], "active_band_count": 3, "active_band_mae_db": 0.32877255883845397}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00011369765125964215, "log_rms_mae_db": 0.7444068586256354, "rms_envelope_nrmse": 0.0454699054945729, "envelope_correlation": 0.9984763532874393, "rms_gain_error_db": -0.09003299707528478, "reference_mean_power": 0.0006915015836222027, "output_mean_power": 0.0006773137298995417}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.214722623247201e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7827908938064025e-05, "output_power": 3.625148906332092e-05, "reference_active": true, "delta_db": -0.18486484582585586}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.619804659862373e-06, "output_power": 6.151844066458644e-06, "reference_active": true, "delta_db": 0.39284098591143946}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0150238495180175e-07, "output_power": 2.3936558290099294e-07, "reference_active": true, "delta_db": 0.7478151492329355}], "active_band_count": 3, "active_band_mae_db": 0.44184032699007697}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00010848831624356859, "log_rms_mae_db": 1.4380593601821605, "rms_envelope_nrmse": 0.1514029490699294, "envelope_correlation": 0.9822094882535213, "rms_gain_error_db": -0.01857678562813073, "reference_mean_power": 0.00023818463717296146, "output_mean_power": 0.00023716798709993953}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.7158903163900278e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.877460217795428e-05, "output_power": 9.854895251381047e-05, "reference_active": true, "delta_db": 0.453632776909786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.954888518804953e-05, "output_power": 4.242075373346965e-05, "reference_active": true, "delta_db": 0.30444134783183174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.296333400161532e-06, "output_power": 5.779421671529616e-06, "reference_active": true, "delta_db": -1.012202883974708}], "active_band_count": 3, "active_band_mae_db": 0.5900923362387752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 7.397564116979367e-05, "log_rms_mae_db": 0.8740837459923035, "rms_envelope_nrmse": 0.08364644228099032, "envelope_correlation": 0.9940492149565491, "rms_gain_error_db": -0.018572891466827394, "reference_mean_power": 0.0002395532500233675, "output_mean_power": 0.00023853097214425838}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.5691901914560294e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.470858674877874e-05, "output_power": 0.00010494090712383027, "reference_active": true, "delta_db": 0.44555457907050366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.833295648941501e-05, "output_power": 1.9704911442968292e-05, "reference_active": true, "delta_db": 0.3134197971054885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.840671065644946e-07, "output_power": 3.972436807001715e-07, "reference_active": true, "delta_db": -0.8584857513315548}], "active_band_count": 3, "active_band_mae_db": 0.539153375835849}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9019854020256165, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1feecd66061d6d159ff1d2adb896b825c7d562ee3bd0859c45a84b89f33ccd28", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9924702685715117, "standard_deviation": 0.0008488654831791938, "minimum": 0.9915100708792914, "maximum": 0.994127178422255, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2802427345828837, "standard_deviation": 0.027841550331733435, "minimum": 1.247410011871001, "maximum": 1.333356356126464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4467277495595019, "standard_deviation": 0.2819111410110632, "minimum": 0.0787203695009962, "maximum": 0.9373016783234047, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10248143061764484, "standard_deviation": 0.0059968004751759655, "minimum": 0.08965649574816899, "maximum": 0.10922181259690283, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975528190847855, "standard_deviation": 0.00048590249257379455, "minimum": 0.9969718044399136, "maximum": 0.9984763532874393, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7807895638356492, "standard_deviation": 0.032419727170856075, "minimum": 0.7430779986175985, "maximum": 0.8273318361045156, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4870059698828863, "standard_deviation": 0.2280003757658249, "minimum": 0.3028897948462555, "maximum": 1.0143380377444104, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059726567560583324, "standard_deviation": 0.006475836573962881, "minimum": 0.0454699054945729, "maximum": 0.06591950116612352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9850998239255278, "standard_deviation": 0.002710451953483445, "minimum": 0.9804549814908501, "maximum": 0.9883650679858096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4203433811145882, "standard_deviation": 0.03014921240587784, "minimum": 1.3852323180916848, "maximum": 1.478911267061897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6121729147794238, "standard_deviation": 0.3545834154357208, "minimum": 0.20802721411604566, "maximum": 1.1441889469260418, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13840006028567925, "standard_deviation": 0.012259847251982361, "minimum": 0.12267623159875501, "maximum": 0.1585089700376432, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9949231081536805, "standard_deviation": 0.0008882501813782917, "minimum": 0.9934379309990246, "maximum": 0.9960869672502733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8638547995362492, "standard_deviation": 0.01590554140207278, "minimum": 0.8446078673735202, "maximum": 0.8896624378370515, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5992724328935468, "standard_deviation": 0.3924852707382902, "minimum": 0.11748060413328083, "maximum": 1.244517040748573, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07855145164970985, "standard_deviation": 0.006503915571896641, "minimum": 0.06925901824293616, "maximum": 0.08766309002797931, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9926868576556682, "standard_deviation": 0.0012842490501939513, "minimum": 0.9909716012336054, "maximum": 0.9951482586335815, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5044823097973261, "standard_deviation": 0.04071005584359558, "minimum": 1.4347950331372294, "maximum": 1.5522839990871415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.33872289963310076, "standard_deviation": 0.20443508810289307, "minimum": 0.12028789762233205, "maximum": 0.789439660166703, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09606688129916567, "standard_deviation": 0.008633534637139596, "minimum": 0.07868430699055469, "maximum": 0.10680086413245554, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974698900100027, "standard_deviation": 0.0006425417704521696, "minimum": 0.9966108797998892, "maximum": 0.998524078084921, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9294435479824862, "standard_deviation": 0.032182964456612485, "minimum": 0.8892420241384319, "maximum": 0.9779930655416396, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.26288381474652356, "standard_deviation": 0.14831871462784707, "minimum": 0.08973273396502914, "maximum": 0.5048106327806782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.055289844734920116, "standard_deviation": 0.007365041259700166, "minimum": 0.043056126010745495, "maximum": 0.06443462718457821, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2226153648291504, "standard_deviation": 0.03187215540132097, "minimum": 0.17015235876454138, "maximum": 0.26563930982687434, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.62240186710725, "standard_deviation": 0.007399687899576351, "minimum": 5.6152916810068945, "maximum": 5.636148705166495, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6205139995773159, "standard_deviation": 0.007270566523515216, "minimum": 0.6105594532767861, "maximum": 0.6323025494071863, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10029275857204077, "standard_deviation": 0.03846443580091126, "minimum": -0.15559802101717513, "maximum": -0.05066114347913517, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949585037752942, "standard_deviation": 0.00036558899759465567, "minimum": 0.9943323999265935, "maximum": 0.9953810240929754, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8252188707343245, "standard_deviation": 0.02728700459220604, "minimum": 0.8004498229028746, "maximum": 0.8710307941079809, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4264056827522486, "standard_deviation": 0.2040875678115217, "minimum": 0.11514851600269153, "maximum": 0.7974262661388788, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07337639856899268, "standard_deviation": 0.0025168331075320722, "minimum": 0.0704201031231679, "maximum": 0.07792172881449781, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983639611799602, "standard_deviation": 0.00018485365928504983, "minimum": 0.9981558105957742, "maximum": 0.998628807582947, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4341493192230326, "standard_deviation": 0.01983019136574394, "minimum": 0.3993224879012774, "maximum": 0.4651265350935128, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43783938186961446, "standard_deviation": 0.1866209144085369, "minimum": 0.23190981382561623, "maximum": 0.7960450589672726, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04118228195238398, "standard_deviation": 0.0030847032733701185, "minimum": 0.036600488815203516, "maximum": 0.04481100399983944, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0047", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.2845141887664795, "predicted_mos": 3.2845141887664795, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "d1231bb0cc4a011864d3cb73644833c093afe980658fe56bb1fc5300af31f99c", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.936270299833268, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8060941944605274, "全帯域_10ms_rms_envelope_nrmse": 0.07379868241081276, "全帯域_10ms_envelope_correlation": 0.994812985761589, "全帯域_10ms_modulation_mae_db": 0.3944105230261024, "全帯域_40ms_log_rms_mae_db": 0.4466869860187963, "全帯域_40ms_rms_envelope_nrmse": 0.0383254505909456, "全帯域_40ms_envelope_correlation": 0.9984996686015655, "全帯域_40ms_modulation_mae_db": 0.32384434722627103, "300–1000 Hz_10ms_log_rms_mae_db": 1.393716929679926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09501665843546916, "300–1000 Hz_10ms_envelope_correlation": 0.9928559516558306, "300–1000 Hz_10ms_modulation_mae_db": 0.15839654801895173, "300–1000 Hz_40ms_log_rms_mae_db": 0.7883968749334552, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04233232055096795, "300–1000 Hz_40ms_envelope_correlation": 0.998524172154126, "300–1000 Hz_40ms_modulation_mae_db": 0.16520880769000942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1908029051101017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08819715088650114, "1000–3000 Hz_10ms_envelope_correlation": 0.9943513079676909, "1000–3000 Hz_10ms_modulation_mae_db": 0.1992425358218396, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6677738259962458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04335658677287043, "1000–3000 Hz_40ms_envelope_correlation": 0.9986641244841608, "1000–3000 Hz_40ms_modulation_mae_db": 0.31345435924922005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5113946134897738, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15593763214343137, "3000–6000 Hz_10ms_envelope_correlation": 0.981076390295545, "3000–6000 Hz_10ms_modulation_mae_db": 0.4961113908768886, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9268322527832232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0877413729314617, "3000–6000 Hz_40ms_envelope_correlation": 0.9934775312505679, "3000–6000 Hz_40ms_modulation_mae_db": 0.41333648234564513, "mr_stft_db": 5.605362878799837, "band_mae_db": 0.1624973183335798, "rms_gain_error_db": -0.05675369441386202, "output_peak": 0.6338498304864236}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8274546964334262, "全帯域_10ms_rms_envelope_nrmse": 0.07591518203961443, "全帯域_10ms_envelope_correlation": 0.9945535592071068, "全帯域_10ms_modulation_mae_db": 0.49710080291598996, "全帯域_40ms_log_rms_mae_db": 0.44096420237346423, "全帯域_40ms_rms_envelope_nrmse": 0.03980530495186485, "全帯域_40ms_envelope_correlation": 0.9984257535155414, "全帯域_40ms_modulation_mae_db": 0.5172163649359771, "300–1000 Hz_10ms_log_rms_mae_db": 1.417295014910592, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10851510322920621, "300–1000 Hz_10ms_envelope_correlation": 0.9906702381055192, "300–1000 Hz_10ms_modulation_mae_db": 0.2505495263094187, "300–1000 Hz_40ms_log_rms_mae_db": 0.8137228330515389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05627852442766638, "300–1000 Hz_40ms_envelope_correlation": 0.9973552999545956, "300–1000 Hz_40ms_modulation_mae_db": 0.36034675794567733, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2248069547601481, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10645844736178227, "1000–3000 Hz_10ms_envelope_correlation": 0.9917086494313926, "1000–3000 Hz_10ms_modulation_mae_db": 0.7007750346781667, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7145860204425625, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06246234065872775, "1000–3000 Hz_40ms_envelope_correlation": 0.9971032923962903, "1000–3000 Hz_40ms_modulation_mae_db": 0.7945805069477583, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4354194355214667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13982222712256534, "3000–6000 Hz_10ms_envelope_correlation": 0.985447992782728, "3000–6000 Hz_10ms_modulation_mae_db": 1.1308881108859892, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8993757468092661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08366454197210446, "3000–6000 Hz_40ms_envelope_correlation": 0.9951137787629191, "3000–6000 Hz_40ms_modulation_mae_db": 1.2575789386500684, "mr_stft_db": 5.597437229092972, "band_mae_db": 0.19238601366664318, "rms_gain_error_db": -0.08093969186811692, "output_peak": 0.6227897687486311}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.876456281362541, "全帯域_10ms_rms_envelope_nrmse": 0.07074347877337003, "全帯域_10ms_envelope_correlation": 0.9952381089912521, "全帯域_10ms_modulation_mae_db": 0.46333660959200335, "全帯域_40ms_log_rms_mae_db": 0.42865619547393324, "全帯域_40ms_rms_envelope_nrmse": 0.0364654514609523, "全帯域_40ms_envelope_correlation": 0.9986476360789659, "全帯域_40ms_modulation_mae_db": 0.2802931884718394, "300–1000 Hz_10ms_log_rms_mae_db": 1.352124135011367, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0807678100295778, "300–1000 Hz_10ms_envelope_correlation": 0.994865687887489, "300–1000 Hz_10ms_modulation_mae_db": 0.24750479362810984, "300–1000 Hz_40ms_log_rms_mae_db": 0.8164629564066906, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047382631301036045, "300–1000 Hz_40ms_envelope_correlation": 0.9981149113001775, "300–1000 Hz_40ms_modulation_mae_db": 0.0614558198149949, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2406646054136399, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1072198239194916, "1000–3000 Hz_10ms_envelope_correlation": 0.9917730942871597, "1000–3000 Hz_10ms_modulation_mae_db": 0.7266554897177956, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6865190177524212, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05676707074484993, "1000–3000 Hz_40ms_envelope_correlation": 0.9978924816709099, "1000–3000 Hz_40ms_modulation_mae_db": 0.5426668433061459, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4289309141662383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13695125983091574, "3000–6000 Hz_10ms_envelope_correlation": 0.985386570487362, "3000–6000 Hz_10ms_modulation_mae_db": 0.40448716240082994, "3000–6000 Hz_40ms_log_rms_mae_db": 0.857477998001156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08273081294379703, "3000–6000 Hz_40ms_envelope_correlation": 0.994201697399564, "3000–6000 Hz_40ms_modulation_mae_db": 0.25915175230975485, "mr_stft_db": 5.606529302787244, "band_mae_db": 0.23957874935160017, "rms_gain_error_db": -0.0575993888098492, "output_peak": 0.6169108340270937}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8136816914439828, "全帯域_10ms_rms_envelope_nrmse": 0.07496870497860787, "全帯域_10ms_envelope_correlation": 0.9949078619575741, "全帯域_10ms_modulation_mae_db": 0.7365729167016547, "全帯域_40ms_log_rms_mae_db": 0.4728521275594271, "全帯域_40ms_rms_envelope_nrmse": 0.047118092820820905, "全帯域_40ms_envelope_correlation": 0.998033775877655, "全帯域_40ms_modulation_mae_db": 0.7794230778119638, "300–1000 Hz_10ms_log_rms_mae_db": 1.3882267185005597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10008168207117887, "300–1000 Hz_10ms_envelope_correlation": 0.992141425386447, "300–1000 Hz_10ms_modulation_mae_db": 0.3099746424961341, "300–1000 Hz_40ms_log_rms_mae_db": 0.8320272586240891, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06547140038537708, "300–1000 Hz_40ms_envelope_correlation": 0.9965129136721029, "300–1000 Hz_40ms_modulation_mae_db": 0.3975744152477927, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2473007527470683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10071585603819004, "1000–3000 Hz_10ms_envelope_correlation": 0.9930240829201351, "1000–3000 Hz_10ms_modulation_mae_db": 0.35183681311569814, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7183418550147234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06088331878588891, "1000–3000 Hz_40ms_envelope_correlation": 0.9977935413353938, "1000–3000 Hz_40ms_modulation_mae_db": 0.5681601083621768, "3000–6000 Hz_10ms_log_rms_mae_db": 1.475852588468392, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13668268643146142, "3000–6000 Hz_10ms_envelope_correlation": 0.9856418020826193, "3000–6000 Hz_10ms_modulation_mae_db": 1.0723961996832196, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8771731253195215, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08238282522800276, "3000–6000 Hz_40ms_envelope_correlation": 0.994600616774753, "3000–6000 Hz_40ms_modulation_mae_db": 1.1395893265442396, "mr_stft_db": 5.59543404930621, "band_mae_db": 0.2649608063742502, "rms_gain_error_db": -0.16693825971477613, "output_peak": 0.6206163262730774}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.814912748715715, "全帯域_10ms_rms_envelope_nrmse": 0.07212897261744056, "全帯域_10ms_envelope_correlation": 0.99530813119838, "全帯域_10ms_modulation_mae_db": 0.13137374977862246, "全帯域_40ms_log_rms_mae_db": 0.43241299313140713, "全帯域_40ms_rms_envelope_nrmse": 0.04550260321800288, "全帯域_40ms_envelope_correlation": 0.9981817342305795, "全帯域_40ms_modulation_mae_db": 0.24697142889152532, "300–1000 Hz_10ms_log_rms_mae_db": 1.3445945913375426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.093469039216034, "300–1000 Hz_10ms_envelope_correlation": 0.993186132314518, "300–1000 Hz_10ms_modulation_mae_db": 0.0627606351087268, "300–1000 Hz_40ms_log_rms_mae_db": 0.7697193257518282, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05811046563882312, "300–1000 Hz_40ms_envelope_correlation": 0.9973166680092423, "300–1000 Hz_40ms_modulation_mae_db": 0.2927652633113796, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2424586747673712, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10259964941306102, "1000–3000 Hz_10ms_envelope_correlation": 0.9928531317641839, "1000–3000 Hz_10ms_modulation_mae_db": 0.2566640860696263, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7071996842019215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06653779921288483, "1000–3000 Hz_40ms_envelope_correlation": 0.9973543931345221, "1000–3000 Hz_40ms_modulation_mae_db": 0.35325198240537165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4547246843104054, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13128879754078887, "3000–6000 Hz_10ms_envelope_correlation": 0.9866334085843173, "3000–6000 Hz_10ms_modulation_mae_db": 0.250015176253859, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8558923929141848, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07303464902222814, "3000–6000 Hz_40ms_envelope_correlation": 0.9955015946462279, "3000–6000 Hz_40ms_modulation_mae_db": 0.23994998118720054, "mr_stft_db": 5.604081043709648, "band_mae_db": 0.23920487529023404, "rms_gain_error_db": -0.16584877285557825, "output_peak": 0.6232091675483992}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8212784936010077, "全帯域_10ms_rms_envelope_nrmse": 0.07702843446413445, "全帯域_10ms_envelope_correlation": 0.9944426764995528, "全帯域_10ms_modulation_mae_db": 0.4806821666746702, "全帯域_40ms_log_rms_mae_db": 0.416372081591344, "全帯域_40ms_rms_envelope_nrmse": 0.04203377685556229, "全帯域_40ms_envelope_correlation": 0.998309229363311, "全帯域_40ms_modulation_mae_db": 0.4989714790664593, "300–1000 Hz_10ms_log_rms_mae_db": 1.38805000962139, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09249918040704498, "300–1000 Hz_10ms_envelope_correlation": 0.993257383619668, "300–1000 Hz_10ms_modulation_mae_db": 0.6616535675705734, "300–1000 Hz_40ms_log_rms_mae_db": 0.8342327310657162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05626224608612854, "300–1000 Hz_40ms_envelope_correlation": 0.9973876423668778, "300–1000 Hz_40ms_modulation_mae_db": 0.31065637741396135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.245128123577655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10253688689048837, "1000–3000 Hz_10ms_envelope_correlation": 0.992457664210574, "1000–3000 Hz_10ms_modulation_mae_db": 0.6283081168127375, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7435710878167817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056675826555410966, "1000–3000 Hz_40ms_envelope_correlation": 0.9978313359353373, "1000–3000 Hz_40ms_modulation_mae_db": 0.25632033691417433, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5001603552530882, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15664521995379713, "3000–6000 Hz_10ms_envelope_correlation": 0.9809137321409306, "3000–6000 Hz_10ms_modulation_mae_db": 0.9487428115793254, "3000–6000 Hz_40ms_log_rms_mae_db": 0.905306961865485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08377720125311845, "3000–6000 Hz_40ms_envelope_correlation": 0.9943040249519417, "3000–6000 Hz_40ms_modulation_mae_db": 0.7157697354376807, "mr_stft_db": 5.606074184921911, "band_mae_db": 0.20399009606208507, "rms_gain_error_db": -0.10200114395995677, "output_peak": 0.6125470289027836}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8221067988480086, "全帯域_10ms_rms_envelope_nrmse": 0.07058339063921158, "全帯域_10ms_envelope_correlation": 0.9953714724540652, "全帯域_10ms_modulation_mae_db": 0.4689736069003039, "全帯域_40ms_log_rms_mae_db": 0.40137220675754887, "全帯域_40ms_rms_envelope_nrmse": 0.04003914528543256, "全帯域_40ms_envelope_correlation": 0.9984957000942825, "全帯域_40ms_modulation_mae_db": 0.387924699117623, "300–1000 Hz_10ms_log_rms_mae_db": 1.4343698625354224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08857768775495584, "300–1000 Hz_10ms_envelope_correlation": 0.9938208448519551, "300–1000 Hz_10ms_modulation_mae_db": 0.17212846338780233, "300–1000 Hz_40ms_log_rms_mae_db": 0.8445632615500657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051340504330994186, "300–1000 Hz_40ms_envelope_correlation": 0.997840306369084, "300–1000 Hz_40ms_modulation_mae_db": 0.25471916132201744, "1000–3000 Hz_10ms_log_rms_mae_db": 1.238841711394062, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10591994110641105, "1000–3000 Hz_10ms_envelope_correlation": 0.991971574376987, "1000–3000 Hz_10ms_modulation_mae_db": 0.6925611030297302, "1000–3000 Hz_40ms_log_rms_mae_db": 0.705311865567794, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06261711530288451, "1000–3000 Hz_40ms_envelope_correlation": 0.9973246045736673, "1000–3000 Hz_40ms_modulation_mae_db": 0.8596690803262095, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4638834484578112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13282629065644658, "3000–6000 Hz_10ms_envelope_correlation": 0.986336271604672, "3000–6000 Hz_10ms_modulation_mae_db": 0.34964465715892046, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8781328166620521, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0754463391536555, "3000–6000 Hz_40ms_envelope_correlation": 0.9953600892672113, "3000–6000 Hz_40ms_modulation_mae_db": 0.19934246026524716, "mr_stft_db": 5.618208259465575, "band_mae_db": 0.21753965687223417, "rms_gain_error_db": -0.11690188038894198, "output_peak": 0.6205413366913403}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8527795197948448, "全帯域_10ms_rms_envelope_nrmse": 0.07495850031086225, "全帯域_10ms_envelope_correlation": 0.9946726340465405, "全帯域_10ms_modulation_mae_db": 0.17654134649315323, "全帯域_40ms_log_rms_mae_db": 0.4440856553903569, "全帯域_40ms_rms_envelope_nrmse": 0.04323287242096386, "全帯域_40ms_envelope_correlation": 0.9981152784385263, "全帯域_40ms_modulation_mae_db": 0.22868906138402786, "300–1000 Hz_10ms_log_rms_mae_db": 1.4360243624159525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10470155098078084, "300–1000 Hz_10ms_envelope_correlation": 0.9913270147163817, "300–1000 Hz_10ms_modulation_mae_db": 0.44058763904673653, "300–1000 Hz_40ms_log_rms_mae_db": 0.8629934018015163, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05864892219287474, "300–1000 Hz_40ms_envelope_correlation": 0.9971526235575148, "300–1000 Hz_40ms_modulation_mae_db": 0.459436365944674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2952355396353683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09783381231415623, "1000–3000 Hz_10ms_envelope_correlation": 0.9932028042805686, "1000–3000 Hz_10ms_modulation_mae_db": 0.4348808672913202, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7424619212311117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05908063823542873, "1000–3000 Hz_40ms_envelope_correlation": 0.9976556561841489, "1000–3000 Hz_40ms_modulation_mae_db": 0.8600674140235246, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4579591542098866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1256013922315854, "3000–6000 Hz_10ms_envelope_correlation": 0.98789053271965, "3000–6000 Hz_10ms_modulation_mae_db": 0.6743501380694994, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8564967969220321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07340050799858146, "3000–6000 Hz_40ms_envelope_correlation": 0.9954955257008977, "3000–6000 Hz_40ms_modulation_mae_db": 0.7688137486378884, "mr_stft_db": 5.601684835781246, "band_mae_db": 0.17576135466565915, "rms_gain_error_db": -0.07878151619834696, "output_peak": 0.6077839864383474}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84000, "output_samples": 84000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.494289961126331, "spectral_convergence": 0.3435593387434393, "spectral_convergence_status": "ok", "amplitude_floor": 1.02822042353332e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.446686358200902, "spectral_convergence": 0.344744769598891, "spectral_convergence_status": "ok", "amplitude_floor": 8.928413196720056e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8751123170722765, "spectral_convergence": 0.3541198490868789, "spectral_convergence_status": "ok", "amplitude_floor": 7.736392334055093e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 1.4784881078912855e-05, "output_power": 1.6990144063584082e-05, "delta_db": 0.603792257127726, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002602315552138397, "output_power": 0.0002544428232699932, "delta_db": -0.09769751376531174, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005931268593866254, "output_power": 0.0005900965728447593, "delta_db": -0.02224498881868219, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005283196568297616, "output_power": 0.0005233199658535061, "delta_db": -0.04129465103193359, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003079318069923152, "output_power": 0.0002984293965950006, "delta_db": -0.13612949701713192, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00026143236338356946, "output_power": 0.0002570258764637111, "delta_db": -0.07382500224069319, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1624973183335798, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05675369441386202, "pitch": null, "output_rms_dbfs": -27.1213004662598, "output_peak": 0.6338498304864236, "reference_rms_dbfs": -27.06454677184594, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.605362878799837}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00019312594202620684, "log_rms_mae_db": 0.8060941944605274, "rms_envelope_nrmse": 0.07379868241081276, "envelope_correlation": 0.994812985761589, "rms_gain_error_db": -0.05675409046859306, "reference_mean_power": 0.0019676998400409485, "output_mean_power": 0.001942153005635275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 4.0378624939923077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018862042103601667, "output_power": 0.00019986821081319954, "reference_active": true, "delta_db": 0.2515501480702389}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.862312285918816e-05, "output_power": 9.640531403340618e-05, "reference_active": true, "delta_db": -0.09877776511119699}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3789377405749525e-05, "output_power": 1.1382935959474128e-05, "reference_active": true, "delta_db": -0.8329036558968715}], "active_band_count": 3, "active_band_mae_db": 0.3944105230261024}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00018244538675777041, "log_rms_mae_db": 0.4466869860187963, "rms_envelope_nrmse": 0.0383254505909456, "envelope_correlation": 0.9984996686015655, "rms_gain_error_db": -0.05675529488098502, "reference_mean_power": 0.001979010526094432, "output_mean_power": 0.0019533163022687526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 3.3102005537920435e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017836230285954771, "output_power": 0.00019050840554995762, "reference_active": true, "delta_db": 0.28611071286458556}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3320761072042273e-05, "output_power": 4.26943410102509e-05, "reference_active": true, "delta_db": -0.06325763252610594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.132775238046075e-07, "output_power": 7.047292549259895e-07, "reference_active": true, "delta_db": -0.6221646962881215}], "active_band_count": 3, "active_band_mae_db": 0.32384434722627103}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00014606965842361196, "log_rms_mae_db": 1.393716929679926, "rms_envelope_nrmse": 0.09501665843546916, "envelope_correlation": 0.9928559516558306, "rms_gain_error_db": -0.06488979697372478, "reference_mean_power": 0.0007714288246159527, "output_mean_power": 0.000759988258472658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 2.155891354220712e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.355627855804233e-05, "output_power": 8.321382272802067e-05, "reference_active": true, "delta_db": -0.017836157101649635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.117824128471733e-05, "output_power": 1.9305530619972236e-05, "reference_active": true, "delta_db": -0.40208149059179965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3214088476749303e-06, "output_power": 3.3639500449841332e-06, "reference_active": true, "delta_db": 0.05527199636340589}], "active_band_count": 3, "active_band_mae_db": 0.15839654801895173}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.0001346093108760191, "log_rms_mae_db": 0.7883968749334552, "rms_envelope_nrmse": 0.04233232055096795, "envelope_correlation": 0.998524172154126, "rms_gain_error_db": -0.06489138427882986, "reference_mean_power": 0.0007758659430163304, "output_mean_power": 0.0007643592934496612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.945223621465206e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.346976557991944e-05, "output_power": 7.318227635204397e-05, "reference_active": true, "delta_db": -0.017027399491539478}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.57892144723787e-06, "output_power": 8.909812770096367e-06, "reference_active": true, "delta_db": -0.31448033892296856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.32297115664878e-07, "output_power": 2.236824695646664e-07, "reference_active": true, "delta_db": -0.16411868465552024}], "active_band_count": 3, "active_band_mae_db": 0.16520880769000942}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00012444218003818797, "log_rms_mae_db": 1.1908029051101017, "rms_envelope_nrmse": 0.08819715088650114, "envelope_correlation": 0.9943513079676909, "rms_gain_error_db": -0.10020838536760296, "reference_mean_power": 0.0006875503513213645, "output_mean_power": 0.000671867555931282}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.3390858183888e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.254326674690693e-05, "output_power": 3.943028820918024e-05, "reference_active": true, "delta_db": -0.33000883294929645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4859458498211302e-05, "output_power": 1.5286312164983908e-05, "reference_active": true, "delta_db": 0.12299740811888013}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.141405389451205e-06, "output_power": 3.2478511795204953e-06, "reference_active": true, "delta_db": 0.14472136639734226}], "active_band_count": 3, "active_band_mae_db": 0.1992425358218396}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00011369765125964215, "log_rms_mae_db": 0.6677738259962458, "rms_envelope_nrmse": 0.04335658677287043, "envelope_correlation": 0.9986641244841608, "rms_gain_error_db": -0.10020545580266854, "reference_mean_power": 0.0006915015836222027, "output_mean_power": 0.0006757291177521494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.214722623247201e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7827908938064025e-05, "output_power": 3.575097943302452e-05, "reference_active": true, "delta_db": -0.2452439068016446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.619804659862373e-06, "output_power": 6.1959394878667434e-06, "reference_active": true, "delta_db": 0.42385947521166595}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0150238495180175e-07, "output_power": 2.1448956099172942e-07, "reference_active": true, "delta_db": 0.2712596957343496}], "active_band_count": 3, "active_band_mae_db": 0.31345435924922005}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00010848831624356859, "log_rms_mae_db": 1.5113946134897738, "rms_envelope_nrmse": 0.15593763214343137, "envelope_correlation": 0.981076390295545, "rms_gain_error_db": -0.06688904815583445, "reference_mean_power": 0.00023818463717296146, "output_mean_power": 0.00023454427753167892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.7158903163900278e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.877460217795428e-05, "output_power": 9.801034236941947e-05, "reference_active": true, "delta_db": 0.4298317148220581}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.954888518804953e-05, "output_power": 4.150950105957481e-05, "reference_active": true, "delta_db": 0.21013267203939354}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.296333400161532e-06, "output_power": 6.001608915207549e-06, "reference_active": true, "delta_db": -0.8483697857692141}], "active_band_count": 3, "active_band_mae_db": 0.4961113908768886}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 7.397564116979367e-05, "log_rms_mae_db": 0.9268322527832232, "rms_envelope_nrmse": 0.0877413729314617, "envelope_correlation": 0.9934775312505679, "rms_gain_error_db": -0.06688508744993607, "reference_mean_power": 0.0002395532500233675, "output_mean_power": 0.0002358921879459246}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.5691901914560294e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.470858674877874e-05, "output_power": 0.00010418691577161583, "reference_active": true, "delta_db": 0.4142382579800362}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.833295648941501e-05, "output_power": 1.9275606821823046e-05, "reference_active": true, "delta_db": 0.21775551401578816}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.840671065644946e-07, "output_power": 4.208273960839953e-07, "reference_active": true, "delta_db": -0.608015675041111}], "active_band_count": 3, "active_band_mae_db": 0.41333648234564513}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9073186830278961, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "89e8b9daeafbb1f8e9e1ea7376583561ade1d8dd8b394f8ade4334ce853dcecb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9926677886548365, "standard_deviation": 0.0008889053727926787, "minimum": 0.9917086494313926, "maximum": 0.9943513079676909, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2406549084256766, "standard_deviation": 0.028750511608545408, "minimum": 1.1908029051101017, "maximum": 1.2952355396353683, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.49886550581711425, "standard_deviation": 0.21416816852068546, "minimum": 0.1992425358218396, "maximum": 0.7266554897177956, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10143519599126022, "standard_deviation": 0.006211864690249861, "minimum": 0.08819715088650114, "maximum": 0.1072198239194916, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9977024287143037, "standard_deviation": 0.00047972154194067084, "minimum": 0.9971032923962903, "maximum": 0.9986641244841608, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7107206597529452, "standard_deviation": 0.02572237564927513, "minimum": 0.6677738259962458, "maximum": 0.7435710878167817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5685213289418227, "standard_deviation": 0.24796514148467638, "minimum": 0.25632033691417433, "maximum": 0.8600674140235246, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05854758703361826, "standard_deviation": 0.006958138583353521, "minimum": 0.04335658677287043, "maximum": 0.06653779921288483, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.984915837587228, "standard_deviation": 0.002552491797430201, "minimum": 0.9809137321409306, "maximum": 0.98789053271965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4660406492346327, "standard_deviation": 0.028859801180574674, "minimum": 1.4289309141662383, "maximum": 1.5113946134897738, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6658294558635665, "standard_deviation": 0.34468764258453916, "minimum": 0.250015176253859, "maximum": 1.1308881108859892, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13946943823887398, "standard_deviation": 0.011236531926946526, "minimum": 0.1256013922315854, "maximum": 0.15664521995379713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947568573442603, "standard_deviation": 0.0007334743496629414, "minimum": 0.9934775312505679, "maximum": 0.9955015946462279, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8820860114096152, "standard_deviation": 0.02625861947644752, "minimum": 0.8558923929141848, "maximum": 0.9268322527832232, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6241915531722156, "standard_deviation": 0.41437885395924495, "minimum": 0.19934246026524716, "maximum": 1.2575789386500684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08027228131286869, "standard_deviation": 0.005515851269027475, "minimum": 0.07303464902222814, "maximum": 0.0877413729314617, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927655848172261, "standard_deviation": 0.0013522698295849256, "minimum": 0.9906702381055192, "maximum": 0.994865687887489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.394300203001594, "standard_deviation": 0.03422823677645963, "minimum": 1.3445945913375426, "maximum": 1.4360243624159525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.28794447694580666, "standard_deviation": 0.1879898459932958, "minimum": 0.0627606351087268, "maximum": 0.6616535675705734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09545358901553097, "standard_deviation": 0.008891048455210444, "minimum": 0.0807678100295778, "maximum": 0.10851510322920621, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997525567172965, "standard_deviation": 0.0006210367209934302, "minimum": 0.9965129136721029, "maximum": 0.998524172154126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8202648303981126, "standard_deviation": 0.03018341109670161, "minimum": 0.7697193257518282, "maximum": 0.8629934018015163, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.28777037108631337, "standard_deviation": 0.12799229188544575, "minimum": 0.0614558198149949, "maximum": 0.459436365944674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0544783768642335, "standard_deviation": 0.007226709754096128, "minimum": 0.04233232055096795, "maximum": 0.06547140038537708, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2119898588270357, "standard_deviation": 0.03494658898436744, "minimum": 0.1624973183335798, "maximum": 0.2649608063742502, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.604351472983081, "standard_deviation": 0.006920522812523057, "minimum": 5.59543404930621, "maximum": 5.618208259465575, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6197810348895121, "standard_deviation": 0.0077874387883967245, "minimum": 0.6077839864383474, "maximum": 0.6338498304864236, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10322054352617853, "standard_deviation": 0.04392374660427864, "minimum": -0.16693825971477613, "maximum": -0.05675369441386202, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949134287645076, "standard_deviation": 0.0003566164322323271, "minimum": 0.9944426764995528, "maximum": 0.9953714724540652, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8293455530825067, "standard_deviation": 0.02355513879769841, "minimum": 0.8060941944605274, "maximum": 0.876456281362541, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41862396526031254, "standard_deviation": 0.19173592274589918, "minimum": 0.13137374977862246, "maximum": 0.7365729167016547, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07376566827925675, "standard_deviation": 0.002392744423963881, "minimum": 0.07058339063921158, "maximum": 0.07702843446413445, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983385970250535, "standard_deviation": 0.00021447022325278873, "minimum": 0.998033775877655, "maximum": 0.9986476360789659, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4354253060370347, "standard_deviation": 0.021416973402291446, "minimum": 0.40137220675754887, "maximum": 0.4728521275594271, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4079167058632108, "standard_deviation": 0.18517787060473648, "minimum": 0.22868906138402786, "maximum": 0.7794230778119638, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04156533720056815, "standard_deviation": 0.0036108555213129902, "minimum": 0.0364654514609523, "maximum": 0.047118092820820905, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0047", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.1341605186462402, "predicted_mos": 2.1341605186462402, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "2742a9567707cbc820e81671d9e14db64b027553c4be21c3be0ea387bf96d221", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.183811999857426, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7638664772489364, "全帯域_10ms_rms_envelope_nrmse": 0.06607419962465326, "全帯域_10ms_envelope_correlation": 0.9958856099961639, "全帯域_10ms_modulation_mae_db": 0.41202918189692683, "全帯域_40ms_log_rms_mae_db": 0.38121015156747645, "全帯域_40ms_rms_envelope_nrmse": 0.0337294678183274, "全帯域_40ms_envelope_correlation": 0.9988940891566326, "全帯域_40ms_modulation_mae_db": 0.3468279321443481, "300–1000 Hz_10ms_log_rms_mae_db": 2.0258038992542264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08765849910017846, "300–1000 Hz_10ms_envelope_correlation": 0.9939443532911625, "300–1000 Hz_10ms_modulation_mae_db": 0.20915537937162695, "300–1000 Hz_40ms_log_rms_mae_db": 1.4542331645604707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04927397833875958, "300–1000 Hz_40ms_envelope_correlation": 0.9979714024142325, "300–1000 Hz_40ms_modulation_mae_db": 0.16947862435126793, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5532772751458566, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10884793478348868, "1000–3000 Hz_10ms_envelope_correlation": 0.9913596210982843, "1000–3000 Hz_10ms_modulation_mae_db": 0.9514648952966782, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0419016018682763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0642357445953685, "1000–3000 Hz_40ms_envelope_correlation": 0.9969985576117341, "1000–3000 Hz_40ms_modulation_mae_db": 1.268391464659882, "3000–6000 Hz_10ms_log_rms_mae_db": 1.666611537079932, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14404866564525703, "3000–6000 Hz_10ms_envelope_correlation": 0.9839599785922625, "3000–6000 Hz_10ms_modulation_mae_db": 0.5775901273368399, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9404752975133944, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07874689388897942, "3000–6000 Hz_40ms_envelope_correlation": 0.9950545646687465, "3000–6000 Hz_40ms_modulation_mae_db": 0.5721924017477594, "mr_stft_db": 5.887752266613333, "band_mae_db": 0.27487469348200155, "rms_gain_error_db": -0.08245538187090522, "output_peak": 0.6332886981915955}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8075636789126953, "全帯域_10ms_rms_envelope_nrmse": 0.074166508746826, "全帯域_10ms_envelope_correlation": 0.9947473910174494, "全帯域_10ms_modulation_mae_db": 0.35501841537307516, "全帯域_40ms_log_rms_mae_db": 0.43025356300226963, "全帯域_40ms_rms_envelope_nrmse": 0.04107168582648393, "全帯域_40ms_envelope_correlation": 0.9982401433778504, "全帯域_40ms_modulation_mae_db": 0.27921074398212264, "300–1000 Hz_10ms_log_rms_mae_db": 1.966444029504718, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10673602798363754, "300–1000 Hz_10ms_envelope_correlation": 0.9910082246365943, "300–1000 Hz_10ms_modulation_mae_db": 0.25950522699149325, "300–1000 Hz_40ms_log_rms_mae_db": 1.3778198591163622, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06172303135320652, "300–1000 Hz_40ms_envelope_correlation": 0.996788901428319, "300–1000 Hz_40ms_modulation_mae_db": 0.19268834632701745, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5524400491213668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11032809262055103, "1000–3000 Hz_10ms_envelope_correlation": 0.9911598319723436, "1000–3000 Hz_10ms_modulation_mae_db": 0.30819304015327864, "1000–3000 Hz_40ms_log_rms_mae_db": 1.03093645435567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06398609266037666, "1000–3000 Hz_40ms_envelope_correlation": 0.9970807487244898, "1000–3000 Hz_40ms_modulation_mae_db": 0.47625162022784545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6906860325597346, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15689197701078736, "3000–6000 Hz_10ms_envelope_correlation": 0.980965043405006, "3000–6000 Hz_10ms_modulation_mae_db": 0.4062849499306956, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9592121047298521, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0814217389026978, "3000–6000 Hz_40ms_envelope_correlation": 0.9943713970795851, "3000–6000 Hz_40ms_modulation_mae_db": 0.2987949148892007, "mr_stft_db": 5.8797648286479465, "band_mae_db": 0.24512777273079142, "rms_gain_error_db": -0.023839521563717367, "output_peak": 0.6168012903348811}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8244445787720572, "全帯域_10ms_rms_envelope_nrmse": 0.07104171174611323, "全帯域_10ms_envelope_correlation": 0.9952644400902534, "全帯域_10ms_modulation_mae_db": 0.3038421066134835, "全帯域_40ms_log_rms_mae_db": 0.4380700735552109, "全帯域_40ms_rms_envelope_nrmse": 0.03863193919513562, "全帯域_40ms_envelope_correlation": 0.9985653488565996, "全帯域_40ms_modulation_mae_db": 0.2176414785716153, "300–1000 Hz_10ms_log_rms_mae_db": 1.8952940320615086, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09319623905549998, "300–1000 Hz_10ms_envelope_correlation": 0.9932700812444172, "300–1000 Hz_10ms_modulation_mae_db": 0.16515946459823241, "300–1000 Hz_40ms_log_rms_mae_db": 1.32380498624687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05334065121080113, "300–1000 Hz_40ms_envelope_correlation": 0.997834429573415, "300–1000 Hz_40ms_modulation_mae_db": 0.11826079058312657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.543093039134228, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1139852300298412, "1000–3000 Hz_10ms_envelope_correlation": 0.9904413276827, "1000–3000 Hz_10ms_modulation_mae_db": 0.39693629701562644, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0059786243567996, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.070483603266288, "1000–3000 Hz_40ms_envelope_correlation": 0.9962274035669406, "1000–3000 Hz_40ms_modulation_mae_db": 0.7369860595320926, "3000–6000 Hz_10ms_log_rms_mae_db": 1.653231158601679, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15326386722852386, "3000–6000 Hz_10ms_envelope_correlation": 0.9816722979802359, "3000–6000 Hz_10ms_modulation_mae_db": 0.40066529722370925, "3000–6000 Hz_40ms_log_rms_mae_db": 0.965408708204949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08265703271322714, "3000–6000 Hz_40ms_envelope_correlation": 0.994247362717768, "3000–6000 Hz_40ms_modulation_mae_db": 0.36235841586964, "mr_stft_db": 5.8763425088679, "band_mae_db": 0.2786170292401877, "rms_gain_error_db": -0.10170030053129883, "output_peak": 0.6161734943993465}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8039371461829076, "全帯域_10ms_rms_envelope_nrmse": 0.06951030705270589, "全帯域_10ms_envelope_correlation": 0.9954282717942443, "全帯域_10ms_modulation_mae_db": 0.582127395430611, "全帯域_40ms_log_rms_mae_db": 0.4112382139186835, "全帯域_40ms_rms_envelope_nrmse": 0.036697943851878606, "全帯域_40ms_envelope_correlation": 0.998659718641349, "全帯域_40ms_modulation_mae_db": 0.6178952405723833, "300–1000 Hz_10ms_log_rms_mae_db": 2.0024979358449375, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0970574916778112, "300–1000 Hz_10ms_envelope_correlation": 0.99257133096304, "300–1000 Hz_10ms_modulation_mae_db": 0.318857170831709, "300–1000 Hz_40ms_log_rms_mae_db": 1.4186586904446674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06025805917977237, "300–1000 Hz_40ms_envelope_correlation": 0.997008557737156, "300–1000 Hz_40ms_modulation_mae_db": 0.1659048798878435, "1000–3000 Hz_10ms_log_rms_mae_db": 1.575579463074254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10636541409860904, "1000–3000 Hz_10ms_envelope_correlation": 0.991728769155874, "1000–3000 Hz_10ms_modulation_mae_db": 0.25757285454250417, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0209458352613667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05492910123354614, "1000–3000 Hz_40ms_envelope_correlation": 0.9978073082543922, "1000–3000 Hz_40ms_modulation_mae_db": 0.29786566912693563, "3000–6000 Hz_10ms_log_rms_mae_db": 1.711203684479156, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14811490912436753, "3000–6000 Hz_10ms_envelope_correlation": 0.9828946280971249, "3000–6000 Hz_10ms_modulation_mae_db": 1.1590981616383849, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0005712159991071, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08475641820015606, "3000–6000 Hz_40ms_envelope_correlation": 0.993869044034391, "3000–6000 Hz_40ms_modulation_mae_db": 1.1933125044347048, "mr_stft_db": 5.880716842479647, "band_mae_db": 0.23765430040725385, "rms_gain_error_db": -0.07566391868809862, "output_peak": 0.6281061882349906}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7789838895074463, "全帯域_10ms_rms_envelope_nrmse": 0.0677566770004192, "全帯域_10ms_envelope_correlation": 0.9957750878326459, "全帯域_10ms_modulation_mae_db": 0.8362806616641979, "全帯域_40ms_log_rms_mae_db": 0.404222590758785, "全帯域_40ms_rms_envelope_nrmse": 0.038311877723238505, "全帯域_40ms_envelope_correlation": 0.998659002381311, "全帯域_40ms_modulation_mae_db": 0.8623727237008404, "300–1000 Hz_10ms_log_rms_mae_db": 1.929210755588336, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09168131408237423, "300–1000 Hz_10ms_envelope_correlation": 0.9933778316523566, "300–1000 Hz_10ms_modulation_mae_db": 0.3033107600287635, "300–1000 Hz_40ms_log_rms_mae_db": 1.341791334535692, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054413326176394425, "300–1000 Hz_40ms_envelope_correlation": 0.9975718294876726, "300–1000 Hz_40ms_modulation_mae_db": 0.15536691729022176, "1000–3000 Hz_10ms_log_rms_mae_db": 1.524367469227301, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10631191535695858, "1000–3000 Hz_10ms_envelope_correlation": 0.9918907011917403, "1000–3000 Hz_10ms_modulation_mae_db": 0.36750352001687975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0001390845391382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0700407863649501, "1000–3000 Hz_40ms_envelope_correlation": 0.9965406705912112, "1000–3000 Hz_40ms_modulation_mae_db": 0.24122908905180449, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7235630666334352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16802248033531392, "3000–6000 Hz_10ms_envelope_correlation": 0.978007326849787, "3000–6000 Hz_10ms_modulation_mae_db": 1.4716307040879482, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0126025383697754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09209247911403078, "3000–6000 Hz_40ms_envelope_correlation": 0.9929072957680355, "3000–6000 Hz_40ms_modulation_mae_db": 1.6210765726595948, "mr_stft_db": 5.8745155814134415, "band_mae_db": 0.2681300531000222, "rms_gain_error_db": -0.1171366252112529, "output_peak": 0.6389667208307015}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8176181979483599, "全帯域_10ms_rms_envelope_nrmse": 0.07057119853854377, "全帯域_10ms_envelope_correlation": 0.9952544890267127, "全帯域_10ms_modulation_mae_db": 0.14710433714684976, "全帯域_40ms_log_rms_mae_db": 0.40948440187171803, "全帯域_40ms_rms_envelope_nrmse": 0.03613635529909417, "全帯域_40ms_envelope_correlation": 0.9986640096819085, "全帯域_40ms_modulation_mae_db": 0.1930144615850137, "300–1000 Hz_10ms_log_rms_mae_db": 1.8826925154987246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08977917195844459, "300–1000 Hz_10ms_envelope_correlation": 0.9936385551195329, "300–1000 Hz_10ms_modulation_mae_db": 0.4460234958888319, "300–1000 Hz_40ms_log_rms_mae_db": 1.3008343330221437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047590129946335644, "300–1000 Hz_40ms_envelope_correlation": 0.9981379974387968, "300–1000 Hz_40ms_modulation_mae_db": 0.23963916563753343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5629525959745545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11129646524807377, "1000–3000 Hz_10ms_envelope_correlation": 0.9908746014847544, "1000–3000 Hz_10ms_modulation_mae_db": 0.5209102739557482, "1000–3000 Hz_40ms_log_rms_mae_db": 1.010772745694539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06787427623261685, "1000–3000 Hz_40ms_envelope_correlation": 0.9964466304820733, "1000–3000 Hz_40ms_modulation_mae_db": 0.6714954407105043, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6455356827722443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15400200605065054, "3000–6000 Hz_10ms_envelope_correlation": 0.9814861935617227, "3000–6000 Hz_10ms_modulation_mae_db": 0.3840514836837914, "3000–6000 Hz_40ms_log_rms_mae_db": 0.967781162376758, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08285479019411611, "3000–6000 Hz_40ms_envelope_correlation": 0.9941982889146691, "3000–6000 Hz_40ms_modulation_mae_db": 0.4901073188151761, "mr_stft_db": 5.877580089402443, "band_mae_db": 0.2024200463552512, "rms_gain_error_db": -0.048689244087599955, "output_peak": 0.6194310605043025}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.803138492809872, "全帯域_10ms_rms_envelope_nrmse": 0.07893459343627968, "全帯域_10ms_envelope_correlation": 0.9941065933548696, "全帯域_10ms_modulation_mae_db": 0.9551528771333399, "全帯域_40ms_log_rms_mae_db": 0.4700244828897138, "全帯域_40ms_rms_envelope_nrmse": 0.04332659506914227, "全帯域_40ms_envelope_correlation": 0.9981460596126077, "全帯域_40ms_modulation_mae_db": 0.650652640915946, "300–1000 Hz_10ms_log_rms_mae_db": 1.982285809445746, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10354751040775707, "300–1000 Hz_10ms_envelope_correlation": 0.9915088719724477, "300–1000 Hz_10ms_modulation_mae_db": 0.1730700902638119, "300–1000 Hz_40ms_log_rms_mae_db": 1.3746515646008728, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06268885154173032, "300–1000 Hz_40ms_envelope_correlation": 0.9966875958761849, "300–1000 Hz_40ms_modulation_mae_db": 0.3044359917735006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5372489364785555, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11130111418200106, "1000–3000 Hz_10ms_envelope_correlation": 0.9909329159911324, "1000–3000 Hz_10ms_modulation_mae_db": 0.3718728627369145, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0160563245600365, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06454928993385359, "1000–3000 Hz_40ms_envelope_correlation": 0.996913711167244, "1000–3000 Hz_40ms_modulation_mae_db": 0.14895327663417504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7311689992368198, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17439505651814258, "3000–6000 Hz_10ms_envelope_correlation": 0.9764363368889298, "3000–6000 Hz_10ms_modulation_mae_db": 1.625265475821232, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0712516865442148, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10621948671694033, "3000–6000 Hz_40ms_envelope_correlation": 0.9909149643283385, "3000–6000 Hz_40ms_modulation_mae_db": 1.6873293253473587, "mr_stft_db": 5.884675332172677, "band_mae_db": 0.27506506287427535, "rms_gain_error_db": -0.0950860966315378, "output_peak": 0.6208121153981705}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8113384748590227, "全帯域_10ms_rms_envelope_nrmse": 0.06918273044018224, "全帯域_10ms_envelope_correlation": 0.9954311612467384, "全帯域_10ms_modulation_mae_db": 0.7107333623077489, "全帯域_40ms_log_rms_mae_db": 0.39992707398045346, "全帯域_40ms_rms_envelope_nrmse": 0.03317419598151115, "全帯域_40ms_envelope_correlation": 0.9988526850064668, "全帯域_40ms_modulation_mae_db": 0.7938616016120758, "300–1000 Hz_10ms_log_rms_mae_db": 1.9293593309930435, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08548157295902861, "300–1000 Hz_10ms_envelope_correlation": 0.9943377661045598, "300–1000 Hz_10ms_modulation_mae_db": 0.6922165069477315, "300–1000 Hz_40ms_log_rms_mae_db": 1.2930008998909215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03973397120226679, "300–1000 Hz_40ms_envelope_correlation": 0.998747175306298, "300–1000 Hz_40ms_modulation_mae_db": 0.16747913791300476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5603623079911328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11905196190888119, "1000–3000 Hz_10ms_envelope_correlation": 0.9895929637630488, "1000–3000 Hz_10ms_modulation_mae_db": 0.3705064478703853, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0379377131460708, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06758885908268761, "1000–3000 Hz_40ms_envelope_correlation": 0.9965970989442372, "1000–3000 Hz_40ms_modulation_mae_db": 0.44574650965368123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6684259070093126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15759350205187195, "3000–6000 Hz_10ms_envelope_correlation": 0.9808078251332534, "3000–6000 Hz_10ms_modulation_mae_db": 1.6900727816695105, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9567320244262963, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09404911846294155, "3000–6000 Hz_40ms_envelope_correlation": 0.9929396384881415, "3000–6000 Hz_40ms_modulation_mae_db": 1.667489975369979, "mr_stft_db": 5.890310878520428, "band_mae_db": 0.21118338613052348, "rms_gain_error_db": -0.0165764401275831, "output_peak": 0.6219536013819404}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84000, "output_samples": 84000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.802037463067905, "spectral_convergence": 0.3487926895317824, "spectral_convergence_status": "ok", "amplitude_floor": 1.02822042353332e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.78531806742956, "spectral_convergence": 0.34734753578712463, "spectral_convergence_status": "ok", "amplitude_floor": 8.928413196720056e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.075901269342533, "spectral_convergence": 0.3500819342885999, "spectral_convergence_status": "ok", "amplitude_floor": 7.736392334055093e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 1.4784881078912855e-05, "output_power": 1.882129904784274e-05, "delta_db": 1.0483175951105297, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002602315552138397, "output_power": 0.00026045229235673365, "delta_db": 0.0036822701261140364, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005931268593866254, "output_power": 0.0005928802923471056, "delta_db": -0.0018057683053406676, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005283196568297616, "output_power": 0.0005207448401994275, "delta_db": -0.06271794052682877, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003079318069923152, "output_power": 0.0002925234469522496, "delta_db": -0.22293868212306275, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00026143236338356946, "output_power": 0.000243433732769951, "delta_db": -0.30978590470013323, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27487469348200155, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08245538187090522, "pitch": null, "output_rms_dbfs": -27.147002153716848, "output_peak": 0.6332886981915955, "reference_rms_dbfs": -27.06454677184594, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.887752266613333}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00019312594202620684, "log_rms_mae_db": 0.7638664772489364, "rms_envelope_nrmse": 0.06607419962465326, "envelope_correlation": 0.9958856099961639, "rms_gain_error_db": -0.08245508628136813, "reference_mean_power": 0.0019676998400409485, "output_mean_power": 0.0019306935321317098}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 4.0378624939923077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018862042103601667, "output_power": 0.0001791140322513099, "reference_active": true, "delta_db": -0.22459099022931378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.862312285918816e-05, "output_power": 9.215539329197974e-05, "reference_active": true, "delta_db": -0.2945799326189592}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3789377405749525e-05, "output_power": 1.1691035945813231e-05, "reference_active": true, "delta_db": -0.7169166228425075}], "active_band_count": 3, "active_band_mae_db": 0.41202918189692683}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00018244538675777041, "log_rms_mae_db": 0.38121015156747645, "rms_envelope_nrmse": 0.0337294678183274, "envelope_correlation": 0.9988940891566326, "rms_gain_error_db": -0.08245466149908728, "reference_mean_power": 0.001979010526094432, "output_mean_power": 0.0019417916893204486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 3.3102005537920435e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017836230285954771, "output_power": 0.00016840056394726223, "reference_active": true, "delta_db": -0.2495952939725989}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3320761072042273e-05, "output_power": 4.077651844742427e-05, "reference_active": true, "delta_db": -0.26285935442110947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.132775238046075e-07, "output_power": 7.201713749831702e-07, "reference_active": true, "delta_db": -0.5280291480393359}], "active_band_count": 3, "active_band_mae_db": 0.3468279321443481}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00014606965842361196, "log_rms_mae_db": 2.0258038992542264, "rms_envelope_nrmse": 0.08765849910017846, "envelope_correlation": 0.9939443532911625, "rms_gain_error_db": -0.0035766251131045196, "reference_mean_power": 0.0007714288246159527, "output_mean_power": 0.0007707937771993105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 2.155891354220712e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.355627855804233e-05, "output_power": 8.013537346478132e-05, "reference_active": true, "delta_db": -0.18154823801615622}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.117824128471733e-05, "output_power": 2.0511612443770917e-05, "reference_active": true, "delta_db": -0.13890089786711313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3214088476749303e-06, "output_power": 3.5647086374095267e-06, "reference_active": true, "delta_db": 0.3070170022316115}], "active_band_count": 3, "active_band_mae_db": 0.20915537937162695}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.0001346093108760191, "log_rms_mae_db": 1.4542331645604707, "rms_envelope_nrmse": 0.04927397833875958, "envelope_correlation": 0.9979714024142325, "rms_gain_error_db": -0.0035778090283028513, "reference_mean_power": 0.0007758659430163304, "output_mean_power": 0.0007752270315903664}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.945223621465206e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.346976557991944e-05, "output_power": 6.968168037639713e-05, "reference_active": true, "delta_db": -0.22990038988605693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.57892144723787e-06, "output_power": 9.390032298363516e-06, "reference_active": true, "delta_db": -0.08649525717909068}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.32297115664878e-07, "output_power": 2.4279952902972743e-07, "reference_active": true, "delta_db": 0.19204022598865614}], "active_band_count": 3, "active_band_mae_db": 0.16947862435126793}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00012444218003818797, "log_rms_mae_db": 1.5532772751458566, "rms_envelope_nrmse": 0.10884793478348868, "envelope_correlation": 0.9913596210982843, "rms_gain_error_db": -0.14454835770400812, "reference_mean_power": 0.0006875503513213645, "output_mean_power": 0.0006650429172410156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.3390858183888e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.254326674690693e-05, "output_power": 4.4452291452481704e-05, "reference_active": true, "delta_db": 0.19063318363361198}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4859458498211302e-05, "output_power": 1.865675497322824e-05, "reference_active": true, "delta_db": 0.9883312445784924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.141405389451205e-06, "output_power": 4.620266325360832e-06, "reference_active": true, "delta_db": 1.6754302576779303}], "active_band_count": 3, "active_band_mae_db": 0.9514648952966782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00011369765125964215, "log_rms_mae_db": 1.0419016018682763, "rms_envelope_nrmse": 0.0642357445953685, "envelope_correlation": 0.9969985576117341, "rms_gain_error_db": -0.1445449393824716, "reference_mean_power": 0.0006915015836222027, "output_mean_power": 0.0006688653296944444}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.214722623247201e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7827908938064025e-05, "output_power": 4.1126967560049055e-05, "reference_active": true, "delta_db": 0.3631435368592496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.619804659862373e-06, "output_power": 7.749992427509396e-06, "reference_active": true, "delta_db": 1.3958005807350564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0150238495180175e-07, "output_power": 3.2277747387611913e-07, "reference_active": true, "delta_db": 2.04623027638534}], "active_band_count": 3, "active_band_mae_db": 1.268391464659882}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00010848831624356859, "log_rms_mae_db": 1.666611537079932, "rms_envelope_nrmse": 0.14404866564525703, "envelope_correlation": 0.9839599785922625, "rms_gain_error_db": -0.2024889109247829, "reference_mean_power": 0.00023818463717296146, "output_mean_power": 0.00022733424162074253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.7158903163900278e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.877460217795428e-05, "output_power": 8.353468520758509e-05, "reference_active": true, "delta_db": -0.26421894440059573}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.954888518804953e-05, "output_power": 3.457570225222862e-05, "reference_active": true, "delta_db": -0.5836323638630397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.296333400161532e-06, "output_power": 5.95131260583614e-06, "reference_active": true, "delta_db": -0.8849190737468844}], "active_band_count": 3, "active_band_mae_db": 0.5775901273368399}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 7.397564116979367e-05, "log_rms_mae_db": 0.9404752975133944, "rms_envelope_nrmse": 0.07874689388897942, "envelope_correlation": 0.9950545646687465, "rms_gain_error_db": -0.2024951404540766, "reference_mean_power": 0.0002395532500233675, "output_mean_power": 0.00022864017995729826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.5691901914560294e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.470858674877874e-05, "output_power": 8.814845722413913e-05, "reference_active": true, "delta_db": -0.3117464040140285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.833295648941501e-05, "output_power": 1.6156199476774446e-05, "reference_active": true, "delta_db": -0.5489330104409517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.840671065644946e-07, "output_power": 3.974804687283799e-07, "reference_active": true, "delta_db": -0.8558977907882982}], "active_band_count": 3, "active_band_mae_db": 0.5721924017477594}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8860865733529408, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8ee58e30d4cdd88c35daca28d47f1db15f33698c01c31ae45a6b07bedd0ebb80", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9909975915424847, "standard_deviation": 0.0007359513829029333, "minimum": 0.9895929637630488, "maximum": 0.9918907011917403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.551165142018406, "standard_deviation": 0.016047306798537542, "minimum": 1.524367469227301, "maximum": 1.575579463074254, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4431200239485019, "standard_deviation": 0.21886021499860595, "minimum": 0.25757285454250417, "maximum": 0.9514648952966782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11093601602855056, "standard_deviation": 0.004182136621369614, "minimum": 0.10631191535695858, "maximum": 0.11905196190888119, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968265161677903, "standard_deviation": 0.0004930050036276296, "minimum": 0.9962274035669406, "maximum": 0.9978073082543922, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0205835479727372, "standard_deviation": 0.015170308714700695, "minimum": 1.0001390845391382, "maximum": 1.0419016018682763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5358648911996151, "standard_deviation": 0.3586223837739592, "minimum": 0.14895327663417504, "maximum": 1.268391464659882, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06546096917121094, "standard_deviation": 0.004956689311292864, "minimum": 0.05492910123354614, "maximum": 0.070483603266288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9807787038135403, "standard_deviation": 0.0024631265946213144, "minimum": 0.9764363368889298, "maximum": 0.9839599785922625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6863032585465392, "standard_deviation": 0.032740517771488356, "minimum": 1.6455356827722443, "maximum": 1.7311689992368198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9643323726740138, "standard_deviation": 0.5823903931365134, "minimum": 0.3840514836837914, "maximum": 1.6900727816695105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15704155799561437, "standard_deviation": 0.009957493361932883, "minimum": 0.14404866564525703, "maximum": 0.17439505651814258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935628194999594, "standard_deviation": 0.0012914277014731627, "minimum": 0.9909149643283385, "maximum": 0.9950545646687465, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9842543422705433, "standard_deviation": 0.042369325967507, "minimum": 0.9404752975133944, "maximum": 1.0712516865442148, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9865826786416767, "standard_deviation": 0.618860168226986, "minimum": 0.2987949148892007, "maximum": 1.6873293253473587, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08784974477413614, "standard_deviation": 0.009099011725578475, "minimum": 0.07874689388897942, "maximum": 0.10621948671694033, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992957126873014, "standard_deviation": 0.0011750111708744427, "minimum": 0.9910082246365943, "maximum": 0.9943377661045598, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.951698538523905, "standard_deviation": 0.050952686794915006, "minimum": 1.8826925154987246, "maximum": 2.0258038992542264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.32091226186527505, "standard_deviation": 0.17574753232904497, "minimum": 0.16515946459823241, "maximum": 0.6922165069477315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09439222840309147, "standard_deviation": 0.007541265832511614, "minimum": 0.08548157295902861, "maximum": 0.10673602798363754, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975934861577593, "standard_deviation": 0.0007210081900441983, "minimum": 0.9966875958761849, "maximum": 0.998747175306298, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.36059935405225, "standard_deviation": 0.056670933936500335, "minimum": 1.2930008998909215, "maximum": 1.4542331645604707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.1891567317204395, "standard_deviation": 0.05783725549188628, "minimum": 0.11826079058312657, "maximum": 0.3044359917735006, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05362774986865835, "standard_deviation": 0.007934940540520257, "minimum": 0.03973397120226679, "maximum": 0.06268885154173032, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24913404304003833, "standard_deviation": 0.030084116091699274, "minimum": 0.2024200463552512, "maximum": 0.2786170292401877, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.881457291014726, "standard_deviation": 0.005620907809736049, "minimum": 5.8745155814134415, "maximum": 5.890310878520428, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.624441646159491, "standard_deviation": 0.008227863740381496, "minimum": 0.6161734943993465, "maximum": 0.6389667208307015, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07014344108899923, "standard_deviation": 0.03680479293407531, "minimum": -0.1171366252112529, "maximum": -0.0165764401275831, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952366305448848, "standard_deviation": 0.0005735990141895782, "minimum": 0.9941065933548696, "maximum": 0.9958856099961639, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8013613670301623, "standard_deviation": 0.020183237894732278, "minimum": 0.7638664772489364, "maximum": 0.8244445787720572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5377860421957791, "standard_deviation": 0.28075199958079466, "minimum": 0.14710433714684976, "maximum": 0.9551528771333399, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0709047408232154, "standard_deviation": 0.004024955614084094, "minimum": 0.06607419962465326, "maximum": 0.07893459343627968, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985851320893406, "standard_deviation": 0.0002663171869502677, "minimum": 0.9981460596126077, "maximum": 0.9988940891566326, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41805381894303884, "standard_deviation": 0.02738002159386374, "minimum": 0.38121015156747645, "maximum": 0.4700244828897138, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4951846028855431, "standard_deviation": 0.2673292957911263, "minimum": 0.1930144615850137, "maximum": 0.8623727237008404, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037635007595601455, "standard_deviation": 0.003462721460464394, "minimum": 0.03317419598151115, "maximum": 0.04332659506914227, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0047", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.436393976211548, "predicted_mos": 2.436393976211548, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "f7e45ec730e8932a44fe03d98fd4b8bf6ace02f605b904c93274de29a91a3634", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.132327600382268, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8096773863509275, "全帯域_10ms_rms_envelope_nrmse": 0.073123215606826, "全帯域_10ms_envelope_correlation": 0.9950046285868842, "全帯域_10ms_modulation_mae_db": 0.7998161522810806, "全帯域_40ms_log_rms_mae_db": 0.4346556059263219, "全帯域_40ms_rms_envelope_nrmse": 0.038130044741919014, "全帯域_40ms_envelope_correlation": 0.9986245197847019, "全帯域_40ms_modulation_mae_db": 0.8647427445128985, "300–1000 Hz_10ms_log_rms_mae_db": 1.3025212392337588, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07690413240605083, "300–1000 Hz_10ms_envelope_correlation": 0.9953230181287948, "300–1000 Hz_10ms_modulation_mae_db": 0.16991947982885328, "300–1000 Hz_40ms_log_rms_mae_db": 0.7945192904635288, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04510647398307644, "300–1000 Hz_40ms_envelope_correlation": 0.9982850818635507, "300–1000 Hz_40ms_modulation_mae_db": 0.2854889680329937, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2280829170219838, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10857923318429331, "1000–3000 Hz_10ms_envelope_correlation": 0.991444385966771, "1000–3000 Hz_10ms_modulation_mae_db": 0.49790323420368865, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7230480329703678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05784364427211117, "1000–3000 Hz_40ms_envelope_correlation": 0.9976489657107198, "1000–3000 Hz_40ms_modulation_mae_db": 0.24039842896226982, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3451008808679186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14690415311456562, "3000–6000 Hz_10ms_envelope_correlation": 0.9836903526877446, "3000–6000 Hz_10ms_modulation_mae_db": 1.4508317701743314, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8717940802132671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0904569563996515, "3000–6000 Hz_40ms_envelope_correlation": 0.9938926014392313, "3000–6000 Hz_40ms_modulation_mae_db": 1.6035091909002424, "mr_stft_db": 5.5445930398475936, "band_mae_db": 0.22988145871584595, "rms_gain_error_db": -0.10418334650869163, "output_peak": 0.6194643852554915}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8903359894010993, "全帯域_10ms_rms_envelope_nrmse": 0.07703631445020473, "全帯域_10ms_envelope_correlation": 0.9943582138156063, "全帯域_10ms_modulation_mae_db": 0.8827693854527343, "全帯域_40ms_log_rms_mae_db": 0.4775424260603735, "全帯域_40ms_rms_envelope_nrmse": 0.04284724078179802, "全帯域_40ms_envelope_correlation": 0.9981258778063287, "全帯域_40ms_modulation_mae_db": 0.8847400892032321, "300–1000 Hz_10ms_log_rms_mae_db": 1.3551577020715009, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08657912552419107, "300–1000 Hz_10ms_envelope_correlation": 0.9940659542092324, "300–1000 Hz_10ms_modulation_mae_db": 0.48094769115027103, "300–1000 Hz_40ms_log_rms_mae_db": 0.7679694118695908, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04292109419696337, "300–1000 Hz_40ms_envelope_correlation": 0.9984493743332838, "300–1000 Hz_40ms_modulation_mae_db": 0.2916134152969328, "1000–3000 Hz_10ms_log_rms_mae_db": 1.209830023707936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10496979642980128, "1000–3000 Hz_10ms_envelope_correlation": 0.9919045713646829, "1000–3000 Hz_10ms_modulation_mae_db": 0.2761836277975628, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6999104686274438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05872128419324779, "1000–3000 Hz_40ms_envelope_correlation": 0.9973770508628681, "1000–3000 Hz_40ms_modulation_mae_db": 0.1594854878824559, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3153991089122694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1332776447651046, "3000–6000 Hz_10ms_envelope_correlation": 0.9863337560867415, "3000–6000 Hz_10ms_modulation_mae_db": 1.1842846409419885, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8378841647518469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07444195310383732, "3000–6000 Hz_40ms_envelope_correlation": 0.9956140657843109, "3000–6000 Hz_40ms_modulation_mae_db": 1.3215698675155896, "mr_stft_db": 5.547793995899922, "band_mae_db": 0.19781414861038737, "rms_gain_error_db": -0.06523548976626142, "output_peak": 0.6484467276404237}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8256061061698791, "全帯域_10ms_rms_envelope_nrmse": 0.07154542415992272, "全帯域_10ms_envelope_correlation": 0.9952261384571153, "全帯域_10ms_modulation_mae_db": 0.5013308642927604, "全帯域_40ms_log_rms_mae_db": 0.4441378493772722, "全帯域_40ms_rms_envelope_nrmse": 0.040819637497716284, "全帯域_40ms_envelope_correlation": 0.9984160347042517, "全帯域_40ms_modulation_mae_db": 0.38308423244248946, "300–1000 Hz_10ms_log_rms_mae_db": 1.3639438921062734, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08668780151148417, "300–1000 Hz_10ms_envelope_correlation": 0.9941023296874977, "300–1000 Hz_10ms_modulation_mae_db": 0.07619954036251081, "300–1000 Hz_40ms_log_rms_mae_db": 0.8463899624286263, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053684991911374484, "300–1000 Hz_40ms_envelope_correlation": 0.9976429566696857, "300–1000 Hz_40ms_modulation_mae_db": 0.39935345706053244, "1000–3000 Hz_10ms_log_rms_mae_db": 1.162950279605718, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10116049965114264, "1000–3000 Hz_10ms_envelope_correlation": 0.9930178159943611, "1000–3000 Hz_10ms_modulation_mae_db": 0.48843699752878905, "1000–3000 Hz_40ms_log_rms_mae_db": 0.687052745828322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061113333336748, "1000–3000 Hz_40ms_envelope_correlation": 0.9978388806122217, "1000–3000 Hz_40ms_modulation_mae_db": 0.5361853202440785, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2989647331858147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14284702717996212, "3000–6000 Hz_10ms_envelope_correlation": 0.9841432382233266, "3000–6000 Hz_10ms_modulation_mae_db": 0.5884927024804615, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7939257116350344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07642212652322307, "3000–6000 Hz_40ms_envelope_correlation": 0.9950072905390569, "3000–6000 Hz_40ms_modulation_mae_db": 0.5422709566027369, "mr_stft_db": 5.520862259117209, "band_mae_db": 0.25677774676935, "rms_gain_error_db": -0.11283725122634575, "output_peak": 0.6253829785144485}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8535077034218852, "全帯域_10ms_rms_envelope_nrmse": 0.08260502786911499, "全帯域_10ms_envelope_correlation": 0.9938009512204012, "全帯域_10ms_modulation_mae_db": 0.878641832127944, "全帯域_40ms_log_rms_mae_db": 0.442415054323443, "全帯域_40ms_rms_envelope_nrmse": 0.04901782975434211, "全帯域_40ms_envelope_correlation": 0.9979242955835198, "全帯域_40ms_modulation_mae_db": 0.7298615740381695, "300–1000 Hz_10ms_log_rms_mae_db": 1.3108851817836853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08988703970785616, "300–1000 Hz_10ms_envelope_correlation": 0.9937776113810787, "300–1000 Hz_10ms_modulation_mae_db": 0.2271126508300035, "300–1000 Hz_40ms_log_rms_mae_db": 0.8196657321621345, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05318288778416495, "300–1000 Hz_40ms_envelope_correlation": 0.9978690073644794, "300–1000 Hz_40ms_modulation_mae_db": 0.3279299348350388, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2094750050146705, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10256630588510238, "1000–3000 Hz_10ms_envelope_correlation": 0.9926771967143356, "1000–3000 Hz_10ms_modulation_mae_db": 0.5008730216474072, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7252693062721878, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060700335814707254, "1000–3000 Hz_40ms_envelope_correlation": 0.9977303274779519, "1000–3000 Hz_40ms_modulation_mae_db": 0.33404839360758265, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3749066155437781, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1624618236530143, "3000–6000 Hz_10ms_envelope_correlation": 0.9799430263128687, "3000–6000 Hz_10ms_modulation_mae_db": 1.154894275653702, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8799724395388295, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10295516584686316, "3000–6000 Hz_40ms_envelope_correlation": 0.9919997833416706, "3000–6000 Hz_40ms_modulation_mae_db": 1.1277744670294902, "mr_stft_db": 5.55115089594972, "band_mae_db": 0.3227914592067821, "rms_gain_error_db": -0.18795545469343006, "output_peak": 0.6258646981379025}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8213754395141062, "全帯域_10ms_rms_envelope_nrmse": 0.07302397004187228, "全帯域_10ms_envelope_correlation": 0.9950325867169292, "全帯域_10ms_modulation_mae_db": 0.6415349837618495, "全帯域_40ms_log_rms_mae_db": 0.43625019481262384, "全帯域_40ms_rms_envelope_nrmse": 0.03866752635042271, "全帯域_40ms_envelope_correlation": 0.9985979959564407, "全帯域_40ms_modulation_mae_db": 0.5662240169159564, "300–1000 Hz_10ms_log_rms_mae_db": 1.3110426750119732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07762352070010996, "300–1000 Hz_10ms_envelope_correlation": 0.995261482955497, "300–1000 Hz_10ms_modulation_mae_db": 0.21432921660832938, "300–1000 Hz_40ms_log_rms_mae_db": 0.809287039042275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04714255886983865, "300–1000 Hz_40ms_envelope_correlation": 0.9981741125095376, "300–1000 Hz_40ms_modulation_mae_db": 0.250667924959871, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2165755310724438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10336620766318966, "1000–3000 Hz_10ms_envelope_correlation": 0.9923765567272115, "1000–3000 Hz_10ms_modulation_mae_db": 0.19126750153125208, "1000–3000 Hz_40ms_log_rms_mae_db": 0.694212327521062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05716391895844984, "1000–3000 Hz_40ms_envelope_correlation": 0.9978510440020643, "1000–3000 Hz_40ms_modulation_mae_db": 0.5335504188271406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3320982791678553, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14726644872500905, "3000–6000 Hz_10ms_envelope_correlation": 0.983095009222425, "3000–6000 Hz_10ms_modulation_mae_db": 1.1209888594470452, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7637096531873786, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.077563490540904, "3000–6000 Hz_40ms_envelope_correlation": 0.9949613417307015, "3000–6000 Hz_40ms_modulation_mae_db": 1.3256797187474214, "mr_stft_db": 5.528757852088579, "band_mae_db": 0.19837533592445342, "rms_gain_error_db": -0.10295487841623294, "output_peak": 0.6297328074418965}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8763508155144727, "全帯域_10ms_rms_envelope_nrmse": 0.07917456251776686, "全帯域_10ms_envelope_correlation": 0.9940266732574919, "全帯域_10ms_modulation_mae_db": 0.16536985245105038, "全帯域_40ms_log_rms_mae_db": 0.44330563575388426, "全帯域_40ms_rms_envelope_nrmse": 0.04193886598168097, "全帯域_40ms_envelope_correlation": 0.9981655253037409, "全帯域_40ms_modulation_mae_db": 0.20671280518198532, "300–1000 Hz_10ms_log_rms_mae_db": 1.3305536508008908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08670767686652234, "300–1000 Hz_10ms_envelope_correlation": 0.9940918579631534, "300–1000 Hz_10ms_modulation_mae_db": 0.4713864772786971, "300–1000 Hz_40ms_log_rms_mae_db": 0.8035922771517352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046128732513056665, "300–1000 Hz_40ms_envelope_correlation": 0.9982963782322882, "300–1000 Hz_40ms_modulation_mae_db": 0.32939523550364386, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2361228091241232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12077167768245842, "1000–3000 Hz_10ms_envelope_correlation": 0.9894180914524537, "1000–3000 Hz_10ms_modulation_mae_db": 0.938269726086844, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7340933396409951, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06760043926103566, "1000–3000 Hz_40ms_envelope_correlation": 0.996540698901025, "1000–3000 Hz_40ms_modulation_mae_db": 0.9685248390565055, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2984665917881737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14313468854960365, "3000–6000 Hz_10ms_envelope_correlation": 0.984260449677012, "3000–6000 Hz_10ms_modulation_mae_db": 0.812059149352693, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8159646274885193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07928224793777633, "3000–6000 Hz_40ms_envelope_correlation": 0.9947139418894724, "3000–6000 Hz_40ms_modulation_mae_db": 1.0502276402557746, "mr_stft_db": 5.546317353459073, "band_mae_db": 0.17269937305811986, "rms_gain_error_db": 0.006292648872901354, "output_peak": 0.6448370849968289}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8294937239794212, "全帯域_10ms_rms_envelope_nrmse": 0.07320790333387893, "全帯域_10ms_envelope_correlation": 0.9952910371429158, "全帯域_10ms_modulation_mae_db": 0.657082880365572, "全帯域_40ms_log_rms_mae_db": 0.42144955489557895, "全帯域_40ms_rms_envelope_nrmse": 0.04423138164957888, "全帯域_40ms_envelope_correlation": 0.9984564182858002, "全帯域_40ms_modulation_mae_db": 0.8162761312170533, "300–1000 Hz_10ms_log_rms_mae_db": 1.3140636685354061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0779485784634434, "300–1000 Hz_10ms_envelope_correlation": 0.995461531402646, "300–1000 Hz_10ms_modulation_mae_db": 0.1301937574290242, "300–1000 Hz_40ms_log_rms_mae_db": 0.7692040873169772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0510841263400919, "300–1000 Hz_40ms_envelope_correlation": 0.998139413059258, "300–1000 Hz_40ms_modulation_mae_db": 0.1684156060642087, "1000–3000 Hz_10ms_log_rms_mae_db": 1.148178243171917, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10032337388112327, "1000–3000 Hz_10ms_envelope_correlation": 0.9932477561823821, "1000–3000 Hz_10ms_modulation_mae_db": 0.6290877208386322, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6864832206574618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061598811530091724, "1000–3000 Hz_40ms_envelope_correlation": 0.9979090794317363, "1000–3000 Hz_40ms_modulation_mae_db": 0.6813162414788637, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3243916831108147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14738598419392693, "3000–6000 Hz_10ms_envelope_correlation": 0.984009269671472, "3000–6000 Hz_10ms_modulation_mae_db": 1.858341103563258, "3000–6000 Hz_40ms_log_rms_mae_db": 0.86282120199665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09627333601398202, "3000–6000 Hz_40ms_envelope_correlation": 0.9935136227460368, "3000–6000 Hz_40ms_modulation_mae_db": 2.003563300765881, "mr_stft_db": 5.540308078572612, "band_mae_db": 0.2823248023202402, "rms_gain_error_db": -0.19400442661352088, "output_peak": 0.6085902408214107}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7898679512875456, "全帯域_10ms_rms_envelope_nrmse": 0.07223951944546327, "全帯域_10ms_envelope_correlation": 0.9951158369208631, "全帯域_10ms_modulation_mae_db": 0.5194342013036638, "全帯域_40ms_log_rms_mae_db": 0.39546417852923843, "全帯域_40ms_rms_envelope_nrmse": 0.037556171434395394, "全帯域_40ms_envelope_correlation": 0.9986728259220838, "全帯域_40ms_modulation_mae_db": 0.3025092797922857, "300–1000 Hz_10ms_log_rms_mae_db": 1.2969169491307007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08603065208552353, "300–1000 Hz_10ms_envelope_correlation": 0.9941449413138956, "300–1000 Hz_10ms_modulation_mae_db": 0.22843872614364347, "300–1000 Hz_40ms_log_rms_mae_db": 0.7773888022851123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04458817223556794, "300–1000 Hz_40ms_envelope_correlation": 0.9983276210752058, "300–1000 Hz_40ms_modulation_mae_db": 0.08872686915148938, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2265864668946265, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10852445299209061, "1000–3000 Hz_10ms_envelope_correlation": 0.9916962599598258, "1000–3000 Hz_10ms_modulation_mae_db": 0.33668893130808225, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7268046389540309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06203285918952731, "1000–3000 Hz_40ms_envelope_correlation": 0.9975672439215395, "1000–3000 Hz_40ms_modulation_mae_db": 0.18892222074991358, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3220711279330666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13961196461929412, "3000–6000 Hz_10ms_envelope_correlation": 0.9848433929078042, "3000–6000 Hz_10ms_modulation_mae_db": 0.8587375654548035, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8167923139268539, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07693381491090005, "3000–6000 Hz_40ms_envelope_correlation": 0.9951002875336238, "3000–6000 Hz_40ms_modulation_mae_db": 0.9453844988646191, "mr_stft_db": 5.542434519715827, "band_mae_db": 0.21288334464555123, "rms_gain_error_db": -0.10863064016451102, "output_peak": 0.6435914457995832}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84000, "output_samples": 84000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.356344110947417, "spectral_convergence": 0.33475147504881403, "spectral_convergence_status": "ok", "amplitude_floor": 1.02822042353332e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.35761608735974, "spectral_convergence": 0.34189436147310315, "spectral_convergence_status": "ok", "amplitude_floor": 8.928413196720056e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.919818921235625, "spectral_convergence": 0.3524095773005507, "spectral_convergence_status": "ok", "amplitude_floor": 7.736392334055093e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 1.4784881078912855e-05, "output_power": 1.7522291316073416e-05, "delta_db": 0.7377306074082408, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002602315552138397, "output_power": 0.0002561465030549166, "delta_db": -0.06871525983694887, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005931268593866254, "output_power": 0.0005916361296231559, "delta_db": -0.010929039106585823, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005283196568297616, "output_power": 0.000512388550224651, "delta_db": -0.1329735299988482, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003079318069923152, "output_power": 0.00029034155495183346, "delta_db": -0.2554535210304803, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00026143236338356946, "output_power": 0.00025119481429248017, "delta_db": -0.17348679491397156, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22988145871584595, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10418334650869163, "pitch": null, "output_rms_dbfs": -27.16873011835463, "output_peak": 0.6194643852554915, "reference_rms_dbfs": -27.06454677184594, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5445930398475936}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00019312594202620684, "log_rms_mae_db": 0.8096773863509275, "rms_envelope_nrmse": 0.073123215606826, "envelope_correlation": 0.9950046285868842, "rms_gain_error_db": -0.10418450646415862, "reference_mean_power": 0.0019676998400409485, "output_mean_power": 0.0019210576572911235}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 4.0378624939923077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018862042103601667, "output_power": 0.0001783225998384171, "reference_active": true, "delta_db": -0.2438232266911153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.862312285918816e-05, "output_power": 8.008095245761836e-05, "reference_active": true, "delta_db": -0.9044952029895719}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3789377405749525e-05, "output_power": 1.0337884016317701e-05, "reference_active": true, "delta_db": -1.2511300271625547}], "active_band_count": 3, "active_band_mae_db": 0.7998161522810806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00018244538675777041, "log_rms_mae_db": 0.4346556059263219, "rms_envelope_nrmse": 0.038130044741919014, "envelope_correlation": 0.9986245197847019, "rms_gain_error_db": -0.10418487107170014, "reference_mean_power": 0.001979010526094432, "output_mean_power": 0.0019321000736365489}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 3.3102005537920435e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017836230285954771, "output_power": 0.00016388012146654929, "reference_active": true, "delta_db": -0.36776793771316924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3320761072042273e-05, "output_power": 3.6176021711654014e-05, "reference_active": true, "delta_db": -0.7827527218463048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.132775238046075e-07, "output_power": 5.832678152009401e-07, "reference_active": true, "delta_db": -1.4437075739792213}], "active_band_count": 3, "active_band_mae_db": 0.8647427445128985}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00014606965842361196, "log_rms_mae_db": 1.3025212392337588, "rms_envelope_nrmse": 0.07690413240605083, "envelope_correlation": 0.9953230181287948, "rms_gain_error_db": -0.025827509412635483, "reference_mean_power": 0.0007714288246159527, "output_mean_power": 0.0007668547489555663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 2.155891354220712e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.355627855804233e-05, "output_power": 8.22180777824014e-05, "reference_active": true, "delta_db": -0.07011770091872054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.117824128471733e-05, "output_power": 1.9251075968581082e-05, "reference_active": true, "delta_db": -0.4143488410823636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3214088476749303e-06, "output_power": 3.34080808465298e-06, "reference_active": true, "delta_db": 0.02529189748547572}], "active_band_count": 3, "active_band_mae_db": 0.16991947982885328}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.0001346093108760191, "log_rms_mae_db": 0.7945192904635288, "rms_envelope_nrmse": 0.04510647398307644, "envelope_correlation": 0.9982850818635507, "rms_gain_error_db": -0.025827897364387575, "reference_mean_power": 0.0007758659430163304, "output_mean_power": 0.0007712654892075079}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.945223621465206e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.346976557991944e-05, "output_power": 7.228381312683389e-05, "reference_active": true, "delta_db": -0.07067599680061601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.57892144723787e-06, "output_power": 9.004200019572796e-06, "reference_active": true, "delta_db": -0.26871477938325194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.32297115664878e-07, "output_power": 2.0622257462702908e-07, "reference_active": true, "delta_db": -0.5170761279151131}], "active_band_count": 3, "active_band_mae_db": 0.2854889680329937}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00012444218003818797, "log_rms_mae_db": 1.2280829170219838, "rms_envelope_nrmse": 0.10857923318429331, "envelope_correlation": 0.991444385966771, "rms_gain_error_db": -0.16768622514966328, "reference_mean_power": 0.0006875503513213645, "output_mean_power": 0.0006615091958112675}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.3390858183888e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.254326674690693e-05, "output_power": 3.9062767803861675e-05, "reference_active": true, "delta_db": -0.37067822678679613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4859458498211302e-05, "output_power": 1.3802525459914264e-05, "reference_active": true, "delta_db": -0.3204442644078237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.141405389451205e-06, "output_power": 3.7790505881575162e-06, "reference_active": true, "delta_db": 0.8025872114164461}], "active_band_count": 3, "active_band_mae_db": 0.49790323420368865}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00011369765125964215, "log_rms_mae_db": 0.7230480329703678, "rms_envelope_nrmse": 0.05784364427211117, "envelope_correlation": 0.9976489657107198, "rms_gain_error_db": -0.16768963889609756, "reference_mean_power": 0.0006915015836222027, "output_mean_power": 0.0006653102511566209}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.214722623247201e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7827908938064025e-05, "output_power": 3.5129662974896677e-05, "reference_active": true, "delta_db": -0.321383516014824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.619804659862373e-06, "output_power": 5.86389741851276e-06, "reference_active": true, "delta_db": 0.18465144180098186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0150238495180175e-07, "output_power": 2.1173674071609454e-07, "reference_active": true, "delta_db": 0.21516032907100352}], "active_band_count": 3, "active_band_mae_db": 0.24039842896226982}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00010848831624356859, "log_rms_mae_db": 1.3451008808679186, "rms_envelope_nrmse": 0.14690415311456562, "envelope_correlation": 0.9836903526877446, "rms_gain_error_db": -0.3009704561153801, "reference_mean_power": 0.00023818463717296146, "output_mean_power": 0.0002222371712964802}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.7158903163900278e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.877460217795428e-05, "output_power": 8.373558082491053e-05, "reference_active": true, "delta_db": -0.25378697677563206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.954888518804953e-05, "output_power": 2.8017453250058047e-05, "reference_active": true, "delta_db": -1.4970559008367847}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.296333400161532e-06, "output_power": 4.008108110533957e-06, "reference_active": true, "delta_db": -2.6016524329105777}], "active_band_count": 3, "active_band_mae_db": 1.4508317701743314}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 7.397564116979367e-05, "log_rms_mae_db": 0.8717940802132671, "rms_envelope_nrmse": 0.0904569563996515, "envelope_correlation": 0.9938926014392313, "rms_gain_error_db": -0.3009650265210816, "reference_mean_power": 0.0002395532500233675, "output_mean_power": 0.00022351442918501507}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.5691901914560294e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.470858674877874e-05, "output_power": 8.48956474231214e-05, "reference_active": true, "delta_db": -0.4750393141942371}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.833295648941501e-05, "output_power": 1.2289952755811024e-05, "reference_active": true, "delta_db": -1.7368229430165727}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.840671065644946e-07, "output_power": 2.660964255920745e-07, "reference_active": true, "delta_db": -2.5986653154899173}], "active_band_count": 3, "active_band_mae_db": 1.6035091909002424}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9086397981400414, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1c87b9a191b257ae4906293be791b7dff8a47b861b277ba922a4397072568f16", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.991972829295253, "standard_deviation": 0.0012108823588282776, "minimum": 0.9894180914524537, "maximum": 0.9932477561823821, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2047251594516772, "standard_deviation": 0.03193981016397189, "minimum": 1.148178243171917, "maximum": 1.2361228091241232, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.48233884511778224, "standard_deviation": 0.23252267237995852, "minimum": 0.19126750153125208, "maximum": 0.938269726086844, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1062826934211502, "standard_deviation": 0.006611912714228766, "minimum": 0.10032337388112327, "maximum": 0.12077167768245842, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975579113650159, "standard_deviation": 0.00044621047187570913, "minimum": 0.996540698901025, "maximum": 0.9979090794317363, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7096092600589838, "standard_deviation": 0.01962407774915642, "minimum": 0.6864832206574618, "maximum": 0.7340933396409951, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4553039188511012, "standard_deviation": 0.27941178621636403, "minimum": 0.1594854878824559, "maximum": 0.9685248390565055, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06084682831948984, "standard_deviation": 0.0032683933868476365, "minimum": 0.05716391895844984, "maximum": 0.06760043926103566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9837898118486743, "standard_deviation": 0.0018231314900828079, "minimum": 0.9799430263128687, "maximum": 0.9863337560867415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3264248775637113, "standard_deviation": 0.0251050403743802, "minimum": 1.2984665917881737, "maximum": 1.3749066155437781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1285787583835352, "standard_deviation": 0.3973025515901493, "minimum": 0.5884927024804615, "maximum": 1.858341103563258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14536121685006004, "standard_deviation": 0.008391204495141026, "minimum": 0.1332776447651046, "maximum": 0.1624618236530143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943503668755129, "standard_deviation": 0.0011668807364794886, "minimum": 0.9919997833416706, "maximum": 0.9956140657843109, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8303580240922975, "standard_deviation": 0.04044317843448993, "minimum": 0.7637096531873786, "maximum": 0.8799724395388295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2399974550852193, "standard_deviation": 0.4393818277449563, "minimum": 0.5422709566027369, "maximum": 2.003563300765881, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08429113640964218, "standard_deviation": 0.010779242810800716, "minimum": 0.07444195310383732, "maximum": 0.10295516584686316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9945285908802244, "standard_deviation": 0.0006903671485056723, "minimum": 0.9937776113810787, "maximum": 0.995461531402646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3231356198342736, "standard_deviation": 0.024601800887437686, "minimum": 1.2969169491307007, "maximum": 1.3639438921062734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.24981594245391658, "standard_deviation": 0.14908419800436978, "minimum": 0.07619954036251081, "maximum": 0.48094769115027103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08354606590814767, "standard_deviation": 0.005154339951791054, "minimum": 0.07690413240605083, "maximum": 0.08988703970785616, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981479931384112, "standard_deviation": 0.0002666080619964627, "minimum": 0.9976429566696857, "maximum": 0.9984493743332838, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7985020753399974, "standard_deviation": 0.027086169933588467, "minimum": 0.7679694118695908, "maximum": 0.8463899624286263, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.26769892636308884, "standard_deviation": 0.09844051337938745, "minimum": 0.08872686915148938, "maximum": 0.39935345706053244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0479798797292668, "standard_deviation": 0.004118516523062972, "minimum": 0.04292109419696337, "maximum": 0.053684991911374484, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23419345865634128, "standard_deviation": 0.05005270221855723, "minimum": 0.17269937305811986, "maximum": 0.3227914592067821, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.540277249331317, "standard_deviation": 0.010310464165919643, "minimum": 5.520862259117209, "maximum": 5.55115089594972, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6307387960759981, "standard_deviation": 0.013880344070007426, "minimum": 0.6085902408214107, "maximum": 0.6484467276404237, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10868860481451154, "standard_deviation": 0.06407726635382868, "minimum": -0.19400442661352088, "maximum": 0.006292648872901354, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947320082647759, "standard_deviation": 0.0005822074494474404, "minimum": 0.9938009512204012, "maximum": 0.9952910371429158, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8370268894549171, "standard_deviation": 0.03394104143133894, "minimum": 0.7898679512875456, "maximum": 0.8903359894010993, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6307475190045818, "standard_deviation": 0.23906855944327968, "minimum": 0.16536985245105038, "maximum": 0.8827693854527343, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07524449217813121, "standard_deviation": 0.003948322571222158, "minimum": 0.07154542415992272, "maximum": 0.08260502786911499, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983729366683585, "standard_deviation": 0.000272040520016217, "minimum": 0.9979242955835198, "maximum": 0.9986728259220838, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.436902562459842, "standard_deviation": 0.02310455832321043, "minimum": 0.39546417852923843, "maximum": 0.4775424260603735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5942688591630088, "standard_deviation": 0.26883640223885263, "minimum": 0.20671280518198532, "maximum": 0.8847400892032321, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04165108727398167, "standard_deviation": 0.0038022443446785593, "minimum": 0.037556171434395394, "maximum": 0.04901782975434211, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0047", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.171109437942505, "predicted_mos": 3.171109437942505, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "ae41311b8054c04dcf2faffd11472e67367a71ed4093f69b77668da8f6878fcd", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.109139900188893, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7927887865706887, "全帯域_10ms_rms_envelope_nrmse": 0.07285090472646934, "全帯域_10ms_envelope_correlation": 0.994947961631494, "全帯域_10ms_modulation_mae_db": 0.385906729549399, "全帯域_40ms_log_rms_mae_db": 0.4215070571558607, "全帯域_40ms_rms_envelope_nrmse": 0.03829168041427465, "全帯域_40ms_envelope_correlation": 0.9985044812353914, "全帯域_40ms_modulation_mae_db": 0.36558249549230853, "300–1000 Hz_10ms_log_rms_mae_db": 1.5382242979460081, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09432888085001723, "300–1000 Hz_10ms_envelope_correlation": 0.9929836866960503, "300–1000 Hz_10ms_modulation_mae_db": 0.23831224538107162, "300–1000 Hz_40ms_log_rms_mae_db": 0.9072010282071586, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04307727073822352, "300–1000 Hz_40ms_envelope_correlation": 0.9985227113512993, "300–1000 Hz_40ms_modulation_mae_db": 0.09695829693884615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2599915004268536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08967995446504166, "1000–3000 Hz_10ms_envelope_correlation": 0.9941227378632463, "1000–3000 Hz_10ms_modulation_mae_db": 0.32929858066395107, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7574261848041964, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.045477927004174426, "1000–3000 Hz_40ms_envelope_correlation": 0.9984744431000557, "1000–3000 Hz_40ms_modulation_mae_db": 0.4419377140320055, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4478412344958758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15141666697406134, "3000–6000 Hz_10ms_envelope_correlation": 0.9822095411922327, "3000–6000 Hz_10ms_modulation_mae_db": 0.5904844100755896, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8827900235228949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08359638925527327, "3000–6000 Hz_40ms_envelope_correlation": 0.9940555407149374, "3000–6000 Hz_40ms_modulation_mae_db": 0.5400182809229557, "mr_stft_db": 5.634275178073307, "band_mae_db": 0.16442208936466327, "rms_gain_error_db": -0.0582229645732808, "output_peak": 0.6323265175531411}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8119499048566514, "全帯域_10ms_rms_envelope_nrmse": 0.07590807907821959, "全帯域_10ms_envelope_correlation": 0.994556109449744, "全帯域_10ms_modulation_mae_db": 0.5233047866324582, "全帯域_40ms_log_rms_mae_db": 0.42665294114045377, "全帯域_40ms_rms_envelope_nrmse": 0.039460947323179396, "全帯域_40ms_envelope_correlation": 0.9984557495044656, "全帯域_40ms_modulation_mae_db": 0.5491846640033552, "300–1000 Hz_10ms_log_rms_mae_db": 1.5535156895235314, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10682393129068779, "300–1000 Hz_10ms_envelope_correlation": 0.9909674136549805, "300–1000 Hz_10ms_modulation_mae_db": 0.3282506304142512, "300–1000 Hz_40ms_log_rms_mae_db": 0.9931739486608826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054684131743945166, "300–1000 Hz_40ms_envelope_correlation": 0.9975341446925079, "300–1000 Hz_40ms_modulation_mae_db": 0.2845334371037378, "1000–3000 Hz_10ms_log_rms_mae_db": 1.275561049864514, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10727982129204838, "1000–3000 Hz_10ms_envelope_correlation": 0.9915637363628703, "1000–3000 Hz_10ms_modulation_mae_db": 0.9371191420814605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8160096125703313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06365915081498268, "1000–3000 Hz_40ms_envelope_correlation": 0.9969700074435053, "1000–3000 Hz_40ms_modulation_mae_db": 1.0141641423413965, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3954460955537755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13877064184202248, "3000–6000 Hz_10ms_envelope_correlation": 0.9854925414063173, "3000–6000 Hz_10ms_modulation_mae_db": 1.1407625506468684, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8835639410063099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08023600169829073, "3000–6000 Hz_40ms_envelope_correlation": 0.9953483894285604, "3000–6000 Hz_40ms_modulation_mae_db": 1.2417361303011, "mr_stft_db": 5.620765168410053, "band_mae_db": 0.19329905660705016, "rms_gain_error_db": -0.08137140496051706, "output_peak": 0.6223432151962481}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8606496485212091, "全帯域_10ms_rms_envelope_nrmse": 0.07082287293123868, "全帯域_10ms_envelope_correlation": 0.9952222526093104, "全帯域_10ms_modulation_mae_db": 0.3963618118927166, "全帯域_40ms_log_rms_mae_db": 0.4269454548113485, "全帯域_40ms_rms_envelope_nrmse": 0.0366171048833567, "全帯域_40ms_envelope_correlation": 0.9986284795666979, "全帯域_40ms_modulation_mae_db": 0.27351672216382766, "300–1000 Hz_10ms_log_rms_mae_db": 1.44178299099732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07871293223531098, "300–1000 Hz_10ms_envelope_correlation": 0.9951447409305879, "300–1000 Hz_10ms_modulation_mae_db": 0.2666977570317097, "300–1000 Hz_40ms_log_rms_mae_db": 0.9037277996456133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04639316810765479, "300–1000 Hz_40ms_envelope_correlation": 0.9982076790420396, "300–1000 Hz_40ms_modulation_mae_db": 0.08982203231473614, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2896894654782627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10923651149307825, "1000–3000 Hz_10ms_envelope_correlation": 0.9915062694606245, "1000–3000 Hz_10ms_modulation_mae_db": 0.5752502971687602, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7558085798244493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058831290647596034, "1000–3000 Hz_40ms_envelope_correlation": 0.9977847355336609, "1000–3000 Hz_40ms_modulation_mae_db": 0.4165729792938555, "3000–6000 Hz_10ms_log_rms_mae_db": 1.402967621223874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14265076720940123, "3000–6000 Hz_10ms_envelope_correlation": 0.9841425993077777, "3000–6000 Hz_10ms_modulation_mae_db": 0.21007635034930341, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8799508933576564, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08765865241207918, "3000–6000 Hz_40ms_envelope_correlation": 0.9934376913813178, "3000–6000 Hz_40ms_modulation_mae_db": 0.1165621736966676, "mr_stft_db": 5.622770825295298, "band_mae_db": 0.24257784558781603, "rms_gain_error_db": -0.05138622199822672, "output_peak": 0.6159414742463533}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8017302610878395, "全帯域_10ms_rms_envelope_nrmse": 0.07300696003853152, "全帯域_10ms_envelope_correlation": 0.9951253932542113, "全帯域_10ms_modulation_mae_db": 0.7941721800718279, "全帯域_40ms_log_rms_mae_db": 0.4591476596440085, "全帯域_40ms_rms_envelope_nrmse": 0.04482076727722045, "全帯域_40ms_envelope_correlation": 0.9981806978599922, "全帯域_40ms_modulation_mae_db": 0.7933718795155208, "300–1000 Hz_10ms_log_rms_mae_db": 1.4819268950390996, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10000101593804227, "300–1000 Hz_10ms_envelope_correlation": 0.9921407900387729, "300–1000 Hz_10ms_modulation_mae_db": 0.42035292826823173, "300–1000 Hz_40ms_log_rms_mae_db": 0.9192911673757931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06444547797232725, "300–1000 Hz_40ms_envelope_correlation": 0.9966091012175438, "300–1000 Hz_40ms_modulation_mae_db": 0.5053730652695276, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3109879519599008, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10136702636395795, "1000–3000 Hz_10ms_envelope_correlation": 0.9928219762931678, "1000–3000 Hz_10ms_modulation_mae_db": 0.18672110857015542, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8219096161820565, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06130448045626722, "1000–3000 Hz_40ms_envelope_correlation": 0.9976380474179418, "1000–3000 Hz_40ms_modulation_mae_db": 0.3347311172600824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4232899538974775, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1355071174966379, "3000–6000 Hz_10ms_envelope_correlation": 0.9858416095435276, "3000–6000 Hz_10ms_modulation_mae_db": 1.0174736538998783, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8524222228218177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08033582019881369, "3000–6000 Hz_40ms_envelope_correlation": 0.9948070127347158, "3000–6000 Hz_40ms_modulation_mae_db": 1.0612543150913838, "mr_stft_db": 5.6225388791440585, "band_mae_db": 0.26165626544541226, "rms_gain_error_db": -0.14872924520129355, "output_peak": 0.6210372970530349}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8087509758586304, "全帯域_10ms_rms_envelope_nrmse": 0.07211357227544535, "全帯域_10ms_envelope_correlation": 0.9952718141582858, "全帯域_10ms_modulation_mae_db": 0.23692331488098492, "全帯域_40ms_log_rms_mae_db": 0.4410003799126965, "全帯域_40ms_rms_envelope_nrmse": 0.044180645005774705, "全帯域_40ms_envelope_correlation": 0.9982639894588299, "全帯域_40ms_modulation_mae_db": 0.3349713993204655, "300–1000 Hz_10ms_log_rms_mae_db": 1.4850079449696383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09469383619634358, "300–1000 Hz_10ms_envelope_correlation": 0.9930348787049996, "300–1000 Hz_10ms_modulation_mae_db": 0.19994418869349753, "300–1000 Hz_40ms_log_rms_mae_db": 0.9475566279741828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058890184964142736, "300–1000 Hz_40ms_envelope_correlation": 0.9972829003553004, "300–1000 Hz_40ms_modulation_mae_db": 0.2382554984214692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2610553909340374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10349489362716875, "1000–3000 Hz_10ms_envelope_correlation": 0.992532932115597, "1000–3000 Hz_10ms_modulation_mae_db": 0.07959427719866058, "1000–3000 Hz_40ms_log_rms_mae_db": 0.777535758070443, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06592677980587748, "1000–3000 Hz_40ms_envelope_correlation": 0.9972070384812624, "1000–3000 Hz_40ms_modulation_mae_db": 0.30311890388297597, "3000–6000 Hz_10ms_log_rms_mae_db": 1.427302820979439, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13234542302314728, "3000–6000 Hz_10ms_envelope_correlation": 0.9864393572273733, "3000–6000 Hz_10ms_modulation_mae_db": 0.20901982357752724, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8572859606098743, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07407239463380785, "3000–6000 Hz_40ms_envelope_correlation": 0.9953538496474025, "3000–6000 Hz_40ms_modulation_mae_db": 0.29091045313512426, "mr_stft_db": 5.6239671957852515, "band_mae_db": 0.24260330411268546, "rms_gain_error_db": -0.15618589156476984, "output_peak": 0.6286774527344287}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8106525836547287, "全帯域_10ms_rms_envelope_nrmse": 0.0779367034350642, "全帯域_10ms_envelope_correlation": 0.9943312744861753, "全帯域_10ms_modulation_mae_db": 0.5303030527206695, "全帯域_40ms_log_rms_mae_db": 0.4178704001871806, "全帯域_40ms_rms_envelope_nrmse": 0.04359988624513783, "全帯域_40ms_envelope_correlation": 0.998196188364934, "全帯域_40ms_modulation_mae_db": 0.5678892615365788, "300–1000 Hz_10ms_log_rms_mae_db": 1.5246811793908035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09666209730871811, "300–1000 Hz_10ms_envelope_correlation": 0.9926411154272063, "300–1000 Hz_10ms_modulation_mae_db": 0.7902771949569897, "300–1000 Hz_40ms_log_rms_mae_db": 0.9585480268536819, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06111936428020482, "300–1000 Hz_40ms_envelope_correlation": 0.9969235076236103, "300–1000 Hz_40ms_modulation_mae_db": 0.4073310443157558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2910866395095426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10226072464550941, "1000–3000 Hz_10ms_envelope_correlation": 0.9924948144169324, "1000–3000 Hz_10ms_modulation_mae_db": 0.6654051389983228, "1000–3000 Hz_40ms_log_rms_mae_db": 0.767078917231748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056479682327688116, "1000–3000 Hz_40ms_envelope_correlation": 0.9978485398480239, "1000–3000 Hz_40ms_modulation_mae_db": 0.35874222114448484, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4933992323967304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15852050869619894, "3000–6000 Hz_10ms_envelope_correlation": 0.9804476959873237, "3000–6000 Hz_10ms_modulation_mae_db": 0.8286764931982386, "3000–6000 Hz_40ms_log_rms_mae_db": 0.902808900448584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08226970638615023, "3000–6000 Hz_40ms_envelope_correlation": 0.9945434026235468, "3000–6000 Hz_40ms_modulation_mae_db": 0.7007768238889723, "mr_stft_db": 5.627135891727897, "band_mae_db": 0.22116131750294765, "rms_gain_error_db": -0.10655684824260148, "output_peak": 0.6140672035472583}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8026681332465748, "全帯域_10ms_rms_envelope_nrmse": 0.07046704050108744, "全帯域_10ms_envelope_correlation": 0.9953749535476164, "全帯域_10ms_modulation_mae_db": 0.43352088446631765, "全帯域_40ms_log_rms_mae_db": 0.39837116769735, "全帯域_40ms_rms_envelope_nrmse": 0.0396015066243609, "全帯域_40ms_envelope_correlation": 0.9985199952240265, "全帯域_40ms_modulation_mae_db": 0.39274074982247015, "300–1000 Hz_10ms_log_rms_mae_db": 1.5585010204736318, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09286450451138528, "300–1000 Hz_10ms_envelope_correlation": 0.9931975738363372, "300–1000 Hz_10ms_modulation_mae_db": 0.12012308645278881, "300–1000 Hz_40ms_log_rms_mae_db": 0.9449144688210666, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054201760385653344, "300–1000 Hz_40ms_envelope_correlation": 0.9975809191569341, "300–1000 Hz_40ms_modulation_mae_db": 0.1545078840403667, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2902802235694435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10612439039502765, "1000–3000 Hz_10ms_envelope_correlation": 0.99189421162841, "1000–3000 Hz_10ms_modulation_mae_db": 0.5266760274581085, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8036413174437135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06335813337209471, "1000–3000 Hz_40ms_envelope_correlation": 0.9971962738759608, "1000–3000 Hz_40ms_modulation_mae_db": 0.5553913382867474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4179163070123584, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12548643231370876, "3000–6000 Hz_10ms_envelope_correlation": 0.9878061820134789, "3000–6000 Hz_10ms_modulation_mae_db": 0.3940843390429578, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8603409135987988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06928300567006945, "3000–6000 Hz_40ms_envelope_correlation": 0.9960754814312023, "3000–6000 Hz_40ms_modulation_mae_db": 0.27589728526476914, "mr_stft_db": 5.64100409401924, "band_mae_db": 0.22396225934547634, "rms_gain_error_db": -0.11272336894177906, "output_peak": 0.6192989471994546}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8575541942908986, "全帯域_10ms_rms_envelope_nrmse": 0.07402342941769623, "全帯域_10ms_envelope_correlation": 0.9948240382119087, "全帯域_10ms_modulation_mae_db": 0.11694961812201632, "全帯域_40ms_log_rms_mae_db": 0.4437186741540782, "全帯域_40ms_rms_envelope_nrmse": 0.04298033958601049, "全帯域_40ms_envelope_correlation": 0.9981570924177353, "全帯域_40ms_modulation_mae_db": 0.23157126118129293, "300–1000 Hz_10ms_log_rms_mae_db": 1.5345405110234414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10465104952155278, "300–1000 Hz_10ms_envelope_correlation": 0.9913528142591215, "300–1000 Hz_10ms_modulation_mae_db": 0.3488481612765022, "300–1000 Hz_40ms_log_rms_mae_db": 0.9730183649937111, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05962612818577689, "300–1000 Hz_40ms_envelope_correlation": 0.9970864562453139, "300–1000 Hz_40ms_modulation_mae_db": 0.32860351960074174, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3433756234354324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10059653845146703, "1000–3000 Hz_10ms_envelope_correlation": 0.9927890628665488, "1000–3000 Hz_10ms_modulation_mae_db": 0.27840551720063145, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8391909455903814, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06285103093648774, "1000–3000 Hz_40ms_envelope_correlation": 0.99728828161962, "1000–3000 Hz_40ms_modulation_mae_db": 0.47392134365876454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4490430658965097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12270276252510998, "3000–6000 Hz_10ms_envelope_correlation": 0.9883650247474889, "3000–6000 Hz_10ms_modulation_mae_db": 0.4978494615405336, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8802339081332783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07078056106988295, "3000–6000 Hz_40ms_envelope_correlation": 0.9957446760359017, "3000–6000 Hz_40ms_modulation_mae_db": 0.561033905515489, "mr_stft_db": 5.630529409177874, "band_mae_db": 0.19330288272588933, "rms_gain_error_db": -0.08797976047756707, "output_peak": 0.610380330821061}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84000, "output_samples": 84000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5111039430386235, "spectral_convergence": 0.3421841801599182, "spectral_convergence_status": "ok", "amplitude_floor": 1.02822042353332e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.463568548638778, "spectral_convergence": 0.3436790293571638, "spectral_convergence_status": "ok", "amplitude_floor": 8.928413196720056e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.928153042542521, "spectral_convergence": 0.3546433690232922, "spectral_convergence_status": "ok", "amplitude_floor": 7.736392334055093e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 1.4784881078912855e-05, "output_power": 1.7199900437332853e-05, "delta_db": 0.6570809731392557, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002602315552138397, "output_power": 0.0002540663444014383, "delta_db": -0.10412818347334755, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005931268593866254, "output_power": 0.0005845700579802947, "delta_db": -0.06311024464175984, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005283196568297616, "output_power": 0.0005207498669277745, "delta_db": -0.06267601846292917, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003079318069923152, "output_power": 0.0003056255531033654, "delta_db": -0.032648880776575503, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00026143236338356946, "output_power": 0.00025743673896025854, "delta_db": -0.06688823569411197, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16442208936466327, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0582229645732808, "pitch": null, "output_rms_dbfs": -27.12276973641922, "output_peak": 0.6323265175531411, "reference_rms_dbfs": -27.06454677184594, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.634275178073307}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00019312594202620684, "log_rms_mae_db": 0.7927887865706887, "rms_envelope_nrmse": 0.07285090472646934, "envelope_correlation": 0.994947961631494, "rms_gain_error_db": -0.058223611816536965, "reference_mean_power": 0.0019676998400409485, "output_mean_power": 0.0019414959508909835}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 4.0378624939923077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018862042103601667, "output_power": 0.00020085117275907985, "reference_active": true, "delta_db": 0.27285661951906925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.862312285918816e-05, "output_power": 9.806056929847335e-05, "reference_active": true, "delta_db": -0.024843399138692997}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3789377405749525e-05, "output_power": 1.1312084513182268e-05, "reference_active": true, "delta_db": -0.8600201699904347}], "active_band_count": 3, "active_band_mae_db": 0.385906729549399}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00018244538675777041, "log_rms_mae_db": 0.4215070571558607, "rms_envelope_nrmse": 0.03829168041427465, "envelope_correlation": 0.9985044812353914, "rms_gain_error_db": -0.058225539141464853, "reference_mean_power": 0.001979010526094432, "output_mean_power": 0.0019526551458083803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 3.3102005537920435e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017836230285954771, "output_power": 0.00019201976753010857, "reference_active": true, "delta_db": 0.32042868627217425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3320761072042273e-05, "output_power": 4.347407490652705e-05, "reference_active": true, "delta_db": 0.015342714993473736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.132775238046075e-07, "output_power": 6.825605151388346e-07, "reference_active": true, "delta_db": -0.7609760852112775}], "active_band_count": 3, "active_band_mae_db": 0.36558249549230853}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00014606965842361196, "log_rms_mae_db": 1.5382242979460081, "rms_envelope_nrmse": 0.09432888085001723, "envelope_correlation": 0.9929836866960503, "rms_gain_error_db": -0.09303880049691524, "reference_mean_power": 0.0007714288246159527, "output_mean_power": 0.0007550782877016688}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 2.155891354220712e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.355627855804233e-05, "output_power": 8.317795045559497e-05, "reference_active": true, "delta_db": -0.01970874129389616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.117824128471733e-05, "output_power": 1.949679814455631e-05, "reference_active": true, "delta_db": -0.35926596589216847}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3214088476749303e-06, "output_power": 3.5885461940059514e-06, "reference_active": true, "delta_db": 0.33596202895715027}], "active_band_count": 3, "active_band_mae_db": 0.23831224538107162}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.0001346093108760191, "log_rms_mae_db": 0.9072010282071586, "rms_envelope_nrmse": 0.04307727073822352, "envelope_correlation": 0.9985227113512993, "rms_gain_error_db": -0.09304078523318436, "reference_mean_power": 0.0007758659430163304, "output_mean_power": 0.0007594210137279318}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.945223621465206e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.346976557991944e-05, "output_power": 7.289463399834214e-05, "reference_active": true, "delta_db": -0.03413094421091624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.57892144723787e-06, "output_power": 9.068125814815311e-06, "reference_active": true, "delta_db": -0.23799074732635178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.32297115664878e-07, "output_power": 2.3330236286486545e-07, "reference_active": true, "delta_db": 0.018753199279270452}], "active_band_count": 3, "active_band_mae_db": 0.09695829693884615}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00012444218003818797, "log_rms_mae_db": 1.2599915004268536, "rms_envelope_nrmse": 0.08967995446504166, "envelope_correlation": 0.9941227378632463, "rms_gain_error_db": -0.08967935072583012, "reference_mean_power": 0.0006875503513213645, "output_mean_power": 0.0006734984076330819}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.3390858183888e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.254326674690693e-05, "output_power": 3.985594182164467e-05, "reference_active": true, "delta_db": -0.2833775853725432}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4859458498211302e-05, "output_power": 1.538170791765436e-05, "reference_active": true, "delta_db": 0.1500157695867204}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.141405389451205e-06, "output_power": 3.569227440269831e-06, "reference_active": true, "delta_db": 0.5545023870325896}], "active_band_count": 3, "active_band_mae_db": 0.32929858066395107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00011369765125964215, "log_rms_mae_db": 0.7574261848041964, "rms_envelope_nrmse": 0.045477927004174426, "envelope_correlation": 0.9984744431000557, "rms_gain_error_db": -0.08967778304253693, "reference_mean_power": 0.0006915015836222027, "output_mean_power": 0.0006773691303687708}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.214722623247201e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7827908938064025e-05, "output_power": 3.624281646749165e-05, "reference_active": true, "delta_db": -0.18590395103634613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.619804659862373e-06, "output_power": 6.150552549173044e-06, "reference_active": true, "delta_db": 0.3919291329611512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0150238495180175e-07, "output_power": 2.3937467218437733e-07, "reference_active": true, "delta_db": 0.747980058098519}], "active_band_count": 3, "active_band_mae_db": 0.4419377140320055}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00010848831624356859, "log_rms_mae_db": 1.4478412344958758, "rms_envelope_nrmse": 0.15141666697406134, "envelope_correlation": 0.9822095411922327, "rms_gain_error_db": -0.015515467794376979, "reference_mean_power": 0.00023818463717296146, "output_mean_power": 0.000237335224440505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.7158903163900278e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.877460217795428e-05, "output_power": 9.862734032732354e-05, "reference_active": true, "delta_db": 0.45708586918169597}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.954888518804953e-05, "output_power": 4.2428338963665175e-05, "reference_active": true, "delta_db": 0.3052178378382519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.296333400161532e-06, "output_power": 5.783486393724579e-06, "reference_active": true, "delta_db": -1.009149523206821}], "active_band_count": 3, "active_band_mae_db": 0.5904844100755896}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 7.397564116979367e-05, "log_rms_mae_db": 0.8827900235228949, "rms_envelope_nrmse": 0.08359638925527327, "envelope_correlation": 0.9940555407149374, "rms_gain_error_db": -0.01551188395333267, "reference_mean_power": 0.0002395532500233675, "output_mean_power": 0.00023869915352817478}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.5691901914560294e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.470858674877874e-05, "output_power": 0.0001050141359533559, "reference_active": true, "delta_db": 0.4485840731376808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.833295648941501e-05, "output_power": 1.970956545439059e-05, "reference_active": true, "delta_db": 0.3144454159379171}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.840671065644946e-07, "output_power": 3.973772838907992e-07, "reference_active": true, "delta_db": -0.8570253536932692}], "active_band_count": 3, "active_band_mae_db": 0.5400182809229557}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9017836220170528, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "edaae1b51c0b658d8546f09e59b0e50e57021b22c87e8257de2621ff0369ccab", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9924657176259246, "standard_deviation": 0.0008488044488217251, "minimum": 0.9915062694606245, "maximum": 0.9941227378632463, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2902534806472483, "standard_deviation": 0.027352400649600057, "minimum": 1.2599915004268536, "maximum": 1.3433756234354324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.44730876116750634, "standard_deviation": 0.28176054012443313, "minimum": 0.07959427719866058, "maximum": 0.9371191420814605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10250498259166238, "standard_deviation": 0.005996078652822888, "minimum": 0.08967995446504166, "maximum": 0.10923651149307825, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975509209150039, "standard_deviation": 0.00048612168823000917, "minimum": 0.9969700074435053, "maximum": 0.9984744431000557, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7923251164646649, "standard_deviation": 0.03200033236480496, "minimum": 0.7558085798244493, "maximum": 0.8391909455903814, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.48732246998753914, "standard_deviation": 0.2277806997778339, "minimum": 0.30311890388297597, "maximum": 1.0141641423413965, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059736059420646054, "standard_deviation": 0.006477598927794505, "minimum": 0.045477927004174426, "maximum": 0.06592677980587748, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.98509306892819, "standard_deviation": 0.0027096830457299173, "minimum": 0.9804476959873237, "maximum": 0.9883650247474889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.432150791432005, "standard_deviation": 0.031136658205997213, "minimum": 1.3954460955537755, "maximum": 1.4933992323967304, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6110533852913621, "standard_deviation": 0.353726366623307, "minimum": 0.20901982357752724, "maximum": 1.1407625506468684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13842504001003597, "standard_deviation": 0.012251297522027174, "minimum": 0.12270276252510998, "maximum": 0.15852050869619894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994920755499698, "standard_deviation": 0.0008852833894520022, "minimum": 0.9934376913813178, "maximum": 0.9960754814312023, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8749245954374018, "standard_deviation": 0.016893160817107195, "minimum": 0.8524222228218177, "maximum": 0.902808900448584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5985236709770577, "standard_deviation": 0.39172266951832024, "minimum": 0.1165621736966676, "maximum": 1.2417361303011, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07852906641554593, "standard_deviation": 0.00648722231647481, "minimum": 0.06928300567006945, "maximum": 0.08765865241207918, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9926828766935071, "standard_deviation": 0.0012843876920762348, "minimum": 0.9909674136549805, "maximum": 0.9951447409305879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5147725661704343, "standard_deviation": 0.04096059461252685, "minimum": 1.44178299099732, "maximum": 1.5585010204736318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.33910077405938027, "standard_deviation": 0.20462199289013241, "minimum": 0.12012308645278881, "maximum": 0.7902771949569897, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09609228098150725, "standard_deviation": 0.008631745962365404, "minimum": 0.07871293223531098, "maximum": 0.10682393129068779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974684274605687, "standard_deviation": 0.0006426154255430172, "minimum": 0.9966091012175438, "maximum": 0.9985227113512993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9434289290665112, "standard_deviation": 0.03177871155096242, "minimum": 0.9037277996456133, "maximum": 0.9931739486608826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.26317309725064764, "standard_deviation": 0.1484619765406418, "minimum": 0.08982203231473614, "maximum": 0.5053730652695276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.055304685797241064, "standard_deviation": 0.0073622744388977595, "minimum": 0.04307727073822352, "maximum": 0.06444547797232725, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21787312758649258, "standard_deviation": 0.032185518007380315, "minimum": 0.16442208936466327, "maximum": 0.26165626544541226, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.627873330204123, "standard_deviation": 0.006982375365113074, "minimum": 5.620765168410053, "maximum": 5.64100409401924, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6205090547938725, "standard_deviation": 0.007333231450021043, "minimum": 0.610380330821061, "maximum": 0.6323265175531411, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10039446324500445, "standard_deviation": 0.038424612883783744, "minimum": -0.15618589156476984, "maximum": -0.05138622199822672, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949567246685932, "standard_deviation": 0.0003669398416452574, "minimum": 0.9943312744861753, "maximum": 0.9953749535476164, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8183430610109026, "standard_deviation": 0.025895652500211915, "minimum": 0.7927887865706887, "maximum": 0.8606496485212091, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42718029729204876, "standard_deviation": 0.20343182266200097, "minimum": 0.11694961812201632, "maximum": 0.7941721800718279, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07339119530046904, "standard_deviation": 0.00252459050284358, "minimum": 0.07046704050108744, "maximum": 0.0779367034350642, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983633342040091, "standard_deviation": 0.0001840056789693689, "minimum": 0.9981570924177353, "maximum": 0.9986284795666979, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4294017168378721, "standard_deviation": 0.018501375437196207, "minimum": 0.39837116769735, "maximum": 0.4591476596440085, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43860355412947744, "standard_deviation": 0.18635399050635673, "minimum": 0.23157126118129293, "maximum": 0.7933718795155208, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04119410966991439, "standard_deviation": 0.0030697538648896687, "minimum": 0.0366171048833567, "maximum": 0.04482076727722045, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1580-141084-0047", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.409174919128418, "predicted_mos": 2.409174919128418, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "d229449f6d6d88316b3611a907b4286a2675c87347ec5919d3564a5cccb7e58f", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.514582199975848, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.810601371395359, "全帯域_10ms_rms_envelope_nrmse": 0.07567412671212112, "全帯域_10ms_envelope_correlation": 0.9946487607367713, "全帯域_10ms_modulation_mae_db": 0.6866495530126809, "全帯域_40ms_log_rms_mae_db": 0.4033005633743024, "全帯域_40ms_rms_envelope_nrmse": 0.04424936593021085, "全帯域_40ms_envelope_correlation": 0.9981258767168418, "全帯域_40ms_modulation_mae_db": 1.0196220900030457, "300–1000 Hz_10ms_log_rms_mae_db": 1.4949804710308017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09507138253769457, "300–1000 Hz_10ms_envelope_correlation": 0.9929959650926111, "300–1000 Hz_10ms_modulation_mae_db": 0.268579112590972, "300–1000 Hz_40ms_log_rms_mae_db": 0.8947428585677779, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05484435705500137, "300–1000 Hz_40ms_envelope_correlation": 0.9976961332350978, "300–1000 Hz_40ms_modulation_mae_db": 0.366668061106703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.278916843951122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10020868857224646, "1000–3000 Hz_10ms_envelope_correlation": 0.9928919585934335, "1000–3000 Hz_10ms_modulation_mae_db": 0.22706564647293484, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7638541236336317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05741261964572413, "1000–3000 Hz_40ms_envelope_correlation": 0.9978721951261779, "1000–3000 Hz_40ms_modulation_mae_db": 0.324945387320194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.402130783856686, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1552215861808699, "3000–6000 Hz_10ms_envelope_correlation": 0.9822890117523269, "3000–6000 Hz_10ms_modulation_mae_db": 1.5275981072438298, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8851150701081815, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10554109227960261, "3000–6000 Hz_40ms_envelope_correlation": 0.9912882048514507, "3000–6000 Hz_40ms_modulation_mae_db": 1.890520886086433, "mr_stft_db": 5.6109333473219465, "band_mae_db": 0.24312548219713306, "rms_gain_error_db": -0.11600822306859257, "output_peak": 0.623443276763886}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.823667095327888, "全帯域_10ms_rms_envelope_nrmse": 0.07187174130534253, "全帯域_10ms_envelope_correlation": 0.9951551683332782, "全帯域_10ms_modulation_mae_db": 0.11372819915767311, "全帯域_40ms_log_rms_mae_db": 0.41286490845043244, "全帯域_40ms_rms_envelope_nrmse": 0.035462410910718355, "全帯域_40ms_envelope_correlation": 0.9988197251442861, "全帯域_40ms_modulation_mae_db": 0.154762161719924, "300–1000 Hz_10ms_log_rms_mae_db": 1.52427422135686, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10039343631085028, "300–1000 Hz_10ms_envelope_correlation": 0.9920306953660593, "300–1000 Hz_10ms_modulation_mae_db": 0.3309867949891535, "300–1000 Hz_40ms_log_rms_mae_db": 0.9293157196594143, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0539095781863475, "300–1000 Hz_40ms_envelope_correlation": 0.9975988052578681, "300–1000 Hz_40ms_modulation_mae_db": 0.544663934203038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.300622418011614, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10230256510761693, "1000–3000 Hz_10ms_envelope_correlation": 0.9924276796029062, "1000–3000 Hz_10ms_modulation_mae_db": 0.3013386641880926, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7397213684954288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05028563994641302, "1000–3000 Hz_40ms_envelope_correlation": 0.99828750327052, "1000–3000 Hz_40ms_modulation_mae_db": 0.15185281522842964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4028118599529107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1397830347988508, "3000–6000 Hz_10ms_envelope_correlation": 0.9847844903165036, "3000–6000 Hz_10ms_modulation_mae_db": 0.07251908977908059, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8389257431141753, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07891307095755942, "3000–6000 Hz_40ms_envelope_correlation": 0.9946985207605216, "3000–6000 Hz_40ms_modulation_mae_db": 0.14252376931235533, "mr_stft_db": 5.610296987232826, "band_mae_db": 0.197100880211452, "rms_gain_error_db": -0.10315899679653473, "output_peak": 0.6498376788391262}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8069252395222765, "全帯域_10ms_rms_envelope_nrmse": 0.07129671094348769, "全帯域_10ms_envelope_correlation": 0.9952756626690705, "全帯域_10ms_modulation_mae_db": 0.4670381901648077, "全帯域_40ms_log_rms_mae_db": 0.39457274401414744, "全帯域_40ms_rms_envelope_nrmse": 0.03980460720945995, "全帯域_40ms_envelope_correlation": 0.9985178807407696, "全帯域_40ms_modulation_mae_db": 0.4425468430793756, "300–1000 Hz_10ms_log_rms_mae_db": 1.378558794027208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08081867458639105, "300–1000 Hz_10ms_envelope_correlation": 0.9948585223421025, "300–1000 Hz_10ms_modulation_mae_db": 0.4050014993233417, "300–1000 Hz_40ms_log_rms_mae_db": 0.8109887455096544, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03935458653528738, "300–1000 Hz_40ms_envelope_correlation": 0.9987492007017801, "300–1000 Hz_40ms_modulation_mae_db": 0.15996964501131947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3250276549512534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11965442870015504, "1000–3000 Hz_10ms_envelope_correlation": 0.9898695621489412, "1000–3000 Hz_10ms_modulation_mae_db": 0.6643490965092301, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7679386494417567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0720420540655562, "1000–3000 Hz_40ms_envelope_correlation": 0.9966313120400838, "1000–3000 Hz_40ms_modulation_mae_db": 0.26334176593091335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4274873746881571, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13828330358016325, "3000–6000 Hz_10ms_envelope_correlation": 0.9850970629803966, "3000–6000 Hz_10ms_modulation_mae_db": 0.5966622117452713, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9080924500353036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08580420350494883, "3000–6000 Hz_40ms_envelope_correlation": 0.9937467347815824, "3000–6000 Hz_40ms_modulation_mae_db": 0.5313194981308523, "mr_stft_db": 5.616986609993727, "band_mae_db": 0.20862915654860217, "rms_gain_error_db": -0.11423138774017043, "output_peak": 0.629391284756194}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8161073660881049, "全帯域_10ms_rms_envelope_nrmse": 0.07609992759536746, "全帯域_10ms_envelope_correlation": 0.9944773485160727, "全帯域_10ms_modulation_mae_db": 0.5233529704572427, "全帯域_40ms_log_rms_mae_db": 0.39811800698204314, "全帯域_40ms_rms_envelope_nrmse": 0.03870581810314137, "全帯域_40ms_envelope_correlation": 0.9984367838864239, "全帯域_40ms_modulation_mae_db": 0.5731682698969276, "300–1000 Hz_10ms_log_rms_mae_db": 1.5209139776473146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09189427935930979, "300–1000 Hz_10ms_envelope_correlation": 0.9933201303310552, "300–1000 Hz_10ms_modulation_mae_db": 0.28521440842217005, "300–1000 Hz_40ms_log_rms_mae_db": 0.8951871905167437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0481728333878697, "300–1000 Hz_40ms_envelope_correlation": 0.9980573706869965, "300–1000 Hz_40ms_modulation_mae_db": 0.4763226610902067, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2230438297885737, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11039646752116009, "1000–3000 Hz_10ms_envelope_correlation": 0.9910275750007733, "1000–3000 Hz_10ms_modulation_mae_db": 0.611935908791461, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7008341663556902, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053526031906067945, "1000–3000 Hz_40ms_envelope_correlation": 0.9977851826300033, "1000–3000 Hz_40ms_modulation_mae_db": 0.5054698475974235, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4252674905325127, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13602126494881578, "3000–6000 Hz_10ms_envelope_correlation": 0.9855852308965288, "3000–6000 Hz_10ms_modulation_mae_db": 1.0034591199043081, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8838950276004451, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07318209535323611, "3000–6000 Hz_40ms_envelope_correlation": 0.9954507358541591, "3000–6000 Hz_40ms_modulation_mae_db": 1.0694274146203158, "mr_stft_db": 5.612604248989212, "band_mae_db": 0.16441145286816988, "rms_gain_error_db": -0.008840657586965953, "output_peak": 0.6336248892588128}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8431139108821566, "全帯域_10ms_rms_envelope_nrmse": 0.08000747416080235, "全帯域_10ms_envelope_correlation": 0.9941585724006821, "全帯域_10ms_modulation_mae_db": 0.5761505395595808, "全帯域_40ms_log_rms_mae_db": 0.4445043045879586, "全帯域_40ms_rms_envelope_nrmse": 0.04466512269582208, "全帯域_40ms_envelope_correlation": 0.9982783747209213, "全帯域_40ms_modulation_mae_db": 0.494992877239798, "300–1000 Hz_10ms_log_rms_mae_db": 1.4776396909103748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09881356796077567, "300–1000 Hz_10ms_envelope_correlation": 0.9924276200272086, "300–1000 Hz_10ms_modulation_mae_db": 0.3630305108024499, "300–1000 Hz_40ms_log_rms_mae_db": 0.8931170593488617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05931561962787092, "300–1000 Hz_40ms_envelope_correlation": 0.9972880808755518, "300–1000 Hz_40ms_modulation_mae_db": 0.18900194549492824, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3391818529998645, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10853098082226821, "1000–3000 Hz_10ms_envelope_correlation": 0.9917177306811757, "1000–3000 Hz_10ms_modulation_mae_db": 0.34947547098078574, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8050816790776575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0670811566572874, "1000–3000 Hz_40ms_envelope_correlation": 0.9971007842468191, "1000–3000 Hz_40ms_modulation_mae_db": 0.6066349023442287, "3000–6000 Hz_10ms_log_rms_mae_db": 1.487071438351044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13525461481006937, "3000–6000 Hz_10ms_envelope_correlation": 0.9860833879636957, "3000–6000 Hz_10ms_modulation_mae_db": 0.9432796629799016, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9174482990765201, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07879513937800753, "3000–6000 Hz_40ms_envelope_correlation": 0.9952881518229353, "3000–6000 Hz_40ms_modulation_mae_db": 0.8191257023373142, "mr_stft_db": 5.605190093643338, "band_mae_db": 0.23767848042937845, "rms_gain_error_db": -0.154540388687702, "output_peak": 0.6225309705360001}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7914902782107706, "全帯域_10ms_rms_envelope_nrmse": 0.07548979429250789, "全帯域_10ms_envelope_correlation": 0.994676570543313, "全帯域_10ms_modulation_mae_db": 0.5738328498234496, "全帯域_40ms_log_rms_mae_db": 0.4417520982851216, "全帯域_40ms_rms_envelope_nrmse": 0.045370365027010064, "全帯域_40ms_envelope_correlation": 0.998011988339587, "全帯域_40ms_modulation_mae_db": 0.34237545393348384, "300–1000 Hz_10ms_log_rms_mae_db": 1.4141487521101517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08418743096170295, "300–1000 Hz_10ms_envelope_correlation": 0.9944633701452984, "300–1000 Hz_10ms_modulation_mae_db": 0.056820082485459895, "300–1000 Hz_40ms_log_rms_mae_db": 0.8810129493368071, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046318499247338635, "300–1000 Hz_40ms_envelope_correlation": 0.9983106044158606, "300–1000 Hz_40ms_modulation_mae_db": 0.08923588206769988, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3505555458694685, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10892356794599276, "1000–3000 Hz_10ms_envelope_correlation": 0.9916783757572722, "1000–3000 Hz_10ms_modulation_mae_db": 0.51002819515568, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8109797206584176, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06707348136730028, "1000–3000 Hz_40ms_envelope_correlation": 0.9971097183274104, "1000–3000 Hz_40ms_modulation_mae_db": 0.2800882944446817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4310247678403543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1347049470626951, "3000–6000 Hz_10ms_envelope_correlation": 0.9858892318281605, "3000–6000 Hz_10ms_modulation_mae_db": 0.6830756979386791, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8821399923218333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07791425180648442, "3000–6000 Hz_40ms_envelope_correlation": 0.9949166872612158, "3000–6000 Hz_40ms_modulation_mae_db": 0.672544683049915, "mr_stft_db": 5.6072027748929605, "band_mae_db": 0.24451912296500203, "rms_gain_error_db": -0.11521278449342143, "output_peak": 0.6351363598520098}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8129568293610634, "全帯域_10ms_rms_envelope_nrmse": 0.07421322986465789, "全帯域_10ms_envelope_correlation": 0.9949741735265027, "全帯域_10ms_modulation_mae_db": 0.6844346330857336, "全帯域_40ms_log_rms_mae_db": 0.4325435857769287, "全帯域_40ms_rms_envelope_nrmse": 0.04048584972626711, "全帯域_40ms_envelope_correlation": 0.9985971113245325, "全帯域_40ms_modulation_mae_db": 0.6168620004312572, "300–1000 Hz_10ms_log_rms_mae_db": 1.5076282709203281, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08287095910492709, "300–1000 Hz_10ms_envelope_correlation": 0.9945736374875149, "300–1000 Hz_10ms_modulation_mae_db": 0.2629896610176334, "300–1000 Hz_40ms_log_rms_mae_db": 0.8787436419180806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04630470831639113, "300–1000 Hz_40ms_envelope_correlation": 0.9982151035229759, "300–1000 Hz_40ms_modulation_mae_db": 0.41601541511282, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3102740555913015, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1084380513780702, "1000–3000 Hz_10ms_envelope_correlation": 0.9921149677249163, "1000–3000 Hz_10ms_modulation_mae_db": 0.3127493537742821, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7862905675419601, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06325246877693412, "1000–3000 Hz_40ms_envelope_correlation": 0.9979353632527436, "1000–3000 Hz_40ms_modulation_mae_db": 0.15185788450943014, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4417013937842134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15034220869660245, "3000–6000 Hz_10ms_envelope_correlation": 0.9823645725484969, "3000–6000 Hz_10ms_modulation_mae_db": 0.8522134603952624, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8420973028838015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07837352654606469, "3000–6000 Hz_40ms_envelope_correlation": 0.9947922345435176, "3000–6000 Hz_40ms_modulation_mae_db": 0.942449284941997, "mr_stft_db": 5.616144147320838, "band_mae_db": 0.2417084581429191, "rms_gain_error_db": -0.15566221459633942, "output_peak": 0.6142072132197359}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.815173411887604, "全帯域_10ms_rms_envelope_nrmse": 0.08149738991145512, "全帯域_10ms_envelope_correlation": 0.993673307334501, "全帯域_10ms_modulation_mae_db": 0.6617262115151006, "全帯域_40ms_log_rms_mae_db": 0.4137183681665254, "全帯域_40ms_rms_envelope_nrmse": 0.05201825288982544, "全帯域_40ms_envelope_correlation": 0.9972139880909485, "全帯域_40ms_modulation_mae_db": 1.0046986476802282, "300–1000 Hz_10ms_log_rms_mae_db": 1.4502149248230056, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10172461934381061, "300–1000 Hz_10ms_envelope_correlation": 0.9918162438674486, "300–1000 Hz_10ms_modulation_mae_db": 0.3862166281099753, "300–1000 Hz_40ms_log_rms_mae_db": 0.8495254110690004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059130204515568774, "300–1000 Hz_40ms_envelope_correlation": 0.997095367864716, "300–1000 Hz_40ms_modulation_mae_db": 0.3553743037126471, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2471158462351435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1066091149762885, "1000–3000 Hz_10ms_envelope_correlation": 0.9921269168372819, "1000–3000 Hz_10ms_modulation_mae_db": 0.6278174721140498, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7040001036472319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06400064509475727, "1000–3000 Hz_40ms_envelope_correlation": 0.9975096395226669, "1000–3000 Hz_40ms_modulation_mae_db": 0.34655273351307603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3893990110677705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16107432773616273, "3000–6000 Hz_10ms_envelope_correlation": 0.9810854603961764, "3000–6000 Hz_10ms_modulation_mae_db": 1.5456897300658732, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8272744541267222, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10877877867790928, "3000–6000 Hz_40ms_envelope_correlation": 0.9909728785047891, "3000–6000 Hz_40ms_modulation_mae_db": 1.915980063116277, "mr_stft_db": 5.609073774581819, "band_mae_db": 0.2110171909603399, "rms_gain_error_db": -0.06777325005395159, "output_peak": 0.6347652118748225}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84000, "output_samples": 84000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.490561014810128, "spectral_convergence": 0.3434583981990579, "spectral_convergence_status": "ok", "amplitude_floor": 1.02822042353332e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.429888230340267, "spectral_convergence": 0.3386272153567612, "spectral_convergence_status": "ok", "amplitude_floor": 8.928413196720056e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.912350796815444, "spectral_convergence": 0.3461841662068641, "spectral_convergence_status": "ok", "amplitude_floor": 7.736392334055093e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 1.4784881078912855e-05, "output_power": 1.7547181397364462e-05, "delta_db": 0.7438953004802934, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002602315552138397, "output_power": 0.00025753618562140046, "delta_db": -0.04521698138948291, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005931268593866254, "output_power": 0.0005674412650783912, "delta_db": -0.1922667616163974, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005283196568297616, "output_power": 0.0005054206689073109, "delta_db": -0.19243771299565687, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003079318069923152, "output_power": 0.0003153429413072478, "delta_db": 0.10328563605864771, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00026143236338356946, "output_power": 0.0002507230711687597, "delta_db": -0.18165050064231994, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24312548219713306, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11600822306859257, "pitch": null, "output_rms_dbfs": -27.18055499491453, "output_peak": 0.623443276763886, "reference_rms_dbfs": -27.06454677184594, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6109333473219465}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00019312594202620684, "log_rms_mae_db": 0.810601371395359, "rms_envelope_nrmse": 0.07567412671212112, "envelope_correlation": 0.9946487607367713, "rms_gain_error_db": -0.11600762180737952, "reference_mean_power": 0.0019676998400409485, "output_mean_power": 0.001915834934283718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 4.0378624939923077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018862042103601667, "output_power": 0.0001916718418157322, "reference_active": true, "delta_db": 0.06969606155379146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.862312285918816e-05, "output_power": 0.00010532486767783194, "reference_active": true, "delta_db": 0.285521720219335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3789377405749525e-05, "output_power": 2.0418213743957395e-05, "reference_active": true, "delta_db": 1.7047308772649163}], "active_band_count": 3, "active_band_mae_db": 0.6866495530126809}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00018244538675777041, "log_rms_mae_db": 0.4033005633743024, "rms_envelope_nrmse": 0.04424936593021085, "envelope_correlation": 0.9981258767168418, "rms_gain_error_db": -0.1160072428157499, "reference_mean_power": 0.001979010526094432, "output_mean_power": 0.0019268476598511778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 3.3102005537920435e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017836230285954771, "output_power": 0.00018759937074033637, "reference_active": true, "delta_db": 0.2192830635495534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3320761072042273e-05, "output_power": 4.865390764777116e-05, "reference_active": true, "delta_db": 0.5042164861319841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.132775238046075e-07, "output_power": 1.3924366383475532e-06, "reference_active": true, "delta_db": 2.3353667203276}], "active_band_count": 3, "active_band_mae_db": 1.0196220900030457}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00014606965842361196, "log_rms_mae_db": 1.4949804710308017, "rms_envelope_nrmse": 0.09507138253769457, "envelope_correlation": 0.9929959650926111, "rms_gain_error_db": -0.15662313945899214, "reference_mean_power": 0.0007714288246159527, "output_mean_power": 0.0007441038445410047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 2.155891354220712e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.355627855804233e-05, "output_power": 8.159690320502799e-05, "reference_active": true, "delta_db": -0.1030541250029999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.117824128471733e-05, "output_power": 1.8740692387271452e-05, "reference_active": true, "delta_db": -0.5310425979006403}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3214088476749303e-06, "output_power": 3.1927012091399777e-06, "reference_active": true, "delta_db": -0.17164061486927573}], "active_band_count": 3, "active_band_mae_db": 0.268579112590972}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.0001346093108760191, "log_rms_mae_db": 0.8947428585677779, "rms_envelope_nrmse": 0.05484435705500137, "envelope_correlation": 0.9976961332350978, "rms_gain_error_db": -0.1566237921262813, "reference_mean_power": 0.0007758659430163304, "output_mean_power": 0.0007483836821532573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.945223621465206e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.346976557991944e-05, "output_power": 7.16009367731683e-05, "reference_active": true, "delta_db": -0.11189949813002352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.57892144723787e-06, "output_power": 8.558691895895307e-06, "reference_active": true, "delta_db": -0.4890921930126874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.32297115664878e-07, "output_power": 2.070821036684365e-07, "reference_active": true, "delta_db": -0.4990124921773983}], "active_band_count": 3, "active_band_mae_db": 0.366668061106703}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00012444218003818797, "log_rms_mae_db": 1.278916843951122, "rms_envelope_nrmse": 0.10020868857224646, "envelope_correlation": 0.9928919585934335, "rms_gain_error_db": -0.21155375270163662, "reference_mean_power": 0.0006875503513213645, "output_mean_power": 0.0006548610091919627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.3390858183888e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.254326674690693e-05, "output_power": 4.186129487486153e-05, "reference_active": true, "delta_db": -0.07018176755334914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4859458498211302e-05, "output_power": 1.5474174501497824e-05, "reference_active": true, "delta_db": 0.17604506720268906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.141405389451205e-06, "output_power": 3.472330019344785e-06, "reference_active": true, "delta_db": 0.43497010466276637}], "active_band_count": 3, "active_band_mae_db": 0.22706564647293484}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00011369765125964215, "log_rms_mae_db": 0.7638541236336317, "rms_envelope_nrmse": 0.05741261964572413, "envelope_correlation": 0.9978721951261779, "rms_gain_error_db": -0.2115540191055783, "reference_mean_power": 0.0006915015836222027, "output_mean_power": 0.0006586243411272249}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.214722623247201e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7827908938064025e-05, "output_power": 3.774288536095959e-05, "reference_active": true, "delta_db": -0.009772369416217381}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.619804659862373e-06, "output_power": 6.315074644414739e-06, "reference_active": true, "delta_db": 0.5065726821306307}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0150238495180175e-07, "output_power": 2.2393877686201186e-07, "reference_active": true, "delta_db": 0.45849111041373397}], "active_band_count": 3, "active_band_mae_db": 0.324945387320194}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00010848831624356859, "log_rms_mae_db": 1.402130783856686, "rms_envelope_nrmse": 0.1552215861808699, "envelope_correlation": 0.9822890117523269, "rms_gain_error_db": 0.181400244340079, "reference_mean_power": 0.00023818463717296146, "output_mean_power": 0.00024834405646968126}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.7158903163900278e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.877460217795428e-05, "output_power": 0.00010140037343683949, "reference_active": true, "delta_db": 0.5775081962636714}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.954888518804953e-05, "output_power": 5.053520953760767e-05, "reference_active": true, "delta_db": 1.0645982475974876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.296333400161532e-06, "output_power": 1.4360628859353537e-05, "reference_active": true, "delta_db": 2.9406878778703307}], "active_band_count": 3, "active_band_mae_db": 1.5275981072438298}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 7.397564116979367e-05, "log_rms_mae_db": 0.8851150701081815, "rms_envelope_nrmse": 0.10554109227960261, "envelope_correlation": 0.9912882048514507, "rms_gain_error_db": 0.18140215134648857, "reference_mean_power": 0.0002395532500233675, "output_mean_power": 0.0002497711551863928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.5691901914560294e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.470858674877874e-05, "output_power": 0.00011432713975685739, "reference_active": true, "delta_db": 0.8175998234917343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.833295648941501e-05, "output_power": 2.545690160967229e-05, "reference_active": true, "delta_db": 1.4257303632029439}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.840671065644946e-07, "output_power": 1.065930329306481e-06, "reference_active": true, "delta_db": 3.4282324715646215}], "active_band_count": 3, "active_band_mae_db": 1.890520886086433}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9013529946442753, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f7c52d7afcfd3bdf79e8cddaf9aac1592079831776b332b1369a2113095f4c22", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9917318457933375, "standard_deviation": 0.0009335084318464304, "minimum": 0.9898695621489412, "maximum": 0.9928919585934335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2968422559247927, "standard_deviation": 0.04457276955136965, "minimum": 1.2230438297885737, "maximum": 1.3505555458694685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.45059497599831455, "standard_deviation": 0.17243268621756375, "minimum": 0.22706564647293484, "maximum": 0.6643490965092301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10813298312797479, "standard_deviation": 0.0058291625895644985, "minimum": 0.10020868857224646, "maximum": 0.11965442870015504, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975289623020531, "standard_deviation": 0.0005463195142367597, "minimum": 0.9966313120400838, "maximum": 0.99828750327052, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7598375473564718, "standard_deviation": 0.042169650738155234, "minimum": 0.7008341663556902, "maximum": 0.8109797206584176, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.32884295386104717, "standard_deviation": 0.1594682681982137, "minimum": 0.15185281522842964, "maximum": 0.6066349023442287, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06183426218250505, "standard_deviation": 0.007443416534013982, "minimum": 0.05028563994641302, "maximum": 0.0720420540655562, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9841473060852857, "standard_deviation": 0.0019333421733850685, "minimum": 0.9810854603961764, "maximum": 0.9860833879636957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.425861765009206, "standard_deviation": 0.03035777220350685, "minimum": 1.3893990110677705, "maximum": 1.487071438351044, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9030621350065258, "standard_deviation": 0.4854121509216558, "minimum": 0.07251908977908059, "maximum": 1.5456897300658732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14383566097677866, "standard_deviation": 0.010243238985512425, "minimum": 0.1347049470626951, "maximum": 0.16107432773616273, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9938942685475214, "standard_deviation": 0.0017813261361031534, "minimum": 0.9909728785047891, "maximum": 0.9954507358541591, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8731235424083728, "standard_deviation": 0.03330139950948527, "minimum": 0.8272744541267222, "maximum": 0.9174482990765201, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9979864126994324, "standard_deviation": 0.625073268185091, "minimum": 0.14252376931235533, "maximum": 1.915980063116277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0859127698129766, "standard_deviation": 0.0135787659313111, "minimum": 0.07318209535323611, "maximum": 0.10877877867790928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933107730824123, "standard_deviation": 0.00119941662889354, "minimum": 0.9918162438674486, "maximum": 0.9948585223421025, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4710448878532556, "standard_deviation": 0.05277955600109524, "minimum": 1.378558794027208, "maximum": 1.52427422135686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2948548372176445, "standard_deviation": 0.11012459848210314, "minimum": 0.056820082485459895, "maximum": 0.4050014993233417, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09197179377068275, "standard_deviation": 0.008368978144699946, "minimum": 0.08081867458639105, "maximum": 0.10172461934381061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978763333201058, "standard_deviation": 0.0005559381396939674, "minimum": 0.997095367864716, "maximum": 0.9987492007017801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8790791969907925, "standard_deviation": 0.035333463255528226, "minimum": 0.8109887455096544, "maximum": 0.9293157196594143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3246564809749203, "standard_deviation": 0.16186753285071553, "minimum": 0.08923588206769988, "maximum": 0.544663934203038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.050918798358959426, "standard_deviation": 0.007030672301782317, "minimum": 0.03935458653528738, "maximum": 0.05931561962787092, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21852377804037457, "standard_deviation": 0.028602623484124463, "minimum": 0.16441145286816988, "maximum": 0.24451912296500203, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.611053997997083, "standard_deviation": 0.004092414301468078, "minimum": 5.605190093643338, "maximum": 5.616986609993727, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6303671106375734, "standard_deviation": 0.010712104422478103, "minimum": 0.6142072132197359, "maximum": 0.6498376788391262, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10442848787795977, "standard_deviation": 0.047756580533521796, "minimum": -0.15566221459633942, "maximum": -0.008840657586965953, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994629945507524, "standard_deviation": 0.0005310993378982275, "minimum": 0.993673307334501, "maximum": 0.9952756626690705, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8150044378344028, "standard_deviation": 0.01467844851442485, "minimum": 0.7914902782107706, "maximum": 0.8431139108821566, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5358641433470336, "standard_deviation": 0.18777174724315346, "minimum": 0.11372819915767311, "maximum": 0.6866495530126809, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07576879934821776, "standard_deviation": 0.00355740674329512, "minimum": 0.07129671094348769, "maximum": 0.08149738991145512, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982502161205389, "standard_deviation": 0.0004925013255800974, "minimum": 0.9972139880909485, "maximum": 0.9988197251442861, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4176718224546825, "standard_deviation": 0.019574763493421642, "minimum": 0.39457274401414744, "maximum": 0.4445043045879586, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.581128542998005, "standard_deviation": 0.30200379358154544, "minimum": 0.154762161719924, "maximum": 1.0196220900030457, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0425952240615569, "standard_deviation": 0.005095615061851898, "minimum": 0.035462410910718355, "maximum": 0.05201825288982544, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1995-1836-0000", "language": "en", "style": "normal", "speaker": "librispeech_1995", "source_utterance_id": "1995-1836-0000", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1995-1836-0000.wav", "source_file_sha256": "2abe5f6447a84974141557fbe150660111440e2ff4d2277563c56f0edb73e396", "decoded_pcm_sha256": "3feb5dcb82941d2a5e9f89efbf0e32e0138e5eec693c118bb9e8031d23d08925", "original_source_sha256": "8120c55c7231b13d3d497b04b2fe3d999cfdac785a0134eb0fc9e41aa90ce591", "corpus_entry_id": "librispeech_1995-1836-0000"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.015101909637451, "predicted_mos": 4.015101909637451, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d82b4a3d4d2f4241d960b592dc1760e20d827057b4dbb01a317f3058a1918c05", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.2669377390011505, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1995-1836-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d82b4a3d4d2f4241d960b592dc1760e20d827057b4dbb01a317f3058a1918c05", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "9b4757ae23cce8e56532839de5fdf9b43c346388e66b2494c20bf772f7621084"}, "rows": [{"case": "librispeech_1995-1836-0000", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.1125595569610596, "predicted_mos": 3.1125595569610596, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1ac507d378afff82f81ad69a2b6800f02ba8c23e42b6fcbc18f6b9148604db8e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.203924967998319, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1995-1836-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1ac507d378afff82f81ad69a2b6800f02ba8c23e42b6fcbc18f6b9148604db8e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6263325371671522, "全帯域_10ms_rms_envelope_nrmse": 0.04876131807310991, "全帯域_10ms_envelope_correlation": 0.9968273130650376, "全帯域_10ms_modulation_mae_db": 0.38068646068498446, "全帯域_40ms_log_rms_mae_db": 0.3358122124543343, "全帯域_40ms_rms_envelope_nrmse": 0.02737526168887273, "全帯域_40ms_envelope_correlation": 0.998953132915758, "全帯域_40ms_modulation_mae_db": 0.5269199789889855, "300–1000 Hz_10ms_log_rms_mae_db": 1.4651104722213808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07875114741600199, "300–1000 Hz_10ms_envelope_correlation": 0.9941138740844263, "300–1000 Hz_10ms_modulation_mae_db": 0.5076870806669684, "300–1000 Hz_40ms_log_rms_mae_db": 0.968051275255213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046808628997082456, "300–1000 Hz_40ms_envelope_correlation": 0.997897035167159, "300–1000 Hz_40ms_modulation_mae_db": 0.4861547449545946, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2602257672173702, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10389570577489457, "1000–3000 Hz_10ms_envelope_correlation": 0.9908354221763729, "1000–3000 Hz_10ms_modulation_mae_db": 0.4333243683911183, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7933067975437885, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06219766925795181, "1000–3000 Hz_40ms_envelope_correlation": 0.9968940277415519, "1000–3000 Hz_40ms_modulation_mae_db": 0.387740229338463, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4065371378030032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15695481741667605, "3000–6000 Hz_10ms_envelope_correlation": 0.9845319670703453, "3000–6000 Hz_10ms_modulation_mae_db": 1.3794790735380138, "3000–6000 Hz_40ms_log_rms_mae_db": 0.803441120940834, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08835168928013182, "3000–6000 Hz_40ms_envelope_correlation": 0.9956506345843184, "3000–6000 Hz_40ms_modulation_mae_db": 1.7325442769534514, "mr_stft_db": 4.99627513854557, "band_mae_db": 0.2162682806486402, "rms_gain_error_db": -0.07038535121395766, "output_peak": 0.29534113824513697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6001081650722166, "全帯域_10ms_rms_envelope_nrmse": 0.048392579107184544, "全帯域_10ms_envelope_correlation": 0.996858394154245, "全帯域_10ms_modulation_mae_db": 0.5142734308915823, "全帯域_40ms_log_rms_mae_db": 0.31736363559220976, "全帯域_40ms_rms_envelope_nrmse": 0.028305406347829406, "全帯域_40ms_envelope_correlation": 0.9988545016875859, "全帯域_40ms_modulation_mae_db": 0.6808987811246615, "300–1000 Hz_10ms_log_rms_mae_db": 1.5095104844369462, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07210292397174552, "300–1000 Hz_10ms_envelope_correlation": 0.9950042786666522, "300–1000 Hz_10ms_modulation_mae_db": 0.6467755786472117, "300–1000 Hz_40ms_log_rms_mae_db": 0.9945506152742175, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041449008337590396, "300–1000 Hz_40ms_envelope_correlation": 0.99824883989363, "300–1000 Hz_40ms_modulation_mae_db": 0.8202633506143338, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3258159890312204, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11424853804752902, "1000–3000 Hz_10ms_envelope_correlation": 0.9897343426321672, "1000–3000 Hz_10ms_modulation_mae_db": 0.8021793570526444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8594585605062134, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07746528122154975, "1000–3000 Hz_40ms_envelope_correlation": 0.9959765761325963, "1000–3000 Hz_40ms_modulation_mae_db": 0.8912916962553972, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4189346417166475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17691771370723416, "3000–6000 Hz_10ms_envelope_correlation": 0.9794800889841765, "3000–6000 Hz_10ms_modulation_mae_db": 1.4164995475335624, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8627046596368843, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1138221259340431, "3000–6000 Hz_40ms_envelope_correlation": 0.9910926817865036, "3000–6000 Hz_40ms_modulation_mae_db": 1.8420466203080281, "mr_stft_db": 4.999195142259526, "band_mae_db": 0.1669456187861942, "rms_gain_error_db": -0.07252630241324552, "output_peak": 0.2982926048661344}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6187703077667897, "全帯域_10ms_rms_envelope_nrmse": 0.04984433734251054, "全帯域_10ms_envelope_correlation": 0.9966777445896143, "全帯域_10ms_modulation_mae_db": 0.345373899172146, "全帯域_40ms_log_rms_mae_db": 0.3472681232335334, "全帯域_40ms_rms_envelope_nrmse": 0.029611022893191827, "全帯域_40ms_envelope_correlation": 0.9987575757933119, "全帯域_40ms_modulation_mae_db": 0.3173266709252207, "300–1000 Hz_10ms_log_rms_mae_db": 1.4115375760913547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07205191984937233, "300–1000 Hz_10ms_envelope_correlation": 0.9950508260210864, "300–1000 Hz_10ms_modulation_mae_db": 0.28298680163255535, "300–1000 Hz_40ms_log_rms_mae_db": 0.9519434747885864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04423031486706473, "300–1000 Hz_40ms_envelope_correlation": 0.9980488450320204, "300–1000 Hz_40ms_modulation_mae_db": 0.1862108831449971, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2760779434666303, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11453278053312833, "1000–3000 Hz_10ms_envelope_correlation": 0.9888965452179188, "1000–3000 Hz_10ms_modulation_mae_db": 0.5694437477091377, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8169097574261827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06897888578787072, "1000–3000 Hz_40ms_envelope_correlation": 0.9962493907106166, "1000–3000 Hz_40ms_modulation_mae_db": 0.5287717765197961, "3000–6000 Hz_10ms_log_rms_mae_db": 1.369097945441805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1697561152556881, "3000–6000 Hz_10ms_envelope_correlation": 0.9814101820085819, "3000–6000 Hz_10ms_modulation_mae_db": 0.3016031336287294, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7823268811938059, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09335944372040193, "3000–6000 Hz_40ms_envelope_correlation": 0.9940467684688058, "3000–6000 Hz_40ms_modulation_mae_db": 0.3617560309242121, "mr_stft_db": 4.977084297998742, "band_mae_db": 0.1276440277500625, "rms_gain_error_db": -0.07309306412963655, "output_peak": 0.30518799890332604}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6011930921692863, "全帯域_10ms_rms_envelope_nrmse": 0.04837906332469577, "全帯域_10ms_envelope_correlation": 0.9968473602847784, "全帯域_10ms_modulation_mae_db": 0.3845494460047411, "全帯域_40ms_log_rms_mae_db": 0.34039568327350617, "全帯域_40ms_rms_envelope_nrmse": 0.027363771001347077, "全帯域_40ms_envelope_correlation": 0.9989251580184237, "全帯域_40ms_modulation_mae_db": 0.49865643279637, "300–1000 Hz_10ms_log_rms_mae_db": 1.4520569680466267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0743207678364306, "300–1000 Hz_10ms_envelope_correlation": 0.9947059224469542, "300–1000 Hz_10ms_modulation_mae_db": 0.27314558170226466, "300–1000 Hz_40ms_log_rms_mae_db": 0.9758927811317796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0418534699194542, "300–1000 Hz_40ms_envelope_correlation": 0.998259488840827, "300–1000 Hz_40ms_modulation_mae_db": 0.5367667107358541, "1000–3000 Hz_10ms_log_rms_mae_db": 1.273275101358776, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11277587814732991, "1000–3000 Hz_10ms_envelope_correlation": 0.9892474841045623, "1000–3000 Hz_10ms_modulation_mae_db": 0.5253313901311277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7959659010978175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060926136949950326, "1000–3000 Hz_40ms_envelope_correlation": 0.9972707784910726, "1000–3000 Hz_40ms_modulation_mae_db": 0.6866827966198671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4502192454974823, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18431849149444657, "3000–6000 Hz_10ms_envelope_correlation": 0.9781825474664846, "3000–6000 Hz_10ms_modulation_mae_db": 1.3103237646244332, "3000–6000 Hz_40ms_log_rms_mae_db": 0.835829797850473, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10882744786202636, "3000–6000 Hz_40ms_envelope_correlation": 0.9927880673728097, "3000–6000 Hz_40ms_modulation_mae_db": 1.7106775759119293, "mr_stft_db": 4.977672904205426, "band_mae_db": 0.18737243620638688, "rms_gain_error_db": -0.06695813627925609, "output_peak": 0.3034928788090354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6365568538736478, "全帯域_10ms_rms_envelope_nrmse": 0.05058817282516019, "全帯域_10ms_envelope_correlation": 0.9965075753010612, "全帯域_10ms_modulation_mae_db": 0.4684728412818871, "全帯域_40ms_log_rms_mae_db": 0.3201460807888821, "全帯域_40ms_rms_envelope_nrmse": 0.02636637827032791, "全帯域_40ms_envelope_correlation": 0.9989537149746237, "全帯域_40ms_modulation_mae_db": 0.6883931610534567, "300–1000 Hz_10ms_log_rms_mae_db": 1.4729126323650386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07894776340196268, "300–1000 Hz_10ms_envelope_correlation": 0.9940137095618712, "300–1000 Hz_10ms_modulation_mae_db": 0.5046620581349086, "300–1000 Hz_40ms_log_rms_mae_db": 0.99942158311174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04502576651132483, "300–1000 Hz_40ms_envelope_correlation": 0.9979567350625467, "300–1000 Hz_40ms_modulation_mae_db": 0.8281436936515888, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2734669987934721, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10778269709700246, "1000–3000 Hz_10ms_envelope_correlation": 0.9900146698428851, "1000–3000 Hz_10ms_modulation_mae_db": 0.44967444057827044, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7983796049756271, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06520518228788945, "1000–3000 Hz_40ms_envelope_correlation": 0.9964125591004402, "1000–3000 Hz_40ms_modulation_mae_db": 0.7326678789974502, "3000–6000 Hz_10ms_log_rms_mae_db": 1.378417609751104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1557509405520334, "3000–6000 Hz_10ms_envelope_correlation": 0.9841402934923665, "3000–6000 Hz_10ms_modulation_mae_db": 1.0648519463771595, "3000–6000 Hz_40ms_log_rms_mae_db": 0.80490598482988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08409793455564131, "3000–6000 Hz_40ms_envelope_correlation": 0.9952028606298196, "3000–6000 Hz_40ms_modulation_mae_db": 1.5350825221018016, "mr_stft_db": 4.9833080030461225, "band_mae_db": 0.1573634423334623, "rms_gain_error_db": -0.0368858250695729, "output_peak": 0.30822434259288267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6285362992820736, "全帯域_10ms_rms_envelope_nrmse": 0.04559553814917304, "全帯域_10ms_envelope_correlation": 0.9972792029330382, "全帯域_10ms_modulation_mae_db": 0.6542475966401708, "全帯域_40ms_log_rms_mae_db": 0.33283370334624024, "全帯域_40ms_rms_envelope_nrmse": 0.02588315119521115, "全帯域_40ms_envelope_correlation": 0.9991339530274537, "全帯域_40ms_modulation_mae_db": 0.8384192768255181, "300–1000 Hz_10ms_log_rms_mae_db": 1.4242858473993745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0668551248378378, "300–1000 Hz_10ms_envelope_correlation": 0.995790167412851, "300–1000 Hz_10ms_modulation_mae_db": 0.5373010104427256, "300–1000 Hz_40ms_log_rms_mae_db": 0.9678096389671631, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038706819955544924, "300–1000 Hz_40ms_envelope_correlation": 0.9985758230193489, "300–1000 Hz_40ms_modulation_mae_db": 0.8466377952992815, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2279563357101408, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10688963089862855, "1000–3000 Hz_10ms_envelope_correlation": 0.9907700749241898, "1000–3000 Hz_10ms_modulation_mae_db": 0.7554814526455774, "1000–3000 Hz_40ms_log_rms_mae_db": 0.775587274553998, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06545657348764591, "1000–3000 Hz_40ms_envelope_correlation": 0.9971017390522484, "1000–3000 Hz_40ms_modulation_mae_db": 0.9552078831594638, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3981711977633002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17806937228465064, "3000–6000 Hz_10ms_envelope_correlation": 0.9800774230126854, "3000–6000 Hz_10ms_modulation_mae_db": 1.9470429889676233, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7882435671094061, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10812696964273774, "3000–6000 Hz_40ms_envelope_correlation": 0.9933644490140403, "3000–6000 Hz_40ms_modulation_mae_db": 2.3872869934731304, "mr_stft_db": 4.982542237749321, "band_mae_db": 0.1999198040430609, "rms_gain_error_db": -0.09576703784821428, "output_peak": 0.3097024192754772}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6170662433979249, "全帯域_10ms_rms_envelope_nrmse": 0.047323297077754956, "全帯域_10ms_envelope_correlation": 0.9969926894628421, "全帯域_10ms_modulation_mae_db": 0.5447307269317844, "全帯域_40ms_log_rms_mae_db": 0.3261873430499878, "全帯域_40ms_rms_envelope_nrmse": 0.024199868191489622, "全帯域_40ms_envelope_correlation": 0.9991826097024389, "全帯域_40ms_modulation_mae_db": 0.62421432671175, "300–1000 Hz_10ms_log_rms_mae_db": 1.4272677165145775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07545365952818849, "300–1000 Hz_10ms_envelope_correlation": 0.9945307957901254, "300–1000 Hz_10ms_modulation_mae_db": 0.7437327855893904, "300–1000 Hz_40ms_log_rms_mae_db": 0.9998141786405469, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04360055737233193, "300–1000 Hz_40ms_envelope_correlation": 0.9980843066226109, "300–1000 Hz_40ms_modulation_mae_db": 1.3580224243494043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2750921176685714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10857899268722636, "1000–3000 Hz_10ms_envelope_correlation": 0.9905241769284195, "1000–3000 Hz_10ms_modulation_mae_db": 0.7111117787375821, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8160243472792156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06795822080807845, "1000–3000 Hz_40ms_envelope_correlation": 0.9968916637031546, "1000–3000 Hz_40ms_modulation_mae_db": 0.6590269513376363, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3893052600918072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17915345094396468, "3000–6000 Hz_10ms_envelope_correlation": 0.979146369737735, "3000–6000 Hz_10ms_modulation_mae_db": 1.705437933470735, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8110474442460626, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11697298831207407, "3000–6000 Hz_40ms_envelope_correlation": 0.9909625388781991, "3000–6000 Hz_40ms_modulation_mae_db": 2.1007021111413517, "mr_stft_db": 4.98548748005243, "band_mae_db": 0.1533364355259757, "rms_gain_error_db": -0.07088443281967105, "output_peak": 0.3064982422052319}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6364012864905988, "全帯域_10ms_rms_envelope_nrmse": 0.04905628351193415, "全帯域_10ms_envelope_correlation": 0.996763352159315, "全帯域_10ms_modulation_mae_db": 0.6632663688990057, "全帯域_40ms_log_rms_mae_db": 0.33248100303168976, "全帯域_40ms_rms_envelope_nrmse": 0.02747666456196674, "全帯域_40ms_envelope_correlation": 0.9989196198370235, "全帯域_40ms_modulation_mae_db": 1.0670697582404725, "300–1000 Hz_10ms_log_rms_mae_db": 1.4517265181236838, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07781935746738022, "300–1000 Hz_10ms_envelope_correlation": 0.9941657271932564, "300–1000 Hz_10ms_modulation_mae_db": 0.9129731961434778, "300–1000 Hz_40ms_log_rms_mae_db": 1.0105091414369352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04417943212278808, "300–1000 Hz_40ms_envelope_correlation": 0.998011498718548, "300–1000 Hz_40ms_modulation_mae_db": 1.2053301997447665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.227883663848237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10400623092853059, "1000–3000 Hz_10ms_envelope_correlation": 0.9917105809425267, "1000–3000 Hz_10ms_modulation_mae_db": 0.6432080357795239, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8013414346044583, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06674110491071619, "1000–3000 Hz_40ms_envelope_correlation": 0.9974126252413472, "1000–3000 Hz_40ms_modulation_mae_db": 0.7591831735864792, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3870800764092206, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16618543818083897, "3000–6000 Hz_10ms_envelope_correlation": 0.9853569245256512, "3000–6000 Hz_10ms_modulation_mae_db": 0.38544611471595963, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8235042173094315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1101459295798198, "3000–6000 Hz_40ms_envelope_correlation": 0.9946174979148531, "3000–6000 Hz_40ms_modulation_mae_db": 0.4962574746633421, "mr_stft_db": 4.98786033705025, "band_mae_db": 0.130259613259041, "rms_gain_error_db": -0.06738953516600896, "output_peak": 0.3055205311382973}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.1051185612971866, "spectral_convergence": 0.13362085707823618, "spectral_convergence_status": "ok", "amplitude_floor": 1.1304927744850054e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.86797743751424, "spectral_convergence": 0.1289781972511515, "spectral_convergence_status": "ok", "amplitude_floor": 1.0019114511304393e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.015729416825285, "spectral_convergence": 0.13312540889902713, "spectral_convergence_status": "ok", "amplitude_floor": 8.812050137930174e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0013595771819321111, "output_power": 0.0013656747371314087, "delta_db": 0.019434085902834266, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002257060285350754, "output_power": 0.0022178279784682564, "delta_db": -0.07615301188855232, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006976179772550826, "output_power": 0.0006697017795839616, "delta_db": -0.1773621063429257, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001064894568705692, "output_power": 0.00010200025790026049, "delta_db": -0.18705342158503363, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.7425121182630722e-05, "output_power": 2.4183023113583566e-05, "delta_db": -0.5463796429462174, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.591013773579702e-05, "output_power": 5.228388137517004e-05, "delta_db": -0.29122741522627765, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2162682806486402, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07038535121395766, "pitch": null, "output_rms_dbfs": -23.534324239535348, "output_peak": 0.29534113824513697, "reference_rms_dbfs": -23.46393888832139, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.99627513854557}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016450226550328756, "log_rms_mae_db": 0.6263325371671522, "rms_envelope_nrmse": 0.04876131807310991, "envelope_correlation": 0.9968273130650376, "rms_gain_error_db": -0.07035498329093377, "reference_mean_power": 0.004505860165227205, "output_mean_power": 0.004433454047452401}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3773993983732174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011178733037365022, "output_power": 0.0010820635607175497, "reference_active": true, "delta_db": -0.1413981266170712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011348254393479664, "output_power": 0.00010382185338835927, "reference_active": true, "delta_db": -0.38640285326915874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1976543085088728e-05, "output_power": 1.9078030494332007e-05, "reference_active": true, "delta_db": -0.6142584021687234}], "active_band_count": 3, "active_band_mae_db": 0.38068646068498446}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014948818281718847, "log_rms_mae_db": 0.3358122124543343, "rms_envelope_nrmse": 0.02737526168887273, "envelope_correlation": 0.998953132915758, "rms_gain_error_db": -0.0713238767310491, "reference_mean_power": 0.004512140106290299, "output_mean_power": 0.004438642720397499}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1895509906107845e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010130343468129068, "output_power": 0.0009822047531953017, "reference_active": true, "delta_db": -0.1342213882712644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.38575581348341e-05, "output_power": 4.86015204065603e-05, "reference_active": true, "delta_db": -0.44596803349010516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.096976553551289e-06, "output_power": 1.665468883716105e-06, "reference_active": true, "delta_db": -1.0005705152055868}], "active_band_count": 3, "active_band_mae_db": 0.5269199789889855}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012259006245866322, "log_rms_mae_db": 1.4651104722213808, "rms_envelope_nrmse": 0.07875114741600199, "envelope_correlation": 0.9941138740844263, "rms_gain_error_db": -0.11305978728297923, "reference_mean_power": 0.001387125934818555, "output_mean_power": 0.0013514809020178304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.931300849598275e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005535817649712056, "output_power": 0.0005226179971346352, "reference_active": true, "delta_db": -0.2499741475768792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.308179193682493e-05, "output_power": 6.262954134598913e-05, "reference_active": true, "delta_db": -0.6702995629063917}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.782827787780852e-06, "output_power": 6.643837664930993e-06, "reference_active": true, "delta_db": 0.6027875315176344}], "active_band_count": 3, "active_band_mae_db": 0.5076870806669684}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001104267603656376, "log_rms_mae_db": 0.968051275255213, "rms_envelope_nrmse": 0.046808628997082456, "envelope_correlation": 0.997897035167159, "rms_gain_error_db": -0.11531127537026169, "reference_mean_power": 0.0013888741987160204, "output_mean_power": 0.0013524828989371202}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.150356442217087e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005149581742784205, "output_power": 0.000486110658314339, "reference_active": true, "delta_db": -0.250368129978138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.177504958330739e-05, "output_power": 2.7423855076370005e-05, "reference_active": true, "delta_db": -0.6395773173639512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.976808620230845e-07, "output_power": 6.998054071835301e-07, "reference_active": true, "delta_db": -0.5685187875216945}], "active_band_count": 3, "active_band_mae_db": 0.4861547449545946}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.111137261258481e-05, "log_rms_mae_db": 1.2602257672173702, "rms_envelope_nrmse": 0.10389570577489457, "envelope_correlation": 0.9908354221763729, "rms_gain_error_db": -0.22619194822446043, "reference_mean_power": 0.00011418247196706928, "output_mean_power": 0.000108387761270839}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.04035362622924e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.425624857591467e-05, "output_power": 4.064402820286506e-05, "reference_active": true, "delta_db": -0.36977853706950536}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.422825900093454e-06, "output_power": 2.947272171180076e-06, "reference_active": true, "delta_db": -0.6496456617005876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0991841496467888e-06, "output_power": 1.0304230535421953e-06, "reference_active": true, "delta_db": -0.28054890640326213}], "active_band_count": 3, "active_band_mae_db": 0.4333243683911183}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.646632245602786e-05, "log_rms_mae_db": 0.7933067975437885, "rms_envelope_nrmse": 0.06219766925795181, "envelope_correlation": 0.9968940277415519, "rms_gain_error_db": -0.2262466139107539, "reference_mean_power": 0.0001146281153210772, "output_mean_power": 0.00010880941879999721}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.508716025767634e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.069122651510939e-05, "output_power": 3.728714602518841e-05, "reference_active": true, "delta_db": -0.3794163701679717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6543344614858788e-06, "output_power": 1.40177750536266e-06, "reference_active": true, "delta_db": -0.7194423015671378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6696400281137344e-08, "output_power": 5.754289234461068e-08, "reference_active": true, "delta_db": 0.06436201628027947}], "active_band_count": 3, "active_band_mae_db": 0.387740229338463}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.742254313282893e-05, "log_rms_mae_db": 1.4065371378030032, "rms_envelope_nrmse": 0.15695481741667605, "envelope_correlation": 0.9845319670703453, "rms_gain_error_db": -0.36498675434959965, "reference_mean_power": 4.027652017807587e-05, "output_mean_power": 3.702996194836022e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.067374336554813e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.634873883610167e-05, "output_power": 1.534327602457111e-05, "reference_active": true, "delta_db": -0.2756615818136408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.842264617940254e-06, "output_power": 4.998838165713632e-06, "reference_active": true, "delta_db": -0.6771214693326005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8971365027113356e-06, "output_power": 9.110310173723597e-07, "reference_active": true, "delta_db": -3.1856541694678}], "active_band_count": 3, "active_band_mae_db": 1.3794790735380138}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.2287916076198396e-05, "log_rms_mae_db": 0.803441120940834, "rms_envelope_nrmse": 0.08835168928013182, "envelope_correlation": 0.9956506345843184, "rms_gain_error_db": -0.36534533263563534, "reference_mean_power": 4.047626840683265e-05, "output_mean_power": 3.721053668644104e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8536334265621397e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7722714720556858e-05, "output_power": 1.6030554397342586e-05, "reference_active": true, "delta_db": -0.4358170461195376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2765957035699146e-06, "output_power": 2.501197894895604e-06, "reference_active": true, "delta_db": -1.1727480296188346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7512938438165208e-07, "output_power": 7.663942637772824e-08, "reference_active": true, "delta_db": -3.5890677551219823}], "active_band_count": 3, "active_band_mae_db": 1.7325442769534514}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9589425828782149, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "efeddb0a71dd208a431bc3b124c5d627edebaf99a25e3ae37e37d22110e010f5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9902166620961304, "standard_deviation": 0.000923575054883874, "minimum": 0.9888965452179188, "maximum": 0.9917105809425267, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2674742396368024, "standard_deviation": 0.03113793607521434, "minimum": 1.227883663848237, "maximum": 1.3258159890312204, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6112193213781227, "standard_deviation": 0.13894725449873677, "minimum": 0.4333243683911183, "maximum": 0.8021793570526444, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10908880676428373, "standard_deviation": 0.004300452717883056, "minimum": 0.10389570577489457, "maximum": 0.11453278053312833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967761700216284, "standard_deviation": 0.0005115797581951529, "minimum": 0.9959765761325963, "maximum": 0.9974126252413472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8071217097484127, "standard_deviation": 0.024881918923158833, "minimum": 0.775587274553998, "maximum": 0.8594585605062134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7000715482268192, "standard_deviation": 0.18321524484141968, "minimum": 0.387740229338463, "maximum": 0.9552078831594638, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06686613183895658, "standard_deviation": 0.005068638549819894, "minimum": 0.060926136949950326, "maximum": 0.07746528122154975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9815407245372533, "standard_deviation": 0.002769308132430684, "minimum": 0.9781825474664846, "maximum": 0.9853569245256512, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3997203893092962, "standard_deviation": 0.025712656523622562, "minimum": 1.369097945441805, "maximum": 1.4502192454974823, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.188835562857027, "standard_deviation": 0.584646371456332, "minimum": 0.3016031336287294, "maximum": 1.9470429889676233, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17088829247944157, "standard_deviation": 0.010578176537813556, "minimum": 0.1557509405520334, "maximum": 0.18431849149444657, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9934656873311687, "standard_deviation": 0.0017656408781934405, "minimum": 0.9909625388781991, "maximum": 0.9956506345843184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8140004591395972, "standard_deviation": 0.02620933041073952, "minimum": 0.7823268811938059, "maximum": 0.8627046596368843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5207942006846558, "standard_deviation": 0.723527962804782, "minimum": 0.3617560309242121, "maximum": 2.3872869934731304, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10296306611085952, "standard_deviation": 0.012469749599129694, "minimum": 0.08409793455564131, "maximum": 0.11697298831207407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9946719126471528, "standard_deviation": 0.0006007696163363629, "minimum": 0.9940137095618712, "maximum": 0.995790167412851, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4518010268998727, "standard_deviation": 0.03147200038844161, "minimum": 1.4115375760913547, "maximum": 1.5095104844369462, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5511580116199378, "standard_deviation": 0.2173175134321258, "minimum": 0.27314558170226466, "maximum": 0.9129731961434778, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07453783303861496, "standard_deviation": 0.004140157238844948, "minimum": 0.0668551248378378, "maximum": 0.07894776340196268, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981353215445864, "standard_deviation": 0.0002192297598626018, "minimum": 0.997897035167159, "maximum": 0.9985758230193489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9834990860757726, "standard_deviation": 0.020388349732928187, "minimum": 0.9519434747885864, "maximum": 1.0105091414369352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7834412253118526, "standard_deviation": 0.38198606859980966, "minimum": 0.1862108831449971, "maximum": 1.3580224243494043, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04323174976039769, "standard_deviation": 0.002495614049907852, "minimum": 0.038706819955544924, "maximum": 0.046808628997082456, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16738870731910296, "standard_deviation": 0.031855056003665005, "minimum": 0.1276440277500625, "maximum": 0.2162682806486402, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.986178192613424, "standard_deviation": 0.008028458267636953, "minimum": 4.977084297998742, "maximum": 4.999195142259526, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3040325195044402, "standard_deviation": 0.00490081894394177, "minimum": 0.29534113824513697, "maximum": 0.3097024192754772, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06923621061744537, "standard_deviation": 0.015985570525302382, "minimum": -0.09576703784821428, "maximum": -0.0368858250695729, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968442039937415, "standard_deviation": 0.00022656858536250255, "minimum": 0.9965075753010612, "maximum": 0.9972792029330382, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6206205981524613, "standard_deviation": 0.01420524980595751, "minimum": 0.6001081650722166, "maximum": 0.6365568538736478, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.49445009631328773, "standard_deviation": 0.12240105594332926, "minimum": 0.345373899172146, "maximum": 0.6632663688990057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04849257367644039, "standard_deviation": 0.0015302520452859427, "minimum": 0.04559553814917304, "maximum": 0.05058817282516019, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989600332445774, "standard_deviation": 0.00013879654151015364, "minimum": 0.9987575757933119, "maximum": 0.9991826097024389, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.33156097309629795, "standard_deviation": 0.010043093300294001, "minimum": 0.31736363559220976, "maximum": 0.3472681232335334, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6552372983333044, "standard_deviation": 0.2271086442302902, "minimum": 0.3173266709252207, "maximum": 1.0670697582404725, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027072690518779557, "standard_deviation": 0.0016243319798285318, "minimum": 0.024199868191489622, "maximum": 0.029611022893191827, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0000", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.965397357940674, "predicted_mos": 2.965397357940674, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a9ede858df0f57e0dd7c2004492234ee30979abd698c76b9fee29a0e3c86ff86", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.106842999812216, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6341889784684466, "全帯域_10ms_rms_envelope_nrmse": 0.04932647329098063, "全帯域_10ms_envelope_correlation": 0.9967657433342425, "全帯域_10ms_modulation_mae_db": 0.43100446026375977, "全帯域_40ms_log_rms_mae_db": 0.3418957664365321, "全帯域_40ms_rms_envelope_nrmse": 0.02749578540802759, "全帯域_40ms_envelope_correlation": 0.9989575999220605, "全帯域_40ms_modulation_mae_db": 0.5397633359621431, "300–1000 Hz_10ms_log_rms_mae_db": 1.4798943653955228, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0799561947010954, "300–1000 Hz_10ms_envelope_correlation": 0.9939556760597509, "300–1000 Hz_10ms_modulation_mae_db": 0.463630497671309, "300–1000 Hz_40ms_log_rms_mae_db": 0.9450783287122855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04656448123294118, "300–1000 Hz_40ms_envelope_correlation": 0.9979449671284811, "300–1000 Hz_40ms_modulation_mae_db": 0.5843215962954779, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2138423163143073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10289618523413925, "1000–3000 Hz_10ms_envelope_correlation": 0.9911962282647923, "1000–3000 Hz_10ms_modulation_mae_db": 0.6611894684865606, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7635070644505011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06266691223675476, "1000–3000 Hz_40ms_envelope_correlation": 0.9970384708408786, "1000–3000 Hz_40ms_modulation_mae_db": 0.7836546444894618, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3164414955917947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1531178402304001, "3000–6000 Hz_10ms_envelope_correlation": 0.9853992119526325, "3000–6000 Hz_10ms_modulation_mae_db": 1.2806375733967377, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7812447439263802, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0856871751779555, "3000–6000 Hz_40ms_envelope_correlation": 0.9961022894018339, "3000–6000 Hz_40ms_modulation_mae_db": 1.7129128698874343, "mr_stft_db": 4.921444436235896, "band_mae_db": 0.2488321451693262, "rms_gain_error_db": -0.07846124268181118, "output_peak": 0.29504993495429954}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5995172604900383, "全帯域_10ms_rms_envelope_nrmse": 0.04916827283164799, "全帯域_10ms_envelope_correlation": 0.9967501356881329, "全帯域_10ms_modulation_mae_db": 0.5641480039633812, "全帯域_40ms_log_rms_mae_db": 0.3234060531789668, "全帯域_40ms_rms_envelope_nrmse": 0.02882295214975359, "全帯域_40ms_envelope_correlation": 0.998806465554459, "全帯域_40ms_modulation_mae_db": 0.7372469090261163, "300–1000 Hz_10ms_log_rms_mae_db": 1.476305177030088, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07400883897415285, "300–1000 Hz_10ms_envelope_correlation": 0.9947395332627548, "300–1000 Hz_10ms_modulation_mae_db": 0.6641018531401867, "300–1000 Hz_40ms_log_rms_mae_db": 0.9529683802333387, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042544872452226754, "300–1000 Hz_40ms_envelope_correlation": 0.9981493894856562, "300–1000 Hz_40ms_modulation_mae_db": 0.9047466949137001, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2793801539271517, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11285058792024284, "1000–3000 Hz_10ms_envelope_correlation": 0.9903103453407095, "1000–3000 Hz_10ms_modulation_mae_db": 0.7271616605399891, "1000–3000 Hz_40ms_log_rms_mae_db": 0.832357820078948, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07772073665241504, "1000–3000 Hz_40ms_envelope_correlation": 0.9962474545155428, "1000–3000 Hz_40ms_modulation_mae_db": 0.7388962113879435, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3392704340946417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17546356887133774, "3000–6000 Hz_10ms_envelope_correlation": 0.979990173998664, "3000–6000 Hz_10ms_modulation_mae_db": 1.5702442348996672, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8479655601668721, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11661416052335782, "3000–6000 Hz_40ms_envelope_correlation": 0.9909853738583446, "3000–6000 Hz_40ms_modulation_mae_db": 2.066837724074176, "mr_stft_db": 4.921119057305484, "band_mae_db": 0.20811297097533624, "rms_gain_error_db": -0.07197428482716983, "output_peak": 0.29790366394254264}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6309844430268059, "全帯域_10ms_rms_envelope_nrmse": 0.050646633633977714, "全帯域_10ms_envelope_correlation": 0.9965630182358781, "全帯域_10ms_modulation_mae_db": 0.31383793428980616, "全帯域_40ms_log_rms_mae_db": 0.35451583193758307, "全帯域_40ms_rms_envelope_nrmse": 0.030686911535581098, "全帯域_40ms_envelope_correlation": 0.9986533361727328, "全帯域_40ms_modulation_mae_db": 0.2729539136700208, "300–1000 Hz_10ms_log_rms_mae_db": 1.437781716359455, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07420748771747036, "300–1000 Hz_10ms_envelope_correlation": 0.9947502828941637, "300–1000 Hz_10ms_modulation_mae_db": 0.26801837289840985, "300–1000 Hz_40ms_log_rms_mae_db": 0.946234977375785, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04559248217340255, "300–1000 Hz_40ms_envelope_correlation": 0.997923525503361, "300–1000 Hz_40ms_modulation_mae_db": 0.23275155781506826, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2405276493874764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11558822221700267, "1000–3000 Hz_10ms_envelope_correlation": 0.9889973476518072, "1000–3000 Hz_10ms_modulation_mae_db": 0.6449462413326391, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8032604822840902, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07132697687233945, "1000–3000 Hz_40ms_envelope_correlation": 0.9962959086314394, "1000–3000 Hz_40ms_modulation_mae_db": 0.6744039745708471, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3101979608650265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16974586107221068, "3000–6000 Hz_10ms_envelope_correlation": 0.9814313111390935, "3000–6000 Hz_10ms_modulation_mae_db": 0.39839846975757515, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7705652758777737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10073929385382716, "3000–6000 Hz_40ms_envelope_correlation": 0.9931519515395167, "3000–6000 Hz_40ms_modulation_mae_db": 0.4824934609552291, "mr_stft_db": 4.913313208623793, "band_mae_db": 0.14608548295783222, "rms_gain_error_db": -0.07271341873477635, "output_peak": 0.3024187610420897}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.61482561222758, "全帯域_10ms_rms_envelope_nrmse": 0.04751886848489078, "全帯域_10ms_envelope_correlation": 0.9969708796791285, "全帯域_10ms_modulation_mae_db": 0.45210125642589905, "全帯域_40ms_log_rms_mae_db": 0.3349039242182348, "全帯域_40ms_rms_envelope_nrmse": 0.026295101823189282, "全帯域_40ms_envelope_correlation": 0.9990262274876569, "全帯域_40ms_modulation_mae_db": 0.5606559422993279, "300–1000 Hz_10ms_log_rms_mae_db": 1.4418815532637639, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07515575211639254, "300–1000 Hz_10ms_envelope_correlation": 0.9945882701608066, "300–1000 Hz_10ms_modulation_mae_db": 0.27563260296719927, "300–1000 Hz_40ms_log_rms_mae_db": 0.9484469885529988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04217775487188686, "300–1000 Hz_40ms_envelope_correlation": 0.998231183549423, "300–1000 Hz_40ms_modulation_mae_db": 0.6693134809883663, "1000–3000 Hz_10ms_log_rms_mae_db": 1.232276466901036, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10980197943747219, "1000–3000 Hz_10ms_envelope_correlation": 0.9899435671409, "1000–3000 Hz_10ms_modulation_mae_db": 0.4663758419669395, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7472924399781617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05927913055450025, "1000–3000 Hz_40ms_envelope_correlation": 0.9975446428193797, "1000–3000 Hz_40ms_modulation_mae_db": 0.6032438670950543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3564530757507711, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1760474916382126, "3000–6000 Hz_10ms_envelope_correlation": 0.9803017421981386, "3000–6000 Hz_10ms_modulation_mae_db": 1.2838122643858116, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8012987560126973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1073113458328075, "3000–6000 Hz_40ms_envelope_correlation": 0.9931514143194692, "3000–6000 Hz_40ms_modulation_mae_db": 1.7263737778807429, "mr_stft_db": 4.908499254734492, "band_mae_db": 0.2019983711150999, "rms_gain_error_db": -0.07075640391452558, "output_peak": 0.3121552068673211}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6531329289350974, "全帯域_10ms_rms_envelope_nrmse": 0.05323194144982414, "全帯域_10ms_envelope_correlation": 0.996125084047048, "全帯域_10ms_modulation_mae_db": 0.5013955292108696, "全帯域_40ms_log_rms_mae_db": 0.33679229775040437, "全帯域_40ms_rms_envelope_nrmse": 0.027921196375965913, "全帯域_40ms_envelope_correlation": 0.9988197945769053, "全帯域_40ms_modulation_mae_db": 0.7353437604797101, "300–1000 Hz_10ms_log_rms_mae_db": 1.4350395040083164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08285420740457715, "300–1000 Hz_10ms_envelope_correlation": 0.9933809166638081, "300–1000 Hz_10ms_modulation_mae_db": 0.4123554370436187, "300–1000 Hz_40ms_log_rms_mae_db": 0.9394863624257216, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04739721547732142, "300–1000 Hz_40ms_envelope_correlation": 0.9977137917364589, "300–1000 Hz_40ms_modulation_mae_db": 0.8280594135561135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2644254611021224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10803212142732473, "1000–3000 Hz_10ms_envelope_correlation": 0.9901922186914687, "1000–3000 Hz_10ms_modulation_mae_db": 0.5390238690848269, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7774974633760448, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06655444340225215, "1000–3000 Hz_40ms_envelope_correlation": 0.9964919213701489, "1000–3000 Hz_40ms_modulation_mae_db": 0.7753248074127018, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2829824571625656, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15703714912222122, "3000–6000 Hz_10ms_envelope_correlation": 0.9839515886191001, "3000–6000 Hz_10ms_modulation_mae_db": 1.1203052834996987, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7580297626634483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08999859923753704, "3000–6000 Hz_40ms_envelope_correlation": 0.9946042873436686, "3000–6000 Hz_40ms_modulation_mae_db": 1.6952466925300624, "mr_stft_db": 4.902111352805928, "band_mae_db": 0.1845151996443759, "rms_gain_error_db": -0.036252748678707376, "output_peak": 0.3074666921791154}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6175149824763484, "全帯域_10ms_rms_envelope_nrmse": 0.04426765297764753, "全帯域_10ms_envelope_correlation": 0.997440576369409, "全帯域_10ms_modulation_mae_db": 0.6304130832026488, "全帯域_40ms_log_rms_mae_db": 0.333263334294995, "全帯域_40ms_rms_envelope_nrmse": 0.02456064123683378, "全帯域_40ms_envelope_correlation": 0.9992343906852393, "全帯域_40ms_modulation_mae_db": 0.8020996957225512, "300–1000 Hz_10ms_log_rms_mae_db": 1.4219041333501434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07007897024585899, "300–1000 Hz_10ms_envelope_correlation": 0.9953625510896058, "300–1000 Hz_10ms_modulation_mae_db": 0.6586515263931347, "300–1000 Hz_40ms_log_rms_mae_db": 0.9289076984699751, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04035661911100607, "300–1000 Hz_40ms_envelope_correlation": 0.9984398892547605, "300–1000 Hz_40ms_modulation_mae_db": 1.019439540533616, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1861622840918118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10575602506807254, "1000–3000 Hz_10ms_envelope_correlation": 0.9911702456048515, "1000–3000 Hz_10ms_modulation_mae_db": 0.7772550099504029, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7337244655713997, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06636455002710853, "1000–3000 Hz_40ms_envelope_correlation": 0.997188897568648, "1000–3000 Hz_40ms_modulation_mae_db": 1.0328674398770943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.329927196045661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18032330081329165, "3000–6000 Hz_10ms_envelope_correlation": 0.9793577765586572, "3000–6000 Hz_10ms_modulation_mae_db": 2.0102729216832422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7827645715363899, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11303633254479534, "3000–6000 Hz_40ms_envelope_correlation": 0.9923857056178573, "3000–6000 Hz_40ms_modulation_mae_db": 2.515986872245564, "mr_stft_db": 4.913880279702813, "band_mae_db": 0.21454301424398084, "rms_gain_error_db": -0.09441497346029164, "output_peak": 0.3078890891982073}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6310996900570086, "全帯域_10ms_rms_envelope_nrmse": 0.04836384398341556, "全帯域_10ms_envelope_correlation": 0.9968688601676682, "全帯域_10ms_modulation_mae_db": 0.6152238182954658, "全帯域_40ms_log_rms_mae_db": 0.3450495069372563, "全帯域_40ms_rms_envelope_nrmse": 0.025654024925811283, "全帯域_40ms_envelope_correlation": 0.9990870485325397, "全帯域_40ms_modulation_mae_db": 0.6758007283224506, "300–1000 Hz_10ms_log_rms_mae_db": 1.386887172748507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07711675522168382, "300–1000 Hz_10ms_envelope_correlation": 0.9942900000914823, "300–1000 Hz_10ms_modulation_mae_db": 0.8378117098996807, "300–1000 Hz_40ms_log_rms_mae_db": 0.9099809007318208, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044015646187908194, "300–1000 Hz_40ms_envelope_correlation": 0.9980549770701577, "300–1000 Hz_40ms_modulation_mae_db": 1.449253266497798, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2242727443803796, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11116979775780524, "1000–3000 Hz_10ms_envelope_correlation": 0.9903434756259691, "1000–3000 Hz_10ms_modulation_mae_db": 0.7938933919319556, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7850520819867949, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0721506836091815, "1000–3000 Hz_40ms_envelope_correlation": 0.9967002188190092, "1000–3000 Hz_40ms_modulation_mae_db": 0.9031015784807167, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3070035319592015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1743769898369018, "3000–6000 Hz_10ms_envelope_correlation": 0.980398681996182, "3000–6000 Hz_10ms_modulation_mae_db": 1.7254819562743222, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7651668667880582, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11466470310518083, "3000–6000 Hz_40ms_envelope_correlation": 0.9915317742177582, "3000–6000 Hz_40ms_modulation_mae_db": 2.116184244257045, "mr_stft_db": 4.914451284451322, "band_mae_db": 0.1882704379447168, "rms_gain_error_db": -0.07679788305034205, "output_peak": 0.30443112038213077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.641154083321718, "全帯域_10ms_rms_envelope_nrmse": 0.04996904469779232, "全帯域_10ms_envelope_correlation": 0.9966361701851381, "全帯域_10ms_modulation_mae_db": 0.6355781485497761, "全帯域_40ms_log_rms_mae_db": 0.3381480126431042, "全帯域_40ms_rms_envelope_nrmse": 0.027880362606573422, "全帯域_40ms_envelope_correlation": 0.9988828783387376, "全帯域_40ms_modulation_mae_db": 0.9962886690736874, "300–1000 Hz_10ms_log_rms_mae_db": 1.4395495004699932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0784801312666749, "300–1000 Hz_10ms_envelope_correlation": 0.9940684757205673, "300–1000 Hz_10ms_modulation_mae_db": 0.8322171970611819, "300–1000 Hz_40ms_log_rms_mae_db": 0.9651845389001117, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044034543299220984, "300–1000 Hz_40ms_envelope_correlation": 0.9980289357608796, "300–1000 Hz_40ms_modulation_mae_db": 1.1686875632189866, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2243836378110338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10473570340106649, "1000–3000 Hz_10ms_envelope_correlation": 0.9918018319889216, "1000–3000 Hz_10ms_modulation_mae_db": 0.4853754093408525, "1000–3000 Hz_40ms_log_rms_mae_db": 0.77616375521575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06841914328910684, "1000–3000 Hz_40ms_envelope_correlation": 0.9974438004994174, "1000–3000 Hz_40ms_modulation_mae_db": 0.5828204645643594, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3194859500907166, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15542416859488528, "3000–6000 Hz_10ms_envelope_correlation": 0.9875711569272247, "3000–6000 Hz_10ms_modulation_mae_db": 0.3989063054285516, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8104069291344226, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10726844638582486, "3000–6000 Hz_40ms_envelope_correlation": 0.9951812101092476, "3000–6000 Hz_40ms_modulation_mae_db": 0.5508012449870611, "mr_stft_db": 4.9239558247515385, "band_mae_db": 0.15601190033173518, "rms_gain_error_db": -0.06537083238148221, "output_peak": 0.30417322894385546}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.029620019402367, "spectral_convergence": 0.1336253542802044, "spectral_convergence_status": "ok", "amplitude_floor": 1.1304927744850054e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.78687573022606, "spectral_convergence": 0.1288610449907388, "spectral_convergence_status": "ok", "amplitude_floor": 1.0019114511304393e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.947837559079261, "spectral_convergence": 0.13259174407040747, "spectral_convergence_status": "ok", "amplitude_floor": 8.812050137930174e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0013595771819321111, "output_power": 0.0013668039209764537, "delta_db": 0.023023488802048696, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002257060285350754, "output_power": 0.0022097730573228526, "delta_db": -0.09195484966326056, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006976179772550826, "output_power": 0.000669843171758736, "delta_db": -0.17644528990529446, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001064894568705692, "output_power": 0.00010144773427954277, "delta_db": -0.21064260002536248, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.7425121182630722e-05, "output_power": 2.3390741603714356e-05, "delta_db": -0.6910456391744353, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.591013773579702e-05, "output_power": 5.217980620558256e-05, "delta_db": -0.29988100344555585, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2488321451693262, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07846124268181118, "pitch": null, "output_rms_dbfs": -23.542400131003202, "output_peak": 0.29504993495429954, "reference_rms_dbfs": -23.46393888832139, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.921444436235896}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016450226550328756, "log_rms_mae_db": 0.6341889784684466, "rms_envelope_nrmse": 0.04932647329098063, "envelope_correlation": 0.9967657433342425, "rms_gain_error_db": -0.07771249400388962, "reference_mean_power": 0.004505860165227205, "output_mean_power": 0.004425949560970824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3773993983732174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011178733037365022, "output_power": 0.0010862823556330775, "reference_active": true, "delta_db": -0.12449859365790755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011348254393479664, "output_power": 0.00010233821463071469, "reference_active": true, "delta_db": -0.44891226700221026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1976543085088728e-05, "output_power": 1.862083392859976e-05, "reference_active": true, "delta_db": -0.7196025201311615}], "active_band_count": 3, "active_band_mae_db": 0.43100446026375977}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014948818281718847, "log_rms_mae_db": 0.3418957664365321, "rms_envelope_nrmse": 0.02749578540802759, "envelope_correlation": 0.9989575999220605, "rms_gain_error_db": -0.07748692052996813, "reference_mean_power": 0.004512140106290299, "output_mean_power": 0.004432348339563126}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1895509906107845e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010130343468129068, "output_power": 0.000985918035551884, "reference_active": true, "delta_db": -0.11783359011197087}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.38575581348341e-05, "output_power": 4.814814947103611e-05, "reference_active": true, "delta_db": -0.4866705882428746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.096976553551289e-06, "output_power": 1.6600263878591723e-06, "reference_active": true, "delta_db": -1.014785829531584}], "active_band_count": 3, "active_band_mae_db": 0.5397633359621431}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012259006245866322, "log_rms_mae_db": 1.4798943653955228, "rms_envelope_nrmse": 0.0799561947010954, "envelope_correlation": 0.9939556760597509, "rms_gain_error_db": -0.12001118371799363, "reference_mean_power": 0.001387125934818555, "output_mean_power": 0.0013493194274244614}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.931300849598275e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005535817649712056, "output_power": 0.0005220472954461146, "reference_active": true, "delta_db": -0.2547192581985186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.308179193682493e-05, "output_power": 6.205893691998723e-05, "reference_active": true, "delta_db": -0.7100485576667047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.782827787780852e-06, "output_power": 6.379000364272606e-06, "reference_active": true, "delta_db": 0.42612367714870364}], "active_band_count": 3, "active_band_mae_db": 0.463630497671309}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001104267603656376, "log_rms_mae_db": 0.9450783287122855, "rms_envelope_nrmse": 0.04656448123294118, "envelope_correlation": 0.9979449671284811, "rms_gain_error_db": -0.12035922496827364, "reference_mean_power": 0.0013888741987160204, "output_mean_power": 0.0013509117762201652}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.150356442217087e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005149581742784205, "output_power": 0.0004860102926226937, "reference_active": true, "delta_db": -0.2512648962944453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.177504958330739e-05, "output_power": 2.706189454591993e-05, "reference_active": true, "delta_db": -0.6972803955770432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.976808620230845e-07, "output_power": 6.628072082518542e-07, "reference_active": true, "delta_db": -0.804419497014945}], "active_band_count": 3, "active_band_mae_db": 0.5843215962954779}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.111137261258481e-05, "log_rms_mae_db": 1.2138423163143073, "rms_envelope_nrmse": 0.10289618523413925, "envelope_correlation": 0.9911962282647923, "rms_gain_error_db": -0.25771313791160105, "reference_mean_power": 0.00011418247196706928, "output_mean_power": 0.00010760392847772353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.04035362622924e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.425624857591467e-05, "output_power": 4.107174769312733e-05, "reference_active": true, "delta_db": -0.32431414324472574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.422825900093454e-06, "output_power": 2.9913965905873144e-06, "reference_active": true, "delta_db": -0.5851081524428996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0991841496467888e-06, "output_power": 8.583330715668664e-07, "reference_active": true, "delta_db": -1.0741461097720566}], "active_band_count": 3, "active_band_mae_db": 0.6611894684865606}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.646632245602786e-05, "log_rms_mae_db": 0.7635070644505011, "rms_envelope_nrmse": 0.06266691223675476, "envelope_correlation": 0.9970384708408786, "rms_gain_error_db": -0.25718732011538264, "reference_mean_power": 0.0001146281153210772, "output_mean_power": 0.00010803697608007993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.508716025767634e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.069122651510939e-05, "output_power": 3.767955063781596e-05, "reference_active": true, "delta_db": -0.3339506575314099}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6543344614858788e-06, "output_power": 1.4095503802607636e-06, "reference_active": true, "delta_db": -0.6954271356678582}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6696400281137344e-08, "output_power": 4.182123676296839e-08, "reference_active": true, "delta_db": -1.3215861402691171}], "active_band_count": 3, "active_band_mae_db": 0.7836546444894618}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.742254313282893e-05, "log_rms_mae_db": 1.3164414955917947, "rms_envelope_nrmse": 0.1531178402304001, "envelope_correlation": 0.9853992119526325, "rms_gain_error_db": -0.3935580129976105, "reference_mean_power": 4.027652017807587e-05, "output_mean_power": 3.6787149723997196e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.067374336554813e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.634873883610167e-05, "output_power": 1.5548772018722236e-05, "reference_active": true, "delta_db": -0.21788160387351704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.842264617940254e-06, "output_power": 5.092508080621331e-06, "reference_active": true, "delta_db": -0.5964949710215292}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8971365027113356e-06, "output_power": 9.448110541738101e-07, "reference_active": true, "delta_db": -3.0275361452951666}], "active_band_count": 3, "active_band_mae_db": 1.2806375733967377}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.2287916076198396e-05, "log_rms_mae_db": 0.7812447439263802, "rms_envelope_nrmse": 0.0856871751779555, "envelope_correlation": 0.9961022894018339, "rms_gain_error_db": -0.3934904231606505, "reference_mean_power": 4.047626840683265e-05, "output_mean_power": 3.6970168062129724e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8536334265621397e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7722714720556858e-05, "output_power": 1.626990502982725e-05, "reference_active": true, "delta_db": -0.3714522888110214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2765957035699146e-06, "output_power": 2.568744830090722e-06, "reference_active": true, "delta_db": -1.0570189230080442}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7512938438165208e-07, "output_power": 7.453019907342814e-08, "reference_active": true, "delta_db": -3.710267397843237}], "active_band_count": 3, "active_band_mae_db": 1.7129128698874343}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9614364380777172, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0bd03ba954734714c19a06f6e200590b1e30c9ea1aedb42a556cee7e7c6bf3d3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9904944075386775, "standard_deviation": 0.0008743372183269245, "minimum": 0.9889973476518072, "maximum": 0.9918018319889216, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2331588392394148, "standard_deviation": 0.02904887657950791, "minimum": 1.1861622840918118, "maximum": 1.2793801539271517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6369026115792709, "standard_deviation": 0.1280436692062098, "minimum": 0.4663758419669395, "maximum": 0.7938933919319556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10885382780789074, "standard_deviation": 0.004314641263553299, "minimum": 0.10289618523413925, "maximum": 0.11558822221700267, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996868914383058, "standard_deviation": 0.000507567234296216, "minimum": 0.9962474545155428, "maximum": 0.9975446428193797, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7773569466177113, "standard_deviation": 0.031067288718728416, "minimum": 0.7337244655713997, "maximum": 0.832357820078948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7617891234847723, "standard_deviation": 0.15092127000417632, "minimum": 0.5828204645643594, "maximum": 1.0328674398770943, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06806032208045731, "standard_deviation": 0.005759414170790819, "minimum": 0.05927913055450025, "maximum": 0.07772073665241504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823002054237115, "standard_deviation": 0.0029874163317527187, "minimum": 0.9793577765586572, "maximum": 0.9875711569272247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3202202626950474, "standard_deviation": 0.022189287278100555, "minimum": 1.2829824571625656, "maximum": 1.3564530757507711, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.223507376165701, "standard_deviation": 0.5818109408314017, "minimum": 0.39839846975757515, "maximum": 2.0102729216832422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16769204627243264, "standard_deviation": 0.010790871707686145, "minimum": 0.1531178402304001, "maximum": 0.18032330081329165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993386750800962, "standard_deviation": 0.0017887259668277345, "minimum": 0.9909853738583446, "maximum": 0.9961022894018339, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7896803082632553, "standard_deviation": 0.029395792223887954, "minimum": 0.7580297626634483, "maximum": 0.8479655601668721, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6083546108521642, "standard_deviation": 0.7280392121771585, "minimum": 0.4824934609552291, "maximum": 2.515986872245564, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10441500708266076, "standard_deviation": 0.011444331914025583, "minimum": 0.0856871751779555, "maximum": 0.11661416052335782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943919632428674, "standard_deviation": 0.0006044333140577838, "minimum": 0.9933809166638081, "maximum": 0.9953625510896058, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4399053903282235, "standard_deviation": 0.029498808608722338, "minimum": 1.386887172748507, "maximum": 1.4798943653955228, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5515523996343401, "standard_deviation": 0.229435349566342, "minimum": 0.26801837289840985, "maximum": 0.8378117098996807, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07648229220598826, "standard_deviation": 0.003984221615315792, "minimum": 0.07007897024585899, "maximum": 0.08285420740457715, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980608324361473, "standard_deviation": 0.0002185382667717081, "minimum": 0.9977137917364589, "maximum": 0.9984398892547605, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9420360219252547, "standard_deviation": 0.01660353698080758, "minimum": 0.9099809007318208, "maximum": 0.9651845389001117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8570716392273908, "standard_deviation": 0.37353194456408856, "minimum": 0.23275155781506826, "maximum": 1.449253266497798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04408545185073925, "standard_deviation": 0.002369715358070624, "minimum": 0.04035661911100607, "maximum": 0.04739721547732142, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1935461902978004, "standard_deviation": 0.03284545609212749, "minimum": 0.14608548295783222, "maximum": 0.2488321451693262, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.914846837326408, "standard_deviation": 0.007287658748031833, "minimum": 4.902111352805928, "maximum": 4.9239558247515385, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.30393596218869523, "standard_deviation": 0.005524009014107097, "minimum": 0.29504993495429954, "maximum": 0.3121552068673211, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07084272346613826, "standard_deviation": 0.016389683433289427, "minimum": -0.09441497346029164, "maximum": -0.036252748678707376, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967650584633306, "standard_deviation": 0.00037353226321867856, "minimum": 0.996125084047048, "maximum": 0.997440576369409, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6278022473753804, "standard_deviation": 0.016715972450842907, "minimum": 0.5995172604900383, "maximum": 0.6531329289350974, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5179627792752008, "standard_deviation": 0.11458754992587201, "minimum": 0.31383793428980616, "maximum": 0.6355781485497761, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04906159141877208, "standard_deviation": 0.002581745929203835, "minimum": 0.04426765297764753, "maximum": 0.05323194144982414, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989334676587913, "standard_deviation": 0.00018260517026196364, "minimum": 0.9986533361727328, "maximum": 0.9992343906852393, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3384968409246346, "standard_deviation": 0.0091194914102435, "minimum": 0.3234060531789668, "maximum": 0.35451583193758307, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.665019119319501, "standard_deviation": 0.21350980584286994, "minimum": 0.2729539136700208, "maximum": 0.9962886690736874, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027414622007716993, "standard_deviation": 0.0019149645914529085, "minimum": 0.02456064123683378, "maximum": 0.030686911535581098, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0000", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.984368085861206, "predicted_mos": 2.984368085861206, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e139a1094916541b8bbcd93bd76e38376f35447562f7cfddff006560db6061f2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.774241500068456, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.612647176397927, "全帯域_10ms_rms_envelope_nrmse": 0.04977218872722114, "全帯域_10ms_envelope_correlation": 0.9966160891013657, "全帯域_10ms_modulation_mae_db": 0.5137834791228694, "全帯域_40ms_log_rms_mae_db": 0.32540774723072247, "全帯域_40ms_rms_envelope_nrmse": 0.027821286841283425, "全帯域_40ms_envelope_correlation": 0.9988271639498343, "全帯域_40ms_modulation_mae_db": 0.677744535712885, "300–1000 Hz_10ms_log_rms_mae_db": 1.672643365071322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07180945210227924, "300–1000 Hz_10ms_envelope_correlation": 0.9951372738676658, "300–1000 Hz_10ms_modulation_mae_db": 0.5480178907462702, "300–1000 Hz_40ms_log_rms_mae_db": 1.20611222271473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03934219631623748, "300–1000 Hz_40ms_envelope_correlation": 0.9985123702871052, "300–1000 Hz_40ms_modulation_mae_db": 0.949203265329999, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5942572433167954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11771478712680448, "1000–3000 Hz_10ms_envelope_correlation": 0.988471010657636, "1000–3000 Hz_10ms_modulation_mae_db": 0.9839098725003131, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0623441391472872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07269996055313929, "1000–3000 Hz_40ms_envelope_correlation": 0.9960279945258699, "1000–3000 Hz_40ms_modulation_mae_db": 1.1048608787570615, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9735888570796443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20759208511598912, "3000–6000 Hz_10ms_envelope_correlation": 0.9731674653920384, "3000–6000 Hz_10ms_modulation_mae_db": 0.4931212865294023, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2673312436013824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.115466652452239, "3000–6000 Hz_40ms_envelope_correlation": 0.9914969644250688, "3000–6000 Hz_40ms_modulation_mae_db": 0.6646422274379921, "mr_stft_db": 5.3037764838767565, "band_mae_db": 0.11999187836378207, "rms_gain_error_db": -0.03584993477456043, "output_peak": 0.3114704581127267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6250554344359037, "全帯域_10ms_rms_envelope_nrmse": 0.04599442085481124, "全帯域_10ms_envelope_correlation": 0.9971626967397209, "全帯域_10ms_modulation_mae_db": 0.4788737172976692, "全帯域_40ms_log_rms_mae_db": 0.33166740113155097, "全帯域_40ms_rms_envelope_nrmse": 0.025143686463459162, "全帯域_40ms_envelope_correlation": 0.9991029974337758, "全帯域_40ms_modulation_mae_db": 0.6224121803434057, "300–1000 Hz_10ms_log_rms_mae_db": 1.6693974204847082, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06862733407688373, "300–1000 Hz_10ms_envelope_correlation": 0.9955366800036037, "300–1000 Hz_10ms_modulation_mae_db": 0.504400557056303, "300–1000 Hz_40ms_log_rms_mae_db": 1.1763275836066536, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040038586583571546, "300–1000 Hz_40ms_envelope_correlation": 0.9984154998747836, "300–1000 Hz_40ms_modulation_mae_db": 0.6956045621236301, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6159068104222754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12455574400604577, "1000–3000 Hz_10ms_envelope_correlation": 0.986406132981284, "1000–3000 Hz_10ms_modulation_mae_db": 0.2983381552003664, "1000–3000 Hz_40ms_log_rms_mae_db": 1.082800370469251, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07406422977615669, "1000–3000 Hz_40ms_envelope_correlation": 0.995135337638345, "1000–3000 Hz_40ms_modulation_mae_db": 0.3872105339074457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9262161440180723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20060762620429562, "3000–6000 Hz_10ms_envelope_correlation": 0.9737840035561413, "3000–6000 Hz_10ms_modulation_mae_db": 0.3924864231929804, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2599740230546836, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11414711611204187, "3000–6000 Hz_40ms_envelope_correlation": 0.9910213625959566, "3000–6000 Hz_40ms_modulation_mae_db": 0.645874102549531, "mr_stft_db": 5.306909358912631, "band_mae_db": 0.07061743985331588, "rms_gain_error_db": -0.06101366676824257, "output_peak": 0.3026180041899744}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6495735909095053, "全帯域_10ms_rms_envelope_nrmse": 0.05582915261158728, "全帯域_10ms_envelope_correlation": 0.9957552722341055, "全帯域_10ms_modulation_mae_db": 0.6749698409069201, "全帯域_40ms_log_rms_mae_db": 0.3531745531906439, "全帯域_40ms_rms_envelope_nrmse": 0.031172065827348546, "全帯域_40ms_envelope_correlation": 0.9985415095534955, "全帯域_40ms_modulation_mae_db": 0.9705227769425903, "300–1000 Hz_10ms_log_rms_mae_db": 1.614323736251481, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06910642729468523, "300–1000 Hz_10ms_envelope_correlation": 0.9954429686652009, "300–1000 Hz_10ms_modulation_mae_db": 0.7091163138667992, "300–1000 Hz_40ms_log_rms_mae_db": 1.1172221675054328, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04152869670894828, "300–1000 Hz_40ms_envelope_correlation": 0.9982258814913717, "300–1000 Hz_40ms_modulation_mae_db": 0.9002331510007863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5987023789398258, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11225501201384895, "1000–3000 Hz_10ms_envelope_correlation": 0.9890837805432409, "1000–3000 Hz_10ms_modulation_mae_db": 0.41312449379203775, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1142825582432048, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06491642977149212, "1000–3000 Hz_40ms_envelope_correlation": 0.9964109754137414, "1000–3000 Hz_40ms_modulation_mae_db": 0.3114238558157338, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9793269539553453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19741321051063038, "3000–6000 Hz_10ms_envelope_correlation": 0.9744497912154344, "3000–6000 Hz_10ms_modulation_mae_db": 0.49914710236173643, "3000–6000 Hz_40ms_log_rms_mae_db": 1.309127215243162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11498759370017085, "3000–6000 Hz_40ms_envelope_correlation": 0.9910278147547296, "3000–6000 Hz_40ms_modulation_mae_db": 0.6058009109187993, "mr_stft_db": 5.323406704821239, "band_mae_db": 0.09850772444921564, "rms_gain_error_db": -0.049551661619199855, "output_peak": 0.3015209306600541}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.60618411544912, "全帯域_10ms_rms_envelope_nrmse": 0.04779640486236563, "全帯域_10ms_envelope_correlation": 0.9969141198892311, "全帯域_10ms_modulation_mae_db": 0.36869387745571897, "全帯域_40ms_log_rms_mae_db": 0.3216335455035496, "全帯域_40ms_rms_envelope_nrmse": 0.025578627283430067, "全帯域_40ms_envelope_correlation": 0.9990572811840255, "全帯域_40ms_modulation_mae_db": 0.4405111006226836, "300–1000 Hz_10ms_log_rms_mae_db": 1.6312240198652037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07471903847585873, "300–1000 Hz_10ms_envelope_correlation": 0.9946029267844598, "300–1000 Hz_10ms_modulation_mae_db": 0.6223034264230284, "300–1000 Hz_40ms_log_rms_mae_db": 1.1539882866491857, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0468835744509302, "300–1000 Hz_40ms_envelope_correlation": 0.9977079284095802, "300–1000 Hz_40ms_modulation_mae_db": 1.1810625275362703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6158850356808483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10934042882381835, "1000–3000 Hz_10ms_envelope_correlation": 0.9897666137728948, "1000–3000 Hz_10ms_modulation_mae_db": 0.854064918537205, "1000–3000 Hz_40ms_log_rms_mae_db": 1.048337632076062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06146744036485209, "1000–3000 Hz_40ms_envelope_correlation": 0.9969600526956307, "1000–3000 Hz_40ms_modulation_mae_db": 1.023636881372893, "3000–6000 Hz_10ms_log_rms_mae_db": 1.94094248225272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2053666087621962, "3000–6000 Hz_10ms_envelope_correlation": 0.9724025163115497, "3000–6000 Hz_10ms_modulation_mae_db": 0.7307912998898832, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2550825968734658, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11495981931069436, "3000–6000 Hz_40ms_envelope_correlation": 0.991338133032025, "3000–6000 Hz_40ms_modulation_mae_db": 1.1194039514427727, "mr_stft_db": 5.339485589531393, "band_mae_db": 0.14488175088940586, "rms_gain_error_db": -0.06211080680498629, "output_peak": 0.30794073577671266}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6433161879517622, "全帯域_10ms_rms_envelope_nrmse": 0.04759237868779845, "全帯域_10ms_envelope_correlation": 0.9968991488817267, "全帯域_10ms_modulation_mae_db": 0.1759403132154993, "全帯域_40ms_log_rms_mae_db": 0.34354692744819515, "全帯域_40ms_rms_envelope_nrmse": 0.02795055192845914, "全帯域_40ms_envelope_correlation": 0.998798609785019, "全帯域_40ms_modulation_mae_db": 0.3253503861821769, "300–1000 Hz_10ms_log_rms_mae_db": 1.6459951475570755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07609968749118202, "300–1000 Hz_10ms_envelope_correlation": 0.9944668703787294, "300–1000 Hz_10ms_modulation_mae_db": 0.5662406330129216, "300–1000 Hz_40ms_log_rms_mae_db": 1.1792889774659268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04658992217705289, "300–1000 Hz_40ms_envelope_correlation": 0.9977894982164893, "300–1000 Hz_40ms_modulation_mae_db": 0.9073078012582526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6718822137922715, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11917070110875794, "1000–3000 Hz_10ms_envelope_correlation": 0.9877711716346863, "1000–3000 Hz_10ms_modulation_mae_db": 0.5890764302324727, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1124279311882477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06548231829474714, "1000–3000 Hz_40ms_envelope_correlation": 0.9965302644142544, "1000–3000 Hz_40ms_modulation_mae_db": 0.6669796854484723, "3000–6000 Hz_10ms_log_rms_mae_db": 2.000906242494322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20078848019184628, "3000–6000 Hz_10ms_envelope_correlation": 0.9742309896010857, "3000–6000 Hz_10ms_modulation_mae_db": 0.500278065604009, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2883153085911703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12081178159276705, "3000–6000 Hz_40ms_envelope_correlation": 0.9901274389864043, "3000–6000 Hz_40ms_modulation_mae_db": 0.7151419862060914, "mr_stft_db": 5.332011116214489, "band_mae_db": 0.13906719641844192, "rms_gain_error_db": -0.01346962012153891, "output_peak": 0.3089405017654667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6323505421284145, "全帯域_10ms_rms_envelope_nrmse": 0.047466180803889514, "全帯域_10ms_envelope_correlation": 0.9969285979874074, "全帯域_10ms_modulation_mae_db": 0.23396961314674006, "全帯域_40ms_log_rms_mae_db": 0.3312603505576883, "全帯域_40ms_rms_envelope_nrmse": 0.025522793233343728, "全帯域_40ms_envelope_correlation": 0.9990303154673977, "全帯域_40ms_modulation_mae_db": 0.36753267700953773, "300–1000 Hz_10ms_log_rms_mae_db": 1.6466163677992696, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0721583292285116, "300–1000 Hz_10ms_envelope_correlation": 0.9950229889719956, "300–1000 Hz_10ms_modulation_mae_db": 0.5707762195315459, "300–1000 Hz_40ms_log_rms_mae_db": 1.1207603098663157, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039570089640353766, "300–1000 Hz_40ms_envelope_correlation": 0.9984167512820313, "300–1000 Hz_40ms_modulation_mae_db": 1.2531240838453408, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6292821089595593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13435191936441565, "1000–3000 Hz_10ms_envelope_correlation": 0.9845713024299074, "1000–3000 Hz_10ms_modulation_mae_db": 0.3611387313339347, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0983933962343393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08472147707264435, "1000–3000 Hz_40ms_envelope_correlation": 0.9941475189087594, "1000–3000 Hz_40ms_modulation_mae_db": 0.4503663323607226, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9576902392348774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2404305462145208, "3000–6000 Hz_10ms_envelope_correlation": 0.961946839622815, "3000–6000 Hz_10ms_modulation_mae_db": 2.043677286960664, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2731888171759567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15243679289900122, "3000–6000 Hz_40ms_envelope_correlation": 0.9846047808380983, "3000–6000 Hz_40ms_modulation_mae_db": 2.541341627338803, "mr_stft_db": 5.331286289082016, "band_mae_db": 0.1689238989495131, "rms_gain_error_db": -0.04521858035916264, "output_peak": 0.300068076881998}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6158291182555904, "全帯域_10ms_rms_envelope_nrmse": 0.04840136153057927, "全帯域_10ms_envelope_correlation": 0.9968638947980295, "全帯域_10ms_modulation_mae_db": 0.46458200148657586, "全帯域_40ms_log_rms_mae_db": 0.3262688245604682, "全帯域_40ms_rms_envelope_nrmse": 0.02494823314059268, "全帯域_40ms_envelope_correlation": 0.9991402110906643, "全帯域_40ms_modulation_mae_db": 0.7693255860679796, "300–1000 Hz_10ms_log_rms_mae_db": 1.6730883044234643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07479720933676375, "300–1000 Hz_10ms_envelope_correlation": 0.9946934342646983, "300–1000 Hz_10ms_modulation_mae_db": 0.992531178985511, "300–1000 Hz_40ms_log_rms_mae_db": 1.2357183648697472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046088039720870486, "300–1000 Hz_40ms_envelope_correlation": 0.9978874687705592, "300–1000 Hz_40ms_modulation_mae_db": 1.6150044800982346, "1000–3000 Hz_10ms_log_rms_mae_db": 1.659757115405414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12466019696369905, "1000–3000 Hz_10ms_envelope_correlation": 0.9868494310761107, "1000–3000 Hz_10ms_modulation_mae_db": 0.4874207636456218, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0826932910714397, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07421615011651994, "1000–3000 Hz_40ms_envelope_correlation": 0.9957621469330119, "1000–3000 Hz_40ms_modulation_mae_db": 0.750744519278918, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9520846720467957, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19149115974966174, "3000–6000 Hz_10ms_envelope_correlation": 0.9763528067464428, "3000–6000 Hz_10ms_modulation_mae_db": 0.8055764404281472, "3000–6000 Hz_40ms_log_rms_mae_db": 1.256827712127966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1063978994752255, "3000–6000 Hz_40ms_envelope_correlation": 0.9922826200286877, "3000–6000 Hz_40ms_modulation_mae_db": 1.1730347817022786, "mr_stft_db": 5.334359482373643, "band_mae_db": 0.1388966363177895, "rms_gain_error_db": -0.07284499920064345, "output_peak": 0.3028598920005624}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6283937687613762, "全帯域_10ms_rms_envelope_nrmse": 0.04914799159050958, "全帯域_10ms_envelope_correlation": 0.996718618628446, "全帯域_10ms_modulation_mae_db": 0.3358969089833401, "全帯域_40ms_log_rms_mae_db": 0.3591289161326723, "全帯域_40ms_rms_envelope_nrmse": 0.029130756687839324, "全帯域_40ms_envelope_correlation": 0.9987298932562637, "全帯域_40ms_modulation_mae_db": 0.523994735299226, "300–1000 Hz_10ms_log_rms_mae_db": 1.6304860283897122, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0750378641898139, "300–1000 Hz_10ms_envelope_correlation": 0.9946248152817596, "300–1000 Hz_10ms_modulation_mae_db": 0.8257003279547228, "300–1000 Hz_40ms_log_rms_mae_db": 1.1154067787680413, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04627748485026567, "300–1000 Hz_40ms_envelope_correlation": 0.9978298157517371, "300–1000 Hz_40ms_modulation_mae_db": 1.1320332405609996, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6468727114619797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13452257955593033, "1000–3000 Hz_10ms_envelope_correlation": 0.9849727083410461, "1000–3000 Hz_10ms_modulation_mae_db": 0.5154104946925445, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0816678837202713, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08103775006200611, "1000–3000 Hz_40ms_envelope_correlation": 0.9953651896066338, "1000–3000 Hz_40ms_modulation_mae_db": 0.5398069696826947, "3000–6000 Hz_10ms_log_rms_mae_db": 2.013212367584729, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20741932725665682, "3000–6000 Hz_10ms_envelope_correlation": 0.9718804138021161, "3000–6000 Hz_10ms_modulation_mae_db": 0.5049702437574108, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3207213592704332, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12519207798773466, "3000–6000 Hz_40ms_envelope_correlation": 0.9894997865100723, "3000–6000 Hz_40ms_modulation_mae_db": 0.80580882454823, "mr_stft_db": 5.324050901262304, "band_mae_db": 0.13898745351390093, "rms_gain_error_db": -0.05134669028718667, "output_peak": 0.30291509553902224}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.471512531654729, "spectral_convergence": 0.1368659653067643, "spectral_convergence_status": "ok", "amplitude_floor": 1.1304927744850054e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.227918098943294, "spectral_convergence": 0.13134164227405737, "spectral_convergence_status": "ok", "amplitude_floor": 1.0019114511304393e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.211898821032245, "spectral_convergence": 0.1300335335977966, "spectral_convergence_status": "ok", "amplitude_floor": 8.812050137930174e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0013595771819321111, "output_power": 0.0013612558510144258, "delta_db": 0.005358923558084946, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002257060285350754, "output_power": 0.0022358156851437776, "delta_db": -0.04107160520542939, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006976179772550826, "output_power": 0.000684451967474416, "delta_db": -0.08274687392718215, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001064894568705692, "output_power": 0.0001005802674501118, "delta_db": -0.24793825897458463, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.7425121182630722e-05, "output_power": 2.6496150036205216e-05, "delta_db": -0.14965780923074298, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.591013773579702e-05, "output_power": 5.845320681500108e-05, "delta_db": 0.19317779928666834, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11999187836378207, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03584993477456043, "pitch": null, "output_rms_dbfs": -23.499788823095955, "output_peak": 0.3114704581127267, "reference_rms_dbfs": -23.46393888832139, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.3037764838767565}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016450226550328756, "log_rms_mae_db": 0.612647176397927, "rms_envelope_nrmse": 0.04977218872722114, "envelope_correlation": 0.9966160891013657, "rms_gain_error_db": -0.035998917830188805, "reference_mean_power": 0.004505860165227205, "output_mean_power": 0.0044686652013387895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3773993983732174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011178733037365022, "output_power": 0.0011023275915120353, "reference_active": true, "delta_db": -0.06081906721410079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011348254393479664, "output_power": 0.00010839858135897129, "reference_active": true, "delta_db": -0.19905464287050317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1976543085088728e-05, "output_power": 1.636107012675826e-05, "reference_active": true, "delta_db": -1.2814767272840042}], "active_band_count": 3, "active_band_mae_db": 0.5137834791228694}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014948818281718847, "log_rms_mae_db": 0.32540774723072247, "rms_envelope_nrmse": 0.027821286841283425, "envelope_correlation": 0.9988271639498343, "rms_gain_error_db": -0.036150291124538624, "reference_mean_power": 0.004512140106290299, "output_mean_power": 0.004474737333120038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1895509906107845e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010130343468129068, "output_power": 0.001001550287717415, "reference_active": true, "delta_db": -0.04951410264465653}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.38575581348341e-05, "output_power": 5.1455365627171126e-05, "reference_active": true, "delta_db": -0.19815990383899812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.096976553551289e-06, "output_power": 1.3900730211953666e-06, "reference_active": true, "delta_db": -1.7855596006550003}], "active_band_count": 3, "active_band_mae_db": 0.677744535712885}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012259006245866322, "log_rms_mae_db": 1.672643365071322, "rms_envelope_nrmse": 0.07180945210227924, "envelope_correlation": 0.9951372738676658, "rms_gain_error_db": -0.04509537633615056, "reference_mean_power": 0.001387125934818555, "output_mean_power": 0.0013727971032311298}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.931300849598275e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005535817649712056, "output_power": 0.0005358976801776068, "reference_active": true, "delta_db": -0.14099898937797722}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.308179193682493e-05, "output_power": 6.379629265784588e-05, "reference_active": true, "delta_db": -0.5901374588926234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.782827787780852e-06, "output_power": 4.686499087905359e-06, "reference_active": true, "delta_db": -0.9129172239682097}], "active_band_count": 3, "active_band_mae_db": 0.5480178907462702}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001104267603656376, "log_rms_mae_db": 1.20611222271473, "rms_envelope_nrmse": 0.03934219631623748, "envelope_correlation": 0.9985123702871052, "rms_gain_error_db": -0.04366425745507094, "reference_mean_power": 0.0013888741987160204, "output_mean_power": 0.001374980326698259}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.150356442217087e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005149581742784205, "output_power": 0.0004981859249553396, "reference_active": true, "delta_db": -0.14380502957222316}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.177504958330739e-05, "output_power": 2.8457045530149463e-05, "reference_active": true, "delta_db": -0.47896428191108464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.976808620230845e-07, "output_power": 4.779089063263837e-07, "reference_active": true, "delta_db": -2.224840484506689}], "active_band_count": 3, "active_band_mae_db": 0.949203265329999}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.111137261258481e-05, "log_rms_mae_db": 1.5942572433167954, "rms_envelope_nrmse": 0.11771478712680448, "envelope_correlation": 0.988471010657636, "rms_gain_error_db": -0.2728839968863344, "reference_mean_power": 0.00011418247196706928, "output_mean_power": 0.00010722870010929374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.04035362622924e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.425624857591467e-05, "output_power": 4.021411270120802e-05, "reference_active": true, "delta_db": -0.41596107042920927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.422825900093454e-06, "output_power": 2.277686663296189e-06, "reference_active": true, "delta_db": -1.7689083073440992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0991841496467888e-06, "output_power": 9.212647054053043e-07, "reference_active": true, "delta_db": -0.766860239727631}], "active_band_count": 3, "active_band_mae_db": 0.9839098725003131}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.646632245602786e-05, "log_rms_mae_db": 1.0623441391472872, "rms_envelope_nrmse": 0.07269996055313929, "envelope_correlation": 0.9960279945258699, "rms_gain_error_db": -0.27450833055607987, "reference_mean_power": 0.0001146281153210772, "output_mean_power": 0.000107606949233514}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.508716025767634e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.069122651510939e-05, "output_power": 3.705672642414828e-05, "reference_active": true, "delta_db": -0.4063372927369408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6543344614858788e-06, "output_power": 1.029679387547947e-06, "reference_active": true, "delta_db": -2.059212976567647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6696400281137344e-08, "output_power": 4.66285863690135e-08, "reference_active": true, "delta_db": -0.8490323669665969}], "active_band_count": 3, "active_band_mae_db": 1.1048608787570615}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.742254313282893e-05, "log_rms_mae_db": 1.9735888570796443, "rms_envelope_nrmse": 0.20759208511598912, "envelope_correlation": 0.9731674653920384, "rms_gain_error_db": 0.23526680718018722, "reference_mean_power": 4.027652017807587e-05, "output_mean_power": 4.2518567276223976e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.067374336554813e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.634873883610167e-05, "output_power": 1.682445649560428e-05, "reference_active": true, "delta_db": 0.12456787266247704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.842264617940254e-06, "output_power": 5.50483990037901e-06, "reference_active": true, "delta_db": -0.2583653118018165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8971365027113356e-06, "output_power": 1.473857557962906e-06, "reference_active": true, "delta_db": -1.0964306751239132}], "active_band_count": 3, "active_band_mae_db": 0.4931212865294023}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.2287916076198396e-05, "log_rms_mae_db": 1.2673312436013824, "rms_envelope_nrmse": 0.115466652452239, "envelope_correlation": 0.9914969644250688, "rms_gain_error_db": 0.23467490844124855, "reference_mean_power": 4.047626840683265e-05, "output_mean_power": 4.272361157500599e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8536334265621397e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7722714720556858e-05, "output_power": 1.7845496846106076e-05, "reference_active": true, "delta_db": 0.029983971115875113}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2765957035699146e-06, "output_power": 2.885341576132744e-06, "reference_active": true, "delta_db": -0.5522562372132408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7512938438165208e-07, "output_power": 1.2652908680737352e-07, "reference_active": true, "delta_db": -1.41168647398486}], "active_band_count": 3, "active_band_mae_db": 0.6646422274379921}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9414303094289991, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2c622048aa131caa4be68325fed25fed6c16402c5317d6a03517415acf3d8d73", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9872365189296008, "standard_deviation": 0.0018781135644769082, "minimum": 0.9845713024299074, "maximum": 0.9897666137728948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.629068202247371, "standard_deviation": 0.028215495950551304, "minimum": 1.5942572433167954, "maximum": 1.6718822137922715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5628104824918121, "standard_deviation": 0.2401776466565983, "minimum": 0.2983381552003664, "maximum": 0.9839098725003131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12207142112041508, "standard_deviation": 0.009293239424578909, "minimum": 0.10934042882381835, "maximum": 0.13452257955593033, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957924350170309, "standard_deviation": 0.0008997448641872212, "minimum": 0.9941475189087594, "maximum": 0.9969600526956307, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.085368400268763, "standard_deviation": 0.02288710325233937, "minimum": 1.048337632076062, "maximum": 1.1142825582432048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6543787070779927, "standard_deviation": 0.29092844861838274, "minimum": 0.3114238558157338, "maximum": 1.1048608787570615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07232571950144472, "standard_deviation": 0.008070425946352777, "minimum": 0.06146744036485209, "maximum": 0.08472147707264435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9722768532809529, "standard_deviation": 0.0043931852127873875, "minimum": 0.961946839622815, "maximum": 0.9763528067464428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9679959948333132, "standard_deviation": 0.029578232288045566, "minimum": 1.9262161440180723, "maximum": 2.013212367584729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7462560185905291, "standard_deviation": 0.5419642755445999, "minimum": 0.3924864231929804, "maximum": 2.043677286960664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20638863050072465, "standard_deviation": 0.014782623417124567, "minimum": 0.19149115974966174, "maximum": 0.2404305462145208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9901748626463803, "standard_deviation": 0.0024045097478952475, "minimum": 0.9846047808380983, "maximum": 0.9922826200286877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2788210344922777, "standard_deviation": 0.024872272872278025, "minimum": 1.2550825968734658, "maximum": 1.3207213592704332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0338810515180623, "standard_deviation": 0.6461828749987806, "minimum": 0.6058009109187993, "maximum": 2.541341627338803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12054996669123433, "standard_deviation": 0.013978017213990127, "minimum": 0.1063978994752255, "maximum": 0.15243679289900122, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949409947772642, "standard_deviation": 0.00040574566209321207, "minimum": 0.9944668703787294, "maximum": 0.9955366800036037, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6479717987302795, "standard_deviation": 0.02211256599532435, "minimum": 1.614323736251481, "maximum": 1.6730883044234643, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6673858184471377, "standard_deviation": 0.1668761330415183, "minimum": 0.504400557056303, "maximum": 0.992531178985511, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07279441777449727, "standard_deviation": 0.0028287762356666835, "minimum": 0.06862733407688373, "maximum": 0.07609968749118202, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980981517604572, "standard_deviation": 0.0003282072869334363, "minimum": 0.9977079284095802, "maximum": 0.9985123702871052, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.163103086430754, "standard_deviation": 0.0444258960021486, "minimum": 1.1154067787680413, "maximum": 1.2357183648697472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0791966389691892, "standard_deviation": 0.28186218151583753, "minimum": 0.6956045621236301, "maximum": 1.6150044800982346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04328982380602879, "standard_deviation": 0.00345696235294386, "minimum": 0.03934219631623748, "maximum": 0.0468835744509302, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1274842473444206, "standard_deviation": 0.030580706346430654, "minimum": 0.07061743985331588, "maximum": 0.1689238989495131, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.324410740759308, "standard_deviation": 0.012897817607611575, "minimum": 5.3037764838767565, "maximum": 5.339485589531393, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.30479171186581466, "standard_deviation": 0.00408348275197203, "minimum": 0.300068076881998, "maximum": 0.3114704581127267, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04892574499194011, "standard_deviation": 0.01828427837871422, "minimum": -0.07284499920064345, "maximum": -0.01346962012153891, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967323047825041, "standard_deviation": 0.00042598857693926874, "minimum": 0.9957552722341055, "maximum": 0.9971626967397209, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6266687417861999, "standard_deviation": 0.014992153001724488, "minimum": 0.60618411544912, "maximum": 0.6495735909095053, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40583871895191664, "standard_deviation": 0.16103787433897557, "minimum": 0.1759403132154993, "maximum": 0.6749698409069201, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04900000995859526, "standard_deviation": 0.0029863981336382722, "minimum": 0.04599442085481124, "maximum": 0.05582915261158728, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989034977150595, "standard_deviation": 0.00021164920991426986, "minimum": 0.9985415095534955, "maximum": 0.9991402110906643, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3365110332194363, "standard_deviation": 0.01383216340800626, "minimum": 0.3216335455035496, "maximum": 0.3591289161326723, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5871742472725606, "standard_deviation": 0.21774027038873664, "minimum": 0.3253503861821769, "maximum": 0.9705227769425903, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02715850017571951, "standard_deviation": 0.002241855970632072, "minimum": 0.02494823314059268, "maximum": 0.031172065827348546, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0000", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.080861806869507, "predicted_mos": 3.080861806869507, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c7eb317f1d219d70e3f4ccff461e963fc3d4049a9de9a5f72c0842c77af946d9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.232897900044918, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.660171200840074, "全帯域_10ms_rms_envelope_nrmse": 0.05011367961800301, "全帯域_10ms_envelope_correlation": 0.9966307383311851, "全帯域_10ms_modulation_mae_db": 0.19255187471868418, "全帯域_40ms_log_rms_mae_db": 0.3486904257776728, "全帯域_40ms_rms_envelope_nrmse": 0.026811737459756183, "全帯域_40ms_envelope_correlation": 0.9989929145483653, "全帯域_40ms_modulation_mae_db": 0.27513676286728084, "300–1000 Hz_10ms_log_rms_mae_db": 1.4170760842968324, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07513165743607514, "300–1000 Hz_10ms_envelope_correlation": 0.9945785111431136, "300–1000 Hz_10ms_modulation_mae_db": 0.397894751221829, "300–1000 Hz_40ms_log_rms_mae_db": 0.9403067140035334, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03891671986461757, "300–1000 Hz_40ms_envelope_correlation": 0.9984991946472302, "300–1000 Hz_40ms_modulation_mae_db": 0.9965780467895015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1370761943045353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1002102354240197, "1000–3000 Hz_10ms_envelope_correlation": 0.9917563309488665, "1000–3000 Hz_10ms_modulation_mae_db": 0.42528196271907986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7099382372394868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05929761960871678, "1000–3000 Hz_40ms_envelope_correlation": 0.9975328004448656, "1000–3000 Hz_40ms_modulation_mae_db": 0.4144177879698497, "3000–6000 Hz_10ms_log_rms_mae_db": 1.181451916824959, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1745455352462076, "3000–6000 Hz_10ms_envelope_correlation": 0.9804949875035992, "3000–6000 Hz_10ms_modulation_mae_db": 0.3622551512563621, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7082947047464563, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11802845811178193, "3000–6000 Hz_40ms_envelope_correlation": 0.9911650222887662, "3000–6000 Hz_40ms_modulation_mae_db": 0.2621631187848737, "mr_stft_db": 4.890480844070775, "band_mae_db": 0.23390691332887723, "rms_gain_error_db": -0.0732697218665381, "output_peak": 0.3002771034508361}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6504281593745551, "全帯域_10ms_rms_envelope_nrmse": 0.051877351273473565, "全帯域_10ms_envelope_correlation": 0.9963271778918693, "全帯域_10ms_modulation_mae_db": 0.5331895859999051, "全帯域_40ms_log_rms_mae_db": 0.36304317345680426, "全帯域_40ms_rms_envelope_nrmse": 0.028815061080024053, "全帯域_40ms_envelope_correlation": 0.9987433253141004, "全帯域_40ms_modulation_mae_db": 0.6949514831674715, "300–1000 Hz_10ms_log_rms_mae_db": 1.3453634783759514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0713078807570218, "300–1000 Hz_10ms_envelope_correlation": 0.9951075160710805, "300–1000 Hz_10ms_modulation_mae_db": 0.7167510070354433, "300–1000 Hz_40ms_log_rms_mae_db": 0.8871520356129705, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04139676791941683, "300–1000 Hz_40ms_envelope_correlation": 0.9982549530741089, "300–1000 Hz_40ms_modulation_mae_db": 0.9375648941283007, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1696770710243545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11034195513923088, "1000–3000 Hz_10ms_envelope_correlation": 0.9905271019096838, "1000–3000 Hz_10ms_modulation_mae_db": 0.7097457797712204, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7450869242606398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07481475235945492, "1000–3000 Hz_40ms_envelope_correlation": 0.9963617224638986, "1000–3000 Hz_40ms_modulation_mae_db": 0.7153329326373615, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2046505484743228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17280357861001572, "3000–6000 Hz_10ms_envelope_correlation": 0.9809402299540434, "3000–6000 Hz_10ms_modulation_mae_db": 0.4392921357878328, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7217033723918451, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10371715274184279, "3000–6000 Hz_40ms_envelope_correlation": 0.9927654692251124, "3000–6000 Hz_40ms_modulation_mae_db": 0.7150110221465469, "mr_stft_db": 4.868221465545228, "band_mae_db": 0.13089536193353643, "rms_gain_error_db": -0.03873775441796219, "output_peak": 0.2954268047624693}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6483497659836395, "全帯域_10ms_rms_envelope_nrmse": 0.0491374088301968, "全帯域_10ms_envelope_correlation": 0.9966972078999374, "全帯域_10ms_modulation_mae_db": 0.291529440457817, "全帯域_40ms_log_rms_mae_db": 0.34550451623878226, "全帯域_40ms_rms_envelope_nrmse": 0.025318388251987824, "全帯域_40ms_envelope_correlation": 0.9990257467069416, "全帯域_40ms_modulation_mae_db": 0.48385536094522896, "300–1000 Hz_10ms_log_rms_mae_db": 1.365016331561411, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0770044248604616, "300–1000 Hz_10ms_envelope_correlation": 0.9942821198439145, "300–1000 Hz_10ms_modulation_mae_db": 0.7370316226678243, "300–1000 Hz_40ms_log_rms_mae_db": 0.9220337869990036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04410685320236719, "300–1000 Hz_40ms_envelope_correlation": 0.9979886085432386, "300–1000 Hz_40ms_modulation_mae_db": 1.122728904398417, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1437832457365227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10718097657283594, "1000–3000 Hz_10ms_envelope_correlation": 0.9910031001270109, "1000–3000 Hz_10ms_modulation_mae_db": 0.8251997274463757, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7503885495817879, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06999750145213517, "1000–3000 Hz_40ms_envelope_correlation": 0.9968675466725218, "1000–3000 Hz_40ms_modulation_mae_db": 0.910166514020433, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2321847081373245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17024112830503066, "3000–6000 Hz_10ms_envelope_correlation": 0.9812050095011468, "3000–6000 Hz_10ms_modulation_mae_db": 0.28081632572895054, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7264610498447047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08133850969962006, "3000–6000 Hz_40ms_envelope_correlation": 0.9954431141006812, "3000–6000 Hz_40ms_modulation_mae_db": 0.4322516914109422, "mr_stft_db": 4.852564976494484, "band_mae_db": 0.12919530027213857, "rms_gain_error_db": -0.011135060553441642, "output_peak": 0.29842491664678866}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.649180169071581, "全帯域_10ms_rms_envelope_nrmse": 0.04990698230864143, "全帯域_10ms_envelope_correlation": 0.996607384211421, "全帯域_10ms_modulation_mae_db": 0.6056386575165135, "全帯域_40ms_log_rms_mae_db": 0.31935624779565236, "全帯域_40ms_rms_envelope_nrmse": 0.026429970602024396, "全帯域_40ms_envelope_correlation": 0.998952868973368, "全帯域_40ms_modulation_mae_db": 0.6851284284041869, "300–1000 Hz_10ms_log_rms_mae_db": 1.3835067937316692, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07115807193866014, "300–1000 Hz_10ms_envelope_correlation": 0.9951687422830109, "300–1000 Hz_10ms_modulation_mae_db": 0.47860768489917, "300–1000 Hz_40ms_log_rms_mae_db": 0.925651747156156, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03944550077010976, "300–1000 Hz_40ms_envelope_correlation": 0.9984538644152698, "300–1000 Hz_40ms_modulation_mae_db": 1.0198715735951944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1698247240399202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11121911177261631, "1000–3000 Hz_10ms_envelope_correlation": 0.9899895139289807, "1000–3000 Hz_10ms_modulation_mae_db": 0.5468887533022446, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7514223403180055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07068958174523701, "1000–3000 Hz_40ms_envelope_correlation": 0.9965095557873597, "1000–3000 Hz_40ms_modulation_mae_db": 0.5926815846115555, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1785718566053096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14406391421841988, "3000–6000 Hz_10ms_envelope_correlation": 0.9867270180476561, "3000–6000 Hz_10ms_modulation_mae_db": 0.5781687861491575, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6978233585052583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07337981820315098, "3000–6000 Hz_40ms_envelope_correlation": 0.9968435094088262, "3000–6000 Hz_40ms_modulation_mae_db": 0.8648711670800697, "mr_stft_db": 4.857145640562963, "band_mae_db": 0.1479779162761867, "rms_gain_error_db": -0.03933055521438181, "output_peak": 0.2966357500649364}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6194190861884125, "全帯域_10ms_rms_envelope_nrmse": 0.05309892980156876, "全帯域_10ms_envelope_correlation": 0.9961962109002217, "全帯域_10ms_modulation_mae_db": 0.6692738314573741, "全帯域_40ms_log_rms_mae_db": 0.35360249614021017, "全帯域_40ms_rms_envelope_nrmse": 0.03077473712613309, "全帯域_40ms_envelope_correlation": 0.9986226614691275, "全帯域_40ms_modulation_mae_db": 0.8743004769277313, "300–1000 Hz_10ms_log_rms_mae_db": 1.3660155017238749, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0726281843355654, "300–1000 Hz_10ms_envelope_correlation": 0.9950164756170695, "300–1000 Hz_10ms_modulation_mae_db": 0.6780524096731654, "300–1000 Hz_40ms_log_rms_mae_db": 0.9286497843072842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042565715419880315, "300–1000 Hz_40ms_envelope_correlation": 0.9982568207670477, "300–1000 Hz_40ms_modulation_mae_db": 1.2069251407656538, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1601124502658808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10642150821248478, "1000–3000 Hz_10ms_envelope_correlation": 0.9908858790100025, "1000–3000 Hz_10ms_modulation_mae_db": 0.6388208013216872, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7412318078999041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06515351797034233, "1000–3000 Hz_40ms_envelope_correlation": 0.9971774225758411, "1000–3000 Hz_40ms_modulation_mae_db": 0.9624503699808328, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1731171055181404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1476813457622991, "3000–6000 Hz_10ms_envelope_correlation": 0.9860712358550815, "3000–6000 Hz_10ms_modulation_mae_db": 1.0902873400934983, "3000–6000 Hz_40ms_log_rms_mae_db": 0.681652923401649, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09159558829272063, "3000–6000 Hz_40ms_envelope_correlation": 0.9947649524451173, "3000–6000 Hz_40ms_modulation_mae_db": 1.4900760809761853, "mr_stft_db": 4.8588486826540995, "band_mae_db": 0.18704001730409922, "rms_gain_error_db": -0.0703382285492358, "output_peak": 0.3279150870894584}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6323304329352762, "全帯域_10ms_rms_envelope_nrmse": 0.0462837485259252, "全帯域_10ms_envelope_correlation": 0.9971053771677082, "全帯域_10ms_modulation_mae_db": 0.37532225325338214, "全帯域_40ms_log_rms_mae_db": 0.3261969789430407, "全帯域_40ms_rms_envelope_nrmse": 0.023566538711456333, "全帯域_40ms_envelope_correlation": 0.9992035162323667, "全帯域_40ms_modulation_mae_db": 0.5017583797208021, "300–1000 Hz_10ms_log_rms_mae_db": 1.3336485079414258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07129095807549944, "300–1000 Hz_10ms_envelope_correlation": 0.9952023240571822, "300–1000 Hz_10ms_modulation_mae_db": 0.5413388729185261, "300–1000 Hz_40ms_log_rms_mae_db": 0.9483800413195961, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04228301447211259, "300–1000 Hz_40ms_envelope_correlation": 0.9982808465459457, "300–1000 Hz_40ms_modulation_mae_db": 0.8127950867187103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1385872752280188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10310511716041679, "1000–3000 Hz_10ms_envelope_correlation": 0.9915365644413802, "1000–3000 Hz_10ms_modulation_mae_db": 0.5658851065004488, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7212632634572401, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06153235514822513, "1000–3000 Hz_40ms_envelope_correlation": 0.9976439338766273, "1000–3000 Hz_40ms_modulation_mae_db": 0.5056900762402846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1367767875918493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16919925573744066, "3000–6000 Hz_10ms_envelope_correlation": 0.9813094526195777, "3000–6000 Hz_10ms_modulation_mae_db": 0.8046295834577758, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6815481092958321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09263498214325494, "3000–6000 Hz_40ms_envelope_correlation": 0.9940978393173265, "3000–6000 Hz_40ms_modulation_mae_db": 1.1716802960267445, "mr_stft_db": 4.862950787362496, "band_mae_db": 0.1383697199950336, "rms_gain_error_db": -0.049551555917869444, "output_peak": 0.30492431433904604}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6406157754261478, "全帯域_10ms_rms_envelope_nrmse": 0.04768190718495839, "全帯域_10ms_envelope_correlation": 0.9969146971055681, "全帯域_10ms_modulation_mae_db": 0.44857040048773594, "全帯域_40ms_log_rms_mae_db": 0.34350910581904576, "全帯域_40ms_rms_envelope_nrmse": 0.024596497830120672, "全帯域_40ms_envelope_correlation": 0.999113734168004, "全帯域_40ms_modulation_mae_db": 0.6420116029415531, "300–1000 Hz_10ms_log_rms_mae_db": 1.3748896856924961, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07502334133546067, "300–1000 Hz_10ms_envelope_correlation": 0.9945894800190457, "300–1000 Hz_10ms_modulation_mae_db": 0.611792330890172, "300–1000 Hz_40ms_log_rms_mae_db": 0.9136303929155506, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040724166180186434, "300–1000 Hz_40ms_envelope_correlation": 0.9983131918081841, "300–1000 Hz_40ms_modulation_mae_db": 0.982083516323511, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1261886242090486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0970866388749229, "1000–3000 Hz_10ms_envelope_correlation": 0.9918829868078746, "1000–3000 Hz_10ms_modulation_mae_db": 0.23970355663486895, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7050652954929251, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05713515575351962, "1000–3000 Hz_40ms_envelope_correlation": 0.9972294803777655, "1000–3000 Hz_40ms_modulation_mae_db": 0.14442155645627058, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1677242808995028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14793531577275817, "3000–6000 Hz_10ms_envelope_correlation": 0.9857052289541858, "3000–6000 Hz_10ms_modulation_mae_db": 0.02636030940130027, "3000–6000 Hz_40ms_log_rms_mae_db": 0.695821470245014, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08412448848623247, "3000–6000 Hz_40ms_envelope_correlation": 0.9951534757290114, "3000–6000 Hz_40ms_modulation_mae_db": 0.12858690563461175, "mr_stft_db": 4.865455430482849, "band_mae_db": 0.10605734004874538, "rms_gain_error_db": -0.05307998557404868, "output_peak": 0.29871939190122704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6147254987128389, "全帯域_10ms_rms_envelope_nrmse": 0.05455403081442762, "全帯域_10ms_envelope_correlation": 0.9960157470258516, "全帯域_10ms_modulation_mae_db": 0.5263669365886307, "全帯域_40ms_log_rms_mae_db": 0.3333944916873494, "全帯域_40ms_rms_envelope_nrmse": 0.03192095728960632, "全帯域_40ms_envelope_correlation": 0.9985626680772437, "全帯域_40ms_modulation_mae_db": 0.5668299028139863, "300–1000 Hz_10ms_log_rms_mae_db": 1.4436717008286517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07975526656266504, "300–1000 Hz_10ms_envelope_correlation": 0.9938806275696609, "300–1000 Hz_10ms_modulation_mae_db": 0.313676777790487, "300–1000 Hz_40ms_log_rms_mae_db": 1.001490456593574, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049071204085134136, "300–1000 Hz_40ms_envelope_correlation": 0.997540376660565, "300–1000 Hz_40ms_modulation_mae_db": 0.6206528572427709, "1000–3000 Hz_10ms_log_rms_mae_db": 1.103458360457312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10709662201387483, "1000–3000 Hz_10ms_envelope_correlation": 0.9906170807221789, "1000–3000 Hz_10ms_modulation_mae_db": 0.6149162076220258, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6782738102000565, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05800331384241432, "1000–3000 Hz_40ms_envelope_correlation": 0.9978812644268009, "1000–3000 Hz_40ms_modulation_mae_db": 0.537295229720561, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1907898905972918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17600350625441383, "3000–6000 Hz_10ms_envelope_correlation": 0.9798907950222987, "3000–6000 Hz_10ms_modulation_mae_db": 0.2758710940868637, "3000–6000 Hz_40ms_log_rms_mae_db": 0.69571478010605, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1202813128543244, "3000–6000 Hz_40ms_envelope_correlation": 0.990040562268676, "3000–6000 Hz_40ms_modulation_mae_db": 0.2542825060282388, "mr_stft_db": 4.868410415761345, "band_mae_db": 0.178183604416812, "rms_gain_error_db": -0.0894347300557882, "output_peak": 0.30636812245358197}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.961849195996627, "spectral_convergence": 0.1351716464912092, "spectral_convergence_status": "ok", "amplitude_floor": 1.1304927744850054e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.713685032932159, "spectral_convergence": 0.13012075960671132, "spectral_convergence_status": "ok", "amplitude_floor": 1.0019114511304393e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.995908303283538, "spectral_convergence": 0.130906991613558, "spectral_convergence_status": "ok", "amplitude_floor": 8.812050137930174e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0013595771819321111, "output_power": 0.001342640771278524, "delta_db": -0.05444036071900536, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002257060285350754, "output_power": 0.0022236623788974455, "delta_db": -0.06474310591020244, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006976179772550826, "output_power": 0.0006847657786588097, "delta_db": -0.08075615377309583, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001064894568705692, "output_power": 0.00010100216459543848, "delta_db": -0.2297593067890616, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.7425121182630722e-05, "output_power": 2.2716361766398946e-05, "delta_db": -0.8180977889058106, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.591013773579702e-05, "output_power": 5.3941880895813505e-05, "delta_db": -0.15564476387608728, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23390691332887723, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0732697218665381, "pitch": null, "output_rms_dbfs": -23.537208610187932, "output_peak": 0.3002771034508361, "reference_rms_dbfs": -23.46393888832139, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.890480844070775}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016450226550328756, "log_rms_mae_db": 0.660171200840074, "rms_envelope_nrmse": 0.05011367961800301, "envelope_correlation": 0.9966307383311851, "rms_gain_error_db": -0.07320670962632607, "reference_mean_power": 0.004505860165227205, "output_mean_power": 0.004430543845231042}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3773993983732174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011178733037365022, "output_power": 0.001089349690206222, "reference_active": true, "delta_db": -0.11225270525559196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011348254393479664, "output_power": 0.00010436407476366856, "reference_active": true, "delta_db": -0.3637803555986168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1976543085088728e-05, "output_power": 2.1468273651786942e-05, "reference_active": true, "delta_db": -0.10162256330184381}], "active_band_count": 3, "active_band_mae_db": 0.19255187471868418}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014948818281718847, "log_rms_mae_db": 0.3486904257776728, "rms_envelope_nrmse": 0.026811737459756183, "envelope_correlation": 0.9989929145483653, "rms_gain_error_db": -0.07346664304958012, "reference_mean_power": 0.004512140106290299, "output_mean_power": 0.004436453277843494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1895509906107845e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010130343468129068, "output_power": 0.0009872420115040854, "reference_active": true, "delta_db": -0.11200542085256193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.38575581348341e-05, "output_power": 5.071530242139234e-05, "reference_active": true, "delta_db": -0.26107639325611864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.096976553551289e-06, "output_power": 1.889560175583236e-06, "reference_active": true, "delta_db": -0.45232847449316194}], "active_band_count": 3, "active_band_mae_db": 0.27513676286728084}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012259006245866322, "log_rms_mae_db": 1.4170760842968324, "rms_envelope_nrmse": 0.07513165743607514, "envelope_correlation": 0.9945785111431136, "rms_gain_error_db": -0.03227962690061401, "reference_mean_power": 0.001387125934818555, "output_mean_power": 0.0013768541218016881}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.931300849598275e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005535817649712056, "output_power": 0.0005357320974864101, "reference_active": true, "delta_db": -0.14234108813552446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.308179193682493e-05, "output_power": 6.472817134704999e-05, "reference_active": true, "delta_db": -0.527158497199069}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.782827787780852e-06, "output_power": 5.125329519061914e-06, "reference_active": true, "delta_db": -0.5241846683308935}], "active_band_count": 3, "active_band_mae_db": 0.397894751221829}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001104267603656376, "log_rms_mae_db": 0.9403067140035334, "rms_envelope_nrmse": 0.03891671986461757, "envelope_correlation": 0.9984991946472302, "rms_gain_error_db": -0.032476709425909765, "reference_mean_power": 0.0013888741987160204, "output_mean_power": 0.0013785268807509415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.150356442217087e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005149581742784205, "output_power": 0.0004977990259777766, "reference_active": true, "delta_db": -0.14717913877545963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.177504958330739e-05, "output_power": 2.856840313502979e-05, "reference_active": true, "delta_db": -0.46200271267391635}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.976808620230845e-07, "output_power": 4.6107751482889223e-07, "reference_active": true, "delta_db": -2.3805522889191284}], "active_band_count": 3, "active_band_mae_db": 0.9965780467895015}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.111137261258481e-05, "log_rms_mae_db": 1.1370761943045353, "rms_envelope_nrmse": 0.1002102354240197, "envelope_correlation": 0.9917563309488665, "rms_gain_error_db": -0.2757860263507322, "reference_mean_power": 0.00011418247196706928, "output_mean_power": 0.0001071570720055301}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.04035362622924e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.425624857591467e-05, "output_power": 4.10035668857908e-05, "reference_active": true, "delta_db": -0.33152960269464526}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.422825900093454e-06, "output_power": 3.0046617934064327e-06, "reference_active": true, "delta_db": -0.5658921494770492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0991841496467888e-06, "output_power": 1.199258680823747e-06, "reference_active": true, "delta_db": 0.37842413598554525}], "active_band_count": 3, "active_band_mae_db": 0.42528196271907986}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.646632245602786e-05, "log_rms_mae_db": 0.7099382372394868, "rms_envelope_nrmse": 0.05929761960871678, "envelope_correlation": 0.9975328004448656, "rms_gain_error_db": -0.2760456870406452, "reference_mean_power": 0.0001146281153210772, "output_mean_power": 0.00010756886425402236}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.508716025767634e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.069122651510939e-05, "output_power": 3.7743629723298234e-05, "reference_active": true, "delta_db": -0.32657117557561977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6543344614858788e-06, "output_power": 1.442705007772805e-06, "reference_active": true, "delta_db": -0.5944577736391548}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6696400281137344e-08, "output_power": 6.106297050044377e-08, "reference_active": true, "delta_db": 0.3222244146947746}], "active_band_count": 3, "active_band_mae_db": 0.4144177879698497}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.742254313282893e-05, "log_rms_mae_db": 1.181451916824959, "rms_envelope_nrmse": 0.1745455352462076, "envelope_correlation": 0.9804949875035992, "rms_gain_error_db": -0.3682145485019042, "reference_mean_power": 4.027652017807587e-05, "output_mean_power": 3.700245050313706e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.067374336554813e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.634873883610167e-05, "output_power": 1.580438213289022e-05, "reference_active": true, "delta_db": -0.14706734314917777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.842264617940254e-06, "output_power": 5.979152191875683e-06, "reference_active": true, "delta_db": 0.1005838408738504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8971365027113356e-06, "output_power": 2.301494849528039e-06, "reference_active": true, "delta_db": 0.8391142697460581}], "active_band_count": 3, "active_band_mae_db": 0.3622551512563621}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.2287916076198396e-05, "log_rms_mae_db": 0.7082947047464563, "rms_envelope_nrmse": 0.11802845811178193, "envelope_correlation": 0.9911650222887662, "rms_gain_error_db": -0.36833569612692524, "reference_mean_power": 4.047626840683265e-05, "output_mean_power": 3.7184923943260396e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8536334265621397e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7722714720556858e-05, "output_power": 1.7471122329655113e-05, "reference_active": true, "delta_db": -0.062094421977432396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2765957035699146e-06, "output_power": 3.4811138056686775e-06, "reference_active": true, "delta_db": 0.2629536415661157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7512938438165208e-07, "output_power": 1.947615177643477e-07, "reference_active": true, "delta_db": 0.46144129281107293}], "active_band_count": 3, "active_band_mae_db": 0.2621631187848737}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9572271811893991, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "39dd79bd7c66e96e2fd3c5a261cabf6f23db9a895682e49d89d107b1f0997f4d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9910248197369973, "standard_deviation": 0.0006589006580348935, "minimum": 0.9899895139289807, "maximum": 0.9918829868078746, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1435884931581992, "standard_deviation": 0.022749059635835817, "minimum": 1.103458360457312, "maximum": 1.1698247240399202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5708052369147438, "standard_deviation": 0.1780828370205034, "minimum": 0.23970355663486895, "maximum": 0.8251997274463757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10533277064630027, "standard_deviation": 0.004879326999391951, "minimum": 0.0970866388749229, "maximum": 0.11121911177261631, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99715046582821, "standard_deviation": 0.0005403242485175027, "minimum": 0.9963617224638986, "maximum": 0.9978812644268009, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7253337785562557, "standard_deviation": 0.026268755262128584, "minimum": 0.6782738102000565, "maximum": 0.7514223403180055, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5978070064546436, "standard_deviation": 0.26611196179882457, "minimum": 0.14442155645627058, "maximum": 0.9624503699808328, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06457797473500566, "standard_deviation": 0.006629216129099618, "minimum": 0.05713515575351962, "maximum": 0.07481475235945492, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9827929946821986, "standard_deviation": 0.002842620633600666, "minimum": 0.9798907950222987, "maximum": 0.9867270180476561, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1831583868310875, "standard_deviation": 0.02787208805123646, "minimum": 1.1367767875918493, "maximum": 1.2321847081373245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4822100907452176, "standard_deviation": 0.33596040015918127, "minimum": 0.02636030940130027, "maximum": 1.0902873400934983, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1628091974883232, "standard_deviation": 0.013676128325774023, "minimum": 0.14406391421841988, "maximum": 0.17600350625441383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937842430979396, "standard_deviation": 0.0022978020173842996, "minimum": 0.990040562268676, "maximum": 0.9968435094088262, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7011274710671012, "standard_deviation": 0.01667977650479977, "minimum": 0.6815481092958321, "maximum": 0.7264610498447047, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6648653485110266, "standard_deviation": 0.48591623295995684, "minimum": 0.12858690563461175, "maximum": 1.4900760809761853, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09563753881661602, "standard_deviation": 0.01703660104432745, "minimum": 0.07337981820315098, "maximum": 0.1202813128543244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9947282245755097, "standard_deviation": 0.000478818657946032, "minimum": 0.9938806275696609, "maximum": 0.9952023240571822, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3786485105190391, "standard_deviation": 0.036293935336443685, "minimum": 1.3336485079414258, "maximum": 1.4436717008286517, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5593931821370771, "standard_deviation": 0.1542044713447005, "minimum": 0.313676777790487, "maximum": 0.7370316226678243, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07416247316267616, "standard_deviation": 0.0031353105394702143, "minimum": 0.07115807193866014, "maximum": 0.07975526656266504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981984820576988, "standard_deviation": 0.0003069044598552228, "minimum": 0.997540376660565, "maximum": 0.9984991946472302, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9334118698634586, "standard_deviation": 0.03305820803996765, "minimum": 0.8871520356129705, "maximum": 1.001490456593574, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9624000024952575, "standard_deviation": 0.18132291072049228, "minimum": 0.6206528572427709, "maximum": 1.2069251407656538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.042313742739228104, "standard_deviation": 0.0032088524048746727, "minimum": 0.03891671986461757, "maximum": 0.049071204085134136, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15645327169692863, "standard_deviation": 0.040955765589273545, "minimum": 0.10605734004874538, "maximum": 0.23390691332887723, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.86550978036678, "standard_deviation": 0.011512535704935881, "minimum": 4.852564976494484, "maximum": 4.890480844070775, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.303586436338543, "standard_deviation": 0.010539610055968068, "minimum": 0.2954268047624693, "maximum": 0.3279150870894584, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05310969901865823, "standard_deviation": 0.024476536186858057, "minimum": -0.0894347300557882, "maximum": -0.011135060553441642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9965618175667204, "standard_deviation": 0.0003648982224407513, "minimum": 0.9960157470258516, "maximum": 0.9971053771677082, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6394025110665655, "standard_deviation": 0.01598364284511742, "minimum": 0.6147254987128389, "maximum": 0.660171200840074, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45530537256000525, "standard_deviation": 0.16115782430395786, "minimum": 0.19255187471868418, "maximum": 0.6692738314573741, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.050331754794649346, "standard_deviation": 0.002749967236636662, "minimum": 0.0462837485259252, "maximum": 0.05455403081442762, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989021794361896, "standard_deviation": 0.00023307366043312646, "minimum": 0.9985626680772437, "maximum": 0.9992035162323667, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3416621794823197, "standard_deviation": 0.0145030572021637, "minimum": 0.31935624779565236, "maximum": 0.36304317345680426, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5904965497235302, "standard_deviation": 0.17821464052790562, "minimum": 0.27513676286728084, "maximum": 0.8743004769277313, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027279236043888606, "standard_deviation": 0.0029734251309682116, "minimum": 0.023566538711456333, "maximum": 0.03192095728960632, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0000", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.057210683822632, "predicted_mos": 3.057210683822632, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bf27783b196e231e4745201d14d03a06ae978662ac4704045b5ae617192ef1c5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.804301199968904, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6260666122747012, "全帯域_10ms_rms_envelope_nrmse": 0.0487975700829312, "全帯域_10ms_envelope_correlation": 0.996822442072205, "全帯域_10ms_modulation_mae_db": 0.3811452722346522, "全帯域_40ms_log_rms_mae_db": 0.33596879391609025, "全帯域_40ms_rms_envelope_nrmse": 0.027402175483250638, "全帯域_40ms_envelope_correlation": 0.9989509214841963, "全帯域_40ms_modulation_mae_db": 0.5274764857461233, "300–1000 Hz_10ms_log_rms_mae_db": 1.4731529337542566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07881583119718394, "300–1000 Hz_10ms_envelope_correlation": 0.9941048769666849, "300–1000 Hz_10ms_modulation_mae_db": 0.5089384890960919, "300–1000 Hz_40ms_log_rms_mae_db": 0.9731649338378213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046843669060026974, "300–1000 Hz_40ms_envelope_correlation": 0.9978943143608968, "300–1000 Hz_40ms_modulation_mae_db": 0.4867454712759998, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2673202566763817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10394549510536093, "1000–3000 Hz_10ms_envelope_correlation": 0.9908226220275443, "1000–3000 Hz_10ms_modulation_mae_db": 0.43374107541778195, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7990555268858647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06219628458777235, "1000–3000 Hz_40ms_envelope_correlation": 0.9968900481350311, "1000–3000 Hz_40ms_modulation_mae_db": 0.3884771360639719, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4140646765588438, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15702050742094176, "3000–6000 Hz_10ms_envelope_correlation": 0.9844833544622807, "3000–6000 Hz_10ms_modulation_mae_db": 1.3784341950542986, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8085188004095252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08815351073435934, "3000–6000 Hz_40ms_envelope_correlation": 0.9956265422799876, "3000–6000 Hz_40ms_modulation_mae_db": 1.7306673961993984, "mr_stft_db": 5.001904603095107, "band_mae_db": 0.2153142444283329, "rms_gain_error_db": -0.07037051023805374, "output_peak": 0.2955624336946905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5984300109562093, "全帯域_10ms_rms_envelope_nrmse": 0.048486347532587064, "全帯域_10ms_envelope_correlation": 0.9968458912567867, "全帯域_10ms_modulation_mae_db": 0.5147479477669107, "全帯域_40ms_log_rms_mae_db": 0.3181027922444234, "全帯域_40ms_rms_envelope_nrmse": 0.028376944228151662, "全帯域_40ms_envelope_correlation": 0.9988484305658917, "全帯域_40ms_modulation_mae_db": 0.6791848744488348, "300–1000 Hz_10ms_log_rms_mae_db": 1.5158092501033245, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07215517736751512, "300–1000 Hz_10ms_envelope_correlation": 0.9949978483791989, "300–1000 Hz_10ms_modulation_mae_db": 0.646811850609878, "300–1000 Hz_40ms_log_rms_mae_db": 0.999095480123388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04147707794308555, "300–1000 Hz_40ms_envelope_correlation": 0.9982468219671093, "300–1000 Hz_40ms_modulation_mae_db": 0.8208090524372736, "1000–3000 Hz_10ms_log_rms_mae_db": 1.331599423873391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11428889620030942, "1000–3000 Hz_10ms_envelope_correlation": 0.9897201484358965, "1000–3000 Hz_10ms_modulation_mae_db": 0.802784739266762, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8646871002738326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07745536922875651, "1000–3000 Hz_40ms_envelope_correlation": 0.9959716033985282, "1000–3000 Hz_40ms_modulation_mae_db": 0.8923988955573142, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4255285568119853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17693263064447345, "3000–6000 Hz_10ms_envelope_correlation": 0.9794731467346206, "3000–6000 Hz_10ms_modulation_mae_db": 1.4157100071796584, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8666451546255326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11370633250390123, "3000–6000 Hz_40ms_envelope_correlation": 0.9910953126771416, "3000–6000 Hz_40ms_modulation_mae_db": 1.8413628687282195, "mr_stft_db": 5.001666095857311, "band_mae_db": 0.1679064661697284, "rms_gain_error_db": -0.07259011463716152, "output_peak": 0.2979561132465982}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6174125468732036, "全帯域_10ms_rms_envelope_nrmse": 0.04988506653459597, "全帯域_10ms_envelope_correlation": 0.9966717890313627, "全帯域_10ms_modulation_mae_db": 0.34432947798217306, "全帯域_40ms_log_rms_mae_db": 0.3456646747622885, "全帯域_40ms_rms_envelope_nrmse": 0.029621549763837712, "全帯域_40ms_envelope_correlation": 0.9987562714235102, "全帯域_40ms_modulation_mae_db": 0.31436131025765257, "300–1000 Hz_10ms_log_rms_mae_db": 1.4179367973380197, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07210981008675424, "300–1000 Hz_10ms_envelope_correlation": 0.9950436303438165, "300–1000 Hz_10ms_modulation_mae_db": 0.282935035541547, "300–1000 Hz_40ms_log_rms_mae_db": 0.9577984052351176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04426188843213183, "300–1000 Hz_40ms_envelope_correlation": 0.998046459940551, "300–1000 Hz_40ms_modulation_mae_db": 0.18658043042072361, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2846288905309788, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1145873051520425, "1000–3000 Hz_10ms_envelope_correlation": 0.9888804068951783, "1000–3000 Hz_10ms_modulation_mae_db": 0.5696069409596703, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8237700131984333, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06897360532659061, "1000–3000 Hz_40ms_envelope_correlation": 0.9962448216124211, "1000–3000 Hz_40ms_modulation_mae_db": 0.5291115171867103, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3766758242353576, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1698912498110373, "3000–6000 Hz_10ms_envelope_correlation": 0.9814018300117878, "3000–6000 Hz_10ms_modulation_mae_db": 0.30819181485467256, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7841645948091112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09338873391817228, "3000–6000 Hz_40ms_envelope_correlation": 0.9940533846669782, "3000–6000 Hz_40ms_modulation_mae_db": 0.3669111078808607, "mr_stft_db": 4.981926990051334, "band_mae_db": 0.12717283570514704, "rms_gain_error_db": -0.07290386668779886, "output_peak": 0.3053496471153892}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6046914046342233, "全帯域_10ms_rms_envelope_nrmse": 0.04843141115321817, "全帯域_10ms_envelope_correlation": 0.996840139587983, "全帯域_10ms_modulation_mae_db": 0.3842853673710487, "全帯域_40ms_log_rms_mae_db": 0.3407896878160423, "全帯域_40ms_rms_envelope_nrmse": 0.027398198864758597, "全帯域_40ms_envelope_correlation": 0.9989220499591929, "全帯域_40ms_modulation_mae_db": 0.49914869225879005, "300–1000 Hz_10ms_log_rms_mae_db": 1.4623346101315184, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07438455583267872, "300–1000 Hz_10ms_envelope_correlation": 0.9946978522986114, "300–1000 Hz_10ms_modulation_mae_db": 0.2746855960602023, "300–1000 Hz_40ms_log_rms_mae_db": 0.9824990380888787, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04189029025820536, "300–1000 Hz_40ms_envelope_correlation": 0.9982571415487532, "300–1000 Hz_40ms_modulation_mae_db": 0.5369594980751887, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2819885001402465, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11283241442588715, "1000–3000 Hz_10ms_envelope_correlation": 0.9892338008727116, "1000–3000 Hz_10ms_modulation_mae_db": 0.5260171862911057, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8023938269271738, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06094094133899592, "1000–3000 Hz_40ms_envelope_correlation": 0.997266863894037, "1000–3000 Hz_40ms_modulation_mae_db": 0.6875651687684586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.458346409878486, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18429827955374764, "3000–6000 Hz_10ms_envelope_correlation": 0.9781566001570202, "3000–6000 Hz_10ms_modulation_mae_db": 1.304979730182194, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8402699455051441, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10858808405757055, "3000–6000 Hz_40ms_envelope_correlation": 0.9927770162456272, "3000–6000 Hz_40ms_modulation_mae_db": 1.704804289142477, "mr_stft_db": 4.982835404009971, "band_mae_db": 0.18663650763697506, "rms_gain_error_db": -0.0667605179804, "output_peak": 0.3034862555928307}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6327568088602497, "全帯域_10ms_rms_envelope_nrmse": 0.0506147032412158, "全帯域_10ms_envelope_correlation": 0.9965039884389765, "全帯域_10ms_modulation_mae_db": 0.47321597919401936, "全帯域_40ms_log_rms_mae_db": 0.3198066146790341, "全帯域_40ms_rms_envelope_nrmse": 0.026377045746802174, "全帯域_40ms_envelope_correlation": 0.9989528765143343, "全帯域_40ms_modulation_mae_db": 0.6943554693888482, "300–1000 Hz_10ms_log_rms_mae_db": 1.4809969184091012, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07900877245760096, "300–1000 Hz_10ms_envelope_correlation": 0.9940055209129626, "300–1000 Hz_10ms_modulation_mae_db": 0.5040456335132176, "300–1000 Hz_40ms_log_rms_mae_db": 1.0052227315886002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04506545157099462, "300–1000 Hz_40ms_envelope_correlation": 0.9979537790639195, "300–1000 Hz_40ms_modulation_mae_db": 0.8282569508897409, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2797091670669833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10782640793276323, "1000–3000 Hz_10ms_envelope_correlation": 0.990001648135381, "1000–3000 Hz_10ms_modulation_mae_db": 0.4494449366296502, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8034126796220064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06519923397449061, "1000–3000 Hz_40ms_envelope_correlation": 0.996408425387663, "1000–3000 Hz_40ms_modulation_mae_db": 0.7322586504700347, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3856353477875805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15588097688725708, "3000–6000 Hz_10ms_envelope_correlation": 0.9841084833308122, "3000–6000 Hz_10ms_modulation_mae_db": 1.0616341597637737, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8075529250995045, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08406400365574147, "3000–6000 Hz_40ms_envelope_correlation": 0.9951891967975887, "3000–6000 Hz_40ms_modulation_mae_db": 1.5327921409043377, "mr_stft_db": 4.987350204075684, "band_mae_db": 0.15615649133339804, "rms_gain_error_db": -0.03673639450564129, "output_peak": 0.30816478070512415}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6265649766618658, "全帯域_10ms_rms_envelope_nrmse": 0.04561663748789504, "全帯域_10ms_envelope_correlation": 0.9972769176384473, "全帯域_10ms_modulation_mae_db": 0.6490273114963786, "全帯域_40ms_log_rms_mae_db": 0.3336142117747224, "全帯域_40ms_rms_envelope_nrmse": 0.025911643091961988, "全帯域_40ms_envelope_correlation": 0.9991322410638507, "全帯域_40ms_modulation_mae_db": 0.8304576807653259, "300–1000 Hz_10ms_log_rms_mae_db": 1.4303225589564146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06690850139525278, "300–1000 Hz_10ms_envelope_correlation": 0.9957842059954038, "300–1000 Hz_10ms_modulation_mae_db": 0.537212466078949, "300–1000 Hz_40ms_log_rms_mae_db": 0.9739280098023564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03873464182034689, "300–1000 Hz_40ms_envelope_correlation": 0.9985742006510236, "300–1000 Hz_40ms_modulation_mae_db": 0.8470443980374626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2382956114750199, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10693566648109493, "1000–3000 Hz_10ms_envelope_correlation": 0.9907578509630797, "1000–3000 Hz_10ms_modulation_mae_db": 0.7559842974473154, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7820306123635886, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06544578384695268, "1000–3000 Hz_40ms_envelope_correlation": 0.9970989865328281, "1000–3000 Hz_40ms_modulation_mae_db": 0.9564631452817173, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4038287837169219, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17795598469313323, "3000–6000 Hz_10ms_envelope_correlation": 0.9800660003845267, "3000–6000 Hz_10ms_modulation_mae_db": 1.9449834108996218, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7900752745726816, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1077636147240871, "3000–6000 Hz_40ms_envelope_correlation": 0.993366064227403, "3000–6000 Hz_40ms_modulation_mae_db": 2.38526940606735, "mr_stft_db": 4.987943845112488, "band_mae_db": 0.2008629371863628, "rms_gain_error_db": -0.09589813916165806, "output_peak": 0.3095669235203176}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6139549060220848, "全帯域_10ms_rms_envelope_nrmse": 0.04734737117670103, "全帯域_10ms_envelope_correlation": 0.9969893966462164, "全帯域_10ms_modulation_mae_db": 0.5446365349244372, "全帯域_40ms_log_rms_mae_db": 0.323855005078629, "全帯域_40ms_rms_envelope_nrmse": 0.024200059199161635, "全帯域_40ms_envelope_correlation": 0.9991826679828002, "全帯域_40ms_modulation_mae_db": 0.6252166983249849, "300–1000 Hz_10ms_log_rms_mae_db": 1.4325244935574701, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07551117608306938, "300–1000 Hz_10ms_envelope_correlation": 0.9945233115220861, "300–1000 Hz_10ms_modulation_mae_db": 0.7437261386702914, "300–1000 Hz_40ms_log_rms_mae_db": 1.0051227680842862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043631082665130515, "300–1000 Hz_40ms_envelope_correlation": 0.9980821466474828, "300–1000 Hz_40ms_modulation_mae_db": 1.3586792199869564, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2793106389864202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10863048138123332, "1000–3000 Hz_10ms_envelope_correlation": 0.9905101194768703, "1000–3000 Hz_10ms_modulation_mae_db": 0.7112680863221458, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8219178832525424, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06795287593723293, "1000–3000 Hz_40ms_envelope_correlation": 0.9968880210192446, "1000–3000 Hz_40ms_modulation_mae_db": 0.6595129059070143, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3969439628696338, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1790284551496781, "3000–6000 Hz_10ms_envelope_correlation": 0.9791589472836156, "3000–6000 Hz_10ms_modulation_mae_db": 1.70282966287087, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8138620102676671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11655332658948363, "3000–6000 Hz_40ms_envelope_correlation": 0.9910021730471839, "3000–6000 Hz_40ms_modulation_mae_db": 2.0988550203695913, "mr_stft_db": 4.991390811896589, "band_mae_db": 0.15301704263332433, "rms_gain_error_db": -0.07086813139566531, "output_peak": 0.30670715673154514}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6304048995408934, "全帯域_10ms_rms_envelope_nrmse": 0.04906966389021549, "全帯域_10ms_envelope_correlation": 0.9967612901845333, "全帯域_10ms_modulation_mae_db": 0.6611050423021453, "全帯域_40ms_log_rms_mae_db": 0.33122830171687523, "全帯域_40ms_rms_envelope_nrmse": 0.027526108501228455, "全帯域_40ms_envelope_correlation": 0.9989152204560252, "全帯域_40ms_modulation_mae_db": 1.0640682571974847, "300–1000 Hz_10ms_log_rms_mae_db": 1.456310087720172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07787242741397468, "300–1000 Hz_10ms_envelope_correlation": 0.9941587308577667, "300–1000 Hz_10ms_modulation_mae_db": 0.9130996045470715, "300–1000 Hz_40ms_log_rms_mae_db": 1.016175595763325, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04421018097967534, "300–1000 Hz_40ms_envelope_correlation": 0.9980092474788551, "300–1000 Hz_40ms_modulation_mae_db": 1.2060590431024953, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2364564108347413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10404050256160781, "1000–3000 Hz_10ms_envelope_correlation": 0.9916998761413751, "1000–3000 Hz_10ms_modulation_mae_db": 0.6435462629673273, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8086395489719757, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06672306924342326, "1000–3000 Hz_40ms_envelope_correlation": 0.9974104049710752, "1000–3000 Hz_40ms_modulation_mae_db": 0.7593282755383024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3952978432829115, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16675244133996603, "3000–6000 Hz_10ms_envelope_correlation": 0.9853246728273881, "3000–6000 Hz_10ms_modulation_mae_db": 0.39183473074262604, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8297199374542051, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11081369294691289, "3000–6000 Hz_40ms_envelope_correlation": 0.9946009210323427, "3000–6000 Hz_40ms_modulation_mae_db": 0.5019708527567003, "mr_stft_db": 4.991941460964187, "band_mae_db": 0.13011015710698626, "rms_gain_error_db": -0.0673754047649362, "output_peak": 0.3054965239399891}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.112039637831192, "spectral_convergence": 0.1335200099551048, "spectral_convergence_status": "ok", "amplitude_floor": 1.1304927744850054e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.873841190855844, "spectral_convergence": 0.1288709669473348, "spectral_convergence_status": "ok", "amplitude_floor": 1.0019114511304393e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.019832980598284, "spectral_convergence": 0.13310506333278282, "spectral_convergence_status": "ok", "amplitude_floor": 8.812050137930174e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0013595771819321111, "output_power": 0.0013655320355939501, "delta_db": 0.01898026094199453, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002257060285350754, "output_power": 0.002217925500114068, "delta_db": -0.0759620494608452, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006976179772550826, "output_power": 0.0006697143715557328, "delta_db": -0.17728044951279132, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001064894568705692, "output_power": 0.00010201427621514309, "delta_db": -0.18645659383106253, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.7425121182630722e-05, "output_power": 2.41996180434155e-05, "delta_db": -0.5434004394286508, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.591013773579702e-05, "output_power": 5.2301000254874585e-05, "delta_db": -0.28980567339465313, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2153142444283329, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07037051023805374, "pitch": null, "output_rms_dbfs": -23.534309398559444, "output_peak": 0.2955624336946905, "reference_rms_dbfs": -23.46393888832139, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.001904603095107}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016450226550328756, "log_rms_mae_db": 0.6260666122747012, "rms_envelope_nrmse": 0.0487975700829312, "envelope_correlation": 0.996822442072205, "rms_gain_error_db": -0.07033765691320143, "reference_mean_power": 0.004505860165227205, "output_mean_power": 0.004433471734956138}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3773993983732174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011178733037365022, "output_power": 0.001081935556490554, "reference_active": true, "delta_db": -0.14191191175743764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011348254393479664, "output_power": 0.00010381190369683092, "reference_active": true, "delta_db": -0.38681907616001887}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1976543085088728e-05, "output_power": 1.9076069497069983e-05, "reference_active": true, "delta_db": -0.6147048287865002}], "active_band_count": 3, "active_band_mae_db": 0.3811452722346522}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014948818281718847, "log_rms_mae_db": 0.33596879391609025, "rms_envelope_nrmse": 0.027402175483250638, "envelope_correlation": 0.9989509214841963, "rms_gain_error_db": -0.07131823633421959, "reference_mean_power": 0.004512140106290299, "output_mean_power": 0.004438648485085661}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1895509906107845e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010130343468129068, "output_power": 0.000982070299167264, "reference_active": true, "delta_db": -0.1348159347678704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.38575581348341e-05, "output_power": 4.859442477470021e-05, "reference_active": true, "delta_db": -0.44660213273409716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.096976553551289e-06, "output_power": 1.6652998220367224e-06, "reference_active": true, "delta_db": -1.0010113897364024}], "active_band_count": 3, "active_band_mae_db": 0.5274764857461233}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012259006245866322, "log_rms_mae_db": 1.4731529337542566, "rms_envelope_nrmse": 0.07881583119718394, "envelope_correlation": 0.9941048769666849, "rms_gain_error_db": -0.11291365384975606, "reference_mean_power": 0.001387125934818555, "output_mean_power": 0.0013515263780427725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.931300849598275e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005535817649712056, "output_power": 0.0005225234344981531, "reference_active": true, "delta_db": -0.2507600322444946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.308179193682493e-05, "output_power": 6.261475591419895e-05, "reference_active": true, "delta_db": -0.6713249559357479}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.782827787780852e-06, "output_power": 6.646810651437436e-06, "reference_active": true, "delta_db": 0.6047304791080332}], "active_band_count": 3, "active_band_mae_db": 0.5089384890960919}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001104267603656376, "log_rms_mae_db": 0.9731649338378213, "rms_envelope_nrmse": 0.046843669060026974, "envelope_correlation": 0.9978943143608968, "rms_gain_error_db": -0.11517708228025143, "reference_mean_power": 0.0013888741987160204, "output_mean_power": 0.0013525246900882824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.150356442217087e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005149581742784205, "output_power": 0.00048603271039131675, "reference_active": true, "delta_db": -0.25106457772456453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.177504958330739e-05, "output_power": 2.7420293411666554e-05, "reference_active": true, "delta_db": -0.6401413926105964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.976808620230845e-07, "output_power": 6.997229657667247e-07, "reference_active": true, "delta_db": -0.5690304434928386}], "active_band_count": 3, "active_band_mae_db": 0.4867454712759998}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.111137261258481e-05, "log_rms_mae_db": 1.2673202566763817, "rms_envelope_nrmse": 0.10394549510536093, "envelope_correlation": 0.9908226220275443, "rms_gain_error_db": -0.22532509131467832, "reference_mean_power": 0.00011418247196706928, "output_mean_power": 0.00010840939775512902}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.04035362622924e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.425624857591467e-05, "output_power": 4.0628802875875536e-05, "reference_active": true, "delta_db": -0.3714057168997901}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.422825900093454e-06, "output_power": 2.946913476680613e-06, "reference_active": true, "delta_db": -0.650174247161939}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0991841496467888e-06, "output_power": 1.0306379524231293e-06, "reference_active": true, "delta_db": -0.2796432621916168}], "active_band_count": 3, "active_band_mae_db": 0.43374107541778195}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.646632245602786e-05, "log_rms_mae_db": 0.7990555268858647, "rms_envelope_nrmse": 0.06219628458777235, "envelope_correlation": 0.9968900481350311, "rms_gain_error_db": -0.22538815475287285, "reference_mean_power": 0.0001146281153210772, "output_mean_power": 0.00010883092901448101}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.508716025767634e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.069122651510939e-05, "output_power": 3.7274956851305054e-05, "reference_active": true, "delta_db": -0.380836311612603}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6543344614858788e-06, "output_power": 1.4015599288968577e-06, "reference_active": true, "delta_db": -0.7201164427356244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6696400281137344e-08, "output_power": 5.7544437782824124e-08, "reference_active": true, "delta_db": 0.06447865384368823}], "active_band_count": 3, "active_band_mae_db": 0.3884771360639719}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.742254313282893e-05, "log_rms_mae_db": 1.4140646765588438, "rms_envelope_nrmse": 0.15702050742094176, "envelope_correlation": 0.9844833544622807, "rms_gain_error_db": -0.3565300565948832, "reference_mean_power": 4.027652017807587e-05, "output_mean_power": 3.710213792463705e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.067374336554813e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.634873883610167e-05, "output_power": 1.5381981410320877e-05, "reference_active": true, "delta_db": -0.2647197406405165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.842264617940254e-06, "output_power": 5.0056397112314794e-06, "reference_active": true, "delta_db": -0.6712163652817198}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8971365027113356e-06, "output_power": 9.081590861653622e-07, "reference_active": true, "delta_db": -3.1993664792406595}], "active_band_count": 3, "active_band_mae_db": 1.3784341950542986}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.2287916076198396e-05, "log_rms_mae_db": 0.8085188004095252, "rms_envelope_nrmse": 0.08815351073435934, "envelope_correlation": 0.9956265422799876, "rms_gain_error_db": -0.356891991407409, "reference_mean_power": 4.047626840683265e-05, "output_mean_power": 3.728303581031291e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8536334265621397e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7722714720556858e-05, "output_power": 1.6067649284940947e-05, "reference_active": true, "delta_db": -0.425779031330725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2765957035699146e-06, "output_power": 2.5026581258550012e-06, "reference_active": true, "delta_db": -1.1702133033463662}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7512938438165208e-07, "output_power": 7.651701785246693e-08, "reference_active": true, "delta_db": -3.5960098539211045}], "active_band_count": 3, "active_band_mae_db": 1.7306673961993984}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9588399112880327, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "19ace1e321b31c08cb6ac59b46974155171e4654e70f53c02895c0825e6cff89", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9902033091185046, "standard_deviation": 0.0009249740162456206, "minimum": 0.9888804068951783, "maximum": 0.9916998761413751, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2749136124480203, "standard_deviation": 0.029946491496334146, "minimum": 1.2364564108347413, "maximum": 1.331599423873391, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6115491906627198, "standard_deviation": 0.1390624715801818, "minimum": 0.43374107541778195, "maximum": 0.802784739266762, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10913589615503741, "standard_deviation": 0.004303653429526474, "minimum": 0.10394549510536093, "maximum": 0.1145873051520425, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967723968688535, "standard_deviation": 0.0005123481828282131, "minimum": 0.9959716033985282, "maximum": 0.9974104049710752, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8132383989369272, "standard_deviation": 0.02462169757912895, "minimum": 0.7820306123635886, "maximum": 0.8646871002738326, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7006394618466905, "standard_deviation": 0.18337746844976793, "minimum": 0.3884771360639719, "maximum": 0.9564631452817173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06686089543552685, "standard_deviation": 0.005063679435799865, "minimum": 0.06094094133899592, "maximum": 0.07745536922875651, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9815216293990066, "standard_deviation": 0.002755823585950195, "minimum": 0.9781566001570202, "maximum": 0.9853246728273881, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4070401756427149, "standard_deviation": 0.025784222569078984, "minimum": 1.3766758242353576, "maximum": 1.458346409878486, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1885747139434644, "standard_deviation": 0.5811051067047351, "minimum": 0.30819181485467256, "maximum": 1.9449834108996218, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17097006568752932, "standard_deviation": 0.010475530305662086, "minimum": 0.15588097688725708, "maximum": 0.18429827955374764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9934638263717817, "standard_deviation": 0.0017503332847869184, "minimum": 0.9910021730471839, "maximum": 0.9956265422799876, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8176010803429214, "standard_deviation": 0.027095513053510597, "minimum": 0.7841645948091112, "maximum": 0.8666451546255326, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5203291352561168, "standard_deviation": 0.7202908004193853, "minimum": 0.3669111078808607, "maximum": 2.38526940606735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10287891239127857, "standard_deviation": 0.012447260527959591, "minimum": 0.08406400365574147, "maximum": 0.11655332658948363, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9946644971595664, "standard_deviation": 0.000601531716242421, "minimum": 0.9940055209129626, "maximum": 0.9957842059954038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4586734562462846, "standard_deviation": 0.03193407883196793, "minimum": 1.4179367973380197, "maximum": 1.5158092501033245, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.551431851764656, "standard_deviation": 0.2170616637889067, "minimum": 0.2746855960602023, "maximum": 0.9130996045470715, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07459578147925372, "standard_deviation": 0.004142689920074446, "minimum": 0.06690850139525278, "maximum": 0.07900877245760096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981330139573239, "standard_deviation": 0.00021957930142947512, "minimum": 0.9978943143608968, "maximum": 0.9985742006510236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9891258703154717, "standard_deviation": 0.020203047431862155, "minimum": 0.9577984052351176, "maximum": 1.016175595763325, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7838917580282302, "standard_deviation": 0.38209526387698994, "minimum": 0.18658043042072361, "maximum": 1.3586792199869564, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04326428534119963, "standard_deviation": 0.002498007822093931, "minimum": 0.03873464182034689, "maximum": 0.046843669060026974, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16714708527503186, "standard_deviation": 0.03190918826786513, "minimum": 0.12717283570514704, "maximum": 0.2153142444283329, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.990869926882834, "standard_deviation": 0.0076084626779555275, "minimum": 4.981926990051334, "maximum": 5.001904603095107, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3040362293183106, "standard_deviation": 0.004894609128759121, "minimum": 0.2955624336946905, "maximum": 0.3095669235203176, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06918788492141437, "standard_deviation": 0.016059352525411248, "minimum": -0.09589813916165806, "maximum": -0.03673639450564129, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968389818570639, "standard_deviation": 0.00022707614492208347, "minimum": 0.9965039884389765, "maximum": 0.9972769176384473, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6187852707279289, "standard_deviation": 0.012432399300029052, "minimum": 0.5984300109562093, "maximum": 0.6327568088602497, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4940616166589707, "standard_deviation": 0.12104237686460805, "minimum": 0.34432947798217306, "maximum": 0.6611050423021453, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04853109638741997, "standard_deviation": 0.0015326415815994656, "minimum": 0.04561663748789504, "maximum": 0.0506147032412158, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989575849312251, "standard_deviation": 0.0001397615327938061, "minimum": 0.9987562714235102, "maximum": 0.9991826679828002, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3311287602485131, "standard_deviation": 0.00989237782281049, "minimum": 0.3181027922444234, "maximum": 0.3456646747622885, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6542836835485055, "standard_deviation": 0.226061158556497, "minimum": 0.31436131025765257, "maximum": 1.0640682571974847, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027101715609894106, "standard_deviation": 0.0016342359030921182, "minimum": 0.024200059199161635, "maximum": 0.029621549763837712, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0000", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.1358258724212646, "predicted_mos": 3.1358258724212646, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7d65fc04489508224a0e6c81d950b3de1e21b4beb8be0c6b5272f1d6cf69c50f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.243220599833876, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6167127021634814, "全帯域_10ms_rms_envelope_nrmse": 0.049357157179843306, "全帯域_10ms_envelope_correlation": 0.9966984426731305, "全帯域_10ms_modulation_mae_db": 0.2896946595474981, "全帯域_40ms_log_rms_mae_db": 0.3451696784298132, "全帯域_40ms_rms_envelope_nrmse": 0.02710481632385283, "全帯域_40ms_envelope_correlation": 0.9989253164077638, "全帯域_40ms_modulation_mae_db": 0.4738411126705085, "300–1000 Hz_10ms_log_rms_mae_db": 1.4277960486250696, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0853601412911151, "300–1000 Hz_10ms_envelope_correlation": 0.9930292824242749, "300–1000 Hz_10ms_modulation_mae_db": 0.9394619678106041, "300–1000 Hz_40ms_log_rms_mae_db": 0.970595933210074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05151043074239627, "300–1000 Hz_40ms_envelope_correlation": 0.9973471329857236, "300–1000 Hz_40ms_modulation_mae_db": 1.356429219656859, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2981445263851081, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10721403643956746, "1000–3000 Hz_10ms_envelope_correlation": 0.9900536799808372, "1000–3000 Hz_10ms_modulation_mae_db": 0.606328820606918, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8076658047902205, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06312391822978122, "1000–3000 Hz_40ms_envelope_correlation": 0.9965813271881645, "1000–3000 Hz_40ms_modulation_mae_db": 0.6281642130467797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4386221923482665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18796892174356133, "3000–6000 Hz_10ms_envelope_correlation": 0.9772008355235045, "3000–6000 Hz_10ms_modulation_mae_db": 1.4824611560528662, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8761136184875439, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1384711666930508, "3000–6000 Hz_40ms_envelope_correlation": 0.9868724474696576, "3000–6000 Hz_40ms_modulation_mae_db": 1.7849166508474248, "mr_stft_db": 4.970358653614762, "band_mae_db": 0.19814943341209587, "rms_gain_error_db": -0.06018063491631955, "output_peak": 0.3063299429109551}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6458495659023864, "全帯域_10ms_rms_envelope_nrmse": 0.04628457091123226, "全帯域_10ms_envelope_correlation": 0.9970714757044281, "全帯域_10ms_modulation_mae_db": 0.5190862269448796, "全帯域_40ms_log_rms_mae_db": 0.34890268907616945, "全帯域_40ms_rms_envelope_nrmse": 0.023589264176458864, "全帯域_40ms_envelope_correlation": 0.999155005971689, "全帯域_40ms_modulation_mae_db": 0.7509606285301874, "300–1000 Hz_10ms_log_rms_mae_db": 1.4212848133215885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07030001350174582, "300–1000 Hz_10ms_envelope_correlation": 0.9952313780981678, "300–1000 Hz_10ms_modulation_mae_db": 0.5436487070336963, "300–1000 Hz_40ms_log_rms_mae_db": 0.9639668927898508, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037741629070625435, "300–1000 Hz_40ms_envelope_correlation": 0.9985512803659798, "300–1000 Hz_40ms_modulation_mae_db": 0.9022537857687395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3025366540533347, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1168951786434679, "1000–3000 Hz_10ms_envelope_correlation": 0.9885797525676463, "1000–3000 Hz_10ms_modulation_mae_db": 0.479940885171611, "1000–3000 Hz_40ms_log_rms_mae_db": 0.811432090201255, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06856963925515935, "1000–3000 Hz_40ms_envelope_correlation": 0.9965438498671542, "1000–3000 Hz_40ms_modulation_mae_db": 0.3807640499258392, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3850640221816855, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1926960597060957, "3000–6000 Hz_10ms_envelope_correlation": 0.976266161452158, "3000–6000 Hz_10ms_modulation_mae_db": 2.088332063344111, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8019182891208306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12830933894180957, "3000–6000 Hz_40ms_envelope_correlation": 0.9897698200492341, "3000–6000 Hz_40ms_modulation_mae_db": 2.7758902103688197, "mr_stft_db": 4.971709639445533, "band_mae_db": 0.21883560171801408, "rms_gain_error_db": -0.0313814205865188, "output_peak": 0.30317538171823777}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6274238635602898, "全帯域_10ms_rms_envelope_nrmse": 0.0468011534334968, "全帯域_10ms_envelope_correlation": 0.9970347084649449, "全帯域_10ms_modulation_mae_db": 0.38052387263516696, "全帯域_40ms_log_rms_mae_db": 0.3387650553844154, "全帯域_40ms_rms_envelope_nrmse": 0.02526237183418722, "全帯域_40ms_envelope_correlation": 0.9990751768816308, "全帯域_40ms_modulation_mae_db": 0.5890081698327041, "300–1000 Hz_10ms_log_rms_mae_db": 1.4428225544901092, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06969825055853758, "300–1000 Hz_10ms_envelope_correlation": 0.9953789839146183, "300–1000 Hz_10ms_modulation_mae_db": 0.24560340601994243, "300–1000 Hz_40ms_log_rms_mae_db": 0.9789290536210326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039760121549856156, "300–1000 Hz_40ms_envelope_correlation": 0.9984503317515593, "300–1000 Hz_40ms_modulation_mae_db": 0.2411749062068738, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2282164574603414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10871655688714883, "1000–3000 Hz_10ms_envelope_correlation": 0.9897322756961666, "1000–3000 Hz_10ms_modulation_mae_db": 0.5513966795090876, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7563669987461341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06644296249925116, "1000–3000 Hz_40ms_envelope_correlation": 0.9961296095714249, "1000–3000 Hz_40ms_modulation_mae_db": 0.6856124493843052, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3626838647770876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1714401659756132, "3000–6000 Hz_10ms_envelope_correlation": 0.9831354245924412, "3000–6000 Hz_10ms_modulation_mae_db": 1.598949313375546, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8254812617083637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11430000803293251, "3000–6000 Hz_40ms_envelope_correlation": 0.9941363790486901, "3000–6000 Hz_40ms_modulation_mae_db": 1.9087564477603776, "mr_stft_db": 4.977192285566963, "band_mae_db": 0.18828781681468296, "rms_gain_error_db": -0.05772684728821226, "output_peak": 0.31374354731755144}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6026530366132865, "全帯域_10ms_rms_envelope_nrmse": 0.04675814877127285, "全帯域_10ms_envelope_correlation": 0.9970338813230107, "全帯域_10ms_modulation_mae_db": 0.39635968420623197, "全帯域_40ms_log_rms_mae_db": 0.3182185593273999, "全帯域_40ms_rms_envelope_nrmse": 0.0247863925800292, "全帯域_40ms_envelope_correlation": 0.9991007724308686, "全帯域_40ms_modulation_mae_db": 0.5115921680970795, "300–1000 Hz_10ms_log_rms_mae_db": 1.4238087776910715, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07131731325095994, "300–1000 Hz_10ms_envelope_correlation": 0.9951533217358581, "300–1000 Hz_10ms_modulation_mae_db": 0.6912086904576604, "300–1000 Hz_40ms_log_rms_mae_db": 0.9743167849079024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04307136992723668, "300–1000 Hz_40ms_envelope_correlation": 0.9981201205751856, "300–1000 Hz_40ms_modulation_mae_db": 1.1582282531195127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2146723904017946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10883626805943122, "1000–3000 Hz_10ms_envelope_correlation": 0.9901030709222107, "1000–3000 Hz_10ms_modulation_mae_db": 0.7125551697909441, "1000–3000 Hz_40ms_log_rms_mae_db": 0.737295135700206, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06292303165282787, "1000–3000 Hz_40ms_envelope_correlation": 0.9970941566153836, "1000–3000 Hz_40ms_modulation_mae_db": 0.9265090348318931, "3000–6000 Hz_10ms_log_rms_mae_db": 1.343143530330153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16448937320234563, "3000–6000 Hz_10ms_envelope_correlation": 0.9823025506949292, "3000–6000 Hz_10ms_modulation_mae_db": 0.6929999330603671, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7968226880106771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09054177510526132, "3000–6000 Hz_40ms_envelope_correlation": 0.9944118208376778, "3000–6000 Hz_40ms_modulation_mae_db": 0.8666789714939451, "mr_stft_db": 4.976668401987898, "band_mae_db": 0.16259974824154028, "rms_gain_error_db": -0.05344756557631298, "output_peak": 0.3092310592271237}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6408272013595687, "全帯域_10ms_rms_envelope_nrmse": 0.049692783739454266, "全帯域_10ms_envelope_correlation": 0.9966820106292682, "全帯域_10ms_modulation_mae_db": 0.612029667277964, "全帯域_40ms_log_rms_mae_db": 0.34388774461656496, "全帯域_40ms_rms_envelope_nrmse": 0.027555772275965073, "全帯域_40ms_envelope_correlation": 0.9989203605104069, "全帯域_40ms_modulation_mae_db": 0.8441338417822223, "300–1000 Hz_10ms_log_rms_mae_db": 1.3802681602859617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07052478074003016, "300–1000 Hz_10ms_envelope_correlation": 0.9952412185618748, "300–1000 Hz_10ms_modulation_mae_db": 0.4440277669334208, "300–1000 Hz_40ms_log_rms_mae_db": 0.9491682948237135, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04056859623508387, "300–1000 Hz_40ms_envelope_correlation": 0.9983425079911917, "300–1000 Hz_40ms_modulation_mae_db": 0.6735569040550148, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2950680908666663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10480861422003085, "1000–3000 Hz_10ms_envelope_correlation": 0.9905111325075543, "1000–3000 Hz_10ms_modulation_mae_db": 0.6939149075564724, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7952023721434196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055980948144409425, "1000–3000 Hz_40ms_envelope_correlation": 0.9974116087480849, "1000–3000 Hz_40ms_modulation_mae_db": 0.8080808445493508, "3000–6000 Hz_10ms_log_rms_mae_db": 1.391344114034082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16414678016937007, "3000–6000 Hz_10ms_envelope_correlation": 0.9834013663334711, "3000–6000 Hz_10ms_modulation_mae_db": 1.1235370416642116, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8032068190166667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0990614063726669, "3000–6000 Hz_40ms_envelope_correlation": 0.9947684547403536, "3000–6000 Hz_40ms_modulation_mae_db": 1.3605044246246913, "mr_stft_db": 4.976559186548983, "band_mae_db": 0.1870876166889137, "rms_gain_error_db": -0.0737508637925373, "output_peak": 0.31396867725572797}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6524737947736948, "全帯域_10ms_rms_envelope_nrmse": 0.05453744718916201, "全帯域_10ms_envelope_correlation": 0.9959622181321887, "全帯域_10ms_modulation_mae_db": 0.5199150780375058, "全帯域_40ms_log_rms_mae_db": 0.3541402586987137, "全帯域_40ms_rms_envelope_nrmse": 0.02992270820343646, "全帯域_40ms_envelope_correlation": 0.9986839565394158, "全帯域_40ms_modulation_mae_db": 0.6537926058706781, "300–1000 Hz_10ms_log_rms_mae_db": 1.445224798599611, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0814183157479192, "300–1000 Hz_10ms_envelope_correlation": 0.9936309754358812, "300–1000 Hz_10ms_modulation_mae_db": 0.39743999439672656, "300–1000 Hz_40ms_log_rms_mae_db": 1.0272400958909587, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04922597644081961, "300–1000 Hz_40ms_envelope_correlation": 0.9975559677246646, "300–1000 Hz_40ms_modulation_mae_db": 0.5353100945035774, "1000–3000 Hz_10ms_log_rms_mae_db": 1.270283552912862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11205707444085902, "1000–3000 Hz_10ms_envelope_correlation": 0.9894007317923077, "1000–3000 Hz_10ms_modulation_mae_db": 0.29445944794326057, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8173978410292962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06294697654791011, "1000–3000 Hz_40ms_envelope_correlation": 0.997011170676548, "1000–3000 Hz_40ms_modulation_mae_db": 0.6823911964963828, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3982844597665616, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1739784053326095, "3000–6000 Hz_10ms_envelope_correlation": 0.9810524495091613, "3000–6000 Hz_10ms_modulation_mae_db": 1.7095854213850987, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8207446747796022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10313260702688944, "3000–6000 Hz_40ms_envelope_correlation": 0.9941107355066429, "3000–6000 Hz_40ms_modulation_mae_db": 2.0826234391677656, "mr_stft_db": 4.973374141571061, "band_mae_db": 0.22035129849929247, "rms_gain_error_db": -0.06361149294362894, "output_peak": 0.32685363722053645}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6280960732661712, "全帯域_10ms_rms_envelope_nrmse": 0.04896044983828827, "全帯域_10ms_envelope_correlation": 0.9967175798856018, "全帯域_10ms_modulation_mae_db": 0.48463851772144584, "全帯域_40ms_log_rms_mae_db": 0.352990459200581, "全帯域_40ms_rms_envelope_nrmse": 0.026800570705989603, "全帯域_40ms_envelope_correlation": 0.9989043198485923, "全帯域_40ms_modulation_mae_db": 0.7089990383332924, "300–1000 Hz_10ms_log_rms_mae_db": 1.438197859647444, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07150589279614422, "300–1000 Hz_10ms_envelope_correlation": 0.9951193352181211, "300–1000 Hz_10ms_modulation_mae_db": 0.722171246573443, "300–1000 Hz_40ms_log_rms_mae_db": 0.9665558255042509, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041288895125366154, "300–1000 Hz_40ms_envelope_correlation": 0.9982902391406479, "300–1000 Hz_40ms_modulation_mae_db": 1.249722937593013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3342163644814566, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11305525705445789, "1000–3000 Hz_10ms_envelope_correlation": 0.9891824676476078, "1000–3000 Hz_10ms_modulation_mae_db": 0.5997598855799656, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8305512489580065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0633162630965822, "1000–3000 Hz_40ms_envelope_correlation": 0.9969540071064533, "1000–3000 Hz_40ms_modulation_mae_db": 0.6459618399590075, "3000–6000 Hz_10ms_log_rms_mae_db": 1.384932549288621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16618993443180322, "3000–6000 Hz_10ms_envelope_correlation": 0.9820359506950006, "3000–6000 Hz_10ms_modulation_mae_db": 0.43386161760941455, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8208128226114746, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09595490776316452, "3000–6000 Hz_40ms_envelope_correlation": 0.9936491393442639, "3000–6000 Hz_40ms_modulation_mae_db": 0.8267575805697117, "mr_stft_db": 4.972990067108024, "band_mae_db": 0.117549168530567, "rms_gain_error_db": -0.03156868349925055, "output_peak": 0.3103505484007056}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6160499180516975, "全帯域_10ms_rms_envelope_nrmse": 0.045458775168472396, "全帯域_10ms_envelope_correlation": 0.9971971534037121, "全帯域_10ms_modulation_mae_db": 0.28755661098306046, "全帯域_40ms_log_rms_mae_db": 0.3183861197027612, "全帯域_40ms_rms_envelope_nrmse": 0.024488790838424444, "全帯域_40ms_envelope_correlation": 0.9991116805854492, "全帯域_40ms_modulation_mae_db": 0.38363846899465875, "300–1000 Hz_10ms_log_rms_mae_db": 1.483403125324645, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07552033600191096, "300–1000 Hz_10ms_envelope_correlation": 0.9946263569958197, "300–1000 Hz_10ms_modulation_mae_db": 0.4825135443381338, "300–1000 Hz_40ms_log_rms_mae_db": 1.0522081462960926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042984524836831906, "300–1000 Hz_40ms_envelope_correlation": 0.9982500060852715, "300–1000 Hz_40ms_modulation_mae_db": 0.9289354571887016, "1000–3000 Hz_10ms_log_rms_mae_db": 1.322050719106203, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12754659569652912, "1000–3000 Hz_10ms_envelope_correlation": 0.9866514174493204, "1000–3000 Hz_10ms_modulation_mae_db": 0.4350706708348932, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8233388456672763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0794575305722509, "1000–3000 Hz_40ms_envelope_correlation": 0.9955659193756886, "1000–3000 Hz_40ms_modulation_mae_db": 0.5520413964042327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4425591145321426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15241662883413037, "3000–6000 Hz_10ms_envelope_correlation": 0.9848068084912756, "3000–6000 Hz_10ms_modulation_mae_db": 0.15879836363741884, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7865673496515063, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07227166628791683, "3000–6000 Hz_40ms_envelope_correlation": 0.9964180715690977, "3000–6000 Hz_40ms_modulation_mae_db": 0.3208559302910718, "mr_stft_db": 4.96259894715903, "band_mae_db": 0.17090354222908546, "rms_gain_error_db": -0.04027649254592619, "output_peak": 0.3105234389468049}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.100271864969162, "spectral_convergence": 0.13786476540674716, "spectral_convergence_status": "ok", "amplitude_floor": 1.1304927744850054e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.827811587309619, "spectral_convergence": 0.13657372551898878, "spectral_convergence_status": "ok", "amplitude_floor": 1.0019114511304393e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.982992508565505, "spectral_convergence": 0.1360571491634685, "spectral_convergence_status": "ok", "amplitude_floor": 8.812050137930174e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0013595771819321111, "output_power": 0.0013529272335041206, "delta_db": -0.021294281968972147, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002257060285350754, "output_power": 0.0022288457748532563, "delta_db": -0.054631406165526517, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006976179772550826, "output_power": 0.0006755060331483723, "delta_db": -0.13988431422741315, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001064894568705692, "output_power": 0.00010350009804846759, "delta_db": -0.12365850791787951, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.7425121182630722e-05, "output_power": 2.3373629060101747e-05, "delta_db": -0.6942240775832126, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.591013773579702e-05, "output_power": 5.7944332382240945e-05, "delta_db": 0.15520401260957123, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19814943341209587, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06018063491631955, "pitch": null, "output_rms_dbfs": -23.524119523237708, "output_peak": 0.3063299429109551, "reference_rms_dbfs": -23.46393888832139, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.970358653614762}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016450226550328756, "log_rms_mae_db": 0.6167127021634814, "rms_envelope_nrmse": 0.049357157179843306, "envelope_correlation": 0.9966984426731305, "rms_gain_error_db": -0.05904911685727553, "reference_mean_power": 0.004505860165227205, "output_mean_power": 0.0044450105698805205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3773993983732174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011178733037365022, "output_power": 0.0010915763362743932, "reference_active": true, "delta_db": -0.1033847250965806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011348254393479664, "output_power": 0.0001064314436097467, "reference_active": true, "delta_db": -0.27859109945553207}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1976543085088728e-05, "output_power": 1.9644843050707874e-05, "reference_active": true, "delta_db": -0.4871081540903818}], "active_band_count": 3, "active_band_mae_db": 0.2896946595474981}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014948818281718847, "log_rms_mae_db": 0.3451696784298132, "rms_envelope_nrmse": 0.02710481632385283, "envelope_correlation": 0.9989253164077638, "rms_gain_error_db": -0.058018704309023755, "reference_mean_power": 0.004512140106290299, "output_mean_power": 0.004452261927148685}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1895509906107845e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010130343468129068, "output_power": 0.0009888454117557145, "reference_active": true, "delta_db": -0.10495767563453998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.38575581348341e-05, "output_power": 5.270818172505108e-05, "reference_active": true, "delta_db": -0.09368624294536476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.096976553551289e-06, "output_power": 1.5823612266926259e-06, "reference_active": true, "delta_db": -1.2228794194316208}], "active_band_count": 3, "active_band_mae_db": 0.4738411126705085}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012259006245866322, "log_rms_mae_db": 1.4277960486250696, "rms_envelope_nrmse": 0.0853601412911151, "envelope_correlation": 0.9930292824242749, "rms_gain_error_db": -0.09873789393196163, "reference_mean_power": 0.001387125934818555, "output_mean_power": 0.0013559450885050318}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.931300849598275e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005535817649712056, "output_power": 0.0005376219207711943, "reference_active": true, "delta_db": -0.1270480784018776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.308179193682493e-05, "output_power": 6.027026385824505e-05, "reference_active": true, "delta_db": -0.8370609476361679}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.782827787780852e-06, "output_power": 3.7732238552900096e-06, "reference_active": true, "delta_db": -1.8542768773937666}], "active_band_count": 3, "active_band_mae_db": 0.9394619678106041}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001104267603656376, "log_rms_mae_db": 0.970595933210074, "rms_envelope_nrmse": 0.05151043074239627, "envelope_correlation": 0.9973471329857236, "rms_gain_error_db": -0.09730831004514648, "reference_mean_power": 0.0013888741987160204, "output_mean_power": 0.001358101031409857}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.150356442217087e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005149581742784205, "output_power": 0.0004994617660701357, "reference_active": true, "delta_db": -0.13269707908048944}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.177504958330739e-05, "output_power": 2.78436695064094e-05, "reference_active": true, "delta_db": -0.5735976670868499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.976808620230845e-07, "output_power": 3.6773069393136783e-07, "reference_active": true, "delta_db": -3.3629929128032376}], "active_band_count": 3, "active_band_mae_db": 1.356429219656859}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.111137261258481e-05, "log_rms_mae_db": 1.2981445263851081, "rms_envelope_nrmse": 0.10721403643956746, "envelope_correlation": 0.9900536799808372, "rms_gain_error_db": -0.18370832711735482, "reference_mean_power": 0.00011418247196706928, "output_mean_power": 0.00010945323654157587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.04035362622924e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.425624857591467e-05, "output_power": 4.0113410765267016e-05, "reference_active": true, "delta_db": -0.4268500698954063}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.422825900093454e-06, "output_power": 3.0152733680378223e-06, "reference_active": true, "delta_db": -0.5505811769950055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0991841496467888e-06, "output_power": 9.055552144176451e-07, "reference_active": true, "delta_db": -0.8415552149303419}], "active_band_count": 3, "active_band_mae_db": 0.606328820606918}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.646632245602786e-05, "log_rms_mae_db": 0.8076658047902205, "rms_envelope_nrmse": 0.06312391822978122, "envelope_correlation": 0.9965813271881645, "rms_gain_error_db": -0.18286756337947063, "reference_mean_power": 0.0001146281153210772, "output_mean_power": 0.00010990169627875025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.508716025767634e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.069122651510939e-05, "output_power": 3.706907141283846e-05, "reference_active": true, "delta_db": -0.4048907354860734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6543344614858788e-06, "output_power": 1.4655893922930855e-06, "reference_active": true, "delta_db": -0.5261100365983146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6696400281137344e-08, "output_power": 4.552042405981631e-08, "reference_active": true, "delta_db": -0.9534918670559511}], "active_band_count": 3, "active_band_mae_db": 0.6281642130467797}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.742254313282893e-05, "log_rms_mae_db": 1.4386221923482665, "rms_envelope_nrmse": 0.18796892174356133, "envelope_correlation": 0.9772008355235045, "rms_gain_error_db": 0.031399331816599436, "reference_mean_power": 4.027652017807587e-05, "output_mean_power": 4.056877315841344e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.067374336554813e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.634873883610167e-05, "output_power": 1.800156454354425e-05, "reference_active": true, "delta_db": 0.4182599571941735}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.842264617940254e-06, "output_power": 7.398794346269221e-06, "reference_active": true, "delta_db": 1.0257973212648406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8971365027113356e-06, "output_power": 3.7881851679096415e-06, "reference_active": true, "delta_db": 3.0033261896995844}], "active_band_count": 3, "active_band_mae_db": 1.4824611560528662}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.2287916076198396e-05, "log_rms_mae_db": 0.8761136184875439, "rms_envelope_nrmse": 0.1384711666930508, "envelope_correlation": 0.9868724474696576, "rms_gain_error_db": 0.031032048899899984, "reference_mean_power": 4.047626840683265e-05, "output_mean_power": 4.0766523021474447e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8536334265621397e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7722714720556858e-05, "output_power": 2.0795068126682088e-05, "reference_active": true, "delta_db": 0.6943010071917993}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2765957035699146e-06, "output_power": 4.714028809854156e-06, "reference_active": true, "delta_db": 1.579693749416018}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7512938438165208e-07, "output_power": 3.5598732952716657e-07, "reference_active": true, "delta_db": 3.0807551959344566}], "active_band_count": 3, "active_band_mae_db": 1.7849166508474248}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9502837969915132, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "75e567ecba67c1caf5b228e00f01618277a15cbbb81089c4e77d2e5d3d2d403c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9892768160704564, "standard_deviation": 0.0012200741974317136, "minimum": 0.9866514174493204, "maximum": 0.9905111325075543, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2831485944584708, "standard_deviation": 0.04265506359527493, "minimum": 1.2146723904017946, "maximum": 1.3342163644814566, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.546678308374144, "standard_deviation": 0.1394460221921951, "minimum": 0.29445944794326057, "maximum": 0.7125551697909441, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11239119768018654, "standard_deviation": 0.007180963193958966, "minimum": 0.10480861422003085, "maximum": 0.12754659569652912, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966614561436128, "standard_deviation": 0.000592953674492289, "minimum": 0.9955659193756886, "maximum": 0.9974116087480849, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7974062921544768, "standard_deviation": 0.033327723609908884, "minimum": 0.737295135700206, "maximum": 0.8305512489580065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6636906280747239, "standard_deviation": 0.16249912645972966, "minimum": 0.3807640499258392, "maximum": 0.9265090348318931, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06534515874977154, "standard_deviation": 0.006751912225285668, "minimum": 0.055980948144409425, "maximum": 0.0794575305722509, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9812751934114927, "standard_deviation": 0.003019649914897679, "minimum": 0.976266161452158, "maximum": 0.9848068084912756, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3933292309073249, "standard_deviation": 0.03401699097512818, "minimum": 1.343143530330153, "maximum": 1.4425591145321426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1610656137661293, "standard_deviation": 0.6771361378982328, "minimum": 0.15879836363741884, "maximum": 2.088332063344111, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17166578367444113, "standard_deviation": 0.013212629840507396, "minimum": 0.15241662883413037, "maximum": 0.1926960597060957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9930171085707022, "standard_deviation": 0.0031105575291496706, "minimum": 0.9868724474696576, "maximum": 0.9964180715690977, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.816458440423333, "standard_deviation": 0.02762903641657129, "minimum": 0.7865673496515063, "maximum": 0.8761136184875439, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.490872956890476, "standard_deviation": 0.7999222678412852, "minimum": 0.3208559302910718, "maximum": 2.7758902103688197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10525535952796147, "standard_deviation": 0.021214431079669742, "minimum": 0.07227166628791683, "maximum": 0.1384711666930508, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.994676356548077, "standard_deviation": 0.0008744552523582772, "minimum": 0.9930292824242749, "maximum": 0.9953789839146183, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4328507672481876, "standard_deviation": 0.028890668532013535, "minimum": 1.3802681602859617, "maximum": 1.483403125324645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5582594154454534, "standard_deviation": 0.21793656671082626, "minimum": 0.24560340601994243, "maximum": 0.9394619678106041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07445563048604537, "standard_deviation": 0.005884899845211238, "minimum": 0.06969825055853758, "maximum": 0.0853601412911151, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998113448327528, "standard_deviation": 0.0004318774664425199, "minimum": 0.9973471329857236, "maximum": 0.9985512803659798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9853726283804844, "standard_deviation": 0.03530480480547208, "minimum": 0.9491682948237135, "maximum": 1.0522081462960926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8807014447615364, "standard_deviation": 0.3805079630829494, "minimum": 0.2411749062068738, "maximum": 1.356429219656859, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04326894299102701, "standard_deviation": 0.00474497908941975, "minimum": 0.037741629070625435, "maximum": 0.05151043074239627, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18297052826677399, "standard_deviation": 0.033380455687704744, "minimum": 0.117549168530567, "maximum": 0.22035129849929247, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.972681415375282, "standard_deviation": 0.0047787026462794355, "minimum": 4.96259894715903, "maximum": 4.977192285566963, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31177202912470536, "standard_deviation": 0.0070682571264873185, "minimum": 0.30317538171823777, "maximum": 0.32685363722053645, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05149300014358832, "standard_deviation": 0.015527351926152452, "minimum": -0.0737508637925373, "maximum": -0.0313814205865188, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967996837770356, "standard_deviation": 0.00039172230260490137, "minimum": 0.9959622181321887, "maximum": 0.9971971534037121, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6287607694613221, "standard_deviation": 0.016862883307428685, "minimum": 0.6026530366132865, "maximum": 0.6524737947736948, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4362255396692191, "standard_deviation": 0.1167331825662716, "minimum": 0.28755661098306046, "maximum": 0.612029667277964, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.048481310778902775, "standard_deviation": 0.0029006092877797234, "minimum": 0.045458775168472396, "maximum": 0.05453744718916201, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989845736469771, "standard_deviation": 0.00015646109486438633, "minimum": 0.9986839565394158, "maximum": 0.999155005971689, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3400575705545523, "standard_deviation": 0.014307273399506952, "minimum": 0.3182185593273999, "maximum": 0.3541402586987137, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6144957542639163, "standard_deviation": 0.1540635331670813, "minimum": 0.38363846899465875, "maximum": 0.8441338417822223, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02618883586729296, "standard_deviation": 0.0020527565554723276, "minimum": 0.023589264176458864, "maximum": 0.02992270820343646, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1995-1836-0014", "language": "en", "style": "normal", "speaker": "librispeech_1995", "source_utterance_id": "1995-1836-0014", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1995-1836-0014.wav", "source_file_sha256": "4853e0b12bcba73372042f38e144e5bbe15822be987af24df9733fa97615e61c", "decoded_pcm_sha256": "af7f0befa255d17d0806307d2e5a0582ff2d024a50adfd0a534360c964de3b82", "original_source_sha256": "c90576861bc6bc0b6837f0b1c340ecc276ac0339ee0a25442a4784f5fc4a4821", "corpus_entry_id": "librispeech_1995-1836-0014"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.4167540073394775, "predicted_mos": 3.4167540073394775, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2fa12ac683c7bb10767e025867eac9e5d76e063b603b6d2cf688cd556c972a24", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.799214733000554, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1995-1836-0014", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "2fa12ac683c7bb10767e025867eac9e5d76e063b603b6d2cf688cd556c972a24", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "9a99a3a90e1f4c343f336f0da5022029b2f18c4ff64d633641d0471804c64565"}, "rows": [{"case": "librispeech_1995-1836-0014", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.6100237369537354, "predicted_mos": 2.6100237369537354, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6cecfe5aa99f42edf610d7f014ce2ebf34030195e0ed5df75e98ce7eaa056c0a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.918155275998288, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1995-1836-0014", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6cecfe5aa99f42edf610d7f014ce2ebf34030195e0ed5df75e98ce7eaa056c0a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5375876830183732, "全帯域_10ms_rms_envelope_nrmse": 0.04817568154433309, "全帯域_10ms_envelope_correlation": 0.9970088796740739, "全帯域_10ms_modulation_mae_db": 0.6242202741246025, "全帯域_40ms_log_rms_mae_db": 0.2941526895473531, "全帯域_40ms_rms_envelope_nrmse": 0.02751309047173752, "全帯域_40ms_envelope_correlation": 0.9989741051495612, "全帯域_40ms_modulation_mae_db": 0.5087942538612675, "300–1000 Hz_10ms_log_rms_mae_db": 1.341215112048321, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07524023709378892, "300–1000 Hz_10ms_envelope_correlation": 0.9950588268962393, "300–1000 Hz_10ms_modulation_mae_db": 0.831825766347405, "300–1000 Hz_40ms_log_rms_mae_db": 0.9095505489780817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04354714263923727, "300–1000 Hz_40ms_envelope_correlation": 0.9982977455123229, "300–1000 Hz_40ms_modulation_mae_db": 0.8665577702764264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4111153996724466, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1691540184216222, "1000–3000 Hz_10ms_envelope_correlation": 0.978970952265362, "1000–3000 Hz_10ms_modulation_mae_db": 0.3378416060307729, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9896742570228033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14287813672126873, "1000–3000 Hz_40ms_envelope_correlation": 0.9843978184773788, "1000–3000 Hz_40ms_modulation_mae_db": 0.2965874307115048, "3000–6000 Hz_10ms_log_rms_mae_db": 1.505458101620603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1824213529757324, "3000–6000 Hz_10ms_envelope_correlation": 0.974554583557369, "3000–6000 Hz_10ms_modulation_mae_db": 0.5500917675517237, "3000–6000 Hz_40ms_log_rms_mae_db": 0.927159131084117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10849254321080712, "3000–6000 Hz_40ms_envelope_correlation": 0.9911711908818146, "3000–6000 Hz_40ms_modulation_mae_db": 1.0330076253136056, "mr_stft_db": 4.911925833317736, "band_mae_db": 0.18539137255632196, "rms_gain_error_db": -0.05974730774072754, "output_peak": 0.45286651001037065}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5364424008752636, "全帯域_10ms_rms_envelope_nrmse": 0.049689103440859576, "全帯域_10ms_envelope_correlation": 0.996804689522229, "全帯域_10ms_modulation_mae_db": 0.24678119215392869, "全帯域_40ms_log_rms_mae_db": 0.3079006850128284, "全帯域_40ms_rms_envelope_nrmse": 0.026818468137015068, "全帯域_40ms_envelope_correlation": 0.9990177250834581, "全帯域_40ms_modulation_mae_db": 0.2656036218783711, "300–1000 Hz_10ms_log_rms_mae_db": 1.4188300869823798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08116475528871443, "300–1000 Hz_10ms_envelope_correlation": 0.9941955051415324, "300–1000 Hz_10ms_modulation_mae_db": 0.6603219472584861, "300–1000 Hz_40ms_log_rms_mae_db": 0.9202631149761561, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047889352018944724, "300–1000 Hz_40ms_envelope_correlation": 0.9978637590258563, "300–1000 Hz_40ms_modulation_mae_db": 0.8191329846573546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3771485354292552, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16261423424935906, "1000–3000 Hz_10ms_envelope_correlation": 0.9805708686907272, "1000–3000 Hz_10ms_modulation_mae_db": 0.176580236922389, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9538052433948243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14005302000045372, "1000–3000 Hz_40ms_envelope_correlation": 0.9849368872567862, "1000–3000 Hz_40ms_modulation_mae_db": 0.1855591625245726, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5215630312371247, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18887545599099373, "3000–6000 Hz_10ms_envelope_correlation": 0.9721874726739278, "3000–6000 Hz_10ms_modulation_mae_db": 0.5985262058906367, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9505096582517023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10752997486672566, "3000–6000 Hz_40ms_envelope_correlation": 0.9905118257138369, "3000–6000 Hz_40ms_modulation_mae_db": 0.9238752565311281, "mr_stft_db": 4.89996171048676, "band_mae_db": 0.10441079727379117, "rms_gain_error_db": -0.05790715447664591, "output_peak": 0.46476225491792605}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5471499291918366, "全帯域_10ms_rms_envelope_nrmse": 0.049739068283898735, "全帯域_10ms_envelope_correlation": 0.9968433593207033, "全帯域_10ms_modulation_mae_db": 0.17546034122155, "全帯域_40ms_log_rms_mae_db": 0.30902524447865276, "全帯域_40ms_rms_envelope_nrmse": 0.027484840257758884, "全帯域_40ms_envelope_correlation": 0.9990231247306425, "全帯域_40ms_modulation_mae_db": 0.12314462986558088, "300–1000 Hz_10ms_log_rms_mae_db": 1.4101927231567764, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08095943848358886, "300–1000 Hz_10ms_envelope_correlation": 0.9944774286450505, "300–1000 Hz_10ms_modulation_mae_db": 0.4492265619629234, "300–1000 Hz_40ms_log_rms_mae_db": 0.9193147568388069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05309786993543177, "300–1000 Hz_40ms_envelope_correlation": 0.9976680396550239, "300–1000 Hz_40ms_modulation_mae_db": 0.5063397601506513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3926846386278333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17217298134356887, "1000–3000 Hz_10ms_envelope_correlation": 0.9781948258206828, "1000–3000 Hz_10ms_modulation_mae_db": 0.715935471326409, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0123071463614015, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14599419651542123, "1000–3000 Hz_40ms_envelope_correlation": 0.9836660579251719, "1000–3000 Hz_40ms_modulation_mae_db": 0.5815516238721533, "3000–6000 Hz_10ms_log_rms_mae_db": 1.477695467804449, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17013589990855138, "3000–6000 Hz_10ms_envelope_correlation": 0.9775723951538556, "3000–6000 Hz_10ms_modulation_mae_db": 0.7976136134729294, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8547052478885926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08256074101497896, "3000–6000 Hz_40ms_envelope_correlation": 0.9946987434072655, "3000–6000 Hz_40ms_modulation_mae_db": 1.2783082290173313, "mr_stft_db": 4.8983205338585245, "band_mae_db": 0.15645439930167068, "rms_gain_error_db": -0.07673243438399581, "output_peak": 0.43945540976879194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5269487433237268, "全帯域_10ms_rms_envelope_nrmse": 0.05599594684083444, "全帯域_10ms_envelope_correlation": 0.9961200796206411, "全帯域_10ms_modulation_mae_db": 0.549760531200703, "全帯域_40ms_log_rms_mae_db": 0.2992339952592994, "全帯域_40ms_rms_envelope_nrmse": 0.03022432520565699, "全帯域_40ms_envelope_correlation": 0.9989631077853821, "全帯域_40ms_modulation_mae_db": 0.6822858026559727, "300–1000 Hz_10ms_log_rms_mae_db": 1.4361082812490482, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0929737266240771, "300–1000 Hz_10ms_envelope_correlation": 0.992828891481871, "300–1000 Hz_10ms_modulation_mae_db": 1.094642067032267, "300–1000 Hz_40ms_log_rms_mae_db": 0.9660299432244925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052766938656177084, "300–1000 Hz_40ms_envelope_correlation": 0.9980052340347674, "300–1000 Hz_40ms_modulation_mae_db": 1.3558379952635173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3912136823695456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16621037369941805, "1000–3000 Hz_10ms_envelope_correlation": 0.9796862243913227, "1000–3000 Hz_10ms_modulation_mae_db": 0.5768761638033307, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9806652134394912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1414339913462037, "1000–3000 Hz_40ms_envelope_correlation": 0.9846500641155157, "1000–3000 Hz_40ms_modulation_mae_db": 0.8387435289165023, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5745614670069208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2246214740721123, "3000–6000 Hz_10ms_envelope_correlation": 0.9604167886017683, "3000–6000 Hz_10ms_modulation_mae_db": 0.791735130548764, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9716857731955729, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1437678431408382, "3000–6000 Hz_40ms_envelope_correlation": 0.9829013475081376, "3000–6000 Hz_40ms_modulation_mae_db": 0.9346906891745813, "mr_stft_db": 4.914412814530357, "band_mae_db": 0.1327677108176197, "rms_gain_error_db": -0.12278371188741091, "output_peak": 0.43453713678223405}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5767999330991889, "全帯域_10ms_rms_envelope_nrmse": 0.05040247833405349, "全帯域_10ms_envelope_correlation": 0.9967219536174371, "全帯域_10ms_modulation_mae_db": 0.5754845268435473, "全帯域_40ms_log_rms_mae_db": 0.31155862640098075, "全帯域_40ms_rms_envelope_nrmse": 0.024670261067746235, "全帯域_40ms_envelope_correlation": 0.9991893718572805, "全帯域_40ms_modulation_mae_db": 0.660255964858469, "300–1000 Hz_10ms_log_rms_mae_db": 1.4379785439282364, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07749778351140339, "300–1000 Hz_10ms_envelope_correlation": 0.9948118981656106, "300–1000 Hz_10ms_modulation_mae_db": 0.5110594742687816, "300–1000 Hz_40ms_log_rms_mae_db": 0.9341921901675481, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04380908112235487, "300–1000 Hz_40ms_envelope_correlation": 0.9983618618137171, "300–1000 Hz_40ms_modulation_mae_db": 0.5093921069293343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4121561418670712, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17089896097411972, "1000–3000 Hz_10ms_envelope_correlation": 0.9786413707441564, "1000–3000 Hz_10ms_modulation_mae_db": 0.7336687466990451, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0073252084358904, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14413231441016813, "1000–3000 Hz_40ms_envelope_correlation": 0.9840763380329166, "1000–3000 Hz_40ms_modulation_mae_db": 0.4364591092414269, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5153567294106576, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18323455457737448, "3000–6000 Hz_10ms_envelope_correlation": 0.9746924536955011, "3000–6000 Hz_10ms_modulation_mae_db": 0.35929739415392953, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9105456740581755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11063331781813178, "3000–6000 Hz_40ms_envelope_correlation": 0.990097046610183, "3000–6000 Hz_40ms_modulation_mae_db": 0.40466929575657895, "mr_stft_db": 4.9112495878530344, "band_mae_db": 0.058446181707426166, "rms_gain_error_db": -0.05281569792730912, "output_peak": 0.45538870778604307}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5686152988710842, "全帯域_10ms_rms_envelope_nrmse": 0.05664089791334566, "全帯域_10ms_envelope_correlation": 0.9959638440647935, "全帯域_10ms_modulation_mae_db": 0.18140748141263904, "全帯域_40ms_log_rms_mae_db": 0.31518827899229834, "全帯域_40ms_rms_envelope_nrmse": 0.033302677413817566, "全帯域_40ms_envelope_correlation": 0.9986189447278927, "全帯域_40ms_modulation_mae_db": 0.1983763751207892, "300–1000 Hz_10ms_log_rms_mae_db": 1.4386050806826378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09750537950245232, "300–1000 Hz_10ms_envelope_correlation": 0.9920918969035796, "300–1000 Hz_10ms_modulation_mae_db": 0.3124238985934468, "300–1000 Hz_40ms_log_rms_mae_db": 0.9859957424431922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06725136003199557, "300–1000 Hz_40ms_envelope_correlation": 0.9963944509254183, "300–1000 Hz_40ms_modulation_mae_db": 0.3062183872403165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4064681525746223, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17603993827724618, "1000–3000 Hz_10ms_envelope_correlation": 0.9772208289574166, "1000–3000 Hz_10ms_modulation_mae_db": 0.6361926967612345, "1000–3000 Hz_40ms_log_rms_mae_db": 1.00387919832587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15164588208974067, "1000–3000 Hz_40ms_envelope_correlation": 0.9823636335299938, "1000–3000 Hz_40ms_modulation_mae_db": 0.501689627590807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5125519330764656, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1856552236882983, "3000–6000 Hz_10ms_envelope_correlation": 0.9730832752445822, "3000–6000 Hz_10ms_modulation_mae_db": 0.8140694506894194, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9088393164732054, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10952436424961082, "3000–6000 Hz_40ms_envelope_correlation": 0.9899505597241403, "3000–6000 Hz_40ms_modulation_mae_db": 0.8448453981529567, "mr_stft_db": 4.912583102201207, "band_mae_db": 0.13451649866038892, "rms_gain_error_db": -0.11296982142302178, "output_peak": 0.4442691429466988}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5585246419578139, "全帯域_10ms_rms_envelope_nrmse": 0.04689760603736866, "全帯域_10ms_envelope_correlation": 0.9972515019505277, "全帯域_10ms_modulation_mae_db": 0.3698069487957065, "全帯域_40ms_log_rms_mae_db": 0.29687125110015106, "全帯域_40ms_rms_envelope_nrmse": 0.02455063039799887, "全帯域_40ms_envelope_correlation": 0.999293649022676, "全帯域_40ms_modulation_mae_db": 0.5979478312966272, "300–1000 Hz_10ms_log_rms_mae_db": 1.4286203351841589, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0785236150240747, "300–1000 Hz_10ms_envelope_correlation": 0.9947523898301022, "300–1000 Hz_10ms_modulation_mae_db": 0.7140397147562422, "300–1000 Hz_40ms_log_rms_mae_db": 0.9510728366020063, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04559377865319584, "300–1000 Hz_40ms_envelope_correlation": 0.9982824218360067, "300–1000 Hz_40ms_modulation_mae_db": 1.0567219676464756, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3739401394523467, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17048275729919807, "1000–3000 Hz_10ms_envelope_correlation": 0.9786313597113807, "1000–3000 Hz_10ms_modulation_mae_db": 0.3142028254702368, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9846035313727214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14374078681445776, "1000–3000 Hz_40ms_envelope_correlation": 0.9841600941154705, "1000–3000 Hz_40ms_modulation_mae_db": 0.20893713817375306, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5381428562670099, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19362127328320877, "3000–6000 Hz_10ms_envelope_correlation": 0.9706977367680623, "3000–6000 Hz_10ms_modulation_mae_db": 0.28804303954771154, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9611887115659549, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11247948348504652, "3000–6000 Hz_40ms_envelope_correlation": 0.9893978627603242, "3000–6000 Hz_40ms_modulation_mae_db": 0.4883808756982311, "mr_stft_db": 4.886344581504752, "band_mae_db": 0.1356959607804392, "rms_gain_error_db": -0.09627340092941067, "output_peak": 0.4507887217499847}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5640710770765187, "全帯域_10ms_rms_envelope_nrmse": 0.05593852885394892, "全帯域_10ms_envelope_correlation": 0.9959987975016199, "全帯域_10ms_modulation_mae_db": 0.4901454181299434, "全帯域_40ms_log_rms_mae_db": 0.31975481279325974, "全帯域_40ms_rms_envelope_nrmse": 0.031817637437855574, "全帯域_40ms_envelope_correlation": 0.998668512594055, "全帯域_40ms_modulation_mae_db": 0.7317990862142655, "300–1000 Hz_10ms_log_rms_mae_db": 1.398839946039187, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08280688905942314, "300–1000 Hz_10ms_envelope_correlation": 0.9939514356283924, "300–1000 Hz_10ms_modulation_mae_db": 0.6185528641361565, "300–1000 Hz_40ms_log_rms_mae_db": 0.9404728172962361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048752231378848924, "300–1000 Hz_40ms_envelope_correlation": 0.9977768826190063, "300–1000 Hz_40ms_modulation_mae_db": 0.6891289337946654, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3942790263708131, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1726109711176907, "1000–3000 Hz_10ms_envelope_correlation": 0.9780756939592455, "1000–3000 Hz_10ms_modulation_mae_db": 0.3161181189873984, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0179632000927725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14748078923514615, "1000–3000 Hz_40ms_envelope_correlation": 0.9833176033790239, "1000–3000 Hz_40ms_modulation_mae_db": 0.5232611524354184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.492433443349815, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18276318502543215, "3000–6000 Hz_10ms_envelope_correlation": 0.9740664840653334, "3000–6000 Hz_10ms_modulation_mae_db": 1.7780990049287373, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9240962195332901, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11049793273994087, "3000–6000 Hz_40ms_envelope_correlation": 0.9900758810552958, "3000–6000 Hz_40ms_modulation_mae_db": 1.6492110466076808, "mr_stft_db": 4.909975183914834, "band_mae_db": 0.11583897404611414, "rms_gain_error_db": -0.08874300575435477, "output_peak": 0.44093466461562564}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.9343540957880885, "spectral_convergence": 0.1505567481045096, "spectral_convergence_status": "ok", "amplitude_floor": 1.7246496227499867e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.770495614925955, "spectral_convergence": 0.15803709066370572, "spectral_convergence_status": "ok", "amplitude_floor": 1.4526318235172953e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.030927789239166, "spectral_convergence": 0.17106179532478977, "spectral_convergence_status": "ok", "amplitude_floor": 1.3962213853324247e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0014645134106370877, "output_power": 0.0014697429482101094, "delta_db": 0.015480322356314458, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024771538682985465, "output_power": 0.002459384312263095, "delta_db": -0.031265851416571615, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013427666217707404, "output_power": 0.0012927575836744803, "delta_db": -0.16483442959673927, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024269361162520585, "output_power": 0.0002350063347629937, "delta_db": -0.13978775490345183, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.4425202041437504e-05, "output_power": 4.175075876929983e-05, "delta_db": -0.26965038944560876, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.4446805694883255e-05, "output_power": 3.076210190857747e-05, "delta_db": -0.4913294876192458, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18539137255632196, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05974730774072754, "pitch": null, "output_rms_dbfs": -22.57321674529947, "output_peak": 0.45286651001037065, "reference_rms_dbfs": -22.513469437558737, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.911925833317736}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002110151907366841, "log_rms_mae_db": 0.5375876830183732, "rms_envelope_nrmse": 0.04817568154433309, "envelope_correlation": 0.9970088796740739, "rms_gain_error_db": -0.05906675766872124, "reference_mean_power": 0.005604891233518836, "output_mean_power": 0.005529177263412521}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2250047113785757e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011643562088722453, "output_power": 0.0011410689612295236, "reference_active": true, "delta_db": -0.08773971298780318}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011188245530999946, "output_power": 0.00010554248802527523, "reference_active": true, "delta_db": -0.25334660705971795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8450792038001933e-05, "output_power": 1.9995713749219857e-05, "reference_active": true, "delta_db": -1.5315745023262863}], "active_band_count": 3, "active_band_mae_db": 0.6242202741246025}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020713167740975364, "log_rms_mae_db": 0.2941526895473531, "rms_envelope_nrmse": 0.02751309047173752, "envelope_correlation": 0.9989741051495612, "rms_gain_error_db": -0.0570870118335166, "reference_mean_power": 0.005603505754865554, "output_mean_power": 0.0055303309457996165}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.048317169639576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010458478077031374, "output_power": 0.0010278332089513058, "reference_active": true, "delta_db": -0.07545844891585013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.131979407845485e-05, "output_power": 5.773260792126139e-05, "reference_active": true, "delta_db": -0.2617951168354096}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6292932861003563e-06, "output_power": 1.2390429139393395e-06, "reference_active": true, "delta_db": -1.189129195832543}], "active_band_count": 3, "active_band_mae_db": 0.5087942538612675}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015831513319989338, "log_rms_mae_db": 1.341215112048321, "rms_envelope_nrmse": 0.07524023709378892, "envelope_correlation": 0.9950588268962393, "rms_gain_error_db": -0.09314305180689417, "reference_mean_power": 0.0022309508329157127, "output_mean_power": 0.0021836131144880055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0339679622838414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046453953265101344, "output_power": 0.00043777947534634796, "reference_active": true, "delta_db": -0.2576728223323032}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.019812862224435e-05, "output_power": 6.217874847882464e-05, "reference_active": true, "delta_db": -0.5268355823450355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.545275737486218e-05, "output_power": 1.0420992811598691e-05, "reference_active": true, "delta_db": -1.7109688943648766}], "active_band_count": 3, "active_band_mae_db": 0.831825766347405}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014375880677218868, "log_rms_mae_db": 0.9095505489780817, "rms_envelope_nrmse": 0.04354714263923727, "envelope_correlation": 0.9982977455123229, "rms_gain_error_db": -0.08908752632127118, "reference_mean_power": 0.0022214967634666704, "output_mean_power": 0.0021763910546699224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.581275628581243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00040407180276463134, "output_power": 0.00038399349450838233, "reference_active": true, "delta_db": -0.2213467848802322}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0351036816934105e-05, "output_power": 3.494606500134909e-05, "reference_active": true, "delta_db": -0.6245641798170788}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1841518228550588e-06, "output_power": 7.907353477869677e-07, "reference_active": true, "delta_db": -1.7537623461319682}], "active_band_count": 3, "active_band_mae_db": 0.8665577702764264}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.952734804355257e-05, "log_rms_mae_db": 1.4111153996724466, "rms_envelope_nrmse": 0.1691540184216222, "envelope_correlation": 0.978970952265362, "rms_gain_error_db": -0.17089995709566877, "reference_mean_power": 0.00025078805819662737, "output_mean_power": 0.0002411109071844088}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4212041991971396e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.171024783262431e-05, "output_power": 6.550143676712284e-05, "reference_active": true, "delta_db": -0.3933039713817773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.275733290040072e-06, "output_power": 7.220763185522176e-06, "reference_active": true, "delta_db": -0.03293668580800977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0724449404458702e-06, "output_power": 1.2277315622922734e-06, "reference_active": true, "delta_db": 0.5872841609025317}], "active_band_count": 3, "active_band_mae_db": 0.3378416060307729}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.039214481073785e-05, "log_rms_mae_db": 0.9896742570228033, "rms_envelope_nrmse": 0.14287813672126873, "envelope_correlation": 0.9843978184773788, "rms_gain_error_db": -0.1722253140056696, "reference_mean_power": 0.0002518169764916092, "output_mean_power": 0.00024202625107969634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3611130645010071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.488974762253847e-05, "output_power": 5.996065485058866e-05, "reference_active": true, "delta_db": -0.34309717769644354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.711335389958727e-06, "output_power": 3.639480818042997e-06, "reference_active": true, "delta_db": -0.08490768150085418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.486818337389537e-08, "output_power": 9.438884012138976e-08, "reference_active": true, "delta_db": 0.46175743293721655}], "active_band_count": 3, "active_band_mae_db": 0.2965874307115048}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.8930089256612995e-05, "log_rms_mae_db": 1.505458101620603, "rms_envelope_nrmse": 0.1824213529757324, "envelope_correlation": 0.974554583557369, "rms_gain_error_db": -0.3713833971799274, "reference_mean_power": 2.8173794881918634e-05, "output_mean_power": 2.5864673987688395e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9380195589864167e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.123435668533653e-06, "output_power": 5.64072726385441e-06, "reference_active": true, "delta_db": -0.35660058043291576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.722245443537651e-06, "output_power": 2.1321935132047145e-06, "reference_active": true, "delta_db": -1.0610066180505369}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.909737577930936e-07, "output_power": 6.549297873957571e-07, "reference_active": true, "delta_db": -0.23266810417171868}], "active_band_count": 3, "active_band_mae_db": 0.5500917675517237}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3848385775447824e-05, "log_rms_mae_db": 0.927159131084117, "rms_envelope_nrmse": 0.10849254321080712, "envelope_correlation": 0.9911711908818146, "rms_gain_error_db": -0.37117186841259725, "reference_mean_power": 2.8310170166312676e-05, "output_mean_power": 2.5991137871207343e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.80878248957878e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.951898475340433e-06, "output_power": 5.537292007843014e-06, "reference_active": true, "delta_db": -0.3135808807461953}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2501096733423988e-06, "output_power": 9.712449413743915e-07, "reference_active": true, "delta_db": -1.0961934592006304}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.567916579161296e-08, "output_power": 4.451463066955063e-08, "reference_active": true, "delta_db": -1.689248535993991}], "active_band_count": 3, "active_band_mae_db": 1.0330076253136056}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9474907073725379, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "819b0567266b87771a9458b876dc84f44021d3e5efaafd7c8c2c20744e18279a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9787490155675367, "standard_deviation": 0.0010259516813498014, "minimum": 0.9772208289574166, "maximum": 0.9805708686907272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3948757145454918, "standard_deviation": 0.014480744233589833, "minimum": 1.3739401394523467, "maximum": 1.4121561418670712, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4759269832501021, "standard_deviation": 0.21390263112032376, "minimum": 0.176580236922389, "maximum": 0.7336687466990451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17002302942277786, "standard_deviation": 0.00411761552272023, "minimum": 0.16261423424935906, "maximum": 0.17603993827724618, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9839460621040321, "standard_deviation": 0.0008212935404488172, "minimum": 0.9823636335299938, "maximum": 0.9849368872567862, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9937778748057218, "standard_deviation": 0.020991808794732284, "minimum": 0.9538052433948243, "maximum": 1.0179632000927725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4465985966832673, "standard_deviation": 0.21650894100083184, "minimum": 0.1855591625245726, "maximum": 0.8387435289165023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14466988964160749, "standard_deviation": 0.00367362290310297, "minimum": 0.14005302000045372, "maximum": 0.15164588208974067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.97215889872005, "standard_deviation": 0.005154338877537981, "minimum": 0.9604167886017683, "maximum": 0.9775723951538556, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5172203787216305, "standard_deviation": 0.029483260595715006, "minimum": 1.477695467804449, "maximum": 1.5745614670069208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7471844508479815, "standard_deviation": 0.4619830374821234, "minimum": 0.28804303954771154, "maximum": 1.7780990049287373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18891605244021292, "standard_deviation": 0.015913441465583107, "minimum": 0.17013589990855138, "maximum": 0.2246214740721123, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9898505572076247, "standard_deviation": 0.0032580501364285397, "minimum": 0.9829013475081376, "maximum": 0.9946987434072655, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9260912165063264, "standard_deviation": 0.03693106893958352, "minimum": 0.8547052478885926, "maximum": 0.9716857731955729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9446235520315117, "standard_deviation": 0.4010332294171924, "minimum": 0.40466929575657895, "maximum": 1.6492110466076808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11068577506576, "standard_deviation": 0.01649594933321075, "minimum": 0.08256074101497896, "maximum": 0.1437678431408382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9940210340865472, "standard_deviation": 0.0010436005995084698, "minimum": 0.9920918969035796, "maximum": 0.9950588268962393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4137987636588432, "standard_deviation": 0.03261874821092308, "minimum": 1.341215112048321, "maximum": 1.4386050806826378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6490115367944636, "standard_deviation": 0.24187599478550045, "minimum": 0.3124238985934468, "maximum": 1.094642067032267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08333397807344037, "standard_deviation": 0.007811022997225419, "minimum": 0.07524023709378892, "maximum": 0.09750537950245232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978312994277649, "standard_deviation": 0.0006356139053356707, "minimum": 0.9963944509254183, "maximum": 0.9983618618137171, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9408614938158151, "standard_deviation": 0.025879897876514867, "minimum": 0.9095505489780817, "maximum": 0.9859957424431922, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7636662382448427, "standard_deviation": 0.3371940674776975, "minimum": 0.3062183872403165, "maximum": 1.3558379952635173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05033846930452325, "standard_deviation": 0.00773986961419732, "minimum": 0.04354714263923727, "maximum": 0.06725136003199557, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1279402368929715, "standard_deviation": 0.03735905363713588, "minimum": 0.058446181707426166, "maximum": 0.18539137255632196, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.905596668458401, "standard_deviation": 0.0098056663591904, "minimum": 4.886344581504752, "maximum": 4.914412814530357, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44787531857220936, "standard_deviation": 0.00989226743023379, "minimum": 0.43453713678223405, "maximum": 0.46476225491792605, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08349656681535957, "standard_deviation": 0.02621024786035456, "minimum": -0.12278371188741091, "maximum": -0.05281569792730912, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9965891381590032, "standard_deviation": 0.0004932617641003702, "minimum": 0.9959638440647935, "maximum": 0.9972515019505277, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5520174634267258, "standard_deviation": 0.01764825637767102, "minimum": 0.5269487433237268, "maximum": 0.5767999330991889, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4016333392353275, "standard_deviation": 0.18290942028165522, "minimum": 0.17546034122155, "maximum": 0.6242202741246025, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05168491390608032, "standard_deviation": 0.0038893125827239895, "minimum": 0.04689760603736866, "maximum": 0.05664089791334566, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989685676188685, "standard_deviation": 0.00023057537454010686, "minimum": 0.9986189447278927, "maximum": 0.999293649022676, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.30671069794810296, "standard_deviation": 0.009123856405511825, "minimum": 0.2941526895473531, "maximum": 0.31975481279325974, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47102594571891787, "standard_deviation": 0.24009953222962893, "minimum": 0.12314462986558088, "maximum": 0.7317990862142655, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028297741298698338, "standard_deviation": 0.003203011237993913, "minimum": 0.02455063039799887, "maximum": 0.033302677413817566, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0014", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.9289023876190186, "predicted_mos": 2.9289023876190186, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6bb444b3f8da1e0385474e21de97263c242f880c4555f5df9b49fad9f95d7f52", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.235314799938351, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5181592945313893, "全帯域_10ms_rms_envelope_nrmse": 0.049352529098432685, "全帯域_10ms_envelope_correlation": 0.9968477513106869, "全帯域_10ms_modulation_mae_db": 0.6634550901383576, "全帯域_40ms_log_rms_mae_db": 0.2902341355939292, "全帯域_40ms_rms_envelope_nrmse": 0.029154130219659354, "全帯域_40ms_envelope_correlation": 0.998827835042143, "全帯域_40ms_modulation_mae_db": 0.5098417665617685, "300–1000 Hz_10ms_log_rms_mae_db": 1.3220628627063888, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07988831017792881, "300–1000 Hz_10ms_envelope_correlation": 0.9944344609462319, "300–1000 Hz_10ms_modulation_mae_db": 0.9817786547608028, "300–1000 Hz_40ms_log_rms_mae_db": 0.8837795611526946, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04595859549293506, "300–1000 Hz_40ms_envelope_correlation": 0.9981184726890322, "300–1000 Hz_40ms_modulation_mae_db": 1.010054371076906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3586831727373545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1685433699023794, "1000–3000 Hz_10ms_envelope_correlation": 0.9791880511146611, "1000–3000 Hz_10ms_modulation_mae_db": 0.37297380949586084, "1000–3000 Hz_40ms_log_rms_mae_db": 0.989880989266975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14300854434654847, "1000–3000 Hz_40ms_envelope_correlation": 0.9844277705652535, "1000–3000 Hz_40ms_modulation_mae_db": 0.2844245563289201, "3000–6000 Hz_10ms_log_rms_mae_db": 1.408288223049827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15895413983795958, "3000–6000 Hz_10ms_envelope_correlation": 0.9826549310879297, "3000–6000 Hz_10ms_modulation_mae_db": 0.5425768013295983, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8792291159274515, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09736940806645178, "3000–6000 Hz_40ms_envelope_correlation": 0.9952930629992054, "3000–6000 Hz_40ms_modulation_mae_db": 1.2919279131482797, "mr_stft_db": 4.792416480161649, "band_mae_db": 0.23342144174604643, "rms_gain_error_db": -0.05277826899166489, "output_peak": 0.4511805698547147}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5583304016109799, "全帯域_10ms_rms_envelope_nrmse": 0.05056876751151745, "全帯域_10ms_envelope_correlation": 0.9967089770620454, "全帯域_10ms_modulation_mae_db": 0.36127448785315774, "全帯域_40ms_log_rms_mae_db": 0.3300251008341769, "全帯域_40ms_rms_envelope_nrmse": 0.0283033986211338, "全帯域_40ms_envelope_correlation": 0.9989245665012412, "全帯域_40ms_modulation_mae_db": 0.40984247554230685, "300–1000 Hz_10ms_log_rms_mae_db": 1.4297271177411326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0841697007956498, "300–1000 Hz_10ms_envelope_correlation": 0.9937830737352991, "300–1000 Hz_10ms_modulation_mae_db": 0.8707583414238406, "300–1000 Hz_40ms_log_rms_mae_db": 0.9110827702039594, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05116343153514321, "300–1000 Hz_40ms_envelope_correlation": 0.9975927260540726, "300–1000 Hz_40ms_modulation_mae_db": 1.1381445164953596, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3209054315289321, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16333684934245263, "1000–3000 Hz_10ms_envelope_correlation": 0.9804256607692594, "1000–3000 Hz_10ms_modulation_mae_db": 0.2016596195434696, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9317020496151618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14076500848806947, "1000–3000 Hz_40ms_envelope_correlation": 0.9847819007269442, "1000–3000 Hz_40ms_modulation_mae_db": 0.2562705571825229, "3000–6000 Hz_10ms_log_rms_mae_db": 1.40313002455328, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15989124370141425, "3000–6000 Hz_10ms_envelope_correlation": 0.9809907098652052, "3000–6000 Hz_10ms_modulation_mae_db": 0.691445006594261, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9174446032326135, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09225771105616963, "3000–6000 Hz_40ms_envelope_correlation": 0.9942883364793501, "3000–6000 Hz_40ms_modulation_mae_db": 1.146779549485628, "mr_stft_db": 4.784339072529064, "band_mae_db": 0.1353116024735352, "rms_gain_error_db": -0.06938234061101733, "output_peak": 0.4687239519810401}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5448494970687545, "全帯域_10ms_rms_envelope_nrmse": 0.04853457595035153, "全帯域_10ms_envelope_correlation": 0.9969959368008967, "全帯域_10ms_modulation_mae_db": 0.1403272562834705, "全帯域_40ms_log_rms_mae_db": 0.3112051620916392, "全帯域_40ms_rms_envelope_nrmse": 0.026178114354974295, "全帯域_40ms_envelope_correlation": 0.9991201312736063, "全帯域_40ms_modulation_mae_db": 0.17464428228506754, "300–1000 Hz_10ms_log_rms_mae_db": 1.3981100284176184, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07865330147080803, "300–1000 Hz_10ms_envelope_correlation": 0.9947977707340621, "300–1000 Hz_10ms_modulation_mae_db": 0.4548149729209057, "300–1000 Hz_40ms_log_rms_mae_db": 0.9057398928924977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050010129875362415, "300–1000 Hz_40ms_envelope_correlation": 0.9979598578411573, "300–1000 Hz_40ms_modulation_mae_db": 0.46181006126744434, "1000–3000 Hz_10ms_log_rms_mae_db": 1.339979214291836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1703142519167461, "1000–3000 Hz_10ms_envelope_correlation": 0.9787216740557345, "1000–3000 Hz_10ms_modulation_mae_db": 0.6265930722168579, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9992061280901579, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14528471602590334, "1000–3000 Hz_40ms_envelope_correlation": 0.9838714369718211, "1000–3000 Hz_40ms_modulation_mae_db": 0.4951063636074558, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4051586369921083, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15907296632488635, "3000–6000 Hz_10ms_envelope_correlation": 0.9809316478584496, "3000–6000 Hz_10ms_modulation_mae_db": 0.5597858489351661, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8732347212075483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08628156182422482, "3000–6000 Hz_40ms_envelope_correlation": 0.9948679649630243, "3000–6000 Hz_40ms_modulation_mae_db": 1.2555557683959693, "mr_stft_db": 4.7904571436593155, "band_mae_db": 0.1961001360580578, "rms_gain_error_db": -0.07326855694183457, "output_peak": 0.4388721911453643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5299324859185107, "全帯域_10ms_rms_envelope_nrmse": 0.05741462189900468, "全帯域_10ms_envelope_correlation": 0.9959148868220704, "全帯域_10ms_modulation_mae_db": 0.5254626779527801, "全帯域_40ms_log_rms_mae_db": 0.30860993454463537, "全帯域_40ms_rms_envelope_nrmse": 0.030292023059614735, "全帯域_40ms_envelope_correlation": 0.9989724395893781, "全帯域_40ms_modulation_mae_db": 0.6028139896696422, "300–1000 Hz_10ms_log_rms_mae_db": 1.4366055427390245, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09775251423584001, "300–1000 Hz_10ms_envelope_correlation": 0.9920078357639099, "300–1000 Hz_10ms_modulation_mae_db": 1.0980160077500851, "300–1000 Hz_40ms_log_rms_mae_db": 0.9575124307081342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054399029801373636, "300–1000 Hz_40ms_envelope_correlation": 0.997867653224624, "300–1000 Hz_40ms_modulation_mae_db": 1.3712264304998483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3497686307226782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16550319462811305, "1000–3000 Hz_10ms_envelope_correlation": 0.9798886074625712, "1000–3000 Hz_10ms_modulation_mae_db": 0.554619771068748, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9857418955530783, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14208863046950204, "1000–3000 Hz_40ms_envelope_correlation": 0.9845010579951771, "1000–3000 Hz_40ms_modulation_mae_db": 0.8577656370694348, "3000–6000 Hz_10ms_log_rms_mae_db": 1.444924176282065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18577401832214216, "3000–6000 Hz_10ms_envelope_correlation": 0.9746569722296456, "3000–6000 Hz_10ms_modulation_mae_db": 0.8497687202343358, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9090938365141302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11559909212051087, "3000–6000 Hz_40ms_envelope_correlation": 0.9913361646910309, "3000–6000 Hz_40ms_modulation_mae_db": 1.0300126432486858, "mr_stft_db": 4.806068765169942, "band_mae_db": 0.17094259282150306, "rms_gain_error_db": -0.12538637615787715, "output_peak": 0.43598461796235294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5728330966356989, "全帯域_10ms_rms_envelope_nrmse": 0.050190540924830454, "全帯域_10ms_envelope_correlation": 0.996767308607314, "全帯域_10ms_modulation_mae_db": 0.5503407198685313, "全帯域_40ms_log_rms_mae_db": 0.3103934461771986, "全帯域_40ms_rms_envelope_nrmse": 0.025488188455324884, "全帯域_40ms_envelope_correlation": 0.9991519055618161, "全帯域_40ms_modulation_mae_db": 0.5648974294421708, "300–1000 Hz_10ms_log_rms_mae_db": 1.4599678762651827, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08217686263400599, "300–1000 Hz_10ms_envelope_correlation": 0.9941951917723096, "300–1000 Hz_10ms_modulation_mae_db": 0.5972826670554606, "300–1000 Hz_40ms_log_rms_mae_db": 0.9674606916982734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04863201836762644, "300–1000 Hz_40ms_envelope_correlation": 0.9980003658320308, "300–1000 Hz_40ms_modulation_mae_db": 0.578087731610828, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3372935086210362, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16695623648791066, "1000–3000 Hz_10ms_envelope_correlation": 0.9796011798897705, "1000–3000 Hz_10ms_modulation_mae_db": 0.4555044477166203, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9746152810817748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1423937880143492, "1000–3000 Hz_40ms_envelope_correlation": 0.9844362266159687, "1000–3000 Hz_40ms_modulation_mae_db": 0.2005407696720557, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4116726353039564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14867499297965828, "3000–6000 Hz_10ms_envelope_correlation": 0.9829268683955843, "3000–6000 Hz_10ms_modulation_mae_db": 0.5504383726964591, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8492267159120376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08151307767236543, "3000–6000 Hz_40ms_envelope_correlation": 0.9945698284248167, "3000–6000 Hz_40ms_modulation_mae_db": 0.6876025470528698, "mr_stft_db": 4.797136608456396, "band_mae_db": 0.09725837480747891, "rms_gain_error_db": -0.06083019804064303, "output_peak": 0.4577348276695034}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5546479550697115, "全帯域_10ms_rms_envelope_nrmse": 0.05410996586448218, "全帯域_10ms_envelope_correlation": 0.9963490680255475, "全帯域_10ms_modulation_mae_db": 0.15884884637211652, "全帯域_40ms_log_rms_mae_db": 0.3029860260870597, "全帯域_40ms_rms_envelope_nrmse": 0.030633860539646434, "全帯域_40ms_envelope_correlation": 0.9988816165377027, "全帯域_40ms_modulation_mae_db": 0.1704447857679551, "300–1000 Hz_10ms_log_rms_mae_db": 1.4310757628205606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09706453319801565, "300–1000 Hz_10ms_envelope_correlation": 0.9922080376392963, "300–1000 Hz_10ms_modulation_mae_db": 0.3075811043964079, "300–1000 Hz_40ms_log_rms_mae_db": 0.9659661894637986, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06445045141356463, "300–1000 Hz_40ms_envelope_correlation": 0.9968001552498603, "300–1000 Hz_40ms_modulation_mae_db": 0.5475149972815688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3494121732103426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17662689362522013, "1000–3000 Hz_10ms_envelope_correlation": 0.9771503041102617, "1000–3000 Hz_10ms_modulation_mae_db": 0.6816155160307753, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9906988895430704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15256325075228921, "1000–3000 Hz_40ms_envelope_correlation": 0.9822206305854457, "1000–3000 Hz_40ms_modulation_mae_db": 0.5266126531609997, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3940620320067976, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15438398577529575, "3000–6000 Hz_10ms_envelope_correlation": 0.9819701754814517, "3000–6000 Hz_10ms_modulation_mae_db": 0.5877786606690208, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8719753981057786, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08817392267913149, "3000–6000 Hz_40ms_envelope_correlation": 0.9943633150432593, "3000–6000 Hz_40ms_modulation_mae_db": 0.7456832279960919, "mr_stft_db": 4.79893537110053, "band_mae_db": 0.17067088152082197, "rms_gain_error_db": -0.11583142823898766, "output_peak": 0.44456477711115666}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5588688756170648, "全帯域_10ms_rms_envelope_nrmse": 0.04744160408366591, "全帯域_10ms_envelope_correlation": 0.9971929061965032, "全帯域_10ms_modulation_mae_db": 0.31414863008709065, "全帯域_40ms_log_rms_mae_db": 0.29650894588605026, "全帯域_40ms_rms_envelope_nrmse": 0.02525190930959432, "全帯域_40ms_envelope_correlation": 0.9992559873107274, "全帯域_40ms_modulation_mae_db": 0.6054463433425642, "300–1000 Hz_10ms_log_rms_mae_db": 1.4130918655184421, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0816224722587865, "300–1000 Hz_10ms_envelope_correlation": 0.9942925956280034, "300–1000 Hz_10ms_modulation_mae_db": 0.6705222956486887, "300–1000 Hz_40ms_log_rms_mae_db": 0.9307304260673239, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04679989526229066, "300–1000 Hz_40ms_envelope_correlation": 0.9981683241177073, "300–1000 Hz_40ms_modulation_mae_db": 1.0480886757246692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3121250913537097, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17095298824161154, "1000–3000 Hz_10ms_envelope_correlation": 0.9785396345484256, "1000–3000 Hz_10ms_modulation_mae_db": 0.3367604246936536, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9535633207006224, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14491461686061774, "1000–3000 Hz_40ms_envelope_correlation": 0.9838969087978024, "1000–3000 Hz_40ms_modulation_mae_db": 0.20267158969319873, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4301927894402826, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17813207282281343, "3000–6000 Hz_10ms_envelope_correlation": 0.9758026280887597, "3000–6000 Hz_10ms_modulation_mae_db": 0.30326272935779225, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8772425969311713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1024412033885177, "3000–6000 Hz_40ms_envelope_correlation": 0.9922334880137578, "3000–6000 Hz_40ms_modulation_mae_db": 0.49733028003740554, "mr_stft_db": 4.776030588650773, "band_mae_db": 0.16046309709214926, "rms_gain_error_db": -0.09680133077635565, "output_peak": 0.45302085830466965}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.572526963410635, "全帯域_10ms_rms_envelope_nrmse": 0.05554653684829872, "全帯域_10ms_envelope_correlation": 0.9960469390475157, "全帯域_10ms_modulation_mae_db": 0.43572588856236544, "全帯域_40ms_log_rms_mae_db": 0.319864499193206, "全帯域_40ms_rms_envelope_nrmse": 0.032137405404514714, "全帯域_40ms_envelope_correlation": 0.9986299270245643, "全帯域_40ms_modulation_mae_db": 0.5761063424064169, "300–1000 Hz_10ms_log_rms_mae_db": 1.4085284498572883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08267930576221412, "300–1000 Hz_10ms_envelope_correlation": 0.993963706334097, "300–1000 Hz_10ms_modulation_mae_db": 0.6209452021805344, "300–1000 Hz_40ms_log_rms_mae_db": 0.9196891332760054, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048537123684996866, "300–1000 Hz_40ms_envelope_correlation": 0.9977838455262577, "300–1000 Hz_40ms_modulation_mae_db": 0.720636885756546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2941183245751668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17077094190548125, "1000–3000 Hz_10ms_envelope_correlation": 0.9785800888236641, "1000–3000 Hz_10ms_modulation_mae_db": 0.30177724583452986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9777575194732921, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14627345658948832, "1000–3000 Hz_40ms_envelope_correlation": 0.9836146336300806, "1000–3000 Hz_40ms_modulation_mae_db": 0.4721329903464217, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3868401321772927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1590355629154617, "3000–6000 Hz_10ms_envelope_correlation": 0.9812453328019214, "3000–6000 Hz_10ms_modulation_mae_db": 1.698112840979306, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8561936770956214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09582035951978905, "3000–6000 Hz_40ms_envelope_correlation": 0.9937528576153744, "3000–6000 Hz_40ms_modulation_mae_db": 1.849455540131353, "mr_stft_db": 4.78488781270783, "band_mae_db": 0.15017488550798672, "rms_gain_error_db": -0.08586222677475813, "output_peak": 0.44231421760147765}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.806565491820261, "spectral_convergence": 0.15124407156852657, "spectral_convergence_status": "ok", "amplitude_floor": 1.7246496227499867e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.651070496097567, "spectral_convergence": 0.16010068214618794, "spectral_convergence_status": "ok", "amplitude_floor": 1.4526318235172953e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.91961345256712, "spectral_convergence": 0.17283202917697188, "spectral_convergence_status": "ok", "amplitude_floor": 1.3962213853324247e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0014645134106370877, "output_power": 0.0014740371515121911, "delta_db": 0.028150766872605218, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024771538682985465, "output_power": 0.002474163693566989, "delta_db": -0.005245539303388365, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013427666217707404, "output_power": 0.001286474807982043, "delta_db": -0.18599250918594412, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024269361162520585, "output_power": 0.00023319619463519533, "delta_db": -0.173368854383061, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.4425202041437504e-05, "output_power": 4.03128362043183e-05, "delta_db": -0.4218605765089448, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.4446805694883255e-05, "output_power": 3.009940529643757e-05, "delta_db": -0.585910404222335, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23342144174604643, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05277826899166489, "pitch": null, "output_rms_dbfs": -22.566247706550403, "output_peak": 0.4511805698547147, "reference_rms_dbfs": -22.513469437558737, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.792416480161649}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002110151907366841, "log_rms_mae_db": 0.5181592945313893, "rms_envelope_nrmse": 0.049352529098432685, "envelope_correlation": 0.9968477513106869, "rms_gain_error_db": -0.05161708485233693, "reference_mean_power": 0.005604891233518836, "output_mean_power": 0.005538669879968599}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2250047113785757e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011643562088722453, "output_power": 0.0011455066358338367, "reference_active": true, "delta_db": -0.0708825400581347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011188245530999946, "output_power": 0.00010598730510368396, "reference_active": true, "delta_db": -0.23508138923835825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8450792038001933e-05, "output_power": 1.9304307508273575e-05, "reference_active": true, "delta_db": -1.6844013411185799}], "active_band_count": 3, "active_band_mae_db": 0.6634550901383576}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020713167740975364, "log_rms_mae_db": 0.2902341355939292, "rms_envelope_nrmse": 0.029154130219659354, "envelope_correlation": 0.998827835042143, "rms_gain_error_db": -0.049055728087864896, "reference_mean_power": 0.005603505754865554, "output_mean_power": 0.005540567490935552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.048317169639576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010458478077031374, "output_power": 0.0010323796627815934, "reference_active": true, "delta_db": -0.05629049745765191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.131979407845485e-05, "output_power": 5.776510706622631e-05, "reference_active": true, "delta_db": -0.25935105121924157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6292932861003563e-06, "output_power": 1.2320005247218646e-06, "reference_active": true, "delta_db": -1.213883751008412}], "active_band_count": 3, "active_band_mae_db": 0.5098417665617685}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015831513319989338, "log_rms_mae_db": 1.3220628627063888, "rms_envelope_nrmse": 0.07988831017792881, "envelope_correlation": 0.9944344609462319, "rms_gain_error_db": -0.10483328027043486, "reference_mean_power": 0.0022309508329157127, "output_mean_power": 0.00217774322397052}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0339679622838414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046453953265101344, "output_power": 0.00043737107984195297, "reference_active": true, "delta_db": -0.26172615751711514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.019812862224435e-05, "output_power": 6.240278548751436e-05, "reference_active": true, "delta_db": -0.5112155882088353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.545275737486218e-05, "output_power": 9.37058196683511e-06, "reference_active": true, "delta_db": -2.1723942185564584}], "active_band_count": 3, "active_band_mae_db": 0.9817786547608028}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014375880677218868, "log_rms_mae_db": 0.8837795611526946, "rms_envelope_nrmse": 0.04595859549293506, "envelope_correlation": 0.9981184726890322, "rms_gain_error_db": -0.10075889521066778, "reference_mean_power": 0.0022214967634666704, "output_mean_power": 0.002170550003929654}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.581275628581243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00040407180276463134, "output_power": 0.00038343669518195667, "reference_active": true, "delta_db": -0.22764872401876185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0351036816934105e-05, "output_power": 3.5115952646450215e-05, "reference_active": true, "delta_db": -0.6035024365520429}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1841518228550588e-06, "output_power": 7.136843814587532e-07, "reference_active": true, "delta_db": -2.1990119526599137}], "active_band_count": 3, "active_band_mae_db": 1.010054371076906}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.952734804355257e-05, "log_rms_mae_db": 1.3586831727373545, "rms_envelope_nrmse": 0.1685433699023794, "envelope_correlation": 0.9791880511146611, "rms_gain_error_db": -0.20957912836282772, "reference_mean_power": 0.00025078805819662737, "output_mean_power": 0.00023897305741650204}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4212041991971396e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.171024783262431e-05, "output_power": 6.513697062007654e-05, "reference_active": true, "delta_db": -0.41753666838725145}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.275733290040072e-06, "output_power": 7.478655149940232e-06, "reference_active": true, "delta_db": 0.1194673728102147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0724449404458702e-06, "output_power": 1.2262153358224563e-06, "reference_active": true, "delta_db": 0.5819173872901163}], "active_band_count": 3, "active_band_mae_db": 0.37297380949586084}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.039214481073785e-05, "log_rms_mae_db": 0.989880989266975, "rms_envelope_nrmse": 0.14300854434654847, "envelope_correlation": 0.9844277705652535, "rms_gain_error_db": -0.20982788082396125, "reference_mean_power": 0.0002518169764916092, "output_mean_power": 0.00023993975831848916}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3611130645010071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.488974762253847e-05, "output_power": 5.967371644735646e-05, "reference_active": true, "delta_db": -0.3639299873270318}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.711335389958727e-06, "output_power": 3.79762932395279e-06, "reference_active": true, "delta_db": 0.09982369277590837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.486818337389537e-08, "output_power": 9.283182853830599e-08, "reference_active": true, "delta_db": 0.38951998888382017}], "active_band_count": 3, "active_band_mae_db": 0.2844245563289201}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.8930089256612995e-05, "log_rms_mae_db": 1.408288223049827, "rms_envelope_nrmse": 0.15895413983795958, "envelope_correlation": 0.9826549310879297, "rms_gain_error_db": -0.588593830495903, "reference_mean_power": 2.8173794881918634e-05, "output_mean_power": 2.4602880963424754e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9380195589864167e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.123435668533653e-06, "output_power": 5.680695035246483e-06, "reference_active": true, "delta_db": -0.3259368453990266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.722245443537651e-06, "output_power": 2.0629377356804825e-06, "reference_active": true, "delta_db": -1.204411594219211}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.909737577930936e-07, "output_power": 6.756524522553356e-07, "reference_active": true, "delta_db": -0.0973819643705574}], "active_band_count": 3, "active_band_mae_db": 0.5425768013295983}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3848385775447824e-05, "log_rms_mae_db": 0.8792291159274515, "rms_envelope_nrmse": 0.09736940806645178, "envelope_correlation": 0.9952930629992054, "rms_gain_error_db": -0.5882257699970941, "reference_mean_power": 2.8310170166312676e-05, "output_mean_power": 2.4724066487811783e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.80878248957878e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.951898475340433e-06, "output_power": 5.516143639341682e-06, "reference_active": true, "delta_db": -0.3301994801424824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2501096733423988e-06, "output_power": 9.101862504077621e-07, "reference_active": true, "delta_db": -1.378178451430738}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.567916579161296e-08, "output_power": 3.987374684769059e-08, "reference_active": true, "delta_db": -2.1674058078716185}], "active_band_count": 3, "active_band_mae_db": 1.2919279131482797}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.948267487666155, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "21e0beab40b15c6ca4be27aa547ae5219074840ddd80097d55945e798ad00c81", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9790119000967936, "standard_deviation": 0.0010074636750263951, "minimum": 0.9771503041102617, "maximum": 0.9804256607692594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3327856933801319, "standard_deviation": 0.02194399715764375, "minimum": 1.2941183245751668, "maximum": 1.3586831727373545, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4414379883250644, "standard_deviation": 0.16802174874946244, "minimum": 0.2016596195434696, "maximum": 0.6816155160307753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16912559075623934, "standard_deviation": 0.0040603550511443135, "minimum": 0.16333684934245263, "maximum": 0.17662689362522013, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9839688207360616, "standard_deviation": 0.0008079737448887424, "minimum": 0.9822206305854457, "maximum": 0.9847819007269442, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9753957591655166, "standard_deviation": 0.022392494209019807, "minimum": 0.9317020496151618, "maximum": 0.9992061280901579, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.41194063963262617, "standard_deviation": 0.2240619245106547, "minimum": 0.2005407696720557, "maximum": 0.8577656370694348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14466150144334597, "standard_deviation": 0.003687151806774556, "minimum": 0.14076500848806947, "maximum": 0.15256325075228921, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9801474082261183, "standard_deviation": 0.003137086453250547, "minimum": 0.9746569722296456, "maximum": 0.9829268683955843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4105335812257014, "standard_deviation": 0.018867699957798228, "minimum": 1.3868401321772927, "maximum": 1.444924176282065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7228961225994924, "standard_deviation": 0.42280425803333965, "minimum": 0.30326272935779225, "maximum": 1.698112840979306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16298987283495395, "standard_deviation": 0.012445584129590978, "minimum": 0.14867499297965828, "maximum": 0.18577401832214216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9938381272787273, "standard_deviation": 0.001364512239522764, "minimum": 0.9913361646910309, "maximum": 0.9952930629992054, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.879205083115794, "standard_deviation": 0.023529893697653237, "minimum": 0.8492267159120376, "maximum": 0.9174446032326135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0630434336870354, "standard_deviation": 0.427233727776159, "minimum": 0.49733028003740554, "maximum": 1.849455540131353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0949320420408951, "standard_deviation": 0.010676863908540087, "minimum": 0.08151307767236543, "maximum": 0.11559909212051087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937103340691511, "standard_deviation": 0.001035503835663127, "minimum": 0.9920078357639099, "maximum": 0.9947977707340621, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4123961882582048, "standard_deviation": 0.04119308784287123, "minimum": 1.3220628627063888, "maximum": 1.4599678762651827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7002124057670907, "standard_deviation": 0.26692742102626316, "minimum": 0.3075811043964079, "maximum": 1.0980160077500851, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08550087506665612, "standard_deviation": 0.007540702371261401, "minimum": 0.07865330147080803, "maximum": 0.09775251423584001, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977864250668427, "standard_deviation": 0.0004389604950909483, "minimum": 0.9968001552498603, "maximum": 0.9981683241177073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.930245136932836, "standard_deviation": 0.030805644556298688, "minimum": 0.8837795611526946, "maximum": 0.9674606916982734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8594454587141462, "standard_deviation": 0.3276990667183699, "minimum": 0.46181006126744434, "maximum": 1.3712264304998483, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05124383442916161, "standard_deviation": 0.005948933802248058, "minimum": 0.04595859549293506, "maximum": 0.06445045141356463, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16429287650344743, "standard_deviation": 0.04039604207998896, "minimum": 0.09725837480747891, "maximum": 0.23342144174604643, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.791283980304437, "standard_deviation": 0.009525617914382242, "minimum": 4.776030588650773, "maximum": 4.806068765169942, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4490495014537849, "standard_deviation": 0.010840253354633234, "minimum": 0.43598461796235294, "maximum": 0.4687239519810401, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08501759081664231, "standard_deviation": 0.02598242821319307, "minimum": -0.12538637615787715, "maximum": -0.05277826899166489, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9966029717340724, "standard_deviation": 0.00045476921621973954, "minimum": 0.9959148868220704, "maximum": 0.9971929061965032, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5512685712328431, "standard_deviation": 0.019378410510272476, "minimum": 0.5181592945313893, "maximum": 0.5728330966356989, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.39369794963973376, "standard_deviation": 0.1863806965270634, "minimum": 0.1403272562834705, "maximum": 0.6634550901383576, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05164489277257295, "standard_deviation": 0.0035948178706182578, "minimum": 0.04744160408366591, "maximum": 0.05741462189900468, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989705511051474, "standard_deviation": 0.00020112456679307998, "minimum": 0.9986299270245643, "maximum": 0.9992559873107274, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.30872840630098686, "standard_deviation": 0.012597112298780078, "minimum": 0.2902341355939292, "maximum": 0.3300251008341769, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4517546768772365, "standard_deviation": 0.1835214396723003, "minimum": 0.1704447857679551, "maximum": 0.6054463433425642, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028429878745557816, "standard_deviation": 0.0025755688112893793, "minimum": 0.02525190930959432, "maximum": 0.032137405404514714, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0014", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.0853970050811768, "predicted_mos": 3.0853970050811768, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c167cddbd1f0e6c36506320a91c6e151196dd7f18ded9da93fdcc842859d1ea0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.3484475002624094, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5529920149692437, "全帯域_10ms_rms_envelope_nrmse": 0.054169426608527395, "全帯域_10ms_envelope_correlation": 0.9963131486061303, "全帯域_10ms_modulation_mae_db": 0.3576123372252415, "全帯域_40ms_log_rms_mae_db": 0.3299772688500289, "全帯域_40ms_rms_envelope_nrmse": 0.03199130302557721, "全帯域_40ms_envelope_correlation": 0.9987251038058503, "全帯域_40ms_modulation_mae_db": 0.5506362055571464, "300–1000 Hz_10ms_log_rms_mae_db": 1.565584749872344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08382144269568544, "300–1000 Hz_10ms_envelope_correlation": 0.9940177878766591, "300–1000 Hz_10ms_modulation_mae_db": 0.7699277795892727, "300–1000 Hz_40ms_log_rms_mae_db": 1.0380168665081462, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05006256139940078, "300–1000 Hz_40ms_envelope_correlation": 0.997922814350492, "300–1000 Hz_40ms_modulation_mae_db": 1.1367631469115518, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6678161564611687, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1727394762796738, "1000–3000 Hz_10ms_envelope_correlation": 0.9780502554383447, "1000–3000 Hz_10ms_modulation_mae_db": 0.3955899136091052, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1908383361488255, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14691436472568742, "1000–3000 Hz_40ms_envelope_correlation": 0.9834702864802746, "1000–3000 Hz_40ms_modulation_mae_db": 0.3968944282513232, "3000–6000 Hz_10ms_log_rms_mae_db": 2.02558084365274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25460246628191996, "3000–6000 Hz_10ms_envelope_correlation": 0.948930433674864, "3000–6000 Hz_10ms_modulation_mae_db": 0.5692006860038331, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3765711359331394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15170391097579938, "3000–6000 Hz_40ms_envelope_correlation": 0.9804414107114052, "3000–6000 Hz_40ms_modulation_mae_db": 1.0404142309428417, "mr_stft_db": 5.246304298521493, "band_mae_db": 0.09543699866707456, "rms_gain_error_db": -0.10767781794652391, "output_peak": 0.44946727782472823}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5423854734257201, "全帯域_10ms_rms_envelope_nrmse": 0.05335883720570378, "全帯域_10ms_envelope_correlation": 0.9964452840748774, "全帯域_10ms_modulation_mae_db": 0.5256666094013459, "全帯域_40ms_log_rms_mae_db": 0.31070751466486823, "全帯域_40ms_rms_envelope_nrmse": 0.03372313450201533, "全帯域_40ms_envelope_correlation": 0.9985798621340575, "全帯域_40ms_modulation_mae_db": 0.3494333263294566, "300–1000 Hz_10ms_log_rms_mae_db": 1.535714361724628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0902510023962857, "300–1000 Hz_10ms_envelope_correlation": 0.9934739154436943, "300–1000 Hz_10ms_modulation_mae_db": 0.681991257512494, "300–1000 Hz_40ms_log_rms_mae_db": 1.042216454621299, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06291876689399727, "300–1000 Hz_40ms_envelope_correlation": 0.997073594427217, "300–1000 Hz_40ms_modulation_mae_db": 0.9221785542149399, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7355205360878772, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17458324326145216, "1000–3000 Hz_10ms_envelope_correlation": 0.9777254711624642, "1000–3000 Hz_10ms_modulation_mae_db": 0.5221281731239236, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2485716056814444, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14710304765205398, "1000–3000 Hz_40ms_envelope_correlation": 0.983462843547862, "1000–3000 Hz_40ms_modulation_mae_db": 0.37535981385872014, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0169200299088996, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2340514457867273, "3000–6000 Hz_10ms_envelope_correlation": 0.957141956868655, "3000–6000 Hz_10ms_modulation_mae_db": 1.0239079854013127, "3000–6000 Hz_40ms_log_rms_mae_db": 1.331799110835195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13231128219317154, "3000–6000 Hz_40ms_envelope_correlation": 0.9851344977457057, "3000–6000 Hz_40ms_modulation_mae_db": 1.1991556839800355, "mr_stft_db": 5.242958013762809, "band_mae_db": 0.1105290491811111, "rms_gain_error_db": -0.10955181474902356, "output_peak": 0.44284355178626666}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5499064713298678, "全帯域_10ms_rms_envelope_nrmse": 0.05007702473533924, "全帯域_10ms_envelope_correlation": 0.9968053182840141, "全帯域_10ms_modulation_mae_db": 0.44018515736675123, "全帯域_40ms_log_rms_mae_db": 0.30738817410197417, "全帯域_40ms_rms_envelope_nrmse": 0.02882872611553275, "全帯域_40ms_envelope_correlation": 0.998914757156931, "全帯域_40ms_modulation_mae_db": 0.34597323433367855, "300–1000 Hz_10ms_log_rms_mae_db": 1.4735188479350512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08348695424062251, "300–1000 Hz_10ms_envelope_correlation": 0.9940462361677354, "300–1000 Hz_10ms_modulation_mae_db": 1.1690859257111879, "300–1000 Hz_40ms_log_rms_mae_db": 1.0057478081680007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054117787235995564, "300–1000 Hz_40ms_envelope_correlation": 0.9974881033331625, "300–1000 Hz_40ms_modulation_mae_db": 1.620584849187317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.681394277913055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17740265222681303, "1000–3000 Hz_10ms_envelope_correlation": 0.9768310703409788, "1000–3000 Hz_10ms_modulation_mae_db": 1.3282680609083861, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2186412916888987, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15043214202669436, "1000–3000 Hz_40ms_envelope_correlation": 0.9826683720967229, "1000–3000 Hz_40ms_modulation_mae_db": 1.4156088395075388, "3000–6000 Hz_10ms_log_rms_mae_db": 1.997359007268777, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2248614443281617, "3000–6000 Hz_10ms_envelope_correlation": 0.960682508641124, "3000–6000 Hz_10ms_modulation_mae_db": 0.535336043492313, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3379233535252177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12782356687938815, "3000–6000 Hz_40ms_envelope_correlation": 0.9861713761969512, "3000–6000 Hz_40ms_modulation_mae_db": 1.0432718700281813, "mr_stft_db": 5.24373967182206, "band_mae_db": 0.09530260602646949, "rms_gain_error_db": -0.08120478943923215, "output_peak": 0.458212755378475}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5752209435024047, "全帯域_10ms_rms_envelope_nrmse": 0.05203514294612316, "全帯域_10ms_envelope_correlation": 0.9965526260284577, "全帯域_10ms_modulation_mae_db": 0.6602612288365209, "全帯域_40ms_log_rms_mae_db": 0.32081522755003394, "全帯域_40ms_rms_envelope_nrmse": 0.025573982464774745, "全帯域_40ms_envelope_correlation": 0.9991787650528632, "全帯域_40ms_modulation_mae_db": 0.881487664628993, "300–1000 Hz_10ms_log_rms_mae_db": 1.659369608058226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07966965916750865, "300–1000 Hz_10ms_envelope_correlation": 0.9945825887690422, "300–1000 Hz_10ms_modulation_mae_db": 0.8459954286923548, "300–1000 Hz_40ms_log_rms_mae_db": 1.0999080964738803, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04372201202860496, "300–1000 Hz_40ms_envelope_correlation": 0.9984017313795872, "300–1000 Hz_40ms_modulation_mae_db": 1.162245481586739, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6968093445253434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17069247925076092, "1000–3000 Hz_10ms_envelope_correlation": 0.9785636151233742, "1000–3000 Hz_10ms_modulation_mae_db": 0.4374753721245863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.20312748389881, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1435261007709918, "1000–3000 Hz_40ms_envelope_correlation": 0.9842186700568566, "1000–3000 Hz_40ms_modulation_mae_db": 0.4121244694092614, "3000–6000 Hz_10ms_log_rms_mae_db": 2.057286707676477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23904846109347463, "3000–6000 Hz_10ms_envelope_correlation": 0.9552207012220781, "3000–6000 Hz_10ms_modulation_mae_db": 0.7284673532078507, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3413340585800462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14328695467113453, "3000–6000 Hz_40ms_envelope_correlation": 0.9825394644028312, "3000–6000 Hz_40ms_modulation_mae_db": 1.249018341257819, "mr_stft_db": 5.254018579768417, "band_mae_db": 0.11880211178270605, "rms_gain_error_db": -0.08279515854982966, "output_peak": 0.46705439271355786}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5441461454274006, "全帯域_10ms_rms_envelope_nrmse": 0.05250824140070207, "全帯域_10ms_envelope_correlation": 0.9963981001272472, "全帯域_10ms_modulation_mae_db": 0.2513853259070577, "全帯域_40ms_log_rms_mae_db": 0.3260861146900136, "全帯域_40ms_rms_envelope_nrmse": 0.02975067348774516, "全帯域_40ms_envelope_correlation": 0.9987382025621158, "全帯域_40ms_modulation_mae_db": 0.19712242734758914, "300–1000 Hz_10ms_log_rms_mae_db": 1.4890769654706377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08280664176907851, "300–1000 Hz_10ms_envelope_correlation": 0.9939098731413044, "300–1000 Hz_10ms_modulation_mae_db": 0.5590690343020424, "300–1000 Hz_40ms_log_rms_mae_db": 1.0099028931518803, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04826144966278107, "300–1000 Hz_40ms_envelope_correlation": 0.9977703291636295, "300–1000 Hz_40ms_modulation_mae_db": 0.429237483926338, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6417215042149533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1736004642525463, "1000–3000 Hz_10ms_envelope_correlation": 0.9780357157886986, "1000–3000 Hz_10ms_modulation_mae_db": 0.6603803917605909, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2223675989421932, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15002595249386985, "1000–3000 Hz_40ms_envelope_correlation": 0.9828189341094136, "1000–3000 Hz_40ms_modulation_mae_db": 0.6987466238277721, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9472640197545896, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25409298009780185, "3000–6000 Hz_10ms_envelope_correlation": 0.9493676443288807, "3000–6000 Hz_10ms_modulation_mae_db": 1.2833746723742523, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2902754265497443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1495643435504473, "3000–6000 Hz_40ms_envelope_correlation": 0.9810134006330077, "3000–6000 Hz_40ms_modulation_mae_db": 0.5803399526164211, "mr_stft_db": 5.26400896520242, "band_mae_db": 0.03159753355814606, "rms_gain_error_db": -0.031053503795889222, "output_peak": 0.4559289220637316}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.542223879245589, "全帯域_10ms_rms_envelope_nrmse": 0.0452770643511135, "全帯域_10ms_envelope_correlation": 0.9973259156384922, "全帯域_10ms_modulation_mae_db": 0.22244873565188028, "全帯域_40ms_log_rms_mae_db": 0.30448726794186004, "全帯域_40ms_rms_envelope_nrmse": 0.023333397574600998, "全帯域_40ms_envelope_correlation": 0.9992317772835908, "全帯域_40ms_modulation_mae_db": 0.26702755299209, "300–1000 Hz_10ms_log_rms_mae_db": 1.5937254696416208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0738976539920843, "300–1000 Hz_10ms_envelope_correlation": 0.9953109292848882, "300–1000 Hz_10ms_modulation_mae_db": 1.2104263938289934, "300–1000 Hz_40ms_log_rms_mae_db": 1.0911100150276274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04065408338509372, "300–1000 Hz_40ms_envelope_correlation": 0.9986064430300416, "300–1000 Hz_40ms_modulation_mae_db": 1.3490330139686115, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6809814551222286, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16856290221146455, "1000–3000 Hz_10ms_envelope_correlation": 0.979125158133545, "1000–3000 Hz_10ms_modulation_mae_db": 0.23963888362872132, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1676638256142373, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14276797186095574, "1000–3000 Hz_40ms_envelope_correlation": 0.9843780470115157, "1000–3000 Hz_40ms_modulation_mae_db": 0.2783400913683189, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9928652030011427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24520494878539786, "3000–6000 Hz_10ms_envelope_correlation": 0.9530817301469614, "3000–6000 Hz_10ms_modulation_mae_db": 0.9229088528400409, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3134508881711213, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1382345235721496, "3000–6000 Hz_40ms_envelope_correlation": 0.9837694029403888, "3000–6000 Hz_40ms_modulation_mae_db": 1.3596604396720842, "mr_stft_db": 5.254853332879873, "band_mae_db": 0.0549271695609587, "rms_gain_error_db": -0.03239600347309057, "output_peak": 0.47013434381830194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5610624481962051, "全帯域_10ms_rms_envelope_nrmse": 0.05541303335698139, "全帯域_10ms_envelope_correlation": 0.9961681810267111, "全帯域_10ms_modulation_mae_db": 0.744325336807353, "全帯域_40ms_log_rms_mae_db": 0.29875335137477493, "全帯域_40ms_rms_envelope_nrmse": 0.030570225247469152, "全帯域_40ms_envelope_correlation": 0.998897630503709, "全帯域_40ms_modulation_mae_db": 0.8871181886785245, "300–1000 Hz_10ms_log_rms_mae_db": 1.5265018370962664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08733725731174126, "300–1000 Hz_10ms_envelope_correlation": 0.9935676460105007, "300–1000 Hz_10ms_modulation_mae_db": 0.9269908040739677, "300–1000 Hz_40ms_log_rms_mae_db": 1.0269499259038242, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05338155528285336, "300–1000 Hz_40ms_envelope_correlation": 0.997715341648476, "300–1000 Hz_40ms_modulation_mae_db": 1.1491779677267917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.746773728314951, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1670325137715055, "1000–3000 Hz_10ms_envelope_correlation": 0.9796681774533639, "1000–3000 Hz_10ms_modulation_mae_db": 0.9870792197011022, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2718019117320052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1420746792298508, "1000–3000 Hz_40ms_envelope_correlation": 0.9845815259582016, "1000–3000 Hz_40ms_modulation_mae_db": 0.7603068674420902, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0715562938916827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22727539671687616, "3000–6000 Hz_10ms_envelope_correlation": 0.9602761538759194, "3000–6000 Hz_10ms_modulation_mae_db": 0.5977820499275373, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4097386262697515, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1287708628730165, "3000–6000 Hz_40ms_envelope_correlation": 0.9862180761642826, "3000–6000 Hz_40ms_modulation_mae_db": 0.8962597807765603, "mr_stft_db": 5.248938193370557, "band_mae_db": 0.13952814874053085, "rms_gain_error_db": -0.11698320928830763, "output_peak": 0.48308416791742187}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5415813466375289, "全帯域_10ms_rms_envelope_nrmse": 0.04989615089115872, "全帯域_10ms_envelope_correlation": 0.996899570703893, "全帯域_10ms_modulation_mae_db": 0.33558994163115424, "全帯域_40ms_log_rms_mae_db": 0.30416630336766903, "全帯域_40ms_rms_envelope_nrmse": 0.028161902499303195, "全帯域_40ms_envelope_correlation": 0.9990539366572535, "全帯域_40ms_modulation_mae_db": 0.39061245256492544, "300–1000 Hz_10ms_log_rms_mae_db": 1.5362825701274971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09331373619438958, "300–1000 Hz_10ms_envelope_correlation": 0.9925106605003987, "300–1000 Hz_10ms_modulation_mae_db": 1.0650881895734157, "300–1000 Hz_40ms_log_rms_mae_db": 1.0636918489509959, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05571439322653725, "300–1000 Hz_40ms_envelope_correlation": 0.9973488806272555, "300–1000 Hz_40ms_modulation_mae_db": 1.2819216107226097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6924177760669608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17353889275893092, "1000–3000 Hz_10ms_envelope_correlation": 0.9778518199476177, "1000–3000 Hz_10ms_modulation_mae_db": 0.4160173296399674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.226520063210724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1481633037936577, "1000–3000 Hz_40ms_envelope_correlation": 0.9831898736284373, "1000–3000 Hz_40ms_modulation_mae_db": 0.3909280241017304, "3000–6000 Hz_10ms_log_rms_mae_db": 2.024970681429765, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24820803124004168, "3000–6000 Hz_10ms_envelope_correlation": 0.9517571448349139, "3000–6000 Hz_10ms_modulation_mae_db": 0.5054120371863334, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3566536971540661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1526347433790736, "3000–6000 Hz_40ms_envelope_correlation": 0.9801829713725732, "3000–6000 Hz_40ms_modulation_mae_db": 1.2392082602611099, "mr_stft_db": 5.2510284010811015, "band_mae_db": 0.10656850507517059, "rms_gain_error_db": -0.1075412021729643, "output_peak": 0.4600932769573412}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.297032374803988, "spectral_convergence": 0.15513969112210058, "spectral_convergence_status": "ok", "amplitude_floor": 1.7246496227499867e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1698218866116665, "spectral_convergence": 0.16185416413315548, "spectral_convergence_status": "ok", "amplitude_floor": 1.4526318235172953e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.272058634148824, "spectral_convergence": 0.16359231300163785, "spectral_convergence_status": "ok", "amplitude_floor": 1.3962213853324247e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0014645134106370877, "output_power": 0.0014640290426612297, "delta_db": -0.001436607762383566, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024771538682985465, "output_power": 0.0023860160620528827, "delta_db": -0.16279620701357184, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013427666217707404, "output_power": 0.0013030094816085286, "delta_db": -0.13052961148568526, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024269361162520585, "output_power": 0.00023619278337652273, "delta_db": -0.11791720570172798, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.4425202041437504e-05, "output_power": 4.403902707330503e-05, "delta_db": -0.037916958011038886, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.4446805694883255e-05, "output_power": 3.542839642190214e-05, "delta_db": 0.12202540202803981, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09543699866707456, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10767781794652391, "pitch": null, "output_rms_dbfs": -22.621147255505264, "output_peak": 0.44946727782472823, "reference_rms_dbfs": -22.513469437558737, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.246304298521493}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002110151907366841, "log_rms_mae_db": 0.5529920149692437, "rms_envelope_nrmse": 0.054169426608527395, "envelope_correlation": 0.9963131486061303, "rms_gain_error_db": -0.106340016221202, "reference_mean_power": 0.005604891233518836, "output_mean_power": 0.005469318157923911}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2250047113785757e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011643562088722453, "output_power": 0.0011470341695925144, "reference_active": true, "delta_db": -0.06509507857505391}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011188245530999946, "output_power": 0.00010929028768558094, "reference_active": true, "delta_db": -0.10180419429233581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8450792038001933e-05, "output_power": 2.309407581882697e-05, "reference_active": true, "delta_db": -0.9059377388083347}], "active_band_count": 3, "active_band_mae_db": 0.3576123372252415}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020713167740975364, "log_rms_mae_db": 0.3299772688500289, "rms_envelope_nrmse": 0.03199130302557721, "envelope_correlation": 0.9987251038058503, "rms_gain_error_db": -0.10547221031969223, "reference_mean_power": 0.005603505754865554, "output_mean_power": 0.005469058908210972}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.048317169639576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010458478077031374, "output_power": 0.001032704900355126, "reference_active": true, "delta_db": -0.054922525538419274}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.131979407845485e-05, "output_power": 6.0447893794716996e-05, "reference_active": true, "delta_db": -0.06219514556404238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6292932861003563e-06, "output_power": 1.1442480298632355e-06, "reference_active": true, "delta_db": -1.5347909455689777}], "active_band_count": 3, "active_band_mae_db": 0.5506362055571464}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015831513319989338, "log_rms_mae_db": 1.565584749872344, "rms_envelope_nrmse": 0.08382144269568544, "envelope_correlation": 0.9940177878766591, "rms_gain_error_db": -0.16043111092341106, "reference_mean_power": 0.0022309508329157127, "output_mean_power": 0.0021500417243459366}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0339679622838414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046453953265101344, "output_power": 0.0004505980525443012, "reference_active": true, "delta_db": -0.13233368278477317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.019812862224435e-05, "output_power": 6.0560596123825436e-05, "reference_active": true, "delta_db": -0.6413539317962152}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.545275737486218e-05, "output_power": 1.0849167475698127e-05, "reference_active": true, "delta_db": -1.5360957241868298}], "active_band_count": 3, "active_band_mae_db": 0.7699277795892727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014375880677218868, "log_rms_mae_db": 1.0380168665081462, "rms_envelope_nrmse": 0.05006256139940078, "envelope_correlation": 0.997922814350492, "rms_gain_error_db": -0.15756959678020582, "reference_mean_power": 0.0022214967634666704, "output_mean_power": 0.0021423416205217814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.581275628581243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00040407180276463134, "output_power": 0.0003961916467108561, "reference_active": true, "delta_db": -0.08553230592230848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0351036816934105e-05, "output_power": 3.489908287917897e-05, "reference_active": true, "delta_db": -0.6304068417923745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1841518228550588e-06, "output_power": 6.367549922193579e-07, "reference_active": true, "delta_db": -2.6943502930199728}], "active_band_count": 3, "active_band_mae_db": 1.1367631469115518}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.952734804355257e-05, "log_rms_mae_db": 1.6678161564611687, "rms_envelope_nrmse": 0.1727394762796738, "envelope_correlation": 0.9780502554383447, "rms_gain_error_db": -0.11030264151777369, "reference_mean_power": 0.00025078805819662737, "output_mean_power": 0.00024449871932618385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4212041991971396e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.171024783262431e-05, "output_power": 6.948100224626683e-05, "reference_active": true, "delta_db": -0.13715149027505713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.275733290040072e-06, "output_power": 6.19245253518228e-06, "reference_active": true, "delta_db": -0.7001408412054169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0724449404458702e-06, "output_power": 1.1623120642311316e-06, "reference_active": true, "delta_db": 0.3494774093468417}], "active_band_count": 3, "active_band_mae_db": 0.3955899136091052}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.039214481073785e-05, "log_rms_mae_db": 1.1908383361488255, "rms_envelope_nrmse": 0.14691436472568742, "envelope_correlation": 0.9834702864802746, "rms_gain_error_db": -0.10978007569469721, "reference_mean_power": 0.0002518169764916092, "output_mean_power": 0.0002455313759378748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3611130645010071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.488974762253847e-05, "output_power": 6.333361550961367e-05, "reference_active": true, "delta_db": -0.10541803857705498}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.711335389958727e-06, "output_power": 2.9278543928021136e-06, "reference_active": true, "delta_db": -1.0298072819700297}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.486818337389537e-08, "output_power": 8.379133496562216e-08, "reference_active": true, "delta_db": -0.055457964206885105}], "active_band_count": 3, "active_band_mae_db": 0.3968944282513232}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.8930089256612995e-05, "log_rms_mae_db": 2.02558084365274, "rms_envelope_nrmse": 0.25460246628191996, "envelope_correlation": 0.948930433674864, "rms_gain_error_db": -0.15564312660238516, "reference_mean_power": 2.8173794881918634e-05, "output_mean_power": 2.7181976693104085e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9380195589864167e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.123435668533653e-06, "output_power": 5.795176499604191e-06, "reference_active": true, "delta_db": -0.23928492098697754}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.722245443537651e-06, "output_power": 2.068271651200443e-06, "reference_active": true, "delta_db": -1.1931970023284326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.909737577930936e-07, "output_power": 6.485590745892778e-07, "reference_active": true, "delta_db": -0.275120134696089}], "active_band_count": 3, "active_band_mae_db": 0.5692006860038331}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3848385775447824e-05, "log_rms_mae_db": 1.3765711359331394, "rms_envelope_nrmse": 0.15170391097579938, "envelope_correlation": 0.9804414107114052, "rms_gain_error_db": -0.15520881296977182, "reference_mean_power": 2.8310170166312676e-05, "output_mean_power": 2.7316282692624244e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.80878248957878e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.951898475340433e-06, "output_power": 5.707027977394328e-06, "reference_active": true, "delta_db": -0.18245513028675414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2501096733423988e-06, "output_power": 8.987744622327259e-07, "reference_active": true, "delta_db": -1.43297391874502}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.567916579161296e-08, "output_power": 4.6435087011381294e-08, "reference_active": true, "delta_db": -1.5058136437967509}], "active_band_count": 3, "active_band_mae_db": 1.0404142309428417}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9376130844790965, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d83495ee72d276434eda7d178a7123df7dc66785a68a4ad41239837f922d85d9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9782314104235483, "standard_deviation": 0.0008787632348625787, "minimum": 0.9768310703409788, "maximum": 0.9796681774533639, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6929293473383171, "standard_deviation": 0.034345447428873424, "minimum": 1.6417215042149533, "maximum": 1.746773728314951, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6233221680620479, "standard_deviation": 0.36203208442632284, "minimum": 0.23963888362872132, "maximum": 1.3282680609083861, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1722690780016434, "standard_deviation": 0.0033585338239704703, "minimum": 0.1670325137715055, "maximum": 0.17740265222681303, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9835985691111605, "standard_deviation": 0.0007204484936755803, "minimum": 0.9826683720967229, "maximum": 0.9845815259582016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2186915146146422, "standard_deviation": 0.03253613389770384, "minimum": 1.1676638256142373, "maximum": 1.2718019117320052, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5910386447208443, "standard_deviation": 0.3730593612030214, "minimum": 0.2783400913683189, "maximum": 1.4156088395075388, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1463759453192202, "standard_deviation": 0.00323879351603785, "minimum": 0.1420746792298508, "maximum": 0.15043214202669436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9545572841991745, "standard_deviation": 0.00456627933606856, "minimum": 0.948930433674864, "maximum": 0.960682508641124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.016725348323009, "standard_deviation": 0.03885382413271474, "minimum": 1.9472640197545896, "maximum": 2.0715562938916827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7707987100541842, "standard_deviation": 0.27984106284730315, "minimum": 0.5054120371863334, "maximum": 1.2833746723742523, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24091814679130014, "standard_deviation": 0.011507423380868258, "minimum": 0.2248614443281617, "maximum": 0.25460246628191996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9831838250208932, "standard_deviation": 0.0025051965734965223, "minimum": 0.9801829713725732, "maximum": 0.9862180761642826, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3447182871272851, "standard_deviation": 0.03693799251774978, "minimum": 1.2902754265497443, "maximum": 1.4097386262697515, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0759160699418817, "standard_deviation": 0.24857470858482134, "minimum": 0.5803399526164211, "maximum": 1.3596604396720842, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1405412735117726, "standard_deviation": 0.010234029561899661, "minimum": 0.12782356687938815, "maximum": 0.1526347433790736, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9939274546492778, "standard_deviation": 0.0008201260147609196, "minimum": 0.9925106605003987, "maximum": 0.9953109292848882, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5474718012407838, "standard_deviation": 0.059246183408824996, "minimum": 1.4735188479350512, "maximum": 1.659369608058226, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.903571851660466, "standard_deviation": 0.23306937706119596, "minimum": 0.5590690343020424, "maximum": 1.2104263938289934, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08432304347092448, "standard_deviation": 0.0060786698693946735, "minimum": 0.0738976539920843, "maximum": 0.09331373619438958, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977909047449827, "standard_deviation": 0.000515939262717241, "minimum": 0.997073594427217, "maximum": 0.9986064430300416, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0471929886007068, "standard_deviation": 0.03506484420787815, "minimum": 1.0057478081680007, "maximum": 1.0999080964738803, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1313927635306125, "standard_deviation": 0.3479377563156891, "minimum": 0.429237483926338, "maximum": 1.620584849187317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.051104076139408, "standard_deviation": 0.007050570437240993, "minimum": 0.04065408338509372, "maximum": 0.06291876689399727, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.09408651532402093, "standard_deviation": 0.0349358383903197, "minimum": 0.03159753355814606, "maximum": 0.13952814874053085, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.250731182051092, "standard_deviation": 0.006931320658394018, "minimum": 5.242958013762809, "maximum": 5.26400896520242, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46085233605747805, "standard_deviation": 0.012561818264705676, "minimum": 0.44284355178626666, "maximum": 0.48308416791742187, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08365043742685763, "standard_deviation": 0.03449680276133627, "minimum": -0.11698320928830763, "maximum": -0.031053503795889222, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9966135180612279, "standard_deviation": 0.00037665558953488856, "minimum": 0.9961681810267111, "maximum": 0.9973259156384922, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.551189840341745, "standard_deviation": 0.01183034414941366, "minimum": 0.5415813466375289, "maximum": 0.5752209435024047, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4421843341034131, "standard_deviation": 0.1886125228557134, "minimum": 0.22244873565188028, "maximum": 0.744325336807353, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05159186518695616, "standard_deviation": 0.0031724366003053925, "minimum": 0.0452770643511135, "maximum": 0.05541303335698139, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989150043945464, "standard_deviation": 0.00022964578041689593, "minimum": 0.9985798621340575, "maximum": 0.9992317772835908, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.31279765281765287, "standard_deviation": 0.011404511395217025, "minimum": 0.29875335137477493, "maximum": 0.3299772688500289, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.48367638155405046, "standard_deviation": 0.26731384973100414, "minimum": 0.19712242734758914, "maximum": 0.8871181886785245, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028991668114627316, "standard_deviation": 0.0033565325928663907, "minimum": 0.023333397574600998, "maximum": 0.03372313450201533, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0014", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.7508528232574463, "predicted_mos": 2.7508528232574463, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "08985b7bbfd4647e234ebb4ed00cec8cc8cac6c102bda33bbd3c30bd9dd174cd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.937279200181365, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.573087409729573, "全帯域_10ms_rms_envelope_nrmse": 0.05239289356608381, "全帯域_10ms_envelope_correlation": 0.9964759746087284, "全帯域_10ms_modulation_mae_db": 0.1374995473491523, "全帯域_40ms_log_rms_mae_db": 0.34464249313080986, "全帯域_40ms_rms_envelope_nrmse": 0.028723014703175134, "全帯域_40ms_envelope_correlation": 0.9989058689289332, "全帯域_40ms_modulation_mae_db": 0.265803477136153, "300–1000 Hz_10ms_log_rms_mae_db": 1.3353164121739052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08177211484583857, "300–1000 Hz_10ms_envelope_correlation": 0.9941803896894535, "300–1000 Hz_10ms_modulation_mae_db": 0.6023215865705995, "300–1000 Hz_40ms_log_rms_mae_db": 0.8730399940796432, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04496792237491614, "300–1000 Hz_40ms_envelope_correlation": 0.9982406357835586, "300–1000 Hz_40ms_modulation_mae_db": 0.5227572103388778, "1000–3000 Hz_10ms_log_rms_mae_db": 1.279270744200884, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17058960626413736, "1000–3000 Hz_10ms_envelope_correlation": 0.9786114504302535, "1000–3000 Hz_10ms_modulation_mae_db": 0.5096256900492936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9475905659505267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14568501484176344, "1000–3000 Hz_40ms_envelope_correlation": 0.9837581073848588, "1000–3000 Hz_40ms_modulation_mae_db": 0.7763938387515036, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3219374375316102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1792104522317132, "3000–6000 Hz_10ms_envelope_correlation": 0.9757170198045958, "3000–6000 Hz_10ms_modulation_mae_db": 0.8815905698183025, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8393269272475108, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11265267423865967, "3000–6000 Hz_40ms_envelope_correlation": 0.9907003651990517, "3000–6000 Hz_40ms_modulation_mae_db": 1.1510522867252464, "mr_stft_db": 4.785104178150525, "band_mae_db": 0.11422829130297633, "rms_gain_error_db": -0.07549500039651456, "output_peak": 0.44757789429385497}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5425915541914295, "全帯域_10ms_rms_envelope_nrmse": 0.05225535909669302, "全帯域_10ms_envelope_correlation": 0.9965373629173713, "全帯域_10ms_modulation_mae_db": 0.10612952417167555, "全帯域_40ms_log_rms_mae_db": 0.29609985466196725, "全帯域_40ms_rms_envelope_nrmse": 0.028605552929104928, "全帯域_40ms_envelope_correlation": 0.998970724232138, "全帯域_40ms_modulation_mae_db": 0.2244714801177646, "300–1000 Hz_10ms_log_rms_mae_db": 1.274216314928508, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08899488331368505, "300–1000 Hz_10ms_envelope_correlation": 0.9931214161595759, "300–1000 Hz_10ms_modulation_mae_db": 0.36558127985410666, "300–1000 Hz_40ms_log_rms_mae_db": 0.85941881533464, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052578004394945525, "300–1000 Hz_40ms_envelope_correlation": 0.9976000596208001, "300–1000 Hz_40ms_modulation_mae_db": 0.4447864293875889, "1000–3000 Hz_10ms_log_rms_mae_db": 1.275578031094668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16833814338385916, "1000–3000 Hz_10ms_envelope_correlation": 0.9793019969437766, "1000–3000 Hz_10ms_modulation_mae_db": 0.6276960783016287, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9119077537595137, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14670282516156571, "1000–3000 Hz_40ms_envelope_correlation": 0.9837004315871078, "1000–3000 Hz_40ms_modulation_mae_db": 0.7509898844975503, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2705832229593212, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17230356708825265, "3000–6000 Hz_10ms_envelope_correlation": 0.9770246076630529, "3000–6000 Hz_10ms_modulation_mae_db": 0.43041729088912123, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7506585531647993, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10258304739857589, "3000–6000 Hz_40ms_envelope_correlation": 0.9911539060776229, "3000–6000 Hz_40ms_modulation_mae_db": 0.7771662124884139, "mr_stft_db": 4.788164750223956, "band_mae_db": 0.1314660685446206, "rms_gain_error_db": -0.09184857952981657, "output_peak": 0.48442872097910217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.56700422102902, "全帯域_10ms_rms_envelope_nrmse": 0.05561917770417348, "全帯域_10ms_envelope_correlation": 0.996088295062442, "全帯域_10ms_modulation_mae_db": 0.1905175510649896, "全帯域_40ms_log_rms_mae_db": 0.3034710768896127, "全帯域_40ms_rms_envelope_nrmse": 0.026366821913694743, "全帯域_40ms_envelope_correlation": 0.9991950827600387, "全帯域_40ms_modulation_mae_db": 0.16341958943166293, "300–1000 Hz_10ms_log_rms_mae_db": 1.3860705282041743, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09755236194804355, "300–1000 Hz_10ms_envelope_correlation": 0.9918598504343916, "300–1000 Hz_10ms_modulation_mae_db": 0.22737314148575116, "300–1000 Hz_40ms_log_rms_mae_db": 0.9017139117155462, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05325580679260152, "300–1000 Hz_40ms_envelope_correlation": 0.9977928681516706, "300–1000 Hz_40ms_modulation_mae_db": 0.5830491015646478, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3030089656095785, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16802909349713813, "1000–3000 Hz_10ms_envelope_correlation": 0.9792770660537653, "1000–3000 Hz_10ms_modulation_mae_db": 0.8946308584047027, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9673095003169, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14585875897557085, "1000–3000 Hz_40ms_envelope_correlation": 0.983675567066564, "1000–3000 Hz_40ms_modulation_mae_db": 0.9742914436990381, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2928190294900284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15562268483650932, "3000–6000 Hz_10ms_envelope_correlation": 0.9812649154526292, "3000–6000 Hz_10ms_modulation_mae_db": 0.32135307811463415, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7728175673019951, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0769500488904807, "3000–6000 Hz_40ms_envelope_correlation": 0.9953267229876723, "3000–6000 Hz_40ms_modulation_mae_db": 0.6976512699800003, "mr_stft_db": 4.772564304869204, "band_mae_db": 0.12118168219243401, "rms_gain_error_db": -0.10256076621982713, "output_peak": 0.4554029894446368}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5633077330717252, "全帯域_10ms_rms_envelope_nrmse": 0.05086941926368248, "全帯域_10ms_envelope_correlation": 0.9967297109439225, "全帯域_10ms_modulation_mae_db": 0.1754319466208869, "全帯域_40ms_log_rms_mae_db": 0.3141466154425696, "全帯域_40ms_rms_envelope_nrmse": 0.030918374994564596, "全帯域_40ms_envelope_correlation": 0.998771731206304, "全帯域_40ms_modulation_mae_db": 0.2861347681493093, "300–1000 Hz_10ms_log_rms_mae_db": 1.3409113886418091, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08361278631272917, "300–1000 Hz_10ms_envelope_correlation": 0.9941956815377286, "300–1000 Hz_10ms_modulation_mae_db": 0.5960248372658346, "300–1000 Hz_40ms_log_rms_mae_db": 0.8983028486830505, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05400937564328293, "300–1000 Hz_40ms_envelope_correlation": 0.9977002892962192, "300–1000 Hz_40ms_modulation_mae_db": 0.7310242060710164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2891453916507556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16514740409038067, "1000–3000 Hz_10ms_envelope_correlation": 0.9799861479968294, "1000–3000 Hz_10ms_modulation_mae_db": 1.2139090116963143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9403175511228801, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14303640385410948, "1000–3000 Hz_40ms_envelope_correlation": 0.9843343001844552, "1000–3000 Hz_40ms_modulation_mae_db": 0.9111958408928857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3081775637892352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1586801718275004, "3000–6000 Hz_10ms_envelope_correlation": 0.9805901008430251, "3000–6000 Hz_10ms_modulation_mae_db": 0.6442153688021243, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8172487981839422, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09432356047386861, "3000–6000 Hz_40ms_envelope_correlation": 0.9929035886579275, "3000–6000 Hz_40ms_modulation_mae_db": 1.2076209425286923, "mr_stft_db": 4.79011491451648, "band_mae_db": 0.13891356148389075, "rms_gain_error_db": -0.09392565531558461, "output_peak": 0.45299870499838935}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5860879310484627, "全帯域_10ms_rms_envelope_nrmse": 0.05082481929661435, "全帯域_10ms_envelope_correlation": 0.9966332792718721, "全帯域_10ms_modulation_mae_db": 0.2594670225247264, "全帯域_40ms_log_rms_mae_db": 0.3341245062897057, "全帯域_40ms_rms_envelope_nrmse": 0.02630485304468513, "全帯域_40ms_envelope_correlation": 0.9990278628278266, "全帯域_40ms_modulation_mae_db": 0.14313456440570657, "300–1000 Hz_10ms_log_rms_mae_db": 1.4016926039584974, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07518851740740072, "300–1000 Hz_10ms_envelope_correlation": 0.9949941361023875, "300–1000 Hz_10ms_modulation_mae_db": 0.49576620293288753, "300–1000 Hz_40ms_log_rms_mae_db": 0.9148455686219092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04109628710336454, "300–1000 Hz_40ms_envelope_correlation": 0.9984011932440858, "300–1000 Hz_40ms_modulation_mae_db": 0.5659912948108518, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3194381135520772, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17268312223777652, "1000–3000 Hz_10ms_envelope_correlation": 0.9784260581143012, "1000–3000 Hz_10ms_modulation_mae_db": 0.7640424109492182, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9922338032930755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15186324123047215, "1000–3000 Hz_40ms_envelope_correlation": 0.9827776894780186, "1000–3000 Hz_40ms_modulation_mae_db": 1.0057029803657478, "3000–6000 Hz_10ms_log_rms_mae_db": 1.280702593691302, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17511289403400213, "3000–6000 Hz_10ms_envelope_correlation": 0.9765361301876458, "3000–6000 Hz_10ms_modulation_mae_db": 0.5859122127211952, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7698888410042805, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10329800672451099, "3000–6000 Hz_40ms_envelope_correlation": 0.9912103133678808, "3000–6000 Hz_40ms_modulation_mae_db": 0.28564732064813103, "mr_stft_db": 4.7866581139125985, "band_mae_db": 0.10513063013582612, "rms_gain_error_db": -0.04284919307478996, "output_peak": 0.43744292404274127}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5358972156971997, "全帯域_10ms_rms_envelope_nrmse": 0.052132968531669426, "全帯域_10ms_envelope_correlation": 0.9965187528414665, "全帯域_10ms_modulation_mae_db": 0.3978629842835608, "全帯域_40ms_log_rms_mae_db": 0.2853931843926983, "全帯域_40ms_rms_envelope_nrmse": 0.024862749741836332, "全帯域_40ms_envelope_correlation": 0.9992102950237497, "全帯域_40ms_modulation_mae_db": 0.6018638347804698, "300–1000 Hz_10ms_log_rms_mae_db": 1.391683681871118, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0847739071449723, "300–1000 Hz_10ms_envelope_correlation": 0.9936744269907299, "300–1000 Hz_10ms_modulation_mae_db": 0.7678684842794729, "300–1000 Hz_40ms_log_rms_mae_db": 0.9538452113494149, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04243391955720351, "300–1000 Hz_40ms_envelope_correlation": 0.9983816605449493, "300–1000 Hz_40ms_modulation_mae_db": 0.9024526548867894, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2338479971295082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16387740952555083, "1000–3000 Hz_10ms_envelope_correlation": 0.9803288090534925, "1000–3000 Hz_10ms_modulation_mae_db": 0.4314461783926779, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8979057697526871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14162009616048316, "1000–3000 Hz_40ms_envelope_correlation": 0.9846018265728981, "1000–3000 Hz_40ms_modulation_mae_db": 0.19894191540152986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3155768475326535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16364682521025944, "3000–6000 Hz_10ms_envelope_correlation": 0.9797791674000031, "3000–6000 Hz_10ms_modulation_mae_db": 1.1468875968363272, "3000–6000 Hz_40ms_log_rms_mae_db": 0.821441243988395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09824486114147357, "3000–6000 Hz_40ms_envelope_correlation": 0.9929748781909774, "3000–6000 Hz_40ms_modulation_mae_db": 1.1683234294298794, "mr_stft_db": 4.7658735988770795, "band_mae_db": 0.13248193078086803, "rms_gain_error_db": -0.07651748844533106, "output_peak": 0.45536654044455316}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5472520918015126, "全帯域_10ms_rms_envelope_nrmse": 0.053111620558618226, "全帯域_10ms_envelope_correlation": 0.9965946069321103, "全帯域_10ms_modulation_mae_db": 0.5711027824491806, "全帯域_40ms_log_rms_mae_db": 0.31598805291722243, "全帯域_40ms_rms_envelope_nrmse": 0.0332685844994078, "全帯域_40ms_envelope_correlation": 0.9987485628717839, "全帯域_40ms_modulation_mae_db": 0.6286888257904814, "300–1000 Hz_10ms_log_rms_mae_db": 1.3284411429262721, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09698936888400185, "300–1000 Hz_10ms_envelope_correlation": 0.9921304996777464, "300–1000 Hz_10ms_modulation_mae_db": 0.7416951564909632, "300–1000 Hz_40ms_log_rms_mae_db": 0.8893164152297937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06572231360584892, "300–1000 Hz_40ms_envelope_correlation": 0.9964915366148934, "300–1000 Hz_40ms_modulation_mae_db": 0.9390517654761429, "1000–3000 Hz_10ms_log_rms_mae_db": 1.242485398434178, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16813115466259732, "1000–3000 Hz_10ms_envelope_correlation": 0.9794769286123908, "1000–3000 Hz_10ms_modulation_mae_db": 0.29657229977443506, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8975441985624717, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1456327944098686, "1000–3000 Hz_40ms_envelope_correlation": 0.984107407778021, "1000–3000 Hz_40ms_modulation_mae_db": 0.34877747056952857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3336076179750433, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16332961265200244, "3000–6000 Hz_10ms_envelope_correlation": 0.9796835988861875, "3000–6000 Hz_10ms_modulation_mae_db": 0.6318940378543058, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8526906072201177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09892845487296871, "3000–6000 Hz_40ms_envelope_correlation": 0.9925673806682247, "3000–6000 Hz_40ms_modulation_mae_db": 1.2182621212171472, "mr_stft_db": 4.799935561611694, "band_mae_db": 0.16692103753689883, "rms_gain_error_db": -0.13249440262947706, "output_peak": 0.4435649418476785}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5411593147372423, "全帯域_10ms_rms_envelope_nrmse": 0.05330855003313025, "全帯域_10ms_envelope_correlation": 0.9962895972041369, "全帯域_10ms_modulation_mae_db": 0.2421507721862862, "全帯域_40ms_log_rms_mae_db": 0.30238397926528165, "全帯域_40ms_rms_envelope_nrmse": 0.02852954566070767, "全帯域_40ms_envelope_correlation": 0.9988482755387121, "全帯域_40ms_modulation_mae_db": 0.27430315835291946, "300–1000 Hz_10ms_log_rms_mae_db": 1.3606306391560339, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08433858353813153, "300–1000 Hz_10ms_envelope_correlation": 0.9937304624217276, "300–1000 Hz_10ms_modulation_mae_db": 0.4387587750166977, "300–1000 Hz_40ms_log_rms_mae_db": 0.918308130301078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04857846022624798, "300–1000 Hz_40ms_envelope_correlation": 0.9978021836229145, "300–1000 Hz_40ms_modulation_mae_db": 0.44719322448624554, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2282267271500074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16493945298472706, "1000–3000 Hz_10ms_envelope_correlation": 0.9800513807273148, "1000–3000 Hz_10ms_modulation_mae_db": 0.4639484595651504, "1000–3000 Hz_40ms_log_rms_mae_db": 0.893804983106824, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14058479254332976, "1000–3000 Hz_40ms_envelope_correlation": 0.9848156639660708, "1000–3000 Hz_40ms_modulation_mae_db": 0.3554346211963406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2860894184839742, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17011144946725784, "3000–6000 Hz_10ms_envelope_correlation": 0.9780036597400188, "3000–6000 Hz_10ms_modulation_mae_db": 0.4015526211018175, "3000–6000 Hz_40ms_log_rms_mae_db": 0.849826069092703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10631398824812795, "3000–6000 Hz_40ms_envelope_correlation": 0.9914510298567658, "3000–6000 Hz_40ms_modulation_mae_db": 0.5944984612123675, "mr_stft_db": 4.778234973197567, "band_mae_db": 0.09132446119242482, "rms_gain_error_db": -0.03894721701165233, "output_peak": 0.4489782281285937}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.757109826764722, "spectral_convergence": 0.15743823255919848, "spectral_convergence_status": "ok", "amplitude_floor": 1.7246496227499867e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.623157380663508, "spectral_convergence": 0.1674219712109507, "spectral_convergence_status": "ok", "amplitude_floor": 1.4526318235172953e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.975045327023345, "spectral_convergence": 0.1684150787125968, "spectral_convergence_status": "ok", "amplitude_floor": 1.3962213853324247e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0014645134106370877, "output_power": 0.001470041358499615, "delta_db": 0.016362005697671725, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024771538682985465, "output_power": 0.0024201637745614618, "delta_db": -0.10108227549941123, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013427666217707404, "output_power": 0.0013100340479117911, "delta_db": -0.10717953945023409, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024269361162520585, "output_power": 0.00023275104680814358, "delta_db": -0.18166701720243555, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.4425202041437504e-05, "output_power": 4.3118391371520407e-05, "delta_db": -0.12966861460927281, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.4446805694883255e-05, "output_power": 3.328188577773641e-05, "delta_db": -0.14941029535883252, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11422829130297633, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07549500039651456, "pitch": null, "output_rms_dbfs": -22.58896443795525, "output_peak": 0.44757789429385497, "reference_rms_dbfs": -22.513469437558737, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.785104178150525}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002110151907366841, "log_rms_mae_db": 0.573087409729573, "rms_envelope_nrmse": 0.05239289356608381, "envelope_correlation": 0.9964759746087284, "rms_gain_error_db": -0.0748329208390206, "reference_mean_power": 0.005604891233518836, "output_mean_power": 0.005509141119175528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2250047113785757e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011643562088722453, "output_power": 0.0011646619515688608, "reference_active": true, "delta_db": 0.0011402432925633145}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011188245530999946, "output_power": 0.00010555156599983793, "reference_active": true, "delta_db": -0.2529730755284699}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8450792038001933e-05, "output_power": 2.7431896204414603e-05, "reference_active": true, "delta_db": -0.15838532322642368}], "active_band_count": 3, "active_band_mae_db": 0.1374995473491523}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020713167740975364, "log_rms_mae_db": 0.34464249313080986, "rms_envelope_nrmse": 0.028723014703175134, "envelope_correlation": 0.9989058689289332, "rms_gain_error_db": -0.07297350804200047, "reference_mean_power": 0.005603505754865554, "output_mean_power": 0.00551013794555298}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.048317169639576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010458478077031374, "output_power": 0.001051285253937028, "reference_active": true, "delta_db": 0.022520823594363845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.131979407845485e-05, "output_power": 5.689291497035419e-05, "reference_active": true, "delta_db": -0.32542501867591334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6292932861003563e-06, "output_power": 1.4691049058028163e-06, "reference_active": true, "delta_db": -0.44946458913818177}], "active_band_count": 3, "active_band_mae_db": 0.265803477136153}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015831513319989338, "log_rms_mae_db": 1.3353164121739052, "rms_envelope_nrmse": 0.08177211484583857, "envelope_correlation": 0.9941803896894535, "rms_gain_error_db": -0.12202630150780044, "reference_mean_power": 0.0022309508329157127, "output_mean_power": 0.0021691389309490346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0339679622838414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046453953265101344, "output_power": 0.0004414022019629159, "reference_active": true, "delta_db": -0.22188183340983522}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.019812862224435e-05, "output_power": 5.74554871682061e-05, "reference_active": true, "delta_db": -0.8699402323201892}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.545275737486218e-05, "output_power": 1.310665042779368e-05, "reference_active": true, "delta_db": -0.7151426939817741}], "active_band_count": 3, "active_band_mae_db": 0.6023215865705995}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014375880677218868, "log_rms_mae_db": 0.8730399940796432, "rms_envelope_nrmse": 0.04496792237491614, "envelope_correlation": 0.9982406357835586, "rms_gain_error_db": -0.11792570680354385, "reference_mean_power": 0.0022214967634666704, "output_mean_power": 0.002161987178957702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.581275628581243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00040407180276463134, "output_power": 0.000386022889898409, "reference_active": true, "delta_db": -0.19845487561779102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0351036816934105e-05, "output_power": 3.2521001025062145e-05, "reference_active": true, "delta_db": -0.936907932718116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1841518228550588e-06, "output_power": 1.0718067938962371e-06, "reference_active": true, "delta_db": -0.43290882268072617}], "active_band_count": 3, "active_band_mae_db": 0.5227572103388778}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.952734804355257e-05, "log_rms_mae_db": 1.279270744200884, "rms_envelope_nrmse": 0.17058960626413736, "envelope_correlation": 0.9786114504302535, "rms_gain_error_db": -0.17306577758192065, "reference_mean_power": 0.00025078805819662737, "output_mean_power": 0.00024099069549057006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4212041991971396e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.171024783262431e-05, "output_power": 6.559805178573411e-05, "reference_active": true, "delta_db": -0.38690282100183276}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.275733290040072e-06, "output_power": 5.84117309093102e-06, "reference_active": true, "delta_db": -0.9537669470940312}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0724449404458702e-06, "output_power": 1.0269617646456918e-06, "reference_active": true, "delta_db": -0.1882073020520167}], "active_band_count": 3, "active_band_mae_db": 0.5096256900492936}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.039214481073785e-05, "log_rms_mae_db": 0.9475905659505267, "rms_envelope_nrmse": 0.14568501484176344, "envelope_correlation": 0.9837581073848588, "rms_gain_error_db": -0.17199965959976043, "reference_mean_power": 0.0002518169764916092, "output_mean_power": 0.000242038826811383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3611130645010071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.488974762253847e-05, "output_power": 5.973224479220279e-05, "reference_active": true, "delta_db": -0.35967248812456337}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.711335389958727e-06, "output_power": 2.9479979409232084e-06, "reference_active": true, "delta_db": -1.0000302706718838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.486818337389537e-08, "output_power": 6.788862776284554e-08, "reference_active": true, "delta_db": -0.9694787574580637}], "active_band_count": 3, "active_band_mae_db": 0.7763938387515036}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.8930089256612995e-05, "log_rms_mae_db": 1.3219374375316102, "rms_envelope_nrmse": 0.1792104522317132, "envelope_correlation": 0.9757170198045958, "rms_gain_error_db": -0.4176402176632281, "reference_mean_power": 2.8173794881918634e-05, "output_mean_power": 2.559065056477399e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9380195589864167e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.123435668533653e-06, "output_power": 5.477153816027029e-06, "reference_active": true, "delta_db": -0.48440222120029697}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.722245443537651e-06, "output_power": 2.207805892383733e-06, "reference_active": true, "delta_db": -0.9096639147112058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.909737577930936e-07, "output_power": 5.180730445512713e-07, "reference_active": true, "delta_db": -1.2507055735434052}], "active_band_count": 3, "active_band_mae_db": 0.8815905698183025}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3848385775447824e-05, "log_rms_mae_db": 0.8393269272475108, "rms_envelope_nrmse": 0.11265267423865967, "envelope_correlation": 0.9907003651990517, "rms_gain_error_db": -0.4172666104208502, "reference_mean_power": 2.8310170166312676e-05, "output_mean_power": 2.5716734355303715e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.80878248957878e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.951898475340433e-06, "output_power": 5.3319239815613e-06, "reference_active": true, "delta_db": -0.4777156565992339}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2501096733423988e-06, "output_power": 9.529796054026863e-07, "reference_active": true, "delta_db": -1.178645093025684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.567916579161296e-08, "output_power": 4.342581891219516e-08, "reference_active": true, "delta_db": -1.7967961105508212}], "active_band_count": 3, "active_band_mae_db": 1.1510522867252464}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9495286982062673, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3a862fc2b3c7417facff6160ba855d965f15327e5925cbc1267ff0e46d90e1ce", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9794324797415155, "standard_deviation": 0.0006785616827756036, "minimum": 0.9784260581143012, "maximum": 0.9803288090534925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.271375171102707, "standard_deviation": 0.033404927360185084, "minimum": 1.2282267271500074, "maximum": 1.3194381135520772, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6502338733916776, "standard_deviation": 0.2972102103654874, "minimum": 0.29657229977443506, "maximum": 1.2139090116963143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16771692333077087, "standard_deviation": 0.002991154289695775, "minimum": 0.16387740952555083, "maximum": 0.17268312223777652, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9839713742522493, "standard_deviation": 0.0006427389514226724, "minimum": 0.9827776894780186, "maximum": 0.9848156639660708, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9310767657331098, "standard_deviation": 0.036630348164061494, "minimum": 0.893804983106824, "maximum": 0.9922338032930755, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6652159994217656, "standard_deviation": 0.3173983829383245, "minimum": 0.19894191540152986, "maximum": 1.0057029803657478, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1451229908971454, "standard_deviation": 0.003510460640977459, "minimum": 0.14058479254332976, "maximum": 0.15186324123047215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9785748999971448, "standard_deviation": 0.0020364701284234484, "minimum": 0.9757170198045958, "maximum": 0.9812649154526292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.301186716431646, "standard_deviation": 0.022012252400367062, "minimum": 1.2705832229593212, "maximum": 1.3336076179750433, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6304778470172284, "standard_deviation": 0.27252851962162217, "minimum": 0.32135307811463415, "maximum": 1.1468875968363272, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16725220716843717, "standard_deviation": 0.008242681887715474, "minimum": 0.15562268483650932, "maximum": 0.1792104522317132, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922860231257653, "standard_deviation": 0.001503866042572095, "minimum": 0.9907003651990517, "maximum": 0.9953267229876723, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8092373259004679, "standard_deviation": 0.039569670637737434, "minimum": 0.7506585531647993, "maximum": 0.8526906072201177, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8875277555287348, "standard_deviation": 0.3497948854484455, "minimum": 0.28564732064813103, "maximum": 1.2182621212171472, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09916183024858327, "standard_deviation": 0.010553511941968215, "minimum": 0.0769500488904807, "maximum": 0.11265267423865967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9934858578767176, "standard_deviation": 0.0010670540744145954, "minimum": 0.9918598504343916, "maximum": 0.9949941361023875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3523703389825397, "standard_deviation": 0.04189265521409072, "minimum": 1.274216314928508, "maximum": 1.4016926039584974, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5294236829870391, "standard_deviation": 0.1848152490660888, "minimum": 0.22737314148575116, "maximum": 0.7678684842794729, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08665281542435035, "standard_deviation": 0.007601774638184077, "minimum": 0.07518851740740072, "maximum": 0.09755236194804355, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978013033598865, "standard_deviation": 0.0006159729877401499, "minimum": 0.9964915366148934, "maximum": 0.9984011932440858, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9010988619143845, "standard_deviation": 0.029127445613104118, "minimum": 0.85941881533464, "maximum": 0.9538452113494149, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6420382358777701, "standard_deviation": 0.19431744322024863, "minimum": 0.4447864293875889, "maximum": 0.9390517654761429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.050330261212301386, "standard_deviation": 0.007959758611915346, "minimum": 0.04109628710336454, "maximum": 0.06572231360584892, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12520595789624245, "standard_deviation": 0.023015068339311585, "minimum": 0.09132446119242482, "maximum": 0.16692103753689883, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.783331299419888, "standard_deviation": 0.010730224454528273, "minimum": 4.7658735988770795, "maximum": 4.799935561611694, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4532201180224438, "standard_deviation": 0.0140171760677873, "minimum": 0.43744292404274127, "maximum": 0.48442872097910217, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08182978782787416, "standard_deviation": 0.030868649281320815, "minimum": -0.13249440262947706, "maximum": -0.03894721701165233, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964834474727562, "standard_deviation": 0.00020472428588105197, "minimum": 0.996088295062442, "maximum": 0.9967297109439225, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5570484339132706, "standard_deviation": 0.017903094275356732, "minimum": 0.5358972156971997, "maximum": 0.5860879310484627, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2600202663313073, "standard_deviation": 0.15435213450769428, "minimum": 0.10612952417167555, "maximum": 0.5711027824491806, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05256435100633313, "standard_deviation": 0.0015304075546370502, "minimum": 0.05082481929661435, "maximum": 0.05561917770417348, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989598004236857, "standard_deviation": 0.00017655225161977505, "minimum": 0.9987485628717839, "maximum": 0.9992102950237497, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.31203122037373343, "standard_deviation": 0.01964498665715324, "minimum": 0.2853931843926983, "maximum": 0.34464249313080986, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3234774622705584, "standard_deviation": 0.18736370067799096, "minimum": 0.14313456440570657, "maximum": 0.6286888257904814, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02844743718589704, "standard_deviation": 0.0027051879051579655, "minimum": 0.024862749741836332, "maximum": 0.0332685844994078, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0014", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.58378267288208, "predicted_mos": 2.58378267288208, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7ea9253c7a0376695cbb476b1037f65d2a9ad594f61c937e626171a77387599e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.199674000032246, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5358404758164731, "全帯域_10ms_rms_envelope_nrmse": 0.048205643378329305, "全帯域_10ms_envelope_correlation": 0.9970052163014651, "全帯域_10ms_modulation_mae_db": 0.6203499647250904, "全帯域_40ms_log_rms_mae_db": 0.2932344136008289, "全帯域_40ms_rms_envelope_nrmse": 0.027525941930530914, "全帯域_40ms_envelope_correlation": 0.9989732267448377, "全帯域_40ms_modulation_mae_db": 0.5050864987845833, "300–1000 Hz_10ms_log_rms_mae_db": 1.3491199451749, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0752652068879021, "300–1000 Hz_10ms_envelope_correlation": 0.995055787997552, "300–1000 Hz_10ms_modulation_mae_db": 0.832066811001183, "300–1000 Hz_40ms_log_rms_mae_db": 0.9170701463296719, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043559067429811235, "300–1000 Hz_40ms_envelope_correlation": 0.9982969031359091, "300–1000 Hz_40ms_modulation_mae_db": 0.8667660578353277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4167009163763327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16917376779099047, "1000–3000 Hz_10ms_envelope_correlation": 0.9789655013933485, "1000–3000 Hz_10ms_modulation_mae_db": 0.33876606878585697, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9940548983635729, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14287846404337015, "1000–3000 Hz_40ms_envelope_correlation": 0.9843980893068325, "1000–3000 Hz_40ms_modulation_mae_db": 0.29684117682559547, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5106610590146028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1824401238169318, "3000–6000 Hz_10ms_envelope_correlation": 0.9745197695188461, "3000–6000 Hz_10ms_modulation_mae_db": 0.5446371462283137, "3000–6000 Hz_40ms_log_rms_mae_db": 0.934081384261376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10833450755966331, "3000–6000 Hz_40ms_envelope_correlation": 0.991154916243526, "3000–6000 Hz_40ms_modulation_mae_db": 1.0279084875965614, "mr_stft_db": 4.914465391818724, "band_mae_db": 0.18423828443419463, "rms_gain_error_db": -0.05980750032040098, "output_peak": 0.45279706291196997}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5384878722538181, "全帯域_10ms_rms_envelope_nrmse": 0.04971956299114768, "全帯域_10ms_envelope_correlation": 0.9968006018630079, "全帯域_10ms_modulation_mae_db": 0.24645029318045375, "全帯域_40ms_log_rms_mae_db": 0.30931522154159485, "全帯域_40ms_rms_envelope_nrmse": 0.026843747508914013, "全帯域_40ms_envelope_correlation": 0.9990157148431734, "全帯域_40ms_modulation_mae_db": 0.26631961151973543, "300–1000 Hz_10ms_log_rms_mae_db": 1.4254992722874815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08119618474932237, "300–1000 Hz_10ms_envelope_correlation": 0.9941914273738806, "300–1000 Hz_10ms_modulation_mae_db": 0.6606886985636478, "300–1000 Hz_40ms_log_rms_mae_db": 0.9267182516389973, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04790880593157512, "300–1000 Hz_40ms_envelope_correlation": 0.9978622196549631, "300–1000 Hz_40ms_modulation_mae_db": 0.8192568757474171, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3819881362441382, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16262888807079, "1000–3000 Hz_10ms_envelope_correlation": 0.9805668255852832, "1000–3000 Hz_10ms_modulation_mae_db": 0.17754291330990288, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9594734326645912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1400523023159557, "1000–3000 Hz_40ms_envelope_correlation": 0.9849371865838668, "1000–3000 Hz_40ms_modulation_mae_db": 0.18555958568987507, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5254006048262203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18894116316317952, "3000–6000 Hz_10ms_envelope_correlation": 0.9721584793813852, "3000–6000 Hz_10ms_modulation_mae_db": 0.588615956345135, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9531839353458791, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10737068889140094, "3000–6000 Hz_40ms_envelope_correlation": 0.9905155482304958, "3000–6000 Hz_40ms_modulation_mae_db": 0.914160182601902, "mr_stft_db": 4.902359100251445, "band_mae_db": 0.10404321580052849, "rms_gain_error_db": -0.057865091727434564, "output_peak": 0.46484818935579}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5471254674211037, "全帯域_10ms_rms_envelope_nrmse": 0.0497830033936914, "全帯域_10ms_envelope_correlation": 0.9968373930387359, "全帯域_10ms_modulation_mae_db": 0.178193710836997, "全帯域_40ms_log_rms_mae_db": 0.3103576987651477, "全帯域_40ms_rms_envelope_nrmse": 0.02750629511608651, "全帯域_40ms_envelope_correlation": 0.9990212512595564, "全帯域_40ms_modulation_mae_db": 0.12242134765869798, "300–1000 Hz_10ms_log_rms_mae_db": 1.4161539114426591, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08098188900361433, "300–1000 Hz_10ms_envelope_correlation": 0.994474420501711, "300–1000 Hz_10ms_modulation_mae_db": 0.44965501541097036, "300–1000 Hz_40ms_log_rms_mae_db": 0.9253304143770663, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05310238709286587, "300–1000 Hz_40ms_envelope_correlation": 0.9976676360504579, "300–1000 Hz_40ms_modulation_mae_db": 0.5067422363081225, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3973693713679085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17219118285452362, "1000–3000 Hz_10ms_envelope_correlation": 0.9781895609883786, "1000–3000 Hz_10ms_modulation_mae_db": 0.7169315420005821, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0187219286392735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14599732090002365, "1000–3000 Hz_40ms_envelope_correlation": 0.9836657424422122, "1000–3000 Hz_40ms_modulation_mae_db": 0.5818057630583381, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4841284924005194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17026794025353356, "3000–6000 Hz_10ms_envelope_correlation": 0.9775219973703131, "3000–6000 Hz_10ms_modulation_mae_db": 0.7904966226361979, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8604794869088924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08253652527982887, "3000–6000 Hz_40ms_envelope_correlation": 0.9946706263191897, "3000–6000 Hz_40ms_modulation_mae_db": 1.2723532270928855, "mr_stft_db": 4.902254580262499, "band_mae_db": 0.15514473251410385, "rms_gain_error_db": -0.07661969874583427, "output_peak": 0.4394448033637356}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5279833919282145, "全帯域_10ms_rms_envelope_nrmse": 0.05598326573978251, "全帯域_10ms_envelope_correlation": 0.9961218878533952, "全帯域_10ms_modulation_mae_db": 0.5458071458222182, "全帯域_40ms_log_rms_mae_db": 0.3009151812280565, "全帯域_40ms_rms_envelope_nrmse": 0.030252650828237318, "全帯域_40ms_envelope_correlation": 0.9989608367790984, "全帯域_40ms_modulation_mae_db": 0.6766990762753561, "300–1000 Hz_10ms_log_rms_mae_db": 1.4446354893812479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09299601748658065, "300–1000 Hz_10ms_envelope_correlation": 0.9928251170650771, "300–1000 Hz_10ms_modulation_mae_db": 1.0952889814010278, "300–1000 Hz_40ms_log_rms_mae_db": 0.9742851937641945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052772080247862216, "300–1000 Hz_40ms_envelope_correlation": 0.9980045713183507, "300–1000 Hz_40ms_modulation_mae_db": 1.3565334886651392, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3936431436428205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16623028773714918, "1000–3000 Hz_10ms_envelope_correlation": 0.9796811571466053, "1000–3000 Hz_10ms_modulation_mae_db": 0.5767435033402469, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9841835548392107, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1414349569312324, "1000–3000 Hz_40ms_envelope_correlation": 0.9846504175550231, "1000–3000 Hz_40ms_modulation_mae_db": 0.8389389610944148, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5795137706862188, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22473305910726335, "3000–6000 Hz_10ms_envelope_correlation": 0.9603642848435016, "3000–6000 Hz_10ms_modulation_mae_db": 0.7835996968157822, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9747636266340386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14374989633765362, "3000–6000 Hz_40ms_envelope_correlation": 0.9828716030833176, "3000–6000 Hz_40ms_modulation_mae_db": 0.9268113794520678, "mr_stft_db": 4.916547111524219, "band_mae_db": 0.1326180894400118, "rms_gain_error_db": -0.12277527074553576, "output_peak": 0.43449938652040504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5792210402268614, "全帯域_10ms_rms_envelope_nrmse": 0.05041448690506384, "全帯域_10ms_envelope_correlation": 0.9967201675928103, "全帯域_10ms_modulation_mae_db": 0.5720355951474358, "全帯域_40ms_log_rms_mae_db": 0.31361082922856254, "全帯域_40ms_rms_envelope_nrmse": 0.024699687557990846, "全帯域_40ms_envelope_correlation": 0.9991872249338709, "全帯域_40ms_modulation_mae_db": 0.6552191717945686, "300–1000 Hz_10ms_log_rms_mae_db": 1.4435114146857744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07754177604391141, "300–1000 Hz_10ms_envelope_correlation": 0.9948061193924359, "300–1000 Hz_10ms_modulation_mae_db": 0.5108484226123665, "300–1000 Hz_40ms_log_rms_mae_db": 0.9408801148780934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04383112676554896, "300–1000 Hz_40ms_envelope_correlation": 0.9983602765305225, "300–1000 Hz_40ms_modulation_mae_db": 0.5089440030043072, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4152361481680478, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17092136501359187, "1000–3000 Hz_10ms_envelope_correlation": 0.9786353829935613, "1000–3000 Hz_10ms_modulation_mae_db": 0.7346261704710463, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0089182429149848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14413985767495116, "1000–3000 Hz_40ms_envelope_correlation": 0.9840751315451429, "1000–3000 Hz_40ms_modulation_mae_db": 0.43690478541528205, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5198576508020007, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18358712458526522, "3000–6000 Hz_10ms_envelope_correlation": 0.974634586549873, "3000–6000 Hz_10ms_modulation_mae_db": 0.3601007330669104, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9163758046786131, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11104833421580283, "3000–6000 Hz_40ms_envelope_correlation": 0.9900511900236064, "3000–6000 Hz_40ms_modulation_mae_db": 0.4045378159718284, "mr_stft_db": 4.91315064943844, "band_mae_db": 0.0594962781617925, "rms_gain_error_db": -0.05271435912618837, "output_peak": 0.4553019486565919}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5661398492914722, "全帯域_10ms_rms_envelope_nrmse": 0.05664737979384719, "全帯域_10ms_envelope_correlation": 0.9959628027056594, "全帯域_10ms_modulation_mae_db": 0.1832895687549538, "全帯域_40ms_log_rms_mae_db": 0.3144051260575939, "全帯域_40ms_rms_envelope_nrmse": 0.03331068098908808, "全帯域_40ms_envelope_correlation": 0.9986181234164565, "全帯域_40ms_modulation_mae_db": 0.20019482824542958, "300–1000 Hz_10ms_log_rms_mae_db": 1.4455524327052396, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09752312012425259, "300–1000 Hz_10ms_envelope_correlation": 0.9920888472673731, "300–1000 Hz_10ms_modulation_mae_db": 0.31317053937898515, "300–1000 Hz_40ms_log_rms_mae_db": 0.9931010746814949, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.067257941576125, "300–1000 Hz_40ms_envelope_correlation": 0.996393496162322, "300–1000 Hz_40ms_modulation_mae_db": 0.30629874645042504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4119861931770572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17605591124608497, "1000–3000 Hz_10ms_envelope_correlation": 0.977215316147481, "1000–3000 Hz_10ms_modulation_mae_db": 0.6371955894468891, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0109279715792836, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15164340924016284, "1000–3000 Hz_40ms_envelope_correlation": 0.9823636496943801, "1000–3000 Hz_40ms_modulation_mae_db": 0.5020639360351121, "3000–6000 Hz_10ms_log_rms_mae_db": 1.51996583577769, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18576003312850853, "3000–6000 Hz_10ms_envelope_correlation": 0.9730508170439753, "3000–6000 Hz_10ms_modulation_mae_db": 0.8078392140370685, "3000–6000 Hz_40ms_log_rms_mae_db": 0.913123908582749, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10954896490326385, "3000–6000 Hz_40ms_envelope_correlation": 0.9899293706059962, "3000–6000 Hz_40ms_modulation_mae_db": 0.8386878951574648, "mr_stft_db": 4.915765702342355, "band_mae_db": 0.1345584450663741, "rms_gain_error_db": -0.1129674015346989, "output_peak": 0.4442249275464226}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5576996960530676, "全帯域_10ms_rms_envelope_nrmse": 0.04688962785617452, "全帯域_10ms_envelope_correlation": 0.9972521017539561, "全帯域_10ms_modulation_mae_db": 0.3729931004274431, "全帯域_40ms_log_rms_mae_db": 0.2973595758093256, "全帯域_40ms_rms_envelope_nrmse": 0.024550430214194154, "全帯域_40ms_envelope_correlation": 0.9992933217028735, "全帯域_40ms_modulation_mae_db": 0.6024121982074779, "300–1000 Hz_10ms_log_rms_mae_db": 1.4364562830995016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07855993555995723, "300–1000 Hz_10ms_envelope_correlation": 0.9947476407414948, "300–1000 Hz_10ms_modulation_mae_db": 0.7145346715393989, "300–1000 Hz_40ms_log_rms_mae_db": 0.9575094297990681, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04561012141962885, "300–1000 Hz_40ms_envelope_correlation": 0.9982811271691353, "300–1000 Hz_40ms_modulation_mae_db": 1.0570947747448665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3788158996603659, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17049769643961443, "1000–3000 Hz_10ms_envelope_correlation": 0.9786267422822514, "1000–3000 Hz_10ms_modulation_mae_db": 0.31477954907459593, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9897193762576291, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14374126753487937, "1000–3000 Hz_40ms_envelope_correlation": 0.9841600878263751, "1000–3000 Hz_40ms_modulation_mae_db": 0.20888532066128374, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5439423328324455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19387745065790762, "3000–6000 Hz_10ms_envelope_correlation": 0.9706181231860352, "3000–6000 Hz_10ms_modulation_mae_db": 0.2790734590303144, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9687367431743853, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11269045369314648, "3000–6000 Hz_40ms_envelope_correlation": 0.9893384419583853, "3000–6000 Hz_40ms_modulation_mae_db": 0.4787889498287461, "mr_stft_db": 4.888655525868735, "band_mae_db": 0.13536570522996047, "rms_gain_error_db": -0.09614783686304683, "output_peak": 0.4509102425775546}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5611516163100964, "全帯域_10ms_rms_envelope_nrmse": 0.05596972759884839, "全帯域_10ms_envelope_correlation": 0.9959942596079732, "全帯域_10ms_modulation_mae_db": 0.48893978078852623, "全帯域_40ms_log_rms_mae_db": 0.31687143026413944, "全帯域_40ms_rms_envelope_nrmse": 0.03183540571617701, "全帯域_40ms_envelope_correlation": 0.9986669875796947, "全帯域_40ms_modulation_mae_db": 0.7331639534600005, "300–1000 Hz_10ms_log_rms_mae_db": 1.405150069793876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08284318623638728, "300–1000 Hz_10ms_envelope_correlation": 0.9939464802161507, "300–1000 Hz_10ms_modulation_mae_db": 0.6185768342446677, "300–1000 Hz_40ms_log_rms_mae_db": 0.9482409206353479, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04877383391237826, "300–1000 Hz_40ms_envelope_correlation": 0.9977750552710449, "300–1000 Hz_40ms_modulation_mae_db": 0.6888743291866956, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3991285100694117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17263641748444125, "1000–3000 Hz_10ms_envelope_correlation": 0.9780688573396303, "1000–3000 Hz_10ms_modulation_mae_db": 0.31692934865401606, "1000–3000 Hz_40ms_log_rms_mae_db": 1.023029574088301, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1474894593878068, "1000–3000 Hz_40ms_envelope_correlation": 0.9833160635948283, "1000–3000 Hz_40ms_modulation_mae_db": 0.5232216655566887, "3000–6000 Hz_10ms_log_rms_mae_db": 1.498066526872694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18276554624175584, "3000–6000 Hz_10ms_envelope_correlation": 0.974054425086162, "3000–6000 Hz_10ms_modulation_mae_db": 1.7727847155074048, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9294144432063495, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11029427361692522, "3000–6000 Hz_40ms_envelope_correlation": 0.9900885728763386, "3000–6000 Hz_40ms_modulation_mae_db": 1.6435374250365349, "mr_stft_db": 4.91271643309008, "band_mae_db": 0.11645427000549013, "rms_gain_error_db": -0.08879586706577838, "output_peak": 0.44090101134591475}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.93662345845634, "spectral_convergence": 0.15039059110573466, "spectral_convergence_status": "ok", "amplitude_floor": 1.7246496227499867e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.774055781418573, "spectral_convergence": 0.15791252767868294, "spectral_convergence_status": "ok", "amplitude_floor": 1.4526318235172953e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.032716935581259, "spectral_convergence": 0.17097687937787726, "spectral_convergence_status": "ok", "amplitude_floor": 1.3962213853324247e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0014645134106370877, "output_power": 0.0014695066479431227, "delta_db": 0.01478202233865017, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024771538682985465, "output_power": 0.0024594510645381163, "delta_db": -0.031147977397714587, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013427666217707404, "output_power": 0.001292792055741144, "delta_db": -0.16471862421358563, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024269361162520585, "output_power": 0.00023502000918905572, "delta_db": -0.1395350572468195, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.4425202041437504e-05, "output_power": 4.1766470171550517e-05, "delta_db": -0.26801638518865195, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.4446805694883255e-05, "output_power": 3.079115580579988e-05, "delta_db": -0.4872296402197459, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18423828443419463, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05980750032040098, "pitch": null, "output_rms_dbfs": -22.57327693787914, "output_peak": 0.45279706291196997, "reference_rms_dbfs": -22.513469437558737, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.914465391818724}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002110151907366841, "log_rms_mae_db": 0.5358404758164731, "rms_envelope_nrmse": 0.048205643378329305, "envelope_correlation": 0.9970052163014651, "rms_gain_error_db": -0.059133547719925766, "reference_mean_power": 0.005604891233518836, "output_mean_power": 0.005529092230972952}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2250047113785757e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011643562088722453, "output_power": 0.0011408143842445643, "reference_active": true, "delta_db": -0.08870874920488533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011188245530999946, "output_power": 0.0001055629633907676, "reference_active": true, "delta_db": -0.25250415242706703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8450792038001933e-05, "output_power": 2.004982848901766e-05, "reference_active": true, "delta_db": -1.5198369925433188}], "active_band_count": 3, "active_band_mae_db": 0.6203499647250904}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020713167740975364, "log_rms_mae_db": 0.2932344136008289, "rms_envelope_nrmse": 0.027525941930530914, "envelope_correlation": 0.9989732267448377, "rms_gain_error_db": -0.05714752773623752, "reference_mean_power": 0.005603505754865554, "output_mean_power": 0.005530253885037445}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.048317169639576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010458478077031374, "output_power": 0.0010276320663019111, "reference_active": true, "delta_db": -0.07630842817882379}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.131979407845485e-05, "output_power": 5.7741992078816296e-05, "reference_active": true, "delta_db": -0.2610892494234634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6292932861003563e-06, "output_power": 1.2422616721676584e-06, "reference_active": true, "delta_db": -1.177861818751463}], "active_band_count": 3, "active_band_mae_db": 0.5050864987845833}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015831513319989338, "log_rms_mae_db": 1.3491199451749, "rms_envelope_nrmse": 0.0752652068879021, "envelope_correlation": 0.995055787997552, "rms_gain_error_db": -0.09297997916727166, "reference_mean_power": 0.0022309508329157127, "output_mean_power": 0.0021836951082168593}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0339679622838414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046453953265101344, "output_power": 0.00043773627966564154, "reference_active": true, "delta_db": -0.2581013616402638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.019812862224435e-05, "output_power": 6.216477652847913e-05, "reference_active": true, "delta_db": -0.5278115786362242}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.545275737486218e-05, "output_power": 1.0422627978277858e-05, "reference_active": true, "delta_db": -1.710287492727061}], "active_band_count": 3, "active_band_mae_db": 0.832066811001183}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014375880677218868, "log_rms_mae_db": 0.9170701463296719, "rms_envelope_nrmse": 0.043559067429811235, "envelope_correlation": 0.9982969031359091, "rms_gain_error_db": -0.08892094921437044, "reference_mean_power": 0.0022214967634666704, "output_mean_power": 0.0021764745334828623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.581275628581243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00040407180276463134, "output_power": 0.00038396541884884553, "reference_active": true, "delta_db": -0.2216643306188385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0351036816934105e-05, "output_power": 3.4939522669884646e-05, "reference_active": true, "delta_db": -0.6253773084053247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1841518228550588e-06, "output_power": 7.908274480691513e-07, "reference_active": true, "delta_db": -1.7532565344818196}], "active_band_count": 3, "active_band_mae_db": 0.8667660578353277}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.952734804355257e-05, "log_rms_mae_db": 1.4167009163763327, "rms_envelope_nrmse": 0.16917376779099047, "envelope_correlation": 0.9789655013933485, "rms_gain_error_db": -0.17052384891514685, "reference_mean_power": 0.00025078805819662737, "output_mean_power": 0.0002411317888016526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4212041991971396e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.171024783262431e-05, "output_power": 6.549517924617344e-05, "reference_active": true, "delta_db": -0.39371888390448095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.275733290040072e-06, "output_power": 7.2202274387096465e-06, "reference_active": true, "delta_db": -0.03325892392939599}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0724449404458702e-06, "output_power": 1.2283073327560843e-06, "reference_active": true, "delta_db": 0.5893203985236941}], "active_band_count": 3, "active_band_mae_db": 0.33876606878585697}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.039214481073785e-05, "log_rms_mae_db": 0.9940548983635729, "rms_envelope_nrmse": 0.14287846404337015, "envelope_correlation": 0.9843980893068325, "rms_gain_error_db": -0.17186423114116942, "reference_mean_power": 0.0002518169764916092, "output_mean_power": 0.00024204637456013125}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3611130645010071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.488974762253847e-05, "output_power": 5.995694261059151e-05, "reference_active": true, "delta_db": -0.3433660632279936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.711335389958727e-06, "output_power": 3.6393163052584795e-06, "reference_active": true, "delta_db": -0.08510399689011483}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.486818337389537e-08, "output_power": 9.439527436874652e-08, "reference_active": true, "delta_db": 0.4620534703586779}], "active_band_count": 3, "active_band_mae_db": 0.29684117682559547}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.8930089256612995e-05, "log_rms_mae_db": 1.5106610590146028, "rms_envelope_nrmse": 0.1824401238169318, "envelope_correlation": 0.9745197695188461, "rms_gain_error_db": -0.364747062750001, "reference_mean_power": 2.8173794881918634e-05, "output_mean_power": 2.5904227296498092e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9380195589864167e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.123435668533653e-06, "output_power": 5.652976310668173e-06, "reference_active": true, "delta_db": -0.34717994088697973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.722245443537651e-06, "output_power": 2.1351136310382263e-06, "reference_active": true, "delta_db": -1.0550628637498278}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.909737577930936e-07, "output_power": 6.550805279900538e-07, "reference_active": true, "delta_db": -0.23166863404813345}], "active_band_count": 3, "active_band_mae_db": 0.5446371462283137}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3848385775447824e-05, "log_rms_mae_db": 0.934081384261376, "rms_envelope_nrmse": 0.10833450755966331, "envelope_correlation": 0.991154916243526, "rms_gain_error_db": -0.36453993974102933, "reference_mean_power": 2.8310170166312676e-05, "output_mean_power": 2.6030858166353422e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.80878248957878e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.951898475340433e-06, "output_power": 5.548966835139036e-06, "reference_active": true, "delta_db": -0.30443385465205014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2501096733423988e-06, "output_power": 9.727748410692493e-07, "reference_active": true, "delta_db": -1.089357858673208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.567916579161296e-08, "output_power": 4.450760787287318e-08, "reference_active": true, "delta_db": -1.689933749464426}], "active_band_count": 3, "active_band_mae_db": 1.0279084875965614}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9473974299255742, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c1a4a24126a02635eea733421908de0d90b745036b3a0c260a2ed6d8e79cd45e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9787436679845674, "standard_deviation": 0.001026482217092365, "minimum": 0.977215316147481, "maximum": 0.9805668255852832, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3993585398382604, "standard_deviation": 0.014504780217851464, "minimum": 1.3788158996603659, "maximum": 1.4167009163763327, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47668933563539206, "standard_deviation": 0.21389926348133467, "minimum": 0.17754291330990288, "maximum": 0.7346261704710463, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17004193957964825, "standard_deviation": 0.004118516819769433, "minimum": 0.16262888807079, "maximum": 0.17605591124608497, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9839457960685826, "standard_deviation": 0.0008215616280523971, "minimum": 0.9823636496943801, "maximum": 0.9849371865838668, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9986286224183558, "standard_deviation": 0.02103867241172362, "minimum": 0.9594734326645912, "maximum": 1.023029574088301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.44677764929207375, "standard_deviation": 0.21657377598031086, "minimum": 0.18555958568987507, "maximum": 0.8389389610944148, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14467212975354776, "standard_deviation": 0.0036738722987683663, "minimum": 0.1400523023159557, "maximum": 0.15164340924016284, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9721153103725115, "standard_deviation": 0.005159260926577685, "minimum": 0.9603642848435016, "maximum": 0.9775219973703131, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5226920341515489, "standard_deviation": 0.029133667079167605, "minimum": 1.4841284924005194, "maximum": 1.5795137706862188, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7408934429583909, "standard_deviation": 0.4619139354867656, "minimum": 0.2790734590303144, "maximum": 1.7727847155074048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1890465551192932, "standard_deviation": 0.01591591394958279, "minimum": 0.17026794025353356, "maximum": 0.22473305910726335, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.989827533667607, "standard_deviation": 0.0032611034583621707, "minimum": 0.9828716030833176, "maximum": 0.9946706263191897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9312699165990355, "standard_deviation": 0.03654631966684963, "minimum": 0.8604794869088924, "maximum": 0.9747636266340386, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9383481703422489, "standard_deviation": 0.40065496585232635, "minimum": 0.4045378159718284, "maximum": 1.6435374250365349, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11069670556221063, "standard_deviation": 0.016508566930459545, "minimum": 0.08253652527982887, "maximum": 0.14374989633765362, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9940169800694594, "standard_deviation": 0.001043251453340958, "minimum": 0.9920888472673731, "maximum": 0.995055787997552, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4207598523213352, "standard_deviation": 0.032444123429594976, "minimum": 1.3491199451749, "maximum": 1.4455524327052396, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6493537467690309, "standard_deviation": 0.24191173325789, "minimum": 0.31317053937898515, "maximum": 1.0952889814010278, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.083363414511491, "standard_deviation": 0.007805395023960541, "minimum": 0.0752652068879021, "maximum": 0.09752312012425259, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978301606615883, "standard_deviation": 0.0006355138253807836, "minimum": 0.996393496162322, "maximum": 0.9983602765305225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9478919432629918, "standard_deviation": 0.026157718812370483, "minimum": 0.9170701463296719, "maximum": 0.9931010746814949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7638138139927876, "standard_deviation": 0.3374238141122026, "minimum": 0.30629874645042504, "maximum": 1.3565334886651392, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05035192054697443, "standard_deviation": 0.0077352921882346245, "minimum": 0.043559067429811235, "maximum": 0.067257941576125, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.127739877581557, "standard_deviation": 0.03667935326438519, "minimum": 0.0594962781617925, "maximum": 0.18423828443419463, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.908239311824562, "standard_deviation": 0.00972856314843424, "minimum": 4.888655525868735, "maximum": 4.916547111524219, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44786594653479805, "standard_deviation": 0.009918397525324685, "minimum": 0.43449938652040504, "maximum": 0.46484818935579, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08346162826611475, "standard_deviation": 0.02622009628424428, "minimum": -0.12277527074553576, "maximum": -0.05271435912618837, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9965868038396254, "standard_deviation": 0.0004928943223624213, "minimum": 0.9959628027056594, "maximum": 0.9972521017539561, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5517061761626384, "standard_deviation": 0.0173235375586715, "minimum": 0.5279833919282145, "maximum": 0.5792210402268614, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4010073949603898, "standard_deviation": 0.18038869484149672, "minimum": 0.178193710836997, "maximum": 0.6203499647250904, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05170158720711061, "standard_deviation": 0.0038850123156654827, "minimum": 0.04688962785617452, "maximum": 0.05664737979384719, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989670859074452, "standard_deviation": 0.00023055840036697607, "minimum": 0.9986181234164565, "maximum": 0.9992933217028735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.30700868456190616, "standard_deviation": 0.008718388169066546, "minimum": 0.2932344136008289, "maximum": 0.31687143026413944, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47018958574323116, "standard_deviation": 0.23908788944160675, "minimum": 0.12242134765869798, "maximum": 0.7331639534600005, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028315604982652353, "standard_deviation": 0.0032024152180078866, "minimum": 0.024550430214194154, "maximum": 0.03331068098908808, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1836-0014", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.9920976161956787, "predicted_mos": 2.9920976161956787, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ef9861474662bdb9f31e0f7d394e206e158224b8203a5d287ef6e5f814424ac3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.044046299997717, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5400589687741075, "全帯域_10ms_rms_envelope_nrmse": 0.051563950973042096, "全帯域_10ms_envelope_correlation": 0.9966719445894092, "全帯域_10ms_modulation_mae_db": 0.15025758264024838, "全帯域_40ms_log_rms_mae_db": 0.3184816394928321, "全帯域_40ms_rms_envelope_nrmse": 0.03052524420599228, "全帯域_40ms_envelope_correlation": 0.9988568428704665, "全帯域_40ms_modulation_mae_db": 0.10477850463865745, "300–1000 Hz_10ms_log_rms_mae_db": 1.3557177661139954, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08720568857021134, "300–1000 Hz_10ms_envelope_correlation": 0.9936212919130812, "300–1000 Hz_10ms_modulation_mae_db": 0.7484451316236003, "300–1000 Hz_40ms_log_rms_mae_db": 0.9354289650256988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054149496993685084, "300–1000 Hz_40ms_envelope_correlation": 0.9976922860056011, "300–1000 Hz_40ms_modulation_mae_db": 0.8038500706494043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4066533386989613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16796831415500993, "1000–3000 Hz_10ms_envelope_correlation": 0.9792767033693852, "1000–3000 Hz_10ms_modulation_mae_db": 0.4462959792846161, "1000–3000 Hz_40ms_log_rms_mae_db": 1.009513234654715, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14245527595137786, "1000–3000 Hz_40ms_envelope_correlation": 0.9844741569985735, "1000–3000 Hz_40ms_modulation_mae_db": 0.533319724224559, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5008837990520152, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15375607702158753, "3000–6000 Hz_10ms_envelope_correlation": 0.981708127356604, "3000–6000 Hz_10ms_modulation_mae_db": 0.7985463267941105, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9053255351818635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0829048081198861, "3000–6000 Hz_40ms_envelope_correlation": 0.994501938880339, "3000–6000 Hz_40ms_modulation_mae_db": 0.9647140842653489, "mr_stft_db": 4.895874703247976, "band_mae_db": 0.1338317794560753, "rms_gain_error_db": -0.10583698457307045, "output_peak": 0.4681887165829229}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5608107370184907, "全帯域_10ms_rms_envelope_nrmse": 0.055841639677762775, "全帯域_10ms_envelope_correlation": 0.9959783025728892, "全帯域_10ms_modulation_mae_db": 0.25747890454114747, "全帯域_40ms_log_rms_mae_db": 0.2965619148089964, "全帯域_40ms_rms_envelope_nrmse": 0.027933200421901467, "全帯域_40ms_envelope_correlation": 0.9989676153726418, "全帯域_40ms_modulation_mae_db": 0.352563179544509, "300–1000 Hz_10ms_log_rms_mae_db": 1.4003548735127742, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08725900994981613, "300–1000 Hz_10ms_envelope_correlation": 0.9933581974390409, "300–1000 Hz_10ms_modulation_mae_db": 0.568077875780613, "300–1000 Hz_40ms_log_rms_mae_db": 0.9235636944561337, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04735789430508803, "300–1000 Hz_40ms_envelope_correlation": 0.9980409744131078, "300–1000 Hz_40ms_modulation_mae_db": 0.8549429210741845, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3921526712847987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16400166285946977, "1000–3000 Hz_10ms_envelope_correlation": 0.9802478137021021, "1000–3000 Hz_10ms_modulation_mae_db": 0.5246020942342328, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9626695102766414, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14039614547868579, "1000–3000 Hz_40ms_envelope_correlation": 0.9849126694011437, "1000–3000 Hz_40ms_modulation_mae_db": 0.3191844862136283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4830511227416971, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17816067525142332, "3000–6000 Hz_10ms_envelope_correlation": 0.9752388379519513, "3000–6000 Hz_10ms_modulation_mae_db": 0.4911398847189512, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8529183232808245, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09538165813351485, "3000–6000 Hz_40ms_envelope_correlation": 0.9923726653537703, "3000–6000 Hz_40ms_modulation_mae_db": 0.8858433734098395, "mr_stft_db": 4.9011165791953415, "band_mae_db": 0.11008865970731524, "rms_gain_error_db": -0.07472063491213983, "output_peak": 0.47078104060016085}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5360991998350474, "全帯域_10ms_rms_envelope_nrmse": 0.0532286447399727, "全帯域_10ms_envelope_correlation": 0.9963472460371157, "全帯域_10ms_modulation_mae_db": 0.37085298639233505, "全帯域_40ms_log_rms_mae_db": 0.3001127161351503, "全帯域_40ms_rms_envelope_nrmse": 0.029107249468305677, "全帯域_40ms_envelope_correlation": 0.9988587627051932, "全帯域_40ms_modulation_mae_db": 0.23133805696297946, "300–1000 Hz_10ms_log_rms_mae_db": 1.3898089716412179, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09789941404171203, "300–1000 Hz_10ms_envelope_correlation": 0.9916328296026425, "300–1000 Hz_10ms_modulation_mae_db": 0.9871135842349851, "300–1000 Hz_40ms_log_rms_mae_db": 0.9562274162349137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060918859258801736, "300–1000 Hz_40ms_envelope_correlation": 0.996697033685658, "300–1000 Hz_40ms_modulation_mae_db": 1.107910268644445, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3797076415320608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16452352093858919, "1000–3000 Hz_10ms_envelope_correlation": 0.9801230222649301, "1000–3000 Hz_10ms_modulation_mae_db": 0.38093189572212793, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9710030872990927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14135349035632533, "1000–3000 Hz_40ms_envelope_correlation": 0.9846523835136755, "1000–3000 Hz_40ms_modulation_mae_db": 0.1948141537288167, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5228155900489175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17584432389944354, "3000–6000 Hz_10ms_envelope_correlation": 0.975950931812876, "3000–6000 Hz_10ms_modulation_mae_db": 0.9551385129850439, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9730347940516707, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10103016324631127, "3000–6000 Hz_40ms_envelope_correlation": 0.9915259814818083, "3000–6000 Hz_40ms_modulation_mae_db": 1.0506867239866233, "mr_stft_db": 4.90411966312718, "band_mae_db": 0.1350753020144391, "rms_gain_error_db": -0.0700574084967518, "output_peak": 0.4399088464374183}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5176957138904065, "全帯域_10ms_rms_envelope_nrmse": 0.05018713444425034, "全帯域_10ms_envelope_correlation": 0.9967720099572174, "全帯域_10ms_modulation_mae_db": 0.08593321128019275, "全帯域_40ms_log_rms_mae_db": 0.2753076960264779, "全帯域_40ms_rms_envelope_nrmse": 0.02729296160105063, "全帯域_40ms_envelope_correlation": 0.999021519410239, "全帯域_40ms_modulation_mae_db": 0.17165914056252662, "300–1000 Hz_10ms_log_rms_mae_db": 1.3769392796773512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08387882197450511, "300–1000 Hz_10ms_envelope_correlation": 0.9940856949807606, "300–1000 Hz_10ms_modulation_mae_db": 0.5176148681220097, "300–1000 Hz_40ms_log_rms_mae_db": 0.8889189235608352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052452368910849784, "300–1000 Hz_40ms_envelope_correlation": 0.9977841745492185, "300–1000 Hz_40ms_modulation_mae_db": 0.6553982245817632, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3807751941485902, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16434755019300074, "1000–3000 Hz_10ms_envelope_correlation": 0.9801632636489492, "1000–3000 Hz_10ms_modulation_mae_db": 0.3255558439094553, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0086732096063502, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1430297594175728, "1000–3000 Hz_40ms_envelope_correlation": 0.9843136877923551, "1000–3000 Hz_40ms_modulation_mae_db": 0.34364513160209526, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4723988499740368, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17652298307106745, "3000–6000 Hz_10ms_envelope_correlation": 0.9757886404780058, "3000–6000 Hz_10ms_modulation_mae_db": 1.065994697005368, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8935588809477178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10019869237980855, "3000–6000 Hz_40ms_envelope_correlation": 0.9918676130607147, "3000–6000 Hz_40ms_modulation_mae_db": 1.4755528575569878, "mr_stft_db": 4.888173775922415, "band_mae_db": 0.15748331827590953, "rms_gain_error_db": -0.07542181956294498, "output_peak": 0.4791261813804369}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.537743564650039, "全帯域_10ms_rms_envelope_nrmse": 0.05214168350066052, "全帯域_10ms_envelope_correlation": 0.9968090845613804, "全帯域_10ms_modulation_mae_db": 0.42208735503905565, "全帯域_40ms_log_rms_mae_db": 0.301730481866472, "全帯域_40ms_rms_envelope_nrmse": 0.031134653817162718, "全帯域_40ms_envelope_correlation": 0.999045945977913, "全帯域_40ms_modulation_mae_db": 0.6414480318558539, "300–1000 Hz_10ms_log_rms_mae_db": 1.3158347428942705, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07982219176833938, "300–1000 Hz_10ms_envelope_correlation": 0.9947820386818217, "300–1000 Hz_10ms_modulation_mae_db": 0.4457036906405132, "300–1000 Hz_40ms_log_rms_mae_db": 0.8798841348529494, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04904624101956615, "300–1000 Hz_40ms_envelope_correlation": 0.9982543845105192, "300–1000 Hz_40ms_modulation_mae_db": 0.7729219733624791, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3567843826834354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17283719377417628, "1000–3000 Hz_10ms_envelope_correlation": 0.9780728283181334, "1000–3000 Hz_10ms_modulation_mae_db": 0.5150371651467284, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9516944588143204, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1466888911993629, "1000–3000 Hz_40ms_envelope_correlation": 0.98365037162377, "1000–3000 Hz_40ms_modulation_mae_db": 0.49265560863217744, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4543365261335903, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17391195422019043, "3000–6000 Hz_10ms_envelope_correlation": 0.9764299380523798, "3000–6000 Hz_10ms_modulation_mae_db": 0.2654168512199602, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8567427016730549, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08708907566294978, "3000–6000 Hz_40ms_envelope_correlation": 0.9936476104448259, "3000–6000 Hz_40ms_modulation_mae_db": 0.7942782438459087, "mr_stft_db": 4.896745058560459, "band_mae_db": 0.14261033554615293, "rms_gain_error_db": -0.15524569953999734, "output_peak": 0.4597472126884494}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5800719794686057, "全帯域_10ms_rms_envelope_nrmse": 0.050072049603572415, "全帯域_10ms_envelope_correlation": 0.9968243812067656, "全帯域_10ms_modulation_mae_db": 0.33737077306714697, "全帯域_40ms_log_rms_mae_db": 0.3158960063229047, "全帯域_40ms_rms_envelope_nrmse": 0.027295513973200933, "全帯域_40ms_envelope_correlation": 0.9990666036378442, "全帯域_40ms_modulation_mae_db": 0.47968758876006895, "300–1000 Hz_10ms_log_rms_mae_db": 1.3789973149077293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08475118537016005, "300–1000 Hz_10ms_envelope_correlation": 0.9938076652727171, "300–1000 Hz_10ms_modulation_mae_db": 0.38895536449028983, "300–1000 Hz_40ms_log_rms_mae_db": 0.9663829257044766, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051860689951223, "300–1000 Hz_40ms_envelope_correlation": 0.9976818683458805, "300–1000 Hz_40ms_modulation_mae_db": 0.5904217060260913, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3574699612961683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16965953594678515, "1000–3000 Hz_10ms_envelope_correlation": 0.9788666940060211, "1000–3000 Hz_10ms_modulation_mae_db": 0.4202561039807369, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9870819113353468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14685673994480322, "1000–3000 Hz_40ms_envelope_correlation": 0.9834308280912527, "1000–3000 Hz_40ms_modulation_mae_db": 0.768773082097932, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4895783915504224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17092323394624076, "3000–6000 Hz_10ms_envelope_correlation": 0.9772591607136197, "3000–6000 Hz_10ms_modulation_mae_db": 0.9749506975586422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8836217879101502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09146258164967892, "3000–6000 Hz_40ms_envelope_correlation": 0.9931217256609849, "3000–6000 Hz_40ms_modulation_mae_db": 1.0114185014245003, "mr_stft_db": 4.887639218505394, "band_mae_db": 0.09042232164827768, "rms_gain_error_db": -0.0902106653663663, "output_peak": 0.4820049061272573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5637392166405127, "全帯域_10ms_rms_envelope_nrmse": 0.05502969731127295, "全帯域_10ms_envelope_correlation": 0.996067788474763, "全帯域_10ms_modulation_mae_db": 0.2344455199598511, "全帯域_40ms_log_rms_mae_db": 0.3198981943401931, "全帯域_40ms_rms_envelope_nrmse": 0.028234472869674527, "全帯域_40ms_envelope_correlation": 0.9989040471084562, "全帯域_40ms_modulation_mae_db": 0.16123448191314174, "300–1000 Hz_10ms_log_rms_mae_db": 1.4398664046284664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0859088028107514, "300–1000 Hz_10ms_envelope_correlation": 0.9935778618798858, "300–1000 Hz_10ms_modulation_mae_db": 0.8139147362906725, "300–1000 Hz_40ms_log_rms_mae_db": 0.974921883789848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048725947972033105, "300–1000 Hz_40ms_envelope_correlation": 0.997934140225655, "300–1000 Hz_40ms_modulation_mae_db": 0.9028389944453465, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4116670808919163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16556019743095982, "1000–3000 Hz_10ms_envelope_correlation": 0.9798729752564097, "1000–3000 Hz_10ms_modulation_mae_db": 0.4579700051507431, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9995178572798112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14143606345326684, "1000–3000 Hz_40ms_envelope_correlation": 0.9846287113502323, "1000–3000 Hz_40ms_modulation_mae_db": 0.7003865126508821, "3000–6000 Hz_10ms_log_rms_mae_db": 1.558240346687779, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18225450717837283, "3000–6000 Hz_10ms_envelope_correlation": 0.974119102839567, "3000–6000 Hz_10ms_modulation_mae_db": 0.5315456737028843, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9404589157188554, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09226660546180737, "3000–6000 Hz_40ms_envelope_correlation": 0.9929014094080597, "3000–6000 Hz_40ms_modulation_mae_db": 1.1349526790921454, "mr_stft_db": 4.9082378783220735, "band_mae_db": 0.09640947287710065, "rms_gain_error_db": -0.05808666154509919, "output_peak": 0.46088834135120654}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5654411216242768, "全帯域_10ms_rms_envelope_nrmse": 0.05552900079449937, "全帯域_10ms_envelope_correlation": 0.9962523687658656, "全帯域_10ms_modulation_mae_db": 0.6152489933993776, "全帯域_40ms_log_rms_mae_db": 0.32877461576174255, "全帯域_40ms_rms_envelope_nrmse": 0.032629167669442065, "全帯域_40ms_envelope_correlation": 0.9988334292987368, "全帯域_40ms_modulation_mae_db": 0.7274854102206006, "300–1000 Hz_10ms_log_rms_mae_db": 1.4030660242567572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09025484691633537, "300–1000 Hz_10ms_envelope_correlation": 0.9931817366564998, "300–1000 Hz_10ms_modulation_mae_db": 0.6040302440209615, "300–1000 Hz_40ms_log_rms_mae_db": 0.9392164362882708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055102134605700355, "300–1000 Hz_40ms_envelope_correlation": 0.9976657662819973, "300–1000 Hz_40ms_modulation_mae_db": 0.7099674314608305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3543872777268762, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1689500362277849, "1000–3000 Hz_10ms_envelope_correlation": 0.9790208845067904, "1000–3000 Hz_10ms_modulation_mae_db": 0.5824398271202184, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9669014769967075, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14440758159359282, "1000–3000 Hz_40ms_envelope_correlation": 0.9840391730311351, "1000–3000 Hz_40ms_modulation_mae_db": 0.7816714374620292, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5477991968012395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1966795687496645, "3000–6000 Hz_10ms_envelope_correlation": 0.9698178984814145, "3000–6000 Hz_10ms_modulation_mae_db": 0.7978404715579398, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9727819660690721, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1181006931076412, "3000–6000 Hz_40ms_envelope_correlation": 0.9884823123904403, "3000–6000 Hz_40ms_modulation_mae_db": 1.0012575328035567, "mr_stft_db": 4.898538353710937, "band_mae_db": 0.18032901709715654, "rms_gain_error_db": -0.1438012888659151, "output_peak": 0.45528420414105325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.934099969086407, "spectral_convergence": 0.15361891690201662, "spectral_convergence_status": "ok", "amplitude_floor": 1.7246496227499867e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.7536964010309255, "spectral_convergence": 0.16434218660883068, "spectral_convergence_status": "ok", "amplitude_floor": 1.4526318235172953e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.999827739626595, "spectral_convergence": 0.1670651234209631, "spectral_convergence_status": "ok", "amplitude_floor": 1.3962213853324247e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0014645134106370877, "output_power": 0.0014738929692236927, "delta_db": 0.027725942879034193, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024771538682985465, "output_power": 0.002399443209209545, "delta_db": -0.13842508236270962, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013427666217707404, "output_power": 0.0012866397587748646, "delta_db": -0.1854356958972425, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024269361162520585, "output_power": 0.00023526601241678533, "delta_db": -0.13499153057710975, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.4425202041437504e-05, "output_power": 4.15431477363913e-05, "delta_db": -0.2913001154523178, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.4446805694883255e-05, "output_power": 3.424819789978383e-05, "delta_db": -0.025112309568037977, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1338317794560753, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10583698457307045, "pitch": null, "output_rms_dbfs": -22.619306422131807, "output_peak": 0.4681887165829229, "reference_rms_dbfs": -22.513469437558737, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.895874703247976}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002110151907366841, "log_rms_mae_db": 0.5400589687741075, "rms_envelope_nrmse": 0.051563950973042096, "envelope_correlation": 0.9966719445894092, "rms_gain_error_db": -0.10352227131762044, "reference_mean_power": 0.005604891233518836, "output_mean_power": 0.005472867856236763}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2250047113785757e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011643562088722453, "output_power": 0.0011310136308987197, "reference_active": true, "delta_db": -0.12618024329495856}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011188245530999946, "output_power": 0.00010904454402726403, "reference_active": true, "delta_db": -0.11158047771904145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8450792038001933e-05, "output_power": 2.708901155697558e-05, "reference_active": true, "delta_db": -0.21301202690674512}], "active_band_count": 3, "active_band_mae_db": 0.15025758264024838}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020713167740975364, "log_rms_mae_db": 0.3184816394928321, "rms_envelope_nrmse": 0.03052524420599228, "envelope_correlation": 0.9988568428704665, "rms_gain_error_db": -0.1033003034479397, "reference_mean_power": 0.005603505754865554, "output_mean_power": 0.005471794668802192}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.048317169639576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010458478077031374, "output_power": 0.0010178907604990088, "reference_active": true, "delta_db": -0.11767318132417906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.131979407845485e-05, "output_power": 5.979799001324474e-05, "reference_active": true, "delta_db": -0.10914101439665852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6292932861003563e-06, "output_power": 1.662460746682926e-06, "reference_active": true, "delta_db": 0.08752131819513478}], "active_band_count": 3, "active_band_mae_db": 0.10477850463865745}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015831513319989338, "log_rms_mae_db": 1.3557177661139954, "rms_envelope_nrmse": 0.08720568857021134, "envelope_correlation": 0.9936212919130812, "rms_gain_error_db": -0.19613085601944852, "reference_mean_power": 0.0022309508329157127, "output_mean_power": 0.002132440457490228}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0339679622838414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046453953265101344, "output_power": 0.00043013678491409717, "reference_active": true, "delta_db": -0.3341609405869964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.019812862224435e-05, "output_power": 6.402806082967045e-05, "reference_active": true, "delta_db": -0.3995518576995153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.545275737486218e-05, "output_power": 1.091047669806362e-05, "reference_active": true, "delta_db": -1.5116225965842889}], "active_band_count": 3, "active_band_mae_db": 0.7484451316236003}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014375880677218868, "log_rms_mae_db": 0.9354289650256988, "rms_envelope_nrmse": 0.054149496993685084, "envelope_correlation": 0.9976922860056011, "rms_gain_error_db": -0.1965093500032836, "reference_mean_power": 0.0022214967634666704, "output_mean_power": 0.0021232187944124482}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.581275628581243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00040407180276463134, "output_power": 0.00037633160343271134, "reference_active": true, "delta_db": -0.3088785437479165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0351036816934105e-05, "output_power": 3.6811212592876765e-05, "reference_active": true, "delta_db": -0.3987457465178001}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1841518228550588e-06, "output_power": 7.998615044974408e-07, "reference_active": true, "delta_db": -1.703925921682496}], "active_band_count": 3, "active_band_mae_db": 0.8038500706494043}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.952734804355257e-05, "log_rms_mae_db": 1.4066533386989613, "rms_envelope_nrmse": 0.16796831415500993, "envelope_correlation": 0.9792767033693852, "rms_gain_error_db": -0.17008292839495265, "reference_mean_power": 0.00025078805819662737, "output_mean_power": 0.0002411562711184839}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4212041991971396e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.171024783262431e-05, "output_power": 6.45114076371105e-05, "reference_active": true, "delta_db": -0.45944705147313664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.275733290040072e-06, "output_power": 5.984449460968949e-06, "reference_active": true, "delta_db": -0.8485256685917696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0724449404458702e-06, "output_power": 1.0801063676868247e-06, "reference_active": true, "delta_db": 0.030915217788942122}], "active_band_count": 3, "active_band_mae_db": 0.4462959792846161}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.039214481073785e-05, "log_rms_mae_db": 1.009513234654715, "rms_envelope_nrmse": 0.14245527595137786, "envelope_correlation": 0.9844741569985735, "rms_gain_error_db": -0.17037066327999323, "reference_mean_power": 0.0002518169764916092, "output_mean_power": 0.00024212963024756678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3611130645010071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.488974762253847e-05, "output_power": 5.826244105633336e-05, "reference_active": true, "delta_db": -0.4678740848023617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.711335389958727e-06, "output_power": 3.014576066254968e-06, "reference_active": true, "delta_db": -0.9030395610023291}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.486818337389537e-08, "output_power": 8.946423625653685e-08, "reference_active": true, "delta_db": 0.22904552686898583}], "active_band_count": 3, "active_band_mae_db": 0.533319724224559}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.8930089256612995e-05, "log_rms_mae_db": 1.5008837990520152, "rms_envelope_nrmse": 0.15375607702158753, "envelope_correlation": 0.981708127356604, "rms_gain_error_db": -0.19778113386562485, "reference_mean_power": 2.8173794881918634e-05, "output_mean_power": 2.6919515244628546e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9380195589864167e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.123435668533653e-06, "output_power": 5.8968774579401745e-06, "reference_active": true, "delta_db": -0.1637305673257029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.722245443537651e-06, "output_power": 2.5184131185280685e-06, "reference_active": true, "delta_db": -0.33800306618064097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.909737577930936e-07, "output_power": 4.467566431935515e-07, "reference_active": true, "delta_db": -1.8939053468759879}], "active_band_count": 3, "active_band_mae_db": 0.7985463267941105}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3848385775447824e-05, "log_rms_mae_db": 0.9053255351818635, "rms_envelope_nrmse": 0.0829048081198861, "envelope_correlation": 0.994501938880339, "rms_gain_error_db": -0.1974262340761447, "reference_mean_power": 2.8310170166312676e-05, "output_mean_power": 2.7052029753375725e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.80878248957878e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.951898475340433e-06, "output_power": 5.7507243811907155e-06, "reference_active": true, "delta_db": -0.14932961237097978}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2501096733423988e-06, "output_power": 1.1643667102203497e-06, "reference_active": true, "delta_db": -0.3085833549313911}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.567916579161296e-08, "output_power": 3.74804416295111e-08, "reference_active": true, "delta_db": -2.436229285493676}], "active_band_count": 3, "active_band_mae_db": 0.9647140842653489}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9477454527436924, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8b1ea041e9cd294d4872eec63a5542e1dfa8f390fc9a4090af2115e2a4e817e8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9794555231340901, "standard_deviation": 0.0007771646530232656, "minimum": 0.9780728283181334, "maximum": 0.9802478137021021, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.379949693532851, "standard_deviation": 0.022564628205968988, "minimum": 1.3543872777268762, "maximum": 1.4116670808919163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.45663611431860734, "standard_deviation": 0.08298192585625151, "minimum": 0.3255558439094553, "maximum": 0.5824398271202184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16723100144072198, "standard_deviation": 0.003155286684811024, "minimum": 0.16400166285946977, "maximum": 0.17283719377417628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9842627477252672, "standard_deviation": 0.0005172057879268367, "minimum": 0.9834308280912527, "maximum": 0.9849126694011437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9821318432828732, "standard_deviation": 0.02217460395036915, "minimum": 0.9516944588143204, "maximum": 1.009513234654715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.516806267076515, "standard_deviation": 0.22068612412173955, "minimum": 0.1948141537288167, "maximum": 0.7816714374620292, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14332799342437344, "standard_deviation": 0.002444748586048786, "minimum": 0.14039614547868579, "maximum": 0.14685673994480322, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9757890797108022, "standard_deviation": 0.003303239814740161, "minimum": 0.9698178984814145, "maximum": 0.981708127356604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5036379778737121, "standard_deviation": 0.03651810720024733, "minimum": 1.4543365261335903, "maximum": 1.558240346687779, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7350716394428625, "standard_deviation": 0.2790165287968406, "minimum": 0.2654168512199602, "maximum": 1.065994697005368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1760066654172488, "standard_deviation": 0.01193227873333428, "minimum": 0.15375607702158753, "maximum": 0.1966795687496645, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9923026570851179, "standard_deviation": 0.0018153134585896982, "minimum": 0.9884823123904403, "maximum": 0.994501938880339, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9098053631041512, "standard_deviation": 0.04769700759627723, "minimum": 0.8529183232808245, "maximum": 0.9730347940516707, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.039837999548114, "standard_deviation": 0.20387867400968482, "minimum": 0.7942782438459087, "maximum": 1.4755528575569878, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09605428472019975, "standard_deviation": 0.010804403803352496, "minimum": 0.0829048081198861, "maximum": 0.1181006931076412, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935059145533063, "standard_deviation": 0.0009031325541122852, "minimum": 0.9916328296026425, "maximum": 0.9947820386818217, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3825731722040704, "standard_deviation": 0.036431492352524124, "minimum": 1.3158347428942705, "maximum": 1.4398664046284664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6342319369004557, "standard_deviation": 0.2015821557678669, "minimum": 0.38895536449028983, "maximum": 0.9871135842349851, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08712249517522884, "standard_deviation": 0.0053035186449639955, "minimum": 0.07982219176833938, "maximum": 0.09789941404171203, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977188285022047, "standard_deviation": 0.0004613920356123123, "minimum": 0.996697033685658, "maximum": 0.9982543845105192, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9330680474891409, "standard_deviation": 0.03445362127813145, "minimum": 0.8798841348529494, "maximum": 0.974921883789848, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7997814487805681, "standard_deviation": 0.16128483034759722, "minimum": 0.5904217060260913, "maximum": 1.107910268644445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0524517041271184, "standard_deviation": 0.004365033676333315, "minimum": 0.04735789430508803, "maximum": 0.060918859258801736, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13078127582780338, "standard_deviation": 0.030645668548169823, "minimum": 0.09042232164827768, "maximum": 0.18032901709715654, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.897555653823972, "standard_deviation": 0.007178141510727187, "minimum": 4.887639218505394, "maximum": 4.9082378783220735, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4644911811636132, "standard_deviation": 0.013620851617223937, "minimum": 0.4399088464374183, "maximum": 0.4820049061272573, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09667264535778562, "standard_deviation": 0.035673902732832614, "minimum": -0.15524569953999734, "maximum": -0.05808666154509919, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964653907706758, "standard_deviation": 0.0003460715060919224, "minimum": 0.9959783025728892, "maximum": 0.9968243812067656, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5502075627376858, "standard_deviation": 0.020472591755221068, "minimum": 0.5176957138904065, "maximum": 0.5800719794686057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3092094157899194, "standard_deviation": 0.16663830752710096, "minimum": 0.08593321128019275, "maximum": 0.6152489933993776, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05294922513062915, "standard_deviation": 0.0023270501299152905, "minimum": 0.050072049603572415, "maximum": 0.055841639677762775, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989443457976863, "standard_deviation": 9.31009419783535e-05, "minimum": 0.9988334292987368, "maximum": 0.9990666036378442, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3070954080943461, "standard_deviation": 0.01707106873846206, "minimum": 0.2753076960264779, "maximum": 0.32877461576174255, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3587742993072922, "standard_deviation": 0.23461179047805858, "minimum": 0.10477850463865745, "maximum": 0.7274854102206006, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.029269058003341286, "standard_deviation": 0.0019650403628444974, "minimum": 0.02729296160105063, "maximum": 0.032629167669442065, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_1995-1837-0023", "language": "en", "style": "normal", "speaker": "librispeech_1995", "source_utterance_id": "1995-1837-0023", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_1995-1837-0023.wav", "source_file_sha256": "45e127676f7df5f1022b53136bfaf1face26c246667f6bc2047a3e720c217a08", "decoded_pcm_sha256": "2fff476dd7493d67e46c49e37f104d20891e347d86560cf73c09f434075020cb", "original_source_sha256": "7f584164eef2f5976f874ee9c027933c892ba7f920f7233bfd2a219d0ebffcd4", "corpus_entry_id": "librispeech_1995-1837-0023"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.168759822845459, "predicted_mos": 4.168759822845459, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7c529ea9767e4426d960d28c18b5ebc84a9715f189f8e459579735d1d433c3e1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7914808679997805, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_1995-1837-0023", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7c529ea9767e4426d960d28c18b5ebc84a9715f189f8e459579735d1d433c3e1", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "44084159233cef136382433cc3d4cb8938a7407571790316e3a868a2292e6e56"}, "rows": [{"case": "librispeech_1995-1837-0023", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.199575424194336, "predicted_mos": 3.199575424194336, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9802a07dec88ac50c5e1d95db80f6311a12e7d4644fa6095e61c99705e85ad05", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.109492894000141, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_1995-1837-0023", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9802a07dec88ac50c5e1d95db80f6311a12e7d4644fa6095e61c99705e85ad05", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5796047841286703, "全帯域_10ms_rms_envelope_nrmse": 0.051757921555778984, "全帯域_10ms_envelope_correlation": 0.9962180608895648, "全帯域_10ms_modulation_mae_db": 0.31145564281647836, "全帯域_40ms_log_rms_mae_db": 0.3376191522832251, "全帯域_40ms_rms_envelope_nrmse": 0.02862188976784378, "全帯域_40ms_envelope_correlation": 0.9987624162740906, "全帯域_40ms_modulation_mae_db": 0.5438162886034258, "300–1000 Hz_10ms_log_rms_mae_db": 1.2589339720393005, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0728911753409451, "300–1000 Hz_10ms_envelope_correlation": 0.9949119386825553, "300–1000 Hz_10ms_modulation_mae_db": 0.2146930974469157, "300–1000 Hz_40ms_log_rms_mae_db": 0.7940516128703915, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038845756500353025, "300–1000 Hz_40ms_envelope_correlation": 0.9984539634809209, "300–1000 Hz_40ms_modulation_mae_db": 0.3808705492180952, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3264140781125655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10090447681696621, "1000–3000 Hz_10ms_envelope_correlation": 0.9922056278110307, "1000–3000 Hz_10ms_modulation_mae_db": 0.6409683147655156, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7979147959129039, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0574298684535108, "1000–3000 Hz_40ms_envelope_correlation": 0.9975088329155223, "1000–3000 Hz_40ms_modulation_mae_db": 0.6939111814996655, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3956021815723954, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16258032569441502, "3000–6000 Hz_10ms_envelope_correlation": 0.9812569127557772, "3000–6000 Hz_10ms_modulation_mae_db": 0.22209091842691775, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7940934454303047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07480246560991072, "3000–6000 Hz_40ms_envelope_correlation": 0.9958527700692843, "3000–6000 Hz_40ms_modulation_mae_db": 0.34219823522496967, "mr_stft_db": 4.665429619351497, "band_mae_db": 0.06792219188960186, "rms_gain_error_db": -0.05888925268672865, "output_peak": 0.3808165281615233}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5818578420594059, "全帯域_10ms_rms_envelope_nrmse": 0.04514605014255378, "全帯域_10ms_envelope_correlation": 0.9971351822576286, "全帯域_10ms_modulation_mae_db": 0.2709540444768087, "全帯域_40ms_log_rms_mae_db": 0.33141734932533523, "全帯域_40ms_rms_envelope_nrmse": 0.021857627628830117, "全帯域_40ms_envelope_correlation": 0.9993044554492421, "全帯域_40ms_modulation_mae_db": 0.2925107875330119, "300–1000 Hz_10ms_log_rms_mae_db": 1.2762594599677466, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06894060116671671, "300–1000 Hz_10ms_envelope_correlation": 0.9955088145616273, "300–1000 Hz_10ms_modulation_mae_db": 0.41997785899429285, "300–1000 Hz_40ms_log_rms_mae_db": 0.7910958730357086, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038613138272038806, "300–1000 Hz_40ms_envelope_correlation": 0.9985108529171487, "300–1000 Hz_40ms_modulation_mae_db": 0.5439720688811484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3807440793962493, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10484447462559816, "1000–3000 Hz_10ms_envelope_correlation": 0.99168462136465, "1000–3000 Hz_10ms_modulation_mae_db": 0.4954039575189502, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8632194549119155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059823630404825426, "1000–3000 Hz_40ms_envelope_correlation": 0.9973892648790917, "1000–3000 Hz_40ms_modulation_mae_db": 0.38582065097532486, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3820791472289085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14412889796114975, "3000–6000 Hz_10ms_envelope_correlation": 0.985151046941277, "3000–6000 Hz_10ms_modulation_mae_db": 0.40403103370300636, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7777253740673568, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07294853415785124, "3000–6000 Hz_40ms_envelope_correlation": 0.9961268516805881, "3000–6000 Hz_40ms_modulation_mae_db": 0.7481713850860552, "mr_stft_db": 4.672387903104057, "band_mae_db": 0.1473814934144778, "rms_gain_error_db": -0.057197387800384705, "output_peak": 0.3980592780013229}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5632403917018659, "全帯域_10ms_rms_envelope_nrmse": 0.04123324322020458, "全帯域_10ms_envelope_correlation": 0.9975989982191271, "全帯域_10ms_modulation_mae_db": 0.3643719286794527, "全帯域_40ms_log_rms_mae_db": 0.3223677732864855, "全帯域_40ms_rms_envelope_nrmse": 0.020296265592484167, "全帯域_40ms_envelope_correlation": 0.9993800600748028, "全帯域_40ms_modulation_mae_db": 0.473982719401347, "300–1000 Hz_10ms_log_rms_mae_db": 1.2203718238142616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07070994862363636, "300–1000 Hz_10ms_envelope_correlation": 0.9952385758779109, "300–1000 Hz_10ms_modulation_mae_db": 0.5768173521275876, "300–1000 Hz_40ms_log_rms_mae_db": 0.7312778594917945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03865774751593637, "300–1000 Hz_40ms_envelope_correlation": 0.998491199615632, "300–1000 Hz_40ms_modulation_mae_db": 0.796864069420682, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3488831949766817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09316098514845926, "1000–3000 Hz_10ms_envelope_correlation": 0.9934487952690227, "1000–3000 Hz_10ms_modulation_mae_db": 0.20280003419817874, "1000–3000 Hz_40ms_log_rms_mae_db": 0.815226420447657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05370214335557442, "1000–3000 Hz_40ms_envelope_correlation": 0.9979487650028366, "1000–3000 Hz_40ms_modulation_mae_db": 0.3128266640525947, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3687135104103332, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14467845189239484, "3000–6000 Hz_10ms_envelope_correlation": 0.9855207798174199, "3000–6000 Hz_10ms_modulation_mae_db": 0.27385118072339315, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7760266464646041, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07601589331382533, "3000–6000 Hz_40ms_envelope_correlation": 0.9959701328101164, "3000–6000 Hz_40ms_modulation_mae_db": 0.5137575326217875, "mr_stft_db": 4.6749203402728545, "band_mae_db": 0.14621185832627145, "rms_gain_error_db": -0.04215035675962124, "output_peak": 0.35479873584916194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5389414336327163, "全帯域_10ms_rms_envelope_nrmse": 0.04725216765311821, "全帯域_10ms_envelope_correlation": 0.9969427154186733, "全帯域_10ms_modulation_mae_db": 0.7887829011656383, "全帯域_40ms_log_rms_mae_db": 0.29462153806266395, "全帯域_40ms_rms_envelope_nrmse": 0.028466192810100648, "全帯域_40ms_envelope_correlation": 0.9988825125764941, "全帯域_40ms_modulation_mae_db": 0.9798861523322776, "300–1000 Hz_10ms_log_rms_mae_db": 1.2456361727787733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07765495172908836, "300–1000 Hz_10ms_envelope_correlation": 0.9943916364202052, "300–1000 Hz_10ms_modulation_mae_db": 1.049877522752064, "300–1000 Hz_40ms_log_rms_mae_db": 0.7894101462086808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051144606969491684, "300–1000 Hz_40ms_envelope_correlation": 0.9974602708235238, "300–1000 Hz_40ms_modulation_mae_db": 1.1709132815613918, "1000–3000 Hz_10ms_log_rms_mae_db": 1.365916338193421, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09728237073981416, "1000–3000 Hz_10ms_envelope_correlation": 0.9929961801005741, "1000–3000 Hz_10ms_modulation_mae_db": 0.8342892272207102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8412818931815835, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056524980698527454, "1000–3000 Hz_40ms_envelope_correlation": 0.9978178825959686, "1000–3000 Hz_40ms_modulation_mae_db": 0.9055376386778633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3657162744368807, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14019563668721344, "3000–6000 Hz_10ms_envelope_correlation": 0.9859525858076785, "3000–6000 Hz_10ms_modulation_mae_db": 0.4377476843947315, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7969078393170373, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07259456388050763, "3000–6000 Hz_40ms_envelope_correlation": 0.9960359700519171, "3000–6000 Hz_40ms_modulation_mae_db": 0.8491417424235198, "mr_stft_db": 4.683108403462971, "band_mae_db": 0.10152566776644774, "rms_gain_error_db": -0.08733809733285516, "output_peak": 0.3686969440849567}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5771038233955245, "全帯域_10ms_rms_envelope_nrmse": 0.04871120253376943, "全帯域_10ms_envelope_correlation": 0.9966831007891817, "全帯域_10ms_modulation_mae_db": 0.41895491118753897, "全帯域_40ms_log_rms_mae_db": 0.318159058574836, "全帯域_40ms_rms_envelope_nrmse": 0.02621064539798848, "全帯域_40ms_envelope_correlation": 0.9990090579045909, "全帯域_40ms_modulation_mae_db": 0.45135797752288753, "300–1000 Hz_10ms_log_rms_mae_db": 1.22142110827928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07957432561824553, "300–1000 Hz_10ms_envelope_correlation": 0.9940096028244882, "300–1000 Hz_10ms_modulation_mae_db": 0.9023155144330232, "300–1000 Hz_40ms_log_rms_mae_db": 0.8019299177325042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04686651130172547, "300–1000 Hz_40ms_envelope_correlation": 0.9978360774665701, "300–1000 Hz_40ms_modulation_mae_db": 1.1914633492177324, "1000–3000 Hz_10ms_log_rms_mae_db": 1.383090730505098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09913516038688634, "1000–3000 Hz_10ms_envelope_correlation": 0.9925904705581494, "1000–3000 Hz_10ms_modulation_mae_db": 0.24321714744932785, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8680597332898226, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057611643814486774, "1000–3000 Hz_40ms_envelope_correlation": 0.9976166568596284, "1000–3000 Hz_40ms_modulation_mae_db": 0.13896168263764216, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3798881703043788, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1488835336512666, "3000–6000 Hz_10ms_envelope_correlation": 0.9842009506262167, "3000–6000 Hz_10ms_modulation_mae_db": 0.42041799381790046, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7925194972271012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07988793755038102, "3000–6000 Hz_40ms_envelope_correlation": 0.9952174571716176, "3000–6000 Hz_40ms_modulation_mae_db": 0.4251671684942626, "mr_stft_db": 4.679934068322315, "band_mae_db": 0.12088988569199037, "rms_gain_error_db": -0.06845813431590497, "output_peak": 0.36605731638474615}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5674152896701387, "全帯域_10ms_rms_envelope_nrmse": 0.04381473714566977, "全帯域_10ms_envelope_correlation": 0.9973344414048637, "全帯域_10ms_modulation_mae_db": 0.38223578432447775, "全帯域_40ms_log_rms_mae_db": 0.31229494652935785, "全帯域_40ms_rms_envelope_nrmse": 0.02242549453573899, "全帯域_40ms_envelope_correlation": 0.9992913730243288, "全帯域_40ms_modulation_mae_db": 0.5180846246943297, "300–1000 Hz_10ms_log_rms_mae_db": 1.2519010194473497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.074761813172379, "300–1000 Hz_10ms_envelope_correlation": 0.9948039028896133, "300–1000 Hz_10ms_modulation_mae_db": 0.6929379858683021, "300–1000 Hz_40ms_log_rms_mae_db": 0.7633933451042323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04648666329892538, "300–1000 Hz_40ms_envelope_correlation": 0.9979396044662293, "300–1000 Hz_40ms_modulation_mae_db": 0.8259704858157891, "1000–3000 Hz_10ms_log_rms_mae_db": 1.290047990318185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09306770481883883, "1000–3000 Hz_10ms_envelope_correlation": 0.9934665508687384, "1000–3000 Hz_10ms_modulation_mae_db": 0.389494940266138, "1000–3000 Hz_40ms_log_rms_mae_db": 0.754936228072689, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05213584481736725, "1000–3000 Hz_40ms_envelope_correlation": 0.9980476569707907, "1000–3000 Hz_40ms_modulation_mae_db": 0.39481326085552043, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4167903403141249, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16234915462720223, "3000–6000 Hz_10ms_envelope_correlation": 0.9818023663620152, "3000–6000 Hz_10ms_modulation_mae_db": 0.46500741523616074, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8207806122225264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0959217025225612, "3000–6000 Hz_40ms_envelope_correlation": 0.9934732505813428, "3000–6000 Hz_40ms_modulation_mae_db": 0.7718321068710793, "mr_stft_db": 4.668681099713486, "band_mae_db": 0.09034628618243405, "rms_gain_error_db": -0.07156923017824908, "output_peak": 0.36421410156250017}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5567441725277672, "全帯域_10ms_rms_envelope_nrmse": 0.04398560724967629, "全帯域_10ms_envelope_correlation": 0.9972636779974491, "全帯域_10ms_modulation_mae_db": 0.448198082476359, "全帯域_40ms_log_rms_mae_db": 0.29909935697194184, "全帯域_40ms_rms_envelope_nrmse": 0.022797864601687628, "全帯域_40ms_envelope_correlation": 0.999211145314311, "全帯域_40ms_modulation_mae_db": 0.8729338553670485, "300–1000 Hz_10ms_log_rms_mae_db": 1.2171264548569791, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07102758909410307, "300–1000 Hz_10ms_envelope_correlation": 0.9952227597620806, "300–1000 Hz_10ms_modulation_mae_db": 1.0205114151060168, "300–1000 Hz_40ms_log_rms_mae_db": 0.756610764288712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03899354998646717, "300–1000 Hz_40ms_envelope_correlation": 0.9985066622275794, "300–1000 Hz_40ms_modulation_mae_db": 1.3853751276027877, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3415287745646134, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09179873351442973, "1000–3000 Hz_10ms_envelope_correlation": 0.9934825909679501, "1000–3000 Hz_10ms_modulation_mae_db": 0.2452241751353772, "1000–3000 Hz_40ms_log_rms_mae_db": 0.798011820702332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04706419821246717, "1000–3000 Hz_40ms_envelope_correlation": 0.9982650136989487, "1000–3000 Hz_40ms_modulation_mae_db": 0.039269501859161886, "3000–6000 Hz_10ms_log_rms_mae_db": 1.350585976822377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14948579071762333, "3000–6000 Hz_10ms_envelope_correlation": 0.9840220048531455, "3000–6000 Hz_10ms_modulation_mae_db": 0.445308882911484, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7948001799787191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08359680643909828, "3000–6000 Hz_40ms_envelope_correlation": 0.9947914590220669, "3000–6000 Hz_40ms_modulation_mae_db": 0.6296515113199826, "mr_stft_db": 4.676689250056389, "band_mae_db": 0.08645803313019886, "rms_gain_error_db": -0.04496884912684026, "output_peak": 0.3575644060021102}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5659571496658322, "全帯域_10ms_rms_envelope_nrmse": 0.04292514001175756, "全帯域_10ms_envelope_correlation": 0.9974618406523815, "全帯域_10ms_modulation_mae_db": 0.28429679104745936, "全帯域_40ms_log_rms_mae_db": 0.3293041164321825, "全帯域_40ms_rms_envelope_nrmse": 0.023401459588880787, "全帯域_40ms_envelope_correlation": 0.9992492416423707, "全帯域_40ms_modulation_mae_db": 0.38101310319298193, "300–1000 Hz_10ms_log_rms_mae_db": 1.2476520287239496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0678181025240678, "300–1000 Hz_10ms_envelope_correlation": 0.9957116356214062, "300–1000 Hz_10ms_modulation_mae_db": 0.31008999792752645, "300–1000 Hz_40ms_log_rms_mae_db": 0.8043409778821035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03887216824786607, "300–1000 Hz_40ms_envelope_correlation": 0.9985773466046536, "300–1000 Hz_40ms_modulation_mae_db": 0.45129155011831834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3125353642623636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09936710991264876, "1000–3000 Hz_10ms_envelope_correlation": 0.9924530236820309, "1000–3000 Hz_10ms_modulation_mae_db": 0.39919812417110684, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7947154847424135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05720424075783387, "1000–3000 Hz_40ms_envelope_correlation": 0.9975071184237584, "1000–3000 Hz_40ms_modulation_mae_db": 0.46123596845792586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3710406852429962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14531265952650194, "3000–6000 Hz_10ms_envelope_correlation": 0.9848778413499614, "3000–6000 Hz_10ms_modulation_mae_db": 1.0421946554975963, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7592863949329467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0726892507934787, "3000–6000 Hz_40ms_envelope_correlation": 0.9960435219607989, "3000–6000 Hz_40ms_modulation_mae_db": 1.270390138727428, "mr_stft_db": 4.668479955059535, "band_mae_db": 0.13153508045520948, "rms_gain_error_db": -0.07863516846658072, "output_peak": 0.35308223990232035}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.591031342499742, "spectral_convergence": 0.12435846902679493, "spectral_convergence_status": "ok", "amplitude_floor": 1.5362649016028353e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.544892251559878, "spectral_convergence": 0.11892335096458231, "spectral_convergence_status": "ok", "amplitude_floor": 1.0749003550909146e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.860365263994872, "spectral_convergence": 0.11606460178533583, "spectral_convergence_status": "ok", "amplitude_floor": 8.625899017329306e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001501246216341323, "output_power": 0.0014872652697176315, "delta_db": -0.040634893291850216, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002746582421816528, "output_power": 0.002711568441894701, "delta_db": -0.05572065428672308, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011276426514964012, "output_power": 0.0011076005864528273, "delta_db": -0.07788317063815528, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016504304929751857, "output_power": 0.0001583333846632302, "delta_db": -0.18024743265844345, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.934184208485853e-05, "output_power": 3.9666300057060574e-05, "delta_db": 0.03567001963381629, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8643647097645308e-05, "output_power": 2.8529267183631635e-05, "delta_db": -0.017376980828622903, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.06792219188960186, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05888925268672865, "pitch": null, "output_rms_dbfs": -22.570422143672033, "output_peak": 0.3808165281615233, "reference_rms_dbfs": -22.511532890985304, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.665429619351497}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018834661138467316, "log_rms_mae_db": 0.5796047841286703, "rms_envelope_nrmse": 0.051757921555778984, "envelope_correlation": 0.9962180608895648, "rms_gain_error_db": -0.05637262713393035, "reference_mean_power": 0.005606081823103195, "output_mean_power": 0.005533783568454463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7511238572638392e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009260227506744139, "output_power": 0.0009221113337226135, "reference_active": true, "delta_db": -0.01838296635753096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010953814679795108, "output_power": 0.0001016395301423672, "reference_active": true, "delta_db": -0.32502740386183093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.583845526426034e-05, "output_power": 2.255126405820423e-05, "reference_active": true, "delta_db": -0.5909565582300733}], "active_band_count": 3, "active_band_mae_db": 0.31145564281647836}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015988223985812614, "log_rms_mae_db": 0.3376191522832251, "rms_envelope_nrmse": 0.02862188976784378, "envelope_correlation": 0.9987624162740906, "rms_gain_error_db": -0.05365613964225086, "reference_mean_power": 0.00558793457857638, "output_mean_power": 0.005519321583040347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5615527006220852e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008311543470725633, "output_power": 0.000832131947746287, "reference_active": true, "delta_db": 0.005105153915053181}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.497505864184299e-05, "output_power": 5.027354047913483e-05, "reference_active": true, "delta_db": -0.3882623010256369}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0542332226205543e-06, "output_power": 1.5446909694691916e-06, "reference_active": true, "delta_db": -1.2380814108695872}], "active_band_count": 3, "active_band_mae_db": 0.5438162886034258}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016200114302549476, "log_rms_mae_db": 1.2589339720393005, "rms_envelope_nrmse": 0.0728911753409451, "envelope_correlation": 0.9949119386825553, "rms_gain_error_db": -0.04727805439133676, "reference_mean_power": 0.0020825301889685474, "output_mean_power": 0.002059982354793469}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.598844168109503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038009245216565237, "output_power": 0.00037516437926531315, "reference_active": true, "delta_db": -0.05667648812849376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.129961338774397e-05, "output_power": 5.899677005407659e-05, "reference_active": true, "delta_db": -0.16629499856237068}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4399990501668035e-05, "output_power": 1.306927285451808e-05, "reference_active": true, "delta_db": -0.42110780564988265}], "active_band_count": 3, "active_band_mae_db": 0.2146930974469157}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012179662863477574, "log_rms_mae_db": 0.7940516128703915, "rms_envelope_nrmse": 0.038845756500353025, "envelope_correlation": 0.9984539634809209, "rms_gain_error_db": -0.04488153117858703, "reference_mean_power": 0.0020746475669658963, "output_mean_power": 0.002053317827994397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.889619852823582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035505972043036534, "output_power": 0.0003518956592579322, "reference_active": true, "delta_db": -0.038874970679720726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2260222284985983e-05, "output_power": 3.134891703848451e-05, "reference_active": true, "delta_db": -0.1244481297406002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1886983564224358e-06, "output_power": 9.487303796236806e-07, "reference_active": true, "delta_db": -0.9792885472339647}], "active_band_count": 3, "active_band_mae_db": 0.3808705492180952}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.2267966954040695e-05, "log_rms_mae_db": 1.3264140781125655, "rms_envelope_nrmse": 0.10090447681696621, "envelope_correlation": 0.9922056278110307, "rms_gain_error_db": -0.14895649471793113, "reference_mean_power": 0.00015543877011727877, "output_mean_power": 0.00015019784509861562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4080839081981722e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.101806003601554e-05, "output_power": 6.7929366319951e-05, "reference_active": true, "delta_db": -0.19311241440184004}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0132703192354966e-06, "output_power": 1.9380305201689243e-06, "reference_active": true, "delta_db": -0.1654147891688447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.589135119213016e-07, "output_power": 5.293641314594475e-07, "reference_active": true, "delta_db": -1.564377740725862}], "active_band_count": 3, "active_band_mae_db": 0.6409683147655156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.184196106800431e-05, "log_rms_mae_db": 0.7979147959129039, "rms_envelope_nrmse": 0.0574298684535108, "envelope_correlation": 0.9975088329155223, "rms_gain_error_db": -0.1496013192541384, "reference_mean_power": 0.00015383740434286893, "output_mean_power": 0.00014862840309374037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3572636139953185e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.549215276795424e-05, "output_power": 6.311966278374678e-05, "reference_active": true, "delta_db": -0.1602459620117621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.98253690709034e-07, "output_power": 9.76846094711593e-07, "reference_active": true, "delta_db": -0.09414779931593079}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.830181704119583e-08, "output_power": 3.171242771047234e-08, "reference_active": true, "delta_db": -1.8273397831713036}], "active_band_count": 3, "active_band_mae_db": 0.6939111814996655}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.781014610300033e-05, "log_rms_mae_db": 1.3956021815723954, "rms_envelope_nrmse": 0.16258032569441502, "envelope_correlation": 0.9812569127557772, "rms_gain_error_db": 0.03340487772284491, "reference_mean_power": 2.2945322907755735e-05, "output_mean_power": 2.312249326603623e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.468236349213924e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8725122830509875e-06, "output_power": 3.99908906442321e-06, "reference_active": true, "delta_db": 0.1396827253786288}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.737803238621202e-07, "output_power": 5.438926443650633e-07, "reference_active": true, "delta_db": -0.23232465859550433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.433384337397866e-07, "output_power": 1.533872254443355e-07, "reference_active": true, "delta_db": 0.29426537130662006}], "active_band_count": 3, "active_band_mae_db": 0.22209091842691775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.4952988731037596e-05, "log_rms_mae_db": 0.7940934454303047, "rms_envelope_nrmse": 0.07480246560991072, "envelope_correlation": 0.9958527700692843, "rms_gain_error_db": 0.03443305002212245, "reference_mean_power": 2.2958960928334722e-05, "output_mean_power": 2.3141714637926883e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.967813463196935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5837220182078332e-06, "output_power": 3.7347750593412633e-06, "reference_active": true, "delta_db": 0.17930134960135946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.664756741318262e-07, "output_power": 2.536155534014401e-07, "reference_active": true, "delta_db": -0.21481685695523922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1431200548978765e-08, "output_power": 9.881985087280438e-09, "reference_active": true, "delta_db": -0.6324764991183102}], "active_band_count": 3, "active_band_mae_db": 0.34219823522496967}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9576588012945946, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b5c145f37c4d8f9dffb49d14abe747e66dc3d9012b028b2f9c7837edd726edfc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9927909825777683, "standard_deviation": 0.0006687552257024014, "minimum": 0.99168462136465, "maximum": 0.9934825909679501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3436450687911472, "standard_deviation": 0.032944955174994704, "minimum": 1.290047990318185, "maximum": 1.383090730505098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4313244900906631, "standard_deviation": 0.21891297419268108, "minimum": 0.20280003419817874, "maximum": 0.8342892272207102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09744512699545518, "standard_deviation": 0.004514921656593808, "minimum": 0.09179873351442973, "maximum": 0.10484447462559816, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9977626489183182, "standard_deviation": 0.00030746373484878854, "minimum": 0.9973892648790917, "maximum": 0.9982650136989487, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8166707289076646, "standard_deviation": 0.038560768952528725, "minimum": 0.754936228072689, "maximum": 0.8680597332898226, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.41654706862696234, "standard_deviation": 0.2802055696511198, "minimum": 0.039269501859161886, "maximum": 0.9055376386778633, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05518706881432414, "standard_deviation": 0.004066613310275998, "minimum": 0.04706419821246717, "maximum": 0.059823630404825426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9840980610641864, "standard_deviation": 0.0017129313184308314, "minimum": 0.9812569127557772, "maximum": 0.9859525858076785, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3788020357915491, "standard_deviation": 0.020236725338561235, "minimum": 1.350585976822377, "maximum": 1.4167903403141249, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4638312205888988, "standard_deviation": 0.24943382574301767, "minimum": 0.22209091842691775, "maximum": 1.0421946554975963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14970180634472088, "standard_deviation": 0.00838750074579866, "minimum": 0.14019563668721344, "maximum": 0.16258032569441502, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954389266684666, "standard_deviation": 0.000924339840683283, "minimum": 0.9934732505813428, "maximum": 0.9961268516805881, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7890174987050745, "standard_deviation": 0.018216297017035747, "minimum": 0.7592863949329467, "maximum": 0.8207806122225264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6937887275961356, "standard_deviation": 0.29226949868584823, "minimum": 0.34219823522496967, "maximum": 1.270390138727428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07855714428345177, "standard_deviation": 0.008024014877195519, "minimum": 0.07259456388050763, "maximum": 0.0959217025225612, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949748583299859, "standard_deviation": 0.0005687081515086906, "minimum": 0.9940096028244882, "maximum": 0.9957116356214062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2424127549884552, "standard_deviation": 0.02108576526505866, "minimum": 1.2171264548569791, "maximum": 1.2762594599677466, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.648402593081966, "standard_deviation": 0.322039019803293, "minimum": 0.2146930974469157, "maximum": 1.049877522752064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07292231340864774, "standard_deviation": 0.004147691271240313, "minimum": 0.0678181025240678, "maximum": 0.07957432561824553, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982219972002822, "standard_deviation": 0.0004184887894157995, "minimum": 0.9974602708235238, "maximum": 0.9985773466046536, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.779013812076766, "standard_deviation": 0.025833050113292844, "minimum": 0.7312778594917945, "maximum": 0.8043409778821035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8433400602294931, "standard_deviation": 0.37453953651851885, "minimum": 0.3808705492180952, "maximum": 1.3853751276027877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.042310017761600496, "standard_deviation": 0.005043819946575514, "minimum": 0.038613138272038806, "maximum": 0.051144606969491684, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11153381210707895, "standard_deviation": 0.029411009382782145, "minimum": 0.06792219188960186, "maximum": 0.1473814934144778, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.673703829917888, "standard_deviation": 0.006098775241288454, "minimum": 4.665429619351497, "maximum": 4.683108403462971, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36791119374358017, "standard_deviation": 0.015081056409250714, "minimum": 0.35308223990232035, "maximum": 0.3980592780013229, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0636508095833956, "standard_deviation": 0.015797986141023807, "minimum": -0.08733809733285516, "maximum": -0.04215035675962124, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9970797522036088, "standard_deviation": 0.0004529268516742643, "minimum": 0.9962180608895648, "maximum": 0.9975989982191271, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5663581108477401, "standard_deviation": 0.014072703709881227, "minimum": 0.5389414336327163, "maximum": 0.5818578420594059, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40865626077177664, "standard_deviation": 0.16591743319208252, "minimum": 0.2709540444768087, "maximum": 0.7887829011656383, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.045603258689066076, "standard_deviation": 0.0034341164534262376, "minimum": 0.04123324322020458, "maximum": 0.051757921555778984, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991362827825289, "standard_deviation": 0.0002237626781755714, "minimum": 0.9987624162740906, "maximum": 0.9993800600748028, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.31811041143325347, "standard_deviation": 0.015343956882624365, "minimum": 0.29462153806266395, "maximum": 0.3376191522832251, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5641981885809138, "standard_deviation": 0.2387164011608471, "minimum": 0.2925107875330119, "maximum": 0.9798861523322776, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.024259679990444323, "standard_deviation": 0.003122409378166139, "minimum": 0.020296265592484167, "maximum": 0.02862188976784378, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1837-0023", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.2000627517700195, "predicted_mos": 3.2000627517700195, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f098367390758124deaf0d583f43cf457d63f945261b6048d043d12eb141f97e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.42071729991585, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5882353903725034, "全帯域_10ms_rms_envelope_nrmse": 0.05437505459796354, "全帯域_10ms_envelope_correlation": 0.9958229362775136, "全帯域_10ms_modulation_mae_db": 0.30587233971086575, "全帯域_40ms_log_rms_mae_db": 0.34682227827794665, "全帯域_40ms_rms_envelope_nrmse": 0.03036259679805499, "全帯域_40ms_envelope_correlation": 0.9986042455814759, "全帯域_40ms_modulation_mae_db": 0.5412082149413817, "300–1000 Hz_10ms_log_rms_mae_db": 1.2867317949795234, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07738625920806923, "300–1000 Hz_10ms_envelope_correlation": 0.9942655137707902, "300–1000 Hz_10ms_modulation_mae_db": 0.24474531490674958, "300–1000 Hz_40ms_log_rms_mae_db": 0.7950428079982776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041902614207447626, "300–1000 Hz_40ms_envelope_correlation": 0.9982081032627481, "300–1000 Hz_40ms_modulation_mae_db": 0.40729676597957215, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2916199112473024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09860161175029157, "1000–3000 Hz_10ms_envelope_correlation": 0.9927040278685416, "1000–3000 Hz_10ms_modulation_mae_db": 0.7293180563514478, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7613939907961375, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057574671053243244, "1000–3000 Hz_40ms_envelope_correlation": 0.9976696005816907, "1000–3000 Hz_40ms_modulation_mae_db": 0.7640371056012766, "3000–6000 Hz_10ms_log_rms_mae_db": 1.324754642520834, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15757020222913368, "3000–6000 Hz_10ms_envelope_correlation": 0.9823972428775869, "3000–6000 Hz_10ms_modulation_mae_db": 0.11869882407017422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8034293314426044, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07632267615044983, "3000–6000 Hz_40ms_envelope_correlation": 0.9958355595108632, "3000–6000 Hz_40ms_modulation_mae_db": 0.31738226425617416, "mr_stft_db": 4.568615583649877, "band_mae_db": 0.09914325025589708, "rms_gain_error_db": -0.06156610432612359, "output_peak": 0.3878036371153884}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5871014754173265, "全帯域_10ms_rms_envelope_nrmse": 0.047088189778261305, "全帯域_10ms_envelope_correlation": 0.9968665917162792, "全帯域_10ms_modulation_mae_db": 0.20640846783290356, "全帯域_40ms_log_rms_mae_db": 0.33219707323950454, "全帯域_40ms_rms_envelope_nrmse": 0.02246114537936715, "全帯域_40ms_envelope_correlation": 0.999248816884473, "全帯域_40ms_modulation_mae_db": 0.2679968928309277, "300–1000 Hz_10ms_log_rms_mae_db": 1.2812610172028585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07009214430022354, "300–1000 Hz_10ms_envelope_correlation": 0.9953540349868072, "300–1000 Hz_10ms_modulation_mae_db": 0.40118390984869245, "300–1000 Hz_40ms_log_rms_mae_db": 0.7964149357210756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039464214666410403, "300–1000 Hz_40ms_envelope_correlation": 0.9984335849595891, "300–1000 Hz_40ms_modulation_mae_db": 0.5983561232675646, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2981175446325288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0964696862240946, "1000–3000 Hz_10ms_envelope_correlation": 0.9931849409992249, "1000–3000 Hz_10ms_modulation_mae_db": 0.9069431853516102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7832149433467788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05542136596310763, "1000–3000 Hz_40ms_envelope_correlation": 0.9980172402035754, "1000–3000 Hz_40ms_modulation_mae_db": 0.9108766715105463, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3596835863955323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14294126823942127, "3000–6000 Hz_10ms_envelope_correlation": 0.9856170814386935, "3000–6000 Hz_10ms_modulation_mae_db": 0.4573508062475083, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8258315260505559, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07793081562627095, "3000–6000 Hz_40ms_envelope_correlation": 0.9959193674950745, "3000–6000 Hz_40ms_modulation_mae_db": 0.6514683989554365, "mr_stft_db": 4.578913215404645, "band_mae_db": 0.19140281546972893, "rms_gain_error_db": -0.05091347613575779, "output_peak": 0.40150244781196387}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.57623240892556, "全帯域_10ms_rms_envelope_nrmse": 0.042200780418296134, "全帯域_10ms_envelope_correlation": 0.9974816077436361, "全帯域_10ms_modulation_mae_db": 0.33421765379756435, "全帯域_40ms_log_rms_mae_db": 0.3271091288926839, "全帯域_40ms_rms_envelope_nrmse": 0.020155833295096504, "全帯域_40ms_envelope_correlation": 0.9993834111126254, "全帯域_40ms_modulation_mae_db": 0.4293523717257804, "300–1000 Hz_10ms_log_rms_mae_db": 1.282375714982855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07624909422745273, "300–1000 Hz_10ms_envelope_correlation": 0.9944647853667331, "300–1000 Hz_10ms_modulation_mae_db": 0.5377335112971476, "300–1000 Hz_40ms_log_rms_mae_db": 0.7681537857650875, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04153595526442393, "300–1000 Hz_40ms_envelope_correlation": 0.9982588481808039, "300–1000 Hz_40ms_modulation_mae_db": 0.813209750450703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3142068901800867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09157715075404109, "1000–3000 Hz_10ms_envelope_correlation": 0.9939127193492582, "1000–3000 Hz_10ms_modulation_mae_db": 0.4865793613243469, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8201280881923991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056791325155453834, "1000–3000 Hz_40ms_envelope_correlation": 0.9979629448604845, "1000–3000 Hz_40ms_modulation_mae_db": 0.8085429844561691, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3167395688562444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1355090757750022, "3000–6000 Hz_10ms_envelope_correlation": 0.9871501581989345, "3000–6000 Hz_10ms_modulation_mae_db": 0.4136817313776408, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8038253251237981, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07537597090968748, "3000–6000 Hz_40ms_envelope_correlation": 0.9960669092605261, "3000–6000 Hz_40ms_modulation_mae_db": 0.46544785402836036, "mr_stft_db": 4.592543659521618, "band_mae_db": 0.17164949014084174, "rms_gain_error_db": -0.0370808455893493, "output_peak": 0.3549970842223683}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5353601588801947, "全帯域_10ms_rms_envelope_nrmse": 0.048088458538372805, "全帯域_10ms_envelope_correlation": 0.9968445347597605, "全帯域_10ms_modulation_mae_db": 0.80910295767496, "全帯域_40ms_log_rms_mae_db": 0.2906961460727043, "全帯域_40ms_rms_envelope_nrmse": 0.029002071828528403, "全帯域_40ms_envelope_correlation": 0.9988505891599146, "全帯域_40ms_modulation_mae_db": 1.044385607409856, "300–1000 Hz_10ms_log_rms_mae_db": 1.2559350083734264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0797263082433853, "300–1000 Hz_10ms_envelope_correlation": 0.9940612324204415, "300–1000 Hz_10ms_modulation_mae_db": 1.0503983205696323, "300–1000 Hz_40ms_log_rms_mae_db": 0.8079466008974473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05182231923917474, "300–1000 Hz_40ms_envelope_correlation": 0.9973754781375463, "300–1000 Hz_40ms_modulation_mae_db": 1.2105088460948166, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3044153826609344, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.091719383897827, "1000–3000 Hz_10ms_envelope_correlation": 0.99400526444985, "1000–3000 Hz_10ms_modulation_mae_db": 1.0517648011556202, "1000–3000 Hz_40ms_log_rms_mae_db": 0.795754027457769, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05367530049720779, "1000–3000 Hz_40ms_envelope_correlation": 0.9982990440127668, "1000–3000 Hz_40ms_modulation_mae_db": 1.1836110666061919, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2774785423487323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12770665414964433, "3000–6000 Hz_10ms_envelope_correlation": 0.9884585691101834, "3000–6000 Hz_10ms_modulation_mae_db": 0.5216292887856371, "3000–6000 Hz_40ms_log_rms_mae_db": 0.763528496570856, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06287720395405981, "3000–6000 Hz_40ms_envelope_correlation": 0.9972933110483705, "3000–6000 Hz_40ms_modulation_mae_db": 0.8656021652507427, "mr_stft_db": 4.581819214981409, "band_mae_db": 0.14583618654960376, "rms_gain_error_db": -0.09147534379069208, "output_peak": 0.3780822858329181}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5785940352390873, "全帯域_10ms_rms_envelope_nrmse": 0.0496604931236463, "全帯域_10ms_envelope_correlation": 0.9965514776700893, "全帯域_10ms_modulation_mae_db": 0.3262673519641606, "全帯域_40ms_log_rms_mae_db": 0.32800190008799973, "全帯域_40ms_rms_envelope_nrmse": 0.02712518216418126, "全帯域_40ms_envelope_correlation": 0.9989370638703815, "全帯域_40ms_modulation_mae_db": 0.347313034274736, "300–1000 Hz_10ms_log_rms_mae_db": 1.2821285624586842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08200254041946405, "300–1000 Hz_10ms_envelope_correlation": 0.9936271394385523, "300–1000 Hz_10ms_modulation_mae_db": 0.8299276959484675, "300–1000 Hz_40ms_log_rms_mae_db": 0.818546325891264, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04897051132524065, "300–1000 Hz_40ms_envelope_correlation": 0.9976183842829541, "300–1000 Hz_40ms_modulation_mae_db": 1.1633006956447973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3004588188314916, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09004385849452295, "1000–3000 Hz_10ms_envelope_correlation": 0.9940945144956576, "1000–3000 Hz_10ms_modulation_mae_db": 0.5653778919641135, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8004736218840984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05324678117084224, "1000–3000 Hz_40ms_envelope_correlation": 0.9981918417858369, "1000–3000 Hz_40ms_modulation_mae_db": 0.7124400892014061, "3000–6000 Hz_10ms_log_rms_mae_db": 1.319321718892928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1363589261973017, "3000–6000 Hz_10ms_envelope_correlation": 0.9868694583933904, "3000–6000 Hz_10ms_modulation_mae_db": 0.3763111298174298, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7971055799545026, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0786728043769293, "3000–6000 Hz_40ms_envelope_correlation": 0.9956921620990807, "3000–6000 Hz_40ms_modulation_mae_db": 0.4392896177216614, "mr_stft_db": 4.591054690155663, "band_mae_db": 0.1533855793749622, "rms_gain_error_db": -0.07118468312495346, "output_peak": 0.3690967448127521}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5533622026648968, "全帯域_10ms_rms_envelope_nrmse": 0.04427905131110387, "全帯域_10ms_envelope_correlation": 0.9972675501524976, "全帯域_10ms_modulation_mae_db": 0.31749936456462075, "全帯域_40ms_log_rms_mae_db": 0.30310474352737443, "全帯域_40ms_rms_envelope_nrmse": 0.02208511950314285, "全帯域_40ms_envelope_correlation": 0.9993062779857191, "全帯域_40ms_modulation_mae_db": 0.3825878457518586, "300–1000 Hz_10ms_log_rms_mae_db": 1.2673476625880178, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07592379837511246, "300–1000 Hz_10ms_envelope_correlation": 0.9946112465358046, "300–1000 Hz_10ms_modulation_mae_db": 0.6731978901642925, "300–1000 Hz_40ms_log_rms_mae_db": 0.7601292439387941, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04611810202295768, "300–1000 Hz_40ms_envelope_correlation": 0.9979450966739585, "300–1000 Hz_40ms_modulation_mae_db": 0.7866184875820084, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3136552402058799, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0942804297359144, "1000–3000 Hz_10ms_envelope_correlation": 0.993439432663794, "1000–3000 Hz_10ms_modulation_mae_db": 0.5982307575352099, "1000–3000 Hz_40ms_log_rms_mae_db": 0.802576008492042, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05592118656298142, "1000–3000 Hz_40ms_envelope_correlation": 0.9978950617908339, "1000–3000 Hz_40ms_modulation_mae_db": 0.6321420611338386, "3000–6000 Hz_10ms_log_rms_mae_db": 1.333900977397397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14573908338595826, "3000–6000 Hz_10ms_envelope_correlation": 0.9850660559658531, "3000–6000 Hz_10ms_modulation_mae_db": 0.5203937162144369, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7701744865658776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0839090000478249, "3000–6000 Hz_40ms_envelope_correlation": 0.9949441344532874, "3000–6000 Hz_40ms_modulation_mae_db": 0.7493843319036663, "mr_stft_db": 4.575032151344158, "band_mae_db": 0.13509681036226617, "rms_gain_error_db": -0.0693814418969717, "output_peak": 0.3592276812368606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5483050261949617, "全帯域_10ms_rms_envelope_nrmse": 0.044747265946821116, "全帯域_10ms_envelope_correlation": 0.9971728087144301, "全帯域_10ms_modulation_mae_db": 0.4431934386832415, "全帯域_40ms_log_rms_mae_db": 0.29868739849467063, "全帯域_40ms_rms_envelope_nrmse": 0.022900916652882655, "全帯域_40ms_envelope_correlation": 0.999209932661051, "全帯域_40ms_modulation_mae_db": 0.8443615036085886, "300–1000 Hz_10ms_log_rms_mae_db": 1.2319444676546702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07183178483406882, "300–1000 Hz_10ms_envelope_correlation": 0.995117396354994, "300–1000 Hz_10ms_modulation_mae_db": 1.1722749239560142, "300–1000 Hz_40ms_log_rms_mae_db": 0.7507781254358532, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03942559058395059, "300–1000 Hz_40ms_envelope_correlation": 0.998471948688249, "300–1000 Hz_40ms_modulation_mae_db": 1.5822003827925073, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2610548486658808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08588323864944476, "1000–3000 Hz_10ms_envelope_correlation": 0.9944000029768563, "1000–3000 Hz_10ms_modulation_mae_db": 0.3358392942498205, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7478101308445023, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0458212750534143, "1000–3000 Hz_40ms_envelope_correlation": 0.9984924206898824, "1000–3000 Hz_40ms_modulation_mae_db": 0.40958887934806815, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2967747102762592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13802417262775013, "3000–6000 Hz_10ms_envelope_correlation": 0.9865485307492746, "3000–6000 Hz_10ms_modulation_mae_db": 0.6400087023405382, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8038474340469619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0850314458683237, "3000–6000 Hz_40ms_envelope_correlation": 0.9949352159685285, "3000–6000 Hz_40ms_modulation_mae_db": 0.7680165423925782, "mr_stft_db": 4.568118324398825, "band_mae_db": 0.14601751188078016, "rms_gain_error_db": -0.0493505348274395, "output_peak": 0.355755688061783}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5811718163087605, "全帯域_10ms_rms_envelope_nrmse": 0.043341214293980156, "全帯域_10ms_envelope_correlation": 0.9974020210326275, "全帯域_10ms_modulation_mae_db": 0.2681114711419183, "全帯域_40ms_log_rms_mae_db": 0.3472449385117774, "全帯域_40ms_rms_envelope_nrmse": 0.02342132535180797, "全帯域_40ms_envelope_correlation": 0.9992398246529224, "全帯域_40ms_modulation_mae_db": 0.33378717431185745, "300–1000 Hz_10ms_log_rms_mae_db": 1.232683851593048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06591433072852197, "300–1000 Hz_10ms_envelope_correlation": 0.9959226918997041, "300–1000 Hz_10ms_modulation_mae_db": 0.3474815838980829, "300–1000 Hz_40ms_log_rms_mae_db": 0.7479513800186858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03585400411207591, "300–1000 Hz_40ms_envelope_correlation": 0.9987749418740381, "300–1000 Hz_40ms_modulation_mae_db": 0.5003201523884121, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2695109867638823, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0958707609997702, "1000–3000 Hz_10ms_envelope_correlation": 0.9931171897918015, "1000–3000 Hz_10ms_modulation_mae_db": 0.6486641154885845, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7858090488022687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057950608508733584, "1000–3000 Hz_40ms_envelope_correlation": 0.9975982180258808, "1000–3000 Hz_40ms_modulation_mae_db": 0.8723296155736202, "3000–6000 Hz_10ms_log_rms_mae_db": 1.304951690230577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13304546860629707, "3000–6000 Hz_10ms_envelope_correlation": 0.9875223192957941, "3000–6000 Hz_10ms_modulation_mae_db": 1.1295099814680716, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7965599315028126, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07075079170553035, "3000–6000 Hz_40ms_envelope_correlation": 0.9966224198998223, "3000–6000 Hz_40ms_modulation_mae_db": 1.3382754535254036, "mr_stft_db": 4.571383255598628, "band_mae_db": 0.16742614366729733, "rms_gain_error_db": -0.07528605747312349, "output_peak": 0.3651725183100545}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.493341825581511, "spectral_convergence": 0.1277997431534897, "spectral_convergence_status": "ok", "amplitude_floor": 1.5362649016028353e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.452905196604884, "spectral_convergence": 0.12073504446929832, "spectral_convergence_status": "ok", "amplitude_floor": 1.0749003550909146e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.759599728763238, "spectral_convergence": 0.11756067662513828, "spectral_convergence_status": "ok", "amplitude_floor": 8.625899017329306e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001501246216341323, "output_power": 0.0014856183365565224, "delta_db": -0.04544674704619139, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002746582421816528, "output_power": 0.0027120405787085006, "delta_db": -0.054964528936700896, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011276426514964012, "output_power": 0.0011079560369667962, "delta_db": -0.0764896589896688, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016504304929751857, "output_power": 0.00015778568749722838, "delta_db": -0.19529632504743047, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.934184208485853e-05, "output_power": 3.792343493741929e-05, "delta_db": -0.1594702390319211, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8643647097645308e-05, "output_power": 2.822988533626905e-05, "delta_db": -0.06319200248346989, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09914325025589708, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06156610432612359, "pitch": null, "output_rms_dbfs": -22.573098995311426, "output_peak": 0.3878036371153884, "reference_rms_dbfs": -22.511532890985304, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.568615583649877}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018834661138467316, "log_rms_mae_db": 0.5882353903725034, "rms_envelope_nrmse": 0.05437505459796354, "envelope_correlation": 0.9958229362775136, "rms_gain_error_db": -0.05846301274470791, "reference_mean_power": 0.005606081823103195, "output_mean_power": 0.005531120638454658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7511238572638392e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009260227506744139, "output_power": 0.0009269613505380944, "reference_active": true, "delta_db": 0.00439970074333126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010953814679795108, "output_power": 0.00010191987883228433, "reference_active": true, "delta_db": -0.3130649043995953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.583845526426034e-05, "output_power": 2.2503563988038456e-05, "reference_active": true, "delta_db": -0.6001524139896707}], "active_band_count": 3, "active_band_mae_db": 0.30587233971086575}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015988223985812614, "log_rms_mae_db": 0.34682227827794665, "rms_envelope_nrmse": 0.03036259679805499, "envelope_correlation": 0.9986042455814759, "rms_gain_error_db": -0.05556837411347784, "reference_mean_power": 0.00558793457857638, "output_mean_power": 0.005516891915105178}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5615527006220852e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008311543470725633, "output_power": 0.0008373946586406834, "reference_active": true, "delta_db": 0.03248505744875208}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.497505864184299e-05, "output_power": 4.9918557067133686e-05, "reference_active": true, "delta_db": -0.4190367804246591}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0542332226205543e-06, "output_power": 1.5683372874153387e-06, "reference_active": true, "delta_db": -1.172102806950734}], "active_band_count": 3, "active_band_mae_db": 0.5412082149413817}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016200114302549476, "log_rms_mae_db": 1.2867317949795234, "rms_envelope_nrmse": 0.07738625920806923, "envelope_correlation": 0.9942655137707902, "rms_gain_error_db": -0.05591055772350174, "reference_mean_power": 0.0020825301889685474, "output_mean_power": 0.002055891779508568}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.598844168109503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038009245216565237, "output_power": 0.000375317391138729, "reference_active": true, "delta_db": -0.05490556667959061}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.129961338774397e-05, "output_power": 5.8290410414747197e-05, "reference_active": true, "delta_db": -0.21860622325157672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4399990501668035e-05, "output_power": 1.2950597031422447e-05, "reference_active": true, "delta_db": -0.46072415478908146}], "active_band_count": 3, "active_band_mae_db": 0.24474531490674958}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012179662863477574, "log_rms_mae_db": 0.7950428079982776, "rms_envelope_nrmse": 0.041902614207447626, "envelope_correlation": 0.9982081032627481, "rms_gain_error_db": -0.054353257886833005, "reference_mean_power": 0.0020746475669658963, "output_mean_power": 0.002048844533143145}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.889619852823582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035505972043036534, "output_power": 0.000351646404744258, "reference_active": true, "delta_db": -0.04195225262051866}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2260222284985983e-05, "output_power": 3.0935694080697296e-05, "reference_active": true, "delta_db": -0.182074911133894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1886983564224358e-06, "output_power": 9.446813657687603e-07, "reference_active": true, "delta_db": -0.9978631341843037}], "active_band_count": 3, "active_band_mae_db": 0.40729676597957215}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.2267966954040695e-05, "log_rms_mae_db": 1.2916199112473024, "rms_envelope_nrmse": 0.09860161175029157, "envelope_correlation": 0.9927040278685416, "rms_gain_error_db": -0.19586685669713028, "reference_mean_power": 0.00015543877011727877, "output_mean_power": 0.0001485842120951927}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4080839081981722e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.101806003601554e-05, "output_power": 6.807470753155273e-05, "reference_active": true, "delta_db": -0.18383020565136704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0132703192354966e-06, "output_power": 1.9132831621031475e-06, "reference_active": true, "delta_db": -0.2212284146569851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.589135119213016e-07, "output_power": 5.033878515304636e-07, "reference_active": true, "delta_db": -1.7828955487459912}], "active_band_count": 3, "active_band_mae_db": 0.7293180563514478}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.184196106800431e-05, "log_rms_mae_db": 0.7613939907961375, "rms_envelope_nrmse": 0.057574671053243244, "envelope_correlation": 0.9976696005816907, "rms_gain_error_db": -0.19536764054087777, "reference_mean_power": 0.00015383740434286893, "output_mean_power": 0.00014707036813188165}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3572636139953185e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.549215276795424e-05, "output_power": 6.32259837003066e-05, "reference_active": true, "delta_db": -0.15293671162489392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.98253690709034e-07, "output_power": 9.292056063173492e-07, "reference_active": true, "delta_db": -0.31129103096909483}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.830181704119583e-08, "output_power": 3.170845716626281e-08, "reference_active": true, "delta_db": -1.827883574209841}], "active_band_count": 3, "active_band_mae_db": 0.7640371056012766}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.781014610300033e-05, "log_rms_mae_db": 1.324754642520834, "rms_envelope_nrmse": 0.15757020222913368, "envelope_correlation": 0.9823972428775869, "rms_gain_error_db": -0.05535126916852948, "reference_mean_power": 2.2945322907755735e-05, "output_mean_power": 2.265473815554086e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.468236349213924e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8725122830509875e-06, "output_power": 3.964160676125102e-06, "reference_active": true, "delta_db": 0.10158445124744353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.737803238621202e-07, "output_power": 6.03099704382419e-07, "reference_active": true, "delta_db": 0.21643464304356907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.433384337397866e-07, "output_power": 1.4208718717011191e-07, "reference_active": true, "delta_db": -0.038077377919510075}], "active_band_count": 3, "active_band_mae_db": 0.11869882407017422}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.4952988731037596e-05, "log_rms_mae_db": 0.8034293314426044, "rms_envelope_nrmse": 0.07632267615044983, "envelope_correlation": 0.9958355595108632, "rms_gain_error_db": -0.04856492736661461, "reference_mean_power": 2.2958960928334722e-05, "output_mean_power": 2.2703652778926368e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.967813463196935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5837220182078332e-06, "output_power": 3.7181805933258638e-06, "reference_active": true, "delta_db": 0.1599616479686561}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.664756741318262e-07, "output_power": 2.8132976839620463e-07, "reference_active": true, "delta_db": 0.23558119111639453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1431200548978765e-08, "output_power": 1.0056143191057857e-08, "reference_active": true, "delta_db": -0.5566039536834719}], "active_band_count": 3, "active_band_mae_db": 0.31738226425617416}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9593571047799557, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c141f38dbf6e92dfbc22032c4d1b4718b0824e162006f27cc651730f7fe3dfb7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9936072615743731, "standard_deviation": 0.0005831059414254325, "minimum": 0.9927040278685416, "maximum": 0.9944000029768563, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2941299528984982, "standard_deviation": 0.01946580178857405, "minimum": 1.2610548486658808, "maximum": 1.3142068901800867, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6653396829275942, "standard_deviation": 0.22900082889273504, "minimum": 0.3358392942498205, "maximum": 1.0517648011556202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09305576506323832, "standard_deviation": 0.0040764867591363795, "minimum": 0.08588323864944476, "maximum": 0.09860161175029157, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.998015796493869, "standard_deviation": 0.00030472173735432943, "minimum": 0.9975982180258808, "maximum": 0.9984924206898824, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7871449824769995, "standard_deviation": 0.02331567272483796, "minimum": 0.7478101308445023, "maximum": 0.8201280881923991, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7866960591788896, "standard_deviation": 0.22453026597484918, "minimum": 0.40958887934806815, "maximum": 1.1836110666061919, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05455031424562301, "standard_deviation": 0.00391024039879599, "minimum": 0.0458212750534143, "maximum": 0.057950608508733584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9862036770037138, "standard_deviation": 0.0018673353154804503, "minimum": 0.9823972428775869, "maximum": 0.9884585691101834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.316700679614813, "standard_deviation": 0.02475184626372867, "minimum": 1.2774785423487323, "maximum": 1.3596835863955323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5221980225401797, "standard_deviation": 0.2885397437972489, "minimum": 0.11869882407017422, "maximum": 1.1295099814680716, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13961185640131357, "standard_deviation": 0.009148899008335147, "minimum": 0.12770665414964433, "maximum": 0.15757020222913368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9959136349669442, "standard_deviation": 0.0007911825716852974, "minimum": 0.9949352159685285, "maximum": 0.9972933110483705, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7955377639072462, "standard_deviation": 0.019961087353608273, "minimum": 0.763528496570856, "maximum": 0.8258315260505559, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6993583285042528, "standard_deviation": 0.3192888662235497, "minimum": 0.31738226425617416, "maximum": 1.3382754535254036, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07635883857988454, "standard_deviation": 0.007113534418108799, "minimum": 0.06287720395405981, "maximum": 0.0850314458683237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9946780050967283, "standard_deviation": 0.0007468143722371718, "minimum": 0.9936271394385523, "maximum": 0.9959226918997041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2650510099791354, "standard_deviation": 0.022540950543199288, "minimum": 1.2319444676546702, "maximum": 1.2867317949795234, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6571178938236348, "standard_deviation": 0.3371855523426781, "minimum": 0.24474531490674958, "maximum": 1.1722749239560142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07489078254203727, "standard_deviation": 0.005292312864170943, "minimum": 0.06591433072852197, "maximum": 0.08200254041946405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981357982574859, "standard_deviation": 0.00046486847476790187, "minimum": 0.9973754781375463, "maximum": 0.9987749418740381, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7806204007083106, "standard_deviation": 0.02719164915457542, "minimum": 0.7479513800186858, "maximum": 0.818546325891264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8827264005250476, "standard_deviation": 0.4040209939290721, "minimum": 0.40729676597957215, "maximum": 1.5822003827925073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04313666392771019, "standard_deviation": 0.005381775949969313, "minimum": 0.03585400411207591, "maximum": 0.05182231923917474, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15124472346267218, "standard_deviation": 0.027587116819761226, "minimum": 0.09914325025589708, "maximum": 0.19140281546972893, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.5784350118818535, "standard_deviation": 0.009518055512161882, "minimum": 4.568118324398825, "maximum": 4.592543659521618, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3714547609255111, "standard_deviation": 0.016587075982404647, "minimum": 0.3549970842223683, "maximum": 0.40150244781196387, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06327981089555136, "standard_deviation": 0.017217634143341543, "minimum": -0.09147534379069208, "maximum": -0.0370808455893493, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9969261910083543, "standard_deviation": 0.0005451145653658029, "minimum": 0.9958229362775136, "maximum": 0.9974816077436361, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5685453142504113, "standard_deviation": 0.019974904477368383, "minimum": 0.5353601588801947, "maximum": 0.5882353903725034, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37633413067127935, "standard_deviation": 0.1871030171793013, "minimum": 0.20640846783290356, "maximum": 0.80910295767496, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04672256350105565, "standard_deviation": 0.003981098183757649, "minimum": 0.042200780418296134, "maximum": 0.05437505459796354, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990975202385703, "standard_deviation": 0.00027022008332524666, "minimum": 0.9986042455814759, "maximum": 0.9993834111126254, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3217329508880827, "standard_deviation": 0.02171052272753632, "minimum": 0.2906961460727043, "maximum": 0.3472449385117774, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5238740806068733, "standard_deviation": 0.2766175890261096, "minimum": 0.2679968928309277, "maximum": 1.044385607409856, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.024689273871632722, "standard_deviation": 0.003661081874819355, "minimum": 0.020155833295096504, "maximum": 0.03036259679805499, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1837-0023", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.0902349948883057, "predicted_mos": 3.0902349948883057, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4bc9c9f438a30ab72a84289b1465d3d0f70295890089be087c08a557524af3cf", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.902030699886382, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5711673566512205, "全帯域_10ms_rms_envelope_nrmse": 0.04553700346591939, "全帯域_10ms_envelope_correlation": 0.9970674739830953, "全帯域_10ms_modulation_mae_db": 0.4148078681702813, "全帯域_40ms_log_rms_mae_db": 0.33473623111826095, "全帯域_40ms_rms_envelope_nrmse": 0.026905055973164744, "全帯域_40ms_envelope_correlation": 0.9988873616656, "全帯域_40ms_modulation_mae_db": 0.4896208230820787, "300–1000 Hz_10ms_log_rms_mae_db": 1.382880986947271, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07211455838388042, "300–1000 Hz_10ms_envelope_correlation": 0.9951531496914117, "300–1000 Hz_10ms_modulation_mae_db": 1.0178593888065721, "300–1000 Hz_40ms_log_rms_mae_db": 0.8304967435385329, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04428013479257596, "300–1000 Hz_40ms_envelope_correlation": 0.998081802363391, "300–1000 Hz_40ms_modulation_mae_db": 1.1327401205983678, "1000–3000 Hz_10ms_log_rms_mae_db": 1.666685048619954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1212422640281838, "1000–3000 Hz_10ms_envelope_correlation": 0.9886603853502238, "1000–3000 Hz_10ms_modulation_mae_db": 0.2736659224660769, "1000–3000 Hz_40ms_log_rms_mae_db": 1.005343531394143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06557190368542333, "1000–3000 Hz_40ms_envelope_correlation": 0.9967219127758945, "1000–3000 Hz_40ms_modulation_mae_db": 0.1906320321259489, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9161440658497129, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19869410580972496, "3000–6000 Hz_10ms_envelope_correlation": 0.9721234634862252, "3000–6000 Hz_10ms_modulation_mae_db": 0.597293602430239, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1993210611158975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10839473649280405, "3000–6000 Hz_40ms_envelope_correlation": 0.9913845668999062, "3000–6000 Hz_40ms_modulation_mae_db": 0.66145176325383, "mr_stft_db": 5.018950159439767, "band_mae_db": 0.07512873359146527, "rms_gain_error_db": -0.04481283421934455, "output_peak": 0.34986560603389455}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5625136181206559, "全帯域_10ms_rms_envelope_nrmse": 0.05030943554468061, "全帯域_10ms_envelope_correlation": 0.9964294328618034, "全帯域_10ms_modulation_mae_db": 0.4869493964748793, "全帯域_40ms_log_rms_mae_db": 0.3299340009310129, "全帯域_40ms_rms_envelope_nrmse": 0.026764071366683892, "全帯域_40ms_envelope_correlation": 0.9989289269851872, "全帯域_40ms_modulation_mae_db": 0.6471554900480784, "300–1000 Hz_10ms_log_rms_mae_db": 1.3154576811489191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0640839053769298, "300–1000 Hz_10ms_envelope_correlation": 0.9961862396369738, "300–1000 Hz_10ms_modulation_mae_db": 0.2662342510601387, "300–1000 Hz_40ms_log_rms_mae_db": 0.840191760050172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036745264548565124, "300–1000 Hz_40ms_envelope_correlation": 0.9987215242153988, "300–1000 Hz_40ms_modulation_mae_db": 0.36262328004164207, "1000–3000 Hz_10ms_log_rms_mae_db": 1.678672857019744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11774172249314953, "1000–3000 Hz_10ms_envelope_correlation": 0.9893473003979648, "1000–3000 Hz_10ms_modulation_mae_db": 0.23027477731546053, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0236176215200092, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06486372650837283, "1000–3000 Hz_40ms_envelope_correlation": 0.9968106706483912, "1000–3000 Hz_40ms_modulation_mae_db": 0.32777443086602026, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8878551128053458, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20494076108467005, "3000–6000 Hz_10ms_envelope_correlation": 0.9701142274961383, "3000–6000 Hz_10ms_modulation_mae_db": 0.7070884603577801, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1924939386266522, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1103887526165347, "3000–6000 Hz_40ms_envelope_correlation": 0.9909771725238538, "3000–6000 Hz_40ms_modulation_mae_db": 0.8921779236810711, "mr_stft_db": 5.010495769003286, "band_mae_db": 0.08467281159124122, "rms_gain_error_db": -0.0598463214586173, "output_peak": 0.3536882942525692}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.539920760462164, "全帯域_10ms_rms_envelope_nrmse": 0.04334715693535961, "全帯域_10ms_envelope_correlation": 0.9973670673722094, "全帯域_10ms_modulation_mae_db": 0.3530312020493718, "全帯域_40ms_log_rms_mae_db": 0.2913732136842774, "全帯域_40ms_rms_envelope_nrmse": 0.02234798178452075, "全帯域_40ms_envelope_correlation": 0.9992712217226367, "全帯域_40ms_modulation_mae_db": 0.5648147555776105, "300–1000 Hz_10ms_log_rms_mae_db": 1.326263888398356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06854628117827326, "300–1000 Hz_10ms_envelope_correlation": 0.9956074412503171, "300–1000 Hz_10ms_modulation_mae_db": 0.5489659422457639, "300–1000 Hz_40ms_log_rms_mae_db": 0.862237258075899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0380846283342106, "300–1000 Hz_40ms_envelope_correlation": 0.9986143510858253, "300–1000 Hz_40ms_modulation_mae_db": 0.7679604335334419, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6625148661252611, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11386731481378366, "1000–3000 Hz_10ms_envelope_correlation": 0.9900342682746149, "1000–3000 Hz_10ms_modulation_mae_db": 0.6068659926608987, "1000–3000 Hz_40ms_log_rms_mae_db": 1.034526559375309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0628529351252808, "1000–3000 Hz_40ms_envelope_correlation": 0.9970208758924236, "1000–3000 Hz_40ms_modulation_mae_db": 0.6710621648814828, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9119882571132212, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2018411150945407, "3000–6000 Hz_10ms_envelope_correlation": 0.9717189845478577, "3000–6000 Hz_10ms_modulation_mae_db": 1.0762822740427518, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2155430804609952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11264447373644719, "3000–6000 Hz_40ms_envelope_correlation": 0.9909126579264081, "3000–6000 Hz_40ms_modulation_mae_db": 1.3779978249638567, "mr_stft_db": 5.01716787770437, "band_mae_db": 0.09655973409915268, "rms_gain_error_db": -0.05249032095687911, "output_peak": 0.3566084872645998}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5635968571376007, "全帯域_10ms_rms_envelope_nrmse": 0.04654696989826734, "全帯域_10ms_envelope_correlation": 0.9970288747316637, "全帯域_10ms_modulation_mae_db": 0.3059936630506595, "全帯域_40ms_log_rms_mae_db": 0.32724348311409845, "全帯域_40ms_rms_envelope_nrmse": 0.025736575875154686, "全帯域_40ms_envelope_correlation": 0.9991078080696084, "全帯域_40ms_modulation_mae_db": 0.3634400047375182, "300–1000 Hz_10ms_log_rms_mae_db": 1.4063023074638286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07721156611829386, "300–1000 Hz_10ms_envelope_correlation": 0.9945644946114837, "300–1000 Hz_10ms_modulation_mae_db": 0.801806597978108, "300–1000 Hz_40ms_log_rms_mae_db": 0.9378996260272524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0450629875395829, "300–1000 Hz_40ms_envelope_correlation": 0.9982171263054833, "300–1000 Hz_40ms_modulation_mae_db": 1.016700315543784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.666253700907799, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10431562943653122, "1000–3000 Hz_10ms_envelope_correlation": 0.9915873572457874, "1000–3000 Hz_10ms_modulation_mae_db": 0.35803977059469627, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0446309311089292, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05599594755741265, "1000–3000 Hz_40ms_envelope_correlation": 0.9974909276591084, "1000–3000 Hz_40ms_modulation_mae_db": 0.2011112223717647, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8461437672657643, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1955078409867527, "3000–6000 Hz_10ms_envelope_correlation": 0.974850734973894, "3000–6000 Hz_10ms_modulation_mae_db": 0.7885222639441158, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1307354245216927, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10553594576315264, "3000–6000 Hz_40ms_envelope_correlation": 0.9931031103827271, "3000–6000 Hz_40ms_modulation_mae_db": 0.7060739667390378, "mr_stft_db": 5.010910400633196, "band_mae_db": 0.10015014920725762, "rms_gain_error_db": -0.08833794164593037, "output_peak": 0.3614772967413357}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5908608251107828, "全帯域_10ms_rms_envelope_nrmse": 0.054247339889624345, "全帯域_10ms_envelope_correlation": 0.9959085465658685, "全帯域_10ms_modulation_mae_db": 0.6457183386256239, "全帯域_40ms_log_rms_mae_db": 0.3494055543575346, "全帯域_40ms_rms_envelope_nrmse": 0.02785941266672734, "全帯域_40ms_envelope_correlation": 0.9989088921610607, "全帯域_40ms_modulation_mae_db": 0.9238238600998424, "300–1000 Hz_10ms_log_rms_mae_db": 1.414902401467199, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08122470035986801, "300–1000 Hz_10ms_envelope_correlation": 0.9938799480362156, "300–1000 Hz_10ms_modulation_mae_db": 0.5561786412598315, "300–1000 Hz_40ms_log_rms_mae_db": 0.9463082335451815, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04531872374330784, "300–1000 Hz_40ms_envelope_correlation": 0.9981251731271498, "300–1000 Hz_40ms_modulation_mae_db": 0.857628403709875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6958329268961432, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11313087457484332, "1000–3000 Hz_10ms_envelope_correlation": 0.9901778384433292, "1000–3000 Hz_10ms_modulation_mae_db": 0.2377247531175547, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0636510856618446, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06384352510665149, "1000–3000 Hz_40ms_envelope_correlation": 0.996891602646981, "1000–3000 Hz_40ms_modulation_mae_db": 0.29296679297356876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9610238642455369, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20104273152174884, "3000–6000 Hz_10ms_envelope_correlation": 0.9711167206621776, "3000–6000 Hz_10ms_modulation_mae_db": 1.1941168628498091, "3000–6000 Hz_40ms_log_rms_mae_db": 1.245223703269431, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10612893640676294, "3000–6000 Hz_40ms_envelope_correlation": 0.9917357633517614, "3000–6000 Hz_40ms_modulation_mae_db": 1.2904688712204462, "mr_stft_db": 5.020194200366714, "band_mae_db": 0.10467503515852235, "rms_gain_error_db": -0.08656993223370874, "output_peak": 0.3503407799085662}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5718363894843815, "全帯域_10ms_rms_envelope_nrmse": 0.05064167202016084, "全帯域_10ms_envelope_correlation": 0.9964527856376811, "全帯域_10ms_modulation_mae_db": 0.15369115555053373, "全帯域_40ms_log_rms_mae_db": 0.3303725329572606, "全帯域_40ms_rms_envelope_nrmse": 0.029570742987657626, "全帯域_40ms_envelope_correlation": 0.9987549802008049, "全帯域_40ms_modulation_mae_db": 0.07258965680831996, "300–1000 Hz_10ms_log_rms_mae_db": 1.3770793593776767, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07805114885538411, "300–1000 Hz_10ms_envelope_correlation": 0.9944836402123445, "300–1000 Hz_10ms_modulation_mae_db": 1.0967754427893712, "300–1000 Hz_40ms_log_rms_mae_db": 0.9116620175200512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04518729830826218, "300–1000 Hz_40ms_envelope_correlation": 0.9982246651735145, "300–1000 Hz_40ms_modulation_mae_db": 1.1485997168973026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6854834131619196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11535459347884028, "1000–3000 Hz_10ms_envelope_correlation": 0.9897123649729598, "1000–3000 Hz_10ms_modulation_mae_db": 0.5049389584801163, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0298807536888368, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06723582289859645, "1000–3000 Hz_40ms_envelope_correlation": 0.9964124778285025, "1000–3000 Hz_40ms_modulation_mae_db": 0.45173915115869345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9117474302904878, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21399416725840853, "3000–6000 Hz_10ms_envelope_correlation": 0.970896465431158, "3000–6000 Hz_10ms_modulation_mae_db": 0.7947572836618649, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2250452410245365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13702179100722597, "3000–6000 Hz_40ms_envelope_correlation": 0.988703468637437, "3000–6000 Hz_40ms_modulation_mae_db": 0.5522291524509556, "mr_stft_db": 5.019933910008266, "band_mae_db": 0.11318694938754632, "rms_gain_error_db": -0.08664680185303805, "output_peak": 0.33979565133834455}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5667783919292614, "全帯域_10ms_rms_envelope_nrmse": 0.04697848626877067, "全帯域_10ms_envelope_correlation": 0.997052319311574, "全帯域_10ms_modulation_mae_db": 0.7128554780525187, "全帯域_40ms_log_rms_mae_db": 0.3483623205315292, "全帯域_40ms_rms_envelope_nrmse": 0.026445952154305247, "全帯域_40ms_envelope_correlation": 0.9991281893087195, "全帯域_40ms_modulation_mae_db": 0.9038910691900268, "300–1000 Hz_10ms_log_rms_mae_db": 1.4683228169287792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0741232715313289, "300–1000 Hz_10ms_envelope_correlation": 0.9949837690363026, "300–1000 Hz_10ms_modulation_mae_db": 0.9322951255024338, "300–1000 Hz_40ms_log_rms_mae_db": 0.9675704335160691, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04273149029647591, "300–1000 Hz_40ms_envelope_correlation": 0.9983892947227424, "300–1000 Hz_40ms_modulation_mae_db": 1.017275558150711, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6685903522592214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10267168509635444, "1000–3000 Hz_10ms_envelope_correlation": 0.9919224241874334, "1000–3000 Hz_10ms_modulation_mae_db": 0.18763973726076635, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0012238781159937, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05791099475834458, "1000–3000 Hz_40ms_envelope_correlation": 0.9974080231198442, "1000–3000 Hz_40ms_modulation_mae_db": 0.05778366533509335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9617639155137963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19108433878178524, "3000–6000 Hz_10ms_envelope_correlation": 0.9740174667825146, "3000–6000 Hz_10ms_modulation_mae_db": 0.9284655419231115, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2030425987298354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0944725165437812, "3000–6000 Hz_40ms_envelope_correlation": 0.9934469322417373, "3000–6000 Hz_40ms_modulation_mae_db": 1.4569532350179484, "mr_stft_db": 5.009211071103437, "band_mae_db": 0.1331759390865323, "rms_gain_error_db": -0.11084574981681784, "output_peak": 0.3442568087435121}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5613423364125573, "全帯域_10ms_rms_envelope_nrmse": 0.04692187569470202, "全帯域_10ms_envelope_correlation": 0.9969002309761736, "全帯域_10ms_modulation_mae_db": 0.24348540674844052, "全帯域_40ms_log_rms_mae_db": 0.3151747894272741, "全帯域_40ms_rms_envelope_nrmse": 0.02360251655330482, "全帯域_40ms_envelope_correlation": 0.9991746352113116, "全帯域_40ms_modulation_mae_db": 0.5077899584384816, "300–1000 Hz_10ms_log_rms_mae_db": 1.332664153801962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06827895383334208, "300–1000 Hz_10ms_envelope_correlation": 0.9956841971838293, "300–1000 Hz_10ms_modulation_mae_db": 0.2874412683953715, "300–1000 Hz_40ms_log_rms_mae_db": 0.8728359440278513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03826400976974097, "300–1000 Hz_40ms_envelope_correlation": 0.9986466794026216, "300–1000 Hz_40ms_modulation_mae_db": 0.4305437682421531, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6422928301432937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11392964504332162, "1000–3000 Hz_10ms_envelope_correlation": 0.9899951490738704, "1000–3000 Hz_10ms_modulation_mae_db": 0.4944450761525599, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9953163172759211, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06400939532586027, "1000–3000 Hz_40ms_envelope_correlation": 0.9967773829374813, "1000–3000 Hz_40ms_modulation_mae_db": 0.5127321713569285, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8803159413566688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1950717990819772, "3000–6000 Hz_10ms_envelope_correlation": 0.9730224607137575, "3000–6000 Hz_10ms_modulation_mae_db": 0.70232169331135, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2251439964951218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10931251385980195, "3000–6000 Hz_40ms_envelope_correlation": 0.9912035181471494, "3000–6000 Hz_40ms_modulation_mae_db": 0.8913889567642866, "mr_stft_db": 5.014826939241787, "band_mae_db": 0.05317178628307334, "rms_gain_error_db": -0.051918672955076316, "output_peak": 0.3674396605301028}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.964081523031311, "spectral_convergence": 0.125446933494274, "spectral_convergence_status": "ok", "amplitude_floor": 1.5362649016028353e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.955150217994633, "spectral_convergence": 0.1233851668909588, "spectral_convergence_status": "ok", "amplitude_floor": 1.0749003550909146e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.137618737293358, "spectral_convergence": 0.11836746517082965, "spectral_convergence_status": "ok", "amplitude_floor": 8.625899017329306e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001501246216341323, "output_power": 0.001506531970546504, "delta_db": 0.015264265544410356, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002746582421816528, "output_power": 0.002708077059324224, "delta_db": -0.06131618039257669, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011276426514964012, "output_power": 0.0011078907269448572, "delta_db": -0.07674566750624327, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016504304929751857, "output_power": 0.00015916914093565642, "delta_db": -0.1573836664590833, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.934184208485853e-05, "output_power": 4.0570098243815026e-05, "delta_db": 0.13351368556386223, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8643647097645308e-05, "output_power": 2.8686872818327026e-05, "delta_db": 0.006548936082615763, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.07512873359146527, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04481283421934455, "pitch": null, "output_rms_dbfs": -22.55634572520465, "output_peak": 0.34986560603389455, "reference_rms_dbfs": -22.511532890985304, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.018950159439767}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018834661138467316, "log_rms_mae_db": 0.5711673566512205, "rms_envelope_nrmse": 0.04553700346591939, "envelope_correlation": 0.9970674739830953, "rms_gain_error_db": -0.042496572026536124, "reference_mean_power": 0.005606081823103195, "output_mean_power": 0.00555149272436978}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7511238572638392e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009260227506744139, "output_power": 0.0009110234167365209, "reference_active": true, "delta_db": -0.07092116506986856}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010953814679795108, "output_power": 9.683754727392489e-05, "reference_active": true, "delta_db": -0.5352160813172058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.583845526426034e-05, "output_power": 2.2306832708170138e-05, "reference_active": true, "delta_db": -0.6382863581237695}], "active_band_count": 3, "active_band_mae_db": 0.4148078681702813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015988223985812614, "log_rms_mae_db": 0.33473623111826095, "rms_envelope_nrmse": 0.026905055973164744, "envelope_correlation": 0.9988873616656, "rms_gain_error_db": -0.04005775301267572, "reference_mean_power": 0.00558793457857638, "output_mean_power": 0.005536630459144159}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5615527006220852e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008311543470725633, "output_power": 0.0008195421621237732, "reference_active": true, "delta_db": -0.06110379479618396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.497505864184299e-05, "output_power": 4.7558787969647486e-05, "reference_active": true, "delta_db": -0.6293492315283866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0542332226205543e-06, "output_power": 1.71715225446761e-06, "reference_active": true, "delta_db": -0.7784094429216657}], "active_band_count": 3, "active_band_mae_db": 0.4896208230820787}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016200114302549476, "log_rms_mae_db": 1.382880986947271, "rms_envelope_nrmse": 0.07211455838388042, "envelope_correlation": 0.9951531496914117, "rms_gain_error_db": -0.10012436858737866, "reference_mean_power": 0.0020825301889685474, "output_mean_power": 0.0020350677355065448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.598844168109503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038009245216565237, "output_power": 0.00036906975610841543, "reference_active": true, "delta_db": -0.1278078767751806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.129961338774397e-05, "output_power": 4.963361866065304e-05, "reference_active": true, "delta_db": -0.9168179580739901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4399990501668035e-05, "output_power": 9.067070080840507e-06, "reference_active": true, "delta_db": -2.0089523315705455}], "active_band_count": 3, "active_band_mae_db": 1.0178593888065721}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012179662863477574, "log_rms_mae_db": 0.8304967435385329, "rms_envelope_nrmse": 0.04428013479257596, "envelope_correlation": 0.998081802363391, "rms_gain_error_db": -0.09937043436892815, "reference_mean_power": 0.0020746475669658963, "output_mean_power": 0.002027716745085687}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.889619852823582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035505972043036534, "output_power": 0.0003437922841539227, "reference_active": true, "delta_db": -0.14005281336097558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2260222284985983e-05, "output_power": 2.623112203380323e-05, "reference_active": true, "delta_db": -0.8985048762079824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1886983564224358e-06, "output_power": 6.904073322195385e-07, "reference_active": true, "delta_db": -2.3596626722261456}], "active_band_count": 3, "active_band_mae_db": 1.1327401205983678}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.2267966954040695e-05, "log_rms_mae_db": 1.666685048619954, "rms_envelope_nrmse": 0.1212422640281838, "envelope_correlation": 0.9886603853502238, "rms_gain_error_db": -0.11426423442778835, "reference_mean_power": 0.00015543877011727877, "output_mean_power": 0.00015140245867759124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4080839081981722e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.101806003601554e-05, "output_power": 6.658710902736593e-05, "reference_active": true, "delta_db": -0.27978644929198326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0132703192354966e-06, "output_power": 1.9016956064959375e-06, "reference_active": true, "delta_db": -0.2476108783095132}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.589135119213016e-07, "output_power": 8.119930820335469e-07, "reference_active": true, "delta_db": 0.29360043979673417}], "active_band_count": 3, "active_band_mae_db": 0.2736659224660769}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.184196106800431e-05, "log_rms_mae_db": 1.005343531394143, "rms_envelope_nrmse": 0.06557190368542333, "envelope_correlation": 0.9967219127758945, "rms_gain_error_db": -0.12322818550276349, "reference_mean_power": 0.00015383740434286893, "output_mean_power": 0.00014953371568702622}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3572636139953185e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.549215276795424e-05, "output_power": 6.165405468607199e-05, "reference_active": true, "delta_db": -0.262276228961783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.98253690709034e-07, "output_power": 1.0118080286311127e-06, "reference_active": true, "delta_db": 0.058571966568923536}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.830181704119583e-08, "output_power": 4.558885434746471e-08, "reference_active": true, "delta_db": -0.2510479008471401}], "active_band_count": 3, "active_band_mae_db": 0.1906320321259489}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.781014610300033e-05, "log_rms_mae_db": 1.9161440658497129, "rms_envelope_nrmse": 0.19869410580972496, "envelope_correlation": 0.9721234634862252, "rms_gain_error_db": 0.06795869509236319, "reference_mean_power": 2.2945322907755735e-05, "output_mean_power": 2.3307196806491516e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.468236349213924e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8725122830509875e-06, "output_power": 3.763769864730625e-06, "reference_active": true, "delta_db": -0.1236974327135537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.737803238621202e-07, "output_power": 4.2189452221818586e-07, "reference_active": true, "delta_db": -1.3354176465282177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.433384337397866e-07, "output_power": 1.547530536405753e-07, "reference_active": true, "delta_db": 0.33276572804894555}], "active_band_count": 3, "active_band_mae_db": 0.597293602430239}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.4952988731037596e-05, "log_rms_mae_db": 1.1993210611158975, "rms_envelope_nrmse": 0.10839473649280405, "envelope_correlation": 0.9913845668999062, "rms_gain_error_db": 0.06826334293739084, "reference_mean_power": 2.2958960928334722e-05, "output_mean_power": 2.3322685890962067e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.967813463196935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5837220182078332e-06, "output_power": 3.4777605994944397e-06, "reference_active": true, "delta_db": -0.13034632056272222}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.664756741318262e-07, "output_power": 1.8948805996316118e-07, "reference_active": true, "delta_db": -1.4807572019581126}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1431200548978765e-08, "output_power": 1.0489786942614489e-08, "reference_active": true, "delta_db": -0.37325176724065545}], "active_band_count": 3, "active_band_mae_db": 0.66145176325383}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9441209289670282, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5fc6adbf3702c12b7ba550c46b8e3200ee260b1fab36700aea3ad2838a89b708", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.990179635993273, "standard_deviation": 0.00108847255420328, "minimum": 0.9886603853502238, "maximum": 0.9919224241874334, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.670790749391667, "standard_deviation": 0.016159993034520446, "minimum": 1.6422928301432937, "maximum": 1.6958329268961432, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3616993735060162, "standard_deviation": 0.15535573950137085, "minimum": 0.18763973726076635, "maximum": 0.6068659926608987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11278171612062599, "standard_deviation": 0.006316732428655602, "minimum": 0.10267168509635444, "maximum": 0.1212422640281838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969417341885783, "standard_deviation": 0.0003586558536900507, "minimum": 0.9964124778285025, "maximum": 0.9974909276591084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0247738347676232, "standard_deviation": 0.023385467048906254, "minimum": 0.9953163172759211, "maximum": 1.0636510856618446, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3382252038836876, "standard_deviation": 0.19838905779256338, "minimum": 0.05778366533509335, "maximum": 0.6710621648814828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0627855313707428, "standard_deviation": 0.0038613312457308644, "minimum": 0.05599594755741265, "maximum": 0.06723582289859645, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9722325655117153, "standard_deviation": 0.0016238365535951188, "minimum": 0.9701142274961383, "maximum": 0.974850734973894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9096227943050668, "standard_deviation": 0.03923503071841293, "minimum": 1.8461437672657643, "maximum": 1.9617639155137963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8486059978151277, "standard_deviation": 0.2030783668991938, "minimum": 0.597293602430239, "maximum": 1.1941168628498091, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20027210745245103, "standard_deviation": 0.0070689817707597444, "minimum": 0.19108433878178524, "maximum": 0.21399416725840853, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9914333987638726, "standard_deviation": 0.0014599774874300634, "minimum": 0.988703468637437, "maximum": 0.9934469322417373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2045686305305203, "standard_deviation": 0.0343155878876326, "minimum": 1.1307354245216927, "maximum": 1.245223703269431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.978592711761429, "standard_deviation": 0.34996381778340424, "minimum": 0.5522291524509556, "maximum": 1.4569532350179484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11048745830331383, "standard_deviation": 0.012038544042094316, "minimum": 0.0944725165437812, "maximum": 0.13702179100722597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9950678599573598, "standard_deviation": 0.0007504248693949793, "minimum": 0.9938799480362156, "maximum": 0.9961862396369738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.377984199441749, "standard_deviation": 0.05208399062594837, "minimum": 1.3154576811489191, "maximum": 1.4683228169287792, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6884445822546988, "standard_deviation": 0.3215089494758086, "minimum": 0.2662342510601387, "maximum": 1.0967754427893712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07295429820466257, "standard_deviation": 0.005788495462945143, "minimum": 0.0640839053769298, "maximum": 0.08122470035986801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983775770495158, "standard_deviation": 0.0002527652399144437, "minimum": 0.998081802363391, "maximum": 0.9987215242153988, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8961502520376261, "standard_deviation": 0.0517351423107663, "minimum": 0.8304967435385329, "maximum": 0.9675704335160691, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8417589495896596, "standard_deviation": 0.3034560065823318, "minimum": 0.36262328004164207, "maximum": 1.1485997168973026, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04195931716659018, "standard_deviation": 0.003648395839014926, "minimum": 0.036745264548565124, "maximum": 0.04531872374330784, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.0950901423005989, "standard_deviation": 0.024383199932337816, "minimum": 0.05317178628307334, "maximum": 0.1331759390865323, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.015211290937603, "standard_deviation": 0.004501175331044813, "minimum": 5.009211071103437, "maximum": 5.020194200366714, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3529340731016156, "standard_deviation": 0.008956423143598083, "minimum": 0.33979565133834455, "maximum": 0.3674396605301028, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07268357189242654, "standard_deviation": 0.023512110372376226, "minimum": -0.11084574981681784, "maximum": -0.04481283421934455, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967758414300085, "standard_deviation": 0.00047315062735027094, "minimum": 0.9959085465658685, "maximum": 0.9973670673722094, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5660020669135781, "standard_deviation": 0.014138003603121134, "minimum": 0.539920760462164, "maximum": 0.5908608251107828, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41456656359028854, "standard_deviation": 0.19292713965756755, "minimum": 0.15369115555053373, "maximum": 0.7128554780525187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.048066242464685605, "standard_deviation": 0.0034504350468761024, "minimum": 0.04334715693535961, "maximum": 0.054247339889624345, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990202519156162, "standard_deviation": 0.00017521690477737353, "minimum": 0.9987549802008049, "maximum": 0.9992712217226367, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.328325265765156, "standard_deviation": 0.018644422275853282, "minimum": 0.2913732136842774, "maximum": 0.3494055543575346, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5591407022477446, "standard_deviation": 0.2782922446929547, "minimum": 0.07258965680831996, "maximum": 0.9238238600998424, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.026154038670189887, "standard_deviation": 0.0022920408416511467, "minimum": 0.02234798178452075, "maximum": 0.029570742987657626, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1837-0023", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8621914386749268, "predicted_mos": 2.8621914386749268, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0193fda4700230558f70e1e5237d1110ddee24ea3aefd36496c4a746cf0975d7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.078287799842656, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5767526868035837, "全帯域_10ms_rms_envelope_nrmse": 0.04881172396111846, "全帯域_10ms_envelope_correlation": 0.9966565069257467, "全帯域_10ms_modulation_mae_db": 0.5544905558410402, "全帯域_40ms_log_rms_mae_db": 0.323307322916064, "全帯域_40ms_rms_envelope_nrmse": 0.026444323665010743, "全帯域_40ms_envelope_correlation": 0.9989733683611318, "全帯域_40ms_modulation_mae_db": 0.7233041700457323, "300–1000 Hz_10ms_log_rms_mae_db": 1.185838311017948, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0751275327021665, "300–1000 Hz_10ms_envelope_correlation": 0.9946158685509372, "300–1000 Hz_10ms_modulation_mae_db": 0.3111598605217633, "300–1000 Hz_40ms_log_rms_mae_db": 0.7346048400958017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0408653019630715, "300–1000 Hz_40ms_envelope_correlation": 0.9983216912059474, "300–1000 Hz_40ms_modulation_mae_db": 0.6156065059503103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2137646949991532, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09134586822090385, "1000–3000 Hz_10ms_envelope_correlation": 0.9937870916762254, "1000–3000 Hz_10ms_modulation_mae_db": 0.3466367881133414, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7149548881140315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05238943607379713, "1000–3000 Hz_40ms_envelope_correlation": 0.9981494129684777, "1000–3000 Hz_40ms_modulation_mae_db": 0.1773241189974334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2010058094615943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13576063730308724, "3000–6000 Hz_10ms_envelope_correlation": 0.9869292517410594, "3000–6000 Hz_10ms_modulation_mae_db": 0.9267591125050556, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7080932353732757, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06510323121318923, "3000–6000 Hz_40ms_envelope_correlation": 0.996924789968233, "3000–6000 Hz_40ms_modulation_mae_db": 0.9822765959453488, "mr_stft_db": 4.557053256751828, "band_mae_db": 0.10571159229420894, "rms_gain_error_db": -0.06222378505682884, "output_peak": 0.3689228202561212}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5753844440737761, "全帯域_10ms_rms_envelope_nrmse": 0.04921403546378484, "全帯域_10ms_envelope_correlation": 0.996618762237953, "全帯域_10ms_modulation_mae_db": 0.42974839301172124, "全帯域_40ms_log_rms_mae_db": 0.3427526500047615, "全帯域_40ms_rms_envelope_nrmse": 0.02968045624082671, "全帯域_40ms_envelope_correlation": 0.9987039402843321, "全帯域_40ms_modulation_mae_db": 0.6250795946153819, "300–1000 Hz_10ms_log_rms_mae_db": 1.163243577186688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06838375854772999, "300–1000 Hz_10ms_envelope_correlation": 0.9956441407087477, "300–1000 Hz_10ms_modulation_mae_db": 0.6770640515154023, "300–1000 Hz_40ms_log_rms_mae_db": 0.737033746247797, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04200026694845431, "300–1000 Hz_40ms_envelope_correlation": 0.9983109823401939, "300–1000 Hz_40ms_modulation_mae_db": 0.8791416334694094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.226949339398645, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09983638869144384, "1000–3000 Hz_10ms_envelope_correlation": 0.9924884167161491, "1000–3000 Hz_10ms_modulation_mae_db": 0.8086446640057409, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7558296142185977, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059636518692668415, "1000–3000 Hz_40ms_envelope_correlation": 0.9974323031036837, "1000–3000 Hz_40ms_modulation_mae_db": 0.9069124662046392, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2078500734190327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15803097590459753, "3000–6000 Hz_10ms_envelope_correlation": 0.982976522480321, "3000–6000 Hz_10ms_modulation_mae_db": 0.7036176227671614, "3000–6000 Hz_40ms_log_rms_mae_db": 0.717220283050853, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09115748398850225, "3000–6000 Hz_40ms_envelope_correlation": 0.994386431675525, "3000–6000 Hz_40ms_modulation_mae_db": 1.015690512581631, "mr_stft_db": 4.563723095338505, "band_mae_db": 0.13643551188466044, "rms_gain_error_db": -0.0732219478049241, "output_peak": 0.3625087706645045}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5574291536244396, "全帯域_10ms_rms_envelope_nrmse": 0.0424101102041193, "全帯域_10ms_envelope_correlation": 0.9974854041621873, "全帯域_10ms_modulation_mae_db": 0.7336880975597414, "全帯域_40ms_log_rms_mae_db": 0.31242579714829166, "全帯域_40ms_rms_envelope_nrmse": 0.02049114338332439, "全帯域_40ms_envelope_correlation": 0.9994040213964904, "全帯域_40ms_modulation_mae_db": 0.8790680835521246, "300–1000 Hz_10ms_log_rms_mae_db": 1.188663512138552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06815145830110268, "300–1000 Hz_10ms_envelope_correlation": 0.9955774729807447, "300–1000 Hz_10ms_modulation_mae_db": 1.0754045697632115, "300–1000 Hz_40ms_log_rms_mae_db": 0.7230490671452446, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035198747266940715, "300–1000 Hz_40ms_envelope_correlation": 0.9987424629250405, "300–1000 Hz_40ms_modulation_mae_db": 1.1398997038954592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1813720110886974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08788066435351463, "1000–3000 Hz_10ms_envelope_correlation": 0.994222435726108, "1000–3000 Hz_10ms_modulation_mae_db": 0.3387629399779119, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7282096032767547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05302591344926493, "1000–3000 Hz_40ms_envelope_correlation": 0.9979899984692716, "1000–3000 Hz_40ms_modulation_mae_db": 0.4220404063935816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2050724554418741, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1432870200864643, "3000–6000 Hz_10ms_envelope_correlation": 0.9854127428735185, "3000–6000 Hz_10ms_modulation_mae_db": 0.593340847844965, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7424744161272141, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07987745798567006, "3000–6000 Hz_40ms_envelope_correlation": 0.9955032134092054, "3000–6000 Hz_40ms_modulation_mae_db": 0.6893817469575048, "mr_stft_db": 4.557094068901276, "band_mae_db": 0.12109391997865386, "rms_gain_error_db": -0.05608249558622473, "output_peak": 0.37391005108620123}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5686730884738429, "全帯域_10ms_rms_envelope_nrmse": 0.04395518192984064, "全帯域_10ms_envelope_correlation": 0.9973373163367175, "全帯域_10ms_modulation_mae_db": 0.46236870187481166, "全帯域_40ms_log_rms_mae_db": 0.3285240811855264, "全帯域_40ms_rms_envelope_nrmse": 0.02410369523544293, "全帯域_40ms_envelope_correlation": 0.9991953833832564, "全帯域_40ms_modulation_mae_db": 0.7306692913180216, "300–1000 Hz_10ms_log_rms_mae_db": 1.120644531421786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06803960093288457, "300–1000 Hz_10ms_envelope_correlation": 0.9957510759190135, "300–1000 Hz_10ms_modulation_mae_db": 0.8993558923231036, "300–1000 Hz_40ms_log_rms_mae_db": 0.7135219187721419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045551764051466055, "300–1000 Hz_40ms_envelope_correlation": 0.9980495053517273, "300–1000 Hz_40ms_modulation_mae_db": 1.2370970363869336, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2789184867471899, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10285754320233732, "1000–3000 Hz_10ms_envelope_correlation": 0.9922356918079016, "1000–3000 Hz_10ms_modulation_mae_db": 0.28185422467894194, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7902440037097179, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06388582306913043, "1000–3000 Hz_40ms_envelope_correlation": 0.9972331399611803, "1000–3000 Hz_40ms_modulation_mae_db": 0.27255044577006277, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1589840444588864, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13473431303101843, "3000–6000 Hz_10ms_envelope_correlation": 0.9870686509624735, "3000–6000 Hz_10ms_modulation_mae_db": 0.33209172735572196, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6939829977354967, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06905914259207727, "3000–6000 Hz_40ms_envelope_correlation": 0.9965123680086181, "3000–6000 Hz_40ms_modulation_mae_db": 0.4271640456757935, "mr_stft_db": 4.547993315139663, "band_mae_db": 0.16019693183366882, "rms_gain_error_db": -0.07931908780010535, "output_peak": 0.3615094464280527}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5586288585666837, "全帯域_10ms_rms_envelope_nrmse": 0.04410389818595574, "全帯域_10ms_envelope_correlation": 0.9972408230322156, "全帯域_10ms_modulation_mae_db": 0.3560355825615991, "全帯域_40ms_log_rms_mae_db": 0.3156914909185759, "全帯域_40ms_rms_envelope_nrmse": 0.02306775624545023, "全帯域_40ms_envelope_correlation": 0.9991821998316751, "全帯域_40ms_modulation_mae_db": 0.588251859768912, "300–1000 Hz_10ms_log_rms_mae_db": 1.2766871937423945, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08326860014046217, "300–1000 Hz_10ms_envelope_correlation": 0.9933657047996693, "300–1000 Hz_10ms_modulation_mae_db": 0.06648105562645039, "300–1000 Hz_40ms_log_rms_mae_db": 0.8283604211902039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049533180808854764, "300–1000 Hz_40ms_envelope_correlation": 0.9974867621101109, "300–1000 Hz_40ms_modulation_mae_db": 0.19004834956796546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1979899118108068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08580641315686607, "1000–3000 Hz_10ms_envelope_correlation": 0.9947503157981223, "1000–3000 Hz_10ms_modulation_mae_db": 0.679589754044676, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7263388746747075, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04905122888578482, "1000–3000 Hz_40ms_envelope_correlation": 0.9986302827300213, "1000–3000 Hz_40ms_modulation_mae_db": 0.8227023942217092, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1761164629485472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12718990256444032, "3000–6000 Hz_10ms_envelope_correlation": 0.9884779513363806, "3000–6000 Hz_10ms_modulation_mae_db": 0.8302358944967726, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6745588735540521, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06587435792685323, "3000–6000 Hz_40ms_envelope_correlation": 0.9968539686551503, "3000–6000 Hz_40ms_modulation_mae_db": 0.9364213529876496, "mr_stft_db": 4.559469902765559, "band_mae_db": 0.12538831639715808, "rms_gain_error_db": -0.03854656980241572, "output_peak": 0.3646175943917074}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.545491593072728, "全帯域_10ms_rms_envelope_nrmse": 0.04233704971397833, "全帯域_10ms_envelope_correlation": 0.997477545998292, "全帯域_10ms_modulation_mae_db": 0.4406111033818025, "全帯域_40ms_log_rms_mae_db": 0.2941604457607665, "全帯域_40ms_rms_envelope_nrmse": 0.020492273799648745, "全帯域_40ms_envelope_correlation": 0.9993827201044713, "全帯域_40ms_modulation_mae_db": 0.8047998564061437, "300–1000 Hz_10ms_log_rms_mae_db": 1.1964906212339141, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06299767558884475, "300–1000 Hz_10ms_envelope_correlation": 0.9962195321468527, "300–1000 Hz_10ms_modulation_mae_db": 0.4324045113474431, "300–1000 Hz_40ms_log_rms_mae_db": 0.7064206790528414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03318263994628193, "300–1000 Hz_40ms_envelope_correlation": 0.9988772255990312, "300–1000 Hz_40ms_modulation_mae_db": 0.7092773364784764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2201402689775729, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08810409588002084, "1000–3000 Hz_10ms_envelope_correlation": 0.9941691957105591, "1000–3000 Hz_10ms_modulation_mae_db": 0.49376242307645857, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7594961129623818, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0489036604335967, "1000–3000 Hz_40ms_envelope_correlation": 0.9983557829409574, "1000–3000 Hz_40ms_modulation_mae_db": 0.35239168295465256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1567676836321668, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13891191863621685, "3000–6000 Hz_10ms_envelope_correlation": 0.9862006392008102, "3000–6000 Hz_10ms_modulation_mae_db": 0.2462285941651253, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6787560677231995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06941541267295794, "3000–6000 Hz_40ms_envelope_correlation": 0.9964766988609128, "3000–6000 Hz_40ms_modulation_mae_db": 0.3381483906678187, "mr_stft_db": 4.5588139333873405, "band_mae_db": 0.10414110130827525, "rms_gain_error_db": -0.05057532280954688, "output_peak": 0.36402976787660596}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5606742232983144, "全帯域_10ms_rms_envelope_nrmse": 0.04918048327168533, "全帯域_10ms_envelope_correlation": 0.9965778397552396, "全帯域_10ms_modulation_mae_db": 0.2749106126502802, "全帯域_40ms_log_rms_mae_db": 0.3103653701678965, "全帯域_40ms_rms_envelope_nrmse": 0.026584313583718534, "全帯域_40ms_envelope_correlation": 0.9989308035796196, "全帯域_40ms_modulation_mae_db": 0.29360776557393503, "300–1000 Hz_10ms_log_rms_mae_db": 1.2885846039238087, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07699892628274231, "300–1000 Hz_10ms_envelope_correlation": 0.9944389560442777, "300–1000 Hz_10ms_modulation_mae_db": 0.7517307659149486, "300–1000 Hz_40ms_log_rms_mae_db": 0.8009869104734555, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041743356456694025, "300–1000 Hz_40ms_envelope_correlation": 0.9983080167035414, "300–1000 Hz_40ms_modulation_mae_db": 0.9758140436440165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.161533544343844, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0913373971235612, "1000–3000 Hz_10ms_envelope_correlation": 0.9937508315758756, "1000–3000 Hz_10ms_modulation_mae_db": 0.2891919704491566, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6826778623257012, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05251170327412798, "1000–3000 Hz_40ms_envelope_correlation": 0.998078633313445, "1000–3000 Hz_40ms_modulation_mae_db": 0.2652631517747088, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1625703294548733, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13290745686058597, "3000–6000 Hz_10ms_envelope_correlation": 0.9874140687052874, "3000–6000 Hz_10ms_modulation_mae_db": 0.45174755705959896, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6668843813677352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0726004518858125, "3000–6000 Hz_40ms_envelope_correlation": 0.9962035369321133, "3000–6000 Hz_40ms_modulation_mae_db": 0.1729403583457855, "mr_stft_db": 4.5595961151575155, "band_mae_db": 0.12696231864670662, "rms_gain_error_db": -0.04645614403727034, "output_peak": 0.3595201053512804}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6032997929559526, "全帯域_10ms_rms_envelope_nrmse": 0.04995845559615617, "全帯域_10ms_envelope_correlation": 0.9964438373680962, "全帯域_10ms_modulation_mae_db": 0.14755644086427486, "全帯域_40ms_log_rms_mae_db": 0.3540821506509715, "全帯域_40ms_rms_envelope_nrmse": 0.025775422454279, "全帯域_40ms_envelope_correlation": 0.998962697704443, "全帯域_40ms_modulation_mae_db": 0.3924986761667775, "300–1000 Hz_10ms_log_rms_mae_db": 1.2259727112097123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08082694814543141, "300–1000 Hz_10ms_envelope_correlation": 0.9937372222220526, "300–1000 Hz_10ms_modulation_mae_db": 0.27662921178706973, "300–1000 Hz_40ms_log_rms_mae_db": 0.764023875760491, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043071578454386396, "300–1000 Hz_40ms_envelope_correlation": 0.998081208502134, "300–1000 Hz_40ms_modulation_mae_db": 0.4903508078214933, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1991413293911908, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09261840993587954, "1000–3000 Hz_10ms_envelope_correlation": 0.9936794796159581, "1000–3000 Hz_10ms_modulation_mae_db": 0.6631208215752252, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7266504528287682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050668264974694927, "1000–3000 Hz_40ms_envelope_correlation": 0.9983764050648544, "1000–3000 Hz_40ms_modulation_mae_db": 0.4847322772134488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1884864989861463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1341415484265004, "3000–6000 Hz_10ms_envelope_correlation": 0.9873592089309939, "3000–6000 Hz_10ms_modulation_mae_db": 0.2778308906825994, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6872425805216951, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05903074960975321, "3000–6000 Hz_40ms_envelope_correlation": 0.9975122376999014, "3000–6000 Hz_40ms_modulation_mae_db": 0.751441410574551, "mr_stft_db": 4.557317424428976, "band_mae_db": 0.10537842679594465, "rms_gain_error_db": -0.027103206370508837, "output_peak": 0.35629942683393645}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.438421997945384, "spectral_convergence": 0.12305902259044114, "spectral_convergence_status": "ok", "amplitude_floor": 1.5362649016028353e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.415859447855407, "spectral_convergence": 0.12027667832441905, "spectral_convergence_status": "ok", "amplitude_floor": 1.0749003550909146e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.816878324454692, "spectral_convergence": 0.11836201676481843, "spectral_convergence_status": "ok", "amplitude_floor": 8.625899017329306e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001501246216341323, "output_power": 0.001500139004997226, "delta_db": -0.0032042258050456146, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002746582421816528, "output_power": 0.0026972383768235023, "delta_db": -0.0787330596316848, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011276426514964012, "output_power": 0.0011058124070135434, "delta_db": -0.08490035752900166, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016504304929751857, "output_power": 0.00015906999185883312, "delta_db": -0.16008980118018282, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.934184208485853e-05, "output_power": 3.7304678942494904e-05, "delta_db": -0.23091384181918248, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8643647097645308e-05, "output_power": 2.9152187002353077e-05, "delta_db": 0.07642826780015619, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10571159229420894, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06222378505682884, "pitch": null, "output_rms_dbfs": -22.573756676042134, "output_peak": 0.3689228202561212, "reference_rms_dbfs": -22.511532890985304, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.557053256751828}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018834661138467316, "log_rms_mae_db": 0.5767526868035837, "rms_envelope_nrmse": 0.04881172396111846, "envelope_correlation": 0.9966565069257467, "rms_gain_error_db": -0.06114317687975646, "reference_mean_power": 0.005606081823103195, "output_mean_power": 0.005527708267712647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7511238572638392e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009260227506744139, "output_power": 0.0009103686846134877, "reference_active": true, "delta_db": -0.07404346432871411}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010953814679795108, "output_power": 9.752462318978937e-05, "reference_active": true, "delta_db": -0.5045110813518068}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.583845526426034e-05, "output_power": 2.0126804766930366e-05, "reference_active": true, "delta_db": -1.0849171218426}], "active_band_count": 3, "active_band_mae_db": 0.5544905558410402}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015988223985812614, "log_rms_mae_db": 0.323307322916064, "rms_envelope_nrmse": 0.026444323665010743, "envelope_correlation": 0.9989733683611318, "rms_gain_error_db": -0.061044434310565075, "reference_mean_power": 0.00558793457857638, "output_mean_power": 0.005509939997701946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5615527006220852e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008311543470725633, "output_power": 0.000820808181147303, "reference_active": true, "delta_db": -0.05440004169452683}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.497505864184299e-05, "output_power": 4.7304093879865924e-05, "reference_active": true, "delta_db": -0.6526697340037425}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0542332226205543e-06, "output_power": 1.4667816739941613e-06, "reference_active": true, "delta_db": -1.4628427344389276}], "active_band_count": 3, "active_band_mae_db": 0.7233041700457323}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016200114302549476, "log_rms_mae_db": 1.185838311017948, "rms_envelope_nrmse": 0.0751275327021665, "envelope_correlation": 0.9946158685509372, "rms_gain_error_db": -0.05153271137420736, "reference_mean_power": 0.0020825301889685474, "output_mean_power": 0.002057965238089258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.598844168109503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038009245216565237, "output_power": 0.00037789458765389574, "reference_active": true, "delta_db": -0.025185736811049714}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.129961338774397e-05, "output_power": 5.7937108436672996e-05, "reference_active": true, "delta_db": -0.24500919109099042}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4399990501668035e-05, "output_power": 1.2360434577232065e-05, "reference_active": true, "delta_db": -0.6632846536632497}], "active_band_count": 3, "active_band_mae_db": 0.3111598605217633}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012179662863477574, "log_rms_mae_db": 0.7346048400958017, "rms_envelope_nrmse": 0.0408653019630715, "envelope_correlation": 0.9983216912059474, "rms_gain_error_db": -0.05196617660490714, "reference_mean_power": 0.0020746475669658963, "output_mean_power": 0.0020499709814350633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.889619852823582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035505972043036534, "output_power": 0.00035337501947908166, "reference_active": true, "delta_db": -0.020655613419307245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2260222284985983e-05, "output_power": 3.02144315026944e-05, "reference_active": true, "delta_db": -0.2845292828749204}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1886983564224358e-06, "output_power": 8.335049820800044e-07, "reference_active": true, "delta_db": -1.5416346215567034}], "active_band_count": 3, "active_band_mae_db": 0.6156065059503103}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.2267966954040695e-05, "log_rms_mae_db": 1.2137646949991532, "rms_envelope_nrmse": 0.09134586822090385, "envelope_correlation": 0.9937870916762254, "rms_gain_error_db": -0.1962691560176803, "reference_mean_power": 0.00015543877011727877, "output_mean_power": 0.0001485704489548427}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4080839081981722e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.101806003601554e-05, "output_power": 6.730122639129533e-05, "reference_active": true, "delta_db": -0.2334582653456354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0132703192354966e-06, "output_power": 1.8671340396204751e-06, "reference_active": true, "delta_db": -0.3272659434973839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.589135119213016e-07, "output_power": 6.796317660792981e-07, "reference_active": true, "delta_db": -0.47918615549700505}], "active_band_count": 3, "active_band_mae_db": 0.3466367881133414}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.184196106800431e-05, "log_rms_mae_db": 0.7149548881140315, "rms_envelope_nrmse": 0.05238943607379713, "envelope_correlation": 0.9981494129684777, "rms_gain_error_db": -0.2010579216861566, "reference_mean_power": 0.00015383740434286893, "output_mean_power": 0.0001468777974759262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3572636139953185e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.549215276795424e-05, "output_power": 6.208784321197018e-05, "reference_active": true, "delta_db": -0.23182692471748478}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.98253690709034e-07, "output_power": 9.503989758089168e-07, "reference_active": true, "delta_db": -0.2133496490505331}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.830181704119583e-08, "output_power": 4.734606502626261e-08, "reference_active": true, "delta_db": -0.08679578322428227}], "active_band_count": 3, "active_band_mae_db": 0.1773241189974334}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.781014610300033e-05, "log_rms_mae_db": 1.2010058094615943, "rms_envelope_nrmse": 0.13576063730308724, "envelope_correlation": 0.9869292517410594, "rms_gain_error_db": -0.015356393505211842, "reference_mean_power": 2.2945322907755735e-05, "output_mean_power": 2.2864332888643015e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.468236349213924e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8725122830509875e-06, "output_power": 3.6105286556114314e-06, "reference_active": true, "delta_db": -0.3042200775212734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.737803238621202e-07, "output_power": 4.6099261847693013e-07, "reference_active": true, "delta_db": -0.9505167987211594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.433384337397866e-07, "output_power": 1.0088082111329868e-07, "reference_active": true, "delta_db": -1.525540461272734}], "active_band_count": 3, "active_band_mae_db": 0.9267591125050556}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.4952988731037596e-05, "log_rms_mae_db": 0.7080932353732757, "rms_envelope_nrmse": 0.06510323121318923, "envelope_correlation": 0.996924789968233, "rms_gain_error_db": -0.014899292165809748, "reference_mean_power": 2.2958960928334722e-05, "output_mean_power": 2.288033083294743e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.967813463196935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5837220182078332e-06, "output_power": 3.3370039910284584e-06, "reference_active": true, "delta_db": -0.30977588917860616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.664756741318262e-07, "output_power": 2.1745256712444306e-07, "reference_active": true, "delta_db": -0.8829303049183126}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1431200548978765e-08, "output_power": 7.632722702418485e-09, "reference_active": true, "delta_db": -1.7541235937391273}], "active_band_count": 3, "active_band_mae_db": 0.9822765959453488}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9624687459538068, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3793c588b3c731747a5d6060628879057aea18216f1d2e2cf1be07bbbad24bf1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9936354323283625, "standard_deviation": 0.0008606988468870797, "minimum": 0.9922356918079016, "maximum": 0.9947503157981223, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2099761983446373, "standard_deviation": 0.035019878827145065, "minimum": 1.161533544343844, "maximum": 1.2789184867471899, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4876954482401815, "standard_deviation": 0.20511159719267477, "minimum": 0.28185422467894194, "maximum": 0.8086446640057409, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09247334757056591, "standard_deviation": 0.005967287534101779, "minimum": 0.08580641315686607, "maximum": 0.10285754320233732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9980307448189865, "standard_deviation": 0.0004776307885964055, "minimum": 0.9972331399611803, "maximum": 0.9986302827300213, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7355501765138326, "standard_deviation": 0.032543735166087565, "minimum": 0.6826778623257012, "maximum": 0.7902440037097179, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4629896179412796, "standard_deviation": 0.26667526319968615, "minimum": 0.1773241189974334, "maximum": 0.9069124662046392, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05375906860663317, "standard_deviation": 0.005296315645875882, "minimum": 0.0489036604335967, "maximum": 0.06388582306913043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9864798795288556, "standard_deviation": 0.001676606869351099, "minimum": 0.982976522480321, "maximum": 0.9884779513363806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1821066697253901, "standard_deviation": 0.021321689276806388, "minimum": 1.1567676836321668, "maximum": 1.2078500734190327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.545231530859625, "standard_deviation": 0.25891673538944887, "minimum": 0.2462285941651253, "maximum": 0.9267591125050556, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1381204716016139, "standard_deviation": 0.009283375100805501, "minimum": 0.12718990256444032, "maximum": 0.15803097590459753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9962966556512074, "standard_deviation": 0.0009672431312540938, "minimum": 0.994386431675525, "maximum": 0.9975122376999014, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6961516044316902, "standard_deviation": 0.025190103736300484, "minimum": 0.6668843813677352, "maximum": 0.7424744161272141, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6641830517170104, "standard_deviation": 0.31868024262016703, "minimum": 0.1729403583457855, "maximum": 1.015690512581631, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07151478598435196, "standard_deviation": 0.009972531692904069, "minimum": 0.05903074960975321, "maximum": 0.09115748398850225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.994918746671537, "standard_deviation": 0.0010333582542307483, "minimum": 0.9933657047996693, "maximum": 0.9962195321468527, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2057656327343504, "standard_deviation": 0.056222147948835524, "minimum": 1.120644531421786, "maximum": 1.2885846039238087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5612787398499242, "standard_deviation": 0.34502957747870744, "minimum": 0.06648105562645039, "maximum": 1.0754045697632115, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07297431258017054, "standard_deviation": 0.00713830600616842, "minimum": 0.06299767558884475, "maximum": 0.08326860014046217, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982722318422159, "standard_deviation": 0.0004301864727183277, "minimum": 0.9974867621101109, "maximum": 0.9988772255990312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7510001823422472, "standard_deviation": 0.043572569866182714, "minimum": 0.7064206790528414, "maximum": 0.8283604211902039, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.779654427151758, "standard_deviation": 0.3485896617077776, "minimum": 0.19004834956796546, "maximum": 1.2370970363869336, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04139335448701871, "standard_deviation": 0.005249819776541557, "minimum": 0.03318263994628193, "maximum": 0.049533180808854764, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12316351489240958, "standard_deviation": 0.019101289892316402, "minimum": 0.10414110130827525, "maximum": 0.16019693183366882, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.557632638983833, "standard_deviation": 0.004461927159826019, "minimum": 4.547993315139663, "maximum": 4.563723095338505, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3639147478610512, "standard_deviation": 0.005484554188447062, "minimum": 0.35629942683393645, "maximum": 0.37391005108620123, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05419106990847809, "standard_deviation": 0.01737444363079248, "minimum": -0.07931908780010535, "maximum": -0.027103206370508837, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996979754477056, "standard_deviation": 0.00044449525854361045, "minimum": 0.9964438373680962, "maximum": 0.9974854041621873, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5682917301086652, "standard_deviation": 0.017471787835913504, "minimum": 0.545491593072728, "maximum": 0.6032997929559526, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4249261859681589, "standard_deviation": 0.17646392560411972, "minimum": 0.14755644086427486, "maximum": 0.7336880975597414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04624636729082985, "standard_deviation": 0.0033298666207090192, "minimum": 0.04233704971397833, "maximum": 0.04995845559615617, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990918918306775, "standard_deviation": 0.0002414440926766476, "minimum": 0.9987039402843321, "maximum": 0.9994040213964904, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.32266366359410675, "standard_deviation": 0.019058384149952742, "minimum": 0.2941604457607665, "maximum": 0.3540821506509715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6296599121808786, "standard_deviation": 0.20098424989291785, "minimum": 0.29360776557393503, "maximum": 0.8790680835521246, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.024579923075962663, "standard_deviation": 0.0031820628002675604, "minimum": 0.02049114338332439, "maximum": 0.02968045624082671, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1837-0023", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.182121992111206, "predicted_mos": 3.182121992111206, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1935679a1231e084e52a6bb388df02d62da3636d01298dc6e33131770b2702e5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.101664999965578, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5762207958016726, "全帯域_10ms_rms_envelope_nrmse": 0.05173989013410455, "全帯域_10ms_envelope_correlation": 0.9962206725805041, "全帯域_10ms_modulation_mae_db": 0.3120908470574089, "全帯域_40ms_log_rms_mae_db": 0.3355106510433129, "全帯域_40ms_rms_envelope_nrmse": 0.02860411066970969, "全帯域_40ms_envelope_correlation": 0.9987640421289028, "全帯域_40ms_modulation_mae_db": 0.5445348917137703, "300–1000 Hz_10ms_log_rms_mae_db": 1.2639289402213472, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07292644505562611, "300–1000 Hz_10ms_envelope_correlation": 0.9949075460325466, "300–1000 Hz_10ms_modulation_mae_db": 0.21474941694928898, "300–1000 Hz_40ms_log_rms_mae_db": 0.7988470437509942, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038855809180445094, "300–1000 Hz_40ms_envelope_correlation": 0.9984534940477737, "300–1000 Hz_40ms_modulation_mae_db": 0.38090056546328466, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3314242501789082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10094780951686907, "1000–3000 Hz_10ms_envelope_correlation": 0.9921976942325929, "1000–3000 Hz_10ms_modulation_mae_db": 0.6383537034361536, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8028794451456651, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05744474949294934, "1000–3000 Hz_40ms_envelope_correlation": 0.9975057346376166, "1000–3000 Hz_40ms_modulation_mae_db": 0.6907280808760472, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4028459742305353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16288471300572996, "3000–6000 Hz_10ms_envelope_correlation": 0.9812224436099293, "3000–6000 Hz_10ms_modulation_mae_db": 0.22575840468905942, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8011268401921184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07513308255075807, "3000–6000 Hz_40ms_envelope_correlation": 0.9958324602159857, "3000–6000 Hz_40ms_modulation_mae_db": 0.3405802364122088, "mr_stft_db": 4.6690344578678085, "band_mae_db": 0.0698939033389444, "rms_gain_error_db": -0.05896455503674429, "output_peak": 0.38081213498468236}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5808344972859252, "全帯域_10ms_rms_envelope_nrmse": 0.04520402936026215, "全帯域_10ms_envelope_correlation": 0.9971276453522758, "全帯域_10ms_modulation_mae_db": 0.26932625750746203, "全帯域_40ms_log_rms_mae_db": 0.3314907422576264, "全帯域_40ms_rms_envelope_nrmse": 0.021878197590191856, "全帯域_40ms_envelope_correlation": 0.9993029288844427, "全帯域_40ms_modulation_mae_db": 0.2916150033241715, "300–1000 Hz_10ms_log_rms_mae_db": 1.2796266443720534, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0689981954679954, "300–1000 Hz_10ms_envelope_correlation": 0.9955017776219155, "300–1000 Hz_10ms_modulation_mae_db": 0.41855975329009104, "300–1000 Hz_40ms_log_rms_mae_db": 0.7924533116836767, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03865304663844084, "300–1000 Hz_40ms_envelope_correlation": 0.9985079357905603, "300–1000 Hz_40ms_modulation_mae_db": 0.5419331704471481, "1000–3000 Hz_10ms_log_rms_mae_db": 1.386057951172854, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10489346674449637, "1000–3000 Hz_10ms_envelope_correlation": 0.9916755032397533, "1000–3000 Hz_10ms_modulation_mae_db": 0.4920358674961373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8664341518665256, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05982835743784184, "1000–3000 Hz_40ms_envelope_correlation": 0.9973873298504622, "1000–3000 Hz_40ms_modulation_mae_db": 0.38262467936828704, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3893608159323132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1444169861067269, "3000–6000 Hz_10ms_envelope_correlation": 0.9850834302570698, "3000–6000 Hz_10ms_modulation_mae_db": 0.39628127370283034, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7831714735352046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07321386449025032, "3000–6000 Hz_40ms_envelope_correlation": 0.9960743372345507, "3000–6000 Hz_40ms_modulation_mae_db": 0.74249445084551, "mr_stft_db": 4.674709234467213, "band_mae_db": 0.14486016588768888, "rms_gain_error_db": -0.05714775450296288, "output_peak": 0.3980065725739011}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5600401137716632, "全帯域_10ms_rms_envelope_nrmse": 0.04126512354332505, "全帯域_10ms_envelope_correlation": 0.9975952089710527, "全帯域_10ms_modulation_mae_db": 0.3641882323329068, "全帯域_40ms_log_rms_mae_db": 0.321502193547775, "全帯域_40ms_rms_envelope_nrmse": 0.020323115880993815, "全帯域_40ms_envelope_correlation": 0.9993784475886456, "全帯域_40ms_modulation_mae_db": 0.47462809933952715, "300–1000 Hz_10ms_log_rms_mae_db": 1.2269990077167636, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07074448963938804, "300–1000 Hz_10ms_envelope_correlation": 0.995234426880921, "300–1000 Hz_10ms_modulation_mae_db": 0.5771665135407918, "300–1000 Hz_40ms_log_rms_mae_db": 0.7352392609353605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03867407099021115, "300–1000 Hz_40ms_envelope_correlation": 0.9984901980635893, "300–1000 Hz_40ms_modulation_mae_db": 0.7973870595108044, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3547673597620666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09322213074989355, "1000–3000 Hz_10ms_envelope_correlation": 0.9934382031798292, "1000–3000 Hz_10ms_modulation_mae_db": 0.2041128627102187, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8181657002950472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053718475181290935, "1000–3000 Hz_40ms_envelope_correlation": 0.9979448679171636, "1000–3000 Hz_40ms_modulation_mae_db": 0.3129696657309957, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3741486223065256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14502204290665774, "3000–6000 Hz_10ms_envelope_correlation": 0.9855020964010293, "3000–6000 Hz_10ms_modulation_mae_db": 0.26316455697745683, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7815957201755778, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07653801029865019, "3000–6000 Hz_40ms_envelope_correlation": 0.9959506462550012, "3000–6000 Hz_40ms_modulation_mae_db": 0.5100068434631071, "mr_stft_db": 4.679042418544442, "band_mae_db": 0.1430997906606316, "rms_gain_error_db": -0.042258588585092224, "output_peak": 0.354785897396594}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5396876885174022, "全帯域_10ms_rms_envelope_nrmse": 0.04729518615794483, "全帯域_10ms_envelope_correlation": 0.9969371096749976, "全帯域_10ms_modulation_mae_db": 0.7910995934529588, "全帯域_40ms_log_rms_mae_db": 0.2955355516092557, "全帯域_40ms_rms_envelope_nrmse": 0.02849234044630559, "全帯域_40ms_envelope_correlation": 0.9988802548295506, "全帯域_40ms_modulation_mae_db": 0.9812505572395005, "300–1000 Hz_10ms_log_rms_mae_db": 1.2520593564679774, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07770445112294169, "300–1000 Hz_10ms_envelope_correlation": 0.9943850308486859, "300–1000 Hz_10ms_modulation_mae_db": 1.05051737595878, "300–1000 Hz_40ms_log_rms_mae_db": 0.7966804456956051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051166707823300746, "300–1000 Hz_40ms_envelope_correlation": 0.9974583038476651, "300–1000 Hz_40ms_modulation_mae_db": 1.1715948710920048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.37316590755102, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0973489243799507, "1000–3000 Hz_10ms_envelope_correlation": 0.992985054048944, "1000–3000 Hz_10ms_modulation_mae_db": 0.8305604100715019, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8477617567444407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056558631008954904, "1000–3000 Hz_40ms_envelope_correlation": 0.9978131615203126, "1000–3000 Hz_40ms_modulation_mae_db": 0.9021146800023327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3727666175295947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14001024071639567, "3000–6000 Hz_10ms_envelope_correlation": 0.9860022774751405, "3000–6000 Hz_10ms_modulation_mae_db": 0.4305434966265754, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8022191749066856, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07201602993454222, "3000–6000 Hz_40ms_envelope_correlation": 0.9960978790835184, "3000–6000 Hz_40ms_modulation_mae_db": 0.8463701345547664, "mr_stft_db": 4.686061940545233, "band_mae_db": 0.10240194375713683, "rms_gain_error_db": -0.08729997070500665, "output_peak": 0.3682593948250488}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5772046489709562, "全帯域_10ms_rms_envelope_nrmse": 0.04877172999487978, "全帯域_10ms_envelope_correlation": 0.9966745584688956, "全帯域_10ms_modulation_mae_db": 0.4174664209103993, "全帯域_40ms_log_rms_mae_db": 0.3177010082784492, "全帯域_40ms_rms_envelope_nrmse": 0.026249398783083376, "全帯域_40ms_envelope_correlation": 0.9990057851814574, "全帯域_40ms_modulation_mae_db": 0.4490655134908425, "300–1000 Hz_10ms_log_rms_mae_db": 1.2258978561521359, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07962346987629391, "300–1000 Hz_10ms_envelope_correlation": 0.9940023596776624, "300–1000 Hz_10ms_modulation_mae_db": 0.9021699438593608, "300–1000 Hz_40ms_log_rms_mae_db": 0.8056726632728135, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04689225576156657, "300–1000 Hz_40ms_envelope_correlation": 0.9978335766715887, "300–1000 Hz_40ms_modulation_mae_db": 1.1916437941032083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3916414374113375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09918735104390627, "1000–3000 Hz_10ms_envelope_correlation": 0.9925821805299612, "1000–3000 Hz_10ms_modulation_mae_db": 0.2439378690275095, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8749874470534894, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057631542054063276, "1000–3000 Hz_40ms_envelope_correlation": 0.9976142273678814, "1000–3000 Hz_40ms_modulation_mae_db": 0.1385662786924315, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3864746680936897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1490878754087018, "3000–6000 Hz_10ms_envelope_correlation": 0.984178018274338, "3000–6000 Hz_10ms_modulation_mae_db": 0.41719158009909735, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7984412553523487, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08034068554307683, "3000–6000 Hz_40ms_envelope_correlation": 0.9951699518888277, "3000–6000 Hz_40ms_modulation_mae_db": 0.42121631331477255, "mr_stft_db": 4.68549533176882, "band_mae_db": 0.1225837892916601, "rms_gain_error_db": -0.06838854208147521, "output_peak": 0.36612294315943117}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5648933605718577, "全帯域_10ms_rms_envelope_nrmse": 0.04384951630261416, "全帯域_10ms_envelope_correlation": 0.9973298805602387, "全帯域_10ms_modulation_mae_db": 0.38135016759408097, "全帯域_40ms_log_rms_mae_db": 0.3122812059095601, "全帯域_40ms_rms_envelope_nrmse": 0.022438584983022966, "全帯域_40ms_envelope_correlation": 0.9992901506695095, "全帯域_40ms_modulation_mae_db": 0.5171994343464125, "300–1000 Hz_10ms_log_rms_mae_db": 1.2560062638997644, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07479974636919923, "300–1000 Hz_10ms_envelope_correlation": 0.9947991898148614, "300–1000 Hz_10ms_modulation_mae_db": 0.6932638996952895, "300–1000 Hz_40ms_log_rms_mae_db": 0.7665776356871716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04649996724905267, "300–1000 Hz_40ms_envelope_correlation": 0.9979387748643689, "300–1000 Hz_40ms_modulation_mae_db": 0.8263822737428254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2960390722297492, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09312764317066989, "1000–3000 Hz_10ms_envelope_correlation": 0.9934568924700807, "1000–3000 Hz_10ms_modulation_mae_db": 0.38578251927581375, "1000–3000 Hz_40ms_log_rms_mae_db": 0.759336298067059, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05216117337243466, "1000–3000 Hz_40ms_envelope_correlation": 0.9980440179072275, "1000–3000 Hz_40ms_modulation_mae_db": 0.39064347965824037, "3000–6000 Hz_10ms_log_rms_mae_db": 1.423874592292165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1629053260311066, "3000–6000 Hz_10ms_envelope_correlation": 0.9817360596028, "3000–6000 Hz_10ms_modulation_mae_db": 0.4565822319021582, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8277286183614879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09652447106763103, "3000–6000 Hz_40ms_envelope_correlation": 0.993434389997125, "3000–6000 Hz_40ms_modulation_mae_db": 0.7643926503934599, "mr_stft_db": 4.672481393323632, "band_mae_db": 0.08821301405013678, "rms_gain_error_db": -0.07143803549533329, "output_peak": 0.3643095087032111}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5511784999050634, "全帯域_10ms_rms_envelope_nrmse": 0.04396759052324682, "全帯域_10ms_envelope_correlation": 0.9972657989124004, "全帯域_10ms_modulation_mae_db": 0.4440044369149206, "全帯域_40ms_log_rms_mae_db": 0.29624027763875505, "全帯域_40ms_rms_envelope_nrmse": 0.022798668830775817, "全帯域_40ms_envelope_correlation": 0.9992108912452734, "全帯域_40ms_modulation_mae_db": 0.8672047071480383, "300–1000 Hz_10ms_log_rms_mae_db": 1.2224649033240118, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07106617744190431, "300–1000 Hz_10ms_envelope_correlation": 0.9952178896041812, "300–1000 Hz_10ms_modulation_mae_db": 1.0205817482272, "300–1000 Hz_40ms_log_rms_mae_db": 0.761848626186829, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03901273907578231, "300–1000 Hz_40ms_envelope_correlation": 0.9985052069032159, "300–1000 Hz_40ms_modulation_mae_db": 1.3856468272994098, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3469785486904144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0918731856133982, "1000–3000 Hz_10ms_envelope_correlation": 0.9934719905033207, "1000–3000 Hz_10ms_modulation_mae_db": 0.246127183274828, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8014594584989607, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04709861279244478, "1000–3000 Hz_40ms_envelope_correlation": 0.9982619535919122, "1000–3000 Hz_40ms_modulation_mae_db": 0.043315073405249684, "3000–6000 Hz_10ms_log_rms_mae_db": 1.358308196480963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1496093965601377, "3000–6000 Hz_10ms_envelope_correlation": 0.9840017603215537, "3000–6000 Hz_10ms_modulation_mae_db": 0.441553480821516, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8004305192089198, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08382237313512894, "3000–6000 Hz_40ms_envelope_correlation": 0.9947525425104995, "3000–6000 Hz_40ms_modulation_mae_db": 0.6273148506989382, "mr_stft_db": 4.679718322542146, "band_mae_db": 0.08696013410035736, "rms_gain_error_db": -0.044840555705794094, "output_peak": 0.3576234523479899}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5630782400576996, "全帯域_10ms_rms_envelope_nrmse": 0.042983816889673385, "全帯域_10ms_envelope_correlation": 0.997454226848772, "全帯域_10ms_modulation_mae_db": 0.2827126104224847, "全帯域_40ms_log_rms_mae_db": 0.32715225327107145, "全帯域_40ms_rms_envelope_nrmse": 0.023423216936371335, "全帯域_40ms_envelope_correlation": 0.999247136692818, "全帯域_40ms_modulation_mae_db": 0.3798758403368627, "300–1000 Hz_10ms_log_rms_mae_db": 1.255048255405005, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06785406742736116, "300–1000 Hz_10ms_envelope_correlation": 0.9957074142894344, "300–1000 Hz_10ms_modulation_mae_db": 0.3104919150374801, "300–1000 Hz_40ms_log_rms_mae_db": 0.8106716352130348, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03888982524336618, "300–1000 Hz_40ms_envelope_correlation": 0.9985761138067928, "300–1000 Hz_40ms_modulation_mae_db": 0.45204102358457926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3176386735682617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09944217178445307, "1000–3000 Hz_10ms_envelope_correlation": 0.9924406787430141, "1000–3000 Hz_10ms_modulation_mae_db": 0.40054855863580047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7988809657319736, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05725824122766713, "1000–3000 Hz_40ms_envelope_correlation": 0.9975010290508456, "1000–3000 Hz_40ms_modulation_mae_db": 0.45670070401813073, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3788556787801427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1456147889427203, "3000–6000 Hz_10ms_envelope_correlation": 0.98482314403625, "3000–6000 Hz_10ms_modulation_mae_db": 1.0330250504400393, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7658216986823699, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07292287584907, "3000–6000 Hz_40ms_envelope_correlation": 0.9960102518622335, "3000–6000 Hz_40ms_modulation_mae_db": 1.2625462645589554, "mr_stft_db": 4.6721380364051885, "band_mae_db": 0.12944311986568083, "rms_gain_error_db": -0.07849714467485741, "output_peak": 0.3531779812024366}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.593900161712388, "spectral_convergence": 0.12420178914577755, "spectral_convergence_status": "ok", "amplitude_floor": 1.5362649016028353e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.550369058357186, "spectral_convergence": 0.11881206545017373, "spectral_convergence_status": "ok", "amplitude_floor": 1.0749003550909146e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.862834153533849, "spectral_convergence": 0.1159936052015742, "spectral_convergence_status": "ok", "amplitude_floor": 8.625899017329306e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001501246216341323, "output_power": 0.0014870914119580786, "delta_db": -0.04114260284908699, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002746582421816528, "output_power": 0.002711646056101496, "delta_db": -0.0555963463813539, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011276426514964012, "output_power": 0.0011076294972392265, "delta_db": -0.07776981180383982, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016504304929751857, "output_power": 0.0001583485290841634, "delta_db": -0.17983205444059086, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.934184208485853e-05, "output_power": 3.968657300323407e-05, "delta_db": 0.03788907699294549, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8643647097645308e-05, "output_power": 2.8465247326842094e-05, "delta_db": -0.02713352756584928, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.0698939033389444, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05896455503674429, "pitch": null, "output_rms_dbfs": -22.570497446022046, "output_peak": 0.38081213498468236, "reference_rms_dbfs": -22.511532890985304, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.6690344578678085}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018834661138467316, "log_rms_mae_db": 0.5762207958016726, "rms_envelope_nrmse": 0.05173989013410455, "envelope_correlation": 0.9962206725805041, "rms_gain_error_db": -0.0564453410137717, "reference_mean_power": 0.005606081823103195, "output_mean_power": 0.005533690917149483}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7511238572638392e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009260227506744139, "output_power": 0.0009220666626783201, "reference_active": true, "delta_db": -0.01859336240475748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010953814679795108, "output_power": 0.00010168563400255935, "reference_active": true, "delta_db": -0.3230578835121733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.583845526426034e-05, "output_power": 2.2532242496478424e-05, "reference_active": true, "delta_db": -0.594621295255296}], "active_band_count": 3, "active_band_mae_db": 0.3120908470574089}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015988223985812614, "log_rms_mae_db": 0.3355106510433129, "rms_envelope_nrmse": 0.02860411066970969, "envelope_correlation": 0.9987640421289028, "rms_gain_error_db": -0.053724221279609206, "reference_mean_power": 0.00558793457857638, "output_mean_power": 0.0055192350607563105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5615527006220852e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008311543470725633, "output_power": 0.0008320337859906331, "reference_active": true, "delta_db": 0.0045928118087388975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.497505864184299e-05, "output_power": 5.029696387777535e-05, "reference_active": true, "delta_db": -0.38623931167582926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0542332226205543e-06, "output_power": 1.5430233336021237e-06, "reference_active": true, "delta_db": -1.2427725516567427}], "active_band_count": 3, "active_band_mae_db": 0.5445348917137703}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016200114302549476, "log_rms_mae_db": 1.2639289402213472, "rms_envelope_nrmse": 0.07292644505562611, "envelope_correlation": 0.9949075460325466, "rms_gain_error_db": -0.04713079340782139, "reference_mean_power": 0.0020825301889685474, "output_mean_power": 0.0020600522060541656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.598844168109503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038009245216565237, "output_power": 0.0003750972251262805, "reference_active": true, "delta_db": -0.057453941537438616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.129961338774397e-05, "output_power": 5.899548549523581e-05, "reference_active": true, "delta_db": -0.16638956016112233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4399990501668035e-05, "output_power": 1.3071388741639633e-05, "reference_active": true, "delta_db": -0.42040474914930603}], "active_band_count": 3, "active_band_mae_db": 0.21474941694928898}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012179662863477574, "log_rms_mae_db": 0.7988470437509942, "rms_envelope_nrmse": 0.038855809180445094, "envelope_correlation": 0.9984534940477737, "rms_gain_error_db": -0.04472988259096044, "reference_mean_power": 0.0020746475669658963, "output_mean_power": 0.0020533895277737154}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.889619852823582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035505972043036534, "output_power": 0.00035183809995278903, "reference_active": true, "delta_db": -0.03958540096203797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2260222284985983e-05, "output_power": 3.1350001753224576e-05, "reference_active": true, "delta_db": -0.12429786060193655}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1886983564224358e-06, "output_power": 9.488330827623455e-07, "reference_active": true, "delta_db": -0.9788184348258795}], "active_band_count": 3, "active_band_mae_db": 0.38090056546328466}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.2267966954040695e-05, "log_rms_mae_db": 1.3314242501789082, "rms_envelope_nrmse": 0.10094780951686907, "envelope_correlation": 0.9921976942325929, "rms_gain_error_db": -0.1482787774407789, "reference_mean_power": 0.00015543877011727877, "output_mean_power": 0.00015022128532674964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4080839081981722e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.101806003601554e-05, "output_power": 6.791561309656884e-05, "reference_active": true, "delta_db": -0.19399179162568952}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0132703192354966e-06, "output_power": 1.940128535544712e-06, "reference_active": true, "delta_db": -0.1607158762235519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.589135119213016e-07, "output_power": 5.298548829099191e-07, "reference_active": true, "delta_db": -1.5603534424592194}], "active_band_count": 3, "active_band_mae_db": 0.6383537034361536}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.184196106800431e-05, "log_rms_mae_db": 0.8028794451456651, "rms_envelope_nrmse": 0.05744474949294934, "envelope_correlation": 0.9975057346376166, "rms_gain_error_db": -0.14892396806714692, "reference_mean_power": 0.00015383740434286893, "output_mean_power": 0.00014865158586043088}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3572636139953185e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.549215276795424e-05, "output_power": 6.310562149879285e-05, "reference_active": true, "delta_db": -0.16121217930081586}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.98253690709034e-07, "output_power": 9.783941167636875e-07, "reference_active": true, "delta_db": -0.08727091972213999}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.830181704119583e-08, "output_power": 3.173900839501422e-08, "reference_active": true, "delta_db": -1.8237011436051855}], "active_band_count": 3, "active_band_mae_db": 0.6907280808760472}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.781014610300033e-05, "log_rms_mae_db": 1.4028459742305353, "rms_envelope_nrmse": 0.16288471300572996, "envelope_correlation": 0.9812224436099293, "rms_gain_error_db": 0.045458175621184434, "reference_mean_power": 2.2945322907755735e-05, "output_mean_power": 2.3186755977347038e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.468236349213924e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8725122830509875e-06, "output_power": 4.007765958643517e-06, "reference_active": true, "delta_db": 0.14909548168361555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.737803238621202e-07, "output_power": 5.448638073955929e-07, "reference_active": true, "delta_db": -0.2245769046076165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.433384337397866e-07, "output_power": 1.5371736717765064e-07, "reference_active": true, "delta_db": 0.3036028277759462}], "active_band_count": 3, "active_band_mae_db": 0.22575840468905942}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.4952988731037596e-05, "log_rms_mae_db": 0.8011268401921184, "rms_envelope_nrmse": 0.07513308255075807, "envelope_correlation": 0.9958324602159857, "rms_gain_error_db": 0.04649068662911564, "reference_mean_power": 2.2958960928334722e-05, "output_mean_power": 2.320605395324497e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.967813463196935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5837220182078332e-06, "output_power": 3.7428307967485026e-06, "reference_active": true, "delta_db": 0.1886587929447084}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.664756741318262e-07, "output_power": 2.5400825047455544e-07, "reference_active": true, "delta_db": -0.20809746339517574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1431200548978765e-08, "output_power": 9.899047285658515e-09, "reference_active": true, "delta_db": -0.6249844528967423}], "active_band_count": 3, "active_band_mae_db": 0.3405802364122088}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9575745334038547, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "34059cdc30b570408062beed703264c8e9402471b2a14d898f7b5960704575c2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.992781024618437, "standard_deviation": 0.0006682139329415131, "minimum": 0.9916755032397533, "maximum": 0.9934719905033207, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3497141500705765, "standard_deviation": 0.033607276606737076, "minimum": 1.2960390722297492, "maximum": 1.3916414374113375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4301823717409954, "standard_deviation": 0.21711879294131703, "minimum": 0.2041128627102187, "maximum": 0.8305604100715019, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09750533537545464, "standard_deviation": 0.004508265925270289, "minimum": 0.0918731856133982, "maximum": 0.10489346674449637, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9977590402304277, "standard_deviation": 0.0003074010950073334, "minimum": 0.9973873298504622, "maximum": 0.9982619535919122, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8212381529253951, "standard_deviation": 0.03910522219954491, "minimum": 0.759336298067059, "maximum": 0.8749874470534894, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4147078302189644, "standard_deviation": 0.27817307802913643, "minimum": 0.043315073405249684, "maximum": 0.9021146800023327, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05521247282095586, "standard_deviation": 0.004062297695471872, "minimum": 0.04709861279244478, "maximum": 0.05982835743784184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9840686537472638, "standard_deviation": 0.001730030075812414, "minimum": 0.9812224436099293, "maximum": 0.9860022774751405, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3858293957057413, "standard_deviation": 0.02022379789354523, "minimum": 1.358308196480963, "maximum": 1.423874592292165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4580125094073416, "standard_deviation": 0.2475744337653295, "minimum": 0.22575840468905942, "maximum": 1.0330250504400393, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1499439212097721, "standard_deviation": 0.008523062162640541, "minimum": 0.14001024071639567, "maximum": 0.1629053260311066, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954153073809677, "standard_deviation": 0.00093642770167884, "minimum": 0.993434389997125, "maximum": 0.9960978790835184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7950669125518391, "standard_deviation": 0.01841224786311298, "minimum": 0.7658216986823699, "maximum": 0.8277286183614879, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6893652180302148, "standard_deviation": 0.290618394859477, "minimum": 0.3405802364122088, "maximum": 1.2625462645589554, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07881392410863844, "standard_deviation": 0.008212219958159173, "minimum": 0.07201602993454222, "maximum": 0.09652447106763103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.994969454346276, "standard_deviation": 0.0005694018216569287, "minimum": 0.9940023596776624, "maximum": 0.9957074142894344, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2477539034448824, "standard_deviation": 0.020582015052402716, "minimum": 1.2224649033240118, "maximum": 1.2796266443720534, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6484375708197854, "standard_deviation": 0.32221665981474884, "minimum": 0.21474941694928898, "maximum": 1.05051737595878, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07296463030008873, "standard_deviation": 0.004150021703829287, "minimum": 0.06785406742736116, "maximum": 0.07962346987629391, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982204504994443, "standard_deviation": 0.0004187026689607889, "minimum": 0.9974583038476651, "maximum": 0.9985761138067928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7834988278031857, "standard_deviation": 0.026195936908233915, "minimum": 0.7352392609353605, "maximum": 0.8106716352130348, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.843441198155408, "standard_deviation": 0.3748092543582763, "minimum": 0.38090056546328466, "maximum": 1.3856468272994098, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04233055274527069, "standard_deviation": 0.005043879379947031, "minimum": 0.03865304663844084, "maximum": 0.051166707823300746, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11093198261902959, "standard_deviation": 0.028061164987613015, "minimum": 0.0698939033389444, "maximum": 0.14486016588768888, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.67733514193306, "standard_deviation": 0.00629858921639301, "minimum": 4.6690344578678085, "maximum": 4.686061940545233, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36788723564916187, "standard_deviation": 0.015041045357861452, "minimum": 0.3531779812024366, "maximum": 0.3980065725739011, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06360439334840826, "standard_deviation": 0.015759213747297468, "minimum": -0.08729997070500665, "maximum": -0.042258588585092224, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.997075637671142, "standard_deviation": 0.00045163346117011854, "minimum": 0.9962206725805041, "maximum": 0.9975952089710527, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.56414223061028, "standard_deviation": 0.01403952000973089, "minimum": 0.5396876885174022, "maximum": 0.5808344972859252, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40777982077407776, "standard_deviation": 0.16686454313547797, "minimum": 0.26932625750746203, "maximum": 0.7910995934529588, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04563461036325634, "standard_deviation": 0.0034256075518482626, "minimum": 0.04126512354332505, "maximum": 0.05173989013410455, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999134954652575, "standard_deviation": 0.00022331079711540973, "minimum": 0.9987640421289028, "maximum": 0.9993784475886456, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3171767354444757, "standard_deviation": 0.01507202219163763, "minimum": 0.2955355516092557, "maximum": 0.3355106510433129, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5631717558673907, "standard_deviation": 0.23841362102605357, "minimum": 0.2916150033241715, "maximum": 0.9812505572395005, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.024275954265056805, "standard_deviation": 0.0031182632759119452, "minimum": 0.020323115880993815, "maximum": 0.02860411066970969, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_1995-1837-0023", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.101205587387085, "predicted_mos": 3.101205587387085, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3b43f7b35a0f57994cb2a7a1f2b89e6de9f12287af67ae17acbac4c7ef9bf3fc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.6003076997585595, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5521917814691599, "全帯域_10ms_rms_envelope_nrmse": 0.04663634632151297, "全帯域_10ms_envelope_correlation": 0.9969620693369189, "全帯域_10ms_modulation_mae_db": 0.6041072522713052, "全帯域_40ms_log_rms_mae_db": 0.32094158496361824, "全帯域_40ms_rms_envelope_nrmse": 0.024606287352781343, "全帯域_40ms_envelope_correlation": 0.9991201644368196, "全帯域_40ms_modulation_mae_db": 0.7472613509409692, "300–1000 Hz_10ms_log_rms_mae_db": 1.2222219904856417, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07589119890223854, "300–1000 Hz_10ms_envelope_correlation": 0.9945476143764544, "300–1000 Hz_10ms_modulation_mae_db": 0.877137134945272, "300–1000 Hz_40ms_log_rms_mae_db": 0.7535997130819952, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042434298261059905, "300–1000 Hz_40ms_envelope_correlation": 0.9982025734573696, "300–1000 Hz_40ms_modulation_mae_db": 1.1040783690159908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2921279760736724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09130262006272061, "1000–3000 Hz_10ms_envelope_correlation": 0.9935468903052089, "1000–3000 Hz_10ms_modulation_mae_db": 0.2827791765009288, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7468802187757809, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047719745921019144, "1000–3000 Hz_40ms_envelope_correlation": 0.9982057740199979, "1000–3000 Hz_40ms_modulation_mae_db": 0.22221610045754156, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3681006850496462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15197590744467157, "3000–6000 Hz_10ms_envelope_correlation": 0.9838611538127032, "3000–6000 Hz_10ms_modulation_mae_db": 0.5204590781585985, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7738120399959628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08188857218160667, "3000–6000 Hz_40ms_envelope_correlation": 0.9951657257687594, "3000–6000 Hz_40ms_modulation_mae_db": 0.8895994014075619, "mr_stft_db": 4.671379263373129, "band_mae_db": 0.09033154537647516, "rms_gain_error_db": -0.0658665715738181, "output_peak": 0.36376530911114946}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5804272240585473, "全帯域_10ms_rms_envelope_nrmse": 0.04308560327417095, "全帯域_10ms_envelope_correlation": 0.9973750504540898, "全帯域_10ms_modulation_mae_db": 0.482845134196262, "全帯域_40ms_log_rms_mae_db": 0.3150189992582414, "全帯域_40ms_rms_envelope_nrmse": 0.019806257625387114, "全帯域_40ms_envelope_correlation": 0.9994136898258398, "全帯域_40ms_modulation_mae_db": 0.4790328780011359, "300–1000 Hz_10ms_log_rms_mae_db": 1.211881131031086, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06707846297714243, "300–1000 Hz_10ms_envelope_correlation": 0.9957076692329346, "300–1000 Hz_10ms_modulation_mae_db": 0.5564433620997117, "300–1000 Hz_40ms_log_rms_mae_db": 0.7430607239995463, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03870620119374288, "300–1000 Hz_40ms_envelope_correlation": 0.9984625284127764, "300–1000 Hz_40ms_modulation_mae_db": 0.8016017816353539, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3583096137249753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09768286301380391, "1000–3000 Hz_10ms_envelope_correlation": 0.9926615521034552, "1000–3000 Hz_10ms_modulation_mae_db": 0.14950053730601162, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8287879138354143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05351928179736194, "1000–3000 Hz_40ms_envelope_correlation": 0.9978027367714162, "1000–3000 Hz_40ms_modulation_mae_db": 0.19567997053103223, "3000–6000 Hz_10ms_log_rms_mae_db": 1.405874805129966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14431180803692797, "3000–6000 Hz_10ms_envelope_correlation": 0.9851069734956658, "3000–6000 Hz_10ms_modulation_mae_db": 1.0357754215273776, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7974718801072279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07165939030745129, "3000–6000 Hz_40ms_envelope_correlation": 0.9962874704428907, "3000–6000 Hz_40ms_modulation_mae_db": 1.4653155918661316, "mr_stft_db": 4.663199634435945, "band_mae_db": 0.09346536479380872, "rms_gain_error_db": -0.04356278056548295, "output_peak": 0.3641273655694364}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5627400068436431, "全帯域_10ms_rms_envelope_nrmse": 0.045259782083385325, "全帯域_10ms_envelope_correlation": 0.9971984999198368, "全帯域_10ms_modulation_mae_db": 0.5874759013799171, "全帯域_40ms_log_rms_mae_db": 0.33445327296556504, "全帯域_40ms_rms_envelope_nrmse": 0.025433641457528928, "全帯域_40ms_envelope_correlation": 0.999120295812585, "全帯域_40ms_modulation_mae_db": 0.8015938551093397, "300–1000 Hz_10ms_log_rms_mae_db": 1.22165345452485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07292533798044988, "300–1000 Hz_10ms_envelope_correlation": 0.9950407313665368, "300–1000 Hz_10ms_modulation_mae_db": 0.9425832102956484, "300–1000 Hz_40ms_log_rms_mae_db": 0.7371667001656768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043399929893258395, "300–1000 Hz_40ms_envelope_correlation": 0.9982044751568689, "300–1000 Hz_40ms_modulation_mae_db": 1.201209264061915, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3603480261340462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09645652986510274, "1000–3000 Hz_10ms_envelope_correlation": 0.9930882579817587, "1000–3000 Hz_10ms_modulation_mae_db": 0.44789337007979185, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7658163226772206, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05208769034908834, "1000–3000 Hz_40ms_envelope_correlation": 0.9982368137148574, "1000–3000 Hz_40ms_modulation_mae_db": 0.5428362278455164, "3000–6000 Hz_10ms_log_rms_mae_db": 1.387852291951372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14163613971571656, "3000–6000 Hz_10ms_envelope_correlation": 0.985644778964924, "3000–6000 Hz_10ms_modulation_mae_db": 0.5301750665411065, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8217237820565545, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06880457743146232, "3000–6000 Hz_40ms_envelope_correlation": 0.9964443209570926, "3000–6000 Hz_40ms_modulation_mae_db": 0.6432557811190921, "mr_stft_db": 4.671064254594683, "band_mae_db": 0.11138078112993244, "rms_gain_error_db": -0.08647863690550009, "output_peak": 0.3798594502259434}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5573041774813141, "全帯域_10ms_rms_envelope_nrmse": 0.04751046148339452, "全帯域_10ms_envelope_correlation": 0.9968596986949132, "全帯域_10ms_modulation_mae_db": 0.7051496784094057, "全帯域_40ms_log_rms_mae_db": 0.3124278732509115, "全帯域_40ms_rms_envelope_nrmse": 0.026294448195981004, "全帯域_40ms_envelope_correlation": 0.9990108951275279, "全帯域_40ms_modulation_mae_db": 0.9611664943943238, "300–1000 Hz_10ms_log_rms_mae_db": 1.1824060817429913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07665910084825828, "300–1000 Hz_10ms_envelope_correlation": 0.9946167316514737, "300–1000 Hz_10ms_modulation_mae_db": 1.245382099382771, "300–1000 Hz_40ms_log_rms_mae_db": 0.7361708253653068, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047212303248426506, "300–1000 Hz_40ms_envelope_correlation": 0.9979708622182983, "300–1000 Hz_40ms_modulation_mae_db": 1.4409502423043605, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2934435493811667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09004251311839447, "1000–3000 Hz_10ms_envelope_correlation": 0.9938771412802773, "1000–3000 Hz_10ms_modulation_mae_db": 0.12987921408444106, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7754924515443794, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04972216185530951, "1000–3000 Hz_40ms_envelope_correlation": 0.9982760688216842, "1000–3000 Hz_40ms_modulation_mae_db": 0.31093865675717713, "3000–6000 Hz_10ms_log_rms_mae_db": 1.396214117485443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1535483918483141, "3000–6000 Hz_10ms_envelope_correlation": 0.9830906219808837, "3000–6000 Hz_10ms_modulation_mae_db": 0.8282081149748101, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7834594168008376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0781387644453871, "3000–6000 Hz_40ms_envelope_correlation": 0.9954959525969984, "3000–6000 Hz_40ms_modulation_mae_db": 1.0565120546643223, "mr_stft_db": 4.670881884431316, "band_mae_db": 0.10549773961527885, "rms_gain_error_db": -0.07338409394953492, "output_peak": 0.34952213737516075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5481838396668507, "全帯域_10ms_rms_envelope_nrmse": 0.046204846966329476, "全帯域_10ms_envelope_correlation": 0.9970530137229899, "全帯域_10ms_modulation_mae_db": 0.5003680871468578, "全帯域_40ms_log_rms_mae_db": 0.319018277856633, "全帯域_40ms_rms_envelope_nrmse": 0.026336217203303255, "全帯域_40ms_envelope_correlation": 0.9990311252808073, "全帯域_40ms_modulation_mae_db": 0.5176568938636085, "300–1000 Hz_10ms_log_rms_mae_db": 1.2200537937061837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06870896627580544, "300–1000 Hz_10ms_envelope_correlation": 0.995641266358425, "300–1000 Hz_10ms_modulation_mae_db": 1.0096018888930474, "300–1000 Hz_40ms_log_rms_mae_db": 0.7459989973420271, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040756781281639544, "300–1000 Hz_40ms_envelope_correlation": 0.998432587103099, "300–1000 Hz_40ms_modulation_mae_db": 1.0898000030634678, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2899793853208497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09193271291214056, "1000–3000 Hz_10ms_envelope_correlation": 0.9936279178318645, "1000–3000 Hz_10ms_modulation_mae_db": 0.12437079886207493, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7415970661654862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048469028870550546, "1000–3000 Hz_40ms_envelope_correlation": 0.9983675857712719, "1000–3000 Hz_40ms_modulation_mae_db": 0.16693122052776765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.370806650551527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15414144444999625, "3000–6000 Hz_10ms_envelope_correlation": 0.9832495091837424, "3000–6000 Hz_10ms_modulation_mae_db": 0.5473852474277221, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7836197117997524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07532274259989183, "3000–6000 Hz_40ms_envelope_correlation": 0.9958493909982146, "3000–6000 Hz_40ms_modulation_mae_db": 0.4561932846717595, "mr_stft_db": 4.672479064188458, "band_mae_db": 0.1096110751285103, "rms_gain_error_db": -0.08301821622126117, "output_peak": 0.3517602088413107}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5648848590571751, "全帯域_10ms_rms_envelope_nrmse": 0.04697948652471222, "全帯域_10ms_envelope_correlation": 0.9968792103154984, "全帯域_10ms_modulation_mae_db": 0.4512334300505003, "全帯域_40ms_log_rms_mae_db": 0.3248961517680336, "全帯域_40ms_rms_envelope_nrmse": 0.02407106928326685, "全帯域_40ms_envelope_correlation": 0.9991211941004395, "全帯域_40ms_modulation_mae_db": 0.38476347872215033, "300–1000 Hz_10ms_log_rms_mae_db": 1.2456210987754808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0725729899607914, "300–1000 Hz_10ms_envelope_correlation": 0.9950339724682671, "300–1000 Hz_10ms_modulation_mae_db": 1.0781813163820704, "300–1000 Hz_40ms_log_rms_mae_db": 0.7576883639567111, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036862553723234576, "300–1000 Hz_40ms_envelope_correlation": 0.998667896065129, "300–1000 Hz_40ms_modulation_mae_db": 1.2012545859582842, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3777049310721126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09001814707241672, "1000–3000 Hz_10ms_envelope_correlation": 0.99381579834064, "1000–3000 Hz_10ms_modulation_mae_db": 0.3035449974961477, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8162312022459052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0492418330013518, "1000–3000 Hz_40ms_envelope_correlation": 0.998177105029485, "1000–3000 Hz_40ms_modulation_mae_db": 0.3444315961909829, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3390674886992953, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14582607311072426, "3000–6000 Hz_10ms_envelope_correlation": 0.9850976998460673, "3000–6000 Hz_10ms_modulation_mae_db": 0.8565107054559092, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7215920134198502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0753946094659156, "3000–6000 Hz_40ms_envelope_correlation": 0.9963534822667598, "3000–6000 Hz_40ms_modulation_mae_db": 0.9925547756770433, "mr_stft_db": 4.666728838550806, "band_mae_db": 0.11775159582437289, "rms_gain_error_db": -0.04661667957220521, "output_peak": 0.3562087486411689}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6009100935587913, "全帯域_10ms_rms_envelope_nrmse": 0.05129776931261788, "全帯域_10ms_envelope_correlation": 0.9962783665277112, "全帯域_10ms_modulation_mae_db": 0.07160382563346195, "全帯域_40ms_log_rms_mae_db": 0.3529636851051961, "全帯域_40ms_rms_envelope_nrmse": 0.026174680259009483, "全帯域_40ms_envelope_correlation": 0.9989669495553893, "全帯域_40ms_modulation_mae_db": 0.20701022772888225, "300–1000 Hz_10ms_log_rms_mae_db": 1.2994779459620656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07804998140938799, "300–1000 Hz_10ms_envelope_correlation": 0.9942321040503231, "300–1000 Hz_10ms_modulation_mae_db": 0.26202030872700904, "300–1000 Hz_40ms_log_rms_mae_db": 0.804255105069688, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04282867664574273, "300–1000 Hz_40ms_envelope_correlation": 0.9982134328977874, "300–1000 Hz_40ms_modulation_mae_db": 0.438787688487647, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3666517525486954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09783378489172498, "1000–3000 Hz_10ms_envelope_correlation": 0.9926374568426063, "1000–3000 Hz_10ms_modulation_mae_db": 0.33590980225205974, "1000–3000 Hz_40ms_log_rms_mae_db": 0.790098731275593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055748249357769994, "1000–3000 Hz_40ms_envelope_correlation": 0.9975852313494342, "1000–3000 Hz_40ms_modulation_mae_db": 0.2399452395266527, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3994923878383527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16102319024828227, "3000–6000 Hz_10ms_envelope_correlation": 0.982716251045187, "3000–6000 Hz_10ms_modulation_mae_db": 0.5549418823078319, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7794286270541658, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09841308912416581, "3000–6000 Hz_40ms_envelope_correlation": 0.9937043185849604, "3000–6000 Hz_40ms_modulation_mae_db": 0.4390391353385063, "mr_stft_db": 4.671267889766372, "band_mae_db": 0.12832846581987703, "rms_gain_error_db": -0.05158637180363978, "output_peak": 0.35883204363262544}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.568271762109805, "全帯域_10ms_rms_envelope_nrmse": 0.04722783408459668, "全帯域_10ms_envelope_correlation": 0.996863897639422, "全帯域_10ms_modulation_mae_db": 0.6518627067935338, "全帯域_40ms_log_rms_mae_db": 0.3318689636878533, "全帯域_40ms_rms_envelope_nrmse": 0.023769644819330457, "全帯域_40ms_envelope_correlation": 0.9991597106127291, "全帯域_40ms_modulation_mae_db": 0.77647039386716, "300–1000 Hz_10ms_log_rms_mae_db": 1.1707858149290216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0669676832016382, "300–1000 Hz_10ms_envelope_correlation": 0.9957687079430351, "300–1000 Hz_10ms_modulation_mae_db": 0.6923644584289322, "300–1000 Hz_40ms_log_rms_mae_db": 0.7163599898194818, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03556802437060149, "300–1000 Hz_40ms_envelope_correlation": 0.9987484673741753, "300–1000 Hz_40ms_modulation_mae_db": 0.7780690672187996, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3636668611889833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10260842114177918, "1000–3000 Hz_10ms_envelope_correlation": 0.9920640439263048, "1000–3000 Hz_10ms_modulation_mae_db": 0.5820932454430335, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8267357785852949, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06140516664245015, "1000–3000 Hz_40ms_envelope_correlation": 0.9972790554087946, "1000–3000 Hz_40ms_modulation_mae_db": 0.49241708843654647, "3000–6000 Hz_10ms_log_rms_mae_db": 1.39780484584663, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14904825875094918, "3000–6000 Hz_10ms_envelope_correlation": 0.9842028063310282, "3000–6000 Hz_10ms_modulation_mae_db": 0.150785784468337, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7870530536421662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08325467154655632, "3000–6000 Hz_40ms_envelope_correlation": 0.9948187563450506, "3000–6000 Hz_40ms_modulation_mae_db": 0.25300090722580904, "mr_stft_db": 4.659520471173746, "band_mae_db": 0.09391787088164306, "rms_gain_error_db": -0.053968682972771076, "output_peak": 0.3730404381937575}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.619192406007867, "spectral_convergence": 0.1250381400241198, "spectral_convergence_status": "ok", "amplitude_floor": 1.5362649016028353e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.542594240381286, "spectral_convergence": 0.11798484638164541, "spectral_convergence_status": "ok", "amplitude_floor": 1.0749003550909146e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.852351143730232, "spectral_convergence": 0.11507687761882872, "spectral_convergence_status": "ok", "amplitude_floor": 8.625899017329306e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001501246216341323, "output_power": 0.001487806104716339, "delta_db": -0.03905589481695845, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002746582421816528, "output_power": 0.0026923419506475472, "delta_db": -0.08662418036875774, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011276426514964012, "output_power": 0.0011130406511091462, "delta_db": -0.05660467911274187, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016504304929751857, "output_power": 0.00016216190221285127, "delta_db": -0.07648408831417985, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.934184208485853e-05, "output_power": 4.085985637027234e-05, "delta_db": 0.16442144847207696, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8643647097645308e-05, "output_power": 2.787073463430461e-05, "delta_db": -0.11879898117413608, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09033154537647516, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0658665715738181, "pitch": null, "output_rms_dbfs": -22.57739946255912, "output_peak": 0.36376530911114946, "reference_rms_dbfs": -22.511532890985304, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.671379263373129}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018834661138467316, "log_rms_mae_db": 0.5521917814691599, "rms_envelope_nrmse": 0.04663634632151297, "envelope_correlation": 0.9969620693369189, "rms_gain_error_db": -0.06368342882665715, "reference_mean_power": 0.005606081823103195, "output_mean_power": 0.005524475975695219}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7511238572638392e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009260227506744139, "output_power": 0.000897284463828203, "reference_active": true, "delta_db": -0.13691508472464195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010953814679795108, "output_power": 9.562887594565074e-05, "reference_active": true, "delta_db": -0.5897633815830969}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.583845526426034e-05, "output_power": 2.0123439715404936e-05, "reference_active": true, "delta_db": -1.0856432905061768}], "active_band_count": 3, "active_band_mae_db": 0.6041072522713052}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015988223985812614, "log_rms_mae_db": 0.32094158496361824, "rms_envelope_nrmse": 0.024606287352781343, "envelope_correlation": 0.9991201644368196, "rms_gain_error_db": -0.062352533185883824, "reference_mean_power": 0.00558793457857638, "output_mean_power": 0.005508280648743205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5615527006220852e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008311543470725633, "output_power": 0.0008059567452823646, "reference_active": true, "delta_db": -0.13369946278139963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.497505864184299e-05, "output_power": 4.894834158297606e-05, "reference_active": true, "delta_db": -0.5042771922209675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0542332226205543e-06, "output_power": 1.419936780737449e-06, "reference_active": true, "delta_db": -1.6038073978205405}], "active_band_count": 3, "active_band_mae_db": 0.7472613509409692}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016200114302549476, "log_rms_mae_db": 1.2222219904856417, "rms_envelope_nrmse": 0.07589119890223854, "envelope_correlation": 0.9945476143764544, "rms_gain_error_db": -0.08124050405781297, "reference_mean_power": 0.0020825301889685474, "output_mean_power": 0.0020439358229983233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.598844168109503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038009245216565237, "output_power": 0.0003630089975792087, "reference_active": true, "delta_db": -0.1997185587816206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.129961338774397e-05, "output_power": 4.933751149386634e-05, "reference_active": true, "delta_db": -0.9428049488438439}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4399990501668035e-05, "output_power": 1.0220529966466894e-05, "reference_active": true, "delta_db": -1.4888878972103514}], "active_band_count": 3, "active_band_mae_db": 0.877137134945272}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012179662863477574, "log_rms_mae_db": 0.7535997130819952, "rms_envelope_nrmse": 0.042434298261059905, "envelope_correlation": 0.9982025734573696, "rms_gain_error_db": -0.0789238944643971, "reference_mean_power": 0.0020746475669658963, "output_mean_power": 0.002037285722485466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.889619852823582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035505972043036534, "output_power": 0.00033711378383647207, "reference_active": true, "delta_db": -0.225248965549778}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2260222284985983e-05, "output_power": 2.5504761852348587e-05, "reference_active": true, "delta_db": -1.0204608280095289}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1886983564224358e-06, "output_power": 7.386167138102094e-07, "reference_active": true, "delta_db": -2.0665253134886656}], "active_band_count": 3, "active_band_mae_db": 1.1040783690159908}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.2267966954040695e-05, "log_rms_mae_db": 1.2921279760736724, "rms_envelope_nrmse": 0.09130262006272061, "envelope_correlation": 0.9935468903052089, "rms_gain_error_db": -0.07426011287987602, "reference_mean_power": 0.00015543877011727877, "output_mean_power": 0.00015280351339321227}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4080839081981722e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.101806003601554e-05, "output_power": 6.977177317841938e-05, "reference_active": true, "delta_db": -0.07689044457131924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0132703192354966e-06, "output_power": 1.8221228553570916e-06, "reference_active": true, "delta_db": -0.4332443534399856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.589135119213016e-07, "output_power": 7.02056415018312e-07, "reference_active": true, "delta_db": -0.3382027314914817}], "active_band_count": 3, "active_band_mae_db": 0.2827791765009288}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.184196106800431e-05, "log_rms_mae_db": 0.7468802187757809, "rms_envelope_nrmse": 0.047719745921019144, "envelope_correlation": 0.9982057740199979, "rms_gain_error_db": -0.08385746130652094, "reference_mean_power": 0.00015383740434286893, "output_mean_power": 0.0001508954683726496}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3572636139953185e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.549215276795424e-05, "output_power": 6.446492669965738e-05, "reference_active": true, "delta_db": -0.06865773011811505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.98253690709034e-07, "output_power": 9.333445389620865e-07, "reference_active": true, "delta_db": -0.291989337837022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.830181704119583e-08, "output_power": 4.5015631314616964e-08, "reference_active": true, "delta_db": -0.3060012334174876}], "active_band_count": 3, "active_band_mae_db": 0.22221610045754156}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.781014610300033e-05, "log_rms_mae_db": 1.3681006850496462, "rms_envelope_nrmse": 0.15197590744467157, "envelope_correlation": 0.9838611538127032, "rms_gain_error_db": 0.10584684598587096, "reference_mean_power": 2.2945322907755735e-05, "output_mean_power": 2.3511419936082695e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.468236349213924e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8725122830509875e-06, "output_power": 3.7561142116000325e-06, "reference_active": true, "delta_db": -0.13254014631095307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.737803238621202e-07, "output_power": 4.6204322427403906e-07, "reference_active": true, "delta_db": -0.9406304546325885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.433384337397866e-07, "output_power": 1.280978944436967e-07, "reference_active": true, "delta_db": -0.4882066335322539}], "active_band_count": 3, "active_band_mae_db": 0.5204590781585985}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.4952988731037596e-05, "log_rms_mae_db": 0.7738120399959628, "rms_envelope_nrmse": 0.08188857218160667, "envelope_correlation": 0.9951657257687594, "rms_gain_error_db": 0.10697344536695817, "reference_mean_power": 2.2958960928334722e-05, "output_mean_power": 2.3531497920847185e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.967813463196935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5837220182078332e-06, "output_power": 3.463721495102741e-06, "reference_active": true, "delta_db": -0.14791350232152706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.664756741318262e-07, "output_power": 2.0485566434985377e-07, "reference_active": true, "delta_db": -1.1420959267334747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1431200548978765e-08, "output_power": 8.32171746147525e-09, "reference_active": true, "delta_db": -1.3787887751676842}], "active_band_count": 3, "active_band_mae_db": 0.8895994014075619}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.959238171730742, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9c1ed4068896305289f5e7755b5183b0a5410d62d60192b4619c7557875a6d42", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9931648823265145, "standard_deviation": 0.0006590655590376884, "minimum": 0.9920640439263048, "maximum": 0.9938771412802773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3377790119305626, "standard_deviation": 0.038476737944202824, "minimum": 1.2899793853208497, "maximum": 1.3777049310721126, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.29449639275306116, "standard_deviation": 0.16267502860759883, "minimum": 0.12437079886207493, "maximum": 0.5820932454430335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09473469900976039, "standard_deviation": 0.004584897079573074, "minimum": 0.09001814707241672, "maximum": 0.10260842114177918, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9979912963608677, "standard_deviation": 0.00039119305786015636, "minimum": 0.9972790554087946, "maximum": 0.9983675857712719, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7864549606381344, "standard_deviation": 0.034721433730563966, "minimum": 0.7415970661654862, "maximum": 0.8287879138354143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.31442451253415216, "standard_deviation": 0.13870314438487594, "minimum": 0.16693122052776765, "maximum": 0.5428362278455164, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.052239144724362675, "standard_deviation": 0.00459734833366477, "minimum": 0.047719745921019144, "maximum": 0.06140516664245015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9841212243325252, "standard_deviation": 0.0010768137723861817, "minimum": 0.982716251045187, "maximum": 0.985644778964924, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3831516590690291, "standard_deviation": 0.022402632393635904, "minimum": 1.3390674886992953, "maximum": 1.405874805129966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6280301626077116, "standard_deviation": 0.27236677498458883, "minimum": 0.150785784468337, "maximum": 1.0357754215273776, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15018890170069776, "standard_deviation": 0.006273632728070752, "minimum": 0.14163613971571656, "maximum": 0.16102319024828227, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955149272450907, "standard_deviation": 0.000938151338799521, "minimum": 0.9937043185849604, "maximum": 0.9964443209570926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7810200656095647, "standard_deviation": 0.02819002348602346, "minimum": 0.7215920134198502, "maximum": 0.8217237820565545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7744338664962782, "standard_deviation": 0.40010573760456003, "minimum": 0.25300090722580904, "maximum": 1.4653155918661316, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07910955213780463, "standard_deviation": 0.009167583236389318, "minimum": 0.06880457743146232, "maximum": 0.09841308912416581, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9950735996809312, "standard_deviation": 0.0005859814264283142, "minimum": 0.9942321040503231, "maximum": 0.9957687079430351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.221762663894665, "standard_deviation": 0.039383204374388414, "minimum": 1.1707858149290216, "maximum": 1.2994779459620656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8329642223943078, "standard_deviation": 0.3156846063354685, "minimum": 0.26202030872700904, "maximum": 1.245382099382771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07235671519446402, "standard_deviation": 0.004375159894200119, "minimum": 0.0669676832016382, "maximum": 0.07804998140938799, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998362852835688, "standard_deviation": 0.0002627611525608652, "minimum": 0.9979708622182983, "maximum": 0.9987484673741753, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7492875523500542, "standard_deviation": 0.025532545510819696, "minimum": 0.7163599898194818, "maximum": 0.804255105069688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0069688752182273, "standard_deviation": 0.3156643716587649, "minimum": 0.438787688487647, "maximum": 1.4409502423043605, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04097109607721325, "standard_deviation": 0.0038114948448337046, "minimum": 0.03556802437060149, "maximum": 0.047212303248426506, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1062855548212373, "standard_deviation": 0.013238733142680871, "minimum": 0.09033154537647516, "maximum": 0.12832846581987703, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.668315162564307, "standard_deviation": 0.004715061052917676, "minimum": 4.659520471173746, "maximum": 4.672479064188458, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3621394626988191, "standard_deviation": 0.010367962729739343, "minimum": 0.34952213737516075, "maximum": 0.3798594502259434, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06306025419552666, "standard_deviation": 0.01658763416461563, "minimum": -0.08647863690550009, "maximum": -0.04356278056548295, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9969337258264225, "standard_deviation": 0.00032132081156683077, "minimum": 0.9962783665277112, "maximum": 0.9973750504540898, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5668642180306609, "standard_deviation": 0.01699418488339359, "minimum": 0.5481838396668507, "maximum": 0.6009100935587913, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5068307519851555, "standard_deviation": 0.1962002030171538, "minimum": 0.07160382563346195, "maximum": 0.7051496784094057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04677526625634, "standard_deviation": 0.0023143975330923196, "minimum": 0.04308560327417095, "maximum": 0.05129776931261788, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991180030940172, "standard_deviation": 0.00013681442771754264, "minimum": 0.9989669495553893, "maximum": 0.9994136898258398, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3264486011070065, "standard_deviation": 0.013156575862047758, "minimum": 0.3124278732509115, "maximum": 0.3529636851051961, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6093694465784462, "standard_deviation": 0.25224361387481914, "minimum": 0.20701022772888225, "maximum": 0.9611664943943238, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.024561530774573556, "standard_deviation": 0.002171161793254875, "minimum": 0.019806257625387114, "maximum": 0.026336217203303255, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2094-142345-0032", "language": "en", "style": "normal", "speaker": "librispeech_2094", "source_utterance_id": "2094-142345-0032", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 51840, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2094-142345-0032.wav", "source_file_sha256": "5c22bcc2a023e7e49df35841d9876e2b41eb68eef63b8454b62e363e085605cb", "decoded_pcm_sha256": "973bdb09bc9497c63b4dfbb3bc77778c87a6ced88d08e7f2f6e22739dd6d5444", "original_source_sha256": "b92bf752cc5264a2a47e0153c236b90edb48dd32b819cfd12d0e5250ca3c7a1e", "corpus_entry_id": "librispeech_2094-142345-0032"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.497432708740234, "predicted_mos": 4.497432708740234, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "80ff88c720d36465c43098db2f00c593a7018953709ca1eca7721afef2c062ba", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.7075543719984125, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2094-142345-0032", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "80ff88c720d36465c43098db2f00c593a7018953709ca1eca7721afef2c062ba", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "8096cbcdd3e093fefedfc7d4466fc50ab9467c5c11071c10b83dc920b31f9ec9"}, "rows": [{"case": "librispeech_2094-142345-0032", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.135708808898926, "predicted_mos": 3.135708808898926, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "df1a5be92df5b80c519c39081a6c264b7e97862feb3bc9ecbdd7d79dcfb4de5a", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.291473779998341, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2094-142345-0032", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "df1a5be92df5b80c519c39081a6c264b7e97862feb3bc9ecbdd7d79dcfb4de5a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0881212290599356, "全帯域_10ms_rms_envelope_nrmse": 0.07124370796898839, "全帯域_10ms_envelope_correlation": 0.9923899449678414, "全帯域_10ms_modulation_mae_db": 0.26248496604001054, "全帯域_40ms_log_rms_mae_db": 0.5902544148134339, "全帯域_40ms_rms_envelope_nrmse": 0.03948563777659963, "全帯域_40ms_envelope_correlation": 0.9973629626359796, "全帯域_40ms_modulation_mae_db": 0.1325763557068889, "300–1000 Hz_10ms_log_rms_mae_db": 1.361533007483515, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10381548055922461, "300–1000 Hz_10ms_envelope_correlation": 0.9892934965177239, "300–1000 Hz_10ms_modulation_mae_db": 0.79820809608843, "300–1000 Hz_40ms_log_rms_mae_db": 0.8891281936141858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061567240668456286, "300–1000 Hz_40ms_envelope_correlation": 0.9959841750153764, "300–1000 Hz_40ms_modulation_mae_db": 0.6865420529417383, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4553390711278265, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10125139861320143, "1000–3000 Hz_10ms_envelope_correlation": 0.9925443265708241, "1000–3000 Hz_10ms_modulation_mae_db": 0.9635096051143378, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9598652222798998, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05547218134143353, "1000–3000 Hz_40ms_envelope_correlation": 0.9976323642728484, "1000–3000 Hz_40ms_modulation_mae_db": 0.8544499699307008, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4240481005056354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13505158878912388, "3000–6000 Hz_10ms_envelope_correlation": 0.9880989276276166, "3000–6000 Hz_10ms_modulation_mae_db": 0.6530078625117244, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8859366943939232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0736113061573928, "3000–6000 Hz_40ms_envelope_correlation": 0.9964865462986998, "3000–6000 Hz_40ms_modulation_mae_db": 0.5331164421115563, "mr_stft_db": 5.30251936544551, "band_mae_db": 0.0526883460789735, "rms_gain_error_db": -0.04121402026461257, "output_peak": 0.5260552826234154}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0124641037990454, "全帯域_10ms_rms_envelope_nrmse": 0.06483758584283271, "全帯域_10ms_envelope_correlation": 0.993815288895616, "全帯域_10ms_modulation_mae_db": 0.2739671765343208, "全帯域_40ms_log_rms_mae_db": 0.5025913925241284, "全帯域_40ms_rms_envelope_nrmse": 0.03691016414265645, "全帯域_40ms_envelope_correlation": 0.9978575042347019, "全帯域_40ms_modulation_mae_db": 0.5333189342260032, "300–1000 Hz_10ms_log_rms_mae_db": 1.2993830343306294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0854993896002495, "300–1000 Hz_10ms_envelope_correlation": 0.9928936465800976, "300–1000 Hz_10ms_modulation_mae_db": 1.078569522895837, "300–1000 Hz_40ms_log_rms_mae_db": 0.7988223272554333, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051519003994712875, "300–1000 Hz_40ms_envelope_correlation": 0.9973204573152702, "300–1000 Hz_40ms_modulation_mae_db": 1.377406647990485, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4535885107304463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09559463151109024, "1000–3000 Hz_10ms_envelope_correlation": 0.9940064417504473, "1000–3000 Hz_10ms_modulation_mae_db": 0.7459885118647009, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9754798208521573, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06529271889464656, "1000–3000 Hz_40ms_envelope_correlation": 0.997519674420678, "1000–3000 Hz_40ms_modulation_mae_db": 0.5984405154281204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4810568788247855, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15453256057879325, "3000–6000 Hz_10ms_envelope_correlation": 0.9844317317199882, "3000–6000 Hz_10ms_modulation_mae_db": 0.8086822646950829, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8991214689868101, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07915942121295479, "3000–6000 Hz_40ms_envelope_correlation": 0.9957817260235221, "3000–6000 Hz_40ms_modulation_mae_db": 0.904287319193018, "mr_stft_db": 5.2896190103004015, "band_mae_db": 0.1232675457334092, "rms_gain_error_db": -0.09861507560837411, "output_peak": 0.5666142061894108}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0349573030801977, "全帯域_10ms_rms_envelope_nrmse": 0.060455967356964194, "全帯域_10ms_envelope_correlation": 0.9946113354503106, "全帯域_10ms_modulation_mae_db": 0.45640649484478885, "全帯域_40ms_log_rms_mae_db": 0.5130040978906255, "全帯域_40ms_rms_envelope_nrmse": 0.03302979754064846, "全帯域_40ms_envelope_correlation": 0.9983061013673887, "全帯域_40ms_modulation_mae_db": 0.5655431806866512, "300–1000 Hz_10ms_log_rms_mae_db": 1.433303655123464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08964769205913325, "300–1000 Hz_10ms_envelope_correlation": 0.9920563484178928, "300–1000 Hz_10ms_modulation_mae_db": 0.6724038674968541, "300–1000 Hz_40ms_log_rms_mae_db": 0.7998796398812463, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04768293756953036, "300–1000 Hz_40ms_envelope_correlation": 0.9976359658623859, "300–1000 Hz_40ms_modulation_mae_db": 0.9530643954262653, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4755135352205453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09161124633500713, "1000–3000 Hz_10ms_envelope_correlation": 0.9948025703668167, "1000–3000 Hz_10ms_modulation_mae_db": 1.4293168103288403, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0225294860126164, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06541330548067532, "1000–3000 Hz_40ms_envelope_correlation": 0.9977510253847139, "1000–3000 Hz_40ms_modulation_mae_db": 1.3316140707980244, "3000–6000 Hz_10ms_log_rms_mae_db": 1.383761202043949, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1237274630685056, "3000–6000 Hz_10ms_envelope_correlation": 0.9899900621726953, "3000–6000 Hz_10ms_modulation_mae_db": 1.0344164265023592, "3000–6000 Hz_40ms_log_rms_mae_db": 0.815527236713455, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06312819780218543, "3000–6000 Hz_40ms_envelope_correlation": 0.9973850385643637, "3000–6000 Hz_40ms_modulation_mae_db": 1.2860501098895905, "mr_stft_db": 5.308775243062529, "band_mae_db": 0.17328591888175682, "rms_gain_error_db": -0.09161642694026051, "output_peak": 0.48557475809561323}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.030523671117341, "全帯域_10ms_rms_envelope_nrmse": 0.06076375754771318, "全帯域_10ms_envelope_correlation": 0.9947020513921128, "全帯域_10ms_modulation_mae_db": 0.42911363806229536, "全帯域_40ms_log_rms_mae_db": 0.49408408897046147, "全帯域_40ms_rms_envelope_nrmse": 0.034774247994155245, "全帯域_40ms_envelope_correlation": 0.9982837496276419, "全帯域_40ms_modulation_mae_db": 0.5443842569849515, "300–1000 Hz_10ms_log_rms_mae_db": 1.3937890307019596, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0873834954018524, "300–1000 Hz_10ms_envelope_correlation": 0.9926199887606436, "300–1000 Hz_10ms_modulation_mae_db": 0.8985682420891666, "300–1000 Hz_40ms_log_rms_mae_db": 0.816889309888865, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048261302586725646, "300–1000 Hz_40ms_envelope_correlation": 0.997759051175366, "300–1000 Hz_40ms_modulation_mae_db": 1.2390242568321088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4890354852758754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09247987594502248, "1000–3000 Hz_10ms_envelope_correlation": 0.9942116751095533, "1000–3000 Hz_10ms_modulation_mae_db": 0.8451786917539562, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0653007163318189, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05843547330997345, "1000–3000 Hz_40ms_envelope_correlation": 0.997916702019632, "1000–3000 Hz_40ms_modulation_mae_db": 0.7597465656423998, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4669831245079992, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17437883695015918, "3000–6000 Hz_10ms_envelope_correlation": 0.9803707721123022, "3000–6000 Hz_10ms_modulation_mae_db": 0.14755946079907506, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8379620039636707, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07304054271332912, "3000–6000 Hz_40ms_envelope_correlation": 0.9964730990781768, "3000–6000 Hz_40ms_modulation_mae_db": 0.2792806955677795, "mr_stft_db": 5.317279699637116, "band_mae_db": 0.13042148403638118, "rms_gain_error_db": -0.12967175079199056, "output_peak": 0.5557677592592766}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1179468383174818, "全帯域_10ms_rms_envelope_nrmse": 0.06783284501986257, "全帯域_10ms_envelope_correlation": 0.9930749078909635, "全帯域_10ms_modulation_mae_db": 0.23044226244895075, "全帯域_40ms_log_rms_mae_db": 0.6082064218529236, "全帯域_40ms_rms_envelope_nrmse": 0.04006622940771523, "全帯域_40ms_envelope_correlation": 0.9972630659557139, "全帯域_40ms_modulation_mae_db": 0.5088936049199312, "300–1000 Hz_10ms_log_rms_mae_db": 1.3557489972330798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07302285332595913, "300–1000 Hz_10ms_envelope_correlation": 0.9947521697202668, "300–1000 Hz_10ms_modulation_mae_db": 0.48597184711294106, "300–1000 Hz_40ms_log_rms_mae_db": 0.8484631817985291, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04159540843398582, "300–1000 Hz_40ms_envelope_correlation": 0.9981981412211165, "300–1000 Hz_40ms_modulation_mae_db": 0.6074710141896392, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4761331029592346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10526037604754156, "1000–3000 Hz_10ms_envelope_correlation": 0.992188522437392, "1000–3000 Hz_10ms_modulation_mae_db": 0.4510978277338085, "1000–3000 Hz_40ms_log_rms_mae_db": 1.003822899821798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06823083437429782, "1000–3000 Hz_40ms_envelope_correlation": 0.9967804017338449, "1000–3000 Hz_40ms_modulation_mae_db": 0.5596058511258165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.382945400868099, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1261179475300319, "3000–6000 Hz_10ms_envelope_correlation": 0.9898102824535777, "3000–6000 Hz_10ms_modulation_mae_db": 0.6416973328897441, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8134585409009173, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05998184318286858, "3000–6000 Hz_40ms_envelope_correlation": 0.9976740236608262, "3000–6000 Hz_40ms_modulation_mae_db": 1.0821736617430184, "mr_stft_db": 5.29263949266183, "band_mae_db": 0.048809890861920883, "rms_gain_error_db": -0.026472459618231258, "output_peak": 0.49746172977766223}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.018487548876141, "全帯域_10ms_rms_envelope_nrmse": 0.05826793964699958, "全帯域_10ms_envelope_correlation": 0.9949347100900906, "全帯域_10ms_modulation_mae_db": 0.15059715204810928, "全帯域_40ms_log_rms_mae_db": 0.5626051699660747, "全帯域_40ms_rms_envelope_nrmse": 0.029326734821952503, "全帯域_40ms_envelope_correlation": 0.9985961382815783, "全帯域_40ms_modulation_mae_db": 0.3282620908881258, "300–1000 Hz_10ms_log_rms_mae_db": 1.4061965379915602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07513672463345777, "300–1000 Hz_10ms_envelope_correlation": 0.9944283252253019, "300–1000 Hz_10ms_modulation_mae_db": 0.8811080787015557, "300–1000 Hz_40ms_log_rms_mae_db": 0.9064384499753343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03929145197097124, "300–1000 Hz_40ms_envelope_correlation": 0.9983630462137907, "300–1000 Hz_40ms_modulation_mae_db": 0.7997337868726252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4864638462574646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09707949309930716, "1000–3000 Hz_10ms_envelope_correlation": 0.9939054110761945, "1000–3000 Hz_10ms_modulation_mae_db": 0.7614967173748278, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0127062117793832, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06274309954541811, "1000–3000 Hz_40ms_envelope_correlation": 0.9978988184256569, "1000–3000 Hz_40ms_modulation_mae_db": 0.8022637161405196, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4199056390941696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15625923569786648, "3000–6000 Hz_10ms_envelope_correlation": 0.9840369332789706, "3000–6000 Hz_10ms_modulation_mae_db": 1.1693006812122682, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8637899120342691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08992949283383976, "3000–6000 Hz_40ms_envelope_correlation": 0.9947467887003592, "3000–6000 Hz_40ms_modulation_mae_db": 1.1769168730635136, "mr_stft_db": 5.291936248433589, "band_mae_db": 0.14432560816416387, "rms_gain_error_db": -0.05949955270281523, "output_peak": 0.5323591553612169}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.009591898142695, "全帯域_10ms_rms_envelope_nrmse": 0.06476103770786489, "全帯域_10ms_envelope_correlation": 0.9937370486177148, "全帯域_10ms_modulation_mae_db": 0.24630540787283542, "全帯域_40ms_log_rms_mae_db": 0.4760992639914905, "全帯域_40ms_rms_envelope_nrmse": 0.036086618327490515, "全帯域_40ms_envelope_correlation": 0.9978605554771002, "全帯域_40ms_modulation_mae_db": 0.22689323751771062, "300–1000 Hz_10ms_log_rms_mae_db": 1.3661278571693636, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08645489327638672, "300–1000 Hz_10ms_envelope_correlation": 0.9924903415281936, "300–1000 Hz_10ms_modulation_mae_db": 0.23209051446701232, "300–1000 Hz_40ms_log_rms_mae_db": 0.8177967193773438, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049598780208934774, "300–1000 Hz_40ms_envelope_correlation": 0.9972546031113382, "300–1000 Hz_40ms_modulation_mae_db": 0.36482619234339886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4703175909399235, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1187514968800434, "1000–3000 Hz_10ms_envelope_correlation": 0.9902749213665516, "1000–3000 Hz_10ms_modulation_mae_db": 0.813257240497624, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0415624257952913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07166133244676325, "1000–3000 Hz_40ms_envelope_correlation": 0.9968129042626991, "1000–3000 Hz_40ms_modulation_mae_db": 0.6240548792422136, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4298341972377873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1389770328524107, "3000–6000 Hz_10ms_envelope_correlation": 0.9880712674684986, "3000–6000 Hz_10ms_modulation_mae_db": 1.3494506109537674, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7910546655170102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07705125005841627, "3000–6000 Hz_40ms_envelope_correlation": 0.9965196894835089, "3000–6000 Hz_40ms_modulation_mae_db": 0.9762347179555236, "mr_stft_db": 5.293510395737454, "band_mae_db": 0.17267233956335457, "rms_gain_error_db": -0.06603931967386929, "output_peak": 0.4753545006984762}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0619167766045727, "全帯域_10ms_rms_envelope_nrmse": 0.06077727715293547, "全帯域_10ms_envelope_correlation": 0.9944667722529535, "全帯域_10ms_modulation_mae_db": 0.18365397289435512, "全帯域_40ms_log_rms_mae_db": 0.5287274397956154, "全帯域_40ms_rms_envelope_nrmse": 0.03112163895674812, "全帯域_40ms_envelope_correlation": 0.9983867181778765, "全帯域_40ms_modulation_mae_db": 0.23546964213884683, "300–1000 Hz_10ms_log_rms_mae_db": 1.3497718337799895, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08006786325705284, "300–1000 Hz_10ms_envelope_correlation": 0.9936236160779637, "300–1000 Hz_10ms_modulation_mae_db": 0.6488127144362453, "300–1000 Hz_40ms_log_rms_mae_db": 0.8371951168624643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046995030676525405, "300–1000 Hz_40ms_envelope_correlation": 0.9975570515773153, "300–1000 Hz_40ms_modulation_mae_db": 0.758118843102849, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4697644172819913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1066008136468955, "1000–3000 Hz_10ms_envelope_correlation": 0.9923482561613743, "1000–3000 Hz_10ms_modulation_mae_db": 1.7180772719760815, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0120422273895346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07385923930916169, "1000–3000 Hz_40ms_envelope_correlation": 0.9965639792181986, "1000–3000 Hz_40ms_modulation_mae_db": 1.4834837918416006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.419059658988861, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15650632451307855, "3000–6000 Hz_10ms_envelope_correlation": 0.9847381140455842, "3000–6000 Hz_10ms_modulation_mae_db": 1.1199655119885263, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8229884927135882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0746531142322865, "3000–6000 Hz_40ms_envelope_correlation": 0.997635365603457, "3000–6000 Hz_40ms_modulation_mae_db": 0.9026526089630416, "mr_stft_db": 5.299512050751953, "band_mae_db": 0.16936805643842456, "rms_gain_error_db": -0.04842889068006486, "output_peak": 0.5608416741149231}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 51840, "output_samples": 51840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.240875568284266, "spectral_convergence": 0.26158701577290117, "spectral_convergence_status": "ok", "amplitude_floor": 1.3508970735580905e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.159919803686594, "spectral_convergence": 0.22824110684546042, "spectral_convergence_status": "ok", "amplitude_floor": 9.199259876037329e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.506762724365671, "spectral_convergence": 0.2314135327013418, "spectral_convergence_status": "ok", "amplitude_floor": 7.633709769762664e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012491242347822075, "output_power": 0.0012176328702300748, "delta_db": -0.11089270922104544, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0036291285647318994, "output_power": 0.0036046279977220864, "delta_db": -0.02941902081450174, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000626458342936399, "output_power": 0.0006270733517052573, "delta_db": 0.004261478783659207, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008149528003014777, "output_power": 0.0008150804857067676, "delta_db": 0.0006803918364903375, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002174945257645064, "output_power": 0.00021189494815061354, "delta_db": -0.11327727746670838, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023982804336185733, "output_power": 0.0002366682744989548, "delta_db": -0.05759919835143596, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.0526883460789735, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04121402026461257, "pitch": null, "output_rms_dbfs": -21.730847807545388, "output_peak": 0.5260552826234154, "reference_rms_dbfs": -21.689633787280776, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.30251936544551}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 51840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00019883564266349852, "log_rms_mae_db": 1.0881212290599356, "rms_envelope_nrmse": 0.07124370796898839, "envelope_correlation": 0.9923899449678414, "rms_gain_error_db": -0.041219866261430906, "reference_mean_power": 0.006787407573323676, "output_mean_power": 0.0067232915124513576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.1392663194097437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011382823319318316, "output_power": 0.0011782963023449957, "reference_active": true, "delta_db": 0.15004519890816992}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00037355144462451696, "output_power": 0.0003467403515830079, "reference_active": true, "delta_db": -0.32346035072496776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.066072767849799e-05, "output_power": 5.6430341811544564e-05, "reference_active": true, "delta_db": -0.3139493484868939}], "active_band_count": 3, "active_band_mae_db": 0.26248496604001054}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00017748806718771712, "log_rms_mae_db": 0.5902544148134339, "rms_envelope_nrmse": 0.03948563777659963, "envelope_correlation": 0.9973629626359796, "rms_gain_error_db": -0.0412568116141055, "reference_mean_power": 0.006850619249749789, "output_mean_power": 0.006785848343475844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.7219585966772077e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000998012429960962, "output_power": 0.0010318548047466337, "reference_active": true, "delta_db": 0.1448264043105299}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018464455345700424, "output_power": 0.0001750359252790533, "reference_active": true, "delta_db": -0.23209306915528807}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.884647670701786e-06, "output_power": 4.861298445822237e-06, "reference_active": true, "delta_db": -0.02080959365484868}], "active_band_count": 3, "active_band_mae_db": 0.1325763557068889}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015076905043222532, "log_rms_mae_db": 1.361533007483515, "rms_envelope_nrmse": 0.10381548055922461, "envelope_correlation": 0.9892934965177239, "rms_gain_error_db": -0.12532479261385007, "reference_mean_power": 0.0024151558510913072, "output_mean_power": 0.0023464574429896187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.3965349131952106e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007806812222828868, "output_power": 0.0007416611437823108, "reference_active": true, "delta_db": -0.22268206835583537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024651476965397817, "output_power": 0.00021623827164905142, "reference_active": true, "delta_db": -0.5691038318389189}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0716739338834284e-05, "output_power": 2.123690740660474e-05, "reference_active": true, "delta_db": -1.6028383880705357}], "active_band_count": 3, "active_band_mae_db": 0.79820809608843}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00011516776273837431, "log_rms_mae_db": 0.8891281936141858, "rms_envelope_nrmse": 0.061567240668456286, "envelope_correlation": 0.9959841750153764, "rms_gain_error_db": -0.12533406678313555, "reference_mean_power": 0.0024377564716401134, "output_mean_power": 0.0023684101377961946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.190139920080037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007614290342229629, "output_power": 0.000709373927440509, "reference_active": true, "delta_db": -0.3075421074638229}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011269364256369602, "output_power": 0.00010205709615425097, "reference_active": true, "delta_db": -0.43056209563978814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5333706288121526e-06, "output_power": 2.6063755868044444e-06, "reference_active": true, "delta_db": -1.3215219557216038}], "active_band_count": 3, "active_band_mae_db": 0.6865420529417383}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00014489100768405854, "log_rms_mae_db": 1.4553390711278265, "rms_envelope_nrmse": 0.10125139861320143, "envelope_correlation": 0.9925443265708241, "rms_gain_error_db": -0.04506201790335687, "reference_mean_power": 0.0008223932804627593, "output_mean_power": 0.0008139043158159303}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 6.859948035043385e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004581143880628001, "output_power": 0.00046184139894729924, "reference_active": true, "delta_db": 0.03518928060530434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.242911398972428e-05, "output_power": 8.049557180797716e-05, "reference_active": true, "delta_db": -0.6003680020029751}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.706958400708153e-06, "output_power": 5.1804718274898825e-06, "reference_active": true, "delta_db": -2.254971532734734}], "active_band_count": 3, "active_band_mae_db": 0.9635096051143378}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00013523948700125142, "log_rms_mae_db": 0.9598652222798998, "rms_envelope_nrmse": 0.05547218134143353, "envelope_correlation": 0.9976323642728484, "rms_gain_error_db": -0.045064733011605594, "reference_mean_power": 0.000830089539077811, "output_mean_power": 0.0008215206179836579}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 6.395142215880047e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004454767193500814, "output_power": 0.00044311028288498397, "reference_active": true, "delta_db": -0.023131842479441035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.066351786105847e-05, "output_power": 3.638631301710505e-05, "reference_active": true, "delta_db": -0.4826689612558505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.874405735020567e-07, "output_power": 5.525662687854964e-07, "reference_active": true, "delta_db": -2.0575491060568107}], "active_band_count": 3, "active_band_mae_db": 0.8544499699307008}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 8.841298696090765e-05, "log_rms_mae_db": 1.4240481005056354, "rms_envelope_nrmse": 0.13505158878912388, "envelope_correlation": 0.9880989276276166, "rms_gain_error_db": -0.17112102372273744, "reference_mean_power": 0.0002494763103331033, "output_mean_power": 0.00023983756705918684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.730062412281665e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013367937231840985, "output_power": 0.00012939214192896604, "reference_active": true, "delta_db": -0.14156495552411957}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8654341368924343e-05, "output_power": 2.1393568904066017e-05, "reference_active": true, "delta_db": -1.2690719027506312}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.319126391736523e-06, "output_power": 7.332280143111151e-06, "reference_active": true, "delta_db": -0.5483867292604224}], "active_band_count": 3, "active_band_mae_db": 0.6530078625117244}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 7.608900761326353e-05, "log_rms_mae_db": 0.8859366943939232, "rms_envelope_nrmse": 0.0736113061573928, "envelope_correlation": 0.9964865462986998, "rms_gain_error_db": -0.17112318033512025, "reference_mean_power": 0.00025181099859470626, "output_mean_power": 0.00024208193231191083}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 2.5627299576593646e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001307843821160996, "output_power": 0.00012526644479669534, "reference_active": true, "delta_db": -0.18721133092813308}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1887209147638657e-05, "output_power": 8.952196337962027e-06, "reference_active": true, "delta_db": -1.2315030540623886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.396397638044017e-07, "output_power": 6.668051545656194e-07, "reference_active": true, "delta_db": 0.18063494134414718}], "active_band_count": 3, "active_band_mae_db": 0.5331164421115563}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8918966930814836, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "273919ad4872780137c2906fd20c12cde1f8f21f870d33091199413eeb88c933", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9930352656048942, "standard_deviation": 0.0014763429289780356, "minimum": 0.9902749213665516, "maximum": 0.9948025703668167, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4720194449741635, "standard_deviation": 0.012835402327973486, "minimum": 1.4535885107304463, "maximum": 1.4890354852758754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9659903345805222, "standard_deviation": 0.4094539786982359, "minimum": 0.4510978277338085, "maximum": 1.7180772719760815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1010786665097636, "standard_deviation": 0.009023610657713163, "minimum": 0.09161124633500713, "maximum": 0.1187514968800434, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997359483717284, "standard_deviation": 0.000550547305908702, "minimum": 0.9965639792181986, "maximum": 0.997916702019632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0116636262828125, "standard_deviation": 0.03368762446852171, "minimum": 0.9598652222798998, "maximum": 1.0653007163318189, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8767074200186744, "standard_deviation": 0.34571952059836564, "minimum": 0.5596058511258165, "maximum": 1.4834837918416006, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06513852308779622, "standard_deviation": 0.00623775024895938, "minimum": 0.05547218134143353, "maximum": 0.07385923930916169, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9861935113599042, "standard_deviation": 0.003347250056424662, "minimum": 0.9803707721123022, "maximum": 0.9899900621726953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4259492752589107, "standard_deviation": 0.03474908205128173, "minimum": 1.382945400868099, "maximum": 1.4810568788247855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8655100189440683, "standard_deviation": 0.38427964483688426, "minimum": 0.14755946079907506, "maximum": 1.3494506109537674, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1456938737474962, "standard_deviation": 0.017540156904637498, "minimum": 0.1237274630685056, "maximum": 0.17437883695015918, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9965877846766142, "standard_deviation": 0.0009985697184635152, "minimum": 0.9947467887003592, "maximum": 0.9976740236608262, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8412298769029555, "standard_deviation": 0.03808250984688182, "minimum": 0.7910546655170102, "maximum": 0.8991214689868101, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8925890535608803, "standard_deviation": 0.33470640814099945, "minimum": 0.2792806955677795, "maximum": 1.2860501098895905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07381939602415916, "standard_deviation": 0.00930146345407957, "minimum": 0.05998184318286858, "maximum": 0.08992949283383976, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927697416035105, "standard_deviation": 0.0016953328783836992, "minimum": 0.9892934965177239, "maximum": 0.9947521697202668, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.370731744226695, "standard_deviation": 0.04064635914218525, "minimum": 1.2993830343306294, "maximum": 1.433303655123464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7119666104110053, "standard_deviation": 0.26516103469264657, "minimum": 0.23209051446701232, "maximum": 1.078569522895837, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08512854901416453, "standard_deviation": 0.009625861908471726, "minimum": 0.07302285332595913, "maximum": 0.10381548055922461, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975090614364949, "standard_deviation": 0.0007280720879723306, "minimum": 0.9959841750153764, "maximum": 0.9983630462137907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8393266173316752, "standard_deviation": 0.04006395958122561, "minimum": 0.7988223272554333, "maximum": 0.9064384499753343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8482733987123887, "standard_deviation": 0.3320923574985253, "minimum": 0.36482619234339886, "maximum": 1.377406647990485, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0483138945137303, "standard_deviation": 0.00672158188619987, "minimum": 0.03929145197097124, "maximum": 0.061567240668456286, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12685489871979808, "standard_deviation": 0.05068806875793076, "minimum": 0.048809890861920883, "maximum": 0.17328591888175682, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.299473938253798, "standard_deviation": 0.009629162162138067, "minimum": 5.2896190103004015, "maximum": 5.317279699637116, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5250036332649992, "standard_deviation": 0.035447496547445966, "minimum": 0.4753545006984762, "maximum": 0.5666142061894108, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07019468703502729, "standard_deviation": 0.03414474572500586, "minimum": -0.12967175079199056, "maximum": -0.026472459618231258, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939665074447004, "standard_deviation": 0.0008864477026588173, "minimum": 0.9923899449678414, "maximum": 0.9949347100900906, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0467511711246762, "standard_deviation": 0.039220938808344985, "minimum": 1.009591898142695, "maximum": 1.1179468383174818, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2791213838432083, "standard_deviation": 0.10909371659873388, "minimum": 0.15059715204810928, "maximum": 0.45640649484478885, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06361751478052012, "standard_deviation": 0.004367874049769203, "minimum": 0.05826793964699958, "maximum": 0.07124370796898839, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979895994697476, "standard_deviation": 0.00048798661234736925, "minimum": 0.9972630659557139, "maximum": 0.9985961382815783, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5344465362255941, "standard_deviation": 0.047642592627533314, "minimum": 0.4760992639914905, "maximum": 0.6082064218529236, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38441766288363866, "standard_deviation": 0.17305596242597882, "minimum": 0.1325763557068889, "maximum": 0.5655431806866512, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03510013362099577, "standard_deviation": 0.003811583155158068, "minimum": 0.029326734821952503, "maximum": 0.04006622940771523, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0032", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.9568934440612793, "predicted_mos": 2.9568934440612793, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "e2b4d86bc71d6e077da42b93dd3272cb04d0fa4e8681d198118e6120b0fc0edc", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.583090499974787, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0867071124683438, "全帯域_10ms_rms_envelope_nrmse": 0.07315190772339793, "全帯域_10ms_envelope_correlation": 0.9919697716857065, "全帯域_10ms_modulation_mae_db": 0.35817111045195665, "全帯域_40ms_log_rms_mae_db": 0.6280982674569836, "全帯域_40ms_rms_envelope_nrmse": 0.04116490934705, "全帯域_40ms_envelope_correlation": 0.9971274798078531, "全帯域_40ms_modulation_mae_db": 0.2769895000829663, "300–1000 Hz_10ms_log_rms_mae_db": 1.3727644898171716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10891897951620277, "300–1000 Hz_10ms_envelope_correlation": 0.988164648148806, "300–1000 Hz_10ms_modulation_mae_db": 0.9373297785359943, "300–1000 Hz_40ms_log_rms_mae_db": 0.7919593690511844, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06509135340835842, "300–1000 Hz_40ms_envelope_correlation": 0.9954427469092084, "300–1000 Hz_40ms_modulation_mae_db": 0.9047093803914454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4367590898897338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10382119560261602, "1000–3000 Hz_10ms_envelope_correlation": 0.992160186030262, "1000–3000 Hz_10ms_modulation_mae_db": 1.0524352186459462, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9010475647037118, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05603930547450895, "1000–3000 Hz_40ms_envelope_correlation": 0.9975948569236315, "1000–3000 Hz_40ms_modulation_mae_db": 0.9916071876574485, "3000–6000 Hz_10ms_log_rms_mae_db": 1.368736509821416, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13792533843564148, "3000–6000 Hz_10ms_envelope_correlation": 0.9875474879037814, "3000–6000 Hz_10ms_modulation_mae_db": 0.6157819953793551, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7843312441249393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0742993592682421, "3000–6000 Hz_40ms_envelope_correlation": 0.9963406797415783, "3000–6000 Hz_40ms_modulation_mae_db": 0.4302230459432193, "mr_stft_db": 5.264282000478324, "band_mae_db": 0.060666218117723845, "rms_gain_error_db": -0.041016218673559514, "output_peak": 0.5314651085865792}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.043573918451804, "全帯域_10ms_rms_envelope_nrmse": 0.06657271437770958, "全帯域_10ms_envelope_correlation": 0.9934561768197031, "全帯域_10ms_modulation_mae_db": 0.18411907679774694, "全帯域_40ms_log_rms_mae_db": 0.5241430747509273, "全帯域_40ms_rms_envelope_nrmse": 0.03808379708938671, "全帯域_40ms_envelope_correlation": 0.9977063804896936, "全帯域_40ms_modulation_mae_db": 0.5354253519491451, "300–1000 Hz_10ms_log_rms_mae_db": 1.3226224502090926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09054111645792132, "300–1000 Hz_10ms_envelope_correlation": 0.9919212547244528, "300–1000 Hz_10ms_modulation_mae_db": 1.0045267650873795, "300–1000 Hz_40ms_log_rms_mae_db": 0.7217751737592732, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05220778335729602, "300–1000 Hz_40ms_envelope_correlation": 0.9971607091178067, "300–1000 Hz_40ms_modulation_mae_db": 1.4059686363513473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.445025784040255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09992710403212, "1000–3000 Hz_10ms_envelope_correlation": 0.9935279909584331, "1000–3000 Hz_10ms_modulation_mae_db": 1.0067298972090473, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8774697133797464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06998605627645231, "1000–3000 Hz_40ms_envelope_correlation": 0.9971870098465083, "1000–3000 Hz_40ms_modulation_mae_db": 0.9031724641861626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.44898206578322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15862732502159854, "3000–6000 Hz_10ms_envelope_correlation": 0.983656794356293, "3000–6000 Hz_10ms_modulation_mae_db": 0.7443160270189034, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8706679454619869, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08621256467039641, "3000–6000 Hz_40ms_envelope_correlation": 0.9950678860538126, "3000–6000 Hz_40ms_modulation_mae_db": 0.8734613549024938, "mr_stft_db": 5.259866855029396, "band_mae_db": 0.13479491174965003, "rms_gain_error_db": -0.09846356745983546, "output_peak": 0.5583929144163748}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0504018085195677, "全帯域_10ms_rms_envelope_nrmse": 0.059472007487777125, "全帯域_10ms_envelope_correlation": 0.9947743938456233, "全帯域_10ms_modulation_mae_db": 0.4585987719227053, "全帯域_40ms_log_rms_mae_db": 0.5280367582205888, "全帯域_40ms_rms_envelope_nrmse": 0.03284787044605559, "全帯域_40ms_envelope_correlation": 0.9983050816642677, "全帯域_40ms_modulation_mae_db": 0.5066653200164137, "300–1000 Hz_10ms_log_rms_mae_db": 1.3933661823216419, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09093678428103656, "300–1000 Hz_10ms_envelope_correlation": 0.9917587255414225, "300–1000 Hz_10ms_modulation_mae_db": 0.7021460818428986, "300–1000 Hz_40ms_log_rms_mae_db": 0.6978870823249621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04772184193788469, "300–1000 Hz_40ms_envelope_correlation": 0.9975435860638313, "300–1000 Hz_40ms_modulation_mae_db": 1.0534827124153883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3984841352250288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09557725727502168, "1000–3000 Hz_10ms_envelope_correlation": 0.9943056109047149, "1000–3000 Hz_10ms_modulation_mae_db": 1.6428687753546598, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8903703352218351, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06557310023353245, "1000–3000 Hz_40ms_envelope_correlation": 0.9978081736885773, "1000–3000 Hz_40ms_modulation_mae_db": 1.4681069596929603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3225217195976244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12151692644068122, "3000–6000 Hz_10ms_envelope_correlation": 0.9904159158468856, "3000–6000 Hz_10ms_modulation_mae_db": 1.0954577656167344, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7439589054812458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06289749781010374, "3000–6000 Hz_40ms_envelope_correlation": 0.9975219955480344, "3000–6000 Hz_40ms_modulation_mae_db": 1.2803661831146702, "mr_stft_db": 5.266354614116191, "band_mae_db": 0.19262893067344775, "rms_gain_error_db": -0.08602902098569266, "output_peak": 0.48898222440558453}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0357954219994687, "全帯域_10ms_rms_envelope_nrmse": 0.06095556856764632, "全帯域_10ms_envelope_correlation": 0.9946350017922216, "全帯域_10ms_modulation_mae_db": 0.38160853216304647, "全帯域_40ms_log_rms_mae_db": 0.5306195771732372, "全帯域_40ms_rms_envelope_nrmse": 0.03472583303252581, "全帯域_40ms_envelope_correlation": 0.9982495596786986, "全帯域_40ms_modulation_mae_db": 0.5531648696961852, "300–1000 Hz_10ms_log_rms_mae_db": 1.3573802212175232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09102609704904797, "300–1000 Hz_10ms_envelope_correlation": 0.9919133122516651, "300–1000 Hz_10ms_modulation_mae_db": 0.9515875564193056, "300–1000 Hz_40ms_log_rms_mae_db": 0.7020498251419027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04992117877755474, "300–1000 Hz_40ms_envelope_correlation": 0.9975242954459727, "300–1000 Hz_40ms_modulation_mae_db": 1.321182635276536, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4027112781716828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09170619971082132, "1000–3000 Hz_10ms_envelope_correlation": 0.9941485675462941, "1000–3000 Hz_10ms_modulation_mae_db": 0.9595933189425461, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9150878186830923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054990578036633105, "1000–3000 Hz_40ms_envelope_correlation": 0.9980212160849766, "1000–3000 Hz_40ms_modulation_mae_db": 0.8614309124880517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.432274873654463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17931761400396684, "3000–6000 Hz_10ms_envelope_correlation": 0.9793991469426023, "3000–6000 Hz_10ms_modulation_mae_db": 0.12946527899120075, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8231401868481947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07838056011544421, "3000–6000 Hz_40ms_envelope_correlation": 0.9960336154235715, "3000–6000 Hz_40ms_modulation_mae_db": 0.47495979720530596, "mr_stft_db": 5.281070512546815, "band_mae_db": 0.12542588475469407, "rms_gain_error_db": -0.1226402438130677, "output_peak": 0.5579172805853484}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1431003214877797, "全帯域_10ms_rms_envelope_nrmse": 0.07044654450201357, "全帯域_10ms_envelope_correlation": 0.9925322823839696, "全帯域_10ms_modulation_mae_db": 0.17419483731700183, "全帯域_40ms_log_rms_mae_db": 0.6237499694160861, "全帯域_40ms_rms_envelope_nrmse": 0.04007323116907819, "全帯域_40ms_envelope_correlation": 0.9972581192701266, "全帯域_40ms_modulation_mae_db": 0.4014027513567397, "300–1000 Hz_10ms_log_rms_mae_db": 1.2955891420622092, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0767699720520278, "300–1000 Hz_10ms_envelope_correlation": 0.9941572417725568, "300–1000 Hz_10ms_modulation_mae_db": 0.6355509400082326, "300–1000 Hz_40ms_log_rms_mae_db": 0.6895880697713512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04233788218475626, "300–1000 Hz_40ms_envelope_correlation": 0.9980943662960424, "300–1000 Hz_40ms_modulation_mae_db": 0.8507232387184657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4093885125061592, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10287608109207795, "1000–3000 Hz_10ms_envelope_correlation": 0.9925657745701182, "1000–3000 Hz_10ms_modulation_mae_db": 0.5926896362611953, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9024351926504752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06579836429866859, "1000–3000 Hz_40ms_envelope_correlation": 0.9970345134223241, "1000–3000 Hz_40ms_modulation_mae_db": 0.8491706170776263, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3549745411910181, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1347455242837927, "3000–6000 Hz_10ms_envelope_correlation": 0.9883032830696178, "3000–6000 Hz_10ms_modulation_mae_db": 0.7868994501975811, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7542918964972184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06475565833240891, "3000–6000 Hz_40ms_envelope_correlation": 0.9972557206098711, "3000–6000 Hz_40ms_modulation_mae_db": 1.2939178471479555, "mr_stft_db": 5.259808690763749, "band_mae_db": 0.06584602319561884, "rms_gain_error_db": -0.022955163171748113, "output_peak": 0.5070394910491776}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0778235210365268, "全帯域_10ms_rms_envelope_nrmse": 0.06076218607229931, "全帯域_10ms_envelope_correlation": 0.99448573482031, "全帯域_10ms_modulation_mae_db": 0.14967940641054814, "全帯域_40ms_log_rms_mae_db": 0.5935308021316117, "全帯域_40ms_rms_envelope_nrmse": 0.030765479116898097, "全帯域_40ms_envelope_correlation": 0.9984427209204241, "全帯域_40ms_modulation_mae_db": 0.343559339088179, "300–1000 Hz_10ms_log_rms_mae_db": 1.3324107294741514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07665459163822787, "300–1000 Hz_10ms_envelope_correlation": 0.9941640428551326, "300–1000 Hz_10ms_modulation_mae_db": 0.9667756364914215, "300–1000 Hz_40ms_log_rms_mae_db": 0.7478611996088121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03820941764860405, "300–1000 Hz_40ms_envelope_correlation": 0.9984148683255389, "300–1000 Hz_40ms_modulation_mae_db": 0.8875562547869288, "1000–3000 Hz_10ms_log_rms_mae_db": 1.452000142987813, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10191125082302198, "1000–3000 Hz_10ms_envelope_correlation": 0.9931026884134038, "1000–3000 Hz_10ms_modulation_mae_db": 0.8563279775327469, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9401728028659195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.065207713760194, "1000–3000 Hz_40ms_envelope_correlation": 0.9975561821192804, "1000–3000 Hz_40ms_modulation_mae_db": 1.0219676854329214, "3000–6000 Hz_10ms_log_rms_mae_db": 1.40723729851963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15465597886143875, "3000–6000 Hz_10ms_envelope_correlation": 0.9848128223277477, "3000–6000 Hz_10ms_modulation_mae_db": 1.352845668724872, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8290032672545606, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09268904097467442, "3000–6000 Hz_40ms_envelope_correlation": 0.9950332315590401, "3000–6000 Hz_40ms_modulation_mae_db": 1.216095396574384, "mr_stft_db": 5.272131105852206, "band_mae_db": 0.1699945005898397, "rms_gain_error_db": -0.05780568470777127, "output_peak": 0.5601309227855562}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0386067028042483, "全帯域_10ms_rms_envelope_nrmse": 0.06390619025088981, "全帯域_10ms_envelope_correlation": 0.9938962117898544, "全帯域_10ms_modulation_mae_db": 0.2575491273649106, "全帯域_40ms_log_rms_mae_db": 0.49613461726067626, "全帯域_40ms_rms_envelope_nrmse": 0.03216461593631987, "全帯域_40ms_envelope_correlation": 0.9983120185976545, "全帯域_40ms_modulation_mae_db": 0.36184057398484754, "300–1000 Hz_10ms_log_rms_mae_db": 1.3540762020821158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08826255923756726, "300–1000 Hz_10ms_envelope_correlation": 0.9921666280926843, "300–1000 Hz_10ms_modulation_mae_db": 0.3483072225000123, "300–1000 Hz_40ms_log_rms_mae_db": 0.6943084249181318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04841514991592622, "300–1000 Hz_40ms_envelope_correlation": 0.9973755439574659, "300–1000 Hz_40ms_modulation_mae_db": 0.5475491243922036, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4145515763474668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11489782528955891, "1000–3000 Hz_10ms_envelope_correlation": 0.9910098548405062, "1000–3000 Hz_10ms_modulation_mae_db": 0.5508853750592776, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9074677319889549, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06851576986425836, "1000–3000 Hz_40ms_envelope_correlation": 0.9972274127577173, "1000–3000 Hz_40ms_modulation_mae_db": 0.9655136878255218, "3000–6000 Hz_10ms_log_rms_mae_db": 1.409202880743114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14144583161775784, "3000–6000 Hz_10ms_envelope_correlation": 0.987481316424059, "3000–6000 Hz_10ms_modulation_mae_db": 1.301598514397958, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7637802571189578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.082748159035018, "3000–6000 Hz_40ms_envelope_correlation": 0.9958100109812494, "3000–6000 Hz_40ms_modulation_mae_db": 0.9926458855976846, "mr_stft_db": 5.2571278603867535, "band_mae_db": 0.17033631276594455, "rms_gain_error_db": -0.06426338200777712, "output_peak": 0.47200542046291916}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0800279332389313, "全帯域_10ms_rms_envelope_nrmse": 0.06621278102831266, "全帯域_10ms_envelope_correlation": 0.9934132794534333, "全帯域_10ms_modulation_mae_db": 0.2216038643910193, "全帯域_40ms_log_rms_mae_db": 0.559070724219325, "全帯域_40ms_rms_envelope_nrmse": 0.037771781458364906, "全帯域_40ms_envelope_correlation": 0.9975880258507623, "全帯域_40ms_modulation_mae_db": 0.2015445039186664, "300–1000 Hz_10ms_log_rms_mae_db": 1.2984534136553256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08367071816309621, "300–1000 Hz_10ms_envelope_correlation": 0.9930166834240762, "300–1000 Hz_10ms_modulation_mae_db": 0.6736790755870549, "300–1000 Hz_40ms_log_rms_mae_db": 0.7270672087780357, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048004062928286295, "300–1000 Hz_40ms_envelope_correlation": 0.9974288654538737, "300–1000 Hz_40ms_modulation_mae_db": 0.8392349763869552, "1000–3000 Hz_10ms_log_rms_mae_db": 1.46796740370807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11307116284655014, "1000–3000 Hz_10ms_envelope_correlation": 0.9914683082593218, "1000–3000 Hz_10ms_modulation_mae_db": 1.8261332722260712, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9513506878168786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07884380292462494, "1000–3000 Hz_40ms_envelope_correlation": 0.9961825327692134, "1000–3000 Hz_40ms_modulation_mae_db": 1.5869315433965434, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3717507913430504, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15720567036202943, "3000–6000 Hz_10ms_envelope_correlation": 0.985410552846138, "3000–6000 Hz_10ms_modulation_mae_db": 1.2808499907837416, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7326236102656694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08225731170367147, "3000–6000 Hz_40ms_envelope_correlation": 0.9978316705395309, "3000–6000 Hz_40ms_modulation_mae_db": 1.0710254862591697, "mr_stft_db": 5.2748197163474595, "band_mae_db": 0.19957951361294266, "rms_gain_error_db": -0.04567857200004511, "output_peak": 0.5636501182363236}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 51840, "output_samples": 51840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.20534891855816, "spectral_convergence": 0.2644116898806423, "spectral_convergence_status": "ok", "amplitude_floor": 1.3508970735580905e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1217873178678754, "spectral_convergence": 0.22857451339140014, "spectral_convergence_status": "ok", "amplitude_floor": 9.199259876037329e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.465709765008937, "spectral_convergence": 0.2309139334619807, "spectral_convergence_status": "ok", "amplitude_floor": 7.633709769762664e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012491242347822075, "output_power": 0.0012133603107674902, "delta_db": -0.12615849378406757, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0036291285647318994, "output_power": 0.0036052736423326497, "delta_db": -0.028641201902715296, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000626458342936399, "output_power": 0.0006325575569035156, "delta_db": 0.04207851055803821, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008149528003014777, "output_power": 0.0008125693151338955, "delta_db": -0.012720382551958643, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002174945257645064, "output_power": 0.00021141022919358662, "delta_db": -0.12322333395671249, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023982804336185733, "output_power": 0.00023811262657821443, "delta_db": -0.031175385952850864, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.060666218117723845, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.041016218673559514, "pitch": null, "output_rms_dbfs": -21.73065000595433, "output_peak": 0.5314651085865792, "reference_rms_dbfs": -21.689633787280776, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.264282000478324}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 51840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00019883564266349852, "log_rms_mae_db": 1.0867071124683438, "rms_envelope_nrmse": 0.07315190772339793, "envelope_correlation": 0.9919697716857065, "rms_gain_error_db": -0.041029009139277865, "reference_mean_power": 0.006787407573323676, "output_mean_power": 0.0067235869839158005}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.1392663194097437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011382823319318316, "output_power": 0.0011831406829215354, "reference_active": true, "delta_db": 0.16786393146080042}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00037355144462451696, "output_power": 0.000340944265020548, "reference_active": true, "delta_db": -0.396670306519897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.066072767849799e-05, "output_power": 5.393984652372392e-05, "reference_active": true, "delta_db": -0.5099790933751726}], "active_band_count": 3, "active_band_mae_db": 0.35817111045195665}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00017748806718771712, "log_rms_mae_db": 0.6280982674569836, "rms_envelope_nrmse": 0.04116490934705, "envelope_correlation": 0.9971274798078531, "rms_gain_error_db": -0.04105849629733347, "reference_mean_power": 0.006850619249749789, "output_mean_power": 0.00678615821809919}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.7219585966772077e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000998012429960962, "output_power": 0.0010354604186562633, "reference_active": true, "delta_db": 0.15997551940310017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018464455345700424, "output_power": 0.0001721629674074174, "reference_active": true, "delta_db": -0.3039676203039399}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.884647670701786e-06, "output_power": 4.488804669778128e-06, "reference_active": true, "delta_db": -0.36702536054185886}], "active_band_count": 3, "active_band_mae_db": 0.2769895000829663}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015076905043222532, "log_rms_mae_db": 1.3727644898171716, "rms_envelope_nrmse": 0.10891897951620277, "envelope_correlation": 0.988164648148806, "rms_gain_error_db": -0.12228577521811045, "reference_mean_power": 0.0024151558510913072, "output_mean_power": 0.00234809997376852}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.3965349131952106e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007806812222828868, "output_power": 0.0007518263563991908, "reference_active": true, "delta_db": -0.16356186921455063}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024651476965397817, "output_power": 0.0002121391898430812, "reference_active": true, "delta_db": -0.6522203870248734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0716739338834284e-05, "output_power": 1.9397886076121497e-05, "reference_active": true, "delta_db": -1.996207079368559}], "active_band_count": 3, "active_band_mae_db": 0.9373297785359943}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00011516776273837431, "log_rms_mae_db": 0.7919593690511844, "rms_envelope_nrmse": 0.06509135340835842, "envelope_correlation": 0.9954427469092084, "rms_gain_error_db": -0.12228949648541952, "reference_mean_power": 0.0024377564716401134, "output_mean_power": 0.002370071065938214}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.190139920080037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007614290342229629, "output_power": 0.0007163224085447931, "reference_active": true, "delta_db": -0.26520895672465206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011269364256369602, "output_power": 0.00010003653457091213, "reference_active": true, "delta_db": -0.5174077804219624}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5333706288121526e-06, "output_power": 2.2648427514853817e-06, "reference_active": true, "delta_db": -1.931511404027722}], "active_band_count": 3, "active_band_mae_db": 0.9047093803914454}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00014489100768405854, "log_rms_mae_db": 1.4367590898897338, "rms_envelope_nrmse": 0.10382119560261602, "envelope_correlation": 0.992160186030262, "rms_gain_error_db": -0.06027835067122962, "reference_mean_power": 0.0008223932804627593, "output_mean_power": 0.0008110576371741521}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 6.859948035043385e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004581143880628001, "output_power": 0.0004642935163100782, "reference_active": true, "delta_db": 0.05818687006239293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.242911398972428e-05, "output_power": 7.879992241718327e-05, "reference_active": true, "delta_db": -0.6928300007793666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.706958400708153e-06, "output_power": 5.003081654855549e-06, "reference_active": true, "delta_db": -2.4062887850960792}], "active_band_count": 3, "active_band_mae_db": 1.0524352186459462}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00013523948700125142, "log_rms_mae_db": 0.9010475647037118, "rms_envelope_nrmse": 0.05603930547450895, "envelope_correlation": 0.9975948569236315, "rms_gain_error_db": -0.06027922280393557, "reference_mean_power": 0.000830089539077811, "output_mean_power": 0.0008186476482749937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 6.395142215880047e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004454767193500814, "output_power": 0.00044556823909030054, "reference_active": true, "delta_db": 0.0008921327240707013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.066351786105847e-05, "output_power": 3.559158491985621e-05, "reference_active": true, "delta_db": -0.5785761949303818}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.874405735020567e-07, "output_power": 5.112154340767982e-07, "reference_active": true, "delta_db": -2.3953532353178932}], "active_band_count": 3, "active_band_mae_db": 0.9916071876574485}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 8.841298696090765e-05, "log_rms_mae_db": 1.368736509821416, "rms_envelope_nrmse": 0.13792533843564148, "envelope_correlation": 0.9875474879037814, "rms_gain_error_db": -0.12863348801260086, "reference_mean_power": 0.0002494763103331033, "output_mean_power": 0.00024219544082268893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.730062412281665e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013367937231840985, "output_power": 0.0001322095518066348, "reference_active": true, "delta_db": -0.04801564728574599}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8654341368924343e-05, "output_power": 2.2017757653640648e-05, "reference_active": true, "delta_db": -1.1441734328718765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.319126391736523e-06, "output_power": 7.154215731849593e-06, "reference_active": true, "delta_db": -0.6551569059804427}], "active_band_count": 3, "active_band_mae_db": 0.6157819953793551}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 7.608900761326353e-05, "log_rms_mae_db": 0.7843312441249393, "rms_envelope_nrmse": 0.0742993592682421, "envelope_correlation": 0.9963406797415783, "rms_gain_error_db": -0.1286331537344201, "reference_mean_power": 0.00025181099859470626, "output_mean_power": 0.00024446201092779353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 2.5627299576593646e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001307843821160996, "output_power": 0.00012777635206959133, "reference_active": true, "delta_db": -0.10105399770074393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1887209147638657e-05, "output_power": 9.158940049874578e-06, "reference_active": true, "delta_db": -1.1323468755988717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.396397638044017e-07, "output_power": 6.481302336725693e-07, "reference_active": true, "delta_db": 0.057268264530042315}], "active_band_count": 3, "active_band_mae_db": 0.4302230459432193}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.892366032425871, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fbc92f37938a7a38197211ea9794279e68e3602f88c6b859f468463bd84c8c3c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9927861226903817, "standard_deviation": 0.0012040286629733967, "minimum": 0.9910098548405062, "maximum": 0.9943056109047149, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.428360990359526, "standard_deviation": 0.025570379009313554, "minimum": 1.3984841352250288, "maximum": 1.46796740370807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0609579339039363, "standard_deviation": 0.45649663752286096, "minimum": 0.5508853750592776, "maximum": 1.8261332722260712, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1029735095839735, "standard_deviation": 0.007900685059011568, "minimum": 0.09170619971082132, "maximum": 0.11489782528955891, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973264872015286, "standard_deviation": 0.0005685322286657304, "minimum": 0.9961825327692134, "maximum": 0.9980212160849766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9106752309138267, "standard_deviation": 0.024581769948245062, "minimum": 0.8774697133797464, "maximum": 0.9513506878168786, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0809876322196543, "standard_deviation": 0.28390380172933266, "minimum": 0.8491706170776263, "maximum": 1.5869315433965434, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06561933635860909, "standard_deviation": 0.007627911181992876, "minimum": 0.054990578036633105, "maximum": 0.07884380292462494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9858784149646407, "standard_deviation": 0.003382621384291439, "minimum": 0.9793991469426023, "maximum": 0.9904159158468856, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.389460085081692, "standard_deviation": 0.04224902834506517, "minimum": 1.3225217195976244, "maximum": 1.44898206578322, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9134018363887932, "standard_deviation": 0.4242130547193217, "minimum": 0.12946527899120075, "maximum": 1.352845668724872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14818002612836337, "standard_deviation": 0.01790454998300806, "minimum": 0.12151692644068122, "maximum": 0.17931761400396684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.996361851307086, "standard_deviation": 0.001079523496306247, "minimum": 0.9950332315590401, "maximum": 0.9978316705395309, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7877246641315966, "standard_deviation": 0.048538552995861975, "minimum": 0.7326236102656694, "maximum": 0.8706679454619869, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9540868745931104, "standard_deviation": 0.34147782203136884, "minimum": 0.4302230459432193, "maximum": 1.2939178471479555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07803001898874491, "standard_deviation": 0.010290178241538243, "minimum": 0.06289749781010374, "maximum": 0.09268904097467442, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921578171013495, "standard_deviation": 0.0018890960242288976, "minimum": 0.988164648148806, "maximum": 0.9941640428551326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.340832853854904, "standard_deviation": 0.03479356057491086, "minimum": 1.2955891420622092, "maximum": 1.3933661823216419, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7774878820590374, "standard_deviation": 0.22804492855754568, "minimum": 0.3483072225000123, "maximum": 1.0045267650873795, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08834760229939097, "standard_deviation": 0.010243602605887928, "minimum": 0.07665459163822787, "maximum": 0.10891897951620277, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973731226962175, "standard_deviation": 0.000880902181318906, "minimum": 0.9954427469092084, "maximum": 0.9984148683255389, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7215620441692067, "standard_deviation": 0.03460822878973996, "minimum": 0.6895880697713512, "maximum": 0.7919593690511844, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9763008698399087, "standard_deviation": 0.27797755983580846, "minimum": 0.5475491243922036, "maximum": 1.4059686363513473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04898858376983334, "standard_deviation": 0.007872712056279426, "minimum": 0.03820941764860405, "maximum": 0.06509135340835842, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1399090369324827, "standard_deviation": 0.0536810388609582, "minimum": 0.060666218117723845, "maximum": 0.19957951361294266, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.266932669440112, "standard_deviation": 0.00839454872183064, "minimum": 5.2571278603867535, "maximum": 5.281070512546815, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5299479350659829, "standard_deviation": 0.0362548008808162, "minimum": 0.47200542046291916, "maximum": 0.5636501182363236, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06735648160243712, "standard_deviation": 0.03296884278575173, "minimum": -0.1226402438130677, "maximum": -0.022955163171748113, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936453565738528, "standard_deviation": 0.001011325466516341, "minimum": 0.9919697716857065, "maximum": 0.9947743938456233, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0695045925008337, "standard_deviation": 0.035897686488704166, "minimum": 1.0357954219994687, "maximum": 1.1431003214877797, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2731905908523669, "standard_deviation": 0.11291405363659757, "minimum": 0.14967940641054814, "maximum": 0.4585987719227053, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06518498750125579, "standard_deviation": 0.004867685947053527, "minimum": 0.059472007487777125, "maximum": 0.07315190772339793, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997873673284935, "standard_deviation": 0.0005194307848555057, "minimum": 0.9971274798078531, "maximum": 0.9984427209204241, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5604229738286794, "standard_deviation": 0.04938846582042788, "minimum": 0.49613461726067626, "maximum": 0.6280982674569836, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.39757402626164284, "standard_deviation": 0.1266546122837528, "minimum": 0.2015445039186664, "maximum": 0.5531648696961852, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03594968969945989, "standard_deviation": 0.00386187359289738, "minimum": 0.030765479116898097, "maximum": 0.04116490934705, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0032", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.609104871749878, "predicted_mos": 2.609104871749878, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "fefc3545a8e2c6f088452d8b4b6bab99d6a649ab8d09582ed2196e8992f34dc8", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.4674665001221, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9312458972312029, "全帯域_10ms_rms_envelope_nrmse": 0.05909645451115113, "全帯域_10ms_envelope_correlation": 0.9948244340782282, "全帯域_10ms_modulation_mae_db": 0.5228298341808016, "全帯域_40ms_log_rms_mae_db": 0.45335931882325115, "全帯域_40ms_rms_envelope_nrmse": 0.031631975120798864, "全帯域_40ms_envelope_correlation": 0.9984191040683471, "全帯域_40ms_modulation_mae_db": 0.6934283214663064, "300–1000 Hz_10ms_log_rms_mae_db": 1.5775714683845843, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06970020214563773, "300–1000 Hz_10ms_envelope_correlation": 0.9953377106884774, "300–1000 Hz_10ms_modulation_mae_db": 0.6636019452143299, "300–1000 Hz_40ms_log_rms_mae_db": 1.0223142008751864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04054441110566042, "300–1000 Hz_40ms_envelope_correlation": 0.998376731912471, "300–1000 Hz_40ms_modulation_mae_db": 0.7586118013609777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8347453352896803, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10622858797121608, "1000–3000 Hz_10ms_envelope_correlation": 0.9920179365180932, "1000–3000 Hz_10ms_modulation_mae_db": 0.8713461655385503, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3269660701890724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06739566098194028, "1000–3000 Hz_40ms_envelope_correlation": 0.9968008363131414, "1000–3000 Hz_40ms_modulation_mae_db": 0.8564697221530558, "3000–6000 Hz_10ms_log_rms_mae_db": 1.975240311938001, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16113208086369557, "3000–6000 Hz_10ms_envelope_correlation": 0.9829381612342037, "3000–6000 Hz_10ms_modulation_mae_db": 0.6225612905779699, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1561910796197974, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07089611282204192, "3000–6000 Hz_40ms_envelope_correlation": 0.9965933220503648, "3000–6000 Hz_40ms_modulation_mae_db": 0.6735756716284089, "mr_stft_db": 5.561571279643684, "band_mae_db": 0.1517656457808466, "rms_gain_error_db": -0.0778793324591042, "output_peak": 0.47452198755309577}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9437102292931393, "全帯域_10ms_rms_envelope_nrmse": 0.06025133834175344, "全帯域_10ms_envelope_correlation": 0.9945576624357536, "全帯域_10ms_modulation_mae_db": 0.37544672410260466, "全帯域_40ms_log_rms_mae_db": 0.5066585427574724, "全帯域_40ms_rms_envelope_nrmse": 0.03449761727287925, "全帯域_40ms_envelope_correlation": 0.9979995631851758, "全帯域_40ms_modulation_mae_db": 0.4199491247787022, "300–1000 Hz_10ms_log_rms_mae_db": 1.5135770526163628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07770170130205702, "300–1000 Hz_10ms_envelope_correlation": 0.9939911097703517, "300–1000 Hz_10ms_modulation_mae_db": 0.19119114392888314, "300–1000 Hz_40ms_log_rms_mae_db": 0.9979161572264188, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04491115061847061, "300–1000 Hz_40ms_envelope_correlation": 0.9977548445710172, "300–1000 Hz_40ms_modulation_mae_db": 0.21362116701696432, "1000–3000 Hz_10ms_log_rms_mae_db": 1.866934296066973, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10579473758551834, "1000–3000 Hz_10ms_envelope_correlation": 0.9924432307854851, "1000–3000 Hz_10ms_modulation_mae_db": 0.8627959871616095, "1000–3000 Hz_40ms_log_rms_mae_db": 1.289102257767008, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06971733018651065, "1000–3000 Hz_40ms_envelope_correlation": 0.9969758974038213, "1000–3000 Hz_40ms_modulation_mae_db": 1.2390827428610127, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9218009704728498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15258087438924745, "3000–6000 Hz_10ms_envelope_correlation": 0.9858346988802603, "3000–6000 Hz_10ms_modulation_mae_db": 0.4806453988006069, "3000–6000 Hz_40ms_log_rms_mae_db": 1.16619609590137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08540127201410605, "3000–6000 Hz_40ms_envelope_correlation": 0.9957955407178093, "3000–6000 Hz_40ms_modulation_mae_db": 0.468311974863913, "mr_stft_db": 5.5715996326267385, "band_mae_db": 0.1555860017195036, "rms_gain_error_db": -0.04694682125798246, "output_peak": 0.4881099218208887}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9686743156391939, "全帯域_10ms_rms_envelope_nrmse": 0.059359688276531014, "全帯域_10ms_envelope_correlation": 0.9947247040882596, "全帯域_10ms_modulation_mae_db": 0.22465612687805878, "全帯域_40ms_log_rms_mae_db": 0.4629933801873231, "全帯域_40ms_rms_envelope_nrmse": 0.03088805899472307, "全帯域_40ms_envelope_correlation": 0.9984222195611483, "全帯域_40ms_modulation_mae_db": 0.46494863933519653, "300–1000 Hz_10ms_log_rms_mae_db": 1.5585537981275854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07810594782279943, "300–1000 Hz_10ms_envelope_correlation": 0.9941669422577086, "300–1000 Hz_10ms_modulation_mae_db": 0.9305426237512909, "300–1000 Hz_40ms_log_rms_mae_db": 1.0325924011081018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04490314706956333, "300–1000 Hz_40ms_envelope_correlation": 0.9980726845634297, "300–1000 Hz_40ms_modulation_mae_db": 1.141049041206644, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7286088029820297, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09515615425423424, "1000–3000 Hz_10ms_envelope_correlation": 0.9935831861432401, "1000–3000 Hz_10ms_modulation_mae_db": 0.8292604661293662, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2345707643869235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05315500372225243, "1000–3000 Hz_40ms_envelope_correlation": 0.9980403411400471, "1000–3000 Hz_40ms_modulation_mae_db": 0.7709438017024501, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9611470097447183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15524746194651093, "3000–6000 Hz_10ms_envelope_correlation": 0.9842005023873064, "3000–6000 Hz_10ms_modulation_mae_db": 0.8559485520564637, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1448952830364212, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06908196487519867, "3000–6000 Hz_40ms_envelope_correlation": 0.9968760216470367, "3000–6000 Hz_40ms_modulation_mae_db": 1.0900655783740716, "mr_stft_db": 5.553881702873098, "band_mae_db": 0.1369047761159277, "rms_gain_error_db": -0.04968562192508574, "output_peak": 0.5102112722547205}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1316496334530097, "全帯域_10ms_rms_envelope_nrmse": 0.06715480413800999, "全帯域_10ms_envelope_correlation": 0.9932817237544966, "全帯域_10ms_modulation_mae_db": 0.8948002781655298, "全帯域_40ms_log_rms_mae_db": 0.5246680207399202, "全帯域_40ms_rms_envelope_nrmse": 0.03505282677404859, "全帯域_40ms_envelope_correlation": 0.9980239447288408, "全帯域_40ms_modulation_mae_db": 1.0755062623710263, "300–1000 Hz_10ms_log_rms_mae_db": 1.7134938472744041, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0892485437624788, "300–1000 Hz_10ms_envelope_correlation": 0.9922955581535348, "300–1000 Hz_10ms_modulation_mae_db": 1.2142595645900398, "300–1000 Hz_40ms_log_rms_mae_db": 1.1548274521166886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05432730212365221, "300–1000 Hz_40ms_envelope_correlation": 0.9970165705419494, "300–1000 Hz_40ms_modulation_mae_db": 1.408620014816611, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8371002538486032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10260638614352866, "1000–3000 Hz_10ms_envelope_correlation": 0.9927002255540576, "1000–3000 Hz_10ms_modulation_mae_db": 1.7309033470013013, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2421536248975849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05813714003842104, "1000–3000 Hz_40ms_envelope_correlation": 0.9978318672363643, "1000–3000 Hz_40ms_modulation_mae_db": 1.3358066530240393, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8735356388721227, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16467864959961329, "3000–6000 Hz_10ms_envelope_correlation": 0.982610127455241, "3000–6000 Hz_10ms_modulation_mae_db": 0.2501798078331803, "3000–6000 Hz_40ms_log_rms_mae_db": 1.096372105108412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08195265764911441, "3000–6000 Hz_40ms_envelope_correlation": 0.995596931109294, "3000–6000 Hz_40ms_modulation_mae_db": 0.3114626312902974, "mr_stft_db": 5.569882284246366, "band_mae_db": 0.08272164566891356, "rms_gain_error_db": -0.0797234547420892, "output_peak": 0.48557629338932107}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0216729500693125, "全帯域_10ms_rms_envelope_nrmse": 0.06327072645689492, "全帯域_10ms_envelope_correlation": 0.9940300602657809, "全帯域_10ms_modulation_mae_db": 0.42052009945364316, "全帯域_40ms_log_rms_mae_db": 0.527764258144601, "全帯域_40ms_rms_envelope_nrmse": 0.03418224658380084, "全帯域_40ms_envelope_correlation": 0.9980861375397396, "全帯域_40ms_modulation_mae_db": 0.23747957256404661, "300–1000 Hz_10ms_log_rms_mae_db": 1.5517247688971048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07911120319267331, "300–1000 Hz_10ms_envelope_correlation": 0.9937793386957458, "300–1000 Hz_10ms_modulation_mae_db": 0.4384088835789641, "300–1000 Hz_40ms_log_rms_mae_db": 1.037175030219439, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04634371271872485, "300–1000 Hz_40ms_envelope_correlation": 0.997623285861629, "300–1000 Hz_40ms_modulation_mae_db": 0.5297086607276785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7858803669933818, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0922565276458491, "1000–3000 Hz_10ms_envelope_correlation": 0.9941349451612014, "1000–3000 Hz_10ms_modulation_mae_db": 0.8014394405383433, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2462853336768072, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055676461922336676, "1000–3000 Hz_40ms_envelope_correlation": 0.9980042434466004, "1000–3000 Hz_40ms_modulation_mae_db": 0.7713568286209096, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9326125773292102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16749673520324454, "3000–6000 Hz_10ms_envelope_correlation": 0.9818939623918975, "3000–6000 Hz_10ms_modulation_mae_db": 0.7838558574551483, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1494780313702915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09365452530737259, "3000–6000 Hz_40ms_envelope_correlation": 0.9941265193807166, "3000–6000 Hz_40ms_modulation_mae_db": 0.7546107529742777, "mr_stft_db": 5.56468590933342, "band_mae_db": 0.08299267622542078, "rms_gain_error_db": -0.046133572237535565, "output_peak": 0.48890928620909513}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0359850068821252, "全帯域_10ms_rms_envelope_nrmse": 0.06303240378682436, "全帯域_10ms_envelope_correlation": 0.9941242543150132, "全帯域_10ms_modulation_mae_db": 0.6010343513622153, "全帯域_40ms_log_rms_mae_db": 0.5519541074973712, "全帯域_40ms_rms_envelope_nrmse": 0.03355805205472524, "全帯域_40ms_envelope_correlation": 0.9982556129923513, "全帯域_40ms_modulation_mae_db": 0.727848089914842, "300–1000 Hz_10ms_log_rms_mae_db": 1.5710979683214077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08250655484590091, "300–1000 Hz_10ms_envelope_correlation": 0.9933945875066158, "300–1000 Hz_10ms_modulation_mae_db": 1.1442548442023646, "300–1000 Hz_40ms_log_rms_mae_db": 1.004208853564739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047658657748829304, "300–1000 Hz_40ms_envelope_correlation": 0.9976526176662409, "300–1000 Hz_40ms_modulation_mae_db": 1.2979147342036732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.816122931448946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10433268173885119, "1000–3000 Hz_10ms_envelope_correlation": 0.9924574315923071, "1000–3000 Hz_10ms_modulation_mae_db": 1.4620074215626297, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3098692935482865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06435153746398335, "1000–3000 Hz_40ms_envelope_correlation": 0.9972793136512699, "1000–3000 Hz_40ms_modulation_mae_db": 1.2016714316278168, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9860090008643094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17591825827196964, "3000–6000 Hz_10ms_envelope_correlation": 0.9796557969733418, "3000–6000 Hz_10ms_modulation_mae_db": 0.09468138793964453, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2444006209547667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09482320020659991, "3000–6000 Hz_40ms_envelope_correlation": 0.9938384172144952, "3000–6000 Hz_40ms_modulation_mae_db": 0.5681832238779645, "mr_stft_db": 5.572530166338594, "band_mae_db": 0.09057788535737778, "rms_gain_error_db": -0.0728032779206602, "output_peak": 0.4989442137054213}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0481654667218525, "全帯域_10ms_rms_envelope_nrmse": 0.060765004243023564, "全帯域_10ms_envelope_correlation": 0.9944440249581256, "全帯域_10ms_modulation_mae_db": 0.4513769705911967, "全帯域_40ms_log_rms_mae_db": 0.5963735596584432, "全帯域_40ms_rms_envelope_nrmse": 0.032373015104375506, "全帯域_40ms_envelope_correlation": 0.9982093527880661, "全帯域_40ms_modulation_mae_db": 0.5888267600172034, "300–1000 Hz_10ms_log_rms_mae_db": 1.6921660153998526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07971708452839733, "300–1000 Hz_10ms_envelope_correlation": 0.9938677315860693, "300–1000 Hz_10ms_modulation_mae_db": 0.8244109858463947, "300–1000 Hz_40ms_log_rms_mae_db": 1.1310680209856492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04532152891810046, "300–1000 Hz_40ms_envelope_correlation": 0.9978672981414114, "300–1000 Hz_40ms_modulation_mae_db": 0.9784821436269189, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8069829872486851, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1025757931331744, "1000–3000 Hz_10ms_envelope_correlation": 0.9923801004738748, "1000–3000 Hz_10ms_modulation_mae_db": 1.2780516708064207, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2814101008503729, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05821958637120441, "1000–3000 Hz_40ms_envelope_correlation": 0.9974451509134444, "1000–3000 Hz_40ms_modulation_mae_db": 1.0442995401352777, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9369454455967179, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14868118060279154, "3000–6000 Hz_10ms_envelope_correlation": 0.9855635214928229, "3000–6000 Hz_10ms_modulation_mae_db": 2.0752184347252123, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1321867451578187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0722321906062677, "3000–6000 Hz_40ms_envelope_correlation": 0.9964310334291927, "3000–6000 Hz_40ms_modulation_mae_db": 2.1964730465313256, "mr_stft_db": 5.565038440911569, "band_mae_db": 0.05676149458232515, "rms_gain_error_db": -0.012513200824109125, "output_peak": 0.5746235264751112}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0259572811427908, "全帯域_10ms_rms_envelope_nrmse": 0.060559871936601305, "全帯域_10ms_envelope_correlation": 0.9945681093870502, "全帯域_10ms_modulation_mae_db": 0.2505164676158938, "全帯域_40ms_log_rms_mae_db": 0.4714832985368487, "全帯域_40ms_rms_envelope_nrmse": 0.029738935040088278, "全帯域_40ms_envelope_correlation": 0.9986211508869033, "全帯域_40ms_modulation_mae_db": 0.20884654785166878, "300–1000 Hz_10ms_log_rms_mae_db": 1.6030684007425247, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07538844366006485, "300–1000 Hz_10ms_envelope_correlation": 0.9944318494881419, "300–1000 Hz_10ms_modulation_mae_db": 0.4150889342988815, "300–1000 Hz_40ms_log_rms_mae_db": 1.0489257197802415, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04239893562524154, "300–1000 Hz_40ms_envelope_correlation": 0.9981480522893266, "300–1000 Hz_40ms_modulation_mae_db": 0.5913040694836944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.802530721649237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11076160153780798, "1000–3000 Hz_10ms_envelope_correlation": 0.991537276972619, "1000–3000 Hz_10ms_modulation_mae_db": 1.0395851101068043, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2389593037875812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059441173121994394, "1000–3000 Hz_40ms_envelope_correlation": 0.9979057709925467, "1000–3000 Hz_40ms_modulation_mae_db": 0.6706402776348391, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9665436106563774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1617216802340315, "3000–6000 Hz_10ms_envelope_correlation": 0.9829873580902865, "3000–6000 Hz_10ms_modulation_mae_db": 0.5304925898581209, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1681280617595327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09020742493095156, "3000–6000 Hz_40ms_envelope_correlation": 0.9944707792062785, "3000–6000 Hz_40ms_modulation_mae_db": 0.5444269482900916, "mr_stft_db": 5.566355525926003, "band_mae_db": 0.15618475006806595, "rms_gain_error_db": -0.07958526835166556, "output_peak": 0.4830673977436212}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 51840, "output_samples": 51840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.569155441220256, "spectral_convergence": 0.25826892142012636, "spectral_convergence_status": "ok", "amplitude_floor": 1.3508970735580905e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.472628914342773, "spectral_convergence": 0.22887467336370157, "spectral_convergence_status": "ok", "amplitude_floor": 9.199259876037329e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6429294833680235, "spectral_convergence": 0.2240397488859873, "spectral_convergence_status": "ok", "amplitude_floor": 7.633709769762664e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012491242347822075, "output_power": 0.0012243516091092256, "delta_db": -0.08699478023711335, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0036291285647318994, "output_power": 0.003580519314163303, "delta_db": -0.05856332941122647, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000626458342936399, "output_power": 0.0006235425474227542, "delta_db": -0.020261044747371112, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008149528003014777, "output_power": 0.0007875130665473146, "delta_db": -0.1487468798626423, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002174945257645064, "output_power": 0.0001945413236519969, "delta_db": -0.484364639660571, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023982804336185733, "output_power": 0.0002460743147282705, "delta_db": 0.11166320076615534, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1517656457808466, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0778793324591042, "pitch": null, "output_rms_dbfs": -21.767513119739878, "output_peak": 0.47452198755309577, "reference_rms_dbfs": -21.689633787280776, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.561571279643684}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 51840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00019883564266349852, "log_rms_mae_db": 0.9312458972312029, "rms_envelope_nrmse": 0.05909645451115113, "envelope_correlation": 0.9948244340782282, "rms_gain_error_db": -0.07788619999591931, "reference_mean_power": 0.006787407573323676, "output_mean_power": 0.006666767485999134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.1392663194097437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011382823319318316, "output_power": 0.0011316760395717663, "reference_active": true, "delta_db": -0.02527874038719059}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00037355144462451696, "output_power": 0.0003244256416402399, "reference_active": true, "delta_db": -0.6123524803958379}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.066072767849799e-05, "output_power": 4.8957788021318716e-05, "reference_active": true, "delta_db": -0.9308582817593765}], "active_band_count": 3, "active_band_mae_db": 0.5228298341808016}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00017748806718771712, "log_rms_mae_db": 0.45335931882325115, "rms_envelope_nrmse": 0.031631975120798864, "envelope_correlation": 0.9984191040683471, "rms_gain_error_db": -0.07784442150828114, "reference_mean_power": 0.006850619249749789, "output_mean_power": 0.006728920362430212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.7219585966772077e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000998012429960962, "output_power": 0.0009973714958846846, "reference_active": true, "delta_db": -0.002789980800773555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018464455345700424, "output_power": 0.00015796197982865105, "reference_active": true, "delta_db": -0.67783933235329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.884647670701786e-06, "output_power": 3.538895010093152e-06, "reference_active": true, "delta_db": -1.3996556512448557}], "active_band_count": 3, "active_band_mae_db": 0.6934283214663064}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015076905043222532, "log_rms_mae_db": 1.5775714683845843, "rms_envelope_nrmse": 0.06970020214563773, "envelope_correlation": 0.9953377106884774, "rms_gain_error_db": -0.10673643597960712, "reference_mean_power": 0.0024151558510913072, "output_mean_power": 0.002356522103172065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.3965349131952106e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007806812222828868, "output_power": 0.0007370461550191007, "reference_active": true, "delta_db": -0.2497904847816712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024651476965397817, "output_power": 0.00022003068967360747, "reference_active": true, "delta_db": -0.49359684623247574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0716739338834284e-05, "output_power": 2.304800088046907e-05, "reference_active": true, "delta_db": -1.2474185046288429}], "active_band_count": 3, "active_band_mae_db": 0.6636019452143299}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00011516776273837431, "log_rms_mae_db": 1.0223142008751864, "rms_envelope_nrmse": 0.04054441110566042, "envelope_correlation": 0.998376731912471, "rms_gain_error_db": -0.10674387556665343, "reference_mean_power": 0.0024377564716401134, "output_mean_power": 0.0023785699644497706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.190139920080037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007614290342229629, "output_power": 0.0007093729281845374, "reference_active": true, "delta_db": -0.30754822513524205}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011269364256369602, "output_power": 0.00010536984456583425, "reference_active": true, "delta_db": -0.2918307727954199}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5333706288121526e-06, "output_power": 2.401837128483198e-06, "reference_active": true, "delta_db": -1.676456406152271}], "active_band_count": 3, "active_band_mae_db": 0.7586118013609777}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00014489100768405854, "log_rms_mae_db": 1.8347453352896803, "rms_envelope_nrmse": 0.10622858797121608, "envelope_correlation": 0.9920179365180932, "rms_gain_error_db": -0.16903081188098248, "reference_mean_power": 0.0008223932804627593, "output_mean_power": 0.0007909999790281792}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 6.859948035043385e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004581143880628001, "output_power": 0.0004288752809922953, "reference_active": true, "delta_db": -0.28642916330759566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.242911398972428e-05, "output_power": 7.446448581105558e-05, "reference_active": true, "delta_db": -0.9385959500034181}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.706958400708153e-06, "output_power": 6.323610780955706e-06, "reference_active": true, "delta_db": -1.3890133833046368}], "active_band_count": 3, "active_band_mae_db": 0.8713461655385503}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00013523948700125142, "log_rms_mae_db": 1.3269660701890724, "rms_envelope_nrmse": 0.06739566098194028, "envelope_correlation": 0.9968008363131414, "rms_gain_error_db": -0.16903549709398533, "reference_mean_power": 0.000830089539077811, "output_mean_power": 0.0007984015862646031}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 6.395142215880047e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004454767193500814, "output_power": 0.0004077899416260158, "reference_active": true, "delta_db": -0.3838850325017504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.066351786105847e-05, "output_power": 3.38562750164878e-05, "reference_active": true, "delta_db": -0.7956577355241172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.874405735020567e-07, "output_power": 6.443957104638504e-07, "reference_active": true, "delta_db": -1.3898663984332997}], "active_band_count": 3, "active_band_mae_db": 0.8564697221530558}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 8.841298696090765e-05, "log_rms_mae_db": 1.975240311938001, "rms_envelope_nrmse": 0.16113208086369557, "envelope_correlation": 0.9829381612342037, "rms_gain_error_db": -0.07612414103668717, "reference_mean_power": 0.0002494763103331033, "output_mean_power": 0.00024514153334792864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.730062412281665e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013367937231840985, "output_power": 0.00013403564046029305, "reference_active": true, "delta_db": 0.011558961735416873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8654341368924343e-05, "output_power": 2.3305843526029372e-05, "reference_active": true, "delta_db": -0.8972560418064088}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.319126391736523e-06, "output_power": 6.670998388260083e-06, "reference_active": true, "delta_db": -0.9588688681920841}], "active_band_count": 3, "active_band_mae_db": 0.6225612905779699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 7.608900761326353e-05, "log_rms_mae_db": 1.1561910796197974, "rms_envelope_nrmse": 0.07089611282204192, "envelope_correlation": 0.9965933220503648, "rms_gain_error_db": -0.07612777620526127, "reference_mean_power": 0.00025181099859470626, "output_mean_power": 0.00024743544811035384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 2.5627299576593646e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001307843821160996, "output_power": 0.00013084593329798974, "reference_active": true, "delta_db": 0.0020434435589292733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1887209147638657e-05, "output_power": 1.0546103031872636e-05, "reference_active": true, "delta_db": -0.5198789397264448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.396397638044017e-07, "output_power": 4.529549301653268e-07, "reference_active": true, "delta_db": -1.4988046315998527}], "active_band_count": 3, "active_band_mae_db": 0.6735756716284089}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8948278032552711, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c3785476dc9e43bd79c3af8352baa05989af0de0cedc3a06c96305ce279184ae", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9926567916501099, "standard_deviation": 0.0008339570679803525, "minimum": 0.991537276972619, "maximum": 0.9941349451612014, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8098632119409421, "standard_deviation": 0.04122548826888814, "minimum": 1.7286088029820297, "maximum": 1.866934296066973, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1094237011056283, "standard_deviation": 0.3453190939176013, "minimum": 0.8014394405383433, "maximum": 1.7309033470013013, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1024640587512725, "standard_deviation": 0.006038424238319848, "minimum": 0.0922565276458491, "maximum": 0.11076160153780798, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975354276371544, "standard_deviation": 0.0004819898193043563, "minimum": 0.9968008363131414, "maximum": 0.9980403411400471, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2711645936379545, "standard_deviation": 0.035607968028738196, "minimum": 1.2345707643869235, "maximum": 1.3269660701890724, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9862838747199252, "standard_deviation": 0.2520993665121141, "minimum": 0.6706402776348391, "maximum": 1.3358066530240393, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06076173672608041, "standard_deviation": 0.005805362125435136, "minimum": 0.05315500372225243, "maximum": 0.06971733018651065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.98321051611317, "standard_deviation": 0.002009151470598211, "minimum": 0.9796557969733418, "maximum": 0.9858346988802603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9442293206842884, "standard_deviation": 0.03621363219724681, "minimum": 1.8735356388721227, "maximum": 1.9860090008643094, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7116979149057934, "standard_deviation": 0.6062858299852041, "minimum": 0.09468138793964453, "maximum": 2.0752184347252123, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16093211513888805, "standard_deviation": 0.008738441699707531, "minimum": 0.14868118060279154, "maximum": 0.17591825827196964, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954660705943985, "standard_deviation": 0.0011801781978873248, "minimum": 0.9938384172144952, "maximum": 0.9968760216470367, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1572310028635513, "standard_deviation": 0.04198986506104072, "minimum": 1.096372105108412, "maximum": 1.2444006209547667, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8258887284787938, "standard_deviation": 0.599368572868166, "minimum": 0.3114626312902974, "maximum": 2.1964730465313256, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0822811685514566, "standard_deviation": 0.010448135421842692, "minimum": 0.06908196487519867, "maximum": 0.09482320020659991, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9939081035183307, "standard_deviation": 0.000868017902462519, "minimum": 0.9922955581535348, "maximum": 0.9953377106884774, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5976566649704782, "standard_deviation": 0.06989429874787413, "minimum": 1.5135770526163628, "maximum": 1.7134938472744041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7277198656763937, "standard_deviation": 0.36530254222024117, "minimum": 0.19119114392888314, "maximum": 1.2142595645900398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07893496015750118, "standard_deviation": 0.0056040266033062355, "minimum": 0.06970020214563773, "maximum": 0.0892485437624788, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978140106934343, "standard_deviation": 0.00041400927822707496, "minimum": 0.9970165705419494, "maximum": 0.998376731912471, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.053628479484558, "standard_deviation": 0.05794111482110791, "minimum": 0.9979161572264188, "maximum": 1.1548274521166886, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8649139540553953, "standard_deviation": 0.41316922784774135, "minimum": 0.21362116701696432, "maximum": 1.408620014816611, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04580110574103034, "standard_deviation": 0.004096634339136961, "minimum": 0.04054441110566042, "maximum": 0.05432730212365221, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11418685943979763, "standard_deviation": 0.040043143964598404, "minimum": 0.05676149458232515, "maximum": 0.15618475006806595, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.565693117737434, "standard_deviation": 0.006065855177537685, "minimum": 5.553881702873098, "maximum": 5.572530166338594, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5004954873939094, "standard_deviation": 0.031803761566803694, "minimum": 0.47452198755309577, "maximum": 0.5746235264751112, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05815881871477901, "standard_deviation": 0.023764072688087116, "minimum": -0.0797234547420892, "maximum": -0.012513200824109125, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943193716603385, "standard_deviation": 0.000500294035834574, "minimum": 0.9932817237544966, "maximum": 0.9948244340782282, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0133825975540782, "standard_deviation": 0.0649551143903848, "minimum": 0.9312458972312029, "maximum": 1.1316496334530097, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.46764760654374293, "standard_deviation": 0.21396729914968057, "minimum": 0.22465612687805878, "maximum": 0.8948002781655298, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.061686286461348716, "standard_deviation": 0.0026865831430207916, "minimum": 0.05909645451115113, "maximum": 0.06715480413800999, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982546357188216, "standard_deviation": 0.00021980218904170293, "minimum": 0.9979995631851758, "maximum": 0.9986211508869033, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5119068107931539, "standard_deviation": 0.04874813098703534, "minimum": 0.45335931882325115, "maximum": 0.5963735596584432, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.552104164787374, "standard_deviation": 0.28486393119427955, "minimum": 0.20884654785166878, "maximum": 1.0755062623710263, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03274034086817995, "standard_deviation": 0.0018892985484057774, "minimum": 0.029738935040088278, "maximum": 0.03505282677404859, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0032", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8793082237243652, "predicted_mos": 2.8793082237243652, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "435cab036a551dd286ba11f9f5a01799211a760451fa35e958d31b21c382ccce", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.022165299858898, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0615365379165203, "全帯域_10ms_rms_envelope_nrmse": 0.06432664840761858, "全帯域_10ms_envelope_correlation": 0.9938205555150348, "全帯域_10ms_modulation_mae_db": 0.5110348057063906, "全帯域_40ms_log_rms_mae_db": 0.5031301562110496, "全帯域_40ms_rms_envelope_nrmse": 0.03533671228825254, "全帯域_40ms_envelope_correlation": 0.997951851965152, "全帯域_40ms_modulation_mae_db": 0.6501443461979418, "300–1000 Hz_10ms_log_rms_mae_db": 1.2565301762115286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07803830222460614, "300–1000 Hz_10ms_envelope_correlation": 0.994149025500482, "300–1000 Hz_10ms_modulation_mae_db": 1.2291823785637221, "300–1000 Hz_40ms_log_rms_mae_db": 0.6800955165835164, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04512935590817461, "300–1000 Hz_40ms_envelope_correlation": 0.9979886749539522, "300–1000 Hz_40ms_modulation_mae_db": 1.3217695924022863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3108325846462123, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0989760352509897, "1000–3000 Hz_10ms_envelope_correlation": 0.9934411945264517, "1000–3000 Hz_10ms_modulation_mae_db": 1.119360451440646, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9138497474730677, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06325457065245516, "1000–3000 Hz_40ms_envelope_correlation": 0.9976422822036738, "1000–3000 Hz_40ms_modulation_mae_db": 0.8626761204425106, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2654780676606812, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1610908754597094, "3000–6000 Hz_10ms_envelope_correlation": 0.9859872138974508, "3000–6000 Hz_10ms_modulation_mae_db": 1.4479308652783327, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7562700810934059, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10676654332708435, "3000–6000 Hz_40ms_envelope_correlation": 0.9961505231227802, "3000–6000 Hz_40ms_modulation_mae_db": 1.4386107295143118, "mr_stft_db": 5.190581080423103, "band_mae_db": 0.24011489980128484, "rms_gain_error_db": -0.06960593045006305, "output_peak": 0.47784353095667187}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1179822032943, "全帯域_10ms_rms_envelope_nrmse": 0.06552764881151682, "全帯域_10ms_envelope_correlation": 0.9935584148850153, "全帯域_10ms_modulation_mae_db": 0.18356112721732623, "全帯域_40ms_log_rms_mae_db": 0.5870184085634181, "全帯域_40ms_rms_envelope_nrmse": 0.03619618769407725, "全帯域_40ms_envelope_correlation": 0.997803807415603, "全帯域_40ms_modulation_mae_db": 0.4871251432892633, "300–1000 Hz_10ms_log_rms_mae_db": 1.2117840735687089, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07275461245068954, "300–1000 Hz_10ms_envelope_correlation": 0.9947649469004024, "300–1000 Hz_10ms_modulation_mae_db": 0.4740960094163524, "300–1000 Hz_40ms_log_rms_mae_db": 0.6670738990456684, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04229001768857456, "300–1000 Hz_40ms_envelope_correlation": 0.9981025920368012, "300–1000 Hz_40ms_modulation_mae_db": 0.525912353131962, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2379327758619043, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08676579129325289, "1000–3000 Hz_10ms_envelope_correlation": 0.9947963107964702, "1000–3000 Hz_10ms_modulation_mae_db": 0.7177254472634026, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8549513260679106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04980197232988166, "1000–3000 Hz_40ms_envelope_correlation": 0.9984492561742098, "1000–3000 Hz_40ms_modulation_mae_db": 0.797362455980327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2564587458385879, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14485714221246793, "3000–6000 Hz_10ms_envelope_correlation": 0.9864889789681295, "3000–6000 Hz_10ms_modulation_mae_db": 0.28782248786943615, "3000–6000 Hz_40ms_log_rms_mae_db": 0.709515591419684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08109828118897554, "3000–6000 Hz_40ms_envelope_correlation": 0.995631004718083, "3000–6000 Hz_40ms_modulation_mae_db": 0.3971875155417348, "mr_stft_db": 5.199792851202289, "band_mae_db": 0.080508342356566, "rms_gain_error_db": 0.00450052151660171, "output_peak": 0.49975477701871335}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.047589456677692, "全帯域_10ms_rms_envelope_nrmse": 0.06242092402069225, "全帯域_10ms_envelope_correlation": 0.9942695448856101, "全帯域_10ms_modulation_mae_db": 0.21052659273314067, "全帯域_40ms_log_rms_mae_db": 0.5873607420485502, "全帯域_40ms_rms_envelope_nrmse": 0.03908373330094259, "全帯域_40ms_envelope_correlation": 0.9975036278538323, "全帯域_40ms_modulation_mae_db": 0.26814264139990324, "300–1000 Hz_10ms_log_rms_mae_db": 1.170646729289101, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07387247545805158, "300–1000 Hz_10ms_envelope_correlation": 0.9945392717859217, "300–1000 Hz_10ms_modulation_mae_db": 0.4391400515933486, "300–1000 Hz_40ms_log_rms_mae_db": 0.6870528055159743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042141403092653434, "300–1000 Hz_40ms_envelope_correlation": 0.9980120124907361, "300–1000 Hz_40ms_modulation_mae_db": 0.47000942893075504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2474072046620588, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08059970689120986, "1000–3000 Hz_10ms_envelope_correlation": 0.995442103550529, "1000–3000 Hz_10ms_modulation_mae_db": 0.5453005462158694, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8447018002107991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04581655503838781, "1000–3000 Hz_40ms_envelope_correlation": 0.998603716164683, "1000–3000 Hz_40ms_modulation_mae_db": 0.48019499058883747, "3000–6000 Hz_10ms_log_rms_mae_db": 1.210801600155907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13315034113959429, "3000–6000 Hz_10ms_envelope_correlation": 0.988408454130745, "3000–6000 Hz_10ms_modulation_mae_db": 1.044971870179734, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7407277091527141, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06848327272023244, "3000–6000 Hz_40ms_envelope_correlation": 0.996933447186946, "3000–6000 Hz_40ms_modulation_mae_db": 1.127985192567765, "mr_stft_db": 5.173897569512279, "band_mae_db": 0.12768805188820886, "rms_gain_error_db": -0.03180184485778277, "output_peak": 0.6119160306255769}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1378458241596, "全帯域_10ms_rms_envelope_nrmse": 0.06445455820428182, "全帯域_10ms_envelope_correlation": 0.9939195760527626, "全帯域_10ms_modulation_mae_db": 0.4632610587559915, "全帯域_40ms_log_rms_mae_db": 0.5569894018485215, "全帯域_40ms_rms_envelope_nrmse": 0.03669842285439127, "全帯域_40ms_envelope_correlation": 0.9979120053785917, "全帯域_40ms_modulation_mae_db": 0.4325217980756766, "300–1000 Hz_10ms_log_rms_mae_db": 1.2212679259520727, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07389180711543795, "300–1000 Hz_10ms_envelope_correlation": 0.9945171658748173, "300–1000 Hz_10ms_modulation_mae_db": 0.16467854036847096, "300–1000 Hz_40ms_log_rms_mae_db": 0.7032522775201155, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042816335031364915, "300–1000 Hz_40ms_envelope_correlation": 0.9979447514736609, "300–1000 Hz_40ms_modulation_mae_db": 0.05759649689477588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2803710760475038, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09112918506604212, "1000–3000 Hz_10ms_envelope_correlation": 0.9939924119032844, "1000–3000 Hz_10ms_modulation_mae_db": 0.7450791118623975, "1000–3000 Hz_40ms_log_rms_mae_db": 0.864138816702613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05361259176441498, "1000–3000 Hz_40ms_envelope_correlation": 0.9978417793458131, "1000–3000 Hz_40ms_modulation_mae_db": 0.7878528050599143, "3000–6000 Hz_10ms_log_rms_mae_db": 1.292902892065836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12095108802047863, "3000–6000 Hz_10ms_envelope_correlation": 0.9907701716035594, "3000–6000 Hz_10ms_modulation_mae_db": 0.47088861695977435, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7484799468935415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06734053331589798, "3000–6000 Hz_40ms_envelope_correlation": 0.9974559490417174, "3000–6000 Hz_40ms_modulation_mae_db": 0.5515536836428608, "mr_stft_db": 5.2136502578412225, "band_mae_db": 0.1487870700694501, "rms_gain_error_db": -0.0781480236810239, "output_peak": 0.5160973008945582}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0608195552273127, "全帯域_10ms_rms_envelope_nrmse": 0.06377684818492872, "全帯域_10ms_envelope_correlation": 0.9939109001498825, "全帯域_10ms_modulation_mae_db": 0.2733699617312603, "全帯域_40ms_log_rms_mae_db": 0.5464683110620323, "全帯域_40ms_rms_envelope_nrmse": 0.0315646765601302, "全帯域_40ms_envelope_correlation": 0.9983630402108837, "全帯域_40ms_modulation_mae_db": 0.42951464468526696, "300–1000 Hz_10ms_log_rms_mae_db": 1.174549888798107, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0641806710299817, "300–1000 Hz_10ms_envelope_correlation": 0.9959273990487674, "300–1000 Hz_10ms_modulation_mae_db": 0.27325777742556284, "300–1000 Hz_40ms_log_rms_mae_db": 0.6488204515158061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031140191537215985, "300–1000 Hz_40ms_envelope_correlation": 0.9989749905198704, "300–1000 Hz_40ms_modulation_mae_db": 0.4227665890067991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3003767960615267, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10152586549562254, "1000–3000 Hz_10ms_envelope_correlation": 0.993093425960027, "1000–3000 Hz_10ms_modulation_mae_db": 0.9103815253535649, "1000–3000 Hz_40ms_log_rms_mae_db": 0.917922668054326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06068147558371812, "1000–3000 Hz_40ms_envelope_correlation": 0.9979395221498051, "1000–3000 Hz_40ms_modulation_mae_db": 0.8880238732067972, "3000–6000 Hz_10ms_log_rms_mae_db": 1.21734619903651, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14117267399761885, "3000–6000 Hz_10ms_envelope_correlation": 0.987044719034669, "3000–6000 Hz_10ms_modulation_mae_db": 0.45773779203575865, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7279715098039706, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09171417851330728, "3000–6000 Hz_40ms_envelope_correlation": 0.9943001222602695, "3000–6000 Hz_40ms_modulation_mae_db": 0.6055273156540499, "mr_stft_db": 5.196913785749025, "band_mae_db": 0.16232727167175395, "rms_gain_error_db": -0.06054841283996004, "output_peak": 0.4995761850333068}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0843426850559845, "全帯域_10ms_rms_envelope_nrmse": 0.07467591181123352, "全帯域_10ms_envelope_correlation": 0.9916134138839434, "全帯域_10ms_modulation_mae_db": 0.13877525330300503, "全帯域_40ms_log_rms_mae_db": 0.6213525932628031, "全帯域_40ms_rms_envelope_nrmse": 0.044263622530984426, "全帯域_40ms_envelope_correlation": 0.9966544742887194, "全帯域_40ms_modulation_mae_db": 0.3409313883907645, "300–1000 Hz_10ms_log_rms_mae_db": 1.2519103052575218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0673098112310778, "300–1000 Hz_10ms_envelope_correlation": 0.9955566576047131, "300–1000 Hz_10ms_modulation_mae_db": 0.8373751967289627, "300–1000 Hz_40ms_log_rms_mae_db": 0.7001003632096883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03736483758542755, "300–1000 Hz_40ms_envelope_correlation": 0.9985618648898534, "300–1000 Hz_40ms_modulation_mae_db": 0.881569563934212, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3246517419828678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09072359174597446, "1000–3000 Hz_10ms_envelope_correlation": 0.9944447422993988, "1000–3000 Hz_10ms_modulation_mae_db": 0.3394219950400373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8988659127978382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05512997232748708, "1000–3000 Hz_40ms_envelope_correlation": 0.9982084480302844, "1000–3000 Hz_40ms_modulation_mae_db": 0.5620875618651117, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2441660188018797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15078307883621536, "3000–6000 Hz_10ms_envelope_correlation": 0.9864668420384933, "3000–6000 Hz_10ms_modulation_mae_db": 0.5968979603879075, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7477105619066017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11399569658767199, "3000–6000 Hz_40ms_envelope_correlation": 0.992321214857783, "3000–6000 Hz_40ms_modulation_mae_db": 0.47074071553070534, "mr_stft_db": 5.2008802207116105, "band_mae_db": 0.10878324062494087, "rms_gain_error_db": -0.022238148035765654, "output_peak": 0.5689073866912377}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0200866754855584, "全帯域_10ms_rms_envelope_nrmse": 0.06184367738604358, "全帯域_10ms_envelope_correlation": 0.9943309438208097, "全帯域_10ms_modulation_mae_db": 0.06717780624291916, "全帯域_40ms_log_rms_mae_db": 0.48763392476540995, "全帯域_40ms_rms_envelope_nrmse": 0.0302440958828737, "全帯域_40ms_envelope_correlation": 0.9985555915205945, "全帯域_40ms_modulation_mae_db": 0.17052975688968797, "300–1000 Hz_10ms_log_rms_mae_db": 1.1708252515260789, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07441872059280356, "300–1000 Hz_10ms_envelope_correlation": 0.9944377543817554, "300–1000 Hz_10ms_modulation_mae_db": 0.1754930008223943, "300–1000 Hz_40ms_log_rms_mae_db": 0.7046239371405811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04567069073499423, "300–1000 Hz_40ms_envelope_correlation": 0.9976475952408209, "300–1000 Hz_40ms_modulation_mae_db": 0.35043120153348967, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3202338861547755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08871291041856984, "1000–3000 Hz_10ms_envelope_correlation": 0.9944079505846035, "1000–3000 Hz_10ms_modulation_mae_db": 0.9600817788940968, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9108308203476103, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053724240534092034, "1000–3000 Hz_40ms_envelope_correlation": 0.9979629639032581, "1000–3000 Hz_40ms_modulation_mae_db": 0.9976834652958756, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2590318334912771, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1534631042038973, "3000–6000 Hz_10ms_envelope_correlation": 0.9848910722968665, "3000–6000 Hz_10ms_modulation_mae_db": 0.7470705830752898, "3000–6000 Hz_40ms_log_rms_mae_db": 0.731389535231675, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08054548178190324, "3000–6000 Hz_40ms_envelope_correlation": 0.9962922053914482, "3000–6000 Hz_40ms_modulation_mae_db": 0.9322526417208579, "mr_stft_db": 5.193749995227132, "band_mae_db": 0.14464802192450443, "rms_gain_error_db": -0.05379891643558416, "output_peak": 0.48063450578032574}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1223707602666517, "全帯域_10ms_rms_envelope_nrmse": 0.07100802044674087, "全帯域_10ms_envelope_correlation": 0.9924795633248503, "全帯域_10ms_modulation_mae_db": 0.6148983832297642, "全帯域_40ms_log_rms_mae_db": 0.6067117194081625, "全帯域_40ms_rms_envelope_nrmse": 0.03990511987774612, "全帯域_40ms_envelope_correlation": 0.9973857502652208, "全帯域_40ms_modulation_mae_db": 0.6133859341759033, "300–1000 Hz_10ms_log_rms_mae_db": 1.2706402430495267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08378233703724765, "300–1000 Hz_10ms_envelope_correlation": 0.9931382298423073, "300–1000 Hz_10ms_modulation_mae_db": 1.0048085504917477, "300–1000 Hz_40ms_log_rms_mae_db": 0.7641098034382018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048616413324531586, "300–1000 Hz_40ms_envelope_correlation": 0.9975733806874771, "300–1000 Hz_40ms_modulation_mae_db": 1.051017671545628, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3227722513870275, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09621105429412839, "1000–3000 Hz_10ms_envelope_correlation": 0.993480257663056, "1000–3000 Hz_10ms_modulation_mae_db": 0.9965719051246893, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8945216676820575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060008727289015074, "1000–3000 Hz_40ms_envelope_correlation": 0.9975213519444378, "1000–3000 Hz_40ms_modulation_mae_db": 1.0012350172597582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2724275163073726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15210995914338546, "3000–6000 Hz_10ms_envelope_correlation": 0.985069721742416, "3000–6000 Hz_10ms_modulation_mae_db": 0.3754918752826621, "3000–6000 Hz_40ms_log_rms_mae_db": 0.79056868693669, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0986442776182656, "3000–6000 Hz_40ms_envelope_correlation": 0.9935103722682044, "3000–6000 Hz_40ms_modulation_mae_db": 0.5534431574671671, "mr_stft_db": 5.189929162542001, "band_mae_db": 0.12423607209456052, "rms_gain_error_db": -0.06894486898183072, "output_peak": 0.5052580187309577}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 51840, "output_samples": 51840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.0918765050559305, "spectral_convergence": 0.26583519798306116, "spectral_convergence_status": "ok", "amplitude_floor": 1.3508970735580905e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.008538822892659, "spectral_convergence": 0.21895010093316647, "spectral_convergence_status": "ok", "amplitude_floor": 9.199259876037329e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.471327913320719, "spectral_convergence": 0.22425802355136523, "spectral_convergence_status": "ok", "amplitude_floor": 7.633709769762664e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012491242347822075, "output_power": 0.001273928565937027, "delta_db": 0.08539441755327247, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0036291285647318994, "output_power": 0.0035855512159975087, "delta_db": -0.05246423353980099, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000626458342936399, "output_power": 0.0006307763706803266, "delta_db": 0.029832186072398632, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008149528003014777, "output_power": 0.0007739973526378905, "delta_db": -0.2239298115877163, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002174945257645064, "output_power": 0.0001965881260392972, "delta_db": -0.43891047621357404, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023982804336185733, "output_power": 0.0002083934847051803, "delta_db": -0.6101582738409467, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24011489980128484, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06960593045006305, "pitch": null, "output_rms_dbfs": -21.75923971773084, "output_peak": 0.47784353095667187, "reference_rms_dbfs": -21.689633787280776, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.190581080423103}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 51840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00019883564266349852, "log_rms_mae_db": 1.0615365379165203, "rms_envelope_nrmse": 0.06432664840761858, "envelope_correlation": 0.9938205555150348, "rms_gain_error_db": -0.06960162806635517, "reference_mean_power": 0.006787407573323676, "output_mean_power": 0.006679497103811888}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.1392663194097437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011382823319318316, "output_power": 0.0011235111831253812, "reference_active": true, "delta_db": -0.0567259529105331}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00037355144462451696, "output_power": 0.00032389175696876716, "reference_active": true, "delta_db": -0.6195052490768185}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.066072767849799e-05, "output_power": 4.979896242789835e-05, "reference_active": true, "delta_db": -0.8568732151318205}], "active_band_count": 3, "active_band_mae_db": 0.5110348057063906}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00017748806718771712, "log_rms_mae_db": 0.5031301562110496, "rms_envelope_nrmse": 0.03533671228825254, "envelope_correlation": 0.997951851965152, "rms_gain_error_db": -0.06960666115502287, "reference_mean_power": 0.006850619249749789, "output_mean_power": 0.006741695988262534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.7219585966772077e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000998012429960962, "output_power": 0.000986444566464961, "reference_active": true, "delta_db": -0.05063265364230546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018464455345700424, "output_power": 0.0001564087332713308, "reference_active": true, "delta_db": -0.720755028373105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.884647670701786e-06, "output_power": 3.7233058088799667e-06, "reference_active": true, "delta_db": -1.1790453565784151}], "active_band_count": 3, "active_band_mae_db": 0.6501443461979418}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015076905043222532, "log_rms_mae_db": 1.2565301762115286, "rms_envelope_nrmse": 0.07803830222460614, "envelope_correlation": 0.994149025500482, "rms_gain_error_db": -0.12525434792897489, "reference_mean_power": 0.0024151558510913072, "output_mean_power": 0.002346495503983397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.3965349131952106e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007806812222828868, "output_power": 0.0007312960169681072, "reference_active": true, "delta_db": -0.28380525380690075}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024651476965397817, "output_power": 0.0002028000414633351, "reference_active": true, "delta_db": -0.847749052014321}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0716739338834284e-05, "output_power": 1.7052019842457293e-05, "reference_active": true, "delta_db": -2.555992829869944}], "active_band_count": 3, "active_band_mae_db": 1.2291823785637221}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00011516776273837431, "log_rms_mae_db": 0.6800955165835164, "rms_envelope_nrmse": 0.04512935590817461, "envelope_correlation": 0.9979886749539522, "rms_gain_error_db": -0.12526265032607833, "reference_mean_power": 0.0024377564716401134, "output_mean_power": 0.0023684490848375856}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.190139920080037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007614290342229629, "output_power": 0.0006990568875396542, "reference_active": true, "delta_db": -0.37116913968888493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011269364256369602, "output_power": 9.15467422858733e-05, "reference_active": true, "delta_db": -0.9025652225744414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5333706288121526e-06, "output_power": 1.9012172810499434e-06, "reference_active": true, "delta_db": -2.691574414943533}], "active_band_count": 3, "active_band_mae_db": 1.3217695924022863}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00014489100768405854, "log_rms_mae_db": 1.3108325846462123, "rms_envelope_nrmse": 0.0989760352509897, "envelope_correlation": 0.9934411945264517, "rms_gain_error_db": -0.280219607954991, "reference_mean_power": 0.0008223932804627593, "output_mean_power": 0.0007710057063898629}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 6.859948035043385e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004581143880628001, "output_power": 0.00040781874458570554, "reference_active": true, "delta_db": -0.5050674854622447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.242911398972428e-05, "output_power": 7.359446099923088e-05, "reference_active": true, "delta_db": -0.9896366107830936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.706958400708153e-06, "output_power": 5.669290924116341e-06, "reference_active": true, "delta_db": -1.8633772580765997}], "active_band_count": 3, "active_band_mae_db": 1.119360451440646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00013523948700125142, "log_rms_mae_db": 0.9138497474730677, "rms_envelope_nrmse": 0.06325457065245516, "envelope_correlation": 0.9976422822036738, "rms_gain_error_db": -0.2802196098522132, "reference_mean_power": 0.000830089539077811, "output_mean_power": 0.0007782210608576564}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 6.395142215880047e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004454767193500814, "output_power": 0.00039066397969262994, "reference_active": true, "delta_db": -0.5702164279117331}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.066351786105847e-05, "output_power": 3.3264933572383304e-05, "reference_active": true, "delta_db": -0.8721828685619288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.874405735020567e-07, "output_power": 6.816734419089026e-07, "reference_active": true, "delta_db": -1.1456290648538698}], "active_band_count": 3, "active_band_mae_db": 0.8626761204425106}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 8.841298696090765e-05, "log_rms_mae_db": 1.2654780676606812, "rms_envelope_nrmse": 0.1610908754597094, "envelope_correlation": 0.9859872138974508, "rms_gain_error_db": -0.706281018906809, "reference_mean_power": 0.0002494763103331033, "output_mean_power": 0.00021203190265222546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.730062412281665e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013367937231840985, "output_power": 0.00011458862853398271, "reference_active": true, "delta_db": -0.6692287613430026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8654341368924343e-05, "output_power": 2.3016863897184184e-05, "reference_active": true, "delta_db": -0.9514428082082732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.319126391736523e-06, "output_power": 4.443913735667236e-06, "reference_active": true, "delta_db": -2.7231210262837227}], "active_band_count": 3, "active_band_mae_db": 1.4479308652783327}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 7.608900761326353e-05, "log_rms_mae_db": 0.7562700810934059, "rms_envelope_nrmse": 0.10676654332708435, "envelope_correlation": 0.9961505231227802, "rms_gain_error_db": -0.7062828809780131, "reference_mean_power": 0.00025181099859470626, "output_mean_power": 0.00021401608103506742}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 2.5627299576593646e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001307843821160996, "output_power": 0.00011127939333720064, "reference_active": true, "delta_db": -0.7014113557914838}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1887209147638657e-05, "output_power": 9.842181675497605e-06, "reference_active": true, "delta_db": -0.819885265445174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.396397638044017e-07, "output_power": 3.3611035882369755e-07, "reference_active": true, "delta_db": -2.7945355673062773}], "active_band_count": 3, "active_band_mae_db": 1.4386107295143118}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8887418909048869, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f54878a85cfc395fe6187a1a85538b4bc7a006a32e82e6073f21872c7cd0b55a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9941372996604776, "standard_deviation": 0.0007866305246776381, "minimum": 0.993093425960027, "maximum": 0.995442103550529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2930722896004845, "standard_deviation": 0.034377884395351874, "minimum": 1.2379327758619043, "maximum": 1.3246517419828678, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.791740345149338, "standard_deviation": 0.2576502086463764, "minimum": 0.3394219950400373, "maximum": 1.119360451440646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09183051755697372, "standard_deviation": 0.006839658340386156, "minimum": 0.08059970689120986, "maximum": 0.10152586549562254, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9980211649895206, "standard_deviation": 0.00037669621064169933, "minimum": 0.9975213519444378, "maximum": 0.998603716164683, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8874728449170278, "standard_deviation": 0.028734011599237035, "minimum": 0.8447018002107991, "maximum": 0.917922668054326, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7971395362123914, "standard_deviation": 0.1890643912130991, "minimum": 0.48019499058883747, "maximum": 1.0012350172597582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.055253763189931485, "standard_deviation": 0.005854348739224072, "minimum": 0.04581655503838781, "maximum": 0.06325457065245516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9868908967140413, "standard_deviation": 0.001921658299486526, "minimum": 0.9848910722968665, "maximum": 0.9907701716035594, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2523266091697565, "standard_deviation": 0.027516299975393222, "minimum": 1.210801600155907, "maximum": 1.292902892065836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6786015063836119, "standard_deviation": 0.39169366754101786, "minimum": 0.28782248786943615, "maximum": 1.4479308652783327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1446972828766709, "standard_deviation": 0.012801987789854363, "minimum": 0.12095108802047863, "maximum": 0.1610908754597094, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9953243548559039, "standard_deviation": 0.0017810362068945058, "minimum": 0.992321214857783, "maximum": 0.9974559490417174, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7440792028047853, "standard_deviation": 0.023811174280649848, "minimum": 0.709515591419684, "maximum": 0.79056868693669, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7596626189549316, "standard_deviation": 0.3685641880271224, "minimum": 0.3971875155417348, "maximum": 1.4386107295143118, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0885735331316673, "standard_deviation": 0.01716867332980469, "minimum": 0.06734053331589798, "maximum": 0.11399569658767199, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9946288063673958, "standard_deviation": 0.0008514503655454216, "minimum": 0.9931382298423073, "maximum": 0.9959273990487674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2160193242065809, "standard_deviation": 0.041021096040514224, "minimum": 1.170646729289101, "maximum": 1.2706402430495267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5747539381763203, "standard_deviation": 0.4016373073154555, "minimum": 0.16467854036847096, "maximum": 1.2291823785637221, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07353109214248699, "standard_deviation": 0.0060139734739070565, "minimum": 0.0641806710299817, "maximum": 0.08378233703724765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981007327866465, "standard_deviation": 0.00046370636463659206, "minimum": 0.9975733806874771, "maximum": 0.9989749905198704, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6943911317461939, "standard_deviation": 0.034137646711267615, "minimum": 0.6488204515158061, "maximum": 0.7641098034382018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6351341121724885, "standard_deviation": 0.41473895619471735, "minimum": 0.05759649689477588, "maximum": 1.3217695924022863, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04189615561286711, "standard_deviation": 0.005439907737330516, "minimum": 0.031140191537215985, "maximum": 0.048616413324531586, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1421366213039087, "standard_deviation": 0.04705259156416726, "minimum": 0.080508342356566, "maximum": 0.24011489980128484, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.194924365401082, "standard_deviation": 0.011345941213460825, "minimum": 5.173897569512279, "maximum": 5.2136502578412225, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5199984669664186, "standard_deviation": 0.046639740107226074, "minimum": 0.47784353095667187, "maximum": 0.6119160306255769, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.047573202970676076, "standard_deviation": 0.02850162390436041, "minimum": -0.0781480236810239, "maximum": 0.00450052151660171, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9934878640647387, "standard_deviation": 0.0009511675556491452, "minimum": 0.9916134138839434, "maximum": 0.9943309438208097, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0815717122604527, "standard_deviation": 0.041286462343746826, "minimum": 1.0200866754855584, "maximum": 1.1378458241596, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3078256236149747, "standard_deviation": 0.19724284694655692, "minimum": 0.06717780624291916, "maximum": 0.6148983832297642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06600427965913203, "standard_deviation": 0.004484742166472501, "minimum": 0.06184367738604358, "maximum": 0.07467591181123352, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977662686123248, "standard_deviation": 0.0005959527234978256, "minimum": 0.9966544742887194, "maximum": 0.9985555915205945, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5620831571462435, "standard_deviation": 0.047888578971696606, "minimum": 0.48763392476540995, "maximum": 0.6213525932628031, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42403695663805097, "standard_deviation": 0.16324179136013134, "minimum": 0.17052975688968797, "maximum": 0.6501443461979418, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03666157137367476, "standard_deviation": 0.004524760090103988, "minimum": 0.0302440958828737, "maximum": 0.044263622530984426, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0032", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.139608144760132, "predicted_mos": 3.139608144760132, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "58793b4ffacbda2615051bb4fd02661efc5d358cbc5d6af5487c0934ad46b076", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.3801039000973105, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.042351709615178, "全帯域_10ms_rms_envelope_nrmse": 0.07043363060295722, "全帯域_10ms_envelope_correlation": 0.9925641671177401, "全帯域_10ms_modulation_mae_db": 0.3217040108109145, "全帯域_40ms_log_rms_mae_db": 0.539712532134478, "全帯域_40ms_rms_envelope_nrmse": 0.03895282373613244, "全帯域_40ms_envelope_correlation": 0.9974435244292735, "全帯域_40ms_modulation_mae_db": 0.18203507236210661, "300–1000 Hz_10ms_log_rms_mae_db": 1.4267069014894356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10489897034688929, "300–1000 Hz_10ms_envelope_correlation": 0.9890705352443623, "300–1000 Hz_10ms_modulation_mae_db": 0.7836002450135027, "300–1000 Hz_40ms_log_rms_mae_db": 0.9752372251627519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0619053041712399, "300–1000 Hz_40ms_envelope_correlation": 0.995943006687498, "300–1000 Hz_40ms_modulation_mae_db": 0.6785482844030447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5129380836269484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10216659287541714, "1000–3000 Hz_10ms_envelope_correlation": 0.9924139112669095, "1000–3000 Hz_10ms_modulation_mae_db": 0.9431020713211025, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0440445037241186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05601154234621324, "1000–3000 Hz_40ms_envelope_correlation": 0.9975824101542171, "1000–3000 Hz_40ms_modulation_mae_db": 0.8444158866547337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4973567300041404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1359266500975392, "3000–6000 Hz_10ms_envelope_correlation": 0.9878865399778947, "3000–6000 Hz_10ms_modulation_mae_db": 0.6200498602912644, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9711582574616042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07323656949480417, "3000–6000 Hz_40ms_envelope_correlation": 0.996417010443358, "3000–6000 Hz_40ms_modulation_mae_db": 0.5365849697058368, "mr_stft_db": 5.335118196392453, "band_mae_db": 0.04673076656796334, "rms_gain_error_db": -0.0430312696150109, "output_peak": 0.5158652929434137}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.983940199584946, "全帯域_10ms_rms_envelope_nrmse": 0.06414023062882936, "全帯域_10ms_envelope_correlation": 0.9939458614147848, "全帯域_10ms_modulation_mae_db": 0.2801208814410706, "全帯域_40ms_log_rms_mae_db": 0.48028999511105136, "全帯域_40ms_rms_envelope_nrmse": 0.03618812932770209, "全帯域_40ms_envelope_correlation": 0.9979451159504268, "全帯域_40ms_modulation_mae_db": 0.5017311522119899, "300–1000 Hz_10ms_log_rms_mae_db": 1.3554867174944392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08615533330053028, "300–1000 Hz_10ms_envelope_correlation": 0.9927830218174688, "300–1000 Hz_10ms_modulation_mae_db": 1.0822449116538504, "300–1000 Hz_40ms_log_rms_mae_db": 0.8794042619766499, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05161031507419096, "300–1000 Hz_40ms_envelope_correlation": 0.9973075285452825, "300–1000 Hz_40ms_modulation_mae_db": 1.384939100235335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4998143958265426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09626404443665985, "1000–3000 Hz_10ms_envelope_correlation": 0.9938825759399355, "1000–3000 Hz_10ms_modulation_mae_db": 0.7237761376607081, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0639415787705169, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06552653905174434, "1000–3000 Hz_40ms_envelope_correlation": 0.9974622285232336, "1000–3000 Hz_40ms_modulation_mae_db": 0.584582956396909, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5452171693390644, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1568770688325193, "3000–6000 Hz_10ms_envelope_correlation": 0.9840572285138021, "3000–6000 Hz_10ms_modulation_mae_db": 0.8000361468411276, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9680740138999737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08062212433099844, "3000–6000 Hz_40ms_envelope_correlation": 0.9956558463188985, "3000–6000 Hz_40ms_modulation_mae_db": 0.9049381190787772, "mr_stft_db": 5.322425947982654, "band_mae_db": 0.1207908000754746, "rms_gain_error_db": -0.09767650173696077, "output_peak": 0.5713460430073578}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0372909571485132, "全帯域_10ms_rms_envelope_nrmse": 0.06026714637769338, "全帯域_10ms_envelope_correlation": 0.9946325538854306, "全帯域_10ms_modulation_mae_db": 0.46876615737275823, "全帯域_40ms_log_rms_mae_db": 0.525660458931644, "全帯域_40ms_rms_envelope_nrmse": 0.033223765072769244, "全帯域_40ms_envelope_correlation": 0.998265643710918, "全帯域_40ms_modulation_mae_db": 0.4920354732887357, "300–1000 Hz_10ms_log_rms_mae_db": 1.4840170550754057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09039692478033816, "300–1000 Hz_10ms_envelope_correlation": 0.9919256487152601, "300–1000 Hz_10ms_modulation_mae_db": 0.6612611063074604, "300–1000 Hz_40ms_log_rms_mae_db": 0.8862338339366784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04790547263780102, "300–1000 Hz_40ms_envelope_correlation": 0.997612340423315, "300–1000 Hz_40ms_modulation_mae_db": 0.9477517015439569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5408680421114647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0918775425565616, "1000–3000 Hz_10ms_envelope_correlation": 0.9947333426913684, "1000–3000 Hz_10ms_modulation_mae_db": 1.4385652918730258, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1111400094472066, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0651108503643791, "1000–3000 Hz_40ms_envelope_correlation": 0.9977410684801353, "1000–3000 Hz_40ms_modulation_mae_db": 1.345278315106147, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4567852323625827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12410119348547288, "3000–6000 Hz_10ms_envelope_correlation": 0.989900412326341, "3000–6000 Hz_10ms_modulation_mae_db": 1.0320605914603422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8896494808747201, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06331400966084938, "3000–6000 Hz_40ms_envelope_correlation": 0.9972912045199634, "3000–6000 Hz_40ms_modulation_mae_db": 1.2861181128093726, "mr_stft_db": 5.339882826818976, "band_mae_db": 0.1606778633549634, "rms_gain_error_db": -0.08701949010444665, "output_peak": 0.48480904540858183}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0183542889597794, "全帯域_10ms_rms_envelope_nrmse": 0.06099407824994006, "全帯域_10ms_envelope_correlation": 0.9946407884018117, "全帯域_10ms_modulation_mae_db": 0.4611552003935153, "全帯域_40ms_log_rms_mae_db": 0.4863557343967248, "全帯域_40ms_rms_envelope_nrmse": 0.034884352952543045, "全帯域_40ms_envelope_correlation": 0.9982473997787641, "全帯域_40ms_modulation_mae_db": 0.6110127121355095, "300–1000 Hz_10ms_log_rms_mae_db": 1.4460332933787856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08829262591062559, "300–1000 Hz_10ms_envelope_correlation": 0.9924656293759013, "300–1000 Hz_10ms_modulation_mae_db": 0.8847107143945454, "300–1000 Hz_40ms_log_rms_mae_db": 0.9039694371739608, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04854782019392118, "300–1000 Hz_40ms_envelope_correlation": 0.9977323316409986, "300–1000 Hz_40ms_modulation_mae_db": 1.230090962355753, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5521009040215643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09325988512070252, "1000–3000 Hz_10ms_envelope_correlation": 0.9940810325298539, "1000–3000 Hz_10ms_modulation_mae_db": 0.8299842244142895, "1000–3000 Hz_40ms_log_rms_mae_db": 1.144784191925591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058632203135111645, "1000–3000 Hz_40ms_envelope_correlation": 0.9978683959506937, "1000–3000 Hz_40ms_modulation_mae_db": 0.7508983293380401, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5108613442085028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17501668835947556, "3000–6000 Hz_10ms_envelope_correlation": 0.9803796179175199, "3000–6000 Hz_10ms_modulation_mae_db": 0.16931101363357762, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8898527209757433, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07247885599298337, "3000–6000 Hz_40ms_envelope_correlation": 0.9966004812969232, "3000–6000 Hz_40ms_modulation_mae_db": 0.28876895908314526, "mr_stft_db": 5.349150501330065, "band_mae_db": 0.13736813015878227, "rms_gain_error_db": -0.1263684120243974, "output_peak": 0.5477713735537502}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1069567712709616, "全帯域_10ms_rms_envelope_nrmse": 0.06524722862423968, "全帯域_10ms_envelope_correlation": 0.9935894730540877, "全帯域_10ms_modulation_mae_db": 0.25794174424244287, "全帯域_40ms_log_rms_mae_db": 0.592842260901427, "全帯域_40ms_rms_envelope_nrmse": 0.038043583080167746, "全帯域_40ms_envelope_correlation": 0.9975286689234695, "全帯域_40ms_modulation_mae_db": 0.5446620254569895, "300–1000 Hz_10ms_log_rms_mae_db": 1.4027608387647312, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07405573412723067, "300–1000 Hz_10ms_envelope_correlation": 0.9946012469767619, "300–1000 Hz_10ms_modulation_mae_db": 0.4729450444368412, "300–1000 Hz_40ms_log_rms_mae_db": 0.923150421112628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042338080073259336, "300–1000 Hz_40ms_envelope_correlation": 0.9981272153074483, "300–1000 Hz_40ms_modulation_mae_db": 0.598612851062308, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5404258467802077, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10625916882592333, "1000–3000 Hz_10ms_envelope_correlation": 0.9920208235838278, "1000–3000 Hz_10ms_modulation_mae_db": 0.4351350589327643, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0819773065617253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0688561046598131, "1000–3000 Hz_40ms_envelope_correlation": 0.996698416982137, "1000–3000 Hz_40ms_modulation_mae_db": 0.5501449495153133, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4596282865710317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12867586852097287, "3000–6000 Hz_10ms_envelope_correlation": 0.9895459845244127, "3000–6000 Hz_10ms_modulation_mae_db": 0.6251343330640641, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9101901593153954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06320848231063586, "3000–6000 Hz_40ms_envelope_correlation": 0.9975188713820732, "3000–6000 Hz_40ms_modulation_mae_db": 1.0747304330394256, "mr_stft_db": 5.326060546690819, "band_mae_db": 0.06067968279302314, "rms_gain_error_db": -0.025389330819290505, "output_peak": 0.5053722938694964}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.998167350987128, "全帯域_10ms_rms_envelope_nrmse": 0.05927359476277326, "全帯域_10ms_envelope_correlation": 0.9947548340991139, "全帯域_10ms_modulation_mae_db": 0.2023580600706513, "全帯域_40ms_log_rms_mae_db": 0.5492432198754507, "全帯域_40ms_rms_envelope_nrmse": 0.02979154963753796, "全帯域_40ms_envelope_correlation": 0.9985482946294234, "全帯域_40ms_modulation_mae_db": 0.40836955178799356, "300–1000 Hz_10ms_log_rms_mae_db": 1.4652538771265973, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07595324766156154, "300–1000 Hz_10ms_envelope_correlation": 0.9943161945394131, "300–1000 Hz_10ms_modulation_mae_db": 0.875923726618462, "300–1000 Hz_40ms_log_rms_mae_db": 0.9976074229880996, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039776304470789586, "300–1000 Hz_40ms_envelope_correlation": 0.9983264287664351, "300–1000 Hz_40ms_modulation_mae_db": 0.7971845667345875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5505298050735603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09783682390971611, "1000–3000 Hz_10ms_envelope_correlation": 0.993765181834271, "1000–3000 Hz_10ms_modulation_mae_db": 0.7374935081300745, "1000–3000 Hz_40ms_log_rms_mae_db": 1.098090639202239, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06293425982842914, "1000–3000 Hz_40ms_envelope_correlation": 0.9978423832019946, "1000–3000 Hz_40ms_modulation_mae_db": 0.7889333617311314, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4843874047564545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1588441902818456, "3000–6000 Hz_10ms_envelope_correlation": 0.9834189490928146, "3000–6000 Hz_10ms_modulation_mae_db": 1.1294013913172434, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9416285913585851, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0917125711765896, "3000–6000 Hz_40ms_envelope_correlation": 0.9943697657863949, "3000–6000 Hz_40ms_modulation_mae_db": 1.1830396147162687, "mr_stft_db": 5.323299389143229, "band_mae_db": 0.13193272147380006, "rms_gain_error_db": -0.05943490663467662, "output_peak": 0.5214960072767505}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9603626868240083, "全帯域_10ms_rms_envelope_nrmse": 0.06391038594515033, "全帯域_10ms_envelope_correlation": 0.9938942538252051, "全帯域_10ms_modulation_mae_db": 0.21155610950270298, "全帯域_40ms_log_rms_mae_db": 0.43980227145743883, "全帯域_40ms_rms_envelope_nrmse": 0.03507637741663858, "全帯域_40ms_envelope_correlation": 0.9979723200091772, "全帯域_40ms_modulation_mae_db": 0.1927930560319054, "300–1000 Hz_10ms_log_rms_mae_db": 1.4105116949791503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08710736574587488, "300–1000 Hz_10ms_envelope_correlation": 0.9923804716158702, "300–1000 Hz_10ms_modulation_mae_db": 0.22232321011818998, "300–1000 Hz_40ms_log_rms_mae_db": 0.886707227196778, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04975182649371101, "300–1000 Hz_40ms_envelope_correlation": 0.9972401183110802, "300–1000 Hz_40ms_modulation_mae_db": 0.361784292684765, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5460543667487536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11941793631263484, "1000–3000 Hz_10ms_envelope_correlation": 0.9901286437531517, "1000–3000 Hz_10ms_modulation_mae_db": 0.8387400200095474, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1211729768450176, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07171266053403973, "1000–3000 Hz_40ms_envelope_correlation": 0.9967684418996794, "1000–3000 Hz_40ms_modulation_mae_db": 0.618150220184175, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4897551448496613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1395962283033978, "3000–6000 Hz_10ms_envelope_correlation": 0.9881753198535831, "3000–6000 Hz_10ms_modulation_mae_db": 1.3506178787403122, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8537421402828069, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07734567124576917, "3000–6000 Hz_40ms_envelope_correlation": 0.9966624220645586, "3000–6000 Hz_40ms_modulation_mae_db": 0.9787030842911633, "mr_stft_db": 5.3265176063044395, "band_mae_db": 0.1789854425531335, "rms_gain_error_db": -0.06409148724639244, "output_peak": 0.4768772076131954}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.040929054567733, "全帯域_10ms_rms_envelope_nrmse": 0.06153520320259538, "全帯域_10ms_envelope_correlation": 0.9943156515184656, "全帯域_10ms_modulation_mae_db": 0.18030286482743496, "全帯域_40ms_log_rms_mae_db": 0.5197781201848262, "全帯域_40ms_rms_envelope_nrmse": 0.03188575183260284, "全帯域_40ms_envelope_correlation": 0.9982895870342982, "全帯域_40ms_modulation_mae_db": 0.20763493705351463, "300–1000 Hz_10ms_log_rms_mae_db": 1.3970865202181826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08135937852195284, "300–1000 Hz_10ms_envelope_correlation": 0.993427458555271, "300–1000 Hz_10ms_modulation_mae_db": 0.6474630968834144, "300–1000 Hz_40ms_log_rms_mae_db": 0.9094493566862977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04788468046691953, "300–1000 Hz_40ms_envelope_correlation": 0.9974706871548685, "300–1000 Hz_40ms_modulation_mae_db": 0.7599834982954176, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5171551486889578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10695320238121694, "1000–3000 Hz_10ms_envelope_correlation": 0.9922503626695777, "1000–3000 Hz_10ms_modulation_mae_db": 1.7062767832841168, "1000–3000 Hz_40ms_log_rms_mae_db": 1.074006533393185, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0736981665441896, "1000–3000 Hz_40ms_envelope_correlation": 0.9965286581785703, "1000–3000 Hz_40ms_modulation_mae_db": 1.4800464765201846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4860901455288877, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1572326039471624, "3000–6000 Hz_10ms_envelope_correlation": 0.9843523433730992, "3000–6000 Hz_10ms_modulation_mae_db": 1.0573013731427545, "3000–6000 Hz_40ms_log_rms_mae_db": 0.908017034332781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07352179371908055, "3000–6000 Hz_40ms_envelope_correlation": 0.9974403343171898, "3000–6000 Hz_40ms_modulation_mae_db": 0.857530622688059, "mr_stft_db": 5.329984969514065, "band_mae_db": 0.1547472567324513, "rms_gain_error_db": -0.042537453706429494, "output_peak": 0.5582839766164207}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 51840, "output_samples": 51840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.270206591150414, "spectral_convergence": 0.258154410090934, "spectral_convergence_status": "ok", "amplitude_floor": 1.3508970735580905e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.195291225430088, "spectral_convergence": 0.23274549390105098, "spectral_convergence_status": "ok", "amplitude_floor": 9.199259876037329e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.539856772596857, "spectral_convergence": 0.23683494929336293, "spectral_convergence_status": "ok", "amplitude_floor": 7.633709769762664e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012491242347822075, "output_power": 0.0012039146083966715, "delta_db": -0.16009950119046598, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0036291285647318994, "output_power": 0.003607383898617449, "delta_db": -0.026099912008284147, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000626458342936399, "output_power": 0.0006283717083490324, "delta_db": 0.013244258594580772, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008149528003014777, "output_power": 0.0008165843585388904, "delta_db": 0.008686004750443609, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002174945257645064, "output_power": 0.0002139912675633843, "delta_db": -0.07052279176082817, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023982804336185733, "output_power": 0.00023992371495866572, "delta_db": 0.0017321311031772992, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.04673076656796334, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0430312696150109, "pitch": null, "output_rms_dbfs": -21.732665056895787, "output_peak": 0.5158652929434137, "reference_rms_dbfs": -21.689633787280776, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.335118196392453}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 51840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00019883564266349852, "log_rms_mae_db": 1.042351709615178, "rms_envelope_nrmse": 0.07043363060295722, "envelope_correlation": 0.9925641671177401, "rms_gain_error_db": -0.04304342610617462, "reference_mean_power": 0.006787407573323676, "output_mean_power": 0.006720469061026634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.1392663194097437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011382823319318316, "output_power": 0.0011828402097821516, "reference_active": true, "delta_db": 0.1667608470910278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00037355144462451696, "output_power": 0.00034463541847153535, "reference_active": true, "delta_db": -0.34990512068223767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.066072767849799e-05, "output_power": 5.4709533204335496e-05, "reference_active": true, "delta_db": -0.448446064659478}], "active_band_count": 3, "active_band_mae_db": 0.3217040108109145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00017748806718771712, "log_rms_mae_db": 0.539712532134478, "rms_envelope_nrmse": 0.03895282373613244, "envelope_correlation": 0.9974435244292735, "rms_gain_error_db": -0.04309654272381651, "reference_mean_power": 0.006850619249749789, "output_mean_power": 0.006782974373630301}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.7219585966772077e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000998012429960962, "output_power": 0.0010382000392058861, "reference_active": true, "delta_db": 0.17145090607928737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018464455345700424, "output_power": 0.00017479346695212025, "reference_active": true, "delta_db": -0.23811305104801084}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.884647670701786e-06, "output_power": 4.733464425999143e-06, "reference_active": true, "delta_db": -0.13654125995902172}], "active_band_count": 3, "active_band_mae_db": 0.18203507236210661}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015076905043222532, "log_rms_mae_db": 1.4267069014894356, "rms_envelope_nrmse": 0.10489897034688929, "envelope_correlation": 0.9890705352443623, "rms_gain_error_db": -0.12137844177353481, "reference_mean_power": 0.0024151558510913072, "output_mean_power": 0.0023485905929902825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.3965349131952106e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007806812222828868, "output_power": 0.0007410955899140707, "reference_active": true, "delta_db": -0.2259950451985618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024651476965397817, "output_power": 0.00021548154012650445, "reference_active": true, "delta_db": -0.584328737073195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0716739338834284e-05, "output_power": 2.1544053409110978e-05, "reference_active": true, "delta_db": -1.5404769527687512}], "active_band_count": 3, "active_band_mae_db": 0.7836002450135027}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00011516776273837431, "log_rms_mae_db": 0.9752372251627519, "rms_envelope_nrmse": 0.0619053041712399, "envelope_correlation": 0.995943006687498, "rms_gain_error_db": -0.12138969749335166, "reference_mean_power": 0.0024377564716401134, "output_mean_power": 0.0023705621632424474}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.190139920080037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007614290342229629, "output_power": 0.0007086444717185974, "reference_active": true, "delta_db": -0.31201029522431034}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011269364256369602, "output_power": 0.0001017401481512687, "reference_active": true, "delta_db": -0.444070509866982}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5333706288121526e-06, "output_power": 2.631678241831046e-06, "reference_active": true, "delta_db": -1.2795640481178419}], "active_band_count": 3, "active_band_mae_db": 0.6785482844030447}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00014489100768405854, "log_rms_mae_db": 1.5129380836269484, "rms_envelope_nrmse": 0.10216659287541714, "envelope_correlation": 0.9924139112669095, "rms_gain_error_db": -0.03261913875869768, "reference_mean_power": 0.0008223932804627593, "output_mean_power": 0.0008162395595502652}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 6.859948035043385e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004581143880628001, "output_power": 0.00046148864299940787, "reference_active": true, "delta_db": 0.031870857893562386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.242911398972428e-05, "output_power": 8.040350408314791e-05, "reference_active": true, "delta_db": -0.605338137349895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.706958400708153e-06, "output_power": 5.256016835769e-06, "reference_active": true, "delta_db": -2.19209721871985}], "active_band_count": 3, "active_band_mae_db": 0.9431020713211025}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00013523948700125142, "log_rms_mae_db": 1.0440445037241186, "rms_envelope_nrmse": 0.05601154234621324, "envelope_correlation": 0.9975824101542171, "rms_gain_error_db": -0.032623027592518304, "reference_mean_power": 0.000830089539077811, "output_mean_power": 0.0008238774916522401}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 6.395142215880047e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004454767193500814, "output_power": 0.000442696149490705, "reference_active": true, "delta_db": -0.027192681381432916}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.066351786105847e-05, "output_power": 3.63749897543727e-05, "reference_active": true, "delta_db": -0.484020677174666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.874405735020567e-07, "output_power": 5.571034532798007e-07, "reference_active": true, "delta_db": -2.022034301408102}], "active_band_count": 3, "active_band_mae_db": 0.8444158866547337}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 8.841298696090765e-05, "log_rms_mae_db": 1.4973567300041404, "rms_envelope_nrmse": 0.1359266500975392, "envelope_correlation": 0.9878865399778947, "rms_gain_error_db": -0.12279852194109354, "reference_mean_power": 0.0002494763103331033, "output_mean_power": 0.00024252106134480542}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.730062412281665e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013367937231840985, "output_power": 0.0001305180694475538, "reference_active": true, "delta_db": -0.10393756377476263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8654341368924343e-05, "output_power": 2.140941399082146e-05, "reference_active": true, "delta_db": -1.2658565034526879}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.319126391736523e-06, "output_power": 7.430912905866044e-06, "reference_active": true, "delta_db": -0.4903555136463427}], "active_band_count": 3, "active_band_mae_db": 0.6200498602912644}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 7.608900761326353e-05, "log_rms_mae_db": 0.9711582574616042, "rms_envelope_nrmse": 0.07323656949480417, "envelope_correlation": 0.996417010443358, "rms_gain_error_db": -0.12280165197327915, "reference_mean_power": 0.00025181099859470626, "output_mean_power": 0.00024479048348190224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 2.5627299576593646e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001307843821160996, "output_power": 0.00012633671484534988, "reference_active": true, "delta_db": -0.15026305249347008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1887209147638657e-05, "output_power": 8.935816801540367e-06, "reference_active": true, "delta_db": -1.2394564742848697}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.396397638044017e-07, "output_power": 6.728821265088505e-07, "reference_active": true, "delta_db": 0.22003538233917058}], "active_band_count": 3, "active_band_mae_db": 0.5365849697058368}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.889758424390996, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d08960c1525ff96b6f8b0ce448811deea709d3daeaea3ae27e4d5c11089a5d25", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9929094842836119, "standard_deviation": 0.0014919276249579561, "minimum": 0.9901286437531517, "maximum": 0.9947333426913684, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5324858241097499, "standard_deviation": 0.01968434290734181, "minimum": 1.4998143958265426, "maximum": 1.5521009040215643, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9566341369532037, "standard_deviation": 0.4137892539252918, "minimum": 0.4351350589327643, "maximum": 1.7062767832841168, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10175439955235405, "standard_deviation": 0.009060837307855204, "minimum": 0.0918775425565616, "maximum": 0.11941793631263484, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973115004213327, "standard_deviation": 0.0005550076906870899, "minimum": 0.9965286581785703, "maximum": 0.9978683959506937, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0923947174837, "standard_deviation": 0.032843017907438855, "minimum": 1.0440445037241186, "maximum": 1.144784191925591, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8703063119308293, "standard_deviation": 0.35186349586414434, "minimum": 0.5501449495153133, "maximum": 1.4800464765201846, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06531029080798999, "standard_deviation": 0.006102870920173526, "minimum": 0.05601154234621324, "maximum": 0.0736981665441896, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9859645494474334, "standard_deviation": 0.0033976435138261766, "minimum": 0.9803796179175199, "maximum": 0.989900412326341, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4912601822025406, "standard_deviation": 0.028279367496731482, "minimum": 1.4567852323625827, "maximum": 1.5452171693390644, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8479890735613358, "standard_deviation": 0.37302747569328454, "minimum": 0.16931101363357762, "maximum": 1.3506178787403122, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14703381147854822, "standard_deviation": 0.017586672081217204, "minimum": 0.12410119348547288, "maximum": 0.17501668835947556, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.99649449201617, "standard_deviation": 0.0010584319491820774, "minimum": 0.9943697657863949, "maximum": 0.9975188713820732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9165390498127013, "standard_deviation": 0.04098813335590521, "minimum": 0.8537421402828069, "maximum": 0.9711582574616042, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8888017394265061, "standard_deviation": 0.33218373026720077, "minimum": 0.28876895908314526, "maximum": 1.2861181128093726, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07443000974146383, "standard_deviation": 0.009266487911096069, "minimum": 0.06320848231063586, "maximum": 0.0917125711765896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9926212758550386, "standard_deviation": 0.0017172650931885965, "minimum": 0.9890705352443623, "maximum": 0.9946012469767619, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.423482112315841, "standard_deviation": 0.04113743266770353, "minimum": 1.3554867174944392, "maximum": 1.4840170550754057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7038090069282833, "standard_deviation": 0.2678062218662679, "minimum": 0.22232321011818998, "maximum": 1.0822449116538504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0860274475493754, "standard_deviation": 0.009622463369934471, "minimum": 0.07405573412723067, "maximum": 0.10489897034688929, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974699571046157, "standard_deviation": 0.0007236250485957601, "minimum": 0.995943006687498, "maximum": 0.9983264287664351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9202198982792305, "standard_deviation": 0.043668349299417215, "minimum": 0.8794042619766499, "maximum": 0.9976074229880996, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8448619071643959, "standard_deviation": 0.3341965594991694, "minimum": 0.361784292684765, "maximum": 1.384939100235335, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04871497544772907, "standard_deviation": 0.006604977369514788, "minimum": 0.039776304470789586, "maximum": 0.0619053041712399, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12398908296369895, "standard_deviation": 0.04712595332967126, "minimum": 0.04673076656796334, "maximum": 0.1789854425531335, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.331554998022087, "standard_deviation": 0.00926726022546806, "minimum": 5.322425947982654, "maximum": 5.349150501330065, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5227276550361208, "standard_deviation": 0.03411135023989265, "minimum": 0.4768772076131954, "maximum": 0.5713460430073578, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0681936064859506, "standard_deviation": 0.033434568507065866, "minimum": -0.1263684120243974, "maximum": -0.025389330819290505, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940421979145799, "standard_deviation": 0.0007280267812182448, "minimum": 0.9925641671177401, "maximum": 0.9947548340991139, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.023544127369781, "standard_deviation": 0.04477544936819825, "minimum": 0.9603626868240083, "maximum": 1.1069567712709616, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2979881285826863, "standard_deviation": 0.11259246876663155, "minimum": 0.18030286482743496, "maximum": 0.46876615737275823, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06322518729927235, "standard_deviation": 0.00357683201743984, "minimum": 0.05927359476277326, "maximum": 0.07043363060295722, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980300693082189, "standard_deviation": 0.0003863055899349898, "minimum": 0.9974435244292735, "maximum": 0.9985482946294234, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5167105741241301, "standard_deviation": 0.04726183753447413, "minimum": 0.43980227145743883, "maximum": 0.592842260901427, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3925342475410931, "standard_deviation": 0.173758911975377, "minimum": 0.18203507236210661, "maximum": 0.6110127121355095, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03475579163201174, "standard_deviation": 0.0030664888723712564, "minimum": 0.02979154963753796, "maximum": 0.03895282373613244, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0032", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.919341802597046, "predicted_mos": 2.919341802597046, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "58613f146cc4a8852966abac42695aaa3034edddac541f80e0ae9f5bec4dd45c", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.794351099990308, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.999893372285791, "全帯域_10ms_rms_envelope_nrmse": 0.058233655848327424, "全帯域_10ms_envelope_correlation": 0.9949259085537673, "全帯域_10ms_modulation_mae_db": 0.47833605401943813, "全帯域_40ms_log_rms_mae_db": 0.49982326817659356, "全帯域_40ms_rms_envelope_nrmse": 0.02962278369065699, "全帯域_40ms_envelope_correlation": 0.9985584486475422, "全帯域_40ms_modulation_mae_db": 0.4027093179765706, "300–1000 Hz_10ms_log_rms_mae_db": 1.2982255017176973, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07772428907410957, "300–1000 Hz_10ms_envelope_correlation": 0.994118679190115, "300–1000 Hz_10ms_modulation_mae_db": 0.9053929908416943, "300–1000 Hz_40ms_log_rms_mae_db": 0.7720205928424012, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04387006790446046, "300–1000 Hz_40ms_envelope_correlation": 0.9980728078501644, "300–1000 Hz_40ms_modulation_mae_db": 1.1772936952185433, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3743843319357332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.099861724597161, "1000–3000 Hz_10ms_envelope_correlation": 0.9927559265490639, "1000–3000 Hz_10ms_modulation_mae_db": 0.3421782775458861, "1000–3000 Hz_40ms_log_rms_mae_db": 0.975494217213893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052587974911277764, "1000–3000 Hz_40ms_envelope_correlation": 0.9978944641003632, "1000–3000 Hz_40ms_modulation_mae_db": 0.49128684738905565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.404452447571172, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1496613184417155, "3000–6000 Hz_10ms_envelope_correlation": 0.9856247700162043, "3000–6000 Hz_10ms_modulation_mae_db": 0.7892451661834062, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8298667328579905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07923472939215608, "3000–6000 Hz_40ms_envelope_correlation": 0.9963719867633727, "3000–6000 Hz_40ms_modulation_mae_db": 1.1560914412971954, "mr_stft_db": 5.281506294974193, "band_mae_db": 0.11390790881827721, "rms_gain_error_db": -0.053032390131944704, "output_peak": 0.4814516593584116}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0201250086443767, "全帯域_10ms_rms_envelope_nrmse": 0.06384889208239647, "全帯域_10ms_envelope_correlation": 0.9939395292580367, "全帯域_10ms_modulation_mae_db": 0.6226967144883343, "全帯域_40ms_log_rms_mae_db": 0.5759552889296814, "全帯域_40ms_rms_envelope_nrmse": 0.032478465189096854, "全帯域_40ms_envelope_correlation": 0.9983241866024453, "全帯域_40ms_modulation_mae_db": 0.4818819300413219, "300–1000 Hz_10ms_log_rms_mae_db": 1.2898720379755944, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09083509836409448, "300–1000 Hz_10ms_envelope_correlation": 0.9919678727155203, "300–1000 Hz_10ms_modulation_mae_db": 1.018759617556565, "300–1000 Hz_40ms_log_rms_mae_db": 0.7939627575637498, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05525705665342789, "300–1000 Hz_40ms_envelope_correlation": 0.9969356903772371, "300–1000 Hz_40ms_modulation_mae_db": 1.2418789313871046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4446168287158525, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09684461217537273, "1000–3000 Hz_10ms_envelope_correlation": 0.9934560454588378, "1000–3000 Hz_10ms_modulation_mae_db": 0.47380535989859895, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0602526629560332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059834226464842585, "1000–3000 Hz_40ms_envelope_correlation": 0.9976162161055184, "1000–3000 Hz_40ms_modulation_mae_db": 0.6611799531393552, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4646889583275724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13736795257596499, "3000–6000 Hz_10ms_envelope_correlation": 0.9880987276759359, "3000–6000 Hz_10ms_modulation_mae_db": 0.8130440118123602, "3000–6000 Hz_40ms_log_rms_mae_db": 0.869026162094541, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06563530423337283, "3000–6000 Hz_40ms_envelope_correlation": 0.9978706705480855, "3000–6000 Hz_40ms_modulation_mae_db": 1.1888655295357138, "mr_stft_db": 5.28936986280623, "band_mae_db": 0.1243454102424509, "rms_gain_error_db": -0.0803557416001098, "output_peak": 0.47876252719482404}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0212810402966486, "全帯域_10ms_rms_envelope_nrmse": 0.0672968586556688, "全帯域_10ms_envelope_correlation": 0.9931893288808774, "全帯域_10ms_modulation_mae_db": 0.5571116814792861, "全帯域_40ms_log_rms_mae_db": 0.5364800460874096, "全帯域_40ms_rms_envelope_nrmse": 0.03836124607071266, "全帯域_40ms_envelope_correlation": 0.9975009841458072, "全帯域_40ms_modulation_mae_db": 0.7437738339080179, "300–1000 Hz_10ms_log_rms_mae_db": 1.3965576399660036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0779836136722683, "300–1000 Hz_10ms_envelope_correlation": 0.9939268228953507, "300–1000 Hz_10ms_modulation_mae_db": 0.2524515802306932, "300–1000 Hz_40ms_log_rms_mae_db": 0.7955133382199362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045154044777999905, "300–1000 Hz_40ms_envelope_correlation": 0.9977459318213509, "300–1000 Hz_40ms_modulation_mae_db": 0.658845836474591, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4721510780863294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10357437756869438, "1000–3000 Hz_10ms_envelope_correlation": 0.992422703336042, "1000–3000 Hz_10ms_modulation_mae_db": 1.3111419164100384, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0140220302082341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05977003816559939, "1000–3000 Hz_40ms_envelope_correlation": 0.9975740520960961, "1000–3000 Hz_40ms_modulation_mae_db": 1.396553623373715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4133320513527616, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16651320820812185, "3000–6000 Hz_10ms_envelope_correlation": 0.9857042001372327, "3000–6000 Hz_10ms_modulation_mae_db": 0.6192824958600059, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8717049655933637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10087706261152231, "3000–6000 Hz_40ms_envelope_correlation": 0.996340043565605, "3000–6000 Hz_40ms_modulation_mae_db": 0.7732248074649183, "mr_stft_db": 5.295148777642743, "band_mae_db": 0.13683635789481532, "rms_gain_error_db": -0.016317042084253854, "output_peak": 0.5343726456794877}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0265687113410984, "全帯域_10ms_rms_envelope_nrmse": 0.06787863367895601, "全帯域_10ms_envelope_correlation": 0.9930867099828117, "全帯域_10ms_modulation_mae_db": 0.9368907112441405, "全帯域_40ms_log_rms_mae_db": 0.47273045544426173, "全帯域_40ms_rms_envelope_nrmse": 0.03242497005025156, "全帯域_40ms_envelope_correlation": 0.9982625679919915, "全帯域_40ms_modulation_mae_db": 1.2407581916037678, "300–1000 Hz_10ms_log_rms_mae_db": 1.3860876631906045, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07602415367654893, "300–1000 Hz_10ms_envelope_correlation": 0.9942968687750463, "300–1000 Hz_10ms_modulation_mae_db": 1.2384878865645148, "300–1000 Hz_40ms_log_rms_mae_db": 0.7881594454829962, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03713041573363276, "300–1000 Hz_40ms_envelope_correlation": 0.998524311116257, "300–1000 Hz_40ms_modulation_mae_db": 1.1865239967099692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.453996707578601, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10009757545178116, "1000–3000 Hz_10ms_envelope_correlation": 0.99273915999157, "1000–3000 Hz_10ms_modulation_mae_db": 0.5853971845504021, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0350364850833413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06186504493031533, "1000–3000 Hz_40ms_envelope_correlation": 0.9971062422477335, "1000–3000 Hz_40ms_modulation_mae_db": 0.27320624145155, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4275085715890314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1431212051560154, "3000–6000 Hz_10ms_envelope_correlation": 0.9865544621787797, "3000–6000 Hz_10ms_modulation_mae_db": 0.9693797134800958, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8178177358828655, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0738374855380646, "3000–6000 Hz_40ms_envelope_correlation": 0.9963527048712154, "3000–6000 Hz_40ms_modulation_mae_db": 0.608068951037099, "mr_stft_db": 5.274548662433166, "band_mae_db": 0.12389758291254117, "rms_gain_error_db": -0.051784676516921126, "output_peak": 0.4880634898954297}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9958719104764093, "全帯域_10ms_rms_envelope_nrmse": 0.06415814977525534, "全帯域_10ms_envelope_correlation": 0.9938666883704379, "全帯域_10ms_modulation_mae_db": 0.13642899438789205, "全帯域_40ms_log_rms_mae_db": 0.5305078155450601, "全帯域_40ms_rms_envelope_nrmse": 0.03427846759940944, "全帯域_40ms_envelope_correlation": 0.9980319868718404, "全帯域_40ms_modulation_mae_db": 0.11089889316222097, "300–1000 Hz_10ms_log_rms_mae_db": 1.3962295057079392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07767286303697933, "300–1000 Hz_10ms_envelope_correlation": 0.9940744049549893, "300–1000 Hz_10ms_modulation_mae_db": 0.3178186350545409, "300–1000 Hz_40ms_log_rms_mae_db": 0.895751683153367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05013869926459612, "300–1000 Hz_40ms_envelope_correlation": 0.9972574666059063, "300–1000 Hz_40ms_modulation_mae_db": 0.40025632126236294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4790890636008927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08651313745942343, "1000–3000 Hz_10ms_envelope_correlation": 0.9945694330073901, "1000–3000 Hz_10ms_modulation_mae_db": 0.15743010582746694, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0116487432705799, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04837592853785454, "1000–3000 Hz_40ms_envelope_correlation": 0.998203819721403, "1000–3000 Hz_40ms_modulation_mae_db": 0.19943688544933824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3883154190541314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15758940126516516, "3000–6000 Hz_10ms_envelope_correlation": 0.983682227439274, "3000–6000 Hz_10ms_modulation_mae_db": 0.9673035814628476, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8065036649834447, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08827617075457524, "3000–6000 Hz_40ms_envelope_correlation": 0.9947399068460604, "3000–6000 Hz_40ms_modulation_mae_db": 0.8928849009787986, "mr_stft_db": 5.261674325682765, "band_mae_db": 0.16340890913997352, "rms_gain_error_db": 0.015152856830952832, "output_peak": 0.6254323482838882}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8972700597411399, "全帯域_10ms_rms_envelope_nrmse": 0.057290495074549225, "全帯域_10ms_envelope_correlation": 0.9951892925110274, "全帯域_10ms_modulation_mae_db": 0.4189292440714984, "全帯域_40ms_log_rms_mae_db": 0.41772557930398074, "全帯域_40ms_rms_envelope_nrmse": 0.03254302324435171, "全帯域_40ms_envelope_correlation": 0.9983674458300554, "全帯域_40ms_modulation_mae_db": 0.33609189276818113, "300–1000 Hz_10ms_log_rms_mae_db": 1.3224050729093273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08840145818362768, "300–1000 Hz_10ms_envelope_correlation": 0.992437610425425, "300–1000 Hz_10ms_modulation_mae_db": 1.3080171362427178, "300–1000 Hz_40ms_log_rms_mae_db": 0.7964503991062191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05710848336207617, "300–1000 Hz_40ms_envelope_correlation": 0.9966759938143446, "300–1000 Hz_40ms_modulation_mae_db": 1.4013017801091223, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4728540105750676, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09013678448270881, "1000–3000 Hz_10ms_envelope_correlation": 0.9941847284234555, "1000–3000 Hz_10ms_modulation_mae_db": 0.3519613965737925, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0623512813397267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05661288355017963, "1000–3000 Hz_40ms_envelope_correlation": 0.9976647067605898, "1000–3000 Hz_40ms_modulation_mae_db": 0.2707926395775844, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3860116393095439, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13832722074034373, "3000–6000 Hz_10ms_envelope_correlation": 0.9893762282588886, "3000–6000 Hz_10ms_modulation_mae_db": 1.150277736747767, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8380657340764792, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08900411363205157, "3000–6000 Hz_40ms_envelope_correlation": 0.9970450169764037, "3000–6000 Hz_40ms_modulation_mae_db": 1.2524671537472971, "mr_stft_db": 5.299932908907967, "band_mae_db": 0.14918178298343523, "rms_gain_error_db": -0.094702831666673, "output_peak": 0.5459263186631731}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.017890112729748, "全帯域_10ms_rms_envelope_nrmse": 0.06557065010881513, "全帯域_10ms_envelope_correlation": 0.9935644987302404, "全帯域_10ms_modulation_mae_db": 0.361213745291896, "全帯域_40ms_log_rms_mae_db": 0.5182921815656001, "全帯域_40ms_rms_envelope_nrmse": 0.038083107872949405, "全帯域_40ms_envelope_correlation": 0.9975151544310207, "全帯域_40ms_modulation_mae_db": 0.5001385473393921, "300–1000 Hz_10ms_log_rms_mae_db": 1.3488847627418543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0756586759917814, "300–1000 Hz_10ms_envelope_correlation": 0.9943572058336284, "300–1000 Hz_10ms_modulation_mae_db": 0.34787066842701647, "300–1000 Hz_40ms_log_rms_mae_db": 0.7469924967884711, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042849712236623254, "300–1000 Hz_40ms_envelope_correlation": 0.9980720737895908, "300–1000 Hz_40ms_modulation_mae_db": 0.6362628486584762, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4102204971045327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09002230292915962, "1000–3000 Hz_10ms_envelope_correlation": 0.9941912155126, "1000–3000 Hz_10ms_modulation_mae_db": 0.7792520779694367, "1000–3000 Hz_40ms_log_rms_mae_db": 0.987043902461649, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05521611040607613, "1000–3000 Hz_40ms_envelope_correlation": 0.9977069306532442, "1000–3000 Hz_40ms_modulation_mae_db": 0.380339926138405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5090567044887666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14177428115094584, "3000–6000 Hz_10ms_envelope_correlation": 0.9868536881484841, "3000–6000 Hz_10ms_modulation_mae_db": 1.023127604498164, "3000–6000 Hz_40ms_log_rms_mae_db": 0.914903240742557, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07053369772540355, "3000–6000 Hz_40ms_envelope_correlation": 0.9967913602187587, "3000–6000 Hz_40ms_modulation_mae_db": 1.307821817958617, "mr_stft_db": 5.301828818069018, "band_mae_db": 0.05819994248377547, "rms_gain_error_db": 0.003139931270748169, "output_peak": 0.5159162467959352}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0585213613291051, "全帯域_10ms_rms_envelope_nrmse": 0.06243336913470787, "全帯域_10ms_envelope_correlation": 0.9942575765159651, "全帯域_10ms_modulation_mae_db": 0.35207159233606156, "全帯域_40ms_log_rms_mae_db": 0.5036657257017113, "全帯域_40ms_rms_envelope_nrmse": 0.03258328835013025, "全帯域_40ms_envelope_correlation": 0.9983412638261063, "全帯域_40ms_modulation_mae_db": 0.2878902794553839, "300–1000 Hz_10ms_log_rms_mae_db": 1.2830408225421608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06745196184552667, "300–1000 Hz_10ms_envelope_correlation": 0.9955406732043414, "300–1000 Hz_10ms_modulation_mae_db": 0.8807754130666142, "300–1000 Hz_40ms_log_rms_mae_db": 0.7119675972761526, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04078070663653025, "300–1000 Hz_40ms_envelope_correlation": 0.9982320177696068, "300–1000 Hz_40ms_modulation_mae_db": 0.9426866515467456, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4434541296860428, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09529583185943981, "1000–3000 Hz_10ms_envelope_correlation": 0.9944456968038428, "1000–3000 Hz_10ms_modulation_mae_db": 0.7858038275279435, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0148764103628076, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06133677838964493, "1000–3000 Hz_40ms_envelope_correlation": 0.9983635956631145, "1000–3000 Hz_40ms_modulation_mae_db": 0.7353885007034603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4617136746441333, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14296151702030602, "3000–6000 Hz_10ms_envelope_correlation": 0.9869825346853806, "3000–6000 Hz_10ms_modulation_mae_db": 1.4988570561449315, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8962913207363039, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07286028409966117, "3000–6000 Hz_40ms_envelope_correlation": 0.9970884416425917, "3000–6000 Hz_40ms_modulation_mae_db": 1.6208339276759975, "mr_stft_db": 5.28416863538947, "band_mae_db": 0.1837777460129941, "rms_gain_error_db": -0.07617842362340406, "output_peak": 0.4893307871789737}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 51840, "output_samples": 51840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.232069709990422, "spectral_convergence": 0.2505136769223686, "spectral_convergence_status": "ok", "amplitude_floor": 1.3508970735580905e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.108500018725052, "spectral_convergence": 0.22068820330437744, "spectral_convergence_status": "ok", "amplitude_floor": 9.199259876037329e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.503949156207104, "spectral_convergence": 0.2319960159303154, "spectral_convergence_status": "ok", "amplitude_floor": 7.633709769762664e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012491242347822075, "output_power": 0.0012481145127735587, "delta_db": -0.0035120127856332257, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0036291285647318994, "output_power": 0.0035888624097187806, "delta_db": -0.04845545083284676, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000626458342936399, "output_power": 0.000624665318572723, "delta_db": -0.012448028026889517, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008149528003014777, "output_power": 0.0008043482286329612, "delta_db": -0.05688346660190674, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002174945257645064, "output_power": 0.00020309516353971826, "delta_db": -0.2974874909754739, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023982804336185733, "output_power": 0.00022564923416761327, "delta_db": -0.2646610036869132, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11390790881827721, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.053032390131944704, "pitch": null, "output_rms_dbfs": -21.742666177412723, "output_peak": 0.4814516593584116, "reference_rms_dbfs": -21.689633787280776, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.281506294974193}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 51840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00019883564266349852, "log_rms_mae_db": 0.999893372285791, "rms_envelope_nrmse": 0.058233655848327424, "envelope_correlation": 0.9949259085537673, "rms_gain_error_db": -0.053019955542184444, "reference_mean_power": 0.006787407573323676, "output_mean_power": 0.0067050486470806555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.1392663194097437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011382823319318316, "output_power": 0.0011301935290898707, "reference_active": true, "delta_db": -0.030971784962081365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00037355144462451696, "output_power": 0.00035093729642808, "reference_active": true, "delta_db": -0.27120894266868156}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.066072767849799e-05, "output_power": 4.673313032090793e-05, "reference_active": true, "delta_db": -1.1328274344275515}], "active_band_count": 3, "active_band_mae_db": 0.47833605401943813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00017748806718771712, "log_rms_mae_db": 0.49982326817659356, "rms_envelope_nrmse": 0.02962278369065699, "envelope_correlation": 0.9985584486475422, "rms_gain_error_db": -0.052982889409094056, "reference_mean_power": 0.006850619249749789, "output_mean_power": 0.006767551067554573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.7219585966772077e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000998012429960962, "output_power": 0.0009996632352628192, "reference_active": true, "delta_db": 0.007177699653318763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018464455345700424, "output_power": 0.00016978526997858903, "reference_active": true, "delta_db": -0.3643649204101765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.884647670701786e-06, "output_power": 4.0287907394897916e-06, "reference_active": true, "delta_db": -0.8365853338662167}], "active_band_count": 3, "active_band_mae_db": 0.4027093179765706}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015076905043222532, "log_rms_mae_db": 1.2982255017176973, "rms_envelope_nrmse": 0.07772428907410957, "envelope_correlation": 0.994118679190115, "rms_gain_error_db": -0.12743629234503906, "reference_mean_power": 0.0024151558510913072, "output_mean_power": 0.0023453168942986218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.3965349131952106e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007806812222828868, "output_power": 0.0007564137484466254, "reference_active": true, "delta_db": -0.1371431950297907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024651476965397817, "output_power": 0.00020742197319544874, "reference_active": true, "delta_db": -0.7498818328747614}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0716739338834284e-05, "output_power": 2.015856985116913e-05, "reference_active": true, "delta_db": -1.8291539446205307}], "active_band_count": 3, "active_band_mae_db": 0.9053929908416943}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00011516776273837431, "log_rms_mae_db": 0.7720205928424012, "rms_envelope_nrmse": 0.04387006790446046, "envelope_correlation": 0.9980728078501644, "rms_gain_error_db": -0.12743956353583902, "reference_mean_power": 0.0024377564716401134, "output_mean_power": 0.002367262190616398}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.190139920080037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007614290342229629, "output_power": 0.0007255281608377634, "reference_active": true, "delta_db": -0.2097515905604208}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011269364256369602, "output_power": 9.54047789292098e-05, "reference_active": true, "delta_db": -0.7232928715822258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5333706288121526e-06, "output_power": 1.9422517985771783e-06, "reference_active": true, "delta_db": -2.5988366235129834}], "active_band_count": 3, "active_band_mae_db": 1.1772936952185433}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00014489100768405854, "log_rms_mae_db": 1.3743843319357332, "rms_envelope_nrmse": 0.099861724597161, "envelope_correlation": 0.9927559265490639, "rms_gain_error_db": -0.060378631180749195, "reference_mean_power": 0.0008223932804627593, "output_mean_power": 0.0008110389097121439}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 6.859948035043385e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004581143880628001, "output_power": 0.0004469112527051869, "reference_active": true, "delta_db": -0.10752642110342742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.242911398972428e-05, "output_power": 8.071424456299422e-05, "reference_active": true, "delta_db": -0.5885860359626753}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.706958400708153e-06, "output_power": 8.069083915185859e-06, "reference_active": true, "delta_db": -0.3304223755715553}], "active_band_count": 3, "active_band_mae_db": 0.3421782775458861}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00013523948700125142, "log_rms_mae_db": 0.975494217213893, "rms_envelope_nrmse": 0.052587974911277764, "envelope_correlation": 0.9978944641003632, "rms_gain_error_db": -0.06038196018402561, "reference_mean_power": 0.000830089539077811, "output_mean_power": 0.0008186282824475876}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 6.395142215880047e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004454767193500814, "output_power": 0.00042751756131822647, "reference_active": true, "delta_db": -0.17871053573413148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.066351786105847e-05, "output_power": 3.563821638108093e-05, "reference_active": true, "delta_db": -0.5728898701572415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.874405735020567e-07, "output_power": 1.0480096805230443e-06, "reference_active": true, "delta_db": 0.722260136275794}], "active_band_count": 3, "active_band_mae_db": 0.49128684738905565}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 8.841298696090765e-05, "log_rms_mae_db": 1.404452447571172, "rms_envelope_nrmse": 0.1496613184417155, "envelope_correlation": 0.9856247700162043, "rms_gain_error_db": -0.29882985508651805, "reference_mean_power": 0.0002494763103331033, "output_mean_power": 0.00023288757936949548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.730062412281665e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013367937231840985, "output_power": 0.00012805145402684443, "reference_active": true, "delta_db": -0.18679883437501282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8654341368924343e-05, "output_power": 2.9428479701678763e-05, "reference_active": true, "delta_db": 0.11577396203367894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.319126391736523e-06, "output_power": 5.170844198343528e-06, "reference_active": true, "delta_db": -2.0651627021415266}], "active_band_count": 3, "active_band_mae_db": 0.7892451661834062}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 7.608900761326353e-05, "log_rms_mae_db": 0.8298667328579905, "rms_envelope_nrmse": 0.07923472939215608, "envelope_correlation": 0.9963719867633727, "rms_gain_error_db": -0.2988337738596222, "reference_mean_power": 0.00025181099859470626, "output_mean_power": 0.00023506681226398066}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 2.5627299576593646e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001307843821160996, "output_power": 0.00012669302470948973, "reference_active": true, "delta_db": -0.13803180170065837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1887209147638657e-05, "output_power": 1.2937976011481874e-05, "reference_active": true, "delta_db": 0.3678643767299068}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.396397638044017e-07, "output_power": 3.233684460851586e-07, "reference_active": true, "delta_db": -2.9623781454610216}], "active_band_count": 3, "active_band_mae_db": 1.1560914412971954}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.893165231481723, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3b410d132f0c072e6a6235790dd75a3475d54c1a15397e75c1b76fe99cb95ef8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9935956136353503, "standard_deviation": 0.000862434137062987, "minimum": 0.992422703336042, "maximum": 0.9945694330073901, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4438458309103814, "standard_deviation": 0.03574532543581479, "minimum": 1.3743843319357332, "maximum": 1.4790890636008927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5983712682879456, "standard_deviation": 0.36102188357017756, "minimum": 0.15743010582746694, "maximum": 1.3111419164100384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09529329331546763, "standard_deviation": 0.005932943188316004, "minimum": 0.08651313745942343, "maximum": 0.10357437756869438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9977662534185079, "standard_deviation": 0.0003916299654645984, "minimum": 0.9971062422477335, "maximum": 0.9983635956631145, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.020090716612033, "standard_deviation": 0.03125594298397937, "minimum": 0.975494217213893, "maximum": 1.0623512813397267, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.551023077152808, "standard_deviation": 0.3915303945593416, "minimum": 0.19943688544933824, "maximum": 1.396553623373715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056949873169473794, "standard_deviation": 0.004710699882355947, "minimum": 0.04837592853785454, "maximum": 0.06186504493031533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9866096048175225, "standard_deviation": 0.0017084906354033291, "minimum": 0.983682227439274, "maximum": 0.9893762282588886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.431884933292139, "standard_deviation": 0.04316067167057739, "minimum": 1.3860116393095439, "maximum": 1.5090567044887666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9788146707736973, "standard_deviation": 0.265843326176638, "minimum": 0.6192824958600059, "maximum": 1.4988570561449315, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1471645130698223, "standard_deviation": 0.010186985973354256, "minimum": 0.13736795257596499, "maximum": 0.16651320820812185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9965750164290116, "standard_deviation": 0.0009042370973809966, "minimum": 0.9947399068460604, "maximum": 0.9978706705480855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8555224446209432, "standard_deviation": 0.038581717886743215, "minimum": 0.8065036649834447, "maximum": 0.914903240742557, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1000323162119545, "standard_deviation": 0.3252300927506629, "minimum": 0.608068951037099, "maximum": 1.6208339276759975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08003235599835092, "standard_deviation": 0.01177713015378475, "minimum": 0.06563530423337283, "maximum": 0.10087706261152231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.993840017249302, "standard_deviation": 0.0011326469236003658, "minimum": 0.9919678727155203, "maximum": 0.9955406732043414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3401628758438977, "standard_deviation": 0.048366419976152236, "minimum": 1.2830408225421608, "maximum": 1.3965576399660036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7836967409980445, "standard_deviation": 0.42265214684971075, "minimum": 0.2524515802306932, "maximum": 1.3080171362427178, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07896901423061704, "standard_deviation": 0.007428212177340971, "minimum": 0.06745196184552667, "maximum": 0.09083509836409448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976895366430572, "standard_deviation": 0.0006622239448261809, "minimum": 0.9966759938143446, "maximum": 0.998524311116257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7876022888041616, "standard_deviation": 0.05276698408017496, "minimum": 0.7119675972761526, "maximum": 0.895751683153367, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9556312576708644, "standard_deviation": 0.3549028615927617, "minimum": 0.40025632126236294, "maximum": 1.4013017801091223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04653614832116835, "standard_deviation": 0.007020457902731748, "minimum": 0.03713041573363276, "maximum": 0.05710848336207617, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13169445506103286, "standard_deviation": 0.03757530700908578, "minimum": 0.05819994248377547, "maximum": 0.1837777460129941, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.286022285738194, "standard_deviation": 0.013554906390620248, "minimum": 5.261674325682765, "maximum": 5.301828818069018, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5199070028812653, "standard_deviation": 0.04950956600601089, "minimum": 0.47876252719482404, "maximum": 0.6254323482838882, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04425978969020069, "standard_deviation": 0.0406328387921165, "minimum": -0.094702831666673, "maximum": 0.015152856830952832, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940024416003955, "standard_deviation": 0.0007597365777547652, "minimum": 0.9930867099828117, "maximum": 0.9951892925110274, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0046776971055396, "standard_deviation": 0.04736491657206234, "minimum": 0.8972700597411399, "maximum": 1.0585213613291051, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4829598421648184, "standard_deviation": 0.23520109756808075, "minimum": 0.13642899438789205, "maximum": 0.9368907112441405, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06333883804483453, "standard_deviation": 0.003885013244509209, "minimum": 0.057290495074549225, "maximum": 0.06787863367895601, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981127547933512, "standard_deviation": 0.0004001405634585565, "minimum": 0.9975009841458072, "maximum": 0.9985584486475422, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5068975450942873, "standard_deviation": 0.04707273780054186, "minimum": 0.41772557930398074, "maximum": 0.5759552889296814, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.513017860781857, "standard_deviation": 0.34637768319377416, "minimum": 0.11089889316222097, "maximum": 1.2407581916037678, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033796919008444865, "standard_deviation": 0.0030124861491055595, "minimum": 0.02962278369065699, "maximum": 0.03836124607071266, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2094-142345-0039", "language": "en", "style": "normal", "speaker": "librispeech_2094", "source_utterance_id": "2094-142345-0039", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2094-142345-0039.wav", "source_file_sha256": "2f04268a2db940cb246c02e7dac1acda4e4829d982faeffb98fd5ce7579fde9a", "decoded_pcm_sha256": "c411d502f97ab6fc22d81b52c4ba95e5e562514c255c2379c8837148877e0ea1", "original_source_sha256": "21d7d00350a4ca2b0109da67cfad724f36979e7c1b2e098f9327bc8c4f433435", "corpus_entry_id": "librispeech_2094-142345-0039"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.157634258270264, "predicted_mos": 4.157634258270264, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "862eea5ada3154f6046f407df3d570df2b5f37f26eaecb34ab0287c9c6e8325a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.768175152999902, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2094-142345-0039", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "862eea5ada3154f6046f407df3d570df2b5f37f26eaecb34ab0287c9c6e8325a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "e88159d23ff9307495753f2c3452423ef53b095a3206159f6a200988e22b9758"}, "rows": [{"case": "librispeech_2094-142345-0039", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.5864269733428955, "predicted_mos": 2.5864269733428955, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d4fa9af815dcdcb6a917be5939859ca547ccc6af13045f07b652cc8f11fe50e2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.699199062000844, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2094-142345-0039", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d4fa9af815dcdcb6a917be5939859ca547ccc6af13045f07b652cc8f11fe50e2", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9211519682495093, "全帯域_10ms_rms_envelope_nrmse": 0.06627934917598212, "全帯域_10ms_envelope_correlation": 0.9927523970318382, "全帯域_10ms_modulation_mae_db": 0.2318632157969814, "全帯域_40ms_log_rms_mae_db": 0.42578884272893663, "全帯域_40ms_rms_envelope_nrmse": 0.03089672591679321, "全帯域_40ms_envelope_correlation": 0.9983257058958135, "全帯域_40ms_modulation_mae_db": 0.13488762153420922, "300–1000 Hz_10ms_log_rms_mae_db": 1.461141020899775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08530882072183069, "300–1000 Hz_10ms_envelope_correlation": 0.9928378511916551, "300–1000 Hz_10ms_modulation_mae_db": 0.8045172684353442, "300–1000 Hz_40ms_log_rms_mae_db": 0.8586916430996869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04811934732701749, "300–1000 Hz_40ms_envelope_correlation": 0.9976122487225081, "300–1000 Hz_40ms_modulation_mae_db": 0.5878279345319412, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4368425570336913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1146086202917707, "1000–3000 Hz_10ms_envelope_correlation": 0.9903099542688057, "1000–3000 Hz_10ms_modulation_mae_db": 0.5826478044840943, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9217627841210673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07351329675405596, "1000–3000 Hz_40ms_envelope_correlation": 0.9964054597980435, "1000–3000 Hz_40ms_modulation_mae_db": 0.5624206734443645, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5989771126299008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16734401241551958, "3000–6000 Hz_10ms_envelope_correlation": 0.9842554082881957, "3000–6000 Hz_10ms_modulation_mae_db": 0.31249121304431376, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9783149828382999, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08262839593019974, "3000–6000 Hz_40ms_envelope_correlation": 0.9968283981714191, "3000–6000 Hz_40ms_modulation_mae_db": 0.45974697073759846, "mr_stft_db": 5.2391404560336055, "band_mae_db": 0.1724530785418595, "rms_gain_error_db": -0.06921608532613266, "output_peak": 0.5042951622658332}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8670859805678351, "全帯域_10ms_rms_envelope_nrmse": 0.061095988808316806, "全帯域_10ms_envelope_correlation": 0.9938904107860531, "全帯域_10ms_modulation_mae_db": 0.2627060353621861, "全帯域_40ms_log_rms_mae_db": 0.44682602429852764, "全帯域_40ms_rms_envelope_nrmse": 0.032361056092268435, "全帯域_40ms_envelope_correlation": 0.9981867761390132, "全帯域_40ms_modulation_mae_db": 0.3818703576182709, "300–1000 Hz_10ms_log_rms_mae_db": 1.324064103918365, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08195862509124564, "300–1000 Hz_10ms_envelope_correlation": 0.9934270656380254, "300–1000 Hz_10ms_modulation_mae_db": 0.9329148553749201, "300–1000 Hz_40ms_log_rms_mae_db": 0.7802463204454153, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043987210643543176, "300–1000 Hz_40ms_envelope_correlation": 0.9980624938948516, "300–1000 Hz_40ms_modulation_mae_db": 0.9500426813043906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4505681918003381, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11885154118998534, "1000–3000 Hz_10ms_envelope_correlation": 0.9892197087402685, "1000–3000 Hz_10ms_modulation_mae_db": 0.5576701627559787, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8977037371445338, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07281845482008371, "1000–3000 Hz_40ms_envelope_correlation": 0.996154180025772, "1000–3000 Hz_40ms_modulation_mae_db": 0.5022679085882488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.589387776220349, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14942749684723006, "3000–6000 Hz_10ms_envelope_correlation": 0.9861508107578126, "3000–6000 Hz_10ms_modulation_mae_db": 0.7129917881951333, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0012942232685018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07450904385279546, "3000–6000 Hz_40ms_envelope_correlation": 0.9964440900961857, "3000–6000 Hz_40ms_modulation_mae_db": 0.9113171770029956, "mr_stft_db": 5.228585405728522, "band_mae_db": 0.11695198708276948, "rms_gain_error_db": -0.0794253157144057, "output_peak": 0.492194022717072}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8758320486688752, "全帯域_10ms_rms_envelope_nrmse": 0.06800424975851241, "全帯域_10ms_envelope_correlation": 0.9923857981886849, "全帯域_10ms_modulation_mae_db": 0.24987612630982303, "全帯域_40ms_log_rms_mae_db": 0.4065213928253638, "全帯域_40ms_rms_envelope_nrmse": 0.0335617464982536, "全帯域_40ms_envelope_correlation": 0.9980081408115773, "全帯域_40ms_modulation_mae_db": 0.37343220895025203, "300–1000 Hz_10ms_log_rms_mae_db": 1.3157472063684996, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08874606690295687, "300–1000 Hz_10ms_envelope_correlation": 0.9922370399034244, "300–1000 Hz_10ms_modulation_mae_db": 0.9528461608351174, "300–1000 Hz_40ms_log_rms_mae_db": 0.742727606619742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04909347418983877, "300–1000 Hz_40ms_envelope_correlation": 0.9975160629711379, "300–1000 Hz_40ms_modulation_mae_db": 0.9663074853293172, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4288922119685579, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11685994178643525, "1000–3000 Hz_10ms_envelope_correlation": 0.9897425074542694, "1000–3000 Hz_10ms_modulation_mae_db": 0.6073925090811115, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9063962861841842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0697335938291021, "1000–3000 Hz_40ms_envelope_correlation": 0.9967168387526875, "1000–3000 Hz_40ms_modulation_mae_db": 0.5131431095873068, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6309779935346451, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14063739478881762, "3000–6000 Hz_10ms_envelope_correlation": 0.9878624036483753, "3000–6000 Hz_10ms_modulation_mae_db": 0.13744709939527652, "3000–6000 Hz_40ms_log_rms_mae_db": 0.953336273905341, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06581187813607399, "3000–6000 Hz_40ms_envelope_correlation": 0.9972982356012217, "3000–6000 Hz_40ms_modulation_mae_db": 0.13378204219550477, "mr_stft_db": 5.249559547899622, "band_mae_db": 0.13822025862900175, "rms_gain_error_db": -0.0704786925158357, "output_peak": 0.49338743227821663}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8956813725459041, "全帯域_10ms_rms_envelope_nrmse": 0.0637709441463012, "全帯域_10ms_envelope_correlation": 0.9933123982685098, "全帯域_10ms_modulation_mae_db": 0.3677168537338373, "全帯域_40ms_log_rms_mae_db": 0.39319334546908613, "全帯域_40ms_rms_envelope_nrmse": 0.030975887615059487, "全帯域_40ms_envelope_correlation": 0.9983074883297766, "全帯域_40ms_modulation_mae_db": 0.2099679139238564, "300–1000 Hz_10ms_log_rms_mae_db": 1.331609048083022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08337456670857316, "300–1000 Hz_10ms_envelope_correlation": 0.9930819389764207, "300–1000 Hz_10ms_modulation_mae_db": 0.4884588454973479, "300–1000 Hz_40ms_log_rms_mae_db": 0.8043986465321317, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04487119007212912, "300–1000 Hz_40ms_envelope_correlation": 0.9978132779120782, "300–1000 Hz_40ms_modulation_mae_db": 0.4538720820208772, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4159159835242658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12005682878480022, "1000–3000 Hz_10ms_envelope_correlation": 0.9889257177663189, "1000–3000 Hz_10ms_modulation_mae_db": 0.8011106863260142, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8660764669652833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06952881567703204, "1000–3000 Hz_40ms_envelope_correlation": 0.9964837460239037, "1000–3000 Hz_40ms_modulation_mae_db": 0.7156015547665403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5840514894986466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1424247950367233, "3000–6000 Hz_10ms_envelope_correlation": 0.9873153676345637, "3000–6000 Hz_10ms_modulation_mae_db": 0.43178476652803416, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9559473651940968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06598161360460732, "3000–6000 Hz_40ms_envelope_correlation": 0.9973001171752273, "3000–6000 Hz_40ms_modulation_mae_db": 0.7422132833203422, "mr_stft_db": 5.231053603736025, "band_mae_db": 0.09653614587095055, "rms_gain_error_db": -0.057474868033333924, "output_peak": 0.5194161996685783}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8614958197896528, "全帯域_10ms_rms_envelope_nrmse": 0.0626985469209515, "全帯域_10ms_envelope_correlation": 0.9935252485048783, "全帯域_10ms_modulation_mae_db": 0.4252381793644215, "全帯域_40ms_log_rms_mae_db": 0.44734305230060045, "全帯域_40ms_rms_envelope_nrmse": 0.032577519152243564, "全帯域_40ms_envelope_correlation": 0.9981220349141147, "全帯域_40ms_modulation_mae_db": 0.6882667961115262, "300–1000 Hz_10ms_log_rms_mae_db": 1.4006611446288368, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07703996994520164, "300–1000 Hz_10ms_envelope_correlation": 0.9942263461868507, "300–1000 Hz_10ms_modulation_mae_db": 0.7587629515938011, "300–1000 Hz_40ms_log_rms_mae_db": 0.828963359213028, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04592781456144456, "300–1000 Hz_40ms_envelope_correlation": 0.9978596262711416, "300–1000 Hz_40ms_modulation_mae_db": 1.1173012162671634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.390412306996713, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11115201847896602, "1000–3000 Hz_10ms_envelope_correlation": 0.9904543195544163, "1000–3000 Hz_10ms_modulation_mae_db": 0.42408501263579756, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8889818084639357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0686499903909404, "1000–3000 Hz_40ms_envelope_correlation": 0.9964032501272586, "1000–3000 Hz_40ms_modulation_mae_db": 0.23988071830681257, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6535996531103956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13495974461165064, "3000–6000 Hz_10ms_envelope_correlation": 0.9887418044364237, "3000–6000 Hz_10ms_modulation_mae_db": 0.7201416887607827, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0341861084614161, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07202585931025576, "3000–6000 Hz_40ms_envelope_correlation": 0.9969355858405907, "3000–6000 Hz_40ms_modulation_mae_db": 0.5531416812376749, "mr_stft_db": 5.236622290305534, "band_mae_db": 0.11622042434988504, "rms_gain_error_db": -0.06845567276131304, "output_peak": 0.501217877611318}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9126234844578324, "全帯域_10ms_rms_envelope_nrmse": 0.061862123186200275, "全帯域_10ms_envelope_correlation": 0.9936463474825996, "全帯域_10ms_modulation_mae_db": 0.10998499603478157, "全帯域_40ms_log_rms_mae_db": 0.4516451086071138, "全帯域_40ms_rms_envelope_nrmse": 0.029001900868565952, "全帯域_40ms_envelope_correlation": 0.9984474795116342, "全帯域_40ms_modulation_mae_db": 0.225998774185984, "300–1000 Hz_10ms_log_rms_mae_db": 1.4147410720059757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07343218858467677, "300–1000 Hz_10ms_envelope_correlation": 0.9947327065942503, "300–1000 Hz_10ms_modulation_mae_db": 0.8020602871336563, "300–1000 Hz_40ms_log_rms_mae_db": 0.805759154902618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03885843451493066, "300–1000 Hz_40ms_envelope_correlation": 0.9984566866422246, "300–1000 Hz_40ms_modulation_mae_db": 0.8595403790206192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4059153576644905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10898273373997673, "1000–3000 Hz_10ms_envelope_correlation": 0.990850920174719, "1000–3000 Hz_10ms_modulation_mae_db": 0.4226853873989576, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8591766606782643, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06472063597302001, "1000–3000 Hz_40ms_envelope_correlation": 0.9968676983296945, "1000–3000 Hz_40ms_modulation_mae_db": 0.18804678249709736, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5958484382605234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1298762289130758, "3000–6000 Hz_10ms_envelope_correlation": 0.9896803430081274, "3000–6000 Hz_10ms_modulation_mae_db": 0.4560548969179104, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9923896535861723, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06505154294002116, "3000–6000 Hz_40ms_envelope_correlation": 0.9976887894291055, "3000–6000 Hz_40ms_modulation_mae_db": 0.24904789057766566, "mr_stft_db": 5.2355324060216235, "band_mae_db": 0.07328927171201349, "rms_gain_error_db": -0.026964783259574434, "output_peak": 0.5059033824629005}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8729991708535251, "全帯域_10ms_rms_envelope_nrmse": 0.05848830386604096, "全帯域_10ms_envelope_correlation": 0.9943705873469462, "全帯域_10ms_modulation_mae_db": 0.1719608107287286, "全帯域_40ms_log_rms_mae_db": 0.39468308285800086, "全帯域_40ms_rms_envelope_nrmse": 0.03012170645901407, "全帯域_40ms_envelope_correlation": 0.9984091993297647, "全帯域_40ms_modulation_mae_db": 0.1994687083945795, "300–1000 Hz_10ms_log_rms_mae_db": 1.3740927256270445, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08152976002707597, "300–1000 Hz_10ms_envelope_correlation": 0.9933924703102776, "300–1000 Hz_10ms_modulation_mae_db": 0.5126627268341998, "300–1000 Hz_40ms_log_rms_mae_db": 0.8142062389795747, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045122468513599745, "300–1000 Hz_40ms_envelope_correlation": 0.9977908942323318, "300–1000 Hz_40ms_modulation_mae_db": 0.6837788714239615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4267310108987956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11459145264067058, "1000–3000 Hz_10ms_envelope_correlation": 0.9904625014795906, "1000–3000 Hz_10ms_modulation_mae_db": 0.7757472680628267, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8921135165858889, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07005488826834337, "1000–3000 Hz_40ms_envelope_correlation": 0.9970347914604799, "1000–3000 Hz_40ms_modulation_mae_db": 0.45313762982540756, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6263473905412202, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1397766876883473, "3000–6000 Hz_10ms_envelope_correlation": 0.9884736682639068, "3000–6000 Hz_10ms_modulation_mae_db": 0.8300160177812647, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9739139767032526, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07657911231982599, "3000–6000 Hz_40ms_envelope_correlation": 0.996681468064463, "3000–6000 Hz_40ms_modulation_mae_db": 0.5675047407631119, "mr_stft_db": 5.228440008349257, "band_mae_db": 0.13211421206931065, "rms_gain_error_db": -0.06379061445924725, "output_peak": 0.4958244541840948}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8432612023494925, "全帯域_10ms_rms_envelope_nrmse": 0.06490630439849208, "全帯域_10ms_envelope_correlation": 0.9931032135760424, "全帯域_10ms_modulation_mae_db": 0.22588233876869857, "全帯域_40ms_log_rms_mae_db": 0.41878115531883847, "全帯域_40ms_rms_envelope_nrmse": 0.033028578134802956, "全帯域_40ms_envelope_correlation": 0.9981056390530317, "全帯域_40ms_modulation_mae_db": 0.1423859353720347, "300–1000 Hz_10ms_log_rms_mae_db": 1.426845351451542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08213682010456655, "300–1000 Hz_10ms_envelope_correlation": 0.9933440141700948, "300–1000 Hz_10ms_modulation_mae_db": 0.5322638764858035, "300–1000 Hz_40ms_log_rms_mae_db": 0.8444756170861771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044928778402739414, "300–1000 Hz_40ms_envelope_correlation": 0.9978875127993215, "300–1000 Hz_40ms_modulation_mae_db": 0.44032074968125823, "1000–3000 Hz_10ms_log_rms_mae_db": 1.386061786118226, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10126551640764912, "1000–3000 Hz_10ms_envelope_correlation": 0.9922427478522995, "1000–3000 Hz_10ms_modulation_mae_db": 0.5506677936903196, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8925590299406693, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06142242737574523, "1000–3000 Hz_40ms_envelope_correlation": 0.9973186260301619, "1000–3000 Hz_40ms_modulation_mae_db": 0.3213935106526854, "3000–6000 Hz_10ms_log_rms_mae_db": 1.662574468633796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15998653323563208, "3000–6000 Hz_10ms_envelope_correlation": 0.9840420571821726, "3000–6000 Hz_10ms_modulation_mae_db": 0.7046022266126158, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0214989764357665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08309686449533042, "3000–6000 Hz_40ms_envelope_correlation": 0.9955721681644665, "3000–6000 Hz_40ms_modulation_mae_db": 0.7659342444236817, "mr_stft_db": 5.2277159651174125, "band_mae_db": 0.09804336834138068, "rms_gain_error_db": -0.07775459153830827, "output_peak": 0.5018226835893109}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.25113076915326, "spectral_convergence": 0.2559225200999497, "spectral_convergence_status": "ok", "amplitude_floor": 1.4935395043887774e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.0824345933010315, "spectral_convergence": 0.22479448044481293, "spectral_convergence_status": "ok", "amplitude_floor": 1.0494802571211602e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.3838560056465266, "spectral_convergence": 0.22741647359793113, "spectral_convergence_status": "ok", "amplitude_floor": 9.412333460941768e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00214444778551191, "output_power": 0.0020946356123404614, "delta_db": -0.1020699320365798, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003131735025595309, "output_power": 0.003101531667830637, "delta_db": -0.04208789682140868, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006987919290953902, "output_power": 0.0006748272984371095, "delta_db": -0.15155237766111876, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004380829302633605, "output_power": 0.00040771056212133503, "delta_db": -0.31204368986281117, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014837331900785606, "output_power": 0.00014617062468588396, "delta_db": -0.06495708660656717, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003193593810232492, "output_power": 0.00034712063108382747, "delta_db": 0.3620074882626715, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1724530785418595, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06921608532613266, "pitch": null, "output_rms_dbfs": -21.69283281658924, "output_peak": 0.5042951622658332, "reference_rms_dbfs": -21.62361673126311, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.2391404560336055}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020729159449367856, "log_rms_mae_db": 0.9211519682495093, "rms_envelope_nrmse": 0.06627934917598212, "envelope_correlation": 0.9927523970318382, "rms_gain_error_db": -0.0692123315956385, "reference_mean_power": 0.0068864212769201745, "output_mean_power": 0.006777544129211211}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4588043207823824e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006447788388285564, "output_power": 0.0006334012181573903, "reference_active": true, "delta_db": -0.07731881530667215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001410929626621107, "output_power": 0.00012921795339118004, "reference_active": true, "delta_db": -0.381824946344526}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.775085636154685e-05, "output_power": 2.628038970362276e-05, "reference_active": true, "delta_db": -0.23644588573974606}], "active_band_count": 3, "active_band_mae_db": 0.2318632157969814}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001768834026970723, "log_rms_mae_db": 0.42578884272893663, "rms_envelope_nrmse": 0.03089672591679321, "envelope_correlation": 0.9983257058958135, "rms_gain_error_db": -0.0692243984193501, "reference_mean_power": 0.006918665749586704, "output_mean_power": 0.006809259884047565}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2539809640994155e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005542590263527943, "output_power": 0.0005439084765008554, "reference_active": true, "delta_db": -0.08186947423684927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.178340930501843e-05, "output_power": 6.743392097342306e-05, "reference_active": true, "delta_db": -0.2714566836866838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.482399562538733e-06, "output_power": 1.5000265597296614e-06, "reference_active": true, "delta_db": 0.051336706679094554}], "active_band_count": 3, "active_band_mae_db": 0.13488762153420922}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016054516244843413, "log_rms_mae_db": 1.461141020899775, "rms_envelope_nrmse": 0.08530882072183069, "envelope_correlation": 0.9928378511916551, "rms_gain_error_db": -0.08494771344328701, "reference_mean_power": 0.0021921303983993327, "output_mean_power": 0.0021496690966755264}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.01765222997539e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005084502896759698, "output_power": 0.0004803666801232054, "reference_active": true, "delta_db": -0.24675623473898206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.688692893438751e-05, "output_power": 8.61053947484605e-05, "reference_active": true, "delta_db": -0.5123482801635606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1104523408685175e-05, "output_power": 1.4418856516340686e-05, "reference_active": true, "delta_db": -1.65444729040349}], "active_band_count": 3, "active_band_mae_db": 0.8045172684353442}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013960343413021469, "log_rms_mae_db": 0.8586916430996869, "rms_envelope_nrmse": 0.04811934732701749, "envelope_correlation": 0.9976122487225081, "rms_gain_error_db": -0.08496797108614355, "reference_mean_power": 0.00220313483790695, "output_mean_power": 0.0021604503040832425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.111026246795902e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004662873703819768, "output_power": 0.00043727784348493134, "reference_active": true, "delta_db": -0.27896179986777314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.729339985490381e-05, "output_power": 4.1867891596611805e-05, "reference_active": true, "delta_db": -0.5291944491527316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4261322714398465e-06, "output_power": 1.144529660835736e-06, "reference_active": true, "delta_db": -0.9553275545753186}], "active_band_count": 3, "active_band_mae_db": 0.5878279345319412}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.530788197178361e-05, "log_rms_mae_db": 1.4368425570336913, "rms_envelope_nrmse": 0.1146086202917707, "envelope_correlation": 0.9903099542688057, "rms_gain_error_db": -0.30543724388446697, "reference_mean_power": 0.0004376816454019562, "output_mean_power": 0.00040795720722835305}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8441025227565861e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.86794553375767e-05, "output_power": 9.352319986044835e-05, "reference_active": true, "delta_db": -0.23307386103491262}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2863459584597701e-05, "output_power": 1.0235359634340956e-05, "reference_active": true, "delta_db": -0.9925467951287937}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6929848290709045e-06, "output_power": 2.3878202508985353e-06, "reference_active": true, "delta_db": -0.5223227572885764}], "active_band_count": 3, "active_band_mae_db": 0.5826478044840943}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.347481502172338e-05, "log_rms_mae_db": 0.9217627841210673, "rms_envelope_nrmse": 0.07351329675405596, "envelope_correlation": 0.9964054597980435, "rms_gain_error_db": -0.3054471557404083, "reference_mean_power": 0.00043987971266632326, "output_mean_power": 0.0004100050605728371}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.705293511824543e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.166386964932698e-05, "output_power": 8.688635533690639e-05, "reference_active": true, "delta_db": -0.23246607268519184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.453379972535193e-06, "output_power": 4.659355839866014e-06, "reference_active": true, "delta_db": -0.6833987963152073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5473641354942359e-07, "output_power": 1.2955459216666657e-07, "reference_active": true, "delta_db": -0.7713971513326944}], "active_band_count": 3, "active_band_mae_db": 0.5624206734443645}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012710775449688844, "log_rms_mae_db": 1.5989771126299008, "rms_envelope_nrmse": 0.16734401241551958, "envelope_correlation": 0.9842554082881957, "rms_gain_error_db": 0.34104749459137434, "reference_mean_power": 0.00030261254733777104, "output_mean_power": 0.0003273344245362468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.325105138813031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3055217502518526e-05, "output_power": 4.205605588848616e-05, "reference_active": true, "delta_db": -0.10197246356176388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2532169127733897e-05, "output_power": 1.11019873497484e-05, "reference_active": true, "delta_db": -0.5262551912936765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7715638592807256e-06, "output_power": 1.6498034728706274e-06, "reference_active": true, "delta_db": -0.3092459842775008}], "active_band_count": 3, "active_band_mae_db": 0.31249121304431376}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010399786917892939, "log_rms_mae_db": 0.9783149828382999, "rms_envelope_nrmse": 0.08262839593019974, "envelope_correlation": 0.9968283981714191, "rms_gain_error_db": 0.34104499546504713, "reference_mean_power": 0.0003041332985815958, "output_mean_power": 0.0003289792239677229}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.904406483303471e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.339234052233676e-05, "output_power": 4.221046741234534e-05, "reference_active": true, "delta_db": -0.11992914712919202}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.902501195013482e-06, "output_power": 5.373470022320722e-06, "reference_active": true, "delta_db": -0.4078125331311915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6308355063930231e-07, "output_power": 1.340479680306065e-07, "reference_active": true, "delta_db": -0.851499231952412}], "active_band_count": 3, "active_band_mae_db": 0.45974697073759846}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9367140018034713, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9c060c4de35cb128cfacdd382cbf74da47e8889ebc12b53b177ea9549d564e7f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.990276047161336, "standard_deviation": 0.0010344074559246082, "minimum": 0.9889257177663189, "maximum": 0.9922427478522995, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4176674257506348, "standard_deviation": 0.02249198929394817, "minimum": 1.386061786118226, "maximum": 1.4505681918003381, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5902508280543874, "standard_deviation": 0.1401733741572988, "minimum": 0.4226853873989576, "maximum": 0.8011106863260142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11329608166503174, "standard_deviation": 0.006101629201726784, "minimum": 0.10126551640764912, "maximum": 0.12005682878480022, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966730738185002, "standard_deviation": 0.0003848727527306843, "minimum": 0.996154180025772, "maximum": 0.9973186260301619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8905962862604784, "standard_deviation": 0.020243805908820623, "minimum": 0.8591766606782643, "maximum": 0.9217627841210673, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.43698648595855794, "standard_deviation": 0.17641907995675846, "minimum": 0.18804678249709736, "maximum": 0.7156015547665403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06880526288604036, "standard_deviation": 0.004007719866841382, "minimum": 0.06142242737574523, "maximum": 0.07351329675405596, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9870652329024472, "standard_deviation": 0.0020771651485434416, "minimum": 0.9840420571821726, "maximum": 0.9896803430081274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6177205403036845, "standard_deviation": 0.03003670056327387, "minimum": 1.5840514894986466, "maximum": 1.662574468633796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5381912121544163, "standard_deviation": 0.2408676577039429, "minimum": 0.13744709939527652, "maximum": 0.8300160177812647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14555411169212457, "standard_deviation": 0.012668744756461433, "minimum": 0.1298762289130758, "maximum": 0.16734401241551958, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9968436065678349, "standard_deviation": 0.0006481372167922606, "minimum": 0.9955721681644665, "maximum": 0.9976887894291055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9888601950491058, "standard_deviation": 0.02919536386458381, "minimum": 0.953336273905341, "maximum": 1.0341861084614161, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.547836003782322, "standard_deviation": 0.2635899519887188, "minimum": 0.13378204219550477, "maximum": 0.9113171770029956, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07321053882363873, "standard_deviation": 0.007313774574113996, "minimum": 0.06505154294002116, "maximum": 0.08309686449533042, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9934099291213749, "standard_deviation": 0.0007775180410475457, "minimum": 0.9922370399034244, "maximum": 0.9947327065942503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3811127091228825, "standard_deviation": 0.05350046503388472, "minimum": 1.3157472063684996, "maximum": 1.461141020899775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7230608715237737, "standard_deviation": 0.18773067628462473, "minimum": 0.4884588454973479, "maximum": 0.9528461608351174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08169085226076592, "standard_deviation": 0.004718253552754832, "minimum": 0.07343218858467677, "maximum": 0.08874606690295687, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978748504306993, "standard_deviation": 0.00028850933975268497, "minimum": 0.9975160629711379, "maximum": 0.9984566866422246, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8099335733597968, "standard_deviation": 0.03663902428945804, "minimum": 0.742727606619742, "maximum": 0.8586916430996869, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7573739249474412, "standard_deviation": 0.25285746438879236, "minimum": 0.44032074968125823, "maximum": 1.1173012162671634, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04511358977815537, "standard_deviation": 0.0030711958905395662, "minimum": 0.03885843451493066, "maximum": 0.04909347418983877, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11797859332464639, "standard_deviation": 0.0303208616630374, "minimum": 0.07328927171201349, "maximum": 0.1724530785418595, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.23458121039895, "standard_deviation": 0.0074012328427083, "minimum": 5.2277159651174125, "maximum": 5.249559547899622, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5017576518471656, "standard_deviation": 0.008721667192006342, "minimum": 0.492194022717072, "maximum": 0.5194161996685783, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06419507795101886, "standard_deviation": 0.01660502162048119, "minimum": -0.0794253157144057, "maximum": -0.026964783259574434, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933733001481941, "standard_deviation": 0.0006321126610533365, "minimum": 0.9923857981886849, "maximum": 0.9943705873469462, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8812663809353283, "standard_deviation": 0.026501081302760368, "minimum": 0.8432612023494925, "maximum": 0.9211519682495093, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.25565356951243223, "standard_deviation": 0.10076589132057105, "minimum": 0.10998499603478157, "maximum": 0.4252381793644215, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06338822628259967, "standard_deviation": 0.0030280955602370145, "minimum": 0.05848830386604096, "maximum": 0.06800424975851241, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982390579980907, "standard_deviation": 0.00015681168206561511, "minimum": 0.9980081408115773, "maximum": 0.9984474795116342, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4230977505508085, "standard_deviation": 0.023809165925014947, "minimum": 0.39319334546908613, "maximum": 0.4516451086071138, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2945347895113391, "standard_deviation": 0.18454577092081895, "minimum": 0.13488762153420922, "maximum": 0.6882667961115262, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03156564009212516, "standard_deviation": 0.0015690839751861345, "minimum": 0.029001900868565952, "maximum": 0.0335617464982536, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0039", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.6345841884613037, "predicted_mos": 2.6345841884613037, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "db885fa0c96ba4e235ae5172df8095c5062235910ff5eb135380b2e547c2bf57", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.38367550028488, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.913491720103044, "全帯域_10ms_rms_envelope_nrmse": 0.06502100865327842, "全帯域_10ms_envelope_correlation": 0.9930189088516167, "全帯域_10ms_modulation_mae_db": 0.20915416379026108, "全帯域_40ms_log_rms_mae_db": 0.41974552107577234, "全帯域_40ms_rms_envelope_nrmse": 0.03025264330176238, "全帯域_40ms_envelope_correlation": 0.9983831638235565, "全帯域_40ms_modulation_mae_db": 0.1773095101712261, "300–1000 Hz_10ms_log_rms_mae_db": 1.3880435639617323, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08693514889996172, "300–1000 Hz_10ms_envelope_correlation": 0.9925456276190715, "300–1000 Hz_10ms_modulation_mae_db": 0.8293349179751862, "300–1000 Hz_40ms_log_rms_mae_db": 0.716902896470577, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04806887380234651, "300–1000 Hz_40ms_envelope_correlation": 0.9975930166406934, "300–1000 Hz_40ms_modulation_mae_db": 0.6477852267812614, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3643183229431612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11655337665930662, "1000–3000 Hz_10ms_envelope_correlation": 0.9899272576639959, "1000–3000 Hz_10ms_modulation_mae_db": 0.4896864564658374, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8362982575809784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07306253241339992, "1000–3000 Hz_40ms_envelope_correlation": 0.9964546054020864, "1000–3000 Hz_40ms_modulation_mae_db": 0.4376899884131145, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5394472622181996, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16294704811407051, "3000–6000 Hz_10ms_envelope_correlation": 0.9849809768195975, "3000–6000 Hz_10ms_modulation_mae_db": 0.3167891539601263, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9094391612684203, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08163859091252429, "3000–6000 Hz_40ms_envelope_correlation": 0.99680995142486, "3000–6000 Hz_40ms_modulation_mae_db": 0.49519684237817546, "mr_stft_db": 5.183534369813756, "band_mae_db": 0.1818930075618356, "rms_gain_error_db": -0.06391485755056503, "output_peak": 0.5207051712345016}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8781514427757728, "全帯域_10ms_rms_envelope_nrmse": 0.0624054136600953, "全帯域_10ms_envelope_correlation": 0.9936446190946185, "全帯域_10ms_modulation_mae_db": 0.3038703139420695, "全帯域_40ms_log_rms_mae_db": 0.44802985344263, "全帯域_40ms_rms_envelope_nrmse": 0.03298114241921102, "全帯域_40ms_envelope_correlation": 0.9981456026477967, "全帯域_40ms_modulation_mae_db": 0.3686106658589558, "300–1000 Hz_10ms_log_rms_mae_db": 1.286592043539278, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08467128929924966, "300–1000 Hz_10ms_envelope_correlation": 0.9929562591041543, "300–1000 Hz_10ms_modulation_mae_db": 1.0482603842942781, "300–1000 Hz_40ms_log_rms_mae_db": 0.678602381196172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04402604660218973, "300–1000 Hz_40ms_envelope_correlation": 0.998042564831055, "300–1000 Hz_40ms_modulation_mae_db": 1.0528752881092724, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4036387036028575, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12029242687582428, "1000–3000 Hz_10ms_envelope_correlation": 0.9889678264876298, "1000–3000 Hz_10ms_modulation_mae_db": 0.5962384609394796, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8199207821207166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07239796480051201, "1000–3000 Hz_40ms_envelope_correlation": 0.9962452854429185, "1000–3000 Hz_40ms_modulation_mae_db": 0.5978226095918728, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5488549390509025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1529651414256904, "3000–6000 Hz_10ms_envelope_correlation": 0.9856408062094965, "3000–6000 Hz_10ms_modulation_mae_db": 0.7496972060552157, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9406722664853149, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07755304425665484, "3000–6000 Hz_40ms_envelope_correlation": 0.996236265986329, "3000–6000 Hz_40ms_modulation_mae_db": 1.007493146346507, "mr_stft_db": 5.179656315706199, "band_mae_db": 0.12169813912340621, "rms_gain_error_db": -0.08842846643263691, "output_peak": 0.49497652471946846}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8729338930547614, "全帯域_10ms_rms_envelope_nrmse": 0.06769619651120698, "全帯域_10ms_envelope_correlation": 0.9924521732153235, "全帯域_10ms_modulation_mae_db": 0.22680496844693177, "全帯域_40ms_log_rms_mae_db": 0.4241659661920051, "全帯域_40ms_rms_envelope_nrmse": 0.0344822346211451, "全帯域_40ms_envelope_correlation": 0.9978893360629892, "全帯域_40ms_modulation_mae_db": 0.27370268472768233, "300–1000 Hz_10ms_log_rms_mae_db": 1.287946941055708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09180153360853949, "300–1000 Hz_10ms_envelope_correlation": 0.9916363815338634, "300–1000 Hz_10ms_modulation_mae_db": 0.868820829997094, "300–1000 Hz_40ms_log_rms_mae_db": 0.6532858808608235, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050563658541575025, "300–1000 Hz_40ms_envelope_correlation": 0.9972915849916735, "300–1000 Hz_40ms_modulation_mae_db": 0.829148408944805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3939413970351882, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11615298072346883, "1000–3000 Hz_10ms_envelope_correlation": 0.989928568563455, "1000–3000 Hz_10ms_modulation_mae_db": 0.7183346543141834, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8253791443919356, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06784132175980682, "1000–3000 Hz_40ms_envelope_correlation": 0.9970034033560828, "1000–3000 Hz_40ms_modulation_mae_db": 0.6646749251388112, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5925089431253043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14156015753084936, "3000–6000 Hz_10ms_envelope_correlation": 0.9878232233594105, "3000–6000 Hz_10ms_modulation_mae_db": 0.06593006289653665, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9091173169704109, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.068604194219296, "3000–6000 Hz_40ms_envelope_correlation": 0.9971434552377214, "3000–6000 Hz_40ms_modulation_mae_db": 0.22989466601192496, "mr_stft_db": 5.198868500605978, "band_mae_db": 0.1572005563213291, "rms_gain_error_db": -0.07052200020356934, "output_peak": 0.4932548458599601}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9044978604962448, "全帯域_10ms_rms_envelope_nrmse": 0.06522936177992106, "全帯域_10ms_envelope_correlation": 0.9930174529948813, "全帯域_10ms_modulation_mae_db": 0.3803984464488184, "全帯域_40ms_log_rms_mae_db": 0.4150448104442176, "全帯域_40ms_rms_envelope_nrmse": 0.0320734503340676, "全帯域_40ms_envelope_correlation": 0.9982027528374305, "全帯域_40ms_modulation_mae_db": 0.2290003506416064, "300–1000 Hz_10ms_log_rms_mae_db": 1.3308018088558635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08781359311926476, "300–1000 Hz_10ms_envelope_correlation": 0.9923208705711022, "300–1000 Hz_10ms_modulation_mae_db": 0.47441188914984594, "300–1000 Hz_40ms_log_rms_mae_db": 0.735749624575353, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046227789829488435, "300–1000 Hz_40ms_envelope_correlation": 0.9976791841849085, "300–1000 Hz_40ms_modulation_mae_db": 0.5264934339695554, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3461226272499232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1195843311216209, "1000–3000 Hz_10ms_envelope_correlation": 0.9891271628776935, "1000–3000 Hz_10ms_modulation_mae_db": 0.923244542497585, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7928081954152255, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07053532648639771, "1000–3000 Hz_40ms_envelope_correlation": 0.9965019572926871, "1000–3000 Hz_40ms_modulation_mae_db": 0.8503689098318703, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5432537142395897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14288130940228563, "3000–6000 Hz_10ms_envelope_correlation": 0.9872730992936466, "3000–6000 Hz_10ms_modulation_mae_db": 0.5001278453397621, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9105568980211983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06810471558392434, "3000–6000 Hz_40ms_envelope_correlation": 0.9972054786685811, "3000–6000 Hz_40ms_modulation_mae_db": 0.8964759267849383, "mr_stft_db": 5.178645925474414, "band_mae_db": 0.12591689818844495, "rms_gain_error_db": -0.06497881937550948, "output_peak": 0.5241595817421292}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8517667426705892, "全帯域_10ms_rms_envelope_nrmse": 0.06254430163462217, "全帯域_10ms_envelope_correlation": 0.9935452980517367, "全帯域_10ms_modulation_mae_db": 0.5410254363906483, "全帯域_40ms_log_rms_mae_db": 0.4257926372500779, "全帯域_40ms_rms_envelope_nrmse": 0.03182579074788492, "全帯域_40ms_envelope_correlation": 0.9981957953106415, "全帯域_40ms_modulation_mae_db": 0.8305263095412757, "300–1000 Hz_10ms_log_rms_mae_db": 1.3312543712696887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07960159110242752, "300–1000 Hz_10ms_envelope_correlation": 0.993811143877858, "300–1000 Hz_10ms_modulation_mae_db": 0.9235213294444412, "300–1000 Hz_40ms_log_rms_mae_db": 0.7403974982391563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04729196032733252, "300–1000 Hz_40ms_envelope_correlation": 0.9976992597888211, "300–1000 Hz_40ms_modulation_mae_db": 1.3162819843498028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3564663928666136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11245843148521198, "1000–3000 Hz_10ms_envelope_correlation": 0.9902515989063193, "1000–3000 Hz_10ms_modulation_mae_db": 0.36837465479224907, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8390274448532276, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06831783835079928, "1000–3000 Hz_40ms_envelope_correlation": 0.9964840060607278, "1000–3000 Hz_40ms_modulation_mae_db": 0.3036175287975063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.596123778388173, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13862056087436356, "3000–6000 Hz_10ms_envelope_correlation": 0.9881036697449755, "3000–6000 Hz_10ms_modulation_mae_db": 0.6308301172861961, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9527221542238236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07502060802970013, "3000–6000 Hz_40ms_envelope_correlation": 0.9966577596903481, "3000–6000 Hz_40ms_modulation_mae_db": 0.37025031405403014, "mr_stft_db": 5.176986212537474, "band_mae_db": 0.12007622229248972, "rms_gain_error_db": -0.0627376073488879, "output_peak": 0.5011588246783101}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8929877798371694, "全帯域_10ms_rms_envelope_nrmse": 0.06094006819009837, "全帯域_10ms_envelope_correlation": 0.9938347516019236, "全帯域_10ms_modulation_mae_db": 0.11545412476637734, "全帯域_40ms_log_rms_mae_db": 0.44656267521394866, "全帯域_40ms_rms_envelope_nrmse": 0.02894514092866712, "全帯域_40ms_envelope_correlation": 0.9984571682811026, "全帯域_40ms_modulation_mae_db": 0.15410305871926896, "300–1000 Hz_10ms_log_rms_mae_db": 1.3322492512710258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07468659569271198, "300–1000 Hz_10ms_envelope_correlation": 0.9945308444002765, "300–1000 Hz_10ms_modulation_mae_db": 0.923967054177754, "300–1000 Hz_40ms_log_rms_mae_db": 0.6681306610001542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03880543141826625, "300–1000 Hz_40ms_envelope_correlation": 0.9984360529184074, "300–1000 Hz_40ms_modulation_mae_db": 1.0210812202799997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3409996547118248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10736096689381328, "1000–3000 Hz_10ms_envelope_correlation": 0.9911779776930808, "1000–3000 Hz_10ms_modulation_mae_db": 0.4040942676574643, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7555647108545294, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06203444542577269, "1000–3000 Hz_40ms_envelope_correlation": 0.9972171108010972, "1000–3000 Hz_40ms_modulation_mae_db": 0.19724058930425667, "3000–6000 Hz_10ms_log_rms_mae_db": 1.516357332706837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1305837800171085, "3000–6000 Hz_10ms_envelope_correlation": 0.9894118302415944, "3000–6000 Hz_10ms_modulation_mae_db": 0.45971665081798035, "3000–6000 Hz_40ms_log_rms_mae_db": 0.923882266062377, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06297913843406568, "3000–6000 Hz_40ms_envelope_correlation": 0.9976450911977042, "3000–6000 Hz_40ms_modulation_mae_db": 0.20572275610633117, "mr_stft_db": 5.177299832644773, "band_mae_db": 0.0836735468379808, "rms_gain_error_db": -0.029422290545598407, "output_peak": 0.505556831291291}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.892858218087111, "全帯域_10ms_rms_envelope_nrmse": 0.059740980702961466, "全帯域_10ms_envelope_correlation": 0.994141194746282, "全帯域_10ms_modulation_mae_db": 0.19819814311340375, "全帯域_40ms_log_rms_mae_db": 0.4151955705490508, "全帯域_40ms_rms_envelope_nrmse": 0.030900326250706867, "全帯域_40ms_envelope_correlation": 0.9983434400566954, "全帯域_40ms_modulation_mae_db": 0.21258910396202424, "300–1000 Hz_10ms_log_rms_mae_db": 1.3367541178380682, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08562230971931281, "300–1000 Hz_10ms_envelope_correlation": 0.9927040168655995, "300–1000 Hz_10ms_modulation_mae_db": 0.55172213842096, "300–1000 Hz_40ms_log_rms_mae_db": 0.7182085928249459, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046555101358525544, "300–1000 Hz_40ms_envelope_correlation": 0.997649036148024, "300–1000 Hz_40ms_modulation_mae_db": 0.7621498755284758, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3724041129806828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11599506131941345, "1000–3000 Hz_10ms_envelope_correlation": 0.9902132462851462, "1000–3000 Hz_10ms_modulation_mae_db": 0.7997044945626421, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8293070389772782, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06988490108047266, "1000–3000 Hz_40ms_envelope_correlation": 0.9970748641548309, "1000–3000 Hz_40ms_modulation_mae_db": 0.46397478646476786, "3000–6000 Hz_10ms_log_rms_mae_db": 1.570142255760194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14367293561015126, "3000–6000 Hz_10ms_envelope_correlation": 0.9878131505865323, "3000–6000 Hz_10ms_modulation_mae_db": 0.7314917170768628, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9366496984758209, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08083345758149003, "3000–6000 Hz_40ms_envelope_correlation": 0.9962723448012766, "3000–6000 Hz_40ms_modulation_mae_db": 0.4200375477454399, "mr_stft_db": 5.174709648446311, "band_mae_db": 0.12763668871363845, "rms_gain_error_db": -0.07206320450312496, "output_peak": 0.49457960850796484}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8650789870718026, "全帯域_10ms_rms_envelope_nrmse": 0.06688679806285126, "全帯域_10ms_envelope_correlation": 0.9926598013095611, "全帯域_10ms_modulation_mae_db": 0.18483771203620325, "全帯域_40ms_log_rms_mae_db": 0.4388179928823627, "全帯域_40ms_rms_envelope_nrmse": 0.034097092022835174, "全帯域_40ms_envelope_correlation": 0.9979579395100779, "全帯域_40ms_modulation_mae_db": 0.1946516795095737, "300–1000 Hz_10ms_log_rms_mae_db": 1.3560887526211436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08236174623897632, "300–1000 Hz_10ms_envelope_correlation": 0.9933009489364223, "300–1000 Hz_10ms_modulation_mae_db": 0.5562738406041075, "300–1000 Hz_40ms_log_rms_mae_db": 0.721423864587256, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04536736601852425, "300–1000 Hz_40ms_envelope_correlation": 0.9978283720057656, "300–1000 Hz_40ms_modulation_mae_db": 0.4967918192651182, "1000–3000 Hz_10ms_log_rms_mae_db": 1.319176825518856, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10042627409530627, "1000–3000 Hz_10ms_envelope_correlation": 0.9924095862536828, "1000–3000 Hz_10ms_modulation_mae_db": 0.7526258085367458, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8003540372490843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06032702280790775, "1000–3000 Hz_40ms_envelope_correlation": 0.9974709760761286, "1000–3000 Hz_40ms_modulation_mae_db": 0.4829123156100405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5817389002088509, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.159512785250272, "3000–6000 Hz_10ms_envelope_correlation": 0.9841986752428009, "3000–6000 Hz_10ms_modulation_mae_db": 0.7702588814430603, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9605211240616288, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08473555394074136, "3000–6000 Hz_40ms_envelope_correlation": 0.9954377285460927, "3000–6000 Hz_40ms_modulation_mae_db": 0.8742234846238427, "mr_stft_db": 5.170999571879986, "band_mae_db": 0.10489017929399669, "rms_gain_error_db": -0.07428141880698957, "output_peak": 0.4995991931209457}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.19976252048103, "spectral_convergence": 0.2555914628741409, "spectral_convergence_status": "ok", "amplitude_floor": 1.4935395043887774e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.027545492770321, "spectral_convergence": 0.22302617929561694, "spectral_convergence_status": "ok", "amplitude_floor": 1.0494802571211602e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.323295096189918, "spectral_convergence": 0.22654662787425867, "spectral_convergence_status": "ok", "amplitude_floor": 9.412333460941768e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00214444778551191, "output_power": 0.0020954311669062783, "delta_db": -0.1004207699571105, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003131735025595309, "output_power": 0.0031105115929002947, "delta_db": -0.02953185218846291, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006987919290953902, "output_power": 0.0006776027573561808, "delta_db": -0.13372715734323487, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004380829302633605, "output_power": 0.0004079867159857389, "delta_db": -0.3091030865956971, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014837331900785606, "output_power": 0.00014268028606218683, "delta_db": -0.16991840269313502, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003193593810232492, "output_power": 0.00034605518085374545, "delta_db": 0.34865677659337313, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1818930075618356, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06391485755056503, "pitch": null, "output_rms_dbfs": -21.687531588813673, "output_peak": 0.5207051712345016, "reference_rms_dbfs": -21.62361673126311, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.183534369813756}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020729159449367856, "log_rms_mae_db": 0.913491720103044, "rms_envelope_nrmse": 0.06502100865327842, "envelope_correlation": 0.9930189088516167, "rms_gain_error_db": -0.0639263935452737, "reference_mean_power": 0.0068864212769201745, "output_mean_power": 0.006785798318715372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4588043207823824e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006447788388285564, "output_power": 0.0006319950444656351, "reference_active": true, "delta_db": -0.08697102851992884}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001410929626621107, "output_power": 0.00012960687519062369, "reference_active": true, "delta_db": -0.3687731290638572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.775085636154685e-05, "output_power": 2.6675007865041173e-05, "reference_active": true, "delta_db": -0.17171833378699722}], "active_band_count": 3, "active_band_mae_db": 0.20915416379026108}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001768834026970723, "log_rms_mae_db": 0.41974552107577234, "rms_envelope_nrmse": 0.03025264330176238, "envelope_correlation": 0.9983831638235565, "rms_gain_error_db": -0.06395597643966155, "reference_mean_power": 0.006918665749586704, "output_mean_power": 0.00681752520265633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2539809640994155e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005542590263527943, "output_power": 0.0005429228788095731, "reference_active": true, "delta_db": -0.08974631279044519}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.178340930501843e-05, "output_power": 6.76061010646644e-05, "reference_active": true, "delta_db": -0.260381907051904}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.482399562538733e-06, "output_power": 1.5457715478346797e-06, "reference_active": true, "delta_db": 0.18180031067132907}], "active_band_count": 3, "active_band_mae_db": 0.1773095101712261}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016054516244843413, "log_rms_mae_db": 1.3880435639617323, "rms_envelope_nrmse": 0.08693514889996172, "envelope_correlation": 0.9925456276190715, "rms_gain_error_db": -0.06975959299789072, "reference_mean_power": 0.0021921303983993327, "output_mean_power": 0.002157200067441214}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.01765222997539e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005084502896759698, "output_power": 0.00047994008674737446, "reference_active": true, "delta_db": -0.25061473428762704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.688692893438751e-05, "output_power": 8.418847468476941e-05, "reference_active": true, "delta_db": -0.6101254900696894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1104523408685175e-05, "output_power": 1.450938805879474e-05, "reference_active": true, "delta_db": -1.6272645295682422}], "active_band_count": 3, "active_band_mae_db": 0.8293349179751862}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013960343413021469, "log_rms_mae_db": 0.716902896470577, "rms_envelope_nrmse": 0.04806887380234651, "envelope_correlation": 0.9975930166406934, "rms_gain_error_db": -0.0697774588961128, "reference_mean_power": 0.00220313483790695, "output_mean_power": 0.002168020238799284}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.111026246795902e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004662873703819768, "output_power": 0.00043703010989735357, "reference_active": true, "delta_db": -0.28142293115521555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.729339985490381e-05, "output_power": 4.07147404172945e-05, "reference_active": true, "delta_db": -0.6504886571660168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4261322714398465e-06, "output_power": 1.1298359732725643e-06, "reference_active": true, "delta_db": -1.0114440920225518}], "active_band_count": 3, "active_band_mae_db": 0.6477852267812614}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.530788197178361e-05, "log_rms_mae_db": 1.3643183229431612, "rms_envelope_nrmse": 0.11655337665930662, "envelope_correlation": 0.9899272576639959, "rms_gain_error_db": -0.3096285623917739, "reference_mean_power": 0.0004376816454019562, "output_mean_power": 0.0004075636830555081}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8441025227565861e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.86794553375767e-05, "output_power": 9.443401519527688e-05, "reference_active": true, "delta_db": -0.1909828809014287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2863459584597701e-05, "output_power": 1.0307748425394657e-05, "reference_active": true, "delta_db": -0.961939758904262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6929848290709045e-06, "output_power": 2.503918759081068e-06, "reference_active": true, "delta_db": -0.31613672959182154}], "active_band_count": 3, "active_band_mae_db": 0.4896864564658374}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.347481502172338e-05, "log_rms_mae_db": 0.8362982575809784, "rms_envelope_nrmse": 0.07306253241339992, "envelope_correlation": 0.9964546054020864, "rms_gain_error_db": -0.30963181750930013, "reference_mean_power": 0.00043987971266632326, "output_mean_power": 0.00040961018883523134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.705293511824543e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.166386964932698e-05, "output_power": 8.776014621556412e-05, "reference_active": true, "delta_db": -0.1890084917989082}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.453379972535193e-06, "output_power": 4.718308778397883e-06, "reference_active": true, "delta_db": -0.6287940002893916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5473641354942359e-07, "output_power": 1.3805933568950526e-07, "reference_active": true, "delta_db": -0.4952674731510437}], "active_band_count": 3, "active_band_mae_db": 0.4376899884131145}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012710775449688844, "log_rms_mae_db": 1.5394472622181996, "rms_envelope_nrmse": 0.16294704811407051, "envelope_correlation": 0.9849809768195975, "rms_gain_error_db": 0.31656971298376685, "reference_mean_power": 0.00030261254733777104, "output_mean_power": 0.0003254946859766122}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.325105138813031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3055217502518526e-05, "output_power": 4.1789954698748567e-05, "reference_active": true, "delta_db": -0.1295388714397541}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2532169127733897e-05, "output_power": 1.1259139238985794e-05, "reference_active": true, "delta_db": -0.4652105728681385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7715638592807256e-06, "output_power": 1.632281318482329e-06, "reference_active": true, "delta_db": -0.35561801757248634}], "active_band_count": 3, "active_band_mae_db": 0.3167891539601263}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010399786917892939, "log_rms_mae_db": 0.9094391612684203, "rms_envelope_nrmse": 0.08163859091252429, "envelope_correlation": 0.99680995142486, "rms_gain_error_db": 0.3165675447512212, "reference_mean_power": 0.0003041332985815958, "output_mean_power": 0.0003271302659612745}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.904406483303471e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.339234052233676e-05, "output_power": 4.1997728444505465e-05, "reference_active": true, "delta_db": -0.14187274904096153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.902501195013482e-06, "output_power": 5.427530760148845e-06, "reference_active": true, "delta_db": -0.36433790210662875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6308355063930231e-07, "output_power": 1.3015839082859646e-07, "reference_active": true, "delta_db": -0.9793798759869361}], "active_band_count": 3, "active_band_mae_db": 0.49519684237817546}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.937921621306513, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a97345efce8c62205352bdbd3992599091d9a32ba1ab508b22c80838d5408117", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9902504030913755, "standard_deviation": 0.0011108889864202664, "minimum": 0.9889678264876298, "maximum": 0.9924095862536828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3621335046136385, "standard_deviation": 0.027838982511249187, "minimum": 1.319176825518856, "maximum": 1.4036387036028575, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6315379174707734, "standard_deviation": 0.1992565141109183, "minimum": 0.36837465479224907, "maximum": 0.923244542497585, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11360298114674569, "standard_deviation": 0.006711961128564301, "minimum": 0.10042627409530627, "maximum": 0.12029242687582428, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968065260733199, "standard_deviation": 0.00044020451741806464, "minimum": 0.9962452854429185, "maximum": 0.9974709760761286, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.812332451430372, "standard_deviation": 0.028158159106607846, "minimum": 0.7555647108545294, "maximum": 0.8390274448532276, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.49978770664403005, "standard_deviation": 0.20533832173767627, "minimum": 0.19724058930425667, "maximum": 0.8503689098318703, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06805016914063361, "standard_deviation": 0.004621696443751408, "minimum": 0.06032702280790775, "maximum": 0.07306253241339992, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9869056789372568, "standard_deviation": 0.0017787883868247173, "minimum": 0.9841986752428009, "maximum": 0.9894118302415944, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5610533907122564, "standard_deviation": 0.02843557815618093, "minimum": 1.516357332706837, "maximum": 1.596123778388173, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5281052043594675, "standard_deviation": 0.24593261598821622, "minimum": 0.06593006289653665, "maximum": 0.7702588814430603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1465929647780989, "standard_deviation": 0.010970530417541714, "minimum": 0.1305837800171085, "maximum": 0.16294704811407051, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.996676009444114, "standard_deviation": 0.0006913646151350018, "minimum": 0.9954377285460927, "maximum": 0.9976450911977042, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9304451106961242, "standard_deviation": 0.020278988080575915, "minimum": 0.9091173169704109, "maximum": 0.9605211240616288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5624118355063987, "standard_deviation": 0.31767724089568966, "minimum": 0.20572275610633117, "maximum": 1.007493146346507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07493366286979959, "standard_deviation": 0.007674789652687357, "minimum": 0.06297913843406568, "maximum": 0.08473555394074136, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9929757616135435, "standard_deviation": 0.0009026908093114432, "minimum": 0.9916363815338634, "maximum": 0.9945308444002765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3312163563015635, "standard_deviation": 0.03325968426946275, "minimum": 1.286592043539278, "maximum": 1.3880435639617323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7720390480079584, "standard_deviation": 0.21335846709703302, "minimum": 0.47441188914984594, "maximum": 1.0482603842942781, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08418672596005553, "standard_deviation": 0.005286039290589341, "minimum": 0.07468659569271198, "maximum": 0.09180153360853949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977773839386685, "standard_deviation": 0.00033995318181832774, "minimum": 0.9972915849916735, "maximum": 0.9984360529184074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7040876749693048, "standard_deviation": 0.03275075865936299, "minimum": 0.6532858808608235, "maximum": 0.7403974982391563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8315759071535364, "standard_deviation": 0.28374693018838854, "minimum": 0.4967918192651182, "maximum": 1.3162819843498028, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04586327848728103, "standard_deviation": 0.003444361810644348, "minimum": 0.03880543141826625, "maximum": 0.050563658541575025, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1278731547916402, "standard_deviation": 0.030149317241220986, "minimum": 0.0836735468379808, "maximum": 0.1818930075618356, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.180087547138612, "standard_deviation": 0.008415731445349132, "minimum": 5.170999571879986, "maximum": 5.198868500605978, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5042488226443214, "standard_deviation": 0.011955061700770651, "minimum": 0.4932548458599601, "maximum": 0.5241595817421292, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0657935830958602, "standard_deviation": 0.01682876253323832, "minimum": -0.08842846643263691, "maximum": -0.029422290545598407, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9932892749832429, "standard_deviation": 0.0005927045502818753, "minimum": 0.9924521732153235, "maximum": 0.994141194746282, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8839708305120619, "standard_deviation": 0.020711684761400426, "minimum": 0.8517667426705892, "maximum": 0.913491720103044, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26996791361683914, "standard_deviation": 0.13556778489510157, "minimum": 0.11545412476637734, "maximum": 0.5410254363906483, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06380801614937938, "standard_deviation": 0.002838978431712879, "minimum": 0.059740980702961466, "maximum": 0.06769619651120698, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981968998162863, "standard_deviation": 0.00019938203236537198, "minimum": 0.9978893360629892, "maximum": 0.9984571682811026, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4291693783812581, "standard_deviation": 0.01347673329163954, "minimum": 0.4150448104442176, "maximum": 0.44802985344263, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3050616703914517, "standard_deviation": 0.22257291924646072, "minimum": 0.15410305871926896, "maximum": 0.8305263095412757, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03194472757828502, "standard_deviation": 0.0018946514694630641, "minimum": 0.02894514092866712, "maximum": 0.0344822346211451, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0039", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.7530453205108643, "predicted_mos": 2.7530453205108643, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d0a5c25cc6324d3e45ebedb8385a1f1b23a34f245ebe69665e278cb946ae0fc0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.035854700021446, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8681229729590632, "全帯域_10ms_rms_envelope_nrmse": 0.0596047785866541, "全帯域_10ms_envelope_correlation": 0.9941446958057542, "全帯域_10ms_modulation_mae_db": 0.2112203332346322, "全帯域_40ms_log_rms_mae_db": 0.44443454779350305, "全帯域_40ms_rms_envelope_nrmse": 0.02874986150523517, "全帯域_40ms_envelope_correlation": 0.998513704313808, "全帯域_40ms_modulation_mae_db": 0.16502655664820584, "300–1000 Hz_10ms_log_rms_mae_db": 1.6484796224794334, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08098327484215993, "300–1000 Hz_10ms_envelope_correlation": 0.9936103625003673, "300–1000 Hz_10ms_modulation_mae_db": 0.5220101507230758, "300–1000 Hz_40ms_log_rms_mae_db": 1.0619202219744936, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04823726258922952, "300–1000 Hz_40ms_envelope_correlation": 0.9975389285890356, "300–1000 Hz_40ms_modulation_mae_db": 0.507020506602303, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6636001943677763, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11052425331407963, "1000–3000 Hz_10ms_envelope_correlation": 0.9907137337774309, "1000–3000 Hz_10ms_modulation_mae_db": 0.7181058535102305, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0076611031739033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06325731476136022, "1000–3000 Hz_40ms_envelope_correlation": 0.9971723343016035, "1000–3000 Hz_40ms_modulation_mae_db": 0.6785599059995597, "3000–6000 Hz_10ms_log_rms_mae_db": 2.216729153056545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15785934577152863, "3000–6000 Hz_10ms_envelope_correlation": 0.9848533840147239, "3000–6000 Hz_10ms_modulation_mae_db": 1.056877440938045, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3140878482249057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08355886231624753, "3000–6000 Hz_40ms_envelope_correlation": 0.9962934537110684, "3000–6000 Hz_40ms_modulation_mae_db": 1.1633312920902092, "mr_stft_db": 5.548223137529628, "band_mae_db": 0.10265304810875296, "rms_gain_error_db": -0.025151045843514297, "output_peak": 0.5643862938570796}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8233434214058909, "全帯域_10ms_rms_envelope_nrmse": 0.05813750189465655, "全帯域_10ms_envelope_correlation": 0.9944423022888129, "全帯域_10ms_modulation_mae_db": 0.4363795208675451, "全帯域_40ms_log_rms_mae_db": 0.4036346382625402, "全帯域_40ms_rms_envelope_nrmse": 0.029154863638798167, "全帯域_40ms_envelope_correlation": 0.9985094846449808, "全帯域_40ms_modulation_mae_db": 0.45675173711320344, "300–1000 Hz_10ms_log_rms_mae_db": 1.6225603161000897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08381771205637992, "300–1000 Hz_10ms_envelope_correlation": 0.9930897676600485, "300–1000 Hz_10ms_modulation_mae_db": 1.103752293854787, "300–1000 Hz_40ms_log_rms_mae_db": 1.0115279097788394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04672487557867456, "300–1000 Hz_40ms_envelope_correlation": 0.9977087051656468, "300–1000 Hz_40ms_modulation_mae_db": 1.3776638885505796, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7804542309007285, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1185735060782021, "1000–3000 Hz_10ms_envelope_correlation": 0.9896068545531601, "1000–3000 Hz_10ms_modulation_mae_db": 1.068793676363507, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1181116800304884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06532425328938524, "1000–3000 Hz_40ms_envelope_correlation": 0.9974255835490642, "1000–3000 Hz_40ms_modulation_mae_db": 0.995824525750522, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1984925012929306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13882397191358137, "3000–6000 Hz_10ms_envelope_correlation": 0.988068193861677, "3000–6000 Hz_10ms_modulation_mae_db": 1.0482270191598493, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3537715562457062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07688329213251366, "3000–6000 Hz_40ms_envelope_correlation": 0.9962184296354963, "3000–6000 Hz_40ms_modulation_mae_db": 1.2648385396983133, "mr_stft_db": 5.554018389029437, "band_mae_db": 0.13404405751377213, "rms_gain_error_db": -0.06542173023605873, "output_peak": 0.4991312429535776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8829414298292158, "全帯域_10ms_rms_envelope_nrmse": 0.06647949037722721, "全帯域_10ms_envelope_correlation": 0.9926878050222878, "全帯域_10ms_modulation_mae_db": 0.30572463191292937, "全帯域_40ms_log_rms_mae_db": 0.4476608074387651, "全帯域_40ms_rms_envelope_nrmse": 0.037187407332519344, "全帯域_40ms_envelope_correlation": 0.9974841614319542, "全帯域_40ms_modulation_mae_db": 0.21557852336063385, "300–1000 Hz_10ms_log_rms_mae_db": 1.5939800280294907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08305103083518779, "300–1000 Hz_10ms_envelope_correlation": 0.9931715765987756, "300–1000 Hz_10ms_modulation_mae_db": 0.32389025782475295, "300–1000 Hz_40ms_log_rms_mae_db": 0.9847800359336989, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04735467409526807, "300–1000 Hz_40ms_envelope_correlation": 0.9975601776751802, "300–1000 Hz_40ms_modulation_mae_db": 0.43277696638657187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7087503701143565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12368845691463952, "1000–3000 Hz_10ms_envelope_correlation": 0.9880445343333903, "1000–3000 Hz_10ms_modulation_mae_db": 0.5108189667330764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.047000642748237, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061867833735877065, "1000–3000 Hz_40ms_envelope_correlation": 0.9970995242822464, "1000–3000 Hz_40ms_modulation_mae_db": 0.5060096090666862, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2240044375253594, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1595016748162286, "3000–6000 Hz_10ms_envelope_correlation": 0.9854056504722173, "3000–6000 Hz_10ms_modulation_mae_db": 1.4667464914875834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.371983221269449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1094122008847432, "3000–6000 Hz_40ms_envelope_correlation": 0.9940103912176556, "3000–6000 Hz_40ms_modulation_mae_db": 1.8227298703346184, "mr_stft_db": 5.552800645050552, "band_mae_db": 0.14558881510960767, "rms_gain_error_db": -0.04905630273143411, "output_peak": 0.548056867390287}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8985346541975422, "全帯域_10ms_rms_envelope_nrmse": 0.06846307986673167, "全帯域_10ms_envelope_correlation": 0.9922115434081026, "全帯域_10ms_modulation_mae_db": 0.4360893262368342, "全帯域_40ms_log_rms_mae_db": 0.45104548045184817, "全帯域_40ms_rms_envelope_nrmse": 0.032764084920861324, "全帯域_40ms_envelope_correlation": 0.9980289523822976, "全帯域_40ms_modulation_mae_db": 0.4961134728952456, "300–1000 Hz_10ms_log_rms_mae_db": 1.6788841433770816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09143610763848652, "300–1000 Hz_10ms_envelope_correlation": 0.991895371375096, "300–1000 Hz_10ms_modulation_mae_db": 0.9890370311585416, "300–1000 Hz_40ms_log_rms_mae_db": 1.0427499826723234, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04897532115725048, "300–1000 Hz_40ms_envelope_correlation": 0.9976357088344167, "300–1000 Hz_40ms_modulation_mae_db": 1.2350107741231697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7889338379290964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12142724884288662, "1000–3000 Hz_10ms_envelope_correlation": 0.9884142616160966, "1000–3000 Hz_10ms_modulation_mae_db": 0.6538417964614114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1219419628561378, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06203600040871156, "1000–3000 Hz_40ms_envelope_correlation": 0.9969584218246788, "1000–3000 Hz_40ms_modulation_mae_db": 0.5457767121715732, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2087716008186606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14943986113172508, "3000–6000 Hz_10ms_envelope_correlation": 0.9860603014744995, "3000–6000 Hz_10ms_modulation_mae_db": 0.8702179020286726, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3259556684273315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07210114528796033, "3000–6000 Hz_40ms_envelope_correlation": 0.9967438520507743, "3000–6000 Hz_40ms_modulation_mae_db": 0.9691023075669826, "mr_stft_db": 5.553323096399079, "band_mae_db": 0.09070112799114682, "rms_gain_error_db": -0.03651179590425373, "output_peak": 0.5034001894932458}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8405535993311449, "全帯域_10ms_rms_envelope_nrmse": 0.057954759065740886, "全帯域_10ms_envelope_correlation": 0.9944797020492856, "全帯域_10ms_modulation_mae_db": 0.20721846789620657, "全帯域_40ms_log_rms_mae_db": 0.4403892121737695, "全帯域_40ms_rms_envelope_nrmse": 0.02814029222482534, "全帯域_40ms_envelope_correlation": 0.9986166482411346, "全帯域_40ms_modulation_mae_db": 0.1854149320258703, "300–1000 Hz_10ms_log_rms_mae_db": 1.6585614595662956, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08520858115877443, "300–1000 Hz_10ms_envelope_correlation": 0.9928789083916123, "300–1000 Hz_10ms_modulation_mae_db": 0.8886856284620928, "300–1000 Hz_40ms_log_rms_mae_db": 1.0278444979158705, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046345714607739494, "300–1000 Hz_40ms_envelope_correlation": 0.9977898550429516, "300–1000 Hz_40ms_modulation_mae_db": 0.8509605922939479, "1000–3000 Hz_10ms_log_rms_mae_db": 1.709924116904163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11708012967484292, "1000–3000 Hz_10ms_envelope_correlation": 0.9893974069446363, "1000–3000 Hz_10ms_modulation_mae_db": 0.4109198997880177, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0493046146278178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06250878874144111, "1000–3000 Hz_40ms_envelope_correlation": 0.997132203149955, "1000–3000 Hz_40ms_modulation_mae_db": 0.40966502529619203, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2000989183267965, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14350958453882243, "3000–6000 Hz_10ms_envelope_correlation": 0.9872847708429779, "3000–6000 Hz_10ms_modulation_mae_db": 0.4938968551248128, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3017096878254963, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07702841608912657, "3000–6000 Hz_40ms_envelope_correlation": 0.9964305220309483, "3000–6000 Hz_40ms_modulation_mae_db": 0.5736192157183067, "mr_stft_db": 5.542170258419026, "band_mae_db": 0.12334055581833775, "rms_gain_error_db": -0.064062786788387, "output_peak": 0.4995453485642737}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8152688829538299, "全帯域_10ms_rms_envelope_nrmse": 0.0571231248416673, "全帯域_10ms_envelope_correlation": 0.9946166565289182, "全帯域_10ms_modulation_mae_db": 0.3565464262522034, "全帯域_40ms_log_rms_mae_db": 0.4144273900839026, "全帯域_40ms_rms_envelope_nrmse": 0.029145205660945986, "全帯域_40ms_envelope_correlation": 0.9984785573198675, "全帯域_40ms_modulation_mae_db": 0.4393487293053003, "300–1000 Hz_10ms_log_rms_mae_db": 1.667500619851542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0798532461154055, "300–1000 Hz_10ms_envelope_correlation": 0.9938384591329403, "300–1000 Hz_10ms_modulation_mae_db": 0.9738457621343866, "300–1000 Hz_40ms_log_rms_mae_db": 1.0533621861737468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042222746134007075, "300–1000 Hz_40ms_envelope_correlation": 0.9982361782929575, "300–1000 Hz_40ms_modulation_mae_db": 1.0906518846650226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7305189784120019, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11176138122361784, "1000–3000 Hz_10ms_envelope_correlation": 0.9903334884826299, "1000–3000 Hz_10ms_modulation_mae_db": 0.8113194008653258, "1000–3000 Hz_40ms_log_rms_mae_db": 1.10094836703056, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06245935521946229, "1000–3000 Hz_40ms_envelope_correlation": 0.99705223567415, "1000–3000 Hz_40ms_modulation_mae_db": 0.6451860166867845, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1905442534346613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1415781668961094, "3000–6000 Hz_10ms_envelope_correlation": 0.9886057841194174, "3000–6000 Hz_10ms_modulation_mae_db": 0.34723427159006137, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3259049078676532, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07505465732163898, "3000–6000 Hz_40ms_envelope_correlation": 0.9972070680592814, "3000–6000 Hz_40ms_modulation_mae_db": 0.5498516620350401, "mr_stft_db": 5.551034206135202, "band_mae_db": 0.12686727184468327, "rms_gain_error_db": -0.05171592176939564, "output_peak": 0.5329157844313309}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8506639998526494, "全帯域_10ms_rms_envelope_nrmse": 0.06492983158755572, "全帯域_10ms_envelope_correlation": 0.9930242468652306, "全帯域_10ms_modulation_mae_db": 0.1776296466944904, "全帯域_40ms_log_rms_mae_db": 0.46738739351106273, "全帯域_40ms_rms_envelope_nrmse": 0.03541026867701033, "全帯域_40ms_envelope_correlation": 0.9976996005393036, "全帯域_40ms_modulation_mae_db": 0.14043042856025414, "300–1000 Hz_10ms_log_rms_mae_db": 1.6872794846733814, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09364842059439502, "300–1000 Hz_10ms_envelope_correlation": 0.9913258704701858, "300–1000 Hz_10ms_modulation_mae_db": 0.7922808131939364, "300–1000 Hz_40ms_log_rms_mae_db": 1.0739617123345526, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054809300773477614, "300–1000 Hz_40ms_envelope_correlation": 0.996753896507875, "300–1000 Hz_40ms_modulation_mae_db": 0.9103477229865731, "1000–3000 Hz_10ms_log_rms_mae_db": 1.791974371750533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11384161706398889, "1000–3000 Hz_10ms_envelope_correlation": 0.9902170396253914, "1000–3000 Hz_10ms_modulation_mae_db": 0.33240713028428875, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1114024196450396, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06741840077789263, "1000–3000 Hz_40ms_envelope_correlation": 0.9968573987456933, "1000–3000 Hz_40ms_modulation_mae_db": 0.27510653659058354, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2159892769865124, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16093792602329307, "3000–6000 Hz_10ms_envelope_correlation": 0.9842587478994287, "3000–6000 Hz_10ms_modulation_mae_db": 0.38956079363600465, "3000–6000 Hz_40ms_log_rms_mae_db": 1.31671403346592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08615491241106957, "3000–6000 Hz_40ms_envelope_correlation": 0.9960682940182097, "3000–6000 Hz_40ms_modulation_mae_db": 0.559135263847009, "mr_stft_db": 5.542270811236882, "band_mae_db": 0.12804470882092753, "rms_gain_error_db": -0.03575965597104798, "output_peak": 0.5118051618485571}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8635619776302877, "全帯域_10ms_rms_envelope_nrmse": 0.065359832557677, "全帯域_10ms_envelope_correlation": 0.992919981992443, "全帯域_10ms_modulation_mae_db": 0.38025732397389705, "全帯域_40ms_log_rms_mae_db": 0.3986698118295788, "全帯域_40ms_rms_envelope_nrmse": 0.03518793292507095, "全帯域_40ms_envelope_correlation": 0.9977169941911157, "全帯域_40ms_modulation_mae_db": 0.311266632398509, "300–1000 Hz_10ms_log_rms_mae_db": 1.6637598281756638, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09527543115977777, "300–1000 Hz_10ms_envelope_correlation": 0.9910883585989636, "300–1000 Hz_10ms_modulation_mae_db": 0.768125395034219, "300–1000 Hz_40ms_log_rms_mae_db": 1.065287255978289, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051729195149078325, "300–1000 Hz_40ms_envelope_correlation": 0.9973161700478708, "300–1000 Hz_40ms_modulation_mae_db": 0.7095217639187196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7683679252258677, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11792991364397622, "1000–3000 Hz_10ms_envelope_correlation": 0.9890429552271617, "1000–3000 Hz_10ms_modulation_mae_db": 0.6739546134396494, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1199900405206535, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06681891648649974, "1000–3000 Hz_40ms_envelope_correlation": 0.9963072519894902, "1000–3000 Hz_40ms_modulation_mae_db": 0.4385900910453013, "3000–6000 Hz_10ms_log_rms_mae_db": 2.232058082722335, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14580261444056553, "3000–6000 Hz_10ms_envelope_correlation": 0.9869579786539702, "3000–6000 Hz_10ms_modulation_mae_db": 1.2117440890425328, "3000–6000 Hz_40ms_log_rms_mae_db": 1.369367853372046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08106494033319665, "3000–6000 Hz_40ms_envelope_correlation": 0.99617742487404, "3000–6000 Hz_40ms_modulation_mae_db": 1.6108222911291026, "mr_stft_db": 5.545562972645887, "band_mae_db": 0.0712447601911954, "rms_gain_error_db": -0.03348504890510028, "output_peak": 0.5893304163780431}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.620900597079012, "spectral_convergence": 0.2637843207451699, "spectral_convergence_status": "ok", "amplitude_floor": 1.4935395043887774e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.484867433641333, "spectral_convergence": 0.23260071653531816, "spectral_convergence_status": "ok", "amplitude_floor": 1.0494802571211602e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.538901381868539, "spectral_convergence": 0.23351900762504957, "spectral_convergence_status": "ok", "amplitude_floor": 9.412333460941768e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00214444778551191, "output_power": 0.0021291822576752217, "delta_db": -0.03102637632714487, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003131735025595309, "output_power": 0.0031544456464620905, "delta_db": 0.03138039057318811, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006987919290953902, "output_power": 0.0006909737537572217, "delta_db": -0.04886329066923728, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004380829302633605, "output_power": 0.00041808694716599036, "delta_db": -0.20289722464036788, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014837331900785606, "output_power": 0.00014637221075974765, "delta_db": -0.058971793325939006, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003193593810232492, "output_power": 0.00030199639349722603, "delta_db": -0.24277921311664058, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10265304810875296, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.025151045843514297, "pitch": null, "output_rms_dbfs": -21.648767777106627, "output_peak": 0.5643862938570796, "reference_rms_dbfs": -21.62361673126311, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.548223137529628}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020729159449367856, "log_rms_mae_db": 0.8681229729590632, "rms_envelope_nrmse": 0.0596047785866541, "envelope_correlation": 0.9941446958057542, "rms_gain_error_db": -0.025125467213298287, "reference_mean_power": 0.0068864212769201745, "output_mean_power": 0.006846695924888528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4588043207823824e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006447788388285564, "output_power": 0.0006416335326603811, "reference_active": true, "delta_db": -0.021237231000571632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001410929626621107, "output_power": 0.0001316844252893212, "reference_active": true, "delta_db": -0.2997094016032729}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.775085636154685e-05, "output_power": 2.5822896396189373e-05, "reference_active": true, "delta_db": -0.3127143671000521}], "active_band_count": 3, "active_band_mae_db": 0.2112203332346322}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001768834026970723, "log_rms_mae_db": 0.44443454779350305, "rms_envelope_nrmse": 0.02874986150523517, "envelope_correlation": 0.998513704313808, "rms_gain_error_db": -0.025016400760865014, "reference_mean_power": 0.006918665749586704, "output_mean_power": 0.006878927142108903}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2539809640994155e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005542590263527943, "output_power": 0.0005537087487467752, "reference_active": true, "delta_db": -0.004313889859065392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.178340930501843e-05, "output_power": 6.726536802017918e-05, "reference_active": true, "delta_db": -0.2823255837760824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.482399562538733e-06, "output_power": 1.4129319915486835e-06, "reference_active": true, "delta_db": -0.20844019630946975}], "active_band_count": 3, "active_band_mae_db": 0.16502655664820584}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016054516244843413, "log_rms_mae_db": 1.6484796224794334, "rms_envelope_nrmse": 0.08098327484215993, "envelope_correlation": 0.9936103625003673, "rms_gain_error_db": 0.00029088606383176764, "reference_mean_power": 0.0021921303983993327, "output_mean_power": 0.0021922772299997704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.01765222997539e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005084502896759698, "output_power": 0.0004981103829900585, "reference_active": true, "delta_db": -0.08922904508190588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.688692893438751e-05, "output_power": 8.396685674669849e-05, "reference_active": true, "delta_db": -0.6215729419585504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1104523408685175e-05, "output_power": 1.7332160524456953e-05, "reference_active": true, "delta_db": -0.8552284651287709}], "active_band_count": 3, "active_band_mae_db": 0.5220101507230758}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013960343413021469, "log_rms_mae_db": 1.0619202219744936, "rms_envelope_nrmse": 0.04823726258922952, "envelope_correlation": 0.9975389285890356, "rms_gain_error_db": 0.00028158708111526053, "reference_mean_power": 0.00220313483790695, "output_mean_power": 0.002203277689001416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.111026246795902e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004662873703819768, "output_power": 0.0004583012337640994, "reference_active": true, "delta_db": -0.07502626155180858}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.729339985490381e-05, "output_power": 4.2411453222185006e-05, "reference_active": true, "delta_db": -0.47317382196129015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4261322714398465e-06, "output_power": 1.1399181374629735e-06, "reference_active": true, "delta_db": -0.9728614362938105}], "active_band_count": 3, "active_band_mae_db": 0.507020506602303}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.530788197178361e-05, "log_rms_mae_db": 1.6636001943677763, "rms_envelope_nrmse": 0.11052425331407963, "envelope_correlation": 0.9907137337774309, "rms_gain_error_db": -0.21968997522359468, "reference_mean_power": 0.0004376816454019562, "output_mean_power": 0.00041609197223044936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8441025227565861e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.86794553375767e-05, "output_power": 9.26121013437508e-05, "reference_active": true, "delta_db": -0.2755900539185856}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2863459584597701e-05, "output_power": 1.0319467145830982e-05, "reference_active": true, "delta_db": -0.9570051363417178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6929848290709045e-06, "output_power": 2.1780190289827993e-06, "reference_active": true, "delta_db": -0.9217223702703881}], "active_band_count": 3, "active_band_mae_db": 0.7181058535102305}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.347481502172338e-05, "log_rms_mae_db": 1.0076611031739033, "rms_envelope_nrmse": 0.06325731476136022, "envelope_correlation": 0.9971723343016035, "rms_gain_error_db": -0.21969507698795254, "reference_mean_power": 0.00043987971266632326, "output_mean_power": 0.0004181811234263337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.705293511824543e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.166386964932698e-05, "output_power": 8.564934411512696e-05, "reference_active": true, "delta_db": -0.29474145739791324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.453379972535193e-06, "output_power": 4.4091000633952575e-06, "reference_active": true, "delta_db": -0.9231580372889202}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5473641354942359e-07, "output_power": 1.2817829910799625e-07, "reference_active": true, "delta_db": -0.8177802233118457}], "active_band_count": 3, "active_band_mae_db": 0.6785599059995597}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012710775449688844, "log_rms_mae_db": 2.216729153056545, "rms_envelope_nrmse": 0.15785934577152863, "envelope_correlation": 0.9848533840147239, "rms_gain_error_db": -0.3136933863205985, "reference_mean_power": 0.00030261254733777104, "output_mean_power": 0.0002815254055357396}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.325105138813031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3055217502518526e-05, "output_power": 4.08431360540459e-05, "reference_active": true, "delta_db": -0.22906706422033385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2532169127733897e-05, "output_power": 9.944148013700477e-06, "reference_active": true, "delta_db": -1.0045866732584476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7715638592807256e-06, "output_power": 1.134119854148842e-06, "reference_active": true, "delta_db": -1.9369785853353534}], "active_band_count": 3, "active_band_mae_db": 1.056877440938045}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010399786917892939, "log_rms_mae_db": 1.3140878482249057, "rms_envelope_nrmse": 0.08355886231624753, "envelope_correlation": 0.9962934537110684, "rms_gain_error_db": -0.3136958935494876, "reference_mean_power": 0.0003041332985815958, "output_mean_power": 0.00028294002196306805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.904406483303471e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.339234052233676e-05, "output_power": 4.064758359751849e-05, "reference_active": true, "delta_db": -0.2837834307962966}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.902501195013482e-06, "output_power": 4.707856955316774e-06, "reference_active": true, "delta_db": -0.9821282501597786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6308355063930231e-07, "output_power": 9.772415835291735e-08, "reference_active": true, "delta_db": -2.2240821953145526}], "active_band_count": 3, "active_band_mae_db": 1.1633312920902092}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9213133429929781, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "29a15464ef4de0c43c714e82c381142e63843e9d521eed432692ba6098c652bb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9894712843199871, "standard_deviation": 0.0009413852357931658, "minimum": 0.9880445343333903, "maximum": 0.9907137337774309, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7428155032005654, "standard_deviation": 0.046723566261764084, "minimum": 1.6636001943677763, "maximum": 1.791974371750533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6475201671806884, "standard_deviation": 0.2342116781617638, "minimum": 0.33240713028428875, "maximum": 1.068793676363507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11685331334452922, "standard_deviation": 0.00458046689042294, "minimum": 0.11052425331407963, "maximum": 0.12368845691463952, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970006191896101, "standard_deviation": 0.0003257721358126945, "minimum": 0.9963072519894902, "maximum": 0.9974255835490642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0845451038291047, "standard_deviation": 0.043645641723642416, "minimum": 1.0076611031739033, "maximum": 1.1219419628561378, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5618398028259004, "standard_deviation": 0.21806013827596316, "minimum": 0.27510653659058354, "maximum": 0.995824525750522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06396135792757873, "standard_deviation": 0.0022336952479853358, "minimum": 0.061867833735877065, "maximum": 0.06741840077789263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.986436851417364, "standard_deviation": 0.0015497814308811663, "minimum": 0.9842587478994287, "maximum": 0.9886057841194174, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2108360280204753, "standard_deviation": 0.013983325092074827, "minimum": 2.1905442534346613, "maximum": 2.232058082722335, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8605631078759453, "standard_deviation": 0.4113445665832027, "minimum": 0.34723427159006137, "maximum": 1.4667464914875834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14968164319148175, "standard_deviation": 0.00867481352672198, "minimum": 0.13882397191358137, "maximum": 0.16093792602329307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9961436794496843, "standard_deviation": 0.0009380101636712636, "minimum": 0.9940103912176556, "maximum": 0.9972070680592814, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3349368470873135, "standard_deviation": 0.026580625601442334, "minimum": 1.3017096878254963, "maximum": 1.371983221269449, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0641788053024477, "standard_deviation": 0.49187745336824934, "minimum": 0.5498516620350401, "maximum": 1.8227298703346184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08265730334706206, "standard_deviation": 0.011745287910989107, "minimum": 0.07210114528796033, "maximum": 0.1094122008847432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9926123343409987, "standard_deviation": 0.0010421629747435128, "minimum": 0.9910883585989636, "maximum": 0.9938384591329403, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6526256877816223, "standard_deviation": 0.03077867546104466, "minimum": 1.5939800280294907, "maximum": 1.6872794846733814, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.795203416548224, "standard_deviation": 0.25937791137325433, "minimum": 0.32389025782475295, "maximum": 1.103752293854787, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08665922555007086, "standard_deviation": 0.005947891086667808, "minimum": 0.0798532461154055, "maximum": 0.09527543115977777, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975674525194919, "standard_deviation": 0.000422095297239461, "minimum": 0.996753896507875, "maximum": 0.9982361782929575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0401792253452267, "standard_deviation": 0.030394673840811647, "minimum": 0.9847800359336989, "maximum": 1.0739617123345526, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.889244262440861, "standard_deviation": 0.335376506093527, "minimum": 0.43277696638657187, "maximum": 1.3776638885505796, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04829988626059064, "standard_deviation": 0.003755325255642169, "minimum": 0.042222746134007075, "maximum": 0.054809300773477614, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11531054317480294, "standard_deviation": 0.024887016388760085, "minimum": 0.0712447601911954, "maximum": 0.14558881510960767, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.548675439555712, "standard_deviation": 0.004869109573550009, "minimum": 5.542170258419026, "maximum": 5.554018389029437, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5310714131145493, "standard_deviation": 0.03369368820319031, "minimum": 0.4991312429535776, "maximum": 0.5893304163780431, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04514553601864897, "standard_deviation": 0.014764016601871603, "minimum": -0.06542173023605873, "maximum": -0.025151045843514297, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9935658667451044, "standard_deviation": 0.0009530507237949074, "minimum": 0.9922115434081026, "maximum": 0.9946166565289182, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.855373867269953, "standard_deviation": 0.02860717018742597, "minimum": 0.8152688829538299, "maximum": 0.8985346541975422, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3138832096335923, "standard_deviation": 0.10468591352388495, "minimum": 0.1776296466944904, "maximum": 0.4363795208675451, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0622565498472388, "standard_deviation": 0.004503759437822768, "minimum": 0.0571231248416673, "maximum": 0.06846307986673167, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981310128830577, "standard_deviation": 0.0004524036875474466, "minimum": 0.9974841614319542, "maximum": 0.9986166482411346, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4334561601931213, "standard_deviation": 0.02476340699984487, "minimum": 0.3986698118295788, "maximum": 0.46738739351106273, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.30124137653840277, "standard_deviation": 0.14466773785781206, "minimum": 0.14043042856025414, "maximum": 0.4961134728952456, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03196748961065833, "standard_deviation": 0.0036052107122412753, "minimum": 0.02814029222482534, "maximum": 0.037187407332519344, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0039", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.7030422687530518, "predicted_mos": 2.7030422687530518, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1f8e36f6a232c2915cc0168815f7f40c178ef51dda27cb0faf1d2a7eb43388cf", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.4581984002143145, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8425972328238035, "全帯域_10ms_rms_envelope_nrmse": 0.06154433036512947, "全帯域_10ms_envelope_correlation": 0.9937903300660641, "全帯域_10ms_modulation_mae_db": 0.5765353077496614, "全帯域_40ms_log_rms_mae_db": 0.39162034113884114, "全帯域_40ms_rms_envelope_nrmse": 0.02903997416246379, "全帯域_40ms_envelope_correlation": 0.9985689389047403, "全帯域_40ms_modulation_mae_db": 0.5631950092709584, "300–1000 Hz_10ms_log_rms_mae_db": 1.2391301399828625, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08250699510910252, "300–1000 Hz_10ms_envelope_correlation": 0.9934384691917758, "300–1000 Hz_10ms_modulation_mae_db": 1.0746691421545407, "300–1000 Hz_40ms_log_rms_mae_db": 0.718792247805641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05074261258861639, "300–1000 Hz_40ms_envelope_correlation": 0.9974780543360041, "300–1000 Hz_40ms_modulation_mae_db": 1.207202044997462, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2075372370173048, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10840491675952861, "1000–3000 Hz_10ms_envelope_correlation": 0.9916290451742951, "1000–3000 Hz_10ms_modulation_mae_db": 0.5296842332842725, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7905148306401124, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07108477600681855, "1000–3000 Hz_40ms_envelope_correlation": 0.9969388353380898, "1000–3000 Hz_40ms_modulation_mae_db": 0.5471500973122962, "3000–6000 Hz_10ms_log_rms_mae_db": 1.331978368851459, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1311342942804412, "3000–6000 Hz_10ms_envelope_correlation": 0.9894676840494487, "3000–6000 Hz_10ms_modulation_mae_db": 0.5098793607491664, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8811741503792752, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0703250416343379, "3000–6000 Hz_40ms_envelope_correlation": 0.9972186155465363, "3000–6000 Hz_40ms_modulation_mae_db": 0.7261811918580404, "mr_stft_db": 5.100503026438907, "band_mae_db": 0.1373028842640018, "rms_gain_error_db": -0.07821410471958251, "output_peak": 0.526293068781317}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9309362888436338, "全帯域_10ms_rms_envelope_nrmse": 0.07196319316736137, "全帯域_10ms_envelope_correlation": 0.991422615347858, "全帯域_10ms_modulation_mae_db": 0.39274570658737185, "全帯域_40ms_log_rms_mae_db": 0.45385442702549056, "全帯域_40ms_rms_envelope_nrmse": 0.038817628824763646, "全帯域_40ms_envelope_correlation": 0.9972233641975972, "全帯域_40ms_modulation_mae_db": 0.3996658822239661, "300–1000 Hz_10ms_log_rms_mae_db": 1.1631726689374833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07929386772616448, "300–1000 Hz_10ms_envelope_correlation": 0.9938355713049584, "300–1000 Hz_10ms_modulation_mae_db": 0.6189142635784695, "300–1000 Hz_40ms_log_rms_mae_db": 0.7061917093286009, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046059878544720205, "300–1000 Hz_40ms_envelope_correlation": 0.997825315344222, "300–1000 Hz_40ms_modulation_mae_db": 0.7676011642790269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.230344289467811, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1048408022203834, "1000–3000 Hz_10ms_envelope_correlation": 0.9916327055015633, "1000–3000 Hz_10ms_modulation_mae_db": 0.4962430239028441, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7852187555218699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05991879452632151, "1000–3000 Hz_40ms_envelope_correlation": 0.9974648491251492, "1000–3000 Hz_40ms_modulation_mae_db": 0.5030115473626696, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3200276274799263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1367319950694375, "3000–6000 Hz_10ms_envelope_correlation": 0.9889758236929811, "3000–6000 Hz_10ms_modulation_mae_db": 0.4419591365134739, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8585263383633731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06611576231498496, "3000–6000 Hz_40ms_envelope_correlation": 0.997614082175405, "3000–6000 Hz_40ms_modulation_mae_db": 0.6707884476869453, "mr_stft_db": 5.083188831482736, "band_mae_db": 0.13085978233210374, "rms_gain_error_db": -0.03635372819079729, "output_peak": 0.6707055323252615}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8707097859564781, "全帯域_10ms_rms_envelope_nrmse": 0.06109359862047484, "全帯域_10ms_envelope_correlation": 0.9939516799960421, "全帯域_10ms_modulation_mae_db": 0.4242660974101355, "全帯域_40ms_log_rms_mae_db": 0.432267225194474, "全帯域_40ms_rms_envelope_nrmse": 0.034335824701867225, "全帯域_40ms_envelope_correlation": 0.9980238584344265, "全帯域_40ms_modulation_mae_db": 0.5595954608359909, "300–1000 Hz_10ms_log_rms_mae_db": 1.2018960703190282, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08145110899592448, "300–1000 Hz_10ms_envelope_correlation": 0.993510795770212, "300–1000 Hz_10ms_modulation_mae_db": 0.4701319414848896, "300–1000 Hz_40ms_log_rms_mae_db": 0.683268989708436, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053400548432203314, "300–1000 Hz_40ms_envelope_correlation": 0.9970523167961889, "300–1000 Hz_40ms_modulation_mae_db": 0.5113458627480946, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2112306331817508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10068681766543577, "1000–3000 Hz_10ms_envelope_correlation": 0.9922622945661952, "1000–3000 Hz_10ms_modulation_mae_db": 0.4024683494488135, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7569925148885068, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05893054915142445, "1000–3000 Hz_40ms_envelope_correlation": 0.9974876644993886, "1000–3000 Hz_40ms_modulation_mae_db": 0.25467660014962573, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3435725007824486, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1332087402283937, "3000–6000 Hz_10ms_envelope_correlation": 0.9889061922220204, "3000–6000 Hz_10ms_modulation_mae_db": 0.22560687303707572, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8587718568759332, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07507792124312679, "3000–6000 Hz_40ms_envelope_correlation": 0.9964206423550188, "3000–6000 Hz_40ms_modulation_mae_db": 0.6502704474151951, "mr_stft_db": 5.094042955907525, "band_mae_db": 0.13793543353134616, "rms_gain_error_db": -0.09863090732617658, "output_peak": 0.4922427966684944}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9018609167288534, "全帯域_10ms_rms_envelope_nrmse": 0.06263813774329949, "全帯域_10ms_envelope_correlation": 0.9935715402783843, "全帯域_10ms_modulation_mae_db": 0.2760933549670483, "全帯域_40ms_log_rms_mae_db": 0.4222369459381648, "全帯域_40ms_rms_envelope_nrmse": 0.031682862656568694, "全帯域_40ms_envelope_correlation": 0.9981798715885387, "全帯域_40ms_modulation_mae_db": 0.14310219734312038, "300–1000 Hz_10ms_log_rms_mae_db": 1.188864397218959, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07677294314014795, "300–1000 Hz_10ms_envelope_correlation": 0.9941235334480903, "300–1000 Hz_10ms_modulation_mae_db": 0.09082312940210219, "300–1000 Hz_40ms_log_rms_mae_db": 0.702650364304715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04191440614915532, "300–1000 Hz_40ms_envelope_correlation": 0.9980806432601914, "300–1000 Hz_40ms_modulation_mae_db": 0.32639908337961987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2280380510190088, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10593634791454538, "1000–3000 Hz_10ms_envelope_correlation": 0.991471924365909, "1000–3000 Hz_10ms_modulation_mae_db": 0.35089761804445735, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7471057688369743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06367137501472607, "1000–3000 Hz_40ms_envelope_correlation": 0.9970987845670333, "1000–3000 Hz_40ms_modulation_mae_db": 0.22742528614071675, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3257038685219766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.137144644988471, "3000–6000 Hz_10ms_envelope_correlation": 0.9882334226997992, "3000–6000 Hz_10ms_modulation_mae_db": 0.6958573833260145, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8499009088033661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06816915362877678, "3000–6000 Hz_40ms_envelope_correlation": 0.997040169797724, "3000–6000 Hz_40ms_modulation_mae_db": 0.4112205198484619, "mr_stft_db": 5.099004246902344, "band_mae_db": 0.08970450320488103, "rms_gain_error_db": -0.017761663506155455, "output_peak": 0.5710509219714474}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8877803161552135, "全帯域_10ms_rms_envelope_nrmse": 0.06537478159149897, "全帯域_10ms_envelope_correlation": 0.9929472381179827, "全帯域_10ms_modulation_mae_db": 0.548765723379079, "全帯域_40ms_log_rms_mae_db": 0.4746335011650534, "全帯域_40ms_rms_envelope_nrmse": 0.034981038080240084, "全帯域_40ms_envelope_correlation": 0.9977613603331849, "全帯域_40ms_modulation_mae_db": 0.574754102282338, "300–1000 Hz_10ms_log_rms_mae_db": 1.2419346625061607, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08650724630250456, "300–1000 Hz_10ms_envelope_correlation": 0.9926626809275776, "300–1000 Hz_10ms_modulation_mae_db": 0.9520073318291469, "300–1000 Hz_40ms_log_rms_mae_db": 0.750393153736861, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04707918081372489, "300–1000 Hz_40ms_envelope_correlation": 0.9977265290506513, "300–1000 Hz_40ms_modulation_mae_db": 1.031500614883229, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2539226548450129, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1075257255597032, "1000–3000 Hz_10ms_envelope_correlation": 0.9913387930969787, "1000–3000 Hz_10ms_modulation_mae_db": 0.2627188754432864, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7953932078116226, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06852933372179161, "1000–3000 Hz_40ms_envelope_correlation": 0.996720070333713, "1000–3000 Hz_40ms_modulation_mae_db": 0.15049991053884024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3510297859150258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1770904983726971, "3000–6000 Hz_10ms_envelope_correlation": 0.9834306314262684, "3000–6000 Hz_10ms_modulation_mae_db": 0.9205582855542606, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9019468827613321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11907623243761857, "3000–6000 Hz_40ms_envelope_correlation": 0.9933943249041925, "3000–6000 Hz_40ms_modulation_mae_db": 1.2757033453351057, "mr_stft_db": 5.098415256508612, "band_mae_db": 0.15571637381540732, "rms_gain_error_db": -0.027472181959223328, "output_peak": 0.5782351644821987}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8733215030431484, "全帯域_10ms_rms_envelope_nrmse": 0.06557745526018292, "全帯域_10ms_envelope_correlation": 0.9928928179743085, "全帯域_10ms_modulation_mae_db": 0.6361694206697355, "全帯域_40ms_log_rms_mae_db": 0.42942144703540475, "全帯域_40ms_rms_envelope_nrmse": 0.031624559300028596, "全帯域_40ms_envelope_correlation": 0.9982193526818359, "全帯域_40ms_modulation_mae_db": 0.48726278900243597, "300–1000 Hz_10ms_log_rms_mae_db": 1.2920231603624175, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08388159475676028, "300–1000 Hz_10ms_envelope_correlation": 0.9931408607464908, "300–1000 Hz_10ms_modulation_mae_db": 0.8371817944637993, "300–1000 Hz_40ms_log_rms_mae_db": 0.763014204644503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044864578074505036, "300–1000 Hz_40ms_envelope_correlation": 0.9980338512053819, "300–1000 Hz_40ms_modulation_mae_db": 0.9047315051073507, "1000–3000 Hz_10ms_log_rms_mae_db": 1.282564156204852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10357076457630629, "1000–3000 Hz_10ms_envelope_correlation": 0.9917477322463567, "1000–3000 Hz_10ms_modulation_mae_db": 0.6443201141995633, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8243923069478245, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06060084190262084, "1000–3000 Hz_40ms_envelope_correlation": 0.9972613555702153, "1000–3000 Hz_40ms_modulation_mae_db": 0.5648425659708056, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3498807931502148, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14239489253591162, "3000–6000 Hz_10ms_envelope_correlation": 0.9874494105550017, "3000–6000 Hz_10ms_modulation_mae_db": 0.3686572646757937, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8250688851963822, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07076791386507293, "3000–6000 Hz_40ms_envelope_correlation": 0.9971081283684414, "3000–6000 Hz_40ms_modulation_mae_db": 0.5293342861910005, "mr_stft_db": 5.110728662790007, "band_mae_db": 0.11781772847637924, "rms_gain_error_db": -0.06215922632159047, "output_peak": 0.49542080773755504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8549063450687964, "全帯域_10ms_rms_envelope_nrmse": 0.06814845170736451, "全帯域_10ms_envelope_correlation": 0.9923074301815039, "全帯域_10ms_modulation_mae_db": 0.3229229456377721, "全帯域_40ms_log_rms_mae_db": 0.38772931296866014, "全帯域_40ms_rms_envelope_nrmse": 0.03574842110620376, "全帯域_40ms_envelope_correlation": 0.9976565579094429, "全帯域_40ms_modulation_mae_db": 0.27980751382684904, "300–1000 Hz_10ms_log_rms_mae_db": 1.1934770936275598, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08407835477971028, "300–1000 Hz_10ms_envelope_correlation": 0.9929576413341537, "300–1000 Hz_10ms_modulation_mae_db": 0.26931317207729405, "300–1000 Hz_40ms_log_rms_mae_db": 0.6627461980695518, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05159729605377117, "300–1000 Hz_40ms_envelope_correlation": 0.9970918704996529, "300–1000 Hz_40ms_modulation_mae_db": 0.09570665838853455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.190013535918568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10423287390024152, "1000–3000 Hz_10ms_envelope_correlation": 0.9919007157026056, "1000–3000 Hz_10ms_modulation_mae_db": 0.23441955376092752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7540400370789097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06372233622171865, "1000–3000 Hz_40ms_envelope_correlation": 0.9972664254646296, "1000–3000 Hz_40ms_modulation_mae_db": 0.23223212868184273, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3599760925055693, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1505270685120724, "3000–6000 Hz_10ms_envelope_correlation": 0.9861595175850661, "3000–6000 Hz_10ms_modulation_mae_db": 0.2688918811501411, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8687858885291331, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07805979689108608, "3000–6000 Hz_40ms_envelope_correlation": 0.9962256430435882, "3000–6000 Hz_40ms_modulation_mae_db": 0.5292350121049293, "mr_stft_db": 5.0974847345873995, "band_mae_db": 0.1563421733312036, "rms_gain_error_db": -0.03482988825419065, "output_peak": 0.5166457206203633}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8195687612853633, "全帯域_10ms_rms_envelope_nrmse": 0.06383668009017265, "全帯域_10ms_envelope_correlation": 0.9932547270627877, "全帯域_10ms_modulation_mae_db": 0.09373155116807026, "全帯域_40ms_log_rms_mae_db": 0.42723453183231436, "全帯域_40ms_rms_envelope_nrmse": 0.034860421942506714, "全帯域_40ms_envelope_correlation": 0.9977416980391116, "全帯域_40ms_modulation_mae_db": 0.06069524450654792, "300–1000 Hz_10ms_log_rms_mae_db": 1.213220312757061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06997594476606617, "300–1000 Hz_10ms_envelope_correlation": 0.9951588036262712, "300–1000 Hz_10ms_modulation_mae_db": 0.32432048544325626, "300–1000 Hz_40ms_log_rms_mae_db": 0.6787027499682403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03745170038156899, "300–1000 Hz_40ms_envelope_correlation": 0.9985120438084174, "300–1000 Hz_40ms_modulation_mae_db": 0.49575416409712797, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2222382133268346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10223076401203733, "1000–3000 Hz_10ms_envelope_correlation": 0.9922730261793717, "1000–3000 Hz_10ms_modulation_mae_db": 0.27982881100134877, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7723059293181568, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061706566171987616, "1000–3000 Hz_40ms_envelope_correlation": 0.9975275332263045, "1000–3000 Hz_40ms_modulation_mae_db": 0.4671980975597816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3358484513822617, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17004124782468644, "3000–6000 Hz_10ms_envelope_correlation": 0.9848917112358052, "3000–6000 Hz_10ms_modulation_mae_db": 0.6037628836328591, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8525466670496141, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11415641548163376, "3000–6000 Hz_40ms_envelope_correlation": 0.9940193543188913, "3000–6000 Hz_40ms_modulation_mae_db": 0.523948131525136, "mr_stft_db": 5.097656605936545, "band_mae_db": 0.15109080591412458, "rms_gain_error_db": 0.010469727007661161, "output_peak": 0.5082878547077687}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.055558539807138, "spectral_convergence": 0.2564186587118922, "spectral_convergence_status": "ok", "amplitude_floor": 1.4935395043887774e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.9100650751682995, "spectral_convergence": 0.2219046495033154, "spectral_convergence_status": "ok", "amplitude_floor": 1.0494802571211602e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.335885464341282, "spectral_convergence": 0.222041152955019, "spectral_convergence_status": "ok", "amplitude_floor": 9.412333460941768e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00214444778551191, "output_power": 0.002133062472980434, "delta_db": -0.023119009821102494, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003131735025595309, "output_power": 0.0030822325680225334, "delta_db": -0.06919604469165917, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006987919290953902, "output_power": 0.0006892670672614505, "delta_db": -0.05960351648960336, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004380829302633605, "output_power": 0.00040680433666819883, "delta_db": -0.3217075738909061, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014837331900785606, "output_power": 0.00014486973797994413, "delta_db": -0.10378137031267468, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003193593810232492, "output_power": 0.000301744038672502, "delta_db": -0.2464097903780651, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1373028842640018, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07821410471958251, "pitch": null, "output_rms_dbfs": -21.70183083598269, "output_peak": 0.526293068781317, "reference_rms_dbfs": -21.62361673126311, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.100503026438907}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020729159449367856, "log_rms_mae_db": 0.8425972328238035, "rms_envelope_nrmse": 0.06154433036512947, "envelope_correlation": 0.9937903300660641, "rms_gain_error_db": -0.07819333025840036, "reference_mean_power": 0.0068864212769201745, "output_mean_power": 0.006763542979421882}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4588043207823824e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006447788388285564, "output_power": 0.0006294981100355292, "reference_active": true, "delta_db": -0.10416345254491671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001410929626621107, "output_power": 0.00012398678706547836, "reference_active": true, "delta_db": -0.5612994680572391}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.775085636154685e-05, "output_power": 2.1720113445056594e-05, "reference_active": true, "delta_db": -1.0641430026468286}], "active_band_count": 3, "active_band_mae_db": 0.5765353077496614}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001768834026970723, "log_rms_mae_db": 0.39162034113884114, "rms_envelope_nrmse": 0.02903997416246379, "envelope_correlation": 0.9985689389047403, "rms_gain_error_db": -0.07817864281205004, "reference_mean_power": 0.006918665749586704, "output_mean_power": 0.0067952350766139836}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2539809640994155e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005542590263527943, "output_power": 0.0005368984637762975, "reference_active": true, "delta_db": -0.1382061320002798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.178340930501843e-05, "output_power": 6.309152336846254e-05, "reference_active": true, "delta_db": -0.5605306727752147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.482399562538733e-06, "output_power": 1.1799957846102848e-06, "reference_active": true, "delta_db": -0.9908482230373806}], "active_band_count": 3, "active_band_mae_db": 0.5631950092709584}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016054516244843413, "log_rms_mae_db": 1.2391301399828625, "rms_envelope_nrmse": 0.08250699510910252, "envelope_correlation": 0.9934384691917758, "rms_gain_error_db": -0.13960354291522412, "reference_mean_power": 0.0021921303983993327, "output_mean_power": 0.0021227850979561743}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.01765222997539e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005084502896759698, "output_power": 0.0004827497958820206, "reference_active": true, "delta_db": -0.2252640027647362}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.688692893438751e-05, "output_power": 7.41304482833349e-05, "reference_active": true, "delta_db": -1.1626856370469834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1104523408685175e-05, "output_power": 1.382833113012567e-05, "reference_active": true, "delta_db": -1.8360577866519026}], "active_band_count": 3, "active_band_mae_db": 1.0746691421545407}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013960343413021469, "log_rms_mae_db": 0.718792247805641, "rms_envelope_nrmse": 0.05074261258861639, "envelope_correlation": 0.9974780543360041, "rms_gain_error_db": -0.13960044228497256, "reference_mean_power": 0.00220313483790695, "output_mean_power": 0.002133442948961926}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.111026246795902e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004662873703819768, "output_power": 0.0004380352630197794, "reference_active": true, "delta_db": -0.27144578750313997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.729339985490381e-05, "output_power": 3.546534391707372e-05, "reference_active": true, "delta_db": -1.2499636027058423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4261322714398465e-06, "output_power": 8.793061951904469e-07, "reference_active": true, "delta_db": -2.1001967447834042}], "active_band_count": 3, "active_band_mae_db": 1.207202044997462}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.530788197178361e-05, "log_rms_mae_db": 1.2075372370173048, "rms_envelope_nrmse": 0.10840491675952861, "envelope_correlation": 0.9916290451742951, "rms_gain_error_db": -0.34508138390730325, "reference_mean_power": 0.0004376816454019562, "output_mean_power": 0.0004042501558860502}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8441025227565861e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.86794553375767e-05, "output_power": 9.029459028883603e-05, "reference_active": true, "delta_db": -0.3856501200070541}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2863459584597701e-05, "output_power": 1.101440514011937e-05, "reference_active": true, "delta_db": -0.6739673936697534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6929848290709045e-06, "output_power": 2.383912924567085e-06, "reference_active": true, "delta_db": -0.5294351861760099}], "active_band_count": 3, "active_band_mae_db": 0.5296842332842725}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.347481502172338e-05, "log_rms_mae_db": 0.7905148306401124, "rms_envelope_nrmse": 0.07108477600681855, "envelope_correlation": 0.9969388353380898, "rms_gain_error_db": -0.3450802015127577, "reference_mean_power": 0.00043987971266632326, "output_mean_power": 0.00040628043852786774}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.705293511824543e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.166386964932698e-05, "output_power": 8.397164316631436e-05, "reference_active": true, "delta_db": -0.38065535726208105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.453379972535193e-06, "output_power": 4.9119941110028105e-06, "reference_active": true, "delta_db": -0.4540792141184309}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5473641354942359e-07, "output_power": 1.2850527477689045e-07, "reference_active": true, "delta_db": -0.806715720556377}], "active_band_count": 3, "active_band_mae_db": 0.5471500973122962}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012710775449688844, "log_rms_mae_db": 1.331978368851459, "rms_envelope_nrmse": 0.1311342942804412, "envelope_correlation": 0.9894676840494487, "rms_gain_error_db": -0.24071005202505402, "reference_mean_power": 0.00030261254733777104, "output_mean_power": 0.0002862964257766519}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.325105138813031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3055217502518526e-05, "output_power": 4.196238734443805e-05, "reference_active": true, "delta_db": -0.11165599186731386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2532169127733897e-05, "output_power": 1.307863677071498e-05, "reference_active": true, "delta_db": 0.18536231141382664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7715638592807256e-06, "output_power": 1.333812656168468e-06, "reference_active": true, "delta_db": -1.2326197789663587}], "active_band_count": 3, "active_band_mae_db": 0.5098793607491664}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010399786917892939, "log_rms_mae_db": 0.8811741503792752, "rms_envelope_nrmse": 0.0703250416343379, "envelope_correlation": 0.9972186155465363, "rms_gain_error_db": -0.24070635608242422, "reference_mean_power": 0.0003041332985815958, "output_mean_power": 0.0002877354267364885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.904406483303471e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.339234052233676e-05, "output_power": 4.242338988060781e-05, "reference_active": true, "delta_db": -0.09807707726259421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.902501195013482e-06, "output_power": 6.235155003612604e-06, "reference_active": true, "delta_db": 0.23811170669013354}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6308355063930231e-07, "output_power": 1.067025206961001e-07, "reference_active": true, "delta_db": -1.8423547916213936}], "active_band_count": 3, "active_band_mae_db": 0.7261811918580404}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9394382381490466, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3e19531c172bf81f18202d663929d6f72a574de4fbf5b9230d62f0bf0160813c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9917820296041594, "standard_deviation": 0.00034336996973926805, "minimum": 0.9913387930969787, "maximum": 0.9922730261793717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2282360963726426, "standard_deviation": 0.02886022561296984, "minimum": 1.190013535918568, "maximum": 1.282564156204852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.40007257238568916, "standard_deviation": 0.14584445664832527, "minimum": 0.23441955376092752, "maximum": 0.6443201141995633, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1046786265760227, "standard_deviation": 0.0025898703284899945, "minimum": 0.10068681766543577, "maximum": 0.10840491675952861, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972206897655654, "standard_deviation": 0.00028628414552987416, "minimum": 0.996720070333713, "maximum": 0.9975275332263045, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7782454188804971, "standard_deviation": 0.025816079180324164, "minimum": 0.7471057688369743, "maximum": 0.8243923069478245, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.36837952921457234, "standard_deviation": 0.16786543429377515, "minimum": 0.15049991053884024, "maximum": 0.5648425659708056, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06352057158967617, "standard_deviation": 0.004279934550547932, "minimum": 0.05893054915142445, "maximum": 0.07108477600681855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9871892991832989, "standard_deviation": 0.0021697670973129073, "minimum": 0.9834306314262684, "maximum": 0.9894676840494487, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3397521860736101, "standard_deviation": 0.013708490340085908, "minimum": 1.3200276274799263, "maximum": 1.3599760925055693, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5043966335798481, "standard_deviation": 0.23153564598292273, "minimum": 0.22560687303707572, "maximum": 0.9205582855542606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14728417272651387, "standard_deviation": 0.01737739828996889, "minimum": 0.1311342942804412, "maximum": 0.1770904983726971, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9961301200637247, "standard_deviation": 0.001567759627774323, "minimum": 0.9933943249041925, "maximum": 0.997614082175405, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8620901972448012, "standard_deviation": 0.022776970416219202, "minimum": 0.8250688851963822, "maximum": 0.9019468827613321, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6645851727456018, "standard_deviation": 0.2667496888259302, "minimum": 0.4112205198484619, "maximum": 1.2757033453351057, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08271852968707973, "standard_deviation": 0.02129405649445303, "minimum": 0.06611576231498496, "maximum": 0.11907623243761857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9936035445436913, "standard_deviation": 0.0007833444321616741, "minimum": 0.9926626809275776, "maximum": 0.9951588036262712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2167148132139416, "standard_deviation": 0.040042905548805234, "minimum": 1.1631726689374833, "maximum": 1.2920231603624175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5796701575541874, "standard_deviation": 0.3513252990561372, "minimum": 0.09082312940210219, "maximum": 1.0746691421545407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08055850694704758, "standard_deviation": 0.005228131512993555, "minimum": 0.06997594476606617, "maximum": 0.08650724630250456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977250780375888, "standard_deviation": 0.0005022396166007283, "minimum": 0.9970523167961889, "maximum": 0.9985120438084174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7082199521958186, "standard_deviation": 0.03482522610576757, "minimum": 0.6627461980695518, "maximum": 0.763014204644503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6675301372350557, "standard_deviation": 0.37554759550186684, "minimum": 0.09570665838853455, "maximum": 1.207202044997462, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.046638775129783164, "standard_deviation": 0.005312368965018847, "minimum": 0.03745170038156899, "maximum": 0.053400548432203314, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13459621060868093, "standard_deviation": 0.022425241223294035, "minimum": 0.08970450320488103, "maximum": 0.1563421733312036, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.097628040069259, "standard_deviation": 0.0075948332323325434, "minimum": 5.083188831482736, "maximum": 5.110728662790007, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5448602334118008, "standard_deviation": 0.060152266468141896, "minimum": 0.4922427966684944, "maximum": 0.6707055323252615, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04311899665875689, "standard_deviation": 0.03494856591375706, "minimum": -0.09863090732617658, "maximum": 0.010469727007661161, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930172973781164, "standard_deviation": 0.0008367909875006413, "minimum": 0.991422615347858, "maximum": 0.9939516799960421, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8727101437381612, "standard_deviation": 0.03488615363419888, "minimum": 0.8195687612853633, "maximum": 0.9309362888436338, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40890376344610924, "standard_deviation": 0.17895462096977205, "minimum": 0.09373155116807026, "maximum": 0.6361694206697355, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06502207856818554, "standard_deviation": 0.003644999593348958, "minimum": 0.06109359862047484, "maximum": 0.07196319316736137, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979218752611098, "standard_deviation": 0.0004141577897571439, "minimum": 0.9972233641975972, "maximum": 0.9985689389047403, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4273747165373004, "standard_deviation": 0.028882133037221307, "minimum": 0.38772931296866014, "maximum": 0.4746335011650534, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3835097749115259, "standard_deviation": 0.201300144088096, "minimum": 0.06069524450654792, "maximum": 0.574754102282338, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033886341346830315, "standard_deviation": 0.003014072179037457, "minimum": 0.02903997416246379, "maximum": 0.038817628824763646, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0039", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.633453369140625, "predicted_mos": 2.633453369140625, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3c760914e0e442a689b08bd5f9469f429846e29de88dd5844ba8debaa7833fd3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.658831700216979, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9086799780485616, "全帯域_10ms_rms_envelope_nrmse": 0.0658802319462866, "全帯域_10ms_envelope_correlation": 0.9928388721010121, "全帯域_10ms_modulation_mae_db": 0.2595165306907507, "全帯域_40ms_log_rms_mae_db": 0.41752570559812796, "全帯域_40ms_rms_envelope_nrmse": 0.03083365709712479, "全帯域_40ms_envelope_correlation": 0.9983308951915476, "全帯域_40ms_modulation_mae_db": 0.14685753195024587, "300–1000 Hz_10ms_log_rms_mae_db": 1.5066156512101898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08596434145596703, "300–1000 Hz_10ms_envelope_correlation": 0.9927294615329789, "300–1000 Hz_10ms_modulation_mae_db": 0.8024719075054593, "300–1000 Hz_40ms_log_rms_mae_db": 0.9054595032626407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048369962374757214, "300–1000 Hz_40ms_envelope_correlation": 0.9975879719546703, "300–1000 Hz_40ms_modulation_mae_db": 0.5902592284415302, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4806173580162505, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11478563678078911, "1000–3000 Hz_10ms_envelope_correlation": 0.990215406952665, "1000–3000 Hz_10ms_modulation_mae_db": 0.5761934593629131, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9687092943568759, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0729918836977746, "1000–3000 Hz_40ms_envelope_correlation": 0.996393804465104, "1000–3000 Hz_40ms_modulation_mae_db": 0.5522794278493405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.657814232626735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16849651644976105, "3000–6000 Hz_10ms_envelope_correlation": 0.9842011811806215, "3000–6000 Hz_10ms_modulation_mae_db": 0.30337282379104974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0260111775266074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08438355088753993, "3000–6000 Hz_40ms_envelope_correlation": 0.9967944127542206, "3000–6000 Hz_40ms_modulation_mae_db": 0.460600426233126, "mr_stft_db": 5.274171527004119, "band_mae_db": 0.1686621126078588, "rms_gain_error_db": -0.06841966762487507, "output_peak": 0.49959941964985877}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8607025374180574, "全帯域_10ms_rms_envelope_nrmse": 0.06097511632622526, "全帯域_10ms_envelope_correlation": 0.9939089035420822, "全帯域_10ms_modulation_mae_db": 0.2698341278323828, "全帯域_40ms_log_rms_mae_db": 0.4376305323798359, "全帯域_40ms_rms_envelope_nrmse": 0.032143740970553523, "全帯域_40ms_envelope_correlation": 0.9982046876779336, "全帯域_40ms_modulation_mae_db": 0.374880559217317, "300–1000 Hz_10ms_log_rms_mae_db": 1.3678995519687875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0826060135006291, "300–1000 Hz_10ms_envelope_correlation": 0.9933193595721214, "300–1000 Hz_10ms_modulation_mae_db": 0.9271149044535251, "300–1000 Hz_40ms_log_rms_mae_db": 0.8355122610722825, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04427355532814172, "300–1000 Hz_40ms_envelope_correlation": 0.9980309181645999, "300–1000 Hz_40ms_modulation_mae_db": 0.9462774187952766, "1000–3000 Hz_10ms_log_rms_mae_db": 1.500837338733267, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11965160812128621, "1000–3000 Hz_10ms_envelope_correlation": 0.9890080920597899, "1000–3000 Hz_10ms_modulation_mae_db": 0.5360630371898885, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9460282525193636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.072661640688781, "1000–3000 Hz_40ms_envelope_correlation": 0.9960999239213542, "1000–3000 Hz_40ms_modulation_mae_db": 0.48368482067676233, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6455012511883236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15010275152169975, "3000–6000 Hz_10ms_envelope_correlation": 0.9860794428608305, "3000–6000 Hz_10ms_modulation_mae_db": 0.6990807234648235, "3000–6000 Hz_40ms_log_rms_mae_db": 1.04805612467319, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07485107871128374, "3000–6000 Hz_40ms_envelope_correlation": 0.9964373756484861, "3000–6000 Hz_40ms_modulation_mae_db": 0.9127843522634237, "mr_stft_db": 5.263994630800378, "band_mae_db": 0.10529379989785399, "rms_gain_error_db": -0.07705360200279998, "output_peak": 0.4930510247079097}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8561310976279011, "全帯域_10ms_rms_envelope_nrmse": 0.06735546878550372, "全帯域_10ms_envelope_correlation": 0.9925357638751359, "全帯域_10ms_modulation_mae_db": 0.27837559404975454, "全帯域_40ms_log_rms_mae_db": 0.39404521017819016, "全帯域_40ms_rms_envelope_nrmse": 0.03284826297066659, "全帯域_40ms_envelope_correlation": 0.9981038947981793, "全帯域_40ms_modulation_mae_db": 0.38002840238507724, "300–1000 Hz_10ms_log_rms_mae_db": 1.3544137265768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0894550613917707, "300–1000 Hz_10ms_envelope_correlation": 0.9921100477426045, "300–1000 Hz_10ms_modulation_mae_db": 0.9534847234462783, "300–1000 Hz_40ms_log_rms_mae_db": 0.7965325328706042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04934045997114771, "300–1000 Hz_40ms_envelope_correlation": 0.9974863776757112, "300–1000 Hz_40ms_modulation_mae_db": 0.972821919817649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4724196329699237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11683371033685568, "1000–3000 Hz_10ms_envelope_correlation": 0.9896908407268039, "1000–3000 Hz_10ms_modulation_mae_db": 0.6004026444669818, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9592966205461505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06902936938539957, "1000–3000 Hz_40ms_envelope_correlation": 0.9967319090447201, "1000–3000 Hz_40ms_modulation_mae_db": 0.5085427463033428, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6931345178503208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14189967972712558, "3000–6000 Hz_10ms_envelope_correlation": 0.9877174913356527, "3000–6000 Hz_10ms_modulation_mae_db": 0.14842340649781696, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0042437143058314, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06680682970715658, "3000–6000 Hz_40ms_envelope_correlation": 0.9972626372116106, "3000–6000 Hz_40ms_modulation_mae_db": 0.12510724321216557, "mr_stft_db": 5.28363612555453, "band_mae_db": 0.13660528879604952, "rms_gain_error_db": -0.07232398115561928, "output_peak": 0.4974455222928253}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8793917886812157, "全帯域_10ms_rms_envelope_nrmse": 0.06374644001532916, "全帯域_10ms_envelope_correlation": 0.9933044786021388, "全帯域_10ms_modulation_mae_db": 0.3861907821650756, "全帯域_40ms_log_rms_mae_db": 0.3904652073882784, "全帯域_40ms_rms_envelope_nrmse": 0.031148369458050204, "全帯域_40ms_envelope_correlation": 0.9982724749804004, "全帯域_40ms_modulation_mae_db": 0.2459824714434641, "300–1000 Hz_10ms_log_rms_mae_db": 1.3736811017723007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0842621967259399, "300–1000 Hz_10ms_envelope_correlation": 0.992938644309668, "300–1000 Hz_10ms_modulation_mae_db": 0.4809462875026129, "300–1000 Hz_40ms_log_rms_mae_db": 0.8539425529760888, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04542711698693901, "300–1000 Hz_40ms_envelope_correlation": 0.9977606994810233, "300–1000 Hz_40ms_modulation_mae_db": 0.44511808722907126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4681692981134715, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12068617628174541, "1000–3000 Hz_10ms_envelope_correlation": 0.9887514058172736, "1000–3000 Hz_10ms_modulation_mae_db": 0.7850428397826471, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9206618312630778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06896465730324508, "1000–3000 Hz_40ms_envelope_correlation": 0.9964828535216927, "1000–3000 Hz_40ms_modulation_mae_db": 0.7065804091545279, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6300972437987256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14276582174102756, "3000–6000 Hz_10ms_envelope_correlation": 0.9872482801062612, "3000–6000 Hz_10ms_modulation_mae_db": 0.4150212598839423, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9959827165497593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06587616425339671, "3000–6000 Hz_40ms_envelope_correlation": 0.9972730041912169, "3000–6000 Hz_40ms_modulation_mae_db": 0.732935445132466, "mr_stft_db": 5.266883365937283, "band_mae_db": 0.08961521468463919, "rms_gain_error_db": -0.05445124275126309, "output_peak": 0.5235250581356827}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8597718085390212, "全帯域_10ms_rms_envelope_nrmse": 0.06236786291989722, "全帯域_10ms_envelope_correlation": 0.9935998680180399, "全帯域_10ms_modulation_mae_db": 0.42787777806326127, "全帯域_40ms_log_rms_mae_db": 0.43957625302287767, "全帯域_40ms_rms_envelope_nrmse": 0.032507298650799384, "全帯域_40ms_envelope_correlation": 0.9981372136992432, "全帯域_40ms_modulation_mae_db": 0.6846286701227244, "300–1000 Hz_10ms_log_rms_mae_db": 1.4480727870175891, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07794630620093937, "300–1000 Hz_10ms_envelope_correlation": 0.9940919985635798, "300–1000 Hz_10ms_modulation_mae_db": 0.7568438570790205, "300–1000 Hz_40ms_log_rms_mae_db": 0.8819843062838106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046401937717180666, "300–1000 Hz_40ms_envelope_correlation": 0.9978155329583576, "300–1000 Hz_40ms_modulation_mae_db": 1.124635369381341, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4394337882255552, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11133926749291152, "1000–3000 Hz_10ms_envelope_correlation": 0.990378333212448, "1000–3000 Hz_10ms_modulation_mae_db": 0.439518075093331, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9427586885725887, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06798361959484082, "1000–3000 Hz_40ms_envelope_correlation": 0.9964260213033007, "1000–3000 Hz_40ms_modulation_mae_db": 0.24416536551720824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7039145440533756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1355398512055152, "3000–6000 Hz_10ms_envelope_correlation": 0.9886056990507911, "3000–6000 Hz_10ms_modulation_mae_db": 0.7310078575524619, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0682542654350866, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07200088479096714, "3000–6000 Hz_40ms_envelope_correlation": 0.9968726302486082, "3000–6000 Hz_40ms_modulation_mae_db": 0.5535415661517242, "mr_stft_db": 5.27191994920205, "band_mae_db": 0.10994549729894176, "rms_gain_error_db": -0.0705978133367623, "output_peak": 0.5025513766364037}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9005815813622118, "全帯域_10ms_rms_envelope_nrmse": 0.061240440678911434, "全帯域_10ms_envelope_correlation": 0.9937739867088979, "全帯域_10ms_modulation_mae_db": 0.12060506551730636, "全帯域_40ms_log_rms_mae_db": 0.4362889852108036, "全帯域_40ms_rms_envelope_nrmse": 0.028038919296608206, "全帯域_40ms_envelope_correlation": 0.9985482564312491, "全帯域_40ms_modulation_mae_db": 0.23567262306266537, "300–1000 Hz_10ms_log_rms_mae_db": 1.4616533272747678, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07428642577220392, "300–1000 Hz_10ms_envelope_correlation": 0.9946161928114798, "300–1000 Hz_10ms_modulation_mae_db": 0.8015488360103884, "300–1000 Hz_40ms_log_rms_mae_db": 0.8631094025418374, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039356323195194505, "300–1000 Hz_40ms_envelope_correlation": 0.9984210431219624, "300–1000 Hz_40ms_modulation_mae_db": 0.8643638908353625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4540768623832596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10953915279443623, "1000–3000 Hz_10ms_envelope_correlation": 0.9907138269241125, "1000–3000 Hz_10ms_modulation_mae_db": 0.4088388379782902, "1000–3000 Hz_40ms_log_rms_mae_db": 0.913906123447784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0644576682169656, "1000–3000 Hz_40ms_envelope_correlation": 0.9968440401057576, "1000–3000 Hz_40ms_modulation_mae_db": 0.1794692210186862, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6487875083366796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1297914341702071, "3000–6000 Hz_10ms_envelope_correlation": 0.9896416756332186, "3000–6000 Hz_10ms_modulation_mae_db": 0.46097608302808496, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0315897335129527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06404801658745789, "3000–6000 Hz_40ms_envelope_correlation": 0.9976953288707568, "3000–6000 Hz_40ms_modulation_mae_db": 0.24537119230961243, "mr_stft_db": 5.272275494735868, "band_mae_db": 0.06463355636302862, "rms_gain_error_db": -0.026164542779415746, "output_peak": 0.5056407347189648}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8622262416620197, "全帯域_10ms_rms_envelope_nrmse": 0.058093780743843006, "全帯域_10ms_envelope_correlation": 0.9944409001260879, "全帯域_10ms_modulation_mae_db": 0.1745394936336191, "全帯域_40ms_log_rms_mae_db": 0.3874965000467327, "全帯域_40ms_rms_envelope_nrmse": 0.029660897936675843, "全帯域_40ms_envelope_correlation": 0.9984480995044877, "全帯域_40ms_modulation_mae_db": 0.20470883957564376, "300–1000 Hz_10ms_log_rms_mae_db": 1.4226626411329522, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0822741018759236, "300–1000 Hz_10ms_envelope_correlation": 0.9932766763164791, "300–1000 Hz_10ms_modulation_mae_db": 0.5089137261440068, "300–1000 Hz_40ms_log_rms_mae_db": 0.868276463257479, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04553090391453806, "300–1000 Hz_40ms_envelope_correlation": 0.9977526912977412, "300–1000 Hz_40ms_modulation_mae_db": 0.6845379236939219, "1000–3000 Hz_10ms_log_rms_mae_db": 1.471229915511812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11480657729374295, "1000–3000 Hz_10ms_envelope_correlation": 0.990353878851139, "1000–3000 Hz_10ms_modulation_mae_db": 0.7588499352843942, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9388947196756947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06921059838905827, "1000–3000 Hz_40ms_envelope_correlation": 0.997055398360295, "1000–3000 Hz_40ms_modulation_mae_db": 0.43692135660606596, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6685012794489726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14066176644622985, "3000–6000 Hz_10ms_envelope_correlation": 0.9884334395553184, "3000–6000 Hz_10ms_modulation_mae_db": 0.8489158213708254, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0105911952861597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07749668364847877, "3000–6000 Hz_40ms_envelope_correlation": 0.9966829044605784, "3000–6000 Hz_40ms_modulation_mae_db": 0.5732978964834894, "mr_stft_db": 5.262734218139044, "band_mae_db": 0.13891814384671966, "rms_gain_error_db": -0.06074965664311449, "output_peak": 0.49729578046652273}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8337871683168008, "全帯域_10ms_rms_envelope_nrmse": 0.06455953201835431, "全帯域_10ms_envelope_correlation": 0.9931824439740333, "全帯域_10ms_modulation_mae_db": 0.23771662114435702, "全帯域_40ms_log_rms_mae_db": 0.4174817640766782, "全帯域_40ms_rms_envelope_nrmse": 0.0334051640980446, "全帯域_40ms_envelope_correlation": 0.9980672392700966, "全帯域_40ms_modulation_mae_db": 0.11412392757959389, "300–1000 Hz_10ms_log_rms_mae_db": 1.4751442148545415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08302876374763968, "300–1000 Hz_10ms_envelope_correlation": 0.993202771743379, "300–1000 Hz_10ms_modulation_mae_db": 0.5250677133364163, "300–1000 Hz_40ms_log_rms_mae_db": 0.9006462282381831, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04542891526816091, "300–1000 Hz_40ms_envelope_correlation": 0.997842482183998, "300–1000 Hz_40ms_modulation_mae_db": 0.43303503578227726, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4398955284323969, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1019568642480224, "1000–3000 Hz_10ms_envelope_correlation": 0.9920867617046552, "1000–3000 Hz_10ms_modulation_mae_db": 0.5364730595529917, "1000–3000 Hz_40ms_log_rms_mae_db": 0.948095893458445, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06126878767024654, "1000–3000 Hz_40ms_envelope_correlation": 0.9972795888538226, "1000–3000 Hz_40ms_modulation_mae_db": 0.30784234394787174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7084593944115822, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16041958935407616, "3000–6000 Hz_10ms_envelope_correlation": 0.9839807632021552, "3000–6000 Hz_10ms_modulation_mae_db": 0.6844116354847705, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0565240594438081, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08326233421339532, "3000–6000 Hz_40ms_envelope_correlation": 0.9955477903512062, "3000–6000 Hz_40ms_modulation_mae_db": 0.7554617391792884, "mr_stft_db": 5.260405804032956, "band_mae_db": 0.08922238947776882, "rms_gain_error_db": -0.08034859453403365, "output_peak": 0.5011896647320923}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.286990252156442, "spectral_convergence": 0.2544928726631055, "spectral_convergence_status": "ok", "amplitude_floor": 1.4935395043887774e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.119318285860638, "spectral_convergence": 0.22616173158343264, "spectral_convergence_status": "ok", "amplitude_floor": 1.0494802571211602e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.416206042995279, "spectral_convergence": 0.23013690664139397, "spectral_convergence_status": "ok", "amplitude_floor": 9.412333460941768e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00214444778551191, "output_power": 0.0020895298022010978, "delta_db": -0.11266906495848039, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003131735025595309, "output_power": 0.0031035224938867068, "delta_db": -0.03930112082328249, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006987919290953902, "output_power": 0.0006753562891899857, "delta_db": -0.1481493178281857, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004380829302633605, "output_power": 0.00040896637574109923, "delta_db": -0.298687285881047, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014837331900785606, "output_power": 0.00014726152126203233, "delta_db": -0.032665290862244234, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003193593810232492, "output_power": 0.0003486018900154986, "delta_db": 0.380500595293913, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1686621126078588, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06841966762487507, "pitch": null, "output_rms_dbfs": -21.692036398887986, "output_peak": 0.49959941964985877, "reference_rms_dbfs": -21.62361673126311, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.274171527004119}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020729159449367856, "log_rms_mae_db": 0.9086799780485616, "rms_envelope_nrmse": 0.0658802319462866, "envelope_correlation": 0.9928388721010121, "rms_gain_error_db": -0.06841809267948867, "reference_mean_power": 0.0068864212769201745, "output_mean_power": 0.006778783721648817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4588043207823824e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006447788388285564, "output_power": 0.0006328434541222449, "reference_active": true, "delta_db": -0.08114483509001133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001410929626621107, "output_power": 0.0001283009186736387, "reference_active": true, "delta_db": -0.41275586785868945}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.775085636154685e-05, "output_power": 2.5990312455091006e-05, "reference_active": true, "delta_db": -0.2846488891235513}], "active_band_count": 3, "active_band_mae_db": 0.2595165306907507}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001768834026970723, "log_rms_mae_db": 0.41752570559812796, "rms_envelope_nrmse": 0.03083365709712479, "envelope_correlation": 0.9983308951915476, "rms_gain_error_db": -0.06845620838492317, "reference_mean_power": 0.006918665749586704, "output_mean_power": 0.006810464428072057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2539809640994155e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005542590263527943, "output_power": 0.0005434525876679803, "reference_active": true, "delta_db": -0.08551113518286174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.178340930501843e-05, "output_power": 6.701109283208153e-05, "reference_active": true, "delta_db": -0.2987738044657691}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.482399562538733e-06, "output_power": 1.463310579639123e-06, "reference_active": true, "delta_db": -0.05628765620210675}], "active_band_count": 3, "active_band_mae_db": 0.14685753195024587}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016054516244843413, "log_rms_mae_db": 1.5066156512101898, "rms_envelope_nrmse": 0.08596434145596703, "envelope_correlation": 0.9927294615329789, "rms_gain_error_db": -0.08135691672826062, "reference_mean_power": 0.0021921303983993327, "output_mean_power": 0.0021514472027812626}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.01765222997539e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005084502896759698, "output_power": 0.0004800051293814058, "reference_active": true, "delta_db": -0.25002620784639096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.688692893438751e-05, "output_power": 8.609045968468998e-05, "reference_active": true, "delta_db": -0.5131016334681484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1104523408685175e-05, "output_power": 1.4452625892018677e-05, "reference_active": true, "delta_db": -1.6442878812018382}], "active_band_count": 3, "active_band_mae_db": 0.8024719075054593}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013960343413021469, "log_rms_mae_db": 0.9054595032626407, "rms_envelope_nrmse": 0.048369962374757214, "envelope_correlation": 0.9975879719546703, "rms_gain_error_db": -0.08137884058321375, "reference_mean_power": 0.00220313483790695, "output_mean_power": 0.002162236498338598}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.111026246795902e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004662873703819768, "output_power": 0.00043698730536543835, "reference_active": true, "delta_db": -0.2818483179198601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.729339985490381e-05, "output_power": 4.1866372864015036e-05, "reference_active": true, "delta_db": -0.5293519897224337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4261322714398465e-06, "output_power": 1.143410220064268e-06, "reference_active": true, "delta_db": -0.9595773776822969}], "active_band_count": 3, "active_band_mae_db": 0.5902592284415302}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.530788197178361e-05, "log_rms_mae_db": 1.4806173580162505, "rms_envelope_nrmse": 0.11478563678078911, "envelope_correlation": 0.990215406952665, "rms_gain_error_db": -0.2891037227398662, "reference_mean_power": 0.0004376816454019562, "output_mean_power": 0.00040949439545609283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8441025227565861e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.86794553375767e-05, "output_power": 9.366622712020889e-05, "reference_active": true, "delta_db": -0.2264371654943849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2863459584597701e-05, "output_power": 1.022768942631722e-05, "reference_active": true, "delta_db": -0.9958025456925448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6929848290709045e-06, "output_power": 2.396623647179345e-06, "reference_active": true, "delta_db": -0.5063406669018097}], "active_band_count": 3, "active_band_mae_db": 0.5761934593629131}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.347481502172338e-05, "log_rms_mae_db": 0.9687092943568759, "rms_envelope_nrmse": 0.0729918836977746, "envelope_correlation": 0.996393804465104, "rms_gain_error_db": -0.2891156256668098, "reference_mean_power": 0.00043987971266632326, "output_mean_power": 0.0004115497764598595}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.705293511824543e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.166386964932698e-05, "output_power": 8.700783210681425e-05, "reference_active": true, "delta_db": -0.2263983943794776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.453379972535193e-06, "output_power": 4.665319528837903e-06, "reference_active": true, "delta_db": -0.6778436483599094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5473641354942359e-07, "output_power": 1.3011665875203106e-07, "reference_active": true, "delta_db": -0.7525962408086346}], "active_band_count": 3, "active_band_mae_db": 0.5522794278493405}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012710775449688844, "log_rms_mae_db": 1.657814232626735, "rms_envelope_nrmse": 0.16849651644976105, "envelope_correlation": 0.9842011811806215, "rms_gain_error_db": 0.3660639171348321, "reference_mean_power": 0.00030261254733777104, "output_mean_power": 0.00032922539173227514}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.325105138813031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3055217502518526e-05, "output_power": 4.21919385446359e-05, "reference_active": true, "delta_db": -0.08796307631835822}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2532169127733897e-05, "output_power": 1.1079328221180192e-05, "reference_active": true, "delta_db": -0.5351281888339103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7715638592807256e-06, "output_power": 1.6582655988294125e-06, "reference_active": true, "delta_db": -0.2870272062208806}], "active_band_count": 3, "active_band_mae_db": 0.30337282379104974}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010399786917892939, "log_rms_mae_db": 1.0260111775266074, "rms_envelope_nrmse": 0.08438355088753993, "envelope_correlation": 0.9967944127542206, "rms_gain_error_db": 0.3660603169873462, "reference_mean_power": 0.0003041332985815958, "output_mean_power": 0.0003308796090656668}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.904406483303471e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.339234052233676e-05, "output_power": 4.231534502618965e-05, "reference_active": true, "delta_db": -0.1091518967182855}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.902501195013482e-06, "output_power": 5.3673123589140176e-06, "reference_active": true, "delta_db": -0.41279213226468114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6308355063930231e-07, "output_power": 1.3379024016341998e-07, "reference_active": true, "delta_db": -0.8598572497164114}], "active_band_count": 3, "active_band_mae_db": 0.460600426233126}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9352604860671652, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "171ab6acb67101fe500400c46283e5bcbffaba63451bd03f043309de8e3682bb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9901498182811108, "standard_deviation": 0.0010445623703006807, "minimum": 0.9887514058172736, "maximum": 0.9920867617046552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.465834965298242, "standard_deviation": 0.020804282685130904, "minimum": 1.4394337882255552, "maximum": 1.500837338733267, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5801727360889297, "standard_deviation": 0.13486329072908307, "minimum": 0.4088388379782902, "maximum": 0.7850428397826471, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1136998741687237, "standard_deviation": 0.006062042370025685, "minimum": 0.1019568642480224, "maximum": 0.12068617628174541, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966641924470059, "standard_deviation": 0.0003872620839035957, "minimum": 0.9960999239213542, "maximum": 0.9972795888538226, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9422939279799976, "standard_deviation": 0.018205402622518563, "minimum": 0.913906123447784, "maximum": 0.9687092943568759, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4274357113842257, "standard_deviation": 0.17432235339906957, "minimum": 0.1794692210186862, "maximum": 0.7065804091545279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06832102811828894, "standard_deviation": 0.003912226988237111, "minimum": 0.06126878767024654, "maximum": 0.0729918836977746, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9869884966156062, "standard_deviation": 0.0020696255640756597, "minimum": 0.9839807632021552, "maximum": 0.9896416756332186, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6695262464643394, "standard_deviation": 0.029169110941562326, "minimum": 1.6300972437987256, "maximum": 1.7084593944115822, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5364012013842219, "standard_deviation": 0.24176726902529114, "minimum": 0.14842340649781696, "maximum": 0.8489158213708254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1462096763269553, "standard_deviation": 0.012864126521642188, "minimum": 0.1297914341702071, "maximum": 0.16849651644976105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9968207604670855, "standard_deviation": 0.0006496675112109554, "minimum": 0.9955477903512062, "maximum": 0.9976953288707568, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0301566233416743, "standard_deviation": 0.025935203401888617, "minimum": 0.9959827165497593, "maximum": 1.0682542654350866, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.544887482620662, "standard_deviation": 0.2642951463352626, "minimum": 0.12510724321216557, "maximum": 0.9127843522634237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0735906928499595, "standard_deviation": 0.007801068410733826, "minimum": 0.06404801658745789, "maximum": 0.08438355088753993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932856440740363, "standard_deviation": 0.0007783156790806525, "minimum": 0.9921100477426045, "maximum": 0.9946161928114798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.426267875225991, "standard_deviation": 0.05595979726441483, "minimum": 1.3544137265768, "maximum": 1.5066156512101898, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7195489944347134, "standard_deviation": 0.18970731566630303, "minimum": 0.4809462875026129, "maximum": 0.9534847234462783, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08247790133387667, "standard_deviation": 0.004660115877101635, "minimum": 0.07428642577220392, "maximum": 0.0894550613917707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978372146047579, "standard_deviation": 0.0002871241271578556, "minimum": 0.9974863776757112, "maximum": 0.9984210431219624, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8631829063128658, "standard_deviation": 0.03558405446096231, "minimum": 0.7965325328706042, "maximum": 0.9054595032626407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7576311092470538, "standard_deviation": 0.257560392025064, "minimum": 0.43303503578227726, "maximum": 1.124635369381341, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04551614684450747, "standard_deviation": 0.0030013681152611642, "minimum": 0.039356323195194505, "maximum": 0.04934045997114771, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11286200037160754, "standard_deviation": 0.03348204122391684, "minimum": 0.06463355636302862, "maximum": 0.1686621126078588, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.269502639425778, "standard_deviation": 0.007567305125149397, "minimum": 5.260405804032956, "maximum": 5.28363612555453, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5025373226675325, "standard_deviation": 0.009288655943737097, "minimum": 0.4930510247079097, "maximum": 0.5235250581356827, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06376363760348544, "standard_deviation": 0.017332235516803855, "minimum": -0.08034859453403365, "maximum": -0.026164542779415746, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9934481521184285, "standard_deviation": 0.0006120158400207805, "minimum": 0.9925357638751359, "maximum": 0.9944409001260879, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8701590252069736, "standard_deviation": 0.02471142842664585, "minimum": 0.8337871683168008, "maximum": 0.9086799780485616, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2693319991370634, "standard_deviation": 0.10076562151722804, "minimum": 0.12060506551730636, "maximum": 0.42787777806326127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06302735917929383, "standard_deviation": 0.002974077446282572, "minimum": 0.058093780743843006, "maximum": 0.06735546878550372, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982640951941422, "standard_deviation": 0.0001703702873991756, "minimum": 0.9980672392700966, "maximum": 0.9985482564312491, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41506376973769055, "standard_deviation": 0.021971935617529574, "minimum": 0.3874965000467327, "maximum": 0.43957625302287767, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2983603781670915, "standard_deviation": 0.18286145818552768, "minimum": 0.11412392757959389, "maximum": 0.6846286701227244, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03132328880981539, "standard_deviation": 0.0017950620549037472, "minimum": 0.028038919296608206, "maximum": 0.0334051640980446, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0039", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.8445310592651367, "predicted_mos": 2.8445310592651367, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5fcf65970d9fdadbd07b93c88583c2bbccc2d24bdb0318b8f576b263b7598165", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.957567499950528, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8623075889145267, "全帯域_10ms_rms_envelope_nrmse": 0.06036168933569856, "全帯域_10ms_envelope_correlation": 0.9939687203989779, "全帯域_10ms_modulation_mae_db": 0.2185923721791346, "全帯域_40ms_log_rms_mae_db": 0.3816064167851395, "全帯域_40ms_rms_envelope_nrmse": 0.027061582521441304, "全帯域_40ms_envelope_correlation": 0.9986829619401756, "全帯域_40ms_modulation_mae_db": 0.12582477264747519, "300–1000 Hz_10ms_log_rms_mae_db": 1.3555051050872982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.081087029938105, "300–1000 Hz_10ms_envelope_correlation": 0.9935567338306767, "300–1000 Hz_10ms_modulation_mae_db": 0.4176798591507647, "300–1000 Hz_40ms_log_rms_mae_db": 0.8370902162870354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046531365486521654, "300–1000 Hz_40ms_envelope_correlation": 0.9977697820616765, "300–1000 Hz_40ms_modulation_mae_db": 0.709969583362369, "1000–3000 Hz_10ms_log_rms_mae_db": 1.366853693468481, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10269383614756794, "1000–3000 Hz_10ms_envelope_correlation": 0.9922991575098438, "1000–3000 Hz_10ms_modulation_mae_db": 0.715188773595349, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8422627730890054, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05883640399807308, "1000–3000 Hz_40ms_envelope_correlation": 0.9979468766918235, "1000–3000 Hz_40ms_modulation_mae_db": 0.6850948715638999, "3000–6000 Hz_10ms_log_rms_mae_db": 1.617167676127501, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11779454062851612, "3000–6000 Hz_10ms_envelope_correlation": 0.9913308884103582, "3000–6000 Hz_10ms_modulation_mae_db": 0.8657706561140678, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9803209801142976, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05758162830359096, "3000–6000 Hz_40ms_envelope_correlation": 0.9978829787293678, "3000–6000 Hz_40ms_modulation_mae_db": 1.2167106909817285, "mr_stft_db": 5.2249496970978395, "band_mae_db": 0.11696549492285369, "rms_gain_error_db": -0.04700189004883553, "output_peak": 0.4846016273288691}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8419484987412663, "全帯域_10ms_rms_envelope_nrmse": 0.06080572249366869, "全帯域_10ms_envelope_correlation": 0.9938729492102878, "全帯域_10ms_modulation_mae_db": 0.021213011894400075, "全帯域_40ms_log_rms_mae_db": 0.4458701145991335, "全帯域_40ms_rms_envelope_nrmse": 0.03041927975029011, "全帯域_40ms_envelope_correlation": 0.9983098501548391, "全帯域_40ms_modulation_mae_db": 0.21466877762726091, "300–1000 Hz_10ms_log_rms_mae_db": 1.3553174131591925, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09140614337628801, "300–1000 Hz_10ms_envelope_correlation": 0.9916701430079312, "300–1000 Hz_10ms_modulation_mae_db": 0.2738331141858866, "300–1000 Hz_40ms_log_rms_mae_db": 0.7842362887982596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049951959399348114, "300–1000 Hz_40ms_envelope_correlation": 0.9972836607255494, "300–1000 Hz_40ms_modulation_mae_db": 0.2792124245119998, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4125833099733327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11232680719873916, "1000–3000 Hz_10ms_envelope_correlation": 0.9909860175556896, "1000–3000 Hz_10ms_modulation_mae_db": 0.6422792809533068, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8845825187501156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0764170965400043, "1000–3000 Hz_40ms_envelope_correlation": 0.996326572830568, "1000–3000 Hz_40ms_modulation_mae_db": 0.403861846545926, "3000–6000 Hz_10ms_log_rms_mae_db": 1.598500468691302, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13837215451736093, "3000–6000 Hz_10ms_envelope_correlation": 0.9881868826425987, "3000–6000 Hz_10ms_modulation_mae_db": 0.9488593804033124, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9655379659718744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06215386294030527, "3000–6000 Hz_40ms_envelope_correlation": 0.9978814437264932, "3000–6000 Hz_40ms_modulation_mae_db": 1.2952170047809017, "mr_stft_db": 5.218440714710121, "band_mae_db": 0.12070211894448442, "rms_gain_error_db": -0.041249358666019365, "output_peak": 0.5288124711327753}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9004400768239725, "全帯域_10ms_rms_envelope_nrmse": 0.06546057996089867, "全帯域_10ms_envelope_correlation": 0.9928986154644593, "全帯域_10ms_modulation_mae_db": 0.36456528765023455, "全帯域_40ms_log_rms_mae_db": 0.48452749511649174, "全帯域_40ms_rms_envelope_nrmse": 0.034255645905342504, "全帯域_40ms_envelope_correlation": 0.9978419455576641, "全帯域_40ms_modulation_mae_db": 0.29366737619128247, "300–1000 Hz_10ms_log_rms_mae_db": 1.4376543718315664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09406440056955073, "300–1000 Hz_10ms_envelope_correlation": 0.9912152108370696, "300–1000 Hz_10ms_modulation_mae_db": 1.0295448598037755, "300–1000 Hz_40ms_log_rms_mae_db": 0.8566387161538823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05303329864367301, "300–1000 Hz_40ms_envelope_correlation": 0.9969417435485327, "300–1000 Hz_40ms_modulation_mae_db": 1.083595379785809, "1000–3000 Hz_10ms_log_rms_mae_db": 1.409438142731429, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11954944966636492, "1000–3000 Hz_10ms_envelope_correlation": 0.9890513012353483, "1000–3000 Hz_10ms_modulation_mae_db": 0.5189499476560857, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9027079399614653, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06969621446893755, "1000–3000 Hz_40ms_envelope_correlation": 0.9965025812569109, "1000–3000 Hz_40ms_modulation_mae_db": 0.47811916948619543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5869508884742851, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1457728096278351, "3000–6000 Hz_10ms_envelope_correlation": 0.9867385229832658, "3000–6000 Hz_10ms_modulation_mae_db": 0.3828163901833968, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9691211747326395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06219988592433326, "3000–6000 Hz_40ms_envelope_correlation": 0.9975095459242536, "3000–6000 Hz_40ms_modulation_mae_db": 0.5965140297905257, "mr_stft_db": 5.215758956657743, "band_mae_db": 0.11154090298466479, "rms_gain_error_db": -0.038045803160857615, "output_peak": 0.5730908629535096}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8229786366150531, "全帯域_10ms_rms_envelope_nrmse": 0.05751460878549288, "全帯域_10ms_envelope_correlation": 0.9946664009513472, "全帯域_10ms_modulation_mae_db": 0.4762155972739373, "全帯域_40ms_log_rms_mae_db": 0.3681942389747876, "全帯域_40ms_rms_envelope_nrmse": 0.030091252563147534, "全帯域_40ms_envelope_correlation": 0.9985366980737693, "全帯域_40ms_modulation_mae_db": 0.41178493189538856, "300–1000 Hz_10ms_log_rms_mae_db": 1.326598379151712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0815824272192747, "300–1000 Hz_10ms_envelope_correlation": 0.9937547539389061, "300–1000 Hz_10ms_modulation_mae_db": 0.8556952326288587, "300–1000 Hz_40ms_log_rms_mae_db": 0.778125340626864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04722974639650341, "300–1000 Hz_40ms_envelope_correlation": 0.9980638884334209, "300–1000 Hz_40ms_modulation_mae_db": 0.9009904109524568, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4081733317526108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11157971906374033, "1000–3000 Hz_10ms_envelope_correlation": 0.9905914197670644, "1000–3000 Hz_10ms_modulation_mae_db": 0.458455072417578, "1000–3000 Hz_40ms_log_rms_mae_db": 0.852518196295827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06158999500203747, "1000–3000 Hz_40ms_envelope_correlation": 0.997478152858651, "1000–3000 Hz_40ms_modulation_mae_db": 0.22615557487412663, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6439631285469332, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11571396057827124, "3000–6000 Hz_10ms_envelope_correlation": 0.9918402321633503, "3000–6000 Hz_10ms_modulation_mae_db": 0.23369580704695422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9870692655895537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.059114048204047476, "3000–6000 Hz_40ms_envelope_correlation": 0.9978661132546005, "3000–6000 Hz_40ms_modulation_mae_db": 0.33682813113495996, "mr_stft_db": 5.21714313393105, "band_mae_db": 0.12424766438253954, "rms_gain_error_db": -0.09680266083571516, "output_peak": 0.5006539304858186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8832263160844298, "全帯域_10ms_rms_envelope_nrmse": 0.06284881334036081, "全帯域_10ms_envelope_correlation": 0.9934785819151327, "全帯域_10ms_modulation_mae_db": 0.4038365622165148, "全帯域_40ms_log_rms_mae_db": 0.41422376142826156, "全帯域_40ms_rms_envelope_nrmse": 0.03204081293561974, "全帯域_40ms_envelope_correlation": 0.9981365922393755, "全帯域_40ms_modulation_mae_db": 0.13786755915045137, "300–1000 Hz_10ms_log_rms_mae_db": 1.3836901211834567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0863682617826808, "300–1000 Hz_10ms_envelope_correlation": 0.9926012157365193, "300–1000 Hz_10ms_modulation_mae_db": 0.6440001994887822, "300–1000 Hz_40ms_log_rms_mae_db": 0.8138954473270106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04994593730462947, "300–1000 Hz_40ms_envelope_correlation": 0.997291914379221, "300–1000 Hz_40ms_modulation_mae_db": 0.8179374344630125, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4080127144052146, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11564981697187746, "1000–3000 Hz_10ms_envelope_correlation": 0.9902396758010819, "1000–3000 Hz_10ms_modulation_mae_db": 0.5257759154198433, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9053496101672267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07418262777756257, "1000–3000 Hz_40ms_envelope_correlation": 0.9964784532657257, "1000–3000 Hz_40ms_modulation_mae_db": 0.5626010528006711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6051082384058697, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11943111022182827, "3000–6000 Hz_10ms_envelope_correlation": 0.9910870198654291, "3000–6000 Hz_10ms_modulation_mae_db": 0.40962584120472595, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9740066746907065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.057942525549317024, "3000–6000 Hz_40ms_envelope_correlation": 0.9978637221045608, "3000–6000 Hz_40ms_modulation_mae_db": 0.5026057793020443, "mr_stft_db": 5.22446437209975, "band_mae_db": 0.07834954531421884, "rms_gain_error_db": -0.03539605309438311, "output_peak": 0.5006103327285606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.846530967321191, "全帯域_10ms_rms_envelope_nrmse": 0.06535116790858135, "全帯域_10ms_envelope_correlation": 0.9929988330465201, "全帯域_10ms_modulation_mae_db": 0.30106587187072287, "全帯域_40ms_log_rms_mae_db": 0.44347228538162214, "全帯域_40ms_rms_envelope_nrmse": 0.03460795740218267, "全帯域_40ms_envelope_correlation": 0.9979472156559169, "全帯域_40ms_modulation_mae_db": 0.2762154889473064, "300–1000 Hz_10ms_log_rms_mae_db": 1.415348820721273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09042658129622358, "300–1000 Hz_10ms_envelope_correlation": 0.9922332004558645, "300–1000 Hz_10ms_modulation_mae_db": 1.12003566502202, "300–1000 Hz_40ms_log_rms_mae_db": 0.869234581963855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05269482095341691, "300–1000 Hz_40ms_envelope_correlation": 0.9974778803883695, "300–1000 Hz_40ms_modulation_mae_db": 1.2425780055777977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4071806896908208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10847680292235866, "1000–3000 Hz_10ms_envelope_correlation": 0.9912259156013753, "1000–3000 Hz_10ms_modulation_mae_db": 0.6111907245508635, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8969010130347866, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06346442381043042, "1000–3000 Hz_40ms_envelope_correlation": 0.997364483024137, "1000–3000 Hz_40ms_modulation_mae_db": 0.5783287129142903, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5848171376917766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15986563634058196, "3000–6000 Hz_10ms_envelope_correlation": 0.9853527732325915, "3000–6000 Hz_10ms_modulation_mae_db": 0.5040523611125924, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9842741981988774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08395698993222644, "3000–6000 Hz_40ms_envelope_correlation": 0.9963963633160666, "3000–6000 Hz_40ms_modulation_mae_db": 0.7546850753754967, "mr_stft_db": 5.230180961623769, "band_mae_db": 0.19422029195839088, "rms_gain_error_db": -0.08453442162260143, "output_peak": 0.48708801833483933}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8944173101643496, "全帯域_10ms_rms_envelope_nrmse": 0.0625469493133782, "全帯域_10ms_envelope_correlation": 0.9935119736635859, "全帯域_10ms_modulation_mae_db": 0.08627473673576723, "全帯域_40ms_log_rms_mae_db": 0.43456346544890717, "全帯域_40ms_rms_envelope_nrmse": 0.0326864589037386, "全帯域_40ms_envelope_correlation": 0.9980082840489309, "全帯域_40ms_modulation_mae_db": 0.075613334951328, "300–1000 Hz_10ms_log_rms_mae_db": 1.398954195069174, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08789004471253566, "300–1000 Hz_10ms_envelope_correlation": 0.9923511730453006, "300–1000 Hz_10ms_modulation_mae_db": 0.18106795019695418, "300–1000 Hz_40ms_log_rms_mae_db": 0.8642894304569153, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05515175147542178, "300–1000 Hz_40ms_envelope_correlation": 0.9967261398191553, "300–1000 Hz_40ms_modulation_mae_db": 0.2045038574968193, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3964702318842985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10645734518924953, "1000–3000 Hz_10ms_envelope_correlation": 0.9913016904130371, "1000–3000 Hz_10ms_modulation_mae_db": 0.40312102057278515, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8702192123245251, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058707485361616386, "1000–3000 Hz_40ms_envelope_correlation": 0.9975221362381674, "1000–3000 Hz_40ms_modulation_mae_db": 0.5051723345249154, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5788666819798258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12367980220033531, "3000–6000 Hz_10ms_envelope_correlation": 0.9904433759732304, "3000–6000 Hz_10ms_modulation_mae_db": 0.8785432565176002, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9455861999494142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05782045092155733, "3000–6000 Hz_40ms_envelope_correlation": 0.997879247094831, "3000–6000 Hz_40ms_modulation_mae_db": 1.4446136869266475, "mr_stft_db": 5.220388851707429, "band_mae_db": 0.08982376561021217, "rms_gain_error_db": -0.0037795808383980813, "output_peak": 0.5628896430297678}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8562868230164793, "全帯域_10ms_rms_envelope_nrmse": 0.06483518992870975, "全帯域_10ms_envelope_correlation": 0.9930596435123213, "全帯域_10ms_modulation_mae_db": 0.25467887952606433, "全帯域_40ms_log_rms_mae_db": 0.4245783705860125, "全帯域_40ms_rms_envelope_nrmse": 0.03459296953876233, "全帯域_40ms_envelope_correlation": 0.9978152486769458, "全帯域_40ms_modulation_mae_db": 0.2148644796622028, "300–1000 Hz_10ms_log_rms_mae_db": 1.379151924944088, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08197181580915583, "300–1000 Hz_10ms_envelope_correlation": 0.9933606812372022, "300–1000 Hz_10ms_modulation_mae_db": 0.6988382779995312, "300–1000 Hz_40ms_log_rms_mae_db": 0.785281963553553, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044428813451477416, "300–1000 Hz_40ms_envelope_correlation": 0.9979349963956434, "300–1000 Hz_40ms_modulation_mae_db": 0.8253708001881793, "1000–3000 Hz_10ms_log_rms_mae_db": 1.392064292420456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10219066058982894, "1000–3000 Hz_10ms_envelope_correlation": 0.9918765832412018, "1000–3000 Hz_10ms_modulation_mae_db": 0.30148100737160394, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8522551118066928, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054685703353551406, "1000–3000 Hz_40ms_envelope_correlation": 0.9977214100807394, "1000–3000 Hz_40ms_modulation_mae_db": 0.347036562759778, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5707648755740895, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14412338842159397, "3000–6000 Hz_10ms_envelope_correlation": 0.9871824397854259, "3000–6000 Hz_10ms_modulation_mae_db": 0.5788250065808735, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9309071402880408, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07307522161241217, "3000–6000 Hz_40ms_envelope_correlation": 0.9969634517248078, "3000–6000 Hz_40ms_modulation_mae_db": 0.6843583048248091, "mr_stft_db": 5.214262274736311, "band_mae_db": 0.09086223392478394, "rms_gain_error_db": -0.031084732420598053, "output_peak": 0.508286940753268}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.25200388220775, "spectral_convergence": 0.2564656653539269, "spectral_convergence_status": "ok", "amplitude_floor": 1.4935395043887774e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.054157462859201, "spectral_convergence": 0.22287315549034206, "spectral_convergence_status": "ok", "amplitude_floor": 1.0494802571211602e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.368687746226568, "spectral_convergence": 0.22412970544815905, "spectral_convergence_status": "ok", "amplitude_floor": 9.412333460941768e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00214444778551191, "output_power": 0.002123955754705217, "delta_db": -0.04170010727814776, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003131735025595309, "output_power": 0.0031380241222209292, "delta_db": 0.008712681978158469, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006987919290953902, "output_power": 0.000681385104658973, "delta_db": -0.10955244836098695, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004380829302633605, "output_power": 0.0004099289772302382, "delta_db": -0.2884771239589117, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014837331900785606, "output_power": 0.00014485890688233942, "delta_db": -0.10410608007184338, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003193593810232492, "output_power": 0.00030857108322057366, "delta_db": -0.14924452788907377, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11696549492285369, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04700189004883553, "pitch": null, "output_rms_dbfs": -21.670618621311945, "output_peak": 0.4846016273288691, "reference_rms_dbfs": -21.62361673126311, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.2249496970978395}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020729159449367856, "log_rms_mae_db": 0.8623075889145267, "rms_envelope_nrmse": 0.06036168933569856, "envelope_correlation": 0.9939687203989779, "rms_gain_error_db": -0.046998309624430856, "reference_mean_power": 0.0068864212769201745, "output_mean_power": 0.006812299859175792}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4588043207823824e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006447788388285564, "output_power": 0.0006219336471867989, "reference_active": true, "delta_db": -0.1566672261947971}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001410929626621107, "output_power": 0.00013706922574684507, "reference_active": true, "delta_db": -0.1256539323859629}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.775085636154685e-05, "output_power": 2.5464244390486264e-05, "reference_active": true, "delta_db": -0.3734559579566438}], "active_band_count": 3, "active_band_mae_db": 0.2185923721791346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001768834026970723, "log_rms_mae_db": 0.3816064167851395, "rms_envelope_nrmse": 0.027061582521441304, "envelope_correlation": 0.9986829619401756, "rms_gain_error_db": -0.04664460129157197, "reference_mean_power": 0.006918665749586704, "output_mean_power": 0.0068447547150872765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2539809640994155e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005542590263527943, "output_power": 0.0005376158801723288, "reference_active": true, "delta_db": -0.13240686035583954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.178340930501843e-05, "output_power": 6.973231846138637e-05, "reference_active": true, "delta_db": -0.12589976137273448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.482399562538733e-06, "output_power": 1.4422764376134311e-06, "reference_active": true, "delta_db": -0.11916769621385156}], "active_band_count": 3, "active_band_mae_db": 0.12582477264747519}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016054516244843413, "log_rms_mae_db": 1.3555051050872982, "rms_envelope_nrmse": 0.081087029938105, "envelope_correlation": 0.9935567338306767, "rms_gain_error_db": -0.0877554421754621, "reference_mean_power": 0.0021921303983993327, "output_mean_power": 0.002148279777374267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.01765222997539e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005084502896759698, "output_power": 0.0004927312605552626, "reference_active": true, "delta_db": -0.13638382822739173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.688692893438751e-05, "output_power": 8.853647386592852e-05, "reference_active": true, "delta_db": -0.3914296872933601}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1104523408685175e-05, "output_power": 1.7858827523964074e-05, "reference_active": true, "delta_db": -0.7252260619315423}], "active_band_count": 3, "active_band_mae_db": 0.4176798591507647}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013960343413021469, "log_rms_mae_db": 0.8370902162870354, "rms_envelope_nrmse": 0.046531365486521654, "envelope_correlation": 0.9977697820616765, "rms_gain_error_db": -0.08777888152591473, "reference_mean_power": 0.00220313483790695, "output_mean_power": 0.002159052435191681}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.111026246795902e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004662873703819768, "output_power": 0.00044967296863872584, "reference_active": true, "delta_db": -0.15756871094450392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.729339985490381e-05, "output_power": 4.3115022900478636e-05, "reference_active": true, "delta_db": -0.4017191483555043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4261322714398465e-06, "output_power": 9.933395295917002e-07, "reference_active": true, "delta_db": -1.5706208907870987}], "active_band_count": 3, "active_band_mae_db": 0.709969583362369}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.530788197178361e-05, "log_rms_mae_db": 1.366853693468481, "rms_envelope_nrmse": 0.10269383614756794, "envelope_correlation": 0.9922991575098438, "rms_gain_error_db": -0.28492830773429173, "reference_mean_power": 0.0004376816454019562, "output_mean_power": 0.0004098882828538553}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8441025227565861e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.86794553375767e-05, "output_power": 9.297890312946243e-05, "reference_active": true, "delta_db": -0.25842325218096085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2863459584597701e-05, "output_power": 1.067363948106676e-05, "reference_active": true, "delta_db": -0.8104525661449669}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6929848290709045e-06, "output_power": 2.1016716045879243e-06, "reference_active": true, "delta_db": -1.0766905024601192}], "active_band_count": 3, "active_band_mae_db": 0.715188773595349}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.347481502172338e-05, "log_rms_mae_db": 0.8422627730890054, "rms_envelope_nrmse": 0.05883640399807308, "envelope_correlation": 0.9979468766918235, "rms_gain_error_db": -0.2849284682287589, "reference_mean_power": 0.00043987971266632326, "output_mean_power": 0.00041194675472027066}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.705293511824543e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.166386964932698e-05, "output_power": 8.616867168416263e-05, "reference_active": true, "delta_db": -0.26848789099303694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.453379972535193e-06, "output_power": 4.665552429465005e-06, "reference_active": true, "delta_db": -0.677626846634079}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5473641354942359e-07, "output_power": 1.1986034928467426e-07, "reference_active": true, "delta_db": -1.1091698770645837}], "active_band_count": 3, "active_band_mae_db": 0.6850948715638999}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012710775449688844, "log_rms_mae_db": 1.617167676127501, "rms_envelope_nrmse": 0.11779454062851612, "envelope_correlation": 0.9913308884103582, "rms_gain_error_db": -0.06532691844112051, "reference_mean_power": 0.00030261254733777104, "output_mean_power": 0.00029809468970829096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.325105138813031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3055217502518526e-05, "output_power": 4.1185431749123296e-05, "reference_active": true, "delta_db": -0.19282164426599152}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2532169127733897e-05, "output_power": 1.022532454375888e-05, "reference_active": true, "delta_db": -0.88349146201621}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7715638592807256e-06, "output_power": 1.2481216762501786e-06, "reference_active": true, "delta_db": -1.5209988620600017}], "active_band_count": 3, "active_band_mae_db": 0.8657706561140678}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010399786917892939, "log_rms_mae_db": 0.9803209801142976, "rms_envelope_nrmse": 0.05758162830359096, "envelope_correlation": 0.9978829787293678, "rms_gain_error_db": -0.0653291311611383, "reference_mean_power": 0.0003041332985815958, "output_mean_power": 0.00029959258423656413}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.904406483303471e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.339234052233676e-05, "output_power": 4.0923590063746177e-05, "reference_active": true, "delta_db": -0.25439350361266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.902501195013482e-06, "output_power": 4.8518204871702165e-06, "reference_active": true, "delta_db": -0.8513135967623828}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6308355063930231e-07, "output_power": 9.077529334740575e-08, "reference_active": true, "delta_db": -2.5444249725701424}], "active_band_count": 3, "active_band_mae_db": 1.2167106909817285}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9322170063789829, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ff768e4630d228fbbea33c61bf9fd1bbba23ada4eea7b98d05be43c89f0ad9d5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9909464701405803, "standard_deviation": 0.001008392843653169, "minimum": 0.9890513012353483, "maximum": 0.9922991575098438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4000970507908304, "standard_deviation": 0.01513758086065227, "minimum": 1.366853693468481, "maximum": 1.4125833099733327, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5220552178171769, "standard_deviation": 0.13446525024347442, "minimum": 0.30148100737160394, "maximum": 0.715188773595349, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10986555471871587, "standard_deviation": 0.0060958265025534185, "minimum": 0.10219066058982894, "maximum": 0.11954944966636492, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971675832808404, "standard_deviation": 0.0006323854569167636, "minimum": 0.996326572830568, "maximum": 0.9979468766918235, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8758495469287055, "standard_deviation": 0.02499107659380887, "minimum": 0.8422627730890054, "maximum": 0.9053496101672267, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.47329626568372535, "standard_deviation": 0.1447886764461418, "minimum": 0.22615557487412663, "maximum": 0.6850948715638999, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06469749378902664, "standard_deviation": 0.007873550627277971, "minimum": 0.054685703353551406, "maximum": 0.0764170965400043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9890202668820313, "standard_deviation": 0.0024587129626997962, "minimum": 0.9853527732325915, "maximum": 0.9918402321633503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5982673869364479, "standard_deviation": 0.023706301558287568, "minimum": 1.5707648755740895, "maximum": 1.6439631285469332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6002735873954405, "standard_deviation": 0.26650295847646005, "minimum": 0.23369580704695422, "maximum": 0.9488593804033124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13309417531704035, "standard_deviation": 0.016205907011294658, "minimum": 0.11571396057827124, "maximum": 0.15986563634058196, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9975303582343726, "standard_deviation": 0.0005607328360991374, "minimum": 0.9963963633160666, "maximum": 0.9978829787293678, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9671029499419255, "standard_deviation": 0.019623283681673827, "minimum": 0.9309071402880408, "maximum": 0.9870692655895537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8539415878896391, "standard_deviation": 0.40899553170535863, "minimum": 0.33682813113495996, "maximum": 1.4446136869266475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06423057667347373, "standard_deviation": 0.009462801201836632, "minimum": 0.05758162830359096, "maximum": 0.08395698993222644, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925928890111838, "standard_deviation": 0.0009107109456716601, "minimum": 0.9912152108370696, "maximum": 0.9937547539389061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.38152754139347, "standard_deviation": 0.03583585489330698, "minimum": 1.326598379151712, "maximum": 1.4376543718315664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6525868948095717, "standard_deviation": 0.3433110700408744, "minimum": 0.18106795019695418, "maximum": 1.12003566502202, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08684958808797678, "standard_deviation": 0.004952164650833997, "minimum": 0.081087029938105, "maximum": 0.09406440056955073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974362507189461, "standard_deviation": 0.00047034395518557985, "minimum": 0.9967261398191553, "maximum": 0.9980638884334209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8235989981459219, "standard_deviation": 0.03815678965762354, "minimum": 0.778125340626864, "maximum": 0.869234581963855, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7580197370423054, "standard_deviation": 0.36002147023943165, "minimum": 0.2045038574968193, "maximum": 1.2425780055777977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04987096163887397, "standard_deviation": 0.0036570748919363093, "minimum": 0.044428813451477416, "maximum": 0.05515175147542178, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11583900225526853, "standard_deviation": 0.03578470462754845, "minimum": 0.07834954531421884, "maximum": 0.19422029195839088, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.220698620320501, "standard_deviation": 0.005422568397860144, "minimum": 5.214262274736311, "maximum": 5.230180961623769, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.518254228343426, "standard_deviation": 0.033668781196435925, "minimum": 0.4846016273288691, "maximum": 0.5730908629535096, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04723681258592605, "standard_deviation": 0.029899566814896084, "minimum": -0.09680266083571516, "maximum": -0.0037795808383980813, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9935569647703291, "standard_deviation": 0.0005976942408449089, "minimum": 0.9928986154644593, "maximum": 0.9946664009513472, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8635170272101585, "standard_deviation": 0.027143593045511248, "minimum": 0.8229786366150531, "maximum": 0.9004400768239725, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26580528991834695, "standard_deviation": 0.1552562094318391, "minimum": 0.021213011894400075, "maximum": 0.4762155972739373, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06246559013334861, "standard_deviation": 0.002796905972964035, "minimum": 0.05751460878549288, "maximum": 0.06546057996089867, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981598495434523, "standard_deviation": 0.00032220234340105797, "minimum": 0.9978152486769458, "maximum": 0.9986829619401756, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4246295185400445, "standard_deviation": 0.03710385837413756, "minimum": 0.3681942389747876, "maximum": 0.48452749511649174, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.21881334013408696, "standard_deviation": 0.10814954867703319, "minimum": 0.075613334951328, "maximum": 0.41178493189538856, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031969494940065596, "standard_deviation": 0.0026621226445314, "minimum": 0.027061582521441304, "maximum": 0.03460795740218267, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2094-142345-0040", "language": "en", "style": "normal", "speaker": "librispeech_2094", "source_utterance_id": "2094-142345-0040", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 38160, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2094-142345-0040.wav", "source_file_sha256": "269d27624a95b857bfeeb64ffd0347c64defe5cdbee2f28455efa9ddc14fa7bc", "decoded_pcm_sha256": "9cbc49d67edc85ca6ba00d13dae38ec545b7aaf549562d1af9d9e72881814d19", "original_source_sha256": "8b3f5a025e4853fabdb000f2820a651f4a66a5ef234d912581ae29fafea72e48", "corpus_entry_id": "librispeech_2094-142345-0040"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.176392078399658, "predicted_mos": 4.176392078399658, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "2ccefe8d31a8f360231ebcccab41b3acc0778bbe0819c93933477e6e89aced02", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.817797656000039, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2094-142345-0040", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "2ccefe8d31a8f360231ebcccab41b3acc0778bbe0819c93933477e6e89aced02", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "a413a9c3cb359ef7f53b985134e93c63b5eacc911b858d84965efaf6c1fb910d"}, "rows": [{"case": "librispeech_2094-142345-0040", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1702492237091064, "predicted_mos": 2.1702492237091064, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "0482565f104fc0356b05dcec8f5a0e0ac1b08b99b9ae45d5d39ea038640764e7", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.6373931460002495, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2094-142345-0040", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "0482565f104fc0356b05dcec8f5a0e0ac1b08b99b9ae45d5d39ea038640764e7", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7493468641434067, "全帯域_10ms_rms_envelope_nrmse": 0.36377156161356844, "全帯域_10ms_envelope_correlation": 0.932027827461512, "全帯域_10ms_modulation_mae_db": 3.7513500601232153, "全帯域_40ms_log_rms_mae_db": 1.0454222391957044, "全帯域_40ms_rms_envelope_nrmse": 0.31129967997062685, "全帯域_40ms_envelope_correlation": 0.949840031459831, "全帯域_40ms_modulation_mae_db": 4.492722166160394, "300–1000 Hz_10ms_log_rms_mae_db": 2.217851754545293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1180568782477763, "300–1000 Hz_10ms_envelope_correlation": 0.9904014673985566, "300–1000 Hz_10ms_modulation_mae_db": 1.0268880357843397, "300–1000 Hz_40ms_log_rms_mae_db": 1.4509228804044112, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06359667718425353, "300–1000 Hz_40ms_envelope_correlation": 0.9971329019545412, "300–1000 Hz_40ms_modulation_mae_db": 1.3411990790507966, "1000–3000 Hz_10ms_log_rms_mae_db": 2.117175921946472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20147290040901394, "1000–3000 Hz_10ms_envelope_correlation": 0.9735242319023095, "1000–3000 Hz_10ms_modulation_mae_db": 0.6978911409172032, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4669322132034717, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12241346534537574, "1000–3000 Hz_40ms_envelope_correlation": 0.989898798020785, "1000–3000 Hz_40ms_modulation_mae_db": 0.6700905561355545, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0831955114911267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18349050472783843, "3000–6000 Hz_10ms_envelope_correlation": 0.9788530251371182, "3000–6000 Hz_10ms_modulation_mae_db": 1.9580106781644797, "3000–6000 Hz_40ms_log_rms_mae_db": 1.388559434441882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11206697760416477, "3000–6000 Hz_40ms_envelope_correlation": 0.9917781200934384, "3000–6000 Hz_40ms_modulation_mae_db": 2.0660616389413042, "mr_stft_db": 5.8753160840430345, "band_mae_db": 0.43815776890709146, "rms_gain_error_db": 1.2759105417834669, "output_peak": 1.2915653835929035}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 37, "at_or_above_full_scale_fraction": 0.0009696016771488469, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0009696016771488469, "longest_full_scale_run_samples": 37, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6755202394258566, "全帯域_10ms_rms_envelope_nrmse": 0.26533227677652677, "全帯域_10ms_envelope_correlation": 0.9622227305158337, "全帯域_10ms_modulation_mae_db": 2.6028670517617845, "全帯域_40ms_log_rms_mae_db": 1.0272248829316006, "全帯域_40ms_rms_envelope_nrmse": 0.21213765385195366, "全帯域_40ms_envelope_correlation": 0.9781168561620965, "全帯域_40ms_modulation_mae_db": 3.1258533791718803, "300–1000 Hz_10ms_log_rms_mae_db": 2.3518678201618037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11293852432847548, "300–1000 Hz_10ms_envelope_correlation": 0.9913679712880015, "300–1000 Hz_10ms_modulation_mae_db": 1.1220974714875758, "300–1000 Hz_40ms_log_rms_mae_db": 1.5392383545325539, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06587357378206303, "300–1000 Hz_40ms_envelope_correlation": 0.996936991152347, "300–1000 Hz_40ms_modulation_mae_db": 1.2097389794657902, "1000–3000 Hz_10ms_log_rms_mae_db": 2.075237468318472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25728685537638024, "1000–3000 Hz_10ms_envelope_correlation": 0.9562786615237681, "1000–3000 Hz_10ms_modulation_mae_db": 0.5045635403937022, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4134172232604847, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.144783820606, "1000–3000 Hz_40ms_envelope_correlation": 0.985291215621177, "1000–3000 Hz_40ms_modulation_mae_db": 0.9852175275514544, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0698005145670892, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17459959196165128, "3000–6000 Hz_10ms_envelope_correlation": 0.9816535784505406, "3000–6000 Hz_10ms_modulation_mae_db": 0.5695657508538289, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3672295097361185, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09493016140651347, "3000–6000 Hz_40ms_envelope_correlation": 0.9946309713019622, "3000–6000 Hz_40ms_modulation_mae_db": 0.6643161716709637, "mr_stft_db": 5.895558710246867, "band_mae_db": 0.3869435605901835, "rms_gain_error_db": 0.9811495969168957, "output_peak": 1.0190174550512712}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 7, "at_or_above_full_scale_fraction": 0.00018343815513626835, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.00018343815513626835, "longest_full_scale_run_samples": 7, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8630736214949146, "全帯域_10ms_rms_envelope_nrmse": 0.1846030633994141, "全帯域_10ms_envelope_correlation": 0.9743596976422334, "全帯域_10ms_modulation_mae_db": 0.6755834926689062, "全帯域_40ms_log_rms_mae_db": 0.9491512413765423, "全帯域_40ms_rms_envelope_nrmse": 0.10282468134592537, "全帯域_40ms_envelope_correlation": 0.9924224837510641, "全帯域_40ms_modulation_mae_db": 1.0045407443216174, "300–1000 Hz_10ms_log_rms_mae_db": 2.307835855234453, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12408545071449288, "300–1000 Hz_10ms_envelope_correlation": 0.9893761369291957, "300–1000 Hz_10ms_modulation_mae_db": 1.3649476096885973, "300–1000 Hz_40ms_log_rms_mae_db": 1.5190460439902636, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06533397264547264, "300–1000 Hz_40ms_envelope_correlation": 0.9969183851569463, "300–1000 Hz_40ms_modulation_mae_db": 1.5110984416461026, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1241491658219442, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15942770461089137, "1000–3000 Hz_10ms_envelope_correlation": 0.983455964287008, "1000–3000 Hz_10ms_modulation_mae_db": 0.8861707322315405, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3662014724375298, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09362206073004388, "1000–3000 Hz_40ms_envelope_correlation": 0.9942265013692196, "1000–3000 Hz_40ms_modulation_mae_db": 1.2546625392998971, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1410034586862254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19923883668534453, "3000–6000 Hz_10ms_envelope_correlation": 0.9751271700699585, "3000–6000 Hz_10ms_modulation_mae_db": 0.7680293803680353, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4309152009769501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11377016316701195, "3000–6000 Hz_40ms_envelope_correlation": 0.9914241506357043, "3000–6000 Hz_40ms_modulation_mae_db": 1.0818675619010107, "mr_stft_db": 5.918596942604822, "band_mae_db": 0.13825689635406116, "rms_gain_error_db": 0.3412450904811094, "output_peak": 0.7278015689911147}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.874256492944367, "全帯域_10ms_rms_envelope_nrmse": 0.22795847850286496, "全帯域_10ms_envelope_correlation": 0.955438162879654, "全帯域_10ms_modulation_mae_db": 1.1773437536085765, "全帯域_40ms_log_rms_mae_db": 0.9563612393817356, "全帯域_40ms_rms_envelope_nrmse": 0.10218734928769171, "全帯域_40ms_envelope_correlation": 0.9908443236172202, "全帯域_40ms_modulation_mae_db": 1.5548680756874482, "300–1000 Hz_10ms_log_rms_mae_db": 2.29029748061467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11943988571361364, "300–1000 Hz_10ms_envelope_correlation": 0.9902894011639496, "300–1000 Hz_10ms_modulation_mae_db": 1.6033835801284575, "300–1000 Hz_40ms_log_rms_mae_db": 1.5908472086709853, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07668560821599567, "300–1000 Hz_40ms_envelope_correlation": 0.9958990659765415, "300–1000 Hz_40ms_modulation_mae_db": 1.9117249617811993, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1625654105975385, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16169193954085234, "1000–3000 Hz_10ms_envelope_correlation": 0.983478643252413, "1000–3000 Hz_10ms_modulation_mae_db": 0.772519654988507, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4183319820425706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09534744729726144, "1000–3000 Hz_40ms_envelope_correlation": 0.9946891365581695, "1000–3000 Hz_40ms_modulation_mae_db": 1.0493977348292352, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1705353287775067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17161756429424835, "3000–6000 Hz_10ms_envelope_correlation": 0.9817677557907166, "3000–6000 Hz_10ms_modulation_mae_db": 0.521760436688142, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4494122510117085, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09629246489143484, "3000–6000 Hz_40ms_envelope_correlation": 0.9940079116610773, "3000–6000 Hz_40ms_modulation_mae_db": 0.6533238625108363, "mr_stft_db": 5.867356891777583, "band_mae_db": 0.29732823669830405, "rms_gain_error_db": -0.2955264106806998, "output_peak": 0.5767840130533178}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8813661824961438, "全帯域_10ms_rms_envelope_nrmse": 0.27169317928863407, "全帯域_10ms_envelope_correlation": 0.9495735209175032, "全帯域_10ms_modulation_mae_db": 2.424008230554382, "全帯域_40ms_log_rms_mae_db": 1.046487598818349, "全帯域_40ms_rms_envelope_nrmse": 0.2071989470047112, "全帯域_40ms_envelope_correlation": 0.9693222009289462, "全帯域_40ms_modulation_mae_db": 2.891678698298465, "300–1000 Hz_10ms_log_rms_mae_db": 2.3047438180263042, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1086112918671564, "300–1000 Hz_10ms_envelope_correlation": 0.9919935782790518, "300–1000 Hz_10ms_modulation_mae_db": 0.7939812904838944, "300–1000 Hz_40ms_log_rms_mae_db": 1.4739959706406571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06664069339351619, "300–1000 Hz_40ms_envelope_correlation": 0.9968528723497204, "300–1000 Hz_40ms_modulation_mae_db": 1.000682956729577, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1132155359771234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22839030085218986, "1000–3000 Hz_10ms_envelope_correlation": 0.9656521664739569, "1000–3000 Hz_10ms_modulation_mae_db": 1.3738278442363552, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4513536576780628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12317915174341112, "1000–3000 Hz_40ms_envelope_correlation": 0.9903167074363352, "1000–3000 Hz_40ms_modulation_mae_db": 1.8164651457080074, "3000–6000 Hz_10ms_log_rms_mae_db": 2.10797808603181, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17343603562051546, "3000–6000 Hz_10ms_envelope_correlation": 0.980903721116908, "3000–6000 Hz_10ms_modulation_mae_db": 0.8452822645200232, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3923506139069928, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08887756162052948, "3000–6000 Hz_40ms_envelope_correlation": 0.9949881890938603, "3000–6000 Hz_40ms_modulation_mae_db": 1.0955196524828044, "mr_stft_db": 5.889546600885701, "band_mae_db": 0.2658155733352109, "rms_gain_error_db": 0.6241860359057426, "output_peak": 1.25703121341984}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 27, "at_or_above_full_scale_fraction": 0.0007075471698113207, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0007075471698113207, "longest_full_scale_run_samples": 27, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7758185657401813, "全帯域_10ms_rms_envelope_nrmse": 0.40272868695592806, "全帯域_10ms_envelope_correlation": 0.9114662989221268, "全帯域_10ms_modulation_mae_db": 2.3538539471315705, "全帯域_40ms_log_rms_mae_db": 1.0020423855977942, "全帯域_40ms_rms_envelope_nrmse": 0.3150322245224675, "全帯域_40ms_envelope_correlation": 0.9478914282857492, "全帯域_40ms_modulation_mae_db": 2.9132190836816982, "300–1000 Hz_10ms_log_rms_mae_db": 2.343201628704335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13005305771906528, "300–1000 Hz_10ms_envelope_correlation": 0.9882947572141016, "300–1000 Hz_10ms_modulation_mae_db": 0.8356410065080294, "300–1000 Hz_40ms_log_rms_mae_db": 1.6855068073527835, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07943187407385101, "300–1000 Hz_40ms_envelope_correlation": 0.9953388230260835, "300–1000 Hz_40ms_modulation_mae_db": 1.2419049569170444, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1507277605811335, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19275364984063603, "1000–3000 Hz_10ms_envelope_correlation": 0.9788423974993795, "1000–3000 Hz_10ms_modulation_mae_db": 1.1077394350020062, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4403131475142703, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1373490403456565, "1000–3000 Hz_40ms_envelope_correlation": 0.991311659527434, "1000–3000 Hz_40ms_modulation_mae_db": 1.3109268963482794, "3000–6000 Hz_10ms_log_rms_mae_db": 2.137437274465393, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15155243684880296, "3000–6000 Hz_10ms_envelope_correlation": 0.9855525976802392, "3000–6000 Hz_10ms_modulation_mae_db": 1.1462601172330238, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4222740793037052, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0786618411286205, "3000–6000 Hz_40ms_envelope_correlation": 0.9960634631911229, "3000–6000 Hz_40ms_modulation_mae_db": 1.3411477344400087, "mr_stft_db": 5.8931760167157305, "band_mae_db": 0.5713521444243376, "rms_gain_error_db": 1.2655341515962522, "output_peak": 1.1074531107004804}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 32, "at_or_above_full_scale_fraction": 0.0008385744234800838, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0008385744234800838, "longest_full_scale_run_samples": 32, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.0700570619818293, "全帯域_10ms_rms_envelope_nrmse": 0.28138491423301903, "全帯域_10ms_envelope_correlation": 0.9337955152077807, "全帯域_10ms_modulation_mae_db": 2.4840776129364355, "全帯域_40ms_log_rms_mae_db": 1.257129379270002, "全帯域_40ms_rms_envelope_nrmse": 0.21774581685779515, "全帯域_40ms_envelope_correlation": 0.959823283122541, "全帯域_40ms_modulation_mae_db": 3.0844104625244317, "300–1000 Hz_10ms_log_rms_mae_db": 2.2636933948775835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11059806629746562, "300–1000 Hz_10ms_envelope_correlation": 0.9916768739584869, "300–1000 Hz_10ms_modulation_mae_db": 1.0767805539429844, "300–1000 Hz_40ms_log_rms_mae_db": 1.5306234472204125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06615032570586667, "300–1000 Hz_40ms_envelope_correlation": 0.9969201327087246, "300–1000 Hz_40ms_modulation_mae_db": 1.1894681103748612, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0773859127444085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15611391080875994, "1000–3000 Hz_10ms_envelope_correlation": 0.9843523194089676, "1000–3000 Hz_10ms_modulation_mae_db": 0.3792128022316478, "1000–3000 Hz_40ms_log_rms_mae_db": 1.383748394313576, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09501851534288464, "1000–3000 Hz_40ms_envelope_correlation": 0.993873055155631, "1000–3000 Hz_40ms_modulation_mae_db": 0.7498813928940997, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1730540421721627, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20397386297585846, "3000–6000 Hz_10ms_envelope_correlation": 0.973827514466672, "3000–6000 Hz_10ms_modulation_mae_db": 2.1045429567344733, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4847262743713798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1367007261114766, "3000–6000 Hz_40ms_envelope_correlation": 0.9882126153698165, "3000–6000 Hz_40ms_modulation_mae_db": 2.242111565352433, "mr_stft_db": 5.863951544330132, "band_mae_db": 0.5105543890477809, "rms_gain_error_db": -0.8888179343933257, "output_peak": 0.6133734485642796}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8169844187214492, "全帯域_10ms_rms_envelope_nrmse": 0.22838164886740747, "全帯域_10ms_envelope_correlation": 0.9554819531483156, "全帯域_10ms_modulation_mae_db": 0.3415352371271662, "全帯域_40ms_log_rms_mae_db": 0.9271678132459853, "全帯域_40ms_rms_envelope_nrmse": 0.1316607966815895, "全帯域_40ms_envelope_correlation": 0.9835617431850286, "全帯域_40ms_modulation_mae_db": 0.37428973883154937, "300–1000 Hz_10ms_log_rms_mae_db": 2.262003961285843, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12522575227652918, "300–1000 Hz_10ms_envelope_correlation": 0.9892319751113773, "300–1000 Hz_10ms_modulation_mae_db": 1.1963650108375339, "300–1000 Hz_40ms_log_rms_mae_db": 1.5158573980314989, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07531845865118725, "300–1000 Hz_40ms_envelope_correlation": 0.9959239686475102, "300–1000 Hz_40ms_modulation_mae_db": 1.4052744580981589, "1000–3000 Hz_10ms_log_rms_mae_db": 2.204451063284206, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2550957029767006, "1000–3000 Hz_10ms_envelope_correlation": 0.9628734138161618, "1000–3000 Hz_10ms_modulation_mae_db": 2.385219912457145, "1000–3000 Hz_40ms_log_rms_mae_db": 1.484347656822226, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19873600780689982, "1000–3000 Hz_40ms_envelope_correlation": 0.9807369622492811, "1000–3000 Hz_40ms_modulation_mae_db": 2.4102098423972, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0527918516027137, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1532617569494819, "3000–6000 Hz_10ms_envelope_correlation": 0.9852844739487298, "3000–6000 Hz_10ms_modulation_mae_db": 1.0325454677614416, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3299554116767267, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08579317614199052, "3000–6000 Hz_40ms_envelope_correlation": 0.9951511657717967, "3000–6000 Hz_40ms_modulation_mae_db": 1.1917263080549731, "mr_stft_db": 5.871604836630273, "band_mae_db": 0.3718893742699913, "rms_gain_error_db": -0.11479711404896258, "output_peak": 0.6997155993858947}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 38160, "output_samples": 38160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.097942602905295, "spectral_convergence": 0.5982854886524577, "spectral_convergence_status": "ok", "amplitude_floor": 5.674431450769592e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.698381329077874, "spectral_convergence": 0.5655289911336402, "spectral_convergence_status": "ok", "amplitude_floor": 3.1139992269810527e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.829624320145933, "spectral_convergence": 0.5226643001459815, "spectral_convergence_status": "ok", "amplitude_floor": 1.28722240515293e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.003603182122246331, "output_power": 0.005538408590618711, "delta_db": 1.8669877821719707, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001490228689925631, "output_power": 0.0015700075584987301, "delta_db": 0.22648823049411088, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00043834980790852295, "output_power": 0.00043650830864540156, "delta_db": -0.018283059490808183, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023807008265260595, "output_power": 0.00022706974103155496, "delta_db": -0.20545558071219056, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.11014495688624e-05, "output_power": 8.581491598386195e-05, "delta_db": 0.2453416497616628, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.691475709405268e-05, "output_power": 2.7329362155396257e-05, "delta_db": 0.06639031081180563, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.43815776890709146, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.2759105417834669, "pitch": null, "output_rms_dbfs": -21.03190675291315, "output_peak": 1.2915653835929035, "reference_rms_dbfs": -22.307817294696616, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8753160840430345}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 38160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.0004033177834583336, "log_rms_mae_db": 1.7493468641434067, "rms_envelope_nrmse": 0.36377156161356844, "envelope_correlation": 0.932027827461512, "rms_gain_error_db": 1.275904138605504, "reference_mean_power": 0.005890133913079112, "output_mean_power": 0.007901609856656285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 4.702783438048151e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019529842144495992, "output_power": 0.0028464263263213406, "reference_active": true, "delta_db": 1.63601214512651}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0010897406087412391, "output_power": 0.0026572703759655428, "reference_active": true, "delta_db": 3.8711261090425575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00042203241052064084, "output_power": 0.0015850282135031523, "reference_active": true, "delta_db": 5.746911926200578}], "active_band_count": 3, "active_band_mae_db": 3.7513500601232153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0002641139400842597, "log_rms_mae_db": 1.0454222391957044, "rms_envelope_nrmse": 0.31129967997062685, "envelope_correlation": 0.949840031459831, "rms_gain_error_db": 1.2759640385660111, "reference_mean_power": 0.0059649962331930016, "output_mean_power": 0.008002147966907501}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.959398727002631e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020291985866854774, "output_power": 0.0038694551381984565, "reference_active": true, "delta_db": 2.8032526472477137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005646977153887027, "output_power": 0.0016632233117432212, "reference_active": true, "delta_db": 4.691345327342364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.714480412149099e-05, "output_power": 0.00014731769832964536, "reference_active": true, "delta_db": 5.9835685238911065}], "active_band_count": 3, "active_band_mae_db": 4.492722166160394}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.00011878678262042868, "log_rms_mae_db": 2.217851754545293, "rms_envelope_nrmse": 0.1180568782477763, "envelope_correlation": 0.9904014673985566, "rms_gain_error_db": -0.05978389636645431, "reference_mean_power": 0.0009978964081743123, "output_mean_power": 0.0009842537311197843}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 1.0827926032603135e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006103304193321489, "output_power": 0.000574678305341699, "reference_active": true, "delta_db": -0.26140213654111044}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011556410212235237, "output_power": 0.0001050810093705929, "reference_active": true, "delta_db": -0.4129871357283085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8931134571049757e-05, "output_power": 1.0878002621735741e-05, "reference_active": true, "delta_db": -2.4062748350836007}], "active_band_count": 3, "active_band_mae_db": 1.0268880357843397}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 9.505044043492436e-05, "log_rms_mae_db": 1.4509228804044112, "rms_envelope_nrmse": 0.06359667718425353, "envelope_correlation": 0.9971329019545412, "rms_gain_error_db": -0.05979063933696312, "reference_mean_power": 0.0010106286473201321, "output_mean_power": 0.000996810354595534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.735673127654544e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005396175923660179, "output_power": 0.0005100246002582613, "reference_active": true, "delta_db": -0.24494975766085766}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.461345336739454e-05, "output_power": 5.8215092774151945e-05, "reference_active": true, "delta_db": -0.45287359303296715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3971113807678808e-06, "output_power": 1.1145789204307934e-06, "reference_active": true, "delta_db": -3.325773886458565}], "active_band_count": 3, "active_band_mae_db": 1.3411990790507966}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 9.972124238081888e-05, "log_rms_mae_db": 2.117175921946472, "rms_envelope_nrmse": 0.20147290040901394, "envelope_correlation": 0.9735242319023095, "rms_gain_error_db": -0.08693802102173334, "reference_mean_power": 0.00028254365881671625, "output_mean_power": 0.0002769438738032201}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 3.4051840450675846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002046408997048573, "output_power": 0.00018905113090386043, "reference_active": true, "delta_db": -0.3441315686098372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.521356834551269e-05, "output_power": 3.475082810219353e-05, "reference_active": true, "delta_db": -1.1430362561714653}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018106896586635e-06, "output_power": 9.219819137731411e-06, "reference_active": true, "delta_db": 0.606505597970307}], "active_band_count": 3, "active_band_mae_db": 0.6978911409172032}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 8.26289228894579e-05, "log_rms_mae_db": 1.4669322132034717, "rms_envelope_nrmse": 0.12241346534537574, "envelope_correlation": 0.989898798020785, "rms_gain_error_db": -0.08694911552403536, "reference_mean_power": 0.0002861487173387815, "output_mean_power": 0.00028047676649603655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.083140770966499e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019461067978585587, "output_power": 0.00017736452530966695, "reference_active": true, "delta_db": -0.4029990875363592}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0239988981192927e-05, "output_power": 1.5481459596078127e-05, "reference_active": true, "delta_db": -1.1639836805951798}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.640717194960459e-07, "output_power": 6.899303269652497e-07, "reference_active": true, "delta_db": -0.44328890027512424}], "active_band_count": 3, "active_band_mae_db": 0.6700905561355545}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 4.3426949519701344e-05, "log_rms_mae_db": 2.0831955114911267, "rms_envelope_nrmse": 0.18349050472783843, "envelope_correlation": 0.9788530251371182, "rms_gain_error_db": -0.09828312070190703, "reference_mean_power": 4.5529138246552346e-05, "output_mean_power": 4.451036116098758e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 6.179039674292901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3063171222465546e-05, "output_power": 3.099253564760902e-05, "reference_active": true, "delta_db": -0.2808739677123735}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2207009907392542e-05, "output_power": 8.203540375818613e-06, "reference_active": true, "delta_db": -1.7260797712625011}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2513877801441335e-06, "output_power": 9.241500151904963e-07, "reference_active": true, "delta_db": -3.8670782955185645}], "active_band_count": 3, "active_band_mae_db": 1.9580106781644797}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 3.683581645514392e-05, "log_rms_mae_db": 1.388559434441882, "rms_envelope_nrmse": 0.11206697760416477, "envelope_correlation": 0.9917781200934384, "rms_gain_error_db": -0.09830919298487087, "reference_mean_power": 4.610980557969544e-05, "output_mean_power": 4.507776464485884e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 5.621286439624602e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.226124623783574e-05, "output_power": 2.869683848149355e-05, "reference_active": true, "delta_db": -0.5084708665214145}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.247687157348319e-06, "output_power": 4.645344345902222e-06, "reference_active": true, "delta_db": -1.9318152646444142}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.30977962238882e-07, "output_power": 9.722560784112847e-08, "reference_active": true, "delta_db": -3.757898785658083}], "active_band_count": 3, "active_band_mae_db": 2.0660616389413042}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8056874503168425, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "02d726ef8ed8479da61b276d7ba4546a157bb415e4b6441022092fd137474376", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9735572247704956, "standard_deviation": 0.010797290052316267, "minimum": 0.9562786615237681, "maximum": 0.9843523194089676, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1281135299089122, "standard_deviation": 0.04351062467942093, "minimum": 2.075237468318472, "maximum": 2.204451063284206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0133931328072634, "standard_deviation": 0.6386998449441149, "minimum": 0.3792128022316478, "maximum": 2.385219912457145, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20152912055192804, "standard_deviation": 0.04174567606457944, "minimum": 0.15611391080875994, "maximum": 0.25728685537638024, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9900430044922541, "standard_deviation": 0.00484980988562778, "minimum": 0.9807369622492811, "maximum": 0.9946891365581695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4280807184090238, "standard_deviation": 0.040481249667701856, "minimum": 1.3662014724375298, "maximum": 1.484347656822226, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2808564543954661, "standard_deviation": 0.5806217246821517, "minimum": 0.6700905561355545, "maximum": 2.4102098423972, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12630618865219165, "standard_deviation": 0.03533511446326242, "minimum": 0.09362206073004388, "maximum": 0.19873600780689982, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9803712295826104, "standard_deviation": 0.004267654245208848, "minimum": 0.973827514466672, "maximum": 0.9855525976802392, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1169745084742533, "standard_deviation": 0.045632923906542125, "minimum": 2.0527918516027137, "maximum": 2.1730540421721627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.118249631540431, "standard_deviation": 0.6023906774602794, "minimum": 0.521760436688142, "maximum": 2.1045429567344733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17639632375796765, "standard_deviation": 0.01897322723122294, "minimum": 0.15155243684880296, "maximum": 0.20397386297585846, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932820733898473, "standard_deviation": 0.0026158852700558827, "minimum": 0.9882126153698165, "maximum": 0.9960634631911229, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.408177846928183, "standard_deviation": 0.048853319772452566, "minimum": 1.3299554116767267, "maximum": 1.4847262743713798, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2920093119192917, "standard_deviation": 0.5854190300662425, "minimum": 0.6533238625108363, "maximum": 2.242111565352433, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10088663400896776, "standard_deviation": 0.01888382627666713, "minimum": 0.0786618411286205, "maximum": 0.1367007261114766, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9903290201678401, "standard_deviation": 0.0013051229460807542, "minimum": 0.9882947572141016, "maximum": 0.9919935782790518, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.2926869641812857, "standard_deviation": 0.04442819830415353, "minimum": 2.217851754545293, "maximum": 2.3518678201618037, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1275105698576766, "standard_deviation": 0.26638579878278873, "minimum": 0.7939812904838944, "maximum": 1.6033835801284575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11862611339557186, "standard_deviation": 0.007578312738934731, "minimum": 0.1086112918671564, "maximum": 0.13005305771906528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964903926215518, "standard_deviation": 0.0006663471906673729, "minimum": 0.9953388230260835, "maximum": 0.9971329019545412, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5382547638554458, "standard_deviation": 0.07283543221072167, "minimum": 1.4509228804044112, "maximum": 1.6855068073527835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3513864930079413, "standard_deviation": 0.2735365110494694, "minimum": 1.000682956729577, "maximum": 1.9117249617811993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06987889795652574, "standard_deviation": 0.006184461672875961, "minimum": 0.06359667718425353, "maximum": 0.07943187407385101, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3725372429533701, "standard_deviation": 0.13888500021947117, "minimum": 0.13825689635406116, "maximum": 0.5713521444243376, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.884388453404268, "standard_deviation": 0.01833144532259204, "minimum": 5.863951544330132, "maximum": 5.918596942604822, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9115927240948878, "standard_deviation": 0.2911335112703182, "minimum": 0.5767840130533178, "maximum": 1.2915653835929035, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": 0.3986104946950598, "standard_deviation": 0.7853683663163313, "minimum": -0.8888179343933257, "maximum": 1.2759105417834669, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.94679571333687, "standard_deviation": 0.019967431083265614, "minimum": 0.9114662989221268, "maximum": 0.9743596976422334, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8383029308685186, "standard_deviation": 0.11725480675803923, "minimum": 1.6755202394258566, "maximum": 2.0700570619818293, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.9763274232390045, "standard_deviation": 1.143107736030095, "minimum": 0.3415352371271662, "maximum": 3.7513500601232153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2782317262046704, "standard_deviation": 0.0725203457243598, "minimum": 0.1846030633994141, "maximum": 0.40272868695592806, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9714777938140595, "standard_deviation": 0.017602049771519898, "minimum": 0.9478914282857492, "maximum": 0.9924224837510641, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.026373347477214, "standard_deviation": 0.10372711271324182, "minimum": 0.9271678132459853, "maximum": 1.257129379270002, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 2.4301977935846857, "standard_deviation": 1.343644827447106, "minimum": 0.37428973883154937, "maximum": 4.492722166160394, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2000108936903451, "standard_deviation": 0.08435866954414882, "minimum": 0.10218734928769171, "maximum": 0.3150322245224675, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0040", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1396148204803467, "predicted_mos": 2.1396148204803467, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "23a6593a0c15b80326874e16545895f11d5c3f6c5a85342daccfbf89ed1848c5", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.870618999935687, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7776605819564655, "全帯域_10ms_rms_envelope_nrmse": 0.3390242953529821, "全帯域_10ms_envelope_correlation": 0.9398906752645622, "全帯域_10ms_modulation_mae_db": 3.5820975942926725, "全帯域_40ms_log_rms_mae_db": 1.0762045933250857, "全帯域_40ms_rms_envelope_nrmse": 0.2845855984792562, "全帯域_40ms_envelope_correlation": 0.9579757603932175, "全帯域_40ms_modulation_mae_db": 4.398457531659112, "300–1000 Hz_10ms_log_rms_mae_db": 1.9575973017263841, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10574966933795428, "300–1000 Hz_10ms_envelope_correlation": 0.9922349312098452, "300–1000 Hz_10ms_modulation_mae_db": 0.8413161469416383, "300–1000 Hz_40ms_log_rms_mae_db": 1.0746825948827172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05190121900726328, "300–1000 Hz_40ms_envelope_correlation": 0.9980376649846031, "300–1000 Hz_40ms_modulation_mae_db": 1.1895729799404668, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6958960440911237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15743725572902037, "1000–3000 Hz_10ms_envelope_correlation": 0.9838495940152092, "1000–3000 Hz_10ms_modulation_mae_db": 0.66325208393688, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8734523847443234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08184185980158112, "1000–3000 Hz_40ms_envelope_correlation": 0.9956692794062182, "1000–3000 Hz_40ms_modulation_mae_db": 1.0389245764239814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7291473741962433, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16450796928844333, "3000–6000 Hz_10ms_envelope_correlation": 0.9827448361198929, "3000–6000 Hz_10ms_modulation_mae_db": 1.820586761249217, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9152921958546643, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08602838981497982, "3000–6000 Hz_40ms_envelope_correlation": 0.9952786947690913, "3000–6000 Hz_40ms_modulation_mae_db": 2.137922607581755, "mr_stft_db": 5.679217519310725, "band_mae_db": 0.40458536137943796, "rms_gain_error_db": 1.1918778949617566, "output_peak": 1.1999095318718092}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 32, "at_or_above_full_scale_fraction": 0.0008385744234800838, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0008385744234800838, "longest_full_scale_run_samples": 32, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7574724210158577, "全帯域_10ms_rms_envelope_nrmse": 0.30853849295299984, "全帯域_10ms_envelope_correlation": 0.9507779742513, "全帯域_10ms_modulation_mae_db": 2.7732427202976964, "全帯域_40ms_log_rms_mae_db": 1.1063687852573407, "全帯域_40ms_rms_envelope_nrmse": 0.24061012305021615, "全帯域_40ms_envelope_correlation": 0.9743511567862428, "全帯域_40ms_modulation_mae_db": 3.29440309434411, "300–1000 Hz_10ms_log_rms_mae_db": 2.0714252606051433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10755913682798939, "300–1000 Hz_10ms_envelope_correlation": 0.9920477509135924, "300–1000 Hz_10ms_modulation_mae_db": 1.1556495421013258, "300–1000 Hz_40ms_log_rms_mae_db": 1.1021796361701763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0589161994340069, "300–1000 Hz_40ms_envelope_correlation": 0.9974700347655246, "300–1000 Hz_40ms_modulation_mae_db": 1.2839357111990726, "1000–3000 Hz_10ms_log_rms_mae_db": 1.67608630340755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24865617832887268, "1000–3000 Hz_10ms_envelope_correlation": 0.959385354385646, "1000–3000 Hz_10ms_modulation_mae_db": 0.6282128305978786, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8897334377091678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13895953825223464, "1000–3000 Hz_40ms_envelope_correlation": 0.9865826005876044, "1000–3000 Hz_40ms_modulation_mae_db": 1.1385083968887433, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7073938223117695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15368569919243516, "3000–6000 Hz_10ms_envelope_correlation": 0.9854227047920415, "3000–6000 Hz_10ms_modulation_mae_db": 0.6722184006414874, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9110633926968404, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08898152013103587, "3000–6000 Hz_40ms_envelope_correlation": 0.9951378360415144, "3000–6000 Hz_40ms_modulation_mae_db": 1.057257670603871, "mr_stft_db": 5.706970130145613, "band_mae_db": 0.4236590034782645, "rms_gain_error_db": 1.1631322633102654, "output_peak": 1.1147615701257438}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 29, "at_or_above_full_scale_fraction": 0.000759958071278826, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.000759958071278826, "longest_full_scale_run_samples": 29, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8850257484239834, "全帯域_10ms_rms_envelope_nrmse": 0.176608706247627, "全帯域_10ms_envelope_correlation": 0.9754545380085238, "全帯域_10ms_modulation_mae_db": 0.7119320582577898, "全帯域_40ms_log_rms_mae_db": 0.960391064372421, "全帯域_40ms_rms_envelope_nrmse": 0.09870436434423871, "全帯域_40ms_envelope_correlation": 0.9921003106070571, "全帯域_40ms_modulation_mae_db": 1.085823105743456, "300–1000 Hz_10ms_log_rms_mae_db": 2.0533052660073188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11465435717247581, "300–1000 Hz_10ms_envelope_correlation": 0.9908451489468478, "300–1000 Hz_10ms_modulation_mae_db": 1.4096264080733005, "300–1000 Hz_40ms_log_rms_mae_db": 1.1220700489979043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057670997473718136, "300–1000 Hz_40ms_envelope_correlation": 0.9975288546666747, "300–1000 Hz_40ms_modulation_mae_db": 1.5992714053560262, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8224076560988585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14137633146316173, "1000–3000 Hz_10ms_envelope_correlation": 0.9871416126594474, "1000–3000 Hz_10ms_modulation_mae_db": 1.0929422812278684, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9720152129648538, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07747382406285235, "1000–3000 Hz_40ms_envelope_correlation": 0.996323016082113, "1000–3000 Hz_40ms_modulation_mae_db": 1.3834452051529398, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7825152127438741, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.173288115971992, "3000–6000 Hz_10ms_envelope_correlation": 0.9806987930785366, "3000–6000 Hz_10ms_modulation_mae_db": 0.9819148256940408, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9511190071435139, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08850947303679711, "3000–6000 Hz_40ms_envelope_correlation": 0.994672945629221, "3000–6000 Hz_40ms_modulation_mae_db": 1.467476140617607, "mr_stft_db": 5.734026931395161, "band_mae_db": 0.1640081855560245, "rms_gain_error_db": 0.2426390442259107, "output_peak": 0.7494582905513366}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.906752784067268, "全帯域_10ms_rms_envelope_nrmse": 0.2686387423452339, "全帯域_10ms_envelope_correlation": 0.9381257539954656, "全帯域_10ms_modulation_mae_db": 1.358578858485199, "全帯域_40ms_log_rms_mae_db": 1.0534145101528058, "全帯域_40ms_rms_envelope_nrmse": 0.1608617286205521, "全帯域_40ms_envelope_correlation": 0.9779756902482079, "全帯域_40ms_modulation_mae_db": 1.3159732665039214, "300–1000 Hz_10ms_log_rms_mae_db": 1.9798237828802503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11090440146398993, "300–1000 Hz_10ms_envelope_correlation": 0.9916649500294238, "300–1000 Hz_10ms_modulation_mae_db": 1.5520463240611881, "300–1000 Hz_40ms_log_rms_mae_db": 1.1049618763395326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06739229158804626, "300–1000 Hz_40ms_envelope_correlation": 0.9969613341380165, "300–1000 Hz_40ms_modulation_mae_db": 1.8050657249230035, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8046240563397518, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13606872851001187, "1000–3000 Hz_10ms_envelope_correlation": 0.9890384647306153, "1000–3000 Hz_10ms_modulation_mae_db": 1.0400677148149755, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9033047080566587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07788811042183924, "1000–3000 Hz_40ms_envelope_correlation": 0.9973826666593462, "1000–3000 Hz_40ms_modulation_mae_db": 1.3762655915505022, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8007396437365695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1656430100116057, "3000–6000 Hz_10ms_envelope_correlation": 0.9825772014032323, "3000–6000 Hz_10ms_modulation_mae_db": 0.5459462328272043, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9546168428541496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09467548286489623, "3000–6000 Hz_40ms_envelope_correlation": 0.9940377615855224, "3000–6000 Hz_40ms_modulation_mae_db": 0.8972769331395489, "mr_stft_db": 5.669779728971116, "band_mae_db": 0.3634081066727031, "rms_gain_error_db": -0.5888566678307511, "output_peak": 0.5806341171288634}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8825502733499178, "全帯域_10ms_rms_envelope_nrmse": 0.3515948578525851, "全帯域_10ms_envelope_correlation": 0.9335793302599549, "全帯域_10ms_modulation_mae_db": 3.309939877101866, "全帯域_40ms_log_rms_mae_db": 1.1031331297456894, "全帯域_40ms_rms_envelope_nrmse": 0.2719605616086244, "全帯域_40ms_envelope_correlation": 0.9630339688361867, "全帯域_40ms_modulation_mae_db": 3.9234369056164664, "300–1000 Hz_10ms_log_rms_mae_db": 2.115129462777489, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10083032731719463, "300–1000 Hz_10ms_envelope_correlation": 0.9930498932565068, "300–1000 Hz_10ms_modulation_mae_db": 0.8215815308694675, "300–1000 Hz_40ms_log_rms_mae_db": 1.1383202199820956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06046443547344789, "300–1000 Hz_40ms_envelope_correlation": 0.9974060742996512, "300–1000 Hz_40ms_modulation_mae_db": 0.9894046785690729, "1000–3000 Hz_10ms_log_rms_mae_db": 1.790759679529265, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2097202911155426, "1000–3000 Hz_10ms_envelope_correlation": 0.9715465712387313, "1000–3000 Hz_10ms_modulation_mae_db": 1.66241507599655, "1000–3000 Hz_40ms_log_rms_mae_db": 1.009180568953093, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10906885965542999, "1000–3000 Hz_40ms_envelope_correlation": 0.9933021029344772, "1000–3000 Hz_40ms_modulation_mae_db": 1.7877269742368471, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7313401801206534, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1533367436107209, "3000–6000 Hz_10ms_envelope_correlation": 0.9851850016726942, "3000–6000 Hz_10ms_modulation_mae_db": 0.8601375435920927, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9480357177710803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07760376946358187, "3000–6000 Hz_40ms_envelope_correlation": 0.9965408010702426, "3000–6000 Hz_40ms_modulation_mae_db": 1.2020827461042112, "mr_stft_db": 5.703440996247686, "band_mae_db": 0.4811638354062169, "rms_gain_error_db": 1.186608727157979, "output_peak": 1.362427125506974}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 32, "at_or_above_full_scale_fraction": 0.0008385744234800838, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0008647798742138365, "longest_full_scale_run_samples": 32, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.798340501900962, "全帯域_10ms_rms_envelope_nrmse": 0.3679618794526209, "全帯域_10ms_envelope_correlation": 0.9221938615317059, "全帯域_10ms_modulation_mae_db": 2.1209171976171226, "全帯域_40ms_log_rms_mae_db": 0.9942694330677944, "全帯域_40ms_rms_envelope_nrmse": 0.2792651879922594, "全帯域_40ms_envelope_correlation": 0.9568517725214938, "全帯域_40ms_modulation_mae_db": 2.6828256515112625, "300–1000 Hz_10ms_log_rms_mae_db": 2.080020355801502, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11412131943411437, "300–1000 Hz_10ms_envelope_correlation": 0.9909108184459978, "300–1000 Hz_10ms_modulation_mae_db": 0.8889262798173396, "300–1000 Hz_40ms_log_rms_mae_db": 1.2183036320277938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06764225385524471, "300–1000 Hz_40ms_envelope_correlation": 0.9965454628661545, "300–1000 Hz_40ms_modulation_mae_db": 1.1896060516703681, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7132370387427494, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.173894816323668, "1000–3000 Hz_10ms_envelope_correlation": 0.9835258871430715, "1000–3000 Hz_10ms_modulation_mae_db": 0.9431104224295551, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9117551237809177, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11896458766129395, "1000–3000 Hz_40ms_envelope_correlation": 0.994652877203023, "1000–3000 Hz_40ms_modulation_mae_db": 1.2382504962647922, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6836407840373364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13463500099458522, "3000–6000 Hz_10ms_envelope_correlation": 0.9883897927469154, "3000–6000 Hz_10ms_modulation_mae_db": 0.9177856795817217, "3000–6000 Hz_40ms_log_rms_mae_db": 0.92070491103658, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06925740152604322, "3000–6000 Hz_40ms_envelope_correlation": 0.9968622217680853, "3000–6000 Hz_40ms_modulation_mae_db": 1.2299450590246097, "mr_stft_db": 5.693904471252725, "band_mae_db": 0.5827912790292225, "rms_gain_error_db": 1.0924024941995285, "output_peak": 1.0847423545116432}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 26, "at_or_above_full_scale_fraction": 0.0006813417190775681, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0006813417190775681, "longest_full_scale_run_samples": 26, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.109878920014013, "全帯域_10ms_rms_envelope_nrmse": 0.2928463156098961, "全帯域_10ms_envelope_correlation": 0.9281875493569014, "全帯域_10ms_modulation_mae_db": 2.2415756026086675, "全帯域_40ms_log_rms_mae_db": 1.2714738400228982, "全帯域_40ms_rms_envelope_nrmse": 0.23091340265416524, "全帯域_40ms_envelope_correlation": 0.9545716627936963, "全帯域_40ms_modulation_mae_db": 2.6023408603413274, "300–1000 Hz_10ms_log_rms_mae_db": 1.9897575092116986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1091772766685831, "300–1000 Hz_10ms_envelope_correlation": 0.9917734790661468, "300–1000 Hz_10ms_modulation_mae_db": 1.094277948473634, "300–1000 Hz_40ms_log_rms_mae_db": 1.0992389447180648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060619309754337394, "300–1000 Hz_40ms_envelope_correlation": 0.9973320182454919, "300–1000 Hz_40ms_modulation_mae_db": 1.2005981838297068, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7648018226623456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14158724844658066, "1000–3000 Hz_10ms_envelope_correlation": 0.9870358514734873, "1000–3000 Hz_10ms_modulation_mae_db": 0.42882829085919033, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9062339933937391, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0793340037743207, "1000–3000 Hz_40ms_envelope_correlation": 0.9957120272976004, "1000–3000 Hz_40ms_modulation_mae_db": 0.8688370897872731, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7682150487295385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18799500370143388, "3000–6000 Hz_10ms_envelope_correlation": 0.9783364425968796, "3000–6000 Hz_10ms_modulation_mae_db": 2.0032004294346546, "3000–6000 Hz_40ms_log_rms_mae_db": 0.985403168158431, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11807142756781502, "3000–6000 Hz_40ms_envelope_correlation": 0.9923056669809958, "3000–6000 Hz_40ms_modulation_mae_db": 2.423305526653835, "mr_stft_db": 5.677057108293826, "band_mae_db": 0.5749664555611819, "rms_gain_error_db": -0.9532718859460962, "output_peak": 0.5629235393505584}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9453888588928492, "全帯域_10ms_rms_envelope_nrmse": 0.24568533377755217, "全帯域_10ms_envelope_correlation": 0.9482784195071281, "全帯域_10ms_modulation_mae_db": 0.7699479816042238, "全帯域_40ms_log_rms_mae_db": 1.0456185927531327, "全帯域_40ms_rms_envelope_nrmse": 0.13875158203153679, "全帯域_40ms_envelope_correlation": 0.9839840033489934, "全帯域_40ms_modulation_mae_db": 1.0518962985513955, "300–1000 Hz_10ms_log_rms_mae_db": 2.008968818065071, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10078982882770766, "300–1000 Hz_10ms_envelope_correlation": 0.9930737502306957, "300–1000 Hz_10ms_modulation_mae_db": 1.253210470424979, "300–1000 Hz_40ms_log_rms_mae_db": 1.1119760678082165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062194818422440665, "300–1000 Hz_40ms_envelope_correlation": 0.9972645799125631, "300–1000 Hz_40ms_modulation_mae_db": 1.4278819613120886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8118791668204517, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24274286875473547, "1000–3000 Hz_10ms_envelope_correlation": 0.9673009805424997, "1000–3000 Hz_10ms_modulation_mae_db": 2.463310558119989, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0114104560027466, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18505331439197797, "1000–3000 Hz_40ms_envelope_correlation": 0.9847426075624531, "1000–3000 Hz_40ms_modulation_mae_db": 2.4142504638499354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7378158811524012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13081721813642058, "3000–6000 Hz_10ms_envelope_correlation": 0.9890259384280415, "3000–6000 Hz_10ms_modulation_mae_db": 0.9357455767990216, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8781092761976733, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07061423109189749, "3000–6000 Hz_40ms_envelope_correlation": 0.9966722506442881, "3000–6000 Hz_40ms_modulation_mae_db": 1.2070003193272474, "mr_stft_db": 5.686512888957645, "band_mae_db": 0.46543426082099254, "rms_gain_error_db": -0.5145628659677739, "output_peak": 0.5854354268520309}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 38160, "output_samples": 38160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.893758325638374, "spectral_convergence": 0.5978607643888982, "spectral_convergence_status": "ok", "amplitude_floor": 5.674431450769592e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.490000597004569, "spectral_convergence": 0.5529685457728922, "spectral_convergence_status": "ok", "amplitude_floor": 3.1139992269810527e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.653893635289229, "spectral_convergence": 0.5093300612473252, "spectral_convergence_status": "ok", "amplitude_floor": 1.28722240515293e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.003603182122246331, "output_power": 0.005403173658444107, "delta_db": 1.7596271200340534, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001490228689925631, "output_power": 0.0015645223609664478, "delta_db": 0.21128854497031319, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00043834980790852295, "output_power": 0.00043597686155344526, "delta_db": -0.023573798540329448, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023807008265260595, "output_power": 0.00022192794517510416, "delta_db": -0.30492830671678006, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.11014495688624e-05, "output_power": 8.200819634348545e-05, "delta_db": 0.04828643642397768, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.691475709405268e-05, "output_power": 2.6424675847076244e-05, "delta_db": -0.07980796159117401, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.40458536137943796, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1918778949617566, "pitch": null, "output_rms_dbfs": -21.115939399734856, "output_peak": 1.1999095318718092, "reference_rms_dbfs": -22.307817294696616, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.679217519310725}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 38160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.0004033177834583336, "log_rms_mae_db": 1.7776605819564655, "rms_envelope_nrmse": 0.3390242953529821, "envelope_correlation": 0.9398906752645622, "rms_gain_error_db": 1.1918628931706685, "reference_mean_power": 0.005890133913079112, "output_mean_power": 0.007750174088481832}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 4.702783438048151e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019529842144495992, "output_power": 0.002657242661921666, "reference_active": true, "delta_db": 1.3373248341216823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0010897406087412391, "output_power": 0.0024791313332656406, "reference_active": true, "delta_db": 3.569763993460861}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00042203241052064084, "output_power": 0.0016190721407378225, "reference_active": true, "delta_db": 5.839203955295473}], "active_band_count": 3, "active_band_mae_db": 3.5820975942926725}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0002641139400842597, "log_rms_mae_db": 1.0762045933250857, "rms_envelope_nrmse": 0.2845855984792562, "envelope_correlation": 0.9579757603932175, "rms_gain_error_db": 1.1919259275987888, "reference_mean_power": 0.0059649962331930016, "output_mean_power": 0.007848791032609964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.959398727002631e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020291985866854774, "output_power": 0.0036370250889811782, "reference_active": true, "delta_db": 2.534217457576364}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005646977153887027, "output_power": 0.0015692759408524653, "reference_active": true, "delta_db": 4.4388328576477925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.714480412149099e-05, "output_power": 0.00015564325347684833, "reference_active": true, "delta_db": 6.22232227975318}], "active_band_count": 3, "active_band_mae_db": 4.398457531659112}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.00011878678262042868, "log_rms_mae_db": 1.9575973017263841, "rms_envelope_nrmse": 0.10574966933795428, "envelope_correlation": 0.9922349312098452, "rms_gain_error_db": -0.036555615851112847, "reference_mean_power": 0.0009978964081743123, "output_mean_power": 0.0009895321244849642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 1.0827926032603135e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006103304193321489, "output_power": 0.0005819606832428193, "reference_active": true, "delta_db": -0.206713709794118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011556410212235237, "output_power": 0.00010788312151656556, "reference_active": true, "delta_db": -0.29869445439189}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8931134571049757e-05, "output_power": 1.189385432508717e-05, "reference_active": true, "delta_db": -2.0185402766389067}], "active_band_count": 3, "active_band_mae_db": 0.8413161469416383}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 9.505044043492436e-05, "log_rms_mae_db": 1.0746825948827172, "rms_envelope_nrmse": 0.05190121900726328, "envelope_correlation": 0.9980376649846031, "rms_gain_error_db": -0.03655646327260649, "reference_mean_power": 0.0010106286473201321, "output_mean_power": 0.001002157447527074}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.735673127654544e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005396175923660179, "output_power": 0.0005162331037557651, "reference_active": true, "delta_db": -0.19240249407224624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.461345336739454e-05, "output_power": 6.145344326143115e-05, "reference_active": true, "delta_db": -0.21776731609732772}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3971113807678808e-06, "output_power": 1.1583326575829448e-06, "reference_active": true, "delta_db": -3.1585491296518264}], "active_band_count": 3, "active_band_mae_db": 1.1895729799404668}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 9.972124238081888e-05, "log_rms_mae_db": 1.6958960440911237, "rms_envelope_nrmse": 0.15743725572902037, "envelope_correlation": 0.9838495940152092, "rms_gain_error_db": -0.20851577067650395, "reference_mean_power": 0.00028254365881671625, "output_mean_power": 0.0002692985330705408}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 3.4051840450675846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002046408997048573, "output_power": 0.00019210239027367653, "reference_active": true, "delta_db": -0.2745966782276588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.521356834551269e-05, "output_power": 3.572973619677357e-05, "reference_active": true, "delta_db": -1.0223897408914966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018106896586635e-06, "output_power": 6.835886742029516e-06, "reference_active": true, "delta_db": -0.6927698326914845}], "active_band_count": 3, "active_band_mae_db": 0.66325208393688}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 8.26289228894579e-05, "log_rms_mae_db": 0.8734523847443234, "rms_envelope_nrmse": 0.08184185980158112, "envelope_correlation": 0.9956692794062182, "rms_gain_error_db": -0.2085184953504535, "reference_mean_power": 0.0002861487173387815, "output_mean_power": 0.0002727344219884181}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.083140770966499e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019461067978585587, "output_power": 0.00018099201034069352, "reference_active": true, "delta_db": -0.315072657456757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0239988981192927e-05, "output_power": 1.6756303910494264e-05, "reference_active": true, "delta_db": -0.8203204313835616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.640717194960459e-07, "output_power": 4.841679767111818e-07, "reference_active": true, "delta_db": -1.9813806404316257}], "active_band_count": 3, "active_band_mae_db": 1.0389245764239814}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 4.3426949519701344e-05, "log_rms_mae_db": 1.7291473741962433, "rms_envelope_nrmse": 0.16450796928844333, "envelope_correlation": 0.9827448361198929, "rms_gain_error_db": -0.22759020444758413, "reference_mean_power": 4.5529138246552346e-05, "output_mean_power": 4.32046418972819e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 6.179039674292901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3063171222465546e-05, "output_power": 3.213174426313722e-05, "reference_active": true, "delta_db": -0.12410204422181817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2207009907392542e-05, "output_power": 8.667129510366658e-06, "reference_active": true, "delta_db": -1.4873401090598177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2513877801441335e-06, "output_power": 9.277233584138754e-07, "reference_active": true, "delta_db": -3.8503181304660155}], "active_band_count": 3, "active_band_mae_db": 1.820586761249217}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 3.683581645514392e-05, "log_rms_mae_db": 0.9152921958546643, "rms_envelope_nrmse": 0.08602838981497982, "envelope_correlation": 0.9952786947690913, "rms_gain_error_db": -0.2275945866795791, "reference_mean_power": 4.610980557969544e-05, "output_mean_power": 4.375561902924427e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 5.621286439624602e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.226124623783574e-05, "output_power": 3.0177945759174076e-05, "reference_active": true, "delta_db": -0.2899146636638615}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.247687157348319e-06, "output_power": 4.971772730190723e-06, "reference_active": true, "delta_db": -1.6368817103533049}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.30977962238882e-07, "output_power": 8.22002996435681e-08, "reference_active": true, "delta_db": -4.486971448728098}], "active_band_count": 3, "active_band_mae_db": 2.137922607581755}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.805266642695944, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "76aa290cd7e1d13ab56258a1eb98bc09a9a946dbfffb48514793afff7145575a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9786030395235884, "standard_deviation": 0.011029364147432687, "minimum": 0.959385354385646, "maximum": 0.9890384647306153, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.759961470961512, "standard_deviation": 0.05717676049910522, "minimum": 1.67608630340755, "maximum": 1.8224076560988585, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1152674072478608, "standard_deviation": 0.6617208101498916, "minimum": 0.42882829085919033, "maximum": 2.463310558119989, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18143546483394918, "standard_deviation": 0.04623045408851365, "minimum": 0.13606872851001187, "maximum": 0.24865617832887268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930458972166044, "standard_deviation": 0.004734437806744854, "minimum": 0.9847426075624531, "maximum": 0.9973826666593462, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9346357357006875, "standard_deviation": 0.05466776096557385, "minimum": 0.8734523847443234, "maximum": 1.0114104560027466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4057760992693769, "standard_deviation": 0.4908675913869127, "minimum": 0.8688370897872731, "maximum": 2.4142504638499354, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10857301225269125, "standard_deviation": 0.03846892088121406, "minimum": 0.07747382406285235, "maximum": 0.18505331439197797, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9840475888547793, "standard_deviation": 0.0036745102598806224, "minimum": 0.9783364425968796, "maximum": 0.9890259384280415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7426009933785482, "standard_deviation": 0.039052861003576425, "minimum": 1.6836407840373364, "maximum": 1.8007396437365695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.09219193122743, "standard_deviation": 0.5286317814988455, "minimum": 0.5459462328272043, "maximum": 2.0032004294346546, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1579885951134546, "standard_deviation": 0.01912134530056368, "minimum": 0.13081721813642058, "maximum": 0.18799500370143388, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9951885223111201, "standard_deviation": 0.0015465631051875006, "minimum": 0.9923056669809958, "maximum": 0.9968622217680853, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9330430639641166, "standard_deviation": 0.03322365370920537, "minimum": 0.8781092761976733, "maximum": 0.985403168158431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4527833753815855, "standard_deviation": 0.5410954382246707, "minimum": 0.8972769331395489, "maximum": 2.423305526653835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08671771193713082, "standard_deviation": 0.015595056381289761, "minimum": 0.06925740152604322, "maximum": 0.11807142756781502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991950090262382, "standard_deviation": 0.0008431512323772672, "minimum": 0.9908451489468478, "maximum": 0.9930737502306957, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.032003469634357, "standard_deviation": 0.055802854641186096, "minimum": 1.9575973017263841, "maximum": 2.115129462777489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.127079331345359, "standard_deviation": 0.2699040641678264, "minimum": 0.8215815308694675, "maximum": 1.5520463240611881, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10797328963125115, "standard_deviation": 0.00534811001060419, "minimum": 0.10078982882770766, "maximum": 0.11465435717247581, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997318252984835, "standard_deviation": 0.0004343012282344978, "minimum": 0.9965454628661545, "maximum": 0.9980376649846031, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1214666276158127, "standard_deviation": 0.04322547909583438, "minimum": 1.0746825948827172, "maximum": 1.2183036320277938, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3356670870999756, "standard_deviation": 0.2623576115810702, "minimum": 0.9894046785690729, "maximum": 1.8050657249230035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06085019062606316, "standard_deviation": 0.005139108161878782, "minimum": 0.05190121900726328, "maximum": 0.06764225385524471, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4325020609880055, "standard_deviation": 0.13312503698536074, "minimum": 0.1640081855560245, "maximum": 0.5827912790292225, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6938637218218116, "standard_deviation": 0.020728290364453755, "minimum": 5.669779728971116, "maximum": 5.734026931395161, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9050364944873699, "standard_deviation": 0.32095008456619306, "minimum": 0.5629235393505584, "maximum": 1.362427125506974, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": 0.3524961255138524, "standard_deviation": 0.9227236874550206, "minimum": -0.9532718859460962, "maximum": 1.1918778949617566, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9420610127719428, "standard_deviation": 0.016514415096983686, "minimum": 0.9221938615317059, "maximum": 0.9754545380085238, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8828837612026645, "standard_deviation": 0.11336137117860093, "minimum": 1.7574724210158577, "maximum": 2.109878920014013, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 2.1085289862831544, "standard_deviation": 1.094087948202488, "minimum": 0.7119320582577898, "maximum": 3.5820975942926725, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.29386232794893713, "standard_deviation": 0.06300280921446193, "minimum": 0.176608706247627, "maximum": 0.3679618794526209, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9701055406918869, "standard_deviation": 0.013992082308425231, "minimum": 0.9545716627936963, "maximum": 0.9921003106070571, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.076359243587146, "standard_deviation": 0.09361467163540206, "minimum": 0.960391064372421, "maximum": 1.2714738400228982, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 2.5443945892838813, "standard_deviation": 1.2977090615060152, "minimum": 1.0518962985513955, "maximum": 4.398457531659112, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.21320656859760614, "standard_deviation": 0.07108295721249368, "minimum": 0.09870436434423871, "maximum": 0.2845855984792562, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0040", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.8816618919372559, "predicted_mos": 1.8816618919372559, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "b573eee06a9e453f00bde05cc7ffff6a7b245986787ccce4aa04ffb08c30083c", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.363163399975747, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6753037016060108, "全帯域_10ms_rms_envelope_nrmse": 0.27658400024625557, "全帯域_10ms_envelope_correlation": 0.9488593881794244, "全帯域_10ms_modulation_mae_db": 2.383343056974653, "全帯域_40ms_log_rms_mae_db": 1.0027857966031894, "全帯域_40ms_rms_envelope_nrmse": 0.22665537248424103, "全帯域_40ms_envelope_correlation": 0.9639764985384217, "全帯域_40ms_modulation_mae_db": 2.8501559493241815, "300–1000 Hz_10ms_log_rms_mae_db": 2.992967858528771, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16562135007882184, "300–1000 Hz_10ms_envelope_correlation": 0.9811974806629782, "300–1000 Hz_10ms_modulation_mae_db": 0.8600813611970354, "300–1000 Hz_40ms_log_rms_mae_db": 2.1674206061813024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10980445402385729, "300–1000 Hz_40ms_envelope_correlation": 0.9912959804956742, "300–1000 Hz_40ms_modulation_mae_db": 1.4731837699513086, "1000–3000 Hz_10ms_log_rms_mae_db": 3.2272622087359477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.32391071153366807, "1000–3000 Hz_10ms_envelope_correlation": 0.9317982884042615, "1000–3000 Hz_10ms_modulation_mae_db": 1.3563247409836834, "1000–3000 Hz_40ms_log_rms_mae_db": 2.2886482199263747, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.21736513371023555, "1000–3000 Hz_40ms_envelope_correlation": 0.9678418169725612, "1000–3000 Hz_40ms_modulation_mae_db": 0.9690541454622795, "3000–6000 Hz_10ms_log_rms_mae_db": 3.483830522811487, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3104410739966044, "3000–6000 Hz_10ms_envelope_correlation": 0.94202031924367, "3000–6000 Hz_10ms_modulation_mae_db": 1.2134323822389474, "3000–6000 Hz_40ms_log_rms_mae_db": 2.451174310134725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21679723161827794, "3000–6000 Hz_40ms_envelope_correlation": 0.9706249514018026, "3000–6000 Hz_40ms_modulation_mae_db": 1.415113526308818, "mr_stft_db": 6.583268672858405, "band_mae_db": 0.3824796363916804, "rms_gain_error_db": 0.6618903507716777, "output_peak": 1.1057984215355112}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 14, "at_or_above_full_scale_fraction": 0.0003668763102725367, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0003668763102725367, "longest_full_scale_run_samples": 10, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8206415767529094, "全帯域_10ms_rms_envelope_nrmse": 0.24394918399502752, "全帯域_10ms_envelope_correlation": 0.9526650236529217, "全帯域_10ms_modulation_mae_db": 2.0770660744049008, "全帯域_40ms_log_rms_mae_db": 0.979966249856139, "全帯域_40ms_rms_envelope_nrmse": 0.1862320103790565, "全帯域_40ms_envelope_correlation": 0.9736520658990517, "全帯域_40ms_modulation_mae_db": 2.77237885900738, "300–1000 Hz_10ms_log_rms_mae_db": 2.861007037088298, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1926451357607597, "300–1000 Hz_10ms_envelope_correlation": 0.9742795014166221, "300–1000 Hz_10ms_modulation_mae_db": 0.5967751312851393, "300–1000 Hz_40ms_log_rms_mae_db": 2.1237178765387648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1323521829720891, "300–1000 Hz_40ms_envelope_correlation": 0.9870535392954428, "300–1000 Hz_40ms_modulation_mae_db": 1.0126562326365096, "1000–3000 Hz_10ms_log_rms_mae_db": 3.300203520465368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.378792594891339, "1000–3000 Hz_10ms_envelope_correlation": 0.9110166630332568, "1000–3000 Hz_10ms_modulation_mae_db": 2.346698360234149, "1000–3000 Hz_40ms_log_rms_mae_db": 2.331336108154585, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.29430093389307266, "1000–3000 Hz_40ms_envelope_correlation": 0.9436010528255228, "1000–3000 Hz_40ms_modulation_mae_db": 1.7369587152731565, "3000–6000 Hz_10ms_log_rms_mae_db": 3.3774395645070183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30532280903345266, "3000–6000 Hz_10ms_envelope_correlation": 0.942817908214545, "3000–6000 Hz_10ms_modulation_mae_db": 1.0207598229525658, "3000–6000 Hz_40ms_log_rms_mae_db": 2.385176457198169, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2228180644269358, "3000–6000 Hz_40ms_envelope_correlation": 0.9683014622162182, "3000–6000 Hz_40ms_modulation_mae_db": 1.4682849228356682, "mr_stft_db": 6.567252092530494, "band_mae_db": 0.5946210959362851, "rms_gain_error_db": -0.8727173759793104, "output_peak": 0.545274557447291}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7011290538233979, "全帯域_10ms_rms_envelope_nrmse": 0.23398628856513673, "全帯域_10ms_envelope_correlation": 0.9578071387800204, "全帯域_10ms_modulation_mae_db": 2.5080422979171964, "全帯域_40ms_log_rms_mae_db": 1.0197945709526464, "全帯域_40ms_rms_envelope_nrmse": 0.18733294333658557, "全帯域_40ms_envelope_correlation": 0.9735538173627427, "全帯域_40ms_modulation_mae_db": 3.055752237843079, "300–1000 Hz_10ms_log_rms_mae_db": 2.8840818653815825, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17379877613872508, "300–1000 Hz_10ms_envelope_correlation": 0.9793077028388397, "300–1000 Hz_10ms_modulation_mae_db": 0.8561119051742653, "300–1000 Hz_40ms_log_rms_mae_db": 2.06193360631818, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11154641410797232, "300–1000 Hz_40ms_envelope_correlation": 0.9909945116862783, "300–1000 Hz_40ms_modulation_mae_db": 1.6360620564151105, "1000–3000 Hz_10ms_log_rms_mae_db": 3.186353914138735, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.3374459417937044, "1000–3000 Hz_10ms_envelope_correlation": 0.9249766031561965, "1000–3000 Hz_10ms_modulation_mae_db": 1.7265097589594376, "1000–3000 Hz_40ms_log_rms_mae_db": 2.1949598542928177, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.2483250319900989, "1000–3000 Hz_40ms_envelope_correlation": 0.9575574568837353, "1000–3000 Hz_40ms_modulation_mae_db": 1.0571547250338273, "3000–6000 Hz_10ms_log_rms_mae_db": 3.4686372714525717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3486534615413484, "3000–6000 Hz_10ms_envelope_correlation": 0.9251598146589527, "3000–6000 Hz_10ms_modulation_mae_db": 1.0094904285034045, "3000–6000 Hz_40ms_log_rms_mae_db": 2.475934520238657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2698722455181291, "3000–6000 Hz_40ms_envelope_correlation": 0.9527690155003748, "3000–6000 Hz_40ms_modulation_mae_db": 1.4332756080864943, "mr_stft_db": 6.585107722797612, "band_mae_db": 0.474553840483477, "rms_gain_error_db": -0.8790572779363498, "output_peak": 0.5814745127330854}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8029355391732662, "全帯域_10ms_rms_envelope_nrmse": 0.22302847147486848, "全帯域_10ms_envelope_correlation": 0.9574458846906478, "全帯域_10ms_modulation_mae_db": 0.5311368960093218, "全帯域_40ms_log_rms_mae_db": 1.0023190074900574, "全帯域_40ms_rms_envelope_nrmse": 0.1243127763469084, "全帯域_40ms_envelope_correlation": 0.9853054948382668, "全帯域_40ms_modulation_mae_db": 0.5972774005738799, "300–1000 Hz_10ms_log_rms_mae_db": 3.019123528270053, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16397839343822387, "300–1000 Hz_10ms_envelope_correlation": 0.9822869604371113, "300–1000 Hz_10ms_modulation_mae_db": 0.6295897703184621, "300–1000 Hz_40ms_log_rms_mae_db": 2.1182861628216036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1142472317776307, "300–1000 Hz_40ms_envelope_correlation": 0.9910175713908775, "300–1000 Hz_40ms_modulation_mae_db": 0.7953325063517808, "1000–3000 Hz_10ms_log_rms_mae_db": 3.3181016003478847, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.38826643153189994, "1000–3000 Hz_10ms_envelope_correlation": 0.9008175461265974, "1000–3000 Hz_10ms_modulation_mae_db": 1.6823819752098472, "1000–3000 Hz_40ms_log_rms_mae_db": 2.378564468839991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.30712800995455614, "1000–3000 Hz_40ms_envelope_correlation": 0.9340765581082698, "1000–3000 Hz_40ms_modulation_mae_db": 1.2206307019294547, "3000–6000 Hz_10ms_log_rms_mae_db": 3.417776995241256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3241325749527456, "3000–6000 Hz_10ms_envelope_correlation": 0.9390306400710384, "3000–6000 Hz_10ms_modulation_mae_db": 0.7146942253319178, "3000–6000 Hz_40ms_log_rms_mae_db": 2.4109728326888638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2397886058329043, "3000–6000 Hz_40ms_envelope_correlation": 0.9656490625145415, "3000–6000 Hz_40ms_modulation_mae_db": 0.9485831666683039, "mr_stft_db": 6.5711140170016415, "band_mae_db": 0.3802546687594716, "rms_gain_error_db": -0.12163965535277467, "output_peak": 0.7791441259875885}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.867068053744539, "全帯域_10ms_rms_envelope_nrmse": 0.2364007333056635, "全帯域_10ms_envelope_correlation": 0.9524927073575697, "全帯域_10ms_modulation_mae_db": 1.49765429465944, "全帯域_40ms_log_rms_mae_db": 0.9883923139328474, "全帯域_40ms_rms_envelope_nrmse": 0.13072706016208868, "全帯域_40ms_envelope_correlation": 0.9865729510926082, "全帯域_40ms_modulation_mae_db": 2.041669378805509, "300–1000 Hz_10ms_log_rms_mae_db": 2.8335179250278437, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1413201689101869, "300–1000 Hz_10ms_envelope_correlation": 0.9863721718647406, "300–1000 Hz_10ms_modulation_mae_db": 1.091788594623034, "300–1000 Hz_40ms_log_rms_mae_db": 1.9640742019917825, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08227569237934411, "300–1000 Hz_40ms_envelope_correlation": 0.9953093435888476, "300–1000 Hz_40ms_modulation_mae_db": 1.476731925382272, "1000–3000 Hz_10ms_log_rms_mae_db": 3.237257123819599, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.31238219023802044, "1000–3000 Hz_10ms_envelope_correlation": 0.9354147825339202, "1000–3000 Hz_10ms_modulation_mae_db": 1.7571230091507006, "1000–3000 Hz_40ms_log_rms_mae_db": 2.315366372236781, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.23742855207482574, "1000–3000 Hz_40ms_envelope_correlation": 0.9609561703718256, "1000–3000 Hz_40ms_modulation_mae_db": 1.3618341742699211, "3000–6000 Hz_10ms_log_rms_mae_db": 3.5023956045973303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3416850283927145, "3000–6000 Hz_10ms_envelope_correlation": 0.9304768252311906, "3000–6000 Hz_10ms_modulation_mae_db": 0.5273280093586354, "3000–6000 Hz_40ms_log_rms_mae_db": 2.5603179945199566, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24545270474524952, "3000–6000 Hz_40ms_envelope_correlation": 0.9629525192385665, "3000–6000 Hz_40ms_modulation_mae_db": 1.1387460045494249, "mr_stft_db": 6.584928422262467, "band_mae_db": 0.3561728955382913, "rms_gain_error_db": -0.5354673789882479, "output_peak": 0.5957186492280293}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.807586864610778, "全帯域_10ms_rms_envelope_nrmse": 0.37082741521843565, "全帯域_10ms_envelope_correlation": 0.9360248862237097, "全帯域_10ms_modulation_mae_db": 2.9996525254517206, "全帯域_40ms_log_rms_mae_db": 0.9534208870420829, "全帯域_40ms_rms_envelope_nrmse": 0.2839237869608102, "全帯域_40ms_envelope_correlation": 0.9698403975669349, "全帯域_40ms_modulation_mae_db": 3.4937335198692523, "300–1000 Hz_10ms_log_rms_mae_db": 2.9904905695419024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16108068695814032, "300–1000 Hz_10ms_envelope_correlation": 0.9822377814416593, "300–1000 Hz_10ms_modulation_mae_db": 1.1659857172992576, "300–1000 Hz_40ms_log_rms_mae_db": 2.1275646157585006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1060775788920664, "300–1000 Hz_40ms_envelope_correlation": 0.9917773503058405, "300–1000 Hz_40ms_modulation_mae_db": 1.4331245564648227, "1000–3000 Hz_10ms_log_rms_mae_db": 3.1663576089857637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.28426645898605885, "1000–3000 Hz_10ms_envelope_correlation": 0.949380618047731, "1000–3000 Hz_10ms_modulation_mae_db": 0.9505803406457479, "1000–3000 Hz_40ms_log_rms_mae_db": 2.2759458890696544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.20533661967783626, "1000–3000 Hz_40ms_envelope_correlation": 0.9725385238552501, "1000–3000 Hz_40ms_modulation_mae_db": 0.7658347127922375, "3000–6000 Hz_10ms_log_rms_mae_db": 3.4238361470141765, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3232777331561853, "3000–6000 Hz_10ms_envelope_correlation": 0.9361770543840303, "3000–6000 Hz_10ms_modulation_mae_db": 1.000225227879539, "3000–6000 Hz_40ms_log_rms_mae_db": 2.343473302192395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2304834516502386, "3000–6000 Hz_40ms_envelope_correlation": 0.9662975300638668, "3000–6000 Hz_40ms_modulation_mae_db": 1.3920418602816362, "mr_stft_db": 6.565896827371044, "band_mae_db": 0.5723287545667233, "rms_gain_error_db": 1.4402613751641873, "output_peak": 0.9654887649312298}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.729261279083814, "全帯域_10ms_rms_envelope_nrmse": 0.3438649015832254, "全帯域_10ms_envelope_correlation": 0.9491184926369348, "全帯域_10ms_modulation_mae_db": 3.0336616135558927, "全帯域_40ms_log_rms_mae_db": 1.074371968194257, "全帯域_40ms_rms_envelope_nrmse": 0.2915147533439686, "全帯域_40ms_envelope_correlation": 0.9675212184773455, "全帯域_40ms_modulation_mae_db": 3.204645189052206, "300–1000 Hz_10ms_log_rms_mae_db": 3.063455005096597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14779254414544568, "300–1000 Hz_10ms_envelope_correlation": 0.985132781152579, "300–1000 Hz_10ms_modulation_mae_db": 1.3606230625422755, "300–1000 Hz_40ms_log_rms_mae_db": 2.200556663242561, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08989259906927614, "300–1000 Hz_40ms_envelope_correlation": 0.9942050105549015, "300–1000 Hz_40ms_modulation_mae_db": 1.6626133082532497, "1000–3000 Hz_10ms_log_rms_mae_db": 3.222015239937031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.3593967118940464, "1000–3000 Hz_10ms_envelope_correlation": 0.9126396092941602, "1000–3000 Hz_10ms_modulation_mae_db": 2.149913566180397, "1000–3000 Hz_40ms_log_rms_mae_db": 2.316975777278782, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.2891947525396842, "1000–3000 Hz_40ms_envelope_correlation": 0.9413486872590214, "1000–3000 Hz_40ms_modulation_mae_db": 1.6016572256712565, "3000–6000 Hz_10ms_log_rms_mae_db": 3.3795759396215717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32642404021767496, "3000–6000 Hz_10ms_envelope_correlation": 0.9361541162450895, "3000–6000 Hz_10ms_modulation_mae_db": 0.755992850187929, "3000–6000 Hz_40ms_log_rms_mae_db": 2.4066988282667348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23261780770662382, "3000–6000 Hz_40ms_envelope_correlation": 0.9664556023264654, "3000–6000 Hz_40ms_modulation_mae_db": 1.2056651227961446, "mr_stft_db": 6.565067147332954, "band_mae_db": 0.615915303538846, "rms_gain_error_db": 1.4492916116435994, "output_peak": 1.0946127493423903}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 14, "at_or_above_full_scale_fraction": 0.0003668763102725367, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0003668763102725367, "longest_full_scale_run_samples": 8, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.643191292738607, "全帯域_10ms_rms_envelope_nrmse": 0.3307116252838244, "全帯域_10ms_envelope_correlation": 0.9290462571206308, "全帯域_10ms_modulation_mae_db": 2.571861819005277, "全帯域_40ms_log_rms_mae_db": 0.9716452021882304, "全帯域_40ms_rms_envelope_nrmse": 0.2382082642435009, "全帯域_40ms_envelope_correlation": 0.9637411588723508, "全帯域_40ms_modulation_mae_db": 3.3024404572423145, "300–1000 Hz_10ms_log_rms_mae_db": 3.026811636204841, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14414892830627557, "300–1000 Hz_10ms_envelope_correlation": 0.9860190276341373, "300–1000 Hz_10ms_modulation_mae_db": 1.044210911543714, "300–1000 Hz_40ms_log_rms_mae_db": 2.1646958946750883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09620197577709884, "300–1000 Hz_40ms_envelope_correlation": 0.9934305332078287, "300–1000 Hz_40ms_modulation_mae_db": 1.3420775333045736, "1000–3000 Hz_10ms_log_rms_mae_db": 3.152307549322841, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.32770276690577976, "1000–3000 Hz_10ms_envelope_correlation": 0.9301442184146027, "1000–3000 Hz_10ms_modulation_mae_db": 1.1801631037148148, "1000–3000 Hz_40ms_log_rms_mae_db": 2.178494288715339, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.24267756770523377, "1000–3000 Hz_40ms_envelope_correlation": 0.9595970121446974, "1000–3000 Hz_40ms_modulation_mae_db": 0.5541373713833823, "3000–6000 Hz_10ms_log_rms_mae_db": 3.3077631388433653, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3020924159775144, "3000–6000 Hz_10ms_envelope_correlation": 0.9480991137936353, "3000–6000 Hz_10ms_modulation_mae_db": 0.6893652443690365, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3406719870394603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2254049361116987, "3000–6000 Hz_40ms_envelope_correlation": 0.9704856419533373, "3000–6000 Hz_40ms_modulation_mae_db": 0.8893034023183564, "mr_stft_db": 6.541642251840343, "band_mae_db": 0.3945772974514426, "rms_gain_error_db": 0.7844977074467657, "output_peak": 1.2135415919753711}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 32, "at_or_above_full_scale_fraction": 0.0008385744234800838, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0008385744234800838, "longest_full_scale_run_samples": 26, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 38160, "output_samples": 38160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.9110077114297015, "spectral_convergence": 0.524455100913614, "spectral_convergence_status": "ok", "amplitude_floor": 5.674431450769592e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.52687615238628, "spectral_convergence": 0.581294141931473, "spectral_convergence_status": "ok", "amplitude_floor": 3.1139992269810527e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.311922154759231, "spectral_convergence": 0.6015054697287731, "spectral_convergence_status": "ok", "amplitude_floor": 1.28722240515293e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.003603182122246331, "output_power": 0.004322105496752204, "delta_db": 0.7900914919744737, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001490228689925631, "output_power": 0.0017259963986926167, "delta_db": 0.6378696502878249, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00043834980790852295, "output_power": 0.0004421685097120623, "delta_db": 0.03766989251330218, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023807008265260595, "output_power": 0.00024158480865751994, "delta_db": 0.06364798765987162, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.11014495688624e-05, "output_power": 8.18804513922669e-05, "delta_db": 0.041516115305668615, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.691475709405268e-05, "output_power": 3.1797921489684824e-05, "delta_db": 0.7240826806089415, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3824796363916804, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6618903507716777, "pitch": null, "output_rms_dbfs": -21.64592694392494, "output_peak": 1.1057984215355112, "reference_rms_dbfs": -22.307817294696616, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.583268672858405}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 38160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.0004033177834583336, "log_rms_mae_db": 1.6753037016060108, "rms_envelope_nrmse": 0.27658400024625557, "envelope_correlation": 0.9488593881794244, "rms_gain_error_db": 0.6619027791462624, "reference_mean_power": 0.005890133913079112, "output_mean_power": 0.006859863065899853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 4.702783438048151e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019529842144495992, "output_power": 0.002685266176506507, "reference_active": true, "delta_db": 1.382886085351094}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0010897406087412391, "output_power": 0.0017452841179878113, "reference_active": true, "delta_db": 2.0454300163591075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00042203241052064084, "output_power": 0.0009942992489455692, "reference_active": true, "delta_db": 3.7217130692137568}], "active_band_count": 3, "active_band_mae_db": 2.383343056974653}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0002641139400842597, "log_rms_mae_db": 1.0027857966031894, "rms_envelope_nrmse": 0.22665537248424103, "envelope_correlation": 0.9639764985384217, "rms_gain_error_db": 0.6619724321489627, "reference_mean_power": 0.0059649962331930016, "output_mean_power": 0.006947161851549356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.959398727002631e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020291985866854774, "output_power": 0.0031651755426975294, "reference_active": true, "delta_db": 1.9307325011206624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005646977153887027, "output_power": 0.0010048996457278768, "reference_active": true, "delta_db": 2.5030666241226363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.714480412149099e-05, "output_power": 9.584400315267334e-05, "reference_active": true, "delta_db": 4.116668722729246}], "active_band_count": 3, "active_band_mae_db": 2.8501559493241815}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.00011878678262042868, "log_rms_mae_db": 2.992967858528771, "rms_envelope_nrmse": 0.16562135007882184, "envelope_correlation": 0.9811974806629782, "rms_gain_error_db": -0.021826883460043654, "reference_mean_power": 0.0009978964081743123, "output_mean_power": 0.0009928937366195767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 1.0827926032603135e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006103304193321489, "output_power": 0.000588982578468568, "reference_active": true, "delta_db": -0.15462567094214533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011556410212235237, "output_power": 8.301776969702674e-05, "reference_active": true, "delta_db": -1.436518876664251}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8931134571049757e-05, "output_power": 1.5075325189145367e-05, "reference_active": true, "delta_db": -0.9890995359847097}], "active_band_count": 3, "active_band_mae_db": 0.8600813611970354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 9.505044043492436e-05, "log_rms_mae_db": 2.1674206061813024, "rms_envelope_nrmse": 0.10980445402385729, "envelope_correlation": 0.9912959804956742, "rms_gain_error_db": -0.021851441241122066, "reference_mean_power": 0.0010106286473201321, "output_mean_power": 0.0010055564602097068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.735673127654544e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005396175923660179, "output_power": 0.0005173856246385753, "reference_active": true, "delta_db": -0.18271742058487545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.461345336739454e-05, "output_power": 4.657916021669603e-05, "reference_active": true, "delta_db": -1.4213129890753653}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3971113807678808e-06, "output_power": 1.2535340168330536e-06, "reference_active": true, "delta_db": -2.8155209001936847}], "active_band_count": 3, "active_band_mae_db": 1.4731837699513086}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 9.972124238081888e-05, "log_rms_mae_db": 3.2272622087359477, "rms_envelope_nrmse": 0.32391071153366807, "envelope_correlation": 0.9317982884042615, "rms_gain_error_db": 0.04538597110520064, "reference_mean_power": 0.00028254365881671625, "output_mean_power": 0.00028551186565037854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 3.4051840450675846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002046408997048573, "output_power": 0.00018469846866519358, "reference_active": true, "delta_db": -0.44529141796806165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.521356834551269e-05, "output_power": 2.5924998233705807e-05, "reference_active": true, "delta_db": -2.4155004830642417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018106896586635e-06, "output_power": 1.0589856474715212e-05, "reference_active": true, "delta_db": 1.2081823219187473}], "active_band_count": 3, "active_band_mae_db": 1.3563247409836834}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 8.26289228894579e-05, "log_rms_mae_db": 2.2886482199263747, "rms_envelope_nrmse": 0.21736513371023555, "envelope_correlation": 0.9678418169725612, "rms_gain_error_db": 0.04536584797866189, "reference_mean_power": 0.0002861487173387815, "output_mean_power": 0.0002891534566039098}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.083140770966499e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019461067978585587, "output_power": 0.00017191143887078155, "reference_active": true, "delta_db": -0.5386189435777939}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0239988981192927e-05, "output_power": 1.1866207838821978e-05, "reference_active": true, "delta_db": -2.3189832091842253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.640717194960459e-07, "output_power": 7.554019422418966e-07, "reference_active": true, "delta_db": -0.049560283624819734}], "active_band_count": 3, "active_band_mae_db": 0.9690541454622795}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 4.3426949519701344e-05, "log_rms_mae_db": 3.483830522811487, "rms_envelope_nrmse": 0.3104410739966044, "envelope_correlation": 0.94202031924367, "rms_gain_error_db": 0.340519778833647, "reference_mean_power": 4.5529138246552346e-05, "output_mean_power": 4.9242649053052387e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 6.179039674292901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3063171222465546e-05, "output_power": 3.060752632202287e-05, "reference_active": true, "delta_db": -0.3351627452132594}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2207009907392542e-05, "output_power": 9.90729306259137e-06, "reference_active": true, "delta_db": -0.9065428722652438}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2513877801441335e-06, "output_power": 1.2959586704197986e-06, "reference_active": true, "delta_db": -2.398591529238339}], "active_band_count": 3, "active_band_mae_db": 1.2134323822389474}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 3.683581645514392e-05, "log_rms_mae_db": 2.451174310134725, "rms_envelope_nrmse": 0.21679723161827794, "envelope_correlation": 0.9706249514018026, "rms_gain_error_db": 0.3404758589834349, "reference_mean_power": 4.610980557969544e-05, "output_mean_power": 4.987017324609259e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 5.621286439624602e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.226124623783574e-05, "output_power": 2.8693509546198344e-05, "reference_active": true, "delta_db": -0.5089746927851079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.247687157348319e-06, "output_power": 5.285919996837348e-06, "reference_active": true, "delta_db": -1.3707885324186975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.30977962238882e-07, "output_power": 1.339716203793596e-07, "reference_active": true, "delta_db": -2.3655773537226485}], "active_band_count": 3, "active_band_mae_db": 1.415113526308818}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7557483589176163, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1146caa8009542edff5449ae3d6513d99a5fab620d6df3c1a69a6fb4c5fbe78a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9245235411263408, "standard_deviation": 0.015615241772770712, "minimum": 0.9008175461265974, "maximum": 0.949380618047731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 3.226232345719146, "standard_deviation": 0.05939244047254283, "minimum": 3.152307549322841, "maximum": 3.3181016003478847, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6437118568848472, "standard_deviation": 0.47049263759806653, "minimum": 0.9505803406457479, "maximum": 2.346698360234149, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3390204759718146, "standard_deviation": 0.03482333350212705, "minimum": 0.28426645898605885, "maximum": 0.38826643153189994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9546896598026104, "standard_deviation": 0.013569795358900271, "minimum": 0.9340765581082698, "maximum": 0.9725385238552501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 2.285036372314291, "standard_deviation": 0.06802508137093735, "minimum": 2.178494288715339, "maximum": 2.378564468839991, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1584077214769395, "standard_deviation": 0.40387561125595967, "minimum": 0.5541373713833823, "maximum": 1.7369587152731565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.25521957519319294, "standard_deviation": 0.037452400352730064, "minimum": 0.20533661967783626, "maximum": 0.30712800995455614, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.937491973980269, "standard_deviation": 0.007256167199736883, "minimum": 0.9251598146589527, "maximum": 0.9480991137936353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 3.420156898011097, "standard_deviation": 0.06468718645583971, "minimum": 3.3077631388433653, "maximum": 3.5023956045973303, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8664110238527469, "standard_deviation": 0.2280845679550093, "minimum": 0.5273280093586354, "maximum": 1.2134323822389474, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.32275364215853003, "standard_deviation": 0.016592448898852794, "minimum": 0.3020924159775144, "maximum": 0.3486534615413484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9654419731518966, "standard_deviation": 0.005720665300816175, "minimum": 0.9527690155003748, "maximum": 0.9706249514018026, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.42180252903487, "standard_deviation": 0.07307922710070115, "minimum": 2.3406719870394603, "maximum": 2.5603179945199566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.236376701730606, "standard_deviation": 0.227561308689885, "minimum": 0.8893034023183564, "maximum": 1.4682849228356682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23540438095125724, "standard_deviation": 0.016668339232401432, "minimum": 0.21679723161827794, "maximum": 0.2698722455181291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9821041759310833, "standard_deviation": 0.004011295705602504, "minimum": 0.9742795014166221, "maximum": 0.9863721718647406, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.958931928142486, "standard_deviation": 0.08638008854001417, "minimum": 2.8335179250278437, "maximum": 3.063455005096597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9506458067478978, "standard_deviation": 0.26418834017054965, "minimum": 0.5967751312851393, "maximum": 1.3606230625422755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16129824796707237, "standard_deviation": 0.017073108122244343, "minimum": 0.1413201689101869, "maximum": 0.1926451357607597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9918854800657113, "standard_deviation": 0.0025282460279168334, "minimum": 0.9870535392954428, "maximum": 0.9953093435888476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.116031203440973, "standard_deviation": 0.07403507463192743, "minimum": 1.9640742019917825, "maximum": 2.200556663242561, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3539727360949534, "standard_deviation": 0.3021581527484195, "minimum": 0.7953325063517808, "maximum": 1.6626133082532497, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10529976612491686, "standard_deviation": 0.015684691232473957, "minimum": 0.08227569237934411, "maximum": 0.1323521829720891, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4713629365832771, "standard_deviation": 0.10800659792011998, "minimum": 0.3561728955382913, "maximum": 0.615915303538846, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.570534644249371, "standard_deviation": 0.014535645713199366, "minimum": 6.541642251840343, "maximum": 6.585107722797612, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8601316716475621, "standard_deviation": 0.26848523775395844, "minimum": 0.545274557447291, "maximum": 1.2135415919753711, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": 0.2408824195961934, "standard_deviation": 0.9710111525903629, "minimum": -0.8790572779363498, "maximum": 1.4492916116435994, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9479324723302325, "standard_deviation": 0.010225684655718305, "minimum": 0.9290462571206308, "maximum": 0.9578071387800204, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.755889670191665, "standard_deviation": 0.07959017796836565, "minimum": 1.643191292738607, "maximum": 1.867068053744539, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 2.2003023222473, "standard_deviation": 0.8362097619789958, "minimum": 0.5311368960093218, "maximum": 3.0336616135558927, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.28241907745905465, "standard_deviation": 0.05785257323592633, "minimum": 0.22302847147486848, "maximum": 0.37082741521843565, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9730204503309652, "standard_deviation": 0.008811750390560483, "minimum": 0.9637411588723508, "maximum": 0.9865729510926082, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9990869995324312, "standard_deviation": 0.03672131034750201, "minimum": 0.9534208870420829, "maximum": 1.074371968194257, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 2.664756623964725, "standard_deviation": 0.9446741015899661, "minimum": 0.5972774005738799, "maximum": 3.4937335198692523, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.208613370907145, "standard_deviation": 0.06310118750294169, "minimum": 0.1243127763469084, "maximum": 0.2915147533439686, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0040", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.1104207038879395, "predicted_mos": 2.1104207038879395, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "4879b311cdcff4394955b3c0bf21eba4a92ceb6b0f834debf209b3f71fbadfe3", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.442251599859446, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8636218773380384, "全帯域_10ms_rms_envelope_nrmse": 0.21507637322330667, "全帯域_10ms_envelope_correlation": 0.9607337707915528, "全帯域_10ms_modulation_mae_db": 0.8279436973025112, "全帯域_40ms_log_rms_mae_db": 1.0031895490475207, "全帯域_40ms_rms_envelope_nrmse": 0.14288815743357894, "全帯域_40ms_envelope_correlation": 0.9805607139485858, "全帯域_40ms_modulation_mae_db": 1.2535225941171075, "300–1000 Hz_10ms_log_rms_mae_db": 2.1307050419353017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11391925636433733, "300–1000 Hz_10ms_envelope_correlation": 0.9912664752066866, "300–1000 Hz_10ms_modulation_mae_db": 0.23165365129829785, "300–1000 Hz_40ms_log_rms_mae_db": 1.313340840098285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06658371427912428, "300–1000 Hz_40ms_envelope_correlation": 0.9969251682906096, "300–1000 Hz_40ms_modulation_mae_db": 0.4570283297972672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6907111424883658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14560029415780032, "1000–3000 Hz_10ms_envelope_correlation": 0.9864022615478508, "1000–3000 Hz_10ms_modulation_mae_db": 0.8334766953152513, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0734126086694187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07460738419752645, "1000–3000 Hz_40ms_envelope_correlation": 0.9967351852219553, "1000–3000 Hz_40ms_modulation_mae_db": 1.2497685745946392, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5535798388021058, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16385069572469094, "3000–6000 Hz_10ms_envelope_correlation": 0.9827027693039911, "3000–6000 Hz_10ms_modulation_mae_db": 0.29245728153568756, "3000–6000 Hz_40ms_log_rms_mae_db": 0.932134650913769, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08220598429109172, "3000–6000 Hz_40ms_envelope_correlation": 0.995506018343497, "3000–6000 Hz_40ms_modulation_mae_db": 0.6341177461149714, "mr_stft_db": 5.623502857190917, "band_mae_db": 0.21534948521430644, "rms_gain_error_db": -0.007311175469634888, "output_peak": 0.7546082575336509}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9463138265161248, "全帯域_10ms_rms_envelope_nrmse": 0.2955853661355619, "全帯域_10ms_envelope_correlation": 0.9244212216581779, "全帯域_10ms_modulation_mae_db": 1.5434274240428454, "全帯域_40ms_log_rms_mae_db": 1.2146094989449987, "全帯域_40ms_rms_envelope_nrmse": 0.19617424659900137, "全帯域_40ms_envelope_correlation": 0.9668632115822717, "全帯域_40ms_modulation_mae_db": 1.8830305947492114, "300–1000 Hz_10ms_log_rms_mae_db": 2.0276762785125095, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11717195442543565, "300–1000 Hz_10ms_envelope_correlation": 0.9904595823104198, "300–1000 Hz_10ms_modulation_mae_db": 1.1847593642550824, "300–1000 Hz_40ms_log_rms_mae_db": 1.0946586336998985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06503264668549467, "300–1000 Hz_40ms_envelope_correlation": 0.9968544175443739, "300–1000 Hz_40ms_modulation_mae_db": 1.4077513530028567, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5053995567636336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17072811900616752, "1000–3000 Hz_10ms_envelope_correlation": 0.9819410478029849, "1000–3000 Hz_10ms_modulation_mae_db": 0.6786917402994638, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8733290992583587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09367967137071614, "1000–3000 Hz_40ms_envelope_correlation": 0.9957536563481199, "1000–3000 Hz_40ms_modulation_mae_db": 0.7059661047819578, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4717838108763608, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15037778132153493, "3000–6000 Hz_10ms_envelope_correlation": 0.9863860574658707, "3000–6000 Hz_10ms_modulation_mae_db": 1.3194490303595268, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9199741610370922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08480690740332798, "3000–6000 Hz_40ms_envelope_correlation": 0.996446608104673, "3000–6000 Hz_40ms_modulation_mae_db": 1.8540370350863304, "mr_stft_db": 5.6277176300371075, "band_mae_db": 0.37998939807767473, "rms_gain_error_db": -0.7241899337179171, "output_peak": 0.7137110346668019}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7597812991759212, "全帯域_10ms_rms_envelope_nrmse": 0.2514877396730804, "全帯域_10ms_envelope_correlation": 0.946002601308471, "全帯域_10ms_modulation_mae_db": 0.7448392760090501, "全帯域_40ms_log_rms_mae_db": 0.9753077565622702, "全帯域_40ms_rms_envelope_nrmse": 0.19053868935698226, "全帯域_40ms_envelope_correlation": 0.9671381635942079, "全帯域_40ms_modulation_mae_db": 0.6326978420661369, "300–1000 Hz_10ms_log_rms_mae_db": 1.9868864676007358, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11006313851767908, "300–1000 Hz_10ms_envelope_correlation": 0.9916174490831219, "300–1000 Hz_10ms_modulation_mae_db": 0.7795231203427212, "300–1000 Hz_40ms_log_rms_mae_db": 1.2025357678160777, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06665028125982432, "300–1000 Hz_40ms_envelope_correlation": 0.9967804267477234, "300–1000 Hz_40ms_modulation_mae_db": 0.8851762214057076, "1000–3000 Hz_10ms_log_rms_mae_db": 1.621454193573784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19408836759961692, "1000–3000 Hz_10ms_envelope_correlation": 0.9772277077001892, "1000–3000 Hz_10ms_modulation_mae_db": 0.4856440265925917, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9976461537211575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13863998437233868, "1000–3000 Hz_40ms_envelope_correlation": 0.9896150850335284, "1000–3000 Hz_40ms_modulation_mae_db": 0.42368715947546054, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5652234397806657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14248475091227059, "3000–6000 Hz_10ms_envelope_correlation": 0.9888181326215468, "3000–6000 Hz_10ms_modulation_mae_db": 0.5802170757404382, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9665510854091885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09599256094996675, "3000–6000 Hz_40ms_envelope_correlation": 0.9954705363381403, "3000–6000 Hz_40ms_modulation_mae_db": 0.30835542052841597, "mr_stft_db": 5.621550691389864, "band_mae_db": 0.5150972105091913, "rms_gain_error_db": -0.576514444102997, "output_peak": 0.7029266291894251}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7941362415881172, "全帯域_10ms_rms_envelope_nrmse": 0.27002928527086806, "全帯域_10ms_envelope_correlation": 0.9506387567174228, "全帯域_10ms_modulation_mae_db": 0.9187941353429753, "全帯域_40ms_log_rms_mae_db": 1.0762224080887255, "全帯域_40ms_rms_envelope_nrmse": 0.17203075740159002, "全帯域_40ms_envelope_correlation": 0.9818150856731197, "全帯域_40ms_modulation_mae_db": 1.0667915130243186, "300–1000 Hz_10ms_log_rms_mae_db": 2.00357309476811, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09327587345245937, "300–1000 Hz_10ms_envelope_correlation": 0.9940924316393472, "300–1000 Hz_10ms_modulation_mae_db": 0.9030649028329218, "300–1000 Hz_40ms_log_rms_mae_db": 1.3242687381197478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056769231619916226, "300–1000 Hz_40ms_envelope_correlation": 0.9977319722923719, "300–1000 Hz_40ms_modulation_mae_db": 0.6098767209451, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5603745877600774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12480624196940598, "1000–3000 Hz_10ms_envelope_correlation": 0.9908866293347649, "1000–3000 Hz_10ms_modulation_mae_db": 0.8452754172500391, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0091612006690718, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07969052007512319, "1000–3000 Hz_40ms_envelope_correlation": 0.9969830879888412, "1000–3000 Hz_40ms_modulation_mae_db": 0.9040311224745826, "3000–6000 Hz_10ms_log_rms_mae_db": 1.546689657339077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16332747065267297, "3000–6000 Hz_10ms_envelope_correlation": 0.9855409303716036, "3000–6000 Hz_10ms_modulation_mae_db": 1.1721329199596056, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9589484004189388, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10667055633545819, "3000–6000 Hz_40ms_envelope_correlation": 0.9946634092462798, "3000–6000 Hz_40ms_modulation_mae_db": 1.2159643408791256, "mr_stft_db": 5.651122363908409, "band_mae_db": 0.48027865610368137, "rms_gain_error_db": 0.607871544313211, "output_peak": 0.8360974096191549}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.831280985146957, "全帯域_10ms_rms_envelope_nrmse": 0.2085381431389324, "全帯域_10ms_envelope_correlation": 0.9629913251980098, "全帯域_10ms_modulation_mae_db": 0.6128972446532255, "全帯域_40ms_log_rms_mae_db": 1.1291941146215356, "全帯域_40ms_rms_envelope_nrmse": 0.09908211979151565, "全帯域_40ms_envelope_correlation": 0.9921276858673458, "全帯域_40ms_modulation_mae_db": 0.6037387536579797, "300–1000 Hz_10ms_log_rms_mae_db": 1.9408600415123252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0955486762600936, "300–1000 Hz_10ms_envelope_correlation": 0.9938427030518745, "300–1000 Hz_10ms_modulation_mae_db": 0.7603268566280343, "300–1000 Hz_40ms_log_rms_mae_db": 1.245411846737451, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05644043014121352, "300–1000 Hz_40ms_envelope_correlation": 0.9978580137599709, "300–1000 Hz_40ms_modulation_mae_db": 1.0347442687874413, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5641659140346091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15665046541636374, "1000–3000 Hz_10ms_envelope_correlation": 0.9839049773737647, "1000–3000 Hz_10ms_modulation_mae_db": 0.5009816294784674, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9382851090402143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07104565271437578, "1000–3000 Hz_40ms_envelope_correlation": 0.9966884885843748, "1000–3000 Hz_40ms_modulation_mae_db": 0.6475213708827986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.511043584464243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11174348429469094, "3000–6000 Hz_10ms_envelope_correlation": 0.9924289701488375, "3000–6000 Hz_10ms_modulation_mae_db": 0.572102632336323, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8872448314026329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07101153671118274, "3000–6000 Hz_40ms_envelope_correlation": 0.9969968049756494, "3000–6000 Hz_40ms_modulation_mae_db": 0.7400305127303292, "mr_stft_db": 5.625879972334047, "band_mae_db": 0.302972275106415, "rms_gain_error_db": -0.3831854027302306, "output_peak": 0.7858866072667626}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9013787490291976, "全帯域_10ms_rms_envelope_nrmse": 0.2525433147753097, "全帯域_10ms_envelope_correlation": 0.9504666582232628, "全帯域_10ms_modulation_mae_db": 1.3774883035395071, "全帯域_40ms_log_rms_mae_db": 1.102794111686386, "全帯域_40ms_rms_envelope_nrmse": 0.15063582316101662, "全帯域_40ms_envelope_correlation": 0.981251528783448, "全帯域_40ms_modulation_mae_db": 1.2604854809683375, "300–1000 Hz_10ms_log_rms_mae_db": 2.183205176725324, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11289084062673478, "300–1000 Hz_10ms_envelope_correlation": 0.9912636516392167, "300–1000 Hz_10ms_modulation_mae_db": 1.226874715226847, "300–1000 Hz_40ms_log_rms_mae_db": 1.479370480316128, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07077974083118074, "300–1000 Hz_40ms_envelope_correlation": 0.9964590399651198, "300–1000 Hz_40ms_modulation_mae_db": 1.5924680686754116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5932858917145603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1435135002250816, "1000–3000 Hz_10ms_envelope_correlation": 0.9868875004880345, "1000–3000 Hz_10ms_modulation_mae_db": 1.3398512622104795, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9224393639193738, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07292209752521644, "1000–3000 Hz_40ms_envelope_correlation": 0.9969968072469505, "1000–3000 Hz_40ms_modulation_mae_db": 1.493950093995274, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5159643013874133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14657003833545448, "3000–6000 Hz_10ms_envelope_correlation": 0.9880602293276425, "3000–6000 Hz_10ms_modulation_mae_db": 1.7740461913913508, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9234958883999264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0944401690298199, "3000–6000 Hz_40ms_envelope_correlation": 0.9962480308934458, "3000–6000 Hz_40ms_modulation_mae_db": 1.9972567214393155, "mr_stft_db": 5.614119770616921, "band_mae_db": 0.19714792012145174, "rms_gain_error_db": 0.29891583547482176, "output_peak": 0.7163771210421944}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7971272182883722, "全帯域_10ms_rms_envelope_nrmse": 0.36369174652973285, "全帯域_10ms_envelope_correlation": 0.9199078179593553, "全帯域_10ms_modulation_mae_db": 1.4790940763457192, "全帯域_40ms_log_rms_mae_db": 0.8911011789084663, "全帯域_40ms_rms_envelope_nrmse": 0.29149614635745036, "全帯域_40ms_envelope_correlation": 0.9485792397833317, "全帯域_40ms_modulation_mae_db": 1.4964260089353607, "300–1000 Hz_10ms_log_rms_mae_db": 1.849136627184853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08703552314647596, "300–1000 Hz_10ms_envelope_correlation": 0.994742810955732, "300–1000 Hz_10ms_modulation_mae_db": 0.8695309888839519, "300–1000 Hz_40ms_log_rms_mae_db": 0.9839575525611634, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04653166233637241, "300–1000 Hz_40ms_envelope_correlation": 0.9983927967304514, "300–1000 Hz_40ms_modulation_mae_db": 0.9622456000020122, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5854596261696436, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14272060407295273, "1000–3000 Hz_10ms_envelope_correlation": 0.9869245790434078, "1000–3000 Hz_10ms_modulation_mae_db": 0.6052547041127937, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9323458923226255, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07775322625137708, "1000–3000 Hz_40ms_envelope_correlation": 0.9963276253125575, "1000–3000 Hz_40ms_modulation_mae_db": 0.9850101994475455, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4376224996954003, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1502855742246112, "3000–6000 Hz_10ms_envelope_correlation": 0.9886701432933712, "3000–6000 Hz_10ms_modulation_mae_db": 1.2672873458203477, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8940845021642932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10382865427155548, "3000–6000 Hz_40ms_envelope_correlation": 0.9965622858400163, "3000–6000 Hz_40ms_modulation_mae_db": 1.4739563078194526, "mr_stft_db": 5.601848477975199, "band_mae_db": 0.49077336032119984, "rms_gain_error_db": 0.9989972519108686, "output_peak": 1.156377047639001}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 28, "at_or_above_full_scale_fraction": 0.0007337526205450734, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.000759958071278826, "longest_full_scale_run_samples": 28, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8582918829532091, "全帯域_10ms_rms_envelope_nrmse": 0.21215487913062267, "全帯域_10ms_envelope_correlation": 0.9615071944786017, "全帯域_10ms_modulation_mae_db": 0.727091198276026, "全帯域_40ms_log_rms_mae_db": 1.1887697043778143, "全帯域_40ms_rms_envelope_nrmse": 0.15794175985604794, "全帯域_40ms_envelope_correlation": 0.9766926428008182, "全帯域_40ms_modulation_mae_db": 0.7861253281222716, "300–1000 Hz_10ms_log_rms_mae_db": 2.118200011941246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1151872902649506, "300–1000 Hz_10ms_envelope_correlation": 0.9908312880219619, "300–1000 Hz_10ms_modulation_mae_db": 0.6312951838534793, "300–1000 Hz_40ms_log_rms_mae_db": 1.4501644973560268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07797036523462149, "300–1000 Hz_40ms_envelope_correlation": 0.9955431964327852, "300–1000 Hz_40ms_modulation_mae_db": 0.9217861052775124, "1000–3000 Hz_10ms_log_rms_mae_db": 1.59907212419962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15519647088574617, "1000–3000 Hz_10ms_envelope_correlation": 0.9846384397168081, "1000–3000 Hz_10ms_modulation_mae_db": 0.7220108879390942, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9665967104463854, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08014577828711263, "1000–3000 Hz_40ms_envelope_correlation": 0.996427929086603, "1000–3000 Hz_40ms_modulation_mae_db": 0.8247799885247437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.584397318337991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14450945408721202, "3000–6000 Hz_10ms_envelope_correlation": 0.9866373954926053, "3000–6000 Hz_10ms_modulation_mae_db": 0.5087851324886857, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9831108976791637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08475444159452208, "3000–6000 Hz_40ms_envelope_correlation": 0.9951114680806012, "3000–6000 Hz_40ms_modulation_mae_db": 0.924992985326699, "mr_stft_db": 5.625057065055613, "band_mae_db": 0.4248676629167199, "rms_gain_error_db": -0.3319993470384752, "output_peak": 0.76748160820911}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 38160, "output_samples": 38160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.751957630376618, "spectral_convergence": 0.5022634835911428, "spectral_convergence_status": "ok", "amplitude_floor": 5.674431450769592e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4014239718545145, "spectral_convergence": 0.48111767965957347, "spectral_convergence_status": "ok", "amplitude_floor": 3.1139992269810527e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.717126969341616, "spectral_convergence": 0.516704208654161, "spectral_convergence_status": "ok", "amplitude_floor": 1.28722240515293e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.003603182122246331, "output_power": 0.0035341931525537896, "delta_db": -0.08395932914118251, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001490228689925631, "output_power": 0.0015528238874963811, "delta_db": 0.17869283105487357, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00043834980790852295, "output_power": 0.00045438131405341306, "delta_db": 0.15599642597108201, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023807008265260595, "output_power": 0.00022686964750188583, "delta_db": -0.20928426461242502, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.11014495688624e-05, "output_power": 7.189143184444014e-05, "delta_db": -0.523514832172969, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.691475709405268e-05, "output_power": 2.7800677994594154e-05, "delta_db": 0.14064922833330673, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21534948521430644, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.007311175469634888, "pitch": null, "output_rms_dbfs": -22.31512847016625, "output_peak": 0.7546082575336509, "reference_rms_dbfs": -22.307817294696616, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.623502857190917}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 38160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.0004033177834583336, "log_rms_mae_db": 1.8636218773380384, "rms_envelope_nrmse": 0.21507637322330667, "envelope_correlation": 0.9607337707915528, "rms_gain_error_db": -0.007312967512550738, "reference_mean_power": 0.005890133913079112, "output_mean_power": 0.005880224021469317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 4.702783438048151e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019529842144495992, "output_power": 0.0017335787719415458, "reference_active": true, "delta_db": -0.5175515268177232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0010897406087412391, "output_power": 0.0008924174603386475, "reference_active": true, "delta_db": -0.8675507624671008}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00042203241052064084, "output_power": 0.0003276973460823207, "reference_active": true, "delta_db": -1.0987288026227096}], "active_band_count": 3, "active_band_mae_db": 0.8279436973025112}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0002641139400842597, "log_rms_mae_db": 1.0031895490475207, "rms_envelope_nrmse": 0.14288815743357894, "envelope_correlation": 0.9805607139485858, "rms_gain_error_db": -0.007334324453157482, "reference_mean_power": 0.0059649962331930016, "output_mean_power": 0.005954931104873353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.959398727002631e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020291985866854774, "output_power": 0.0016537419727862177, "reference_active": true, "delta_db": -0.8885680201581209}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005646977153887027, "output_power": 0.00048826919514109355, "reference_active": true, "delta_db": -0.6315670515544696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.714480412149099e-05, "output_power": 2.2174549295687157e-05, "reference_active": true, "delta_db": -2.240432710638732}], "active_band_count": 3, "active_band_mae_db": 1.2535225941171075}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.00011878678262042868, "log_rms_mae_db": 2.1307050419353017, "rms_envelope_nrmse": 0.11391925636433733, "envelope_correlation": 0.9912664752066866, "rms_gain_error_db": 0.09169923473328409, "reference_mean_power": 0.0009978964081743123, "output_mean_power": 0.0010191905375618847}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 1.0827926032603135e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006103304193321489, "output_power": 0.0006104497703016249, "reference_active": true, "delta_db": 0.0008491859232072292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011556410212235237, "output_power": 0.0001046910133537936, "reference_active": true, "delta_db": -0.4291354581787329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8931134571049757e-05, "output_power": 1.781061892802806e-05, "reference_active": true, "delta_db": -0.2649763097929535}], "active_band_count": 3, "active_band_mae_db": 0.23165365129829785}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 9.505044043492436e-05, "log_rms_mae_db": 1.313340840098285, "rms_envelope_nrmse": 0.06658371427912428, "envelope_correlation": 0.9969251682906096, "rms_gain_error_db": 0.09169675489705284, "reference_mean_power": 0.0010106286473201321, "output_mean_power": 0.0010321938808012753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.735673127654544e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005396175923660179, "output_power": 0.0005380283959761039, "reference_active": true, "delta_db": -0.012809025052423933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.461345336739454e-05, "output_power": 6.0878387342855546e-05, "reference_active": true, "delta_db": -0.2585981367554547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3971113807678808e-06, "output_power": 1.860888838644233e-06, "reference_active": true, "delta_db": -1.099677827583923}], "active_band_count": 3, "active_band_mae_db": 0.4570283297972672}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 9.972124238081888e-05, "log_rms_mae_db": 1.6907111424883658, "rms_envelope_nrmse": 0.14560029415780032, "envelope_correlation": 0.9864022615478508, "rms_gain_error_db": -0.27739337011920057, "reference_mean_power": 0.00028254365881671625, "output_mean_power": 0.00026506124148668403}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 3.4051840450675846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002046408997048573, "output_power": 0.00018427232910529488, "reference_active": true, "delta_db": -0.4553231138524339}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.521356834551269e-05, "output_power": 4.101909812227932e-05, "reference_active": true, "delta_db": -0.4228267637677491}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018106896586635e-06, "output_power": 5.51878939519938e-06, "reference_active": true, "delta_db": -1.6222802083255707}], "active_band_count": 3, "active_band_mae_db": 0.8334766953152513}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 8.26289228894579e-05, "log_rms_mae_db": 1.0734126086694187, "rms_envelope_nrmse": 0.07460738419752645, "envelope_correlation": 0.9967351852219553, "rms_gain_error_db": -0.2774033598985828, "reference_mean_power": 0.0002861487173387815, "output_mean_power": 0.00026844261916887777}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.083140770966499e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019461067978585587, "output_power": 0.00017523428791244396, "reference_active": true, "delta_db": -0.45547581566732676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0239988981192927e-05, "output_power": 1.85137882684469e-05, "reference_active": true, "delta_db": -0.38714979089192636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.640717194960459e-07, "output_power": 3.9126059745879965e-07, "reference_active": true, "delta_db": -2.9066801172246644}], "active_band_count": 3, "active_band_mae_db": 1.2497685745946392}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 4.3426949519701344e-05, "log_rms_mae_db": 1.5535798388021058, "rms_envelope_nrmse": 0.16385069572469094, "envelope_correlation": 0.9827027693039911, "rms_gain_error_db": -0.154131383290171, "reference_mean_power": 4.5529138246552346e-05, "output_mean_power": 4.394164305724297e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 6.179039674292901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3063171222465546e-05, "output_power": 3.167679079391059e-05, "reference_active": true, "delta_db": -0.1860332981818893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2207009907392542e-05, "output_power": 1.0877506174813968e-05, "reference_active": true, "delta_db": -0.5007995855303753}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2513877801441335e-06, "output_power": 2.352362246045925e-06, "reference_active": true, "delta_db": 0.190538960894798}], "active_band_count": 3, "active_band_mae_db": 0.29245728153568756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 3.683581645514392e-05, "log_rms_mae_db": 0.932134650913769, "rms_envelope_nrmse": 0.08220598429109172, "envelope_correlation": 0.995506018343497, "rms_gain_error_db": -0.15415130297101967, "reference_mean_power": 4.610980557969544e-05, "output_mean_power": 4.4501859754527737e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 5.621286439624602e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.226124623783574e-05, "output_power": 3.110882596121463e-05, "reference_active": true, "delta_db": -0.15797518722735313}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.247687157348319e-06, "output_power": 6.410847996454441e-06, "reference_active": true, "delta_db": -0.5328395925244047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.30977962238882e-07, "output_power": 1.7474980495447177e-07, "reference_active": true, "delta_db": -1.2115384585931563}], "active_band_count": 3, "active_band_mae_db": 0.6341177461149714}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8124342832032336, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b28a4c9bd6e0611f6acbde20155bd8ea6889e517c1e5362bce058c4beb332722", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9848516428759756, "standard_deviation": 0.004049776762768743, "minimum": 0.9772277077001892, "maximum": 0.9908866293347649, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5899903795880368, "standard_deviation": 0.05338286955634088, "minimum": 1.5053995567636336, "maximum": 1.6907111424883658, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7513982953997725, "standard_deviation": 0.2732524531308913, "minimum": 0.4856440265925917, "maximum": 1.3398512622104795, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1541630079166419, "standard_deviation": 0.020891464885163046, "minimum": 0.12480624196940598, "maximum": 0.19408836759961692, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9956909831028664, "standard_deviation": 0.002487941057912978, "minimum": 0.9896150850335284, "maximum": 0.9969968072469505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9641520172558258, "standard_deviation": 0.061830426554953404, "minimum": 0.8733290992583587, "maximum": 1.0734126086694187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9043393267721253, "standard_deviation": 0.3414535523033582, "minimum": 0.42368715947546054, "maximum": 1.493950093995274, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08606053934922331, "standard_deviation": 0.022349376609166915, "minimum": 0.07104565271437578, "maximum": 0.13863998437233868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9874055785031837, "standard_deviation": 0.002840531134321675, "minimum": 0.9827027693039911, "maximum": 0.9924289701488375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.523288056335407, "standard_deviation": 0.04950178160493093, "minimum": 1.4376224996954003, "maximum": 1.584397318337991, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9358097012039956, "standard_deviation": 0.5169314553024664, "minimum": 0.29245728153568756, "maximum": 1.7740461913913508, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14664365619414227, "standard_deviation": 0.016219555488348882, "minimum": 0.11174348429469094, "maximum": 0.16385069572469094, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9958756452277878, "standard_deviation": 0.0008059922220966289, "minimum": 0.9946634092462798, "maximum": 0.9969968049756494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9331930521781255, "standard_deviation": 0.03418123880081779, "minimum": 0.8872448314026329, "maximum": 0.9831108976791637, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.14358888374058, "standard_deviation": 0.5994861488116406, "minimum": 0.30835542052841597, "maximum": 1.9972567214393155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09046385132336561, "standard_deviation": 0.011945329412973738, "minimum": 0.07101153671118274, "maximum": 0.10667055633545819, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922645489885451, "standard_deviation": 0.0016776826868251655, "minimum": 0.9904595823104198, "maximum": 0.994742810955732, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0300303425225508, "standard_deviation": 0.10994324242899126, "minimum": 1.849136627184853, "maximum": 2.183205176725324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.823378597915167, "standard_deviation": 0.3150551880789831, "minimum": 0.23165365129829785, "maximum": 1.226874715226847, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10563656913227079, "standard_deviation": 0.011745239517685038, "minimum": 0.08703552314647596, "maximum": 0.11717195442543565, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970681289704257, "standard_deviation": 0.0008995746749424437, "minimum": 0.9955431964327852, "maximum": 0.9983927967304514, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2617135445880971, "standard_deviation": 0.16830682955515586, "minimum": 0.9839575525611634, "maximum": 1.479370480316128, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9838845834866636, "standard_deviation": 0.37495729375392145, "minimum": 0.4570283297972672, "maximum": 1.5924680686754116, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06334475904846845, "standard_deviation": 0.009761741083334182, "minimum": 0.04653166233637241, "maximum": 0.07797036523462149, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.37580949604633007, "standard_deviation": 0.12474452335673074, "minimum": 0.19714792012145174, "maximum": 0.5150972105091913, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6238498535635095, "standard_deviation": 0.013879429034572984, "minimum": 5.601848477975199, "maximum": 5.651122363908409, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8041832143957626, "standard_deviation": 0.14897098638464057, "minimum": 0.7029266291894251, "maximum": 1.156377047639001, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.014676958920044184, "standard_deviation": 0.6058121741416559, "minimum": -0.7241899337179171, "maximum": 0.9989972519108686, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9470836682918568, "standard_deviation": 0.01658218606683872, "minimum": 0.9199078179593553, "maximum": 0.9629913251980098, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8439915100044924, "standard_deviation": 0.06128123731054175, "minimum": 1.7597812991759212, "maximum": 1.9463138265161248, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0289469194389826, "standard_deviation": 0.3753609770067381, "minimum": 0.6128972446532255, "maximum": 1.5434274240428454, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2586383559846768, "standard_deviation": 0.0523555452443284, "minimum": 0.2085381431389324, "maximum": 0.36369174652973285, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9743785340041411, "standard_deviation": 0.013286157217639299, "minimum": 0.9485792397833317, "maximum": 0.9921276858673458, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0726485402797148, "standard_deviation": 0.1102044804504301, "minimum": 0.8911011789084663, "maximum": 1.2146094989449987, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1228522644550907, "standard_deviation": 0.44396011824393, "minimum": 0.6037387536579797, "maximum": 1.8830305947492114, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1750984624946479, "standard_deviation": 0.05599640573034307, "minimum": 0.09908211979151565, "maximum": 0.29149614635745036, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0040", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.133728504180908, "predicted_mos": 2.133728504180908, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "554992762eb0282cd280385e79f8afa70c0916c2ac1119002727fa0b47b74b1c", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.5943991001695395, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.70907842462923, "全帯域_10ms_rms_envelope_nrmse": 0.3567066219884458, "全帯域_10ms_envelope_correlation": 0.9306641149322759, "全帯域_10ms_modulation_mae_db": 3.5740162986930586, "全帯域_40ms_log_rms_mae_db": 0.9887263326612729, "全帯域_40ms_rms_envelope_nrmse": 0.3002398554525687, "全帯域_40ms_envelope_correlation": 0.9502605712575078, "全帯域_40ms_modulation_mae_db": 4.260355495789128, "300–1000 Hz_10ms_log_rms_mae_db": 2.3565554949130947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12130208450080439, "300–1000 Hz_10ms_envelope_correlation": 0.9899102774156417, "300–1000 Hz_10ms_modulation_mae_db": 1.003630124031931, "300–1000 Hz_40ms_log_rms_mae_db": 1.6429676962187807, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06674590601999912, "300–1000 Hz_40ms_envelope_correlation": 0.9968682454573164, "300–1000 Hz_40ms_modulation_mae_db": 1.3291228611659542, "1000–3000 Hz_10ms_log_rms_mae_db": 2.27847610687973, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20559883348893623, "1000–3000 Hz_10ms_envelope_correlation": 0.9726094811455679, "1000–3000 Hz_10ms_modulation_mae_db": 0.7121184539097953, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6625700130235077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12715858669163407, "1000–3000 Hz_40ms_envelope_correlation": 0.9891575264543184, "1000–3000 Hz_40ms_modulation_mae_db": 0.6108517195365429, "3000–6000 Hz_10ms_log_rms_mae_db": 2.258361635192704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19157129516932456, "3000–6000 Hz_10ms_envelope_correlation": 0.9772384231489453, "3000–6000 Hz_10ms_modulation_mae_db": 1.8587040323572452, "3000–6000 Hz_40ms_log_rms_mae_db": 1.594439975053868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11929898191502068, "3000–6000 Hz_40ms_envelope_correlation": 0.9907442569445308, "3000–6000 Hz_40ms_modulation_mae_db": 1.9685382724699927, "mr_stft_db": 5.944842621753686, "band_mae_db": 0.44201811953417186, "rms_gain_error_db": 1.170497946204621, "output_peak": 1.295377598419888}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 36, "at_or_above_full_scale_fraction": 0.0009433962264150943, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0009433962264150943, "longest_full_scale_run_samples": 36, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6196414162514108, "全帯域_10ms_rms_envelope_nrmse": 0.2417337690061246, "全帯域_10ms_envelope_correlation": 0.9664296296548734, "全帯域_10ms_modulation_mae_db": 2.285739053413018, "全帯域_40ms_log_rms_mae_db": 0.999828170780426, "全帯域_40ms_rms_envelope_nrmse": 0.18866458798050617, "全帯域_40ms_envelope_correlation": 0.9813292803663863, "全帯域_40ms_modulation_mae_db": 2.7154035603211137, "300–1000 Hz_10ms_log_rms_mae_db": 2.4784406090440725, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11688285237610033, "300–1000 Hz_10ms_envelope_correlation": 0.9908277489246051, "300–1000 Hz_10ms_modulation_mae_db": 1.071819064927696, "300–1000 Hz_40ms_log_rms_mae_db": 1.7473015140326675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06953499813419958, "300–1000 Hz_40ms_envelope_correlation": 0.9966439851705191, "300–1000 Hz_40ms_modulation_mae_db": 1.182936935365633, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2378701996443855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25963014901855475, "1000–3000 Hz_10ms_envelope_correlation": 0.9557151570244895, "1000–3000 Hz_10ms_modulation_mae_db": 0.4490827008398773, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6065679762296696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14728864050791207, "1000–3000 Hz_40ms_envelope_correlation": 0.984839606262649, "1000–3000 Hz_40ms_modulation_mae_db": 0.934609540656953, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2412250595785395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18259090809542822, "3000–6000 Hz_10ms_envelope_correlation": 0.9808297212406057, "3000–6000 Hz_10ms_modulation_mae_db": 0.5847744803475536, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5787661300764704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10400247866230056, "3000–6000 Hz_40ms_envelope_correlation": 0.994233876223959, "3000–6000 Hz_40ms_modulation_mae_db": 0.6624342078563518, "mr_stft_db": 5.961746792512585, "band_mae_db": 0.3804309691346925, "rms_gain_error_db": 0.8404474322262565, "output_peak": 0.987264112646472}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8150007370231198, "全帯域_10ms_rms_envelope_nrmse": 0.16595216083694664, "全帯域_10ms_envelope_correlation": 0.9795399657181857, "全帯域_10ms_modulation_mae_db": 0.5972884112246559, "全帯域_40ms_log_rms_mae_db": 0.923364553727624, "全帯域_40ms_rms_envelope_nrmse": 0.09108057787564183, "全帯域_40ms_envelope_correlation": 0.9943954290455622, "全帯域_40ms_modulation_mae_db": 0.8654778783386928, "300–1000 Hz_10ms_log_rms_mae_db": 2.428407504941166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12655958173952775, "300–1000 Hz_10ms_envelope_correlation": 0.9890008895369273, "300–1000 Hz_10ms_modulation_mae_db": 1.3495266732939168, "300–1000 Hz_40ms_log_rms_mae_db": 1.7043671973107284, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06748855396561332, "300–1000 Hz_40ms_envelope_correlation": 0.9967448794376198, "300–1000 Hz_40ms_modulation_mae_db": 1.4983373777266686, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2592507242428983, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16244956454579473, "1000–3000 Hz_10ms_envelope_correlation": 0.9827895641367246, "1000–3000 Hz_10ms_modulation_mae_db": 0.8085527595701052, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5492158307902557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09600622061575298, "1000–3000 Hz_40ms_envelope_correlation": 0.9938106236699301, "1000–3000 Hz_40ms_modulation_mae_db": 1.1832829128827898, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2974972649391927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20854463553749372, "3000–6000 Hz_10ms_envelope_correlation": 0.9735613563825226, "3000–6000 Hz_10ms_modulation_mae_db": 0.6834014643250802, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6442715608504652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12390616393116341, "3000–6000 Hz_40ms_envelope_correlation": 0.9903999452974803, "3000–6000 Hz_40ms_modulation_mae_db": 0.990079420979059, "mr_stft_db": 5.984005997151631, "band_mae_db": 0.17006333890350336, "rms_gain_error_db": 0.3416851595440473, "output_peak": 0.7336355812338892}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.84919980558523, "全帯域_10ms_rms_envelope_nrmse": 0.22766072396151227, "全帯域_10ms_envelope_correlation": 0.9558256705607507, "全帯域_10ms_modulation_mae_db": 0.7719110998062674, "全帯域_40ms_log_rms_mae_db": 0.9357414111013606, "全帯域_40ms_rms_envelope_nrmse": 0.09620285532021595, "全帯域_40ms_envelope_correlation": 0.9915145793015042, "全帯域_40ms_modulation_mae_db": 1.0857876179062451, "300–1000 Hz_10ms_log_rms_mae_db": 2.4245076386014905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12217054109689261, "300–1000 Hz_10ms_envelope_correlation": 0.9898583673387477, "300–1000 Hz_10ms_modulation_mae_db": 1.5945588080660276, "300–1000 Hz_40ms_log_rms_mae_db": 1.7631828056787917, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07905749031973511, "300–1000 Hz_40ms_envelope_correlation": 0.9956365208844683, "300–1000 Hz_40ms_modulation_mae_db": 1.9289517533744387, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3173464595633932, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1638459113144053, "1000–3000 Hz_10ms_envelope_correlation": 0.9829027873664798, "1000–3000 Hz_10ms_modulation_mae_db": 0.7140847739331991, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5990703863429367, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09647704390231611, "1000–3000 Hz_40ms_envelope_correlation": 0.9943430160463405, "1000–3000 Hz_40ms_modulation_mae_db": 0.9907155173659131, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3407333432096262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17908143687211478, "3000–6000 Hz_10ms_envelope_correlation": 0.9809048668735135, "3000–6000 Hz_10ms_modulation_mae_db": 0.4962604827801629, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6661264852150215, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10405026725792728, "3000–6000 Hz_40ms_envelope_correlation": 0.9935858396612626, "3000–6000 Hz_40ms_modulation_mae_db": 0.6051213859134795, "mr_stft_db": 5.936122401954152, "band_mae_db": 0.28839448440568977, "rms_gain_error_db": -0.17367788437630757, "output_peak": 0.6076377972223729}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8146965701350368, "全帯域_10ms_rms_envelope_nrmse": 0.28261272728998443, "全帯域_10ms_envelope_correlation": 0.946449449185759, "全帯域_10ms_modulation_mae_db": 2.7411972797329045, "全帯域_40ms_log_rms_mae_db": 0.9958710908816021, "全帯域_40ms_rms_envelope_nrmse": 0.2278235008117946, "全帯域_40ms_envelope_correlation": 0.9631556176937467, "全帯域_40ms_modulation_mae_db": 3.259776063005711, "300–1000 Hz_10ms_log_rms_mae_db": 2.4370924822494566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11177795038822659, "300–1000 Hz_10ms_envelope_correlation": 0.9915569837592035, "300–1000 Hz_10ms_modulation_mae_db": 0.7912699298555496, "300–1000 Hz_40ms_log_rms_mae_db": 1.6543294158141375, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0690582313850767, "300–1000 Hz_40ms_envelope_correlation": 0.9966380348634986, "300–1000 Hz_40ms_modulation_mae_db": 1.0061125434341565, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2494625730848066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23183333096460143, "1000–3000 Hz_10ms_envelope_correlation": 0.9645638007229249, "1000–3000 Hz_10ms_modulation_mae_db": 1.2932476982932493, "1000–3000 Hz_40ms_log_rms_mae_db": 1.617798580530013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.125199126292882, "1000–3000 Hz_40ms_envelope_correlation": 0.9897783896168745, "1000–3000 Hz_40ms_modulation_mae_db": 1.7346158737080624, "3000–6000 Hz_10ms_log_rms_mae_db": 2.269534392419131, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17795552617291513, "3000–6000 Hz_10ms_envelope_correlation": 0.9799888164786215, "3000–6000 Hz_10ms_modulation_mae_db": 0.7843098987901191, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5752341248955521, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09220559021978951, "3000–6000 Hz_40ms_envelope_correlation": 0.9944691759820902, "3000–6000 Hz_40ms_modulation_mae_db": 1.036346347298573, "mr_stft_db": 5.957926899926028, "band_mae_db": 0.26749737639800797, "rms_gain_error_db": 0.6665767023038236, "output_peak": 1.3336211464092085}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 31, "at_or_above_full_scale_fraction": 0.0008123689727463312, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0008123689727463312, "longest_full_scale_run_samples": 31, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7391295987701463, "全帯域_10ms_rms_envelope_nrmse": 0.3693428110613418, "全帯域_10ms_envelope_correlation": 0.9204262304282219, "全帯域_10ms_modulation_mae_db": 2.1066232793341593, "全帯域_40ms_log_rms_mae_db": 0.9898542758488194, "全帯域_40ms_rms_envelope_nrmse": 0.277226929658479, "全帯域_40ms_envelope_correlation": 0.9566687131605379, "全帯域_40ms_modulation_mae_db": 2.6220451214629352, "300–1000 Hz_10ms_log_rms_mae_db": 2.496903878194542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13355872308884578, "300–1000 Hz_10ms_envelope_correlation": 0.9877168736299958, "300–1000 Hz_10ms_modulation_mae_db": 0.8162475887172004, "300–1000 Hz_40ms_log_rms_mae_db": 1.8882388937938708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08235535914971907, "300–1000 Hz_40ms_envelope_correlation": 0.9950363815786698, "300–1000 Hz_40ms_modulation_mae_db": 1.2428385186783293, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2995268759049887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19362240454016674, "1000–3000 Hz_10ms_envelope_correlation": 0.9782428450388723, "1000–3000 Hz_10ms_modulation_mae_db": 1.0595583164454367, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6053740285448872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1364760429804377, "1000–3000 Hz_40ms_envelope_correlation": 0.9910285634596369, "1000–3000 Hz_40ms_modulation_mae_db": 1.258479759939006, "3000–6000 Hz_10ms_log_rms_mae_db": 2.31148797467258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15525882324869614, "3000–6000 Hz_10ms_envelope_correlation": 0.9850567686928513, "3000–6000 Hz_10ms_modulation_mae_db": 1.0791885750408403, "3000–6000 Hz_40ms_log_rms_mae_db": 1.621186979064986, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0817785410367126, "3000–6000 Hz_40ms_envelope_correlation": 0.9957989649179912, "3000–6000 Hz_40ms_modulation_mae_db": 1.26286436505841, "mr_stft_db": 5.961330457017042, "band_mae_db": 0.4825342796200245, "rms_gain_error_db": 1.0708870879191492, "output_peak": 1.071396172995438}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 24, "at_or_above_full_scale_fraction": 0.0006289308176100629, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0006289308176100629, "longest_full_scale_run_samples": 24, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.012948525242078, "全帯域_10ms_rms_envelope_nrmse": 0.279941989088224, "全帯域_10ms_envelope_correlation": 0.9354838373768551, "全帯域_10ms_modulation_mae_db": 2.9754207883415287, "全帯域_40ms_log_rms_mae_db": 1.2222649678476898, "全帯域_40ms_rms_envelope_nrmse": 0.2202172352500994, "全帯域_40ms_envelope_correlation": 0.9599135534178954, "全帯域_40ms_modulation_mae_db": 3.6659683701421315, "300–1000 Hz_10ms_log_rms_mae_db": 2.4144214223072042, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11294180639520134, "300–1000 Hz_10ms_envelope_correlation": 0.9913753229475447, "300–1000 Hz_10ms_modulation_mae_db": 1.066949498361527, "300–1000 Hz_40ms_log_rms_mae_db": 1.7183062326880785, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06832224160218682, "300–1000 Hz_40ms_envelope_correlation": 0.9967524729229903, "300–1000 Hz_40ms_modulation_mae_db": 1.1923694556776512, "1000–3000 Hz_10ms_log_rms_mae_db": 2.239002335928957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1609909870876958, "1000–3000 Hz_10ms_envelope_correlation": 0.9836067554192369, "1000–3000 Hz_10ms_modulation_mae_db": 0.3455220677244964, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5782894177274576, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10015337027971234, "1000–3000 Hz_40ms_envelope_correlation": 0.9933782455745936, "1000–3000 Hz_40ms_modulation_mae_db": 0.7245707479272077, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3520005196565226, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20934276207978111, "3000–6000 Hz_10ms_envelope_correlation": 0.9722551387353799, "3000–6000 Hz_10ms_modulation_mae_db": 2.012389331110445, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6969154258906678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14067152401638505, "3000–6000 Hz_40ms_envelope_correlation": 0.9870516710714815, "3000–6000 Hz_40ms_modulation_mae_db": 2.143808296986648, "mr_stft_db": 5.931801537321021, "band_mae_db": 0.5074784755606352, "rms_gain_error_db": -0.9537880104096084, "output_peak": 0.5177675634125425}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7572304423786427, "全帯域_10ms_rms_envelope_nrmse": 0.2289110993614791, "全帯域_10ms_envelope_correlation": 0.9561057802464379, "全帯域_10ms_modulation_mae_db": 0.6497717036590418, "全帯域_40ms_log_rms_mae_db": 0.922997824567352, "全帯域_40ms_rms_envelope_nrmse": 0.13884088754067814, "全帯域_40ms_envelope_correlation": 0.9820301511521163, "全帯域_40ms_modulation_mae_db": 0.732910142861125, "300–1000 Hz_10ms_log_rms_mae_db": 2.385659577043488, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1300671657256955, "300–1000 Hz_10ms_envelope_correlation": 0.9884134716600224, "300–1000 Hz_10ms_modulation_mae_db": 1.1660652625908894, "300–1000 Hz_40ms_log_rms_mae_db": 1.7209322844906463, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07944646584526681, "300–1000 Hz_40ms_envelope_correlation": 0.9954733057285547, "300–1000 Hz_40ms_modulation_mae_db": 1.3913330105176582, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3395451599623005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.256555751909615, "1000–3000 Hz_10ms_envelope_correlation": 0.9618510850309875, "1000–3000 Hz_10ms_modulation_mae_db": 2.3142075173160452, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6532128745103862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19847348651596067, "1000–3000 Hz_40ms_envelope_correlation": 0.9802262863629929, "1000–3000 Hz_40ms_modulation_mae_db": 2.3544346784965025, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2294092193289026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15982875361863338, "3000–6000 Hz_10ms_envelope_correlation": 0.9844124678923808, "3000–6000 Hz_10ms_modulation_mae_db": 0.9675442250296048, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5433390392950936, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09277208537972685, "3000–6000 Hz_40ms_envelope_correlation": 0.994600500867778, "3000–6000 Hz_40ms_modulation_mae_db": 1.117477646489011, "mr_stft_db": 5.938219841118894, "band_mae_db": 0.3459525784044705, "rms_gain_error_db": 0.0249318352159781, "output_peak": 0.7404888819429698}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 38160, "output_samples": 38160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.169126287752205, "spectral_convergence": 0.6002153965337679, "spectral_convergence_status": "ok", "amplitude_floor": 5.674431450769592e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.771619788415548, "spectral_convergence": 0.5844024008372948, "spectral_convergence_status": "ok", "amplitude_floor": 3.1139992269810527e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.893781789093306, "spectral_convergence": 0.5275514445400181, "spectral_convergence_status": "ok", "amplitude_floor": 1.28722240515293e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.003603182122246331, "output_power": 0.005333438371125817, "delta_db": 1.7032106708176102, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001490228689925631, "output_power": 0.001579252119549835, "delta_db": 0.2519854825128432, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00043834980790852295, "output_power": 0.00043740337028743147, "delta_db": -0.009386955287401634, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023807008265260595, "output_power": 0.00023003316531088633, "delta_db": -0.14914367258318473, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.11014495688624e-05, "output_power": 8.777606505197396e-05, "delta_db": 0.34347491131582875, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.691475709405268e-05, "output_power": 2.8150179412198243e-05, "delta_db": 0.19490702468816296, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.44201811953417186, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.170497946204621, "pitch": null, "output_rms_dbfs": -21.137319348491992, "output_peak": 1.295377598419888, "reference_rms_dbfs": -22.307817294696616, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.944842621753686}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 38160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.0004033177834583336, "log_rms_mae_db": 1.70907842462923, "rms_envelope_nrmse": 0.3567066219884458, "envelope_correlation": 0.9306641149322759, "rms_gain_error_db": 1.1704823701344345, "reference_mean_power": 0.005890133913079112, "output_mean_power": 0.007712113378587561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 4.702783438048151e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019529842144495992, "output_power": 0.00277582850913893, "reference_active": true, "delta_db": 1.5269389887762808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0010897406087412391, "output_power": 0.0025621982952824, "reference_active": true, "delta_db": 3.7128960289838044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00042203241052064084, "output_power": 0.0014913074666327083, "reference_active": true, "delta_db": 5.482213878319091}], "active_band_count": 3, "active_band_mae_db": 3.5740162986930586}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0002641139400842597, "log_rms_mae_db": 0.9887263326612729, "rms_envelope_nrmse": 0.3002398554525687, "envelope_correlation": 0.9502605712575078, "rms_gain_error_db": 1.1705179328572517, "reference_mean_power": 0.0059649962331930016, "output_mean_power": 0.007810196615641528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.959398727002631e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020291985866854774, "output_power": 0.0037263943233063306, "reference_active": true, "delta_db": 2.639642584338826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005646977153887027, "output_power": 0.0015961668297044176, "reference_active": true, "delta_db": 4.512622506919588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.714480412149099e-05, "output_power": 0.00013576199526939273, "reference_active": true, "delta_db": 5.628801396108968}], "active_band_count": 3, "active_band_mae_db": 4.260355495789128}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.00011878678262042868, "log_rms_mae_db": 2.3565554949130947, "rms_envelope_nrmse": 0.12130208450080439, "envelope_correlation": 0.9899102774156417, "rms_gain_error_db": -0.04847763837269635, "reference_mean_power": 0.0009978964081743123, "output_mean_power": 0.0009868194382746923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 1.0827926032603135e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006103304193321489, "output_power": 0.00057305027117542, "reference_active": true, "delta_db": -0.27372293683216925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011556410212235237, "output_power": 0.00010474207703789323, "reference_active": true, "delta_db": -0.4270176766354153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8931134571049757e-05, "output_power": 1.112145648179649e-05, "reference_active": true, "delta_db": -2.3101497586282083}], "active_band_count": 3, "active_band_mae_db": 1.003630124031931}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 9.505044043492436e-05, "log_rms_mae_db": 1.6429676962187807, "rms_envelope_nrmse": 0.06674590601999912, "envelope_correlation": 0.9968682454573164, "rms_gain_error_db": -0.04848777618547968, "reference_mean_power": 0.0010106286473201321, "output_mean_power": 0.0009994080125475418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.735673127654544e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005396175923660179, "output_power": 0.0005086504761620688, "reference_active": true, "delta_db": -0.2566664451151474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.461345336739454e-05, "output_power": 5.786444168172734e-05, "reference_active": true, "delta_db": -0.47911186120643834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3971113807678808e-06, "output_power": 1.1337810295602968e-06, "reference_active": true, "delta_db": -3.2515902771762772}], "active_band_count": 3, "active_band_mae_db": 1.3291228611659542}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 9.972124238081888e-05, "log_rms_mae_db": 2.27847610687973, "rms_envelope_nrmse": 0.20559883348893623, "envelope_correlation": 0.9726094811455679, "rms_gain_error_db": -0.02623740578052125, "reference_mean_power": 0.00028254365881671625, "output_mean_power": 0.0002808418493480834}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 3.4051840450675846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002046408997048573, "output_power": 0.00019015882752241534, "reference_active": true, "delta_db": -0.31875945552943763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.521356834551269e-05, "output_power": 3.516319687811526e-05, "reference_active": true, "delta_db": -1.0918043150060446}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018106896586635e-06, "output_power": 9.476566205654316e-06, "reference_active": true, "delta_db": 0.7257915911939035}], "active_band_count": 3, "active_band_mae_db": 0.7121184539097953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 8.26289228894579e-05, "log_rms_mae_db": 1.6625700130235077, "rms_envelope_nrmse": 0.12715858669163407, "envelope_correlation": 0.9891575264543184, "rms_gain_error_db": -0.026251751869574774, "reference_mean_power": 0.0002861487173387815, "output_mean_power": 0.00028442425443454296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.083140770966499e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019461067978585587, "output_power": 0.0001783971555706428, "reference_active": true, "delta_db": -0.3777874412157124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0239988981192927e-05, "output_power": 1.5676778081668107e-05, "reference_active": true, "delta_db": -1.1095346116240385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.640717194960459e-07, "output_power": 7.05684884169811e-07, "reference_active": true, "delta_db": -0.34523310576987787}], "active_band_count": 3, "active_band_mae_db": 0.6108517195365429}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 4.3426949519701344e-05, "log_rms_mae_db": 2.258361635192704, "rms_envelope_nrmse": 0.19157129516932456, "envelope_correlation": 0.9772384231489453, "rms_gain_error_db": 0.03335170857786272, "reference_mean_power": 4.5529138246552346e-05, "output_mean_power": 4.588012591550936e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 6.179039674292901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3063171222465546e-05, "output_power": 3.1775489988401395e-05, "reference_active": true, "delta_db": -0.17252249933336355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2207009907392542e-05, "output_power": 8.231432677331285e-06, "reference_active": true, "delta_db": -1.7113386649335305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2513877801441335e-06, "output_power": 9.621110691414953e-07, "reference_active": true, "delta_db": -3.6922509328048414}], "active_band_count": 3, "active_band_mae_db": 1.8587040323572452}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 3.683581645514392e-05, "log_rms_mae_db": 1.594439975053868, "rms_envelope_nrmse": 0.11929898191502068, "envelope_correlation": 0.9907442569445308, "rms_gain_error_db": 0.03332101871240293, "reference_mean_power": 4.610980557969544e-05, "output_mean_power": 4.646494130786164e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 5.621286439624602e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.226124623783574e-05, "output_power": 2.9334826855122675e-05, "reference_active": true, "delta_db": -0.4129761078788574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.247687157348319e-06, "output_power": 4.674973277085248e-06, "reference_active": true, "delta_db": -1.904203062605159}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.30977962238882e-07, "output_power": 1.0109437248609605e-07, "reference_active": true, "delta_db": -3.588435646925961}], "active_band_count": 3, "active_band_mae_db": 1.9685382724699927}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8012420591482878, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8c728e7e81860e72f8c40a19fe43367916f1fdc67f62f7669b1db4bf655ce565", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9727851844856604, "standard_deviation": 0.010875321653626115, "minimum": 0.9557151570244895, "maximum": 0.9836067554192369, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.277560054401432, "standard_deviation": 0.037975883320576484, "minimum": 2.2378701996443855, "maximum": 2.3395451599623005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9620467860040256, "standard_deviation": 0.6254467904574815, "minimum": 0.3455220677244964, "maximum": 2.3142075173160452, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20431586660872125, "standard_deviation": 0.04128218597103414, "minimum": 0.1609909870876958, "maximum": 0.25963014901855475, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9895702821809169, "standard_deviation": 0.0048886147975658175, "minimum": 0.9802262863629929, "maximum": 0.9943430160463405, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.6090123884623893, "standard_deviation": 0.03691086705443317, "minimum": 1.5492158307902557, "maximum": 1.6625700130235077, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2239450938141223, "standard_deviation": 0.5738350588302858, "minimum": 0.6108517195365429, "maximum": 2.3544346784965025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.128404064723326, "standard_deviation": 0.034247810043942534, "minimum": 0.09600622061575298, "maximum": 0.19847348651596067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792809449306026, "standard_deviation": 0.0046518371452900845, "minimum": 0.9722551387353799, "maximum": 0.9850567686928513, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2875311761246495, "standard_deviation": 0.04528825548506901, "minimum": 2.2294092193289026, "maximum": 2.3520005196565226, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0583215612226313, "standard_deviation": 0.5751199043407215, "minimum": 0.4962604827801629, "maximum": 2.012389331110445, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1830217675992984, "standard_deviation": 0.019883164395922005, "minimum": 0.15525882324869614, "maximum": 0.20934276207978111, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926105288708218, "standard_deviation": 0.0029383600989460706, "minimum": 0.9870516710714815, "maximum": 0.9957989649179912, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6150349650427656, "standard_deviation": 0.05164628843091403, "minimum": 1.5433390392950936, "maximum": 1.6969154258906678, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2233337428814406, "standard_deviation": 0.5608078421094066, "minimum": 0.6051213859134795, "maximum": 2.143808296986648, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10733570405237824, "standard_deviation": 0.019438803063372593, "minimum": 0.0817785410367126, "maximum": 0.14067152401638505, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.989832491901586, "standard_deviation": 0.0013903996383176397, "minimum": 0.9877168736299958, "maximum": 0.9915569837592035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.4277485759118145, "standard_deviation": 0.04546260022682429, "minimum": 2.3565554949130947, "maximum": 2.496903878194542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.107508368730592, "standard_deviation": 0.26666429679828624, "minimum": 0.7912699298555496, "maximum": 1.5945588080660276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12190758816391178, "standard_deviation": 0.007861726892378173, "minimum": 0.11177795038822659, "maximum": 0.13355872308884578, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962242282554545, "standard_deviation": 0.0007203743409838262, "minimum": 0.9950363815786698, "maximum": 0.9968682454573164, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7299532550034626, "standard_deviation": 0.0761733347172248, "minimum": 1.6429676962187807, "maximum": 1.8882388937938708, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3465003069925614, "standard_deviation": 0.2781918716518475, "minimum": 1.0061125434341565, "maximum": 1.9289517533744387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07275115580272457, "standard_deviation": 0.006371727714616831, "minimum": 0.06674590601999912, "maximum": 0.08235535914971907, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3605462027451495, "standard_deviation": 0.11579779833006201, "minimum": 0.17006333890350336, "maximum": 0.5074784755606352, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.95199956859438, "standard_deviation": 0.01749598420778622, "minimum": 5.931801537321021, "maximum": 5.984005997151631, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9108986067853476, "standard_deviation": 0.3082416469005725, "minimum": 0.5177675634125425, "maximum": 1.3336211464092085, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": 0.37344503357849496, "standard_deviation": 0.7180427125853031, "minimum": -0.9537880104096084, "maximum": 1.170497946204621, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9488655847629199, "standard_deviation": 0.01954948226610792, "minimum": 0.9204262304282219, "maximum": 0.9795399657181857, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7896156900018618, "standard_deviation": 0.11568209381686716, "minimum": 1.6196414162514108, "maximum": 2.012948525242078, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.9627459892755792, "standard_deviation": 1.1562981245884933, "minimum": 0.5972884112246559, "maximum": 3.5740162986930586, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.26910773782425734, "standard_deviation": 0.06839426885854205, "minimum": 0.16595216083694664, "maximum": 0.3693428110613418, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9724084869244072, "standard_deviation": 0.01690576810594601, "minimum": 0.9502605712575078, "maximum": 0.9943954290455622, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9973310784270184, "standard_deviation": 0.09674955326364816, "minimum": 0.922997824567352, "maximum": 1.2222649678476898, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 2.400965531228385, "standard_deviation": 1.3529169685295765, "minimum": 0.732910142861125, "maximum": 4.260355495789128, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.19253705373624797, "standard_deviation": 0.07860894433911261, "minimum": 0.09108057787564183, "maximum": 0.3002398554525687, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2094-142345-0040", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.7880301475524902, "predicted_mos": 1.7880301475524902, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "880c59534f2737547deec2cc9d647227fd9a16a61370778c22c0cce6ca475dff", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.489181800279766, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5970125062919474, "全帯域_10ms_rms_envelope_nrmse": 0.22358251165020324, "全帯域_10ms_envelope_correlation": 0.9568943578518199, "全帯域_10ms_modulation_mae_db": 0.8881624059254266, "全帯域_40ms_log_rms_mae_db": 0.805767572532643, "全帯域_40ms_rms_envelope_nrmse": 0.11601912603587314, "全帯域_40ms_envelope_correlation": 0.9875412273805585, "全帯域_40ms_modulation_mae_db": 1.3743784706256958, "300–1000 Hz_10ms_log_rms_mae_db": 2.349702433762901, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13364946069341785, "300–1000 Hz_10ms_envelope_correlation": 0.9877547538717634, "300–1000 Hz_10ms_modulation_mae_db": 0.9512239518428901, "300–1000 Hz_40ms_log_rms_mae_db": 1.5695103356872608, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08761928461225002, "300–1000 Hz_40ms_envelope_correlation": 0.9944887485388645, "300–1000 Hz_40ms_modulation_mae_db": 1.0755534594744747, "1000–3000 Hz_10ms_log_rms_mae_db": 2.161588538694025, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15808650305133437, "1000–3000 Hz_10ms_envelope_correlation": 0.98378842712466, "1000–3000 Hz_10ms_modulation_mae_db": 1.2109398509814209, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3420339388474913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07741510450011783, "1000–3000 Hz_40ms_envelope_correlation": 0.9959129970695889, "1000–3000 Hz_40ms_modulation_mae_db": 1.4047565073090829, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1903862906776537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1701317899002066, "3000–6000 Hz_10ms_envelope_correlation": 0.9816037088574515, "3000–6000 Hz_10ms_modulation_mae_db": 1.103333662746316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4307340517813043, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08716009848313448, "3000–6000 Hz_40ms_envelope_correlation": 0.9949102824631952, "3000–6000 Hz_40ms_modulation_mae_db": 1.6033291089207236, "mr_stft_db": 5.87942417526904, "band_mae_db": 0.21404723699244288, "rms_gain_error_db": -0.23510411156097177, "output_peak": 0.6737772326843414}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8747216974259653, "全帯域_10ms_rms_envelope_nrmse": 0.4062798385038208, "全帯域_10ms_envelope_correlation": 0.8660083882984441, "全帯域_10ms_modulation_mae_db": 1.6652220373713806, "全帯域_40ms_log_rms_mae_db": 1.033786610064867, "全帯域_40ms_rms_envelope_nrmse": 0.2444539944739864, "全帯域_40ms_envelope_correlation": 0.9459002123765855, "全帯域_40ms_modulation_mae_db": 2.238769026612768, "300–1000 Hz_10ms_log_rms_mae_db": 2.4052227764410796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11924495799668618, "300–1000 Hz_10ms_envelope_correlation": 0.9903477060897071, "300–1000 Hz_10ms_modulation_mae_db": 0.8601897089014222, "300–1000 Hz_40ms_log_rms_mae_db": 1.6099503975993945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07131315105284584, "300–1000 Hz_40ms_envelope_correlation": 0.9964167940295032, "300–1000 Hz_40ms_modulation_mae_db": 1.0517297987907666, "1000–3000 Hz_10ms_log_rms_mae_db": 2.16376568101994, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16805068769635617, "1000–3000 Hz_10ms_envelope_correlation": 0.98147554868365, "1000–3000 Hz_10ms_modulation_mae_db": 0.5842597702007569, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4202269823731923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09161764066491793, "1000–3000 Hz_40ms_envelope_correlation": 0.9941917539551187, "1000–3000 Hz_40ms_modulation_mae_db": 0.21345975779347695, "3000–6000 Hz_10ms_log_rms_mae_db": 2.041189800121792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16682433158772664, "3000–6000 Hz_10ms_envelope_correlation": 0.982409349968953, "3000–6000 Hz_10ms_modulation_mae_db": 1.0303736077639858, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3422069679573798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09876044170563356, "3000–6000 Hz_40ms_envelope_correlation": 0.9934018700781422, "3000–6000 Hz_40ms_modulation_mae_db": 1.0670414524887846, "mr_stft_db": 5.86695881873553, "band_mae_db": 0.25062624265476513, "rms_gain_error_db": 0.12814462640357227, "output_peak": 0.8167517704513565}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7742362846916644, "全帯域_10ms_rms_envelope_nrmse": 0.2358499951925767, "全帯域_10ms_envelope_correlation": 0.9533273613115963, "全帯域_10ms_modulation_mae_db": 0.9478330088204675, "全帯域_40ms_log_rms_mae_db": 1.140673872010934, "全帯域_40ms_rms_envelope_nrmse": 0.10723575830409474, "全帯域_40ms_envelope_correlation": 0.9892331955325936, "全帯域_40ms_modulation_mae_db": 0.6217926362276441, "300–1000 Hz_10ms_log_rms_mae_db": 2.306070245348673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0975164242739657, "300–1000 Hz_10ms_envelope_correlation": 0.9937824464452882, "300–1000 Hz_10ms_modulation_mae_db": 0.4323838343063924, "300–1000 Hz_40ms_log_rms_mae_db": 1.6090683946914133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05914149392728609, "300–1000 Hz_40ms_envelope_correlation": 0.9976764635005674, "300–1000 Hz_40ms_modulation_mae_db": 0.5393740483286223, "1000–3000 Hz_10ms_log_rms_mae_db": 2.098073064479112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18527323795462722, "1000–3000 Hz_10ms_envelope_correlation": 0.9775410724285707, "1000–3000 Hz_10ms_modulation_mae_db": 0.5458929332729426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4124342301352508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11839189014593836, "1000–3000 Hz_40ms_envelope_correlation": 0.9904379039877782, "1000–3000 Hz_40ms_modulation_mae_db": 0.325838026163486, "3000–6000 Hz_10ms_log_rms_mae_db": 2.22019641340001, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20297798329719743, "3000–6000 Hz_10ms_envelope_correlation": 0.9745858258289096, "3000–6000 Hz_10ms_modulation_mae_db": 1.9749848940892403, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4984556007683538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1271313681352413, "3000–6000 Hz_40ms_envelope_correlation": 0.9907094970637478, "3000–6000 Hz_40ms_modulation_mae_db": 2.025046848382576, "mr_stft_db": 5.876734640802693, "band_mae_db": 0.16924745525947893, "rms_gain_error_db": 0.07297104362722336, "output_peak": 0.7537140368273907}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6093157175146064, "全帯域_10ms_rms_envelope_nrmse": 0.2220421303878298, "全帯域_10ms_envelope_correlation": 0.9620995516243619, "全帯域_10ms_modulation_mae_db": 2.8283713018937067, "全帯域_40ms_log_rms_mae_db": 0.8897307046774755, "全帯域_40ms_rms_envelope_nrmse": 0.17070028586027278, "全帯域_40ms_envelope_correlation": 0.9789976835045013, "全帯域_40ms_modulation_mae_db": 3.1893862698254574, "300–1000 Hz_10ms_log_rms_mae_db": 2.3005935690355224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1178866213183093, "300–1000 Hz_10ms_envelope_correlation": 0.9905574912678695, "300–1000 Hz_10ms_modulation_mae_db": 0.8809346697416446, "300–1000 Hz_40ms_log_rms_mae_db": 1.5211261386103268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07009185276254852, "300–1000 Hz_40ms_envelope_correlation": 0.9965524299993693, "300–1000 Hz_40ms_modulation_mae_db": 0.9277600078247908, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1110514023480134, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20528090548702746, "1000–3000 Hz_10ms_envelope_correlation": 0.9746126847876982, "1000–3000 Hz_10ms_modulation_mae_db": 1.9590102132696525, "1000–3000 Hz_40ms_log_rms_mae_db": 1.470273288926264, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14523631500282302, "1000–3000 Hz_40ms_envelope_correlation": 0.9886663691704641, "1000–3000 Hz_40ms_modulation_mae_db": 2.464507128781641, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0586888697963315, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1769259624926416, "3000–6000 Hz_10ms_envelope_correlation": 0.9804930505427699, "3000–6000 Hz_10ms_modulation_mae_db": 1.282474834517747, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3607746873223567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09596535472121402, "3000–6000 Hz_40ms_envelope_correlation": 0.9946547281257507, "3000–6000 Hz_40ms_modulation_mae_db": 1.587443129127223, "mr_stft_db": 5.882906616704463, "band_mae_db": 0.4162242192284695, "rms_gain_error_db": -0.8251139393909134, "output_peak": 0.5469662966946154}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6636576073805212, "全帯域_10ms_rms_envelope_nrmse": 0.18598693728751145, "全帯域_10ms_envelope_correlation": 0.9707118949960315, "全帯域_10ms_modulation_mae_db": 0.767770347259618, "全帯域_40ms_log_rms_mae_db": 0.9170993238208148, "全帯域_40ms_rms_envelope_nrmse": 0.099678667134272, "全帯域_40ms_envelope_correlation": 0.9916818053627742, "全帯域_40ms_modulation_mae_db": 0.7367835458875694, "300–1000 Hz_10ms_log_rms_mae_db": 2.3875430971198663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13233631053623068, "300–1000 Hz_10ms_envelope_correlation": 0.9880076713119574, "300–1000 Hz_10ms_modulation_mae_db": 0.7816455096931586, "300–1000 Hz_40ms_log_rms_mae_db": 1.5903541185615906, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07314594273070153, "300–1000 Hz_40ms_envelope_correlation": 0.9962257154379461, "300–1000 Hz_40ms_modulation_mae_db": 0.9194845723760148, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1582200058188468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19740752375520573, "1000–3000 Hz_10ms_envelope_correlation": 0.9750721298020759, "1000–3000 Hz_10ms_modulation_mae_db": 0.464896633894899, "1000–3000 Hz_40ms_log_rms_mae_db": 1.442436001540528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11370001885527768, "1000–3000 Hz_40ms_envelope_correlation": 0.991322348296697, "1000–3000 Hz_40ms_modulation_mae_db": 0.46902024083954547, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1500399554998166, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18033807713079833, "3000–6000 Hz_10ms_envelope_correlation": 0.9804527727287258, "3000–6000 Hz_10ms_modulation_mae_db": 0.6371941481755216, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4259234414701691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09925371080272913, "3000–6000 Hz_40ms_envelope_correlation": 0.9940971725698723, "3000–6000 Hz_40ms_modulation_mae_db": 0.9022812816536288, "mr_stft_db": 5.889466445146662, "band_mae_db": 0.2710335570501064, "rms_gain_error_db": -0.3414920108918962, "output_peak": 0.6218979702916685}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.729109911378192, "全帯域_10ms_rms_envelope_nrmse": 0.19848017213976973, "全帯域_10ms_envelope_correlation": 0.9664264582555352, "全帯域_10ms_modulation_mae_db": 0.5716514336638748, "全帯域_40ms_log_rms_mae_db": 1.0151631624841289, "全帯域_40ms_rms_envelope_nrmse": 0.11544672656713946, "全帯域_40ms_envelope_correlation": 0.9881625823609704, "全帯域_40ms_modulation_mae_db": 0.8351227836219958, "300–1000 Hz_10ms_log_rms_mae_db": 2.419123048633604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12086461333325378, "300–1000 Hz_10ms_envelope_correlation": 0.9902882369306657, "300–1000 Hz_10ms_modulation_mae_db": 0.9265751187887834, "300–1000 Hz_40ms_log_rms_mae_db": 1.737829448959928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08039439780791584, "300–1000 Hz_40ms_envelope_correlation": 0.9954779124941725, "300–1000 Hz_40ms_modulation_mae_db": 1.1299571520940062, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1222821627212802, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21375114676885007, "1000–3000 Hz_10ms_envelope_correlation": 0.9699349969563826, "1000–3000 Hz_10ms_modulation_mae_db": 0.7279199948325354, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4101347584231136, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1201452296433801, "1000–3000 Hz_40ms_envelope_correlation": 0.9904601241808009, "1000–3000 Hz_40ms_modulation_mae_db": 0.7538711121573346, "3000–6000 Hz_10ms_log_rms_mae_db": 2.150841751991159, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18397504066632805, "3000–6000 Hz_10ms_envelope_correlation": 0.978376727444089, "3000–6000 Hz_10ms_modulation_mae_db": 1.6154235088902311, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3580206774520418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09905708135931017, "3000–6000 Hz_40ms_envelope_correlation": 0.9934306703138163, "3000–6000 Hz_40ms_modulation_mae_db": 2.017761482626044, "mr_stft_db": 5.8840606896700764, "band_mae_db": 0.22279682787464494, "rms_gain_error_db": -0.3240837063883003, "output_peak": 0.6602129672512476}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7876616873433313, "全帯域_10ms_rms_envelope_nrmse": 0.3308087121719512, "全帯域_10ms_envelope_correlation": 0.9444900145493446, "全帯域_10ms_modulation_mae_db": 1.9239471789680653, "全帯域_40ms_log_rms_mae_db": 1.0235621319658896, "全帯域_40ms_rms_envelope_nrmse": 0.26987735449619504, "全帯域_40ms_envelope_correlation": 0.9662566167983067, "全帯域_40ms_modulation_mae_db": 2.180106548482065, "300–1000 Hz_10ms_log_rms_mae_db": 2.316374796647363, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10381466250853018, "300–1000 Hz_10ms_envelope_correlation": 0.9927237853994928, "300–1000 Hz_10ms_modulation_mae_db": 0.45884827733929184, "300–1000 Hz_40ms_log_rms_mae_db": 1.5940863517516863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06471128033630809, "300–1000 Hz_40ms_envelope_correlation": 0.9971053766962089, "300–1000 Hz_40ms_modulation_mae_db": 0.35778086487102484, "1000–3000 Hz_10ms_log_rms_mae_db": 2.13784661537522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1807780134203505, "1000–3000 Hz_10ms_envelope_correlation": 0.9785510422543413, "1000–3000 Hz_10ms_modulation_mae_db": 0.2570389981930806, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4057974852314195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10562776060397347, "1000–3000 Hz_40ms_envelope_correlation": 0.9924622130014636, "1000–3000 Hz_40ms_modulation_mae_db": 0.47002317227015555, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0559365589264202, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1777715140805253, "3000–6000 Hz_10ms_envelope_correlation": 0.9811608075996234, "3000–6000 Hz_10ms_modulation_mae_db": 1.067971974611601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3677438369372166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09779211182750526, "3000–6000 Hz_40ms_envelope_correlation": 0.9943924544649668, "3000–6000 Hz_40ms_modulation_mae_db": 1.2158603788703772, "mr_stft_db": 5.88419194806254, "band_mae_db": 0.4211378035243951, "rms_gain_error_db": 1.2307389732156206, "output_peak": 0.8979083476737221}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8680230232310273, "全帯域_10ms_rms_envelope_nrmse": 0.23801920966834134, "全帯域_10ms_envelope_correlation": 0.9534549947344101, "全帯域_10ms_modulation_mae_db": 0.6014517935660872, "全帯域_40ms_log_rms_mae_db": 0.9405971919784759, "全帯域_40ms_rms_envelope_nrmse": 0.11988093651676052, "全帯域_40ms_envelope_correlation": 0.9871125790134135, "全帯域_40ms_modulation_mae_db": 0.5974859195140522, "300–1000 Hz_10ms_log_rms_mae_db": 2.2895601571419224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10934655701391174, "300–1000 Hz_10ms_envelope_correlation": 0.9918844561901791, "300–1000 Hz_10ms_modulation_mae_db": 0.9800975232298175, "300–1000 Hz_40ms_log_rms_mae_db": 1.5433144622879895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05997684713047246, "300–1000 Hz_40ms_envelope_correlation": 0.9974733930255242, "300–1000 Hz_40ms_modulation_mae_db": 1.1082073898837512, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1021605936584415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2505717503585725, "1000–3000 Hz_10ms_envelope_correlation": 0.9602855265588298, "1000–3000 Hz_10ms_modulation_mae_db": 1.5466734226677834, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4023482561394616, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13666458190544403, "1000–3000 Hz_40ms_envelope_correlation": 0.9877184701385299, "1000–3000 Hz_40ms_modulation_mae_db": 1.396731264402679, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1249267357575996, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17298465662988066, "3000–6000 Hz_10ms_envelope_correlation": 0.9816980526828988, "3000–6000 Hz_10ms_modulation_mae_db": 1.0025580241009715, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3857020979222674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10336834010735889, "3000–6000 Hz_40ms_envelope_correlation": 0.9932049507754286, "3000–6000 Hz_40ms_modulation_mae_db": 0.957946031714415, "mr_stft_db": 5.861126188273123, "band_mae_db": 0.13283335912154157, "rms_gain_error_db": 0.1415493911474517, "output_peak": 0.9779252990359312}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 38160, "output_samples": 38160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.092927720301303, "spectral_convergence": 0.4995032694821619, "spectral_convergence_status": "ok", "amplitude_floor": 5.674431450769592e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.702110461427573, "spectral_convergence": 0.5532577539766023, "spectral_convergence_status": "ok", "amplitude_floor": 3.1139992269810527e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.843234344078245, "spectral_convergence": 0.6049181328715696, "spectral_convergence_status": "ok", "amplitude_floor": 1.28722240515293e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.003603182122246331, "output_power": 0.003202288047660252, "delta_db": -0.512258196403971, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001490228689925631, "output_power": 0.0015827836747666771, "delta_db": 0.2616864200144654, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00043834980790852295, "output_power": 0.00043765723455930984, "delta_db": -0.006867087408272371, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023807008265260595, "output_power": 0.0002416851248914173, "delta_db": 0.06545098778167725, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.11014495688624e-05, "output_power": 7.585246889296558e-05, "delta_db": -0.2905889561661391, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.691475709405268e-05, "output_power": 2.7844129315729518e-05, "delta_db": 0.1474317741801322, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21404723699244288, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.23510411156097177, "pitch": null, "output_rms_dbfs": -22.54292140625759, "output_peak": 0.6737772326843414, "reference_rms_dbfs": -22.307817294696616, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.87942417526904}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 38160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.0004033177834583336, "log_rms_mae_db": 1.5970125062919474, "rms_envelope_nrmse": 0.22358251165020324, "envelope_correlation": 0.9568943578518199, "rms_gain_error_db": -0.2351034703973987, "reference_mean_power": 0.005890133913079112, "output_mean_power": 0.005579751037415932}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 4.702783438048151e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019529842144495992, "output_power": 0.0019060133461904177, "reference_active": true, "delta_db": -0.10572795685516438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0010897406087412391, "output_power": 0.0008400118253667614, "reference_active": true, "delta_db": -1.1303773506880472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00042203241052064084, "output_power": 0.00030374393394624004, "reference_active": true, "delta_db": -1.4283819102330686}], "active_band_count": 3, "active_band_mae_db": 0.8881624059254266}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0002641139400842597, "log_rms_mae_db": 0.805767572532643, "rms_envelope_nrmse": 0.11601912603587314, "envelope_correlation": 0.9875412273805585, "rms_gain_error_db": -0.2350763993930457, "reference_mean_power": 0.0059649962331930016, "output_mean_power": 0.00565070368131744}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.959398727002631e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020291985866854774, "output_power": 0.0018731148843585644, "reference_active": true, "delta_db": -0.34760136248654155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005646977153887027, "output_power": 0.0004101443765611229, "reference_active": true, "delta_db": -1.3887926933485035}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.714480412149099e-05, "output_power": 2.1439958327409445e-05, "reference_active": true, "delta_db": -2.3867413560420423}], "active_band_count": 3, "active_band_mae_db": 1.3743784706256958}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.00011878678262042868, "log_rms_mae_db": 2.349702433762901, "rms_envelope_nrmse": 0.13364946069341785, "envelope_correlation": 0.9877547538717634, "rms_gain_error_db": -0.1350192516116931, "reference_mean_power": 0.0009978964081743123, "output_mean_power": 0.0009673497734196369}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 1.0827926032603135e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006103304193321489, "output_power": 0.0005765473946838497, "reference_active": true, "delta_db": -0.24730002084480973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011556410212235237, "output_power": 8.70994046560355e-05, "reference_active": true, "delta_db": -1.228077628667377}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8931134571049757e-05, "output_power": 1.378310999620463e-05, "reference_active": true, "delta_db": -1.378294206016484}], "active_band_count": 3, "active_band_mae_db": 0.9512239518428901}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 9.505044043492436e-05, "log_rms_mae_db": 1.5695103356872608, "rms_envelope_nrmse": 0.08761928461225002, "envelope_correlation": 0.9944887485388645, "rms_gain_error_db": -0.13504030182668855, "reference_mean_power": 0.0010106286473201321, "output_mean_power": 0.000979687517090001}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.735673127654544e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005396175923660179, "output_power": 0.0005065949500945065, "reference_active": true, "delta_db": -0.2742524361527797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.461345336739454e-05, "output_power": 4.7017514007941794e-05, "reference_active": true, "delta_db": -1.380632905903176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3971113807678808e-06, "output_power": 1.6692089712165506e-06, "reference_active": true, "delta_db": -1.5717750363674685}], "active_band_count": 3, "active_band_mae_db": 1.0755534594744747}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 9.972124238081888e-05, "log_rms_mae_db": 2.161588538694025, "rms_envelope_nrmse": 0.15808650305133437, "envelope_correlation": 0.98378842712466, "rms_gain_error_db": -0.013067016539867554, "reference_mean_power": 0.00028254365881671625, "output_mean_power": 0.0002816948214150563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 3.4051840450675846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002046408997048573, "output_power": 0.0002048413657833426, "reference_active": true, "delta_db": 0.0042522632107863155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.521356834551269e-05, "output_power": 3.5370880246696024e-05, "reference_active": true, "delta_db": -1.0662291589132058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018106896586635e-06, "output_power": 4.444654721001666e-06, "reference_active": true, "delta_db": -2.5623381308202706}], "active_band_count": 3, "active_band_mae_db": 1.2109398509814209}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 8.26289228894579e-05, "log_rms_mae_db": 1.3420339388474913, "rms_envelope_nrmse": 0.07741510450011783, "envelope_correlation": 0.9959129970695889, "rms_gain_error_db": -0.013085708388457035, "reference_mean_power": 0.0002861487173387815, "output_mean_power": 0.00028528782149920396}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.083140770966499e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019461067978585587, "output_power": 0.00019212497903374446, "reference_active": true, "delta_db": -0.055828365531331456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0239988981192927e-05, "output_power": 1.563875026649669e-05, "reference_active": true, "delta_db": -1.1200822723386519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.640717194960459e-07, "output_power": 3.7957555619875537e-07, "reference_active": true, "delta_db": -3.0383588840572653}], "active_band_count": 3, "active_band_mae_db": 1.4047565073090829}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 4.3426949519701344e-05, "log_rms_mae_db": 2.1903862906776537, "rms_envelope_nrmse": 0.1701317899002066, "envelope_correlation": 0.9816037088574515, "rms_gain_error_db": -0.03914114095510304, "reference_mean_power": 4.5529138246552346e-05, "output_mean_power": 4.512064675707953e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 6.179039674292901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3063171222465546e-05, "output_power": 3.1597660723577354e-05, "reference_active": true, "delta_db": -0.19689574581747027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2207009907392542e-05, "output_power": 1.0213177736749489e-05, "reference_active": true, "delta_db": -0.7744840718800832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2513877801441335e-06, "output_power": 1.3139782832372153e-06, "reference_active": true, "delta_db": -2.3386211705413946}], "active_band_count": 3, "active_band_mae_db": 1.103333662746316}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 3.683581645514392e-05, "log_rms_mae_db": 1.4307340517813043, "rms_envelope_nrmse": 0.08716009848313448, "envelope_correlation": 0.9949102824631952, "rms_gain_error_db": -0.03916620200226917, "reference_mean_power": 4.610980557969544e-05, "output_mean_power": 4.569584060162132e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 5.621286439624602e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.226124623783574e-05, "output_power": 3.039078029450701e-05, "reference_active": true, "delta_db": -0.25939289088628836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.247687157348319e-06, "output_power": 5.775275229422386e-06, "reference_active": true, "delta_db": -0.9862675296385852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.30977962238882e-07, "output_power": 1.0165713244594421e-07, "reference_active": true, "delta_db": -3.5643269062372966}], "active_band_count": 3, "active_band_mae_db": 1.6033291089207236}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7977447174386263, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5c1678c38ba37bf32c3cd456b1901f211b292b796296d933e6c38e2857f0e1de", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.975157678574526, "standard_deviation": 0.007371396768794969, "minimum": 0.9602855265588298, "maximum": 0.98378842712466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.13187350801436, "standard_deviation": 0.027205017806675415, "minimum": 2.098073064479112, "maximum": 2.16376568101994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.912078977164134, "standard_deviation": 0.5967309215870245, "minimum": 0.2570389981930806, "maximum": 1.9590102132696525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1948999710615405, "standard_deviation": 0.029122854495584254, "minimum": 0.15808650305133437, "maximum": 0.2505717503585725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9913965224750552, "standard_deviation": 0.0027289625742981266, "minimum": 0.9877184701385299, "maximum": 0.9959129970695889, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4132106177020902, "standard_deviation": 0.036661832261541336, "minimum": 1.3420339388474913, "maximum": 1.470273288926264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9372759012146752, "standard_deviation": 0.7682135536094843, "minimum": 0.21345975779347695, "maximum": 2.464507128781641, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11359981766523405, "standard_deviation": 0.022215731223145095, "minimum": 0.07741510450011783, "maximum": 0.14523631500282302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9800975369566777, "standard_deviation": 0.002533433487061204, "minimum": 0.9745858258289096, "maximum": 0.982409349968953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.124025797021348, "standard_deviation": 0.0663497302521707, "minimum": 2.041189800121792, "maximum": 2.22019641340001, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2142893318619516, "standard_deviation": 0.4122463809036711, "minimum": 0.6371941481755216, "maximum": 1.9749848940892403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17899116947316307, "standard_deviation": 0.01115073191685546, "minimum": 0.16682433158772664, "maximum": 0.20297798329719743, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993600203231865, "standard_deviation": 0.001324680823462016, "minimum": 0.9907094970637478, "maximum": 0.9949102824631952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3961951702013862, "standard_deviation": 0.05220852630026785, "minimum": 1.3422069679573798, "maximum": 1.4984556007683538, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4220887142229715, "standard_deviation": 0.4524207275175482, "minimum": 0.9022812816536288, "maximum": 2.025046848382576, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10106106339276585, "standard_deviation": 0.011510546464288562, "minimum": 0.08716009848313448, "maximum": 0.1271313681352413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9906683184383653, "standard_deviation": 0.00211218124257918, "minimum": 0.9877547538717634, "maximum": 0.9937824464452882, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3467737655163665, "standard_deviation": 0.05114168349556517, "minimum": 2.2895601571419224, "maximum": 2.419123048633604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.783987324230425, "standard_deviation": 0.2175546132387936, "minimum": 0.4323838343063924, "maximum": 0.9800975232298175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11683245095928818, "standard_deviation": 0.012789401796347539, "minimum": 0.0975164242739657, "maximum": 0.13364946069341785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964271042152695, "standard_deviation": 0.0010576409328579928, "minimum": 0.9944887485388645, "maximum": 0.9976764635005674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5969049560186988, "standard_deviation": 0.06502036923450821, "minimum": 1.5211261386103268, "maximum": 1.737829448959928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8887309117054314, "standard_deviation": 0.2864209073778399, "minimum": 0.35778086487102484, "maximum": 1.1299571520940062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07079928129504104, "standard_deviation": 0.009793167828852435, "minimum": 0.05914149392728609, "maximum": 0.08761928461225002, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2622433377132306, "standard_deviation": 0.10585964829959697, "minimum": 0.13283335912154157, "maximum": 0.4211378035243951, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.878108690333016, "standard_deviation": 0.009567762375873692, "minimum": 5.861126188273123, "maximum": 5.889466445146662, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7436442401137842, "standard_deviation": 0.14612992502303312, "minimum": 0.5469662966946154, "maximum": 0.9779252990359312, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.019048716729776705, "standard_deviation": 0.5989671632350899, "minimum": -0.8251139393909134, "maximum": 1.2307389732156206, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.946676627702693, "standard_deviation": 0.03362045107889595, "minimum": 0.8660083882984441, "maximum": 0.9707118949960315, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7379673044071569, "standard_deviation": 0.10784982617387964, "minimum": 1.5970125062919474, "maximum": 1.8747216974259653, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.2743011884335784, "standard_deviation": 0.7971999257643156, "minimum": 0.5716514336638748, "maximum": 2.8283713018937067, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2551311883752505, "standard_deviation": 0.07494574812117276, "minimum": 0.18598693728751145, "maximum": 0.4062798385038208, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9793607377912129, "standard_deviation": 0.015789755288426955, "minimum": 0.9459002123765855, "maximum": 0.9916818053627742, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9707975711919035, "standard_deviation": 0.10363416641641952, "minimum": 0.805767572532643, "maximum": 1.140673872010934, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.471728150099656, "standard_deviation": 0.9622947170516999, "minimum": 0.5974859195140522, "maximum": 3.1893862698254574, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.15541160617357427, "standard_deviation": 0.06665814554381273, "minimum": 0.099678667134272, "maximum": 0.26987735449619504, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2300-131720-0022", "language": "en", "style": "normal", "speaker": "librispeech_2300", "source_utterance_id": "2300-131720-0022", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2300-131720-0022.wav", "source_file_sha256": "d1770cf4e456b9411242e6393fe2fdc6b26846c13b4651d7fa54228f5c4fb888", "decoded_pcm_sha256": "b7f1420236bac4a5c404bc851efe87bc6142a76f2a2935e40197b946d8182998", "original_source_sha256": "93212060c39673d7a3ee80b52a8f2612e70e3c121fd588c690e3d51be10e6963", "corpus_entry_id": "librispeech_2300-131720-0022"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.092949390411377, "predicted_mos": 4.092949390411377, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "da124c3f2b63d741f237fa8475795c16d790bec46f1fb7b67c7b8f66670c478c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.6488731190002, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2300-131720-0022", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "da124c3f2b63d741f237fa8475795c16d790bec46f1fb7b67c7b8f66670c478c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "ba50eb8b1e006b7341f273bf21fa396079a09996fabc4f27cde2993eeda0f3d7"}, "rows": [{"case": "librispeech_2300-131720-0022", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.763442039489746, "predicted_mos": 2.763442039489746, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7a894e0665a839ad10f54cfd7ba7ee5935439e1c77100038fa911c93bd70ee33", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.103265180998278, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2300-131720-0022", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7a894e0665a839ad10f54cfd7ba7ee5935439e1c77100038fa911c93bd70ee33", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0417502573875692, "全帯域_10ms_rms_envelope_nrmse": 0.08382373836796746, "全帯域_10ms_envelope_correlation": 0.991147591269651, "全帯域_10ms_modulation_mae_db": 0.20299708711793696, "全帯域_40ms_log_rms_mae_db": 0.6241652517656205, "全帯域_40ms_rms_envelope_nrmse": 0.04154620051249543, "全帯域_40ms_envelope_correlation": 0.997658549299896, "全帯域_40ms_modulation_mae_db": 0.28135862563355074, "300–1000 Hz_10ms_log_rms_mae_db": 1.4797185431828028, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12268940679621876, "300–1000 Hz_10ms_envelope_correlation": 0.9871572660373104, "300–1000 Hz_10ms_modulation_mae_db": 0.4219753846184928, "300–1000 Hz_40ms_log_rms_mae_db": 1.0227586658420913, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0707769998154564, "300–1000 Hz_40ms_envelope_correlation": 0.9952840079845986, "300–1000 Hz_40ms_modulation_mae_db": 0.7741158242437997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9982725494735623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21218837574116334, "1000–3000 Hz_10ms_envelope_correlation": 0.9647966930282954, "1000–3000 Hz_10ms_modulation_mae_db": 0.6876418960195853, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3836830116205376, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12460660797162146, "1000–3000 Hz_40ms_envelope_correlation": 0.987865270780557, "1000–3000 Hz_40ms_modulation_mae_db": 1.009350406942252, "3000–6000 Hz_10ms_log_rms_mae_db": 2.112544840040399, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3344645188777203, "3000–6000 Hz_10ms_envelope_correlation": 0.9412456167804539, "3000–6000 Hz_10ms_modulation_mae_db": 0.6964795486766748, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6408192809398694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2657283094342505, "3000–6000 Hz_40ms_envelope_correlation": 0.9627144580512698, "3000–6000 Hz_40ms_modulation_mae_db": 0.871411832777555, "mr_stft_db": 5.910808071307574, "band_mae_db": 0.3478893490755343, "rms_gain_error_db": -0.06246656175035299, "output_peak": 0.6635848481139347}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0330264003535305, "全帯域_10ms_rms_envelope_nrmse": 0.09189648567536614, "全帯域_10ms_envelope_correlation": 0.9895081125735843, "全帯域_10ms_modulation_mae_db": 0.3215596084304329, "全帯域_40ms_log_rms_mae_db": 0.5882256153499827, "全帯域_40ms_rms_envelope_nrmse": 0.05337413761607973, "全帯域_40ms_envelope_correlation": 0.9963822707657006, "全帯域_40ms_modulation_mae_db": 0.19673014000201391, "300–1000 Hz_10ms_log_rms_mae_db": 1.5999601838367987, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15613682882509414, "300–1000 Hz_10ms_envelope_correlation": 0.9796255138040488, "300–1000 Hz_10ms_modulation_mae_db": 0.6916982196484877, "300–1000 Hz_40ms_log_rms_mae_db": 1.1208502576525137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09772268739384393, "300–1000 Hz_40ms_envelope_correlation": 0.9920275516560046, "300–1000 Hz_40ms_modulation_mae_db": 0.880857152262319, "1000–3000 Hz_10ms_log_rms_mae_db": 1.967053883249205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22808744895531657, "1000–3000 Hz_10ms_envelope_correlation": 0.959476611655421, "1000–3000 Hz_10ms_modulation_mae_db": 1.0167421716585094, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3975039552083233, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14445553675800463, "1000–3000 Hz_40ms_envelope_correlation": 0.982244578198666, "1000–3000 Hz_40ms_modulation_mae_db": 0.7876509588175805, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0808949721579837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3167630557766564, "3000–6000 Hz_10ms_envelope_correlation": 0.9403237055672812, "3000–6000 Hz_10ms_modulation_mae_db": 0.31881178805593, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5871660433167296, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24348108055326187, "3000–6000 Hz_40ms_envelope_correlation": 0.9634789469279671, "3000–6000 Hz_40ms_modulation_mae_db": 0.3180232005296888, "mr_stft_db": 5.922776811270267, "band_mae_db": 0.3779159746859364, "rms_gain_error_db": -0.1426005907967488, "output_peak": 0.6666055275275578}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0217543008599383, "全帯域_10ms_rms_envelope_nrmse": 0.08140736534222127, "全帯域_10ms_envelope_correlation": 0.991687786808874, "全帯域_10ms_modulation_mae_db": 0.4227681126255007, "全帯域_40ms_log_rms_mae_db": 0.5862656261101648, "全帯域_40ms_rms_envelope_nrmse": 0.04068811129830565, "全帯域_40ms_envelope_correlation": 0.9978121280366835, "全帯域_40ms_modulation_mae_db": 0.6421448523499199, "300–1000 Hz_10ms_log_rms_mae_db": 1.5197196104463568, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16113962088924494, "300–1000 Hz_10ms_envelope_correlation": 0.9779773533843742, "300–1000 Hz_10ms_modulation_mae_db": 0.6719653082909339, "300–1000 Hz_40ms_log_rms_mae_db": 1.0772982275440748, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0928111493044087, "300–1000 Hz_40ms_envelope_correlation": 0.992600023847519, "300–1000 Hz_40ms_modulation_mae_db": 1.0335232233035587, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9719142590011234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23965721293971887, "1000–3000 Hz_10ms_envelope_correlation": 0.9546054111680383, "1000–3000 Hz_10ms_modulation_mae_db": 0.711699816400298, "1000–3000 Hz_40ms_log_rms_mae_db": 1.366660116074818, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.152975555870934, "1000–3000 Hz_40ms_envelope_correlation": 0.9805342390157815, "1000–3000 Hz_40ms_modulation_mae_db": 0.880692646712494, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1004482057196774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3105938124394905, "3000–6000 Hz_10ms_envelope_correlation": 0.9454237377107317, "3000–6000 Hz_10ms_modulation_mae_db": 0.3867758567157056, "3000–6000 Hz_40ms_log_rms_mae_db": 1.539239483233614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23519715118697682, "3000–6000 Hz_40ms_envelope_correlation": 0.9679476841158047, "3000–6000 Hz_40ms_modulation_mae_db": 0.4001982232931756, "mr_stft_db": 5.925085105403561, "band_mae_db": 0.3522799997927786, "rms_gain_error_db": -0.08508435537953814, "output_peak": 0.6010667132541003}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9758807350203711, "全帯域_10ms_rms_envelope_nrmse": 0.08872021994448966, "全帯域_10ms_envelope_correlation": 0.9900599795492349, "全帯域_10ms_modulation_mae_db": 0.5388751761346556, "全帯域_40ms_log_rms_mae_db": 0.5743719716149414, "全帯域_40ms_rms_envelope_nrmse": 0.05796651476843911, "全帯域_40ms_envelope_correlation": 0.9953353384708573, "全帯域_40ms_modulation_mae_db": 0.5328325794764893, "300–1000 Hz_10ms_log_rms_mae_db": 1.5112215711618013, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13452817883416082, "300–1000 Hz_10ms_envelope_correlation": 0.9845096352175554, "300–1000 Hz_10ms_modulation_mae_db": 0.5209946194787222, "300–1000 Hz_40ms_log_rms_mae_db": 1.0591959945441765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08043182040608275, "300–1000 Hz_40ms_envelope_correlation": 0.9939461286368244, "300–1000 Hz_40ms_modulation_mae_db": 0.9456907312220976, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9561452618635355, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22523401713486674, "1000–3000 Hz_10ms_envelope_correlation": 0.96001904260317, "1000–3000 Hz_10ms_modulation_mae_db": 0.5666638384095686, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3589056915249957, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13243330306832776, "1000–3000 Hz_40ms_envelope_correlation": 0.9853025988277974, "1000–3000 Hz_40ms_modulation_mae_db": 0.9813468117566743, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1185241141288587, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30488547410621736, "3000–6000 Hz_10ms_envelope_correlation": 0.9450801153123839, "3000–6000 Hz_10ms_modulation_mae_db": 0.3268332802045271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6058916978315732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24011252815968043, "3000–6000 Hz_40ms_envelope_correlation": 0.9647689309741478, "3000–6000 Hz_40ms_modulation_mae_db": 0.4963021340754387, "mr_stft_db": 5.920572106848908, "band_mae_db": 0.30930039937285664, "rms_gain_error_db": -0.031624711388652194, "output_peak": 0.5858646206871734}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.008323846709179, "全帯域_10ms_rms_envelope_nrmse": 0.09002829300247937, "全帯域_10ms_envelope_correlation": 0.9898950023513179, "全帯域_10ms_modulation_mae_db": 0.1312005709902162, "全帯域_40ms_log_rms_mae_db": 0.5552745664511072, "全帯域_40ms_rms_envelope_nrmse": 0.04949762936929161, "全帯域_40ms_envelope_correlation": 0.9966532326925562, "全帯域_40ms_modulation_mae_db": 0.17102732749964633, "300–1000 Hz_10ms_log_rms_mae_db": 1.5694653858273007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1637578552413451, "300–1000 Hz_10ms_envelope_correlation": 0.9770629945297339, "300–1000 Hz_10ms_modulation_mae_db": 0.569048907266124, "300–1000 Hz_40ms_log_rms_mae_db": 1.0870984399706807, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10367292142193318, "300–1000 Hz_40ms_envelope_correlation": 0.9899517412140605, "300–1000 Hz_40ms_modulation_mae_db": 0.3816897902369103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9728870215745853, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22693822196037763, "1000–3000 Hz_10ms_envelope_correlation": 0.9600666521520791, "1000–3000 Hz_10ms_modulation_mae_db": 0.7611893340555159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.410715068271149, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14568202451389897, "1000–3000 Hz_40ms_envelope_correlation": 0.982021193666622, "1000–3000 Hz_40ms_modulation_mae_db": 0.7645648287709923, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1154785186445495, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3002135755318034, "3000–6000 Hz_10ms_envelope_correlation": 0.9490421141241003, "3000–6000 Hz_10ms_modulation_mae_db": 0.2695523033178036, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5861073755545727, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23165282954613836, "3000–6000 Hz_40ms_envelope_correlation": 0.9689203478198529, "3000–6000 Hz_40ms_modulation_mae_db": 0.1613713587339455, "mr_stft_db": 5.918736820634646, "band_mae_db": 0.3288935398692066, "rms_gain_error_db": -0.017327064114027655, "output_peak": 0.7264022819729882}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0080824925677727, "全帯域_10ms_rms_envelope_nrmse": 0.0871910326292576, "全帯域_10ms_envelope_correlation": 0.990413283370126, "全帯域_10ms_modulation_mae_db": 0.35082176905000645, "全帯域_40ms_log_rms_mae_db": 0.5666118194346887, "全帯域_40ms_rms_envelope_nrmse": 0.04909960834987639, "全帯域_40ms_envelope_correlation": 0.9967059965600148, "全帯域_40ms_modulation_mae_db": 0.39522576707319557, "300–1000 Hz_10ms_log_rms_mae_db": 1.61441500181249, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15316366858202385, "300–1000 Hz_10ms_envelope_correlation": 0.9798990230355991, "300–1000 Hz_10ms_modulation_mae_db": 0.5936906295599425, "300–1000 Hz_40ms_log_rms_mae_db": 1.1438946434094686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09366973260396269, "300–1000 Hz_40ms_envelope_correlation": 0.991739597074312, "300–1000 Hz_40ms_modulation_mae_db": 0.803561450150606, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0192563499155285, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24748433972788633, "1000–3000 Hz_10ms_envelope_correlation": 0.9515216373234262, "1000–3000 Hz_10ms_modulation_mae_db": 0.24915897582911337, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4531719419512372, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15850163805779807, "1000–3000 Hz_40ms_envelope_correlation": 0.9795505970121563, "1000–3000 Hz_40ms_modulation_mae_db": 0.29352378254557215, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0903428935875636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3043897288515745, "3000–6000 Hz_10ms_envelope_correlation": 0.9490053056054125, "3000–6000 Hz_10ms_modulation_mae_db": 1.1159538743075244, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5350957736327149, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.238025018809094, "3000–6000 Hz_40ms_envelope_correlation": 0.9683379561446207, "3000–6000 Hz_40ms_modulation_mae_db": 1.4633940990849454, "mr_stft_db": 5.928384888563055, "band_mae_db": 0.3477884295869232, "rms_gain_error_db": -0.06651505841122901, "output_peak": 0.8666366911459993}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.049806166100614, "全帯域_10ms_rms_envelope_nrmse": 0.09149316048245956, "全帯域_10ms_envelope_correlation": 0.9895385536386933, "全帯域_10ms_modulation_mae_db": 0.21050036143020115, "全帯域_40ms_log_rms_mae_db": 0.6173215335409236, "全帯域_40ms_rms_envelope_nrmse": 0.04723621607493072, "全帯域_40ms_envelope_correlation": 0.997064960782277, "全帯域_40ms_modulation_mae_db": 0.07129077281826142, "300–1000 Hz_10ms_log_rms_mae_db": 1.6083366357681148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14201526025064978, "300–1000 Hz_10ms_envelope_correlation": 0.9830785418909281, "300–1000 Hz_10ms_modulation_mae_db": 0.5208606868372224, "300–1000 Hz_40ms_log_rms_mae_db": 1.145009581310403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0884550429909207, "300–1000 Hz_40ms_envelope_correlation": 0.9932864569449309, "300–1000 Hz_40ms_modulation_mae_db": 0.8703077219459173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9580885484248567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22141942299272033, "1000–3000 Hz_10ms_envelope_correlation": 0.9613702325939996, "1000–3000 Hz_10ms_modulation_mae_db": 1.0206234504310192, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3374107923373386, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13781689622007645, "1000–3000 Hz_40ms_envelope_correlation": 0.9843443951939251, "1000–3000 Hz_40ms_modulation_mae_db": 0.9186964854367369, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1027527232855086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3225831442736531, "3000–6000 Hz_10ms_envelope_correlation": 0.9365202203482847, "3000–6000 Hz_10ms_modulation_mae_db": 1.1746767273457162, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5904651818044948, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25636286831465427, "3000–6000 Hz_40ms_envelope_correlation": 0.9586170661560689, "3000–6000 Hz_40ms_modulation_mae_db": 1.4281610525247117, "mr_stft_db": 5.921334410016949, "band_mae_db": 0.36783668074217984, "rms_gain_error_db": -0.10464276095930745, "output_peak": 0.6378673842225417}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0542726488042893, "全帯域_10ms_rms_envelope_nrmse": 0.08755849395156828, "全帯域_10ms_envelope_correlation": 0.9905060717933836, "全帯域_10ms_modulation_mae_db": 0.2644144339545646, "全帯域_40ms_log_rms_mae_db": 0.5664782458467235, "全帯域_40ms_rms_envelope_nrmse": 0.04704336088930732, "全帯域_40ms_envelope_correlation": 0.997217330096053, "全帯域_40ms_modulation_mae_db": 0.4033125071158155, "300–1000 Hz_10ms_log_rms_mae_db": 1.5086713589516791, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13485099872185016, "300–1000 Hz_10ms_envelope_correlation": 0.9849438119885437, "300–1000 Hz_10ms_modulation_mae_db": 0.38944169625824915, "300–1000 Hz_40ms_log_rms_mae_db": 0.9835417531304036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07765137603848167, "300–1000 Hz_40ms_envelope_correlation": 0.995201949245222, "300–1000 Hz_40ms_modulation_mae_db": 0.5314503579119156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.991885205683066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2643818853750139, "1000–3000 Hz_10ms_envelope_correlation": 0.9477627408781095, "1000–3000 Hz_10ms_modulation_mae_db": 0.7078780930471167, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3805636988566556, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17052469995250624, "1000–3000 Hz_40ms_envelope_correlation": 0.9764418950014664, "1000–3000 Hz_40ms_modulation_mae_db": 0.5443536156902941, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1323408499374215, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3115844668122891, "3000–6000 Hz_10ms_envelope_correlation": 0.9408678663307056, "3000–6000 Hz_10ms_modulation_mae_db": 1.3680480978165228, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6270007573558574, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24117333033349483, "3000–6000 Hz_40ms_envelope_correlation": 0.9634671955801454, "3000–6000 Hz_40ms_modulation_mae_db": 1.4184669120066904, "mr_stft_db": 5.920425575003494, "band_mae_db": 0.3061097808307536, "rms_gain_error_db": -0.13622600334950918, "output_peak": 0.6366499286698541}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.931007239987484, "spectral_convergence": 0.2801279007334203, "spectral_convergence_status": "ok", "amplitude_floor": 2.1903561449018732e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6492954692891235, "spectral_convergence": 0.2777750740029396, "spectral_convergence_status": "ok", "amplitude_floor": 2.0456135368211886e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.152121504646114, "spectral_convergence": 0.29158745445521955, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394418151882858e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0025307845601282375, "output_power": 0.0024605136408113203, "delta_db": -0.12229399184283782, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026710985360254404, "output_power": 0.0026313747247331144, "delta_db": -0.06507210485129722, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004653266401104453, "output_power": 0.0004907835715624274, "delta_db": 0.23132100016824306, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012444324831517497, "output_power": 0.00011041278090719858, "delta_db": -0.5195199052075323, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00016257063671523992, "output_power": 0.0001655713626160185, "delta_db": 0.07943116278681064, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.589634923518972e-05, "output_power": 4.5921923921535825e-05, "delta_db": 1.0696979295964844, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3478893490755343, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06246656175035299, "pitch": null, "output_rms_dbfs": -22.288111356192314, "output_peak": 0.6635848481139347, "reference_rms_dbfs": -22.225644794441955, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.910808071307574}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031508938028130336, "log_rms_mae_db": 1.0417502573875692, "rms_envelope_nrmse": 0.08382373836796746, "envelope_correlation": 0.991147591269651, "rms_gain_error_db": -0.06244265259724279, "reference_mean_power": 0.0059947180836205155, "output_mean_power": 0.00590914298567642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.905621652404067e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011843610245185533, "output_power": 0.0012128070438471882, "reference_active": true, "delta_db": 0.10307603633226078}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002763334576078544, "output_power": 0.0002571220155289491, "reference_active": true, "delta_db": -0.31294207362752136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0009701982183084e-05, "output_power": 5.2281926431539905e-05, "reference_active": true, "delta_db": 0.19297315139402874}], "active_band_count": 3, "active_band_mae_db": 0.20299708711793696}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002456604014165908, "log_rms_mae_db": 0.6241652517656205, "rms_envelope_nrmse": 0.04154620051249543, "envelope_correlation": 0.997658549299896, "rms_gain_error_db": -0.06233084769434954, "reference_mean_power": 0.006022035579915752, "output_mean_power": 0.005936223342792207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5974133279358877e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001059507007170014, "output_power": 0.0010936285136858516, "reference_active": true, "delta_db": 0.13765991560439447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014348331799351297, "output_power": 0.00013275870421458873, "reference_active": true, "delta_db": -0.3373840615840428}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4072600706208148e-06, "output_power": 3.129695560652532e-06, "reference_active": true, "delta_db": -0.36903189971221495}], "active_band_count": 3, "active_band_mae_db": 0.28135862563355074}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026889287764829997, "log_rms_mae_db": 1.4797185431828028, "rms_envelope_nrmse": 0.12268940679621876, "envelope_correlation": 0.9871572660373104, "rms_gain_error_db": -0.04119645632080362, "reference_mean_power": 0.0016839373335741825, "output_mean_power": 0.0016680393047620267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1138357532326787e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006324020549599114, "output_power": 0.0006348747162986016, "reference_active": true, "delta_db": 0.016947592855955228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002190350723697022, "output_power": 0.00019347263127353375, "reference_active": true, "delta_db": -0.5389412237545955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1194141233781235e-05, "output_power": 2.6489227936596555e-05, "reference_active": true, "delta_db": -0.7100373372449276}], "active_band_count": 3, "active_band_mae_db": 0.4219753846184928}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001970017370857221, "log_rms_mae_db": 1.0227586658420913, "rms_envelope_nrmse": 0.0707769998154564, "envelope_correlation": 0.9952840079845986, "rms_gain_error_db": -0.04121415116566939, "reference_mean_power": 0.001692399182790771, "output_mean_power": 0.0016764144354007406}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.64481232375879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970882254118313, "output_power": 0.0006029004504422341, "reference_active": true, "delta_db": 0.042071013649516195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011301601543605377, "output_power": 0.00010153937155037393, "reference_active": true, "delta_db": -0.4650552033921599}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.966963236644965e-06, "output_power": 1.9533948548821374e-06, "reference_active": true, "delta_db": -1.815221255689723}], "active_band_count": 3, "active_band_mae_db": 0.7741158242437997}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.658803471872848e-05, "log_rms_mae_db": 1.9982725494735623, "rms_envelope_nrmse": 0.21218837574116334, "envelope_correlation": 0.9647966930282954, "rms_gain_error_db": -0.3572486400334818, "reference_mean_power": 0.00020257298157917204, "output_mean_power": 0.00018657637733091262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0818748871802829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.592086384233793e-05, "output_power": 5.6114682711547063e-05, "reference_active": true, "delta_db": -0.6994637818657679}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.925318155530915e-05, "output_power": 1.465645618501692e-05, "reference_active": true, "delta_db": -1.184735321614506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.819375654274508e-06, "output_power": 3.6653860023488792e-06, "reference_active": true, "delta_db": -0.17872658457848173}], "active_band_count": 3, "active_band_mae_db": 0.6876418960195853}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.479997453028037e-05, "log_rms_mae_db": 1.3836830116205376, "rms_envelope_nrmse": 0.12460660797162146, "envelope_correlation": 0.987865270780557, "rms_gain_error_db": -0.3573407553857077, "reference_mean_power": 0.00020359039158760131, "output_mean_power": 0.00018750946822217285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.438814668171717e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.33475246376368e-05, "output_power": 5.419965142937338e-05, "reference_active": true, "delta_db": -0.6773315559957067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.661338887813947e-06, "output_power": 6.798936992663501e-06, "reference_active": true, "delta_db": -1.5259629950634004}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1289641878321164e-07, "output_power": 2.5877655415045744e-07, "reference_active": true, "delta_db": -0.8247566697676494}], "active_band_count": 3, "active_band_mae_db": 1.009350406942252}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.558280881155462e-05, "log_rms_mae_db": 2.112544840040399, "rms_envelope_nrmse": 0.3344645188777203, "envelope_correlation": 0.9412456167804539, "rms_gain_error_db": 0.6357411390222634, "reference_mean_power": 8.701244577979696e-05, "output_mean_power": 0.00010072922442030137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.56825886194788e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4501869786005614e-05, "output_power": 2.457199271313728e-05, "reference_active": true, "delta_db": 0.01241150359258524}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.528335672589267e-06, "output_power": 7.198911712351241e-06, "reference_active": true, "delta_db": 2.013282350144363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7809461115210634e-06, "output_power": 1.7549966806928613e-06, "reference_active": true, "delta_db": -0.0637447922930765}], "active_band_count": 3, "active_band_mae_db": 0.6964795486766748}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.713713309096251e-05, "log_rms_mae_db": 1.6408192809398694, "rms_envelope_nrmse": 0.2657283094342505, "envelope_correlation": 0.9627144580512698, "rms_gain_error_db": 0.6356686005901304, "reference_mean_power": 8.74494743564098e-05, "output_mean_power": 0.00010123345599310207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.348274544921528e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.135032255455729e-05, "output_power": 2.2047348626058138e-05, "reference_active": true, "delta_db": 0.13951928874315606}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4135743662495686e-06, "output_power": 3.863198448428105e-06, "reference_active": true, "delta_db": 2.04286333269331}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6206121669861435e-07, "output_power": 1.790046788596411e-07, "reference_active": true, "delta_db": 0.4318528768961988}], "active_band_count": 3, "active_band_mae_db": 0.871411832777555}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8920875634294372, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "44162cb60730b67c8dec38c8ed00f40fa149f45c6dd0df9a9bf61ff7082895a0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9574523776753174, "standard_deviation": 0.0056565542128790315, "minimum": 0.9477627408781095, "maximum": 0.9647966930282954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9794378848981828, "standard_deviation": 0.021864561183330227, "minimum": 1.9561452618635355, "maximum": 2.0192563499155285, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7151996969813408, "standard_deviation": 0.24681072054157957, "minimum": 0.24915897582911337, "maximum": 1.0206234504310192, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23317386560338296, "standard_deviation": 0.01659224322182819, "minimum": 0.21218837574116334, "maximum": 0.2643818853750139, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9822880959621214, "standard_deviation": 0.003570545856748444, "minimum": 0.9764418950014664, "maximum": 0.987865270780557, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3860767844806319, "standard_deviation": 0.03537384783033615, "minimum": 1.3374107923373386, "maximum": 1.4531719419512372, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7725224420840745, "standard_deviation": 0.24335619726566557, "minimum": 0.29352378254557215, "maximum": 1.009350406942252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14587453280164595, "standard_deviation": 0.014723482003533551, "minimum": 0.12460660797162146, "maximum": 0.17052469995250624, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9434385852224192, "standard_deviation": 0.004440194934875219, "minimum": 0.9365202203482847, "maximum": 0.9490421141241003, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1066658896877453, "standard_deviation": 0.01646033507005, "minimum": 2.0808949721579837, "maximum": 2.1323408499374215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7071414345550505, "standard_deviation": 0.4491548749933526, "minimum": 0.2695523033178036, "maximum": 1.3680480978165228, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3131847220836756, "standard_deviation": 0.011197692739433574, "minimum": 0.3002135755318034, "maximum": 0.3344645188777203, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9647815732212346, "standard_deviation": 0.003494899733381509, "minimum": 0.9586170661560689, "maximum": 0.9689203478198529, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5889731992086782, "standard_deviation": 0.03743617659478142, "minimum": 1.5350957736327149, "maximum": 1.6408192809398694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8196661016282689, "standard_deviation": 0.5492081307781013, "minimum": 0.1613713587339455, "maximum": 1.4633940990849454, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.24396663954219389, "standard_deviation": 0.01142657292863076, "minimum": 0.23165282954613836, "maximum": 0.2657283094342505, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9817817674860118, "standard_deviation": 0.003644015403080707, "minimum": 0.9770629945297339, "maximum": 0.9871572660373104, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.551438536373418, "standard_deviation": 0.05275147923234563, "minimum": 1.4797185431828028, "maximum": 1.61441500181249, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5474594314947718, "standard_deviation": 0.10759295029055892, "minimum": 0.38944169625824915, "maximum": 0.6916982196484877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14603522726757345, "standard_deviation": 0.014705576584049062, "minimum": 0.12268940679621876, "maximum": 0.1637578552413451, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.993004682075434, "standard_deviation": 0.0018137746127798256, "minimum": 0.9899517412140605, "maximum": 0.9952840079845986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0799559454254766, "standard_deviation": 0.05738187485754636, "minimum": 0.9835417531304036, "maximum": 1.145009581310403, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7776495314096405, "standard_deviation": 0.21746218344250498, "minimum": 0.3816897902369103, "maximum": 1.0335232233035587, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08814896624688626, "standard_deviation": 0.011061495466498927, "minimum": 0.0707769998154564, "maximum": 0.10367292142193318, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.34225176924452116, "standard_deviation": 0.025801450576888513, "minimum": 0.3061097808307536, "maximum": 0.3779159746859364, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.921015473631057, "standard_deviation": 0.0051269886186040775, "minimum": 5.910808071307574, "maximum": 5.928384888563055, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6730847494492687, "standard_deviation": 0.08928209645236844, "minimum": 0.5858646206871734, "maximum": 0.8666366911459993, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08081088826867068, "standard_deviation": 0.04545859024134731, "minimum": -0.1426005907967488, "maximum": -0.017327064114027655, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9903445476693581, "standard_deviation": 0.0007664673233155056, "minimum": 0.9895081125735843, "maximum": 0.991687786808874, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0241121059754081, "standard_deviation": 0.026181929768365732, "minimum": 0.9758807350203711, "maximum": 1.0542726488042893, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30539213996668935, "standard_deviation": 0.1320864202737879, "minimum": 0.1312005709902162, "maximum": 0.5388751761346556, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08776484867447618, "standard_deviation": 0.00365138798712398, "minimum": 0.08140736534222127, "maximum": 0.09189648567536614, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968537258380048, "standard_deviation": 0.000786756030460051, "minimum": 0.9953353384708573, "maximum": 0.9978121280366835, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.584839328764269, "standard_deviation": 0.0246927846795389, "minimum": 0.5552745664511072, "maximum": 0.6241652517656205, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33674032149611155, "standard_deviation": 0.19278809491236834, "minimum": 0.07129077281826142, "maximum": 0.6421448523499199, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04830647235984075, "standard_deviation": 0.005701217539405453, "minimum": 0.04068811129830565, "maximum": 0.05796651476843911, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0022", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5912277698516846, "predicted_mos": 2.5912277698516846, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8b7a56c1a026ac770b7d82113457084d4f4d86b08410ff49f9275af0669866e5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.948247199878097, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0804967366563487, "全帯域_10ms_rms_envelope_nrmse": 0.0860126398828003, "全帯域_10ms_envelope_correlation": 0.9906862139779473, "全帯域_10ms_modulation_mae_db": 0.17071919233977065, "全帯域_40ms_log_rms_mae_db": 0.6728536238632572, "全帯域_40ms_rms_envelope_nrmse": 0.04562043110915466, "全帯域_40ms_envelope_correlation": 0.9971563791857768, "全帯域_40ms_modulation_mae_db": 0.3801583020978241, "300–1000 Hz_10ms_log_rms_mae_db": 1.4072173809638402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13318687186666153, "300–1000 Hz_10ms_envelope_correlation": 0.984945192546889, "300–1000 Hz_10ms_modulation_mae_db": 0.38104122043629457, "300–1000 Hz_40ms_log_rms_mae_db": 0.9149107084020788, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0803619091879121, "300–1000 Hz_40ms_envelope_correlation": 0.9939501890310468, "300–1000 Hz_40ms_modulation_mae_db": 0.7307287626418608, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8865027192107415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20535342282797306, "1000–3000 Hz_10ms_envelope_correlation": 0.9684972399693315, "1000–3000 Hz_10ms_modulation_mae_db": 0.7747420770979673, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2383783437902247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12928226644896007, "1000–3000 Hz_40ms_envelope_correlation": 0.9890215397443491, "1000–3000 Hz_40ms_modulation_mae_db": 1.0470892566127816, "3000–6000 Hz_10ms_log_rms_mae_db": 2.025838718864493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3162810538559697, "3000–6000 Hz_10ms_envelope_correlation": 0.9464575831886363, "3000–6000 Hz_10ms_modulation_mae_db": 0.97422772074554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5398116764049492, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24747560283073913, "3000–6000 Hz_40ms_envelope_correlation": 0.9669260691657491, "3000–6000 Hz_40ms_modulation_mae_db": 0.8416843221554982, "mr_stft_db": 5.817323447466947, "band_mae_db": 0.35351502814225594, "rms_gain_error_db": -0.06026799544704467, "output_peak": 0.6385122193124718}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0119656414455158, "全帯域_10ms_rms_envelope_nrmse": 0.09470856918782952, "全帯域_10ms_envelope_correlation": 0.9888228678723306, "全帯域_10ms_modulation_mae_db": 0.2946470184006013, "全帯域_40ms_log_rms_mae_db": 0.5806591301429809, "全帯域_40ms_rms_envelope_nrmse": 0.053200201812444656, "全帯域_40ms_envelope_correlation": 0.9963836990749579, "全帯域_40ms_modulation_mae_db": 0.2255108164274954, "300–1000 Hz_10ms_log_rms_mae_db": 1.5291837026682336, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16231192093763058, "300–1000 Hz_10ms_envelope_correlation": 0.9777277784136416, "300–1000 Hz_10ms_modulation_mae_db": 0.7105949715490345, "300–1000 Hz_40ms_log_rms_mae_db": 1.000221379943645, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09722570357548711, "300–1000 Hz_40ms_envelope_correlation": 0.9918816784731467, "300–1000 Hz_40ms_modulation_mae_db": 0.9336239602211188, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9069962410965993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21415827269254448, "1000–3000 Hz_10ms_envelope_correlation": 0.9640295063718508, "1000–3000 Hz_10ms_modulation_mae_db": 0.9986812395618708, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2856121010510722, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13738856393911383, "1000–3000 Hz_40ms_envelope_correlation": 0.9841636797908302, "1000–3000 Hz_40ms_modulation_mae_db": 0.8137698356612676, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0064408851956417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30630912171897956, "3000–6000 Hz_10ms_envelope_correlation": 0.9437198718304007, "3000–6000 Hz_10ms_modulation_mae_db": 0.492081038632938, "3000–6000 Hz_40ms_log_rms_mae_db": 1.508088971084779, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23446469106474604, "3000–6000 Hz_40ms_envelope_correlation": 0.9658092411166721, "3000–6000 Hz_40ms_modulation_mae_db": 0.6857268996217033, "mr_stft_db": 5.830555124357968, "band_mae_db": 0.3865154378930608, "rms_gain_error_db": -0.139562972888422, "output_peak": 0.6693560971976221}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0212589539848205, "全帯域_10ms_rms_envelope_nrmse": 0.08355284646801511, "全帯域_10ms_envelope_correlation": 0.9912675390573463, "全帯域_10ms_modulation_mae_db": 0.3669281069150241, "全帯域_40ms_log_rms_mae_db": 0.57770896649286, "全帯域_40ms_rms_envelope_nrmse": 0.03953416304255952, "全帯域_40ms_envelope_correlation": 0.9980059273287105, "全帯域_40ms_modulation_mae_db": 0.4905659111553043, "300–1000 Hz_10ms_log_rms_mae_db": 1.4760546430248858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16199540822348904, "300–1000 Hz_10ms_envelope_correlation": 0.9777748635992877, "300–1000 Hz_10ms_modulation_mae_db": 0.7498781381152476, "300–1000 Hz_40ms_log_rms_mae_db": 0.9470270981454239, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09195868781548852, "300–1000 Hz_40ms_envelope_correlation": 0.9928145904502562, "300–1000 Hz_40ms_modulation_mae_db": 1.119122169846866, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8733620607086066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20910507434757433, "1000–3000 Hz_10ms_envelope_correlation": 0.967748842138341, "1000–3000 Hz_10ms_modulation_mae_db": 0.6334980411054867, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2421278851493591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14260083291007344, "1000–3000 Hz_40ms_envelope_correlation": 0.9865089228887969, "1000–3000 Hz_40ms_modulation_mae_db": 0.9989805931050459, "3000–6000 Hz_10ms_log_rms_mae_db": 1.999104860042527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3043325500990618, "3000–6000 Hz_10ms_envelope_correlation": 0.9475402585713116, "3000–6000 Hz_10ms_modulation_mae_db": 0.22594280328584512, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4701464151766888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23117676005642432, "3000–6000 Hz_40ms_envelope_correlation": 0.968947977831409, "3000–6000 Hz_40ms_modulation_mae_db": 0.38061870729751424, "mr_stft_db": 5.835961742305315, "band_mae_db": 0.41357377864068323, "rms_gain_error_db": -0.10176586028553575, "output_peak": 0.5867377939847485}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0028209095622758, "全帯域_10ms_rms_envelope_nrmse": 0.08785346082042417, "全帯域_10ms_envelope_correlation": 0.9902548882727588, "全帯域_10ms_modulation_mae_db": 0.40205458946298184, "全帯域_40ms_log_rms_mae_db": 0.5810903283629804, "全帯域_40ms_rms_envelope_nrmse": 0.05500606425658546, "全帯域_40ms_envelope_correlation": 0.9958145838263079, "全帯域_40ms_modulation_mae_db": 0.2702015089066158, "300–1000 Hz_10ms_log_rms_mae_db": 1.4947194350224104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1362866591583508, "300–1000 Hz_10ms_envelope_correlation": 0.9841101260129386, "300–1000 Hz_10ms_modulation_mae_db": 0.4990345169389904, "300–1000 Hz_40ms_log_rms_mae_db": 0.9950558492163789, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07996493989117376, "300–1000 Hz_40ms_envelope_correlation": 0.9940580006292975, "300–1000 Hz_40ms_modulation_mae_db": 0.9264647336683058, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8505162256320513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20644714706040287, "1000–3000 Hz_10ms_envelope_correlation": 0.968379068236558, "1000–3000 Hz_10ms_modulation_mae_db": 0.6837142228662444, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2128259506087873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1338366871560358, "1000–3000 Hz_40ms_envelope_correlation": 0.9882864458128703, "1000–3000 Hz_40ms_modulation_mae_db": 1.0883754108706445, "3000–6000 Hz_10ms_log_rms_mae_db": 2.048390299883273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29927461305320513, "3000–6000 Hz_10ms_envelope_correlation": 0.9473875282420133, "3000–6000 Hz_10ms_modulation_mae_db": 0.341486713644648, "3000–6000 Hz_40ms_log_rms_mae_db": 1.552713182132495, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23309115965758453, "3000–6000 Hz_40ms_envelope_correlation": 0.9669572518954862, "3000–6000 Hz_40ms_modulation_mae_db": 0.36628354472379154, "mr_stft_db": 5.831588425088873, "band_mae_db": 0.38024190740415426, "rms_gain_error_db": -0.04472038878911935, "output_peak": 0.5842741816503302}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.053589085657165, "全帯域_10ms_rms_envelope_nrmse": 0.09443559194531695, "全帯域_10ms_envelope_correlation": 0.9889001404292046, "全帯域_10ms_modulation_mae_db": 0.059885205882913795, "全帯域_40ms_log_rms_mae_db": 0.5979666106928886, "全帯域_40ms_rms_envelope_nrmse": 0.051600292275916095, "全帯域_40ms_envelope_correlation": 0.9963764468487847, "全帯域_40ms_modulation_mae_db": 0.0771187741157045, "300–1000 Hz_10ms_log_rms_mae_db": 1.4988451332694306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16586440732056634, "300–1000 Hz_10ms_envelope_correlation": 0.9765462573096542, "300–1000 Hz_10ms_modulation_mae_db": 0.61391824966651, "300–1000 Hz_40ms_log_rms_mae_db": 0.9893794147110302, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10964935822743076, "300–1000 Hz_40ms_envelope_correlation": 0.9886712084604447, "300–1000 Hz_40ms_modulation_mae_db": 0.36754375277810575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.868193810669703, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2108809791065595, "1000–3000 Hz_10ms_envelope_correlation": 0.9654272723863887, "1000–3000 Hz_10ms_modulation_mae_db": 0.7546296436864095, "1000–3000 Hz_40ms_log_rms_mae_db": 1.279207510876788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13176200175125613, "1000–3000 Hz_40ms_envelope_correlation": 0.9864754508813298, "1000–3000 Hz_40ms_modulation_mae_db": 0.8248920675922272, "3000–6000 Hz_10ms_log_rms_mae_db": 2.014209132877424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30554979898327683, "3000–6000 Hz_10ms_envelope_correlation": 0.9458745403180794, "3000–6000 Hz_10ms_modulation_mae_db": 0.2434375720432778, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5098194635655748, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23521729908887673, "3000–6000 Hz_40ms_envelope_correlation": 0.9668831170217811, "3000–6000 Hz_40ms_modulation_mae_db": 0.578337683276801, "mr_stft_db": 5.827483751256377, "band_mae_db": 0.3559996587947272, "rms_gain_error_db": -0.019334663027592017, "output_peak": 0.7228262584516734}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.012290611083842, "全帯域_10ms_rms_envelope_nrmse": 0.0836451736133276, "全帯域_10ms_envelope_correlation": 0.9911986068041844, "全帯域_10ms_modulation_mae_db": 0.33729108535088675, "全帯域_40ms_log_rms_mae_db": 0.5764338684876006, "全帯域_40ms_rms_envelope_nrmse": 0.04615095390700253, "全帯域_40ms_envelope_correlation": 0.9971307963307137, "全帯域_40ms_modulation_mae_db": 0.3523945400094462, "300–1000 Hz_10ms_log_rms_mae_db": 1.5743936728901742, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14882780954469696, "300–1000 Hz_10ms_envelope_correlation": 0.9810231957985015, "300–1000 Hz_10ms_modulation_mae_db": 0.6207657863029576, "300–1000 Hz_40ms_log_rms_mae_db": 1.0494989556161294, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0887474837045816, "300–1000 Hz_40ms_envelope_correlation": 0.9925983957947753, "300–1000 Hz_40ms_modulation_mae_db": 0.8497810135468741, "1000–3000 Hz_10ms_log_rms_mae_db": 1.919930739926492, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24062400695287817, "1000–3000 Hz_10ms_envelope_correlation": 0.9554465661142242, "1000–3000 Hz_10ms_modulation_mae_db": 0.14341840071124493, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3259419093411977, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16938945406525258, "1000–3000 Hz_40ms_envelope_correlation": 0.9784828848852539, "1000–3000 Hz_40ms_modulation_mae_db": 0.32753268759580717, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0212012490190383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29725657005207257, "3000–6000 Hz_10ms_envelope_correlation": 0.9502221800442013, "3000–6000 Hz_10ms_modulation_mae_db": 1.0058455447617853, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4765168517338045, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2305530586136178, "3000–6000 Hz_40ms_envelope_correlation": 0.9693467153822865, "3000–6000 Hz_40ms_modulation_mae_db": 1.3866478983303068, "mr_stft_db": 5.836888160292811, "band_mae_db": 0.3980707061012739, "rms_gain_error_db": -0.0781213691005272, "output_peak": 0.773078545706367}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1028793134939323, "全帯域_10ms_rms_envelope_nrmse": 0.08992497139482396, "全帯域_10ms_envelope_correlation": 0.9898930623911542, "全帯域_10ms_modulation_mae_db": 0.19571713351021236, "全帯域_40ms_log_rms_mae_db": 0.6466030700451395, "全帯域_40ms_rms_envelope_nrmse": 0.044546747182045376, "全帯域_40ms_envelope_correlation": 0.9973621347339963, "全帯域_40ms_modulation_mae_db": 0.14077705819558883, "300–1000 Hz_10ms_log_rms_mae_db": 1.5885811754357024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13647288539686978, "300–1000 Hz_10ms_envelope_correlation": 0.9842652008569921, "300–1000 Hz_10ms_modulation_mae_db": 0.5723726093818333, "300–1000 Hz_40ms_log_rms_mae_db": 1.0806070055957477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07759959182605142, "300–1000 Hz_40ms_envelope_correlation": 0.9947726823841307, "300–1000 Hz_40ms_modulation_mae_db": 0.9455243207831264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8455788464356544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21089110273093292, "1000–3000 Hz_10ms_envelope_correlation": 0.9660743920693129, "1000–3000 Hz_10ms_modulation_mae_db": 1.0383171989929942, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2082449945261697, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13454520802470352, "1000–3000 Hz_40ms_envelope_correlation": 0.9870001366521163, "1000–3000 Hz_40ms_modulation_mae_db": 0.9719590768917898, "3000–6000 Hz_10ms_log_rms_mae_db": 2.070422268116421, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3210217834779835, "3000–6000 Hz_10ms_envelope_correlation": 0.9368552308395292, "3000–6000 Hz_10ms_modulation_mae_db": 1.4717835748570351, "3000–6000 Hz_40ms_log_rms_mae_db": 1.606606313270911, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25310838883196324, "3000–6000 Hz_40ms_envelope_correlation": 0.9595957123948334, "3000–6000 Hz_40ms_modulation_mae_db": 1.848317070009364, "mr_stft_db": 5.824329376690063, "band_mae_db": 0.39790678945668717, "rms_gain_error_db": -0.08676488394333777, "output_peak": 0.636386860243157}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1004496618683068, "全帯域_10ms_rms_envelope_nrmse": 0.08718047800384023, "全帯域_10ms_envelope_correlation": 0.9905886076024301, "全帯域_10ms_modulation_mae_db": 0.26551812623976806, "全帯域_40ms_log_rms_mae_db": 0.6128578483762892, "全帯域_40ms_rms_envelope_nrmse": 0.047196253936651916, "全帯域_40ms_envelope_correlation": 0.9971793763664355, "全帯域_40ms_modulation_mae_db": 0.44587582566425926, "300–1000 Hz_10ms_log_rms_mae_db": 1.4795099738654316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12892754374674856, "300–1000 Hz_10ms_envelope_correlation": 0.9861577076253055, "300–1000 Hz_10ms_modulation_mae_db": 0.3946647598477146, "300–1000 Hz_40ms_log_rms_mae_db": 0.9575405426100367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07091663761506553, "300–1000 Hz_40ms_envelope_correlation": 0.9959386960524671, "300–1000 Hz_40ms_modulation_mae_db": 0.6730601227143066, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9039938753692345, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20843675627828312, "1000–3000 Hz_10ms_envelope_correlation": 0.9659310969431252, "1000–3000 Hz_10ms_modulation_mae_db": 0.6567146934229052, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2279859624738099, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13048082478685022, "1000–3000 Hz_40ms_envelope_correlation": 0.9859578696480336, "1000–3000 Hz_40ms_modulation_mae_db": 0.6312894914631992, "3000–6000 Hz_10ms_log_rms_mae_db": 2.050419613952767, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31262274762041403, "3000–6000 Hz_10ms_envelope_correlation": 0.9401436474463798, "3000–6000 Hz_10ms_modulation_mae_db": 1.5953029870754332, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5562264986647365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24545915194343146, "3000–6000 Hz_40ms_envelope_correlation": 0.9620456182047845, "3000–6000 Hz_40ms_modulation_mae_db": 1.7223875602104923, "mr_stft_db": 5.828773587973653, "band_mae_db": 0.3480750132866887, "rms_gain_error_db": -0.1308993190378711, "output_peak": 0.6358129156175711}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.829828290758853, "spectral_convergence": 0.2796754601667017, "spectral_convergence_status": "ok", "amplitude_floor": 2.1903561449018732e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.557353469601518, "spectral_convergence": 0.2837786293612543, "spectral_convergence_status": "ok", "amplitude_floor": 2.0456135368211886e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.064788582040471, "spectral_convergence": 0.2936491926423304, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394418151882858e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0025307845601282375, "output_power": 0.002454868804020957, "delta_db": -0.13226889237001557, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026710985360254404, "output_power": 0.002652671248121384, "delta_db": -0.030064791446780835, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004653266401104453, "output_power": 0.0004905716494987913, "delta_db": 0.22944529640091932, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012444324831517497, "output_power": 0.00010971137007311589, "delta_db": -0.5471970032821128, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00016257063671523992, "output_power": 0.00015485523944945803, "delta_db": -0.2111620259168639, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.589634923518972e-05, "output_power": 4.4889575421130093e-05, "delta_db": 0.9709521594368432, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.35351502814225594, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06026799544704467, "pitch": null, "output_rms_dbfs": -22.285912789889004, "output_peak": 0.6385122193124718, "reference_rms_dbfs": -22.225644794441955, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.817323447466947}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031508938028130336, "log_rms_mae_db": 1.0804967366563487, "rms_envelope_nrmse": 0.0860126398828003, "envelope_correlation": 0.9906862139779473, "rms_gain_error_db": -0.06018936939414393, "reference_mean_power": 0.0059947180836205155, "output_mean_power": 0.005912209666912584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.905621652404067e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011843610245185533, "output_power": 0.001229513647144733, "reference_active": true, "delta_db": 0.1624924661321894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002763334576078544, "output_power": 0.00025684907487123656, "reference_active": true, "delta_db": -0.3175546533966288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0009701982183084e-05, "output_power": 4.96413085195777e-05, "reference_active": true, "delta_db": -0.03211045749049376}], "active_band_count": 3, "active_band_mae_db": 0.17071919233977065}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002456604014165908, "log_rms_mae_db": 0.6728536238632572, "rms_envelope_nrmse": 0.04562043110915466, "envelope_correlation": 0.9971563791857768, "rms_gain_error_db": -0.0600428969810066, "reference_mean_power": 0.006022035579915752, "output_mean_power": 0.005939351488602322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5974133279358877e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001059507007170014, "output_power": 0.0011112095127086077, "reference_active": true, "delta_db": 0.20692117289156142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014348331799351297, "output_power": 0.00013427718852436097, "reference_active": true, "delta_db": -0.2879917152530882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4072600706208148e-06, "output_power": 2.9366289916090707e-06, "reference_active": true, "delta_db": -0.6455620181488226}], "active_band_count": 3, "active_band_mae_db": 0.3801583020978241}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026889287764829997, "log_rms_mae_db": 1.4072173809638402, "rms_envelope_nrmse": 0.13318687186666153, "envelope_correlation": 0.984945192546889, "rms_gain_error_db": -0.0016087347116765683, "reference_mean_power": 0.0016839373335741825, "output_mean_power": 0.0016833136768450308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1138357532326787e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006324020549599114, "output_power": 0.00064004300616685, "reference_active": true, "delta_db": 0.05215883867255797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002190350723697022, "output_power": 0.00019326170837620057, "reference_active": true, "delta_db": -0.5436784632551984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1194141233781235e-05, "output_power": 2.7500741142420497e-05, "reference_active": true, "delta_db": -0.5472863593811274}], "active_band_count": 3, "active_band_mae_db": 0.38104122043629457}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001970017370857221, "log_rms_mae_db": 0.9149107084020788, "rms_envelope_nrmse": 0.0803619091879121, "envelope_correlation": 0.9939501890310468, "rms_gain_error_db": -0.0016235430626362773, "reference_mean_power": 0.001692399182790771, "output_mean_power": 0.0016917666236537312}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.64481232375879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970882254118313, "output_power": 0.0006073300113932585, "reference_active": true, "delta_db": 0.07386235470942495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011301601543605377, "output_power": 0.00010185147334044342, "reference_active": true, "delta_db": -0.45172675769153486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.966963236644965e-06, "output_power": 2.021400866341234e-06, "reference_active": true, "delta_db": -1.6665971755246227}], "active_band_count": 3, "active_band_mae_db": 0.7307287626418608}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.658803471872848e-05, "log_rms_mae_db": 1.8865027192107415, "rms_envelope_nrmse": 0.20535342282797306, "envelope_correlation": 0.9684972399693315, "rms_gain_error_db": -0.5868488878698936, "reference_mean_power": 0.00020257298157917204, "output_mean_power": 0.0001769687696538281}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0818748871802829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.592086384233793e-05, "output_power": 5.6113229373635086e-05, "reference_active": true, "delta_db": -0.6995762630879208}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.925318155530915e-05, "output_power": 1.4439708365788009e-05, "reference_active": true, "delta_db": -1.2494408418062364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.819375654274508e-06, "output_power": 3.5032526792255907e-06, "reference_active": true, "delta_db": -0.3752091263997445}], "active_band_count": 3, "active_band_mae_db": 0.7747420770979673}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.479997453028037e-05, "log_rms_mae_db": 1.2383783437902247, "rms_envelope_nrmse": 0.12928226644896007, "envelope_correlation": 0.9890215397443491, "rms_gain_error_db": -0.5869415202122396, "reference_mean_power": 0.00020359039158760131, "output_mean_power": 0.00017785379057220986}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.438814668171717e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.33475246376368e-05, "output_power": 5.425958067938088e-05, "reference_active": true, "delta_db": -0.6725321590747434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.661338887813947e-06, "output_power": 6.56676506982714e-06, "reference_active": true, "delta_db": -1.6768583658494118}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1289641878321164e-07, "output_power": 2.607431260746215e-07, "reference_active": true, "delta_db": -0.7918772449141899}], "active_band_count": 3, "active_band_mae_db": 1.0470892566127816}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.558280881155462e-05, "log_rms_mae_db": 2.025838718864493, "rms_envelope_nrmse": 0.3162810538559697, "envelope_correlation": 0.9464575831886363, "rms_gain_error_db": 0.5469433254133336, "reference_mean_power": 8.701244577979696e-05, "output_mean_power": 9.869058167666744e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.56825886194788e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4501869786005614e-05, "output_power": 2.432861279071739e-05, "reference_active": true, "delta_db": -0.030818811854354422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.528335672589267e-06, "output_power": 6.730564828668178e-06, "reference_active": true, "delta_db": 1.721128993418737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7809461115210634e-06, "output_power": 1.3601200505880914e-06, "reference_active": true, "delta_db": -1.1707353569635282}], "active_band_count": 3, "active_band_mae_db": 0.97422772074554}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.713713309096251e-05, "log_rms_mae_db": 1.5398116764049492, "rms_envelope_nrmse": 0.24747560283073913, "envelope_correlation": 0.9669260691657491, "rms_gain_error_db": 0.5468775310674869, "reference_mean_power": 8.74494743564098e-05, "output_mean_power": 9.918476220914322e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.348274544921528e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.135032255455729e-05, "output_power": 2.1589059448842865e-05, "reference_active": true, "delta_db": 0.048292816216146006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4135743662495686e-06, "output_power": 3.540715437676761e-06, "reference_active": true, "delta_db": 1.6643033982569888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6206121669861435e-07, "output_power": 1.3441056909875708e-07, "reference_active": true, "delta_db": -0.8124567519933595}], "active_band_count": 3, "active_band_mae_db": 0.8416843221554982}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8930093308258282, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6e660986740441d8794914ad3f2cd3e8b627f87363fcc20200ee985969a8d486", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9651917480286416, "standard_deviation": 0.004231253826678292, "minimum": 0.9554465661142242, "maximum": 0.9684972399693315, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8818843148811353, "standard_deviation": 0.02712511857627863, "minimum": 1.8455788464356544, "maximum": 1.919930739926492, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7104644396806404, "standard_deviation": 0.2745667188192607, "minimum": 0.14341840071124493, "maximum": 1.0383171989929942, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21323709524964357, "standard_deviation": 0.011403752533640003, "minimum": 0.20535342282797306, "maximum": 0.24062400695287817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9857371162879476, "standard_deviation": 0.003276994921507133, "minimum": 0.9784828848852539, "maximum": 0.9890215397443491, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.252540582227176, "standard_deviation": 0.0407849393186752, "minimum": 1.2082449945261697, "maximum": 1.3259419093411977, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8379860524740954, "standard_deviation": 0.2547634695191393, "minimum": 0.32753268759580717, "maximum": 1.0883754108706445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1386607298852807, "standard_deviation": 0.013115231548093445, "minimum": 0.12928226644896007, "maximum": 0.16938945406525258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9447751050600689, "standard_deviation": 0.0043665820191588385, "minimum": 0.9368552308395292, "maximum": 0.9502221800442013, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0295033784939482, "standard_deviation": 0.024613520638182663, "minimum": 1.999104860042527, "maximum": 2.070422268116421, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.793763494380813, "standard_deviation": 0.5477966866718648, "minimum": 0.22594280328584512, "maximum": 1.5953029870754332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30783102985762034, "standard_deviation": 0.00822149181613709, "minimum": 0.29725657005207257, "maximum": 0.3210217834779835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9658139628766252, "standard_deviation": 0.0033563682260791482, "minimum": 0.9595957123948334, "maximum": 0.9693467153822865, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5274911715042423, "standard_deviation": 0.04543190562625829, "minimum": 1.4701464151766888, "maximum": 1.606606313270911, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9762504607031839, "standard_deviation": 0.5944181052592127, "minimum": 0.36628354472379154, "maximum": 1.848317070009364, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2388182640109229, "standard_deviation": 0.008574755709655808, "minimum": 0.2305530586136178, "maximum": 0.25310838883196324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9815687902704013, "standard_deviation": 0.0037958254244386466, "minimum": 0.9765462573096542, "maximum": 0.9861577076253055, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5060631396425137, "standard_deviation": 0.05802670835918069, "minimum": 1.4072173809638402, "maximum": 1.5885811754357024, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5677837815298228, "standard_deviation": 0.13526659982091174, "minimum": 0.38104122043629457, "maximum": 0.7498781381152476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14673418827437668, "standard_deviation": 0.014933044338933726, "minimum": 0.12892754374674856, "maximum": 0.16586440732056634, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9930856801594456, "standard_deviation": 0.0022015116560305183, "minimum": 0.9886712084604447, "maximum": 0.9959386960524671, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9917801192800588, "standard_deviation": 0.0539755451386639, "minimum": 0.9149107084020788, "maximum": 1.0806070055957477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8182311045250705, "standard_deviation": 0.228227576053231, "minimum": 0.36754375277810575, "maximum": 1.119122169846866, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08705303898039884, "standard_deviation": 0.012454355294527191, "minimum": 0.07091663761506553, "maximum": 0.10964935822743076, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.37923728996494144, "standard_deviation": 0.02422788436647897, "minimum": 0.3480750132866887, "maximum": 0.41357377864068323, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.829112951929002, "standard_deviation": 0.0063269502205442795, "minimum": 5.817323447466947, "maximum": 5.836888160292811, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6558731090204926, "standard_deviation": 0.06479776617131117, "minimum": 0.5842741816503302, "maximum": 0.773078545706367, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08267968156493123, "standard_deviation": 0.04126287597335567, "minimum": -0.139562972888422, "maximum": -0.019334663027592017, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9902014908009196, "standard_deviation": 0.0009417012362512482, "minimum": 0.9888228678723306, "maximum": 0.9912675390573463, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.048218864219026, "standard_deviation": 0.041712395977101906, "minimum": 1.0028209095622758, "maximum": 1.1028793134939323, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2615950572627699, "standard_deviation": 0.1139842286181064, "minimum": 0.059885205882913795, "maximum": 0.40205458946298184, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08841421641454723, "standard_deviation": 0.00434476754046342, "minimum": 0.08355284646801511, "maximum": 0.09470856918782952, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969261679619604, "standard_deviation": 0.0006914804899263556, "minimum": 0.9958145838263079, "maximum": 0.9980059273287105, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6057716808079996, "standard_deviation": 0.03618672773132992, "minimum": 0.5764338684876006, "maximum": 0.6728536238632572, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2978253420715298, "standard_deviation": 0.14550032190348255, "minimum": 0.0771187741157045, "maximum": 0.4905659111553043, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04785688844029502, "standard_deviation": 0.005102491969486799, "minimum": 0.03953416304255952, "maximum": 0.05500606425658546, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0022", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.230555534362793, "predicted_mos": 2.230555534362793, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5a1900615c458aecf424e43a52109db2e315adb03e267b1f631aa345580ca85b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.500287699978799, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0064588632990623, "全帯域_10ms_rms_envelope_nrmse": 0.08663815292604876, "全帯域_10ms_envelope_correlation": 0.990568957782517, "全帯域_10ms_modulation_mae_db": 0.18019113038759701, "全帯域_40ms_log_rms_mae_db": 0.5834637435278145, "全帯域_40ms_rms_envelope_nrmse": 0.045061423469099304, "全帯域_40ms_envelope_correlation": 0.9972164415456454, "全帯域_40ms_modulation_mae_db": 0.20701055743117905, "300–1000 Hz_10ms_log_rms_mae_db": 1.8016714225683899, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1415316851268612, "300–1000 Hz_10ms_envelope_correlation": 0.9830620464214638, "300–1000 Hz_10ms_modulation_mae_db": 0.45566154656535396, "300–1000 Hz_40ms_log_rms_mae_db": 1.316798522438083, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08102378428288877, "300–1000 Hz_40ms_envelope_correlation": 0.9942950468100498, "300–1000 Hz_40ms_modulation_mae_db": 0.9776091199857708, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5281437719928093, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.27995245137920277, "1000–3000 Hz_10ms_envelope_correlation": 0.9385364674708779, "1000–3000 Hz_10ms_modulation_mae_db": 0.7278025740144551, "1000–3000 Hz_40ms_log_rms_mae_db": 1.8307058410385055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1735931371509715, "1000–3000 Hz_40ms_envelope_correlation": 0.9742576597093547, "1000–3000 Hz_40ms_modulation_mae_db": 0.9042356086492407, "3000–6000 Hz_10ms_log_rms_mae_db": 2.745487732916695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31477977036474125, "3000–6000 Hz_10ms_envelope_correlation": 0.9416328775915529, "3000–6000 Hz_10ms_modulation_mae_db": 0.8200963094318675, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1457801424406866, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23954969154349146, "3000–6000 Hz_40ms_envelope_correlation": 0.9651984382029759, "3000–6000 Hz_40ms_modulation_mae_db": 0.9840829867661753, "mr_stft_db": 6.305889057134574, "band_mae_db": 0.2590779716789833, "rms_gain_error_db": -0.04833733663217091, "output_peak": 0.7088605508541533}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.012108172967644, "全帯域_10ms_rms_envelope_nrmse": 0.08198926381614673, "全帯域_10ms_envelope_correlation": 0.9915670403900486, "全帯域_10ms_modulation_mae_db": 0.26381163571017685, "全帯域_40ms_log_rms_mae_db": 0.5655461912829501, "全帯域_40ms_rms_envelope_nrmse": 0.04332044677897451, "全帯域_40ms_envelope_correlation": 0.9975030682605971, "全帯域_40ms_modulation_mae_db": 0.26032448217850024, "300–1000 Hz_10ms_log_rms_mae_db": 1.7348023405815258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1487035614914893, "300–1000 Hz_10ms_envelope_correlation": 0.9820829174016621, "300–1000 Hz_10ms_modulation_mae_db": 0.550948816912005, "300–1000 Hz_40ms_log_rms_mae_db": 1.2432323476951666, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09259357487069511, "300–1000 Hz_40ms_envelope_correlation": 0.9936032995651886, "300–1000 Hz_40ms_modulation_mae_db": 1.0994394053756105, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4534655296056798, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2428391185241409, "1000–3000 Hz_10ms_envelope_correlation": 0.9533516930273851, "1000–3000 Hz_10ms_modulation_mae_db": 1.5438207791861858, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7157115278144568, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14114749888341702, "1000–3000 Hz_40ms_envelope_correlation": 0.9836647354674773, "1000–3000 Hz_40ms_modulation_mae_db": 1.6498254813064925, "3000–6000 Hz_10ms_log_rms_mae_db": 2.80927201683408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35859648322131804, "3000–6000 Hz_10ms_envelope_correlation": 0.9294034105985035, "3000–6000 Hz_10ms_modulation_mae_db": 0.4981965838629118, "3000–6000 Hz_40ms_log_rms_mae_db": 2.096644395086863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2817954512833317, "3000–6000 Hz_40ms_envelope_correlation": 0.9556950653313349, "3000–6000 Hz_40ms_modulation_mae_db": 0.45451750652825035, "mr_stft_db": 6.28227194048505, "band_mae_db": 0.37404989538253125, "rms_gain_error_db": -0.08560577785790674, "output_peak": 0.6255662811318446}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.059548944320331, "全帯域_10ms_rms_envelope_nrmse": 0.08808776065726576, "全帯域_10ms_envelope_correlation": 0.9902962242391494, "全帯域_10ms_modulation_mae_db": 0.08039478019739452, "全帯域_40ms_log_rms_mae_db": 0.5620867529255663, "全帯域_40ms_rms_envelope_nrmse": 0.047035135165340775, "全帯域_40ms_envelope_correlation": 0.9970827474847787, "全帯域_40ms_modulation_mae_db": 0.09493018485136907, "300–1000 Hz_10ms_log_rms_mae_db": 1.8168826495463961, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13896747628886963, "300–1000 Hz_10ms_envelope_correlation": 0.9838432289789917, "300–1000 Hz_10ms_modulation_mae_db": 0.5276885133473951, "300–1000 Hz_40ms_log_rms_mae_db": 1.246295561600607, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08362520279037952, "300–1000 Hz_40ms_envelope_correlation": 0.9941654046438015, "300–1000 Hz_40ms_modulation_mae_db": 0.8657334876414623, "1000–3000 Hz_10ms_log_rms_mae_db": 2.475081026871279, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2572971542325421, "1000–3000 Hz_10ms_envelope_correlation": 0.949538052106533, "1000–3000 Hz_10ms_modulation_mae_db": 0.9497695310250674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7948910494363184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16562523057324977, "1000–3000 Hz_40ms_envelope_correlation": 0.9772976819082069, "1000–3000 Hz_40ms_modulation_mae_db": 0.7408116702280356, "3000–6000 Hz_10ms_log_rms_mae_db": 2.740836459703933, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.33512619673853394, "3000–6000 Hz_10ms_envelope_correlation": 0.9372730269619864, "3000–6000 Hz_10ms_modulation_mae_db": 0.6688596321643505, "3000–6000 Hz_40ms_log_rms_mae_db": 2.130272223528085, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25462205062590965, "3000–6000 Hz_40ms_envelope_correlation": 0.9631049874454513, "3000–6000 Hz_40ms_modulation_mae_db": 0.854688710909382, "mr_stft_db": 6.298675986969344, "band_mae_db": 0.3953228799650965, "rms_gain_error_db": -0.10221495734813424, "output_peak": 0.6144735893361035}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9990237931864521, "全帯域_10ms_rms_envelope_nrmse": 0.08358894048151798, "全帯域_10ms_envelope_correlation": 0.9912088175669913, "全帯域_10ms_modulation_mae_db": 0.40506990395053344, "全帯域_40ms_log_rms_mae_db": 0.5989015120158873, "全帯域_40ms_rms_envelope_nrmse": 0.047713759694624755, "全帯域_40ms_envelope_correlation": 0.99684626098598, "全帯域_40ms_modulation_mae_db": 0.43092359139738584, "300–1000 Hz_10ms_log_rms_mae_db": 1.8149812042228046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1611765616032448, "300–1000 Hz_10ms_envelope_correlation": 0.9783536337985236, "300–1000 Hz_10ms_modulation_mae_db": 0.5413801614223697, "300–1000 Hz_40ms_log_rms_mae_db": 1.29936652090187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09973432218542685, "300–1000 Hz_40ms_envelope_correlation": 0.9918846076188118, "300–1000 Hz_40ms_modulation_mae_db": 0.8666477585257623, "1000–3000 Hz_10ms_log_rms_mae_db": 2.49236595832403, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.32264062190501125, "1000–3000 Hz_10ms_envelope_correlation": 0.9220789091529189, "1000–3000 Hz_10ms_modulation_mae_db": 1.2175163543372598, "1000–3000 Hz_40ms_log_rms_mae_db": 1.758645515464873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.21528858072864376, "1000–3000 Hz_40ms_envelope_correlation": 0.9624135223486028, "1000–3000 Hz_40ms_modulation_mae_db": 1.1958003100025423, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7753953420297126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31880610948558913, "3000–6000 Hz_10ms_envelope_correlation": 0.943458294316399, "3000–6000 Hz_10ms_modulation_mae_db": 0.19081403246017634, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1012132766618015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24147353924081244, "3000–6000 Hz_40ms_envelope_correlation": 0.9670882600409313, "3000–6000 Hz_40ms_modulation_mae_db": 0.48384038043539057, "mr_stft_db": 6.293081420087602, "band_mae_db": 0.4278198023935227, "rms_gain_error_db": -0.027765826761741073, "output_peak": 0.6005062561189572}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9576530151146273, "全帯域_10ms_rms_envelope_nrmse": 0.07701365951372943, "全帯域_10ms_envelope_correlation": 0.9925305700204219, "全帯域_10ms_modulation_mae_db": 0.2022618368986203, "全帯域_40ms_log_rms_mae_db": 0.4960454832465878, "全帯域_40ms_rms_envelope_nrmse": 0.04008035647965451, "全帯域_40ms_envelope_correlation": 0.997806122931256, "全帯域_40ms_modulation_mae_db": 0.15702591965209814, "300–1000 Hz_10ms_log_rms_mae_db": 1.670603108903109, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1169513120651795, "300–1000 Hz_10ms_envelope_correlation": 0.9884388487751302, "300–1000 Hz_10ms_modulation_mae_db": 0.41009192550996376, "300–1000 Hz_40ms_log_rms_mae_db": 1.133163339490189, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06578679951390821, "300–1000 Hz_40ms_envelope_correlation": 0.9961840040653699, "300–1000 Hz_40ms_modulation_mae_db": 0.6654663450169305, "1000–3000 Hz_10ms_log_rms_mae_db": 2.515710225449831, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24985882401691253, "1000–3000 Hz_10ms_envelope_correlation": 0.9509644496840898, "1000–3000 Hz_10ms_modulation_mae_db": 0.9198871705959175, "1000–3000 Hz_40ms_log_rms_mae_db": 1.8386519434861743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1528670134085282, "1000–3000 Hz_40ms_envelope_correlation": 0.98003407261649, "1000–3000 Hz_40ms_modulation_mae_db": 0.40690686042869406, "3000–6000 Hz_10ms_log_rms_mae_db": 2.817802549292179, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3428511806228803, "3000–6000 Hz_10ms_envelope_correlation": 0.9387237993089022, "3000–6000 Hz_10ms_modulation_mae_db": 1.1270291590165211, "3000–6000 Hz_40ms_log_rms_mae_db": 2.15154930380749, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2691199092072592, "3000–6000 Hz_40ms_envelope_correlation": 0.9623904173279328, "3000–6000 Hz_40ms_modulation_mae_db": 1.250598163699901, "mr_stft_db": 6.305057131844208, "band_mae_db": 0.3633122396286284, "rms_gain_error_db": -0.05295601969625787, "output_peak": 0.6371501177925658}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0296742005085515, "全帯域_10ms_rms_envelope_nrmse": 0.08993469837369054, "全帯域_10ms_envelope_correlation": 0.9902758862287339, "全帯域_10ms_modulation_mae_db": 0.34910448421409673, "全帯域_40ms_log_rms_mae_db": 0.5878741444657742, "全帯域_40ms_rms_envelope_nrmse": 0.05119869819214084, "全帯域_40ms_envelope_correlation": 0.9970745361005454, "全帯域_40ms_modulation_mae_db": 0.7074965920937982, "300–1000 Hz_10ms_log_rms_mae_db": 1.7474707515996604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14667533414256764, "300–1000 Hz_10ms_envelope_correlation": 0.9830248068648891, "300–1000 Hz_10ms_modulation_mae_db": 0.6733640039368715, "300–1000 Hz_40ms_log_rms_mae_db": 1.2456852842597577, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09278621900981437, "300–1000 Hz_40ms_envelope_correlation": 0.9940782645888775, "300–1000 Hz_40ms_modulation_mae_db": 1.1508500565805566, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5804312626545562, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.28125765513659085, "1000–3000 Hz_10ms_envelope_correlation": 0.9368351650253506, "1000–3000 Hz_10ms_modulation_mae_db": 0.9238101066494853, "1000–3000 Hz_40ms_log_rms_mae_db": 1.8756023577026992, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1775723440238648, "1000–3000 Hz_40ms_envelope_correlation": 0.9735884975610685, "1000–3000 Hz_40ms_modulation_mae_db": 1.2075149079703131, "3000–6000 Hz_10ms_log_rms_mae_db": 2.812919811809371, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3224776603044207, "3000–6000 Hz_10ms_envelope_correlation": 0.9412675876255905, "3000–6000 Hz_10ms_modulation_mae_db": 1.144032293848613, "3000–6000 Hz_40ms_log_rms_mae_db": 2.144394167278699, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24935883296717304, "3000–6000 Hz_40ms_envelope_correlation": 0.9641705680615548, "3000–6000 Hz_40ms_modulation_mae_db": 1.6601314106794574, "mr_stft_db": 6.304202154131095, "band_mae_db": 0.38463644871525515, "rms_gain_error_db": -0.20860448938464285, "output_peak": 0.6653381799770017}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0554928504979575, "全帯域_10ms_rms_envelope_nrmse": 0.09264268597507741, "全帯域_10ms_envelope_correlation": 0.9892562862958612, "全帯域_10ms_modulation_mae_db": 0.3029477627389945, "全帯域_40ms_log_rms_mae_db": 0.5728414617199242, "全帯域_40ms_rms_envelope_nrmse": 0.04717629847316531, "全帯域_40ms_envelope_correlation": 0.9969424387651459, "全帯域_40ms_modulation_mae_db": 0.438301654282536, "300–1000 Hz_10ms_log_rms_mae_db": 1.871017892527061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1478096695989812, "300–1000 Hz_10ms_envelope_correlation": 0.9813128131710315, "300–1000 Hz_10ms_modulation_mae_db": 0.7202048803320972, "300–1000 Hz_40ms_log_rms_mae_db": 1.332871762057588, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09345734523677185, "300–1000 Hz_40ms_envelope_correlation": 0.9917345677764797, "300–1000 Hz_40ms_modulation_mae_db": 0.886625524963664, "1000–3000 Hz_10ms_log_rms_mae_db": 2.484811717947384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2547553439273231, "1000–3000 Hz_10ms_envelope_correlation": 0.9487668814892669, "1000–3000 Hz_10ms_modulation_mae_db": 1.2545449098869244, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7763279914412324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1590341642711252, "1000–3000 Hz_40ms_envelope_correlation": 0.9784365789206682, "1000–3000 Hz_40ms_modulation_mae_db": 1.5324355990760476, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7193795078203298, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31821281646775124, "3000–6000 Hz_10ms_envelope_correlation": 0.9409199202727456, "3000–6000 Hz_10ms_modulation_mae_db": 0.8476055598017037, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0668090984014853, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24343892511575102, "3000–6000 Hz_40ms_envelope_correlation": 0.9644127227684871, "3000–6000 Hz_40ms_modulation_mae_db": 1.0442913330200039, "mr_stft_db": 6.2963942545422285, "band_mae_db": 0.3057842283932093, "rms_gain_error_db": -0.020178585843532916, "output_peak": 0.6252356140786}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0436326413135202, "全帯域_10ms_rms_envelope_nrmse": 0.09455831346746255, "全帯域_10ms_envelope_correlation": 0.9888327529774701, "全帯域_10ms_modulation_mae_db": 0.3943973103075356, "全帯域_40ms_log_rms_mae_db": 0.5565564029036663, "全帯域_40ms_rms_envelope_nrmse": 0.04770082234572796, "全帯域_40ms_envelope_correlation": 0.9969122509846216, "全帯域_40ms_modulation_mae_db": 0.28667012173368844, "300–1000 Hz_10ms_log_rms_mae_db": 1.7994825446971905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14351792896755353, "300–1000 Hz_10ms_envelope_correlation": 0.9824150203982135, "300–1000 Hz_10ms_modulation_mae_db": 0.39827820835268063, "300–1000 Hz_40ms_log_rms_mae_db": 1.3191519565637948, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08585382954178898, "300–1000 Hz_40ms_envelope_correlation": 0.993236185551003, "300–1000 Hz_40ms_modulation_mae_db": 0.8625382670781091, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4772647209431384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2258745726907767, "1000–3000 Hz_10ms_envelope_correlation": 0.9603301787390626, "1000–3000 Hz_10ms_modulation_mae_db": 0.9666836743254167, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7935524274631227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1502373273617456, "1000–3000 Hz_40ms_envelope_correlation": 0.9809009568266719, "1000–3000 Hz_40ms_modulation_mae_db": 0.9159513602336453, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8198944176589, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3403734770577068, "3000–6000 Hz_10ms_envelope_correlation": 0.9296345559056387, "3000–6000 Hz_10ms_modulation_mae_db": 1.3681838206859378, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1450035931840086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.265109295655483, "3000–6000 Hz_40ms_envelope_correlation": 0.9560714915314528, "3000–6000 Hz_40ms_modulation_mae_db": 1.925797714994501, "mr_stft_db": 6.304669163082088, "band_mae_db": 0.28574434862393056, "rms_gain_error_db": -0.03866787385819335, "output_peak": 0.6849016606120762}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.369579538530757, "spectral_convergence": 0.27983259381886644, "spectral_convergence_status": "ok", "amplitude_floor": 2.1903561449018732e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.120781392979894, "spectral_convergence": 0.28877932714843313, "spectral_convergence_status": "ok", "amplitude_floor": 2.0456135368211886e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.42730623989307, "spectral_convergence": 0.3012555237349037, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394418151882858e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0025307845601282375, "output_power": 0.0024851151919307642, "delta_db": -0.07908651960106967, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026710985360254404, "output_power": 0.0026631004357812755, "delta_db": -0.01302363643792924, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004653266401104453, "output_power": 0.0004485693916575655, "delta_db": -0.1592828078304155, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012444324831517497, "output_power": 0.00011727973165319373, "delta_db": -0.25748375286985636, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00016257063671523992, "output_power": 0.00016466838245181087, "delta_db": 0.0556811275006401, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.589634923518972e-05, "output_power": 4.508595574165988e-05, "delta_db": 0.9899099858339887, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2590779716789833, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04833733663217091, "pitch": null, "output_rms_dbfs": -22.27398213107413, "output_peak": 0.7088605508541533, "reference_rms_dbfs": -22.225644794441955, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.305889057134574}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031508938028130336, "log_rms_mae_db": 1.0064588632990623, "rms_envelope_nrmse": 0.08663815292604876, "envelope_correlation": 0.990568957782517, "rms_gain_error_db": -0.04832558559632418, "reference_mean_power": 0.0059947180836205155, "output_mean_power": 0.005928382349529874}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.905621652404067e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011843610245185533, "output_power": 0.0011638235666559002, "reference_active": true, "delta_db": -0.07596959873822826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002763334576078544, "output_power": 0.0002651624465392381, "reference_active": true, "delta_db": -0.17921453733084736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0009701982183084e-05, "output_power": 5.340638620643938e-05, "reference_active": true, "delta_db": 0.2853892550937154}], "active_band_count": 3, "active_band_mae_db": 0.18019113038759701}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002456604014165908, "log_rms_mae_db": 0.5834637435278145, "rms_envelope_nrmse": 0.045061423469099304, "envelope_correlation": 0.9972164415456454, "rms_gain_error_db": -0.04832866430113153, "reference_mean_power": 0.006022035579915752, "output_mean_power": 0.005955393336919703}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5974133279358877e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001059507007170014, "output_power": 0.0010474744570687863, "reference_active": true, "delta_db": -0.04960391926240845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014348331799351297, "output_power": 0.00013756987654664916, "reference_active": true, "delta_db": -0.1827806345441163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4072600706208148e-06, "output_power": 3.1155919423084666e-06, "reference_active": true, "delta_db": -0.38864711848701244}], "active_band_count": 3, "active_band_mae_db": 0.20701055743117905}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026889287764829997, "log_rms_mae_db": 1.8016714225683899, "rms_envelope_nrmse": 0.1415316851268612, "envelope_correlation": 0.9830620464214638, "rms_gain_error_db": -0.22134963588591938, "reference_mean_power": 0.0016839373335741825, "output_mean_power": 0.0016002615187253727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1138357532326787e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006324020549599114, "output_power": 0.0005865240826130973, "reference_active": true, "delta_db": -0.3270742358098922}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002190350723697022, "output_power": 0.00018853348418090818, "reference_active": true, "delta_db": -0.651251670274861}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1194141233781235e-05, "output_power": 2.852378542674204e-05, "reference_active": true, "delta_db": -0.3886587336113087}], "active_band_count": 3, "active_band_mae_db": 0.45566154656535396}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001970017370857221, "log_rms_mae_db": 1.316798522438083, "rms_envelope_nrmse": 0.08102378428288877, "envelope_correlation": 0.9942950468100498, "rms_gain_error_db": -0.2213685318416385, "reference_mean_power": 0.001692399182790771, "output_mean_power": 0.0016082958961487722}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.64481232375879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970882254118313, "output_power": 0.0005566402717356128, "reference_active": true, "delta_db": -0.3046388355943907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011301601543605377, "output_power": 9.80213067770291e-05, "reference_active": true, "delta_db": -0.6181950343295142}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.966963236644965e-06, "output_power": 1.8677245014486236e-06, "reference_active": true, "delta_db": -2.0099934900334078}], "active_band_count": 3, "active_band_mae_db": 0.9776091199857708}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.658803471872848e-05, "log_rms_mae_db": 2.5281437719928093, "rms_envelope_nrmse": 0.27995245137920277, "envelope_correlation": 0.9385364674708779, "rms_gain_error_db": -0.04393997176829683, "reference_mean_power": 0.00020257298157917204, "output_mean_power": 0.0002005337721231354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0818748871802829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.592086384233793e-05, "output_power": 5.7848853927578956e-05, "reference_active": true, "delta_db": -0.5672813047330534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.925318155530915e-05, "output_power": 1.3803623996178126e-05, "reference_active": true, "delta_db": -1.4450938537901545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.819375654274508e-06, "output_power": 3.6718854063611606e-06, "reference_active": true, "delta_db": -0.1710325635201574}], "active_band_count": 3, "active_band_mae_db": 0.7278025740144551}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.479997453028037e-05, "log_rms_mae_db": 1.8307058410385055, "rms_envelope_nrmse": 0.1735931371509715, "envelope_correlation": 0.9742576597093547, "rms_gain_error_db": -0.043998473503713616, "reference_mean_power": 0.00020359039158760131, "output_mean_power": 0.00020153822548737472}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.438814668171717e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.33475246376368e-05, "output_power": 5.434065979714742e-05, "reference_active": true, "delta_db": -0.666047418372263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.661338887813947e-06, "output_power": 6.915637989379863e-06, "reference_active": true, "delta_db": -1.452050646638098}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1289641878321164e-07, "output_power": 2.7285990049370557e-07, "reference_active": true, "delta_db": -0.594608760937361}], "active_band_count": 3, "active_band_mae_db": 0.9042356086492407}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.558280881155462e-05, "log_rms_mae_db": 2.745487732916695, "rms_envelope_nrmse": 0.31477977036474125, "envelope_correlation": 0.9416328775915529, "rms_gain_error_db": 0.13356495734762455, "reference_mean_power": 8.701244577979696e-05, "output_mean_power": 8.973004217552658e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.56825886194788e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4501869786005614e-05, "output_power": 2.25320539433938e-05, "reference_active": true, "delta_db": -0.3639844520010226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.528335672589267e-06, "output_power": 3.30953799117328e-06, "reference_active": true, "delta_db": -1.3617124153427178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7809461115210634e-06, "output_power": 1.5038051664073155e-06, "reference_active": true, "delta_db": -0.7345920609518618}], "active_band_count": 3, "active_band_mae_db": 0.8200963094318675}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.713713309096251e-05, "log_rms_mae_db": 2.1457801424406866, "rms_envelope_nrmse": 0.23954969154349146, "envelope_correlation": 0.9651984382029759, "rms_gain_error_db": 0.1335258206580145, "reference_mean_power": 8.74494743564098e-05, "output_mean_power": 9.017990748269444e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.348274544921528e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.135032255455729e-05, "output_power": 2.0069737722546954e-05, "reference_active": true, "delta_db": -0.26862743575395764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4135743662495686e-06, "output_power": 1.6454353492786003e-06, "reference_active": true, "delta_db": -1.6637986157585734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6206121669861435e-07, "output_power": 1.2814356564399608e-07, "reference_active": true, "delta_db": -1.0198229087859947}], "active_band_count": 3, "active_band_mae_db": 0.9840829867661753}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.879452901978408, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "eb8dece3249ca7d733a164b291486821b55f1ee0eb3f479a9ba22aa381d807b5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9450502245869357, "standard_deviation": 0.012005481713251226, "minimum": 0.9220789091529189, "maximum": 0.9603301787390626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.5009092767235885, "standard_deviation": 0.03980976979708253, "minimum": 2.4534655296056798, "maximum": 2.5804312626545562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.062979387502589, "standard_deviation": 0.25803204977940186, "minimum": 0.7278025740144551, "maximum": 1.5438207791861858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2643094677265625, "standard_deviation": 0.029813610312000673, "minimum": 0.2258745726907767, "maximum": 0.32264062190501125, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9763242131698175, "standard_deviation": 0.006538509440093939, "minimum": 0.9624135223486028, "maximum": 0.9836647354674773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.7980110817309227, "standard_deviation": 0.05009087174258495, "minimum": 1.7157115278144568, "maximum": 1.8756023577026992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0691852247368765, "standard_deviation": 0.41127698594298984, "minimum": 0.40690686042869406, "maximum": 1.6498254813064925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16692066205019324, "standard_deviation": 0.02299504740512013, "minimum": 0.14114749888341702, "maximum": 0.21528858072864376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9377891840726649, "standard_deviation": 0.0054349749948087, "minimum": 0.9294034105985035, "maximum": 0.943458294316399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.78012347975815, "standard_deviation": 0.04032363133680136, "minimum": 2.7193795078203298, "maximum": 2.8198944176589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8331021739090102, "standard_deviation": 0.3819936246475053, "minimum": 0.19081403246017634, "maximum": 1.3681838206859378, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.33140296178286766, "standard_deviation": 0.015369422641362, "minimum": 0.31477977036474125, "maximum": 0.35859648322131804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9622664938387651, "standard_deviation": 0.004180869340619766, "minimum": 0.9556950653313349, "maximum": 0.9670882600409313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.12270827504864, "standard_deviation": 0.030833115773553865, "minimum": 2.0668090984014853, "maximum": 2.15154930380749, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0822435258791328, "standard_deviation": 0.5194591549313011, "minimum": 0.45451750652825035, "maximum": 1.925797714994501, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.25555846195490145, "standard_deviation": 0.015139221068280976, "minimum": 0.23954969154349146, "maximum": 0.2817954512833317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9828166644762382, "standard_deviation": 0.0028194792030563917, "minimum": 0.9783536337985236, "maximum": 0.9884388487751302, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7821139893307671, "standard_deviation": 0.06181816647401457, "minimum": 1.670603108903109, "maximum": 1.871017892527061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5347022570473421, "standard_deviation": 0.11611738562613788, "minimum": 0.39827820835268063, "maximum": 0.7202048803320972, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14316669116059336, "standard_deviation": 0.012516025785020491, "minimum": 0.1169513120651795, "maximum": 0.1611765616032448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936476725774477, "standard_deviation": 0.0014265435425484412, "minimum": 0.9917345677764797, "maximum": 0.9961840040653699, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.267070661875882, "standard_deviation": 0.0654069721562385, "minimum": 1.133163339490189, "maximum": 1.332871762057588, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9218637456459833, "standard_deviation": 0.152959226609546, "minimum": 0.6654663450169305, "maximum": 1.1508500565805566, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08685763467895921, "standard_deviation": 0.010466849220969394, "minimum": 0.06578679951390821, "maximum": 0.09973432218542685, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.34946847684764465, "standard_deviation": 0.05905658322563611, "minimum": 0.2590779716789833, "maximum": 0.4278198023935227, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.298780138534523, "standard_deviation": 0.008147878923590009, "minimum": 6.28227194048505, "maximum": 6.305889057134574, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6452540312376628, "standard_deviation": 0.03746410435900377, "minimum": 0.6005062561189572, "maximum": 0.7088605508541533, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0730413584228225, "standard_deviation": 0.06144209636913584, "minimum": -0.20860448938464285, "maximum": -0.020178585843532916, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9905670669376492, "standard_deviation": 0.0012042348776031004, "minimum": 0.9888327529774701, "maximum": 0.9925305700204219, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.020449060151018, "standard_deviation": 0.03390560673766852, "minimum": 0.9576530151146273, "maximum": 1.059548944320331, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2722723555506186, "standard_deviation": 0.1131767659710345, "minimum": 0.08039478019739452, "maximum": 0.40506990395053344, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08680668440136739, "standard_deviation": 0.005794892476111525, "minimum": 0.07701365951372943, "maximum": 0.09455831346746255, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971729833823213, "standard_deviation": 0.00032894299109536765, "minimum": 0.99684626098598, "maximum": 0.997806122931256, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5654144615110214, "standard_deviation": 0.03141511958157716, "minimum": 0.4960454832465878, "maximum": 0.5989015120158873, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3228353879525694, "standard_deviation": 0.1967755258158977, "minimum": 0.09493018485136907, "maximum": 0.7074965920937982, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04616086757484099, "standard_deviation": 0.0033425289507810155, "minimum": 0.04008035647965451, "maximum": 0.05119869819214084, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0022", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.388336181640625, "predicted_mos": 2.388336181640625, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c441a21d4d1cb471127b31b80d607b7d6c48f5fcd4407499a5311a8ece5d885f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.967980400193483, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0918972033684138, "全帯域_10ms_rms_envelope_nrmse": 0.08763675438936784, "全帯域_10ms_envelope_correlation": 0.9903772459569928, "全帯域_10ms_modulation_mae_db": 0.23636209482161374, "全帯域_40ms_log_rms_mae_db": 0.6503969823123052, "全帯域_40ms_rms_envelope_nrmse": 0.04654868884455955, "全帯域_40ms_envelope_correlation": 0.9971217334209351, "全帯域_40ms_modulation_mae_db": 0.39337764400853487, "300–1000 Hz_10ms_log_rms_mae_db": 1.4989152976474125, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14348555878235225, "300–1000 Hz_10ms_envelope_correlation": 0.9824338001394176, "300–1000 Hz_10ms_modulation_mae_db": 0.5289911884964332, "300–1000 Hz_40ms_log_rms_mae_db": 1.0291163668621113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09971121311406919, "300–1000 Hz_40ms_envelope_correlation": 0.9907537997690345, "300–1000 Hz_40ms_modulation_mae_db": 0.6662700271168626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6875100538502181, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23497270887569155, "1000–3000 Hz_10ms_envelope_correlation": 0.9566349669460286, "1000–3000 Hz_10ms_modulation_mae_db": 1.404879169873522, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1673781932145944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15425045360565845, "1000–3000 Hz_40ms_envelope_correlation": 0.9797211444297114, "1000–3000 Hz_40ms_modulation_mae_db": 1.4138697756846552, "3000–6000 Hz_10ms_log_rms_mae_db": 1.88377941157485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3165371206771419, "3000–6000 Hz_10ms_envelope_correlation": 0.9425040690010085, "3000–6000 Hz_10ms_modulation_mae_db": 0.5687240524868015, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4400646101556716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2474970515647639, "3000–6000 Hz_40ms_envelope_correlation": 0.9637850629919033, "3000–6000 Hz_40ms_modulation_mae_db": 0.6620216173340219, "mr_stft_db": 5.790875585772386, "band_mae_db": 0.29094056273008423, "rms_gain_error_db": -0.09303415233724491, "output_peak": 0.6243842893009348}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0966300929174202, "全帯域_10ms_rms_envelope_nrmse": 0.08801061634481797, "全帯域_10ms_envelope_correlation": 0.990295625083838, "全帯域_10ms_modulation_mae_db": 0.1815330237587888, "全帯域_40ms_log_rms_mae_db": 0.6324938782108736, "全帯域_40ms_rms_envelope_nrmse": 0.051914312287372644, "全帯域_40ms_envelope_correlation": 0.9964031234707129, "全帯域_40ms_modulation_mae_db": 0.21225645447718575, "300–1000 Hz_10ms_log_rms_mae_db": 1.4035734227866898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1220830794777576, "300–1000 Hz_10ms_envelope_correlation": 0.9876276286690957, "300–1000 Hz_10ms_modulation_mae_db": 0.44009880044958205, "300–1000 Hz_40ms_log_rms_mae_db": 0.9174975897657786, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07700408696001583, "300–1000 Hz_40ms_envelope_correlation": 0.9950262084836349, "300–1000 Hz_40ms_modulation_mae_db": 0.6540472756064769, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7571151984619704, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21397392232189757, "1000–3000 Hz_10ms_envelope_correlation": 0.9646725356134622, "1000–3000 Hz_10ms_modulation_mae_db": 0.7792933341446497, "1000–3000 Hz_40ms_log_rms_mae_db": 1.219941568721537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14864922929564486, "1000–3000 Hz_40ms_envelope_correlation": 0.9828282575379299, "1000–3000 Hz_40ms_modulation_mae_db": 0.9781954209683489, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9435293691827917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3152927008567636, "3000–6000 Hz_10ms_envelope_correlation": 0.9460701620931794, "3000–6000 Hz_10ms_modulation_mae_db": 1.3058070490377451, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4884243302918752, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24382691334129106, "3000–6000 Hz_40ms_envelope_correlation": 0.9672902259425522, "3000–6000 Hz_40ms_modulation_mae_db": 1.3240337304047676, "mr_stft_db": 5.798192454722147, "band_mae_db": 0.3598360699298248, "rms_gain_error_db": -0.10182568819986935, "output_peak": 0.5813230263925286}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0166117501355172, "全帯域_10ms_rms_envelope_nrmse": 0.0954138896121347, "全帯域_10ms_envelope_correlation": 0.9885166228494652, "全帯域_10ms_modulation_mae_db": 0.2532051113470341, "全帯域_40ms_log_rms_mae_db": 0.5995213636248486, "全帯域_40ms_rms_envelope_nrmse": 0.04857359315414818, "全帯域_40ms_envelope_correlation": 0.9968033318359283, "全帯域_40ms_modulation_mae_db": 0.44237094476384486, "300–1000 Hz_10ms_log_rms_mae_db": 1.5102382552149978, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1399545395408605, "300–1000 Hz_10ms_envelope_correlation": 0.9836430235690468, "300–1000 Hz_10ms_modulation_mae_db": 0.14853294749683985, "300–1000 Hz_40ms_log_rms_mae_db": 1.0258080188204777, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08998901032846712, "300–1000 Hz_40ms_envelope_correlation": 0.9931089413550338, "300–1000 Hz_40ms_modulation_mae_db": 0.32286392444054596, "1000–3000 Hz_10ms_log_rms_mae_db": 1.725423725590253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.205365801549806, "1000–3000 Hz_10ms_envelope_correlation": 0.9670779764331109, "1000–3000 Hz_10ms_modulation_mae_db": 0.8488839343697979, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2161934475137393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1276436055544407, "1000–3000 Hz_40ms_envelope_correlation": 0.9869957667731294, "1000–3000 Hz_40ms_modulation_mae_db": 0.9812807439138296, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8146783301489273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.300835392431884, "3000–6000 Hz_10ms_envelope_correlation": 0.9461307257186925, "3000–6000 Hz_10ms_modulation_mae_db": 1.2884660503952874, "3000–6000 Hz_40ms_log_rms_mae_db": 1.358028145683228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23387305169624054, "3000–6000 Hz_40ms_envelope_correlation": 0.9662929155207131, "3000–6000 Hz_40ms_modulation_mae_db": 1.2201441642923647, "mr_stft_db": 5.765225735496423, "band_mae_db": 0.3760579361120289, "rms_gain_error_db": -0.07671472687680296, "output_peak": 0.5602454238967071}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0801082395561465, "全帯域_10ms_rms_envelope_nrmse": 0.08436722738926947, "全帯域_10ms_envelope_correlation": 0.9910496264530194, "全帯域_10ms_modulation_mae_db": 0.2575959872982333, "全帯域_40ms_log_rms_mae_db": 0.6228896492127036, "全帯域_40ms_rms_envelope_nrmse": 0.04153030137010387, "全帯域_40ms_envelope_correlation": 0.9976400095515078, "全帯域_40ms_modulation_mae_db": 0.27607885436150276, "300–1000 Hz_10ms_log_rms_mae_db": 1.4922254192435016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14794727639046085, "300–1000 Hz_10ms_envelope_correlation": 0.9813521006552532, "300–1000 Hz_10ms_modulation_mae_db": 0.5656038962008395, "300–1000 Hz_40ms_log_rms_mae_db": 1.092763133383735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08993219346159256, "300–1000 Hz_40ms_envelope_correlation": 0.9926674674262685, "300–1000 Hz_40ms_modulation_mae_db": 0.7877434969153455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7922833012878145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2032988274424379, "1000–3000 Hz_10ms_envelope_correlation": 0.9677243866158982, "1000–3000 Hz_10ms_modulation_mae_db": 0.8092879552875932, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2492072765947495, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1387542867404171, "1000–3000 Hz_40ms_envelope_correlation": 0.9842658918291277, "1000–3000 Hz_40ms_modulation_mae_db": 0.8998009560379653, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9227703441851831, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3033770496621065, "3000–6000 Hz_10ms_envelope_correlation": 0.9471881432535462, "3000–6000 Hz_10ms_modulation_mae_db": 1.2261299137260404, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4528610467539917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24045911966390174, "3000–6000 Hz_40ms_envelope_correlation": 0.9658268450551412, "3000–6000 Hz_40ms_modulation_mae_db": 1.5171426958243195, "mr_stft_db": 5.79598374287798, "band_mae_db": 0.25901238927485287, "rms_gain_error_db": -0.0423016079956757, "output_peak": 0.5779443632718201}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0647990760522934, "全帯域_10ms_rms_envelope_nrmse": 0.09218965650066333, "全帯域_10ms_envelope_correlation": 0.9894091082553794, "全帯域_10ms_modulation_mae_db": 0.322866307167381, "全帯域_40ms_log_rms_mae_db": 0.6524099436583051, "全帯域_40ms_rms_envelope_nrmse": 0.0492198838648081, "全帯域_40ms_envelope_correlation": 0.9968861145024509, "全帯域_40ms_modulation_mae_db": 0.39130495010039196, "300–1000 Hz_10ms_log_rms_mae_db": 1.4708370122181473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13771433841334407, "300–1000 Hz_10ms_envelope_correlation": 0.9838993336056024, "300–1000 Hz_10ms_modulation_mae_db": 0.6051038592145018, "300–1000 Hz_40ms_log_rms_mae_db": 1.0606686473020244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0828103328783966, "300–1000 Hz_40ms_envelope_correlation": 0.9938691596224207, "300–1000 Hz_40ms_modulation_mae_db": 0.8540170836199444, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7433424449656272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23057272438435927, "1000–3000 Hz_10ms_envelope_correlation": 0.9582015424710145, "1000–3000 Hz_10ms_modulation_mae_db": 0.35604009242111007, "1000–3000 Hz_40ms_log_rms_mae_db": 1.219921131325452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1587912447884696, "1000–3000 Hz_40ms_envelope_correlation": 0.9784614540528189, "1000–3000 Hz_40ms_modulation_mae_db": 0.5261760036645365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8328968085629216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30238346873982724, "3000–6000 Hz_10ms_envelope_correlation": 0.944206152506327, "3000–6000 Hz_10ms_modulation_mae_db": 1.0591606408746959, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4222604759123127, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23717656996400752, "3000–6000 Hz_40ms_envelope_correlation": 0.9646246798724356, "3000–6000 Hz_40ms_modulation_mae_db": 1.1768103213768697, "mr_stft_db": 5.782087912044201, "band_mae_db": 0.2749322674657663, "rms_gain_error_db": -0.12815206279666613, "output_peak": 0.6889104918780302}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0105043422353874, "全帯域_10ms_rms_envelope_nrmse": 0.08169610913766508, "全帯域_10ms_envelope_correlation": 0.9916044259623518, "全帯域_10ms_modulation_mae_db": 0.12168059573114255, "全帯域_40ms_log_rms_mae_db": 0.5764818929884343, "全帯域_40ms_rms_envelope_nrmse": 0.04396097600729901, "全帯域_40ms_envelope_correlation": 0.9973529551891085, "全帯域_40ms_modulation_mae_db": 0.31254416023659837, "300–1000 Hz_10ms_log_rms_mae_db": 1.4748268967289757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15151274947450785, "300–1000 Hz_10ms_envelope_correlation": 0.9805429763012563, "300–1000 Hz_10ms_modulation_mae_db": 0.4302613017489773, "300–1000 Hz_40ms_log_rms_mae_db": 1.005249453763281, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08533857933500856, "300–1000 Hz_40ms_envelope_correlation": 0.993641446289221, "300–1000 Hz_40ms_modulation_mae_db": 0.5911260209315817, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7030571282502718, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1805512178086063, "1000–3000 Hz_10ms_envelope_correlation": 0.9744710928673396, "1000–3000 Hz_10ms_modulation_mae_db": 0.6092297604916129, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1284720645018838, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11272672468764565, "1000–3000 Hz_40ms_envelope_correlation": 0.9893871367753879, "1000–3000 Hz_40ms_modulation_mae_db": 0.7328209513779175, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8596525997271698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3155495202904559, "3000–6000 Hz_10ms_envelope_correlation": 0.9420126976489382, "3000–6000 Hz_10ms_modulation_mae_db": 0.5844064841328954, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4006485644671638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2322362331270109, "3000–6000 Hz_40ms_envelope_correlation": 0.9675484911662763, "3000–6000 Hz_40ms_modulation_mae_db": 0.4443531551833715, "mr_stft_db": 5.767711305010773, "band_mae_db": 0.22765192086177835, "rms_gain_error_db": -0.05424568965089799, "output_peak": 0.6694799511389737}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0570337004348935, "全帯域_10ms_rms_envelope_nrmse": 0.09318097000022198, "全帯域_10ms_envelope_correlation": 0.9891092443233649, "全帯域_10ms_modulation_mae_db": 0.47915411732407404, "全帯域_40ms_log_rms_mae_db": 0.5795468860516316, "全帯域_40ms_rms_envelope_nrmse": 0.050096618704680476, "全帯域_40ms_envelope_correlation": 0.9966124715262956, "全帯域_40ms_modulation_mae_db": 0.3398126357377518, "300–1000 Hz_10ms_log_rms_mae_db": 1.54892579015338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1522952233188687, "300–1000 Hz_10ms_envelope_correlation": 0.9807528054151973, "300–1000 Hz_10ms_modulation_mae_db": 0.6033998978652056, "300–1000 Hz_40ms_log_rms_mae_db": 1.0882016744140102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08924554537487148, "300–1000 Hz_40ms_envelope_correlation": 0.9937207150583809, "300–1000 Hz_40ms_modulation_mae_db": 0.7710296148695116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7609170253971433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21053918783234715, "1000–3000 Hz_10ms_envelope_correlation": 0.9654405340890151, "1000–3000 Hz_10ms_modulation_mae_db": 0.42631774792578175, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2110576165927154, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12832252183146814, "1000–3000 Hz_40ms_envelope_correlation": 0.9869065834074223, "1000–3000 Hz_40ms_modulation_mae_db": 0.5718316954713256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8322108283329144, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29856802022567835, "3000–6000 Hz_10ms_envelope_correlation": 0.9535019708105692, "3000–6000 Hz_10ms_modulation_mae_db": 0.6237571217560594, "3000–6000 Hz_40ms_log_rms_mae_db": 1.393122033800517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23510565432689898, "3000–6000 Hz_40ms_envelope_correlation": 0.9712751871808858, "3000–6000 Hz_40ms_modulation_mae_db": 0.6979570344051281, "mr_stft_db": 5.77799535805723, "band_mae_db": 0.27734252282908783, "rms_gain_error_db": -0.08111127959002602, "output_peak": 0.6471362582013894}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.046479698731575, "全帯域_10ms_rms_envelope_nrmse": 0.09102166419303045, "全帯域_10ms_envelope_correlation": 0.9897447326526732, "全帯域_10ms_modulation_mae_db": 0.43374160913160703, "全帯域_40ms_log_rms_mae_db": 0.6058207143082069, "全帯域_40ms_rms_envelope_nrmse": 0.05606812834986307, "全帯域_40ms_envelope_correlation": 0.9960111951454543, "全帯域_40ms_modulation_mae_db": 0.3346418143028067, "300–1000 Hz_10ms_log_rms_mae_db": 1.578839206246664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14744275677610413, "300–1000 Hz_10ms_envelope_correlation": 0.9817584492830509, "300–1000 Hz_10ms_modulation_mae_db": 0.8993844575490991, "300–1000 Hz_40ms_log_rms_mae_db": 1.0772249680886232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09319261967376208, "300–1000 Hz_40ms_envelope_correlation": 0.992580601989102, "300–1000 Hz_40ms_modulation_mae_db": 0.5363233396971122, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7491298790124428, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22174006651146128, "1000–3000 Hz_10ms_envelope_correlation": 0.9613380470035409, "1000–3000 Hz_10ms_modulation_mae_db": 0.7284956212179866, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2475582200406727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15198344501424563, "1000–3000 Hz_40ms_envelope_correlation": 0.9806623903168334, "1000–3000 Hz_40ms_modulation_mae_db": 0.5655540880698844, "3000–6000 Hz_10ms_log_rms_mae_db": 1.853795921223763, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2933087276108999, "3000–6000 Hz_10ms_envelope_correlation": 0.9508127807347909, "3000–6000 Hz_10ms_modulation_mae_db": 0.609548691205941, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4067074599277296, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2281874702392798, "3000–6000 Hz_40ms_envelope_correlation": 0.9693826646265378, "3000–6000 Hz_40ms_modulation_mae_db": 1.1019775435847177, "mr_stft_db": 5.794645952765006, "band_mae_db": 0.28255864354548943, "rms_gain_error_db": -0.14984403354476242, "output_peak": 0.6048942746166469}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.759809688688768, "spectral_convergence": 0.28414664181180393, "spectral_convergence_status": "ok", "amplitude_floor": 2.1903561449018732e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.50118793609707, "spectral_convergence": 0.3043066155918604, "spectral_convergence_status": "ok", "amplitude_floor": 2.0456135368211886e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.111629132531319, "spectral_convergence": 0.32121000271891775, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394418151882858e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0025307845601282375, "output_power": 0.0024512802743417795, "delta_db": -0.13862205812526524, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026710985360254404, "output_power": 0.0026345694468408874, "delta_db": -0.059802583156291514, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004653266401104453, "output_power": 0.00045311777928183725, "delta_db": -0.11546813540526402, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012444324831517497, "output_power": 0.0001245219219925992, "delta_db": 0.0027447650666512, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00016257063671523992, "output_power": 0.0001527928269869447, "delta_db": -0.2693914034268979, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.589634923518972e-05, "output_power": 4.688260391556878e-05, "delta_db": 1.1596144312001353, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29094056273008423, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09303415233724491, "pitch": null, "output_rms_dbfs": -22.318678946779205, "output_peak": 0.6243842893009348, "reference_rms_dbfs": -22.225644794441955, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.790875585772386}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031508938028130336, "log_rms_mae_db": 1.0918972033684138, "rms_envelope_nrmse": 0.08763675438936784, "envelope_correlation": 0.9903772459569928, "rms_gain_error_db": -0.09308946796440933, "reference_mean_power": 0.0059947180836205155, "output_mean_power": 0.005867590778289754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.905621652404067e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011843610245185533, "output_power": 0.0011732134527120454, "reference_active": true, "delta_db": -0.041070727987346504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002763334576078544, "output_power": 0.00025660133316978426, "reference_active": true, "delta_db": -0.3217456272684043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0009701982183084e-05, "output_power": 4.617716577134224e-05, "reference_active": true, "delta_db": -0.34626992920909044}], "active_band_count": 3, "active_band_mae_db": 0.23636209482161374}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002456604014165908, "log_rms_mae_db": 0.6503969823123052, "rms_envelope_nrmse": 0.04654868884455955, "envelope_correlation": 0.9971217334209351, "rms_gain_error_db": -0.093205274967744, "reference_mean_power": 0.006022035579915752, "output_mean_power": 0.005894171791242394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5974133279358877e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001059507007170014, "output_power": 0.0010599096115653248, "reference_active": true, "delta_db": 0.001649971674572601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014348331799351297, "output_power": 0.00012882928501278595, "reference_active": true, "delta_db": -0.46786814487673445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4072600706208148e-06, "output_power": 2.8929691488501716e-06, "reference_active": true, "delta_db": -0.7106148154742975}], "active_band_count": 3, "active_band_mae_db": 0.39337764400853487}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026889287764829997, "log_rms_mae_db": 1.4989152976474125, "rms_envelope_nrmse": 0.14348555878235225, "envelope_correlation": 0.9824338001394176, "rms_gain_error_db": -0.15833085916074224, "reference_mean_power": 0.0016839373335741825, "output_mean_power": 0.0016236515804753345}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1138357532326787e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006324020549599114, "output_power": 0.0005860702515428365, "reference_active": true, "delta_db": -0.33043594980817165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002190350723697022, "output_power": 0.00018166348504620322, "reference_active": true, "delta_db": -0.8124601910940975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1194141233781235e-05, "output_power": 2.816211656050019e-05, "reference_active": true, "delta_db": -0.44407742458703076}], "active_band_count": 3, "active_band_mae_db": 0.5289911884964332}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001970017370857221, "log_rms_mae_db": 1.0291163668621113, "rms_envelope_nrmse": 0.09971121311406919, "envelope_correlation": 0.9907537997690345, "rms_gain_error_db": -0.1583511993609461, "reference_mean_power": 0.001692399182790771, "output_mean_power": 0.001631802848913883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.64481232375879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970882254118313, "output_power": 0.000556351507077837, "reference_active": true, "delta_db": -0.3068923814451661}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011301601543605377, "output_power": 8.95269707491168e-05, "reference_active": true, "delta_db": -1.0118610159717818}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.966963236644965e-06, "output_power": 2.536918390039192e-06, "reference_active": true, "delta_db": -0.6800566839336399}], "active_band_count": 3, "active_band_mae_db": 0.6662700271168626}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.658803471872848e-05, "log_rms_mae_db": 1.6875100538502181, "rms_envelope_nrmse": 0.23497270887569155, "envelope_correlation": 0.9566349669460286, "rms_gain_error_db": -0.1660125035869096, "reference_mean_power": 0.00020257298157917204, "output_mean_power": 0.00019497560201736896}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0818748871802829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.592086384233793e-05, "output_power": 6.883368252840024e-05, "reference_active": true, "delta_db": 0.18778114643291752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.925318155530915e-05, "output_power": 2.212491329218246e-05, "reference_active": true, "delta_db": 0.603790711727279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.819375654274508e-06, "output_power": 8.400380161195486e-06, "reference_active": true, "delta_db": 3.423065651460369}], "active_band_count": 3, "active_band_mae_db": 1.404879169873522}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.479997453028037e-05, "log_rms_mae_db": 1.1673781932145944, "rms_envelope_nrmse": 0.15425045360565845, "envelope_correlation": 0.9797211444297114, "rms_gain_error_db": -0.16607084008030937, "reference_mean_power": 0.00020359039158760131, "output_mean_power": 0.00019595222252587473}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.438814668171717e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.33475246376368e-05, "output_power": 7.078466546123765e-05, "reference_active": true, "delta_db": 0.48209534766129375}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.661338887813947e-06, "output_power": 1.224982622657992e-05, "reference_active": true, "delta_db": 1.0309261194469148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1289641878321164e-07, "output_power": 5.864883348630043e-07, "reference_active": true, "delta_db": 2.7285878599457574}], "active_band_count": 3, "active_band_mae_db": 1.4138697756846552}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.558280881155462e-05, "log_rms_mae_db": 1.88377941157485, "rms_envelope_nrmse": 0.3165371206771419, "envelope_correlation": 0.9425040690010085, "rms_gain_error_db": 0.2703814271798065, "reference_mean_power": 8.701244577979696e-05, "output_mean_power": 9.260181924450161e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.56825886194788e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4501869786005614e-05, "output_power": 2.217500060811754e-05, "reference_active": true, "delta_db": -0.4333558708391447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.528335672589267e-06, "output_power": 4.705822417729144e-06, "reference_active": true, "delta_db": 0.16696921919981997}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7809461115210634e-06, "output_power": 1.3805942869193844e-06, "reference_active": true, "delta_db": -1.1058470674214398}], "active_band_count": 3, "active_band_mae_db": 0.5687240524868015}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.713713309096251e-05, "log_rms_mae_db": 1.4400646101556716, "rms_envelope_nrmse": 0.2474970515647639, "envelope_correlation": 0.9637850629919033, "rms_gain_error_db": 0.27031967041171245, "reference_mean_power": 8.74494743564098e-05, "output_mean_power": 9.306559759550369e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.348274544921528e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.135032255455729e-05, "output_power": 1.9229076029349786e-05, "reference_active": true, "delta_db": -0.4544602402322315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4135743662495686e-06, "output_power": 2.5911844812818873e-06, "reference_active": true, "delta_db": 0.3083764930330072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6206121669861435e-07, "output_power": 1.222802402233249e-07, "reference_active": true, "delta_db": -1.2232281187368268}], "active_band_count": 3, "active_band_mae_db": 0.6620216173340219}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8945058807254953, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7a59c0c77abcf29eef3d42cb890456d3f3e9c2ad2a40c72ab735526e5a10236c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9644451352549261, "standard_deviation": 0.005718177472291811, "minimum": 0.9566349669460286, "maximum": 0.9744710928673396, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7398473446019675, "standard_deviation": 0.03357459447423981, "minimum": 1.6875100538502181, "maximum": 1.7922833012878145, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7453034519665067, "standard_deviation": 0.321402010816046, "minimum": 0.35604009242111007, "maximum": 1.404879169873522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21262680709082588, "standard_deviation": 0.017234502266244307, "minimum": 0.1805512178086063, "maximum": 0.23497270887569155, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9836535781402951, "standard_deviation": 0.0039106345287248075, "minimum": 0.9784614540528189, "maximum": 0.9893871367753879, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.207466189813168, "standard_deviation": 0.040702110402848414, "minimum": 1.1284720645018838, "maximum": 1.2492072765947495, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8336912043985578, "standard_deviation": 0.29972609801469013, "minimum": 0.5261760036645365, "maximum": 1.4138697756846552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14014018893974875, "standard_deviation": 0.016075129569946778, "minimum": 0.11272672468764565, "maximum": 0.1587912447884696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9465533377208815, "standard_deviation": 0.003959882071112869, "minimum": 0.9420126976489382, "maximum": 0.9535019708105692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8679142016173151, "standard_deviation": 0.04564011617089588, "minimum": 1.8146783301489273, "maximum": 1.9435293691827917, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9082500004519334, "standard_deviation": 0.34158217474991837, "minimum": 0.5687240524868015, "maximum": 1.3058070490377451, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3057315000618447, "standard_deviation": 0.008871598807472438, "minimum": 0.2933087276108999, "maximum": 0.3165371206771419, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9670032590445556, "standard_deviation": 0.0024551267497202813, "minimum": 0.9637850629919033, "maximum": 0.9712751871808858, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4202645833740613, "standard_deviation": 0.04014790715066414, "minimum": 1.358028145683228, "maximum": 1.4884243302918752, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.018055032800695, "standard_deviation": 0.37304807640982207, "minimum": 0.4443531551833715, "maximum": 1.5171426958243195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2372952579904243, "standard_deviation": 0.006345016173617399, "minimum": 0.2281874702392798, "maximum": 0.2474970515647639, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9827512647047401, "standard_deviation": 0.002323996712566505, "minimum": 0.9805429763012563, "maximum": 0.9876276286690957, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4972976625299712, "standard_deviation": 0.05277935837594397, "minimum": 1.4035734227866898, "maximum": 1.578839206246664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5276720436276848, "standard_deviation": 0.21120249524621643, "minimum": 0.14853294749683985, "maximum": 0.8993844575490991, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.142804440271782, "standard_deviation": 0.009839730992562433, "minimum": 0.1220830794777576, "maximum": 0.1522952233188687, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.993171042499137, "standard_deviation": 0.0012479442990207057, "minimum": 0.9907537997690345, "maximum": 0.9950262084836349, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0370662315500052, "standard_deviation": 0.057799767287911384, "minimum": 0.9174975897657786, "maximum": 1.092763133383735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6479275978996726, "standard_deviation": 0.16856439417564675, "minimum": 0.32286392444054596, "maximum": 0.8540170836199444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08840294764077293, "standard_deviation": 0.00683461465715383, "minimum": 0.07700408696001583, "maximum": 0.09971121311406919, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.29354153909361413, "standard_deviation": 0.04996245270880858, "minimum": 0.22765192086177835, "maximum": 0.3760579361120289, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.784089755843269, "standard_deviation": 0.012890128021178615, "minimum": 5.765225735496423, "maximum": 5.798192454722147, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6192897598371289, "standard_deviation": 0.046309304220040155, "minimum": 0.5602454238967071, "maximum": 0.6889104918780302, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09090365512399318, "standard_deviation": 0.035833790175737186, "minimum": -0.14984403354476242, "maximum": -0.0423016079956757, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900133289421356, "standard_deviation": 0.001022067491123405, "minimum": 0.9885166228494652, "maximum": 0.9916044259623518, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0580080129289557, "standard_deviation": 0.032246331933795364, "minimum": 1.0105043422353874, "maximum": 1.0966300929174202, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2857673558224843, "standard_deviation": 0.1211582305203564, "minimum": 0.12168059573114255, "maximum": 0.47915411732407404, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08918961094589636, "standard_deviation": 0.004632781156655949, "minimum": 0.08169610913766508, "maximum": 0.0954138896121347, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968538668302992, "standard_deviation": 0.0005229029702191101, "minimum": 0.9960111951454543, "maximum": 0.9976400095515078, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6149451637959136, "standard_deviation": 0.029483951690595308, "minimum": 0.5764818929884343, "maximum": 0.6524099436583051, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3377984322485771, "standard_deviation": 0.07281022356682589, "minimum": 0.21225645447718575, "maximum": 0.44237094476384486, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04848906282285436, "standard_deviation": 0.004551185578403488, "minimum": 0.04153030137010387, "maximum": 0.05606812834986307, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0022", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.7640480995178223, "predicted_mos": 2.7640480995178223, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "945bfccdf42a6b444111727c5dbec3bb0ec76b2958e5cfd92a8f996fecf5c10e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.428174099884927, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0214411435965665, "全帯域_10ms_rms_envelope_nrmse": 0.0841293005532006, "全帯域_10ms_envelope_correlation": 0.991079299329291, "全帯域_10ms_modulation_mae_db": 0.2058144111283093, "全帯域_40ms_log_rms_mae_db": 0.6009881177606599, "全帯域_40ms_rms_envelope_nrmse": 0.04154247720777786, "全帯域_40ms_envelope_correlation": 0.9976576045656774, "全帯域_40ms_modulation_mae_db": 0.2825313746872642, "300–1000 Hz_10ms_log_rms_mae_db": 1.5137868622927044, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12315968185152462, "300–1000 Hz_10ms_envelope_correlation": 0.9870570070302477, "300–1000 Hz_10ms_modulation_mae_db": 0.42425613267997303, "300–1000 Hz_40ms_log_rms_mae_db": 1.062800662048859, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07112198401134993, "300–1000 Hz_40ms_envelope_correlation": 0.9952325532222255, "300–1000 Hz_40ms_modulation_mae_db": 0.7788245469189837, "1000–3000 Hz_10ms_log_rms_mae_db": 2.052090491836376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21275400900763422, "1000–3000 Hz_10ms_envelope_correlation": 0.9645559751049342, "1000–3000 Hz_10ms_modulation_mae_db": 0.6700654199587319, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4525493483197813, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12511966762151, "1000–3000 Hz_40ms_envelope_correlation": 0.9876795747546868, "1000–3000 Hz_40ms_modulation_mae_db": 0.9892348403748277, "3000–6000 Hz_10ms_log_rms_mae_db": 2.168260841234532, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.33481018542051577, "3000–6000 Hz_10ms_envelope_correlation": 0.9413652818469503, "3000–6000 Hz_10ms_modulation_mae_db": 0.699697904867151, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7100237137952239, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26605719863447225, "3000–6000 Hz_40ms_envelope_correlation": 0.9628429467924551, "3000–6000 Hz_40ms_modulation_mae_db": 0.900134338426758, "mr_stft_db": 5.937592249038193, "band_mae_db": 0.3495358181279455, "rms_gain_error_db": -0.061282966418962394, "output_peak": 0.666804140098842}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0326429671538542, "全帯域_10ms_rms_envelope_nrmse": 0.09279192549682393, "全帯域_10ms_envelope_correlation": 0.989283800454306, "全帯域_10ms_modulation_mae_db": 0.25636986636614734, "全帯域_40ms_log_rms_mae_db": 0.5943666943459696, "全帯域_40ms_rms_envelope_nrmse": 0.05462546184090806, "全帯域_40ms_envelope_correlation": 0.9961831009205758, "全帯域_40ms_modulation_mae_db": 0.1038456350072165, "300–1000 Hz_10ms_log_rms_mae_db": 1.6309663409292037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15656002189604196, "300–1000 Hz_10ms_envelope_correlation": 0.9794913022115525, "300–1000 Hz_10ms_modulation_mae_db": 0.6875342571800124, "300–1000 Hz_40ms_log_rms_mae_db": 1.1584645699083704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09790803142372272, "300–1000 Hz_40ms_envelope_correlation": 0.9919704551927061, "300–1000 Hz_40ms_modulation_mae_db": 0.8814526170137746, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0190720116653544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22829548796095814, "1000–3000 Hz_10ms_envelope_correlation": 0.9594107555635764, "1000–3000 Hz_10ms_modulation_mae_db": 1.0169058240619455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.460856609569469, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14434984712457363, "1000–3000 Hz_40ms_envelope_correlation": 0.9822686358792697, "1000–3000 Hz_40ms_modulation_mae_db": 0.7872469270139434, "3000–6000 Hz_10ms_log_rms_mae_db": 2.129362973798265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3169299694032388, "3000–6000 Hz_10ms_envelope_correlation": 0.9403779235561526, "3000–6000 Hz_10ms_modulation_mae_db": 0.3130041778581259, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6562994363322407, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24348218694044704, "3000–6000 Hz_40ms_envelope_correlation": 0.9635685507782414, "3000–6000 Hz_40ms_modulation_mae_db": 0.26890534556516005, "mr_stft_db": 5.948687642780079, "band_mae_db": 0.3801375884831271, "rms_gain_error_db": -0.13915689170529572, "output_peak": 0.6663641088192371}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0095141146348285, "全帯域_10ms_rms_envelope_nrmse": 0.0810495439654607, "全帯域_10ms_envelope_correlation": 0.9917639260903424, "全帯域_10ms_modulation_mae_db": 0.41635996783197876, "全帯域_40ms_log_rms_mae_db": 0.5617369493066716, "全帯域_40ms_rms_envelope_nrmse": 0.04013844918334177, "全帯域_40ms_envelope_correlation": 0.9978773734338243, "全帯域_40ms_modulation_mae_db": 0.5990874245693849, "300–1000 Hz_10ms_log_rms_mae_db": 1.5553372252592177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16161388646010996, "300–1000 Hz_10ms_envelope_correlation": 0.9778275532168573, "300–1000 Hz_10ms_modulation_mae_db": 0.6717830540284232, "300–1000 Hz_40ms_log_rms_mae_db": 1.1239393283552048, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09316147295505353, "300–1000 Hz_40ms_envelope_correlation": 0.9925151234271838, "300–1000 Hz_40ms_modulation_mae_db": 1.03478793236094, "1000–3000 Hz_10ms_log_rms_mae_db": 2.017781465736147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24008879430110192, "1000–3000 Hz_10ms_envelope_correlation": 0.9544122190541853, "1000–3000 Hz_10ms_modulation_mae_db": 0.7195572768457775, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4298703710416234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1531608716510052, "1000–3000 Hz_40ms_envelope_correlation": 0.9804394719835215, "1000–3000 Hz_40ms_modulation_mae_db": 0.8911770985242212, "3000–6000 Hz_10ms_log_rms_mae_db": 2.149339850625527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31074166061408, "3000–6000 Hz_10ms_envelope_correlation": 0.9455304523724392, "3000–6000 Hz_10ms_modulation_mae_db": 0.4135524991448243, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6054941935787563, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23529455114277958, "3000–6000 Hz_40ms_envelope_correlation": 0.9680582746902711, "3000–6000 Hz_40ms_modulation_mae_db": 0.40110613754492813, "mr_stft_db": 5.950090232739356, "band_mae_db": 0.3512290956494388, "rms_gain_error_db": -0.086051432082042, "output_peak": 0.6012808850895878}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9531445933452452, "全帯域_10ms_rms_envelope_nrmse": 0.0830456316276146, "全帯域_10ms_envelope_correlation": 0.9912966488255552, "全帯域_10ms_modulation_mae_db": 0.4043457320935997, "全帯域_40ms_log_rms_mae_db": 0.5517373963925809, "全帯域_40ms_rms_envelope_nrmse": 0.0498511048340547, "全帯域_40ms_envelope_correlation": 0.9965625378275627, "全帯域_40ms_modulation_mae_db": 0.37845328403292394, "300–1000 Hz_10ms_log_rms_mae_db": 1.54449773286273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13517204286237602, "300–1000 Hz_10ms_envelope_correlation": 0.9843581234932142, "300–1000 Hz_10ms_modulation_mae_db": 0.5184978153611152, "300–1000 Hz_40ms_log_rms_mae_db": 1.0987041503975379, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08083219728364707, "300–1000 Hz_40ms_envelope_correlation": 0.9938765312635431, "300–1000 Hz_40ms_modulation_mae_db": 0.9447454752903779, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0104080709428365, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22544362504081128, "1000–3000 Hz_10ms_envelope_correlation": 0.9599293583860111, "1000–3000 Hz_10ms_modulation_mae_db": 0.5721703123716989, "1000–3000 Hz_40ms_log_rms_mae_db": 1.417335897461417, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1322534062867917, "1000–3000 Hz_40ms_envelope_correlation": 0.9853147654678068, "1000–3000 Hz_40ms_modulation_mae_db": 0.9695592206487417, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1798820201665112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30520636987976724, "3000–6000 Hz_10ms_envelope_correlation": 0.9450899226641798, "3000–6000 Hz_10ms_modulation_mae_db": 0.2704790866730056, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6746602914347724, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24039009484951815, "3000–6000 Hz_40ms_envelope_correlation": 0.9647919241846057, "3000–6000 Hz_40ms_modulation_mae_db": 0.4360514402274898, "mr_stft_db": 5.945255930856966, "band_mae_db": 0.3106089406751589, "rms_gain_error_db": -0.038784569197683076, "output_peak": 0.5847435507002408}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9959444863237831, "全帯域_10ms_rms_envelope_nrmse": 0.09030549709196797, "全帯域_10ms_envelope_correlation": 0.9898372385222669, "全帯域_10ms_modulation_mae_db": 0.13652308114364511, "全帯域_40ms_log_rms_mae_db": 0.5362520650197293, "全帯域_40ms_rms_envelope_nrmse": 0.04968779803425551, "全帯域_40ms_envelope_correlation": 0.9966322406281974, "全帯域_40ms_modulation_mae_db": 0.18189867524100164, "300–1000 Hz_10ms_log_rms_mae_db": 1.6054529031713045, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16408449264520084, "300–1000 Hz_10ms_envelope_correlation": 0.9769574188566701, "300–1000 Hz_10ms_modulation_mae_db": 0.5758574977179022, "300–1000 Hz_40ms_log_rms_mae_db": 1.1262395140432857, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10379844911559566, "300–1000 Hz_40ms_envelope_correlation": 0.9899106338667796, "300–1000 Hz_40ms_modulation_mae_db": 0.387246297101524, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0260967363874163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22752231986102756, "1000–3000 Hz_10ms_envelope_correlation": 0.9599043501577629, "1000–3000 Hz_10ms_modulation_mae_db": 0.7639133970507269, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4766463753503087, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14593914529158158, "1000–3000 Hz_40ms_envelope_correlation": 0.9819801905777664, "1000–3000 Hz_40ms_modulation_mae_db": 0.7631069633813626, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1714637210251233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3003175040370119, "3000–6000 Hz_10ms_envelope_correlation": 0.9491623509295105, "3000–6000 Hz_10ms_modulation_mae_db": 0.2668461526966454, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6546847442560249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2316475461751169, "3000–6000 Hz_40ms_envelope_correlation": 0.96905905889664, "3000–6000 Hz_40ms_modulation_mae_db": 0.1477555012622603, "mr_stft_db": 5.944889722458721, "band_mae_db": 0.3307997709036708, "rms_gain_error_db": -0.017859238250145026, "output_peak": 0.7265616262191001}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.990890932751887, "全帯域_10ms_rms_envelope_nrmse": 0.0876032224970928, "全帯域_10ms_envelope_correlation": 0.9903202107048067, "全帯域_10ms_modulation_mae_db": 0.3461587702665225, "全帯域_40ms_log_rms_mae_db": 0.5496320807839664, "全帯域_40ms_rms_envelope_nrmse": 0.04890074885925585, "全帯域_40ms_envelope_correlation": 0.9967288200106751, "全帯域_40ms_modulation_mae_db": 0.41665570332830604, "300–1000 Hz_10ms_log_rms_mae_db": 1.6421403545248177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15357691236202184, "300–1000 Hz_10ms_envelope_correlation": 0.9797855294378652, "300–1000 Hz_10ms_modulation_mae_db": 0.5916425805286616, "300–1000 Hz_40ms_log_rms_mae_db": 1.1873776212228346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09383708935079167, "300–1000 Hz_40ms_envelope_correlation": 0.9917009972816403, "300–1000 Hz_40ms_modulation_mae_db": 0.8036966560959242, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0729694287869496, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24782547185934276, "1000–3000 Hz_10ms_envelope_correlation": 0.9513723348254024, "1000–3000 Hz_10ms_modulation_mae_db": 0.2488652114845374, "1000–3000 Hz_40ms_log_rms_mae_db": 1.524298730976472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15856369858226765, "1000–3000 Hz_40ms_envelope_correlation": 0.979504957039406, "1000–3000 Hz_40ms_modulation_mae_db": 0.28788437486811663, "3000–6000 Hz_10ms_log_rms_mae_db": 2.139328988595071, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3046797912189866, "3000–6000 Hz_10ms_envelope_correlation": 0.9490991575438744, "3000–6000 Hz_10ms_modulation_mae_db": 1.0678900232052573, "3000–6000 Hz_40ms_log_rms_mae_db": 1.603903174420623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2382597910638659, "3000–6000 Hz_40ms_envelope_correlation": 0.9684466683595415, "3000–6000 Hz_40ms_modulation_mae_db": 1.4174413769226295, "mr_stft_db": 5.953597943228768, "band_mae_db": 0.3458646063414193, "rms_gain_error_db": -0.06447921198568839, "output_peak": 0.8706850583046789}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0097086044127648, "全帯域_10ms_rms_envelope_nrmse": 0.09060191727468865, "全帯域_10ms_envelope_correlation": 0.9897416100169313, "全帯域_10ms_modulation_mae_db": 0.15986080697873925, "全帯域_40ms_log_rms_mae_db": 0.5764062194245458, "全帯域_40ms_rms_envelope_nrmse": 0.04666537131609115, "全帯域_40ms_envelope_correlation": 0.9971363869738239, "全帯域_40ms_modulation_mae_db": 0.06185860959657886, "300–1000 Hz_10ms_log_rms_mae_db": 1.646109412651561, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1424093111126724, "300–1000 Hz_10ms_envelope_correlation": 0.9829606558779165, "300–1000 Hz_10ms_modulation_mae_db": 0.5166672436366642, "300–1000 Hz_40ms_log_rms_mae_db": 1.1847888492925505, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08859746778053874, "300–1000 Hz_40ms_envelope_correlation": 0.9932364372931907, "300–1000 Hz_40ms_modulation_mae_db": 0.8702082341227099, "1000–3000 Hz_10ms_log_rms_mae_db": 2.008262336275735, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22181604693603318, "1000–3000 Hz_10ms_envelope_correlation": 0.9612099290984393, "1000–3000 Hz_10ms_modulation_mae_db": 1.0180818899090143, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3958024442578958, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13804795314342555, "1000–3000 Hz_40ms_envelope_correlation": 0.9842537219387977, "1000–3000 Hz_40ms_modulation_mae_db": 0.9135139280787006, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1559054495680257, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3224631598975565, "3000–6000 Hz_10ms_envelope_correlation": 0.9366259411979181, "3000–6000 Hz_10ms_modulation_mae_db": 1.1337571205513752, "3000–6000 Hz_40ms_log_rms_mae_db": 1.653226082075682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25608281225240176, "3000–6000 Hz_40ms_envelope_correlation": 0.9587415031554145, "3000–6000 Hz_40ms_modulation_mae_db": 1.384278811141402, "mr_stft_db": 5.9466648102372694, "band_mae_db": 0.3663709021400143, "rms_gain_error_db": -0.10326577776965469, "output_peak": 0.6382358745030754}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0497080351571733, "全帯域_10ms_rms_envelope_nrmse": 0.08742787684445129, "全帯域_10ms_envelope_correlation": 0.9905323753579456, "全帯域_10ms_modulation_mae_db": 0.2536068857760173, "全帯域_40ms_log_rms_mae_db": 0.5615125641456913, "全帯域_40ms_rms_envelope_nrmse": 0.046032731448827156, "全帯域_40ms_envelope_correlation": 0.9973437165102965, "全帯域_40ms_modulation_mae_db": 0.39245818132907134, "300–1000 Hz_10ms_log_rms_mae_db": 1.5313440735688197, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1350841542703508, "300–1000 Hz_10ms_envelope_correlation": 0.9848654672254289, "300–1000 Hz_10ms_modulation_mae_db": 0.39423563142832574, "300–1000 Hz_40ms_log_rms_mae_db": 1.0088023277681215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07760737288478917, "300–1000 Hz_40ms_envelope_correlation": 0.9951787184657557, "300–1000 Hz_40ms_modulation_mae_db": 0.5258762491353766, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0439124900594274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2652651744429258, "1000–3000 Hz_10ms_envelope_correlation": 0.9475033590422808, "1000–3000 Hz_10ms_modulation_mae_db": 0.7025004576550739, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4469878853970575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1711806142766458, "1000–3000 Hz_40ms_envelope_correlation": 0.9763296066596963, "1000–3000 Hz_40ms_modulation_mae_db": 0.5346632065160545, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1888748909870666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.311544640797427, "3000–6000 Hz_10ms_envelope_correlation": 0.9409433009049494, "3000–6000 Hz_10ms_modulation_mae_db": 1.3599609568260558, "3000–6000 Hz_40ms_log_rms_mae_db": 1.696099070686873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24097788319632063, "3000–6000 Hz_40ms_envelope_correlation": 0.9635610135895757, "3000–6000 Hz_40ms_modulation_mae_db": 1.383103852696311, "mr_stft_db": 5.946945477979267, "band_mae_db": 0.30561928182376263, "rms_gain_error_db": -0.1348137011910058, "output_peak": 0.6367294782857404}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.959097531250677, "spectral_convergence": 0.2791917713506903, "spectral_convergence_status": "ok", "amplitude_floor": 2.1903561449018732e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.676906747007233, "spectral_convergence": 0.2796646366293311, "spectral_convergence_status": "ok", "amplitude_floor": 2.0456135368211886e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.176772468856668, "spectral_convergence": 0.2938517014138143, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394418151882858e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0025307845601282375, "output_power": 0.002458297902376768, "delta_db": -0.12620665664857675, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026710985360254404, "output_power": 0.0026333928851813484, "delta_db": -0.06174251443388173, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004653266401104453, "output_power": 0.0004914025870643458, "delta_db": 0.23679521806326667, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012444324831517497, "output_power": 0.00011096941281725479, "delta_db": -0.4976805086101233, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00016257063671523992, "output_power": 0.00016606898401005655, "delta_db": 0.092464220428747, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.589634923518972e-05, "output_power": 4.605564414750492e-05, "delta_db": 1.0823257905830772, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3495358181279455, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.061282966418962394, "pitch": null, "output_rms_dbfs": -22.28692776086092, "output_peak": 0.666804140098842, "reference_rms_dbfs": -22.225644794441955, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.937592249038193}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031508938028130336, "log_rms_mae_db": 1.0214411435965665, "rms_envelope_nrmse": 0.0841293005532006, "envelope_correlation": 0.991079299329291, "rms_gain_error_db": -0.06125926674426872, "reference_mean_power": 0.0059947180836205155, "output_mean_power": 0.005910753355898762}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.905621652404067e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011843610245185533, "output_power": 0.0012109996544612672, "reference_active": true, "delta_db": 0.09659912228634894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002763334576078544, "output_power": 0.00025673416474935066, "reference_active": true, "delta_db": -0.31949805142182514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0009701982183084e-05, "output_power": 5.238281972918837e-05, "reference_active": true, "delta_db": 0.20134605967675384}], "active_band_count": 3, "active_band_mae_db": 0.2058144111283093}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002456604014165908, "log_rms_mae_db": 0.6009881177606599, "rms_envelope_nrmse": 0.04154247720777786, "envelope_correlation": 0.9976576045656774, "rms_gain_error_db": -0.061148235097726183, "reference_mean_power": 0.006022035579915752, "output_mean_power": 0.005937840035776226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5974133279358877e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001059507007170014, "output_power": 0.0010918895257039922, "reference_active": true, "delta_db": 0.13074866591082343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014348331799351297, "output_power": 0.00013269045187404584, "reference_active": true, "delta_db": -0.33961737932855873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4072600706208148e-06, "output_power": 3.123794644027472e-06, "reference_active": true, "delta_db": -0.37722807882241033}], "active_band_count": 3, "active_band_mae_db": 0.2825313746872642}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026889287764829997, "log_rms_mae_db": 1.5137868622927044, "rms_envelope_nrmse": 0.12315968185152462, "envelope_correlation": 0.9870570070302477, "rms_gain_error_db": -0.03665670404041854, "reference_mean_power": 0.0016839373335741825, "output_mean_power": 0.0016697838455643908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1138357532326787e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006324020549599114, "output_power": 0.0006344723216753284, "reference_active": true, "delta_db": 0.01419408628823025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002190350723697022, "output_power": 0.0001933038827965298, "reference_active": true, "delta_db": -0.5427308301229499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1194141233781235e-05, "output_power": 2.6453837775004923e-05, "reference_active": true, "delta_db": -0.715843481628739}], "active_band_count": 3, "active_band_mae_db": 0.42425613267997303}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001970017370857221, "log_rms_mae_db": 1.062800662048859, "rms_envelope_nrmse": 0.07112198401134993, "envelope_correlation": 0.9952325532222255, "rms_gain_error_db": -0.036674748656806276, "reference_mean_power": 0.001692399182790771, "output_mean_power": 0.001678167600287371}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.64481232375879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970882254118313, "output_power": 0.0006025404380901358, "reference_active": true, "delta_db": 0.03947691908927132}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011301601543605377, "output_power": 0.00010144622593447126, "reference_active": true, "delta_db": -0.46904096694786035}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.966963236644965e-06, "output_power": 1.947675447606746e-06, "reference_active": true, "delta_db": -1.8279557547198197}], "active_band_count": 3, "active_band_mae_db": 0.7788245469189837}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.658803471872848e-05, "log_rms_mae_db": 2.052090491836376, "rms_envelope_nrmse": 0.21275400900763422, "envelope_correlation": 0.9645559751049342, "rms_gain_error_db": -0.34130684054766236, "reference_mean_power": 0.00020257298157917204, "output_mean_power": 0.000187262508298631}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0818748871802829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.592086384233793e-05, "output_power": 5.6308904887370225e-05, "reference_active": true, "delta_db": -0.6844580868971483}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.925318155530915e-05, "output_power": 1.469312924838389e-05, "reference_active": true, "delta_db": -1.173882073019302}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.819375654274508e-06, "output_power": 3.688134625786668e-06, "reference_active": true, "delta_db": -0.15185609995974553}], "active_band_count": 3, "active_band_mae_db": 0.6700654199587319}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.479997453028037e-05, "log_rms_mae_db": 1.4525493483197813, "rms_envelope_nrmse": 0.12511966762151, "envelope_correlation": 0.9876795747546868, "rms_gain_error_db": -0.3414012049402134, "reference_mean_power": 0.00020359039158760131, "output_mean_power": 0.00018819893315248884}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.438814668171717e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.33475246376368e-05, "output_power": 5.4369515833838765e-05, "reference_active": true, "delta_db": -0.6637418351596082}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.661338887813947e-06, "output_power": 6.811986211966479e-06, "reference_active": true, "delta_db": -1.5176355574787572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1289641878321164e-07, "output_power": 2.610765587720049e-07, "reference_active": true, "delta_db": -0.7863271284861179}], "active_band_count": 3, "active_band_mae_db": 0.9892348403748277}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.558280881155462e-05, "log_rms_mae_db": 2.168260841234532, "rms_envelope_nrmse": 0.33481018542051577, "envelope_correlation": 0.9413652818469503, "rms_gain_error_db": 0.6518656400137184, "reference_mean_power": 8.701244577979696e-05, "output_mean_power": 0.0001011039073751656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.56825886194788e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4501869786005614e-05, "output_power": 2.4613066644005462e-05, "reference_active": true, "delta_db": 0.01966500126277622}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.528335672589267e-06, "output_power": 7.269964064708509e-06, "reference_active": true, "delta_db": 2.055936517721915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7809461115210634e-06, "output_power": 1.771338487583478e-06, "reference_active": true, "delta_db": -0.02349219561676143}], "active_band_count": 3, "active_band_mae_db": 0.699697904867151}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.713713309096251e-05, "log_rms_mae_db": 1.7100237137952239, "rms_envelope_nrmse": 0.26605719863447225, "envelope_correlation": 0.9628429467924551, "rms_gain_error_db": 0.6517916138991277, "reference_mean_power": 8.74494743564098e-05, "output_mean_power": 0.00010160997973378273}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.348274544921528e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.135032255455729e-05, "output_power": 2.208940556490624e-05, "reference_active": true, "delta_db": 0.147795883762601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4135743662495686e-06, "output_power": 3.907179360264501e-06, "reference_active": true, "delta_db": 2.092026634057125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6206121669861435e-07, "output_power": 1.8019268011697274e-07, "reference_active": true, "delta_db": 0.46058049746054824}], "active_band_count": 3, "active_band_mae_db": 0.900134338426758}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8911755319375176, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4773dc17e78b6faa569c9d6566ff48c58bc7d260aa6e5ffe6ef669923aeb548b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.957287285154074, "standard_deviation": 0.005676235529731404, "minimum": 0.9475033590422808, "maximum": 0.9645559751049342, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0313241289612805, "standard_deviation": 0.02284797947694767, "minimum": 2.008262336275735, "maximum": 2.0729694287869496, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7140074736671882, "standard_deviation": 0.24648050710614985, "minimum": 0.2488652114845374, "maximum": 1.0180818899090143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23362636617622937, "standard_deviation": 0.01669972454344401, "minimum": 0.21275400900763422, "maximum": 0.2652651744429258, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9822213655376189, "standard_deviation": 0.0035620797660139325, "minimum": 0.9763296066596963, "maximum": 0.9876795747546868, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.450543457796753, "standard_deviation": 0.03922631507521645, "minimum": 1.3958024442578958, "maximum": 1.524298730976472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7670483199257461, "standard_deviation": 0.2423875550866577, "minimum": 0.28788437486811663, "maximum": 0.9892348403748277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14607690049722513, "standard_deviation": 0.01480376161991501, "minimum": 0.12511966762151, "maximum": 0.1711806142766458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9435242913769968, "standard_deviation": 0.0044425287391658035, "minimum": 0.9366259411979181, "maximum": 0.9491623509295105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.160302342000015, "standard_deviation": 0.02043911999257203, "minimum": 2.129362973798265, "maximum": 2.1888748909870666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6906484902278051, "standard_deviation": 0.441145532552865, "minimum": 0.2668461526966454, "maximum": 1.3599609568260558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.313336660158573, "standard_deviation": 0.01119823926203071, "minimum": 0.3003175040370119, "maximum": 0.33481018542051577, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9648837425558431, "standard_deviation": 0.0034966327729956735, "minimum": 0.9587415031554145, "maximum": 0.96905905889664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6567988383225245, "standard_deviation": 0.03806766223343758, "minimum": 1.603903174420623, "maximum": 1.7100237137952239, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7923471004733673, "standard_deviation": 0.5439594683751418, "minimum": 0.1477555012622603, "maximum": 1.4174413769226295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2440240080318653, "standard_deviation": 0.011440906819140725, "minimum": 0.2316475461751169, "maximum": 0.26605719863447225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.981662882168719, "standard_deviation": 0.00365265113897157, "minimum": 0.9769574188566701, "maximum": 0.9870570070302477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.583704363157545, "standard_deviation": 0.053435889557571135, "minimum": 1.5137868622927044, "maximum": 1.646109412651561, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5475592765701347, "standard_deviation": 0.10574859726857921, "minimum": 0.39423563142832574, "maximum": 0.6875342571800124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1464575629325373, "standard_deviation": 0.014682665196410505, "minimum": 0.12315968185152462, "maximum": 0.16408449264520084, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.992952681251628, "standard_deviation": 0.0018151766600204553, "minimum": 0.9899106338667796, "maximum": 0.9952325532222255, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1188896278795957, "standard_deviation": 0.061355693865576766, "minimum": 1.0088023277681215, "maximum": 1.1873776212228346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7783547510049513, "standard_deviation": 0.21707922564548202, "minimum": 0.387246297101524, "maximum": 1.03478793236094, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08835800810068606, "standard_deviation": 0.011032779882199376, "minimum": 0.07112198401134993, "maximum": 0.10379844911559566, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3425207505180672, "standard_deviation": 0.025739206929487463, "minimum": 0.30561928182376263, "maximum": 0.3801375884831271, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.946715501164828, "standard_deviation": 0.004650639989029867, "minimum": 5.937592249038193, "maximum": 5.953597943228768, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6739255902525628, "standard_deviation": 0.09061750660790494, "minimum": 0.5847435507002408, "maximum": 0.8706850583046789, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08071172357505964, "standard_deviation": 0.04347306510951484, "minimum": -0.13915689170529572, "maximum": -0.017859238250145026, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9904818886626806, "standard_deviation": 0.0008528979083025572, "minimum": 0.989283800454306, "maximum": 0.9917639260903424, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0078743596720128, "standard_deviation": 0.029227573056932955, "minimum": 0.9531445933452452, "maximum": 1.0497080351571733, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2723799401981199, "standard_deviation": 0.10673458105016685, "minimum": 0.13652308114364511, "maximum": 0.41635996783197876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08711936441891258, "standard_deviation": 0.004091198997426893, "minimum": 0.0810495439654607, "maximum": 0.09279192549682393, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970152226088291, "standard_deviation": 0.0005861008361731121, "minimum": 0.9961831009205758, "maximum": 0.9978773734338243, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5665790108974769, "standard_deviation": 0.02245320869460336, "minimum": 0.5362520650197293, "maximum": 0.6009881177606599, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.30209861097396845, "standard_deviation": 0.18014715131556153, "minimum": 0.06185860959657886, "maximum": 0.5990874245693849, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04718051784056401, "standard_deviation": 0.004702064857639217, "minimum": 0.04013844918334177, "maximum": 0.05462546184090806, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0022", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.317538022994995, "predicted_mos": 2.317538022994995, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a8120da196ea612fd12a4624466bec7b30815eebbf09a79d31ceb4c559737063", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.979163299780339, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9743659278134723, "全帯域_10ms_rms_envelope_nrmse": 0.08464951844983032, "全帯域_10ms_envelope_correlation": 0.9910070075669911, "全帯域_10ms_modulation_mae_db": 0.2635475488845142, "全帯域_40ms_log_rms_mae_db": 0.5392915408024948, "全帯域_40ms_rms_envelope_nrmse": 0.04739090606871332, "全帯域_40ms_envelope_correlation": 0.9969753009879522, "全帯域_40ms_modulation_mae_db": 0.3358560613130827, "300–1000 Hz_10ms_log_rms_mae_db": 1.5654122081291144, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13539993894486108, "300–1000 Hz_10ms_envelope_correlation": 0.9845070016229898, "300–1000 Hz_10ms_modulation_mae_db": 0.6833883491604548, "300–1000 Hz_40ms_log_rms_mae_db": 1.047164874427629, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09009104561936487, "300–1000 Hz_40ms_envelope_correlation": 0.9927175728314647, "300–1000 Hz_40ms_modulation_mae_db": 0.9449683046006484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9226818076701626, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21495736168097163, "1000–3000 Hz_10ms_envelope_correlation": 0.9645911871649624, "1000–3000 Hz_10ms_modulation_mae_db": 0.29060900946207835, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4061514312429804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.141334813559516, "1000–3000 Hz_40ms_envelope_correlation": 0.9833196924692733, "1000–3000 Hz_40ms_modulation_mae_db": 0.2190860667224662, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1457527694327547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3119708097511159, "3000–6000 Hz_10ms_envelope_correlation": 0.94581303914961, "3000–6000 Hz_10ms_modulation_mae_db": 0.36061679366628024, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6423025845302122, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25045043970413977, "3000–6000 Hz_40ms_envelope_correlation": 0.9642783808156393, "3000–6000 Hz_40ms_modulation_mae_db": 0.8694079081049765, "mr_stft_db": 5.919911051429566, "band_mae_db": 0.2767845637326265, "rms_gain_error_db": -0.08267427733611168, "output_peak": 0.8190118222403688}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0322325516964446, "全帯域_10ms_rms_envelope_nrmse": 0.09451998405109972, "全帯域_10ms_envelope_correlation": 0.988883162705242, "全帯域_10ms_modulation_mae_db": 0.3986543611374033, "全帯域_40ms_log_rms_mae_db": 0.5728346781080017, "全帯域_40ms_rms_envelope_nrmse": 0.049844528071790156, "全帯域_40ms_envelope_correlation": 0.9966343487963955, "全帯域_40ms_modulation_mae_db": 0.4899637609252294, "300–1000 Hz_10ms_log_rms_mae_db": 1.5485533303662284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16670709348215948, "300–1000 Hz_10ms_envelope_correlation": 0.9766117129566563, "300–1000 Hz_10ms_modulation_mae_db": 0.4936790456978441, "300–1000 Hz_40ms_log_rms_mae_db": 1.058448315872354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10787439679888422, "300–1000 Hz_40ms_envelope_correlation": 0.9892292413300272, "300–1000 Hz_40ms_modulation_mae_db": 0.8157679291226178, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8892522324788068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26282689657203107, "1000–3000 Hz_10ms_envelope_correlation": 0.9487236389374706, "1000–3000 Hz_10ms_modulation_mae_db": 0.8211812200317863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3601030956017888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16875861381256663, "1000–3000 Hz_40ms_envelope_correlation": 0.977254793310963, "1000–3000 Hz_40ms_modulation_mae_db": 0.6728032867230103, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1076505463196202, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3145385146138578, "3000–6000 Hz_10ms_envelope_correlation": 0.9429505057462703, "3000–6000 Hz_10ms_modulation_mae_db": 0.34129263702840396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5801991169924545, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24300595435663358, "3000–6000 Hz_40ms_envelope_correlation": 0.964917232660621, "3000–6000 Hz_40ms_modulation_mae_db": 0.5369036003163453, "mr_stft_db": 5.8989205490865, "band_mae_db": 0.33913624150491134, "rms_gain_error_db": -0.005368742272873265, "output_peak": 0.8498082991360283}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0308359484876415, "全帯域_10ms_rms_envelope_nrmse": 0.08876432630158419, "全帯域_10ms_envelope_correlation": 0.9901758212770174, "全帯域_10ms_modulation_mae_db": 0.3927866912496445, "全帯域_40ms_log_rms_mae_db": 0.5875615981303481, "全帯域_40ms_rms_envelope_nrmse": 0.045931499290202966, "全帯域_40ms_envelope_correlation": 0.9971328698480886, "全帯域_40ms_modulation_mae_db": 0.5000493784680352, "300–1000 Hz_10ms_log_rms_mae_db": 1.511287031489635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14881483803675602, "300–1000 Hz_10ms_envelope_correlation": 0.981040128392658, "300–1000 Hz_10ms_modulation_mae_db": 0.792566758769985, "300–1000 Hz_40ms_log_rms_mae_db": 1.0572314335304598, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08200480338910486, "300–1000 Hz_40ms_envelope_correlation": 0.99373254026433, "300–1000 Hz_40ms_modulation_mae_db": 0.8831547437726951, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9486322031604533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20476725067652257, "1000–3000 Hz_10ms_envelope_correlation": 0.9671883536900371, "1000–3000 Hz_10ms_modulation_mae_db": 0.6850685969536706, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3331897120561618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12650885116714694, "1000–3000 Hz_40ms_envelope_correlation": 0.9863660197003994, "1000–3000 Hz_40ms_modulation_mae_db": 0.2020969217449667, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0954028690182946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29978609475359086, "3000–6000 Hz_10ms_envelope_correlation": 0.9486338699177741, "3000–6000 Hz_10ms_modulation_mae_db": 0.4558039411131136, "3000–6000 Hz_40ms_log_rms_mae_db": 1.586580249446712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22908830469238753, "3000–6000 Hz_40ms_envelope_correlation": 0.9691940987722523, "3000–6000 Hz_40ms_modulation_mae_db": 0.7120369924097907, "mr_stft_db": 5.89593296854454, "band_mae_db": 0.20433319589838392, "rms_gain_error_db": 0.00961987483609462, "output_peak": 0.6808488059535003}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.012250526553434, "全帯域_10ms_rms_envelope_nrmse": 0.07581382663734408, "全帯域_10ms_envelope_correlation": 0.9927755304524798, "全帯域_10ms_modulation_mae_db": 0.25063908566827386, "全帯域_40ms_log_rms_mae_db": 0.5000324009977929, "全帯域_40ms_rms_envelope_nrmse": 0.0399457923362365, "全帯域_40ms_envelope_correlation": 0.997826577362179, "全帯域_40ms_modulation_mae_db": 0.34910968418087723, "300–1000 Hz_10ms_log_rms_mae_db": 1.5979579380283087, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1511946988091812, "300–1000 Hz_10ms_envelope_correlation": 0.980714713861258, "300–1000 Hz_10ms_modulation_mae_db": 0.6012575473293884, "300–1000 Hz_40ms_log_rms_mae_db": 1.1077568077557822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08602072179561235, "300–1000 Hz_40ms_envelope_correlation": 0.9937547984672674, "300–1000 Hz_40ms_modulation_mae_db": 0.8910651116855307, "1000–3000 Hz_10ms_log_rms_mae_db": 2.008022741657965, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22167660168183703, "1000–3000 Hz_10ms_envelope_correlation": 0.9621020964177578, "1000–3000 Hz_10ms_modulation_mae_db": 0.39238811064838663, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4139579076501634, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13710188580637073, "1000–3000 Hz_40ms_envelope_correlation": 0.9841559692048831, "1000–3000 Hz_40ms_modulation_mae_db": 0.24792283665277812, "3000–6000 Hz_10ms_log_rms_mae_db": 2.125323422472323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3160275678442749, "3000–6000 Hz_10ms_envelope_correlation": 0.9405688584739695, "3000–6000 Hz_10ms_modulation_mae_db": 1.0349722280077762, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6008801020119707, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25217515706244126, "3000–6000 Hz_40ms_envelope_correlation": 0.9608434697038841, "3000–6000 Hz_40ms_modulation_mae_db": 0.8750605195663012, "mr_stft_db": 5.899597777141736, "band_mae_db": 0.23908976942148508, "rms_gain_error_db": -0.05392027030180202, "output_peak": 0.5865585601989121}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0620020680662847, "全帯域_10ms_rms_envelope_nrmse": 0.08694679659777554, "全帯域_10ms_envelope_correlation": 0.9907151514686591, "全帯域_10ms_modulation_mae_db": 0.5178572894021097, "全帯域_40ms_log_rms_mae_db": 0.5761700750378679, "全帯域_40ms_rms_envelope_nrmse": 0.047723595652009725, "全帯域_40ms_envelope_correlation": 0.9972522391069812, "全帯域_40ms_modulation_mae_db": 0.6673050359944318, "300–1000 Hz_10ms_log_rms_mae_db": 1.523225257619732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12950282343864775, "300–1000 Hz_10ms_envelope_correlation": 0.9865862503754077, "300–1000 Hz_10ms_modulation_mae_db": 0.5132168235617246, "300–1000 Hz_40ms_log_rms_mae_db": 1.0695242277332437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07611084097074479, "300–1000 Hz_40ms_envelope_correlation": 0.9959804188506312, "300–1000 Hz_40ms_modulation_mae_db": 0.9311840352234967, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9793412408467597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22735046377971996, "1000–3000 Hz_10ms_envelope_correlation": 0.9596074242397857, "1000–3000 Hz_10ms_modulation_mae_db": 0.8744432180265145, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3544642323746583, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13410788424085401, "1000–3000 Hz_40ms_envelope_correlation": 0.984672770878686, "1000–3000 Hz_40ms_modulation_mae_db": 0.4459709010931204, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1447668565988667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3205933630181098, "3000–6000 Hz_10ms_envelope_correlation": 0.941657709675596, "3000–6000 Hz_10ms_modulation_mae_db": 0.7147209950263257, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6625014286088, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24979282796555402, "3000–6000 Hz_40ms_envelope_correlation": 0.963622206657056, "3000–6000 Hz_40ms_modulation_mae_db": 0.753582158888468, "mr_stft_db": 5.908635498850736, "band_mae_db": 0.3643897399139766, "rms_gain_error_db": -0.1599953035966768, "output_peak": 0.6036414481212102}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0699818605280556, "全帯域_10ms_rms_envelope_nrmse": 0.09036962569806845, "全帯域_10ms_envelope_correlation": 0.9898407077412923, "全帯域_10ms_modulation_mae_db": 0.18752792319669234, "全帯域_40ms_log_rms_mae_db": 0.6133071263339487, "全帯域_40ms_rms_envelope_nrmse": 0.049934556922742845, "全帯域_40ms_envelope_correlation": 0.9966062306194641, "全帯域_40ms_modulation_mae_db": 0.20713902455356017, "300–1000 Hz_10ms_log_rms_mae_db": 1.532246586236989, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14385128940217043, "300–1000 Hz_10ms_envelope_correlation": 0.9825089124261916, "300–1000 Hz_10ms_modulation_mae_db": 0.34665693942696624, "300–1000 Hz_40ms_log_rms_mae_db": 1.0738191128014871, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0983606102563631, "300–1000 Hz_40ms_envelope_correlation": 0.9909915824295253, "300–1000 Hz_40ms_modulation_mae_db": 0.29791393871460997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9259676837626032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23241755246161963, "1000–3000 Hz_10ms_envelope_correlation": 0.9582003303383057, "1000–3000 Hz_10ms_modulation_mae_db": 0.4163017420856551, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3101475752107141, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14261991338202193, "1000–3000 Hz_40ms_envelope_correlation": 0.9828012888385258, "1000–3000 Hz_40ms_modulation_mae_db": 0.32964156083717505, "3000–6000 Hz_10ms_log_rms_mae_db": 2.097711349888484, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3183039778444672, "3000–6000 Hz_10ms_envelope_correlation": 0.9424123634021786, "3000–6000 Hz_10ms_modulation_mae_db": 0.8827481239728431, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6245258420860968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2551320234379697, "3000–6000 Hz_40ms_envelope_correlation": 0.961980364372811, "3000–6000 Hz_40ms_modulation_mae_db": 1.1566287398227153, "mr_stft_db": 5.927350816281218, "band_mae_db": 0.280014184669893, "rms_gain_error_db": -0.041333726140424784, "output_peak": 0.7215268418897207}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9461770742745885, "全帯域_10ms_rms_envelope_nrmse": 0.08318719753702451, "全帯域_10ms_envelope_correlation": 0.9913484754901983, "全帯域_10ms_modulation_mae_db": 0.1540641329906235, "全帯域_40ms_log_rms_mae_db": 0.5540450177875277, "全帯域_40ms_rms_envelope_nrmse": 0.049876759597753714, "全帯域_40ms_envelope_correlation": 0.9967308810186583, "全帯域_40ms_modulation_mae_db": 0.25604241048937265, "300–1000 Hz_10ms_log_rms_mae_db": 1.4979651122767725, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15040763686019618, "300–1000 Hz_10ms_envelope_correlation": 0.9806615748476029, "300–1000 Hz_10ms_modulation_mae_db": 0.20510499307397734, "300–1000 Hz_40ms_log_rms_mae_db": 0.9886540303515101, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08779952972293967, "300–1000 Hz_40ms_envelope_correlation": 0.9929858183380281, "300–1000 Hz_40ms_modulation_mae_db": 0.3188689414415911, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9510033811516088, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21421266492484303, "1000–3000 Hz_10ms_envelope_correlation": 0.9641323885031317, "1000–3000 Hz_10ms_modulation_mae_db": 0.8717409520574405, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3471964001656531, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1318519724018513, "1000–3000 Hz_40ms_envelope_correlation": 0.9861320509668716, "1000–3000 Hz_40ms_modulation_mae_db": 0.6678524026149484, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0849099324769473, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3039311776858078, "3000–6000 Hz_10ms_envelope_correlation": 0.9469553407302751, "3000–6000 Hz_10ms_modulation_mae_db": 0.2674496090944873, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5699276317500952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23093215359521546, "3000–6000 Hz_40ms_envelope_correlation": 0.9685323985188757, "3000–6000 Hz_40ms_modulation_mae_db": 0.2749014677134183, "mr_stft_db": 5.894188767789875, "band_mae_db": 0.36277722655487055, "rms_gain_error_db": -0.10261702131093195, "output_peak": 0.6301646791754385}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0077812955095484, "全帯域_10ms_rms_envelope_nrmse": 0.08623507965153804, "全帯域_10ms_envelope_correlation": 0.9907817411162613, "全帯域_10ms_modulation_mae_db": 0.2583589067959669, "全帯域_40ms_log_rms_mae_db": 0.592625939486505, "全帯域_40ms_rms_envelope_nrmse": 0.0503206405734578, "全帯域_40ms_envelope_correlation": 0.996735764394457, "全帯域_40ms_modulation_mae_db": 0.5506306697909292, "300–1000 Hz_10ms_log_rms_mae_db": 1.5664104730586197, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14814409827271838, "300–1000 Hz_10ms_envelope_correlation": 0.9812662238203769, "300–1000 Hz_10ms_modulation_mae_db": 0.38128412523940564, "300–1000 Hz_40ms_log_rms_mae_db": 1.1205525779259782, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08785823191948167, "300–1000 Hz_40ms_envelope_correlation": 0.9928888406496816, "300–1000 Hz_40ms_modulation_mae_db": 0.7222449871089633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.958854821175978, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.212561608602656, "1000–3000 Hz_10ms_envelope_correlation": 0.9645023746131428, "1000–3000 Hz_10ms_modulation_mae_db": 0.8981685803117735, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3584182977642207, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14259329096094786, "1000–3000 Hz_40ms_envelope_correlation": 0.9826625504242364, "1000–3000 Hz_40ms_modulation_mae_db": 0.578903820508122, "3000–6000 Hz_10ms_log_rms_mae_db": 2.083313614967426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3164478009422314, "3000–6000 Hz_10ms_envelope_correlation": 0.9432028292048317, "3000–6000 Hz_10ms_modulation_mae_db": 0.8630347141076524, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5597233475623824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24147152556191298, "3000–6000 Hz_40ms_envelope_correlation": 0.9660532756986084, "3000–6000 Hz_40ms_modulation_mae_db": 0.5251883489499125, "mr_stft_db": 5.916086134512018, "band_mae_db": 0.3264609990145776, "rms_gain_error_db": -0.1277992686081559, "output_peak": 0.5836944408924606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.941474714648121, "spectral_convergence": 0.28286583335495236, "spectral_convergence_status": "ok", "amplitude_floor": 2.1903561449018732e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.646212060722872, "spectral_convergence": 0.28126183129686433, "spectral_convergence_status": "ok", "amplitude_floor": 2.0456135368211886e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.172046378917706, "spectral_convergence": 0.2908176377000449, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394418151882858e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0025307845601282375, "output_power": 0.0024897378016177397, "delta_db": -0.07101563003133955, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026710985360254404, "output_power": 0.0025885246426572174, "delta_db": -0.13637605505638375, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004653266401104453, "output_power": 0.0004603964841481434, "delta_db": -0.046259183052798164, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012444324831517497, "output_power": 0.00011987978551889052, "delta_db": -0.16225381616597967, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00016257063671523992, "output_power": 0.00017394395817329868, "delta_db": 0.2936724174875637, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.589634923518972e-05, "output_power": 4.468515923249305e-05, "delta_db": 0.9511302806016942, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2767845637326265, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08267427733611168, "pitch": null, "output_rms_dbfs": -22.30831907177807, "output_peak": 0.8190118222403688, "reference_rms_dbfs": -22.225644794441955, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.919911051429566}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031508938028130336, "log_rms_mae_db": 0.9743659278134723, "rms_envelope_nrmse": 0.08464951844983032, "envelope_correlation": 0.9910070075669911, "rms_gain_error_db": -0.08261663958821999, "reference_mean_power": 0.0059947180836205155, "output_mean_power": 0.005881757300253226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.905621652404067e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011843610245185533, "output_power": 0.001114572656099116, "reference_active": true, "delta_db": -0.26375722726368445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002763334576078544, "output_power": 0.0002479773336002345, "reference_active": true, "delta_db": -0.47021485246549855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0009701982183084e-05, "output_power": 4.9361370568051177e-05, "reference_active": true, "delta_db": -0.05667056692435972}], "active_band_count": 3, "active_band_mae_db": 0.2635475488845142}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002456604014165908, "log_rms_mae_db": 0.5392915408024948, "rms_envelope_nrmse": 0.04739090606871332, "envelope_correlation": 0.9969753009879522, "rms_gain_error_db": -0.08253885357887203, "reference_mean_power": 0.006022035579915752, "output_mean_power": 0.005908665870956917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5974133279358877e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001059507007170014, "output_power": 0.0009898817424157677, "reference_active": true, "delta_db": -0.2952051920353447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014348331799351297, "output_power": 0.00012706282377581197, "reference_active": true, "delta_db": -0.5278290837041317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4072600706208148e-06, "output_power": 3.5551561010881844e-06, "reference_active": true, "delta_db": 0.18453390819977172}], "active_band_count": 3, "active_band_mae_db": 0.3358560613130827}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026889287764829997, "log_rms_mae_db": 1.5654122081291144, "rms_envelope_nrmse": 0.13539993894486108, "envelope_correlation": 0.9845070016229898, "rms_gain_error_db": -0.2145512213875864, "reference_mean_power": 0.0016839373335741825, "output_mean_power": 0.0016027685182732415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1138357532326787e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006324020549599114, "output_power": 0.0005704682526155568, "reference_active": true, "delta_db": -0.4476179217154893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002190350723697022, "output_power": 0.00018209548475387293, "reference_active": true, "delta_db": -0.8021448343329157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1194141233781235e-05, "output_power": 2.594375324113837e-05, "reference_active": true, "delta_db": -0.8004022914329594}], "active_band_count": 3, "active_band_mae_db": 0.6833883491604548}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001970017370857221, "log_rms_mae_db": 1.047164874427629, "rms_envelope_nrmse": 0.09009104561936487, "envelope_correlation": 0.9927175728314647, "rms_gain_error_db": -0.21459532065834896, "reference_mean_power": 0.001692399182790771, "output_mean_power": 0.001610806134519666}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.64481232375879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970882254118313, "output_power": 0.00053690105207826, "reference_active": true, "delta_db": -0.4614425199214512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011301601543605377, "output_power": 9.528743230895942e-05, "reference_active": true, "delta_db": -0.7410436721088175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.966963236644965e-06, "output_power": 2.037371810485304e-06, "reference_active": true, "delta_db": -1.6324187217716761}], "active_band_count": 3, "active_band_mae_db": 0.9449683046006484}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.658803471872848e-05, "log_rms_mae_db": 1.9226818076701626, "rms_envelope_nrmse": 0.21495736168097163, "envelope_correlation": 0.9645911871649624, "rms_gain_error_db": 0.03972010346921875, "reference_mean_power": 0.00020257298157917204, "output_mean_power": 0.00020443419038659546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0818748871802829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.592086384233793e-05, "output_power": 6.570344634058083e-05, "reference_active": true, "delta_db": -0.014347396317248889}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.925318155530915e-05, "output_power": 1.7414828894744532e-05, "reference_active": true, "delta_db": -0.4358329443667819}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.819375654274508e-06, "output_power": 4.208787938850644e-06, "reference_active": true, "delta_db": 0.42164668770220426}], "active_band_count": 3, "active_band_mae_db": 0.29060900946207835}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.479997453028037e-05, "log_rms_mae_db": 1.4061514312429804, "rms_envelope_nrmse": 0.141334813559516, "envelope_correlation": 0.9833196924692733, "rms_gain_error_db": 0.03966857039652838, "reference_mean_power": 0.00020359039158760131, "output_mean_power": 0.00020545851022827735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.438814668171717e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.33475246376368e-05, "output_power": 6.435640729433473e-05, "reference_active": true, "delta_db": 0.06862142454352153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.661338887813947e-06, "output_power": 8.61865459735448e-06, "reference_active": true, "delta_db": -0.49597839788791953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1289641878321164e-07, "output_power": 3.196439069167437e-07, "reference_active": true, "delta_db": 0.0926583777359575}], "active_band_count": 3, "active_band_mae_db": 0.2190860667224662}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.558280881155462e-05, "log_rms_mae_db": 2.1457527694327547, "rms_envelope_nrmse": 0.3119708097511159, "envelope_correlation": 0.94581303914961, "rms_gain_error_db": 0.4058036026383587, "reference_mean_power": 8.701244577979696e-05, "output_mean_power": 9.553483028067502e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.56825886194788e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4501869786005614e-05, "output_power": 2.667809385671239e-05, "reference_active": true, "delta_db": 0.36955568620811285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.528335672589267e-06, "output_power": 4.259186055773925e-06, "reference_active": true, "delta_db": -0.26612000430575355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7809461115210634e-06, "output_power": 1.6070645071559088e-06, "reference_active": true, "delta_db": -0.44617469048497416}], "active_band_count": 3, "active_band_mae_db": 0.36061679366628024}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.713713309096251e-05, "log_rms_mae_db": 1.6423025845302122, "rms_envelope_nrmse": 0.25045043970413977, "envelope_correlation": 0.9642783808156393, "rms_gain_error_db": 0.4057350301588689, "reference_mean_power": 8.74494743564098e-05, "output_mean_power": 9.601314737094004e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.348274544921528e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.135032255455729e-05, "output_power": 2.3747627457462308e-05, "reference_active": true, "delta_db": 0.46215786675697434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4135743662495686e-06, "output_power": 2.207774961259565e-06, "reference_active": true, "delta_db": -0.38705881068176395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6206121669861435e-07, "output_power": 1.0808822634629327e-07, "reference_active": true, "delta_db": -1.7590070468761911}], "active_band_count": 3, "active_band_mae_db": 0.8694079081049765}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8872508854315312, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "aa702bf3e1228ad7522c30f45d00ee453cd34911cc652790b3ca7006209f3bd4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9611309742380743, "standard_deviation": 0.005792047173470866, "minimum": 0.9487236389374706, "maximum": 0.9671883536900371, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.947969513988042, "standard_deviation": 0.03642992606376134, "minimum": 1.8892522324788068, "maximum": 2.008022741657965, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6562376786971631, "standard_deviation": 0.2511384493115792, "minimum": 0.29060900946207835, "maximum": 0.8981685803117735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22384630004752512, "standard_deviation": 0.018020284462163194, "minimum": 0.20476725067652257, "maximum": 0.26282689657203107, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9834206419742297, "standard_deviation": 0.002856613960879649, "minimum": 0.977254793310963, "maximum": 0.9863660197003994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3604535815082925, "standard_deviation": 0.034725267897496595, "minimum": 1.3101475752107141, "maximum": 1.4139579076501634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4205347246120734, "standard_deviation": 0.19888485450631116, "minimum": 0.2020969217449667, "maximum": 0.6728032867230103, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14060965316640942, "standard_deviation": 0.012710922347666314, "minimum": 0.12650885116714694, "maximum": 0.16875861381256663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9440243145375632, "standard_deviation": 0.0028032057209678458, "minimum": 0.9405688584739695, "maximum": 0.9486338699177741, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.11060392014684, "standard_deviation": 0.02515712361842162, "minimum": 2.083313614967426, "maximum": 2.1457527694327547, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6150798802521104, "standard_deviation": 0.2940583933717541, "minimum": 0.2674496090944873, "maximum": 1.0349722280077762, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.31269991330668195, "standard_deviation": 0.007234851315613142, "minimum": 0.29978609475359086, "maximum": 0.3205933630181098, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9649276783999685, "standard_deviation": 0.002926643440464032, "minimum": 0.9608434697038841, "maximum": 0.9691940987722523, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6033300378735904, "standard_deviation": 0.03646571043807247, "minimum": 1.5597233475623824, "maximum": 1.6625014286088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.712963716971491, "standard_deviation": 0.26923033559493503, "minimum": 0.2749014677134183, "maximum": 1.1566287398227153, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2440060482970318, "standard_deviation": 0.00975725701143534, "minimum": 0.22908830469238753, "maximum": 0.2551320234379697, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9817370647878927, "standard_deviation": 0.0029510633117051837, "minimum": 0.9766117129566563, "maximum": 0.9865862503754077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.542882242150675, "standard_deviation": 0.033041079723342574, "minimum": 1.4979651122767725, "maximum": 1.5979579380283087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5021443227824682, "standard_deviation": 0.19057362957244198, "minimum": 0.20510499307397734, "maximum": 0.792566758769985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1467528021558363, "standard_deviation": 0.011163190448205058, "minimum": 0.12950282343864775, "maximum": 0.16670709348215948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9927851016451195, "standard_deviation": 0.0019987305534692763, "minimum": 0.9892292413300272, "maximum": 0.9959804188506312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0653939225498055, "standard_deviation": 0.04010933536993055, "minimum": 0.9886540303515101, "maximum": 1.1205525779259782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7256459989587691, "standard_deviation": 0.26701430264337755, "minimum": 0.29791393871460997, "maximum": 0.9449683046006484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08951502255906194, "standard_deviation": 0.009780566237679842, "minimum": 0.07611084097074479, "maximum": 0.10787439679888422, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.29912324008884056, "standard_deviation": 0.0586920314330845, "minimum": 0.20433319589838392, "maximum": 0.3643897399139766, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.907577945454523, "standard_deviation": 0.012364053819080125, "minimum": 5.894188767789875, "maximum": 5.927350816281218, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6844068622009549, "standard_deviation": 0.10423135791189479, "minimum": 0.5836944408924606, "maximum": 0.8498082991360283, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07051109184136023, "standard_deviation": 0.05886930402630261, "minimum": -0.1599953035966768, "maximum": 0.00961987483609462, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9906909497272677, "standard_deviation": 0.0011428598813923428, "minimum": 0.988883162705242, "maximum": 0.9927755304524798, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0169534066161836, "standard_deviation": 0.041742070828673844, "minimum": 0.9461770742745885, "maximum": 1.0699818605280556, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3029294924156536, "standard_deviation": 0.12258041280438794, "minimum": 0.1540641329906235, "maximum": 0.5178572894021097, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0863107943655331, "standard_deviation": 0.005516183772707155, "minimum": 0.07581382663734408, "maximum": 0.09451998405109972, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969867765167719, "standard_deviation": 0.00041312759452660324, "minimum": 0.9966062306194641, "maximum": 0.997826577362179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5669835470855609, "standard_deviation": 0.03537997660527957, "minimum": 0.5000324009977929, "maximum": 0.6133071263339487, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4195120032144398, "standard_deviation": 0.15758109193504222, "minimum": 0.20713902455356017, "maximum": 0.6673050359944318, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04762103481411338, "standard_deviation": 0.0034743926029319246, "minimum": 0.0399457923362365, "maximum": 0.0503206405734578, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2300-131720-0028", "language": "en", "style": "normal", "speaker": "librispeech_2300", "source_utterance_id": "2300-131720-0028", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2300-131720-0028.wav", "source_file_sha256": "fa099f84b852191aab695e6179235ef08c00fc3a19b148dc7b67e2fff1153c4f", "decoded_pcm_sha256": "9eaf55b481a213cff53140518cebb83838d1187d3e2187282fcf13a3699505c4", "original_source_sha256": "72f44e8aee815c88d0ad1b8f6d9a15fea55954dbd8e4d996b97112fe34bda0a1", "corpus_entry_id": "librispeech_2300-131720-0028"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.25172233581543, "predicted_mos": 4.25172233581543, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1ea34655dd23b722e2110567cb5f5b1feb53239efe3125fb4b9384ca828f89dc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.647483476997877, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2300-131720-0028", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1ea34655dd23b722e2110567cb5f5b1feb53239efe3125fb4b9384ca828f89dc", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "cf9bc8e7c16c92de5a6a34b62b32a7e149958c948fb1e3cc5e20a5dd2b9cedfe"}, "rows": [{"case": "librispeech_2300-131720-0028", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.2682998180389404, "predicted_mos": 3.2682998180389404, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "12fe49dd66f1e0df8a3db2bf36d7d115c065d1a015cf35ad6197fe43ee6901a7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.109720184002072, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2300-131720-0028", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "12fe49dd66f1e0df8a3db2bf36d7d115c065d1a015cf35ad6197fe43ee6901a7", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8653710242357356, "全帯域_10ms_rms_envelope_nrmse": 0.06602843583456812, "全帯域_10ms_envelope_correlation": 0.994967877416979, "全帯域_10ms_modulation_mae_db": 0.2726041244760672, "全帯域_40ms_log_rms_mae_db": 0.49609737410737037, "全帯域_40ms_rms_envelope_nrmse": 0.037054164756168705, "全帯域_40ms_envelope_correlation": 0.998618281190554, "全帯域_40ms_modulation_mae_db": 0.39055100563793316, "300–1000 Hz_10ms_log_rms_mae_db": 1.6982920095500391, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10003760239607817, "300–1000 Hz_10ms_envelope_correlation": 0.9928921741669249, "300–1000 Hz_10ms_modulation_mae_db": 0.6870793239582422, "300–1000 Hz_40ms_log_rms_mae_db": 1.1074734098681747, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06010171857292344, "300–1000 Hz_40ms_envelope_correlation": 0.9978901415811319, "300–1000 Hz_40ms_modulation_mae_db": 0.7772798386314399, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7526026807973691, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17810389087151166, "1000–3000 Hz_10ms_envelope_correlation": 0.9787354176349465, "1000–3000 Hz_10ms_modulation_mae_db": 0.9966820251912084, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0476140973045318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12511094583150384, "1000–3000 Hz_40ms_envelope_correlation": 0.9914481320728251, "1000–3000 Hz_40ms_modulation_mae_db": 1.2993183046885373, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9858249131250556, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17016767248424194, "3000–6000 Hz_10ms_envelope_correlation": 0.9831673701889492, "3000–6000 Hz_10ms_modulation_mae_db": 1.1312668590285884, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4240890496024585, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10432422816025363, "3000–6000 Hz_40ms_envelope_correlation": 0.994576450810199, "3000–6000 Hz_40ms_modulation_mae_db": 1.7513811980533536, "mr_stft_db": 5.7143274256989995, "band_mae_db": 0.28484955380572946, "rms_gain_error_db": -0.16084294096866153, "output_peak": 0.5171409095233951}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8419614748302352, "全帯域_10ms_rms_envelope_nrmse": 0.05784944284475447, "全帯域_10ms_envelope_correlation": 0.9961123266346915, "全帯域_10ms_modulation_mae_db": 0.18277017197848544, "全帯域_40ms_log_rms_mae_db": 0.4746771676187012, "全帯域_40ms_rms_envelope_nrmse": 0.03301188643359843, "全帯域_40ms_envelope_correlation": 0.9988555423545104, "全帯域_40ms_modulation_mae_db": 0.17347304671053318, "300–1000 Hz_10ms_log_rms_mae_db": 1.6314018238411991, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09165387145341014, "300–1000 Hz_10ms_envelope_correlation": 0.993952427190611, "300–1000 Hz_10ms_modulation_mae_db": 0.5644946472213761, "300–1000 Hz_40ms_log_rms_mae_db": 1.088847444552916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05550853493586776, "300–1000 Hz_40ms_envelope_correlation": 0.9980734062019034, "300–1000 Hz_40ms_modulation_mae_db": 0.7801723441774295, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7678965858936606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17400390764181026, "1000–3000 Hz_10ms_envelope_correlation": 0.9780940736158132, "1000–3000 Hz_10ms_modulation_mae_db": 0.8252709118527779, "1000–3000 Hz_40ms_log_rms_mae_db": 1.071755532399953, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11152405466506998, "1000–3000 Hz_40ms_envelope_correlation": 0.9919740244711289, "1000–3000 Hz_40ms_modulation_mae_db": 0.8449626136324291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9525428191024743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1499937017391481, "3000–6000 Hz_10ms_envelope_correlation": 0.9864815780700182, "3000–6000 Hz_10ms_modulation_mae_db": 1.2134363566671997, "3000–6000 Hz_40ms_log_rms_mae_db": 1.364385248627285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08548641450990699, "3000–6000 Hz_40ms_envelope_correlation": 0.9958335502333066, "3000–6000 Hz_40ms_modulation_mae_db": 1.7943545069086138, "mr_stft_db": 5.710524980276593, "band_mae_db": 0.21491468219068588, "rms_gain_error_db": -0.12984279821505967, "output_peak": 0.5030951663795199}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8481570911396928, "全帯域_10ms_rms_envelope_nrmse": 0.06310318902133256, "全帯域_10ms_envelope_correlation": 0.9951474620977797, "全帯域_10ms_modulation_mae_db": 0.19783791222508226, "全帯域_40ms_log_rms_mae_db": 0.4713968525319679, "全帯域_40ms_rms_envelope_nrmse": 0.03343299582392394, "全帯域_40ms_envelope_correlation": 0.9985838043995817, "全帯域_40ms_modulation_mae_db": 0.35072571435524996, "300–1000 Hz_10ms_log_rms_mae_db": 1.7242246475360419, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10011167658251319, "300–1000 Hz_10ms_envelope_correlation": 0.9924679097202287, "300–1000 Hz_10ms_modulation_mae_db": 0.828026120873195, "300–1000 Hz_40ms_log_rms_mae_db": 1.1394476720798499, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059769071988305944, "300–1000 Hz_40ms_envelope_correlation": 0.9974244336460902, "300–1000 Hz_40ms_modulation_mae_db": 1.2444706483977006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7532326909077665, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16274088771274337, "1000–3000 Hz_10ms_envelope_correlation": 0.980489550244131, "1000–3000 Hz_10ms_modulation_mae_db": 0.8543342152457036, "1000–3000 Hz_40ms_log_rms_mae_db": 1.059812475646686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09715484112919524, "1000–3000 Hz_40ms_envelope_correlation": 0.993640433328356, "1000–3000 Hz_40ms_modulation_mae_db": 0.8157151646666486, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9773614454519042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20842480212408548, "3000–6000 Hz_10ms_envelope_correlation": 0.9737617065938936, "3000–6000 Hz_10ms_modulation_mae_db": 1.5897563803250794, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4162865518397285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11525359071844632, "3000–6000 Hz_40ms_envelope_correlation": 0.9917525264618627, "3000–6000 Hz_40ms_modulation_mae_db": 1.239256393411382, "mr_stft_db": 5.700164554703851, "band_mae_db": 0.2547729833935305, "rms_gain_error_db": -0.06488611973914939, "output_peak": 0.4722054631846819}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8137793577716168, "全帯域_10ms_rms_envelope_nrmse": 0.06524301554534066, "全帯域_10ms_envelope_correlation": 0.9948922028626423, "全帯域_10ms_modulation_mae_db": 0.3662339201178411, "全帯域_40ms_log_rms_mae_db": 0.4272766020372869, "全帯域_40ms_rms_envelope_nrmse": 0.036402784920629865, "全帯域_40ms_envelope_correlation": 0.9984248789653369, "全帯域_40ms_modulation_mae_db": 0.312913652519546, "300–1000 Hz_10ms_log_rms_mae_db": 1.7089435308860643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10284884705376256, "300–1000 Hz_10ms_envelope_correlation": 0.9922521751838362, "300–1000 Hz_10ms_modulation_mae_db": 0.7226063897977536, "300–1000 Hz_40ms_log_rms_mae_db": 1.1402451721835498, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06261684805062567, "300–1000 Hz_40ms_envelope_correlation": 0.9974060138210901, "300–1000 Hz_40ms_modulation_mae_db": 0.9557029435032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7524312518808416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16768183834871206, "1000–3000 Hz_10ms_envelope_correlation": 0.9794465768767425, "1000–3000 Hz_10ms_modulation_mae_db": 0.529491785946436, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0570512639218672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10135932617379374, "1000–3000 Hz_40ms_envelope_correlation": 0.9933156168187095, "1000–3000 Hz_40ms_modulation_mae_db": 0.6819922837136753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9718365428193423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15983188496001532, "3000–6000 Hz_10ms_envelope_correlation": 0.984860056227588, "3000–6000 Hz_10ms_modulation_mae_db": 0.7052805255228006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3828459168055898, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09517498682055982, "3000–6000 Hz_40ms_envelope_correlation": 0.994545442290386, "3000–6000 Hz_40ms_modulation_mae_db": 0.27229227564341507, "mr_stft_db": 5.710898734555699, "band_mae_db": 0.20647195868691195, "rms_gain_error_db": -0.10601214863733784, "output_peak": 0.550808823856709}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8290694246712376, "全帯域_10ms_rms_envelope_nrmse": 0.060422578868233615, "全帯域_10ms_envelope_correlation": 0.9955685752318176, "全帯域_10ms_modulation_mae_db": 0.21497695019240873, "全帯域_40ms_log_rms_mae_db": 0.4468263455916706, "全帯域_40ms_rms_envelope_nrmse": 0.03327066929244937, "全帯域_40ms_envelope_correlation": 0.9986239428410715, "全帯域_40ms_modulation_mae_db": 0.32159096936308673, "300–1000 Hz_10ms_log_rms_mae_db": 1.6733441199185228, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09660081451079354, "300–1000 Hz_10ms_envelope_correlation": 0.9930026769263958, "300–1000 Hz_10ms_modulation_mae_db": 0.7954806668399789, "300–1000 Hz_40ms_log_rms_mae_db": 1.1081811430429218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05685134991649281, "300–1000 Hz_40ms_envelope_correlation": 0.997695479006008, "300–1000 Hz_40ms_modulation_mae_db": 1.4186178339446258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8009047386539478, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1744059072483451, "1000–3000 Hz_10ms_envelope_correlation": 0.9786062178964627, "1000–3000 Hz_10ms_modulation_mae_db": 1.2089852047870266, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0962965971985426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1100297935639532, "1000–3000 Hz_40ms_envelope_correlation": 0.9930505519694217, "1000–3000 Hz_40ms_modulation_mae_db": 1.6695173459238581, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9553175818765098, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15364696438821612, "3000–6000 Hz_10ms_envelope_correlation": 0.9856313637283408, "3000–6000 Hz_10ms_modulation_mae_db": 0.33797119275993787, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3422936753345784, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08995990610027842, "3000–6000 Hz_40ms_envelope_correlation": 0.9950421237730198, "3000–6000 Hz_40ms_modulation_mae_db": 0.40257210893834666, "mr_stft_db": 5.689604124732983, "band_mae_db": 0.1975559044288581, "rms_gain_error_db": -0.07460810168905306, "output_peak": 0.55846606315777}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7885219243747288, "全帯域_10ms_rms_envelope_nrmse": 0.05835648132150806, "全帯域_10ms_envelope_correlation": 0.9959695237764923, "全帯域_10ms_modulation_mae_db": 0.40840876876728865, "全帯域_40ms_log_rms_mae_db": 0.41964495787007555, "全帯域_40ms_rms_envelope_nrmse": 0.03506393644698395, "全帯域_40ms_envelope_correlation": 0.9985830275660941, "全帯域_40ms_modulation_mae_db": 0.5237202221035994, "300–1000 Hz_10ms_log_rms_mae_db": 1.669359630080976, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09693699186317344, "300–1000 Hz_10ms_envelope_correlation": 0.9930718977170613, "300–1000 Hz_10ms_modulation_mae_db": 1.0232089453479887, "300–1000 Hz_40ms_log_rms_mae_db": 1.1174211827496474, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060359121452159326, "300–1000 Hz_40ms_envelope_correlation": 0.9974900159689932, "300–1000 Hz_40ms_modulation_mae_db": 1.398783308755892, "1000–3000 Hz_10ms_log_rms_mae_db": 1.765137648954697, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1674802614037033, "1000–3000 Hz_10ms_envelope_correlation": 0.9797389863561135, "1000–3000 Hz_10ms_modulation_mae_db": 0.26809718392978604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0712616838173739, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10693429294879872, "1000–3000 Hz_40ms_envelope_correlation": 0.9926496104305952, "1000–3000 Hz_40ms_modulation_mae_db": 0.4611491554462075, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9857164402607674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14680729723291666, "3000–6000 Hz_10ms_envelope_correlation": 0.9869973599054738, "3000–6000 Hz_10ms_modulation_mae_db": 0.6706198429666245, "3000–6000 Hz_40ms_log_rms_mae_db": 1.405125046320097, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09547554787346027, "3000–6000 Hz_40ms_envelope_correlation": 0.9945673810336213, "3000–6000 Hz_40ms_modulation_mae_db": 0.8315084996132572, "mr_stft_db": 5.6958599816367474, "band_mae_db": 0.19142229174306366, "rms_gain_error_db": -0.11037951993130818, "output_peak": 0.5051649642345006}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7963751495784764, "全帯域_10ms_rms_envelope_nrmse": 0.058733597542238544, "全帯域_10ms_envelope_correlation": 0.9960332940347277, "全帯域_10ms_modulation_mae_db": 0.5103410798979109, "全帯域_40ms_log_rms_mae_db": 0.4241471854965123, "全帯域_40ms_rms_envelope_nrmse": 0.03541048409074925, "全帯域_40ms_envelope_correlation": 0.9986884459616715, "全帯域_40ms_modulation_mae_db": 0.3817630007893698, "300–1000 Hz_10ms_log_rms_mae_db": 1.7386688844155713, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09690131347030932, "300–1000 Hz_10ms_envelope_correlation": 0.99346709520159, "300–1000 Hz_10ms_modulation_mae_db": 0.9924617073437974, "300–1000 Hz_40ms_log_rms_mae_db": 1.1674027284467634, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06426655804641852, "300–1000 Hz_40ms_envelope_correlation": 0.9975329755060326, "300–1000 Hz_40ms_modulation_mae_db": 1.22506792541596, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8353443771293025, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1820726530069026, "1000–3000 Hz_10ms_envelope_correlation": 0.9767889280143922, "1000–3000 Hz_10ms_modulation_mae_db": 0.8480585529090271, "1000–3000 Hz_40ms_log_rms_mae_db": 1.093893212581736, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12096131412819636, "1000–3000 Hz_40ms_envelope_correlation": 0.99138291275526, "1000–3000 Hz_40ms_modulation_mae_db": 1.1870455319236197, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9873592649883973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15306269884672555, "3000–6000 Hz_10ms_envelope_correlation": 0.9858505340553937, "3000–6000 Hz_10ms_modulation_mae_db": 0.04699492853807221, "3000–6000 Hz_40ms_log_rms_mae_db": 1.408381551117062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09228805587444439, "3000–6000 Hz_40ms_envelope_correlation": 0.9949919269780413, "3000–6000 Hz_40ms_modulation_mae_db": 0.19911784333972457, "mr_stft_db": 5.701963753850923, "band_mae_db": 0.23254636892295424, "rms_gain_error_db": -0.13916661940012404, "output_peak": 0.5072153601974317}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8273069507668372, "全帯域_10ms_rms_envelope_nrmse": 0.0668576994045399, "全帯域_10ms_envelope_correlation": 0.9945489565272939, "全帯域_10ms_modulation_mae_db": 0.28895826527270035, "全帯域_40ms_log_rms_mae_db": 0.4249321011261522, "全帯域_40ms_rms_envelope_nrmse": 0.034747224440503544, "全帯域_40ms_envelope_correlation": 0.9984682367005907, "全帯域_40ms_modulation_mae_db": 0.22800292697988045, "300–1000 Hz_10ms_log_rms_mae_db": 1.6905669716476919, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11270262786169562, "300–1000 Hz_10ms_envelope_correlation": 0.990555695543916, "300–1000 Hz_10ms_modulation_mae_db": 0.8065068206475655, "300–1000 Hz_40ms_log_rms_mae_db": 1.1283203424979427, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06657734067119575, "300–1000 Hz_40ms_envelope_correlation": 0.9969759219629921, "300–1000 Hz_40ms_modulation_mae_db": 1.2332703768060134, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7707699406652155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1741941087684657, "1000–3000 Hz_10ms_envelope_correlation": 0.9773096642054042, "1000–3000 Hz_10ms_modulation_mae_db": 0.4817808105950591, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0458050782443353, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1106317062680911, "1000–3000 Hz_40ms_envelope_correlation": 0.9902268403468848, "1000–3000 Hz_40ms_modulation_mae_db": 0.36011192716164303, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0284634177753813, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18956687835919048, "3000–6000 Hz_10ms_envelope_correlation": 0.9793174873694731, "3000–6000 Hz_10ms_modulation_mae_db": 0.5162566949100865, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4175990207186946, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1074909978166276, "3000–6000 Hz_40ms_envelope_correlation": 0.9934749695235104, "3000–6000 Hz_40ms_modulation_mae_db": 0.33563810665188204, "mr_stft_db": 5.714053109496095, "band_mae_db": 0.1084305393568052, "rms_gain_error_db": -0.0631357688836496, "output_peak": 0.5555542462120583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.987405693743143, "spectral_convergence": 0.21186003566701483, "spectral_convergence_status": "ok", "amplitude_floor": 1.711947899928738e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4405322149611495, "spectral_convergence": 0.21277472508764741, "spectral_convergence_status": "ok", "amplitude_floor": 1.4743066880059505e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.715044368392706, "spectral_convergence": 0.21453523917277084, "spectral_convergence_status": "ok", "amplitude_floor": 1.2829193968481007e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015568083457567275, "output_power": 0.0015629357303633294, "delta_db": 0.01705968617933225, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001958578248681761, "output_power": 0.00184642681091738, "delta_db": -0.25608829401373706, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00020973665094298736, "output_power": 0.00021067740385813112, "delta_db": 0.019436290349504122, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018883580517861778, "output_power": 0.00016048383208584175, "delta_db": -0.7065305836935016, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010837478072549288, "output_power": 9.5819481788722e-05, "delta_db": -0.5347441400243413, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.95174204959434e-05, "output_power": 7.6372753812934e-05, "delta_db": -0.17523832857396038, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.28484955380572946, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16084294096866153, "pitch": null, "output_rms_dbfs": -24.031043868947933, "output_peak": 0.5171409095233951, "reference_rms_dbfs": -23.870200927979273, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7143274256989995}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023872023988588299, "log_rms_mae_db": 0.8653710242357356, "rms_envelope_nrmse": 0.06602843583456812, "envelope_correlation": 0.994967877416979, "rms_gain_error_db": -0.16003098143189287, "reference_mean_power": 0.004104010412713017, "output_mean_power": 0.003955536131167457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.972155761177528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037154075128907306, "output_power": 0.0003599776930513474, "reference_active": true, "delta_db": -0.1373086552504287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.499306084417706e-05, "output_power": 5.0179622765469695e-05, "reference_active": true, "delta_db": -0.39780500533567414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6530449727349195e-05, "output_power": 1.548869393160986e-05, "reference_active": true, "delta_db": -0.2826987128420987}], "active_band_count": 3, "active_band_mae_db": 0.2726041244760672}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00021497588341920653, "log_rms_mae_db": 0.49609737410737037, "rms_envelope_nrmse": 0.037054164756168705, "envelope_correlation": 0.998618281190554, "rms_gain_error_db": -0.15945097888998716, "reference_mean_power": 0.00411739985369924, "output_mean_power": 0.003968971193714897}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.973189563381595e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033993681960372955, "output_power": 0.0003289564956028968, "reference_active": true, "delta_db": -0.14259740469860127}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4133967175911375e-05, "output_power": 2.2044885332296465e-05, "reference_active": true, "delta_db": -0.3932087356156553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.827774931478282e-07, "output_power": 7.625469079868764e-07, "reference_active": true, "delta_db": -0.635846876599543}], "active_band_count": 3, "active_band_mae_db": 0.39055100563793316}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001937632671887944, "log_rms_mae_db": 1.6982920095500391, "rms_envelope_nrmse": 0.10003760239607817, "envelope_correlation": 0.9928921741669249, "rms_gain_error_db": -0.29568615376780955, "reference_mean_power": 0.0014622039835396937, "output_mean_power": 0.0013659642947079192}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.890315707731239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020584065582039213, "output_power": 0.00018996281395747223, "reference_active": true, "delta_db": -0.3486256264279133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2632804837837524e-05, "output_power": 3.472916212346048e-05, "reference_active": true, "delta_db": -0.8904960044309297}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.198246611681044e-06, "output_power": 6.7843671730371695e-06, "reference_active": true, "delta_db": -0.8221163410158838}], "active_band_count": 3, "active_band_mae_db": 0.6870793239582422}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001823230477151681, "log_rms_mae_db": 1.1074734098681747, "rms_envelope_nrmse": 0.06010171857292344, "envelope_correlation": 0.9978901415811319, "rms_gain_error_db": -0.29471031387429414, "reference_mean_power": 0.0014670926652510829, "output_mean_power": 0.0013708391987190586}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5556250379109994e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020148633268320183, "output_power": 0.00018394342453844487, "reference_active": true, "delta_db": -0.3956132457138936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9312508746169674e-05, "output_power": 1.6720601673966843e-05, "reference_active": true, "delta_db": -0.6258679241789693}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.933225179066708e-07, "output_power": 5.127426822799325e-07, "reference_active": true, "delta_db": -1.3103583460014565}], "active_band_count": 3, "active_band_mae_db": 0.7772798386314399}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.475794683802334e-05, "log_rms_mae_db": 1.7526026807973691, "rms_envelope_nrmse": 0.17810389087151166, "envelope_correlation": 0.9787354176349465, "rms_gain_error_db": -0.6759698098210808, "reference_mean_power": 0.000205358190626383, "output_mean_power": 0.00017575797823848925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.94981889493768e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.427037048631366e-05, "output_power": 4.850046151853892e-05, "reference_active": true, "delta_db": -0.48816915115856185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.925199890017768e-06, "output_power": 6.776906847690578e-06, "reference_active": true, "delta_db": -1.1958643534805793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.574401681634841e-06, "output_power": 1.9057900418940573e-06, "reference_active": true, "delta_db": -1.3060125709344839}], "active_band_count": 3, "active_band_mae_db": 0.9966820251912084}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.971385946015731e-05, "log_rms_mae_db": 1.0476140973045318, "rms_envelope_nrmse": 0.12511094583150384, "envelope_correlation": 0.9914481320728251, "rms_gain_error_db": -0.6761595060750174, "reference_mean_power": 0.00020637707638038015, "output_mean_power": 0.0001766222874968535}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.280870922430589e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.267995669092476e-05, "output_power": 4.6967692137630486e-05, "reference_active": true, "delta_db": -0.49846188570164235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.180545251002319e-06, "output_power": 3.094846064963772e-06, "reference_active": true, "delta_db": -1.3059387622733385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0193067246169046e-07, "output_power": 1.246943756236314e-07, "reference_active": true, "delta_db": -2.093554266090631}], "active_band_count": 3, "active_band_mae_db": 1.2993183046885373}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.517166954195252e-05, "log_rms_mae_db": 1.9858249131250556, "rms_envelope_nrmse": 0.17016767248424194, "envelope_correlation": 0.9831673701889492, "rms_gain_error_db": -0.4467645892733273, "reference_mean_power": 0.00014718641611289376, "output_mean_power": 0.00013279791968020705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4339203545791116e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.0687614509069e-05, "output_power": 6.382926021807329e-05, "reference_active": true, "delta_db": -0.44323514816781706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.343067234178686e-06, "output_power": 5.2282101520704754e-06, "reference_active": true, "delta_db": -2.521364380382656}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3595191848035655e-06, "output_power": 2.1374866354805436e-06, "reference_active": true, "delta_db": -0.4292010485352916}], "active_band_count": 3, "active_band_mae_db": 1.1312668590285884}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.54521457203503e-05, "log_rms_mae_db": 1.4240890496024585, "rms_envelope_nrmse": 0.10432422816025363, "envelope_correlation": 0.994576450810199, "rms_gain_error_db": -0.44676611057214666, "reference_mean_power": 0.0001479201087162099, "output_mean_power": 0.00013345984197592217}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3826641859038576e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.726675521598142e-05, "output_power": 6.020822387478451e-05, "reference_active": true, "delta_db": -0.48144663048391817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.27086384707959e-06, "output_power": 2.225654548961085e-06, "reference_active": true, "delta_db": -2.830579695863695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6847652723937028e-07, "output_power": 1.0772777130417464e-07, "reference_active": true, "delta_db": -1.9421172678124476}], "active_band_count": 3, "active_band_mae_db": 1.7513811980533536}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.896621133188683, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8ea0da36818d77f8c2d6c7d64b8faf3b67e5b11c196d79df36c67962c6f4d250", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9786511768555007, "standard_deviation": 0.0012411236954934556, "minimum": 0.9767889280143922, "maximum": 0.980489550244131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.77478998936035, "standard_deviation": 0.02918553557215045, "minimum": 1.7524312518808416, "maximum": 1.8353443771293025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7515875863071282, "standard_deviation": 0.3048329343218302, "minimum": 0.26809718392978604, "maximum": 1.2089852047870266, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17258543187527425, "standard_deviation": 0.006272363927921408, "minimum": 0.16274088771274337, "maximum": 0.1820726530069026, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9922110152741477, "standard_deviation": 0.0011601371509597142, "minimum": 0.9902268403468848, "maximum": 0.993640433328356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0679362426393784, "standard_deviation": 0.01923900767019508, "minimum": 1.0458050782443353, "maximum": 1.0962965971985426, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9149765408945774, "standard_deviation": 0.44303215362212683, "minimum": 0.36011192716164303, "maximum": 1.6695173459238581, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11046328433857527, "standard_deviation": 0.00924704296913569, "minimum": 0.09715484112919524, "maximum": 0.12511094583150384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9832584320173913, "standard_deviation": 0.004551457866342565, "minimum": 0.9737617065938936, "maximum": 0.9869973599054738, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9805528031749788, "standard_deviation": 0.023615209280442803, "minimum": 1.9525428191024743, "maximum": 2.0284634177753813, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7764478475897987, "standard_deviation": 0.5052090635898713, "minimum": 0.04699492853807221, "maximum": 1.5897563803250794, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16643773751681745, "standard_deviation": 0.021888276341312164, "minimum": 0.14680729723291666, "maximum": 0.20842480212408548, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943480463879933, "standard_deviation": 0.0012387364306570947, "minimum": 0.9917525264618627, "maximum": 0.9958335502333066, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3951257575456868, "standard_deviation": 0.02916554519754587, "minimum": 1.3422936753345784, "maximum": 1.4240890496024585, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8532651165699969, "standard_deviation": 0.6631337247278162, "minimum": 0.19911784333972457, "maximum": 1.7943545069086138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09818171598424719, "standard_deviation": 0.009969309503938734, "minimum": 0.08548641450990699, "maximum": 0.11525359071844632, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927077564563205, "standard_deviation": 0.0010195120908444684, "minimum": 0.990555695543916, "maximum": 0.993952427190611, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6918502022345134, "standard_deviation": 0.03397000047359604, "minimum": 1.6314018238411991, "maximum": 1.7386688844155713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8024830777537372, "standard_deviation": 0.15206495998167588, "minimum": 0.5644946472213761, "maximum": 1.0232089453479887, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.099724218148967, "standard_deviation": 0.006199934624708815, "minimum": 0.09165387145341014, "maximum": 0.11270262786169562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975610484617802, "standard_deviation": 0.00033376432648331733, "minimum": 0.9969759219629921, "maximum": 0.9980734062019034, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1246673869277206, "standard_deviation": 0.02450592509426513, "minimum": 1.088847444552916, "maximum": 1.1674027284467634, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1291706524540328, "standard_deviation": 0.25788031388572596, "minimum": 0.7772798386314399, "maximum": 1.4186178339446258, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060756317954248656, "standard_deviation": 0.003664824790352793, "minimum": 0.05550853493586776, "maximum": 0.06657734067119575, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21137053531606737, "standard_deviation": 0.05203478346619795, "minimum": 0.1084305393568052, "maximum": 0.28484955380572946, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.704674583118987, "standard_deviation": 0.0091542823840645, "minimum": 5.689604124732983, "maximum": 5.7143274256989995, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5212063745932584, "standard_deviation": 0.030806015639922968, "minimum": 0.4722054631846819, "maximum": 0.55846606315777, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10610925218304292, "standard_deviation": 0.0362697856077307, "minimum": -0.16084294096866153, "maximum": -0.0631357688836496, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.995405027322803, "standard_deviation": 0.0005968586290915559, "minimum": 0.9945489565272939, "maximum": 0.9961123266346915, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8263177996710701, "standard_deviation": 0.026015617404378435, "minimum": 0.7885219243747288, "maximum": 0.8653710242357356, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3052663991159731, "standard_deviation": 0.1149174742349759, "minimum": 0.18277017197848544, "maximum": 0.5103410798979109, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06207430504781449, "standard_deviation": 0.003687343995802335, "minimum": 0.05784944284475447, "maximum": 0.0668576994045399, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986057699974263, "standard_deviation": 0.00013220385045571497, "minimum": 0.9984248789653369, "maximum": 0.9988555423545104, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4481248232974671, "standard_deviation": 0.029052284271910714, "minimum": 0.41964495787007555, "maximum": 0.49609737410737037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33534256730739986, "standard_deviation": 0.10646348340805856, "minimum": 0.17347304671053318, "maximum": 0.5237202221035994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03479926827562588, "standard_deviation": 0.001488057960468118, "minimum": 0.03301188643359843, "maximum": 0.037054164756168705, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0028", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.3172154426574707, "predicted_mos": 3.3172154426574707, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "772475ec542aef58b200a457e988276ffd8c8c62741a1913b784be16d3d5f700", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.06386290024966, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.866645652516317, "全帯域_10ms_rms_envelope_nrmse": 0.06704736959080712, "全帯域_10ms_envelope_correlation": 0.9948148620421612, "全帯域_10ms_modulation_mae_db": 0.26922176473661974, "全帯域_40ms_log_rms_mae_db": 0.48800255167808543, "全帯域_40ms_rms_envelope_nrmse": 0.03797307365276729, "全帯域_40ms_envelope_correlation": 0.9985472509557038, "全帯域_40ms_modulation_mae_db": 0.38660748327926703, "300–1000 Hz_10ms_log_rms_mae_db": 1.6734187951793116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10196428827796852, "300–1000 Hz_10ms_envelope_correlation": 0.9925871757203012, "300–1000 Hz_10ms_modulation_mae_db": 0.6206413229998781, "300–1000 Hz_40ms_log_rms_mae_db": 1.0692777202964683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05952269486155204, "300–1000 Hz_40ms_envelope_correlation": 0.9979562841711919, "300–1000 Hz_40ms_modulation_mae_db": 0.6881901336751252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7391088211353987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18189723662112348, "1000–3000 Hz_10ms_envelope_correlation": 0.978168000015691, "1000–3000 Hz_10ms_modulation_mae_db": 1.0226188350520318, "1000–3000 Hz_40ms_log_rms_mae_db": 1.065319776720188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12814776341293202, "1000–3000 Hz_40ms_envelope_correlation": 0.991461414391764, "1000–3000 Hz_40ms_modulation_mae_db": 1.3691148754276197, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9148593938213436, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16946601374664358, "3000–6000 Hz_10ms_envelope_correlation": 0.9834892368290558, "3000–6000 Hz_10ms_modulation_mae_db": 1.1395213800827715, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4215799008834575, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10521633340828976, "3000–6000 Hz_40ms_envelope_correlation": 0.9946642708232029, "3000–6000 Hz_40ms_modulation_mae_db": 1.7568659142885714, "mr_stft_db": 5.641526232745943, "band_mae_db": 0.3049949516382974, "rms_gain_error_db": -0.16435093332756587, "output_peak": 0.5249090362014645}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8559541036032979, "全帯域_10ms_rms_envelope_nrmse": 0.05816046101548565, "全帯域_10ms_envelope_correlation": 0.9960349332296796, "全帯域_10ms_modulation_mae_db": 0.1885943686881554, "全帯域_40ms_log_rms_mae_db": 0.4828991305708418, "全帯域_40ms_rms_envelope_nrmse": 0.03198931275125704, "全帯域_40ms_envelope_correlation": 0.9989057619340612, "全帯域_40ms_modulation_mae_db": 0.19971983262453397, "300–1000 Hz_10ms_log_rms_mae_db": 1.6165492570310938, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09159082635633435, "300–1000 Hz_10ms_envelope_correlation": 0.9939008998115129, "300–1000 Hz_10ms_modulation_mae_db": 0.5205379505648741, "300–1000 Hz_40ms_log_rms_mae_db": 1.058340056204881, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05432673566864837, "300–1000 Hz_40ms_envelope_correlation": 0.9981106034008321, "300–1000 Hz_40ms_modulation_mae_db": 0.7191317277154524, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7572769070537024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17408478515282338, "1000–3000 Hz_10ms_envelope_correlation": 0.9780732746994362, "1000–3000 Hz_10ms_modulation_mae_db": 0.9383739248735168, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0679892164160893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10947213726892543, "1000–3000 Hz_40ms_envelope_correlation": 0.992348259649089, "1000–3000 Hz_40ms_modulation_mae_db": 0.9502232331266581, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9218549403731364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1514891340454768, "3000–6000 Hz_10ms_envelope_correlation": 0.9861129366401292, "3000–6000 Hz_10ms_modulation_mae_db": 1.120592716032542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4048705192073816, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0850019105018367, "3000–6000 Hz_40ms_envelope_correlation": 0.9957302428563292, "3000–6000 Hz_40ms_modulation_mae_db": 1.727732642341093, "mr_stft_db": 5.639804406235801, "band_mae_db": 0.20444491328450207, "rms_gain_error_db": -0.12189932363129397, "output_peak": 0.5049577926731188}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8481069085595394, "全帯域_10ms_rms_envelope_nrmse": 0.06315333551970619, "全帯域_10ms_envelope_correlation": 0.9951399714929746, "全帯域_10ms_modulation_mae_db": 0.20065495712557924, "全帯域_40ms_log_rms_mae_db": 0.47596808182652384, "全帯域_40ms_rms_envelope_nrmse": 0.03276748113612581, "全帯域_40ms_envelope_correlation": 0.9986462038242193, "全帯域_40ms_modulation_mae_db": 0.35457266639307666, "300–1000 Hz_10ms_log_rms_mae_db": 1.7131659221797328, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10414241176301609, "300–1000 Hz_10ms_envelope_correlation": 0.9918250062776243, "300–1000 Hz_10ms_modulation_mae_db": 0.7999251897521577, "300–1000 Hz_40ms_log_rms_mae_db": 1.1205478721287743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06170666464831186, "300–1000 Hz_40ms_envelope_correlation": 0.9972377953378797, "300–1000 Hz_40ms_modulation_mae_db": 1.2497585724105569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7166553008539944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16283060182353054, "1000–3000 Hz_10ms_envelope_correlation": 0.9806015544493957, "1000–3000 Hz_10ms_modulation_mae_db": 0.8414751943805854, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0293147190709064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.096140783026247, "1000–3000 Hz_40ms_envelope_correlation": 0.9939893794433576, "1000–3000 Hz_40ms_modulation_mae_db": 0.8218836306027004, "3000–6000 Hz_10ms_log_rms_mae_db": 1.922136543072342, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2082161797363136, "3000–6000 Hz_10ms_envelope_correlation": 0.9737786510754896, "3000–6000 Hz_10ms_modulation_mae_db": 1.5787465157567926, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4307891649548163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11476915062491194, "3000–6000 Hz_40ms_envelope_correlation": 0.9918323768925977, "3000–6000 Hz_40ms_modulation_mae_db": 1.1632820926278797, "mr_stft_db": 5.631540061125473, "band_mae_db": 0.2594770392027146, "rms_gain_error_db": -0.06691375449892163, "output_peak": 0.47628180580518975}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8166014202657393, "全帯域_10ms_rms_envelope_nrmse": 0.06561524472966652, "全帯域_10ms_envelope_correlation": 0.9948143526440493, "全帯域_10ms_modulation_mae_db": 0.41204406202225813, "全帯域_40ms_log_rms_mae_db": 0.4278918273333664, "全帯域_40ms_rms_envelope_nrmse": 0.0362395862282923, "全帯域_40ms_envelope_correlation": 0.9984151705740961, "全帯域_40ms_modulation_mae_db": 0.3982168153696699, "300–1000 Hz_10ms_log_rms_mae_db": 1.675033356736747, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10304123642726408, "300–1000 Hz_10ms_envelope_correlation": 0.9921248145345223, "300–1000 Hz_10ms_modulation_mae_db": 0.6561181157634756, "300–1000 Hz_40ms_log_rms_mae_db": 1.0990483139691383, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06295787010818184, "300–1000 Hz_40ms_envelope_correlation": 0.9972561673842624, "300–1000 Hz_40ms_modulation_mae_db": 0.8736327144953222, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7264319236382595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16595981816290217, "1000–3000 Hz_10ms_envelope_correlation": 0.980290125812232, "1000–3000 Hz_10ms_modulation_mae_db": 0.8905485957971359, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0560903943810571, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10214807727239965, "1000–3000 Hz_40ms_envelope_correlation": 0.993654825908262, "1000–3000 Hz_40ms_modulation_mae_db": 1.0407255368511559, "3000–6000 Hz_10ms_log_rms_mae_db": 1.937435618378075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15597994791001205, "3000–6000 Hz_10ms_envelope_correlation": 0.9855633043656485, "3000–6000 Hz_10ms_modulation_mae_db": 0.6473282811263298, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4191186365266304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09173010422234738, "3000–6000 Hz_40ms_envelope_correlation": 0.9949306775097371, "3000–6000 Hz_40ms_modulation_mae_db": 0.2627342261631645, "mr_stft_db": 5.648381088041757, "band_mae_db": 0.21224811792887424, "rms_gain_error_db": -0.0990223147866974, "output_peak": 0.5572239116328754}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8510508301198393, "全帯域_10ms_rms_envelope_nrmse": 0.06068651320216233, "全帯域_10ms_envelope_correlation": 0.9955311587948094, "全帯域_10ms_modulation_mae_db": 0.21398528719760326, "全帯域_40ms_log_rms_mae_db": 0.4473894128027936, "全帯域_40ms_rms_envelope_nrmse": 0.03315715977523819, "全帯域_40ms_envelope_correlation": 0.9986357123220824, "全帯域_40ms_modulation_mae_db": 0.3050348053600012, "300–1000 Hz_10ms_log_rms_mae_db": 1.6531612848391786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09723453707381984, "300–1000 Hz_10ms_envelope_correlation": 0.9928689764670923, "300–1000 Hz_10ms_modulation_mae_db": 0.8181133424027189, "300–1000 Hz_40ms_log_rms_mae_db": 1.0578642748735556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05878501164617633, "300–1000 Hz_40ms_envelope_correlation": 0.997465120920496, "300–1000 Hz_40ms_modulation_mae_db": 1.4706413519130699, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7600450995111228, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17947547116684043, "1000–3000 Hz_10ms_envelope_correlation": 0.9777566717115056, "1000–3000 Hz_10ms_modulation_mae_db": 1.1896102468450913, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0810634085033568, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11577001865824556, "1000–3000 Hz_40ms_envelope_correlation": 0.992706975878901, "1000–3000 Hz_40ms_modulation_mae_db": 1.5226798804812118, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9068447264656616, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15550411973328743, "3000–6000 Hz_10ms_envelope_correlation": 0.9852686782734357, "3000–6000 Hz_10ms_modulation_mae_db": 0.3353707502107908, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3334028935709739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09451955606368358, "3000–6000 Hz_40ms_envelope_correlation": 0.9944869356608342, "3000–6000 Hz_40ms_modulation_mae_db": 0.4008762287096566, "mr_stft_db": 5.625630887511061, "band_mae_db": 0.2105310042871017, "rms_gain_error_db": -0.07559919860310307, "output_peak": 0.5582634801050076}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.806527556278323, "全帯域_10ms_rms_envelope_nrmse": 0.05961295277121616, "全帯域_10ms_envelope_correlation": 0.9957835124642781, "全帯域_10ms_modulation_mae_db": 0.43171433929010944, "全帯域_40ms_log_rms_mae_db": 0.43191422115780276, "全帯域_40ms_rms_envelope_nrmse": 0.033984076347458855, "全帯域_40ms_envelope_correlation": 0.998681390135171, "全帯域_40ms_modulation_mae_db": 0.48390188773124954, "300–1000 Hz_10ms_log_rms_mae_db": 1.6517922976544779, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09559114265090828, "300–1000 Hz_10ms_envelope_correlation": 0.9932460551481779, "300–1000 Hz_10ms_modulation_mae_db": 1.0858103546179894, "300–1000 Hz_40ms_log_rms_mae_db": 1.0935996087934765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057936593218792715, "300–1000 Hz_40ms_envelope_correlation": 0.9976946403854071, "300–1000 Hz_40ms_modulation_mae_db": 1.552305749913607, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7263883904859174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16265316598384844, "1000–3000 Hz_10ms_envelope_correlation": 0.9816884321582985, "1000–3000 Hz_10ms_modulation_mae_db": 0.6555676912226253, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0492484770491872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10407841565954352, "1000–3000 Hz_40ms_envelope_correlation": 0.9939839607127993, "1000–3000 Hz_40ms_modulation_mae_db": 1.0586264837217294, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9506482738671203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14807517915038923, "3000–6000 Hz_10ms_envelope_correlation": 0.9868383337357596, "3000–6000 Hz_10ms_modulation_mae_db": 0.60155249143335, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4315404639393823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09551815729575482, "3000–6000 Hz_40ms_envelope_correlation": 0.9946663005924299, "3000–6000 Hz_40ms_modulation_mae_db": 0.7108808366262961, "mr_stft_db": 5.62418381055856, "band_mae_db": 0.21090355993021112, "rms_gain_error_db": -0.11083811763319797, "output_peak": 0.5063604419292309}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8226399265134933, "全帯域_10ms_rms_envelope_nrmse": 0.05973518650537008, "全帯域_10ms_envelope_correlation": 0.9958569985403188, "全帯域_10ms_modulation_mae_db": 0.4436226458273363, "全帯域_40ms_log_rms_mae_db": 0.4230325076508876, "全帯域_40ms_rms_envelope_nrmse": 0.03530021038851508, "全帯域_40ms_envelope_correlation": 0.9986672932326989, "全帯域_40ms_modulation_mae_db": 0.3333412748683027, "300–1000 Hz_10ms_log_rms_mae_db": 1.6866509619596055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09757333291231957, "300–1000 Hz_10ms_envelope_correlation": 0.9932476052403635, "300–1000 Hz_10ms_modulation_mae_db": 0.9658201462811847, "300–1000 Hz_40ms_log_rms_mae_db": 1.106930665763503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06329251517630108, "300–1000 Hz_40ms_envelope_correlation": 0.9975082830370081, "300–1000 Hz_40ms_modulation_mae_db": 1.2128249739016088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.824546724063164, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18343312964781788, "1000–3000 Hz_10ms_envelope_correlation": 0.976610352506393, "1000–3000 Hz_10ms_modulation_mae_db": 0.8185354947221546, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1041927375255005, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12198745370642529, "1000–3000 Hz_40ms_envelope_correlation": 0.9914368388701995, "1000–3000 Hz_40ms_modulation_mae_db": 1.152265940746849, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9291004547588366, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15996787918392766, "3000–6000 Hz_10ms_envelope_correlation": 0.9844324667637643, "3000–6000 Hz_10ms_modulation_mae_db": 0.25281513523705235, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4386081618496156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09962589328555002, "3000–6000 Hz_40ms_envelope_correlation": 0.9939273781191548, "3000–6000 Hz_40ms_modulation_mae_db": 0.4294106635759746, "mr_stft_db": 5.627583682337341, "band_mae_db": 0.2226970897416872, "rms_gain_error_db": -0.13339216269882095, "output_peak": 0.5047276741536424}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8266808524462225, "全帯域_10ms_rms_envelope_nrmse": 0.06524983380491481, "全帯域_10ms_envelope_correlation": 0.9948178369221087, "全帯域_10ms_modulation_mae_db": 0.2464967246397417, "全帯域_40ms_log_rms_mae_db": 0.414708680139154, "全帯域_40ms_rms_envelope_nrmse": 0.03373416292789605, "全帯域_40ms_envelope_correlation": 0.9985670208302757, "全帯域_40ms_modulation_mae_db": 0.21560241624780763, "300–1000 Hz_10ms_log_rms_mae_db": 1.676513515647917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11531643911234075, "300–1000 Hz_10ms_envelope_correlation": 0.9900665745402758, "300–1000 Hz_10ms_modulation_mae_db": 0.7941153273511711, "300–1000 Hz_40ms_log_rms_mae_db": 1.1211771135710746, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06813271473351672, "300–1000 Hz_40ms_envelope_correlation": 0.9967804875772569, "300–1000 Hz_40ms_modulation_mae_db": 1.2300296844571594, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7156823559325671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17237041856366409, "1000–3000 Hz_10ms_envelope_correlation": 0.9777094299503, "1000–3000 Hz_10ms_modulation_mae_db": 0.28560792169846855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0084237637328353, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10824812888672047, "1000–3000 Hz_40ms_envelope_correlation": 0.9906508584886963, "1000–3000 Hz_40ms_modulation_mae_db": 0.1789017101164034, "3000–6000 Hz_10ms_log_rms_mae_db": 1.972170933704354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1880025027455641, "3000–6000 Hz_10ms_envelope_correlation": 0.9795889579626493, "3000–6000 Hz_10ms_modulation_mae_db": 0.569600963606233, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4274509429313216, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10581818744840424, "3000–6000 Hz_40ms_envelope_correlation": 0.9936338298793885, "3000–6000 Hz_40ms_modulation_mae_db": 0.317193952519403, "mr_stft_db": 5.64937865646829, "band_mae_db": 0.09742459078914116, "rms_gain_error_db": -0.06538948548050137, "output_peak": 0.558419849071228}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.903660190687437, "spectral_convergence": 0.21297523890547587, "spectral_convergence_status": "ok", "amplitude_floor": 1.711947899928738e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.373857237600428, "spectral_convergence": 0.2153476451423553, "spectral_convergence_status": "ok", "amplitude_floor": 1.4743066880059505e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.647061269949967, "spectral_convergence": 0.21606666476313272, "spectral_convergence_status": "ok", "amplitude_floor": 1.2829193968481007e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015568083457567275, "output_power": 0.001563191069089929, "delta_db": 0.017769140448713736, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001958578248681761, "output_power": 0.0018478416423616702, "delta_db": -0.2527617706335836, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00020973665094298736, "output_power": 0.0002093545710036875, "delta_db": -0.007918813211440065, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018883580517861778, "output_power": 0.00015865596619260298, "delta_db": -0.756279361867419, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010837478072549288, "output_power": 9.473955617018593e-05, "delta_db": -0.5839688593619048, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.95174204959434e-05, "output_power": 7.574171018389018e-05, "delta_db": -0.211271764306723, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3049949516382974, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16435093332756587, "pitch": null, "output_rms_dbfs": -24.034551861306838, "output_peak": 0.5249090362014645, "reference_rms_dbfs": -23.870200927979273, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.641526232745943}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023872023988588299, "log_rms_mae_db": 0.866645652516317, "rms_envelope_nrmse": 0.06704736959080712, "envelope_correlation": 0.9948148620421612, "rms_gain_error_db": -0.1637494650977595, "reference_mean_power": 0.004104010412713017, "output_mean_power": 0.003952150801152742}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.972155761177528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037154075128907306, "output_power": 0.00036114445783307363, "reference_active": true, "delta_db": -0.12325500434154629}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.499306084417706e-05, "output_power": 5.090667248582746e-05, "reference_active": true, "delta_db": -0.3353318232558393}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6530449727349195e-05, "output_power": 1.5253756961575165e-05, "reference_active": true, "delta_db": -0.34907846661247355}], "active_band_count": 3, "active_band_mae_db": 0.26922176473661974}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00021497588341920653, "log_rms_mae_db": 0.48800255167808543, "rms_envelope_nrmse": 0.03797307365276729, "envelope_correlation": 0.9985472509557038, "rms_gain_error_db": -0.16319288187312198, "reference_mean_power": 0.00411739985369924, "output_mean_power": 0.0039655529810642895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.973189563381595e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033993681960372955, "output_power": 0.00033045689051370116, "reference_active": true, "delta_db": -0.12283394921913321}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4133967175911375e-05, "output_power": 2.2404457301584778e-05, "reference_active": true, "delta_db": -0.3229428916936671}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.827774931478282e-07, "output_power": 7.489394219884792e-07, "reference_active": true, "delta_db": -0.7140456089250008}], "active_band_count": 3, "active_band_mae_db": 0.38660748327926703}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001937632671887944, "log_rms_mae_db": 1.6734187951793116, "rms_envelope_nrmse": 0.10196428827796852, "envelope_correlation": 0.9925871757203012, "rms_gain_error_db": -0.2971243433367648, "reference_mean_power": 0.0014622039835396937, "output_mean_power": 0.0013655120231645306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.890315707731239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020584065582039213, "output_power": 0.00018952255630028065, "reference_active": true, "delta_db": -0.3587025132805305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2632804837837524e-05, "output_power": 3.4230323648042925e-05, "reference_active": true, "delta_db": -0.9533290066605732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.198246611681044e-06, "output_power": 7.223234815587564e-06, "reference_active": true, "delta_db": -0.5498924490585304}], "active_band_count": 3, "active_band_mae_db": 0.6206413229998781}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001823230477151681, "log_rms_mae_db": 1.0692777202964683, "rms_envelope_nrmse": 0.05952269486155204, "envelope_correlation": 0.9979562841711919, "rms_gain_error_db": -0.29602072360311826, "reference_mean_power": 0.0014670926652510829, "output_mean_power": 0.0013704256337040606}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5556250379109994e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020148633268320183, "output_power": 0.00018354784790976993, "reference_active": true, "delta_db": -0.40496295419605743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9312508746169674e-05, "output_power": 1.6530827980800264e-05, "reference_active": true, "delta_db": -0.6754408679448477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.933225179066708e-07, "output_power": 5.527371407738872e-07, "reference_active": true, "delta_db": -0.9841665788844702}], "active_band_count": 3, "active_band_mae_db": 0.6881901336751252}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.475794683802334e-05, "log_rms_mae_db": 1.7391088211353987, "rms_envelope_nrmse": 0.18189723662112348, "envelope_correlation": 0.978168000015691, "rms_gain_error_db": -0.7328908092571862, "reference_mean_power": 0.000205358190626383, "output_mean_power": 0.000173469428728099}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.94981889493768e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.427037048631366e-05, "output_power": 4.8020006812599995e-05, "reference_active": true, "delta_db": -0.5314056904355059}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.925199890017768e-06, "output_power": 6.6603419731880685e-06, "reference_active": true, "delta_db": -1.271214229548129}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.574401681634841e-06, "output_power": 1.923767803015485e-06, "reference_active": true, "delta_db": -1.2652365851724603}], "active_band_count": 3, "active_band_mae_db": 1.0226188350520318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.971385946015731e-05, "log_rms_mae_db": 1.065319776720188, "rms_envelope_nrmse": 0.12814776341293202, "envelope_correlation": 0.991461414391764, "rms_gain_error_db": -0.7330630877932308, "reference_mean_power": 0.00020637707638038015, "output_mean_power": 0.0001743231829269111}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.280870922430589e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.267995669092476e-05, "output_power": 4.657824755540731e-05, "reference_active": true, "delta_db": -0.534622644951463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.180545251002319e-06, "output_power": 3.01288780476027e-06, "reference_active": true, "delta_db": -1.4224996915323198}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0193067246169046e-07, "output_power": 1.230778957551854e-07, "reference_active": true, "delta_db": -2.1502222897990757}], "active_band_count": 3, "active_band_mae_db": 1.3691148754276197}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.517166954195252e-05, "log_rms_mae_db": 1.9148593938213436, "rms_envelope_nrmse": 0.16946601374664358, "envelope_correlation": 0.9834892368290558, "rms_gain_error_db": -0.4814978818555349, "reference_mean_power": 0.00014718641611289376, "output_mean_power": 0.00013174008595713291}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4339203545791116e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.0687614509069e-05, "output_power": 6.357760521486659e-05, "reference_active": true, "delta_db": -0.46039160531754636}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.343067234178686e-06, "output_power": 5.335050405766399e-06, "reference_active": true, "delta_db": -2.4335094709892564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3595191848035655e-06, "output_power": 2.0910152802345065e-06, "reference_active": true, "delta_db": -0.5246630639415121}], "active_band_count": 3, "active_band_mae_db": 1.1395213800827715}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.54521457203503e-05, "log_rms_mae_db": 1.4215799008834575, "rms_envelope_nrmse": 0.10521633340828976, "envelope_correlation": 0.9946642708232029, "rms_gain_error_db": -0.48147529134442557, "reference_mean_power": 0.0001479201087162099, "output_mean_power": 0.00013239747061384715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3826641859038576e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.726675521598142e-05, "output_power": 6.009678101937944e-05, "reference_active": true, "delta_db": -0.4894926849329835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.27086384707959e-06, "output_power": 2.257095076379659e-06, "reference_active": true, "delta_db": -2.769658731685803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6847652723937028e-07, "output_power": 1.060217013901965e-07, "reference_active": true, "delta_db": -2.011446326246928}], "active_band_count": 3, "active_band_mae_db": 1.7568659142885714}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8939841865350497, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7b66ee3917ea701e4c79b5fea02a1dd2cf7cee90ccc7aa27196437be69298501", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9788622301629065, "standard_deviation": 0.0017638480537849985, "minimum": 0.976610352506393, "maximum": 0.9816884321582985, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7457669403342657, "standard_deviation": 0.036033458411617765, "minimum": 1.7156823559325671, "maximum": 1.824546724063164, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8302922380739511, "standard_deviation": 0.2696252378775702, "minimum": 0.28560792169846855, "maximum": 1.1896102468450913, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17283807839031878, "standard_deviation": 0.008378065895565834, "minimum": 0.16265316598384844, "maximum": 0.18343312964781788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9925290641678837, "standard_deviation": 0.0012784709464213583, "minimum": 0.9906508584886963, "maximum": 0.9939893794433576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.05770531167489, "standard_deviation": 0.02971109208465269, "minimum": 1.0084237637328353, "maximum": 1.1041927375255005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.011802661384291, "standard_deviation": 0.4044417221077011, "minimum": 0.1789017101164034, "maximum": 1.5226798804812118, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11074909723642987, "standard_deviation": 0.010650202837922455, "minimum": 0.096140783026247, "maximum": 0.12814776341293202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9831340707057414, "standard_deviation": 0.004390844510554629, "minimum": 0.9737786510754896, "maximum": 0.9868383337357596, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9318813605551086, "standard_deviation": 0.02113257784381003, "minimum": 1.9068447264656616, "maximum": 1.972170933704354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7806910291857329, "standard_deviation": 0.45553474094192437, "minimum": 0.25281513523705235, "maximum": 1.5787465157567926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1670876195314518, "standard_deviation": 0.020865835853879706, "minimum": 0.14807517915038923, "maximum": 0.2082161797363136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9942340015417093, "standard_deviation": 0.0011577138908335932, "minimum": 0.9918323768925977, "maximum": 0.9957302428563292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4134200854829473, "standard_deviation": 0.033882653263537996, "minimum": 1.3334028935709739, "maximum": 1.4386081618496156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8461220696065048, "standard_deviation": 0.6230765611703714, "minimum": 0.2627342261631645, "maximum": 1.7568659142885714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09902491160634731, "standard_deviation": 0.009374577102505237, "minimum": 0.0850019105018367, "maximum": 0.11476915062491194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9924833884674837, "standard_deviation": 0.0011794373951533208, "minimum": 0.9900665745402758, "maximum": 0.9939008998115129, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6682856739035081, "standard_deviation": 0.02845681807229449, "minimum": 1.6165492570310938, "maximum": 1.7131659221797328, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7826352187166812, "standard_deviation": 0.1844315171122786, "minimum": 0.5205379505648741, "maximum": 1.0858103546179894, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10080677682174644, "standard_deviation": 0.007204582290513939, "minimum": 0.09159082635633435, "maximum": 0.11531643911234075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975011727767917, "standard_deviation": 0.00042528890165258816, "minimum": 0.9967804875772569, "maximum": 0.9981106034008321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.090848203200109, "standard_deviation": 0.026036720420644765, "minimum": 1.0578642748735556, "maximum": 1.1211771135710746, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1245643635602378, "standard_deviation": 0.32843421125624045, "minimum": 0.6881901336751252, "maximum": 1.552305749913607, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06083260000768512, "standard_deviation": 0.004161819920461698, "minimum": 0.05432673566864837, "maximum": 0.06813271473351672, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21534015835031617, "standard_deviation": 0.05869452416783237, "minimum": 0.09742459078914116, "maximum": 0.3049949516382974, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.636003603128028, "standard_deviation": 0.010107696684004849, "minimum": 5.62418381055856, "maximum": 5.64937865646829, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5238929989464697, "standard_deviation": 0.031134462774385488, "minimum": 0.47628180580518975, "maximum": 0.558419849071228, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10467566133251278, "standard_deviation": 0.03498402121346845, "minimum": -0.16435093332756587, "maximum": -0.06538948548050137, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9953492032662974, "standard_deviation": 0.0005137436513949447, "minimum": 0.9948143526440493, "maximum": 0.9960349332296796, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8367759062878465, "standard_deviation": 0.02143388057563796, "minimum": 0.806527556278323, "maximum": 0.866645652516317, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30079176869092544, "standard_deviation": 0.10955368645596048, "minimum": 0.1885943686881554, "maximum": 0.4436226458273363, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06240761214241611, "standard_deviation": 0.0033037432746293445, "minimum": 0.05816046101548565, "maximum": 0.06704736959080712, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986332254760385, "standard_deviation": 0.00014002744773690808, "minimum": 0.9984151705740961, "maximum": 0.9989057619340612, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4489758016449319, "standard_deviation": 0.029247468519437575, "minimum": 0.414708680139154, "maximum": 0.48800255167808543, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33462464773423856, "standard_deviation": 0.09461999437664266, "minimum": 0.19971983262453397, "maximum": 0.48390188773124954, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.034393132900943826, "standard_deviation": 0.0019856070850854737, "minimum": 0.03198931275125704, "maximum": 0.03797307365276729, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0028", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.9920647144317627, "predicted_mos": 2.9920647144317627, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3729dd607aab4f460a0137167946843e2ba867aa66f3e28d0cd3d05fa247240d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.680826000403613, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8350445948536438, "全帯域_10ms_rms_envelope_nrmse": 0.05740433525690986, "全帯域_10ms_envelope_correlation": 0.996015471066278, "全帯域_10ms_modulation_mae_db": 0.1529086331877275, "全帯域_40ms_log_rms_mae_db": 0.44714923008569096, "全帯域_40ms_rms_envelope_nrmse": 0.03172886276865382, "全帯域_40ms_envelope_correlation": 0.998755819870021, "全帯域_40ms_modulation_mae_db": 0.22444509250589853, "300–1000 Hz_10ms_log_rms_mae_db": 1.896215386111906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08730233268198724, "300–1000 Hz_10ms_envelope_correlation": 0.9943196790063048, "300–1000 Hz_10ms_modulation_mae_db": 0.770962491469903, "300–1000 Hz_40ms_log_rms_mae_db": 1.2558363446142549, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050491071953506994, "300–1000 Hz_40ms_envelope_correlation": 0.9982228489246773, "300–1000 Hz_40ms_modulation_mae_db": 0.8078898086954364, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0498306116078635, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1838521641652527, "1000–3000 Hz_10ms_envelope_correlation": 0.9753713529190658, "1000–3000 Hz_10ms_modulation_mae_db": 0.7737049292698913, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2089356437288736, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10907328188877553, "1000–3000 Hz_40ms_envelope_correlation": 0.9926397674144897, "1000–3000 Hz_40ms_modulation_mae_db": 0.933214895941906, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3808465118087736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16686233560118618, "3000–6000 Hz_10ms_envelope_correlation": 0.9830768430352944, "3000–6000 Hz_10ms_modulation_mae_db": 1.2604229232538111, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6497140561027335, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09885329303037309, "3000–6000 Hz_40ms_envelope_correlation": 0.9941262826147714, "3000–6000 Hz_40ms_modulation_mae_db": 1.573916039099104, "mr_stft_db": 5.955660043182296, "band_mae_db": 0.29750451738668016, "rms_gain_error_db": -0.07435703880549803, "output_peak": 0.4987401306343883}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8006928134108786, "全帯域_10ms_rms_envelope_nrmse": 0.05626685062031117, "全帯域_10ms_envelope_correlation": 0.9962434035876454, "全帯域_10ms_modulation_mae_db": 0.5572168301203585, "全帯域_40ms_log_rms_mae_db": 0.40282522557432215, "全帯域_40ms_rms_envelope_nrmse": 0.02798832662712395, "全帯域_40ms_envelope_correlation": 0.9991395670938189, "全帯域_40ms_modulation_mae_db": 0.4176392959730783, "300–1000 Hz_10ms_log_rms_mae_db": 1.8906861411918217, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09070275134002762, "300–1000 Hz_10ms_envelope_correlation": 0.993803978012303, "300–1000 Hz_10ms_modulation_mae_db": 0.7124968420813929, "300–1000 Hz_40ms_log_rms_mae_db": 1.2412003766646542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05334765611755085, "300–1000 Hz_40ms_envelope_correlation": 0.9979305478952206, "300–1000 Hz_40ms_modulation_mae_db": 0.851676056476526, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0165533050520037, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19576370525509273, "1000–3000 Hz_10ms_envelope_correlation": 0.9723076435993548, "1000–3000 Hz_10ms_modulation_mae_db": 0.9842431949027165, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2056435714219007, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11975562639021846, "1000–3000 Hz_40ms_envelope_correlation": 0.9914358441819526, "1000–3000 Hz_40ms_modulation_mae_db": 1.1251116426941914, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3642295018205575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16223669171302435, "3000–6000 Hz_10ms_envelope_correlation": 0.98439922378289, "3000–6000 Hz_10ms_modulation_mae_db": 0.8033533104370406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6285720873729062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08836676083764448, "3000–6000 Hz_40ms_envelope_correlation": 0.9959918873140864, "3000–6000 Hz_40ms_modulation_mae_db": 0.8526346820808518, "mr_stft_db": 5.931146827457432, "band_mae_db": 0.24497146085243363, "rms_gain_error_db": -0.1043150763996635, "output_peak": 0.48418762097718}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.788925991847601, "全帯域_10ms_rms_envelope_nrmse": 0.06211876508864031, "全帯域_10ms_envelope_correlation": 0.9953539671438071, "全帯域_10ms_modulation_mae_db": 0.2039492922896361, "全帯域_40ms_log_rms_mae_db": 0.4071280682935555, "全帯域_40ms_rms_envelope_nrmse": 0.03199378677013625, "全帯域_40ms_envelope_correlation": 0.9987948804252453, "全帯域_40ms_modulation_mae_db": 0.22445622730518835, "300–1000 Hz_10ms_log_rms_mae_db": 1.9309424631119139, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10078967553168315, "300–1000 Hz_10ms_envelope_correlation": 0.9925810827157807, "300–1000 Hz_10ms_modulation_mae_db": 0.8483685230052579, "300–1000 Hz_40ms_log_rms_mae_db": 1.265648606238868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06029447052187843, "300–1000 Hz_40ms_envelope_correlation": 0.9976377761493044, "300–1000 Hz_40ms_modulation_mae_db": 1.0510576642536629, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0187625885599476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18345791320557525, "1000–3000 Hz_10ms_envelope_correlation": 0.9748540404714702, "1000–3000 Hz_10ms_modulation_mae_db": 1.0988284517022855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.203758159415862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10478378027343521, "1000–3000 Hz_40ms_envelope_correlation": 0.9925113100532934, "1000–3000 Hz_40ms_modulation_mae_db": 1.2853908558841758, "3000–6000 Hz_10ms_log_rms_mae_db": 2.394136789793377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18580623422601675, "3000–6000 Hz_10ms_envelope_correlation": 0.9789117949773202, "3000–6000 Hz_10ms_modulation_mae_db": 0.8663296562685178, "3000–6000 Hz_40ms_log_rms_mae_db": 1.634178048643922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09906543152231448, "3000–6000 Hz_40ms_envelope_correlation": 0.994080175489416, "3000–6000 Hz_40ms_modulation_mae_db": 0.6864975121534528, "mr_stft_db": 5.948060754863772, "band_mae_db": 0.22329350146215632, "rms_gain_error_db": -0.0941287191434533, "output_peak": 0.5046480138278644}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8275980798163434, "全帯域_10ms_rms_envelope_nrmse": 0.06100143747498549, "全帯域_10ms_envelope_correlation": 0.9954763529148946, "全帯域_10ms_modulation_mae_db": 0.3746946483207241, "全帯域_40ms_log_rms_mae_db": 0.44010186289729863, "全帯域_40ms_rms_envelope_nrmse": 0.035827937516885164, "全帯域_40ms_envelope_correlation": 0.9983716570700045, "全帯域_40ms_modulation_mae_db": 0.4302829397819705, "300–1000 Hz_10ms_log_rms_mae_db": 1.9429284106653175, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.103405989698479, "300–1000 Hz_10ms_envelope_correlation": 0.9918337209744711, "300–1000 Hz_10ms_modulation_mae_db": 0.586013092049296, "300–1000 Hz_40ms_log_rms_mae_db": 1.2827428266074725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06526379972713017, "300–1000 Hz_40ms_envelope_correlation": 0.9967202709372375, "300–1000 Hz_40ms_modulation_mae_db": 0.9229654291136544, "1000–3000 Hz_10ms_log_rms_mae_db": 2.055377728983501, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16790724905964372, "1000–3000 Hz_10ms_envelope_correlation": 0.978779871817763, "1000–3000 Hz_10ms_modulation_mae_db": 0.7282600916204993, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2449656427668954, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09436954718165506, "1000–3000 Hz_40ms_envelope_correlation": 0.9935881581926023, "1000–3000 Hz_40ms_modulation_mae_db": 0.8370301058578589, "3000–6000 Hz_10ms_log_rms_mae_db": 2.370775351804398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16862053945876368, "3000–6000 Hz_10ms_envelope_correlation": 0.9831075944780824, "3000–6000 Hz_10ms_modulation_mae_db": 1.1970209980611484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6488241192191284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09937792781984879, "3000–6000 Hz_40ms_envelope_correlation": 0.9946929269290133, "3000–6000 Hz_40ms_modulation_mae_db": 1.456286948350168, "mr_stft_db": 5.934241858470891, "band_mae_db": 0.1813165398850514, "rms_gain_error_db": -0.05906107295501623, "output_peak": 0.46875310542613147}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8251579307806363, "全帯域_10ms_rms_envelope_nrmse": 0.062392325857342223, "全帯域_10ms_envelope_correlation": 0.9954149019201111, "全帯域_10ms_modulation_mae_db": 0.22185542029385194, "全帯域_40ms_log_rms_mae_db": 0.447280676004464, "全帯域_40ms_rms_envelope_nrmse": 0.03470013451266049, "全帯域_40ms_envelope_correlation": 0.9986957497396186, "全帯域_40ms_modulation_mae_db": 0.27849603826193814, "300–1000 Hz_10ms_log_rms_mae_db": 1.8688884923668139, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10859815057208992, "300–1000 Hz_10ms_envelope_correlation": 0.9912485982840908, "300–1000 Hz_10ms_modulation_mae_db": 0.37087684038991897, "300–1000 Hz_40ms_log_rms_mae_db": 1.2140704221514722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0657161331570276, "300–1000 Hz_40ms_envelope_correlation": 0.9970429762307951, "300–1000 Hz_40ms_modulation_mae_db": 0.6707110131921024, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0077827380239563, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18262878328100823, "1000–3000 Hz_10ms_envelope_correlation": 0.9752239484065226, "1000–3000 Hz_10ms_modulation_mae_db": 1.170742182099115, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2160675439930968, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11506997087404348, "1000–3000 Hz_40ms_envelope_correlation": 0.9907325801896218, "1000–3000 Hz_40ms_modulation_mae_db": 1.3616425452244567, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3904736264958766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17278831031986994, "3000–6000 Hz_10ms_envelope_correlation": 0.9819547746533631, "3000–6000 Hz_10ms_modulation_mae_db": 0.49029069058564173, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6588554323077986, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09107974429608881, "3000–6000 Hz_40ms_envelope_correlation": 0.9948912850046878, "3000–6000 Hz_40ms_modulation_mae_db": 0.42023610440774134, "mr_stft_db": 5.95300267535326, "band_mae_db": 0.23225161828274046, "rms_gain_error_db": -0.1247873614758546, "output_peak": 0.48247039742719366}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7817012871204309, "全帯域_10ms_rms_envelope_nrmse": 0.05881515084562952, "全帯域_10ms_envelope_correlation": 0.9958656808849381, "全帯域_10ms_modulation_mae_db": 0.3097130367270409, "全帯域_40ms_log_rms_mae_db": 0.42619431147727027, "全帯域_40ms_rms_envelope_nrmse": 0.0334586122385995, "全帯域_40ms_envelope_correlation": 0.9986797438612723, "全帯域_40ms_modulation_mae_db": 0.49865873826975937, "300–1000 Hz_10ms_log_rms_mae_db": 1.7997398147349073, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0907333982786633, "300–1000 Hz_10ms_envelope_correlation": 0.994163836463466, "300–1000 Hz_10ms_modulation_mae_db": 1.0029844225273787, "300–1000 Hz_40ms_log_rms_mae_db": 1.1816690342716565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057817980657294994, "300–1000 Hz_40ms_envelope_correlation": 0.9979367565938556, "300–1000 Hz_40ms_modulation_mae_db": 1.2960300015467159, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0600948941438517, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19096558918722098, "1000–3000 Hz_10ms_envelope_correlation": 0.9722923756165291, "1000–3000 Hz_10ms_modulation_mae_db": 0.6769716811267852, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2466181556642186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.105513930508263, "1000–3000 Hz_40ms_envelope_correlation": 0.9918992587274761, "1000–3000 Hz_40ms_modulation_mae_db": 0.6723984264769188, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4314710837516924, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18003958184927876, "3000–6000 Hz_10ms_envelope_correlation": 0.9802914329573714, "3000–6000 Hz_10ms_modulation_mae_db": 0.4596096302491502, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6370690407018058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09921825912962214, "3000–6000 Hz_40ms_envelope_correlation": 0.994237241695581, "3000–6000 Hz_40ms_modulation_mae_db": 1.1990358680829263, "mr_stft_db": 5.936939536708643, "band_mae_db": 0.1893662775069931, "rms_gain_error_db": -0.09908852166502316, "output_peak": 0.4695617157014714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8317207083889996, "全帯域_10ms_rms_envelope_nrmse": 0.063064035719062, "全帯域_10ms_envelope_correlation": 0.9951232565089418, "全帯域_10ms_modulation_mae_db": 0.45253682881086577, "全帯域_40ms_log_rms_mae_db": 0.44593684318597254, "全帯域_40ms_rms_envelope_nrmse": 0.032486372785983875, "全帯域_40ms_envelope_correlation": 0.9986188270445675, "全帯域_40ms_modulation_mae_db": 0.48444742170806726, "300–1000 Hz_10ms_log_rms_mae_db": 1.9532357197830001, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10078931701860552, "300–1000 Hz_10ms_envelope_correlation": 0.9921287654684088, "300–1000 Hz_10ms_modulation_mae_db": 0.4483116905147863, "300–1000 Hz_40ms_log_rms_mae_db": 1.264114655920136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06277674124307563, "300–1000 Hz_40ms_envelope_correlation": 0.9967969356721085, "300–1000 Hz_40ms_modulation_mae_db": 0.10191062715140624, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0087836529412737, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1835000744904914, "1000–3000 Hz_10ms_envelope_correlation": 0.9745376045790135, "1000–3000 Hz_10ms_modulation_mae_db": 0.7711656248389223, "1000–3000 Hz_40ms_log_rms_mae_db": 1.155576635393281, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10548541096211185, "1000–3000 Hz_40ms_envelope_correlation": 0.9919385492473745, "1000–3000 Hz_40ms_modulation_mae_db": 0.9582944394006163, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3892045930612023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15888785458110516, "3000–6000 Hz_10ms_envelope_correlation": 0.984733490796903, "3000–6000 Hz_10ms_modulation_mae_db": 1.4497586209243207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.600088069807254, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0878825929069499, "3000–6000 Hz_40ms_envelope_correlation": 0.9952064551124034, "3000–6000 Hz_40ms_modulation_mae_db": 1.8536102213815386, "mr_stft_db": 5.9477316232850965, "band_mae_db": 0.12141401457205837, "rms_gain_error_db": -0.029185980506413245, "output_peak": 0.5506368635440209}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8033771998012441, "全帯域_10ms_rms_envelope_nrmse": 0.06162368961054639, "全帯域_10ms_envelope_correlation": 0.9954115627296095, "全帯域_10ms_modulation_mae_db": 0.4361585056176316, "全帯域_40ms_log_rms_mae_db": 0.426183779294568, "全帯域_40ms_rms_envelope_nrmse": 0.0333704708865304, "全帯域_40ms_envelope_correlation": 0.9986440168376675, "全帯域_40ms_modulation_mae_db": 0.46089089902416047, "300–1000 Hz_10ms_log_rms_mae_db": 1.9224054287996546, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10275573347862144, "300–1000 Hz_10ms_envelope_correlation": 0.9921077951446851, "300–1000 Hz_10ms_modulation_mae_db": 0.9575520674617577, "300–1000 Hz_40ms_log_rms_mae_db": 1.247324916490043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06423460002567531, "300–1000 Hz_40ms_envelope_correlation": 0.9970434128362169, "300–1000 Hz_40ms_modulation_mae_db": 1.1576304149458625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.96534266918227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.182093772116477, "1000–3000 Hz_10ms_envelope_correlation": 0.9754548542547479, "1000–3000 Hz_10ms_modulation_mae_db": 0.3621923853371721, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2204686899927706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11836418752620104, "1000–3000 Hz_40ms_envelope_correlation": 0.9901841782411244, "1000–3000 Hz_40ms_modulation_mae_db": 0.5040685765937417, "3000–6000 Hz_10ms_log_rms_mae_db": 2.333792524667696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.160301237030639, "3000–6000 Hz_10ms_envelope_correlation": 0.9843390204429228, "3000–6000 Hz_10ms_modulation_mae_db": 0.060047795556982496, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6105558558406012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09413980447561533, "3000–6000 Hz_40ms_envelope_correlation": 0.994511315582392, "3000–6000 Hz_40ms_modulation_mae_db": 0.3728290488988457, "mr_stft_db": 5.935358017039117, "band_mae_db": 0.1622165513324032, "rms_gain_error_db": -0.0852663308534733, "output_peak": 0.5453114418122875}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2891377027222255, "spectral_convergence": 0.21435543480500122, "spectral_convergence_status": "ok", "amplitude_floor": 1.711947899928738e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.737116120558231, "spectral_convergence": 0.2155794892907944, "spectral_convergence_status": "ok", "amplitude_floor": 1.4743066880059505e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.840726306266431, "spectral_convergence": 0.21425272074680834, "spectral_convergence_status": "ok", "amplitude_floor": 1.2829193968481007e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015568083457567275, "output_power": 0.0015853061253130337, "delta_db": 0.07877986554492275, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001958578248681761, "output_power": 0.0018817398285153112, "delta_db": -0.17381349873030208, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00020973665094298736, "output_power": 0.00021749126374329751, "delta_db": 0.157674878189147, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018883580517861778, "output_power": 0.00016615444908694554, "delta_db": -0.5557236960287283, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010837478072549288, "output_power": 9.624996218366877e-05, "delta_db": -0.5152766423035038, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.95174204959434e-05, "output_power": 8.527822174180342e-05, "delta_db": 0.303758523523477, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29750451738668016, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07435703880549803, "pitch": null, "output_rms_dbfs": -23.94455796678477, "output_peak": 0.4987401306343883, "reference_rms_dbfs": -23.870200927979273, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.955660043182296}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023872023988588299, "log_rms_mae_db": 0.8350445948536438, "rms_envelope_nrmse": 0.05740433525690986, "envelope_correlation": 0.996015471066278, "rms_gain_error_db": -0.07332972847780153, "reference_mean_power": 0.004104010412713017, "output_mean_power": 0.004035296782692453}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.972155761177528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037154075128907306, "output_power": 0.0003572436067677345, "reference_active": true, "delta_db": -0.1704198954780126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.499306084417706e-05, "output_power": 5.198599130419048e-05, "reference_active": true, "delta_db": -0.24421562826698184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6530449727349195e-05, "output_power": 1.6699124658818838e-05, "reference_active": true, "delta_db": 0.044090375818187956}], "active_band_count": 3, "active_band_mae_db": 0.1529086331877275}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00021497588341920653, "log_rms_mae_db": 0.44714923008569096, "rms_envelope_nrmse": 0.03172886276865382, "envelope_correlation": 0.998755819870021, "rms_gain_error_db": -0.07266346244808776, "reference_mean_power": 0.00411739985369924, "output_mean_power": 0.004049083179732257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.973189563381595e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033993681960372955, "output_power": 0.00032606311982710443, "reference_active": true, "delta_db": -0.18096527267562756}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4133967175911375e-05, "output_power": 2.27908430583015e-05, "reference_active": true, "delta_db": -0.24868327215642927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.827774931478282e-07, "output_power": 9.337270174892053e-07, "reference_active": true, "delta_db": 0.24368673268563878}], "active_band_count": 3, "active_band_mae_db": 0.22444509250589853}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001937632671887944, "log_rms_mae_db": 1.896215386111906, "rms_envelope_nrmse": 0.08730233268198724, "envelope_correlation": 0.9943196790063048, "rms_gain_error_db": -0.18065038518732132, "reference_mean_power": 0.0014622039835396937, "output_mean_power": 0.001402629356306902}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.890315707731239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020584065582039213, "output_power": 0.0001861979910897951, "reference_active": true, "delta_db": -0.43556165871561997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2632804837837524e-05, "output_power": 3.59518052074977e-05, "reference_active": true, "delta_db": -0.7402320407588013}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.198246611681044e-06, "output_power": 6.309741985959257e-06, "reference_active": true, "delta_db": -1.1370937749352878}], "active_band_count": 3, "active_band_mae_db": 0.770962491469903}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001823230477151681, "log_rms_mae_db": 1.2558363446142549, "rms_envelope_nrmse": 0.050491071953506994, "envelope_correlation": 0.9982228489246773, "rms_gain_error_db": -0.1795797834865009, "reference_mean_power": 0.0014670926652510829, "output_mean_power": 0.001407665826477568}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5556250379109994e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020148633268320183, "output_power": 0.0001810102716529579, "reference_active": true, "delta_db": -0.46542372057378734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9312508746169674e-05, "output_power": 1.6764609588381672e-05, "reference_active": true, "delta_db": -0.6144524935014268}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.933225179066708e-07, "output_power": 5.088104055254752e-07, "reference_active": true, "delta_db": -1.3437932120110951}], "active_band_count": 3, "active_band_mae_db": 0.8078898086954364}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.475794683802334e-05, "log_rms_mae_db": 2.0498306116078635, "rms_envelope_nrmse": 0.1838521641652527, "envelope_correlation": 0.9753713529190658, "rms_gain_error_db": -0.5083033157294082, "reference_mean_power": 0.000205358190626383, "output_mean_power": 0.00018267608588823034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.94981889493768e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.427037048631366e-05, "output_power": 4.8566428956714816e-05, "reference_active": true, "delta_db": -0.4822661499074584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.925199890017768e-06, "output_power": 7.219398455930519e-06, "reference_active": true, "delta_db": -0.9211693918296182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.574401681634841e-06, "output_power": 2.084051207350127e-06, "reference_active": true, "delta_db": -0.9176792460725974}], "active_band_count": 3, "active_band_mae_db": 0.7737049292698913}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.971385946015731e-05, "log_rms_mae_db": 1.2089356437288736, "rms_envelope_nrmse": 0.10907328188877553, "envelope_correlation": 0.9926397674144897, "rms_gain_error_db": -0.508351663579775, "reference_mean_power": 0.00020637707638038015, "output_mean_power": 0.0001835803905382448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.280870922430589e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.267995669092476e-05, "output_power": 4.742223501821122e-05, "reference_active": true, "delta_db": -0.45663390749884525}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.180545251002319e-06, "output_power": 3.3446637407047887e-06, "reference_active": true, "delta_db": -0.9688046663405723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0193067246169046e-07, "output_power": 1.4715725568941457e-07, "reference_active": true, "delta_db": -1.3742061139863002}], "active_band_count": 3, "active_band_mae_db": 0.933214895941906}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.517166954195252e-05, "log_rms_mae_db": 2.3808465118087736, "rms_envelope_nrmse": 0.16686233560118618, "envelope_correlation": 0.9830768430352944, "rms_gain_error_db": -0.19078098352792733, "reference_mean_power": 0.00014718641611289376, "output_mean_power": 0.00014086063190046357}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4339203545791116e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.0687614509069e-05, "output_power": 6.426772516499596e-05, "reference_active": true, "delta_db": -0.41350398000976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.343067234178686e-06, "output_power": 4.5813454210448275e-06, "reference_active": true, "delta_db": -3.0949643647112834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3595191848035655e-06, "output_power": 2.2158658805189285e-06, "reference_active": true, "delta_db": -0.27280042504038965}], "active_band_count": 3, "active_band_mae_db": 1.2604229232538111}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.54521457203503e-05, "log_rms_mae_db": 1.6497140561027335, "rms_envelope_nrmse": 0.09885329303037309, "envelope_correlation": 0.9941262826147714, "rms_gain_error_db": -0.19075323215267945, "reference_mean_power": 0.0001479201087162099, "output_mean_power": 0.00014156369641882459}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3826641859038576e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.726675521598142e-05, "output_power": 6.043058892109169e-05, "reference_active": true, "delta_db": -0.4654365223422618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.27086384707959e-06, "output_power": 2.057210536368554e-06, "reference_active": true, "delta_db": -3.1723698659412363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6847652723937028e-07, "output_power": 1.312638774954667e-07, "reference_active": true, "delta_db": -1.0839417290138142}], "active_band_count": 3, "active_band_mae_db": 1.573916039099104}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8796969694545811, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "33c482df096fbd8b5230b6b277b07c74dfcf6aa4031d898a1bfb218304277278", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9748527114580584, "standard_deviation": 0.002046737976591211, "minimum": 0.9722923756165291, "maximum": 0.978779871817763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0228160235618335, "standard_deviation": 0.03153572205909151, "minimum": 1.96534266918227, "maximum": 2.0600948941438517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8207635676121734, "standard_deviation": 0.25931935611021945, "minimum": 0.3621923853371721, "maximum": 1.170742182099115, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18377115634509528, "standard_deviation": 0.0080315578511198, "minimum": 0.16790724905964372, "maximum": 0.19576370525509273, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9918662057809918, "standard_deviation": 0.0010879560152231233, "minimum": 0.9901841782411244, "maximum": 0.9935881581926023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2127542552971124, "standard_deviation": 0.028460953688248167, "minimum": 1.155576635393281, "maximum": 1.2466181556642186, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9596439360092333, "standard_deviation": 0.2929863129596832, "minimum": 0.5040685765937417, "maximum": 1.3616425452244567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10905196695058796, "standard_deviation": 0.008422177308121937, "minimum": 0.09436954718165506, "maximum": 0.11975562639021846, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9826017718905184, "standard_deviation": 0.002092395805481798, "minimum": 0.9789117949773202, "maximum": 0.984733490796903, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3818662479004464, "standard_deviation": 0.02799318725744907, "minimum": 2.333792524667696, "maximum": 2.4314710837516924, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8233542031670766, "standard_deviation": 0.47073854575074514, "minimum": 0.060047795556982496, "maximum": 1.4497586209243207, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16944284809748547, "standard_deviation": 0.009602709114662157, "minimum": 0.15888785458110516, "maximum": 0.18580623422601675, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947171962177939, "standard_deviation": 0.0006461783295508778, "minimum": 0.994080175489416, "maximum": 0.9959918873140864, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6334820887495187, "standard_deviation": 0.020074477687295266, "minimum": 1.600088069807254, "maximum": 1.6588554323077986, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0518808030568287, "standard_deviation": 0.5521102461779491, "minimum": 0.3728290488988457, "maximum": 1.8536102213815386, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09474797675230713, "standard_deviation": 0.005051069527753527, "minimum": 0.0878825929069499, "maximum": 0.09937792781984879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927734320086887, "standard_deviation": 0.0011643495879016562, "minimum": 0.9912485982840908, "maximum": 0.9943196790063048, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9006302320956667, "standard_deviation": 0.049647693013614415, "minimum": 1.7997398147349073, "maximum": 1.9532357197830001, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7121957461874615, "standard_deviation": 0.22953676248706967, "minimum": 0.37087684038991897, "maximum": 1.0029844225273787, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09813466857501965, "standard_deviation": 0.0075599833684518855, "minimum": 0.08730233268198724, "maximum": 0.10859815057208992, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997416440654927, "standard_deviation": 0.0005832645514205986, "minimum": 0.9967202709372375, "maximum": 0.9982228489246773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2440758978698196, "standard_deviation": 0.03232456606193444, "minimum": 1.1816690342716565, "maximum": 1.2827428266074725, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8574838769219209, "standard_deviation": 0.36497311707395846, "minimum": 0.10191062715140624, "maximum": 1.2960300015467159, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0599928066753925, "standard_deviation": 0.005676326707314904, "minimum": 0.050491071953506994, "maximum": 0.0657161331570276, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20654181016006457, "standard_deviation": 0.05455724239953927, "minimum": 0.12141401457205837, "maximum": 0.29750451738668016, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.9427676670450635, "standard_deviation": 0.009412982969859736, "minimum": 5.931146827457432, "maximum": 5.955660043182296, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5005386611688172, "standard_deviation": 0.03184396025595202, "minimum": 0.46875310542613147, "maximum": 0.5506368635440209, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08377376272554943, "standard_deviation": 0.02958588772178307, "minimum": -0.1247873614758546, "maximum": -0.029185980506413245, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956130745945282, "standard_deviation": 0.000383345831230502, "minimum": 0.9951232565089418, "maximum": 0.9962434035876454, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8117773257524722, "standard_deviation": 0.020662669110388755, "minimum": 0.7817012871204309, "maximum": 0.8350445948536438, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33862914942097955, "standard_deviation": 0.14074863539708588, "minimum": 0.1529086331877275, "maximum": 0.5572168301203585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.060335823809178374, "standard_deviation": 0.0025189418012234164, "minimum": 0.05626685062031117, "maximum": 0.063064035719062, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987125327427769, "standard_deviation": 0.00021448990802968015, "minimum": 0.9983716570700045, "maximum": 0.9991395670938189, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4303499996016428, "standard_deviation": 0.017882189436657207, "minimum": 0.40282522557432215, "maximum": 0.447280676004464, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3774145816037576, "standard_deviation": 0.11595464745230838, "minimum": 0.22444509250589853, "maximum": 0.49865873826975937, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03269431301332168, "standard_deviation": 0.0023451056445697427, "minimum": 0.02798832662712395, "maximum": 0.035827937516885164, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0028", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.2277913093566895, "predicted_mos": 3.2277913093566895, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "177283ce4440a790f60d7cf4fbdbeb097970fd57f72f015a36a729c265e8d546", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.230523299891502, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8270963655008483, "全帯域_10ms_rms_envelope_nrmse": 0.06175216382318108, "全帯域_10ms_envelope_correlation": 0.9953890072321555, "全帯域_10ms_modulation_mae_db": 0.23427513861688576, "全帯域_40ms_log_rms_mae_db": 0.4233493834144555, "全帯域_40ms_rms_envelope_nrmse": 0.0316686179172709, "全帯域_40ms_envelope_correlation": 0.9987911950138733, "全帯域_40ms_modulation_mae_db": 0.2171728122297382, "300–1000 Hz_10ms_log_rms_mae_db": 1.5290187742190606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09003145534486108, "300–1000 Hz_10ms_envelope_correlation": 0.9940755477713366, "300–1000 Hz_10ms_modulation_mae_db": 0.32044092716918515, "300–1000 Hz_40ms_log_rms_mae_db": 1.0292921689967367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05234644458610738, "300–1000 Hz_40ms_envelope_correlation": 0.9982341124782694, "300–1000 Hz_40ms_modulation_mae_db": 0.48700021984712344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7297549766093818, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18125538637452138, "1000–3000 Hz_10ms_envelope_correlation": 0.9756137537515331, "1000–3000 Hz_10ms_modulation_mae_db": 0.7357851601073802, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0683759304068434, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11006026049659363, "1000–3000 Hz_40ms_envelope_correlation": 0.991623803171215, "1000–3000 Hz_40ms_modulation_mae_db": 0.659203593382469, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8426926255254756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16811295590093006, "3000–6000 Hz_10ms_envelope_correlation": 0.9827842209969283, "3000–6000 Hz_10ms_modulation_mae_db": 0.23565181304876592, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3756309454213438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08815824516733993, "3000–6000 Hz_40ms_envelope_correlation": 0.9952476179949112, "3000–6000 Hz_40ms_modulation_mae_db": 0.2032456214456498, "mr_stft_db": 5.596163839051411, "band_mae_db": 0.22332902077772177, "rms_gain_error_db": -0.0849612454335568, "output_peak": 0.49681021906016865}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9014534559924205, "全帯域_10ms_rms_envelope_nrmse": 0.06236615604175751, "全帯域_10ms_envelope_correlation": 0.9954019897548546, "全帯域_10ms_modulation_mae_db": 0.3138724115111154, "全帯域_40ms_log_rms_mae_db": 0.489298067240278, "全帯域_40ms_rms_envelope_nrmse": 0.03361188968888446, "全帯域_40ms_envelope_correlation": 0.9987563020958281, "全帯域_40ms_modulation_mae_db": 0.2319024238353171, "300–1000 Hz_10ms_log_rms_mae_db": 1.5611566626123914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09319695928022023, "300–1000 Hz_10ms_envelope_correlation": 0.9937766263527913, "300–1000 Hz_10ms_modulation_mae_db": 0.6972776767732839, "300–1000 Hz_40ms_log_rms_mae_db": 1.0480077533433645, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05863707214871024, "300–1000 Hz_40ms_envelope_correlation": 0.9978318963166611, "300–1000 Hz_40ms_modulation_mae_db": 0.8190290000335964, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6638214249353434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1581662748215775, "1000–3000 Hz_10ms_envelope_correlation": 0.9817168983314122, "1000–3000 Hz_10ms_modulation_mae_db": 0.8639631073713279, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0341374781111279, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09855316262755763, "1000–3000 Hz_40ms_envelope_correlation": 0.9934709227766116, "1000–3000 Hz_40ms_modulation_mae_db": 0.7319371010453715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7911227014262785, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15690698001781783, "3000–6000 Hz_10ms_envelope_correlation": 0.9850233844000369, "3000–6000 Hz_10ms_modulation_mae_db": 0.6735434211040706, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3286344989978387, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09149491066006063, "3000–6000 Hz_40ms_envelope_correlation": 0.9948494645888515, "3000–6000 Hz_40ms_modulation_mae_db": 0.43078000804901073, "mr_stft_db": 5.593940410850688, "band_mae_db": 0.16656117589498823, "rms_gain_error_db": -0.12314674318065981, "output_peak": 0.5407606979192294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8202716586005204, "全帯域_10ms_rms_envelope_nrmse": 0.06223340601090436, "全帯域_10ms_envelope_correlation": 0.9952792776496545, "全帯域_10ms_modulation_mae_db": 0.37434896099814585, "全帯域_40ms_log_rms_mae_db": 0.4600148953960961, "全帯域_40ms_rms_envelope_nrmse": 0.037923333287752485, "全帯域_40ms_envelope_correlation": 0.9981620291492946, "全帯域_40ms_modulation_mae_db": 0.3777728860113994, "300–1000 Hz_10ms_log_rms_mae_db": 1.526620279205081, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0962341059018748, "300–1000 Hz_10ms_envelope_correlation": 0.9929765735420212, "300–1000 Hz_10ms_modulation_mae_db": 0.19735066370321833, "300–1000 Hz_40ms_log_rms_mae_db": 1.0483988906681005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06137436512303025, "300–1000 Hz_40ms_envelope_correlation": 0.9971398481917207, "300–1000 Hz_40ms_modulation_mae_db": 0.2548845282163204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.735957683639644, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16936366110072215, "1000–3000 Hz_10ms_envelope_correlation": 0.9794184692792657, "1000–3000 Hz_10ms_modulation_mae_db": 0.7470863627080387, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0029031959925163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09730658091979755, "1000–3000 Hz_40ms_envelope_correlation": 0.9945926946333975, "1000–3000 Hz_40ms_modulation_mae_db": 0.6798935574924232, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8074983899822499, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16107690986513395, "3000–6000 Hz_10ms_envelope_correlation": 0.9879982434610234, "3000–6000 Hz_10ms_modulation_mae_db": 1.3037440996322907, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3102779060827912, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11188968429647167, "3000–6000 Hz_40ms_envelope_correlation": 0.9954300658396688, "3000–6000 Hz_40ms_modulation_mae_db": 1.2879963477356275, "mr_stft_db": 5.5853197228899445, "band_mae_db": 0.2655350400715522, "rms_gain_error_db": -0.06536152266753954, "output_peak": 0.4795761924678535}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8782785342644897, "全帯域_10ms_rms_envelope_nrmse": 0.06372872213772922, "全帯域_10ms_envelope_correlation": 0.9951493361442385, "全帯域_10ms_modulation_mae_db": 0.28922281828264307, "全帯域_40ms_log_rms_mae_db": 0.46323746043633157, "全帯域_40ms_rms_envelope_nrmse": 0.032777570557907185, "全帯域_40ms_envelope_correlation": 0.9987902530155736, "全帯域_40ms_modulation_mae_db": 0.4009171912548614, "300–1000 Hz_10ms_log_rms_mae_db": 1.5309619971549362, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10409675350003351, "300–1000 Hz_10ms_envelope_correlation": 0.9921713453221049, "300–1000 Hz_10ms_modulation_mae_db": 0.7147787209730291, "300–1000 Hz_40ms_log_rms_mae_db": 1.0272885207693407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06135998667817585, "300–1000 Hz_40ms_envelope_correlation": 0.9976886472193129, "300–1000 Hz_40ms_modulation_mae_db": 0.8390324798984409, "1000–3000 Hz_10ms_log_rms_mae_db": 1.790692800158497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16984278905453726, "1000–3000 Hz_10ms_envelope_correlation": 0.9788390340346248, "1000–3000 Hz_10ms_modulation_mae_db": 0.36593149633103095, "1000–3000 Hz_40ms_log_rms_mae_db": 1.115208050464745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1059386036991803, "1000–3000 Hz_40ms_envelope_correlation": 0.9924132884174715, "1000–3000 Hz_40ms_modulation_mae_db": 0.47946426899791567, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8481492695286932, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14905496724187248, "3000–6000 Hz_10ms_envelope_correlation": 0.9864939296514353, "3000–6000 Hz_10ms_modulation_mae_db": 0.28995581265262893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3425102205942168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07761015841460081, "3000–6000 Hz_40ms_envelope_correlation": 0.9962493288703201, "3000–6000 Hz_40ms_modulation_mae_db": 0.14642762681099944, "mr_stft_db": 5.613550297439979, "band_mae_db": 0.1800612081926297, "rms_gain_error_db": -0.10972855755192082, "output_peak": 0.5139452515684684}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8731776943068064, "全帯域_10ms_rms_envelope_nrmse": 0.0572029126056831, "全帯域_10ms_envelope_correlation": 0.9959959466069661, "全帯域_10ms_modulation_mae_db": 0.25957525153370037, "全帯域_40ms_log_rms_mae_db": 0.5102442996337744, "全帯域_40ms_rms_envelope_nrmse": 0.03008725364981253, "全帯域_40ms_envelope_correlation": 0.9988244700531771, "全帯域_40ms_modulation_mae_db": 0.407004631460895, "300–1000 Hz_10ms_log_rms_mae_db": 1.5709382273958703, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09494663160687586, "300–1000 Hz_10ms_envelope_correlation": 0.9931494322511273, "300–1000 Hz_10ms_modulation_mae_db": 0.47146465026551265, "300–1000 Hz_40ms_log_rms_mae_db": 1.058534256903558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05588234877418785, "300–1000 Hz_40ms_envelope_correlation": 0.9976543091624143, "300–1000 Hz_40ms_modulation_mae_db": 0.6530011769566868, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6347981097472501, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14619608939360612, "1000–3000 Hz_10ms_envelope_correlation": 0.9848910322571484, "1000–3000 Hz_10ms_modulation_mae_db": 1.00256010214102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9726807896702281, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09327295739656487, "1000–3000 Hz_40ms_envelope_correlation": 0.9946698629274269, "1000–3000 Hz_40ms_modulation_mae_db": 1.0055317773478285, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8887253019613655, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1446116464572822, "3000–6000 Hz_10ms_envelope_correlation": 0.9873510621842849, "3000–6000 Hz_10ms_modulation_mae_db": 0.5370571336195885, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3578137548455744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08421368440558949, "3000–6000 Hz_40ms_envelope_correlation": 0.995599053208768, "3000–6000 Hz_40ms_modulation_mae_db": 0.8392603563130155, "mr_stft_db": 5.5975286481019895, "band_mae_db": 0.172902923071093, "rms_gain_error_db": -0.03439552017737354, "output_peak": 0.5415838743098151}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7875537172131252, "全帯域_10ms_rms_envelope_nrmse": 0.059232419625685526, "全帯域_10ms_envelope_correlation": 0.9957488895298331, "全帯域_10ms_modulation_mae_db": 0.35432486545300934, "全帯域_40ms_log_rms_mae_db": 0.39085212308466794, "全帯域_40ms_rms_envelope_nrmse": 0.03045464958971145, "全帯域_40ms_envelope_correlation": 0.9988670749040269, "全帯域_40ms_modulation_mae_db": 0.3173688915174238, "300–1000 Hz_10ms_log_rms_mae_db": 1.5661359603341294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09344314801591504, "300–1000 Hz_10ms_envelope_correlation": 0.9933983545417412, "300–1000 Hz_10ms_modulation_mae_db": 0.4756586769056166, "300–1000 Hz_40ms_log_rms_mae_db": 1.0341143220216085, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055188322121836164, "300–1000 Hz_40ms_envelope_correlation": 0.9977548146109295, "300–1000 Hz_40ms_modulation_mae_db": 0.9161757337845721, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6951446856994337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1596262319106834, "1000–3000 Hz_10ms_envelope_correlation": 0.9809483464067673, "1000–3000 Hz_10ms_modulation_mae_db": 0.5432908868846339, "1000–3000 Hz_40ms_log_rms_mae_db": 0.982258912379273, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08862828366997594, "1000–3000 Hz_40ms_envelope_correlation": 0.9945240882194258, "1000–3000 Hz_40ms_modulation_mae_db": 0.5132358414644419, "3000–6000 Hz_10ms_log_rms_mae_db": 1.772207040669569, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14466170540313383, "3000–6000 Hz_10ms_envelope_correlation": 0.9873903976437679, "3000–6000 Hz_10ms_modulation_mae_db": 1.1064753376646168, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2789046703018645, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08892674402107589, "3000–6000 Hz_40ms_envelope_correlation": 0.9950991589977431, "3000–6000 Hz_40ms_modulation_mae_db": 0.9339018283157032, "mr_stft_db": 5.598783026314937, "band_mae_db": 0.10732996088039903, "rms_gain_error_db": -0.0756678824615764, "output_peak": 0.5407354462567511}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8803640118457812, "全帯域_10ms_rms_envelope_nrmse": 0.05885951989979053, "全帯域_10ms_envelope_correlation": 0.9958095154880383, "全帯域_10ms_modulation_mae_db": 0.22898377086650193, "全帯域_40ms_log_rms_mae_db": 0.4488566373244517, "全帯域_40ms_rms_envelope_nrmse": 0.031027743968542278, "全帯域_40ms_envelope_correlation": 0.9988141619473786, "全帯域_40ms_modulation_mae_db": 0.2393661354211051, "300–1000 Hz_10ms_log_rms_mae_db": 1.5945687480855413, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09897672588952387, "300–1000 Hz_10ms_envelope_correlation": 0.9924728065453405, "300–1000 Hz_10ms_modulation_mae_db": 0.35762228096104237, "300–1000 Hz_40ms_log_rms_mae_db": 1.166784754423832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057906042465575625, "300–1000 Hz_40ms_envelope_correlation": 0.9973705885649141, "300–1000 Hz_40ms_modulation_mae_db": 0.665987242045103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.772576104431228, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18960465287863867, "1000–3000 Hz_10ms_envelope_correlation": 0.9756602981417365, "1000–3000 Hz_10ms_modulation_mae_db": 0.5455970317083578, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1160030635639295, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13203492769377959, "1000–3000 Hz_40ms_envelope_correlation": 0.9904839156517347, "1000–3000 Hz_40ms_modulation_mae_db": 0.5332806810524299, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8423598387197238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15471350318080299, "3000–6000 Hz_10ms_envelope_correlation": 0.9859127341734413, "3000–6000 Hz_10ms_modulation_mae_db": 0.5070019722047151, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3880021948723635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0961005798050345, "3000–6000 Hz_40ms_envelope_correlation": 0.995052858398562, "3000–6000 Hz_40ms_modulation_mae_db": 0.5446912183766727, "mr_stft_db": 5.597634114354576, "band_mae_db": 0.26316066069464483, "rms_gain_error_db": -0.07216852566812386, "output_peak": 0.5102142302390517}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8996974979098407, "全帯域_10ms_rms_envelope_nrmse": 0.062146769521626184, "全帯域_10ms_envelope_correlation": 0.9953464804982359, "全帯域_10ms_modulation_mae_db": 0.1599193729329683, "全帯域_40ms_log_rms_mae_db": 0.48684085989917075, "全帯域_40ms_rms_envelope_nrmse": 0.03010992059885523, "全帯域_40ms_envelope_correlation": 0.9989483087653891, "全帯域_40ms_modulation_mae_db": 0.18776177417351056, "300–1000 Hz_10ms_log_rms_mae_db": 1.6227893750369455, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0980418733142339, "300–1000 Hz_10ms_envelope_correlation": 0.9930237714375129, "300–1000 Hz_10ms_modulation_mae_db": 0.5222197634770288, "300–1000 Hz_40ms_log_rms_mae_db": 1.0895108346326037, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05937934792239603, "300–1000 Hz_40ms_envelope_correlation": 0.9977454637584278, "300–1000 Hz_40ms_modulation_mae_db": 0.6899865446098309, "1000–3000 Hz_10ms_log_rms_mae_db": 1.736921115585614, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18430828213858286, "1000–3000 Hz_10ms_envelope_correlation": 0.975896262757504, "1000–3000 Hz_10ms_modulation_mae_db": 0.6725709168606159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1126689941143604, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12377379421569965, "1000–3000 Hz_40ms_envelope_correlation": 0.9904875005385677, "1000–3000 Hz_40ms_modulation_mae_db": 0.61356258247818, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8275047307163301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14926570694800692, "3000–6000 Hz_10ms_envelope_correlation": 0.9865736542366587, "3000–6000 Hz_10ms_modulation_mae_db": 0.4313212927903883, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3352551882839114, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08159272106794961, "3000–6000 Hz_40ms_envelope_correlation": 0.995877706547738, "3000–6000 Hz_40ms_modulation_mae_db": 0.43659160883659437, "mr_stft_db": 5.604701127182203, "band_mae_db": 0.20798935411697791, "rms_gain_error_db": -0.09288899730978992, "output_peak": 0.5296808485855167}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.799787429548736, "spectral_convergence": 0.20435491609626963, "spectral_convergence_status": "ok", "amplitude_floor": 1.711947899928738e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.313286835788074, "spectral_convergence": 0.20431383897806663, "spectral_convergence_status": "ok", "amplitude_floor": 1.4743066880059505e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.675417251817427, "spectral_convergence": 0.21154484874258564, "spectral_convergence_status": "ok", "amplitude_floor": 1.2829193968481007e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015568083457567275, "output_power": 0.0015736490931737616, "delta_db": 0.04672744717893903, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001958578248681761, "output_power": 0.0018985688851290705, "delta_db": -0.13514567946213152, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00020973665094298736, "output_power": 0.00019786263263640557, "delta_db": -0.25310545715809757, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018883580517861778, "output_power": 0.00017062615712363995, "delta_db": -0.440387347765511, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010837478072549288, "output_power": 0.00010001228313296395, "delta_db": -0.34874890085159865, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.95174204959434e-05, "output_power": 8.1667302806438e-05, "delta_db": 0.11585929225005266, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22332902077772177, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0849612454335568, "pitch": null, "output_rms_dbfs": -23.95516217341283, "output_peak": 0.49681021906016865, "reference_rms_dbfs": -23.870200927979273, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.596163839051411}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023872023988588299, "log_rms_mae_db": 0.8270963655008483, "rms_envelope_nrmse": 0.06175216382318108, "envelope_correlation": 0.9953890072321555, "rms_gain_error_db": -0.08466970466424105, "reference_mean_power": 0.004104010412713017, "output_mean_power": 0.004024773858615444}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.972155761177528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037154075128907306, "output_power": 0.0003587607378922386, "reference_active": true, "delta_db": -0.15201546353098994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.499306084417706e-05, "output_power": 5.1815241993589214e-05, "reference_active": true, "delta_db": -0.25850361847924436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6530449727349195e-05, "output_power": 1.768134517524501e-05, "reference_active": true, "delta_db": 0.2923063338404229}], "active_band_count": 3, "active_band_mae_db": 0.23427513861688576}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00021497588341920653, "log_rms_mae_db": 0.4233493834144555, "rms_envelope_nrmse": 0.0316686179172709, "envelope_correlation": 0.9987911950138733, "rms_gain_error_db": -0.08404329755537748, "reference_mean_power": 0.00411739985369924, "output_mean_power": 0.004038487239768918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.973189563381595e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033993681960372955, "output_power": 0.00032875288308917543, "reference_active": true, "delta_db": -0.1452863672357153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4133967175911375e-05, "output_power": 2.3171553628759365e-05, "reference_active": true, "delta_db": -0.17673563959685795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.827774931478282e-07, "output_power": 9.523594489615605e-07, "reference_active": true, "delta_db": 0.32949642985664135}], "active_band_count": 3, "active_band_mae_db": 0.2171728122297382}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001937632671887944, "log_rms_mae_db": 1.5290187742190606, "rms_envelope_nrmse": 0.09003145534486108, "envelope_correlation": 0.9940755477713366, "rms_gain_error_db": -0.22101834014049765, "reference_mean_power": 0.0014622039835396937, "output_mean_power": 0.0013896522301351552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.890315707731239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020584065582039213, "output_power": 0.00019370995412202198, "reference_active": true, "delta_db": -0.26379218624310663}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2632804837837524e-05, "output_power": 3.680607106732955e-05, "reference_active": true, "delta_db": -0.6382444573512419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.198246611681044e-06, "output_power": 8.310929379481792e-06, "reference_active": true, "delta_db": 0.059286137913206895}], "active_band_count": 3, "active_band_mae_db": 0.32044092716918515}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001823230477151681, "log_rms_mae_db": 1.0292921689967367, "rms_envelope_nrmse": 0.05234644458610738, "envelope_correlation": 0.9982341124782694, "rms_gain_error_db": -0.22073327910460427, "reference_mean_power": 0.0014670926652510829, "output_mean_power": 0.0013943898664298218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5556250379109994e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020148633268320183, "output_power": 0.00018882003408169005, "reference_active": true, "delta_db": -0.28197520519995234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9312508746169674e-05, "output_power": 1.7211937376564904e-05, "reference_active": true, "delta_db": -0.5000893616204958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.933225179066708e-07, "output_power": 5.929822536229255e-07, "reference_active": true, "delta_db": -0.6789360927209221}], "active_band_count": 3, "active_band_mae_db": 0.48700021984712344}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.475794683802334e-05, "log_rms_mae_db": 1.7297549766093818, "rms_envelope_nrmse": 0.18125538637452138, "envelope_correlation": 0.9756137537515331, "rms_gain_error_db": -0.4353955285796584, "reference_mean_power": 0.000205358190626383, "output_mean_power": 0.00018576867193764257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.94981889493768e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.427037048631366e-05, "output_power": 4.677023439526312e-05, "reference_active": true, "delta_db": -0.6459323993930027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.925199890017768e-06, "output_power": 7.63806938247019e-06, "reference_active": true, "delta_db": -0.6763435232608643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.574401681634841e-06, "output_power": 2.099753697424422e-06, "reference_active": true, "delta_db": -0.8850795576682736}], "active_band_count": 3, "active_band_mae_db": 0.7357851601073802}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.971385946015731e-05, "log_rms_mae_db": 1.0683759304068434, "rms_envelope_nrmse": 0.11006026049659363, "envelope_correlation": 0.991623803171215, "rms_gain_error_db": -0.43536558570404704, "reference_mean_power": 0.00020637707638038015, "output_mean_power": 0.00018669165135075074}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.280870922430589e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.267995669092476e-05, "output_power": 4.551619026822178e-05, "reference_active": true, "delta_db": -0.6347950504334748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.180545251002319e-06, "output_power": 3.619064411697204e-06, "reference_active": true, "delta_db": -0.6263661597815426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0193067246169046e-07, "output_power": 1.7122112039582795e-07, "reference_active": true, "delta_db": -0.7164495699323896}], "active_band_count": 3, "active_band_mae_db": 0.659203593382469}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.517166954195252e-05, "log_rms_mae_db": 1.8426926255254756, "rms_envelope_nrmse": 0.16811295590093006, "envelope_correlation": 0.9827842209969283, "rms_gain_error_db": -0.1414343902011566, "reference_mean_power": 0.00014718641611289376, "output_mean_power": 0.00014247028452915296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4339203545791116e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.0687614509069e-05, "output_power": 6.940185166685243e-05, "reference_active": true, "delta_db": -0.07972268051794934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.343067234178686e-06, "output_power": 8.809108969488452e-06, "reference_active": true, "delta_db": -0.2555749182802871}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3595191848035655e-06, "output_power": 2.5703324366295586e-06, "reference_active": true, "delta_db": 0.3716578403480613}], "active_band_count": 3, "active_band_mae_db": 0.23565181304876592}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.54521457203503e-05, "log_rms_mae_db": 1.3756309454213438, "rms_envelope_nrmse": 0.08815824516733993, "envelope_correlation": 0.9952476179949112, "rms_gain_error_db": -0.1413814127164709, "reference_mean_power": 0.0001479201087162099, "output_mean_power": 0.00014318221483162227}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3826641859038576e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.726675521598142e-05, "output_power": 6.634622200372293e-05, "reference_active": true, "delta_db": -0.05984281292669079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.27086384707959e-06, "output_power": 4.004718836497969e-06, "reference_active": true, "delta_db": -0.2794369604339673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6847652723937028e-07, "output_power": 1.7930199057416432e-07, "reference_active": true, "delta_db": 0.27045709097629134}], "active_band_count": 3, "active_band_mae_db": 0.2032456214456498}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8915586154390818, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6da6d3f27aae9816dc8968c823ca498be508873e2a09f1a70d5ff6941b0125b4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.979123011869999, "standard_deviation": 0.003342695224878325, "minimum": 0.9756137537515331, "maximum": 0.9848910322571484, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7199583626007988, "standard_deviation": 0.05268642725516909, "minimum": 1.6347981097472501, "maximum": 1.790692800158497, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6845981330140507, "standard_deviation": 0.2002778861148296, "minimum": 0.36593149633103095, "maximum": 1.00256010214102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16979542095910866, "standard_deviation": 0.014784770990157613, "minimum": 0.14619608939360612, "maximum": 0.18960465287863867, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927832595419813, "standard_deviation": 0.0017878719324760433, "minimum": 0.9904839156517347, "maximum": 0.9946698629274269, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.050529551837878, "standard_deviation": 0.060837221069389685, "minimum": 0.9726807896702281, "maximum": 1.1160030635639295, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6520136754076324, "standard_deviation": 0.16767331351382925, "minimum": 0.47946426899791567, "maximum": 1.0055317773478285, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10619607133989364, "standard_deviation": 0.015140731335996626, "minimum": 0.08862828366997594, "maximum": 0.13203492769377959, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9861909533434471, "standard_deviation": 0.0016619919100331884, "minimum": 0.9827842209969283, "maximum": 0.9879982434610234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8275324873162107, "standard_deviation": 0.03665911657180641, "minimum": 1.772207040669569, "maximum": 1.8887253019613655, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6355938603396332, "standard_deviation": 0.38116104964715886, "minimum": 0.23565181304876592, "maximum": 1.3037440996322907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15355054687687253, "standard_deviation": 0.00827054619147066, "minimum": 0.1446116464572822, "maximum": 0.16811295590093006, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954256568058204, "standard_deviation": 0.0004663146012478894, "minimum": 0.9948494645888515, "maximum": 0.9962493288703201, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.339628672424988, "standard_deviation": 0.03518123080963933, "minimum": 1.2789046703018645, "maximum": 1.3880021948723635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6028618269854091, "standard_deviation": 0.38978897350253516, "minimum": 0.14642762681099944, "maximum": 1.2879963477356275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08999834097976532, "standard_deviation": 0.010561560960051567, "minimum": 0.07761015841460081, "maximum": 0.11188968429647167, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931305572204969, "standard_deviation": 0.0006290947839437432, "minimum": 0.9921713453221049, "maximum": 0.9940755477713366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5627737530054944, "standard_deviation": 0.03412547170265052, "minimum": 1.526620279205081, "maximum": 1.6227893750369455, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4696016700284896, "standard_deviation": 0.17875198035839268, "minimum": 0.19735066370321833, "maximum": 0.7147787209730291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09612095660669229, "standard_deviation": 0.004303702095339425, "minimum": 0.09003145534486108, "maximum": 0.10409675350003351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976774600378312, "standard_deviation": 0.00032252983332433017, "minimum": 0.9971398481917207, "maximum": 0.9982341124782694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.062741437719893, "standard_deviation": 0.046563900019385644, "minimum": 1.0272885207693407, "maximum": 1.166784754423832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6656371156739592, "standard_deviation": 0.21305469940361557, "minimum": 0.2548845282163204, "maximum": 0.9161757337845721, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05775924122750242, "standard_deviation": 0.0031369250144372644, "minimum": 0.05234644458610738, "maximum": 0.06137436512303025, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19835866796250085, "standard_deviation": 0.05306105629251529, "minimum": 0.10732996088039903, "maximum": 0.2655350400715522, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.598452648273216, "standard_deviation": 0.00816515318826841, "minimum": 5.5853197228899445, "maximum": 5.613550297439979, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5191633450508568, "standard_deviation": 0.023050315339966043, "minimum": 0.4795761924678535, "maximum": 0.5415838743098151, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0822898743063176, "standard_deviation": 0.02743400763779049, "minimum": -0.12314674318065981, "maximum": -0.03439552017737354, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955150553629971, "standard_deviation": 0.00029738688041150814, "minimum": 0.9951493361442385, "maximum": 0.9959959466069661, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8584866169542291, "standard_deviation": 0.04158959573167929, "minimum": 0.7875537172131252, "maximum": 0.9014534559924205, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27681532377437124, "standard_deviation": 0.07081880433656389, "minimum": 0.1599193729329683, "maximum": 0.37434896099814585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0609402587082947, "standard_deviation": 0.002229982188250133, "minimum": 0.0572029126056831, "maximum": 0.06372872213772922, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987442243680676, "standard_deviation": 0.00024245628549220038, "minimum": 0.9981620291492946, "maximum": 0.9989483087653891, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45908671580365323, "standard_deviation": 0.03846329962688335, "minimum": 0.39085212308466794, "maximum": 0.5102442996337744, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.29740834323803134, "standard_deviation": 0.08920469649843743, "minimum": 0.18776177417351056, "maximum": 0.407004631460895, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03220762240734207, "standard_deviation": 0.002635418849227216, "minimum": 0.03008725364981253, "maximum": 0.037923333287752485, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0028", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.249767303466797, "predicted_mos": 3.249767303466797, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "803f6ab95ff7dfe5f80df00f784e179cfbd0d98555e95504919a6546a27ca834", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.051444699987769, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8518566701071905, "全帯域_10ms_rms_envelope_nrmse": 0.06592494178921919, "全帯域_10ms_envelope_correlation": 0.9949907477594534, "全帯域_10ms_modulation_mae_db": 0.2775441302092923, "全帯域_40ms_log_rms_mae_db": 0.49101132800409064, "全帯域_40ms_rms_envelope_nrmse": 0.036828152302997644, "全帯域_40ms_envelope_correlation": 0.9986460767718954, "全帯域_40ms_modulation_mae_db": 0.3892458093246732, "300–1000 Hz_10ms_log_rms_mae_db": 1.7158501455456512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10020813071637892, "300–1000 Hz_10ms_envelope_correlation": 0.9928576345041884, "300–1000 Hz_10ms_modulation_mae_db": 0.6862090759269664, "300–1000 Hz_40ms_log_rms_mae_db": 1.1178642166821882, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06008920046118952, "300–1000 Hz_40ms_envelope_correlation": 0.9978834370093633, "300–1000 Hz_40ms_modulation_mae_db": 0.7776529120750318, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7771209729249409, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1782262339639615, "1000–3000 Hz_10ms_envelope_correlation": 0.9786737331352349, "1000–3000 Hz_10ms_modulation_mae_db": 0.9898517352671119, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0712086962193077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12510398623094318, "1000–3000 Hz_40ms_envelope_correlation": 0.9914187630027286, "1000–3000 Hz_40ms_modulation_mae_db": 1.291641609151841, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0167669539055573, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17030006419240015, "3000–6000 Hz_10ms_envelope_correlation": 0.9830884832355697, "3000–6000 Hz_10ms_modulation_mae_db": 1.1140837570822437, "3000–6000 Hz_40ms_log_rms_mae_db": 1.454441458231002, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10419812028618873, "3000–6000 Hz_40ms_envelope_correlation": 0.9945294366195109, "3000–6000 Hz_40ms_modulation_mae_db": 1.7316423693638923, "mr_stft_db": 5.727069996437467, "band_mae_db": 0.2804046743123147, "rms_gain_error_db": -0.162175821727561, "output_peak": 0.5176862870743543}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8268173717099145, "全帯域_10ms_rms_envelope_nrmse": 0.057962160120314105, "全帯域_10ms_envelope_correlation": 0.9960938530100799, "全帯域_10ms_modulation_mae_db": 0.18951607242905047, "全帯域_40ms_log_rms_mae_db": 0.4584864426197898, "全帯域_40ms_rms_envelope_nrmse": 0.03314717533568964, "全帯域_40ms_envelope_correlation": 0.9988402691596323, "全帯域_40ms_modulation_mae_db": 0.17667884109316834, "300–1000 Hz_10ms_log_rms_mae_db": 1.6564651211404555, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09188128314405748, "300–1000 Hz_10ms_envelope_correlation": 0.9939105176609356, "300–1000 Hz_10ms_modulation_mae_db": 0.5640812980556799, "300–1000 Hz_40ms_log_rms_mae_db": 1.1154258432312858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05559132186231039, "300–1000 Hz_40ms_envelope_correlation": 0.9980557512606361, "300–1000 Hz_40ms_modulation_mae_db": 0.7813377578000608, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7926135777207706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1741499670502947, "1000–3000 Hz_10ms_envelope_correlation": 0.9780354608814466, "1000–3000 Hz_10ms_modulation_mae_db": 0.8212536227684262, "1000–3000 Hz_40ms_log_rms_mae_db": 1.095820503189689, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11160045205707779, "1000–3000 Hz_40ms_envelope_correlation": 0.9919377403885283, "1000–3000 Hz_40ms_modulation_mae_db": 0.8419438196990722, "3000–6000 Hz_10ms_log_rms_mae_db": 1.978119292100297, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15028364749664364, "3000–6000 Hz_10ms_envelope_correlation": 0.9864032965430816, "3000–6000 Hz_10ms_modulation_mae_db": 1.196011875522798, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3943386659593011, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0854186236766497, "3000–6000 Hz_40ms_envelope_correlation": 0.9958053196942039, "3000–6000 Hz_40ms_modulation_mae_db": 1.777156453746671, "mr_stft_db": 5.721174986733138, "band_mae_db": 0.20974254899039682, "rms_gain_error_db": -0.12952355073462488, "output_peak": 0.5037365699842349}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8277642491477639, "全帯域_10ms_rms_envelope_nrmse": 0.06303536662249638, "全帯域_10ms_envelope_correlation": 0.9951567600112193, "全帯域_10ms_modulation_mae_db": 0.19824800939825338, "全帯域_40ms_log_rms_mae_db": 0.45428395857888443, "全帯域_40ms_rms_envelope_nrmse": 0.03335048833770338, "全帯域_40ms_envelope_correlation": 0.998589200539744, "全帯域_40ms_modulation_mae_db": 0.34869771915051667, "300–1000 Hz_10ms_log_rms_mae_db": 1.739350558595185, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10030964201415418, "300–1000 Hz_10ms_envelope_correlation": 0.9924301039254992, "300–1000 Hz_10ms_modulation_mae_db": 0.8271812475814714, "300–1000 Hz_40ms_log_rms_mae_db": 1.1461656398639881, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05981203043146401, "300–1000 Hz_40ms_envelope_correlation": 0.9974120983826125, "300–1000 Hz_40ms_modulation_mae_db": 1.2443973722212862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7792492592240567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1628518383242331, "1000–3000 Hz_10ms_envelope_correlation": 0.9804474817628812, "1000–3000 Hz_10ms_modulation_mae_db": 0.8529392307186822, "1000–3000 Hz_40ms_log_rms_mae_db": 1.081457071569328, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09716095420988263, "1000–3000 Hz_40ms_envelope_correlation": 0.9936218099148004, "1000–3000 Hz_40ms_modulation_mae_db": 0.8158144677495546, "3000–6000 Hz_10ms_log_rms_mae_db": 2.006441813223179, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20850651046779228, "3000–6000 Hz_10ms_envelope_correlation": 0.9737634504421451, "3000–6000 Hz_10ms_modulation_mae_db": 1.5727683483613795, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4446016133964867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1150877222970187, "3000–6000 Hz_40ms_envelope_correlation": 0.991781971580915, "3000–6000 Hz_40ms_modulation_mae_db": 1.2146428580127964, "mr_stft_db": 5.712780799070873, "band_mae_db": 0.25266748856140997, "rms_gain_error_db": -0.06390203007289319, "output_peak": 0.4722509973817107}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.808493823415538, "全帯域_10ms_rms_envelope_nrmse": 0.06535186252132781, "全帯域_10ms_envelope_correlation": 0.9948724163241324, "全帯域_10ms_modulation_mae_db": 0.36142448211412165, "全帯域_40ms_log_rms_mae_db": 0.4201734324016691, "全帯域_40ms_rms_envelope_nrmse": 0.03652514716355659, "全帯域_40ms_envelope_correlation": 0.9984108774778213, "全帯域_40ms_modulation_mae_db": 0.31457384612156714, "300–1000 Hz_10ms_log_rms_mae_db": 1.7285835890527836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10296877714530268, "300–1000 Hz_10ms_envelope_correlation": 0.9922264400523907, "300–1000 Hz_10ms_modulation_mae_db": 0.7214790308024028, "300–1000 Hz_40ms_log_rms_mae_db": 1.1592035280978514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06252982875960196, "300–1000 Hz_40ms_envelope_correlation": 0.9974076837553973, "300–1000 Hz_40ms_modulation_mae_db": 0.9551275948493249, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7733614476017017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16781368171021344, "1000–3000 Hz_10ms_envelope_correlation": 0.9793909509936448, "1000–3000 Hz_10ms_modulation_mae_db": 0.523770103952897, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0826678552867395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10141597110316876, "1000–3000 Hz_40ms_envelope_correlation": 0.9932820358854129, "1000–3000 Hz_40ms_modulation_mae_db": 0.6768485655413518, "3000–6000 Hz_10ms_log_rms_mae_db": 1.995814508407356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15981066985466424, "3000–6000 Hz_10ms_envelope_correlation": 0.9848918376925436, "3000–6000 Hz_10ms_modulation_mae_db": 0.6926322505399042, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4119164109131652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09506327085431837, "3000–6000 Hz_40ms_envelope_correlation": 0.994577866804886, "3000–6000 Hz_40ms_modulation_mae_db": 0.26387682829811165, "mr_stft_db": 5.722471864582272, "band_mae_db": 0.20431339097723733, "rms_gain_error_db": -0.10544902850218063, "output_peak": 0.5518399350995127}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8120579372866382, "全帯域_10ms_rms_envelope_nrmse": 0.060277079361930955, "全帯域_10ms_envelope_correlation": 0.9955922311006412, "全帯域_10ms_modulation_mae_db": 0.22623849456549983, "全帯域_40ms_log_rms_mae_db": 0.4266863171668675, "全帯域_40ms_rms_envelope_nrmse": 0.03311306753489981, "全帯域_40ms_envelope_correlation": 0.9986398917438296, "全帯域_40ms_modulation_mae_db": 0.32466813773530595, "300–1000 Hz_10ms_log_rms_mae_db": 1.6916694392251137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09681907013621091, "300–1000 Hz_10ms_envelope_correlation": 0.9929630107916472, "300–1000 Hz_10ms_modulation_mae_db": 0.795835045262273, "300–1000 Hz_40ms_log_rms_mae_db": 1.1322538562110918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05693819983136715, "300–1000 Hz_40ms_envelope_correlation": 0.9976788992816648, "300–1000 Hz_40ms_modulation_mae_db": 1.4200381860356555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.822911392259463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17449940962310148, "1000–3000 Hz_10ms_envelope_correlation": 0.9785552015658885, "1000–3000 Hz_10ms_modulation_mae_db": 1.2064497908174863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1252461864195427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10997745445416866, "1000–3000 Hz_40ms_envelope_correlation": 0.9930312829084288, "1000–3000 Hz_40ms_modulation_mae_db": 1.6664632340284184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9876895143311302, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15406551443292318, "3000–6000 Hz_10ms_envelope_correlation": 0.9855527189115155, "3000–6000 Hz_10ms_modulation_mae_db": 0.3229743059569851, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3761214235196906, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09033587647169439, "3000–6000 Hz_40ms_envelope_correlation": 0.99498946063273, "3000–6000 Hz_40ms_modulation_mae_db": 0.3859998986653929, "mr_stft_db": 5.7025928635967595, "band_mae_db": 0.19449384151091512, "rms_gain_error_db": -0.07559252104772657, "output_peak": 0.5588181957145346}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7847789347706304, "全帯域_10ms_rms_envelope_nrmse": 0.058795707104383116, "全帯域_10ms_envelope_correlation": 0.9959072674216183, "全帯域_10ms_modulation_mae_db": 0.4047820262756994, "全帯域_40ms_log_rms_mae_db": 0.4137499335179461, "全帯域_40ms_rms_envelope_nrmse": 0.03524362967226694, "全帯域_40ms_envelope_correlation": 0.9985682813224971, "全帯域_40ms_modulation_mae_db": 0.5191761989573073, "300–1000 Hz_10ms_log_rms_mae_db": 1.692647539408286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09701815701294277, "300–1000 Hz_10ms_envelope_correlation": 0.9930553273100792, "300–1000 Hz_10ms_modulation_mae_db": 1.020416398309923, "300–1000 Hz_40ms_log_rms_mae_db": 1.138992922042794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060309357768688836, "300–1000 Hz_40ms_envelope_correlation": 0.9974891787927082, "300–1000 Hz_40ms_modulation_mae_db": 1.3952495645480463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7875743126592298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16770886242660513, "1000–3000 Hz_10ms_envelope_correlation": 0.9796548333544931, "1000–3000 Hz_10ms_modulation_mae_db": 0.26951324278941713, "1000–3000 Hz_40ms_log_rms_mae_db": 1.091100887601681, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10710783798351113, "1000–3000 Hz_40ms_envelope_correlation": 0.9925922535744844, "1000–3000 Hz_40ms_modulation_mae_db": 0.45259832137066475, "3000–6000 Hz_10ms_log_rms_mae_db": 2.015486756980413, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1469423986694379, "3000–6000 Hz_10ms_envelope_correlation": 0.9869546496432448, "3000–6000 Hz_10ms_modulation_mae_db": 0.656945683541145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.434618098924459, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0953418463694667, "3000–6000 Hz_40ms_envelope_correlation": 0.9945564249054685, "3000–6000 Hz_40ms_modulation_mae_db": 0.8168853752898219, "mr_stft_db": 5.7100704474767285, "band_mae_db": 0.18981981383707522, "rms_gain_error_db": -0.111184045085524, "output_peak": 0.5050260224460934}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7899959593502656, "全帯域_10ms_rms_envelope_nrmse": 0.05900397468193975, "全帯域_10ms_envelope_correlation": 0.9959923076536592, "全帯域_10ms_modulation_mae_db": 0.5204267466819144, "全帯域_40ms_log_rms_mae_db": 0.4225880581525095, "全帯域_40ms_rms_envelope_nrmse": 0.035715678459146506, "全帯域_40ms_envelope_correlation": 0.9986571033881199, "全帯域_40ms_modulation_mae_db": 0.3830235736479892, "300–1000 Hz_10ms_log_rms_mae_db": 1.7573795662125262, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09708169963932908, "300–1000 Hz_10ms_envelope_correlation": 0.9934326297982813, "300–1000 Hz_10ms_modulation_mae_db": 0.9925126080394874, "300–1000 Hz_40ms_log_rms_mae_db": 1.1859854814215915, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06428391271176258, "300–1000 Hz_40ms_envelope_correlation": 0.9975230038577027, "300–1000 Hz_40ms_modulation_mae_db": 1.227240606424244, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8640994653899863, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18224198820737403, "1000–3000 Hz_10ms_envelope_correlation": 0.9767060123872323, "1000–3000 Hz_10ms_modulation_mae_db": 0.8401069145682843, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1216273114628454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12102441593731504, "1000–3000 Hz_40ms_envelope_correlation": 0.9913316577818772, "1000–3000 Hz_40ms_modulation_mae_db": 1.1797586396089814, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0126780891665015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15355516587565737, "3000–6000 Hz_10ms_envelope_correlation": 0.9857374688167663, "3000–6000 Hz_10ms_modulation_mae_db": 0.04612746585419955, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4368157154970047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09276262762978568, "3000–6000 Hz_40ms_envelope_correlation": 0.9949058270833605, "3000–6000 Hz_40ms_modulation_mae_db": 0.20339302948473761, "mr_stft_db": 5.714857850143652, "band_mae_db": 0.22735032462784913, "rms_gain_error_db": -0.13861356730270794, "output_peak": 0.5075088979174839}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8243324621616751, "全帯域_10ms_rms_envelope_nrmse": 0.06683387680792782, "全帯域_10ms_envelope_correlation": 0.9945517389163, "全帯域_10ms_modulation_mae_db": 0.286342439351356, "全帯域_40ms_log_rms_mae_db": 0.4207252410337759, "全帯域_40ms_rms_envelope_nrmse": 0.034641031577823826, "全帯域_40ms_envelope_correlation": 0.9984771664843248, "全帯域_40ms_modulation_mae_db": 0.23866875804641097, "300–1000 Hz_10ms_log_rms_mae_db": 1.7052579012238216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1127250674245854, "300–1000 Hz_10ms_envelope_correlation": 0.9905443822287623, "300–1000 Hz_10ms_modulation_mae_db": 0.8052639608204201, "300–1000 Hz_40ms_log_rms_mae_db": 1.1413296224887994, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06638446437923426, "300–1000 Hz_40ms_envelope_correlation": 0.9969887283264711, "300–1000 Hz_40ms_modulation_mae_db": 1.2321423178980184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7968350995684716, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1744897214689701, "1000–3000 Hz_10ms_envelope_correlation": 0.9772312941693355, "1000–3000 Hz_10ms_modulation_mae_db": 0.49119704697278266, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0705599628046134, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11092211856803119, "1000–3000 Hz_40ms_envelope_correlation": 0.9901703811631473, "1000–3000 Hz_40ms_modulation_mae_db": 0.3702497835635706, "3000–6000 Hz_10ms_log_rms_mae_db": 2.061167394089315, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18961611536370065, "3000–6000 Hz_10ms_envelope_correlation": 0.9793530400219338, "3000–6000 Hz_10ms_modulation_mae_db": 0.5287300269419762, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4472112542688922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10721385479629678, "3000–6000 Hz_40ms_envelope_correlation": 0.9935462704327513, "3000–6000 Hz_40ms_modulation_mae_db": 0.3458781421040724, "mr_stft_db": 5.727644467936595, "band_mae_db": 0.10880698009946166, "rms_gain_error_db": -0.06283520448585148, "output_peak": 0.5552157356747882}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.999773902430508, "spectral_convergence": 0.21088459093840176, "spectral_convergence_status": "ok", "amplitude_floor": 1.711947899928738e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.453801383737769, "spectral_convergence": 0.21211621692691976, "spectral_convergence_status": "ok", "amplitude_floor": 1.4743066880059505e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7276347031441235, "spectral_convergence": 0.21401164816829615, "spectral_convergence_status": "ok", "amplitude_floor": 1.2829193968481007e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015568083457567275, "output_power": 0.0015603894112878698, "delta_db": 0.009978434510136816, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001958578248681761, "output_power": 0.0018470407089797016, "delta_db": -0.25464459623144287, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00020973665094298736, "output_power": 0.00021088341849167058, "delta_db": 0.023681040853659744, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018883580517861778, "output_power": 0.00016061608862849363, "delta_db": -0.7029529876735668, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010837478072549288, "output_power": 9.60660397782127e-05, "delta_db": -0.5235834404196443, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.95174204959434e-05, "output_power": 7.650741504508137e-05, "delta_db": -0.16758754618543753, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2804046743123147, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.162175821727561, "pitch": null, "output_rms_dbfs": -24.03237674970683, "output_peak": 0.5176862870743543, "reference_rms_dbfs": -23.870200927979273, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.727069996437467}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023872023988588299, "log_rms_mae_db": 0.8518566701071905, "rms_envelope_nrmse": 0.06592494178921919, "envelope_correlation": 0.9949907477594534, "rms_gain_error_db": -0.16137183882374287, "reference_mean_power": 0.004104010412713017, "output_mean_power": 0.003954315072322327}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.972155761177528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037154075128907306, "output_power": 0.0003598643727890605, "reference_active": true, "delta_db": -0.13867602088078}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.499306084417706e-05, "output_power": 5.007781048960383e-05, "reference_active": true, "delta_db": -0.4066256031432415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6530449727349195e-05, "output_power": 1.5472182965125415e-05, "reference_active": true, "delta_db": -0.28733076660385554}], "active_band_count": 3, "active_band_mae_db": 0.2775441302092923}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00021497588341920653, "log_rms_mae_db": 0.49101132800409064, "rms_envelope_nrmse": 0.036828152302997644, "envelope_correlation": 0.9986460767718954, "rms_gain_error_db": -0.16080057189488456, "reference_mean_power": 0.00411739985369924, "output_mean_power": 0.003967738006605828}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.973189563381595e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033993681960372955, "output_power": 0.0003288370270685816, "reference_active": true, "delta_db": -0.14417493697875908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4133967175911375e-05, "output_power": 2.2030360584538998e-05, "reference_active": true, "delta_db": -0.3960711214738565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.827774931478282e-07, "output_power": 7.640154044568874e-07, "reference_active": true, "delta_db": -0.6274913695214039}], "active_band_count": 3, "active_band_mae_db": 0.3892458093246732}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001937632671887944, "log_rms_mae_db": 1.7158501455456512, "rms_envelope_nrmse": 0.10020813071637892, "envelope_correlation": 0.9928576345041884, "rms_gain_error_db": -0.294149158080541, "reference_mean_power": 0.0014622039835396937, "output_mean_power": 0.0013664478036796224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.890315707731239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020584065582039213, "output_power": 0.00018992623904282024, "reference_active": true, "delta_db": -0.3494618855163533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2632804837837524e-05, "output_power": 3.470598535388417e-05, "reference_active": true, "delta_db": -0.8933952691044704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.198246611681044e-06, "output_power": 6.794288294877037e-06, "reference_active": true, "delta_db": -0.8157700731600758}], "active_band_count": 3, "active_band_mae_db": 0.6862090759269664}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001823230477151681, "log_rms_mae_db": 1.1178642166821882, "rms_envelope_nrmse": 0.06008920046118952, "envelope_correlation": 0.9978834370093633, "rms_gain_error_db": -0.29317536997694793, "reference_mean_power": 0.0014670926652510829, "output_mean_power": 0.0013713237853842767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5556250379109994e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020148633268320183, "output_power": 0.00018389650201112182, "reference_active": true, "delta_db": -0.39672123825809824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9312508746169674e-05, "output_power": 1.671556433342278e-05, "reference_active": true, "delta_db": -0.6271765007770291}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.933225179066708e-07, "output_power": 5.128958745271566e-07, "reference_active": true, "delta_db": -1.309060997189968}], "active_band_count": 3, "active_band_mae_db": 0.7776529120750318}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.475794683802334e-05, "log_rms_mae_db": 1.7771209729249409, "rms_envelope_nrmse": 0.1782262339639615, "envelope_correlation": 0.9786737331352349, "rms_gain_error_db": -0.6717661919907927, "reference_mean_power": 0.000205358190626383, "output_mean_power": 0.0001759281800430022}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.94981889493768e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.427037048631366e-05, "output_power": 4.850799206850649e-05, "reference_active": true, "delta_db": -0.4874948849101578}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.925199890017768e-06, "output_power": 6.788442182914725e-06, "reference_active": true, "delta_db": -1.188478279401744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.574401681634841e-06, "output_power": 1.9112526751385045e-06, "reference_active": true, "delta_db": -1.2935820414894341}], "active_band_count": 3, "active_band_mae_db": 0.9898517352671119}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.971385946015731e-05, "log_rms_mae_db": 1.0712086962193077, "rms_envelope_nrmse": 0.12510398623094318, "envelope_correlation": 0.9914187630027286, "rms_gain_error_db": -0.6719645254751726, "reference_mean_power": 0.00020637707638038015, "output_mean_power": 0.000176792974682051}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.280870922430589e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.267995669092476e-05, "output_power": 4.697344070923758e-05, "reference_active": true, "delta_db": -0.4979303671101063}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.180545251002319e-06, "output_power": 3.101282664461385e-06, "reference_active": true, "delta_db": -1.2969157713974964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0193067246169046e-07, "output_power": 1.2508188648644956e-07, "reference_active": true, "delta_db": -2.08007868894792}], "active_band_count": 3, "active_band_mae_db": 1.291641609151841}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.517166954195252e-05, "log_rms_mae_db": 2.0167669539055573, "rms_envelope_nrmse": 0.17030006419240015, "envelope_correlation": 0.9830884832355697, "rms_gain_error_db": -0.4358707473733827, "reference_mean_power": 0.00014718641611289376, "output_mean_power": 0.00013313144809131577}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4339203545791116e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.0687614509069e-05, "output_power": 6.401534702095528e-05, "reference_active": true, "delta_db": -0.4305922169614169}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.343067234178686e-06, "output_power": 5.267953314109098e-06, "reference_active": true, "delta_db": -2.488475566273054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3595191848035655e-06, "output_power": 2.1404503780384954e-06, "reference_active": true, "delta_db": -0.4231834880122596}], "active_band_count": 3, "active_band_mae_db": 1.1140837570822437}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.54521457203503e-05, "log_rms_mae_db": 1.454441458231002, "rms_envelope_nrmse": 0.10419812028618873, "envelope_correlation": 0.9945294366195109, "rms_gain_error_db": -0.4358744837874501, "reference_mean_power": 0.0001479201087162099, "output_mean_power": 0.0001337949645949322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3826641859038576e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.726675521598142e-05, "output_power": 6.038695312749739e-05, "reference_active": true, "delta_db": -0.46857361400530184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.27086384707959e-06, "output_power": 2.2458537388545386e-06, "reference_active": true, "delta_db": -2.791342571314975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6847652723937028e-07, "output_power": 1.0790419016764708e-07, "reference_active": true, "delta_db": -1.9350109227713999}], "active_band_count": 3, "active_band_mae_db": 1.7316423693638923}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8963480036579521, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1e300dca7f3106b80dd59a002085efb801d6b52b64e556c3b1869a75ccd31027", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9785868710312696, "standard_deviation": 0.0012499588224782264, "minimum": 0.9767060123872323, "maximum": 0.9804474817628812, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7992206909185775, "standard_deviation": 0.03049928314998723, "minimum": 1.7733614476017017, "maximum": 1.8640994653899863, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7493852109818859, "standard_deviation": 0.3020623211408782, "minimum": 0.26951324278941713, "maximum": 1.2064497908174863, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1727477128468442, "standard_deviation": 0.006277941201664744, "minimum": 0.1628518383242331, "maximum": 0.18224198820737403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.992173240577426, "standard_deviation": 0.0011701033457722923, "minimum": 0.9901703811631473, "maximum": 0.9936218099148004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0924610593192183, "standard_deviation": 0.020999953438057453, "minimum": 1.0705599628046134, "maximum": 1.1252461864195427, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9119148050891818, "standard_deviation": 0.44062229944815806, "minimum": 0.3702497835635706, "maximum": 1.6664632340284184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1105391488180123, "standard_deviation": 0.009240023214363766, "minimum": 0.09716095420988263, "maximum": 0.12510398623094318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.98321811816335, "standard_deviation": 0.004520623064100003, "minimum": 0.9737634504421451, "maximum": 0.9869546496432448, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0092705402754687, "standard_deviation": 0.025142280345238906, "minimum": 1.978119292100297, "maximum": 2.061167394089315, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.766284214225079, "standard_deviation": 0.49928634475976397, "minimum": 0.04612746585419955, "maximum": 1.5727683483613795, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16663501079415244, "standard_deviation": 0.02179645624432337, "minimum": 0.1469423986694379, "maximum": 0.20850651046779228, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943365722192282, "standard_deviation": 0.0012072311779239115, "minimum": 0.991781971580915, "maximum": 0.9958053196942039, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4250080800887501, "standard_deviation": 0.02797165379278668, "minimum": 1.3761214235196906, "maximum": 1.454441458231002, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.842434369370687, "standard_deviation": 0.6548272757745618, "minimum": 0.20339302948473761, "maximum": 1.777156453746671, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09817774279767738, "standard_deviation": 0.009821255203181306, "minimum": 0.0854186236766497, "maximum": 0.1150877222970187, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992677505783973, "standard_deviation": 0.0010115139439853499, "minimum": 0.9905443822287623, "maximum": 0.9939105176609356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7109004825504779, "standard_deviation": 0.031633963937859985, "minimum": 1.6564651211404555, "maximum": 1.7573795662125262, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.801622333099828, "standard_deviation": 0.1517417281644698, "minimum": 0.5640812980556799, "maximum": 1.020416398309923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09987647840412019, "standard_deviation": 0.006143496647133936, "minimum": 0.09188128314405748, "maximum": 0.1127250674245854, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975548475833196, "standard_deviation": 0.00032560192197719117, "minimum": 0.9969887283264711, "maximum": 0.9980557512606361, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1421526387549488, "standard_deviation": 0.022806801017609898, "minimum": 1.1154258432312858, "maximum": 1.1859854814215915, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1291482889814586, "standard_deviation": 0.257388925436434, "minimum": 0.7776529120750318, "maximum": 1.4200381860356555, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060742289525702334, "standard_deviation": 0.0035867898872964533, "minimum": 0.05559132186231039, "maximum": 0.06638446437923426, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2084498828645825, "standard_deviation": 0.05068002642523069, "minimum": 0.10880698009946166, "maximum": 0.2804046743123147, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.717332909497186, "standard_deviation": 0.0087763436274341, "minimum": 5.7025928635967595, "maximum": 5.727644467936595, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5215103301615891, "standard_deviation": 0.03087440076146238, "minimum": 0.4722509973817107, "maximum": 0.5588181957145346, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10615947111988372, "standard_deviation": 0.03656648157391798, "minimum": -0.162175821727561, "maximum": -0.06283520448585148, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9953946652746379, "standard_deviation": 0.0005795133678487792, "minimum": 0.9945517389163, "maximum": 0.9960938530100799, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8157621759937019, "standard_deviation": 0.02182049794191487, "minimum": 0.7847789347706304, "maximum": 0.8518566701071905, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3080653001281484, "standard_deviation": 0.11433754459916262, "minimum": 0.18951607242905047, "maximum": 0.5204267466819144, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06214812112619239, "standard_deviation": 0.0035740982397177, "minimum": 0.057962160120314105, "maximum": 0.06683387680792782, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998603608360983, "standard_deviation": 0.00012905061454444267, "minimum": 0.9984108774778213, "maximum": 0.9988402691596323, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.43846308893444164, "standard_deviation": 0.026894353706155533, "minimum": 0.4137499335179461, "maximum": 0.49101132800409064, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33684161050961736, "standard_deviation": 0.10295596261605723, "minimum": 0.17667884109316834, "maximum": 0.5191761989573073, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03482054629801054, "standard_deviation": 0.0015035841538989369, "minimum": 0.03311306753489981, "maximum": 0.036828152302997644, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0028", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.2671265602111816, "predicted_mos": 3.2671265602111816, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5c6901bc2873d21b2740a499e5396edf0d7d81f5d8343fdeb947c4b64996242a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.279204899910837, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8126591344417411, "全帯域_10ms_rms_envelope_nrmse": 0.05666277615420221, "全帯域_10ms_envelope_correlation": 0.9961497050628954, "全帯域_10ms_modulation_mae_db": 0.31349863447750526, "全帯域_40ms_log_rms_mae_db": 0.424404042383378, "全帯域_40ms_rms_envelope_nrmse": 0.029392789678717196, "全帯域_40ms_envelope_correlation": 0.9989869586397003, "全帯域_40ms_modulation_mae_db": 0.27863706700421137, "300–1000 Hz_10ms_log_rms_mae_db": 1.707453215402983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0954073719870867, "300–1000 Hz_10ms_envelope_correlation": 0.9931898094678159, "300–1000 Hz_10ms_modulation_mae_db": 0.5729221116220942, "300–1000 Hz_40ms_log_rms_mae_db": 1.0787810014461567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05467530525491525, "300–1000 Hz_40ms_envelope_correlation": 0.9979075953982184, "300–1000 Hz_40ms_modulation_mae_db": 0.6082546870144081, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7439136116831357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17123343337662333, "1000–3000 Hz_10ms_envelope_correlation": 0.977881491801865, "1000–3000 Hz_10ms_modulation_mae_db": 0.7999111637093591, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0328860358019463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09096894302198218, "1000–3000 Hz_40ms_envelope_correlation": 0.9941187147116402, "1000–3000 Hz_40ms_modulation_mae_db": 0.9380162190953666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9440210603306274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.142414394944891, "3000–6000 Hz_10ms_envelope_correlation": 0.9877712629557569, "3000–6000 Hz_10ms_modulation_mae_db": 0.29525075098496223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3889901983797273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08025240426767875, "3000–6000 Hz_40ms_envelope_correlation": 0.996010218221362, "3000–6000 Hz_40ms_modulation_mae_db": 0.4038254513346146, "mr_stft_db": 5.698277337323296, "band_mae_db": 0.11294600843518306, "rms_gain_error_db": -0.0885790831403209, "output_peak": 0.5480335328186979}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8271721387374457, "全帯域_10ms_rms_envelope_nrmse": 0.06108086223303781, "全帯域_10ms_envelope_correlation": 0.9955519170186301, "全帯域_10ms_modulation_mae_db": 0.4368513594397411, "全帯域_40ms_log_rms_mae_db": 0.4549780528115946, "全帯域_40ms_rms_envelope_nrmse": 0.03400985980165617, "全帯域_40ms_envelope_correlation": 0.9986750202840006, "全帯域_40ms_modulation_mae_db": 0.42356733381240524, "300–1000 Hz_10ms_log_rms_mae_db": 1.7147100085396503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10300067964521088, "300–1000 Hz_10ms_envelope_correlation": 0.9922570231999344, "300–1000 Hz_10ms_modulation_mae_db": 0.807609826667128, "300–1000 Hz_40ms_log_rms_mae_db": 1.1959028081552698, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061510312282380034, "300–1000 Hz_40ms_envelope_correlation": 0.9975632819660453, "300–1000 Hz_40ms_modulation_mae_db": 1.0551345422385678, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8796114038406027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20180887009579296, "1000–3000 Hz_10ms_envelope_correlation": 0.9703375991488832, "1000–3000 Hz_10ms_modulation_mae_db": 0.7505639657708013, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1565798543177936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1231374835855916, "1000–3000 Hz_40ms_envelope_correlation": 0.9907404046625897, "1000–3000 Hz_40ms_modulation_mae_db": 0.5535723025929403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9784442154774688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15181492781668507, "3000–6000 Hz_10ms_envelope_correlation": 0.9862050345591329, "3000–6000 Hz_10ms_modulation_mae_db": 0.6329039500247912, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3631849663292899, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08692214816945365, "3000–6000 Hz_40ms_envelope_correlation": 0.9953682101485885, "3000–6000 Hz_40ms_modulation_mae_db": 0.5331011822402902, "mr_stft_db": 5.687275876964253, "band_mae_db": 0.16278211327117856, "rms_gain_error_db": -0.10623462734742065, "output_peak": 0.5341759990532762}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8524244349260509, "全帯域_10ms_rms_envelope_nrmse": 0.07245741364462045, "全帯域_10ms_envelope_correlation": 0.9936160961051144, "全帯域_10ms_modulation_mae_db": 0.21686297154858955, "全帯域_40ms_log_rms_mae_db": 0.47221791609009967, "全帯域_40ms_rms_envelope_nrmse": 0.0459331254309983, "全帯域_40ms_envelope_correlation": 0.9973074554080372, "全帯域_40ms_modulation_mae_db": 0.10936118219353802, "300–1000 Hz_10ms_log_rms_mae_db": 1.6646308208854836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10937081590744292, "300–1000 Hz_10ms_envelope_correlation": 0.9908583461827123, "300–1000 Hz_10ms_modulation_mae_db": 0.46030114850170434, "300–1000 Hz_40ms_log_rms_mae_db": 1.0665694716740994, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07503897799417432, "300–1000 Hz_40ms_envelope_correlation": 0.9955887341701877, "300–1000 Hz_40ms_modulation_mae_db": 0.5056036317532983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.748714966282962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16004404668606945, "1000–3000 Hz_10ms_envelope_correlation": 0.981310972291983, "1000–3000 Hz_10ms_modulation_mae_db": 0.871664755769773, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0272368678553137, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10259372634411955, "1000–3000 Hz_40ms_envelope_correlation": 0.9928683464187824, "1000–3000 Hz_40ms_modulation_mae_db": 0.9159856835638899, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9315469966792418, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1448602058529713, "3000–6000 Hz_10ms_envelope_correlation": 0.9874833816919962, "3000–6000 Hz_10ms_modulation_mae_db": 0.39546200008543414, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3380102544433483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0894739748844538, "3000–6000 Hz_40ms_envelope_correlation": 0.9954806555113429, "3000–6000 Hz_40ms_modulation_mae_db": 0.4235992802144939, "mr_stft_db": 5.694740029931236, "band_mae_db": 0.2275815323742166, "rms_gain_error_db": -0.08326760239978745, "output_peak": 0.48918021796061517}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8038140277947631, "全帯域_10ms_rms_envelope_nrmse": 0.06431657674396819, "全帯域_10ms_envelope_correlation": 0.9950851051624638, "全帯域_10ms_modulation_mae_db": 0.08024364333088785, "全帯域_40ms_log_rms_mae_db": 0.42510546886562983, "全帯域_40ms_rms_envelope_nrmse": 0.03251236247559786, "全帯域_40ms_envelope_correlation": 0.9988415664283817, "全帯域_40ms_modulation_mae_db": 0.1040076151372376, "300–1000 Hz_10ms_log_rms_mae_db": 1.701757431311826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10457258394797847, "300–1000 Hz_10ms_envelope_correlation": 0.9918961170920787, "300–1000 Hz_10ms_modulation_mae_db": 0.5019359561593185, "300–1000 Hz_40ms_log_rms_mae_db": 1.0771534280532291, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05830402652810692, "300–1000 Hz_40ms_envelope_correlation": 0.9977558203217256, "300–1000 Hz_40ms_modulation_mae_db": 0.7964415585957321, "1000–3000 Hz_10ms_log_rms_mae_db": 1.735947382534066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1601392185053702, "1000–3000 Hz_10ms_envelope_correlation": 0.9821780294949147, "1000–3000 Hz_10ms_modulation_mae_db": 0.44780909736332114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0278572619673882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10285873793205844, "1000–3000 Hz_40ms_envelope_correlation": 0.9939766699603164, "1000–3000 Hz_40ms_modulation_mae_db": 0.5762582128645632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9720552023832973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.153970602150309, "3000–6000 Hz_10ms_envelope_correlation": 0.985560469930694, "3000–6000 Hz_10ms_modulation_mae_db": 0.527133256796568, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4074302084994954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08683109390765152, "3000–6000 Hz_40ms_envelope_correlation": 0.9953395219544163, "3000–6000 Hz_40ms_modulation_mae_db": 0.8579647757119245, "mr_stft_db": 5.684564993769301, "band_mae_db": 0.2142742579550407, "rms_gain_error_db": -0.12021851444620585, "output_peak": 0.4617640625861066}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8163859888732503, "全帯域_10ms_rms_envelope_nrmse": 0.057877942473123854, "全帯域_10ms_envelope_correlation": 0.9960125578522838, "全帯域_10ms_modulation_mae_db": 0.09160916576788558, "全帯域_40ms_log_rms_mae_db": 0.41696536252489946, "全帯域_40ms_rms_envelope_nrmse": 0.03173912395097857, "全帯域_40ms_envelope_correlation": 0.9988417256711906, "全帯域_40ms_modulation_mae_db": 0.16581863154651535, "300–1000 Hz_10ms_log_rms_mae_db": 1.6613542648236974, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09695015428318386, "300–1000 Hz_10ms_envelope_correlation": 0.9930818144468866, "300–1000 Hz_10ms_modulation_mae_db": 0.3768513145731416, "300–1000 Hz_40ms_log_rms_mae_db": 1.1108612135340774, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058502891654576096, "300–1000 Hz_40ms_envelope_correlation": 0.9976933646480113, "300–1000 Hz_40ms_modulation_mae_db": 0.5002254762580808, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8145898832962006, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16965477058690012, "1000–3000 Hz_10ms_envelope_correlation": 0.9782321140472415, "1000–3000 Hz_10ms_modulation_mae_db": 0.476215801459436, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1325492561354893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10726997280827646, "1000–3000 Hz_40ms_envelope_correlation": 0.9912451069617452, "1000–3000 Hz_40ms_modulation_mae_db": 0.5745924088555764, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9576978617371807, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1541347265773417, "3000–6000 Hz_10ms_envelope_correlation": 0.985725958177637, "3000–6000 Hz_10ms_modulation_mae_db": 0.7171699124422494, "3000–6000 Hz_40ms_log_rms_mae_db": 1.369259695632379, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08832710544390955, "3000–6000 Hz_40ms_envelope_correlation": 0.9955798683486838, "3000–6000 Hz_40ms_modulation_mae_db": 1.139652126359971, "mr_stft_db": 5.6976222371688054, "band_mae_db": 0.16784097347712243, "rms_gain_error_db": -0.10349859796464493, "output_peak": 0.5379808183178797}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8611354849458196, "全帯域_10ms_rms_envelope_nrmse": 0.06417522630225789, "全帯域_10ms_envelope_correlation": 0.995071657369363, "全帯域_10ms_modulation_mae_db": 0.21980103929452977, "全帯域_40ms_log_rms_mae_db": 0.47611651780252345, "全帯域_40ms_rms_envelope_nrmse": 0.03430394397353758, "全帯域_40ms_envelope_correlation": 0.9986373651221715, "全帯域_40ms_modulation_mae_db": 0.36801907314612065, "300–1000 Hz_10ms_log_rms_mae_db": 1.637984855301785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10280746991563985, "300–1000 Hz_10ms_envelope_correlation": 0.9925040383628078, "300–1000 Hz_10ms_modulation_mae_db": 0.7674649971297738, "300–1000 Hz_40ms_log_rms_mae_db": 1.1036129624240425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06432116720002055, "300–1000 Hz_40ms_envelope_correlation": 0.9975172539690317, "300–1000 Hz_40ms_modulation_mae_db": 1.1132909940301323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8002753657178456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1702032649944522, "1000–3000 Hz_10ms_envelope_correlation": 0.9788178780383976, "1000–3000 Hz_10ms_modulation_mae_db": 0.5208513713793257, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1110690289734655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10632344142553807, "1000–3000 Hz_40ms_envelope_correlation": 0.9924557235308344, "1000–3000 Hz_40ms_modulation_mae_db": 0.4531836579310827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9849326033393033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1630466832612575, "3000–6000 Hz_10ms_envelope_correlation": 0.983876370212968, "3000–6000 Hz_10ms_modulation_mae_db": 0.6363829801930564, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3885876328559086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10107209175331933, "3000–6000 Hz_40ms_envelope_correlation": 0.9936340729640089, "3000–6000 Hz_40ms_modulation_mae_db": 0.657033009901412, "mr_stft_db": 5.714203088032299, "band_mae_db": 0.15025044048191252, "rms_gain_error_db": -0.10941279547824362, "output_peak": 0.4988002237835352}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8340867980616419, "全帯域_10ms_rms_envelope_nrmse": 0.06231192759059052, "全帯域_10ms_envelope_correlation": 0.9952901681391463, "全帯域_10ms_modulation_mae_db": 0.20599430186585913, "全帯域_40ms_log_rms_mae_db": 0.441915301177636, "全帯域_40ms_rms_envelope_nrmse": 0.03336240757902731, "全帯域_40ms_envelope_correlation": 0.9986413377584414, "全帯域_40ms_modulation_mae_db": 0.20900355669336315, "300–1000 Hz_10ms_log_rms_mae_db": 1.6429701510882668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10387643819366178, "300–1000 Hz_10ms_envelope_correlation": 0.9920928060578852, "300–1000 Hz_10ms_modulation_mae_db": 0.8108197519526209, "300–1000 Hz_40ms_log_rms_mae_db": 1.078801024266995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06208064703271154, "300–1000 Hz_40ms_envelope_correlation": 0.9974740827106077, "300–1000 Hz_40ms_modulation_mae_db": 0.9657954419634261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7923923530472952, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17704413764790378, "1000–3000 Hz_10ms_envelope_correlation": 0.9772847649336778, "1000–3000 Hz_10ms_modulation_mae_db": 0.8006376050961954, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0456462768245967, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11014153373737842, "1000–3000 Hz_40ms_envelope_correlation": 0.9923374909928674, "1000–3000 Hz_40ms_modulation_mae_db": 0.777878657227272, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9115591918237071, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1569203332265365, "3000–6000 Hz_10ms_envelope_correlation": 0.9861425983635893, "3000–6000 Hz_10ms_modulation_mae_db": 0.4453711137303486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3304194838394467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08902944066402334, "3000–6000 Hz_40ms_envelope_correlation": 0.995812539985034, "3000–6000 Hz_40ms_modulation_mae_db": 0.3193554922273756, "mr_stft_db": 5.685759385437219, "band_mae_db": 0.21367799938937768, "rms_gain_error_db": -0.07596460080541231, "output_peak": 0.466964990567165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8163957533168873, "全帯域_10ms_rms_envelope_nrmse": 0.060093460280277136, "全帯域_10ms_envelope_correlation": 0.9956215659752169, "全帯域_10ms_modulation_mae_db": 0.3172087289365431, "全帯域_40ms_log_rms_mae_db": 0.4319778086399639, "全帯域_40ms_rms_envelope_nrmse": 0.030646688147277883, "全帯域_40ms_envelope_correlation": 0.9988547705134962, "全帯域_40ms_modulation_mae_db": 0.3123466170613414, "300–1000 Hz_10ms_log_rms_mae_db": 1.630987659393569, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09807524579420397, "300–1000 Hz_10ms_envelope_correlation": 0.9927724252740233, "300–1000 Hz_10ms_modulation_mae_db": 0.5307754801932425, "300–1000 Hz_40ms_log_rms_mae_db": 1.0521897410155494, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05880284625987657, "300–1000 Hz_40ms_envelope_correlation": 0.9974999068629467, "300–1000 Hz_40ms_modulation_mae_db": 0.6133811783159524, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7259997510619536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1768256048643787, "1000–3000 Hz_10ms_envelope_correlation": 0.9786800394186786, "1000–3000 Hz_10ms_modulation_mae_db": 0.40678352448012833, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0887787408066467, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1217109452935291, "1000–3000 Hz_40ms_envelope_correlation": 0.991708148112196, "1000–3000 Hz_40ms_modulation_mae_db": 0.3756274996268361, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9551555624010506, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15461739181851142, "3000–6000 Hz_10ms_envelope_correlation": 0.9854816712749285, "3000–6000 Hz_10ms_modulation_mae_db": 0.5599546002631469, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3120942865233072, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07671715612459779, "3000–6000 Hz_40ms_envelope_correlation": 0.9963344285249701, "3000–6000 Hz_40ms_modulation_mae_db": 0.9635756801935002, "mr_stft_db": 5.688067652707159, "band_mae_db": 0.2299961617638023, "rms_gain_error_db": -0.0765389366995318, "output_peak": 0.4930792250228169}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.989199952054278, "spectral_convergence": 0.2037472195624694, "spectral_convergence_status": "ok", "amplitude_floor": 1.711947899928738e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4066880954629255, "spectral_convergence": 0.2049850130228095, "spectral_convergence_status": "ok", "amplitude_floor": 1.4743066880059505e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.698943964452684, "spectral_convergence": 0.21412432729601322, "spectral_convergence_status": "ok", "amplitude_floor": 1.2829193968481007e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015568083457567275, "output_power": 0.0015605492801714443, "delta_db": 0.010423365865021609, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001958578248681761, "output_power": 0.0018876725160906637, "delta_db": -0.16014274388160607, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00020973665094298736, "output_power": 0.00020754953115648524, "delta_db": -0.0455257227915173, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018883580517861778, "output_power": 0.0001744584588080472, "delta_db": -0.3439231286153992, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010837478072549288, "output_power": 0.00010765134896221969, "delta_db": -0.02908755508036263, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.95174204959434e-05, "output_power": 8.115581374712511e-05, "delta_db": 0.08857353437719163, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11294600843518306, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0885790831403209, "pitch": null, "output_rms_dbfs": -23.958780011119593, "output_peak": 0.5480335328186979, "reference_rms_dbfs": -23.870200927979273, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.698277337323296}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023872023988588299, "log_rms_mae_db": 0.8126591344417411, "rms_envelope_nrmse": 0.05666277615420221, "envelope_correlation": 0.9961497050628954, "rms_gain_error_db": -0.08835229279211627, "reference_mean_power": 0.004104010412713017, "output_mean_power": 0.00402136250920592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.972155761177528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037154075128907306, "output_power": 0.0003564819531096394, "reference_active": true, "delta_db": -0.1796890634886235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.499306084417706e-05, "output_power": 5.00423250037802e-05, "reference_active": true, "delta_db": -0.4097041349990429}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6530449727349195e-05, "output_power": 1.5246648871086522e-05, "reference_active": true, "delta_db": -0.3511027049448494}], "active_band_count": 3, "active_band_mae_db": 0.31349863447750526}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00021497588341920653, "log_rms_mae_db": 0.424404042383378, "rms_envelope_nrmse": 0.029392789678717196, "envelope_correlation": 0.9989869586397003, "rms_gain_error_db": -0.08811345839818704, "reference_mean_power": 0.00411739985369924, "output_mean_power": 0.004034704186266871}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.973189563381595e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033993681960372955, "output_power": 0.00032516575126237293, "reference_active": true, "delta_db": -0.19293410513571435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4133967175911375e-05, "output_power": 2.1900615055435094e-05, "reference_active": true, "delta_db": -0.42172406013655456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.827774931478282e-07, "output_power": 8.389304317915522e-07, "reference_active": true, "delta_db": -0.22125303574036528}], "active_band_count": 3, "active_band_mae_db": 0.27863706700421137}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001937632671887944, "log_rms_mae_db": 1.707453215402983, "rms_envelope_nrmse": 0.0954073719870867, "envelope_correlation": 0.9931898094678159, "rms_gain_error_db": -0.19389329350217333, "reference_mean_power": 0.0014622039835396937, "output_mean_power": 0.0013983588437022365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.890315707731239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020584065582039213, "output_power": 0.00018576637126621636, "reference_active": true, "delta_db": -0.4456405915792793}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2632804837837524e-05, "output_power": 3.6063638959160725e-05, "reference_active": true, "delta_db": -0.726743595095034}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.198246611681044e-06, "output_power": 7.229075547734088e-06, "reference_active": true, "delta_db": -0.5463821481919693}], "active_band_count": 3, "active_band_mae_db": 0.5729221116220942}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001823230477151681, "log_rms_mae_db": 1.0787810014461567, "rms_envelope_nrmse": 0.05467530525491525, "envelope_correlation": 0.9979075953982184, "rms_gain_error_db": -0.1936339238912997, "reference_mean_power": 0.0014670926652510829, "output_mean_power": 0.0014031178623906322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5556250379109994e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020148633268320183, "output_power": 0.00018080363914469273, "reference_active": true, "delta_db": -0.4703842467005976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9312508746169674e-05, "output_power": 1.6920739555456104e-05, "reference_active": true, "delta_db": -0.5741935260593121}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.933225179066708e-07, "output_power": 5.793175385231586e-07, "reference_active": true, "delta_db": -0.7801862882833146}], "active_band_count": 3, "active_band_mae_db": 0.6082546870144081}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.475794683802334e-05, "log_rms_mae_db": 1.7439136116831357, "rms_envelope_nrmse": 0.17123343337662333, "envelope_correlation": 0.977881491801865, "rms_gain_error_db": -0.32364977250683447, "reference_mean_power": 0.000205358190626383, "output_mean_power": 0.00019061060154520446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.94981889493768e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.427037048631366e-05, "output_power": 4.782371254920082e-05, "reference_active": true, "delta_db": -0.5491949905976085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.925199890017768e-06, "output_power": 7.366422953818829e-06, "reference_active": true, "delta_db": -0.8336130053271479}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.574401681634841e-06, "output_power": 2.0369659103392176e-06, "reference_active": true, "delta_db": -1.0169254952033209}], "active_band_count": 3, "active_band_mae_db": 0.7999111637093591}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.971385946015731e-05, "log_rms_mae_db": 1.0328860358019463, "rms_envelope_nrmse": 0.09096894302198218, "envelope_correlation": 0.9941187147116402, "rms_gain_error_db": -0.3236672060763768, "reference_mean_power": 0.00020637707638038015, "output_mean_power": 0.00019155554810847432}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.280870922430589e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.267995669092476e-05, "output_power": 4.6321818375002496e-05, "reference_active": true, "delta_db": -0.5585980985251189}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.180545251002319e-06, "output_power": 3.542665284601004e-06, "reference_active": true, "delta_db": -0.7190280708746906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0193067246169046e-07, "output_power": 1.4176206890106984e-07, "reference_active": true, "delta_db": -1.5364224878862907}], "active_band_count": 3, "active_band_mae_db": 0.9380162190953666}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.517166954195252e-05, "log_rms_mae_db": 1.9440210603306274, "rms_envelope_nrmse": 0.142414394944891, "envelope_correlation": 0.9877712629557569, "rms_gain_error_db": 0.027135149023979593, "reference_mean_power": 0.00014718641611289376, "output_mean_power": 0.00014810893038599504}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4339203545791116e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.0687614509069e-05, "output_power": 7.055765444775714e-05, "reference_active": true, "delta_db": -0.007991907104543644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.343067234178686e-06, "output_power": 8.575450300376993e-06, "reference_active": true, "delta_db": -0.3723253981702038}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3595191848035655e-06, "output_power": 2.1002936439939025e-06, "reference_active": true, "delta_db": -0.5054349476801392}], "active_band_count": 3, "active_band_mae_db": 0.29525075098496223}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.54521457203503e-05, "log_rms_mae_db": 1.3889901983797273, "rms_envelope_nrmse": 0.08025240426767875, "envelope_correlation": 0.996010218221362, "rms_gain_error_db": 0.02715900431218081, "reference_mean_power": 0.0001479201087162099, "output_mean_power": 0.00014884803912702997}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3826641859038576e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.726675521598142e-05, "output_power": 6.740560653797808e-05, "reference_active": true, "delta_db": 0.008955420923845976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.27086384707959e-06, "output_power": 3.825969121524127e-06, "reference_active": true, "delta_db": -0.47774265945879846}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6847652723937028e-07, "output_power": 1.4258097315818807e-07, "reference_active": true, "delta_db": -0.7247782736211993}], "active_band_count": 3, "active_band_mae_db": 0.4038254513346146}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8933387147270149, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "aa4f7eb059d359440ec017b5eea2e0f1777c85d3d48e8d1345c177948a3276d3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9780903611469551, "standard_deviation": 0.003561227599488676, "minimum": 0.9703375991488832, "maximum": 0.9821780294949147, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7801805896830076, "standard_deviation": 0.05185272071235626, "minimum": 1.7259997510619536, "maximum": 1.8796114038406027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6343046606285425, "standard_deviation": 0.1887395793991151, "minimum": 0.40678352448012833, "maximum": 0.871664755769773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17336916834468635, "standard_deviation": 0.01317595476092606, "minimum": 0.16004404668606945, "maximum": 0.20180887009579296, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9924313256688715, "standard_deviation": 0.0012086044223341438, "minimum": 0.9907404046625897, "maximum": 0.9941187147116402, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.07782541533533, "standard_deviation": 0.05145474554436087, "minimum": 1.0272368678553137, "maximum": 1.1565798543177936, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6456393302196909, "standard_deviation": 0.20851240183585443, "minimum": 0.3756274996268361, "maximum": 0.9380162190953666, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10812559801855923, "standard_deviation": 0.010492544690892764, "minimum": 0.09096894302198218, "maximum": 0.1231374835855916, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9860308433958378, "standard_deviation": 0.0012212002519547754, "minimum": 0.983876370212968, "maximum": 0.9877712629557569, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9544265867714847, "standard_deviation": 0.024781411357716315, "minimum": 1.9115591918237071, "maximum": 1.9849326033393033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5262035705650696, "standard_deviation": 0.14045795636130604, "minimum": 0.29525075098496223, "maximum": 0.7171699124422494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15272240820606295, "standard_deviation": 0.006549260677785863, "minimum": 0.142414394944891, "maximum": 0.1630466832612575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954449394573008, "standard_deviation": 0.0008072510820322036, "minimum": 0.9936340729640089, "maximum": 0.9963344285249701, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3622470908128628, "standard_deviation": 0.03298552258002373, "minimum": 1.3120942865233072, "maximum": 1.4074302084994954, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6622633747729478, "standard_deviation": 0.29641547067300367, "minimum": 0.3193554922273756, "maximum": 1.139652126359971, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08732817690188596, "standard_deviation": 0.007175992436964766, "minimum": 0.07671715612459779, "maximum": 0.10107209175331933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992331547510518, "standard_deviation": 0.0007508198866433292, "minimum": 0.9908583461827123, "maximum": 0.9931898094678159, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6702310508434077, "standard_deviation": 0.033361563294474775, "minimum": 1.630987659393569, "maximum": 1.7147100085396503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.603585073349878, "standard_deviation": 0.1690060941386223, "minimum": 0.3768513145731416, "maximum": 0.8108197519526209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10175759495930106, "standard_deviation": 0.00462939062402393, "minimum": 0.0954073719870867, "maximum": 0.10937081590744292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973750050058467, "standard_deviation": 0.0007369715092085862, "minimum": 0.9955887341701877, "maximum": 0.9979075953982184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0954839563211776, "standard_deviation": 0.04470576868287835, "minimum": 1.0521897410155494, "maximum": 1.1959028081552698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7697659387711997, "standard_deviation": 0.2482969769615984, "minimum": 0.5002254762580808, "maximum": 1.1132909940301323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.061654521775845156, "standard_deviation": 0.00614651065369605, "minimum": 0.05467530525491525, "maximum": 0.07503897799417432, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18491868589347923, "standard_deviation": 0.04260210369347328, "minimum": 0.11294600843518306, "maximum": 0.2299961617638023, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.693813825166696, "standard_deviation": 0.009835685642628428, "minimum": 5.684564993769301, "maximum": 5.714203088032299, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5037473837637616, "standard_deviation": 0.032787457742749056, "minimum": 0.4617640625861066, "maximum": 0.5480335328186979, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09546434478519594, "standard_deviation": 0.016572157554958302, "minimum": -0.12021851444620585, "maximum": -0.07596460080541231, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952998465856392, "standard_deviation": 0.0007865433599918649, "minimum": 0.9936160961051144, "maximum": 0.9961497050628954, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8280092201372, "standard_deviation": 0.020078753418908125, "minimum": 0.8038140277947631, "maximum": 0.8611354849458196, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2352587305826927, "standard_deviation": 0.11926890296831237, "minimum": 0.08024364333088785, "maximum": 0.4368513594397411, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.062372023177759764, "standard_deviation": 0.004904848196972879, "minimum": 0.05666277615420221, "maximum": 0.07245741364462045, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985982749781774, "standard_deviation": 0.0005359652660745839, "minimum": 0.9973074554080372, "maximum": 0.9989869586397003, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44296005878696565, "standard_deviation": 0.02255261424610987, "minimum": 0.41696536252489946, "maximum": 0.47611651780252345, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24634513457434162, "standard_deviation": 0.11863190584973812, "minimum": 0.1040076151372376, "maximum": 0.42356733381240524, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033987537629723855, "standard_deviation": 0.005108717431379495, "minimum": 0.029392789678717196, "maximum": 0.0459331254309983, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2300-131720-0041", "language": "en", "style": "normal", "speaker": "librispeech_2300", "source_utterance_id": "2300-131720-0041", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 60000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2300-131720-0041.wav", "source_file_sha256": "7419afc0b0a3875c8e60b61a5038de3939d12404eae1de343dea2db9867e20ee", "decoded_pcm_sha256": "976092b4e8947717eb460c0f8df2d08babdc9b088a74a4efec5fca50e34afea2", "original_source_sha256": "4ede50e44cbe5dfb258311f80fd68b7efe2e080265efa5e6bc07aea531a22e28", "corpus_entry_id": "librispeech_2300-131720-0041"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.769171953201294, "predicted_mos": 3.769171953201294, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "38e132aca81796256d7fc76b4e832530b7cd5e1090a913b11748397e1dfb1333", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.881456294999225, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2300-131720-0041", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "38e132aca81796256d7fc76b4e832530b7cd5e1090a913b11748397e1dfb1333", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "1fb0d0ccb72963d69461dc23b437371942e1244437c77215b6a1e1be34891b3d"}, "rows": [{"case": "librispeech_2300-131720-0041", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.22967529296875, "predicted_mos": 2.22967529296875, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "ba147ce875b25dfd549d4aa1beebfdaae255b4bb3e71908056da4012c1e8089e", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.325816111999302, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2300-131720-0041", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ba147ce875b25dfd549d4aa1beebfdaae255b4bb3e71908056da4012c1e8089e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6233909026513077, "全帯域_10ms_rms_envelope_nrmse": 0.061818950763648316, "全帯域_10ms_envelope_correlation": 0.9956977794627472, "全帯域_10ms_modulation_mae_db": 0.17679778566024296, "全帯域_40ms_log_rms_mae_db": 0.3427053919417756, "全帯域_40ms_rms_envelope_nrmse": 0.034443376613536414, "全帯域_40ms_envelope_correlation": 0.9986217393483501, "全帯域_40ms_modulation_mae_db": 0.12181485163300075, "300–1000 Hz_10ms_log_rms_mae_db": 1.3799903784499539, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10783996031132792, "300–1000 Hz_10ms_envelope_correlation": 0.9905736258789739, "300–1000 Hz_10ms_modulation_mae_db": 0.7721006506275699, "300–1000 Hz_40ms_log_rms_mae_db": 0.9741147482765135, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07042381726240066, "300–1000 Hz_40ms_envelope_correlation": 0.9959121667358857, "300–1000 Hz_40ms_modulation_mae_db": 1.102406669187181, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4366555320580459, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1377762602285102, "1000–3000 Hz_10ms_envelope_correlation": 0.9874859272648188, "1000–3000 Hz_10ms_modulation_mae_db": 0.8341654472181427, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9332381264496951, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08468459649818919, "1000–3000 Hz_40ms_envelope_correlation": 0.9955468717877457, "1000–3000 Hz_40ms_modulation_mae_db": 0.889743605448459, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4999052218659679, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21818843197742957, "3000–6000 Hz_10ms_envelope_correlation": 0.9681286996106039, "3000–6000 Hz_10ms_modulation_mae_db": 0.6909813134878963, "3000–6000 Hz_40ms_log_rms_mae_db": 1.267844264141507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1740703763769428, "3000–6000 Hz_40ms_envelope_correlation": 0.9787952564106599, "3000–6000 Hz_40ms_modulation_mae_db": 0.8218529522158425, "mr_stft_db": 4.746852286423121, "band_mae_db": 0.2363930779036407, "rms_gain_error_db": -0.07323495251471442, "output_peak": 0.48822292327003175}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6720041548065809, "全帯域_10ms_rms_envelope_nrmse": 0.07360174547022857, "全帯域_10ms_envelope_correlation": 0.9941236522782961, "全帯域_10ms_modulation_mae_db": 0.2423994852913494, "全帯域_40ms_log_rms_mae_db": 0.3720427516401231, "全帯域_40ms_rms_envelope_nrmse": 0.044595380315594665, "全帯域_40ms_envelope_correlation": 0.9979413140811184, "全帯域_40ms_modulation_mae_db": 0.3770110315223607, "300–1000 Hz_10ms_log_rms_mae_db": 1.361937950820094, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10958435127342551, "300–1000 Hz_10ms_envelope_correlation": 0.9905819504325256, "300–1000 Hz_10ms_modulation_mae_db": 1.2037376698979354, "300–1000 Hz_40ms_log_rms_mae_db": 0.9249893969509521, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06957130303406903, "300–1000 Hz_40ms_envelope_correlation": 0.9964535520829648, "300–1000 Hz_40ms_modulation_mae_db": 1.3950822766144764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.421749354570662, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17293993717109463, "1000–3000 Hz_10ms_envelope_correlation": 0.9821370057248835, "1000–3000 Hz_10ms_modulation_mae_db": 0.6310409132646416, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9655272572539761, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11376581328889715, "1000–3000 Hz_40ms_envelope_correlation": 0.9943147728027129, "1000–3000 Hz_40ms_modulation_mae_db": 1.0825347371954483, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5285350397319772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22800588000490127, "3000–6000 Hz_10ms_envelope_correlation": 0.9652273021739511, "3000–6000 Hz_10ms_modulation_mae_db": 0.7833742338112556, "3000–6000 Hz_40ms_log_rms_mae_db": 1.345633245852368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17807597491592103, "3000–6000 Hz_40ms_envelope_correlation": 0.9777191110209345, "3000–6000 Hz_40ms_modulation_mae_db": 0.6609147176339402, "mr_stft_db": 4.761378516663882, "band_mae_db": 0.3976497274707807, "rms_gain_error_db": -0.1612963012377483, "output_peak": 0.4685216796093205}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6291117755720218, "全帯域_10ms_rms_envelope_nrmse": 0.06982465897329845, "全帯域_10ms_envelope_correlation": 0.9945024381075506, "全帯域_10ms_modulation_mae_db": 0.10182390812344494, "全帯域_40ms_log_rms_mae_db": 0.36569030344900594, "全帯域_40ms_rms_envelope_nrmse": 0.03892235852821944, "全帯域_40ms_envelope_correlation": 0.9982459809707289, "全帯域_40ms_modulation_mae_db": 0.2672792399378932, "300–1000 Hz_10ms_log_rms_mae_db": 1.3872179955800719, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11192474302439982, "300–1000 Hz_10ms_envelope_correlation": 0.9898569578752298, "300–1000 Hz_10ms_modulation_mae_db": 0.4393432668830097, "300–1000 Hz_40ms_log_rms_mae_db": 1.0267522988392812, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06759542063055927, "300–1000 Hz_40ms_envelope_correlation": 0.996340116687793, "300–1000 Hz_40ms_modulation_mae_db": 0.6127294217828502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4204186934379868, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15927756509198393, "1000–3000 Hz_10ms_envelope_correlation": 0.9839274173101002, "1000–3000 Hz_10ms_modulation_mae_db": 1.07562007656691, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9379585006210308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1064221598494037, "1000–3000 Hz_40ms_envelope_correlation": 0.9937878633853433, "1000–3000 Hz_40ms_modulation_mae_db": 0.393361360238959, "3000–6000 Hz_10ms_log_rms_mae_db": 1.553916963240597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23064689396627375, "3000–6000 Hz_10ms_envelope_correlation": 0.9655476111586371, "3000–6000 Hz_10ms_modulation_mae_db": 0.9548606290906573, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3112864662188015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18550272626553682, "3000–6000 Hz_40ms_envelope_correlation": 0.9765787805693237, "3000–6000 Hz_40ms_modulation_mae_db": 1.1189333333017744, "mr_stft_db": 4.747613688543605, "band_mae_db": 0.28654763709010794, "rms_gain_error_db": -0.08198302540376812, "output_peak": 0.4787795537262777}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6329420765629823, "全帯域_10ms_rms_envelope_nrmse": 0.06980404513796375, "全帯域_10ms_envelope_correlation": 0.9946018645905564, "全帯域_10ms_modulation_mae_db": 0.232454891947702, "全帯域_40ms_log_rms_mae_db": 0.3899385365291348, "全帯域_40ms_rms_envelope_nrmse": 0.036479299527698954, "全帯域_40ms_envelope_correlation": 0.9985742182164458, "全帯域_40ms_modulation_mae_db": 0.2325160644448007, "300–1000 Hz_10ms_log_rms_mae_db": 1.3109004048136836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10098342442943241, "300–1000 Hz_10ms_envelope_correlation": 0.9917944055993635, "300–1000 Hz_10ms_modulation_mae_db": 0.5377996756103535, "300–1000 Hz_40ms_log_rms_mae_db": 0.8916440128862826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05215177240439318, "300–1000 Hz_40ms_envelope_correlation": 0.9979931894211979, "300–1000 Hz_40ms_modulation_mae_db": 0.5820557802936522, "1000–3000 Hz_10ms_log_rms_mae_db": 1.358710216659551, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15398522832600597, "1000–3000 Hz_10ms_envelope_correlation": 0.9861518575982064, "1000–3000 Hz_10ms_modulation_mae_db": 1.5038814266571494, "1000–3000 Hz_40ms_log_rms_mae_db": 0.940360389310182, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10611575899649969, "1000–3000 Hz_40ms_envelope_correlation": 0.9950143130441783, "1000–3000 Hz_40ms_modulation_mae_db": 0.9880100029937394, "3000–6000 Hz_10ms_log_rms_mae_db": 1.556654638032106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2287039708819333, "3000–6000 Hz_10ms_envelope_correlation": 0.9648506117329501, "3000–6000 Hz_10ms_modulation_mae_db": 0.7653990527728235, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3092863012748468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1798482852558477, "3000–6000 Hz_40ms_envelope_correlation": 0.9775819798537148, "3000–6000 Hz_40ms_modulation_mae_db": 1.076505646195715, "mr_stft_db": 4.749037344772556, "band_mae_db": 0.28645892758504615, "rms_gain_error_db": -0.11779414253150455, "output_peak": 0.5253473760666484}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6892926257883126, "全帯域_10ms_rms_envelope_nrmse": 0.06949905970548188, "全帯域_10ms_envelope_correlation": 0.9946229544913153, "全帯域_10ms_modulation_mae_db": 0.26635032873876613, "全帯域_40ms_log_rms_mae_db": 0.38179246650919085, "全帯域_40ms_rms_envelope_nrmse": 0.03510138544770321, "全帯域_40ms_envelope_correlation": 0.9986569239989931, "全帯域_40ms_modulation_mae_db": 0.3702168695555786, "300–1000 Hz_10ms_log_rms_mae_db": 1.3512030533601707, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11826466134979896, "300–1000 Hz_10ms_envelope_correlation": 0.9886198302804948, "300–1000 Hz_10ms_modulation_mae_db": 0.5547632216016459, "300–1000 Hz_40ms_log_rms_mae_db": 0.9587013979887105, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07213150075848125, "300–1000 Hz_40ms_envelope_correlation": 0.9957426305809662, "300–1000 Hz_40ms_modulation_mae_db": 0.4151805014104431, "1000–3000 Hz_10ms_log_rms_mae_db": 1.54532436159011, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16070179251819172, "1000–3000 Hz_10ms_envelope_correlation": 0.9829254219456788, "1000–3000 Hz_10ms_modulation_mae_db": 1.2828369753653979, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0279960049107308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09950791854499737, "1000–3000 Hz_40ms_envelope_correlation": 0.993792073175887, "1000–3000 Hz_40ms_modulation_mae_db": 0.7234780055985209, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6065005657124973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23592312685541283, "3000–6000 Hz_10ms_envelope_correlation": 0.9625316565353536, "3000–6000 Hz_10ms_modulation_mae_db": 0.9204531045998965, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3581102268994747, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19127428339096575, "3000–6000 Hz_40ms_envelope_correlation": 0.9743668859403617, "3000–6000 Hz_40ms_modulation_mae_db": 1.3330272966458738, "mr_stft_db": 4.768542406317661, "band_mae_db": 0.23234764059788024, "rms_gain_error_db": -0.10644567051524031, "output_peak": 0.494153601339792}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6196045663072838, "全帯域_10ms_rms_envelope_nrmse": 0.0631410847489821, "全帯域_10ms_envelope_correlation": 0.9954740311855, "全帯域_10ms_modulation_mae_db": 0.16548445466054382, "全帯域_40ms_log_rms_mae_db": 0.3605302432213803, "全帯域_40ms_rms_envelope_nrmse": 0.03010631508658423, "全帯域_40ms_envelope_correlation": 0.9989219617365397, "全帯域_40ms_modulation_mae_db": 0.1060523828284629, "300–1000 Hz_10ms_log_rms_mae_db": 1.3624450072322065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09452610679875208, "300–1000 Hz_10ms_envelope_correlation": 0.9927469408908502, "300–1000 Hz_10ms_modulation_mae_db": 0.5694371525291942, "300–1000 Hz_40ms_log_rms_mae_db": 0.9848722610371943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05034791197598028, "300–1000 Hz_40ms_envelope_correlation": 0.9979958535958077, "300–1000 Hz_40ms_modulation_mae_db": 0.8655979484625753, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4432990371083598, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14138965003137577, "1000–3000 Hz_10ms_envelope_correlation": 0.9880561207131618, "1000–3000 Hz_10ms_modulation_mae_db": 1.0303805920556146, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0056489793924803, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10025352367934426, "1000–3000 Hz_40ms_envelope_correlation": 0.9949697064544835, "1000–3000 Hz_40ms_modulation_mae_db": 1.2073475380604322, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5520664886089197, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2210622850728827, "3000–6000 Hz_10ms_envelope_correlation": 0.9671794245524055, "3000–6000 Hz_10ms_modulation_mae_db": 0.8881361668553284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3847135649022906, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18085795480901837, "3000–6000 Hz_40ms_envelope_correlation": 0.9770851868202011, "3000–6000 Hz_40ms_modulation_mae_db": 0.9006133637961374, "mr_stft_db": 4.750772686793162, "band_mae_db": 0.2302328135329558, "rms_gain_error_db": -0.04702838754462278, "output_peak": 0.5254180948248188}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6432734518703386, "全帯域_10ms_rms_envelope_nrmse": 0.07368550451882702, "全帯域_10ms_envelope_correlation": 0.9938233864915849, "全帯域_10ms_modulation_mae_db": 0.8442586348063473, "全帯域_40ms_log_rms_mae_db": 0.3823597771967239, "全帯域_40ms_rms_envelope_nrmse": 0.03650901112667752, "全帯域_40ms_envelope_correlation": 0.9983706401462958, "全帯域_40ms_modulation_mae_db": 0.581608113599118, "300–1000 Hz_10ms_log_rms_mae_db": 1.3529935561011661, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1102268150477637, "300–1000 Hz_10ms_envelope_correlation": 0.9903281385800977, "300–1000 Hz_10ms_modulation_mae_db": 0.9210681688531466, "300–1000 Hz_40ms_log_rms_mae_db": 0.9569477888440773, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06033848661152104, "300–1000 Hz_40ms_envelope_correlation": 0.9973995673095954, "300–1000 Hz_40ms_modulation_mae_db": 0.7658242906322642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4262706817289255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19993174302784528, "1000–3000 Hz_10ms_envelope_correlation": 0.973188792238568, "1000–3000 Hz_10ms_modulation_mae_db": 1.142613213577964, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9134209943486394, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12415216355715848, "1000–3000 Hz_40ms_envelope_correlation": 0.9891603180854547, "1000–3000 Hz_40ms_modulation_mae_db": 0.8407769762344669, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6393854436476043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23619293087949803, "3000–6000 Hz_10ms_envelope_correlation": 0.962703321389204, "3000–6000 Hz_10ms_modulation_mae_db": 1.3235509373423922, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3898682401743483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18313428269970924, "3000–6000 Hz_40ms_envelope_correlation": 0.9765135999663423, "3000–6000 Hz_40ms_modulation_mae_db": 1.1504684742259175, "mr_stft_db": 4.758660167305052, "band_mae_db": 0.17530149735042358, "rms_gain_error_db": -0.022353563228725858, "output_peak": 0.48643231896092626}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6851386261158123, "全帯域_10ms_rms_envelope_nrmse": 0.07371060531559186, "全帯域_10ms_envelope_correlation": 0.9938588455178312, "全帯域_10ms_modulation_mae_db": 0.3811862316029775, "全帯域_40ms_log_rms_mae_db": 0.3800894443527571, "全帯域_40ms_rms_envelope_nrmse": 0.04153479188567189, "全帯域_40ms_envelope_correlation": 0.9979516445836042, "全帯域_40ms_modulation_mae_db": 0.2535169680210337, "300–1000 Hz_10ms_log_rms_mae_db": 1.4052807952850543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1126676795629189, "300–1000 Hz_10ms_envelope_correlation": 0.9896285329646937, "300–1000 Hz_10ms_modulation_mae_db": 0.6632157390711009, "300–1000 Hz_40ms_log_rms_mae_db": 0.9690335466149872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06456539381965193, "300–1000 Hz_40ms_envelope_correlation": 0.9965873163976336, "300–1000 Hz_40ms_modulation_mae_db": 0.4707222508469388, "1000–3000 Hz_10ms_log_rms_mae_db": 1.381916040123164, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17730444215003555, "1000–3000 Hz_10ms_envelope_correlation": 0.9803998123503912, "1000–3000 Hz_10ms_modulation_mae_db": 0.8652990820604204, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8445127331360024, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11687814435240174, "1000–3000 Hz_40ms_envelope_correlation": 0.9930552276536189, "1000–3000 Hz_40ms_modulation_mae_db": 0.4799825232383488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5392228964521686, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22563082596143136, "3000–6000 Hz_10ms_envelope_correlation": 0.9659622085423635, "3000–6000 Hz_10ms_modulation_mae_db": 0.7729351023007499, "3000–6000 Hz_40ms_log_rms_mae_db": 1.33905221987154, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18148336608262594, "3000–6000 Hz_40ms_envelope_correlation": 0.976923123659115, "3000–6000 Hz_40ms_modulation_mae_db": 0.8224830603175137, "mr_stft_db": 4.742756572688905, "band_mae_db": 0.30070090458544224, "rms_gain_error_db": -0.07172010578016408, "output_peak": 0.4722925884932119}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.847527575853446, "spectral_convergence": 0.22062903378286747, "spectral_convergence_status": "ok", "amplitude_floor": 1.419137961520426e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.498962350081633, "spectral_convergence": 0.2146250426850191, "spectral_convergence_status": "ok", "amplitude_floor": 1.2001614819603722e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.894066933334286, "spectral_convergence": 0.2294893214000301, "spectral_convergence_status": "ok", "amplitude_floor": 7.977266337453151e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015734976003854988, "output_power": 0.001606067161927811, "delta_db": 0.08897617547641187, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002272380907829781, "output_power": 0.002174881963690609, "delta_db": -0.19045440072445383, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00035321559220182435, "output_power": 0.0003624991642403426, "delta_db": 0.11267142977659116, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024119869501752642, "output_power": 0.00022143612524391098, "delta_db": -0.37126480338352785, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012307274117421543, "output_power": 0.00011731480709846604, "delta_db": -0.20809042827174473, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.88440946982043e-05, "output_power": 4.30541606710649e-05, "delta_db": 0.4469012297891148, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2363930779036407, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07323495251471442, "pitch": null, "output_rms_dbfs": -23.443570982925163, "output_peak": 0.48822292327003175, "reference_rms_dbfs": -23.37033603041045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.746852286423121}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00019859470734653442, "log_rms_mae_db": 0.6233909026513077, "rms_envelope_nrmse": 0.061818950763648316, "envelope_correlation": 0.9956977794627472, "rms_gain_error_db": -0.07323495251474953, "reference_mean_power": 0.004608354103444978, "output_mean_power": 0.004531295109685084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.3747785986813366e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009662478923264104, "output_power": 0.0009397552026559229, "reference_active": true, "delta_db": -0.12073820876343203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.594410683545364e-05, "output_power": 9.174616241789016e-05, "reference_active": true, "delta_db": -0.19430396744889974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7446486319031935e-05, "output_power": 1.8333403304340013e-05, "reference_active": true, "delta_db": 0.21535118076839704}], "active_band_count": 3, "active_band_mae_db": 0.17679778566024296}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00016219647203140816, "log_rms_mae_db": 0.3427053919417756, "rms_envelope_nrmse": 0.034443376613536414, "envelope_correlation": 0.9986217393483501, "rms_gain_error_db": -0.07323495251474756, "reference_mean_power": 0.004645568268479526, "output_mean_power": 0.0045678869948238615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1771130170213754e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008502808342452142, "output_power": 0.0008283128569740124, "reference_active": true, "delta_db": -0.11367987493560747}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.133174041000048e-05, "output_power": 3.939744963966057e-05, "reference_active": true, "delta_db": -0.20815584023992728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2770439378674992e-06, "output_power": 1.2899317241870556e-06, "reference_active": true, "delta_db": 0.04360883972346744}], "active_band_count": 3, "active_band_mae_db": 0.12181485163300075}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015290889340024391, "log_rms_mae_db": 1.3799903784499539, "rms_envelope_nrmse": 0.10783996031132792, "envelope_correlation": 0.9905736258789739, "rms_gain_error_db": -0.1935273679105319, "reference_mean_power": 0.0016143224036409894, "output_mean_power": 0.0015439653059032168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 8.858385684123232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006875363814202012, "output_power": 0.0006367145371980597, "reference_active": true, "delta_db": -0.33350918470426627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.173936122300265e-05, "output_power": 5.0780829351690776e-05, "reference_active": true, "delta_db": -0.8486234232076874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0160658568889886e-05, "output_power": 7.825370764797443e-06, "reference_active": true, "delta_db": -1.1341693439707563}], "active_band_count": 3, "active_band_mae_db": 0.7721006506275699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012027897252950328, "log_rms_mae_db": 0.9741147482765135, "rms_envelope_nrmse": 0.07042381726240066, "envelope_correlation": 0.9959121667358857, "rms_gain_error_db": -0.1935273679105309, "reference_mean_power": 0.0016273586545452231, "output_mean_power": 0.0015564333972025697}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.855745033342305e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006306037982733972, "output_power": 0.0005819780366471244, "reference_active": true, "delta_db": -0.34849987219273376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.290965975606541e-05, "output_power": 1.964251463261294e-05, "reference_active": true, "delta_db": -0.6682155394851665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0400487875049354e-06, "output_power": 6.137666016693322e-07, "reference_active": true, "delta_db": -2.2905045958836427}], "active_band_count": 3, "active_band_mae_db": 1.102406669187181}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.957022642296078e-05, "log_rms_mae_db": 1.4366555320580459, "rms_envelope_nrmse": 0.1377762602285102, "envelope_correlation": 0.9874859272648188, "rms_gain_error_db": -0.29483279791635214, "reference_mean_power": 0.000306890935740228, "output_mean_power": 0.00028674825362401917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.0486708619890815e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.911544017889007e-05, "output_power": 5.757498134311742e-05, "reference_active": true, "delta_db": -0.7934127237235036}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0167211042917879e-05, "output_power": 7.2810221010860576e-06, "reference_active": true, "delta_db": -1.450094890643783}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3309723892934117e-06, "output_power": 2.4742068332609074e-06, "reference_active": true, "delta_db": 0.25898872728714173}], "active_band_count": 3, "active_band_mae_db": 0.8341654472181427}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 8.985216395929287e-05, "log_rms_mae_db": 0.9332381264496951, "rms_envelope_nrmse": 0.08468459649818919, "envelope_correlation": 0.9955468717877457, "rms_gain_error_db": -0.2948327979163532, "reference_mean_power": 0.00030936919363315046, "output_mean_power": 0.0002890638519036209}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 9.09439257266203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.275291186820264e-05, "output_power": 5.145830671263331e-05, "reference_active": true, "delta_db": -0.8617839159321496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.365294966430701e-06, "output_power": 3.290062078127456e-06, "reference_active": true, "delta_db": -1.2280950219570061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9190263566647066e-07, "output_power": 1.6793675288692934e-07, "reference_active": true, "delta_db": -0.5793518784562208}], "active_band_count": 3, "active_band_mae_db": 0.889743605448459}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.809631559896483e-05, "log_rms_mae_db": 1.4999052218659679, "rms_envelope_nrmse": 0.21818843197742957, "envelope_correlation": 0.9681286996106039, "rms_gain_error_db": -0.10053885783305852, "reference_mean_power": 6.929937891252573e-05, "output_mean_power": 6.771353027649071e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.858105660801035e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.988342411251772e-05, "output_power": 3.534936216879851e-05, "reference_active": true, "delta_db": -0.5241085478439803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1366606898498195e-05, "output_power": 1.0709905127290859e-05, "reference_active": true, "delta_db": -0.25845216969858736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.736645328964402e-06, "output_power": 2.337488348972909e-06, "reference_active": true, "delta_db": 1.290383222921121}], "active_band_count": 3, "active_band_mae_db": 0.6909813134878963}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.651257368418169e-05, "log_rms_mae_db": 1.267844264141507, "rms_envelope_nrmse": 0.1740703763769428, "envelope_correlation": 0.9787952564106599, "rms_gain_error_db": -0.10053885783305952, "reference_mean_power": 6.985899704640885e-05, "output_mean_power": 6.826034209568174e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.431140169416232e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0602300645347785e-05, "output_power": 3.41103343233194e-05, "reference_active": true, "delta_db": -0.7566466662307045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067895142462218e-06, "output_power": 5.803270981180121e-06, "reference_active": true, "delta_db": -0.19365217035047705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.231130378903163e-07, "output_power": 1.7451391036142463e-07, "reference_active": true, "delta_db": 1.5152600200663464}], "active_band_count": 3, "active_band_mae_db": 0.8218529522158425}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9276573883621017, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "859511930891111bc9d49cceee2c53f1a9194063dfb28d32fa7a0e5e9c3822b3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9830340443932261, "standard_deviation": 0.004784558968603424, "minimum": 0.973188792238568, "maximum": 0.9880561207131618, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4292929896596005, "standard_deviation": 0.05488293095118583, "minimum": 1.358710216659551, "maximum": 1.54532436159011, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0457297158457801, "standard_deviation": 0.27410106071316814, "minimum": 0.6310409132646416, "maximum": 1.5038814266571494, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16291332731813038, "standard_deviation": 0.020251848675642697, "minimum": 0.1377762602285102, "maximum": 0.19993174302784528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.993705143298678, "standard_deviation": 0.002006022480215866, "minimum": 0.9891603180854547, "maximum": 0.9955468717877457, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9460828731778421, "standard_deviation": 0.0564206113279074, "minimum": 0.8445127331360024, "maximum": 1.0279960049107308, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8256543436260468, "standard_deviation": 0.28263529095515466, "minimum": 0.393361360238959, "maximum": 1.2073475380604322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10647250984586144, "standard_deviation": 0.012165900560941518, "minimum": 0.08468459649818919, "maximum": 0.12415216355715848, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9652663544619335, "standard_deviation": 0.0019507600544883623, "minimum": 0.9625316565353536, "maximum": 0.9681286996106039, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5595234071614796, "standard_deviation": 0.04408884479941512, "minimum": 1.4999052218659679, "maximum": 1.6393854436476043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8874613175326249, "standard_deviation": 0.19757808603923452, "minimum": 0.6909813134878963, "maximum": 1.3235509373423922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22804429319997035, "standard_deviation": 0.006407559807479108, "minimum": 0.21818843197742957, "maximum": 0.23619293087949803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9769454905300816, "standard_deviation": 0.0012770201783168407, "minimum": 0.9743668859403617, "maximum": 0.9787952564106599, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.338224316166897, "standard_deviation": 0.04105010950564197, "minimum": 1.267844264141507, "maximum": 1.3898682401743483, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9855998555415894, "standard_deviation": 0.2203868180902065, "minimum": 0.6609147176339402, "maximum": 1.3330272966458738, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18178090622457094, "standard_deviation": 0.00512224786198802, "minimum": 0.1740703763769428, "maximum": 0.19127428339096575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9905162978127786, "standard_deviation": 0.0012801659159278075, "minimum": 0.9886198302804948, "maximum": 0.9927469408908502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3639961427053002, "standard_deviation": 0.0283095445059625, "minimum": 1.3109004048136836, "maximum": 1.4052807952850543, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7076831931342444, "standard_deviation": 0.25102457650423965, "minimum": 0.4393432668830097, "maximum": 1.2037376698979354, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10825221772472741, "standard_deviation": 0.007370964468226589, "minimum": 0.09452610679875208, "maximum": 0.11826466134979896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968030491014805, "standard_deviation": 0.0008860993597944674, "minimum": 0.9957426305809662, "maximum": 0.9979958535958077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9608819314297498, "standard_deviation": 0.040104137814096025, "minimum": 0.8916440128862826, "maximum": 1.0267522988392812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7761998924037976, "standard_deviation": 0.33500363610848255, "minimum": 0.4151805014104431, "maximum": 1.3950822766144764, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06339070081213208, "standard_deviation": 0.00835694399756853, "minimum": 0.05034791197598028, "maximum": 0.07213150075848125, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.26820402826453466, "standard_deviation": 0.06638876394322166, "minimum": 0.17530149735042358, "maximum": 0.3976497274707807, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.7532017086884935, "standard_deviation": 0.00874949823353615, "minimum": 4.742756572688905, "maximum": 4.768542406317661, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4923960170363784, "standard_deviation": 0.02201252595408735, "minimum": 0.4685216796093205, "maximum": 0.5254180948248188, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08523201859456105, "standard_deviation": 0.043171354508837904, "minimum": -0.1612963012377483, "maximum": -0.022353563228725858, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945881190156728, "standard_deviation": 0.0006925434868078101, "minimum": 0.9938233864915849, "maximum": 0.9956977794627472, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.64934477245933, "standard_deviation": 0.028445048032294192, "minimum": 0.6196045663072838, "maximum": 0.6892926257883126, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30134446510392177, "standard_deviation": 0.23432384407420265, "minimum": 0.10182390812344494, "maximum": 0.8442586348063473, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06938570682925274, "standard_deviation": 0.004653717807029128, "minimum": 0.061818950763648316, "maximum": 0.07371060531559186, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984105528852596, "standard_deviation": 0.0003490973315155254, "minimum": 0.9979413140811184, "maximum": 0.9989219617365397, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37189361435501145, "standard_deviation": 0.01521230950585369, "minimum": 0.3427053919417756, "maximum": 0.3899385365291348, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28875194019278105, "standard_deviation": 0.15413496583924996, "minimum": 0.1060523828284629, "maximum": 0.581608113599118, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03721148981646079, "standard_deviation": 0.0044684278068563045, "minimum": 0.03010631508658423, "maximum": 0.044595380315594665, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0041", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1092872619628906, "predicted_mos": 2.1092872619628906, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "786887e22c2711db88e27f74c3cdb8c6cf80263e2373557bd629b61d1f939b8d", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.572340599726886, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6135577322262153, "全帯域_10ms_rms_envelope_nrmse": 0.06130544540081657, "全帯域_10ms_envelope_correlation": 0.9957752779778958, "全帯域_10ms_modulation_mae_db": 0.21557010175320004, "全帯域_40ms_log_rms_mae_db": 0.31893036735515995, "全帯域_40ms_rms_envelope_nrmse": 0.03262284421047361, "全帯域_40ms_envelope_correlation": 0.9987778037601861, "全帯域_40ms_modulation_mae_db": 0.13934236275526765, "300–1000 Hz_10ms_log_rms_mae_db": 1.3332227211753613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11311835832305758, "300–1000 Hz_10ms_envelope_correlation": 0.9895909879182414, "300–1000 Hz_10ms_modulation_mae_db": 0.6261833564243781, "300–1000 Hz_40ms_log_rms_mae_db": 0.8822260001795691, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07309259864890473, "300–1000 Hz_40ms_envelope_correlation": 0.9955731049015651, "300–1000 Hz_40ms_modulation_mae_db": 0.8830736676089033, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4365439640606432, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.134840392046491, "1000–3000 Hz_10ms_envelope_correlation": 0.9883198422484021, "1000–3000 Hz_10ms_modulation_mae_db": 0.8326980868028375, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9246938531653288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08411673458505753, "1000–3000 Hz_40ms_envelope_correlation": 0.9959582562386112, "1000–3000 Hz_40ms_modulation_mae_db": 0.9637484223272211, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4942188101323925, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21876861175325282, "3000–6000 Hz_10ms_envelope_correlation": 0.9679074850469827, "3000–6000 Hz_10ms_modulation_mae_db": 0.6566773438213322, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2983331519322414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17513019220249718, "3000–6000 Hz_40ms_envelope_correlation": 0.9784840718786406, "3000–6000 Hz_40ms_modulation_mae_db": 0.8641008429768876, "mr_stft_db": 4.701441902385049, "band_mae_db": 0.23951497910935848, "rms_gain_error_db": -0.07570307963361625, "output_peak": 0.4879830518130639}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6982987887882428, "全帯域_10ms_rms_envelope_nrmse": 0.0769517068496941, "全帯域_10ms_envelope_correlation": 0.9935297359364079, "全帯域_10ms_modulation_mae_db": 0.2908472603608043, "全帯域_40ms_log_rms_mae_db": 0.3936763038595652, "全帯域_40ms_rms_envelope_nrmse": 0.048301929567919136, "全帯域_40ms_envelope_correlation": 0.9974963599464128, "全帯域_40ms_modulation_mae_db": 0.2928063408853759, "300–1000 Hz_10ms_log_rms_mae_db": 1.3310829117792382, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11172897749243277, "300–1000 Hz_10ms_envelope_correlation": 0.9901604038517628, "300–1000 Hz_10ms_modulation_mae_db": 1.2144640317187008, "300–1000 Hz_40ms_log_rms_mae_db": 0.8623068481762333, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07051490052476325, "300–1000 Hz_40ms_envelope_correlation": 0.996308409409982, "300–1000 Hz_40ms_modulation_mae_db": 1.4258150810956145, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4002748823026272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1690711403626292, "1000–3000 Hz_10ms_envelope_correlation": 0.9828740885043078, "1000–3000 Hz_10ms_modulation_mae_db": 0.6183658170014042, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9317352837644142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1081506703952788, "1000–3000 Hz_40ms_envelope_correlation": 0.9949692185430097, "1000–3000 Hz_40ms_modulation_mae_db": 1.0528211645899432, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5340703724479001, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22978710224653812, "3000–6000 Hz_10ms_envelope_correlation": 0.9649308810479352, "3000–6000 Hz_10ms_modulation_mae_db": 0.7939267670153959, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3782972672062326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17992149399574747, "3000–6000 Hz_40ms_envelope_correlation": 0.9773153157023156, "3000–6000 Hz_40ms_modulation_mae_db": 0.7516528249575766, "mr_stft_db": 4.710030397148483, "band_mae_db": 0.3892894131629317, "rms_gain_error_db": -0.15733772915315297, "output_peak": 0.4736533155362734}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6408340456831015, "全帯域_10ms_rms_envelope_nrmse": 0.07382710241196014, "全帯域_10ms_envelope_correlation": 0.993861748858729, "全帯域_10ms_modulation_mae_db": 0.09754700321681449, "全帯域_40ms_log_rms_mae_db": 0.37134217318191326, "全帯域_40ms_rms_envelope_nrmse": 0.04291310002631147, "全帯域_40ms_envelope_correlation": 0.99787281394942, "全帯域_40ms_modulation_mae_db": 0.25696791354141824, "300–1000 Hz_10ms_log_rms_mae_db": 1.3450848570727276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11191785470800521, "300–1000 Hz_10ms_envelope_correlation": 0.9898952908012181, "300–1000 Hz_10ms_modulation_mae_db": 0.5256779495210826, "300–1000 Hz_40ms_log_rms_mae_db": 0.9790753333364136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06841065176831422, "300–1000 Hz_40ms_envelope_correlation": 0.9962856648983123, "300–1000 Hz_40ms_modulation_mae_db": 0.7017442740373414, "1000–3000 Hz_10ms_log_rms_mae_db": 1.37030528006631, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15031938244512352, "1000–3000 Hz_10ms_envelope_correlation": 0.9860449825936474, "1000–3000 Hz_10ms_modulation_mae_db": 1.1016212790808686, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8645792749247347, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09657117303982168, "1000–3000 Hz_40ms_envelope_correlation": 0.9953990227454717, "1000–3000 Hz_40ms_modulation_mae_db": 0.4815606932573308, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5593839629155641, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23434443166089397, "3000–6000 Hz_10ms_envelope_correlation": 0.9647779739173302, "3000–6000 Hz_10ms_modulation_mae_db": 1.1178879959379528, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3765023293351781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19007700173071163, "3000–6000 Hz_40ms_envelope_correlation": 0.9756345151324017, "3000–6000 Hz_40ms_modulation_mae_db": 1.2065192616935894, "mr_stft_db": 4.702422390994062, "band_mae_db": 0.28187824041981574, "rms_gain_error_db": -0.0926815548672681, "output_peak": 0.4756623902669271}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.636386525270405, "全帯域_10ms_rms_envelope_nrmse": 0.06988261566992145, "全帯域_10ms_envelope_correlation": 0.9946102645403517, "全帯域_10ms_modulation_mae_db": 0.19246665455234527, "全帯域_40ms_log_rms_mae_db": 0.3987480786248886, "全帯域_40ms_rms_envelope_nrmse": 0.037733708227128404, "全帯域_40ms_envelope_correlation": 0.9984903456118124, "全帯域_40ms_modulation_mae_db": 0.3275993934837652, "300–1000 Hz_10ms_log_rms_mae_db": 1.2665486102870744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10294738363463007, "300–1000 Hz_10ms_envelope_correlation": 0.9916003782562449, "300–1000 Hz_10ms_modulation_mae_db": 0.547519577529295, "300–1000 Hz_40ms_log_rms_mae_db": 0.8183983033618623, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05692989451870355, "300–1000 Hz_40ms_envelope_correlation": 0.9977026971431614, "300–1000 Hz_40ms_modulation_mae_db": 0.5788412401086793, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3497631549938567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14828439298693427, "1000–3000 Hz_10ms_envelope_correlation": 0.987817563581468, "1000–3000 Hz_10ms_modulation_mae_db": 1.0494898879523857, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8975474407993214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10166674055188353, "1000–3000 Hz_40ms_envelope_correlation": 0.9961671734919707, "1000–3000 Hz_40ms_modulation_mae_db": 1.0802849955052298, "3000–6000 Hz_10ms_log_rms_mae_db": 1.535492527586917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22825742854717462, "3000–6000 Hz_10ms_envelope_correlation": 0.9649921382552253, "3000–6000 Hz_10ms_modulation_mae_db": 0.6243453891209703, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3193584735863269, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18031659962220742, "3000–6000 Hz_40ms_envelope_correlation": 0.977233191832743, "3000–6000 Hz_40ms_modulation_mae_db": 1.0356849222438882, "mr_stft_db": 4.704145606999055, "band_mae_db": 0.2924065080382095, "rms_gain_error_db": -0.12668052818123515, "output_peak": 0.5191759923670292}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6938502508512939, "全帯域_10ms_rms_envelope_nrmse": 0.06883029359573833, "全帯域_10ms_envelope_correlation": 0.9947277070725776, "全帯域_10ms_modulation_mae_db": 0.18284833277547033, "全帯域_40ms_log_rms_mae_db": 0.3996932293056524, "全帯域_40ms_rms_envelope_nrmse": 0.035836242269625236, "全帯域_40ms_envelope_correlation": 0.9985855324647223, "全帯域_40ms_modulation_mae_db": 0.33636773822829746, "300–1000 Hz_10ms_log_rms_mae_db": 1.3044329217523074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11991993839334476, "300–1000 Hz_10ms_envelope_correlation": 0.9882569264780012, "300–1000 Hz_10ms_modulation_mae_db": 0.44423272965783195, "300–1000 Hz_40ms_log_rms_mae_db": 0.8949315761143405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07410786081893231, "300–1000 Hz_40ms_envelope_correlation": 0.9954359758014766, "300–1000 Hz_40ms_modulation_mae_db": 0.5204741309975632, "1000–3000 Hz_10ms_log_rms_mae_db": 1.482240030553464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14851515814844643, "1000–3000 Hz_10ms_envelope_correlation": 0.9859372413278208, "1000–3000 Hz_10ms_modulation_mae_db": 0.8491590894871538, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9775107248515289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09158147730128678, "1000–3000 Hz_40ms_envelope_correlation": 0.9953848045013879, "1000–3000 Hz_40ms_modulation_mae_db": 1.2734009794188623, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5897407693204124, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2348704748486634, "3000–6000 Hz_10ms_envelope_correlation": 0.9629504851323611, "3000–6000 Hz_10ms_modulation_mae_db": 1.008721080556924, "3000–6000 Hz_40ms_log_rms_mae_db": 1.384470937351204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19181784533127352, "3000–6000 Hz_40ms_envelope_correlation": 0.9741120817050792, "3000–6000 Hz_40ms_modulation_mae_db": 1.4391250171984427, "mr_stft_db": 4.710060018516882, "band_mae_db": 0.25721696746904005, "rms_gain_error_db": -0.10363028099343041, "output_peak": 0.4877989245153096}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6039251540154869, "全帯域_10ms_rms_envelope_nrmse": 0.06442654220531059, "全帯域_10ms_envelope_correlation": 0.9952879682981765, "全帯域_10ms_modulation_mae_db": 0.14093236526952316, "全帯域_40ms_log_rms_mae_db": 0.3622028275808208, "全帯域_40ms_rms_envelope_nrmse": 0.03202510447627498, "全帯域_40ms_envelope_correlation": 0.9987812073021918, "全帯域_40ms_modulation_mae_db": 0.14796552782316855, "300–1000 Hz_10ms_log_rms_mae_db": 1.3304170209711816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09945139310834863, "300–1000 Hz_10ms_envelope_correlation": 0.9919787938447837, "300–1000 Hz_10ms_modulation_mae_db": 0.621966556753308, "300–1000 Hz_40ms_log_rms_mae_db": 0.9165857798983313, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05284321057600421, "300–1000 Hz_40ms_envelope_correlation": 0.9978150049177599, "300–1000 Hz_40ms_modulation_mae_db": 0.9136127285504028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.400771708157786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13547078376117433, "1000–3000 Hz_10ms_envelope_correlation": 0.9891736635196099, "1000–3000 Hz_10ms_modulation_mae_db": 0.8275835196732185, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9629234694128516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09582702593888068, "1000–3000 Hz_40ms_envelope_correlation": 0.9955435572588505, "1000–3000 Hz_40ms_modulation_mae_db": 1.4134980992743362, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5397210422803262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21954558690792758, "3000–6000 Hz_10ms_envelope_correlation": 0.9676808349520706, "3000–6000 Hz_10ms_modulation_mae_db": 0.8163348025345933, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4003887253277825, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18100053796540969, "3000–6000 Hz_40ms_envelope_correlation": 0.9769935116854572, "3000–6000 Hz_40ms_modulation_mae_db": 0.9086869900503584, "mr_stft_db": 4.691570162391309, "band_mae_db": 0.22450949477299675, "rms_gain_error_db": -0.04909930393958526, "output_peak": 0.5203313994501951}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6589207152961435, "全帯域_10ms_rms_envelope_nrmse": 0.07494729322800302, "全帯域_10ms_envelope_correlation": 0.9936036652799963, "全帯域_10ms_modulation_mae_db": 0.8775254596766153, "全帯域_40ms_log_rms_mae_db": 0.39226512954583004, "全帯域_40ms_rms_envelope_nrmse": 0.0373587902356094, "全帯域_40ms_envelope_correlation": 0.9983008226213747, "全帯域_40ms_modulation_mae_db": 0.5559494412044091, "300–1000 Hz_10ms_log_rms_mae_db": 1.3215329930218396, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11581669110728333, "300–1000 Hz_10ms_envelope_correlation": 0.9893249125517144, "300–1000 Hz_10ms_modulation_mae_db": 0.9647378887434886, "300–1000 Hz_40ms_log_rms_mae_db": 0.9214170787406896, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06350962123605694, "300–1000 Hz_40ms_envelope_correlation": 0.997141243185894, "300–1000 Hz_40ms_modulation_mae_db": 0.8012955564535161, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4119778648672852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19299190384674142, "1000–3000 Hz_10ms_envelope_correlation": 0.9749605958418341, "1000–3000 Hz_10ms_modulation_mae_db": 0.6533273253121995, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9090195960782692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12741667992200845, "1000–3000 Hz_40ms_envelope_correlation": 0.9886873227279349, "1000–3000 Hz_40ms_modulation_mae_db": 0.3956575465025633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6105972384574232, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23321274865361297, "3000–6000 Hz_10ms_envelope_correlation": 0.9636395822828836, "3000–6000 Hz_10ms_modulation_mae_db": 1.3445688600419032, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3811553964143017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1823504167309455, "3000–6000 Hz_40ms_envelope_correlation": 0.9766616477709819, "3000–6000 Hz_40ms_modulation_mae_db": 1.2425859410868176, "mr_stft_db": 4.704189966277233, "band_mae_db": 0.17604868369349927, "rms_gain_error_db": -0.029439046931302892, "output_peak": 0.4830388097843099}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6722267817338158, "全帯域_10ms_rms_envelope_nrmse": 0.07375265054180076, "全帯域_10ms_envelope_correlation": 0.9938762301925608, "全帯域_10ms_modulation_mae_db": 0.32778533546339883, "全帯域_40ms_log_rms_mae_db": 0.38206602443866383, "全帯域_40ms_rms_envelope_nrmse": 0.042561884232650636, "全帯域_40ms_envelope_correlation": 0.9978813167292061, "全帯域_40ms_modulation_mae_db": 0.23868150094719806, "300–1000 Hz_10ms_log_rms_mae_db": 1.3735430272687825, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11142758673306458, "300–1000 Hz_10ms_envelope_correlation": 0.9899108454670359, "300–1000 Hz_10ms_modulation_mae_db": 0.6528352767324251, "300–1000 Hz_40ms_log_rms_mae_db": 0.9348185769478992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06467090387455954, "300–1000 Hz_40ms_envelope_correlation": 0.996638790466002, "300–1000 Hz_40ms_modulation_mae_db": 0.45400322162387935, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3473173773536684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16626712579361685, "1000–3000 Hz_10ms_envelope_correlation": 0.9832286097865255, "1000–3000 Hz_10ms_modulation_mae_db": 0.5858920656387268, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7884008321389983, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1093672051373518, "1000–3000 Hz_40ms_envelope_correlation": 0.9943769854085112, "1000–3000 Hz_40ms_modulation_mae_db": 0.3249927159399063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5499925921285873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2239499508922487, "3000–6000 Hz_10ms_envelope_correlation": 0.9664159235458951, "3000–6000 Hz_10ms_modulation_mae_db": 0.7091907504228697, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3641539669543161, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1786812577620642, "3000–6000 Hz_40ms_envelope_correlation": 0.9775834385588775, "3000–6000 Hz_40ms_modulation_mae_db": 0.8269802362011088, "mr_stft_db": 4.689339753424735, "band_mae_db": 0.3189265263451227, "rms_gain_error_db": -0.0873818540677057, "output_peak": 0.46973114587466525}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.796186149465905, "spectral_convergence": 0.22204228527309727, "spectral_convergence_status": "ok", "amplitude_floor": 1.419137961520426e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.460643651566013, "spectral_convergence": 0.21597289175070405, "spectral_convergence_status": "ok", "amplitude_floor": 1.2001614819603722e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.847495906123231, "spectral_convergence": 0.22873604702218972, "spectral_convergence_status": "ok", "amplitude_floor": 7.977266337453151e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015734976003854988, "output_power": 0.0016071668614374614, "delta_db": 0.09194884067270143, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002272380907829781, "output_power": 0.0021771000762495557, "delta_db": -0.1860273873812731, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00035321559220182435, "output_power": 0.0003615875556235432, "delta_db": 0.10173608763470614, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024119869501752642, "output_power": 0.00021822017657920133, "delta_db": -0.4348005090520106, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012307274117421543, "output_power": 0.00011625848243919073, "delta_db": -0.24737223991655116, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.88440946982043e-05, "output_power": 4.234922677119269e-05, "delta_db": 0.3752048099989086, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23951497910935848, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07570307963361625, "pitch": null, "output_rms_dbfs": -23.44603911004406, "output_peak": 0.4879830518130639, "reference_rms_dbfs": -23.37033603041045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.701441902385049}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00019859470734653442, "log_rms_mae_db": 0.6135577322262153, "rms_envelope_nrmse": 0.06130544540081657, "envelope_correlation": 0.9957752779778958, "rms_gain_error_db": -0.0757030796336543, "reference_mean_power": 0.004608354103444978, "output_mean_power": 0.004528720673331215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.3747785986813366e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009662478923264104, "output_power": 0.000942720755140016, "reference_active": true, "delta_db": -0.1070549089611216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.594410683545364e-05, "output_power": 9.217244010669521e-05, "reference_active": true, "delta_db": -0.17417219426544284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7446486319031935e-05, "output_power": 1.8978255730659715e-05, "reference_active": true, "delta_db": 0.36548320203303575}], "active_band_count": 3, "active_band_mae_db": 0.21557010175320004}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00016219647203140816, "log_rms_mae_db": 0.31893036735515995, "rms_envelope_nrmse": 0.03262284421047361, "envelope_correlation": 0.9987778037601861, "rms_gain_error_db": -0.07570307963365333, "reference_mean_power": 0.004645568268479526, "output_mean_power": 0.004565291768943581}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1771130170213754e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008502808342452142, "output_power": 0.0008314400153523729, "reference_active": true, "delta_db": -0.0973146767492512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.133174041000048e-05, "output_power": 3.969301249803384e-05, "reference_active": true, "delta_db": -0.17569632172932648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2770439378674992e-06, "output_power": 1.3204058774053573e-06, "reference_active": true, "delta_db": 0.14501608978722524}], "active_band_count": 3, "active_band_mae_db": 0.13934236275526765}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015290889340024391, "log_rms_mae_db": 1.3332227211753613, "rms_envelope_nrmse": 0.11311835832305758, "envelope_correlation": 0.9895909879182414, "rms_gain_error_db": -0.19792296507890234, "reference_mean_power": 0.0016143224036409894, "output_mean_power": 0.0015424034126479487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 8.858385684123232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006875363814202012, "output_power": 0.0006395351665336131, "reference_active": true, "delta_db": -0.31431253773416523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.173936122300265e-05, "output_power": 5.241658185211683e-05, "reference_active": true, "delta_db": -0.7109343513462107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0160658568889886e-05, "output_power": 8.348174912097552e-06, "reference_active": true, "delta_db": -0.853303180192758}], "active_band_count": 3, "active_band_mae_db": 0.6261833564243781}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012027897252950328, "log_rms_mae_db": 0.8822260001795691, "rms_envelope_nrmse": 0.07309259864890473, "envelope_correlation": 0.9955731049015651, "rms_gain_error_db": -0.19792296507890283, "reference_mean_power": 0.0016273586545452231, "output_mean_power": 0.0015548588910811753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.855745033342305e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006306037982733972, "output_power": 0.0005852312151736948, "reference_active": true, "delta_db": -0.3242909972385146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.290965975606541e-05, "output_power": 2.0535005599347408e-05, "reference_active": true, "delta_db": -0.4752381368132178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0400487875049354e-06, "output_power": 6.793358364201765e-07, "reference_active": true, "delta_db": -1.8496918687749773}], "active_band_count": 3, "active_band_mae_db": 0.8830736676089033}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.957022642296078e-05, "log_rms_mae_db": 1.4365439640606432, "rms_envelope_nrmse": 0.134840392046491, "envelope_correlation": 0.9883198422484021, "rms_gain_error_db": -0.3544556334609562, "reference_mean_power": 0.000306890935740228, "output_mean_power": 0.0002828384822233616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.0486708619890815e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.911544017889007e-05, "output_power": 5.705693454545421e-05, "reference_active": true, "delta_db": -0.8326664333933536}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0167211042917879e-05, "output_power": 7.461263969862994e-06, "reference_active": true, "delta_db": -1.3438943335353535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3309723892934117e-06, "output_power": 2.164630532333627e-06, "reference_active": true, "delta_db": -0.32153349347980553}], "active_band_count": 3, "active_band_mae_db": 0.8326980868028375}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 8.985216395929287e-05, "log_rms_mae_db": 0.9246938531653288, "rms_envelope_nrmse": 0.08411673458505753, "envelope_correlation": 0.9959582562386112, "rms_gain_error_db": -0.3544556334609583, "reference_mean_power": 0.00030936919363315046, "output_mean_power": 0.00028512250765181394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 9.09439257266203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.275291186820264e-05, "output_power": 5.093390395171726e-05, "reference_active": true, "delta_db": -0.9062691780740052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.365294966430701e-06, "output_power": 3.3644612116770526e-06, "reference_active": true, "delta_db": -1.1309806889697331}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9190263566647066e-07, "output_power": 1.5764543511690052e-07, "reference_active": true, "delta_db": -0.8539953999379253}], "active_band_count": 3, "active_band_mae_db": 0.9637484223272211}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.809631559896483e-05, "log_rms_mae_db": 1.4942188101323925, "rms_envelope_nrmse": 0.21876861175325282, "envelope_correlation": 0.9679074850469827, "rms_gain_error_db": -0.145738890861894, "reference_mean_power": 6.929937891252573e-05, "output_mean_power": 6.701244337070976e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.858105660801035e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.988342411251772e-05, "output_power": 3.571133200409507e-05, "reference_active": true, "delta_db": -0.4798638740288925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1366606898498195e-05, "output_power": 1.0754553325711797e-05, "reference_active": true, "delta_db": -0.24038463334777443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.736645328964402e-06, "output_power": 2.315738334411416e-06, "reference_active": true, "delta_db": 1.2497835240873298}], "active_band_count": 3, "active_band_mae_db": 0.6566773438213322}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.651257368418169e-05, "log_rms_mae_db": 1.2983331519322414, "rms_envelope_nrmse": 0.17513019220249718, "envelope_correlation": 0.9784840718786406, "rms_gain_error_db": -0.145738890861894, "reference_mean_power": 6.985899704640885e-05, "output_mean_power": 6.755359365364955e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.431140169416232e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0602300645347785e-05, "output_power": 3.4419595718284974e-05, "reference_active": true, "delta_db": -0.7174487773704908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067895142462218e-06, "output_power": 5.840389252991499e-06, "reference_active": true, "delta_db": -0.16596274123890664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.231130378903163e-07, "output_power": 1.82470704559166e-07, "reference_active": true, "delta_db": 1.7088910103212653}], "active_band_count": 3, "active_band_mae_db": 0.8641008429768876}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9282973682511934, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d0a5a920c1cdddff1bcf1a415a578b98e9f094e263f99f4efcb141314a970af3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.984794573425452, "standard_deviation": 0.004574139300756511, "minimum": 0.9749605958418341, "maximum": 0.9891736635196099, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.399899282794455, "standard_deviation": 0.045393389765162453, "minimum": 1.3473173773536684, "maximum": 1.482240030553464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8147671338685993, "standard_deviation": 0.19126225198878663, "minimum": 0.5858920656387268, "maximum": 1.1016212790808686, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15572003492389463, "standard_deviation": 0.019497410385380606, "minimum": 0.134840392046491, "maximum": 0.19299190384674142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945607926144685, "standard_deviation": 0.002437184005276919, "minimum": 0.9886873227279349, "maximum": 0.9961671734919707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9070513093919308, "standard_deviation": 0.05974086011237663, "minimum": 0.7884008321389983, "maximum": 0.9775107248515289, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8732455771019241, "standard_deviation": 0.4170284652980402, "minimum": 0.3249927159399063, "maximum": 1.4134980992743362, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10183721335894616, "standard_deviation": 0.013276981746554634, "minimum": 0.08411673458505753, "maximum": 0.12741667992200845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9654119130225856, "standard_deviation": 0.0017885823794580654, "minimum": 0.9629504851323611, "maximum": 0.9679074850469827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5516521644086905, "standard_deviation": 0.03585485342016064, "minimum": 1.4942188101323925, "maximum": 1.6105972384574232, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8839566236814926, "standard_deviation": 0.2520568953108837, "minimum": 0.6243453891209703, "maximum": 1.3445688600419032, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22784204193878901, "standard_deviation": 0.0064452298157415665, "minimum": 0.21876861175325282, "maximum": 0.2348704748486634, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9767522217833121, "standard_deviation": 0.0013376239750022931, "minimum": 0.9741120817050792, "maximum": 0.9784840718786406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.362832531013448, "standard_deviation": 0.035238669855642, "minimum": 1.2983331519322414, "maximum": 1.4003887253277825, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0344170045510837, "standard_deviation": 0.24039993514754435, "minimum": 0.7516528249575766, "maximum": 1.4391250171984427, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18241191816760707, "standard_deviation": 0.0056915310618291136, "minimum": 0.17513019220249718, "maximum": 0.19181784533127352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9900898173961252, "standard_deviation": 0.00120227638358657, "minimum": 0.9882569264780012, "maximum": 0.9919787938447837, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.325733132916064, "standard_deviation": 0.031032416205635, "minimum": 1.2665486102870744, "maximum": 1.3735430272687825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6997021708850637, "standard_deviation": 0.2584505090518385, "minimum": 0.44423272965783195, "maximum": 1.2144640317187008, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11079102293752086, "standard_deviation": 0.00661897199805519, "minimum": 0.09945139310834863, "maximum": 0.11991993839334476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966126113405192, "standard_deviation": 0.0008927207182257543, "minimum": 0.9954359758014766, "maximum": 0.9978150049177599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9012199370944174, "standard_deviation": 0.048709532077380324, "minimum": 0.8183983033618623, "maximum": 0.9790753333364136, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7848574875594874, "standard_deviation": 0.3084604123516723, "minimum": 0.45400322162387935, "maximum": 1.4258150810956145, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06550995524577985, "standard_deviation": 0.007593730353285514, "minimum": 0.05284321057600421, "maximum": 0.07410786081893231, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2724738516263717, "standard_deviation": 0.06459426340833999, "minimum": 0.17604868369349927, "maximum": 0.3892894131629317, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.701650024767101, "standard_deviation": 0.00762911846831739, "minimum": 4.689339753424735, "maximum": 4.710060018516882, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4896718787009716, "standard_deviation": 0.019680844894679375, "minimum": 0.46973114587466525, "maximum": 0.5203313994501951, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09024417222091209, "standard_deviation": 0.04075025301575608, "minimum": -0.15733772915315297, "maximum": -0.029439046931302892, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994409074769587, "standard_deviation": 0.0008269029252735186, "minimum": 0.9935297359364079, "maximum": 0.9957752779778958, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.652249999233088, "standard_deviation": 0.03486504233494935, "minimum": 0.6039251540154869, "maximum": 0.6982987887882428, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2906903141335215, "standard_deviation": 0.24853978170701643, "minimum": 0.09754700321681449, "maximum": 0.8775254596766153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07049045623790562, "standard_deviation": 0.00544645838292473, "minimum": 0.06130544540081657, "maximum": 0.0769517068496941, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982732752981658, "standard_deviation": 0.0004744207113132189, "minimum": 0.9974963599464128, "maximum": 0.9987812073021918, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37736551673656177, "standard_deviation": 0.027115441583320632, "minimum": 0.31893036735515995, "maximum": 0.3996932293056524, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28696002735861253, "standard_deviation": 0.13137892466511317, "minimum": 0.13934236275526765, "maximum": 0.5559494412044091, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03866920040574911, "standard_deviation": 0.005571272599798369, "minimum": 0.03202510447627498, "maximum": 0.048301929567919136, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0041", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.9981296062469482, "predicted_mos": 1.9981296062469482, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "70fa201219efb114abab4401eb7f57ea9565d8e5d3e0b368e2567a13b5c6cb0b", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.12793800001964, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6791577866122688, "全帯域_10ms_rms_envelope_nrmse": 0.08305280731510677, "全帯域_10ms_envelope_correlation": 0.9923520665657278, "全帯域_10ms_modulation_mae_db": 0.38925746061699656, "全帯域_40ms_log_rms_mae_db": 0.40178765331379473, "全帯域_40ms_rms_envelope_nrmse": 0.05194812821072675, "全帯域_40ms_envelope_correlation": 0.9970024366652623, "全帯域_40ms_modulation_mae_db": 0.23431453425832935, "300–1000 Hz_10ms_log_rms_mae_db": 1.6483846127481072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12838792654671885, "300–1000 Hz_10ms_envelope_correlation": 0.9869245347332897, "300–1000 Hz_10ms_modulation_mae_db": 0.6848690835337551, "300–1000 Hz_40ms_log_rms_mae_db": 1.2151905912557832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08400197039234297, "300–1000 Hz_40ms_envelope_correlation": 0.9946636812839733, "300–1000 Hz_40ms_modulation_mae_db": 0.48007678127850023, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7331887791202305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16644925700596208, "1000–3000 Hz_10ms_envelope_correlation": 0.9813174845461146, "1000–3000 Hz_10ms_modulation_mae_db": 1.2856262041327806, "1000–3000 Hz_40ms_log_rms_mae_db": 1.198531916886735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10675007331541224, "1000–3000 Hz_40ms_envelope_correlation": 0.9924124082046846, "1000–3000 Hz_40ms_modulation_mae_db": 0.8403494225331131, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9667504426691431, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2397171276060397, "3000–6000 Hz_10ms_envelope_correlation": 0.9613099448302318, "3000–6000 Hz_10ms_modulation_mae_db": 1.1792686144391389, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6272437315752286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19183287160419796, "3000–6000 Hz_40ms_envelope_correlation": 0.9744354661711991, "3000–6000 Hz_40ms_modulation_mae_db": 1.095874010240159, "mr_stft_db": 5.023869474214615, "band_mae_db": 0.22988447038355186, "rms_gain_error_db": -0.15037272808454324, "output_peak": 0.49192600783009344}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7000438792794653, "全帯域_10ms_rms_envelope_nrmse": 0.07881912009837184, "全帯域_10ms_envelope_correlation": 0.9933727973383125, "全帯域_10ms_modulation_mae_db": 0.3675450841311998, "全帯域_40ms_log_rms_mae_db": 0.4693533649804597, "全帯域_40ms_rms_envelope_nrmse": 0.05112130649971583, "全帯域_40ms_envelope_correlation": 0.9973667551103306, "全帯域_40ms_modulation_mae_db": 0.5067474456567835, "300–1000 Hz_10ms_log_rms_mae_db": 1.5633637763136863, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12547821810486093, "300–1000 Hz_10ms_envelope_correlation": 0.9880299304558143, "300–1000 Hz_10ms_modulation_mae_db": 1.0711354287013515, "300–1000 Hz_40ms_log_rms_mae_db": 1.1807218969784035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08973858977843328, "300–1000 Hz_40ms_envelope_correlation": 0.9942566931945938, "300–1000 Hz_40ms_modulation_mae_db": 1.4880004790422747, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6926811471482714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14801273941669665, "1000–3000 Hz_10ms_envelope_correlation": 0.9855618159508033, "1000–3000 Hz_10ms_modulation_mae_db": 0.6907374293610706, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0984654029247094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08904859070744318, "1000–3000 Hz_40ms_envelope_correlation": 0.9952025010663588, "1000–3000 Hz_40ms_modulation_mae_db": 1.0879973607253555, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9924104731724175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23156951128286277, "3000–6000 Hz_10ms_envelope_correlation": 0.9639506199435036, "3000–6000 Hz_10ms_modulation_mae_db": 1.5177954453736036, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6061798893643935, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18050306964312407, "3000–6000 Hz_40ms_envelope_correlation": 0.977491988337787, "3000–6000 Hz_40ms_modulation_mae_db": 1.3182467113100387, "mr_stft_db": 5.008210257685665, "band_mae_db": 0.26538466438051217, "rms_gain_error_db": -0.19895886991327594, "output_peak": 0.46445423003734077}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6201419816550535, "全帯域_10ms_rms_envelope_nrmse": 0.07339020550250455, "全帯域_10ms_envelope_correlation": 0.9939253099852257, "全帯域_10ms_modulation_mae_db": 0.3987042936149983, "全帯域_40ms_log_rms_mae_db": 0.36633326659815313, "全帯域_40ms_rms_envelope_nrmse": 0.03965824402564447, "全帯域_40ms_envelope_correlation": 0.9981746662290079, "全帯域_40ms_modulation_mae_db": 0.1527025336864283, "300–1000 Hz_10ms_log_rms_mae_db": 1.5516732510791447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11372660331357246, "300–1000 Hz_10ms_envelope_correlation": 0.989643881334326, "300–1000 Hz_10ms_modulation_mae_db": 0.5012387226565111, "300–1000 Hz_40ms_log_rms_mae_db": 1.0671200475613611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07113012718735733, "300–1000 Hz_40ms_envelope_correlation": 0.9960600813825687, "300–1000 Hz_40ms_modulation_mae_db": 0.8484407356989584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5991011723546513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1787186623792502, "1000–3000 Hz_10ms_envelope_correlation": 0.9786153705438412, "1000–3000 Hz_10ms_modulation_mae_db": 0.9883894550749973, "1000–3000 Hz_40ms_log_rms_mae_db": 1.035463438457759, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11653036213576194, "1000–3000 Hz_40ms_envelope_correlation": 0.9912372990762675, "1000–3000 Hz_40ms_modulation_mae_db": 0.4899779415991536, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9067008170486612, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2273515289268592, "3000–6000 Hz_10ms_envelope_correlation": 0.9657282785698267, "3000–6000 Hz_10ms_modulation_mae_db": 0.7094695745621928, "3000–6000 Hz_40ms_log_rms_mae_db": 1.543718365315378, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18376039261588545, "3000–6000 Hz_40ms_envelope_correlation": 0.9764988251871756, "3000–6000 Hz_40ms_modulation_mae_db": 0.9651135768447118, "mr_stft_db": 4.984931283536493, "band_mae_db": 0.28297763306392815, "rms_gain_error_db": -0.08659803352476432, "output_peak": 0.4621738560346104}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.662115349015901, "全帯域_10ms_rms_envelope_nrmse": 0.08164646501693636, "全帯域_10ms_envelope_correlation": 0.9925027110801731, "全帯域_10ms_modulation_mae_db": 0.10179969403527844, "全帯域_40ms_log_rms_mae_db": 0.3556837278584222, "全帯域_40ms_rms_envelope_nrmse": 0.04144625118868995, "全帯域_40ms_envelope_correlation": 0.9979245492529597, "全帯域_40ms_modulation_mae_db": 0.14268584260463107, "300–1000 Hz_10ms_log_rms_mae_db": 1.7109173361330854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12540932694795826, "300–1000 Hz_10ms_envelope_correlation": 0.9869250291347679, "300–1000 Hz_10ms_modulation_mae_db": 0.4521866376476495, "300–1000 Hz_40ms_log_rms_mae_db": 1.1953333412537517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07096826364639142, "300–1000 Hz_40ms_envelope_correlation": 0.9955210088285623, "300–1000 Hz_40ms_modulation_mae_db": 0.790876500237856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7713665631683442, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17223801172338934, "1000–3000 Hz_10ms_envelope_correlation": 0.9798083571653541, "1000–3000 Hz_10ms_modulation_mae_db": 1.6017844718563186, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2431060818076827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10734648153675447, "1000–3000 Hz_40ms_envelope_correlation": 0.992039224424598, "1000–3000 Hz_40ms_modulation_mae_db": 1.1857307027426514, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9923596700836912, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24377246127160074, "3000–6000 Hz_10ms_envelope_correlation": 0.9599200938072415, "3000–6000 Hz_10ms_modulation_mae_db": 0.9691042589764995, "3000–6000 Hz_40ms_log_rms_mae_db": 1.607728134696046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19039878381499212, "3000–6000 Hz_40ms_envelope_correlation": 0.9747583280661369, "3000–6000 Hz_40ms_modulation_mae_db": 0.8383611735133295, "mr_stft_db": 5.014402028844343, "band_mae_db": 0.13279674045144682, "rms_gain_error_db": 0.032632432974694384, "output_peak": 0.5177706580730052}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6608071808009321, "全帯域_10ms_rms_envelope_nrmse": 0.07571849162610675, "全帯域_10ms_envelope_correlation": 0.9935067428591895, "全帯域_10ms_modulation_mae_db": 0.6197156930773663, "全帯域_40ms_log_rms_mae_db": 0.3618554306655044, "全帯域_40ms_rms_envelope_nrmse": 0.03927259163760487, "全帯域_40ms_envelope_correlation": 0.9981770861206615, "全帯域_40ms_modulation_mae_db": 0.35901240943997564, "300–1000 Hz_10ms_log_rms_mae_db": 1.5621429132685507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10975748296233082, "300–1000 Hz_10ms_envelope_correlation": 0.9905284860738135, "300–1000 Hz_10ms_modulation_mae_db": 0.5494938085646869, "300–1000 Hz_40ms_log_rms_mae_db": 1.1051662170568541, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0674540810931925, "300–1000 Hz_40ms_envelope_correlation": 0.9966858962021848, "300–1000 Hz_40ms_modulation_mae_db": 0.8362061444609382, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7591715520951188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16359622685134304, "1000–3000 Hz_10ms_envelope_correlation": 0.9818337982602134, "1000–3000 Hz_10ms_modulation_mae_db": 1.2124858676255876, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1701077423272006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10204638298954628, "1000–3000 Hz_40ms_envelope_correlation": 0.9928775557961266, "1000–3000 Hz_40ms_modulation_mae_db": 0.6583735434303458, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9984198327679696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22396345635590256, "3000–6000 Hz_10ms_envelope_correlation": 0.9667832055989142, "3000–6000 Hz_10ms_modulation_mae_db": 0.6991397817046208, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6158644792256156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1770357298184652, "3000–6000 Hz_40ms_envelope_correlation": 0.9782805280049336, "3000–6000 Hz_40ms_modulation_mae_db": 0.9037260128640794, "mr_stft_db": 5.0238834357752005, "band_mae_db": 0.21471905411758632, "rms_gain_error_db": -0.07160089843562382, "output_peak": 0.46749266888179686}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6340191848907688, "全帯域_10ms_rms_envelope_nrmse": 0.06628793719837209, "全帯域_10ms_envelope_correlation": 0.9951018517808974, "全帯域_10ms_modulation_mae_db": 0.19510303015456043, "全帯域_40ms_log_rms_mae_db": 0.35252558614301444, "全帯域_40ms_rms_envelope_nrmse": 0.036729747984377135, "全帯域_40ms_envelope_correlation": 0.9984899203963163, "全帯域_40ms_modulation_mae_db": 0.11295268097094664, "300–1000 Hz_10ms_log_rms_mae_db": 1.6196965328796753, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11393057329589004, "300–1000 Hz_10ms_envelope_correlation": 0.9895601524605161, "300–1000 Hz_10ms_modulation_mae_db": 0.8207954124485625, "300–1000 Hz_40ms_log_rms_mae_db": 1.1686988159493361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0715673069749138, "300–1000 Hz_40ms_envelope_correlation": 0.9959412444400259, "300–1000 Hz_40ms_modulation_mae_db": 1.1683022030396535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7648166041429463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18870707443582613, "1000–3000 Hz_10ms_envelope_correlation": 0.9765684328272484, "1000–3000 Hz_10ms_modulation_mae_db": 1.421590380057495, "1000–3000 Hz_40ms_log_rms_mae_db": 1.203442049923716, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1273808678395438, "1000–3000 Hz_40ms_envelope_correlation": 0.9900699443474262, "1000–3000 Hz_40ms_modulation_mae_db": 0.9382490107039435, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9632290532291312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23330709474787106, "3000–6000 Hz_10ms_envelope_correlation": 0.9634932951303369, "3000–6000 Hz_10ms_modulation_mae_db": 0.9299175004288633, "3000–6000 Hz_40ms_log_rms_mae_db": 1.628618200361429, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18909756572554473, "3000–6000 Hz_40ms_envelope_correlation": 0.9749299383832102, "3000–6000 Hz_40ms_modulation_mae_db": 1.0216327721990217, "mr_stft_db": 5.017695430562238, "band_mae_db": 0.3069431399096642, "rms_gain_error_db": -0.09878243943401406, "output_peak": 0.44656673082195575}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6385454816651028, "全帯域_10ms_rms_envelope_nrmse": 0.0683583006218325, "全帯域_10ms_envelope_correlation": 0.9948126038681498, "全帯域_10ms_modulation_mae_db": 0.13828288530184876, "全帯域_40ms_log_rms_mae_db": 0.35248059126503883, "全帯域_40ms_rms_envelope_nrmse": 0.035822512321098454, "全帯域_40ms_envelope_correlation": 0.9986164102075452, "全帯域_40ms_modulation_mae_db": 0.2752270315032063, "300–1000 Hz_10ms_log_rms_mae_db": 1.621846466292689, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11240305898017988, "300–1000 Hz_10ms_envelope_correlation": 0.9897457248037952, "300–1000 Hz_10ms_modulation_mae_db": 0.3424645016623451, "300–1000 Hz_40ms_log_rms_mae_db": 1.1656763080901549, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06089280650368966, "300–1000 Hz_40ms_envelope_correlation": 0.9971202541171986, "300–1000 Hz_40ms_modulation_mae_db": 0.667591183620082, "1000–3000 Hz_10ms_log_rms_mae_db": 1.680417997317449, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1720188151521808, "1000–3000 Hz_10ms_envelope_correlation": 0.9821238729474671, "1000–3000 Hz_10ms_modulation_mae_db": 0.9528866862552768, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1635224221844294, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11769312794345091, "1000–3000 Hz_40ms_envelope_correlation": 0.9933474085752231, "1000–3000 Hz_40ms_modulation_mae_db": 1.1195341427628531, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9665986237453446, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21089948263409994, "3000–6000 Hz_10ms_envelope_correlation": 0.9705427886884672, "3000–6000 Hz_10ms_modulation_mae_db": 0.6890683361543674, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5425485762956137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17167624774191514, "3000–6000 Hz_40ms_envelope_correlation": 0.9795648141440966, "3000–6000 Hz_40ms_modulation_mae_db": 0.8063148329627431, "mr_stft_db": 5.025467447946025, "band_mae_db": 0.3970652135514776, "rms_gain_error_db": -0.11366836766753424, "output_peak": 0.45267500056923876}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6419458940153675, "全帯域_10ms_rms_envelope_nrmse": 0.06979560279186794, "全帯域_10ms_envelope_correlation": 0.9944508822763697, "全帯域_10ms_modulation_mae_db": 0.5207981433822302, "全帯域_40ms_log_rms_mae_db": 0.3404970214235505, "全帯域_40ms_rms_envelope_nrmse": 0.03556101284870169, "全帯域_40ms_envelope_correlation": 0.9984644025483553, "全帯域_40ms_modulation_mae_db": 0.35637542743200545, "300–1000 Hz_10ms_log_rms_mae_db": 1.631077308332385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11817828646696421, "300–1000 Hz_10ms_envelope_correlation": 0.988519218547333, "300–1000 Hz_10ms_modulation_mae_db": 0.7636512876013208, "300–1000 Hz_40ms_log_rms_mae_db": 1.124843282048608, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06264456367102005, "300–1000 Hz_40ms_envelope_correlation": 0.9967777325218709, "300–1000 Hz_40ms_modulation_mae_db": 1.0146182576513638, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7058285498527659, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15128442934319247, "1000–3000 Hz_10ms_envelope_correlation": 0.9851062994776147, "1000–3000 Hz_10ms_modulation_mae_db": 1.156740059473447, "1000–3000 Hz_40ms_log_rms_mae_db": 1.118728107967153, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09587936044770155, "1000–3000 Hz_40ms_envelope_correlation": 0.9945677742793856, "1000–3000 Hz_40ms_modulation_mae_db": 0.8678027107149225, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9485959698689128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23078542909395242, "3000–6000 Hz_10ms_envelope_correlation": 0.9644831351706324, "3000–6000 Hz_10ms_modulation_mae_db": 1.279935482210143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6075042723870285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18305037282424022, "3000–6000 Hz_40ms_envelope_correlation": 0.9773400111145734, "3000–6000 Hz_40ms_modulation_mae_db": 1.1210943334600223, "mr_stft_db": 5.011265509902988, "band_mae_db": 0.28981891990173253, "rms_gain_error_db": -0.02590507651068615, "output_peak": 0.47464663330540074}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.161142328854875, "spectral_convergence": 0.2287776272774462, "spectral_convergence_status": "ok", "amplitude_floor": 1.419137961520426e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.83313971395685, "spectral_convergence": 0.2304285422778039, "spectral_convergence_status": "ok", "amplitude_floor": 1.2001614819603722e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.077326379832119, "spectral_convergence": 0.23760857251915293, "spectral_convergence_status": "ok", "amplitude_floor": 7.977266337453151e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015734976003854988, "output_power": 0.0016014301110014457, "delta_db": 0.07641905158035817, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002272380907829781, "output_power": 0.0021276936657248216, "delta_db": -0.28572031088288463, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00035321559220182435, "output_power": 0.00033624647089803097, "delta_db": -0.21382131781609195, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024119869501752642, "output_power": 0.00022781483816633735, "delta_db": -0.24792946384803924, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012307274117421543, "output_power": 0.00011216934016753126, "delta_db": -0.40287708499734204, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.88440946982043e-05, "output_power": 4.023268005143485e-05, "delta_db": 0.15253959317659524, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22988447038355186, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15037272808454324, "pitch": null, "output_rms_dbfs": -23.520708758494994, "output_peak": 0.49192600783009344, "reference_rms_dbfs": -23.37033603041045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.023869474214615}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00019859470734653442, "log_rms_mae_db": 0.6791577866122688, "rms_envelope_nrmse": 0.08305280731510677, "envelope_correlation": 0.9923520665657278, "rms_gain_error_db": -0.15037272808457997, "reference_mean_power": 0.004608354103444978, "output_mean_power": 0.004451522469301969}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.3747785986813366e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009662478923264104, "output_power": 0.0009421424359630159, "reference_active": true, "delta_db": -0.10971993883084374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.594410683545364e-05, "output_power": 9.094661546584773e-05, "reference_active": true, "delta_db": -0.2323176262561721}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7446486319031935e-05, "output_power": 2.10999514447467e-05, "reference_active": true, "delta_db": 0.8257348167639738}], "active_band_count": 3, "active_band_mae_db": 0.38925746061699656}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00016219647203140816, "log_rms_mae_db": 0.40178765331379473, "rms_envelope_nrmse": 0.05194812821072675, "envelope_correlation": 0.9970024366652623, "rms_gain_error_db": -0.15037272808457897, "reference_mean_power": 0.004645568268479526, "output_mean_power": 0.004487470160844112}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1771130170213754e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008502808342452142, "output_power": 0.0008343563997862237, "reference_active": true, "delta_db": -0.08210788497951065}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.133174041000048e-05, "output_power": 3.979383639191849e-05, "reference_active": true, "delta_db": -0.16467883019377672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2770439378674992e-06, "output_power": 1.149714877025399e-06, "reference_active": true, "delta_db": -0.4561568876017006}], "active_band_count": 3, "active_band_mae_db": 0.23431453425832935}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015290889340024391, "log_rms_mae_db": 1.6483846127481072, "rms_envelope_nrmse": 0.12838792654671885, "envelope_correlation": 0.9869245347332897, "rms_gain_error_db": -0.3085687724435779, "reference_mean_power": 0.0016143224036409894, "output_mean_power": 0.001503603738039669}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 8.858385684123232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006875363814202012, "output_power": 0.0006168519955581201, "reference_active": true, "delta_db": -0.47114710031523815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.173936122300265e-05, "output_power": 5.3619003064726264e-05, "reference_active": true, "delta_db": -0.6124339689919743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0160658568889886e-05, "output_power": 1.2706457144353181e-05, "reference_active": true, "delta_db": 0.9710261812940529}], "active_band_count": 3, "active_band_mae_db": 0.6848690835337551}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012027897252950328, "log_rms_mae_db": 1.2151905912557832, "rms_envelope_nrmse": 0.08400197039234297, "envelope_correlation": 0.9946636812839733, "rms_gain_error_db": -0.3085687724435779, "reference_mean_power": 0.0016273586545452231, "output_mean_power": 0.0015157458947398545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.855745033342305e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006306037982733972, "output_power": 0.0005650257661683565, "reference_active": true, "delta_db": -0.4768832941351524}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.290965975606541e-05, "output_power": 2.02250102920117e-05, "reference_active": true, "delta_db": -0.5412988800533518}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0400487875049354e-06, "output_power": 9.437324644752242e-07, "reference_active": true, "delta_db": -0.4220481696469963}], "active_band_count": 3, "active_band_mae_db": 0.48007678127850023}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.957022642296078e-05, "log_rms_mae_db": 1.7331887791202305, "rms_envelope_nrmse": 0.16644925700596208, "envelope_correlation": 0.9813174845461146, "rms_gain_error_db": -0.2844328206615312, "reference_mean_power": 0.000306890935740228, "output_mean_power": 0.0002874357477033918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.0486708619890815e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.911544017889007e-05, "output_power": 5.749531290715032e-05, "reference_active": true, "delta_db": -0.7994263639729099}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0167211042917879e-05, "output_power": 7.468921631161241e-06, "reference_active": true, "delta_db": -1.3394393583239168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3309723892934117e-06, "output_power": 3.4620905160609874e-06, "reference_active": true, "delta_db": 1.718012890101515}], "active_band_count": 3, "active_band_mae_db": 1.2856262041327806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 8.985216395929287e-05, "log_rms_mae_db": 1.198531916886735, "rms_envelope_nrmse": 0.10675007331541224, "envelope_correlation": 0.9924124082046846, "rms_gain_error_db": -0.28443282066153064, "reference_mean_power": 0.00030936919363315046, "output_mean_power": 0.00028975689775214065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 9.09439257266203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.275291186820264e-05, "output_power": 5.2025089637055744e-05, "reference_active": true, "delta_db": -0.8142104565013704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.365294966430701e-06, "output_power": 3.3603966885261785e-06, "reference_active": true, "delta_db": -1.1362304658932973}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9190263566647066e-07, "output_power": 2.188475424524987e-07, "reference_active": true, "delta_db": 0.5706073452046715}], "active_band_count": 3, "active_band_mae_db": 0.8403494225331131}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.809631559896483e-05, "log_rms_mae_db": 1.9667504426691431, "rms_envelope_nrmse": 0.2397171276060397, "envelope_correlation": 0.9613099448302318, "rms_gain_error_db": -0.24642687246436273, "reference_mean_power": 6.929937891252573e-05, "output_mean_power": 6.547668070954355e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.858105660801035e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.988342411251772e-05, "output_power": 3.287487649824125e-05, "reference_active": true, "delta_db": -0.8392830675686489}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1366606898498195e-05, "output_power": 8.851970193607893e-06, "reference_active": true, "delta_db": -1.0859089776161097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.736645328964402e-06, "output_power": 2.5175175824324087e-06, "reference_active": true, "delta_db": 1.6126137981326578}], "active_band_count": 3, "active_band_mae_db": 1.1792686144391389}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.651257368418169e-05, "log_rms_mae_db": 1.6272437315752286, "rms_envelope_nrmse": 0.19183287160419796, "envelope_correlation": 0.9744354661711991, "rms_gain_error_db": -0.24642687246436326, "reference_mean_power": 6.985899704640885e-05, "output_mean_power": 6.60054291405762e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.431140169416232e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0602300645347785e-05, "output_power": 3.138821819080633e-05, "reference_active": true, "delta_db": -1.1178397978754284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067895142462218e-06, "output_power": 5.095595038335849e-06, "reference_active": true, "delta_db": -0.7584316120132574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.231130378903163e-07, "output_power": 1.7038804110669795e-07, "reference_active": true, "delta_db": 1.4113506208317907}], "active_band_count": 3, "active_band_mae_db": 1.095874010240159}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9086065356532813, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "138314d331a74c13730ff3dedbf5c4dca5bb22d5e5add274053ba020baea77bf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9813669289648321, "standard_deviation": 0.0030565884231550508, "minimum": 0.9765684328272484, "maximum": 0.9855618159508033, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7133215456499722, "standard_deviation": 0.057481678285474805, "minimum": 1.5991011723546513, "maximum": 1.7713665631683442, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1637800692296216, "standard_deviation": 0.2865034369938035, "minimum": 0.6907374293610706, "maximum": 1.6017844718563186, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16762815203848008, "standard_deviation": 0.013508605807411778, "minimum": 0.14801273941669665, "maximum": 0.18870707443582613, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927192644712588, "standard_deviation": 0.0016807654481027568, "minimum": 0.9900699443474262, "maximum": 0.9952025010663588, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.153920895309923, "standard_deviation": 0.06665122635915473, "minimum": 1.035463438457759, "maximum": 1.2431060818076827, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8985018544015424, "standard_deviation": 0.23836394831229965, "minimum": 0.4899779415991536, "maximum": 1.1857307027426514, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10783440586445181, "standard_deviation": 0.012453474846712756, "minimum": 0.08904859070744318, "maximum": 0.1273808678395438, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9645264202173943, "standard_deviation": 0.0032884854795952115, "minimum": 0.9599200938072415, "maximum": 0.9705427886884672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9668831103231588, "standard_deviation": 0.02991665759159095, "minimum": 1.9067008170486612, "maximum": 1.9984198327679696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9967123742311786, "standard_deviation": 0.306126996967504, "minimum": 0.6890683361543674, "maximum": 1.5177954453736036, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23017076148989854, "standard_deviation": 0.01004398287914429, "minimum": 0.21089948263409994, "maximum": 0.24377246127160074, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.976662487426139, "standard_deviation": 0.0018440105304961343, "minimum": 0.9744354661711991, "maximum": 0.9795648141440966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5974257061525916, "standard_deviation": 0.03460471847181866, "minimum": 1.5425485762956137, "maximum": 1.628618200361429, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0087954279242632, "standard_deviation": 0.16865540963210632, "minimum": 0.8063148329627431, "maximum": 1.3182467113100387, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18341937922354562, "standard_deviation": 0.006960605254769473, "minimum": 0.17167624774191514, "maximum": 0.19183287160419796, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.988734619692957, "standard_deviation": 0.0013544931231324183, "minimum": 0.9869245347332897, "maximum": 0.9905284860738135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6136377746309156, "standard_deviation": 0.05358262269452177, "minimum": 1.5516732510791447, "maximum": 1.7109173361330854, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6482293603520228, "standard_deviation": 0.23499943169517734, "minimum": 0.3424645016623451, "maximum": 1.0711354287013515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11840893457730942, "standard_deviation": 0.007086323293873811, "minimum": 0.10975748296233082, "maximum": 0.12838792654671885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958783239963722, "standard_deviation": 0.0010193282958631277, "minimum": 0.9942566931945938, "maximum": 0.9971202541171986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1528438125242815, "standard_deviation": 0.049701206128836835, "minimum": 1.0671200475613611, "maximum": 1.2151905912557832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9117640356287033, "standard_deviation": 0.31139000803205075, "minimum": 0.48007678127850023, "maximum": 1.4880004790422747, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07229971365591763, "standard_deviation": 0.009939487851614203, "minimum": 0.06089280650368966, "maximum": 0.08973858977843328, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2649487294699875, "standard_deviation": 0.07688463008507573, "minimum": 0.13279674045144682, "maximum": 0.3970652135514776, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.013715608558446, "standard_deviation": 0.013232809990316938, "minimum": 4.984931283536493, "maximum": 5.025467447946025, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47221322319418024, "standard_deviation": 0.022968159484502187, "minimum": 0.44656673082195575, "maximum": 0.5177706580730052, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08915674757446843, "standard_deviation": 0.07135976162180871, "minimum": -0.19895886991327594, "maximum": 0.032632432974694384, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9937531207192557, "standard_deviation": 0.0010134752998597354, "minimum": 0.9923520665657278, "maximum": 0.9951018517808974, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6545970922418576, "standard_deviation": 0.026164423719468057, "minimum": 0.6201419816550535, "maximum": 0.7000438792794653, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34140078553930986, "standard_deviation": 0.18360231681252429, "minimum": 0.10179969403527844, "maximum": 0.6197156930773663, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07463361627138734, "standard_deviation": 0.0062420304919733, "minimum": 0.06628793719837209, "maximum": 0.08305280731510677, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980270283163049, "standard_deviation": 0.0005724011236604835, "minimum": 0.9970024366652623, "maximum": 0.9986164102075452, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3750645802809922, "standard_deviation": 0.04216259582552177, "minimum": 0.3404970214235505, "maximum": 0.4693533649804597, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2675022381940383, "standard_deviation": 0.1348626767311812, "minimum": 0.11295268097094664, "maximum": 0.5067474456567835, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041444974339569896, "standard_deviation": 0.006550089861099952, "minimum": 0.03556101284870169, "maximum": 0.05194812821072675, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0041", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.1315624713897705, "predicted_mos": 2.1315624713897705, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "15ded1a85433abafbe6f3733dd8afdfda0ca725fe546269621cd2ac835002da3", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.481470699887723, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6904469399350384, "全帯域_10ms_rms_envelope_nrmse": 0.07307111125523598, "全帯域_10ms_envelope_correlation": 0.9939828906552126, "全帯域_10ms_modulation_mae_db": 0.23280230183073317, "全帯域_40ms_log_rms_mae_db": 0.39350433604707663, "全帯域_40ms_rms_envelope_nrmse": 0.03828495468355469, "全帯域_40ms_envelope_correlation": 0.9983214128894201, "全帯域_40ms_modulation_mae_db": 0.31176101716926224, "300–1000 Hz_10ms_log_rms_mae_db": 1.3206376259837966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11351298676846079, "300–1000 Hz_10ms_envelope_correlation": 0.9899292581584005, "300–1000 Hz_10ms_modulation_mae_db": 0.585338920037525, "300–1000 Hz_40ms_log_rms_mae_db": 0.9326500031854785, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07241019058781883, "300–1000 Hz_40ms_envelope_correlation": 0.9961935311928994, "300–1000 Hz_40ms_modulation_mae_db": 0.8812414855599368, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2977355714793428, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1541237563805572, "1000–3000 Hz_10ms_envelope_correlation": 0.9840784683668249, "1000–3000 Hz_10ms_modulation_mae_db": 0.8412684807672193, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8692426093476301, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09248818857337118, "1000–3000 Hz_40ms_envelope_correlation": 0.9943805040114483, "1000–3000 Hz_40ms_modulation_mae_db": 0.5160544148554761, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3574862608784866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22195466332281308, "3000–6000 Hz_10ms_envelope_correlation": 0.9675070922487262, "3000–6000 Hz_10ms_modulation_mae_db": 0.7358074848924628, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2863932986859614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1799863112726294, "3000–6000 Hz_40ms_envelope_correlation": 0.9774443763308345, "3000–6000 Hz_40ms_modulation_mae_db": 0.902980633985396, "mr_stft_db": 4.654218475176699, "band_mae_db": 0.17267897925862308, "rms_gain_error_db": -0.08998405363719969, "output_peak": 0.4576714656298787}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6798778645250199, "全帯域_10ms_rms_envelope_nrmse": 0.07366956565116865, "全帯域_10ms_envelope_correlation": 0.9940285215584491, "全帯域_10ms_modulation_mae_db": 0.5044201626703732, "全帯域_40ms_log_rms_mae_db": 0.42678876502582397, "全帯域_40ms_rms_envelope_nrmse": 0.04423434317790022, "全帯域_40ms_envelope_correlation": 0.9978959139676052, "全帯域_40ms_modulation_mae_db": 0.2602868308084031, "300–1000 Hz_10ms_log_rms_mae_db": 1.2753834832149706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11284649297465482, "300–1000 Hz_10ms_envelope_correlation": 0.9902307813879914, "300–1000 Hz_10ms_modulation_mae_db": 0.7763886318437043, "300–1000 Hz_40ms_log_rms_mae_db": 0.9143820570576844, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07320996050521576, "300–1000 Hz_40ms_envelope_correlation": 0.9963038293692116, "300–1000 Hz_40ms_modulation_mae_db": 0.9037158566560599, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3826682840314009, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1495954217522222, "1000–3000 Hz_10ms_envelope_correlation": 0.9857538524415972, "1000–3000 Hz_10ms_modulation_mae_db": 0.32835952097731985, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9510801286849279, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09839643922355402, "1000–3000 Hz_40ms_envelope_correlation": 0.9945668342494408, "1000–3000 Hz_40ms_modulation_mae_db": 0.4162116564001997, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3836240357688132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22044926611737645, "3000–6000 Hz_10ms_envelope_correlation": 0.9682187050391644, "3000–6000 Hz_10ms_modulation_mae_db": 0.9775858192042411, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2486598211406863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17136545379959325, "3000–6000 Hz_40ms_envelope_correlation": 0.9797398813744208, "3000–6000 Hz_40ms_modulation_mae_db": 1.0478798999134202, "mr_stft_db": 4.668376798220426, "band_mae_db": 0.3103461570466121, "rms_gain_error_db": -0.14257030818442185, "output_peak": 0.4583501901858671}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6455016598796126, "全帯域_10ms_rms_envelope_nrmse": 0.0788274104810062, "全帯域_10ms_envelope_correlation": 0.9931762362261372, "全帯域_10ms_modulation_mae_db": 0.2570224038594197, "全帯域_40ms_log_rms_mae_db": 0.37188572390143393, "全帯域_40ms_rms_envelope_nrmse": 0.048372962775394934, "全帯域_40ms_envelope_correlation": 0.9974955635333627, "全帯域_40ms_modulation_mae_db": 0.25349469996657786, "300–1000 Hz_10ms_log_rms_mae_db": 1.2070411511703967, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12283400195400151, "300–1000 Hz_10ms_envelope_correlation": 0.9881197281038, "300–1000 Hz_10ms_modulation_mae_db": 0.4785758717054458, "300–1000 Hz_40ms_log_rms_mae_db": 0.8432485651111588, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0833726593508759, "300–1000 Hz_40ms_envelope_correlation": 0.9947253350449294, "300–1000 Hz_40ms_modulation_mae_db": 0.754311299449907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3404221232891815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14024980994438765, "1000–3000 Hz_10ms_envelope_correlation": 0.9883800961206345, "1000–3000 Hz_10ms_modulation_mae_db": 1.064880334835943, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8828915203468454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09853749006379503, "1000–3000 Hz_40ms_envelope_correlation": 0.995310534589514, "1000–3000 Hz_40ms_modulation_mae_db": 1.3619485160451161, "3000–6000 Hz_10ms_log_rms_mae_db": 1.377048117567034, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2245448419649521, "3000–6000 Hz_10ms_envelope_correlation": 0.9661271553520389, "3000–6000 Hz_10ms_modulation_mae_db": 0.627258944549953, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2090542147871775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1758388381491303, "3000–6000 Hz_40ms_envelope_correlation": 0.9783860921879654, "3000–6000 Hz_40ms_modulation_mae_db": 0.8852030895787409, "mr_stft_db": 4.6684362549367355, "band_mae_db": 0.29685317155268925, "rms_gain_error_db": -0.15775422437246717, "output_peak": 0.5136300582275206}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6631880294145578, "全帯域_10ms_rms_envelope_nrmse": 0.08006544324151632, "全帯域_10ms_envelope_correlation": 0.9927684399175235, "全帯域_10ms_modulation_mae_db": 0.8001087450689938, "全帯域_40ms_log_rms_mae_db": 0.37091541783400384, "全帯域_40ms_rms_envelope_nrmse": 0.03898369477601384, "全帯域_40ms_envelope_correlation": 0.9982702802281657, "全帯域_40ms_modulation_mae_db": 0.4972357790256474, "300–1000 Hz_10ms_log_rms_mae_db": 1.3046308611734574, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12979773033890446, "300–1000 Hz_10ms_envelope_correlation": 0.9864347649207127, "300–1000 Hz_10ms_modulation_mae_db": 1.0739730713332405, "300–1000 Hz_40ms_log_rms_mae_db": 0.969833786615883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07646452973535671, "300–1000 Hz_40ms_envelope_correlation": 0.9955196411016027, "300–1000 Hz_40ms_modulation_mae_db": 0.4665499303193039, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3222574443692086, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15225711368322692, "1000–3000 Hz_10ms_envelope_correlation": 0.9849304018377073, "1000–3000 Hz_10ms_modulation_mae_db": 1.2152956563963662, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8573457215480008, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09716626111688041, "1000–3000 Hz_40ms_envelope_correlation": 0.994445341899904, "1000–3000 Hz_40ms_modulation_mae_db": 0.5818159427415266, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4789463381629802, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22207373867189512, "3000–6000 Hz_10ms_envelope_correlation": 0.9668767178075682, "3000–6000 Hz_10ms_modulation_mae_db": 0.924780005856303, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3449748409178288, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18079018140871836, "3000–6000 Hz_40ms_envelope_correlation": 0.9770787932537377, "3000–6000 Hz_40ms_modulation_mae_db": 1.1396067780651578, "mr_stft_db": 4.6795004199654775, "band_mae_db": 0.20475303512404433, "rms_gain_error_db": -0.09649971032871518, "output_peak": 0.4815250350254572}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7337865036270845, "全帯域_10ms_rms_envelope_nrmse": 0.0753712096355003, "全帯域_10ms_envelope_correlation": 0.9935801679470035, "全帯域_10ms_modulation_mae_db": 0.5208609038636798, "全帯域_40ms_log_rms_mae_db": 0.45532453865968575, "全帯域_40ms_rms_envelope_nrmse": 0.036883979987135725, "全帯域_40ms_envelope_correlation": 0.9983559447112728, "全帯域_40ms_modulation_mae_db": 0.39906829073552225, "300–1000 Hz_10ms_log_rms_mae_db": 1.2207899323248421, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11018911294222096, "300–1000 Hz_10ms_envelope_correlation": 0.9899420229933331, "300–1000 Hz_10ms_modulation_mae_db": 0.9132375608947073, "300–1000 Hz_40ms_log_rms_mae_db": 0.8857089858162328, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058208790050549075, "300–1000 Hz_40ms_envelope_correlation": 0.996982778000529, "300–1000 Hz_40ms_modulation_mae_db": 0.46428625721382205, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3834059478865666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1392652450005681, "1000–3000 Hz_10ms_envelope_correlation": 0.9875524967699025, "1000–3000 Hz_10ms_modulation_mae_db": 1.268704994098566, "1000–3000 Hz_40ms_log_rms_mae_db": 0.925109668294899, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0962568414493164, "1000–3000 Hz_40ms_envelope_correlation": 0.9944304141435909, "1000–3000 Hz_40ms_modulation_mae_db": 1.2954917193706024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3739176312604378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21798674649739566, "3000–6000 Hz_10ms_envelope_correlation": 0.9693513305731315, "3000–6000 Hz_10ms_modulation_mae_db": 0.7399052820216466, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2392569380168446, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17693271312036582, "3000–6000 Hz_40ms_envelope_correlation": 0.9788026279550732, "3000–6000 Hz_40ms_modulation_mae_db": 1.1067808427539738, "mr_stft_db": 4.6679501513448285, "band_mae_db": 0.2829977377212984, "rms_gain_error_db": -0.01979303429273591, "output_peak": 0.48945725149928093}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6534255362936451, "全帯域_10ms_rms_envelope_nrmse": 0.07724988055858832, "全帯域_10ms_envelope_correlation": 0.9932889033081879, "全帯域_10ms_modulation_mae_db": 0.3812659008265233, "全帯域_40ms_log_rms_mae_db": 0.3659757661671226, "全帯域_40ms_rms_envelope_nrmse": 0.038126551438612453, "全帯域_40ms_envelope_correlation": 0.9983855561823052, "全帯域_40ms_modulation_mae_db": 0.11934132456388084, "300–1000 Hz_10ms_log_rms_mae_db": 1.3079249786739093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12594202557595832, "300–1000 Hz_10ms_envelope_correlation": 0.9870348563390196, "300–1000 Hz_10ms_modulation_mae_db": 0.7476252527930128, "300–1000 Hz_40ms_log_rms_mae_db": 0.9132540483415477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0719942756068257, "300–1000 Hz_40ms_envelope_correlation": 0.9958111897788465, "300–1000 Hz_40ms_modulation_mae_db": 0.4286475289193552, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3596897086633009, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1450439348430552, "1000–3000 Hz_10ms_envelope_correlation": 0.9862486379903872, "1000–3000 Hz_10ms_modulation_mae_db": 1.1138422960491159, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8879064327679116, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0899318009460264, "1000–3000 Hz_40ms_envelope_correlation": 0.995089776053158, "1000–3000 Hz_40ms_modulation_mae_db": 0.5540953759386684, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3920295574903676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21962360053473845, "3000–6000 Hz_10ms_envelope_correlation": 0.967879177543285, "3000–6000 Hz_10ms_modulation_mae_db": 0.6196860750108373, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2690883833815179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17922580446155065, "3000–6000 Hz_40ms_envelope_correlation": 0.9774717616543057, "3000–6000 Hz_40ms_modulation_mae_db": 0.8662982497143269, "mr_stft_db": 4.670579403877906, "band_mae_db": 0.26254644672534355, "rms_gain_error_db": -0.10361389899579729, "output_peak": 0.5987229435715782}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6725362641782779, "全帯域_10ms_rms_envelope_nrmse": 0.06760624696748006, "全帯域_10ms_envelope_correlation": 0.9948074081586786, "全帯域_10ms_modulation_mae_db": 0.4124164478074936, "全帯域_40ms_log_rms_mae_db": 0.39018806167476033, "全帯域_40ms_rms_envelope_nrmse": 0.033456422541242974, "全帯域_40ms_envelope_correlation": 0.9986708185215072, "全帯域_40ms_modulation_mae_db": 0.3724676171039949, "300–1000 Hz_10ms_log_rms_mae_db": 1.2937721116416956, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10701116008584438, "300–1000 Hz_10ms_envelope_correlation": 0.9906316003146558, "300–1000 Hz_10ms_modulation_mae_db": 0.5861417607718914, "300–1000 Hz_40ms_log_rms_mae_db": 0.9066109736238425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06440240625016545, "300–1000 Hz_40ms_envelope_correlation": 0.996534877721037, "300–1000 Hz_40ms_modulation_mae_db": 0.8119586492923991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3377489361967663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1484857499491284, "1000–3000 Hz_10ms_envelope_correlation": 0.9863502721229304, "1000–3000 Hz_10ms_modulation_mae_db": 1.131603028164388, "1000–3000 Hz_40ms_log_rms_mae_db": 0.872900330679359, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0899561885900271, "1000–3000 Hz_40ms_envelope_correlation": 0.9961894150835936, "1000–3000 Hz_40ms_modulation_mae_db": 0.627688915783362, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4136098933625967, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23339028797018696, "3000–6000 Hz_10ms_envelope_correlation": 0.9659305810384734, "3000–6000 Hz_10ms_modulation_mae_db": 1.6406895055950728, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2815440858822063, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18574425184387075, "3000–6000 Hz_40ms_envelope_correlation": 0.977388723673704, "3000–6000 Hz_40ms_modulation_mae_db": 1.786802471245173, "mr_stft_db": 4.6609622572034946, "band_mae_db": 0.17910784160430693, "rms_gain_error_db": -0.04954496847871018, "output_peak": 0.4581061416070739}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6661488412639586, "全帯域_10ms_rms_envelope_nrmse": 0.0780314504430898, "全帯域_10ms_envelope_correlation": 0.9931711567473597, "全帯域_10ms_modulation_mae_db": 0.20155779638927696, "全帯域_40ms_log_rms_mae_db": 0.3835616588918374, "全帯域_40ms_rms_envelope_nrmse": 0.04129960946440262, "全帯域_40ms_envelope_correlation": 0.9980113020197299, "全帯域_40ms_modulation_mae_db": 0.2945134498678277, "300–1000 Hz_10ms_log_rms_mae_db": 1.214645184730207, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12576344485575566, "300–1000 Hz_10ms_envelope_correlation": 0.9869579754608316, "300–1000 Hz_10ms_modulation_mae_db": 0.7982263760866278, "300–1000 Hz_40ms_log_rms_mae_db": 0.924965046662035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08863318505524953, "300–1000 Hz_40ms_envelope_correlation": 0.9931107084274773, "300–1000 Hz_40ms_modulation_mae_db": 1.1098071019707023, "1000–3000 Hz_10ms_log_rms_mae_db": 1.272529087841668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13464379646133434, "1000–3000 Hz_10ms_envelope_correlation": 0.9890788874470247, "1000–3000 Hz_10ms_modulation_mae_db": 0.9429544736276827, "1000–3000 Hz_40ms_log_rms_mae_db": 0.857868521887273, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09052970710751956, "1000–3000 Hz_40ms_envelope_correlation": 0.9960035629819715, "1000–3000 Hz_40ms_modulation_mae_db": 1.5332241490069005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4202008451236017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21437903785341392, "3000–6000 Hz_10ms_envelope_correlation": 0.9692045490888769, "3000–6000 Hz_10ms_modulation_mae_db": 0.8637206119383459, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2879926528136345, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17352825288416932, "3000–6000 Hz_40ms_envelope_correlation": 0.9790491871179801, "3000–6000 Hz_40ms_modulation_mae_db": 0.982162353140839, "mr_stft_db": 4.667392407455453, "band_mae_db": 0.2649439932765737, "rms_gain_error_db": -0.0709436495002232, "output_peak": 0.4947191309569322}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.70715333892951, "spectral_convergence": 0.22725664512318564, "spectral_convergence_status": "ok", "amplitude_floor": 1.419137961520426e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.3903439142525365, "spectral_convergence": 0.2196543800252866, "spectral_convergence_status": "ok", "amplitude_floor": 1.2001614819603722e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.86515817234805, "spectral_convergence": 0.2293555712829778, "spectral_convergence_status": "ok", "amplitude_floor": 7.977266337453151e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015734976003854988, "output_power": 0.0016410576568525864, "delta_db": 0.18257754863510084, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002272380907829781, "output_power": 0.0021242569846472072, "delta_db": -0.2927407684366642, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00035321559220182435, "output_power": 0.0003540340630013354, "delta_db": 0.010051825628856136, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024119869501752642, "output_power": 0.00022804352370855744, "delta_db": -0.2435721070146497, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012307274117421543, "output_power": 0.00011984526582332443, "delta_db": -0.11540990627695168, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.88440946982043e-05, "output_power": 4.059730235678336e-05, "delta_db": 0.19172171955951595, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17267897925862308, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08998405363719969, "pitch": null, "output_rms_dbfs": -23.460320084047652, "output_peak": 0.4576714656298787, "reference_rms_dbfs": -23.37033603041045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.654218475176699}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00019859470734653442, "log_rms_mae_db": 0.6904469399350384, "rms_envelope_nrmse": 0.07307111125523598, "envelope_correlation": 0.9939828906552126, "rms_gain_error_db": -0.08998405363716538, "reference_mean_power": 0.004608354103444978, "output_mean_power": 0.004513853267413048}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.3747785986813366e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009662478923264104, "output_power": 0.0009209921150949132, "reference_active": true, "delta_db": -0.20832647521975559}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.594410683545364e-05, "output_power": 8.922215403380572e-05, "reference_active": true, "delta_db": -0.315456003394412}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7446486319031935e-05, "output_power": 1.675890084860112e-05, "reference_active": true, "delta_db": -0.17462442687803198}], "active_band_count": 3, "active_band_mae_db": 0.23280230183073317}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00016219647203140816, "log_rms_mae_db": 0.39350433604707663, "rms_envelope_nrmse": 0.03828495468355469, "envelope_correlation": 0.9983214128894201, "rms_gain_error_db": -0.08998405363716488, "reference_mean_power": 0.004645568268479526, "output_mean_power": 0.004550304303219883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1771130170213754e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008502808342452142, "output_power": 0.0008070394990654283, "reference_active": true, "delta_db": -0.2266759898459531}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.133174041000048e-05, "output_power": 3.902469680059553e-05, "reference_active": true, "delta_db": -0.24944155546458546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2770439378674992e-06, "output_power": 1.1489186763292624e-06, "reference_active": true, "delta_db": -0.4591655061972482}], "active_band_count": 3, "active_band_mae_db": 0.31176101716926224}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015290889340024391, "log_rms_mae_db": 1.3206376259837966, "rms_envelope_nrmse": 0.11351298676846079, "envelope_correlation": 0.9899292581584005, "rms_gain_error_db": -0.29583897040101775, "reference_mean_power": 0.0016143224036409894, "output_mean_power": 0.0015080174844881234}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 8.858385684123232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006875363814202012, "output_power": 0.0006119467115618814, "reference_active": true, "delta_db": -0.5058207873158462}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.173936122300265e-05, "output_power": 4.6908538828425535e-05, "reference_active": true, "delta_db": -1.1931022664988373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0160658568889886e-05, "output_power": 1.0027957559825613e-05, "reference_active": true, "delta_db": -0.0570937062978913}], "active_band_count": 3, "active_band_mae_db": 0.585338920037525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012027897252950328, "log_rms_mae_db": 0.9326500031854785, "rms_envelope_nrmse": 0.07241019058781883, "envelope_correlation": 0.9961935311928994, "rms_gain_error_db": -0.29583897040101675, "reference_mean_power": 0.0016273586545452231, "output_mean_power": 0.0015201952838245012}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.855745033342305e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006306037982733972, "output_power": 0.0005576208071470073, "reference_active": true, "delta_db": -0.5341761155428646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.290965975606541e-05, "output_power": 1.7968697683287202e-05, "reference_active": true, "delta_db": -1.0550203738886221}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0400487875049354e-06, "output_power": 8.15832336881538e-07, "reference_active": true, "delta_db": -1.0545279672483232}], "active_band_count": 3, "active_band_mae_db": 0.8812414855599368}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.957022642296078e-05, "log_rms_mae_db": 1.2977355714793428, "rms_envelope_nrmse": 0.1541237563805572, "envelope_correlation": 0.9840784683668249, "rms_gain_error_db": -0.25705985165550455, "reference_mean_power": 0.000306890935740228, "output_mean_power": 0.0002892531360586165}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.0486708619890815e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.911544017889007e-05, "output_power": 6.220480565372794e-05, "reference_active": true, "delta_db": -0.4575114072072251}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0167211042917879e-05, "output_power": 8.333551920849036e-06, "reference_active": true, "delta_db": -0.8637169282644734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3309723892934117e-06, "output_power": 3.074644014500428e-06, "reference_active": true, "delta_db": 1.2025771068299593}], "active_band_count": 3, "active_band_mae_db": 0.8412684807672193}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 8.985216395929287e-05, "log_rms_mae_db": 0.8692426093476301, "rms_envelope_nrmse": 0.09248818857337118, "envelope_correlation": 0.9943805040114483, "rms_gain_error_db": -0.25705985165550455, "reference_mean_power": 0.00030936919363315046, "output_mean_power": 0.0002915889621909876}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 9.09439257266203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.275291186820264e-05, "output_power": 5.658515263540731e-05, "reference_active": true, "delta_db": -0.44931391094468703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.365294966430701e-06, "output_power": 3.6554405157425547e-06, "reference_active": true, "delta_db": -0.7707387354437966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9190263566647066e-07, "output_power": 2.0696266087001847e-07, "reference_active": true, "delta_db": 0.3281105981779448}], "active_band_count": 3, "active_band_mae_db": 0.5160544148554761}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.809631559896483e-05, "log_rms_mae_db": 1.3574862608784866, "rms_envelope_nrmse": 0.22195466332281308, "envelope_correlation": 0.9675070922487262, "rms_gain_error_db": 0.004232799443730445, "reference_mean_power": 6.929937891252573e-05, "output_mean_power": 6.936695365194081e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.858105660801035e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.988342411251772e-05, "output_power": 3.95475780144589e-05, "reference_active": true, "delta_db": -0.03672545289059146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1366606898498195e-05, "output_power": 1.2791836315246432e-05, "reference_active": true, "delta_db": 0.5130205286637061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.736645328964402e-06, "output_power": 2.5437754982457064e-06, "reference_active": true, "delta_db": 1.657676473123091}], "active_band_count": 3, "active_band_mae_db": 0.7358074848924628}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.651257368418169e-05, "log_rms_mae_db": 1.2863932986859614, "rms_envelope_nrmse": 0.1799863112726294, "envelope_correlation": 0.9774443763308345, "rms_gain_error_db": 0.004232799443729482, "reference_mean_power": 6.985899704640885e-05, "output_mean_power": 6.992711747685552e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.431140169416232e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0602300645347785e-05, "output_power": 3.885702855609284e-05, "reference_active": true, "delta_db": -0.1908105616259591}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067895142462218e-06, "output_power": 7.267398981671404e-06, "reference_active": true, "delta_db": 0.7834093644008217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.231130378903163e-07, "output_power": 1.835592377286565e-07, "reference_active": true, "delta_db": 1.734721975929407}], "active_band_count": 3, "active_band_mae_db": 0.902980633985396}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9233199730243344, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e57ca85fe4ac1889cdd2ab8f7ff603e402a96292acb253ab9c58d578635e4b44", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.986546639137126, "standard_deviation": 0.0017006536971120305, "minimum": 0.9840784683668249, "maximum": 0.9890788874470247, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3370571379696794, "standard_deviation": 0.03906351891143742, "minimum": 1.272529087841668, "maximum": 1.3834059478865666, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9883635981145751, "standard_deviation": 0.3003595584117218, "minimum": 0.32835952097731985, "maximum": 1.268704994098566, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14545810350181, "standard_deviation": 0.00686817417219631, "minimum": 0.13464379646133434, "maximum": 0.1541237563805572, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950520478765776, "standard_deviation": 0.0007278292600339875, "minimum": 0.9943805040114483, "maximum": 0.9961894150835936, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8880431166946059, "standard_deviation": 0.03340177720039127, "minimum": 0.8573457215480008, "maximum": 0.9510801286849279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8608163362677315, "standard_deviation": 0.4527708158829792, "minimum": 0.4162116564001997, "maximum": 1.5332241490069005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09415786463381126, "standard_deviation": 0.0038185230425175367, "minimum": 0.0899318009460264, "maximum": 0.09853749006379503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.967636913586408, "standard_deviation": 0.001286166601324524, "minimum": 0.9659305810384734, "maximum": 0.9693513305731315, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3996078349517898, "standard_deviation": 0.038117851726432526, "minimum": 1.3574862608784866, "maximum": 1.4789463381629802, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8911792161336078, "standard_deviation": 0.3298270924757904, "minimum": 0.6196860750108373, "maximum": 1.6406895055950728, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22180027286659648, "standard_deviation": 0.00558237742188927, "minimum": 0.21437903785341392, "maximum": 0.23339028797018696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9781701804435027, "standard_deviation": 0.0009635946523828983, "minimum": 0.9770787932537377, "maximum": 0.9797398813744208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2708705294532323, "standard_deviation": 0.040455733812705355, "minimum": 1.2090542147871775, "maximum": 1.3449748409178288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0897142897996286, "standard_deviation": 0.29967577998784595, "minimum": 0.8662982497143269, "maximum": 1.786802471245173, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1779264758675035, "standard_deviation": 0.004519588776528801, "minimum": 0.17136545379959325, "maximum": 0.18574425184387075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9886601234598431, "standard_deviation": 0.001706738971022301, "minimum": 0.9864347649207127, "maximum": 0.9906316003146558, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2681031661141593, "standard_deviation": 0.046624156907749806, "minimum": 1.2070411511703967, "maximum": 1.3206376259837966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7449384306832694, "standard_deviation": 0.1934874415516763, "minimum": 0.4785758717054458, "maximum": 1.0739730713332405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11848711943697511, "standard_deviation": 0.008556192922087908, "minimum": 0.10701116008584438, "maximum": 0.12979773033890446, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956477363295666, "standard_deviation": 0.0012321970125735299, "minimum": 0.9931107084274773, "maximum": 0.996982778000529, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9113316833017329, "standard_deviation": 0.03662542915907432, "minimum": 0.8432485651111588, "maximum": 0.969833786615883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7275647636726859, "standard_deviation": 0.24939597921181517, "minimum": 0.4286475289193552, "maximum": 1.1098071019707023, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07358699964275713, "standard_deviation": 0.009675720015960343, "minimum": 0.058208790050549075, "maximum": 0.08863318505524953, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24677842028868643, "standard_deviation": 0.053818428740042897, "minimum": 0.17267897925862308, "maximum": 0.3103461570466121, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.667177021022628, "standard_deviation": 0.007306150320620423, "minimum": 4.654218475176699, "maximum": 4.6795004199654775, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4940227770879486, "standard_deviation": 0.04691641325642316, "minimum": 0.4576714656298787, "maximum": 0.5987229435715782, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0913379809737838, "standard_deviation": 0.045507154094276164, "minimum": -0.15775422437246717, "maximum": -0.01979303429273591, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993600465564819, "standard_deviation": 0.0006480980538322623, "minimum": 0.9927684399175235, "maximum": 0.9948074081586786, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6756139548896494, "standard_deviation": 0.027450694000808667, "minimum": 0.6455016598796126, "maximum": 0.7337865036270845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41380683278956165, "standard_deviation": 0.19725075171872847, "minimum": 0.20155779638927696, "maximum": 0.8001087450689938, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07548653977919821, "standard_deviation": 0.0040171368754310975, "minimum": 0.06760624696748006, "maximum": 0.08006544324151632, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998175849006671, "standard_deviation": 0.00036247958458809595, "minimum": 0.9974955635333627, "maximum": 0.9986708185215072, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.39476803352521805, "standard_deviation": 0.031080805961413285, "minimum": 0.3659757661671226, "maximum": 0.45532453865968575, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31352112615513955, "standard_deviation": 0.11279015089460742, "minimum": 0.11934132456388084, "maximum": 0.4972357790256474, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03995531485553218, "standard_deviation": 0.004624797005375428, "minimum": 0.033456422541242974, "maximum": 0.048372962775394934, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0041", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2284398078918457, "predicted_mos": 2.2284398078918457, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "def4033d3025046d5d3f1ecc6c5edc2718b650d94fa859e9b4a0bd956686b060", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.282414000015706, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6144160200716067, "全帯域_10ms_rms_envelope_nrmse": 0.061563019829726257, "全帯域_10ms_envelope_correlation": 0.995736715962354, "全帯域_10ms_modulation_mae_db": 0.17548836412540506, "全帯域_40ms_log_rms_mae_db": 0.33778313453333747, "全帯域_40ms_rms_envelope_nrmse": 0.034424508618633894, "全帯域_40ms_envelope_correlation": 0.9986259489370308, "全帯域_40ms_modulation_mae_db": 0.1209212643961698, "300–1000 Hz_10ms_log_rms_mae_db": 1.389862666278159, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10802436801549023, "300–1000 Hz_10ms_envelope_correlation": 0.9905342004858879, "300–1000 Hz_10ms_modulation_mae_db": 0.7675773498040236, "300–1000 Hz_40ms_log_rms_mae_db": 0.9858211563907299, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07043667838943359, "300–1000 Hz_40ms_envelope_correlation": 0.9959026720645339, "300–1000 Hz_40ms_modulation_mae_db": 1.1002404955035685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4520378155715874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13794642124282977, "1000–3000 Hz_10ms_envelope_correlation": 0.9874299743655642, "1000–3000 Hz_10ms_modulation_mae_db": 0.8365143682308497, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9463749199392888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08474600602978141, "1000–3000 Hz_40ms_envelope_correlation": 0.9955108597262882, "1000–3000 Hz_40ms_modulation_mae_db": 0.8879246273725204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5122177847331377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21813574732324917, "3000–6000 Hz_10ms_envelope_correlation": 0.9681546942961491, "3000–6000 Hz_10ms_modulation_mae_db": 0.6872327325522946, "3000–6000 Hz_40ms_log_rms_mae_db": 1.279897865563536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1739432138607237, "3000–6000 Hz_40ms_envelope_correlation": 0.9788375056237565, "3000–6000 Hz_40ms_modulation_mae_db": 0.817246445272089, "mr_stft_db": 4.755872328424308, "band_mae_db": 0.2337850601778643, "rms_gain_error_db": -0.07460370376666854, "output_peak": 0.4877579612323477}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6673363602955606, "全帯域_10ms_rms_envelope_nrmse": 0.0741009372621703, "全帯域_10ms_envelope_correlation": 0.9940431577405139, "全帯域_10ms_modulation_mae_db": 0.25774982220713405, "全帯域_40ms_log_rms_mae_db": 0.36987114846238445, "全帯域_40ms_rms_envelope_nrmse": 0.044967133376763946, "全帯域_40ms_envelope_correlation": 0.9979058840919279, "全帯域_40ms_modulation_mae_db": 0.3688271715584612, "300–1000 Hz_10ms_log_rms_mae_db": 1.3672346612100494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1098718458896986, "300–1000 Hz_10ms_envelope_correlation": 0.9905199016649151, "300–1000 Hz_10ms_modulation_mae_db": 1.194441416266695, "300–1000 Hz_40ms_log_rms_mae_db": 0.9287790726611617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06963872144880488, "300–1000 Hz_40ms_envelope_correlation": 0.9964356754016152, "300–1000 Hz_40ms_modulation_mae_db": 1.388070901093145, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4382781701375433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17308568195578333, "1000–3000 Hz_10ms_envelope_correlation": 0.9820560293080629, "1000–3000 Hz_10ms_modulation_mae_db": 0.6251750963494073, "1000–3000 Hz_40ms_log_rms_mae_db": 0.978966075376095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11368484933563965, "1000–3000 Hz_40ms_envelope_correlation": 0.9942787288697864, "1000–3000 Hz_40ms_modulation_mae_db": 1.0773128670390881, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5424558716628989, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22799729795847606, "3000–6000 Hz_10ms_envelope_correlation": 0.9652365786862792, "3000–6000 Hz_10ms_modulation_mae_db": 0.7806625178588185, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3575699118573337, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17794631644462608, "3000–6000 Hz_40ms_envelope_correlation": 0.9777573593824825, "3000–6000 Hz_40ms_modulation_mae_db": 0.6601950581328949, "mr_stft_db": 4.772548026782493, "band_mae_db": 0.39535867197463154, "rms_gain_error_db": -0.16258011714472065, "output_peak": 0.46803469588403657}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6229214360023835, "全帯域_10ms_rms_envelope_nrmse": 0.06974161050642101, "全帯域_10ms_envelope_correlation": 0.9945131277863531, "全帯域_10ms_modulation_mae_db": 0.10249682056086525, "全帯域_40ms_log_rms_mae_db": 0.3628574757196351, "全帯域_40ms_rms_envelope_nrmse": 0.03932184805523449, "全帯域_40ms_envelope_correlation": 0.9982037551284727, "全帯域_40ms_modulation_mae_db": 0.2715366147976455, "300–1000 Hz_10ms_log_rms_mae_db": 1.3977681610718287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11218890337616791, "300–1000 Hz_10ms_envelope_correlation": 0.9897982152510302, "300–1000 Hz_10ms_modulation_mae_db": 0.43395789810239194, "300–1000 Hz_40ms_log_rms_mae_db": 1.0349433299057802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06773337204147656, "300–1000 Hz_40ms_envelope_correlation": 0.9963128772464702, "300–1000 Hz_40ms_modulation_mae_db": 0.6085763177325476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4350608042060067, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15944967902057158, "1000–3000 Hz_10ms_envelope_correlation": 0.9838615604657827, "1000–3000 Hz_10ms_modulation_mae_db": 1.0789190748576523, "1000–3000 Hz_40ms_log_rms_mae_db": 0.955485513488203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10646095937683321, "1000–3000 Hz_40ms_envelope_correlation": 0.9937524378547246, "1000–3000 Hz_40ms_modulation_mae_db": 0.39661034322567995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5661759155538695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23062804072527998, "3000–6000 Hz_10ms_envelope_correlation": 0.9655813764559323, "3000–6000 Hz_10ms_modulation_mae_db": 0.957437261861279, "3000–6000 Hz_40ms_log_rms_mae_db": 1.324389366653793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18538851349422666, "3000–6000 Hz_40ms_envelope_correlation": 0.9766370669405512, "3000–6000 Hz_40ms_modulation_mae_db": 1.1166624160538605, "mr_stft_db": 4.755804001801676, "band_mae_db": 0.28435093925369465, "rms_gain_error_db": -0.08032749973232418, "output_peak": 0.4792541383202861}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.624719288321401, "全帯域_10ms_rms_envelope_nrmse": 0.06962574391165892, "全帯域_10ms_envelope_correlation": 0.994626236084257, "全帯域_10ms_modulation_mae_db": 0.238667646301106, "全帯域_40ms_log_rms_mae_db": 0.3737857621572658, "全帯域_40ms_rms_envelope_nrmse": 0.036231746358447975, "全帯域_40ms_envelope_correlation": 0.9985926937761175, "全帯域_40ms_modulation_mae_db": 0.22974108125813453, "300–1000 Hz_10ms_log_rms_mae_db": 1.3301997787129127, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10128666164267562, "300–1000 Hz_10ms_envelope_correlation": 0.9917356581975999, "300–1000 Hz_10ms_modulation_mae_db": 0.5331644827677331, "300–1000 Hz_40ms_log_rms_mae_db": 0.9071722174644071, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05227432727772697, "300–1000 Hz_40ms_envelope_correlation": 0.9979742945724467, "300–1000 Hz_40ms_modulation_mae_db": 0.5797240178886822, "1000–3000 Hz_10ms_log_rms_mae_db": 1.372548956175876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15414582515547895, "1000–3000 Hz_10ms_envelope_correlation": 0.9860779929957489, "1000–3000 Hz_10ms_modulation_mae_db": 1.5061841398217741, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9583471773736594, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10615904985175245, "1000–3000 Hz_40ms_envelope_correlation": 0.9949648039884004, "1000–3000 Hz_40ms_modulation_mae_db": 0.989845233530834, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5682274557716398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22872045467599716, "3000–6000 Hz_10ms_envelope_correlation": 0.9648363085700309, "3000–6000 Hz_10ms_modulation_mae_db": 0.7641607260904033, "3000–6000 Hz_40ms_log_rms_mae_db": 1.325453536067303, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17972974414658388, "3000–6000 Hz_40ms_envelope_correlation": 0.9776003809228371, "3000–6000 Hz_40ms_modulation_mae_db": 1.0739751987629818, "mr_stft_db": 4.759290386507042, "band_mae_db": 0.28550527040655216, "rms_gain_error_db": -0.1169121851120923, "output_peak": 0.5261608872035687}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6813767092975447, "全帯域_10ms_rms_envelope_nrmse": 0.06981242907912887, "全帯域_10ms_envelope_correlation": 0.9945704025380554, "全帯域_10ms_modulation_mae_db": 0.26381529673863796, "全帯域_40ms_log_rms_mae_db": 0.38203613100308403, "全帯域_40ms_rms_envelope_nrmse": 0.035119907248009244, "全帯域_40ms_envelope_correlation": 0.9986526745346973, "全帯域_40ms_modulation_mae_db": 0.36646602298574965, "300–1000 Hz_10ms_log_rms_mae_db": 1.3642818133256898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11841594172343016, "300–1000 Hz_10ms_envelope_correlation": 0.9885806748702051, "300–1000 Hz_10ms_modulation_mae_db": 0.559290543008853, "300–1000 Hz_40ms_log_rms_mae_db": 0.9717172311014644, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07210699320246616, "300–1000 Hz_40ms_envelope_correlation": 0.9957353857692245, "300–1000 Hz_40ms_modulation_mae_db": 0.4121039044887865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.557126878296821, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16080007401033924, "1000–3000 Hz_10ms_envelope_correlation": 0.9828718144486988, "1000–3000 Hz_10ms_modulation_mae_db": 1.2850321846419028, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0390358957195083, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09951080904717714, "1000–3000 Hz_40ms_envelope_correlation": 0.9937543603350084, "1000–3000 Hz_40ms_modulation_mae_db": 0.7260097476700255, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6228131067976803, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23594226828270837, "3000–6000 Hz_10ms_envelope_correlation": 0.962519511287709, "3000–6000 Hz_10ms_modulation_mae_db": 0.9202629507912231, "3000–6000 Hz_40ms_log_rms_mae_db": 1.370697879802445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19121993620049962, "3000–6000 Hz_40ms_envelope_correlation": 0.9743743197592775, "3000–6000 Hz_40ms_modulation_mae_db": 1.3316465568796312, "mr_stft_db": 4.779451169187933, "band_mae_db": 0.22960866084282328, "rms_gain_error_db": -0.10570686242810211, "output_peak": 0.49432874341996014}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6180809318953523, "全帯域_10ms_rms_envelope_nrmse": 0.0640041963534917, "全帯域_10ms_envelope_correlation": 0.9953502632608295, "全帯域_10ms_modulation_mae_db": 0.18825177612133961, "全帯域_40ms_log_rms_mae_db": 0.3580750016933853, "全帯域_40ms_rms_envelope_nrmse": 0.030272337118242208, "全帯域_40ms_envelope_correlation": 0.9989122582881472, "全帯域_40ms_modulation_mae_db": 0.10781173497511883, "300–1000 Hz_10ms_log_rms_mae_db": 1.3751260147101434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09475335175030404, "300–1000 Hz_10ms_envelope_correlation": 0.9927072225248978, "300–1000 Hz_10ms_modulation_mae_db": 0.5761537160521973, "300–1000 Hz_40ms_log_rms_mae_db": 0.993240211532359, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050383871212006, "300–1000 Hz_40ms_envelope_correlation": 0.9979879374968734, "300–1000 Hz_40ms_modulation_mae_db": 0.8622618292461027, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4561541901906443, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14142840271344245, "1000–3000 Hz_10ms_envelope_correlation": 0.9880076526432657, "1000–3000 Hz_10ms_modulation_mae_db": 1.0328379971029158, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0187526559316706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10016480229496312, "1000–3000 Hz_40ms_envelope_correlation": 0.9949376576417853, "1000–3000 Hz_40ms_modulation_mae_db": 1.2061383298579689, "3000–6000 Hz_10ms_log_rms_mae_db": 1.567437508304456, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22109788896299853, "3000–6000 Hz_10ms_envelope_correlation": 0.967164465247021, "3000–6000 Hz_10ms_modulation_mae_db": 0.8841658672147276, "3000–6000 Hz_40ms_log_rms_mae_db": 1.398582767356309, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1808333031801942, "3000–6000 Hz_40ms_envelope_correlation": 0.9770862003943306, "3000–6000 Hz_40ms_modulation_mae_db": 0.8960936013968186, "mr_stft_db": 4.759628433313407, "band_mae_db": 0.22624411227384436, "rms_gain_error_db": -0.04894078192605304, "output_peak": 0.5271779335757606}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6415838233084894, "全帯域_10ms_rms_envelope_nrmse": 0.07341286004888165, "全帯域_10ms_envelope_correlation": 0.993869164583716, "全帯域_10ms_modulation_mae_db": 0.8347896758138725, "全帯域_40ms_log_rms_mae_db": 0.3840343715291635, "全帯域_40ms_rms_envelope_nrmse": 0.03660682312347483, "全帯域_40ms_envelope_correlation": 0.9983610935779811, "全帯域_40ms_modulation_mae_db": 0.5821953050997813, "300–1000 Hz_10ms_log_rms_mae_db": 1.3679806049289356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1104077549979241, "300–1000 Hz_10ms_envelope_correlation": 0.9902852750452636, "300–1000 Hz_10ms_modulation_mae_db": 0.9281025186848041, "300–1000 Hz_40ms_log_rms_mae_db": 0.9652361497535147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06026188544492232, "300–1000 Hz_40ms_envelope_correlation": 0.9973979873097004, "300–1000 Hz_40ms_modulation_mae_db": 0.7598734467815049, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4398747617539862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19994679868070547, "1000–3000 Hz_10ms_envelope_correlation": 0.9731837451324198, "1000–3000 Hz_10ms_modulation_mae_db": 1.14297237514412, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9290784549259796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1240495694429242, "1000–3000 Hz_40ms_envelope_correlation": 0.9891740986442876, "1000–3000 Hz_40ms_modulation_mae_db": 0.8411939426780859, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6560205003216397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23625043329915085, "3000–6000 Hz_10ms_envelope_correlation": 0.962701077015159, "3000–6000 Hz_10ms_modulation_mae_db": 1.3202934360648098, "3000–6000 Hz_40ms_log_rms_mae_db": 1.406883111421142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18312294690072042, "3000–6000 Hz_40ms_envelope_correlation": 0.9765320082700454, "3000–6000 Hz_40ms_modulation_mae_db": 1.1485149288218277, "mr_stft_db": 4.769234725220301, "band_mae_db": 0.17446445948762554, "rms_gain_error_db": -0.02120601274383651, "output_peak": 0.48772212827547684}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6861098788177561, "全帯域_10ms_rms_envelope_nrmse": 0.07376175758475227, "全帯域_10ms_envelope_correlation": 0.9938495944954405, "全帯域_10ms_modulation_mae_db": 0.3786496331196189, "全帯域_40ms_log_rms_mae_db": 0.3797891881675319, "全帯域_40ms_rms_envelope_nrmse": 0.04155974962130391, "全帯域_40ms_envelope_correlation": 0.9979482419308707, "全帯域_40ms_modulation_mae_db": 0.25555829450733586, "300–1000 Hz_10ms_log_rms_mae_db": 1.4160732553998878, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11279443026997674, "300–1000 Hz_10ms_envelope_correlation": 0.9896002161812144, "300–1000 Hz_10ms_modulation_mae_db": 0.6693175416641535, "300–1000 Hz_40ms_log_rms_mae_db": 0.9762111715601314, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06449465664830337, "300–1000 Hz_40ms_envelope_correlation": 0.99659027675486, "300–1000 Hz_40ms_modulation_mae_db": 0.4748854636465151, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3922304320752832, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17725695754247942, "1000–3000 Hz_10ms_envelope_correlation": 0.9803780026046093, "1000–3000 Hz_10ms_modulation_mae_db": 0.866924248503277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8594604839426909, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.116741580833326, "1000–3000 Hz_40ms_envelope_correlation": 0.9930430789497542, "1000–3000 Hz_40ms_modulation_mae_db": 0.48114660443358237, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5514314005547205, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22554988938192388, "3000–6000 Hz_10ms_envelope_correlation": 0.9660004039111207, "3000–6000 Hz_10ms_modulation_mae_db": 0.7697532880681853, "3000–6000 Hz_40ms_log_rms_mae_db": 1.35431243880383, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18130146320406199, "3000–6000 Hz_40ms_envelope_correlation": 0.9769833462575449, "3000–6000 Hz_40ms_modulation_mae_db": 0.8181141891977769, "mr_stft_db": 4.753521747707087, "band_mae_db": 0.298543561551914, "rms_gain_error_db": -0.07132696492325258, "output_peak": 0.4734265365093165}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.856568383687981, "spectral_convergence": 0.21890327662785503, "spectral_convergence_status": "ok", "amplitude_floor": 1.419137961520426e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.508883164107508, "spectral_convergence": 0.21363419548125, "spectral_convergence_status": "ok", "amplitude_floor": 1.2001614819603722e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.902165437477436, "spectral_convergence": 0.22862718948435043, "spectral_convergence_status": "ok", "amplitude_floor": 7.977266337453151e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015734976003854988, "output_power": 0.0016028749477429852, "delta_db": 0.08033556174480082, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002272380907829781, "output_power": 0.0021759498952223164, "delta_db": -0.18832240943807432, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00035321559220182435, "output_power": 0.000362710609248266, "delta_db": 0.11520392192234766, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024119869501752642, "output_power": 0.0002216479342650031, "delta_db": -0.3671126568952766, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012307274117421543, "output_power": 0.00011753965867398084, "delta_db": -0.19977448435225997, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.88440946982043e-05, "output_power": 4.3104353616414416e-05, "delta_db": 0.45196132671442646, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2337850601778643, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07460370376666854, "pitch": null, "output_rms_dbfs": -23.44493973417712, "output_peak": 0.4877579612323477, "reference_rms_dbfs": -23.37033603041045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.755872328424308}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00019859470734653442, "log_rms_mae_db": 0.6144160200716067, "rms_envelope_nrmse": 0.061563019829726257, "envelope_correlation": 0.995736715962354, "rms_gain_error_db": -0.07460370376661264, "reference_mean_power": 0.004608354103444978, "output_mean_power": 0.004529867221731334}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.3747785986813366e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009662478923264104, "output_power": 0.0009389927643466691, "reference_active": true, "delta_db": -0.12426313892430534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.594410683545364e-05, "output_power": 9.16158227575116e-05, "reference_active": true, "delta_db": -0.20047818123213132}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7446486319031935e-05, "output_power": 1.8275966421141257e-05, "reference_active": true, "delta_db": 0.20172377221977855}], "active_band_count": 3, "active_band_mae_db": 0.17548836412540506}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00016219647203140816, "log_rms_mae_db": 0.33778313453333747, "rms_envelope_nrmse": 0.034424508618633894, "envelope_correlation": 0.9986259489370308, "rms_gain_error_db": -0.07460370376661117, "reference_mean_power": 0.004645568268479526, "output_mean_power": 0.004566447576146392}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1771130170213754e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008502808342452142, "output_power": 0.0008276908286640603, "reference_active": true, "delta_db": -0.11694247003304069}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.133174041000048e-05, "output_power": 3.9372201236710836e-05, "reference_active": true, "delta_db": -0.21093996898278025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2770439378674992e-06, "output_power": 1.2873421089022347e-06, "reference_active": true, "delta_db": 0.03488135417268851}], "active_band_count": 3, "active_band_mae_db": 0.1209212643961698}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015290889340024391, "log_rms_mae_db": 1.389862666278159, "rms_envelope_nrmse": 0.10802436801549023, "envelope_correlation": 0.9905342004858879, "rms_gain_error_db": -0.19111817790342522, "reference_mean_power": 0.0016143224036409894, "output_mean_power": 0.0015448220374207542}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 8.858385684123232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006875363814202012, "output_power": 0.0006366893216142741, "reference_active": true, "delta_db": -0.3336811802392523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.173936122300265e-05, "output_power": 5.078215500889482e-05, "reference_active": true, "delta_db": -0.8485100500900564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0160658568889886e-05, "output_power": 7.849966004995522e-06, "reference_active": true, "delta_db": -1.1205408190827617}], "active_band_count": 3, "active_band_mae_db": 0.7675773498040236}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012027897252950328, "log_rms_mae_db": 0.9858211563907299, "rms_envelope_nrmse": 0.07043667838943359, "envelope_correlation": 0.9959026720645339, "rms_gain_error_db": -0.19111817790342422, "reference_mean_power": 0.0016273586545452231, "output_mean_power": 0.0015572970471442057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.855745033342305e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006306037982733972, "output_power": 0.0005820022882332405, "reference_active": true, "delta_db": -0.3483189012638605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.290965975606541e-05, "output_power": 1.9648665770010223e-05, "reference_active": true, "delta_db": -0.6668557406582952}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0400487875049354e-06, "output_power": 6.144676558587581e-07, "reference_active": true, "delta_db": -2.28554684458855}], "active_band_count": 3, "active_band_mae_db": 1.1002404955035685}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.957022642296078e-05, "log_rms_mae_db": 1.4520378155715874, "rms_envelope_nrmse": 0.13794642124282977, "envelope_correlation": 0.9874299743655642, "rms_gain_error_db": -0.2893536656078991, "reference_mean_power": 0.000306890935740228, "output_mean_power": 0.0002871102483502492}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.0486708619890815e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.911544017889007e-05, "output_power": 5.7594736252683824e-05, "reference_active": true, "delta_db": -0.7919228445298355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0167211042917879e-05, "output_power": 7.276939969075981e-06, "reference_active": true, "delta_db": -1.452530461562704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3309723892934117e-06, "output_power": 2.4776850999795236e-06, "reference_active": true, "delta_db": 0.2650897986000099}], "active_band_count": 3, "active_band_mae_db": 0.8365143682308497}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 8.985216395929287e-05, "log_rms_mae_db": 0.9463749199392888, "rms_envelope_nrmse": 0.08474600602978141, "envelope_correlation": 0.9955108597262882, "rms_gain_error_db": -0.2893536656078996, "reference_mean_power": 0.00030936919363315046, "output_mean_power": 0.00028942876987124715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 9.09439257266203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.275291186820264e-05, "output_power": 5.147684533871394e-05, "reference_active": true, "delta_db": -0.860219586752956}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.365294966430701e-06, "output_power": 3.2882272056335663e-06, "reference_active": true, "delta_db": -1.2305177643612517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9190263566647066e-07, "output_power": 1.681811378583692e-07, "reference_active": true, "delta_db": -0.5730365310033536}], "active_band_count": 3, "active_band_mae_db": 0.8879246273725204}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.809631559896483e-05, "log_rms_mae_db": 1.5122177847331377, "rms_envelope_nrmse": 0.21813574732324917, "envelope_correlation": 0.9681546942961491, "rms_gain_error_db": -0.09450007320425112, "reference_mean_power": 6.929937891252573e-05, "output_mean_power": 6.78077501811346e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.858105660801035e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.988342411251772e-05, "output_power": 3.540180717478806e-05, "reference_active": true, "delta_db": -0.5176700448234326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1366606898498195e-05, "output_power": 1.0721584981914721e-05, "reference_active": true, "delta_db": -0.25371848399691965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.736645328964402e-06, "output_power": 2.3374487605445584e-06, "reference_active": true, "delta_db": 1.2903096688365312}], "active_band_count": 3, "active_band_mae_db": 0.6872327325522946}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.651257368418169e-05, "log_rms_mae_db": 1.279897865563536, "rms_envelope_nrmse": 0.1739432138607237, "envelope_correlation": 0.9788375056237565, "rms_gain_error_db": -0.09450007320425162, "reference_mean_power": 6.985899704640885e-05, "output_mean_power": 6.835532286092841e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.431140169416232e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0602300645347785e-05, "output_power": 3.4166389597119585e-05, "reference_active": true, "delta_db": -0.7495155386179876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067895142462218e-06, "output_power": 5.808429123643199e-06, "reference_active": true, "delta_db": -0.1897937294060621}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.231130378903163e-07, "output_power": 1.7440023054591436e-07, "reference_active": true, "delta_db": 1.5124300677922176}], "active_band_count": 3, "active_band_mae_db": 0.817246445272089}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9273817637764903, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "16753c20ba50222302d096bfecfc992a0d46007682340bd5434633334f7c6f09", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.982983346495519, "standard_deviation": 0.004766798995568594, "minimum": 0.9731837451324198, "maximum": 0.9880076526432657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4429140010509685, "standard_deviation": 0.05466206864188624, "minimum": 1.372548956175876, "maximum": 1.557126878296821, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0468199355814873, "standard_deviation": 0.2758381899019086, "minimum": 0.6251750963494073, "maximum": 1.5061841398217741, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16300748004020377, "standard_deviation": 0.020209275761587774, "minimum": 0.13794642124282977, "maximum": 0.19994679868070547, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9936770032512544, "standard_deviation": 0.001987908733865903, "minimum": 0.9891740986442876, "maximum": 0.9955108597262882, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9606876470871369, "standard_deviation": 0.05519778462792183, "minimum": 0.8594604839426909, "maximum": 1.0390358957195083, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8257727119759732, "standard_deviation": 0.28078234120995826, "minimum": 0.39661034322567995, "maximum": 1.2061383298579689, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10643970327654964, "standard_deviation": 0.012111434296220242, "minimum": 0.08474600602978141, "maximum": 0.1240495694429242, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9652743019336751, "standard_deviation": 0.001960120826128286, "minimum": 0.962519511287709, "maximum": 0.9681546942961491, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5733474429625052, "standard_deviation": 0.045615936941018606, "minimum": 1.5122177847331377, "maximum": 1.6560205003216397, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8854960975627176, "standard_deviation": 0.1977920523985188, "minimum": 0.6872327325522946, "maximum": 1.3202934360648098, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.228040252576223, "standard_deviation": 0.0064310367183874045, "minimum": 0.21813574732324917, "maximum": 0.23625043329915085, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9769760234438531, "standard_deviation": 0.001284996923215245, "minimum": 0.9743743197592775, "maximum": 0.9788375056237565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3522233596907114, "standard_deviation": 0.04180477182136577, "minimum": 1.279897865563536, "maximum": 1.406883111421142, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9828060493147351, "standard_deviation": 0.22087771818301077, "minimum": 0.6601950581328949, "maximum": 1.3316465568796312, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18168567967895455, "standard_deviation": 0.005145139577388111, "minimum": 0.1739432138607237, "maximum": 0.19121993620049962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904701705276268, "standard_deviation": 0.0012775623183190368, "minimum": 0.9885806748702051, "maximum": 0.9927072225248978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3760658694547008, "standard_deviation": 0.02577024674594515, "minimum": 1.3301997787129127, "maximum": 1.4160732553998878, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7077506832938565, "standard_deviation": 0.24936494856241123, "minimum": 0.43395789810239194, "maximum": 1.194441416266695, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10846790720820843, "standard_deviation": 0.007339892024960692, "minimum": 0.09475335175030404, "maximum": 0.11841594172343016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967921383269656, "standard_deviation": 0.0008863942470464728, "minimum": 0.9957353857692245, "maximum": 0.9979879374968734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9703900675461936, "standard_deviation": 0.03912900314951755, "minimum": 0.9071722174644071, "maximum": 1.0349433299057802, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7732170470476065, "standard_deviation": 0.3331767398941204, "minimum": 0.4121039044887865, "maximum": 1.388070901093145, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06341631320814248, "standard_deviation": 0.008343264531267163, "minimum": 0.050383871212006, "maximum": 0.07210699320246616, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.26598259199611873, "standard_deviation": 0.06636627503432747, "minimum": 0.17446445948762554, "maximum": 0.39535867197463154, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.763168852368031, "standard_deviation": 0.009396059550706378, "minimum": 4.753521747707087, "maximum": 4.779451169187933, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4929828780525941, "standard_deviation": 0.02242514307566369, "minimum": 0.46803469588403657, "maximum": 0.5271779335757606, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08520051597213124, "standard_deviation": 0.043346315997143156, "minimum": -0.16258011714472065, "maximum": -0.02120601274383651, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.99456983280644, "standard_deviation": 0.0006828669516274833, "minimum": 0.9938495944954405, "maximum": 0.995736715962354, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6445680560012618, "standard_deviation": 0.029474357076461646, "minimum": 0.6144160200716067, "maximum": 0.6861098788177561, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3049886293734974, "standard_deviation": 0.22861104133546703, "minimum": 0.10249682056086525, "maximum": 0.8347896758138725, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06950281932202887, "standard_deviation": 0.004598337007115436, "minimum": 0.061563019829726257, "maximum": 0.0741009372621703, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984003187831556, "standard_deviation": 0.00035910555903056264, "minimum": 0.9979058840919279, "maximum": 0.9989122582881472, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.36852902665822346, "standard_deviation": 0.015426537405111513, "minimum": 0.33778313453333747, "maximum": 0.3840343715291635, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28788218619729955, "standard_deviation": 0.1532231978853884, "minimum": 0.10781173497511883, "maximum": 0.5821953050997813, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03731300669001381, "standard_deviation": 0.00455204628892999, "minimum": 0.030272337118242208, "maximum": 0.044967133376763946, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2300-131720-0041", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.89534592628479, "predicted_mos": 1.89534592628479, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "04a5266dfcb79719fc9e9cd06842f2101da6d824fc40e443f73cdd6ddda6d1c9", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.351109300274402, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6393010753372121, "全帯域_10ms_rms_envelope_nrmse": 0.07228622275115816, "全帯域_10ms_envelope_correlation": 0.9941206327022669, "全帯域_10ms_modulation_mae_db": 0.3286953918100276, "全帯域_40ms_log_rms_mae_db": 0.35418775891475585, "全帯域_40ms_rms_envelope_nrmse": 0.03957202355541885, "全帯域_40ms_envelope_correlation": 0.9982205153698553, "全帯域_40ms_modulation_mae_db": 0.1826976624588639, "300–1000 Hz_10ms_log_rms_mae_db": 1.4163026777092018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1017375520212387, "300–1000 Hz_10ms_envelope_correlation": 0.9917498935132177, "300–1000 Hz_10ms_modulation_mae_db": 0.5269127420395416, "300–1000 Hz_40ms_log_rms_mae_db": 0.9723525194933084, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06039304706078387, "300–1000 Hz_40ms_envelope_correlation": 0.9972351900410706, "300–1000 Hz_40ms_modulation_mae_db": 0.27072283264946523, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3918111296973585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1399882939906608, "1000–3000 Hz_10ms_envelope_correlation": 0.9870595171860483, "1000–3000 Hz_10ms_modulation_mae_db": 1.2236481670059527, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9446469518408709, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08403596797095107, "1000–3000 Hz_40ms_envelope_correlation": 0.995614094521529, "1000–3000 Hz_40ms_modulation_mae_db": 0.8599736688296232, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5436056675830778, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22161840114943113, "3000–6000 Hz_10ms_envelope_correlation": 0.967467366779165, "3000–6000 Hz_10ms_modulation_mae_db": 0.7875824035519682, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3173780149799166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17527677125623028, "3000–6000 Hz_40ms_envelope_correlation": 0.9786651902102893, "3000–6000 Hz_40ms_modulation_mae_db": 1.0038941997361044, "mr_stft_db": 4.733965667646889, "band_mae_db": 0.19333343840305905, "rms_gain_error_db": -0.08976847270888463, "output_peak": 0.5390867411505121}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.601534781446843, "全帯域_10ms_rms_envelope_nrmse": 0.06323085418983314, "全帯域_10ms_envelope_correlation": 0.9956159502560226, "全帯域_10ms_modulation_mae_db": 0.6013973355597904, "全帯域_40ms_log_rms_mae_db": 0.3443420194823897, "全帯域_40ms_rms_envelope_nrmse": 0.03409979081164887, "全帯域_40ms_envelope_correlation": 0.9988016119499401, "全帯域_40ms_modulation_mae_db": 0.5054369235672683, "300–1000 Hz_10ms_log_rms_mae_db": 1.3541489527648432, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11463118689332247, "300–1000 Hz_10ms_envelope_correlation": 0.9898648962061637, "300–1000 Hz_10ms_modulation_mae_db": 1.0508047357024186, "300–1000 Hz_40ms_log_rms_mae_db": 1.0045007887387007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07264343102403385, "300–1000 Hz_40ms_envelope_correlation": 0.9963532919369652, "300–1000 Hz_40ms_modulation_mae_db": 0.6498550153986425, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3269642640988848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13435176701966636, "1000–3000 Hz_10ms_envelope_correlation": 0.9877561506791243, "1000–3000 Hz_10ms_modulation_mae_db": 0.9773122777408977, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8344981756619657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07890566169482556, "1000–3000 Hz_40ms_envelope_correlation": 0.9956215707930137, "1000–3000 Hz_40ms_modulation_mae_db": 0.593101924268457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.610212194232195, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23596766362715194, "3000–6000 Hz_10ms_envelope_correlation": 0.9626168114735697, "3000–6000 Hz_10ms_modulation_mae_db": 1.0042799583066948, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3871192868475484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18907876911852997, "3000–6000 Hz_40ms_envelope_correlation": 0.9753277234835398, "3000–6000 Hz_40ms_modulation_mae_db": 1.195357111020203, "mr_stft_db": 4.749415907483653, "band_mae_db": 0.21080272618054882, "rms_gain_error_db": -0.12296544372083554, "output_peak": 0.4719643875457619}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6380803889495171, "全帯域_10ms_rms_envelope_nrmse": 0.06947544623767554, "全帯域_10ms_envelope_correlation": 0.9945287098433471, "全帯域_10ms_modulation_mae_db": 0.27859437645701557, "全帯域_40ms_log_rms_mae_db": 0.34693029974288375, "全帯域_40ms_rms_envelope_nrmse": 0.036138006517770174, "全帯域_40ms_envelope_correlation": 0.9984008088671811, "全帯域_40ms_modulation_mae_db": 0.37592931784849704, "300–1000 Hz_10ms_log_rms_mae_db": 1.4138591565539191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12017505344146966, "300–1000 Hz_10ms_envelope_correlation": 0.9880386927606478, "300–1000 Hz_10ms_modulation_mae_db": 0.4197460804329529, "300–1000 Hz_40ms_log_rms_mae_db": 1.008450297664412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07396662627392468, "300–1000 Hz_40ms_envelope_correlation": 0.9952238714674354, "300–1000 Hz_40ms_modulation_mae_db": 1.1234213057437978, "1000–3000 Hz_10ms_log_rms_mae_db": 1.479376619470418, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1654541018158039, "1000–3000 Hz_10ms_envelope_correlation": 0.9817510784508913, "1000–3000 Hz_10ms_modulation_mae_db": 1.010299818347007, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0073258026857792, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11250145897822066, "1000–3000 Hz_40ms_envelope_correlation": 0.9916931409466858, "1000–3000 Hz_40ms_modulation_mae_db": 0.530315318426923, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4742524113118454, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2210731147605373, "3000–6000 Hz_10ms_envelope_correlation": 0.968486354015104, "3000–6000 Hz_10ms_modulation_mae_db": 1.4272010692155666, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2917704756336583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1774931742393174, "3000–6000 Hz_40ms_envelope_correlation": 0.978634924326761, "3000–6000 Hz_40ms_modulation_mae_db": 1.5308474558540406, "mr_stft_db": 4.755490441222782, "band_mae_db": 0.2855328218214435, "rms_gain_error_db": -0.004368340220606825, "output_peak": 0.4697774789463148}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6710601714955604, "全帯域_10ms_rms_envelope_nrmse": 0.07819042479282176, "全帯域_10ms_envelope_correlation": 0.9931689087309126, "全帯域_10ms_modulation_mae_db": 0.4730071737147831, "全帯域_40ms_log_rms_mae_db": 0.4112027617499155, "全帯域_40ms_rms_envelope_nrmse": 0.041948208533782505, "全帯域_40ms_envelope_correlation": 0.9980327616076499, "全帯域_40ms_modulation_mae_db": 0.6550167345469652, "300–1000 Hz_10ms_log_rms_mae_db": 1.429011887784276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1327589472973435, "300–1000 Hz_10ms_envelope_correlation": 0.9855568288586768, "300–1000 Hz_10ms_modulation_mae_db": 0.7406697238158776, "300–1000 Hz_40ms_log_rms_mae_db": 1.047474464139365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08054015303768662, "300–1000 Hz_40ms_envelope_correlation": 0.9946241391678509, "300–1000 Hz_40ms_modulation_mae_db": 0.8525077504114663, "1000–3000 Hz_10ms_log_rms_mae_db": 1.383106509093181, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13756778853217402, "1000–3000 Hz_10ms_envelope_correlation": 0.9875688774625231, "1000–3000 Hz_10ms_modulation_mae_db": 0.947829869977655, "1000–3000 Hz_40ms_log_rms_mae_db": 0.902648240374727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08435512708569563, "1000–3000 Hz_40ms_envelope_correlation": 0.9956800409653074, "1000–3000 Hz_40ms_modulation_mae_db": 0.38603323451919586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.538842546769357, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21212889353087394, "3000–6000 Hz_10ms_envelope_correlation": 0.9712871447886038, "3000–6000 Hz_10ms_modulation_mae_db": 1.1952873500794075, "3000–6000 Hz_40ms_log_rms_mae_db": 1.353869663556502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17396340681684241, "3000–6000 Hz_40ms_envelope_correlation": 0.9797636148429384, "3000–6000 Hz_40ms_modulation_mae_db": 1.3051566057504453, "mr_stft_db": 4.737313397691797, "band_mae_db": 0.27118126576571067, "rms_gain_error_db": -0.11398190145071985, "output_peak": 0.4960710983488806}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6626314542492268, "全帯域_10ms_rms_envelope_nrmse": 0.07340804861154643, "全帯域_10ms_envelope_correlation": 0.993996942753549, "全帯域_10ms_modulation_mae_db": 0.1371027263205298, "全帯域_40ms_log_rms_mae_db": 0.4071767234968927, "全帯域_40ms_rms_envelope_nrmse": 0.03798607565784827, "全帯域_40ms_envelope_correlation": 0.9983332736392357, "全帯域_40ms_modulation_mae_db": 0.28575116068642153, "300–1000 Hz_10ms_log_rms_mae_db": 1.3205325480873082, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10086741088661182, "300–1000 Hz_10ms_envelope_correlation": 0.9918217008186354, "300–1000 Hz_10ms_modulation_mae_db": 0.6363968963077414, "300–1000 Hz_40ms_log_rms_mae_db": 0.9413244004402388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05872918189329043, "300–1000 Hz_40ms_envelope_correlation": 0.9973132793304448, "300–1000 Hz_40ms_modulation_mae_db": 0.9751714900315346, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3337119053716215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14949299391794452, "1000–3000 Hz_10ms_envelope_correlation": 0.9849559308616772, "1000–3000 Hz_10ms_modulation_mae_db": 1.2689620729449973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8284245333549383, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07746359980153882, "1000–3000 Hz_40ms_envelope_correlation": 0.9958115669436515, "1000–3000 Hz_40ms_modulation_mae_db": 0.8791515221562981, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5005031524222654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22084911455676234, "3000–6000 Hz_10ms_envelope_correlation": 0.9672316785277516, "3000–6000 Hz_10ms_modulation_mae_db": 0.6829109838154594, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2582869200080558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17855371826867925, "3000–6000 Hz_40ms_envelope_correlation": 0.9777684744506246, "3000–6000 Hz_40ms_modulation_mae_db": 0.8268510160431507, "mr_stft_db": 4.7527831200694, "band_mae_db": 0.12133562177686796, "rms_gain_error_db": -0.027663297682036406, "output_peak": 0.4979624231737684}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.676736647353869, "全帯域_10ms_rms_envelope_nrmse": 0.07981123385042557, "全帯域_10ms_envelope_correlation": 0.9928430280703036, "全帯域_10ms_modulation_mae_db": 0.22518373709344874, "全帯域_40ms_log_rms_mae_db": 0.35103812024702896, "全帯域_40ms_rms_envelope_nrmse": 0.037920363252313426, "全帯域_40ms_envelope_correlation": 0.998368249724738, "全帯域_40ms_modulation_mae_db": 0.2714610689653908, "300–1000 Hz_10ms_log_rms_mae_db": 1.3256787400673442, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13680108904623173, "300–1000 Hz_10ms_envelope_correlation": 0.9845657092930862, "300–1000 Hz_10ms_modulation_mae_db": 1.0161154977710571, "300–1000 Hz_40ms_log_rms_mae_db": 0.9548906763099911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07619898401618043, "300–1000 Hz_40ms_envelope_correlation": 0.9951897979800093, "300–1000 Hz_40ms_modulation_mae_db": 0.5285324713823972, "1000–3000 Hz_10ms_log_rms_mae_db": 1.437444336359287, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14652888617000967, "1000–3000 Hz_10ms_envelope_correlation": 0.9859015341463747, "1000–3000 Hz_10ms_modulation_mae_db": 0.8793200016071651, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9558167491366366, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0906195517940581, "1000–3000 Hz_40ms_envelope_correlation": 0.9949886896184101, "1000–3000 Hz_40ms_modulation_mae_db": 0.480167956746197, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5904758388649878, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22707621429315927, "3000–6000 Hz_10ms_envelope_correlation": 0.9653664764790421, "3000–6000 Hz_10ms_modulation_mae_db": 0.9084799661267033, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3947972910296453, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18316286806688606, "3000–6000 Hz_40ms_envelope_correlation": 0.9764840129054313, "3000–6000 Hz_40ms_modulation_mae_db": 1.167766122847335, "mr_stft_db": 4.737462361256857, "band_mae_db": 0.20838544169152928, "rms_gain_error_db": -0.09097728908632662, "output_peak": 0.5186431521052611}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6568308846090989, "全帯域_10ms_rms_envelope_nrmse": 0.0799210497242299, "全帯域_10ms_envelope_correlation": 0.9927412524485404, "全帯域_10ms_modulation_mae_db": 0.5221285600772897, "全帯域_40ms_log_rms_mae_db": 0.3834822894464773, "全帯域_40ms_rms_envelope_nrmse": 0.038585556666717806, "全帯域_40ms_envelope_correlation": 0.9982435850452461, "全帯域_40ms_modulation_mae_db": 0.19417073689520348, "300–1000 Hz_10ms_log_rms_mae_db": 1.3504498697953022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12245184322262236, "300–1000 Hz_10ms_envelope_correlation": 0.9876806876658338, "300–1000 Hz_10ms_modulation_mae_db": 0.3896035669857523, "300–1000 Hz_40ms_log_rms_mae_db": 0.9338903403508377, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07308407719743662, "300–1000 Hz_40ms_envelope_correlation": 0.9955400249143641, "300–1000 Hz_40ms_modulation_mae_db": 0.6633884908595169, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4103994132470137, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1555366712178573, "1000–3000 Hz_10ms_envelope_correlation": 0.9841853160944956, "1000–3000 Hz_10ms_modulation_mae_db": 1.215027990503328, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9036943361191968, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09830835411971581, "1000–3000 Hz_40ms_envelope_correlation": 0.9941647664102108, "1000–3000 Hz_40ms_modulation_mae_db": 0.7946976371187447, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5336330995545078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23341208281575246, "3000–6000 Hz_10ms_envelope_correlation": 0.963651899141656, "3000–6000 Hz_10ms_modulation_mae_db": 0.7323750605227817, "3000–6000 Hz_40ms_log_rms_mae_db": 1.310569638163966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1782909560855155, "3000–6000 Hz_40ms_envelope_correlation": 0.9777091712274534, "3000–6000 Hz_40ms_modulation_mae_db": 0.8729726846470721, "mr_stft_db": 4.758002675104784, "band_mae_db": 0.1761208061851234, "rms_gain_error_db": -0.06620259740715546, "output_peak": 0.4215249371236868}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6162361897947063, "全帯域_10ms_rms_envelope_nrmse": 0.07391683923383369, "全帯域_10ms_envelope_correlation": 0.9938893355051007, "全帯域_10ms_modulation_mae_db": 0.2551779359223904, "全帯域_40ms_log_rms_mae_db": 0.32456431525698554, "全帯域_40ms_rms_envelope_nrmse": 0.04633511168921502, "全帯域_40ms_envelope_correlation": 0.9975433578025047, "全帯域_40ms_modulation_mae_db": 0.39103894386382204, "300–1000 Hz_10ms_log_rms_mae_db": 1.3402865802926516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.107228213948986, "300–1000 Hz_10ms_envelope_correlation": 0.9905750209399335, "300–1000 Hz_10ms_modulation_mae_db": 0.4835790953458461, "300–1000 Hz_40ms_log_rms_mae_db": 0.9375509023153655, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07063200185711974, "300–1000 Hz_40ms_envelope_correlation": 0.9957374356018375, "300–1000 Hz_40ms_modulation_mae_db": 0.7633823922055983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4186343509677681, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1485499786632337, "1000–3000 Hz_10ms_envelope_correlation": 0.9853080509916403, "1000–3000 Hz_10ms_modulation_mae_db": 1.0372553815000527, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9867885162096373, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08913959157864453, "1000–3000 Hz_40ms_envelope_correlation": 0.9949654421338527, "1000–3000 Hz_40ms_modulation_mae_db": 0.47522367255409437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5790669873479362, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22150063916377255, "3000–6000 Hz_10ms_envelope_correlation": 0.967327852053211, "3000–6000 Hz_10ms_modulation_mae_db": 0.8876844812441128, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3262552655844682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17651007515636646, "3000–6000 Hz_40ms_envelope_correlation": 0.9782090181593647, "3000–6000 Hz_40ms_modulation_mae_db": 1.1596741322643798, "mr_stft_db": 4.732879704855112, "band_mae_db": 0.19018534045695587, "rms_gain_error_db": -0.11059468741122352, "output_peak": 0.4547366864858381}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.844853723810778, "spectral_convergence": 0.2225249366504399, "spectral_convergence_status": "ok", "amplitude_floor": 1.419137961520426e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.469817784250744, "spectral_convergence": 0.21951002789857185, "spectral_convergence_status": "ok", "amplitude_floor": 1.2001614819603722e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.8872254948791465, "spectral_convergence": 0.23767377160293351, "spectral_convergence_status": "ok", "amplitude_floor": 7.977266337453151e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015734976003854988, "output_power": 0.0016210796013201606, "delta_db": 0.12938255979922839, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002272380907829781, "output_power": 0.0021624218385671154, "delta_db": -0.2154071304301529, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00035321559220182435, "output_power": 0.00033958511719983756, "delta_db": -0.17091218226834504, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024119869501752642, "output_power": 0.00021968107575784558, "delta_db": -0.40582307168037046, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012307274117421543, "output_power": 0.00012483281725182414, "delta_db": 0.061668982093034924, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.88440946982043e-05, "output_power": 4.045811780538929e-05, "delta_db": 0.1768067041472227, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19333343840305905, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08976847270888463, "pitch": null, "output_rms_dbfs": -23.460104503119332, "output_peak": 0.5390867411505121, "reference_rms_dbfs": -23.37033603041045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.733965667646889}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00019859470734653442, "log_rms_mae_db": 0.6393010753372121, "rms_envelope_nrmse": 0.07228622275115816, "envelope_correlation": 0.9941206327022669, "rms_gain_error_db": -0.08976847270894629, "reference_mean_power": 0.004608354103444978, "output_mean_power": 0.004514077337685691}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.3747785986813366e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009662478923264104, "output_power": 0.000922063132554387, "reference_active": true, "delta_db": -0.20327901909360904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.594410683545364e-05, "output_power": 9.257843290258843e-05, "reference_active": true, "delta_db": -0.15508479165682873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7446486319031935e-05, "output_power": 2.0159521085529097e-05, "reference_active": true, "delta_db": 0.627722364679645}], "active_band_count": 3, "active_band_mae_db": 0.3286953918100276}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00016219647203140816, "log_rms_mae_db": 0.35418775891475585, "rms_envelope_nrmse": 0.03957202355541885, "envelope_correlation": 0.9982205153698553, "rms_gain_error_db": -0.08976847270894482, "reference_mean_power": 0.004645568268479526, "output_mean_power": 0.004550530182942911}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1771130170213754e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008502808342452142, "output_power": 0.0008134306080161695, "reference_active": true, "delta_db": -0.19241879830414602}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.133174041000048e-05, "output_power": 4.0271906671989475e-05, "reference_active": true, "delta_db": -0.11281501325181795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2770439378674992e-06, "output_power": 1.350491196257924e-06, "reference_active": true, "delta_db": 0.24285917582062774}], "active_band_count": 3, "active_band_mae_db": 0.1826976624588639}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015290889340024391, "log_rms_mae_db": 1.4163026777092018, "rms_envelope_nrmse": 0.1017375520212387, "envelope_correlation": 0.9917498935132177, "rms_gain_error_db": -0.22163463328841546, "reference_mean_power": 0.0016143224036409894, "output_mean_power": 0.001534005125269351}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 8.858385684123232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006875363814202012, "output_power": 0.0006473177537073903, "reference_active": true, "delta_db": -0.2617816568567515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.173936122300265e-05, "output_power": 5.4772385947572136e-05, "reference_active": true, "delta_db": -0.5200047210333323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0160658568889886e-05, "output_power": 1.221285055297867e-05, "reference_active": true, "delta_db": 0.7989518482285413}], "active_band_count": 3, "active_band_mae_db": 0.5269127420395416}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012027897252950328, "log_rms_mae_db": 0.9723525194933084, "rms_envelope_nrmse": 0.06039304706078387, "envelope_correlation": 0.9972351900410706, "rms_gain_error_db": -0.22163463328841393, "reference_mean_power": 0.0016273586545452231, "output_mean_power": 0.0015463927844236124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.855745033342305e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006306037982733972, "output_power": 0.0005964425476346019, "reference_active": true, "delta_db": -0.24187965698346725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.290965975606541e-05, "output_power": 2.1185642878419215e-05, "reference_active": true, "delta_db": -0.33976992086728064}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0400487875049354e-06, "output_power": 9.862836189068018e-07, "reference_active": true, "delta_db": -0.23051892009764777}], "active_band_count": 3, "active_band_mae_db": 0.27072283264946523}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.957022642296078e-05, "log_rms_mae_db": 1.3918111296973585, "rms_envelope_nrmse": 0.1399882939906608, "envelope_correlation": 0.9870595171860483, "rms_gain_error_db": -0.2922499369815868, "reference_mean_power": 0.000306890935740228, "output_mean_power": 0.0002869188409036283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.0486708619890815e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.911544017889007e-05, "output_power": 5.771253038391382e-05, "reference_active": true, "delta_db": -0.783049620754452}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0167211042917879e-05, "output_power": 7.810385586222197e-06, "reference_active": true, "delta_db": -1.145293635623648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3309723892934117e-06, "output_power": 3.481747350115835e-06, "reference_active": true, "delta_db": 1.7426012446397579}], "active_band_count": 3, "active_band_mae_db": 1.2236481670059527}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 8.985216395929287e-05, "log_rms_mae_db": 0.9446469518408709, "rms_envelope_nrmse": 0.08403596797095107, "envelope_correlation": 0.995614094521529, "rms_gain_error_db": -0.29224993698158624, "reference_mean_power": 0.00030936919363315046, "output_mean_power": 0.00028923581673865093}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 9.09439257266203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.275291186820264e-05, "output_power": 5.208558273102655e-05, "reference_active": true, "delta_db": -0.80916355414952}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.365294966430701e-06, "output_power": 3.6043167119937464e-06, "reference_active": true, "delta_db": -0.8319064913211632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9190263566647066e-07, "output_power": 2.3821331654291143e-07, "reference_active": true, "delta_db": 0.9388509610181862}], "active_band_count": 3, "active_band_mae_db": 0.8599736688296232}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.809631559896483e-05, "log_rms_mae_db": 1.5436056675830778, "rms_envelope_nrmse": 0.22161840114943113, "envelope_correlation": 0.967467366779165, "rms_gain_error_db": 0.015640656856120254, "reference_mean_power": 6.929937891252573e-05, "output_mean_power": 6.95494032508451e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.858105660801035e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.988342411251772e-05, "output_power": 3.845770932747883e-05, "reference_active": true, "delta_db": -0.15809024124145704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1366606898498195e-05, "output_power": 1.0291556095318442e-05, "reference_active": true, "delta_db": -0.43149795100130006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.736645328964402e-06, "output_power": 2.612323963863649e-06, "reference_active": true, "delta_db": 1.7731590184131474}], "active_band_count": 3, "active_band_mae_db": 0.7875824035519682}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.651257368418169e-05, "log_rms_mae_db": 1.3173780149799166, "rms_envelope_nrmse": 0.17527677125623028, "envelope_correlation": 0.9786651902102893, "rms_gain_error_db": 0.015640656856121215, "reference_mean_power": 6.985899704640885e-05, "output_mean_power": 7.011104042379945e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.431140169416232e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0602300645347785e-05, "output_power": 3.7444155673939345e-05, "reference_active": true, "delta_db": -0.3516660052030643}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067895142462218e-06, "output_power": 5.681928422344756e-06, "reference_active": true, "delta_db": -0.28542308832308744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.231130378903163e-07, "output_power": 2.1269799620238272e-07, "reference_active": true, "delta_db": 2.3745935056821614}], "active_band_count": 3, "active_band_mae_db": 1.0038941997361044}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9249693315435656, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "177afa8b122ebbb75f05ac6d7b45aed115712d57244a41b8a2841ec1b735eb6e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9855608069840969, "standard_deviation": 0.002001780480810021, "minimum": 0.9817510784508913, "maximum": 0.9877561506791243, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3976810660381918, "standard_deviation": 0.050984410891940925, "minimum": 1.3269642640988848, "maximum": 1.479376619470418, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0699569474533819, "standard_deviation": 0.14576816570969073, "minimum": 0.8793200016071651, "maximum": 1.2689620729449973, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14718381016591878, "standard_deviation": 0.01014259879718969, "minimum": 0.13435176701966636, "maximum": 0.1654541018158039, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9948174140415826, "standard_deviation": 0.0013759210946944758, "minimum": 0.9916931409466858, "maximum": 0.9958115669436515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9204804131729689, "standard_deviation": 0.06572513477083326, "minimum": 0.8284245333549383, "maximum": 1.0073258026857792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6248331168274417, "standard_deviation": 0.19239087693025145, "minimum": 0.38603323451919586, "maximum": 0.8791515221562981, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08941616412795628, "standard_deviation": 0.011473083888750426, "minimum": 0.07746359980153882, "maximum": 0.11250145897822066, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9666794479072629, "standard_deviation": 0.0027577667794297057, "minimum": 0.9626168114735697, "maximum": 0.9712871447886038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5463239872607715, "standard_deviation": 0.04570319440905234, "minimum": 1.4742524113118454, "maximum": 1.610212194232195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9532251591078367, "standard_deviation": 0.251121374981606, "minimum": 0.6829109838154594, "maximum": 1.4272010692155666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22420326548718011, "standard_deviation": 0.007674461897300904, "minimum": 0.21212889353087394, "maximum": 0.23596766362715194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9778202662008003, "standard_deviation": 0.0013793716488435957, "minimum": 0.9753277234835398, "maximum": 0.9797636148429384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.33000581947547, "standard_deviation": 0.04657743368398217, "minimum": 1.2582869200080558, "maximum": 1.3947972910296453, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1328149160203413, "standard_deviation": 0.23044631716932423, "minimum": 0.8268510160431507, "maximum": 1.5308474558540406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17904121737604592, "standard_deviation": 0.004888752834497757, "minimum": 0.17396340681684241, "maximum": 0.18907876911852997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9887316787570244, "standard_deviation": 0.0027349519233829194, "minimum": 0.9845657092930862, "maximum": 0.9918217008186354, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3687838016318556, "standard_deviation": 0.04385998314391462, "minimum": 1.3205325480873082, "maximum": 1.429011887784276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6579785423001485, "standard_deviation": 0.25808396762054164, "minimum": 0.3896035669857523, "maximum": 1.0508047357024186, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11708141209472828, "standard_deviation": 0.013493278277673297, "minimum": 0.10086741088661182, "maximum": 0.13680108904623173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959021288049972, "standard_deviation": 0.0009805898405592602, "minimum": 0.9946241391678509, "maximum": 0.9973132793304448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9750542986815274, "standard_deviation": 0.04119573904358887, "minimum": 0.9338903403508377, "maximum": 1.047474464139365, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7283727185853024, "standard_deviation": 0.265484926998234, "minimum": 0.27072283264946523, "maximum": 1.1234213057437978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07077343779505704, "standard_deviation": 0.0075279885035014005, "minimum": 0.05872918189329043, "maximum": 0.08054015303768662, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20710968278515482, "standard_deviation": 0.05218694541132096, "minimum": 0.12133562177686796, "maximum": 0.2855328218214435, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 4.744664159416409, "standard_deviation": 0.01030153733611843, "minimum": 4.732879704855112, "maximum": 4.758002675104784, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.48372086311000295, "standard_deviation": 0.037189371459512476, "minimum": 0.4215249371236868, "maximum": 0.5390867411505121, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07831525371097361, "standard_deviation": 0.04272787237058859, "minimum": -0.12296544372083554, "maximum": -0.004368340220606825, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9938630950387554, "standard_deviation": 0.0009547951125834049, "minimum": 0.9927412524485404, "maximum": 0.9956159502560226, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6453014491545042, "standard_deviation": 0.0265475291324775, "minimum": 0.601534781446843, "maximum": 0.676736647353869, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3526609046194094, "standard_deviation": 0.16184260646192672, "minimum": 0.1371027263205298, "maximum": 0.6013973355597904, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07378001492394053, "standard_deviation": 0.00567881053972557, "minimum": 0.06323085418983314, "maximum": 0.0799210497242299, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982430205007939, "standard_deviation": 0.0003575697846589093, "minimum": 0.9975433578025047, "maximum": 0.9988016119499401, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.36536553604216615, "standard_deviation": 0.03151926113904116, "minimum": 0.32456431525698554, "maximum": 0.4112027617499155, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3576878186040541, "standard_deviation": 0.16108944187620008, "minimum": 0.1826976624588639, "maximum": 0.6550167345469652, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03907314208558936, "standard_deviation": 0.003729203055793898, "minimum": 0.03409979081164887, "maximum": 0.04633511168921502, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_237-126133-0003", "language": "en", "style": "normal", "speaker": "librispeech_237", "source_utterance_id": "237-126133-0003", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_237-126133-0003.wav", "source_file_sha256": "eee807661a1075bd1daa41143c9bbd0b4555d3ffc5f9c0d07c7541b89930b351", "decoded_pcm_sha256": "15627325ed4198a6a91f3a9dd72c16085a4986165c018e492c56a9076ad969f4", "original_source_sha256": "18c38673f71d4012a9c023d7548bff3c0f78172415f67c69abf403590c4e84bc", "corpus_entry_id": "librispeech_237-126133-0003"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.541547775268555, "predicted_mos": 4.541547775268555, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "780d0e36696d3b72761adc5acb2a0730fca079902a3a71a337a87dae2388a789", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.56334877900008, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_237-126133-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "780d0e36696d3b72761adc5acb2a0730fca079902a3a71a337a87dae2388a789", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "3090a528faff51d26d0ea1fe3914063151c1e8eaf9354b24e51173247b7a69fc"}, "rows": [{"case": "librispeech_237-126133-0003", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.3949594497680664, "predicted_mos": 2.3949594497680664, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "37888a12d609cf26f0e3046f482db39445b8cd36d008c1ea21968c158726c486", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.482658105000155, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_237-126133-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "37888a12d609cf26f0e3046f482db39445b8cd36d008c1ea21968c158726c486", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.968250897643232, "全帯域_10ms_rms_envelope_nrmse": 0.0781406535662062, "全帯域_10ms_envelope_correlation": 0.9938717300066603, "全帯域_10ms_modulation_mae_db": 0.3885887408054462, "全帯域_40ms_log_rms_mae_db": 0.5033273483124855, "全帯域_40ms_rms_envelope_nrmse": 0.04447663332756445, "全帯域_40ms_envelope_correlation": 0.9979676932724085, "全帯域_40ms_modulation_mae_db": 0.5573476534174213, "300–1000 Hz_10ms_log_rms_mae_db": 1.5125491133237599, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10297436558200063, "300–1000 Hz_10ms_envelope_correlation": 0.9910205245217738, "300–1000 Hz_10ms_modulation_mae_db": 0.7625126785519504, "300–1000 Hz_40ms_log_rms_mae_db": 0.8802994083871603, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06178330473028664, "300–1000 Hz_40ms_envelope_correlation": 0.996808245329466, "300–1000 Hz_40ms_modulation_mae_db": 0.4641310133085322, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5659655392300411, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13393429287349873, "1000–3000 Hz_10ms_envelope_correlation": 0.9853607577158263, "1000–3000 Hz_10ms_modulation_mae_db": 0.8788393425718938, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8515766092338539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07701427800400369, "1000–3000 Hz_40ms_envelope_correlation": 0.9947376529754964, "1000–3000 Hz_40ms_modulation_mae_db": 1.0023677571356788, "3000–6000 Hz_10ms_log_rms_mae_db": 1.65632058006543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21105934268131082, "3000–6000 Hz_10ms_envelope_correlation": 0.9610950946155086, "3000–6000 Hz_10ms_modulation_mae_db": 0.2527199543566097, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9886553762489769, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11474495727201602, "3000–6000 Hz_40ms_envelope_correlation": 0.987235258496505, "3000–6000 Hz_40ms_modulation_mae_db": 0.40486957120289596, "mr_stft_db": 5.638096264392064, "band_mae_db": 0.20460916942912913, "rms_gain_error_db": -0.12397842735698765, "output_peak": 0.23613277954677747}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9723738351570965, "全帯域_10ms_rms_envelope_nrmse": 0.07602990320260539, "全帯域_10ms_envelope_correlation": 0.9943437942354325, "全帯域_10ms_modulation_mae_db": 0.3325643524757759, "全帯域_40ms_log_rms_mae_db": 0.5307289666961418, "全帯域_40ms_rms_envelope_nrmse": 0.042499112243192146, "全帯域_40ms_envelope_correlation": 0.9983404547565499, "全帯域_40ms_modulation_mae_db": 0.5419820476098685, "300–1000 Hz_10ms_log_rms_mae_db": 1.5345410422722612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10229097530069366, "300–1000 Hz_10ms_envelope_correlation": 0.9913483903733551, "300–1000 Hz_10ms_modulation_mae_db": 0.3629131750078412, "300–1000 Hz_40ms_log_rms_mae_db": 0.967956787635875, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0655104383262822, "300–1000 Hz_40ms_envelope_correlation": 0.9966000090516594, "300–1000 Hz_40ms_modulation_mae_db": 0.5969680866238901, "1000–3000 Hz_10ms_log_rms_mae_db": 1.589908098412392, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14569467950096163, "1000–3000 Hz_10ms_envelope_correlation": 0.9827061410206566, "1000–3000 Hz_10ms_modulation_mae_db": 0.6707770057557471, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8824881359996656, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07806026957996183, "1000–3000 Hz_40ms_envelope_correlation": 0.9946038635313553, "1000–3000 Hz_40ms_modulation_mae_db": 0.618702157651917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7088759379496923, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19696243674566097, "3000–6000 Hz_10ms_envelope_correlation": 0.9661073781527953, "3000–6000 Hz_10ms_modulation_mae_db": 1.1140199383456781, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0104794675413684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11833722045467288, "3000–6000 Hz_40ms_envelope_correlation": 0.9865636191543584, "3000–6000 Hz_40ms_modulation_mae_db": 1.1319450923315177, "mr_stft_db": 5.628271489529463, "band_mae_db": 0.2592253301526439, "rms_gain_error_db": -0.16836742463028315, "output_peak": 0.24134421127902306}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9518027087649398, "全帯域_10ms_rms_envelope_nrmse": 0.06874692456544915, "全帯域_10ms_envelope_correlation": 0.9952626316602466, "全帯域_10ms_modulation_mae_db": 0.3428550856420472, "全帯域_40ms_log_rms_mae_db": 0.47591583745562177, "全帯域_40ms_rms_envelope_nrmse": 0.03341942582851839, "全帯域_40ms_envelope_correlation": 0.9989029186892573, "全帯域_40ms_modulation_mae_db": 0.2931507806212379, "300–1000 Hz_10ms_log_rms_mae_db": 1.4630455485317884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08644814663676405, "300–1000 Hz_10ms_envelope_correlation": 0.9935844323904893, "300–1000 Hz_10ms_modulation_mae_db": 0.848268591144488, "300–1000 Hz_40ms_log_rms_mae_db": 0.7929563795949984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04467408639546606, "300–1000 Hz_40ms_envelope_correlation": 0.9982799286251283, "300–1000 Hz_40ms_modulation_mae_db": 0.530388879078385, "1000–3000 Hz_10ms_log_rms_mae_db": 1.589352787591837, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13331282427310753, "1000–3000 Hz_10ms_envelope_correlation": 0.9856710885300607, "1000–3000 Hz_10ms_modulation_mae_db": 1.190522734882356, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8926504578579991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07184856067905533, "1000–3000 Hz_40ms_envelope_correlation": 0.9958087943934774, "1000–3000 Hz_40ms_modulation_mae_db": 1.253363630620734, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7402770469151831, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20285541265763504, "3000–6000 Hz_10ms_envelope_correlation": 0.9640686141737471, "3000–6000 Hz_10ms_modulation_mae_db": 0.4936312171697306, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0428773195100964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12347544985444414, "3000–6000 Hz_40ms_envelope_correlation": 0.9852767938251412, "3000–6000 Hz_40ms_modulation_mae_db": 0.5146012316682703, "mr_stft_db": 5.645425630182761, "band_mae_db": 0.24313772531807346, "rms_gain_error_db": -0.11148926112921617, "output_peak": 0.24596651996125724}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9499117629738472, "全帯域_10ms_rms_envelope_nrmse": 0.0718875379542136, "全帯域_10ms_envelope_correlation": 0.9948743382844543, "全帯域_10ms_modulation_mae_db": 0.2655267992323261, "全帯域_40ms_log_rms_mae_db": 0.4954846078232961, "全帯域_40ms_rms_envelope_nrmse": 0.039644554326866074, "全帯域_40ms_envelope_correlation": 0.9984743449403022, "全帯域_40ms_modulation_mae_db": 0.3827140651772762, "300–1000 Hz_10ms_log_rms_mae_db": 1.574655473323641, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09276327191184817, "300–1000 Hz_10ms_envelope_correlation": 0.992730385440825, "300–1000 Hz_10ms_modulation_mae_db": 0.48951209477159124, "300–1000 Hz_40ms_log_rms_mae_db": 0.9515083959931925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05818551058696963, "300–1000 Hz_40ms_envelope_correlation": 0.9971281415862268, "300–1000 Hz_40ms_modulation_mae_db": 0.466560234541928, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5846324989453109, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12463196886547047, "1000–3000 Hz_10ms_envelope_correlation": 0.9873664402069692, "1000–3000 Hz_10ms_modulation_mae_db": 0.6351813530855926, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8713748661044971, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07057351670742598, "1000–3000 Hz_40ms_envelope_correlation": 0.9956947701294326, "1000–3000 Hz_40ms_modulation_mae_db": 0.46503525742973534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.719266633415449, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20405033621453883, "3000–6000 Hz_10ms_envelope_correlation": 0.9638625193222504, "3000–6000 Hz_10ms_modulation_mae_db": 0.4053238104935053, "3000–6000 Hz_40ms_log_rms_mae_db": 1.007132248820749, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11836467283499273, "3000–6000 Hz_40ms_envelope_correlation": 0.9864417486105603, "3000–6000 Hz_40ms_modulation_mae_db": 0.40107783274168735, "mr_stft_db": 5.633834427978461, "band_mae_db": 0.19638737126813535, "rms_gain_error_db": -0.13741027323316485, "output_peak": 0.22956288513034334}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.959672219956736, "全帯域_10ms_rms_envelope_nrmse": 0.07986305282046968, "全帯域_10ms_envelope_correlation": 0.9934623105555209, "全帯域_10ms_modulation_mae_db": 0.5402867315550574, "全帯域_40ms_log_rms_mae_db": 0.5323646624796419, "全帯域_40ms_rms_envelope_nrmse": 0.04425391813020228, "全帯域_40ms_envelope_correlation": 0.9978189919364144, "全帯域_40ms_modulation_mae_db": 0.7194194647301443, "300–1000 Hz_10ms_log_rms_mae_db": 1.5335668001080471, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09830906067647563, "300–1000 Hz_10ms_envelope_correlation": 0.9915662496737081, "300–1000 Hz_10ms_modulation_mae_db": 0.2689635473866031, "300–1000 Hz_40ms_log_rms_mae_db": 0.937903319405932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05859955885931256, "300–1000 Hz_40ms_envelope_correlation": 0.9967674473513701, "300–1000 Hz_40ms_modulation_mae_db": 0.6053061690104979, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6286241223987774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14759277902169038, "1000–3000 Hz_10ms_envelope_correlation": 0.9822196309895225, "1000–3000 Hz_10ms_modulation_mae_db": 0.7725316729048447, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8728361761719972, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07780773493265117, "1000–3000 Hz_40ms_envelope_correlation": 0.9947836247940537, "1000–3000 Hz_40ms_modulation_mae_db": 0.9415429132956277, "3000–6000 Hz_10ms_log_rms_mae_db": 1.740757104850651, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20368566055935, "3000–6000 Hz_10ms_envelope_correlation": 0.9643428470155501, "3000–6000 Hz_10ms_modulation_mae_db": 0.06876661709813887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0439927312227948, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1171294338459687, "3000–6000 Hz_40ms_envelope_correlation": 0.9867939136929532, "3000–6000 Hz_40ms_modulation_mae_db": 0.13447508971791802, "mr_stft_db": 5.646300644232664, "band_mae_db": 0.17452531581311204, "rms_gain_error_db": -0.05770763091759554, "output_peak": 0.23478812960740286}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9807857669166379, "全帯域_10ms_rms_envelope_nrmse": 0.07389286194230484, "全帯域_10ms_envelope_correlation": 0.994520810411248, "全帯域_10ms_modulation_mae_db": 0.32454886387793636, "全帯域_40ms_log_rms_mae_db": 0.5324404849180865, "全帯域_40ms_rms_envelope_nrmse": 0.04094342112509971, "全帯域_40ms_envelope_correlation": 0.9982917674949928, "全帯域_40ms_modulation_mae_db": 0.5762450706295111, "300–1000 Hz_10ms_log_rms_mae_db": 1.5206495818042098, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09151180216087063, "300–1000 Hz_10ms_envelope_correlation": 0.9927893852790725, "300–1000 Hz_10ms_modulation_mae_db": 0.5235730949771554, "300–1000 Hz_40ms_log_rms_mae_db": 0.9039979992188508, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05505560685527309, "300–1000 Hz_40ms_envelope_correlation": 0.9972811405179599, "300–1000 Hz_40ms_modulation_mae_db": 0.5376429533407233, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5812972487014787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13405897937242775, "1000–3000 Hz_10ms_envelope_correlation": 0.9854872042183991, "1000–3000 Hz_10ms_modulation_mae_db": 0.6770685698861959, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8618585572084854, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07526425980956049, "1000–3000 Hz_40ms_envelope_correlation": 0.9953246738797324, "1000–3000 Hz_40ms_modulation_mae_db": 0.7954273298695256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7454602430787372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21980998663537363, "3000–6000 Hz_10ms_envelope_correlation": 0.9576488329537112, "3000–6000 Hz_10ms_modulation_mae_db": 0.7370363735432744, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0668024501597468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1327174836603707, "3000–6000 Hz_40ms_envelope_correlation": 0.9829599024052236, "3000–6000 Hz_40ms_modulation_mae_db": 0.8086187749686453, "mr_stft_db": 5.634484494754368, "band_mae_db": 0.2432943316474058, "rms_gain_error_db": -0.1189442864406173, "output_peak": 0.2317549682804538}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9502864036477493, "全帯域_10ms_rms_envelope_nrmse": 0.07593648361341654, "全帯域_10ms_envelope_correlation": 0.9941197701179001, "全帯域_10ms_modulation_mae_db": 0.4209483282696702, "全帯域_40ms_log_rms_mae_db": 0.4814940754254504, "全帯域_40ms_rms_envelope_nrmse": 0.043645583944311755, "全帯域_40ms_envelope_correlation": 0.9979171540033028, "全帯域_40ms_modulation_mae_db": 0.6351010650425158, "300–1000 Hz_10ms_log_rms_mae_db": 1.5420755636256935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09626786642754286, "300–1000 Hz_10ms_envelope_correlation": 0.9919486946574135, "300–1000 Hz_10ms_modulation_mae_db": 0.7344866471370265, "300–1000 Hz_40ms_log_rms_mae_db": 0.9411374086852697, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057993412031036175, "300–1000 Hz_40ms_envelope_correlation": 0.9968909675066726, "300–1000 Hz_40ms_modulation_mae_db": 0.5050158541562743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6038545407093612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13818675660716925, "1000–3000 Hz_10ms_envelope_correlation": 0.9844674883119756, "1000–3000 Hz_10ms_modulation_mae_db": 0.8271994919957346, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8960439712493106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08344734141128404, "1000–3000 Hz_40ms_envelope_correlation": 0.9940093105465434, "1000–3000 Hz_40ms_modulation_mae_db": 1.0413357230860874, "3000–6000 Hz_10ms_log_rms_mae_db": 1.682753261190368, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19950310057231274, "3000–6000 Hz_10ms_envelope_correlation": 0.9652197677754272, "3000–6000 Hz_10ms_modulation_mae_db": 0.29417694186056226, "3000–6000 Hz_40ms_log_rms_mae_db": 1.002696321890545, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12283170053544791, "3000–6000 Hz_40ms_envelope_correlation": 0.9854971264805882, "3000–6000 Hz_40ms_modulation_mae_db": 0.36795608447984823, "mr_stft_db": 5.643100137563446, "band_mae_db": 0.22608782072284242, "rms_gain_error_db": -0.0777180399790364, "output_peak": 0.23850229303184056}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9293700384614165, "全帯域_10ms_rms_envelope_nrmse": 0.07419576845788187, "全帯域_10ms_envelope_correlation": 0.9944489581253434, "全帯域_10ms_modulation_mae_db": 0.3133476458702608, "全帯域_40ms_log_rms_mae_db": 0.4602407355803857, "全帯域_40ms_rms_envelope_nrmse": 0.03972107595258074, "全帯域_40ms_envelope_correlation": 0.9983707460694257, "全帯域_40ms_modulation_mae_db": 0.32621353068711784, "300–1000 Hz_10ms_log_rms_mae_db": 1.5257500419642613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09921236602289074, "300–1000 Hz_10ms_envelope_correlation": 0.9915444220269152, "300–1000 Hz_10ms_modulation_mae_db": 0.9890165020507826, "300–1000 Hz_40ms_log_rms_mae_db": 0.8947090971138362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05861960076903218, "300–1000 Hz_40ms_envelope_correlation": 0.9969662574355375, "300–1000 Hz_40ms_modulation_mae_db": 0.5355696516808719, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5605600505692259, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13046930047651725, "1000–3000 Hz_10ms_envelope_correlation": 0.9865723700216704, "1000–3000 Hz_10ms_modulation_mae_db": 0.8319513335994544, "1000–3000 Hz_40ms_log_rms_mae_db": 0.876406470819366, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08064710766165863, "1000–3000 Hz_40ms_envelope_correlation": 0.9949706383060961, "1000–3000 Hz_40ms_modulation_mae_db": 0.9657214180623495, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6252041359835876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20099225483152733, "3000–6000 Hz_10ms_envelope_correlation": 0.9652336334055729, "3000–6000 Hz_10ms_modulation_mae_db": 0.6012203741019237, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9814688146634756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1250287823449116, "3000–6000 Hz_40ms_envelope_correlation": 0.9849367251180415, "3000–6000 Hz_40ms_modulation_mae_db": 0.599463323337894, "mr_stft_db": 5.6121206718681265, "band_mae_db": 0.2962396448982902, "rms_gain_error_db": -0.11058627641390976, "output_peak": 0.23197398731331614}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6513586120379715, "spectral_convergence": 0.23934292602823756, "spectral_convergence_status": "ok", "amplitude_floor": 6.60007588392934e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.461577887944173, "spectral_convergence": 0.2371099056957538, "spectral_convergence_status": "ok", "amplitude_floor": 5.570424644238683e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8013522931940456, "spectral_convergence": 0.2404663298771505, "spectral_convergence_status": "ok", "amplitude_floor": 3.8555660806566813e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.162305217059986e-05, "output_power": 3.35197431070069e-05, "delta_db": 0.25296897032063453, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000304634409424632, "output_power": 0.00029743654431596696, "delta_db": -0.10384629985455111, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004181285613649292, "output_power": 0.00039919685307046067, "delta_db": -0.20122725164726035, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001153043826104485, "output_power": 0.00011540048082603506, "delta_db": 0.003618036181994691, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.6981779049220075e-05, "output_power": 5.460322379873347e-05, "delta_db": -0.18517719981963532, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.999437610841641e-06, "output_power": 3.350592089276231e-06, "delta_db": 0.48081725875069875, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20460916942912913, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12397842735698765, "pitch": null, "output_rms_dbfs": -30.440682682000325, "output_peak": 0.23613277954677747, "reference_rms_dbfs": -30.31670425464334, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.638096264392064}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.667347842324586e-05, "log_rms_mae_db": 0.968250897643232, "rms_envelope_nrmse": 0.0781406535662062, "envelope_correlation": 0.9938717300066603, "rms_gain_error_db": -0.12349364693722811, "reference_mean_power": 0.0009300339703354519, "output_mean_power": 0.0009039604863715627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3130295372770493e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001632046380308234, "output_power": 0.0001547381597976567, "reference_active": true, "delta_db": -0.23135068825736124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.195140055898603e-05, "output_power": 2.768479103094335e-05, "reference_active": true, "delta_db": -0.6224865005472778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.743176048402797e-06, "output_power": 3.4837529533614196e-06, "reference_active": true, "delta_db": -0.3119290336116995}], "active_band_count": 3, "active_band_mae_db": 0.3885887408054462}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.141277463219576e-05, "log_rms_mae_db": 0.5033273483124855, "rms_envelope_nrmse": 0.04447663332756445, "envelope_correlation": 0.9979676932724085, "rms_gain_error_db": -0.12369133907718347, "reference_mean_power": 0.0009316560347730994, "output_mean_power": 0.0009054958568858885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0476212509594803e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014768349458157853, "output_power": 0.00013952016600253428, "reference_active": true, "delta_db": -0.24694976041168795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4986865819787958e-05, "output_power": 1.3181177606779069e-05, "reference_active": true, "delta_db": -0.5575660713385241}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9796240655945466e-07, "output_power": 3.2590373343833e-07, "reference_active": true, "delta_db": -0.8675271285020519}], "active_band_count": 3, "active_band_mae_db": 0.5573476534174213}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.73108999878072e-05, "log_rms_mae_db": 1.5125491133237599, "rms_envelope_nrmse": 0.10297436558200063, "envelope_correlation": 0.9910205245217738, "rms_gain_error_db": -0.2028146491482874, "reference_mean_power": 0.0005508019269041907, "output_mean_power": 0.0005256709604084421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0420364913089255e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.387221376866869e-05, "output_power": 8.450058097464167e-05, "reference_active": true, "delta_db": -0.45677364957141375}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.257079739763329e-05, "output_power": 1.8219045687601566e-05, "reference_active": true, "delta_db": -0.9302127750505972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1515320396484854e-06, "output_power": 2.647298898159801e-06, "reference_active": true, "delta_db": 0.9005516110338403}], "active_band_count": 3, "active_band_mae_db": 0.7625126785519504}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091336805501339e-05, "log_rms_mae_db": 0.8802994083871603, "rms_envelope_nrmse": 0.06178330473028664, "envelope_correlation": 0.996808245329466, "rms_gain_error_db": -0.2033231432124991, "reference_mean_power": 0.0005517580103658142, "output_mean_power": 0.0005265217700155656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.88495020416125e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.537091697949908e-05, "output_power": 7.644082001769771e-05, "reference_active": true, "delta_db": -0.4798460896186157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0155340206288246e-05, "output_power": 8.326386518189486e-06, "reference_active": true, "delta_db": -0.8623790960120036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.398901151759038e-07, "output_power": 2.37134950320317e-07, "reference_active": true, "delta_db": -0.050167854294977283}], "active_band_count": 3, "active_band_mae_db": 0.4641310133085322}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.045309710707616e-05, "log_rms_mae_db": 1.5659655392300411, "rms_envelope_nrmse": 0.13393429287349873, "envelope_correlation": 0.9853607577158263, "rms_gain_error_db": -0.06170554094005159, "reference_mean_power": 0.00014468505517057894, "output_mean_power": 0.00014264387235789252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.678016723697031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.684581121209281e-05, "output_power": 3.498736418766044e-05, "reference_active": true, "delta_db": -0.22476896911637564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.802139012131204e-06, "output_power": 6.859162731164887e-06, "reference_active": true, "delta_db": -0.5594257748770998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.902802552587548e-06, "output_power": 1.2421138289312804e-06, "reference_active": true, "delta_db": -1.8523232837222061}], "active_band_count": 3, "active_band_mae_db": 0.8788393425718938}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.539248843957361e-05, "log_rms_mae_db": 0.8515766092338539, "rms_envelope_nrmse": 0.07701427800400369, "envelope_correlation": 0.9947376529754964, "rms_gain_error_db": -0.06091267010165556, "reference_mean_power": 0.0001448248558759811, "output_mean_power": 0.000142807770145294}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.24555278166651e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.61104008282169e-05, "output_power": 3.3991046382215004e-05, "reference_active": true, "delta_db": -0.2626777502189195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6607268103199764e-06, "output_power": 3.2413841264112983e-06, "reference_active": true, "delta_db": -0.5283681894454733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4925257777124562e-07, "output_power": 8.960167068935651e-08, "reference_active": true, "delta_db": -2.2160573317426433}], "active_band_count": 3, "active_band_mae_db": 1.0023677571356788}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.667503736683173e-05, "log_rms_mae_db": 1.65632058006543, "rms_envelope_nrmse": 0.21105934268131082, "envelope_correlation": 0.9610950946155086, "rms_gain_error_db": -0.0950962848439396, "reference_mean_power": 8.017666159457313e-06, "output_mean_power": 7.844013647358817e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.8921445305362116e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.737362730986327e-06, "output_power": 2.6250298047277758e-06, "reference_active": true, "delta_db": -0.18198111296642813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.579169756744011e-07, "output_power": 7.186900654675204e-07, "reference_active": true, "delta_db": -0.23079992819177636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.809426009650544e-07, "output_power": 1.6711019524138074e-07, "reference_active": true, "delta_db": -0.3453788219116246}], "active_band_count": 3, "active_band_mae_db": 0.2527199543566097}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4241724067408686e-05, "log_rms_mae_db": 0.9886553762489769, "rms_envelope_nrmse": 0.11474495727201602, "envelope_correlation": 0.987235258496505, "rms_gain_error_db": -0.09517810017775868, "reference_mean_power": 8.048582997885633e-06, "output_mean_power": 7.874112526897374e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.450576990854383e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.750016160596199e-06, "output_power": 2.6576473890323702e-06, "reference_active": true, "delta_db": -0.14837886822685184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.547469624529576e-07, "output_power": 3.485260239800213e-07, "reference_active": true, "delta_db": -0.07683473624091886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2784081824019638e-08, "output_power": 1.0179584007322829e-08, "reference_active": true, "delta_db": -0.9893951091409172}], "active_band_count": 3, "active_band_mae_db": 0.40486957120289596}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8969585982091297, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "20aa578ac5fc3d18dbc3dadecbd151755579e3123b0987194baeeb522fd53575", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9849813901268851, "standard_deviation": 0.0017793778834085034, "minimum": 0.9822196309895225, "maximum": 0.9873664402069692, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5880243608198028, "standard_deviation": 0.02138720289101061, "minimum": 1.5605600505692259, "maximum": 1.6286241223987774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8105089380852274, "standard_deviation": 0.1770114041357294, "minimum": 0.6351813530855926, "maximum": 1.190522734882356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13598519762385539, "standard_deviation": 0.007636714674512637, "minimum": 0.12463196886547047, "maximum": 0.14759277902169038, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949916660695235, "standard_deviation": 0.0005972086727600523, "minimum": 0.9940093105465434, "maximum": 0.9958087943934774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8756544055806468, "standard_deviation": 0.014871776909335735, "minimum": 0.8515766092338539, "maximum": 0.8960439712493106, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8854370233939569, "standard_deviation": 0.2504075314448237, "minimum": 0.46503525742973534, "maximum": 1.253363630620734, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07683288359820015, "standard_deviation": 0.004264015582191669, "minimum": 0.07057351670742598, "maximum": 0.08344734141128404, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9634473359268203, "standard_deviation": 0.0027755096809236, "minimum": 0.9576488329537112, "maximum": 0.9661073781527953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7023643679311373, "standard_deviation": 0.044004938612589334, "minimum": 1.6252041359835876, "maximum": 1.7454602430787372, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.49586190337117786, "standard_deviation": 0.32553082484518253, "minimum": 0.06876661709813887, "maximum": 1.1140199383456781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20486481636221368, "standard_deviation": 0.007309622747768095, "minimum": 0.19696243674566097, "maximum": 0.21980998663537363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9857131359729214, "standard_deviation": 0.001371094398080432, "minimum": 0.9829599024052236, "maximum": 0.987235258496505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.018013091257219, "standard_deviation": 0.0299539749472615, "minimum": 0.9814688146634756, "maximum": 1.0668024501597468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5453758750560846, "standard_deviation": 0.3066207489169323, "minimum": 0.13447508971791802, "maximum": 1.1319450923315177, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12157871260035308, "standard_deviation": 0.005713097848827849, "minimum": 0.11474495727201602, "maximum": 0.1327174836603707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920665605454441, "standard_deviation": 0.0008794369751330173, "minimum": 0.9910205245217738, "maximum": 0.9935844323904893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5258541456192078, "standard_deviation": 0.031461298845065724, "minimum": 1.4630455485317884, "maximum": 1.574655473323641, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6224057913784298, "standard_deviation": 0.24994169315019152, "minimum": 0.2689635473866031, "maximum": 0.9890165020507826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09622223183988579, "standard_deviation": 0.0056745558656703, "minimum": 0.08644814663676405, "maximum": 0.10297436558200063, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970902671755025, "standard_deviation": 0.0005254062687462572, "minimum": 0.9966000090516594, "maximum": 0.9982799286251283, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9088085995043893, "standard_deviation": 0.055638994660074804, "minimum": 0.7929563795949984, "maximum": 0.967956787635875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5301978552176378, "standard_deviation": 0.05241667171920751, "minimum": 0.4641310133085322, "maximum": 0.6053061690104979, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05755268981920732, "standard_deviation": 0.006049163404454815, "minimum": 0.04467408639546606, "maximum": 0.0655104383262822, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23043833865620406, "standard_deviation": 0.03865019174360721, "minimum": 0.17452531581311204, "maximum": 0.2962396448982902, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.635204220062669, "standard_deviation": 0.01122258292497132, "minimum": 5.6121206718681265, "maximum": 5.646300644232664, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2362532217688018, "standard_deviation": 0.005487549481606477, "minimum": 0.22956288513034334, "maximum": 0.24596651996125724, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11327520251260134, "standard_deviation": 0.03407311223530226, "minimum": -0.16836742463028315, "maximum": -0.05770763091759554, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943630429246008, "standard_deviation": 0.0005625470385955515, "minimum": 0.9934623105555209, "maximum": 0.9952626316602466, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.957806704190207, "standard_deviation": 0.01610845780614909, "minimum": 0.9293700384614165, "maximum": 0.9807857669166379, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.366083318466065, "standard_deviation": 0.0846373271173061, "minimum": 0.2655267992323261, "maximum": 0.5402867315550574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07483664826531841, "standard_deviation": 0.0035030218554337874, "minimum": 0.06874692456544915, "maximum": 0.07986305282046968, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982605088953318, "standard_deviation": 0.0003538578545749371, "minimum": 0.9978189919364144, "maximum": 0.9989029186892573, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5014995898363888, "standard_deviation": 0.02818538356324051, "minimum": 0.4602407355803857, "maximum": 0.5324404849180865, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5040217097393866, "standard_deviation": 0.15300639087242823, "minimum": 0.2931507806212379, "maximum": 0.7194194647301443, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041075465609791945, "standard_deviation": 0.0036413702205291077, "minimum": 0.03341942582851839, "maximum": 0.04447663332756445, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-126133-0003", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5358200073242188, "predicted_mos": 2.5358200073242188, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d1a09e0b23bd935ed983674013cab4cd5d7db5122394a350c638ddb7002dd2e9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.348977899644524, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9733871398464375, "全帯域_10ms_rms_envelope_nrmse": 0.08013796297091898, "全帯域_10ms_envelope_correlation": 0.9935295856032041, "全帯域_10ms_modulation_mae_db": 0.37617971398789546, "全帯域_40ms_log_rms_mae_db": 0.512100142048742, "全帯域_40ms_rms_envelope_nrmse": 0.045036426076469294, "全帯域_40ms_envelope_correlation": 0.9978959192757616, "全帯域_40ms_modulation_mae_db": 0.5159306013300188, "300–1000 Hz_10ms_log_rms_mae_db": 1.5392520410989283, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10684585885074778, "300–1000 Hz_10ms_envelope_correlation": 0.9902485175871075, "300–1000 Hz_10ms_modulation_mae_db": 0.703099032240787, "300–1000 Hz_40ms_log_rms_mae_db": 0.9187725041217512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06314567108480446, "300–1000 Hz_40ms_envelope_correlation": 0.9965815684311172, "300–1000 Hz_40ms_modulation_mae_db": 0.5284039084139894, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5936475648967199, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1323672597649862, "1000–3000 Hz_10ms_envelope_correlation": 0.9857060048197752, "1000–3000 Hz_10ms_modulation_mae_db": 1.0706998085819481, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8514347324210138, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07499519583803724, "1000–3000 Hz_40ms_envelope_correlation": 0.9950444837226707, "1000–3000 Hz_40ms_modulation_mae_db": 1.1975536694198687, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6023198003796586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20345734564846787, "3000–6000 Hz_10ms_envelope_correlation": 0.9639410231461533, "3000–6000 Hz_10ms_modulation_mae_db": 0.34979792720312125, "3000–6000 Hz_40ms_log_rms_mae_db": 1.000062625039905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10902062159707218, "3000–6000 Hz_40ms_envelope_correlation": 0.9890051013243161, "3000–6000 Hz_40ms_modulation_mae_db": 0.5219148271393598, "mr_stft_db": 5.558443072979904, "band_mae_db": 0.17161653782218247, "rms_gain_error_db": -0.12050526170075493, "output_peak": 0.23635210415067623}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9696841378711754, "全帯域_10ms_rms_envelope_nrmse": 0.07589894216723504, "全帯域_10ms_envelope_correlation": 0.9943655273918858, "全帯域_10ms_modulation_mae_db": 0.2273858601209101, "全帯域_40ms_log_rms_mae_db": 0.5204824139433376, "全帯域_40ms_rms_envelope_nrmse": 0.04296081978525674, "全帯域_40ms_envelope_correlation": 0.9982967892079145, "全帯域_40ms_modulation_mae_db": 0.43657629498011286, "300–1000 Hz_10ms_log_rms_mae_db": 1.5877910966865552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10499897933407595, "300–1000 Hz_10ms_envelope_correlation": 0.99081107311013, "300–1000 Hz_10ms_modulation_mae_db": 0.460631891219319, "300–1000 Hz_40ms_log_rms_mae_db": 0.9830440403272599, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06538557379977294, "300–1000 Hz_40ms_envelope_correlation": 0.9965722865350729, "300–1000 Hz_40ms_modulation_mae_db": 0.5096729739079184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6024331157828366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13846152784441595, "1000–3000 Hz_10ms_envelope_correlation": 0.9844080389684283, "1000–3000 Hz_10ms_modulation_mae_db": 0.6408296154688304, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8815716029169159, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07298496515141623, "1000–3000 Hz_40ms_envelope_correlation": 0.995340401442851, "1000–3000 Hz_40ms_modulation_mae_db": 0.5662054146122432, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6756945235009177, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18584613024925498, "3000–6000 Hz_10ms_envelope_correlation": 0.9701323720496959, "3000–6000 Hz_10ms_modulation_mae_db": 1.2047821829258758, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0139646202451285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11521232830566294, "3000–6000 Hz_40ms_envelope_correlation": 0.9878297818380214, "3000–6000 Hz_40ms_modulation_mae_db": 1.2908857415783914, "mr_stft_db": 5.557493943295657, "band_mae_db": 0.21219895158779534, "rms_gain_error_db": -0.16889229113460852, "output_peak": 0.24208216920793899}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9730059011361153, "全帯域_10ms_rms_envelope_nrmse": 0.07046909199588396, "全帯域_10ms_envelope_correlation": 0.9949967731390232, "全帯域_10ms_modulation_mae_db": 0.3338881580159711, "全帯域_40ms_log_rms_mae_db": 0.495891949984333, "全帯域_40ms_rms_envelope_nrmse": 0.033986970052496866, "全帯域_40ms_envelope_correlation": 0.9988373885339679, "全帯域_40ms_modulation_mae_db": 0.2590182419805904, "300–1000 Hz_10ms_log_rms_mae_db": 1.5190072441669271, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08863234804556384, "300–1000 Hz_10ms_envelope_correlation": 0.9932226314856325, "300–1000 Hz_10ms_modulation_mae_db": 0.7839073003248359, "300–1000 Hz_40ms_log_rms_mae_db": 0.8306378454307591, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04452391950228949, "300–1000 Hz_40ms_envelope_correlation": 0.9982596041990951, "300–1000 Hz_40ms_modulation_mae_db": 0.4475055829162112, "1000–3000 Hz_10ms_log_rms_mae_db": 1.590816811313515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1310285126811309, "1000–3000 Hz_10ms_envelope_correlation": 0.986202845943351, "1000–3000 Hz_10ms_modulation_mae_db": 1.270356723540431, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8392149383659698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07035260975627358, "1000–3000 Hz_40ms_envelope_correlation": 0.9960460364490188, "1000–3000 Hz_40ms_modulation_mae_db": 1.2999343484542096, "3000–6000 Hz_10ms_log_rms_mae_db": 1.679229406460379, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19298067531419583, "3000–6000 Hz_10ms_envelope_correlation": 0.967578481213242, "3000–6000 Hz_10ms_modulation_mae_db": 0.6150047327774391, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0042746572086938, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11907775544831829, "3000–6000 Hz_40ms_envelope_correlation": 0.9866867342556889, "3000–6000 Hz_40ms_modulation_mae_db": 0.6866470011941127, "mr_stft_db": 5.569858739985836, "band_mae_db": 0.20010698867899904, "rms_gain_error_db": -0.103841635944227, "output_peak": 0.2446720573374879}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9442043205780665, "全帯域_10ms_rms_envelope_nrmse": 0.07280237995660128, "全帯域_10ms_envelope_correlation": 0.9947448843127533, "全帯域_10ms_modulation_mae_db": 0.12385139807307198, "全帯域_40ms_log_rms_mae_db": 0.5029364029831256, "全帯域_40ms_rms_envelope_nrmse": 0.03973760052213287, "全帯域_40ms_envelope_correlation": 0.9984744595060424, "全帯域_40ms_modulation_mae_db": 0.286204884633567, "300–1000 Hz_10ms_log_rms_mae_db": 1.6300197252122641, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09202225261775135, "300–1000 Hz_10ms_envelope_correlation": 0.9928015946083181, "300–1000 Hz_10ms_modulation_mae_db": 0.4630323522987248, "300–1000 Hz_40ms_log_rms_mae_db": 0.952848803208832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05630380536181015, "300–1000 Hz_40ms_envelope_correlation": 0.9972727341077441, "300–1000 Hz_40ms_modulation_mae_db": 0.463115168247453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5952164765571912, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1277684322980473, "1000–3000 Hz_10ms_envelope_correlation": 0.9867917476547698, "1000–3000 Hz_10ms_modulation_mae_db": 0.5223305116573692, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8502334035684016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07285163971822343, "1000–3000 Hz_40ms_envelope_correlation": 0.9955358175429101, "1000–3000 Hz_40ms_modulation_mae_db": 0.5212275615326484, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6568919271149085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18877339194606055, "3000–6000 Hz_10ms_envelope_correlation": 0.9689566180391617, "3000–6000 Hz_10ms_modulation_mae_db": 0.4693158089038965, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9549718647441745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10831667928067402, "3000–6000 Hz_40ms_envelope_correlation": 0.9887677232570271, "3000–6000 Hz_40ms_modulation_mae_db": 0.3957642269839749, "mr_stft_db": 5.565093240082831, "band_mae_db": 0.1671055331994932, "rms_gain_error_db": -0.1395405709082, "output_peak": 0.22860452416040422}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9804772885486972, "全帯域_10ms_rms_envelope_nrmse": 0.08137928716252718, "全帯域_10ms_envelope_correlation": 0.9932135459640393, "全帯域_10ms_modulation_mae_db": 0.5396421318810914, "全帯域_40ms_log_rms_mae_db": 0.5459920360407005, "全帯域_40ms_rms_envelope_nrmse": 0.045652489986816015, "全帯域_40ms_envelope_correlation": 0.9976830072977524, "全帯域_40ms_modulation_mae_db": 0.7081384219644248, "300–1000 Hz_10ms_log_rms_mae_db": 1.6110628097640578, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10300582045017352, "300–1000 Hz_10ms_envelope_correlation": 0.9907294738119888, "300–1000 Hz_10ms_modulation_mae_db": 0.31109514580869374, "300–1000 Hz_40ms_log_rms_mae_db": 0.9716178909551278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061754703478050524, "300–1000 Hz_40ms_envelope_correlation": 0.9963892004509661, "300–1000 Hz_40ms_modulation_mae_db": 0.6260324674468375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.653916430243542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1480300642110587, "1000–3000 Hz_10ms_envelope_correlation": 0.9821901379963401, "1000–3000 Hz_10ms_modulation_mae_db": 0.9153147012570094, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8750954095290253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07896016857852638, "1000–3000 Hz_40ms_envelope_correlation": 0.994781966210183, "1000–3000 Hz_40ms_modulation_mae_db": 1.144838776800902, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6795526940387306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1876504493540728, "3000–6000 Hz_10ms_envelope_correlation": 0.969621740860901, "3000–6000 Hz_10ms_modulation_mae_db": 0.0827025531688323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0118075869932517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1066615046601551, "3000–6000 Hz_40ms_envelope_correlation": 0.9891050610434408, "3000–6000 Hz_40ms_modulation_mae_db": 0.29632149015036696, "mr_stft_db": 5.574999317539469, "band_mae_db": 0.13865188665259676, "rms_gain_error_db": -0.06292537316944037, "output_peak": 0.2424191290191037}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9744298416183778, "全帯域_10ms_rms_envelope_nrmse": 0.0762860900757478, "全帯域_10ms_envelope_correlation": 0.9941605770894206, "全帯域_10ms_modulation_mae_db": 0.3791824798081282, "全帯域_40ms_log_rms_mae_db": 0.5349905295441084, "全帯域_40ms_rms_envelope_nrmse": 0.04285314088709496, "全帯域_40ms_envelope_correlation": 0.9981246586201243, "全帯域_40ms_modulation_mae_db": 0.6284889416113226, "300–1000 Hz_10ms_log_rms_mae_db": 1.5980736401899087, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10005752945204426, "300–1000 Hz_10ms_envelope_correlation": 0.9913342761723138, "300–1000 Hz_10ms_modulation_mae_db": 0.5934556095252865, "300–1000 Hz_40ms_log_rms_mae_db": 0.9530204177101536, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060333780309291554, "300–1000 Hz_40ms_envelope_correlation": 0.9966862510418102, "300–1000 Hz_40ms_modulation_mae_db": 0.47904850182483844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5826593097950485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13100666518848633, "1000–3000 Hz_10ms_envelope_correlation": 0.9863206505204911, "1000–3000 Hz_10ms_modulation_mae_db": 0.8164098222526665, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8526009706100393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07435709153562106, "1000–3000 Hz_40ms_envelope_correlation": 0.9956725468681255, "1000–3000 Hz_40ms_modulation_mae_db": 1.0622394498913232, "3000–6000 Hz_10ms_log_rms_mae_db": 1.672512254893043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2020695582551295, "3000–6000 Hz_10ms_envelope_correlation": 0.9644436358117786, "3000–6000 Hz_10ms_modulation_mae_db": 0.5557080348810693, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0222061177142057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11852564309556675, "3000–6000 Hz_40ms_envelope_correlation": 0.9868174495356257, "3000–6000 Hz_40ms_modulation_mae_db": 0.6541190982257298, "mr_stft_db": 5.560108809574757, "band_mae_db": 0.22704246276831122, "rms_gain_error_db": -0.12506855122322602, "output_peak": 0.23232961925201748}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9682816000157135, "全帯域_10ms_rms_envelope_nrmse": 0.07624688858638612, "全帯域_10ms_envelope_correlation": 0.9940688090137229, "全帯域_10ms_modulation_mae_db": 0.2925279295390552, "全帯域_40ms_log_rms_mae_db": 0.49297557392896607, "全帯域_40ms_rms_envelope_nrmse": 0.042571833355178085, "全帯域_40ms_envelope_correlation": 0.9980200330191192, "全帯域_40ms_modulation_mae_db": 0.4876607533715786, "300–1000 Hz_10ms_log_rms_mae_db": 1.6010037590390482, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09729728561695006, "300–1000 Hz_10ms_envelope_correlation": 0.9917543219201712, "300–1000 Hz_10ms_modulation_mae_db": 0.877986014422171, "300–1000 Hz_40ms_log_rms_mae_db": 0.9447255710908389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055387012105012005, "300–1000 Hz_40ms_envelope_correlation": 0.9971397813552203, "300–1000 Hz_40ms_modulation_mae_db": 0.35017828945523455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6028225011023776, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13370014967190919, "1000–3000 Hz_10ms_envelope_correlation": 0.9856648938606956, "1000–3000 Hz_10ms_modulation_mae_db": 0.9702567777880192, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8779988474037993, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08010737587107836, "1000–3000 Hz_40ms_envelope_correlation": 0.9947986201650231, "1000–3000 Hz_40ms_modulation_mae_db": 1.270188342275496, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6104563692997798, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19267882416492546, "3000–6000 Hz_10ms_envelope_correlation": 0.9678247811995307, "3000–6000 Hz_10ms_modulation_mae_db": 0.4319610264917129, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9482831592048104, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11902812607012982, "3000–6000 Hz_40ms_envelope_correlation": 0.9870352220685055, "3000–6000 Hz_40ms_modulation_mae_db": 0.537476536342653, "mr_stft_db": 5.58074363914544, "band_mae_db": 0.20208077820357928, "rms_gain_error_db": -0.07498914814652788, "output_peak": 0.23146326622601457}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9289894543012163, "全帯域_10ms_rms_envelope_nrmse": 0.07448866529321449, "全帯域_10ms_envelope_correlation": 0.9943848116965626, "全帯域_10ms_modulation_mae_db": 0.2917145906515965, "全帯域_40ms_log_rms_mae_db": 0.4584582440285208, "全帯域_40ms_rms_envelope_nrmse": 0.04010868538926913, "全帯域_40ms_envelope_correlation": 0.9983100564771754, "全帯域_40ms_modulation_mae_db": 0.3258020594325652, "300–1000 Hz_10ms_log_rms_mae_db": 1.5521727660963587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09869632810762709, "300–1000 Hz_10ms_envelope_correlation": 0.9915873132487567, "300–1000 Hz_10ms_modulation_mae_db": 0.8879430328464735, "300–1000 Hz_40ms_log_rms_mae_db": 0.8967002456816445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05933352774621353, "300–1000 Hz_40ms_envelope_correlation": 0.9968192911473079, "300–1000 Hz_40ms_modulation_mae_db": 0.4604257097802253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5670874671864818, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13034428095848005, "1000–3000 Hz_10ms_envelope_correlation": 0.9866964253753895, "1000–3000 Hz_10ms_modulation_mae_db": 1.0252577808177765, "1000–3000 Hz_40ms_log_rms_mae_db": 0.860215957066437, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08048962987642422, "1000–3000 Hz_40ms_envelope_correlation": 0.9951236247378766, "1000–3000 Hz_40ms_modulation_mae_db": 1.1809896082981224, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5759092401461148, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1863836073791597, "3000–6000 Hz_10ms_envelope_correlation": 0.9698346189735633, "3000–6000 Hz_10ms_modulation_mae_db": 0.5857084031751386, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9624364179119811, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11601011491941095, "3000–6000 Hz_40ms_envelope_correlation": 0.9869931626935439, "3000–6000 Hz_40ms_modulation_mae_db": 0.6944379807659867, "mr_stft_db": 5.544042714250961, "band_mae_db": 0.25937258006006836, "rms_gain_error_db": -0.10214999753955675, "output_peak": 0.2331944733683972}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.574395457971397, "spectral_convergence": 0.24076655938846256, "spectral_convergence_status": "ok", "amplitude_floor": 6.60007588392934e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.37590798260745, "spectral_convergence": 0.23844134144314172, "spectral_convergence_status": "ok", "amplitude_floor": 5.570424644238683e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.725025778360865, "spectral_convergence": 0.24028305455957996, "spectral_convergence_status": "ok", "amplitude_floor": 3.8555660806566813e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.162305217059986e-05, "output_power": 3.30835120813327e-05, "delta_db": 0.19607822649765372, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000304634409424632, "output_power": 0.00029836424589357153, "delta_db": -0.09032177930242379, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004181285613649292, "output_power": 0.00040076599937598017, "delta_db": -0.18418964912087882, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001153043826104485, "output_power": 0.00011507012955960894, "delta_db": -0.008832126562387889, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.6981779049220075e-05, "output_power": 5.373573435628152e-05, "delta_db": -0.2547281603178572, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.999437610841641e-06, "output_power": 3.2106632825055372e-06, "delta_db": 0.2955492851318934, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17161653782218247, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12050526170075493, "pitch": null, "output_rms_dbfs": -30.437209516344094, "output_peak": 0.23635210415067623, "reference_rms_dbfs": -30.31670425464334, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.558443072979904}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.667347842324586e-05, "log_rms_mae_db": 0.9733871398464375, "rms_envelope_nrmse": 0.08013796297091898, "envelope_correlation": 0.9935295856032041, "rms_gain_error_db": -0.12016044482051799, "reference_mean_power": 0.0009300339703354519, "output_mean_power": 0.000904654540681687}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3130295372770493e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001632046380308234, "output_power": 0.0001553511576225124, "reference_active": true, "delta_db": -0.21418002776773531}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.195140055898603e-05, "output_power": 2.7537389367615973e-05, "reference_active": true, "delta_db": -0.6456713442386706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.743176048402797e-06, "output_power": 3.5186127839896538e-06, "reference_active": true, "delta_db": -0.26868776995728044}], "active_band_count": 3, "active_band_mae_db": 0.37617971398789546}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.141277463219576e-05, "log_rms_mae_db": 0.512100142048742, "rms_envelope_nrmse": 0.045036426076469294, "envelope_correlation": 0.9978959192757616, "rms_gain_error_db": -0.12042054317944725, "reference_mean_power": 0.0009316560347730994, "output_mean_power": 0.0009061780685660082}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0476212509594803e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014768349458157853, "output_power": 0.0001400218567480723, "reference_active": true, "delta_db": -0.23136128078412352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4986865819787958e-05, "output_power": 1.3078193582858694e-05, "reference_active": true, "delta_db": -0.5916305752384036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9796240655945466e-07, "output_power": 3.367922223724472e-07, "reference_active": true, "delta_db": -0.7247999479675294}], "active_band_count": 3, "active_band_mae_db": 0.5159306013300188}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.73108999878072e-05, "log_rms_mae_db": 1.5392520410989283, "rms_envelope_nrmse": 0.10684585885074778, "envelope_correlation": 0.9902485175871075, "rms_gain_error_db": -0.1895827151756609, "reference_mean_power": 0.0005508019269041907, "output_mean_power": 0.0005272749988215628}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0420364913089255e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.387221376866869e-05, "output_power": 8.491656050584592e-05, "reference_active": true, "delta_db": -0.4354466484831576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.257079739763329e-05, "output_power": 1.8097969015742318e-05, "reference_active": true, "delta_db": -0.9591706206897926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1515320396484854e-06, "output_power": 2.536388646673546e-06, "reference_active": true, "delta_db": 0.7146798275494105}], "active_band_count": 3, "active_band_mae_db": 0.703099032240787}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091336805501339e-05, "log_rms_mae_db": 0.9187725041217512, "rms_envelope_nrmse": 0.06314567108480446, "envelope_correlation": 0.9965815684311172, "rms_gain_error_db": -0.19030246373783, "reference_mean_power": 0.0005517580103658142, "output_mean_power": 0.0005281027153967347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.88495020416125e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.537091697949908e-05, "output_power": 7.672167478981417e-05, "reference_active": true, "delta_db": -0.4639187159347399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0155340206288246e-05, "output_power": 8.277365844126468e-06, "reference_active": true, "delta_db": -0.8880232632308833}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.398901151759038e-07, "output_power": 2.2734498489486714e-07, "reference_active": true, "delta_db": -0.23326974607634504}], "active_band_count": 3, "active_band_mae_db": 0.5284039084139894}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.045309710707616e-05, "log_rms_mae_db": 1.5936475648967199, "rms_envelope_nrmse": 0.1323672597649862, "envelope_correlation": 0.9857060048197752, "rms_gain_error_db": -0.08424746148903092, "reference_mean_power": 0.00014468505517057894, "output_mean_power": 0.0001419054019190836}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.678016723697031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.684581121209281e-05, "output_power": 3.530023007036184e-05, "reference_active": true, "delta_db": -0.1861058663978482}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.802139012131204e-06, "output_power": 6.757178156093937e-06, "reference_active": true, "delta_db": -0.6244831449989864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.902802552587548e-06, "output_power": 1.0945678403172906e-06, "reference_active": true, "delta_db": -2.40151041434901}], "active_band_count": 3, "active_band_mae_db": 1.0706998085819481}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.539248843957361e-05, "log_rms_mae_db": 0.8514347324210138, "rms_envelope_nrmse": 0.07499519583803724, "envelope_correlation": 0.9950444837226707, "rms_gain_error_db": -0.08288298094203395, "reference_mean_power": 0.0001448248558759811, "output_mean_power": 0.00014208715119999133}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.24555278166651e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.61104008282169e-05, "output_power": 3.4159902489494855e-05, "reference_active": true, "delta_db": -0.2411568684400036}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6607268103199764e-06, "output_power": 3.1600085458770156e-06, "reference_active": true, "delta_db": -0.6387906280380963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4925257777124562e-07, "output_power": 7.991908251593285e-08, "reference_active": true, "delta_db": -2.712713511781506}], "active_band_count": 3, "active_band_mae_db": 1.1975536694198687}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.667503736683173e-05, "log_rms_mae_db": 1.6023198003796586, "rms_envelope_nrmse": 0.20345734564846787, "envelope_correlation": 0.9639410231461533, "rms_gain_error_db": -0.25900202122996796, "reference_mean_power": 8.017666159457313e-06, "output_mean_power": 7.5534918849163e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.8921445305362116e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.737362730986327e-06, "output_power": 2.5913847296209085e-06, "reference_active": true, "delta_db": -0.2380045481879511}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.579169756744011e-07, "output_power": 7.069951252416301e-07, "reference_active": true, "delta_db": -0.3020521508643349}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.809426009650544e-07, "output_power": 1.6091892421959867e-07, "reference_active": true, "delta_db": -0.5093370825570775}], "active_band_count": 3, "active_band_mae_db": 0.34979792720312125}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4241724067408686e-05, "log_rms_mae_db": 1.000062625039905, "rms_envelope_nrmse": 0.10902062159707218, "envelope_correlation": 0.9890051013243161, "rms_gain_error_db": -0.25895293518280316, "reference_mean_power": 8.048582997885633e-06, "output_mean_power": 7.582704528672813e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.450576990854383e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.750016160596199e-06, "output_power": 2.6201106258647364e-06, "reference_active": true, "delta_db": -0.21015617574687992}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.547469624529576e-07, "output_power": 3.448196406055793e-07, "reference_active": true, "delta_db": -0.12326690600477695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2784081824019638e-08, "output_power": 9.625812585142072e-09, "reference_active": true, "delta_db": -1.2323213996664224}], "active_band_count": 3, "active_band_mae_db": 0.5219148271393598}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8965217235969226, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "31effcaf8635ac32d530123367d5faa5531562d88d0f312ee4635c7e5d8ab674", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.985497593142405, "standard_deviation": 0.0015345847453416316, "minimum": 0.9821901379963401, "maximum": 0.9867917476547698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.598574959609714, "standard_deviation": 0.025174031205324778, "minimum": 1.5670874671864818, "maximum": 1.653916430243542, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9039319676705063, "standard_deviation": 0.24018735525698387, "minimum": 0.5223305116573692, "maximum": 1.270356723540431, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13408836157731432, "standard_deviation": 0.006430149177729917, "minimum": 0.1277684322980473, "maximum": 0.1480300642110587, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9952929371423324, "standard_deviation": 0.00044212832947310035, "minimum": 0.994781966210183, "maximum": 0.9960460364490188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8610457327352004, "standard_deviation": 0.015417878159310497, "minimum": 0.8392149383659698, "maximum": 0.8815716029169159, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0303971464106016, "standard_deviation": 0.3092867284469219, "minimum": 0.5212275615326484, "maximum": 1.2999343484542096, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07563733454070007, "standard_deviation": 0.0037671911646504595, "minimum": 0.07035260975627358, "maximum": 0.08048962987642422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9677916589117533, "standard_deviation": 0.002403632661850117, "minimum": 0.9639410231461533, "maximum": 0.9701323720496959, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6440707769791914, "standard_deviation": 0.04138240502593177, "minimum": 1.5759092401461148, "maximum": 1.6795526940387306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5368725836908856, "standard_deviation": 0.31887867673533726, "minimum": 0.0827025531688323, "maximum": 1.2047821829258758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19247999778890834, "standard_deviation": 0.006876353740979859, "minimum": 0.18584613024925498, "maximum": 0.20345734564846787, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9877800295020212, "standard_deviation": 0.0010371854758235456, "minimum": 0.9866867342556889, "maximum": 0.9891050610434408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9897508811327689, "standard_deviation": 0.029565060930034675, "minimum": 0.9482831592048104, "maximum": 1.0222061177142057, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6346958627975718, "standard_deviation": 0.30023494672062856, "minimum": 0.29632149015036696, "maximum": 1.2908857415783914, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11398159667212376, "standard_deviation": 0.005182422869384277, "minimum": 0.1066615046601551, "maximum": 0.11907775544831829, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915611502430524, "standard_deviation": 0.0010258097614790143, "minimum": 0.9902485175871075, "maximum": 0.9932226314856325, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.579797885281756, "standard_deviation": 0.03863737490044356, "minimum": 1.5190072441669271, "maximum": 1.6300197252122641, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6351437973357864, "standard_deviation": 0.21261987479180647, "minimum": 0.31109514580869374, "maximum": 0.8879430328464735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09894455030936672, "standard_deviation": 0.006254760944378185, "minimum": 0.08863234804556384, "maximum": 0.10684585885074778, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969650896585418, "standard_deviation": 0.0006015420157360365, "minimum": 0.9963892004509661, "maximum": 0.9982596041990951, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9314209148157959, "standard_deviation": 0.04914437188538157, "minimum": 0.8306378454307591, "maximum": 0.9830440403272599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48304782524908846, "standard_deviation": 0.0784249511762596, "minimum": 0.35017828945523455, "maximum": 0.6260324674468375, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.058270999173405585, "standard_deviation": 0.006470474252984945, "minimum": 0.04452391950228949, "maximum": 0.06538557379977294, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19727196487162818, "standard_deviation": 0.03779885492823079, "minimum": 0.13865188665259676, "maximum": 0.25937258006006836, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.563847934606857, "standard_deviation": 0.011484848928279981, "minimum": 5.544042714250961, "maximum": 5.58074363914544, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.23638966784025503, "standard_deviation": 0.005963334613350962, "minimum": 0.22860452416040422, "maximum": 0.2446720573374879, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11223910372081769, "standard_deviation": 0.034183514375600486, "minimum": -0.16889229113460852, "maximum": -0.06292537316944037, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941830642763265, "standard_deviation": 0.0005894693436069489, "minimum": 0.9932135459640393, "maximum": 0.9949967731390232, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9640574604894749, "standard_deviation": 0.01779939656338463, "minimum": 0.9289894543012163, "maximum": 0.9804772885486972, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.320546532759715, "standard_deviation": 0.12171296186798265, "minimum": 0.12385139807307198, "maximum": 0.5396421318810914, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07596366352606436, "standard_deviation": 0.0035717496570786176, "minimum": 0.07046909199588396, "maximum": 0.08137928716252718, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982052889922323, "standard_deviation": 0.0003587000206851349, "minimum": 0.9976830072977524, "maximum": 0.9988373885339679, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5079784115627293, "standard_deviation": 0.027247189724577164, "minimum": 0.4584582440285208, "maximum": 0.5459920360407005, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4559775249130225, "standard_deviation": 0.1616350710207235, "minimum": 0.2590182419805904, "maximum": 0.7081384219644248, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041613495756839244, "standard_deviation": 0.0037080543780623515, "minimum": 0.033986970052496866, "maximum": 0.045652489986816015, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-126133-0003", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.837935209274292, "predicted_mos": 1.837935209274292, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e12253abf0034779718863dead3b37fd4d834e1f9e32caae4152a7128462f16a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.347659000195563, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9410963031210183, "全帯域_10ms_rms_envelope_nrmse": 0.07346596064974466, "全帯域_10ms_envelope_correlation": 0.9944988463416258, "全帯域_10ms_modulation_mae_db": 0.25881389654977666, "全帯域_40ms_log_rms_mae_db": 0.5120190940495198, "全帯域_40ms_rms_envelope_nrmse": 0.04151607360118068, "全帯域_40ms_envelope_correlation": 0.9981256048281743, "全帯域_40ms_modulation_mae_db": 0.46173941070913155, "300–1000 Hz_10ms_log_rms_mae_db": 1.738227007541424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09330267298379508, "300–1000 Hz_10ms_envelope_correlation": 0.9924725826071364, "300–1000 Hz_10ms_modulation_mae_db": 0.860867773710939, "300–1000 Hz_40ms_log_rms_mae_db": 0.9729747831115879, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052949570289131956, "300–1000 Hz_40ms_envelope_correlation": 0.9974718919420684, "300–1000 Hz_40ms_modulation_mae_db": 0.39995807080551365, "1000–3000 Hz_10ms_log_rms_mae_db": 1.89129553533547, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14483616226715584, "1000–3000 Hz_10ms_envelope_correlation": 0.9828523764293149, "1000–3000 Hz_10ms_modulation_mae_db": 0.6447300848001318, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0597186698237482, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08705999382806331, "1000–3000 Hz_40ms_envelope_correlation": 0.993273951968726, "1000–3000 Hz_40ms_modulation_mae_db": 0.8245078679897707, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2371901747583767, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2500783783294953, "3000–6000 Hz_10ms_envelope_correlation": 0.9455114562365777, "3000–6000 Hz_10ms_modulation_mae_db": 0.6889804942362895, "3000–6000 Hz_40ms_log_rms_mae_db": 1.387898704741599, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1435621672617043, "3000–6000 Hz_40ms_envelope_correlation": 0.9803814660167048, "3000–6000 Hz_40ms_modulation_mae_db": 0.746442064873408, "mr_stft_db": 5.987883270691519, "band_mae_db": 0.21983943694599606, "rms_gain_error_db": -0.07745255624246483, "output_peak": 0.23544420854631123}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9286591718650125, "全帯域_10ms_rms_envelope_nrmse": 0.07055323365429028, "全帯域_10ms_envelope_correlation": 0.9949742437932695, "全帯域_10ms_modulation_mae_db": 0.29186568026562026, "全帯域_40ms_log_rms_mae_db": 0.4692423127742486, "全帯域_40ms_rms_envelope_nrmse": 0.0375964160452096, "全帯域_40ms_envelope_correlation": 0.9985258266296948, "全帯域_40ms_modulation_mae_db": 0.36592875461085, "300–1000 Hz_10ms_log_rms_mae_db": 1.6403513270512342, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09338948346926408, "300–1000 Hz_10ms_envelope_correlation": 0.9925646012784836, "300–1000 Hz_10ms_modulation_mae_db": 0.5627535407908945, "300–1000 Hz_40ms_log_rms_mae_db": 0.9347263685690229, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05606738294634358, "300–1000 Hz_40ms_envelope_correlation": 0.9972825521505119, "300–1000 Hz_40ms_modulation_mae_db": 0.45863059340239426, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8693078904779243, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14330886651597521, "1000–3000 Hz_10ms_envelope_correlation": 0.9832326099958912, "1000–3000 Hz_10ms_modulation_mae_db": 0.82227145225505, "1000–3000 Hz_40ms_log_rms_mae_db": 1.023350600519382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07833727882709841, "1000–3000 Hz_40ms_envelope_correlation": 0.9945388664848893, "1000–3000 Hz_40ms_modulation_mae_db": 0.7664172677005726, "3000–6000 Hz_10ms_log_rms_mae_db": 2.151404729773228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23058562883634637, "3000–6000 Hz_10ms_envelope_correlation": 0.9552219103714303, "3000–6000 Hz_10ms_modulation_mae_db": 0.8015578610980162, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3553396752655593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13710277667330717, "3000–6000 Hz_40ms_envelope_correlation": 0.9825885875359431, "3000–6000 Hz_40ms_modulation_mae_db": 0.7981029585317089, "mr_stft_db": 5.973287096715527, "band_mae_db": 0.24995128701349845, "rms_gain_error_db": -0.09889357639224819, "output_peak": 0.23999267590443962}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9494206055518762, "全帯域_10ms_rms_envelope_nrmse": 0.07514645702063058, "全帯域_10ms_envelope_correlation": 0.9943063283023905, "全帯域_10ms_modulation_mae_db": 0.3587257110932749, "全帯域_40ms_log_rms_mae_db": 0.4826238610029484, "全帯域_40ms_rms_envelope_nrmse": 0.0410030553684232, "全帯域_40ms_envelope_correlation": 0.9982581105963444, "全帯域_40ms_modulation_mae_db": 0.668098963346652, "300–1000 Hz_10ms_log_rms_mae_db": 1.6889371661762405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09397545475148897, "300–1000 Hz_10ms_envelope_correlation": 0.9925165963216586, "300–1000 Hz_10ms_modulation_mae_db": 0.591843589260827, "300–1000 Hz_40ms_log_rms_mae_db": 0.9610959421829292, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05231184614561012, "300–1000 Hz_40ms_envelope_correlation": 0.997745220291638, "300–1000 Hz_40ms_modulation_mae_db": 0.5142618568711866, "1000–3000 Hz_10ms_log_rms_mae_db": 1.86481490199985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14336890689982193, "1000–3000 Hz_10ms_envelope_correlation": 0.9832126699777434, "1000–3000 Hz_10ms_modulation_mae_db": 0.7839011129129873, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0148233334390984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07877729465382319, "1000–3000 Hz_40ms_envelope_correlation": 0.9944881801030965, "1000–3000 Hz_40ms_modulation_mae_db": 1.1003519328471418, "3000–6000 Hz_10ms_log_rms_mae_db": 2.177959317747372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23482980345543764, "3000–6000 Hz_10ms_envelope_correlation": 0.9524643601649174, "3000–6000 Hz_10ms_modulation_mae_db": 0.6986537097576635, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3739882214102037, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14098125988750373, "3000–6000 Hz_40ms_envelope_correlation": 0.9812703968718532, "3000–6000 Hz_40ms_modulation_mae_db": 0.7788814779976526, "mr_stft_db": 5.962884638234431, "band_mae_db": 0.3009064569902154, "rms_gain_error_db": -0.11077339815416867, "output_peak": 0.2375948256261635}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9691554463912426, "全帯域_10ms_rms_envelope_nrmse": 0.06970510820251478, "全帯域_10ms_envelope_correlation": 0.9951819752469924, "全帯域_10ms_modulation_mae_db": 0.3949920863110085, "全帯域_40ms_log_rms_mae_db": 0.5006073378840109, "全帯域_40ms_rms_envelope_nrmse": 0.03836551759460206, "全帯域_40ms_envelope_correlation": 0.998563860627937, "全帯域_40ms_modulation_mae_db": 0.6244154603830027, "300–1000 Hz_10ms_log_rms_mae_db": 1.6820795680904097, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08379670426341373, "300–1000 Hz_10ms_envelope_correlation": 0.9941992464271395, "300–1000 Hz_10ms_modulation_mae_db": 0.44165158770299434, "300–1000 Hz_40ms_log_rms_mae_db": 0.9581928716957661, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05168960127170599, "300–1000 Hz_40ms_envelope_correlation": 0.9978856224811061, "300–1000 Hz_40ms_modulation_mae_db": 0.6429207301527856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.863386466291447, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14989950473696942, "1000–3000 Hz_10ms_envelope_correlation": 0.9816252872209542, "1000–3000 Hz_10ms_modulation_mae_db": 1.3271238269240737, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0593196878181095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0845173711836178, "1000–3000 Hz_40ms_envelope_correlation": 0.9937128640065355, "1000–3000 Hz_40ms_modulation_mae_db": 1.5782971797759668, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1981198608913006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2583142297739018, "3000–6000 Hz_10ms_envelope_correlation": 0.9420863561760708, "3000–6000 Hz_10ms_modulation_mae_db": 0.6900869274955767, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4117262422778083, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15289984354115613, "3000–6000 Hz_40ms_envelope_correlation": 0.9776136562610863, "3000–6000 Hz_40ms_modulation_mae_db": 0.7662854654561099, "mr_stft_db": 5.982530329893656, "band_mae_db": 0.3154249820601467, "rms_gain_error_db": -0.1299801863306038, "output_peak": 0.23451829477963496}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9429137477916498, "全帯域_10ms_rms_envelope_nrmse": 0.07206806648781161, "全帯域_10ms_envelope_correlation": 0.9947895172496841, "全帯域_10ms_modulation_mae_db": 0.2295192664122534, "全帯域_40ms_log_rms_mae_db": 0.4968320347449383, "全帯域_40ms_rms_envelope_nrmse": 0.03991377305933783, "全帯域_40ms_envelope_correlation": 0.9983797542424143, "全帯域_40ms_modulation_mae_db": 0.1963634303267039, "300–1000 Hz_10ms_log_rms_mae_db": 1.7947710112129291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09712415703666627, "300–1000 Hz_10ms_envelope_correlation": 0.9918940367488297, "300–1000 Hz_10ms_modulation_mae_db": 0.6377204193545026, "300–1000 Hz_40ms_log_rms_mae_db": 1.0003002323545565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0598158956083668, "300–1000 Hz_40ms_envelope_correlation": 0.9968198767826253, "300–1000 Hz_40ms_modulation_mae_db": 0.3860416810895855, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8999189663927412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13939998652397909, "1000–3000 Hz_10ms_envelope_correlation": 0.9841500099441582, "1000–3000 Hz_10ms_modulation_mae_db": 0.42885167975805744, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0216907724567834, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0758943117112583, "1000–3000 Hz_40ms_envelope_correlation": 0.99497793949273, "1000–3000 Hz_40ms_modulation_mae_db": 0.48541147843360033, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1818807970245033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23784812024949184, "3000–6000 Hz_10ms_envelope_correlation": 0.9527839139434465, "3000–6000 Hz_10ms_modulation_mae_db": 0.5235326784102716, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3849780051423406, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14114037909596294, "3000–6000 Hz_40ms_envelope_correlation": 0.9819416700661477, "3000–6000 Hz_40ms_modulation_mae_db": 0.6304556558176501, "mr_stft_db": 5.99963317620849, "band_mae_db": 0.24385542370754706, "rms_gain_error_db": -0.11717223844473558, "output_peak": 0.23865035958859915}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.987579743598927, "全帯域_10ms_rms_envelope_nrmse": 0.08036389759377932, "全帯域_10ms_envelope_correlation": 0.9934096654134354, "全帯域_10ms_modulation_mae_db": 0.26395104032968403, "全帯域_40ms_log_rms_mae_db": 0.5108429730044628, "全帯域_40ms_rms_envelope_nrmse": 0.03913520855407385, "全帯域_40ms_envelope_correlation": 0.9983507690826338, "全帯域_40ms_modulation_mae_db": 0.46177464298014015, "300–1000 Hz_10ms_log_rms_mae_db": 1.7192485326051434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10200040674978508, "300–1000 Hz_10ms_envelope_correlation": 0.9910214469275058, "300–1000 Hz_10ms_modulation_mae_db": 0.3210286186759915, "300–1000 Hz_40ms_log_rms_mae_db": 0.9916679273163631, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06050055265623945, "300–1000 Hz_40ms_envelope_correlation": 0.9967116687573178, "300–1000 Hz_40ms_modulation_mae_db": 0.576960509088491, "1000–3000 Hz_10ms_log_rms_mae_db": 1.905277152471007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14756016074593864, "1000–3000 Hz_10ms_envelope_correlation": 0.9821967155715399, "1000–3000 Hz_10ms_modulation_mae_db": 0.7847440748187143, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0310836644928543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07412419333654491, "1000–3000 Hz_40ms_envelope_correlation": 0.9951519987789157, "1000–3000 Hz_40ms_modulation_mae_db": 1.0100118139685867, "3000–6000 Hz_10ms_log_rms_mae_db": 2.170474018424603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2453198026723636, "3000–6000 Hz_10ms_envelope_correlation": 0.9495549775183254, "3000–6000 Hz_10ms_modulation_mae_db": 0.9324677447802872, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3643286762898628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15046839615343185, "3000–6000 Hz_40ms_envelope_correlation": 0.97919485053992, "3000–6000 Hz_40ms_modulation_mae_db": 0.9458824300955092, "mr_stft_db": 5.95666916144594, "band_mae_db": 0.26383019045122846, "rms_gain_error_db": -0.08114869309067706, "output_peak": 0.23480544591869842}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9475675942086, "全帯域_10ms_rms_envelope_nrmse": 0.0799240745659641, "全帯域_10ms_envelope_correlation": 0.9934767308602074, "全帯域_10ms_modulation_mae_db": 0.5768088110235049, "全帯域_40ms_log_rms_mae_db": 0.5182773019784682, "全帯域_40ms_rms_envelope_nrmse": 0.046995198189481134, "全帯域_40ms_envelope_correlation": 0.9975685688451965, "全帯域_40ms_modulation_mae_db": 0.2705523233924153, "300–1000 Hz_10ms_log_rms_mae_db": 1.7629594816727476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09388413911847482, "300–1000 Hz_10ms_envelope_correlation": 0.99251875534157, "300–1000 Hz_10ms_modulation_mae_db": 0.3297377839067297, "300–1000 Hz_40ms_log_rms_mae_db": 1.0127400863930824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05841900047358949, "300–1000 Hz_40ms_envelope_correlation": 0.9970611255060897, "300–1000 Hz_40ms_modulation_mae_db": 0.2739407636513854, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8674949923955064, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17411501421729064, "1000–3000 Hz_10ms_envelope_correlation": 0.9753261096479932, "1000–3000 Hz_10ms_modulation_mae_db": 1.197602602138008, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0386216642573174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.119160447828694, "1000–3000 Hz_40ms_envelope_correlation": 0.987372299016563, "1000–3000 Hz_40ms_modulation_mae_db": 1.3971285298524465, "3000–6000 Hz_10ms_log_rms_mae_db": 2.134180784916038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2344936830227992, "3000–6000 Hz_10ms_envelope_correlation": 0.9524153915280617, "3000–6000 Hz_10ms_modulation_mae_db": 0.3352848314094741, "3000–6000 Hz_40ms_log_rms_mae_db": 1.309432881369257, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13529253327537252, "3000–6000 Hz_40ms_envelope_correlation": 0.9825684681575282, "3000–6000 Hz_40ms_modulation_mae_db": 0.39442923353595666, "mr_stft_db": 5.980093584453307, "band_mae_db": 0.28201200362579965, "rms_gain_error_db": -0.07159561209234613, "output_peak": 0.24126454959076798}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9435360038123635, "全帯域_10ms_rms_envelope_nrmse": 0.07424742276020393, "全帯域_10ms_envelope_correlation": 0.9945534628334399, "全帯域_10ms_modulation_mae_db": 0.5369306932470551, "全帯域_40ms_log_rms_mae_db": 0.49917316865334993, "全帯域_40ms_rms_envelope_nrmse": 0.04278661207635717, "全帯域_40ms_envelope_correlation": 0.9982122728172039, "全帯域_40ms_modulation_mae_db": 0.7456693837624945, "300–1000 Hz_10ms_log_rms_mae_db": 1.7277569301642273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09975720605282444, "300–1000 Hz_10ms_envelope_correlation": 0.9915457170410095, "300–1000 Hz_10ms_modulation_mae_db": 0.8412861158701613, "300–1000 Hz_40ms_log_rms_mae_db": 0.9673346803310183, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06004583815945739, "300–1000 Hz_40ms_envelope_correlation": 0.9969342366223104, "300–1000 Hz_40ms_modulation_mae_db": 0.5083754534898535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8389109694612944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14224263402629364, "1000–3000 Hz_10ms_envelope_correlation": 0.9835199312177445, "1000–3000 Hz_10ms_modulation_mae_db": 0.7433052160571904, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0505545512906143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07888674503160527, "1000–3000 Hz_40ms_envelope_correlation": 0.9946534257603462, "1000–3000 Hz_40ms_modulation_mae_db": 0.828747760132886, "3000–6000 Hz_10ms_log_rms_mae_db": 2.181940329029307, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2390164642613218, "3000–6000 Hz_10ms_envelope_correlation": 0.9500412799696633, "3000–6000 Hz_10ms_modulation_mae_db": 0.6786009687448229, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3669858775980057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13874115406584983, "3000–6000 Hz_40ms_envelope_correlation": 0.9817308341738181, "3000–6000 Hz_40ms_modulation_mae_db": 0.8593019507809917, "mr_stft_db": 5.984465286157882, "band_mae_db": 0.29126265936650636, "rms_gain_error_db": -0.13987445288320643, "output_peak": 0.22740811002140568}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.063464187211519, "spectral_convergence": 0.23910746025028057, "spectral_convergence_status": "ok", "amplitude_floor": 6.60007588392934e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.892786257166861, "spectral_convergence": 0.2419758367592836, "spectral_convergence_status": "ok", "amplitude_floor": 5.570424644238683e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.007399367696176, "spectral_convergence": 0.24395758458920286, "spectral_convergence_status": "ok", "amplitude_floor": 3.8555660806566813e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.162305217059986e-05, "output_power": 3.296499417061035e-05, "delta_db": 0.18049218744881249, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000304634409424632, "output_power": 0.0002973979027595144, "delta_db": -0.10441055146592335, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004181285613649292, "output_power": 0.00040964985224797667, "delta_db": -0.08897031629441901, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001153043826104485, "output_power": 0.00011489484405542229, "delta_db": -0.015452747152726522, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.6981779049220075e-05, "output_power": 5.4760520956988374e-05, "delta_db": -0.1726843329910338, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.999437610841641e-06, "output_power": 3.570610531093606e-06, "delta_db": 0.7570264863230611, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21983943694599606, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07745255624246483, "pitch": null, "output_rms_dbfs": -30.394156810885807, "output_peak": 0.23544420854631123, "reference_rms_dbfs": -30.31670425464334, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.987883270691519}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.667347842324586e-05, "log_rms_mae_db": 0.9410963031210183, "rms_envelope_nrmse": 0.07346596064974466, "envelope_correlation": 0.9944988463416258, "rms_gain_error_db": -0.07734169596763878, "reference_mean_power": 0.0009300339703354519, "output_mean_power": 0.0009136179891444433}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3130295372770493e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001632046380308234, "output_power": 0.000157433182356765, "reference_active": true, "delta_db": -0.15636222216522294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.195140055898603e-05, "output_power": 2.898264883600762e-05, "reference_active": true, "delta_db": -0.42351824997566234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.743176048402797e-06, "output_power": 3.916484165459022e-06, "reference_active": true, "delta_db": 0.1965612175084447}], "active_band_count": 3, "active_band_mae_db": 0.25881389654977666}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.141277463219576e-05, "log_rms_mae_db": 0.5120190940495198, "rms_envelope_nrmse": 0.04151607360118068, "envelope_correlation": 0.9981256048281743, "rms_gain_error_db": -0.0778550635035385, "reference_mean_power": 0.0009316560347730994, "output_mean_power": 0.0009151032443814789}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0476212509594803e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014768349458157853, "output_power": 0.00014314296177292982, "reference_active": true, "delta_db": -0.13561961394224983}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4986865819787958e-05, "output_power": 1.3723235324731544e-05, "reference_active": true, "delta_db": -0.3825430836182097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9796240655945466e-07, "output_power": 3.2593912476277505e-07, "reference_active": true, "delta_db": -0.8670555345669351}], "active_band_count": 3, "active_band_mae_db": 0.46173941070913155}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.73108999878072e-05, "log_rms_mae_db": 1.738227007541424, "rms_envelope_nrmse": 0.09330267298379508, "envelope_correlation": 0.9924725826071364, "rms_gain_error_db": -0.11896196777592213, "reference_mean_power": 0.0005508019269041907, "output_mean_power": 0.0005359191230526249}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0420364913089255e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.387221376866869e-05, "output_power": 8.712780248031708e-05, "reference_active": true, "delta_db": -0.3238029937675633}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.257079739763329e-05, "output_power": 1.8911235778401254e-05, "reference_active": true, "delta_db": -0.7682699313228893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1515320396484854e-06, "output_power": 3.0325003341215093e-06, "reference_active": true, "delta_db": 1.4905303960423646}], "active_band_count": 3, "active_band_mae_db": 0.860867773710939}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091336805501339e-05, "log_rms_mae_db": 0.9729747831115879, "rms_envelope_nrmse": 0.052949570289131956, "envelope_correlation": 0.9974718919420684, "rms_gain_error_db": -0.11992314774315437, "reference_mean_power": 0.0005517580103658142, "output_mean_power": 0.0005367305706190539}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.88495020416125e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.537091697949908e-05, "output_power": 7.947962663100235e-05, "reference_active": true, "delta_db": -0.31054127986329955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0155340206288246e-05, "output_power": 8.647076244542896e-06, "reference_active": true, "delta_db": -0.6982518862066883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.398901151759038e-07, "output_power": 2.506804453564838e-07, "reference_active": true, "delta_db": 0.19108104634655312}], "active_band_count": 3, "active_band_mae_db": 0.39995807080551365}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.045309710707616e-05, "log_rms_mae_db": 1.89129553533547, "rms_envelope_nrmse": 0.14483616226715584, "envelope_correlation": 0.9828523764293149, "rms_gain_error_db": -0.07742147994473611, "reference_mean_power": 0.00014468505517057894, "output_mean_power": 0.0001421286157341458}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.678016723697031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.684581121209281e-05, "output_power": 3.611383963236154e-05, "reference_active": true, "delta_db": -0.08714457386443028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.802139012131204e-06, "output_power": 7.406829631183015e-06, "reference_active": true, "delta_db": -0.22581328643262485}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.902802552587548e-06, "output_power": 1.3099975575833707e-06, "reference_active": true, "delta_db": -1.6212323941033404}], "active_band_count": 3, "active_band_mae_db": 0.6447300848001318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.539248843957361e-05, "log_rms_mae_db": 1.0597186698237482, "rms_envelope_nrmse": 0.08705999382806331, "envelope_correlation": 0.993273951968726, "rms_gain_error_db": -0.07841615784324475, "reference_mean_power": 0.0001448248558759811, "output_mean_power": 0.00014223336642912363}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.24555278166651e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.61104008282169e-05, "output_power": 3.531884269501875e-05, "reference_active": true, "delta_db": -0.09625844697475458}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6607268103199764e-06, "output_power": 3.6014771708117567e-06, "reference_active": true, "delta_db": -0.07086653724797196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4925257777124562e-07, "output_power": 8.775704250471982e-08, "reference_active": true, "delta_db": -2.3063986197465853}], "active_band_count": 3, "active_band_mae_db": 0.8245078679897707}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.667503736683173e-05, "log_rms_mae_db": 2.2371901747583767, "rms_envelope_nrmse": 0.2500783783294953, "envelope_correlation": 0.9455114562365777, "rms_gain_error_db": -0.02850508233192168, "reference_mean_power": 8.017666159457313e-06, "output_mean_power": 7.965214228249e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.8921445305362116e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.737362730986327e-06, "output_power": 2.4520888138001744e-06, "reference_active": true, "delta_db": -0.4779615394355679}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.579169756744011e-07, "output_power": 6.402037906909913e-07, "reference_active": true, "delta_db": -0.7330339311319225}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.809426009650544e-07, "output_power": 1.485751671741426e-07, "reference_active": true, "delta_db": -0.8559460121413782}], "active_band_count": 3, "active_band_mae_db": 0.6889804942362895}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4241724067408686e-05, "log_rms_mae_db": 1.387898704741599, "rms_envelope_nrmse": 0.1435621672617043, "envelope_correlation": 0.9803814660167048, "rms_gain_error_db": -0.02786440590780648, "reference_mean_power": 8.048582997885633e-06, "output_mean_power": 7.997108463344895e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.450576990854383e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.750016160596199e-06, "output_power": 2.456250004841049e-06, "reference_active": true, "delta_db": -0.4906267741376201}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.547469624529576e-07, "output_power": 3.107707780599307e-07, "reference_active": true, "delta_db": -0.5747851035722056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2784081824019638e-08, "output_power": 9.756141936148786e-09, "reference_active": true, "delta_db": -1.1739143169103985}], "active_band_count": 3, "active_band_mae_db": 0.746442064873408}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8697344121923888, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "911a27b24fb3bacedf90b69d6aff8a6acccd30ddfa10595b441a5ad2d2b0c6cc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9820144637506674, "standard_deviation": 0.00281240671295717, "minimum": 0.9753261096479932, "maximum": 0.9841500099441582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.875050859353155, "standard_deviation": 0.0221454232226277, "minimum": 1.8389109694612944, "maximum": 1.905277152471007, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8415662562080266, "standard_deviation": 0.289873712554727, "minimum": 0.42885167975805744, "maximum": 1.3271238269240737, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14809140449167807, "standard_deviation": 0.010995452480430525, "minimum": 0.13939998652397909, "maximum": 0.17411501421729064, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9935211907014754, "standard_deviation": 0.00256183738073801, "minimum": 0.987372299016563, "maximum": 0.9951519987789157, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0373953680122385, "standard_deviation": 0.0175128429254847, "minimum": 1.0148233334390984, "maximum": 1.0597186698237482, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9988592288376215, "standard_deviation": 0.3549217400150751, "minimum": 0.48541147843360033, "maximum": 1.5782971797759668, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08459470455008815, "standard_deviation": 0.014598341966242927, "minimum": 0.07412419333654491, "maximum": 0.119160447828694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9500099557385617, "standard_deviation": 0.0043077579821170055, "minimum": 0.9420863561760708, "maximum": 0.9552219103714303, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.179143751570591, "standard_deviation": 0.030734998873094048, "minimum": 2.134180784916038, "maximum": 2.2371901747583767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6686456519915502, "standard_deviation": 0.17781522903631344, "minimum": 0.3352848314094741, "maximum": 0.9324677447802872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24131076382514471, "standard_deviation": 0.009279924417594655, "minimum": 0.23058562883634637, "maximum": 0.2583142297739018, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9809112412028752, "standard_deviation": 0.0017534914590692937, "minimum": 0.9776136562610863, "maximum": 0.9825885875359431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3693347855118296, "standard_deviation": 0.029815382459924503, "minimum": 1.309432881369257, "maximum": 1.4117262422778083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7399726546361234, "standard_deviation": 0.16633004689146366, "minimum": 0.39442923353595666, "maximum": 0.9458824300955092, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14252356374428604, "standard_deviation": 0.006235125153326563, "minimum": 0.13529253327537252, "maximum": 0.15289984354115613, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9923416228366666, "standard_deviation": 0.0009378006693730647, "minimum": 0.9910214469275058, "maximum": 0.9941992464271395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7192913780642947, "standard_deviation": 0.048656129256560994, "minimum": 1.6403513270512342, "maximum": 1.7947710112129291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.57336117865913, "standard_deviation": 0.20681653124436467, "minimum": 0.3210286186759915, "maximum": 0.860867773710939, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09465377805321407, "standard_deviation": 0.0054572775322327205, "minimum": 0.08379670426341373, "maximum": 0.10200040674978508, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972390243167084, "standard_deviation": 0.0004323982326708414, "minimum": 0.9967116687573178, "maximum": 0.9978856224811061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9748791114942907, "standard_deviation": 0.025362360103832544, "minimum": 0.9347263685690229, "maximum": 1.0127400863930824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.47013620731889944, "standard_deviation": 0.11667851260704169, "minimum": 0.2739407636513854, "maximum": 0.6429207301527856, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056474960943805594, "standard_deviation": 0.0037172066693206426, "minimum": 0.05168960127170599, "maximum": 0.06050055265623945, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27088530502011726, "standard_deviation": 0.03217895335054455, "minimum": 0.21983943694599606, "maximum": 0.3154249820601467, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.978430817975094, "standard_deviation": 0.013818133641870059, "minimum": 5.95666916144594, "maximum": 5.99963317620849, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.23620980874700256, "standard_deviation": 0.00431979741048151, "minimum": 0.22740811002140568, "maximum": 0.24126454959076798, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10336133920380633, "standard_deviation": 0.025301437152847578, "minimum": -0.13987445288320643, "maximum": -0.07159561209234613, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943988462551306, "standard_deviation": 0.0006511789950404934, "minimum": 0.9934096654134354, "maximum": 0.9951819752469924, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9512410770425863, "standard_deviation": 0.018515643402031427, "minimum": 0.9286591718650125, "maximum": 0.987579743598927, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36395089815402226, "standard_deviation": 0.1312267728811405, "minimum": 0.2295192664122534, "maximum": 0.5768088110235049, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07443427761686741, "standard_deviation": 0.0039606008977480294, "minimum": 0.06970510820251478, "maximum": 0.08036389759377932, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982480959586999, "standard_deviation": 0.00031233271899328057, "minimum": 0.9975685688451965, "maximum": 0.998563860627937, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49870226051149336, "standard_deviation": 0.01622022315276757, "minimum": 0.4692423127742486, "maximum": 0.5182773019784682, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4743177961889238, "standard_deviation": 0.1944056369824964, "minimum": 0.1963634303267039, "maximum": 0.7456693837624945, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04091398181108319, "standard_deviation": 0.0029886391956908216, "minimum": 0.0375964160452096, "maximum": 0.046995198189481134, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-126133-0003", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.3724162578582764, "predicted_mos": 2.3724162578582764, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1b22c95459685f684ca95741387ad4fa5a1dc723691f40c4fd18c549ded1ce64", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.51739800022915, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9493383828416113, "全帯域_10ms_rms_envelope_nrmse": 0.07254674005082493, "全帯域_10ms_envelope_correlation": 0.9946416981885671, "全帯域_10ms_modulation_mae_db": 0.3199400141225349, "全帯域_40ms_log_rms_mae_db": 0.48429265218251266, "全帯域_40ms_rms_envelope_nrmse": 0.040416914734179715, "全帯域_40ms_envelope_correlation": 0.998220880147701, "全帯域_40ms_modulation_mae_db": 0.40868209239311276, "300–1000 Hz_10ms_log_rms_mae_db": 1.4614433947070857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08765346308085604, "300–1000 Hz_10ms_envelope_correlation": 0.9933452573848327, "300–1000 Hz_10ms_modulation_mae_db": 0.6625614912645719, "300–1000 Hz_40ms_log_rms_mae_db": 0.8526949567173222, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049101716340515925, "300–1000 Hz_40ms_envelope_correlation": 0.9977948592383455, "300–1000 Hz_40ms_modulation_mae_db": 0.4193156758075623, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4723130721048021, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13909050660898906, "1000–3000 Hz_10ms_envelope_correlation": 0.9842534313322593, "1000–3000 Hz_10ms_modulation_mae_db": 0.7889115586959438, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8072768479350911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.083206884366123, "1000–3000 Hz_40ms_envelope_correlation": 0.9940268694083205, "1000–3000 Hz_40ms_modulation_mae_db": 1.1206529274553927, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5544162798840306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20614192614277493, "3000–6000 Hz_10ms_envelope_correlation": 0.9630335686925237, "3000–6000 Hz_10ms_modulation_mae_db": 0.19361564425896138, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9337003140691444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11857364885676219, "3000–6000 Hz_40ms_envelope_correlation": 0.9864551332797727, "3000–6000 Hz_40ms_modulation_mae_db": 0.32140644652762435, "mr_stft_db": 5.5053501386458015, "band_mae_db": 0.1972180219379349, "rms_gain_error_db": -0.06683968051686905, "output_peak": 0.22956216055079284}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9560942565177178, "全帯域_10ms_rms_envelope_nrmse": 0.07385875143200142, "全帯域_10ms_envelope_correlation": 0.9944039803086709, "全帯域_10ms_modulation_mae_db": 0.40989042408944876, "全帯域_40ms_log_rms_mae_db": 0.5060500079433756, "全帯域_40ms_rms_envelope_nrmse": 0.04481290824203561, "全帯域_40ms_envelope_correlation": 0.9977438717987519, "全帯域_40ms_modulation_mae_db": 0.4352380045876781, "300–1000 Hz_10ms_log_rms_mae_db": 1.4086894625704462, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09272177951821564, "300–1000 Hz_10ms_envelope_correlation": 0.9925210059715294, "300–1000 Hz_10ms_modulation_mae_db": 0.5900121795831968, "300–1000 Hz_40ms_log_rms_mae_db": 0.895093832844614, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060836916817440634, "300–1000 Hz_40ms_envelope_correlation": 0.9965365416333741, "300–1000 Hz_40ms_modulation_mae_db": 0.3394650875696931, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4656736083204513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12756509059291798, "1000–3000 Hz_10ms_envelope_correlation": 0.9867714268040924, "1000–3000 Hz_10ms_modulation_mae_db": 0.48954525095526463, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7883646938528176, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06694083504424587, "1000–3000 Hz_40ms_envelope_correlation": 0.9960251826075601, "1000–3000 Hz_40ms_modulation_mae_db": 0.4809260344416275, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5559542523351295, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1811348496640806, "3000–6000 Hz_10ms_envelope_correlation": 0.9720712776032877, "3000–6000 Hz_10ms_modulation_mae_db": 0.7185413408615418, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9650801987617779, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11422757814934562, "3000–6000 Hz_40ms_envelope_correlation": 0.9888127859569797, "3000–6000 Hz_40ms_modulation_mae_db": 0.8989631278484538, "mr_stft_db": 5.503607485664809, "band_mae_db": 0.1911432520468849, "rms_gain_error_db": -0.04129665465176593, "output_peak": 0.24491684966258495}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9288589033533748, "全帯域_10ms_rms_envelope_nrmse": 0.0768809997214458, "全帯域_10ms_envelope_correlation": 0.9939854095963964, "全帯域_10ms_modulation_mae_db": 0.1776819079550895, "全帯域_40ms_log_rms_mae_db": 0.4879937181022463, "全帯域_40ms_rms_envelope_nrmse": 0.04187319628394836, "全帯域_40ms_envelope_correlation": 0.998112885670331, "全帯域_40ms_modulation_mae_db": 0.24007042504518106, "300–1000 Hz_10ms_log_rms_mae_db": 1.4217892075872085, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09294320460660856, "300–1000 Hz_10ms_envelope_correlation": 0.9925159593449498, "300–1000 Hz_10ms_modulation_mae_db": 0.33672073253044243, "300–1000 Hz_40ms_log_rms_mae_db": 0.8767628023525916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05818580038385284, "300–1000 Hz_40ms_envelope_correlation": 0.9968817787887034, "300–1000 Hz_40ms_modulation_mae_db": 0.2860840162032378, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4757295466388145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13211576080175105, "1000–3000 Hz_10ms_envelope_correlation": 0.9858881504765283, "1000–3000 Hz_10ms_modulation_mae_db": 0.16205943396261002, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8356408335443132, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07169156769739511, "1000–3000 Hz_40ms_envelope_correlation": 0.995738984760104, "1000–3000 Hz_40ms_modulation_mae_db": 0.10047627045976165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5221095815507084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18540626988442047, "3000–6000 Hz_10ms_envelope_correlation": 0.970223128297086, "3000–6000 Hz_10ms_modulation_mae_db": 0.7717202422282581, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9286857614342579, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11173512900732595, "3000–6000 Hz_40ms_envelope_correlation": 0.9878747488629686, "3000–6000 Hz_40ms_modulation_mae_db": 0.7733904411705296, "mr_stft_db": 5.503983142241687, "band_mae_db": 0.2293784785343532, "rms_gain_error_db": -0.08958936588957146, "output_peak": 0.23165517026024554}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9533679274318824, "全帯域_10ms_rms_envelope_nrmse": 0.07469446376810832, "全帯域_10ms_envelope_correlation": 0.9945384341863407, "全帯域_10ms_modulation_mae_db": 0.4567535916428252, "全帯域_40ms_log_rms_mae_db": 0.5026020195466064, "全帯域_40ms_rms_envelope_nrmse": 0.042865013703800685, "全帯域_40ms_envelope_correlation": 0.9982823319541234, "全帯域_40ms_modulation_mae_db": 0.7526707665205702, "300–1000 Hz_10ms_log_rms_mae_db": 1.385558574277589, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09935832217249556, "300–1000 Hz_10ms_envelope_correlation": 0.9916938257052338, "300–1000 Hz_10ms_modulation_mae_db": 0.563663989744103, "300–1000 Hz_40ms_log_rms_mae_db": 0.85787547702725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059824307719778244, "300–1000 Hz_40ms_envelope_correlation": 0.9970598744405692, "300–1000 Hz_40ms_modulation_mae_db": 0.9093892744796319, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4465045921304487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12761993487784742, "1000–3000 Hz_10ms_envelope_correlation": 0.9867952411749685, "1000–3000 Hz_10ms_modulation_mae_db": 0.7698987422055191, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7972867367622856, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0676437126431176, "1000–3000 Hz_40ms_envelope_correlation": 0.9961272094160999, "1000–3000 Hz_40ms_modulation_mae_db": 0.800970810728273, "3000–6000 Hz_10ms_log_rms_mae_db": 1.461579840974228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18481798129614757, "3000–6000 Hz_10ms_envelope_correlation": 0.9702234534780747, "3000–6000 Hz_10ms_modulation_mae_db": 0.3374914996837088, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9096715796372394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11389754797109336, "3000–6000 Hz_40ms_envelope_correlation": 0.9875706715965461, "3000–6000 Hz_40ms_modulation_mae_db": 0.37552340806435863, "mr_stft_db": 5.510812722990753, "band_mae_db": 0.2424181115136083, "rms_gain_error_db": -0.16307330902669626, "output_peak": 0.23585592062769092}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9606928192014629, "全帯域_10ms_rms_envelope_nrmse": 0.08045454280964597, "全帯域_10ms_envelope_correlation": 0.9933696325927862, "全帯域_10ms_modulation_mae_db": 0.3073715268460604, "全帯域_40ms_log_rms_mae_db": 0.5294909341634505, "全帯域_40ms_rms_envelope_nrmse": 0.047507399834486756, "全帯域_40ms_envelope_correlation": 0.9974891239976045, "全帯域_40ms_modulation_mae_db": 0.7176539748785314, "300–1000 Hz_10ms_log_rms_mae_db": 1.4401497974255908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10050943931886577, "300–1000 Hz_10ms_envelope_correlation": 0.9911708758205503, "300–1000 Hz_10ms_modulation_mae_db": 0.5284013129086577, "300–1000 Hz_40ms_log_rms_mae_db": 0.9437335306188918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06441956818500973, "300–1000 Hz_40ms_envelope_correlation": 0.9960473494282495, "300–1000 Hz_40ms_modulation_mae_db": 0.6460692416671058, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4524807256049925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12724063414623904, "1000–3000 Hz_10ms_envelope_correlation": 0.9869212586839595, "1000–3000 Hz_10ms_modulation_mae_db": 0.4448634954038339, "1000–3000 Hz_40ms_log_rms_mae_db": 0.801839302299976, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07514870768384836, "1000–3000 Hz_40ms_envelope_correlation": 0.995242007712247, "1000–3000 Hz_40ms_modulation_mae_db": 0.6468216163882681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5344987588814714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1803126121746468, "3000–6000 Hz_10ms_envelope_correlation": 0.9717084644065374, "3000–6000 Hz_10ms_modulation_mae_db": 0.3096097677070915, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9728053264212196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11506456408012672, "3000–6000 Hz_40ms_envelope_correlation": 0.987356636975375, "3000–6000 Hz_40ms_modulation_mae_db": 0.2330386902017688, "mr_stft_db": 5.514719430578265, "band_mae_db": 0.21002857363320568, "rms_gain_error_db": -0.06522746255691757, "output_peak": 0.22990508633053053}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9598153720852443, "全帯域_10ms_rms_envelope_nrmse": 0.07193003484115038, "全帯域_10ms_envelope_correlation": 0.9947002325419398, "全帯域_10ms_modulation_mae_db": 0.314320649250276, "全帯域_40ms_log_rms_mae_db": 0.5023141472407281, "全帯域_40ms_rms_envelope_nrmse": 0.038440557349179824, "全帯域_40ms_envelope_correlation": 0.998356530991759, "全帯域_40ms_modulation_mae_db": 0.3975199140294607, "300–1000 Hz_10ms_log_rms_mae_db": 1.4442679763919626, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09272394443552771, "300–1000 Hz_10ms_envelope_correlation": 0.9925491895421688, "300–1000 Hz_10ms_modulation_mae_db": 0.4399203495040373, "300–1000 Hz_40ms_log_rms_mae_db": 0.8770826688864003, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05601215804903207, "300–1000 Hz_40ms_envelope_correlation": 0.9971189994203186, "300–1000 Hz_40ms_modulation_mae_db": 0.5362898191772943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4493247600938346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13823214903222733, "1000–3000 Hz_10ms_envelope_correlation": 0.9844044908899519, "1000–3000 Hz_10ms_modulation_mae_db": 0.795685094813749, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8165790473841728, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07606729332143727, "1000–3000 Hz_40ms_envelope_correlation": 0.9948929797292404, "1000–3000 Hz_40ms_modulation_mae_db": 0.6405211447267153, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5160113544732348, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20076148277339256, "3000–6000 Hz_10ms_envelope_correlation": 0.9651740036009292, "3000–6000 Hz_10ms_modulation_mae_db": 0.3516479939319231, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9028038690111654, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11645162414395983, "3000–6000 Hz_40ms_envelope_correlation": 0.9868409158330466, "3000–6000 Hz_40ms_modulation_mae_db": 0.42366935981256565, "mr_stft_db": 5.498747305099688, "band_mae_db": 0.2579563334769636, "rms_gain_error_db": -0.051222342331391855, "output_peak": 0.22920018699847672}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9419256298529045, "全帯域_10ms_rms_envelope_nrmse": 0.06901208544256802, "全帯域_10ms_envelope_correlation": 0.9951959703268015, "全帯域_10ms_modulation_mae_db": 0.32746301871448796, "全帯域_40ms_log_rms_mae_db": 0.46780807658076234, "全帯域_40ms_rms_envelope_nrmse": 0.03556329753401972, "全帯域_40ms_envelope_correlation": 0.998697081340644, "全帯域_40ms_modulation_mae_db": 0.4370897641591278, "300–1000 Hz_10ms_log_rms_mae_db": 1.4320957169622208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0898636060661417, "300–1000 Hz_10ms_envelope_correlation": 0.9930619486063856, "300–1000 Hz_10ms_modulation_mae_db": 0.5729853549979144, "300–1000 Hz_40ms_log_rms_mae_db": 0.847153062783749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048446516203850785, "300–1000 Hz_40ms_envelope_correlation": 0.9979598549305709, "300–1000 Hz_40ms_modulation_mae_db": 0.3561867369586354, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4279848130381032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14806774885226714, "1000–3000 Hz_10ms_envelope_correlation": 0.982137927441788, "1000–3000 Hz_10ms_modulation_mae_db": 0.0763677111539507, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8271448823427323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08931112332641006, "1000–3000 Hz_40ms_envelope_correlation": 0.9929600501200284, "1000–3000 Hz_40ms_modulation_mae_db": 0.35595227753403963, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4775276586723118, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19261543530319747, "3000–6000 Hz_10ms_envelope_correlation": 0.9685715567899121, "3000–6000 Hz_10ms_modulation_mae_db": 0.445637316988497, "3000–6000 Hz_40ms_log_rms_mae_db": 0.887651086754371, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11653958959924819, "3000–6000 Hz_40ms_envelope_correlation": 0.9871778874451219, "3000–6000 Hz_40ms_modulation_mae_db": 0.5760426483982515, "mr_stft_db": 5.505164042777275, "band_mae_db": 0.22021814633823997, "rms_gain_error_db": -0.09664989498717688, "output_peak": 0.23216988516668152}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9386861208468801, "全帯域_10ms_rms_envelope_nrmse": 0.07508064340161265, "全帯域_10ms_envelope_correlation": 0.9942745421150362, "全帯域_10ms_modulation_mae_db": 0.4133432857515547, "全帯域_40ms_log_rms_mae_db": 0.482149981123716, "全帯域_40ms_rms_envelope_nrmse": 0.04021473643518253, "全帯域_40ms_envelope_correlation": 0.9982727539463994, "全帯域_40ms_modulation_mae_db": 0.6798340565374573, "300–1000 Hz_10ms_log_rms_mae_db": 1.4450256298826338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09210533427993708, "300–1000 Hz_10ms_envelope_correlation": 0.9926202620491521, "300–1000 Hz_10ms_modulation_mae_db": 0.24546572267566, "300–1000 Hz_40ms_log_rms_mae_db": 0.8411849856044685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05544343956393934, "300–1000 Hz_40ms_envelope_correlation": 0.997124802680582, "300–1000 Hz_40ms_modulation_mae_db": 0.6827135509492446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4901937780132375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1304489650224922, "1000–3000 Hz_10ms_envelope_correlation": 0.9862348689861117, "1000–3000 Hz_10ms_modulation_mae_db": 0.4204978563930655, "1000–3000 Hz_40ms_log_rms_mae_db": 0.853246322272864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07775117672435589, "1000–3000 Hz_40ms_envelope_correlation": 0.994853993462987, "1000–3000 Hz_40ms_modulation_mae_db": 0.7029362434366843, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5073770556260384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1866034556784781, "3000–6000 Hz_10ms_envelope_correlation": 0.9699335934014215, "3000–6000 Hz_10ms_modulation_mae_db": 0.9430675809848553, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8930910230205501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1107431643246673, "3000–6000 Hz_40ms_envelope_correlation": 0.9891560047971222, "3000–6000 Hz_40ms_modulation_mae_db": 1.1491881026787016, "mr_stft_db": 5.5211038311553615, "band_mae_db": 0.2400587935401792, "rms_gain_error_db": -0.08977928890529126, "output_peak": 0.23107422880214745}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.447713908779389, "spectral_convergence": 0.23478593801326045, "spectral_convergence_status": "ok", "amplitude_floor": 6.60007588392934e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.324110426404969, "spectral_convergence": 0.23734046770811695, "spectral_convergence_status": "ok", "amplitude_floor": 5.570424644238683e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.744226080753045, "spectral_convergence": 0.24589225510570747, "spectral_convergence_status": "ok", "amplitude_floor": 3.8555660806566813e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.162305217059986e-05, "output_power": 3.480610263907848e-05, "delta_db": 0.4165161200011688, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000304634409424632, "output_power": 0.0003018210860186595, "delta_db": -0.040293792529096414, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004181285613649292, "output_power": 0.00040917148881984815, "delta_db": -0.09404469846781936, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001153043826104485, "output_power": 0.00011470881141474133, "delta_db": -0.02249035000062677, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.6981779049220075e-05, "output_power": 5.182622844421323e-05, "delta_db": -0.41186399354265923, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.999437610841641e-06, "output_power": 3.139422385233318e-06, "delta_db": 0.19809917708623875, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1972180219379349, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06683968051686905, "pitch": null, "output_rms_dbfs": -30.383543935160212, "output_peak": 0.22956216055079284, "reference_rms_dbfs": -30.31670425464334, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5053501386458015}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.667347842324586e-05, "log_rms_mae_db": 0.9493383828416113, "rms_envelope_nrmse": 0.07254674005082493, "envelope_correlation": 0.9946416981885671, "rms_gain_error_db": -0.06653429442386617, "reference_mean_power": 0.0009300339703354519, "output_mean_power": 0.0009158943551996218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3130295372770493e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001632046380308234, "output_power": 0.00015905478396588898, "reference_active": true, "delta_db": -0.11185760470303682}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.195140055898603e-05, "output_power": 2.875572202106234e-05, "reference_active": true, "delta_db": -0.457656231188601}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.743176048402797e-06, "output_power": 3.421445547842336e-06, "reference_active": true, "delta_db": -0.3903062064759669}], "active_band_count": 3, "active_band_mae_db": 0.3199400141225349}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.141277463219576e-05, "log_rms_mae_db": 0.48429265218251266, "rms_envelope_nrmse": 0.040416914734179715, "envelope_correlation": 0.998220880147701, "rms_gain_error_db": -0.0667755089503741, "reference_mean_power": 0.0009316560347730994, "output_mean_power": 0.000917440801218858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0476212509594803e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014768349458157853, "output_power": 0.00014385499586777524, "reference_active": true, "delta_db": -0.11407011475222231}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4986865819787958e-05, "output_power": 1.3339057932002823e-05, "reference_active": true, "delta_db": -0.5058566029392886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9796240655945466e-07, "output_power": 3.461227077000881e-07, "reference_active": true, "delta_db": -0.6061195594878276}], "active_band_count": 3, "active_band_mae_db": 0.40868209239311276}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.73108999878072e-05, "log_rms_mae_db": 1.4614433947070857, "rms_envelope_nrmse": 0.08765346308085604, "envelope_correlation": 0.9933452573848327, "rms_gain_error_db": -0.09005448056668917, "reference_mean_power": 0.0005508019269041907, "output_mean_power": 0.0005394982035072638}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0420364913089255e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.387221376866869e-05, "output_power": 8.946687103718219e-05, "reference_active": true, "delta_db": -0.20874811018199893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.257079739763329e-05, "output_power": 1.8893356614583157e-05, "reference_active": true, "delta_db": -0.7723778040836611}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1515320396484854e-06, "output_power": 2.7127119075048456e-06, "reference_active": true, "delta_db": 1.0065585595280555}], "active_band_count": 3, "active_band_mae_db": 0.6625614912645719}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091336805501339e-05, "log_rms_mae_db": 0.8526949567173222, "rms_envelope_nrmse": 0.049101716340515925, "envelope_correlation": 0.9977948592383455, "rms_gain_error_db": -0.09029642273412895, "reference_mean_power": 0.0005517580103658142, "output_mean_power": 0.0005404045595680864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.88495020416125e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.537091697949908e-05, "output_power": 8.113939852006064e-05, "reference_active": true, "delta_db": -0.220781622674502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0155340206288246e-05, "output_power": 8.455252843816377e-06, "reference_active": true, "delta_db": -0.7956787778785475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.398901151759038e-07, "output_power": 2.5360687093291953e-07, "reference_active": true, "delta_db": 0.2414866268696374}], "active_band_count": 3, "active_band_mae_db": 0.4193156758075623}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.045309710707616e-05, "log_rms_mae_db": 1.4723130721048021, "rms_envelope_nrmse": 0.13909050660898906, "envelope_correlation": 0.9842534313322593, "rms_gain_error_db": -0.15724174151142384, "reference_mean_power": 0.00014468505517057894, "output_mean_power": 0.00013954025107698498}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.678016723697031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.684581121209281e-05, "output_power": 3.514816058212196e-05, "reference_active": true, "delta_db": -0.2048552066094178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.802139012131204e-06, "output_power": 7.703480360153071e-06, "reference_active": true, "delta_db": -0.05526704267367361}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.902802552587548e-06, "output_power": 1.1714735373227701e-06, "reference_active": true, "delta_db": -2.1066124268047397}], "active_band_count": 3, "active_band_mae_db": 0.7889115586959438}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.539248843957361e-05, "log_rms_mae_db": 0.8072768479350911, "rms_envelope_nrmse": 0.083206884366123, "envelope_correlation": 0.9940268694083205, "rms_gain_error_db": -0.1587102959067768, "reference_mean_power": 0.0001448248558759811, "output_mean_power": 0.00013962785791382898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.24555278166651e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.61104008282169e-05, "output_power": 3.4282981956475206e-05, "reference_active": true, "delta_db": -0.22553719039848433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6607268103199764e-06, "output_power": 3.6653218121695597e-06, "reference_active": true, "delta_db": 0.005447914181648867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4925257777124562e-07, "output_power": 7.258125492885306e-08, "reference_active": true, "delta_db": -3.130973677786045}], "active_band_count": 3, "active_band_mae_db": 1.1206529274553927}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.667503736683173e-05, "log_rms_mae_db": 1.5544162798840306, "rms_envelope_nrmse": 0.20614192614277493, "envelope_correlation": 0.9630335686925237, "rms_gain_error_db": -0.15184648906723058, "reference_mean_power": 8.017666159457313e-06, "output_mean_power": 7.742180999214526e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.8921445305362116e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.737362730986327e-06, "output_power": 2.635571295163701e-06, "reference_active": true, "delta_db": -0.16457581205821137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.579169756744011e-07, "output_power": 8.272940730451009e-07, "reference_active": true, "delta_db": 0.38038278520647867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.809426009650544e-07, "output_power": 1.8244403233797266e-07, "reference_active": true, "delta_db": 0.0358883355121942}], "active_band_count": 3, "active_band_mae_db": 0.19361564425896138}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4241724067408686e-05, "log_rms_mae_db": 0.9337003140691444, "rms_envelope_nrmse": 0.11857364885676219, "envelope_correlation": 0.9864551332797727, "rms_gain_error_db": -0.1514947490604259, "reference_mean_power": 8.048582997885633e-06, "output_mean_power": 7.772665033633681e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.450576990854383e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.750016160596199e-06, "output_power": 2.7007339953534855e-06, "reference_active": true, "delta_db": -0.07853434863283545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.547469624529576e-07, "output_power": 4.102232972047336e-07, "reference_active": true, "delta_db": 0.6310163550975855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2784081824019638e-08, "output_power": 1.2055984420339108e-08, "reference_active": true, "delta_db": -0.25466863585245214}], "active_band_count": 3, "active_band_mae_db": 0.32140644652762435}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8959026498274456, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "38f2b7c1354426b7c62ace99c7295e677d07bc607e7bf0ef7490e498220db7af", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9854258494737074, "standard_deviation": 0.0016910885936615232, "minimum": 0.982137927441788, "maximum": 0.9869212586839595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4600256119930854, "standard_deviation": 0.01971394597257083, "minimum": 1.4279848130381032, "maximum": 1.4901937780132375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.49347864294799204, "standard_deviation": 0.2794235748787753, "minimum": 0.0763677111539507, "maximum": 0.795685094813749, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1337975987418414, "standard_deviation": 0.007419671672906575, "minimum": 0.12724063414623904, "maximum": 0.14806774885226714, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949834096520734, "standard_deviation": 0.001073393460436107, "minimum": 0.9929600501200284, "maximum": 0.9961272094160999, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8159223332992815, "standard_deviation": 0.02167309528522206, "minimum": 0.7883646938528176, "maximum": 0.853246322272864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6061571656463453, "standard_deviation": 0.30445675408474104, "minimum": 0.10047627045976165, "maximum": 1.1206529274553927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07597016260086664, "standard_deviation": 0.007591089778959183, "minimum": 0.06694083504424587, "maximum": 0.08931112332641006, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9688673807837216, "standard_deviation": 0.003182820293967876, "minimum": 0.9630335686925237, "maximum": 0.9720712776032877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5161843477996442, "standard_deviation": 0.03373730891122941, "minimum": 1.461579840974228, "maximum": 1.5559542523351295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5089164233306046, "standard_deviation": 0.2669289669719299, "minimum": 0.19361564425896138, "maximum": 0.9430675809848553, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18972425161464232, "standard_deviation": 0.009369647473449214, "minimum": 0.1803126121746468, "maximum": 0.20614192614277493, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9876555980933666, "standard_deviation": 0.0009310479322878067, "minimum": 0.9864551332797727, "maximum": 0.9891560047971222, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9241861448887158, "standard_deviation": 0.0318848504607557, "minimum": 0.887651086754371, "maximum": 0.9728053264212196, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5939027780877817, "standard_deviation": 0.31961685113867616, "minimum": 0.2330386902017688, "maximum": 1.1491881026787016, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11465410576656615, "standard_deviation": 0.0025880938780869127, "minimum": 0.1107431643246673, "maximum": 0.11857364885676219, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9924347905531004, "standard_deviation": 0.0007000300716100968, "minimum": 0.9911708758205503, "maximum": 0.9933452573848327, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.429877469975592, "standard_deviation": 0.02396997681842623, "minimum": 1.385558574277589, "maximum": 1.4614433947070857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4924663916510729, "standard_deviation": 0.1410766615348952, "minimum": 0.24546572267566, "maximum": 0.6625614912645719, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09348488668483101, "standard_deviation": 0.004380455107970333, "minimum": 0.08765346308085604, "maximum": 0.10050943931886577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970655075700892, "standard_deviation": 0.0006197265879419684, "minimum": 0.9960473494282495, "maximum": 0.9979598549305709, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.873947664604411, "standard_deviation": 0.033458540244508624, "minimum": 0.8411849856044685, "maximum": 0.9437335306188918, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5219391753515507, "standard_deviation": 0.21331067075493934, "minimum": 0.2860840162032378, "maximum": 0.9093892744796319, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05653380290792745, "standard_deviation": 0.005559544218997484, "minimum": 0.048446516203850785, "maximum": 0.06441956818500973, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2235524638776712, "standard_deviation": 0.02324961872008089, "minimum": 0.1911432520468849, "maximum": 0.2579563334769636, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.507936012394205, "standard_deviation": 0.0071797921127558706, "minimum": 5.498747305099688, "maximum": 5.5211038311553615, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.23304243604989383, "standard_deviation": 0.005239891545825816, "minimum": 0.22920018699847672, "maximum": 0.24491684966258495, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08295974985821003, "standard_deviation": 0.03782181446324037, "minimum": -0.16307330902669626, "maximum": -0.04129665465176593, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943887374820672, "standard_deviation": 0.0005411064517101602, "minimum": 0.9933696325927862, "maximum": 0.9951959703268015, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9485974265163848, "standard_deviation": 0.011237730641342272, "minimum": 0.9288589033533748, "maximum": 0.9606928192014629, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3408455522965347, "standard_deviation": 0.08661934680773402, "minimum": 0.1776819079550895, "maximum": 0.4567535916428252, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07430728268341968, "standard_deviation": 0.003425783209967815, "minimum": 0.06901208544256802, "maximum": 0.08045454280964597, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981469324809144, "standard_deviation": 0.00037465350069594414, "minimum": 0.9974891239976045, "maximum": 0.998697081340644, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4953376921104248, "standard_deviation": 0.01886175778070014, "minimum": 0.46780807658076234, "maximum": 0.5294909341634505, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.50859487476889, "standard_deviation": 0.18420540087345538, "minimum": 0.24007042504518106, "maximum": 0.7526707665205702, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04146175301460415, "standard_deviation": 0.0037126165328827208, "minimum": 0.03556329753401972, "maximum": 0.047507399834486756, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-126133-0003", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.3816978931427, "predicted_mos": 2.3816978931427, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0fac2ada9abcc6d5ab12f12fb843772ca0afb06c9cc16e36a3205a1d9a9b5bb8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.892946099862456, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.967677397197783, "全帯域_10ms_rms_envelope_nrmse": 0.07830117035848565, "全帯域_10ms_envelope_correlation": 0.9938451734642316, "全帯域_10ms_modulation_mae_db": 0.388417667607529, "全帯域_40ms_log_rms_mae_db": 0.5017316372106703, "全帯域_40ms_rms_envelope_nrmse": 0.04457497609820027, "全帯域_40ms_envelope_correlation": 0.9979566530741625, "全帯域_40ms_modulation_mae_db": 0.5583883231859774, "300–1000 Hz_10ms_log_rms_mae_db": 1.5333452373714755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10307757547548868, "300–1000 Hz_10ms_envelope_correlation": 0.9910003536117239, "300–1000 Hz_10ms_modulation_mae_db": 0.7633168270552603, "300–1000 Hz_40ms_log_rms_mae_db": 0.8928348728743134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06181447985500663, "300–1000 Hz_40ms_envelope_correlation": 0.9968028239298015, "300–1000 Hz_40ms_modulation_mae_db": 0.46476427466517906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5817649766136128, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13400724658291727, "1000–3000 Hz_10ms_envelope_correlation": 0.9853438091366308, "1000–3000 Hz_10ms_modulation_mae_db": 0.8788613039418776, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8691188439145279, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.077026638278054, "1000–3000 Hz_40ms_envelope_correlation": 0.9947340243202796, "1000–3000 Hz_40ms_modulation_mae_db": 1.0026342395821446, "3000–6000 Hz_10ms_log_rms_mae_db": 1.680542470829995, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21167518710893046, "3000–6000 Hz_10ms_envelope_correlation": 0.9608902273995144, "3000–6000 Hz_10ms_modulation_mae_db": 0.24977641425361274, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0109432420748428, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11553932094263854, "3000–6000 Hz_40ms_envelope_correlation": 0.9870512901041556, "3000–6000 Hz_40ms_modulation_mae_db": 0.40246547775184444, "mr_stft_db": 5.6443876924439875, "band_mae_db": 0.20882951124235763, "rms_gain_error_db": -0.12350431627678517, "output_peak": 0.23617001229344264}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9659007413048588, "全帯域_10ms_rms_envelope_nrmse": 0.07613073280094818, "全帯域_10ms_envelope_correlation": 0.9943275661805222, "全帯域_10ms_modulation_mae_db": 0.3322005058827969, "全帯域_40ms_log_rms_mae_db": 0.5264137470938711, "全帯域_40ms_rms_envelope_nrmse": 0.04248171785994787, "全帯域_40ms_envelope_correlation": 0.9983415409356016, "全帯域_40ms_modulation_mae_db": 0.5407973637025684, "300–1000 Hz_10ms_log_rms_mae_db": 1.5524281996339657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10234445996453694, "300–1000 Hz_10ms_envelope_correlation": 0.9913367572861262, "300–1000 Hz_10ms_modulation_mae_db": 0.3634449471893045, "300–1000 Hz_40ms_log_rms_mae_db": 0.9785473153261384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06552058580355852, "300–1000 Hz_40ms_envelope_correlation": 0.9965961803559262, "300–1000 Hz_40ms_modulation_mae_db": 0.5977641216560797, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6090546303440052, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14577784196177404, "1000–3000 Hz_10ms_envelope_correlation": 0.9826832527354147, "1000–3000 Hz_10ms_modulation_mae_db": 0.6698296613041141, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9005311226151024, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07809158577209979, "1000–3000 Hz_40ms_envelope_correlation": 0.9945963090691331, "1000–3000 Hz_40ms_modulation_mae_db": 0.6182685200132682, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7257964667258494, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19733248887007232, "3000–6000 Hz_10ms_envelope_correlation": 0.9659900589462216, "3000–6000 Hz_10ms_modulation_mae_db": 1.1083057164683439, "3000–6000 Hz_40ms_log_rms_mae_db": 1.027936778887107, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11860306990535482, "3000–6000 Hz_40ms_envelope_correlation": 0.9864948065486112, "3000–6000 Hz_40ms_modulation_mae_db": 1.1268617389693827, "mr_stft_db": 5.634378287820255, "band_mae_db": 0.26235908084006293, "rms_gain_error_db": -0.16825534295256378, "output_peak": 0.24133536940672973}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9449566955547173, "全帯域_10ms_rms_envelope_nrmse": 0.06877846908551812, "全帯域_10ms_envelope_correlation": 0.9952587648139305, "全帯域_10ms_modulation_mae_db": 0.34280171768340084, "全帯域_40ms_log_rms_mae_db": 0.4662436405586273, "全帯域_40ms_rms_envelope_nrmse": 0.03345841086921372, "全帯域_40ms_envelope_correlation": 0.9989007079950856, "全帯域_40ms_modulation_mae_db": 0.29551787528253654, "300–1000 Hz_10ms_log_rms_mae_db": 1.4733581869444483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0865159017181841, "300–1000 Hz_10ms_envelope_correlation": 0.9935733019909975, "300–1000 Hz_10ms_modulation_mae_db": 0.8489435007767633, "300–1000 Hz_40ms_log_rms_mae_db": 0.7964316386734314, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044703234879848085, "300–1000 Hz_40ms_envelope_correlation": 0.9982763946093678, "300–1000 Hz_40ms_modulation_mae_db": 0.5303393080753058, "1000–3000 Hz_10ms_log_rms_mae_db": 1.612062654805289, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1333873519300814, "1000–3000 Hz_10ms_envelope_correlation": 0.985651034675872, "1000–3000 Hz_10ms_modulation_mae_db": 1.1906404397306904, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9143879441182876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07185816031884365, "1000–3000 Hz_40ms_envelope_correlation": 0.9958026751770338, "1000–3000 Hz_40ms_modulation_mae_db": 1.2537698074544166, "3000–6000 Hz_10ms_log_rms_mae_db": 1.760954214547126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2033169628121492, "3000–6000 Hz_10ms_envelope_correlation": 0.9639272906923487, "3000–6000 Hz_10ms_modulation_mae_db": 0.48970699715255606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0621785531171744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12372880195004773, "3000–6000 Hz_40ms_envelope_correlation": 0.9852198511207544, "3000–6000 Hz_40ms_modulation_mae_db": 0.5104277406768029, "mr_stft_db": 5.651267245919146, "band_mae_db": 0.24545874045075997, "rms_gain_error_db": -0.11169838868278681, "output_peak": 0.2458755662587669}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9385737310253506, "全帯域_10ms_rms_envelope_nrmse": 0.07194347968210875, "全帯域_10ms_envelope_correlation": 0.9948657345192954, "全帯域_10ms_modulation_mae_db": 0.26338236760192224, "全帯域_40ms_log_rms_mae_db": 0.47612010363842105, "全帯域_40ms_rms_envelope_nrmse": 0.03969232104113, "全帯域_40ms_envelope_correlation": 0.9984697354878164, "全帯域_40ms_modulation_mae_db": 0.3816276137160717, "300–1000 Hz_10ms_log_rms_mae_db": 1.593248808697195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09281585089865042, "300–1000 Hz_10ms_envelope_correlation": 0.9927198419179913, "300–1000 Hz_10ms_modulation_mae_db": 0.48994366131338923, "300–1000 Hz_40ms_log_rms_mae_db": 0.9628622133354442, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05818961100598939, "300–1000 Hz_40ms_envelope_correlation": 0.9971252901773763, "300–1000 Hz_40ms_modulation_mae_db": 0.46753195438626144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6051206085769654, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12470491216340372, "1000–3000 Hz_10ms_envelope_correlation": 0.9873496419748099, "1000–3000 Hz_10ms_modulation_mae_db": 0.634622589062555, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8920157864852406, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07058943684072516, "1000–3000 Hz_40ms_envelope_correlation": 0.9956897739528318, "1000–3000 Hz_40ms_modulation_mae_db": 0.46506834247103956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7421189928602956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20459154168225416, "3000–6000 Hz_10ms_envelope_correlation": 0.9637105797422498, "3000–6000 Hz_10ms_modulation_mae_db": 0.40675542035629064, "3000–6000 Hz_40ms_log_rms_mae_db": 1.030569408264589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11888313880396055, "3000–6000 Hz_40ms_envelope_correlation": 0.9863439284267809, "3000–6000 Hz_40ms_modulation_mae_db": 0.40617337812837323, "mr_stft_db": 5.640546152530764, "band_mae_db": 0.20057922180006918, "rms_gain_error_db": -0.13735267494960474, "output_peak": 0.2294951303892298}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9540326592170657, "全帯域_10ms_rms_envelope_nrmse": 0.07985197598677611, "全帯域_10ms_envelope_correlation": 0.9934639526990372, "全帯域_10ms_modulation_mae_db": 0.5429141693964802, "全帯域_40ms_log_rms_mae_db": 0.528706625788348, "全帯域_40ms_rms_envelope_nrmse": 0.044258198622935396, "全帯域_40ms_envelope_correlation": 0.9978178872245582, "全帯域_40ms_modulation_mae_db": 0.7209544603825419, "300–1000 Hz_10ms_log_rms_mae_db": 1.5468646256982903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09841292163725128, "300–1000 Hz_10ms_envelope_correlation": 0.9915479444498045, "300–1000 Hz_10ms_modulation_mae_db": 0.26979949817985743, "300–1000 Hz_40ms_log_rms_mae_db": 0.9486730711450584, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05865481127279558, "300–1000 Hz_40ms_envelope_correlation": 0.9967604176685552, "300–1000 Hz_40ms_modulation_mae_db": 0.6063583395631463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.645085676477562, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14767497804576946, "1000–3000 Hz_10ms_envelope_correlation": 0.9821967132594526, "1000–3000 Hz_10ms_modulation_mae_db": 0.7714580516924897, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8913329807493268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07783731975280625, "1000–3000 Hz_40ms_envelope_correlation": 0.9947755338594174, "1000–3000 Hz_40ms_modulation_mae_db": 0.9410468333981709, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7587729825737242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20429182516495722, "3000–6000 Hz_10ms_envelope_correlation": 0.9641759380525894, "3000–6000 Hz_10ms_modulation_mae_db": 0.07452781750625043, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0603058871674533, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11767010308710793, "3000–6000 Hz_40ms_envelope_correlation": 0.9867006857984337, "3000–6000 Hz_40ms_modulation_mae_db": 0.1348026238423895, "mr_stft_db": 5.652907610910492, "band_mae_db": 0.17937342221573757, "rms_gain_error_db": -0.05722492335867414, "output_peak": 0.2347662395844505}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9705195746405637, "全帯域_10ms_rms_envelope_nrmse": 0.07390920704257917, "全帯域_10ms_envelope_correlation": 0.994519534475171, "全帯域_10ms_modulation_mae_db": 0.3251336060136946, "全帯域_40ms_log_rms_mae_db": 0.5317033487966804, "全帯域_40ms_rms_envelope_nrmse": 0.0408378119202478, "全帯域_40ms_envelope_correlation": 0.9983032458060024, "全帯域_40ms_modulation_mae_db": 0.5780295356168018, "300–1000 Hz_10ms_log_rms_mae_db": 1.5359267934906746, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09157968538772403, "300–1000 Hz_10ms_envelope_correlation": 0.9927776206647366, "300–1000 Hz_10ms_modulation_mae_db": 0.5242824899292824, "300–1000 Hz_40ms_log_rms_mae_db": 0.9161814785602995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05509462397212344, "300–1000 Hz_40ms_envelope_correlation": 0.9972760377244158, "300–1000 Hz_40ms_modulation_mae_db": 0.5392095047390436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5989757578840742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13415183541117306, "1000–3000 Hz_10ms_envelope_correlation": 0.985465080057545, "1000–3000 Hz_10ms_modulation_mae_db": 0.6775211810869166, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8767315280523195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07530352765598318, "1000–3000 Hz_40ms_envelope_correlation": 0.9953173341664644, "1000–3000 Hz_40ms_modulation_mae_db": 0.7961574322715759, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7599481680946854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2203860724872459, "3000–6000 Hz_10ms_envelope_correlation": 0.9574438258989139, "3000–6000 Hz_10ms_modulation_mae_db": 0.7335804167977019, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0836396325838191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1331992144444772, "3000–6000 Hz_40ms_envelope_correlation": 0.9828269071178775, "3000–6000 Hz_40ms_modulation_mae_db": 0.8067774948943581, "mr_stft_db": 5.641588955604704, "band_mae_db": 0.2446100955361213, "rms_gain_error_db": -0.11961575653642648, "output_peak": 0.2317188795119929}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9423700143402622, "全帯域_10ms_rms_envelope_nrmse": 0.0758626318725952, "全帯域_10ms_envelope_correlation": 0.9941311104300652, "全帯域_10ms_modulation_mae_db": 0.4211340984901477, "全帯域_40ms_log_rms_mae_db": 0.47445258994713646, "全帯域_40ms_rms_envelope_nrmse": 0.04362573722937739, "全帯域_40ms_envelope_correlation": 0.9979185445525114, "全帯域_40ms_modulation_mae_db": 0.6346555488738114, "300–1000 Hz_10ms_log_rms_mae_db": 1.5553335224138913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09639173553597646, "300–1000 Hz_10ms_envelope_correlation": 0.9919273796208402, "300–1000 Hz_10ms_modulation_mae_db": 0.7354623530718946, "300–1000 Hz_40ms_log_rms_mae_db": 0.952773239525591, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058080684945184584, "300–1000 Hz_40ms_envelope_correlation": 0.996880560086397, "300–1000 Hz_40ms_modulation_mae_db": 0.5061140437918873, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6168559907319877, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1382671089501228, "1000–3000 Hz_10ms_envelope_correlation": 0.9844467793203711, "1000–3000 Hz_10ms_modulation_mae_db": 0.8272446991988073, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9114983421161486, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08346698828471498, "1000–3000 Hz_40ms_envelope_correlation": 0.9940024049419709, "1000–3000 Hz_40ms_modulation_mae_db": 1.0417469560010268, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6996770890693946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.199901214321179, "3000–6000 Hz_10ms_envelope_correlation": 0.9650938108768815, "3000–6000 Hz_10ms_modulation_mae_db": 0.28924277421558636, "3000–6000 Hz_40ms_log_rms_mae_db": 1.023026298877268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12312624996086745, "3000–6000 Hz_40ms_envelope_correlation": 0.9854209672729556, "3000–6000 Hz_40ms_modulation_mae_db": 0.36760210426399803, "mr_stft_db": 5.647874710664145, "band_mae_db": 0.2294267846462826, "rms_gain_error_db": -0.07733732815951162, "output_peak": 0.23843530915680083}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9186926954743795, "全帯域_10ms_rms_envelope_nrmse": 0.07401540987667163, "全帯域_10ms_envelope_correlation": 0.994478188355698, "全帯域_10ms_modulation_mae_db": 0.3156816583535204, "全帯域_40ms_log_rms_mae_db": 0.46051790930264797, "全帯域_40ms_rms_envelope_nrmse": 0.03970779300941097, "全帯域_40ms_envelope_correlation": 0.9983738796492746, "全帯域_40ms_modulation_mae_db": 0.32203706662300174, "300–1000 Hz_10ms_log_rms_mae_db": 1.542408961527671, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09926032909782903, "300–1000 Hz_10ms_envelope_correlation": 0.9915342070496314, "300–1000 Hz_10ms_modulation_mae_db": 0.9897333851164217, "300–1000 Hz_40ms_log_rms_mae_db": 0.9104887824394633, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05863257909670916, "300–1000 Hz_40ms_envelope_correlation": 0.9969626004768727, "300–1000 Hz_40ms_modulation_mae_db": 0.5357670797748767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.577704956166133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1305150004772249, "1000–3000 Hz_10ms_envelope_correlation": 0.9865574098876623, "1000–3000 Hz_10ms_modulation_mae_db": 0.8321068115412832, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8932341457834336, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08062831761380365, "1000–3000 Hz_40ms_envelope_correlation": 0.994966622895435, "1000–3000 Hz_40ms_modulation_mae_db": 0.9664866249125684, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6452664383491704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20148731244175705, "3000–6000 Hz_10ms_envelope_correlation": 0.9651093911861484, "3000–6000 Hz_10ms_modulation_mae_db": 0.5994814894632593, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0014129042923896, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12545107410187542, "3000–6000 Hz_40ms_envelope_correlation": 0.9848654726279689, "3000–6000 Hz_40ms_modulation_mae_db": 0.5991653230489049, "mr_stft_db": 5.6178801571304895, "band_mae_db": 0.29793192334488877, "rms_gain_error_db": -0.11126786939938887, "output_peak": 0.23195844771041765}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.657769854814471, "spectral_convergence": 0.23932140969413185, "spectral_convergence_status": "ok", "amplitude_floor": 6.60007588392934e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.470228474210613, "spectral_convergence": 0.23724976757119637, "spectral_convergence_status": "ok", "amplitude_floor": 5.570424644238683e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.80516474830688, "spectral_convergence": 0.24060773161430046, "spectral_convergence_status": "ok", "amplitude_floor": 3.8555660806566813e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.162305217059986e-05, "output_power": 3.3442962164812446e-05, "delta_db": 0.24300952976677934, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000304634409424632, "output_power": 0.0002974879861715343, "delta_db": -0.10309524949698291, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004181285613649292, "output_power": 0.00039924528088504526, "delta_db": -0.20070042742854535, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001153043826104485, "output_power": 0.00011542011598942025, "delta_db": 0.0043569167886361995, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.6981779049220075e-05, "output_power": 5.462974851624226e-05, "delta_db": -0.18306803105302716, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.999437610841641e-06, "output_power": 3.3799830642809102e-06, "delta_db": 0.5187469129201748, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20882951124235763, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12350431627678517, "pitch": null, "output_rms_dbfs": -30.440208570920127, "output_peak": 0.23617001229344264, "reference_rms_dbfs": -30.31670425464334, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6443876924439875}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.667347842324586e-05, "log_rms_mae_db": 0.967677397197783, "rms_envelope_nrmse": 0.07830117035848565, "envelope_correlation": 0.9938451734642316, "rms_gain_error_db": -0.12301934617045981, "reference_mean_power": 0.0009300339703354519, "output_mean_power": 0.0009040592149031845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3130295372770493e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001632046380308234, "output_power": 0.00015468930336188743, "reference_active": true, "delta_db": -0.23272212959096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.195140055898603e-05, "output_power": 2.7678017042427184e-05, "reference_active": true, "delta_db": -0.6235492738964286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.743176048402797e-06, "output_power": 3.4861180808102386e-06, "reference_active": true, "delta_db": -0.3089815993351983}], "active_band_count": 3, "active_band_mae_db": 0.388417667607529}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.141277463219576e-05, "log_rms_mae_db": 0.5017316372106703, "rms_envelope_nrmse": 0.04457497609820027, "envelope_correlation": 0.9979566530741625, "rms_gain_error_db": -0.12321659918146771, "reference_mean_power": 0.0009316560347730994, "output_mean_power": 0.0009055948446747924}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0476212509594803e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014768349458157853, "output_power": 0.00013948425520830313, "reference_active": true, "delta_db": -0.24806772548147338}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4986865819787958e-05, "output_power": 1.3177367288055502e-05, "reference_active": true, "delta_db": -0.5588216796538291}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9796240655945466e-07, "output_power": 3.2584757406848527e-07, "reference_active": true, "delta_db": -0.8682755644226298}], "active_band_count": 3, "active_band_mae_db": 0.5583883231859774}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.73108999878072e-05, "log_rms_mae_db": 1.5333452373714755, "rms_envelope_nrmse": 0.10307757547548868, "envelope_correlation": 0.9910003536117239, "rms_gain_error_db": -0.20225550105702622, "reference_mean_power": 0.0005508019269041907, "output_mean_power": 0.0005257386441687958}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0420364913089255e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.387221376866869e-05, "output_power": 8.447632939706844e-05, "reference_active": true, "delta_db": -0.4580202490466494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.257079739763329e-05, "output_power": 1.8214598563566813e-05, "reference_active": true, "delta_db": -0.9312729827119061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1515320396484854e-06, "output_power": 2.6473632921971064e-06, "reference_active": true, "delta_db": 0.9006572494072252}], "active_band_count": 3, "active_band_mae_db": 0.7633168270552603}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091336805501339e-05, "log_rms_mae_db": 0.8928348728743134, "rms_envelope_nrmse": 0.06181447985500663, "envelope_correlation": 0.9968028239298015, "rms_gain_error_db": -0.20276273031996553, "reference_mean_power": 0.0005517580103658142, "output_mean_power": 0.000526589716682884}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.88495020416125e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.537091697949908e-05, "output_power": 7.642059529459844e-05, "reference_active": true, "delta_db": -0.4809952986159739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0155340206288246e-05, "output_power": 8.323445270846842e-06, "reference_active": true, "delta_db": -0.8639134868938992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.398901151759038e-07, "output_power": 2.3717775236010812e-07, "reference_active": true, "delta_db": -0.049384038485664}], "active_band_count": 3, "active_band_mae_db": 0.46476427466517906}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.045309710707616e-05, "log_rms_mae_db": 1.5817649766136128, "rms_envelope_nrmse": 0.13400724658291727, "envelope_correlation": 0.9853438091366308, "rms_gain_error_db": -0.0607250125230325, "reference_mean_power": 0.00014468505517057894, "output_mean_power": 0.00014267608141569623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.678016723697031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.684581121209281e-05, "output_power": 3.497765898449339e-05, "reference_active": true, "delta_db": -0.2259738329232126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.802139012131204e-06, "output_power": 6.859126093411212e-06, "reference_active": true, "delta_db": -0.5594489724849849}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.902802552587548e-06, "output_power": 1.2424462645776109e-06, "reference_active": true, "delta_db": -1.8511611064174351}], "active_band_count": 3, "active_band_mae_db": 0.8788613039418776}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.539248843957361e-05, "log_rms_mae_db": 0.8691188439145279, "rms_envelope_nrmse": 0.077026638278054, "envelope_correlation": 0.9947340243202796, "rms_gain_error_db": -0.05992955611666366, "reference_mean_power": 0.0001448248558759811, "output_mean_power": 0.00014284010125099018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.24555278166651e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.61104008282169e-05, "output_power": 3.3982633759198354e-05, "reference_active": true, "delta_db": -0.26375274152894335}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6607268103199764e-06, "output_power": 3.2416567520436947e-06, "reference_active": true, "delta_db": -0.5280029293458848}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4925257777124562e-07, "output_power": 8.959981972593265e-08, "reference_active": true, "delta_db": -2.2161470478716057}], "active_band_count": 3, "active_band_mae_db": 1.0026342395821446}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.667503736683173e-05, "log_rms_mae_db": 1.680542470829995, "rms_envelope_nrmse": 0.21167518710893046, "envelope_correlation": 0.9608902273995144, "rms_gain_error_db": -0.0833763356299532, "reference_mean_power": 8.017666159457313e-06, "output_mean_power": 7.865210231933934e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.8921445305362116e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.737362730986327e-06, "output_power": 2.6241085041961194e-06, "reference_active": true, "delta_db": -0.18350561362780818}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.579169756744011e-07, "output_power": 7.195063152555911e-07, "reference_active": true, "delta_db": -0.225870242348763}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.809426009650544e-07, "output_power": 1.673190885411446e-07, "reference_active": true, "delta_db": -0.3399533867842671}], "active_band_count": 3, "active_band_mae_db": 0.24977641425361274}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4241724067408686e-05, "log_rms_mae_db": 1.0109432420748428, "rms_envelope_nrmse": 0.11553932094263854, "envelope_correlation": 0.9870512901041556, "rms_gain_error_db": -0.08344805913054228, "reference_mean_power": 8.048582997885633e-06, "output_mean_power": 7.895408793329155e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.450576990854383e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.750016160596199e-06, "output_power": 2.6565255782115864e-06, "reference_active": true, "delta_db": -0.1502124414260895}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.547469624529576e-07, "output_power": 3.489340024438828e-07, "reference_active": true, "delta_db": -0.07175393421395568}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2784081824019638e-08, "output_power": 1.0188882077558107e-08, "reference_active": true, "delta_db": -0.9854300576154883}], "active_band_count": 3, "active_band_mae_db": 0.40246547775184444}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8968095875087768, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9689336cb1501b322e614861a970c71ce79ad54a0c52d8733a6cc864f59beb48", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9849617151309698, "standard_deviation": 0.001781843147177735, "minimum": 0.9821967132594526, "maximum": 0.9873496419748099, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6058281564499537, "standard_deviation": 0.02114390192176313, "minimum": 1.577704956166133, "maximum": 1.645085676477562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8102855921948418, "standard_deviation": 0.17722283840243105, "minimum": 0.634622589062555, "maximum": 1.1906404397306904, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13606078444030834, "standard_deviation": 0.007642909583137384, "minimum": 0.12470491216340372, "maximum": 0.14767497804576946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949855847978208, "standard_deviation": 0.0005975567568565345, "minimum": 0.9940024049419709, "maximum": 0.9958026751770338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8936063367292983, "standard_deviation": 0.01554801189783351, "minimum": 0.8691188439145279, "maximum": 0.9143879441182876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8856473445130264, "standard_deviation": 0.2505872644732747, "minimum": 0.46506834247103956, "maximum": 1.2537698074544166, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07685024681462883, "standard_deviation": 0.004261328519420336, "minimum": 0.07058943684072516, "maximum": 0.08346698828471498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9632926403493585, "standard_deviation": 0.0028071020553162964, "minimum": 0.9574438258989139, "maximum": 0.9659900589462216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7216346028812801, "standard_deviation": 0.042816792456583166, "minimum": 1.6452664383491704, "maximum": 1.760954214547126, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4939221307767001, "standard_deviation": 0.32316608407875613, "minimum": 0.07452781750625043, "maximum": 1.1083057164683439, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20537282561106818, "standard_deviation": 0.007375690168285775, "minimum": 0.19733248887007232, "maximum": 0.2203860724872459, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9856154886271922, "standard_deviation": 0.0013667522378257451, "minimum": 0.9828269071178775, "maximum": 0.9870512901041556, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0375015881580805, "standard_deviation": 0.02832091942715666, "minimum": 1.0014129042923896, "maximum": 1.0836396325838191, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5442844851970068, "standard_deviation": 0.3048504058304249, "minimum": 0.1348026238423895, "maximum": 1.1268617389693827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12202512164954121, "standard_deviation": 0.005648347444170134, "minimum": 0.11553932094263854, "maximum": 0.1331992144444772, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920521758239814, "standard_deviation": 0.0008817305807152674, "minimum": 0.9910003536117239, "maximum": 0.9935733019909975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5416142919722016, "standard_deviation": 0.03329328435224805, "minimum": 1.4733581869444483, "maximum": 1.593248808697195, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6231158328290216, "standard_deviation": 0.2499853946483359, "minimum": 0.26979949817985743, "maximum": 0.9897333851164217, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09629980746445513, "standard_deviation": 0.00568024988259431, "minimum": 0.0865159017181841, "maximum": 0.10307757547548868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997085038128589, "standard_deviation": 0.0005262045778959142, "minimum": 0.9965961803559262, "maximum": 0.9982763946093678, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9198490764849674, "standard_deviation": 0.057660501331625966, "minimum": 0.7964316386734314, "maximum": 0.9785473153261384, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5309810783314725, "standard_deviation": 0.05245618156477336, "minimum": 0.46476427466517906, "maximum": 0.6063583395631463, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05758632635390192, "standard_deviation": 0.006045742650157362, "minimum": 0.044703234879848085, "maximum": 0.06552058580355852, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23357109750953498, "standard_deviation": 0.037586875955733665, "minimum": 0.17937342221573757, "maximum": 0.29793192334488877, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.641353851627998, "standard_deviation": 0.01123687175650797, "minimum": 5.6178801571304895, "maximum": 5.652907610910492, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.23621936928897888, "standard_deviation": 0.005478076544417317, "minimum": 0.2294951303892298, "maximum": 0.2458755662587669, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11328207503946769, "standard_deviation": 0.03419911835113011, "minimum": -0.16825534295256378, "maximum": -0.05722492335867414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943612531172439, "standard_deviation": 0.0005637048726188184, "minimum": 0.9934639526990372, "maximum": 0.9952587648139305, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9503404385943726, "standard_deviation": 0.01771226605760281, "minimum": 0.9186926954743795, "maximum": 0.9705195746405637, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3664582238786865, "standard_deviation": 0.08556718869240186, "minimum": 0.26338236760192224, "maximum": 0.5429141693964802, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07484913458821035, "standard_deviation": 0.003515049573058441, "minimum": 0.06877846908551812, "maximum": 0.07985197598677611, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982602743406266, "standard_deviation": 0.0003545737264442512, "minimum": 0.9978178872245582, "maximum": 0.9989007079950856, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49573620029205034, "standard_deviation": 0.03001348336918516, "minimum": 0.46051790930264797, "maximum": 0.5317033487966804, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5040009734229138, "standard_deviation": 0.15375440737775387, "minimum": 0.29551787528253654, "maximum": 0.7209544603825419, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04107962083130792, "standard_deviation": 0.00363941272982349, "minimum": 0.03345841086921372, "maximum": 0.04457497609820027, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-126133-0003", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.4083499908447266, "predicted_mos": 2.4083499908447266, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "643ba200c0fbcf53f0b5721660993e3128ab21d594332ab1c2552d5d6d8e4141", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.381102200131863, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9220268556104315, "全帯域_10ms_rms_envelope_nrmse": 0.07207395342523243, "全帯域_10ms_envelope_correlation": 0.994814067627064, "全帯域_10ms_modulation_mae_db": 0.3645197239599418, "全帯域_40ms_log_rms_mae_db": 0.49651588181086614, "全帯域_40ms_rms_envelope_nrmse": 0.042036239248708834, "全帯域_40ms_envelope_correlation": 0.9982036568616085, "全帯域_40ms_modulation_mae_db": 0.6433885838001584, "300–1000 Hz_10ms_log_rms_mae_db": 1.5055484823324639, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0992311659720801, "300–1000 Hz_10ms_envelope_correlation": 0.9916353600587735, "300–1000 Hz_10ms_modulation_mae_db": 0.6645490783130074, "300–1000 Hz_40ms_log_rms_mae_db": 0.9045852016492485, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06184921919003151, "300–1000 Hz_40ms_envelope_correlation": 0.9967121668710663, "300–1000 Hz_40ms_modulation_mae_db": 0.5653475506706195, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5778825245332682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1241709488300984, "1000–3000 Hz_10ms_envelope_correlation": 0.9874818924418656, "1000–3000 Hz_10ms_modulation_mae_db": 0.655418421844927, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8388510999587264, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06470022809228827, "1000–3000 Hz_40ms_envelope_correlation": 0.9964453228242088, "1000–3000 Hz_40ms_modulation_mae_db": 0.9434845297764974, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6121429384520782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2054392363350027, "3000–6000 Hz_10ms_envelope_correlation": 0.9632129922912114, "3000–6000 Hz_10ms_modulation_mae_db": 0.21860468367045907, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9866993499076192, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13213147567883216, "3000–6000 Hz_40ms_envelope_correlation": 0.9835849602183251, "3000–6000 Hz_40ms_modulation_mae_db": 0.1891184625935828, "mr_stft_db": 5.6113627418568734, "band_mae_db": 0.27572438832689766, "rms_gain_error_db": -0.12660138497432136, "output_peak": 0.23470452526381694}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9033116950882891, "全帯域_10ms_rms_envelope_nrmse": 0.06913181322632816, "全帯域_10ms_envelope_correlation": 0.9951188570544608, "全帯域_10ms_modulation_mae_db": 0.5680170049381211, "全帯域_40ms_log_rms_mae_db": 0.45934807342895156, "全帯域_40ms_rms_envelope_nrmse": 0.0360123721513662, "全帯域_40ms_envelope_correlation": 0.9985792712772456, "全帯域_40ms_modulation_mae_db": 0.675228823122452, "300–1000 Hz_10ms_log_rms_mae_db": 1.5054010440326475, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09079193259203054, "300–1000 Hz_10ms_envelope_correlation": 0.9928777352485781, "300–1000 Hz_10ms_modulation_mae_db": 0.30741758979531125, "300–1000 Hz_40ms_log_rms_mae_db": 0.8638150835495716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05673152468301286, "300–1000 Hz_40ms_envelope_correlation": 0.9970612778117112, "300–1000 Hz_40ms_modulation_mae_db": 0.535704174171428, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6034109237927643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1370778651972201, "1000–3000 Hz_10ms_envelope_correlation": 0.9846656920335115, "1000–3000 Hz_10ms_modulation_mae_db": 0.15497797638942204, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8780640258740747, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07750705363759723, "1000–3000 Hz_40ms_envelope_correlation": 0.9946878720635202, "1000–3000 Hz_40ms_modulation_mae_db": 0.38923344035876206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6588693769489562, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19941472393363538, "3000–6000 Hz_10ms_envelope_correlation": 0.965438301217773, "3000–6000 Hz_10ms_modulation_mae_db": 0.735487699712237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0025755457062402, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11531993088475115, "3000–6000 Hz_40ms_envelope_correlation": 0.9873209054092907, "3000–6000 Hz_40ms_modulation_mae_db": 0.8774778402938069, "mr_stft_db": 5.609206532509325, "band_mae_db": 0.18428865626277346, "rms_gain_error_db": -0.061707143255890926, "output_peak": 0.2240562729051889}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9660858857543915, "全帯域_10ms_rms_envelope_nrmse": 0.0709983288016396, "全帯域_10ms_envelope_correlation": 0.9948275233664771, "全帯域_10ms_modulation_mae_db": 0.2856451892627239, "全帯域_40ms_log_rms_mae_db": 0.47916966206835304, "全帯域_40ms_rms_envelope_nrmse": 0.036275407290573825, "全帯域_40ms_envelope_correlation": 0.9985116191552627, "全帯域_40ms_modulation_mae_db": 0.35802826687380734, "300–1000 Hz_10ms_log_rms_mae_db": 1.4724047552682773, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08874055327071778, "300–1000 Hz_10ms_envelope_correlation": 0.9931398733860162, "300–1000 Hz_10ms_modulation_mae_db": 0.31322281364667676, "300–1000 Hz_40ms_log_rms_mae_db": 0.82287900836348, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04893799680994895, "300–1000 Hz_40ms_envelope_correlation": 0.9977615343623589, "300–1000 Hz_40ms_modulation_mae_db": 0.3942246017329259, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6322766731292193, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15406545244992031, "1000–3000 Hz_10ms_envelope_correlation": 0.9811696587988366, "1000–3000 Hz_10ms_modulation_mae_db": 0.2908582028715393, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9307891590743003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10107957867929929, "1000–3000 Hz_40ms_envelope_correlation": 0.9912345273277318, "1000–3000 Hz_40ms_modulation_mae_db": 0.5222057742025389, "3000–6000 Hz_10ms_log_rms_mae_db": 1.658111349897318, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19632597802587673, "3000–6000 Hz_10ms_envelope_correlation": 0.9665040846658594, "3000–6000 Hz_10ms_modulation_mae_db": 0.6515744370274081, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9974569519485983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11598007259428626, "3000–6000 Hz_40ms_envelope_correlation": 0.9877382644700243, "3000–6000 Hz_40ms_modulation_mae_db": 0.6426218282784862, "mr_stft_db": 5.616911581702812, "band_mae_db": 0.13684616112129927, "rms_gain_error_db": -0.01401772857733445, "output_peak": 0.24001550914439324}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9689504773768087, "全帯域_10ms_rms_envelope_nrmse": 0.07115143753334817, "全帯域_10ms_envelope_correlation": 0.9948572154454105, "全帯域_10ms_modulation_mae_db": 0.30734219642211774, "全帯域_40ms_log_rms_mae_db": 0.49957303400264597, "全帯域_40ms_rms_envelope_nrmse": 0.03873568017326826, "全帯域_40ms_envelope_correlation": 0.9983921888340348, "全帯域_40ms_modulation_mae_db": 0.5241938143530508, "300–1000 Hz_10ms_log_rms_mae_db": 1.5289191463454792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08957918873816394, "300–1000 Hz_10ms_envelope_correlation": 0.9930387748781119, "300–1000 Hz_10ms_modulation_mae_db": 0.3955973790507687, "300–1000 Hz_40ms_log_rms_mae_db": 0.8985598273260423, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052132342881442806, "300–1000 Hz_40ms_envelope_correlation": 0.9975043576032103, "300–1000 Hz_40ms_modulation_mae_db": 0.6196816793549581, "1000–3000 Hz_10ms_log_rms_mae_db": 1.580854794579698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13664417052529348, "1000–3000 Hz_10ms_envelope_correlation": 0.9849256398236288, "1000–3000 Hz_10ms_modulation_mae_db": 0.9006198461277295, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8554754519756054, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07643037165353847, "1000–3000 Hz_40ms_envelope_correlation": 0.9951995941250542, "1000–3000 Hz_40ms_modulation_mae_db": 0.9921857412636825, "3000–6000 Hz_10ms_log_rms_mae_db": 1.73564036566679, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20789932635779707, "3000–6000 Hz_10ms_envelope_correlation": 0.9629381592481502, "3000–6000 Hz_10ms_modulation_mae_db": 0.7043080150169027, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0165909580766355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1159544441222882, "3000–6000 Hz_40ms_envelope_correlation": 0.9871249798569259, "3000–6000 Hz_40ms_modulation_mae_db": 0.8973489321643124, "mr_stft_db": 5.620415834853778, "band_mae_db": 0.2047090551736225, "rms_gain_error_db": -0.08361786027195817, "output_peak": 0.2406649199023635}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.968604838114856, "全帯域_10ms_rms_envelope_nrmse": 0.06976076369256863, "全帯域_10ms_envelope_correlation": 0.995062061155714, "全帯域_10ms_modulation_mae_db": 0.41961621676710675, "全帯域_40ms_log_rms_mae_db": 0.47658992343986045, "全帯域_40ms_rms_envelope_nrmse": 0.033486790975189674, "全帯域_40ms_envelope_correlation": 0.9988257956647456, "全帯域_40ms_modulation_mae_db": 0.6274995418230719, "300–1000 Hz_10ms_log_rms_mae_db": 1.527551520728436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09798215183733329, "300–1000 Hz_10ms_envelope_correlation": 0.9916936575684082, "300–1000 Hz_10ms_modulation_mae_db": 0.6784456559814775, "300–1000 Hz_40ms_log_rms_mae_db": 0.850466976691433, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052294178060895403, "300–1000 Hz_40ms_envelope_correlation": 0.9975502612961585, "300–1000 Hz_40ms_modulation_mae_db": 0.4463577125119477, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6135906060538057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13703249309416654, "1000–3000 Hz_10ms_envelope_correlation": 0.9846762112224532, "1000–3000 Hz_10ms_modulation_mae_db": 0.5633280555714785, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8778149181000212, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08486754288655833, "1000–3000 Hz_40ms_envelope_correlation": 0.9936262297700958, "1000–3000 Hz_40ms_modulation_mae_db": 0.7113868752911242, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6419234659832094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20887657712422658, "3000–6000 Hz_10ms_envelope_correlation": 0.9617824864797728, "3000–6000 Hz_10ms_modulation_mae_db": 0.579022249577242, "3000–6000 Hz_40ms_log_rms_mae_db": 1.012672505609932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12690830382280305, "3000–6000 Hz_40ms_envelope_correlation": 0.9845159858527878, "3000–6000 Hz_40ms_modulation_mae_db": 0.557043190655751, "mr_stft_db": 5.61628846192869, "band_mae_db": 0.2142443570677457, "rms_gain_error_db": -0.08326079493057101, "output_peak": 0.23436182929729277}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9484548318959497, "全帯域_10ms_rms_envelope_nrmse": 0.07138289580647438, "全帯域_10ms_envelope_correlation": 0.9947996917685886, "全帯域_10ms_modulation_mae_db": 0.24928009267540183, "全帯域_40ms_log_rms_mae_db": 0.5041570301791857, "全帯域_40ms_rms_envelope_nrmse": 0.041383260405362776, "全帯域_40ms_envelope_correlation": 0.9981214514250842, "全帯域_40ms_modulation_mae_db": 0.2961541263803944, "300–1000 Hz_10ms_log_rms_mae_db": 1.492107707573423, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09252579846088387, "300–1000 Hz_10ms_envelope_correlation": 0.9926074910339363, "300–1000 Hz_10ms_modulation_mae_db": 0.5431021565877358, "300–1000 Hz_40ms_log_rms_mae_db": 0.8999519250410603, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05851811103983159, "300–1000 Hz_40ms_envelope_correlation": 0.9968829158003534, "300–1000 Hz_40ms_modulation_mae_db": 0.9026483954310054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6359305096620325, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13878405957132103, "1000–3000 Hz_10ms_envelope_correlation": 0.9843727155900707, "1000–3000 Hz_10ms_modulation_mae_db": 0.036462630038391, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8858742367643373, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07681653031387434, "1000–3000 Hz_40ms_envelope_correlation": 0.9948203608140559, "1000–3000 Hz_40ms_modulation_mae_db": 0.03979410393570784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6870011740878748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2091283096188402, "3000–6000 Hz_10ms_envelope_correlation": 0.9617255647160744, "3000–6000 Hz_10ms_modulation_mae_db": 0.706991705207514, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0412538684434984, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13413480702985697, "3000–6000 Hz_40ms_envelope_correlation": 0.9828493083899436, "3000–6000 Hz_40ms_modulation_mae_db": 0.8539619450713175, "mr_stft_db": 5.617247934462632, "band_mae_db": 0.21008451998719294, "rms_gain_error_db": -0.06949832987369667, "output_peak": 0.2397428454212977}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9979230224317523, "全帯域_10ms_rms_envelope_nrmse": 0.06974595540611403, "全帯域_10ms_envelope_correlation": 0.9951054164377249, "全帯域_10ms_modulation_mae_db": 0.33865833735860223, "全帯域_40ms_log_rms_mae_db": 0.5428315624487597, "全帯域_40ms_rms_envelope_nrmse": 0.03787289496074892, "全帯域_40ms_envelope_correlation": 0.9985247836294154, "全帯域_40ms_modulation_mae_db": 0.20058429732975105, "300–1000 Hz_10ms_log_rms_mae_db": 1.5411406563176826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09564509779503573, "300–1000 Hz_10ms_envelope_correlation": 0.9922494597088478, "300–1000 Hz_10ms_modulation_mae_db": 0.9709585090256706, "300–1000 Hz_40ms_log_rms_mae_db": 0.9242892741624521, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05704841776064639, "300–1000 Hz_40ms_envelope_correlation": 0.9972740814745256, "300–1000 Hz_40ms_modulation_mae_db": 0.35497665847787613, "1000–3000 Hz_10ms_log_rms_mae_db": 1.593446541837049, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12584525777849012, "1000–3000 Hz_10ms_envelope_correlation": 0.9871618362712306, "1000–3000 Hz_10ms_modulation_mae_db": 1.065819254393988, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8525238268640556, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06595933395767133, "1000–3000 Hz_40ms_envelope_correlation": 0.9963455513233771, "1000–3000 Hz_40ms_modulation_mae_db": 0.9411949639168536, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7018552338965909, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2113634809872488, "3000–6000 Hz_10ms_envelope_correlation": 0.961000942728401, "3000–6000 Hz_10ms_modulation_mae_db": 0.429363877600808, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0350352854552034, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13081517911988202, "3000–6000 Hz_40ms_envelope_correlation": 0.9841549355125068, "3000–6000 Hz_40ms_modulation_mae_db": 0.5040904395510059, "mr_stft_db": 5.632882081689101, "band_mae_db": 0.24327164327621534, "rms_gain_error_db": -0.10302419908827873, "output_peak": 0.2400211873038911}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.940526286246707, "全帯域_10ms_rms_envelope_nrmse": 0.06825951042649864, "全帯域_10ms_envelope_correlation": 0.9953234423427163, "全帯域_10ms_modulation_mae_db": 0.2883700185928719, "全帯域_40ms_log_rms_mae_db": 0.4770743823259386, "全帯域_40ms_rms_envelope_nrmse": 0.03481738808219438, "全帯域_40ms_envelope_correlation": 0.9987802810014641, "全帯域_40ms_modulation_mae_db": 0.534513857351651, "300–1000 Hz_10ms_log_rms_mae_db": 1.4451774380682032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08404624510648159, "300–1000 Hz_10ms_envelope_correlation": 0.993981627492762, "300–1000 Hz_10ms_modulation_mae_db": 0.5417031681892546, "300–1000 Hz_40ms_log_rms_mae_db": 0.8040304067807865, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04777903497555483, "300–1000 Hz_40ms_envelope_correlation": 0.9980460194592091, "300–1000 Hz_40ms_modulation_mae_db": 0.45773629988880876, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5691382904346551, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12063615626168908, "1000–3000 Hz_10ms_envelope_correlation": 0.9883719468300207, "1000–3000 Hz_10ms_modulation_mae_db": 0.5967116503270887, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8216273565683636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061549300411268916, "1000–3000 Hz_40ms_envelope_correlation": 0.9970870047076454, "1000–3000 Hz_40ms_modulation_mae_db": 0.5269174542118541, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6840965952500653, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19509125376692735, "3000–6000 Hz_10ms_envelope_correlation": 0.9669786060909069, "3000–6000 Hz_10ms_modulation_mae_db": 0.8070935208749876, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0365953229861555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11422799177015537, "3000–6000 Hz_40ms_envelope_correlation": 0.9873846016521797, "3000–6000 Hz_40ms_modulation_mae_db": 1.0013940710696192, "mr_stft_db": 5.623995958882247, "band_mae_db": 0.21196629460683927, "rms_gain_error_db": -0.10684933889550878, "output_peak": 0.23750095748518088}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6500118346795505, "spectral_convergence": 0.23495652792816382, "spectral_convergence_status": "ok", "amplitude_floor": 6.60007588392934e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.441052168014751, "spectral_convergence": 0.23503456618361213, "spectral_convergence_status": "ok", "amplitude_floor": 5.570424644238683e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.743024222876319, "spectral_convergence": 0.23758599161758684, "spectral_convergence_status": "ok", "amplitude_floor": 3.8555660806566813e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.162305217059986e-05, "output_power": 3.364579936375044e-05, "delta_db": 0.2692706628187477, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000304634409424632, "output_power": 0.0002960582721826156, "delta_db": -0.12401756475841427, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004181285613649292, "output_power": 0.00040244065414638886, "delta_db": -0.16607987932357515, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001153043826104485, "output_power": 0.00011451712212914562, "delta_db": -0.02975389275016033, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.6981779049220075e-05, "output_power": 5.2751444164483124e-05, "delta_db": -0.33501650668605437, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.999437610841641e-06, "output_power": 3.54862914647877e-06, "delta_db": 0.7302078236244343, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27572438832689766, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12660138497432136, "pitch": null, "output_rms_dbfs": -30.443305639617662, "output_peak": 0.23470452526381694, "reference_rms_dbfs": -30.31670425464334, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6113627418568734}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.667347842324586e-05, "log_rms_mae_db": 0.9220268556104315, "rms_envelope_nrmse": 0.07207395342523243, "envelope_correlation": 0.994814067627064, "rms_gain_error_db": -0.12582934962047734, "reference_mean_power": 0.0009300339703354519, "output_mean_power": 0.0009034744531945022}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3130295372770493e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001632046380308234, "output_power": 0.0001524276341431327, "reference_active": true, "delta_db": -0.2966878767136543}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.195140055898603e-05, "output_power": 2.9083388392800854e-05, "reference_active": true, "delta_db": -0.4084489669289349}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.743176048402797e-06, "output_power": 3.42293002099672e-06, "reference_active": true, "delta_db": -0.38842232823723627}], "active_band_count": 3, "active_band_mae_db": 0.3645197239599418}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.141277463219576e-05, "log_rms_mae_db": 0.49651588181086614, "rms_envelope_nrmse": 0.042036239248708834, "envelope_correlation": 0.9982036568616085, "rms_gain_error_db": -0.12418320200323113, "reference_mean_power": 0.0009316560347730994, "output_mean_power": 0.0009053933101945729}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0476212509594803e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014768349458157853, "output_power": 0.0001377277839221176, "reference_active": true, "delta_db": -0.30310400733657306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4986865819787958e-05, "output_power": 1.3290410010010355e-05, "reference_active": true, "delta_db": -0.5217243984407657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9796240655945466e-07, "output_power": 3.0853775410925914e-07, "reference_active": true, "delta_db": -1.1053373456231366}], "active_band_count": 3, "active_band_mae_db": 0.6433885838001584}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.73108999878072e-05, "log_rms_mae_db": 1.5055484823324639, "rms_envelope_nrmse": 0.0992311659720801, "envelope_correlation": 0.9916353600587735, "rms_gain_error_db": -0.18755693419191174, "reference_mean_power": 0.0005508019269041907, "output_mean_power": 0.0005275210053605936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0420364913089255e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.387221376866869e-05, "output_power": 8.534984548275461e-05, "reference_active": true, "delta_db": -0.41334320656705903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.257079739763329e-05, "output_power": 1.8434592716567155e-05, "reference_active": true, "delta_db": -0.8791335542955279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1515320396484854e-06, "output_power": 2.528511104047961e-06, "reference_active": true, "delta_db": 0.7011704740764352}], "active_band_count": 3, "active_band_mae_db": 0.6645490783130074}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091336805501339e-05, "log_rms_mae_db": 0.9045852016492485, "rms_envelope_nrmse": 0.06184921919003151, "envelope_correlation": 0.9967121668710663, "rms_gain_error_db": -0.1862734315910084, "reference_mean_power": 0.0005517580103658142, "output_mean_power": 0.0005285928736153613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.88495020416125e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.537091697949908e-05, "output_power": 7.733858632710264e-05, "reference_active": true, "delta_db": -0.42913716921233447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0155340206288246e-05, "output_power": 8.17673168852796e-06, "reference_active": true, "delta_db": -0.9411472991568888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.398901151759038e-07, "output_power": 2.2255457920071878e-07, "reference_active": true, "delta_db": -0.3257581836426351}], "active_band_count": 3, "active_band_mae_db": 0.5653475506706195}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.045309710707616e-05, "log_rms_mae_db": 1.5778825245332682, "rms_envelope_nrmse": 0.1241709488300984, "envelope_correlation": 0.9874818924418656, "rms_gain_error_db": -0.13632803090625767, "reference_mean_power": 0.00014468505517057894, "output_mean_power": 0.00014021383604712195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.678016723697031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.684581121209281e-05, "output_power": 3.535781058121364e-05, "reference_active": true, "delta_db": -0.17902757681236564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.802139012131204e-06, "output_power": 7.214346871610252e-06, "reference_active": true, "delta_db": -0.34016664235598637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.902802552587548e-06, "output_power": 1.3636019722880656e-06, "reference_active": true, "delta_db": -1.447061046366429}], "active_band_count": 3, "active_band_mae_db": 0.655418421844927}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.539248843957361e-05, "log_rms_mae_db": 0.8388510999587264, "rms_envelope_nrmse": 0.06470022809228827, "envelope_correlation": 0.9964453228242088, "rms_gain_error_db": -0.13376938262521118, "reference_mean_power": 0.0001448248558759811, "output_mean_power": 0.00014043202771277528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.24555278166651e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.61104008282169e-05, "output_power": 3.4556819089048566e-05, "reference_active": true, "delta_db": -0.19098549759992767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6607268103199764e-06, "output_power": 3.3574381606066113e-06, "reference_active": true, "delta_db": -0.37559297615560844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4925257777124562e-07, "output_power": 8.862052715757357e-08, "reference_active": true, "delta_db": -2.2638751155739563}], "active_band_count": 3, "active_band_mae_db": 0.9434845297764974}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.667503736683173e-05, "log_rms_mae_db": 1.6121429384520782, "rms_envelope_nrmse": 0.2054392363350027, "envelope_correlation": 0.9632129922912114, "rms_gain_error_db": -0.1823947037260804, "reference_mean_power": 8.017666159457313e-06, "output_mean_power": 7.687913685121031e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.8921445305362116e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.737362730986327e-06, "output_power": 2.7071239179931734e-06, "reference_active": true, "delta_db": -0.048242141107187456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.579169756744011e-07, "output_power": 7.222053666079066e-07, "reference_active": true, "delta_db": -0.20960923131774556}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.809426009650544e-07, "output_power": 1.6509904492862841e-07, "reference_active": true, "delta_db": -0.3979626785864442}], "active_band_count": 3, "active_band_mae_db": 0.21860468367045907}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4241724067408686e-05, "log_rms_mae_db": 0.9866993499076192, "rms_envelope_nrmse": 0.13213147567883216, "envelope_correlation": 0.9835849602183251, "rms_gain_error_db": -0.18203369152411275, "reference_mean_power": 8.048582997885633e-06, "output_mean_power": 7.718200525975978e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.450576990854383e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.750016160596199e-06, "output_power": 2.7393318787723767e-06, "reference_active": true, "delta_db": -0.016905943894901583}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.547469624529576e-07, "output_power": 3.5212353413443256e-07, "reference_active": true, "delta_db": -0.03223633406749778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2784081824019638e-08, "output_power": 1.1346142420821832e-08, "reference_active": true, "delta_db": -0.5182131098183491}], "active_band_count": 3, "active_band_mae_db": 0.1891184625935828}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8928997747833578, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "492563946ffd6bbf35070ef16572e2122497de3b89beb7eb6dc4f5e07f92830d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9853531991264523, "standard_deviation": 0.0022832459497921257, "minimum": 0.9811696587988366, "maximum": 0.9883719468300207, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6008163580028114, "standard_deviation": 0.02499973470310641, "minimum": 1.5691382904346551, "maximum": 1.6359305096620325, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5330245046955704, "standard_deviation": 0.3559872790032329, "minimum": 0.036462630038391, "maximum": 1.065819254393988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13428205046352487, "standard_deviation": 0.010638451026194918, "minimum": 0.12063615626168908, "maximum": 0.15406545244992031, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949308078694612, "standard_deviation": 0.0018690072102249843, "minimum": 0.9912345273277318, "maximum": 0.9970870047076454, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8676275093974356, "standard_deviation": 0.0334982531482304, "minimum": 0.8216273565683636, "maximum": 0.9307891590743003, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6333003603696277, "standard_deviation": 0.3298023637496934, "minimum": 0.03979410393570784, "maximum": 0.9921857412636825, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07611374245401202, "standard_deviation": 0.012806029692148002, "minimum": 0.061549300411268916, "maximum": 0.10107957867929929, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9636976421797686, "standard_deviation": 0.002308726802728387, "minimum": 0.961000942728401, "maximum": 0.9669786060909069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6724550625228605, "standard_deviation": 0.038104950742619145, "minimum": 1.6121429384520782, "maximum": 1.73564036566679, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6040557735859449, "standard_deviation": 0.1931392901877824, "minimum": 0.21860468367045907, "maximum": 0.8070935208749876, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20419236076869435, "standard_deviation": 0.006330906857903245, "minimum": 0.19509125376692735, "maximum": 0.2113634809872488, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9855842426702479, "standard_deviation": 0.0019975720385614288, "minimum": 0.9828493083899436, "maximum": 0.9877382644700243, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0161099735167354, "standard_deviation": 0.020064825973948448, "minimum": 0.9866993499076192, "maximum": 1.0412538684434984, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6903820887097352, "standard_deviation": 0.26922030881607656, "minimum": 0.1891184625935828, "maximum": 1.0013940710696192, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1231840256278569, "standard_deviation": 0.008604812142596877, "minimum": 0.11422799177015537, "maximum": 0.13413480702985697, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9926529974219293, "standard_deviation": 0.0007858089263049383, "minimum": 0.9916353600587735, "maximum": 0.993981627492762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5022813438333267, "standard_deviation": 0.0319275445919926, "minimum": 1.4451774380682032, "maximum": 1.5411406563176826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5518745438237378, "standard_deviation": 0.22221739086616138, "minimum": 0.30741758979531125, "maximum": 0.9709585090256706, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09231776672159087, "standard_deviation": 0.005099541467868087, "minimum": 0.08404624510648159, "maximum": 0.0992311659720801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973490768348241, "standard_deviation": 0.00045227212604512766, "minimum": 0.9967121668710663, "maximum": 0.9980460194592091, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8710722129455093, "standard_deviation": 0.042827249605410365, "minimum": 0.8040304067807865, "maximum": 0.9242892741624521, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5345846340299463, "standard_deviation": 0.1729201589831301, "minimum": 0.35497665847787613, "maximum": 0.9026483954310054, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05441135317517054, "standard_deviation": 0.004901110799062971, "minimum": 0.04777903497555483, "maximum": 0.06184921919003151, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21014188447782325, "standard_deviation": 0.040512742050676086, "minimum": 0.13684616112129927, "maximum": 0.27572438832689766, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.618538890985683, "standard_deviation": 0.007435424803923091, "minimum": 5.609206532509325, "maximum": 5.632882081689101, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.23638350584042814, "standard_deviation": 0.005556918018607582, "minimum": 0.2240562729051889, "maximum": 0.2406649199023635, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08107209748344502, "standard_deviation": 0.0343007501003332, "minimum": -0.12660138497432136, "maximum": -0.01401772857733445, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949885343997695, "standard_deviation": 0.00019179018817312978, "minimum": 0.9947996917685886, "maximum": 0.9953234423427163, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9519854865648982, "standard_deviation": 0.02996075661286901, "minimum": 0.9033116950882891, "maximum": 0.9979230224317523, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35268109749711096, "standard_deviation": 0.10181479384157996, "minimum": 0.24928009267540183, "maximum": 0.5680170049381211, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07031308228977551, "standard_deviation": 0.0012904491546709607, "minimum": 0.06825951042649864, "maximum": 0.07207395342523243, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984923809811077, "standard_deviation": 0.0002490010196648953, "minimum": 0.9981214514250842, "maximum": 0.9988257956647456, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49190744371307016, "standard_deviation": 0.02535554891579173, "minimum": 0.45934807342895156, "maximum": 0.5428315624487597, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4824489138792921, "standard_deviation": 0.17655135141142764, "minimum": 0.20058429732975105, "maximum": 0.675228823122452, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03757750416092661, "standard_deviation": 0.0030305585571555205, "minimum": 0.033486790975189674, "maximum": 0.042036239248708834, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_237-134500-0029", "language": "en", "style": "normal", "speaker": "librispeech_237", "source_utterance_id": "237-134500-0029", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 33040, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_237-134500-0029.wav", "source_file_sha256": "6bfb755c356a9d66ceffbf0026a5e07880b0e14b92ce7eb099222b4c5c5f128c", "decoded_pcm_sha256": "8562d47701848db95c8610733017a69a5928db4edb524e0f205ab4182cf7b4f7", "original_source_sha256": "a71914689325868d93f40677ab17bdfbb8c8209801864e31a4f4303cf7fd0fa0", "corpus_entry_id": "librispeech_237-134500-0029"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 2.9236948490142822, "predicted_mos": 2.9236948490142822, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "533cd620dbfe671a4659bf680292208f5744c534963802a1e31ce0e69d1c8a5c", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.063360731001012, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_237-134500-0029", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "533cd620dbfe671a4659bf680292208f5744c534963802a1e31ce0e69d1c8a5c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "71d86eaff9e13fc4a8e0e41c271e36be5ba1a6cd179c3ba3a6931b89dbd78206"}, "rows": [{"case": "librispeech_237-134500-0029", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.4701690673828125, "predicted_mos": 1.4701690673828125, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "6dec62fe4c738d77da9e3dce3dd1e7dfc998b5de43e6c5574ea51ef28d74aba6", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.295469864002371, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_237-134500-0029", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6dec62fe4c738d77da9e3dce3dd1e7dfc998b5de43e6c5574ea51ef28d74aba6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.937387330303347, "全帯域_10ms_rms_envelope_nrmse": 0.08343144999421376, "全帯域_10ms_envelope_correlation": 0.9935761733146378, "全帯域_10ms_modulation_mae_db": 0.5889672161151785, "全帯域_40ms_log_rms_mae_db": 0.4744744315264625, "全帯域_40ms_rms_envelope_nrmse": 0.03511766606250344, "全帯域_40ms_envelope_correlation": 0.998866683501684, "全帯域_40ms_modulation_mae_db": 0.6407570034888036, "300–1000 Hz_10ms_log_rms_mae_db": 1.4045989000811383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12871728755380826, "300–1000 Hz_10ms_envelope_correlation": 0.9852280293257165, "300–1000 Hz_10ms_modulation_mae_db": 0.7077768127007215, "300–1000 Hz_40ms_log_rms_mae_db": 0.874270887660961, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06929437709669033, "300–1000 Hz_40ms_envelope_correlation": 0.9954609811004032, "300–1000 Hz_40ms_modulation_mae_db": 0.5956921544367401, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6872186855767073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11410786645071691, "1000–3000 Hz_10ms_envelope_correlation": 0.9909781022496471, "1000–3000 Hz_10ms_modulation_mae_db": 0.6040923042697369, "1000–3000 Hz_40ms_log_rms_mae_db": 0.905672178460243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05524204611322223, "1000–3000 Hz_40ms_envelope_correlation": 0.9987068407548351, "1000–3000 Hz_40ms_modulation_mae_db": 0.946975723137354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.990618011639307, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41065702712258206, "3000–6000 Hz_10ms_envelope_correlation": 0.8960657405064039, "3000–6000 Hz_10ms_modulation_mae_db": 4.693983740718466, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6222857451698358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33302063287095723, "3000–6000 Hz_40ms_envelope_correlation": 0.9273471152833557, "3000–6000 Hz_40ms_modulation_mae_db": 5.105524798566557, "mr_stft_db": 6.018994587996083, "band_mae_db": 0.8587035991249653, "rms_gain_error_db": -0.10264482442129333, "output_peak": 0.9611528957333749}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9955028981795185, "全帯域_10ms_rms_envelope_nrmse": 0.07615062496687068, "全帯域_10ms_envelope_correlation": 0.9946147673257355, "全帯域_10ms_modulation_mae_db": 0.3919322073856288, "全帯域_40ms_log_rms_mae_db": 0.5024332143636567, "全帯域_40ms_rms_envelope_nrmse": 0.03375289976935187, "全帯域_40ms_envelope_correlation": 0.9988674092286485, "全帯域_40ms_modulation_mae_db": 0.37343712266320295, "300–1000 Hz_10ms_log_rms_mae_db": 1.3654102619018929, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11215636442699481, "300–1000 Hz_10ms_envelope_correlation": 0.9887701455419585, "300–1000 Hz_10ms_modulation_mae_db": 0.8393095401482951, "300–1000 Hz_40ms_log_rms_mae_db": 0.8146950413651017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053435487854847516, "300–1000 Hz_40ms_envelope_correlation": 0.9973051152424077, "300–1000 Hz_40ms_modulation_mae_db": 0.47058213377927416, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6773610011519022, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12618471985348803, "1000–3000 Hz_10ms_envelope_correlation": 0.9879683749932353, "1000–3000 Hz_10ms_modulation_mae_db": 0.647380255253994, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8936591475301089, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05527075610163389, "1000–3000 Hz_40ms_envelope_correlation": 0.9976129758771128, "1000–3000 Hz_40ms_modulation_mae_db": 0.7560968099752493, "3000–6000 Hz_10ms_log_rms_mae_db": 1.943017731828683, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4082644519673068, "3000–6000 Hz_10ms_envelope_correlation": 0.8974386671481246, "3000–6000 Hz_10ms_modulation_mae_db": 4.5826936157520155, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5381587362898037, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32618164053397686, "3000–6000 Hz_40ms_envelope_correlation": 0.9305501515446307, "3000–6000 Hz_40ms_modulation_mae_db": 4.646094931499046, "mr_stft_db": 6.030056600608584, "band_mae_db": 0.8468460657292134, "rms_gain_error_db": -0.05759633038262996, "output_peak": 0.9936790666502452}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.981078483012451, "全帯域_10ms_rms_envelope_nrmse": 0.09274359055922933, "全帯域_10ms_envelope_correlation": 0.9921626454786406, "全帯域_10ms_modulation_mae_db": 0.6412327296738367, "全帯域_40ms_log_rms_mae_db": 0.5463432261192893, "全帯域_40ms_rms_envelope_nrmse": 0.05038057543573619, "全帯域_40ms_envelope_correlation": 0.9976834016270163, "全帯域_40ms_modulation_mae_db": 0.527338831166048, "300–1000 Hz_10ms_log_rms_mae_db": 1.3498037862721104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11622005744684237, "300–1000 Hz_10ms_envelope_correlation": 0.987984216223485, "300–1000 Hz_10ms_modulation_mae_db": 0.5573766423648402, "300–1000 Hz_40ms_log_rms_mae_db": 0.8198450745125865, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056373981748615035, "300–1000 Hz_40ms_envelope_correlation": 0.9971327560071289, "300–1000 Hz_40ms_modulation_mae_db": 0.3955942380854432, "1000–3000 Hz_10ms_log_rms_mae_db": 1.753782856598585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13537670897507256, "1000–3000 Hz_10ms_envelope_correlation": 0.9866514186754386, "1000–3000 Hz_10ms_modulation_mae_db": 1.167571226507169, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9890538354419449, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07810856088351423, "1000–3000 Hz_40ms_envelope_correlation": 0.9959163214019109, "1000–3000 Hz_40ms_modulation_mae_db": 0.9337555487146139, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8901587593078621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4074071961890832, "3000–6000 Hz_10ms_envelope_correlation": 0.9076862658083176, "3000–6000 Hz_10ms_modulation_mae_db": 4.721099314546202, "3000–6000 Hz_40ms_log_rms_mae_db": 1.498530076228854, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32547598611648515, "3000–6000 Hz_40ms_envelope_correlation": 0.9396341251987734, "3000–6000 Hz_40ms_modulation_mae_db": 5.071660072965408, "mr_stft_db": 6.01659070761347, "band_mae_db": 0.9009333237816413, "rms_gain_error_db": -0.15029868415843803, "output_peak": 0.9458751026182252}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9333325292728647, "全帯域_10ms_rms_envelope_nrmse": 0.08630043630424715, "全帯域_10ms_envelope_correlation": 0.9932347799887051, "全帯域_10ms_modulation_mae_db": 0.23696709788317274, "全帯域_40ms_log_rms_mae_db": 0.5117592092609146, "全帯域_40ms_rms_envelope_nrmse": 0.04821673991728104, "全帯域_40ms_envelope_correlation": 0.9978835839633013, "全帯域_40ms_modulation_mae_db": 0.44125956296186475, "300–1000 Hz_10ms_log_rms_mae_db": 1.4526369094013838, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11776421108819579, "300–1000 Hz_10ms_envelope_correlation": 0.9876610874471824, "300–1000 Hz_10ms_modulation_mae_db": 0.506179010983718, "300–1000 Hz_40ms_log_rms_mae_db": 0.87383061566176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05702885679860787, "300–1000 Hz_40ms_envelope_correlation": 0.9970180908962946, "300–1000 Hz_40ms_modulation_mae_db": 0.5126841964743969, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6984089503905855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13037729678667698, "1000–3000 Hz_10ms_envelope_correlation": 0.9879534426192248, "1000–3000 Hz_10ms_modulation_mae_db": 0.43926075459229913, "1000–3000 Hz_40ms_log_rms_mae_db": 0.995243204526823, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0836043056706133, "1000–3000 Hz_40ms_envelope_correlation": 0.9954667936141885, "1000–3000 Hz_40ms_modulation_mae_db": 0.543449798562456, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9484923080460168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4178882572007082, "3000–6000 Hz_10ms_envelope_correlation": 0.9017934691162779, "3000–6000 Hz_10ms_modulation_mae_db": 4.819339521118826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5820382123088725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3370147330045906, "3000–6000 Hz_40ms_envelope_correlation": 0.9342195977639651, "3000–6000 Hz_40ms_modulation_mae_db": 5.1623487653338405, "mr_stft_db": 6.002067463054402, "band_mae_db": 0.933594730013494, "rms_gain_error_db": -0.14777446463431712, "output_peak": 0.9292395159123369}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.010093744143292, "全帯域_10ms_rms_envelope_nrmse": 0.08130203773876936, "全帯域_10ms_envelope_correlation": 0.9938955178512733, "全帯域_10ms_modulation_mae_db": 0.4134468089042853, "全帯域_40ms_log_rms_mae_db": 0.5334329850277048, "全帯域_40ms_rms_envelope_nrmse": 0.03817237431429929, "全帯域_40ms_envelope_correlation": 0.9986104631482315, "全帯域_40ms_modulation_mae_db": 0.4165075875939805, "300–1000 Hz_10ms_log_rms_mae_db": 1.4834173596508848, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11655334107320228, "300–1000 Hz_10ms_envelope_correlation": 0.9880250561354006, "300–1000 Hz_10ms_modulation_mae_db": 0.48576867209496744, "300–1000 Hz_40ms_log_rms_mae_db": 0.9302945622088903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06256535977711973, "300–1000 Hz_40ms_envelope_correlation": 0.9965468961167991, "300–1000 Hz_40ms_modulation_mae_db": 0.46836519138613414, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7181153549715116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12345360973323571, "1000–3000 Hz_10ms_envelope_correlation": 0.9885343152240764, "1000–3000 Hz_10ms_modulation_mae_db": 0.385257304127473, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8925297400674981, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05993242003360923, "1000–3000 Hz_40ms_envelope_correlation": 0.9972961959706503, "1000–3000 Hz_40ms_modulation_mae_db": 0.6443135762369296, "3000–6000 Hz_10ms_log_rms_mae_db": 1.950370178661535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4105016357191796, "3000–6000 Hz_10ms_envelope_correlation": 0.9013004184183813, "3000–6000 Hz_10ms_modulation_mae_db": 4.568771811473389, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5356743263115191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3280735753463951, "3000–6000 Hz_40ms_envelope_correlation": 0.9342944508528565, "3000–6000 Hz_40ms_modulation_mae_db": 4.7813821412115844, "mr_stft_db": 6.01046992002103, "band_mae_db": 0.8671611727138723, "rms_gain_error_db": -0.0953571187338033, "output_peak": 0.9322980098946192}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9867621072802428, "全帯域_10ms_rms_envelope_nrmse": 0.08519013003862103, "全帯域_10ms_envelope_correlation": 0.9933807710081113, "全帯域_10ms_modulation_mae_db": 0.48393035382677424, "全帯域_40ms_log_rms_mae_db": 0.5110360825205225, "全帯域_40ms_rms_envelope_nrmse": 0.04221170988291334, "全帯域_40ms_envelope_correlation": 0.9984073957582162, "全帯域_40ms_modulation_mae_db": 0.4823129432936939, "300–1000 Hz_10ms_log_rms_mae_db": 1.428242808854869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1265153890768103, "300–1000 Hz_10ms_envelope_correlation": 0.9857465957992421, "300–1000 Hz_10ms_modulation_mae_db": 0.23901431179417862, "300–1000 Hz_40ms_log_rms_mae_db": 0.8722516121087753, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06099605923959467, "300–1000 Hz_40ms_envelope_correlation": 0.9966707716962282, "300–1000 Hz_40ms_modulation_mae_db": 0.6035152136888015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5969726016546117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11378209588002086, "1000–3000 Hz_10ms_envelope_correlation": 0.9906286090453609, "1000–3000 Hz_10ms_modulation_mae_db": 0.6786981437297929, "1000–3000 Hz_40ms_log_rms_mae_db": 0.927918172682159, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06012792232562874, "1000–3000 Hz_40ms_envelope_correlation": 0.997731632265104, "1000–3000 Hz_40ms_modulation_mae_db": 0.5416079684998725, "3000–6000 Hz_10ms_log_rms_mae_db": 1.924611360806887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3983033424641557, "3000–6000 Hz_10ms_envelope_correlation": 0.9085085762533064, "3000–6000 Hz_10ms_modulation_mae_db": 4.62959388089858, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5932512142414026, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31866531852141006, "3000–6000 Hz_40ms_envelope_correlation": 0.9393625113896077, "3000–6000 Hz_40ms_modulation_mae_db": 4.860361478796271, "mr_stft_db": 6.002499965541557, "band_mae_db": 0.8716062183193051, "rms_gain_error_db": -0.14024371677455574, "output_peak": 0.9441228654922749}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9895330659116138, "全帯域_10ms_rms_envelope_nrmse": 0.07465872743702598, "全帯域_10ms_envelope_correlation": 0.994998096736336, "全帯域_10ms_modulation_mae_db": 0.43243940285539284, "全帯域_40ms_log_rms_mae_db": 0.5362381993365282, "全帯域_40ms_rms_envelope_nrmse": 0.045874083279989035, "全帯域_40ms_envelope_correlation": 0.9980729837457523, "全帯域_40ms_modulation_mae_db": 0.3183895892072941, "300–1000 Hz_10ms_log_rms_mae_db": 1.3581645745947868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10748142155180687, "300–1000 Hz_10ms_envelope_correlation": 0.9900315811132928, "300–1000 Hz_10ms_modulation_mae_db": 0.35685078811623705, "300–1000 Hz_40ms_log_rms_mae_db": 0.7459324746143999, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0626318489816746, "300–1000 Hz_40ms_envelope_correlation": 0.9967214831517447, "300–1000 Hz_40ms_modulation_mae_db": 0.44087348430401274, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6410588447321686, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.111976634076666, "1000–3000 Hz_10ms_envelope_correlation": 0.9908411818531767, "1000–3000 Hz_10ms_modulation_mae_db": 0.4128507399726509, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8695914538133872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06278383847118148, "1000–3000 Hz_40ms_envelope_correlation": 0.9972808234584333, "1000–3000 Hz_40ms_modulation_mae_db": 0.4130986497703278, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8339028201669803, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41800457955397324, "3000–6000 Hz_10ms_envelope_correlation": 0.8990782327120208, "3000–6000 Hz_10ms_modulation_mae_db": 4.753595054471565, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4733415336730937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32969359123713954, "3000–6000 Hz_40ms_envelope_correlation": 0.9349228123636812, "3000–6000 Hz_40ms_modulation_mae_db": 5.176215899346343, "mr_stft_db": 6.011485619038642, "band_mae_db": 0.8806095156603889, "rms_gain_error_db": -0.13577649402287034, "output_peak": 0.9177653472364745}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9621999322935938, "全帯域_10ms_rms_envelope_nrmse": 0.07257883306447988, "全帯域_10ms_envelope_correlation": 0.9954528968505013, "全帯域_10ms_modulation_mae_db": 0.2732061214426435, "全帯域_40ms_log_rms_mae_db": 0.4961198186088187, "全帯域_40ms_rms_envelope_nrmse": 0.03838621996871954, "全帯域_40ms_envelope_correlation": 0.9989632866571266, "全帯域_40ms_modulation_mae_db": 0.3410278309387183, "300–1000 Hz_10ms_log_rms_mae_db": 1.3242931810987866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10859299728868327, "300–1000 Hz_10ms_envelope_correlation": 0.9898323598978285, "300–1000 Hz_10ms_modulation_mae_db": 0.3499843526227931, "300–1000 Hz_40ms_log_rms_mae_db": 0.7972136182344648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05758507763600568, "300–1000 Hz_40ms_envelope_correlation": 0.9973797234673237, "300–1000 Hz_40ms_modulation_mae_db": 0.21624259094608853, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6341695275487544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1083692380413878, "1000–3000 Hz_10ms_envelope_correlation": 0.9916190487985395, "1000–3000 Hz_10ms_modulation_mae_db": 0.5794412490408248, "1000–3000 Hz_40ms_log_rms_mae_db": 0.954634834760971, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05952545386428209, "1000–3000 Hz_40ms_envelope_correlation": 0.9978476957616489, "1000–3000 Hz_40ms_modulation_mae_db": 0.45180600363863427, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8926117841634325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40562309736560614, "3000–6000 Hz_10ms_envelope_correlation": 0.9047713194093606, "3000–6000 Hz_10ms_modulation_mae_db": 4.503245865994216, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4784606405591143, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3348456129127278, "3000–6000 Hz_40ms_envelope_correlation": 0.9322223436400257, "3000–6000 Hz_40ms_modulation_mae_db": 4.4675858622938085, "mr_stft_db": 6.020844583653253, "band_mae_db": 0.8929915654827564, "rms_gain_error_db": -0.19364881704190845, "output_peak": 0.9607179653810083}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 33040, "output_samples": 33040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8707191182385055, "spectral_convergence": 0.3105939557773417, "spectral_convergence_status": "ok", "amplitude_floor": 1.5756171103077008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.682224063291211, "spectral_convergence": 0.33709000727987154, "spectral_convergence_status": "ok", "amplitude_floor": 1.2505050051947005e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.504040582458532, "spectral_convergence": 0.35630991281052804, "spectral_convergence_status": "ok", "amplitude_floor": 9.121352204658788e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021685618865241799, "output_power": 0.00021892284794753852, "delta_db": 0.041192686262728234, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010795690773652512, "output_power": 0.0011144138611271827, "delta_db": 0.13796068873314446, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002989928450689294, "output_power": 0.0029701727598701027, "delta_db": -0.028790849779088926, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0020580355715967367, "output_power": 0.001889199782500513, "delta_db": -0.3717499003943609, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0009751360590152492, "output_power": 0.0009222282629313754, "delta_db": -0.24226788641202407, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.9392123549665832e-05, "output_power": 5.255951242189713e-05, "delta_db": 4.330259583168445, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8587035991249653, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10264482442129333, "pitch": null, "output_rms_dbfs": -21.44632478534282, "output_peak": 0.9611528957333749, "reference_rms_dbfs": -21.343679960921527, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.018994587996083}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 33040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00032988134947798775, "log_rms_mae_db": 0.937387330303347, "rms_envelope_nrmse": 0.08343144999421376, "envelope_correlation": 0.9935761733146378, "rms_gain_error_db": -0.10264403193294797, "reference_mean_power": 0.007356724541990913, "output_mean_power": 0.007184889476260956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 2.980285612038863e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020727352824950612, "output_power": 0.0019722425648863247, "reference_active": true, "delta_db": -0.21583512611613728}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00028987718896809884, "output_power": 0.00023593548190642083, "reference_active": true, "delta_db": -0.894207826224924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.012977878291207e-05, "output_power": 6.028601806708838e-05, "reference_active": true, "delta_db": -0.6568586960044744}], "active_band_count": 3, "active_band_mae_db": 0.5889672161151785}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0002960474071300256, "log_rms_mae_db": 0.4744744315264625, "rms_envelope_nrmse": 0.03511766606250344, "envelope_correlation": 0.998866683501684, "rms_gain_error_db": -0.10264577909758635, "reference_mean_power": 0.007465415825536721, "output_mean_power": 0.007291039064469566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 2.5597102021075326e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001877761726334087, "output_power": 0.0017922025151566952, "reference_active": true, "delta_db": -0.20253400291015608}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012702101262043316, "output_power": 0.00010440340872536486, "reference_active": true, "delta_db": -0.8516089222146437}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.962314408759724e-06, "output_power": 3.2444129425041963e-06, "reference_active": true, "delta_db": -0.8681280853416111}], "active_band_count": 3, "active_band_mae_db": 0.6407570034888036}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00021933697126022545, "log_rms_mae_db": 1.4045989000811383, "rms_envelope_nrmse": 0.12871728755380826, "envelope_correlation": 0.9852280293257165, "rms_gain_error_db": -0.09933884508056506, "reference_mean_power": 0.003363603458339666, "output_mean_power": 0.0032875389415050345}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.3609823101859838e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008546335077009212, "output_power": 0.0007549103053232756, "reference_active": true, "delta_db": -0.5388456225143655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016660612815788933, "output_power": 0.00013325285491635436, "reference_active": true, "delta_db": -0.9701444924525722}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.188726605138031e-05, "output_power": 4.8252080927686825e-05, "reference_active": true, "delta_db": 0.6143403231352267}], "active_band_count": 3, "active_band_mae_db": 0.7077768127007215}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0001893291578274055, "log_rms_mae_db": 0.874270887660961, "rms_envelope_nrmse": 0.06929437709669033, "envelope_correlation": 0.9954609811004032, "rms_gain_error_db": -0.09934077899394234, "reference_mean_power": 0.0034133051942474998, "output_mean_power": 0.003336115236958516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.15224102535942e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007711766819603955, "output_power": 0.0006896218736420404, "reference_active": true, "delta_db": -0.4854286122461271}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.304889773698623e-05, "output_power": 6.176269422728964e-05, "reference_active": true, "delta_db": -0.7288743444212691}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527118205585199e-06, "output_power": 2.2148698217439324e-06, "reference_active": true, "delta_db": -0.5727735066428239}], "active_band_count": 3, "active_band_mae_db": 0.5956921544367401}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.0002256735995145706, "log_rms_mae_db": 1.6872186855767073, "rms_envelope_nrmse": 0.11410786645071691, "envelope_correlation": 0.9909781022496471, "rms_gain_error_db": -0.3458858723257482, "reference_mean_power": 0.002711357875845494, "output_mean_power": 0.0025037920564744674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.6613986227009806e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012279896352023905, "output_power": 0.0011349391107157393, "reference_active": true, "delta_db": -0.34222138843999544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888327196960883, "output_power": 9.301041294299469e-05, "reference_active": true, "delta_db": -1.0658917693013212}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.234146057131384e-05, "output_power": 2.9467497384832876e-05, "reference_active": true, "delta_db": -0.40416375506789404}], "active_band_count": 3, "active_band_mae_db": 0.6040923042697369}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.00020374269470838238, "log_rms_mae_db": 0.905672178460243, "rms_envelope_nrmse": 0.05524204611322223, "envelope_correlation": 0.9987068407548351, "rms_gain_error_db": -0.345889121347841, "reference_mean_power": 0.002751412553659036, "output_mean_power": 0.0025407784798793767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.4886745667367597e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001137287164515336, "output_power": 0.00105333681681719, "reference_active": true, "delta_db": -0.33302873548874845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.907501205097428e-05, "output_power": 3.967023683344157e-05, "reference_active": true, "delta_db": -0.9239562167518168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5268858664777347e-06, "output_power": 1.7546536517708468e-06, "reference_active": true, "delta_db": -1.5839422171714967}], "active_band_count": 3, "active_band_mae_db": 0.946975723137354}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 4.718157231228245e-05, "log_rms_mae_db": 1.990618011639307, "rms_envelope_nrmse": 0.41065702712258206, "envelope_correlation": 0.8960657405064039, "rms_gain_error_db": 0.7490846307880211, "reference_mean_power": 9.509723574878595e-05, "output_mean_power": 0.00011299945692966872}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 5.720070068605411e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.50238140104046e-05, "output_power": 4.756860230475429e-05, "reference_active": true, "delta_db": 1.3295695288795093}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.975419310826897e-06, "output_power": 8.621986896256051e-06, "reference_active": true, "delta_db": 2.387776710407685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.471999704474668e-06, "output_power": 1.60091480563772e-05, "reference_active": true, "delta_db": 10.364604982868205}], "active_band_count": 3, "active_band_mae_db": 4.693983740718466}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 4.015479902310721e-05, "log_rms_mae_db": 1.6222857451698358, "rms_envelope_nrmse": 0.33302063287095723, "envelope_correlation": 0.9273471152833557, "rms_gain_error_db": 0.7491159640951822, "reference_mean_power": 9.64997641207967e-05, "output_mean_power": 0.00011466684101488949}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 5.087796481772889e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1685652820939687e-05, "output_power": 5.540329504841018e-05, "reference_active": true, "delta_db": 2.426729353281259}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.336227531732081e-06, "output_power": 4.876895271879687e-06, "reference_active": true, "delta_db": 3.1962829184895094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1955736150290575e-07, "output_power": 1.1141214468216786e-06, "reference_active": true, "delta_db": 9.693562123928901}], "active_band_count": 3, "active_band_mae_db": 5.105524798566557}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8410185210022357, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "57641b26881e9133e82cb5d1054cc22438a39a967ac526ecae05fc053c0616cc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9893968116823374, "standard_deviation": 0.0018301903149600806, "minimum": 0.9866514186754386, "maximum": 0.9916190487985395, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6758859778281034, "standard_deviation": 0.050241767812626334, "minimum": 1.5969726016546117, "maximum": 1.753782856598585, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6143189971867427, "standard_deviation": 0.24967146964952872, "minimum": 0.385257304127473, "maximum": 1.167571226507169, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1204535212246581, "standard_deviation": 0.009751870071458226, "minimum": 0.1083692380413878, "maximum": 0.13537670897507256, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972324098879854, "standard_deviation": 0.001055765044814477, "minimum": 0.9954667936141885, "maximum": 0.9987068407548351, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9285378209103919, "standard_deviation": 0.04675750490622896, "minimum": 0.8695914538133872, "maximum": 0.995243204526823, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6538880098169297, "standard_deviation": 0.20646155468899138, "minimum": 0.4130986497703278, "maximum": 0.946975723137354, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06432441293296065, "standard_deviation": 0.01061318991111918, "minimum": 0.05524204611322223, "maximum": 0.0836043056706133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9020803361715242, "standard_deviation": 0.0045890202869195325, "minimum": 0.8960657405064039, "maximum": 0.9085085762533064, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.921722869327588, "standard_deviation": 0.04825784277168509, "minimum": 1.8339028201669803, "maximum": 1.990618011639307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 4.659040350621657, "standard_deviation": 0.10611652671996939, "minimum": 4.503245865994216, "maximum": 4.819339521118826, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.40958119844782437, "standard_deviation": 0.006450713065474849, "minimum": 0.3983033424641557, "maximum": 0.41800457955397324, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.934069138504612, "standard_deviation": 0.004153665382869622, "minimum": 0.9273471152833557, "maximum": 0.9396341251987734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.540217560597812, "standard_deviation": 0.055221331910127396, "minimum": 1.4733415336730937, "maximum": 1.6222857451698358, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.908896743751607, "standard_deviation": 0.2627567272186491, "minimum": 4.4675858622938085, "maximum": 5.176215899346343, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3291213863179603, "standard_deviation": 0.005891416803594431, "minimum": 0.31866531852141006, "maximum": 0.3370147330045906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9879098839355133, "standard_deviation": 0.001727525567199217, "minimum": 0.9852280293257165, "maximum": 0.9900315811132928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3958209727319817, "standard_deviation": 0.05555695559700963, "minimum": 1.3242931810987866, "maximum": 1.4834173596508848, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5052825163532189, "standard_deviation": 0.1973834697191135, "minimum": 0.23901431179417862, "maximum": 0.8393095401482951, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11675013368829298, "standard_deviation": 0.007685530565465326, "minimum": 0.10748142155180687, "maximum": 0.12871728755380826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967794772097913, "standard_deviation": 0.0006124936694467468, "minimum": 0.9954609811004032, "maximum": 0.9973797234673237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8410417357958675, "standard_deviation": 0.05760646390710061, "minimum": 0.7459324746143999, "maximum": 0.9302945622088903, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.46294365038761137, "standard_deviation": 0.12295692596220835, "minimum": 0.21624259094608853, "maximum": 0.6035152136888015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059988881141644426, "standard_deviation": 0.004953853629118446, "minimum": 0.053435487854847516, "maximum": 0.06929437709669033, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.8815557738532045, "standard_deviation": 0.027381951979331295, "minimum": 0.8468460657292134, "maximum": 0.933594730013494, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.0141261809408775, "standard_deviation": 0.009483917646058562, "minimum": 6.002067463054402, "maximum": 6.030056600608584, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9481063461148198, "standard_deviation": 0.02376498567446283, "minimum": 0.9177653472364745, "maximum": 0.9936790666502452, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12791755627122703, "standard_deviation": 0.04153049714426293, "minimum": -0.19364881704190845, "maximum": -0.05759633038262996, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939144560692426, "standard_deviation": 0.0010664980430781588, "minimum": 0.9921626454786406, "maximum": 0.9954528968505013, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9744862612996155, "standard_deviation": 0.02764847242519416, "minimum": 0.9333325292728647, "maximum": 1.010093744143292, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4327652422608641, "standard_deviation": 0.1395463516130739, "minimum": 0.23696709788317274, "maximum": 0.6412327296738367, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08154447876293214, "standard_deviation": 0.00678063925924654, "minimum": 0.07257883306447988, "maximum": 0.09274359055922933, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998419400953747, "standard_deviation": 0.0004901943310688208, "minimum": 0.9976834016270163, "maximum": 0.9989632866571266, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5139796458454872, "standard_deviation": 0.02374816476501854, "minimum": 0.4744744315264625, "maximum": 0.5463432261192893, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44262880891420076, "standard_deviation": 0.10639228274893126, "minimum": 0.3183895892072941, "maximum": 0.6407570034888036, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04151403357884921, "standard_deviation": 0.0061544089006406715, "minimum": 0.03375289976935187, "maximum": 0.05038057543573619, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0029", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.608553171157837, "predicted_mos": 1.608553171157837, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "24b82bacdbdc1a38c32ed549cfca670784fb0881b35ee263d1f90c0476e477c6", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.930225200019777, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9272420871342977, "全帯域_10ms_rms_envelope_nrmse": 0.07946948016240427, "全帯域_10ms_envelope_correlation": 0.9942263921613126, "全帯域_10ms_modulation_mae_db": 0.6499183908932195, "全帯域_40ms_log_rms_mae_db": 0.48665426550696333, "全帯域_40ms_rms_envelope_nrmse": 0.03332442398885131, "全帯域_40ms_envelope_correlation": 0.9990532859648944, "全帯域_40ms_modulation_mae_db": 0.6560712295622848, "300–1000 Hz_10ms_log_rms_mae_db": 1.3467554817218401, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12226297255241018, "300–1000 Hz_10ms_envelope_correlation": 0.9866316854390285, "300–1000 Hz_10ms_modulation_mae_db": 0.585905000584169, "300–1000 Hz_40ms_log_rms_mae_db": 0.7621514536032734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0625565406694916, "300–1000 Hz_40ms_envelope_correlation": 0.9962780751178785, "300–1000 Hz_40ms_modulation_mae_db": 0.6373513304883127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6980155518160185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12022260424047229, "1000–3000 Hz_10ms_envelope_correlation": 0.9900105566200781, "1000–3000 Hz_10ms_modulation_mae_db": 0.6014031658017545, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8882525516195505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058340176981591586, "1000–3000 Hz_40ms_envelope_correlation": 0.9986144900384936, "1000–3000 Hz_40ms_modulation_mae_db": 0.913005270358668, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9681420092984039, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40384188444731434, "3000–6000 Hz_10ms_envelope_correlation": 0.8998538815396165, "3000–6000 Hz_10ms_modulation_mae_db": 4.627263663277197, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6035358699270268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32435437273737294, "3000–6000 Hz_40ms_envelope_correlation": 0.9313686761948712, "3000–6000 Hz_40ms_modulation_mae_db": 5.0875145440747005, "mr_stft_db": 5.966291881721211, "band_mae_db": 0.8671489710823045, "rms_gain_error_db": -0.12251929271490493, "output_peak": 0.9586073318389958}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0366518551984163, "全帯域_10ms_rms_envelope_nrmse": 0.07777821193294952, "全帯域_10ms_envelope_correlation": 0.9943758151179741, "全帯域_10ms_modulation_mae_db": 0.4399848614033934, "全帯域_40ms_log_rms_mae_db": 0.5196486429792156, "全帯域_40ms_rms_envelope_nrmse": 0.03512572685011461, "全帯域_40ms_envelope_correlation": 0.9987682243974254, "全帯域_40ms_modulation_mae_db": 0.42129410339739465, "300–1000 Hz_10ms_log_rms_mae_db": 1.2832386869875965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10637530390865488, "300–1000 Hz_10ms_envelope_correlation": 0.9899085917785957, "300–1000 Hz_10ms_modulation_mae_db": 0.4861164901208677, "300–1000 Hz_40ms_log_rms_mae_db": 0.7396025960265418, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0517153518687788, "300–1000 Hz_40ms_envelope_correlation": 0.9974642914225975, "300–1000 Hz_40ms_modulation_mae_db": 0.23146421061962266, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7480834411502826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13423434669404374, "1000–3000 Hz_10ms_envelope_correlation": 0.986358061205878, "1000–3000 Hz_10ms_modulation_mae_db": 0.5993413464584474, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9259446400994374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061986788502319, "1000–3000 Hz_40ms_envelope_correlation": 0.9969849340949926, "1000–3000 Hz_40ms_modulation_mae_db": 0.6581028484237188, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9258653385181996, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39770987931498136, "3000–6000 Hz_10ms_envelope_correlation": 0.9013361448163824, "3000–6000 Hz_10ms_modulation_mae_db": 4.36876075933882, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5838131209768433, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31286655472919933, "3000–6000 Hz_40ms_envelope_correlation": 0.9349828065562305, "3000–6000 Hz_40ms_modulation_mae_db": 4.432017681665695, "mr_stft_db": 5.966359221303871, "band_mae_db": 0.8366622788413428, "rms_gain_error_db": -0.056701683244253794, "output_peak": 0.9881262081147384}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.96206380763377, "全帯域_10ms_rms_envelope_nrmse": 0.08983939035058267, "全帯域_10ms_envelope_correlation": 0.9927607298268777, "全帯域_10ms_modulation_mae_db": 0.6168549537997605, "全帯域_40ms_log_rms_mae_db": 0.51491344006994, "全帯域_40ms_rms_envelope_nrmse": 0.04760127801391469, "全帯域_40ms_envelope_correlation": 0.9980917264527923, "全帯域_40ms_modulation_mae_db": 0.637921927693084, "300–1000 Hz_10ms_log_rms_mae_db": 1.3438325414703964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11726252746648853, "300–1000 Hz_10ms_envelope_correlation": 0.987813443050133, "300–1000 Hz_10ms_modulation_mae_db": 0.6546947512356537, "300–1000 Hz_40ms_log_rms_mae_db": 0.793477497658219, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05473752781771798, "300–1000 Hz_40ms_envelope_correlation": 0.9973997703452434, "300–1000 Hz_40ms_modulation_mae_db": 0.38729204065719863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7981090175117065, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12988738339547587, "1000–3000 Hz_10ms_envelope_correlation": 0.9880247859587316, "1000–3000 Hz_10ms_modulation_mae_db": 0.9567936783247669, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0083591138436512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0729206244115573, "1000–3000 Hz_40ms_envelope_correlation": 0.9968712881097463, "1000–3000 Hz_40ms_modulation_mae_db": 1.2343264236386635, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8376692343523175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39540633659144864, "3000–6000 Hz_10ms_envelope_correlation": 0.9119696942882051, "3000–6000 Hz_10ms_modulation_mae_db": 4.603455325360443, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4450515987987367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31678954217550465, "3000–6000 Hz_40ms_envelope_correlation": 0.9417188766182792, "3000–6000 Hz_40ms_modulation_mae_db": 4.892294080210436, "mr_stft_db": 5.95924550805782, "band_mae_db": 0.901963073217662, "rms_gain_error_db": -0.17856677233595047, "output_peak": 0.9449830677358153}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9549344306744678, "全帯域_10ms_rms_envelope_nrmse": 0.08889222525693458, "全帯域_10ms_envelope_correlation": 0.9929028855042107, "全帯域_10ms_modulation_mae_db": 0.34497579781067, "全帯域_40ms_log_rms_mae_db": 0.5405694682767539, "全帯域_40ms_rms_envelope_nrmse": 0.05033174256594054, "全帯域_40ms_envelope_correlation": 0.9977963822185361, "全帯域_40ms_modulation_mae_db": 0.656135309336821, "300–1000 Hz_10ms_log_rms_mae_db": 1.4310530100621839, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11559414642755898, "300–1000 Hz_10ms_envelope_correlation": 0.9881325720951365, "300–1000 Hz_10ms_modulation_mae_db": 0.39683564796429255, "300–1000 Hz_40ms_log_rms_mae_db": 0.8180185923033396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057672702042850084, "300–1000 Hz_40ms_envelope_correlation": 0.996979125810107, "300–1000 Hz_40ms_modulation_mae_db": 0.6589390386514585, "1000–3000 Hz_10ms_log_rms_mae_db": 1.753960242759907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13400574852409486, "1000–3000 Hz_10ms_envelope_correlation": 0.9875523283953391, "1000–3000 Hz_10ms_modulation_mae_db": 0.6543523853636913, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0161046580580464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08377680743881527, "1000–3000 Hz_40ms_envelope_correlation": 0.9958954714201798, "1000–3000 Hz_40ms_modulation_mae_db": 1.0115704308683162, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9347270623710457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4022196511392601, "3000–6000 Hz_10ms_envelope_correlation": 0.9065789881108808, "3000–6000 Hz_10ms_modulation_mae_db": 4.479264559029871, "3000–6000 Hz_40ms_log_rms_mae_db": 1.569954385962437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3214489244143515, "3000–6000 Hz_40ms_envelope_correlation": 0.9381613755489403, "3000–6000 Hz_40ms_modulation_mae_db": 4.798262164250946, "mr_stft_db": 5.954252678319438, "band_mae_db": 0.9036531097338454, "rms_gain_error_db": -0.1801622136235456, "output_peak": 0.9247684587649024}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0051606964397102, "全帯域_10ms_rms_envelope_nrmse": 0.08144039192058364, "全帯域_10ms_envelope_correlation": 0.9938369032182173, "全帯域_10ms_modulation_mae_db": 0.5240590135082136, "全帯域_40ms_log_rms_mae_db": 0.5345853459699407, "全帯域_40ms_rms_envelope_nrmse": 0.037834754626612205, "全帯域_40ms_envelope_correlation": 0.9985790406137414, "全帯域_40ms_modulation_mae_db": 0.5311375421964156, "300–1000 Hz_10ms_log_rms_mae_db": 1.432232410454782, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1180052901281614, "300–1000 Hz_10ms_envelope_correlation": 0.9875593334619455, "300–1000 Hz_10ms_modulation_mae_db": 0.3140643474104157, "300–1000 Hz_40ms_log_rms_mae_db": 0.8988433450063668, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06467928912522682, "300–1000 Hz_40ms_envelope_correlation": 0.9960298570955838, "300–1000 Hz_40ms_modulation_mae_db": 0.508186212391038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7284444246425028, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12627345314720548, "1000–3000 Hz_10ms_envelope_correlation": 0.9879937332575356, "1000–3000 Hz_10ms_modulation_mae_db": 0.538736554245982, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8822513280867061, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059433522185190206, "1000–3000 Hz_40ms_envelope_correlation": 0.9973453599038002, "1000–3000 Hz_40ms_modulation_mae_db": 0.9479540750701609, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9235314874166405, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3979490629799711, "3000–6000 Hz_10ms_envelope_correlation": 0.9048897328328047, "3000–6000 Hz_10ms_modulation_mae_db": 4.349627526784114, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5530990175394042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31617205447880387, "3000–6000 Hz_40ms_envelope_correlation": 0.9369257315969487, "3000–6000 Hz_40ms_modulation_mae_db": 4.5708709245311505, "mr_stft_db": 5.968141661688772, "band_mae_db": 0.8516506788948371, "rms_gain_error_db": -0.06685367586799293, "output_peak": 0.9271646035530848}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9936037567234617, "全帯域_10ms_rms_envelope_nrmse": 0.08523871246849209, "全帯域_10ms_envelope_correlation": 0.9934480223391644, "全帯域_10ms_modulation_mae_db": 0.506339037390314, "全帯域_40ms_log_rms_mae_db": 0.5285792760566436, "全帯域_40ms_rms_envelope_nrmse": 0.045195701693525166, "全帯域_40ms_envelope_correlation": 0.9982345483083399, "全帯域_40ms_modulation_mae_db": 0.5260428823031359, "300–1000 Hz_10ms_log_rms_mae_db": 1.4064622040760275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12308187446126599, "300–1000 Hz_10ms_envelope_correlation": 0.9866186074051713, "300–1000 Hz_10ms_modulation_mae_db": 0.2049647219851021, "300–1000 Hz_40ms_log_rms_mae_db": 0.8243400961215098, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06752030152369101, "300–1000 Hz_40ms_envelope_correlation": 0.9959443970798498, "300–1000 Hz_40ms_modulation_mae_db": 0.6774246447235205, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6437277107919257, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11398559880859586, "1000–3000 Hz_10ms_envelope_correlation": 0.9907144008221069, "1000–3000 Hz_10ms_modulation_mae_db": 0.6485112683844513, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9129834689272549, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059442614365810754, "1000–3000 Hz_40ms_envelope_correlation": 0.9979541636140715, "1000–3000 Hz_40ms_modulation_mae_db": 0.6509469469124334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8752266595299218, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3905609227259569, "3000–6000 Hz_10ms_envelope_correlation": 0.9095224385477562, "3000–6000 Hz_10ms_modulation_mae_db": 4.414957765589116, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5148419179846142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3102487282131396, "3000–6000 Hz_40ms_envelope_correlation": 0.9402893946908877, "3000–6000 Hz_40ms_modulation_mae_db": 4.6202071729701695, "mr_stft_db": 5.942547549648829, "band_mae_db": 0.8876112889895267, "rms_gain_error_db": -0.16564497075109266, "output_peak": 0.9427420615545832}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9818442239969848, "全帯域_10ms_rms_envelope_nrmse": 0.07460679795865853, "全帯域_10ms_envelope_correlation": 0.994906866739699, "全帯域_10ms_modulation_mae_db": 0.40438258145647366, "全帯域_40ms_log_rms_mae_db": 0.5282696186814073, "全帯域_40ms_rms_envelope_nrmse": 0.04389417835085382, "全帯域_40ms_envelope_correlation": 0.9981358746501845, "全帯域_40ms_modulation_mae_db": 0.2675448067112928, "300–1000 Hz_10ms_log_rms_mae_db": 1.4194077976437138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1062652358963445, "300–1000 Hz_10ms_envelope_correlation": 0.9901359569146774, "300–1000 Hz_10ms_modulation_mae_db": 0.503138373700823, "300–1000 Hz_40ms_log_rms_mae_db": 0.7917342676208667, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058556376964886574, "300–1000 Hz_40ms_envelope_correlation": 0.997046789672566, "300–1000 Hz_40ms_modulation_mae_db": 0.596079186603972, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6715672084701343, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11360930716610255, "1000–3000 Hz_10ms_envelope_correlation": 0.9903650924574096, "1000–3000 Hz_10ms_modulation_mae_db": 0.3774499158710161, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8762951767768316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062221893520643945, "1000–3000 Hz_40ms_envelope_correlation": 0.9970657442185155, "1000–3000 Hz_40ms_modulation_mae_db": 0.24636111123917526, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8225574527037371, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4067555851093078, "3000–6000 Hz_10ms_envelope_correlation": 0.9011919417623969, "3000–6000 Hz_10ms_modulation_mae_db": 4.479241152050278, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4134632405673122, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31597027545996387, "3000–6000 Hz_40ms_envelope_correlation": 0.9375018501997517, "3000–6000 Hz_40ms_modulation_mae_db": 4.902193527855901, "mr_stft_db": 5.962168505743091, "band_mae_db": 0.8585172771703021, "rms_gain_error_db": -0.09641939681028391, "output_peak": 0.923508364420297}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9614575568442247, "全帯域_10ms_rms_envelope_nrmse": 0.0714079902207116, "全帯域_10ms_envelope_correlation": 0.995534386817774, "全帯域_10ms_modulation_mae_db": 0.2571431719571516, "全帯域_40ms_log_rms_mae_db": 0.5020361370439362, "全帯域_40ms_rms_envelope_nrmse": 0.03791790072820434, "全帯域_40ms_envelope_correlation": 0.998904865703459, "全帯域_40ms_modulation_mae_db": 0.3073933355973193, "300–1000 Hz_10ms_log_rms_mae_db": 1.2768473270716478, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1048436144372503, "300–1000 Hz_10ms_envelope_correlation": 0.9904152128328718, "300–1000 Hz_10ms_modulation_mae_db": 0.356703317098777, "300–1000 Hz_40ms_log_rms_mae_db": 0.7280225496655419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05237454345994536, "300–1000 Hz_40ms_envelope_correlation": 0.9977580198671718, "300–1000 Hz_40ms_modulation_mae_db": 0.16322630978081418, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6495735336350439, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11068705000273088, "1000–3000 Hz_10ms_envelope_correlation": 0.991140556086508, "1000–3000 Hz_10ms_modulation_mae_db": 0.779083937991316, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9308531780053785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060746891086878724, "1000–3000 Hz_40ms_envelope_correlation": 0.9976181649956725, "1000–3000 Hz_40ms_modulation_mae_db": 0.38089596580448193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8677239887880737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3974009123954693, "3000–6000 Hz_10ms_envelope_correlation": 0.9058996598113785, "3000–6000 Hz_10ms_modulation_mae_db": 4.281373284511313, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4773495339485576, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3241047804556745, "3000–6000 Hz_40ms_envelope_correlation": 0.9341615290762549, "3000–6000 Hz_40ms_modulation_mae_db": 4.286583748113782, "mr_stft_db": 5.9615818797036715, "band_mae_db": 0.8910895979747018, "rms_gain_error_db": -0.17374050969650792, "output_peak": 0.9514291532755259}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 33040, "output_samples": 33040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.826420845037825, "spectral_convergence": 0.30632476734974634, "spectral_convergence_status": "ok", "amplitude_floor": 1.5756171103077008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.642524442611484, "spectral_convergence": 0.3343826160007296, "spectral_convergence_status": "ok", "amplitude_floor": 1.2505050051947005e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.429930357514322, "spectral_convergence": 0.35535674137205703, "spectral_convergence_status": "ok", "amplitude_floor": 9.121352204658788e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021685618865241799, "output_power": 0.00021398858690273477, "delta_db": -0.05781209740355589, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010795690773652512, "output_power": 0.00111287847399006, "delta_db": 0.1319730574164272, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002989928450689294, "output_power": 0.0029692416916385043, "delta_db": -0.030152458071866263, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0020580355715967367, "output_power": 0.0018752884616422471, "delta_db": -0.4038479546031845, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0009751360590152492, "output_power": 0.0009113557500777607, "delta_db": -0.2937727822439038, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.9392123549665832e-05, "output_power": 5.201863083800468e-05, "delta_db": 4.285335476754889, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8671489710823045, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12251929271490493, "pitch": null, "output_rms_dbfs": -21.466199253636432, "output_peak": 0.9586073318389958, "reference_rms_dbfs": -21.343679960921527, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.966291881721211}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 33040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00032988134947798775, "log_rms_mae_db": 0.9272420871342977, "rms_envelope_nrmse": 0.07946948016240427, "envelope_correlation": 0.9942263921613126, "rms_gain_error_db": -0.1225183332312988, "reference_mean_power": 0.007356724541990913, "output_mean_power": 0.007152084908939394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 2.980285612038863e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020727352824950612, "output_power": 0.0019672022799328386, "reference_active": true, "delta_db": -0.22694821043725527}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00028987718896809884, "output_power": 0.00023475186411612257, "reference_active": true, "delta_db": -0.9160499157931192}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.012977878291207e-05, "output_power": 5.824072389201178e-05, "reference_active": true, "delta_db": -0.8067570464492837}], "active_band_count": 3, "active_band_mae_db": 0.6499183908932195}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0002960474071300256, "log_rms_mae_db": 0.48665426550696333, "rms_envelope_nrmse": 0.03332442398885131, "envelope_correlation": 0.9990532859648944, "rms_gain_error_db": -0.1225197435242194, "reference_mean_power": 0.007465415825536721, "output_mean_power": 0.007257750405294572}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 2.5597102021075326e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001877761726334087, "output_power": 0.001784825208897299, "reference_active": true, "delta_db": -0.22044791462467728}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012702101262043316, "output_power": 0.00010477198710335116, "reference_active": true, "delta_db": -0.8363038983384194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.962314408759724e-06, "output_power": 3.2122012455834783e-06, "reference_active": true, "delta_db": -0.9114618757237578}], "active_band_count": 3, "active_band_mae_db": 0.6560712295622848}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00021933697126022545, "log_rms_mae_db": 1.3467554817218401, "rms_envelope_nrmse": 0.12226297255241018, "envelope_correlation": 0.9866316854390285, "rms_gain_error_db": -0.0890602317175601, "reference_mean_power": 0.003363603458339666, "output_mean_power": 0.003295328900231235}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.3609823101859838e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008546335077009212, "output_power": 0.0007717567584911672, "reference_active": true, "delta_db": -0.44299474832113905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016660612815788933, "output_power": 0.00012956619613432613, "reference_active": true, "delta_db": -1.0919926298638742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.188726605138031e-05, "output_power": 4.409149022727993e-05, "reference_active": true, "delta_db": 0.22272762356749365}], "active_band_count": 3, "active_band_mae_db": 0.585905000584169}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0001893291578274055, "log_rms_mae_db": 0.7621514536032734, "rms_envelope_nrmse": 0.0625565406694916, "envelope_correlation": 0.9962780751178785, "rms_gain_error_db": -0.08906180793455919, "reference_mean_power": 0.0034133051942474998, "output_mean_power": 0.0033440205746368907}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.15224102535942e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007711766819603955, "output_power": 0.0006993032580423077, "reference_active": true, "delta_db": -0.42488337732037246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.304889773698623e-05, "output_power": 6.026870776421116e-05, "reference_active": true, "delta_db": -0.8352178732688729}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527118205585199e-06, "output_power": 2.1748548748473494e-06, "reference_active": true, "delta_db": -0.6519527408756929}], "active_band_count": 3, "active_band_mae_db": 0.6373513304883127}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.0002256735995145706, "log_rms_mae_db": 1.6980155518160185, "rms_envelope_nrmse": 0.12022260424047229, "envelope_correlation": 0.9900105566200781, "rms_gain_error_db": -0.37416470884299735, "reference_mean_power": 0.002711357875845494, "output_mean_power": 0.0024875417218409505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.6613986227009806e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012279896352023905, "output_power": 0.001106397956253384, "reference_active": true, "delta_db": -0.452833363053748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888327196960883, "output_power": 9.313877991997763e-05, "reference_active": true, "delta_db": -1.05990204914817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.234146057131384e-05, "output_power": 3.024211918628711e-05, "reference_active": true, "delta_db": -0.2914740852033454}], "active_band_count": 3, "active_band_mae_db": 0.6014031658017545}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.00020374269470838238, "log_rms_mae_db": 0.8882525516195505, "rms_envelope_nrmse": 0.058340176981591586, "envelope_correlation": 0.9986144900384936, "rms_gain_error_db": -0.3741677849859798, "reference_mean_power": 0.002751412553659036, "output_mean_power": 0.002524288193144205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.4886745667367597e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001137287164515336, "output_power": 0.0010274574927700572, "reference_active": true, "delta_db": -0.4410627410080395}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.907501205097428e-05, "output_power": 3.980966262660472e-05, "reference_active": true, "delta_db": -0.9087191781723646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5268858664777347e-06, "output_power": 1.835110571702551e-06, "reference_active": true, "delta_db": -1.3892338918956}], "active_band_count": 3, "active_band_mae_db": 0.913005270358668}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 4.718157231228245e-05, "log_rms_mae_db": 1.9681420092984039, "rms_envelope_nrmse": 0.40384188444731434, "envelope_correlation": 0.8998538815396165, "rms_gain_error_db": 0.743112532135693, "reference_mean_power": 9.509723574878595e-05, "output_mean_power": 0.00011284417516872603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 5.720070068605411e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.50238140104046e-05, "output_power": 4.7615372204412233e-05, "reference_active": true, "delta_db": 1.3338374555089747}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.975419310826897e-06, "output_power": 8.371100533189776e-06, "reference_active": true, "delta_db": 2.2595287020903703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.471999704474668e-06, "output_power": 1.573077814114549e-05, "reference_active": true, "delta_db": 10.288424832232245}], "active_band_count": 3, "active_band_mae_db": 4.627263663277197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 4.015479902310721e-05, "log_rms_mae_db": 1.6035358699270268, "rms_envelope_nrmse": 0.32435437273737294, "envelope_correlation": 0.9313686761948712, "rms_gain_error_db": 0.7431504768779779, "reference_mean_power": 9.64997641207967e-05, "output_mean_power": 0.00011450944228769675}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 5.087796481772889e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1685652820939687e-05, "output_power": 5.543692305815512e-05, "reference_active": true, "delta_db": 2.429364580776692}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.336227531732081e-06, "output_power": 4.798456666542472e-06, "reference_active": true, "delta_db": 3.125864395169403}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1955736150290575e-07, "output_power": 1.1176550575632102e-06, "reference_active": true, "delta_db": 9.707314656278006}], "active_band_count": 3, "active_band_mae_db": 5.0875145440747005}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8430108255786105, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0620a20688d033c3d9f53c48c2f7b96f823971179b084f1b260c2506e8ac2897", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9890199393504484, "standard_deviation": 0.0017501788017441557, "minimum": 0.986358061205878, "maximum": 0.991140556086508, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7114351413471902, "standard_deviation": 0.054946533456785206, "minimum": 1.6437277107919257, "maximum": 1.7981090175117065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6444590315551783, "standard_deviation": 0.16993764370229617, "minimum": 0.3774499158710161, "maximum": 0.9567936783247669, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1228631864973402, "standard_deviation": 0.00951769820511261, "minimum": 0.11068705000273088, "maximum": 0.13423434669404374, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972937020494339, "standard_deviation": 0.000807944479987961, "minimum": 0.9958954714201798, "maximum": 0.9986144900384936, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.930130514427107, "standard_deviation": 0.054468543909181474, "minimum": 0.8762951767768316, "maximum": 1.0161046580580464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7553953840394523, "standard_deviation": 0.33309401069256744, "minimum": 0.24636111123917526, "maximum": 1.2343264236386635, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06485866481160085, "standard_deviation": 0.008914311856568638, "minimum": 0.058340176981591586, "maximum": 0.08377680743881527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9051553102136776, "standard_deviation": 0.004252033907000277, "minimum": 0.8998538815396165, "maximum": 0.9119696942882051, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8944304041222926, "standard_deviation": 0.05121637226886124, "minimum": 1.8225574527037371, "maximum": 1.9681420092984039, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 4.450493004492644, "standard_deviation": 0.1213129560461712, "minimum": 4.281373284511313, "maximum": 4.627263663277197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3989805293379637, "standard_deviation": 0.005118795465092143, "minimum": 0.3905609227259569, "maximum": 0.4067555851093078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9368887800602705, "standard_deviation": 0.003347342483824994, "minimum": 0.9313686761948712, "maximum": 0.9417188766182792, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5201385857131164, "standard_deviation": 0.0691315515952385, "minimum": 1.4134632405673122, "maximum": 1.6035358699270268, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.698742980459098, "standard_deviation": 0.268178592849407, "minimum": 4.286583748113782, "maximum": 5.0875145440747005, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3177444040830013, "standard_deviation": 0.005132246682565389, "minimum": 0.3102487282131396, "maximum": 0.32435437273737294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.988401925372195, "standard_deviation": 0.0015479494085847644, "minimum": 0.9866186074051713, "maximum": 0.9904152128328718, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3674786824360234, "standard_deviation": 0.06408710459756799, "minimum": 1.2768473270716478, "maximum": 1.432232410454782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4378028312625126, "standard_deviation": 0.14801522077537677, "minimum": 0.2049647219851021, "maximum": 0.6546947512356537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11421137065976684, "standard_deviation": 0.007381728783158147, "minimum": 0.1048436144372503, "maximum": 0.12308187446126599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968625408013747, "standard_deviation": 0.0006946830771216714, "minimum": 0.9959443970798498, "maximum": 0.9977580198671718, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7945237997507074, "standard_deviation": 0.05453594390435512, "minimum": 0.7280225496655419, "maximum": 0.8988433450063668, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48249537173949214, "standard_deviation": 0.2002443191055508, "minimum": 0.16322630978081418, "maximum": 0.6774246447235205, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05872657918407353, "standard_deviation": 0.005781630221715108, "minimum": 0.0517153518687788, "maximum": 0.06752030152369101, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.8747870344880653, "standard_deviation": 0.024824686503299123, "minimum": 0.8366622788413428, "maximum": 0.9036531097338454, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.960073610773338, "standard_deviation": 0.008380798856208604, "minimum": 5.942547549648829, "maximum": 5.968141661688772, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9451661561572429, "standard_deviation": 0.02164831915218978, "minimum": 0.923508364420297, "maximum": 0.9881262081147384, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13007606438056651, "standard_deviation": 0.05155087400598079, "minimum": -0.1801622136235456, "maximum": -0.056701683244253794, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939990002156538, "standard_deviation": 0.0009591635704637635, "minimum": 0.9927607298268777, "maximum": 0.995534386817774, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9778698018306666, "standard_deviation": 0.03395767129470026, "minimum": 0.9272420871342977, "maximum": 1.0366518551984163, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.46795722602739953, "standard_deviation": 0.13325172663147045, "minimum": 0.2571431719571516, "maximum": 0.6499183908932195, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08108415003391461, "standard_deviation": 0.006591424224533971, "minimum": 0.0714079902207116, "maximum": 0.08983939035058267, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984454935386716, "standard_deviation": 0.000445498573173002, "minimum": 0.9977963822185361, "maximum": 0.9990532859648944, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5194070243231, "standard_deviation": 0.017871157435849973, "minimum": 0.48665426550696333, "maximum": 0.5405694682767539, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5004426420997186, "standard_deviation": 0.15433381939950008, "minimum": 0.2675448067112928, "maximum": 0.656135309336821, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04140321335225208, "standard_deviation": 0.0061907475897439975, "minimum": 0.03332442398885131, "maximum": 0.05033174256594054, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0029", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.2541604042053223, "predicted_mos": 1.2541604042053223, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "26958e1c8889c3f431c165c577dd8db73b61cdb97f85f3d0d25989bbea912263", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.103649300057441, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9387537264748378, "全帯域_10ms_rms_envelope_nrmse": 0.0828741446975953, "全帯域_10ms_envelope_correlation": 0.9939120245047007, "全帯域_10ms_modulation_mae_db": 0.6800866549184611, "全帯域_40ms_log_rms_mae_db": 0.47212825316103585, "全帯域_40ms_rms_envelope_nrmse": 0.04363748678362944, "全帯域_40ms_envelope_correlation": 0.9984656498617864, "全帯域_40ms_modulation_mae_db": 0.7399528356773897, "300–1000 Hz_10ms_log_rms_mae_db": 1.5214109923154115, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12456470517599093, "300–1000 Hz_10ms_envelope_correlation": 0.9863002341960646, "300–1000 Hz_10ms_modulation_mae_db": 0.39459631095453657, "300–1000 Hz_40ms_log_rms_mae_db": 0.9398827039231177, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05874260628719312, "300–1000 Hz_40ms_envelope_correlation": 0.9970816192243288, "300–1000 Hz_40ms_modulation_mae_db": 0.9213315558810464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8283265933030368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11262470050573026, "1000–3000 Hz_10ms_envelope_correlation": 0.9914116889339207, "1000–3000 Hz_10ms_modulation_mae_db": 1.0183463177468042, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9519237325805249, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06253639472065674, "1000–3000 Hz_40ms_envelope_correlation": 0.998121701501055, "1000–3000 Hz_40ms_modulation_mae_db": 1.0288572245028609, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4173605602215935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41184075618331906, "3000–6000 Hz_10ms_envelope_correlation": 0.9000409715481358, "3000–6000 Hz_10ms_modulation_mae_db": 4.40493496359237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8442943924940813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3294158641011207, "3000–6000 Hz_40ms_envelope_correlation": 0.9335786774372226, "3000–6000 Hz_40ms_modulation_mae_db": 4.6175954237687735, "mr_stft_db": 6.289332248434803, "band_mae_db": 0.9123519985128903, "rms_gain_error_db": -0.17735364720424923, "output_peak": 0.9352803879234428}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.89615779705776, "全帯域_10ms_rms_envelope_nrmse": 0.081383952712438, "全帯域_10ms_envelope_correlation": 0.9940790805960141, "全帯域_10ms_modulation_mae_db": 0.2636785148234274, "全帯域_40ms_log_rms_mae_db": 0.49133242960705253, "全帯域_40ms_rms_envelope_nrmse": 0.04736189843419704, "全帯域_40ms_envelope_correlation": 0.9980506695286464, "全帯域_40ms_modulation_mae_db": 0.3536032013964912, "300–1000 Hz_10ms_log_rms_mae_db": 1.4567625505278932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13054516212144518, "300–1000 Hz_10ms_envelope_correlation": 0.9848488892290193, "300–1000 Hz_10ms_modulation_mae_db": 0.7224171484589714, "300–1000 Hz_40ms_log_rms_mae_db": 0.9054012323059195, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0771512393589629, "300–1000 Hz_40ms_envelope_correlation": 0.9944002075977376, "300–1000 Hz_40ms_modulation_mae_db": 0.7093454505328006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9210301818044468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12079469942734947, "1000–3000 Hz_10ms_envelope_correlation": 0.9895114597883431, "1000–3000 Hz_10ms_modulation_mae_db": 0.5317904771591557, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0706855625007805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06503119373181743, "1000–3000 Hz_40ms_envelope_correlation": 0.9974330410162924, "1000–3000 Hz_40ms_modulation_mae_db": 0.6842391166371705, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3230752546510876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4177818109692262, "3000–6000 Hz_10ms_envelope_correlation": 0.8966717682729745, "3000–6000 Hz_10ms_modulation_mae_db": 4.562611166184552, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7656332333769413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33080543843935334, "3000–6000 Hz_40ms_envelope_correlation": 0.9325247094070813, "3000–6000 Hz_40ms_modulation_mae_db": 4.9444371688011985, "mr_stft_db": 6.304148340683632, "band_mae_db": 0.9152032152139985, "rms_gain_error_db": -0.17116826505019384, "output_peak": 0.9825515120638382}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9257362461528049, "全帯域_10ms_rms_envelope_nrmse": 0.06728868402591401, "全帯域_10ms_envelope_correlation": 0.995825055311682, "全帯域_10ms_modulation_mae_db": 0.22728462957652443, "全帯域_40ms_log_rms_mae_db": 0.5098281342651743, "全帯域_40ms_rms_envelope_nrmse": 0.035124617927208134, "全帯域_40ms_envelope_correlation": 0.9987703815908735, "全帯域_40ms_modulation_mae_db": 0.6751430563610757, "300–1000 Hz_10ms_log_rms_mae_db": 1.4760767073838987, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12408186022383862, "300–1000 Hz_10ms_envelope_correlation": 0.986429788555305, "300–1000 Hz_10ms_modulation_mae_db": 0.8377984391271731, "300–1000 Hz_40ms_log_rms_mae_db": 0.9902495984587123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07318039931643106, "300–1000 Hz_40ms_envelope_correlation": 0.9948530004760655, "300–1000 Hz_40ms_modulation_mae_db": 0.5765255789253313, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7069283850897923, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09619175332968513, "1000–3000 Hz_10ms_envelope_correlation": 0.9930100254467243, "1000–3000 Hz_10ms_modulation_mae_db": 0.6550679709287592, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8432957732357041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047506203300759976, "1000–3000 Hz_40ms_envelope_correlation": 0.9981941295045558, "1000–3000 Hz_40ms_modulation_mae_db": 1.3802038427670826, "3000–6000 Hz_10ms_log_rms_mae_db": 2.398770833177153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4364483482643409, "3000–6000 Hz_10ms_envelope_correlation": 0.8944386633430644, "3000–6000 Hz_10ms_modulation_mae_db": 5.062084427587308, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8557551143646767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.35876924364924656, "3000–6000 Hz_40ms_envelope_correlation": 0.9267102760105406, "3000–6000 Hz_40ms_modulation_mae_db": 5.388008704311548, "mr_stft_db": 6.296452365152201, "band_mae_db": 0.8439833339682937, "rms_gain_error_db": 0.03337798623371076, "output_peak": 1.018172382845083}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 3.026634382566586e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.026634382566586e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9588052807027992, "全帯域_10ms_rms_envelope_nrmse": 0.09132823473847224, "全帯域_10ms_envelope_correlation": 0.9926879991118989, "全帯域_10ms_modulation_mae_db": 0.6231887575528682, "全帯域_40ms_log_rms_mae_db": 0.48748306129465013, "全帯域_40ms_rms_envelope_nrmse": 0.04782139415301153, "全帯域_40ms_envelope_correlation": 0.9983110068518621, "全帯域_40ms_modulation_mae_db": 0.7331578166960718, "300–1000 Hz_10ms_log_rms_mae_db": 1.4718042930460933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13017714656986137, "300–1000 Hz_10ms_envelope_correlation": 0.9859585402166335, "300–1000 Hz_10ms_modulation_mae_db": 0.8337392888967119, "300–1000 Hz_40ms_log_rms_mae_db": 0.9116902947553773, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07411404844063021, "300–1000 Hz_40ms_envelope_correlation": 0.9963418620403846, "300–1000 Hz_40ms_modulation_mae_db": 0.49740372190966936, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9153635961620534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1398959786874067, "1000–3000 Hz_10ms_envelope_correlation": 0.9854252713126616, "1000–3000 Hz_10ms_modulation_mae_db": 0.7512291488511135, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0314688765571451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0631341403306259, "1000–3000 Hz_40ms_envelope_correlation": 0.9973631698621982, "1000–3000 Hz_40ms_modulation_mae_db": 1.1364067467092165, "3000–6000 Hz_10ms_log_rms_mae_db": 2.369289619345296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.43145527238334586, "3000–6000 Hz_10ms_envelope_correlation": 0.8952890041885715, "3000–6000 Hz_10ms_modulation_mae_db": 4.893156911543485, "3000–6000 Hz_40ms_log_rms_mae_db": 1.828550741047257, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.35393419502600554, "3000–6000 Hz_40ms_envelope_correlation": 0.927454485446894, "3000–6000 Hz_40ms_modulation_mae_db": 5.081119457587288, "mr_stft_db": 6.297938184730168, "band_mae_db": 0.9266598133684609, "rms_gain_error_db": -0.2304785823074443, "output_peak": 0.9578923437590566}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9313451874630914, "全帯域_10ms_rms_envelope_nrmse": 0.08288326772424243, "全帯域_10ms_envelope_correlation": 0.9937966919910997, "全帯域_10ms_modulation_mae_db": 0.6461991119058498, "全帯域_40ms_log_rms_mae_db": 0.4818889884406214, "全帯域_40ms_rms_envelope_nrmse": 0.03915830689880625, "全帯域_40ms_envelope_correlation": 0.9987329081034069, "全帯域_40ms_modulation_mae_db": 0.8570292126769053, "300–1000 Hz_10ms_log_rms_mae_db": 1.3850381602461594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12199966426529611, "300–1000 Hz_10ms_envelope_correlation": 0.9869622882046539, "300–1000 Hz_10ms_modulation_mae_db": 0.5199343929628638, "300–1000 Hz_40ms_log_rms_mae_db": 0.8187507516940634, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06843647192875835, "300–1000 Hz_40ms_envelope_correlation": 0.9959381462786581, "300–1000 Hz_40ms_modulation_mae_db": 0.7181228011242151, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8508031995700784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11573222472666102, "1000–3000 Hz_10ms_envelope_correlation": 0.9902582070706324, "1000–3000 Hz_10ms_modulation_mae_db": 0.7319922308129666, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9995588960699384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057517417984872504, "1000–3000 Hz_40ms_envelope_correlation": 0.9979291468591331, "1000–3000 Hz_40ms_modulation_mae_db": 1.2806097431901777, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4048029137384326, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4227854909116827, "3000–6000 Hz_10ms_envelope_correlation": 0.8968387225684145, "3000–6000 Hz_10ms_modulation_mae_db": 4.643856834308617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.868559877482093, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33692733953657117, "3000–6000 Hz_40ms_envelope_correlation": 0.9324338078157113, "3000–6000 Hz_40ms_modulation_mae_db": 5.067938460906487, "mr_stft_db": 6.313752241299931, "band_mae_db": 0.9030165883485374, "rms_gain_error_db": -0.15560770502972754, "output_peak": 0.9458108753520603}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9678631792855806, "全帯域_10ms_rms_envelope_nrmse": 0.09366830232216077, "全帯域_10ms_envelope_correlation": 0.9919478074932044, "全帯域_10ms_modulation_mae_db": 0.524079702707574, "全帯域_40ms_log_rms_mae_db": 0.5165113019771875, "全帯域_40ms_rms_envelope_nrmse": 0.04911597432246103, "全帯域_40ms_envelope_correlation": 0.9977531096632061, "全帯域_40ms_modulation_mae_db": 0.6941412678230594, "300–1000 Hz_10ms_log_rms_mae_db": 1.548223887235334, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13305867311318845, "300–1000 Hz_10ms_envelope_correlation": 0.9848567571314464, "300–1000 Hz_10ms_modulation_mae_db": 0.7363093149876493, "300–1000 Hz_40ms_log_rms_mae_db": 1.0250073012158079, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0797621302251828, "300–1000 Hz_40ms_envelope_correlation": 0.994932948628603, "300–1000 Hz_40ms_modulation_mae_db": 0.6580382579096119, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9215191173707957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12706049615538562, "1000–3000 Hz_10ms_envelope_correlation": 0.9877793832285839, "1000–3000 Hz_10ms_modulation_mae_db": 0.425516055640049, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9670588200755849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0633861980582489, "1000–3000 Hz_40ms_envelope_correlation": 0.9968290182168914, "1000–3000 Hz_40ms_modulation_mae_db": 0.7860545382492082, "3000–6000 Hz_10ms_log_rms_mae_db": 2.269002687363097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4182206785821177, "3000–6000 Hz_10ms_envelope_correlation": 0.8958989887076632, "3000–6000 Hz_10ms_modulation_mae_db": 4.444027156639276, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7828207032462446, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33920206884355253, "3000–6000 Hz_40ms_envelope_correlation": 0.928405091397875, "3000–6000 Hz_40ms_modulation_mae_db": 4.878729385257618, "mr_stft_db": 6.266325746682692, "band_mae_db": 0.9845060328113625, "rms_gain_error_db": -0.14150810883530673, "output_peak": 0.9396134694169793}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9659941610213308, "全帯域_10ms_rms_envelope_nrmse": 0.09514055563227503, "全帯域_10ms_envelope_correlation": 0.9915605957767808, "全帯域_10ms_modulation_mae_db": 0.2476168428584372, "全帯域_40ms_log_rms_mae_db": 0.5120100183278582, "全帯域_40ms_rms_envelope_nrmse": 0.04977673355563312, "全帯域_40ms_envelope_correlation": 0.99749435640221, "全帯域_40ms_modulation_mae_db": 0.5253619768343613, "300–1000 Hz_10ms_log_rms_mae_db": 1.604052529626055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11930592433955621, "300–1000 Hz_10ms_envelope_correlation": 0.9872184809152881, "300–1000 Hz_10ms_modulation_mae_db": 0.5276303753089286, "300–1000 Hz_40ms_log_rms_mae_db": 0.9323291962268392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053254039013918196, "300–1000 Hz_40ms_envelope_correlation": 0.9972675993158294, "300–1000 Hz_40ms_modulation_mae_db": 0.5194301593442714, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8777507186678524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13168218872473547, "1000–3000 Hz_10ms_envelope_correlation": 0.9869275777376173, "1000–3000 Hz_10ms_modulation_mae_db": 0.6675002055722871, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0477975618786122, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07644570159114746, "1000–3000 Hz_40ms_envelope_correlation": 0.9954570802514306, "1000–3000 Hz_40ms_modulation_mae_db": 1.1461799533471166, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2833222215188327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4166047270228308, "3000–6000 Hz_10ms_envelope_correlation": 0.9002998877865083, "3000–6000 Hz_10ms_modulation_mae_db": 4.828774225978571, "3000–6000 Hz_40ms_log_rms_mae_db": 1.749648910430599, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.342281664358007, "3000–6000 Hz_40ms_envelope_correlation": 0.9302640081010071, "3000–6000 Hz_40ms_modulation_mae_db": 5.169554365506974, "mr_stft_db": 6.251694077786076, "band_mae_db": 0.8761600707398137, "rms_gain_error_db": -0.059229228112225446, "output_peak": 0.9569018018503486}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9421217079544887, "全帯域_10ms_rms_envelope_nrmse": 0.08245403145792232, "全帯域_10ms_envelope_correlation": 0.9940445228017706, "全帯域_10ms_modulation_mae_db": 0.35315021577700173, "全帯域_40ms_log_rms_mae_db": 0.5065849289585885, "全帯域_40ms_rms_envelope_nrmse": 0.04449742188842885, "全帯域_40ms_envelope_correlation": 0.9984815864325092, "全帯域_40ms_modulation_mae_db": 0.5800048004851838, "300–1000 Hz_10ms_log_rms_mae_db": 1.5620710941404767, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11705615840454514, "300–1000 Hz_10ms_envelope_correlation": 0.9883405747411356, "300–1000 Hz_10ms_modulation_mae_db": 0.44818507312362454, "300–1000 Hz_40ms_log_rms_mae_db": 0.897037326759124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07033566381438992, "300–1000 Hz_40ms_envelope_correlation": 0.9960694916535683, "300–1000 Hz_40ms_modulation_mae_db": 0.8452887719393264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.820382193130744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10879678422254305, "1000–3000 Hz_10ms_envelope_correlation": 0.9916280975580511, "1000–3000 Hz_10ms_modulation_mae_db": 0.22135147680394973, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0091551551519233, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054913071506536154, "1000–3000 Hz_40ms_envelope_correlation": 0.9983798096953254, "1000–3000 Hz_40ms_modulation_mae_db": 0.35726257063032624, "3000–6000 Hz_10ms_log_rms_mae_db": 2.289328618784512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4268437626052544, "3000–6000 Hz_10ms_envelope_correlation": 0.8925430003079275, "3000–6000 Hz_10ms_modulation_mae_db": 4.726911433016006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.804494874215036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34777133466875293, "3000–6000 Hz_40ms_envelope_correlation": 0.9253365992138503, "3000–6000 Hz_40ms_modulation_mae_db": 5.150859668668264, "mr_stft_db": 6.3191920253309535, "band_mae_db": 0.9281625120565756, "rms_gain_error_db": -0.19873726248889162, "output_peak": 0.9288831207812248}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 33040, "output_samples": 33040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.189008198071371, "spectral_convergence": 0.32053140222297266, "spectral_convergence_status": "ok", "amplitude_floor": 1.5756171103077008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.010267367953003, "spectral_convergence": 0.3260124810375904, "spectral_convergence_status": "ok", "amplitude_floor": 1.2505050051947005e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.668721179280035, "spectral_convergence": 0.3423536666060758, "spectral_convergence_status": "ok", "amplitude_floor": 9.121352204658788e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021685618865241799, "output_power": 0.00022102104943155513, "delta_db": 0.08261816077756634, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010795690773652512, "output_power": 0.0010824484974876233, "delta_db": 0.011568055913511215, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002989928450689294, "output_power": 0.0028763803568333697, "delta_db": -0.1681448149567326, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0020580355715967367, "output_power": 0.0018853975971027846, "delta_db": -0.38049927697596897, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0009751360590152492, "output_power": 0.000924106431786283, "delta_db": -0.2334322332289668, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.9392123549665832e-05, "output_power": 5.58998077621233e-05, "delta_db": 4.597849449224595, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9123519985128903, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.17735364720424923, "pitch": null, "output_rms_dbfs": -21.52103360812578, "output_peak": 0.9352803879234428, "reference_rms_dbfs": -21.343679960921527, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.289332248434803}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 33040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00032988134947798775, "log_rms_mae_db": 0.9387537264748378, "rms_envelope_nrmse": 0.0828741446975953, "envelope_correlation": 0.9939120245047007, "rms_gain_error_db": -0.17735324354507442, "reference_mean_power": 0.007356724541990913, "output_mean_power": 0.007062348925397509}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 2.980285612038863e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020727352824950612, "output_power": 0.0019102642533792816, "reference_active": true, "delta_db": -0.3545039126730645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00028987718896809884, "output_power": 0.0002602544231827109, "reference_active": true, "delta_db": -0.4681592182904343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.012977878291207e-05, "output_power": 5.298376131284204e-05, "reference_active": true, "delta_db": -1.2175968337918845}], "active_band_count": 3, "active_band_mae_db": 0.6800866549184611}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0002960474071300256, "log_rms_mae_db": 0.47212825316103585, "rms_envelope_nrmse": 0.04363748678362944, "envelope_correlation": 0.9984656498617864, "rms_gain_error_db": -0.17735148204683804, "reference_mean_power": 0.007465415825536721, "output_mean_power": 0.007166693888943425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 2.5597102021075326e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001877761726334087, "output_power": 0.0017269162006090793, "reference_active": true, "delta_db": -0.36369219020873306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012702101262043316, "output_power": 0.0001222646417517553, "reference_active": true, "delta_db": -0.165746909763235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.962314408759724e-06, "output_power": 2.6847694233875915e-06, "reference_active": true, "delta_db": -1.690419407060201}], "active_band_count": 3, "active_band_mae_db": 0.7399528356773897}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00021933697126022545, "log_rms_mae_db": 1.5214109923154115, "rms_envelope_nrmse": 0.12456470517599093, "envelope_correlation": 0.9863002341960646, "rms_gain_error_db": -0.2124442889266956, "reference_mean_power": 0.003363603458339666, "output_mean_power": 0.003203025243704067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.3609823101859838e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008546335077009212, "output_power": 0.000741934609642455, "reference_active": true, "delta_db": -0.6141428579156438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016660612815788933, "output_power": 0.00017857951934716724, "reference_active": true, "delta_db": 0.3014067799400852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.188726605138031e-05, "output_power": 3.937840177415566e-05, "reference_active": true, "delta_db": -0.2682392950078807}], "active_band_count": 3, "active_band_mae_db": 0.39459631095453657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0001893291578274055, "log_rms_mae_db": 0.9398827039231177, "rms_envelope_nrmse": 0.05874260628719312, "envelope_correlation": 0.9970816192243288, "rms_gain_error_db": -0.21244293687942334, "reference_mean_power": 0.0034133051942474998, "output_mean_power": 0.0032503552338103755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.15224102535942e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007711766819603955, "output_power": 0.0006793855372820196, "reference_active": true, "delta_db": -0.5503759194988702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.304889773698623e-05, "output_power": 8.324237413735009e-05, "reference_active": true, "delta_db": 0.5673079105044678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527118205585199e-06, "output_power": 1.7297943438147642e-06, "reference_active": true, "delta_db": -1.6463108376398012}], "active_band_count": 3, "active_band_mae_db": 0.9213315558810464}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.0002256735995145706, "log_rms_mae_db": 1.8283265933030368, "rms_envelope_nrmse": 0.11262470050573026, "envelope_correlation": 0.9914116889339207, "rms_gain_error_db": -0.35197643348242524, "reference_mean_power": 0.002711357875845494, "output_mean_power": 0.0025002831906472383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.6613986227009806e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012279896352023905, "output_power": 0.0011096953489351182, "reference_active": true, "delta_db": -0.43990935394448416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888327196960883, "output_power": 8.822218571365279e-05, "reference_active": true, "delta_db": -1.2954293620246333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.234146057131384e-05, "output_power": 2.386654784607536e-05, "reference_active": true, "delta_db": -1.3197002372712952}], "active_band_count": 3, "active_band_mae_db": 1.0183463177468042}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.00020374269470838238, "log_rms_mae_db": 0.9519237325805249, "rms_envelope_nrmse": 0.06253639472065674, "envelope_correlation": 0.998121701501055, "rms_gain_error_db": -0.35197452507696875, "reference_mean_power": 0.002751412553659036, "output_mean_power": 0.0025372207935689633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.4886745667367597e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001137287164515336, "output_power": 0.0010247819209436532, "reference_active": true, "delta_db": -0.45238682659697277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.907501205097428e-05, "output_power": 4.0002187441097014e-05, "reference_active": true, "delta_db": -0.8877667439987968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5268858664777347e-06, "output_power": 1.6902222561041802e-06, "reference_active": true, "delta_db": -1.7464181029128127}], "active_band_count": 3, "active_band_mae_db": 1.0288572245028609}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 4.718157231228245e-05, "log_rms_mae_db": 2.4173605602215935, "rms_envelope_nrmse": 0.41184075618331906, "envelope_correlation": 0.9000409715481358, "rms_gain_error_db": 0.9243287682065581, "reference_mean_power": 9.509723574878595e-05, "output_mean_power": 0.00011765239437136715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 5.720070068605411e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.50238140104046e-05, "output_power": 4.934077839285457e-05, "reference_active": true, "delta_db": 1.4884255861024998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.975419310826897e-06, "output_power": 7.247890257220953e-06, "reference_active": true, "delta_db": 1.6338192148320896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.471999704474668e-06, "output_power": 1.503708767053371e-05, "reference_active": true, "delta_db": 10.092560089842522}], "active_band_count": 3, "active_band_mae_db": 4.40493496359237}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 4.015479902310721e-05, "log_rms_mae_db": 1.8442943924940813, "rms_envelope_nrmse": 0.3294158641011207, "envelope_correlation": 0.9335786774372226, "rms_gain_error_db": 0.9243423657433578, "reference_mean_power": 9.64997641207967e-05, "output_mean_power": 0.00011938794816424937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 5.087796481772889e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1685652820939687e-05, "output_power": 5.681720732663818e-05, "reference_active": true, "delta_db": 2.536172241727831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.336227531732081e-06, "output_power": 3.781817062035252e-06, "reference_active": true, "delta_db": 2.0918537922122544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1955736150290575e-07, "output_power": 1.0001205379613496e-06, "reference_active": true, "delta_db": 9.224760237366237}], "active_band_count": 3, "active_band_mae_db": 4.6175954237687735}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8357575962685241, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "18b3ea7b9ffea66bf14fe02ea09e8f0d65e185bf25ab877ff2713d2603111bcb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9894939638845668, "standard_deviation": 0.002598341297553376, "minimum": 0.9854252713126616, "maximum": 0.9930100254467243, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.85526299813735, "standard_deviation": 0.07250444610554128, "minimum": 1.7069283850897923, "maximum": 1.9215191173707957, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6253492354393856, "standard_deviation": 0.2379152986341622, "minimum": 0.22135147680394973, "maximum": 1.0183463177468042, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11909735322243709, "standard_deviation": 0.013842177195154997, "minimum": 0.09619175332968513, "maximum": 0.1398959786874067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974633871133602, "standard_deviation": 0.000960861240831419, "minimum": 0.9954570802514306, "maximum": 0.9983798096953254, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9901180472562765, "standard_deviation": 0.07124083588911352, "minimum": 0.8432957732357041, "maximum": 1.0706855625007805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.974976717004145, "standard_deviation": 0.34177782891938213, "minimum": 0.35726257063032624, "maximum": 1.3802038427670826, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06130879015308313, "standard_deviation": 0.008434013109138152, "minimum": 0.047506203300759976, "maximum": 0.07644570159114746, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8965026258404074, "standard_deviation": 0.002641874055003773, "minimum": 0.8925430003079275, "maximum": 0.9002998877865083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3443690886000006, "standard_deviation": 0.060297216132842, "minimum": 2.269002687363097, "maximum": 2.4173605602215935, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 4.695794639856273, "standard_deviation": 0.22683961930640478, "minimum": 4.40493496359237, "maximum": 5.062084427587308, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.4227476058652647, "standard_deviation": 0.008299399753037777, "minimum": 0.41184075618331906, "maximum": 0.4364483482643409, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9295884568537728, "standard_deviation": 0.003056894714522897, "minimum": 0.9253365992138503, "maximum": 0.9335786774372226, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8124697308321163, "standard_deviation": 0.04370994182898673, "minimum": 1.749648910430599, "maximum": 1.868559877482093, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 5.037280329351018, "standard_deviation": 0.22871146653294017, "minimum": 4.6175954237687735, "maximum": 5.388008704311548, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.34238839357782624, "standard_deviation": 0.01050885594816855, "minimum": 0.3294158641011207, "maximum": 0.35876924364924656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9863644441486933, "standard_deviation": 0.0011792895016192748, "minimum": 0.9848488892290193, "maximum": 0.9883405747411356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5031800268151652, "standard_deviation": 0.06952476068043194, "minimum": 1.3850381602461594, "maximum": 1.604052529626055, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6275762929775573, "standard_deviation": 0.17547935834798026, "minimum": 0.39459631095453657, "maximum": 0.8377984391271731, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12509866177671525, "standard_deviation": 0.005706858059870277, "minimum": 0.11705615840454514, "maximum": 0.13305867311318845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958606094018969, "standard_deviation": 0.0010529100596557886, "minimum": 0.9944002075977376, "maximum": 0.9972675993158294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9275435506673702, "standard_deviation": 0.06223744738394496, "minimum": 0.8187507516940634, "maximum": 1.0250073012158079, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6806857871957841, "standard_deviation": 0.1503260206665659, "minimum": 0.49740372190966936, "maximum": 0.9213315558810464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06937207479818332, "standard_deviation": 0.00910289465415238, "minimum": 0.053254039013918196, "maximum": 0.0797621302251828, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.9112554456274915, "standard_deviation": 0.040936676519637685, "minimum": 0.8439833339682937, "maximum": 0.9845060328113625, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.292354403762557, "standard_deviation": 0.023006461045876038, "minimum": 6.251694077786076, "maximum": 6.3191920253309535, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9581382367490041, "standard_deviation": 0.0294598270061871, "minimum": 0.9288831207812248, "maximum": 1.018172382845083, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.137588101599291, "standard_deviation": 0.08517744175371088, "minimum": -0.2304785823074443, "maximum": 0.03337798623371076, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9934817221983939, "standard_deviation": 0.001369190328415336, "minimum": 0.9915605957767808, "maximum": 0.995825055311682, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9408471607640867, "standard_deviation": 0.023923999765827854, "minimum": 0.89615779705776, "maximum": 0.9678631792855806, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.445660553765018, "standard_deviation": 0.19325117522874843, "minimum": 0.22728462957652443, "maximum": 0.6800866549184611, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08462764666387751, "standard_deviation": 0.0089375291900464, "minimum": 0.06728868402591401, "maximum": 0.09514055563227503, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982574585543126, "standard_deviation": 0.0004574839671307548, "minimum": 0.99749435640221, "maximum": 0.9987703815908735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4972208895040211, "standard_deviation": 0.016176142252358975, "minimum": 0.47212825316103585, "maximum": 0.5165113019771875, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6447992709938173, "standard_deviation": 0.1551575702183722, "minimum": 0.3536032013964912, "maximum": 0.8570292126769053, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04456172924542193, "standard_deviation": 0.00514806737142897, "minimum": 0.035124617927208134, "maximum": 0.04977673355563312, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0029", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.2596595287322998, "predicted_mos": 1.2596595287322998, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "b37d2e4a1efeb3d882c38d97c6ceec162679f72dc7462cd814665049e89895cd", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.513951499946415, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0323992368692203, "全帯域_10ms_rms_envelope_nrmse": 0.08174420583148541, "全帯域_10ms_envelope_correlation": 0.9938291105575725, "全帯域_10ms_modulation_mae_db": 0.18496403667695124, "全帯域_40ms_log_rms_mae_db": 0.5592867893855576, "全帯域_40ms_rms_envelope_nrmse": 0.044369953737143045, "全帯域_40ms_envelope_correlation": 0.998069724408746, "全帯域_40ms_modulation_mae_db": 0.16892438469382112, "300–1000 Hz_10ms_log_rms_mae_db": 1.2437190098512827, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11279887561557822, "300–1000 Hz_10ms_envelope_correlation": 0.9886364562050547, "300–1000 Hz_10ms_modulation_mae_db": 0.13705432154625827, "300–1000 Hz_40ms_log_rms_mae_db": 0.7219667352093065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06092843246505786, "300–1000 Hz_40ms_envelope_correlation": 0.9964890225651962, "300–1000 Hz_40ms_modulation_mae_db": 0.17636087685061755, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6059407210100176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1205991003430937, "1000–3000 Hz_10ms_envelope_correlation": 0.9892213908869012, "1000–3000 Hz_10ms_modulation_mae_db": 0.6802332114360117, "1000–3000 Hz_40ms_log_rms_mae_db": 0.941362112537038, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06966114526577256, "1000–3000 Hz_40ms_envelope_correlation": 0.9964653313951517, "1000–3000 Hz_40ms_modulation_mae_db": 0.7355540758132092, "3000–6000 Hz_10ms_log_rms_mae_db": 1.787625696662694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4065839053847036, "3000–6000 Hz_10ms_envelope_correlation": 0.9000196815112872, "3000–6000 Hz_10ms_modulation_mae_db": 4.303154518346673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.455923854213244, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3246524981409441, "3000–6000 Hz_40ms_envelope_correlation": 0.9326883302033497, "3000–6000 Hz_40ms_modulation_mae_db": 4.632238700055701, "mr_stft_db": 5.897637510398887, "band_mae_db": 0.8632996100432334, "rms_gain_error_db": -0.08633521808648537, "output_peak": 0.9887801740507222}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9962995071692808, "全帯域_10ms_rms_envelope_nrmse": 0.08239571561990168, "全帯域_10ms_envelope_correlation": 0.9938792701340682, "全帯域_10ms_modulation_mae_db": 0.3436261326066456, "全帯域_40ms_log_rms_mae_db": 0.5484077523486206, "全帯域_40ms_rms_envelope_nrmse": 0.044886300453682464, "全帯域_40ms_envelope_correlation": 0.9980904028730191, "全帯域_40ms_modulation_mae_db": 0.1858509390269003, "300–1000 Hz_10ms_log_rms_mae_db": 1.2770688300342727, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10272841462375559, "300–1000 Hz_10ms_envelope_correlation": 0.9905695565702348, "300–1000 Hz_10ms_modulation_mae_db": 0.6954311516091783, "300–1000 Hz_40ms_log_rms_mae_db": 0.7099304281186535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04181732454713716, "300–1000 Hz_40ms_envelope_correlation": 0.9982918368108546, "300–1000 Hz_40ms_modulation_mae_db": 0.2822008275463544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5811968589529144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12283155908279653, "1000–3000 Hz_10ms_envelope_correlation": 0.9890483557344123, "1000–3000 Hz_10ms_modulation_mae_db": 0.44599855805923355, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8897651176952506, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06977088875039285, "1000–3000 Hz_40ms_envelope_correlation": 0.9963471057380345, "1000–3000 Hz_40ms_modulation_mae_db": 0.2610390173214894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7209220414772213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4117105209567694, "3000–6000 Hz_10ms_envelope_correlation": 0.9009215280187512, "3000–6000 Hz_10ms_modulation_mae_db": 4.815075813846028, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5176660755832863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3335934337914343, "3000–6000 Hz_40ms_envelope_correlation": 0.931651825665828, "3000–6000 Hz_40ms_modulation_mae_db": 4.955361727208138, "mr_stft_db": 5.911742067485385, "band_mae_db": 0.8853183619382085, "rms_gain_error_db": 0.09867653186350858, "output_peak": 0.9136068154415073}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9776453665568765, "全帯域_10ms_rms_envelope_nrmse": 0.08520450092146258, "全帯域_10ms_envelope_correlation": 0.9932460740370986, "全帯域_10ms_modulation_mae_db": 0.48601237019561766, "全帯域_40ms_log_rms_mae_db": 0.5828611379960975, "全帯域_40ms_rms_envelope_nrmse": 0.04847702921879638, "全帯域_40ms_envelope_correlation": 0.9976282415058412, "全帯域_40ms_modulation_mae_db": 0.5249201769922585, "300–1000 Hz_10ms_log_rms_mae_db": 1.2520573373052584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11420856642864793, "300–1000 Hz_10ms_envelope_correlation": 0.9884552564970976, "300–1000 Hz_10ms_modulation_mae_db": 0.3634361779609045, "300–1000 Hz_40ms_log_rms_mae_db": 0.7550364919239014, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06448678890806879, "300–1000 Hz_40ms_envelope_correlation": 0.9962188458063101, "300–1000 Hz_40ms_modulation_mae_db": 0.7360245621018237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.496599381775019, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11503511837432287, "1000–3000 Hz_10ms_envelope_correlation": 0.9899779683884081, "1000–3000 Hz_10ms_modulation_mae_db": 0.47234763937026875, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8301747519968342, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06097174561546023, "1000–3000 Hz_40ms_envelope_correlation": 0.9969856254710348, "1000–3000 Hz_40ms_modulation_mae_db": 0.4769849031992166, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7411170784954597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39883020742516795, "3000–6000 Hz_10ms_envelope_correlation": 0.9043758487705392, "3000–6000 Hz_10ms_modulation_mae_db": 4.141772553058138, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4325799823057277, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32364529660199165, "3000–6000 Hz_40ms_envelope_correlation": 0.9335966116482002, "3000–6000 Hz_40ms_modulation_mae_db": 4.330464310680065, "mr_stft_db": 5.885668487858024, "band_mae_db": 0.8071561459081792, "rms_gain_error_db": -0.06532664920806774, "output_peak": 0.9549923814093553}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9837591673638845, "全帯域_10ms_rms_envelope_nrmse": 0.0832063973547107, "全帯域_10ms_envelope_correlation": 0.9936301284138224, "全帯域_10ms_modulation_mae_db": 0.4304632989273389, "全帯域_40ms_log_rms_mae_db": 0.5966510104858981, "全帯域_40ms_rms_envelope_nrmse": 0.04186662811005531, "全帯域_40ms_envelope_correlation": 0.9983602167822894, "全帯域_40ms_modulation_mae_db": 0.6231492668173061, "300–1000 Hz_10ms_log_rms_mae_db": 1.2868573054915937, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11939732859744262, "300–1000 Hz_10ms_envelope_correlation": 0.9876279904295949, "300–1000 Hz_10ms_modulation_mae_db": 0.8191105834226308, "300–1000 Hz_40ms_log_rms_mae_db": 0.7944731871846903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06442616156747917, "300–1000 Hz_40ms_envelope_correlation": 0.9965889483743673, "300–1000 Hz_40ms_modulation_mae_db": 0.6052482600628422, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5654845633607661, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11321602073828889, "1000–3000 Hz_10ms_envelope_correlation": 0.9904192830987139, "1000–3000 Hz_10ms_modulation_mae_db": 0.5799757024105718, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9061637288395378, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057623794157791654, "1000–3000 Hz_40ms_envelope_correlation": 0.9975449521216011, "1000–3000 Hz_40ms_modulation_mae_db": 0.9076500650602354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7933903793470936, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41124466378792107, "3000–6000 Hz_10ms_envelope_correlation": 0.9012033736269651, "3000–6000 Hz_10ms_modulation_mae_db": 4.501416475912268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4511406150241477, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32724665376911133, "3000–6000 Hz_40ms_envelope_correlation": 0.9348374750430506, "3000–6000 Hz_40ms_modulation_mae_db": 4.750947572565749, "mr_stft_db": 5.905456805152329, "band_mae_db": 0.876585627244725, "rms_gain_error_db": -0.10957320720111867, "output_peak": 0.976606988436263}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0178116508426056, "全帯域_10ms_rms_envelope_nrmse": 0.0902082645472812, "全帯域_10ms_envelope_correlation": 0.9928882352820849, "全帯域_10ms_modulation_mae_db": 0.4881388355922336, "全帯域_40ms_log_rms_mae_db": 0.4672159525003696, "全帯域_40ms_rms_envelope_nrmse": 0.04477669512747552, "全帯域_40ms_envelope_correlation": 0.9986184519005369, "全帯域_40ms_modulation_mae_db": 0.5810835959275026, "300–1000 Hz_10ms_log_rms_mae_db": 1.2843776888564533, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12329049928438747, "300–1000 Hz_10ms_envelope_correlation": 0.9867532736037333, "300–1000 Hz_10ms_modulation_mae_db": 0.42793830642706904, "300–1000 Hz_40ms_log_rms_mae_db": 0.6794128192646257, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05966889887178445, "300–1000 Hz_40ms_envelope_correlation": 0.9972277081422258, "300–1000 Hz_40ms_modulation_mae_db": 0.425817860910011, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6163086768395176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1315371016165686, "1000–3000 Hz_10ms_envelope_correlation": 0.9876554709197288, "1000–3000 Hz_10ms_modulation_mae_db": 0.936414788843523, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8891354009111847, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06907938817039176, "1000–3000 Hz_40ms_envelope_correlation": 0.9973171025236202, "1000–3000 Hz_40ms_modulation_mae_db": 0.6119623923262354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7583505901066272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39754378293056675, "3000–6000 Hz_10ms_envelope_correlation": 0.9041456836260948, "3000–6000 Hz_10ms_modulation_mae_db": 4.547923032170287, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4352027830817722, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32075872804020283, "3000–6000 Hz_40ms_envelope_correlation": 0.9340278169861564, "3000–6000 Hz_40ms_modulation_mae_db": 4.707763052325945, "mr_stft_db": 5.915980095824864, "band_mae_db": 0.9136015719627352, "rms_gain_error_db": -0.22888797312616674, "output_peak": 0.9685265660407676}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0697279927306058, "全帯域_10ms_rms_envelope_nrmse": 0.0938917879806681, "全帯域_10ms_envelope_correlation": 0.991798612871149, "全帯域_10ms_modulation_mae_db": 0.5649561984382271, "全帯域_40ms_log_rms_mae_db": 0.5446647030691983, "全帯域_40ms_rms_envelope_nrmse": 0.042528369986180314, "全帯域_40ms_envelope_correlation": 0.9982104555404168, "全帯域_40ms_modulation_mae_db": 0.501623518501109, "300–1000 Hz_10ms_log_rms_mae_db": 1.413229985259003, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12564058267723016, "300–1000 Hz_10ms_envelope_correlation": 0.9859831461726072, "300–1000 Hz_10ms_modulation_mae_db": 0.31430620507194523, "300–1000 Hz_40ms_log_rms_mae_db": 0.8704161912484726, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06925076943412277, "300–1000 Hz_40ms_envelope_correlation": 0.995564703427151, "300–1000 Hz_40ms_modulation_mae_db": 0.4110176078231987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.766419707522182, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1487747914022013, "1000–3000 Hz_10ms_envelope_correlation": 0.9832397670003695, "1000–3000 Hz_10ms_modulation_mae_db": 0.30698023493968624, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9591761040202879, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06543054102169997, "1000–3000 Hz_40ms_envelope_correlation": 0.9967864603968314, "1000–3000 Hz_40ms_modulation_mae_db": 0.2634541783243868, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8139804850080796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4016404353726633, "3000–6000 Hz_10ms_envelope_correlation": 0.9074197851572711, "3000–6000 Hz_10ms_modulation_mae_db": 4.556604689998401, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4557485662468335, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3201776913843883, "3000–6000 Hz_40ms_envelope_correlation": 0.9393193577669555, "3000–6000 Hz_40ms_modulation_mae_db": 4.661965102851418, "mr_stft_db": 5.921647582496578, "band_mae_db": 0.8446540199381599, "rms_gain_error_db": -0.07808308877967644, "output_peak": 0.9612532190295493}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0618590949591824, "全帯域_10ms_rms_envelope_nrmse": 0.08700874065668461, "全帯域_10ms_envelope_correlation": 0.9929612712234086, "全帯域_10ms_modulation_mae_db": 0.33004992381284826, "全帯域_40ms_log_rms_mae_db": 0.5459240730194612, "全帯域_40ms_rms_envelope_nrmse": 0.042020755610412, "全帯域_40ms_envelope_correlation": 0.9982379557757912, "全帯域_40ms_modulation_mae_db": 0.7651575214006091, "300–1000 Hz_10ms_log_rms_mae_db": 1.2694213927139768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1218720377384775, "300–1000 Hz_10ms_envelope_correlation": 0.9867304883175086, "300–1000 Hz_10ms_modulation_mae_db": 0.6244083764144496, "300–1000 Hz_40ms_log_rms_mae_db": 0.680619746684192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05424514713835635, "300–1000 Hz_40ms_envelope_correlation": 0.9972908330466269, "300–1000 Hz_40ms_modulation_mae_db": 0.3354020220161278, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6590547749726148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11778137596093236, "1000–3000 Hz_10ms_envelope_correlation": 0.9895895258587121, "1000–3000 Hz_10ms_modulation_mae_db": 0.446050398477287, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9548838782608636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05863903470142073, "1000–3000 Hz_40ms_envelope_correlation": 0.9974203237143808, "1000–3000 Hz_40ms_modulation_mae_db": 0.7582215570163098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.785599427087395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3940230814650414, "3000–6000 Hz_10ms_envelope_correlation": 0.9067075171795244, "3000–6000 Hz_10ms_modulation_mae_db": 4.324140546970316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4612301903029719, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.318312054447974, "3000–6000 Hz_40ms_envelope_correlation": 0.9358935859802316, "3000–6000 Hz_40ms_modulation_mae_db": 4.51483142664162, "mr_stft_db": 5.892789415511571, "band_mae_db": 0.878838131088493, "rms_gain_error_db": -0.06258417917101392, "output_peak": 0.9448564230569212}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9405673302235211, "全帯域_10ms_rms_envelope_nrmse": 0.08847147749794428, "全帯域_10ms_envelope_correlation": 0.9927325436590518, "全帯域_10ms_modulation_mae_db": 0.35577954241183246, "全帯域_40ms_log_rms_mae_db": 0.5101983003293918, "全帯域_40ms_rms_envelope_nrmse": 0.035364872814224256, "全帯域_40ms_envelope_correlation": 0.998728392627627, "全帯域_40ms_modulation_mae_db": 0.5538705184261222, "300–1000 Hz_10ms_log_rms_mae_db": 1.2864108918444181, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12291145293560453, "300–1000 Hz_10ms_envelope_correlation": 0.9865599745072112, "300–1000 Hz_10ms_modulation_mae_db": 0.4350607587091046, "300–1000 Hz_40ms_log_rms_mae_db": 0.7055353572533012, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05024809974406305, "300–1000 Hz_40ms_envelope_correlation": 0.9975342509824956, "300–1000 Hz_40ms_modulation_mae_db": 0.3075088192269893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.528296687612869, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09853761166618287, "1000–3000 Hz_10ms_envelope_correlation": 0.9927429913249215, "1000–3000 Hz_10ms_modulation_mae_db": 0.9931652182914331, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8459284692585763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04193306104843518, "1000–3000 Hz_40ms_envelope_correlation": 0.9987609778318675, "1000–3000 Hz_40ms_modulation_mae_db": 1.1179868676795486, "3000–6000 Hz_10ms_log_rms_mae_db": 1.716474842985289, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4009651167170017, "3000–6000 Hz_10ms_envelope_correlation": 0.905025169083495, "3000–6000 Hz_10ms_modulation_mae_db": 4.461453792292775, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4679215570121802, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3288054688740074, "3000–6000 Hz_40ms_envelope_correlation": 0.9327170268463155, "3000–6000 Hz_40ms_modulation_mae_db": 4.648543156440458, "mr_stft_db": 5.915899429998393, "band_mae_db": 0.824208921411662, "rms_gain_error_db": 0.01549637889442475, "output_peak": 0.9470550491591003}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 33040, "output_samples": 33040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7150994021388115, "spectral_convergence": 0.3326463101208378, "spectral_convergence_status": "ok", "amplitude_floor": 1.5756171103077008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.528962581056885, "spectral_convergence": 0.3369043588818854, "spectral_convergence_status": "ok", "amplitude_floor": 1.2505050051947005e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.448850548000964, "spectral_convergence": 0.37051741654063236, "spectral_convergence_status": "ok", "amplitude_floor": 9.121352204658788e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021685618865241799, "output_power": 0.000224574936622039, "delta_db": 0.1518946521095384, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010795690773652512, "output_power": 0.0011025743776736608, "delta_db": 0.09157459498964697, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002989928450689294, "output_power": 0.002929935459550548, "delta_db": -0.0880274188927064, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0020580355715967367, "output_power": 0.001971914363833976, "delta_db": -0.18564826441914956, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0009751360590152492, "output_power": 0.0009123770779802826, "delta_db": -0.2889085048472932, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.9392123549665832e-05, "output_power": 5.3088418989719416e-05, "delta_db": 4.373744225001066, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8632996100432334, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08633521808648537, "pitch": null, "output_rms_dbfs": -21.430015179008013, "output_peak": 0.9887801740507222, "reference_rms_dbfs": -21.343679960921527, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.897637510398887}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 33040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00032988134947798775, "log_rms_mae_db": 1.0323992368692203, "rms_envelope_nrmse": 0.08174420583148541, "envelope_correlation": 0.9938291105575725, "rms_gain_error_db": -0.0863351135065623, "reference_mean_power": 0.007356724541990913, "output_mean_power": 0.00721192138061659}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 2.980285612038863e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020727352824950612, "output_power": 0.002039495872501055, "reference_active": true, "delta_db": -0.0702100937411258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00028987718896809884, "output_power": 0.0002711902648046354, "reference_active": true, "delta_db": -0.2893994601409067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.012977878291207e-05, "output_power": 6.704620696536375e-05, "reference_active": true, "delta_db": -0.19528255614882128}], "active_band_count": 3, "active_band_mae_db": 0.18496403667695124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0002960474071300256, "log_rms_mae_db": 0.5592867893855576, "rms_envelope_nrmse": 0.044369953737143045, "envelope_correlation": 0.998069724408746, "rms_gain_error_db": -0.08633693290970983, "reference_mean_power": 0.007465415825536721, "output_mean_power": 0.0073184702165651015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 2.5597102021075326e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001877761726334087, "output_power": 0.0018478983966051434, "reference_active": true, "delta_db": -0.06962394129511021}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012702101262043316, "output_power": 0.0001240761009890485, "reference_active": true, "delta_db": -0.10187433045239519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.962314408759724e-06, "output_power": 3.667933434453381e-06, "reference_active": true, "delta_db": -0.33527488233395797}], "active_band_count": 3, "active_band_mae_db": 0.16892438469382112}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00021933697126022545, "log_rms_mae_db": 1.2437190098512827, "rms_envelope_nrmse": 0.11279887561557822, "envelope_correlation": 0.9886364562050547, "rms_gain_error_db": -0.13347553048990843, "reference_mean_power": 0.003363603458339666, "output_mean_power": 0.0032617993157494354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.3609823101859838e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008546335077009212, "output_power": 0.0008363364708402205, "reference_active": true, "delta_db": -0.09398880393042175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016660612815788933, "output_power": 0.00015952044195903937, "reference_active": true, "delta_db": -0.18874627514927372}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.188726605138031e-05, "output_power": 4.314443676310498e-05, "reference_active": true, "delta_db": 0.12842788555907936}], "active_band_count": 3, "active_band_mae_db": 0.13705432154625827}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0001893291578274055, "log_rms_mae_db": 0.7219667352093065, "rms_envelope_nrmse": 0.06092843246505786, "envelope_correlation": 0.9964890225651962, "rms_gain_error_db": -0.13347696766646522, "reference_mean_power": 0.0034133051942474998, "output_mean_power": 0.0033099956640577867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.15224102535942e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007711766819603955, "output_power": 0.0007549418514062898, "reference_active": true, "delta_db": -0.09240387501282117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.304889773698623e-05, "output_power": 7.174990245656857e-05, "reference_active": true, "delta_db": -0.07792352096209845}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527118205585199e-06, "output_power": 2.3267516630529413e-06, "reference_active": true, "delta_db": -0.35875523457693304}], "active_band_count": 3, "active_band_mae_db": 0.17636087685061755}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.0002256735995145706, "log_rms_mae_db": 1.6059407210100176, "rms_envelope_nrmse": 0.1205991003430937, "envelope_correlation": 0.9892213908869012, "rms_gain_error_db": -0.21571888330209774, "reference_mean_power": 0.002711357875845494, "output_mean_power": 0.002579971783131778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.6613986227009806e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012279896352023905, "output_power": 0.0011217569817057109, "reference_active": true, "delta_db": -0.392959199554494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888327196960883, "output_power": 0.0001373144666170604, "reference_active": true, "delta_db": 0.6259554480414399}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.234146057131384e-05, "output_power": 4.092023581884521e-05, "reference_active": true, "delta_db": 1.0217849867121012}], "active_band_count": 3, "active_band_mae_db": 0.6802332114360117}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.00020374269470838238, "log_rms_mae_db": 0.941362112537038, "rms_envelope_nrmse": 0.06966114526577256, "envelope_correlation": 0.9964653313951517, "rms_gain_error_db": -0.21572123330876708, "reference_mean_power": 0.002751412553659036, "output_mean_power": 0.002618084087805539}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.4886745667367597e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001137287164515336, "output_power": 0.0010546914039499063, "reference_active": true, "delta_db": -0.3274473130754085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.907501205097428e-05, "output_power": 6.113603503889531e-05, "reference_active": true, "delta_db": 0.9543685433317651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5268858664777347e-06, "output_power": 3.126585168475499e-06, "reference_active": true, "delta_db": 0.924846371032454}], "active_band_count": 3, "active_band_mae_db": 0.7355540758132092}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 4.718157231228245e-05, "log_rms_mae_db": 1.787625696662694, "rms_envelope_nrmse": 0.4065839053847036, "envelope_correlation": 0.9000196815112872, "rms_gain_error_db": 0.8036815993238271, "reference_mean_power": 9.509723574878595e-05, "output_mean_power": 0.00011442898695870997}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 5.720070068605411e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.50238140104046e-05, "output_power": 4.684065605876517e-05, "reference_active": true, "delta_db": 1.2625953128727716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.975419310826897e-06, "output_power": 6.998001439204974e-06, "reference_active": true, "delta_db": 1.4814434002923398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.471999704474668e-06, "output_power": 1.529150415260247e-05, "reference_active": true, "delta_db": 10.16542484187491}], "active_band_count": 3, "active_band_mae_db": 4.303154518346673}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 4.015479902310721e-05, "log_rms_mae_db": 1.455923854213244, "rms_envelope_nrmse": 0.3246524981409441, "envelope_correlation": 0.9326883302033497, "rms_gain_error_db": 0.8037087298432664, "reference_mean_power": 9.64997641207967e-05, "output_mean_power": 0.00011611735236483584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 5.087796481772889e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1685652820939687e-05, "output_power": 5.4652730680062047e-05, "reference_active": true, "delta_db": 2.367492066626352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.336227531732081e-06, "output_power": 3.770823847011852e-06, "reference_active": true, "delta_db": 2.0792110727922144}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1955736150290575e-07, "output_power": 1.0533619296866297e-06, "reference_active": true, "delta_db": 9.450012960748536}], "active_band_count": 3, "active_band_mae_db": 4.632238700055701}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8424010135953403, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "eb0e4023364d0b228f589a954aaf8d81a09ddb18aef4cc4d312dcfea4e44f379", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9889868441515209, "standard_deviation": 0.0027351239650238563, "minimum": 0.9832397670003695, "maximum": 0.9927429913249215, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6024126715057374, "standard_deviation": 0.08350843333895, "minimum": 1.496599381775019, "maximum": 1.766419707522182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6076457189785018, "standard_deviation": 0.24602471287241048, "minimum": 0.30698023493968624, "maximum": 0.9931652182914331, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12103908489804838, "standard_deviation": 0.01462415434628362, "minimum": 0.09853761166618287, "maximum": 0.1487747914022013, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972034848990653, "standard_deviation": 0.0007655007851880681, "minimum": 0.9963471057380345, "maximum": 0.9987609778318675, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9020736954399466, "standard_deviation": 0.04818202630066086, "minimum": 0.8301747519968342, "maximum": 0.9591761040202879, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6416066320925788, "standard_deviation": 0.3014139257040643, "minimum": 0.2610390173214894, "maximum": 1.1179868676795486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06163869984142062, "standard_deviation": 0.009366418099961249, "minimum": 0.04193306104843518, "maximum": 0.06977088875039285, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.903727323371741, "standard_deviation": 0.0027460428983950213, "minimum": 0.9000196815112872, "maximum": 0.9074197851572711, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7646825676462323, "standard_deviation": 0.03596172543307331, "minimum": 1.716474842985289, "maximum": 1.8139804850080796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 4.456442677824361, "standard_deviation": 0.20321872981536052, "minimum": 4.141772553058138, "maximum": 4.815075813846028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.40281771425497936, "standard_deviation": 0.006438255270962067, "minimum": 0.3940230814650414, "maximum": 0.4117105209567694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9343415037675109, "standard_deviation": 0.002411090956351773, "minimum": 0.931651825665828, "maximum": 0.9393193577669555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4596767029712705, "standard_deviation": 0.026372990057287257, "minimum": 1.4325799823057277, "maximum": 1.5176660755832863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.650264381096136, "standard_deviation": 0.18031977853388748, "minimum": 4.330464310680065, "maximum": 4.955361727208138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3246489781312567, "standard_deviation": 0.005075403752905208, "minimum": 0.318312054447974, "maximum": 0.3335934337914343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9876645177878803, "standard_deviation": 0.001500491345721846, "minimum": 0.9859831461726072, "maximum": 0.9905695565702348, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2891428051695324, "standard_deviation": 0.05265302806088582, "minimum": 1.2437190098512827, "maximum": 1.413229985259003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47709323514519253, "standard_deviation": 0.22218911050781703, "minimum": 0.13705432154625827, "maximum": 0.8191105834226308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1178559697376405, "standard_deviation": 0.007577386331165147, "minimum": 0.10272841462375559, "maximum": 0.12564058267723016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969007686444035, "standard_deviation": 0.0008545897867999894, "minimum": 0.995564703427151, "maximum": 0.9982918368108546, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7396738696108929, "standard_deviation": 0.06526316682484756, "minimum": 0.6794128192646257, "maximum": 0.8704161912484726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.40994760456724555, "standard_deviation": 0.18191610516932155, "minimum": 0.17636087685061755, "maximum": 0.7360245621018237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0581339528345087, "standard_deviation": 0.008922211860166939, "minimum": 0.04181732454713716, "maximum": 0.06925076943412277, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.8617077986919246, "standard_deviation": 0.03472470163078959, "minimum": 0.8071561459081792, "maximum": 0.9136015719627352, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.905852674340754, "standard_deviation": 0.012724339930329136, "minimum": 5.885668487858024, "maximum": 5.921647582496578, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9569597020780233, "standard_deviation": 0.02292447092516956, "minimum": 0.9136068154415073, "maximum": 0.9887801740507222, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06457717560182444, "standard_deviation": 0.09467733476135977, "minimum": -0.22888797312616674, "maximum": 0.09867653186350858, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993120655772282, "standard_deviation": 0.0006897771078805249, "minimum": 0.991798612871149, "maximum": 0.9938792701340682, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0100086683393972, "standard_deviation": 0.04401874387198483, "minimum": 0.9405673302235211, "maximum": 1.0697279927306058, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.39799879233271185, "standard_deviation": 0.1192476678232447, "minimum": 0.18496403667695124, "maximum": 0.5649561984382271, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08651638630126732, "standard_deviation": 0.004217684074923232, "minimum": 0.08174420583148541, "maximum": 0.0938917879806681, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982429801767835, "standard_deviation": 0.00034281040352465215, "minimum": 0.9976282415058412, "maximum": 0.998728392627627, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5444012148918244, "standard_deviation": 0.040666996294133693, "minimum": 0.4672159525003696, "maximum": 0.5966510104858981, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.48807249022320365, "standard_deviation": 0.20796189830814057, "minimum": 0.16892438469382112, "maximum": 0.7651575214006091, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04303632563224616, "standard_deviation": 0.0037671492499334394, "minimum": 0.035364872814224256, "maximum": 0.04847702921879638, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0029", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.4635074138641357, "predicted_mos": 1.4635074138641357, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "3cbf2ed618eb1f85220002b6e407cb35b0412474ab07099eb6b5251f0050647c", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.549660900142044, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9393330528090001, "全帯域_10ms_rms_envelope_nrmse": 0.08343826727297636, "全帯域_10ms_envelope_correlation": 0.9935752514795789, "全帯域_10ms_modulation_mae_db": 0.5881710148248279, "全帯域_40ms_log_rms_mae_db": 0.4768737778726002, "全帯域_40ms_rms_envelope_nrmse": 0.035138663983703196, "全帯域_40ms_envelope_correlation": 0.9988663281248787, "全帯域_40ms_modulation_mae_db": 0.65089723101378, "300–1000 Hz_10ms_log_rms_mae_db": 1.4275678433607153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12879107814327628, "300–1000 Hz_10ms_envelope_correlation": 0.9852078181684387, "300–1000 Hz_10ms_modulation_mae_db": 0.7084057126240926, "300–1000 Hz_40ms_log_rms_mae_db": 0.8947400891992194, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06931174468958719, "300–1000 Hz_40ms_envelope_correlation": 0.9954548832223276, "300–1000 Hz_40ms_modulation_mae_db": 0.5984729361084659, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7044475323024288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11415862623035511, "1000–3000 Hz_10ms_envelope_correlation": 0.9909674061641267, "1000–3000 Hz_10ms_modulation_mae_db": 0.6043892535705532, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9135264772085459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055231874387190835, "1000–3000 Hz_40ms_envelope_correlation": 0.9987051984467252, "1000–3000 Hz_40ms_modulation_mae_db": 0.9464521977452828, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0070869724094558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4106760456570007, "3000–6000 Hz_10ms_envelope_correlation": 0.8961359607517009, "3000–6000 Hz_10ms_modulation_mae_db": 4.695434515996749, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6328792338935796, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33303322148954384, "3000–6000 Hz_40ms_envelope_correlation": 0.9274429860995922, "3000–6000 Hz_40ms_modulation_mae_db": 5.1087600236815005, "mr_stft_db": 6.0253572925479615, "band_mae_db": 0.8560761133915635, "rms_gain_error_db": -0.10277972981978822, "output_peak": 0.9612942841299922}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9772916475791857, "全帯域_10ms_rms_envelope_nrmse": 0.07611662211061346, "全帯域_10ms_envelope_correlation": 0.994619141626431, "全帯域_10ms_modulation_mae_db": 0.3850237922968165, "全帯域_40ms_log_rms_mae_db": 0.49347073314965156, "全帯域_40ms_rms_envelope_nrmse": 0.033716941682731025, "全帯域_40ms_envelope_correlation": 0.9988693262129856, "全帯域_40ms_modulation_mae_db": 0.37036536183228996, "300–1000 Hz_10ms_log_rms_mae_db": 1.3830387552852594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11220399878206232, "300–1000 Hz_10ms_envelope_correlation": 0.9887579970299811, "300–1000 Hz_10ms_modulation_mae_db": 0.839045660979666, "300–1000 Hz_40ms_log_rms_mae_db": 0.8377007206256353, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05343081028220292, "300–1000 Hz_40ms_envelope_correlation": 0.9973031065864879, "300–1000 Hz_40ms_modulation_mae_db": 0.4697652565017991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.696021031003438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1262545781683977, "1000–3000 Hz_10ms_envelope_correlation": 0.9879566528901096, "1000–3000 Hz_10ms_modulation_mae_db": 0.64778625562373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9071563106027845, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05530178533280629, "1000–3000 Hz_40ms_envelope_correlation": 0.9976111181074905, "1000–3000 Hz_40ms_modulation_mae_db": 0.7567751522192853, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9620159961837909, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40838427996441107, "3000–6000 Hz_10ms_envelope_correlation": 0.8974659836605917, "3000–6000 Hz_10ms_modulation_mae_db": 4.584692618755674, "3000–6000 Hz_40ms_log_rms_mae_db": 1.538459456506385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32630253839197154, "3000–6000 Hz_40ms_envelope_correlation": 0.930604491354405, "3000–6000 Hz_40ms_modulation_mae_db": 4.650363534915163, "mr_stft_db": 6.037438846591321, "band_mae_db": 0.8464038271036186, "rms_gain_error_db": -0.057169587443785516, "output_peak": 0.9939251791245401}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9621759651086558, "全帯域_10ms_rms_envelope_nrmse": 0.09274364693563308, "全帯域_10ms_envelope_correlation": 0.9921630051310864, "全帯域_10ms_modulation_mae_db": 0.6367631215496021, "全帯域_40ms_log_rms_mae_db": 0.5351304448924022, "全帯域_40ms_rms_envelope_nrmse": 0.05037483426938323, "全帯域_40ms_envelope_correlation": 0.997684069818208, "全帯域_40ms_modulation_mae_db": 0.5297040524344809, "300–1000 Hz_10ms_log_rms_mae_db": 1.3668076030805245, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11631350484858487, "300–1000 Hz_10ms_envelope_correlation": 0.9879621157229903, "300–1000 Hz_10ms_modulation_mae_db": 0.55734750505268, "300–1000 Hz_40ms_log_rms_mae_db": 0.8379711972168657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05638090612989305, "300–1000 Hz_40ms_envelope_correlation": 0.9971285587811632, "300–1000 Hz_40ms_modulation_mae_db": 0.3968401710634337, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7721735539958248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13543106450263195, "1000–3000 Hz_10ms_envelope_correlation": 0.9866392919613862, "1000–3000 Hz_10ms_modulation_mae_db": 1.1675986653599073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0034174715777304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07811641054208934, "1000–3000 Hz_40ms_envelope_correlation": 0.9959138593357004, "1000–3000 Hz_40ms_modulation_mae_db": 0.9340070250312454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9111421261441082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4073793056236417, "3000–6000 Hz_10ms_envelope_correlation": 0.9077761932496283, "3000–6000 Hz_10ms_modulation_mae_db": 4.721820914617619, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5147254229506029, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3254226274555604, "3000–6000 Hz_40ms_envelope_correlation": 0.9397574390687589, "3000–6000 Hz_40ms_modulation_mae_db": 5.073960628039264, "mr_stft_db": 6.022056890372458, "band_mae_db": 0.9005214462322012, "rms_gain_error_db": -0.15014964670393793, "output_peak": 0.9461226283001858}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9312669776454522, "全帯域_10ms_rms_envelope_nrmse": 0.08640938069499739, "全帯域_10ms_envelope_correlation": 0.993216679477481, "全帯域_10ms_modulation_mae_db": 0.23407319599096443, "全帯域_40ms_log_rms_mae_db": 0.5108327473277496, "全帯域_40ms_rms_envelope_nrmse": 0.0481845638662404, "全帯域_40ms_envelope_correlation": 0.997886151174451, "全帯域_40ms_modulation_mae_db": 0.4407155061048287, "300–1000 Hz_10ms_log_rms_mae_db": 1.467958613096897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11780695189069762, "300–1000 Hz_10ms_envelope_correlation": 0.9876487171713352, "300–1000 Hz_10ms_modulation_mae_db": 0.506101588477465, "300–1000 Hz_40ms_log_rms_mae_db": 0.8884725611817804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05700896020590949, "300–1000 Hz_40ms_envelope_correlation": 0.9970170834693799, "300–1000 Hz_40ms_modulation_mae_db": 0.5139241391850513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7184877120902846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13042632904977408, "1000–3000 Hz_10ms_envelope_correlation": 0.9879435962065005, "1000–3000 Hz_10ms_modulation_mae_db": 0.43981303754888207, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0101281187273599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08362638221260281, "1000–3000 Hz_40ms_envelope_correlation": 0.9954627959342295, "1000–3000 Hz_40ms_modulation_mae_db": 0.546003988358496, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9790780922937998, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4179876309280853, "3000–6000 Hz_10ms_envelope_correlation": 0.9018523653820129, "3000–6000 Hz_10ms_modulation_mae_db": 4.820808556350161, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5986503416694684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3370971572284813, "3000–6000 Hz_40ms_envelope_correlation": 0.9343150826245926, "3000–6000 Hz_40ms_modulation_mae_db": 5.16508823787698, "mr_stft_db": 6.012366487182626, "band_mae_db": 0.935211750399109, "rms_gain_error_db": -0.14735800567984503, "output_peak": 0.9290067338210087}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9886201851691616, "全帯域_10ms_rms_envelope_nrmse": 0.0810626033207693, "全帯域_10ms_envelope_correlation": 0.9939323296443423, "全帯域_10ms_modulation_mae_db": 0.4047406038833201, "全帯域_40ms_log_rms_mae_db": 0.5265868778622173, "全帯域_40ms_rms_envelope_nrmse": 0.0380133482454154, "全帯域_40ms_envelope_correlation": 0.9986229393596846, "全帯域_40ms_modulation_mae_db": 0.41532884916311313, "300–1000 Hz_10ms_log_rms_mae_db": 1.5002566980940082, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1166092063792039, "300–1000 Hz_10ms_envelope_correlation": 0.9880108231659135, "300–1000 Hz_10ms_modulation_mae_db": 0.4863822746647272, "300–1000 Hz_40ms_log_rms_mae_db": 0.9424336985613835, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06255361943800711, "300–1000 Hz_40ms_envelope_correlation": 0.9965455276274257, "300–1000 Hz_40ms_modulation_mae_db": 0.4697668129858645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7380712832846332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12348746310783756, "1000–3000 Hz_10ms_envelope_correlation": 0.9885272873543741, "1000–3000 Hz_10ms_modulation_mae_db": 0.3860821463934255, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9073647480843191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05993030345388022, "1000–3000 Hz_40ms_envelope_correlation": 0.9972956436627421, "1000–3000 Hz_40ms_modulation_mae_db": 0.6462863753800786, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9630623909128724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41053018116082685, "3000–6000 Hz_10ms_envelope_correlation": 0.9013712726664922, "3000–6000 Hz_10ms_modulation_mae_db": 4.570024175609755, "3000–6000 Hz_40ms_log_rms_mae_db": 1.545695918087953, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32809160279310756, "3000–6000 Hz_40ms_envelope_correlation": 0.9343928371657853, "3000–6000 Hz_40ms_modulation_mae_db": 4.784164784647732, "mr_stft_db": 6.015772455637998, "band_mae_db": 0.8651217651281323, "rms_gain_error_db": -0.09525006827663841, "output_peak": 0.9324447426201572}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9756478020522582, "全帯域_10ms_rms_envelope_nrmse": 0.08506234162978307, "全帯域_10ms_envelope_correlation": 0.9934023978797591, "全帯域_10ms_modulation_mae_db": 0.48215858657615157, "全帯域_40ms_log_rms_mae_db": 0.4997596939860024, "全帯域_40ms_rms_envelope_nrmse": 0.04214372061517662, "全帯域_40ms_envelope_correlation": 0.9984154293919001, "全帯域_40ms_modulation_mae_db": 0.4874290184132075, "300–1000 Hz_10ms_log_rms_mae_db": 1.4470921746989203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1265821790112905, "300–1000 Hz_10ms_envelope_correlation": 0.9857294615185435, "300–1000 Hz_10ms_modulation_mae_db": 0.23832644268320788, "300–1000 Hz_40ms_log_rms_mae_db": 0.8862068142728179, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06098442401320078, "300–1000 Hz_40ms_envelope_correlation": 0.9966700856678067, "300–1000 Hz_40ms_modulation_mae_db": 0.6051462389657006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6113454542738477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11380393950677342, "1000–3000 Hz_10ms_envelope_correlation": 0.9906222642033622, "1000–3000 Hz_10ms_modulation_mae_db": 0.6791058985444275, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9380217086510535, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06011091783914278, "1000–3000 Hz_40ms_envelope_correlation": 0.9977297381883795, "1000–3000 Hz_40ms_modulation_mae_db": 0.5409941690084253, "3000–6000 Hz_10ms_log_rms_mae_db": 1.944642067217527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39834138617845055, "3000–6000 Hz_10ms_envelope_correlation": 0.908586275500878, "3000–6000 Hz_10ms_modulation_mae_db": 4.630587739820519, "3000–6000 Hz_40ms_log_rms_mae_db": 1.604319959572333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31870151374383443, "3000–6000 Hz_40ms_envelope_correlation": 0.9394676303764026, "3000–6000 Hz_40ms_modulation_mae_db": 4.86333262186665, "mr_stft_db": 6.01021297220652, "band_mae_db": 0.8687239468064938, "rms_gain_error_db": -0.14080181511648143, "output_peak": 0.9440380988200313}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9810696232249836, "全帯域_10ms_rms_envelope_nrmse": 0.0746313198701911, "全帯域_10ms_envelope_correlation": 0.9950039630802298, "全帯域_10ms_modulation_mae_db": 0.4299389162613681, "全帯域_40ms_log_rms_mae_db": 0.5367698609535912, "全帯域_40ms_rms_envelope_nrmse": 0.045805561863147756, "全帯域_40ms_envelope_correlation": 0.9980809444542732, "全帯域_40ms_modulation_mae_db": 0.31799684765351127, "300–1000 Hz_10ms_log_rms_mae_db": 1.376073793749872, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10753691173875825, "300–1000 Hz_10ms_envelope_correlation": 0.9900192794693233, "300–1000 Hz_10ms_modulation_mae_db": 0.35690717386633103, "300–1000 Hz_40ms_log_rms_mae_db": 0.7614474876983421, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06262004472083184, "300–1000 Hz_40ms_envelope_correlation": 0.996720687933247, "300–1000 Hz_40ms_modulation_mae_db": 0.4403096577517245, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6606226934519257, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11202517643589355, "1000–3000 Hz_10ms_envelope_correlation": 0.9908344568458964, "1000–3000 Hz_10ms_modulation_mae_db": 0.41307080895513043, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8800279823434302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06279695212804419, "1000–3000 Hz_40ms_envelope_correlation": 0.9972805129420435, "1000–3000 Hz_40ms_modulation_mae_db": 0.41281410571305077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8585839910653206, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4180529622032991, "3000–6000 Hz_10ms_envelope_correlation": 0.8991455868172054, "3000–6000 Hz_10ms_modulation_mae_db": 4.755190210354933, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4852964805420146, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.329727848371621, "3000–6000 Hz_40ms_envelope_correlation": 0.9350253096098178, "3000–6000 Hz_40ms_modulation_mae_db": 5.1794158347446215, "mr_stft_db": 6.0153520976859625, "band_mae_db": 0.8824078062064865, "rms_gain_error_db": -0.13555957256477572, "output_peak": 0.9177066239934801}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9541008617044247, "全帯域_10ms_rms_envelope_nrmse": 0.07243406288657436, "全帯域_10ms_envelope_correlation": 0.995471480414206, "全帯域_10ms_modulation_mae_db": 0.27438466522976385, "全帯域_40ms_log_rms_mae_db": 0.4901226323826876, "全帯域_40ms_rms_envelope_nrmse": 0.03830068795462007, "全帯域_40ms_envelope_correlation": 0.9989680577912823, "全帯域_40ms_modulation_mae_db": 0.3421734797849126, "300–1000 Hz_10ms_log_rms_mae_db": 1.3412798248221751, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10864644810252905, "300–1000 Hz_10ms_envelope_correlation": 0.9898187789087658, "300–1000 Hz_10ms_modulation_mae_db": 0.35023075473842585, "300–1000 Hz_40ms_log_rms_mae_db": 0.8165829578409566, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05756945745264211, "300–1000 Hz_40ms_envelope_correlation": 0.9973778325176873, "300–1000 Hz_40ms_modulation_mae_db": 0.21616107937369108, "1000–3000 Hz_10ms_log_rms_mae_db": 1.649606350235679, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10839176466346583, "1000–3000 Hz_10ms_envelope_correlation": 0.9916129441070309, "1000–3000 Hz_10ms_modulation_mae_db": 0.5796856829466427, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9592975785934627, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05951506151626517, "1000–3000 Hz_40ms_envelope_correlation": 0.9978458410976423, "1000–3000 Hz_40ms_modulation_mae_db": 0.4524226510535878, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9191135930140073, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40566053446610517, "3000–6000 Hz_10ms_envelope_correlation": 0.9048434676323367, "3000–6000 Hz_10ms_modulation_mae_db": 4.504403483404648, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4956654323190437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33486385390864587, "3000–6000 Hz_40ms_envelope_correlation": 0.9323275929163581, "3000–6000 Hz_40ms_modulation_mae_db": 4.470363664580467, "mr_stft_db": 6.026974837281872, "band_mae_db": 0.8984276413305793, "rms_gain_error_db": -0.19326232051382566, "output_peak": 0.9609145253259319}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 33040, "output_samples": 33040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.877425183997666, "spectral_convergence": 0.31041024380467613, "spectral_convergence_status": "ok", "amplitude_floor": 1.5756171103077008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.691130563720352, "spectral_convergence": 0.33715893627457966, "spectral_convergence_status": "ok", "amplitude_floor": 1.2505050051947005e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.507516129925866, "spectral_convergence": 0.35629069341785335, "spectral_convergence_status": "ok", "amplitude_floor": 9.121352204658788e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021685618865241799, "output_power": 0.0002176176336180823, "delta_db": 0.015222628774538014, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010795690773652512, "output_power": 0.001114682979947605, "delta_db": 0.13900933603370044, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002989928450689294, "output_power": 0.0029704985008446585, "delta_db": -0.02831458202511737, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0020580355715967367, "output_power": 0.0018894136605933707, "delta_db": -0.37125825933593504, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0009751360590152492, "output_power": 0.0009223712581915172, "delta_db": -0.24159454727438331, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.9392123549665832e-05, "output_power": 5.2690352270117345e-05, "delta_db": 4.341057326905707, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8560761133915635, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10277972981978822, "pitch": null, "output_rms_dbfs": -21.446459690741317, "output_peak": 0.9612942841299922, "reference_rms_dbfs": -21.343679960921527, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0253572925479615}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 33040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00032988134947798775, "log_rms_mae_db": 0.9393330528090001, "rms_envelope_nrmse": 0.08343826727297636, "envelope_correlation": 0.9935752514795789, "rms_gain_error_db": -0.10277893167416946, "reference_mean_power": 0.007356724541990913, "output_mean_power": 0.007184666304031416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 2.980285612038863e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020727352824950612, "output_power": 0.0019734997045242784, "reference_active": true, "delta_db": -0.2130677440169485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00028987718896809884, "output_power": 0.00023547314967687692, "reference_active": true, "delta_db": -0.9027264823131497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.012977878291207e-05, "output_power": 6.039911666834052e-05, "reference_active": true, "delta_db": -0.6487188181443854}], "active_band_count": 3, "active_band_mae_db": 0.5881710148248279}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0002960474071300256, "log_rms_mae_db": 0.4768737778726002, "rms_envelope_nrmse": 0.035138663983703196, "envelope_correlation": 0.9988663281248787, "rms_gain_error_db": -0.1027807464308343, "reference_mean_power": 0.007465415825536721, "output_mean_power": 0.007290812481620964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 2.5597102021075326e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001877761726334087, "output_power": 0.0017934319625517899, "reference_active": true, "delta_db": -0.1995557727323496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012702101262043316, "output_power": 0.0001042599211855317, "reference_active": true, "delta_db": -0.8575817835334321}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.962314408759724e-06, "output_power": 3.2239887676848647e-06, "reference_active": true, "delta_db": -0.8955541367755584}], "active_band_count": 3, "active_band_mae_db": 0.65089723101378}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00021933697126022545, "log_rms_mae_db": 1.4275678433607153, "rms_envelope_nrmse": 0.12879107814327628, "envelope_correlation": 0.9852078181684387, "rms_gain_error_db": -0.09880884977786336, "reference_mean_power": 0.003363603458339666, "output_mean_power": 0.003287940163852974}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.3609823101859838e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008546335077009212, "output_power": 0.000754754508623876, "reference_active": true, "delta_db": -0.5397420022466483}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016660612815788933, "output_power": 0.0001331673043223713, "reference_active": true, "delta_db": -0.9729336321538302}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.188726605138031e-05, "output_power": 4.8232099366157373e-05, "reference_active": true, "delta_db": 0.6125415034717991}], "active_band_count": 3, "active_band_mae_db": 0.7084057126240926}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0001893291578274055, "log_rms_mae_db": 0.8947400891992194, "rms_envelope_nrmse": 0.06931174468958719, "envelope_correlation": 0.9954548832223276, "rms_gain_error_db": -0.09881084250659498, "reference_mean_power": 0.0034133051942474998, "output_mean_power": 0.003336522342535858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.15224102535942e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007711766819603955, "output_power": 0.0006894807933663277, "reference_active": true, "delta_db": -0.4863171666217642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.304889773698623e-05, "output_power": 6.172886210157073e-05, "reference_active": true, "delta_db": -0.7312539574596293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527118205585199e-06, "output_power": 2.212283539761944e-06, "reference_active": true, "delta_db": -0.577847684244004}], "active_band_count": 3, "active_band_mae_db": 0.5984729361084659}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.0002256735995145706, "log_rms_mae_db": 1.7044475323024288, "rms_envelope_nrmse": 0.11415862623035511, "envelope_correlation": 0.9909674061641267, "rms_gain_error_db": -0.3453931117148881, "reference_mean_power": 0.002711357875845494, "output_mean_power": 0.002504076158656394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.6613986227009806e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012279896352023905, "output_power": 0.0011344925620054935, "reference_active": true, "delta_db": -0.34393048279691874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888327196960883, "output_power": 9.301103951678024e-05, "reference_active": true, "delta_db": -1.065862512725376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.234146057131384e-05, "output_power": 2.9472851279522197e-05, "reference_active": true, "delta_db": -0.4033747651893648}], "active_band_count": 3, "active_band_mae_db": 0.6043892535705532}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.00020374269470838238, "log_rms_mae_db": 0.9135264772085459, "rms_envelope_nrmse": 0.055231874387190835, "envelope_correlation": 0.9987051984467252, "rms_gain_error_db": -0.34539645569004573, "reference_mean_power": 0.002751412553659036, "output_mean_power": 0.0025410667233076822}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.4886745667367597e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001137287164515336, "output_power": 0.0010528818991195842, "reference_active": true, "delta_db": -0.33490478233390897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.907501205097428e-05, "output_power": 3.9686717227327984e-05, "reference_active": true, "delta_db": -0.9221523813319549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5268858664777347e-06, "output_power": 1.7553175028288223e-06, "reference_active": true, "delta_db": -1.5822994295699846}], "active_band_count": 3, "active_band_mae_db": 0.9464521977452828}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 4.718157231228245e-05, "log_rms_mae_db": 2.0070869724094558, "rms_envelope_nrmse": 0.4106760456570007, "envelope_correlation": 0.8961359607517009, "rms_gain_error_db": 0.7538038636658568, "reference_mean_power": 9.509723574878595e-05, "output_mean_power": 0.00011312231379687163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 5.720070068605411e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.50238140104046e-05, "output_power": 4.750483488340623e-05, "reference_active": true, "delta_db": 1.323743749526511}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.975419310826897e-06, "output_power": 8.637349649172122e-06, "reference_active": true, "delta_db": 2.3955081322532443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.471999704474668e-06, "output_power": 1.601816966567243e-05, "reference_active": true, "delta_db": 10.367051666210491}], "active_band_count": 3, "active_band_mae_db": 4.695434515996749}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 4.015479902310721e-05, "log_rms_mae_db": 1.6328792338935796, "rms_envelope_nrmse": 0.33303322148954384, "envelope_correlation": 0.9274429860995922, "rms_gain_error_db": 0.7538348970400354, "reference_mean_power": 9.64997641207967e-05, "output_mean_power": 0.00011479150279124682}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 5.087796481772889e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1685652820939687e-05, "output_power": 5.532221610654499e-05, "reference_active": true, "delta_db": 2.4203690948409387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.336227531732081e-06, "output_power": 4.893767107994411e-06, "reference_active": true, "delta_db": 3.211281599893481}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1955736150290575e-07, "output_power": 1.1143952690619634e-06, "reference_active": true, "delta_db": 9.694629376310083}], "active_band_count": 3, "active_band_mae_db": 5.1087600236815005}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8404729931574609, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "356d2c94134b5441ed77f9b0e688be4776b7691d96ef6afe4159228fe7595cff", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9893879874665983, "standard_deviation": 0.0018319320264887423, "minimum": 0.9866392919613862, "maximum": 0.9916129441070309, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6938469513297578, "standard_deviation": 0.05166194145661992, "minimum": 1.6113454542738477, "maximum": 1.7721735539958248, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6146914686178373, "standard_deviation": 0.24950750657133533, "minimum": 0.3860821463934255, "maximum": 1.1675986653599073, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12049736770814115, "standard_deviation": 0.0097613742496658, "minimum": 0.10839176466346583, "maximum": 0.13543106450263195, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972305884643691, "standard_deviation": 0.0010564464910589994, "minimum": 0.9954627959342295, "maximum": 0.9987051984467252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9398675494735858, "standard_deviation": 0.047444940239424326, "minimum": 0.8800279823434302, "maximum": 1.0101281187273599, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6544694580636815, "standard_deviation": 0.20625580871619717, "minimum": 0.41281410571305077, "maximum": 0.9464521977452828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0643287109265027, "standard_deviation": 0.010619334596705126, "minimum": 0.055231874387190835, "maximum": 0.08362638221260281, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9021471382076057, "standard_deviation": 0.004600692083334829, "minimum": 0.8961359607517009, "maximum": 0.908586275500878, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9430906536551102, "standard_deviation": 0.046116396504647134, "minimum": 1.8585839910653206, "maximum": 2.0070869724094558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 4.660370276863757, "standard_deviation": 0.10612600945596376, "minimum": 4.504403483404648, "maximum": 4.820808556350161, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.40962654077272753, "standard_deviation": 0.006464271377975566, "minimum": 0.39834138617845055, "maximum": 0.4180529622032991, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9341666711519641, "standard_deviation": 0.004165274231775463, "minimum": 0.9274429860995922, "maximum": 0.9397574390687589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5519615306926724, "standard_deviation": 0.05436713301854086, "minimum": 1.4852964805420146, "maximum": 1.6328792338935796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.911931166294048, "standard_deviation": 0.262601612299653, "minimum": 4.470363664580467, "maximum": 5.1794158347446215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3291550454228458, "standard_deviation": 0.005898074287641131, "minimum": 0.31870151374383443, "maximum": 0.3370971572284813, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9878943738944114, "standard_deviation": 0.0017298772524399085, "minimum": 0.9852078181684387, "maximum": 0.9900192794693233, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4137594132735465, "standard_deviation": 0.055475615282900716, "minimum": 1.3412798248221751, "maximum": 1.5002566980940082, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5053433891358245, "standard_deviation": 0.1975012944987698, "minimum": 0.23832644268320788, "maximum": 0.839045660979666, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11681128486205035, "standard_deviation": 0.007692031444349748, "minimum": 0.10753691173875825, "maximum": 0.12879107814327628, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967772207256906, "standard_deviation": 0.0006135604604203646, "minimum": 0.9954548832223276, "maximum": 0.9973778325176873, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8581944408246251, "standard_deviation": 0.056129639741603655, "minimum": 0.7614474876983421, "maximum": 0.9424336985613835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.46379828649196636, "standard_deviation": 0.12366985157312801, "minimum": 0.21616107937369108, "maximum": 0.6051462389657006, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05998249586653431, "standard_deviation": 0.004959360346586291, "minimum": 0.05343081028220292, "maximum": 0.06931174468958719, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.8816117870747731, "standard_deviation": 0.028892970487146865, "minimum": 0.8464038271036186, "maximum": 0.935211750399109, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.0206914849383395, "standard_deviation": 0.009066417262547092, "minimum": 6.01021297220652, "maximum": 6.037438846591321, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9481816020169159, "standard_deviation": 0.023880659256975766, "minimum": 0.9177066239934801, "maximum": 0.9939251791245401, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12779134326488473, "standard_deviation": 0.041525830889020274, "minimum": -0.19326232051382566, "maximum": -0.057169587443785516, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939230310916394, "standard_deviation": 0.001071674203742336, "minimum": 0.9921630051310864, "maximum": 0.995471480414206, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9636882644116402, "standard_deviation": 0.020668528141233888, "minimum": 0.9312669776454522, "maximum": 0.9886201851691616, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4294067370766018, "standard_deviation": 0.13925880626563786, "minimum": 0.23407319599096443, "maximum": 0.6367631215496021, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08148728059019225, "standard_deviation": 0.006819190721658065, "minimum": 0.07243406288657436, "maximum": 0.09274364693563308, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984241557909579, "standard_deviation": 0.0004904930109883959, "minimum": 0.997684069818208, "maximum": 0.9989680577912823, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5086933460533627, "standard_deviation": 0.022299458067407685, "minimum": 0.4768737778726002, "maximum": 0.5367698609535912, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4443262933000155, "standard_deviation": 0.10989423100390013, "minimum": 0.31799684765351127, "maximum": 0.65089723101378, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041459790310052216, "standard_deviation": 0.006162331570912351, "minimum": 0.033716941682731025, "maximum": 0.05037483426938323, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0029", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.4281394481658936, "predicted_mos": 1.4281394481658936, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "6a3df7524b81a3751a237ff7183b71f8f9850007b6115c0960f3d8ff1fb8532e", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.535966099705547, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9772417898532961, "全帯域_10ms_rms_envelope_nrmse": 0.07967857267870056, "全帯域_10ms_envelope_correlation": 0.9940850421795502, "全帯域_10ms_modulation_mae_db": 0.3888396156816587, "全帯域_40ms_log_rms_mae_db": 0.5195726917956206, "全帯域_40ms_rms_envelope_nrmse": 0.034145419223581816, "全帯域_40ms_envelope_correlation": 0.9988161852927961, "全帯域_40ms_modulation_mae_db": 0.4921350101992683, "300–1000 Hz_10ms_log_rms_mae_db": 1.3331719733861185, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11407688720259418, "300–1000 Hz_10ms_envelope_correlation": 0.9884450379497227, "300–1000 Hz_10ms_modulation_mae_db": 0.35034801219652006, "300–1000 Hz_40ms_log_rms_mae_db": 0.8421129907779148, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06187019354622509, "300–1000 Hz_40ms_envelope_correlation": 0.9964204065297607, "300–1000 Hz_40ms_modulation_mae_db": 0.50128146092747, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6319483458759063, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10014042423555619, "1000–3000 Hz_10ms_envelope_correlation": 0.9924925895248036, "1000–3000 Hz_10ms_modulation_mae_db": 0.5944457230956489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8906522958110052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0480798321558335, "1000–3000 Hz_40ms_envelope_correlation": 0.998237436028306, "1000–3000 Hz_40ms_modulation_mae_db": 0.5671448743554038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9426330907364657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.42480821890797194, "3000–6000 Hz_10ms_envelope_correlation": 0.8958244492610408, "3000–6000 Hz_10ms_modulation_mae_db": 4.735163725461253, "3000–6000 Hz_40ms_log_rms_mae_db": 1.547310185499043, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34868099899986343, "3000–6000 Hz_40ms_envelope_correlation": 0.9266901684364537, "3000–6000 Hz_40ms_modulation_mae_db": 5.042614430287258, "mr_stft_db": 5.994718013702804, "band_mae_db": 0.8397902018670064, "rms_gain_error_db": -0.037372685503421724, "output_peak": 0.9298354001333361}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9919101927113501, "全帯域_10ms_rms_envelope_nrmse": 0.07894997087485382, "全帯域_10ms_envelope_correlation": 0.9942523466975284, "全帯域_10ms_modulation_mae_db": 0.2713185916266822, "全帯域_40ms_log_rms_mae_db": 0.5377754271665695, "全帯域_40ms_rms_envelope_nrmse": 0.04393845502010361, "全帯域_40ms_envelope_correlation": 0.9980755732962533, "全帯域_40ms_modulation_mae_db": 0.4576428470096821, "300–1000 Hz_10ms_log_rms_mae_db": 1.369633993500826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1147094147418419, "300–1000 Hz_10ms_envelope_correlation": 0.9882791053680086, "300–1000 Hz_10ms_modulation_mae_db": 0.9300598430506074, "300–1000 Hz_40ms_log_rms_mae_db": 0.8522547023742189, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05153578750620619, "300–1000 Hz_40ms_envelope_correlation": 0.9974301231381862, "300–1000 Hz_40ms_modulation_mae_db": 0.24408118006985238, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7338019729687002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12096819964624814, "1000–3000 Hz_10ms_envelope_correlation": 0.9890476242863403, "1000–3000 Hz_10ms_modulation_mae_db": 0.677795047642269, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8587711204674303, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06871622333460899, "1000–3000 Hz_40ms_envelope_correlation": 0.9962069893882547, "1000–3000 Hz_40ms_modulation_mae_db": 1.1028044077406636, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9450215528647448, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4092617741361972, "3000–6000 Hz_10ms_envelope_correlation": 0.9019609953140508, "3000–6000 Hz_10ms_modulation_mae_db": 4.7583085977607995, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5569962686264647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32802790080047656, "3000–6000 Hz_40ms_envelope_correlation": 0.9344211049151513, "3000–6000 Hz_40ms_modulation_mae_db": 4.906839418998037, "mr_stft_db": 6.006205678167581, "band_mae_db": 0.7824316495665992, "rms_gain_error_db": 0.018033087304951165, "output_peak": 0.9942227713570566}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9464652149859811, "全帯域_10ms_rms_envelope_nrmse": 0.08058104016954333, "全帯域_10ms_envelope_correlation": 0.9939640567958192, "全帯域_10ms_modulation_mae_db": 0.16865061146514346, "全帯域_40ms_log_rms_mae_db": 0.4670782429419734, "全帯域_40ms_rms_envelope_nrmse": 0.041868055523841874, "全帯域_40ms_envelope_correlation": 0.9982052802038403, "全帯域_40ms_modulation_mae_db": 0.4159901575234131, "300–1000 Hz_10ms_log_rms_mae_db": 1.2433949552526715, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10873082209651544, "300–1000 Hz_10ms_envelope_correlation": 0.9894247025868096, "300–1000 Hz_10ms_modulation_mae_db": 0.23160841604093066, "300–1000 Hz_40ms_log_rms_mae_db": 0.6890545736882664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05772007972277287, "300–1000 Hz_40ms_envelope_correlation": 0.9967475996984405, "300–1000 Hz_40ms_modulation_mae_db": 0.31284369667025097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.601040195343078, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11087861187893408, "1000–3000 Hz_10ms_envelope_correlation": 0.9907248067982879, "1000–3000 Hz_10ms_modulation_mae_db": 0.47915926004631754, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8577324679041318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05627395476707816, "1000–3000 Hz_40ms_envelope_correlation": 0.997520640025986, "1000–3000 Hz_40ms_modulation_mae_db": 0.7692464054049362, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9274808794772194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4151793784908255, "3000–6000 Hz_10ms_envelope_correlation": 0.8978799392678468, "3000–6000 Hz_10ms_modulation_mae_db": 4.571811038809554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.511585817593585, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32867695310294576, "3000–6000 Hz_40ms_envelope_correlation": 0.9330496609012607, "3000–6000 Hz_40ms_modulation_mae_db": 4.955659311832974, "mr_stft_db": 5.9743427854954385, "band_mae_db": 0.909026122194552, "rms_gain_error_db": 0.00307053467742281, "output_peak": 0.9456825405882753}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0333354725985744, "全帯域_10ms_rms_envelope_nrmse": 0.08355011199808776, "全帯域_10ms_envelope_correlation": 0.9934942598782118, "全帯域_10ms_modulation_mae_db": 0.7798423284658383, "全帯域_40ms_log_rms_mae_db": 0.5757446105194782, "全帯域_40ms_rms_envelope_nrmse": 0.05107959796273177, "全帯域_40ms_envelope_correlation": 0.9973324152165105, "全帯域_40ms_modulation_mae_db": 1.110923926094828, "300–1000 Hz_10ms_log_rms_mae_db": 1.4595447847292575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10311114805428237, "300–1000 Hz_10ms_envelope_correlation": 0.9905466366764625, "300–1000 Hz_10ms_modulation_mae_db": 0.2744005840402243, "300–1000 Hz_40ms_log_rms_mae_db": 0.7752558868202871, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04965476358586469, "300–1000 Hz_40ms_envelope_correlation": 0.9976033039452992, "300–1000 Hz_40ms_modulation_mae_db": 0.8513978771394987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7422466313619434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13308921138233074, "1000–3000 Hz_10ms_envelope_correlation": 0.9867078865062566, "1000–3000 Hz_10ms_modulation_mae_db": 0.8373489964097515, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0187120335906532, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08700907718560605, "1000–3000 Hz_40ms_envelope_correlation": 0.9941069697128557, "1000–3000 Hz_40ms_modulation_mae_db": 1.1094267268876405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9053972340487053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40534882279434636, "3000–6000 Hz_10ms_envelope_correlation": 0.8955818892512649, "3000–6000 Hz_10ms_modulation_mae_db": 4.364591115901433, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6324446499048635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33160806942953763, "3000–6000 Hz_40ms_envelope_correlation": 0.924994693377154, "3000–6000 Hz_40ms_modulation_mae_db": 4.669561480969173, "mr_stft_db": 5.996242298618701, "band_mae_db": 0.9077596191787016, "rms_gain_error_db": -0.026832047192529095, "output_peak": 0.9654869492654234}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9575945949548891, "全帯域_10ms_rms_envelope_nrmse": 0.08698076839579763, "全帯域_10ms_envelope_correlation": 0.9929504267436422, "全帯域_10ms_modulation_mae_db": 0.2552222659921946, "全帯域_40ms_log_rms_mae_db": 0.5144074056696099, "全帯域_40ms_rms_envelope_nrmse": 0.043860607283682136, "全帯域_40ms_envelope_correlation": 0.9980239055557659, "全帯域_40ms_modulation_mae_db": 0.31498076128777314, "300–1000 Hz_10ms_log_rms_mae_db": 1.445880210876074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1134959938436065, "300–1000 Hz_10ms_envelope_correlation": 0.988611517578141, "300–1000 Hz_10ms_modulation_mae_db": 0.5032851974148194, "300–1000 Hz_40ms_log_rms_mae_db": 0.8115568931342286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051837712492594265, "300–1000 Hz_40ms_envelope_correlation": 0.9976963961485229, "300–1000 Hz_40ms_modulation_mae_db": 0.4311719796098472, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7194647930396632, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1377335217285881, "1000–3000 Hz_10ms_envelope_correlation": 0.9856695461835467, "1000–3000 Hz_10ms_modulation_mae_db": 0.31050698897708146, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9255398461814545, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07489978057143445, "1000–3000 Hz_40ms_envelope_correlation": 0.9954151944018095, "1000–3000 Hz_40ms_modulation_mae_db": 0.301991987601928, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9231947896293657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40998136211050373, "3000–6000 Hz_10ms_envelope_correlation": 0.899269882205505, "3000–6000 Hz_10ms_modulation_mae_db": 4.579506686311485, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6184256560338477, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.328030951809584, "3000–6000 Hz_40ms_envelope_correlation": 0.9323565203999317, "3000–6000 Hz_40ms_modulation_mae_db": 4.883907655498227, "mr_stft_db": 5.998678285318296, "band_mae_db": 0.8623678215945118, "rms_gain_error_db": -0.013579259197811092, "output_peak": 0.9392612940778042}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0175662853021221, "全帯域_10ms_rms_envelope_nrmse": 0.07872772251384468, "全帯域_10ms_envelope_correlation": 0.9944113841360364, "全帯域_10ms_modulation_mae_db": 0.4628880093906247, "全帯域_40ms_log_rms_mae_db": 0.5602554937568885, "全帯域_40ms_rms_envelope_nrmse": 0.04750772843421435, "全帯域_40ms_envelope_correlation": 0.9979276276220597, "全帯域_40ms_modulation_mae_db": 0.4329483469337016, "300–1000 Hz_10ms_log_rms_mae_db": 1.4007370793340794, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11741129760040364, "300–1000 Hz_10ms_envelope_correlation": 0.9882857272738511, "300–1000 Hz_10ms_modulation_mae_db": 0.35312115593996724, "300–1000 Hz_40ms_log_rms_mae_db": 0.8893439092166616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07882258410803014, "300–1000 Hz_40ms_envelope_correlation": 0.9948043296642982, "300–1000 Hz_40ms_modulation_mae_db": 0.6523604176849797, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7787144646568356, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11186361877863642, "1000–3000 Hz_10ms_envelope_correlation": 0.99067026420912, "1000–3000 Hz_10ms_modulation_mae_db": 0.650615388014493, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0068947088778784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06842779189611324, "1000–3000 Hz_40ms_envelope_correlation": 0.9964142053239213, "1000–3000 Hz_40ms_modulation_mae_db": 0.7174348307928438, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9581504954144773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3973385003600713, "3000–6000 Hz_10ms_envelope_correlation": 0.9069405423426117, "3000–6000 Hz_10ms_modulation_mae_db": 4.52752059900424, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5810032817249595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3280380340135429, "3000–6000 Hz_40ms_envelope_correlation": 0.9334808325662927, "3000–6000 Hz_40ms_modulation_mae_db": 4.784524290514541, "mr_stft_db": 6.015843635039233, "band_mae_db": 0.9273568782509646, "rms_gain_error_db": -0.13835792908611494, "output_peak": 0.9592800860410279}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9897097183421665, "全帯域_10ms_rms_envelope_nrmse": 0.07923837053616418, "全帯域_10ms_envelope_correlation": 0.9941966867527492, "全帯域_10ms_modulation_mae_db": 0.5359841259826342, "全帯域_40ms_log_rms_mae_db": 0.48997096444163446, "全帯域_40ms_rms_envelope_nrmse": 0.040340062450319646, "全帯域_40ms_envelope_correlation": 0.9984212519523401, "全帯域_40ms_modulation_mae_db": 0.7353328523007732, "300–1000 Hz_10ms_log_rms_mae_db": 1.3874432178696243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11114986270938415, "300–1000 Hz_10ms_envelope_correlation": 0.9890381776094083, "300–1000 Hz_10ms_modulation_mae_db": 0.3245464929658222, "300–1000 Hz_40ms_log_rms_mae_db": 0.8412475029026985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06276257937128611, "300–1000 Hz_40ms_envelope_correlation": 0.9963345113329364, "300–1000 Hz_40ms_modulation_mae_db": 0.4934654287463584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7327629911441638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11130145623944929, "1000–3000 Hz_10ms_envelope_correlation": 0.9907990933944334, "1000–3000 Hz_10ms_modulation_mae_db": 0.7420481130959685, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9486532085708645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056052999997895606, "1000–3000 Hz_40ms_envelope_correlation": 0.9977678084319073, "1000–3000 Hz_40ms_modulation_mae_db": 1.1649816969397941, "3000–6000 Hz_10ms_log_rms_mae_db": 1.904424090678042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40253257785997215, "3000–6000 Hz_10ms_envelope_correlation": 0.9068559017672095, "3000–6000 Hz_10ms_modulation_mae_db": 4.632651907379825, "3000–6000 Hz_40ms_log_rms_mae_db": 1.600356779540691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32628093860941626, "3000–6000 Hz_40ms_envelope_correlation": 0.9362186917186354, "3000–6000 Hz_40ms_modulation_mae_db": 4.818468140868995, "mr_stft_db": 5.990520162893913, "band_mae_db": 0.8630859089637997, "rms_gain_error_db": -0.08942243242859889, "output_peak": 0.9777884706317792}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9695951768196323, "全帯域_10ms_rms_envelope_nrmse": 0.06594527165883303, "全帯域_10ms_envelope_correlation": 0.9959893954307965, "全帯域_10ms_modulation_mae_db": 0.4338593792492658, "全帯域_40ms_log_rms_mae_db": 0.4867029427011145, "全帯域_40ms_rms_envelope_nrmse": 0.0333804789893827, "全帯域_40ms_envelope_correlation": 0.9988784340693663, "全帯域_40ms_modulation_mae_db": 0.6101307640845591, "300–1000 Hz_10ms_log_rms_mae_db": 1.3481888060450027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09472829022540384, "300–1000 Hz_10ms_envelope_correlation": 0.9919659946059713, "300–1000 Hz_10ms_modulation_mae_db": 0.25886110325890055, "300–1000 Hz_40ms_log_rms_mae_db": 0.8181154521154546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041975855774529364, "300–1000 Hz_40ms_envelope_correlation": 0.9983214236610153, "300–1000 Hz_40ms_modulation_mae_db": 0.35407565753721676, "1000–3000 Hz_10ms_log_rms_mae_db": 1.662698617026366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11652011447205665, "1000–3000 Hz_10ms_envelope_correlation": 0.9899103412835832, "1000–3000 Hz_10ms_modulation_mae_db": 0.28151415404286667, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8692128231321948, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060066253997426745, "1000–3000 Hz_40ms_envelope_correlation": 0.9971694137635638, "1000–3000 Hz_40ms_modulation_mae_db": 0.41092304667140334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9421690679727246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4044988940484983, "3000–6000 Hz_10ms_envelope_correlation": 0.9055043734420279, "3000–6000 Hz_10ms_modulation_mae_db": 4.566141260335141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5753658022884987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3259196593510861, "3000–6000 Hz_40ms_envelope_correlation": 0.9363550755744326, "3000–6000 Hz_40ms_modulation_mae_db": 4.8829408399137675, "mr_stft_db": 5.95749028411849, "band_mae_db": 0.8352475504939978, "rms_gain_error_db": 0.03662163627590296, "output_peak": 0.9672543798858483}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 33040, "output_samples": 33040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8845862353656795, "spectral_convergence": 0.3146189902022789, "spectral_convergence_status": "ok", "amplitude_floor": 1.5756171103077008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.618358023729062, "spectral_convergence": 0.31031070057664323, "spectral_convergence_status": "ok", "amplitude_floor": 1.2505050051947005e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.481209782013671, "spectral_convergence": 0.34757648365868854, "spectral_convergence_status": "ok", "amplitude_floor": 9.121352204658788e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021685618865241799, "output_power": 0.0002205585128990876, "delta_db": 0.07352004327708174, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010795690773652512, "output_power": 0.0011094400470689173, "delta_db": 0.11853402052113833, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002989928450689294, "output_power": 0.0029235297738028753, "delta_db": -0.0975327459858231, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0020580355715967367, "output_power": 0.0020038615114667657, "delta_db": -0.11585173164713627, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0009751360590152492, "output_power": 0.0009629884768937473, "delta_db": -0.05444125875716293, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.9392123549665832e-05, "output_power": 5.5655938511319164e-05, "delta_db": 4.578861411013696, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8397902018670064, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.037372685503421724, "pitch": null, "output_rms_dbfs": -21.381052646424948, "output_peak": 0.9298354001333361, "reference_rms_dbfs": -21.343679960921527, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.994718013702804}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 33040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00032988134947798775, "log_rms_mae_db": 0.9772417898532961, "rms_envelope_nrmse": 0.07967857267870056, "envelope_correlation": 0.9940850421795502, "rms_gain_error_db": -0.03737154938418242, "reference_mean_power": 0.007356724541990913, "output_mean_power": 0.007293690660651434}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 2.980285612038863e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020727352824950612, "output_power": 0.0020478431217166924, "reference_active": true, "delta_db": -0.05247156352209026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00028987718896809884, "output_power": 0.0002714860888796359, "reference_active": true, "delta_db": -0.28466460205319644}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.012977878291207e-05, "output_power": 5.793809357028589e-05, "reference_active": true, "delta_db": -0.8293826814696894}], "active_band_count": 3, "active_band_mae_db": 0.3888396156816587}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0002960474071300256, "log_rms_mae_db": 0.5195726917956206, "rms_envelope_nrmse": 0.034145419223581816, "envelope_correlation": 0.9988161852927961, "rms_gain_error_db": -0.03737047887469247, "reference_mean_power": 0.007465415825536721, "output_mean_power": 0.007401452480041261}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 2.5597102021075326e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001877761726334087, "output_power": 0.001853598067854786, "reference_active": true, "delta_db": -0.05624914678941298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012702101262043316, "output_power": 0.0001210753678202623, "reference_active": true, "delta_db": -0.20819773559800078}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.962314408759724e-06, "output_power": 2.9974580854733893e-06, "reference_active": true, "delta_db": -1.2119581482103912}], "active_band_count": 3, "active_band_mae_db": 0.4921350101992683}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00021933697126022545, "log_rms_mae_db": 1.3331719733861185, "rms_envelope_nrmse": 0.11407688720259418, "envelope_correlation": 0.9884450379497227, "rms_gain_error_db": -0.13050493124969442, "reference_mean_power": 0.003363603458339666, "output_mean_power": 0.0032640311684588434}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.3609823101859838e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008546335077009212, "output_power": 0.0007539484088848857, "reference_active": true, "delta_db": -0.5443828721986844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016660612815788933, "output_power": 0.00016767024866479232, "reference_active": true, "delta_db": 0.027650367023031308}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.188726605138031e-05, "output_power": 4.6771693492138217e-05, "reference_active": true, "delta_db": 0.47901079736784435}], "active_band_count": 3, "active_band_mae_db": 0.35034801219652006}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0001893291578274055, "log_rms_mae_db": 0.8421129907779148, "rms_envelope_nrmse": 0.06187019354622509, "envelope_correlation": 0.9964204065297607, "rms_gain_error_db": -0.13050497606153041, "reference_mean_power": 0.0034133051942474998, "output_mean_power": 0.0033122615565474908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.15224102535942e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007711766819603955, "output_power": 0.0006884403248034798, "reference_active": true, "delta_db": -0.49287588519225345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.304889773698623e-05, "output_power": 7.624475068574116e-05, "reference_active": true, "delta_db": 0.18596281569671094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527118205585199e-06, "output_power": 2.089897508706261e-06, "reference_active": true, "delta_db": -0.8250056818934453}], "active_band_count": 3, "active_band_mae_db": 0.50128146092747}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.0002256735995145706, "log_rms_mae_db": 1.6319483458759063, "rms_envelope_nrmse": 0.10014042423555619, "envelope_correlation": 0.9924925895248036, "rms_gain_error_db": -0.11107592122999692, "reference_mean_power": 0.002711357875845494, "output_mean_power": 0.0026428910021163415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.6613986227009806e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012279896352023905, "output_power": 0.0012115079208580338, "reference_active": true, "delta_db": -0.05868443268311073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888327196960883, "output_power": 0.00011621777186279634, "reference_active": true, "delta_db": -0.0984820461715326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.234146057131384e-05, "output_power": 2.224039827077298e-05, "reference_active": true, "delta_db": -1.6261706904323034}], "active_band_count": 3, "active_band_mae_db": 0.5944457230956489}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.00020374269470838238, "log_rms_mae_db": 0.8906522958110052, "rms_envelope_nrmse": 0.0480798321558335, "envelope_correlation": 0.998237436028306, "rms_gain_error_db": -0.11107446840471619, "reference_mean_power": 0.002751412553659036, "output_mean_power": 0.002681935121335085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.4886745667367597e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001137287164515336, "output_power": 0.0011249433303214423, "reference_active": true, "delta_db": -0.04739492542910778}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.907501205097428e-05, "output_power": 5.019693515592225e-05, "reference_active": true, "delta_db": 0.09816786579098992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5268858664777347e-06, "output_power": 1.7660314896894618e-06, "reference_active": true, "delta_db": -1.5558718318461138}], "active_band_count": 3, "active_band_mae_db": 0.5671448743554038}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 4.718157231228245e-05, "log_rms_mae_db": 1.9426330907364657, "rms_envelope_nrmse": 0.42480821890797194, "envelope_correlation": 0.8958244492610408, "rms_gain_error_db": 0.9959937865100909, "reference_mean_power": 9.509723574878595e-05, "output_mean_power": 0.00011960993981659563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 5.720070068605411e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.50238140104046e-05, "output_power": 5.4542815306484686e-05, "reference_active": true, "delta_db": 1.923741129601545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.975419310826897e-06, "output_power": 8.042134932706225e-06, "reference_active": true, "delta_db": 2.085416680818073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.471999704474668e-06, "output_power": 1.540072121874246e-05, "reference_active": true, "delta_db": 10.19633336596414}], "active_band_count": 3, "active_band_mae_db": 4.735163725461253}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 4.015479902310721e-05, "log_rms_mae_db": 1.547310185499043, "rms_envelope_nrmse": 0.34868099899986343, "envelope_correlation": 0.9266901684364537, "rms_gain_error_db": 0.9960157094202657, "reference_mean_power": 9.64997641207967e-05, "output_mean_power": 0.00012137460308938337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 5.087796481772889e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1685652820939687e-05, "output_power": 6.277678463564726e-05, "reference_active": true, "delta_db": 2.969364084607618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.336227531732081e-06, "output_power": 4.275429408862164e-06, "reference_active": true, "delta_db": 2.6246460258747217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1955736150290575e-07, "output_power": 1.0738896089275032e-06, "reference_active": true, "delta_db": 9.533833180379434}], "active_band_count": 3, "active_band_mae_db": 5.042614430287258}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8484594014171448, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fb425cc79c0a08a173b483070ca789e1a251536fcdd14cb9f508aa8baf61c7e2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895027690232965, "standard_deviation": 0.0022789139750902295, "minimum": 0.9856695461835467, "maximum": 0.9924925895248036, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7003347514270821, "standard_deviation": 0.06140053188621886, "minimum": 1.601040195343078, "maximum": 1.7787144646568356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5716792089155496, "standard_deviation": 0.19949308392659323, "minimum": 0.28151415404286667, "maximum": 0.8373489964097515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11781189479522496, "standard_deviation": 0.012421281622322335, "minimum": 0.10014042423555619, "maximum": 0.1377335217285881, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966048321345755, "standard_deviation": 0.0013640558166491579, "minimum": 0.9941069697128557, "maximum": 0.998237436028306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9220210630669516, "standard_deviation": 0.06456516200576938, "minimum": 0.8577324679041318, "maximum": 1.0187120335906532, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7679942470493266, "standard_deviation": 0.33248614542455185, "minimum": 0.301991987601928, "maximum": 1.1649816969397941, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06494073923824958, "standard_deviation": 0.012391031989274277, "minimum": 0.0480798321558335, "maximum": 0.08700907718560605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9012272466064447, "standard_deviation": 0.0047697218975769, "minimum": 0.8955818892512649, "maximum": 0.9069405423426117, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.931058900102718, "standard_deviation": 0.019379254631491104, "minimum": 1.904424090678042, "maximum": 1.9581504954144773, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 4.591961866370466, "standard_deviation": 0.1235843916953616, "minimum": 4.364591115901433, "maximum": 4.7583085977607995, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.4086186910885483, "standard_deviation": 0.008446149602384684, "minimum": 0.3973385003600713, "maximum": 0.42480821890797194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.932195843486164, "standard_deviation": 0.004191988050749867, "minimum": 0.924994693377154, "maximum": 0.9363550755744326, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.577936055151494, "standard_deviation": 0.03947525484952037, "minimum": 1.511585817593585, "maximum": 1.6324446499048635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.868064446110371, "standard_deviation": 0.11273138024035707, "minimum": 4.669561480969173, "maximum": 5.042614430287258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3306579382645566, "standard_deviation": 0.007482830212600434, "minimum": 0.3259196593510861, "maximum": 0.34868099899986343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9893246124560469, "standard_deviation": 0.0013102251127661878, "minimum": 0.9882791053680086, "maximum": 0.9919659946059713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3734993776242068, "standard_deviation": 0.06843990759640853, "minimum": 1.2433949552526715, "maximum": 1.4595447847292575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.40327885061347396, "standard_deviation": 0.22864587934091543, "minimum": 0.23160841604093066, "maximum": 0.9300598430506074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10967671455925401, "standard_deviation": 0.0074520101900732594, "minimum": 0.09472829022540384, "maximum": 0.11741129760040364, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969197617648073, "standard_deviation": 0.0010949287092861333, "minimum": 0.9948043296642982, "maximum": 0.9983214236610153, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8148677388787162, "standard_deviation": 0.06071272094072289, "minimum": 0.6890545736882664, "maximum": 0.8893439092166616, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4800847122981842, "standard_deviation": 0.1964392645539276, "minimum": 0.24408118006985238, "maximum": 0.8513978771394987, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05702244451343859, "standard_deviation": 0.011147389228434527, "minimum": 0.041975855774529364, "maximum": 0.07882258410803014, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.8658832190137666, "standard_deviation": 0.04780569437219224, "minimum": 0.7824316495665992, "maximum": 0.9273568782509646, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.991755142919307, "standard_deviation": 0.01830188283506826, "minimum": 5.95749028411849, "maximum": 6.015843635039233, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9598514864975689, "standard_deviation": 0.021097355572336912, "minimum": 0.9298354001333361, "maximum": 0.9942227713570566, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03097988689377485, "standard_deviation": 0.05786449460340941, "minimum": -0.13835792908611494, "maximum": 0.03662163627590296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941679498267917, "standard_deviation": 0.0008758293738622002, "minimum": 0.9929504267436422, "maximum": 0.9959893954307965, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9854273056960015, "standard_deviation": 0.029274163699843543, "minimum": 0.9464652149859811, "maximum": 1.0333354725985744, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41207561598175524, "standard_deviation": 0.1919349827304331, "minimum": 0.16865061146514346, "maximum": 0.7798423284658383, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07920647860322812, "standard_deviation": 0.006066001925118318, "minimum": 0.06594527165883303, "maximum": 0.08698076839579763, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982100841511166, "standard_deviation": 0.0005016409094696685, "minimum": 0.9973324152165105, "maximum": 0.9988784340693663, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5189384723741111, "standard_deviation": 0.037564281169396814, "minimum": 0.4670782429419734, "maximum": 0.5757446105194782, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5712605831792499, "standard_deviation": 0.2529131689798858, "minimum": 0.31498076128777314, "maximum": 1.110923926094828, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.042015050610982235, "standard_deviation": 0.006079887340018294, "minimum": 0.0333804789893827, "maximum": 0.05107959796273177, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_237-134500-0031", "language": "en", "style": "normal", "speaker": "librispeech_237", "source_utterance_id": "237-134500-0031", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_237-134500-0031.wav", "source_file_sha256": "333f8e275cefdd131b539d529fefb535a352f1a67b2598410b10ecad419df8c2", "decoded_pcm_sha256": "66fcbe0885d7123cd5c5c14ff8b544f94ed65d51857cb70d13de972dd00f9421", "original_source_sha256": "7456acb116b3a8c6d43c44ebf79a8f7b59f6ed546ffb63ce78a6dc58bb6d5abe", "corpus_entry_id": "librispeech_237-134500-0031"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.765350818634033, "predicted_mos": 3.765350818634033, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "49402d92c425e74fcf31e0293bc450adca73cd447e7dddd8a7ee60679fa3221f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.971693428000435, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_237-134500-0031", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "49402d92c425e74fcf31e0293bc450adca73cd447e7dddd8a7ee60679fa3221f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "66b8f0fa3a5988c5f43c9ed60894944381007f86fe6b413cf7370344f5a1b084"}, "rows": [{"case": "librispeech_237-134500-0031", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1054608821868896, "predicted_mos": 2.1054608821868896, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7d8e6d25eac63b0e7e823c9aa60b139d89bbe5f95bffeb692dbc20ac4b457d4b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.306427561001328, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_237-134500-0031", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7d8e6d25eac63b0e7e823c9aa60b139d89bbe5f95bffeb692dbc20ac4b457d4b", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1406275996837765, "全帯域_10ms_rms_envelope_nrmse": 0.11772920703333926, "全帯域_10ms_envelope_correlation": 0.9865086881821642, "全帯域_10ms_modulation_mae_db": 0.7514019807827799, "全帯域_40ms_log_rms_mae_db": 0.5964000324857751, "全帯域_40ms_rms_envelope_nrmse": 0.06367385267142067, "全帯域_40ms_envelope_correlation": 0.9956000254664908, "全帯域_40ms_modulation_mae_db": 0.7404732818522156, "300–1000 Hz_10ms_log_rms_mae_db": 1.7631312487188993, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14113944393167532, "300–1000 Hz_10ms_envelope_correlation": 0.9828865147125885, "300–1000 Hz_10ms_modulation_mae_db": 1.3555055709918982, "300–1000 Hz_40ms_log_rms_mae_db": 0.9427413575680339, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07574593650172398, "300–1000 Hz_40ms_envelope_correlation": 0.99452292227587, "300–1000 Hz_40ms_modulation_mae_db": 0.9299963150034314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7365496326565828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1523978284366089, "1000–3000 Hz_10ms_envelope_correlation": 0.9817061710873463, "1000–3000 Hz_10ms_modulation_mae_db": 0.598127078191154, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0127172510713975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0894983163828254, "1000–3000 Hz_40ms_envelope_correlation": 0.993431613399466, "1000–3000 Hz_40ms_modulation_mae_db": 0.6517413417704296, "3000–6000 Hz_10ms_log_rms_mae_db": 1.94625089911462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5255874311349522, "3000–6000 Hz_10ms_envelope_correlation": 0.8039976732716476, "3000–6000 Hz_10ms_modulation_mae_db": 2.902334306203464, "3000–6000 Hz_40ms_log_rms_mae_db": 1.309076851850252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.44486547565689327, "3000–6000 Hz_40ms_envelope_correlation": 0.8386594494491253, "3000–6000 Hz_40ms_modulation_mae_db": 3.872807850599134, "mr_stft_db": 5.834428130315772, "band_mae_db": 0.37777164738217944, "rms_gain_error_db": -0.008180711277544987, "output_peak": 0.8110286208073898}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1301308851325085, "全帯域_10ms_rms_envelope_nrmse": 0.10923464475289374, "全帯域_10ms_envelope_correlation": 0.9883918061919956, "全帯域_10ms_modulation_mae_db": 0.3566701137403507, "全帯域_40ms_log_rms_mae_db": 0.5796530001741458, "全帯域_40ms_rms_envelope_nrmse": 0.0521917381219901, "全帯域_40ms_envelope_correlation": 0.9970461492431147, "全帯域_40ms_modulation_mae_db": 0.21272133139935256, "300–1000 Hz_10ms_log_rms_mae_db": 1.688673267104968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1361155001341231, "300–1000 Hz_10ms_envelope_correlation": 0.9841273797268358, "300–1000 Hz_10ms_modulation_mae_db": 0.8845479526853192, "300–1000 Hz_40ms_log_rms_mae_db": 0.9005720341634081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06802877133244138, "300–1000 Hz_40ms_envelope_correlation": 0.9955918612449904, "300–1000 Hz_40ms_modulation_mae_db": 0.24515297984763273, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7711070750460916, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1431052034642313, "1000–3000 Hz_10ms_envelope_correlation": 0.983848915270221, "1000–3000 Hz_10ms_modulation_mae_db": 0.3247800618288412, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0810235808948647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08352838019201936, "1000–3000 Hz_40ms_envelope_correlation": 0.9941830064860707, "1000–3000 Hz_40ms_modulation_mae_db": 0.17577431157317447, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9489883535212877, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5136025593960709, "3000–6000 Hz_10ms_envelope_correlation": 0.8222641472295539, "3000–6000 Hz_10ms_modulation_mae_db": 2.796089602754724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2912822971158844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4305193156733373, "3000–6000 Hz_40ms_envelope_correlation": 0.8592546066990239, "3000–6000 Hz_40ms_modulation_mae_db": 3.7827571509426363, "mr_stft_db": 5.8399863846199365, "band_mae_db": 0.40315617474469834, "rms_gain_error_db": 0.006059193901533356, "output_peak": 0.7420997250076589}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1627044245940807, "全帯域_10ms_rms_envelope_nrmse": 0.11926597121177336, "全帯域_10ms_envelope_correlation": 0.9860983196247726, "全帯域_10ms_modulation_mae_db": 1.3291315580355565, "全帯域_40ms_log_rms_mae_db": 0.6135565056415652, "全帯域_40ms_rms_envelope_nrmse": 0.04822801971800879, "全帯域_40ms_envelope_correlation": 0.9975095246250185, "全帯域_40ms_modulation_mae_db": 0.9755585221295346, "300–1000 Hz_10ms_log_rms_mae_db": 1.6961041328436057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15134727419447805, "300–1000 Hz_10ms_envelope_correlation": 0.9803588879246967, "300–1000 Hz_10ms_modulation_mae_db": 2.03677622291908, "300–1000 Hz_40ms_log_rms_mae_db": 0.9363684413203448, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06837938308377028, "300–1000 Hz_40ms_envelope_correlation": 0.9955578583375964, "300–1000 Hz_40ms_modulation_mae_db": 1.450890357824891, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7759596323420421, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15967445885340734, "1000–3000 Hz_10ms_envelope_correlation": 0.9804432442415939, "1000–3000 Hz_10ms_modulation_mae_db": 0.5897303413470175, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0769925214728173, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08983212493036846, "1000–3000 Hz_40ms_envelope_correlation": 0.9942968527341349, "1000–3000 Hz_40ms_modulation_mae_db": 0.7592415198413712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9884231771188565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5270774052531915, "3000–6000 Hz_10ms_envelope_correlation": 0.7998526701695385, "3000–6000 Hz_10ms_modulation_mae_db": 2.6841201193619604, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3345473422790846, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.44127189909036074, "3000–6000 Hz_40ms_envelope_correlation": 0.8382988839247437, "3000–6000 Hz_40ms_modulation_mae_db": 3.7519321543019193, "mr_stft_db": 5.843465456030116, "band_mae_db": 0.40901666142993154, "rms_gain_error_db": -0.061369395857904255, "output_peak": 0.7204653124077511}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1860467806524315, "全帯域_10ms_rms_envelope_nrmse": 0.11779169512617423, "全帯域_10ms_envelope_correlation": 0.9865824477975456, "全帯域_10ms_modulation_mae_db": 0.7477232215858264, "全帯域_40ms_log_rms_mae_db": 0.6707292755193729, "全帯域_40ms_rms_envelope_nrmse": 0.07986918391516924, "全帯域_40ms_envelope_correlation": 0.9933550787998588, "全帯域_40ms_modulation_mae_db": 0.8459741373771008, "300–1000 Hz_10ms_log_rms_mae_db": 1.733558815533829, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15005928384364906, "300–1000 Hz_10ms_envelope_correlation": 0.9809631163101689, "300–1000 Hz_10ms_modulation_mae_db": 0.4949307946321078, "300–1000 Hz_40ms_log_rms_mae_db": 1.018407724422998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10656961778677665, "300–1000 Hz_40ms_envelope_correlation": 0.9898529631248203, "300–1000 Hz_40ms_modulation_mae_db": 0.4756192192367437, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8040308070851978, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16080365405378877, "1000–3000 Hz_10ms_envelope_correlation": 0.9796561747744248, "1000–3000 Hz_10ms_modulation_mae_db": 0.8545621894164368, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0567917606738513, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10199025602248263, "1000–3000 Hz_40ms_envelope_correlation": 0.9910725539002061, "1000–3000 Hz_40ms_modulation_mae_db": 0.6634162049983678, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0121483077251603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5125349285696407, "3000–6000 Hz_10ms_envelope_correlation": 0.8084327707467125, "3000–6000 Hz_10ms_modulation_mae_db": 2.625224172760145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3884489229309953, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.43036300519602855, "3000–6000 Hz_40ms_envelope_correlation": 0.843931603945577, "3000–6000 Hz_40ms_modulation_mae_db": 3.5857112302095806, "mr_stft_db": 5.83609639923887, "band_mae_db": 0.5036990529888686, "rms_gain_error_db": -0.17004671493257523, "output_peak": 0.6096713463144762}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1376576950992443, "全帯域_10ms_rms_envelope_nrmse": 0.10878774213120998, "全帯域_10ms_envelope_correlation": 0.9884793069937288, "全帯域_10ms_modulation_mae_db": 0.17794123006118878, "全帯域_40ms_log_rms_mae_db": 0.6364964905353818, "全帯域_40ms_rms_envelope_nrmse": 0.05993283391385843, "全帯域_40ms_envelope_correlation": 0.9961043963073282, "全帯域_40ms_modulation_mae_db": 0.0794650551402653, "300–1000 Hz_10ms_log_rms_mae_db": 1.8509923995162407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13911174194812623, "300–1000 Hz_10ms_envelope_correlation": 0.9833589794493899, "300–1000 Hz_10ms_modulation_mae_db": 0.5939844310568427, "300–1000 Hz_40ms_log_rms_mae_db": 1.0767586972490557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07822322838806585, "300–1000 Hz_40ms_envelope_correlation": 0.994195239547801, "300–1000 Hz_40ms_modulation_mae_db": 0.15823914400815067, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7800721690187213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13957277299593185, "1000–3000 Hz_10ms_envelope_correlation": 0.984927504383317, "1000–3000 Hz_10ms_modulation_mae_db": 0.09017182380691431, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0339899075056789, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0884139278906223, "1000–3000 Hz_40ms_envelope_correlation": 0.9934625095933769, "1000–3000 Hz_40ms_modulation_mae_db": 0.09841246283674537, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9448246145169426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5180320897350568, "3000–6000 Hz_10ms_envelope_correlation": 0.8077975064991435, "3000–6000 Hz_10ms_modulation_mae_db": 2.832287534869583, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3321467902455237, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4377940854706986, "3000–6000 Hz_40ms_envelope_correlation": 0.8418652046889906, "3000–6000 Hz_40ms_modulation_mae_db": 3.8847526918129027, "mr_stft_db": 5.840221248084404, "band_mae_db": 0.3808175047588012, "rms_gain_error_db": -0.010271426863292486, "output_peak": 0.7812271912288817}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1386801281119279, "全帯域_10ms_rms_envelope_nrmse": 0.11092493005770601, "全帯域_10ms_envelope_correlation": 0.9880958680881233, "全帯域_10ms_modulation_mae_db": 0.43861590666410705, "全帯域_40ms_log_rms_mae_db": 0.640983220334446, "全帯域_40ms_rms_envelope_nrmse": 0.05434350815188214, "全帯域_40ms_envelope_correlation": 0.9970872527517219, "全帯域_40ms_modulation_mae_db": 0.34188625984428955, "300–1000 Hz_10ms_log_rms_mae_db": 1.7717948014252511, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13669455932203767, "300–1000 Hz_10ms_envelope_correlation": 0.9842388500120512, "300–1000 Hz_10ms_modulation_mae_db": 0.9458918427908997, "300–1000 Hz_40ms_log_rms_mae_db": 0.9955292372436693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07214504325488136, "300–1000 Hz_40ms_envelope_correlation": 0.9957350468211296, "300–1000 Hz_40ms_modulation_mae_db": 0.47757978391935235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7536224449580515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13988329308329164, "1000–3000 Hz_10ms_envelope_correlation": 0.98454988635343, "1000–3000 Hz_10ms_modulation_mae_db": 0.4505834446788515, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0350976264474416, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07517467496953337, "1000–3000 Hz_40ms_envelope_correlation": 0.9952693756682867, "1000–3000 Hz_40ms_modulation_mae_db": 0.49748692223183055, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0273198332082463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5180731531858914, "3000–6000 Hz_10ms_envelope_correlation": 0.8135984308706334, "3000–6000 Hz_10ms_modulation_mae_db": 2.702000940482202, "3000–6000 Hz_40ms_log_rms_mae_db": 1.402052577488033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4353713413751661, "3000–6000 Hz_40ms_envelope_correlation": 0.8490467515546216, "3000–6000 Hz_40ms_modulation_mae_db": 3.727718697776693, "mr_stft_db": 5.8484160158127665, "band_mae_db": 0.4747639375827206, "rms_gain_error_db": -0.14773542115691932, "output_peak": 0.6722356429096845}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1450689051875436, "全帯域_10ms_rms_envelope_nrmse": 0.11689526378439216, "全帯域_10ms_envelope_correlation": 0.9867127235493091, "全帯域_10ms_modulation_mae_db": 0.9221595073339252, "全帯域_40ms_log_rms_mae_db": 0.6077607066003782, "全帯域_40ms_rms_envelope_nrmse": 0.060227886473266, "全帯域_40ms_envelope_correlation": 0.9963023949528351, "全帯域_40ms_modulation_mae_db": 0.35489221148185135, "300–1000 Hz_10ms_log_rms_mae_db": 1.778623220023065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14687801954735566, "300–1000 Hz_10ms_envelope_correlation": 0.9816215656317065, "300–1000 Hz_10ms_modulation_mae_db": 1.1886665121546247, "300–1000 Hz_40ms_log_rms_mae_db": 1.0423757015203334, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08830500704689162, "300–1000 Hz_40ms_envelope_correlation": 0.993088996550266, "300–1000 Hz_40ms_modulation_mae_db": 0.33149099639418345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7287251787121274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1537699483295077, "1000–3000 Hz_10ms_envelope_correlation": 0.9812749298584594, "1000–3000 Hz_10ms_modulation_mae_db": 1.2571696508096406, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0232880648160008, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08720879150618686, "1000–3000 Hz_40ms_envelope_correlation": 0.993497521732394, "1000–3000 Hz_40ms_modulation_mae_db": 0.88245563533839, "3000–6000 Hz_10ms_log_rms_mae_db": 2.037315972190415, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5113970269413218, "3000–6000 Hz_10ms_envelope_correlation": 0.8119639039244744, "3000–6000 Hz_10ms_modulation_mae_db": 2.6257263407772498, "3000–6000 Hz_40ms_log_rms_mae_db": 1.383232977303793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4274804378254837, "3000–6000 Hz_40ms_envelope_correlation": 0.849145998361023, "3000–6000 Hz_40ms_modulation_mae_db": 3.5337346966140895, "mr_stft_db": 5.84691147430464, "band_mae_db": 0.4539389921399864, "rms_gain_error_db": -0.14119946962519508, "output_peak": 0.6962348384585582}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1534489496986364, "全帯域_10ms_rms_envelope_nrmse": 0.12619983211013874, "全帯域_10ms_envelope_correlation": 0.9848265932605692, "全帯域_10ms_modulation_mae_db": 0.8062120121368602, "全帯域_40ms_log_rms_mae_db": 0.6028368497548772, "全帯域_40ms_rms_envelope_nrmse": 0.07285297716122896, "全帯域_40ms_envelope_correlation": 0.9944786469636289, "全帯域_40ms_modulation_mae_db": 0.2755672716672596, "300–1000 Hz_10ms_log_rms_mae_db": 1.75358985801837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15789358809408155, "300–1000 Hz_10ms_envelope_correlation": 0.978845531340177, "300–1000 Hz_10ms_modulation_mae_db": 1.2767619942827901, "300–1000 Hz_40ms_log_rms_mae_db": 0.9784821191538187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08834024911708072, "300–1000 Hz_40ms_envelope_correlation": 0.9926720944017069, "300–1000 Hz_40ms_modulation_mae_db": 0.5125974790943874, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7555512025472617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15459771019632274, "1000–3000 Hz_10ms_envelope_correlation": 0.9811953963864886, "1000–3000 Hz_10ms_modulation_mae_db": 0.7335296653678869, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0160476670516079, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09155029188835305, "1000–3000 Hz_40ms_envelope_correlation": 0.9928138171993779, "1000–3000 Hz_40ms_modulation_mae_db": 0.6594653773577209, "3000–6000 Hz_10ms_log_rms_mae_db": 2.01814147297719, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5333379520800374, "3000–6000 Hz_10ms_envelope_correlation": 0.8010559597649047, "3000–6000 Hz_10ms_modulation_mae_db": 2.6799350702249174, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3568623875750383, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4477064164373213, "3000–6000 Hz_40ms_envelope_correlation": 0.8396712304819294, "3000–6000 Hz_40ms_modulation_mae_db": 3.6633992824329824, "mr_stft_db": 5.832392587747762, "band_mae_db": 0.41153571370818903, "rms_gain_error_db": 0.08652124827128509, "output_peak": 0.7806981459437656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.925541617030301, "spectral_convergence": 0.3708900180948589, "spectral_convergence_status": "ok", "amplitude_floor": 2.051312383514402e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.640649163364185, "spectral_convergence": 0.4037739652389937, "spectral_convergence_status": "ok", "amplitude_floor": 1.425885904641783e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.937093610552832, "spectral_convergence": 0.3920732992406542, "spectral_convergence_status": "ok", "amplitude_floor": 8.483800566678623e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001304494586920238, "output_power": 0.00015336444187413078, "delta_db": 0.7028239715747533, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014301207186140645, "output_power": 0.0014545291657547719, "delta_db": 0.0734973555010181, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.004308309383381418, "output_power": 0.004279562341100558, "delta_db": -0.029075256047406507, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.001304992137209625, "output_power": 0.0012683345566593157, "delta_db": -0.12374069717044169, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002442775973124686, "output_power": 0.0002447942466539104, "delta_db": 0.009175667922841978, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0021123865382114e-05, "output_power": 1.3606553193599644e-05, "delta_db": 1.328316936076615, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.37777164738217944, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.008180711277544987, "pitch": null, "output_rms_dbfs": -21.29936212245409, "output_peak": 0.8110286208073898, "reference_rms_dbfs": -21.29118141117655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.834428130315772}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002631236493882886, "log_rms_mae_db": 1.1406275996837765, "rms_envelope_nrmse": 0.11772920703333926, "envelope_correlation": 0.9865086881821642, "rms_gain_error_db": -0.008180567643691622, "reference_mean_power": 0.007434364903539191, "output_mean_power": 0.0074203743776803084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.450307530079474e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015621941098031631, "output_power": 0.0015833190075557515, "reference_active": true, "delta_db": 0.058334294709896685}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031400948871223737, "output_power": 0.0003430978150183171, "reference_active": true, "delta_db": 0.38475180561578676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9119711962951405e-05, "output_power": 7.453629402377152e-05, "reference_active": true, "delta_db": 1.811119842022656}], "active_band_count": 3, "active_band_mae_db": 0.7514019807827799}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024033810696394756, "log_rms_mae_db": 0.5964000324857751, "rms_envelope_nrmse": 0.06367385267142067, "envelope_correlation": 0.9956000254664908, "rms_gain_error_db": -0.00818109500839293, "reference_mean_power": 0.007471745584669175, "output_mean_power": 0.007457683807551346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.136699874330791e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014162642987993615, "output_power": 0.0014645571811204335, "reference_active": true, "delta_db": 0.14562025040511495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001495506387635467, "output_power": 0.0001625791412460475, "reference_active": true, "delta_db": 0.3627655291225368}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0546566248780076e-06, "output_power": 4.53174855264944e-06, "reference_active": true, "delta_db": 1.7130340660289953}], "active_band_count": 3, "active_band_mae_db": 0.7404732818522156}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002303175773961387, "log_rms_mae_db": 1.7631312487188993, "rms_envelope_nrmse": 0.14113944393167532, "envelope_correlation": 0.9828865147125885, "rms_gain_error_db": -0.039800796424402936, "reference_mean_power": 0.004944923373977616, "output_mean_power": 0.0048998127850024085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6955719701225893e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012875788279854436, "output_power": 0.0012756407390720939, "reference_active": true, "delta_db": -0.04045446355801715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026650258378785487, "output_power": 0.00029697568988230796, "reference_active": true, "delta_db": 0.47019475940767724}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0302791253002337e-05, "output_power": 6.871782151937598e-05, "reference_active": true, "delta_db": 3.5558674900100002}], "active_band_count": 3, "active_band_mae_db": 1.3555055709918982}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020459712012168156, "log_rms_mae_db": 0.9427413575680339, "rms_envelope_nrmse": 0.07574593650172398, "envelope_correlation": 0.99452292227587, "rms_gain_error_db": -0.039800717999673475, "reference_mean_power": 0.004969790940376444, "output_mean_power": 0.004924453583309683}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.46585734788702e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011840763589577269, "output_power": 0.0012050648555003106, "reference_active": true, "delta_db": 0.07630710714870151}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012134699925967192, "output_power": 0.00013011798683361676, "reference_active": true, "delta_db": 0.30308294011282133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.482336680977429e-06, "output_power": 4.324347385021336e-06, "reference_active": true, "delta_db": 2.4105988977487716}], "active_band_count": 3, "active_band_mae_db": 0.9299963150034314}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014083569934759876, "log_rms_mae_db": 1.7365496326565828, "rms_envelope_nrmse": 0.1523978284366089, "envelope_correlation": 0.9817061710873463, "rms_gain_error_db": -0.19297456587110207, "reference_mean_power": 0.0013181968247743324, "output_mean_power": 0.001260906270859268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.706116677050846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003562978060875054, "output_power": 0.0003344118128625088, "reference_active": true, "delta_db": -0.2753153799469568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927249258823322e-05, "output_power": 7.050119220497425e-05, "reference_active": true, "delta_db": 0.07635646295528804}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3444445693793863e-05, "output_power": 1.8741909670602895e-05, "reference_active": true, "delta_db": 1.442709391671217}], "active_band_count": 3, "active_band_mae_db": 0.598127078191154}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010879087849669345, "log_rms_mae_db": 1.0127172510713975, "rms_envelope_nrmse": 0.0894983163828254, "envelope_correlation": 0.993431613399466, "rms_gain_error_db": -0.19297733174580262, "reference_mean_power": 0.0013248216670349005, "output_mean_power": 0.001267242381767385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.107952744901008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003355694425017771, "output_power": 0.0003188676219520945, "reference_active": true, "delta_db": -0.22171983489257172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.409137696567018e-05, "output_power": 3.6443244467016587e-05, "reference_active": true, "delta_db": 0.2897249143787393}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.097157965658087e-07, "output_power": 1.1290431482342776e-06, "reference_active": true, "delta_db": 1.443779276039978}], "active_band_count": 3, "active_band_mae_db": 0.6517413417704296}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.557718084358035e-05, "log_rms_mae_db": 1.94625089911462, "rms_envelope_nrmse": 0.5255874311349522, "envelope_correlation": 0.8039976732716476, "rms_gain_error_db": 1.0722597306531558, "reference_mean_power": 2.3934708718272692e-05, "output_mean_power": 3.063755610895397e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3699291941082093e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.018836822609012e-06, "output_power": 5.8541626483357535e-06, "reference_active": true, "delta_db": 1.6336441159665607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.745214583455981e-06, "output_power": 6.989980569594655e-06, "reference_active": true, "delta_db": 1.682201120688824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3999346848964708e-06, "output_power": 4.844216470804165e-06, "reference_active": true, "delta_db": 5.391157681955009}], "active_band_count": 3, "active_band_mae_db": 2.902334306203464}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.6234588308267947e-05, "log_rms_mae_db": 1.309076851850252, "rms_envelope_nrmse": 0.44486547565689327, "envelope_correlation": 0.8386594494491253, "rms_gain_error_db": 1.0722599063401896, "reference_mean_power": 2.405439007374668e-05, "output_mean_power": 3.079075513466874e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1165079820636279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.688364032412796e-06, "output_power": 7.5636147239168285e-06, "reference_active": true, "delta_db": 3.1189562031641227}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313243490564173e-06, "output_power": 5.5617990274863475e-06, "reference_active": true, "delta_db": 2.249619384265807}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.200263616160695e-07, "output_power": 5.061292514985568e-07, "reference_active": true, "delta_db": 6.249847964367471}], "active_band_count": 3, "active_band_mae_db": 3.872807850599134}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8538517291895117, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b67335b82794ca2dd9bfbc83f8217f616f986186fc9f792a8416c60eb5962b1e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9822002777944101, "standard_deviation": 0.0019765140576656205, "minimum": 0.9796561747744248, "maximum": 0.984927504383317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7632022677957595, "standard_deviation": 0.02457101646887118, "minimum": 1.7287251787121274, "maximum": 1.8040308070851978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6123317819308429, "standard_deviation": 0.352823815063267, "minimum": 0.09017182380691431, "maximum": 1.2571696508096406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15047560867663629, "standard_deviation": 0.008520718764822714, "minimum": 0.13957277299593185, "maximum": 0.16080365405378877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9935034063391641, "standard_deviation": 0.0012306501982443355, "minimum": 0.9910725539002061, "maximum": 0.9952693756682867, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0419935474917075, "standard_deviation": 0.02661647209281665, "minimum": 1.0127172510713975, "maximum": 1.0810235808948647, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5484992219935036, "standard_deviation": 0.2767726613188727, "minimum": 0.09841246283674537, "maximum": 0.88245563533839, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08839959547279894, "standard_deviation": 0.0075397778529149, "minimum": 0.07517467496953337, "maximum": 0.10199025602248263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8086203828095762, "standard_deviation": 0.007351262783740983, "minimum": 0.7998526701695385, "maximum": 0.8222641472295539, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9904265787965896, "standard_deviation": 0.03882533461319794, "minimum": 1.9448246145169426, "maximum": 2.037315972190415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.7309647609292806, "standard_deviation": 0.10125541983626038, "minimum": 2.625224172760145, "maximum": 2.902334306203464, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5199553182870204, "standard_deviation": 0.007905198456863063, "minimum": 0.5113970269413218, "maximum": 0.5333379520800374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8449842161381294, "standard_deviation": 0.007183602895741474, "minimum": 0.8382988839247437, "maximum": 0.8592546066990239, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3497062683485757, "standard_deviation": 0.039674781178246084, "minimum": 1.2912822971158844, "maximum": 1.402052577488033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.725351719336242, "standard_deviation": 0.12604655625149583, "minimum": 3.5337346966140895, "maximum": 3.8847526918129027, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.4369214970906612, "standard_deviation": 0.00731580494973383, "minimum": 0.4274804378254837, "maximum": 0.4477064164373213, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9820501031384519, "standard_deviation": 0.0019277339209217734, "minimum": 0.978845531340177, "maximum": 0.9842388500120512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7545584678980286, "standard_deviation": 0.051337666207407935, "minimum": 1.688673267104968, "maximum": 1.8509923995162407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0971331651891951, "standard_deviation": 0.48905199837588714, "minimum": 0.4949307946321078, "maximum": 2.03677622291908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14490492637694083, "standard_deviation": 0.007865600157686878, "minimum": 0.1361155001341231, "maximum": 0.15789358809408155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9939021227880225, "standard_deviation": 0.0019999083657925567, "minimum": 0.9898529631248203, "maximum": 0.9957350468211296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9864044140802077, "standard_deviation": 0.05885111608168104, "minimum": 0.9005720341634081, "maximum": 1.0767586972490557, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5726957844160966, "standard_deviation": 0.42396267400566173, "minimum": 0.15823914400815067, "maximum": 1.450890357824891, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08071715456395398, "standard_deviation": 0.013102161074588578, "minimum": 0.06802877133244138, "maximum": 0.10656961778677665, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4268374605919219, "standard_deviation": 0.045633575664438326, "minimum": 0.37777164738217944, "maximum": 0.5036990529888686, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.840239712019283, "standard_deviation": 0.005786106238874956, "minimum": 5.832392587747762, "maximum": 5.8484160158127665, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7267076028847708, "standard_deviation": 0.06640412108302057, "minimum": 0.6096713463144762, "maximum": 0.8110286208073898, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05577783719257661, "standard_deviation": 0.09043613563416922, "minimum": -0.17004671493257523, "maximum": 0.08652124827128509, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9869619692110261, "standard_deviation": 0.001273343128168309, "minimum": 0.9848265932605692, "maximum": 0.9884793069937288, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1492956710200186, "standard_deviation": 0.0179527583172325, "minimum": 1.1301308851325085, "maximum": 1.1860467806524315, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6912319412925743, "standard_deviation": 0.3624301722961508, "minimum": 0.17794123006118878, "maximum": 1.3291315580355565, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11585366077595344, "standard_deviation": 0.005917005485653697, "minimum": 0.10878774213120998, "maximum": 0.12619983211013874, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959354336387496, "standard_deviation": 0.0014189686898118466, "minimum": 0.9933550787998588, "maximum": 0.9975095246250185, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6185520101307428, "standard_deviation": 0.029129874805876678, "minimum": 0.5796530001741458, "maximum": 0.6707292755193729, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4783172588614837, "standard_deviation": 0.32857380749994264, "minimum": 0.0794650551402653, "maximum": 0.9755585221295346, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06141500001585304, "standard_deviation": 0.010617127007980393, "minimum": 0.04822801971800879, "maximum": 0.07986918391516924, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0031", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.2612509727478027, "predicted_mos": 2.2612509727478027, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0e6382ad419127a588e29fde0d2cac83acd103b2e02fa28bd9ebb1f8afab8e8e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.132841099984944, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1650715052372727, "全帯域_10ms_rms_envelope_nrmse": 0.11931808367960137, "全帯域_10ms_envelope_correlation": 0.9861483671880863, "全帯域_10ms_modulation_mae_db": 0.7778133132977381, "全帯域_40ms_log_rms_mae_db": 0.6096407966130737, "全帯域_40ms_rms_envelope_nrmse": 0.06425969644217386, "全帯域_40ms_envelope_correlation": 0.9955201864634498, "全帯域_40ms_modulation_mae_db": 0.7678571422623182, "300–1000 Hz_10ms_log_rms_mae_db": 1.8312489209575047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14414953615413398, "300–1000 Hz_10ms_envelope_correlation": 0.9821472636352695, "300–1000 Hz_10ms_modulation_mae_db": 1.3243555459066738, "300–1000 Hz_40ms_log_rms_mae_db": 0.9760663811510946, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07741518797930917, "300–1000 Hz_40ms_envelope_correlation": 0.9942719896105542, "300–1000 Hz_40ms_modulation_mae_db": 0.8570260261601069, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7394379202494066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15225105068839406, "1000–3000 Hz_10ms_envelope_correlation": 0.9817615308391471, "1000–3000 Hz_10ms_modulation_mae_db": 0.5710186561157925, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9995303817827209, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09129109321107112, "1000–3000 Hz_40ms_envelope_correlation": 0.9931683532826713, "1000–3000 Hz_40ms_modulation_mae_db": 0.6219465375514956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.818690569533946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5083078454906332, "3000–6000 Hz_10ms_envelope_correlation": 0.8098615961373651, "3000–6000 Hz_10ms_modulation_mae_db": 2.7238225216616505, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1856476590659832, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4269784368707315, "3000–6000 Hz_40ms_envelope_correlation": 0.844405626442713, "3000–6000 Hz_40ms_modulation_mae_db": 3.75035997833157, "mr_stft_db": 5.756345677129261, "band_mae_db": 0.3003565347591684, "rms_gain_error_db": -0.005234483859750754, "output_peak": 0.8111830375274082}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.12268138440065, "全帯域_10ms_rms_envelope_nrmse": 0.11110390219818833, "全帯域_10ms_envelope_correlation": 0.9880075288221443, "全帯域_10ms_modulation_mae_db": 0.28996616041215895, "全帯域_40ms_log_rms_mae_db": 0.571100557279644, "全帯域_40ms_rms_envelope_nrmse": 0.05396505294322558, "全帯域_40ms_envelope_correlation": 0.9968473921607277, "全帯域_40ms_modulation_mae_db": 0.17906180003056407, "300–1000 Hz_10ms_log_rms_mae_db": 1.7469470006679868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14030538697322248, "300–1000 Hz_10ms_envelope_correlation": 0.9831361935363033, "300–1000 Hz_10ms_modulation_mae_db": 0.8134399419292185, "300–1000 Hz_40ms_log_rms_mae_db": 0.9203721083080119, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07183976523733995, "300–1000 Hz_40ms_envelope_correlation": 0.9950798010063335, "300–1000 Hz_40ms_modulation_mae_db": 0.2877022313182294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7873324699711761, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14237890175949391, "1000–3000 Hz_10ms_envelope_correlation": 0.9839956247151802, "1000–3000 Hz_10ms_modulation_mae_db": 0.42546077266961885, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0826379960942942, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08190820759432661, "1000–3000 Hz_40ms_envelope_correlation": 0.9943636387940059, "1000–3000 Hz_40ms_modulation_mae_db": 0.2877128345826847, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8415955993245445, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5045347070982587, "3000–6000 Hz_10ms_envelope_correlation": 0.822577229508949, "3000–6000 Hz_10ms_modulation_mae_db": 2.721338171523295, "3000–6000 Hz_40ms_log_rms_mae_db": 1.17845679355807, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.42263976704926925, "3000–6000 Hz_40ms_envelope_correlation": 0.858160289367483, "3000–6000 Hz_40ms_modulation_mae_db": 3.7438350623778387, "mr_stft_db": 5.751331710799434, "band_mae_db": 0.3291265054491314, "rms_gain_error_db": 0.015045308441758606, "output_peak": 0.738574485946807}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.176272635957756, "全帯域_10ms_rms_envelope_nrmse": 0.11770161375192523, "全帯域_10ms_envelope_correlation": 0.9864530860375018, "全帯域_10ms_modulation_mae_db": 1.2782082284445908, "全帯域_40ms_log_rms_mae_db": 0.6061673759446503, "全帯域_40ms_rms_envelope_nrmse": 0.045963342614124006, "全帯域_40ms_envelope_correlation": 0.9977371734485453, "全帯域_40ms_modulation_mae_db": 0.9462506819577922, "300–1000 Hz_10ms_log_rms_mae_db": 1.735748381999071, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15059175595434132, "300–1000 Hz_10ms_envelope_correlation": 0.980529828444351, "300–1000 Hz_10ms_modulation_mae_db": 2.036257025268809, "300–1000 Hz_40ms_log_rms_mae_db": 0.9258583527905151, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06372930072455163, "300–1000 Hz_40ms_envelope_correlation": 0.9961256711728934, "300–1000 Hz_40ms_modulation_mae_db": 1.407827882059883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7976676253527777, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16283461135014424, "1000–3000 Hz_10ms_envelope_correlation": 0.9796408122420086, "1000–3000 Hz_10ms_modulation_mae_db": 0.6103588240180257, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0805447621054098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09430382931986125, "1000–3000 Hz_40ms_envelope_correlation": 0.9936024664745862, "1000–3000 Hz_40ms_modulation_mae_db": 0.8763094024942252, "3000–6000 Hz_10ms_log_rms_mae_db": 1.851551777523602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.511757102100624, "3000–6000 Hz_10ms_envelope_correlation": 0.8040528941008563, "3000–6000 Hz_10ms_modulation_mae_db": 2.5359896239133835, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2114551295626483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4280311801864768, "3000–6000 Hz_40ms_envelope_correlation": 0.8403926847361464, "3000–6000 Hz_40ms_modulation_mae_db": 3.63772402786855, "mr_stft_db": 5.754699429850724, "band_mae_db": 0.3506599928963854, "rms_gain_error_db": -0.058625299106177836, "output_peak": 0.7209716757775879}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1901954876751717, "全帯域_10ms_rms_envelope_nrmse": 0.11745667858691544, "全帯域_10ms_envelope_correlation": 0.9866677197826219, "全帯域_10ms_modulation_mae_db": 0.775464167972504, "全帯域_40ms_log_rms_mae_db": 0.6715462451276945, "全帯域_40ms_rms_envelope_nrmse": 0.08143915161579414, "全帯域_40ms_envelope_correlation": 0.9930836172994014, "全帯域_40ms_modulation_mae_db": 0.7794228309126027, "300–1000 Hz_10ms_log_rms_mae_db": 1.779234614688467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15213114627445426, "300–1000 Hz_10ms_envelope_correlation": 0.9803318293120542, "300–1000 Hz_10ms_modulation_mae_db": 0.5635263898341988, "300–1000 Hz_40ms_log_rms_mae_db": 1.033424839732723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10901523034082854, "300–1000 Hz_40ms_envelope_correlation": 0.9892244889176696, "300–1000 Hz_40ms_modulation_mae_db": 0.5087320880747069, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7859873224179796, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15285849486769995, "1000–3000 Hz_10ms_envelope_correlation": 0.9815649807808775, "1000–3000 Hz_10ms_modulation_mae_db": 0.8894820380786671, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0242726801237547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09397880111952085, "1000–3000 Hz_40ms_envelope_correlation": 0.9924417664384032, "1000–3000 Hz_40ms_modulation_mae_db": 0.66478072362117, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8613900408162203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.49883708784564773, "3000–6000 Hz_10ms_envelope_correlation": 0.8115682255219534, "3000–6000 Hz_10ms_modulation_mae_db": 2.4726183200623284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2373025636054884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.41920333579252883, "3000–6000 Hz_40ms_envelope_correlation": 0.8451180777988703, "3000–6000 Hz_40ms_modulation_mae_db": 3.4894026295695526, "mr_stft_db": 5.7495010591024425, "band_mae_db": 0.38646536803731735, "rms_gain_error_db": -0.17407591932633082, "output_peak": 0.6064615446238055}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.168557481601368, "全帯域_10ms_rms_envelope_nrmse": 0.1099747475681772, "全帯域_10ms_envelope_correlation": 0.98822791436131, "全帯域_10ms_modulation_mae_db": 0.16670831205793527, "全帯域_40ms_log_rms_mae_db": 0.6455607611138114, "全帯域_40ms_rms_envelope_nrmse": 0.06220034675415124, "全帯域_40ms_envelope_correlation": 0.9958040361311943, "全帯域_40ms_modulation_mae_db": 0.10110625290176455, "300–1000 Hz_10ms_log_rms_mae_db": 1.8910105453387516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14216925063428257, "300–1000 Hz_10ms_envelope_correlation": 0.982609240273946, "300–1000 Hz_10ms_modulation_mae_db": 0.5574686610728284, "300–1000 Hz_40ms_log_rms_mae_db": 1.0864744846771768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08183331238657643, "300–1000 Hz_40ms_envelope_correlation": 0.9936157036672295, "300–1000 Hz_40ms_modulation_mae_db": 0.17231041292043522, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8157983951043772, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1393105071804068, "1000–3000 Hz_10ms_envelope_correlation": 0.9848266799334956, "1000–3000 Hz_10ms_modulation_mae_db": 0.04021512035182186, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0461571596914288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08551626532521761, "1000–3000 Hz_40ms_envelope_correlation": 0.9937847498571778, "1000–3000 Hz_40ms_modulation_mae_db": 0.13203977741137574, "3000–6000 Hz_10ms_log_rms_mae_db": 1.837098896489439, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5029761108149976, "3000–6000 Hz_10ms_envelope_correlation": 0.8106875459801897, "3000–6000 Hz_10ms_modulation_mae_db": 2.6618995295475556, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2395520959384523, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4253735372725588, "3000–6000 Hz_40ms_envelope_correlation": 0.842218279562351, "3000–6000 Hz_40ms_modulation_mae_db": 3.7342977025975492, "mr_stft_db": 5.755563718166272, "band_mae_db": 0.30205131391511414, "rms_gain_error_db": -0.012121454999582419, "output_peak": 0.7814605538889862}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1545313768499392, "全帯域_10ms_rms_envelope_nrmse": 0.11385005226048367, "全帯域_10ms_envelope_correlation": 0.987380804570249, "全帯域_10ms_modulation_mae_db": 0.33724601563855544, "全帯域_40ms_log_rms_mae_db": 0.6515022753713471, "全帯域_40ms_rms_envelope_nrmse": 0.05562056163828718, "全帯域_40ms_envelope_correlation": 0.9968276935242523, "全帯域_40ms_modulation_mae_db": 0.3232179649153483, "300–1000 Hz_10ms_log_rms_mae_db": 1.7772107542570372, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13729596052057766, "300–1000 Hz_10ms_envelope_correlation": 0.9839436251852183, "300–1000 Hz_10ms_modulation_mae_db": 0.762188130179417, "300–1000 Hz_40ms_log_rms_mae_db": 0.9955133538945675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0711112735447519, "300–1000 Hz_40ms_envelope_correlation": 0.9956427530069755, "300–1000 Hz_40ms_modulation_mae_db": 0.3530218785167178, "1000–3000 Hz_10ms_log_rms_mae_db": 1.76335351567736, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1417581474823207, "1000–3000 Hz_10ms_envelope_correlation": 0.9841452060513626, "1000–3000 Hz_10ms_modulation_mae_db": 0.5906962979009646, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0229621190032991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07672603723357864, "1000–3000 Hz_40ms_envelope_correlation": 0.9951093614015057, "1000–3000 Hz_40ms_modulation_mae_db": 0.5617992374070341, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9057752626029634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5054011126120769, "3000–6000 Hz_10ms_envelope_correlation": 0.8136420416164918, "3000–6000 Hz_10ms_modulation_mae_db": 2.58435851393013, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2788365129446368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4274770937741643, "3000–6000 Hz_40ms_envelope_correlation": 0.8448241714607454, "3000–6000 Hz_40ms_modulation_mae_db": 3.7054013665636245, "mr_stft_db": 5.760411286294203, "band_mae_db": 0.3675914651590187, "rms_gain_error_db": -0.1252025049283744, "output_peak": 0.7093450144471318}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1290557942636097, "全帯域_10ms_rms_envelope_nrmse": 0.11125655474930776, "全帯域_10ms_envelope_correlation": 0.9880569931020915, "全帯域_10ms_modulation_mae_db": 0.829670131685449, "全帯域_40ms_log_rms_mae_db": 0.616165008806535, "全帯域_40ms_rms_envelope_nrmse": 0.060380087913217556, "全帯域_40ms_envelope_correlation": 0.9963857823462225, "全帯域_40ms_modulation_mae_db": 0.3323537494599971, "300–1000 Hz_10ms_log_rms_mae_db": 1.8317417132429932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1418847596275233, "300–1000 Hz_10ms_envelope_correlation": 0.9829757879824257, "300–1000 Hz_10ms_modulation_mae_db": 0.9998298520088102, "300–1000 Hz_40ms_log_rms_mae_db": 1.0734308616112036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08925472595582314, "300–1000 Hz_40ms_envelope_correlation": 0.9930435842102368, "300–1000 Hz_40ms_modulation_mae_db": 0.2530517710891604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7487563112264606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14964660536231433, "1000–3000 Hz_10ms_envelope_correlation": 0.9822817566969176, "1000–3000 Hz_10ms_modulation_mae_db": 1.186740164595909, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0068707668140198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08345172056517482, "1000–3000 Hz_40ms_envelope_correlation": 0.994097413085962, "1000–3000 Hz_40ms_modulation_mae_db": 0.6984704531494993, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9062636689470607, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5040487806017842, "3000–6000 Hz_10ms_envelope_correlation": 0.8110618835149621, "3000–6000 Hz_10ms_modulation_mae_db": 2.6369299344641655, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2493772825740115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4244252543462986, "3000–6000 Hz_40ms_envelope_correlation": 0.8444475315639852, "3000–6000 Hz_40ms_modulation_mae_db": 3.638146228341539, "mr_stft_db": 5.752533995363829, "band_mae_db": 0.4014775897388933, "rms_gain_error_db": -0.16275706573659643, "output_peak": 0.6783403219316767}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.150224442687217, "全帯域_10ms_rms_envelope_nrmse": 0.12425588531743735, "全帯域_10ms_envelope_correlation": 0.9853178096952827, "全帯域_10ms_modulation_mae_db": 0.7945517024763137, "全帯域_40ms_log_rms_mae_db": 0.6059559111586791, "全帯域_40ms_rms_envelope_nrmse": 0.07229188412412024, "全帯域_40ms_envelope_correlation": 0.9945842222387278, "全帯域_40ms_modulation_mae_db": 0.26569098659850526, "300–1000 Hz_10ms_log_rms_mae_db": 1.8176387684911612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1573237661356003, "300–1000 Hz_10ms_envelope_correlation": 0.9789578391993764, "300–1000 Hz_10ms_modulation_mae_db": 1.2628390343033369, "300–1000 Hz_40ms_log_rms_mae_db": 1.0111093899133752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08788330431492446, "300–1000 Hz_40ms_envelope_correlation": 0.9927179278189364, "300–1000 Hz_40ms_modulation_mae_db": 0.45651691381916876, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7927819503991678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15301792030652062, "1000–3000 Hz_10ms_envelope_correlation": 0.9816315485904049, "1000–3000 Hz_10ms_modulation_mae_db": 0.7186561912924591, "1000–3000 Hz_40ms_log_rms_mae_db": 1.023694977046743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09312207881164394, "1000–3000 Hz_40ms_envelope_correlation": 0.9926094871970289, "1000–3000 Hz_40ms_modulation_mae_db": 0.6468205055184505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.905977411754168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5104892278419936, "3000–6000 Hz_10ms_envelope_correlation": 0.8061815867916198, "3000–6000 Hz_10ms_modulation_mae_db": 2.4484675836006278, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2640204528830274, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4284456200365053, "3000–6000 Hz_40ms_envelope_correlation": 0.8415401327302198, "3000–6000 Hz_40ms_modulation_mae_db": 3.4692039802353185, "mr_stft_db": 5.752468680431066, "band_mae_db": 0.3390824453812303, "rms_gain_error_db": 0.09505423691533996, "output_peak": 0.7816325655093084}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.854531929656164, "spectral_convergence": 0.37194238081492487, "spectral_convergence_status": "ok", "amplitude_floor": 2.051312383514402e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.565225157501371, "spectral_convergence": 0.4047771625551727, "spectral_convergence_status": "ok", "amplitude_floor": 1.425885904641783e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.849279944230245, "spectral_convergence": 0.39251981785935175, "spectral_convergence_status": "ok", "amplitude_floor": 8.483800566678623e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001304494586920238, "output_power": 0.00014989203123126982, "delta_db": 0.6033626358588154, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014301207186140645, "output_power": 0.0014560652268533441, "delta_db": 0.07808131898699025, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.004308309383381418, "output_power": 0.00428365315565141, "delta_db": -0.02492583735824884, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.001304992137209625, "output_power": 0.0012755135657599346, "delta_db": -0.09922813065738764, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002442775973124686, "output_power": 0.00024162961780559682, "delta_db": -0.04733472714911291, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0021123865382114e-05, "output_power": 1.2469156795593543e-05, "delta_db": 0.9492065585444552, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3003565347591684, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.005234483859750754, "pitch": null, "output_rms_dbfs": -21.296415895036297, "output_peak": 0.8111830375274082, "reference_rms_dbfs": -21.29118141117655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.756345677129261}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002631236493882886, "log_rms_mae_db": 1.1650715052372727, "rms_envelope_nrmse": 0.11931808367960137, "envelope_correlation": 0.9861483671880863, "rms_gain_error_db": -0.0052343159222348465, "reference_mean_power": 0.007434364903539191, "output_mean_power": 0.007425410064076636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.450307530079474e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015621941098031631, "output_power": 0.0015840997655110227, "reference_active": true, "delta_db": 0.06047533700861044}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031400948871223737, "output_power": 0.0003413912504806854, "reference_active": true, "delta_db": 0.36309614638793286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9119711962951405e-05, "output_power": 7.625049319716498e-05, "reference_active": true, "delta_db": 1.909868456496671}], "active_band_count": 3, "active_band_mae_db": 0.7778133132977381}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024033810696394756, "log_rms_mae_db": 0.6096407966130737, "rms_envelope_nrmse": 0.06425969644217386, "envelope_correlation": 0.9955201864634498, "rms_gain_error_db": -0.005234868185525739, "reference_mean_power": 0.007471745584669175, "output_mean_power": 0.0074627447704499125}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.136699874330791e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014162642987993615, "output_power": 0.001465772194515339, "reference_active": true, "delta_db": 0.14922171341792595}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001495506387635467, "output_power": 0.00016205466220015072, "reference_active": true, "delta_db": 0.34873257560572934}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0546566248780076e-06, "output_power": 4.629393653082301e-06, "reference_active": true, "delta_db": 1.8056171377632992}], "active_band_count": 3, "active_band_mae_db": 0.7678571422623182}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002303175773961387, "log_rms_mae_db": 1.8312489209575047, "rms_envelope_nrmse": 0.14414953615413398, "envelope_correlation": 0.9821472636352695, "rms_gain_error_db": -0.023608034474428913, "reference_mean_power": 0.004944923373977616, "output_mean_power": 0.004918115941634734}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6955719701225893e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012875788279854436, "output_power": 0.001274736023973817, "reference_active": true, "delta_db": -0.043535677259856284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026650258378785487, "output_power": 0.00029463644282155235, "reference_active": true, "delta_db": 0.43585038651233243}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0302791253002337e-05, "output_power": 6.774085775434112e-05, "reference_active": true, "delta_db": 3.4936805739478327}], "active_band_count": 3, "active_band_mae_db": 1.3243555459066738}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020459712012168156, "log_rms_mae_db": 0.9760663811510946, "rms_envelope_nrmse": 0.07741518797930917, "envelope_correlation": 0.9942719896105542, "rms_gain_error_db": -0.02360798910244944, "reference_mean_power": 0.004969790940376444, "output_mean_power": 0.004942848747552811}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.46585734788702e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011840763589577269, "output_power": 0.001203796962057034, "reference_active": true, "delta_db": 0.07173532828706178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012134699925967192, "output_power": 0.0001290358872060571, "reference_active": true, "delta_db": 0.26681470911971894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.482336680977429e-06, "output_power": 4.150624917157757e-06, "reference_active": true, "delta_db": 2.2325280410735404}], "active_band_count": 3, "active_band_mae_db": 0.8570260261601069}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014083569934759876, "log_rms_mae_db": 1.7394379202494066, "rms_envelope_nrmse": 0.15225105068839406, "envelope_correlation": 0.9817615308391471, "rms_gain_error_db": -0.2064360095972684, "reference_mean_power": 0.0013181968247743324, "output_mean_power": 0.001257004001579419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.706116677050846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003562978060875054, "output_power": 0.0003357833670973611, "reference_active": true, "delta_db": -0.2575396845909772}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927249258823322e-05, "output_power": 7.156654133328327e-05, "reference_active": true, "delta_db": 0.1414921428269078}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3444445693793863e-05, "output_power": 1.819471743379075e-05, "reference_active": true, "delta_db": 1.3140241409294924}], "active_band_count": 3, "active_band_mae_db": 0.5710186561157925}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010879087849669345, "log_rms_mae_db": 0.9995303817827209, "rms_envelope_nrmse": 0.09129109321107112, "envelope_correlation": 0.9931683532826713, "rms_gain_error_db": -0.206438768079314, "reference_mean_power": 0.0013248216670349005, "output_mean_power": 0.0012633205055587707}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.107952744901008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003355694425017771, "output_power": 0.00032044742227308284, "reference_active": true, "delta_db": -0.20025623992126293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.409137696567018e-05, "output_power": 3.715518185279046e-05, "reference_active": true, "delta_db": 0.3737484833532335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.097157965658087e-07, "output_power": 1.0902249101770552e-06, "reference_active": true, "delta_db": 1.2918348893799905}], "active_band_count": 3, "active_band_mae_db": 0.6219465375514956}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.557718084358035e-05, "log_rms_mae_db": 1.818690569533946, "rms_envelope_nrmse": 0.5083078454906332, "envelope_correlation": 0.8098615961373651, "rms_gain_error_db": 0.8615431277456664, "reference_mean_power": 2.3934708718272692e-05, "output_mean_power": 2.9186529646606116e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3699291941082093e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.018836822609012e-06, "output_power": 5.351970253338397e-06, "reference_active": true, "delta_db": 1.2441331809368976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.745214583455981e-06, "output_power": 6.859803992422779e-06, "reference_active": true, "delta_db": 1.6005585018803754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3999346848964708e-06, "output_power": 4.772933263730049e-06, "reference_active": true, "delta_db": 5.326775882167678}], "active_band_count": 3, "active_band_mae_db": 2.7238225216616505}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.6234588308267947e-05, "log_rms_mae_db": 1.1856476590659832, "rms_envelope_nrmse": 0.4269784368707315, "envelope_correlation": 0.844405626442713, "rms_gain_error_db": 0.861535803156855, "reference_mean_power": 2.405439007374668e-05, "output_mean_power": 2.9332422350220572e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1165079820636279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.688364032412796e-06, "output_power": 7.042946371355314e-06, "reference_active": true, "delta_db": 2.809206031727562}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313243490564173e-06, "output_power": 5.581662849619995e-06, "reference_active": true, "delta_db": 2.265102467663888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.200263616160695e-07, "output_power": 4.976841216177876e-07, "reference_active": true, "delta_db": 6.17677143560326}], "active_band_count": 3, "active_band_mae_db": 3.75035997833157}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8572646438691762, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4326f33d9d2d22adee007fb1fe852ace007d60f99d7bfb6d3a9bf05b099b38a9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9824810174811742, "standard_deviation": 0.0017217326543811416, "minimum": 0.9796408122420086, "maximum": 0.9848266799334956, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7788894387998382, "standard_deviation": 0.026009203184416393, "minimum": 1.7394379202494066, "maximum": 1.8157983951043772, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6290785081279073, "standard_deviation": 0.3338595773104996, "minimum": 0.04021512035182186, "maximum": 1.186740164595909, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14925702987466183, "standard_deviation": 0.0077741025441037965, "minimum": 0.1393105071804068, "maximum": 0.16283461135014424, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9936471545664176, "standard_deviation": 0.000897140336127103, "minimum": 0.9924417664384032, "maximum": 0.9951093614015057, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0358338553327089, "standard_deviation": 0.03139554293315668, "minimum": 0.9995303817827209, "maximum": 1.0826379960942942, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5612349339669919, "standard_deviation": 0.23870518803581037, "minimum": 0.13203977741137574, "maximum": 0.8763094024942252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08753725414754936, "standard_deviation": 0.00656730209805844, "minimum": 0.07672603723357864, "maximum": 0.09430382931986125, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8112041253965484, "standard_deviation": 0.0055275798952527325, "minimum": 0.8040528941008563, "maximum": 0.822577229508949, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.866042903373993, "standard_deviation": 0.03525508453919768, "minimum": 1.818690569533946, "maximum": 1.9062636689470607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.598178024837892, "standard_deviation": 0.10606866626716878, "minimum": 2.4484675836006278, "maximum": 2.7238225216616505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.505793996800752, "standard_deviation": 0.004226587045014566, "minimum": 0.49883708784564773, "maximum": 0.511757102100624, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8451383492078143, "standard_deviation": 0.0055379141974461615, "minimum": 0.8403926847361464, "maximum": 0.858160289367483, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2305810612665398, "standard_deviation": 0.03591157533553314, "minimum": 1.17845679355807, "maximum": 1.2788365129446368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.646046371985693, "standard_deviation": 0.11192946529778247, "minimum": 3.4692039802353185, "maximum": 3.75035997833157, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.4253217781660667, "standard_deviation": 0.003160562912055019, "minimum": 0.41920333579252883, "maximum": 0.4284456200365053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9818289509461181, "standard_deviation": 0.0017062287169665714, "minimum": 0.9789578391993764, "maximum": 0.9839436251852183, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8013475874553717, "standard_deviation": 0.051348933921721136, "minimum": 1.735748381999071, "maximum": 1.8910105453387516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0399880725629114, "standard_deviation": 0.4946903072924415, "minimum": 0.5574686610728284, "maximum": 2.036257025268809, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14573144528426699, "standard_deviation": 0.006861370802743252, "minimum": 0.13729596052057766, "maximum": 0.1573237661356003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9937152399263536, "standard_deviation": 0.0021817420153756478, "minimum": 0.9892244889176696, "maximum": 0.9961256711728934, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0027812215098333, "standard_deviation": 0.06150402767134772, "minimum": 0.9203721083080119, "maximum": 1.0864744846771768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.537023650494801, "standard_deviation": 0.41016666880739105, "minimum": 0.17231041292043522, "maximum": 1.407827882059883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08151026256051315, "standard_deviation": 0.01410070777826438, "minimum": 0.06372930072455163, "maximum": 0.10901523034082854, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.34710140191703237, "standard_deviation": 0.03691842647497693, "minimum": 0.3003565347591684, "maximum": 0.4014775897388933, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.754106944642154, "standard_deviation": 0.0034080588983372373, "minimum": 5.7495010591024425, "maximum": 5.760411286294203, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.728496149956589, "standard_deviation": 0.06589909674294402, "minimum": 0.6064615446238055, "maximum": 0.8111830375274082, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.053489647824964265, "standard_deviation": 0.09445463040216182, "minimum": -0.17407591932633082, "maximum": 0.09505423691533996, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9870325279449109, "standard_deviation": 0.0010516135697110132, "minimum": 0.9853178096952827, "maximum": 0.98822791436131, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.157073763584123, "standard_deviation": 0.022936883073938492, "minimum": 1.12268138440065, "maximum": 1.1901954876751717, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6562035039981556, "standard_deviation": 0.366567483168751, "minimum": 0.16670831205793527, "maximum": 1.2782082284445908, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11561468976400455, "standard_deviation": 0.004933804680875288, "minimum": 0.1099747475681772, "maximum": 0.12425588531743735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9958487629515651, "standard_deviation": 0.001471884413551121, "minimum": 0.9930836172994014, "maximum": 0.9977371734485453, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6222048664269293, "standard_deviation": 0.03199469472011987, "minimum": 0.571100557279644, "maximum": 0.6715462451276945, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4618701761298616, "standard_deviation": 0.31930920687828324, "minimum": 0.10110625290176455, "maximum": 0.9462506819577922, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06201501550563673, "standard_deviation": 0.011045387924200258, "minimum": 0.045963342614124006, "maximum": 0.08143915161579414, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0031", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.802995204925537, "predicted_mos": 1.802995204925537, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5fab8e9d8d5dcd48e2db58b15d92fd86deefa9ad7d45bb29934b52f0c9d107b5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.2424125000834465, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1250415343905724, "全帯域_10ms_rms_envelope_nrmse": 0.11034294166791345, "全帯域_10ms_envelope_correlation": 0.9884782983736937, "全帯域_10ms_modulation_mae_db": 0.12390429961112805, "全帯域_40ms_log_rms_mae_db": 0.5935586779971588, "全帯域_40ms_rms_envelope_nrmse": 0.06818046555430249, "全帯域_40ms_envelope_correlation": 0.9952446913206716, "全帯域_40ms_modulation_mae_db": 0.2528202825164753, "300–1000 Hz_10ms_log_rms_mae_db": 1.8356428685741855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13328499394761095, "300–1000 Hz_10ms_envelope_correlation": 0.9848937244398297, "300–1000 Hz_10ms_modulation_mae_db": 0.34674314423413044, "300–1000 Hz_40ms_log_rms_mae_db": 1.0159289659418018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07946904967564478, "300–1000 Hz_40ms_envelope_correlation": 0.994098089416172, "300–1000 Hz_40ms_modulation_mae_db": 0.41887347372576555, "1000–3000 Hz_10ms_log_rms_mae_db": 2.059554627803852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17121364979015036, "1000–3000 Hz_10ms_envelope_correlation": 0.9768275374071629, "1000–3000 Hz_10ms_modulation_mae_db": 0.38898323905018556, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1969597082892534, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10105296886071138, "1000–3000 Hz_40ms_envelope_correlation": 0.991202217395395, "1000–3000 Hz_40ms_modulation_mae_db": 0.42706630367583887, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5396410242102654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5682681082197503, "3000–6000 Hz_10ms_envelope_correlation": 0.7833193032780169, "3000–6000 Hz_10ms_modulation_mae_db": 2.874824671900182, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7763542958220175, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.46672800587954133, "3000–6000 Hz_40ms_envelope_correlation": 0.836576690429935, "3000–6000 Hz_40ms_modulation_mae_db": 3.804885925083847, "mr_stft_db": 6.2704493607257525, "band_mae_db": 0.5857772384234369, "rms_gain_error_db": 0.08788794533673767, "output_peak": 0.8550769804117537}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0877355920469796, "全帯域_10ms_rms_envelope_nrmse": 0.10366218166608895, "全帯域_10ms_envelope_correlation": 0.9895962029720813, "全帯域_10ms_modulation_mae_db": 0.23929788234759797, "全帯域_40ms_log_rms_mae_db": 0.5829164395796298, "全帯域_40ms_rms_envelope_nrmse": 0.052650701720858416, "全帯域_40ms_envelope_correlation": 0.9971972698167056, "全帯域_40ms_modulation_mae_db": 0.2995510387192932, "300–1000 Hz_10ms_log_rms_mae_db": 1.8148584085130757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13182696466891142, "300–1000 Hz_10ms_envelope_correlation": 0.9854497619926555, "300–1000 Hz_10ms_modulation_mae_db": 0.458050165868262, "300–1000 Hz_40ms_log_rms_mae_db": 1.0450237731431495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07984324238587827, "300–1000 Hz_40ms_envelope_correlation": 0.9946288350811817, "300–1000 Hz_40ms_modulation_mae_db": 0.29387375827790174, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0868285033341594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1705474537488152, "1000–3000 Hz_10ms_envelope_correlation": 0.9769275049791586, "1000–3000 Hz_10ms_modulation_mae_db": 0.7949969439747834, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2152184506865704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09356327868586382, "1000–3000 Hz_40ms_envelope_correlation": 0.9924976526671603, "1000–3000 Hz_40ms_modulation_mae_db": 0.6400317605236342, "3000–6000 Hz_10ms_log_rms_mae_db": 2.626132393156561, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5710096265293755, "3000–6000 Hz_10ms_envelope_correlation": 0.784530084422444, "3000–6000 Hz_10ms_modulation_mae_db": 2.654885874213162, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9033854944015043, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.46623118862349505, "3000–6000 Hz_40ms_envelope_correlation": 0.8404490088172739, "3000–6000 Hz_40ms_modulation_mae_db": 3.566090557205452, "mr_stft_db": 6.242896469732813, "band_mae_db": 0.5676105488713677, "rms_gain_error_db": -0.13384639288546918, "output_peak": 0.6956246745947273}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1309286866936865, "全帯域_10ms_rms_envelope_nrmse": 0.09839232590263206, "全帯域_10ms_envelope_correlation": 0.9905678802242008, "全帯域_10ms_modulation_mae_db": 0.12263361104314867, "全帯域_40ms_log_rms_mae_db": 0.58488908804217, "全帯域_40ms_rms_envelope_nrmse": 0.05656484639749273, "全帯域_40ms_envelope_correlation": 0.9965955787086838, "全帯域_40ms_modulation_mae_db": 0.19965161442408533, "300–1000 Hz_10ms_log_rms_mae_db": 1.8183337044119832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13297067004048638, "300–1000 Hz_10ms_envelope_correlation": 0.984981792142386, "300–1000 Hz_10ms_modulation_mae_db": 0.7423940445586491, "300–1000 Hz_40ms_log_rms_mae_db": 1.0210926152698765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0773858810846723, "300–1000 Hz_40ms_envelope_correlation": 0.9947435634673204, "300–1000 Hz_40ms_modulation_mae_db": 0.4522013428605461, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0214559677778636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1655949681983684, "1000–3000 Hz_10ms_envelope_correlation": 0.9787543023190539, "1000–3000 Hz_10ms_modulation_mae_db": 0.7840654160780874, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1530301164230197, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08447980614890097, "1000–3000 Hz_40ms_envelope_correlation": 0.9940336639065094, "1000–3000 Hz_40ms_modulation_mae_db": 0.4750674888894461, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6250465605890407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5607590219725591, "3000–6000 Hz_10ms_envelope_correlation": 0.7807895201226751, "3000–6000 Hz_10ms_modulation_mae_db": 2.6927723524238623, "3000–6000 Hz_40ms_log_rms_mae_db": 1.833182320483505, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4590610396154517, "3000–6000 Hz_40ms_envelope_correlation": 0.8340273044451924, "3000–6000 Hz_40ms_modulation_mae_db": 3.511680076729681, "mr_stft_db": 6.237964740782625, "band_mae_db": 0.544311433764448, "rms_gain_error_db": -0.08178487751077711, "output_peak": 0.6870598966001523}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1639034685613838, "全帯域_10ms_rms_envelope_nrmse": 0.10869998683167827, "全帯域_10ms_envelope_correlation": 0.988515898442483, "全帯域_10ms_modulation_mae_db": 0.8412031387901742, "全帯域_40ms_log_rms_mae_db": 0.6438393691457084, "全帯域_40ms_rms_envelope_nrmse": 0.054387678478083394, "全帯域_40ms_envelope_correlation": 0.9969796480635315, "全帯域_40ms_modulation_mae_db": 0.25090029805524444, "300–1000 Hz_10ms_log_rms_mae_db": 1.8673972343846688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1477781321992862, "300–1000 Hz_10ms_envelope_correlation": 0.981265264862105, "300–1000 Hz_10ms_modulation_mae_db": 1.816403496077059, "300–1000 Hz_40ms_log_rms_mae_db": 1.045662005371659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06756409058687785, "300–1000 Hz_40ms_envelope_correlation": 0.9959933760258859, "300–1000 Hz_40ms_modulation_mae_db": 0.558457498989973, "1000–3000 Hz_10ms_log_rms_mae_db": 2.117565119510908, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16975349588793323, "1000–3000 Hz_10ms_envelope_correlation": 0.9771454296509499, "1000–3000 Hz_10ms_modulation_mae_db": 0.5036603708922405, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2549190464509317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09444160669132393, "1000–3000 Hz_40ms_envelope_correlation": 0.9925094147973753, "1000–3000 Hz_40ms_modulation_mae_db": 0.4298747605905769, "3000–6000 Hz_10ms_log_rms_mae_db": 2.637977257645796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5697923058875939, "3000–6000 Hz_10ms_envelope_correlation": 0.779418769620545, "3000–6000 Hz_10ms_modulation_mae_db": 2.5982449310419757, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8959013652615322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4656096723343962, "3000–6000 Hz_40ms_envelope_correlation": 0.8348254316178496, "3000–6000 Hz_40ms_modulation_mae_db": 3.4723575446246735, "mr_stft_db": 6.254524482306896, "band_mae_db": 0.6501896036942798, "rms_gain_error_db": -0.12419338795929673, "output_peak": 0.71707660182539}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1473578346535653, "全帯域_10ms_rms_envelope_nrmse": 0.11249477607228295, "全帯域_10ms_envelope_correlation": 0.9879285225126104, "全帯域_10ms_modulation_mae_db": 0.6192784988306043, "全帯域_40ms_log_rms_mae_db": 0.6054802321038423, "全帯域_40ms_rms_envelope_nrmse": 0.060987941210226285, "全帯域_40ms_envelope_correlation": 0.9965078044217628, "全帯域_40ms_modulation_mae_db": 0.27814226077304943, "300–1000 Hz_10ms_log_rms_mae_db": 1.963121884974282, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1511015440625784, "300–1000 Hz_10ms_envelope_correlation": 0.9807615956634033, "300–1000 Hz_10ms_modulation_mae_db": 0.9025530048299335, "300–1000 Hz_40ms_log_rms_mae_db": 1.0861867462668586, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08455249431528507, "300–1000 Hz_40ms_envelope_correlation": 0.9941215145481095, "300–1000 Hz_40ms_modulation_mae_db": 0.6121583036248489, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0985969563658946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15787429308743728, "1000–3000 Hz_10ms_envelope_correlation": 0.9803766643198644, "1000–3000 Hz_10ms_modulation_mae_db": 0.7403979224342553, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2211273134092377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08900258849135548, "1000–3000 Hz_40ms_envelope_correlation": 0.9935835358303942, "1000–3000 Hz_40ms_modulation_mae_db": 0.6697691380626877, "3000–6000 Hz_10ms_log_rms_mae_db": 2.608828803388974, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5644659022091855, "3000–6000 Hz_10ms_envelope_correlation": 0.7790773478726194, "3000–6000 Hz_10ms_modulation_mae_db": 2.5369783377492734, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8130178648832267, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4539520772631191, "3000–6000 Hz_40ms_envelope_correlation": 0.8391518828733883, "3000–6000 Hz_40ms_modulation_mae_db": 3.395187665823491, "mr_stft_db": 6.25368354095524, "band_mae_db": 0.5853121370098447, "rms_gain_error_db": -0.20782471759555593, "output_peak": 0.7194445349178075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.111339245780435, "全帯域_10ms_rms_envelope_nrmse": 0.12777167784305235, "全帯域_10ms_envelope_correlation": 0.984275847749139, "全帯域_10ms_modulation_mae_db": 0.43593740689820365, "全帯域_40ms_log_rms_mae_db": 0.570356961901961, "全帯域_40ms_rms_envelope_nrmse": 0.07833987562714949, "全帯域_40ms_envelope_correlation": 0.9934595285523279, "全帯域_40ms_modulation_mae_db": 0.30530319677825907, "300–1000 Hz_10ms_log_rms_mae_db": 1.8535951982652188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16018490268069288, "300–1000 Hz_10ms_envelope_correlation": 0.9779298867721975, "300–1000 Hz_10ms_modulation_mae_db": 0.4325124324823768, "300–1000 Hz_40ms_log_rms_mae_db": 1.0498270229991746, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09901244172441394, "300–1000 Hz_40ms_envelope_correlation": 0.9906591179046098, "300–1000 Hz_40ms_modulation_mae_db": 0.18714746542674968, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1394982230569206, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17608110660447396, "1000–3000 Hz_10ms_envelope_correlation": 0.9760627089958245, "1000–3000 Hz_10ms_modulation_mae_db": 0.9608451899157516, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2496361083356713, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10235071168599688, "1000–3000 Hz_40ms_envelope_correlation": 0.9912596918668569, "1000–3000 Hz_40ms_modulation_mae_db": 0.5507152812064516, "3000–6000 Hz_10ms_log_rms_mae_db": 2.613108098116545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5551191422712655, "3000–6000 Hz_10ms_envelope_correlation": 0.7860768642143829, "3000–6000 Hz_10ms_modulation_mae_db": 2.6311951388561146, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8625640313513003, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4564315312016104, "3000–6000 Hz_40ms_envelope_correlation": 0.8366287161809335, "3000–6000 Hz_40ms_modulation_mae_db": 3.441371115198374, "mr_stft_db": 6.254000311849434, "band_mae_db": 0.5631643712228632, "rms_gain_error_db": 0.03819206948795553, "output_peak": 0.9166098290805559}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0983938832252123, "全帯域_10ms_rms_envelope_nrmse": 0.09431293445490407, "全帯域_10ms_envelope_correlation": 0.9913676885220526, "全帯域_10ms_modulation_mae_db": 0.3325505998771148, "全帯域_40ms_log_rms_mae_db": 0.5533210910145782, "全帯域_40ms_rms_envelope_nrmse": 0.04904987022997725, "全帯域_40ms_envelope_correlation": 0.9974066224774305, "全帯域_40ms_modulation_mae_db": 0.2883115535101635, "300–1000 Hz_10ms_log_rms_mae_db": 1.7940543110956062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11788805621869351, "300–1000 Hz_10ms_envelope_correlation": 0.9880861336597812, "300–1000 Hz_10ms_modulation_mae_db": 0.7264557846577754, "300–1000 Hz_40ms_log_rms_mae_db": 1.0081905668884408, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07013402879622531, "300–1000 Hz_40ms_envelope_correlation": 0.9953241475656258, "300–1000 Hz_40ms_modulation_mae_db": 0.5084064768207728, "1000–3000 Hz_10ms_log_rms_mae_db": 2.002711525115026, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1568963725672738, "1000–3000 Hz_10ms_envelope_correlation": 0.9804985462110329, "1000–3000 Hz_10ms_modulation_mae_db": 0.7052191784125438, "1000–3000 Hz_40ms_log_rms_mae_db": 1.161011608551643, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07773722083771974, "1000–3000 Hz_40ms_envelope_correlation": 0.9948839501651878, "1000–3000 Hz_40ms_modulation_mae_db": 0.44111609801821156, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6777487722532234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5921177446444931, "3000–6000 Hz_10ms_envelope_correlation": 0.7678935523090371, "3000–6000 Hz_10ms_modulation_mae_db": 2.837073306743345, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8542467796018363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.484340432786689, "3000–6000 Hz_40ms_envelope_correlation": 0.8272781760162174, "3000–6000 Hz_40ms_modulation_mae_db": 3.6960591359779635, "mr_stft_db": 6.2490140438655475, "band_mae_db": 0.5596594859088014, "rms_gain_error_db": 0.016747904541364874, "output_peak": 0.7315064045476116}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1247316826708857, "全帯域_10ms_rms_envelope_nrmse": 0.11619977127910128, "全帯域_10ms_envelope_correlation": 0.9869971778090819, "全帯域_10ms_modulation_mae_db": 0.692414250266478, "全帯域_40ms_log_rms_mae_db": 0.5877515612525266, "全帯域_40ms_rms_envelope_nrmse": 0.06683341012467024, "全帯域_40ms_envelope_correlation": 0.9952677066641532, "全帯域_40ms_modulation_mae_db": 0.3106032639840102, "300–1000 Hz_10ms_log_rms_mae_db": 1.841234473889559, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15967672379938944, "300–1000 Hz_10ms_envelope_correlation": 0.9784982299943923, "300–1000 Hz_10ms_modulation_mae_db": 1.3212681855961668, "300–1000 Hz_40ms_log_rms_mae_db": 1.037887568952174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.100705261974837, "300–1000 Hz_40ms_envelope_correlation": 0.9906321753265523, "300–1000 Hz_40ms_modulation_mae_db": 0.7319071219664474, "1000–3000 Hz_10ms_log_rms_mae_db": 2.111985193845342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14929091974377154, "1000–3000 Hz_10ms_envelope_correlation": 0.9826138510081031, "1000–3000 Hz_10ms_modulation_mae_db": 0.41683959018779576, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2394372613845785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08669217556991063, "1000–3000 Hz_40ms_envelope_correlation": 0.9941061621115515, "1000–3000 Hz_40ms_modulation_mae_db": 0.38822115256792306, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5930618047786753, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5594412850040353, "3000–6000 Hz_10ms_envelope_correlation": 0.7901976619527611, "3000–6000 Hz_10ms_modulation_mae_db": 2.9508752701111924, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8093637239356228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.45913891325872175, "3000–6000 Hz_40ms_envelope_correlation": 0.8422363958965885, "3000–6000 Hz_40ms_modulation_mae_db": 3.8268960926169484, "mr_stft_db": 6.240883457787546, "band_mae_db": 0.5933226148832197, "rms_gain_error_db": 0.04966821013065359, "output_peak": 0.7950505936391665}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.381293371921063, "spectral_convergence": 0.38034168602261476, "spectral_convergence_status": "ok", "amplitude_floor": 2.051312383514402e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.144774415778777, "spectral_convergence": 0.39529276784761086, "spectral_convergence_status": "ok", "amplitude_floor": 1.425885904641783e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.285280294477418, "spectral_convergence": 0.3988312641380039, "spectral_convergence_status": "ok", "amplitude_floor": 8.483800566678623e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001304494586920238, "output_power": 0.0001595927643312765, "delta_db": 0.8757091596783334, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014301207186140645, "output_power": 0.0014269611595442307, "delta_db": -0.009605462406478308, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.004308309383381418, "output_power": 0.004435778673337484, "delta_db": 0.1266298574932165, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.001304992137209625, "output_power": 0.0012862100155650368, "delta_db": -0.06296007927524401, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002442775973124686, "output_power": 0.00025462176898074184, "delta_db": 0.18011891397468938, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0021123865382114e-05, "output_power": 1.6860887318334346e-05, "delta_db": 2.2596399577126594, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5857772384234369, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.08788794533673767, "pitch": null, "output_rms_dbfs": -21.20329346583981, "output_peak": 0.8550769804117537, "reference_rms_dbfs": -21.29118141117655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.2704493607257525}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002631236493882886, "log_rms_mae_db": 1.1250415343905724, "rms_envelope_nrmse": 0.11034294166791345, "envelope_correlation": 0.9884782983736937, "rms_gain_error_db": 0.08788792473910843, "reference_mean_power": 0.007434364903539191, "output_mean_power": 0.0075863463520424}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.450307530079474e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015621941098031631, "output_power": 0.0016148712347759488, "reference_active": true, "delta_db": 0.1440290266401957}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031400948871223737, "output_power": 0.0002987686852898234, "reference_active": true, "delta_db": -0.21607695732415522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9119711962951405e-05, "output_power": 4.898861033554126e-05, "reference_active": true, "delta_db": -0.011606914869033279}], "active_band_count": 3, "active_band_mae_db": 0.12390429961112805}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024033810696394756, "log_rms_mae_db": 0.5935586779971588, "rms_envelope_nrmse": 0.06818046555430249, "envelope_correlation": 0.9952446913206716, "rms_gain_error_db": 0.08788860761498754, "reference_mean_power": 0.007471745584669175, "output_mean_power": 0.007624492408950748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.136699874330791e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014162642987993615, "output_power": 0.0014742349980230642, "reference_active": true, "delta_db": 0.17422409453604037}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001495506387635467, "output_power": 0.0001499432798956072, "reference_active": true, "delta_db": 0.011387341384500684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0546566248780076e-06, "output_power": 2.677179291238793e-06, "reference_active": true, "delta_db": -0.5728494116288849}], "active_band_count": 3, "active_band_mae_db": 0.2528202825164753}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002303175773961387, "log_rms_mae_db": 1.8356428685741855, "rms_envelope_nrmse": 0.13328499394761095, "envelope_correlation": 0.9848937244398297, "rms_gain_error_db": 0.01842345687533026, "reference_mean_power": 0.004944923373977616, "output_mean_power": 0.004965945075987156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6955719701225893e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012875788279854436, "output_power": 0.0012963710963312352, "reference_active": true, "delta_db": 0.0295551266436783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026650258378785487, "output_power": 0.00023628310738653537, "reference_active": true, "delta_db": -0.5226875026145904}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0302791253002337e-05, "output_power": 3.390637137027057e-05, "reference_active": true, "delta_db": 0.4879868034441225}], "active_band_count": 3, "active_band_mae_db": 0.34674314423413044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020459712012168156, "log_rms_mae_db": 1.0159289659418018, "rms_envelope_nrmse": 0.07946904967564478, "envelope_correlation": 0.994098089416172, "rms_gain_error_db": 0.01842332559854717, "reference_mean_power": 0.004969790940376444, "output_mean_power": 0.00499091820773509}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.46585734788702e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011840763589577269, "output_power": 0.0011999808782891404, "reference_active": true, "delta_db": 0.0579461545344242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012134699925967192, "output_power": 0.0001164641706309142, "reference_active": true, "delta_db": -0.17836702991606904}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.482336680977429e-06, "output_power": 1.9625917149031238e-06, "reference_active": true, "delta_db": -1.0203072367268033}], "active_band_count": 3, "active_band_mae_db": 0.41887347372576555}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014083569934759876, "log_rms_mae_db": 2.059554627803852, "rms_envelope_nrmse": 0.17121364979015036, "envelope_correlation": 0.9768275374071629, "rms_gain_error_db": -0.04678553459133952, "reference_mean_power": 0.0013181968247743324, "output_mean_power": 0.0013040724130244862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.706116677050846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003562978060875054, "output_power": 0.0003530245303823433, "reference_active": true, "delta_db": -0.040082642101925074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927249258823322e-05, "output_power": 6.781618824279996e-05, "reference_active": true, "delta_db": -0.09227439204572407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3444445693793863e-05, "output_power": 1.7060909118255065e-05, "reference_active": true, "delta_db": 1.0345926830029075}], "active_band_count": 3, "active_band_mae_db": 0.38898323905018556}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010879087849669345, "log_rms_mae_db": 1.1969597082892534, "rms_envelope_nrmse": 0.10105296886071138, "envelope_correlation": 0.991202217395395, "rms_gain_error_db": -0.04678220206470817, "reference_mean_power": 0.0013248216670349005, "output_mean_power": 0.0013106272761378719}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.107952744901008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003355694425017771, "output_power": 0.00033591290552383973, "reference_active": true, "delta_db": 0.0044428303638033495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.409137696567018e-05, "output_power": 3.2813870586639284e-05, "reference_active": true, "delta_db": -0.16587081963408778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.097157965658087e-07, "output_power": 1.045733782159498e-06, "reference_active": true, "delta_db": 1.1108852610296254}], "active_band_count": 3, "active_band_mae_db": 0.42706630367583887}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.557718084358035e-05, "log_rms_mae_db": 2.5396410242102654, "rms_envelope_nrmse": 0.5682681082197503, "envelope_correlation": 0.7833193032780169, "rms_gain_error_db": 1.3844606169485718, "reference_mean_power": 2.3934708718272692e-05, "output_mean_power": 3.292109016732764e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3699291941082093e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.018836822609012e-06, "output_power": 6.098931916881664e-06, "reference_active": true, "delta_db": 1.8115341247196048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.745214583455981e-06, "output_power": 6.231882206478515e-06, "reference_active": true, "delta_db": 1.1836337936547432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3999346848964708e-06, "output_power": 5.1172706065235405e-06, "reference_active": true, "delta_db": 5.6293060973261975}], "active_band_count": 3, "active_band_mae_db": 2.874824671900182}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.6234588308267947e-05, "log_rms_mae_db": 1.7763542958220175, "rms_envelope_nrmse": 0.46672800587954133, "envelope_correlation": 0.836576690429935, "rms_gain_error_db": 1.384486836386364, "reference_mean_power": 2.405439007374668e-05, "output_mean_power": 3.308590611028865e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1165079820636279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.688364032412796e-06, "output_power": 7.85067267112445e-06, "reference_active": true, "delta_db": 3.2807309181201116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313243490564173e-06, "output_power": 5.003296217025581e-06, "reference_active": true, "delta_db": 1.790028623823612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.200263616160695e-07, "output_power": 5.172094660055108e-07, "reference_active": true, "delta_db": 6.343898233307819}], "active_band_count": 3, "active_band_mae_db": 3.804885925083847}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8338941610009247, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6897173f8545b321711728296dba753ecbcd6bd3d0eac5bf0a8243fa15bde01b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9786508181113938, "standard_deviation": 0.002310462133000638, "minimum": 0.9760627089958245, "maximum": 0.9826138510081031, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0797745146012456, "standard_deviation": 0.048085078133560076, "minimum": 2.002711525115026, "maximum": 2.1394982230569206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6618759813682054, "standard_deviation": 0.20340804362140907, "minimum": 0.38898323905018556, "maximum": 0.9608451899157516, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16465653245352796, "standard_deviation": 0.009082166514094934, "minimum": 0.14929091974377154, "maximum": 0.17608110660447396, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930095360925538, "standard_deviation": 0.0013593502575010548, "minimum": 0.991202217395395, "maximum": 0.9948839501651878, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2114174516913632, "standard_deviation": 0.03855561751205712, "minimum": 1.1530301164230197, "maximum": 1.2549190464509317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5027327479418462, "standard_deviation": 0.10539929429764608, "minimum": 0.38822115256792306, "maximum": 0.6697691380626877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09116504462147285, "standard_deviation": 0.00835856462501852, "minimum": 0.07773722083771974, "maximum": 0.10235071168599688, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.7814128879740602, "standard_deviation": 0.00659944306211952, "minimum": 0.7678935523090371, "maximum": 0.7901976619527611, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.615193089267385, "standard_deviation": 0.03944589545828853, "minimum": 2.5396410242102654, "maximum": 2.6777487722532234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.722106235379888, "standard_deviation": 0.14742235821525276, "minimum": 2.5369783377492734, "maximum": 2.9508752701111924, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5676216420922822, "standard_deviation": 0.011319718393416788, "minimum": 0.5551191422712655, "maximum": 0.5921177446444931, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8363967007846723, "standard_deviation": 0.004620310980733162, "minimum": 0.8272781760162174, "maximum": 0.8422363958965885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8435019844675682, "standard_deviation": 0.04394891849414801, "minimum": 1.7763542958220175, "maximum": 1.9033854944015043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.589316014157554, "standard_deviation": 0.16653113274566247, "minimum": 3.395187665823491, "maximum": 3.8268960926169484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.463936607620378, "standard_deviation": 0.009524241898896725, "minimum": 0.4539520772631191, "maximum": 0.484340432786689, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9827332986908438, "standard_deviation": 0.0036411632044066563, "minimum": 0.9779298867721975, "maximum": 0.9880861336597812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8485297605135722, "standard_deviation": 0.051748602117788214, "minimum": 1.7940543110956062, "maximum": 1.963121884974282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8432975322880442, "standard_deviation": 0.502346219950271, "minimum": 0.34674314423413044, "maximum": 1.816403496077059, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14183899845220616, "standard_deviation": 0.015125812396628151, "minimum": 0.11788805621869351, "maximum": 0.16018490268069288, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9937751024169321, "standard_deviation": 0.0020283253855539665, "minimum": 0.9906321753265523, "maximum": 0.9959933760258859, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0387249081041419, "standard_deviation": 0.02457432433658981, "minimum": 1.0081905668884408, "maximum": 1.0861867462668586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4703781802116257, "standard_deviation": 0.1740524097899861, "minimum": 0.18714746542674968, "maximum": 0.7319071219664474, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08233331131797932, "standard_deviation": 0.012109504025668495, "minimum": 0.06756409058687785, "maximum": 0.100705261974837, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.5811684292222827, "standard_deviation": 0.03224702239134052, "minimum": 0.544311433764448, "maximum": 0.6501896036942798, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.250427051000732, "standard_deviation": 0.010324384040692771, "minimum": 6.237964740782625, "maximum": 6.2704493607257525, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7646811894521457, "standard_deviation": 0.08313757672980501, "minimum": 0.6870598966001523, "maximum": 0.9166098290805559, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04439415580679841, "standard_deviation": 0.10649101497134761, "minimum": -0.20782471759555593, "maximum": 0.08788794533673767, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9884659395756679, "standard_deviation": 0.0022078235292033663, "minimum": 0.984275847749139, "maximum": 0.9913676885220526, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1236789910028402, "standard_deviation": 0.02483589832513431, "minimum": 1.0877355920469796, "maximum": 1.1639034685613838, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4259024609580562, "standard_deviation": 0.26931365211875646, "minimum": 0.12263361104314867, "maximum": 0.8412031387901742, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10898457446470668, "standard_deviation": 0.01052237480396172, "minimum": 0.09431293445490407, "maximum": 0.12777167784305235, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9960823562531583, "standard_deviation": 0.0013358149440453769, "minimum": 0.9934595285523279, "maximum": 0.9974066224774305, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5902641776296969, "standard_deviation": 0.026626963140708722, "minimum": 0.5533210910145782, "maximum": 0.6438393691457084, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.27316043859507255, "standard_deviation": 0.03722277283788357, "minimum": 0.19965161442408533, "maximum": 0.3106032639840102, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06087434866784504, "standard_deviation": 0.009725097517262601, "minimum": 0.04904987022997725, "maximum": 0.07833987562714949, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0031", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.0776617527008057, "predicted_mos": 2.0776617527008057, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c769cb8ee15fba572896cbe9c79c0e98230ac1aaed1de7ca1a09ffd706115d82", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.709319399669766, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.106906753904374, "全帯域_10ms_rms_envelope_nrmse": 0.11107420834874333, "全帯域_10ms_envelope_correlation": 0.9880481922782575, "全帯域_10ms_modulation_mae_db": 0.8808486852722616, "全帯域_40ms_log_rms_mae_db": 0.5764795469191394, "全帯域_40ms_rms_envelope_nrmse": 0.06425034682823973, "全帯域_40ms_envelope_correlation": 0.9955733726418278, "全帯域_40ms_modulation_mae_db": 0.7497687237989169, "300–1000 Hz_10ms_log_rms_mae_db": 1.604759113455014, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1484148874572268, "300–1000 Hz_10ms_envelope_correlation": 0.9810793143227198, "300–1000 Hz_10ms_modulation_mae_db": 1.3925513134174434, "300–1000 Hz_40ms_log_rms_mae_db": 0.9370938959903897, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08876412391869522, "300–1000 Hz_40ms_envelope_correlation": 0.9925343070899835, "300–1000 Hz_40ms_modulation_mae_db": 0.9868279336660178, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6000576899238868, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17721226964259046, "1000–3000 Hz_10ms_envelope_correlation": 0.9756551729287021, "1000–3000 Hz_10ms_modulation_mae_db": 0.4498308713089601, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9815582476445694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10351335279246981, "1000–3000 Hz_40ms_envelope_correlation": 0.9910120856880005, "1000–3000 Hz_40ms_modulation_mae_db": 0.5096726715463066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7619560494179214, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5129244612111703, "3000–6000 Hz_10ms_envelope_correlation": 0.8194833317124034, "3000–6000 Hz_10ms_modulation_mae_db": 3.0512854423494105, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2452333473232224, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4392656416517366, "3000–6000 Hz_40ms_envelope_correlation": 0.8485478457292429, "3000–6000 Hz_40ms_modulation_mae_db": 4.046154482302202, "mr_stft_db": 5.680702648818941, "band_mae_db": 0.36676099856723665, "rms_gain_error_db": 0.005187815464959623, "output_peak": 0.7867718701071794}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1752015087343635, "全帯域_10ms_rms_envelope_nrmse": 0.1269212835242522, "全帯域_10ms_envelope_correlation": 0.9847837266758479, "全帯域_10ms_modulation_mae_db": 0.5372735195285258, "全帯域_40ms_log_rms_mae_db": 0.6247926040874466, "全帯域_40ms_rms_envelope_nrmse": 0.08133259262476952, "全帯域_40ms_envelope_correlation": 0.9931885314788782, "全帯域_40ms_modulation_mae_db": 0.18146007602412517, "300–1000 Hz_10ms_log_rms_mae_db": 1.633734926552547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16288743408208972, "300–1000 Hz_10ms_envelope_correlation": 0.9778186602349401, "300–1000 Hz_10ms_modulation_mae_db": 1.151058480488794, "300–1000 Hz_40ms_log_rms_mae_db": 0.9960649288287997, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11448177299424502, "300–1000 Hz_40ms_envelope_correlation": 0.9879727928596106, "300–1000 Hz_40ms_modulation_mae_db": 0.38189916031953036, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6297461998624208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16630181741739963, "1000–3000 Hz_10ms_envelope_correlation": 0.9781763827630267, "1000–3000 Hz_10ms_modulation_mae_db": 0.8375229357234254, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9783402394536284, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08365077808298889, "1000–3000 Hz_40ms_envelope_correlation": 0.9940133160028904, "1000–3000 Hz_40ms_modulation_mae_db": 0.5058957345723597, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6493470664159129, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4923895278106378, "3000–6000 Hz_10ms_envelope_correlation": 0.8381896860836384, "3000–6000 Hz_10ms_modulation_mae_db": 3.178084275813245, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1696174533270478, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4207329141140221, "3000–6000 Hz_40ms_envelope_correlation": 0.8653891424528591, "3000–6000 Hz_40ms_modulation_mae_db": 4.185893903764045, "mr_stft_db": 5.688978867446562, "band_mae_db": 0.482429291857646, "rms_gain_error_db": 0.10735122989513986, "output_peak": 0.915853814554273}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1940689502375987, "全帯域_10ms_rms_envelope_nrmse": 0.1165027761227978, "全帯域_10ms_envelope_correlation": 0.986930665588992, "全帯域_10ms_modulation_mae_db": 0.37744069852048207, "全帯域_40ms_log_rms_mae_db": 0.6379654790560197, "全帯域_40ms_rms_envelope_nrmse": 0.0633478141297244, "全帯域_40ms_envelope_correlation": 0.9960647731458816, "全帯域_40ms_modulation_mae_db": 0.1914624383087873, "300–1000 Hz_10ms_log_rms_mae_db": 1.71790728498324, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14247403990628504, "300–1000 Hz_10ms_envelope_correlation": 0.9829055777636345, "300–1000 Hz_10ms_modulation_mae_db": 0.8802644853267627, "300–1000 Hz_40ms_log_rms_mae_db": 1.040510131162445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08304538698483611, "300–1000 Hz_40ms_envelope_correlation": 0.9942115047646474, "300–1000 Hz_40ms_modulation_mae_db": 0.44764494681644695, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6190594477502487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15077277630100536, "1000–3000 Hz_10ms_envelope_correlation": 0.9821805450233745, "1000–3000 Hz_10ms_modulation_mae_db": 0.32907821129649756, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9499218029955812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07636127710887027, "1000–3000 Hz_40ms_envelope_correlation": 0.9954558799163109, "1000–3000 Hz_40ms_modulation_mae_db": 0.12123676493220774, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7299294250285475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4994215763107318, "3000–6000 Hz_10ms_envelope_correlation": 0.8212495350567894, "3000–6000 Hz_10ms_modulation_mae_db": 2.799731990724601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1590239873809958, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.42335178745384505, "3000–6000 Hz_40ms_envelope_correlation": 0.8517232300852541, "3000–6000 Hz_40ms_modulation_mae_db": 3.8132177254257265, "mr_stft_db": 5.6751100891768855, "band_mae_db": 0.45653325074213474, "rms_gain_error_db": -0.1859303721658797, "output_peak": 0.737354194699711}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1758054396883129, "全帯域_10ms_rms_envelope_nrmse": 0.11220914306562918, "全帯域_10ms_envelope_correlation": 0.9876858350580645, "全帯域_10ms_modulation_mae_db": 0.4875115464118471, "全帯域_40ms_log_rms_mae_db": 0.6269836236960425, "全帯域_40ms_rms_envelope_nrmse": 0.05870346690694565, "全帯域_40ms_envelope_correlation": 0.9962825345994075, "全帯域_40ms_modulation_mae_db": 0.19413893635941704, "300–1000 Hz_10ms_log_rms_mae_db": 1.5530559218818898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13813068137057294, "300–1000 Hz_10ms_envelope_correlation": 0.9836062717831363, "300–1000 Hz_10ms_modulation_mae_db": 0.9513607195278633, "300–1000 Hz_40ms_log_rms_mae_db": 0.9133845423860488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08275665743941049, "300–1000 Hz_40ms_envelope_correlation": 0.9936029508098938, "300–1000 Hz_40ms_modulation_mae_db": 0.3418082242493304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6754655729854084, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16036045478099256, "1000–3000 Hz_10ms_envelope_correlation": 0.979685401954549, "1000–3000 Hz_10ms_modulation_mae_db": 0.20248695847389678, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0182735592367183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0974335971581276, "1000–3000 Hz_40ms_envelope_correlation": 0.992079330995022, "1000–3000 Hz_40ms_modulation_mae_db": 0.16996699561195516, "3000–6000 Hz_10ms_log_rms_mae_db": 1.738911698611574, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5089034388960965, "3000–6000 Hz_10ms_envelope_correlation": 0.815334661568739, "3000–6000 Hz_10ms_modulation_mae_db": 2.9718603201018063, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2030749723957268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.43608134229172474, "3000–6000 Hz_40ms_envelope_correlation": 0.843391290807169, "3000–6000 Hz_40ms_modulation_mae_db": 3.9486275774982134, "mr_stft_db": 5.6814844030439815, "band_mae_db": 0.4337951811040215, "rms_gain_error_db": -0.06344146175304644, "output_peak": 0.8911077548468623}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0897890686656517, "全帯域_10ms_rms_envelope_nrmse": 0.09685047621621624, "全帯域_10ms_envelope_correlation": 0.9908889439904598, "全帯域_10ms_modulation_mae_db": 0.17800633568559152, "全帯域_40ms_log_rms_mae_db": 0.6205873149814417, "全帯域_40ms_rms_envelope_nrmse": 0.05298753235179565, "全帯域_40ms_envelope_correlation": 0.9970639973926778, "全帯域_40ms_modulation_mae_db": 0.27058225528867885, "300–1000 Hz_10ms_log_rms_mae_db": 1.5856637137419614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12089620631086709, "300–1000 Hz_10ms_envelope_correlation": 0.9874617915376501, "300–1000 Hz_10ms_modulation_mae_db": 0.34253446240752333, "300–1000 Hz_40ms_log_rms_mae_db": 0.891995599358277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07159460705713429, "300–1000 Hz_40ms_envelope_correlation": 0.9952240568105193, "300–1000 Hz_40ms_modulation_mae_db": 0.2022782997259529, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6329094415512815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15645389374843258, "1000–3000 Hz_10ms_envelope_correlation": 0.9810067414353971, "1000–3000 Hz_10ms_modulation_mae_db": 0.527516089139754, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0111997702792033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08729093080058861, "1000–3000 Hz_40ms_envelope_correlation": 0.9942544961564841, "1000–3000 Hz_40ms_modulation_mae_db": 0.28935176724703093, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7142461210168096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5065558959290731, "3000–6000 Hz_10ms_envelope_correlation": 0.8147939652906854, "3000–6000 Hz_10ms_modulation_mae_db": 2.6118490198204998, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1807193398717424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.43070967999384274, "3000–6000 Hz_40ms_envelope_correlation": 0.8445904949088033, "3000–6000 Hz_40ms_modulation_mae_db": 3.6520779567892294, "mr_stft_db": 5.677741022887903, "band_mae_db": 0.3597815926384849, "rms_gain_error_db": -0.103200781235998, "output_peak": 0.7214207882370958}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1876772567227991, "全帯域_10ms_rms_envelope_nrmse": 0.12010135725593538, "全帯域_10ms_envelope_correlation": 0.9867334763170875, "全帯域_10ms_modulation_mae_db": 0.4395164403231027, "全帯域_40ms_log_rms_mae_db": 0.6354866925424714, "全帯域_40ms_rms_envelope_nrmse": 0.0809871114951399, "全帯域_40ms_envelope_correlation": 0.9936328412282782, "全帯域_40ms_modulation_mae_db": 0.21492996646849727, "300–1000 Hz_10ms_log_rms_mae_db": 1.6543619339367712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13398177987204005, "300–1000 Hz_10ms_envelope_correlation": 0.9847014555350411, "300–1000 Hz_10ms_modulation_mae_db": 0.6423645805564352, "300–1000 Hz_40ms_log_rms_mae_db": 0.932154548794387, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08385862786875757, "300–1000 Hz_40ms_envelope_correlation": 0.9934005814685756, "300–1000 Hz_40ms_modulation_mae_db": 0.36202950442680154, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6753358478411307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18069972191087946, "1000–3000 Hz_10ms_envelope_correlation": 0.977255686139947, "1000–3000 Hz_10ms_modulation_mae_db": 1.4901276090478268, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9743014553846974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12393752771141452, "1000–3000 Hz_40ms_envelope_correlation": 0.9893140822966559, "1000–3000 Hz_40ms_modulation_mae_db": 1.1245368674469975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7349849025542472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.49938657416412163, "3000–6000 Hz_10ms_envelope_correlation": 0.8117364366415196, "3000–6000 Hz_10ms_modulation_mae_db": 2.521925178820827, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2135001729942507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.42565719570172683, "3000–6000 Hz_40ms_envelope_correlation": 0.8405030265158947, "3000–6000 Hz_40ms_modulation_mae_db": 3.54240666974887, "mr_stft_db": 5.690556863008125, "band_mae_db": 0.4120402226391257, "rms_gain_error_db": 0.1597887969162301, "output_peak": 0.9420923094838549}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1598129614629484, "全帯域_10ms_rms_envelope_nrmse": 0.12151805057012449, "全帯域_10ms_envelope_correlation": 0.9856852907771092, "全帯域_10ms_modulation_mae_db": 0.4198365698139667, "全帯域_40ms_log_rms_mae_db": 0.6125799591970921, "全帯域_40ms_rms_envelope_nrmse": 0.07182727612644667, "全帯域_40ms_envelope_correlation": 0.9947193014739117, "全帯域_40ms_modulation_mae_db": 0.11193206061151235, "300–1000 Hz_10ms_log_rms_mae_db": 1.6599975700481928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15466067177253007, "300–1000 Hz_10ms_envelope_correlation": 0.97986238174915, "300–1000 Hz_10ms_modulation_mae_db": 1.2289872230026757, "300–1000 Hz_40ms_log_rms_mae_db": 0.935737218945803, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09464531737189971, "300–1000 Hz_40ms_envelope_correlation": 0.9924556555539708, "300–1000 Hz_40ms_modulation_mae_db": 0.3211299389410032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6297461326932774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15974247834984567, "1000–3000 Hz_10ms_envelope_correlation": 0.9799981390044928, "1000–3000 Hz_10ms_modulation_mae_db": 0.884155275432323, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9685071057513488, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07595224193017218, "1000–3000 Hz_40ms_envelope_correlation": 0.9950717338630782, "1000–3000 Hz_40ms_modulation_mae_db": 0.84049160455281, "3000–6000 Hz_10ms_log_rms_mae_db": 1.732939155862712, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.49807365840046947, "3000–6000 Hz_10ms_envelope_correlation": 0.8152431344153748, "3000–6000 Hz_10ms_modulation_mae_db": 2.5078605690117306, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2074996872771067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.42322417804719503, "3000–6000 Hz_40ms_envelope_correlation": 0.8440300649999758, "3000–6000 Hz_40ms_modulation_mae_db": 3.5660713227614855, "mr_stft_db": 5.687974694406436, "band_mae_db": 0.37607487380023935, "rms_gain_error_db": -0.17088084868324693, "output_peak": 0.7962448790212423}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.162865795754597, "全帯域_10ms_rms_envelope_nrmse": 0.11931763098121531, "全帯域_10ms_envelope_correlation": 0.986097734558697, "全帯域_10ms_modulation_mae_db": 0.9259075829227296, "全帯域_40ms_log_rms_mae_db": 0.6513566010355937, "全帯域_40ms_rms_envelope_nrmse": 0.05777807578671479, "全帯域_40ms_envelope_correlation": 0.9963762556884586, "全帯域_40ms_modulation_mae_db": 0.4821056667363189, "300–1000 Hz_10ms_log_rms_mae_db": 1.6879888742090678, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14930165858236058, "300–1000 Hz_10ms_envelope_correlation": 0.9808442696671913, "300–1000 Hz_10ms_modulation_mae_db": 1.5107040209999507, "300–1000 Hz_40ms_log_rms_mae_db": 0.9888201713758995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07713018372960605, "300–1000 Hz_40ms_envelope_correlation": 0.9943498805159787, "300–1000 Hz_40ms_modulation_mae_db": 0.9831188229335406, "1000–3000 Hz_10ms_log_rms_mae_db": 1.645817248171533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1529715838000075, "1000–3000 Hz_10ms_envelope_correlation": 0.9815069181517065, "1000–3000 Hz_10ms_modulation_mae_db": 0.9479882744573634, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9901587189588857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0659765161361593, "1000–3000 Hz_40ms_envelope_correlation": 0.9963083689338185, "1000–3000 Hz_40ms_modulation_mae_db": 0.6820014984407324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.716102643416722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4919585654668093, "3000–6000 Hz_10ms_envelope_correlation": 0.8235526883310268, "3000–6000 Hz_10ms_modulation_mae_db": 2.7058764481481865, "3000–6000 Hz_40ms_log_rms_mae_db": 1.182053574876682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4175309321436024, "3000–6000 Hz_40ms_envelope_correlation": 0.8532451231066367, "3000–6000 Hz_40ms_modulation_mae_db": 3.6982324575218697, "mr_stft_db": 5.685889075128762, "band_mae_db": 0.4017922779772996, "rms_gain_error_db": -0.03765233827155158, "output_peak": 0.9373237162277412}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6870832184014715, "spectral_convergence": 0.3808028019580035, "spectral_convergence_status": "ok", "amplitude_floor": 2.051312383514402e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.470479630308516, "spectral_convergence": 0.3878379560042066, "spectral_convergence_status": "ok", "amplitude_floor": 1.425885904641783e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.884545097746838, "spectral_convergence": 0.3799393267584744, "spectral_convergence_status": "ok", "amplitude_floor": 8.483800566678623e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001304494586920238, "output_power": 0.0001403333704453993, "delta_db": 0.3171867465943356, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014301207186140645, "output_power": 0.0014555038338459486, "delta_db": 0.07640655267899978, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.004308309383381418, "output_power": 0.004251286769819008, "delta_db": -0.05786481589031199, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.001304992137209625, "output_power": 0.0013112956288731982, "delta_db": 0.020927185295053927, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002442775973124686, "output_power": 0.00026487042428233204, "delta_db": 0.3514982754135884, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0021123865382114e-05, "output_power": 1.3758930337176271e-05, "delta_db": 1.37668241553113, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.36676099856723665, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.005187815464959623, "pitch": null, "output_rms_dbfs": -21.285993595711588, "output_peak": 0.7867718701071794, "reference_rms_dbfs": -21.29118141117655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.680702648818941}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002631236493882886, "log_rms_mae_db": 1.106906753904374, "rms_envelope_nrmse": 0.11107420834874333, "envelope_correlation": 0.9880481922782575, "rms_gain_error_db": 0.005187779044791533, "reference_mean_power": 0.007434364903539191, "output_mean_power": 0.007443250783620202}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.450307530079474e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015621941098031631, "output_power": 0.0016267920368701597, "reference_active": true, "delta_db": 0.17597041889877565}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031400948871223737, "output_power": 0.0002850286251487146, "reference_active": true, "delta_db": -0.4205429377259883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9119711962951405e-05, "output_power": 7.867904527499275e-05, "reference_active": true, "delta_db": 2.046032699192021}], "active_band_count": 3, "active_band_mae_db": 0.8808486852722616}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024033810696394756, "log_rms_mae_db": 0.5764795469191394, "rms_envelope_nrmse": 0.06425034682823973, "envelope_correlation": 0.9955733726418278, "rms_gain_error_db": 0.00518779873792652, "reference_mean_power": 0.007471745584669175, "output_mean_power": 0.007480676177706419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.136699874330791e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014162642987993615, "output_power": 0.0014971487005039604, "reference_active": true, "delta_db": 0.24120630094667278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001495506387635467, "output_power": 0.00013822439060036428, "reference_active": true, "delta_db": -0.3420358842795896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0546566248780076e-06, "output_power": 4.483000597304156e-06, "reference_active": true, "delta_db": 1.6660639861704885}], "active_band_count": 3, "active_band_mae_db": 0.7497687237989169}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002303175773961387, "log_rms_mae_db": 1.604759113455014, "rms_envelope_nrmse": 0.1484148874572268, "envelope_correlation": 0.9810793143227198, "rms_gain_error_db": -0.06919154096719224, "reference_mean_power": 0.004944923373977616, "output_mean_power": 0.004866765403441093}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6955719701225893e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012875788279854436, "output_power": 0.0013225831161767145, "reference_active": true, "delta_db": 0.11649147441482417}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026650258378785487, "output_power": 0.00024377113001832783, "reference_active": true, "delta_db": -0.3871915340659893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0302791253002337e-05, "output_power": 7.061219750667488e-05, "reference_active": true, "delta_db": 3.6739709317715166}], "active_band_count": 3, "active_band_mae_db": 1.3925513134174434}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020459712012168156, "log_rms_mae_db": 0.9370938959903897, "rms_envelope_nrmse": 0.08876412391869522, "envelope_correlation": 0.9925343070899835, "rms_gain_error_db": -0.0691913930319229, "reference_mean_power": 0.004969790940376444, "output_mean_power": 0.004891240087182052}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.46585734788702e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011840763589577269, "output_power": 0.001238234355761529, "reference_active": true, "delta_db": 0.19423139516301235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012134699925967192, "output_power": 0.00011759092454983661, "reference_active": true, "delta_db": -0.13655236321808195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.482336680977429e-06, "output_power": 4.5481071634052675e-06, "reference_active": true, "delta_db": 2.6297000426169594}], "active_band_count": 3, "active_band_mae_db": 0.9868279336660178}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014083569934759876, "log_rms_mae_db": 1.6000576899238868, "rms_envelope_nrmse": 0.17721226964259046, "envelope_correlation": 0.9756551729287021, "rms_gain_error_db": 0.059005131003866104, "reference_mean_power": 0.0013181968247743324, "output_mean_power": 0.0013362286349013137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.706116677050846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003562978060875054, "output_power": 0.0003540839194091404, "reference_active": true, "delta_db": -0.027069442983745574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927249258823322e-05, "output_power": 7.303128398582621e-05, "reference_active": true, "delta_db": 0.22948121314262884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3444445693793863e-05, "output_power": 1.729167630259755e-05, "reference_active": true, "delta_db": 1.092941957800506}], "active_band_count": 3, "active_band_mae_db": 0.4498308713089601}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010879087849669345, "log_rms_mae_db": 0.9815582476445694, "rms_envelope_nrmse": 0.10351335279246981, "envelope_correlation": 0.9910120856880005, "rms_gain_error_db": 0.059004842851633085, "reference_mean_power": 0.0013248216670349005, "output_mean_power": 0.0013429440102593255}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.107952744901008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003355694425017771, "output_power": 0.00033936096442774917, "reference_active": true, "delta_db": 0.04879478996346824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.409137696567018e-05, "output_power": 3.679606915420288e-05, "reference_active": true, "delta_db": 0.33156883467990006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.097157965658087e-07, "output_power": 1.0548678371308062e-06, "reference_active": true, "delta_db": 1.1486543899955513}], "active_band_count": 3, "active_band_mae_db": 0.5096726715463066}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.557718084358035e-05, "log_rms_mae_db": 1.7619560494179214, "rms_envelope_nrmse": 0.5129244612111703, "envelope_correlation": 0.8194833317124034, "rms_gain_error_db": 1.1003330568347958, "reference_mean_power": 2.3934708718272692e-05, "output_mean_power": 3.083624249139695e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3699291941082093e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.018836822609012e-06, "output_power": 6.028070950890329e-06, "reference_active": true, "delta_db": 1.7607798252369977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.745214583455981e-06, "output_power": 6.967692332995036e-06, "reference_active": true, "delta_db": 1.6683310916924938}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3999346848964708e-06, "output_power": 5.230971029502818e-06, "reference_active": true, "delta_db": 5.724745410118739}], "active_band_count": 3, "active_band_mae_db": 3.0512854423494105}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.6234588308267947e-05, "log_rms_mae_db": 1.2452333473232224, "rms_envelope_nrmse": 0.4392656416517366, "envelope_correlation": 0.8485478457292429, "rms_gain_error_db": 1.1003644566331934, "reference_mean_power": 2.405439007374668e-05, "output_mean_power": 3.0990657832119404e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1165079820636279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.688364032412796e-06, "output_power": 7.860004169824084e-06, "reference_active": true, "delta_db": 3.2858899814892295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313243490564173e-06, "output_power": 5.670369865400707e-06, "reference_active": true, "delta_db": 2.3335803446798273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.200263616160695e-07, "output_power": 5.38487905591391e-07, "reference_active": true, "delta_db": 6.51899312073755}], "active_band_count": 3, "active_band_mae_db": 4.046154482302202}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8625814007814653, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0ac11677ff7941f9042e962025288535efa2647a30eb1e29837d59d80e3311ba", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9794331234251494, "standard_deviation": 0.0022456831886403057, "minimum": 0.9756551729287021, "maximum": 0.9821805450233745, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6385171975973982, "standard_deviation": 0.026232838897250567, "minimum": 1.6000576899238868, "maximum": 1.6754655729854084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7085882781100059, "standard_deviation": 0.41670012689490354, "minimum": 0.20248695847389678, "maximum": 1.4901276090478268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16306437449389416, "standard_deviation": 0.010932532669299088, "minimum": 0.15077277630100536, "maximum": 0.18069972191087946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934386617315325, "standard_deviation": 0.0024117371753300592, "minimum": 0.9893140822966559, "maximum": 0.9963083689338185, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.984032612463079, "standard_deviation": 0.022315387937350765, "minimum": 0.9499218029955812, "maximum": 1.0182735592367183, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5303942380438, "standard_deviation": 0.3440513395666783, "minimum": 0.12123676493220774, "maximum": 1.1245368674469975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08926452771509889, "standard_deviation": 0.018491045023530207, "minimum": 0.0659765161361593, "maximum": 0.12393752771141452, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8199479298875221, "standard_deviation": 0.008324450195881407, "minimum": 0.8117364366415196, "maximum": 0.8381896860836384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7223021327905559, "standard_deviation": 0.03296374627673015, "minimum": 1.6493470664159129, "maximum": 1.7619560494179214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.793559155598788, "standard_deviation": 0.25134704137714603, "minimum": 2.5078605690117306, "maximum": 3.178084275813245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5012017122736387, "standard_deviation": 0.007603997140364062, "minimum": 0.4919585654668093, "maximum": 0.5129244612111703, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8489275273257295, "standard_deviation": 0.00794628065788706, "minimum": 0.8405030265158947, "maximum": 0.8653891424528591, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1950903169308469, "standard_deviation": 0.027767200653406498, "minimum": 1.1590239873809958, "maximum": 1.2452333473232224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.806585261976455, "standard_deviation": 0.23444115532470103, "minimum": 3.54240666974887, "maximum": 4.185893903764045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.4270692089247119, "standard_deviation": 0.007605910960363809, "minimum": 0.4175309321436024, "maximum": 0.4392656416517366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.982284965324183, "standard_deviation": 0.0030264537423881715, "minimum": 0.9778186602349401, "maximum": 0.9874617915376501, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6371836673510856, "standard_deviation": 0.05439021336960414, "minimum": 1.5530559218818898, "maximum": 1.71790728498324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.012478160715931, "standard_deviation": 0.3901995290031149, "minimum": 0.34253446240752333, "maximum": 1.5107040209999507, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14384341991924654, "standard_deviation": 0.013031364570587378, "minimum": 0.12089620631086709, "maximum": 0.16288743408208972, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9929689662341474, "standard_deviation": 0.0022212818927199317, "minimum": 0.9879727928596106, "maximum": 0.9952240568105193, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9544701296052562, "standard_deviation": 0.049360000506657495, "minimum": 0.891995599358277, "maximum": 1.040510131162445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.503342103884828, "standard_deviation": 0.3051224442396542, "minimum": 0.2022782997259529, "maximum": 0.9868279336660178, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08703458467057307, "standard_deviation": 0.013066061998376231, "minimum": 0.07159460705713429, "maximum": 0.11448177299424502, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.41115096116577354, "standard_deviation": 0.04404689347612718, "minimum": 0.3597815926384849, "maximum": 0.482429291857646, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6835547079897, "standard_deviation": 0.005618441014297248, "minimum": 5.6751100891768855, "maximum": 5.690556863008125, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.841021165897245, "standard_deviation": 0.0907092231706329, "minimum": 0.7214207882370958, "maximum": 0.9420923094838549, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03609724497917413, "standard_deviation": 0.12335943850818919, "minimum": -0.1859303721658797, "maximum": 0.1597887969162301, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9871067331555644, "standard_deviation": 0.0018567282767893267, "minimum": 0.9847837266758479, "maximum": 0.9908889439904598, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1565159668963307, "standard_deviation": 0.03792328579179383, "minimum": 1.0897890686656517, "maximum": 1.1940689502375987, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5307926723098133, "standard_deviation": 0.2533173888397347, "minimum": 0.17800633568559152, "maximum": 0.9259075829227296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11556186576061425, "standard_deviation": 0.00911518743910343, "minimum": 0.09685047621621624, "maximum": 0.1269212835242522, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9953627009561652, "standard_deviation": 0.001385562036456891, "minimum": 0.9931885314788782, "maximum": 0.9970639973926778, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6232789776894059, "standard_deviation": 0.022299874404695375, "minimum": 0.5764795469191394, "maximum": 0.6513566010355937, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.29954751544953173, "standard_deviation": 0.21245614190463513, "minimum": 0.11193206061151235, "maximum": 0.7497687237989169, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06640177703122205, "standard_deviation": 0.01064107123843912, "minimum": 0.05298753235179565, "maximum": 0.08133259262476952, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0031", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.055853843688965, "predicted_mos": 2.055853843688965, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7dc3a66061bb62cd0b501018654279f240b68694dfd1c0bf67a176990433ec4e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.306308799888939, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1313897823318908, "全帯域_10ms_rms_envelope_nrmse": 0.11776880135201662, "全帯域_10ms_envelope_correlation": 0.9864986333879798, "全帯域_10ms_modulation_mae_db": 0.7497756632099036, "全帯域_40ms_log_rms_mae_db": 0.5924276107462031, "全帯域_40ms_rms_envelope_nrmse": 0.06374526648373703, "全帯域_40ms_envelope_correlation": 0.9955897757671035, "全帯域_40ms_modulation_mae_db": 0.7333675940637198, "300–1000 Hz_10ms_log_rms_mae_db": 1.7814965345848413, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14122056601549252, "300–1000 Hz_10ms_envelope_correlation": 0.9828644597330676, "300–1000 Hz_10ms_modulation_mae_db": 1.3563116542161813, "300–1000 Hz_40ms_log_rms_mae_db": 0.960305495597917, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07580272761727984, "300–1000 Hz_40ms_envelope_correlation": 0.9945135844152833, "300–1000 Hz_40ms_modulation_mae_db": 0.9286520691064549, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7620143299242936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1524754122580923, "1000–3000 Hz_10ms_envelope_correlation": 0.9816877151208535, "1000–3000 Hz_10ms_modulation_mae_db": 0.6012823360063929, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0367554480336596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08955719396103858, "1000–3000 Hz_40ms_envelope_correlation": 0.9934238453043457, "1000–3000 Hz_40ms_modulation_mae_db": 0.655038933067977, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9773904613386608, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5261753175118734, "3000–6000 Hz_10ms_envelope_correlation": 0.8041960148171042, "3000–6000 Hz_10ms_modulation_mae_db": 2.9135737465959792, "3000–6000 Hz_40ms_log_rms_mae_db": 1.338329167862102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.44527163023144894, "3000–6000 Hz_40ms_envelope_correlation": 0.8390019791882372, "3000–6000 Hz_40ms_modulation_mae_db": 3.881435213350342, "mr_stft_db": 5.84513427489687, "band_mae_db": 0.3791083989663279, "rms_gain_error_db": -0.007833154437560252, "output_peak": 0.8111427281255559}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.120824431810155, "全帯域_10ms_rms_envelope_nrmse": 0.10921411731810408, "全帯域_10ms_envelope_correlation": 0.9883952844052296, "全帯域_10ms_modulation_mae_db": 0.3546193999732476, "全帯域_40ms_log_rms_mae_db": 0.5781642449897497, "全帯域_40ms_rms_envelope_nrmse": 0.0522239673851974, "全帯域_40ms_envelope_correlation": 0.9970420852536165, "全帯域_40ms_modulation_mae_db": 0.21751326410476315, "300–1000 Hz_10ms_log_rms_mae_db": 1.705529676870187, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1361575892113207, "300–1000 Hz_10ms_envelope_correlation": 0.9841155875773354, "300–1000 Hz_10ms_modulation_mae_db": 0.8865441779077322, "300–1000 Hz_40ms_log_rms_mae_db": 0.9113934403942069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06803291971003238, "300–1000 Hz_40ms_envelope_correlation": 0.9955899655016351, "300–1000 Hz_40ms_modulation_mae_db": 0.24650600139411616, "1000–3000 Hz_10ms_log_rms_mae_db": 1.797396009317311, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14317854098346183, "1000–3000 Hz_10ms_envelope_correlation": 0.9838312312960186, "1000–3000 Hz_10ms_modulation_mae_db": 0.3280303901037777, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1014315332445865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08357101532104645, "1000–3000 Hz_40ms_envelope_correlation": 0.9941761307778663, "1000–3000 Hz_40ms_modulation_mae_db": 0.1762677104743763, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9794865482221033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5141237039104766, "3000–6000 Hz_10ms_envelope_correlation": 0.822509602796412, "3000–6000 Hz_10ms_modulation_mae_db": 2.808296751071906, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3178211371818593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4308955278167619, "3000–6000 Hz_40ms_envelope_correlation": 0.859623524588346, "3000–6000 Hz_40ms_modulation_mae_db": 3.792084583897953, "mr_stft_db": 5.850670554116398, "band_mae_db": 0.40219758981289977, "rms_gain_error_db": 0.006028110172693544, "output_peak": 0.7416205392218569}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1502344744679365, "全帯域_10ms_rms_envelope_nrmse": 0.11907578528342001, "全帯域_10ms_envelope_correlation": 0.9861433295186017, "全帯域_10ms_modulation_mae_db": 1.3261378791300729, "全帯域_40ms_log_rms_mae_db": 0.603943481115247, "全帯域_40ms_rms_envelope_nrmse": 0.04817359045575635, "全帯域_40ms_envelope_correlation": 0.9975148001494958, "全帯域_40ms_modulation_mae_db": 0.9696107270352087, "300–1000 Hz_10ms_log_rms_mae_db": 1.7134245783900224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15137957642270172, "300–1000 Hz_10ms_envelope_correlation": 0.9803465185123224, "300–1000 Hz_10ms_modulation_mae_db": 2.038066988714148, "300–1000 Hz_40ms_log_rms_mae_db": 0.9463664828541831, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06840722725126736, "300–1000 Hz_40ms_envelope_correlation": 0.9955518747723658, "300–1000 Hz_40ms_modulation_mae_db": 1.4524721250756845, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7998913597365864, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15973277970374813, "1000–3000 Hz_10ms_envelope_correlation": 0.9804251155364312, "1000–3000 Hz_10ms_modulation_mae_db": 0.5928819797115902, "1000–3000 Hz_40ms_log_rms_mae_db": 1.097732927406117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08984932425454757, "1000–3000 Hz_40ms_envelope_correlation": 0.994291360799728, "1000–3000 Hz_40ms_modulation_mae_db": 0.7610174177186421, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0173380935984837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5275790085809554, "3000–6000 Hz_10ms_envelope_correlation": 0.8000524755962108, "3000–6000 Hz_10ms_modulation_mae_db": 2.6938667795981197, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3656336609548834, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4416205519735676, "3000–6000 Hz_40ms_envelope_correlation": 0.8386253360637381, "3000–6000 Hz_40ms_modulation_mae_db": 3.759142185326246, "mr_stft_db": 5.853785106943138, "band_mae_db": 0.40996318215886984, "rms_gain_error_db": -0.06091101988415851, "output_peak": 0.7197392043043501}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1782981820207838, "全帯域_10ms_rms_envelope_nrmse": 0.11771927572470721, "全帯域_10ms_envelope_correlation": 0.9865999638255626, "全帯域_10ms_modulation_mae_db": 0.75335724875402, "全帯域_40ms_log_rms_mae_db": 0.6622153703570741, "全帯域_40ms_rms_envelope_nrmse": 0.07980180628932634, "全帯域_40ms_envelope_correlation": 0.9933677807022901, "全帯域_40ms_modulation_mae_db": 0.8435699774086954, "300–1000 Hz_10ms_log_rms_mae_db": 1.7511189714444537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15013974859702867, "300–1000 Hz_10ms_envelope_correlation": 0.9809391045962864, "300–1000 Hz_10ms_modulation_mae_db": 0.4944391271142139, "300–1000 Hz_40ms_log_rms_mae_db": 1.0348016667663682, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10662866888017852, "300–1000 Hz_40ms_envelope_correlation": 0.9898379365659583, "300–1000 Hz_40ms_modulation_mae_db": 0.47570057768942325, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8312996849885381, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16090012706640172, "1000–3000 Hz_10ms_envelope_correlation": 0.9796287558307988, "1000–3000 Hz_10ms_modulation_mae_db": 0.8570474561269007, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0778167245892398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1020512886815963, "1000–3000 Hz_40ms_envelope_correlation": 0.991058222444518, "1000–3000 Hz_40ms_modulation_mae_db": 0.664780543864058, "3000–6000 Hz_10ms_log_rms_mae_db": 2.041151975505996, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5130666451731369, "3000–6000 Hz_10ms_envelope_correlation": 0.8086165093544762, "3000–6000 Hz_10ms_modulation_mae_db": 2.637036161144613, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4175793655821187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4306754730375352, "3000–6000 Hz_40ms_envelope_correlation": 0.8442845234727693, "3000–6000 Hz_40ms_modulation_mae_db": 3.5944535741084973, "mr_stft_db": 5.847428031520773, "band_mae_db": 0.4956090738969207, "rms_gain_error_db": -0.17073892574966887, "output_peak": 0.6096750336218861}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1296797520707487, "全帯域_10ms_rms_envelope_nrmse": 0.10892494231211045, "全帯域_10ms_envelope_correlation": 0.9884503537858726, "全帯域_10ms_modulation_mae_db": 0.176237814588451, "全帯域_40ms_log_rms_mae_db": 0.6399068174088881, "全帯域_40ms_rms_envelope_nrmse": 0.06008089239733516, "全帯域_40ms_envelope_correlation": 0.9960850080078162, "全帯域_40ms_modulation_mae_db": 0.06312946501917176, "300–1000 Hz_10ms_log_rms_mae_db": 1.871243920470146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13913535688724868, "300–1000 Hz_10ms_envelope_correlation": 0.9833501583151957, "300–1000 Hz_10ms_modulation_mae_db": 0.5956087646899841, "300–1000 Hz_40ms_log_rms_mae_db": 1.0896389144285237, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07820588592754929, "300–1000 Hz_40ms_envelope_correlation": 0.9941952489732715, "300–1000 Hz_40ms_modulation_mae_db": 0.15914472884439942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8107891289398055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13965563450274973, "1000–3000 Hz_10ms_envelope_correlation": 0.9849087843980364, "1000–3000 Hz_10ms_modulation_mae_db": 0.0896602168707159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0574692439834272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08846199563201194, "1000–3000 Hz_40ms_envelope_correlation": 0.9934533699133746, "1000–3000 Hz_40ms_modulation_mae_db": 0.09793839391960063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9716015807150766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5184431196831687, "3000–6000 Hz_10ms_envelope_correlation": 0.8080635407067948, "3000–6000 Hz_10ms_modulation_mae_db": 2.843290873698987, "3000–6000 Hz_40ms_log_rms_mae_db": 1.361068448720073, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4380390558716621, "3000–6000 Hz_40ms_envelope_correlation": 0.8422607497582123, "3000–6000 Hz_40ms_modulation_mae_db": 3.893101251901676, "mr_stft_db": 5.851867593227475, "band_mae_db": 0.3764461866556526, "rms_gain_error_db": -0.010470625872340612, "output_peak": 0.7809696989144199}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1360748105198122, "全帯域_10ms_rms_envelope_nrmse": 0.11083551989396796, "全帯域_10ms_envelope_correlation": 0.9881176480984835, "全帯域_10ms_modulation_mae_db": 0.43564895718610125, "全帯域_40ms_log_rms_mae_db": 0.6345476974214591, "全帯域_40ms_rms_envelope_nrmse": 0.05418049001933382, "全帯域_40ms_envelope_correlation": 0.997110106240923, "全帯域_40ms_modulation_mae_db": 0.3442172922177105, "300–1000 Hz_10ms_log_rms_mae_db": 1.7902357783368612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13675179673384588, "300–1000 Hz_10ms_envelope_correlation": 0.9842235776321954, "300–1000 Hz_10ms_modulation_mae_db": 0.9476542575443281, "300–1000 Hz_40ms_log_rms_mae_db": 1.0107059239808376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07215010334246981, "300–1000 Hz_40ms_envelope_correlation": 0.99573311821857, "300–1000 Hz_40ms_modulation_mae_db": 0.4794462001936129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7766875580624237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1399699149087828, "1000–3000 Hz_10ms_envelope_correlation": 0.9845315016353899, "1000–3000 Hz_10ms_modulation_mae_db": 0.4538352339669975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.052981595925027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07523318941424952, "1000–3000 Hz_40ms_envelope_correlation": 0.9952624056728967, "1000–3000 Hz_40ms_modulation_mae_db": 0.49781050609724004, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0589720641822544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5187406546033032, "3000–6000 Hz_10ms_envelope_correlation": 0.8137717180996472, "3000–6000 Hz_10ms_modulation_mae_db": 2.714439635294427, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4382286303267269, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4358752771914411, "3000–6000 Hz_40ms_envelope_correlation": 0.8493626177304775, "3000–6000 Hz_40ms_modulation_mae_db": 3.7371151126896596, "mr_stft_db": 5.859848610253337, "band_mae_db": 0.4690300673045314, "rms_gain_error_db": -0.14839328553446673, "output_peak": 0.6728569505164159}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1374884676380648, "全帯域_10ms_rms_envelope_nrmse": 0.11652615203521645, "全帯域_10ms_envelope_correlation": 0.9868002650575765, "全帯域_10ms_modulation_mae_db": 0.9061774615590364, "全帯域_40ms_log_rms_mae_db": 0.5950817756865686, "全帯域_40ms_rms_envelope_nrmse": 0.05990325667582808, "全帯域_40ms_envelope_correlation": 0.9963467110421577, "全帯域_40ms_modulation_mae_db": 0.31133984310531404, "300–1000 Hz_10ms_log_rms_mae_db": 1.7936937582325383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1469470561678577, "300–1000 Hz_10ms_envelope_correlation": 0.98160123456017, "300–1000 Hz_10ms_modulation_mae_db": 1.1903821558860201, "300–1000 Hz_40ms_log_rms_mae_db": 1.055168645393391, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08831662637543279, "300–1000 Hz_40ms_envelope_correlation": 0.9930846615970582, "300–1000 Hz_40ms_modulation_mae_db": 0.33248474412858803, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7532365797026328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15385889405335076, "1000–3000 Hz_10ms_envelope_correlation": 0.9812532038715542, "1000–3000 Hz_10ms_modulation_mae_db": 1.2594898401794963, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0456984836442114, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08726244400268646, "1000–3000 Hz_40ms_envelope_correlation": 0.9934894568200175, "1000–3000 Hz_40ms_modulation_mae_db": 0.8833119467604877, "3000–6000 Hz_10ms_log_rms_mae_db": 2.070289755908095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5118582713169593, "3000–6000 Hz_10ms_envelope_correlation": 0.8121853862216444, "3000–6000 Hz_10ms_modulation_mae_db": 2.63661489392897, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4126892701102018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4277252221166665, "3000–6000 Hz_40ms_envelope_correlation": 0.849531450276251, "3000–6000 Hz_40ms_modulation_mae_db": 3.542348228850749, "mr_stft_db": 5.8582895299767905, "band_mae_db": 0.44688888802370785, "rms_gain_error_db": -0.1420776159623136, "output_peak": 0.6963573077598089}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1400759177691977, "全帯域_10ms_rms_envelope_nrmse": 0.12619077479178725, "全帯域_10ms_envelope_correlation": 0.9848246326235689, "全帯域_10ms_modulation_mae_db": 0.8051203625889075, "全帯域_40ms_log_rms_mae_db": 0.5995666454476625, "全帯域_40ms_rms_envelope_nrmse": 0.07285540163546247, "全帯域_40ms_envelope_correlation": 0.9944754083872613, "全帯域_40ms_modulation_mae_db": 0.2728238084020777, "300–1000 Hz_10ms_log_rms_mae_db": 1.7720671238500523, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15793027570571283, "300–1000 Hz_10ms_envelope_correlation": 0.9788316143425645, "300–1000 Hz_10ms_modulation_mae_db": 1.2760584832442907, "300–1000 Hz_40ms_log_rms_mae_db": 0.9892481488720866, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08834910082601356, "300–1000 Hz_40ms_envelope_correlation": 0.9926676824485442, "300–1000 Hz_40ms_modulation_mae_db": 0.5117337974752997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7784386218474737, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15465265055495392, "1000–3000 Hz_10ms_envelope_correlation": 0.9811802560404037, "1000–3000 Hz_10ms_modulation_mae_db": 0.733126615823576, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0361031592086873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09155991701780836, "1000–3000 Hz_40ms_envelope_correlation": 0.9928091220831963, "1000–3000 Hz_40ms_modulation_mae_db": 0.6578740375074742, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0443046302939902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5339388771286043, "3000–6000 Hz_10ms_envelope_correlation": 0.8012663759615375, "3000–6000 Hz_10ms_modulation_mae_db": 2.6923724893769894, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3808445137328222, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.44808914459725474, "3000–6000 Hz_40ms_envelope_correlation": 0.8400617154433304, "3000–6000 Hz_40ms_modulation_mae_db": 3.6730580994784567, "mr_stft_db": 5.841444333757629, "band_mae_db": 0.408433372328947, "rms_gain_error_db": 0.08614613549500874, "output_peak": 0.7818962295094707}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.937058720009307, "spectral_convergence": 0.3708652426229753, "spectral_convergence_status": "ok", "amplitude_floor": 2.051312383514402e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.653736079730178, "spectral_convergence": 0.4037261742302693, "spectral_convergence_status": "ok", "amplitude_floor": 1.425885904641783e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.944608024951125, "spectral_convergence": 0.39212761224799975, "spectral_convergence_status": "ok", "amplitude_floor": 8.483800566678623e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001304494586920238, "output_power": 0.00015236226974333348, "delta_db": 0.6743515242084237, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014301207186140645, "output_power": 0.001455037181684023, "delta_db": 0.07501392869914414, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.004308309383381418, "output_power": 0.004280087582771394, "delta_db": -0.02854226798919223, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.001304992137209625, "output_power": 0.0012685486189892624, "delta_db": -0.12300778136420995, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002442775973124686, "output_power": 0.0002450423328286729, "delta_db": 0.013574786749857303, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0021123865382114e-05, "output_power": 1.370668526989912e-05, "delta_db": 1.3601601047871401, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3791083989663279, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.007833154437560252, "pitch": null, "output_rms_dbfs": -21.299014565614108, "output_peak": 0.8111427281255559, "reference_rms_dbfs": -21.29118141117655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.84513427489687}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002631236493882886, "log_rms_mae_db": 1.1313897823318908, "rms_envelope_nrmse": 0.11776880135201662, "envelope_correlation": 0.9864986333879798, "rms_gain_error_db": -0.007833012831770014, "reference_mean_power": 0.007434364903539191, "output_mean_power": 0.007420968235103398}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.450307530079474e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015621941098031631, "output_power": 0.0015828051785060848, "reference_active": true, "delta_db": 0.05692466508226923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031400948871223737, "output_power": 0.00034295580959734125, "reference_active": true, "delta_db": 0.38295392366026165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9119711962951405e-05, "output_power": 7.450761324092718e-05, "reference_active": true, "delta_db": 1.8094484008871803}], "active_band_count": 3, "active_band_mae_db": 0.7497756632099036}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024033810696394756, "log_rms_mae_db": 0.5924276107462031, "rms_envelope_nrmse": 0.06374526648373703, "envelope_correlation": 0.9955897757671035, "rms_gain_error_db": -0.00783356584601937, "reference_mean_power": 0.007471745584669175, "output_mean_power": 0.007458280606823747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.136699874330791e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014162642987993615, "output_power": 0.001464282812829073, "reference_active": true, "delta_db": 0.14480657239237432}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001495506387635467, "output_power": 0.00016247942223198477, "reference_active": true, "delta_db": 0.36010093733677134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0546566248780076e-06, "output_power": 4.5131724184938074e-06, "reference_active": true, "delta_db": 1.6951952724620134}], "active_band_count": 3, "active_band_mae_db": 0.7333675940637198}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002303175773961387, "log_rms_mae_db": 1.7814965345848413, "rms_envelope_nrmse": 0.14122056601549252, "envelope_correlation": 0.9828644597330676, "rms_gain_error_db": -0.039076439406911564, "reference_mean_power": 0.004944923373977616, "output_mean_power": 0.004900630089832456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6955719701225893e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012875788279854436, "output_power": 0.0012749476280054055, "reference_active": true, "delta_db": -0.04281481556858674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026650258378785487, "output_power": 0.00029688183264537857, "reference_active": true, "delta_db": 0.46882198297972155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0302791253002337e-05, "output_power": 6.874046260877573e-05, "reference_active": true, "delta_db": 3.5572981641002355}], "active_band_count": 3, "active_band_mae_db": 1.3563116542161813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020459712012168156, "log_rms_mae_db": 0.960305495597917, "rms_envelope_nrmse": 0.07580272761727984, "envelope_correlation": 0.9945135844152833, "rms_gain_error_db": -0.03907636585500365, "reference_mean_power": 0.004969790940376444, "output_mean_power": 0.00492527499277944}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.46585734788702e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011840763589577269, "output_power": 0.0012044250729040615, "reference_active": true, "delta_db": 0.0740007762159747}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012134699925967192, "output_power": 0.00013008046994050025, "reference_active": true, "delta_db": 0.3018305591438055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.482336680977429e-06, "output_power": 4.323875414871525e-06, "reference_active": true, "delta_db": 2.4101248719595847}], "active_band_count": 3, "active_band_mae_db": 0.9286520691064549}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014083569934759876, "log_rms_mae_db": 1.7620143299242936, "rms_envelope_nrmse": 0.1524754122580923, "envelope_correlation": 0.9816877151208535, "rms_gain_error_db": -0.1917605892869239, "reference_mean_power": 0.0013181968247743324, "output_mean_power": 0.0012612587792861485}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.706116677050846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003562978060875054, "output_power": 0.0003340307212499538, "reference_active": true, "delta_db": -0.28026736890390713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927249258823322e-05, "output_power": 7.050835381592209e-05, "reference_active": true, "delta_db": 0.0767976030350377}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3444445693793863e-05, "output_power": 1.8759493346484065e-05, "reference_active": true, "delta_db": 1.4467820360802341}], "active_band_count": 3, "active_band_mae_db": 0.6012823360063929}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010879087849669345, "log_rms_mae_db": 1.0367554480336596, "rms_envelope_nrmse": 0.08955719396103858, "envelope_correlation": 0.9934238453043457, "rms_gain_error_db": -0.19176340611954054, "reference_mean_power": 0.0013248216670349005, "output_mean_power": 0.0012675966466916798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.107952744901008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003355694425017771, "output_power": 0.00031848191857043593, "reference_active": true, "delta_db": -0.2269762557450044}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.409137696567018e-05, "output_power": 3.646826182158942e-05, "reference_active": true, "delta_db": 0.29270521200862804}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.097157965658087e-07, "output_power": 1.1294737580223867e-06, "reference_active": true, "delta_db": 1.4454353314502986}], "active_band_count": 3, "active_band_mae_db": 0.655038933067977}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.557718084358035e-05, "log_rms_mae_db": 1.9773904613386608, "rms_envelope_nrmse": 0.5261753175118734, "envelope_correlation": 0.8041960148171042, "rms_gain_error_db": 1.090161797709781, "reference_mean_power": 2.3934708718272692e-05, "output_mean_power": 3.076410792993629e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3699291941082093e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.018836822609012e-06, "output_power": 5.8785622719100894e-06, "reference_active": true, "delta_db": 1.65170750231048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.745214583455981e-06, "output_power": 7.010408069350605e-06, "reference_active": true, "delta_db": 1.6948744213340559}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3999346848964708e-06, "output_power": 4.8475433932756134e-06, "reference_active": true, "delta_db": 5.394139316143402}], "active_band_count": 3, "active_band_mae_db": 2.9135737465959792}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.6234588308267947e-05, "log_rms_mae_db": 1.338329167862102, "rms_envelope_nrmse": 0.44527163023144894, "envelope_correlation": 0.8390019791882372, "rms_gain_error_db": 1.0901616377341503, "reference_mean_power": 2.405439007374668e-05, "output_mean_power": 3.091793737158313e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1165079820636279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.688364032412796e-06, "output_power": 7.587900529847687e-06, "reference_active": true, "delta_db": 3.1328785077788366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313243490564173e-06, "output_power": 5.574156351305109e-06, "reference_active": true, "delta_db": 2.259257928182092}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.200263616160695e-07, "output_power": 5.063998423988623e-07, "reference_active": true, "delta_db": 6.252169204090098}], "active_band_count": 3, "active_band_mae_db": 3.881435213350342}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8530044173442837, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2d85935139db8ed8b643e2e09b130e2a218ca0a09284fbd67de692bca78d2bc3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9821808204661857, "standard_deviation": 0.0019781533012198024, "minimum": 0.9796287558307988, "maximum": 0.9849087843980364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.788719159064883, "standard_deviation": 0.025974990065044962, "minimum": 1.7532365797026328, "maximum": 1.8312996849885381, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6144192585986809, "standard_deviation": 0.35312689586305357, "minimum": 0.0896602168707159, "maximum": 1.2594898401794963, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15055299425394264, "standard_deviation": 0.00851804843181589, "minimum": 0.13965563450274973, "maximum": 0.16090012706640172, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934954892269929, "standard_deviation": 0.0012327084414337013, "minimum": 0.991058222444518, "maximum": 0.9952624056728967, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0632486395043694, "standard_deviation": 0.026035460574975553, "minimum": 1.0361031592086873, "maximum": 1.1014315332445865, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.549254936176232, "standard_deviation": 0.27728906664112174, "minimum": 0.09793839391960063, "maximum": 0.8833119467604877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08844329603562315, "standard_deviation": 0.0075379920748963785, "minimum": 0.07523318941424952, "maximum": 0.1020512886815963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8088327029442284, "standard_deviation": 0.007359841503792366, "minimum": 0.8000524755962108, "maximum": 0.822509602796412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0200668887205824, "standard_deviation": 0.03944683938001626, "minimum": 1.9716015807150766, "maximum": 2.070289755908095, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.742436416338749, "standard_deviation": 0.10124498267150187, "minimum": 2.63661489392897, "maximum": 2.9135737465959792, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5204906997385598, "standard_deviation": 0.007935829789774388, "minimum": 0.5118582713169593, "maximum": 0.5339388771286043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8453439870651702, "standard_deviation": 0.007186117335487404, "minimum": 0.8386253360637381, "maximum": 0.859623524588346, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3790242743088486, "standard_deviation": 0.04145898477482567, "minimum": 1.3178211371818593, "maximum": 1.4382286303267269, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.7340922812004473, "standard_deviation": 0.12591576337566995, "minimum": 3.542348228850749, "maximum": 3.893101251901676, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.43727398535454226, "standard_deviation": 0.007346216874298797, "minimum": 0.4277252221166665, "maximum": 0.44808914459725474, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9820340319086422, "standard_deviation": 0.001928661366317333, "minimum": 0.9788316143425645, "maximum": 0.9842235776321954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7723512927723877, "standard_deviation": 0.052135977648056366, "minimum": 1.705529676870187, "maximum": 1.871243920470146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0981332011646123, "standard_deviation": 0.4891231346255154, "minimum": 0.4944391271142139, "maximum": 2.038066988714148, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14495774571765108, "standard_deviation": 0.007864786173498882, "minimum": 0.1361575892113207, "maximum": 0.15793027570571283, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9938967590615858, "standard_deviation": 0.0020032941594892843, "minimum": 0.9898379365659583, "maximum": 0.99573311821857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9997035897859392, "standard_deviation": 0.05954915965504722, "minimum": 0.9113934403942069, "maximum": 1.0896389144285237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5732675304884474, "standard_deviation": 0.42386894604888586, "minimum": 0.15914472884439942, "maximum": 1.4524721250756845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08073665749127795, "standard_deviation": 0.01311312679335335, "minimum": 0.06803291971003238, "maximum": 0.10662866888017852, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.42345959489348217, "standard_deviation": 0.04287598323475704, "minimum": 0.3764461866556526, "maximum": 0.4956090738969207, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.851058504336551, "standard_deviation": 0.0063076177194132665, "minimum": 5.841444333757629, "maximum": 5.859848610253337, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7267822114967206, "standard_deviation": 0.0664487075108418, "minimum": 0.6096750336218861, "maximum": 0.8111427281255559, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05603129772160079, "standard_deviation": 0.0906964999729271, "minimum": -0.17073892574966887, "maximum": 0.08614613549500874, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9869787638378593, "standard_deviation": 0.0012656603671561005, "minimum": 0.9848246326235689, "maximum": 0.9884503537858726, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1405082273285738, "standard_deviation": 0.017482625274323066, "minimum": 1.120824431810155, "maximum": 1.1782981820207838, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6883843483737175, "standard_deviation": 0.3612180515357395, "minimum": 0.176237814588451, "maximum": 1.3261378791300729, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11578192108891626, "standard_deviation": 0.00588059498574519, "minimum": 0.10892494231211045, "maximum": 0.12619077479178725, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959414594438329, "standard_deviation": 0.0014209734041642278, "minimum": 0.9933677807022901, "maximum": 0.9975148001494958, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6132317053966065, "standard_deviation": 0.028859993578923382, "minimum": 0.5781642449897497, "maximum": 0.6622153703570741, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4694464964195826, "standard_deviation": 0.3311649167921271, "minimum": 0.06312946501917176, "maximum": 0.9696107270352087, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06137058391774708, "standard_deviation": 0.010627357873223824, "minimum": 0.04817359045575635, "maximum": 0.07980180628932634, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_237-134500-0031", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.898712396621704, "predicted_mos": 1.898712396621704, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d880fae7b90e2efc71a114acc365df6d7cf178482c79a674880c13d7ffa58767", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.501089700032026, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1777618562690773, "全帯域_10ms_rms_envelope_nrmse": 0.11792516237415418, "全帯域_10ms_envelope_correlation": 0.9865183889460664, "全帯域_10ms_modulation_mae_db": 0.8557508915538184, "全帯域_40ms_log_rms_mae_db": 0.6394386349180887, "全帯域_40ms_rms_envelope_nrmse": 0.05808011591611139, "全帯域_40ms_envelope_correlation": 0.9966818314656484, "全帯域_40ms_modulation_mae_db": 0.5719431668687001, "300–1000 Hz_10ms_log_rms_mae_db": 1.6882037692891565, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13987825818842792, "300–1000 Hz_10ms_envelope_correlation": 0.9835478766838913, "300–1000 Hz_10ms_modulation_mae_db": 1.1411910771324179, "300–1000 Hz_40ms_log_rms_mae_db": 0.969913199545557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07573649906312213, "300–1000 Hz_40ms_envelope_correlation": 0.9953559130465507, "300–1000 Hz_40ms_modulation_mae_db": 0.5839370775437228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.717332680613221, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14574045809546785, "1000–3000 Hz_10ms_envelope_correlation": 0.9832066702699583, "1000–3000 Hz_10ms_modulation_mae_db": 0.9314263119044859, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9660597820549138, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06819198882860084, "1000–3000 Hz_40ms_envelope_correlation": 0.9960235275426604, "1000–3000 Hz_40ms_modulation_mae_db": 0.43594869393034913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9747421352342192, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.515799000481429, "3000–6000 Hz_10ms_envelope_correlation": 0.8096213483887622, "3000–6000 Hz_10ms_modulation_mae_db": 2.7505340213975917, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3463435978314282, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.432199742226843, "3000–6000 Hz_40ms_envelope_correlation": 0.8462717147314246, "3000–6000 Hz_40ms_modulation_mae_db": 3.7138724311531264, "mr_stft_db": 5.826212464950007, "band_mae_db": 0.438140409039919, "rms_gain_error_db": -0.15884845134539555, "output_peak": 0.7233823225154417}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1792846995372137, "全帯域_10ms_rms_envelope_nrmse": 0.12512622521589883, "全帯域_10ms_envelope_correlation": 0.9851634737140359, "全帯域_10ms_modulation_mae_db": 0.5041232939908417, "全帯域_40ms_log_rms_mae_db": 0.6256887290219683, "全帯域_40ms_rms_envelope_nrmse": 0.07510248076181202, "全帯域_40ms_envelope_correlation": 0.9947371426792188, "全帯域_40ms_modulation_mae_db": 0.491240854771051, "300–1000 Hz_10ms_log_rms_mae_db": 1.706152233984911, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16012979663645488, "300–1000 Hz_10ms_envelope_correlation": 0.9791011283560958, "300–1000 Hz_10ms_modulation_mae_db": 0.8918611983921837, "300–1000 Hz_40ms_log_rms_mae_db": 0.998894396912093, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11302654169397973, "300–1000 Hz_40ms_envelope_correlation": 0.9897278937177879, "300–1000 Hz_40ms_modulation_mae_db": 0.46223216192967237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7878267152490759, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16406196417391672, "1000–3000 Hz_10ms_envelope_correlation": 0.9787475065011288, "1000–3000 Hz_10ms_modulation_mae_db": 0.8567831779948446, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0215489458578593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08601709830685737, "1000–3000 Hz_40ms_envelope_correlation": 0.9936288255648802, "1000–3000 Hz_40ms_modulation_mae_db": 0.7910849348342323, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9826955339393608, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5257730954991611, "3000–6000 Hz_10ms_envelope_correlation": 0.7986188503039537, "3000–6000 Hz_10ms_modulation_mae_db": 2.774291003008814, "3000–6000 Hz_40ms_log_rms_mae_db": 1.345036266967754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4437408008304061, "3000–6000 Hz_40ms_envelope_correlation": 0.8342476594322488, "3000–6000 Hz_40ms_modulation_mae_db": 3.7646989061770086, "mr_stft_db": 5.825550595176343, "band_mae_db": 0.5146100850934826, "rms_gain_error_db": -0.2604474021807713, "output_peak": 0.686657679145839}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.14292673603984, "全帯域_10ms_rms_envelope_nrmse": 0.11100267776609861, "全帯域_10ms_envelope_correlation": 0.9881570934042467, "全帯域_10ms_modulation_mae_db": 0.489409851749536, "全帯域_40ms_log_rms_mae_db": 0.5871390142208132, "全帯域_40ms_rms_envelope_nrmse": 0.054764881281378824, "全帯域_40ms_envelope_correlation": 0.9971616064184392, "全帯域_40ms_modulation_mae_db": 0.4348516131692039, "300–1000 Hz_10ms_log_rms_mae_db": 1.6543760954931204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.135545934171783, "300–1000 Hz_10ms_envelope_correlation": 0.9846182255601529, "300–1000 Hz_10ms_modulation_mae_db": 0.8681314061256752, "300–1000 Hz_40ms_log_rms_mae_db": 0.9488581969206183, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07325200465481205, "300–1000 Hz_40ms_envelope_correlation": 0.9957107354416986, "300–1000 Hz_40ms_modulation_mae_db": 0.38443087684741206, "1000–3000 Hz_10ms_log_rms_mae_db": 1.705349534515189, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15379436274221894, "1000–3000 Hz_10ms_envelope_correlation": 0.981281201276788, "1000–3000 Hz_10ms_modulation_mae_db": 0.6427672702196089, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9890292107394677, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08174735949475015, "1000–3000 Hz_40ms_envelope_correlation": 0.9943832728169968, "1000–3000 Hz_40ms_modulation_mae_db": 0.4647225524715437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9667635502738117, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.511264059967281, "3000–6000 Hz_10ms_envelope_correlation": 0.8101530592820868, "3000–6000 Hz_10ms_modulation_mae_db": 2.6252147554299925, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3251170051489463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.430729127756832, "3000–6000 Hz_40ms_envelope_correlation": 0.8445553105796021, "3000–6000 Hz_40ms_modulation_mae_db": 3.615222250875982, "mr_stft_db": 5.831266033715342, "band_mae_db": 0.4793025455984668, "rms_gain_error_db": -0.1807790866531302, "output_peak": 0.7632375799622398}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1519563567295577, "全帯域_10ms_rms_envelope_nrmse": 0.10240384106500901, "全帯域_10ms_envelope_correlation": 0.9898000429979347, "全帯域_10ms_modulation_mae_db": 0.14010767502960222, "全帯域_40ms_log_rms_mae_db": 0.5958525002436442, "全帯域_40ms_rms_envelope_nrmse": 0.05650942631074202, "全帯域_40ms_envelope_correlation": 0.9966326023152066, "全帯域_40ms_modulation_mae_db": 0.302208733292268, "300–1000 Hz_10ms_log_rms_mae_db": 1.6887649012462738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12340130094676709, "300–1000 Hz_10ms_envelope_correlation": 0.9869668689500343, "300–1000 Hz_10ms_modulation_mae_db": 0.7752361614197482, "300–1000 Hz_40ms_log_rms_mae_db": 0.9414094344929475, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06969066010290215, "300–1000 Hz_40ms_envelope_correlation": 0.9955248697194755, "300–1000 Hz_40ms_modulation_mae_db": 0.21126634716094128, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7348445426163337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16817553081695336, "1000–3000 Hz_10ms_envelope_correlation": 0.9777551372634831, "1000–3000 Hz_10ms_modulation_mae_db": 1.0315716487221838, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9881689430226877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.097016208737779, "1000–3000 Hz_40ms_envelope_correlation": 0.9924302227290648, "1000–3000 Hz_40ms_modulation_mae_db": 0.5861188805650888, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9376574754294902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5115725678168273, "3000–6000 Hz_10ms_envelope_correlation": 0.8138186421573204, "3000–6000 Hz_10ms_modulation_mae_db": 2.9424775815212385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2897526530252659, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4299866440109363, "3000–6000 Hz_40ms_envelope_correlation": 0.8490289979767351, "3000–6000 Hz_40ms_modulation_mae_db": 3.9264006561254328, "mr_stft_db": 5.825813229970266, "band_mae_db": 0.3758933165373593, "rms_gain_error_db": -0.09595793688882241, "output_peak": 0.8577768501882717}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1027862997610185, "全帯域_10ms_rms_envelope_nrmse": 0.11310952986599644, "全帯域_10ms_envelope_correlation": 0.9875619556240341, "全帯域_10ms_modulation_mae_db": 1.0988904972885494, "全帯域_40ms_log_rms_mae_db": 0.5690659176562151, "全帯域_40ms_rms_envelope_nrmse": 0.05079762056524598, "全帯域_40ms_envelope_correlation": 0.9972147798166848, "全帯域_40ms_modulation_mae_db": 1.0561801913774707, "300–1000 Hz_10ms_log_rms_mae_db": 1.6799277109539703, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14966307934026526, "300–1000 Hz_10ms_envelope_correlation": 0.9807259196556277, "300–1000 Hz_10ms_modulation_mae_db": 1.7251270578194415, "300–1000 Hz_40ms_log_rms_mae_db": 0.9040666824527109, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06358729786138186, "300–1000 Hz_40ms_envelope_correlation": 0.9962986266569152, "300–1000 Hz_40ms_modulation_mae_db": 1.4108155667673843, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7767468544310154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16599261892438644, "1000–3000 Hz_10ms_envelope_correlation": 0.9784645650079566, "1000–3000 Hz_10ms_modulation_mae_db": 0.25090226907830804, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0491883375239528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10844334380732891, "1000–3000 Hz_40ms_envelope_correlation": 0.9899933595581184, "1000–3000 Hz_40ms_modulation_mae_db": 0.2226039171635571, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0178651393926543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.511254891241323, "3000–6000 Hz_10ms_envelope_correlation": 0.8111816655690384, "3000–6000 Hz_10ms_modulation_mae_db": 2.7678509175506307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3320150018887846, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4281241374057425, "3000–6000 Hz_40ms_envelope_correlation": 0.8476321723961212, "3000–6000 Hz_40ms_modulation_mae_db": 3.701186203170861, "mr_stft_db": 5.814886873407544, "band_mae_db": 0.4219267727842901, "rms_gain_error_db": -0.006729529415270643, "output_peak": 0.8061077691769912}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1054436446702591, "全帯域_10ms_rms_envelope_nrmse": 0.10509395652566558, "全帯域_10ms_envelope_correlation": 0.9893793232329233, "全帯域_10ms_modulation_mae_db": 0.4595476137556182, "全帯域_40ms_log_rms_mae_db": 0.6034043472486759, "全帯域_40ms_rms_envelope_nrmse": 0.060376952151454666, "全帯域_40ms_envelope_correlation": 0.9963616697611681, "全帯域_40ms_modulation_mae_db": 0.6851316753103647, "300–1000 Hz_10ms_log_rms_mae_db": 1.6647764853105598, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14432949080944613, "300–1000 Hz_10ms_envelope_correlation": 0.982576709189336, "300–1000 Hz_10ms_modulation_mae_db": 0.2446728460627697, "300–1000 Hz_40ms_log_rms_mae_db": 0.9340288483565811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09265165814398756, "300–1000 Hz_40ms_envelope_correlation": 0.9927619261371143, "300–1000 Hz_40ms_modulation_mae_db": 0.32859560945493577, "1000–3000 Hz_10ms_log_rms_mae_db": 1.755777833996393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.157843855754363, "1000–3000 Hz_10ms_envelope_correlation": 0.9802821208290587, "1000–3000 Hz_10ms_modulation_mae_db": 1.3582289020731204, "1000–3000 Hz_40ms_log_rms_mae_db": 1.040866848226131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08830950154645403, "1000–3000 Hz_40ms_envelope_correlation": 0.9933000161676803, "1000–3000 Hz_40ms_modulation_mae_db": 0.9910387624511653, "3000–6000 Hz_10ms_log_rms_mae_db": 1.951195542741253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5008757404424226, "3000–6000 Hz_10ms_envelope_correlation": 0.8256421516221218, "3000–6000 Hz_10ms_modulation_mae_db": 2.832958320984486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.341087715529962, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4238237305586603, "3000–6000 Hz_40ms_envelope_correlation": 0.8571073765449428, "3000–6000 Hz_40ms_modulation_mae_db": 3.73998732998299, "mr_stft_db": 5.819436597174761, "band_mae_db": 0.4744123918152341, "rms_gain_error_db": -0.1631005517085464, "output_peak": 0.7328730387687582}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.131988633596028, "全帯域_10ms_rms_envelope_nrmse": 0.10251765799574174, "全帯域_10ms_envelope_correlation": 0.9898340738134985, "全帯域_10ms_modulation_mae_db": 0.45271942626328454, "全帯域_40ms_log_rms_mae_db": 0.5797873820119469, "全帯域_40ms_rms_envelope_nrmse": 0.04819944268708495, "全帯域_40ms_envelope_correlation": 0.9977140906547702, "全帯域_40ms_modulation_mae_db": 0.26229372765735853, "300–1000 Hz_10ms_log_rms_mae_db": 1.7017924355769627, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14631301449342907, "300–1000 Hz_10ms_envelope_correlation": 0.9818534470767006, "300–1000 Hz_10ms_modulation_mae_db": 1.4780992794871908, "300–1000 Hz_40ms_log_rms_mae_db": 0.9310718499991081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08219269123929575, "300–1000 Hz_40ms_envelope_correlation": 0.9942512275038968, "300–1000 Hz_40ms_modulation_mae_db": 0.9404200532920192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7728689852659199, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15451845152006885, "1000–3000 Hz_10ms_envelope_correlation": 0.9811339518825815, "1000–3000 Hz_10ms_modulation_mae_db": 0.5933856390626547, "1000–3000 Hz_40ms_log_rms_mae_db": 1.026553747451855, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08278016200347248, "1000–3000 Hz_40ms_envelope_correlation": 0.9941052030472851, "1000–3000 Hz_40ms_modulation_mae_db": 0.58238016540858, "3000–6000 Hz_10ms_log_rms_mae_db": 1.952365779713859, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.525340764959239, "3000–6000 Hz_10ms_envelope_correlation": 0.8096879313142827, "3000–6000 Hz_10ms_modulation_mae_db": 2.7064860056640896, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3251234916999435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.44462798569406536, "3000–6000 Hz_40ms_envelope_correlation": 0.8448572150625738, "3000–6000 Hz_40ms_modulation_mae_db": 3.6301007515991857, "mr_stft_db": 5.813570635964194, "band_mae_db": 0.39089322636263635, "rms_gain_error_db": -0.135944955109155, "output_peak": 0.7990176021877091}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.125791097032, "全帯域_10ms_rms_envelope_nrmse": 0.11893004552349871, "全帯域_10ms_envelope_correlation": 0.9862105221443177, "全帯域_10ms_modulation_mae_db": 0.6657223154973932, "全帯域_40ms_log_rms_mae_db": 0.5545810334383553, "全帯域_40ms_rms_envelope_nrmse": 0.05679373991341029, "全帯域_40ms_envelope_correlation": 0.9966624581934641, "全帯域_40ms_modulation_mae_db": 0.39494738549668623, "300–1000 Hz_10ms_log_rms_mae_db": 1.7415473297462538, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1568742616316229, "300–1000 Hz_10ms_envelope_correlation": 0.9788209611517525, "300–1000 Hz_10ms_modulation_mae_db": 1.401719479096365, "300–1000 Hz_40ms_log_rms_mae_db": 0.9242666749917198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08130783266340783, "300–1000 Hz_40ms_envelope_correlation": 0.994012345341037, "300–1000 Hz_40ms_modulation_mae_db": 0.816687248226965, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6998675089109685, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1336284025622272, "1000–3000 Hz_10ms_envelope_correlation": 0.9858967695044835, "1000–3000 Hz_10ms_modulation_mae_db": 0.49681548948485427, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9511158914965284, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06064039830481578, "1000–3000 Hz_40ms_envelope_correlation": 0.9969316651641839, "1000–3000 Hz_40ms_modulation_mae_db": 0.42337733885600787, "3000–6000 Hz_10ms_log_rms_mae_db": 2.005584540073506, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5216316504557919, "3000–6000 Hz_10ms_envelope_correlation": 0.8043200796081763, "3000–6000 Hz_10ms_modulation_mae_db": 2.6731463050094617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.355841149001978, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.43554355628663216, "3000–6000 Hz_40ms_envelope_correlation": 0.8427280635468841, "3000–6000 Hz_40ms_modulation_mae_db": 3.6769759664797, "mr_stft_db": 5.826310387114432, "band_mae_db": 0.38574869073400836, "rms_gain_error_db": -0.12367321927549096, "output_peak": 0.8034803337096431}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.922475737223668, "spectral_convergence": 0.36700673563732505, "spectral_convergence_status": "ok", "amplitude_floor": 2.051312383514402e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.630326752644275, "spectral_convergence": 0.37470989411628153, "spectral_convergence_status": "ok", "amplitude_floor": 1.425885904641783e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.925834904982078, "spectral_convergence": 0.3800760254492041, "spectral_convergence_status": "ok", "amplitude_floor": 8.483800566678623e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001304494586920238, "output_power": 0.00015565772527893125, "delta_db": 0.7672839824099338, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014301207186140645, "output_power": 0.0014059203688284545, "delta_db": -0.07411975464876977, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.004308309383381418, "output_power": 0.004051526245436202, "delta_db": -0.26688226229656364, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.001304992137209625, "output_power": 0.0012887958646381108, "delta_db": -0.05423761095310109, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002442775973124686, "output_power": 0.0002455092609021707, "delta_db": 0.02184239201701358, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0021123865382114e-05, "output_power": 1.397539447894716e-05, "delta_db": 1.444476451914132, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.438140409039919, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15884845134539555, "pitch": null, "output_rms_dbfs": -21.450029862521944, "output_peak": 0.7233823225154417, "reference_rms_dbfs": -21.29118141117655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.826212464950007}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002631236493882886, "log_rms_mae_db": 1.1777618562690773, "rms_envelope_nrmse": 0.11792516237415418, "envelope_correlation": 0.9865183889460664, "rms_gain_error_db": -0.15884839215746732, "reference_mean_power": 0.007434364903539191, "output_mean_power": 0.007167356964447251}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.450307530079474e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015621941098031631, "output_power": 0.001416562277699288, "reference_active": true, "delta_db": -0.4249932339589434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031400948871223737, "output_power": 0.0003210410940362262, "reference_active": true, "delta_db": 0.09617854961225136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9119711962951405e-05, "output_power": 7.86799183493096e-05, "reference_active": true, "delta_db": 2.0460808910902606}], "active_band_count": 3, "active_band_mae_db": 0.8557508915538184}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024033810696394756, "log_rms_mae_db": 0.6394386349180887, "rms_envelope_nrmse": 0.05808011591611139, "envelope_correlation": 0.9966818314656484, "rms_gain_error_db": -0.15884871774354323, "reference_mean_power": 0.007471745584669175, "output_mean_power": 0.007203394564688979}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.136699874330791e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014162642987993615, "output_power": 0.001302726120811952, "reference_active": true, "delta_db": -0.3629118636932905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001495506387635467, "output_power": 0.00015390605016101034, "reference_active": true, "delta_db": 0.12467420267136507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0546566248780076e-06, "output_power": 4.053094610838095e-06, "reference_active": true, "delta_db": 1.2282434342414446}], "active_band_count": 3, "active_band_mae_db": 0.5719431668687001}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002303175773961387, "log_rms_mae_db": 1.6882037692891565, "rms_envelope_nrmse": 0.13987825818842792, "envelope_correlation": 0.9835478766838913, "rms_gain_error_db": -0.27801037210733903, "reference_mean_power": 0.004944923373977616, "output_mean_power": 0.004638296733024684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6955719701225893e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012875788279854436, "output_power": 0.0011712657478566747, "reference_active": true, "delta_db": -0.4111838375371786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026650258378785487, "output_power": 0.0002608463207002802, "reference_active": true, "delta_db": -0.09316708667740967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0302791253002337e-05, "output_power": 5.9347832847217745e-05, "reference_active": true, "delta_db": 2.9192223071826655}], "active_band_count": 3, "active_band_mae_db": 1.1411910771324179}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020459712012168156, "log_rms_mae_db": 0.969913199545557, "rms_envelope_nrmse": 0.07573649906312213, "envelope_correlation": 0.9953559130465507, "rms_gain_error_db": -0.2780104284036734, "reference_mean_power": 0.004969790940376444, "output_mean_power": 0.004661622241724128}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.46585734788702e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011840763589577269, "output_power": 0.00109152999128348, "reference_active": true, "delta_db": -0.35344037115522564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012134699925967192, "output_power": 0.00011967313288429466, "reference_active": true, "delta_db": -0.0603238087677255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.482336680977429e-06, "output_power": 3.3780446844477712e-06, "reference_active": true, "delta_db": 1.3380470527082173}], "active_band_count": 3, "active_band_mae_db": 0.5839370775437228}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014083569934759876, "log_rms_mae_db": 1.717332680613221, "rms_envelope_nrmse": 0.14574045809546785, "envelope_correlation": 0.9832066702699583, "rms_gain_error_db": -0.06554418067734152, "reference_mean_power": 0.0013181968247743324, "output_mean_power": 0.001298451830982041}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.706116677050846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003562978060875054, "output_power": 0.00032344394562704507, "reference_active": true, "delta_db": -0.4201412195213621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927249258823322e-05, "output_power": 6.268604440602353e-05, "reference_active": true, "delta_db": -0.4338994886405855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3444445693793863e-05, "output_power": 2.1016806415485857e-05, "reference_active": true, "delta_db": 1.94023822755151}], "active_band_count": 3, "active_band_mae_db": 0.9314263119044859}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010879087849669345, "log_rms_mae_db": 0.9660597820549138, "rms_envelope_nrmse": 0.06819198882860084, "envelope_correlation": 0.9960235275426604, "rms_gain_error_db": -0.0655449544464001, "reference_mean_power": 0.0013248216670349005, "output_mean_power": 0.001304977208615941}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.107952744901008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003355694425017771, "output_power": 0.00030936450743934804, "reference_active": true, "delta_db": -0.35311919617030924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.409137696567018e-05, "output_power": 3.152224962953425e-05, "reference_active": true, "delta_db": -0.3402733896973084}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.097157965658087e-07, "output_power": 9.327772847957055e-07, "reference_active": true, "delta_db": 0.6144534959234298}], "active_band_count": 3, "active_band_mae_db": 0.43594869393034913}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.557718084358035e-05, "log_rms_mae_db": 1.9747421352342192, "rms_envelope_nrmse": 0.515799000481429, "envelope_correlation": 0.8096213483887622, "rms_gain_error_db": 1.043045182727074, "reference_mean_power": 2.3934708718272692e-05, "output_mean_power": 3.043215202696174e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3699291941082093e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.018836822609012e-06, "output_power": 5.781103980986082e-06, "reference_active": true, "delta_db": 1.5791040798920155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.745214583455981e-06, "output_power": 6.8687385317461286e-06, "reference_active": true, "delta_db": 1.6062112820898669}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3999346848964708e-06, "output_power": 4.495069842496959e-06, "reference_active": true, "delta_db": 5.066286702210893}], "active_band_count": 3, "active_band_mae_db": 2.7505340213975917}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.6234588308267947e-05, "log_rms_mae_db": 1.3463435978314282, "rms_envelope_nrmse": 0.432199742226843, "envelope_correlation": 0.8462717147314246, "rms_gain_error_db": 1.0430698313578788, "reference_mean_power": 2.405439007374668e-05, "output_mean_power": 3.058449630319342e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1165079820636279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.688364032412796e-06, "output_power": 7.3150080048012975e-06, "reference_active": true, "delta_db": 2.973810274204861}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313243490564173e-06, "output_power": 5.3765182988041614e-06, "reference_active": true, "delta_db": 2.1024777485343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.200263616160695e-07, "output_power": 4.85075755222857e-07, "reference_active": true, "delta_db": 6.065329270720219}], "active_band_count": 3, "active_band_mae_db": 3.7138724311531264}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8586358324606407, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6a4bfb010b8ecc31f11ffdd91e0d17feff10e14617593d6b916b4c54a04164a8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9808459903169298, "standard_deviation": 0.002707177046025162, "minimum": 0.9777551372634831, "maximum": 0.9858967695044835, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7438268319497645, "standard_deviation": 0.03422934581881423, "minimum": 1.6998675089109685, "maximum": 1.7878267152490759, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7702350885675076, "standard_deviation": 0.34630608503469645, "minimum": 0.25090226907830804, "maximum": 1.3582289020731204, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1554694555737003, "standard_deviation": 0.011508044568541062, "minimum": 0.1336284025622272, "maximum": 0.16817553081695336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938495115738588, "standard_deviation": 0.0021317743143294296, "minimum": 0.9899933595581184, "maximum": 0.9969316651641839, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0040664632966745, "standard_deviation": 0.035714518852827344, "minimum": 0.9511158914965284, "maximum": 1.0491883375239528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5621594057100655, "standard_deviation": 0.23824021853728788, "minimum": 0.2226039171635571, "maximum": 0.9910387624511653, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08414325762875732, "standard_deviation": 0.015078374562351383, "minimum": 0.06064039830481578, "maximum": 0.10844334380732891, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8103804660307178, "standard_deviation": 0.007760278793963918, "minimum": 0.7986188503039537, "maximum": 0.8256421516221218, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9736087120997694, "standard_deviation": 0.02766612138150835, "minimum": 1.9376574754294902, "maximum": 2.0178651393926543, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.7591198638207883, "standard_deviation": 0.09823051296570583, "minimum": 2.6252147554299925, "maximum": 2.9424775815212385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5154389713579344, "standard_deviation": 0.008496148003243772, "minimum": 0.5008757404424226, "maximum": 0.5257730954991611, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8458035637838166, "standard_deviation": 0.006405287129642815, "minimum": 0.8342476594322488, "maximum": 0.8571073765449428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3325396101367577, "standard_deviation": 0.020367485003853648, "minimum": 1.2897526530252659, "maximum": 1.355841149001978, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.7210555619455357, "standard_deviation": 0.09731585381052212, "minimum": 3.615222250875982, "maximum": 3.9264006561254328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.4335969655962647, "standard_deviation": 0.00733922079789107, "minimum": 0.4238237305586603, "maximum": 0.44462798569406536, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9822763920779489, "standard_deviation": 0.002772228876656321, "minimum": 0.9788209611517525, "maximum": 0.9869668689500343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.690692620200151, "standard_deviation": 0.02689858415685634, "minimum": 1.6543760954931204, "maximum": 1.7415473297462538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0657548131919738, "standard_deviation": 0.47074914256561606, "minimum": 0.2446728460627697, "maximum": 1.7251270578194415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14451689202727452, "standard_deviation": 0.011796337132948134, "minimum": 0.12340130094676709, "maximum": 0.16012979663645488, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942054421955595, "standard_deviation": 0.0021341898023589374, "minimum": 0.9897278937177879, "maximum": 0.9962986266569152, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.944063660458917, "standard_deviation": 0.029193454018031716, "minimum": 0.9040666824527109, "maximum": 0.998894396912093, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6422981176528816, "standard_deviation": 0.3957687367790751, "minimum": 0.21126634716094128, "maximum": 1.4108155667673843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08143064817786114, "standard_deviation": 0.015493470327641996, "minimum": 0.06358729786138186, "maximum": 0.11302654169397973, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4351159297456746, "standard_deviation": 0.05056704279593609, "minimum": 0.3758933165373593, "maximum": 0.5146100850934826, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.822880852184111, "standard_deviation": 0.006226047877476738, "minimum": 5.813570635964194, "maximum": 5.831266033715342, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7715666469568616, "standard_deviation": 0.05541422872833329, "minimum": 0.686657679145839, "maximum": 0.8577768501882717, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1406851415720728, "standard_deviation": 0.07274160105349994, "minimum": -0.2604474021807713, "maximum": -0.006729529415270643, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9878281092346322, "standard_deviation": 0.0017699873906718304, "minimum": 0.9851634737140359, "maximum": 0.9898340738134985, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1397424154543745, "standard_deviation": 0.029192246898024618, "minimum": 1.1027862997610185, "maximum": 1.1792846995372137, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5832839456410804, "standard_deviation": 0.290498628518264, "minimum": 0.14010767502960222, "maximum": 1.0988904972885494, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11201363704150788, "standard_deviation": 0.008343942213260758, "minimum": 0.10240384106500901, "maximum": 0.12512622521589883, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9966457726630751, "standard_deviation": 0.0008828923176363822, "minimum": 0.9947371426792188, "maximum": 0.9977140906547702, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5943696948449635, "standard_deviation": 0.028244494339346334, "minimum": 0.5545810334383553, "maximum": 0.6394386349180887, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5248496684928878, "standard_deviation": 0.25490296680221847, "minimum": 0.26229372765735853, "maximum": 1.0561801913774707, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05757808244840501, "standard_deviation": 0.008095811196845645, "minimum": 0.04819944268708495, "maximum": 0.07510248076181202, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_260-123286-0008", "language": "en", "style": "normal", "speaker": "librispeech_260", "source_utterance_id": "260-123286-0008", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 59760, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_260-123286-0008.wav", "source_file_sha256": "37cf2ed280012ad7d7ab19beb9fc14715587ad14c65e0efe0f09f55ef4f8e87c", "decoded_pcm_sha256": "931193da803cf63a31b439032de16719f19aca2c3130d571b4b8044d43a1413c", "original_source_sha256": "735769549e4dbbfb3f4ba43f89866cde8085933d8b16e67550cc65264bf84fc3", "corpus_entry_id": "librispeech_260-123286-0008"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.9420230388641357, "predicted_mos": 3.9420230388641357, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "1f879e0baae5d4d51551094678fdeb23ba773ef4d866a1b2f2aa67a9fcaca2d0", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.57172277599966, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_260-123286-0008", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1f879e0baae5d4d51551094678fdeb23ba773ef4d866a1b2f2aa67a9fcaca2d0", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "fde9082de1536b3cb7b9b64030ee3c897ee296663d8ee9fc2fea3b658c7a179e"}, "rows": [{"case": "librispeech_260-123286-0008", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1337292194366455, "predicted_mos": 2.1337292194366455, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "1909d51b82262bfc04a2da32ae3b9a3101cf634acd34d07449102bd3a0bf0175", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.457493030000478, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_260-123286-0008", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1909d51b82262bfc04a2da32ae3b9a3101cf634acd34d07449102bd3a0bf0175", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0200142781986736, "全帯域_10ms_rms_envelope_nrmse": 0.0938958735602321, "全帯域_10ms_envelope_correlation": 0.992238786560505, "全帯域_10ms_modulation_mae_db": 0.4661362029640705, "全帯域_40ms_log_rms_mae_db": 0.5175589774192852, "全帯域_40ms_rms_envelope_nrmse": 0.05268733243164517, "全帯域_40ms_envelope_correlation": 0.9974249024564145, "全帯域_40ms_modulation_mae_db": 0.44857485581215323, "300–1000 Hz_10ms_log_rms_mae_db": 1.5539489721763138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14089527372541039, "300–1000 Hz_10ms_envelope_correlation": 0.9860089099892293, "300–1000 Hz_10ms_modulation_mae_db": 1.063495372919211, "300–1000 Hz_40ms_log_rms_mae_db": 0.9013202432752945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08000164124245146, "300–1000 Hz_40ms_envelope_correlation": 0.9952534987704859, "300–1000 Hz_40ms_modulation_mae_db": 0.9125269952944879, "1000–3000 Hz_10ms_log_rms_mae_db": 1.804161698500498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21286396249015543, "1000–3000 Hz_10ms_envelope_correlation": 0.9644744835387802, "1000–3000 Hz_10ms_modulation_mae_db": 0.6434374549836069, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2259232130481463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14660296895064276, "1000–3000 Hz_40ms_envelope_correlation": 0.982004213280334, "1000–3000 Hz_40ms_modulation_mae_db": 0.775742575374438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1901931581382206, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22698983173263232, "3000–6000 Hz_10ms_envelope_correlation": 0.9679049387167982, "3000–6000 Hz_10ms_modulation_mae_db": 0.33436149413562627, "3000–6000 Hz_40ms_log_rms_mae_db": 1.888653276506732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15723924848695994, "3000–6000 Hz_40ms_envelope_correlation": 0.9840874705910243, "3000–6000 Hz_40ms_modulation_mae_db": 0.29410983561012966, "mr_stft_db": 5.574681038826739, "band_mae_db": 0.15862775320015468, "rms_gain_error_db": -0.0761416551763979, "output_peak": 0.5002645106449802}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0587688919055753, "全帯域_10ms_rms_envelope_nrmse": 0.10011701192559794, "全帯域_10ms_envelope_correlation": 0.9912884666709505, "全帯域_10ms_modulation_mae_db": 0.46675222035994296, "全帯域_40ms_log_rms_mae_db": 0.5784582593374062, "全帯域_40ms_rms_envelope_nrmse": 0.05492616403054361, "全帯域_40ms_envelope_correlation": 0.9973923906168566, "全帯域_40ms_modulation_mae_db": 0.2819086159642106, "300–1000 Hz_10ms_log_rms_mae_db": 1.7441231337662417, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15654539166316234, "300–1000 Hz_10ms_envelope_correlation": 0.982955566244646, "300–1000 Hz_10ms_modulation_mae_db": 1.4960816917203907, "300–1000 Hz_40ms_log_rms_mae_db": 1.0726319339804142, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0848249330958489, "300–1000 Hz_40ms_envelope_correlation": 0.9953066898990892, "300–1000 Hz_40ms_modulation_mae_db": 1.0568826348154177, "1000–3000 Hz_10ms_log_rms_mae_db": 1.860834443578954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21964886652890694, "1000–3000 Hz_10ms_envelope_correlation": 0.9622600984457029, "1000–3000 Hz_10ms_modulation_mae_db": 0.6306249036673551, "1000–3000 Hz_40ms_log_rms_mae_db": 1.206396971617456, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14056655636190832, "1000–3000 Hz_40ms_envelope_correlation": 0.9840144579318024, "1000–3000 Hz_40ms_modulation_mae_db": 0.5314533666506679, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2351671354006837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22316985290223645, "3000–6000 Hz_10ms_envelope_correlation": 0.9670402906229445, "3000–6000 Hz_10ms_modulation_mae_db": 0.11275214225448761, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9047417773653479, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15176571560037044, "3000–6000 Hz_40ms_envelope_correlation": 0.9840687583955513, "3000–6000 Hz_40ms_modulation_mae_db": 0.13599656453151932, "mr_stft_db": 5.579725924190872, "band_mae_db": 0.17326689945533225, "rms_gain_error_db": -0.15322107602144827, "output_peak": 0.4662714301608534}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.998346135243834, "全帯域_10ms_rms_envelope_nrmse": 0.09644844027584105, "全帯域_10ms_envelope_correlation": 0.9920000484382109, "全帯域_10ms_modulation_mae_db": 0.31509010773153573, "全帯域_40ms_log_rms_mae_db": 0.5573154860327575, "全帯域_40ms_rms_envelope_nrmse": 0.06213991805963035, "全帯域_40ms_envelope_correlation": 0.9966696466394253, "全帯域_40ms_modulation_mae_db": 0.3010210330377494, "300–1000 Hz_10ms_log_rms_mae_db": 1.6188353379453295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1497280668793862, "300–1000 Hz_10ms_envelope_correlation": 0.9847633041233183, "300–1000 Hz_10ms_modulation_mae_db": 1.0695378808618037, "300–1000 Hz_40ms_log_rms_mae_db": 0.9881054560359017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09359956225612054, "300–1000 Hz_40ms_envelope_correlation": 0.9944807889356011, "300–1000 Hz_40ms_modulation_mae_db": 0.8428957066578181, "1000–3000 Hz_10ms_log_rms_mae_db": 1.767531063810179, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.191420115769772, "1000–3000 Hz_10ms_envelope_correlation": 0.9715913682684754, "1000–3000 Hz_10ms_modulation_mae_db": 0.7492344685452567, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2070666962730483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12290253433314327, "1000–3000 Hz_40ms_envelope_correlation": 0.9879924799947635, "1000–3000 Hz_40ms_modulation_mae_db": 0.9121242842474352, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2453909362089344, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24582909342136816, "3000–6000 Hz_10ms_envelope_correlation": 0.9601087568538509, "3000–6000 Hz_10ms_modulation_mae_db": 0.5484438346517617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9514370846681535, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15872363159322134, "3000–6000 Hz_40ms_envelope_correlation": 0.9825523727292618, "3000–6000 Hz_40ms_modulation_mae_db": 0.5478132832950832, "mr_stft_db": 5.576431966521856, "band_mae_db": 0.17750739774080673, "rms_gain_error_db": -0.17438495442134969, "output_peak": 0.47085085869830107}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.024744416368248, "全帯域_10ms_rms_envelope_nrmse": 0.09173613752064999, "全帯域_10ms_envelope_correlation": 0.992628192870897, "全帯域_10ms_modulation_mae_db": 0.1536878691420159, "全帯域_40ms_log_rms_mae_db": 0.5842453580335285, "全帯域_40ms_rms_envelope_nrmse": 0.05201905909617722, "全帯域_40ms_envelope_correlation": 0.9975208551008433, "全帯域_40ms_modulation_mae_db": 0.09779259225236932, "300–1000 Hz_10ms_log_rms_mae_db": 1.6430762859695327, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1430246554229477, "300–1000 Hz_10ms_envelope_correlation": 0.9859986882397471, "300–1000 Hz_10ms_modulation_mae_db": 0.27263325456121357, "300–1000 Hz_40ms_log_rms_mae_db": 1.0333043078136595, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08835524703724189, "300–1000 Hz_40ms_envelope_correlation": 0.9949322561406598, "300–1000 Hz_40ms_modulation_mae_db": 0.6625417969576315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.809330755907883, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19728461689297871, "1000–3000 Hz_10ms_envelope_correlation": 0.9696275900492962, "1000–3000 Hz_10ms_modulation_mae_db": 0.4062237980610002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.193853462174276, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1155601069154866, "1000–3000 Hz_40ms_envelope_correlation": 0.9887157761001473, "1000–3000 Hz_40ms_modulation_mae_db": 0.5465669956891965, "3000–6000 Hz_10ms_log_rms_mae_db": 2.235746285662008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23693567069693464, "3000–6000 Hz_10ms_envelope_correlation": 0.9635385105718182, "3000–6000 Hz_10ms_modulation_mae_db": 0.4498396656075574, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8738335593268314, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15570441085701733, "3000–6000 Hz_40ms_envelope_correlation": 0.9835099107027854, "3000–6000 Hz_40ms_modulation_mae_db": 0.31790694320248175, "mr_stft_db": 5.578535725341467, "band_mae_db": 0.13909293913508822, "rms_gain_error_db": -0.0827726528987868, "output_peak": 0.4832949643412854}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.071729450597361, "全帯域_10ms_rms_envelope_nrmse": 0.09938341601954448, "全帯域_10ms_envelope_correlation": 0.9913791785739623, "全帯域_10ms_modulation_mae_db": 0.18611296603652913, "全帯域_40ms_log_rms_mae_db": 0.5847912185486261, "全帯域_40ms_rms_envelope_nrmse": 0.054583857176274446, "全帯域_40ms_envelope_correlation": 0.9973856501033904, "全帯域_40ms_modulation_mae_db": 0.15441645821452285, "300–1000 Hz_10ms_log_rms_mae_db": 1.6308013728717128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14983593782169063, "300–1000 Hz_10ms_envelope_correlation": 0.9845452882416881, "300–1000 Hz_10ms_modulation_mae_db": 0.29583598591136534, "300–1000 Hz_40ms_log_rms_mae_db": 0.9712646545076487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08807847188715681, "300–1000 Hz_40ms_envelope_correlation": 0.995005851437168, "300–1000 Hz_40ms_modulation_mae_db": 0.2165273137521834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8154564879229524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23353245525535368, "1000–3000 Hz_10ms_envelope_correlation": 0.9573223801599843, "1000–3000 Hz_10ms_modulation_mae_db": 0.8296063997124792, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2507722944697621, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14397824393361516, "1000–3000 Hz_40ms_envelope_correlation": 0.9835938034034349, "1000–3000 Hz_40ms_modulation_mae_db": 1.0108939952736438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.167641753634564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22712997973665972, "3000–6000 Hz_10ms_envelope_correlation": 0.9657962428710672, "3000–6000 Hz_10ms_modulation_mae_db": 0.511931830739465, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9029553829166985, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15390317537526632, "3000–6000 Hz_40ms_envelope_correlation": 0.9836753715540496, "3000–6000 Hz_40ms_modulation_mae_db": 0.4999684431451616, "mr_stft_db": 5.5670995788306845, "band_mae_db": 0.17829869354347003, "rms_gain_error_db": -0.13360068592432886, "output_peak": 0.5092480994824069}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9647509300341365, "全帯域_10ms_rms_envelope_nrmse": 0.08493328497853286, "全帯域_10ms_envelope_correlation": 0.9936917666645015, "全帯域_10ms_modulation_mae_db": 0.3497344686692497, "全帯域_40ms_log_rms_mae_db": 0.5239732805391376, "全帯域_40ms_rms_envelope_nrmse": 0.04694580323377159, "全帯域_40ms_envelope_correlation": 0.9980136957221474, "全帯域_40ms_modulation_mae_db": 0.361140822622658, "300–1000 Hz_10ms_log_rms_mae_db": 1.6189501451669412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1339962973389865, "300–1000 Hz_10ms_envelope_correlation": 0.987368354655864, "300–1000 Hz_10ms_modulation_mae_db": 0.5756916793498893, "300–1000 Hz_40ms_log_rms_mae_db": 0.9677878770823676, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07454592725665214, "300–1000 Hz_40ms_envelope_correlation": 0.9960051460076664, "300–1000 Hz_40ms_modulation_mae_db": 1.0262776619114047, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8026547867271998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20680458933574716, "1000–3000 Hz_10ms_envelope_correlation": 0.9665597811739698, "1000–3000 Hz_10ms_modulation_mae_db": 0.6941264162985682, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2503472639892994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13345595000548668, "1000–3000 Hz_40ms_envelope_correlation": 0.9852998629066485, "1000–3000 Hz_40ms_modulation_mae_db": 0.7507507711378616, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1557424944938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2336817879542591, "3000–6000 Hz_10ms_envelope_correlation": 0.9642353236880011, "3000–6000 Hz_10ms_modulation_mae_db": 0.39791069277349345, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8666703320907867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15631229350842352, "3000–6000 Hz_40ms_envelope_correlation": 0.9831118539572011, "3000–6000 Hz_40ms_modulation_mae_db": 0.4591843341035899, "mr_stft_db": 5.570130799457421, "band_mae_db": 0.1411667006152493, "rms_gain_error_db": -0.0892942247685696, "output_peak": 0.4644521562362647}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9773234347042804, "全帯域_10ms_rms_envelope_nrmse": 0.10320669817784571, "全帯域_10ms_envelope_correlation": 0.9906396222738161, "全帯域_10ms_modulation_mae_db": 0.36208416251550785, "全帯域_40ms_log_rms_mae_db": 0.4975748852289342, "全帯域_40ms_rms_envelope_nrmse": 0.05951401075186895, "全帯域_40ms_envelope_correlation": 0.9967389499265439, "全帯域_40ms_modulation_mae_db": 0.16457192626584202, "300–1000 Hz_10ms_log_rms_mae_db": 1.6453247286540544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18185221400697468, "300–1000 Hz_10ms_envelope_correlation": 0.9767928272627854, "300–1000 Hz_10ms_modulation_mae_db": 0.30156454893249995, "300–1000 Hz_40ms_log_rms_mae_db": 0.9944416269877774, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1125632294345015, "300–1000 Hz_40ms_envelope_correlation": 0.9912300150398424, "300–1000 Hz_40ms_modulation_mae_db": 0.28590732815643094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.816117812862572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20882610925756384, "1000–3000 Hz_10ms_envelope_correlation": 0.9658501386894729, "1000–3000 Hz_10ms_modulation_mae_db": 0.6322185338482642, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2905359782084704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12671334350286387, "1000–3000 Hz_40ms_envelope_correlation": 0.9864760020478889, "1000–3000 Hz_40ms_modulation_mae_db": 0.4593061974794915, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1543792827350194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22475683582292139, "3000–6000 Hz_10ms_envelope_correlation": 0.967033932539659, "3000–6000 Hz_10ms_modulation_mae_db": 0.1665653964722173, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8607017344971364, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1542054382983421, "3000–6000 Hz_40ms_envelope_correlation": 0.9835764272709203, "3000–6000 Hz_40ms_modulation_mae_db": 0.21598001253559398, "mr_stft_db": 5.571858403468295, "band_mae_db": 0.14979429381386497, "rms_gain_error_db": -0.10055198010473025, "output_peak": 0.48192073633167587}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9881597508131418, "全帯域_10ms_rms_envelope_nrmse": 0.09479151567003163, "全帯域_10ms_envelope_correlation": 0.9921355137181584, "全帯域_10ms_modulation_mae_db": 0.08350203786272921, "全帯域_40ms_log_rms_mae_db": 0.5631891844053841, "全帯域_40ms_rms_envelope_nrmse": 0.057248540633641355, "全帯域_40ms_envelope_correlation": 0.997030171350042, "全帯域_40ms_modulation_mae_db": 0.07626769782739302, "300–1000 Hz_10ms_log_rms_mae_db": 1.6761123505622757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1487501594246225, "300–1000 Hz_10ms_envelope_correlation": 0.984470688538578, "300–1000 Hz_10ms_modulation_mae_db": 0.3593796521081582, "300–1000 Hz_40ms_log_rms_mae_db": 1.026231330123369, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08571649527699432, "300–1000 Hz_40ms_envelope_correlation": 0.9948395218126339, "300–1000 Hz_40ms_modulation_mae_db": 0.6514947090775097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7703579175154796, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21944528538618502, "1000–3000 Hz_10ms_envelope_correlation": 0.9634491423724352, "1000–3000 Hz_10ms_modulation_mae_db": 0.48484058702739824, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1524009351012252, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1313875407848432, "1000–3000 Hz_40ms_envelope_correlation": 0.9857595247869891, "1000–3000 Hz_40ms_modulation_mae_db": 0.5249055672781409, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1501773909718325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2160294156336995, "3000–6000 Hz_10ms_envelope_correlation": 0.9697062534191286, "3000–6000 Hz_10ms_modulation_mae_db": 0.15972447371271298, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8402068898936674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14956811574697682, "3000–6000 Hz_40ms_envelope_correlation": 0.9848688183672759, "3000–6000 Hz_40ms_modulation_mae_db": 0.16610417274670083, "mr_stft_db": 5.5730202191578, "band_mae_db": 0.1069557758324735, "rms_gain_error_db": -0.11480468460435944, "output_peak": 0.4627043548984992}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59760, "output_samples": 59760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.731050233884645, "spectral_convergence": 0.29574753387520836, "spectral_convergence_status": "ok", "amplitude_floor": 1.6701466153856754e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.418747180659531, "spectral_convergence": 0.2918429147598279, "spectral_convergence_status": "ok", "amplitude_floor": 1.1197689318182202e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.574245701936045, "spectral_convergence": 0.3055780997681949, "spectral_convergence_status": "ok", "amplitude_floor": 1.0416079453540003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0010765709682483164, "output_power": 0.0010686547560830744, "delta_db": -0.03205241133623844, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019015067355649686, "output_power": 0.0018695975043178826, "delta_db": -0.07349748464179803, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009390993860770502, "output_power": 0.0009199315616566686, "delta_db": -0.08956037440367942, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028731828855173817, "output_power": 0.00026642628919785094, "delta_db": -0.327861948959925, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.911907847527107e-05, "output_power": 5.865081505243641e-05, "delta_db": -0.034536040784773445, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.5637714608274427e-05, "output_power": 6.0924939751067086e-05, "delta_db": 0.39425825907451384, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15862775320015468, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0761416551763979, "pitch": null, "output_rms_dbfs": -23.722056052522568, "output_peak": 0.5002645106449802, "reference_rms_dbfs": -23.64591439734617, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.574681038826739}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00019933639001390812, "log_rms_mae_db": 1.0200142781986736, "rms_envelope_nrmse": 0.0938958735602321, "envelope_correlation": 0.992238786560505, "rms_gain_error_db": -0.07614166368270814, "reference_mean_power": 0.004325042032530898, "output_mean_power": 0.004249875094035413}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 2.983288998803909e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014406782878671795, "output_power": 0.001375343184893879, "reference_active": true, "delta_db": -0.2015593129599247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010200208961765019, "output_power": 0.00010345719725726474, "reference_active": true, "delta_db": 0.061516400066968964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7770617096993514e-05, "output_power": 3.6067730836648034e-05, "reference_active": true, "delta_db": 1.1353328958653177}], "active_band_count": 3, "active_band_mae_db": 0.4661362029640705}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00018454723874230657, "log_rms_mae_db": 0.5175589774192852, "rms_envelope_nrmse": 0.05268733243164517, "envelope_correlation": 0.9974249024564145, "rms_gain_error_db": -0.07614170050683976, "reference_mean_power": 0.004360109839005934, "output_mean_power": 0.004284333404290724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 2.740771226026708e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013018843976497208, "output_power": 0.0012505087629412926, "reference_active": true, "delta_db": -0.17485682860757326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649956473480408e-05, "output_power": 4.8682747043482934e-05, "reference_active": true, "delta_db": 0.19926188742228257}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9034855452992526e-06, "output_power": 2.3807301302059922e-06, "reference_active": true, "delta_db": 0.971605851406604}], "active_band_count": 3, "active_band_mae_db": 0.44857485581215323}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00017949843182759416, "log_rms_mae_db": 1.5539489721763138, "rms_envelope_nrmse": 0.14089527372541039, "envelope_correlation": 0.9860089099892293, "rms_gain_error_db": -0.0744736159708041, "reference_mean_power": 0.001722884360599687, "output_mean_power": 0.001693591897009151}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.0185329105758934e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006952905258544651, "output_power": 0.0006581272071364427, "reference_active": true, "delta_db": -0.23856466647287655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.45148729518806e-05, "output_power": 5.846378807941786e-05, "reference_active": true, "delta_db": -0.42772894924447386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5243024554692387e-05, "output_power": 2.7257774825074158e-05, "reference_active": true, "delta_db": 2.5241925030402825}], "active_band_count": 3, "active_band_mae_db": 1.063495372919211}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00016126256915979383, "log_rms_mae_db": 0.9013202432752945, "rms_envelope_nrmse": 0.08000164124245146, "envelope_correlation": 0.9952534987704859, "rms_gain_error_db": -0.07447362070469771, "reference_mean_power": 0.0017368536839123, "output_mean_power": 0.001707323712158017}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 9.447738359038237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006450702777726167, "output_power": 0.0006199940553331733, "reference_active": true, "delta_db": -0.17219506435213391}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9818892488309525e-05, "output_power": 2.6548956627325404e-05, "reference_active": true, "delta_db": -0.5044405116434224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0738004145337803e-06, "output_power": 1.7259098537487438e-06, "reference_active": true, "delta_db": 2.0609454098879074}], "active_band_count": 3, "active_band_mae_db": 0.9125269952944879}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 6.626476450042254e-05, "log_rms_mae_db": 1.804161698500498, "rms_envelope_nrmse": 0.21286396249015543, "envelope_correlation": 0.9644744835387802, "rms_gain_error_db": -0.21976118848390444, "reference_mean_power": 0.00025607969857314127, "output_mean_power": 0.00024344397524435707}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.9689769875456006e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013521942671011384, "output_power": 0.00011210582171440927, "reference_active": true, "delta_db": -0.8141092400348102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.463778228624424e-05, "output_power": 1.2659125223641156e-05, "reference_active": true, "delta_db": -0.630715875171205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5896173962011905e-06, "output_power": 4.9984270357307355e-06, "reference_active": true, "delta_db": -0.48548724974480567}], "active_band_count": 3, "active_band_mae_db": 0.6434374549836069}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 5.564377890100535e-05, "log_rms_mae_db": 1.2259232130481463, "rms_envelope_nrmse": 0.14660296895064276, "envelope_correlation": 0.982004213280334, "rms_gain_error_db": -0.21976118679543657, "reference_mean_power": 0.0002581560174430691, "output_mean_power": 0.0002454178425474673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 1.774101433771277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012602347588101508, "output_power": 0.00010418011976562204, "reference_active": true, "delta_db": -0.8266660145271307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.970334241514461e-06, "output_power": 6.01830562575434e-06, "reference_active": true, "delta_db": -0.6377936528273955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.812565026245218e-07, "output_power": 3.1256507789426783e-07, "reference_active": true, "delta_db": -0.862768058768788}], "active_band_count": 3, "active_band_mae_db": 0.775742575374438}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 3.541043489510122e-05, "log_rms_mae_db": 2.1901931581382206, "rms_envelope_nrmse": 0.22698983173263232, "envelope_correlation": 0.9679049387167982, "rms_gain_error_db": 0.2800779855577838, "reference_mean_power": 6.312530293815556e-05, "output_mean_power": 6.733041229579158e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 4.34877201852775e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2352362117577493e-05, "output_power": 2.3459469128847887e-05, "reference_active": true, "delta_db": 0.20994755535907184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.837592049886246e-06, "output_power": 9.613466809427065e-06, "reference_active": true, "delta_db": 0.36546080918746865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1202466049559064e-06, "output_power": 2.3396666696909823e-06, "reference_active": true, "delta_db": 0.4276761178603383}], "active_band_count": 3, "active_band_mae_db": 0.33436149413562627}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 3.3704303759041684e-05, "log_rms_mae_db": 1.888653276506732, "rms_envelope_nrmse": 0.15723924848695994, "envelope_correlation": 0.9840874705910243, "rms_gain_error_db": 0.2800780163056841, "reference_mean_power": 6.363712790649308e-05, "output_mean_power": 6.787633310524632e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 3.9430492064624054e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.36389645521954e-05, "output_power": 2.48916288066446e-05, "reference_active": true, "delta_db": 0.22424866593682247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.125341563464706e-06, "output_power": 4.359322915637046e-06, "reference_active": true, "delta_db": 0.2395912805475645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9054004218332769e-07, "output_power": 2.0981436214650599e-07, "reference_active": true, "delta_db": 0.41848956034600193}], "active_band_count": 3, "active_band_mae_db": 0.29410983561012966}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.853330125066422, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ad5306c22ff31ac2c4f1f10cf876cce748863e17c5fc3b8bb42fe0bfc75df9d2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9651418728372646, "standard_deviation": 0.004423792637035427, "minimum": 0.9573223801599843, "maximum": 0.9715913682684754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8058056208532147, "standard_deviation": 0.029213964547607725, "minimum": 1.767531063810179, "maximum": 1.860834443578954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6337890702679911, "standard_deviation": 0.1359568395048533, "minimum": 0.4062237980610002, "maximum": 0.8296063997124792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21122825011458285, "standard_deviation": 0.013366833797623316, "minimum": 0.191420115769772, "maximum": 0.23353245525535368, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9854820150565011, "standard_deviation": 0.0022574327467991236, "minimum": 0.982004213280334, "maximum": 0.9887157761001473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2221621018602105, "standard_deviation": 0.04216716644077607, "minimum": 1.1524009351012252, "maximum": 1.2905359782084704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6889679691413595, "standard_deviation": 0.20337577552347055, "minimum": 0.4593061974794915, "maximum": 1.0108939952736438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13264590559849873, "standard_deviation": 0.010758723804351838, "minimum": 0.1155601069154866, "maximum": 0.14660296895064276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9656705311604085, "standard_deviation": 0.0029897253752055944, "minimum": 0.9601087568538509, "maximum": 0.9697062534191286, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1918048046556327, "standard_deviation": 0.04088847355902431, "minimum": 2.1501773909718325, "maximum": 2.2453909362089344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.33519119129341524, "standard_deviation": 0.1700819124724208, "minimum": 0.11275214225448761, "maximum": 0.5484438346517617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2293153084875889, "standard_deviation": 0.009220209992744685, "minimum": 0.2160294156336995, "maximum": 0.24582909342136816, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9836813729460087, "standard_deviation": 0.0006932791328788137, "minimum": 0.9825523727292618, "maximum": 0.9848688183672759, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8861500046581694, "standard_deviation": 0.034197495009285823, "minimum": 1.8402068898936674, "maximum": 1.9514370846681535, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.32963294864628256, "standard_deviation": 0.15673156808370137, "minimum": 0.13599656453151932, "maximum": 0.5478132832950832, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15467775368332223, "standard_deviation": 0.0029777160014806112, "minimum": 0.14956811574697682, "maximum": 0.15872363159322134, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.984112953411982, "standard_deviation": 0.003238384081405456, "minimum": 0.9767928272627854, "maximum": 0.987368354655864, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6413965408890503, "standard_deviation": 0.05418009115947354, "minimum": 1.5539489721763138, "maximum": 1.7441231337662417, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6792775082955664, "standard_deviation": 0.46831221151183444, "minimum": 0.27263325456121357, "maximum": 1.4960816917203907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15057849953539762, "standard_deviation": 0.01436295499003755, "minimum": 0.1339962973389865, "maximum": 0.18185221400697468, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946317210053933, "standard_deviation": 0.001444019535000684, "minimum": 0.9912300150398424, "maximum": 0.9960051460076664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9943859287258041, "standard_deviation": 0.05153906435189134, "minimum": 0.9013202432752945, "maximum": 1.0726319339804142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7068817683278604, "standard_deviation": 0.3180307316614062, "minimum": 0.2165273137521834, "maximum": 1.0568826348154177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08846068843587095, "standard_deviation": 0.011297427552896951, "minimum": 0.07454592725665214, "maximum": 0.1125632294345015, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15308880666705496, "standard_deviation": 0.024339188310285472, "minimum": 0.1069557758324735, "maximum": 0.17829869354347003, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.573935456974391, "standard_deviation": 0.0042734544774948304, "minimum": 5.5670995788306845, "maximum": 5.579725924190872, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47987588884928334, "standard_deviation": 0.01729401096177916, "minimum": 0.4627043548984992, "maximum": 0.5092480994824069, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11559648923999635, "standard_deviation": 0.035352724195435926, "minimum": -0.17438495442134969, "maximum": -0.0761416551763979, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920001969713752, "standard_deviation": 0.0009315432580009357, "minimum": 0.9906396222738161, "maximum": 0.9936917666645015, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0129796609831563, "standard_deviation": 0.03810442744172943, "minimum": 0.9647509300341365, "maximum": 1.071729450597361, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2978875044101976, "standard_deviation": 0.14301667012752645, "minimum": 0.08350203786272921, "maximum": 0.46675222035994296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09556404726603447, "standard_deviation": 0.005676454341448609, "minimum": 0.08493328497853286, "maximum": 0.10320669817784571, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997272032739458, "standard_deviation": 0.00044211890004606223, "minimum": 0.9966696466394253, "maximum": 0.9980136957221474, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5508883311931325, "standard_deviation": 0.03357806799704523, "minimum": 0.4975748852289342, "maximum": 0.5847912185486261, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2357117502496123, "standard_deviation": 0.13285680034492875, "minimum": 0.07626769782739302, "maximum": 0.44857485581215323, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05500808567669409, "standard_deviation": 0.004716131693076308, "minimum": 0.04694580323377159, "maximum": 0.06213991805963035, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123286-0008", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1705446243286133, "predicted_mos": 2.1705446243286133, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "60c01b407b52cf28eab8effeb89d092378b6089343e5ec61b4d473758c96079d", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.3368315999396145, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.01199272398416, "全帯域_10ms_rms_envelope_nrmse": 0.0967235858429101, "全帯域_10ms_envelope_correlation": 0.9917531030188541, "全帯域_10ms_modulation_mae_db": 0.47707500704508443, "全帯域_40ms_log_rms_mae_db": 0.5350302044775912, "全帯域_40ms_rms_envelope_nrmse": 0.05500787372185427, "全帯域_40ms_envelope_correlation": 0.9971691117469511, "全帯域_40ms_modulation_mae_db": 0.4672971663795369, "300–1000 Hz_10ms_log_rms_mae_db": 1.5527515132491876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1498220121244794, "300–1000 Hz_10ms_envelope_correlation": 0.9842262250375421, "300–1000 Hz_10ms_modulation_mae_db": 1.2254481321004806, "300–1000 Hz_40ms_log_rms_mae_db": 0.8875993701776562, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08470350914822566, "300–1000 Hz_40ms_envelope_correlation": 0.994658516861379, "300–1000 Hz_40ms_modulation_mae_db": 1.0427724566939018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7951827414383148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19303746025022261, "1000–3000 Hz_10ms_envelope_correlation": 0.9714090555284377, "1000–3000 Hz_10ms_modulation_mae_db": 0.8266415013546925, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2130893764708175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13417932720364914, "1000–3000 Hz_40ms_envelope_correlation": 0.9859317670238393, "1000–3000 Hz_40ms_modulation_mae_db": 0.9807435749102428, "3000–6000 Hz_10ms_log_rms_mae_db": 2.169063775249636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22597778122538154, "3000–6000 Hz_10ms_envelope_correlation": 0.9681797708523037, "3000–6000 Hz_10ms_modulation_mae_db": 0.3010412348553928, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8921755606805748, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1597684653300656, "3000–6000 Hz_40ms_envelope_correlation": 0.9833946728246192, "3000–6000 Hz_40ms_modulation_mae_db": 0.23284602867568463, "mr_stft_db": 5.546344429439827, "band_mae_db": 0.2206965257770428, "rms_gain_error_db": -0.06534622418967753, "output_peak": 0.5076762635588578}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0634091088119664, "全帯域_10ms_rms_envelope_nrmse": 0.10147701314306003, "全帯域_10ms_envelope_correlation": 0.9910278337126426, "全帯域_10ms_modulation_mae_db": 0.5577589260563379, "全帯域_40ms_log_rms_mae_db": 0.5771211730706617, "全帯域_40ms_rms_envelope_nrmse": 0.05747489120971368, "全帯域_40ms_envelope_correlation": 0.997095671609402, "全帯域_40ms_modulation_mae_db": 0.36265441882001886, "300–1000 Hz_10ms_log_rms_mae_db": 1.7477783883259232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16152360664281212, "300–1000 Hz_10ms_envelope_correlation": 0.9816717555037761, "300–1000 Hz_10ms_modulation_mae_db": 1.391921934058507, "300–1000 Hz_40ms_log_rms_mae_db": 1.0565973456535571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09179405346932501, "300–1000 Hz_40ms_envelope_correlation": 0.9941436675160765, "300–1000 Hz_40ms_modulation_mae_db": 1.0550901270571393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.872505475008142, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2085774204223928, "1000–3000 Hz_10ms_envelope_correlation": 0.9664241877189436, "1000–3000 Hz_10ms_modulation_mae_db": 0.7225414649184492, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2031299035953695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1290048733896212, "1000–3000 Hz_40ms_envelope_correlation": 0.9873581979429599, "1000–3000 Hz_40ms_modulation_mae_db": 0.4681696218774925, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2656240729426984, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22642489281005432, "3000–6000 Hz_10ms_envelope_correlation": 0.9660343189804819, "3000–6000 Hz_10ms_modulation_mae_db": 0.14598180880439934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9132052445546648, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15444886215175252, "3000–6000 Hz_40ms_envelope_correlation": 0.9833128385684302, "3000–6000 Hz_40ms_modulation_mae_db": 0.07699986969878828, "mr_stft_db": 5.556320799471642, "band_mae_db": 0.18764676655903137, "rms_gain_error_db": -0.14728043426894882, "output_peak": 0.46399298386054544}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.996781640009741, "全帯域_10ms_rms_envelope_nrmse": 0.09686491116020177, "全帯域_10ms_envelope_correlation": 0.9918734231166854, "全帯域_10ms_modulation_mae_db": 0.2997119454844954, "全帯域_40ms_log_rms_mae_db": 0.5494097836745949, "全帯域_40ms_rms_envelope_nrmse": 0.06272316618280459, "全帯域_40ms_envelope_correlation": 0.9965284943935192, "全帯域_40ms_modulation_mae_db": 0.3152971083961173, "300–1000 Hz_10ms_log_rms_mae_db": 1.6265670201814728, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1510328546408181, "300–1000 Hz_10ms_envelope_correlation": 0.9843305138449636, "300–1000 Hz_10ms_modulation_mae_db": 0.8878471981758363, "300–1000 Hz_40ms_log_rms_mae_db": 1.015104685455069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09573662765700336, "300–1000 Hz_40ms_envelope_correlation": 0.9939628832568824, "300–1000 Hz_40ms_modulation_mae_db": 0.6475391984544864, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7496262520237993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18383436925802368, "1000–3000 Hz_10ms_envelope_correlation": 0.9739381547013802, "1000–3000 Hz_10ms_modulation_mae_db": 0.9267978364760355, "1000–3000 Hz_40ms_log_rms_mae_db": 1.193562246056181, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12093207702499836, "1000–3000 Hz_40ms_envelope_correlation": 0.9884673739691374, "1000–3000 Hz_40ms_modulation_mae_db": 1.0916785174509904, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2523112054891667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24462058930161307, "3000–6000 Hz_10ms_envelope_correlation": 0.9605280101720158, "3000–6000 Hz_10ms_modulation_mae_db": 0.4915100248833231, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9402990024222293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15842394956611283, "3000–6000 Hz_40ms_envelope_correlation": 0.9824624485971313, "3000–6000 Hz_40ms_modulation_mae_db": 0.5191667422883883, "mr_stft_db": 5.551187240172467, "band_mae_db": 0.23011114923550438, "rms_gain_error_db": -0.155315576064204, "output_peak": 0.4712028285420472}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0162764737800163, "全帯域_10ms_rms_envelope_nrmse": 0.09001069717891995, "全帯域_10ms_envelope_correlation": 0.992910273625424, "全帯域_10ms_modulation_mae_db": 0.09506770919156203, "全帯域_40ms_log_rms_mae_db": 0.5912037923529483, "全帯域_40ms_rms_envelope_nrmse": 0.049725552090430065, "全帯域_40ms_envelope_correlation": 0.9977646670521336, "全帯域_40ms_modulation_mae_db": 0.15459922999705425, "300–1000 Hz_10ms_log_rms_mae_db": 1.652033424358841, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13996701962322425, "300–1000 Hz_10ms_envelope_correlation": 0.9866117972077306, "300–1000 Hz_10ms_modulation_mae_db": 0.5136700977689294, "300–1000 Hz_40ms_log_rms_mae_db": 1.04009557284075, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08578659439014276, "300–1000 Hz_40ms_envelope_correlation": 0.9952529443961381, "300–1000 Hz_40ms_modulation_mae_db": 0.8728206870776297, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8079122772125913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19158905424618605, "1000–3000 Hz_10ms_envelope_correlation": 0.971338764255363, "1000–3000 Hz_10ms_modulation_mae_db": 0.4713198243251661, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2018286239168414, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11371806911854654, "1000–3000 Hz_40ms_envelope_correlation": 0.989324306827133, "1000–3000 Hz_40ms_modulation_mae_db": 0.6618630758839193, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2200475855388024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23396241276801188, "3000–6000 Hz_10ms_envelope_correlation": 0.9642471461182118, "3000–6000 Hz_10ms_modulation_mae_db": 0.4137332783855739, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8773334463365208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15515815333322883, "3000–6000 Hz_40ms_envelope_correlation": 0.983351733773038, "3000–6000 Hz_40ms_modulation_mae_db": 0.30486588574107953, "mr_stft_db": 5.54585539252304, "band_mae_db": 0.18912919865240674, "rms_gain_error_db": -0.10109691341205372, "output_peak": 0.4711281282675864}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0679583042492022, "全帯域_10ms_rms_envelope_nrmse": 0.10116900543588475, "全帯域_10ms_envelope_correlation": 0.991079698081881, "全帯域_10ms_modulation_mae_db": 0.20832933510329443, "全帯域_40ms_log_rms_mae_db": 0.5666990954980011, "全帯域_40ms_rms_envelope_nrmse": 0.05444389984979211, "全帯域_40ms_envelope_correlation": 0.9974333096120666, "全帯域_40ms_modulation_mae_db": 0.08978962246241028, "300–1000 Hz_10ms_log_rms_mae_db": 1.6257347218602587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15563704618962476, "300–1000 Hz_10ms_envelope_correlation": 0.9831947107840596, "300–1000 Hz_10ms_modulation_mae_db": 0.3441057006179657, "300–1000 Hz_40ms_log_rms_mae_db": 0.9494852836507371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09044604655506029, "300–1000 Hz_40ms_envelope_correlation": 0.994593975870355, "300–1000 Hz_40ms_modulation_mae_db": 0.21157627147702818, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8282423691906844, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22844298841566935, "1000–3000 Hz_10ms_envelope_correlation": 0.9602817379819304, "1000–3000 Hz_10ms_modulation_mae_db": 1.0652624405352915, "1000–3000 Hz_40ms_log_rms_mae_db": 1.264029937054078, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14545513019543946, "1000–3000 Hz_40ms_envelope_correlation": 0.985005347435483, "1000–3000 Hz_40ms_modulation_mae_db": 1.2531053517555846, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1801808930026136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2277397886756061, "3000–6000 Hz_10ms_envelope_correlation": 0.9655478124095134, "3000–6000 Hz_10ms_modulation_mae_db": 0.4046241018878523, "3000–6000 Hz_40ms_log_rms_mae_db": 1.89226146076157, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1555320051619166, "3000–6000 Hz_40ms_envelope_correlation": 0.9831323181836583, "3000–6000 Hz_40ms_modulation_mae_db": 0.44412471258407127, "mr_stft_db": 5.543487452741908, "band_mae_db": 0.2526275678634991, "rms_gain_error_db": -0.1429745981063958, "output_peak": 0.5070190431471137}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9919831746458982, "全帯域_10ms_rms_envelope_nrmse": 0.08712540776952454, "全帯域_10ms_envelope_correlation": 0.9933658518090793, "全帯域_10ms_modulation_mae_db": 0.28918996019477544, "全帯域_40ms_log_rms_mae_db": 0.5439079603550955, "全帯域_40ms_rms_envelope_nrmse": 0.04798259423945327, "全帯域_40ms_envelope_correlation": 0.997931756553161, "全帯域_40ms_modulation_mae_db": 0.34399501805542215, "300–1000 Hz_10ms_log_rms_mae_db": 1.666515124418138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13358441480550787, "300–1000 Hz_10ms_envelope_correlation": 0.9874193724054555, "300–1000 Hz_10ms_modulation_mae_db": 0.6120078860169698, "300–1000 Hz_40ms_log_rms_mae_db": 0.9908368651959496, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07776034664148485, "300–1000 Hz_40ms_envelope_correlation": 0.9955754756078967, "300–1000 Hz_40ms_modulation_mae_db": 1.0344710778922752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8151602184976507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18833350545056252, "1000–3000 Hz_10ms_envelope_correlation": 0.9734735508513893, "1000–3000 Hz_10ms_modulation_mae_db": 0.8093632470923238, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2698007513146314, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12338812523783542, "1000–3000 Hz_40ms_envelope_correlation": 0.989216830926339, "1000–3000 Hz_40ms_modulation_mae_db": 0.8686228523647902, "3000–6000 Hz_10ms_log_rms_mae_db": 2.168737693373773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23596406620909224, "3000–6000 Hz_10ms_envelope_correlation": 0.9637032125312173, "3000–6000 Hz_10ms_modulation_mae_db": 0.27488572931025224, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8765567225972697, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15668629662166247, "3000–6000 Hz_40ms_envelope_correlation": 0.9830167110401739, "3000–6000 Hz_40ms_modulation_mae_db": 0.2834754137606012, "mr_stft_db": 5.542171479227954, "band_mae_db": 0.2367202689632919, "rms_gain_error_db": -0.09887262428109933, "output_peak": 0.46477853175087724}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9990331128812299, "全帯域_10ms_rms_envelope_nrmse": 0.10345546062728948, "全帯域_10ms_envelope_correlation": 0.9905737925448925, "全帯域_10ms_modulation_mae_db": 0.3881549974875805, "全帯域_40ms_log_rms_mae_db": 0.49973758600714735, "全帯域_40ms_rms_envelope_nrmse": 0.05834709638814984, "全帯域_40ms_envelope_correlation": 0.9968490269137207, "全帯域_40ms_modulation_mae_db": 0.1946386864518305, "300–1000 Hz_10ms_log_rms_mae_db": 1.6644366227320728, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18269273284855123, "300–1000 Hz_10ms_envelope_correlation": 0.9764384566492282, "300–1000 Hz_10ms_modulation_mae_db": 0.26896580597230463, "300–1000 Hz_40ms_log_rms_mae_db": 0.9871691974312421, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11134849081040336, "300–1000 Hz_40ms_envelope_correlation": 0.9912030053125128, "300–1000 Hz_40ms_modulation_mae_db": 0.17571249922745755, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7900768655155592, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19898059760322118, "1000–3000 Hz_10ms_envelope_correlation": 0.9690390704597156, "1000–3000 Hz_10ms_modulation_mae_db": 0.22689910283841544, "1000–3000 Hz_40ms_log_rms_mae_db": 1.237865063558277, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12018325550433202, "1000–3000 Hz_40ms_envelope_correlation": 0.9879062569760717, "1000–3000 Hz_40ms_modulation_mae_db": 0.16724015436390527, "3000–6000 Hz_10ms_log_rms_mae_db": 2.17665205667435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22888788586004796, "3000–6000 Hz_10ms_envelope_correlation": 0.9659325824041851, "3000–6000 Hz_10ms_modulation_mae_db": 0.24798832225870773, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8555375762498714, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1555846096388872, "3000–6000 Hz_40ms_envelope_correlation": 0.9832679599022494, "3000–6000 Hz_40ms_modulation_mae_db": 0.1951180415033242, "mr_stft_db": 5.539352371743405, "band_mae_db": 0.17351727688238486, "rms_gain_error_db": -0.09150597049423183, "output_peak": 0.47089301850286214}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9974135966592675, "全帯域_10ms_rms_envelope_nrmse": 0.095292603711208, "全帯域_10ms_envelope_correlation": 0.9920347134099692, "全帯域_10ms_modulation_mae_db": 0.11986224808034661, "全帯域_40ms_log_rms_mae_db": 0.5701572833983126, "全帯域_40ms_rms_envelope_nrmse": 0.057587090421410705, "全帯域_40ms_envelope_correlation": 0.9969741224820985, "全帯域_40ms_modulation_mae_db": 0.11780416779453419, "300–1000 Hz_10ms_log_rms_mae_db": 1.6688957997034113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15349602738373377, "300–1000 Hz_10ms_envelope_correlation": 0.983367531254648, "300–1000 Hz_10ms_modulation_mae_db": 0.3179519541075586, "300–1000 Hz_40ms_log_rms_mae_db": 1.0138622406517552, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08877732462439296, "300–1000 Hz_40ms_envelope_correlation": 0.9942928908487384, "300–1000 Hz_40ms_modulation_mae_db": 0.5616217246401618, "1000–3000 Hz_10ms_log_rms_mae_db": 1.781487065089658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18920934224651265, "1000–3000 Hz_10ms_envelope_correlation": 0.9720458471684282, "1000–3000 Hz_10ms_modulation_mae_db": 0.12664269934118974, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1617122447903792, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11012217364189784, "1000–3000 Hz_40ms_envelope_correlation": 0.989859329177038, "1000–3000 Hz_40ms_modulation_mae_db": 0.132660179923948, "3000–6000 Hz_10ms_log_rms_mae_db": 2.159399110801793, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21689373213853608, "3000–6000 Hz_10ms_envelope_correlation": 0.9691426185508867, "3000–6000 Hz_10ms_modulation_mae_db": 0.07637449576574336, "3000–6000 Hz_40ms_log_rms_mae_db": 1.826707379693776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14560116869680587, "3000–6000 Hz_40ms_envelope_correlation": 0.9853613523050038, "3000–6000 Hz_40ms_modulation_mae_db": 0.12023223723252684, "mr_stft_db": 5.546109724019186, "band_mae_db": 0.17997700544102493, "rms_gain_error_db": -0.10485268387076524, "output_peak": 0.46748255785522164}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59760, "output_samples": 59760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.703158995979921, "spectral_convergence": 0.2989939867153913, "spectral_convergence_status": "ok", "amplitude_floor": 1.6701466153856754e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.401423356489642, "spectral_convergence": 0.2936207685463451, "spectral_convergence_status": "ok", "amplitude_floor": 1.1197689318182202e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.534450935849917, "spectral_convergence": 0.31040080829581934, "spectral_convergence_status": "ok", "amplitude_floor": 1.0416079453540003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0010765709682483164, "output_power": 0.0010710328828989998, "delta_db": -0.02239859308838732, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019015067355649686, "output_power": 0.0018740801507549656, "delta_db": -0.06309707241767913, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009390993860770502, "output_power": 0.0009352994994503007, "delta_db": -0.017608547117101147, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028731828855173817, "output_power": 0.00025735613278664716, "delta_db": -0.47828748908010316, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.911907847527107e-05, "output_power": 5.526648052164358e-05, "delta_db": -0.2926584709535878, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.5637714608274427e-05, "output_power": 6.17137846453358e-05, "delta_db": 0.4501289820053981, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2206965257770428, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06534622418967753, "pitch": null, "output_rms_dbfs": -23.71126062153585, "output_peak": 0.5076762635588578, "reference_rms_dbfs": -23.64591439734617, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.546344429439827}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00019933639001390812, "log_rms_mae_db": 1.01199272398416, "rms_envelope_nrmse": 0.0967235858429101, "envelope_correlation": 0.9917531030188541, "rms_gain_error_db": -0.06534622792887425, "reference_mean_power": 0.004325042032530898, "output_mean_power": 0.00426045232324273}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 2.983288998803909e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014406782878671795, "output_power": 0.00137723192424874, "reference_active": true, "delta_db": -0.1955992996939368}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010200208961765019, "output_power": 0.00010367408589869603, "reference_active": true, "delta_db": 0.0706114602772459}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7770617096993514e-05, "output_power": 3.6315075928051905e-05, "reference_active": true, "delta_db": 1.1650142611640706}], "active_band_count": 3, "active_band_mae_db": 0.47707500704508443}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00018454723874230657, "log_rms_mae_db": 0.5350302044775912, "rms_envelope_nrmse": 0.05500787372185427, "envelope_correlation": 0.9971691117469511, "rms_gain_error_db": -0.06534626160339459, "reference_mean_power": 0.004360109839005934, "output_mean_power": 0.004294996397591166}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 2.740771226026708e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013018843976497208, "output_power": 0.0012540786714435284, "reference_active": true, "delta_db": -0.16247640545228126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649956473480408e-05, "output_power": 4.868345348337739e-05, "reference_active": true, "delta_db": 0.19932490784339466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9034855452992526e-06, "output_power": 2.4185696671319057e-06, "reference_active": true, "delta_db": 1.0400901858429348}], "active_band_count": 3, "active_band_mae_db": 0.4672971663795369}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00017949843182759416, "log_rms_mae_db": 1.5527515132491876, "rms_envelope_nrmse": 0.1498220121244794, "envelope_correlation": 0.9842262250375421, "rms_gain_error_db": -0.025426549646940202, "reference_mean_power": 0.001722884360599687, "output_mean_power": 0.0017128268953394705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.0185329105758934e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006952905258544651, "output_power": 0.0006637182086085087, "reference_active": true, "delta_db": -0.20182579309282783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.45148729518806e-05, "output_power": 5.652643884921986e-05, "reference_active": true, "delta_db": -0.5740822047755839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5243024554692387e-05, "output_power": 2.9724515813587002e-05, "reference_active": true, "delta_db": 2.90043639843303}], "active_band_count": 3, "active_band_mae_db": 1.2254481321004806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00016126256915979383, "log_rms_mae_db": 0.8875993701776562, "rms_envelope_nrmse": 0.08470350914822566, "envelope_correlation": 0.994658516861379, "rms_gain_error_db": -0.025426550095504732, "reference_mean_power": 0.0017368536839123, "output_mean_power": 0.001726714671512643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 9.447738359038237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006450702777726167, "output_power": 0.0006258606114059781, "reference_active": true, "delta_db": -0.13129411772225014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9818892488309525e-05, "output_power": 2.5598220142096573e-05, "reference_active": true, "delta_db": -0.6628173953110006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0738004145337803e-06, "output_power": 1.837994269416128e-06, "reference_active": true, "delta_db": 2.334205857048455}], "active_band_count": 3, "active_band_mae_db": 1.0427724566939018}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 6.626476450042254e-05, "log_rms_mae_db": 1.7951827414383148, "rms_envelope_nrmse": 0.19303746025022261, "envelope_correlation": 0.9714090555284377, "rms_gain_error_db": -0.38080817523300103, "reference_mean_power": 0.00025607969857314127, "output_mean_power": 0.0002345818097280039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.9689769875456006e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013521942671011384, "output_power": 0.00010987645053090409, "reference_active": true, "delta_db": -0.9013446887923131}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.463778228624424e-05, "output_power": 1.1873960749513996e-05, "reference_active": true, "delta_db": -0.9087967443831699}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5896173962011905e-06, "output_power": 4.790752939951899e-06, "reference_active": true, "delta_db": -0.6697830708885946}], "active_band_count": 3, "active_band_mae_db": 0.8266415013546925}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 5.564377890100535e-05, "log_rms_mae_db": 1.2130893764708175, "rms_envelope_nrmse": 0.13417932720364914, "envelope_correlation": 0.9859317670238393, "rms_gain_error_db": -0.38080817263419614, "reference_mean_power": 0.0002581560174430691, "output_mean_power": 0.00023648382178531312}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 1.774101433771277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012602347588101508, "output_power": 0.00010203495346492135, "reference_active": true, "delta_db": -0.9170248306739642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.970334241514461e-06, "output_power": 5.656821122266025e-06, "reference_active": true, "delta_db": -0.9068115801804169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.812565026245218e-07, "output_power": 2.946984341377362e-07, "reference_active": true, "delta_db": -1.118394313876347}], "active_band_count": 3, "active_band_mae_db": 0.9807435749102428}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 3.541043489510122e-05, "log_rms_mae_db": 2.169063775249636, "rms_envelope_nrmse": 0.22597778122538154, "envelope_correlation": 0.9681797708523037, "rms_gain_error_db": 0.27631836395824405, "reference_mean_power": 6.312530293815556e-05, "output_mean_power": 6.72721505986834e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 4.34877201852775e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2352362117577493e-05, "output_power": 2.39579630133403e-05, "reference_active": true, "delta_db": 0.3012646552960035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.837592049886246e-06, "output_power": 9.468563321552413e-06, "reference_active": true, "delta_db": 0.2995013757032409}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1202466049559064e-06, "output_power": 2.273118852212104e-06, "reference_active": true, "delta_db": 0.30235767356693405}], "active_band_count": 3, "active_band_mae_db": 0.3010412348553928}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 3.3704303759041684e-05, "log_rms_mae_db": 1.8921755606805748, "rms_envelope_nrmse": 0.1597684653300656, "envelope_correlation": 0.9833946728246192, "rms_gain_error_db": 0.2763183960779059, "reference_mean_power": 6.363712790649308e-05, "output_mean_power": 6.781759903867708e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 3.9430492064624054e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.36389645521954e-05, "output_power": 2.5336070449698978e-05, "reference_active": true, "delta_db": 0.3011080856461485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.125341563464706e-06, "output_power": 4.222712098065889e-06, "reference_active": true, "delta_db": 0.10131560103594309}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9054004218332769e-07, "output_power": 2.039847483057445e-07, "reference_active": true, "delta_db": 0.2961143993449623}], "active_band_count": 3, "active_band_mae_db": 0.23284602867568463}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8497096350985429, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f37b8ff0d4668637219e85a06a8018ea89d33c5de265ba1ee65965ad1ebb5490", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9697437960831985, "standard_deviation": 0.004522438469077419, "minimum": 0.9602817379819304, "maximum": 0.9739381547013802, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.80502415799705, "standard_deviation": 0.036148685399572444, "minimum": 1.7496262520237993, "maximum": 1.872505475008142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6469335146101954, "standard_deviation": 0.33714974996370906, "minimum": 0.12664269934118974, "maximum": 1.0652624405352915, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19775059223659885, "standard_deviation": 0.01452205320698861, "minimum": 0.18383436925802368, "maximum": 0.22844298841566935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9878836762847502, "standard_deviation": 0.0017099703045986851, "minimum": 0.985005347435483, "maximum": 0.989859329177038, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2181272683445719, "standard_deviation": 0.036778927805379016, "minimum": 1.1617122447903792, "maximum": 1.2698007513146314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7030104160663592, "standard_deviation": 0.41874352622624644, "minimum": 0.132660179923948, "maximum": 1.2531053517555846, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12462287891453999, "standard_deviation": 0.01139268245258376, "minimum": 0.11012217364189784, "maximum": 0.14545513019543946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.965414434002352, "standard_deviation": 0.00268288093157267, "minimum": 0.9605280101720158, "maximum": 0.9691426185508867, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.199002049134104, "standard_deviation": 0.04133082054524382, "minimum": 2.159399110801793, "maximum": 2.2656240729426984, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.2945173745189056, "standard_deviation": 0.14009388292376848, "minimum": 0.07637449576574336, "maximum": 0.4915100248833231, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2300588936235429, "standard_deviation": 0.008212220251238536, "minimum": 0.21689373213853608, "maximum": 0.24462058930161307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9834125043992881, "standard_deviation": 0.0008426665421579329, "minimum": 0.9824624485971313, "maximum": 0.9853613523050038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8842595491620595, "standard_deviation": 0.03453667989456913, "minimum": 1.826707379693776, "maximum": 1.9402990024222293, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.272103616435558, "standard_deviation": 0.15138996687688205, "minimum": 0.07699986969878828, "maximum": 0.5191667422883883, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.155150438812554, "standard_deviation": 0.004249635732853901, "minimum": 0.14560116869680587, "maximum": 0.1597684653300656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9834075453359254, "standard_deviation": 0.0033702936911854083, "minimum": 0.9764384566492282, "maximum": 0.9874193724054555, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.650589076853663, "standard_deviation": 0.05476621096534623, "minimum": 1.5527515132491876, "maximum": 1.7477783883259232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6952398386023191, "standard_deviation": 0.42942326469713427, "minimum": 0.26896580597230463, "maximum": 1.391921934058507, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15346946428234393, "standard_deviation": 0.014738384918477279, "minimum": 0.13358441480550787, "maximum": 0.18269273284855123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942104199587474, "standard_deviation": 0.001331766534063648, "minimum": 0.9912030053125128, "maximum": 0.9955754756078967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9925938201320894, "standard_deviation": 0.05375426417564444, "minimum": 0.8875993701776562, "maximum": 1.0565973456535571, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.70020050531501, "standard_deviation": 0.36265988207207966, "minimum": 0.17571249922745755, "maximum": 1.0550901270571393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09079412416200477, "standard_deviation": 0.00988613118209, "minimum": 0.07776034664148485, "maximum": 0.11134849081040336, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20880321992177325, "standard_deviation": 0.02977564854703632, "minimum": 0.17351727688238486, "maximum": 0.2526275678634991, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.546353611167429, "standard_deviation": 0.005314925671630238, "minimum": 5.539352371743405, "maximum": 5.556320799471642, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47802166943563895, "standard_deviation": 0.018316017141413145, "minimum": 0.46399298386054544, "maximum": 0.5076762635588578, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11340562808592203, "standard_deviation": 0.03163053912712595, "minimum": -0.155315576064204, "maximum": -0.06534622418967753, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918273361649286, "standard_deviation": 0.000953187307426073, "minimum": 0.9905737925448925, "maximum": 0.9933658518090793, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0181060168776852, "standard_deviation": 0.030489867472309622, "minimum": 0.9919831746458982, "maximum": 1.0679583042492022, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30439376608043456, "standard_deviation": 0.16404676530188028, "minimum": 0.09506770919156203, "maximum": 0.5577589260563379, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09651483560862482, "standard_deviation": 0.005689362637358151, "minimum": 0.08712540776952454, "maximum": 0.10345546062728948, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972182700453815, "standard_deviation": 0.0004695043874217862, "minimum": 0.9965284943935192, "maximum": 0.997931756553161, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5541583598542941, "standard_deviation": 0.028705442701291686, "minimum": 0.49973758600714735, "maximum": 0.5912037923529483, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.25575942729461554, "standard_deviation": 0.1352520237168146, "minimum": 0.08978962246241028, "maximum": 0.4672971663795369, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05541152051295107, "standard_deviation": 0.004776232438380952, "minimum": 0.04798259423945327, "maximum": 0.06272316618280459, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123286-0008", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.052403688430786, "predicted_mos": 2.052403688430786, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "00844d6faa6e0b4e87ea50b1c3decf47fbb277c58058befbcda885f2c40cf0be", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.828179499600083, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9710375033726912, "全帯域_10ms_rms_envelope_nrmse": 0.09321348132722815, "全帯域_10ms_envelope_correlation": 0.9924439796171697, "全帯域_10ms_modulation_mae_db": 0.06909209306026332, "全帯域_40ms_log_rms_mae_db": 0.5314423804222205, "全帯域_40ms_rms_envelope_nrmse": 0.05997841369857791, "全帯域_40ms_envelope_correlation": 0.9966470614380761, "全帯域_40ms_modulation_mae_db": 0.19430772037769736, "300–1000 Hz_10ms_log_rms_mae_db": 1.7617628737855953, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15780237161118918, "300–1000 Hz_10ms_envelope_correlation": 0.98265054115332, "300–1000 Hz_10ms_modulation_mae_db": 0.44228171519206433, "300–1000 Hz_40ms_log_rms_mae_db": 1.1082154467011107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08522991078790146, "300–1000 Hz_40ms_envelope_correlation": 0.9946499656581153, "300–1000 Hz_40ms_modulation_mae_db": 0.36539386923022815, "1000–3000 Hz_10ms_log_rms_mae_db": 2.073599748936407, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2588606123474968, "1000–3000 Hz_10ms_envelope_correlation": 0.9495019170892697, "1000–3000 Hz_10ms_modulation_mae_db": 0.4347653711593115, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3639539455929077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16681828730347031, "1000–3000 Hz_40ms_envelope_correlation": 0.9773259522874471, "1000–3000 Hz_40ms_modulation_mae_db": 0.6233903674913885, "3000–6000 Hz_10ms_log_rms_mae_db": 2.667607280108835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25280896172843004, "3000–6000 Hz_10ms_envelope_correlation": 0.9582937840606323, "3000–6000 Hz_10ms_modulation_mae_db": 0.5751506357144281, "3000–6000 Hz_40ms_log_rms_mae_db": 2.113884224327628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15747694341869314, "3000–6000 Hz_40ms_envelope_correlation": 0.9831967249842828, "3000–6000 Hz_40ms_modulation_mae_db": 0.4648573194896774, "mr_stft_db": 5.857795746966336, "band_mae_db": 0.1201365858829808, "rms_gain_error_db": 0.04105258925741037, "output_peak": 0.5202910545021306}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9926203744347467, "全帯域_10ms_rms_envelope_nrmse": 0.08677580651899408, "全帯域_10ms_envelope_correlation": 0.9935853035877129, "全帯域_10ms_modulation_mae_db": 0.565066106956482, "全帯域_40ms_log_rms_mae_db": 0.5439048472392835, "全帯域_40ms_rms_envelope_nrmse": 0.051334609752289595, "全帯域_40ms_envelope_correlation": 0.9978405362873708, "全帯域_40ms_modulation_mae_db": 0.44929251454248864, "300–1000 Hz_10ms_log_rms_mae_db": 1.814981488633744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16186880251298963, "300–1000 Hz_10ms_envelope_correlation": 0.9824746894801938, "300–1000 Hz_10ms_modulation_mae_db": 0.9468335563987544, "300–1000 Hz_40ms_log_rms_mae_db": 1.072859281704113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09582789402198222, "300–1000 Hz_40ms_envelope_correlation": 0.9948608582779401, "300–1000 Hz_40ms_modulation_mae_db": 0.34323469228110853, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1284311445218367, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2095503528983826, "1000–3000 Hz_10ms_envelope_correlation": 0.9659185981603494, "1000–3000 Hz_10ms_modulation_mae_db": 0.7112314324365547, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3755901419212266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12290821733713195, "1000–3000 Hz_40ms_envelope_correlation": 0.9872875080319539, "1000–3000 Hz_40ms_modulation_mae_db": 1.0698810358281365, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7252790497858728, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2436717131697765, "3000–6000 Hz_10ms_envelope_correlation": 0.9616888386935037, "3000–6000 Hz_10ms_modulation_mae_db": 0.5584319821855763, "3000–6000 Hz_40ms_log_rms_mae_db": 2.181632618289073, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15670266604295202, "3000–6000 Hz_40ms_envelope_correlation": 0.9836226745243326, "3000–6000 Hz_40ms_modulation_mae_db": 0.541519640642282, "mr_stft_db": 5.856648648874001, "band_mae_db": 0.1468327113867741, "rms_gain_error_db": -0.1688533637412359, "output_peak": 0.5114813102209846}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0678677288510843, "全帯域_10ms_rms_envelope_nrmse": 0.1064688131968296, "全帯域_10ms_envelope_correlation": 0.9901657139607689, "全帯域_10ms_modulation_mae_db": 0.19629697403971424, "全帯域_40ms_log_rms_mae_db": 0.6171757793943919, "全帯域_40ms_rms_envelope_nrmse": 0.059150991656215755, "全帯域_40ms_envelope_correlation": 0.9969934806984413, "全帯域_40ms_modulation_mae_db": 0.2550176852258699, "300–1000 Hz_10ms_log_rms_mae_db": 1.6991478851451212, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15731000266592954, "300–1000 Hz_10ms_envelope_correlation": 0.9832459198032176, "300–1000 Hz_10ms_modulation_mae_db": 0.6302180381199972, "300–1000 Hz_40ms_log_rms_mae_db": 1.014727786816806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09635131779325264, "300–1000 Hz_40ms_envelope_correlation": 0.9943340302730935, "300–1000 Hz_40ms_modulation_mae_db": 0.4102756749071399, "1000–3000 Hz_10ms_log_rms_mae_db": 2.130117266251843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23028423236369736, "1000–3000 Hz_10ms_envelope_correlation": 0.9585034492595867, "1000–3000 Hz_10ms_modulation_mae_db": 0.44886953759870024, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3624115166707131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.144737894908657, "1000–3000 Hz_40ms_envelope_correlation": 0.982195155709108, "1000–3000 Hz_40ms_modulation_mae_db": 0.6979228383074089, "3000–6000 Hz_10ms_log_rms_mae_db": 2.784204972936162, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25417511871062065, "3000–6000 Hz_10ms_envelope_correlation": 0.9586996822496515, "3000–6000 Hz_10ms_modulation_mae_db": 0.250575042814178, "3000–6000 Hz_40ms_log_rms_mae_db": 2.19432502253119, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16474910743229232, "3000–6000 Hz_40ms_envelope_correlation": 0.9820267639349122, "3000–6000 Hz_40ms_modulation_mae_db": 0.23915162733740616, "mr_stft_db": 5.856345974738865, "band_mae_db": 0.15605666183321057, "rms_gain_error_db": -0.17192691155990852, "output_peak": 0.4612507690443047}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9764683975794402, "全帯域_10ms_rms_envelope_nrmse": 0.08507683016344114, "全帯域_10ms_envelope_correlation": 0.993664468161521, "全帯域_10ms_modulation_mae_db": 0.1851176473039657, "全帯域_40ms_log_rms_mae_db": 0.4951790293023337, "全帯域_40ms_rms_envelope_nrmse": 0.04288349609395776, "全帯域_40ms_envelope_correlation": 0.9983464045923321, "全帯域_40ms_modulation_mae_db": 0.1237824885923939, "300–1000 Hz_10ms_log_rms_mae_db": 1.7816221739461477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14661487792076527, "300–1000 Hz_10ms_envelope_correlation": 0.9849140374506206, "300–1000 Hz_10ms_modulation_mae_db": 0.20549022839649878, "300–1000 Hz_40ms_log_rms_mae_db": 0.9696054302295524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07496402493690224, "300–1000 Hz_40ms_envelope_correlation": 0.9960544435064191, "300–1000 Hz_40ms_modulation_mae_db": 0.41712365013925484, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0703415433089956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23276833024526614, "1000–3000 Hz_10ms_envelope_correlation": 0.9573916859419634, "1000–3000 Hz_10ms_modulation_mae_db": 0.4043314914900897, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3417077582613326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14600982482580938, "1000–3000 Hz_40ms_envelope_correlation": 0.982083688033648, "1000–3000 Hz_40ms_modulation_mae_db": 0.6160488380406465, "3000–6000 Hz_10ms_log_rms_mae_db": 2.718715151723155, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2621798919114851, "3000–6000 Hz_10ms_envelope_correlation": 0.9546243859683966, "3000–6000 Hz_10ms_modulation_mae_db": 0.7323208440853278, "3000–6000 Hz_40ms_log_rms_mae_db": 2.149524571224983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17227366278839434, "3000–6000 Hz_40ms_envelope_correlation": 0.9796151888370175, "3000–6000 Hz_40ms_modulation_mae_db": 0.6932863593324395, "mr_stft_db": 5.858534041425144, "band_mae_db": 0.13486370350533725, "rms_gain_error_db": -0.09075753096506353, "output_peak": 0.5112879848253031}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0226968307216973, "全帯域_10ms_rms_envelope_nrmse": 0.09430101019367221, "全帯域_10ms_envelope_correlation": 0.9925017661709268, "全帯域_10ms_modulation_mae_db": 0.20644096949019186, "全帯域_40ms_log_rms_mae_db": 0.5561324373658304, "全帯域_40ms_rms_envelope_nrmse": 0.05493540488970213, "全帯域_40ms_envelope_correlation": 0.9976370973451926, "全帯域_40ms_modulation_mae_db": 0.20851386925634982, "300–1000 Hz_10ms_log_rms_mae_db": 1.7776436772815565, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.153091575473427, "300–1000 Hz_10ms_envelope_correlation": 0.9843984366581557, "300–1000 Hz_10ms_modulation_mae_db": 0.8194887013762138, "300–1000 Hz_40ms_log_rms_mae_db": 1.0392256035368626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09405734960349686, "300–1000 Hz_40ms_envelope_correlation": 0.9949065960474662, "300–1000 Hz_40ms_modulation_mae_db": 0.5842150306387296, "1000–3000 Hz_10ms_log_rms_mae_db": 2.051028771662012, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2333375758296429, "1000–3000 Hz_10ms_envelope_correlation": 0.9573643653478805, "1000–3000 Hz_10ms_modulation_mae_db": 0.5555793112923259, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3393989851348578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14645332529027646, "1000–3000 Hz_40ms_envelope_correlation": 0.981867052996502, "1000–3000 Hz_40ms_modulation_mae_db": 0.4988757209675068, "3000–6000 Hz_10ms_log_rms_mae_db": 2.796209116213005, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25057085987379196, "3000–6000 Hz_10ms_envelope_correlation": 0.9595616994425926, "3000–6000 Hz_10ms_modulation_mae_db": 0.4994318458552878, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2265131307394035, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16159242856671496, "3000–6000 Hz_40ms_envelope_correlation": 0.9826827129451778, "3000–6000 Hz_40ms_modulation_mae_db": 0.3518252184455262, "mr_stft_db": 5.860631525103249, "band_mae_db": 0.16562721025735008, "rms_gain_error_db": -0.21108919957601452, "output_peak": 0.5649166135915186}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9911480202974577, "全帯域_10ms_rms_envelope_nrmse": 0.09336632041388326, "全帯域_10ms_envelope_correlation": 0.9923252570107759, "全帯域_10ms_modulation_mae_db": 0.497008911765378, "全帯域_40ms_log_rms_mae_db": 0.5374645143019108, "全帯域_40ms_rms_envelope_nrmse": 0.04903512145804313, "全帯域_40ms_envelope_correlation": 0.9977242500295567, "全帯域_40ms_modulation_mae_db": 0.14647071511168172, "300–1000 Hz_10ms_log_rms_mae_db": 1.6961002602884323, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1539073381272265, "300–1000 Hz_10ms_envelope_correlation": 0.9832882091582743, "300–1000 Hz_10ms_modulation_mae_db": 1.2455022184404398, "300–1000 Hz_40ms_log_rms_mae_db": 1.1304991410120646, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09026755407132481, "300–1000 Hz_40ms_envelope_correlation": 0.9941259151526285, "300–1000 Hz_40ms_modulation_mae_db": 0.8886878686867314, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0768075177177954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23547876967653789, "1000–3000 Hz_10ms_envelope_correlation": 0.9588930675058688, "1000–3000 Hz_10ms_modulation_mae_db": 0.379300016676466, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3675986853773148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14633975703204635, "1000–3000 Hz_40ms_envelope_correlation": 0.9831652348285103, "1000–3000 Hz_40ms_modulation_mae_db": 0.5622726724274366, "3000–6000 Hz_10ms_log_rms_mae_db": 2.718914173536752, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25023078426444323, "3000–6000 Hz_10ms_envelope_correlation": 0.9623394315487835, "3000–6000 Hz_10ms_modulation_mae_db": 0.36117319132943954, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1673630370206185, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16163209598454956, "3000–6000 Hz_40ms_envelope_correlation": 0.9843108819796035, "3000–6000 Hz_40ms_modulation_mae_db": 0.27079017944837414, "mr_stft_db": 5.876557666038523, "band_mae_db": 0.19203329604047456, "rms_gain_error_db": -0.0028366609869471984, "output_peak": 0.5044849303444655}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0691156324392896, "全帯域_10ms_rms_envelope_nrmse": 0.09558485292531793, "全帯域_10ms_envelope_correlation": 0.9920152171074088, "全帯域_10ms_modulation_mae_db": 0.18285359432035356, "全帯域_40ms_log_rms_mae_db": 0.6091840739866737, "全帯域_40ms_rms_envelope_nrmse": 0.052406250416354716, "全帯域_40ms_envelope_correlation": 0.9975538175388676, "全帯域_40ms_modulation_mae_db": 0.11640169681671249, "300–1000 Hz_10ms_log_rms_mae_db": 1.8783197021728906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16806917785007608, "300–1000 Hz_10ms_envelope_correlation": 0.9808692099660011, "300–1000 Hz_10ms_modulation_mae_db": 1.0166160637154487, "300–1000 Hz_40ms_log_rms_mae_db": 1.1653092645662626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09880787609107197, "300–1000 Hz_40ms_envelope_correlation": 0.9942793696811925, "300–1000 Hz_40ms_modulation_mae_db": 0.8877167849015987, "1000–3000 Hz_10ms_log_rms_mae_db": 2.003450209105035, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21325776600829374, "1000–3000 Hz_10ms_envelope_correlation": 0.9654667085575279, "1000–3000 Hz_10ms_modulation_mae_db": 0.49300351567937284, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2871275386718581, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12560015925885232, "1000–3000 Hz_40ms_envelope_correlation": 0.987064126330033, "1000–3000 Hz_40ms_modulation_mae_db": 0.5178228731980699, "3000–6000 Hz_10ms_log_rms_mae_db": 2.800973221175377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2681595253222602, "3000–6000 Hz_10ms_envelope_correlation": 0.9537292488951195, "3000–6000 Hz_10ms_modulation_mae_db": 0.4532108431207898, "3000–6000 Hz_40ms_log_rms_mae_db": 2.243866090307688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17798008019929923, "3000–6000 Hz_40ms_envelope_correlation": 0.978717608673348, "3000–6000 Hz_40ms_modulation_mae_db": 0.523636444318084, "mr_stft_db": 5.857437771104856, "band_mae_db": 0.21697745361751633, "rms_gain_error_db": -0.12234019171273477, "output_peak": 0.5245904817740585}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.011665191014068, "全帯域_10ms_rms_envelope_nrmse": 0.0840982275173259, "全帯域_10ms_envelope_correlation": 0.9937876314062551, "全帯域_10ms_modulation_mae_db": 0.3612125971993918, "全帯域_40ms_log_rms_mae_db": 0.5465984079427716, "全帯域_40ms_rms_envelope_nrmse": 0.04542574474478194, "全帯域_40ms_envelope_correlation": 0.9981075723693251, "全帯域_40ms_modulation_mae_db": 0.5043405003568824, "300–1000 Hz_10ms_log_rms_mae_db": 1.7091873580796135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13221248856994636, "300–1000 Hz_10ms_envelope_correlation": 0.9880035416417209, "300–1000 Hz_10ms_modulation_mae_db": 0.6456845492347214, "300–1000 Hz_40ms_log_rms_mae_db": 0.9909169672945846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07415935091323728, "300–1000 Hz_40ms_envelope_correlation": 0.9964930020119701, "300–1000 Hz_40ms_modulation_mae_db": 0.5766078951571738, "1000–3000 Hz_10ms_log_rms_mae_db": 2.068098218035949, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20639805047754495, "1000–3000 Hz_10ms_envelope_correlation": 0.9667944985282838, "1000–3000 Hz_10ms_modulation_mae_db": 0.3586200940438565, "1000–3000 Hz_40ms_log_rms_mae_db": 1.351523795399346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12775686721400561, "1000–3000 Hz_40ms_envelope_correlation": 0.9863100711976857, "1000–3000 Hz_40ms_modulation_mae_db": 0.39695222599668284, "3000–6000 Hz_10ms_log_rms_mae_db": 2.740412601584262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2608139922677523, "3000–6000 Hz_10ms_envelope_correlation": 0.9556460142044932, "3000–6000 Hz_10ms_modulation_mae_db": 0.5518882630834879, "3000–6000 Hz_40ms_log_rms_mae_db": 2.265997952420175, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1666308790623035, "3000–6000 Hz_40ms_envelope_correlation": 0.9813163605450224, "3000–6000 Hz_40ms_modulation_mae_db": 0.4151382867193389, "mr_stft_db": 5.875599968506073, "band_mae_db": 0.15761099554873945, "rms_gain_error_db": -0.07708743542006284, "output_peak": 0.5484656772745992}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59760, "output_samples": 59760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.091988329755992, "spectral_convergence": 0.30486380453062023, "spectral_convergence_status": "ok", "amplitude_floor": 1.6701466153856754e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.756723690976561, "spectral_convergence": 0.3120062884926032, "spectral_convergence_status": "ok", "amplitude_floor": 1.1197689318182202e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.724675220166456, "spectral_convergence": 0.3190542118746817, "spectral_convergence_status": "ok", "amplitude_floor": 1.0416079453540003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0010765709682483164, "output_power": 0.0010759725731483935, "delta_db": -0.002414628940982377, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019015067355649686, "output_power": 0.001992170542397389, "delta_db": 0.20228645914195012, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009390993860770502, "output_power": 0.0008888670930499812, "delta_db": -0.2387472825035955, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028731828855173817, "output_power": 0.0002851527416262179, "delta_db": -0.0328571933429569, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.911907847527107e-05, "output_power": 6.185307221916005e-05, "delta_db": 0.19633619917209158, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.5637714608274427e-05, "output_power": 5.625835869101257e-05, "delta_db": 0.04817775219630842, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1201365858829808, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.04105258925741037, "pitch": null, "output_rms_dbfs": -23.604861808088764, "output_peak": 0.5202910545021306, "reference_rms_dbfs": -23.64591439734617, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.857795746966336}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00019933639001390812, "log_rms_mae_db": 0.9710375033726912, "rms_envelope_nrmse": 0.09321348132722815, "envelope_correlation": 0.9924439796171697, "rms_gain_error_db": 0.04105259173264993, "reference_mean_power": 0.004325042032530898, "output_mean_power": 0.0043661192339810575}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 2.983288998803909e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014406782878671795, "output_power": 0.0014133581576307069, "reference_active": true, "delta_db": -0.08314781157522376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010200208961765019, "output_power": 0.00010288383725438863, "reference_active": true, "delta_db": 0.037380849243405734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7770617096993514e-05, "output_power": 2.833089485210129e-05, "reference_active": true, "delta_db": 0.08674761836216047}], "active_band_count": 3, "active_band_mae_db": 0.06909209306026332}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00018454723874230657, "log_rms_mae_db": 0.5314423804222205, "rms_envelope_nrmse": 0.05997841369857791, "envelope_correlation": 0.9966470614380761, "rms_gain_error_db": 0.04105259396815848, "reference_mean_power": 0.004360109839005934, "output_mean_power": 0.004401520100144121}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 2.740771226026708e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013018843976497208, "output_power": 0.0012738018513012378, "reference_active": true, "delta_db": -0.09470546517663977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649956473480408e-05, "output_power": 4.901821909757108e-05, "reference_active": true, "delta_db": 0.22908641020019987}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9034855452992526e-06, "output_power": 2.020517944337207e-06, "reference_active": true, "delta_db": 0.2591312857562525}], "active_band_count": 3, "active_band_mae_db": 0.19430772037769736}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00017949843182759416, "log_rms_mae_db": 1.7617628737855953, "rms_envelope_nrmse": 0.15780237161118918, "envelope_correlation": 0.98265054115332, "rms_gain_error_db": 0.043169497945722286, "reference_mean_power": 0.001722884360599687, "output_mean_power": 0.0017400954784431704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.0185329105758934e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006952905258544651, "output_power": 0.0006597653538599376, "reference_active": true, "delta_db": -0.227768058767173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.45148729518806e-05, "output_power": 5.641403469649695e-05, "reference_active": true, "delta_db": -0.5827268505533106}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5243024554692387e-05, "output_power": 1.716746508692908e-05, "reference_active": true, "delta_db": 0.5163502362557092}], "active_band_count": 3, "active_band_mae_db": 0.44228171519206433}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00016126256915979383, "log_rms_mae_db": 1.1082154467011107, "rms_envelope_nrmse": 0.08522991078790146, "envelope_correlation": 0.9946499656581153, "rms_gain_error_db": 0.04316950186272573, "reference_mean_power": 0.0017368536839123, "output_mean_power": 0.001754204352848765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 9.447738359038237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006450702777726167, "output_power": 0.0006168874424191845, "reference_active": true, "delta_db": -0.19401102142604476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9818892488309525e-05, "output_power": 2.536530954805965e-05, "reference_active": true, "delta_db": -0.7025134252431559}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0738004145337803e-06, "output_power": 1.0255523902895778e-06, "reference_active": true, "delta_db": -0.19965716102148398}], "active_band_count": 3, "active_band_mae_db": 0.36539386923022815}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 6.626476450042254e-05, "log_rms_mae_db": 2.073599748936407, "rms_envelope_nrmse": 0.2588606123474968, "envelope_correlation": 0.9495019170892697, "rms_gain_error_db": 0.18548249362809283, "reference_mean_power": 0.00025607969857314127, "output_mean_power": 0.00026725349863510907}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.9689769875456006e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013521942671011384, "output_power": 0.00013044683676686252, "reference_active": true, "delta_db": -0.15605538060917745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.463778228624424e-05, "output_power": 1.6340344737372756e-05, "reference_active": true, "delta_db": 0.4778593131004246}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5896173962011905e-06, "output_power": 4.790092939810416e-06, "reference_active": true, "delta_db": -0.6703814197683324}], "active_band_count": 3, "active_band_mae_db": 0.4347653711593115}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 5.564377890100535e-05, "log_rms_mae_db": 1.3639539455929077, "rms_envelope_nrmse": 0.16681828730347031, "envelope_correlation": 0.9773259522874471, "rms_gain_error_db": 0.18548249813912043, "reference_mean_power": 0.0002581560174430691, "output_mean_power": 0.0002694204160324096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 1.774101433771277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012602347588101508, "output_power": 0.00012153596250413842, "reference_active": true, "delta_db": -0.15746649081396974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.970334241514461e-06, "output_power": 8.024812969723094e-06, "reference_active": true, "delta_db": 0.6118131542270454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.812565026245218e-07, "output_power": 2.958885188723032e-07, "reference_active": true, "delta_db": -1.1008914574331505}], "active_band_count": 3, "active_band_mae_db": 0.6233903674913885}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 3.541043489510122e-05, "log_rms_mae_db": 2.667607280108835, "rms_envelope_nrmse": 0.25280896172843004, "envelope_correlation": 0.9582937840606323, "rms_gain_error_db": 0.04539813071293419, "reference_mean_power": 6.312530293815556e-05, "output_mean_power": 6.378863199049155e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 4.34877201852775e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2352362117577493e-05, "output_power": 2.1769072821400373e-05, "reference_active": true, "delta_db": -0.11483492419680039}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.837592049886246e-06, "output_power": 7.874494239701294e-06, "reference_active": true, "delta_db": -0.5011128067707609}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1202466049559064e-06, "output_power": 2.7373907418112645e-06, "reference_active": true, "delta_db": 1.1095041761757232}], "active_band_count": 3, "active_band_mae_db": 0.5751506357144281}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 3.3704303759041684e-05, "log_rms_mae_db": 2.113884224327628, "rms_envelope_nrmse": 0.15747694341869314, "envelope_correlation": 0.9831967249842828, "rms_gain_error_db": 0.04539814247122932, "reference_mean_power": 6.363712790649308e-05, "output_mean_power": 6.430583545755806e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 3.9430492064624054e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.36389645521954e-05, "output_power": 2.2913848570977526e-05, "reference_active": true, "delta_db": -0.13530410726687703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.125341563464706e-06, "output_power": 3.746137799237039e-06, "reference_active": true, "delta_db": -0.4187616382542639}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9054004218332769e-07, "output_power": 2.312261128601299e-07, "reference_active": true, "delta_db": 0.8405062129478914}], "active_band_count": 3, "active_band_mae_db": 0.4648573194896774}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8355915850421766, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a9cc51412c9f9a952745617b69c225332eb4645314f26f4c83fc1bffd8d17a21", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9599792862988412, "standard_deviation": 0.005837086664947227, "minimum": 0.9495019170892697, "maximum": 0.9667944985282838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0752343024424844, "standard_deviation": 0.0408147680982915, "minimum": 2.003450209105035, "maximum": 2.130117266251843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4732125962970847, "standard_deviation": 0.11501261856047822, "minimum": 0.3586200940438565, "maximum": 0.7112314324365547, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2274919612308578, "standard_deviation": 0.01726704596365643, "minimum": 0.20639805047754495, "maximum": 0.2588606123474968, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.983412348676861, "standard_deviation": 0.0033698083602078845, "minimum": 0.9773259522874471, "maximum": 0.9872875080319539, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3486640458786945, "standard_deviation": 0.027865648032819438, "minimum": 1.2871275386718581, "maximum": 1.3755901419212266, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6228958215321596, "standard_deviation": 0.20235010155336866, "minimum": 0.39695222599668284, "maximum": 1.0698810358281365, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14082804164628115, "standard_deviation": 0.01465565530853528, "minimum": 0.12290821733713195, "maximum": 0.16681828730347031, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9580728856328966, "standard_deviation": 0.0031754820160276058, "minimum": 0.9537292488951195, "maximum": 0.9623394315487835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.744039445882928, "standard_deviation": 0.04637730424961258, "minimum": 2.667607280108835, "maximum": 2.800973221175377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.49777283102356445, "standard_deviation": 0.14607377727561552, "minimum": 0.250575042814178, "maximum": 0.7323208440853278, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25532635590607, "standard_deviation": 0.007872631222738152, "minimum": 0.2436717131697765, "maximum": 0.2681595253222602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.981936114552962, "standard_deviation": 0.0019555175165716075, "minimum": 0.978717608673348, "maximum": 0.9843108819796035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.192888330857595, "standard_deviation": 0.050676475470230894, "minimum": 2.113884224327628, "maximum": 2.265997952420175, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.43752563446664106, "standard_deviation": 0.15089689365060632, "minimum": 0.23915162733740616, "maximum": 0.6932863593324395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1648797329368999, "standard_deviation": 0.007293755688010738, "minimum": 0.15670266604295202, "maximum": 0.17798008019929923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.983730573163938, "standard_deviation": 0.002120194603332926, "minimum": 0.9808692099660011, "maximum": 0.9880035416417209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7648456774166377, "standard_deviation": 0.06322824636896208, "minimum": 1.6961002602884323, "maximum": 1.8783197021728906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7440143838592673, "standard_deviation": 0.33301714831144286, "minimum": 0.20549022839649878, "maximum": 1.2455022184404398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1538595793414437, "standard_deviation": 0.010791436591614204, "minimum": 0.13221248856994636, "maximum": 0.16806917785007608, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949630225761033, "standard_deviation": 0.0008625066698344449, "minimum": 0.9941259151526285, "maximum": 0.9964930020119701, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0614198652326694, "standard_deviation": 0.06965109460928594, "minimum": 0.9696054302295524, "maximum": 1.1653092645662626, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5591569332427456, "standard_deviation": 0.2216755285043194, "minimum": 0.34323469228110853, "maximum": 0.8886878686867314, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08870815977739618, "standard_deviation": 0.009671944626877088, "minimum": 0.07415935091323728, "maximum": 0.09880787609107197, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1612673272590479, "standard_deviation": 0.0309537414269918, "minimum": 0.1201365858829808, "maximum": 0.21697745361751633, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.862443917844631, "standard_deviation": 0.008521305302499053, "minimum": 5.856345974738865, "maximum": 5.876557666038523, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5183461026971705, "standard_deviation": 0.03086312638737854, "minimum": 0.4612507690443047, "maximum": 0.5649166135915186, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10047983808806961, "standard_deviation": 0.08676674396694835, "minimum": -0.21108919957601452, "maximum": 0.04105258925741037, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9925611671278174, "standard_deviation": 0.001188006466775519, "minimum": 0.9901657139607689, "maximum": 0.9937876314062551, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0128274598388094, "standard_deviation": 0.03825338679173191, "minimum": 0.9710375033726912, "maximum": 1.0691156324392896, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.28288611176696754, "standard_deviation": 0.17329733311751394, "minimum": 0.06909209306026332, "maximum": 0.565066106956482, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09236066778208653, "standard_deviation": 0.007245211015616817, "minimum": 0.0840982275173259, "maximum": 0.1064688131968296, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976062775373953, "standard_deviation": 0.0005564166783778371, "minimum": 0.9966470614380761, "maximum": 0.9983464045923321, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.554635183744427, "standard_deviation": 0.040420382945468694, "minimum": 0.4951790293023337, "maximum": 0.6171757793943919, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24976589878500952, "standard_deviation": 0.14820436849881588, "minimum": 0.11640169681671249, "maximum": 0.5043405003568824, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05189375408874036, "standard_deviation": 0.006083324127844128, "minimum": 0.04288349609395776, "maximum": 0.05997841369857791, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123286-0008", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.2299551963806152, "predicted_mos": 2.2299551963806152, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "6405e82ebd91cad60ff698d8555838740fed6c70649552d2c5290f43541f6433", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.632705199997872, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0474216118810995, "全帯域_10ms_rms_envelope_nrmse": 0.0917958769362239, "全帯域_10ms_envelope_correlation": 0.9926393330297789, "全帯域_10ms_modulation_mae_db": 0.1633133051614921, "全帯域_40ms_log_rms_mae_db": 0.5262051472570018, "全帯域_40ms_rms_envelope_nrmse": 0.049006597298342815, "全帯域_40ms_envelope_correlation": 0.9977628878167233, "全帯域_40ms_modulation_mae_db": 0.17831468199216238, "300–1000 Hz_10ms_log_rms_mae_db": 1.552388449459996, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14700788147150015, "300–1000 Hz_10ms_envelope_correlation": 0.9848280248236652, "300–1000 Hz_10ms_modulation_mae_db": 0.6547801078718604, "300–1000 Hz_40ms_log_rms_mae_db": 0.9069436064813616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07662274116305184, "300–1000 Hz_40ms_envelope_correlation": 0.9956345204619288, "300–1000 Hz_40ms_modulation_mae_db": 0.35420956589030467, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6576825611274753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19109073053480863, "1000–3000 Hz_10ms_envelope_correlation": 0.9715640055959344, "1000–3000 Hz_10ms_modulation_mae_db": 0.2530982161309263, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1661532444628715, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12341513006649012, "1000–3000 Hz_40ms_envelope_correlation": 0.9870769345529204, "1000–3000 Hz_40ms_modulation_mae_db": 0.1778417127233458, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9441154190811758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22217176665967076, "3000–6000 Hz_10ms_envelope_correlation": 0.9672024648537687, "3000–6000 Hz_10ms_modulation_mae_db": 0.4965883869535837, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7968735289965363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15443735027411526, "3000–6000 Hz_40ms_envelope_correlation": 0.9835217864960364, "3000–6000 Hz_40ms_modulation_mae_db": 0.49582553472410623, "mr_stft_db": 5.464742460990874, "band_mae_db": 0.11823891989611823, "rms_gain_error_db": -0.04170810402693972, "output_peak": 0.5057258599711991}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0179958981936987, "全帯域_10ms_rms_envelope_nrmse": 0.08646626118064477, "全帯域_10ms_envelope_correlation": 0.9937307500052911, "全帯域_10ms_modulation_mae_db": 0.2642250192222418, "全帯域_40ms_log_rms_mae_db": 0.5720002387598357, "全帯域_40ms_rms_envelope_nrmse": 0.054219084255007606, "全帯域_40ms_envelope_correlation": 0.9976607965914872, "全帯域_40ms_modulation_mae_db": 0.23968720397917045, "300–1000 Hz_10ms_log_rms_mae_db": 1.584301489070613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15785556362933426, "300–1000 Hz_10ms_envelope_correlation": 0.9830911443555646, "300–1000 Hz_10ms_modulation_mae_db": 0.7352504390190148, "300–1000 Hz_40ms_log_rms_mae_db": 1.0173912800866907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11165162214579256, "300–1000 Hz_40ms_envelope_correlation": 0.9918077556641065, "300–1000 Hz_40ms_modulation_mae_db": 0.5309005100341159, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7192539234521935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.204822799096673, "1000–3000 Hz_10ms_envelope_correlation": 0.9672928320191952, "1000–3000 Hz_10ms_modulation_mae_db": 0.5408600970457028, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1958175643327364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12245507847331649, "1000–3000 Hz_40ms_envelope_correlation": 0.9880989232966145, "1000–3000 Hz_40ms_modulation_mae_db": 0.5896233509466632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9732103065704452, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22340398738527928, "3000–6000 Hz_10ms_envelope_correlation": 0.9678202518145536, "3000–6000 Hz_10ms_modulation_mae_db": 0.35195138573275625, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7818423257911933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14898829491481677, "3000–6000 Hz_40ms_envelope_correlation": 0.9848923387917151, "3000–6000 Hz_40ms_modulation_mae_db": 0.27537270501941724, "mr_stft_db": 5.458018364500648, "band_mae_db": 0.2420279190378476, "rms_gain_error_db": -0.1981202775801464, "output_peak": 0.5692135869286847}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9939853421143687, "全帯域_10ms_rms_envelope_nrmse": 0.0940567874053704, "全帯域_10ms_envelope_correlation": 0.9922743179538616, "全帯域_10ms_modulation_mae_db": 0.30375705613392945, "全帯域_40ms_log_rms_mae_db": 0.5665945298626892, "全帯域_40ms_rms_envelope_nrmse": 0.05090942346083057, "全帯域_40ms_envelope_correlation": 0.9976977611774495, "全帯域_40ms_modulation_mae_db": 0.3663138106314445, "300–1000 Hz_10ms_log_rms_mae_db": 1.5686965336288403, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1474726694323475, "300–1000 Hz_10ms_envelope_correlation": 0.9853070439562146, "300–1000 Hz_10ms_modulation_mae_db": 0.3787070594223763, "300–1000 Hz_40ms_log_rms_mae_db": 0.9638052979397804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09803641271767657, "300–1000 Hz_40ms_envelope_correlation": 0.9938759467422962, "300–1000 Hz_40ms_modulation_mae_db": 0.5687193794936111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6472607753586197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19127503888936037, "1000–3000 Hz_10ms_envelope_correlation": 0.9714449607465219, "1000–3000 Hz_10ms_modulation_mae_db": 0.5241639823901669, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1500972573885488, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11564786527739823, "1000–3000 Hz_40ms_envelope_correlation": 0.9890003131360544, "1000–3000 Hz_40ms_modulation_mae_db": 0.605356541005989, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9758728201270275, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20812371680169045, "3000–6000 Hz_10ms_envelope_correlation": 0.971854548981017, "3000–6000 Hz_10ms_modulation_mae_db": 0.24376358229381168, "3000–6000 Hz_40ms_log_rms_mae_db": 1.76217278074392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14567668070143708, "3000–6000 Hz_40ms_envelope_correlation": 0.9854405413631773, "3000–6000 Hz_40ms_modulation_mae_db": 0.27670151421522593, "mr_stft_db": 5.470507812558682, "band_mae_db": 0.14662602234185573, "rms_gain_error_db": -0.12104309350347302, "output_peak": 0.472236638732377}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0600877417468664, "全帯域_10ms_rms_envelope_nrmse": 0.08312091554411094, "全帯域_10ms_envelope_correlation": 0.9939793680252851, "全帯域_10ms_modulation_mae_db": 0.2082216393303645, "全帯域_40ms_log_rms_mae_db": 0.5467756747968926, "全帯域_40ms_rms_envelope_nrmse": 0.04378067902087544, "全帯域_40ms_envelope_correlation": 0.9983126218946802, "全帯域_40ms_modulation_mae_db": 0.345739754040854, "300–1000 Hz_10ms_log_rms_mae_db": 1.6517517380397249, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13582366873575671, "300–1000 Hz_10ms_envelope_correlation": 0.9870260059572636, "300–1000 Hz_10ms_modulation_mae_db": 0.8489506923984639, "300–1000 Hz_40ms_log_rms_mae_db": 1.0426193384992706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06836598242321063, "300–1000 Hz_40ms_envelope_correlation": 0.9966955670664054, "300–1000 Hz_40ms_modulation_mae_db": 0.6996229663376455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6235073491513805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19862391901987206, "1000–3000 Hz_10ms_envelope_correlation": 0.9707954861535687, "1000–3000 Hz_10ms_modulation_mae_db": 1.2226439576029742, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1076687124957305, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13279662281783464, "1000–3000 Hz_40ms_envelope_correlation": 0.9880064168405208, "1000–3000 Hz_40ms_modulation_mae_db": 1.3886608373852531, "3000–6000 Hz_10ms_log_rms_mae_db": 2.007041718516646, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2306644815023338, "3000–6000 Hz_10ms_envelope_correlation": 0.9655234364202158, "3000–6000 Hz_10ms_modulation_mae_db": 0.18401044393648994, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8368250918919249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15439578953979705, "3000–6000 Hz_40ms_envelope_correlation": 0.9836465190800393, "3000–6000 Hz_40ms_modulation_mae_db": 0.3387699300587131, "mr_stft_db": 5.475285215671259, "band_mae_db": 0.2961024931670713, "rms_gain_error_db": -0.10819281430699432, "output_peak": 0.45465099740613407}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0455227334243486, "全帯域_10ms_rms_envelope_nrmse": 0.08289472584643506, "全帯域_10ms_envelope_correlation": 0.9940108037409717, "全帯域_10ms_modulation_mae_db": 0.33229843849476765, "全帯域_40ms_log_rms_mae_db": 0.5930385588037863, "全帯域_40ms_rms_envelope_nrmse": 0.05042958095880785, "全帯域_40ms_envelope_correlation": 0.9977081312858237, "全帯域_40ms_modulation_mae_db": 0.3850694620384618, "300–1000 Hz_10ms_log_rms_mae_db": 1.5534920657354676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13156710020126403, "300–1000 Hz_10ms_envelope_correlation": 0.9878468735399828, "300–1000 Hz_10ms_modulation_mae_db": 0.5250829372961728, "300–1000 Hz_40ms_log_rms_mae_db": 0.9616939189958107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07225901444612198, "300–1000 Hz_40ms_envelope_correlation": 0.9962889649215938, "300–1000 Hz_40ms_modulation_mae_db": 0.7735906432721302, "1000–3000 Hz_10ms_log_rms_mae_db": 1.704869138273884, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22359883683265636, "1000–3000 Hz_10ms_envelope_correlation": 0.9623421163957218, "1000–3000 Hz_10ms_modulation_mae_db": 0.7413361212297623, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2679821270985518, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14559939060852795, "1000–3000 Hz_40ms_envelope_correlation": 0.9853312171114427, "1000–3000 Hz_40ms_modulation_mae_db": 0.7535091775657259, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9644979039768782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2115846919787426, "3000–6000 Hz_10ms_envelope_correlation": 0.9705148788874928, "3000–6000 Hz_10ms_modulation_mae_db": 0.15337979101399193, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7632357338142615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14763458335675791, "3000–6000 Hz_40ms_envelope_correlation": 0.9848759883721047, "3000–6000 Hz_40ms_modulation_mae_db": 0.26275711736718516, "mr_stft_db": 5.481631941712531, "band_mae_db": 0.2382760798477632, "rms_gain_error_db": -0.10603800324228255, "output_peak": 0.460470839387781}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9906745892177531, "全帯域_10ms_rms_envelope_nrmse": 0.10928812620244709, "全帯域_10ms_envelope_correlation": 0.989672296944158, "全帯域_10ms_modulation_mae_db": 0.426937091077267, "全帯域_40ms_log_rms_mae_db": 0.5583390840442654, "全帯域_40ms_rms_envelope_nrmse": 0.055283884664209845, "全帯域_40ms_envelope_correlation": 0.9974846984135548, "全帯域_40ms_modulation_mae_db": 0.4760907321437415, "300–1000 Hz_10ms_log_rms_mae_db": 1.5631791673551085, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17083253216733785, "300–1000 Hz_10ms_envelope_correlation": 0.9797953873489978, "300–1000 Hz_10ms_modulation_mae_db": 0.3340012188426739, "300–1000 Hz_40ms_log_rms_mae_db": 0.9511073350334909, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08772979734329069, "300–1000 Hz_40ms_envelope_correlation": 0.9953059385688918, "300–1000 Hz_40ms_modulation_mae_db": 0.2959994967568473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6744019928351268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20124285209339102, "1000–3000 Hz_10ms_envelope_correlation": 0.9683942033307351, "1000–3000 Hz_10ms_modulation_mae_db": 0.841695514336009, "1000–3000 Hz_40ms_log_rms_mae_db": 1.174248723179066, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11932164090853248, "1000–3000 Hz_40ms_envelope_correlation": 0.9886140258374637, "1000–3000 Hz_40ms_modulation_mae_db": 0.560318844678165, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0150378526299613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22902565647710585, "3000–6000 Hz_10ms_envelope_correlation": 0.9660461743614134, "3000–6000 Hz_10ms_modulation_mae_db": 0.40114492330873325, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8097486077491305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1542808436575685, "3000–6000 Hz_40ms_envelope_correlation": 0.9836278064780185, "3000–6000 Hz_40ms_modulation_mae_db": 0.5014646865783147, "mr_stft_db": 5.48216549466845, "band_mae_db": 0.15272677993089265, "rms_gain_error_db": -0.18666550045301888, "output_peak": 0.4781687173385062}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0440433330850776, "全帯域_10ms_rms_envelope_nrmse": 0.09503811802989727, "全帯域_10ms_envelope_correlation": 0.9928973632502088, "全帯域_10ms_modulation_mae_db": 0.2273315560036965, "全帯域_40ms_log_rms_mae_db": 0.6093362546684395, "全帯域_40ms_rms_envelope_nrmse": 0.06806192579750549, "全帯域_40ms_envelope_correlation": 0.9966830602965594, "全帯域_40ms_modulation_mae_db": 0.15277626020722526, "300–1000 Hz_10ms_log_rms_mae_db": 1.5795440784836692, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15238083136686198, "300–1000 Hz_10ms_envelope_correlation": 0.9869338816571773, "300–1000 Hz_10ms_modulation_mae_db": 0.9423101487106281, "300–1000 Hz_40ms_log_rms_mae_db": 0.9810532494825273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11103595037114537, "300–1000 Hz_40ms_envelope_correlation": 0.9949879053222842, "300–1000 Hz_40ms_modulation_mae_db": 1.0237838213718482, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6714463255885363, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21125780363294389, "1000–3000 Hz_10ms_envelope_correlation": 0.965143896068235, "1000–3000 Hz_10ms_modulation_mae_db": 0.4114228122437004, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1629583858898054, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12594019216342464, "1000–3000 Hz_40ms_envelope_correlation": 0.9870186142387887, "1000–3000 Hz_40ms_modulation_mae_db": 0.5995336922329201, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9937168073668898, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22635629392472406, "3000–6000 Hz_10ms_envelope_correlation": 0.9664984175015914, "3000–6000 Hz_10ms_modulation_mae_db": 0.14819860030310691, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7853283361638168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15461751413271493, "3000–6000 Hz_40ms_envelope_correlation": 0.9834227578197581, "3000–6000 Hz_40ms_modulation_mae_db": 0.11561128301278083, "mr_stft_db": 5.458668631663255, "band_mae_db": 0.2488770842749329, "rms_gain_error_db": -0.312648946135044, "output_peak": 0.48531397834539153}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9670136566092277, "全帯域_10ms_rms_envelope_nrmse": 0.09602644261968687, "全帯域_10ms_envelope_correlation": 0.9919101730050909, "全帯域_10ms_modulation_mae_db": 0.415693983728431, "全帯域_40ms_log_rms_mae_db": 0.48714694118350416, "全帯域_40ms_rms_envelope_nrmse": 0.046712329894804845, "全帯域_40ms_envelope_correlation": 0.9979885026984137, "全帯域_40ms_modulation_mae_db": 0.4222722016647979, "300–1000 Hz_10ms_log_rms_mae_db": 1.653230597045695, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16661963009064745, "300–1000 Hz_10ms_envelope_correlation": 0.980413873323711, "300–1000 Hz_10ms_modulation_mae_db": 0.49409731387312683, "300–1000 Hz_40ms_log_rms_mae_db": 1.0191789591654694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08723499339817388, "300–1000 Hz_40ms_envelope_correlation": 0.9943829399068581, "300–1000 Hz_40ms_modulation_mae_db": 0.4767001829045487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5888766930864955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19123837453543005, "1000–3000 Hz_10ms_envelope_correlation": 0.9715049104535531, "1000–3000 Hz_10ms_modulation_mae_db": 0.3203895432457167, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0877060458679817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11591760439031543, "1000–3000 Hz_40ms_envelope_correlation": 0.98903837134657, "1000–3000 Hz_40ms_modulation_mae_db": 0.44060145980056714, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9968677982948588, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.251715975461677, "3000–6000 Hz_10ms_envelope_correlation": 0.9583841905957754, "3000–6000 Hz_10ms_modulation_mae_db": 0.313195680002898, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7864945659476594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16461653270981141, "3000–6000 Hz_40ms_envelope_correlation": 0.9810469347464955, "3000–6000 Hz_40ms_modulation_mae_db": 0.46297640398413925, "mr_stft_db": 5.4644932194605, "band_mae_db": 0.14636204280381834, "rms_gain_error_db": -0.06944573233032092, "output_peak": 0.5228628963356201}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59760, "output_samples": 59760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.576890551726625, "spectral_convergence": 0.29496506533741346, "spectral_convergence_status": "ok", "amplitude_floor": 1.6701466153856754e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.288034494023393, "spectral_convergence": 0.29383264432176864, "spectral_convergence_status": "ok", "amplitude_floor": 1.1197689318182202e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.529302337222606, "spectral_convergence": 0.31445370855830695, "spectral_convergence_status": "ok", "amplitude_floor": 1.0416079453540003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0010765709682483164, "output_power": 0.0010522922036579443, "delta_db": -0.09906311313797439, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019015067355649686, "output_power": 0.0019181907330612954, "delta_db": 0.03793920428537564, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009390993860770502, "output_power": 0.0009114667140631415, "delta_db": -0.1297074339286954, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028731828855173817, "output_power": 0.00028694572800556435, "delta_db": -0.005635074022167888, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.911907847527107e-05, "output_power": 5.4475592380881445e-05, "delta_db": -0.3552569426141402, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.5637714608274427e-05, "output_power": 5.459917837201126e-05, "delta_db": -0.08183175138835583, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11823891989611823, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04170810402693972, "pitch": null, "output_rms_dbfs": -23.687622501373113, "output_peak": 0.5057258599711991, "reference_rms_dbfs": -23.64591439734617, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.464742460990874}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00019933639001390812, "log_rms_mae_db": 1.0474216118810995, "rms_envelope_nrmse": 0.0917958769362239, "envelope_correlation": 0.9926393330297789, "rms_gain_error_db": -0.04170810461319202, "reference_mean_power": 0.004325042032530898, "output_mean_power": 0.004283704672340001}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 2.983288998803909e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014406782878671795, "output_power": 0.0014160530948255222, "reference_active": true, "delta_db": -0.07487473564327583}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010200208961765019, "output_power": 0.00010551092719139838, "reference_active": true, "delta_db": 0.1468837065236785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7770617096993514e-05, "output_power": 2.9539536578644973e-05, "reference_active": true, "delta_db": 0.2681814733175219}], "active_band_count": 3, "active_band_mae_db": 0.1633133051614921}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00018454723874230657, "log_rms_mae_db": 0.5262051472570018, "rms_envelope_nrmse": 0.049006597298342815, "envelope_correlation": 0.9977628878167233, "rms_gain_error_db": -0.041708090111420563, "reference_mean_power": 0.004360109839005934, "output_mean_power": 0.004318437326423539}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 2.740771226026708e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013018843976497208, "output_power": 0.0012891111461829695, "reference_active": true, "delta_db": -0.04282058749147853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649956473480408e-05, "output_power": 4.7460438926727724e-05, "reference_active": true, "delta_db": 0.08882862731199632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9034855452992526e-06, "output_power": 1.7346829109361456e-06, "reference_active": true, "delta_db": -0.4032948311730123}], "active_band_count": 3, "active_band_mae_db": 0.17831468199216238}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00017949843182759416, "log_rms_mae_db": 1.552388449459996, "rms_envelope_nrmse": 0.14700788147150015, "envelope_correlation": 0.9848280248236652, "rms_gain_error_db": -0.04120233404774089, "reference_mean_power": 0.001722884360599687, "output_mean_power": 0.0017066163237197901}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.0185329105758934e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006952905258544651, "output_power": 0.0006772292626878086, "reference_active": true, "delta_db": -0.11430596198673212}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.45148729518806e-05, "output_power": 6.208385462075476e-05, "reference_active": true, "delta_db": -0.16681173089411905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5243024554692387e-05, "output_power": 2.24591549948553e-05, "reference_active": true, "delta_db": 1.68322263073473}], "active_band_count": 3, "active_band_mae_db": 0.6547801078718604}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00016126256915979383, "log_rms_mae_db": 0.9069436064813616, "rms_envelope_nrmse": 0.07662274116305184, "envelope_correlation": 0.9956345204619288, "rms_gain_error_db": -0.04120233031978472, "reference_mean_power": 0.0017368536839123, "output_mean_power": 0.0017204537455957007}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 9.447738359038237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006450702777726167, "output_power": 0.0006363082149670652, "reference_active": true, "delta_db": -0.059395016533808034}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9818892488309525e-05, "output_power": 2.8252397543379058e-05, "reference_active": true, "delta_db": -0.23436200526044376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0738004145337803e-06, "output_power": 1.2817718659420163e-06, "reference_active": true, "delta_db": 0.7688716758766623}], "active_band_count": 3, "active_band_mae_db": 0.35420956589030467}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 6.626476450042254e-05, "log_rms_mae_db": 1.6576825611274753, "rms_envelope_nrmse": 0.19109073053480863, "envelope_correlation": 0.9715640055959344, "rms_gain_error_db": -0.07722594547472415, "reference_mean_power": 0.00025607969857314127, "output_mean_power": 0.000251566353937577}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.9689769875456006e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013521942671011384, "output_power": 0.0001260696132644568, "reference_active": true, "delta_db": -0.30428669746489734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.463778228624424e-05, "output_power": 1.5538816857868126e-05, "reference_active": true, "delta_db": 0.25942664647706054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5896173962011905e-06, "output_power": 5.847095838446657e-06, "reference_active": true, "delta_db": 0.19558130445082106}], "active_band_count": 3, "active_band_mae_db": 0.2530982161309263}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 5.564377890100535e-05, "log_rms_mae_db": 1.1661532444628715, "rms_envelope_nrmse": 0.12341513006649012, "envelope_correlation": 0.9870769345529204, "rms_gain_error_db": -0.07722593504495195, "reference_mean_power": 0.0002581560174430691, "output_mean_power": 0.00025360607878337416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 1.774101433771277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012602347588101508, "output_power": 0.00011879320787168751, "reference_active": true, "delta_db": -0.2565984371027207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.970334241514461e-06, "output_power": 7.308217586072074e-06, "reference_active": true, "delta_db": 0.20557865144794588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.812565026245218e-07, "output_power": 3.87571705988818e-07, "reference_active": true, "delta_db": 0.07134804961937088}], "active_band_count": 3, "active_band_mae_db": 0.1778417127233458}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 3.541043489510122e-05, "log_rms_mae_db": 1.9441154190811758, "rms_envelope_nrmse": 0.22217176665967076, "envelope_correlation": 0.9672024648537687, "rms_gain_error_db": -0.1897819203242665, "reference_mean_power": 6.312530293815556e-05, "output_mean_power": 6.0426199928476546e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 4.34877201852775e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2352362117577493e-05, "output_power": 2.052649570767435e-05, "reference_active": true, "delta_db": -0.3700861182380106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.837592049886246e-06, "output_power": 7.792454909361698e-06, "reference_active": true, "delta_db": -0.5465965238479707}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1202466049559064e-06, "output_power": 2.419327312508492e-06, "reference_active": true, "delta_db": 0.5730825187747699}], "active_band_count": 3, "active_band_mae_db": 0.4965883869535837}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 3.3704303759041684e-05, "log_rms_mae_db": 1.7968735289965363, "rms_envelope_nrmse": 0.15443735027411526, "envelope_correlation": 0.9835217864960364, "rms_gain_error_db": -0.1897819097456665, "reference_mean_power": 6.363712790649308e-05, "output_mean_power": 6.091614050368458e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 3.9430492064624054e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.36389645521954e-05, "output_power": 2.1565804831000288e-05, "reference_active": true, "delta_db": -0.39862778846862684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.125341563464706e-06, "output_power": 3.5566233898240237e-06, "reference_active": true, "delta_db": -0.6442203213778341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9054004218332769e-07, "output_power": 2.1108098224722546e-07, "reference_active": true, "delta_db": 0.4446284943258576}], "active_band_count": 3, "active_band_mae_db": 0.49582553472410623}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8560394377746497, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7849e96de5460b2b0e9fe52c19c792421cc485c85d5b37331dd1e308e4e03af8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9685603013454331, "standard_deviation": 0.0034430465981917144, "minimum": 0.9623421163957218, "maximum": 0.9715640055959344, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6609123448592138, "standard_deviation": 0.04211199441548239, "minimum": 1.5888766930864955, "maximum": 1.7192539234521935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6069512805281199, "standard_deviation": 0.3182596417469757, "minimum": 0.2530982161309263, "maximum": 1.2226439576029742, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20164379432939192, "standard_deviation": 0.01147040007396234, "minimum": 0.19109073053480863, "maximum": 0.22359883683265636, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9877731020450469, "standard_deviation": 0.0012530764786562473, "minimum": 0.9853312171114427, "maximum": 0.98903837134657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1640790075894114, "standard_deviation": 0.05493192619440609, "minimum": 1.0877060458679817, "maximum": 1.2679821270985518, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6394307020423287, "standard_deviation": 0.3460902927097637, "minimum": 0.1778417127233458, "maximum": 1.3886608373852531, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12513669058823002, "standard_deviation": 0.009978843632623491, "minimum": 0.11564786527739823, "maximum": 0.14559939060852795, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9667305454269786, "standard_deviation": 0.0040289622621775, "minimum": 0.9583841905957754, "maximum": 0.971854548981017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9837950783204854, "standard_deviation": 0.023626588547169827, "minimum": 1.9441154190811758, "maximum": 2.0150378526299613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.2865290991931715, "standard_deviation": 0.12618348483929848, "minimum": 0.14819860030310691, "maximum": 0.4965883869535837, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.225380821273903, "standard_deviation": 0.013298982367557928, "minimum": 0.20812371680169045, "maximum": 0.251715975461677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9838093341434181, "standard_deviation": 0.0013571989892747225, "minimum": 0.9810469347464955, "maximum": 0.9854405413631773, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7903151213873052, "standard_deviation": 0.024554277960266515, "minimum": 1.76217278074392, "maximum": 1.8368250918919249, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3411848968699853, "standard_deviation": 0.13625924043938523, "minimum": 0.11561128301278083, "maximum": 0.5014646865783147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15308094866087735, "standard_deviation": 0.005875797895992827, "minimum": 0.14567668070143708, "maximum": 0.16461653270981141, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9844052793703221, "standard_deviation": 0.0030474316680052975, "minimum": 0.9797953873489978, "maximum": 0.9878468735399828, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5883230148523895, "standard_deviation": 0.04114667037952077, "minimum": 1.552388449459996, "maximum": 1.653230597045695, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6141474896792896, "standard_deviation": 0.2188929540258636, "minimum": 0.3340012188426739, "maximum": 0.9423101487106281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15119498463688125, "standard_deviation": 0.013743783886941145, "minimum": 0.13156710020126403, "maximum": 0.17083253216733785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9948724423317956, "standard_deviation": 0.0015462976181167083, "minimum": 0.9918077556641065, "maximum": 0.9966955670664054, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9804741232105502, "standard_deviation": 0.044115100934434805, "minimum": 0.9069436064813616, "maximum": 1.0426193384992706, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5904408207576315, "standard_deviation": 0.23685603100022978, "minimum": 0.2959994967568473, "maximum": 1.0237838213718482, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08911706425105793, "standard_deviation": 0.016665510558264336, "minimum": 0.06836598242321063, "maximum": 0.11165162214579256, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1986546676625375, "standard_deviation": 0.06490325434633068, "minimum": 0.11823891989611823, "maximum": 0.2961024931670713, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.469439142653275, "standard_deviation": 0.009553623583872663, "minimum": 5.458018364500648, "maximum": 5.48216549466845, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4935804393057117, "standard_deviation": 0.0379821517844966, "minimum": 0.45465099740613407, "maximum": 0.5692135869286847, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14298280894727747, "standard_deviation": 0.08656607476053016, "minimum": -0.312648946135044, "maximum": -0.04170810402693972, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9926393007443307, "standard_deviation": 0.0014344042955785749, "minimum": 0.989672296944158, "maximum": 0.9940108037409717, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0208431132840552, "standard_deviation": 0.03366025221907517, "minimum": 0.9670136566092277, "maximum": 1.0600877417468664, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2927222611440238, "standard_deviation": 0.09546785228160336, "minimum": 0.1633133051614921, "maximum": 0.426937091077267, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09233590672060205, "standard_deviation": 0.008619392640922093, "minimum": 0.08289472584643506, "maximum": 0.10928812620244709, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976623075218365, "standard_deviation": 0.0004680136838626927, "minimum": 0.9966830602965594, "maximum": 0.9983126218946802, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5574295536720519, "standard_deviation": 0.03833601451465472, "minimum": 0.48714694118350416, "maximum": 0.6093362546684395, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.32078301333723225, "standard_deviation": 0.11732081023090668, "minimum": 0.15277626020722526, "maximum": 0.4760907321437415, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.052300438168798064, "standard_deviation": 0.007378582078105014, "minimum": 0.04378067902087544, "maximum": 0.06806192579750549, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123286-0008", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.1716151237487793, "predicted_mos": 2.1716151237487793, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "bffaf9307b1cf37427ab859f75647b67c96988548127ac05891d5eb86887cafd", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.853291499894112, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0024887062011667, "全帯域_10ms_rms_envelope_nrmse": 0.09347200124155028, "全帯域_10ms_envelope_correlation": 0.9923087515036229, "全帯域_10ms_modulation_mae_db": 0.4734385436206899, "全帯域_40ms_log_rms_mae_db": 0.515987775763132, "全帯域_40ms_rms_envelope_nrmse": 0.05280500980488158, "全帯域_40ms_envelope_correlation": 0.9974131047059355, "全帯域_40ms_modulation_mae_db": 0.45597065435195755, "300–1000 Hz_10ms_log_rms_mae_db": 1.5660031770957128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14100366763065625, "300–1000 Hz_10ms_envelope_correlation": 0.9859831340261657, "300–1000 Hz_10ms_modulation_mae_db": 1.0676773673675815, "300–1000 Hz_40ms_log_rms_mae_db": 0.9099463931013742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0800270198954429, "300–1000 Hz_40ms_envelope_correlation": 0.9952466260671909, "300–1000 Hz_40ms_modulation_mae_db": 0.9137641774035138, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8134303085390568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21310724667926015, "1000–3000 Hz_10ms_envelope_correlation": 0.9643944184996958, "1000–3000 Hz_10ms_modulation_mae_db": 0.6319212022835518, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2331957019672128, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14669840527389713, "1000–3000 Hz_40ms_envelope_correlation": 0.9819752436192398, "1000–3000 Hz_40ms_modulation_mae_db": 0.7698862616130056, "3000–6000 Hz_10ms_log_rms_mae_db": 2.213003652470715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22829651881878418, "3000–6000 Hz_10ms_envelope_correlation": 0.967784895832316, "3000–6000 Hz_10ms_modulation_mae_db": 0.3818366877945709, "3000–6000 Hz_40ms_log_rms_mae_db": 1.915395066120604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1587821367669744, "3000–6000 Hz_40ms_envelope_correlation": 0.9839795205201757, "3000–6000 Hz_40ms_modulation_mae_db": 0.34054226358886924, "mr_stft_db": 5.5875096541303835, "band_mae_db": 0.15664006772609415, "rms_gain_error_db": -0.07591203900179223, "output_peak": 0.5019204180729975}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.047158323319303, "全帯域_10ms_rms_envelope_nrmse": 0.1001103400778519, "全帯域_10ms_envelope_correlation": 0.9912867950476656, "全帯域_10ms_modulation_mae_db": 0.46791977434504545, "全帯域_40ms_log_rms_mae_db": 0.5751211134963171, "全帯域_40ms_rms_envelope_nrmse": 0.05507083076091156, "全帯域_40ms_envelope_correlation": 0.9973736811431423, "全帯域_40ms_modulation_mae_db": 0.27682248688917305, "300–1000 Hz_10ms_log_rms_mae_db": 1.7624961441047193, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15681172116645564, "300–1000 Hz_10ms_envelope_correlation": 0.9828832172485377, "300–1000 Hz_10ms_modulation_mae_db": 1.5023703923485792, "300–1000 Hz_40ms_log_rms_mae_db": 1.0856307523916249, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08497908383198509, "300–1000 Hz_40ms_envelope_correlation": 0.9952742204573845, "300–1000 Hz_40ms_modulation_mae_db": 1.0587219555306373, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8712790506868002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21981589402915663, "1000–3000 Hz_10ms_envelope_correlation": 0.9621811855633358, "1000–3000 Hz_10ms_modulation_mae_db": 0.6361473327494506, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2110534160222322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14058579102581448, "1000–3000 Hz_40ms_envelope_correlation": 0.9839760961204078, "1000–3000 Hz_40ms_modulation_mae_db": 0.527522969526598, "3000–6000 Hz_10ms_log_rms_mae_db": 2.249828275538796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22363159287135354, "3000–6000 Hz_10ms_envelope_correlation": 0.9669785815847498, "3000–6000 Hz_10ms_modulation_mae_db": 0.1330454699172527, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9215205945142104, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15221393991754473, "3000–6000 Hz_40ms_envelope_correlation": 0.9839965428751012, "3000–6000 Hz_40ms_modulation_mae_db": 0.17652447980839625, "mr_stft_db": 5.591034096507499, "band_mae_db": 0.17191692447700715, "rms_gain_error_db": -0.1521389392259092, "output_peak": 0.466590949355304}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9924309369232487, "全帯域_10ms_rms_envelope_nrmse": 0.09662461106395716, "全帯域_10ms_envelope_correlation": 0.9919680920589179, "全帯域_10ms_modulation_mae_db": 0.31056036869455533, "全帯域_40ms_log_rms_mae_db": 0.5580240233490845, "全帯域_40ms_rms_envelope_nrmse": 0.06206998588302105, "全帯域_40ms_envelope_correlation": 0.9966770162596668, "全帯域_40ms_modulation_mae_db": 0.29479295763211444, "300–1000 Hz_10ms_log_rms_mae_db": 1.6337141299117157, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14987216793026137, "300–1000 Hz_10ms_envelope_correlation": 0.9847193308603688, "300–1000 Hz_10ms_modulation_mae_db": 1.0748246302399307, "300–1000 Hz_40ms_log_rms_mae_db": 0.9990031845205888, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09358453270271491, "300–1000 Hz_40ms_envelope_correlation": 0.9944689419962173, "300–1000 Hz_40ms_modulation_mae_db": 0.8449726569121361, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7827888594503645, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19151758449805426, "1000–3000 Hz_10ms_envelope_correlation": 0.9715554117630024, "1000–3000 Hz_10ms_modulation_mae_db": 0.7411632136411527, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2202225842509922, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12281839361942345, "1000–3000 Hz_40ms_envelope_correlation": 0.9879983084505133, "1000–3000 Hz_40ms_modulation_mae_db": 0.9077254269604452, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2625657807624435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24579457930741502, "3000–6000 Hz_10ms_envelope_correlation": 0.9602322047083989, "3000–6000 Hz_10ms_modulation_mae_db": 0.5398503639162838, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9719699825121877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1581147172482575, "3000–6000 Hz_40ms_envelope_correlation": 0.9827509181360768, "3000–6000 Hz_40ms_modulation_mae_db": 0.5318446966080478, "mr_stft_db": 5.588215469432519, "band_mae_db": 0.17118698260753876, "rms_gain_error_db": -0.173867223858213, "output_peak": 0.47099752112623944}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0206719437500675, "全帯域_10ms_rms_envelope_nrmse": 0.091532659247925, "全帯域_10ms_envelope_correlation": 0.9926705344599293, "全帯域_10ms_modulation_mae_db": 0.09802409612778284, "全帯域_40ms_log_rms_mae_db": 0.5840254815603517, "全帯域_40ms_rms_envelope_nrmse": 0.0523503652923107, "全帯域_40ms_envelope_correlation": 0.9975011955902188, "全帯域_40ms_modulation_mae_db": 0.18487982803345004, "300–1000 Hz_10ms_log_rms_mae_db": 1.6578223296102346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14300797254980754, "300–1000 Hz_10ms_envelope_correlation": 0.9859888211519833, "300–1000 Hz_10ms_modulation_mae_db": 0.2698736020625461, "300–1000 Hz_40ms_log_rms_mae_db": 1.0551003212612042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0882363203481154, "300–1000 Hz_40ms_envelope_correlation": 0.9949318822805039, "300–1000 Hz_40ms_modulation_mae_db": 0.6591922458281753, "1000–3000 Hz_10ms_log_rms_mae_db": 1.825254490040084, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19749699977911273, "1000–3000 Hz_10ms_envelope_correlation": 0.9695809453488529, "1000–3000 Hz_10ms_modulation_mae_db": 0.39388743901884476, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2084627265952192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11555232935652254, "1000–3000 Hz_40ms_envelope_correlation": 0.9887224821765005, "1000–3000 Hz_40ms_modulation_mae_db": 0.5426196086597009, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2580751250473194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23742688801272438, "3000–6000 Hz_10ms_envelope_correlation": 0.963561775866444, "3000–6000 Hz_10ms_modulation_mae_db": 0.4618555254216854, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8995453692620674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15607356458687982, "3000–6000 Hz_40ms_envelope_correlation": 0.9835546484701867, "3000–6000 Hz_40ms_modulation_mae_db": 0.3263397193335884, "mr_stft_db": 5.590291824906134, "band_mae_db": 0.13093622820880474, "rms_gain_error_db": -0.08532833810525572, "output_peak": 0.48385814787803566}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0626324420594586, "全帯域_10ms_rms_envelope_nrmse": 0.09886289553996794, "全帯域_10ms_envelope_correlation": 0.9914683342981532, "全帯域_10ms_modulation_mae_db": 0.2092155542872004, "全帯域_40ms_log_rms_mae_db": 0.580026827776263, "全帯域_40ms_rms_envelope_nrmse": 0.054443158699693565, "全帯域_40ms_envelope_correlation": 0.997394771220812, "全帯域_40ms_modulation_mae_db": 0.1730535343076203, "300–1000 Hz_10ms_log_rms_mae_db": 1.6471731565214762, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15009327624498692, "300–1000 Hz_10ms_envelope_correlation": 0.9844801860501704, "300–1000 Hz_10ms_modulation_mae_db": 0.3010042367296066, "300–1000 Hz_40ms_log_rms_mae_db": 0.9871810644221191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08826194952616394, "300–1000 Hz_40ms_envelope_correlation": 0.99496993596685, "300–1000 Hz_40ms_modulation_mae_db": 0.2079847540520624, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8296603672240304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2336558915945088, "1000–3000 Hz_10ms_envelope_correlation": 0.9572560115778207, "1000–3000 Hz_10ms_modulation_mae_db": 0.8212226375827152, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2636615274583136, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14388765628835995, "1000–3000 Hz_40ms_envelope_correlation": 0.9835773234172625, "1000–3000 Hz_40ms_modulation_mae_db": 1.005254316490382, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1816699653705593, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22750286761288735, "3000–6000 Hz_10ms_envelope_correlation": 0.9657426297764908, "3000–6000 Hz_10ms_modulation_mae_db": 0.5069716643317126, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9232027828720453, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15367827920788327, "3000–6000 Hz_40ms_envelope_correlation": 0.9837372640425279, "3000–6000 Hz_40ms_modulation_mae_db": 0.48572993212948345, "mr_stft_db": 5.5770694335242945, "band_mae_db": 0.17344826128132032, "rms_gain_error_db": -0.13244572420248932, "output_peak": 0.5093584601053812}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9565234522879603, "全帯域_10ms_rms_envelope_nrmse": 0.08455642617915769, "全帯域_10ms_envelope_correlation": 0.9937489161861393, "全帯域_10ms_modulation_mae_db": 0.33114589894423296, "全帯域_40ms_log_rms_mae_db": 0.5154340046103862, "全帯域_40ms_rms_envelope_nrmse": 0.04679986617726871, "全帯域_40ms_envelope_correlation": 0.998026332158533, "全帯域_40ms_modulation_mae_db": 0.35506974198237645, "300–1000 Hz_10ms_log_rms_mae_db": 1.6377458878588782, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13429910726057945, "300–1000 Hz_10ms_envelope_correlation": 0.987305529089181, "300–1000 Hz_10ms_modulation_mae_db": 0.5705499651477526, "300–1000 Hz_40ms_log_rms_mae_db": 0.9888322758203396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07481382682581919, "300–1000 Hz_40ms_envelope_correlation": 0.9959669967296765, "300–1000 Hz_40ms_modulation_mae_db": 1.0207422480674266, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8136916068419335, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20674714143854, "1000–3000 Hz_10ms_envelope_correlation": 0.9665582396707747, "1000–3000 Hz_10ms_modulation_mae_db": 0.6837162664160555, "1000–3000 Hz_40ms_log_rms_mae_db": 1.261586555889033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1331328758749222, "1000–3000 Hz_40ms_envelope_correlation": 0.9853380003824798, "1000–3000 Hz_40ms_modulation_mae_db": 0.7428508248525066, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1795046651877747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2347369559388719, "3000–6000 Hz_10ms_envelope_correlation": 0.9640602974523157, "3000–6000 Hz_10ms_modulation_mae_db": 0.44866734674297604, "3000–6000 Hz_40ms_log_rms_mae_db": 1.892731221332942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1574527884755463, "3000–6000 Hz_40ms_envelope_correlation": 0.9829541463080852, "3000–6000 Hz_40ms_modulation_mae_db": 0.509912952054916, "mr_stft_db": 5.582295228158858, "band_mae_db": 0.1396014675647898, "rms_gain_error_db": -0.08884285499211575, "output_peak": 0.4639334000973583}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9757194235306285, "全帯域_10ms_rms_envelope_nrmse": 0.10302900529882841, "全帯域_10ms_envelope_correlation": 0.9906704868169246, "全帯域_10ms_modulation_mae_db": 0.4056064742066512, "全帯域_40ms_log_rms_mae_db": 0.49221721954559644, "全帯域_40ms_rms_envelope_nrmse": 0.059149356737781426, "全帯域_40ms_envelope_correlation": 0.9967801713826212, "全帯域_40ms_modulation_mae_db": 0.20542333820294223, "300–1000 Hz_10ms_log_rms_mae_db": 1.6616486498691827, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18211217133238788, "300–1000 Hz_10ms_envelope_correlation": 0.9767124242475985, "300–1000 Hz_10ms_modulation_mae_db": 0.3104058832684744, "300–1000 Hz_40ms_log_rms_mae_db": 1.0083588874350373, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11268478254590249, "300–1000 Hz_40ms_envelope_correlation": 0.9911947843757843, "300–1000 Hz_40ms_modulation_mae_db": 0.2802440465128056, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8307541397920009, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20942495845837902, "1000–3000 Hz_10ms_envelope_correlation": 0.9656585786635322, "1000–3000 Hz_10ms_modulation_mae_db": 0.6442431316012538, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3048600968979047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12727917733208569, "1000–3000 Hz_40ms_envelope_correlation": 0.9863414359784675, "1000–3000 Hz_40ms_modulation_mae_db": 0.4696122408470105, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1760825714025116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22460272578386198, "3000–6000 Hz_10ms_envelope_correlation": 0.9672168689588262, "3000–6000 Hz_10ms_modulation_mae_db": 0.14574812874106946, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8783069478744172, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1537520282829881, "3000–6000 Hz_40ms_envelope_correlation": 0.9837713850036397, "3000–6000 Hz_40ms_modulation_mae_db": 0.19815596512742883, "mr_stft_db": 5.583831697689184, "band_mae_db": 0.14168450054200132, "rms_gain_error_db": -0.10118479995834816, "output_peak": 0.4873049004019477}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9900068967545341, "全帯域_10ms_rms_envelope_nrmse": 0.09613769107844963, "全帯域_10ms_envelope_correlation": 0.9918966261576407, "全帯域_10ms_modulation_mae_db": 0.12579648774380195, "全帯域_40ms_log_rms_mae_db": 0.5626941305662424, "全帯域_40ms_rms_envelope_nrmse": 0.05966375439625993, "全帯域_40ms_envelope_correlation": 0.9967492250011515, "全帯域_40ms_modulation_mae_db": 0.13944985710285548, "300–1000 Hz_10ms_log_rms_mae_db": 1.6919978394065676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14920544702657043, "300–1000 Hz_10ms_envelope_correlation": 0.9843654456278162, "300–1000 Hz_10ms_modulation_mae_db": 0.3624400189827719, "300–1000 Hz_40ms_log_rms_mae_db": 1.0426264555996343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08610919631964191, "300–1000 Hz_40ms_envelope_correlation": 0.9947784426520037, "300–1000 Hz_40ms_modulation_mae_db": 0.6443826594995407, "1000–3000 Hz_10ms_log_rms_mae_db": 1.784349329967435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21999698018015862, "1000–3000 Hz_10ms_envelope_correlation": 0.963333874202728, "1000–3000 Hz_10ms_modulation_mae_db": 0.49628736985231564, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1682971904005213, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13191975886619214, "1000–3000 Hz_40ms_envelope_correlation": 0.9856967858783979, "1000–3000 Hz_40ms_modulation_mae_db": 0.5331949384016487, "3000–6000 Hz_10ms_log_rms_mae_db": 2.16966235956036, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21697166998877987, "3000–6000 Hz_10ms_envelope_correlation": 0.9695890693608188, "3000–6000 Hz_10ms_modulation_mae_db": 0.19143935215597738, "3000–6000 Hz_40ms_log_rms_mae_db": 1.866327680528179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15052149568251638, "3000–6000 Hz_40ms_envelope_correlation": 0.9847802399815342, "3000–6000 Hz_40ms_modulation_mae_db": 0.2040814734498897, "mr_stft_db": 5.582443721468447, "band_mae_db": 0.10033111095095804, "rms_gain_error_db": -0.10910457861205884, "output_peak": 0.46263359664782905}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59760, "output_samples": 59760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.747039244140863, "spectral_convergence": 0.2952039111890726, "spectral_convergence_status": "ok", "amplitude_floor": 1.6701466153856754e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.433452210199866, "spectral_convergence": 0.2911227330014557, "spectral_convergence_status": "ok", "amplitude_floor": 1.1197689318182202e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.582037508050423, "spectral_convergence": 0.30431335436833074, "spectral_convergence_status": "ok", "amplitude_floor": 1.0416079453540003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0010765709682483164, "output_power": 0.0010659248000099823, "delta_db": -0.04316097368899355, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019015067355649686, "output_power": 0.0018705675891270961, "delta_db": -0.07124462960176076, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009390993860770502, "output_power": 0.0009206754450823676, "delta_db": -0.08604996198682859, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028731828855173817, "output_power": 0.00026693462315803665, "delta_db": -0.31958362576063193, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.911907847527107e-05, "output_power": 5.907214114334772e-05, "delta_db": -0.0034494314191841147, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.5637714608274427e-05, "output_power": 6.123566815254079e-05, "delta_db": 0.4163517838991659, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15664006772609415, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07591203900179223, "pitch": null, "output_rms_dbfs": -23.72182643634796, "output_peak": 0.5019204180729975, "reference_rms_dbfs": -23.64591439734617, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5875096541303835}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00019933639001390812, "log_rms_mae_db": 1.0024887062011667, "rms_envelope_nrmse": 0.09347200124155028, "envelope_correlation": 0.9923087515036229, "rms_gain_error_db": -0.07591205007516111, "reference_mean_power": 0.004325042032530898, "output_mean_power": 0.004250099792941194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 2.983288998803909e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014406782878671795, "output_power": 0.0013756889057598629, "reference_active": true, "delta_db": -0.20046776142422254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010200208961765019, "output_power": 0.00010393258464884826, "reference_active": true, "delta_db": 0.08142658856681752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7770617096993514e-05, "output_power": 3.609338869317696e-05, "reference_active": true, "delta_db": 1.1384212808710297}], "active_band_count": 3, "active_band_mae_db": 0.4734385436206899}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00018454723874230657, "log_rms_mae_db": 0.515987775763132, "rms_envelope_nrmse": 0.05280500980488158, "envelope_correlation": 0.9974131047059355, "rms_gain_error_db": -0.07591209092921605, "reference_mean_power": 0.004360109839005934, "output_mean_power": 0.004284559921096563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 2.740771226026708e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013018843976497208, "output_power": 0.0012510111149961114, "reference_active": true, "delta_db": -0.173112539223847}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649956473480408e-05, "output_power": 4.905221884820653e-05, "reference_active": true, "delta_db": 0.23209769575942357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9034855452992526e-06, "output_power": 2.3758540381748254e-06, "reference_active": true, "delta_db": 0.9627017280726022}], "active_band_count": 3, "active_band_mae_db": 0.45597065435195755}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00017949843182759416, "log_rms_mae_db": 1.5660031770957128, "rms_envelope_nrmse": 0.14100366763065625, "envelope_correlation": 0.9859831340261657, "rms_gain_error_db": -0.07116056459597986, "reference_mean_power": 0.001722884360599687, "output_mean_power": 0.0016948843605157015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.0185329105758934e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006952905258544651, "output_power": 0.000657629152311052, "reference_active": true, "delta_db": -0.24185254680315776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.45148729518806e-05, "output_power": 5.8454936253948636e-05, "reference_active": true, "delta_db": -0.4283865511786927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5243024554692387e-05, "output_power": 2.731180793183059e-05, "reference_active": true, "delta_db": 2.5327930041208937}], "active_band_count": 3, "active_band_mae_db": 1.0676773673675815}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00016126256915979383, "log_rms_mae_db": 0.9099463931013742, "rms_envelope_nrmse": 0.0800270198954429, "envelope_correlation": 0.9952466260671909, "rms_gain_error_db": -0.07116057039137465, "reference_mean_power": 0.0017368536839123, "output_mean_power": 0.0017086266546723552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 9.447738359038237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006450702777726167, "output_power": 0.0006195119375276623, "reference_active": true, "delta_db": -0.1755735250344662}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9818892488309525e-05, "output_power": 2.656219170350774e-05, "reference_active": true, "delta_db": -0.5022760242040876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0738004145337803e-06, "output_power": 1.72690268831044e-06, "reference_active": true, "delta_db": 2.0634429829719876}], "active_band_count": 3, "active_band_mae_db": 0.9137641774035138}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 6.626476450042254e-05, "log_rms_mae_db": 1.8134303085390568, "rms_envelope_nrmse": 0.21310724667926015, "envelope_correlation": 0.9643944184996958, "rms_gain_error_db": -0.20811403765565398, "reference_mean_power": 0.00025607969857314127, "output_mean_power": 0.00024409773307977424}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.9689769875456006e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013521942671011384, "output_power": 0.00011203617306895921, "reference_active": true, "delta_db": -0.8168082454928254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.463778228624424e-05, "output_power": 1.2679103135338872e-05, "reference_active": true, "delta_db": -0.6238674890210545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5896173962011905e-06, "output_power": 5.0335373397679796e-06, "reference_active": true, "delta_db": -0.4550878723367752}], "active_band_count": 3, "active_band_mae_db": 0.6319212022835518}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 5.564377890100535e-05, "log_rms_mae_db": 1.2331957019672128, "rms_envelope_nrmse": 0.14669840527389713, "envelope_correlation": 0.9819752436192398, "rms_gain_error_db": -0.20811403766088576, "reference_mean_power": 0.0002581560174430691, "output_mean_power": 0.00024607690101869414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 1.774101433771277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012602347588101508, "output_power": 0.00010409363045861298, "reference_active": true, "delta_db": -0.8302729818858479}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.970334241514461e-06, "output_power": 6.017602869142235e-06, "reference_active": true, "delta_db": -0.6383008074293673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.812565026245218e-07, "output_power": 3.1412952679794024e-07, "reference_active": true, "delta_db": -0.8410849955238014}], "active_band_count": 3, "active_band_mae_db": 0.7698862616130056}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 3.541043489510122e-05, "log_rms_mae_db": 2.213003652470715, "rms_envelope_nrmse": 0.22829651881878418, "envelope_correlation": 0.967784895832316, "rms_gain_error_db": 0.30893850025764175, "reference_mean_power": 6.312530293815556e-05, "output_mean_power": 6.77793383999315e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 4.34877201852775e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2352362117577493e-05, "output_power": 2.3645246707745038e-05, "reference_active": true, "delta_db": 0.24420425100946777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.837592049886246e-06, "output_power": 9.72039283309131e-06, "reference_active": true, "delta_db": 0.4134986623309257}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1202466049559064e-06, "output_power": 2.3722862490496553e-06, "reference_active": true, "delta_db": 0.48780715004331926}], "active_band_count": 3, "active_band_mae_db": 0.3818366877945709}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 3.3704303759041684e-05, "log_rms_mae_db": 1.915395066120604, "rms_envelope_nrmse": 0.1587821367669744, "envelope_correlation": 0.9839795205201757, "rms_gain_error_db": 0.30893852888728124, "reference_mean_power": 6.363712790649308e-05, "output_mean_power": 6.832889910776002e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 3.9430492064624054e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.36389645521954e-05, "output_power": 2.5115884555658202e-05, "reference_active": true, "delta_db": 0.26320028766566533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.125341563464706e-06, "output_power": 4.4142664160084224e-06, "reference_active": true, "delta_db": 0.29398628369459145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9054004218332769e-07, "output_power": 2.1204609499218098e-07, "reference_active": true, "delta_db": 0.4644402194063508}], "active_band_count": 3, "active_band_mae_db": 0.34054226358886924}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8527406166189484, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "56a37445d1c210a106cfb38662d5d4e7b3869e68e466daa6f161c23d014cd8ee", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9650648331612178, "standard_deviation": 0.0044375744553273215, "minimum": 0.9572560115778207, "maximum": 0.9715554117630024, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.818901019067713, "standard_deviation": 0.028271790015283127, "minimum": 1.7827888594503645, "maximum": 1.8712790506868002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6310735741431674, "standard_deviation": 0.13398940811407511, "minimum": 0.39388743901884476, "maximum": 0.8212226375827152, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21147033708214627, "standard_deviation": 0.013400900327939017, "minimum": 0.19151758449805426, "maximum": 0.2336558915945088, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9854532095029086, "standard_deviation": 0.002262245425944607, "minimum": 0.9819752436192398, "maximum": 0.9887224821765005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2339174749351787, "standard_deviation": 0.041999905518227174, "minimum": 1.1682971904005213, "maximum": 1.3048600968979047, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6873333234189123, "standard_deviation": 0.19899251159332113, "minimum": 0.4696122408470105, "maximum": 1.005254316490382, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13273429845465218, "standard_deviation": 0.010724907737028482, "minimum": 0.11555232935652254, "maximum": 0.14669840527389713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9656457904425451, "standard_deviation": 0.0029409190391034216, "minimum": 0.9602322047083989, "maximum": 0.9695890693608188, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.21129904941756, "standard_deviation": 0.039939714726866096, "minimum": 2.16966235956036, "maximum": 2.2625657807624435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3511768173776911, "standard_deviation": 0.16812505445156647, "minimum": 0.1330454699172527, "maximum": 0.5398503639162838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22987047479183476, "standard_deviation": 0.009067162242142558, "minimum": 0.21697166998877987, "maximum": 0.24579457930741502, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9836905831671658, "standard_deviation": 0.0006343662766609305, "minimum": 0.9827509181360768, "maximum": 0.9847802399815342, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.9086249556270816, "standard_deviation": 0.032690164816076846, "minimum": 1.866327680528179, "maximum": 1.9719699825121877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.34664143526257746, "standard_deviation": 0.14738863196152183, "minimum": 0.17652447980839625, "maximum": 0.5318446966080478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1550736187710738, "standard_deviation": 0.0029822064569849155, "minimum": 0.15052149568251638, "maximum": 0.1587821367669744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9840547610377277, "standard_deviation": 0.0032519948301446074, "minimum": 0.9767124242475985, "maximum": 0.987305529089181, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6573251642973108, "standard_deviation": 0.055633580189179056, "minimum": 1.5660031770957128, "maximum": 1.7624961441047193, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6823932620184054, "standard_deviation": 0.46960353856926695, "minimum": 0.2698736020625461, "maximum": 1.5023703923485792, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15080069139271318, "standard_deviation": 0.014389773511478354, "minimum": 0.13429910726057945, "maximum": 0.18211217133238788, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946039788157014, "standard_deviation": 0.001445819678668292, "minimum": 0.9911947843757843, "maximum": 0.9959669967296765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0095849168189903, "standard_deviation": 0.05334561269980377, "minimum": 0.9099463931013742, "maximum": 1.0856307523916249, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7037505929757872, "standard_deviation": 0.32097826236698895, "minimum": 0.2079847540520624, "maximum": 1.0587219555306373, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08858708899947323, "standard_deviation": 0.011263323662857082, "minimum": 0.07481382682581919, "maximum": 0.11268478254590249, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1482181929198143, "standard_deviation": 0.025369992166264, "minimum": 0.10033111095095804, "maximum": 0.17344826128132032, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.585336390727165, "standard_deviation": 0.004754775566471238, "minimum": 5.5770694335242945, "maximum": 5.591034096507499, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.48082467421063657, "standard_deviation": 0.01783177318978, "minimum": 0.46263359664782905, "maximum": 0.5093584601053812, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11485306224452277, "standard_deviation": 0.03477872833871168, "minimum": -0.173867223858213, "maximum": -0.07591203900179223, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920023170661242, "standard_deviation": 0.0009391674218029932, "minimum": 0.9906704868169246, "maximum": 0.9937489161861393, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.005954015603296, "standard_deviation": 0.035724261828069696, "minimum": 0.9565234522879603, "maximum": 1.0626324420594586, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.302713399746245, "standard_deviation": 0.14630513101881953, "minimum": 0.09802409612778284, "maximum": 0.4734385436206899, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.095540703715961, "standard_deviation": 0.005738625370894379, "minimum": 0.08455642617915769, "maximum": 0.10302900529882841, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972394371827601, "standard_deviation": 0.00046714666289265916, "minimum": 0.9966770162596668, "maximum": 0.998026332158533, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5479413220834217, "standard_deviation": 0.03499494645209594, "minimum": 0.49221721954559644, "maximum": 0.5840254815603517, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2606827998128112, "standard_deviation": 0.10666787342500764, "minimum": 0.13944985710285548, "maximum": 0.45597065435195755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055294040969016064, "standard_deviation": 0.0048939702848420875, "minimum": 0.04679986617726871, "maximum": 0.06206998588302105, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123286-0008", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6135973930358887, "predicted_mos": 2.6135973930358887, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "7466a3ff4dbd4e2ecd1c863f2d1b5e00ca53376b2fba5de4241ae93d46fee142", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.414061299990863, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9775937420465108, "全帯域_10ms_rms_envelope_nrmse": 0.08567822362785257, "全帯域_10ms_envelope_correlation": 0.9936735391477467, "全帯域_10ms_modulation_mae_db": 0.42032905833752476, "全帯域_40ms_log_rms_mae_db": 0.49371216454640793, "全帯域_40ms_rms_envelope_nrmse": 0.04410179700539794, "全帯域_40ms_envelope_correlation": 0.9983894930160713, "全帯域_40ms_modulation_mae_db": 0.45282144205320335, "300–1000 Hz_10ms_log_rms_mae_db": 1.5128839276938415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12741058954882623, "300–1000 Hz_10ms_envelope_correlation": 0.9886972277214297, "300–1000 Hz_10ms_modulation_mae_db": 0.6099802170013096, "300–1000 Hz_40ms_log_rms_mae_db": 0.8876509710495756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061230786089973205, "300–1000 Hz_40ms_envelope_correlation": 0.997547173799546, "300–1000 Hz_40ms_modulation_mae_db": 0.38758915840696045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.661692513539925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18133792908815735, "1000–3000 Hz_10ms_envelope_correlation": 0.9744783549779096, "1000–3000 Hz_10ms_modulation_mae_db": 0.5653163241106346, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1078287264077007, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11777893975747003, "1000–3000 Hz_40ms_envelope_correlation": 0.988732732774181, "1000–3000 Hz_40ms_modulation_mae_db": 0.8676018720424602, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1855943221311667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2256932363619981, "3000–6000 Hz_10ms_envelope_correlation": 0.9668153334077858, "3000–6000 Hz_10ms_modulation_mae_db": 0.28729603580090324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8121730337584268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15169987738083582, "3000–6000 Hz_40ms_envelope_correlation": 0.9843883819419124, "3000–6000 Hz_40ms_modulation_mae_db": 0.4003816259219628, "mr_stft_db": 5.558288162977693, "band_mae_db": 0.14035133508833117, "rms_gain_error_db": -0.14404772129539345, "output_peak": 0.48136441014293085}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0313725756504024, "全帯域_10ms_rms_envelope_nrmse": 0.1049686918761357, "全帯域_10ms_envelope_correlation": 0.9903408105224517, "全帯域_10ms_modulation_mae_db": 0.4512209379500707, "全帯域_40ms_log_rms_mae_db": 0.5645215783358791, "全帯域_40ms_rms_envelope_nrmse": 0.05086317149205071, "全帯域_40ms_envelope_correlation": 0.9975610015921531, "全帯域_40ms_modulation_mae_db": 0.5665682190703111, "300–1000 Hz_10ms_log_rms_mae_db": 1.6952727244292014, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16151998286997954, "300–1000 Hz_10ms_envelope_correlation": 0.9816496183775503, "300–1000 Hz_10ms_modulation_mae_db": 0.6538483712587332, "300–1000 Hz_40ms_log_rms_mae_db": 1.002786269762905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08075936220943786, "300–1000 Hz_40ms_envelope_correlation": 0.9951783880520081, "300–1000 Hz_40ms_modulation_mae_db": 0.4827679904403399, "1000–3000 Hz_10ms_log_rms_mae_db": 1.777081300636661, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2081208433009762, "1000–3000 Hz_10ms_envelope_correlation": 0.9669412312087227, "1000–3000 Hz_10ms_modulation_mae_db": 0.7609807347685497, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1970550956294395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12420600670188424, "1000–3000 Hz_40ms_envelope_correlation": 0.9872326044189842, "1000–3000 Hz_40ms_modulation_mae_db": 0.9070943148839974, "3000–6000 Hz_10ms_log_rms_mae_db": 2.148110785864551, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22960721956020758, "3000–6000 Hz_10ms_envelope_correlation": 0.9650100935443126, "3000–6000 Hz_10ms_modulation_mae_db": 0.36999619061886974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8417473512030562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15026847602487897, "3000–6000 Hz_40ms_envelope_correlation": 0.9845156030472604, "3000–6000 Hz_40ms_modulation_mae_db": 0.4325129918233199, "mr_stft_db": 5.5529175866524225, "band_mae_db": 0.06463856019070958, "rms_gain_error_db": -0.01755556469647379, "output_peak": 0.48188495468804304}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9986171756806389, "全帯域_10ms_rms_envelope_nrmse": 0.08854309263026022, "全帯域_10ms_envelope_correlation": 0.9930957644289202, "全帯域_10ms_modulation_mae_db": 0.21845813906897524, "全帯域_40ms_log_rms_mae_db": 0.5046972367622807, "全帯域_40ms_rms_envelope_nrmse": 0.04368954772197075, "全帯域_40ms_envelope_correlation": 0.9981953541752875, "全帯域_40ms_modulation_mae_db": 0.14491297248753138, "300–1000 Hz_10ms_log_rms_mae_db": 1.6223556099520808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14150203576194104, "300–1000 Hz_10ms_envelope_correlation": 0.9858686593423607, "300–1000 Hz_10ms_modulation_mae_db": 0.5960592502236107, "300–1000 Hz_40ms_log_rms_mae_db": 0.9501586173595419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07735301474060853, "300–1000 Hz_40ms_envelope_correlation": 0.995606976361575, "300–1000 Hz_40ms_modulation_mae_db": 0.4059671342437179, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7717696712256692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2156784930339443, "1000–3000 Hz_10ms_envelope_correlation": 0.9637899320767287, "1000–3000 Hz_10ms_modulation_mae_db": 0.45676382361937934, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2378692454635023, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14646528714722332, "1000–3000 Hz_40ms_envelope_correlation": 0.9827541050514064, "1000–3000 Hz_40ms_modulation_mae_db": 0.6829339306449617, "3000–6000 Hz_10ms_log_rms_mae_db": 2.204192848156128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22271492230065063, "3000–6000 Hz_10ms_envelope_correlation": 0.9681803403458704, "3000–6000 Hz_10ms_modulation_mae_db": 0.2878092892675762, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9087344677305202, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15039817933525804, "3000–6000 Hz_40ms_envelope_correlation": 0.9848158474816053, "3000–6000 Hz_40ms_modulation_mae_db": 0.30175410350592885, "mr_stft_db": 5.558942885904429, "band_mae_db": 0.10625108980633446, "rms_gain_error_db": -0.02089840431727461, "output_peak": 0.5128312416176727}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9817690145399892, "全帯域_10ms_rms_envelope_nrmse": 0.08820712871150445, "全帯域_10ms_envelope_correlation": 0.9931978728170698, "全帯域_10ms_modulation_mae_db": 0.29934729847762737, "全帯域_40ms_log_rms_mae_db": 0.49860334417580127, "全帯域_40ms_rms_envelope_nrmse": 0.046613887317253126, "全帯域_40ms_envelope_correlation": 0.9980651634848638, "全帯域_40ms_modulation_mae_db": 0.21933708112827768, "300–1000 Hz_10ms_log_rms_mae_db": 1.6253730504669037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15979783849443416, "300–1000 Hz_10ms_envelope_correlation": 0.9822861733277112, "300–1000 Hz_10ms_modulation_mae_db": 0.17567426732010438, "300–1000 Hz_40ms_log_rms_mae_db": 0.9571574820315234, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08755034233362677, "300–1000 Hz_40ms_envelope_correlation": 0.9950612174544895, "300–1000 Hz_40ms_modulation_mae_db": 0.4074210219328163, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7760043247961992, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18042629610818647, "1000–3000 Hz_10ms_envelope_correlation": 0.9747004734973086, "1000–3000 Hz_10ms_modulation_mae_db": 0.4148666400825114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1876093950423678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.115257536997656, "1000–3000 Hz_40ms_envelope_correlation": 0.9889311641030464, "1000–3000 Hz_40ms_modulation_mae_db": 0.7486106030916106, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1929926402694435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2276292995365088, "3000–6000 Hz_10ms_envelope_correlation": 0.9662713113708715, "3000–6000 Hz_10ms_modulation_mae_db": 0.4656704999050673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8735119846681303, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15377743236296504, "3000–6000 Hz_40ms_envelope_correlation": 0.9837989547926271, "3000–6000 Hz_40ms_modulation_mae_db": 0.4199059810025288, "mr_stft_db": 5.563252588119442, "band_mae_db": 0.09669096643708602, "rms_gain_error_db": -0.10384975904950718, "output_peak": 0.4565126662685395}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0306428039032487, "全帯域_10ms_rms_envelope_nrmse": 0.08835099883217978, "全帯域_10ms_envelope_correlation": 0.9931494609864057, "全帯域_10ms_modulation_mae_db": 0.1575581765943573, "全帯域_40ms_log_rms_mae_db": 0.621164123789859, "全帯域_40ms_rms_envelope_nrmse": 0.05093510877097236, "全帯域_40ms_envelope_correlation": 0.9976178193116028, "全帯域_40ms_modulation_mae_db": 0.13511021967944184, "300–1000 Hz_10ms_log_rms_mae_db": 1.5289714935600702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1332093744914211, "300–1000 Hz_10ms_envelope_correlation": 0.9875910882980512, "300–1000 Hz_10ms_modulation_mae_db": 0.6056572333878624, "300–1000 Hz_40ms_log_rms_mae_db": 0.86847955859658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08929713354998202, "300–1000 Hz_40ms_envelope_correlation": 0.9943127077770308, "300–1000 Hz_40ms_modulation_mae_db": 0.5173036420576923, "1000–3000 Hz_10ms_log_rms_mae_db": 1.794007236138612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19150031234798684, "1000–3000 Hz_10ms_envelope_correlation": 0.971420797956538, "1000–3000 Hz_10ms_modulation_mae_db": 0.5873820950086276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2203275584821334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1251802140211744, "1000–3000 Hz_40ms_envelope_correlation": 0.9871068844080666, "1000–3000 Hz_40ms_modulation_mae_db": 0.7886693339248212, "3000–6000 Hz_10ms_log_rms_mae_db": 2.183309721855085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2352175849892082, "3000–6000 Hz_10ms_envelope_correlation": 0.963225201505727, "3000–6000 Hz_10ms_modulation_mae_db": 0.44321106849472264, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8804693028823067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1596426479805058, "3000–6000 Hz_40ms_envelope_correlation": 0.9824093053648609, "3000–6000 Hz_40ms_modulation_mae_db": 0.4953148735653073, "mr_stft_db": 5.555615918160787, "band_mae_db": 0.09935069300538323, "rms_gain_error_db": -0.08829997214725283, "output_peak": 0.48756163823789234}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0081737980083285, "全帯域_10ms_rms_envelope_nrmse": 0.08840659030829258, "全帯域_10ms_envelope_correlation": 0.9931515954062854, "全帯域_10ms_modulation_mae_db": 0.2537324500689618, "全帯域_40ms_log_rms_mae_db": 0.5182645456546078, "全帯域_40ms_rms_envelope_nrmse": 0.04397115073138651, "全帯域_40ms_envelope_correlation": 0.9982709118143807, "全帯域_40ms_modulation_mae_db": 0.5023681898441333, "300–1000 Hz_10ms_log_rms_mae_db": 1.5975845473115438, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1594558885989612, "300–1000 Hz_10ms_envelope_correlation": 0.9820278519661305, "300–1000 Hz_10ms_modulation_mae_db": 0.9073954297204594, "300–1000 Hz_40ms_log_rms_mae_db": 0.94195491562059, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08727654740397406, "300–1000 Hz_40ms_envelope_correlation": 0.9945100584820284, "300–1000 Hz_40ms_modulation_mae_db": 0.5101648149931884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.830321189217636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2053669632464936, "1000–3000 Hz_10ms_envelope_correlation": 0.9673753021624195, "1000–3000 Hz_10ms_modulation_mae_db": 0.4513314527070234, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2223871374510984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12150226683781634, "1000–3000 Hz_40ms_envelope_correlation": 0.987627704459928, "1000–3000 Hz_40ms_modulation_mae_db": 0.3927228710247534, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1670475673443597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23483640873125863, "3000–6000 Hz_10ms_envelope_correlation": 0.9680709930927134, "3000–6000 Hz_10ms_modulation_mae_db": 0.8185718326475099, "3000–6000 Hz_40ms_log_rms_mae_db": 1.882892929353174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16418776844796246, "3000–6000 Hz_40ms_envelope_correlation": 0.9846240184512488, "3000–6000 Hz_40ms_modulation_mae_db": 0.7881910424911777, "mr_stft_db": 5.569205116131671, "band_mae_db": 0.1576159963031045, "rms_gain_error_db": -0.0832382581773932, "output_peak": 0.4612310960229919}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9353277193830831, "全帯域_10ms_rms_envelope_nrmse": 0.08658628867067555, "全帯域_10ms_envelope_correlation": 0.9934002268248097, "全帯域_10ms_modulation_mae_db": 0.2598713588830613, "全帯域_40ms_log_rms_mae_db": 0.5040148885558745, "全帯域_40ms_rms_envelope_nrmse": 0.04563788633144866, "全帯域_40ms_envelope_correlation": 0.9980699044857813, "全帯域_40ms_modulation_mae_db": 0.1270539044852122, "300–1000 Hz_10ms_log_rms_mae_db": 1.6598454958215103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1463539598325804, "300–1000 Hz_10ms_envelope_correlation": 0.9850961627317212, "300–1000 Hz_10ms_modulation_mae_db": 0.4488472584022871, "300–1000 Hz_40ms_log_rms_mae_db": 0.9732423731698516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08212347154482237, "300–1000 Hz_40ms_envelope_correlation": 0.9954928449205513, "300–1000 Hz_40ms_modulation_mae_db": 0.7500565467421869, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7675131348206583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21530395609377184, "1000–3000 Hz_10ms_envelope_correlation": 0.9643514518763607, "1000–3000 Hz_10ms_modulation_mae_db": 0.9983841011736144, "1000–3000 Hz_40ms_log_rms_mae_db": 1.218215887410137, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14628844477765315, "1000–3000 Hz_40ms_envelope_correlation": 0.9835019607089677, "1000–3000 Hz_40ms_modulation_mae_db": 1.0036982133054988, "3000–6000 Hz_10ms_log_rms_mae_db": 2.149693134373617, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23388821954551958, "3000–6000 Hz_10ms_envelope_correlation": 0.9643770857525127, "3000–6000 Hz_10ms_modulation_mae_db": 0.5930520828429128, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8840762778288174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15638820924811345, "3000–6000 Hz_40ms_envelope_correlation": 0.9832458636402512, "3000–6000 Hz_40ms_modulation_mae_db": 0.6821143098666406, "mr_stft_db": 5.5483555849905555, "band_mae_db": 0.178925535206898, "rms_gain_error_db": -0.06569827348990298, "output_peak": 0.46201520161172294}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0190621823907648, "全帯域_10ms_rms_envelope_nrmse": 0.09261272382467597, "全帯域_10ms_envelope_correlation": 0.992718927119936, "全帯域_10ms_modulation_mae_db": 0.3074991266227905, "全帯域_40ms_log_rms_mae_db": 0.5421193734052024, "全帯域_40ms_rms_envelope_nrmse": 0.04979107988212213, "全帯域_40ms_envelope_correlation": 0.9980830366141901, "全帯域_40ms_modulation_mae_db": 0.316412283894515, "300–1000 Hz_10ms_log_rms_mae_db": 1.6133136116288544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15343351048202247, "300–1000 Hz_10ms_envelope_correlation": 0.984031678442594, "300–1000 Hz_10ms_modulation_mae_db": 0.44477136884426954, "300–1000 Hz_40ms_log_rms_mae_db": 1.0117629829714376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08950786681115014, "300–1000 Hz_40ms_envelope_correlation": 0.9951804569007973, "300–1000 Hz_40ms_modulation_mae_db": 0.5876759120492233, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7550807922020606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21032491589045246, "1000–3000 Hz_10ms_envelope_correlation": 0.965349232011575, "1000–3000 Hz_10ms_modulation_mae_db": 0.3885494884225605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2222487836907654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12875037954835977, "1000–3000 Hz_40ms_envelope_correlation": 0.9863838390595389, "1000–3000 Hz_40ms_modulation_mae_db": 0.6830132740502103, "3000–6000 Hz_10ms_log_rms_mae_db": 2.222081450857259, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2530025498031009, "3000–6000 Hz_10ms_envelope_correlation": 0.9604384971136783, "3000–6000 Hz_10ms_modulation_mae_db": 0.5426045731937804, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8896818939517768, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17518071608613497, "3000–6000 Hz_40ms_envelope_correlation": 0.9802405160896799, "3000–6000 Hz_40ms_modulation_mae_db": 0.5696125191051467, "mr_stft_db": 5.5649031497004815, "band_mae_db": 0.20010047015671795, "rms_gain_error_db": -0.19426078973424607, "output_peak": 0.5004289183214654}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59760, "output_samples": 59760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.74291137466499, "spectral_convergence": 0.29070925644367884, "spectral_convergence_status": "ok", "amplitude_floor": 1.6701466153856754e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.381009918767072, "spectral_convergence": 0.29156560209177346, "spectral_convergence_status": "ok", "amplitude_floor": 1.1197689318182202e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.550943195501017, "spectral_convergence": 0.3122763315627482, "spectral_convergence_status": "ok", "amplitude_floor": 1.0416079453540003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0010765709682483164, "output_power": 0.0010453150054029092, "delta_db": -0.1279547937496624, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019015067355649686, "output_power": 0.0018506622090875596, "delta_db": -0.11770711449843936, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009390993860770502, "output_power": 0.0009031704765292063, "delta_db": -0.16941824007351416, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028731828855173817, "output_power": 0.0002636959204252039, "delta_db": -0.3725985985596347, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.911907847527107e-05, "output_power": 5.97797672293774e-05, "delta_db": 0.04826563819970266, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.5637714608274427e-05, "output_power": 5.5716733226047694e-05, "delta_db": 0.006163625449033939, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14035133508833117, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14404772129539345, "pitch": null, "output_rms_dbfs": -23.78996211864156, "output_peak": 0.48136441014293085, "reference_rms_dbfs": -23.64591439734617, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.558288162977693}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00019933639001390812, "log_rms_mae_db": 0.9775937420465108, "rms_envelope_nrmse": 0.08567822362785257, "envelope_correlation": 0.9936735391477467, "rms_gain_error_db": -0.14404772044232322, "reference_mean_power": 0.004325042032530898, "output_mean_power": 0.004183941084260726}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 2.983288998803909e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014406782878671795, "output_power": 0.0014121710588697906, "reference_active": true, "delta_db": -0.08679704279649719}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010200208961765019, "output_power": 0.00010879248628363352, "reference_active": true, "delta_db": 0.27989833174088385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7770617096993514e-05, "output_power": 3.4120446873189926e-05, "reference_active": true, "delta_db": 0.8942918004751933}], "active_band_count": 3, "active_band_mae_db": 0.42032905833752476}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00018454723874230657, "log_rms_mae_db": 0.49371216454640793, "rms_envelope_nrmse": 0.04410179700539794, "envelope_correlation": 0.9983894930160713, "rms_gain_error_db": -0.14404770230830505, "reference_mean_power": 0.004360109839005934, "output_mean_power": 0.004217864849929077}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 2.740771226026708e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013018843976497208, "output_power": 0.0012812738037874581, "reference_active": true, "delta_db": -0.0693047536342137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649956473480408e-05, "output_power": 5.1176490403256916e-05, "reference_active": true, "delta_db": 0.41621611755570886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9034855452992526e-06, "output_power": 2.3272547678241818e-06, "reference_active": true, "delta_db": 0.8729434549696875}], "active_band_count": 3, "active_band_mae_db": 0.45282144205320335}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00017949843182759416, "log_rms_mae_db": 1.5128839276938415, "rms_envelope_nrmse": 0.12741058954882623, "envelope_correlation": 0.9886972277214297, "rms_gain_error_db": -0.19636881411907708, "reference_mean_power": 0.001722884360599687, "output_mean_power": 0.0016467180553271766}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.0185329105758934e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006952905258544651, "output_power": 0.000655789331133612, "reference_active": true, "delta_db": -0.25401964689715173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.45148729518806e-05, "output_power": 6.289425473302872e-05, "reference_active": true, "delta_db": -0.1104887105443832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5243024554692387e-05, "output_power": 2.136064579656362e-05, "reference_active": true, "delta_db": 1.4654322935623938}], "active_band_count": 3, "active_band_mae_db": 0.6099802170013096}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00016126256915979383, "log_rms_mae_db": 0.8876509710495756, "rms_envelope_nrmse": 0.061230786089973205, "envelope_correlation": 0.997547173799546, "rms_gain_error_db": -0.1963688196037081, "reference_mean_power": 0.0017368536839123, "output_mean_power": 0.0016600698123189189}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 9.447738359038237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006450702777726167, "output_power": 0.0006125711581263418, "reference_active": true, "delta_db": -0.2245048689009564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9818892488309525e-05, "output_power": 2.9603798661285985e-05, "reference_active": true, "delta_db": -0.03144067303145153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0738004145337803e-06, "output_power": 1.3231398835225762e-06, "reference_active": true, "delta_db": 0.9068219332884736}], "active_band_count": 3, "active_band_mae_db": 0.38758915840696045}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 6.626476450042254e-05, "log_rms_mae_db": 1.661692513539925, "rms_envelope_nrmse": 0.18133792908815735, "envelope_correlation": 0.9744783549779096, "rms_gain_error_db": -0.2219390147868597, "reference_mean_power": 0.00025607969857314127, "output_mean_power": 0.00024332192769266616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.9689769875456006e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013521942671011384, "output_power": 0.00012353034210053538, "reference_active": true, "delta_db": -0.3926544618438329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.463778228624424e-05, "output_power": 1.3962641872480863e-05, "reference_active": true, "delta_db": -0.2050768447727107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5896173962011905e-06, "output_power": 4.340705436611838e-06, "reference_active": true, "delta_db": -1.0982176657153602}], "active_band_count": 3, "active_band_mae_db": 0.5653163241106346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 5.564377890100535e-05, "log_rms_mae_db": 1.1078287264077007, "rms_envelope_nrmse": 0.11777893975747003, "envelope_correlation": 0.988732732774181, "rms_gain_error_db": -0.2219390088718506, "reference_mean_power": 0.0002581560174430691, "output_mean_power": 0.00024529480566114003}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 1.774101433771277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012602347588101508, "output_power": 0.00011539984820254116, "reference_active": true, "delta_db": -0.3824621626801593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.970334241514461e-06, "output_power": 6.485105241079254e-06, "reference_active": true, "delta_db": -0.31336575628115176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.812565026245218e-07, "output_power": 2.457646921105649e-07, "reference_active": true, "delta_db": -1.9069776971660697}], "active_band_count": 3, "active_band_mae_db": 0.8676018720424602}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 3.541043489510122e-05, "log_rms_mae_db": 2.1855943221311667, "rms_envelope_nrmse": 0.2256932363619981, "envelope_correlation": 0.9668153334077858, "rms_gain_error_db": 0.04520961054505474, "reference_mean_power": 6.312530293815556e-05, "output_mean_power": 6.378586308986904e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 4.34877201852775e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2352362117577493e-05, "output_power": 2.147108357942261e-05, "reference_active": true, "delta_db": -0.17469462085122234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.837592049886246e-06, "output_power": 7.884541444612086e-06, "reference_active": true, "delta_db": -0.4955750996039973}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1202466049559064e-06, "output_power": 2.215890089700108e-06, "reference_active": true, "delta_db": 0.19161838694748998}], "active_band_count": 3, "active_band_mae_db": 0.28729603580090324}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 3.3704303759041684e-05, "log_rms_mae_db": 1.8121730337584268, "rms_envelope_nrmse": 0.15169987738083582, "envelope_correlation": 0.9843883819419124, "rms_gain_error_db": 0.04520963001734228, "reference_mean_power": 6.363712790649308e-05, "output_mean_power": 6.430304422067775e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 3.9430492064624054e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.36389645521954e-05, "output_power": 2.235267633015135e-05, "reference_active": true, "delta_db": -0.2429891988541371}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.125341563464706e-06, "output_power": 3.5469885422512035e-06, "reference_active": true, "delta_db": -0.6560012681434093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9054004218332769e-07, "output_power": 2.04268640323286e-07, "reference_active": true, "delta_db": 0.3021544107683421}], "active_band_count": 3, "active_band_mae_db": 0.4003816259219628}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8453042559321198, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ec497e913d897c196c72f745fff9cc394970a2ceff40dd8e08d6f280dca5e78b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9685508469709454, "standard_deviation": 0.0044031516777902095, "minimum": 0.9637899320767287, "maximum": 0.9747004734973086, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7666837703221776, "standard_deviation": 0.04807081320552899, "minimum": 1.661692513539925, "maximum": 1.830321189217636, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5779468324866126, "standard_deviation": 0.2082926304622404, "minimum": 0.3885494884225605, "maximum": 0.9983841011736144, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2010074636387461, "standard_deviation": 0.014525396948430587, "minimum": 0.18042629610818647, "maximum": 0.2156784930339443, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9865338743730149, "standard_deviation": 0.0022708296765920647, "minimum": 0.9827541050514064, "maximum": 0.9889311641030464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.201692728697143, "standard_deviation": 0.04107505468910098, "minimum": 1.1078287264077007, "maximum": 1.2378692454635023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7592930516210392, "standard_deviation": 0.18528356447605848, "minimum": 0.3927228710247534, "maximum": 1.0036982133054988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12817863447365463, "standard_deviation": 0.011994929757948658, "minimum": 0.115257536997656, "maximum": 0.14646528714722332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.965298607016684, "standard_deviation": 0.002621047892608869, "minimum": 0.9604384971136783, "maximum": 0.9681803403458704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1816278088564514, "standard_deviation": 0.02577151613916781, "minimum": 2.148110785864551, "maximum": 2.222081450857259, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.47602644659641774, "standard_deviation": 0.17685198173755054, "minimum": 0.28729603580090324, "maximum": 0.8185718326475099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23282368010355653, "standard_deviation": 0.00932323751154369, "minimum": 0.22271492230065063, "maximum": 0.2530025498031009, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9835048113511808, "standard_deviation": 0.0015484170862645457, "minimum": 0.9802405160896799, "maximum": 0.9848158474816053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.871660905172026, "standard_deviation": 0.030458746913278684, "minimum": 1.8121730337584268, "maximum": 1.9087344677305202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5112234309102517, "standard_deviation": 0.1604790297504616, "minimum": 0.30175410350592885, "maximum": 0.7881910424911777, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1576929133583318, "standard_deviation": 0.008561483438153057, "minimum": 0.15026847602487897, "maximum": 0.17518071608613497, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9846560575259435, "standard_deviation": 0.002633897142742086, "minimum": 0.9816496183775503, "maximum": 0.9886972277214297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.606950057608001, "standard_deviation": 0.06121724712390812, "minimum": 1.5128839276938415, "maximum": 1.6952727244292014, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5552791745198296, "standard_deviation": 0.2099437083105634, "minimum": 0.17567426732010438, "maximum": 0.9073954297204594, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14783539751002076, "standard_deviation": 0.012929158689851138, "minimum": 0.12741058954882623, "maximum": 0.16151998286997954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9953612279685033, "standard_deviation": 0.0009882215974849977, "minimum": 0.9943127077770308, "maximum": 0.997547173799546, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9491491463202506, "standard_deviation": 0.05040469720818775, "minimum": 0.86847955859658, "maximum": 1.0117629829714376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5061182776082657, "standard_deviation": 0.11991949771819271, "minimum": 0.38758915840696045, "maximum": 0.7500565467421869, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08188731558544687, "standard_deviation": 0.009431497670799056, "minimum": 0.061230786089973205, "maximum": 0.08950786681115014, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13049058077432063, "standard_deviation": 0.04638449899774882, "minimum": 0.06463856019070958, "maximum": 0.20010047015671795, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.558935124079685, "standard_deviation": 0.006753874159804419, "minimum": 5.5483555849905555, "maximum": 5.569205116131671, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4804787658639073, "standard_deviation": 0.019940518740052, "minimum": 0.4565126662685395, "maximum": 0.5128312416176727, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08973109286343052, "standard_deviation": 0.05933574889855997, "minimum": -0.19426078973424607, "maximum": -0.01755556469647379, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9928410246567032, "standard_deviation": 0.001045779687894603, "minimum": 0.9903408105224517, "maximum": 0.9936735391477467, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9978198764503707, "standard_deviation": 0.03236885519281197, "minimum": 0.9353277193830831, "maximum": 1.0313725756504024, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2960020682504211, "standard_deviation": 0.0985359918012661, "minimum": 0.1575581765943573, "maximum": 0.4512209379500707, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0904192173101971, "standard_deviation": 0.006215285523040207, "minimum": 0.08567822362785257, "maximum": 0.1049686918761357, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980315855617913, "standard_deviation": 0.0002952214107837849, "minimum": 0.9975610015921531, "maximum": 0.9983894930160713, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5308871569032391, "standard_deviation": 0.043750842280399306, "minimum": 0.49371216454640793, "maximum": 0.621164123789859, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.30807303908032824, "standard_deviation": 0.17839416566555025, "minimum": 0.1270539044852122, "maximum": 0.5665682190703111, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04695045365657527, "standard_deviation": 0.003133156843195328, "minimum": 0.04368954772197075, "maximum": 0.05093510877097236, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_260-123288-0003", "language": "en", "style": "normal", "speaker": "librispeech_260", "source_utterance_id": "260-123288-0003", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_260-123288-0003.wav", "source_file_sha256": "5f9aea48abd2b1462bd025080b267915f82f4cd9a104bdfde9dc44a6c27358af", "decoded_pcm_sha256": "d480f00d02d595a6352c610e02536f099ff03473f6693b5628683647e9479766", "original_source_sha256": "5e8ca953fe1b2dd885f0cfbf4d706040e55637eee883bea8f0dc6e08afac3240", "corpus_entry_id": "librispeech_260-123288-0003"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.6410365104675293, "predicted_mos": 3.6410365104675293, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8333faa6f2e05fbbee30ea66662dbc72bef5a6df1ed83ed9a3d7de66ccf556b9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.081714187999751, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_260-123288-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "8333faa6f2e05fbbee30ea66662dbc72bef5a6df1ed83ed9a3d7de66ccf556b9", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "380f18150746b92f548e2cc642146570a9fdb278b3fbf7470f430b6e5ea125e7"}, "rows": [{"case": "librispeech_260-123288-0003", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.5770514011383057, "predicted_mos": 2.5770514011383057, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ac8e34ad3047e037ed219c036185f189fcb6a3037909e459549581a8d410f601", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.900330308002594, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_260-123288-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ac8e34ad3047e037ed219c036185f189fcb6a3037909e459549581a8d410f601", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2765138622172967, "全帯域_10ms_rms_envelope_nrmse": 0.12026408032058064, "全帯域_10ms_envelope_correlation": 0.9860439056248117, "全帯域_10ms_modulation_mae_db": 0.22538767058725376, "全帯域_40ms_log_rms_mae_db": 0.6332518139430388, "全帯域_40ms_rms_envelope_nrmse": 0.06519606543796727, "全帯域_40ms_envelope_correlation": 0.9954924285177731, "全帯域_40ms_modulation_mae_db": 0.33106489671632133, "300–1000 Hz_10ms_log_rms_mae_db": 1.9130295790749174, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18080642611691147, "300–1000 Hz_10ms_envelope_correlation": 0.9749519660741504, "300–1000 Hz_10ms_modulation_mae_db": 0.2827842872622816, "300–1000 Hz_40ms_log_rms_mae_db": 1.0828956820618547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0955956031679762, "300–1000 Hz_40ms_envelope_correlation": 0.992757835455141, "300–1000 Hz_40ms_modulation_mae_db": 0.2880062626078011, "1000–3000 Hz_10ms_log_rms_mae_db": 1.869903931471678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2439948444664359, "1000–3000 Hz_10ms_envelope_correlation": 0.9504361538413042, "1000–3000 Hz_10ms_modulation_mae_db": 0.2996122102035359, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0960047413605796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13878784673269406, "1000–3000 Hz_40ms_envelope_correlation": 0.9819439766180097, "1000–3000 Hz_40ms_modulation_mae_db": 0.6150524508854178, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1022229833547157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17314633764108087, "3000–6000 Hz_10ms_envelope_correlation": 0.9802677474827838, "3000–6000 Hz_10ms_modulation_mae_db": 0.19125853481995905, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2186398674261394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09205552653082538, "3000–6000 Hz_40ms_envelope_correlation": 0.994160455844028, "3000–6000 Hz_40ms_modulation_mae_db": 0.4980267923360882, "mr_stft_db": 5.665854939239647, "band_mae_db": 0.1243953768838329, "rms_gain_error_db": -0.08742348024412233, "output_peak": 0.3737158899566247}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2529297033360065, "全帯域_10ms_rms_envelope_nrmse": 0.11057338353587512, "全帯域_10ms_envelope_correlation": 0.9881809411935998, "全帯域_10ms_modulation_mae_db": 0.5489508798998743, "全帯域_40ms_log_rms_mae_db": 0.6383973651026533, "全帯域_40ms_rms_envelope_nrmse": 0.053705874744088104, "全帯域_40ms_envelope_correlation": 0.9969242556550069, "全帯域_40ms_modulation_mae_db": 0.5603485781558829, "300–1000 Hz_10ms_log_rms_mae_db": 1.9553624393892808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16355682105153607, "300–1000 Hz_10ms_envelope_correlation": 0.9796325334452308, "300–1000 Hz_10ms_modulation_mae_db": 0.4505425088190944, "300–1000 Hz_40ms_log_rms_mae_db": 1.150676691835796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08813278487989783, "300–1000 Hz_40ms_envelope_correlation": 0.9939572253251614, "300–1000 Hz_40ms_modulation_mae_db": 0.2856645654933749, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9058619322763368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22256929590962174, "1000–3000 Hz_10ms_envelope_correlation": 0.9589359558334227, "1000–3000 Hz_10ms_modulation_mae_db": 0.5349605781475418, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1376421821356846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12000686858765092, "1000–3000 Hz_40ms_envelope_correlation": 0.9865239747795881, "1000–3000 Hz_40ms_modulation_mae_db": 0.6780073115363576, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1403663078106403, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17430108008016953, "3000–6000 Hz_10ms_envelope_correlation": 0.9795228463172557, "3000–6000 Hz_10ms_modulation_mae_db": 0.8369351993861093, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2514894521538726, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10004515836581411, "3000–6000 Hz_40ms_envelope_correlation": 0.9932776218608133, "3000–6000 Hz_40ms_modulation_mae_db": 0.7587845829343635, "mr_stft_db": 5.695599195750046, "band_mae_db": 0.10265420951242621, "rms_gain_error_db": -0.0670494831322086, "output_peak": 0.37392639281711815}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2698935934029416, "全帯域_10ms_rms_envelope_nrmse": 0.1369269796717826, "全帯域_10ms_envelope_correlation": 0.9822452689964801, "全帯域_10ms_modulation_mae_db": 0.25795026779500646, "全帯域_40ms_log_rms_mae_db": 0.6817219387253987, "全帯域_40ms_rms_envelope_nrmse": 0.07932148492209161, "全帯域_40ms_envelope_correlation": 0.9934357288269049, "全帯域_40ms_modulation_mae_db": 0.44806667909156045, "300–1000 Hz_10ms_log_rms_mae_db": 1.820651235579377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1895646451721366, "300–1000 Hz_10ms_envelope_correlation": 0.9730902060968322, "300–1000 Hz_10ms_modulation_mae_db": 0.39581025573721096, "300–1000 Hz_40ms_log_rms_mae_db": 1.0788821425383646, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11940015541808784, "300–1000 Hz_40ms_envelope_correlation": 0.9884143302598835, "300–1000 Hz_40ms_modulation_mae_db": 0.30472088151222915, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8245011617523597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2231767162127035, "1000–3000 Hz_10ms_envelope_correlation": 0.9584866005126322, "1000–3000 Hz_10ms_modulation_mae_db": 0.7099392823259567, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0558403908826055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12244105086223281, "1000–3000 Hz_40ms_envelope_correlation": 0.9859778189171383, "1000–3000 Hz_40ms_modulation_mae_db": 0.7889724624596091, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1092349917586146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16751171059128503, "3000–6000 Hz_10ms_envelope_correlation": 0.9809437149895642, "3000–6000 Hz_10ms_modulation_mae_db": 0.43536281900573853, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2151476491933315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09247685781934156, "3000–6000 Hz_40ms_envelope_correlation": 0.99383743482714, "3000–6000 Hz_40ms_modulation_mae_db": 0.6148437655340161, "mr_stft_db": 5.685791269653186, "band_mae_db": 0.18078154426911716, "rms_gain_error_db": 0.03783500233779167, "output_peak": 0.5231111630702706}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2587267198833205, "全帯域_10ms_rms_envelope_nrmse": 0.11030947707110256, "全帯域_10ms_envelope_correlation": 0.9882762934294322, "全帯域_10ms_modulation_mae_db": 0.14553950541249272, "全帯域_40ms_log_rms_mae_db": 0.6356569725699238, "全帯域_40ms_rms_envelope_nrmse": 0.05595533940838971, "全帯域_40ms_envelope_correlation": 0.9966452975062623, "全帯域_40ms_modulation_mae_db": 0.3348976764771668, "300–1000 Hz_10ms_log_rms_mae_db": 1.947651384656509, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15846463764970062, "300–1000 Hz_10ms_envelope_correlation": 0.980758316863758, "300–1000 Hz_10ms_modulation_mae_db": 0.3583104914369563, "300–1000 Hz_40ms_log_rms_mae_db": 1.106640701985225, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08660758036630138, "300–1000 Hz_40ms_envelope_correlation": 0.99377286103799, "300–1000 Hz_40ms_modulation_mae_db": 0.35018098271034565, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8947182526495785, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22268993330897577, "1000–3000 Hz_10ms_envelope_correlation": 0.9586956840986297, "1000–3000 Hz_10ms_modulation_mae_db": 1.178005499872441, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1027632650866959, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11843232742231528, "1000–3000 Hz_40ms_envelope_correlation": 0.9882862157215916, "1000–3000 Hz_40ms_modulation_mae_db": 1.3568371343925876, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1143542615463784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17406785583004497, "3000–6000 Hz_10ms_envelope_correlation": 0.9794104751789358, "3000–6000 Hz_10ms_modulation_mae_db": 0.5681023251670317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2667215571075054, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10566123236576938, "3000–6000 Hz_40ms_envelope_correlation": 0.992020961255984, "3000–6000 Hz_40ms_modulation_mae_db": 0.7766781913429833, "mr_stft_db": 5.674019475940141, "band_mae_db": 0.1422729782428774, "rms_gain_error_db": -0.03014232347148276, "output_peak": 0.38425222106010803}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2544900386315636, "全帯域_10ms_rms_envelope_nrmse": 0.12078742676383707, "全帯域_10ms_envelope_correlation": 0.9860361371134441, "全帯域_10ms_modulation_mae_db": 0.40160806907900365, "全帯域_40ms_log_rms_mae_db": 0.6704921949184961, "全帯域_40ms_rms_envelope_nrmse": 0.06577344159700818, "全帯域_40ms_envelope_correlation": 0.9953746273215276, "全帯域_40ms_modulation_mae_db": 0.4230436126245459, "300–1000 Hz_10ms_log_rms_mae_db": 1.7826372422692316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1688586237178235, "300–1000 Hz_10ms_envelope_correlation": 0.9784806710641057, "300–1000 Hz_10ms_modulation_mae_db": 0.22508662935876542, "300–1000 Hz_40ms_log_rms_mae_db": 1.0625598032206374, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09778162930380795, "300–1000 Hz_40ms_envelope_correlation": 0.9921011030285277, "300–1000 Hz_40ms_modulation_mae_db": 0.29502612512430687, "1000–3000 Hz_10ms_log_rms_mae_db": 1.925688908212013, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21832998967050904, "1000–3000 Hz_10ms_envelope_correlation": 0.9599745191808121, "1000–3000 Hz_10ms_modulation_mae_db": 1.2849508570121875, "1000–3000 Hz_40ms_log_rms_mae_db": 1.158535994840525, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11634512303228958, "1000–3000 Hz_40ms_envelope_correlation": 0.9875465833233912, "1000–3000 Hz_40ms_modulation_mae_db": 1.6205703023736377, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1392979327246864, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18253649983329726, "3000–6000 Hz_10ms_envelope_correlation": 0.9773035302765173, "3000–6000 Hz_10ms_modulation_mae_db": 0.8060704159977696, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2870298554803756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10165299408381835, "3000–6000 Hz_40ms_envelope_correlation": 0.9924800032473211, "3000–6000 Hz_40ms_modulation_mae_db": 1.042507269052682, "mr_stft_db": 5.683582738674008, "band_mae_db": 0.10965401951953387, "rms_gain_error_db": 0.04407100171990025, "output_peak": 0.42097883276677056}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.248815414378774, "全帯域_10ms_rms_envelope_nrmse": 0.10648649245106193, "全帯域_10ms_envelope_correlation": 0.9890483371156387, "全帯域_10ms_modulation_mae_db": 0.3170965373438614, "全帯域_40ms_log_rms_mae_db": 0.6686559226997086, "全帯域_40ms_rms_envelope_nrmse": 0.05584972608483722, "全帯域_40ms_envelope_correlation": 0.9966544785905493, "全帯域_40ms_modulation_mae_db": 0.4396393684868322, "300–1000 Hz_10ms_log_rms_mae_db": 1.8535187643513957, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17123117968803508, "300–1000 Hz_10ms_envelope_correlation": 0.9774857413044251, "300–1000 Hz_10ms_modulation_mae_db": 0.20744621084479617, "300–1000 Hz_40ms_log_rms_mae_db": 1.0099848813972319, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0847091833077384, "300–1000 Hz_40ms_envelope_correlation": 0.9941712881451416, "300–1000 Hz_40ms_modulation_mae_db": 0.2551331672692921, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8449189901506409, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19975255606808434, "1000–3000 Hz_10ms_envelope_correlation": 0.9667943301450896, "1000–3000 Hz_10ms_modulation_mae_db": 0.8504266883292817, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0753791296672761, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10105967416703648, "1000–3000 Hz_40ms_envelope_correlation": 0.9905249050342045, "1000–3000 Hz_40ms_modulation_mae_db": 1.1566866053517477, "3000–6000 Hz_10ms_log_rms_mae_db": 2.110937628189943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18445501191224983, "3000–6000 Hz_10ms_envelope_correlation": 0.9778481536387766, "3000–6000 Hz_10ms_modulation_mae_db": 0.12709499479956876, "3000–6000 Hz_40ms_log_rms_mae_db": 1.281352482479855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11039885883510088, "3000–6000 Hz_40ms_envelope_correlation": 0.9917109181853436, "3000–6000 Hz_40ms_modulation_mae_db": 0.14015072398432935, "mr_stft_db": 5.6853099699093015, "band_mae_db": 0.09064477253771204, "rms_gain_error_db": -0.0403981298463454, "output_peak": 0.40149092222092464}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.288030161428826, "全帯域_10ms_rms_envelope_nrmse": 0.13677548190812888, "全帯域_10ms_envelope_correlation": 0.9825398513127764, "全帯域_10ms_modulation_mae_db": 0.044158451972737676, "全帯域_40ms_log_rms_mae_db": 0.641490617486561, "全帯域_40ms_rms_envelope_nrmse": 0.08633199352410607, "全帯域_40ms_envelope_correlation": 0.992357366122276, "全帯域_40ms_modulation_mae_db": 0.2303193589150574, "300–1000 Hz_10ms_log_rms_mae_db": 1.8199419410111537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17846425208270372, "300–1000 Hz_10ms_envelope_correlation": 0.9768194329297883, "300–1000 Hz_10ms_modulation_mae_db": 0.3439200333843024, "300–1000 Hz_40ms_log_rms_mae_db": 1.013355818449709, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11881336818777563, "300–1000 Hz_40ms_envelope_correlation": 0.9889917329561847, "300–1000 Hz_40ms_modulation_mae_db": 0.21008364402604732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8497098686068638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22342394493832815, "1000–3000 Hz_10ms_envelope_correlation": 0.9582562396010148, "1000–3000 Hz_10ms_modulation_mae_db": 0.9752866192759301, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0974913020320927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1186808285052237, "1000–3000 Hz_40ms_envelope_correlation": 0.9868079981624563, "1000–3000 Hz_40ms_modulation_mae_db": 1.1640542891353205, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1135451683048223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1778829175859025, "3000–6000 Hz_10ms_envelope_correlation": 0.9785625466761254, "3000–6000 Hz_10ms_modulation_mae_db": 0.4773394234201429, "3000–6000 Hz_40ms_log_rms_mae_db": 1.223103724178485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10020299561360661, "3000–6000 Hz_40ms_envelope_correlation": 0.9930809712528266, "3000–6000 Hz_40ms_modulation_mae_db": 0.40262293470927557, "mr_stft_db": 5.6912899956286145, "band_mae_db": 0.15639974113480734, "rms_gain_error_db": 0.09815278685591, "output_peak": 0.5191409298672859}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2401700935817528, "全帯域_10ms_rms_envelope_nrmse": 0.1149681868310443, "全帯域_10ms_envelope_correlation": 0.9872767217357769, "全帯域_10ms_modulation_mae_db": 0.4684413927116195, "全帯域_40ms_log_rms_mae_db": 0.6054148493356175, "全帯域_40ms_rms_envelope_nrmse": 0.05905696113267731, "全帯域_40ms_envelope_correlation": 0.9964104778040629, "全帯域_40ms_modulation_mae_db": 0.5684643887343241, "300–1000 Hz_10ms_log_rms_mae_db": 1.857440402241185, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17456529797795228, "300–1000 Hz_10ms_envelope_correlation": 0.9770185567043432, "300–1000 Hz_10ms_modulation_mae_db": 0.5604128639796304, "300–1000 Hz_40ms_log_rms_mae_db": 1.0413259479689208, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09680012893233157, "300–1000 Hz_40ms_envelope_correlation": 0.9931721851512916, "300–1000 Hz_40ms_modulation_mae_db": 0.4925981675542817, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8595093292733023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22927504108783037, "1000–3000 Hz_10ms_envelope_correlation": 0.9568133822368964, "1000–3000 Hz_10ms_modulation_mae_db": 0.7815838449619573, "1000–3000 Hz_40ms_log_rms_mae_db": 1.117528639108084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13863934197445413, "1000–3000 Hz_40ms_envelope_correlation": 0.9823835171516258, "1000–3000 Hz_40ms_modulation_mae_db": 0.876284430756122, "3000–6000 Hz_10ms_log_rms_mae_db": 2.117684192710916, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18453405693184374, "3000–6000 Hz_10ms_envelope_correlation": 0.9773410623853905, "3000–6000 Hz_10ms_modulation_mae_db": 0.11056749549739671, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2910166383836514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11137218334722257, "3000–6000 Hz_40ms_envelope_correlation": 0.9912303828165656, "3000–6000 Hz_40ms_modulation_mae_db": 0.38660197397793644, "mr_stft_db": 5.672452835851998, "band_mae_db": 0.16829470678128788, "rms_gain_error_db": -0.12792871001741535, "output_peak": 0.38507683158359574}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.044091670631526, "spectral_convergence": 0.35641710530185955, "spectral_convergence_status": "ok", "amplitude_floor": 1.3116896377207775e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.455297339754661, "spectral_convergence": 0.35186377689908177, "spectral_convergence_status": "ok", "amplitude_floor": 8.457684224428572e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.498175807332755, "spectral_convergence": 0.36728559785485665, "spectral_convergence_status": "ok", "amplitude_floor": 6.717993311155654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008437417957183396, "output_power": 0.0008428333933591539, "delta_db": -0.004678287367755134, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011741200681074208, "output_power": 0.0011241330465115947, "delta_db": -0.18894795994148023, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004760668283187788, "output_power": 0.00047050199258196824, "delta_db": -0.05106454477239901, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001833223674641814, "output_power": 0.0001823154470208793, "delta_db": -0.023919905457861077, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.526890589901711e-05, "output_power": 6.390755724813554e-05, "delta_db": -0.0915411495240373, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6955820206525705e-05, "output_power": 4.0392886291603505e-05, "delta_db": 0.3862204142394646, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1243953768838329, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08742348024412233, "pitch": null, "output_rms_dbfs": -25.64779453127612, "output_peak": 0.3737158899566247, "reference_rms_dbfs": -25.560371051032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.665854939239647}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001544797211376954, "log_rms_mae_db": 1.2765138622172967, "rms_envelope_nrmse": 0.12026408032058064, "envelope_correlation": 0.9860439056248117, "rms_gain_error_db": -0.08742120103739479, "reference_mean_power": 0.002781784143506645, "output_mean_power": 0.0027263481089386194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.487428460245723e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008814675226478596, "output_power": 0.0008508341662181179, "reference_active": true, "delta_db": -0.15361394334963518}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001997406840389717, "output_power": 0.00019016008802046186, "reference_active": true, "delta_db": -0.21347163098532673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0345220429658976e-05, "output_power": 3.75737333404563e-05, "reference_active": true, "delta_db": -0.3090774374267994}], "active_band_count": 3, "active_band_mae_db": 0.22538767058725376}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013989081102856241, "log_rms_mae_db": 0.6332518139430388, "rms_envelope_nrmse": 0.06519606543796727, "envelope_correlation": 0.9954924285177731, "rms_gain_error_db": -0.08754430754642155, "reference_mean_power": 0.002795409730019158, "output_mean_power": 0.002739624502250603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3091173503801554e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008259246223406256, "output_power": 0.0007991076185772066, "reference_active": true, "delta_db": -0.14335142240310986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.583418527096058e-05, "output_power": 8.911568343204243e-05, "reference_active": true, "delta_db": -0.3156631300457231}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329771424834818e-06, "output_power": 2.9443972824351685e-06, "reference_active": true, "delta_db": -0.5341801377001311}], "active_band_count": 3, "active_band_mae_db": 0.33106489671632133}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000132208474335363, "log_rms_mae_db": 1.9130295790749174, "rms_envelope_nrmse": 0.18080642611691147, "envelope_correlation": 0.9749519660741504, "rms_gain_error_db": -0.23123942819497492, "reference_mean_power": 0.001270855716483403, "output_mean_power": 0.0012049590946928465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.478174051796597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005667765898001897, "output_power": 0.0005335889753876276, "reference_active": true, "delta_db": -0.26205055864324567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011084740591226797, "output_power": 9.990383972152573e-05, "reference_active": true, "delta_db": -0.45143353769726535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2190579526022385e-05, "output_power": 2.2890512743155273e-05, "reference_active": true, "delta_db": 0.13486876544633378}], "active_band_count": 3, "active_band_mae_db": 0.2827842872622816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012422751968092996, "log_rms_mae_db": 1.0828956820618547, "rms_envelope_nrmse": 0.0955956031679762, "envelope_correlation": 0.992757835455141, "rms_gain_error_db": -0.2312397687289493, "reference_mean_power": 0.0012772403358267607, "output_mean_power": 0.001211012562736551}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.710312666254066e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005574547781532276, "output_power": 0.0005248630032450159, "reference_active": true, "delta_db": -0.2616368118948216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.253668134122962e-05, "output_power": 4.805911048331623e-05, "reference_active": true, "delta_db": -0.38686907352829397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7043530583570796e-06, "output_power": 1.6218409958063098e-06, "reference_active": true, "delta_db": -0.21551290240028773}], "active_band_count": 3, "active_band_mae_db": 0.2880062626078011}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.199705053096058e-05, "log_rms_mae_db": 1.869903931471678, "rms_envelope_nrmse": 0.2439948444664359, "envelope_correlation": 0.9504361538413042, "rms_gain_error_db": -0.08298768130144599, "reference_mean_power": 0.00020286667816607547, "output_mean_power": 0.00019902697846123314}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3687094204678632e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.826983537989376e-05, "output_power": 7.195555079901298e-05, "reference_active": true, "delta_db": -0.3653011846265447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2413787132715612e-05, "output_power": 2.1636298475924943e-05, "reference_active": true, "delta_db": -0.15332279085972353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388555256663533e-06, "output_power": 5.853038623562954e-06, "reference_active": true, "delta_db": -0.38021265512433944}], "active_band_count": 3, "active_band_mae_db": 0.2996122102035359}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.313537286332709e-05, "log_rms_mae_db": 1.0960047413605796, "rms_envelope_nrmse": 0.13878784673269406, "envelope_correlation": 0.9819439766180097, "rms_gain_error_db": -0.08323716550744194, "reference_mean_power": 0.00020386380860129322, "output_mean_power": 0.00019999374685878164}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.180981609421061e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.558718162537124e-05, "output_power": 7.116263927873002e-05, "reference_active": true, "delta_db": -0.26196105544546144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0576597086854303e-05, "output_power": 1.0260449866531458e-05, "reference_active": true, "delta_db": -0.13179557603035585}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.38427493679316e-07, "output_power": 3.8546695754908466e-07, "reference_active": true, "delta_db": -1.4514007211804363}], "active_band_count": 3, "active_band_mae_db": 0.6150524508854178}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.0553751915640445e-05, "log_rms_mae_db": 2.1022229833547157, "rms_envelope_nrmse": 0.17314633764108087, "envelope_correlation": 0.9802677474827838, "rms_gain_error_db": 0.16041522381276122, "reference_mean_power": 5.25173391881524e-05, "output_mean_power": 5.449344153523864e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.578963632678691e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2240876569039177e-05, "output_power": 2.2256512349777382e-05, "reference_active": true, "delta_db": 0.0030521042871626287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.96335110256818e-06, "output_power": 8.420861992543364e-06, "reference_active": true, "delta_db": 0.2426068576009982}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4165297947437381e-06, "output_power": 1.3134515261602685e-06, "reference_active": true, "delta_db": -0.32811664257171635}], "active_band_count": 3, "active_band_mae_db": 0.19125853481995905}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.394748258818145e-05, "log_rms_mae_db": 1.2186398674261394, "rms_envelope_nrmse": 0.09205552653082538, "envelope_correlation": 0.994160455844028, "rms_gain_error_db": 0.15584526111672903, "reference_mean_power": 5.256947927681562e-05, "output_mean_power": 5.449017482401667e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.187575200111859e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.180909445738007e-05, "output_power": 2.2150295928339037e-05, "reference_active": true, "delta_db": 0.06741899311364953}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0063771970727005e-06, "output_power": 4.274554761485653e-06, "reference_active": true, "delta_db": 0.2813904987038107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0500724932970203e-07, "output_power": 8.066631050418877e-08, "reference_active": true, "delta_db": -1.1452708851908044}], "active_band_count": 3, "active_band_mae_db": 0.4980267923360882}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8207086687243562, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2f0b51ac1af5975299a52921d6132f3c775360218aa956f1ae751765fbb6b048", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9585491081812252, "standard_deviation": 0.004457883374204724, "minimum": 0.9504361538413042, "maximum": 0.9667943301450896, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8718515467990966, "standard_deviation": 0.03420913216515325, "minimum": 1.8245011617523597, "maximum": 1.925688908212013, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.826845697516104, "standard_deviation": 0.3239603442641725, "minimum": 0.2996122102035359, "maximum": 1.2849508570121875, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2229015402078111, "standard_deviation": 0.012205545176098108, "minimum": 0.19975255606808434, "maximum": 0.2439948444664359, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9862493737135006, "standard_deviation": 0.002879834974115808, "minimum": 0.9819439766180097, "maximum": 0.9905249050342045, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.105148205639193, "standard_deviation": 0.032786685474510614, "minimum": 1.0558403908826055, "maximum": 1.158535994840525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0320581233613502, "standard_deviation": 0.3520115718936818, "minimum": 0.6150524508854178, "maximum": 1.6205703023736377, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12179913266048711, "standard_deviation": 0.012286833155818718, "minimum": 0.10105967416703648, "maximum": 0.13878784673269406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9789000096181687, "standard_deviation": 0.0013572705886327995, "minimum": 0.9773035302765173, "maximum": 0.9809437149895642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1184554333000896, "standard_deviation": 0.013944008930248584, "minimum": 2.1022229833547157, "maximum": 2.1403663078106403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4440914010117145, "standard_deviation": 0.28724423679085775, "minimum": 0.11056749549739671, "maximum": 0.8369351993861093, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17730443380073424, "standard_deviation": 0.006138754392808181, "minimum": 0.16751171059128503, "maximum": 0.18453405693184374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927248436612528, "standard_deviation": 0.0010384113826193719, "minimum": 0.9912303828165656, "maximum": 0.994160455844028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.254312653300402, "standard_deviation": 0.03182798412810663, "minimum": 1.2151476491933315, "maximum": 1.2910166383836514, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5775270292339593, "standard_deviation": 0.28152056190811975, "minimum": 0.14015072398432935, "maximum": 1.042507269052682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10173322587018735, "standard_deviation": 0.007250298108441074, "minimum": 0.09205552653082538, "maximum": 0.11137218334722257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9772796780603292, "standard_deviation": 0.002460302856181262, "minimum": 0.9730902060968322, "maximum": 0.980758316863758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8687791235716311, "standard_deviation": 0.06342469082211431, "minimum": 1.7826372422692316, "maximum": 1.9553624393892808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3530391601028797, "standard_deviation": 0.11768964771422048, "minimum": 0.20744621084479617, "maximum": 0.5604128639796304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17318898543209993, "standard_deviation": 0.009914494102403365, "minimum": 0.15846463764970062, "maximum": 0.1895646451721366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9921673201699152, "standard_deviation": 0.00224657020013913, "minimum": 0.9884143302598835, "maximum": 0.9941712881451416, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0682902086822175, "standard_deviation": 0.047450740143332934, "minimum": 1.0099848813972319, "maximum": 1.150676691835796, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.31017672453720985, "standard_deviation": 0.08390302541469392, "minimum": 0.21008364402604732, "maximum": 0.4925981675542817, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09848005419548961, "standard_deviation": 0.013629709142390546, "minimum": 0.0847091833077384, "maximum": 0.11940015541808784, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13438716861019934, "standard_deviation": 0.03269084855216921, "minimum": 0.09064477253771204, "maximum": 0.18078154426911716, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.681737552580868, "standard_deviation": 0.010099688094430606, "minimum": 5.665854939239647, "maximum": 5.695599195750046, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4227116479178373, "standard_deviation": 0.0626875280780503, "minimum": 0.3737158899566247, "maximum": 0.5231111630702706, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.02161041697474656, "standard_deviation": 0.07589132331343565, "minimum": -0.12792871001741535, "maximum": 0.09815278685591, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.986205932065245, "standard_deviation": 0.002580005458061229, "minimum": 0.9822452689964801, "maximum": 0.9890483371156387, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2611961983575601, "standard_deviation": 0.015789344584808043, "minimum": 1.2401700935817528, "maximum": 1.288030161428826, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3011415968502312, "standard_deviation": 0.16801497828205197, "minimum": 0.044158451972737676, "maximum": 0.5489508798998743, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11963643856917663, "standard_deviation": 0.011700263493436175, "minimum": 0.10648649245106193, "maximum": 0.1369269796717826, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9954118325430453, "standard_deviation": 0.0016730191696455942, "minimum": 0.992357366122276, "maximum": 0.9969242556550069, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6468852093476747, "standard_deviation": 0.02500779905074166, "minimum": 0.6054148493356175, "maximum": 0.6817219387253987, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4169805699002114, "standard_deviation": 0.11586924463294902, "minimum": 0.2303193589150574, "maximum": 0.5684643887343241, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0651488608563957, "standard_deviation": 0.011886538781473324, "minimum": 0.053705874744088104, "maximum": 0.08633199352410607, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123288-0003", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.416579008102417, "predicted_mos": 2.416579008102417, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2b9207a29b80fcbe10b7279e3037ced1de73e89af09f5a65ed843382b1c12c43", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.496673999819905, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.275252638731993, "全帯域_10ms_rms_envelope_nrmse": 0.12456059761804607, "全帯域_10ms_envelope_correlation": 0.9850425108120634, "全帯域_10ms_modulation_mae_db": 0.16526539098122436, "全帯域_40ms_log_rms_mae_db": 0.6369454785623813, "全帯域_40ms_rms_envelope_nrmse": 0.06748168319710915, "全帯域_40ms_envelope_correlation": 0.9952009098504022, "全帯域_40ms_modulation_mae_db": 0.2945145408624431, "300–1000 Hz_10ms_log_rms_mae_db": 1.9212186896968435, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18653820080858616, "300–1000 Hz_10ms_envelope_correlation": 0.9733094741671794, "300–1000 Hz_10ms_modulation_mae_db": 0.3430765531628444, "300–1000 Hz_40ms_log_rms_mae_db": 1.0670072619882383, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0991740863720888, "300–1000 Hz_40ms_envelope_correlation": 0.9921362326145705, "300–1000 Hz_40ms_modulation_mae_db": 0.2759198685985381, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8768647477956462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23048730526362612, "1000–3000 Hz_10ms_envelope_correlation": 0.955519364151122, "1000–3000 Hz_10ms_modulation_mae_db": 0.4993132174833952, "1000–3000 Hz_40ms_log_rms_mae_db": 1.112895597088014, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1306949238806011, "1000–3000 Hz_40ms_envelope_correlation": 0.9844597600385222, "1000–3000 Hz_40ms_modulation_mae_db": 0.9077723326746917, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0920538816444925, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1755317212223695, "3000–6000 Hz_10ms_envelope_correlation": 0.9798581544381849, "3000–6000 Hz_10ms_modulation_mae_db": 0.2584238852822263, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2257879114514907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09549220641980882, "3000–6000 Hz_40ms_envelope_correlation": 0.9938239649487727, "3000–6000 Hz_40ms_modulation_mae_db": 0.5197404509938905, "mr_stft_db": 5.612350996014516, "band_mae_db": 0.1554934186826564, "rms_gain_error_db": -0.10174796829309413, "output_peak": 0.3751058336816822}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2731236853931087, "全帯域_10ms_rms_envelope_nrmse": 0.110645614295001, "全帯域_10ms_envelope_correlation": 0.9881727574563203, "全帯域_10ms_modulation_mae_db": 0.5464390459688281, "全帯域_40ms_log_rms_mae_db": 0.6421362394408722, "全帯域_40ms_rms_envelope_nrmse": 0.0523740994807452, "全帯域_40ms_envelope_correlation": 0.9970979605878262, "全帯域_40ms_modulation_mae_db": 0.5347823697550204, "300–1000 Hz_10ms_log_rms_mae_db": 1.9761733675962374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1663128877374927, "300–1000 Hz_10ms_envelope_correlation": 0.978925374339762, "300–1000 Hz_10ms_modulation_mae_db": 0.4428258428809337, "300–1000 Hz_40ms_log_rms_mae_db": 1.137179228321712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08692038446416708, "300–1000 Hz_40ms_envelope_correlation": 0.9941141004897573, "300–1000 Hz_40ms_modulation_mae_db": 0.2981315188232861, "1000–3000 Hz_10ms_log_rms_mae_db": 1.936944196329738, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21185511571509855, "1000–3000 Hz_10ms_envelope_correlation": 0.9624129531307597, "1000–3000 Hz_10ms_modulation_mae_db": 0.4764290091080166, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1406610803198787, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11182092527691437, "1000–3000 Hz_40ms_envelope_correlation": 0.9885255255431038, "1000–3000 Hz_40ms_modulation_mae_db": 0.7398306229651345, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0799142618937947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17157320216120434, "3000–6000 Hz_10ms_envelope_correlation": 0.9801493673398977, "3000–6000 Hz_10ms_modulation_mae_db": 0.9144986698228829, "3000–6000 Hz_40ms_log_rms_mae_db": 1.175667368709333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09878797228333075, "3000–6000 Hz_40ms_envelope_correlation": 0.9934894876323934, "3000–6000 Hz_40ms_modulation_mae_db": 0.7966415178863705, "mr_stft_db": 5.635168296411049, "band_mae_db": 0.1675132994481837, "rms_gain_error_db": -0.07875795708559681, "output_peak": 0.37217555008211034}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2730412023832076, "全帯域_10ms_rms_envelope_nrmse": 0.13596947584308422, "全帯域_10ms_envelope_correlation": 0.9825038264869991, "全帯域_10ms_modulation_mae_db": 0.328653390196413, "全帯域_40ms_log_rms_mae_db": 0.7076346345433333, "全帯域_40ms_rms_envelope_nrmse": 0.08052072680783741, "全帯域_40ms_envelope_correlation": 0.9932525745802048, "全帯域_40ms_modulation_mae_db": 0.5270221745027477, "300–1000 Hz_10ms_log_rms_mae_db": 1.91190478384759, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19207765293937967, "300–1000 Hz_10ms_envelope_correlation": 0.9726026387419747, "300–1000 Hz_10ms_modulation_mae_db": 0.3375380685484259, "300–1000 Hz_40ms_log_rms_mae_db": 1.154375567552596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12378070303766021, "300–1000 Hz_40ms_envelope_correlation": 0.9877179239617915, "300–1000 Hz_40ms_modulation_mae_db": 0.35527150712408456, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9166644289057695, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2073001817554714, "1000–3000 Hz_10ms_envelope_correlation": 0.964174445741326, "1000–3000 Hz_10ms_modulation_mae_db": 0.8764426166139678, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0916153464398761, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11485948922530313, "1000–3000 Hz_40ms_envelope_correlation": 0.9881528036283064, "1000–3000 Hz_40ms_modulation_mae_db": 0.975912255455595, "3000–6000 Hz_10ms_log_rms_mae_db": 2.061362651338548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16705174680326657, "3000–6000 Hz_10ms_envelope_correlation": 0.9809818412088898, "3000–6000 Hz_10ms_modulation_mae_db": 0.518037598106274, "3000–6000 Hz_40ms_log_rms_mae_db": 1.135470136097323, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09397812246985608, "3000–6000 Hz_40ms_envelope_correlation": 0.9935454809536534, "3000–6000 Hz_40ms_modulation_mae_db": 0.7639710439098613, "mr_stft_db": 5.633330490450444, "band_mae_db": 0.19941842814289532, "rms_gain_error_db": 0.041604230143852776, "output_peak": 0.5098735015867935}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2534550022619635, "全帯域_10ms_rms_envelope_nrmse": 0.10955582610529667, "全帯域_10ms_envelope_correlation": 0.9884472271163923, "全帯域_10ms_modulation_mae_db": 0.14291149397665207, "全帯域_40ms_log_rms_mae_db": 0.629083132189232, "全帯域_40ms_rms_envelope_nrmse": 0.05840991827690327, "全帯域_40ms_envelope_correlation": 0.9963571834999235, "全帯域_40ms_modulation_mae_db": 0.3349433625452248, "300–1000 Hz_10ms_log_rms_mae_db": 1.9327594237116519, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16234035008014613, "300–1000 Hz_10ms_envelope_correlation": 0.9798339503719429, "300–1000 Hz_10ms_modulation_mae_db": 0.3689648858253163, "300–1000 Hz_40ms_log_rms_mae_db": 1.0758811469178715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09013203930947196, "300–1000 Hz_40ms_envelope_correlation": 0.9932346417112479, "300–1000 Hz_40ms_modulation_mae_db": 0.35382226766988706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9306613162954356, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21685436951751189, "1000–3000 Hz_10ms_envelope_correlation": 0.9621891414761997, "1000–3000 Hz_10ms_modulation_mae_db": 1.3344727185007048, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0960976714194879, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12132775985669993, "1000–3000 Hz_40ms_envelope_correlation": 0.9897939272522324, "1000–3000 Hz_40ms_modulation_mae_db": 1.5930314970574717, "3000–6000 Hz_10ms_log_rms_mae_db": 2.067831908500817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17099603546707967, "3000–6000 Hz_10ms_envelope_correlation": 0.9800367837249436, "3000–6000 Hz_10ms_modulation_mae_db": 0.5900445040703352, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1790700061526105, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10083695041971272, "3000–6000 Hz_40ms_envelope_correlation": 0.9926799227395512, "3000–6000 Hz_40ms_modulation_mae_db": 0.8322043470931478, "mr_stft_db": 5.614374237422942, "band_mae_db": 0.18219140822591187, "rms_gain_error_db": -0.03147994032289202, "output_peak": 0.37964101125555827}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.23825532420377, "全帯域_10ms_rms_envelope_nrmse": 0.12271791887344922, "全帯域_10ms_envelope_correlation": 0.9855339588550424, "全帯域_10ms_modulation_mae_db": 0.41261693426066715, "全帯域_40ms_log_rms_mae_db": 0.6644300816270183, "全帯域_40ms_rms_envelope_nrmse": 0.06524706578799215, "全帯域_40ms_envelope_correlation": 0.9954171259754756, "全帯域_40ms_modulation_mae_db": 0.4274670082574275, "300–1000 Hz_10ms_log_rms_mae_db": 1.8056565938280804, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17108656303362418, "300–1000 Hz_10ms_envelope_correlation": 0.9779179733533541, "300–1000 Hz_10ms_modulation_mae_db": 0.2771645251390746, "300–1000 Hz_40ms_log_rms_mae_db": 1.0614897382042796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0958019813604423, "300–1000 Hz_40ms_envelope_correlation": 0.9924339119585031, "300–1000 Hz_40ms_modulation_mae_db": 0.450954931289733, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9390067037937644, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20206977625498268, "1000–3000 Hz_10ms_envelope_correlation": 0.9662507493805702, "1000–3000 Hz_10ms_modulation_mae_db": 1.3470864052145934, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1359522877764694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11085088553112901, "1000–3000 Hz_40ms_envelope_correlation": 0.9897831616893416, "1000–3000 Hz_40ms_modulation_mae_db": 1.7886245307946143, "3000–6000 Hz_10ms_log_rms_mae_db": 2.084421346495522, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17821149727993565, "3000–6000 Hz_10ms_envelope_correlation": 0.978349438615917, "3000–6000 Hz_10ms_modulation_mae_db": 0.7520464178099783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.198557428732675, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09327495673113957, "3000–6000 Hz_40ms_envelope_correlation": 0.9936407150243427, "3000–6000 Hz_40ms_modulation_mae_db": 0.9256619452285223, "mr_stft_db": 5.627915965126494, "band_mae_db": 0.13332080521203415, "rms_gain_error_db": 0.020665022135851377, "output_peak": 0.42157084910859977}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2528583644359406, "全帯域_10ms_rms_envelope_nrmse": 0.10885168247287665, "全帯域_10ms_envelope_correlation": 0.9885673419267007, "全帯域_10ms_modulation_mae_db": 0.2947488680994703, "全帯域_40ms_log_rms_mae_db": 0.6751689463057108, "全帯域_40ms_rms_envelope_nrmse": 0.05745363665394209, "全帯域_40ms_envelope_correlation": 0.9964477177206162, "全帯域_40ms_modulation_mae_db": 0.4616257730337004, "300–1000 Hz_10ms_log_rms_mae_db": 1.8671976707277447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17226272378055787, "300–1000 Hz_10ms_envelope_correlation": 0.9772083592435291, "300–1000 Hz_10ms_modulation_mae_db": 0.22769518915010542, "300–1000 Hz_40ms_log_rms_mae_db": 1.0010600857718182, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08615810415297982, "300–1000 Hz_40ms_envelope_correlation": 0.9938917162778819, "300–1000 Hz_40ms_modulation_mae_db": 0.3277204842184348, "1000–3000 Hz_10ms_log_rms_mae_db": 1.893569491128251, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20463250677192793, "1000–3000 Hz_10ms_envelope_correlation": 0.9650415337080842, "1000–3000 Hz_10ms_modulation_mae_db": 0.90718842655605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0673406548223385, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10217343973330746, "1000–3000 Hz_40ms_envelope_correlation": 0.9904999911058296, "1000–3000 Hz_40ms_modulation_mae_db": 1.305579644552273, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0392200130458322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18165432464619105, "3000–6000 Hz_10ms_envelope_correlation": 0.978441323162695, "3000–6000 Hz_10ms_modulation_mae_db": 0.3459709188976467, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1825133610044742, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11036449995722022, "3000–6000 Hz_40ms_envelope_correlation": 0.9916758225793241, "3000–6000 Hz_40ms_modulation_mae_db": 0.37253617481101786, "mr_stft_db": 5.627817889339642, "band_mae_db": 0.07338641887632538, "rms_gain_error_db": -0.016256910684108918, "output_peak": 0.38362248727113923}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3037953341909516, "全帯域_10ms_rms_envelope_nrmse": 0.1386861489297342, "全帯域_10ms_envelope_correlation": 0.9820797470824205, "全帯域_10ms_modulation_mae_db": 0.041258926592985974, "全帯域_40ms_log_rms_mae_db": 0.6682013200808307, "全帯域_40ms_rms_envelope_nrmse": 0.08670947517238782, "全帯域_40ms_envelope_correlation": 0.9923059825362762, "全帯域_40ms_modulation_mae_db": 0.14420617788882453, "300–1000 Hz_10ms_log_rms_mae_db": 1.8535088163449993, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18180354217470224, "300–1000 Hz_10ms_envelope_correlation": 0.9760100871478332, "300–1000 Hz_10ms_modulation_mae_db": 0.351977191596412, "300–1000 Hz_40ms_log_rms_mae_db": 1.0273198288772907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11987944783779893, "300–1000 Hz_40ms_envelope_correlation": 0.9888600227604183, "300–1000 Hz_40ms_modulation_mae_db": 0.24288038806849496, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8615540718237056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21292615375967958, "1000–3000 Hz_10ms_envelope_correlation": 0.9620258326362403, "1000–3000 Hz_10ms_modulation_mae_db": 0.9361038915114968, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1250934453555053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11518499153502064, "1000–3000 Hz_40ms_envelope_correlation": 0.9878925053515961, "1000–3000 Hz_40ms_modulation_mae_db": 1.2449053001618797, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0764592453399797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17762462394677944, "3000–6000 Hz_10ms_envelope_correlation": 0.9784502588341903, "3000–6000 Hz_10ms_modulation_mae_db": 0.5213782140700481, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1847506824568401, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10215981144962928, "3000–6000 Hz_40ms_envelope_correlation": 0.9925493258686084, "3000–6000 Hz_40ms_modulation_mae_db": 0.47272986668482186, "mr_stft_db": 5.633402718103979, "band_mae_db": 0.18086085229400672, "rms_gain_error_db": 0.10438002448682591, "output_peak": 0.516373917448644}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2897508343164397, "全帯域_10ms_rms_envelope_nrmse": 0.12126365816258387, "全帯域_10ms_envelope_correlation": 0.9858381532096563, "全帯域_10ms_modulation_mae_db": 0.4128928256734136, "全帯域_40ms_log_rms_mae_db": 0.6351116919709261, "全帯域_40ms_rms_envelope_nrmse": 0.06257263002621655, "全帯域_40ms_envelope_correlation": 0.9959817798998755, "全帯域_40ms_modulation_mae_db": 0.47189052132469383, "300–1000 Hz_10ms_log_rms_mae_db": 1.9119143292046594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1782772398523133, "300–1000 Hz_10ms_envelope_correlation": 0.9759698428717787, "300–1000 Hz_10ms_modulation_mae_db": 0.4991485710617332, "300–1000 Hz_40ms_log_rms_mae_db": 1.083810092379295, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0974951716536369, "300–1000 Hz_40ms_envelope_correlation": 0.9930404508378194, "300–1000 Hz_40ms_modulation_mae_db": 0.5991629389341683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8845929613051815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21890332724902803, "1000–3000 Hz_10ms_envelope_correlation": 0.9599744121321548, "1000–3000 Hz_10ms_modulation_mae_db": 0.8116527039645968, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1421133590330643, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12979988783869273, "1000–3000 Hz_40ms_envelope_correlation": 0.9844187200531813, "1000–3000 Hz_40ms_modulation_mae_db": 0.9547250362828873, "3000–6000 Hz_10ms_log_rms_mae_db": 2.045175021139451, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17888560364129014, "3000–6000 Hz_10ms_envelope_correlation": 0.978566650464791, "3000–6000 Hz_10ms_modulation_mae_db": 0.1574701428916844, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1689983564976207, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10820622471838853, "3000–6000 Hz_40ms_envelope_correlation": 0.991621530782772, "3000–6000 Hz_40ms_modulation_mae_db": 0.460230226523418, "mr_stft_db": 5.624806461153096, "band_mae_db": 0.16588642197139536, "rms_gain_error_db": -0.13950436896570761, "output_peak": 0.3847453070020932}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.992196756556515, "spectral_convergence": 0.3558484852868338, "spectral_convergence_status": "ok", "amplitude_floor": 1.3116896377207775e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.406527307350797, "spectral_convergence": 0.3497184129319849, "spectral_convergence_status": "ok", "amplitude_floor": 8.457684224428572e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.438328924136236, "spectral_convergence": 0.36699276274759873, "spectral_convergence_status": "ok", "amplitude_floor": 6.717993311155654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008437417957183396, "output_power": 0.0008443623536555636, "delta_db": 0.0031929890116960094, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011741200681074208, "output_power": 0.0011225606155001355, "delta_db": -0.19502709956341482, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004760668283187788, "output_power": 0.0004706140219056302, "delta_db": -0.05003058686373388, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001833223674641814, "output_power": 0.00017401273983740965, "delta_db": -0.22634412124740588, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.526890589901711e-05, "output_power": 6.356629369772377e-05, "delta_db": -0.1147944253592795, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6955820206525705e-05, "output_power": 3.999815643045495e-05, "delta_db": 0.3435712900504082, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1554934186826564, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10174796829309413, "pitch": null, "output_rms_dbfs": -25.66211901932509, "output_peak": 0.3751058336816822, "reference_rms_dbfs": -25.560371051032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.612350996014516}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001544797211376954, "log_rms_mae_db": 1.275252638731993, "rms_envelope_nrmse": 0.12456059761804607, "envelope_correlation": 0.9850425108120634, "rms_gain_error_db": -0.10174606262138308, "reference_mean_power": 0.002781784143506645, "output_mean_power": 0.0027173702788076226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.487428460245723e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008814675226478596, "output_power": 0.0008487479690791709, "reference_active": true, "delta_db": -0.16427567415776617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001997406840389717, "output_power": 0.00018990009886805289, "reference_active": true, "delta_db": -0.21941341998369263}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0345220429658976e-05, "output_power": 3.9317091637734954e-05, "reference_active": true, "delta_db": -0.11210707880221432}], "active_band_count": 3, "active_band_mae_db": 0.16526539098122436}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013989081102856241, "log_rms_mae_db": 0.6369454785623813, "rms_envelope_nrmse": 0.06748168319710915, "envelope_correlation": 0.9952009098504022, "rms_gain_error_db": -0.10185485514044199, "reference_mean_power": 0.002795409730019158, "output_mean_power": 0.002730611952981166}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3091173503801554e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008259246223406256, "output_power": 0.0007957673359856268, "reference_active": true, "delta_db": -0.16154304820383436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.583418527096058e-05, "output_power": 8.984099737739027e-05, "reference_active": true, "delta_db": -0.2804589043604529}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329771424834818e-06, "output_power": 3.007878261559611e-06, "reference_active": true, "delta_db": -0.44154167002304207}], "active_band_count": 3, "active_band_mae_db": 0.2945145408624431}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000132208474335363, "log_rms_mae_db": 1.9212186896968435, "rms_envelope_nrmse": 0.18653820080858616, "envelope_correlation": 0.9733094741671794, "rms_gain_error_db": -0.22368791675312377, "reference_mean_power": 0.001270855716483403, "output_mean_power": 0.001207056099894349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.478174051796597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005667765898001897, "output_power": 0.0005396186160238604, "reference_active": true, "delta_db": -0.21324980041686345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011084740591226797, "output_power": 9.932386453895743e-05, "reference_active": true, "delta_db": -0.4767192512376134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2190579526022385e-05, "output_power": 2.3993560814834746e-05, "reference_active": true, "delta_db": 0.3392606078340565}], "active_band_count": 3, "active_band_mae_db": 0.3430765531628444}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012422751968092996, "log_rms_mae_db": 1.0670072619882383, "rms_envelope_nrmse": 0.0991740863720888, "envelope_correlation": 0.9921362326145705, "rms_gain_error_db": -0.22368734470804702, "reference_mean_power": 0.0012772403358267607, "output_mean_power": 0.0012131203577751418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.710312666254066e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005574547781532276, "output_power": 0.0005314483195615951, "reference_active": true, "delta_db": -0.20748604159112904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.253668134122962e-05, "output_power": 4.845603739704016e-05, "reference_active": true, "delta_db": -0.3511473979146765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7043530583570796e-06, "output_power": 1.601942589045265e-06, "reference_active": true, "delta_db": -0.2691261662898089}], "active_band_count": 3, "active_band_mae_db": 0.2759198685985381}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.199705053096058e-05, "log_rms_mae_db": 1.8768647477956462, "rms_envelope_nrmse": 0.23048730526362612, "envelope_correlation": 0.955519364151122, "rms_gain_error_db": -0.2748761323740555, "reference_mean_power": 0.00020286667816607547, "output_mean_power": 0.0001904246230994137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3687094204678632e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.826983537989376e-05, "output_power": 6.872807509160693e-05, "reference_active": true, "delta_db": -0.5646023986257523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2413787132715612e-05, "output_power": 2.0470998052282935e-05, "reference_active": true, "delta_db": -0.39376226085720606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388555256663533e-06, "output_power": 5.642156862710444e-06, "reference_active": true, "delta_db": -0.539574992967227}], "active_band_count": 3, "active_band_mae_db": 0.4993132174833952}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.313537286332709e-05, "log_rms_mae_db": 1.112895597088014, "rms_envelope_nrmse": 0.1306949238806011, "envelope_correlation": 0.9844597600385222, "rms_gain_error_db": -0.27506902929023797, "reference_mean_power": 0.00020386380860129322, "output_mean_power": 0.00019135209902355768}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.180981609421061e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.558718162537124e-05, "output_power": 6.747130941375694e-05, "reference_active": true, "delta_db": -0.49329013783763687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0576597086854303e-05, "output_power": 9.272907292066374e-06, "reference_active": true, "delta_db": -0.5713004246548267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.38427493679316e-07, "output_power": 3.6749766547065553e-07, "reference_active": true, "delta_db": -1.6587264355316114}], "active_band_count": 3, "active_band_mae_db": 0.9077723326746917}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.0553751915640445e-05, "log_rms_mae_db": 2.0920538816444925, "rms_envelope_nrmse": 0.1755317212223695, "envelope_correlation": 0.9798581544381849, "rms_gain_error_db": 0.17916006038645918, "reference_mean_power": 5.25173391881524e-05, "output_mean_power": 5.472915216200632e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.578963632678691e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2240876569039177e-05, "output_power": 2.3085299574422266e-05, "reference_active": true, "delta_db": 0.16183614808983926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.96335110256818e-06, "output_power": 8.36473391841258e-06, "reference_active": true, "delta_db": 0.21356266535923757}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4165297947437381e-06, "output_power": 1.2919283530483799e-06, "reference_active": true, "delta_db": -0.3998728423976022}], "active_band_count": 3, "active_band_mae_db": 0.2584238852822263}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.394748258818145e-05, "log_rms_mae_db": 1.2257879114514907, "rms_envelope_nrmse": 0.09549220641980882, "envelope_correlation": 0.9938239649487727, "rms_gain_error_db": 0.1751480613179857, "reference_mean_power": 5.256947927681562e-05, "output_mean_power": 5.473290272644397e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.187575200111859e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.180909445738007e-05, "output_power": 2.292068995629187e-05, "reference_active": true, "delta_db": 0.21590053112389765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0063771970727005e-06, "output_power": 4.147803616674937e-06, "reference_active": true, "delta_db": 0.15066350603207182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0500724932970203e-07, "output_power": 7.97909343497457e-08, "reference_active": true, "delta_db": -1.1926573158257021}], "active_band_count": 3, "active_band_mae_db": 0.5197404509938905}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8169535615697329, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f3cf04af48f88d90afa6d8e46a56ef27de9abb231768545eac0ff40460e67fff", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9621985540445571, "standard_deviation": 0.0033389742123327874, "minimum": 0.955519364151122, "maximum": 0.9662507493805702, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9049822396721865, "standard_deviation": 0.029755412394600333, "minimum": 1.8615540718237056, "maximum": 1.9390067037937644, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8985861236191027, "standard_deviation": 0.3247668275578323, "minimum": 0.4764290091080166, "maximum": 1.3470864052145934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21312859203591578, "standard_deviation": 0.009092130588182422, "minimum": 0.20206977625498268, "maximum": 0.23048730526362612, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9879407993327642, "standard_deviation": 0.0023387940897851896, "minimum": 0.9844187200531813, "maximum": 0.9904999911058296, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1139711802818293, "standard_deviation": 0.027007975715184096, "minimum": 1.0673406548223385, "maximum": 1.1421133590330643, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1887976524930686, "standard_deviation": 0.3626569005732904, "minimum": 0.7398306229651345, "maximum": 1.7886245307946143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11708903785970855, "standard_deviation": 0.009727659441121878, "minimum": 0.10217343973330746, "maximum": 0.1306949238806011, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793542272236886, "standard_deviation": 0.0010199181392498851, "minimum": 0.978349438615917, "maximum": 0.9809818412088898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0683047911748047, "standard_deviation": 0.01872494551157275, "minimum": 2.0392200130458322, "maximum": 2.0920538816444925, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5072337938688845, "standard_deviation": 0.25162869401431137, "minimum": 0.1574701428916844, "maximum": 0.9144986698228829, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17519109439601455, "standard_deviation": 0.004891980507439989, "minimum": 0.16705174680326657, "maximum": 0.18165432464619105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9928782813161773, "standard_deviation": 0.0008837612749728704, "minimum": 0.991621530782772, "maximum": 0.9938239649487727, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1813519063877957, "standard_deviation": 0.02557553875414366, "minimum": 1.135470136097323, "maximum": 1.2257879114514907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6429644466413813, "standard_deviation": 0.2086268251355416, "minimum": 0.37253617481101786, "maximum": 0.9256619452285223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10038759305613575, "standard_deviation": 0.006349430324067137, "minimum": 0.09327495673113957, "maximum": 0.11036449995722022, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9764722125296693, "standard_deviation": 0.002546534645916861, "minimum": 0.9726026387419747, "maximum": 0.9798339503719429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.897541709369726, "standard_deviation": 0.0530487988190953, "minimum": 1.8056565938280804, "maximum": 1.9761733675962374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.35604885342060566, "standard_deviation": 0.08569006236974268, "minimum": 0.22769518915010542, "maximum": 0.4991485710617332, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17633739505085028, "standard_deviation": 0.01018427440685866, "minimum": 0.16234035008014613, "maximum": 0.19207765293937967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9919286250764987, "standard_deviation": 0.002360889832104009, "minimum": 0.9877179239617915, "maximum": 0.9941141004897573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0760153687516376, "standard_deviation": 0.05102352984662185, "minimum": 1.0010600857718182, "maximum": 1.154375567552596, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3629829880908284, "standard_deviation": 0.11409093882966957, "minimum": 0.24288038806849496, "maximum": 0.5991629389341683, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09991773977353074, "standard_deviation": 0.014369068364275844, "minimum": 0.08615810415297982, "maximum": 0.12378070303766021, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15725888160667612, "standard_deviation": 0.03918169611176538, "minimum": 0.07338641887632538, "maximum": 0.19941842814289532, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6261458817527705, "standard_deviation": 0.00863777068672901, "minimum": 5.612350996014516, "maximum": 5.635168296411049, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41788855717957757, "standard_deviation": 0.06074502831516887, "minimum": 0.37217555008211034, "maximum": 0.516373917448644, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.025137233573108677, "standard_deviation": 0.08041882164931426, "minimum": -0.13950436896570761, "maximum": 0.10438002448682591, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9857731903681994, "standard_deviation": 0.0025534650314349983, "minimum": 0.9820797470824205, "maximum": 0.9885673419267007, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2699415482396719, "standard_deviation": 0.021226958377757232, "minimum": 1.23825532420377, "maximum": 1.3037953341909516, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2930983594687068, "standard_deviation": 0.16750447926997059, "minimum": 0.041258926592985974, "maximum": 0.5464390459688281, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12153136528750899, "standard_deviation": 0.01156403867236244, "minimum": 0.10885168247287665, "maximum": 0.1386861489297342, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.995257654331325, "standard_deviation": 0.0016603640975623554, "minimum": 0.9923059825362762, "maximum": 0.9970979605878262, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6573389405900381, "standard_deviation": 0.02660638540097366, "minimum": 0.629083132189232, "maximum": 0.7076346345433333, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3995564910212603, "standard_deviation": 0.1335289126037876, "minimum": 0.14420617788882453, "maximum": 0.5347823697550204, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06634615442539171, "standard_deviation": 0.011769204838912724, "minimum": 0.0523740994807452, "maximum": 0.08670947517238782, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123288-0003", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.4153120517730713, "predicted_mos": 2.4153120517730713, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b06722f948b5bc561575679cbbdfeba37498e5860b0a17f85dba29f8f9476af0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.4504498997703195, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.213244468239772, "全帯域_10ms_rms_envelope_nrmse": 0.12163749879483002, "全帯域_10ms_envelope_correlation": 0.9858020105829903, "全帯域_10ms_modulation_mae_db": 0.3155856673707164, "全帯域_40ms_log_rms_mae_db": 0.623262006281624, "全帯域_40ms_rms_envelope_nrmse": 0.06399384496076124, "全帯域_40ms_envelope_correlation": 0.9956121001423223, "全帯域_40ms_modulation_mae_db": 0.4607512244241317, "300–1000 Hz_10ms_log_rms_mae_db": 2.0464454377471952, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18617389158034273, "300–1000 Hz_10ms_envelope_correlation": 0.9737123993000729, "300–1000 Hz_10ms_modulation_mae_db": 0.3104911103760157, "300–1000 Hz_40ms_log_rms_mae_db": 1.2269569477708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1053058547582696, "300–1000 Hz_40ms_envelope_correlation": 0.9908020466004027, "300–1000 Hz_40ms_modulation_mae_db": 0.47200370193366714, "1000–3000 Hz_10ms_log_rms_mae_db": 2.178143244033655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24387263571486897, "1000–3000 Hz_10ms_envelope_correlation": 0.9510974941057633, "1000–3000 Hz_10ms_modulation_mae_db": 0.5762672895287219, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3198058196109215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14583569111804723, "1000–3000 Hz_40ms_envelope_correlation": 0.9804029482202139, "1000–3000 Hz_40ms_modulation_mae_db": 0.8095930900192453, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6941570114879907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20897755217220332, "3000–6000 Hz_10ms_envelope_correlation": 0.970535463932341, "3000–6000 Hz_10ms_modulation_mae_db": 0.6180266639751492, "3000–6000 Hz_40ms_log_rms_mae_db": 1.613872980561298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12409370537060897, "3000–6000 Hz_40ms_envelope_correlation": 0.9893150796709999, "3000–6000 Hz_40ms_modulation_mae_db": 0.7154103416263881, "mr_stft_db": 5.996203450414023, "band_mae_db": 0.17682358592314162, "rms_gain_error_db": 0.014351284404398253, "output_peak": 0.4433936842939487}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2117536508779143, "全帯域_10ms_rms_envelope_nrmse": 0.10725728394659584, "全帯域_10ms_envelope_correlation": 0.9889252859147095, "全帯域_10ms_modulation_mae_db": 0.37747712853098564, "全帯域_40ms_log_rms_mae_db": 0.6425785397774447, "全帯域_40ms_rms_envelope_nrmse": 0.05858346455767312, "全帯域_40ms_envelope_correlation": 0.996418830855713, "全帯域_40ms_modulation_mae_db": 0.48097325586622786, "300–1000 Hz_10ms_log_rms_mae_db": 2.0693186553466822, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17178775765321508, "300–1000 Hz_10ms_envelope_correlation": 0.9778253086641623, "300–1000 Hz_10ms_modulation_mae_db": 0.7076467959819684, "300–1000 Hz_40ms_log_rms_mae_db": 1.1903852188726811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10904559036039436, "300–1000 Hz_40ms_envelope_correlation": 0.9910548610131122, "300–1000 Hz_40ms_modulation_mae_db": 0.7569185516308979, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2271433009300274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24292157649427687, "1000–3000 Hz_10ms_envelope_correlation": 0.9538120553535896, "1000–3000 Hz_10ms_modulation_mae_db": 0.34683694431708884, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3602317203750112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1436513606828579, "1000–3000 Hz_40ms_envelope_correlation": 0.982544606118954, "1000–3000 Hz_40ms_modulation_mae_db": 0.5268134488980236, "3000–6000 Hz_10ms_log_rms_mae_db": 2.748391608772054, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23007056150958946, "3000–6000 Hz_10ms_envelope_correlation": 0.9640751867602106, "3000–6000 Hz_10ms_modulation_mae_db": 0.4810270792773304, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6333012882655358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1344679271243694, "3000–6000 Hz_40ms_envelope_correlation": 0.9871052585797999, "3000–6000 Hz_40ms_modulation_mae_db": 0.6171454115129965, "mr_stft_db": 6.019229253568011, "band_mae_db": 0.23266419631580848, "rms_gain_error_db": -0.10768749773700845, "output_peak": 0.3917380744266909}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2664324865648395, "全帯域_10ms_rms_envelope_nrmse": 0.11034899535739125, "全帯域_10ms_envelope_correlation": 0.9883546013185922, "全帯域_10ms_modulation_mae_db": 0.29081732940445243, "全帯域_40ms_log_rms_mae_db": 0.6463903416507124, "全帯域_40ms_rms_envelope_nrmse": 0.061654437848287606, "全帯域_40ms_envelope_correlation": 0.9961717467497813, "全帯域_40ms_modulation_mae_db": 0.3655035018622126, "300–1000 Hz_10ms_log_rms_mae_db": 2.0816882626358306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17546005319154198, "300–1000 Hz_10ms_envelope_correlation": 0.9767677781200574, "300–1000 Hz_10ms_modulation_mae_db": 0.6339021685804047, "300–1000 Hz_40ms_log_rms_mae_db": 1.198797903384686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10076522306426551, "300–1000 Hz_40ms_envelope_correlation": 0.9925501996507224, "300–1000 Hz_40ms_modulation_mae_db": 0.4448876582053656, "1000–3000 Hz_10ms_log_rms_mae_db": 2.231838550365595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2476674092544153, "1000–3000 Hz_10ms_envelope_correlation": 0.9488793580146669, "1000–3000 Hz_10ms_modulation_mae_db": 1.1361413704049281, "1000–3000 Hz_40ms_log_rms_mae_db": 1.347415840498163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14348972425909395, "1000–3000 Hz_40ms_envelope_correlation": 0.9806062283689411, "1000–3000 Hz_40ms_modulation_mae_db": 1.4269649578795676, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6688875075954543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2108791745890294, "3000–6000 Hz_10ms_envelope_correlation": 0.9700561746032754, "3000–6000 Hz_10ms_modulation_mae_db": 0.6883600257840067, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5934697015481665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11575322185295636, "3000–6000 Hz_40ms_envelope_correlation": 0.99053513337663, "3000–6000 Hz_40ms_modulation_mae_db": 0.7597547916105314, "mr_stft_db": 5.9995909471906925, "band_mae_db": 0.13323561193158837, "rms_gain_error_db": -0.15039013175886937, "output_peak": 0.3844247304632079}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2993591461386698, "全帯域_10ms_rms_envelope_nrmse": 0.116693862080847, "全帯域_10ms_envelope_correlation": 0.9868718772360283, "全帯域_10ms_modulation_mae_db": 0.571158717258283, "全帯域_40ms_log_rms_mae_db": 0.6990367910244222, "全帯域_40ms_rms_envelope_nrmse": 0.06313896950742835, "全帯域_40ms_envelope_correlation": 0.9956921308842384, "全帯域_40ms_modulation_mae_db": 0.6122601138520409, "300–1000 Hz_10ms_log_rms_mae_db": 2.1152179184859934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16490626593082638, "300–1000 Hz_10ms_envelope_correlation": 0.9792010459662628, "300–1000 Hz_10ms_modulation_mae_db": 0.47161948000018583, "300–1000 Hz_40ms_log_rms_mae_db": 1.1649058872350029, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08996252748605056, "300–1000 Hz_40ms_envelope_correlation": 0.993577431747064, "300–1000 Hz_40ms_modulation_mae_db": 0.5465109996284995, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1972082244952955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23355596140281862, "1000–3000 Hz_10ms_envelope_correlation": 0.9540880976898499, "1000–3000 Hz_10ms_modulation_mae_db": 0.6924131178011077, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3250840269837796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12909130208267894, "1000–3000 Hz_40ms_envelope_correlation": 0.9844720825797261, "1000–3000 Hz_40ms_modulation_mae_db": 0.8998274439918864, "3000–6000 Hz_10ms_log_rms_mae_db": 2.684544240322085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22336300970408043, "3000–6000 Hz_10ms_envelope_correlation": 0.9662793587161181, "3000–6000 Hz_10ms_modulation_mae_db": 0.47792007274512266, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6027160144939632, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12824519361586414, "3000–6000 Hz_40ms_envelope_correlation": 0.9884641123479357, "3000–6000 Hz_40ms_modulation_mae_db": 0.631596962972958, "mr_stft_db": 6.007950454025365, "band_mae_db": 0.11163902370843658, "rms_gain_error_db": 0.0012033770760811434, "output_peak": 0.3974642835750528}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.201924862255435, "全帯域_10ms_rms_envelope_nrmse": 0.11067276013814475, "全帯域_10ms_envelope_correlation": 0.9882281211403143, "全帯域_10ms_modulation_mae_db": 0.41072208902763013, "全帯域_40ms_log_rms_mae_db": 0.6138075135430158, "全帯域_40ms_rms_envelope_nrmse": 0.05395181093856322, "全帯域_40ms_envelope_correlation": 0.9970420237399825, "全帯域_40ms_modulation_mae_db": 0.47267929992641355, "300–1000 Hz_10ms_log_rms_mae_db": 2.0935867388916662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16010149141236804, "300–1000 Hz_10ms_envelope_correlation": 0.9804779383404458, "300–1000 Hz_10ms_modulation_mae_db": 0.25223577483403953, "300–1000 Hz_40ms_log_rms_mae_db": 1.1872539374404085, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08316049383565423, "300–1000 Hz_40ms_envelope_correlation": 0.9946835551438851, "300–1000 Hz_40ms_modulation_mae_db": 0.17497025731261961, "1000–3000 Hz_10ms_log_rms_mae_db": 2.205594839134232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23958214188862428, "1000–3000 Hz_10ms_envelope_correlation": 0.9516025801912085, "1000–3000 Hz_10ms_modulation_mae_db": 1.0133774758492802, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3016925057797464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1290357686813318, "1000–3000 Hz_40ms_envelope_correlation": 0.9846436732496875, "1000–3000 Hz_40ms_modulation_mae_db": 1.3891102977906156, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6606978022740315, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21312372053664208, "3000–6000 Hz_10ms_envelope_correlation": 0.9697351561211094, "3000–6000 Hz_10ms_modulation_mae_db": 0.7034056579858458, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6106742065262536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1181984986112644, "3000–6000 Hz_40ms_envelope_correlation": 0.99022601731166, "3000–6000 Hz_40ms_modulation_mae_db": 0.6671074280546484, "mr_stft_db": 6.013780914290102, "band_mae_db": 0.1846352760357787, "rms_gain_error_db": -0.12782161359067426, "output_peak": 0.40123491599935945}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2245959916775786, "全帯域_10ms_rms_envelope_nrmse": 0.11952736601877792, "全帯域_10ms_envelope_correlation": 0.9862237174036996, "全帯域_10ms_modulation_mae_db": 0.3217386392327839, "全帯域_40ms_log_rms_mae_db": 0.6348964446558992, "全帯域_40ms_rms_envelope_nrmse": 0.0638487281168451, "全帯域_40ms_envelope_correlation": 0.995591247320613, "全帯域_40ms_modulation_mae_db": 0.3398069841629123, "300–1000 Hz_10ms_log_rms_mae_db": 2.0380547798640443, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17773675027594926, "300–1000 Hz_10ms_envelope_correlation": 0.9758552161965138, "300–1000 Hz_10ms_modulation_mae_db": 0.23627866044983892, "300–1000 Hz_40ms_log_rms_mae_db": 1.1348303144496517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10937972302268209, "300–1000 Hz_40ms_envelope_correlation": 0.9899833880152277, "300–1000 Hz_40ms_modulation_mae_db": 0.23294966639187345, "1000–3000 Hz_10ms_log_rms_mae_db": 2.191576232261875, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2602733464300729, "1000–3000 Hz_10ms_envelope_correlation": 0.9438041486057106, "1000–3000 Hz_10ms_modulation_mae_db": 0.822502224442344, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3518578777139763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13859305972612154, "1000–3000 Hz_40ms_envelope_correlation": 0.9820252435054244, "1000–3000 Hz_40ms_modulation_mae_db": 0.983389455223335, "3000–6000 Hz_10ms_log_rms_mae_db": 2.691913099644677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2299723566215752, "3000–6000 Hz_10ms_envelope_correlation": 0.9639822691291899, "3000–6000 Hz_10ms_modulation_mae_db": 1.1365416115701923, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6192668136593902, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1341283475199577, "3000–6000 Hz_40ms_envelope_correlation": 0.987416994273011, "3000–6000 Hz_40ms_modulation_mae_db": 1.286430253637177, "mr_stft_db": 6.008852185315599, "band_mae_db": 0.12576808198329287, "rms_gain_error_db": -0.009343613967535514, "output_peak": 0.43871955161118176}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.276389058192018, "全帯域_10ms_rms_envelope_nrmse": 0.13236054798863184, "全帯域_10ms_envelope_correlation": 0.98333783922769, "全帯域_10ms_modulation_mae_db": 0.4892607207806794, "全帯域_40ms_log_rms_mae_db": 0.6952643035074446, "全帯域_40ms_rms_envelope_nrmse": 0.07778657036656424, "全帯域_40ms_envelope_correlation": 0.993606959318822, "全帯域_40ms_modulation_mae_db": 0.6024945756809653, "300–1000 Hz_10ms_log_rms_mae_db": 2.0326773157806066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16355434132880836, "300–1000 Hz_10ms_envelope_correlation": 0.9796505390894767, "300–1000 Hz_10ms_modulation_mae_db": 0.5712932957888707, "300–1000 Hz_40ms_log_rms_mae_db": 1.1441649247212438, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09435819899356002, "300–1000 Hz_40ms_envelope_correlation": 0.9925811002595042, "300–1000 Hz_40ms_modulation_mae_db": 0.8785718311605564, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1699676552425675, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2351800542368451, "1000–3000 Hz_10ms_envelope_correlation": 0.9535103858691786, "1000–3000 Hz_10ms_modulation_mae_db": 0.8439326007826929, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2590116321613845, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12899780788920506, "1000–3000 Hz_40ms_envelope_correlation": 0.9844638859428009, "1000–3000 Hz_40ms_modulation_mae_db": 0.9382543219472433, "3000–6000 Hz_10ms_log_rms_mae_db": 2.718851591911233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19660852618107102, "3000–6000 Hz_10ms_envelope_correlation": 0.9740195840494946, "3000–6000 Hz_10ms_modulation_mae_db": 0.9624169804296395, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6318282916613887, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1130104184627766, "3000–6000 Hz_40ms_envelope_correlation": 0.9911295328571513, "3000–6000 Hz_40ms_modulation_mae_db": 1.1703229171610474, "mr_stft_db": 6.0138428288058465, "band_mae_db": 0.07968673134953257, "rms_gain_error_db": 0.07626258119475644, "output_peak": 0.42357510087173456}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2107252882870092, "全帯域_10ms_rms_envelope_nrmse": 0.11368202457826107, "全帯域_10ms_envelope_correlation": 0.9879259792724833, "全帯域_10ms_modulation_mae_db": 0.4305648892497333, "全帯域_40ms_log_rms_mae_db": 0.636788261463453, "全帯域_40ms_rms_envelope_nrmse": 0.06364773919838258, "全帯域_40ms_envelope_correlation": 0.9963086060158918, "全帯域_40ms_modulation_mae_db": 0.556032551315433, "300–1000 Hz_10ms_log_rms_mae_db": 2.114611667970694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17755782999211014, "300–1000 Hz_10ms_envelope_correlation": 0.9775915777688142, "300–1000 Hz_10ms_modulation_mae_db": 0.5395926385036005, "300–1000 Hz_40ms_log_rms_mae_db": 1.2751042652176972, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1125953578049104, "300–1000 Hz_40ms_envelope_correlation": 0.9922807442421447, "300–1000 Hz_40ms_modulation_mae_db": 0.6635335530456731, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2123714824142624, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23055293380341013, "1000–3000 Hz_10ms_envelope_correlation": 0.9555567077311005, "1000–3000 Hz_10ms_modulation_mae_db": 0.8015222567818112, "1000–3000 Hz_40ms_log_rms_mae_db": 1.29818726874508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11862183216537603, "1000–3000 Hz_40ms_envelope_correlation": 0.986753648294643, "1000–3000 Hz_40ms_modulation_mae_db": 0.8850791749852155, "3000–6000 Hz_10ms_log_rms_mae_db": 2.604391618180023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20758688089299357, "3000–6000 Hz_10ms_envelope_correlation": 0.970871237502966, "3000–6000 Hz_10ms_modulation_mae_db": 0.6155991440073417, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5468734057006295, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12035043320770021, "3000–6000 Hz_40ms_envelope_correlation": 0.9896529191896034, "3000–6000 Hz_40ms_modulation_mae_db": 0.6364170780159789, "mr_stft_db": 6.0158499512927035, "band_mae_db": 0.20274948329033382, "rms_gain_error_db": -0.2309701916874085, "output_peak": 0.3788109954949943}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.427151675889137, "spectral_convergence": 0.3562454187648366, "spectral_convergence_status": "ok", "amplitude_floor": 1.3116896377207775e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.870929455339768, "spectral_convergence": 0.3598559221085244, "spectral_convergence_status": "ok", "amplitude_floor": 8.457684224428572e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.690529220013166, "spectral_convergence": 0.37576695886222866, "spectral_convergence_status": "ok", "amplitude_floor": 6.717993311155654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008437417957183396, "output_power": 0.0008471493530878833, "delta_db": 0.017504206190704058, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011741200681074208, "output_power": 0.001120609498682468, "delta_db": -0.20258211713249122, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004760668283187788, "output_power": 0.0005329903341673281, "delta_db": 0.49051411637296544, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001833223674641814, "output_power": 0.0001815844366403426, "delta_db": -0.04136834200745631, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.526890589901711e-05, "output_power": 6.857375252778016e-05, "delta_db": 0.21451583109473765, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6955820206525705e-05, "output_power": 3.7768395550193994e-05, "delta_db": 0.09445690274049512, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17682358592314162, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.014351284404398253, "pitch": null, "output_rms_dbfs": -25.546019766627598, "output_peak": 0.4433936842939487, "reference_rms_dbfs": -25.560371051032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.996203450414023}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001544797211376954, "log_rms_mae_db": 1.213244468239772, "rms_envelope_nrmse": 0.12163749879483002, "envelope_correlation": 0.9858020105829903, "rms_gain_error_db": 0.014353209274872443, "reference_mean_power": 0.002781784143506645, "output_mean_power": 0.002790993006063829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.487428460245723e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008814675226478596, "output_power": 0.0008831285784219824, "reference_active": true, "delta_db": 0.008176234430860919}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001997406840389717, "output_power": 0.00019087065439547186, "reference_active": true, "delta_db": -0.19727370367540634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0345220429658976e-05, "output_power": 3.401428812843949e-05, "reference_active": true, "delta_db": -0.7413070640058821}], "active_band_count": 3, "active_band_mae_db": 0.3155856673707164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013989081102856241, "log_rms_mae_db": 0.623262006281624, "rms_envelope_nrmse": 0.06399384496076124, "envelope_correlation": 0.9956121001423223, "rms_gain_error_db": 0.01426488412982525, "reference_mean_power": 0.002795409730019158, "output_mean_power": 0.0028046066593411176}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3091173503801554e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008259246223406256, "output_power": 0.0008302962160510292, "reference_active": true, "delta_db": 0.022926452688189748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.583418527096058e-05, "output_power": 8.936865003252483e-05, "reference_active": true, "delta_db": -0.303352573693276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329771424834818e-06, "output_power": 2.611060635115345e-06, "reference_active": true, "delta_db": -1.0559746468909295}], "active_band_count": 3, "active_band_mae_db": 0.4607512244241317}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000132208474335363, "log_rms_mae_db": 2.0464454377471952, "rms_envelope_nrmse": 0.18617389158034273, "envelope_correlation": 0.9737123993000729, "rms_gain_error_db": 0.019404152587458524, "reference_mean_power": 0.001270855716483403, "output_mean_power": 0.0012765465670976591}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.478174051796597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005667765898001897, "output_power": 0.0005554641316131207, "reference_active": true, "delta_db": -0.08755883630167273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011084740591226797, "output_power": 9.82675825798374e-05, "reference_active": true, "delta_db": -0.5231526169315447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2190579526022385e-05, "output_power": 2.3891577834898164e-05, "reference_active": true, "delta_db": 0.3207618778948297}], "active_band_count": 3, "active_band_mae_db": 0.3104911103760157}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012422751968092996, "log_rms_mae_db": 1.2269569477708, "rms_envelope_nrmse": 0.1053058547582696, "envelope_correlation": 0.9908020466004027, "rms_gain_error_db": 0.019391236541140826, "reference_mean_power": 0.0012772403358267607, "output_mean_power": 0.0012829559610043897}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.710312666254066e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005574547781532276, "output_power": 0.0005499806471120431, "reference_active": true, "delta_db": -0.058622346364816114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.253668134122962e-05, "output_power": 4.484887084451004e-05, "reference_active": true, "delta_db": -0.6871112229394659}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7043530583570796e-06, "output_power": 1.46060172970559e-06, "reference_active": true, "delta_db": -0.6702775364967194}], "active_band_count": 3, "active_band_mae_db": 0.47200370193366714}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.199705053096058e-05, "log_rms_mae_db": 2.178143244033655, "rms_envelope_nrmse": 0.24387263571486897, "envelope_correlation": 0.9510974941057633, "rms_gain_error_db": 0.04277862383566793, "reference_mean_power": 0.00020286667816607547, "output_mean_power": 0.00020487481777803026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3687094204678632e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.826983537989376e-05, "output_power": 7.438968493805908e-05, "reference_active": true, "delta_db": -0.22081701047211352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2413787132715612e-05, "output_power": 2.0704511192313635e-05, "reference_active": true, "delta_db": -0.3445026124531626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388555256663533e-06, "output_power": 4.887136068862132e-06, "reference_active": true, "delta_db": -1.1634822456608895}], "active_band_count": 3, "active_band_mae_db": 0.5762672895287219}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.313537286332709e-05, "log_rms_mae_db": 1.3198058196109215, "rms_envelope_nrmse": 0.14583569111804723, "envelope_correlation": 0.9804029482202139, "rms_gain_error_db": 0.04266922549663309, "reference_mean_power": 0.00020386380860129322, "output_mean_power": 0.00020587663254565316}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.180981609421061e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.558718162537124e-05, "output_power": 7.313150145542269e-05, "reference_active": true, "delta_db": -0.14343662389692566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0576597086854303e-05, "output_power": 1.02788234452116e-05, "reference_active": true, "delta_db": -0.12402553875494494}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.38427493679316e-07, "output_power": 3.27337316216014e-07, "reference_active": true, "delta_db": -2.161317107405865}], "active_band_count": 3, "active_band_mae_db": 0.8095930900192453}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.0553751915640445e-05, "log_rms_mae_db": 2.6941570114879907, "rms_envelope_nrmse": 0.20897755217220332, "envelope_correlation": 0.970535463932341, "rms_gain_error_db": -0.134534399283032, "reference_mean_power": 5.25173391881524e-05, "output_mean_power": 5.091541340443509e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.578963632678691e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2240876569039177e-05, "output_power": 2.021793476222293e-05, "reference_active": true, "delta_db": -0.4141510902780253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.96335110256818e-06, "output_power": 7.1148291026633055e-06, "reference_active": true, "delta_db": -0.489313912612916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4165297947437381e-06, "output_power": 1.138057525224387e-06, "reference_active": true, "delta_db": -0.9506149890345061}], "active_band_count": 3, "active_band_mae_db": 0.6180266639751492}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.394748258818145e-05, "log_rms_mae_db": 1.613872980561298, "rms_envelope_nrmse": 0.12409370537060897, "envelope_correlation": 0.9893150796709999, "rms_gain_error_db": -0.13937593592875358, "reference_mean_power": 5.256947927681562e-05, "output_mean_power": 5.090917762173201e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.187575200111859e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.180909445738007e-05, "output_power": 1.9709418915429476e-05, "reference_active": true, "delta_db": -0.439638131230501}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0063771970727005e-06, "output_power": 3.647148848263523e-06, "reference_active": true, "delta_db": -0.40798347145772396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0500724932970203e-07, "output_power": 7.786787804842503e-08, "reference_active": true, "delta_db": -1.2986094221909394}], "active_band_count": 3, "active_band_mae_db": 0.7154103416263881}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8005944515977695, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7eed9b1e052dbd9d7eabe78224ec0664c9104f11075c46ee4c40dc4f9060e12a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9515438534451335, "standard_deviation": 0.0037567733183911873, "minimum": 0.9438041486057106, "maximum": 0.9555567077311005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2017304411096887, "standard_deviation": 0.021939090525909217, "minimum": 2.1699676552425675, "maximum": 2.231838550365595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7791241599884968, "standard_deviation": 0.2462464854867686, "minimum": 0.34683694431708884, "maximum": 1.1361413704049281, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24170075740316652, "standard_deviation": 0.009444342422172562, "minimum": 0.23055293380341013, "maximum": 0.2602733464300729, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9832390395350489, "standard_deviation": 0.0022128443761467585, "minimum": 0.9804029482202139, "maximum": 0.986753648294643, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3204108364835077, "standard_deviation": 0.03367962366497538, "minimum": 1.2590116321613845, "maximum": 1.3602317203750112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9823790238418916, "standard_deviation": 0.2973002302063252, "minimum": 0.5268134488980236, "maximum": 1.4269649578795676, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13466456832558904, "standard_deviation": 0.009644384618231808, "minimum": 0.11862183216537603, "maximum": 0.14583569111804723, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9686943038518382, "standard_deviation": 0.0035616420586821475, "minimum": 0.9639822691291899, "maximum": 0.9740195840494946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.6839793100234437, "standard_deviation": 0.04248674052288736, "minimum": 2.604391618180023, "maximum": 2.748391608772054, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7104121544718285, "standard_deviation": 0.2298517536155172, "minimum": 0.47792007274512266, "maximum": 1.1365416115701923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21507272277589806, "standard_deviation": 0.011778691766850372, "minimum": 0.19660852618107102, "maximum": 0.23007056150958946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.989230630950849, "standard_deviation": 0.0014578773445503093, "minimum": 0.9871052585797999, "maximum": 0.9911295328571513, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6065003378020781, "standard_deviation": 0.027604465879775938, "minimum": 1.5468734057006295, "maximum": 1.6333012882655358, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8105231480739657, "standard_deviation": 0.2639961688090359, "minimum": 0.6171454115129965, "maximum": 1.286430253637177, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12353096822068721, "standard_deviation": 0.008143257097777717, "minimum": 0.1130104184627766, "maximum": 0.1344679271243694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9776352254307257, "standard_deviation": 0.0022020055500049794, "minimum": 0.9737123993000729, "maximum": 0.9804779383404458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0739500970903393, "standard_deviation": 0.03288887382528674, "minimum": 2.0326773157806066, "maximum": 2.1152179184859934, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.46538249056436554, "standard_deviation": 0.17958899400425168, "minimum": 0.23627866044983892, "maximum": 0.7076467959819684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17215979767064526, "standard_deviation": 0.008780956936650822, "minimum": 0.16010149141236804, "maximum": 0.18617389158034273, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.992189165834008, "standard_deviation": 0.0015361272208701585, "minimum": 0.9899833880152277, "maximum": 0.9946835551438851, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1902999248865214, "standard_deviation": 0.045461423975876235, "minimum": 1.1348303144496517, "maximum": 1.2751042652176972, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5212932774136441, "standard_deviation": 0.24359932486112992, "minimum": 0.17497025731261961, "maximum": 0.8785718311605564, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10057162116572335, "standard_deviation": 0.010492979578915193, "minimum": 0.08316049383565423, "maximum": 0.1125953578049104, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15590024881723913, "standard_deviation": 0.05146459499691475, "minimum": 0.07968673134953257, "maximum": 0.23266419631580848, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.009412498112793, "standard_deviation": 0.008017068470521067, "minimum": 5.996203450414023, "maximum": 6.019229253568011, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40742016709202133, "standard_deviation": 0.024684602140510842, "minimum": 0.3788109954949943, "maximum": 0.4433936842939487, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06679947575828253, "standard_deviation": 0.10303447096454348, "minimum": -0.2309701916874085, "maximum": 0.07626258119475644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9869586790120635, "standard_deviation": 0.0018267598410923237, "minimum": 0.98333783922769, "maximum": 0.9889252859147095, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2380531190291544, "standard_deviation": 0.036982135492965085, "minimum": 1.201924862255435, "maximum": 1.2993591461386698, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.400915647606908, "standard_deviation": 0.09560634062311464, "minimum": 0.29081732940445243, "maximum": 0.571158717258283, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11652254236293497, "standard_deviation": 0.008038387424121923, "minimum": 0.10725728394659584, "maximum": 0.13236054798863184, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9958054556284205, "standard_deviation": 0.0010157445844649062, "minimum": 0.993606959318822, "maximum": 0.9970420237399825, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6490030252380019, "standard_deviation": 0.031489462829481944, "minimum": 0.6138075135430158, "maximum": 0.6990367910244222, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.48631268838629216, "standard_deviation": 0.10081694329163561, "minimum": 0.3398069841629123, "maximum": 0.6122601138520409, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06332569568681318, "standard_deviation": 0.006794189068466739, "minimum": 0.05395181093856322, "maximum": 0.07778657036656424, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123288-0003", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.660795211791992, "predicted_mos": 2.660795211791992, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ea3cf48eaf8953fc33a2c6b9324ba71c43620cdf35a6cfae930eb3be31a3b650", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.129498099908233, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2133609809879449, "全帯域_10ms_rms_envelope_nrmse": 0.11130086651255629, "全帯域_10ms_envelope_correlation": 0.9881535978760422, "全帯域_10ms_modulation_mae_db": 0.5847866467203937, "全帯域_40ms_log_rms_mae_db": 0.6492658974072028, "全帯域_40ms_rms_envelope_nrmse": 0.06490663499425747, "全帯域_40ms_envelope_correlation": 0.9957121311220528, "全帯域_40ms_modulation_mae_db": 0.6259885925918608, "300–1000 Hz_10ms_log_rms_mae_db": 1.8043473982060232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18146628422285413, "300–1000 Hz_10ms_envelope_correlation": 0.9750634687119515, "300–1000 Hz_10ms_modulation_mae_db": 0.38357818409792466, "300–1000 Hz_40ms_log_rms_mae_db": 1.114070676082758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.127018825213804, "300–1000 Hz_40ms_envelope_correlation": 0.9872629459519814, "300–1000 Hz_40ms_modulation_mae_db": 0.5493450372441239, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7347672924617363, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21201907729362446, "1000–3000 Hz_10ms_envelope_correlation": 0.962438168072245, "1000–3000 Hz_10ms_modulation_mae_db": 0.7980042280857139, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0033424122274348, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13105048239002662, "1000–3000 Hz_40ms_envelope_correlation": 0.9845187518469369, "1000–3000 Hz_40ms_modulation_mae_db": 0.9347534930849871, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8185847719857746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17029411070991113, "3000–6000 Hz_10ms_envelope_correlation": 0.9801875793071393, "3000–6000 Hz_10ms_modulation_mae_db": 0.6283192021430246, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0591479377177804, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09894735781422023, "3000–6000 Hz_40ms_envelope_correlation": 0.9929452526829231, "3000–6000 Hz_40ms_modulation_mae_db": 0.5952025765622279, "mr_stft_db": 5.550620209993568, "band_mae_db": 0.1813940413045683, "rms_gain_error_db": -0.1559174034728275, "output_peak": 0.3847583807986789}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2572586159114025, "全帯域_10ms_rms_envelope_nrmse": 0.11201836553666512, "全帯域_10ms_envelope_correlation": 0.9879339020898557, "全帯域_10ms_modulation_mae_db": 0.47213632900095454, "全帯域_40ms_log_rms_mae_db": 0.6425860159311758, "全帯域_40ms_rms_envelope_nrmse": 0.060060934693167096, "全帯域_40ms_envelope_correlation": 0.9962798057801295, "全帯域_40ms_modulation_mae_db": 0.6965606701358652, "300–1000 Hz_10ms_log_rms_mae_db": 1.7547443579257322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15074158918230002, "300–1000 Hz_10ms_envelope_correlation": 0.9827086329151227, "300–1000 Hz_10ms_modulation_mae_db": 0.4104912536437193, "300–1000 Hz_40ms_log_rms_mae_db": 1.0555830961654535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08545465957135108, "300–1000 Hz_40ms_envelope_correlation": 0.994241436631397, "300–1000 Hz_40ms_modulation_mae_db": 0.7514223546251171, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7171167676231764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21359998682186232, "1000–3000 Hz_10ms_envelope_correlation": 0.9621794999827933, "1000–3000 Hz_10ms_modulation_mae_db": 0.5885033678839883, "1000–3000 Hz_40ms_log_rms_mae_db": 1.002851038653528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11121551462203136, "1000–3000 Hz_40ms_envelope_correlation": 0.9884403373715922, "1000–3000 Hz_40ms_modulation_mae_db": 0.7176897151186349, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7948838852125262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1662904493543284, "3000–6000 Hz_10ms_envelope_correlation": 0.9813229087040801, "3000–6000 Hz_10ms_modulation_mae_db": 0.288749951390413, "3000–6000 Hz_40ms_log_rms_mae_db": 1.060791951428974, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09151354162498346, "3000–6000 Hz_40ms_envelope_correlation": 0.9939298808334974, "3000–6000 Hz_40ms_modulation_mae_db": 0.43615623088168515, "mr_stft_db": 5.556769620934472, "band_mae_db": 0.12374070832000973, "rms_gain_error_db": -0.12765268929391416, "output_peak": 0.37611740128699367}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2852757614331398, "全帯域_10ms_rms_envelope_nrmse": 0.1164504602542006, "全帯域_10ms_envelope_correlation": 0.9869344001568782, "全帯域_10ms_modulation_mae_db": 0.2729293475432692, "全帯域_40ms_log_rms_mae_db": 0.6784286823065203, "全帯域_40ms_rms_envelope_nrmse": 0.0621301433329947, "全帯域_40ms_envelope_correlation": 0.9959764109505809, "全帯域_40ms_modulation_mae_db": 0.3593330328403363, "300–1000 Hz_10ms_log_rms_mae_db": 1.7912093014518846, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17085569874713671, "300–1000 Hz_10ms_envelope_correlation": 0.9776822859972052, "300–1000 Hz_10ms_modulation_mae_db": 0.5671277847457189, "300–1000 Hz_40ms_log_rms_mae_db": 1.1052535626226072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09945984016869377, "300–1000 Hz_40ms_envelope_correlation": 0.9920126180718437, "300–1000 Hz_40ms_modulation_mae_db": 0.6059103693602517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.723601524994786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21796041497131002, "1000–3000 Hz_10ms_envelope_correlation": 0.9604328600339707, "1000–3000 Hz_10ms_modulation_mae_db": 0.6820099432943724, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9825766977519214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11109648328304018, "1000–3000 Hz_40ms_envelope_correlation": 0.9885529348844615, "1000–3000 Hz_40ms_modulation_mae_db": 0.741835285750276, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8445325704357274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17599389963165954, "3000–6000 Hz_10ms_envelope_correlation": 0.978825494514079, "3000–6000 Hz_10ms_modulation_mae_db": 0.32580660130980155, "3000–6000 Hz_40ms_log_rms_mae_db": 1.161432016645716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08724309340995565, "3000–6000 Hz_40ms_envelope_correlation": 0.9945623780495889, "3000–6000 Hz_40ms_modulation_mae_db": 0.31507652304391726, "mr_stft_db": 5.560528179416849, "band_mae_db": 0.11732248153774837, "rms_gain_error_db": -0.11872838030762535, "output_peak": 0.39107653515751906}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3187935592686058, "全帯域_10ms_rms_envelope_nrmse": 0.11932776993070669, "全帯域_10ms_envelope_correlation": 0.9863065023597688, "全帯域_10ms_modulation_mae_db": 0.2231508706522379, "全帯域_40ms_log_rms_mae_db": 0.6466315854764719, "全帯域_40ms_rms_envelope_nrmse": 0.05757522992567534, "全帯域_40ms_envelope_correlation": 0.9966393245533957, "全帯域_40ms_modulation_mae_db": 0.17226048346414027, "300–1000 Hz_10ms_log_rms_mae_db": 1.8495011212965538, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18754488934198651, "300–1000 Hz_10ms_envelope_correlation": 0.9732313284619906, "300–1000 Hz_10ms_modulation_mae_db": 0.26106879976322833, "300–1000 Hz_40ms_log_rms_mae_db": 1.0549506524029622, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10089240407031344, "300–1000 Hz_40ms_envelope_correlation": 0.9923247875324515, "300–1000 Hz_40ms_modulation_mae_db": 0.332600009574292, "1000–3000 Hz_10ms_log_rms_mae_db": 1.692206690931341, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19495508853119567, "1000–3000 Hz_10ms_envelope_correlation": 0.968366984673006, "1000–3000 Hz_10ms_modulation_mae_db": 0.926576978543757, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0055420037639113, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10660190869251336, "1000–3000 Hz_40ms_envelope_correlation": 0.9898763842896278, "1000–3000 Hz_40ms_modulation_mae_db": 1.1479605786938576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.879704791691718, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18035635142327439, "3000–6000 Hz_10ms_envelope_correlation": 0.9778011451283605, "3000–6000 Hz_10ms_modulation_mae_db": 0.6090940009795084, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1510191463142454, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10183339255198008, "3000–6000 Hz_40ms_envelope_correlation": 0.9926898440169483, "3000–6000 Hz_40ms_modulation_mae_db": 0.45303557629020935, "mr_stft_db": 5.5516956396653265, "band_mae_db": 0.16167621324297726, "rms_gain_error_db": -0.13944057078324346, "output_peak": 0.38310754185342205}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2344701125249793, "全帯域_10ms_rms_envelope_nrmse": 0.10892969819850418, "全帯域_10ms_envelope_correlation": 0.9885520346657032, "全帯域_10ms_modulation_mae_db": 0.5163388084085526, "全帯域_40ms_log_rms_mae_db": 0.6680349807938376, "全帯域_40ms_rms_envelope_nrmse": 0.05658594209007194, "全帯域_40ms_envelope_correlation": 0.9965909981021119, "全帯域_40ms_modulation_mae_db": 0.48897753485625745, "300–1000 Hz_10ms_log_rms_mae_db": 1.7899839699020443, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16314357019944936, "300–1000 Hz_10ms_envelope_correlation": 0.9796039689565742, "300–1000 Hz_10ms_modulation_mae_db": 0.2752439332717053, "300–1000 Hz_40ms_log_rms_mae_db": 1.0823771212110678, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08903963538325972, "300–1000 Hz_40ms_envelope_correlation": 0.9934856683836688, "300–1000 Hz_40ms_modulation_mae_db": 0.14287207632532023, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7243860558385824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20564635482526308, "1000–3000 Hz_10ms_envelope_correlation": 0.96468048240728, "1000–3000 Hz_10ms_modulation_mae_db": 1.0290096425047945, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0000597599615575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11367598959120444, "1000–3000 Hz_40ms_envelope_correlation": 0.9881084367541197, "1000–3000 Hz_40ms_modulation_mae_db": 1.232335997728098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8741560677996743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18123611667193293, "3000–6000 Hz_10ms_envelope_correlation": 0.9778207662322955, "3000–6000 Hz_10ms_modulation_mae_db": 0.8583071378827647, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1055430135898852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10234109170261478, "3000–6000 Hz_40ms_envelope_correlation": 0.9930792228590478, "3000–6000 Hz_40ms_modulation_mae_db": 0.9974184423864099, "mr_stft_db": 5.551536020465868, "band_mae_db": 0.14972633724188047, "rms_gain_error_db": -0.0486711099290478, "output_peak": 0.3879389832521291}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2795104654802967, "全帯域_10ms_rms_envelope_nrmse": 0.12884384645411637, "全帯域_10ms_envelope_correlation": 0.9841976747843177, "全帯域_10ms_modulation_mae_db": 0.15312629559877883, "全帯域_40ms_log_rms_mae_db": 0.6722785362514924, "全帯域_40ms_rms_envelope_nrmse": 0.07324399713055035, "全帯域_40ms_envelope_correlation": 0.9943157959756804, "全帯域_40ms_modulation_mae_db": 0.128565416035369, "300–1000 Hz_10ms_log_rms_mae_db": 1.7529877573758348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16627235994436254, "300–1000 Hz_10ms_envelope_correlation": 0.9789290123976422, "300–1000 Hz_10ms_modulation_mae_db": 0.29413893462777535, "300–1000 Hz_40ms_log_rms_mae_db": 1.0278776123478734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09793631627558413, "300–1000 Hz_40ms_envelope_correlation": 0.9919935374334555, "300–1000 Hz_40ms_modulation_mae_db": 0.2903264851713558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7442872046403943, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21076114167027613, "1000–3000 Hz_10ms_envelope_correlation": 0.9629004425230696, "1000–3000 Hz_10ms_modulation_mae_db": 0.6574119991082356, "1000–3000 Hz_40ms_log_rms_mae_db": 1.034545443506677, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11111972309090935, "1000–3000 Hz_40ms_envelope_correlation": 0.9886445259502127, "1000–3000 Hz_40ms_modulation_mae_db": 0.7946336544952984, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7920457610776999, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1690408526304598, "3000–6000 Hz_10ms_envelope_correlation": 0.9804965506676737, "3000–6000 Hz_10ms_modulation_mae_db": 0.48987280977576986, "3000–6000 Hz_40ms_log_rms_mae_db": 1.078271876356913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0913907563993936, "3000–6000 Hz_40ms_envelope_correlation": 0.9939387741068337, "3000–6000 Hz_40ms_modulation_mae_db": 0.6049493945820478, "mr_stft_db": 5.556021069795108, "band_mae_db": 0.14285428326394148, "rms_gain_error_db": 0.06306374785030679, "output_peak": 0.5647706478043107}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2549498286324077, "全帯域_10ms_rms_envelope_nrmse": 0.10890944574909914, "全帯域_10ms_envelope_correlation": 0.9885464195530669, "全帯域_10ms_modulation_mae_db": 0.13036589661289613, "全帯域_40ms_log_rms_mae_db": 0.6360387459832552, "全帯域_40ms_rms_envelope_nrmse": 0.05324007762518448, "全帯域_40ms_envelope_correlation": 0.9969891877997286, "全帯域_40ms_modulation_mae_db": 0.3260842333337531, "300–1000 Hz_10ms_log_rms_mae_db": 1.8001255401730774, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16213108339064128, "300–1000 Hz_10ms_envelope_correlation": 0.9799598446410053, "300–1000 Hz_10ms_modulation_mae_db": 0.1656660161253366, "300–1000 Hz_40ms_log_rms_mae_db": 1.1055646424344494, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08439552709555875, "300–1000 Hz_40ms_envelope_correlation": 0.9944191024670765, "300–1000 Hz_40ms_modulation_mae_db": 0.35632274802886155, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7617946763097416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20529891744138928, "1000–3000 Hz_10ms_envelope_correlation": 0.9651582373515939, "1000–3000 Hz_10ms_modulation_mae_db": 0.4716945363888869, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0595418268595471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1106997996821264, "1000–3000 Hz_40ms_envelope_correlation": 0.9886752462201672, "1000–3000 Hz_40ms_modulation_mae_db": 0.699318246272158, "3000–6000 Hz_10ms_log_rms_mae_db": 1.880919666241063, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17902142657509043, "3000–6000 Hz_10ms_envelope_correlation": 0.9786857870295322, "3000–6000 Hz_10ms_modulation_mae_db": 0.24453138526755533, "3000–6000 Hz_40ms_log_rms_mae_db": 1.169488113480084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10284350176383722, "3000–6000 Hz_40ms_envelope_correlation": 0.9925206491798918, "3000–6000 Hz_40ms_modulation_mae_db": 0.18313210968154678, "mr_stft_db": 5.570545241567607, "band_mae_db": 0.12772884415731914, "rms_gain_error_db": -0.07336512066463144, "output_peak": 0.3855103065559246}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2873245287465636, "全帯域_10ms_rms_envelope_nrmse": 0.12869205745449372, "全帯域_10ms_envelope_correlation": 0.984004458362738, "全帯域_10ms_modulation_mae_db": 0.6326253443484128, "全帯域_40ms_log_rms_mae_db": 0.654309311143152, "全帯域_40ms_rms_envelope_nrmse": 0.07142375652782774, "全帯域_40ms_envelope_correlation": 0.9946565613633851, "全帯域_40ms_modulation_mae_db": 0.6293952697177578, "300–1000 Hz_10ms_log_rms_mae_db": 1.812657938645083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2022614543463512, "300–1000 Hz_10ms_envelope_correlation": 0.9685204662827128, "300–1000 Hz_10ms_modulation_mae_db": 0.33180168866839627, "300–1000 Hz_40ms_log_rms_mae_db": 1.0179810755821042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13707286819092662, "300–1000 Hz_40ms_envelope_correlation": 0.9844119890430263, "300–1000 Hz_40ms_modulation_mae_db": 0.43658008381929486, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7146266866147257, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.204688369951032, "1000–3000 Hz_10ms_envelope_correlation": 0.965045275554471, "1000–3000 Hz_10ms_modulation_mae_db": 0.7316158110836705, "1000–3000 Hz_40ms_log_rms_mae_db": 0.992473626514131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09979540947283742, "1000–3000 Hz_40ms_envelope_correlation": 0.9914220516198309, "1000–3000 Hz_40ms_modulation_mae_db": 0.8249306266799112, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8166811529724223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17415959021990482, "3000–6000 Hz_10ms_envelope_correlation": 0.9792772041611406, "3000–6000 Hz_10ms_modulation_mae_db": 0.3009532334874256, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1250763224630462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09651726635493275, "3000–6000 Hz_40ms_envelope_correlation": 0.9933664805256975, "3000–6000 Hz_40ms_modulation_mae_db": 0.24966321359782895, "mr_stft_db": 5.551023827403348, "band_mae_db": 0.20671053389492647, "rms_gain_error_db": -0.13422432370052859, "output_peak": 0.4045430166351233}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.851689314464895, "spectral_convergence": 0.3508755929758644, "spectral_convergence_status": "ok", "amplitude_floor": 1.3116896377207775e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.319465713697666, "spectral_convergence": 0.3498067240513574, "spectral_convergence_status": "ok", "amplitude_floor": 8.457684224428572e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.480705601818141, "spectral_convergence": 0.37004695304073737, "spectral_convergence_status": "ok", "amplitude_floor": 6.717993311155654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008437417957183396, "output_power": 0.0008554608137435081, "delta_db": 0.0599055760693151, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011741200681074208, "output_power": 0.001100343858779863, "delta_db": -0.28184087177283823, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004760668283187788, "output_power": 0.0004560524009503192, "delta_db": -0.18653175015453147, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001833223674641814, "output_power": 0.00016793532482199775, "delta_db": -0.3807339855510142, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.526890589901711e-05, "output_power": 6.587599978930746e-05, "delta_db": 0.040208867973420166, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6955820206525705e-05, "output_power": 3.579056265863743e-05, "delta_db": -0.13914319630629074, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1813940413045683, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1559174034728275, "pitch": null, "output_rms_dbfs": -25.716288454504827, "output_peak": 0.3847583807986789, "reference_rms_dbfs": -25.560371051032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.550620209993568}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001544797211376954, "log_rms_mae_db": 1.2133609809879449, "rms_envelope_nrmse": 0.11130086651255629, "envelope_correlation": 0.9881535978760422, "rms_gain_error_db": -0.1559180228611449, "reference_mean_power": 0.002781784143506645, "output_mean_power": 0.002683685532705981}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.487428460245723e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008814675226478596, "output_power": 0.0008396287185491002, "reference_active": true, "delta_db": -0.2111903079181837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001997406840389717, "output_power": 0.00017416227159236988, "reference_active": true, "delta_db": -0.5951245228541566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0345220429658976e-05, "output_power": 3.2433034323476305e-05, "reference_active": true, "delta_db": -0.9480451093888408}], "active_band_count": 3, "active_band_mae_db": 0.5847866467203937}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013989081102856241, "log_rms_mae_db": 0.6492658974072028, "rms_envelope_nrmse": 0.06490663499425747, "envelope_correlation": 0.9957121311220528, "rms_gain_error_db": -0.15589387280075895, "reference_mean_power": 0.002795409730019158, "output_mean_power": 0.002696845614303835}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3091173503801554e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008259246223406256, "output_power": 0.0007865716506284692, "reference_active": true, "delta_db": -0.21202123779017779}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.583418527096058e-05, "output_power": 8.397557858075973e-05, "reference_active": true, "delta_db": -0.5736745031671271}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329771424834818e-06, "output_power": 2.5893300906992384e-06, "reference_active": true, "delta_db": -1.0922700368182774}], "active_band_count": 3, "active_band_mae_db": 0.6259885925918608}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000132208474335363, "log_rms_mae_db": 1.8043473982060232, "rms_envelope_nrmse": 0.18146628422285413, "envelope_correlation": 0.9750634687119515, "rms_gain_error_db": -0.339734066187709, "reference_mean_power": 0.001270855716483403, "output_mean_power": 0.0011752299216365714}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.478174051796597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005667765898001897, "output_power": 0.0005293354945037998, "reference_active": true, "delta_db": -0.2968088728882185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011084740591226797, "output_power": 9.68115403083399e-05, "reference_active": true, "delta_db": -0.5879840409749353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2190579526022385e-05, "output_power": 2.0872500518301077e-05, "reference_active": true, "delta_db": -0.2659416384306202}], "active_band_count": 3, "active_band_mae_db": 0.38357818409792466}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012422751968092996, "log_rms_mae_db": 1.114070676082758, "rms_envelope_nrmse": 0.127018825213804, "envelope_correlation": 0.9872629459519814, "rms_gain_error_db": -0.33973115257041436, "reference_mean_power": 0.0012772403358267607, "output_mean_power": 0.0011811349214146578}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.710312666254066e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005574547781532276, "output_power": 0.000521228669101575, "reference_active": true, "delta_db": -0.29181347147478254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.253668134122962e-05, "output_power": 4.578920448003661e-05, "reference_active": true, "delta_db": -0.5969953725691391}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7043530583570796e-06, "output_power": 1.430991115841174e-06, "reference_active": true, "delta_db": -0.7592262676884499}], "active_band_count": 3, "active_band_mae_db": 0.5493450372441239}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.199705053096058e-05, "log_rms_mae_db": 1.7347672924617363, "rms_envelope_nrmse": 0.21201907729362446, "envelope_correlation": 0.962438168072245, "rms_gain_error_db": -0.31394106377520375, "reference_mean_power": 0.00020286667816607547, "output_mean_power": 0.000188719427995672}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3687094204678632e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.826983537989376e-05, "output_power": 6.971028164138855e-05, "reference_active": true, "delta_db": -0.5029758303251903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2413787132715612e-05, "output_power": 1.892586161577816e-05, "reference_active": true, "delta_db": -0.7345958280970517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388555256663533e-06, "output_power": 4.895066013061662e-06, "reference_active": true, "delta_db": -1.1564410258348998}], "active_band_count": 3, "active_band_mae_db": 0.7980042280857139}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.313537286332709e-05, "log_rms_mae_db": 1.0033424122274348, "rms_envelope_nrmse": 0.13105048239002662, "envelope_correlation": 0.9845187518469369, "rms_gain_error_db": -0.3140527239033302, "reference_mean_power": 0.00020386380860129322, "output_mean_power": 0.0001896421459631912}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.180981609421061e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.558718162537124e-05, "output_power": 6.730936119861472e-05, "reference_active": true, "delta_db": -0.5037268346902457}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0576597086854303e-05, "output_power": 8.974610099351912e-06, "reference_active": true, "delta_db": -0.7133037051664265}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.38427493679316e-07, "output_power": 3.735977346597368e-07, "reference_active": true, "delta_db": -1.5872299393982894}], "active_band_count": 3, "active_band_mae_db": 0.9347534930849871}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.0553751915640445e-05, "log_rms_mae_db": 1.8185847719857746, "rms_envelope_nrmse": 0.17029411070991113, "envelope_correlation": 0.9801875793071393, "rms_gain_error_db": -0.14933554233704513, "reference_mean_power": 5.25173391881524e-05, "output_mean_power": 5.0742184494634566e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.578963632678691e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2240876569039177e-05, "output_power": 2.1009575033900205e-05, "reference_active": true, "delta_db": -0.2473463198004094}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.96335110256818e-06, "output_power": 7.0851070049281335e-06, "reference_active": true, "delta_db": -0.507494504116949}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4165297947437381e-06, "output_power": 1.0919783689255086e-06, "reference_active": true, "delta_db": -1.1301167825117155}], "active_band_count": 3, "active_band_mae_db": 0.6283192021430246}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.394748258818145e-05, "log_rms_mae_db": 1.0591479377177804, "rms_envelope_nrmse": 0.09894735781422023, "envelope_correlation": 0.9929452526829231, "rms_gain_error_db": -0.14934055000032842, "reference_mean_power": 5.256947927681562e-05, "output_mean_power": 5.07925036134752e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.187575200111859e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.180909445738007e-05, "output_power": 2.0511988639538305e-05, "reference_active": true, "delta_db": -0.2662986616134124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0063771970727005e-06, "output_power": 3.6836246939436056e-06, "reference_active": true, "delta_db": -0.3647645965500685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0500724932970203e-07, "output_power": 8.049424571567722e-08, "reference_active": true, "delta_db": -1.1545444715232027}], "active_band_count": 3, "active_band_mae_db": 0.5952025765622279}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8276746140844288, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "eec7efe0dc2b0acf81137f5a1224d22b4db2003b5851a49bd6e13a20da86123c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9639002438248037, "standard_deviation": 0.0024370699414868733, "minimum": 0.9604328600339707, "maximum": 0.968366984673006, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7265983624268104, "standard_deviation": 0.020867921837715214, "minimum": 1.692206690931341, "maximum": 1.7617946763097416, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7356033133616773, "standard_deviation": 0.1799763304158929, "minimum": 0.4716945363888869, "maximum": 1.0290096425047945, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20811616893824414, "standard_deviation": 0.007052970438562477, "minimum": 0.19495508853119567, "maximum": 0.21796041497131002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9885298336171187, "standard_deviation": 0.001945225597456095, "minimum": 0.9845187518469369, "maximum": 0.9914220516198309, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0101166011548384, "standard_deviation": 0.02487134650380859, "minimum": 0.9825766977519214, "maximum": 1.0595418268595471, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8866821997279026, "standard_deviation": 0.2024790357695695, "minimum": 0.699318246272158, "maximum": 1.232335997728098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11190691385308614, "standard_deviation": 0.00884508812784636, "minimum": 0.09979540947283742, "maximum": 0.13105048239002662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793021794680377, "standard_deviation": 0.0012733130743094677, "minimum": 0.9778011451283605, "maximum": 0.9813229087040801, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8376885834270755, "standard_deviation": 0.03728803526984257, "minimum": 1.7920457610776999, "maximum": 1.880919666241063, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.46820429027953286, "standard_deviation": 0.216669563434445, "minimum": 0.24453138526755533, "maximum": 0.8583071378827647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17454909965207016, "standard_deviation": 0.005572025322122151, "minimum": 0.1662904493543284, "maximum": 0.18123611667193293, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933790602818036, "standard_deviation": 0.000707819814488336, "minimum": 0.9925206491798918, "maximum": 0.9945623780495889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1138462972495806, "standard_deviation": 0.044736537743710274, "minimum": 1.0591479377177804, "maximum": 1.169488113480084, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.47932925837823415, "standard_deviation": 0.25846716265043995, "minimum": 0.18313210968154678, "maximum": 0.9974184423864099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09657875020273972, "standard_deviation": 0.005920945213814858, "minimum": 0.08724309340995565, "maximum": 0.10284350176383722, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9769623760455255, "standard_deviation": 0.004506513279300025, "minimum": 0.9685204662827128, "maximum": 0.9827086329151227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7944446731220292, "standard_deviation": 0.03120995293190336, "minimum": 1.7529877573758348, "maximum": 1.8495011212965538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3361395743679756, "standard_deviation": 0.12031040365467968, "minimum": 0.1656660161253366, "maximum": 0.5671277847457189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1730521161718852, "standard_deviation": 0.016475181913578506, "minimum": 0.15074158918230002, "maximum": 0.2022614543463512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9912690106893626, "standard_deviation": 0.003564343693257061, "minimum": 0.9844119890430263, "maximum": 0.9944191024670765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0704573048561594, "standard_deviation": 0.036870421915272095, "minimum": 1.0179810755821042, "maximum": 1.114070676082758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4331723955185771, "standard_deviation": 0.1946898785481822, "minimum": 0.14287207632532023, "maximum": 0.7514223546251171, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10265875949618644, "standard_deviation": 0.019377474031508586, "minimum": 0.08439552709555875, "maximum": 0.13707286819092662, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1513941803704214, "standard_deviation": 0.030768642856151045, "minimum": 0.11732248153774837, "maximum": 0.20671053389492647, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.556092476155268, "standard_deviation": 0.006811249854812156, "minimum": 5.550620209993568, "maximum": 5.570545241567607, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4097278516680127, "standard_deviation": 0.06317328185089634, "minimum": 0.37611740128699367, "maximum": 0.5647706478043107, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09186698128768893, "standard_deviation": 0.07208500150205405, "minimum": -0.1559174034728275, "maximum": 0.06306374785030679, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9868286237310463, "standard_deviation": 0.001854714696490956, "minimum": 0.984004458362738, "maximum": 0.9885520346657032, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2663679816231674, "standard_deviation": 0.033291822266584614, "minimum": 1.2133609809879449, "maximum": 1.3187935592686058, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.373182442360687, "standard_deviation": 0.20082590671742376, "minimum": 0.13036589661289613, "maximum": 0.6326253443484128, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11680906376129277, "standard_deviation": 0.008196631744114285, "minimum": 0.10890944574909914, "maximum": 0.12884384645411637, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9958950269558831, "standard_deviation": 0.0009600177973869542, "minimum": 0.9943157959756804, "maximum": 0.9969891877997286, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6559467194116385, "standard_deviation": 0.015246615420116598, "minimum": 0.6360387459832552, "maximum": 0.6784286823065203, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4283956541219175, "standard_deviation": 0.21572705131744405, "minimum": 0.128565416035369, "maximum": 0.6965606701358652, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06239583953996614, "standard_deviation": 0.007089564149942114, "minimum": 0.05324007762518448, "maximum": 0.07324399713055035, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123288-0003", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.5726873874664307, "predicted_mos": 2.5726873874664307, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f6478415a93c8074efa6ea3074561b6c3d059c24a5bab9110bf6bbb3c3ee5724", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.598890599794686, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2627115612100588, "全帯域_10ms_rms_envelope_nrmse": 0.12006464300075749, "全帯域_10ms_envelope_correlation": 0.9860853791754497, "全帯域_10ms_modulation_mae_db": 0.22260964948843973, "全帯域_40ms_log_rms_mae_db": 0.6231858673293276, "全帯域_40ms_rms_envelope_nrmse": 0.06521256324725029, "全帯域_40ms_envelope_correlation": 0.9954838834892263, "全帯域_40ms_modulation_mae_db": 0.33094349464445433, "300–1000 Hz_10ms_log_rms_mae_db": 1.9339362704582106, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18099737649567238, "300–1000 Hz_10ms_envelope_correlation": 0.9748876661138243, "300–1000 Hz_10ms_modulation_mae_db": 0.28557411053042764, "300–1000 Hz_40ms_log_rms_mae_db": 1.104299810973081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09571660446304633, "300–1000 Hz_40ms_envelope_correlation": 0.9927242775094457, "300–1000 Hz_40ms_modulation_mae_db": 0.2867649601552795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8931300737701346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24428345239587568, "1000–3000 Hz_10ms_envelope_correlation": 0.9503227591866946, "1000–3000 Hz_10ms_modulation_mae_db": 0.2931821813266217, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1149488007880435, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13889924267925283, "1000–3000 Hz_40ms_envelope_correlation": 0.9818949670343439, "1000–3000 Hz_40ms_modulation_mae_db": 0.6112858965896029, "3000–6000 Hz_10ms_log_rms_mae_db": 2.129390280453204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17382663283669356, "3000–6000 Hz_10ms_envelope_correlation": 0.980193325392092, "3000–6000 Hz_10ms_modulation_mae_db": 0.18605325859154612, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2426403940045438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09258766579410113, "3000–6000 Hz_40ms_envelope_correlation": 0.9941495255064062, "3000–6000 Hz_40ms_modulation_mae_db": 0.4946654098276229, "mr_stft_db": 5.676720155988272, "band_mae_db": 0.12177761288615775, "rms_gain_error_db": -0.08607294105295392, "output_peak": 0.3750992250337336}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2406857495910526, "全帯域_10ms_rms_envelope_nrmse": 0.11055448292839053, "全帯域_10ms_envelope_correlation": 0.9881850432233589, "全帯域_10ms_modulation_mae_db": 0.5349457760790444, "全帯域_40ms_log_rms_mae_db": 0.6303323096067995, "全帯域_40ms_rms_envelope_nrmse": 0.05367429436230922, "全帯域_40ms_envelope_correlation": 0.9969276084587434, "全帯域_40ms_modulation_mae_db": 0.5521358373519597, "300–1000 Hz_10ms_log_rms_mae_db": 1.9737072984738786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16362438327973972, "300–1000 Hz_10ms_envelope_correlation": 0.9796040025919238, "300–1000 Hz_10ms_modulation_mae_db": 0.44928930938259876, "300–1000 Hz_40ms_log_rms_mae_db": 1.1663153770834904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08808822779285137, "300–1000 Hz_40ms_envelope_correlation": 0.9939489753721152, "300–1000 Hz_40ms_modulation_mae_db": 0.28452953354245486, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9296262224433924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2230042976868913, "1000–3000 Hz_10ms_envelope_correlation": 0.9587941864512395, "1000–3000 Hz_10ms_modulation_mae_db": 0.5276063247080526, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1541799035687428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12033866915117568, "1000–3000 Hz_40ms_envelope_correlation": 0.9864444353091479, "1000–3000 Hz_40ms_modulation_mae_db": 0.6714736557145929, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1628758563092254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17471816876493906, "3000–6000 Hz_10ms_envelope_correlation": 0.9794095783875212, "3000–6000 Hz_10ms_modulation_mae_db": 0.8308541771638148, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2760466225715474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09996076144341806, "3000–6000 Hz_40ms_envelope_correlation": 0.9932472490413298, "3000–6000 Hz_40ms_modulation_mae_db": 0.7586194131607527, "mr_stft_db": 5.705527923037875, "band_mae_db": 0.09811874067476763, "rms_gain_error_db": -0.06720850646307025, "output_peak": 0.37347274485589316}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.253331447249811, "全帯域_10ms_rms_envelope_nrmse": 0.13696966017170095, "全帯域_10ms_envelope_correlation": 0.9822312571275089, "全帯域_10ms_modulation_mae_db": 0.2672698870158894, "全帯域_40ms_log_rms_mae_db": 0.6690051975666967, "全帯域_40ms_rms_envelope_nrmse": 0.07932956358859418, "全帯域_40ms_envelope_correlation": 0.9934344510748954, "全帯域_40ms_modulation_mae_db": 0.45619928526733083, "300–1000 Hz_10ms_log_rms_mae_db": 1.8340682774234045, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18971772078073348, "300–1000 Hz_10ms_envelope_correlation": 0.9730448100737265, "300–1000 Hz_10ms_modulation_mae_db": 0.39912292441236846, "300–1000 Hz_40ms_log_rms_mae_db": 1.0890102200091407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11949387290800129, "300–1000 Hz_40ms_envelope_correlation": 0.9883914176406323, "300–1000 Hz_40ms_modulation_mae_db": 0.30522506676191924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8418555511296948, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2234527158210522, "1000–3000 Hz_10ms_envelope_correlation": 0.9583791204668987, "1000–3000 Hz_10ms_modulation_mae_db": 0.6996302109010575, "1000–3000 Hz_40ms_log_rms_mae_db": 1.070488222561944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12260522542426923, "1000–3000 Hz_40ms_envelope_correlation": 0.9859142425928269, "1000–3000 Hz_40ms_modulation_mae_db": 0.7823645129064237, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1322719592211063, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16822322433554474, "3000–6000 Hz_10ms_envelope_correlation": 0.9808191671782194, "3000–6000 Hz_10ms_modulation_mae_db": 0.429760582038889, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2387306314227822, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09297293296729991, "3000–6000 Hz_40ms_envelope_correlation": 0.9937795183266586, "3000–6000 Hz_40ms_modulation_mae_db": 0.6154687884956876, "mr_stft_db": 5.694985647303501, "band_mae_db": 0.18222048327556206, "rms_gain_error_db": 0.03703677906886293, "output_peak": 0.5219312861246871}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2452714897789279, "全帯域_10ms_rms_envelope_nrmse": 0.11040096617737723, "全帯域_10ms_envelope_correlation": 0.988253562739048, "全帯域_10ms_modulation_mae_db": 0.13486267081406964, "全帯域_40ms_log_rms_mae_db": 0.6292127034290418, "全帯域_40ms_rms_envelope_nrmse": 0.05599015567470552, "全帯域_40ms_envelope_correlation": 0.9966355751034565, "全帯域_40ms_modulation_mae_db": 0.3252991842880822, "300–1000 Hz_10ms_log_rms_mae_db": 1.9652359215587039, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15869480318228663, "300–1000 Hz_10ms_envelope_correlation": 0.9806996018937102, "300–1000 Hz_10ms_modulation_mae_db": 0.360462437900021, "300–1000 Hz_40ms_log_rms_mae_db": 1.1279445051811596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08675673309222161, "300–1000 Hz_40ms_envelope_correlation": 0.9937456224713387, "300–1000 Hz_40ms_modulation_mae_db": 0.3493573698223897, "1000–3000 Hz_10ms_log_rms_mae_db": 1.918513094594126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22283405693015623, "1000–3000 Hz_10ms_envelope_correlation": 0.9586048311428077, "1000–3000 Hz_10ms_modulation_mae_db": 1.1761907905715179, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1189618915725377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11825799193551872, "1000–3000 Hz_40ms_envelope_correlation": 0.9882644874117058, "1000–3000 Hz_40ms_modulation_mae_db": 1.3599622612138014, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1363841665806764, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1747567476397266, "3000–6000 Hz_10ms_envelope_correlation": 0.9792681789572593, "3000–6000 Hz_10ms_modulation_mae_db": 0.5651126525277576, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2917659380936188, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10602083604044575, "3000–6000 Hz_40ms_envelope_correlation": 0.9919583279841234, "3000–6000 Hz_40ms_modulation_mae_db": 0.7750414483014979, "mr_stft_db": 5.6832539547442815, "band_mae_db": 0.14369264707233206, "rms_gain_error_db": -0.027900556266903468, "output_peak": 0.3837814637240846}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.253382116868141, "全帯域_10ms_rms_envelope_nrmse": 0.12058154967009337, "全帯域_10ms_envelope_correlation": 0.9860791806938612, "全帯域_10ms_modulation_mae_db": 0.43432391260533293, "全帯域_40ms_log_rms_mae_db": 0.6701747277847891, "全帯域_40ms_rms_envelope_nrmse": 0.06543285637219162, "全帯域_40ms_envelope_correlation": 0.9954200663329152, "全帯域_40ms_modulation_mae_db": 0.42762881348080223, "300–1000 Hz_10ms_log_rms_mae_db": 1.8040183736937971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16915746833075998, "300–1000 Hz_10ms_envelope_correlation": 0.9784075980962321, "300–1000 Hz_10ms_modulation_mae_db": 0.22374669412061587, "300–1000 Hz_40ms_log_rms_mae_db": 1.0790675280021989, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09803989758071431, "300–1000 Hz_40ms_envelope_correlation": 0.9920619045013588, "300–1000 Hz_40ms_modulation_mae_db": 0.2941668437573993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9473903430214574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21867236589537717, "1000–3000 Hz_10ms_envelope_correlation": 0.9598446061913486, "1000–3000 Hz_10ms_modulation_mae_db": 1.2812585347911298, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1769002662071602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11646371610912633, "1000–3000 Hz_40ms_envelope_correlation": 0.9874898635410657, "1000–3000 Hz_40ms_modulation_mae_db": 1.6215651403557, "3000–6000 Hz_10ms_log_rms_mae_db": 2.161465445044294, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18322207385897465, "3000–6000 Hz_10ms_envelope_correlation": 0.9771709233267075, "3000–6000 Hz_10ms_modulation_mae_db": 0.803905945770004, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3063100187080403, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10185361236473627, "3000–6000 Hz_40ms_envelope_correlation": 0.9924592146683718, "3000–6000 Hz_40ms_modulation_mae_db": 1.044770525093017, "mr_stft_db": 5.693642371729903, "band_mae_db": 0.10635786174004602, "rms_gain_error_db": 0.042603729578850336, "output_peak": 0.4210949614582579}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2463803389674601, "全帯域_10ms_rms_envelope_nrmse": 0.10666210360623912, "全帯域_10ms_envelope_correlation": 0.989013550633831, "全帯域_10ms_modulation_mae_db": 0.3169093126980587, "全帯域_40ms_log_rms_mae_db": 0.6637893785031017, "全帯域_40ms_rms_envelope_nrmse": 0.055837800678007324, "全帯域_40ms_envelope_correlation": 0.9966581672176439, "全帯域_40ms_modulation_mae_db": 0.4313988991406199, "300–1000 Hz_10ms_log_rms_mae_db": 1.875546316085209, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17134747521102667, "300–1000 Hz_10ms_envelope_correlation": 0.9774515948608992, "300–1000 Hz_10ms_modulation_mae_db": 0.21035240659432622, "300–1000 Hz_40ms_log_rms_mae_db": 1.0283778223209297, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08477336651465543, "300–1000 Hz_40ms_envelope_correlation": 0.994159289538039, "300–1000 Hz_40ms_modulation_mae_db": 0.2532419822998737, "1000–3000 Hz_10ms_log_rms_mae_db": 1.862542531584615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19999181192730575, "1000–3000 Hz_10ms_envelope_correlation": 0.9667133493725457, "1000–3000 Hz_10ms_modulation_mae_db": 0.8443425323197786, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0865516729610114, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10099290039884048, "1000–3000 Hz_40ms_envelope_correlation": 0.9905104578450901, "1000–3000 Hz_40ms_modulation_mae_db": 1.1542971929402392, "3000–6000 Hz_10ms_log_rms_mae_db": 2.136293988637878, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18550140486792474, "3000–6000 Hz_10ms_envelope_correlation": 0.9777056070463119, "3000–6000 Hz_10ms_modulation_mae_db": 0.12270496618950139, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3014772484776964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11131772527387448, "3000–6000 Hz_40ms_envelope_correlation": 0.9916546003761377, "3000–6000 Hz_40ms_modulation_mae_db": 0.1385984066375631, "mr_stft_db": 5.6962588145787985, "band_mae_db": 0.09412810456632181, "rms_gain_error_db": -0.039606128925725605, "output_peak": 0.4032060237425199}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2736815957445606, "全帯域_10ms_rms_envelope_nrmse": 0.13669163470242018, "全帯域_10ms_envelope_correlation": 0.9825614213847573, "全帯域_10ms_modulation_mae_db": 0.05467913946067276, "全帯域_40ms_log_rms_mae_db": 0.6377136436415711, "全帯域_40ms_rms_envelope_nrmse": 0.08650038465238205, "全帯域_40ms_envelope_correlation": 0.9923307249129673, "全帯域_40ms_modulation_mae_db": 0.2449268758810048, "300–1000 Hz_10ms_log_rms_mae_db": 1.8391891515940462, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17868705931337592, "300–1000 Hz_10ms_envelope_correlation": 0.9767626683539724, "300–1000 Hz_10ms_modulation_mae_db": 0.3439246914661853, "300–1000 Hz_40ms_log_rms_mae_db": 1.0378095421613625, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11901128138337265, "300–1000 Hz_40ms_envelope_correlation": 0.9889549378937065, "300–1000 Hz_40ms_modulation_mae_db": 0.21155296623774303, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8687474163128932, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22373132245649868, "1000–3000 Hz_10ms_envelope_correlation": 0.9581474039757844, "1000–3000 Hz_10ms_modulation_mae_db": 0.9711505035402431, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1069414982997559, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11864486208511284, "1000–3000 Hz_40ms_envelope_correlation": 0.9867947748682957, "1000–3000 Hz_40ms_modulation_mae_db": 1.1632292966988016, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1329674207569402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1784937232768038, "3000–6000 Hz_10ms_envelope_correlation": 0.978426550525731, "3000–6000 Hz_10ms_modulation_mae_db": 0.4770162945966268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2428920290653265, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10047981022245461, "3000–6000 Hz_40ms_envelope_correlation": 0.9930248112477432, "3000–6000 Hz_40ms_modulation_mae_db": 0.40831480116984104, "mr_stft_db": 5.700521750280025, "band_mae_db": 0.15869587441791252, "rms_gain_error_db": 0.09842544983631768, "output_peak": 0.5169557488905561}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2393364708588825, "全帯域_10ms_rms_envelope_nrmse": 0.11544611342603092, "全帯域_10ms_envelope_correlation": 0.9871675733459868, "全帯域_10ms_modulation_mae_db": 0.4558249903039899, "全帯域_40ms_log_rms_mae_db": 0.6140050627447559, "全帯域_40ms_rms_envelope_nrmse": 0.05948671089948907, "全帯域_40ms_envelope_correlation": 0.9963536386206397, "全帯域_40ms_modulation_mae_db": 0.5692440620390632, "300–1000 Hz_10ms_log_rms_mae_db": 1.87514020287163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17464028378829108, "300–1000 Hz_10ms_envelope_correlation": 0.9769841414423445, "300–1000 Hz_10ms_modulation_mae_db": 0.5637343073237703, "300–1000 Hz_40ms_log_rms_mae_db": 1.0544813609631698, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09677451139050593, "300–1000 Hz_40ms_envelope_correlation": 0.993161269226449, "300–1000 Hz_40ms_modulation_mae_db": 0.49309907537125436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8791021211217096, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2296177151053475, "1000–3000 Hz_10ms_envelope_correlation": 0.9566947842832687, "1000–3000 Hz_10ms_modulation_mae_db": 0.7807460020029596, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1321667215092572, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13880467997364046, "1000–3000 Hz_40ms_envelope_correlation": 0.9823257407212819, "1000–3000 Hz_40ms_modulation_mae_db": 0.8786669217077097, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1475505064545533, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18551202421587215, "3000–6000 Hz_10ms_envelope_correlation": 0.9771855372125166, "3000–6000 Hz_10ms_modulation_mae_db": 0.10812986561477433, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3163303008681662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11220727511850642, "3000–6000 Hz_40ms_envelope_correlation": 0.9911609269154265, "3000–6000 Hz_40ms_modulation_mae_db": 0.3879876510524052, "mr_stft_db": 5.682606264234682, "band_mae_db": 0.1692396558938212, "rms_gain_error_db": -0.12738075964685536, "output_peak": 0.38490438250026265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.057136432499027, "spectral_convergence": 0.35607512099031824, "spectral_convergence_status": "ok", "amplitude_floor": 1.3116896377207775e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.468742205098093, "spectral_convergence": 0.3511607170516739, "spectral_convergence_status": "ok", "amplitude_floor": 8.457684224428572e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.504281830367695, "spectral_convergence": 0.3667022356359908, "spectral_convergence_status": "ok", "amplitude_floor": 6.717993311155654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008437417957183396, "output_power": 0.0008417355524001815, "delta_db": -0.010338920483299187, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011741200681074208, "output_power": 0.0011250181703928311, "delta_db": -0.18552974213222867, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004760668283187788, "output_power": 0.0004709042398680211, "delta_db": -0.04735320781067068, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001833223674641814, "output_power": 0.00018263090053028834, "delta_db": -0.016411966086892084, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.526890589901711e-05, "output_power": 6.415574258994999e-05, "delta_db": -0.07470796476051958, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6955820206525705e-05, "output_power": 4.0486966019071923e-05, "delta_db": 0.3963238760433363, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12177761288615775, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08607294105295392, "pitch": null, "output_rms_dbfs": -25.64644399208495, "output_peak": 0.3750992250337336, "reference_rms_dbfs": -25.560371051032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.676720155988272}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001544797211376954, "log_rms_mae_db": 1.2627115612100588, "rms_envelope_nrmse": 0.12006464300075749, "envelope_correlation": 0.9860853791754497, "rms_gain_error_db": -0.08607101741040254, "reference_mean_power": 0.002781784143506645, "output_mean_power": 0.0027271958385316113}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.487428460245723e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008814675226478596, "output_power": 0.0008479712651763872, "reference_active": true, "delta_db": -0.1682517974409038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001997406840389717, "output_power": 0.00018990720161983897, "reference_active": true, "delta_db": -0.21925098572276994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0345220429658976e-05, "output_power": 3.7823305110652455e-05, "reference_active": true, "delta_db": -0.28032616530164545}], "active_band_count": 3, "active_band_mae_db": 0.22260964948843973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013989081102856241, "log_rms_mae_db": 0.6231858673293276, "rms_envelope_nrmse": 0.06521256324725029, "envelope_correlation": 0.9954838834892263, "rms_gain_error_db": -0.08620091515818903, "reference_mean_power": 0.002795409730019158, "output_mean_power": 0.0027404720746094747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3091173503801554e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008259246223406256, "output_power": 0.0007964500035684011, "reference_active": true, "delta_db": -0.15781894879887112}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.583418527096058e-05, "output_power": 8.906159450020314e-05, "reference_active": true, "delta_db": -0.3182998888540328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329771424834818e-06, "output_power": 2.9562642898889966e-06, "reference_active": true, "delta_db": -0.516711646280459}], "active_band_count": 3, "active_band_mae_db": 0.33094349464445433}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000132208474335363, "log_rms_mae_db": 1.9339362704582106, "rms_envelope_nrmse": 0.18099737649567238, "envelope_correlation": 0.9748876661138243, "rms_gain_error_db": -0.22800232878046062, "reference_mean_power": 0.001270855716483403, "output_mean_power": 0.0012058575694808044}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.478174051796597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005667765898001897, "output_power": 0.0005330378131324806, "reference_active": true, "delta_db": -0.2665388529543338}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011084740591226797, "output_power": 9.994953552393513e-05, "reference_active": true, "delta_db": -0.44944753819598054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2190579526022385e-05, "output_power": 2.2921457968035083e-05, "reference_active": true, "delta_db": 0.14073594044096846}], "active_band_count": 3, "active_band_mae_db": 0.28557411053042764}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012422751968092996, "log_rms_mae_db": 1.104299810973081, "rms_envelope_nrmse": 0.09571660446304633, "envelope_correlation": 0.9927242775094457, "rms_gain_error_db": -0.22800347080908268, "reference_mean_power": 0.0012772403358267607, "output_mean_power": 0.001211915327617919}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.710312666254066e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005574547781532276, "output_power": 0.0005243067695344114, "reference_active": true, "delta_db": -0.2662417721227015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.253668134122962e-05, "output_power": 4.811863759244739e-05, "reference_active": true, "delta_db": -0.38149313182917133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7043530583570796e-06, "output_power": 1.6229441193771004e-06, "reference_active": true, "delta_db": -0.21255997651396574}], "active_band_count": 3, "active_band_mae_db": 0.2867649601552795}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.199705053096058e-05, "log_rms_mae_db": 1.8931300737701346, "rms_envelope_nrmse": 0.24428345239587568, "envelope_correlation": 0.9503227591866946, "rms_gain_error_db": -0.07358299345723827, "reference_mean_power": 0.00020286667816607547, "output_mean_power": 0.00019945844025330269}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3687094204678632e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.826983537989376e-05, "output_power": 7.195614430599172e-05, "reference_active": true, "delta_db": -0.3652653631034015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2413787132715612e-05, "output_power": 2.1657866919215667e-05, "reference_active": true, "delta_db": -0.14899562270953978}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388555256663533e-06, "output_power": 5.873190662844762e-06, "reference_active": true, "delta_db": -0.3652855581669237}], "active_band_count": 3, "active_band_mae_db": 0.2931821813266217}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.313537286332709e-05, "log_rms_mae_db": 1.1149488007880435, "rms_envelope_nrmse": 0.13889924267925283, "envelope_correlation": 0.9818949670343439, "rms_gain_error_db": -0.07385351689094213, "reference_mean_power": 0.00020386380860129322, "output_mean_power": 0.0002004263335053912}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.180981609421061e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.558718162537124e-05, "output_power": 7.117550803438508e-05, "reference_active": true, "delta_db": -0.26117576637282947}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0576597086854303e-05, "output_power": 1.0272490820045735e-05, "reference_active": true, "delta_db": -0.12670198486154705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.38427493679316e-07, "output_power": 3.859483906047042e-07, "reference_active": true, "delta_db": -1.445979938534432}], "active_band_count": 3, "active_band_mae_db": 0.6112858965896029}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.0553751915640445e-05, "log_rms_mae_db": 2.129390280453204, "rms_envelope_nrmse": 0.17382663283669356, "envelope_correlation": 0.980193325392092, "rms_gain_error_db": 0.17562158531175887, "reference_mean_power": 5.25173391881524e-05, "output_mean_power": 5.468457898007664e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.578963632678691e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2240876569039177e-05, "output_power": 2.2278665344962843e-05, "reference_active": true, "delta_db": 0.007372699761658687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.96335110256818e-06, "output_power": 8.401131653100254e-06, "reference_active": true, "delta_db": 0.2324192643156896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4165297947437381e-06, "output_power": 1.3164032097928966e-06, "reference_active": true, "delta_db": -0.3183678116972901}], "active_band_count": 3, "active_band_mae_db": 0.18605325859154612}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.394748258818145e-05, "log_rms_mae_db": 1.2426403940045438, "rms_envelope_nrmse": 0.09258766579410113, "envelope_correlation": 0.9941495255064062, "rms_gain_error_db": 0.17084206922245915, "reference_mean_power": 5.256947927681562e-05, "output_mean_power": 5.467866242225801e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.187575200111859e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.180909445738007e-05, "output_power": 2.216385609549308e-05, "reference_active": true, "delta_db": 0.07007688260652797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0063771970727005e-06, "output_power": 4.265705045728837e-06, "reference_active": true, "delta_db": 0.2723898730038649}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0500724932970203e-07, "output_power": 8.073583381122573e-08, "reference_active": true, "delta_db": -1.141529473872476}], "active_band_count": 3, "active_band_mae_db": 0.4946654098276229}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.820405893601631, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "66d066d044999c71f521be3397803793e74b8d3ef8f8036c2b73d9df2d18d906", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9584376301338235, "standard_deviation": 0.004465718561293783, "minimum": 0.9503227591866946, "maximum": 0.9667133493725457, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8926134192472528, "standard_deviation": 0.03639913622862342, "minimum": 1.8418555511296948, "maximum": 1.9473903430214574, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8217633850201701, "standard_deviation": 0.32559973466577935, "minimum": 0.2931821813266217, "maximum": 1.2812585347911298, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22319846727731307, "standard_deviation": 0.012220222467441035, "minimum": 0.19999181192730575, "maximum": 0.24428345239587568, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9862048711654696, "standard_deviation": 0.0028919610274524003, "minimum": 0.9818949670343439, "maximum": 0.9905104578450901, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1201423721835566, "standard_deviation": 0.0344779359922771, "minimum": 1.070488222561944, "maximum": 1.1769002662071602, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0303556097658588, "standard_deviation": 0.35458444694550434, "minimum": 0.6112858965896029, "maximum": 1.6215651403557, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12187591096961707, "standard_deviation": 0.012353097085395452, "minimum": 0.10099290039884048, "maximum": 0.13889924267925283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9787723585032948, "standard_deviation": 0.0013732442305008806, "minimum": 0.9771709233267075, "maximum": 0.9808191671782194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1423999529322346, "standard_deviation": 0.01332857721170646, "minimum": 2.129390280453204, "maximum": 2.1628758563092254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4404422178116143, "standard_deviation": 0.287258982288617, "minimum": 0.10812986561477433, "maximum": 0.8308541771638148, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1780317499745599, "standard_deviation": 0.006262117104218625, "minimum": 0.16822322433554474, "maximum": 0.18551202421587215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926792717582746, "standard_deviation": 0.0010513726999541678, "minimum": 0.9911609269154265, "maximum": 0.9941495255064062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2770241479014652, "standard_deviation": 0.031696109324476154, "minimum": 1.2387306314227822, "maximum": 1.3163303008681662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5779333054672985, "standard_deviation": 0.28174037592999035, "minimum": 0.1385984066375631, "maximum": 1.044770525093017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10217507740310458, "standard_deviation": 0.007401749309824005, "minimum": 0.09258766579410113, "maximum": 0.11220727511850642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9772302604283292, "standard_deviation": 0.002460866707785273, "minimum": 0.9730448100737265, "maximum": 0.9806996018937102, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.88760522651986, "standard_deviation": 0.06334095331725709, "minimum": 1.8040183736937971, "maximum": 1.9737072984738786, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3545258602162892, "standard_deviation": 0.11803651390631965, "minimum": 0.21035240659432622, "maximum": 0.5637343073237703, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17335832129773573, "standard_deviation": 0.00991018936623762, "minimum": 0.15869480318228663, "maximum": 0.18971772078073348, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9921434617691357, "standard_deviation": 0.002252452950477701, "minimum": 0.9883914176406323, "maximum": 0.994159289538039, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0859132708368167, "standard_deviation": 0.04665665624452389, "minimum": 1.0283778223209297, "maximum": 1.1663153770834904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.30974222474353924, "standard_deviation": 0.08404856926503705, "minimum": 0.21155296623774303, "maximum": 0.49309907537125436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09858181189067111, "standard_deviation": 0.013664698517798526, "minimum": 0.08477336651465543, "maximum": 0.11949387290800129, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13427887256586513, "standard_deviation": 0.033934742388244475, "minimum": 0.09412810456632181, "maximum": 0.18222048327556206, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.691689610237168, "standard_deviation": 0.009876468308328442, "minimum": 5.676720155988272, "maximum": 5.705527923037875, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42255572954124937, "standard_deviation": 0.061812778020941046, "minimum": 0.37347274485589316, "maximum": 0.5219312861246871, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.021262866733934706, "standard_deviation": 0.07536195788947037, "minimum": -0.12738075964685536, "maximum": 0.09842544983631768, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9861971210404752, "standard_deviation": 0.002564249406438658, "minimum": 0.9822312571275089, "maximum": 0.989013550633831, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.251847596283612, "standard_deviation": 0.011671436401999783, "minimum": 1.2393364708588825, "maximum": 1.2736815957445606, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3026781673081872, "standard_deviation": 0.16573024198884886, "minimum": 0.05467913946067276, "maximum": 0.5349457760790444, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11967139421037623, "standard_deviation": 0.011625434960930568, "minimum": 0.10666210360623912, "maximum": 0.13696966017170095, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.995405514401311, "standard_deviation": 0.0016751832637996878, "minimum": 0.9923307249129673, "maximum": 0.9969276084587434, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6421773613257604, "standard_deviation": 0.02221110482942873, "minimum": 0.6140050627447559, "maximum": 0.6701747277847891, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.41722205651166466, "standard_deviation": 0.11258035872114178, "minimum": 0.2449268758810048, "maximum": 0.5692440620390632, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06518304118436616, "standard_deviation": 0.011900531825396847, "minimum": 0.05367429436230922, "maximum": 0.08650038465238205, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123288-0003", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6319801807403564, "predicted_mos": 2.6319801807403564, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6e89210b8113c8123a7b48a90696111de56b0550d35be722cba7022fd93eda89", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.475436900276691, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2643299667685435, "全帯域_10ms_rms_envelope_nrmse": 0.12044866664449591, "全帯域_10ms_envelope_correlation": 0.9859664759626184, "全帯域_10ms_modulation_mae_db": 0.28342491174505063, "全帯域_40ms_log_rms_mae_db": 0.6576562668999375, "全帯域_40ms_rms_envelope_nrmse": 0.06715431066589814, "全帯域_40ms_envelope_correlation": 0.9951700900272702, "全帯域_40ms_modulation_mae_db": 0.437435068869535, "300–1000 Hz_10ms_log_rms_mae_db": 1.8282512098333188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1659287477896968, "300–1000 Hz_10ms_envelope_correlation": 0.9788921548511028, "300–1000 Hz_10ms_modulation_mae_db": 0.2461904319897712, "300–1000 Hz_40ms_log_rms_mae_db": 1.0771674120416244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09198320407861854, "300–1000 Hz_40ms_envelope_correlation": 0.9930720804317296, "300–1000 Hz_40ms_modulation_mae_db": 0.45383694009188424, "1000–3000 Hz_10ms_log_rms_mae_db": 1.827959124539441, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2067810809732303, "1000–3000 Hz_10ms_envelope_correlation": 0.9642136774535398, "1000–3000 Hz_10ms_modulation_mae_db": 0.45280375753878266, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0529793090946158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11284139421309884, "1000–3000 Hz_40ms_envelope_correlation": 0.9883703559912862, "1000–3000 Hz_40ms_modulation_mae_db": 0.5285037927907426, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0732719961345647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18394427001155758, "3000–6000 Hz_10ms_envelope_correlation": 0.978022263597412, "3000–6000 Hz_10ms_modulation_mae_db": 0.30131738653205037, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2452755962195583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10997292353858477, "3000–6000 Hz_40ms_envelope_correlation": 0.9918319735681463, "3000–6000 Hz_40ms_modulation_mae_db": 0.4263530712237062, "mr_stft_db": 5.671327424878755, "band_mae_db": 0.14939029721203806, "rms_gain_error_db": -0.07853762613335147, "output_peak": 0.37970572121809537}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2406799862518891, "全帯域_10ms_rms_envelope_nrmse": 0.11947093240540613, "全帯域_10ms_envelope_correlation": 0.9861861327846524, "全帯域_10ms_modulation_mae_db": 0.2755881300128901, "全帯域_40ms_log_rms_mae_db": 0.6600230176054742, "全帯域_40ms_rms_envelope_nrmse": 0.06287818161618226, "全帯域_40ms_envelope_correlation": 0.995759476909167, "全帯域_40ms_modulation_mae_db": 0.1872563197017566, "300–1000 Hz_10ms_log_rms_mae_db": 1.852216806453517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17408353037624583, "300–1000 Hz_10ms_envelope_correlation": 0.9771476162560992, "300–1000 Hz_10ms_modulation_mae_db": 0.47836644177073345, "300–1000 Hz_40ms_log_rms_mae_db": 1.0881940803293486, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10467585802069859, "300–1000 Hz_40ms_envelope_correlation": 0.991783715183479, "300–1000 Hz_40ms_modulation_mae_db": 0.4903311752720472, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8709035300858208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21114524353803685, "1000–3000 Hz_10ms_envelope_correlation": 0.9636510188524664, "1000–3000 Hz_10ms_modulation_mae_db": 0.42143654804753133, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0822935645391178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11292433951261557, "1000–3000 Hz_40ms_envelope_correlation": 0.9884290614104456, "1000–3000 Hz_40ms_modulation_mae_db": 0.6607883666151574, "3000–6000 Hz_10ms_log_rms_mae_db": 2.087596175113754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16663311246133014, "3000–6000 Hz_10ms_envelope_correlation": 0.9811325344641124, "3000–6000 Hz_10ms_modulation_mae_db": 0.830086283360496, "3000–6000 Hz_40ms_log_rms_mae_db": 1.191503521817022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08429064144636703, "3000–6000 Hz_40ms_envelope_correlation": 0.9950449296594641, "3000–6000 Hz_40ms_modulation_mae_db": 0.894655399900727, "mr_stft_db": 5.671122819749098, "band_mae_db": 0.11202524619119615, "rms_gain_error_db": -0.06838468056662397, "output_peak": 0.37368764760535456}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2969527506533043, "全帯域_10ms_rms_envelope_nrmse": 0.12361709326215756, "全帯域_10ms_envelope_correlation": 0.9852117266211807, "全帯域_10ms_modulation_mae_db": 0.1759035953443304, "全帯域_40ms_log_rms_mae_db": 0.6636007559114229, "全帯域_40ms_rms_envelope_nrmse": 0.06685072904210697, "全帯域_40ms_envelope_correlation": 0.9952559957548944, "全帯域_40ms_modulation_mae_db": 0.3241421930639475, "300–1000 Hz_10ms_log_rms_mae_db": 1.8225888857736712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17558127259532147, "300–1000 Hz_10ms_envelope_correlation": 0.9763183553912217, "300–1000 Hz_10ms_modulation_mae_db": 0.28496192265524684, "300–1000 Hz_40ms_log_rms_mae_db": 1.0084881816461742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10014060485726883, "300–1000 Hz_40ms_envelope_correlation": 0.9917839018674345, "300–1000 Hz_40ms_modulation_mae_db": 0.4537246326580546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8687805520579377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20960926272753191, "1000–3000 Hz_10ms_envelope_correlation": 0.9635883522207412, "1000–3000 Hz_10ms_modulation_mae_db": 0.9783871620368808, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1121253186718707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12069384091298778, "1000–3000 Hz_40ms_envelope_correlation": 0.987648513244342, "1000–3000 Hz_40ms_modulation_mae_db": 1.095132340579691, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0430605185519135, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17005913944458864, "3000–6000 Hz_10ms_envelope_correlation": 0.9803260163884153, "3000–6000 Hz_10ms_modulation_mae_db": 0.6297281745854243, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1869028897353615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09198159569924506, "3000–6000 Hz_40ms_envelope_correlation": 0.9939606984132341, "3000–6000 Hz_40ms_modulation_mae_db": 0.7160764980780714, "mr_stft_db": 5.66059138881402, "band_mae_db": 0.13969990653977737, "rms_gain_error_db": -0.09318812406599503, "output_peak": 0.39382508008978223}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.28314153156666, "全帯域_10ms_rms_envelope_nrmse": 0.1189446166840182, "全帯域_10ms_envelope_correlation": 0.9863328762546428, "全帯域_10ms_modulation_mae_db": 0.4329022371090181, "全帯域_40ms_log_rms_mae_db": 0.6585795162355211, "全帯域_40ms_rms_envelope_nrmse": 0.0648722919354892, "全帯域_40ms_envelope_correlation": 0.9954626195466852, "全帯域_40ms_modulation_mae_db": 0.7400972611478854, "300–1000 Hz_10ms_log_rms_mae_db": 1.8479081119025502, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16253353765083645, "300–1000 Hz_10ms_envelope_correlation": 0.9797663054703013, "300–1000 Hz_10ms_modulation_mae_db": 0.8986007083304558, "300–1000 Hz_40ms_log_rms_mae_db": 1.071146517076228, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09777458251008263, "300–1000 Hz_40ms_envelope_correlation": 0.9920846192928855, "300–1000 Hz_40ms_modulation_mae_db": 1.131319346584291, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8769355393797582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21808328943897762, "1000–3000 Hz_10ms_envelope_correlation": 0.9602973838736406, "1000–3000 Hz_10ms_modulation_mae_db": 0.89881945152954, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1056336962965112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13024237675139938, "1000–3000 Hz_40ms_envelope_correlation": 0.984968136474442, "1000–3000 Hz_40ms_modulation_mae_db": 1.0459424619205724, "3000–6000 Hz_10ms_log_rms_mae_db": 2.122248040621616, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17049173239216742, "3000–6000 Hz_10ms_envelope_correlation": 0.9803725885164182, "3000–6000 Hz_10ms_modulation_mae_db": 0.26037040775165227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2566469915217615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09357614079063542, "3000–6000 Hz_40ms_envelope_correlation": 0.9936597334710183, "3000–6000 Hz_40ms_modulation_mae_db": 0.4415004933157481, "mr_stft_db": 5.656697973902838, "band_mae_db": 0.19195328767905273, "rms_gain_error_db": -0.029993631555526607, "output_peak": 0.438750002024913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2484323480029067, "全帯域_10ms_rms_envelope_nrmse": 0.11017613686451219, "全帯域_10ms_envelope_correlation": 0.9882819419359485, "全帯域_10ms_modulation_mae_db": 0.5243616281657774, "全帯域_40ms_log_rms_mae_db": 0.5882862802833069, "全帯域_40ms_rms_envelope_nrmse": 0.049998973089885504, "全帯域_40ms_envelope_correlation": 0.9973884976687821, "全帯域_40ms_modulation_mae_db": 0.5725499709727598, "300–1000 Hz_10ms_log_rms_mae_db": 1.7839826572943236, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17376278235633788, "300–1000 Hz_10ms_envelope_correlation": 0.9768675795679345, "300–1000 Hz_10ms_modulation_mae_db": 0.3545748271167392, "300–1000 Hz_40ms_log_rms_mae_db": 1.0276333261836637, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09226148804785163, "300–1000 Hz_40ms_envelope_correlation": 0.9931746497800198, "300–1000 Hz_40ms_modulation_mae_db": 0.4288852860570908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.919936655142622, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23212308963206443, "1000–3000 Hz_10ms_envelope_correlation": 0.9551627138228879, "1000–3000 Hz_10ms_modulation_mae_db": 0.38053809321979104, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1535743825254157, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12243644759344643, "1000–3000 Hz_40ms_envelope_correlation": 0.9859009834410656, "1000–3000 Hz_40ms_modulation_mae_db": 0.39023295901043004, "3000–6000 Hz_10ms_log_rms_mae_db": 2.067612731419564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18061169105087693, "3000–6000 Hz_10ms_envelope_correlation": 0.9777618809941132, "3000–6000 Hz_10ms_modulation_mae_db": 0.31517804917415176, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2274647803054992, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0966054418972491, "3000–6000 Hz_40ms_envelope_correlation": 0.9932782154104742, "3000–6000 Hz_40ms_modulation_mae_db": 0.24487415257066522, "mr_stft_db": 5.672634837304293, "band_mae_db": 0.09282105180401896, "rms_gain_error_db": -0.09056630069330349, "output_peak": 0.3966866861829503}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.231643743113681, "全帯域_10ms_rms_envelope_nrmse": 0.12026871234776156, "全帯域_10ms_envelope_correlation": 0.9861142579021911, "全帯域_10ms_modulation_mae_db": 0.2506769259585438, "全帯域_40ms_log_rms_mae_db": 0.6458321234571579, "全帯域_40ms_rms_envelope_nrmse": 0.06099355106971042, "全帯域_40ms_envelope_correlation": 0.9960075650389464, "全帯域_40ms_modulation_mae_db": 0.31986115625559847, "300–1000 Hz_10ms_log_rms_mae_db": 1.8921420394685677, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17262562854986158, "300–1000 Hz_10ms_envelope_correlation": 0.9772280865691504, "300–1000 Hz_10ms_modulation_mae_db": 0.4120570573521845, "300–1000 Hz_40ms_log_rms_mae_db": 1.1186535659052312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09910511589787649, "300–1000 Hz_40ms_envelope_correlation": 0.9917813199634934, "300–1000 Hz_40ms_modulation_mae_db": 0.2279406354236173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8472452890700288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22899001998172488, "1000–3000 Hz_10ms_envelope_correlation": 0.9582923732997248, "1000–3000 Hz_10ms_modulation_mae_db": 0.22735735574151697, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0583336889130261, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1336041506036726, "1000–3000 Hz_40ms_envelope_correlation": 0.9844093496905465, "1000–3000 Hz_40ms_modulation_mae_db": 0.39616759427025566, "3000–6000 Hz_10ms_log_rms_mae_db": 2.105176811485891, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18011341890870003, "3000–6000 Hz_10ms_envelope_correlation": 0.977891108117869, "3000–6000 Hz_10ms_modulation_mae_db": 0.290958464338463, "3000–6000 Hz_40ms_log_rms_mae_db": 1.234241734685445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09669808291140677, "3000–6000 Hz_40ms_envelope_correlation": 0.9932395776091303, "3000–6000 Hz_40ms_modulation_mae_db": 0.3807731220235288, "mr_stft_db": 5.677111484020145, "band_mae_db": 0.14505231689885065, "rms_gain_error_db": 0.015425016217864315, "output_peak": 0.4167217141998564}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2676552500280072, "全帯域_10ms_rms_envelope_nrmse": 0.13251554036309005, "全帯域_10ms_envelope_correlation": 0.9832266441323005, "全帯域_10ms_modulation_mae_db": 0.1477338814476369, "全帯域_40ms_log_rms_mae_db": 0.6504219401112824, "全帯域_40ms_rms_envelope_nrmse": 0.07611970799271402, "全帯域_40ms_envelope_correlation": 0.9938252105857543, "全帯域_40ms_modulation_mae_db": 0.24330932916180023, "300–1000 Hz_10ms_log_rms_mae_db": 1.8896879128700663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19262058739378685, "300–1000 Hz_10ms_envelope_correlation": 0.9716917222278195, "300–1000 Hz_10ms_modulation_mae_db": 0.3023647326548679, "300–1000 Hz_40ms_log_rms_mae_db": 1.0350353580837668, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11090123195806927, "300–1000 Hz_40ms_envelope_correlation": 0.9897042557754526, "300–1000 Hz_40ms_modulation_mae_db": 0.2604597692365785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8019359999333215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2253169109373694, "1000–3000 Hz_10ms_envelope_correlation": 0.9574756218590409, "1000–3000 Hz_10ms_modulation_mae_db": 0.5915837189143559, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0182454385239763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12931909000095398, "1000–3000 Hz_40ms_envelope_correlation": 0.9845025791105593, "1000–3000 Hz_40ms_modulation_mae_db": 0.7569305233024224, "3000–6000 Hz_10ms_log_rms_mae_db": 2.113919636653365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18934179039651747, "3000–6000 Hz_10ms_envelope_correlation": 0.9755736793161118, "3000–6000 Hz_10ms_modulation_mae_db": 0.5664325938882996, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2529012858302424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10799288528598264, "3000–6000 Hz_40ms_envelope_correlation": 0.9914641672566502, "3000–6000 Hz_40ms_modulation_mae_db": 0.6430615783605921, "mr_stft_db": 5.669693602130649, "band_mae_db": 0.1540338466120022, "rms_gain_error_db": 0.014928156393842738, "output_peak": 0.47020702972713563}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.24976123704499, "全帯域_10ms_rms_envelope_nrmse": 0.12189236955764868, "全帯域_10ms_envelope_correlation": 0.9856460258182063, "全帯域_10ms_modulation_mae_db": 0.24206539828117654, "全帯域_40ms_log_rms_mae_db": 0.642042065812284, "全帯域_40ms_rms_envelope_nrmse": 0.06639949696016778, "全帯域_40ms_envelope_correlation": 0.9953624631402678, "全帯域_40ms_modulation_mae_db": 0.20739679253500887, "300–1000 Hz_10ms_log_rms_mae_db": 1.8477992611918743, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17742320138200468, "300–1000 Hz_10ms_envelope_correlation": 0.9758370353195084, "300–1000 Hz_10ms_modulation_mae_db": 0.27027779285623593, "300–1000 Hz_40ms_log_rms_mae_db": 1.0640716163164354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0972736403118432, "300–1000 Hz_40ms_envelope_correlation": 0.992326243071274, "300–1000 Hz_40ms_modulation_mae_db": 0.3036238814881101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.90025444221805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20031808208642313, "1000–3000 Hz_10ms_envelope_correlation": 0.9664574706727876, "1000–3000 Hz_10ms_modulation_mae_db": 0.23943418545055825, "1000–3000 Hz_40ms_log_rms_mae_db": 1.127258549401184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11144137481560674, "1000–3000 Hz_40ms_envelope_correlation": 0.9884419177795287, "1000–3000 Hz_40ms_modulation_mae_db": 0.42426312204383243, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0823289299288006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17308117157213593, "3000–6000 Hz_10ms_envelope_correlation": 0.979666260773518, "3000–6000 Hz_10ms_modulation_mae_db": 0.9225531694247452, "3000–6000 Hz_40ms_log_rms_mae_db": 1.215148480532365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09117177110657845, "3000–6000 Hz_40ms_envelope_correlation": 0.9939949736555485, "3000–6000 Hz_40ms_modulation_mae_db": 1.02573712663697, "mr_stft_db": 5.666560262988921, "band_mae_db": 0.13184617979901223, "rms_gain_error_db": -0.11355870395115442, "output_peak": 0.36880680849145714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.03371583874462, "spectral_convergence": 0.35381807083101263, "spectral_convergence_status": "ok", "amplitude_floor": 1.3116896377207775e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.457018524517663, "spectral_convergence": 0.3537651490741041, "spectral_convergence_status": "ok", "amplitude_floor": 8.457684224428572e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.523247911373983, "spectral_convergence": 0.38133944458219776, "spectral_convergence_status": "ok", "amplitude_floor": 6.717993311155654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008437417957183396, "output_power": 0.0008306765988064134, "delta_db": -0.06777586894211687, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011741200681074208, "output_power": 0.0011552373809156661, "delta_db": -0.07041277836318661, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004760668283187788, "output_power": 0.0004665707751607042, "delta_db": -0.08750389309561424, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001833223674641814, "output_power": 0.00017082775611417807, "delta_db": -0.3065702087515751, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.526890589901711e-05, "output_power": 6.74810485807276e-05, "delta_db": 0.14475489734764335, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6955820206525705e-05, "output_power": 3.887006431631726e-05, "delta_db": 0.21932413677209214, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14939029721203806, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07853762613335147, "pitch": null, "output_rms_dbfs": -25.638908677165347, "output_peak": 0.37970572121809537, "reference_rms_dbfs": -25.560371051032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.671327424878755}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001544797211376954, "log_rms_mae_db": 1.2643299667685435, "rms_envelope_nrmse": 0.12044866664449591, "envelope_correlation": 0.9859664759626184, "rms_gain_error_db": -0.07853251970349877, "reference_mean_power": 0.002781784143506645, "output_mean_power": 0.0027319338248324703}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.487428460245723e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008814675226478596, "output_power": 0.0008643272502091065, "reference_active": true, "delta_db": -0.08528110005580114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001997406840389717, "output_power": 0.00019444065952579893, "reference_active": true, "delta_db": -0.11679447348917457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0345220429658976e-05, "output_power": 3.4751389686979964e-05, "reference_active": true, "delta_db": -0.6481991616901762}], "active_band_count": 3, "active_band_mae_db": 0.28342491174505063}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013989081102856241, "log_rms_mae_db": 0.6576562668999375, "rms_envelope_nrmse": 0.06715431066589814, "envelope_correlation": 0.9951700900272702, "rms_gain_error_db": -0.078566718769977, "reference_mean_power": 0.002795409730019158, "output_mean_power": 0.0027452936189308226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3091173503801554e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008259246223406256, "output_power": 0.0008141508441253714, "reference_active": true, "delta_db": -0.06235535931616885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.583418527096058e-05, "output_power": 9.291897326842234e-05, "reference_active": true, "delta_db": -0.1341605276438996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329771424834818e-06, "output_power": 2.575345435974003e-06, "reference_active": true, "delta_db": -1.1157893196485364}], "active_band_count": 3, "active_band_mae_db": 0.437435068869535}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000132208474335363, "log_rms_mae_db": 1.8282512098333188, "rms_envelope_nrmse": 0.1659287477896968, "envelope_correlation": 0.9788921548511028, "rms_gain_error_db": -0.15300402322244872, "reference_mean_power": 0.001270855716483403, "output_mean_power": 0.0012268623653729157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.478174051796597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005667765898001897, "output_power": 0.0005501117757581661, "reference_active": true, "delta_db": -0.12960962192310752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011084740591226797, "output_power": 0.00010438522026234185, "reference_active": true, "delta_db": -0.2608652211336609}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2190579526022385e-05, "output_power": 2.0481369802112922e-05, "reference_active": true, "delta_db": -0.34809645291254526}], "active_band_count": 3, "active_band_mae_db": 0.2461904319897712}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012422751968092996, "log_rms_mae_db": 1.0771674120416244, "rms_envelope_nrmse": 0.09198320407861854, "envelope_correlation": 0.9930720804317296, "rms_gain_error_db": -0.15301168650092994, "reference_mean_power": 0.0012772403358267607, "output_mean_power": 0.0012330237919356323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.710312666254066e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005574547781532276, "output_power": 0.0005426782572224287, "reference_active": true, "delta_db": -0.11667220690990175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.253668134122962e-05, "output_power": 4.755808592827486e-05, "reference_active": true, "delta_db": -0.4323826838819306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7043530583570796e-06, "output_power": 1.413559101749885e-06, "reference_active": true, "delta_db": -0.8124559294838203}], "active_band_count": 3, "active_band_mae_db": 0.45383694009188424}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.199705053096058e-05, "log_rms_mae_db": 1.827959124539441, "rms_envelope_nrmse": 0.2067810809732303, "envelope_correlation": 0.9642136774535398, "rms_gain_error_db": -0.22046785879547032, "reference_mean_power": 0.00020286667816607547, "output_mean_power": 0.00019282526284853378}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3687094204678632e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.826983537989376e-05, "output_power": 6.782126263162421e-05, "reference_active": true, "delta_db": -0.6222854959805825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2413787132715612e-05, "output_power": 2.1889158951141342e-05, "reference_active": true, "delta_db": -0.10286168095869219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388555256663533e-06, "output_power": 5.521743720186402e-06, "reference_active": true, "delta_db": -0.6332640956770732}], "active_band_count": 3, "active_band_mae_db": 0.45280375753878266}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.313537286332709e-05, "log_rms_mae_db": 1.0529793090946158, "rms_envelope_nrmse": 0.11284139421309884, "envelope_correlation": 0.9883703559912862, "rms_gain_error_db": -0.22052582047600996, "reference_mean_power": 0.00020386380860129322, "output_mean_power": 0.00019377045160164175}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.180981609421061e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.558718162537124e-05, "output_power": 6.736521675065241e-05, "reference_active": true, "delta_db": -0.5001244087842}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0576597086854303e-05, "output_power": 1.1011806617011689e-05, "reference_active": true, "delta_db": 0.17512615643992063}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.38427493679316e-07, "output_power": 4.366175082415439e-07, "reference_active": true, "delta_db": -0.910260813148107}], "active_band_count": 3, "active_band_mae_db": 0.5285037927907426}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.0553751915640445e-05, "log_rms_mae_db": 2.0732719961345647, "rms_envelope_nrmse": 0.18394427001155758, "envelope_correlation": 0.978022263597412, "rms_gain_error_db": 0.21346014915254088, "reference_mean_power": 5.25173391881524e-05, "output_mean_power": 5.5163108250669035e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.578963632678691e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2240876569039177e-05, "output_power": 2.2989839522948128e-05, "reference_active": true, "delta_db": 0.1438403986558185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.96335110256818e-06, "output_power": 8.68189556476184e-06, "reference_active": true, "delta_db": 0.3751869326412606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4165297947437381e-06, "output_power": 1.296382712211481e-06, "reference_active": true, "delta_db": -0.384924828299072}], "active_band_count": 3, "active_band_mae_db": 0.30131738653205037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.394748258818145e-05, "log_rms_mae_db": 1.2452755962195583, "rms_envelope_nrmse": 0.10997292353858477, "envelope_correlation": 0.9918319735681463, "rms_gain_error_db": 0.21317084528993588, "reference_mean_power": 5.256947927681562e-05, "output_mean_power": 5.521419690455006e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.187575200111859e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.180909445738007e-05, "output_power": 2.2765978159070438e-05, "reference_active": true, "delta_db": 0.18648681398181585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0063771970727005e-06, "output_power": 4.4650741472282256e-06, "reference_active": true, "delta_db": 0.4707683985288636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0500724932970203e-07, "output_power": 9.099947403144688e-08, "reference_active": true, "delta_db": -0.6218040011604391}], "active_band_count": 3, "active_band_mae_db": 0.4263530712237062}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8276707995364667, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8099b30cd730b9e34e7f01f3a8475bd1da72a75e3d94f9002b2b52e0c2c91508", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9611423265068537, "standard_deviation": 0.003927870840564324, "minimum": 0.9551627138228879, "maximum": 0.9664574706727876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8642438915533726, "standard_deviation": 0.03800801388358764, "minimum": 1.8019359999333215, "maximum": 1.919936655142622, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5237950340598696, "standard_deviation": 0.28200014786209543, "minimum": 0.22735735574151697, "maximum": 0.9783871620368808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2165458724144198, "standard_deviation": 0.011419137757222766, "minimum": 0.20031808208642313, "maximum": 0.23212308963206443, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9865838621427769, "standard_deviation": 0.0018251931227432538, "minimum": 0.9844093496905465, "maximum": 0.9884419177795287, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0888054934957148, "standard_deviation": 0.04429418228484201, "minimum": 1.0182454385239763, "maximum": 1.1535743825254157, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.662245145066638, "standard_deviation": 0.2835722339436804, "minimum": 0.39023295901043004, "maximum": 1.095132340579691, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12168787680047266, "standard_deviation": 0.00874243815405525, "minimum": 0.11144137481560674, "maximum": 0.1336041506036726, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9788432915209963, "standard_deviation": 0.0018476118469276462, "minimum": 0.9755736793161118, "maximum": 0.9811325344641124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0869018549886835, "standard_deviation": 0.02622590661622957, "minimum": 2.0430605185519135, "maximum": 2.122248040621616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5145780661319104, "standard_deviation": 0.26235344308308217, "minimum": 0.26037040775165227, "maximum": 0.9225531694247452, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17678454077973427, "standard_deviation": 0.007894588113018908, "minimum": 0.16663311246133014, "maximum": 0.18934179039651747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933092836304582, "standard_deviation": 0.001172452732844662, "minimum": 0.9914641672566502, "maximum": 0.9950449296594641, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2262606600809067, "standard_deviation": 0.026566199291719845, "minimum": 1.1869028897353615, "maximum": 1.2566469915217615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5966289302637511, "standard_deviation": 0.27065032442188613, "minimum": 0.24487415257066522, "maximum": 1.02573712663697, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09653618533450616, "standard_deviation": 0.008619490788841448, "minimum": 0.08429064144636703, "maximum": 0.10997292353858477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9767186069566423, "standard_deviation": 0.0024108542622618364, "minimum": 0.9716917222278195, "maximum": 0.9797663054703013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8455721105984861, "standard_deviation": 0.0354381440074797, "minimum": 1.7839826572943236, "maximum": 1.8921420394685677, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.40592423934077937, "standard_deviation": 0.21378348543703443, "minimum": 0.2461904319897712, "maximum": 0.8986007083304558, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17431991101176142, "standard_deviation": 0.008933910347121586, "minimum": 0.16253353765083645, "maximum": 0.19262058739378685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.991963848170721, "standard_deviation": 0.001071983336335308, "minimum": 0.9897042557754526, "maximum": 0.9931746497800198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0612987571978092, "standard_deviation": 0.03581998262256978, "minimum": 1.0084881816461742, "maximum": 1.1186535659052312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4687652083514592, "standard_deviation": 0.28539242307997154, "minimum": 0.2279406354236173, "maximum": 1.131319346584291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09926446571028866, "standard_deviation": 0.006253366092605209, "minimum": 0.09198320407861854, "maximum": 0.11090123195806927, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13960276659199353, "standard_deviation": 0.029480136236412526, "minimum": 0.09282105180401896, "maximum": 0.19195328767905273, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6682174742235905, "standard_deviation": 0.006681645451066457, "minimum": 5.656697973902838, "maximum": 5.677111484020145, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40479883619244306, "standard_deviation": 0.03515739161472568, "minimum": 0.36880680849145714, "maximum": 0.47020702972713563, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.055484486794280995, "standard_deviation": 0.04979767134553801, "minimum": -0.11355870395115442, "maximum": 0.015425016217864315, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9858707601764676, "standard_deviation": 0.0013975203395864483, "minimum": 0.9832266441323005, "maximum": 0.9882819419359485, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2603246016787477, "standard_deviation": 0.022017940330266002, "minimum": 1.231643743113681, "maximum": 1.2969527506533043, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.291582088508053, "standard_deviation": 0.12687395326200204, "minimum": 0.1477338814476369, "maximum": 0.5243616281657774, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12091675851613629, "standard_deviation": 0.0061499726924661, "minimum": 0.11017613686451219, "maximum": 0.13251554036309005, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.995528989833971, "standard_deviation": 0.000991266334849243, "minimum": 0.9938252105857543, "maximum": 0.9973884976687821, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6458052457895483, "standard_deviation": 0.024396608101514103, "minimum": 0.5882862802833069, "maximum": 0.6636007559114229, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3790060114635365, "standard_deviation": 0.19361210114818772, "minimum": 0.1872563197017566, "maximum": 0.7400972611478854, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06440840529651928, "standard_deviation": 0.007333618670833673, "minimum": 0.049998973089885504, "maximum": 0.07611970799271402, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_260-123440-0020", "language": "en", "style": "normal", "speaker": "librispeech_260", "source_utterance_id": "260-123440-0020", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 79920, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_260-123440-0020.wav", "source_file_sha256": "c25e52bfc012e6b96dbb4ac7a1aa31f39c68f36da62cb684f3f0d2594dcdc165", "decoded_pcm_sha256": "c10a9cdd8fd4960bfb88b1611c0123695ebcd137816c47ff78925a8699616669", "original_source_sha256": "44be2bb2c15059058798a25f87de6310601074e1bc4b68bda0ad3ad1860a9040", "corpus_entry_id": "librispeech_260-123440-0020"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.3061909675598145, "predicted_mos": 3.3061909675598145, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "8e022374cf85d7c8f80cd89ab4e054fbbad65546d5e9832f785e1e3ce8bf180a", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.658392861998436, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_260-123440-0020", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "8e022374cf85d7c8f80cd89ab4e054fbbad65546d5e9832f785e1e3ce8bf180a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "66ad247eff03a910dee5b9feed83094cbef18c21f694b77ba10c47f4d65c570a"}, "rows": [{"case": "librispeech_260-123440-0020", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.774533748626709, "predicted_mos": 1.774533748626709, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "d295446d46513f0e85a4d149d9b599bf411009537197c5a17d34f2d2f040c812", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.132898231000581, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_260-123440-0020", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d295446d46513f0e85a4d149d9b599bf411009537197c5a17d34f2d2f040c812", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2785857094423847, "全帯域_10ms_rms_envelope_nrmse": 0.08469395054899394, "全帯域_10ms_envelope_correlation": 0.9929363386606525, "全帯域_10ms_modulation_mae_db": 0.21314875855649593, "全帯域_40ms_log_rms_mae_db": 0.6211080267390748, "全帯域_40ms_rms_envelope_nrmse": 0.040601153433828335, "全帯域_40ms_envelope_correlation": 0.9982596830869493, "全帯域_40ms_modulation_mae_db": 0.16305020461194494, "300–1000 Hz_10ms_log_rms_mae_db": 1.9310523478006567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13342254462747574, "300–1000 Hz_10ms_envelope_correlation": 0.9858578984022875, "300–1000 Hz_10ms_modulation_mae_db": 0.48416193777608396, "300–1000 Hz_40ms_log_rms_mae_db": 1.164436366688805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07873781570547037, "300–1000 Hz_40ms_envelope_correlation": 0.9948072144532468, "300–1000 Hz_40ms_modulation_mae_db": 0.7558246726926335, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0276498319940988, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1727470684875362, "1000–3000 Hz_10ms_envelope_correlation": 0.9791683513222121, "1000–3000 Hz_10ms_modulation_mae_db": 0.27374231198367605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.174398160250224, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09936348595973965, "1000–3000 Hz_40ms_envelope_correlation": 0.9930215791141529, "1000–3000 Hz_40ms_modulation_mae_db": 0.5154090881217527, "3000–6000 Hz_10ms_log_rms_mae_db": 2.313273993059565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3094501417714141, "3000–6000 Hz_10ms_envelope_correlation": 0.935612817064538, "3000–6000 Hz_10ms_modulation_mae_db": 3.014598123562409, "3000–6000 Hz_40ms_log_rms_mae_db": 1.586526278286622, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23257683208894162, "3000–6000 Hz_40ms_envelope_correlation": 0.9612494163719266, "3000–6000 Hz_40ms_modulation_mae_db": 4.299480462838393, "mr_stft_db": 6.116209984120999, "band_mae_db": 0.09873209144306992, "rms_gain_error_db": -0.050459756589471265, "output_peak": 0.4574988113667936}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3196874714574949, "全帯域_10ms_rms_envelope_nrmse": 0.10421833588725028, "全帯域_10ms_envelope_correlation": 0.9893123883938076, "全帯域_10ms_modulation_mae_db": 0.30174604261690025, "全帯域_40ms_log_rms_mae_db": 0.6884285871164375, "全帯域_40ms_rms_envelope_nrmse": 0.05556247112115406, "全帯域_40ms_envelope_correlation": 0.9967007282832012, "全帯域_40ms_modulation_mae_db": 0.4616851455098339, "300–1000 Hz_10ms_log_rms_mae_db": 1.9857896346726016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.145144577971966, "300–1000 Hz_10ms_envelope_correlation": 0.983141418032737, "300–1000 Hz_10ms_modulation_mae_db": 0.9179528896870824, "300–1000 Hz_40ms_log_rms_mae_db": 1.1851687848764936, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07942239635294439, "300–1000 Hz_40ms_envelope_correlation": 0.9946840416905076, "300–1000 Hz_40ms_modulation_mae_db": 1.258594411600261, "1000–3000 Hz_10ms_log_rms_mae_db": 2.056985843485257, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18352618922022754, "1000–3000 Hz_10ms_envelope_correlation": 0.9766086570593793, "1000–3000 Hz_10ms_modulation_mae_db": 0.5183379842908904, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2384047313111683, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10880704741603558, "1000–3000 Hz_40ms_envelope_correlation": 0.9918223731988774, "1000–3000 Hz_40ms_modulation_mae_db": 0.5984737924566113, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2773158844672303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3124408176245865, "3000–6000 Hz_10ms_envelope_correlation": 0.9374051064384371, "3000–6000 Hz_10ms_modulation_mae_db": 3.1151608876322316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5920435403918363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23268901041456302, "3000–6000 Hz_40ms_envelope_correlation": 0.9630208827740392, "3000–6000 Hz_40ms_modulation_mae_db": 4.366311009287231, "mr_stft_db": 6.119326728354196, "band_mae_db": 0.15473751426165322, "rms_gain_error_db": -0.025481912503361664, "output_peak": 0.44908966277977014}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3457844829064105, "全帯域_10ms_rms_envelope_nrmse": 0.0949175762354324, "全帯域_10ms_envelope_correlation": 0.9911275411065558, "全帯域_10ms_modulation_mae_db": 0.4550269773451401, "全帯域_40ms_log_rms_mae_db": 0.6253615975111991, "全帯域_40ms_rms_envelope_nrmse": 0.04222222725332929, "全帯域_40ms_envelope_correlation": 0.9981189496398699, "全帯域_40ms_modulation_mae_db": 0.6518444710861556, "300–1000 Hz_10ms_log_rms_mae_db": 1.9251817862039489, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1502560346102553, "300–1000 Hz_10ms_envelope_correlation": 0.9818566637118803, "300–1000 Hz_10ms_modulation_mae_db": 0.6979219112408795, "300–1000 Hz_40ms_log_rms_mae_db": 1.1018770567938205, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08149485614575167, "300–1000 Hz_40ms_envelope_correlation": 0.9942597331187786, "300–1000 Hz_40ms_modulation_mae_db": 0.6262834879625228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9897003118545418, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.168770302988073, "1000–3000 Hz_10ms_envelope_correlation": 0.9803848272792914, "1000–3000 Hz_10ms_modulation_mae_db": 0.6030289266884458, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1778708083224603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10111939199733404, "1000–3000 Hz_40ms_envelope_correlation": 0.9930165195029304, "1000–3000 Hz_40ms_modulation_mae_db": 0.8919982898210694, "3000–6000 Hz_10ms_log_rms_mae_db": 2.241592994114101, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30930645038253135, "3000–6000 Hz_10ms_envelope_correlation": 0.9377758494063233, "3000–6000 Hz_10ms_modulation_mae_db": 3.139992736703006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5849955542415013, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23172929190482414, "3000–6000 Hz_40ms_envelope_correlation": 0.9626438092863179, "3000–6000 Hz_40ms_modulation_mae_db": 4.379966900242668, "mr_stft_db": 6.11330840034672, "band_mae_db": 0.16314531899592463, "rms_gain_error_db": -0.04454546757811485, "output_peak": 0.4600598349747203}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.331236987821297, "全帯域_10ms_rms_envelope_nrmse": 0.08350983123805213, "全帯域_10ms_envelope_correlation": 0.9931478072471094, "全帯域_10ms_modulation_mae_db": 0.31613963830451475, "全帯域_40ms_log_rms_mae_db": 0.6747679230330077, "全帯域_40ms_rms_envelope_nrmse": 0.04171386713692894, "全帯域_40ms_envelope_correlation": 0.9981895572460247, "全帯域_40ms_modulation_mae_db": 0.545358646691216, "300–1000 Hz_10ms_log_rms_mae_db": 1.9166965437245695, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13613898878857875, "300–1000 Hz_10ms_envelope_correlation": 0.985240498065927, "300–1000 Hz_10ms_modulation_mae_db": 0.7540656819013488, "300–1000 Hz_40ms_log_rms_mae_db": 1.0693413550519264, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07953602574840407, "300–1000 Hz_40ms_envelope_correlation": 0.9946615937016978, "300–1000 Hz_40ms_modulation_mae_db": 0.8685569234742622, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9852180873078085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16057568730978442, "1000–3000 Hz_10ms_envelope_correlation": 0.9825803033165986, "1000–3000 Hz_10ms_modulation_mae_db": 0.31941641724692277, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1578600970591582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08452521872977156, "1000–3000 Hz_40ms_envelope_correlation": 0.994992225557153, "1000–3000 Hz_40ms_modulation_mae_db": 0.258144297564373, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2855291088860388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2947072100346623, "3000–6000 Hz_10ms_envelope_correlation": 0.9440573174430127, "3000–6000 Hz_10ms_modulation_mae_db": 3.380029932083652, "3000–6000 Hz_40ms_log_rms_mae_db": 1.605280458462091, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22521508854081698, "3000–6000 Hz_40ms_envelope_correlation": 0.9651267246400668, "3000–6000 Hz_40ms_modulation_mae_db": 4.642335285768745, "mr_stft_db": 6.105691762880046, "band_mae_db": 0.14070495132556646, "rms_gain_error_db": -0.0684472262928277, "output_peak": 0.49473723291906824}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3562713322203812, "全帯域_10ms_rms_envelope_nrmse": 0.09515979187553739, "全帯域_10ms_envelope_correlation": 0.9911690574850943, "全帯域_10ms_modulation_mae_db": 0.07990796828245635, "全帯域_40ms_log_rms_mae_db": 0.7415634463816674, "全帯域_40ms_rms_envelope_nrmse": 0.04944207369190748, "全帯域_40ms_envelope_correlation": 0.9975529169177483, "全帯域_40ms_modulation_mae_db": 0.10426010563702287, "300–1000 Hz_10ms_log_rms_mae_db": 1.9616661615280084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14115256852082503, "300–1000 Hz_10ms_envelope_correlation": 0.9841751561628466, "300–1000 Hz_10ms_modulation_mae_db": 0.5808360044825457, "300–1000 Hz_40ms_log_rms_mae_db": 1.2111541285772611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08359305469033221, "300–1000 Hz_40ms_envelope_correlation": 0.9942476877442347, "300–1000 Hz_40ms_modulation_mae_db": 0.8695761872535898, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9577404933203513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17343827264074332, "1000–3000 Hz_10ms_envelope_correlation": 0.9793304936642753, "1000–3000 Hz_10ms_modulation_mae_db": 0.5429092451624574, "1000–3000 Hz_40ms_log_rms_mae_db": 1.142120410463423, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08789481712322385, "1000–3000 Hz_40ms_envelope_correlation": 0.9943770261605687, "1000–3000 Hz_40ms_modulation_mae_db": 0.2920814457780934, "3000–6000 Hz_10ms_log_rms_mae_db": 2.303033294482233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32190705257001667, "3000–6000 Hz_10ms_envelope_correlation": 0.9383423754585469, "3000–6000 Hz_10ms_modulation_mae_db": 3.41119722062021, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6480698416352366, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2364522690363873, "3000–6000 Hz_40ms_envelope_correlation": 0.965539605423521, "3000–6000 Hz_40ms_modulation_mae_db": 4.6161911488045035, "mr_stft_db": 6.1084916997662475, "band_mae_db": 0.19303129278802592, "rms_gain_error_db": -0.12094625066413371, "output_peak": 0.4547415965108811}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2801187578438342, "全帯域_10ms_rms_envelope_nrmse": 0.09210720179786307, "全帯域_10ms_envelope_correlation": 0.9916656314287973, "全帯域_10ms_modulation_mae_db": 0.39056535302469636, "全帯域_40ms_log_rms_mae_db": 0.6699211176275793, "全帯域_40ms_rms_envelope_nrmse": 0.04723260024299231, "全帯域_40ms_envelope_correlation": 0.9976576512625527, "全帯域_40ms_modulation_mae_db": 0.6026723269643737, "300–1000 Hz_10ms_log_rms_mae_db": 1.9750338577278514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15375430003781848, "300–1000 Hz_10ms_envelope_correlation": 0.9810895393896213, "300–1000 Hz_10ms_modulation_mae_db": 0.6270874668860243, "300–1000 Hz_40ms_log_rms_mae_db": 1.2288493565862166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10261520264046864, "300–1000 Hz_40ms_envelope_correlation": 0.9907242796443825, "300–1000 Hz_40ms_modulation_mae_db": 0.7244533365781122, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0227442027092675, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1755847936673159, "1000–3000 Hz_10ms_envelope_correlation": 0.9793234868448232, "1000–3000 Hz_10ms_modulation_mae_db": 0.42335072798869805, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1358548412147746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08902782010243894, "1000–3000 Hz_40ms_envelope_correlation": 0.9945697153668253, "1000–3000 Hz_40ms_modulation_mae_db": 0.06598719270532091, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2220540287049517, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31384107457973465, "3000–6000 Hz_10ms_envelope_correlation": 0.9353504844348178, "3000–6000 Hz_10ms_modulation_mae_db": 3.276730100617396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5567955252852435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23986632664807184, "3000–6000 Hz_40ms_envelope_correlation": 0.9595949114851432, "3000–6000 Hz_40ms_modulation_mae_db": 4.469171704236989, "mr_stft_db": 6.0951146994419965, "band_mae_db": 0.10098842126280504, "rms_gain_error_db": -0.04789836642569544, "output_peak": 0.49094653223842577}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2995863917642383, "全帯域_10ms_rms_envelope_nrmse": 0.09237210931333444, "全帯域_10ms_envelope_correlation": 0.9916147025501895, "全帯域_10ms_modulation_mae_db": 0.3218921890109639, "全帯域_40ms_log_rms_mae_db": 0.6446065086725021, "全帯域_40ms_rms_envelope_nrmse": 0.04890015194275374, "全帯域_40ms_envelope_correlation": 0.9974433664849319, "全帯域_40ms_modulation_mae_db": 0.5202986146686026, "300–1000 Hz_10ms_log_rms_mae_db": 1.9865791488409077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1507851871661517, "300–1000 Hz_10ms_envelope_correlation": 0.9820031059996098, "300–1000 Hz_10ms_modulation_mae_db": 0.46297668564419614, "300–1000 Hz_40ms_log_rms_mae_db": 1.1740778347887044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08999445621602524, "300–1000 Hz_40ms_envelope_correlation": 0.9929825608962616, "300–1000 Hz_40ms_modulation_mae_db": 0.7830196523962459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9089274922874426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16052894302584128, "1000–3000 Hz_10ms_envelope_correlation": 0.9819602443321533, "1000–3000 Hz_10ms_modulation_mae_db": 0.6883241764292078, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0833866692321739, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07641360246632872, "1000–3000 Hz_40ms_envelope_correlation": 0.9956672795180431, "1000–3000 Hz_40ms_modulation_mae_db": 1.016424297226229, "3000–6000 Hz_10ms_log_rms_mae_db": 2.251129839208703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30801092184995105, "3000–6000 Hz_10ms_envelope_correlation": 0.9365743604787875, "3000–6000 Hz_10ms_modulation_mae_db": 3.2798314947495943, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5542759684905594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2355856355461299, "3000–6000 Hz_40ms_envelope_correlation": 0.9603393934121864, "3000–6000 Hz_40ms_modulation_mae_db": 4.571830124341816, "mr_stft_db": 6.1084137481565515, "band_mae_db": 0.11117560101009986, "rms_gain_error_db": 0.008721822411668068, "output_peak": 0.44599285973510017}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3079930084044202, "全帯域_10ms_rms_envelope_nrmse": 0.08695182596797747, "全帯域_10ms_envelope_correlation": 0.9926645181806362, "全帯域_10ms_modulation_mae_db": 0.12495191632919156, "全帯域_40ms_log_rms_mae_db": 0.717850500671845, "全帯域_40ms_rms_envelope_nrmse": 0.045395873467844246, "全帯域_40ms_envelope_correlation": 0.9979901590465872, "全帯域_40ms_modulation_mae_db": 0.4178801023657914, "300–1000 Hz_10ms_log_rms_mae_db": 1.9289904222403986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14596398275116862, "300–1000 Hz_10ms_envelope_correlation": 0.9829115667536594, "300–1000 Hz_10ms_modulation_mae_db": 0.2574731109615343, "300–1000 Hz_40ms_log_rms_mae_db": 1.176258573502501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08770693409894531, "300–1000 Hz_40ms_envelope_correlation": 0.9933321234631565, "300–1000 Hz_40ms_modulation_mae_db": 0.5462501803023314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9479898906411277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16471972446886682, "1000–3000 Hz_10ms_envelope_correlation": 0.9812875520136165, "1000–3000 Hz_10ms_modulation_mae_db": 0.42608270806148396, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1885461433303346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09565541076856833, "1000–3000 Hz_40ms_envelope_correlation": 0.9938135118604758, "1000–3000 Hz_40ms_modulation_mae_db": 0.18697289946809303, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2324678680100036, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31557494057018465, "3000–6000 Hz_10ms_envelope_correlation": 0.9333673699462998, "3000–6000 Hz_10ms_modulation_mae_db": 3.137931741427644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5745226807261168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2380518435316041, "3000–6000 Hz_40ms_envelope_correlation": 0.9594220660217739, "3000–6000 Hz_40ms_modulation_mae_db": 4.410156622959512, "mr_stft_db": 6.110476040004862, "band_mae_db": 0.11736271156888663, "rms_gain_error_db": -0.12750221523543123, "output_peak": 0.46382606429464013}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79920, "output_samples": 79920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.272392126635151, "spectral_convergence": 0.2616787868288194, "spectral_convergence_status": "ok", "amplitude_floor": 1.5443736044991425e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.967571550704384, "spectral_convergence": 0.2527655792385915, "spectral_convergence_status": "ok", "amplitude_floor": 1.1257392826698756e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.108666275023462, "spectral_convergence": 0.25929421169778777, "spectral_convergence_status": "ok", "amplitude_floor": 1.1201855379193364e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001035002994445706, "output_power": 0.001053492556510214, "delta_db": 0.07689865146502603, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0032595449236567883, "output_power": 0.003183669137281575, "delta_db": -0.10229043575197076, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007961593328208376, "output_power": 0.0008060557338859363, "delta_db": 0.0536508128250698, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023616410570225164, "output_power": 0.00022111148843446138, "delta_db": -0.28602582310561336, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.38734826098075e-05, "output_power": 5.38722840847876e-05, "delta_db": -9.661870132926684e-05, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3554110211771778e-05, "output_power": 1.378523053499321e-05, "delta_db": 0.07343020680941029, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09873209144306992, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.050459756589471265, "pitch": null, "output_rms_dbfs": -22.731109644843087, "output_peak": 0.4574988113667936, "reference_rms_dbfs": -22.68064988825362, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.116209984120999}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001944445221026879, "log_rms_mae_db": 1.2785857094423847, "rms_envelope_nrmse": 0.08469395054899394, "envelope_correlation": 0.9929363386606525, "rms_gain_error_db": -0.050459075552325164, "reference_mean_power": 0.005399702914122853, "output_mean_power": 0.0053373288082058205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 2.8891675057826967e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000788720956533661, "output_power": 0.0007872944364718414, "reference_active": true, "delta_db": -0.007861978545546882}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001646344464707057, "output_power": 0.0001515620937840078, "reference_active": true, "delta_db": -0.359301115426133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.1066196116820634e-05, "output_power": 5.437031222282686e-05, "reference_active": true, "delta_db": 0.27228318169780785}], "active_band_count": 3, "active_band_mae_db": 0.21314875855649593}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00017595749087279735, "log_rms_mae_db": 0.6211080267390748, "rms_envelope_nrmse": 0.040601153433828335, "envelope_correlation": 0.9982596830869493, "rms_gain_error_db": -0.05045949752047051, "reference_mean_power": 0.005432355653378944, "output_mean_power": 0.0053696038410245076}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 2.6677500098154174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007158752148089715, "output_power": 0.0007078817226264442, "reference_active": true, "delta_db": -0.04876627366510136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.294187442575086e-05, "output_power": 6.813047412431591e-05, "reference_active": true, "delta_db": -0.29635508182197695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7590528739181784e-06, "output_power": 3.885808198783779e-06, "reference_active": true, "delta_db": 0.14402925834875652}], "active_band_count": 3, "active_band_mae_db": 0.16305020461194494}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001659932762873133, "log_rms_mae_db": 1.9310523478006567, "rms_envelope_nrmse": 0.13342254462747574, "envelope_correlation": 0.9858578984022875, "rms_gain_error_db": -0.1647930156276374, "reference_mean_power": 0.0017617711849040457, "output_mean_power": 0.001696173215528407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.325463642910379e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006157322295664234, "output_power": 0.000589546107656394, "reference_active": true, "delta_db": -0.18874110287886486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015692982702212587, "output_power": 0.0001316912446734594, "reference_active": true, "delta_db": -0.7614859364090762}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0206834209715196e-05, "output_power": 3.5815746024658514e-05, "reference_active": true, "delta_db": -0.5022587740403108}], "active_band_count": 3, "active_band_mae_db": 0.48416193777608396}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00014319111528307625, "log_rms_mae_db": 1.164436366688805, "rms_envelope_nrmse": 0.07873781570547037, "envelope_correlation": 0.9948072144532468, "rms_gain_error_db": -0.1647932971424884, "reference_mean_power": 0.0017724260962696937, "output_mean_power": 0.0017064312902162317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.2006657592514347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005739716253322745, "output_power": 0.0005496990021202199, "reference_active": true, "delta_db": -0.18765474710237484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.210701544963183e-05, "output_power": 6.367319927448867e-05, "reference_active": true, "delta_db": -0.5402084859955706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.009713658025178e-06, "output_power": 2.8128869704008793e-06, "reference_active": true, "delta_db": -1.539610784979955}], "active_band_count": 3, "active_band_mae_db": 0.7558246726926335}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.00012416177010932925, "log_rms_mae_db": 2.0276498319940988, "rms_envelope_nrmse": 0.1727470684875362, "envelope_correlation": 0.9791683513222121, "rms_gain_error_db": -0.22274828411661213, "reference_mean_power": 0.0002459141129048244, "output_mean_power": 0.0002336192498147329}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.7078032497142614e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.578711086190898e-05, "output_power": 8.681348720104681e-05, "reference_active": true, "delta_db": -0.4271987253674339}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5502167493014184e-05, "output_power": 1.5626185652274426e-05, "reference_active": true, "delta_db": 0.03460554880545472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901592249182956e-06, "output_power": 5.432840236109835e-06, "reference_active": true, "delta_db": -0.35942266177813953}], "active_band_count": 3, "active_band_mae_db": 0.27374231198367605}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 9.745648796239735e-05, "log_rms_mae_db": 1.174398160250224, "rms_envelope_nrmse": 0.09936348595973965, "envelope_correlation": 0.9930215791141529, "rms_gain_error_db": -0.22275565458571164, "reference_mean_power": 0.00024740043253105237, "output_mean_power": 0.00023503085967778393}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.5717362127850378e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.047075579844295e-05, "output_power": 8.13194091049893e-05, "reference_active": true, "delta_db": -0.46314004212311133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.5496928110188975e-06, "output_power": 8.085242903801536e-06, "reference_active": true, "delta_db": 0.2976379087817882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.070423026829593e-07, "output_power": 3.396992712162978e-07, "reference_active": true, "delta_db": -0.7854493134603586}], "active_band_count": 3, "active_band_mae_db": 0.5154090881217527}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 3.6263821941295626e-05, "log_rms_mae_db": 2.313273993059565, "rms_envelope_nrmse": 0.3094501417714141, "envelope_correlation": 0.935612817064538, "rms_gain_error_db": -0.10833200846839006, "reference_mean_power": 2.0515613806993198e-05, "output_mean_power": 2.0010194697917066e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 4.904304308942738e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.621343716362641e-06, "output_power": 2.0640617112742997e-06, "reference_active": true, "delta_db": 1.048475849158864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.453801656353405e-07, "output_power": 9.134230866769138e-07, "reference_active": true, "delta_db": 0.88294152046336}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.409211497093947e-07, "output_power": 1.2391178896909092e-06, "reference_active": true, "delta_db": 7.112377001065003}], "active_band_count": 3, "active_band_mae_db": 3.014598123562409}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 2.5114070553475023e-05, "log_rms_mae_db": 1.586526278286622, "rms_envelope_nrmse": 0.23257683208894162, "envelope_correlation": 0.9612494163719266, "rms_gain_error_db": -0.10833236516968046, "reference_mean_power": 2.063955290066712e-05, "output_mean_power": 2.01310787961055e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 4.623298998188318e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6281370009575077e-06, "output_power": 2.6022109746613096e-06, "reference_active": true, "delta_db": 2.036515579806763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7397133003103647e-07, "output_power": 6.730848414159955e-07, "reference_active": true, "delta_db": 2.5523150097400777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7174687173869334e-08, "output_power": 1.1637237969908562e-07, "reference_active": true, "delta_db": 8.30961079896834}], "active_band_count": 3, "active_band_mae_db": 4.299480462838393}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8571229423423884, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1a1ffe447291a2977ee73591ecb7fbaa38a43081f6e10bfcc585d2f565c64860", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9800804894790437, "standard_deviation": 0.001903130206003381, "minimum": 0.9766086570593793, "maximum": 0.9825803033165986, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9871195191999869, "standard_deviation": 0.04827251746802949, "minimum": 1.9089274922874426, "maximum": 2.056985843485257, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47439906223147277, "standard_deviation": 0.14048742939564246, "minimum": 0.27374231198367605, "maximum": 0.6883241764292078, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16998637272604855, "standard_deviation": 0.007941090136129857, "minimum": 0.16052894302584128, "maximum": 0.18352618922022754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939100287848783, "standard_deviation": 0.001246003665738052, "minimum": 0.9918223731988774, "maximum": 0.9956672795180431, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1623052326479648, "standard_deviation": 0.045076433928397976, "minimum": 1.0833866692321739, "maximum": 1.2384047313111683, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.47818641289269287, "standard_deviation": 0.34112460431416175, "minimum": 0.06598719270532091, "maximum": 1.016424297226229, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09285084932043008, "standard_deviation": 0.010364266034203292, "minimum": 0.07641360246632872, "maximum": 0.10880704741603558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9373107100838454, "standard_deviation": 0.003153768200707784, "minimum": 0.9333673699462998, "maximum": 0.9440573174430127, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2657996263666034, "standard_deviation": 0.03378056361187477, "minimum": 2.2220540287049517, "maximum": 2.313273993059565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.2194340296745176, "standard_deviation": 0.13901646223578087, "minimum": 3.014598123562409, "maximum": 3.41119722062021, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3106548261728852, "standard_deviation": 0.007826220957147042, "minimum": 0.2947072100346623, "maximum": 0.32190705257001667, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9621171011768719, "standard_deviation": 0.0023725630644067814, "minimum": 0.9594220660217739, "maximum": 0.965539605423521, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5878137309399007, "standard_deviation": 0.029825537946600922, "minimum": 1.5542759684905594, "maximum": 1.6480698416352366, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.469430407309982, "standard_deviation": 0.12702883570097712, "minimum": 4.299480462838393, "maximum": 4.642335285768745, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23402078721391736, "standard_deviation": 0.004558046514332454, "minimum": 0.22521508854081698, "maximum": 0.23986632664807184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9832844808148211, "standard_deviation": 0.0016860633336215668, "minimum": 0.9810895393896213, "maximum": 0.9858578984022875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.951373737842368, "standard_deviation": 0.029012578576499873, "minimum": 1.9166965437245695, "maximum": 1.9865791488409077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5978094610724619, "standard_deviation": 0.20177784532660134, "minimum": 0.2574731109615343, "maximum": 0.9179528896870824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14457727305927995, "standard_deviation": 0.007217145875380686, "minimum": 0.13342254462747574, "maximum": 0.15375430003781848, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9937124043390333, "standard_deviation": 0.0013750802614745155, "minimum": 0.9907242796443825, "maximum": 0.9948072144532468, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1638954321082162, "standard_deviation": 0.05338341877313843, "minimum": 1.0693413550519264, "maximum": 1.2288493565862166, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8040698565324949, "standard_deviation": 0.2145529271149445, "minimum": 0.5462501803023314, "maximum": 1.258594411600261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08538759269979274, "standard_deviation": 0.008069437775732356, "minimum": 0.07873781570547037, "maximum": 0.10261520264046864, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13498473783200396, "standard_deviation": 0.033663733366382606, "minimum": 0.09873209144306992, "maximum": 0.19303129278802592, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.109629132883953, "standard_deviation": 0.007373192722521807, "minimum": 6.0951146994419965, "maximum": 6.119326728354196, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46461157435242495, "standard_deviation": 0.018353725978658808, "minimum": 0.44599285973510017, "maximum": 0.49473723291906824, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05956992160967097, "standard_deviation": 0.045874621082526844, "minimum": -0.12750221523543123, "maximum": 0.008721822411668068, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9917047481316053, "standard_deviation": 0.0012463403138690862, "minimum": 0.9893123883938076, "maximum": 0.9931478072471094, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3149080177325576, "standard_deviation": 0.028700250343784508, "minimum": 1.2785857094423847, "maximum": 1.3562713322203812, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2754223554337949, "standard_deviation": 0.12804644172310828, "minimum": 0.07990796828245635, "maximum": 0.4550269773451401, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09174132785805514, "standard_deviation": 0.006743720044125886, "minimum": 0.08350983123805213, "maximum": 0.10421833588725028, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977391264959832, "standard_deviation": 0.0005193687157379155, "minimum": 0.9967007282832012, "maximum": 0.9982596830869493, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6729509634691642, "standard_deviation": 0.04264726645549462, "minimum": 0.6211080267390748, "maximum": 0.7415634463816674, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4333812021918676, "standard_deviation": 0.19961398863240234, "minimum": 0.10426010563702287, "maximum": 0.6518444710861556, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0463838022863423, "standard_deviation": 0.0049884937791281414, "minimum": 0.040601153433828335, "maximum": 0.05556247112115406, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123440-0020", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.8893160820007324, "predicted_mos": 1.8893160820007324, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "a5c55c219db0b6aa7021e414868ae830fd08006ee136dc8e5b3f4cd06fa32e17", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.365324000362307, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2873975099885442, "全帯域_10ms_rms_envelope_nrmse": 0.08470487835352453, "全帯域_10ms_envelope_correlation": 0.9929380928515038, "全帯域_10ms_modulation_mae_db": 0.24248607600749084, "全帯域_40ms_log_rms_mae_db": 0.6250399404942557, "全帯域_40ms_rms_envelope_nrmse": 0.03976557048426788, "全帯域_40ms_envelope_correlation": 0.9983401970834672, "全帯域_40ms_modulation_mae_db": 0.18531725550179143, "300–1000 Hz_10ms_log_rms_mae_db": 1.947179258287305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13517220331357913, "300–1000 Hz_10ms_envelope_correlation": 0.985427070539843, "300–1000 Hz_10ms_modulation_mae_db": 0.5117187743173834, "300–1000 Hz_40ms_log_rms_mae_db": 1.1778237108590035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07534804950180254, "300–1000 Hz_40ms_envelope_correlation": 0.9951786301341083, "300–1000 Hz_40ms_modulation_mae_db": 0.8047251566022018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.928149098282262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15940210164047425, "1000–3000 Hz_10ms_envelope_correlation": 0.9826662118306166, "1000–3000 Hz_10ms_modulation_mae_db": 0.4066858899919235, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0526290024904084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0861403379462724, "1000–3000 Hz_40ms_envelope_correlation": 0.995388096360951, "1000–3000 Hz_40ms_modulation_mae_db": 0.46533561681378827, "3000–6000 Hz_10ms_log_rms_mae_db": 2.204641102568429, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3087349584432217, "3000–6000 Hz_10ms_envelope_correlation": 0.9351602206172964, "3000–6000 Hz_10ms_modulation_mae_db": 2.812376828655746, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4719173539868302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2307529834724104, "3000–6000 Hz_40ms_envelope_correlation": 0.9615634670854488, "3000–6000 Hz_40ms_modulation_mae_db": 4.21040459385626, "mr_stft_db": 6.008345752498861, "band_mae_db": 0.1752839834739652, "rms_gain_error_db": -0.05659091327360422, "output_peak": 0.45750340141189033}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.331111724400691, "全帯域_10ms_rms_envelope_nrmse": 0.1057473941436117, "全帯域_10ms_envelope_correlation": 0.988986824534817, "全帯域_10ms_modulation_mae_db": 0.38475301810400464, "全帯域_40ms_log_rms_mae_db": 0.6934742779967198, "全帯域_40ms_rms_envelope_nrmse": 0.05866094980643491, "全帯域_40ms_envelope_correlation": 0.9963288950132252, "全帯域_40ms_modulation_mae_db": 0.5632487269373203, "300–1000 Hz_10ms_log_rms_mae_db": 2.029195264391437, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15166546032140776, "300–1000 Hz_10ms_envelope_correlation": 0.9815661998981109, "300–1000 Hz_10ms_modulation_mae_db": 0.9310981944381922, "300–1000 Hz_40ms_log_rms_mae_db": 1.1941648615534566, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08605357654584812, "300–1000 Hz_40ms_envelope_correlation": 0.9937092705886731, "300–1000 Hz_40ms_modulation_mae_db": 1.2736325949357055, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9301543863932202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17424936828899726, "1000–3000 Hz_10ms_envelope_correlation": 0.9792875643481072, "1000–3000 Hz_10ms_modulation_mae_db": 1.0543802026291982, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1301246921736814, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0972986982466042, "1000–3000 Hz_40ms_envelope_correlation": 0.9941114867929903, "1000–3000 Hz_40ms_modulation_mae_db": 1.358842352286194, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1214361330509117, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3031916712888192, "3000–6000 Hz_10ms_envelope_correlation": 0.938407016882095, "3000–6000 Hz_10ms_modulation_mae_db": 2.8484451956866574, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4057606777145955, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2263452026600196, "3000–6000 Hz_40ms_envelope_correlation": 0.9631724214598388, "3000–6000 Hz_40ms_modulation_mae_db": 4.225920180920365, "mr_stft_db": 6.019194973993751, "band_mae_db": 0.1803552632413965, "rms_gain_error_db": -0.03829359145744186, "output_peak": 0.44793289764336913}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3751246099873153, "全帯域_10ms_rms_envelope_nrmse": 0.09699015051523582, "全帯域_10ms_envelope_correlation": 0.9907401524626399, "全帯域_10ms_modulation_mae_db": 0.3993437551858608, "全帯域_40ms_log_rms_mae_db": 0.6409733381396392, "全帯域_40ms_rms_envelope_nrmse": 0.043914614132295964, "全帯域_40ms_envelope_correlation": 0.997970112543512, "全帯域_40ms_modulation_mae_db": 0.5904706698733817, "300–1000 Hz_10ms_log_rms_mae_db": 1.9301671933667495, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.155447857009973, "300–1000 Hz_10ms_envelope_correlation": 0.9805679710633552, "300–1000 Hz_10ms_modulation_mae_db": 0.711853263388139, "300–1000 Hz_40ms_log_rms_mae_db": 1.1071820164225814, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0875338084882107, "300–1000 Hz_40ms_envelope_correlation": 0.9933650438824103, "300–1000 Hz_40ms_modulation_mae_db": 0.6451060582058923, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9011037373039554, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15409619491690532, "1000–3000 Hz_10ms_envelope_correlation": 0.9841267042905991, "1000–3000 Hz_10ms_modulation_mae_db": 0.8875365897632138, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0726134369925187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08795920544479421, "1000–3000 Hz_40ms_envelope_correlation": 0.995426331502016, "1000–3000 Hz_40ms_modulation_mae_db": 1.1601504620178384, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1234009887487186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30704286236216893, "3000–6000 Hz_10ms_envelope_correlation": 0.9373233119530747, "3000–6000 Hz_10ms_modulation_mae_db": 2.8961114799274594, "3000–6000 Hz_40ms_log_rms_mae_db": 1.414167827514377, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2357014813454766, "3000–6000 Hz_40ms_envelope_correlation": 0.9602920406833412, "3000–6000 Hz_40ms_modulation_mae_db": 4.234766904918879, "mr_stft_db": 6.006787489834256, "band_mae_db": 0.16387607392180295, "rms_gain_error_db": -0.05170426401108884, "output_peak": 0.45776573276373095}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3386805095166159, "全帯域_10ms_rms_envelope_nrmse": 0.08375060732612079, "全帯域_10ms_envelope_correlation": 0.9930971683316379, "全帯域_10ms_modulation_mae_db": 0.26443115951722634, "全帯域_40ms_log_rms_mae_db": 0.6836204746553312, "全帯域_40ms_rms_envelope_nrmse": 0.04108951335845655, "全帯域_40ms_envelope_correlation": 0.9982250404807306, "全帯域_40ms_modulation_mae_db": 0.4222413681540518, "300–1000 Hz_10ms_log_rms_mae_db": 1.9737024575932907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1369021443334644, "300–1000 Hz_10ms_envelope_correlation": 0.9850509273817095, "300–1000 Hz_10ms_modulation_mae_db": 0.7935248020873855, "300–1000 Hz_40ms_log_rms_mae_db": 1.0900945267014621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0783661452850921, "300–1000 Hz_40ms_envelope_correlation": 0.9947583285114426, "300–1000 Hz_40ms_modulation_mae_db": 0.8959216320583637, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9278167201295506, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15392709375873645, "1000–3000 Hz_10ms_envelope_correlation": 0.9836644674931088, "1000–3000 Hz_10ms_modulation_mae_db": 0.43899724935027856, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1270866347557709, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08124991036246623, "1000–3000 Hz_40ms_envelope_correlation": 0.9952083414675846, "1000–3000 Hz_40ms_modulation_mae_db": 0.3802227147017927, "3000–6000 Hz_10ms_log_rms_mae_db": 2.150922597666269, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29089264399788345, "3000–6000 Hz_10ms_envelope_correlation": 0.9431276961764994, "3000–6000 Hz_10ms_modulation_mae_db": 3.0527308896381107, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4265103715028995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22411678146852115, "3000–6000 Hz_40ms_envelope_correlation": 0.963792552885035, "3000–6000 Hz_40ms_modulation_mae_db": 4.447569433234402, "mr_stft_db": 5.995887491333981, "band_mae_db": 0.10677524839921089, "rms_gain_error_db": -0.05580327609793274, "output_peak": 0.4927829679606267}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.33634028000423, "全帯域_10ms_rms_envelope_nrmse": 0.09478235416105499, "全帯域_10ms_envelope_correlation": 0.991219787939337, "全帯域_10ms_modulation_mae_db": 0.14182510154181685, "全帯域_40ms_log_rms_mae_db": 0.7359611962714921, "全帯域_40ms_rms_envelope_nrmse": 0.049348581007923144, "全帯域_40ms_envelope_correlation": 0.9975287207444274, "全帯域_40ms_modulation_mae_db": 0.09565578354820166, "300–1000 Hz_10ms_log_rms_mae_db": 1.987381193109282, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14252389069390714, "300–1000 Hz_10ms_envelope_correlation": 0.983771978044929, "300–1000 Hz_10ms_modulation_mae_db": 0.6093424581855412, "300–1000 Hz_40ms_log_rms_mae_db": 1.2064425786610036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08360763340724699, "300–1000 Hz_40ms_envelope_correlation": 0.9941054720647158, "300–1000 Hz_40ms_modulation_mae_db": 0.8759071694172181, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8395967832710713, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17112242412195225, "1000–3000 Hz_10ms_envelope_correlation": 0.9797401470380428, "1000–3000 Hz_10ms_modulation_mae_db": 0.4269478093679251, "1000–3000 Hz_40ms_log_rms_mae_db": 1.036341282790597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08565182084226924, "1000–3000 Hz_40ms_envelope_correlation": 0.9945800096182911, "1000–3000 Hz_40ms_modulation_mae_db": 0.2287163750357735, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1395040143552815, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3045662279229362, "3000–6000 Hz_10ms_envelope_correlation": 0.9425006954531919, "3000–6000 Hz_10ms_modulation_mae_db": 3.0173020487243725, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4211324778605066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22862129756829677, "3000–6000 Hz_40ms_envelope_correlation": 0.9657254446235873, "3000–6000 Hz_40ms_modulation_mae_db": 4.353525761444086, "mr_stft_db": 5.995599199764538, "band_mae_db": 0.2305637620507277, "rms_gain_error_db": -0.10944086547038621, "output_peak": 0.4535405543444284}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3192125952734646, "全帯域_10ms_rms_envelope_nrmse": 0.09313321758261897, "全帯域_10ms_envelope_correlation": 0.9914704213143029, "全帯域_10ms_modulation_mae_db": 0.3035872403468188, "全帯域_40ms_log_rms_mae_db": 0.6899604162380876, "全帯域_40ms_rms_envelope_nrmse": 0.04704165604575603, "全帯域_40ms_envelope_correlation": 0.9976745941254783, "全帯域_40ms_modulation_mae_db": 0.49950341300121154, "300–1000 Hz_10ms_log_rms_mae_db": 2.0351395209633267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15210653276257294, "300–1000 Hz_10ms_envelope_correlation": 0.9815280122865228, "300–1000 Hz_10ms_modulation_mae_db": 0.6363800432517076, "300–1000 Hz_40ms_log_rms_mae_db": 1.242232941281232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09996488256938366, "300–1000 Hz_40ms_envelope_correlation": 0.9912131868998424, "300–1000 Hz_40ms_modulation_mae_db": 0.7471804662240787, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9640988330119753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17114453491566692, "1000–3000 Hz_10ms_envelope_correlation": 0.9799891311145806, "1000–3000 Hz_10ms_modulation_mae_db": 0.37807211409114877, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1401434045110552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08686401015532685, "1000–3000 Hz_40ms_envelope_correlation": 0.9946220419224706, "1000–3000 Hz_40ms_modulation_mae_db": 0.0783814932304354, "3000–6000 Hz_10ms_log_rms_mae_db": 2.123481542327658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3070905619368732, "3000–6000 Hz_10ms_envelope_correlation": 0.9361495362980495, "3000–6000 Hz_10ms_modulation_mae_db": 2.9827705676254808, "3000–6000 Hz_40ms_log_rms_mae_db": 1.432642492678635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23355452775162258, "3000–6000 Hz_40ms_envelope_correlation": 0.9605503846103441, "3000–6000 Hz_40ms_modulation_mae_db": 4.303635534589354, "mr_stft_db": 5.983297005497754, "band_mae_db": 0.1072402159054378, "rms_gain_error_db": -0.05691981724989616, "output_peak": 0.49268311338186715}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3311990679957593, "全帯域_10ms_rms_envelope_nrmse": 0.09328807300684068, "全帯域_10ms_envelope_correlation": 0.9914375572203695, "全帯域_10ms_modulation_mae_db": 0.36583158634138896, "全帯域_40ms_log_rms_mae_db": 0.6729553810005684, "全帯域_40ms_rms_envelope_nrmse": 0.04866985203658741, "全帯域_40ms_envelope_correlation": 0.9974625536589374, "全帯域_40ms_modulation_mae_db": 0.6433932758898403, "300–1000 Hz_10ms_log_rms_mae_db": 2.022815650601422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15114622307027975, "300–1000 Hz_10ms_envelope_correlation": 0.9819131852651936, "300–1000 Hz_10ms_modulation_mae_db": 0.38453579066355575, "300–1000 Hz_40ms_log_rms_mae_db": 1.1704747110615736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08884241915947187, "300–1000 Hz_40ms_envelope_correlation": 0.9931664372405349, "300–1000 Hz_40ms_modulation_mae_db": 0.7248903791155575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9056122134516154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1534214511532762, "1000–3000 Hz_10ms_envelope_correlation": 0.9835393618706169, "1000–3000 Hz_10ms_modulation_mae_db": 0.9310766568422353, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0985430476520266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0781838844894025, "1000–3000 Hz_40ms_envelope_correlation": 0.995612513104184, "1000–3000 Hz_40ms_modulation_mae_db": 1.1771601367201194, "3000–6000 Hz_10ms_log_rms_mae_db": 2.102084567291774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30045374769089556, "3000–6000 Hz_10ms_envelope_correlation": 0.9385527777489064, "3000–6000 Hz_10ms_modulation_mae_db": 2.9766701790142185, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3555042724926225, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23418699403046817, "3000–6000 Hz_40ms_envelope_correlation": 0.9605585874306126, "3000–6000 Hz_40ms_modulation_mae_db": 4.393871557063315, "mr_stft_db": 6.007029230219172, "band_mae_db": 0.21305528735278215, "rms_gain_error_db": 0.0008444260083275868, "output_peak": 0.44610990023965025}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3093155542467516, "全帯域_10ms_rms_envelope_nrmse": 0.08650088643770276, "全帯域_10ms_envelope_correlation": 0.9927093837994291, "全帯域_10ms_modulation_mae_db": 0.13286616960577244, "全帯域_40ms_log_rms_mae_db": 0.7169247569907552, "全帯域_40ms_rms_envelope_nrmse": 0.046088165765349576, "全帯域_40ms_envelope_correlation": 0.9978747746594518, "全帯域_40ms_modulation_mae_db": 0.3156317735222651, "300–1000 Hz_10ms_log_rms_mae_db": 1.8883048763756007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1421142884912789, "300–1000 Hz_10ms_envelope_correlation": 0.9838304830277038, "300–1000 Hz_10ms_modulation_mae_db": 0.22400714651122108, "300–1000 Hz_40ms_log_rms_mae_db": 1.0847238218048887, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08406042822348181, "300–1000 Hz_40ms_envelope_correlation": 0.9939125888647202, "300–1000 Hz_40ms_modulation_mae_db": 0.42466445582283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8810462577811955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15313656096939293, "1000–3000 Hz_10ms_envelope_correlation": 0.9841476589194355, "1000–3000 Hz_10ms_modulation_mae_db": 0.3405133848627144, "1000–3000 Hz_40ms_log_rms_mae_db": 1.095855426392426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08506511920821716, "1000–3000 Hz_40ms_envelope_correlation": 0.9956026416241852, "1000–3000 Hz_40ms_modulation_mae_db": 0.2783675891283557, "3000–6000 Hz_10ms_log_rms_mae_db": 2.097413977114052, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30436636041028786, "3000–6000 Hz_10ms_envelope_correlation": 0.9369230461538949, "3000–6000 Hz_10ms_modulation_mae_db": 2.90321833851814, "3000–6000 Hz_40ms_log_rms_mae_db": 1.411664646228329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23731610286253577, "3000–6000 Hz_40ms_envelope_correlation": 0.9593361818928188, "3000–6000 Hz_40ms_modulation_mae_db": 4.311534271138741, "mr_stft_db": 6.016291730509227, "band_mae_db": 0.21368712282481453, "rms_gain_error_db": -0.11273382201333766, "output_peak": 0.4656364315991206}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79920, "output_samples": 79920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.187533762124067, "spectral_convergence": 0.26036092814077644, "spectral_convergence_status": "ok", "amplitude_floor": 1.5443736044991425e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.863775442551977, "spectral_convergence": 0.25134556323117013, "spectral_convergence_status": "ok", "amplitude_floor": 1.1257392826698756e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.973728052820536, "spectral_convergence": 0.2570760442653579, "spectral_convergence_status": "ok", "amplitude_floor": 1.1201855379193364e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001035002994445706, "output_power": 0.0010428617543523984, "delta_db": 0.03285134259150038, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0032595449236567883, "output_power": 0.003194383224002848, "delta_db": -0.08769954544605163, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007961593328208376, "output_power": 0.000805031157212131, "delta_db": 0.048126988403603646, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023616410570225164, "output_power": 0.00022055257665185344, "delta_db": -0.2970175451775145, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.38734826098075e-05, "output_power": 4.847283824324359e-05, "delta_db": -0.45876601256001076, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3554110211771778e-05, "output_power": 1.3162754036550522e-05, "delta_db": -0.12724246666511016, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1752839834739652, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05659091327360422, "pitch": null, "output_rms_dbfs": -22.73724080152722, "output_peak": 0.45750340141189033, "reference_rms_dbfs": -22.68064988825362, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.008345752498861}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001944445221026879, "log_rms_mae_db": 1.2873975099885442, "rms_envelope_nrmse": 0.08470487835352453, "envelope_correlation": 0.9929380928515038, "rms_gain_error_db": -0.05659017995039463, "reference_mean_power": 0.005399702914122853, "output_mean_power": 0.0053297992093545386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 2.8891675057826967e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000788720956533661, "output_power": 0.0007862594257825252, "reference_active": true, "delta_db": -0.013575154391472024}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001646344464707057, "output_power": 0.00014938651067057332, "reference_active": true, "delta_db": -0.42209324582262364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.1066196116820634e-05, "output_power": 5.461506961547662e-05, "reference_active": true, "delta_db": 0.29178982780837687}], "active_band_count": 3, "active_band_mae_db": 0.24248607600749084}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00017595749087279735, "log_rms_mae_db": 0.6250399404942557, "rms_envelope_nrmse": 0.03976557048426788, "envelope_correlation": 0.9983401970834672, "rms_gain_error_db": -0.05659027053063585, "reference_mean_power": 0.005432355653378944, "output_mean_power": 0.005362029119548041}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 2.6677500098154174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007158752148089715, "output_power": 0.0007072777477818209, "reference_active": true, "delta_db": -0.052473318187492776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.294187442575086e-05, "output_power": 6.789591979885752e-05, "reference_active": true, "delta_db": -0.31133243250735826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7590528739181784e-06, "output_power": 3.9290995818784175e-06, "reference_active": true, "delta_db": 0.1921460158105232}], "active_band_count": 3, "active_band_mae_db": 0.18531725550179143}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001659932762873133, "log_rms_mae_db": 1.947179258287305, "rms_envelope_nrmse": 0.13517220331357913, "envelope_correlation": 0.985427070539843, "rms_gain_error_db": -0.12350978562850971, "reference_mean_power": 0.0017617711849040457, "output_mean_power": 0.001712373601266353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.325463642910379e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006157322295664234, "output_power": 0.0005920135354048899, "reference_active": true, "delta_db": -0.17060250412647388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015692982702212587, "output_power": 0.00013077083697546544, "reference_active": true, "delta_db": -0.7919459267004075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0206834209715196e-05, "output_power": 3.524025886238639e-05, "reference_active": true, "delta_db": -0.5726078921252689}], "active_band_count": 3, "active_band_mae_db": 0.5117187743173834}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00014319111528307625, "log_rms_mae_db": 1.1778237108590035, "rms_envelope_nrmse": 0.07534804950180254, "envelope_correlation": 0.9951786301341083, "rms_gain_error_db": -0.12350999982907068, "reference_mean_power": 0.0017724260962696937, "output_mean_power": 0.0017227296789506766}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.2006657592514347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005739716253322745, "output_power": 0.0005511399467132877, "reference_active": true, "delta_db": -0.17628533391866666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.210701544963183e-05, "output_power": 6.400981648756941e-05, "reference_active": true, "delta_db": -0.5173093813689856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.009713658025178e-06, "output_power": 2.6980828685680658e-06, "reference_active": true, "delta_db": -1.7205807545189533}], "active_band_count": 3, "active_band_mae_db": 0.8047251566022018}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.00012416177010932925, "log_rms_mae_db": 1.928149098282262, "rms_envelope_nrmse": 0.15940210164047425, "envelope_correlation": 0.9826662118306166, "rms_gain_error_db": -0.3422538060361743, "reference_mean_power": 0.0002459141129048244, "output_mean_power": 0.0002272783526149847}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.7078032497142614e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.578711086190898e-05, "output_power": 8.750523714577656e-05, "reference_active": true, "delta_db": -0.3927302812046182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5502167493014184e-05, "output_power": 1.4722589422019194e-05, "reference_active": true, "delta_db": -0.22408224092433138}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901592249182956e-06, "output_power": 5.136232626893137e-06, "reference_active": true, "delta_db": -0.6032451478468209}], "active_band_count": 3, "active_band_mae_db": 0.4066858899919235}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 9.745648796239735e-05, "log_rms_mae_db": 1.0526290024904084, "rms_envelope_nrmse": 0.0861403379462724, "envelope_correlation": 0.995388096360951, "rms_gain_error_db": -0.3422584602636328, "reference_mean_power": 0.00024740043253105237, "output_mean_power": 0.00022865179154927357}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.5717362127850378e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.047075579844295e-05, "output_power": 8.220983463824511e-05, "reference_active": true, "delta_db": -0.4158444381529669}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.5496928110188975e-06, "output_power": 7.457959154234112e-06, "reference_active": true, "delta_db": -0.053092805206309865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.070423026829593e-07, "output_power": 3.2880058819475034e-07, "reference_active": true, "delta_db": -0.9270696070820881}], "active_band_count": 3, "active_band_mae_db": 0.46533561681378827}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 3.6263821941295626e-05, "log_rms_mae_db": 2.204641102568429, "rms_envelope_nrmse": 0.3087349584432217, "envelope_correlation": 0.9351602206172964, "rms_gain_error_db": -0.293431249027675, "reference_mean_power": 2.0515613806993198e-05, "output_mean_power": 1.9175265692983743e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 4.904304308942738e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.621343716362641e-06, "output_power": 1.949537804911083e-06, "reference_active": true, "delta_db": 0.800565685662489}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.453801656353405e-07, "output_power": 8.548348899021031e-07, "reference_active": true, "delta_db": 0.5950440737676731}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.409211497093947e-07, "output_power": 1.2190653846941282e-06, "reference_active": true, "delta_db": 7.041520726537075}], "active_band_count": 3, "active_band_mae_db": 2.812376828655746}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 2.5114070553475023e-05, "log_rms_mae_db": 1.4719173539868302, "rms_envelope_nrmse": 0.2307529834724104, "envelope_correlation": 0.9615634670854488, "rms_gain_error_db": -0.2934298208822748, "reference_mean_power": 2.063955290066712e-05, "output_mean_power": 1.929111380842478e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 4.623298998188318e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6281370009575077e-06, "output_power": 2.5050595437112094e-06, "reference_active": true, "delta_db": 1.8712710707106281}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7397133003103647e-07, "output_power": 6.578328715697183e-07, "reference_active": true, "delta_db": 2.4527726223954986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7174687173869334e-08, "output_power": 1.1630699745167223e-07, "reference_active": true, "delta_db": 8.307170088462653}], "active_band_count": 3, "active_band_mae_db": 4.21040459385626}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.857423143959344, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "134db64ac1f48e62a7e2e070df8306a7caa8cfb655684f0f2f7628f8205829b5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9821451558631384, "standard_deviation": 0.0021064021334394503, "minimum": 0.9792875643481072, "maximum": 0.9841476589194355, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9096972537031056, "standard_deviation": 0.037554347933972755, "minimum": 1.8395967832710713, "maximum": 1.9640988330119753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6080262371123297, "standard_deviation": 0.2947298107345349, "minimum": 0.3405133848627144, "maximum": 1.0543802026291982, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1613124662206752, "standard_deviation": 0.009255497364157962, "minimum": 0.15313656096939293, "maximum": 0.17424936828899726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995068932799084, "standard_deviation": 0.0005586743025003053, "minimum": 0.9941114867929903, "maximum": 0.995612513104184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0941671159698105, "standard_deviation": 0.037883775670762856, "minimum": 1.036341282790597, "maximum": 1.1401434045110552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6408970924917872, "standard_deviation": 0.5055640213058707, "minimum": 0.0783814932304354, "maximum": 1.358842352286194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0860516233369191, "standard_deviation": 0.005566406902628529, "minimum": 0.0781838844894025, "maximum": 0.0972986982466042, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.938518037660376, "standard_deviation": 0.0028779990230435887, "minimum": 0.9351602206172964, "maximum": 0.9431276961764994, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1328606153903866, "standard_deviation": 0.033878792226659855, "minimum": 2.097413977114052, "maximum": 2.204641102568429, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.936203190973773, "standard_deviation": 0.08425438225280794, "minimum": 2.812376828655746, "maximum": 3.0527308896381107, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30329237925663577, "standard_deviation": 0.005640561206761794, "minimum": 0.29089264399788345, "maximum": 0.3087349584432217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9618738850838783, "standard_deviation": 0.002162081790646506, "minimum": 0.9593361818928188, "maximum": 0.9657254446235873, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4174125149973495, "standard_deviation": 0.03231828452943013, "minimum": 1.3555042724926225, "maximum": 1.4719173539868302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.310153529645675, "standard_deviation": 0.08497531598554818, "minimum": 4.21040459385626, "maximum": 4.447569433234402, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23132442139491888, "standard_deviation": 0.00466908845976957, "minimum": 0.22411678146852115, "maximum": 0.23731610286253577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9829569784384209, "standard_deviation": 0.0018002492458690923, "minimum": 0.9805679710633552, "maximum": 0.985427070539843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9767356768360518, "standard_deviation": 0.05246920287637462, "minimum": 1.8883048763756007, "maximum": 2.0351395209633267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6003075591053907, "standard_deviation": 0.2257014698453708, "minimum": 0.22400714651122108, "maximum": 0.9310981944381922, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14588482499955785, "standard_deviation": 0.007673897532791429, "minimum": 0.13517220331357913, "maximum": 0.155447857009973, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936761197733059, "standard_deviation": 0.001200753095139315, "minimum": 0.9912131868998424, "maximum": 0.9951786301341083, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1591423960431502, "standard_deviation": 0.058371893265247964, "minimum": 1.0847238218048887, "maximum": 1.242232941281232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7990034890477309, "standard_deviation": 0.24278307128219123, "minimum": 0.42466445582283, "maximum": 1.2736325949357055, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08547211789756723, "standard_deviation": 0.007410885800937765, "minimum": 0.07534804950180254, "maximum": 0.09996488256938366, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17385461964626722, "standard_deviation": 0.046934971953422125, "minimum": 0.10677524839921089, "maximum": 0.2305637620507277, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.004054109206443, "standard_deviation": 0.011848945546955758, "minimum": 5.983297005497754, "maximum": 6.019194973993751, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46424437491808546, "standard_deviation": 0.01860130458426896, "minimum": 0.44610990023965025, "maximum": 0.4927829679606267, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.060080265445670016, "standard_deviation": 0.036850968383380506, "minimum": -0.11273382201333766, "maximum": 0.0008444260083275868, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9915749235567546, "standard_deviation": 0.0013634761554700188, "minimum": 0.988986824534817, "maximum": 0.9930971683316379, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3285477314266714, "standard_deviation": 0.025345521998586112, "minimum": 1.2873975099885442, "maximum": 1.3751246099873153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27939051333129744, "standard_deviation": 0.10387280953889455, "minimum": 0.13286616960577244, "maximum": 0.3993437551858608, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09236219519083877, "standard_deviation": 0.007319634096906762, "minimum": 0.08375060732612079, "maximum": 0.1057473941436117, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976756110386538, "standard_deviation": 0.0006268469737894344, "minimum": 0.9963288950132252, "maximum": 0.9983401970834672, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6823637227233561, "standard_deviation": 0.03652315257016196, "minimum": 0.6250399404942557, "maximum": 0.7359611962714921, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.41443278330350797, "standard_deviation": 0.19882898465557372, "minimum": 0.09565578354820166, "maximum": 0.6433932758898403, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04682236282963393, "standard_deviation": 0.005875532253785517, "minimum": 0.03976557048426788, "maximum": 0.05866094980643491, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123440-0020", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.5955967903137207, "predicted_mos": 1.5955967903137207, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "96ff6b7526de34f216dd64ef2ab6d994c404d84d025e60bc08648c6dcf22699c", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.021377900149673, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.274788081026821, "全帯域_10ms_rms_envelope_nrmse": 0.09536234161803507, "全帯域_10ms_envelope_correlation": 0.991059739238193, "全帯域_10ms_modulation_mae_db": 0.2490286995400779, "全帯域_40ms_log_rms_mae_db": 0.6515460826331889, "全帯域_40ms_rms_envelope_nrmse": 0.047358938177530445, "全帯域_40ms_envelope_correlation": 0.9976476774519126, "全帯域_40ms_modulation_mae_db": 0.13375980550850727, "300–1000 Hz_10ms_log_rms_mae_db": 2.08088327145862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1529221185457651, "300–1000 Hz_10ms_envelope_correlation": 0.9812337636701823, "300–1000 Hz_10ms_modulation_mae_db": 0.2982018466769641, "300–1000 Hz_40ms_log_rms_mae_db": 1.2457929729640933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08046188196251772, "300–1000 Hz_40ms_envelope_correlation": 0.9943216680253004, "300–1000 Hz_40ms_modulation_mae_db": 0.6262484610366471, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4004926341388844, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20233433087788227, "1000–3000 Hz_10ms_envelope_correlation": 0.9715013063460426, "1000–3000 Hz_10ms_modulation_mae_db": 0.9418875165784245, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4680104117064154, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10564231587180822, "1000–3000 Hz_40ms_envelope_correlation": 0.9917988352439474, "1000–3000 Hz_40ms_modulation_mae_db": 0.8285089844266785, "3000–6000 Hz_10ms_log_rms_mae_db": 2.888175594665606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3350260923973704, "3000–6000 Hz_10ms_envelope_correlation": 0.9294882384628411, "3000–6000 Hz_10ms_modulation_mae_db": 3.2259450641399696, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0420365904159103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2496479103718468, "3000–6000 Hz_40ms_envelope_correlation": 0.9586623238173324, "3000–6000 Hz_40ms_modulation_mae_db": 4.49107368678455, "mr_stft_db": 6.47038622414573, "band_mae_db": 0.15944542080187019, "rms_gain_error_db": -0.053430439628243565, "output_peak": 0.4821590070619358}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3320821953332307, "全帯域_10ms_rms_envelope_nrmse": 0.09392021324441109, "全帯域_10ms_envelope_correlation": 0.9913447008805396, "全帯域_10ms_modulation_mae_db": 0.33242025463295355, "全帯域_40ms_log_rms_mae_db": 0.676519567604611, "全帯域_40ms_rms_envelope_nrmse": 0.04924769832733546, "全帯域_40ms_envelope_correlation": 0.997482524697382, "全帯域_40ms_modulation_mae_db": 0.5764078324759033, "300–1000 Hz_10ms_log_rms_mae_db": 2.026261614501478, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15326273393792905, "300–1000 Hz_10ms_envelope_correlation": 0.9811187424447488, "300–1000 Hz_10ms_modulation_mae_db": 0.4342140515359593, "300–1000 Hz_40ms_log_rms_mae_db": 1.1898810046275712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08814438127159242, "300–1000 Hz_40ms_envelope_correlation": 0.9932396503476633, "300–1000 Hz_40ms_modulation_mae_db": 0.8826870844051462, "1000–3000 Hz_10ms_log_rms_mae_db": 2.455610591992841, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22363102134228377, "1000–3000 Hz_10ms_envelope_correlation": 0.965058355599366, "1000–3000 Hz_10ms_modulation_mae_db": 0.5132550424601414, "1000–3000 Hz_40ms_log_rms_mae_db": 1.509914135501496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13227035901683862, "1000–3000 Hz_40ms_envelope_correlation": 0.9870332246236073, "1000–3000 Hz_40ms_modulation_mae_db": 0.6514982532618291, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9832875292902203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36016264522605945, "3000–6000 Hz_10ms_envelope_correlation": 0.9182827661632433, "3000–6000 Hz_10ms_modulation_mae_db": 3.5104569850253218, "3000–6000 Hz_40ms_log_rms_mae_db": 2.15379812558376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2725496597194777, "3000–6000 Hz_40ms_envelope_correlation": 0.9504815494412184, "3000–6000 Hz_40ms_modulation_mae_db": 4.573349780329511, "mr_stft_db": 6.495183163643859, "band_mae_db": 0.2285779503511006, "rms_gain_error_db": -0.06845049316742677, "output_peak": 0.4576996379087453}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3217826147138008, "全帯域_10ms_rms_envelope_nrmse": 0.087901554055288, "全帯域_10ms_envelope_correlation": 0.9924563432203133, "全帯域_10ms_modulation_mae_db": 0.3363313098555931, "全帯域_40ms_log_rms_mae_db": 0.6753320342251603, "全帯域_40ms_rms_envelope_nrmse": 0.05097589419562684, "全帯域_40ms_envelope_correlation": 0.9973462816498566, "全帯域_40ms_modulation_mae_db": 0.389905108437331, "300–1000 Hz_10ms_log_rms_mae_db": 1.986594124818849, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14641919092790712, "300–1000 Hz_10ms_envelope_correlation": 0.982948322482094, "300–1000 Hz_10ms_modulation_mae_db": 0.4853175022216935, "300–1000 Hz_40ms_log_rms_mae_db": 1.2657551871080241, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09899512651577107, "300–1000 Hz_40ms_envelope_correlation": 0.9916597745324925, "300–1000 Hz_40ms_modulation_mae_db": 0.5627270226265823, "1000–3000 Hz_10ms_log_rms_mae_db": 2.361269867669897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19390593708198658, "1000–3000 Hz_10ms_envelope_correlation": 0.973704819355291, "1000–3000 Hz_10ms_modulation_mae_db": 0.1278178066761076, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4256689868914392, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11768075566458744, "1000–3000 Hz_40ms_envelope_correlation": 0.9896958117983994, "1000–3000 Hz_40ms_modulation_mae_db": 0.2274977573217655, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8998488693654094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34708984258012976, "3000–6000 Hz_10ms_envelope_correlation": 0.9249945489476649, "3000–6000 Hz_10ms_modulation_mae_db": 3.2352222969775455, "3000–6000 Hz_40ms_log_rms_mae_db": 2.136773324915134, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2551576099677012, "3000–6000 Hz_40ms_envelope_correlation": 0.95750797498292, "3000–6000 Hz_40ms_modulation_mae_db": 4.503084176912226, "mr_stft_db": 6.470125554115561, "band_mae_db": 0.22750498372852115, "rms_gain_error_db": -0.10806488804597539, "output_peak": 0.4679401092277686}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3069669022174548, "全帯域_10ms_rms_envelope_nrmse": 0.08653391368999984, "全帯域_10ms_envelope_correlation": 0.9926851348940567, "全帯域_10ms_modulation_mae_db": 0.19192723824185035, "全帯域_40ms_log_rms_mae_db": 0.6739243639785772, "全帯域_40ms_rms_envelope_nrmse": 0.04568086777556999, "全帯域_40ms_envelope_correlation": 0.9978868313031675, "全帯域_40ms_modulation_mae_db": 0.21575967726180045, "300–1000 Hz_10ms_log_rms_mae_db": 2.000889020304532, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15485326187824489, "300–1000 Hz_10ms_envelope_correlation": 0.9807551282022566, "300–1000 Hz_10ms_modulation_mae_db": 0.2764900402752661, "300–1000 Hz_40ms_log_rms_mae_db": 1.1787859720456526, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08974189553525129, "300–1000 Hz_40ms_envelope_correlation": 0.9930998122233253, "300–1000 Hz_40ms_modulation_mae_db": 0.33943671162825656, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3390667622813464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20549920835328286, "1000–3000 Hz_10ms_envelope_correlation": 0.9703929875416726, "1000–3000 Hz_10ms_modulation_mae_db": 0.30832074250487157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4285805954969135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12469475327935571, "1000–3000 Hz_40ms_envelope_correlation": 0.9884301166804051, "1000–3000 Hz_40ms_modulation_mae_db": 0.15535181097999187, "3000–6000 Hz_10ms_log_rms_mae_db": 2.933848543810635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3471787351188515, "3000–6000 Hz_10ms_envelope_correlation": 0.9264424703377124, "3000–6000 Hz_10ms_modulation_mae_db": 3.4467287886108466, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1712451996103814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2505809460380646, "3000–6000 Hz_40ms_envelope_correlation": 0.960122262061167, "3000–6000 Hz_40ms_modulation_mae_db": 4.606264480178687, "mr_stft_db": 6.470827243486631, "band_mae_db": 0.240982299704232, "rms_gain_error_db": -0.10035563596015096, "output_peak": 0.43755862895805625}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3267628996771634, "全帯域_10ms_rms_envelope_nrmse": 0.09200132588005129, "全帯域_10ms_envelope_correlation": 0.9917878602740081, "全帯域_10ms_modulation_mae_db": 0.650508376080151, "全帯域_40ms_log_rms_mae_db": 0.6671262525806254, "全帯域_40ms_rms_envelope_nrmse": 0.04805577915119452, "全帯域_40ms_envelope_correlation": 0.9977578411160833, "全帯域_40ms_modulation_mae_db": 0.8470986352543681, "300–1000 Hz_10ms_log_rms_mae_db": 2.0851897154148307, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15366557061434175, "300–1000 Hz_10ms_envelope_correlation": 0.9814393094218976, "300–1000 Hz_10ms_modulation_mae_db": 1.5792818440633127, "300–1000 Hz_40ms_log_rms_mae_db": 1.2466350974473523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09817724484317693, "300–1000 Hz_40ms_envelope_correlation": 0.9922566881319926, "300–1000 Hz_40ms_modulation_mae_db": 1.9312557789609182, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3966884624868188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19902824863628818, "1000–3000 Hz_10ms_envelope_correlation": 0.9722915960661094, "1000–3000 Hz_10ms_modulation_mae_db": 0.7288913735834157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4544046498098118, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11295760660320828, "1000–3000 Hz_40ms_envelope_correlation": 0.9905064078005907, "1000–3000 Hz_40ms_modulation_mae_db": 0.9162697678504294, "3000–6000 Hz_10ms_log_rms_mae_db": 2.910640004431042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.364173230047225, "3000–6000 Hz_10ms_envelope_correlation": 0.9232570774358357, "3000–6000 Hz_10ms_modulation_mae_db": 3.5095449842296156, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1538947305603617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2730548391744277, "3000–6000 Hz_40ms_envelope_correlation": 0.9558299160580579, "3000–6000 Hz_40ms_modulation_mae_db": 4.692840646150306, "mr_stft_db": 6.4774133478891995, "band_mae_db": 0.29732800759715916, "rms_gain_error_db": -0.13833069890496905, "output_peak": 0.4729870918698867}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3443679881425579, "全帯域_10ms_rms_envelope_nrmse": 0.10377698943412592, "全帯域_10ms_envelope_correlation": 0.9893846971332123, "全帯域_10ms_modulation_mae_db": 0.16495006132367715, "全帯域_40ms_log_rms_mae_db": 0.6984072374928709, "全帯域_40ms_rms_envelope_nrmse": 0.056076910783410504, "全帯域_40ms_envelope_correlation": 0.9966269524186382, "全帯域_40ms_modulation_mae_db": 0.16394408327825744, "300–1000 Hz_10ms_log_rms_mae_db": 2.0830363802362384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16377059984551734, "300–1000 Hz_10ms_envelope_correlation": 0.9785526290586364, "300–1000 Hz_10ms_modulation_mae_db": 0.31256380816206947, "300–1000 Hz_40ms_log_rms_mae_db": 1.2987781599063488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10865066851750085, "300–1000 Hz_40ms_envelope_correlation": 0.9896305106733485, "300–1000 Hz_40ms_modulation_mae_db": 0.391264298310805, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3345720864837287, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17208237615230387, "1000–3000 Hz_10ms_envelope_correlation": 0.9797604377922434, "1000–3000 Hz_10ms_modulation_mae_db": 0.5482006886919576, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4085017136189584, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09269628245066032, "1000–3000 Hz_40ms_envelope_correlation": 0.9937963364225999, "1000–3000 Hz_40ms_modulation_mae_db": 0.8855879543324391, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9250119054349377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34745690635180043, "3000–6000 Hz_10ms_envelope_correlation": 0.9236050448960038, "3000–6000 Hz_10ms_modulation_mae_db": 3.342194598668551, "3000–6000 Hz_40ms_log_rms_mae_db": 2.188238596854339, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2611848099356147, "3000–6000 Hz_40ms_envelope_correlation": 0.9543247506018657, "3000–6000 Hz_40ms_modulation_mae_db": 4.505307276365788, "mr_stft_db": 6.470546811257812, "band_mae_db": 0.16138594855210556, "rms_gain_error_db": -0.015896636116992845, "output_peak": 0.45966721717445574}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.298774159393251, "全帯域_10ms_rms_envelope_nrmse": 0.08679368555948379, "全帯域_10ms_envelope_correlation": 0.9926610048309857, "全帯域_10ms_modulation_mae_db": 0.1280152931808669, "全帯域_40ms_log_rms_mae_db": 0.7487517609877534, "全帯域_40ms_rms_envelope_nrmse": 0.04794931404451515, "全帯域_40ms_envelope_correlation": 0.9976908350298016, "全帯域_40ms_modulation_mae_db": 0.16256477175356593, "300–1000 Hz_10ms_log_rms_mae_db": 2.0748475021675485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14580524433459485, "300–1000 Hz_10ms_envelope_correlation": 0.9831289289474345, "300–1000 Hz_10ms_modulation_mae_db": 0.40057389181486985, "300–1000 Hz_40ms_log_rms_mae_db": 1.3128324449367739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08597594191971229, "300–1000 Hz_40ms_envelope_correlation": 0.9938691602042603, "300–1000 Hz_40ms_modulation_mae_db": 0.4756729695675836, "1000–3000 Hz_10ms_log_rms_mae_db": 2.363861959478457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18668758257392667, "1000–3000 Hz_10ms_envelope_correlation": 0.9761385994590123, "1000–3000 Hz_10ms_modulation_mae_db": 1.1657801527126617, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4392518786064332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11631192573195408, "1000–3000 Hz_40ms_envelope_correlation": 0.9910802797220993, "1000–3000 Hz_40ms_modulation_mae_db": 1.2205228294977548, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8782004166226414, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3457138676579115, "3000–6000 Hz_10ms_envelope_correlation": 0.9290927758486931, "3000–6000 Hz_10ms_modulation_mae_db": 3.2268602127717707, "3000–6000 Hz_40ms_log_rms_mae_db": 2.086135985421863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2534032315251775, "3000–6000 Hz_40ms_envelope_correlation": 0.9608079933025443, "3000–6000 Hz_40ms_modulation_mae_db": 4.442174056906097, "mr_stft_db": 6.475195326352366, "band_mae_db": 0.2736200775047339, "rms_gain_error_db": -0.1132706141264913, "output_peak": 0.49079852036050736}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3144200036334042, "全帯域_10ms_rms_envelope_nrmse": 0.09018610337954105, "全帯域_10ms_envelope_correlation": 0.991989777742459, "全帯域_10ms_modulation_mae_db": 0.4463789038964059, "全帯域_40ms_log_rms_mae_db": 0.6798792123990525, "全帯域_40ms_rms_envelope_nrmse": 0.05009726601468175, "全帯域_40ms_envelope_correlation": 0.9973326164736657, "全帯域_40ms_modulation_mae_db": 0.4499517296919522, "300–1000 Hz_10ms_log_rms_mae_db": 2.120505091103913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1426679784801213, "300–1000 Hz_10ms_envelope_correlation": 0.9836835728527515, "300–1000 Hz_10ms_modulation_mae_db": 1.1758798690329604, "300–1000 Hz_40ms_log_rms_mae_db": 1.3057436543670942, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07897998562965405, "300–1000 Hz_40ms_envelope_correlation": 0.9945805326556086, "300–1000 Hz_40ms_modulation_mae_db": 1.4069565283812207, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3394929281897126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18719071499493076, "1000–3000 Hz_10ms_envelope_correlation": 0.9758948273924336, "1000–3000 Hz_10ms_modulation_mae_db": 0.5085646402562057, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3997160785238993, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11241372249598114, "1000–3000 Hz_40ms_envelope_correlation": 0.9907278794759943, "1000–3000 Hz_40ms_modulation_mae_db": 0.8088461107389495, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8873790116814666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3569364185071803, "3000–6000 Hz_10ms_envelope_correlation": 0.916896264553144, "3000–6000 Hz_10ms_modulation_mae_db": 3.3903984343936924, "3000–6000 Hz_40ms_log_rms_mae_db": 2.114052587924562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2619882477497553, "3000–6000 Hz_40ms_envelope_correlation": 0.9525339621511584, "3000–6000 Hz_40ms_modulation_mae_db": 4.604520994223602, "mr_stft_db": 6.465445240636375, "band_mae_db": 0.13530020187763495, "rms_gain_error_db": -0.04703314611535089, "output_peak": 0.45280530639416094}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79920, "output_samples": 79920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.674859980409335, "spectral_convergence": 0.27481058723487134, "spectral_convergence_status": "ok", "amplitude_floor": 1.5443736044991425e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.380748818286704, "spectral_convergence": 0.25736578584660874, "spectral_convergence_status": "ok", "amplitude_floor": 1.1257392826698756e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.355549873741152, "spectral_convergence": 0.26217694034912153, "spectral_convergence_status": "ok", "amplitude_floor": 1.1201855379193364e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001035002994445706, "output_power": 0.0010503037849399046, "delta_db": 0.06373324252986803, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0032595449236567883, "output_power": 0.0031755523963500083, "delta_db": -0.11337687988112619, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007961593328208376, "output_power": 0.0007998246275805736, "delta_db": 0.019947820614168797, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023616410570225164, "output_power": 0.0002311940988784346, "delta_db": -0.09237145672594671, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.38734826098075e-05, "output_power": 5.635702382412927e-05, "delta_db": 0.19572999162566063, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3554110211771778e-05, "output_power": 1.5108533992028647e-05, "delta_db": 0.4715131334344507, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15944542080187019, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.053430439628243565, "pitch": null, "output_rms_dbfs": -22.734080327881863, "output_peak": 0.4821590070619358, "reference_rms_dbfs": -22.68064988825362, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.47038622414573}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001944445221026879, "log_rms_mae_db": 1.274788081026821, "rms_envelope_nrmse": 0.09536234161803507, "envelope_correlation": 0.991059739238193, "rms_gain_error_db": -0.05342994405075279, "reference_mean_power": 0.005399702914122853, "output_mean_power": 0.005333678962200825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 2.8891675057826967e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000788720956533661, "output_power": 0.0008086734164103378, "reference_active": true, "delta_db": 0.10849786442341178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001646344464707057, "output_power": 0.00015173668072504968, "reference_active": true, "delta_db": -0.3543012829273663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.1066196116820634e-05, "output_power": 5.452079801276864e-05, "reference_active": true, "delta_db": 0.28428695126945563}], "active_band_count": 3, "active_band_mae_db": 0.2490286995400779}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00017595749087279735, "log_rms_mae_db": 0.6515460826331889, "rms_envelope_nrmse": 0.047358938177530445, "envelope_correlation": 0.9976476774519126, "rms_gain_error_db": -0.05342954823921185, "reference_mean_power": 0.005432355653378944, "output_mean_power": 0.005365932934669183}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 2.6677500098154174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007158752148089715, "output_power": 0.0007282458483366037, "reference_active": true, "delta_db": 0.07440690995487664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.294187442575086e-05, "output_power": 7.027292131300112e-05, "reference_active": true, "delta_db": -0.1618891146285416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7590528739181784e-06, "output_power": 3.618929270835703e-06, "reference_active": true, "delta_db": -0.1649833919421036}], "active_band_count": 3, "active_band_mae_db": 0.13375980550850727}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001659932762873133, "log_rms_mae_db": 2.08088327145862, "rms_envelope_nrmse": 0.1529221185457651, "envelope_correlation": 0.9812337636701823, "rms_gain_error_db": -0.04550472294294497, "reference_mean_power": 0.0017617711849040457, "output_mean_power": 0.001743407982710633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.325463642910379e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006157322295664234, "output_power": 0.0006193586906491517, "reference_active": true, "delta_db": 0.02550349188562226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015692982702212587, "output_power": 0.00014032254675067673, "reference_active": true, "delta_db": -0.48578037822945835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0206834209715196e-05, "output_power": 3.681016059150733e-05, "reference_active": true, "delta_db": -0.38332166991581174}], "active_band_count": 3, "active_band_mae_db": 0.2982018466769641}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00014319111528307625, "log_rms_mae_db": 1.2457929729640933, "rms_envelope_nrmse": 0.08046188196251772, "envelope_correlation": 0.9943216680253004, "rms_gain_error_db": -0.04550597585946457, "reference_mean_power": 0.0017724260962696937, "output_mean_power": 0.0017539513303501065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.2006657592514347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005739716253322745, "output_power": 0.000581309737536826, "reference_active": true, "delta_db": 0.05517174604118331}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.210701544963183e-05, "output_power": 6.431113817829414e-05, "reference_active": true, "delta_db": -0.496913244444128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.009713658025178e-06, "output_power": 2.9542505196624153e-06, "reference_active": true, "delta_db": -1.32666039262463}], "active_band_count": 3, "active_band_mae_db": 0.6262484610366471}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.00012416177010932925, "log_rms_mae_db": 2.4004926341388844, "rms_envelope_nrmse": 0.20233433087788227, "envelope_correlation": 0.9715013063460426, "rms_gain_error_db": 0.000633246100176179, "reference_mean_power": 0.0002459141129048244, "output_mean_power": 0.0002459499723304237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.7078032497142614e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.578711086190898e-05, "output_power": 9.43743213955105e-05, "reference_active": true, "delta_db": -0.06453232406900253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5502167493014184e-05, "output_power": 1.786146614901274e-05, "reference_active": true, "delta_db": 0.6152467994016384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901592249182956e-06, "output_power": 9.672918813952427e-06, "reference_active": true, "delta_db": 2.1458834262646325}], "active_band_count": 3, "active_band_mae_db": 0.9418875165784245}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 9.745648796239735e-05, "log_rms_mae_db": 1.4680104117064154, "rms_envelope_nrmse": 0.10564231587180822, "envelope_correlation": 0.9917988352439474, "rms_gain_error_db": 0.0006332201613751129, "reference_mean_power": 0.00024740043253105237, "output_mean_power": 0.000247436507215321}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.5717362127850378e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.047075579844295e-05, "output_power": 9.072898578268095e-05, "reference_active": true, "delta_db": 0.012378376719534463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.5496928110188975e-06, "output_power": 9.397777357576853e-06, "reference_active": true, "delta_db": 0.9509587094550589}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.070423026829593e-07, "output_power": 5.779077694461782e-07, "reference_active": true, "delta_db": 1.5221898671054421}], "active_band_count": 3, "active_band_mae_db": 0.8285089844266785}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 3.6263821941295626e-05, "log_rms_mae_db": 2.888175594665606, "rms_envelope_nrmse": 0.3350260923973704, "envelope_correlation": 0.9294882384628411, "rms_gain_error_db": 0.3636065808076811, "reference_mean_power": 2.0515613806993198e-05, "output_mean_power": 2.2307205762324668e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 4.904304308942738e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.621343716362641e-06, "output_power": 2.2077832151874607e-06, "reference_active": true, "delta_db": 1.3408133450740345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.453801656353405e-07, "output_power": 9.623722851545075e-07, "reference_active": true, "delta_db": 1.1096527538418022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.409211497093947e-07, "output_power": 1.2723653582090662e-06, "reference_active": true, "delta_db": 7.227369093504072}], "active_band_count": 3, "active_band_mae_db": 3.2259450641399696}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 2.5114070553475023e-05, "log_rms_mae_db": 2.0420365904159103, "rms_envelope_nrmse": 0.2496479103718468, "envelope_correlation": 0.9586623238173324, "rms_gain_error_db": 0.36360094647884433, "reference_mean_power": 2.063955290066712e-05, "output_mean_power": 2.2441939120833325e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 4.623298998188318e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6281370009575077e-06, "output_power": 2.751265495331774e-06, "reference_active": true, "delta_db": 2.278415553818102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7397133003103647e-07, "output_power": 7.040748558267618e-07, "reference_active": true, "delta_db": 2.747805259943719}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7174687173869334e-08, "output_power": 1.2011268078523797e-07, "reference_active": true, "delta_db": 8.447000246591829}], "active_band_count": 3, "active_band_mae_db": 4.49107368678455}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8247360647520995, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5e621e2cfee52d2b79dd344c554fe1c080e2aa4b08ce26f15bad33f8b97c4c5c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9730928661940214, "standard_deviation": 0.004423577691812443, "minimum": 0.965058355599366, "maximum": 0.9797604377922434, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.373881911590211, "standard_deviation": 0.041588578036192946, "minimum": 2.3345720864837287, "maximum": 2.455610591992841, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6053397454329732, "standard_deviation": 0.3338049542978998, "minimum": 0.1278178066761076, "maximum": 1.1657801527126617, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19629492750161062, "standard_deviation": 0.015320709280593736, "minimum": 0.17208237615230387, "maximum": 0.22363102134228377, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9903836114709554, "standard_deviation": 0.0020626888930434665, "minimum": 0.9870332246236073, "maximum": 0.9937963364225999, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.441756056269421, "standard_deviation": 0.03546964571500114, "minimum": 1.3997160785238993, "maximum": 1.509914135501496, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7117604335512298, "standard_deviation": 0.3589006086859435, "minimum": 0.15535181097999187, "maximum": 1.2205228294977548, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11433346513929923, "standard_deviation": 0.011913313458312154, "minimum": 0.09269628245066032, "maximum": 0.13227035901683862, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9240073983306423, "standard_deviation": 0.004580536803556782, "minimum": 0.916896264553144, "maximum": 0.9294882384628411, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.913298984412745, "standard_deviation": 0.0341860210641905, "minimum": 2.8782004166226414, "maximum": 2.9832875292902203, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.360918920602164, "standard_deviation": 0.12248827067916258, "minimum": 3.2259450641399696, "maximum": 3.5104569850253218, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.35046721723581603, "standard_deviation": 0.009380923579972902, "minimum": 0.3350260923973704, "maximum": 0.364173230047225, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9562838415520329, "standard_deviation": 0.003667022470128409, "minimum": 0.9504815494412184, "maximum": 0.9608079933025443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.130771892660789, "standard_deviation": 0.048088094934425195, "minimum": 2.0420365904159103, "maximum": 2.188238596854339, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.552326887231346, "standard_deviation": 0.08136413402246921, "minimum": 4.442174056906097, "maximum": 4.692840646150306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2596959068102582, "standard_deviation": 0.009222568127316035, "minimum": 0.2496479103718468, "maximum": 0.2730548391744277, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9816075496350002, "standard_deviation": 0.0016426570137338219, "minimum": 0.9785526290586364, "maximum": 0.9836835728527515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0572758400007514, "standard_deviation": 0.046971211061510615, "minimum": 1.986594124818849, "maximum": 2.120505091103913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.620315356722887, "standard_deviation": 0.48492960554209485, "minimum": 0.2764900402752661, "maximum": 1.5792818440633127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15167083732055267, "standard_deviation": 0.006629450719493108, "minimum": 0.1426679784801213, "maximum": 0.16377059984551734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9928322245992489, "standard_deviation": 0.0016276499515205532, "minimum": 0.9896305106733485, "maximum": 0.9945805326556086, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.255525561675364, "standard_deviation": 0.050924740432335636, "minimum": 1.1787859720456526, "maximum": 1.3128324449367739, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.827031106864645, "standard_deviation": 0.563152741901964, "minimum": 0.33943671162825656, "maximum": 1.9312557789609182, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09114089077439708, "standard_deviation": 0.010123052976603868, "minimum": 0.07897998562965405, "maximum": 0.10865066851750085, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2155181112646697, "standard_deviation": 0.05799103091034342, "minimum": 0.13530020187763495, "maximum": 0.29732800759715916, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.474390363940941, "standard_deviation": 0.009132615098809126, "minimum": 6.465445240636375, "maximum": 6.495183163643859, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4652019398694396, "standard_deviation": 0.016968795198393314, "minimum": 0.43755862895805625, "maximum": 0.49079852036050736, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0806040690082001, "standard_deviation": 0.040959670449167984, "minimum": -0.13833069890496905, "maximum": -0.015896636116992845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.991671157276721, "standard_deviation": 0.0010985646564116865, "minimum": 0.9893846971332123, "maximum": 0.9926851348940567, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3149931055172106, "standard_deviation": 0.021659022542626877, "minimum": 1.274788081026821, "maximum": 1.3443679881425579, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31244501709394695, "standard_deviation": 0.172175133102294, "minimum": 0.1280152931808669, "maximum": 0.650508376080151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09205951585761701, "standard_deviation": 0.005741241709455345, "minimum": 0.08653391368999984, "maximum": 0.10377698943412592, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974714450175635, "standard_deviation": 0.00039315356542827027, "minimum": 0.9966269524186382, "maximum": 0.9978868313031675, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.68393581398773, "standard_deviation": 0.02924745659283045, "minimum": 0.6515460826331889, "maximum": 0.7487517609877534, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3674239554577107, "standard_deviation": 0.25129681563362405, "minimum": 0.13375980550850727, "maximum": 0.8470986352543681, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04943033355873308, "standard_deviation": 0.003150417027261581, "minimum": 0.04568086777556999, "maximum": 0.056076910783410504, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123440-0020", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.9838910102844238, "predicted_mos": 1.9838910102844238, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "700165a43acf07e5d948698ef68a1bd544b68cf2aa3c36f95f510759edc33c87", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.532686499878764, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3108609504018915, "全帯域_10ms_rms_envelope_nrmse": 0.0865968157332224, "全帯域_10ms_envelope_correlation": 0.9926347940925041, "全帯域_10ms_modulation_mae_db": 0.35626830793742653, "全帯域_40ms_log_rms_mae_db": 0.646793474810485, "全帯域_40ms_rms_envelope_nrmse": 0.04433272910668952, "全帯域_40ms_envelope_correlation": 0.997959760198665, "全帯域_40ms_modulation_mae_db": 0.28665871878322674, "300–1000 Hz_10ms_log_rms_mae_db": 1.8875456108185558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14014214527360744, "300–1000 Hz_10ms_envelope_correlation": 0.9843414607096128, "300–1000 Hz_10ms_modulation_mae_db": 0.7334401466123143, "300–1000 Hz_40ms_log_rms_mae_db": 1.1134143754524213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08234122599580267, "300–1000 Hz_40ms_envelope_correlation": 0.994268842555585, "300–1000 Hz_40ms_modulation_mae_db": 0.8190518158853374, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8000036019082628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15243654834148876, "1000–3000 Hz_10ms_envelope_correlation": 0.9837480253913637, "1000–3000 Hz_10ms_modulation_mae_db": 0.41037317759729786, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0757033629673285, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08279193431971629, "1000–3000 Hz_40ms_envelope_correlation": 0.9949241638721025, "1000–3000 Hz_40ms_modulation_mae_db": 0.6583147306077523, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9802028344135154, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2973168057869667, "3000–6000 Hz_10ms_envelope_correlation": 0.9414570619143926, "3000–6000 Hz_10ms_modulation_mae_db": 3.101502537101315, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3696087476989, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22780876624131788, "3000–6000 Hz_40ms_envelope_correlation": 0.9630811105255704, "3000–6000 Hz_40ms_modulation_mae_db": 4.322526440410997, "mr_stft_db": 5.947894571450043, "band_mae_db": 0.12558558242083945, "rms_gain_error_db": -0.07707469710600927, "output_peak": 0.47053365331467517}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3776273883899073, "全帯域_10ms_rms_envelope_nrmse": 0.09821760015441543, "全帯域_10ms_envelope_correlation": 0.990497519038596, "全帯域_10ms_modulation_mae_db": 0.07209809817748856, "全帯域_40ms_log_rms_mae_db": 0.6951665235614947, "全帯域_40ms_rms_envelope_nrmse": 0.04643286618970012, "全帯域_40ms_envelope_correlation": 0.9976910503876422, "全帯域_40ms_modulation_mae_db": 0.16060910107314644, "300–1000 Hz_10ms_log_rms_mae_db": 1.8176561367432602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13450256732865523, "300–1000 Hz_10ms_envelope_correlation": 0.9855369395019928, "300–1000 Hz_10ms_modulation_mae_db": 0.5245588092650894, "300–1000 Hz_40ms_log_rms_mae_db": 1.13970597840789, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07355979355020031, "300–1000 Hz_40ms_envelope_correlation": 0.9953645414877458, "300–1000 Hz_40ms_modulation_mae_db": 0.7105225451718203, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7930546105708007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16693651307095106, "1000–3000 Hz_10ms_envelope_correlation": 0.980445915826794, "1000–3000 Hz_10ms_modulation_mae_db": 0.2830980495610166, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0731649980558151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10688981344855138, "1000–3000 Hz_40ms_envelope_correlation": 0.991436557799121, "1000–3000 Hz_40ms_modulation_mae_db": 0.3164631631768953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9306096195347384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29904384055255867, "3000–6000 Hz_10ms_envelope_correlation": 0.9404737025019835, "3000–6000 Hz_10ms_modulation_mae_db": 3.183465274212217, "3000–6000 Hz_40ms_log_rms_mae_db": 1.337236280020344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22211770176730825, "3000–6000 Hz_40ms_envelope_correlation": 0.9648019216235026, "3000–6000 Hz_40ms_modulation_mae_db": 4.466712251122277, "mr_stft_db": 5.964132917207497, "band_mae_db": 0.10699235206036563, "rms_gain_error_db": -0.02175279536172551, "output_peak": 0.5268438556996099}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2947083840482199, "全帯域_10ms_rms_envelope_nrmse": 0.09205574092385643, "全帯域_10ms_envelope_correlation": 0.991649156406789, "全帯域_10ms_modulation_mae_db": 0.6023727968983245, "全帯域_40ms_log_rms_mae_db": 0.6533997147815668, "全帯域_40ms_rms_envelope_nrmse": 0.04641024045702429, "全帯域_40ms_envelope_correlation": 0.9976985765687869, "全帯域_40ms_modulation_mae_db": 0.6670643803030941, "300–1000 Hz_10ms_log_rms_mae_db": 1.8594473798988054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1515867551744454, "300–1000 Hz_10ms_envelope_correlation": 0.9817374453443021, "300–1000 Hz_10ms_modulation_mae_db": 0.8543832265379637, "300–1000 Hz_40ms_log_rms_mae_db": 1.1183919668004036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08703103663387564, "300–1000 Hz_40ms_envelope_correlation": 0.9933881820552898, "300–1000 Hz_40ms_modulation_mae_db": 0.7698774374750572, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7377428860633144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1457965680571644, "1000–3000 Hz_10ms_envelope_correlation": 0.9852019632072824, "1000–3000 Hz_10ms_modulation_mae_db": 0.1867706912672141, "1000–3000 Hz_40ms_log_rms_mae_db": 1.013707738983503, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07489897349203867, "1000–3000 Hz_40ms_envelope_correlation": 0.996074406044751, "1000–3000 Hz_40ms_modulation_mae_db": 0.30019769508911925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.958430353674784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29776637985323673, "3000–6000 Hz_10ms_envelope_correlation": 0.9401382058748112, "3000–6000 Hz_10ms_modulation_mae_db": 3.0765647830983056, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4225224953684172, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2290360972002519, "3000–6000 Hz_40ms_envelope_correlation": 0.9621221602517941, "3000–6000 Hz_40ms_modulation_mae_db": 4.273151494778909, "mr_stft_db": 5.959988003629565, "band_mae_db": 0.15644781409419367, "rms_gain_error_db": -0.03202230119397482, "output_peak": 0.432550490311778}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3088161426140852, "全帯域_10ms_rms_envelope_nrmse": 0.08485117967249428, "全帯域_10ms_envelope_correlation": 0.9929389705697479, "全帯域_10ms_modulation_mae_db": 0.3191499650931052, "全帯域_40ms_log_rms_mae_db": 0.6412171344605359, "全帯域_40ms_rms_envelope_nrmse": 0.04537166722868492, "全帯域_40ms_envelope_correlation": 0.9978699360014007, "全帯域_40ms_modulation_mae_db": 0.3859433319727699, "300–1000 Hz_10ms_log_rms_mae_db": 1.9512413735668857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15166770795776355, "300–1000 Hz_10ms_envelope_correlation": 0.9816266425470442, "300–1000 Hz_10ms_modulation_mae_db": 0.67227339168628, "300–1000 Hz_40ms_log_rms_mae_db": 1.1836654555369983, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09672537531490415, "300–1000 Hz_40ms_envelope_correlation": 0.9918985623564551, "300–1000 Hz_40ms_modulation_mae_db": 0.937433896405893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7964759865419304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17557822484577523, "1000–3000 Hz_10ms_envelope_correlation": 0.9785672824102417, "1000–3000 Hz_10ms_modulation_mae_db": 0.31425808699427055, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0798709470530667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10110907507774544, "1000–3000 Hz_40ms_envelope_correlation": 0.9929084164427258, "1000–3000 Hz_40ms_modulation_mae_db": 0.635107285568811, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9281728369966207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29663448075939625, "3000–6000 Hz_10ms_envelope_correlation": 0.9430909734917368, "3000–6000 Hz_10ms_modulation_mae_db": 3.020204116442777, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3436550024473184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21932754225970236, "3000–6000 Hz_40ms_envelope_correlation": 0.9667261888898019, "3000–6000 Hz_40ms_modulation_mae_db": 4.294178004022433, "mr_stft_db": 5.950917409515989, "band_mae_db": 0.20277116229687306, "rms_gain_error_db": -0.0824014902775543, "output_peak": 0.440324707463522}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3288018961582047, "全帯域_10ms_rms_envelope_nrmse": 0.09444476526715119, "全帯域_10ms_envelope_correlation": 0.9912284588090599, "全帯域_10ms_modulation_mae_db": 0.2935936437712368, "全帯域_40ms_log_rms_mae_db": 0.7137067080719915, "全帯域_40ms_rms_envelope_nrmse": 0.04704715788823925, "全帯域_40ms_envelope_correlation": 0.9976951983722637, "全帯域_40ms_modulation_mae_db": 0.5559383812529909, "300–1000 Hz_10ms_log_rms_mae_db": 1.8603284722788433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14141962754197349, "300–1000 Hz_10ms_envelope_correlation": 0.9839752576633037, "300–1000 Hz_10ms_modulation_mae_db": 0.695417723260945, "300–1000 Hz_40ms_log_rms_mae_db": 1.0875540564575883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08107192339165369, "300–1000 Hz_40ms_envelope_correlation": 0.994367476718387, "300–1000 Hz_40ms_modulation_mae_db": 0.9611163713718192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7113734283094215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19911514084244256, "1000–3000 Hz_10ms_envelope_correlation": 0.9727497872988764, "1000–3000 Hz_10ms_modulation_mae_db": 1.2253564807650419, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0020669984768882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12849015760062166, "1000–3000 Hz_40ms_envelope_correlation": 0.9878524497599132, "1000–3000 Hz_40ms_modulation_mae_db": 1.1545129584007088, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9862007535122252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29739604815032744, "3000–6000 Hz_10ms_envelope_correlation": 0.94195117337493, "3000–6000 Hz_10ms_modulation_mae_db": 3.0168062290580266, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3856971682923567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23057031854077953, "3000–6000 Hz_40ms_envelope_correlation": 0.9625006131460159, "3000–6000 Hz_40ms_modulation_mae_db": 4.325508747298291, "mr_stft_db": 5.963119884824958, "band_mae_db": 0.13134497315040525, "rms_gain_error_db": -0.07630714875649874, "output_peak": 0.434657296515874}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.375340794194695, "全帯域_10ms_rms_envelope_nrmse": 0.09015012110578598, "全帯域_10ms_envelope_correlation": 0.9920255089333411, "全帯域_10ms_modulation_mae_db": 0.4431105727001175, "全帯域_40ms_log_rms_mae_db": 0.7289530365210837, "全帯域_40ms_rms_envelope_nrmse": 0.050303898572157855, "全帯域_40ms_envelope_correlation": 0.9973680380693948, "全帯域_40ms_modulation_mae_db": 0.6646506541619198, "300–1000 Hz_10ms_log_rms_mae_db": 2.037088395839473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15589504806184223, "300–1000 Hz_10ms_envelope_correlation": 0.9804571432971892, "300–1000 Hz_10ms_modulation_mae_db": 0.9689363819386574, "300–1000 Hz_40ms_log_rms_mae_db": 1.2964143532537715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09886329601599271, "300–1000 Hz_40ms_envelope_correlation": 0.9914570140570315, "300–1000 Hz_40ms_modulation_mae_db": 1.1878369647542455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8652625153156046, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13926759261647678, "1000–3000 Hz_10ms_envelope_correlation": 0.9871970548170089, "1000–3000 Hz_10ms_modulation_mae_db": 0.769623133780058, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1327937174638283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08409220693822082, "1000–3000 Hz_40ms_envelope_correlation": 0.9958210517696648, "1000–3000 Hz_40ms_modulation_mae_db": 1.0385743217025147, "3000–6000 Hz_10ms_log_rms_mae_db": 1.92327023597422, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3133779826040043, "3000–6000 Hz_10ms_envelope_correlation": 0.9344784620725916, "3000–6000 Hz_10ms_modulation_mae_db": 3.203085410515003, "3000–6000 Hz_40ms_log_rms_mae_db": 1.373796068361449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23008853259691522, "3000–6000 Hz_40ms_envelope_correlation": 0.962072200003169, "3000–6000 Hz_40ms_modulation_mae_db": 4.486946665199552, "mr_stft_db": 5.968246504055606, "band_mae_db": 0.1880357190408335, "rms_gain_error_db": -0.08435477147653267, "output_peak": 0.4732911545878552}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3352873778305943, "全帯域_10ms_rms_envelope_nrmse": 0.09591905577987074, "全帯域_10ms_envelope_correlation": 0.9910383157356117, "全帯域_10ms_modulation_mae_db": 0.5272201946982835, "全帯域_40ms_log_rms_mae_db": 0.6823141225007405, "全帯域_40ms_rms_envelope_nrmse": 0.054714150975191486, "全帯域_40ms_envelope_correlation": 0.996986765940697, "全帯域_40ms_modulation_mae_db": 0.6265674017116446, "300–1000 Hz_10ms_log_rms_mae_db": 1.844139395835595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.144982259841094, "300–1000 Hz_10ms_envelope_correlation": 0.9831217013050765, "300–1000 Hz_10ms_modulation_mae_db": 1.0137713283793426, "300–1000 Hz_40ms_log_rms_mae_db": 1.135199622427962, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09518868784407206, "300–1000 Hz_40ms_envelope_correlation": 0.9920703017964067, "300–1000 Hz_40ms_modulation_mae_db": 1.2565418656637866, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8154693239483728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17119161683456302, "1000–3000 Hz_10ms_envelope_correlation": 0.9795721725783144, "1000–3000 Hz_10ms_modulation_mae_db": 0.5015528745094812, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0751175365551235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11246028526051713, "1000–3000 Hz_40ms_envelope_correlation": 0.9908948718431853, "1000–3000 Hz_40ms_modulation_mae_db": 0.7019310891244368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9389413659135175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2915391296643917, "3000–6000 Hz_10ms_envelope_correlation": 0.9453767051472952, "3000–6000 Hz_10ms_modulation_mae_db": 3.1713196915402175, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3561033460341316, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22567028253325985, "3000–6000 Hz_40ms_envelope_correlation": 0.9649969756073249, "3000–6000 Hz_40ms_modulation_mae_db": 4.438005476480714, "mr_stft_db": 5.950844232162438, "band_mae_db": 0.16024398716857347, "rms_gain_error_db": -0.11512472689782619, "output_peak": 0.45637782929055637}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3124727628070554, "全帯域_10ms_rms_envelope_nrmse": 0.08643330238352841, "全帯域_10ms_envelope_correlation": 0.9926923541743324, "全帯域_10ms_modulation_mae_db": 0.18175643104769093, "全帯域_40ms_log_rms_mae_db": 0.652825749953633, "全帯域_40ms_rms_envelope_nrmse": 0.04605649540962337, "全帯域_40ms_envelope_correlation": 0.9978266608535322, "全帯域_40ms_modulation_mae_db": 0.1373065563489494, "300–1000 Hz_10ms_log_rms_mae_db": 1.8567773454262297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13961210948118166, "300–1000 Hz_10ms_envelope_correlation": 0.9844135994542905, "300–1000 Hz_10ms_modulation_mae_db": 0.8628116264223458, "300–1000 Hz_40ms_log_rms_mae_db": 1.198152605579988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07846118228451837, "300–1000 Hz_40ms_envelope_correlation": 0.9947759478522884, "300–1000 Hz_40ms_modulation_mae_db": 0.7839320015590828, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6858812523628512, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14308827238077051, "1000–3000 Hz_10ms_envelope_correlation": 0.9856981964942225, "1000–3000 Hz_10ms_modulation_mae_db": 0.15259897518979973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9747068147356361, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07579829375557877, "1000–3000 Hz_40ms_envelope_correlation": 0.9957079923336369, "1000–3000 Hz_40ms_modulation_mae_db": 0.07166478028575816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.952018750036524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29829122978275324, "3000–6000 Hz_10ms_envelope_correlation": 0.9409374195783616, "3000–6000 Hz_10ms_modulation_mae_db": 3.0442395191933334, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3533384572854734, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22725206319647911, "3000–6000 Hz_40ms_envelope_correlation": 0.9631989485432587, "3000–6000 Hz_40ms_modulation_mae_db": 4.328155451532347, "mr_stft_db": 5.9519381394178845, "band_mae_db": 0.1487235070872064, "rms_gain_error_db": -0.09014241462367976, "output_peak": 0.4763634759612963}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79920, "output_samples": 79920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.040819702530389, "spectral_convergence": 0.2597090195300295, "spectral_convergence_status": "ok", "amplitude_floor": 1.5443736044991425e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.778845775464492, "spectral_convergence": 0.2474754619141008, "spectral_convergence_status": "ok", "amplitude_floor": 1.1257392826698756e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0240182363552455, "spectral_convergence": 0.2519749823250791, "spectral_convergence_status": "ok", "amplitude_floor": 1.1201855379193364e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001035002994445706, "output_power": 0.0010206438765670769, "delta_db": -0.060673719563400286, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0032595449236567883, "output_power": 0.0031867399841666, "delta_db": -0.09810341436979125, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007961593328208376, "output_power": 0.0007928966709819833, "delta_db": -0.017833959541770993, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023616410570225164, "output_power": 0.0002295609659338072, "delta_db": -0.12315846919598666, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.38734826098075e-05, "output_power": 5.671525205618269e-05, "delta_db": 0.22324815445488239, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3554110211771778e-05, "output_power": 1.2853500978044389e-05, "delta_db": -0.230495777399205, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12558558242083945, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07707469710600927, "pitch": null, "output_rms_dbfs": -22.75772458535963, "output_peak": 0.47053365331467517, "reference_rms_dbfs": -22.68064988825362, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.947894571450043}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001944445221026879, "log_rms_mae_db": 1.3108609504018915, "rms_envelope_nrmse": 0.0865968157332224, "envelope_correlation": 0.9926347940925041, "rms_gain_error_db": -0.07707466577624603, "reference_mean_power": 0.005399702914122853, "output_mean_power": 0.005304719194903107}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 2.8891675057826967e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000788720956533661, "output_power": 0.0007658348795559254, "reference_active": true, "delta_db": -0.12788238122618464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001646344464707057, "output_power": 0.00014880737193527212, "reference_active": true, "delta_db": -0.43896261088390875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.1066196116820634e-05, "output_power": 4.5492260317313655e-05, "reference_active": true, "delta_db": -0.5019599317021862}], "active_band_count": 3, "active_band_mae_db": 0.35626830793742653}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00017595749087279735, "log_rms_mae_db": 0.646793474810485, "rms_envelope_nrmse": 0.04433272910668952, "envelope_correlation": 0.997959760198665, "rms_gain_error_db": -0.0770759091336118, "reference_mean_power": 0.005432355653378944, "output_mean_power": 0.005336796026776268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 2.6677500098154174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007158752148089715, "output_power": 0.0006971614058635373, "reference_active": true, "delta_db": -0.11503989429140069}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.294187442575086e-05, "output_power": 6.737067437621439e-05, "reference_active": true, "delta_db": -0.3450602478022473}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7590528739181784e-06, "output_power": 3.4283948417929036e-06, "reference_active": true, "delta_db": -0.39987601425603214}], "active_band_count": 3, "active_band_mae_db": 0.28665871878322674}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001659932762873133, "log_rms_mae_db": 1.8875456108185558, "rms_envelope_nrmse": 0.14014214527360744, "envelope_correlation": 0.9843414607096128, "rms_gain_error_db": -0.12409611532003233, "reference_mean_power": 0.0017617711849040457, "output_mean_power": 0.001712142433762384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.325463642910379e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006157322295664234, "output_power": 0.0005657562857563924, "reference_active": true, "delta_db": -0.3676249884821115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015692982702212587, "output_power": 0.00012214009839481806, "reference_active": true, "delta_db": -1.0884723051761667}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0206834209715196e-05, "output_power": 3.3874864329793875e-05, "reference_active": true, "delta_db": -0.7442231461786648}], "active_band_count": 3, "active_band_mae_db": 0.7334401466123143}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00014319111528307625, "log_rms_mae_db": 1.1134143754524213, "rms_envelope_nrmse": 0.08234122599580267, "envelope_correlation": 0.994268842555585, "rms_gain_error_db": -0.12409700089824954, "reference_mean_power": 0.0017724260962696937, "output_mean_power": 0.0017224968471124247}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.2006657592514347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005739716253322745, "output_power": 0.0005256046095756338, "reference_active": true, "delta_db": -0.38231257942868074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.210701544963183e-05, "output_power": 5.577616561210166e-05, "reference_active": true, "delta_db": -1.1152886528663577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.009713658025178e-06, "output_power": 3.214829451609562e-06, "reference_active": true, "delta_db": -0.9595542153609736}], "active_band_count": 3, "active_band_mae_db": 0.8190518158853374}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.00012416177010932925, "log_rms_mae_db": 1.8000036019082628, "rms_envelope_nrmse": 0.15243654834148876, "envelope_correlation": 0.9837480253913637, "rms_gain_error_db": -0.07808563333649514, "reference_mean_power": 0.0002459141129048244, "output_mean_power": 0.00024153211820992989}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.7078032497142614e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.578711086190898e-05, "output_power": 8.642005169193581e-05, "reference_active": true, "delta_db": -0.4469255248616709}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5502167493014184e-05, "output_power": 1.4683684049769309e-05, "reference_active": true, "delta_db": -0.2355739373198364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901592249182956e-06, "output_power": 5.201243583221593e-06, "reference_active": true, "delta_db": -0.5486200706103863}], "active_band_count": 3, "active_band_mae_db": 0.41037317759729786}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 9.745648796239735e-05, "log_rms_mae_db": 1.0757033629673285, "rms_envelope_nrmse": 0.08279193431971629, "envelope_correlation": 0.9949241638721025, "rms_gain_error_db": -0.07809247871427606, "reference_mean_power": 0.00024740043253105237, "output_mean_power": 0.00024299156979259404}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.5717362127850378e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.047075579844295e-05, "output_power": 8.233428497840058e-05, "reference_active": true, "delta_db": -0.40927500176929743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.5496928110188975e-06, "output_power": 6.735981872832808e-06, "reference_active": true, "delta_db": -0.49528371242076186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.070423026829593e-07, "output_power": 3.1812734334189754e-07, "reference_active": true, "delta_db": -1.0703854776331974}], "active_band_count": 3, "active_band_mae_db": 0.6583147306077523}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 3.6263821941295626e-05, "log_rms_mae_db": 1.9802028344135154, "rms_envelope_nrmse": 0.2973168057869667, "envelope_correlation": 0.9414570619143926, "rms_gain_error_db": 0.02452095574762033, "reference_mean_power": 2.0515613806993198e-05, "output_mean_power": 2.063177584478116e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 4.904304308942738e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.621343716362641e-06, "output_power": 2.096080986465259e-06, "reference_active": true, "delta_db": 1.115329657490963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.453801656353405e-07, "output_power": 9.350168516211462e-07, "reference_active": true, "delta_db": 0.9844160603253421}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.409211497093947e-07, "output_power": 1.2657592690303782e-06, "reference_active": true, "delta_db": 7.20476189348764}], "active_band_count": 3, "active_band_mae_db": 3.101502537101315}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 2.5114070553475023e-05, "log_rms_mae_db": 1.3696087476989, "rms_envelope_nrmse": 0.22780876624131788, "envelope_correlation": 0.9630811105255704, "rms_gain_error_db": 0.024518543714133915, "reference_mean_power": 2.063955290066712e-05, "output_mean_power": 2.0756405169577718e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 4.623298998188318e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6281370009575077e-06, "output_power": 2.6581286139249087e-06, "reference_active": true, "delta_db": 2.1288504434908506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7397133003103647e-07, "output_power": 6.715769974583587e-07, "reference_active": true, "delta_db": 2.542575035481593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7174687173869334e-08, "output_power": 1.1601234879159105e-07, "reference_active": true, "delta_db": 8.296153842260548}], "active_band_count": 3, "active_band_mae_db": 4.322526440410997}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8512250139074506, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f662d19d5f238182552fb0f9eef252838aef448bfa19cc223c5e5fa56a2e744e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.981647549753013, "standard_deviation": 0.004760851311802942, "minimum": 0.9727497872988764, "maximum": 0.9871970548170089, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7756579506275698, "standard_deviation": 0.05924835871543757, "minimum": 1.6858812523628512, "maximum": 1.8652625153156046, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4804539337080225, "standard_deviation": 0.3593971606832375, "minimum": 0.15259897518979973, "maximum": 1.2253564807650419, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16167630962370405, "standard_deviation": 0.020348201916762607, "minimum": 0.13926759261647678, "maximum": 0.19911514084244256, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9932024887331375, "standard_deviation": 0.0029640584700897313, "minimum": 0.9878524497599132, "maximum": 0.996074406044751, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0533915142863985, "standard_deviation": 0.05177835925972038, "minimum": 0.9747068147356361, "maximum": 1.1327937174638283, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6095957529944995, "standard_deviation": 0.37100850382607886, "minimum": 0.07166478028575816, "maximum": 1.1545129584007088, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09581634248662377, "standard_deviation": 0.01942231031461273, "minimum": 0.07489897349203867, "maximum": 0.12849015760062166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9409879629945128, "standard_deviation": 0.0031212233584809327, "minimum": 0.9344784620725916, "maximum": 0.9453767051472952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9497308437570182, "standard_deviation": 0.023864073379054326, "minimum": 1.92327023597422, "maximum": 1.9862007535122252, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.1021484451451493, "standard_deviation": 0.07520629771730716, "minimum": 3.0168062290580266, "maximum": 3.203085410515003, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29892073714420436, "standard_deviation": 0.006272027361652997, "minimum": 0.2915391296643917, "maximum": 0.3133779826040043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9636875148238047, "standard_deviation": 0.0016589525878085017, "minimum": 0.962072200003169, "maximum": 0.9667261888898019, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3677446956885488, "standard_deviation": 0.0273002613415672, "minimum": 1.337236280020344, "maximum": 1.4225224953684172, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.36689806635569, "standard_deviation": 0.08340121240313973, "minimum": 4.273151494778909, "maximum": 4.486946665199552, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.22648391304200177, "standard_deviation": 0.003954931188620377, "minimum": 0.21932754225970236, "maximum": 0.23057031854077953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9831512737278515, "standard_deviation": 0.0017305237128108911, "minimum": 0.9804571432971892, "maximum": 0.9855369395019928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.889278013800956, "standard_deviation": 0.07135604132516132, "minimum": 1.8176561367432602, "maximum": 2.037088395839473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7906990792628672, "standard_deviation": 0.16388064621430676, "minimum": 0.5245588092650894, "maximum": 1.0137713283793426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1449760275825704, "standard_deviation": 0.00738903375782494, "minimum": 0.13450256732865523, "maximum": 0.15589504806184223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9934488586098986, "standard_deviation": 0.0014746526649011991, "minimum": 0.9914570140570315, "maximum": 0.9953645414877458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1590623017396278, "standard_deviation": 0.06632634282181032, "minimum": 1.0875540564575883, "maximum": 1.2964143532537715, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9282891122858803, "standard_deviation": 0.20055797617365803, "minimum": 0.7105225451718203, "maximum": 1.2565418656637866, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08665531512887745, "standard_deviation": 0.009348969789417844, "minimum": 0.07355979355020031, "maximum": 0.09886329601599271, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1525181371649113, "standard_deviation": 0.031898629577624564, "minimum": 0.10699235206036563, "maximum": 0.20277116229687306, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.957135207782997, "standard_deviation": 0.0076253329151065264, "minimum": 5.947894571450043, "maximum": 5.968246504055606, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46386780789314586, "standard_deviation": 0.03093717825489131, "minimum": 0.432550490311778, "maximum": 0.5268438556996099, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07239754321172515, "standard_deviation": 0.030742614417807138, "minimum": -0.11512472689782619, "maximum": -0.02175279536172551, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918381347199978, "standard_deviation": 0.0008828269550656227, "minimum": 0.990497519038596, "maximum": 0.9929389705697479, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3304894620555816, "standard_deviation": 0.03098246556805137, "minimum": 1.2947083840482199, "maximum": 1.3776273883899073, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34944625129045914, "standard_deviation": 0.174631943797805, "minimum": 0.07209809817748856, "maximum": 0.6023727968983245, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0910835726275406, "standard_deviation": 0.004898819503399986, "minimum": 0.08485117967249428, "maximum": 0.09821760015441543, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976369982990478, "standard_deviation": 0.0003161486058089785, "minimum": 0.996986765940697, "maximum": 0.997959760198665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6767970580826914, "standard_deviation": 0.03324937995569697, "minimum": 0.6412171344605359, "maximum": 0.7289530365210837, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43559231570096774, "standard_deviation": 0.22240185588757377, "minimum": 0.1373065563489494, "maximum": 0.6670643803030941, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04758365072841385, "standard_deviation": 0.003359982337308409, "minimum": 0.04433272910668952, "maximum": 0.054714150975191486, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123440-0020", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.7928903102874756, "predicted_mos": 1.7928903102874756, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "8fdde44507a3b55bc3ce5d618ebf2a267e8222801a715f4dc0ca0b5ae9582f7d", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.095909500028938, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2762667215393595, "全帯域_10ms_rms_envelope_nrmse": 0.08477368463722848, "全帯域_10ms_envelope_correlation": 0.9929227142728585, "全帯域_10ms_modulation_mae_db": 0.21473612015406426, "全帯域_40ms_log_rms_mae_db": 0.6150083139100644, "全帯域_40ms_rms_envelope_nrmse": 0.04067630020508455, "全帯域_40ms_envelope_correlation": 0.9982525273433883, "全帯域_40ms_modulation_mae_db": 0.16437098470136138, "300–1000 Hz_10ms_log_rms_mae_db": 1.9432462822431933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13359282715356607, "300–1000 Hz_10ms_envelope_correlation": 0.9858198065573394, "300–1000 Hz_10ms_modulation_mae_db": 0.48161880976254196, "300–1000 Hz_40ms_log_rms_mae_db": 1.1798379993689871, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07887870351296145, "300–1000 Hz_40ms_envelope_correlation": 0.9947825311623155, "300–1000 Hz_40ms_modulation_mae_db": 0.7532909171936328, "1000–3000 Hz_10ms_log_rms_mae_db": 2.037552009012728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17290620428956438, "1000–3000 Hz_10ms_envelope_correlation": 0.9791271524934376, "1000–3000 Hz_10ms_modulation_mae_db": 0.27134372255913985, "1000–3000 Hz_40ms_log_rms_mae_db": 1.189271002438303, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09943580856511837, "1000–3000 Hz_40ms_envelope_correlation": 0.9930018613583264, "1000–3000 Hz_40ms_modulation_mae_db": 0.5140149542897011, "3000–6000 Hz_10ms_log_rms_mae_db": 2.333124831030886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3088994167417813, "3000–6000 Hz_10ms_envelope_correlation": 0.9359851885936538, "3000–6000 Hz_10ms_modulation_mae_db": 3.0215976701010514, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6123375849295687, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2317430683632279, "3000–6000 Hz_40ms_envelope_correlation": 0.9615982234328391, "3000–6000 Hz_40ms_modulation_mae_db": 4.304189145353883, "mr_stft_db": 6.122947295973577, "band_mae_db": 0.1030746518179358, "rms_gain_error_db": -0.04999919395102753, "output_peak": 0.45781398891155084}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3149660224137292, "全帯域_10ms_rms_envelope_nrmse": 0.10412349021399235, "全帯域_10ms_envelope_correlation": 0.9893315902242841, "全帯域_10ms_modulation_mae_db": 0.30068510346291216, "全帯域_40ms_log_rms_mae_db": 0.689053658107068, "全帯域_40ms_rms_envelope_nrmse": 0.0555664229097151, "全帯域_40ms_envelope_correlation": 0.9967002403380596, "全帯域_40ms_modulation_mae_db": 0.4650495247615935, "300–1000 Hz_10ms_log_rms_mae_db": 1.9973751787864622, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14525260735512047, "300–1000 Hz_10ms_envelope_correlation": 0.9831161510626937, "300–1000 Hz_10ms_modulation_mae_db": 0.9164641880596506, "300–1000 Hz_40ms_log_rms_mae_db": 1.1957748537058976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07943897094977648, "300–1000 Hz_40ms_envelope_correlation": 0.9946804751608607, "300–1000 Hz_40ms_modulation_mae_db": 1.2580366928511626, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0669594630813544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.183749633803477, "1000–3000 Hz_10ms_envelope_correlation": 0.9765483465750272, "1000–3000 Hz_10ms_modulation_mae_db": 0.5152103789305366, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2466762553765378, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10895063897741122, "1000–3000 Hz_40ms_envelope_correlation": 0.9917898628991717, "1000–3000 Hz_40ms_modulation_mae_db": 0.5945913673436388, "3000–6000 Hz_10ms_log_rms_mae_db": 2.298759723292339, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31270101018630897, "3000–6000 Hz_10ms_envelope_correlation": 0.9376141034666156, "3000–6000 Hz_10ms_modulation_mae_db": 3.1229515704272157, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6194254409198705, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2328598440382161, "3000–6000 Hz_40ms_envelope_correlation": 0.9632091860361006, "3000–6000 Hz_40ms_modulation_mae_db": 4.371791893084753, "mr_stft_db": 6.12640452429918, "band_mae_db": 0.15850680827987898, "rms_gain_error_db": -0.025113854449136556, "output_peak": 0.44876901280837767}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3411324474748905, "全帯域_10ms_rms_envelope_nrmse": 0.09497163009795, "全帯域_10ms_envelope_correlation": 0.9911172408745279, "全帯域_10ms_modulation_mae_db": 0.45656153717843356, "全帯域_40ms_log_rms_mae_db": 0.6149873651889969, "全帯域_40ms_rms_envelope_nrmse": 0.04218341596066513, "全帯域_40ms_envelope_correlation": 0.9981213011001757, "全帯域_40ms_modulation_mae_db": 0.652676533369513, "300–1000 Hz_10ms_log_rms_mae_db": 1.9329693680305966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1504329158222176, "300–1000 Hz_10ms_envelope_correlation": 0.98181347207334, "300–1000 Hz_10ms_modulation_mae_db": 0.696286499946876, "300–1000 Hz_40ms_log_rms_mae_db": 1.1113232609479178, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08160628304059579, "300–1000 Hz_40ms_envelope_correlation": 0.9942412033426662, "300–1000 Hz_40ms_modulation_mae_db": 0.6246748190784389, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0019868783728985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16896228982840522, "1000–3000 Hz_10ms_envelope_correlation": 0.9803423886346196, "1000–3000 Hz_10ms_modulation_mae_db": 0.5997631651765343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1930598487106332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10119678952781108, "1000–3000 Hz_40ms_envelope_correlation": 0.9930015023197131, "1000–3000 Hz_40ms_modulation_mae_db": 0.8882189435577464, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2620883351717107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3098783556391621, "3000–6000 Hz_10ms_envelope_correlation": 0.9378506950865577, "3000–6000 Hz_10ms_modulation_mae_db": 3.1491937821631892, "3000–6000 Hz_40ms_log_rms_mae_db": 1.612923841546364, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2323677516826966, "3000–6000 Hz_40ms_envelope_correlation": 0.9626594647368625, "3000–6000 Hz_40ms_modulation_mae_db": 4.386404556625515, "mr_stft_db": 6.120281778151437, "band_mae_db": 0.16861616720897835, "rms_gain_error_db": -0.043713116209846276, "output_peak": 0.4598920994423113}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.319583575725213, "全帯域_10ms_rms_envelope_nrmse": 0.08353462271458048, "全帯域_10ms_envelope_correlation": 0.993144756460958, "全帯域_10ms_modulation_mae_db": 0.31779816862617155, "全帯域_40ms_log_rms_mae_db": 0.6564792237930265, "全帯域_40ms_rms_envelope_nrmse": 0.04171361856217174, "全帯域_40ms_envelope_correlation": 0.9981917399375889, "全帯域_40ms_modulation_mae_db": 0.546297120365459, "300–1000 Hz_10ms_log_rms_mae_db": 1.9254249061085442, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13626533965291845, "300–1000 Hz_10ms_envelope_correlation": 0.9852148766357836, "300–1000 Hz_10ms_modulation_mae_db": 0.7536555062606061, "300–1000 Hz_40ms_log_rms_mae_db": 1.0857531717866962, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07957433284013637, "300–1000 Hz_40ms_envelope_correlation": 0.9946564944621353, "300–1000 Hz_40ms_modulation_mae_db": 0.868251307536689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9946636932330508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16080185175717304, "1000–3000 Hz_10ms_envelope_correlation": 0.982550756097606, "1000–3000 Hz_10ms_modulation_mae_db": 0.3204117511185803, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1694909375407583, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08471802833340336, "1000–3000 Hz_40ms_envelope_correlation": 0.9949823735512784, "1000–3000 Hz_40ms_modulation_mae_db": 0.2551428681166326, "3000–6000 Hz_10ms_log_rms_mae_db": 2.303820580587909, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2948762449396225, "3000–6000 Hz_10ms_envelope_correlation": 0.9442724484664627, "3000–6000 Hz_10ms_modulation_mae_db": 3.388746256927151, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6302594357840927, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22532282573452062, "3000–6000 Hz_40ms_envelope_correlation": 0.9653145397908687, "3000–6000 Hz_40ms_modulation_mae_db": 4.648710975519752, "mr_stft_db": 6.113478744851332, "band_mae_db": 0.1490494654031643, "rms_gain_error_db": -0.06936728028427483, "output_peak": 0.49393121364148157}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3437214129181665, "全帯域_10ms_rms_envelope_nrmse": 0.095233115531665, "全帯域_10ms_envelope_correlation": 0.9911542131281401, "全帯域_10ms_modulation_mae_db": 0.07491973860121402, "全帯域_40ms_log_rms_mae_db": 0.7274859957077531, "全帯域_40ms_rms_envelope_nrmse": 0.049441395951695215, "全帯域_40ms_envelope_correlation": 0.9975530992620346, "全帯域_40ms_modulation_mae_db": 0.10707727924253947, "300–1000 Hz_10ms_log_rms_mae_db": 1.9717992018710708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14126876718786735, "300–1000 Hz_10ms_envelope_correlation": 0.9841466163645628, "300–1000 Hz_10ms_modulation_mae_db": 0.5787299233225681, "300–1000 Hz_40ms_log_rms_mae_db": 1.222543699806582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08364790638087069, "300–1000 Hz_40ms_envelope_correlation": 0.9942361616220097, "300–1000 Hz_40ms_modulation_mae_db": 0.8679094651591255, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9705299221355947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17376919865747487, "1000–3000 Hz_10ms_envelope_correlation": 0.9792753079740818, "1000–3000 Hz_10ms_modulation_mae_db": 0.5537033814560064, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1557707518169653, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08822642218675934, "1000–3000 Hz_40ms_envelope_correlation": 0.9943521432253358, "1000–3000 Hz_40ms_modulation_mae_db": 0.30263349564388653, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3234669584827006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3228139587780662, "3000–6000 Hz_10ms_envelope_correlation": 0.9385071887942185, "3000–6000 Hz_10ms_modulation_mae_db": 3.420666023959985, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6774801653871636, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23749798130417182, "3000–6000 Hz_40ms_envelope_correlation": 0.9656620842179947, "3000–6000 Hz_40ms_modulation_mae_db": 4.623652318192053, "mr_stft_db": 6.117472450013563, "band_mae_db": 0.19935358409266457, "rms_gain_error_db": -0.12107930803026734, "output_peak": 0.4542789760967214}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2750507607929253, "全帯域_10ms_rms_envelope_nrmse": 0.09228482821553273, "全帯域_10ms_envelope_correlation": 0.9916342069163346, "全帯域_10ms_modulation_mae_db": 0.3910139090001555, "全帯域_40ms_log_rms_mae_db": 0.6665259929964081, "全帯域_40ms_rms_envelope_nrmse": 0.04735445175100379, "全帯域_40ms_envelope_correlation": 0.9976464794458062, "全帯域_40ms_modulation_mae_db": 0.6047498556936365, "300–1000 Hz_10ms_log_rms_mae_db": 1.9868061913079378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15392604586215847, "300–1000 Hz_10ms_envelope_correlation": 0.9810527604130032, "300–1000 Hz_10ms_modulation_mae_db": 0.6256588226230552, "300–1000 Hz_40ms_log_rms_mae_db": 1.247064026431833, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1027684599400068, "300–1000 Hz_40ms_envelope_correlation": 0.990700060777939, "300–1000 Hz_40ms_modulation_mae_db": 0.722775592001979, "1000–3000 Hz_10ms_log_rms_mae_db": 2.03342747355878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1757794031068639, "1000–3000 Hz_10ms_envelope_correlation": 0.9793009118098405, "1000–3000 Hz_10ms_modulation_mae_db": 0.42576965855688176, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1485567716419633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08913182159500467, "1000–3000 Hz_40ms_envelope_correlation": 0.9945727678465834, "1000–3000 Hz_40ms_modulation_mae_db": 0.06993349728916355, "3000–6000 Hz_10ms_log_rms_mae_db": 2.240783339816203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3134414944102736, "3000–6000 Hz_10ms_envelope_correlation": 0.9357548219593084, "3000–6000 Hz_10ms_modulation_mae_db": 3.2838580043138443, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5877695627272213, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23922337657562628, "3000–6000 Hz_40ms_envelope_correlation": 0.9599868182665354, "3000–6000 Hz_40ms_modulation_mae_db": 4.4738602201074995, "mr_stft_db": 6.103801132915212, "band_mae_db": 0.10949145757959688, "rms_gain_error_db": -0.048236065663029536, "output_peak": 0.4907077957561094}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2838789124042072, "全帯域_10ms_rms_envelope_nrmse": 0.09227851369380798, "全帯域_10ms_envelope_correlation": 0.9916326049476916, "全帯域_10ms_modulation_mae_db": 0.3243849407259762, "全帯域_40ms_log_rms_mae_db": 0.6370346788035139, "全帯域_40ms_rms_envelope_nrmse": 0.04887553980066739, "全帯域_40ms_envelope_correlation": 0.9974463943116466, "全帯域_40ms_modulation_mae_db": 0.527962666613096, "300–1000 Hz_10ms_log_rms_mae_db": 2.0027061999251505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15096891521511743, "300–1000 Hz_10ms_envelope_correlation": 0.9819699304983924, "300–1000 Hz_10ms_modulation_mae_db": 0.4607014697232854, "300–1000 Hz_40ms_log_rms_mae_db": 1.1853355129155543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09010760200363382, "300–1000 Hz_40ms_envelope_correlation": 0.9929733036819183, "300–1000 Hz_40ms_modulation_mae_db": 0.7816038254891015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9232028734676825, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1607392498859568, "1000–3000 Hz_10ms_envelope_correlation": 0.9819211743646876, "1000–3000 Hz_10ms_modulation_mae_db": 0.6855663707120763, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0988980880166161, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07655829281814937, "1000–3000 Hz_40ms_envelope_correlation": 0.995651242845992, "1000–3000 Hz_40ms_modulation_mae_db": 1.0144115925346262, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2694781256991927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.307672817217779, "3000–6000 Hz_10ms_envelope_correlation": 0.9368802393236598, "3000–6000 Hz_10ms_modulation_mae_db": 3.288196119210632, "3000–6000 Hz_40ms_log_rms_mae_db": 1.57913130027769, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2350480238636529, "3000–6000 Hz_40ms_envelope_correlation": 0.9606193963950276, "3000–6000 Hz_40ms_modulation_mae_db": 4.577750798908338, "mr_stft_db": 6.116364539029997, "band_mae_db": 0.10450710935572992, "rms_gain_error_db": 0.00939657106604332, "output_peak": 0.44604494030588193}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3082523749899806, "全帯域_10ms_rms_envelope_nrmse": 0.08703562944443075, "全帯域_10ms_envelope_correlation": 0.9926492848283732, "全帯域_10ms_modulation_mae_db": 0.1268524756946157, "全帯域_40ms_log_rms_mae_db": 0.7144641136960403, "全帯域_40ms_rms_envelope_nrmse": 0.045390114045731914, "全帯域_40ms_envelope_correlation": 0.9979899186359854, "全帯域_40ms_modulation_mae_db": 0.4205164113904189, "300–1000 Hz_10ms_log_rms_mae_db": 1.9392215619569608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1461851765981198, "300–1000 Hz_10ms_envelope_correlation": 0.9828626819735792, "300–1000 Hz_10ms_modulation_mae_db": 0.2578310406531344, "300–1000 Hz_40ms_log_rms_mae_db": 1.1845968408385106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08787869051394824, "300–1000 Hz_40ms_envelope_correlation": 0.9933085653133649, "300–1000 Hz_40ms_modulation_mae_db": 0.5475464387134399, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9637977994611289, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16488014776262186, "1000–3000 Hz_10ms_envelope_correlation": 0.9812511834480209, "1000–3000 Hz_10ms_modulation_mae_db": 0.42726116634818895, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2048650886261687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09573679263920151, "1000–3000 Hz_40ms_envelope_correlation": 0.9937988822709948, "1000–3000 Hz_40ms_modulation_mae_db": 0.18623516333663517, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2492033666062743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31531075248173407, "3000–6000 Hz_10ms_envelope_correlation": 0.9336655189517709, "3000–6000 Hz_10ms_modulation_mae_db": 3.1456081448127584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5978300767705373, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23753230923408364, "3000–6000 Hz_40ms_envelope_correlation": 0.9597045930159436, "3000–6000 Hz_40ms_modulation_mae_db": 4.415327527615104, "mr_stft_db": 6.117091806054226, "band_mae_db": 0.11574569189213951, "rms_gain_error_db": -0.12725172242355812, "output_peak": 0.4639738051629353}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79920, "output_samples": 79920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2764846279305155, "spectral_convergence": 0.26140565214256795, "spectral_convergence_status": "ok", "amplitude_floor": 1.5443736044991425e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.975607657013561, "spectral_convergence": 0.252482082617658, "spectral_convergence_status": "ok", "amplitude_floor": 1.1257392826698756e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.116749602976654, "spectral_convergence": 0.25893439618852543, "spectral_convergence_status": "ok", "amplitude_floor": 1.1201855379193364e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001035002994445706, "output_power": 0.001051946943011967, "delta_db": 0.07052229598981329, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0032595449236567883, "output_power": 0.003184726420505586, "delta_db": -0.10084840134927434, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007961593328208376, "output_power": 0.0008066277195346742, "delta_db": 0.05673151934393872, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023616410570225164, "output_power": 0.0002213608262020946, "delta_db": -0.281131233116905, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.38734826098075e-05, "output_power": 5.403131779543154e-05, "delta_db": 0.012705087965741269, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3554110211771778e-05, "output_power": 1.385868265050842e-05, "delta_db": 0.09650937314194225, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1030746518179358, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04999919395102753, "pitch": null, "output_rms_dbfs": -22.73064908220465, "output_peak": 0.45781398891155084, "reference_rms_dbfs": -22.68064988825362, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.122947295973577}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001944445221026879, "log_rms_mae_db": 1.2762667215393595, "rms_envelope_nrmse": 0.08477368463722848, "envelope_correlation": 0.9929227142728585, "rms_gain_error_db": -0.04999852666440638, "reference_mean_power": 0.005399702914122853, "output_mean_power": 0.0053378948368543845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 2.8891675057826967e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000788720956533661, "output_power": 0.0007869602738346017, "reference_active": true, "delta_db": -0.00970570803405988}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001646344464707057, "output_power": 0.00015155603252589637, "reference_active": true, "delta_db": -0.35947480157176737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.1066196116820634e-05, "output_power": 5.440468421930167e-05, "reference_active": true, "delta_db": 0.2750278508563655}], "active_band_count": 3, "active_band_mae_db": 0.21473612015406426}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00017595749087279735, "log_rms_mae_db": 0.6150083139100644, "rms_envelope_nrmse": 0.04067630020508455, "envelope_correlation": 0.9982525273433883, "rms_gain_error_db": -0.04999912426406652, "reference_mean_power": 0.005432355653378944, "output_mean_power": 0.005370173075296806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 2.6677500098154174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007158752148089715, "output_power": 0.0007075551650124634, "reference_active": true, "delta_db": -0.050770208613104606}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.294187442575086e-05, "output_power": 6.815126938538622e-05, "reference_active": true, "delta_db": -0.2950297000107261}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7590528739181784e-06, "output_power": 3.888747446878081e-06, "reference_active": true, "delta_db": 0.14731304548025342}], "active_band_count": 3, "active_band_mae_db": 0.16437098470136138}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001659932762873133, "log_rms_mae_db": 1.9432462822431933, "rms_envelope_nrmse": 0.13359282715356607, "envelope_correlation": 0.9858198065573394, "rms_gain_error_db": -0.1621180108831099, "reference_mean_power": 0.0017617711849040457, "output_mean_power": 0.0016972182826947018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.325463642910379e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006157322295664234, "output_power": 0.0005897359857070575, "reference_active": true, "delta_db": -0.18734257421783684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015692982702212587, "output_power": 0.00013177157577359518, "reference_active": true, "delta_db": -0.7588375661732977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0206834209715196e-05, "output_power": 3.584530253965147e-05, "reference_active": true, "delta_db": -0.49867628889649135}], "active_band_count": 3, "active_band_mae_db": 0.48161880976254196}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00014319111528307625, "log_rms_mae_db": 1.1798379993689871, "rms_envelope_nrmse": 0.07887870351296145, "envelope_correlation": 0.9947825311623155, "rms_gain_error_db": -0.16211835236262467, "reference_mean_power": 0.0017724260962696937, "output_mean_power": 0.0017074826541381995}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.2006657592514347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005739716253322745, "output_power": 0.0005499224746139289, "reference_active": true, "delta_db": -0.1858895419974442}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.210701544963183e-05, "output_power": 6.373089936772943e-05, "reference_active": true, "delta_db": -0.5362747296910102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.009713658025178e-06, "output_power": 2.8141193464750575e-06, "reference_active": true, "delta_db": -1.5377084798924443}], "active_band_count": 3, "active_band_mae_db": 0.7532909171936328}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.00012416177010932925, "log_rms_mae_db": 2.037552009012728, "rms_envelope_nrmse": 0.17290620428956438, "envelope_correlation": 0.9791271524934376, "rms_gain_error_db": -0.21711050226061018, "reference_mean_power": 0.0002459141129048244, "output_mean_power": 0.00023392271893165573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.7078032497142614e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.578711086190898e-05, "output_power": 8.691556127790191e-05, "reference_active": true, "delta_db": -0.4220953516008963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5502167493014184e-05, "output_power": 1.5654612485837696e-05, "reference_active": true, "delta_db": 0.04249896654917566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901592249182956e-06, "output_power": 5.445346437134289e-06, "reference_active": true, "delta_db": -0.3494368495273476}], "active_band_count": 3, "active_band_mae_db": 0.27134372255913985}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 9.745648796239735e-05, "log_rms_mae_db": 1.189271002438303, "rms_envelope_nrmse": 0.09943580856511837, "envelope_correlation": 0.9930018613583264, "rms_gain_error_db": -0.2171184756070418, "reference_mean_power": 0.00024740043253105237, "output_mean_power": 0.000235336129793283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.5717362127850378e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.047075579844295e-05, "output_power": 8.142145108783894e-05, "reference_active": true, "delta_db": -0.4576938037501199}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.5496928110188975e-06, "output_power": 8.103287658735717e-06, "reference_active": true, "delta_db": 0.30731975199860906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.070423026829593e-07, "output_power": 3.4035835483972466e-07, "reference_active": true, "delta_db": -0.7770313071203743}], "active_band_count": 3, "active_band_mae_db": 0.5140149542897011}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 3.6263821941295626e-05, "log_rms_mae_db": 2.333124831030886, "rms_envelope_nrmse": 0.3088994167417813, "envelope_correlation": 0.9359851885936538, "rms_gain_error_db": -0.08395929413084173, "reference_mean_power": 2.0515613806993198e-05, "output_mean_power": 2.012280810729988e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 4.904304308942738e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.621343716362641e-06, "output_power": 2.068627767865392e-06, "reference_active": true, "delta_db": 1.0580725730283076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.453801656353405e-07, "output_power": 9.153574855931786e-07, "reference_active": true, "delta_db": 0.8921290534149329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.409211497093947e-07, "output_power": 1.239749852816124e-06, "reference_active": true, "delta_db": 7.114591383859913}], "active_band_count": 3, "active_band_mae_db": 3.0215976701010514}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 2.5114070553475023e-05, "log_rms_mae_db": 1.6123375849295687, "rms_envelope_nrmse": 0.2317430683632279, "envelope_correlation": 0.9615982234328391, "rms_gain_error_db": -0.0839602396493247, "reference_mean_power": 2.063955290066712e-05, "output_mean_power": 2.024436977249627e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 4.623298998188318e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6281370009575077e-06, "output_power": 2.607063091236952e-06, "reference_active": true, "delta_db": 2.044605951083436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7397133003103647e-07, "output_power": 6.740082126738221e-07, "reference_active": true, "delta_db": 2.558268794500678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7174687173869334e-08, "output_power": 1.1637457406259905e-07, "reference_active": true, "delta_db": 8.309692690477535}], "active_band_count": 3, "active_band_mae_db": 4.304189145353883}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8571012707021751, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "07d3d92b202be1e41529b8ae4d08d458cb31a093744ee713d40b75a4bdc5a74b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9800396526746652, "standard_deviation": 0.0019107611010032077, "minimum": 0.9765483465750272, "maximum": 0.982550756097606, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9990150140404024, "standard_deviation": 0.04646452533035218, "minimum": 1.9232028734676825, "maximum": 2.0669594630813544, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47487869935724303, "standard_deviation": 0.14030123294725322, "minimum": 0.27134372255913985, "maximum": 0.6855663707120763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17019849738644213, "standard_deviation": 0.007950280815334516, "minimum": 0.1607392498859568, "maximum": 0.183749633803477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938938295396744, "standard_deviation": 0.0012519689616002853, "minimum": 0.9917898628991717, "maximum": 0.995651242845992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1758235930209933, "standard_deviation": 0.04382540260457122, "minimum": 1.0988980880166161, "maximum": 1.2466762553765378, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.47814773526400384, "standard_deviation": 0.33868888592720237, "minimum": 0.06993349728916355, "maximum": 1.0144115925346262, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09299432433035737, "standard_deviation": 0.01033157635278469, "minimum": 0.07655829281814937, "maximum": 0.10895063897741122, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9375662755802809, "standard_deviation": 0.0031032011747839797, "minimum": 0.9336655189517709, "maximum": 0.9442724484664627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2850906575859016, "standard_deviation": 0.034520324680177066, "minimum": 2.240783339816203, "maximum": 2.333124831030886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.227602196489478, "standard_deviation": 0.1395594657955025, "minimum": 3.0215976701010514, "maximum": 3.420666023959985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.310699256299341, "standard_deviation": 0.007955224015884367, "minimum": 0.2948762449396225, "maximum": 0.3228139587780662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9623442882365215, "standard_deviation": 0.002290822340618559, "minimum": 0.9597045930159436, "maximum": 0.9656620842179947, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6146446760428135, "standard_deviation": 0.03044775717554931, "minimum": 1.57913130027769, "maximum": 1.6774801653871636, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.475210929425862, "standard_deviation": 0.12765461354196142, "minimum": 4.304189145353883, "maximum": 4.648710975519752, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2339493975995245, "standard_deviation": 0.004430831873019138, "minimum": 0.22532282573452062, "maximum": 0.23922337657562628, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9832495369473369, "standard_deviation": 0.0016888943184879708, "minimum": 0.9810527604130032, "maximum": 0.9858198065573394, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9624436112787396, "standard_deviation": 0.030866693557938536, "minimum": 1.9254249061085442, "maximum": 2.0027061999251505, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5963682825439647, "standard_deviation": 0.20161269335291127, "minimum": 0.2578310406531344, "maximum": 0.9164641880596506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14473657435588572, "standard_deviation": 0.007231598414698886, "minimum": 0.13359282715356607, "maximum": 0.15392604586215847, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936973494404012, "standard_deviation": 0.001378889047775082, "minimum": 0.990700060777939, "maximum": 0.9947825311623155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1765286707252474, "standard_deviation": 0.05359714682107817, "minimum": 1.0857531717866962, "maximum": 1.247064026431833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8030111322529462, "standard_deviation": 0.21445961977973382, "minimum": 0.5475464387134399, "maximum": 1.2580366928511626, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08548761864774121, "standard_deviation": 0.008100873861521429, "minimum": 0.07887870351296145, "maximum": 0.1027684599400068, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13854311695376104, "standard_deviation": 0.035639899247385406, "minimum": 0.1030746518179358, "maximum": 0.19935358409266457, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.1172302839110655, "standard_deviation": 0.006776071420226802, "minimum": 6.103801132915212, "maximum": 6.12640452429918, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4644264790156712, "standard_deviation": 0.01817272894406887, "minimum": 0.44604494030588193, "maximum": 0.49393121364148157, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05942049624313711, "standard_deviation": 0.0460979259913225, "minimum": -0.12725172242355812, "maximum": 0.00939657106604332, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.991698326456646, "standard_deviation": 0.0012386166189047208, "minimum": 0.9893315902242841, "maximum": 0.993144756460958, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.307856528532309, "standard_deviation": 0.027336167369114216, "minimum": 1.2750507607929253, "maximum": 1.3437214129181665, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27586899918044283, "standard_deviation": 0.12926141447743816, "minimum": 0.07491973860121402, "maximum": 0.45656153717843356, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09177943931864846, "standard_deviation": 0.0067035105797033975, "minimum": 0.08353462271458048, "maximum": 0.10412349021399235, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977377125468356, "standard_deviation": 0.0005190022953082091, "minimum": 0.9967002403380596, "maximum": 0.9982525273433883, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6651299177753589, "standard_deviation": 0.04274018756710546, "minimum": 0.6149873651889969, "maximum": 0.7274859957077531, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4360875470172022, "standard_deviation": 0.1996796913605883, "minimum": 0.10707727924253947, "maximum": 0.652676533369513, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04640015739834185, "standard_deviation": 0.004983329845892839, "minimum": 0.04067630020508455, "maximum": 0.0555664229097151, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_260-123440-0020", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.0344855785369873, "predicted_mos": 2.0344855785369873, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "0cdb6a47585e1f3a2b2b4261651df6da30c543e9b802919d1decb061919132e1", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.950100699905306, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.310437062300642, "全帯域_10ms_rms_envelope_nrmse": 0.0896419238971087, "全帯域_10ms_envelope_correlation": 0.9924173978416362, "全帯域_10ms_modulation_mae_db": 0.3377866587452499, "全帯域_40ms_log_rms_mae_db": 0.6271383545095685, "全帯域_40ms_rms_envelope_nrmse": 0.04839879643310693, "全帯域_40ms_envelope_correlation": 0.9979714656279703, "全帯域_40ms_modulation_mae_db": 0.5011683331188641, "300–1000 Hz_10ms_log_rms_mae_db": 1.8520518122869156, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.147145262163971, "300–1000 Hz_10ms_envelope_correlation": 0.9832110975733902, "300–1000 Hz_10ms_modulation_mae_db": 0.7613232040231036, "300–1000 Hz_40ms_log_rms_mae_db": 1.08297483062997, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08558469768182292, "300–1000 Hz_40ms_envelope_correlation": 0.9946329087812342, "300–1000 Hz_40ms_modulation_mae_db": 0.8774749434277505, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9623564135218108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17276790092189637, "1000–3000 Hz_10ms_envelope_correlation": 0.9805518199343441, "1000–3000 Hz_10ms_modulation_mae_db": 0.7505278098166582, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1636236552411998, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11133245186290712, "1000–3000 Hz_40ms_envelope_correlation": 0.993013884362812, "1000–3000 Hz_40ms_modulation_mae_db": 0.8258065354728794, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2479301909922085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31420827825497866, "3000–6000 Hz_10ms_envelope_correlation": 0.9356231227722794, "3000–6000 Hz_10ms_modulation_mae_db": 3.1198145630200753, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5489896767265254, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2334472083383536, "3000–6000 Hz_40ms_envelope_correlation": 0.9619500192014214, "3000–6000 Hz_40ms_modulation_mae_db": 4.432043940916625, "mr_stft_db": 6.08925830051652, "band_mae_db": 0.25788991170312475, "rms_gain_error_db": -0.1920434388053366, "output_peak": 0.45772498083025864}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3091561932036604, "全帯域_10ms_rms_envelope_nrmse": 0.09162490626980162, "全帯域_10ms_envelope_correlation": 0.9917821234037505, "全帯域_10ms_modulation_mae_db": 0.2664417052150234, "全帯域_40ms_log_rms_mae_db": 0.6922497399253713, "全帯域_40ms_rms_envelope_nrmse": 0.046083106982566804, "全帯域_40ms_envelope_correlation": 0.99784699283318, "全帯域_40ms_modulation_mae_db": 0.20282993522397585, "300–1000 Hz_10ms_log_rms_mae_db": 1.945274830896593, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14817358952588616, "300–1000 Hz_10ms_envelope_correlation": 0.9824501279576215, "300–1000 Hz_10ms_modulation_mae_db": 0.6473334264990559, "300–1000 Hz_40ms_log_rms_mae_db": 1.152823151209076, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08118272331249318, "300–1000 Hz_40ms_envelope_correlation": 0.9945111575053287, "300–1000 Hz_40ms_modulation_mae_db": 0.8512713192314453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8621476791122213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1666365191956889, "1000–3000 Hz_10ms_envelope_correlation": 0.9807782350185451, "1000–3000 Hz_10ms_modulation_mae_db": 0.29639421410151084, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0793184793790818, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09609088759433526, "1000–3000 Hz_40ms_envelope_correlation": 0.9931463870333849, "1000–3000 Hz_40ms_modulation_mae_db": 0.434147511902145, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2157102966596836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3088094319514379, "3000–6000 Hz_10ms_envelope_correlation": 0.9371995872215465, "3000–6000 Hz_10ms_modulation_mae_db": 3.355538658852625, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5442208721958741, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23562978933917184, "3000–6000 Hz_40ms_envelope_correlation": 0.9610014942097178, "3000–6000 Hz_40ms_modulation_mae_db": 4.578071309058669, "mr_stft_db": 6.077813192648699, "band_mae_db": 0.11251612675872276, "rms_gain_error_db": -0.09886711333021457, "output_peak": 0.46266636635725417}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.308973876164866, "全帯域_10ms_rms_envelope_nrmse": 0.0926890337058633, "全帯域_10ms_envelope_correlation": 0.9915279958133915, "全帯域_10ms_modulation_mae_db": 0.48332915895182804, "全帯域_40ms_log_rms_mae_db": 0.6207938297080338, "全帯域_40ms_rms_envelope_nrmse": 0.04528003006023159, "全帯域_40ms_envelope_correlation": 0.9978153306063136, "全帯域_40ms_modulation_mae_db": 0.7637155273279094, "300–1000 Hz_10ms_log_rms_mae_db": 1.8612604957575094, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1510244486862333, "300–1000 Hz_10ms_envelope_correlation": 0.9817692003588684, "300–1000 Hz_10ms_modulation_mae_db": 0.736049614698921, "300–1000 Hz_40ms_log_rms_mae_db": 1.1617267916438625, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08360137099985043, "300–1000 Hz_40ms_envelope_correlation": 0.9938578550245818, "300–1000 Hz_40ms_modulation_mae_db": 1.0627085557949645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9479527802753944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15709909709407174, "1000–3000 Hz_10ms_envelope_correlation": 0.9827334057085401, "1000–3000 Hz_10ms_modulation_mae_db": 0.6087538568333223, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0816160708633493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07730257080843735, "1000–3000 Hz_40ms_envelope_correlation": 0.9955612161201735, "1000–3000 Hz_40ms_modulation_mae_db": 0.5266516168320284, "3000–6000 Hz_10ms_log_rms_mae_db": 2.264751246599994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30608982622833075, "3000–6000 Hz_10ms_envelope_correlation": 0.9389342326100372, "3000–6000 Hz_10ms_modulation_mae_db": 3.2073279582211254, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5813159267474415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22696441638633216, "3000–6000 Hz_40ms_envelope_correlation": 0.9641263824161662, "3000–6000 Hz_40ms_modulation_mae_db": 4.480436091525562, "mr_stft_db": 6.093295061011934, "band_mae_db": 0.15889774874452942, "rms_gain_error_db": -0.03911232617864241, "output_peak": 0.4626026166781899}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2700079263703035, "全帯域_10ms_rms_envelope_nrmse": 0.0877667409092501, "全帯域_10ms_envelope_correlation": 0.9924382151002709, "全帯域_10ms_modulation_mae_db": 0.5418009431073997, "全帯域_40ms_log_rms_mae_db": 0.6635441019488495, "全帯域_40ms_rms_envelope_nrmse": 0.04803342989115089, "全帯域_40ms_envelope_correlation": 0.9976013274207014, "全帯域_40ms_modulation_mae_db": 0.3366275056022186, "300–1000 Hz_10ms_log_rms_mae_db": 1.9933723843191884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1548359431329937, "300–1000 Hz_10ms_envelope_correlation": 0.9808456866522515, "300–1000 Hz_10ms_modulation_mae_db": 0.9204491817175074, "300–1000 Hz_40ms_log_rms_mae_db": 1.2662816979499707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10025939060955619, "300–1000 Hz_40ms_envelope_correlation": 0.9914449309363982, "300–1000 Hz_40ms_modulation_mae_db": 0.9212822483334954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.953928155132201, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1549586392511421, "1000–3000 Hz_10ms_envelope_correlation": 0.9832321266077594, "1000–3000 Hz_10ms_modulation_mae_db": 0.8426872035199336, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1730283917051505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09060541148253849, "1000–3000 Hz_40ms_envelope_correlation": 0.993860843850615, "1000–3000 Hz_40ms_modulation_mae_db": 0.9741837758301437, "3000–6000 Hz_10ms_log_rms_mae_db": 2.304736199286552, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3142990806337229, "3000–6000 Hz_10ms_envelope_correlation": 0.9384468003400436, "3000–6000 Hz_10ms_modulation_mae_db": 3.077266707450532, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6029222881141179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23222383624224083, "3000–6000 Hz_40ms_envelope_correlation": 0.9645289149481109, "3000–6000 Hz_40ms_modulation_mae_db": 4.3775098743027945, "mr_stft_db": 6.106728827831545, "band_mae_db": 0.10898346278093857, "rms_gain_error_db": -0.0823733690994998, "output_peak": 0.45879543978059123}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2987574524007695, "全帯域_10ms_rms_envelope_nrmse": 0.09016092931682933, "全帯域_10ms_envelope_correlation": 0.9920313174008608, "全帯域_10ms_modulation_mae_db": 0.11912743209267235, "全帯域_40ms_log_rms_mae_db": 0.6280055420908156, "全帯域_40ms_rms_envelope_nrmse": 0.04189397844214653, "全帯域_40ms_envelope_correlation": 0.9982104872292641, "全帯域_40ms_modulation_mae_db": 0.2019479107107747, "300–1000 Hz_10ms_log_rms_mae_db": 1.8370766092254422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1417640426299757, "300–1000 Hz_10ms_envelope_correlation": 0.9838677431283207, "300–1000 Hz_10ms_modulation_mae_db": 0.4052956303180051, "300–1000 Hz_40ms_log_rms_mae_db": 1.008586467758187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07592205673415714, "300–1000 Hz_40ms_envelope_correlation": 0.9949853805229004, "300–1000 Hz_40ms_modulation_mae_db": 0.6650232359900953, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0003446353401766, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1698236377358254, "1000–3000 Hz_10ms_envelope_correlation": 0.9797816218916038, "1000–3000 Hz_10ms_modulation_mae_db": 0.5092568717781153, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1512021498178613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07960738125139045, "1000–3000 Hz_40ms_envelope_correlation": 0.9953815868833422, "1000–3000 Hz_40ms_modulation_mae_db": 0.20160521851088717, "3000–6000 Hz_10ms_log_rms_mae_db": 2.225030701733534, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30183922810959163, "3000–6000 Hz_10ms_envelope_correlation": 0.9433696663087261, "3000–6000 Hz_10ms_modulation_mae_db": 3.2242762351350964, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5651137570032103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22685418189714865, "3000–6000 Hz_40ms_envelope_correlation": 0.9662028700658981, "3000–6000 Hz_40ms_modulation_mae_db": 4.490413219733767, "mr_stft_db": 6.086074892960274, "band_mae_db": 0.10745137994194083, "rms_gain_error_db": -0.08585026350321927, "output_peak": 0.45231451809859047}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2966252712529254, "全帯域_10ms_rms_envelope_nrmse": 0.09098327283371865, "全帯域_10ms_envelope_correlation": 0.991855081714081, "全帯域_10ms_modulation_mae_db": 0.17244630340119316, "全帯域_40ms_log_rms_mae_db": 0.6738420279370348, "全帯域_40ms_rms_envelope_nrmse": 0.0455380278684282, "全帯域_40ms_envelope_correlation": 0.9978322534006744, "全帯域_40ms_modulation_mae_db": 0.43563043019832676, "300–1000 Hz_10ms_log_rms_mae_db": 1.9367591763424785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14383369293749645, "300–1000 Hz_10ms_envelope_correlation": 0.983465964461699, "300–1000 Hz_10ms_modulation_mae_db": 0.6754902414517835, "300–1000 Hz_40ms_log_rms_mae_db": 1.2101859679953604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08335236587959814, "300–1000 Hz_40ms_envelope_correlation": 0.9939131164116429, "300–1000 Hz_40ms_modulation_mae_db": 0.8309796376383091, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9829145244260489, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1609361022982278, "1000–3000 Hz_10ms_envelope_correlation": 0.9822481153706759, "1000–3000 Hz_10ms_modulation_mae_db": 0.13676720903796485, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1503066300766742, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09293606089729207, "1000–3000 Hz_40ms_envelope_correlation": 0.9943379914556939, "1000–3000 Hz_40ms_modulation_mae_db": 0.49299624991489915, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2670575391349046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29935647251097447, "3000–6000 Hz_10ms_envelope_correlation": 0.945945019544868, "3000–6000 Hz_10ms_modulation_mae_db": 3.1931119935526673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5733297099498003, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22675150745481287, "3000–6000 Hz_40ms_envelope_correlation": 0.9676405757744795, "3000–6000 Hz_40ms_modulation_mae_db": 4.501797217848905, "mr_stft_db": 6.085135397375866, "band_mae_db": 0.18811918943128847, "rms_gain_error_db": -0.07130014983770654, "output_peak": 0.4739980698534624}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2756064914878573, "全帯域_10ms_rms_envelope_nrmse": 0.08948674803975347, "全帯域_10ms_envelope_correlation": 0.99221531351289, "全帯域_10ms_modulation_mae_db": 0.47368448871899954, "全帯域_40ms_log_rms_mae_db": 0.6595839737487833, "全帯域_40ms_rms_envelope_nrmse": 0.04622311162283963, "全帯域_40ms_envelope_correlation": 0.9978970125047972, "全帯域_40ms_modulation_mae_db": 0.37921101186482237, "300–1000 Hz_10ms_log_rms_mae_db": 1.9764455913408894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1534903279130649, "300–1000 Hz_10ms_envelope_correlation": 0.9811425572756477, "300–1000 Hz_10ms_modulation_mae_db": 0.37618680794841813, "300–1000 Hz_40ms_log_rms_mae_db": 1.30233075224948, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09391058930558575, "300–1000 Hz_40ms_envelope_correlation": 0.9925300322016323, "300–1000 Hz_40ms_modulation_mae_db": 0.7767117462952867, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8990014087309566, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1635412798828695, "1000–3000 Hz_10ms_envelope_correlation": 0.981245110966642, "1000–3000 Hz_10ms_modulation_mae_db": 0.32101522149992145, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0740867061422936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08769256718196587, "1000–3000 Hz_40ms_envelope_correlation": 0.9942786229421648, "1000–3000 Hz_40ms_modulation_mae_db": 0.22170241773025234, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2062358533699364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3082090683953522, "3000–6000 Hz_10ms_envelope_correlation": 0.9385361239792112, "3000–6000 Hz_10ms_modulation_mae_db": 3.1976062285236946, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5314652127342987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23040873696824063, "3000–6000 Hz_40ms_envelope_correlation": 0.9632953039890808, "3000–6000 Hz_40ms_modulation_mae_db": 4.393772070339669, "mr_stft_db": 6.0911624117360015, "band_mae_db": 0.15897415215170652, "rms_gain_error_db": -0.12384790989595011, "output_peak": 0.44294502025023125}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2779907421572652, "全帯域_10ms_rms_envelope_nrmse": 0.08346031674284393, "全帯域_10ms_envelope_correlation": 0.9932057894767629, "全帯域_10ms_modulation_mae_db": 0.5464093825496573, "全帯域_40ms_log_rms_mae_db": 0.646181437865325, "全帯域_40ms_rms_envelope_nrmse": 0.043474600205736416, "全帯域_40ms_envelope_correlation": 0.9981032284951655, "全帯域_40ms_modulation_mae_db": 0.650756237942025, "300–1000 Hz_10ms_log_rms_mae_db": 1.9044155992395648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1487063872358036, "300–1000 Hz_10ms_envelope_correlation": 0.9822754704473509, "300–1000 Hz_10ms_modulation_mae_db": 0.6697572634621985, "300–1000 Hz_40ms_log_rms_mae_db": 1.2517837589964547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09579280635898757, "300–1000 Hz_40ms_envelope_correlation": 0.9920591988844449, "300–1000 Hz_40ms_modulation_mae_db": 0.7962664372318963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9360277081367452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17236248373348834, "1000–3000 Hz_10ms_envelope_correlation": 0.979335382071435, "1000–3000 Hz_10ms_modulation_mae_db": 0.5526011292692226, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1032887011436432, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09705077760601276, "1000–3000 Hz_40ms_envelope_correlation": 0.9933760830894861, "1000–3000 Hz_40ms_modulation_mae_db": 0.9176037383973634, "3000–6000 Hz_10ms_log_rms_mae_db": 2.278486736584321, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30976351666219637, "3000–6000 Hz_10ms_envelope_correlation": 0.9369316773890106, "3000–6000 Hz_10ms_modulation_mae_db": 3.1737375408262225, "3000–6000 Hz_40ms_log_rms_mae_db": 1.596822951178998, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2347495616599275, "3000–6000 Hz_40ms_envelope_correlation": 0.9612152681700825, "3000–6000 Hz_40ms_modulation_mae_db": 4.484482392970653, "mr_stft_db": 6.094995554124666, "band_mae_db": 0.14340952052376912, "rms_gain_error_db": -0.10133132761972447, "output_peak": 0.4576880368074884}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79920, "output_samples": 79920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.247066730737792, "spectral_convergence": 0.26286259230946396, "spectral_convergence_status": "ok", "amplitude_floor": 1.5443736044991425e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.923041070978857, "spectral_convergence": 0.2544958929406846, "spectral_convergence_status": "ok", "amplitude_floor": 1.1257392826698756e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.09766709983291, "spectral_convergence": 0.2522244493530525, "spectral_convergence_status": "ok", "amplitude_floor": 1.1201855379193364e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001035002994445706, "output_power": 0.0010234738202644842, "delta_db": -0.04864868089762588, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0032595449236567883, "output_power": 0.0030969749687357684, "delta_db": -0.2221927575423382, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007961593328208376, "output_power": 0.0007699929947636308, "delta_db": -0.14513216331905754, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023616410570225164, "output_power": 0.0002013054970184597, "delta_db": -0.6935825613476188, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.38734826098075e-05, "output_power": 5.4356307170152404e-05, "delta_db": 0.03874893109582101, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3554110211771778e-05, "output_power": 1.4858482266438358e-05, "delta_db": 0.3990343760162869, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25788991170312475, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1920434388053366, "pitch": null, "output_rms_dbfs": -22.872693327058954, "output_peak": 0.45772498083025864, "reference_rms_dbfs": -22.68064988825362, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.08925830051652}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001944445221026879, "log_rms_mae_db": 1.310437062300642, "rms_envelope_nrmse": 0.0896419238971087, "envelope_correlation": 0.9924173978416362, "rms_gain_error_db": -0.1920429137292494, "reference_mean_power": 0.005399702914122853, "output_mean_power": 0.005166132904514954}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 2.8891675057826967e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000788720956533661, "output_power": 0.0008227578234214702, "reference_active": true, "delta_db": 0.18348640292182955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001646344464707057, "output_power": 0.00016492924367537872, "reference_active": true, "delta_db": 0.007769595816845564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.1066196116820634e-05, "output_power": 4.2259379020818315e-05, "reference_active": true, "delta_db": -0.8221039774970745}], "active_band_count": 3, "active_band_mae_db": 0.3377866587452499}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00017595749087279735, "log_rms_mae_db": 0.6271383545095685, "rms_envelope_nrmse": 0.04839879643310693, "envelope_correlation": 0.9979714656279703, "rms_gain_error_db": -0.19204267714597412, "reference_mean_power": 0.005432355653378944, "output_mean_power": 0.0051973734972264925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 2.6677500098154174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007158752148089715, "output_power": 0.0007436008081532732, "reference_active": true, "delta_db": 0.16502526658079825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.294187442575086e-05, "output_power": 7.116105241359304e-05, "reference_active": true, "delta_db": -0.10734557096591957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7590528739181784e-06, "output_power": 2.8311639407677983e-06, "reference_active": true, "delta_db": -1.2311341618098746}], "active_band_count": 3, "active_band_mae_db": 0.5011683331188641}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001659932762873133, "log_rms_mae_db": 1.8520518122869156, "rms_envelope_nrmse": 0.147145262163971, "envelope_correlation": 0.9832110975733902, "rms_gain_error_db": -0.3229468667966299, "reference_mean_power": 0.0017617711849040457, "output_mean_power": 0.0016355160704129667}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.325463642910379e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006157322295664234, "output_power": 0.0005849678988939963, "reference_active": true, "delta_db": -0.2225985258062232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015692982702212587, "output_power": 0.00014405071455716, "reference_active": true, "delta_db": -0.37190079120583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0206834209715196e-05, "output_power": 2.7249143100102212e-05, "reference_active": true, "delta_db": -1.6894702950572578}], "active_band_count": 3, "active_band_mae_db": 0.7613232040231036}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00014319111528307625, "log_rms_mae_db": 1.08297483062997, "rms_envelope_nrmse": 0.08558469768182292, "envelope_correlation": 0.9946329087812342, "rms_gain_error_db": -0.3229472631658417, "reference_mean_power": 0.0017724260962696937, "output_mean_power": 0.0016454072607945084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.2006657592514347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005739716253322745, "output_power": 0.000547672441361401, "reference_active": true, "delta_db": -0.2036953532356412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.210701544963183e-05, "output_power": 6.259779935141131e-05, "reference_active": true, "delta_db": -0.6141845454034482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.009713658025178e-06, "output_power": 2.6403339906745137e-06, "reference_active": true, "delta_db": -1.814544931644162}], "active_band_count": 3, "active_band_mae_db": 0.8774749434277505}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.00012416177010932925, "log_rms_mae_db": 1.9623564135218108, "rms_envelope_nrmse": 0.17276790092189637, "envelope_correlation": 0.9805518199343441, "rms_gain_error_db": -0.5150624148688999, "reference_mean_power": 0.0002459141129048244, "output_mean_power": 0.0002184123602004188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.7078032497142614e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.578711086190898e-05, "output_power": 8.476216552102985e-05, "reference_active": true, "delta_db": -0.5310503060943094}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5502167493014184e-05, "output_power": 1.3042133365526984e-05, "reference_active": true, "delta_db": -0.7504378802999174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901592249182956e-06, "output_power": 4.720192122558605e-06, "reference_active": true, "delta_db": -0.9700952430557481}], "active_band_count": 3, "active_band_mae_db": 0.7505278098166582}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 9.745648796239735e-05, "log_rms_mae_db": 1.1636236552411998, "rms_envelope_nrmse": 0.11133245186290712, "envelope_correlation": 0.993013884362812, "rms_gain_error_db": -0.515063754559244, "reference_mean_power": 0.00024740043253105237, "output_mean_power": 0.00021973238980423573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.5717362127850378e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.047075579844295e-05, "output_power": 8.01226189070794e-05, "reference_active": true, "delta_db": -0.5275308214873213}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.5496928110188975e-06, "output_power": 6.368637156744631e-06, "reference_active": true, "delta_db": -0.7388277467367859}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.070423026829593e-07, "output_power": 3.0798774970950516e-07, "reference_active": true, "delta_db": -1.2110610381945313}], "active_band_count": 3, "active_band_mae_db": 0.8258065354728794}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 3.6263821941295626e-05, "log_rms_mae_db": 2.2479301909922085, "rms_envelope_nrmse": 0.31420827825497866, "envelope_correlation": 0.9356231227722794, "rms_gain_error_db": 0.1443905477355495, "reference_mean_power": 2.0515613806993198e-05, "output_mean_power": 2.120916495880951e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 4.904304308942738e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.621343716362641e-06, "output_power": 2.059571336744191e-06, "reference_active": true, "delta_db": 1.0390174631763431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.453801656353405e-07, "output_power": 9.255331504501391e-07, "reference_active": true, "delta_db": 0.9401414667540995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.409211497093947e-07, "output_power": 1.317963560580161e-06, "reference_active": true, "delta_db": 7.3802847591297835}], "active_band_count": 3, "active_band_mae_db": 3.1198145630200753}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 2.5114070553475023e-05, "log_rms_mae_db": 1.5489896767265254, "rms_envelope_nrmse": 0.2334472083383536, "envelope_correlation": 0.9619500192014214, "rms_gain_error_db": 0.1443838208647383, "reference_mean_power": 2.063955290066712e-05, "output_mean_power": 2.133726088964612e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 4.623298998188318e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6281370009575077e-06, "output_power": 2.6359726242534645e-06, "reference_active": true, "delta_db": 2.0924994943900996}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7397133003103647e-07, "output_power": 7.079553071697721e-07, "reference_active": true, "delta_db": 2.771675328618543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7174687173869334e-08, "output_power": 1.196973499402914e-07, "reference_active": true, "delta_db": 8.431956999741232}], "active_band_count": 3, "active_band_mae_db": 4.432043940916625}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8437316547927324, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0ed18dd3cad39d74e5f62b27f0978ed30c8a7b1b7504d981e4070e9fe935d390", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9812382271961932, "standard_deviation": 0.0013967503853728272, "minimum": 0.979335382071435, "maximum": 0.9832321266077594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9430841630844444, "standard_deviation": 0.04459617508736664, "minimum": 1.8621476791122213, "maximum": 2.0003446353401766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5022504394820811, "standard_deviation": 0.23912551527618706, "minimum": 0.13676720903796485, "maximum": 0.8426872035199336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16476570751415126, "standard_deviation": 0.006781796065512595, "minimum": 0.1549586392511421, "maximum": 0.17276790092189637, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941195769672091, "standard_deviation": 0.0009660947706585872, "minimum": 0.993013884362812, "maximum": 0.9955612161201735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1220588480461569, "standard_deviation": 0.04155684395999817, "minimum": 1.0740867061422936, "maximum": 1.1730283917051505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5743371330738247, "standard_deviation": 0.30065876733137614, "minimum": 0.20160521851088717, "maximum": 0.9741837758301437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09157726358560993, "standard_deviation": 0.010727309948150595, "minimum": 0.07730257080843735, "maximum": 0.11133245186290712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9393732787707153, "standard_deviation": 0.003498180432860733, "minimum": 0.9356231227722794, "maximum": 0.945945019544868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.251242345545142, "standard_deviation": 0.033853869713813656, "minimum": 2.2062358533699364, "maximum": 2.304736199286552, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.1935849856977545, "standard_deviation": 0.08171566263047439, "minimum": 3.077266707450532, "maximum": 3.355538658852625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30782186284332314, "standard_deviation": 0.005319648151920516, "minimum": 0.29935647251097447, "maximum": 0.3142990806337229, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9637451035968696, "standard_deviation": 0.002370267034535361, "minimum": 0.9610014942097178, "maximum": 0.9676405757744795, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5680225493312832, "standard_deviation": 0.025427782209283788, "minimum": 1.5314652127342987, "maximum": 1.6029222881141179, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 4.467315764587081, "standard_deviation": 0.06455779717418288, "minimum": 4.3775098743027945, "maximum": 4.578071309058669, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23087865478577851, "standard_deviation": 0.003678819535079767, "minimum": 0.22675150745481287, "maximum": 0.23562978933917184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9823784809818937, "standard_deviation": 0.001092683828905502, "minimum": 0.9808456866522515, "maximum": 0.9838677431283207, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9133320624260728, "standard_deviation": 0.05894313889052578, "minimum": 1.8370766092254422, "maximum": 1.9933723843191884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.648985671264874, "standard_deviation": 0.18080495651359002, "minimum": 0.37618680794841813, "maximum": 0.9204491817175074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1486217117781781, "standard_deviation": 0.004479060433422065, "minimum": 0.1417640426299757, "maximum": 0.1548359431329937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9934918225335204, "standard_deviation": 0.0013120606348218263, "minimum": 0.9914449309363982, "maximum": 0.9949853805229004, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1795866773040453, "standard_deviation": 0.09880793212957097, "minimum": 1.008586467758187, "maximum": 1.30233075224948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8477147654929054, "standard_deviation": 0.11580614364258597, "minimum": 0.6650232359900953, "maximum": 1.0627085557949645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08745075011025642, "standard_deviation": 0.008303839488262594, "minimum": 0.07592205673415714, "maximum": 0.10025939060955619, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15453018650450256, "standard_deviation": 0.05075156132549328, "minimum": 0.10745137994194083, "maximum": 0.25788991170312475, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.090557954775688, "standard_deviation": 0.008474049268191574, "minimum": 6.077813192648699, "maximum": 6.106728827831545, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4585918810820083, "standard_deviation": 0.008911289062754449, "minimum": 0.44294502025023125, "maximum": 0.4739980698534624, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09934073728378673, "standard_deviation": 0.044852455997203826, "minimum": -0.1920434388053366, "maximum": -0.03911232617864241, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9921841542829555, "standard_deviation": 0.0005192686418038846, "minimum": 0.9915279958133915, "maximum": 0.9932057894767629, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2934443769172863, "standard_deviation": 0.016567597172249426, "minimum": 1.2700079263703035, "maximum": 1.310437062300642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3676282590977529, "standard_deviation": 0.16814591141385443, "minimum": 0.11912743209267235, "maximum": 0.5464093825496573, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08947673396439615, "standard_deviation": 0.0028490609218734396, "minimum": 0.08346031674284393, "maximum": 0.0926890337058633, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979097622647584, "standard_deviation": 0.00018740009326610017, "minimum": 0.9976013274207014, "maximum": 0.9982104872292641, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6514173759667227, "standard_deviation": 0.025317129284426254, "minimum": 0.6207938297080338, "maximum": 0.6922497399253713, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43398586149861457, "standard_deviation": 0.1999931585270892, "minimum": 0.2019479107107747, "maximum": 0.7637155273279094, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045615635188275876, "standard_deviation": 0.0021624288152617146, "minimum": 0.04189397844214653, "maximum": 0.04839879643310693, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2830-3980-0000", "language": "en", "style": "normal", "speaker": "librispeech_2830", "source_utterance_id": "2830-3980-0000", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 59680, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2830-3980-0000.wav", "source_file_sha256": "bd51f8c8974eebc1fa2d68a089b0b8c3adab983570542f2ac3443ca689ac701c", "decoded_pcm_sha256": "8f71058a78243905dd373f5302bc82f25ff2a7e7631413b5ce924104f0f14bea", "original_source_sha256": "a73adb84b9560fdfc8f6c7f5801e567267ac8beb6fd51df4f5dec7217eb43ad8", "corpus_entry_id": "librispeech_2830-3980-0000"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.7120254039764404, "predicted_mos": 3.7120254039764404, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "8f63ce651ea69baf52fe115eb6d0f6053551b53ce699899b6417e104cd136b1d", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.100034063001658, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2830-3980-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "8f63ce651ea69baf52fe115eb6d0f6053551b53ce699899b6417e104cd136b1d", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "fd97b4748ed6a2437861ffee06416e2dbd09229870db6e19899ec2d39491809a"}, "rows": [{"case": "librispeech_2830-3980-0000", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.7048709392547607, "predicted_mos": 2.7048709392547607, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "86473689a844682efebfd0f27c42973665824d05fb87f21f6a99370bdb876e72", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.279120926999894, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2830-3980-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "86473689a844682efebfd0f27c42973665824d05fb87f21f6a99370bdb876e72", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.083501141404912, "全帯域_10ms_rms_envelope_nrmse": 0.06013482689746816, "全帯域_10ms_envelope_correlation": 0.9965956953131183, "全帯域_10ms_modulation_mae_db": 0.2636380883956992, "全帯域_40ms_log_rms_mae_db": 0.6061985705962667, "全帯域_40ms_rms_envelope_nrmse": 0.03281708638094683, "全帯域_40ms_envelope_correlation": 0.9991806814189281, "全帯域_40ms_modulation_mae_db": 0.20513496070071738, "300–1000 Hz_10ms_log_rms_mae_db": 1.5423431525829354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08472147719816023, "300–1000 Hz_10ms_envelope_correlation": 0.9944325881830304, "300–1000 Hz_10ms_modulation_mae_db": 0.4623893192083573, "300–1000 Hz_40ms_log_rms_mae_db": 0.9426393992366696, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05097014492151829, "300–1000 Hz_40ms_envelope_correlation": 0.997922576743686, "300–1000 Hz_40ms_modulation_mae_db": 0.9869972790395227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6703276292771434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17629518568327873, "1000–3000 Hz_10ms_envelope_correlation": 0.980142437763523, "1000–3000 Hz_10ms_modulation_mae_db": 0.42317164537150287, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0858400336259741, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11865687686210448, "1000–3000 Hz_40ms_envelope_correlation": 0.9930353416208488, "1000–3000 Hz_40ms_modulation_mae_db": 0.693876257939213, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9372676252619343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3073129710309251, "3000–6000 Hz_10ms_envelope_correlation": 0.937183516991418, "3000–6000 Hz_10ms_modulation_mae_db": 1.8481357102875873, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5349744114835528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25990136693934374, "3000–6000 Hz_40ms_envelope_correlation": 0.9525065928779299, "3000–6000 Hz_40ms_modulation_mae_db": 2.362989968245047, "mr_stft_db": 5.609850308494974, "band_mae_db": 0.3889402520377929, "rms_gain_error_db": -0.1603763580523536, "output_peak": 0.9656757190888365}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2223539739499862, "全帯域_10ms_rms_envelope_nrmse": 0.06847542634950854, "全帯域_10ms_envelope_correlation": 0.9952769381203989, "全帯域_10ms_modulation_mae_db": 0.5396875742607313, "全帯域_40ms_log_rms_mae_db": 0.7046880936316177, "全帯域_40ms_rms_envelope_nrmse": 0.04088545730189477, "全帯域_40ms_envelope_correlation": 0.9982755586108473, "全帯域_40ms_modulation_mae_db": 0.29547138636069475, "300–1000 Hz_10ms_log_rms_mae_db": 1.5352465853412331, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09837539209689049, "300–1000 Hz_10ms_envelope_correlation": 0.9924089447163773, "300–1000 Hz_10ms_modulation_mae_db": 0.9985919716791661, "300–1000 Hz_40ms_log_rms_mae_db": 0.9598932211384428, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05410185678748455, "300–1000 Hz_40ms_envelope_correlation": 0.9976329927173802, "300–1000 Hz_40ms_modulation_mae_db": 1.0658414736575326, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7469105797366065, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16669420749277913, "1000–3000 Hz_10ms_envelope_correlation": 0.979914432661045, "1000–3000 Hz_10ms_modulation_mae_db": 0.4525857068479335, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1073146174297868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08898154450023642, "1000–3000 Hz_40ms_envelope_correlation": 0.99414605687467, "1000–3000 Hz_40ms_modulation_mae_db": 0.5866877337790991, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8928175061538548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30186792839215165, "3000–6000 Hz_10ms_envelope_correlation": 0.937328845620058, "3000–6000 Hz_10ms_modulation_mae_db": 1.4679127082126764, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4207084728876767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24608325952777874, "3000–6000 Hz_40ms_envelope_correlation": 0.9560223538981177, "3000–6000 Hz_40ms_modulation_mae_db": 2.05307910461512, "mr_stft_db": 5.593069592125896, "band_mae_db": 0.15689851803608226, "rms_gain_error_db": -0.08389565074625141, "output_peak": 0.9821627434152921}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1981767792432392, "全帯域_10ms_rms_envelope_nrmse": 0.07681555479561458, "全帯域_10ms_envelope_correlation": 0.9940906030000892, "全帯域_10ms_modulation_mae_db": 0.15457451766699334, "全帯域_40ms_log_rms_mae_db": 0.6627235728758832, "全帯域_40ms_rms_envelope_nrmse": 0.03994762026036907, "全帯域_40ms_envelope_correlation": 0.9984299401394044, "全帯域_40ms_modulation_mae_db": 0.19653962900012603, "300–1000 Hz_10ms_log_rms_mae_db": 1.521437426712907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09821765964633485, "300–1000 Hz_10ms_envelope_correlation": 0.9923800919127089, "300–1000 Hz_10ms_modulation_mae_db": 0.4314433599129712, "300–1000 Hz_40ms_log_rms_mae_db": 0.9083397496442304, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04843455760317351, "300–1000 Hz_40ms_envelope_correlation": 0.9980696460192684, "300–1000 Hz_40ms_modulation_mae_db": 1.0673309113979759, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6880797749419114, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15628608579967843, "1000–3000 Hz_10ms_envelope_correlation": 0.9864102585414917, "1000–3000 Hz_10ms_modulation_mae_db": 0.8862220793884231, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0324368858970727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10918853748738995, "1000–3000 Hz_40ms_envelope_correlation": 0.9960265860509672, "1000–3000 Hz_40ms_modulation_mae_db": 0.8009603436129087, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9060152360174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2952682340713875, "3000–6000 Hz_10ms_envelope_correlation": 0.9400908742503138, "3000–6000 Hz_10ms_modulation_mae_db": 1.4758272304533868, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5086943180241903, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24491397998391537, "3000–6000 Hz_40ms_envelope_correlation": 0.9565312864982977, "3000–6000 Hz_40ms_modulation_mae_db": 2.0322387190586766, "mr_stft_db": 5.598898407255853, "band_mae_db": 0.29697573027186963, "rms_gain_error_db": -0.11391842465289623, "output_peak": 0.9453421016583393}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.088976485957185, "全帯域_10ms_rms_envelope_nrmse": 0.06509744547405484, "全帯域_10ms_envelope_correlation": 0.996066628821184, "全帯域_10ms_modulation_mae_db": 0.4231404261626042, "全帯域_40ms_log_rms_mae_db": 0.6131781139871973, "全帯域_40ms_rms_envelope_nrmse": 0.04075462410053844, "全帯域_40ms_envelope_correlation": 0.9986666035396666, "全帯域_40ms_modulation_mae_db": 0.39241808270659445, "300–1000 Hz_10ms_log_rms_mae_db": 1.4906706548210007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09017510992352427, "300–1000 Hz_10ms_envelope_correlation": 0.9937134360167422, "300–1000 Hz_10ms_modulation_mae_db": 1.2293636915828172, "300–1000 Hz_40ms_log_rms_mae_db": 0.9318617013074018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05249085590635476, "300–1000 Hz_40ms_envelope_correlation": 0.9978487699912854, "300–1000 Hz_40ms_modulation_mae_db": 1.4472452605188748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6837823945299497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14002912835922723, "1000–3000 Hz_10ms_envelope_correlation": 0.9867379544798589, "1000–3000 Hz_10ms_modulation_mae_db": 0.8972877102193073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0349544602470413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08669985502859699, "1000–3000 Hz_40ms_envelope_correlation": 0.9955375697192594, "1000–3000 Hz_40ms_modulation_mae_db": 0.7375538185845335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9292844287318607, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2943153614264983, "3000–6000 Hz_10ms_envelope_correlation": 0.9414546035983206, "3000–6000 Hz_10ms_modulation_mae_db": 1.6408701364691265, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5149780336404945, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24736113117003908, "3000–6000 Hz_40ms_envelope_correlation": 0.9562589068268449, "3000–6000 Hz_40ms_modulation_mae_db": 2.2800496786549496, "mr_stft_db": 5.599351005845659, "band_mae_db": 0.30552926200717273, "rms_gain_error_db": -0.18798148264062065, "output_peak": 0.9502003394980983}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0018115036693014, "全帯域_10ms_rms_envelope_nrmse": 0.06467757284676759, "全帯域_10ms_envelope_correlation": 0.9958512936868862, "全帯域_10ms_modulation_mae_db": 0.22584912461465115, "全帯域_40ms_log_rms_mae_db": 0.4601504400777502, "全帯域_40ms_rms_envelope_nrmse": 0.035163024059996606, "全帯域_40ms_envelope_correlation": 0.9988180662877804, "全帯域_40ms_modulation_mae_db": 0.13584798298763687, "300–1000 Hz_10ms_log_rms_mae_db": 1.6072586390470391, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09394857629238394, "300–1000 Hz_10ms_envelope_correlation": 0.9932344518906203, "300–1000 Hz_10ms_modulation_mae_db": 0.6283592628500373, "300–1000 Hz_40ms_log_rms_mae_db": 1.0242958770370958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05845816096851934, "300–1000 Hz_40ms_envelope_correlation": 0.9974177629094838, "300–1000 Hz_40ms_modulation_mae_db": 0.8854578355240571, "1000–3000 Hz_10ms_log_rms_mae_db": 1.603678567308997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16778405181560024, "1000–3000 Hz_10ms_envelope_correlation": 0.9813678502152893, "1000–3000 Hz_10ms_modulation_mae_db": 0.42144488349828446, "1000–3000 Hz_40ms_log_rms_mae_db": 1.004291632697727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10847483915409684, "1000–3000 Hz_40ms_envelope_correlation": 0.993670044806076, "1000–3000 Hz_40ms_modulation_mae_db": 0.35269302693834254, "3000–6000 Hz_10ms_log_rms_mae_db": 1.976529753316851, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3059349246029593, "3000–6000 Hz_10ms_envelope_correlation": 0.9346226056425907, "3000–6000 Hz_10ms_modulation_mae_db": 1.7648406683902695, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5551872288791888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25861699703798274, "3000–6000 Hz_40ms_envelope_correlation": 0.9506123585617603, "3000–6000 Hz_40ms_modulation_mae_db": 2.2825883090430525, "mr_stft_db": 5.577455207895627, "band_mae_db": 0.2748444813108894, "rms_gain_error_db": -0.11134430526738508, "output_peak": 0.9652354373417342}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1761751199417272, "全帯域_10ms_rms_envelope_nrmse": 0.06856305002169126, "全帯域_10ms_envelope_correlation": 0.9953274973769601, "全帯域_10ms_modulation_mae_db": 0.08956362655249996, "全帯域_40ms_log_rms_mae_db": 0.6720602104187332, "全帯域_40ms_rms_envelope_nrmse": 0.04683155991453481, "全帯域_40ms_envelope_correlation": 0.997776935459864, "全帯域_40ms_modulation_mae_db": 0.4712975408639523, "300–1000 Hz_10ms_log_rms_mae_db": 1.5676371437664338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08854700241430022, "300–1000 Hz_10ms_envelope_correlation": 0.9938148185709774, "300–1000 Hz_10ms_modulation_mae_db": 0.6347294598152188, "300–1000 Hz_40ms_log_rms_mae_db": 0.9941857817901926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05661419646381768, "300–1000 Hz_40ms_envelope_correlation": 0.9973083777824587, "300–1000 Hz_40ms_modulation_mae_db": 1.096017282486521, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6128347207209315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18496917541175373, "1000–3000 Hz_10ms_envelope_correlation": 0.976695092325609, "1000–3000 Hz_10ms_modulation_mae_db": 0.6166550193109666, "1000–3000 Hz_40ms_log_rms_mae_db": 1.013683042558936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11949586919036909, "1000–3000 Hz_40ms_envelope_correlation": 0.9916624372752382, "1000–3000 Hz_40ms_modulation_mae_db": 0.7374182459116727, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0427681482673825, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3064897798575648, "3000–6000 Hz_10ms_envelope_correlation": 0.9372432436693, "3000–6000 Hz_10ms_modulation_mae_db": 1.8645345014580945, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5591064610963818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25900387779036727, "3000–6000 Hz_40ms_envelope_correlation": 0.9525714739607793, "3000–6000 Hz_40ms_modulation_mae_db": 2.365551792604115, "mr_stft_db": 5.590651285669083, "band_mae_db": 0.3963166111719927, "rms_gain_error_db": -0.11566534099925549, "output_peak": 1.0409052636111327}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 3.351206434316354e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.351206434316354e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0780472779218044, "全帯域_10ms_rms_envelope_nrmse": 0.05862790585353065, "全帯域_10ms_envelope_correlation": 0.9965827189508405, "全帯域_10ms_modulation_mae_db": 0.27845965116180404, "全帯域_40ms_log_rms_mae_db": 0.5648578014140345, "全帯域_40ms_rms_envelope_nrmse": 0.03316126746861509, "全帯域_40ms_envelope_correlation": 0.9989163477056586, "全帯域_40ms_modulation_mae_db": 0.2749906404894534, "300–1000 Hz_10ms_log_rms_mae_db": 1.437295169248397, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09477762322100834, "300–1000 Hz_10ms_envelope_correlation": 0.9929659348996389, "300–1000 Hz_10ms_modulation_mae_db": 1.0768243901567252, "300–1000 Hz_40ms_log_rms_mae_db": 0.943575114632045, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05495006653716893, "300–1000 Hz_40ms_envelope_correlation": 0.9975430565046235, "300–1000 Hz_40ms_modulation_mae_db": 1.4597152918191447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6486337404633096, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15103799942149707, "1000–3000 Hz_10ms_envelope_correlation": 0.9842074527452773, "1000–3000 Hz_10ms_modulation_mae_db": 0.4836285372450808, "1000–3000 Hz_40ms_log_rms_mae_db": 1.019315147607538, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08491554345896501, "1000–3000 Hz_40ms_envelope_correlation": 0.995655826947196, "1000–3000 Hz_40ms_modulation_mae_db": 0.44018303663739183, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9403176541274545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31019535446172664, "3000–6000 Hz_10ms_envelope_correlation": 0.9369658082461134, "3000–6000 Hz_10ms_modulation_mae_db": 2.6241331753359973, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5057690965584618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2565294283645989, "3000–6000 Hz_40ms_envelope_correlation": 0.9545329744403106, "3000–6000 Hz_40ms_modulation_mae_db": 3.104318295606206, "mr_stft_db": 5.613268936993798, "band_mae_db": 0.2502415906564736, "rms_gain_error_db": -0.09139989093510353, "output_peak": 1.0137462074671109}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.020072419000151, "全帯域_10ms_rms_envelope_nrmse": 0.06596831394999503, "全帯域_10ms_envelope_correlation": 0.9957380985505205, "全帯域_10ms_modulation_mae_db": 0.22534890615194994, "全帯域_40ms_log_rms_mae_db": 0.4779581944252752, "全帯域_40ms_rms_envelope_nrmse": 0.0349340866231444, "全帯域_40ms_envelope_correlation": 0.9989190365172353, "全帯域_40ms_modulation_mae_db": 0.22230475388086082, "300–1000 Hz_10ms_log_rms_mae_db": 1.4197513508552229, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09012172154791816, "300–1000 Hz_10ms_envelope_correlation": 0.9938389202230181, "300–1000 Hz_10ms_modulation_mae_db": 0.5003556491439008, "300–1000 Hz_40ms_log_rms_mae_db": 0.857067696973589, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05579173723396664, "300–1000 Hz_40ms_envelope_correlation": 0.997724620879388, "300–1000 Hz_40ms_modulation_mae_db": 0.6369888743894602, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7153449250995083, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1622994025841201, "1000–3000 Hz_10ms_envelope_correlation": 0.9835310616197173, "1000–3000 Hz_10ms_modulation_mae_db": 0.5118964181842249, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1106510791234396, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11353270566339088, "1000–3000 Hz_40ms_envelope_correlation": 0.9936241154929834, "1000–3000 Hz_40ms_modulation_mae_db": 0.8638404215162389, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8830560125810258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2892400828571387, "3000–6000 Hz_10ms_envelope_correlation": 0.94356343233448, "3000–6000 Hz_10ms_modulation_mae_db": 1.7634475455845673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4921726393323322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24661819830136045, "3000–6000 Hz_40ms_envelope_correlation": 0.9566201230111767, "3000–6000 Hz_40ms_modulation_mae_db": 2.358505981721122, "mr_stft_db": 5.599899898691112, "band_mae_db": 0.34234002711882655, "rms_gain_error_db": -0.13023387757567273, "output_peak": 0.9242332913275049}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59680, "output_samples": 59680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.758669224967529, "spectral_convergence": 0.2010070943814744, "spectral_convergence_status": "ok", "amplitude_floor": 3.189740694195829e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4447495865042645, "spectral_convergence": 0.18979139531194147, "spectral_convergence_status": "ok", "amplitude_floor": 2.205927306089595e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.62613211401313, "spectral_convergence": 0.18676301068241596, "spectral_convergence_status": "ok", "amplitude_floor": 1.9576049396551346e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004284845572396228, "output_power": 0.004370359754241673, "delta_db": 0.08582013806162642, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.012674374150346739, "output_power": 0.012016719932880074, "delta_db": -0.23140584021930952, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0031087438744962816, "output_power": 0.002994704684027339, "delta_db": -0.16230940663171964, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005542642121172655, "output_power": 0.00048792024859701323, "delta_db": -0.5536799622902591, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001593466145796079, "output_power": 0.0001389548954703287, "delta_db": -0.5946898918787389, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.162629787041768e-05, "output_power": 0.00010779379616665556, "delta_db": 0.705736273145104, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3889402520377929, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1603763580523536, "pitch": null, "output_rms_dbfs": -16.964485863930243, "output_peak": 0.9656757190888365, "reference_rms_dbfs": -16.80410950587789, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.609850308494974}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00043785897255075944, "log_rms_mae_db": 1.083501141404912, "rms_envelope_nrmse": 0.06013482689746816, "envelope_correlation": 0.9965956953131183, "rms_gain_error_db": -0.16037732249121422, "reference_mean_power": 0.0209012125553907, "output_mean_power": 0.020143445116746674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.005551294872497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00444581511703579, "output_power": 0.004277423637738609, "reference_active": true, "delta_db": -0.16769133151669052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00039454721092167534, "output_power": 0.00040136090220884574, "reference_active": true, "delta_db": 0.07436086585430417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.0001084016816849342, "output_power": 0.00012300484737271975, "reference_active": true, "delta_db": 0.5488620678161029}], "active_band_count": 3, "active_band_mae_db": 0.2636380883956992}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0004107294113879301, "log_rms_mae_db": 0.6061985705962667, "rms_envelope_nrmse": 0.03281708638094683, "envelope_correlation": 0.9991806814189281, "rms_gain_error_db": -0.16037588622237287, "reference_mean_power": 0.021070817398047652, "output_mean_power": 0.020306907700463398}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.012296407388435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00391054354874771, "output_power": 0.003781228449545148, "reference_active": true, "delta_db": -0.14604209840478516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019349823972573515, "output_power": 0.00019432004510060025, "reference_active": true, "delta_db": 0.01840584045052721}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9805555260958754e-06, "output_power": 5.390708306318692e-06, "reference_active": true, "delta_db": -0.45095694324683977}], "active_band_count": 3, "active_band_mae_db": 0.20513496070071738}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.0003730035430310483, "log_rms_mae_db": 1.5423431525829354, "rms_envelope_nrmse": 0.08472147719816023, "envelope_correlation": 0.9944325881830304, "rms_gain_error_db": -0.0817889334936584, "reference_mean_power": 0.007423998211444209, "output_mean_power": 0.007285493335178682}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.725628531347034e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002177971361103236, "output_power": 0.0020623368944547907, "reference_active": true, "delta_db": -0.23692553554311452}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00032785256002961345, "output_power": 0.0002937205572346815, "reference_active": true, "delta_db": -0.47744235732462814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.051887548388713e-05, "output_power": 6.896324593232448e-05, "reference_active": true, "delta_db": -0.6728000647573293}], "active_band_count": 3, "active_band_mae_db": 0.4623893192083573}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00035212990311193784, "log_rms_mae_db": 0.9426393992366696, "rms_envelope_nrmse": 0.05097014492151829, "envelope_correlation": 0.997922576743686, "rms_gain_error_db": -0.08178881694744969, "reference_mean_power": 0.007484242812710004, "output_mean_power": 0.007344614187818853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.215922155621447e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018646593945503163, "output_power": 0.001780718887587223, "reference_active": true, "delta_db": -0.20004148418814505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015694821766981044, "output_power": 0.00014966857654892337, "reference_active": true, "delta_db": -0.20625759904010293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505413793795639e-06, "output_power": 3.6124816591164143e-06, "reference_active": true, "delta_db": -2.55469275389032}], "active_band_count": 3, "active_band_mae_db": 0.9869972790395227}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00013449969207959865, "log_rms_mae_db": 1.6703276292771434, "rms_envelope_nrmse": 0.17629518568327873, "envelope_correlation": 0.980142437763523, "rms_gain_error_db": -0.6397660620313961, "reference_mean_power": 0.0006066736643199691, "output_mean_power": 0.0005235745600198446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.4525665594490304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019778914796660728, "output_power": 0.0001818221141844817, "reference_active": true, "delta_db": -0.3655575628795252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.140914614473558e-05, "output_power": 2.588868839616264e-05, "reference_active": true, "delta_db": -0.8394608205294479}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0263986880205243e-05, "output_power": 1.0417553596559394e-05, "reference_active": true, "delta_db": 0.06449655270553535}], "active_band_count": 3, "active_band_mae_db": 0.42317164537150287}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00012636239720129053, "log_rms_mae_db": 1.0858400336259741, "rms_envelope_nrmse": 0.11865687686210448, "envelope_correlation": 0.9930353416208488, "rms_gain_error_db": -0.6397796822086228, "reference_mean_power": 0.0006115927691385325, "output_mean_power": 0.0005278182152992372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.048532457224497e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001773914005108231, "output_power": 0.00016376371521342094, "reference_active": true, "delta_db": -0.3471488014535743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4831454807134908e-05, "output_power": 1.2866329067183912e-05, "reference_active": true, "delta_db": -0.6172909810996896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.85797297398325e-07, "output_power": 6.283100313166704e-07, "reference_active": true, "delta_db": 1.1171889912643753}], "active_band_count": 3, "active_band_mae_db": 0.693876257939213}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 5.5319997387638595e-05, "log_rms_mae_db": 1.9372676252619343, "rms_envelope_nrmse": 0.3073129710309251, "envelope_correlation": 0.937183516991418, "rms_gain_error_db": 0.25068166418923854, "reference_mean_power": 9.493685130389184e-05, "output_mean_power": 0.00010057799878149362}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.144674032958852e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.902158206464317e-05, "output_power": 2.536612378191167e-05, "reference_active": true, "delta_db": -0.5846697621139751}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.982855070886748e-06, "output_power": 1.572321307160371e-05, "reference_active": true, "delta_db": 1.9728653360429724}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.211971373682015e-06, "output_power": 6.389382225948575e-06, "reference_active": true, "delta_db": 2.986872032705815}], "active_band_count": 3, "active_band_mae_db": 1.8481357102875873}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 4.1914797815135623e-05, "log_rms_mae_db": 1.5349744114835528, "rms_envelope_nrmse": 0.25990136693934374, "envelope_correlation": 0.9525065928779299, "rms_gain_error_db": 0.2506825346780207, "reference_mean_power": 9.570694886911743e-05, "output_mean_power": 0.00010139387586533238}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.518176746918335e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6494345211181707e-05, "output_power": 2.3389414736867027e-05, "reference_active": true, "delta_db": -0.541338359215167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.515847029872695e-06, "output_power": 8.504724789551537e-06, "reference_active": true, "delta_db": 2.74921042351872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6527459125158204e-07, "output_power": 6.361181192289291e-07, "reference_active": true, "delta_db": 3.7984211220012547}], "active_band_count": 3, "active_band_mae_db": 2.362989968245047}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8897187570818353, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3b8a9d83e1f9d25795c4e2b2a04b79ed70974b0082334b5a63d3b25154762363", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9823758175439764, "standard_deviation": 0.0034708473054234877, "minimum": 0.976695092325609, "maximum": 0.9867379544798589, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6711990415097946, "standard_deviation": 0.04869695875083867, "minimum": 1.603678567308997, "maximum": 1.7469105797366065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5866115000082155, "standard_deviation": 0.19832590945081674, "minimum": 0.42144488349828446, "maximum": 0.8972877102193073, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16317440457099183, "standard_deviation": 0.014202693397328527, "minimum": 0.14002912835922723, "maximum": 0.18496917541175373, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941697473484048, "standard_deviation": 0.0014951673928363938, "minimum": 0.9916624372752382, "maximum": 0.9960265860509672, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0510608623984394, "standard_deviation": 0.043298945874759014, "minimum": 1.004291632697727, "maximum": 1.1106510791234396, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.651651610614925, "standard_deviation": 0.17814549148481193, "minimum": 0.35269302693834254, "maximum": 0.8638404215162389, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10374322141814371, "standard_deviation": 0.01454697615348117, "minimum": 0.08491554345896501, "maximum": 0.11949586919036909, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9385566162940744, "standard_deviation": 0.002901855686581076, "minimum": 0.9346226056425907, "maximum": 0.94356343233448, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9385070455572206, "standard_deviation": 0.05158078785067489, "minimum": 1.8830560125810258, "maximum": 2.0427681482673825, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.8062127095239633, "standard_deviation": 0.3642277515449142, "minimum": 1.4679127082126764, "maximum": 2.6241331753359973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.301328079587544, "standard_deviation": 0.007508325905562857, "minimum": 0.2892400828571387, "maximum": 0.31019535446172664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9544570087594022, "standard_deviation": 0.0022936341396969043, "minimum": 0.9506123585617603, "maximum": 0.9566201230111767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5114488327377849, "standard_deviation": 0.04374509971647308, "minimum": 1.4207084728876767, "maximum": 1.5591064610963818, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.3549152311935364, "standard_deviation": 0.33120071550854246, "minimum": 2.0322387190586766, "maximum": 3.104318295606206, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2523785298894233, "standard_deviation": 0.006658425239979805, "minimum": 0.24491397998391537, "maximum": 0.25990136693934374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933486483016392, "standard_deviation": 0.000731566977259848, "minimum": 0.9923800919127089, "maximum": 0.9944325881830304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.515205015296896, "standard_deviation": 0.06345219204728748, "minimum": 1.4197513508552229, "maximum": 1.6072586390470391, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7452571380436492, "standard_deviation": 0.31002904154483674, "minimum": 0.4314433599129712, "maximum": 1.2293636915828172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09236057029256506, "standard_deviation": 0.004803796250933226, "minimum": 0.08472147719816023, "maximum": 0.09837539209689049, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976834754434467, "standard_deviation": 0.0002588104021550309, "minimum": 0.9973083777824587, "maximum": 0.9980696460192684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9452323177199584, "standard_deviation": 0.05096880699797394, "minimum": 0.857067696973589, "maximum": 1.0242958770370958, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.080699276104136, "standard_deviation": 0.2731879030501377, "minimum": 0.6369888743894602, "maximum": 1.4597152918191447, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05397644705275046, "standard_deviation": 0.0032350710911208265, "minimum": 0.04843455760317351, "maximum": 0.05845816096851934, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3015108090763875, "standard_deviation": 0.07800011728106715, "minimum": 0.15689851803608226, "maximum": 0.3963166111719927, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5978055803715, "standard_deviation": 0.01121167013539439, "minimum": 5.577455207895627, "maximum": 5.613268936993798, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9734376379260061, "standard_deviation": 0.038042666195579285, "minimum": 0.9242332913275049, "maximum": 1.0409052636111327, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12435191635869233, "standard_deviation": 0.034758631540748, "minimum": -0.18798148264062065, "maximum": -0.08389565074625141, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956911842274997, "standard_deviation": 0.0008148178876804355, "minimum": 0.9940906030000892, "maximum": 0.9965956953131183, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1086393376360384, "standard_deviation": 0.08169610045192467, "minimum": 1.0018115036693014, "maximum": 1.2223539739499862, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27503273937086664, "standard_deviation": 0.14437853844378978, "minimum": 0.08956362655249996, "maximum": 0.5396875742607313, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06604501202357882, "standard_deviation": 0.005618457182550678, "minimum": 0.05862790585353065, "maximum": 0.07681555479561458, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986228962099231, "standard_deviation": 0.0004469681891582138, "minimum": 0.997776935459864, "maximum": 0.9991806814189281, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5952268746783447, "standard_deviation": 0.08928559031955399, "minimum": 0.4601504400777502, "maximum": 0.7046880936316177, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2742506221237545, "standard_deviation": 0.11078392240311738, "minimum": 0.13584798298763687, "maximum": 0.4712975408639523, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038061840763755, "standard_deviation": 0.004860880022468573, "minimum": 0.03281708638094683, "maximum": 0.04683155991453481, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0000", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.6748080253601074, "predicted_mos": 2.6748080253601074, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "1885481195d8d4e4cdd85d6ecdadebf31c85148dae1dc675c72bf61a65b021f6", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.793165300041437, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.072502347351038, "全帯域_10ms_rms_envelope_nrmse": 0.06264362212907126, "全帯域_10ms_envelope_correlation": 0.9962391025292354, "全帯域_10ms_modulation_mae_db": 0.3468526217514078, "全帯域_40ms_log_rms_mae_db": 0.6242080921132576, "全帯域_40ms_rms_envelope_nrmse": 0.03376489681896584, "全帯域_40ms_envelope_correlation": 0.9990670873700785, "全帯域_40ms_modulation_mae_db": 0.14625478566800235, "300–1000 Hz_10ms_log_rms_mae_db": 1.5389650467853158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08823232067528958, "300–1000 Hz_10ms_envelope_correlation": 0.9939536007366454, "300–1000 Hz_10ms_modulation_mae_db": 0.464935450196392, "300–1000 Hz_40ms_log_rms_mae_db": 0.9144796801098478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053177259491313106, "300–1000 Hz_40ms_envelope_correlation": 0.9977328640209161, "300–1000 Hz_40ms_modulation_mae_db": 0.9529726894121501, "1000–3000 Hz_10ms_log_rms_mae_db": 1.661862157916618, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17346336801037865, "1000–3000 Hz_10ms_envelope_correlation": 0.9816164172088955, "1000–3000 Hz_10ms_modulation_mae_db": 0.7881504489761461, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0838924480831331, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11771806846804941, "1000–3000 Hz_40ms_envelope_correlation": 0.9940138957828055, "1000–3000 Hz_40ms_modulation_mae_db": 0.5349308520203581, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8873936847812869, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3083905676416406, "3000–6000 Hz_10ms_envelope_correlation": 0.9360490524918954, "3000–6000 Hz_10ms_modulation_mae_db": 1.8350706240784025, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5436258212428722, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25904934875724883, "3000–6000 Hz_40ms_envelope_correlation": 0.9521151299234683, "3000–6000 Hz_40ms_modulation_mae_db": 2.3370755426422467, "mr_stft_db": 5.556444297710843, "band_mae_db": 0.40280405490555765, "rms_gain_error_db": -0.15052950689558384, "output_peak": 0.9686837272670071}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1986788931322183, "全帯域_10ms_rms_envelope_nrmse": 0.07036989312002706, "全帯域_10ms_envelope_correlation": 0.9949959681499028, "全帯域_10ms_modulation_mae_db": 0.4834324129368859, "全帯域_40ms_log_rms_mae_db": 0.6905137302799335, "全帯域_40ms_rms_envelope_nrmse": 0.0414072568884916, "全帯域_40ms_envelope_correlation": 0.9982165844796128, "全帯域_40ms_modulation_mae_db": 0.40441830196623907, "300–1000 Hz_10ms_log_rms_mae_db": 1.5327683649577195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09682160591509792, "300–1000 Hz_10ms_envelope_correlation": 0.9926468330816748, "300–1000 Hz_10ms_modulation_mae_db": 1.0692435359031613, "300–1000 Hz_40ms_log_rms_mae_db": 0.961844979755681, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05356918053804865, "300–1000 Hz_40ms_envelope_correlation": 0.997672470075839, "300–1000 Hz_40ms_modulation_mae_db": 1.1530890132768017, "1000–3000 Hz_10ms_log_rms_mae_db": 1.696606158880663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1639346679065694, "1000–3000 Hz_10ms_envelope_correlation": 0.9806329833734463, "1000–3000 Hz_10ms_modulation_mae_db": 0.5217202576492729, "1000–3000 Hz_40ms_log_rms_mae_db": 1.048777435669767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08684845863539013, "1000–3000 Hz_40ms_envelope_correlation": 0.9945303108836215, "1000–3000 Hz_40ms_modulation_mae_db": 0.516869564614073, "3000–6000 Hz_10ms_log_rms_mae_db": 1.861484250563006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30671423273436715, "3000–6000 Hz_10ms_envelope_correlation": 0.9345151410970277, "3000–6000 Hz_10ms_modulation_mae_db": 1.3851488549502147, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4522679502232392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25002358895499655, "3000–6000 Hz_40ms_envelope_correlation": 0.9538865581482253, "3000–6000 Hz_40ms_modulation_mae_db": 1.9772447039534082, "mr_stft_db": 5.536677922299326, "band_mae_db": 0.15752525017631716, "rms_gain_error_db": -0.07704589453240449, "output_peak": 0.9804394444617504}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1584102853612752, "全帯域_10ms_rms_envelope_nrmse": 0.07617738852611876, "全帯域_10ms_envelope_correlation": 0.9941709921639975, "全帯域_10ms_modulation_mae_db": 0.14227049961951815, "全帯域_40ms_log_rms_mae_db": 0.6762693033778318, "全帯域_40ms_rms_envelope_nrmse": 0.03974085738666333, "全帯域_40ms_envelope_correlation": 0.9984185523376972, "全帯域_40ms_modulation_mae_db": 0.18985052445690323, "300–1000 Hz_10ms_log_rms_mae_db": 1.570208908689039, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09855234622168173, "300–1000 Hz_10ms_envelope_correlation": 0.9923233401755456, "300–1000 Hz_10ms_modulation_mae_db": 0.45362006094040846, "300–1000 Hz_40ms_log_rms_mae_db": 0.9165914709267031, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04818694526157111, "300–1000 Hz_40ms_envelope_correlation": 0.9980772239683818, "300–1000 Hz_40ms_modulation_mae_db": 1.0251073616049846, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6815227153849375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15802952056113276, "1000–3000 Hz_10ms_envelope_correlation": 0.986629869111295, "1000–3000 Hz_10ms_modulation_mae_db": 1.032284079896188, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0364174858829616, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11206813079026147, "1000–3000 Hz_40ms_envelope_correlation": 0.9962375163134788, "1000–3000 Hz_40ms_modulation_mae_db": 0.6993271208866848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7980286808250194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30267464219462076, "3000–6000 Hz_10ms_envelope_correlation": 0.9359899613236465, "3000–6000 Hz_10ms_modulation_mae_db": 1.5187067677952968, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4561404206831539, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24953254695526822, "3000–6000 Hz_40ms_envelope_correlation": 0.9540032852961777, "3000–6000 Hz_40ms_modulation_mae_db": 2.0983469036584714, "mr_stft_db": 5.536803418804983, "band_mae_db": 0.3016231343766015, "rms_gain_error_db": -0.10217628072822281, "output_peak": 0.9519427844848896}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0690293957519683, "全帯域_10ms_rms_envelope_nrmse": 0.07120561428543531, "全帯域_10ms_envelope_correlation": 0.9951692411455421, "全帯域_10ms_modulation_mae_db": 0.5953415317247862, "全帯域_40ms_log_rms_mae_db": 0.6178009945936219, "全帯域_40ms_rms_envelope_nrmse": 0.04605201055489487, "全帯域_40ms_envelope_correlation": 0.9981290432516909, "全帯域_40ms_modulation_mae_db": 0.5998275208763183, "300–1000 Hz_10ms_log_rms_mae_db": 1.493990677238612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0912005819121944, "300–1000 Hz_10ms_envelope_correlation": 0.9935563719835963, "300–1000 Hz_10ms_modulation_mae_db": 1.1687192449224575, "300–1000 Hz_40ms_log_rms_mae_db": 0.905461686424389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05158717172552475, "300–1000 Hz_40ms_envelope_correlation": 0.9979093237997402, "300–1000 Hz_40ms_modulation_mae_db": 1.2489775400604393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6337951287527317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13312341393585414, "1000–3000 Hz_10ms_envelope_correlation": 0.9879961632745373, "1000–3000 Hz_10ms_modulation_mae_db": 0.893982562959306, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9812605717999096, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0820752240321447, "1000–3000 Hz_40ms_envelope_correlation": 0.9959740278593389, "1000–3000 Hz_40ms_modulation_mae_db": 0.7751731920659787, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8880535166434278, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.293161628762876, "3000–6000 Hz_10ms_envelope_correlation": 0.9409745887519648, "3000–6000 Hz_10ms_modulation_mae_db": 1.4820017772223768, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5365106346853736, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24693791996263856, "3000–6000 Hz_40ms_envelope_correlation": 0.9555778984834883, "3000–6000 Hz_40ms_modulation_mae_db": 2.130800340177697, "mr_stft_db": 5.540659226041577, "band_mae_db": 0.28647470135731923, "rms_gain_error_db": -0.18023276504893782, "output_peak": 0.9565789012137573}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0175695614987905, "全帯域_10ms_rms_envelope_nrmse": 0.06435838440901429, "全帯域_10ms_envelope_correlation": 0.9959597853657891, "全帯域_10ms_modulation_mae_db": 0.1612425881743322, "全帯域_40ms_log_rms_mae_db": 0.46062443893727323, "全帯域_40ms_rms_envelope_nrmse": 0.03476845102425714, "全帯域_40ms_envelope_correlation": 0.9989277800754278, "全帯域_40ms_modulation_mae_db": 0.3526736538041139, "300–1000 Hz_10ms_log_rms_mae_db": 1.579945641004694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09438730879346255, "300–1000 Hz_10ms_envelope_correlation": 0.9931781877629964, "300–1000 Hz_10ms_modulation_mae_db": 0.8202830534271213, "300–1000 Hz_40ms_log_rms_mae_db": 1.0023499257406687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05904111027374485, "300–1000 Hz_40ms_envelope_correlation": 0.9973664716589649, "300–1000 Hz_40ms_modulation_mae_db": 1.0392794334931683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5792840592717377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16258783580858827, "1000–3000 Hz_10ms_envelope_correlation": 0.9831359356941969, "1000–3000 Hz_10ms_modulation_mae_db": 0.555789958904172, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9579820576826267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1068169364132703, "1000–3000 Hz_40ms_envelope_correlation": 0.9944850334685059, "1000–3000 Hz_40ms_modulation_mae_db": 0.4400354719678445, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8777038066719876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3044221924345591, "3000–6000 Hz_10ms_envelope_correlation": 0.9348463289012982, "3000–6000 Hz_10ms_modulation_mae_db": 1.7649533029091302, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5200056870521923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2569698778240997, "3000–6000 Hz_40ms_envelope_correlation": 0.9508584178952532, "3000–6000 Hz_40ms_modulation_mae_db": 2.2382014521266114, "mr_stft_db": 5.516562644950491, "band_mae_db": 0.2987148147033419, "rms_gain_error_db": -0.13454344201654495, "output_peak": 0.9500830192589917}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1789270367725835, "全帯域_10ms_rms_envelope_nrmse": 0.06968629277719747, "全帯域_10ms_envelope_correlation": 0.995165795954036, "全帯域_10ms_modulation_mae_db": 0.12368974677423332, "全帯域_40ms_log_rms_mae_db": 0.6465230430170485, "全帯域_40ms_rms_envelope_nrmse": 0.04611789705704821, "全帯域_40ms_envelope_correlation": 0.997850051951161, "全帯域_40ms_modulation_mae_db": 0.527388303226446, "300–1000 Hz_10ms_log_rms_mae_db": 1.603400102514351, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0910007604807338, "300–1000 Hz_10ms_envelope_correlation": 0.9934557718395637, "300–1000 Hz_10ms_modulation_mae_db": 0.6154644304896598, "300–1000 Hz_40ms_log_rms_mae_db": 0.9835463246720644, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056094349404371475, "300–1000 Hz_40ms_envelope_correlation": 0.9973478188910685, "300–1000 Hz_40ms_modulation_mae_db": 1.2252663068114746, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5962117547049368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18331806821727323, "1000–3000 Hz_10ms_envelope_correlation": 0.9775433130883738, "1000–3000 Hz_10ms_modulation_mae_db": 0.40586249644481803, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9823195035702706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1207072215149423, "1000–3000 Hz_40ms_envelope_correlation": 0.9918429811054983, "1000–3000 Hz_40ms_modulation_mae_db": 0.48035495010140633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9868194303626523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3033131723066818, "3000–6000 Hz_10ms_envelope_correlation": 0.9371790931525638, "3000–6000 Hz_10ms_modulation_mae_db": 1.6816957231465712, "3000–6000 Hz_40ms_log_rms_mae_db": 1.578286869496916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2586713324022175, "3000–6000 Hz_40ms_envelope_correlation": 0.9514750798553133, "3000–6000 Hz_40ms_modulation_mae_db": 2.2000914889556165, "mr_stft_db": 5.532752448161794, "band_mae_db": 0.40075662721094113, "rms_gain_error_db": -0.11500921940752466, "output_peak": 1.0376014819365729}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.351206434316354e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.102092456372107, "全帯域_10ms_rms_envelope_nrmse": 0.061436030465349456, "全帯域_10ms_envelope_correlation": 0.9962427621368134, "全帯域_10ms_modulation_mae_db": 0.07623285761755129, "全帯域_40ms_log_rms_mae_db": 0.589265567351489, "全帯域_40ms_rms_envelope_nrmse": 0.03585520688730298, "全帯域_40ms_envelope_correlation": 0.9987162805855129, "全帯域_40ms_modulation_mae_db": 0.11522598208756736, "300–1000 Hz_10ms_log_rms_mae_db": 1.475354063776825, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09692972078504968, "300–1000 Hz_10ms_envelope_correlation": 0.9926256144456657, "300–1000 Hz_10ms_modulation_mae_db": 1.1077969144266198, "300–1000 Hz_40ms_log_rms_mae_db": 0.9236804699952363, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05558908859508795, "300–1000 Hz_40ms_envelope_correlation": 0.9974809440330711, "300–1000 Hz_40ms_modulation_mae_db": 1.5773077162548133, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6701960380619745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15193240645009173, "1000–3000 Hz_10ms_envelope_correlation": 0.9841800162042714, "1000–3000 Hz_10ms_modulation_mae_db": 0.599256925830704, "1000–3000 Hz_40ms_log_rms_mae_db": 1.008586333773563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08467668036626995, "1000–3000 Hz_40ms_envelope_correlation": 0.9958974302067073, "1000–3000 Hz_40ms_modulation_mae_db": 0.3647559678360525, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8694816524508644, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3036285594776946, "3000–6000 Hz_10ms_envelope_correlation": 0.9380063544537343, "3000–6000 Hz_10ms_modulation_mae_db": 2.460577962381796, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4921931514670264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2515279950144344, "3000–6000 Hz_40ms_envelope_correlation": 0.9548135039429031, "3000–6000 Hz_40ms_modulation_mae_db": 2.940908080014244, "mr_stft_db": 5.55498391325014, "band_mae_db": 0.25645103659534185, "rms_gain_error_db": -0.09166220942677358, "output_peak": 1.0141713382362867}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0279500426140582, "全帯域_10ms_rms_envelope_nrmse": 0.06796696239765607, "全帯域_10ms_envelope_correlation": 0.9954379707249046, "全帯域_10ms_modulation_mae_db": 0.20596905833960086, "全帯域_40ms_log_rms_mae_db": 0.5329067561432238, "全帯域_40ms_rms_envelope_nrmse": 0.038185970082997275, "全帯域_40ms_envelope_correlation": 0.99862744636721, "全帯域_40ms_modulation_mae_db": 0.16868654697553465, "300–1000 Hz_10ms_log_rms_mae_db": 1.4550758667378545, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09243367130493867, "300–1000 Hz_10ms_envelope_correlation": 0.9934751433112267, "300–1000 Hz_10ms_modulation_mae_db": 0.3689440029111084, "300–1000 Hz_40ms_log_rms_mae_db": 0.8683724452062689, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05747779572627317, "300–1000 Hz_40ms_envelope_correlation": 0.9975377057289607, "300–1000 Hz_40ms_modulation_mae_db": 0.5385885025037103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6944322443466637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16358079106593773, "1000–3000 Hz_10ms_envelope_correlation": 0.984040361798624, "1000–3000 Hz_10ms_modulation_mae_db": 0.5815888150966325, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1204738965429852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11680964649255288, "1000–3000 Hz_40ms_envelope_correlation": 0.993951866156038, "1000–3000 Hz_40ms_modulation_mae_db": 0.5697271609203994, "3000–6000 Hz_10ms_log_rms_mae_db": 1.82485084400035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2882074627981032, "3000–6000 Hz_10ms_envelope_correlation": 0.9433080050656484, "3000–6000 Hz_10ms_modulation_mae_db": 1.7320401001606474, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4933312682656188, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24629015881516794, "3000–6000 Hz_40ms_envelope_correlation": 0.956051781435057, "3000–6000 Hz_40ms_modulation_mae_db": 2.3048444718899894, "mr_stft_db": 5.557317004791201, "band_mae_db": 0.356930879976193, "rms_gain_error_db": -0.1235784087919814, "output_peak": 0.9264926569751106}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59680, "output_samples": 59680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.700088846499647, "spectral_convergence": 0.20071682948703776, "spectral_convergence_status": "ok", "amplitude_floor": 3.189740694195829e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.398526982088858, "spectral_convergence": 0.1920982592579735, "spectral_convergence_status": "ok", "amplitude_floor": 2.205927306089595e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.570717064544024, "spectral_convergence": 0.1877941621492442, "spectral_convergence_status": "ok", "amplitude_floor": 1.9576049396551346e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004284845572396228, "output_power": 0.004370101065955281, "delta_db": 0.08556306489887838, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.012674374150346739, "output_power": 0.012084499326302175, "delta_db": -0.20697861615508162, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0031087438744962816, "output_power": 0.002987444620678401, "delta_db": -0.17285079180319568, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005542642121172655, "output_power": 0.0004776484926395081, "delta_db": -0.6460842631343794, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001593466145796079, "output_power": 0.00013645076789371442, "delta_db": -0.6736685698552957, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.162629787041768e-05, "output_power": 0.00010597124624329944, "delta_db": 0.6316790235865155, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.40280405490555765, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15052950689558384, "pitch": null, "output_rms_dbfs": -16.954639012773473, "output_peak": 0.9686837272670071, "reference_rms_dbfs": -16.80410950587789, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.556444297710843}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00043785897255075944, "log_rms_mae_db": 1.072502347351038, "rms_envelope_nrmse": 0.06264362212907126, "envelope_correlation": 0.9962391025292354, "rms_gain_error_db": -0.15053094486048438, "reference_mean_power": 0.0209012125553907, "output_mean_power": 0.020189166392308246}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.005551294872497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00444581511703579, "output_power": 0.004265027019815226, "reference_active": true, "delta_db": -0.18029611380571878}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00039454721092167534, "output_power": 0.0004096577239041283, "reference_active": true, "delta_db": 0.16322170016493884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.0001084016816849342, "output_power": 0.00012727409710493418, "reference_active": true, "delta_db": 0.6970400512835657}], "active_band_count": 3, "active_band_mae_db": 0.3468526217514078}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0004107294113879301, "log_rms_mae_db": 0.6242080921132576, "rms_envelope_nrmse": 0.03376489681896584, "envelope_correlation": 0.9990670873700785, "rms_gain_error_db": -0.15053032845001643, "reference_mean_power": 0.021070817398047652, "output_mean_power": 0.020352996158605748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.012296407388435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00391054354874771, "output_power": 0.003783627085129863, "reference_active": true, "delta_db": -0.1432880095033538}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019349823972573515, "output_power": 0.0001975637637464897, "reference_active": true, "delta_db": 0.09030272673502299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9805555260958754e-06, "output_power": 5.7045864344912724e-06, "reference_active": true, "delta_db": -0.20517362076563028}], "active_band_count": 3, "active_band_mae_db": 0.14625478566800235}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.0003730035430310483, "log_rms_mae_db": 1.5389650467853158, "rms_envelope_nrmse": 0.08823232067528958, "envelope_correlation": 0.9939536007366454, "rms_gain_error_db": -0.07843301041879208, "reference_mean_power": 0.007423998211444209, "output_mean_power": 0.007291125228996967}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.725628531347034e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002177971361103236, "output_power": 0.0020566406235741786, "reference_active": true, "delta_db": -0.2489375486077081}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00032785256002961345, "output_power": 0.0002903949949001902, "reference_active": true, "delta_db": -0.52689452380137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.051887548388713e-05, "output_power": 6.982328432166843e-05, "reference_active": true, "delta_db": -0.6189742781800978}], "active_band_count": 3, "active_band_mae_db": 0.464935450196392}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00035212990311193784, "log_rms_mae_db": 0.9144796801098478, "rms_envelope_nrmse": 0.053177259491313106, "envelope_correlation": 0.9977328640209161, "rms_gain_error_db": -0.07843365529335866, "reference_mean_power": 0.007484242812710004, "output_mean_power": 0.007350290495060225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.215922155621447e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018646593945503163, "output_power": 0.0017796518255795719, "reference_active": true, "delta_db": -0.20264469151109626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015694821766981044, "output_power": 0.0001478255625138631, "reference_active": true, "delta_db": -0.26006847868098637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505413793795639e-06, "output_power": 3.7467475880328677e-06, "reference_active": true, "delta_db": -2.3962048980443678}], "active_band_count": 3, "active_band_mae_db": 0.9529726894121501}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00013449969207959865, "log_rms_mae_db": 1.661862157916618, "rms_envelope_nrmse": 0.17346336801037865, "envelope_correlation": 0.9816164172088955, "rms_gain_error_db": -0.7177362174885265, "reference_mean_power": 0.0006066736643199691, "output_mean_power": 0.0005142585499336684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.4525665594490304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019778914796660728, "output_power": 0.0001787356008909167, "reference_active": true, "delta_db": -0.4399139491149356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.140914614473558e-05, "output_power": 2.3800281858303162e-05, "reference_active": true, "delta_db": -1.204740300137055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0263986880205243e-05, "output_power": 8.696337346562353e-06, "reference_active": true, "delta_db": -0.7197970976764477}], "active_band_count": 3, "active_band_mae_db": 0.7881504489761461}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00012636239720129053, "log_rms_mae_db": 1.0838924480831331, "rms_envelope_nrmse": 0.11771806846804941, "envelope_correlation": 0.9940138957828055, "rms_gain_error_db": -0.7177486015653582, "reference_mean_power": 0.0006115927691385325, "output_mean_power": 0.0005184268450221089}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.048532457224497e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001773914005108231, "output_power": 0.00016105746256135281, "reference_active": true, "delta_db": -0.4195170994203914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4831454807134908e-05, "output_power": 1.1402492480691204e-05, "reference_active": true, "delta_db": -1.1418395824284657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.85797297398325e-07, "output_power": 4.906803674390299e-07, "reference_active": true, "delta_db": 0.043435874212217195}], "active_band_count": 3, "active_band_mae_db": 0.5349308520203581}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 5.5319997387638595e-05, "log_rms_mae_db": 1.8873936847812869, "rms_envelope_nrmse": 0.3083905676416406, "envelope_correlation": 0.9360490524918954, "rms_gain_error_db": 0.16852548233607945, "reference_mean_power": 9.493685130389184e-05, "output_mean_power": 9.869323211242945e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.144674032958852e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.902158206464317e-05, "output_power": 2.527514644754847e-05, "reference_active": true, "delta_db": -0.6002740297871902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.982855070886748e-06, "output_power": 1.533751172356775e-05, "reference_active": true, "delta_db": 1.8650014154024888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.211971373682015e-06, "output_power": 6.467929965413097e-06, "reference_active": true, "delta_db": 3.0399364270455282}], "active_band_count": 3, "active_band_mae_db": 1.8350706240784025}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 4.1914797815135623e-05, "log_rms_mae_db": 1.5436258212428722, "rms_envelope_nrmse": 0.25904934875724883, "envelope_correlation": 0.9521151299234683, "rms_gain_error_db": 0.16852646230658755, "reference_mean_power": 9.570694886911743e-05, "output_mean_power": 9.94938226953671e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.518176746918335e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6494345211181707e-05, "output_power": 2.3163654321374245e-05, "reference_active": true, "delta_db": -0.5834611551088462}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.515847029872695e-06, "output_power": 8.355752033213618e-06, "reference_active": true, "delta_db": 2.67246321119511}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6527459125158204e-07, "output_power": 6.298336794388816e-07, "reference_active": true, "delta_db": 3.7553022616227842}], "active_band_count": 3, "active_band_mae_db": 2.3370755426422467}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8895595201822599, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3d12b057647233c80fd691e3be78d1a2a43b140edc4bfd300f828713ba29bb67", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.983221882469205, "standard_deviation": 0.0033280091242774068, "minimum": 0.9775433130883738, "maximum": 0.9879961632745373, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.651738782165033, "standard_deviation": 0.0444344802766252, "minimum": 1.5792840592717377, "maximum": 1.696606158880663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6723294432196549, "standard_deviation": 0.21144219663526267, "minimum": 0.40586249644481803, "maximum": 1.032284079896188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16124625899447825, "standard_deviation": 0.014827048619965807, "minimum": 0.13312341393585414, "maximum": 0.18331806821727323, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946166327219993, "standard_deviation": 0.0014458485838003856, "minimum": 0.9918429811054983, "maximum": 0.9962375163134788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0274637166256522, "standard_deviation": 0.055711408145084435, "minimum": 0.9579820576826267, "maximum": 1.1204738965429852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5476467850515997, "standard_deviation": 0.13412632754953946, "minimum": 0.3647559678360525, "maximum": 0.7751731920659787, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10346504583911015, "standard_deviation": 0.016258650480672154, "minimum": 0.0820752240321447, "maximum": 0.1207072215149423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9376085656547224, "standard_deviation": 0.003080064756054883, "minimum": 0.9345151410970277, "maximum": 0.9433080050656484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8742269832873244, "standard_deviation": 0.05539824948651574, "minimum": 1.7980286808250194, "maximum": 1.9868194303626523, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7325243890805546, "standard_deviation": 0.33238585024767636, "minimum": 1.3851488549502147, "maximum": 2.460577962381796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3013140572938179, "standard_deviation": 0.006952377601922522, "minimum": 0.2882074627981032, "maximum": 0.3083905676416406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9535977068724858, "standard_deviation": 0.001922684736552978, "minimum": 0.9508584178952532, "maximum": 0.956051781435057, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.509045225389549, "standard_deviation": 0.04378225585818848, "minimum": 1.4522679502232392, "maximum": 1.578286869496916, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.278439122927286, "standard_deviation": 0.29172063661981573, "minimum": 1.9772447039534082, "maximum": 2.940908080014244, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.25237534608575896, "standard_deviation": 0.005157752231208425, "minimum": 0.24629015881516794, "maximum": 0.25904934875724883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931518579171144, "standard_deviation": 0.0005634609229051347, "minimum": 0.9923233401755456, "maximum": 0.9939536007366454, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5312135839630514, "standard_deviation": 0.05274145493292776, "minimum": 1.4550758667378545, "maximum": 1.603400102514351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7586258366521161, "standard_deviation": 0.3255789784819884, "minimum": 0.3689440029111084, "maximum": 1.1687192449224575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09369478951105603, "standard_deviation": 0.0035702150835459994, "minimum": 0.08823232067528958, "maximum": 0.09855234622168173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976406027721179, "standard_deviation": 0.0002589334457883964, "minimum": 0.9973478188910685, "maximum": 0.9980772239683818, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9345408728538573, "standard_deviation": 0.04441936024154896, "minimum": 0.8683724452062689, "maximum": 1.0023499257406687, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0950735704271928, "standard_deviation": 0.2962693491661766, "minimum": 0.5385885025037103, "maximum": 1.5773077162548133, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05434036262699188, "standard_deviation": 0.003463450739633095, "minimum": 0.04818694526157111, "maximum": 0.05904111027374485, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3076600624127017, "standard_deviation": 0.08095419410139779, "minimum": 0.15752525017631716, "maximum": 0.40280405490555765, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.541525109501294, "standard_deviation": 0.014140767287857074, "minimum": 5.516562644950491, "maximum": 5.557317004791201, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9732491692292957, "standard_deviation": 0.036533916129791184, "minimum": 0.9264926569751106, "maximum": 1.0376014819365729, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12184721585599669, "standard_deviation": 0.03326889205773535, "minimum": -0.18023276504893782, "maximum": -0.07704589453240449, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954227022712776, "standard_deviation": 0.0007077956264197879, "minimum": 0.9941709921639975, "maximum": 0.9962427621368134, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1031450023567548, "standard_deviation": 0.06866615443482661, "minimum": 1.0175695614987905, "maximum": 1.1986788931322183, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26687891461728946, "standard_deviation": 0.18838572143566734, "minimum": 0.07623285761755129, "maximum": 0.5953415317247862, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06798052351373371, "standard_deviation": 0.004937343468193654, "minimum": 0.061436030465349456, "maximum": 0.07617738852611876, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984941033022989, "standard_deviation": 0.0004165033068410564, "minimum": 0.997850051951161, "maximum": 0.9990670873700785, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6047639907267099, "standard_deviation": 0.07647385757488565, "minimum": 0.46062443893727323, "maximum": 0.6905137302799335, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3130407023826406, "standard_deviation": 0.18557799957880333, "minimum": 0.11522598208756736, "maximum": 0.5998275208763183, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03948656833757766, "standard_deviation": 0.004791506731054887, "minimum": 0.03376489681896584, "maximum": 0.04611789705704821, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0000", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.922340154647827, "predicted_mos": 2.922340154647827, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "e01c24b8cdda4cbca28c79d74f33ef6d066637915962117443e452e19216f0e4", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.854645799845457, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1290268466730426, "全帯域_10ms_rms_envelope_nrmse": 0.06341837090979724, "全帯域_10ms_envelope_correlation": 0.9959423889431023, "全帯域_10ms_modulation_mae_db": 0.15373567113572192, "全帯域_40ms_log_rms_mae_db": 0.5663145729200888, "全帯域_40ms_rms_envelope_nrmse": 0.032708547545639, "全帯域_40ms_envelope_correlation": 0.9989121981644609, "全帯域_40ms_modulation_mae_db": 0.43368409347943965, "300–1000 Hz_10ms_log_rms_mae_db": 1.746729304415926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08489668465626758, "300–1000 Hz_10ms_envelope_correlation": 0.9944993383986735, "300–1000 Hz_10ms_modulation_mae_db": 0.8322595426175899, "300–1000 Hz_40ms_log_rms_mae_db": 1.096697333384167, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049728067258803875, "300–1000 Hz_40ms_envelope_correlation": 0.9980715697513318, "300–1000 Hz_40ms_modulation_mae_db": 1.2702505790768417, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9557594262530416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1640857797053984, "1000–3000 Hz_10ms_envelope_correlation": 0.9834580211939243, "1000–3000 Hz_10ms_modulation_mae_db": 1.2778265930920851, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1312443879777871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11147491732456713, "1000–3000 Hz_40ms_envelope_correlation": 0.9943481445895133, "1000–3000 Hz_40ms_modulation_mae_db": 0.8525227324975276, "3000–6000 Hz_10ms_log_rms_mae_db": 2.465661930825273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3138495741469869, "3000–6000 Hz_10ms_envelope_correlation": 0.9344577516313454, "3000–6000 Hz_10ms_modulation_mae_db": 1.3992711677098775, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8877282260593318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26041371009094305, "3000–6000 Hz_40ms_envelope_correlation": 0.9527317082395839, "3000–6000 Hz_40ms_modulation_mae_db": 1.986650046699718, "mr_stft_db": 5.909413407208781, "band_mae_db": 0.35016346033376705, "rms_gain_error_db": -0.06979902040665889, "output_peak": 0.9748558964818065}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0375960841954266, "全帯域_10ms_rms_envelope_nrmse": 0.06639660073897993, "全帯域_10ms_envelope_correlation": 0.995671259998739, "全帯域_10ms_modulation_mae_db": 0.3091724093571758, "全帯域_40ms_log_rms_mae_db": 0.5153659483516789, "全帯域_40ms_rms_envelope_nrmse": 0.03678721903230723, "全帯域_40ms_envelope_correlation": 0.9987521454050258, "全帯域_40ms_modulation_mae_db": 0.30307251284178505, "300–1000 Hz_10ms_log_rms_mae_db": 1.6757045302811617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09547234866747023, "300–1000 Hz_10ms_envelope_correlation": 0.9930038732717691, "300–1000 Hz_10ms_modulation_mae_db": 1.1460119200306604, "300–1000 Hz_40ms_log_rms_mae_db": 1.0052962410664783, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05430100118190158, "300–1000 Hz_40ms_envelope_correlation": 0.9978055112218198, "300–1000 Hz_40ms_modulation_mae_db": 1.1636368652829445, "1000–3000 Hz_10ms_log_rms_mae_db": 1.966526718689003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1675458748849416, "1000–3000 Hz_10ms_envelope_correlation": 0.9802486359571927, "1000–3000 Hz_10ms_modulation_mae_db": 0.6484923153585813, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2153105902819947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09294358694172596, "1000–3000 Hz_40ms_envelope_correlation": 0.9944829637383853, "1000–3000 Hz_40ms_modulation_mae_db": 0.8761136494966987, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4583619671007795, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.33343542446891455, "3000–6000 Hz_10ms_envelope_correlation": 0.9337446474264626, "3000–6000 Hz_10ms_modulation_mae_db": 2.0003354201946246, "3000–6000 Hz_40ms_log_rms_mae_db": 1.802752993125389, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26793721831885436, "3000–6000 Hz_40ms_envelope_correlation": 0.9563905058159862, "3000–6000 Hz_40ms_modulation_mae_db": 2.598034847939525, "mr_stft_db": 5.888776116382691, "band_mae_db": 0.38384472432122047, "rms_gain_error_db": -0.12932994142528892, "output_peak": 1.0230732569431051}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1272959978857424, "全帯域_10ms_rms_envelope_nrmse": 0.06828462471607778, "全帯域_10ms_envelope_correlation": 0.9954104185334756, "全帯域_10ms_modulation_mae_db": 0.1825220664437011, "全帯域_40ms_log_rms_mae_db": 0.6215273698602416, "全帯域_40ms_rms_envelope_nrmse": 0.03774481078047812, "全帯域_40ms_envelope_correlation": 0.9986758576031596, "全帯域_40ms_modulation_mae_db": 0.19007358661958743, "300–1000 Hz_10ms_log_rms_mae_db": 1.6974884259387037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09904508817793023, "300–1000 Hz_10ms_envelope_correlation": 0.9924616127551964, "300–1000 Hz_10ms_modulation_mae_db": 0.7731249613252036, "300–1000 Hz_40ms_log_rms_mae_db": 1.0371948133815396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05546478338474082, "300–1000 Hz_40ms_envelope_correlation": 0.997672773668817, "300–1000 Hz_40ms_modulation_mae_db": 1.3371557767455118, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9098536981961174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14413716126689968, "1000–3000 Hz_10ms_envelope_correlation": 0.98527337091766, "1000–3000 Hz_10ms_modulation_mae_db": 0.34391408098264803, "1000–3000 Hz_40ms_log_rms_mae_db": 1.148122841965324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0819946580926648, "1000–3000 Hz_40ms_envelope_correlation": 0.9953627526605291, "1000–3000 Hz_40ms_modulation_mae_db": 0.2689713642399591, "3000–6000 Hz_10ms_log_rms_mae_db": 2.505110741774698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3145249471974874, "3000–6000 Hz_10ms_envelope_correlation": 0.9337209852913969, "3000–6000 Hz_10ms_modulation_mae_db": 1.4332390931266088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8840260660557369, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2588461244249383, "3000–6000 Hz_40ms_envelope_correlation": 0.9528435135834322, "3000–6000 Hz_40ms_modulation_mae_db": 1.898479486483069, "mr_stft_db": 5.9225252507800965, "band_mae_db": 0.28322102738025695, "rms_gain_error_db": -0.13129209720231858, "output_peak": 1.0074574245354608}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1399670088093847, "全帯域_10ms_rms_envelope_nrmse": 0.07078299942068343, "全帯域_10ms_envelope_correlation": 0.9951355952392013, "全帯域_10ms_modulation_mae_db": 0.0915588398767374, "全帯域_40ms_log_rms_mae_db": 0.6082460322036704, "全帯域_40ms_rms_envelope_nrmse": 0.04479426883636695, "全帯域_40ms_envelope_correlation": 0.9981421562338174, "全帯域_40ms_modulation_mae_db": 0.2654091837498001, "300–1000 Hz_10ms_log_rms_mae_db": 1.7353986407153563, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0927299999607143, "300–1000 Hz_10ms_envelope_correlation": 0.9935959911775999, "300–1000 Hz_10ms_modulation_mae_db": 1.0086776543514684, "300–1000 Hz_40ms_log_rms_mae_db": 1.13371415576771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05987528462317407, "300–1000 Hz_40ms_envelope_correlation": 0.9973848095192329, "300–1000 Hz_40ms_modulation_mae_db": 1.5227021868894022, "1000–3000 Hz_10ms_log_rms_mae_db": 2.013610791720482, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16799769102339712, "1000–3000 Hz_10ms_envelope_correlation": 0.9836893129693598, "1000–3000 Hz_10ms_modulation_mae_db": 1.0272632335308023, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2507210120816143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11696416946753697, "1000–3000 Hz_40ms_envelope_correlation": 0.9948598736603153, "1000–3000 Hz_40ms_modulation_mae_db": 0.7954777064176642, "3000–6000 Hz_10ms_log_rms_mae_db": 2.403754097672059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32012238707958435, "3000–6000 Hz_10ms_envelope_correlation": 0.9323827768589615, "3000–6000 Hz_10ms_modulation_mae_db": 1.5785379534767625, "3000–6000 Hz_40ms_log_rms_mae_db": 1.84643579530557, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2609234331632294, "3000–6000 Hz_40ms_envelope_correlation": 0.9529447464985475, "3000–6000 Hz_40ms_modulation_mae_db": 2.238496153971244, "mr_stft_db": 5.90699836726512, "band_mae_db": 0.3814414180993095, "rms_gain_error_db": -0.1554274809259929, "output_peak": 0.9910467153411825}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1258941831665372, "全帯域_10ms_rms_envelope_nrmse": 0.06738747791029592, "全帯域_10ms_envelope_correlation": 0.9954831979327655, "全帯域_10ms_modulation_mae_db": 0.541691393005066, "全帯域_40ms_log_rms_mae_db": 0.6326815286064644, "全帯域_40ms_rms_envelope_nrmse": 0.03948206492394729, "全帯域_40ms_envelope_correlation": 0.9984598120504908, "全帯域_40ms_modulation_mae_db": 0.602085413929952, "300–1000 Hz_10ms_log_rms_mae_db": 1.6621815643945008, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09964205549910761, "300–1000 Hz_10ms_envelope_correlation": 0.9921612556812163, "300–1000 Hz_10ms_modulation_mae_db": 1.1245737083747172, "300–1000 Hz_40ms_log_rms_mae_db": 1.0431010716862392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056290374895111094, "300–1000 Hz_40ms_envelope_correlation": 0.9973858517376138, "300–1000 Hz_40ms_modulation_mae_db": 1.439573523215417, "1000–3000 Hz_10ms_log_rms_mae_db": 2.01124504154441, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19460349826103043, "1000–3000 Hz_10ms_envelope_correlation": 0.9758992568698591, "1000–3000 Hz_10ms_modulation_mae_db": 0.46002808123142813, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2134087165647545, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14116351093800109, "1000–3000 Hz_40ms_envelope_correlation": 0.9896422947471415, "1000–3000 Hz_40ms_modulation_mae_db": 0.5946310861281279, "3000–6000 Hz_10ms_log_rms_mae_db": 2.554746201581906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3328083622120699, "3000–6000 Hz_10ms_envelope_correlation": 0.9246527829633683, "3000–6000 Hz_10ms_modulation_mae_db": 1.692219623844603, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9321767784438864, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2628302534206694, "3000–6000 Hz_40ms_envelope_correlation": 0.9505993613544466, "3000–6000 Hz_40ms_modulation_mae_db": 2.223661150621529, "mr_stft_db": 5.912433065288689, "band_mae_db": 0.2952802824675978, "rms_gain_error_db": -0.10727809245865276, "output_peak": 0.9886952460548581}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1670227729322775, "全帯域_10ms_rms_envelope_nrmse": 0.07593556704665304, "全帯域_10ms_envelope_correlation": 0.9943957228094343, "全帯域_10ms_modulation_mae_db": 0.5048915133150588, "全帯域_40ms_log_rms_mae_db": 0.7068488280894374, "全帯域_40ms_rms_envelope_nrmse": 0.043655072120076244, "全帯域_40ms_envelope_correlation": 0.9982944051464526, "全帯域_40ms_modulation_mae_db": 0.4292244177026802, "300–1000 Hz_10ms_log_rms_mae_db": 1.7193139258081032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0901342197000477, "300–1000 Hz_10ms_envelope_correlation": 0.9941923225300813, "300–1000 Hz_10ms_modulation_mae_db": 1.6575665138626776, "300–1000 Hz_40ms_log_rms_mae_db": 1.0282583126810991, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05567486775776255, "300–1000 Hz_40ms_envelope_correlation": 0.9981057431542391, "300–1000 Hz_40ms_modulation_mae_db": 2.221469361719332, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0216124240718716, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20450522891234066, "1000–3000 Hz_10ms_envelope_correlation": 0.9703419783512397, "1000–3000 Hz_10ms_modulation_mae_db": 0.4844973453865136, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2366431364601222, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10987974261407984, "1000–3000 Hz_40ms_envelope_correlation": 0.9926990811022405, "1000–3000 Hz_40ms_modulation_mae_db": 0.7448293276100401, "3000–6000 Hz_10ms_log_rms_mae_db": 2.528213243898624, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32474222005518677, "3000–6000 Hz_10ms_envelope_correlation": 0.9279498209317767, "3000–6000 Hz_10ms_modulation_mae_db": 1.4312944914089023, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9242205595844109, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26761059426195866, "3000–6000 Hz_40ms_envelope_correlation": 0.9484535611016802, "3000–6000 Hz_40ms_modulation_mae_db": 2.0711795903498422, "mr_stft_db": 5.9046559368757485, "band_mae_db": 0.39850247069835554, "rms_gain_error_db": -0.16819827332653817, "output_peak": 0.9758714270402213}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.131112803853806, "全帯域_10ms_rms_envelope_nrmse": 0.06300652224169366, "全帯域_10ms_envelope_correlation": 0.9960937273145587, "全帯域_10ms_modulation_mae_db": 0.6148195119372795, "全帯域_40ms_log_rms_mae_db": 0.686074735565444, "全帯域_40ms_rms_envelope_nrmse": 0.03325892957614788, "全帯域_40ms_envelope_correlation": 0.9989876638493247, "全帯域_40ms_modulation_mae_db": 0.49280561762688985, "300–1000 Hz_10ms_log_rms_mae_db": 1.7176578354531664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09159653735774274, "300–1000 Hz_10ms_envelope_correlation": 0.993470395754736, "300–1000 Hz_10ms_modulation_mae_db": 0.7375672703271295, "300–1000 Hz_40ms_log_rms_mae_db": 1.0494878657804327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054144307157397466, "300–1000 Hz_40ms_envelope_correlation": 0.997653287248063, "300–1000 Hz_40ms_modulation_mae_db": 1.0337868461453281, "1000–3000 Hz_10ms_log_rms_mae_db": 2.005760776564141, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19752385933608468, "1000–3000 Hz_10ms_envelope_correlation": 0.9733743797075803, "1000–3000 Hz_10ms_modulation_mae_db": 0.5725497770677909, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2210217246382553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11577205545564133, "1000–3000 Hz_40ms_envelope_correlation": 0.9928763226877536, "1000–3000 Hz_40ms_modulation_mae_db": 0.5508956827872066, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4836250212366884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32811169579738514, "3000–6000 Hz_10ms_envelope_correlation": 0.9258301521918664, "3000–6000 Hz_10ms_modulation_mae_db": 1.666330713199945, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8504444466290262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2585933220931519, "3000–6000 Hz_40ms_envelope_correlation": 0.9516870585465689, "3000–6000 Hz_40ms_modulation_mae_db": 2.206338398836276, "mr_stft_db": 5.917985546378138, "band_mae_db": 0.3306636087504891, "rms_gain_error_db": -0.11951272853862197, "output_peak": 0.8853682812526045}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0693124236812723, "全帯域_10ms_rms_envelope_nrmse": 0.07239735663729846, "全帯域_10ms_envelope_correlation": 0.9947509315389034, "全帯域_10ms_modulation_mae_db": 0.21896652574919825, "全帯域_40ms_log_rms_mae_db": 0.5889872504062567, "全帯域_40ms_rms_envelope_nrmse": 0.03960089363237017, "全帯域_40ms_envelope_correlation": 0.9984687961801197, "全帯域_40ms_modulation_mae_db": 0.1680547779243651, "300–1000 Hz_10ms_log_rms_mae_db": 1.66222106182892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09571761292934954, "300–1000 Hz_10ms_envelope_correlation": 0.9929110015039531, "300–1000 Hz_10ms_modulation_mae_db": 0.8071025154821584, "300–1000 Hz_40ms_log_rms_mae_db": 1.0211679493615977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0511296646995802, "300–1000 Hz_40ms_envelope_correlation": 0.9980393343640622, "300–1000 Hz_40ms_modulation_mae_db": 1.0860774502932655, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0426624565309517, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18569313471767934, "1000–3000 Hz_10ms_envelope_correlation": 0.9760990520223967, "1000–3000 Hz_10ms_modulation_mae_db": 0.6311474335814827, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2759147803784214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12510127947202387, "1000–3000 Hz_40ms_envelope_correlation": 0.9898848331116845, "1000–3000 Hz_40ms_modulation_mae_db": 0.33012110188064675, "3000–6000 Hz_10ms_log_rms_mae_db": 2.455195621871696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32400120665956045, "3000–6000 Hz_10ms_envelope_correlation": 0.9267335548265809, "3000–6000 Hz_10ms_modulation_mae_db": 1.4074661319019652, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8704003974832837, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.261974671838755, "3000–6000 Hz_40ms_envelope_correlation": 0.9498728142646944, "3000–6000 Hz_40ms_modulation_mae_db": 2.0764899786718445, "mr_stft_db": 5.903858815830465, "band_mae_db": 0.2578582012859415, "rms_gain_error_db": -0.10246589235091864, "output_peak": 1.096125462230808}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 3, "at_or_above_full_scale_fraction": 5.0268096514745307e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 5.0268096514745307e-05, "longest_full_scale_run_samples": 3, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59680, "output_samples": 59680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.109040158479727, "spectral_convergence": 0.21868388022772586, "spectral_convergence_status": "ok", "amplitude_floor": 3.189740694195829e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.827561324913165, "spectral_convergence": 0.19438016466840807, "spectral_convergence_status": "ok", "amplitude_floor": 2.205927306089595e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.791638738233454, "spectral_convergence": 0.18934112328833552, "spectral_convergence_status": "ok", "amplitude_floor": 1.9576049396551346e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004284845572396228, "output_power": 0.00445322427387718, "delta_db": 0.1673939329004837, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.012674374150346739, "output_power": 0.012367429877643813, "delta_db": -0.10647066820660529, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0031087438744962816, "output_power": 0.0029882023775270198, "delta_db": -0.1717493558587958, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005542642121172655, "output_power": 0.00047083754536055754, "delta_db": -0.708457509444284, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001593466145796079, "output_power": 0.0001520129541967898, "delta_db": -0.20462241938978662, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.162629787041768e-05, "output_power": 0.00010870482645142046, "delta_db": 0.742286876202647, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.35016346033376705, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06979902040665889, "pitch": null, "output_rms_dbfs": -16.873908526284545, "output_peak": 0.9748558964818065, "reference_rms_dbfs": -16.80410950587789, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.909413407208781}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00043785897255075944, "log_rms_mae_db": 1.1290268466730426, "rms_envelope_nrmse": 0.06341837090979724, "envelope_correlation": 0.9959423889431023, "rms_gain_error_db": -0.06979932529183408, "reference_mean_power": 0.0209012125553907, "output_mean_power": 0.02056797564208879}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.005551294872497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00444581511703579, "output_power": 0.004237650685112281, "reference_active": true, "delta_db": -0.2082624382826431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00039454721092167534, "output_power": 0.00041025251846405316, "reference_active": true, "delta_db": 0.1695227811176359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.0001084016816849342, "output_power": 0.00010633931055498685, "reference_active": true, "delta_db": -0.08342179400688679}], "active_band_count": 3, "active_band_mae_db": 0.15373567113572192}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0004107294113879301, "log_rms_mae_db": 0.5663145729200888, "rms_envelope_nrmse": 0.032708547545639, "envelope_correlation": 0.9989121981644609, "rms_gain_error_db": -0.06979897488302347, "reference_mean_power": 0.021070817398047652, "output_mean_power": 0.020734878075654738}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.012296407388435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00391054354874771, "output_power": 0.0037393308722779403, "reference_active": true, "delta_db": -0.19443231536801758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019349823972573515, "output_power": 0.00019563328551909235, "reference_active": true, "delta_db": 0.04765730099734508}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9805555260958754e-06, "output_power": 4.6864637329605185e-06, "reference_active": true, "delta_db": -1.0589626640729561}], "active_band_count": 3, "active_band_mae_db": 0.43368409347943965}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.0003730035430310483, "log_rms_mae_db": 1.746729304415926, "rms_envelope_nrmse": 0.08489668465626758, "envelope_correlation": 0.9944993383986735, "rms_gain_error_db": -0.03313497226373362, "reference_mean_power": 0.007423998211444209, "output_mean_power": 0.007367571535970078}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.725628531347034e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002177971361103236, "output_power": 0.002105150426055952, "reference_active": true, "delta_db": -0.14769030454237297}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00032785256002961345, "output_power": 0.00028280424997396736, "reference_active": true, "delta_db": -0.6419264745620012}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.051887548388713e-05, "output_power": 5.434774338086321e-05, "reference_active": true, "delta_db": -1.7071618487483957}], "active_band_count": 3, "active_band_mae_db": 0.8322595426175899}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00035212990311193784, "log_rms_mae_db": 1.096697333384167, "rms_envelope_nrmse": 0.049728067258803875, "envelope_correlation": 0.9980715697513318, "rms_gain_error_db": -0.03314204836505501, "reference_mean_power": 0.007484242812710004, "output_mean_power": 0.007427346141834506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.215922155621447e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018646593945503163, "output_power": 0.0018090553304850579, "reference_active": true, "delta_db": -0.13147663497165607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015694821766981044, "output_power": 0.00014075371719439843, "reference_active": true, "delta_db": -0.47296515094576475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505413793795639e-06, "output_power": 3.10916615181038e-06, "reference_active": true, "delta_db": -3.206309951313104}], "active_band_count": 3, "active_band_mae_db": 1.2702505790768417}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00013449969207959865, "log_rms_mae_db": 1.9557594262530416, "rms_envelope_nrmse": 0.1640857797053984, "envelope_correlation": 0.9834580211939243, "rms_gain_error_db": -0.6017713369340323, "reference_mean_power": 0.0006066736643199691, "output_mean_power": 0.000528175204277914}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.4525665594490304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019778914796660728, "output_power": 0.00016511585523782885, "reference_active": true, "delta_db": -0.7841368125874775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.140914614473558e-05, "output_power": 2.411913220055268e-05, "reference_active": true, "delta_db": -1.1469445233778266}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0263986880205243e-05, "output_power": 6.623327713652771e-06, "reference_active": true, "delta_db": -1.9023984433109518}], "active_band_count": 3, "active_band_mae_db": 1.2778265930920851}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00012636239720129053, "log_rms_mae_db": 1.1312443879777871, "rms_envelope_nrmse": 0.11147491732456713, "envelope_correlation": 0.9943481445895133, "rms_gain_error_db": -0.6017845278461625, "reference_mean_power": 0.0006115927691385325, "output_mean_power": 0.0005324562011414741}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.048532457224497e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001773914005108231, "output_power": 0.00014659702377342493, "reference_active": true, "delta_db": -0.8280740919407106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4831454807134908e-05, "output_power": 1.1864830623560655e-05, "reference_active": true, "delta_db": -0.9692220993798022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.85797297398325e-07, "output_power": 4.0778185184016283e-07, "reference_active": true, "delta_db": -0.7602720061720701}], "active_band_count": 3, "active_band_mae_db": 0.8525227324975276}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 5.5319997387638595e-05, "log_rms_mae_db": 2.465661930825273, "rms_envelope_nrmse": 0.3138495741469869, "envelope_correlation": 0.9344577516313454, "rms_gain_error_db": 0.27933480432344315, "reference_mean_power": 9.493685130389184e-05, "output_mean_power": 0.0001012437689740823}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.144674032958852e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.902158206464317e-05, "output_power": 2.501948597900303e-05, "reference_active": true, "delta_db": -0.6444270066453455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.982855070886748e-06, "output_power": 1.3315154910504521e-05, "reference_active": true, "delta_db": 1.2509145729376392}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.211971373682015e-06, "output_power": 5.457815600628024e-06, "reference_active": true, "delta_db": 2.3024719235466478}], "active_band_count": 3, "active_band_mae_db": 1.3992711677098775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 4.1914797815135623e-05, "log_rms_mae_db": 1.8877282260593318, "rms_envelope_nrmse": 0.26041371009094305, "envelope_correlation": 0.9527317082395839, "rms_gain_error_db": 0.2793341143364107, "reference_mean_power": 9.570694886911743e-05, "output_mean_power": 0.0001020650100353833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.518176746918335e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6494345211181707e-05, "output_power": 2.2638881248870673e-05, "reference_active": true, "delta_db": -0.6829822929513646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.515847029872695e-06, "output_power": 7.491074335354982e-06, "reference_active": true, "delta_db": 2.198048844995597}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6527459125158204e-07, "output_power": 5.389984971034239e-07, "reference_active": true, "delta_db": 3.078919002152192}], "active_band_count": 3, "active_band_mae_db": 1.986650046699718}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8720001339433806, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "520e92408806a60675f81d675d40171deda6533cb1cacca9b8326cfd435c4f67", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9785480009986516, "standard_deviation": 0.005421946255849252, "minimum": 0.9703419783512397, "maximum": 0.98527337091766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9908789166962524, "standard_deviation": 0.04337636716620866, "minimum": 1.9098536981961174, "maximum": 2.0426624565309517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6807148575289166, "standard_deviation": 0.31452273059232605, "minimum": 0.34391408098264803, "maximum": 1.2778265930920851, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1782615285134715, "standard_deviation": 0.02059005943343855, "minimum": 0.14413716126689968, "maximum": 0.20450522891234066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930195332871954, "standard_deviation": 0.0022084141247260334, "minimum": 0.9896422947471415, "maximum": 0.9953627526605291, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2115483987935343, "standard_deviation": 0.04911526493919878, "minimum": 1.1312443879777871, "maximum": 1.2759147803784214, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6266953313822339, "standard_deviation": 0.23248565440111432, "minimum": 0.2689713642399591, "maximum": 0.8761136494966987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11191174003828012, "standard_deviation": 0.01824124675075431, "minimum": 0.0819946580926648, "maximum": 0.14116351093800109, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9299340590152199, "standard_deviation": 0.004039727084283482, "minimum": 0.9246527829633683, "maximum": 0.9344577516313454, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.4818336032452155, "standard_deviation": 0.047254303438563745, "minimum": 2.403754097672059, "maximum": 2.554746201581906, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5760868243579111, "standard_deviation": 0.20794596621384773, "minimum": 1.3992711677098775, "maximum": 2.0003354201946246, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.32394947720214695, "standard_deviation": 0.007482309100308019, "minimum": 0.3138495741469869, "maximum": 0.33343542446891455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9519404086756175, "standard_deviation": 0.002406819458870956, "minimum": 0.9484535611016802, "maximum": 0.9563905058159862, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8747731578358293, "standard_deviation": 0.04241729560311815, "minimum": 1.802752993125389, "maximum": 1.9321767784438864, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.162416206696631, "standard_deviation": 0.21297069992857173, "minimum": 1.898479486483069, "maximum": 2.598034847939525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.26239116595156253, "standard_deviation": 0.0036139195681744994, "minimum": 0.2585933220931519, "maximum": 0.26793721831885436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932869738841532, "standard_deviation": 0.0008096368506957005, "minimum": 0.9921612556812163, "maximum": 0.9944993383986735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7020869111044798, "standard_deviation": 0.0328214586799227, "minimum": 1.6621815643945008, "maximum": 1.746729304415926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0108605107964506, "standard_deviation": 0.30567326796601546, "minimum": 0.7375672703271295, "maximum": 1.6575665138626776, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09365431836857874, "standard_deviation": 0.0048792839605944615, "minimum": 0.08489668465626758, "maximum": 0.09964205549910761, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977648600831475, "standard_deviation": 0.00029185977479960723, "minimum": 0.9973848095192329, "maximum": 0.9981057431542391, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.051864717888658, "standard_deviation": 0.042552703306108226, "minimum": 1.0052962410664783, "maximum": 1.13371415576771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3843315736710053, "standard_deviation": 0.3777340499292862, "minimum": 1.0337868461453281, "maximum": 2.221469361719332, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05457604386980895, "standard_deviation": 0.003131877733506395, "minimum": 0.049728067258803875, "maximum": 0.05987528462317407, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3351218991671172, "standard_deviation": 0.05214453431465228, "minimum": 0.2578582012859415, "maximum": 0.39850247069835554, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.908330813251216, "standard_deviation": 0.010214565714852089, "minimum": 5.888776116382691, "maximum": 5.9225252507800965, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9928117137350059, "standard_deviation": 0.05852658109912127, "minimum": 0.8853682812526045, "maximum": 1.096125462230808, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12291294082937386, "standard_deviation": 0.030971242598329417, "minimum": -0.16819827332653817, "maximum": -0.06979902040665889, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9953604052887726, "standard_deviation": 0.0005785803157350952, "minimum": 0.9943957228094343, "maximum": 0.9960937273145587, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1159035151496863, "standard_deviation": 0.041624443700613974, "minimum": 1.0375960841954266, "maximum": 1.1670227729322775, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3271697413524923, "standard_deviation": 0.1996044735625499, "minimum": 0.0915588398767374, "maximum": 0.6148195119372795, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06845118995268494, "standard_deviation": 0.004430087156645403, "minimum": 0.06300652224169366, "maximum": 0.07593556704665304, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985866293291064, "standard_deviation": 0.00029643105383835955, "minimum": 0.9981421562338174, "maximum": 0.9989876638493247, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6157557832504104, "standard_deviation": 0.061942361320730985, "minimum": 0.5153659483516789, "maximum": 0.7068488280894374, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36055120048431244, "standard_deviation": 0.15323545390219867, "minimum": 0.1680547779243651, "maximum": 0.602085413929952, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03850397580591661, "standard_deviation": 0.0043569738960413645, "minimum": 0.032708547545639, "maximum": 0.04479426883636695, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0000", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.0705482959747314, "predicted_mos": 3.0705482959747314, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "bb68d5d3a5cf33643575864f768a10a98dc45e3c9813c9071f84663af46b5f5f", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.111384500283748, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1267809446845005, "全帯域_10ms_rms_envelope_nrmse": 0.07017533693804222, "全帯域_10ms_envelope_correlation": 0.9951053118484466, "全帯域_10ms_modulation_mae_db": 0.7170639374798538, "全帯域_40ms_log_rms_mae_db": 0.6316145492247145, "全帯域_40ms_rms_envelope_nrmse": 0.03540401577436102, "全帯域_40ms_envelope_correlation": 0.998821989948605, "全帯域_40ms_modulation_mae_db": 0.5029876169023151, "300–1000 Hz_10ms_log_rms_mae_db": 1.4273924038052266, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09658512498357767, "300–1000 Hz_10ms_envelope_correlation": 0.9926780863988128, "300–1000 Hz_10ms_modulation_mae_db": 0.5197305176208142, "300–1000 Hz_40ms_log_rms_mae_db": 0.8802522495064116, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050418697215183786, "300–1000 Hz_40ms_envelope_correlation": 0.9979608840335539, "300–1000 Hz_40ms_modulation_mae_db": 0.7205242480652451, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5669896028688826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15186336468273615, "1000–3000 Hz_10ms_envelope_correlation": 0.9835501636774026, "1000–3000 Hz_10ms_modulation_mae_db": 0.20677424862800778, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9903838933854144, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08258710573218482, "1000–3000 Hz_40ms_envelope_correlation": 0.9952689986456554, "1000–3000 Hz_40ms_modulation_mae_db": 0.4346362228879214, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7800042234245002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3011889185243542, "3000–6000 Hz_10ms_envelope_correlation": 0.9367236059671118, "3000–6000 Hz_10ms_modulation_mae_db": 1.5568186540747349, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4663065036978968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25773874197838115, "3000–6000 Hz_40ms_envelope_correlation": 0.9509075562364948, "3000–6000 Hz_40ms_modulation_mae_db": 2.1566254974125383, "mr_stft_db": 5.468726297066037, "band_mae_db": 0.22423851568434175, "rms_gain_error_db": -0.11872985004476891, "output_peak": 1.02687751337168}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 3.351206434316354e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.351206434316354e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.253775714726038, "全帯域_10ms_rms_envelope_nrmse": 0.0713951364383694, "全帯域_10ms_envelope_correlation": 0.9948887892858571, "全帯域_10ms_modulation_mae_db": 0.23633803895352337, "全帯域_40ms_log_rms_mae_db": 0.6874254648491951, "全帯域_40ms_rms_envelope_nrmse": 0.0405876196451248, "全帯域_40ms_envelope_correlation": 0.9983417402027042, "全帯域_40ms_modulation_mae_db": 0.1520544759863759, "300–1000 Hz_10ms_log_rms_mae_db": 1.4574030329712804, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09804684260023333, "300–1000 Hz_10ms_envelope_correlation": 0.9924619579916818, "300–1000 Hz_10ms_modulation_mae_db": 0.7062971409443981, "300–1000 Hz_40ms_log_rms_mae_db": 0.9548127781210213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056150594638025776, "300–1000 Hz_40ms_envelope_correlation": 0.9974636423857344, "300–1000 Hz_40ms_modulation_mae_db": 0.719591862275248, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4799519789708127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1342442412709133, "1000–3000 Hz_10ms_envelope_correlation": 0.9880011154193653, "1000–3000 Hz_10ms_modulation_mae_db": 1.0683229542693324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9637255080118785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0778864842262564, "1000–3000 Hz_40ms_envelope_correlation": 0.9967970520961279, "1000–3000 Hz_40ms_modulation_mae_db": 0.6419601912913765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7020647915059357, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2976875725520977, "3000–6000 Hz_10ms_envelope_correlation": 0.9414409544151847, "3000–6000 Hz_10ms_modulation_mae_db": 1.5463038802638307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4320582980969878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2527863354012423, "3000–6000 Hz_40ms_envelope_correlation": 0.9552159323293367, "3000–6000 Hz_40ms_modulation_mae_db": 1.9851301652279183, "mr_stft_db": 5.467988498359149, "band_mae_db": 0.29038810937611254, "rms_gain_error_db": -0.10309874521576928, "output_peak": 1.0109925241680935}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.258496674796481, "全帯域_10ms_rms_envelope_nrmse": 0.06740011364508045, "全帯域_10ms_envelope_correlation": 0.9954611052072685, "全帯域_10ms_modulation_mae_db": 0.2278545455096872, "全帯域_40ms_log_rms_mae_db": 0.6794247124263506, "全帯域_40ms_rms_envelope_nrmse": 0.035138180001246966, "全帯域_40ms_envelope_correlation": 0.9987959105630719, "全帯域_40ms_modulation_mae_db": 0.2120358136407702, "300–1000 Hz_10ms_log_rms_mae_db": 1.3696160835679871, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0984282206969644, "300–1000 Hz_10ms_envelope_correlation": 0.9923234379839747, "300–1000 Hz_10ms_modulation_mae_db": 0.7814671792757685, "300–1000 Hz_40ms_log_rms_mae_db": 0.9220269311429792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058664647552602374, "300–1000 Hz_40ms_envelope_correlation": 0.9970946154181981, "300–1000 Hz_40ms_modulation_mae_db": 0.8106756758544397, "1000–3000 Hz_10ms_log_rms_mae_db": 1.504237721765677, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1584000719947894, "1000–3000 Hz_10ms_envelope_correlation": 0.9841989377538649, "1000–3000 Hz_10ms_modulation_mae_db": 0.5698414827277863, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9517674669087776, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10046428165528605, "1000–3000 Hz_40ms_envelope_correlation": 0.9955966803455872, "1000–3000 Hz_40ms_modulation_mae_db": 0.47104077976392794, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6955319585438708, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29515426781487125, "3000–6000 Hz_10ms_envelope_correlation": 0.939666039865196, "3000–6000 Hz_10ms_modulation_mae_db": 1.7905979158297332, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3950641223080795, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25271377515761273, "3000–6000 Hz_40ms_envelope_correlation": 0.9530974029711531, "3000–6000 Hz_40ms_modulation_mae_db": 2.231172065454146, "mr_stft_db": 5.4616748962727435, "band_mae_db": 0.3344856175595797, "rms_gain_error_db": -0.10325935823604944, "output_peak": 1.0381185873765437}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1843948016037236, "全帯域_10ms_rms_envelope_nrmse": 0.07009994341802303, "全帯域_10ms_envelope_correlation": 0.994994524761445, "全帯域_10ms_modulation_mae_db": 0.5248865020686942, "全帯域_40ms_log_rms_mae_db": 0.6261011530068851, "全帯域_40ms_rms_envelope_nrmse": 0.03457503873338389, "全帯域_40ms_envelope_correlation": 0.9987143650260036, "全帯域_40ms_modulation_mae_db": 0.4245930192331371, "300–1000 Hz_10ms_log_rms_mae_db": 1.5152580503849762, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08765767816056191, "300–1000 Hz_10ms_envelope_correlation": 0.9939861622042866, "300–1000 Hz_10ms_modulation_mae_db": 0.8102568623832985, "300–1000 Hz_40ms_log_rms_mae_db": 0.9801315333307555, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04874662500791948, "300–1000 Hz_40ms_envelope_correlation": 0.9980231405892993, "300–1000 Hz_40ms_modulation_mae_db": 1.074200492911275, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5289785944317886, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14099217412033407, "1000–3000 Hz_10ms_envelope_correlation": 0.9879635294730333, "1000–3000 Hz_10ms_modulation_mae_db": 0.5735188189736936, "1000–3000 Hz_40ms_log_rms_mae_db": 1.048347430243155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0943401737940517, "1000–3000 Hz_40ms_envelope_correlation": 0.996203671634159, "1000–3000 Hz_40ms_modulation_mae_db": 0.460618553103384, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7090133526998954, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28357547609539147, "3000–6000 Hz_10ms_envelope_correlation": 0.9451680375536451, "3000–6000 Hz_10ms_modulation_mae_db": 1.7383820900451046, "3000–6000 Hz_40ms_log_rms_mae_db": 1.362152365397205, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23902519140495435, "3000–6000 Hz_40ms_envelope_correlation": 0.9586936223259335, "3000–6000 Hz_40ms_modulation_mae_db": 2.2656567367671148, "mr_stft_db": 5.465556989036528, "band_mae_db": 0.23861377076305704, "rms_gain_error_db": -0.0375847149757566, "output_peak": 0.9967945268109311}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1929088325579829, "全帯域_10ms_rms_envelope_nrmse": 0.06549182474010477, "全帯域_10ms_envelope_correlation": 0.9957707570334899, "全帯域_10ms_modulation_mae_db": 0.6027139676079544, "全帯域_40ms_log_rms_mae_db": 0.6238909385183503, "全帯域_40ms_rms_envelope_nrmse": 0.03553226741647621, "全帯域_40ms_envelope_correlation": 0.998829592035534, "全帯域_40ms_modulation_mae_db": 0.5180049665557893, "300–1000 Hz_10ms_log_rms_mae_db": 1.460346250040985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10673830756470463, "300–1000 Hz_10ms_envelope_correlation": 0.991037961587557, "300–1000 Hz_10ms_modulation_mae_db": 0.41736511960926165, "300–1000 Hz_40ms_log_rms_mae_db": 0.9392517219471745, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0641653722365868, "300–1000 Hz_40ms_envelope_correlation": 0.9965970485683335, "300–1000 Hz_40ms_modulation_mae_db": 0.44733286815170886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.53221973106608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14542005340321254, "1000–3000 Hz_10ms_envelope_correlation": 0.985276192998521, "1000–3000 Hz_10ms_modulation_mae_db": 0.4477469240388084, "1000–3000 Hz_40ms_log_rms_mae_db": 1.039554722741665, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0828846113327623, "1000–3000 Hz_40ms_envelope_correlation": 0.995635556007418, "1000–3000 Hz_40ms_modulation_mae_db": 0.3456947402131787, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6833677013083255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29572292471013273, "3000–6000 Hz_10ms_envelope_correlation": 0.9407120220279165, "3000–6000 Hz_10ms_modulation_mae_db": 1.7816214701636166, "3000–6000 Hz_40ms_log_rms_mae_db": 1.367976328398849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25175178272736626, "3000–6000 Hz_40ms_envelope_correlation": 0.9543767479114063, "3000–6000 Hz_40ms_modulation_mae_db": 2.2804602443384994, "mr_stft_db": 5.479561817097507, "band_mae_db": 0.25566657455400843, "rms_gain_error_db": -0.12205263871645299, "output_peak": 0.9514321825850983}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0710202686772934, "全帯域_10ms_rms_envelope_nrmse": 0.05786123055584296, "全帯域_10ms_envelope_correlation": 0.9966214672067187, "全帯域_10ms_modulation_mae_db": 0.3651240194398155, "全帯域_40ms_log_rms_mae_db": 0.5028723923894068, "全帯域_40ms_rms_envelope_nrmse": 0.029574494175354186, "全帯域_40ms_envelope_correlation": 0.9990992805303242, "全帯域_40ms_modulation_mae_db": 0.18715028482031348, "300–1000 Hz_10ms_log_rms_mae_db": 1.4386689017201548, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08769043560987742, "300–1000 Hz_10ms_envelope_correlation": 0.9939942229091239, "300–1000 Hz_10ms_modulation_mae_db": 0.3720633579778458, "300–1000 Hz_40ms_log_rms_mae_db": 0.8716140942743714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05252708034068754, "300–1000 Hz_40ms_envelope_correlation": 0.9977106586440974, "300–1000 Hz_40ms_modulation_mae_db": 0.6379096405590338, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5219761924244208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16978585843370522, "1000–3000 Hz_10ms_envelope_correlation": 0.9792789106711116, "1000–3000 Hz_10ms_modulation_mae_db": 0.48563778179343337, "1000–3000 Hz_40ms_log_rms_mae_db": 1.000260689857958, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08529086561210925, "1000–3000 Hz_40ms_envelope_correlation": 0.9948091201040926, "1000–3000 Hz_40ms_modulation_mae_db": 0.6329584744870633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7372971721507946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3077935618098919, "3000–6000 Hz_10ms_envelope_correlation": 0.9393185261267936, "3000–6000 Hz_10ms_modulation_mae_db": 2.0054508740141763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4308565584524147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2577346228072937, "3000–6000 Hz_40ms_envelope_correlation": 0.9551038037680114, "3000–6000 Hz_40ms_modulation_mae_db": 2.504530289170537, "mr_stft_db": 5.478784051015304, "band_mae_db": 0.2096742382161553, "rms_gain_error_db": -0.06318477870282119, "output_peak": 1.0346599129739769}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 3, "at_or_above_full_scale_fraction": 5.0268096514745307e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 5.0268096514745307e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0529298391624202, "全帯域_10ms_rms_envelope_nrmse": 0.0678093081392164, "全帯域_10ms_envelope_correlation": 0.9953193752866161, "全帯域_10ms_modulation_mae_db": 0.2221259400052439, "全帯域_40ms_log_rms_mae_db": 0.5621017620209435, "全帯域_40ms_rms_envelope_nrmse": 0.03483896830154592, "全帯域_40ms_envelope_correlation": 0.9986799054611384, "全帯域_40ms_modulation_mae_db": 0.3692628346952878, "300–1000 Hz_10ms_log_rms_mae_db": 1.3943662742311917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08106426708408346, "300–1000 Hz_10ms_envelope_correlation": 0.9948264035390804, "300–1000 Hz_10ms_modulation_mae_db": 0.8345680787502966, "300–1000 Hz_40ms_log_rms_mae_db": 0.8523128461546113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040490237542040695, "300–1000 Hz_40ms_envelope_correlation": 0.998630719933858, "300–1000 Hz_40ms_modulation_mae_db": 1.0945915164140085, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5291299044456934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16291864600023967, "1000–3000 Hz_10ms_envelope_correlation": 0.9827686811162999, "1000–3000 Hz_10ms_modulation_mae_db": 0.44943011674241085, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9630826338746522, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09907537423902883, "1000–3000 Hz_40ms_envelope_correlation": 0.9954268493595113, "1000–3000 Hz_40ms_modulation_mae_db": 0.3305736118794473, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7244957032201516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.296808827263061, "3000–6000 Hz_10ms_envelope_correlation": 0.9390268979145752, "3000–6000 Hz_10ms_modulation_mae_db": 1.5909369315469, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3223490150137838, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24985630414586435, "3000–6000 Hz_40ms_envelope_correlation": 0.954247775578035, "3000–6000 Hz_40ms_modulation_mae_db": 2.131470790730907, "mr_stft_db": 5.478956119917381, "band_mae_db": 0.19101242085490802, "rms_gain_error_db": -0.0007194167834606884, "output_peak": 1.0336166832049747}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 3.351206434316354e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.351206434316354e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.21185358206208, "全帯域_10ms_rms_envelope_nrmse": 0.07619181736011199, "全帯域_10ms_envelope_correlation": 0.9940868997121893, "全帯域_10ms_modulation_mae_db": 0.7338497541166644, "全帯域_40ms_log_rms_mae_db": 0.674598765009667, "全帯域_40ms_rms_envelope_nrmse": 0.050068305500107935, "全帯域_40ms_envelope_correlation": 0.997289373360894, "全帯域_40ms_modulation_mae_db": 0.5224778791300582, "300–1000 Hz_10ms_log_rms_mae_db": 1.4846388679648639, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1049976050918286, "300–1000 Hz_10ms_envelope_correlation": 0.9914008902258831, "300–1000 Hz_10ms_modulation_mae_db": 0.2627213276190092, "300–1000 Hz_40ms_log_rms_mae_db": 0.9326255396744172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06961530225796918, "300–1000 Hz_40ms_envelope_correlation": 0.9960933632895486, "300–1000 Hz_40ms_modulation_mae_db": 0.3355756107217002, "1000–3000 Hz_10ms_log_rms_mae_db": 1.520004146616031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14350103593166386, "1000–3000 Hz_10ms_envelope_correlation": 0.9872932681075749, "1000–3000 Hz_10ms_modulation_mae_db": 0.9942592379121988, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9695275546282398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0899370528657143, "1000–3000 Hz_40ms_envelope_correlation": 0.9966842875764226, "1000–3000 Hz_40ms_modulation_mae_db": 0.7382793773457691, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7224788209256228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2890517704249047, "3000–6000 Hz_10ms_envelope_correlation": 0.9420069397667257, "3000–6000 Hz_10ms_modulation_mae_db": 1.6819496138666807, "3000–6000 Hz_40ms_log_rms_mae_db": 1.378472492021456, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24435264571334467, "3000–6000 Hz_40ms_envelope_correlation": 0.956050271873842, "3000–6000 Hz_40ms_modulation_mae_db": 2.0313234003115124, "mr_stft_db": 5.47221517899744, "band_mae_db": 0.3153460792291207, "rms_gain_error_db": -0.04705868605064481, "output_peak": 0.9732188751414469}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59680, "output_samples": 59680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.56159197710853, "spectral_convergence": 0.21537505164189658, "spectral_convergence_status": "ok", "amplitude_floor": 3.189740694195829e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.274729849858236, "spectral_convergence": 0.19441786709422404, "spectral_convergence_status": "ok", "amplitude_floor": 2.205927306089595e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.569857064231343, "spectral_convergence": 0.19373159787391359, "spectral_convergence_status": "ok", "amplitude_floor": 1.9576049396551346e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004284845572396228, "output_power": 0.0043104040411516545, "delta_db": 0.02582806820253382, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.012674374150346739, "output_power": 0.012361481684245782, "delta_db": -0.10855993737109214, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0031087438744962816, "output_power": 0.0028656389003140106, "delta_db": -0.3536347855881412, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005542642121172655, "output_power": 0.0005269511083124763, "delta_db": -0.21946515572586595, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001593466145796079, "output_power": 0.0001475462467159736, "delta_db": -0.33414674658884425, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.162629787041768e-05, "output_power": 9.826520838285247e-05, "delta_db": 0.30379640062957336, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22423851568434175, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11872985004476891, "pitch": null, "output_rms_dbfs": -16.922839355922655, "output_peak": 1.02687751337168, "reference_rms_dbfs": -16.80410950587789, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.468726297066037}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00043785897255075944, "log_rms_mae_db": 1.1267809446845005, "rms_envelope_nrmse": 0.07017533693804222, "envelope_correlation": 0.9951053118484466, "rms_gain_error_db": -0.1187310404498729, "reference_mean_power": 0.0209012125553907, "output_mean_power": 0.020337538018255833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.005551294872497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00444581511703579, "output_power": 0.0042022350548218375, "reference_active": true, "delta_db": -0.24471056935856775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00039454721092167534, "output_power": 0.0004522446044983947, "reference_active": true, "delta_db": 0.5927441654023214}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.0001084016816849342, "output_power": 0.00014669311385128703, "reference_active": true, "delta_db": 1.313737077678672}], "active_band_count": 3, "active_band_mae_db": 0.7170639374798538}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0004107294113879301, "log_rms_mae_db": 0.6316145492247145, "rms_envelope_nrmse": 0.03540401577436102, "envelope_correlation": 0.998821989948605, "rms_gain_error_db": -0.11872962701068318, "reference_mean_power": 0.021070817398047652, "output_mean_power": 0.020502575543894917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.012296407388435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00391054354874771, "output_power": 0.003727141363658052, "reference_active": true, "delta_db": -0.20861261596145925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019349823972573515, "output_power": 0.00022618533400853527, "reference_active": true, "delta_db": 0.6778742304070572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9805555260958754e-06, "output_power": 6.90222523764948e-06, "reference_active": true, "delta_db": 0.6224760043384289}], "active_band_count": 3, "active_band_mae_db": 0.5029876169023151}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.0003730035430310483, "log_rms_mae_db": 1.4273924038052266, "rms_envelope_nrmse": 0.09658512498357767, "envelope_correlation": 0.9926780863988128, "rms_gain_error_db": -0.08933380531621266, "reference_mean_power": 0.007423998211444209, "output_mean_power": 0.007272847447180678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.725628531347034e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002177971361103236, "output_power": 0.002055749112297809, "reference_active": true, "delta_db": -0.2508205337273812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00032785256002961345, "output_power": 0.00028519264050258085, "reference_active": true, "delta_db": -0.6054026499381506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.051887548388713e-05, "output_power": 6.848585216668007e-05, "reference_active": true, "delta_db": -0.7029683691969107}], "active_band_count": 3, "active_band_mae_db": 0.5197305176208142}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00035212990311193784, "log_rms_mae_db": 0.8802522495064116, "rms_envelope_nrmse": 0.050418697215183786, "envelope_correlation": 0.9979608840335539, "rms_gain_error_db": -0.0893331795916067, "reference_mean_power": 0.007484242812710004, "output_mean_power": 0.007331866539654145}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.215922155621447e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018646593945503163, "output_power": 0.0017742451884746643, "reference_active": true, "delta_db": -0.21585877431944897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015694821766981044, "output_power": 0.00014198878055819242, "reference_active": true, "delta_db": -0.4350235878963865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505413793795639e-06, "output_power": 4.594157585900966e-06, "reference_active": true, "delta_db": -1.5106903819798998}], "active_band_count": 3, "active_band_mae_db": 0.7205242480652451}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00013449969207959865, "log_rms_mae_db": 1.5669896028688826, "rms_envelope_nrmse": 0.15186336468273615, "envelope_correlation": 0.9835501636774026, "rms_gain_error_db": -0.24696314014380105, "reference_mean_power": 0.0006066736643199691, "output_mean_power": 0.0005731375048169028}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.4525665594490304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019778914796660728, "output_power": 0.00018281470635157313, "reference_active": true, "delta_db": -0.3419133041847441}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.140914614473558e-05, "output_power": 2.9796295423529925e-05, "reference_active": true, "delta_db": -0.2289385874120958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0263986880205243e-05, "output_power": 1.0147732321407487e-05, "reference_active": true, "delta_db": -0.04947085428718344}], "active_band_count": 3, "active_band_mae_db": 0.20677424862800778}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00012636239720129053, "log_rms_mae_db": 0.9903838933854144, "rms_envelope_nrmse": 0.08258710573218482, "envelope_correlation": 0.9952689986456554, "rms_gain_error_db": -0.2469650350204221, "reference_mean_power": 0.0006115927691385325, "output_mean_power": 0.0005777844355947467}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.048532457224497e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001773914005108231, "output_power": 0.00016518283814235794, "reference_active": true, "delta_db": -0.30967638689049487}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4831454807134908e-05, "output_power": 1.4234739903415581e-05, "reference_active": true, "delta_db": -0.17834216618404206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.85797297398325e-07, "output_power": 5.861976943892616e-07, "reference_active": true, "delta_db": 0.8158901155892272}], "active_band_count": 3, "active_band_mae_db": 0.4346362228879214}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 5.5319997387638595e-05, "log_rms_mae_db": 1.7800042234245002, "rms_envelope_nrmse": 0.3011889185243542, "envelope_correlation": 0.9367236059671118, "rms_gain_error_db": -0.07090185513387667, "reference_mean_power": 9.493685130389184e-05, "output_mean_power": 9.339951868794823e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.144674032958852e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.902158206464317e-05, "output_power": 2.677778067044032e-05, "reference_active": true, "delta_db": -0.34946503558746594}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.982855070886748e-06, "output_power": 1.129915485141111e-05, "reference_active": true, "delta_db": 0.5379119439707193}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.211971373682015e-06, "output_power": 7.675020855911333e-06, "reference_active": true, "delta_db": 3.7830789826660194}], "active_band_count": 3, "active_band_mae_db": 1.5568186540747349}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 4.1914797815135623e-05, "log_rms_mae_db": 1.4663065036978968, "rms_envelope_nrmse": 0.25773874197838115, "envelope_correlation": 0.9509075562364948, "rms_gain_error_db": -0.07090304503086428, "reference_mean_power": 9.570694886911743e-05, "output_mean_power": 9.41571201020447e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.518176746918335e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6494345211181707e-05, "output_power": 2.4283959087133904e-05, "reference_active": true, "delta_db": -0.3783369818948874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.515847029872695e-06, "output_power": 6.7018851960246235e-06, "reference_active": true, "delta_db": 1.714577618747502}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6527459125158204e-07, "output_power": 7.267613838216747e-07, "reference_active": true, "delta_db": 4.376961891595225}], "active_band_count": 3, "active_band_mae_db": 2.1566254974125383}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8992087204479589, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "eacc5e3124f03aa326e64b4163f0335b5f6e63b15b12b353b7669b13b5f636ad", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9847913499021467, "standard_deviation": 0.0030054630290106995, "minimum": 0.9792789106711116, "maximum": 0.9880011154193653, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5229359840736731, "standard_deviation": 0.02480942693988072, "minimum": 1.4799519789708127, "maximum": 1.5669896028688826, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5994414456357089, "standard_deviation": 0.2903034193003907, "minimum": 0.20677424862800778, "maximum": 1.0683229542693324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1508906807296993, "standard_deviation": 0.012063260360030041, "minimum": 0.1342442412709133, "maximum": 0.16978585843370522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9958027769711217, "standard_deviation": 0.0006979965218269061, "minimum": 0.9948091201040926, "maximum": 0.9967970520961279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9908312374564676, "standard_deviation": 0.03636272774074816, "minimum": 0.9517674669087776, "maximum": 1.048347430243155, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5069702438715085, "standard_deviation": 0.1480330134341838, "minimum": 0.3305736118794473, "maximum": 0.7382793773457691, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08905824368217422, "standard_deviation": 0.008264402214508111, "minimum": 0.0778864842262564, "maximum": 0.10046428165528605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9405078779546436, "standard_deviation": 0.0024932436068712603, "minimum": 0.9367236059671118, "maximum": 0.9451680375536451, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7192817154723872, "standard_deviation": 0.030011702277785494, "minimum": 1.6833677013083255, "maximum": 1.7800042234245002, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7115076787255972, "standard_deviation": 0.15349640249032206, "minimum": 1.5463038802638307, "maximum": 2.0054508740141763, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2958729148993381, "standard_deviation": 0.007297989893734393, "minimum": 0.28357547609539147, "maximum": 0.3077935618098919, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9547116391242765, "standard_deviation": 0.002250725568411429, "minimum": 0.9509075562364948, "maximum": 0.9586936223259335, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.394404460423334, "standard_deviation": 0.0464471505343747, "minimum": 1.3223490150137838, "maximum": 1.4663065036978968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.1982961486766466, "standard_deviation": 0.1628628998309568, "minimum": 1.9851301652279183, "maximum": 2.504530289170537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.25074492491700745, "standard_deviation": 0.006390057883225618, "minimum": 0.23902519140495435, "maximum": 0.25773874197838115, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.99283864035505, "standard_deviation": 0.0013280517430786234, "minimum": 0.991037961587557, "maximum": 0.9948264035390804, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4434612330858334, "standard_deviation": 0.04701740251797143, "minimum": 1.3696160835679871, "maximum": 1.5152580503849762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5880586980225866, "standard_deviation": 0.22291237820145088, "minimum": 0.2627213276190092, "maximum": 0.8345680787502966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09515106022397893, "standard_deviation": 0.008965410198470417, "minimum": 0.08106426708408346, "maximum": 0.10673830756470463, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974467591078279, "standard_deviation": 0.0008234010235719866, "minimum": 0.9960933632895486, "maximum": 0.998630719933858, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9166284617689677, "standard_deviation": 0.04435822439220423, "minimum": 0.8523128461546113, "maximum": 0.9801315333307555, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7300502393690824, "standard_deviation": 0.26781111612385633, "minimum": 0.3355756107217002, "maximum": 1.0945915164140085, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.055097319598876954, "standard_deviation": 0.00917317846008367, "minimum": 0.040490237542040695, "maximum": 0.06961530225796918, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25742816577966043, "standard_deviation": 0.051448753648130424, "minimum": 0.19101242085490802, "maximum": 0.3344856175595797, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.471682980970261, "standard_deviation": 0.006821853092892179, "minimum": 5.4616748962727435, "maximum": 5.479561817097507, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 1.0082138507040932, "standard_deviation": 0.03201395191758668, "minimum": 0.9514321825850983, "maximum": 1.0381185873765437, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07446102359071549, "standard_deviation": 0.04399656239351844, "minimum": -0.12205263871645299, "maximum": -0.0007194167834606884, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952810287927539, "standard_deviation": 0.0007332183417116311, "minimum": 0.9940868997121893, "maximum": 0.9966214672067187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.169020082283815, "standard_deviation": 0.0780991985375672, "minimum": 1.0529298391624202, "maximum": 1.258496674796481, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4537445881476796, "standard_deviation": 0.21871107948592697, "minimum": 0.2221259400052439, "maximum": 0.7338497541166644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0683030889043489, "standard_deviation": 0.005298538736275452, "minimum": 0.05786123055584296, "maximum": 0.07619181736011199, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985715196410344, "standard_deviation": 0.0005589619897501348, "minimum": 0.997289373360894, "maximum": 0.9990992805303242, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6235037171806891, "standard_deviation": 0.06349201330530609, "minimum": 0.5028723923894068, "maximum": 0.6874254648491951, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36107086137050587, "standard_deviation": 0.15632874654663548, "minimum": 0.1520544759863759, "maximum": 0.5224778791300582, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03696486119345011, "standard_deviation": 0.006065101055188893, "minimum": 0.029574494175354186, "maximum": 0.050068305500107935, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0000", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.661423683166504, "predicted_mos": 2.661423683166504, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "875e728789bb4dafb59e40de8dd85a29e958fee7755ff817b02a10641cb53e54", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.855871899984777, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0830353657937166, "全帯域_10ms_rms_envelope_nrmse": 0.05999301911278997, "全帯域_10ms_envelope_correlation": 0.9966228823348916, "全帯域_10ms_modulation_mae_db": 0.26996623407858805, "全帯域_40ms_log_rms_mae_db": 0.5982089490057763, "全帯域_40ms_rms_envelope_nrmse": 0.03296440296824048, "全帯域_40ms_envelope_correlation": 0.9991819606172591, "全帯域_40ms_modulation_mae_db": 0.2402284508534095, "300–1000 Hz_10ms_log_rms_mae_db": 1.5810854897488154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08502188435995693, "300–1000 Hz_10ms_envelope_correlation": 0.9944008451792308, "300–1000 Hz_10ms_modulation_mae_db": 0.46469227222132553, "300–1000 Hz_40ms_log_rms_mae_db": 0.9814672788096308, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05116186448855497, "300–1000 Hz_40ms_envelope_correlation": 0.9979121261542526, "300–1000 Hz_40ms_modulation_mae_db": 0.9920316014118314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7252752612853925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17637119347295235, "1000–3000 Hz_10ms_envelope_correlation": 0.979994835799342, "1000–3000 Hz_10ms_modulation_mae_db": 0.4379902123362087, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1528797250901184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11836933269113144, "1000–3000 Hz_40ms_envelope_correlation": 0.9929376842996878, "1000–3000 Hz_40ms_modulation_mae_db": 0.7039991203312542, "3000–6000 Hz_10ms_log_rms_mae_db": 2.026988224977522, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30954476255931573, "3000–6000 Hz_10ms_envelope_correlation": 0.937283092538675, "3000–6000 Hz_10ms_modulation_mae_db": 1.8810609923324657, "3000–6000 Hz_40ms_log_rms_mae_db": 1.635976551612618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26172911886037703, "3000–6000 Hz_40ms_envelope_correlation": 0.9527306300212929, "3000–6000 Hz_40ms_modulation_mae_db": 2.396312858457405, "mr_stft_db": 5.642779679548096, "band_mae_db": 0.39194420912138317, "rms_gain_error_db": -0.16263577389065972, "output_peak": 0.9692718217192645}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.190836013353681, "全帯域_10ms_rms_envelope_nrmse": 0.06861197727252474, "全帯域_10ms_envelope_correlation": 0.9952636500157604, "全帯域_10ms_modulation_mae_db": 0.5536168888010944, "全帯域_40ms_log_rms_mae_db": 0.6674741974770456, "全帯域_40ms_rms_envelope_nrmse": 0.041160083426977655, "全帯域_40ms_envelope_correlation": 0.9982581803078796, "全帯域_40ms_modulation_mae_db": 0.32343032659418863, "300–1000 Hz_10ms_log_rms_mae_db": 1.5737343027928772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09899372830389318, "300–1000 Hz_10ms_envelope_correlation": 0.9923175437384162, "300–1000 Hz_10ms_modulation_mae_db": 0.999427266494289, "300–1000 Hz_40ms_log_rms_mae_db": 0.9954726971717173, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05448065868140355, "300–1000 Hz_40ms_envelope_correlation": 0.9976012805342395, "300–1000 Hz_40ms_modulation_mae_db": 1.0679777501124674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.807995490374177, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16744418479801818, "1000–3000 Hz_10ms_envelope_correlation": 0.9797282671789135, "1000–3000 Hz_10ms_modulation_mae_db": 0.46572931864239037, "1000–3000 Hz_40ms_log_rms_mae_db": 1.177250389085997, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0897299877882321, "1000–3000 Hz_40ms_envelope_correlation": 0.9940184838269954, "1000–3000 Hz_40ms_modulation_mae_db": 0.6000062859687371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9578143459252093, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30062521665143455, "3000–6000 Hz_10ms_envelope_correlation": 0.9384378093422607, "3000–6000 Hz_10ms_modulation_mae_db": 1.490139030184558, "3000–6000 Hz_40ms_log_rms_mae_db": 1.504495544105067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2441617942462889, "3000–6000 Hz_40ms_envelope_correlation": 0.957178069448276, "3000–6000 Hz_40ms_modulation_mae_db": 2.07814690031833, "mr_stft_db": 5.623756407197205, "band_mae_db": 0.15942769026352402, "rms_gain_error_db": -0.0859652237310604, "output_peak": 0.9813128484481234}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1468550696695778, "全帯域_10ms_rms_envelope_nrmse": 0.07652440319640105, "全帯域_10ms_envelope_correlation": 0.9941336981099241, "全帯域_10ms_modulation_mae_db": 0.1572062299711706, "全帯域_40ms_log_rms_mae_db": 0.6461678615174941, "全帯域_40ms_rms_envelope_nrmse": 0.03970408239045055, "全帯域_40ms_envelope_correlation": 0.9984481104632963, "全帯域_40ms_modulation_mae_db": 0.19084078962542864, "300–1000 Hz_10ms_log_rms_mae_db": 1.5447006241749348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09862171068202079, "300–1000 Hz_10ms_envelope_correlation": 0.9923229073330472, "300–1000 Hz_10ms_modulation_mae_db": 0.4337257929994229, "300–1000 Hz_40ms_log_rms_mae_db": 0.9241031257351859, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04870026335630673, "300–1000 Hz_40ms_envelope_correlation": 0.9980519223297216, "300–1000 Hz_40ms_modulation_mae_db": 1.061432606351142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7368823196760441, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15603970244761384, "1000–3000 Hz_10ms_envelope_correlation": 0.9863282465457262, "1000–3000 Hz_10ms_modulation_mae_db": 0.8762995119119129, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0907427970977341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10850901961451666, "1000–3000 Hz_40ms_envelope_correlation": 0.9959836387200272, "1000–3000 Hz_40ms_modulation_mae_db": 0.8103404666355769, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9852531788389796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2962182929154833, "3000–6000 Hz_10ms_envelope_correlation": 0.9404064207965573, "3000–6000 Hz_10ms_modulation_mae_db": 1.5011055041832184, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6052037845590243, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24523628207001022, "3000–6000 Hz_40ms_envelope_correlation": 0.9569755433884184, "3000–6000 Hz_40ms_modulation_mae_db": 2.0575289614659638, "mr_stft_db": 5.631581708276876, "band_mae_db": 0.3019964612947274, "rms_gain_error_db": -0.1130203089162613, "output_peak": 0.9520088145775584}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0373674464917637, "全帯域_10ms_rms_envelope_nrmse": 0.06361405594874768, "全帯域_10ms_envelope_correlation": 0.9962722574727867, "全帯域_10ms_modulation_mae_db": 0.3559929173807286, "全帯域_40ms_log_rms_mae_db": 0.5822695935610215, "全帯域_40ms_rms_envelope_nrmse": 0.039323291851560235, "全帯域_40ms_envelope_correlation": 0.9988010177568734, "全帯域_40ms_modulation_mae_db": 0.3954955387405766, "300–1000 Hz_10ms_log_rms_mae_db": 1.5203807564053957, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09066677635967983, "300–1000 Hz_10ms_envelope_correlation": 0.9936482315845573, "300–1000 Hz_10ms_modulation_mae_db": 1.2270083816635744, "300–1000 Hz_40ms_log_rms_mae_db": 0.953365311711115, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052876727685638066, "300–1000 Hz_40ms_envelope_correlation": 0.9978165004064795, "300–1000 Hz_40ms_modulation_mae_db": 1.4492860906506142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7415693528831857, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1404300108443828, "1000–3000 Hz_10ms_envelope_correlation": 0.9865826193846841, "1000–3000 Hz_10ms_modulation_mae_db": 0.8879186399846759, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0943144742389823, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08688699006747541, "1000–3000 Hz_40ms_envelope_correlation": 0.9954169857746297, "1000–3000 Hz_40ms_modulation_mae_db": 0.735744450603863, "3000–6000 Hz_10ms_log_rms_mae_db": 1.995645667552954, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.295942780419793, "3000–6000 Hz_10ms_envelope_correlation": 0.9416575544637849, "3000–6000 Hz_10ms_modulation_mae_db": 1.677241330701067, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5962876886694022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24855070627969536, "3000–6000 Hz_40ms_envelope_correlation": 0.9565642343003603, "3000–6000 Hz_40ms_modulation_mae_db": 2.3153653915516244, "mr_stft_db": 5.631501718352365, "band_mae_db": 0.31082812962557405, "rms_gain_error_db": -0.18937118379319207, "output_peak": 0.9500809496669811}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9902142046864346, "全帯域_10ms_rms_envelope_nrmse": 0.06457548833190045, "全帯域_10ms_envelope_correlation": 0.9958609956448472, "全帯域_10ms_modulation_mae_db": 0.24498961962075963, "全帯域_40ms_log_rms_mae_db": 0.4720306207220476, "全帯域_40ms_rms_envelope_nrmse": 0.03495559052173846, "全帯域_40ms_envelope_correlation": 0.9988296522067722, "全帯域_40ms_modulation_mae_db": 0.15583127979041686, "300–1000 Hz_10ms_log_rms_mae_db": 1.6460118007255131, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09445529034571393, "300–1000 Hz_10ms_envelope_correlation": 0.9931688532523614, "300–1000 Hz_10ms_modulation_mae_db": 0.6287210342762815, "300–1000 Hz_40ms_log_rms_mae_db": 1.0698624660283331, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05882523465880995, "300–1000 Hz_40ms_envelope_correlation": 0.9973921637306934, "300–1000 Hz_40ms_modulation_mae_db": 0.8900386698573906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6549190165910603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16812101441877622, "1000–3000 Hz_10ms_envelope_correlation": 0.981196643601983, "1000–3000 Hz_10ms_modulation_mae_db": 0.41002204241677825, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0798714127874884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10848735378166555, "1000–3000 Hz_40ms_envelope_correlation": 0.9935649141043583, "1000–3000 Hz_40ms_modulation_mae_db": 0.361578323686953, "3000–6000 Hz_10ms_log_rms_mae_db": 2.054369027359723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3055274332694667, "3000–6000 Hz_10ms_envelope_correlation": 0.9353064102773682, "3000–6000 Hz_10ms_modulation_mae_db": 1.7929140370930243, "3000–6000 Hz_40ms_log_rms_mae_db": 1.645460144760536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25772448333982145, "3000–6000 Hz_40ms_envelope_correlation": 0.9513299394189457, "3000–6000 Hz_40ms_modulation_mae_db": 2.315087099600641, "mr_stft_db": 5.612722713442611, "band_mae_db": 0.2807478678316299, "rms_gain_error_db": -0.10969554880955579, "output_peak": 0.9684059671885347}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1412618246468627, "全帯域_10ms_rms_envelope_nrmse": 0.06590395228716847, "全帯域_10ms_envelope_correlation": 0.9957295874631087, "全帯域_10ms_modulation_mae_db": 0.033647392177803395, "全帯域_40ms_log_rms_mae_db": 0.6312201594018071, "全帯域_40ms_rms_envelope_nrmse": 0.043399371844428394, "全帯域_40ms_envelope_correlation": 0.9981573037120842, "全帯域_40ms_modulation_mae_db": 0.49067568446270443, "300–1000 Hz_10ms_log_rms_mae_db": 1.6116079914227803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0890777428929904, "300–1000 Hz_10ms_envelope_correlation": 0.9937449073923627, "300–1000 Hz_10ms_modulation_mae_db": 0.6325842410356697, "300–1000 Hz_40ms_log_rms_mae_db": 1.0376303927654207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05698892217971777, "300–1000 Hz_40ms_envelope_correlation": 0.9972744658015635, "300–1000 Hz_40ms_modulation_mae_db": 1.1008875802309042, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6628360998982537, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1851630160814953, "1000–3000 Hz_10ms_envelope_correlation": 0.9765622154656358, "1000–3000 Hz_10ms_modulation_mae_db": 0.6300900713826264, "1000–3000 Hz_40ms_log_rms_mae_db": 1.07585930241259, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11940051236784728, "1000–3000 Hz_40ms_envelope_correlation": 0.9915862344681361, "1000–3000 Hz_40ms_modulation_mae_db": 0.7478259213955011, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1047792328396335, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3076438707405118, "3000–6000 Hz_10ms_envelope_correlation": 0.9376740961464154, "3000–6000 Hz_10ms_modulation_mae_db": 1.8893663912483103, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6505422551689692, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2601990368570953, "3000–6000 Hz_40ms_envelope_correlation": 0.9529257712127934, "3000–6000 Hz_40ms_modulation_mae_db": 2.385475402572333, "mr_stft_db": 5.6223343697312815, "band_mae_db": 0.39391229893837415, "rms_gain_error_db": -0.12694392869278967, "output_peak": 1.0366963463995664}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 3.351206434316354e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.351206434316354e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0459685968300798, "全帯域_10ms_rms_envelope_nrmse": 0.058898170624205756, "全帯域_10ms_envelope_correlation": 0.9965383218030487, "全帯域_10ms_modulation_mae_db": 0.21579321229096723, "全帯域_40ms_log_rms_mae_db": 0.5377012237170635, "全帯域_40ms_rms_envelope_nrmse": 0.03337233452153531, "全帯域_40ms_envelope_correlation": 0.9988868496784795, "全帯域_40ms_modulation_mae_db": 0.22547321424211408, "300–1000 Hz_10ms_log_rms_mae_db": 1.4598652759186679, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09530907053880011, "300–1000 Hz_10ms_envelope_correlation": 0.992891187296973, "300–1000 Hz_10ms_modulation_mae_db": 1.0718183459935022, "300–1000 Hz_40ms_log_rms_mae_db": 0.9642623722194514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055260792506745496, "300–1000 Hz_40ms_envelope_correlation": 0.9975164876093118, "300–1000 Hz_40ms_modulation_mae_db": 1.4591253310900962, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6909987912302558, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15089110218052404, "1000–3000 Hz_10ms_envelope_correlation": 0.9841791590172124, "1000–3000 Hz_10ms_modulation_mae_db": 0.4759168851485121, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0818403543283965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08441854917072644, "1000–3000 Hz_40ms_envelope_correlation": 0.9956441040128126, "1000–3000 Hz_40ms_modulation_mae_db": 0.446132479789096, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0098878107207887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.313618971082037, "3000–6000 Hz_10ms_envelope_correlation": 0.9368456767564064, "3000–6000 Hz_10ms_modulation_mae_db": 2.6749789426549273, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5943286004231905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2598086810675398, "3000–6000 Hz_40ms_envelope_correlation": 0.9544689423137828, "3000–6000 Hz_40ms_modulation_mae_db": 3.148917566342052, "mr_stft_db": 5.64493718617286, "band_mae_db": 0.26108775128819883, "rms_gain_error_db": -0.08575529479557502, "output_peak": 1.016143182613958}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9916706437339422, "全帯域_10ms_rms_envelope_nrmse": 0.06588237346234198, "全帯域_10ms_envelope_correlation": 0.9957486078466361, "全帯域_10ms_modulation_mae_db": 0.23651424593052484, "全帯域_40ms_log_rms_mae_db": 0.4527011427808232, "全帯域_40ms_rms_envelope_nrmse": 0.03493872637523865, "全帯域_40ms_envelope_correlation": 0.9989188396000037, "全帯域_40ms_modulation_mae_db": 0.21630151279982882, "300–1000 Hz_10ms_log_rms_mae_db": 1.4554328345505418, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09063470018226931, "300–1000 Hz_10ms_envelope_correlation": 0.9937748068687624, "300–1000 Hz_10ms_modulation_mae_db": 0.5029504383021414, "300–1000 Hz_40ms_log_rms_mae_db": 0.9082803058135103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05608477246208165, "300–1000 Hz_40ms_envelope_correlation": 0.9977071274418335, "300–1000 Hz_40ms_modulation_mae_db": 0.6392091252947102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.769510441586968, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1621184582706335, "1000–3000 Hz_10ms_envelope_correlation": 0.9834505349653497, "1000–3000 Hz_10ms_modulation_mae_db": 0.5315789808055638, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1752248013151112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11290244623588712, "1000–3000 Hz_40ms_envelope_correlation": 0.9935939829571089, "1000–3000 Hz_40ms_modulation_mae_db": 0.8750643556675423, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9505195749580686, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2902517787469827, "3000–6000 Hz_10ms_envelope_correlation": 0.9440181407546825, "3000–6000 Hz_10ms_modulation_mae_db": 1.788904250689918, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5904814229810664, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24750903326243748, "3000–6000 Hz_40ms_envelope_correlation": 0.9570501456935537, "3000–6000 Hz_40ms_modulation_mae_db": 2.383808637347565, "mr_stft_db": 5.632457386545911, "band_mae_db": 0.3498115296022553, "rms_gain_error_db": -0.12896930838712595, "output_peak": 0.922255508574378}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59680, "output_samples": 59680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.789870235105237, "spectral_convergence": 0.19810514813399546, "spectral_convergence_status": "ok", "amplitude_floor": 3.189740694195829e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.479149724188453, "spectral_convergence": 0.1883938719278663, "spectral_convergence_status": "ok", "amplitude_floor": 2.205927306089595e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.659319079350599, "spectral_convergence": 0.184590828670329, "spectral_convergence_status": "ok", "amplitude_floor": 1.9576049396551346e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004284845572396228, "output_power": 0.004350248626474489, "delta_db": 0.06578904179695264, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.012674374150346739, "output_power": 0.01201999406636754, "delta_db": -0.23022270169591535, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0031087438744962816, "output_power": 0.0029962422455383224, "delta_db": -0.16008019478151034, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005542642121172655, "output_power": 0.0004893780638435625, "delta_db": -0.540723394602804, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001593466145796079, "output_power": 0.00014008589871055338, "delta_db": -0.5594842039428354, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.162629787041768e-05, "output_power": 0.00011004155307464569, "delta_db": 0.7953657179082813, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.39194420912138317, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16263577389065972, "pitch": null, "output_rms_dbfs": -16.96674527976855, "output_peak": 0.9692718217192645, "reference_rms_dbfs": -16.80410950587789, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.642779679548096}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00043785897255075944, "log_rms_mae_db": 1.0830353657937166, "rms_envelope_nrmse": 0.05999301911278997, "envelope_correlation": 0.9966228823348916, "rms_gain_error_db": -0.16263693626811349, "reference_mean_power": 0.0209012125553907, "output_mean_power": 0.020132967302944425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.005551294872497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00444581511703579, "output_power": 0.0042643516359459665, "reference_active": true, "delta_db": -0.1809838907236792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00039454721092167534, "output_power": 0.00039260050207930354, "reference_active": true, "delta_db": -0.021481270105334748}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.0001084016816849342, "output_power": 0.00012467499911411617, "reference_active": true, "delta_db": 0.6074335414067502}], "active_band_count": 3, "active_band_mae_db": 0.26996623407858805}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0004107294113879301, "log_rms_mae_db": 0.5982089490057763, "rms_envelope_nrmse": 0.03296440296824048, "envelope_correlation": 0.9991819606172591, "rms_gain_error_db": -0.16263658862096533, "reference_mean_power": 0.021070817398047652, "output_mean_power": 0.020296339772397913}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.012296407388435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00391054354874771, "output_power": 0.003769816075583962, "reference_active": true, "delta_db": -0.15916964597533362}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019349823972573515, "output_power": 0.00019043689215309727, "reference_active": true, "delta_db": -0.0692593318531658}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9805555260958754e-06, "output_power": 5.339688094248433e-06, "reference_active": true, "delta_db": -0.49225637473172906}], "active_band_count": 3, "active_band_mae_db": 0.2402284508534095}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.0003730035430310483, "log_rms_mae_db": 1.5810854897488154, "rms_envelope_nrmse": 0.08502188435995693, "envelope_correlation": 0.9944008451792308, "rms_gain_error_db": -0.07978490041624055, "reference_mean_power": 0.007423998211444209, "output_mean_power": 0.007288855970301717}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.725628531347034e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002177971361103236, "output_power": 0.0020599485726018247, "reference_active": true, "delta_db": -0.24195786591468485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00032785256002961345, "output_power": 0.00029321337348639623, "reference_active": true, "delta_db": -0.48494804566491445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.051887548388713e-05, "output_power": 6.905269142062506e-05, "reference_active": true, "delta_db": -0.6671709050843772}], "active_band_count": 3, "active_band_mae_db": 0.46469227222132553}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00035212990311193784, "log_rms_mae_db": 0.9814672788096308, "rms_envelope_nrmse": 0.05116186448855497, "envelope_correlation": 0.9979121261542526, "rms_gain_error_db": -0.07978586232328358, "reference_mean_power": 0.007484242812710004, "output_mean_power": 0.007348002285620139}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.215922155621447e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018646593945503163, "output_power": 0.0017784984122262894, "reference_active": true, "delta_db": -0.20546031782674393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015694821766981044, "output_power": 0.00014947040545089236, "reference_active": true, "delta_db": -0.212011755610575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505413793795639e-06, "output_power": 3.609214164700863e-06, "reference_active": true, "delta_db": -2.5586227307981755}], "active_band_count": 3, "active_band_mae_db": 0.9920316014118314}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00013449969207959865, "log_rms_mae_db": 1.7252752612853925, "rms_envelope_nrmse": 0.17637119347295235, "envelope_correlation": 0.979994835799342, "rms_gain_error_db": -0.6223369102263765, "reference_mean_power": 0.0006066736643199691, "output_mean_power": 0.0005256799969033394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.4525665594490304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019778914796660728, "output_power": 0.00018176045862527396, "reference_active": true, "delta_db": -0.3670304975800153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.140914614473558e-05, "output_power": 2.589754939381223e-05, "reference_active": true, "delta_db": -0.8379746023428565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0263986880205243e-05, "output_power": 1.052477068005507e-05, "reference_active": true, "delta_db": 0.10896553708575421}], "active_band_count": 3, "active_band_mae_db": 0.4379902123362087}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00012636239720129053, "log_rms_mae_db": 1.1528797250901184, "rms_envelope_nrmse": 0.11836933269113144, "envelope_correlation": 0.9929376842996878, "rms_gain_error_db": -0.6223542388593999, "reference_mean_power": 0.0006115927691385325, "output_mean_power": 0.0005299402645662885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.048532457224497e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001773914005108231, "output_power": 0.000163684255221368, "reference_active": true, "delta_db": -0.34925655840269904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4831454807134908e-05, "output_power": 1.2876749604312153e-05, "reference_active": true, "delta_db": -0.6137750208851457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.85797297398325e-07, "output_power": 6.329241580672113e-07, "reference_active": true, "delta_db": 1.1489657817059178}], "active_band_count": 3, "active_band_mae_db": 0.7039991203312542}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 5.5319997387638595e-05, "log_rms_mae_db": 2.026988224977522, "rms_envelope_nrmse": 0.30954476255931573, "envelope_correlation": 0.937283092538675, "rms_gain_error_db": 0.3304493875280772, "reference_mean_power": 9.493685130389184e-05, "output_mean_power": 0.00010244240424348991}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.144674032958852e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.902158206464317e-05, "output_power": 2.5858367753593036e-05, "reference_active": true, "delta_db": -0.5011997597939617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.982855070886748e-06, "output_power": 1.6225509850180725e-05, "reference_active": true, "delta_db": 2.1094358625727887}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.211971373682015e-06, "output_power": 6.456934806288111e-06, "reference_active": true, "delta_db": 3.0325473546306467}], "active_band_count": 3, "active_band_mae_db": 1.8810609923324657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 4.1914797815135623e-05, "log_rms_mae_db": 1.635976551612618, "rms_envelope_nrmse": 0.26172911886037703, "envelope_correlation": 0.9527306300212929, "rms_gain_error_db": 0.3304489728325785, "reference_mean_power": 9.570694886911743e-05, "output_mean_power": 0.00010327337460748624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.518176746918335e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6494345211181707e-05, "output_power": 2.3858355561921798e-05, "reference_active": true, "delta_db": -0.4551268410311707}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.515847029872695e-06, "output_power": 8.782695488103952e-06, "reference_active": true, "delta_db": 2.888886030604796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6527459125158204e-07, "output_power": 6.429663202734754e-07, "reference_active": true, "delta_db": 3.8449257037362483}], "active_band_count": 3, "active_band_mae_db": 2.396312858457405}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8893619745056927, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2fdef899ef400ce99839c586517f80d80e8546b6415551cad5e17ce0f7e3de4b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9822528152448559, "standard_deviation": 0.0034943201982570024, "minimum": 0.9765622154656358, "maximum": 0.9865826193846841, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7237483466906671, "standard_deviation": 0.052378408035750605, "minimum": 1.6549190165910603, "maximum": 1.807995490374177, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5894432078285835, "standard_deviation": 0.19261281511995704, "minimum": 0.41002204241677825, "maximum": 0.8879186399846759, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16332233531429954, "standard_deviation": 0.014244724026824761, "minimum": 0.1404300108443828, "maximum": 0.1851630160814953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9940932535204694, "standard_deviation": 0.0015064361646382229, "minimum": 0.9915862344681361, "maximum": 0.9959836387200272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1159979070445523, "standard_deviation": 0.04441833835710343, "minimum": 1.07585930241259, "maximum": 1.177250389085997, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6600864255098154, "standard_deviation": 0.1783397515602643, "minimum": 0.361578323686953, "maximum": 0.8750643556675423, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10358802396468525, "standard_deviation": 0.014352393697986307, "minimum": 0.08441854917072644, "maximum": 0.11940051236784728, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9389536501345188, "standard_deviation": 0.0028663249145629093, "minimum": 0.9353064102773682, "maximum": 0.9440181407546825, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.01065713289661, "standard_deviation": 0.05115325268759485, "minimum": 1.9505195749580686, "maximum": 2.1047792328396335, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.836963809885936, "standard_deviation": 0.37221529065420994, "minimum": 1.490139030184558, "maximum": 2.6749789426549273, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30242163829812807, "standard_deviation": 0.00797161518291671, "minimum": 0.2902517787469827, "maximum": 0.313618971082037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9549029094746779, "standard_deviation": 0.0023429256450622164, "minimum": 0.9513299394189457, "maximum": 0.957178069448276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.602846999034984, "standard_deviation": 0.046499375055204906, "minimum": 1.504495544105067, "maximum": 1.6505422551689692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.385080352206989, "standard_deviation": 0.3369681037961938, "minimum": 2.0575289614659638, "maximum": 3.148917566342052, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2531148919979082, "standard_deviation": 0.007415553677918191, "minimum": 0.2441617942462889, "maximum": 0.26172911886037703, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932836603307138, "standard_deviation": 0.0007424507004791067, "minimum": 0.9923175437384162, "maximum": 0.9944008451792308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.549102384467441, "standard_deviation": 0.06814856650850952, "minimum": 1.4554328345505418, "maximum": 1.6460118007255131, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7451159716232758, "standard_deviation": 0.30800871575330335, "minimum": 0.4337257929994229, "maximum": 1.2270083816635744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09284761295816557, "standard_deviation": 0.004853059682825436, "minimum": 0.08502188435995693, "maximum": 0.09899372830389318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976590092510119, "standard_deviation": 0.0002641138547203204, "minimum": 0.9972744658015635, "maximum": 0.9980519223297216, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9793054937817955, "standard_deviation": 0.054588693439407185, "minimum": 0.9082803058135103, "maximum": 1.0698624660283331, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0824985943748944, "standard_deviation": 0.2723191941615915, "minimum": 0.6392091252947102, "maximum": 1.4591253310900962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054297404502407276, "standard_deviation": 0.0032750096816071385, "minimum": 0.04870026335630673, "maximum": 0.05882523465880995, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3062194922457083, "standard_deviation": 0.07672921356505517, "minimum": 0.15942769026352402, "maximum": 0.39391229893837415, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6302588961584, "standard_deviation": 0.010643228560631823, "minimum": 5.612722713442611, "maximum": 5.64493718617286, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9745219298985456, "standard_deviation": 0.03692666986170067, "minimum": 0.922255508574378, "maximum": 1.0366963463995664, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12529457137702749, "standard_deviation": 0.03589775309005548, "minimum": -0.18937118379319207, "maximum": -0.08575529479557502, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9957712500863753, "standard_deviation": 0.0008030497159669844, "minimum": 0.9941336981099241, "maximum": 0.9966228823348916, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0784011456507572, "standard_deviation": 0.074962115534827, "minimum": 0.9902142046864346, "maximum": 1.190836013353681, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.25846584253145455, "standard_deviation": 0.15120431246336968, "minimum": 0.033647392177803395, "maximum": 0.5536168888010944, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06550043002951, "standard_deviation": 0.005475317997356603, "minimum": 0.058898170624205756, "maximum": 0.07652440319640105, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986852392928309, "standard_deviation": 0.0003572780237025806, "minimum": 0.9981573037120842, "maximum": 0.9991819606172591, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5734717185228848, "standard_deviation": 0.0795530078839424, "minimum": 0.4527011427808232, "maximum": 0.6674741974770456, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.27978459963858343, "standard_deviation": 0.11438591647429865, "minimum": 0.15583127979041686, "maximum": 0.49067568446270443, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037477235487521215, "standard_deviation": 0.003910068145870975, "minimum": 0.03296440296824048, "maximum": 0.043399371844428394, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0000", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.4048266410827637, "predicted_mos": 3.4048266410827637, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "048e47e447c3445e4758fcc2c00c994f13391ec45cc020e6c9df4e56e8604908", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.546553700231016, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.093110782609653, "全帯域_10ms_rms_envelope_nrmse": 0.05933918422036813, "全帯域_10ms_envelope_correlation": 0.9964987773775227, "全帯域_10ms_modulation_mae_db": 0.14007563353495298, "全帯域_40ms_log_rms_mae_db": 0.6234091215503451, "全帯域_40ms_rms_envelope_nrmse": 0.032213861194093224, "全帯域_40ms_envelope_correlation": 0.9989939281412175, "全帯域_40ms_modulation_mae_db": 0.2531415558305166, "300–1000 Hz_10ms_log_rms_mae_db": 1.4406332227173089, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07967229102292069, "300–1000 Hz_10ms_envelope_correlation": 0.9950735688116494, "300–1000 Hz_10ms_modulation_mae_db": 0.8603529856381403, "300–1000 Hz_40ms_log_rms_mae_db": 0.873985928488553, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048081488800795436, "300–1000 Hz_40ms_envelope_correlation": 0.9981551831947455, "300–1000 Hz_40ms_modulation_mae_db": 0.903269882974545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5954488812465357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14665892744325895, "1000–3000 Hz_10ms_envelope_correlation": 0.9858929947636621, "1000–3000 Hz_10ms_modulation_mae_db": 0.9703323232775093, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9816186365527166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09160510462244258, "1000–3000 Hz_40ms_envelope_correlation": 0.9955873262730232, "1000–3000 Hz_40ms_modulation_mae_db": 0.9698521479964154, "3000–6000 Hz_10ms_log_rms_mae_db": 1.924751957390154, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28833064653581375, "3000–6000 Hz_10ms_envelope_correlation": 0.944724141017692, "3000–6000 Hz_10ms_modulation_mae_db": 2.089198844651923, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5008932201816192, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2456296587128345, "3000–6000 Hz_40ms_envelope_correlation": 0.9577234548183283, "3000–6000 Hz_40ms_modulation_mae_db": 2.5155005209074535, "mr_stft_db": 5.570192010109959, "band_mae_db": 0.3207470896262557, "rms_gain_error_db": -0.09571939192019449, "output_peak": 1.0295954606518227}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.675603217158177e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.675603217158177e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1437498269866255, "全帯域_10ms_rms_envelope_nrmse": 0.07019311953433513, "全帯域_10ms_envelope_correlation": 0.995002159172482, "全帯域_10ms_modulation_mae_db": 0.4706828983112312, "全帯域_40ms_log_rms_mae_db": 0.6340614446452824, "全帯域_40ms_rms_envelope_nrmse": 0.04128469378047211, "全帯域_40ms_envelope_correlation": 0.9982064802032077, "全帯域_40ms_modulation_mae_db": 0.23140059178943892, "300–1000 Hz_10ms_log_rms_mae_db": 1.4747676873075348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08937970636920871, "300–1000 Hz_10ms_envelope_correlation": 0.9937436386842875, "300–1000 Hz_10ms_modulation_mae_db": 0.8807636275801439, "300–1000 Hz_40ms_log_rms_mae_db": 0.9243148591044529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048124951824219116, "300–1000 Hz_40ms_envelope_correlation": 0.9981155359420958, "300–1000 Hz_40ms_modulation_mae_db": 1.4219886267570272, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6656942028163628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16568615586098023, "1000–3000 Hz_10ms_envelope_correlation": 0.9813162353382926, "1000–3000 Hz_10ms_modulation_mae_db": 0.49823630961498494, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9863882544260865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09554038066097567, "1000–3000 Hz_40ms_envelope_correlation": 0.9949623983155242, "1000–3000 Hz_40ms_modulation_mae_db": 0.4683236973339518, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9101677811322897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29699427083001767, "3000–6000 Hz_10ms_envelope_correlation": 0.9406443983548747, "3000–6000 Hz_10ms_modulation_mae_db": 1.6762376920963689, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4789341013968147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24294006865812476, "3000–6000 Hz_40ms_envelope_correlation": 0.9580860422010374, "3000–6000 Hz_40ms_modulation_mae_db": 2.1716002942704526, "mr_stft_db": 5.592166556632442, "band_mae_db": 0.3175870423014913, "rms_gain_error_db": -0.04959303679882013, "output_peak": 0.9395647980689941}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0741157913247958, "全帯域_10ms_rms_envelope_nrmse": 0.07098785146810592, "全帯域_10ms_envelope_correlation": 0.9949584672510974, "全帯域_10ms_modulation_mae_db": 0.18773013451106443, "全帯域_40ms_log_rms_mae_db": 0.5758283686416721, "全帯域_40ms_rms_envelope_nrmse": 0.037529569195034085, "全帯域_40ms_envelope_correlation": 0.9986188710729922, "全帯域_40ms_modulation_mae_db": 0.20466799371469094, "300–1000 Hz_10ms_log_rms_mae_db": 1.513159440158798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08983342687818315, "300–1000 Hz_10ms_envelope_correlation": 0.9937615428129669, "300–1000 Hz_10ms_modulation_mae_db": 0.8009620584834343, "300–1000 Hz_40ms_log_rms_mae_db": 0.9810757188683513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05418404897306285, "300–1000 Hz_40ms_envelope_correlation": 0.9976821365993414, "300–1000 Hz_40ms_modulation_mae_db": 1.01699024981216, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7017394304747575, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17349361902103805, "1000–3000 Hz_10ms_envelope_correlation": 0.9817124069932295, "1000–3000 Hz_10ms_modulation_mae_db": 0.5984357495617074, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1266428855550124, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1197660723954232, "1000–3000 Hz_40ms_envelope_correlation": 0.9937678050500718, "1000–3000 Hz_40ms_modulation_mae_db": 0.8533702818222606, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9464244956682186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29704809010308303, "3000–6000 Hz_10ms_envelope_correlation": 0.9414779749047214, "3000–6000 Hz_10ms_modulation_mae_db": 1.6891280745496877, "3000–6000 Hz_40ms_log_rms_mae_db": 1.508744347802004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2508070133587583, "3000–6000 Hz_40ms_envelope_correlation": 0.9558734524433341, "3000–6000 Hz_40ms_modulation_mae_db": 2.205525913972316, "mr_stft_db": 5.582728814943194, "band_mae_db": 0.40041196944927715, "rms_gain_error_db": -0.10784811281875639, "output_peak": 0.9786715494452936}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0581328556839982, "全帯域_10ms_rms_envelope_nrmse": 0.06594816609164, "全帯域_10ms_envelope_correlation": 0.9956723559538267, "全帯域_10ms_modulation_mae_db": 0.2299689943599481, "全帯域_40ms_log_rms_mae_db": 0.5267957275894767, "全帯域_40ms_rms_envelope_nrmse": 0.03427669401974882, "全帯域_40ms_envelope_correlation": 0.9988760172650423, "全帯域_40ms_modulation_mae_db": 0.32174607136002736, "300–1000 Hz_10ms_log_rms_mae_db": 1.4854155255143873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09188942239379039, "300–1000 Hz_10ms_envelope_correlation": 0.9933906732631306, "300–1000 Hz_10ms_modulation_mae_db": 0.768582374630547, "300–1000 Hz_40ms_log_rms_mae_db": 0.9572485625064454, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050591274784588955, "300–1000 Hz_40ms_envelope_correlation": 0.9978968766220205, "300–1000 Hz_40ms_modulation_mae_db": 1.0412649194356687, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6266689045630935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15534410363683623, "1000–3000 Hz_10ms_envelope_correlation": 0.983450502890088, "1000–3000 Hz_10ms_modulation_mae_db": 0.3619970259879533, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9825211355349344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09291065283377087, "1000–3000 Hz_40ms_envelope_correlation": 0.994841150382605, "1000–3000 Hz_40ms_modulation_mae_db": 0.4101473492339489, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9286475776474628, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30347380627177684, "3000–6000 Hz_10ms_envelope_correlation": 0.9363007380641977, "3000–6000 Hz_10ms_modulation_mae_db": 1.8420752084944108, "3000–6000 Hz_40ms_log_rms_mae_db": 1.531903060592158, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2638229350380072, "3000–6000 Hz_40ms_envelope_correlation": 0.9489180553959631, "3000–6000 Hz_40ms_modulation_mae_db": 2.474989383788937, "mr_stft_db": 5.583906652528043, "band_mae_db": 0.34450773420298336, "rms_gain_error_db": -0.10800169388257017, "output_peak": 1.0572419392152375}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 3, "at_or_above_full_scale_fraction": 5.0268096514745307e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 5.0268096514745307e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.137902198152178, "全帯域_10ms_rms_envelope_nrmse": 0.062009968637351996, "全帯域_10ms_envelope_correlation": 0.9962456746119362, "全帯域_10ms_modulation_mae_db": 0.6034300886369129, "全帯域_40ms_log_rms_mae_db": 0.6254574637594692, "全帯域_40ms_rms_envelope_nrmse": 0.031395171418457185, "全帯域_40ms_envelope_correlation": 0.9991462699760979, "全帯域_40ms_modulation_mae_db": 0.9983050438060318, "300–1000 Hz_10ms_log_rms_mae_db": 1.5432372462556658, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08977116384196183, "300–1000 Hz_10ms_envelope_correlation": 0.9937252992981302, "300–1000 Hz_10ms_modulation_mae_db": 1.1535388435078267, "300–1000 Hz_40ms_log_rms_mae_db": 0.9454865104981842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04880101223217221, "300–1000 Hz_40ms_envelope_correlation": 0.998104169442585, "300–1000 Hz_40ms_modulation_mae_db": 1.7297489630596143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7091808308459284, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16204361922083657, "1000–3000 Hz_10ms_envelope_correlation": 0.984350383191543, "1000–3000 Hz_10ms_modulation_mae_db": 1.0076434684452626, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0671395057456539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10954273077230242, "1000–3000 Hz_40ms_envelope_correlation": 0.9951968957805232, "1000–3000 Hz_40ms_modulation_mae_db": 0.8113421439790759, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8511978002802054, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2899642180065023, "3000–6000 Hz_10ms_envelope_correlation": 0.9432344395580222, "3000–6000 Hz_10ms_modulation_mae_db": 1.734088782603349, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4239551657049958, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2431912618091213, "3000–6000 Hz_40ms_envelope_correlation": 0.9577538260900635, "3000–6000 Hz_40ms_modulation_mae_db": 2.2087221589102533, "mr_stft_db": 5.570960144851146, "band_mae_db": 0.30773175271676534, "rms_gain_error_db": -0.12708983137780902, "output_peak": 1.0271344859508313}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 3.351206434316354e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.351206434316354e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1338139452453737, "全帯域_10ms_rms_envelope_nrmse": 0.06891333920601167, "全帯域_10ms_envelope_correlation": 0.9951656189675157, "全帯域_10ms_modulation_mae_db": 0.45629140972773313, "全帯域_40ms_log_rms_mae_db": 0.6294205547961589, "全帯域_40ms_rms_envelope_nrmse": 0.03915757012647212, "全帯域_40ms_envelope_correlation": 0.9983466280167376, "全帯域_40ms_modulation_mae_db": 0.19651551647877843, "300–1000 Hz_10ms_log_rms_mae_db": 1.5541290068203195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10205425835071312, "300–1000 Hz_10ms_envelope_correlation": 0.9918944560695386, "300–1000 Hz_10ms_modulation_mae_db": 0.7834328380755871, "300–1000 Hz_40ms_log_rms_mae_db": 0.955066805580596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05778423516339956, "300–1000 Hz_40ms_envelope_correlation": 0.9972839913510863, "300–1000 Hz_40ms_modulation_mae_db": 0.9208157942823988, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7274951955807756, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1496884565603063, "1000–3000 Hz_10ms_envelope_correlation": 0.9844473310599119, "1000–3000 Hz_10ms_modulation_mae_db": 0.8060669899469994, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0659760297860048, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0934821641823289, "1000–3000 Hz_40ms_envelope_correlation": 0.9943250003224966, "1000–3000 Hz_40ms_modulation_mae_db": 0.5990531170178991, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9174921349937428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3038271913674113, "3000–6000 Hz_10ms_envelope_correlation": 0.9353856869685917, "3000–6000 Hz_10ms_modulation_mae_db": 1.6960007812005626, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4982725554714922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2520166690367916, "3000–6000 Hz_40ms_envelope_correlation": 0.9530752143435226, "3000–6000 Hz_40ms_modulation_mae_db": 2.2527185445911453, "mr_stft_db": 5.591576875147267, "band_mae_db": 0.22644993136994665, "rms_gain_error_db": -0.03887561898398789, "output_peak": 1.0973982329122485}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 3.351206434316354e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.351206434316354e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1506220638900557, "全帯域_10ms_rms_envelope_nrmse": 0.07216034338038066, "全帯域_10ms_envelope_correlation": 0.9947286836440852, "全帯域_10ms_modulation_mae_db": 0.45597457164789273, "全帯域_40ms_log_rms_mae_db": 0.5838165296134208, "全帯域_40ms_rms_envelope_nrmse": 0.0383847459807051, "全帯域_40ms_envelope_correlation": 0.9984638721941369, "全帯域_40ms_modulation_mae_db": 0.43574253209376596, "300–1000 Hz_10ms_log_rms_mae_db": 1.5213286719074182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08960241182059123, "300–1000 Hz_10ms_envelope_correlation": 0.9936992794802363, "300–1000 Hz_10ms_modulation_mae_db": 0.6747514761572155, "300–1000 Hz_40ms_log_rms_mae_db": 0.9397851430347053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046803709239823466, "300–1000 Hz_40ms_envelope_correlation": 0.9982183916606778, "300–1000 Hz_40ms_modulation_mae_db": 0.9809758448332692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.676084811295925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15674001245720023, "1000–3000 Hz_10ms_envelope_correlation": 0.9853144470598516, "1000–3000 Hz_10ms_modulation_mae_db": 0.8020262465132176, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0142805009898126, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10331160936565931, "1000–3000 Hz_40ms_envelope_correlation": 0.9959507868041785, "1000–3000 Hz_40ms_modulation_mae_db": 0.8647809619251712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8994151642595347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2926712103265454, "3000–6000 Hz_10ms_envelope_correlation": 0.940655088399302, "3000–6000 Hz_10ms_modulation_mae_db": 1.6584501188562422, "3000–6000 Hz_40ms_log_rms_mae_db": 1.490560506185801, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24374420063414903, "3000–6000 Hz_40ms_envelope_correlation": 0.9565488999783385, "3000–6000 Hz_40ms_modulation_mae_db": 2.2493890191580594, "mr_stft_db": 5.57482150217306, "band_mae_db": 0.23770510139189108, "rms_gain_error_db": -0.07430935819602831, "output_peak": 1.040667358176731}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 3, "at_or_above_full_scale_fraction": 5.0268096514745307e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 5.0268096514745307e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1468552266238101, "全帯域_10ms_rms_envelope_nrmse": 0.07329504555798777, "全帯域_10ms_envelope_correlation": 0.9948085604271599, "全帯域_10ms_modulation_mae_db": 0.12828656613477904, "全帯域_40ms_log_rms_mae_db": 0.6148540263266713, "全帯域_40ms_rms_envelope_nrmse": 0.041456599473557364, "全帯域_40ms_envelope_correlation": 0.9985081155060535, "全帯域_40ms_modulation_mae_db": 0.44069986290495605, "300–1000 Hz_10ms_log_rms_mae_db": 1.493925163344916, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09561930472174511, "300–1000 Hz_10ms_envelope_correlation": 0.992812172489408, "300–1000 Hz_10ms_modulation_mae_db": 0.2826835698010641, "300–1000 Hz_40ms_log_rms_mae_db": 0.9183437305046386, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05243102530539379, "300–1000 Hz_40ms_envelope_correlation": 0.9977832456677519, "300–1000 Hz_40ms_modulation_mae_db": 0.4134883564312137, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7034764891566578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17010438830082525, "1000–3000 Hz_10ms_envelope_correlation": 0.9818072420958508, "1000–3000 Hz_10ms_modulation_mae_db": 0.7403680055857716, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0345458225459134, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11911028123160869, "1000–3000 Hz_40ms_envelope_correlation": 0.9928939664451347, "1000–3000 Hz_40ms_modulation_mae_db": 0.4531343458950757, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9116903868979516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29097929179113896, "3000–6000 Hz_10ms_envelope_correlation": 0.9421291468573744, "3000–6000 Hz_10ms_modulation_mae_db": 1.9581776344391022, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4809281142527047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2497478222505474, "3000–6000 Hz_40ms_envelope_correlation": 0.954971637000169, "3000–6000 Hz_40ms_modulation_mae_db": 2.541781551944935, "mr_stft_db": 5.598472747609488, "band_mae_db": 0.32515897181104503, "rms_gain_error_db": -0.16946424191454823, "output_peak": 0.9862483731158352}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59680, "output_samples": 59680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.71538960040214, "spectral_convergence": 0.20792840396107495, "spectral_convergence_status": "ok", "amplitude_floor": 3.189740694195829e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.413440230838857, "spectral_convergence": 0.1822905896796328, "spectral_convergence_status": "ok", "amplitude_floor": 2.205927306089595e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.581746199088878, "spectral_convergence": 0.1844591079268966, "spectral_convergence_status": "ok", "amplitude_floor": 1.9576049396551346e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004284845572396228, "output_power": 0.004357957271526784, "delta_db": 0.07347793414248417, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.012674374150346739, "output_power": 0.012381573071716407, "delta_db": -0.10150698395361231, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0031087438744962816, "output_power": 0.002922237025405292, "delta_db": -0.26869503477530277, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005542642121172655, "output_power": 0.0004984461830953161, "delta_db": -0.46098563162311557, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001593466145796079, "output_power": 0.000149356778620244, "delta_db": -0.2811790301943198, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.162629787041768e-05, "output_power": 0.00010861353074229458, "delta_db": 0.7386379230686995, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3207470896262557, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09571939192019449, "pitch": null, "output_rms_dbfs": -16.899828897798084, "output_peak": 1.0295954606518227, "reference_rms_dbfs": -16.80410950587789, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.570192010109959}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00043785897255075944, "log_rms_mae_db": 1.093110782609653, "rms_envelope_nrmse": 0.05933918422036813, "envelope_correlation": 0.9964987773775227, "rms_gain_error_db": -0.0957194059846067, "reference_mean_power": 0.0209012125553907, "output_mean_power": 0.02044558499589032}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.005551294872497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00444581511703579, "output_power": 0.004222453908953763, "reference_active": true, "delta_db": -0.22386480596849062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00039454721092167534, "output_power": 0.00038876732219940283, "reference_active": true, "delta_db": -0.06409224468120313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.0001084016816849342, "output_power": 0.00010514994132202874, "reference_active": true, "delta_db": -0.13226984995516522}], "active_band_count": 3, "active_band_mae_db": 0.14007563353495298}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0004107294113879301, "log_rms_mae_db": 0.6234091215503451, "rms_envelope_nrmse": 0.032213861194093224, "envelope_correlation": 0.9989939281412175, "rms_gain_error_db": -0.0957168686890956, "reference_mean_power": 0.021070817398047652, "output_mean_power": 0.020611504648069726}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.012296407388435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00391054354874771, "output_power": 0.003734532579218436, "reference_active": true, "delta_db": -0.20000874179894987}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019349823972573515, "output_power": 0.00019389686617899966, "reference_active": true, "delta_db": 0.008937713845962952}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9805555260958754e-06, "output_power": 5.26858153840917e-06, "reference_active": true, "delta_db": -0.550478211846637}], "active_band_count": 3, "active_band_mae_db": 0.2531415558305166}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.0003730035430310483, "log_rms_mae_db": 1.4406332227173089, "rms_envelope_nrmse": 0.07967229102292069, "envelope_correlation": 0.9950735688116494, "rms_gain_error_db": -0.09816880253538499, "reference_mean_power": 0.007423998211444209, "output_mean_power": 0.007258067090693512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.725628531347034e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002177971361103236, "output_power": 0.0020651890167884307, "reference_active": true, "delta_db": -0.23092358078864644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00032785256002961345, "output_power": 0.00029632517748530706, "reference_active": true, "delta_db": -0.43910025998958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.051887548388713e-05, "output_power": 5.185542396289857e-05, "reference_active": true, "delta_db": -1.911035116136195}], "active_band_count": 3, "active_band_mae_db": 0.8603529856381403}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00035212990311193784, "log_rms_mae_db": 0.873985928488553, "rms_envelope_nrmse": 0.048081488800795436, "envelope_correlation": 0.9981551831947455, "rms_gain_error_db": -0.09817314063849511, "reference_mean_power": 0.007484242812710004, "output_mean_power": 0.007316957877684211}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.215922155621447e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018646593945503163, "output_power": 0.001778264175703042, "reference_active": true, "delta_db": -0.20603234155415728}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015694821766981044, "output_power": 0.0001524407050715157, "reference_active": true, "delta_db": -0.12655439331853122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505413793795639e-06, "output_power": 3.7631595753558523e-06, "reference_active": true, "delta_db": -2.3772229140509467}], "active_band_count": 3, "active_band_mae_db": 0.903269882974545}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00013449969207959865, "log_rms_mae_db": 1.5954488812465357, "rms_envelope_nrmse": 0.14665892744325895, "envelope_correlation": 0.9858929947636621, "rms_gain_error_db": -0.4673566567577194, "reference_mean_power": 0.0006066736643199691, "output_mean_power": 0.0005447778946426281}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.4525665594490304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019778914796660728, "output_power": 0.0001682072977400706, "reference_active": true, "delta_db": -0.7035762571122612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.140914614473558e-05, "output_power": 2.5198057751161348e-05, "reference_active": true, "delta_db": -0.9568906334257082}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0263986880205243e-05, "output_power": 7.695964935738504e-06, "reference_active": true, "delta_db": -1.2505300792945582}], "active_band_count": 3, "active_band_mae_db": 0.9703323232775093}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00012636239720129053, "log_rms_mae_db": 0.9816186365527166, "rms_envelope_nrmse": 0.09160510462244258, "envelope_correlation": 0.9955873262730232, "rms_gain_error_db": -0.46736360264364796, "reference_mean_power": 0.0006115927691385325, "output_mean_power": 0.0005491942503373283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.048532457224497e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001773914005108231, "output_power": 0.0001502205912916292, "reference_active": true, "delta_db": -0.7220309540299987}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4831454807134908e-05, "output_power": 1.1880056914940803e-05, "reference_active": true, "delta_db": -0.9636523149604014}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.85797297398325e-07, "output_power": 3.6649476611719016e-07, "reference_active": true, "delta_db": -1.2238731749988458}], "active_band_count": 3, "active_band_mae_db": 0.9698521479964154}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 5.5319997387638595e-05, "log_rms_mae_db": 1.924751957390154, "rms_envelope_nrmse": 0.28833064653581375, "envelope_correlation": 0.944724141017692, "rms_gain_error_db": 0.25689472858304274, "reference_mean_power": 9.493685130389184e-05, "output_mean_power": 0.00010072198974038946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.144674032958852e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.902158206464317e-05, "output_power": 2.5548251752512883e-05, "reference_active": true, "delta_db": -0.5535989656329764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.982855070886748e-06, "output_power": 1.7341210169046306e-05, "reference_active": true, "delta_db": 2.3982463559853167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.211971373682015e-06, "output_power": 6.892024097481831e-06, "reference_active": true, "delta_db": 3.315751212337477}], "active_band_count": 3, "active_band_mae_db": 2.089198844651923}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 4.1914797815135623e-05, "log_rms_mae_db": 1.5008932201816192, "rms_envelope_nrmse": 0.2456296587128345, "envelope_correlation": 0.9577234548183283, "rms_gain_error_db": 0.25689187363028565, "reference_mean_power": 9.570694886911743e-05, "output_mean_power": 0.00010153894776058853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.518176746918335e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6494345211181707e-05, "output_power": 2.3894312830415044e-05, "reference_active": true, "delta_db": -0.4485864541719367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.515847029872695e-06, "output_power": 8.968581109006512e-06, "reference_active": true, "delta_db": 2.979845179357577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6527459125158204e-07, "output_power": 6.847036562189097e-07, "reference_active": true, "delta_db": 4.118069929192847}], "active_band_count": 3, "active_band_mae_db": 2.5155005209074535}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8948018272107698, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b2532a0eafa44c2439ece61724470c3a66a5aef75b8df5b5492c970b276a25ea", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9835364429240536, "standard_deviation": 0.001750950812284103, "minimum": 0.9813162353382926, "maximum": 0.9858929947636621, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6757235932475045, "standard_deviation": 0.04503624507551463, "minimum": 1.5954488812465357, "maximum": 1.7274951955807756, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7231382648666758, "standard_deviation": 0.22424478514470042, "minimum": 0.3619970259879533, "maximum": 1.0076434684452626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15996991031266022, "standard_deviation": 0.009541600321721145, "minimum": 0.14665892744325895, "maximum": 0.17349361902103805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946906661716947, "standard_deviation": 0.0009970076435297608, "minimum": 0.9928939664451347, "maximum": 0.9959507868041785, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0323890963920168, "standard_deviation": 0.051730832317315616, "minimum": 0.9816186365527166, "maximum": 1.1266428855550124, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6787505056504748, "standard_deviation": 0.22074598466370726, "minimum": 0.4101473492339489, "maximum": 0.9698521479964154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10315862450806396, "standard_deviation": 0.01171823465282506, "minimum": 0.09160510462244258, "maximum": 0.1197660723954232, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9405689517655971, "standard_deviation": 0.0032233068323506858, "minimum": 0.9353856869685917, "maximum": 0.944724141017692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9112234122836949, "standard_deviation": 0.028053208289638275, "minimum": 1.8511978002802054, "maximum": 1.9464244956682186, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.792919642111456, "standard_deviation": 0.15713124398961764, "minimum": 1.6584501188562422, "maximum": 2.089198844651923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29541109065403615, "standard_deviation": 0.005951869537451531, "minimum": 0.28833064653581375, "maximum": 0.3038271913674113, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9553688227838445, "standard_deviation": 0.0031005493417570187, "minimum": 0.9489180553959631, "maximum": 0.9580860422010374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4892738839484487, "standard_deviation": 0.031298297654416066, "minimum": 1.4239551657049958, "maximum": 1.531903060592158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.3275284234429443, "standard_deviation": 0.15491843663989918, "minimum": 2.1716002942704526, "maximum": 2.541781551944935, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.24898745368729175, "standard_deviation": 0.006990706135806609, "minimum": 0.24294006865812476, "maximum": 0.2638229350380072, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935125788636685, "standard_deviation": 0.0009067933967988948, "minimum": 0.9918944560695386, "maximum": 0.9950735688116494, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5033244955032936, "standard_deviation": 0.03732055767477536, "minimum": 1.4406332227173089, "maximum": 1.5541290068203195, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7756334717342449, "standard_deviation": 0.24347809639225224, "minimum": 0.2826835698010641, "maximum": 1.1535388435078267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09097774817488927, "standard_deviation": 0.006325551820444645, "minimum": 0.07967229102292069, "maximum": 0.10205425835071312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997904941310038, "standard_deviation": 0.00031509856285725135, "minimum": 0.9972839913510863, "maximum": 0.9982183916606778, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9369134073232408, "standard_deviation": 0.032157076618476145, "minimum": 0.873985928488553, "maximum": 0.9810757188683513, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0535678296982371, "standard_deviation": 0.38722972027160785, "minimum": 0.4134883564312137, "maximum": 1.7297489630596143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05085021829043192, "standard_deviation": 0.0037307111699922276, "minimum": 0.046803709239823466, "maximum": 0.05778423516339956, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.31003744910870695, "standard_deviation": 0.05603604142654663, "minimum": 0.22644993136994665, "maximum": 0.40041196944927715, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5831031629993255, "standard_deviation": 0.010513225883991301, "minimum": 5.570192010109959, "maximum": 5.598472747609488, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 1.0195652746921242, "standard_deviation": 0.0496603965620128, "minimum": 0.9395647980689941, "maximum": 1.0973982329122485, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09636266073658933, "standard_deviation": 0.042315654868919715, "minimum": -0.16946424191454823, "maximum": -0.03887561898398789, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9953850371757033, "standard_deviation": 0.0006764923534726407, "minimum": 0.9947286836440852, "maximum": 0.9964987773775227, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1172878363145613, "standard_deviation": 0.03651099553813234, "minimum": 1.0581328556839982, "maximum": 1.1506220638900557, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33405503710806433, "standard_deviation": 0.18254148202101217, "minimum": 0.12828656613477904, "maximum": 0.6034300886369129, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06785587726202266, "standard_deviation": 0.004997874403460114, "minimum": 0.05933918422036813, "maximum": 0.07329504555798777, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986450227969357, "standard_deviation": 0.0003294896983869286, "minimum": 0.9982064802032077, "maximum": 0.9991462699760979, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6017054046153121, "standard_deviation": 0.037079515567395846, "minimum": 0.5267957275894767, "maximum": 0.6340614446452824, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38527739599727573, "standard_deviation": 0.2658802793067077, "minimum": 0.19651551647877843, "maximum": 0.9983050438060318, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.036962363148567504, "standard_deviation": 0.003906025469404433, "minimum": 0.031395171418457185, "maximum": 0.041456599473557364, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2830-3980-0038", "language": "en", "style": "normal", "speaker": "librispeech_2830", "source_utterance_id": "2830-3980-0038", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 88640, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2830-3980-0038.wav", "source_file_sha256": "457a1c367660ef48f57da0ebb3350ffdc77c5e023e485fb3252f9d969390a686", "decoded_pcm_sha256": "77d64124560df236e40a3ec37cc728b327f805fee264fb8362c33791320222e1", "original_source_sha256": "2c8978e75dd10716d066062bb2fbfd1a002ac3190e42b0f251f15d0aa2d6d272", "corpus_entry_id": "librispeech_2830-3980-0038"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.049129962921143, "predicted_mos": 4.049129962921143, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "9fae48abb3c4bca97d7bc79471e9a2e3bfa5c9f304abfc7de9d09035f1ab13a3", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.196500830999867, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2830-3980-0038", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9fae48abb3c4bca97d7bc79471e9a2e3bfa5c9f304abfc7de9d09035f1ab13a3", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "34283f63866e01407d3dd8060fcc0517a3925172939da95984cc41b8a93e3ecd"}, "rows": [{"case": "librispeech_2830-3980-0038", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.178510904312134, "predicted_mos": 3.178510904312134, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "e6ef2b8396a9a0be6dd520ae463c3f33957bc2effa867cbdebc4a102fbec07c1", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.048176927000895, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2830-3980-0038", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "e6ef2b8396a9a0be6dd520ae463c3f33957bc2effa867cbdebc4a102fbec07c1", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6389901266967921, "全帯域_10ms_rms_envelope_nrmse": 0.09665249429537519, "全帯域_10ms_envelope_correlation": 0.9896582381255375, "全帯域_10ms_modulation_mae_db": 0.5696363322238032, "全帯域_40ms_log_rms_mae_db": 0.7383768326852473, "全帯域_40ms_rms_envelope_nrmse": 0.046340090763661586, "全帯域_40ms_envelope_correlation": 0.9975513137232447, "全帯域_40ms_modulation_mae_db": 0.43136777167878754, "300–1000 Hz_10ms_log_rms_mae_db": 3.001359074508102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09804481178877072, "300–1000 Hz_10ms_envelope_correlation": 0.9940167904639269, "300–1000 Hz_10ms_modulation_mae_db": 0.6353646630345141, "300–1000 Hz_40ms_log_rms_mae_db": 2.243186359652414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06503980980418535, "300–1000 Hz_40ms_envelope_correlation": 0.9974063506371075, "300–1000 Hz_40ms_modulation_mae_db": 0.8671889115276805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.848728890482448, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14906141255912367, "1000–3000 Hz_10ms_envelope_correlation": 0.9863095716090039, "1000–3000 Hz_10ms_modulation_mae_db": 0.43353440348374433, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1225613551177707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10393580946907664, "1000–3000 Hz_40ms_envelope_correlation": 0.9940249769529311, "1000–3000 Hz_40ms_modulation_mae_db": 0.49099846560933164, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8142491244614052, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1414704173671724, "3000–6000 Hz_10ms_envelope_correlation": 0.9861544930182277, "3000–6000 Hz_10ms_modulation_mae_db": 0.5587484491349536, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1234278829119078, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07606161332467117, "3000–6000 Hz_40ms_envelope_correlation": 0.9959253100176699, "3000–6000 Hz_40ms_modulation_mae_db": 0.8338255484940674, "mr_stft_db": 5.8862495177735665, "band_mae_db": 0.1679394622000823, "rms_gain_error_db": -0.10426257820153229, "output_peak": 0.5692346967899763}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.548956403296094, "全帯域_10ms_rms_envelope_nrmse": 0.09224915550094846, "全帯域_10ms_envelope_correlation": 0.9905451033123416, "全帯域_10ms_modulation_mae_db": 0.21038422562119155, "全帯域_40ms_log_rms_mae_db": 0.6860652379205029, "全帯域_40ms_rms_envelope_nrmse": 0.05069868748318049, "全帯域_40ms_envelope_correlation": 0.9970738658900664, "全帯域_40ms_modulation_mae_db": 0.24521520988211062, "300–1000 Hz_10ms_log_rms_mae_db": 2.8698903496830477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1158576656026754, "300–1000 Hz_10ms_envelope_correlation": 0.9914224820882026, "300–1000 Hz_10ms_modulation_mae_db": 0.32425831468350114, "300–1000 Hz_40ms_log_rms_mae_db": 2.14045470339281, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0890474751087725, "300–1000 Hz_40ms_envelope_correlation": 0.9948285451994, "300–1000 Hz_40ms_modulation_mae_db": 0.39495151865963146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.898215841695086, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15058260700421694, "1000–3000 Hz_10ms_envelope_correlation": 0.9859730542458485, "1000–3000 Hz_10ms_modulation_mae_db": 0.6590982259078203, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1603914038409882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0927321164678738, "1000–3000 Hz_40ms_envelope_correlation": 0.9956822542572241, "1000–3000 Hz_40ms_modulation_mae_db": 0.8521852228976131, "3000–6000 Hz_10ms_log_rms_mae_db": 1.772465005190347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1637587328644197, "3000–6000 Hz_10ms_envelope_correlation": 0.9818522710980997, "3000–6000 Hz_10ms_modulation_mae_db": 0.777644282443298, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1471847810416067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09160711959028542, "3000–6000 Hz_40ms_envelope_correlation": 0.9941493640325725, "3000–6000 Hz_40ms_modulation_mae_db": 1.273125332158748, "mr_stft_db": 5.890645480259541, "band_mae_db": 0.17778522045953357, "rms_gain_error_db": -0.11799528957746892, "output_peak": 0.6033292236547295}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6151900743561658, "全帯域_10ms_rms_envelope_nrmse": 0.09769736952383355, "全帯域_10ms_envelope_correlation": 0.9893818500362526, "全帯域_10ms_modulation_mae_db": 0.49847764577321424, "全帯域_40ms_log_rms_mae_db": 0.7439514473846051, "全帯域_40ms_rms_envelope_nrmse": 0.04972054148127892, "全帯域_40ms_envelope_correlation": 0.9971521440148752, "全帯域_40ms_modulation_mae_db": 0.5820373813114516, "300–1000 Hz_10ms_log_rms_mae_db": 2.8219929071985783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10587575096347854, "300–1000 Hz_10ms_envelope_correlation": 0.9928838190606241, "300–1000 Hz_10ms_modulation_mae_db": 0.3817196830513831, "300–1000 Hz_40ms_log_rms_mae_db": 2.047789054769293, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06940171384144975, "300–1000 Hz_40ms_envelope_correlation": 0.996980758120759, "300–1000 Hz_40ms_modulation_mae_db": 0.36412193536223253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9219650226591343, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1515523066690611, "1000–3000 Hz_10ms_envelope_correlation": 0.984785626297065, "1000–3000 Hz_10ms_modulation_mae_db": 0.3834021987433611, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1763828608533722, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0822586107596479, "1000–3000 Hz_40ms_envelope_correlation": 0.9956645863320331, "1000–3000 Hz_40ms_modulation_mae_db": 0.38064690315141053, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7896528262025928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1541757338666399, "3000–6000 Hz_10ms_envelope_correlation": 0.9834935032009308, "3000–6000 Hz_10ms_modulation_mae_db": 0.8657831104123709, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1488867209657077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08285287954792649, "3000–6000 Hz_40ms_envelope_correlation": 0.9950714748884963, "3000–6000 Hz_40ms_modulation_mae_db": 1.083006542954826, "mr_stft_db": 5.893256024850891, "band_mae_db": 0.19626797228185766, "rms_gain_error_db": -0.10666472116141601, "output_peak": 0.6037706624483472}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5775295978754618, "全帯域_10ms_rms_envelope_nrmse": 0.0931403702647315, "全帯域_10ms_envelope_correlation": 0.990434942826807, "全帯域_10ms_modulation_mae_db": 0.47025472262815526, "全帯域_40ms_log_rms_mae_db": 0.7323399729929341, "全帯域_40ms_rms_envelope_nrmse": 0.04427139990914492, "全帯域_40ms_envelope_correlation": 0.9977479422126562, "全帯域_40ms_modulation_mae_db": 0.42362989497015485, "300–1000 Hz_10ms_log_rms_mae_db": 2.8857584369801454, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10042824251280022, "300–1000 Hz_10ms_envelope_correlation": 0.9936845284253477, "300–1000 Hz_10ms_modulation_mae_db": 0.20311522142113014, "300–1000 Hz_40ms_log_rms_mae_db": 2.2222465949824084, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06536306395618005, "300–1000 Hz_40ms_envelope_correlation": 0.99737845640453, "300–1000 Hz_40ms_modulation_mae_db": 0.10601971144496597, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8240776064950068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13966410055967626, "1000–3000 Hz_10ms_envelope_correlation": 0.9870111364539765, "1000–3000 Hz_10ms_modulation_mae_db": 0.4215676586195747, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0964304601574044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07781915031620333, "1000–3000 Hz_40ms_envelope_correlation": 0.9958931017560572, "1000–3000 Hz_40ms_modulation_mae_db": 0.6055070144671711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.752121639598867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16882935126084941, "3000–6000 Hz_10ms_envelope_correlation": 0.9802183137692537, "3000–6000 Hz_10ms_modulation_mae_db": 0.4428272366892783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1450297283659925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.099415189547284, "3000–6000 Hz_40ms_envelope_correlation": 0.9928453652974082, "3000–6000 Hz_40ms_modulation_mae_db": 1.1763210763195784, "mr_stft_db": 5.8799160579041825, "band_mae_db": 0.060270647379312564, "rms_gain_error_db": -0.07947694953251304, "output_peak": 0.5847405495640255}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5588156874891632, "全帯域_10ms_rms_envelope_nrmse": 0.09591218639180189, "全帯域_10ms_envelope_correlation": 0.9897423793353707, "全帯域_10ms_modulation_mae_db": 0.4461098748789089, "全帯域_40ms_log_rms_mae_db": 0.712900494461468, "全帯域_40ms_rms_envelope_nrmse": 0.048773796902992765, "全帯域_40ms_envelope_correlation": 0.9972063357576982, "全帯域_40ms_modulation_mae_db": 0.2718157966692143, "300–1000 Hz_10ms_log_rms_mae_db": 2.848941199033011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10698389846181965, "300–1000 Hz_10ms_envelope_correlation": 0.9928349418201045, "300–1000 Hz_10ms_modulation_mae_db": 1.360287507775795, "300–1000 Hz_40ms_log_rms_mae_db": 2.2089123849341883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07289128853529432, "300–1000 Hz_40ms_envelope_correlation": 0.9967262843240082, "300–1000 Hz_40ms_modulation_mae_db": 1.5322038074945945, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8956363315716176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13864247277792105, "1000–3000 Hz_10ms_envelope_correlation": 0.9873931221982548, "1000–3000 Hz_10ms_modulation_mae_db": 0.5601791836691808, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1359192682238015, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07810126199244642, "1000–3000 Hz_40ms_envelope_correlation": 0.9961827029437348, "1000–3000 Hz_40ms_modulation_mae_db": 0.36316586216607316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7402967960389983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16160535168493403, "3000–6000 Hz_10ms_envelope_correlation": 0.9819781942354139, "3000–6000 Hz_10ms_modulation_mae_db": 0.9993431765679178, "3000–6000 Hz_40ms_log_rms_mae_db": 1.100766216020107, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0762775848489116, "3000–6000 Hz_40ms_envelope_correlation": 0.9962072859920424, "3000–6000 Hz_40ms_modulation_mae_db": 1.2811907404401304, "mr_stft_db": 5.877978592812647, "band_mae_db": 0.21061063331372054, "rms_gain_error_db": -0.08453738447470037, "output_peak": 0.5781651972041014}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5311806925310347, "全帯域_10ms_rms_envelope_nrmse": 0.09137910010542424, "全帯域_10ms_envelope_correlation": 0.9907196040138503, "全帯域_10ms_modulation_mae_db": 0.4697033340664956, "全帯域_40ms_log_rms_mae_db": 0.6622320985110404, "全帯域_40ms_rms_envelope_nrmse": 0.04796793977909018, "全帯域_40ms_envelope_correlation": 0.9974013227359472, "全帯域_40ms_modulation_mae_db": 0.8050240388133606, "300–1000 Hz_10ms_log_rms_mae_db": 2.806322758914066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1074923032758313, "300–1000 Hz_10ms_envelope_correlation": 0.9926924946028852, "300–1000 Hz_10ms_modulation_mae_db": 1.1105471641030122, "300–1000 Hz_40ms_log_rms_mae_db": 2.053453141137599, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07065677059261748, "300–1000 Hz_40ms_envelope_correlation": 0.9968635737348507, "300–1000 Hz_40ms_modulation_mae_db": 1.3032162504877844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8809601485121183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15834871623130184, "1000–3000 Hz_10ms_envelope_correlation": 0.9845475913589262, "1000–3000 Hz_10ms_modulation_mae_db": 1.0852053137599562, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1228606848488394, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09633966620520117, "1000–3000 Hz_40ms_envelope_correlation": 0.995490079667691, "1000–3000 Hz_40ms_modulation_mae_db": 1.2051823588593202, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8003187479385103, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17217617198096533, "3000–6000 Hz_10ms_envelope_correlation": 0.9799476442001803, "3000–6000 Hz_10ms_modulation_mae_db": 0.6695828473235127, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1317578690721057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0856065244966607, "3000–6000 Hz_40ms_envelope_correlation": 0.994914471210862, "3000–6000 Hz_40ms_modulation_mae_db": 0.8121938943567453, "mr_stft_db": 5.872836243108004, "band_mae_db": 0.2264955634865834, "rms_gain_error_db": -0.11676867755604615, "output_peak": 0.5932676666270951}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.582586565601576, "全帯域_10ms_rms_envelope_nrmse": 0.09128009394083045, "全帯域_10ms_envelope_correlation": 0.9908758409453456, "全帯域_10ms_modulation_mae_db": 0.5761423414809926, "全帯域_40ms_log_rms_mae_db": 0.7122208701796087, "全帯域_40ms_rms_envelope_nrmse": 0.050081174500671705, "全帯域_40ms_envelope_correlation": 0.9970899468773047, "全帯域_40ms_modulation_mae_db": 0.461312478169023, "300–1000 Hz_10ms_log_rms_mae_db": 2.923501838836301, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12087376611112037, "300–1000 Hz_10ms_envelope_correlation": 0.9909180551982264, "300–1000 Hz_10ms_modulation_mae_db": 0.5836187371181211, "300–1000 Hz_40ms_log_rms_mae_db": 2.238357790790354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08583322661580778, "300–1000 Hz_40ms_envelope_correlation": 0.9953905864793915, "300–1000 Hz_40ms_modulation_mae_db": 0.9156789471507389, "1000–3000 Hz_10ms_log_rms_mae_db": 1.904129867707675, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14523631231884546, "1000–3000 Hz_10ms_envelope_correlation": 0.9866090875742735, "1000–3000 Hz_10ms_modulation_mae_db": 0.5967578812172629, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1676934045834446, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0935957807054582, "1000–3000 Hz_40ms_envelope_correlation": 0.9949606916515187, "1000–3000 Hz_40ms_modulation_mae_db": 0.3402488086868026, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7154171379630292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15832376480619753, "3000–6000 Hz_10ms_envelope_correlation": 0.9827338607138799, "3000–6000 Hz_10ms_modulation_mae_db": 1.0206364108358768, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1021976908638331, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08592253734506435, "3000–6000 Hz_40ms_envelope_correlation": 0.9950404744096789, "3000–6000 Hz_40ms_modulation_mae_db": 1.4542920094922849, "mr_stft_db": 5.896623097889097, "band_mae_db": 0.17129428887089881, "rms_gain_error_db": -0.04663860797719463, "output_peak": 0.586603677277274}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6761284556555656, "全帯域_10ms_rms_envelope_nrmse": 0.09570994246421524, "全帯域_10ms_envelope_correlation": 0.9897741582443133, "全帯域_10ms_modulation_mae_db": 0.2927095566163218, "全帯域_40ms_log_rms_mae_db": 0.7934655817376184, "全帯域_40ms_rms_envelope_nrmse": 0.049811026773772145, "全帯域_40ms_envelope_correlation": 0.9970448767524412, "全帯域_40ms_modulation_mae_db": 0.537866669678993, "300–1000 Hz_10ms_log_rms_mae_db": 2.785759570298062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09777590792161349, "300–1000 Hz_10ms_envelope_correlation": 0.9940239195817636, "300–1000 Hz_10ms_modulation_mae_db": 1.0267296510939357, "300–1000 Hz_40ms_log_rms_mae_db": 2.112107263834456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06726136341957784, "300–1000 Hz_40ms_envelope_correlation": 0.9971489182397338, "300–1000 Hz_40ms_modulation_mae_db": 1.3202081320925274, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9341967731802359, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14581881058657217, "1000–3000 Hz_10ms_envelope_correlation": 0.9866725721313523, "1000–3000 Hz_10ms_modulation_mae_db": 0.4753681257024575, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1901687848450397, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08978323921662988, "1000–3000 Hz_40ms_envelope_correlation": 0.9957057250599101, "1000–3000 Hz_40ms_modulation_mae_db": 0.46552964890023435, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7747655152609678, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.159555398024697, "3000–6000 Hz_10ms_envelope_correlation": 0.9824074894520842, "3000–6000 Hz_10ms_modulation_mae_db": 0.5481639571701629, "3000–6000 Hz_40ms_log_rms_mae_db": 1.165874584548503, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08516056981657202, "3000–6000 Hz_40ms_envelope_correlation": 0.994776901547534, "3000–6000 Hz_40ms_modulation_mae_db": 0.5364099578272588, "mr_stft_db": 5.878367286320749, "band_mae_db": 0.11782446931895435, "rms_gain_error_db": -0.051947800499028696, "output_peak": 0.5754641736452663}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88640, "output_samples": 88640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.096978985292122, "spectral_convergence": 0.3413435471922017, "spectral_convergence_status": "ok", "amplitude_floor": 1.5935254482600421e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.709189966831276, "spectral_convergence": 0.29263490522949265, "spectral_convergence_status": "ok", "amplitude_floor": 9.146897175620752e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.852579601197302, "spectral_convergence": 0.3062090819151416, "spectral_convergence_status": "ok", "amplitude_floor": 7.905477970735852e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007242342814613898, "output_power": 0.0007165117659333351, "delta_db": -0.04655751797553117, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001748392652036385, "output_power": 0.0017091897418226225, "delta_db": -0.0984869499965025, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009820880914630636, "output_power": 0.0009650386978420505, "delta_db": -0.07605716128763076, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002665715938534463, "output_power": 0.00023788924415121785, "delta_db": -0.49439062243019943, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.016487980563797e-05, "output_power": 6.291158856098689e-05, "delta_db": 0.1938759813197721, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00015485832573279259, "output_power": 0.00015139366598530585, "delta_db": -0.0982685401908579, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1679394622000823, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10426257820153229, "pitch": null, "output_rms_dbfs": -24.153368494691346, "output_peak": 0.5692346967899763, "reference_rms_dbfs": -24.04910591648981, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8862495177735665}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00022148173042756648, "log_rms_mae_db": 1.6389901266967921, "rms_envelope_nrmse": 0.09665249429537519, "envelope_correlation": 0.9896582381255375, "rms_gain_error_db": -0.10438200433906891, "reference_mean_power": 0.0039398267431541775, "output_mean_power": 0.003846262524000084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.3094240146918987e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011389510990882513, "output_power": 0.0011210344000093576, "reference_active": true, "delta_db": -0.06886138488460795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017715594495014008, "output_power": 0.00017164098336265414, "reference_active": true, "delta_db": -0.1373473682723674}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5198627564616767e-05, "output_power": 3.561614510753985e-05, "reference_active": true, "delta_db": 1.502700243514434}], "active_band_count": 3, "active_band_mae_db": 0.5696363322238032}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.00020322184022683357, "log_rms_mae_db": 0.7383768326852473, "rms_envelope_nrmse": 0.046340090763661586, "envelope_correlation": 0.9975513137232447, "rms_gain_error_db": -0.10477056528067494, "reference_mean_power": 0.0039603691212238695, "output_mean_power": 0.0038659711535099354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.1092095592585075e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010485738336262066, "output_power": 0.0010351042266236183, "reference_active": true, "delta_db": -0.05614934083332252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.703453851742181e-05, "output_power": 8.158102417001914e-05, "reference_active": true, "delta_db": -0.2810247774657911}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.852514808001562e-06, "output_power": 2.309163094245835e-06, "reference_active": true, "delta_db": 0.9569291967372489}], "active_band_count": 3, "active_band_mae_db": 0.43136777167878754}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00017913726163896074, "log_rms_mae_db": 3.001359074508102, "rms_envelope_nrmse": 0.09804481178877072, "envelope_correlation": 0.9940167904639269, "rms_gain_error_db": 0.015243912588198068, "reference_mean_power": 0.001348381907516537, "output_mean_power": 0.0013531230987597267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 1.724352265721046e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008753679492081999, "output_power": 0.0008451361983880329, "reference_active": true, "delta_db": -0.15263937719936196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012951638097036193, "output_power": 0.00011693808140644475, "reference_active": true, "delta_db": -0.44368736454789454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6972503859652748e-05, "output_power": 1.2553626157478501e-05, "reference_active": true, "delta_db": -1.3097672473562858}], "active_band_count": 3, "active_band_mae_db": 0.6353646630345141}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.0001658599445391014, "log_rms_mae_db": 2.243186359652414, "rms_envelope_nrmse": 0.06503980980418535, "envelope_correlation": 0.9974063506371075, "rms_gain_error_db": 0.015231107582813345, "reference_mean_power": 0.001355729112927364, "output_mean_power": 0.001360492127108453}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 1.6447728018068126e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008253872294632201, "output_power": 0.0008003573906015755, "reference_active": true, "delta_db": -0.13373785487206077}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.001034181994634e-05, "output_power": 5.4569405931739915e-05, "reference_active": true, "delta_db": -0.41276874647473144}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.081003459949387e-06, "output_power": 1.2964828879358605e-06, "reference_active": true, "delta_db": -2.0550601332362493}], "active_band_count": 3, "active_band_mae_db": 0.8671889115276805}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00010266849739942118, "log_rms_mae_db": 1.848728890482448, "rms_envelope_nrmse": 0.14906141255912367, "envelope_correlation": 0.9863095716090039, "rms_gain_error_db": -0.4073840024407957, "reference_mean_power": 0.0002406311059478081, "output_mean_power": 0.00021908536469027255}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.4743465614681157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013528556332001226, "output_power": 0.00011871440052962348, "reference_active": true, "delta_db": -0.5674805055778933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.422298628989197e-05, "output_power": 1.4088952935099895e-05, "reference_active": true, "delta_db": -0.0411207305818769}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7979674926628947e-06, "output_power": 2.385846326042801e-06, "reference_active": true, "delta_db": -0.6920019742914629}], "active_band_count": 3, "active_band_mae_db": 0.43353440348374433}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 8.271272308600344e-05, "log_rms_mae_db": 1.1225613551177707, "rms_envelope_nrmse": 0.10393580946907664, "envelope_correlation": 0.9940249769529311, "rms_gain_error_db": -0.40740225442050465, "reference_mean_power": 0.00024194203098044546, "output_mean_power": 0.00022027798574170262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.355449255956094e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012689841334701795, "output_power": 0.00011280307596919694, "reference_active": true, "delta_db": -0.5113524963842218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.520296796235256e-06, "output_power": 6.676434386741096e-06, "reference_active": true, "delta_db": 0.10277220535655907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.927693257279988e-07, "output_power": 1.5817974051773964e-07, "reference_active": true, "delta_db": -0.8588706950872143}], "active_band_count": 3, "active_band_mae_db": 0.49099846560933164}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 5.887039168463902e-05, "log_rms_mae_db": 1.8142491244614052, "rms_envelope_nrmse": 0.1414704173671724, "envelope_correlation": 0.9861544930182277, "rms_gain_error_db": -0.1247940214409472, "reference_mean_power": 0.00013131711251995087, "output_mean_power": 0.00012759742863243103}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 6.872487837717039e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.115138542486786e-05, "output_power": 2.9182267870527805e-05, "reference_active": true, "delta_db": -0.2835832658494834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.223707279035506e-06, "output_power": 2.9039280858605588e-06, "reference_active": true, "delta_db": -0.4536974275676456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5460080528529549e-06, "output_power": 1.245418424602009e-06, "reference_active": true, "delta_db": -0.9389646539877315}], "active_band_count": 3, "active_band_mae_db": 0.5587484491349536}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 5.4692297362327305e-05, "log_rms_mae_db": 1.1234278829119078, "rms_envelope_nrmse": 0.07606161332467117, "envelope_correlation": 0.9959253100176699, "rms_gain_error_db": -0.12480472970713066, "reference_mean_power": 0.00013203257062264164, "output_mean_power": 0.00012829230436739436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 6.492145466124568e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.913804296383342e-05, "output_power": 2.7407135253340016e-05, "reference_active": true, "delta_db": -0.2659673630004194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5333654526802403e-06, "output_power": 1.334031864974443e-06, "reference_active": true, "delta_db": -0.6047947085032095}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.807699107498543e-08, "output_power": 6.050496475038851e-08, "reference_active": true, "delta_db": -1.6307145739785733}], "active_band_count": 3, "active_band_mae_db": 0.8338255484940674}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7329680644420824, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d94e889d9230acfb7bf339e7b6412757d63ad1cc90f460297178fa204674bb4f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9861627202335875, "standard_deviation": 0.001018083587874797, "minimum": 0.9845475913589262, "maximum": 0.9873931221982548, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8884888102879152, "standard_deviation": 0.03661879823517096, "minimum": 1.8240776064950068, "maximum": 1.9341967731802359, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5768891238879198, "standard_deviation": 0.22622261844431107, "minimum": 0.3834021987433611, "maximum": 1.0852053137599562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1473633423383398, "standard_deviation": 0.006477595388150389, "minimum": 0.13864247277792105, "maximum": 0.15834871623130184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9954505148276376, "standard_deviation": 0.0006729779251412503, "minimum": 0.9940249769529311, "maximum": 0.9961827029437348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1465510278088327, "standard_deviation": 0.03205541063092308, "minimum": 1.0964304601574044, "maximum": 1.1901687848450397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5879330355922445, "standard_deviation": 0.30008308122068295, "minimum": 0.3402488086868026, "maximum": 1.2051823588593202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08932070439156717, "standard_deviation": 0.00926614321092774, "minimum": 0.07781915031620333, "maximum": 0.10393580946907664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823482212110088, "standard_deviation": 0.0019505958294827295, "minimum": 0.9799476442001803, "maximum": 0.9861544930182277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7699108490818398, "standard_deviation": 0.032674526551251575, "minimum": 1.7154171379630292, "maximum": 1.8142491244614052, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7353411838221714, "standard_deviation": 0.2158054491910387, "minimum": 0.4428272366892783, "maximum": 1.0206364108358768, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1599868652319844, "standard_deviation": 0.009435460457226754, "minimum": 0.1414704173671724, "maximum": 0.17217617198096533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9948663309245331, "standard_deviation": 0.001042021791859838, "minimum": 0.9928453652974082, "maximum": 0.9962072859920424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1331406842237204, "standard_deviation": 0.0231759038849866, "minimum": 1.100766216020107, "maximum": 1.165874584548503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0562956377554549, "standard_deviation": 0.30483563154170223, "minimum": 0.5364099578272588, "maximum": 1.4542920094922849, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08536300231467198, "standard_deviation": 0.0076739605342280555, "minimum": 0.07606161332467117, "maximum": 0.099415189547284, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9928096289051351, "standard_deviation": 0.0011476238056357616, "minimum": 0.9909180551982264, "maximum": 0.9940239195817636, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.8679407669314143, "standard_deviation": 0.0699300314939242, "minimum": 2.785759570298062, "maximum": 3.001359074508102, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.703205117785174, "standard_deviation": 0.4171843772073201, "minimum": 0.20311522142113014, "maximum": 1.360287507775795, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10666654332976372, "standard_deviation": 0.008280992900597252, "minimum": 0.09777590792161349, "maximum": 0.12087376611112037, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965904341424726, "standard_deviation": 0.0009552788431884942, "minimum": 0.9948285451994, "maximum": 0.9974063506371075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.1583134116866907, "standard_deviation": 0.08099689528036567, "minimum": 2.047789054769293, "maximum": 2.243186359652414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8504486517775194, "standard_deviation": 0.5200499355759732, "minimum": 0.10601971144496597, "maximum": 1.5322038074945945, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07318683898423564, "standard_deviation": 0.009219571323325513, "minimum": 0.06503980980418535, "maximum": 0.0890474751087725, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16606103216386792, "standard_deviation": 0.05376696057246824, "minimum": 0.060270647379312564, "maximum": 0.2264955634865834, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.884484037614834, "standard_deviation": 0.008465542376049868, "minimum": 5.872836243108004, "maximum": 5.896623097889097, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.586821980901352, "standard_deviation": 0.012643044383157057, "minimum": 0.5692346967899763, "maximum": 0.6037706624483472, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08853650112248751, "standard_deviation": 0.027859475786751853, "minimum": -0.11799528957746892, "maximum": -0.04663860797719463, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901415146049773, "standard_deviation": 0.0005640811208895012, "minimum": 0.9893818500362526, "maximum": 0.9908758409453456, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5911722004377316, "standard_deviation": 0.04894444350418776, "minimum": 1.5311806925310347, "maximum": 1.6761284556555656, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4416772541611354, "standard_deviation": 0.12816857805368534, "minimum": 0.21038422562119155, "maximum": 0.5761423414809926, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09425258906089506, "standard_deviation": 0.002531166938980922, "minimum": 0.09128009394083045, "maximum": 0.09769736952383355, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972834684955292, "standard_deviation": 0.0002571469810848384, "minimum": 0.9970448767524412, "maximum": 0.9977479422126562, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7226940669841281, "standard_deviation": 0.039645570512870175, "minimum": 0.6622320985110404, "maximum": 0.7934655817376184, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4697836551466369, "standard_deviation": 0.17845458840748868, "minimum": 0.24521520988211062, "maximum": 0.8050240388133606, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04845808219922409, "standard_deviation": 0.0021854425770645694, "minimum": 0.04427139990914492, "maximum": 0.05069868748318049, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0038", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.91097092628479, "predicted_mos": 2.91097092628479, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "2d3ac475e049193f0656a7e97f77290c7406739e2c0caac75c9e8d2d33ab366f", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.876046100165695, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6572703879056538, "全帯域_10ms_rms_envelope_nrmse": 0.09814044440557099, "全帯域_10ms_envelope_correlation": 0.9893141238396537, "全帯域_10ms_modulation_mae_db": 0.562365369068756, "全帯域_40ms_log_rms_mae_db": 0.7482729882284052, "全帯域_40ms_rms_envelope_nrmse": 0.04623612615001749, "全帯域_40ms_envelope_correlation": 0.9975248791785006, "全帯域_40ms_modulation_mae_db": 0.42682360050560786, "300–1000 Hz_10ms_log_rms_mae_db": 2.7364766126570514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10005571912182623, "300–1000 Hz_10ms_envelope_correlation": 0.9937552794803848, "300–1000 Hz_10ms_modulation_mae_db": 0.6355388073524547, "300–1000 Hz_40ms_log_rms_mae_db": 1.97118870001957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06642988586028128, "300–1000 Hz_40ms_envelope_correlation": 0.9972686885896541, "300–1000 Hz_40ms_modulation_mae_db": 0.8434732933122495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.705351252791054, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14707279071711796, "1000–3000 Hz_10ms_envelope_correlation": 0.9866431704372586, "1000–3000 Hz_10ms_modulation_mae_db": 0.4513341120148377, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9479950031829747, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1022384141210075, "1000–3000 Hz_40ms_envelope_correlation": 0.9942091918224842, "1000–3000 Hz_40ms_modulation_mae_db": 0.48654928997508295, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6321209945828505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14447294069784447, "3000–6000 Hz_10ms_envelope_correlation": 0.9855571530843668, "3000–6000 Hz_10ms_modulation_mae_db": 0.6452407340390468, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9378724120129597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07848457049661864, "3000–6000 Hz_40ms_envelope_correlation": 0.9957047110600479, "3000–6000 Hz_40ms_modulation_mae_db": 0.9504944191589816, "mr_stft_db": 5.813223879179053, "band_mae_db": 0.1551122225107939, "rms_gain_error_db": -0.09128339783384498, "output_peak": 0.5722414147551796}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5678307688890136, "全帯域_10ms_rms_envelope_nrmse": 0.09307806726567833, "全帯域_10ms_envelope_correlation": 0.9903539900310917, "全帯域_10ms_modulation_mae_db": 0.33759673681340235, "全帯域_40ms_log_rms_mae_db": 0.708097901103595, "全帯域_40ms_rms_envelope_nrmse": 0.05215405486905844, "全帯域_40ms_envelope_correlation": 0.9968711802998993, "全帯域_40ms_modulation_mae_db": 0.1756512257209967, "300–1000 Hz_10ms_log_rms_mae_db": 2.6020459832470424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11636010215636311, "300–1000 Hz_10ms_envelope_correlation": 0.991351903348693, "300–1000 Hz_10ms_modulation_mae_db": 0.44463516902014305, "300–1000 Hz_40ms_log_rms_mae_db": 1.8796975244009506, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08787911123611479, "300–1000 Hz_40ms_envelope_correlation": 0.9949753815702435, "300–1000 Hz_40ms_modulation_mae_db": 0.5040080530589436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7556794294047455, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15159339865189855, "1000–3000 Hz_10ms_envelope_correlation": 0.985951269415294, "1000–3000 Hz_10ms_modulation_mae_db": 0.7257023212791877, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9851427330746123, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0945709899423378, "1000–3000 Hz_40ms_envelope_correlation": 0.9956712409472765, "1000–3000 Hz_40ms_modulation_mae_db": 0.9983698191461933, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6006034853426274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16338131302686904, "3000–6000 Hz_10ms_envelope_correlation": 0.98176598084769, "3000–6000 Hz_10ms_modulation_mae_db": 0.8473309345115246, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9568700595625814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08997583243055335, "3000–6000 Hz_40ms_envelope_correlation": 0.9942893695547195, "3000–6000 Hz_40ms_modulation_mae_db": 1.3988758249009325, "mr_stft_db": 5.817977432414028, "band_mae_db": 0.1922104751692094, "rms_gain_error_db": -0.11117818230914175, "output_peak": 0.6093568007565099}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.695583242616428, "全帯域_10ms_rms_envelope_nrmse": 0.10007381640990644, "全帯域_10ms_envelope_correlation": 0.9889049896639731, "全帯域_10ms_modulation_mae_db": 0.5013202221402336, "全帯域_40ms_log_rms_mae_db": 0.7520123522036818, "全帯域_40ms_rms_envelope_nrmse": 0.05060617812784962, "全帯域_40ms_envelope_correlation": 0.9970684860648644, "全帯域_40ms_modulation_mae_db": 0.6046524835522843, "300–1000 Hz_10ms_log_rms_mae_db": 2.5968266780260914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11317063684265237, "300–1000 Hz_10ms_envelope_correlation": 0.9918065007435587, "300–1000 Hz_10ms_modulation_mae_db": 0.41782768824130395, "300–1000 Hz_40ms_log_rms_mae_db": 1.8327126890195788, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07566899807382223, "300–1000 Hz_40ms_envelope_correlation": 0.9963372967447194, "300–1000 Hz_40ms_modulation_mae_db": 0.3975597799138941, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7641097110151756, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15325551272096288, "1000–3000 Hz_10ms_envelope_correlation": 0.9844332114400317, "1000–3000 Hz_10ms_modulation_mae_db": 0.47635809443793736, "1000–3000 Hz_40ms_log_rms_mae_db": 0.992732845913451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08181141733394123, "1000–3000 Hz_40ms_envelope_correlation": 0.9957541612331446, "1000–3000 Hz_40ms_modulation_mae_db": 0.6061128048500186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6555227449908028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15586462656343789, "3000–6000 Hz_10ms_envelope_correlation": 0.9831671922739686, "3000–6000 Hz_10ms_modulation_mae_db": 1.1105507654457274, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9994494564480156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08595734224603055, "3000–6000 Hz_40ms_envelope_correlation": 0.9948058422693568, "3000–6000 Hz_40ms_modulation_mae_db": 1.4042870202419355, "mr_stft_db": 5.8195670597298585, "band_mae_db": 0.20811427140470162, "rms_gain_error_db": -0.11298519021101569, "output_peak": 0.6096301346505867}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5713998939827392, "全帯域_10ms_rms_envelope_nrmse": 0.09311822351588413, "全帯域_10ms_envelope_correlation": 0.9904138307640922, "全帯域_10ms_modulation_mae_db": 0.4989887031299958, "全帯域_40ms_log_rms_mae_db": 0.727956352311525, "全帯域_40ms_rms_envelope_nrmse": 0.044663609458851304, "全帯域_40ms_envelope_correlation": 0.9976914461281511, "全帯域_40ms_modulation_mae_db": 0.4285576136059188, "300–1000 Hz_10ms_log_rms_mae_db": 2.6487584631646657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10267976143503715, "300–1000 Hz_10ms_envelope_correlation": 0.9933809693956905, "300–1000 Hz_10ms_modulation_mae_db": 0.15533137257123905, "300–1000 Hz_40ms_log_rms_mae_db": 1.9495544853319244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0659994755546435, "300–1000 Hz_40ms_envelope_correlation": 0.9973054908195115, "300–1000 Hz_40ms_modulation_mae_db": 0.20800587088384861, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7188092987719916, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13986569402554028, "1000–3000 Hz_10ms_envelope_correlation": 0.9869941522545972, "1000–3000 Hz_10ms_modulation_mae_db": 0.47321885734570107, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9289820720961607, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08059803151546716, "1000–3000 Hz_40ms_envelope_correlation": 0.9956823760515549, "1000–3000 Hz_40ms_modulation_mae_db": 0.7019872874413137, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5891340021470262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1671120989268428, "3000–6000 Hz_10ms_envelope_correlation": 0.9806808554657502, "3000–6000 Hz_10ms_modulation_mae_db": 0.6654260812098611, "3000–6000 Hz_40ms_log_rms_mae_db": 0.967078760224844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10020616825075589, "3000–6000 Hz_40ms_envelope_correlation": 0.9928296214224145, "3000–6000 Hz_40ms_modulation_mae_db": 1.4383003340608573, "mr_stft_db": 5.810228428057311, "band_mae_db": 0.0880084250110237, "rms_gain_error_db": -0.07811298714462712, "output_peak": 0.5838756015099213}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5779107849991665, "全帯域_10ms_rms_envelope_nrmse": 0.09697227833444107, "全帯域_10ms_envelope_correlation": 0.9895370908041479, "全帯域_10ms_modulation_mae_db": 0.3960794627018867, "全帯域_40ms_log_rms_mae_db": 0.706788933466951, "全帯域_40ms_rms_envelope_nrmse": 0.04968111413705388, "全帯域_40ms_envelope_correlation": 0.9971272777115372, "全帯域_40ms_modulation_mae_db": 0.2616392106388879, "300–1000 Hz_10ms_log_rms_mae_db": 2.5813698426091136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10766205187266856, "300–1000 Hz_10ms_envelope_correlation": 0.9926842731145807, "300–1000 Hz_10ms_modulation_mae_db": 1.4522426440787417, "300–1000 Hz_40ms_log_rms_mae_db": 1.9149329626092022, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0725889741998146, "300–1000 Hz_40ms_envelope_correlation": 0.9966994569258434, "300–1000 Hz_40ms_modulation_mae_db": 1.6786628411188342, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7513541325845199, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13632824599006918, "1000–3000 Hz_10ms_envelope_correlation": 0.987854569175481, "1000–3000 Hz_10ms_modulation_mae_db": 0.6079168799900562, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9949278284506449, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07779844439084382, "1000–3000 Hz_40ms_envelope_correlation": 0.9962840980440731, "1000–3000 Hz_40ms_modulation_mae_db": 0.4408312112708221, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5859822403558688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1600342646529901, "3000–6000 Hz_10ms_envelope_correlation": 0.9823787812325726, "3000–6000 Hz_10ms_modulation_mae_db": 1.172602348162448, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9216309665913672, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07802929722679572, "3000–6000 Hz_40ms_envelope_correlation": 0.9960853038688342, "3000–6000 Hz_40ms_modulation_mae_db": 1.5371667398021664, "mr_stft_db": 5.804311705192156, "band_mae_db": 0.22421614302313927, "rms_gain_error_db": -0.09619125441102645, "output_peak": 0.6028261381021172}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5712282259333, "全帯域_10ms_rms_envelope_nrmse": 0.09362598085655013, "全帯域_10ms_envelope_correlation": 0.9902562435417095, "全帯域_10ms_modulation_mae_db": 0.37372708719589137, "全帯域_40ms_log_rms_mae_db": 0.6853270855029945, "全帯域_40ms_rms_envelope_nrmse": 0.05014176167754786, "全帯域_40ms_envelope_correlation": 0.9971594827960092, "全帯域_40ms_modulation_mae_db": 0.6405837643642557, "300–1000 Hz_10ms_log_rms_mae_db": 2.6180905103031065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1077915724796023, "300–1000 Hz_10ms_envelope_correlation": 0.9926479308041648, "300–1000 Hz_10ms_modulation_mae_db": 1.186538929582303, "300–1000 Hz_40ms_log_rms_mae_db": 1.8454119370394289, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07004306013956528, "300–1000 Hz_40ms_envelope_correlation": 0.9969080108597037, "300–1000 Hz_40ms_modulation_mae_db": 1.4001396021571464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7702743220073118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16111986537115075, "1000–3000 Hz_10ms_envelope_correlation": 0.9841457259416828, "1000–3000 Hz_10ms_modulation_mae_db": 1.0701580083828686, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0177708151403415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09994795104648059, "1000–3000 Hz_40ms_envelope_correlation": 0.9952611914283627, "1000–3000 Hz_40ms_modulation_mae_db": 1.2686044567279549, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6137090467807789, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1714315658304545, "3000–6000 Hz_10ms_envelope_correlation": 0.979897540406932, "3000–6000 Hz_10ms_modulation_mae_db": 0.7077220347439002, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9384155882727347, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08873002426492962, "3000–6000 Hz_40ms_envelope_correlation": 0.9944063445347098, "3000–6000 Hz_40ms_modulation_mae_db": 0.8546172761622017, "mr_stft_db": 5.798875224939312, "band_mae_db": 0.23898558732770891, "rms_gain_error_db": -0.12136618831767587, "output_peak": 0.5943712956207228}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6059950076765903, "全帯域_10ms_rms_envelope_nrmse": 0.09133086171687665, "全帯域_10ms_envelope_correlation": 0.9908808360902799, "全帯域_10ms_modulation_mae_db": 0.5156841651889484, "全帯域_40ms_log_rms_mae_db": 0.7535988772106539, "全帯域_40ms_rms_envelope_nrmse": 0.050897142326164595, "全帯域_40ms_envelope_correlation": 0.9969999217304444, "全帯域_40ms_modulation_mae_db": 0.427934338866518, "300–1000 Hz_10ms_log_rms_mae_db": 2.636309276127534, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12190903183105585, "300–1000 Hz_10ms_envelope_correlation": 0.9907170759248345, "300–1000 Hz_10ms_modulation_mae_db": 0.6956960700671173, "300–1000 Hz_40ms_log_rms_mae_db": 1.899036383188532, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08561976839566804, "300–1000 Hz_40ms_envelope_correlation": 0.995375258850316, "300–1000 Hz_40ms_modulation_mae_db": 0.9848759225433735, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7448762773445425, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14259534557765752, "1000–3000 Hz_10ms_envelope_correlation": 0.9871477116172009, "1000–3000 Hz_10ms_modulation_mae_db": 0.692653573372155, "1000–3000 Hz_40ms_log_rms_mae_db": 0.969392870961153, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09119078807968857, "1000–3000 Hz_40ms_envelope_correlation": 0.9953059096949214, "1000–3000 Hz_40ms_modulation_mae_db": 0.45017568852946277, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5508301783583922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1572352034749696, "3000–6000 Hz_10ms_envelope_correlation": 0.9830321562922758, "3000–6000 Hz_10ms_modulation_mae_db": 1.1875669062915055, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9327988666309895, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08472449416637903, "3000–6000 Hz_40ms_envelope_correlation": 0.9953049401784699, "3000–6000 Hz_40ms_modulation_mae_db": 1.6701001783442304, "mr_stft_db": 5.813856201615335, "band_mae_db": 0.1964904696128976, "rms_gain_error_db": -0.05338602531082161, "output_peak": 0.5886940628242427}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7386459991246845, "全帯域_10ms_rms_envelope_nrmse": 0.09683289292408666, "全帯域_10ms_envelope_correlation": 0.9895650375837852, "全帯域_10ms_modulation_mae_db": 0.20472186744466822, "全帯域_40ms_log_rms_mae_db": 0.7962525950949904, "全帯域_40ms_rms_envelope_nrmse": 0.05064008086748725, "全帯域_40ms_envelope_correlation": 0.9969506117086707, "全帯域_40ms_modulation_mae_db": 0.4219608169477982, "300–1000 Hz_10ms_log_rms_mae_db": 2.548799404435182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10002594393639817, "300–1000 Hz_10ms_envelope_correlation": 0.993695757578739, "300–1000 Hz_10ms_modulation_mae_db": 1.1466714866339223, "300–1000 Hz_40ms_log_rms_mae_db": 1.8462601487162618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06582080829036774, "300–1000 Hz_40ms_envelope_correlation": 0.9972458820717067, "300–1000 Hz_40ms_modulation_mae_db": 1.4621049730869498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7647967727905975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14041368203832538, "1000–3000 Hz_10ms_envelope_correlation": 0.987643886716703, "1000–3000 Hz_10ms_modulation_mae_db": 0.5600130803730642, "1000–3000 Hz_40ms_log_rms_mae_db": 0.960924240070773, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08435075839821311, "1000–3000 Hz_40ms_envelope_correlation": 0.9962682019039525, "1000–3000 Hz_40ms_modulation_mae_db": 0.7024010152259786, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5850662583036808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16112697992486447, "3000–6000 Hz_10ms_envelope_correlation": 0.9819941613599058, "3000–6000 Hz_10ms_modulation_mae_db": 0.6486005567716201, "3000–6000 Hz_40ms_log_rms_mae_db": 0.954226271818425, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08634466675618652, "3000–6000 Hz_40ms_envelope_correlation": 0.9946074919205395, "3000–6000 Hz_40ms_modulation_mae_db": 0.6019579329758744, "mr_stft_db": 5.797106651726494, "band_mae_db": 0.13366635677965635, "rms_gain_error_db": -0.04713190755440093, "output_peak": 0.5695568320281544}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88640, "output_samples": 88640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.022045910267693, "spectral_convergence": 0.3402556419317259, "spectral_convergence_status": "ok", "amplitude_floor": 1.5935254482600421e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6422602336434995, "spectral_convergence": 0.2918666166913912, "spectral_convergence_status": "ok", "amplitude_floor": 9.146897175620752e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.775365493625969, "spectral_convergence": 0.3066030290732164, "spectral_convergence_status": "ok", "amplitude_floor": 7.905477970735852e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007242342814613898, "output_power": 0.0007180868274338987, "delta_db": -0.037021180823588316, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001748392652036385, "output_power": 0.001714910461996517, "delta_db": -0.08397522798019794, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009820880914630636, "output_power": 0.0009713080910410274, "delta_db": -0.047934384922243, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002665715938534463, "output_power": 0.00023797904793671518, "delta_db": -0.4927514592580796, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.016487980563797e-05, "output_power": 6.204537783930508e-05, "delta_db": 0.13366380213904966, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00015485832573279259, "output_power": 0.0001501073062990209, "delta_db": -0.1353272799416047, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1551122225107939, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09128339783384498, "pitch": null, "output_rms_dbfs": -24.140389314323656, "output_peak": 0.5722414147551796, "reference_rms_dbfs": -24.04910591648981, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.813223879179053}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00022148173042756648, "log_rms_mae_db": 1.6572703879056538, "rms_envelope_nrmse": 0.09814044440557099, "envelope_correlation": 0.9893141238396537, "rms_gain_error_db": -0.09149672673157175, "reference_mean_power": 0.0039398267431541775, "output_mean_power": 0.003857691118268872}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.3094240146918987e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011389510990882513, "output_power": 0.0011319009664381026, "reference_active": true, "delta_db": -0.02696647298618802}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017715594495014008, "output_power": 0.00017253537562698535, "reference_active": true, "delta_db": -0.11477577151602081}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5198627564616767e-05, "output_power": 3.596766743449697e-05, "reference_active": true, "delta_db": 1.5453538627040593}], "active_band_count": 3, "active_band_mae_db": 0.562365369068756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.00020322184022683357, "log_rms_mae_db": 0.7482729882284052, "rms_envelope_nrmse": 0.04623612615001749, "envelope_correlation": 0.9975248791785006, "rms_gain_error_db": -0.0918420005002857, "reference_mean_power": 0.0039603691212238695, "output_mean_power": 0.0038774969567756524}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.1092095592585075e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010485738336262066, "output_power": 0.0010455524751475664, "reference_active": true, "delta_db": -0.012531815488830016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.703453851742181e-05, "output_power": 8.22485084551145e-05, "reference_active": true, "delta_db": -0.24563599973621145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.852514808001562e-06, "output_power": 2.344185562229651e-06, "reference_active": true, "delta_db": 1.022302986291782}], "active_band_count": 3, "active_band_mae_db": 0.42682360050560786}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00017913726163896074, "log_rms_mae_db": 2.7364766126570514, "rms_envelope_nrmse": 0.10005571912182623, "envelope_correlation": 0.9937552794803848, "rms_gain_error_db": 0.04536986116560918, "reference_mean_power": 0.001348381907516537, "output_mean_power": 0.0013625420141384428}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 1.724352265721046e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008753679492081999, "output_power": 0.0008541993340966874, "reference_active": true, "delta_db": -0.10631412744989316}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012951638097036193, "output_power": 0.00011759703124223347, "reference_active": true, "delta_db": -0.41928342542576913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6972503859652748e-05, "output_power": 1.2349348003824335e-05, "reference_active": true, "delta_db": -1.3810188691817018}], "active_band_count": 3, "active_band_mae_db": 0.6355388073524547}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.0001658599445391014, "log_rms_mae_db": 1.97118870001957, "rms_envelope_nrmse": 0.06642988586028128, "envelope_correlation": 0.9972686885896541, "rms_gain_error_db": 0.04536575681327962, "reference_mean_power": 0.001355729112927364, "output_mean_power": 0.001369965081927601}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 1.6447728018068126e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008253872294632201, "output_power": 0.0008079664632432283, "reference_active": true, "delta_db": -0.09264410489806212}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.001034181994634e-05, "output_power": 5.4774008550089606e-05, "reference_active": true, "delta_db": -0.3965157526824084}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.081003459949387e-06, "output_power": 1.30060913512762e-06, "reference_active": true, "delta_db": -2.041260022356278}], "active_band_count": 3, "active_band_mae_db": 0.8434732933122495}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00010266849739942118, "log_rms_mae_db": 1.705351252791054, "rms_envelope_nrmse": 0.14707279071711796, "envelope_correlation": 0.9866431704372586, "rms_gain_error_db": -0.4192730645816033, "reference_mean_power": 0.0002406311059478081, "output_mean_power": 0.00021848642604622274}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.4743465614681157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013528556332001226, "output_power": 0.00012062070737793229, "reference_active": true, "delta_db": -0.49829583303886826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.422298628989197e-05, "output_power": 1.3613940342369331e-05, "reference_active": true, "delta_db": -0.19006948253465453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7979674926628947e-06, "output_power": 2.400374267640044e-06, "reference_active": true, "delta_db": -0.6656370204709904}], "active_band_count": 3, "active_band_mae_db": 0.4513341120148377}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 8.271272308600344e-05, "log_rms_mae_db": 0.9479950031829747, "rms_envelope_nrmse": 0.1022384141210075, "envelope_correlation": 0.9942091918224842, "rms_gain_error_db": -0.41927769299163387, "reference_mean_power": 0.00024194203098044546, "output_mean_power": 0.0002196764758049566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.355449255956094e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012689841334701795, "output_power": 0.00011468138036636024, "reference_active": true, "delta_db": -0.4396328029429006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.520296796235256e-06, "output_power": 6.417970094925497e-06, "reference_active": true, "delta_db": -0.06869675623144057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.927693257279988e-07, "output_power": 1.5484817712826047e-07, "reference_active": true, "delta_db": -0.9513183107509077}], "active_band_count": 3, "active_band_mae_db": 0.48654928997508295}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 5.887039168463902e-05, "log_rms_mae_db": 1.6321209945828505, "rms_envelope_nrmse": 0.14447294069784447, "envelope_correlation": 0.9855571530843668, "rms_gain_error_db": -0.1560001915576945, "reference_mean_power": 0.00013131711251995087, "output_mean_power": 0.00012668386519773024}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 6.872487837717039e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.115138542486786e-05, "output_power": 2.944282801791562e-05, "reference_active": true, "delta_db": -0.2449784405208423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.223707279035506e-06, "output_power": 2.7962938403696934e-06, "reference_active": true, "delta_db": -0.6177279382177934}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5460080528529549e-06, "output_power": 1.2075640336023776e-06, "reference_active": true, "delta_db": -1.0730158233785048}], "active_band_count": 3, "active_band_mae_db": 0.6452407340390468}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 5.4692297362327305e-05, "log_rms_mae_db": 0.9378724120129597, "rms_envelope_nrmse": 0.07848457049661864, "envelope_correlation": 0.9957047110600479, "rms_gain_error_db": -0.15600263908125664, "reference_mean_power": 0.00013203257062264164, "output_mean_power": 0.00012737400808710076}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 6.492145466124568e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.913804296383342e-05, "output_power": 2.7674760974678307e-05, "reference_active": true, "delta_db": -0.22376500739543073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5333654526802403e-06, "output_power": 1.2707958375583942e-06, "reference_active": true, "delta_db": -0.8156989056392219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.807699107498543e-08, "output_power": 5.8031064238259693e-08, "reference_active": true, "delta_db": -1.8120193444422918}], "active_band_count": 3, "active_band_mae_db": 0.9504944191589816}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.738651926696187, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4511fca9a4f2dab6d05f12791d0e91ec1c730a653911c7e51045cab8e24ac8cb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9863517121247811, "standard_deviation": 0.001402393934470404, "minimum": 0.9841457259416828, "maximum": 0.987854569175481, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7469063995887422, "standard_deviation": 0.023235521613197203, "minimum": 1.705351252791054, "maximum": 1.7702743220073118, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.632169365899476, "standard_deviation": 0.2043384175421284, "minimum": 0.4513341120148377, "maximum": 1.0701580083828686, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14653056688659033, "standard_deviation": 0.008327312444513223, "minimum": 0.13632824599006918, "maximum": 0.16111986537115075, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955545463907213, "standard_deviation": 0.0006621620634841247, "minimum": 0.9942091918224842, "maximum": 0.9962840980440731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9747335511112638, "standard_deviation": 0.02858498189147624, "minimum": 0.9289820720961607, "maximum": 1.0177708151403415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7068789466458534, "standard_deviation": 0.29177373591642936, "minimum": 0.4408312112708221, "maximum": 1.2686044567279549, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08906334935349747, "standard_deviation": 0.009260327882787335, "minimum": 0.07779844439084382, "maximum": 0.1022384141210075, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823092276204327, "standard_deviation": 0.0017185875275955397, "minimum": 0.979897540406932, "maximum": 0.9855571530843668, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6016211188577536, "standard_deviation": 0.032153797704593014, "minimum": 1.5508301783583922, "maximum": 1.6555227449908028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8731300451469541, "standard_deviation": 0.24449252640187855, "minimum": 0.6452407340390468, "maximum": 1.1875669062915055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1600823741372841, "standard_deviation": 0.008104213048547486, "minimum": 0.14447294069784447, "maximum": 0.1714315658304545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947542031011365, "standard_deviation": 0.0010031299006081785, "minimum": 0.9928296214224145, "maximum": 0.9960853038688342, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9510427976952396, "standard_deviation": 0.024422619786354308, "minimum": 0.9216309665913672, "maximum": 0.9994494564480156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2319749657058976, "standard_deviation": 0.37867528588163074, "minimum": 0.6019579329758744, "maximum": 1.6701001783442304, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08655654947978116, "standard_deviation": 0.00701385187724597, "minimum": 0.07802929722679572, "maximum": 0.10020616825075589, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925049612988308, "standard_deviation": 0.0011221916326069382, "minimum": 0.9907170759248345, "maximum": 0.9937552794803848, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.6210845963212233, "standard_deviation": 0.05615697613784084, "minimum": 2.548799404435182, "maximum": 2.7364766126570514, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.766810270943403, "standard_deviation": 0.4491365903300845, "minimum": 0.15533137257123905, "maximum": 1.4522426440787417, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10870685245945047, "standard_deviation": 0.007939508754488737, "minimum": 0.10002594393639817, "maximum": 0.12190903183105585, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965144333039623, "standard_deviation": 0.0008957521498157615, "minimum": 0.9949753815702435, "maximum": 0.9973054908195115, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.892349353790681, "standard_deviation": 0.050831292141589844, "minimum": 1.8327126890195788, "maximum": 1.97118870001957, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9348537920094051, "standard_deviation": 0.5426403716149504, "minimum": 0.20800587088384861, "maximum": 1.6786628411188342, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07375626021878467, "standard_deviation": 0.008749053757620879, "minimum": 0.06582080829036774, "maximum": 0.08787911123611479, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17960049385489135, "standard_deviation": 0.050530318133715425, "minimum": 0.0880084250110237, "maximum": 0.23898558732770891, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.809393322856693, "standard_deviation": 0.008450656926752984, "minimum": 5.797106651726494, "maximum": 5.8195670597298585, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5913190350309294, "standard_deviation": 0.015594098398515626, "minimum": 0.5695568320281544, "maximum": 0.6096301346505867, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0889543916365693, "standard_deviation": 0.02754596697991069, "minimum": -0.12136618831767587, "maximum": -0.04713190755440093, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899032677898416, "standard_deviation": 0.0006693484914060441, "minimum": 0.9889049896639731, "maximum": 0.9908808360902799, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6232330388909468, "standard_deviation": 0.06603579972599521, "minimum": 1.5678307688890136, "maximum": 1.7386459991246845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4238104517104728, "standard_deviation": 0.11832626920963651, "minimum": 0.20472186744466822, "maximum": 0.562365369068756, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0953965706786243, "standard_deviation": 0.003028027373416348, "minimum": 0.09133086171687665, "maximum": 0.10007381640990644, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971741607022596, "standard_deviation": 0.0002869292451868274, "minimum": 0.9968711802998993, "maximum": 0.9976914461281511, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7347883856403497, "standard_deviation": 0.03505023200582248, "minimum": 0.6853270855029945, "maximum": 0.7962525950949904, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42347538177528343, "standard_deviation": 0.15474779779869752, "minimum": 0.1756512257209967, "maximum": 0.6405837643642557, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0493775084517538, "standard_deviation": 0.002560345994583681, "minimum": 0.044663609458851304, "maximum": 0.05215405486905844, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0038", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.2475521564483643, "predicted_mos": 3.2475521564483643, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "4e7f905e5fccc9baf8947fe3ff2b8318a97fcccaa4253b71220ca53788615e18", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.606708699837327, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6245068631280555, "全帯域_10ms_rms_envelope_nrmse": 0.09242585549065854, "全帯域_10ms_envelope_correlation": 0.9904959300943577, "全帯域_10ms_modulation_mae_db": 0.48996261273271297, "全帯域_40ms_log_rms_mae_db": 0.720665900593459, "全帯域_40ms_rms_envelope_nrmse": 0.044787759285402885, "全帯域_40ms_envelope_correlation": 0.9976098352741553, "全帯域_40ms_modulation_mae_db": 0.3226923820600542, "300–1000 Hz_10ms_log_rms_mae_db": 3.223479360110898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10998475731716564, "300–1000 Hz_10ms_envelope_correlation": 0.9924715929540553, "300–1000 Hz_10ms_modulation_mae_db": 0.2696025753459224, "300–1000 Hz_40ms_log_rms_mae_db": 2.3202911345496275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07626183714831956, "300–1000 Hz_40ms_envelope_correlation": 0.9964017664205266, "300–1000 Hz_40ms_modulation_mae_db": 0.5815312382215801, "1000–3000 Hz_10ms_log_rms_mae_db": 2.269171204677822, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1589574210323212, "1000–3000 Hz_10ms_envelope_correlation": 0.9837418333570899, "1000–3000 Hz_10ms_modulation_mae_db": 1.1225989232807696, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2786116976718955, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09349525489662959, "1000–3000 Hz_40ms_envelope_correlation": 0.9948992973396292, "1000–3000 Hz_40ms_modulation_mae_db": 1.0211159729106207, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2709445066583114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1726518951178407, "3000–6000 Hz_10ms_envelope_correlation": 0.9793742524060632, "3000–6000 Hz_10ms_modulation_mae_db": 0.831849523173798, "3000–6000 Hz_40ms_log_rms_mae_db": 1.427585075572395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08578521190891131, "3000–6000 Hz_40ms_envelope_correlation": 0.9948371130919189, "3000–6000 Hz_40ms_modulation_mae_db": 1.2685655177689266, "mr_stft_db": 6.156959056943106, "band_mae_db": 0.1700448225665608, "rms_gain_error_db": -0.044037487928098, "output_peak": 0.5965393263075871}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6039580314726025, "全帯域_10ms_rms_envelope_nrmse": 0.09244009431042123, "全帯域_10ms_envelope_correlation": 0.9905026220279735, "全帯域_10ms_modulation_mae_db": 0.1913763511898853, "全帯域_40ms_log_rms_mae_db": 0.706581129292743, "全帯域_40ms_rms_envelope_nrmse": 0.046087333143756855, "全帯域_40ms_envelope_correlation": 0.997563345206125, "全帯域_40ms_modulation_mae_db": 0.20306881834904736, "300–1000 Hz_10ms_log_rms_mae_db": 3.087220160932823, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11622800451465577, "300–1000 Hz_10ms_envelope_correlation": 0.991385027148904, "300–1000 Hz_10ms_modulation_mae_db": 0.7785832517284668, "300–1000 Hz_40ms_log_rms_mae_db": 2.1915425523801613, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08469958325078279, "300–1000 Hz_40ms_envelope_correlation": 0.9953196751670919, "300–1000 Hz_40ms_modulation_mae_db": 1.096251661572676, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3081669073138444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1678105154752002, "1000–3000 Hz_10ms_envelope_correlation": 0.9819793714487447, "1000–3000 Hz_10ms_modulation_mae_db": 0.5647101042423468, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3645792937950503, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1127750334800626, "1000–3000 Hz_40ms_envelope_correlation": 0.9923009595440392, "1000–3000 Hz_40ms_modulation_mae_db": 0.6655227925586654, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2496074896928957, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18088506530033618, "3000–6000 Hz_10ms_envelope_correlation": 0.9781845534611934, "3000–6000 Hz_10ms_modulation_mae_db": 0.13200782787837081, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4383072633631577, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08834165176829763, "3000–6000 Hz_40ms_envelope_correlation": 0.9948361580497346, "3000–6000 Hz_40ms_modulation_mae_db": 0.7778722630050945, "mr_stft_db": 6.1323238442771135, "band_mae_db": 0.12542881313761803, "rms_gain_error_db": -0.1005292768682094, "output_peak": 0.5898611347721517}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.585859609842032, "全帯域_10ms_rms_envelope_nrmse": 0.10047987575702255, "全帯域_10ms_envelope_correlation": 0.988785775370415, "全帯域_10ms_modulation_mae_db": 0.31732980660254434, "全帯域_40ms_log_rms_mae_db": 0.6830354513900818, "全帯域_40ms_rms_envelope_nrmse": 0.05108983206873512, "全帯域_40ms_envelope_correlation": 0.9970299620048557, "全帯域_40ms_modulation_mae_db": 0.3777210045161105, "300–1000 Hz_10ms_log_rms_mae_db": 3.1220672651636026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10207711618840028, "300–1000 Hz_10ms_envelope_correlation": 0.9934306118878734, "300–1000 Hz_10ms_modulation_mae_db": 0.8245602234875274, "300–1000 Hz_40ms_log_rms_mae_db": 2.2400763981639384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06655602516583643, "300–1000 Hz_40ms_envelope_correlation": 0.9972027735482359, "300–1000 Hz_40ms_modulation_mae_db": 0.7831002594841404, "1000–3000 Hz_10ms_log_rms_mae_db": 2.279446558628823, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.153382699526772, "1000–3000 Hz_10ms_envelope_correlation": 0.9857035878097097, "1000–3000 Hz_10ms_modulation_mae_db": 0.6819997798605743, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2555077848626075, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0961621057950695, "1000–3000 Hz_40ms_envelope_correlation": 0.9955382154117587, "1000–3000 Hz_40ms_modulation_mae_db": 0.49037020440080487, "3000–6000 Hz_10ms_log_rms_mae_db": 2.217514514712245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15904587400447523, "3000–6000 Hz_10ms_envelope_correlation": 0.9833479335297318, "3000–6000 Hz_10ms_modulation_mae_db": 1.014549764313223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.435551865001774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09015853303436396, "3000–6000 Hz_40ms_envelope_correlation": 0.9954820472691235, "3000–6000 Hz_40ms_modulation_mae_db": 1.2972689301214066, "mr_stft_db": 6.149398942943662, "band_mae_db": 0.268018319266724, "rms_gain_error_db": -0.12094702191437422, "output_peak": 0.5826855022599576}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6413422517622478, "全帯域_10ms_rms_envelope_nrmse": 0.10218683630403036, "全帯域_10ms_envelope_correlation": 0.9885577909678595, "全帯域_10ms_modulation_mae_db": 0.08874649823781565, "全帯域_40ms_log_rms_mae_db": 0.8391734938675727, "全帯域_40ms_rms_envelope_nrmse": 0.059041057704793495, "全帯域_40ms_envelope_correlation": 0.9962009869051706, "全帯域_40ms_modulation_mae_db": 0.5018464064052083, "300–1000 Hz_10ms_log_rms_mae_db": 3.2489133764687903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10552354223407952, "300–1000 Hz_10ms_envelope_correlation": 0.9931906470110431, "300–1000 Hz_10ms_modulation_mae_db": 1.3903789466547025, "300–1000 Hz_40ms_log_rms_mae_db": 2.4166478195011107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06464056848307992, "300–1000 Hz_40ms_envelope_correlation": 0.9976291638913597, "300–1000 Hz_40ms_modulation_mae_db": 1.7294602196993418, "1000–3000 Hz_10ms_log_rms_mae_db": 2.311306951094884, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1556856763705516, "1000–3000 Hz_10ms_envelope_correlation": 0.9846257141284182, "1000–3000 Hz_10ms_modulation_mae_db": 1.0553372947591901, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3986571242979187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09379099388113205, "1000–3000 Hz_40ms_envelope_correlation": 0.9950414006914237, "1000–3000 Hz_40ms_modulation_mae_db": 1.1860735088053709, "3000–6000 Hz_10ms_log_rms_mae_db": 2.232938266471991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1733913156524791, "3000–6000 Hz_10ms_envelope_correlation": 0.9793139511602126, "3000–6000 Hz_10ms_modulation_mae_db": 1.229900681391219, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4026082519190306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09743278327292168, "3000–6000 Hz_40ms_envelope_correlation": 0.9932348017926448, "3000–6000 Hz_40ms_modulation_mae_db": 1.6227833554780442, "mr_stft_db": 6.154606854870463, "band_mae_db": 0.14051612629940596, "rms_gain_error_db": -0.17773293290471376, "output_peak": 0.5796761778896474}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.542762806979313, "全帯域_10ms_rms_envelope_nrmse": 0.09258190127985957, "全帯域_10ms_envelope_correlation": 0.9905193503522352, "全帯域_10ms_modulation_mae_db": 0.33809823096689723, "全帯域_40ms_log_rms_mae_db": 0.7371769599673283, "全帯域_40ms_rms_envelope_nrmse": 0.046580287016094894, "全帯域_40ms_envelope_correlation": 0.9975342590800818, "全帯域_40ms_modulation_mae_db": 0.20333178162619192, "300–1000 Hz_10ms_log_rms_mae_db": 3.2181699959922034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11527160681353472, "300–1000 Hz_10ms_envelope_correlation": 0.9915898604314753, "300–1000 Hz_10ms_modulation_mae_db": 1.0691714891522395, "300–1000 Hz_40ms_log_rms_mae_db": 2.3913965923091505, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07498350366895801, "300–1000 Hz_40ms_envelope_correlation": 0.9964145443411325, "300–1000 Hz_40ms_modulation_mae_db": 1.4998594962755707, "1000–3000 Hz_10ms_log_rms_mae_db": 2.293148235190288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15876381524162156, "1000–3000 Hz_10ms_envelope_correlation": 0.9837854352816383, "1000–3000 Hz_10ms_modulation_mae_db": 0.8790320472264934, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3409584939517278, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0813987614439392, "1000–3000 Hz_40ms_envelope_correlation": 0.996446624223307, "1000–3000 Hz_40ms_modulation_mae_db": 0.8899046483191887, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2584320126669404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16927270597384092, "3000–6000 Hz_10ms_envelope_correlation": 0.9802861038938364, "3000–6000 Hz_10ms_modulation_mae_db": 1.0016792111331936, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4313512399247579, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09160416234400721, "3000–6000 Hz_40ms_envelope_correlation": 0.994223429374119, "3000–6000 Hz_40ms_modulation_mae_db": 1.2106793476798927, "mr_stft_db": 6.137734121385222, "band_mae_db": 0.12222670876520329, "rms_gain_error_db": -0.10450046668076642, "output_peak": 0.607059308800692}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4949142756017528, "全帯域_10ms_rms_envelope_nrmse": 0.0898993553923883, "全帯域_10ms_envelope_correlation": 0.9909452717174151, "全帯域_10ms_modulation_mae_db": 0.6008320179909731, "全帯域_40ms_log_rms_mae_db": 0.6784033381688866, "全帯域_40ms_rms_envelope_nrmse": 0.04542763459658846, "全帯域_40ms_envelope_correlation": 0.9975098282778336, "全帯域_40ms_modulation_mae_db": 0.42198021086565346, "300–1000 Hz_10ms_log_rms_mae_db": 3.3152925972200986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10510541269343138, "300–1000 Hz_10ms_envelope_correlation": 0.9930420333313353, "300–1000 Hz_10ms_modulation_mae_db": 0.1299321585035928, "300–1000 Hz_40ms_log_rms_mae_db": 2.453453852309134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06514909525726298, "300–1000 Hz_40ms_envelope_correlation": 0.9973307127532008, "300–1000 Hz_40ms_modulation_mae_db": 0.6060969864278642, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2205233310466497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15364593585394687, "1000–3000 Hz_10ms_envelope_correlation": 0.9844860270130007, "1000–3000 Hz_10ms_modulation_mae_db": 0.84166652198659, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2707218157131241, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07691636355686135, "1000–3000 Hz_40ms_envelope_correlation": 0.9964362506162513, "1000–3000 Hz_40ms_modulation_mae_db": 0.927614180658166, "3000–6000 Hz_10ms_log_rms_mae_db": 2.199918464234935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15935159055385092, "3000–6000 Hz_10ms_envelope_correlation": 0.9830229650205086, "3000–6000 Hz_10ms_modulation_mae_db": 0.822775907543534, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3789798158783937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07848607690732684, "3000–6000 Hz_40ms_envelope_correlation": 0.9958757163217606, "3000–6000 Hz_40ms_modulation_mae_db": 1.178340744058017, "mr_stft_db": 6.141241058238652, "band_mae_db": 0.18599752871537492, "rms_gain_error_db": -0.0292441070714978, "output_peak": 0.5692199992142174}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.627792174613486, "全帯域_10ms_rms_envelope_nrmse": 0.09576641987019424, "全帯域_10ms_envelope_correlation": 0.9897804717731922, "全帯域_10ms_modulation_mae_db": 0.18995490895096243, "全帯域_40ms_log_rms_mae_db": 0.8542776862963277, "全帯域_40ms_rms_envelope_nrmse": 0.05238539260053629, "全帯域_40ms_envelope_correlation": 0.996832474328016, "全帯域_40ms_modulation_mae_db": 0.36704894250821624, "300–1000 Hz_10ms_log_rms_mae_db": 3.2288688630128863, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1083060845538419, "300–1000 Hz_10ms_envelope_correlation": 0.9926460110990944, "300–1000 Hz_10ms_modulation_mae_db": 1.0915998193386238, "300–1000 Hz_40ms_log_rms_mae_db": 2.3730475299065685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06624958783710179, "300–1000 Hz_40ms_envelope_correlation": 0.9973343718068058, "300–1000 Hz_40ms_modulation_mae_db": 1.3695090032417012, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2273521876291196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1522275983851065, "1000–3000 Hz_10ms_envelope_correlation": 0.9846402894403664, "1000–3000 Hz_10ms_modulation_mae_db": 1.1121913375545351, "1000–3000 Hz_40ms_log_rms_mae_db": 1.279757014269712, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07013200344691478, "1000–3000 Hz_40ms_envelope_correlation": 0.9968096362309986, "1000–3000 Hz_40ms_modulation_mae_db": 1.1074126794886459, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2794758173413565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19774815260693784, "3000–6000 Hz_10ms_envelope_correlation": 0.9731372292718269, "3000–6000 Hz_10ms_modulation_mae_db": 0.9425387486519526, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4324716140872542, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10890919719375672, "3000–6000 Hz_40ms_envelope_correlation": 0.9924236519734405, "3000–6000 Hz_40ms_modulation_mae_db": 1.2610092941228073, "mr_stft_db": 6.15216040029492, "band_mae_db": 0.11740574879845261, "rms_gain_error_db": -0.11075764679379153, "output_peak": 0.5693900141946979}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5947980802447208, "全帯域_10ms_rms_envelope_nrmse": 0.09967040863321118, "全帯域_10ms_envelope_correlation": 0.9889534584901737, "全帯域_10ms_modulation_mae_db": 0.6038806838538523, "全帯域_40ms_log_rms_mae_db": 0.7626815026716454, "全帯域_40ms_rms_envelope_nrmse": 0.04943047271125709, "全帯域_40ms_envelope_correlation": 0.9971535957588815, "全帯域_40ms_modulation_mae_db": 0.31342529934699154, "300–1000 Hz_10ms_log_rms_mae_db": 3.092981826059118, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1116165322072583, "300–1000 Hz_10ms_envelope_correlation": 0.9920830820800203, "300–1000 Hz_10ms_modulation_mae_db": 0.27134405823429364, "300–1000 Hz_40ms_log_rms_mae_db": 2.288800495987043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0676450398419245, "300–1000 Hz_40ms_envelope_correlation": 0.99714318260378, "300–1000 Hz_40ms_modulation_mae_db": 0.650678604827834, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2190848095710387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14317514885495286, "1000–3000 Hz_10ms_envelope_correlation": 0.9873666358527532, "1000–3000 Hz_10ms_modulation_mae_db": 1.189971149883843, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2846458172397766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08110904712118244, "1000–3000 Hz_40ms_envelope_correlation": 0.996836468981179, "1000–3000 Hz_40ms_modulation_mae_db": 1.1658700319912179, "3000–6000 Hz_10ms_log_rms_mae_db": 2.26232788945513, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17043212104391936, "3000–6000 Hz_10ms_envelope_correlation": 0.9805909152927736, "3000–6000 Hz_10ms_modulation_mae_db": 0.910125996378523, "3000–6000 Hz_40ms_log_rms_mae_db": 1.445914485773191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09681272737562216, "3000–6000 Hz_40ms_envelope_correlation": 0.9945322176369286, "3000–6000 Hz_40ms_modulation_mae_db": 0.9694647553085782, "mr_stft_db": 6.1417679079917775, "band_mae_db": 0.23988028596213415, "rms_gain_error_db": -0.09178001188886122, "output_peak": 0.6159233652510216}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88640, "output_samples": 88640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.433859468593011, "spectral_convergence": 0.3475372132965209, "spectral_convergence_status": "ok", "amplitude_floor": 1.5935254482600421e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.058410658628378, "spectral_convergence": 0.2990032310230916, "spectral_convergence_status": "ok", "amplitude_floor": 9.146897175620752e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.97860704360793, "spectral_convergence": 0.3122216740622351, "spectral_convergence_status": "ok", "amplitude_floor": 7.905477970735852e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007242342814613898, "output_power": 0.0007592895631924265, "delta_db": 0.20528352132305527, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001748392652036385, "output_power": 0.001715575337836763, "delta_db": -0.08229178179191084, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009820880914630636, "output_power": 0.0009619029458989246, "delta_db": -0.09019190140362016, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002665715938534463, "output_power": 0.0002511316942209758, "delta_db": -0.25912342112694253, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.016487980563797e-05, "output_power": 5.578522438527457e-05, "delta_db": -0.32823869142870465, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00015485832573279259, "output_power": 0.00015290461645051033, "delta_db": -0.05513961832513138, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1700448225665608, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.044037487928098, "pitch": null, "output_rms_dbfs": -24.09314340441791, "output_peak": 0.5965393263075871, "reference_rms_dbfs": -24.04910591648981, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.156959056943106}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00022148173042756648, "log_rms_mae_db": 1.6245068631280555, "rms_envelope_nrmse": 0.09242585549065854, "envelope_correlation": 0.9904959300943577, "rms_gain_error_db": -0.044059757218731274, "reference_mean_power": 0.0039398267431541775, "output_mean_power": 0.003900058739790613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.3094240146918987e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011389510990882513, "output_power": 0.0011391584800181835, "reference_active": true, "delta_db": 0.0007906941316290582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017715594495014008, "output_power": 0.00016519563296440965, "reference_active": true, "delta_db": -0.30357168520270883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5198627564616767e-05, "output_power": 3.295561598468399e-05, "reference_active": true, "delta_db": 1.1655254588638009}], "active_band_count": 3, "active_band_mae_db": 0.48996261273271297}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.00020322184022683357, "log_rms_mae_db": 0.720665900593459, "rms_envelope_nrmse": 0.044787759285402885, "envelope_correlation": 0.9976098352741553, "rms_gain_error_db": -0.044039643379940076, "reference_mean_power": 0.0039603691212238695, "output_mean_power": 0.0039204119231503335}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.1092095592585075e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010485738336262066, "output_power": 0.0010468746219212002, "reference_active": true, "delta_db": -0.007043441482598292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.703453851742181e-05, "output_power": 8.379651390910629e-05, "reference_active": true, "delta_db": -0.1646567935703497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.852514808001562e-06, "output_power": 2.2253553974989983e-06, "reference_active": true, "delta_db": 0.7963769111272144}], "active_band_count": 3, "active_band_mae_db": 0.3226923820600542}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00017913726163896074, "log_rms_mae_db": 3.223479360110898, "rms_envelope_nrmse": 0.10998475731716564, "envelope_correlation": 0.9924715929540553, "rms_gain_error_db": 0.012199214055319457, "reference_mean_power": 0.001348381907516537, "output_mean_power": 0.0013521748002491573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 1.724352265721046e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008753679492081999, "output_power": 0.000835753059280956, "reference_active": true, "delta_db": -0.20112666269406143}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012951638097036193, "output_power": 0.00011556406290989792, "reference_active": true, "delta_db": -0.49501898558018115}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6972503859652748e-05, "output_power": 1.7418554988294682e-05, "reference_active": true, "delta_db": 0.11266207776352474}], "active_band_count": 3, "active_band_mae_db": 0.2696025753459224}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.0001658599445391014, "log_rms_mae_db": 2.3202911345496275, "rms_envelope_nrmse": 0.07626183714831956, "envelope_correlation": 0.9964017664205266, "rms_gain_error_db": 0.01218624733140332, "reference_mean_power": 0.001355729112927364, "output_mean_power": 0.001359538613595061}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 1.6447728018068126e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008253872294632201, "output_power": 0.0007919542361508297, "reference_active": true, "delta_db": -0.17957658928879908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.001034181994634e-05, "output_power": 5.5749378535220595e-05, "reference_active": true, "delta_db": -0.31986070061542254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.081003459949387e-06, "output_power": 1.5622738370095184e-06, "reference_active": true, "delta_db": -1.245156424760519}], "active_band_count": 3, "active_band_mae_db": 0.5815312382215801}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00010266849739942118, "log_rms_mae_db": 2.269171204677822, "rms_envelope_nrmse": 0.1589574210323212, "envelope_correlation": 0.9837418333570899, "rms_gain_error_db": -0.33156069371420505, "reference_mean_power": 0.0002406311059478081, "output_mean_power": 0.0002229439532181039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.4743465614681157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013528556332001226, "output_power": 0.00012668624200554466, "reference_active": true, "delta_db": -0.2852200084429232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.422298628989197e-05, "output_power": 1.1027638670123889e-05, "reference_active": true, "delta_db": -1.1050826376281937}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7979674926628947e-06, "output_power": 1.77457047730246e-06, "reference_active": true, "delta_db": -1.9774941237711918}], "active_band_count": 3, "active_band_mae_db": 1.1225989232807696}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 8.271272308600344e-05, "log_rms_mae_db": 1.2786116976718955, "rms_envelope_nrmse": 0.09349525489662959, "envelope_correlation": 0.9948992973396292, "rms_gain_error_db": -0.33159201150594225, "reference_mean_power": 0.00024194203098044546, "output_mean_power": 0.00022415690464195123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.355449255956094e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012689841334701795, "output_power": 0.00012078421417435409, "reference_active": true, "delta_db": -0.2144601388229848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.520296796235256e-06, "output_power": 5.133453861362493e-06, "reference_active": true, "delta_db": -1.0385770174708522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.927693257279988e-07, "output_power": 1.2705941517360138e-07, "reference_active": true, "delta_db": -1.8103107624380255}], "active_band_count": 3, "active_band_mae_db": 1.0211159729106207}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 5.887039168463902e-05, "log_rms_mae_db": 2.2709445066583114, "rms_envelope_nrmse": 0.1726518951178407, "envelope_correlation": 0.9793742524060632, "rms_gain_error_db": -0.09105098350875804, "reference_mean_power": 0.00013131711251995087, "output_mean_power": 0.00012859267370390963}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 6.872487837717039e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.115138542486786e-05, "output_power": 2.766725999763872e-05, "reference_active": true, "delta_db": -0.5151121490698292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.223707279035506e-06, "output_power": 3.5375258711392827e-06, "reference_active": true, "delta_db": 0.4034402482801835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5460080528529549e-06, "output_power": 1.0752566434806633e-06, "reference_active": true, "delta_db": -1.5769961721713814}], "active_band_count": 3, "active_band_mae_db": 0.831849523173798}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 5.4692297362327305e-05, "log_rms_mae_db": 1.427585075572395, "rms_envelope_nrmse": 0.08578521190891131, "envelope_correlation": 0.9948371130919189, "rms_gain_error_db": -0.09106342795443077, "reference_mean_power": 0.00013203257062264164, "output_mean_power": 0.00012929291770116276}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 6.492145466124568e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.913804296383342e-05, "output_power": 2.581491304913903e-05, "reference_active": true, "delta_db": -0.5258971275297916}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5333654526802403e-06, "output_power": 1.6990466655626433e-06, "reference_active": true, "delta_db": 0.4455963304235838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.807699107498543e-08, "output_power": 4.586085636733663e-08, "reference_active": true, "delta_db": -2.8342030953534043}], "active_band_count": 3, "active_band_mae_db": 1.2685655177689266}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7160308366594427, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e264ba20530bb0406482bc9054d1e0adb0af2fc0e5357b9ca5cc24ce055e83a5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9845411117914651, "standard_deviation": 0.0015642146087172988, "minimum": 0.9819793714487447, "maximum": 0.9873666358527532, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2660250231440586, "standard_deviation": 0.038769413470293335, "minimum": 2.2190848095710387, "maximum": 2.311306951094884, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9309383948492927, "standard_deviation": 0.22643724231762172, "minimum": 0.5647101042423468, "maximum": 1.189971149883843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1554561013425591, "standard_deviation": 0.007017378260914021, "minimum": 0.14317514885495286, "maximum": 0.1678105154752002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955386066298233, "standard_deviation": 0.001515091346865301, "minimum": 0.9923009595440392, "maximum": 0.996836468981179, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3091798802252264, "standard_deviation": 0.05188849052116849, "minimum": 1.2555077848626075, "maximum": 1.3986571242979187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.931735502391585, "standard_deviation": 0.24663375893148282, "minimum": 0.49037020440080487, "maximum": 1.1860735088053709, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08822244545272395, "standard_deviation": 0.013506261591783027, "minimum": 0.07013200344691478, "maximum": 0.1127750334800626, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9796572380045183, "standard_deviation": 0.0031868887141867546, "minimum": 0.9731372292718269, "maximum": 0.9833479335297318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2463948701542256, "standard_deviation": 0.02746769522752568, "minimum": 2.199918464234935, "maximum": 2.2794758173413565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8606784575579767, "standard_deviation": 0.32121586106322175, "minimum": 0.13200782787837081, "maximum": 1.229900681391219, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17284734003171004, "standard_deviation": 0.012399404435395832, "minimum": 0.15904587400447523, "maximum": 0.19774815260693784, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944306419387088, "standard_deviation": 0.001136074211678672, "minimum": 0.9924236519734405, "maximum": 0.9958757163217606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4240962014399943, "standard_deviation": 0.0221726975733877, "minimum": 1.3789798158783937, "maximum": 1.445914485773191, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1982480259428459, "standard_deviation": 0.24742309528069187, "minimum": 0.7778722630050945, "maximum": 1.6227833554780442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09219129297565093, "standard_deviation": 0.009074978101440222, "minimum": 0.07848607690732684, "maximum": 0.10890919719375672, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9924798582429751, "standard_deviation": 0.0007465206566061654, "minimum": 0.991385027148904, "maximum": 0.9934306118878734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 3.192124180620053, "standard_deviation": 0.08207070670556774, "minimum": 3.087220160932823, "maximum": 3.3152925972200986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7281465653056711, "standard_deviation": 0.45912635110931727, "minimum": 0.1299321585035928, "maximum": 1.3903789466547025, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10926413206529595, "standard_deviation": 0.0049928001143287925, "minimum": 0.10207711618840028, "maximum": 0.11622800451465577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968470238165167, "standard_deviation": 0.0007573978562805731, "minimum": 0.9953196751670919, "maximum": 0.9976291638913597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.3344070468883418, "standard_deviation": 0.09044677465788939, "minimum": 2.1915425523801613, "maximum": 2.453453852309134, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0395609337188385, "standard_deviation": 0.4495393034470567, "minimum": 0.5815312382215801, "maximum": 1.7294602196993418, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07077315508165824, "standard_deviation": 0.0071620895314814135, "minimum": 0.06464056848307992, "maximum": 0.08469958325078279, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1711897941889342, "standard_deviation": 0.05687049881963501, "minimum": 0.11740574879845261, "maximum": 0.268018319266724, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.145774023368114, "standard_deviation": 0.008776541910529376, "minimum": 6.1323238442771135, "maximum": 6.156959056943106, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5887943535862465, "standard_deviation": 0.01694890395238856, "minimum": 0.5692199992142174, "maximum": 0.6159233652510216, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09744111900628905, "standard_deviation": 0.045936030449921485, "minimum": -0.17773293290471376, "maximum": -0.0292441070714978, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9898175838492027, "standard_deviation": 0.0009329970043333704, "minimum": 0.9885577909678595, "maximum": 0.9909452717174151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5894917617055264, "standard_deviation": 0.04893227224258774, "minimum": 1.4949142756017528, "maximum": 1.6413422517622478, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35252263881570545, "standard_deviation": 0.1953564526522927, "minimum": 0.08874649823781565, "maximum": 0.6038806838538523, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09568134337972325, "standard_deviation": 0.0045577101724316145, "minimum": 0.0898993553923883, "maximum": 0.10218683630403036, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.99717928585439, "standard_deviation": 0.00048821004482643053, "minimum": 0.9962009869051706, "maximum": 0.9976098352741553, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7477494327810055, "standard_deviation": 0.06704635793716948, "minimum": 0.6784033381688866, "maximum": 0.8542776862963277, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3388893557096842, "standard_deviation": 0.10246517474021977, "minimum": 0.20306881834904736, "maximum": 0.5018464064052083, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04935372114089563, "standard_deviation": 0.004791046472482588, "minimum": 0.044787759285402885, "maximum": 0.059041057704793495, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0038", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.1120524406433105, "predicted_mos": 3.1120524406433105, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "83ce445e9363638781417c0e97dec165bb6f5ed17d70920b7bb016a9d4d1284e", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.594779000151902, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.66888256255846, "全帯域_10ms_rms_envelope_nrmse": 0.09820275320455546, "全帯域_10ms_envelope_correlation": 0.9892574047392005, "全帯域_10ms_modulation_mae_db": 0.46040431710500923, "全帯域_40ms_log_rms_mae_db": 0.816202141185851, "全帯域_40ms_rms_envelope_nrmse": 0.05475383988201322, "全帯域_40ms_envelope_correlation": 0.9963875720757103, "全帯域_40ms_modulation_mae_db": 0.29458498063432353, "300–1000 Hz_10ms_log_rms_mae_db": 2.763999825256344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10196844388275265, "300–1000 Hz_10ms_envelope_correlation": 0.9934157235132571, "300–1000 Hz_10ms_modulation_mae_db": 0.486787506920274, "300–1000 Hz_40ms_log_rms_mae_db": 2.098783513966244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06315021146787328, "300–1000 Hz_40ms_envelope_correlation": 0.9975145496481376, "300–1000 Hz_40ms_modulation_mae_db": 0.6689183436174487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5888094332358456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1358219305543121, "1000–3000 Hz_10ms_envelope_correlation": 0.9877446538182904, "1000–3000 Hz_10ms_modulation_mae_db": 0.414237932410388, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9088613426950185, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07315825611126366, "1000–3000 Hz_40ms_envelope_correlation": 0.9962862334468722, "1000–3000 Hz_40ms_modulation_mae_db": 0.6087669550917671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.481676030702798, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15547602174090702, "3000–6000 Hz_10ms_envelope_correlation": 0.9835089187622175, "3000–6000 Hz_10ms_modulation_mae_db": 0.4080485894338893, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9229744657288691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0892336200833374, "3000–6000 Hz_40ms_envelope_correlation": 0.9948403709036931, "3000–6000 Hz_40ms_modulation_mae_db": 0.4326577940334703, "mr_stft_db": 5.7419005630919955, "band_mae_db": 0.14475552723894017, "rms_gain_error_db": -0.026921749999119435, "output_peak": 0.5821549008447178}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6541276044660536, "全帯域_10ms_rms_envelope_nrmse": 0.0990469024628959, "全帯域_10ms_envelope_correlation": 0.9890278754626416, "全帯域_10ms_modulation_mae_db": 0.743386007471745, "全帯域_40ms_log_rms_mae_db": 0.7517860030011323, "全帯域_40ms_rms_envelope_nrmse": 0.05027215185583397, "全帯域_40ms_envelope_correlation": 0.9970073463525435, "全帯域_40ms_modulation_mae_db": 0.38608254899928984, "300–1000 Hz_10ms_log_rms_mae_db": 2.6821737127038623, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1062937163888254, "300–1000 Hz_10ms_envelope_correlation": 0.9928807120524415, "300–1000 Hz_10ms_modulation_mae_db": 0.7951308896378578, "300–1000 Hz_40ms_log_rms_mae_db": 2.1242622404211566, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06937425355550125, "300–1000 Hz_40ms_envelope_correlation": 0.9970514641797967, "300–1000 Hz_40ms_modulation_mae_db": 1.1453437075238748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.629496166677292, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15108401795584192, "1000–3000 Hz_10ms_envelope_correlation": 0.9847523172745546, "1000–3000 Hz_10ms_modulation_mae_db": 0.7013538396401672, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9659960575920045, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0855626134924132, "1000–3000 Hz_40ms_envelope_correlation": 0.9950733323714516, "1000–3000 Hz_40ms_modulation_mae_db": 0.5934405355346748, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4991229866245084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1690017792316126, "3000–6000 Hz_10ms_envelope_correlation": 0.9804540540150901, "3000–6000 Hz_10ms_modulation_mae_db": 0.3839395734746649, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9667969844768431, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10079682511813892, "3000–6000 Hz_40ms_envelope_correlation": 0.9927405140801356, "3000–6000 Hz_40ms_modulation_mae_db": 0.653998228572816, "mr_stft_db": 5.745328453856222, "band_mae_db": 0.21782435170722123, "rms_gain_error_db": -0.07635978785619645, "output_peak": 0.5987847101276116}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.677687846198622, "全帯域_10ms_rms_envelope_nrmse": 0.10216429860485209, "全帯域_10ms_envelope_correlation": 0.9885440897550822, "全帯域_10ms_modulation_mae_db": 0.2880297083192592, "全帯域_40ms_log_rms_mae_db": 0.7583525446018046, "全帯域_40ms_rms_envelope_nrmse": 0.05609374025487586, "全帯域_40ms_envelope_correlation": 0.996454263671177, "全帯域_40ms_modulation_mae_db": 0.18990874761242058, "300–1000 Hz_10ms_log_rms_mae_db": 2.7080135503326446, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11269918022056477, "300–1000 Hz_10ms_envelope_correlation": 0.9917788291008479, "300–1000 Hz_10ms_modulation_mae_db": 0.5646018249777756, "300–1000 Hz_40ms_log_rms_mae_db": 2.034907123566676, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07207017102141684, "300–1000 Hz_40ms_envelope_correlation": 0.9965789497727411, "300–1000 Hz_40ms_modulation_mae_db": 0.5709386943842928, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6261575297643394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13581975114897632, "1000–3000 Hz_10ms_envelope_correlation": 0.9880079200361421, "1000–3000 Hz_10ms_modulation_mae_db": 0.4660670628868595, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9155846926192095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0808715818370017, "1000–3000 Hz_40ms_envelope_correlation": 0.9960193722449918, "1000–3000 Hz_40ms_modulation_mae_db": 0.6403792768145785, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5217140283566783, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1725982065747348, "3000–6000 Hz_10ms_envelope_correlation": 0.9792837585136277, "3000–6000 Hz_10ms_modulation_mae_db": 0.2364781259126911, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9363763556356419, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08678475700684425, "3000–6000 Hz_40ms_envelope_correlation": 0.9946171113332862, "3000–6000 Hz_40ms_modulation_mae_db": 0.5209369685280854, "mr_stft_db": 5.7448352600949875, "band_mae_db": 0.16329494419891047, "rms_gain_error_db": -0.1333135582760805, "output_peak": 0.57344714887453}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.663640297821593, "全帯域_10ms_rms_envelope_nrmse": 0.10259960037813885, "全帯域_10ms_envelope_correlation": 0.9883855213428107, "全帯域_10ms_modulation_mae_db": 0.9235514123421332, "全帯域_40ms_log_rms_mae_db": 0.7558966915179488, "全帯域_40ms_rms_envelope_nrmse": 0.05532762959874048, "全帯域_40ms_envelope_correlation": 0.9964443741332548, "全帯域_40ms_modulation_mae_db": 1.122728431554939, "300–1000 Hz_10ms_log_rms_mae_db": 2.800901724631694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11907422220902338, "300–1000 Hz_10ms_envelope_correlation": 0.9908711978949797, "300–1000 Hz_10ms_modulation_mae_db": 0.9931383917759798, "300–1000 Hz_40ms_log_rms_mae_db": 2.1321047253460166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08145925856860681, "300–1000 Hz_40ms_envelope_correlation": 0.9956583099436196, "300–1000 Hz_40ms_modulation_mae_db": 1.281323188422939, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5566826919558352, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13976344037327898, "1000–3000 Hz_10ms_envelope_correlation": 0.9874521209841997, "1000–3000 Hz_10ms_modulation_mae_db": 0.7192263611710334, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8581871922496402, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07684261088194186, "1000–3000 Hz_40ms_envelope_correlation": 0.9967357021421799, "1000–3000 Hz_40ms_modulation_mae_db": 0.7903179491918791, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4668677555102856, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1556186924971755, "3000–6000 Hz_10ms_envelope_correlation": 0.9833517340593836, "3000–6000 Hz_10ms_modulation_mae_db": 0.8463518428532458, "3000–6000 Hz_40ms_log_rms_mae_db": 0.898789664850592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08914341135690271, "3000–6000 Hz_40ms_envelope_correlation": 0.9946729431213495, "3000–6000 Hz_40ms_modulation_mae_db": 0.9357195543760497, "mr_stft_db": 5.74415455112486, "band_mae_db": 0.21787720155429371, "rms_gain_error_db": -0.09826325994998814, "output_peak": 0.5974603425301392}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6115222228859272, "全帯域_10ms_rms_envelope_nrmse": 0.09692933628051115, "全帯域_10ms_envelope_correlation": 0.9895092110268082, "全帯域_10ms_modulation_mae_db": 0.2984576708190835, "全帯域_40ms_log_rms_mae_db": 0.7819730823095704, "全帯域_40ms_rms_envelope_nrmse": 0.04711232530736276, "全帯域_40ms_envelope_correlation": 0.9974226515496025, "全帯域_40ms_modulation_mae_db": 0.2813866656214101, "300–1000 Hz_10ms_log_rms_mae_db": 2.8898254166137196, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1085628524079618, "300–1000 Hz_10ms_envelope_correlation": 0.9925517221038779, "300–1000 Hz_10ms_modulation_mae_db": 0.9018191506639647, "300–1000 Hz_40ms_log_rms_mae_db": 2.2746565112925485, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06997409057378377, "300–1000 Hz_40ms_envelope_correlation": 0.9969661149511184, "300–1000 Hz_40ms_modulation_mae_db": 1.1000194839808466, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6535341669701658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1380749803873007, "1000–3000 Hz_10ms_envelope_correlation": 0.9874284808783925, "1000–3000 Hz_10ms_modulation_mae_db": 0.29472415626297943, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9630309244232907, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07885853289828747, "1000–3000 Hz_40ms_envelope_correlation": 0.9960393222655183, "1000–3000 Hz_40ms_modulation_mae_db": 0.2443785951278968, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4528930489787026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16309803932431152, "3000–6000 Hz_10ms_envelope_correlation": 0.9815405478135295, "3000–6000 Hz_10ms_modulation_mae_db": 0.8329954536497729, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8998170599062685, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08592607965665289, "3000–6000 Hz_40ms_envelope_correlation": 0.994695535230537, "3000–6000 Hz_40ms_modulation_mae_db": 1.100219666378576, "mr_stft_db": 5.738710567218334, "band_mae_db": 0.18966403383384237, "rms_gain_error_db": -0.09144165477063128, "output_peak": 0.6388960216570276}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6202268709261745, "全帯域_10ms_rms_envelope_nrmse": 0.09767086388984948, "全帯域_10ms_envelope_correlation": 0.9893217668221493, "全帯域_10ms_modulation_mae_db": 0.6672168914408473, "全帯域_40ms_log_rms_mae_db": 0.7469504500650326, "全帯域_40ms_rms_envelope_nrmse": 0.051561111551381225, "全帯域_40ms_envelope_correlation": 0.9968980723232664, "全帯域_40ms_modulation_mae_db": 0.39533248761721734, "300–1000 Hz_10ms_log_rms_mae_db": 2.6933449587603158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09881904716791476, "300–1000 Hz_10ms_envelope_correlation": 0.9937936484296801, "300–1000 Hz_10ms_modulation_mae_db": 0.3782089973241902, "300–1000 Hz_40ms_log_rms_mae_db": 2.022520437518739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06321275877552372, "300–1000 Hz_40ms_envelope_correlation": 0.9974630465862891, "300–1000 Hz_40ms_modulation_mae_db": 0.4153701316091753, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6676628223295744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16151236905594335, "1000–3000 Hz_10ms_envelope_correlation": 0.9836321695566097, "1000–3000 Hz_10ms_modulation_mae_db": 0.7015254146831303, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0081987052916799, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10453996851437465, "1000–3000 Hz_40ms_envelope_correlation": 0.9940266628639883, "1000–3000 Hz_40ms_modulation_mae_db": 0.7851446762558346, "3000–6000 Hz_10ms_log_rms_mae_db": 1.464698342469134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15856477818778597, "3000–6000 Hz_10ms_envelope_correlation": 0.9828190670538403, "3000–6000 Hz_10ms_modulation_mae_db": 0.6390959632069191, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9407928592529672, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07789432681074769, "3000–6000 Hz_40ms_envelope_correlation": 0.9962534236884497, "3000–6000 Hz_40ms_modulation_mae_db": 0.5272472184335035, "mr_stft_db": 5.741343717643207, "band_mae_db": 0.24945318712905862, "rms_gain_error_db": -0.09181657580123873, "output_peak": 0.6030848920678934}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6681850352607346, "全帯域_10ms_rms_envelope_nrmse": 0.0986331390721238, "全帯域_10ms_envelope_correlation": 0.9891895645177593, "全帯域_10ms_modulation_mae_db": 0.8899150521769207, "全帯域_40ms_log_rms_mae_db": 0.750075035934618, "全帯域_40ms_rms_envelope_nrmse": 0.05035968715485568, "全帯域_40ms_envelope_correlation": 0.9969433457398331, "全帯域_40ms_modulation_mae_db": 0.9342055661681553, "300–1000 Hz_10ms_log_rms_mae_db": 2.890871272389366, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11947179487763367, "300–1000 Hz_10ms_envelope_correlation": 0.9910001128740382, "300–1000 Hz_10ms_modulation_mae_db": 0.5398449729421458, "300–1000 Hz_40ms_log_rms_mae_db": 2.2946393860416467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08467156568930334, "300–1000 Hz_40ms_envelope_correlation": 0.995477397663159, "300–1000 Hz_40ms_modulation_mae_db": 0.49839412338241024, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7115814441913766, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14479437107326196, "1000–3000 Hz_10ms_envelope_correlation": 0.988010999473393, "1000–3000 Hz_10ms_modulation_mae_db": 0.47898159881676855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.017946097116067, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10116510322195937, "1000–3000 Hz_40ms_envelope_correlation": 0.9954297213154097, "1000–3000 Hz_40ms_modulation_mae_db": 0.4300954477149032, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4445719455422643, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15458105414110881, "3000–6000 Hz_10ms_envelope_correlation": 0.983508341281588, "3000–6000 Hz_10ms_modulation_mae_db": 1.2374115965005599, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8550878952454967, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07950021095503972, "3000–6000 Hz_40ms_envelope_correlation": 0.9954294781646877, "3000–6000 Hz_40ms_modulation_mae_db": 1.205971848303448, "mr_stft_db": 5.747200945935571, "band_mae_db": 0.1663792253455692, "rms_gain_error_db": -0.029588441869096836, "output_peak": 0.606871452817919}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5771612565216915, "全帯域_10ms_rms_envelope_nrmse": 0.09213389197947187, "全帯域_10ms_envelope_correlation": 0.9905874145451794, "全帯域_10ms_modulation_mae_db": 0.074530975115679, "全帯域_40ms_log_rms_mae_db": 0.7765281500055446, "全帯域_40ms_rms_envelope_nrmse": 0.04759017412074845, "全帯域_40ms_envelope_correlation": 0.9974289032899994, "全帯域_40ms_modulation_mae_db": 0.25821086303080615, "300–1000 Hz_10ms_log_rms_mae_db": 2.8537605379730167, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11959801248079881, "300–1000 Hz_10ms_envelope_correlation": 0.9908984238467954, "300–1000 Hz_10ms_modulation_mae_db": 0.6994679942426482, "300–1000 Hz_40ms_log_rms_mae_db": 2.2696075690685085, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0875510248095998, "300–1000 Hz_40ms_envelope_correlation": 0.9950780602349724, "300–1000 Hz_40ms_modulation_mae_db": 0.9562715453812644, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6067619201647516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13670274906533786, "1000–3000 Hz_10ms_envelope_correlation": 0.9883013696857773, "1000–3000 Hz_10ms_modulation_mae_db": 0.5330172327996641, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8920574626610479, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08845413991955113, "1000–3000 Hz_40ms_envelope_correlation": 0.9956759236021608, "1000–3000 Hz_40ms_modulation_mae_db": 0.9079547381354797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4376215995597945, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1549326842208892, "3000–6000 Hz_10ms_envelope_correlation": 0.9835784039467959, "3000–6000 Hz_10ms_modulation_mae_db": 0.20165957508885912, "3000–6000 Hz_40ms_log_rms_mae_db": 0.902266680290772, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08737722218058605, "3000–6000 Hz_40ms_envelope_correlation": 0.9950380243749752, "3000–6000 Hz_40ms_modulation_mae_db": 0.47670428738883386, "mr_stft_db": 5.728145201833115, "band_mae_db": 0.1841503562464307, "rms_gain_error_db": -0.11323072908236594, "output_peak": 0.5989786641914129}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88640, "output_samples": 88640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.861807447333543, "spectral_convergence": 0.3407736171458889, "spectral_convergence_status": "ok", "amplitude_floor": 1.5935254482600421e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.545355782249414, "spectral_convergence": 0.2890483236660716, "spectral_convergence_status": "ok", "amplitude_floor": 9.146897175620752e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.818538459693028, "spectral_convergence": 0.3076595637164965, "spectral_convergence_status": "ok", "amplitude_floor": 7.905477970735852e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007242342814613898, "output_power": 0.0007657981116232862, "delta_db": 0.24235212893068322, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001748392652036385, "output_power": 0.0017175738874691837, "delta_db": -0.07723543778672358, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009820880914630636, "output_power": 0.0009598596653976861, "delta_db": -0.09942702519922215, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002665715938534463, "output_power": 0.0002593629832464528, "delta_db": -0.11905875793567797, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.016487980563797e-05, "output_power": 6.185230691520346e-05, "delta_db": 0.12012848790117411, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00015485832573279259, "output_power": 0.00014753716260362386, "delta_db": -0.21033132568016003, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14475552723894017, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.026921749999119435, "pitch": null, "output_rms_dbfs": -24.07602766648893, "output_peak": 0.5821549008447178, "reference_rms_dbfs": -24.04910591648981, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7419005630919955}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00022148173042756648, "log_rms_mae_db": 1.66888256255846, "rms_envelope_nrmse": 0.09820275320455546, "envelope_correlation": 0.9892574047392005, "rms_gain_error_db": -0.02694504060613603, "reference_mean_power": 0.0039398267431541775, "output_mean_power": 0.003915458450627487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.3094240146918987e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011389510990882513, "output_power": 0.001121909361339196, "reference_active": true, "delta_db": -0.06547306188471264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017715594495014008, "output_power": 0.00015803827169190296, "reference_active": true, "delta_db": -0.4959345926104475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5198627564616767e-05, "output_power": 3.0433886553711857e-05, "reference_active": true, "delta_db": 0.8198052968198677}], "active_band_count": 3, "active_band_mae_db": 0.46040431710500923}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.00020322184022683357, "log_rms_mae_db": 0.816202141185851, "rms_envelope_nrmse": 0.05475383988201322, "envelope_correlation": 0.9963875720757103, "rms_gain_error_db": -0.02668003129039776, "reference_mean_power": 0.0039603691212238695, "output_mean_power": 0.003936113948572516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.1092095592585075e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010485738336262066, "output_power": 0.0010325915750858078, "reference_active": true, "delta_db": -0.0667043870258431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.703453851742181e-05, "output_power": 8.044098768220776e-05, "reference_active": true, "delta_db": -0.34214236529638986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.852514808001562e-06, "output_power": 2.066581356599568e-06, "reference_active": true, "delta_db": 0.47490818958073777}], "active_band_count": 3, "active_band_mae_db": 0.29458498063432353}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00017913726163896074, "log_rms_mae_db": 2.763999825256344, "rms_envelope_nrmse": 0.10196844388275265, "envelope_correlation": 0.9934157235132571, "rms_gain_error_db": -0.010370995970341282, "reference_mean_power": 0.001348381907516537, "output_mean_power": 0.0013451657995178685}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 1.724352265721046e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008753679492081999, "output_power": 0.0008409557271764499, "reference_active": true, "delta_db": -0.1741750867501602}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012951638097036193, "output_power": 0.00010567185299629197, "reference_active": true, "delta_db": -0.8836537757066647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6972503859652748e-05, "output_power": 1.5470079669328473e-05, "reference_active": true, "delta_db": -0.4025336583039971}], "active_band_count": 3, "active_band_mae_db": 0.486787506920274}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.0001658599445391014, "log_rms_mae_db": 2.098783513966244, "rms_envelope_nrmse": 0.06315021146787328, "envelope_correlation": 0.9975145496481376, "rms_gain_error_db": -0.010376369434767902, "reference_mean_power": 0.001355729112927364, "output_mean_power": 0.0013524938072380006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 1.6447728018068126e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008253872294632201, "output_power": 0.0007944761774008289, "reference_active": true, "delta_db": -0.16576865820237197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.001034181994634e-05, "output_power": 5.058746993258582e-05, "reference_active": true, "delta_db": -0.7418314125170884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.081003459949387e-06, "output_power": 1.6156872814827598e-06, "reference_active": true, "delta_db": -1.0991549601328856}], "active_band_count": 3, "active_band_mae_db": 0.6689183436174487}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00010266849739942118, "log_rms_mae_db": 1.5888094332358456, "rms_envelope_nrmse": 0.1358219305543121, "envelope_correlation": 0.9877446538182904, "rms_gain_error_db": 0.017186765750558275, "reference_mean_power": 0.0002406311059478081, "output_mean_power": 0.00024158526601144147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.4743465614681157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013528556332001226, "output_power": 0.0001282908011799551, "reference_active": true, "delta_db": -0.230559370325763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.422298628989197e-05, "output_power": 1.240861682239884e-05, "reference_active": true, "delta_db": -0.5926741763209766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7979674926628947e-06, "output_power": 2.5403563359473775e-06, "reference_active": true, "delta_db": -0.4194802505844244}], "active_band_count": 3, "active_band_mae_db": 0.414237932410388}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 8.271272308600344e-05, "log_rms_mae_db": 0.9088613426950185, "rms_envelope_nrmse": 0.07315825611126366, "envelope_correlation": 0.9962862334468722, "rms_gain_error_db": 0.017178363024147852, "reference_mean_power": 0.00024194203098044546, "output_mean_power": 0.00024290091921135763}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.355449255956094e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012689841334701795, "output_power": 0.00012188983028424346, "reference_active": true, "delta_db": -0.17488719652755108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.520296796235256e-06, "output_power": 5.702550192044792e-06, "reference_active": true, "delta_db": -0.5819824826717015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.927693257279988e-07, "output_power": 1.5069360054902539e-07, "reference_active": true, "delta_db": -1.0694311860760488}], "active_band_count": 3, "active_band_mae_db": 0.6087669550917671}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 5.887039168463902e-05, "log_rms_mae_db": 1.481676030702798, "rms_envelope_nrmse": 0.15547602174090702, "envelope_correlation": 0.9835089187622175, "rms_gain_error_db": -0.2875582706243405, "reference_mean_power": 0.00013131711251995087, "output_mean_power": 0.00012290385284855702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 6.872487837717039e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.115138542486786e-05, "output_power": 2.9682416201325497e-05, "reference_active": true, "delta_db": -0.20978115824674762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.223707279035506e-06, "output_power": 3.0741219465178964e-06, "reference_active": true, "delta_db": -0.20634508443748778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5460080528529549e-06, "output_power": 1.8621458963601328e-06, "reference_active": true, "delta_db": 0.8080195256174325}], "active_band_count": 3, "active_band_mae_db": 0.4080485894338893}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 5.4692297362327305e-05, "log_rms_mae_db": 0.9229744657288691, "rms_envelope_nrmse": 0.0892336200833374, "envelope_correlation": 0.9948403709036931, "rms_gain_error_db": -0.28756368027535145, "reference_mean_power": 0.00013203257062264164, "output_mean_power": 0.00012357331886607477}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 6.492145466124568e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.913804296383342e-05, "output_power": 2.7838933105713248e-05, "reference_active": true, "delta_db": -0.19807791896499255}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5333654526802403e-06, "output_power": 1.4156221529796372e-06, "reference_active": true, "delta_db": -0.34698323941977327}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.807699107498543e-08, "output_power": 1.0474991808770293e-07, "reference_active": true, "delta_db": 0.7529122237156451}], "active_band_count": 3, "active_band_mae_db": 0.4326577940334703}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7542824154699507, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0b78d8602964f5a0773bce3a50da42c500898fa5bbe28ea722a7e36c12c024b0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.98691625396342, "standard_deviation": 0.0017325603139640758, "minimum": 0.9836321695566097, "maximum": 0.9883013696857773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6300857719111477, "standard_deviation": 0.04818852206512473, "minimum": 1.5566826919558352, "maximum": 1.7115814441913766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5386416998338739, "standard_deviation": 0.1555851950977756, "minimum": 0.29472415626297943, "maximum": 0.7192263611710334, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14294670120178166, "standard_deviation": 0.009170787130058233, "minimum": 0.13581975114897632, "maximum": 0.16151236905594335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9956607837815716, "standard_deviation": 0.0008355992926974701, "minimum": 0.9940266628639883, "maximum": 0.9967357021421799, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9412328093309947, "standard_deviation": 0.05666727830085252, "minimum": 0.8581871922496402, "maximum": 1.017946097116067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6250597717333768, "standard_deviation": 0.21293733785119226, "minimum": 0.2443785951278968, "maximum": 0.9079547381354797, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08618160085959914, "standard_deviation": 0.011374975376762188, "minimum": 0.07315825611126366, "maximum": 0.10453996851437465, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9822556031807591, "standard_deviation": 0.001647408931020717, "minimum": 0.9792837585136277, "maximum": 0.9835784039467959, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.471145717218021, "standard_deviation": 0.02847011088885231, "minimum": 1.4376215995597945, "maximum": 1.5217140283566783, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5982475900150752, "standard_deviation": 0.3579952675369662, "minimum": 0.20165957508885912, "maximum": 1.2374115965005599, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16048390698981568, "standard_deviation": 0.007012602104081216, "minimum": 0.15458105414110881, "maximum": 0.1725982065747348, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947859251121394, "standard_deviation": 0.0009904811693724352, "minimum": 0.9927405140801356, "maximum": 0.9962534236884497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9153627456734313, "standard_deviation": 0.03404321111333418, "minimum": 0.8550878952454967, "maximum": 0.9667969844768431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7316819457518478, "standard_deviation": 0.30450735894671777, "minimum": 0.4326577940334703, "maximum": 1.205971848303448, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0870820566460312, "standard_deviation": 0.0069732533797636055, "minimum": 0.07789432681074769, "maximum": 0.10079682511813892, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921487962269897, "standard_deviation": 0.0011747784752927694, "minimum": 0.9908711978949797, "maximum": 0.9937936484296801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.78536137483262, "standard_deviation": 0.08661799441507627, "minimum": 2.6821737127038623, "maximum": 2.890871272389366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6698749660606045, "standard_deviation": 0.21429033350706245, "minimum": 0.3782089973241902, "maximum": 0.9931383917759798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11081090870443441, "standard_deviation": 0.008207355344336907, "minimum": 0.09881904716791476, "maximum": 0.11959801248079881, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964734866224793, "standard_deviation": 0.0009452758768860491, "minimum": 0.9950780602349724, "maximum": 0.9975145496481376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.1564351884026918, "standard_deviation": 0.109242643298055, "minimum": 2.022520437518739, "maximum": 2.2946393860416467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8295724022877815, "standard_deviation": 0.33100027659167386, "minimum": 0.4153701316091753, "maximum": 1.281323188422939, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0739329168077011, "standard_deviation": 0.009475753464182037, "minimum": 0.06315021146787328, "maximum": 0.0875510248095998, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1916748534067833, "standard_deviation": 0.0346741316177276, "minimum": 0.14475552723894017, "maximum": 0.24945318712905862, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.741452407599787, "standard_deviation": 0.0059955551281184296, "minimum": 5.728145201833115, "maximum": 5.747200945935571, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5999597666389065, "standard_deviation": 0.019280235464559196, "minimum": 0.57344714887453, "maximum": 0.6388960216570276, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08261696970058965, "standard_deviation": 0.03754577237080679, "minimum": -0.1333135582760805, "maximum": -0.026921749999119435, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9892278560264539, "standard_deviation": 0.0006714350159717897, "minimum": 0.9883855213428107, "maximum": 0.9905874145451794, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6426792120799072, "standard_deviation": 0.035658435103942884, "minimum": 1.5771612565216915, "maximum": 1.677687846198622, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5431865043488346, "standard_deviation": 0.30980399684112964, "minimum": 0.074530975115679, "maximum": 0.9235514123421332, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09842259823404982, "standard_deviation": 0.0032573618986346206, "minimum": 0.09213389197947187, "maximum": 0.10259960037813885, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968733161419234, "standard_deviation": 0.00041909274418418685, "minimum": 0.9963875720757103, "maximum": 0.9974289032899994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7672205123276878, "standard_deviation": 0.023462453404044136, "minimum": 0.7469504500650326, "maximum": 0.816202141185851, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4828050364048202, "standard_deviation": 0.3469248589860834, "minimum": 0.18990874761242058, "maximum": 1.122728431554939, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05163383246572645, "standard_deviation": 0.003454757992501301, "minimum": 0.04711232530736276, "maximum": 0.05609374025487586, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0038", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.2113091945648193, "predicted_mos": 3.2113091945648193, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "99285af7225f8a0e1662a21a94ae3839cbeee22116ed69c386c4df24561d607f", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.971473400015384, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.602713770590481, "全帯域_10ms_rms_envelope_nrmse": 0.09573032778398897, "全帯域_10ms_envelope_correlation": 0.9898410096261309, "全帯域_10ms_modulation_mae_db": 0.5854751836291481, "全帯域_40ms_log_rms_mae_db": 0.7156073305175713, "全帯域_40ms_rms_envelope_nrmse": 0.0457469874181413, "全帯域_40ms_envelope_correlation": 0.9976097387418481, "全帯域_40ms_modulation_mae_db": 0.4319650763111033, "300–1000 Hz_10ms_log_rms_mae_db": 3.131175358841015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09926146515393007, "300–1000 Hz_10ms_envelope_correlation": 0.9939020780747121, "300–1000 Hz_10ms_modulation_mae_db": 0.6317878328815522, "300–1000 Hz_40ms_log_rms_mae_db": 2.377995718330487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06604857156352563, "300–1000 Hz_40ms_envelope_correlation": 0.9973561252959736, "300–1000 Hz_40ms_modulation_mae_db": 0.8783484381207874, "1000–3000 Hz_10ms_log_rms_mae_db": 1.943192473376617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14998223068252528, "1000–3000 Hz_10ms_envelope_correlation": 0.9860944115639545, "1000–3000 Hz_10ms_modulation_mae_db": 0.39450336175479217, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2277830945337622, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10433577724579605, "1000–3000 Hz_40ms_envelope_correlation": 0.9939226374130807, "1000–3000 Hz_40ms_modulation_mae_db": 0.491239346184926, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9179339279161227, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14254757917428304, "3000–6000 Hz_10ms_envelope_correlation": 0.9859628236072165, "3000–6000 Hz_10ms_modulation_mae_db": 0.5164993244754866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2403050005370895, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07624092020680073, "3000–6000 Hz_40ms_envelope_correlation": 0.9958362875346748, "3000–6000 Hz_40ms_modulation_mae_db": 0.7966952600247215, "mr_stft_db": 5.9253730119606205, "band_mae_db": 0.1711563251691882, "rms_gain_error_db": -0.10187892605144636, "output_peak": 0.5688475769844377}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.550911563788756, "全帯域_10ms_rms_envelope_nrmse": 0.09269261532447179, "全帯域_10ms_envelope_correlation": 0.9904534591846839, "全帯域_10ms_modulation_mae_db": 0.2065701606379954, "全帯域_40ms_log_rms_mae_db": 0.6764552362300981, "全帯域_40ms_rms_envelope_nrmse": 0.0509690673638621, "全帯域_40ms_envelope_correlation": 0.997042762346886, "全帯域_40ms_modulation_mae_db": 0.28652656581348973, "300–1000 Hz_10ms_log_rms_mae_db": 2.995235362199529, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11697937879998183, "300–1000 Hz_10ms_envelope_correlation": 0.9912823353091569, "300–1000 Hz_10ms_modulation_mae_db": 0.33435726554959544, "300–1000 Hz_40ms_log_rms_mae_db": 2.267067688979899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08995009534395085, "300–1000 Hz_40ms_envelope_correlation": 0.9947476545733898, "300–1000 Hz_40ms_modulation_mae_db": 0.39011721321527704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9877011331402736, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15111407101535632, "1000–3000 Hz_10ms_envelope_correlation": 0.9858236854314304, "1000–3000 Hz_10ms_modulation_mae_db": 0.6259117249471163, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2666451007219144, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0926961523865851, "1000–3000 Hz_40ms_envelope_correlation": 0.9956333407278894, "1000–3000 Hz_40ms_modulation_mae_db": 0.8307028841308671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8821729633038842, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16623292340826618, "3000–6000 Hz_10ms_envelope_correlation": 0.9816446629906033, "3000–6000 Hz_10ms_modulation_mae_db": 0.7236897516433763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.260267054861189, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09416104725964132, "3000–6000 Hz_40ms_envelope_correlation": 0.9940629148790965, "3000–6000 Hz_40ms_modulation_mae_db": 1.2112585198454624, "mr_stft_db": 5.931357154537341, "band_mae_db": 0.18143651005327613, "rms_gain_error_db": -0.1189511329269749, "output_peak": 0.6071144750356546}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5719265410978023, "全帯域_10ms_rms_envelope_nrmse": 0.09732675256024566, "全帯域_10ms_envelope_correlation": 0.9894603866990154, "全帯域_10ms_modulation_mae_db": 0.45979804642705097, "全帯域_40ms_log_rms_mae_db": 0.7279414962174257, "全帯域_40ms_rms_envelope_nrmse": 0.049617693456706906, "全帯域_40ms_envelope_correlation": 0.997164156825765, "全帯域_40ms_modulation_mae_db": 0.5540804326045573, "300–1000 Hz_10ms_log_rms_mae_db": 2.9344421506265412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10675008931631624, "300–1000 Hz_10ms_envelope_correlation": 0.9927911386035814, "300–1000 Hz_10ms_modulation_mae_db": 0.38210603046765623, "300–1000 Hz_40ms_log_rms_mae_db": 2.173419571175802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07013842477004982, "300–1000 Hz_40ms_envelope_correlation": 0.9969374059370829, "300–1000 Hz_40ms_modulation_mae_db": 0.3672000331340839, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0117317660547305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15229001410564663, "1000–3000 Hz_10ms_envelope_correlation": 0.9846261118712465, "1000–3000 Hz_10ms_modulation_mae_db": 0.35348248886334327, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2834799826910215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08206922249084887, "1000–3000 Hz_40ms_envelope_correlation": 0.9956472220353245, "1000–3000 Hz_40ms_modulation_mae_db": 0.4008679050272566, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8917234555997133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15561485846085454, "3000–6000 Hz_10ms_envelope_correlation": 0.9832602295955377, "3000–6000 Hz_10ms_modulation_mae_db": 0.8095213967649055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2537531096419985, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08345266515745622, "3000–6000 Hz_40ms_envelope_correlation": 0.9949771951466477, "3000–6000 Hz_40ms_modulation_mae_db": 1.0259195649693935, "mr_stft_db": 5.932489368355184, "band_mae_db": 0.174938612219103, "rms_gain_error_db": -0.10648867572918133, "output_peak": 0.6033610061419934}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5649846582844815, "全帯域_10ms_rms_envelope_nrmse": 0.09342486691147543, "全帯域_10ms_envelope_correlation": 0.9903601573137215, "全帯域_10ms_modulation_mae_db": 0.49593739859191305, "全帯域_40ms_log_rms_mae_db": 0.7241700442980413, "全帯域_40ms_rms_envelope_nrmse": 0.04450549034415261, "全帯域_40ms_envelope_correlation": 0.997712802458249, "全帯域_40ms_modulation_mae_db": 0.4321311066993014, "300–1000 Hz_10ms_log_rms_mae_db": 3.012660328224219, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10174724380029795, "300–1000 Hz_10ms_envelope_correlation": 0.9935544492582223, "300–1000 Hz_10ms_modulation_mae_db": 0.21607846034647885, "300–1000 Hz_40ms_log_rms_mae_db": 2.357209381518723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06661892703242694, "300–1000 Hz_40ms_envelope_correlation": 0.9973088946211067, "300–1000 Hz_40ms_modulation_mae_db": 0.1016489141745125, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9187943611082743, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14070828186261872, "1000–3000 Hz_10ms_envelope_correlation": 0.986852624563892, "1000–3000 Hz_10ms_modulation_mae_db": 0.3844568469531353, "1000–3000 Hz_40ms_log_rms_mae_db": 1.209355717483817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07852485409197221, "1000–3000 Hz_40ms_envelope_correlation": 0.9958250854216434, "1000–3000 Hz_40ms_modulation_mae_db": 0.5721260524023322, "3000–6000 Hz_10ms_log_rms_mae_db": 1.856655447475851, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16981752677948148, "3000–6000 Hz_10ms_envelope_correlation": 0.98010710232753, "3000–6000 Hz_10ms_modulation_mae_db": 0.3844475320227523, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2562380262761392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09930252111574571, "3000–6000 Hz_40ms_envelope_correlation": 0.9928774743333394, "3000–6000 Hz_40ms_modulation_mae_db": 1.1246051695722874, "mr_stft_db": 5.919610607149708, "band_mae_db": 0.08144840494497803, "rms_gain_error_db": -0.07957764556211239, "output_peak": 0.5825444416219279}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.53447229720015, "全帯域_10ms_rms_envelope_nrmse": 0.09608670215362129, "全帯域_10ms_envelope_correlation": 0.9896967040664304, "全帯域_10ms_modulation_mae_db": 0.43776159020035843, "全帯域_40ms_log_rms_mae_db": 0.7017911270538096, "全帯域_40ms_rms_envelope_nrmse": 0.04884451330756624, "全帯域_40ms_envelope_correlation": 0.9971965169023067, "全帯域_40ms_modulation_mae_db": 0.2669931616776697, "300–1000 Hz_10ms_log_rms_mae_db": 2.9751678770332166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10803300911614946, "300–1000 Hz_10ms_envelope_correlation": 0.9927184568718972, "300–1000 Hz_10ms_modulation_mae_db": 1.3501358275422224, "300–1000 Hz_40ms_log_rms_mae_db": 2.3434874789757862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07370418805898558, "300–1000 Hz_40ms_envelope_correlation": 0.996672526005423, "300–1000 Hz_40ms_modulation_mae_db": 1.5343635724393743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9735353591623814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13973258434843375, "1000–3000 Hz_10ms_envelope_correlation": 0.9871693543066967, "1000–3000 Hz_10ms_modulation_mae_db": 0.5912969885724682, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2306164955055041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0783923029388846, "1000–3000 Hz_40ms_envelope_correlation": 0.996102666042836, "1000–3000 Hz_40ms_modulation_mae_db": 0.3868311765609578, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8405726402737717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16248027186464306, "3000–6000 Hz_10ms_envelope_correlation": 0.9817143917386355, "3000–6000 Hz_10ms_modulation_mae_db": 0.9476829753410456, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2221530660694782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07590233440087206, "3000–6000 Hz_40ms_envelope_correlation": 0.9960567465965089, "3000–6000 Hz_40ms_modulation_mae_db": 1.2295772666814975, "mr_stft_db": 5.919168389163698, "band_mae_db": 0.17587913893059523, "rms_gain_error_db": -0.08391898324580604, "output_peak": 0.5825221010844005}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5048091732357671, "全帯域_10ms_rms_envelope_nrmse": 0.09086457649606401, "全帯域_10ms_envelope_correlation": 0.9908212771897389, "全帯域_10ms_modulation_mae_db": 0.4526566331172979, "全帯域_40ms_log_rms_mae_db": 0.6613995002428541, "全帯域_40ms_rms_envelope_nrmse": 0.047984747680208974, "全帯域_40ms_envelope_correlation": 0.9974029952265604, "全帯域_40ms_modulation_mae_db": 0.789409267952229, "300–1000 Hz_10ms_log_rms_mae_db": 2.920548164223229, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10867642371081508, "300–1000 Hz_10ms_envelope_correlation": 0.9925565869466839, "300–1000 Hz_10ms_modulation_mae_db": 1.0965154660377303, "300–1000 Hz_40ms_log_rms_mae_db": 2.1752193493964036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07163875952056968, "300–1000 Hz_40ms_envelope_correlation": 0.996796261264361, "300–1000 Hz_40ms_modulation_mae_db": 1.299225025745595, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9736220559546809, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15932957171106082, "1000–3000 Hz_10ms_envelope_correlation": 0.9842512788512978, "1000–3000 Hz_10ms_modulation_mae_db": 1.0472486203509295, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2221293023984696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09693716235789031, "1000–3000 Hz_40ms_envelope_correlation": 0.9953008218292296, "1000–3000 Hz_40ms_modulation_mae_db": 1.1755378486738814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.911914438579178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.174152039263515, "3000–6000 Hz_10ms_envelope_correlation": 0.9798530958004774, "3000–6000 Hz_10ms_modulation_mae_db": 0.6472885982045363, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2415485070859402, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0878335998837767, "3000–6000 Hz_40ms_envelope_correlation": 0.994901487056735, "3000–6000 Hz_40ms_modulation_mae_db": 0.7611551753365123, "mr_stft_db": 5.913843465151494, "band_mae_db": 0.2130375226402934, "rms_gain_error_db": -0.11609635934329615, "output_peak": 0.5954708497392776}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.563185470327513, "全帯域_10ms_rms_envelope_nrmse": 0.0903836529426479, "全帯域_10ms_envelope_correlation": 0.9910598576651181, "全帯域_10ms_modulation_mae_db": 0.5504562830628345, "全帯域_40ms_log_rms_mae_db": 0.6906423373950643, "全帯域_40ms_rms_envelope_nrmse": 0.04932547366804975, "全帯域_40ms_envelope_correlation": 0.9971908031000185, "全帯域_40ms_modulation_mae_db": 0.4357849852958962, "300–1000 Hz_10ms_log_rms_mae_db": 3.046247997868197, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1222582905444646, "300–1000 Hz_10ms_envelope_correlation": 0.9907547970484885, "300–1000 Hz_10ms_modulation_mae_db": 0.5735619372859825, "300–1000 Hz_40ms_log_rms_mae_db": 2.3719155092364335, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08698439970400672, "300–1000 Hz_40ms_envelope_correlation": 0.9953080215767729, "300–1000 Hz_40ms_modulation_mae_db": 0.9169405827788125, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9909464551511669, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14560148369127532, "1000–3000 Hz_10ms_envelope_correlation": 0.9865002335172358, "1000–3000 Hz_10ms_modulation_mae_db": 0.6121531763267921, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2727951808462854, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09341754347710948, "1000–3000 Hz_40ms_envelope_correlation": 0.994930038377855, "1000–3000 Hz_40ms_modulation_mae_db": 0.34403460088491805, "3000–6000 Hz_10ms_log_rms_mae_db": 1.816080307781199, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15950254075434073, "3000–6000 Hz_10ms_envelope_correlation": 0.982393278669551, "3000–6000 Hz_10ms_modulation_mae_db": 0.9628694008814177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.208193048853508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08568308965318627, "3000–6000 Hz_40ms_envelope_correlation": 0.9948726807036706, "3000–6000 Hz_40ms_modulation_mae_db": 1.4003658879220893, "mr_stft_db": 5.936456179802576, "band_mae_db": 0.15566638415808778, "rms_gain_error_db": -0.04677215889342113, "output_peak": 0.5877959855194497}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6559353898728415, "全帯域_10ms_rms_envelope_nrmse": 0.09517465229797008, "全帯域_10ms_envelope_correlation": 0.9898741683430342, "全帯域_10ms_modulation_mae_db": 0.25933455004375117, "全帯域_40ms_log_rms_mae_db": 0.7739440697570211, "全帯域_40ms_rms_envelope_nrmse": 0.049373379530331614, "全帯域_40ms_envelope_correlation": 0.9970866198515351, "全帯域_40ms_modulation_mae_db": 0.5266414480848458, "300–1000 Hz_10ms_log_rms_mae_db": 2.9118085971816763, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09896047431648992, "300–1000 Hz_10ms_envelope_correlation": 0.9939078159246497, "300–1000 Hz_10ms_modulation_mae_db": 1.0194681094254094, "300–1000 Hz_40ms_log_rms_mae_db": 2.2406323760226776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06821546351760426, "300–1000 Hz_40ms_envelope_correlation": 0.997092026337881, "300–1000 Hz_40ms_modulation_mae_db": 1.325716218655452, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0346377243919758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14646053251779187, "1000–3000 Hz_10ms_envelope_correlation": 0.986507564873034, "1000–3000 Hz_10ms_modulation_mae_db": 0.45054986417868426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2971811350867846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08996563789385445, "1000–3000 Hz_40ms_envelope_correlation": 0.9956282973696888, "1000–3000 Hz_40ms_modulation_mae_db": 0.4469667307115286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8855585468106297, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16131232977784515, "3000–6000 Hz_10ms_envelope_correlation": 0.982219034618503, "3000–6000 Hz_10ms_modulation_mae_db": 0.5277113182547696, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2797107142918975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08646963582710358, "3000–6000 Hz_40ms_envelope_correlation": 0.9947185129681445, "3000–6000 Hz_40ms_modulation_mae_db": 0.48832547016803796, "mr_stft_db": 5.9166498167250365, "band_mae_db": 0.10604573097278762, "rms_gain_error_db": -0.04742715367335153, "output_peak": 0.5762734511897383}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88640, "output_samples": 88640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.135142052281203, "spectral_convergence": 0.340578225140041, "spectral_convergence_status": "ok", "amplitude_floor": 1.5935254482600421e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.751448653973347, "spectral_convergence": 0.29243375475178013, "spectral_convergence_status": "ok", "amplitude_floor": 9.146897175620752e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.889528329627312, "spectral_convergence": 0.3069148462374022, "spectral_convergence_status": "ok", "amplitude_floor": 7.905477970735852e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007242342814613898, "output_power": 0.0007109551986426736, "delta_db": -0.08036843985556355, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001748392652036385, "output_power": 0.001712054524640596, "delta_db": -0.09121380862699065, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009820880914630636, "output_power": 0.0009659361045159967, "delta_db": -0.07202045615712979, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002665715938534463, "output_power": 0.00023858988757476514, "delta_db": -0.481618361414676, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.016487980563797e-05, "output_power": 6.39514658777557e-05, "delta_db": 0.26507450279786154, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00015485832573279259, "output_power": 0.0001535572483806163, "delta_db": -0.036642382162907626, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1711563251691882, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10187892605144636, "pitch": null, "output_rms_dbfs": -24.150984842541256, "output_peak": 0.5688475769844377, "reference_rms_dbfs": -24.04910591648981, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9253730119606205}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00022148173042756648, "log_rms_mae_db": 1.602713770590481, "rms_envelope_nrmse": 0.09573032778398897, "envelope_correlation": 0.9898410096261309, "rms_gain_error_db": -0.10199220399640412, "reference_mean_power": 0.0039398267431541775, "output_mean_power": 0.0038483795964795236}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.3094240146918987e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011389510990882513, "output_power": 0.0011177763635251917, "reference_active": true, "delta_db": -0.08150156232881986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017715594495014008, "output_power": 0.00017008465406593564, "reference_active": true, "delta_db": -0.1769059980498917}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5198627564616767e-05, "output_power": 3.557776701495241e-05, "reference_active": true, "delta_db": 1.4980179905087327}], "active_band_count": 3, "active_band_mae_db": 0.5854751836291481}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.00020322184022683357, "log_rms_mae_db": 0.7156073305175713, "rms_envelope_nrmse": 0.0457469874181413, "envelope_correlation": 0.9976097387418481, "rms_gain_error_db": -0.10235837198660071, "reference_mean_power": 0.0039603691212238695, "output_mean_power": 0.0038681190186966845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.1092095592585075e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010485738336262066, "output_power": 0.001031843805638221, "reference_active": true, "delta_db": -0.06985054662044664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.703453851742181e-05, "output_power": 8.062915936692569e-05, "reference_active": true, "delta_db": -0.33199499000686805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.852514808001562e-06, "output_power": 2.27597065072576e-06, "reference_active": true, "delta_db": 0.8940496923059951}], "active_band_count": 3, "active_band_mae_db": 0.4319650763111033}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00017913726163896074, "log_rms_mae_db": 3.131175358841015, "rms_envelope_nrmse": 0.09926146515393007, "envelope_correlation": 0.9939020780747121, "rms_gain_error_db": 0.020819221215717415, "reference_mean_power": 0.001348381907516537, "output_mean_power": 0.0013548613025997143}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 1.724352265721046e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008753679492081999, "output_power": 0.0008439577438341448, "reference_active": true, "delta_db": -0.15869938819312976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012951638097036193, "output_power": 0.00011664351777869399, "reference_active": true, "delta_db": -0.45464091794101297}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6972503859652748e-05, "output_power": 1.2634079252429212e-05, "reference_active": true, "delta_db": -1.2820231925105139}], "active_band_count": 3, "active_band_mae_db": 0.6317878328815522}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.0001658599445391014, "log_rms_mae_db": 2.377995718330487, "rms_envelope_nrmse": 0.06604857156352563, "envelope_correlation": 0.9973561252959736, "rms_gain_error_db": 0.02080539300896418, "reference_mean_power": 0.001355729112927364, "output_mean_power": 0.0013622394761589477}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 1.6447728018068126e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008253872294632201, "output_power": 0.0007991719079343024, "reference_active": true, "delta_db": -0.14017535714886142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.001034181994634e-05, "output_power": 5.446269807860914e-05, "reference_active": true, "delta_db": -0.4212694806405176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.081003459949387e-06, "output_power": 1.2909599067653045e-06, "reference_active": true, "delta_db": -2.073600476572983}], "active_band_count": 3, "active_band_mae_db": 0.8783484381207874}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00010266849739942118, "log_rms_mae_db": 1.943192473376617, "rms_envelope_nrmse": 0.14998223068252528, "envelope_correlation": 0.9860944115639545, "rms_gain_error_db": -0.3841278273803306, "reference_mean_power": 0.0002406311059478081, "output_mean_power": 0.00022026169874810096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.4743465614681157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013528556332001226, "output_power": 0.00011823626151932841, "reference_active": true, "delta_db": -0.5850076493135151}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.422298628989197e-05, "output_power": 1.4202741391294916e-05, "reference_active": true, "delta_db": -0.006186120779441482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7979674926628947e-06, "output_power": 2.4412431233504364e-06, "reference_active": true, "delta_db": -0.5923163151714198}], "active_band_count": 3, "active_band_mae_db": 0.39450336175479217}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 8.271272308600344e-05, "log_rms_mae_db": 1.2277830945337622, "rms_envelope_nrmse": 0.10433577724579605, "envelope_correlation": 0.9939226374130807, "rms_gain_error_db": -0.3841478966591308, "reference_mean_power": 0.00024194203098044546, "output_mean_power": 0.00022146063066444318}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.355449255956094e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012689841334701795, "output_power": 0.00011232919923818201, "reference_active": true, "delta_db": -0.5296352906331093}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.520296796235256e-06, "output_power": 6.741447396921649e-06, "reference_active": true, "delta_db": 0.14485785342333252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.927693257279988e-07, "output_power": 1.6036716023725633e-07, "reference_active": true, "delta_db": -0.7992248944983359}], "active_band_count": 3, "active_band_mae_db": 0.491239346184926}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 5.887039168463902e-05, "log_rms_mae_db": 1.9179339279161227, "rms_envelope_nrmse": 0.14254757917428304, "envelope_correlation": 0.9859628236072165, "rms_gain_error_db": -0.051731229948140584, "reference_mean_power": 0.00013131711251995087, "output_mean_power": 0.00012976220049022597}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 6.872487837717039e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.115138542486786e-05, "output_power": 2.9462034778363908e-05, "reference_active": true, "delta_db": -0.24214628372509844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.223707279035506e-06, "output_power": 2.930727070957883e-06, "reference_active": true, "delta_db": -0.41380223965761737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5460080528529549e-06, "output_power": 1.258510394041243e-06, "reference_active": true, "delta_db": -0.8935494500437442}], "active_band_count": 3, "active_band_mae_db": 0.5164993244754866}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 5.4692297362327305e-05, "log_rms_mae_db": 1.2403050005370895, "rms_envelope_nrmse": 0.07624092020680073, "envelope_correlation": 0.9958362875346748, "rms_gain_error_db": -0.05174290990637799, "reference_mean_power": 0.00013203257062264164, "output_mean_power": 0.00013046883604461793}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 6.492145466124568e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.913804296383342e-05, "output_power": 2.7664841432788434e-05, "reference_active": true, "delta_db": -0.2253219402841779}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5333654526802403e-06, "output_power": 1.3465557427682378e-06, "reference_active": true, "delta_db": -0.5642133779206552}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.807699107498543e-08, "output_power": 6.092666739610851e-08, "reference_active": true, "delta_db": -1.6005504618693311}], "active_band_count": 3, "active_band_mae_db": 0.7966952600247215}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7309631895435401, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "06b078b1309854450db740f497a87ae26015cef551eba170fce97154c5e183c5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9859781581223486, "standard_deviation": 0.0010408781930547146, "minimum": 0.9842512788512978, "maximum": 0.9871693543066967, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9792701660425125, "standard_deviation": 0.036548306535993, "minimum": 1.9187943611082743, "maximum": 2.0346377243919758, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5574503839934076, "standard_deviation": 0.2262225121326669, "minimum": 0.35348248886334327, "maximum": 1.0472486203509295, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1481523462418386, "standard_deviation": 0.006434800179977811, "minimum": 0.13973258434843375, "maximum": 0.15932957171106082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953737636521934, "standard_deviation": 0.0006813764623385424, "minimum": 0.9939226374130807, "maximum": 0.996102666042836, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2512482511584448, "standard_deviation": 0.032577880248336584, "minimum": 1.209355717483817, "maximum": 1.2971811350867846, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5810383180720835, "standard_deviation": 0.28491329892016937, "minimum": 0.34403460088491805, "maximum": 1.1755378486738814, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08954233161036763, "standard_deviation": 0.009259959166977519, "minimum": 0.0783923029388846, "maximum": 0.10433577724579605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9821443274185068, "standard_deviation": 0.0019171499058730243, "minimum": 0.9798530958004774, "maximum": 0.9859628236072165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8753264659675437, "standard_deviation": 0.03513453616029663, "minimum": 1.816080307781199, "maximum": 1.9179339279161227, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6899637871985362, "standard_deviation": 0.20972950403449128, "minimum": 0.3844475320227523, "maximum": 0.9628694008814177, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16145750868540365, "standard_deviation": 0.00963962515362435, "minimum": 0.14254757917428304, "maximum": 0.174152039263515, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947879124023522, "standard_deviation": 0.0009960747792049226, "minimum": 0.9928774743333394, "maximum": 0.9960567465965089, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.245271065952155, "standard_deviation": 0.02252277851988772, "minimum": 1.208193048853508, "maximum": 1.2797107142918975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0047377893150002, "standard_deviation": 0.30091240272382724, "minimum": 0.48832547016803796, "maximum": 1.4003658879220893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08613072668807283, "standard_deviation": 0.008024939046914457, "minimum": 0.07590233440087206, "maximum": 0.09930252111574571, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992683457254674, "standard_deviation": 0.0011621720517026538, "minimum": 0.9907547970484885, "maximum": 0.9939078159246497, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.9909107295247033, "standard_deviation": 0.07355532683506144, "minimum": 2.9118085971816763, "maximum": 3.131175358841015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7005013661920784, "standard_deviation": 0.40909825857163473, "minimum": 0.21607846034647885, "maximum": 1.3501358275422224, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10783329684480564, "standard_deviation": 0.008305752435472791, "minimum": 0.09896047431648992, "maximum": 0.1222582905444646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965273644514989, "standard_deviation": 0.0009660044866993985, "minimum": 0.9947476545733898, "maximum": 0.9973561252959736, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.2883683842045266, "standard_deviation": 0.08580605908688368, "minimum": 2.173419571175802, "maximum": 2.377995718330487, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8516949997829868, "standard_deviation": 0.521856790473798, "minimum": 0.1016489141745125, "maximum": 1.5343635724393743, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07416235368888993, "standard_deviation": 0.009215905091618612, "minimum": 0.06604857156352563, "maximum": 0.08995009534395085, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15745107863603866, "standard_deviation": 0.04296260169002696, "minimum": 0.08144840494497803, "maximum": 0.2130375226402934, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.924368499105707, "standard_deviation": 0.00829753809890374, "minimum": 5.913843465151494, "maximum": 5.936456179802576, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.58799123591461, "standard_deviation": 0.013208633631926347, "minimum": 0.5688475769844377, "maximum": 0.6071144750356546, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08763887942819873, "standard_deviation": 0.028571394248725596, "minimum": -0.1189511329269749, "maximum": -0.04677215889342113, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901958775109841, "standard_deviation": 0.0005670061102994768, "minimum": 0.9894603866990154, "maximum": 0.9910598576651181, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5686173580497242, "standard_deviation": 0.045342382277124324, "minimum": 1.5048091732357671, "maximum": 1.6559353898728415, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4309987307137937, "standard_deviation": 0.1328431338546167, "minimum": 0.2065701606379954, "maximum": 0.5854751836291481, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09396051830881064, "standard_deviation": 0.002527897479054233, "minimum": 0.0903836529426479, "maximum": 0.09732675256024566, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973007994316461, "standard_deviation": 0.0002477715659067327, "minimum": 0.997042762346886, "maximum": 0.997712802458249, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7089938927139857, "standard_deviation": 0.03497803410546343, "minimum": 0.6613995002428541, "maximum": 0.7739440697570211, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46544150555488656, "standard_deviation": 0.16520959305332206, "minimum": 0.2669931616776697, "maximum": 0.789409267952229, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04829591909612744, "standard_deviation": 0.0021506111691851426, "minimum": 0.04450549034415261, "maximum": 0.0509690673638621, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0038", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.974628448486328, "predicted_mos": 2.974628448486328, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "16e5e1f317a2224bb3194d95470aab387175b8500e9d707f47ebfab31637da8b", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7748822001740336, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5310344602658432, "全帯域_10ms_rms_envelope_nrmse": 0.09447071989624016, "全帯域_10ms_envelope_correlation": 0.9900256588364703, "全帯域_10ms_modulation_mae_db": 0.50189842710293, "全帯域_40ms_log_rms_mae_db": 0.680057471248844, "全帯域_40ms_rms_envelope_nrmse": 0.04655321093179907, "全帯域_40ms_envelope_correlation": 0.9974031065576993, "全帯域_40ms_modulation_mae_db": 0.25549301156322585, "300–1000 Hz_10ms_log_rms_mae_db": 2.9084478381162464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11118175954277647, "300–1000 Hz_10ms_envelope_correlation": 0.9920792731047147, "300–1000 Hz_10ms_modulation_mae_db": 0.6230219994084947, "300–1000 Hz_40ms_log_rms_mae_db": 2.1651012571401425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06642718260320082, "300–1000 Hz_40ms_envelope_correlation": 0.9971675080071413, "300–1000 Hz_40ms_modulation_mae_db": 0.8204993656339982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8292234793520632, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.139942725093648, "1000–3000 Hz_10ms_envelope_correlation": 0.9870383865152583, "1000–3000 Hz_10ms_modulation_mae_db": 0.4938017934014882, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0348119184392055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06815517878496333, "1000–3000 Hz_40ms_envelope_correlation": 0.997062519617864, "1000–3000 Hz_40ms_modulation_mae_db": 0.6366859248109886, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7711980566076218, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15788744748857234, "3000–6000 Hz_10ms_envelope_correlation": 0.9828176107199967, "3000–6000 Hz_10ms_modulation_mae_db": 0.8047452467115349, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1214790264823198, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08203744300355047, "3000–6000 Hz_40ms_envelope_correlation": 0.9951460878339595, "3000–6000 Hz_40ms_modulation_mae_db": 0.8779844702897734, "mr_stft_db": 5.852095348858653, "band_mae_db": 0.09399998236632778, "rms_gain_error_db": -0.04896771547262278, "output_peak": 0.613572255133392}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5806549818235631, "全帯域_10ms_rms_envelope_nrmse": 0.09489833347067173, "全帯域_10ms_envelope_correlation": 0.9899031854591158, "全帯域_10ms_modulation_mae_db": 0.49198145171728536, "全帯域_40ms_log_rms_mae_db": 0.7827856062473596, "全帯域_40ms_rms_envelope_nrmse": 0.05125856691081565, "全帯域_40ms_envelope_correlation": 0.9968638913149265, "全帯域_40ms_modulation_mae_db": 0.3126137895783759, "300–1000 Hz_10ms_log_rms_mae_db": 2.9831306483484803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10496057675563937, "300–1000 Hz_10ms_envelope_correlation": 0.9930202915564106, "300–1000 Hz_10ms_modulation_mae_db": 0.9356171793835801, "300–1000 Hz_40ms_log_rms_mae_db": 2.271816628933739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06596764408551702, "300–1000 Hz_40ms_envelope_correlation": 0.9972931004789483, "300–1000 Hz_40ms_modulation_mae_db": 1.0765873539258877, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8789282225032016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16045149070857898, "1000–3000 Hz_10ms_envelope_correlation": 0.9845046916605945, "1000–3000 Hz_10ms_modulation_mae_db": 0.5157267540229583, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1176815427588314, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10368254201972187, "1000–3000 Hz_40ms_envelope_correlation": 0.9949185892380655, "1000–3000 Hz_40ms_modulation_mae_db": 0.5517812050277505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7745570219006237, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1380237452017188, "3000–6000 Hz_10ms_envelope_correlation": 0.9868145533059653, "3000–6000 Hz_10ms_modulation_mae_db": 0.3873467072772119, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0895095438751359, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0668875856145327, "3000–6000 Hz_40ms_envelope_correlation": 0.996827140944521, "3000–6000 Hz_40ms_modulation_mae_db": 0.6865537329844497, "mr_stft_db": 5.867715365650739, "band_mae_db": 0.2448691301400926, "rms_gain_error_db": -0.06408557295339432, "output_peak": 0.6102454169362879}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5061368326534859, "全帯域_10ms_rms_envelope_nrmse": 0.09336042619986783, "全帯域_10ms_envelope_correlation": 0.9903199248543963, "全帯域_10ms_modulation_mae_db": 0.5184803631380913, "全帯域_40ms_log_rms_mae_db": 0.6343016561342348, "全帯域_40ms_rms_envelope_nrmse": 0.04580634047034668, "全帯域_40ms_envelope_correlation": 0.997705503331141, "全帯域_40ms_modulation_mae_db": 0.13256093190422205, "300–1000 Hz_10ms_log_rms_mae_db": 2.9768592785167756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09504978906505601, "300–1000 Hz_10ms_envelope_correlation": 0.9944603787318286, "300–1000 Hz_10ms_modulation_mae_db": 0.5898557175564796, "300–1000 Hz_40ms_log_rms_mae_db": 2.2193631665225273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05849263255941556, "300–1000 Hz_40ms_envelope_correlation": 0.9980490013809755, "300–1000 Hz_40ms_modulation_mae_db": 0.7717277752797198, "1000–3000 Hz_10ms_log_rms_mae_db": 1.898210931741387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1500577012008404, "1000–3000 Hz_10ms_envelope_correlation": 0.9852188487773321, "1000–3000 Hz_10ms_modulation_mae_db": 0.6148325445597669, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1006932417763973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08177728276301625, "1000–3000 Hz_40ms_envelope_correlation": 0.9958546661817295, "1000–3000 Hz_40ms_modulation_mae_db": 0.7776596787626753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7375337793816912, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14436600007663952, "3000–6000 Hz_10ms_envelope_correlation": 0.9855468424587241, "3000–6000 Hz_10ms_modulation_mae_db": 0.6268216673055181, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1530888937876118, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.076863398645466, "3000–6000 Hz_40ms_envelope_correlation": 0.9957478841198503, "3000–6000 Hz_40ms_modulation_mae_db": 0.7268558561712407, "mr_stft_db": 5.848708475677211, "band_mae_db": 0.11887150816050639, "rms_gain_error_db": -0.12535845824174277, "output_peak": 0.6089531630739636}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5608238173518005, "全帯域_10ms_rms_envelope_nrmse": 0.09081464852065521, "全帯域_10ms_envelope_correlation": 0.9908057888052952, "全帯域_10ms_modulation_mae_db": 0.36767033766841495, "全帯域_40ms_log_rms_mae_db": 0.6319066732139494, "全帯域_40ms_rms_envelope_nrmse": 0.04560637893353217, "全帯域_40ms_envelope_correlation": 0.9976175164433914, "全帯域_40ms_modulation_mae_db": 0.2775268776190513, "300–1000 Hz_10ms_log_rms_mae_db": 2.7727757951901535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1128335308151104, "300–1000 Hz_10ms_envelope_correlation": 0.9919096459380669, "300–1000 Hz_10ms_modulation_mae_db": 0.8433900188861525, "300–1000 Hz_40ms_log_rms_mae_db": 2.041743374566024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06950331692197963, "300–1000 Hz_40ms_envelope_correlation": 0.996951981831456, "300–1000 Hz_40ms_modulation_mae_db": 1.1682893366687535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8780108275942995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17536571579994142, "1000–3000 Hz_10ms_envelope_correlation": 0.9800396923734217, "1000–3000 Hz_10ms_modulation_mae_db": 0.46455938447583156, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0663249888165816, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11131055242082323, "1000–3000 Hz_40ms_envelope_correlation": 0.9925540289933311, "1000–3000 Hz_40ms_modulation_mae_db": 0.370779405853185, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7795641430832274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17283763913117214, "3000–6000 Hz_10ms_envelope_correlation": 0.9794566325038551, "3000–6000 Hz_10ms_modulation_mae_db": 0.6866046904223699, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1177135012834458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08001314010320155, "3000–6000 Hz_40ms_envelope_correlation": 0.9953998664022653, "3000–6000 Hz_40ms_modulation_mae_db": 0.621515707867377, "mr_stft_db": 5.853500398722457, "band_mae_db": 0.11115945194608656, "rms_gain_error_db": -0.10076769005632336, "output_peak": 0.5680638852282501}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6068650514351324, "全帯域_10ms_rms_envelope_nrmse": 0.09843424029267389, "全帯域_10ms_envelope_correlation": 0.989281856012913, "全帯域_10ms_modulation_mae_db": 0.39678185070135635, "全帯域_40ms_log_rms_mae_db": 0.7470836667209639, "全帯域_40ms_rms_envelope_nrmse": 0.04893993828465814, "全帯域_40ms_envelope_correlation": 0.9971771829572884, "全帯域_40ms_modulation_mae_db": 0.4298525407575286, "300–1000 Hz_10ms_log_rms_mae_db": 2.94347521860694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10198269125144727, "300–1000 Hz_10ms_envelope_correlation": 0.9934061356353395, "300–1000 Hz_10ms_modulation_mae_db": 0.1746118606376363, "300–1000 Hz_40ms_log_rms_mae_db": 2.159085217666328, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06343787956455989, "300–1000 Hz_40ms_envelope_correlation": 0.9974689742972854, "300–1000 Hz_40ms_modulation_mae_db": 0.12929360156398062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9256623191122455, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16855078568466145, "1000–3000 Hz_10ms_envelope_correlation": 0.9809736075307254, "1000–3000 Hz_10ms_modulation_mae_db": 0.3651249887500591, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1341051237595268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07483339761638798, "1000–3000 Hz_40ms_envelope_correlation": 0.9962561842014922, "1000–3000 Hz_40ms_modulation_mae_db": 0.3094750535216296, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7511016345637715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.155487993398525, "3000–6000 Hz_10ms_envelope_correlation": 0.9832467192104297, "3000–6000 Hz_10ms_modulation_mae_db": 0.33238367839186117, "3000–6000 Hz_40ms_log_rms_mae_db": 1.083370182193719, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0765236831889771, "3000–6000 Hz_40ms_envelope_correlation": 0.9957698073037122, "3000–6000 Hz_40ms_modulation_mae_db": 0.4255951338078303, "mr_stft_db": 5.877163606478692, "band_mae_db": 0.07269908502214986, "rms_gain_error_db": -0.047479541310258555, "output_peak": 0.5960734972912931}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5968020388267004, "全帯域_10ms_rms_envelope_nrmse": 0.0934351666834842, "全帯域_10ms_envelope_correlation": 0.9903981306465324, "全帯域_10ms_modulation_mae_db": 0.36831182629991993, "全帯域_40ms_log_rms_mae_db": 0.7618602808761739, "全帯域_40ms_rms_envelope_nrmse": 0.0493582977776089, "全帯域_40ms_envelope_correlation": 0.9973055355598076, "全帯域_40ms_modulation_mae_db": 0.3062698483914165, "300–1000 Hz_10ms_log_rms_mae_db": 2.7852940039362575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10721130529885115, "300–1000 Hz_10ms_envelope_correlation": 0.9926507416402104, "300–1000 Hz_10ms_modulation_mae_db": 0.7855973208454713, "300–1000 Hz_40ms_log_rms_mae_db": 2.1137707496142, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07369086886705317, "300–1000 Hz_40ms_envelope_correlation": 0.9964822768543417, "300–1000 Hz_40ms_modulation_mae_db": 1.0410671313790243, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9201737862555124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15165307321237062, "1000–3000 Hz_10ms_envelope_correlation": 0.9859378524456186, "1000–3000 Hz_10ms_modulation_mae_db": 0.831606398832739, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1498323145453204, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09052796088976103, "1000–3000 Hz_40ms_envelope_correlation": 0.9961856394289429, "1000–3000 Hz_40ms_modulation_mae_db": 0.9864315907280474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7468318047161713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14688576953439328, "3000–6000 Hz_10ms_envelope_correlation": 0.9850783004962332, "3000–6000 Hz_10ms_modulation_mae_db": 0.5751056592660164, "3000–6000 Hz_40ms_log_rms_mae_db": 1.123821028413948, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07491959974307505, "3000–6000 Hz_40ms_envelope_correlation": 0.9959384743207064, "3000–6000 Hz_40ms_modulation_mae_db": 0.735256773218163, "mr_stft_db": 5.87722073434902, "band_mae_db": 0.2260177010985045, "rms_gain_error_db": -0.1327838131599818, "output_peak": 0.6025390892434775}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.649537860119299, "全帯域_10ms_rms_envelope_nrmse": 0.09225292656883836, "全帯域_10ms_envelope_correlation": 0.9905440165637899, "全帯域_10ms_modulation_mae_db": 0.5020151987119054, "全帯域_40ms_log_rms_mae_db": 0.7931670424852528, "全帯域_40ms_rms_envelope_nrmse": 0.04715842346499585, "全帯域_40ms_envelope_correlation": 0.9973272667899356, "全帯域_40ms_modulation_mae_db": 0.4518606990800647, "300–1000 Hz_10ms_log_rms_mae_db": 2.8537567281057887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10819044639226773, "300–1000 Hz_10ms_envelope_correlation": 0.992594360293731, "300–1000 Hz_10ms_modulation_mae_db": 0.24465573422828446, "300–1000 Hz_40ms_log_rms_mae_db": 2.1356803733508394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07859053576910689, "300–1000 Hz_40ms_envelope_correlation": 0.9960054882640377, "300–1000 Hz_40ms_modulation_mae_db": 0.5123323145707626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.845658114834364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13086384666514803, "1000–3000 Hz_10ms_envelope_correlation": 0.9888655987306166, "1000–3000 Hz_10ms_modulation_mae_db": 0.5644375251439829, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0749120873744702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07794747952163973, "1000–3000 Hz_40ms_envelope_correlation": 0.9962331783670385, "1000–3000 Hz_40ms_modulation_mae_db": 0.5574969259392962, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7557718334980108, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16424145126210452, "3000–6000 Hz_10ms_envelope_correlation": 0.9814339601043717, "3000–6000 Hz_10ms_modulation_mae_db": 0.8731640139492765, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0996287252706072, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09755568098505296, "3000–6000 Hz_40ms_envelope_correlation": 0.9935248156230455, "3000–6000 Hz_40ms_modulation_mae_db": 1.0632701076230726, "mr_stft_db": 5.864151250497616, "band_mae_db": 0.13016318514747469, "rms_gain_error_db": -0.02575272849297116, "output_peak": 0.602111687647066}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.551193768593091, "全帯域_10ms_rms_envelope_nrmse": 0.09179789187771276, "全帯域_10ms_envelope_correlation": 0.9907424169259115, "全帯域_10ms_modulation_mae_db": 0.2155093465614405, "全帯域_40ms_log_rms_mae_db": 0.6861125640096267, "全帯域_40ms_rms_envelope_nrmse": 0.046037943407157465, "全帯域_40ms_envelope_correlation": 0.9976756961234478, "全帯域_40ms_modulation_mae_db": 0.33731961235375146, "300–1000 Hz_10ms_log_rms_mae_db": 2.7455221224723148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09939177592878119, "300–1000 Hz_10ms_envelope_correlation": 0.9936634322567317, "300–1000 Hz_10ms_modulation_mae_db": 0.29622899207364656, "300–1000 Hz_40ms_log_rms_mae_db": 1.973272665975906, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05607263731519714, "300–1000 Hz_40ms_envelope_correlation": 0.998003017688303, "300–1000 Hz_40ms_modulation_mae_db": 0.46494051646783535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8446095835586147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15173317985076193, "1000–3000 Hz_10ms_envelope_correlation": 0.9849426531779889, "1000–3000 Hz_10ms_modulation_mae_db": 0.5414842940346575, "1000–3000 Hz_40ms_log_rms_mae_db": 1.08961081205558, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07930399288079915, "1000–3000 Hz_40ms_envelope_correlation": 0.9962935337318639, "1000–3000 Hz_40ms_modulation_mae_db": 0.5019004651020138, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7383026244537692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15793867184790597, "3000–6000 Hz_10ms_envelope_correlation": 0.9827799256622601, "3000–6000 Hz_10ms_modulation_mae_db": 0.6725043889759584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1044171817495059, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08666188812132236, "3000–6000 Hz_40ms_envelope_correlation": 0.9948755087438546, "3000–6000 Hz_40ms_modulation_mae_db": 0.9657724338764379, "mr_stft_db": 5.841077163261416, "band_mae_db": 0.1631305283404608, "rms_gain_error_db": -0.1272425395656971, "output_peak": 0.5884100454484309}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88640, "output_samples": 88640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0792104820908905, "spectral_convergence": 0.3418308020467015, "spectral_convergence_status": "ok", "amplitude_floor": 1.5935254482600421e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.663281600356039, "spectral_convergence": 0.2923941461998816, "spectral_convergence_status": "ok", "amplitude_floor": 9.146897175620752e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.81379396412903, "spectral_convergence": 0.30896527970403687, "spectral_convergence_status": "ok", "amplitude_floor": 7.905477970735852e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007242342814613898, "output_power": 0.0007382931228425678, "delta_db": 0.08349744817597726, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001748392652036385, "output_power": 0.0017164998070386822, "delta_db": -0.07995213673260794, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009820880914630636, "output_power": 0.0009674018434017178, "delta_db": -0.06543534326855993, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002665715938534463, "output_power": 0.0002520457991529804, "delta_db": -0.2433440522241404, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.016487980563797e-05, "output_power": 6.0408517730930544e-05, "delta_db": 0.017551259682447633, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00015485832573279259, "output_power": 0.00015752755231947994, "delta_db": 0.07421965411423354, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09399998236632778, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04896771547262278, "pitch": null, "output_rms_dbfs": -24.098073631962436, "output_peak": 0.613572255133392, "reference_rms_dbfs": -24.04910591648981, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.852095348858653}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00022148173042756648, "log_rms_mae_db": 1.5310344602658432, "rms_envelope_nrmse": 0.09447071989624016, "envelope_correlation": 0.9900256588364703, "rms_gain_error_db": -0.04895289206731711, "reference_mean_power": 0.0039398267431541775, "output_mean_power": 0.0038956670729379875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.3094240146918987e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011389510990882513, "output_power": 0.0011223490952849936, "reference_active": true, "delta_db": -0.0637711720863635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017715594495014008, "output_power": 0.00016445042668415352, "reference_active": true, "delta_db": -0.3232072620424735}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5198627564616767e-05, "output_power": 3.260232500079962e-05, "reference_active": true, "delta_db": 1.1187168471799527}], "active_band_count": 3, "active_band_mae_db": 0.50189842710293}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.00020322184022683357, "log_rms_mae_db": 0.680057471248844, "rms_envelope_nrmse": 0.04655321093179907, "envelope_correlation": 0.9974031065576993, "rms_gain_error_db": -0.04854206127296949, "reference_mean_power": 0.0039603691212238695, "output_mean_power": 0.003916349659653579}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.1092095592585075e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010485738336262066, "output_power": 0.0010358014506967493, "reference_active": true, "delta_db": -0.0532250108421703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.703453851742181e-05, "output_power": 8.249224428978126e-05, "reference_active": true, "delta_db": -0.23278511693883142}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.852514808001562e-06, "output_power": 2.06922910723049e-06, "reference_active": true, "delta_db": 0.4804689069086759}], "active_band_count": 3, "active_band_mae_db": 0.25549301156322585}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00017913726163896074, "log_rms_mae_db": 2.9084478381162464, "rms_envelope_nrmse": 0.11118175954277647, "envelope_correlation": 0.9920792731047147, "rms_gain_error_db": -0.020700130175536596, "reference_mean_power": 0.001348381907516537, "output_mean_power": 0.0013419702976811615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 1.724352265721046e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008753679492081999, "output_power": 0.0008452627861922844, "reference_active": true, "delta_db": -0.15198892265618072}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012951638097036193, "output_power": 0.00010804219061191622, "reference_active": true, "delta_db": -0.7873131942797517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6972503859652748e-05, "output_power": 1.3701544576560743e-05, "reference_active": true, "delta_db": -0.9297638812895516}], "active_band_count": 3, "active_band_mae_db": 0.6230219994084947}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.0001658599445391014, "log_rms_mae_db": 2.1651012571401425, "rms_envelope_nrmse": 0.06642718260320082, "envelope_correlation": 0.9971675080071413, "rms_gain_error_db": -0.020713401716140528, "reference_mean_power": 0.001355729112927364, "output_mean_power": 0.0013492784435904875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 1.6447728018068126e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008253872294632201, "output_power": 0.0007967296243204672, "reference_active": true, "delta_db": -0.15346779535149618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.001034181994634e-05, "output_power": 5.1621477804791195e-05, "reference_active": true, "delta_db": -0.6539566727893643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.081003459949387e-06, "output_power": 1.4218881650036489e-06, "reference_active": true, "delta_db": -1.6540736287611344}], "active_band_count": 3, "active_band_mae_db": 0.8204993656339982}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00010266849739942118, "log_rms_mae_db": 1.8292234793520632, "rms_envelope_nrmse": 0.139942725093648, "envelope_correlation": 0.9870383865152583, "rms_gain_error_db": -0.18048688858265782, "reference_mean_power": 0.0002406311059478081, "output_mean_power": 0.00023083575450633464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.4743465614681157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013528556332001226, "output_power": 0.00012778381850123716, "reference_active": true, "delta_db": -0.2477559255909674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.422298628989197e-05, "output_power": 1.3042312606137665e-05, "reference_active": true, "delta_db": -0.37636185972554087}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7979674926628947e-06, "output_power": 2.2967509515545633e-06, "reference_active": true, "delta_db": -0.8572875948879564}], "active_band_count": 3, "active_band_mae_db": 0.4938017934014882}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 8.271272308600344e-05, "log_rms_mae_db": 1.0348119184392055, "rms_envelope_nrmse": 0.06815517878496333, "envelope_correlation": 0.997062519617864, "rms_gain_error_db": -0.1805027092598742, "reference_mean_power": 0.00024194203098044546, "output_mean_power": 0.0002320924703385511}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.355449255956094e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012689841334701795, "output_power": 0.00012142102895153574, "reference_active": true, "delta_db": -0.19162283130270152}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.520296796235256e-06, "output_power": 5.702592083857402e-06, "reference_active": true, "delta_db": -0.5819505788468946}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.927693257279988e-07, "output_power": 1.483848260303094e-07, "reference_active": true, "delta_db": -1.1364843642833697}], "active_band_count": 3, "active_band_mae_db": 0.6366859248109886}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 5.887039168463902e-05, "log_rms_mae_db": 1.7711980566076218, "rms_envelope_nrmse": 0.15788744748857234, "envelope_correlation": 0.9828176107199967, "rms_gain_error_db": 0.014215045947864269, "reference_mean_power": 0.00013131711251995087, "output_mean_power": 0.00013174763539050156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 6.872487837717039e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.115138542486786e-05, "output_power": 2.8981244056987936e-05, "reference_active": true, "delta_db": -0.31360342030042837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.223707279035506e-06, "output_power": 3.803729510704452e-06, "reference_active": true, "delta_db": 0.7185402612710755}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5460080528529549e-06, "output_power": 1.1246115665564943e-06, "reference_active": true, "delta_db": -1.3820920585631007}], "active_band_count": 3, "active_band_mae_db": 0.8047452467115349}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 5.4692297362327305e-05, "log_rms_mae_db": 1.1214790264823198, "rms_envelope_nrmse": 0.08203744300355047, "envelope_correlation": 0.9951460878339595, "rms_gain_error_db": 0.014206311300376664, "reference_mean_power": 0.00013203257062264164, "output_mean_power": 0.0001324651727032531}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 6.492145466124568e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.913804296383342e-05, "output_power": 2.721373217064991e-05, "reference_active": true, "delta_db": -0.2967227305303479}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5333654526802403e-06, "output_power": 1.853711324831616e-06, "reference_active": true, "delta_db": 0.823964289596884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.807699107498543e-08, "output_power": 6.216355324126939e-08, "reference_active": true, "delta_db": -1.5132663907420885}], "active_band_count": 3, "active_band_mae_db": 0.8779844702897734}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.739035398616385, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "44190d989bead0b55e12396068f03766780296d6956671c921fd1a86965e1a23", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9846901664014446, "standard_deviation": 0.0029355570737930625, "minimum": 0.9800396923734217, "maximum": 0.9888655987306166, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8775596581189609, "standard_deviation": 0.035853239724760025, "minimum": 1.8292234793520632, "maximum": 1.9256623191122455, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5489467104026855, "standard_deviation": 0.13606157697722696, "minimum": 0.3651249887500591, "maximum": 0.831606398832739, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15357731477699385, "standard_deviation": 0.01447899184212171, "minimum": 0.13086384666514803, "maximum": 0.17536571579994142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995669792470041, "standard_deviation": 0.0013919444628927753, "minimum": 0.9925540289933311, "maximum": 0.997062519617864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0959965036907393, "standard_deviation": 0.037705898853275396, "minimum": 1.0348119184392055, "maximum": 1.1498323145453204, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5865262812181984, "standard_deviation": 0.21761294985595758, "minimum": 0.3094750535216296, "maximum": 0.9864315907280474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08594229836213907, "standard_deviation": 0.014853637079190947, "minimum": 0.06815517878496333, "maximum": 0.11131055242082323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9833968180577295, "standard_deviation": 0.002367860484504815, "minimum": 0.9794566325038551, "maximum": 0.9868145533059653, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7568576122756108, "standard_deviation": 0.01641875708278083, "minimum": 1.7375337793816912, "maximum": 1.7795641430832274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6198345065374684, "standard_deviation": 0.18682621440779915, "minimum": 0.33238367839186117, "maximum": 0.8731640139492765, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15470858974262894, "standard_deviation": 0.0112582264010436, "minimum": 0.1380237452017188, "maximum": 0.17283763913117214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954036981614893, "standard_deviation": 0.0009591838936142546, "minimum": 0.9935248156230455, "maximum": 0.996827140944521, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1116285103820367, "standard_deviation": 0.02231418691744546, "minimum": 1.083370182193719, "maximum": 1.1530888937876118, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.762850526979793, "standard_deviation": 0.20223042490433538, "minimum": 0.4255951338078303, "maximum": 1.0632701076230726, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08018280242564726, "standard_deviation": 0.009072215239360958, "minimum": 0.0668875856145327, "maximum": 0.09755568098505296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9929730323946292, "standard_deviation": 0.0008498338708703054, "minimum": 0.9919096459380669, "maximum": 0.9944603787318286, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.87115770416162, "standard_deviation": 0.09522140247136954, "minimum": 2.7455221224723148, "maximum": 2.9831306483484803, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5616223528774682, "standard_deviation": 0.2914680987231905, "minimum": 0.1746118606376363, "maximum": 0.9356171793835801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10510023438124119, "standard_deviation": 0.006025320158563677, "minimum": 0.09504978906505601, "maximum": 0.1128335308151104, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971776686003111, "standard_deviation": 0.0007017549059122364, "minimum": 0.9960054882640377, "maximum": 0.9980490013809755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.1349791792212134, "standard_deviation": 0.09458615609940438, "minimum": 1.973272665975906, "maximum": 2.271816628933739, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7480921744362452, "standard_deviation": 0.35752717242969123, "minimum": 0.12929360156398062, "maximum": 1.1682893366687535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06652283721075376, "standard_deviation": 0.007458633569979245, "minimum": 0.05607263731519714, "maximum": 0.07859053576910689, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1451138215277004, "standard_deviation": 0.061809337505453345, "minimum": 0.07269908502214986, "maximum": 0.2448691301400926, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.860204042936975, "standard_deviation": 0.013405064910233909, "minimum": 5.841077163261416, "maximum": 5.87722073434902, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5987461300002701, "standard_deviation": 0.01481949672594412, "minimum": 0.5680638852282501, "maximum": 0.613572255133392, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08405475740662398, "standard_deviation": 0.04241254037969278, "minimum": -0.1327838131599818, "maximum": -0.02575272849297116, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.990252622263053, "standard_deviation": 0.0005032055234125172, "minimum": 0.989281856012913, "maximum": 0.9908057888052952, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5728811013836144, "standard_deviation": 0.04549013907142884, "minimum": 1.5061368326534859, "maximum": 1.649537860119299, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42033110023766795, "standard_deviation": 0.10425124195125715, "minimum": 0.2155093465614405, "maximum": 0.5184803631380913, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09368304418876802, "standard_deviation": 0.0023501817696439816, "minimum": 0.09081464852065521, "maximum": 0.09843424029267389, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973844623847047, "standard_deviation": 0.00028435540106337367, "minimum": 0.9968638913149265, "maximum": 0.997705503331141, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7146593701170506, "standard_deviation": 0.06480904064161537, "minimum": 0.6319066732139494, "maximum": 0.7931670424852528, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31293716390595455, "standard_deviation": 0.10049685106189099, "minimum": 0.13256093190422205, "maximum": 0.4518606990800647, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04758988752261424, "standard_deviation": 0.0020421926898926105, "minimum": 0.04560637893353217, "maximum": 0.05125856691081565, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2830-3980-0042", "language": "en", "style": "normal", "speaker": "librispeech_2830", "source_utterance_id": "2830-3980-0042", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 48320, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2830-3980-0042.wav", "source_file_sha256": "f6f5e38352b5f6d86e9207c9fe05dfe27163f88895c48bc59cbf61a4d42bd4be", "decoded_pcm_sha256": "3b4a08b0741e0ca396487185085d64365ab56b16d5868e61a4550a5dc24466f2", "original_source_sha256": "ff30195a24b3f6c24a6ad4748f2ab49f4a036975c0b62f5b1c2f4203488ee9ae", "corpus_entry_id": "librispeech_2830-3980-0042"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.767183303833008, "predicted_mos": 4.767183303833008, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "de6b2b3874f099d5354f5235e16a8b2d0e4f870086a1d0af14e9607a95e00862", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.002317393998965, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2830-3980-0042", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "de6b2b3874f099d5354f5235e16a8b2d0e4f870086a1d0af14e9607a95e00862", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0042d8afd9bfdec43aeb47b3c8d03f059e19efc4a4103d4bd165e674531c4a9b"}, "rows": [{"case": "librispeech_2830-3980-0042", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1061484813690186, "predicted_mos": 2.1061484813690186, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "5ab6bc954874a83deb452c2c60cf191b1e7e1be3d84c28e9bebb1e25cec8bc29", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3785196730023017, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2830-3980-0042", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5ab6bc954874a83deb452c2c60cf191b1e7e1be3d84c28e9bebb1e25cec8bc29", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7684721306981401, "全帯域_10ms_rms_envelope_nrmse": 0.08625284546091547, "全帯域_10ms_envelope_correlation": 0.9927713047080733, "全帯域_10ms_modulation_mae_db": 0.33885507522162367, "全帯域_40ms_log_rms_mae_db": 0.8159095486241686, "全帯域_40ms_rms_envelope_nrmse": 0.04359281723760568, "全帯域_40ms_envelope_correlation": 0.9981961298713675, "全帯域_40ms_modulation_mae_db": 0.22396665722492795, "300–1000 Hz_10ms_log_rms_mae_db": 2.8338879406044035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09556994503940032, "300–1000 Hz_10ms_envelope_correlation": 0.9940578539574854, "300–1000 Hz_10ms_modulation_mae_db": 0.16893219865544687, "300–1000 Hz_40ms_log_rms_mae_db": 2.063601785668921, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05001359124121518, "300–1000 Hz_40ms_envelope_correlation": 0.9984124459500747, "300–1000 Hz_40ms_modulation_mae_db": 0.13070714571362974, "1000–3000 Hz_10ms_log_rms_mae_db": 1.803815493985327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1545758729705217, "1000–3000 Hz_10ms_envelope_correlation": 0.9844130217313815, "1000–3000 Hz_10ms_modulation_mae_db": 1.0919325058352822, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2737609080478673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.104195659422328, "1000–3000 Hz_40ms_envelope_correlation": 0.9934297945528445, "1000–3000 Hz_40ms_modulation_mae_db": 1.3672053385387162, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6894462141515998, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15985903499518878, "3000–6000 Hz_10ms_envelope_correlation": 0.9839319492384072, "3000–6000 Hz_10ms_modulation_mae_db": 0.9499232273170004, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1264470393297779, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09541978284495206, "3000–6000 Hz_40ms_envelope_correlation": 0.9954229372899872, "3000–6000 Hz_40ms_modulation_mae_db": 0.9565058744187298, "mr_stft_db": 5.8419920990861405, "band_mae_db": 0.24638343298627943, "rms_gain_error_db": -0.12088290230473661, "output_peak": 0.5561347830141948}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7485528421391439, "全帯域_10ms_rms_envelope_nrmse": 0.0891860446714551, "全帯域_10ms_envelope_correlation": 0.9920876785811895, "全帯域_10ms_modulation_mae_db": 0.40189493571080076, "全帯域_40ms_log_rms_mae_db": 0.8292810713410975, "全帯域_40ms_rms_envelope_nrmse": 0.04464976418786901, "全帯域_40ms_envelope_correlation": 0.9978982804614794, "全帯域_40ms_modulation_mae_db": 0.25900316589567635, "300–1000 Hz_10ms_log_rms_mae_db": 2.7687480073092394, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09092251576061455, "300–1000 Hz_10ms_envelope_correlation": 0.994601512872577, "300–1000 Hz_10ms_modulation_mae_db": 0.22557049850973987, "300–1000 Hz_40ms_log_rms_mae_db": 2.1379461100723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045497538861594994, "300–1000 Hz_40ms_envelope_correlation": 0.9986401745858119, "300–1000 Hz_40ms_modulation_mae_db": 0.30136299374440645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.811281632300326, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1737294194760101, "1000–3000 Hz_10ms_envelope_correlation": 0.97928344468194, "1000–3000 Hz_10ms_modulation_mae_db": 0.866501770694471, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1986475191843702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08927048373911661, "1000–3000 Hz_40ms_envelope_correlation": 0.9946112885547098, "1000–3000 Hz_40ms_modulation_mae_db": 0.6740027303791746, "3000–6000 Hz_10ms_log_rms_mae_db": 1.699687069754186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2090902479061562, "3000–6000 Hz_10ms_envelope_correlation": 0.9703633217778634, "3000–6000 Hz_10ms_modulation_mae_db": 1.1825828548125223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0681475055765304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11325323675430633, "3000–6000 Hz_40ms_envelope_correlation": 0.9920179836950165, "3000–6000 Hz_40ms_modulation_mae_db": 1.6792126308957107, "mr_stft_db": 5.8412476229553825, "band_mae_db": 0.10799831516086443, "rms_gain_error_db": -0.024149995989078715, "output_peak": 0.5456228547684705}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7965506444284096, "全帯域_10ms_rms_envelope_nrmse": 0.09285951666516884, "全帯域_10ms_envelope_correlation": 0.9914890100042425, "全帯域_10ms_modulation_mae_db": 0.18811078855306143, "全帯域_40ms_log_rms_mae_db": 0.8428394993100424, "全帯域_40ms_rms_envelope_nrmse": 0.04423949413746068, "全帯域_40ms_envelope_correlation": 0.9980567432389719, "全帯域_40ms_modulation_mae_db": 0.054790746304605294, "300–1000 Hz_10ms_log_rms_mae_db": 2.549648065645066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09366827515837058, "300–1000 Hz_10ms_envelope_correlation": 0.994442315000638, "300–1000 Hz_10ms_modulation_mae_db": 0.6941916245653559, "300–1000 Hz_40ms_log_rms_mae_db": 1.7715850435727043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04911466652344099, "300–1000 Hz_40ms_envelope_correlation": 0.9986783933184779, "300–1000 Hz_40ms_modulation_mae_db": 0.5090747724859053, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7939265367732276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15693833258478057, "1000–3000 Hz_10ms_envelope_correlation": 0.9832066834467429, "1000–3000 Hz_10ms_modulation_mae_db": 0.3674990969557821, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2350475112425645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08427207859273422, "1000–3000 Hz_40ms_envelope_correlation": 0.995308717587546, "1000–3000 Hz_40ms_modulation_mae_db": 0.31414021464254954, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7435344775441695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17985957291988533, "3000–6000 Hz_10ms_envelope_correlation": 0.9779232503982475, "3000–6000 Hz_10ms_modulation_mae_db": 0.7576732408666934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.177965729732743, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11134648088628095, "3000–6000 Hz_40ms_envelope_correlation": 0.9913116480791712, "3000–6000 Hz_40ms_modulation_mae_db": 1.0419653399645272, "mr_stft_db": 5.845506419608237, "band_mae_db": 0.21307267007762407, "rms_gain_error_db": -0.10385858571606157, "output_peak": 0.569523439688512}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7221452791212661, "全帯域_10ms_rms_envelope_nrmse": 0.09372987030008949, "全帯域_10ms_envelope_correlation": 0.9913166065431179, "全帯域_10ms_modulation_mae_db": 0.11514126178629482, "全帯域_40ms_log_rms_mae_db": 0.7136429242940017, "全帯域_40ms_rms_envelope_nrmse": 0.04867150786819215, "全帯域_40ms_envelope_correlation": 0.9975236129315344, "全帯域_40ms_modulation_mae_db": 0.08758572535327884, "300–1000 Hz_10ms_log_rms_mae_db": 2.685017104807666, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11517260264797036, "300–1000 Hz_10ms_envelope_correlation": 0.9915486491065586, "300–1000 Hz_10ms_modulation_mae_db": 0.2438332391314276, "300–1000 Hz_40ms_log_rms_mae_db": 2.004651416323934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07389269485517816, "300–1000 Hz_40ms_envelope_correlation": 0.9965140320422706, "300–1000 Hz_40ms_modulation_mae_db": 0.43613577192602343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8234911013490749, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1420730523025336, "1000–3000 Hz_10ms_envelope_correlation": 0.9863641578758775, "1000–3000 Hz_10ms_modulation_mae_db": 0.3821542998176441, "1000–3000 Hz_40ms_log_rms_mae_db": 1.231120761665809, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07635593175496491, "1000–3000 Hz_40ms_envelope_correlation": 0.9963280917130667, "1000–3000 Hz_40ms_modulation_mae_db": 0.3524262399790686, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6755900138079938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1639436473873438, "3000–6000 Hz_10ms_envelope_correlation": 0.9816252510745581, "3000–6000 Hz_10ms_modulation_mae_db": 0.2536502494542197, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0710040816510493, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0788046208712181, "3000–6000 Hz_40ms_envelope_correlation": 0.9955337075376635, "3000–6000 Hz_40ms_modulation_mae_db": 0.1644973825910166, "mr_stft_db": 5.830284862297901, "band_mae_db": 0.11127852836305523, "rms_gain_error_db": 0.010838826171504144, "output_peak": 0.5911280249998675}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7281136613388095, "全帯域_10ms_rms_envelope_nrmse": 0.0954686524019387, "全帯域_10ms_envelope_correlation": 0.9909638212477773, "全帯域_10ms_modulation_mae_db": 0.30626540585481493, "全帯域_40ms_log_rms_mae_db": 0.7435568861289078, "全帯域_40ms_rms_envelope_nrmse": 0.0532420076405881, "全帯域_40ms_envelope_correlation": 0.9970526869102749, "全帯域_40ms_modulation_mae_db": 0.18533593477364732, "300–1000 Hz_10ms_log_rms_mae_db": 2.644015338984766, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09859087980752353, "300–1000 Hz_10ms_envelope_correlation": 0.9936534948565602, "300–1000 Hz_10ms_modulation_mae_db": 0.351281461405776, "300–1000 Hz_40ms_log_rms_mae_db": 1.9478680857075974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0584145791869459, "300–1000 Hz_40ms_envelope_correlation": 0.9977059983039278, "300–1000 Hz_40ms_modulation_mae_db": 0.7992116206308045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8216293306983509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1499341927751659, "1000–3000 Hz_10ms_envelope_correlation": 0.9849119968361971, "1000–3000 Hz_10ms_modulation_mae_db": 0.92964769308775, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2473733873406894, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09394244518601447, "1000–3000 Hz_40ms_envelope_correlation": 0.9942909931634518, "1000–3000 Hz_40ms_modulation_mae_db": 1.072133915237311, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7216128617690167, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19887925808085236, "3000–6000 Hz_10ms_envelope_correlation": 0.973242814346507, "3000–6000 Hz_10ms_modulation_mae_db": 1.3024949991950565, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0763080903799906, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10594347381406616, "3000–6000 Hz_40ms_envelope_correlation": 0.9929888134284485, "3000–6000 Hz_40ms_modulation_mae_db": 1.3341065497675733, "mr_stft_db": 5.839606110754396, "band_mae_db": 0.16715723587017842, "rms_gain_error_db": -0.0577067504537923, "output_peak": 0.6001946039129424}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7503050662598008, "全帯域_10ms_rms_envelope_nrmse": 0.09313019972412538, "全帯域_10ms_envelope_correlation": 0.9914979981742436, "全帯域_10ms_modulation_mae_db": 0.4296192005592703, "全帯域_40ms_log_rms_mae_db": 0.7839805518968486, "全帯域_40ms_rms_envelope_nrmse": 0.04802593838788691, "全帯域_40ms_envelope_correlation": 0.9977918803610445, "全帯域_40ms_modulation_mae_db": 0.2283454209482424, "300–1000 Hz_10ms_log_rms_mae_db": 2.6589397119780775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10505218899855451, "300–1000 Hz_10ms_envelope_correlation": 0.9929550924064087, "300–1000 Hz_10ms_modulation_mae_db": 0.7898820662830218, "300–1000 Hz_40ms_log_rms_mae_db": 2.010584923757891, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06179763946432305, "300–1000 Hz_40ms_envelope_correlation": 0.9977430915131822, "300–1000 Hz_40ms_modulation_mae_db": 0.4593018789896081, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7668375039801678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15822554220239624, "1000–3000 Hz_10ms_envelope_correlation": 0.983271360643526, "1000–3000 Hz_10ms_modulation_mae_db": 1.1878277577169458, "1000–3000 Hz_40ms_log_rms_mae_db": 1.160168310984324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09268687348819821, "1000–3000 Hz_40ms_envelope_correlation": 0.994744185314158, "1000–3000 Hz_40ms_modulation_mae_db": 1.1349381304946224, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7674015484911494, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20387761213202354, "3000–6000 Hz_10ms_envelope_correlation": 0.9728183925787507, "3000–6000 Hz_10ms_modulation_mae_db": 1.3538668992439167, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1675260067825124, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12629271385563326, "3000–6000 Hz_40ms_envelope_correlation": 0.9892157295922812, "3000–6000 Hz_40ms_modulation_mae_db": 0.9451029936454942, "mr_stft_db": 5.840387999682261, "band_mae_db": 0.16083938312435556, "rms_gain_error_db": -0.14057288955006608, "output_peak": 0.5612312073540894}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.673759934339921, "全帯域_10ms_rms_envelope_nrmse": 0.09443261416388127, "全帯域_10ms_envelope_correlation": 0.9912041069734331, "全帯域_10ms_modulation_mae_db": 0.4074138197640151, "全帯域_40ms_log_rms_mae_db": 0.8003041265866407, "全帯域_40ms_rms_envelope_nrmse": 0.04538664383798551, "全帯域_40ms_envelope_correlation": 0.9980034634815009, "全帯域_40ms_modulation_mae_db": 0.5181972088258969, "300–1000 Hz_10ms_log_rms_mae_db": 2.699093609424769, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10554548085570521, "300–1000 Hz_10ms_envelope_correlation": 0.9927215783765595, "300–1000 Hz_10ms_modulation_mae_db": 0.7118455896181434, "300–1000 Hz_40ms_log_rms_mae_db": 1.9439951490363547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05317690942977685, "300–1000 Hz_40ms_envelope_correlation": 0.9982513142400646, "300–1000 Hz_40ms_modulation_mae_db": 0.4270734270878879, "1000–3000 Hz_10ms_log_rms_mae_db": 1.82957355627391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14894196099218404, "1000–3000 Hz_10ms_envelope_correlation": 0.9849998997148398, "1000–3000 Hz_10ms_modulation_mae_db": 0.8857047191309103, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2808596808436221, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09107959587273384, "1000–3000 Hz_40ms_envelope_correlation": 0.9944973407313817, "1000–3000 Hz_40ms_modulation_mae_db": 1.2243518761360586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6867653360607973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15163650277706478, "3000–6000 Hz_10ms_envelope_correlation": 0.9843435171111387, "3000–6000 Hz_10ms_modulation_mae_db": 1.0194236415980653, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1169088528112558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08505497846496257, "3000–6000 Hz_40ms_envelope_correlation": 0.9947644707200279, "3000–6000 Hz_40ms_modulation_mae_db": 0.8257473576008016, "mr_stft_db": 5.851120740843418, "band_mae_db": 0.164340282319255, "rms_gain_error_db": -0.12498596213508865, "output_peak": 0.5593545851424505}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7134379867452296, "全帯域_10ms_rms_envelope_nrmse": 0.09550975333153028, "全帯域_10ms_envelope_correlation": 0.9911999020555952, "全帯域_10ms_modulation_mae_db": 0.36033884681182293, "全帯域_40ms_log_rms_mae_db": 0.8268076142957677, "全帯域_40ms_rms_envelope_nrmse": 0.05063277616836811, "全帯域_40ms_envelope_correlation": 0.997702798358793, "全帯域_40ms_modulation_mae_db": 0.47918022026421775, "300–1000 Hz_10ms_log_rms_mae_db": 2.6714185352763113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10736974129908454, "300–1000 Hz_10ms_envelope_correlation": 0.9925586099494653, "300–1000 Hz_10ms_modulation_mae_db": 0.6340685573911554, "300–1000 Hz_40ms_log_rms_mae_db": 1.9814595704893554, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06226920128739359, "300–1000 Hz_40ms_envelope_correlation": 0.9976075545474701, "300–1000 Hz_40ms_modulation_mae_db": 0.5082504480487768, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7929165428380458, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16899012192720866, "1000–3000 Hz_10ms_envelope_correlation": 0.9822445139246886, "1000–3000 Hz_10ms_modulation_mae_db": 1.647476648265143, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2424713938007412, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12428684071923655, "1000–3000 Hz_40ms_envelope_correlation": 0.9914220320174455, "1000–3000 Hz_40ms_modulation_mae_db": 1.8749302947973312, "3000–6000 Hz_10ms_log_rms_mae_db": 1.63635296819423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15717004388233854, "3000–6000 Hz_10ms_envelope_correlation": 0.9831042389889053, "3000–6000 Hz_10ms_modulation_mae_db": 0.47966767295244517, "3000–6000 Hz_40ms_log_rms_mae_db": 1.028804547515415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08453923025500698, "3000–6000 Hz_40ms_envelope_correlation": 0.9949128945319118, "3000–6000 Hz_40ms_modulation_mae_db": 0.4252556066843503, "mr_stft_db": 5.830507746530151, "band_mae_db": 0.2676783669153016, "rms_gain_error_db": -0.1781525744440886, "output_peak": 0.5366437497223129}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48320, "output_samples": 48320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.032563579489195, "spectral_convergence": 0.3191501160049946, "spectral_convergence_status": "ok", "amplitude_floor": 1.6806044167073843e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.65915669974862, "spectral_convergence": 0.27753990050132, "spectral_convergence_status": "ok", "amplitude_floor": 1.0917469705551035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.834256018020606, "spectral_convergence": 0.2972371613539697, "spectral_convergence_status": "ok", "amplitude_floor": 1.0075473322757857e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0009427012142024427, "output_power": 0.000933931903757705, "delta_db": -0.04058855000088893, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024186117972299723, "output_power": 0.002340405010104689, "delta_db": -0.14275147760352916, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0014180970240603914, "output_power": 0.0014122823276292262, "delta_db": -0.017844209647511017, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023968394187503082, "output_power": 0.00021150856179790107, "delta_db": -0.5431098637831617, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.504236566794763e-05, "output_power": 5.6607033341799416e-05, "delta_db": 0.12173303812557898, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00019111878985838765, "output_power": 0.00016598775112969065, "delta_db": -0.6122734587570068, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24638343298627943, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12088290230473661, "pitch": null, "output_rms_dbfs": -22.906687320683844, "output_peak": 0.5561347830141948, "reference_rms_dbfs": -22.785804418379104, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8419920990861405}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00023057786100875665, "log_rms_mae_db": 1.7684721306981401, "rms_envelope_nrmse": 0.08625284546091547, "envelope_correlation": 0.9927713047080733, "rms_gain_error_db": -0.12111122662194498, "reference_mean_power": 0.005273848963429562, "output_mean_power": 0.0051288094645655785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.9211277816293244e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014051118338624634, "output_power": 0.001389612524352809, "reference_active": true, "delta_db": -0.04817171744646868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001364841166272218, "output_power": 0.0001501595812398148, "reference_active": true, "delta_db": 0.414709352787379}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.767297964659028e-05, "output_power": 3.316359137996562e-05, "reference_active": true, "delta_db": -0.5536841554310232}], "active_band_count": 3, "active_band_mae_db": 0.33885507522162367}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00020448142520679275, "log_rms_mae_db": 0.8159095486241686, "rms_envelope_nrmse": 0.04359281723760568, "envelope_correlation": 0.9981961298713675, "rms_gain_error_db": -0.12111405996552187, "reference_mean_power": 0.005325056781288949, "output_mean_power": 0.00517860560489038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.674469134143454e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013611842323655793, "output_power": 0.0013389392374710794, "reference_active": true, "delta_db": -0.07156040988921439}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.483051409804231e-05, "output_power": 7.28129965349805e-05, "reference_active": true, "delta_db": 0.504294387897566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6716298904493925e-06, "output_power": 1.709010163801894e-06, "reference_active": true, "delta_db": 0.09604517388800349}], "active_band_count": 3, "active_band_mae_db": 0.22396665722492795}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00020294603022145288, "log_rms_mae_db": 2.8338879406044035, "rms_envelope_nrmse": 0.09556994503940032, "envelope_correlation": 0.9940578539574854, "rms_gain_error_db": -0.0013401042240173888, "reference_mean_power": 0.002254450133685994, "output_mean_power": 0.0022537545844228467}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.5930281169536517e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007344814515063544, "output_power": 0.0007382425200819912, "reference_active": true, "delta_db": 0.022182227097710776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010026570767730039, "output_power": 0.0001001713519481879, "reference_active": true, "delta_db": -0.004088882125198996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.614259763381322e-05, "output_power": 1.8031255631085765e-05, "reference_active": true, "delta_db": 0.4805254867434309}], "active_band_count": 3, "active_band_mae_db": 0.16893219865544687}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00017314576936579135, "log_rms_mae_db": 2.063601785668921, "rms_envelope_nrmse": 0.05001359124121518, "envelope_correlation": 0.9984124459500747, "rms_gain_error_db": -0.0013547717474009797, "reference_mean_power": 0.002277104155750591, "output_mean_power": 0.002276393929076733}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.4781762283606311e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007359452368517945, "output_power": 0.0007420199759416943, "reference_active": true, "delta_db": 0.03570098266987332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.796294327084698e-05, "output_power": 4.733041633581395e-05, "reference_active": true, "delta_db": -0.05765501250925582}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4655779178142778e-06, "output_power": 1.3681457590456506e-06, "reference_active": true, "delta_db": -0.2987654419617601}], "active_band_count": 3, "active_band_mae_db": 0.13070714571362974}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 8.794013668187323e-05, "log_rms_mae_db": 1.803815493985327, "rms_envelope_nrmse": 0.1545758729705217, "envelope_correlation": 0.9844130217313815, "rms_gain_error_db": -0.4006394711123744, "reference_mean_power": 0.0002815001552852329, "output_mean_power": 0.00025669339364164137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.4039983283954285e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014419558923291274, "output_power": 0.000127364912134188, "reference_active": true, "delta_db": -0.5390217572052778}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9143053212884802e-05, "output_power": 1.3743912603119004e-05, "reference_active": true, "delta_db": -1.4390082173525505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.190405195331938e-06, "output_power": 3.1079875127848698e-06, "reference_active": true, "delta_db": -1.2977675429480182}], "active_band_count": 3, "active_band_mae_db": 1.0919325058352822}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 7.034799916708663e-05, "log_rms_mae_db": 1.2737609080478673, "rms_envelope_nrmse": 0.104195659422328, "envelope_correlation": 0.9934297945528445, "rms_gain_error_db": -0.4006670316073601, "reference_mean_power": 0.00028432562712779955, "output_mean_power": 0.00025926822985845275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.2325770901894182e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001384543016382475, "output_power": 0.00012101261897614897, "reference_active": true, "delta_db": -0.5847579327356799}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.792343358100607e-06, "output_power": 6.809340394098341e-06, "reference_active": true, "delta_db": -1.5778158834815699}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0751348414140544e-07, "output_power": 1.9677047950458184e-07, "reference_active": true, "delta_db": -1.9390421993988982}], "active_band_count": 3, "active_band_mae_db": 1.3672053385387162}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 5.049105990445221e-05, "log_rms_mae_db": 1.6894462141515998, "rms_envelope_nrmse": 0.15985903499518878, "envelope_correlation": 0.9839319492384072, "rms_gain_error_db": -0.4854395775589029, "reference_mean_power": 0.00010138440655269997, "output_mean_power": 9.066239878686724e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.3079247184162883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1617157093216335e-05, "output_power": 4.644542701968303e-05, "reference_active": true, "delta_db": -0.4585112118455371}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0705364401401438e-05, "output_power": 1.3759008266666648e-05, "reference_active": true, "delta_db": -1.7749574662979248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.974689061207133e-06, "output_power": 2.5811394942785316e-06, "reference_active": true, "delta_db": -0.6163010038075393}], "active_band_count": 3, "active_band_mae_db": 0.9499232273170004}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 4.0590764349045424e-05, "log_rms_mae_db": 1.1264470393297779, "rms_envelope_nrmse": 0.09541978284495206, "envelope_correlation": 0.9954229372899872, "rms_gain_error_db": -0.4854614435440857, "reference_mean_power": 0.00010240265187942684, "output_mean_power": 9.157249752318214e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.2076196938321023e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.154744016518151e-05, "output_power": 5.4974101458001233e-05, "reference_active": true, "delta_db": -0.49051855169577335}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.945675894844695e-06, "output_power": 6.6022363713630425e-06, "reference_active": true, "delta_db": -1.7794323377760892}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5378071346519667e-07, "output_power": 2.2105581340385605e-07, "reference_active": true, "delta_db": -0.5995667337843269}], "active_band_count": 3, "active_band_mae_db": 0.9565058744187298}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7868355369008211, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c827cc72321d803836c362d8834a09ed9d2d868259776ba800598a83067f5911", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9835868848568992, "standard_deviation": 0.0021618096256995566, "minimum": 0.97928344468194, "maximum": 0.9863641578758775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8054339622748037, "standard_deviation": 0.020655950843054324, "minimum": 1.7668375039801678, "maximum": 1.82957355627391, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9198430614379911, "standard_deviation": 0.41904331875967055, "minimum": 0.3674990969557821, "maximum": 1.647476648265143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15667606190385008, "standard_deviation": 0.010473439557010312, "minimum": 0.1420730523025336, "maximum": 0.1737294194760101, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9943290554543256, "standard_deviation": 0.0014391017181595146, "minimum": 0.9914220320174455, "maximum": 0.9963280917130667, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2336811841387485, "standard_deviation": 0.039164236322252514, "minimum": 1.160168310984324, "maximum": 1.2808596808436221, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.001766092525604, "standard_deviation": 0.5305109172831239, "minimum": 0.31414021464254954, "maximum": 1.8749302947973312, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09451123859691585, "standard_deviation": 0.014430738867566285, "minimum": 0.07635593175496491, "maximum": 0.12428684071923655, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9784190919392972, "standard_deviation": 0.0056191438502255515, "minimum": 0.9703633217778634, "maximum": 0.9843435171111387, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.702548811221643, "standard_deviation": 0.04103446720742649, "minimum": 1.63635296819423, "maximum": 1.7674015484911494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9124103481799899, "standard_deviation": 0.3924423213052563, "minimum": 0.2536502494542197, "maximum": 1.3538668992439167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17803949001010666, "standard_deviation": 0.023087535276638697, "minimum": 0.15163650277706478, "maximum": 0.2090902479061562, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932710231093135, "standard_deviation": 0.002286947719227552, "minimum": 0.9892157295922812, "maximum": 0.9955337075376635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1041389817224092, "standard_deviation": 0.052060632450044976, "minimum": 1.028804547515415, "maximum": 1.177965729732743, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9215492169460254, "standard_deviation": 0.47646721014819865, "minimum": 0.1644973825910166, "maximum": 1.6792126308957107, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1000818147183033, "standard_deviation": 0.01674847797795672, "minimum": 0.0788046208712181, "maximum": 0.12629271385563326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933173883157815, "standard_deviation": 0.0010533025413988876, "minimum": 0.9915486491065586, "maximum": 0.994601512872577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.6888460392537876, "standard_deviation": 0.08462110926217786, "minimum": 2.549648065645066, "maximum": 2.8338879406044035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47745065444500834, "standard_deviation": 0.2544414613890563, "minimum": 0.16893219865544687, "maximum": 0.7898820662830218, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10148645369590295, "standard_deviation": 0.00816984034695736, "minimum": 0.09092251576061455, "maximum": 0.11517260264797036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99794412556266, "standard_deviation": 0.0007164285483648114, "minimum": 0.9965140320422706, "maximum": 0.9986783933184779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.9827115105786324, "standard_deviation": 0.10636075903850237, "minimum": 1.7715850435727043, "maximum": 2.1379461100723, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44638975732838027, "standard_deviation": 0.19032779413941747, "minimum": 0.13070714571362974, "maximum": 0.7992116206308045, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05677210260623359, "standard_deviation": 0.009224513364913955, "minimum": 0.045497538861594994, "maximum": 0.07389269485517816, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17984352685211422, "standard_deviation": 0.05839769773627958, "minimum": 0.10799831516086443, "maximum": 0.2676783669153016, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.840081700219736, "standard_deviation": 0.007012800797794469, "minimum": 5.830284862297901, "maximum": 5.851120740843418, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.564979156075355, "standard_deviation": 0.021534056522190908, "minimum": 0.5366437497223129, "maximum": 0.6001946039129424, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09243385430267606, "standard_deviation": 0.06346931972209716, "minimum": -0.1781525744440886, "maximum": 0.010838826171504144, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.991566303535959, "standard_deviation": 0.0005886145434826705, "minimum": 0.9909638212477773, "maximum": 0.9927713047080733, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.73766719313384, "standard_deviation": 0.037220981332081564, "minimum": 1.673759934339921, "maximum": 1.7965506444284096, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.318454916782713, "standard_deviation": 0.11204493325290728, "minimum": 0.11514126178629482, "maximum": 0.4296192005592703, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09257118708988807, "standard_deviation": 0.003242886578501019, "minimum": 0.08625284546091547, "maximum": 0.09550975333153028, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977781994518709, "standard_deviation": 0.00036143978855409384, "minimum": 0.9970526869102749, "maximum": 0.9981961298713675, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7945402778096844, "standard_deviation": 0.04523901929746753, "minimum": 0.7136429242940017, "maximum": 0.8428394993100424, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2545506349488116, "standard_deviation": 0.16655408960535836, "minimum": 0.054790746304605294, "maximum": 0.5181972088258969, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04730511868324452, "standard_deviation": 0.0034354713627198423, "minimum": 0.04359281723760568, "maximum": 0.0532420076405881, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0042", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.2162296772003174, "predicted_mos": 2.2162296772003174, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "288f2b59ccdad3971a38d12ee54ed9447ac6729149b10e16eb8bb2e35b9e946b", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8717026002705097, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7466728648878098, "全帯域_10ms_rms_envelope_nrmse": 0.08582874412121064, "全帯域_10ms_envelope_correlation": 0.9928300604714005, "全帯域_10ms_modulation_mae_db": 0.38766129574178526, "全帯域_40ms_log_rms_mae_db": 0.8312161554787698, "全帯域_40ms_rms_envelope_nrmse": 0.042999210424827745, "全帯域_40ms_envelope_correlation": 0.9982225790607652, "全帯域_40ms_modulation_mae_db": 0.2435197987517904, "300–1000 Hz_10ms_log_rms_mae_db": 2.5478423381296986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10079745190429167, "300–1000 Hz_10ms_envelope_correlation": 0.9933938841628468, "300–1000 Hz_10ms_modulation_mae_db": 0.25331547546932337, "300–1000 Hz_40ms_log_rms_mae_db": 1.696687784661483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053917318163021415, "300–1000 Hz_40ms_envelope_correlation": 0.9981494066720795, "300–1000 Hz_40ms_modulation_mae_db": 0.14186326762657062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6866534339534278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15886099727039277, "1000–3000 Hz_10ms_envelope_correlation": 0.9837856045284823, "1000–3000 Hz_10ms_modulation_mae_db": 1.0733984917819683, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1174344469346937, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11065190248827797, "1000–3000 Hz_40ms_envelope_correlation": 0.9928146942930907, "1000–3000 Hz_40ms_modulation_mae_db": 1.3334083266245418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.536338332609064, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16007973574994408, "3000–6000 Hz_10ms_envelope_correlation": 0.9834967839754499, "3000–6000 Hz_10ms_modulation_mae_db": 0.7154587424124688, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8829447703329112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0933597311330155, "3000–6000 Hz_40ms_envelope_correlation": 0.9953121379826504, "3000–6000 Hz_40ms_modulation_mae_db": 0.8105791240906797, "mr_stft_db": 5.777753738239571, "band_mae_db": 0.22818155090801603, "rms_gain_error_db": -0.10988851551434697, "output_peak": 0.5559392858967748}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.752452759531447, "全帯域_10ms_rms_envelope_nrmse": 0.08917241824958609, "全帯域_10ms_envelope_correlation": 0.9920872421513616, "全帯域_10ms_modulation_mae_db": 0.4346264152221773, "全帯域_40ms_log_rms_mae_db": 0.8276014882845669, "全帯域_40ms_rms_envelope_nrmse": 0.04473713442222712, "全帯域_40ms_envelope_correlation": 0.9978849106003413, "全帯域_40ms_modulation_mae_db": 0.30362111480818427, "300–1000 Hz_10ms_log_rms_mae_db": 2.5277843782561558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09136505607805462, "300–1000 Hz_10ms_envelope_correlation": 0.9945512301650911, "300–1000 Hz_10ms_modulation_mae_db": 0.18441468440921316, "300–1000 Hz_40ms_log_rms_mae_db": 1.8326491554794557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04613717656650343, "300–1000 Hz_40ms_envelope_correlation": 0.9985967695719554, "300–1000 Hz_40ms_modulation_mae_db": 0.4388509510694029, "1000–3000 Hz_10ms_log_rms_mae_db": 1.669615899150451, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1746892357895033, "1000–3000 Hz_10ms_envelope_correlation": 0.9789879364857518, "1000–3000 Hz_10ms_modulation_mae_db": 0.839673254527828, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9301338852559047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08972856957491418, "1000–3000 Hz_40ms_envelope_correlation": 0.9944694580556513, "1000–3000 Hz_40ms_modulation_mae_db": 0.6824638974078016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5330382271484995, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20422748872317717, "3000–6000 Hz_10ms_envelope_correlation": 0.9716292579351045, "3000–6000 Hz_10ms_modulation_mae_db": 0.961498264818751, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8738208622405315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10797426440572347, "3000–6000 Hz_40ms_envelope_correlation": 0.9926095002754376, "3000–6000 Hz_40ms_modulation_mae_db": 1.379883501454394, "mr_stft_db": 5.785482762334309, "band_mae_db": 0.10320947517077939, "rms_gain_error_db": -0.021434965913634106, "output_peak": 0.5465585471423915}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7452518741700027, "全帯域_10ms_rms_envelope_nrmse": 0.09251945367688778, "全帯域_10ms_envelope_correlation": 0.991512749388149, "全帯域_10ms_modulation_mae_db": 0.10408652385917738, "全帯域_40ms_log_rms_mae_db": 0.7787811763219136, "全帯域_40ms_rms_envelope_nrmse": 0.04395477579578223, "全帯域_40ms_envelope_correlation": 0.9980463537017368, "全帯域_40ms_modulation_mae_db": 0.0896638852991604, "300–1000 Hz_10ms_log_rms_mae_db": 2.3023691084670768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09572671984308331, "300–1000 Hz_10ms_envelope_correlation": 0.9941194027770047, "300–1000 Hz_10ms_modulation_mae_db": 0.6623773817982924, "300–1000 Hz_40ms_log_rms_mae_db": 1.5071287283076067, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04939467493874522, "300–1000 Hz_40ms_envelope_correlation": 0.9985914852372295, "300–1000 Hz_40ms_modulation_mae_db": 0.506725288906387, "1000–3000 Hz_10ms_log_rms_mae_db": 1.654008776401875, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1541120489794744, "1000–3000 Hz_10ms_envelope_correlation": 0.9839148478236749, "1000–3000 Hz_10ms_modulation_mae_db": 0.29634629279462305, "1000–3000 Hz_40ms_log_rms_mae_db": 1.034226235111834, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08310108787833526, "1000–3000 Hz_40ms_envelope_correlation": 0.9956007738399558, "1000–3000 Hz_40ms_modulation_mae_db": 0.446709951666083, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5207086008042419, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17406873373965473, "3000–6000 Hz_10ms_envelope_correlation": 0.979358535545996, "3000–6000 Hz_10ms_modulation_mae_db": 0.7409956736614234, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9244380923693417, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10774294576454528, "3000–6000 Hz_40ms_envelope_correlation": 0.9919801903179098, "3000–6000 Hz_40ms_modulation_mae_db": 1.0232182844747875, "mr_stft_db": 5.786805744662215, "band_mae_db": 0.21609248004692938, "rms_gain_error_db": -0.09069848097157836, "output_peak": 0.5713292673381587}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6940271693535331, "全帯域_10ms_rms_envelope_nrmse": 0.09124037223913761, "全帯域_10ms_envelope_correlation": 0.9917524657207276, "全帯域_10ms_modulation_mae_db": 0.08505625390435427, "全帯域_40ms_log_rms_mae_db": 0.7165071926491945, "全帯域_40ms_rms_envelope_nrmse": 0.04734564253953709, "全帯域_40ms_envelope_correlation": 0.9976447178245467, "全帯域_40ms_modulation_mae_db": 0.024297404256394248, "300–1000 Hz_10ms_log_rms_mae_db": 2.498290455348523, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11323564463392269, "300–1000 Hz_10ms_envelope_correlation": 0.9918650920633525, "300–1000 Hz_10ms_modulation_mae_db": 0.20540740035398705, "300–1000 Hz_40ms_log_rms_mae_db": 1.7622960588882024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07490301602191253, "300–1000 Hz_40ms_envelope_correlation": 0.996434282894107, "300–1000 Hz_40ms_modulation_mae_db": 0.5116316996474881, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6772307831813396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14105486270997583, "1000–3000 Hz_10ms_envelope_correlation": 0.9868509402008143, "1000–3000 Hz_10ms_modulation_mae_db": 0.22693191969150237, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0500128559926822, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08041022038161764, "1000–3000 Hz_40ms_envelope_correlation": 0.9962549195428644, "1000–3000 Hz_40ms_modulation_mae_db": 0.3345251229540816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4734789297011943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16244193604783985, "3000–6000 Hz_10ms_envelope_correlation": 0.9819378533339601, "3000–6000 Hz_10ms_modulation_mae_db": 0.21408668847935516, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8245605194934469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08254203128763493, "3000–6000 Hz_40ms_envelope_correlation": 0.9951541888365828, "3000–6000 Hz_40ms_modulation_mae_db": 0.1230997813528434, "mr_stft_db": 5.773287226386123, "band_mae_db": 0.11765490671046212, "rms_gain_error_db": 0.00811796162462125, "output_peak": 0.5893682956975448}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.787526836501094, "全帯域_10ms_rms_envelope_nrmse": 0.09508259927079361, "全帯域_10ms_envelope_correlation": 0.9910816873770769, "全帯域_10ms_modulation_mae_db": 0.24267249857516024, "全帯域_40ms_log_rms_mae_db": 0.7661546435380712, "全帯域_40ms_rms_envelope_nrmse": 0.051754007036261554, "全帯域_40ms_envelope_correlation": 0.9972443421095459, "全帯域_40ms_modulation_mae_db": 0.11773893592640831, "300–1000 Hz_10ms_log_rms_mae_db": 2.4250071906193496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09761599095461487, "300–1000 Hz_10ms_envelope_correlation": 0.9938170198566981, "300–1000 Hz_10ms_modulation_mae_db": 0.37834464791339717, "300–1000 Hz_40ms_log_rms_mae_db": 1.6286906098594132, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057657398410905614, "300–1000 Hz_40ms_envelope_correlation": 0.997795096872602, "300–1000 Hz_40ms_modulation_mae_db": 0.8064795903244977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6540832526954778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.150894033950404, "1000–3000 Hz_10ms_envelope_correlation": 0.9844743295844878, "1000–3000 Hz_10ms_modulation_mae_db": 0.5110571914419451, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9987167862012086, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08685135412079802, "1000–3000 Hz_40ms_envelope_correlation": 0.9949110210880157, "1000–3000 Hz_40ms_modulation_mae_db": 0.5750978184905454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5996679939346952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19540072617420917, "3000–6000 Hz_10ms_envelope_correlation": 0.9743800796757716, "3000–6000 Hz_10ms_modulation_mae_db": 1.2759835567557274, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8723098586771737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10413357507355057, "3000–6000 Hz_40ms_envelope_correlation": 0.9935748077851402, "3000–6000 Hz_40ms_modulation_mae_db": 1.3102638615789364, "mr_stft_db": 5.785411138613482, "band_mae_db": 0.15632015844421587, "rms_gain_error_db": -0.05761982745442474, "output_peak": 0.6094563879027796}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.742843818221954, "全帯域_10ms_rms_envelope_nrmse": 0.09231299840283853, "全帯域_10ms_envelope_correlation": 0.991638573718404, "全帯域_10ms_modulation_mae_db": 0.30805430528496874, "全帯域_40ms_log_rms_mae_db": 0.7771405147777497, "全帯域_40ms_rms_envelope_nrmse": 0.04752370916794745, "全帯域_40ms_envelope_correlation": 0.99783991160645, "全帯域_40ms_modulation_mae_db": 0.27910442543004416, "300–1000 Hz_10ms_log_rms_mae_db": 2.426164038172639, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10197540639713443, "300–1000 Hz_10ms_envelope_correlation": 0.9933898845681477, "300–1000 Hz_10ms_modulation_mae_db": 0.7242364850490389, "300–1000 Hz_40ms_log_rms_mae_db": 1.7283147408081965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059975447147679446, "300–1000 Hz_40ms_envelope_correlation": 0.9978984988745198, "300–1000 Hz_40ms_modulation_mae_db": 0.49459261999932, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6518601253042338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1605473141696835, "1000–3000 Hz_10ms_envelope_correlation": 0.9826228757542459, "1000–3000 Hz_10ms_modulation_mae_db": 1.1660659851452229, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9956073318295727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09196919311187789, "1000–3000 Hz_40ms_envelope_correlation": 0.9946833601236497, "1000–3000 Hz_40ms_modulation_mae_db": 1.132182771998034, "3000–6000 Hz_10ms_log_rms_mae_db": 1.606104901059874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2008452543855697, "3000–6000 Hz_10ms_envelope_correlation": 0.9743118662659931, "3000–6000 Hz_10ms_modulation_mae_db": 1.5070273753468093, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9702187408008162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13762904168413223, "3000–6000 Hz_40ms_envelope_correlation": 0.9877827446405673, "3000–6000 Hz_40ms_modulation_mae_db": 0.9167296881325786, "mr_stft_db": 5.778626866460219, "band_mae_db": 0.19263956323686238, "rms_gain_error_db": -0.14039277977428632, "output_peak": 0.5590106340990864}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6868195551861225, "全帯域_10ms_rms_envelope_nrmse": 0.09344642780425982, "全帯域_10ms_envelope_correlation": 0.9914281079851053, "全帯域_10ms_modulation_mae_db": 0.3592197010081575, "全帯域_40ms_log_rms_mae_db": 0.8304272038114302, "全帯域_40ms_rms_envelope_nrmse": 0.04536586645652557, "全帯域_40ms_envelope_correlation": 0.9980315978662445, "全帯域_40ms_modulation_mae_db": 0.48014440210143167, "300–1000 Hz_10ms_log_rms_mae_db": 2.4860877415257137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10267425740544013, "300–1000 Hz_10ms_envelope_correlation": 0.9930900551262553, "300–1000 Hz_10ms_modulation_mae_db": 0.7366637948850933, "300–1000 Hz_40ms_log_rms_mae_db": 1.6500346238085253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04875140371391534, "300–1000 Hz_40ms_envelope_correlation": 0.9985208995540624, "300–1000 Hz_40ms_modulation_mae_db": 0.3512598597437619, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6953239082689637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14634424803137436, "1000–3000 Hz_10ms_envelope_correlation": 0.9853672047333992, "1000–3000 Hz_10ms_modulation_mae_db": 0.6316636211648408, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0603564487721762, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08761671074330793, "1000–3000 Hz_40ms_envelope_correlation": 0.9947142345155531, "1000–3000 Hz_40ms_modulation_mae_db": 0.9789575567919909, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5126285943876543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14443748625515115, "3000–6000 Hz_10ms_envelope_correlation": 0.9857503699213517, "3000–6000 Hz_10ms_modulation_mae_db": 1.093444149476485, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8742645934668808, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07849587019491659, "3000–6000 Hz_40ms_envelope_correlation": 0.9956469837192287, "3000–6000 Hz_40ms_modulation_mae_db": 0.9098057703646115, "mr_stft_db": 5.7900672436124525, "band_mae_db": 0.15584514761348578, "rms_gain_error_db": -0.1333503688706515, "output_peak": 0.5582359533259308}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6318818800411836, "全帯域_10ms_rms_envelope_nrmse": 0.09423020810825951, "全帯域_10ms_envelope_correlation": 0.991428460252032, "全帯域_10ms_modulation_mae_db": 0.3648681489378253, "全帯域_40ms_log_rms_mae_db": 0.7480461366147649, "全帯域_40ms_rms_envelope_nrmse": 0.048505548059275774, "全帯域_40ms_envelope_correlation": 0.9979144654627126, "全帯域_40ms_modulation_mae_db": 0.5503515131943423, "300–1000 Hz_10ms_log_rms_mae_db": 2.4875623281792176, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11127083790090879, "300–1000 Hz_10ms_envelope_correlation": 0.9919535267629583, "300–1000 Hz_10ms_modulation_mae_db": 0.5157165409604891, "300–1000 Hz_40ms_log_rms_mae_db": 1.69586156466607, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060167605348418265, "300–1000 Hz_40ms_envelope_correlation": 0.9977670993931956, "300–1000 Hz_40ms_modulation_mae_db": 0.34796886026923673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6370183232492508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15745448537845294, "1000–3000 Hz_10ms_envelope_correlation": 0.984457231876369, "1000–3000 Hz_10ms_modulation_mae_db": 1.5762990577713836, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0177728742409158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10965210619729261, "1000–3000 Hz_40ms_envelope_correlation": 0.9934167456550042, "1000–3000 Hz_40ms_modulation_mae_db": 1.7126639400250587, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4804278050927537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.153516726933528, "3000–6000 Hz_10ms_envelope_correlation": 0.9839303769468117, "3000–6000 Hz_10ms_modulation_mae_db": 0.22520347951203554, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8494970968923086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0848355625828974, "3000–6000 Hz_40ms_envelope_correlation": 0.9950008118977273, "3000–6000 Hz_40ms_modulation_mae_db": 0.11854409612813571, "mr_stft_db": 5.7760262563859, "band_mae_db": 0.25914502602519324, "rms_gain_error_db": -0.17698343176524733, "output_peak": 0.5368989620472666}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48320, "output_samples": 48320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.981057851321068, "spectral_convergence": 0.32018458687812035, "spectral_convergence_status": "ok", "amplitude_floor": 1.6806044167073843e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.590256161749404, "spectral_convergence": 0.2775323921688711, "spectral_convergence_status": "ok", "amplitude_floor": 1.0917469705551035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.761947201648242, "spectral_convergence": 0.2974574988582244, "spectral_convergence_status": "ok", "amplitude_floor": 1.0075473322757857e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0009427012142024427, "output_power": 0.0009325391694649223, "delta_db": -0.04706983909291595, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024186117972299723, "output_power": 0.0023555604818035324, "delta_db": -0.11471907151878567, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0014180970240603914, "output_power": 0.0014119504899652022, "delta_db": -0.018864771876932675, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023968394187503082, "output_power": 0.0002072861755302105, "delta_db": -0.6306859974640545, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.504236566794763e-05, "output_power": 5.584136339561251e-05, "delta_db": 0.06258922079937383, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00019111878985838765, "output_power": 0.0001705247205038687, "delta_db": -0.4951604046960335, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22818155090801603, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10988851551434697, "pitch": null, "output_rms_dbfs": -22.895692933893454, "output_peak": 0.5559392858967748, "reference_rms_dbfs": -22.785804418379104, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.777753738239571}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00023057786100875665, "log_rms_mae_db": 1.7466728648878098, "rms_envelope_nrmse": 0.08582874412121064, "envelope_correlation": 0.9928300604714005, "rms_gain_error_db": -0.11018115010277157, "reference_mean_power": 0.005273848963429562, "output_mean_power": 0.005141733617093455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.9211277816293244e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014051118338624634, "output_power": 0.0014016347802882183, "reference_active": true, "delta_db": -0.010760258481833682}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001364841166272218, "output_power": 0.0001509026246554539, "reference_active": true, "delta_db": 0.43614680352939056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.767297964659028e-05, "output_power": 3.1946426723212615e-05, "reference_active": true, "delta_db": -0.7160768252141315}], "active_band_count": 3, "active_band_mae_db": 0.38766129574178526}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00020448142520679275, "log_rms_mae_db": 0.8312161554787698, "rms_envelope_nrmse": 0.042999210424827745, "envelope_correlation": 0.9982225790607652, "rms_gain_error_db": -0.11015102427025075, "reference_mean_power": 0.005325056781288949, "output_mean_power": 0.00519169463964873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.674469134143454e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013611842323655793, "output_power": 0.0013495929123005933, "reference_active": true, "delta_db": -0.03714120899805674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.483051409804231e-05, "output_power": 7.302377677922697e-05, "reference_active": true, "delta_db": 0.5168482532404705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6716298904493925e-06, "output_power": 1.7409933873122021e-06, "reference_active": true, "delta_db": 0.176569934016844}], "active_band_count": 3, "active_band_mae_db": 0.2435197987517904}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00020294603022145288, "log_rms_mae_db": 2.5478423381296986, "rms_envelope_nrmse": 0.10079745190429167, "envelope_correlation": 0.9933938841628468, "rms_gain_error_db": 0.013343247888300467, "reference_mean_power": 0.002254450133685994, "output_mean_power": 0.002261387349569893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.5930281169536517e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007344814515063544, "output_power": 0.0007430211298132112, "reference_active": true, "delta_db": 0.05020331621497966}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010026570767730039, "output_power": 9.931208228649167e-05, "reference_active": true, "delta_db": -0.041503355363004896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.614259763381322e-05, "output_power": 1.8827705532555672e-05, "reference_active": true, "delta_db": 0.6682397548299855}], "active_band_count": 3, "active_band_mae_db": 0.25331547546932337}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00017314576936579135, "log_rms_mae_db": 1.696687784661483, "rms_envelope_nrmse": 0.053917318163021415, "envelope_correlation": 0.9981494066720795, "rms_gain_error_db": 0.013332523556444952, "reference_mean_power": 0.002277104155750591, "output_mean_power": 0.002284105440496493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.4781762283606311e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007359452368517945, "output_power": 0.0007463678331277903, "reference_active": true, "delta_db": 0.06107415145780502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.796294327084698e-05, "output_power": 4.664119559850917e-05, "reference_active": true, "delta_db": -0.12136151369586759}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4655779178142778e-06, "output_power": 1.3857774786280923e-06, "reference_active": true, "delta_db": -0.2431541377260392}], "active_band_count": 3, "active_band_mae_db": 0.14186326762657062}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 8.794013668187323e-05, "log_rms_mae_db": 1.6866534339534278, "rms_envelope_nrmse": 0.15886099727039277, "envelope_correlation": 0.9837856045284823, "rms_gain_error_db": -0.46671103954112947, "reference_mean_power": 0.0002815001552852329, "output_mean_power": 0.0002528177342142549}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.4039983283954285e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014419558923291274, "output_power": 0.00012565286844391015, "reference_active": true, "delta_db": -0.5977956879319339}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9143053212884802e-05, "output_power": 1.3653889592608148e-05, "reference_active": true, "delta_db": -1.46754819813654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.190405195331938e-06, "output_power": 3.2119654168860983e-06, "reference_active": true, "delta_db": -1.1548515892774314}], "active_band_count": 3, "active_band_mae_db": 1.0733984917819683}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 7.034799916708663e-05, "log_rms_mae_db": 1.1174344469346937, "rms_envelope_nrmse": 0.11065190248827797, "envelope_correlation": 0.9928146942930907, "rms_gain_error_db": -0.46671804247358883, "reference_mean_power": 0.00028432562712779955, "output_mean_power": 0.0002553549032582922}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.2325770901894182e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001384543016382475, "output_power": 0.00011930318868697503, "reference_active": true, "delta_db": -0.6465440192005355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.792343358100607e-06, "output_power": 6.748744019114004e-06, "reference_active": true, "delta_db": -1.616636775442723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0751348414140544e-07, "output_power": 2.0613880085918015e-07, "reference_active": true, "delta_db": -1.7370441852303662}], "active_band_count": 3, "active_band_mae_db": 1.3334083266245418}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 5.049105990445221e-05, "log_rms_mae_db": 1.536338332609064, "rms_envelope_nrmse": 0.16007973574994408, "envelope_correlation": 0.9834967839754499, "rms_gain_error_db": -0.45005865370205006, "reference_mean_power": 0.00010138440655269997, "output_mean_power": 9.1404020284611e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.3079247184162883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1617157093216335e-05, "output_power": 4.899662884315728e-05, "reference_active": true, "delta_db": -0.2262788140228581}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0705364401401438e-05, "output_power": 1.4382074018092723e-05, "reference_active": true, "delta_db": -1.5826135867339965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.974689061207133e-06, "output_power": 2.752283672684405e-06, "reference_active": true, "delta_db": -0.3374838264805516}], "active_band_count": 3, "active_band_mae_db": 0.7154587424124688}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 4.0590764349045424e-05, "log_rms_mae_db": 0.8829447703329112, "rms_envelope_nrmse": 0.0933597311330155, "envelope_correlation": 0.9953121379826504, "rms_gain_error_db": -0.450065066280124, "reference_mean_power": 0.00010240265187942684, "output_mean_power": 9.232189216588158e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.2076196938321023e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.154744016518151e-05, "output_power": 5.799578189742437e-05, "reference_active": true, "delta_db": -0.25813586839153696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.945675894844695e-06, "output_power": 6.915568379523966e-06, "reference_active": true, "delta_db": -1.5780642265375324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5378071346519667e-07, "output_power": 2.212610078411442e-07, "reference_active": true, "delta_db": -0.5955372773429699}], "active_band_count": 3, "active_band_mae_db": 0.8105791240906797}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7909098351240533, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b5a78f53df5d2dc2984f297092849564d9940c316442423915895103031ff495", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9838076213734033, "standard_deviation": 0.0023054454245115498, "minimum": 0.9789879364857518, "maximum": 0.9868509402008143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6657243127756274, "standard_deviation": 0.019823236990963028, "minimum": 1.6370183232492508, "maximum": 1.6953239082689637, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7901794767899142, "standard_deviation": 0.4638492455388225, "minimum": 0.22693191969150237, "maximum": 1.5762990577713836, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15549465328490764, "standard_deviation": 0.0101731393825952, "minimum": 0.14105486270997583, "maximum": 0.1746892357895033, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994608150889223, "standard_deviation": 0.0010994126054652527, "minimum": 0.9928146942930907, "maximum": 0.9962549195428644, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0255326080423735, "standard_deviation": 0.05491454374322321, "minimum": 0.9301338852559047, "maximum": 1.1174344469346937, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8995011732447671, "standard_deviation": 0.4760737772560652, "minimum": 0.3345251229540816, "maximum": 1.7126639400250587, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09249764306205269, "standard_deviation": 0.011476092438409592, "minimum": 0.08041022038161764, "maximum": 0.11065190248827797, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793493904500549, "standard_deviation": 0.005285157353906525, "minimum": 0.9716292579351045, "maximum": 0.9857503699213517, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5327991730922472, "standard_deviation": 0.04880345039952485, "minimum": 1.4734789297011943, "maximum": 1.606104901059874, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8417122413078819, "standard_deviation": 0.46430195054157375, "minimum": 0.21408668847935516, "maximum": 1.5070273753468093, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17437726100113424, "standard_deviation": 0.02302822879264872, "minimum": 0.14443748625515115, "maximum": 0.20422748872317717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933826706819056, "standard_deviation": 0.0026340475954604423, "minimum": 0.9877827446405673, "maximum": 0.9956469837192287, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8840068167841764, "standard_deviation": 0.0449262542927507, "minimum": 0.8245605194934469, "maximum": 0.9702187408008162, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8240155134471208, "standard_deviation": 0.4766276814719223, "minimum": 0.11854409612813571, "maximum": 1.379883501454394, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.099589127765802, "standard_deviation": 0.019259448137680737, "minimum": 0.07849587019491659, "maximum": 0.13762904168413223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932725119352943, "standard_deviation": 0.0009585095502890029, "minimum": 0.9918650920633525, "maximum": 0.9945512301650911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.462638447337297, "standard_deviation": 0.07780799189471693, "minimum": 2.3023691084670768, "maximum": 2.5478423381296986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4575595513548543, "standard_deviation": 0.23315470458271031, "minimum": 0.18441468440921316, "maximum": 0.7366637948850933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10183267063968132, "standard_deviation": 0.0074214083577061616, "minimum": 0.09136505607805462, "maximum": 0.11323564463392269, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979691923837188, "standard_deviation": 0.0007121451549083818, "minimum": 0.996434282894107, "maximum": 0.9985967695719554, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6877079083098692, "standard_deviation": 0.09701145043342838, "minimum": 1.5071287283076067, "maximum": 1.8326491554794557, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44992151719833307, "standard_deviation": 0.18947856701457852, "minimum": 0.14186326762657062, "maximum": 0.8064795903244977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05636300503888766, "standard_deviation": 0.009180532464429743, "minimum": 0.04613717656650343, "maximum": 0.07490301602191253, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17863603851949303, "standard_deviation": 0.05472114593956487, "minimum": 0.10320947517077939, "maximum": 0.25914502602519324, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.781682622086784, "standard_deviation": 0.006001691155586256, "minimum": 5.773287226386123, "maximum": 5.7900672436124525, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5658496666812416, "standard_deviation": 0.02359765517023832, "minimum": 0.5368989620472666, "maximum": 0.6094563879027796, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0902813010799435, "standard_deviation": 0.06297049863187473, "minimum": -0.17698343176524733, "maximum": 0.00811796162462125, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9917199183830321, "standard_deviation": 0.0005339044123475167, "minimum": 0.9910816873770769, "maximum": 0.9928300604714005, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7234345947366434, "standard_deviation": 0.04917890886099073, "minimum": 1.6318818800411836, "maximum": 1.787526836501094, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2857806428167008, "standard_deviation": 0.13084359343649135, "minimum": 0.08505625390435427, "maximum": 0.4346264152221773, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0917291527341217, "standard_deviation": 0.003000142522004109, "minimum": 0.08582874412121064, "maximum": 0.09508259927079361, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978536097790429, "standard_deviation": 0.00029873113738528067, "minimum": 0.9972443421095459, "maximum": 0.9982225790607652, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7844843139345576, "standard_deviation": 0.04229955865092725, "minimum": 0.7165071926491945, "maximum": 0.8312161554787698, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.26105518497096947, "standard_deviation": 0.1852050390368784, "minimum": 0.024297404256394248, "maximum": 0.5503515131943423, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04652323673779807, "standard_deviation": 0.0028413586536725263, "minimum": 0.042999210424827745, "maximum": 0.051754007036261554, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0042", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5028038024902344, "predicted_mos": 2.5028038024902344, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "17467180e29a1a2af3df89341056b02f68725059866718f4fc4a4d49ffc08784", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.4586575999855995, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6327843263323516, "全帯域_10ms_rms_envelope_nrmse": 0.08476675653950026, "全帯域_10ms_envelope_correlation": 0.9929627734209389, "全帯域_10ms_modulation_mae_db": 0.5142710125971607, "全帯域_40ms_log_rms_mae_db": 0.7376874749365954, "全帯域_40ms_rms_envelope_nrmse": 0.044151238533544114, "全帯域_40ms_envelope_correlation": 0.9981320226925924, "全帯域_40ms_modulation_mae_db": 0.5429874706178577, "300–1000 Hz_10ms_log_rms_mae_db": 3.2194074901140493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09451079270340723, "300–1000 Hz_10ms_envelope_correlation": 0.9943047928780386, "300–1000 Hz_10ms_modulation_mae_db": 1.1157167342313075, "300–1000 Hz_40ms_log_rms_mae_db": 2.4340715690557984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04860321551121798, "300–1000 Hz_40ms_envelope_correlation": 0.9985548911182659, "300–1000 Hz_40ms_modulation_mae_db": 1.6431009619768637, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2029892519276384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1533603911200396, "1000–3000 Hz_10ms_envelope_correlation": 0.9843894595360678, "1000–3000 Hz_10ms_modulation_mae_db": 0.6835960972255101, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4860193879375363, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08891310784158371, "1000–3000 Hz_40ms_envelope_correlation": 0.9952759944987427, "1000–3000 Hz_40ms_modulation_mae_db": 0.9349229155725912, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2370937753977396, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20777404963098453, "3000–6000 Hz_10ms_envelope_correlation": 0.9713470499069812, "3000–6000 Hz_10ms_modulation_mae_db": 0.861429583940713, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3393120396307994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10441685840126785, "3000–6000 Hz_40ms_envelope_correlation": 0.9925041936838677, "3000–6000 Hz_40ms_modulation_mae_db": 0.8402761908611204, "mr_stft_db": 6.118272097962202, "band_mae_db": 0.12408777389282771, "rms_gain_error_db": -0.11888611546811186, "output_peak": 0.5633865983057829}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6059047184647146, "全帯域_10ms_rms_envelope_nrmse": 0.08728777117807794, "全帯域_10ms_envelope_correlation": 0.9924341408923363, "全帯域_10ms_modulation_mae_db": 0.17327365782725793, "全帯域_40ms_log_rms_mae_db": 0.7181863185745888, "全帯域_40ms_rms_envelope_nrmse": 0.04486936324660382, "全帯域_40ms_envelope_correlation": 0.9979436191561194, "全帯域_40ms_modulation_mae_db": 0.2925117595802111, "300–1000 Hz_10ms_log_rms_mae_db": 3.0061652839409745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0979181177251806, "300–1000 Hz_10ms_envelope_correlation": 0.99379885624597, "300–1000 Hz_10ms_modulation_mae_db": 0.4432208747546013, "300–1000 Hz_40ms_log_rms_mae_db": 2.268895200629511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05764189909995711, "300–1000 Hz_40ms_envelope_correlation": 0.9978457749802068, "300–1000 Hz_40ms_modulation_mae_db": 0.6146385584450337, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2328468362839957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1607428389645972, "1000–3000 Hz_10ms_envelope_correlation": 0.9825368647997368, "1000–3000 Hz_10ms_modulation_mae_db": 1.0013715173832154, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5125060206690688, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09332954370178138, "1000–3000 Hz_40ms_envelope_correlation": 0.9943064871091771, "1000–3000 Hz_40ms_modulation_mae_db": 1.0839559602483388, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2532853837676847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20381438121546946, "3000–6000 Hz_10ms_envelope_correlation": 0.9721878035503074, "3000–6000 Hz_10ms_modulation_mae_db": 0.6117630066048202, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4172675192365314, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11682456175500716, "3000–6000 Hz_40ms_envelope_correlation": 0.9903781931602019, "3000–6000 Hz_40ms_modulation_mae_db": 0.6149366351928217, "mr_stft_db": 6.086300351356713, "band_mae_db": 0.09487119636893572, "rms_gain_error_db": -0.07691769932278794, "output_peak": 0.5398265266877491}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6577944752687288, "全帯域_10ms_rms_envelope_nrmse": 0.08401813231213857, "全帯域_10ms_envelope_correlation": 0.993037794234297, "全帯域_10ms_modulation_mae_db": 0.2886884357119887, "全帯域_40ms_log_rms_mae_db": 0.7111933328763856, "全帯域_40ms_rms_envelope_nrmse": 0.04229064476056087, "全帯域_40ms_envelope_correlation": 0.9982132929893985, "全帯域_40ms_modulation_mae_db": 0.13320593560481522, "300–1000 Hz_10ms_log_rms_mae_db": 2.831825406050438, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07677740507556827, "300–1000 Hz_10ms_envelope_correlation": 0.9961797201766517, "300–1000 Hz_10ms_modulation_mae_db": 0.3909206175485296, "300–1000 Hz_40ms_log_rms_mae_db": 2.111524522714291, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04371435420244862, "300–1000 Hz_40ms_envelope_correlation": 0.9987560599604419, "300–1000 Hz_40ms_modulation_mae_db": 0.6068649924028754, "1000–3000 Hz_10ms_log_rms_mae_db": 2.231597781275806, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1875692889720046, "1000–3000 Hz_10ms_envelope_correlation": 0.9757277266630318, "1000–3000 Hz_10ms_modulation_mae_db": 0.8548177936781644, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4902887787745611, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08470024786172314, "1000–3000 Hz_40ms_envelope_correlation": 0.9949748752428724, "1000–3000 Hz_40ms_modulation_mae_db": 0.3708183752201555, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2203495204822397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.208556908007705, "3000–6000 Hz_10ms_envelope_correlation": 0.9706574132359385, "3000–6000 Hz_10ms_modulation_mae_db": 2.2147139920308687, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4137754114183883, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12172648124188387, "3000–6000 Hz_40ms_envelope_correlation": 0.9902987541706668, "3000–6000 Hz_40ms_modulation_mae_db": 2.215795410463662, "mr_stft_db": 6.0962730989172185, "band_mae_db": 0.1425847302322355, "rms_gain_error_db": -0.08161229536988981, "output_peak": 0.546993991100552}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.759420069672387, "全帯域_10ms_rms_envelope_nrmse": 0.08715956019177776, "全帯域_10ms_envelope_correlation": 0.9924475083090339, "全帯域_10ms_modulation_mae_db": 0.17393394176817586, "全帯域_40ms_log_rms_mae_db": 0.744275102352744, "全帯域_40ms_rms_envelope_nrmse": 0.04219742208173556, "全帯域_40ms_envelope_correlation": 0.9981311762619524, "全帯域_40ms_modulation_mae_db": 0.08916375578086844, "300–1000 Hz_10ms_log_rms_mae_db": 3.050291059978424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07879856521441958, "300–1000 Hz_10ms_envelope_correlation": 0.9960511598679032, "300–1000 Hz_10ms_modulation_mae_db": 0.4450520630732642, "300–1000 Hz_40ms_log_rms_mae_db": 2.2640599652892957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04178347835798527, "300–1000 Hz_40ms_envelope_correlation": 0.998946908035209, "300–1000 Hz_40ms_modulation_mae_db": 0.5692415987808376, "1000–3000 Hz_10ms_log_rms_mae_db": 2.151113446380769, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1512205471373872, "1000–3000 Hz_10ms_envelope_correlation": 0.9856469612385401, "1000–3000 Hz_10ms_modulation_mae_db": 0.9642790685371779, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5647025191138877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.101038120146239, "1000–3000 Hz_40ms_envelope_correlation": 0.9944477874972877, "1000–3000 Hz_40ms_modulation_mae_db": 1.5093862677735934, "3000–6000 Hz_10ms_log_rms_mae_db": 2.216636509345871, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2116492775643451, "3000–6000 Hz_10ms_envelope_correlation": 0.9702186670605911, "3000–6000 Hz_10ms_modulation_mae_db": 0.5241434456633844, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3534698381203996, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11648928378664256, "3000–6000 Hz_40ms_envelope_correlation": 0.9905842028263074, "3000–6000 Hz_40ms_modulation_mae_db": 0.6601169623279802, "mr_stft_db": 6.086258296952885, "band_mae_db": 0.165852279740099, "rms_gain_error_db": -0.04474824909896348, "output_peak": 0.5698791160862584}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.748289892719111, "全帯域_10ms_rms_envelope_nrmse": 0.08739922505576772, "全帯域_10ms_envelope_correlation": 0.9924932072052172, "全帯域_10ms_modulation_mae_db": 0.24150607946366567, "全帯域_40ms_log_rms_mae_db": 0.7822461213845759, "全帯域_40ms_rms_envelope_nrmse": 0.0417198890018022, "全帯域_40ms_envelope_correlation": 0.9982889519631472, "全帯域_40ms_modulation_mae_db": 0.2477548973590017, "300–1000 Hz_10ms_log_rms_mae_db": 2.943135656418409, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08218257840252567, "300–1000 Hz_10ms_envelope_correlation": 0.995708045228264, "300–1000 Hz_10ms_modulation_mae_db": 0.3066910015819038, "300–1000 Hz_40ms_log_rms_mae_db": 2.1511641317053507, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043120068075536935, "300–1000 Hz_40ms_envelope_correlation": 0.9989144683465585, "300–1000 Hz_40ms_modulation_mae_db": 0.5608326783355779, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2653820104894127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15911260248902484, "1000–3000 Hz_10ms_envelope_correlation": 0.9842779308897293, "1000–3000 Hz_10ms_modulation_mae_db": 0.6152691322458305, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5252294320386657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0984553713998141, "1000–3000 Hz_40ms_envelope_correlation": 0.9953678772461197, "1000–3000 Hz_40ms_modulation_mae_db": 1.0220236749476632, "3000–6000 Hz_10ms_log_rms_mae_db": 2.256539760613436, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20536626957469828, "3000–6000 Hz_10ms_envelope_correlation": 0.9714155607553008, "3000–6000 Hz_10ms_modulation_mae_db": 1.1041298096714214, "3000–6000 Hz_40ms_log_rms_mae_db": 1.460832015495249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10909909533263265, "3000–6000 Hz_40ms_envelope_correlation": 0.991490865100527, "3000–6000 Hz_40ms_modulation_mae_db": 1.368188174632432, "mr_stft_db": 6.0919451021246145, "band_mae_db": 0.22105472382058303, "rms_gain_error_db": -0.08859228290765757, "output_peak": 0.5541631398584548}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.63264649665022, "全帯域_10ms_rms_envelope_nrmse": 0.08387301526317252, "全帯域_10ms_envelope_correlation": 0.9929809496972042, "全帯域_10ms_modulation_mae_db": 0.13190431033551384, "全帯域_40ms_log_rms_mae_db": 0.7631920970976042, "全帯域_40ms_rms_envelope_nrmse": 0.04105860705306181, "全帯域_40ms_envelope_correlation": 0.9982196891955618, "全帯域_40ms_modulation_mae_db": 0.2017538351431086, "300–1000 Hz_10ms_log_rms_mae_db": 3.1680942659846085, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0857907591477344, "300–1000 Hz_10ms_envelope_correlation": 0.9952566000265897, "300–1000 Hz_10ms_modulation_mae_db": 0.32987869995811353, "300–1000 Hz_40ms_log_rms_mae_db": 2.355361459162212, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04650908545102878, "300–1000 Hz_40ms_envelope_correlation": 0.998608840566428, "300–1000 Hz_40ms_modulation_mae_db": 0.13825276758565025, "1000–3000 Hz_10ms_log_rms_mae_db": 2.261495791716818, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17548198617909674, "1000–3000 Hz_10ms_envelope_correlation": 0.9795673277738195, "1000–3000 Hz_10ms_modulation_mae_db": 0.6820881007526457, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5266537810604843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10366286616757263, "1000–3000 Hz_40ms_envelope_correlation": 0.9936583647241798, "1000–3000 Hz_40ms_modulation_mae_db": 1.0030170196646655, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1761307345385594, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19990882311654398, "3000–6000 Hz_10ms_envelope_correlation": 0.9733547113508868, "3000–6000 Hz_10ms_modulation_mae_db": 1.1408885971310105, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3749051665460041, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11552467125423677, "3000–6000 Hz_40ms_envelope_correlation": 0.9906666320319557, "3000–6000 Hz_40ms_modulation_mae_db": 0.8778406712318726, "mr_stft_db": 6.095402686288462, "band_mae_db": 0.15899318624753514, "rms_gain_error_db": -0.031522368729892516, "output_peak": 0.567371092276354}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7120130953231072, "全帯域_10ms_rms_envelope_nrmse": 0.10212616522930641, "全帯域_10ms_envelope_correlation": 0.9895656898439646, "全帯域_10ms_modulation_mae_db": 0.12273407607996646, "全帯域_40ms_log_rms_mae_db": 0.9120199645235465, "全帯域_40ms_rms_envelope_nrmse": 0.05265038500224752, "全帯域_40ms_envelope_correlation": 0.9971094185175107, "全帯域_40ms_modulation_mae_db": 0.16732329381572697, "300–1000 Hz_10ms_log_rms_mae_db": 3.0890851834216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10827038246817242, "300–1000 Hz_10ms_envelope_correlation": 0.9922854396408275, "300–1000 Hz_10ms_modulation_mae_db": 0.358076582450465, "300–1000 Hz_40ms_log_rms_mae_db": 2.3033334847911653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06346184189165141, "300–1000 Hz_40ms_envelope_correlation": 0.997309553408359, "300–1000 Hz_40ms_modulation_mae_db": 0.38820129191320807, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1647384615961625, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1356026181786775, "1000–3000 Hz_10ms_envelope_correlation": 0.9875656908851241, "1000–3000 Hz_10ms_modulation_mae_db": 0.38231376960725605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4788679154221647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07626864574524024, "1000–3000 Hz_40ms_envelope_correlation": 0.9959888378387217, "1000–3000 Hz_40ms_modulation_mae_db": 0.563154521434408, "3000–6000 Hz_10ms_log_rms_mae_db": 2.229272562956168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2065402202819678, "3000–6000 Hz_10ms_envelope_correlation": 0.9708269918953394, "3000–6000 Hz_10ms_modulation_mae_db": 1.3337888042600508, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4035741803993727, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10531981550001973, "3000–6000 Hz_40ms_envelope_correlation": 0.9922023704478732, "3000–6000 Hz_40ms_modulation_mae_db": 1.611746638034301, "mr_stft_db": 6.085060779065977, "band_mae_db": 0.08590076794893782, "rms_gain_error_db": -0.022694057760711643, "output_peak": 0.5891521255478182}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.673669081863147, "全帯域_10ms_rms_envelope_nrmse": 0.09290722224986896, "全帯域_10ms_envelope_correlation": 0.9916554419586026, "全帯域_10ms_modulation_mae_db": 0.451079142360616, "全帯域_40ms_log_rms_mae_db": 0.7755098913427592, "全帯域_40ms_rms_envelope_nrmse": 0.04858011542575961, "全帯域_40ms_envelope_correlation": 0.9979212205985186, "全帯域_40ms_modulation_mae_db": 0.26026317939642013, "300–1000 Hz_10ms_log_rms_mae_db": 3.0473322180738194, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09094668528944468, "300–1000 Hz_10ms_envelope_correlation": 0.9948407596557192, "300–1000 Hz_10ms_modulation_mae_db": 0.5358424884333989, "300–1000 Hz_40ms_log_rms_mae_db": 2.192734630449277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05390252382307354, "300–1000 Hz_40ms_envelope_correlation": 0.9983757153849263, "300–1000 Hz_40ms_modulation_mae_db": 1.0210163663776808, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1814958933803967, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16933754031786494, "1000–3000 Hz_10ms_envelope_correlation": 0.9803765616214174, "1000–3000 Hz_10ms_modulation_mae_db": 0.7441382232019138, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4450276835644669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1053368037016751, "1000–3000 Hz_40ms_envelope_correlation": 0.9923487656059227, "1000–3000 Hz_40ms_modulation_mae_db": 1.1330560892248207, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1767096071257406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21564008628111758, "3000–6000 Hz_10ms_envelope_correlation": 0.9683361666121281, "3000–6000 Hz_10ms_modulation_mae_db": 0.8960261014720151, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3533681820957333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11987602653494979, "3000–6000 Hz_40ms_envelope_correlation": 0.9896557466512877, "3000–6000 Hz_40ms_modulation_mae_db": 0.7372194161119321, "mr_stft_db": 6.108957263064053, "band_mae_db": 0.17401909239859756, "rms_gain_error_db": -0.18444635881626126, "output_peak": 0.5507166548667695}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48320, "output_samples": 48320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.386172983725045, "spectral_convergence": 0.3297684588561558, "spectral_convergence_status": "ok", "amplitude_floor": 1.6806044167073843e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.028670819229128, "spectral_convergence": 0.2896200214401102, "spectral_convergence_status": "ok", "amplitude_floor": 1.0917469705551035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9399724909324325, "spectral_convergence": 0.30000172520441326, "spectral_convergence_status": "ok", "amplitude_floor": 1.0075473322757857e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0009427012142024427, "output_power": 0.000921595049982271, "delta_db": -0.09833933026446294, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024186117972299723, "output_power": 0.002345168747871083, "delta_db": -0.13392068827071887, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0014180970240603914, "output_power": 0.001380979326804092, "delta_db": -0.11518768378047972, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023968394187503082, "output_power": 0.0002325976417554213, "delta_db": -0.13033631324346584, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.504236566794763e-05, "output_power": 5.182109671590138e-05, "delta_db": -0.2619049124844624, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00019111878985838765, "output_power": 0.00019090601636700385, "delta_db": -0.004837715313376526, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12408777389282771, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11888611546811186, "pitch": null, "output_rms_dbfs": -22.90469053384722, "output_peak": 0.5633865983057829, "reference_rms_dbfs": -22.785804418379104, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.118272097962202}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00023057786100875665, "log_rms_mae_db": 1.6327843263323516, "rms_envelope_nrmse": 0.08476675653950026, "envelope_correlation": 0.9929627734209389, "rms_gain_error_db": -0.11891212490627891, "reference_mean_power": 0.005273848963429562, "output_mean_power": 0.0051314071558132305}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.9211277816293244e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014051118338624634, "output_power": 0.0013392977645183993, "reference_active": true, "delta_db": -0.20833747491621732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001364841166272218, "output_power": 0.00011621117692406714, "reference_active": true, "delta_db": -0.6983421368102218}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.767297964659028e-05, "output_power": 3.2539931049292684e-05, "reference_active": true, "delta_db": -0.6361334260650429}], "active_band_count": 3, "active_band_mae_db": 0.5142710125971607}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00020448142520679275, "log_rms_mae_db": 0.7376874749365954, "rms_envelope_nrmse": 0.044151238533544114, "envelope_correlation": 0.9981320226925924, "rms_gain_error_db": -0.11921867223772284, "reference_mean_power": 0.005325056781288949, "output_mean_power": 0.005180866192604244}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.674469134143454e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013611842323655793, "output_power": 0.001294984677625035, "reference_active": true, "delta_db": -0.21652279852546696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.483051409804231e-05, "output_power": 5.238848594501534e-05, "reference_active": true, "delta_db": -0.9254361822983016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6716298904493925e-06, "output_power": 1.4943068423478113e-06, "reference_active": true, "delta_db": -0.48700343102980426}], "active_band_count": 3, "active_band_mae_db": 0.5429874706178577}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00020294603022145288, "log_rms_mae_db": 3.2194074901140493, "rms_envelope_nrmse": 0.09451079270340723, "envelope_correlation": 0.9943047928780386, "rms_gain_error_db": -0.052482738142468725, "reference_mean_power": 0.002254450133685994, "output_mean_power": 0.0022273699682569034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.5930281169536517e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007344814515063544, "output_power": 0.0007144665338417041, "reference_active": true, "delta_db": -0.11998941733518695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010026570767730039, "output_power": 8.052035108529551e-05, "reference_active": true, "delta_db": -0.9524676358484464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.614259763381322e-05, "output_power": 9.561017362057603e-06, "reference_active": true, "delta_db": -2.2746931495102887}], "active_band_count": 3, "active_band_mae_db": 1.1157167342313075}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00017314576936579135, "log_rms_mae_db": 2.4340715690557984, "rms_envelope_nrmse": 0.04860321551121798, "envelope_correlation": 0.9985548911182659, "rms_gain_error_db": -0.0525012816508473, "reference_mean_power": 0.002277104155750591, "output_mean_power": 0.002249742267146914}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.4781762283606311e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007359452368517945, "output_power": 0.0007085070778300901, "reference_active": true, "delta_db": -0.1650130571236138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.796294327084698e-05, "output_power": 3.949231672005271e-05, "reference_active": true, "delta_db": -0.8439321464416896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4655779178142778e-06, "output_power": 5.942554104989171e-07, "reference_active": true, "delta_db": -3.920357682365288}], "active_band_count": 3, "active_band_mae_db": 1.6431009619768637}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 8.794013668187323e-05, "log_rms_mae_db": 2.2029892519276384, "rms_envelope_nrmse": 0.1533603911200396, "envelope_correlation": 0.9843894595360678, "rms_gain_error_db": -0.280088201896737, "reference_mean_power": 0.0002815001552852329, "output_mean_power": 0.00026391849050150007}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.4039983283954285e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014419558923291274, "output_power": 0.00012577759892148506, "reference_active": true, "delta_db": -0.5934867620370523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9143053212884802e-05, "output_power": 1.3954830653109597e-05, "reference_active": true, "delta_db": -1.3728663603657651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.190405195331938e-06, "output_power": 4.2726718211451475e-06, "reference_active": true, "delta_db": 0.08443516927371304}], "active_band_count": 3, "active_band_mae_db": 0.6835960972255101}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 7.034799916708663e-05, "log_rms_mae_db": 1.4860193879375363, "rms_envelope_nrmse": 0.08891310784158371, "envelope_correlation": 0.9952759944987427, "rms_gain_error_db": -0.28015208523955026, "reference_mean_power": 0.00028432562712779955, "output_mean_power": 0.0002665635706584494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.2325770901894182e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001384543016382475, "output_power": 0.00011991544453221676, "reference_active": true, "delta_db": -0.6243133165066335}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.792343358100607e-06, "output_power": 7.0458530429668986e-06, "reference_active": true, "delta_db": -1.4295305238601996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0751348414140544e-07, "output_power": 2.5868525411343257e-07, "reference_active": true, "delta_db": -0.7509249063509406}], "active_band_count": 3, "active_band_mae_db": 0.9349229155725912}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 5.049105990445221e-05, "log_rms_mae_db": 2.2370937753977396, "rms_envelope_nrmse": 0.20777404963098453, "envelope_correlation": 0.9713470499069812, "rms_gain_error_db": 0.14540024148084088, "reference_mean_power": 0.00010138440655269997, "output_mean_power": 0.00010483617993063315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.3079247184162883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1617157093216335e-05, "output_power": 5.344776470786741e-05, "reference_active": true, "delta_db": 0.15135465537991682}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0705364401401438e-05, "output_power": 1.3722776136736731e-05, "reference_active": true, "delta_db": -1.7864089957781464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.974689061207133e-06, "output_power": 2.563238787913698e-06, "reference_active": true, "delta_db": -0.646525100664076}], "active_band_count": 3, "active_band_mae_db": 0.861429583940713}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 4.0590764349045424e-05, "log_rms_mae_db": 1.3393120396307994, "rms_envelope_nrmse": 0.10441685840126785, "envelope_correlation": 0.9925041936838677, "rms_gain_error_db": 0.1453680789033118, "reference_mean_power": 0.00010240265187942684, "output_mean_power": 0.00010588830865750011}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.2076196938321023e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.154744016518151e-05, "output_power": 6.22001307759333e-05, "reference_active": true, "delta_db": 0.04581303003375474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.945675894844695e-06, "output_power": 6.910091781335702e-06, "reference_active": true, "delta_db": -1.5815048673018683}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5378071346519667e-07, "output_power": 2.0658916793746917e-07, "reference_active": true, "delta_db": -0.8935106752477379}], "active_band_count": 3, "active_band_mae_db": 0.8402761908611204}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7415719624129796, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1b8e6dbc6dc7945d81db68a3e1eed29817aa5142da54d1cbb6a3c10b17e9255c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9825110654259334, "standard_deviation": 0.00380355674367154, "minimum": 0.9757277266630318, "maximum": 0.9875656908851241, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2114574341313746, "standard_deviation": 0.0432076391380484, "minimum": 2.151113446380769, "maximum": 2.2653820104894127, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7409842128289642, "standard_deviation": 0.20065645446393712, "minimum": 0.38231376960725605, "maximum": 1.0013715173832154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16155347666983658, "standard_deviation": 0.015970399646133168, "minimum": 0.1356026181786775, "maximum": 0.1875692889720046, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994546123720378, "standard_deviation": 0.0011437305215941597, "minimum": 0.9923487656059227, "maximum": 0.9959888378387217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.5036619398226045, "standard_deviation": 0.03652938601138374, "minimum": 1.4450276835644669, "maximum": 1.5647025191138877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9525418530107795, "standard_deviation": 0.3500288459336228, "minimum": 0.3708183752201555, "maximum": 1.5093862677735934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09396308832070366, "standard_deviation": 0.01013114693471653, "minimum": 0.07626864574524024, "maximum": 0.1053368037016751, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9710430455459341, "standard_deviation": 0.0014681203153759333, "minimum": 0.9683361666121281, "maximum": 0.9733547113508868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2207522317784303, "standard_deviation": 0.030768089776510486, "minimum": 2.1761307345385594, "maximum": 2.256539760613436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0858604175967854, "standard_deviation": 0.5299381806261012, "minimum": 0.5241434456633844, "maximum": 2.2147139920308687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20740625195910398, "standard_deviation": 0.004800046012896138, "minimum": 0.19990882311654398, "maximum": 0.21564008628111758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.990972619759086, "standard_deviation": 0.0009937427430211968, "minimum": 0.9896557466512877, "maximum": 0.9925041936838677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3895630441178097, "standard_deviation": 0.04137314393909065, "minimum": 1.3393120396307994, "maximum": 1.460832015495249, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1157650123570153, "standard_deviation": 0.5674536057555757, "minimum": 0.6149366351928217, "maximum": 2.215795410463662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11365959922583005, "standard_deviation": 0.00655743412852522, "minimum": 0.10441685840126785, "maximum": 0.12172648124188387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9948031717149954, "standard_deviation": 0.0013133475267905735, "minimum": 0.9922854396408275, "maximum": 0.9961797201766517, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 3.0444170704977904, "standard_deviation": 0.12239997525451714, "minimum": 2.831825406050438, "maximum": 3.2194074901140493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.490674882753948, "standard_deviation": 0.2630289111905935, "minimum": 0.3066910015819038, "maximum": 1.1157167342313075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08939941075330661, "standard_deviation": 0.010650286481215911, "minimum": 0.07677740507556827, "maximum": 0.10827038246817242, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984140264750494, "standard_deviation": 0.0005673781005665258, "minimum": 0.997309553408359, "maximum": 0.998946908035209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.2601431204746127, "standard_deviation": 0.10675927361480936, "minimum": 2.111524522714291, "maximum": 2.4340715690557984, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6927686519772159, "standard_deviation": 0.45627012496563213, "minimum": 0.13825276758565025, "maximum": 1.6431009619768637, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04984205830161245, "standard_deviation": 0.007774881368743329, "minimum": 0.04178347835798527, "maximum": 0.06346184189165141, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14592046883121895, "standard_deviation": 0.04423807083617129, "minimum": 0.08590076794893782, "maximum": 0.22105472382058303, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.096058709466515, "standard_deviation": 0.011888971597423874, "minimum": 6.085060779065977, "maximum": 6.118272097962202, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5601861555912173, "standard_deviation": 0.015626638898948438, "minimum": 0.5398265266877491, "maximum": 0.5891521255478182, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08117742843428452, "standard_deviation": 0.05260736808104688, "minimum": -0.18444635881626126, "maximum": -0.022694057760711643, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9921971881951993, "standard_deviation": 0.0011546012233446868, "minimum": 0.9895656898439646, "maximum": 0.993037794234297, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.677815269536721, "standard_deviation": 0.056634998678468274, "minimum": 1.6059047184647146, "maximum": 1.759420069672387, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2621738320180431, "standard_deviation": 0.1475860513899728, "minimum": 0.12273407607996646, "maximum": 0.5142710125971607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08869223100245127, "standard_deviation": 0.0061494666468539195, "minimum": 0.08387301526317252, "maximum": 0.10212616522930641, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979949239218501, "standard_deviation": 0.0003806389215312223, "minimum": 0.9971094185175107, "maximum": 0.9982889519631472, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7680387878861, "standard_deviation": 0.06348714689841166, "minimum": 0.7111933328763856, "maximum": 0.9120199645235465, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24187051591225123, "standard_deviation": 0.13933661911660725, "minimum": 0.08916375578086844, "maximum": 0.5429874706178577, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04468970813816443, "standard_deviation": 0.0040142885180062, "minimum": 0.04105860705306181, "maximum": 0.05265038500224752, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0042", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.315295934677124, "predicted_mos": 2.315295934677124, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "fc96079ffab82e050dac049463f01f76eb284a3fead0b8ff8a6ba1e0031c523f", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.682234500069171, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6763212338180102, "全帯域_10ms_rms_envelope_nrmse": 0.09621894815990352, "全帯域_10ms_envelope_correlation": 0.9908521317046555, "全帯域_10ms_modulation_mae_db": 0.09941559259344816, "全帯域_40ms_log_rms_mae_db": 0.8502027684792045, "全帯域_40ms_rms_envelope_nrmse": 0.050347773641961246, "全帯域_40ms_envelope_correlation": 0.9974270307368687, "全帯域_40ms_modulation_mae_db": 0.23818447641435206, "300–1000 Hz_10ms_log_rms_mae_db": 2.5195112784895883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10196735460637378, "300–1000 Hz_10ms_envelope_correlation": 0.9931470157819233, "300–1000 Hz_10ms_modulation_mae_db": 0.40175144917037525, "300–1000 Hz_40ms_log_rms_mae_db": 1.8827568660072662, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051178370483220895, "300–1000 Hz_40ms_envelope_correlation": 0.9982786894065176, "300–1000 Hz_40ms_modulation_mae_db": 0.15484576712582568, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5803626455131328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15953799584767123, "1000–3000 Hz_10ms_envelope_correlation": 0.9829837674068012, "1000–3000 Hz_10ms_modulation_mae_db": 0.577216416225579, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0220898276172643, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09671308414208271, "1000–3000 Hz_40ms_envelope_correlation": 0.994274771962198, "1000–3000 Hz_40ms_modulation_mae_db": 0.6867311085085589, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3270971407505412, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1845409285376995, "3000–6000 Hz_10ms_envelope_correlation": 0.9771693501324448, "3000–6000 Hz_10ms_modulation_mae_db": 0.20496843307684723, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8877769509124216, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10896476958461647, "3000–6000 Hz_40ms_envelope_correlation": 0.9925386622959631, "3000–6000 Hz_40ms_modulation_mae_db": 0.2928967881971148, "mr_stft_db": 5.708310326941469, "band_mae_db": 0.17752969892773232, "rms_gain_error_db": -0.09932759636855298, "output_peak": 0.5589818477914061}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7310210020805201, "全帯域_10ms_rms_envelope_nrmse": 0.08726585809491919, "全帯域_10ms_envelope_correlation": 0.9924214708605462, "全帯域_10ms_modulation_mae_db": 0.10885351145871279, "全帯域_40ms_log_rms_mae_db": 0.7778160234327828, "全帯域_40ms_rms_envelope_nrmse": 0.04640560035720733, "全帯域_40ms_envelope_correlation": 0.9977190099919659, "全帯域_40ms_modulation_mae_db": 0.08924593249432934, "300–1000 Hz_10ms_log_rms_mae_db": 2.5699640341007517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07824425390287053, "300–1000 Hz_10ms_envelope_correlation": 0.9960250510948184, "300–1000 Hz_10ms_modulation_mae_db": 0.8492507087809901, "300–1000 Hz_40ms_log_rms_mae_db": 1.985696712853066, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044704007981137904, "300–1000 Hz_40ms_envelope_correlation": 0.9986904226076094, "300–1000 Hz_40ms_modulation_mae_db": 0.9898380458675713, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5624277585255075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14869606997221552, "1000–3000 Hz_10ms_envelope_correlation": 0.9848632544839331, "1000–3000 Hz_10ms_modulation_mae_db": 0.4901843718035613, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0245701076117648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08690084813107371, "1000–3000 Hz_40ms_envelope_correlation": 0.9947846069725103, "1000–3000 Hz_40ms_modulation_mae_db": 0.4632545306961943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3418812298825717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1830306335480732, "3000–6000 Hz_10ms_envelope_correlation": 0.9773469395789449, "3000–6000 Hz_10ms_modulation_mae_db": 0.7256383252523829, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8418791687982923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1146405241609672, "3000–6000 Hz_40ms_envelope_correlation": 0.9906701592211333, "3000–6000 Hz_40ms_modulation_mae_db": 0.6993323691050258, "mr_stft_db": 5.71631838928948, "band_mae_db": 0.15018722812812987, "rms_gain_error_db": 0.007335372007622142, "output_peak": 0.5312397580081125}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9013197158263824, "全帯域_10ms_rms_envelope_nrmse": 0.09189100101225574, "全帯域_10ms_envelope_correlation": 0.9917228164919034, "全帯域_10ms_modulation_mae_db": 0.22688071706412924, "全帯域_40ms_log_rms_mae_db": 0.8860486462550268, "全帯域_40ms_rms_envelope_nrmse": 0.05005267171690701, "全帯域_40ms_envelope_correlation": 0.9974519057499176, "全帯域_40ms_modulation_mae_db": 0.21125890490837096, "300–1000 Hz_10ms_log_rms_mae_db": 2.563745345349332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08516913609359793, "300–1000 Hz_10ms_envelope_correlation": 0.995318677992383, "300–1000 Hz_10ms_modulation_mae_db": 0.4626619498009909, "300–1000 Hz_40ms_log_rms_mae_db": 2.015738517860192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05741495274328644, "300–1000 Hz_40ms_envelope_correlation": 0.997876183175543, "300–1000 Hz_40ms_modulation_mae_db": 0.5776115738096331, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4931127870176661, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1716305676100772, "1000–3000 Hz_10ms_envelope_correlation": 0.9798302385431001, "1000–3000 Hz_10ms_modulation_mae_db": 1.0911602239573002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0040923925902803, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09979247276047402, "1000–3000 Hz_40ms_envelope_correlation": 0.9932698426979898, "1000–3000 Hz_40ms_modulation_mae_db": 1.2502186378094007, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3919940204993397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15082707643819548, "3000–6000 Hz_10ms_envelope_correlation": 0.9845468175424366, "3000–6000 Hz_10ms_modulation_mae_db": 1.07781146218184, "3000–6000 Hz_40ms_log_rms_mae_db": 0.83104668314916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07974273156519875, "3000–6000 Hz_40ms_envelope_correlation": 0.9956655114427947, "3000–6000 Hz_40ms_modulation_mae_db": 0.696591672485816, "mr_stft_db": 5.686536699828085, "band_mae_db": 0.22709668665395666, "rms_gain_error_db": -0.06895886621678263, "output_peak": 0.5631022290870938}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7561489726155415, "全帯域_10ms_rms_envelope_nrmse": 0.09428530754783157, "全帯域_10ms_envelope_correlation": 0.9912946407768483, "全帯域_10ms_modulation_mae_db": 0.1890265897687695, "全帯域_40ms_log_rms_mae_db": 0.8860671229706499, "全帯域_40ms_rms_envelope_nrmse": 0.05123147676612825, "全帯域_40ms_envelope_correlation": 0.9972823729127744, "全帯域_40ms_modulation_mae_db": 0.4123526484184242, "300–1000 Hz_10ms_log_rms_mae_db": 2.7466224907226935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.098898276474685, "300–1000 Hz_10ms_envelope_correlation": 0.993719517482696, "300–1000 Hz_10ms_modulation_mae_db": 0.13160426321490035, "300–1000 Hz_40ms_log_rms_mae_db": 2.09034849071535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0633943957062841, "300–1000 Hz_40ms_envelope_correlation": 0.9973832953775997, "300–1000 Hz_40ms_modulation_mae_db": 0.6030668609350616, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5715685121188383, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1414492305722308, "1000–3000 Hz_10ms_envelope_correlation": 0.9871881941554437, "1000–3000 Hz_10ms_modulation_mae_db": 0.7362615802117949, "1000–3000 Hz_40ms_log_rms_mae_db": 0.99861167583166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08863978076357372, "1000–3000 Hz_40ms_envelope_correlation": 0.9957123337835988, "1000–3000 Hz_40ms_modulation_mae_db": 0.48390203595892456, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3935356322029784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16470516275186317, "3000–6000 Hz_10ms_envelope_correlation": 0.981732946724972, "3000–6000 Hz_10ms_modulation_mae_db": 0.8190768680677776, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8761098462495168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0947130248909243, "3000–6000 Hz_40ms_envelope_correlation": 0.9936427353404383, "3000–6000 Hz_40ms_modulation_mae_db": 1.1187932465379367, "mr_stft_db": 5.708948506428256, "band_mae_db": 0.24051573894994624, "rms_gain_error_db": 0.001588478509136885, "output_peak": 0.5704982899359139}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7747028501668412, "全帯域_10ms_rms_envelope_nrmse": 0.08987853684217088, "全帯域_10ms_envelope_correlation": 0.99224871376582, "全帯域_10ms_modulation_mae_db": 0.2955852435591662, "全帯域_40ms_log_rms_mae_db": 0.9639554825511585, "全帯域_40ms_rms_envelope_nrmse": 0.04789499775366772, "全帯域_40ms_envelope_correlation": 0.9978820528737938, "全帯域_40ms_modulation_mae_db": 0.3812891064189003, "300–1000 Hz_10ms_log_rms_mae_db": 2.5863842899457246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08818059209030733, "300–1000 Hz_10ms_envelope_correlation": 0.9949168128209436, "300–1000 Hz_10ms_modulation_mae_db": 0.4662629429919387, "300–1000 Hz_40ms_log_rms_mae_db": 1.952186113946086, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04769889937072646, "300–1000 Hz_40ms_envelope_correlation": 0.9985126403424285, "300–1000 Hz_40ms_modulation_mae_db": 0.4429142041646563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.465118858966631, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12265492797814877, "1000–3000 Hz_10ms_envelope_correlation": 0.9897176087908244, "1000–3000 Hz_10ms_modulation_mae_db": 0.23710135026800794, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9206300673667587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06170049992736896, "1000–3000 Hz_40ms_envelope_correlation": 0.9974017952248944, "1000–3000 Hz_40ms_modulation_mae_db": 0.2762963945464201, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3280144447652866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14989170113446476, "3000–6000 Hz_10ms_envelope_correlation": 0.9847807950872909, "3000–6000 Hz_10ms_modulation_mae_db": 0.9113956724745633, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7917382983021862, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08588721311889456, "3000–6000 Hz_40ms_envelope_correlation": 0.9950074338598859, "3000–6000 Hz_40ms_modulation_mae_db": 1.1084565920555876, "mr_stft_db": 5.6762684674640695, "band_mae_db": 0.16347839178092977, "rms_gain_error_db": -0.12560583714124846, "output_peak": 0.575554194912818}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.775525292878677, "全帯域_10ms_rms_envelope_nrmse": 0.09640933333759136, "全帯域_10ms_envelope_correlation": 0.9908149054405971, "全帯域_10ms_modulation_mae_db": 0.1547003096565683, "全帯域_40ms_log_rms_mae_db": 0.8189073783308523, "全帯域_40ms_rms_envelope_nrmse": 0.049379721054809776, "全帯域_40ms_envelope_correlation": 0.9974458232855448, "全帯域_40ms_modulation_mae_db": 0.2822252578858854, "300–1000 Hz_10ms_log_rms_mae_db": 2.5097525383066244, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10089013813800136, "300–1000 Hz_10ms_envelope_correlation": 0.9933375552135553, "300–1000 Hz_10ms_modulation_mae_db": 0.2568337928790408, "300–1000 Hz_40ms_log_rms_mae_db": 1.863909833321976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05687784356562062, "300–1000 Hz_40ms_envelope_correlation": 0.9978438313671002, "300–1000 Hz_40ms_modulation_mae_db": 0.5341228447034799, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6544354263335443, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15542353135963463, "1000–3000 Hz_10ms_envelope_correlation": 0.9847501258534065, "1000–3000 Hz_10ms_modulation_mae_db": 1.1178591682365389, "1000–3000 Hz_40ms_log_rms_mae_db": 1.155660175005517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10445993445432225, "1000–3000 Hz_40ms_envelope_correlation": 0.9940551591201386, "1000–3000 Hz_40ms_modulation_mae_db": 1.2845775472614858, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3141359606228518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19219781914906178, "3000–6000 Hz_10ms_envelope_correlation": 0.9762487160041624, "3000–6000 Hz_10ms_modulation_mae_db": 1.391233701532088, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7378338654259562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10561195838311119, "3000–6000 Hz_40ms_envelope_correlation": 0.9928269110353892, "3000–6000 Hz_40ms_modulation_mae_db": 1.6648027708553677, "mr_stft_db": 5.692144317742861, "band_mae_db": 0.21362286173993958, "rms_gain_error_db": 0.008463032235702952, "output_peak": 0.5368228253870713}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6516945074700802, "全帯域_10ms_rms_envelope_nrmse": 0.09132367054655918, "全帯域_10ms_envelope_correlation": 0.9918573360947568, "全帯域_10ms_modulation_mae_db": 0.3463760725306229, "全帯域_40ms_log_rms_mae_db": 0.8066955744824845, "全帯域_40ms_rms_envelope_nrmse": 0.044802912818482386, "全帯域_40ms_envelope_correlation": 0.9981755808527841, "全帯域_40ms_modulation_mae_db": 0.15725131362233005, "300–1000 Hz_10ms_log_rms_mae_db": 2.566258330422823, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10720938845697124, "300–1000 Hz_10ms_envelope_correlation": 0.9929513835495596, "300–1000 Hz_10ms_modulation_mae_db": 0.4450223424925652, "300–1000 Hz_40ms_log_rms_mae_db": 1.8973322364028635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0663819701090953, "300–1000 Hz_40ms_envelope_correlation": 0.9977138857823158, "300–1000 Hz_40ms_modulation_mae_db": 0.5370387384811915, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4645526960318995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1369366427200464, "1000–3000 Hz_10ms_envelope_correlation": 0.9874776314250544, "1000–3000 Hz_10ms_modulation_mae_db": 0.5166699907819965, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9836993099055582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0861496735840161, "1000–3000 Hz_40ms_envelope_correlation": 0.9952556590253574, "1000–3000 Hz_40ms_modulation_mae_db": 0.6010584138893115, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3663410064110046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17478252734356622, "3000–6000 Hz_10ms_envelope_correlation": 0.9793948574956656, "3000–6000 Hz_10ms_modulation_mae_db": 1.0418332806305541, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8803457544318583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09708444105828837, "3000–6000 Hz_40ms_envelope_correlation": 0.9933159601637858, "3000–6000 Hz_40ms_modulation_mae_db": 0.8950082787478465, "mr_stft_db": 5.683618627530637, "band_mae_db": 0.33200330412948764, "rms_gain_error_db": -0.1564694058197123, "output_peak": 0.5761820256589829}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8983995303243035, "全帯域_10ms_rms_envelope_nrmse": 0.09804844659894754, "全帯域_10ms_envelope_correlation": 0.9904506766891983, "全帯域_10ms_modulation_mae_db": 0.44303053844561563, "全帯域_40ms_log_rms_mae_db": 0.9316675641060991, "全帯域_40ms_rms_envelope_nrmse": 0.050455993073412156, "全帯域_40ms_envelope_correlation": 0.9974080850153633, "全帯域_40ms_modulation_mae_db": 0.6098009383475229, "300–1000 Hz_10ms_log_rms_mae_db": 2.720380956724535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09533234972334846, "300–1000 Hz_10ms_envelope_correlation": 0.9940362142587212, "300–1000 Hz_10ms_modulation_mae_db": 0.5881942563724939, "300–1000 Hz_40ms_log_rms_mae_db": 1.980231660924975, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04993623070025424, "300–1000 Hz_40ms_envelope_correlation": 0.9983867694214621, "300–1000 Hz_40ms_modulation_mae_db": 0.4870486438510626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5229735612694972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14520967045929808, "1000–3000 Hz_10ms_envelope_correlation": 0.9865681754452857, "1000–3000 Hz_10ms_modulation_mae_db": 0.5865242057732248, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0569836928871814, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09009571813317431, "1000–3000 Hz_40ms_envelope_correlation": 0.9957289350588875, "1000–3000 Hz_40ms_modulation_mae_db": 0.42941509386975757, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2957872556089833, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14791617325197867, "3000–6000 Hz_10ms_envelope_correlation": 0.9857772173141374, "3000–6000 Hz_10ms_modulation_mae_db": 0.6441177075277483, "3000–6000 Hz_40ms_log_rms_mae_db": 0.783142692971925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0793924095837915, "3000–6000 Hz_40ms_envelope_correlation": 0.99661474724376, "3000–6000 Hz_40ms_modulation_mae_db": 0.8549330572027661, "mr_stft_db": 5.702340772916173, "band_mae_db": 0.24121092672355474, "rms_gain_error_db": -0.0954656449082693, "output_peak": 0.5609067926682108}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48320, "output_samples": 48320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.824004658357366, "spectral_convergence": 0.3313567699038239, "spectral_convergence_status": "ok", "amplitude_floor": 1.6806044167073843e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4979243404953175, "spectral_convergence": 0.2796091223336061, "spectral_convergence_status": "ok", "amplitude_floor": 1.0917469705551035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.803001981971721, "spectral_convergence": 0.3010447710299712, "spectral_convergence_status": "ok", "amplitude_floor": 1.0075473322757857e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0009427012142024427, "output_power": 0.0009523578459695533, "delta_db": 0.04426097611337943, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024186117972299723, "output_power": 0.002335296958712728, "delta_db": -0.1522405315686618, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0014180970240603914, "output_power": 0.0014105386384259129, "delta_db": -0.023209584955238994, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023968394187503082, "output_power": 0.00021638757308342803, "delta_db": -0.4440662248117308, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.504236566794763e-05, "output_power": 5.560145598608143e-05, "delta_db": 0.0438907302810696, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00019111878985838765, "output_power": 0.00017601609116203783, "delta_db": -0.35751014583631324, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17752969892773232, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09932759636855298, "pitch": null, "output_rms_dbfs": -22.88513201474766, "output_peak": 0.5589818477914061, "reference_rms_dbfs": -22.785804418379104, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.708310326941469}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00023057786100875665, "log_rms_mae_db": 1.6763212338180102, "rms_envelope_nrmse": 0.09621894815990352, "envelope_correlation": 0.9908521317046555, "rms_gain_error_db": -0.09940782964534564, "reference_mean_power": 0.005273848963429562, "output_mean_power": 0.005154504285173164}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.9211277816293244e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014051118338624634, "output_power": 0.0014019073685019832, "reference_active": true, "delta_db": -0.009915730016341521}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001364841166272218, "output_power": 0.0001323155911793922, "reference_active": true, "delta_db": -0.1347109164612042}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.767297964659028e-05, "output_power": 3.636369092716276e-05, "reference_active": true, "delta_db": -0.15362013130279875}], "active_band_count": 3, "active_band_mae_db": 0.09941559259344816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00020448142520679275, "log_rms_mae_db": 0.8502027684792045, "rms_envelope_nrmse": 0.050347773641961246, "envelope_correlation": 0.9974270307368687, "rms_gain_error_db": -0.09909180649504613, "reference_mean_power": 0.005325056781288949, "output_mean_power": 0.0052049320282302825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.674469134143454e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013611842323655793, "output_power": 0.0013478874338119736, "reference_active": true, "delta_db": -0.042632852311583855}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.483051409804231e-05, "output_power": 6.703066313372759e-05, "reference_active": true, "delta_db": 0.14494050367605543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6716298904493925e-06, "output_power": 1.8872878130052144e-06, "reference_active": true, "delta_db": 0.526980073255417}], "active_band_count": 3, "active_band_mae_db": 0.23818447641435206}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00020294603022145288, "log_rms_mae_db": 2.5195112784895883, "rms_envelope_nrmse": 0.10196735460637378, "envelope_correlation": 0.9931470157819233, "rms_gain_error_db": -0.0492004420397285, "reference_mean_power": 0.002254450133685994, "output_mean_power": 0.0022290539986708343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.5930281169536517e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007344814515063544, "output_power": 0.0006998823030500561, "reference_active": true, "delta_db": -0.20955820399609154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010026570767730039, "output_power": 9.681414833007084e-05, "reference_active": true, "delta_db": -0.15213594078691609}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.614259763381322e-05, "output_power": 1.960331154869145e-05, "reference_active": true, "delta_db": 0.8435602027281182}], "active_band_count": 3, "active_band_mae_db": 0.40175144917037525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00017314576936579135, "log_rms_mae_db": 1.8827568660072662, "rms_envelope_nrmse": 0.051178370483220895, "envelope_correlation": 0.9982786894065176, "rms_gain_error_db": -0.04920725581515759, "reference_mean_power": 0.002277104155750591, "output_mean_power": 0.00225144929327762}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.4781762283606311e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007359452368517945, "output_power": 0.0007015140121711899, "reference_active": true, "delta_db": -0.2080914869432698}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.796294327084698e-05, "output_power": 4.987978063975109e-05, "reference_active": true, "delta_db": 0.17018708929274867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4655779178142778e-06, "output_power": 1.4949779436102049e-06, "reference_active": true, "delta_db": 0.08625872514145863}], "active_band_count": 3, "active_band_mae_db": 0.15484576712582568}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 8.794013668187323e-05, "log_rms_mae_db": 1.5803626455131328, "rms_envelope_nrmse": 0.15953799584767123, "envelope_correlation": 0.9829837674068012, "rms_gain_error_db": -0.304578321373176, "reference_mean_power": 0.0002815001552852329, "output_mean_power": 0.00026243442701285853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.4039983283954285e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014419558923291274, "output_power": 0.0001260716218416768, "reference_active": true, "delta_db": -0.5833463622042903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9143053212884802e-05, "output_power": 1.5570963495392555e-05, "reference_active": true, "delta_db": -0.8969572003060966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.190405195331938e-06, "output_power": 3.954772002218524e-06, "reference_active": true, "delta_db": -0.2513456861663501}], "active_band_count": 3, "active_band_mae_db": 0.577216416225579}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 7.034799916708663e-05, "log_rms_mae_db": 1.0220898276172643, "rms_envelope_nrmse": 0.09671308414208271, "envelope_correlation": 0.994274771962198, "rms_gain_error_db": -0.30460007237188486, "reference_mean_power": 0.00028432562712779955, "output_mean_power": 0.00026506720487695824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.2325770901894182e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001384543016382475, "output_power": 0.00011973985486270122, "reference_active": true, "delta_db": -0.6306772599193388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.792343358100607e-06, "output_power": 7.576265074005194e-06, "reference_active": true, "delta_db": -1.1143147203577892}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0751348414140544e-07, "output_power": 2.859855100185048e-07, "reference_active": true, "delta_db": -0.31520134524854837}], "active_band_count": 3, "active_band_mae_db": 0.6867311085085589}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 5.049105990445221e-05, "log_rms_mae_db": 1.3270971407505412, "rms_envelope_nrmse": 0.1845409285376995, "envelope_correlation": 0.9771693501324448, "rms_gain_error_db": -0.39886080956005315, "reference_mean_power": 0.00010138440655269997, "output_mean_power": 9.248793488843949e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.3079247184162883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1617157093216335e-05, "output_power": 4.553975233265796e-05, "reference_active": true, "delta_db": -0.5440341694434138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0705364401401438e-05, "output_power": 2.0714404264876524e-05, "reference_active": true, "delta_db": 0.0018956952262847594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.974689061207133e-06, "output_power": 2.927817706798147e-06, "reference_active": true, "delta_db": -0.06897543456084321}], "active_band_count": 3, "active_band_mae_db": 0.20496843307684723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 4.0590764349045424e-05, "log_rms_mae_db": 0.8877769509124216, "rms_envelope_nrmse": 0.10896476958461647, "envelope_correlation": 0.9925386622959631, "rms_gain_error_db": -0.39886621142752, "reference_mean_power": 0.00010240265187942684, "output_mean_power": 9.341671309408384e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.2076196938321023e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.154744016518151e-05, "output_power": 5.518762453007363e-05, "reference_active": true, "delta_db": -0.47368293891713453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.945675894844695e-06, "output_power": 1.0828830739647502e-05, "reference_active": true, "delta_db": 0.369472629781315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5378071346519667e-07, "output_power": 2.5171270363118146e-07, "reference_active": true, "delta_db": -0.03553479589289484}], "active_band_count": 3, "active_band_mae_db": 0.2928967881971148}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7827696644268818, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "039d6742c1c4f3e16944ea8c0cc934cb2f418bd87c6440dd4a422149aa0eafb3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9854223745129811, "standard_deviation": 0.0030492954702331793, "minimum": 0.9798302385431001, "maximum": 0.9897176087908244, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5393190307220896, "standard_deviation": 0.06547438629525668, "minimum": 1.4645526960318995, "maximum": 1.6544354263335443, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6691221634072504, "standard_deviation": 0.30264552473688566, "minimum": 0.23710135026800794, "maximum": 1.1178591682365389, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14769232956491535, "standard_deviation": 0.014937174074743107, "minimum": 0.12265492797814877, "maximum": 0.1716305676100772, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950603879806968, "standard_deviation": 0.0012706630639306897, "minimum": 0.9932698426979898, "maximum": 0.9974017952248944, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0207921561019981, "standard_deviation": 0.06728865940529631, "minimum": 0.9206300673667587, "maximum": 1.155660175005517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6844317203175067, "standard_deviation": 0.379470622322141, "minimum": 0.2762963945464201, "maximum": 1.2845775472614858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08930650148701072, "standard_deviation": 0.012946875093990588, "minimum": 0.06170049992736896, "maximum": 0.10445993445432225, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9808747049850068, "standard_deviation": 0.00384108731870298, "minimum": 0.9762487160041624, "maximum": 0.9857772173141374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3448483363429447, "standard_deviation": 0.035882011840202885, "minimum": 1.2957872556089833, "maximum": 1.3935356322029784, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8520094313429751, "standard_deviation": 0.35055963880885377, "minimum": 0.20496843307684723, "maximum": 1.391233701532088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16848650276936283, "standard_deviation": 0.017591401112577947, "minimum": 0.14791617325197867, "maximum": 0.19219781914906178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937852650753938, "standard_deviation": 0.0019068260668804915, "minimum": 0.9906701592211333, "maximum": 0.99661474724376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8287341575301646, "standard_deviation": 0.05382848649440116, "minimum": 0.7378338654259562, "maximum": 0.8877769509124216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9163518468984326, "standard_deviation": 0.40113794435701755, "minimum": 0.2928967881971148, "maximum": 1.6648027708553677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09575463404322404, "standard_deviation": 0.01337734818880232, "minimum": 0.0793924095837915, "maximum": 0.1146405241609672, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9941815285243251, "standard_deviation": 0.0011188426442582757, "minimum": 0.9929513835495596, "maximum": 0.9960250510948184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.597827408007759, "standard_deviation": 0.08792484938033174, "minimum": 2.5097525383066244, "maximum": 2.7466224907226935, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4501977132129119, "standard_deviation": 0.21379730221060148, "minimum": 0.13160426321490035, "maximum": 0.8492507087809901, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09448643618576946, "standard_deviation": 0.009779053373167148, "minimum": 0.07824425390287053, "maximum": 0.10720938845697124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998085714685072, "standard_deviation": 0.00044878741313868155, "minimum": 0.9973832953775997, "maximum": 0.9986904226076094, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.9585250540039718, "standard_deviation": 0.07591148616785498, "minimum": 1.863909833321976, "maximum": 2.09034849071535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5408108348673102, "standard_deviation": 0.22939599520516862, "minimum": 0.15484576712582568, "maximum": 0.9898380458675713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05469833383245325, "standard_deviation": 0.0076364308227071815, "minimum": 0.044704007981137904, "maximum": 0.0663819701090953, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2182056046292096, "standard_deviation": 0.0577594183961226, "minimum": 0.15018722812812987, "maximum": 0.33200330412948764, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.696810763517629, "standard_deviation": 0.014212890070149042, "minimum": 5.6762684674640695, "maximum": 5.71631838928948, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5591609954312011, "standard_deviation": 0.016836140000022978, "minimum": 0.5312397580081125, "maximum": 0.5761820256589829, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06605505846276297, "standard_deviation": 0.06460459420027141, "minimum": -0.1564694058197123, "maximum": 0.008463032235702952, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9914578364780406, "standard_deviation": 0.0007176484144694548, "minimum": 0.9904506766891983, "maximum": 0.9924214708605462, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7706416381475445, "standard_deviation": 0.09123264864521602, "minimum": 1.6516945074700802, "maximum": 1.9013197158263824, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2329835718846291, "standard_deviation": 0.12085585505805953, "minimum": 0.09941559259344816, "maximum": 0.44303053844561563, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09316513776752237, "standard_deviation": 0.003694971884377414, "minimum": 0.08726585809491919, "maximum": 0.09804844659894754, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975989826773766, "standard_deviation": 0.00030197949324312556, "minimum": 0.9972823729127744, "maximum": 0.9981755808527841, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8651700700760323, "standard_deviation": 0.06377116349839469, "minimum": 0.7778160234327828, "maximum": 0.9639554825511585, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2977010723137644, "standard_deviation": 0.16560362491122266, "minimum": 0.08924593249432934, "maximum": 0.6098009383475229, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04882139339782199, "standard_deviation": 0.002251364085997979, "minimum": 0.044802912818482386, "maximum": 0.05123147676612825, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0042", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.063138723373413, "predicted_mos": 2.063138723373413, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "f997ce118b9f136718768b50e6e59cef9a2b3092dbf8d15f8714341473b5e969", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.589927299879491, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7470581462578256, "全帯域_10ms_rms_envelope_nrmse": 0.08613500469300539, "全帯域_10ms_envelope_correlation": 0.9927724490679601, "全帯域_10ms_modulation_mae_db": 0.3325574582951114, "全帯域_40ms_log_rms_mae_db": 0.7895163185823865, "全帯域_40ms_rms_envelope_nrmse": 0.04326129730529666, "全帯域_40ms_envelope_correlation": 0.9982146301112265, "全帯域_40ms_modulation_mae_db": 0.24328840065512114, "300–1000 Hz_10ms_log_rms_mae_db": 2.9620398403418835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09652610464974551, "300–1000 Hz_10ms_envelope_correlation": 0.9939579216821233, "300–1000 Hz_10ms_modulation_mae_db": 0.1934256492267691, "300–1000 Hz_40ms_log_rms_mae_db": 2.212877344504041, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05076479481631307, "300–1000 Hz_40ms_envelope_correlation": 0.9983828745911622, "300–1000 Hz_40ms_modulation_mae_db": 0.11598763602532232, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9069891701901245, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15503511729351604, "1000–3000 Hz_10ms_envelope_correlation": 0.9842401467892337, "1000–3000 Hz_10ms_modulation_mae_db": 1.0819142249731588, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3924047921270137, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10391559526448554, "1000–3000 Hz_40ms_envelope_correlation": 0.9933753376544742, "1000–3000 Hz_40ms_modulation_mae_db": 1.3604897498357473, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8147661669783843, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1590806973174034, "3000–6000 Hz_10ms_envelope_correlation": 0.9838000232438278, "3000–6000 Hz_10ms_modulation_mae_db": 0.90091642458558, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2722816454805477, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09255739422427309, "3000–6000 Hz_40ms_envelope_correlation": 0.9954095893551598, "3000–6000 Hz_40ms_modulation_mae_db": 0.9108075930645539, "mr_stft_db": 5.883938651128723, "band_mae_db": 0.2504236988076882, "rms_gain_error_db": -0.11992058033552455, "output_peak": 0.5557188932077991}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6973239997526492, "全帯域_10ms_rms_envelope_nrmse": 0.08852726680956742, "全帯域_10ms_envelope_correlation": 0.9921960705421672, "全帯域_10ms_modulation_mae_db": 0.4182695676451124, "全帯域_40ms_log_rms_mae_db": 0.7798643834663195, "全帯域_40ms_rms_envelope_nrmse": 0.043550268563118964, "全帯域_40ms_envelope_correlation": 0.9979964940656536, "全帯域_40ms_modulation_mae_db": 0.2634638317310195, "300–1000 Hz_10ms_log_rms_mae_db": 2.903398920211958, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0916407151096829, "300–1000 Hz_10ms_envelope_correlation": 0.9945346670113221, "300–1000 Hz_10ms_modulation_mae_db": 0.23756496776271083, "300–1000 Hz_40ms_log_rms_mae_db": 2.291036274015993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045981327647090925, "300–1000 Hz_40ms_envelope_correlation": 0.9986269779105836, "300–1000 Hz_40ms_modulation_mae_db": 0.29669006737324904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9207695458067728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17485303404219296, "1000–3000 Hz_10ms_envelope_correlation": 0.979036727186446, "1000–3000 Hz_10ms_modulation_mae_db": 0.8824299996550465, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3414729794393299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08959440423070926, "1000–3000 Hz_40ms_envelope_correlation": 0.9945651595109488, "1000–3000 Hz_40ms_modulation_mae_db": 0.6819180038360395, "3000–6000 Hz_10ms_log_rms_mae_db": 1.808053571905873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20940402168706257, "3000–6000 Hz_10ms_envelope_correlation": 0.9701175914655054, "3000–6000 Hz_10ms_modulation_mae_db": 1.1418404559637894, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1928386343846438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11237834237786039, "3000–6000 Hz_40ms_envelope_correlation": 0.9918543555516738, "3000–6000 Hz_40ms_modulation_mae_db": 1.6393548426768587, "mr_stft_db": 5.876878213560666, "band_mae_db": 0.10678055992519077, "rms_gain_error_db": -0.02077325859986613, "output_peak": 0.5455612935974214}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7360378323112118, "全帯域_10ms_rms_envelope_nrmse": 0.09251856525422127, "全帯域_10ms_envelope_correlation": 0.9915285570768498, "全帯域_10ms_modulation_mae_db": 0.163795489811451, "全帯域_40ms_log_rms_mae_db": 0.8138333663753078, "全帯域_40ms_rms_envelope_nrmse": 0.04391213279473706, "全帯域_40ms_envelope_correlation": 0.9980752358672936, "全帯域_40ms_modulation_mae_db": 0.05397798792707013, "300–1000 Hz_10ms_log_rms_mae_db": 2.667649503942691, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0941914118888516, "300–1000 Hz_10ms_envelope_correlation": 0.9943800407261858, "300–1000 Hz_10ms_modulation_mae_db": 0.7087876399541887, "300–1000 Hz_40ms_log_rms_mae_db": 1.9133656065043771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0492759693695472, "300–1000 Hz_40ms_envelope_correlation": 0.9986686665803438, "300–1000 Hz_40ms_modulation_mae_db": 0.5036163606900771, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9003492320024993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15804387687892457, "1000–3000 Hz_10ms_envelope_correlation": 0.9829411692265211, "1000–3000 Hz_10ms_modulation_mae_db": 0.387789381082907, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3604836312356006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08442920770663546, "1000–3000 Hz_40ms_envelope_correlation": 0.9952385095733198, "1000–3000 Hz_40ms_modulation_mae_db": 0.2903829220464023, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8707250235198372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18053126685412757, "3000–6000 Hz_10ms_envelope_correlation": 0.9777414816211535, "3000–6000 Hz_10ms_modulation_mae_db": 0.7110164831025031, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3238994769976355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11134575638583512, "3000–6000 Hz_40ms_envelope_correlation": 0.9911923145537537, "3000–6000 Hz_40ms_modulation_mae_db": 0.9992039104211733, "mr_stft_db": 5.8844971425276436, "band_mae_db": 0.1839595264961403, "rms_gain_error_db": -0.10108337648685956, "output_peak": 0.5665537172885278}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6771993145956303, "全帯域_10ms_rms_envelope_nrmse": 0.09302663514799855, "全帯域_10ms_envelope_correlation": 0.991444548191975, "全帯域_10ms_modulation_mae_db": 0.06915091037728564, "全帯域_40ms_log_rms_mae_db": 0.6845313398500258, "全帯域_40ms_rms_envelope_nrmse": 0.04837685848684423, "全帯域_40ms_envelope_correlation": 0.9975508136681438, "全帯域_40ms_modulation_mae_db": 0.07410924358358771, "300–1000 Hz_10ms_log_rms_mae_db": 2.8193544980730816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11627124141736339, "300–1000 Hz_10ms_envelope_correlation": 0.991423024414471, "300–1000 Hz_10ms_modulation_mae_db": 0.2586622656184384, "300–1000 Hz_40ms_log_rms_mae_db": 2.151516275242438, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07491463731614748, "300–1000 Hz_40ms_envelope_correlation": 0.9964491373889486, "300–1000 Hz_40ms_modulation_mae_db": 0.43084275919506304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9321020219838778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14307738894704375, "1000–3000 Hz_10ms_envelope_correlation": 0.9861321772913119, "1000–3000 Hz_10ms_modulation_mae_db": 0.40861388772085067, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3526181975200733, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07667741585391802, "1000–3000 Hz_40ms_envelope_correlation": 0.9962384976951477, "1000–3000 Hz_40ms_modulation_mae_db": 0.37661542172186113, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7986376616959037, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16493784626281002, "3000–6000 Hz_10ms_envelope_correlation": 0.9814962949857011, "3000–6000 Hz_10ms_modulation_mae_db": 0.2567026638643865, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2175891601894864, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.078638284553698, "3000–6000 Hz_40ms_envelope_correlation": 0.9955518613895796, "3000–6000 Hz_40ms_modulation_mae_db": 0.1754652212392568, "mr_stft_db": 5.867599814457722, "band_mae_db": 0.12260624265000315, "rms_gain_error_db": 0.01182510071176845, "output_peak": 0.5923035668408497}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7177406872627945, "全帯域_10ms_rms_envelope_nrmse": 0.09480354451217071, "全帯域_10ms_envelope_correlation": 0.9910807433949741, "全帯域_10ms_modulation_mae_db": 0.3062960754972613, "全帯域_40ms_log_rms_mae_db": 0.7445746725213956, "全帯域_40ms_rms_envelope_nrmse": 0.05303167428790314, "全帯域_40ms_envelope_correlation": 0.9970743447959006, "全帯域_40ms_modulation_mae_db": 0.18719563494539884, "300–1000 Hz_10ms_log_rms_mae_db": 2.770981048614961, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09925750021027979, "300–1000 Hz_10ms_envelope_correlation": 0.9935837122444318, "300–1000 Hz_10ms_modulation_mae_db": 0.3446174154682436, "300–1000 Hz_40ms_log_rms_mae_db": 2.1009139616079056, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05898227796606657, "300–1000 Hz_40ms_envelope_correlation": 0.9976753217572867, "300–1000 Hz_40ms_modulation_mae_db": 0.7954453374522035, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9223900553465079, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15040438200653, "1000–3000 Hz_10ms_envelope_correlation": 0.9847701160518229, "1000–3000 Hz_10ms_modulation_mae_db": 0.9180736767932198, "1000–3000 Hz_40ms_log_rms_mae_db": 1.384176937522279, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09366376085171071, "1000–3000 Hz_40ms_envelope_correlation": 0.994260306393918, "1000–3000 Hz_40ms_modulation_mae_db": 1.0681313446366087, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8257675988525626, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1983784175242094, "3000–6000 Hz_10ms_envelope_correlation": 0.9732316459564937, "3000–6000 Hz_10ms_modulation_mae_db": 1.2652461847513279, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2155595748465882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10391029866230267, "3000–6000 Hz_40ms_envelope_correlation": 0.9930021310286199, "3000–6000 Hz_40ms_modulation_mae_db": 1.2965429485490576, "mr_stft_db": 5.877621353939898, "band_mae_db": 0.1526097941608053, "rms_gain_error_db": -0.05886511042759489, "output_peak": 0.5967107951073318}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7457895876528113, "全帯域_10ms_rms_envelope_nrmse": 0.09342616170382874, "全帯域_10ms_envelope_correlation": 0.9914641344069168, "全帯域_10ms_modulation_mae_db": 0.4263391612839585, "全帯域_40ms_log_rms_mae_db": 0.7673466952594992, "全帯域_40ms_rms_envelope_nrmse": 0.04816440341411659, "全帯域_40ms_envelope_correlation": 0.9978007537202327, "全帯域_40ms_modulation_mae_db": 0.21105448886331454, "300–1000 Hz_10ms_log_rms_mae_db": 2.793507791312494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10566391307674956, "300–1000 Hz_10ms_envelope_correlation": 0.9928687113369037, "300–1000 Hz_10ms_modulation_mae_db": 0.8081268590517041, "300–1000 Hz_40ms_log_rms_mae_db": 2.160933833450796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0619695014593475, "300–1000 Hz_40ms_envelope_correlation": 0.9977257346567637, "300–1000 Hz_40ms_modulation_mae_db": 0.4725741074814587, "1000–3000 Hz_10ms_log_rms_mae_db": 1.871358224961606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15896828958238146, "1000–3000 Hz_10ms_envelope_correlation": 0.9830750800079234, "1000–3000 Hz_10ms_modulation_mae_db": 1.201803539096395, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2742825717711705, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09245938635004557, "1000–3000 Hz_40ms_envelope_correlation": 0.9947223761816428, "1000–3000 Hz_40ms_modulation_mae_db": 1.1493505397077854, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8906361167833388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.206246906169288, "3000–6000 Hz_10ms_envelope_correlation": 0.9725692296610615, "3000–6000 Hz_10ms_modulation_mae_db": 1.403061162311431, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3108464882389868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12856352586706132, "3000–6000 Hz_40ms_envelope_correlation": 0.9891091225675174, "3000–6000 Hz_40ms_modulation_mae_db": 0.9947741741471704, "mr_stft_db": 5.879732740928987, "band_mae_db": 0.14534573293092756, "rms_gain_error_db": -0.14591315363751434, "output_peak": 0.5623265863059312}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.632761239280791, "全帯域_10ms_rms_envelope_nrmse": 0.09313271570140469, "全帯域_10ms_envelope_correlation": 0.9914520373763762, "全帯域_10ms_modulation_mae_db": 0.3625241233289112, "全帯域_40ms_log_rms_mae_db": 0.762647711378576, "全帯域_40ms_rms_envelope_nrmse": 0.04412528182142858, "全帯域_40ms_envelope_correlation": 0.9981271083746628, "全帯域_40ms_modulation_mae_db": 0.48159161342055473, "300–1000 Hz_10ms_log_rms_mae_db": 2.828988749828827, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1060848427005606, "300–1000 Hz_10ms_envelope_correlation": 0.9926417886466131, "300–1000 Hz_10ms_modulation_mae_db": 0.7267544006274115, "300–1000 Hz_40ms_log_rms_mae_db": 2.088985470806244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05324523897895059, "300–1000 Hz_40ms_envelope_correlation": 0.9982407895204233, "300–1000 Hz_40ms_modulation_mae_db": 0.435405235036812, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9282076045244712, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1495778333120044, "1000–3000 Hz_10ms_envelope_correlation": 0.984852863561613, "1000–3000 Hz_10ms_modulation_mae_db": 0.8734419142348241, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4135667381158359, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09081706423852463, "1000–3000 Hz_40ms_envelope_correlation": 0.9944939012885908, "1000–3000 Hz_40ms_modulation_mae_db": 1.21355375160814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8068170332300688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15312556566836386, "3000–6000 Hz_10ms_envelope_correlation": 0.9841670013888416, "3000–6000 Hz_10ms_modulation_mae_db": 0.9926354787900843, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2563787942252629, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08612673041797098, "3000–6000 Hz_40ms_envelope_correlation": 0.9946823929691947, "3000–6000 Hz_40ms_modulation_mae_db": 0.8017890316758999, "mr_stft_db": 5.889213058003083, "band_mae_db": 0.149022845507333, "rms_gain_error_db": -0.12597341676193402, "output_peak": 0.5585016519978394}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7155841810371337, "全帯域_10ms_rms_envelope_nrmse": 0.09543761375713161, "全帯域_10ms_envelope_correlation": 0.9912029962581602, "全帯域_10ms_modulation_mae_db": 0.34702732165813116, "全帯域_40ms_log_rms_mae_db": 0.8131069091431999, "全帯域_40ms_rms_envelope_nrmse": 0.0503887530583669, "全帯域_40ms_envelope_correlation": 0.9977211550614632, "全帯域_40ms_modulation_mae_db": 0.49883136856870847, "300–1000 Hz_10ms_log_rms_mae_db": 2.8058513542301027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10800000964086252, "300–1000 Hz_10ms_envelope_correlation": 0.9924757278676414, "300–1000 Hz_10ms_modulation_mae_db": 0.6268850357048881, "300–1000 Hz_40ms_log_rms_mae_db": 2.1330741732643475, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06244039224222803, "300–1000 Hz_40ms_envelope_correlation": 0.9975975985611134, "300–1000 Hz_40ms_modulation_mae_db": 0.5024745532451834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9002405409639396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16928902763698891, "1000–3000 Hz_10ms_envelope_correlation": 0.9820723659458614, "1000–3000 Hz_10ms_modulation_mae_db": 1.6281060896117596, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3766316803652985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12378751395649908, "1000–3000 Hz_40ms_envelope_correlation": 0.9913949886546263, "1000–3000 Hz_40ms_modulation_mae_db": 1.8634364998279438, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7470774983653483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15845651956156703, "3000–6000 Hz_10ms_envelope_correlation": 0.9828519117850435, "3000–6000 Hz_10ms_modulation_mae_db": 0.5080278743875998, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1754808010377809, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08519935587615547, "3000–6000 Hz_40ms_envelope_correlation": 0.9947744540529511, "3000–6000 Hz_40ms_modulation_mae_db": 0.4511319893841481, "mr_stft_db": 5.868600698602791, "band_mae_db": 0.2490291799219803, "rms_gain_error_db": -0.17702172599210392, "output_peak": 0.534693934956466}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48320, "output_samples": 48320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.074726364623212, "spectral_convergence": 0.31732712077344527, "spectral_convergence_status": "ok", "amplitude_floor": 1.6806044167073843e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.702560867017591, "spectral_convergence": 0.27763938171153696, "spectral_convergence_status": "ok", "amplitude_floor": 1.0917469705551035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.874528721745367, "spectral_convergence": 0.2980876302045022, "spectral_convergence_status": "ok", "amplitude_floor": 1.0075473322757857e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0009427012142024427, "output_power": 0.0009266881697493866, "delta_db": -0.07440447671474917, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024186117972299723, "output_power": 0.0023435829206521043, "delta_db": -0.1368584255850635, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0014180970240603914, "output_power": 0.0014144074902782662, "delta_db": -0.011313980687322393, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023968394187503082, "output_power": 0.00021240655754968876, "delta_db": -0.5247101811427197, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.504236566794763e-05, "output_power": 5.736949321379437e-05, "delta_db": 0.17983922274483755, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00019111878985838765, "output_power": 0.00016740244478840997, "delta_db": -0.5754159059714369, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2504236988076882, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11992058033552455, "pitch": null, "output_rms_dbfs": -22.90572499871463, "output_peak": 0.5557188932077991, "reference_rms_dbfs": -22.785804418379104, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.883938651128723}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00023057786100875665, "log_rms_mae_db": 1.7470581462578256, "rms_envelope_nrmse": 0.08613500469300539, "envelope_correlation": 0.9927724490679601, "rms_gain_error_db": -0.12015453972319995, "reference_mean_power": 0.005273848963429562, "output_mean_power": 0.005129939390341323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.9211277816293244e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014051118338624634, "output_power": 0.0013874707812251122, "reference_active": true, "delta_db": -0.05487045352367537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001364841166272218, "output_power": 0.00015050916138919352, "reference_active": true, "delta_db": 0.424808227511887}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.767297964659028e-05, "output_power": 3.3437253794209266e-05, "reference_active": true, "delta_db": -0.5179936938497719}], "active_band_count": 3, "active_band_mae_db": 0.3325574582951114}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00020448142520679275, "log_rms_mae_db": 0.7895163185823865, "rms_envelope_nrmse": 0.04326129730529666, "envelope_correlation": 0.9982146301112265, "rms_gain_error_db": -0.1202794160731605, "reference_mean_power": 0.005325056781288949, "output_mean_power": 0.005179600944938006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.674469134143454e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013611842323655793, "output_power": 0.0013373264725520535, "reference_active": true, "delta_db": -0.07679468081456876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.483051409804231e-05, "output_power": 7.277391656503077e-05, "reference_active": true, "delta_db": 0.5019628300194073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6716298904493925e-06, "output_power": 1.7308159841980695e-06, "reference_active": true, "delta_db": 0.15110769113138733}], "active_band_count": 3, "active_band_mae_db": 0.24328840065512114}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00020294603022145288, "log_rms_mae_db": 2.9620398403418835, "rms_envelope_nrmse": 0.09652610464974551, "envelope_correlation": 0.9939579216821233, "rms_gain_error_db": 0.005252007984870242, "reference_mean_power": 0.002254450133685994, "output_mean_power": 0.00225717813344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.5930281169536517e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007344814515063544, "output_power": 0.0007383514589500462, "reference_active": true, "delta_db": 0.022823047120258844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010026570767730039, "output_power": 0.00010017123827946192, "reference_active": true, "delta_db": -0.004093810253602252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.614259763381322e-05, "output_power": 1.8336203947213466e-05, "reference_active": true, "delta_db": 0.5533600903064463}], "active_band_count": 3, "active_band_mae_db": 0.1934256492267691}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00017314576936579135, "log_rms_mae_db": 2.212877344504041, "rms_envelope_nrmse": 0.05076479481631307, "envelope_correlation": 0.9983828745911622, "rms_gain_error_db": 0.005235126997409482, "reference_mean_power": 0.002277104155750591, "output_mean_power": 0.0022798507062513906}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.4781762283606311e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007359452368517945, "output_power": 0.0007422783406336475, "reference_active": true, "delta_db": 0.03721289385260631}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.796294327084698e-05, "output_power": 4.734360560988885e-05, "reference_active": true, "delta_db": -0.05644495955695239}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4655779178142778e-06, "output_power": 1.3822239293521505e-06, "reference_active": true, "delta_db": -0.25430505466640824}], "active_band_count": 3, "active_band_mae_db": 0.11598763602532232}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 8.794013668187323e-05, "log_rms_mae_db": 1.9069891701901245, "rms_envelope_nrmse": 0.15503511729351604, "envelope_correlation": 0.9842401467892337, "rms_gain_error_db": -0.37966991182503274, "reference_mean_power": 0.0002815001552852329, "output_mean_power": 0.00025793581406901965}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.4039983283954285e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014419558923291274, "output_power": 0.00012770526145847248, "reference_active": true, "delta_db": -0.5274318547797283}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9143053212884802e-05, "output_power": 1.3714908727463059e-05, "reference_active": true, "delta_db": -1.4481828484731538}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.190405195331938e-06, "output_power": 3.127830587508803e-06, "reference_active": true, "delta_db": -1.2701279716665945}], "active_band_count": 3, "active_band_mae_db": 1.0819142249731588}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 7.034799916708663e-05, "log_rms_mae_db": 1.3924047921270137, "rms_envelope_nrmse": 0.10391559526448554, "envelope_correlation": 0.9933753376544742, "rms_gain_error_db": -0.3797050018731053, "reference_mean_power": 0.00028432562712779955, "output_mean_power": 0.0002605226610579872}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.2325770901894182e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001384543016382475, "output_power": 0.00012133754464366291, "reference_active": true, "delta_db": -0.5731125096936398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.792343358100607e-06, "output_power": 6.7958254144781385e-06, "reference_active": true, "delta_db": -1.586444198763154}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0751348414140544e-07, "output_power": 1.9754812399271784e-07, "reference_active": true, "delta_db": -1.9219125410504483}], "active_band_count": 3, "active_band_mae_db": 1.3604897498357473}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 5.049105990445221e-05, "log_rms_mae_db": 1.8147661669783843, "rms_envelope_nrmse": 0.1590806973174034, "envelope_correlation": 0.9838000232438278, "rms_gain_error_db": -0.4272466119180545, "reference_mean_power": 0.00010138440655269997, "output_mean_power": 9.188539829376863e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.3079247184162883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1617157093216335e-05, "output_power": 4.674438729563606e-05, "reference_active": true, "delta_db": -0.43064609707345886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0705364401401438e-05, "output_power": 1.3902774496180275e-05, "reference_active": true, "delta_db": -1.7298139967799997}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.974689061207133e-06, "output_power": 2.625503843799093e-06, "reference_active": true, "delta_db": -0.5422891799032814}], "active_band_count": 3, "active_band_mae_db": 0.90091642458558}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 4.0590764349045424e-05, "log_rms_mae_db": 1.2722816454805477, "rms_envelope_nrmse": 0.09255739422427309, "envelope_correlation": 0.9954095893551598, "rms_gain_error_db": -0.42727264848425683, "reference_mean_power": 0.00010240265187942684, "output_mean_power": 9.280768477376496e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.2076196938321023e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.154744016518151e-05, "output_power": 5.53295611643297e-05, "reference_active": true, "delta_db": -0.46252769088195345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.945675894844695e-06, "output_power": 6.675640707970623e-06, "reference_active": true, "delta_db": -1.7314134821741058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5378071346519667e-07, "output_power": 2.241870144352967e-07, "reference_active": true, "delta_db": -0.5384816061376023}], "active_band_count": 3, "active_band_mae_db": 0.9108075930645539}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7850917950006867, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "53ef5b722fed9570344f68160a9d2309f93551b247da3639f711bc7b3ce2570e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9833900807575917, "standard_deviation": 0.0021802602491752454, "minimum": 0.979036727186446, "maximum": 0.9861321772913119, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9103007994724748, "standard_deviation": 0.01992638316392165, "minimum": 1.871358224961606, "maximum": 1.9321020219838778, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9227715891460202, "standard_deviation": 0.40625976930798613, "minimum": 0.387789381082907, "maximum": 1.6281060896117596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15740611871244775, "standard_deviation": 0.010485949823906254, "minimum": 0.14307738894704375, "maximum": 0.17485303404219296, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942861346190837, "standard_deviation": 0.001425232382562451, "minimum": 0.9913949886546263, "maximum": 0.9962384976951477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3619546910120752, "standard_deviation": 0.04225008782631341, "minimum": 1.2742825717711705, "maximum": 1.4135667381158359, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0004847791525662, "standard_deviation": 0.5266290539956038, "minimum": 0.2903829220464023, "maximum": 1.8634364998279438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09441804355656604, "standard_deviation": 0.014182060205889796, "minimum": 0.07667741585391802, "maximum": 0.12378751395649908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9782468975134535, "standard_deviation": 0.005623665532042698, "minimum": 0.9701175914655054, "maximum": 0.9841670013888416, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8203100839164146, "standard_deviation": 0.044260473325849564, "minimum": 1.7470774983653483, "maximum": 1.8906361167833388, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8974308409695877, "standard_deviation": 0.3885283334726475, "minimum": 0.2567026638643865, "maximum": 1.403061162311431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17877015513060399, "standard_deviation": 0.023084587847895046, "minimum": 0.15312556566836386, "maximum": 0.20940402168706257, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9931970276835562, "standard_deviation": 0.002320894256429544, "minimum": 0.9891091225675174, "maximum": 0.9955518613895796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2456093219251163, "standard_deviation": 0.05425271855719825, "minimum": 1.1754808010377809, "maximum": 1.3238994769976355, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9086337138947649, "standard_deviation": 0.45643251107944566, "minimum": 0.1754652212392568, "maximum": 1.6393548426768587, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09983996104564463, "standard_deviation": 0.01705517113353709, "minimum": 0.078638284553698, "maximum": 0.12856352586706132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932331992412116, "standard_deviation": 0.0010683939508182507, "minimum": 0.991423024414471, "maximum": 0.9945346670113221, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.8189714633194995, "standard_deviation": 0.08760218512728633, "minimum": 2.667649503942691, "maximum": 2.9620398403418835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48810302917679427, "standard_deviation": 0.2535933570975107, "minimum": 0.1934256492267691, "maximum": 0.8081268590517041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10220446733676197, "standard_deviation": 0.008239433564108155, "minimum": 0.0916407151096829, "maximum": 0.11627124141736339, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979208876208282, "standard_deviation": 0.0007311886496455201, "minimum": 0.9964491373889486, "maximum": 0.9986686665803438, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.1315878674245177, "standard_deviation": 0.10932854650770384, "minimum": 1.9133656065043771, "maximum": 2.291036274015993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44412950706242116, "standard_deviation": 0.19303695598767034, "minimum": 0.11598763602532232, "maximum": 0.7954453374522035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.057196767474461424, "standard_deviation": 0.009357163281254218, "minimum": 0.045981327647090925, "maximum": 0.07491463731614748, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16997219755000859, "standard_deviation": 0.05411839375561145, "minimum": 0.10678055992519077, "maximum": 0.2504236988076882, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.878510209143689, "standard_deviation": 0.007578592933123616, "minimum": 5.867599814457722, "maximum": 5.889213058003083, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5640463049127709, "standard_deviation": 0.021307384860931523, "minimum": 0.534693934956466, "maximum": 0.5967107951073318, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09221569019120363, "standard_deviation": 0.06447815861001796, "minimum": -0.17702172599210392, "maximum": 0.01182510071176845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9916426920394223, "standard_deviation": 0.0005621663674085743, "minimum": 0.9910807433949741, "maximum": 0.9927724490679601, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7086868735188558, "standard_deviation": 0.038935603447065446, "minimum": 1.632761239280791, "maximum": 1.7470581462578256, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3032450134871528, "standard_deviation": 0.12474457779093553, "minimum": 0.06915091037728564, "maximum": 0.4263391612839585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09212593844741604, "standard_deviation": 0.0031768826524780397, "minimum": 0.08613500469300539, "maximum": 0.09543761375713161, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978200669580721, "standard_deviation": 0.00037510906063428274, "minimum": 0.9970743447959006, "maximum": 0.9982146301112265, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7694276745720888, "standard_deviation": 0.041863003175209385, "minimum": 0.6845313398500258, "maximum": 0.8138333663753078, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2516890712118469, "standard_deviation": 0.16500153702580939, "minimum": 0.05397798792707013, "maximum": 0.49883136856870847, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04685133371647651, "standard_deviation": 0.0036765079823158553, "minimum": 0.04326129730529666, "maximum": 0.05303167428790314, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2830-3980-0042", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6828958988189697, "predicted_mos": 2.6828958988189697, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "74806cf27f57e7db073f97ac794b9e0f2e26101551d4474e2398123169c51214", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.040252299979329, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7443089632526425, "全帯域_10ms_rms_envelope_nrmse": 0.09490362461957506, "全帯域_10ms_envelope_correlation": 0.991051791271481, "全帯域_10ms_modulation_mae_db": 0.20369480436979134, "全帯域_40ms_log_rms_mae_db": 0.8019779014229123, "全帯域_40ms_rms_envelope_nrmse": 0.05024094882153537, "全帯域_40ms_envelope_correlation": 0.997329739604791, "全帯域_40ms_modulation_mae_db": 0.48951996387476937, "300–1000 Hz_10ms_log_rms_mae_db": 2.6618194933495483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09203856555783603, "300–1000 Hz_10ms_envelope_correlation": 0.9945275447829804, "300–1000 Hz_10ms_modulation_mae_db": 0.30239146012860735, "300–1000 Hz_40ms_log_rms_mae_db": 1.9968397284375174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05788582719679428, "300–1000 Hz_40ms_envelope_correlation": 0.9978064340174805, "300–1000 Hz_40ms_modulation_mae_db": 0.12128839999849683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.723174374015662, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14311141343234102, "1000–3000 Hz_10ms_envelope_correlation": 0.9859445663966246, "1000–3000 Hz_10ms_modulation_mae_db": 0.5727716663091112, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2090002564556361, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07482593722376839, "1000–3000 Hz_40ms_envelope_correlation": 0.996019832581278, "1000–3000 Hz_40ms_modulation_mae_db": 0.2136311968425957, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7020181786494442, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18720784186039832, "3000–6000 Hz_10ms_envelope_correlation": 0.9763970398566946, "3000–6000 Hz_10ms_modulation_mae_db": 0.3808413576190363, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0671770850323228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10250230479948212, "3000–6000 Hz_40ms_envelope_correlation": 0.9925836361491259, "3000–6000 Hz_40ms_modulation_mae_db": 0.5484464079738604, "mr_stft_db": 5.803156802196811, "band_mae_db": 0.16226478021474594, "rms_gain_error_db": 0.01577246233019561, "output_peak": 0.5367699696235401}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6922281883259387, "全帯域_10ms_rms_envelope_nrmse": 0.09497532527760727, "全帯域_10ms_envelope_correlation": 0.9910412505522871, "全帯域_10ms_modulation_mae_db": 0.445056405316776, "全帯域_40ms_log_rms_mae_db": 0.9181263411326035, "全帯域_40ms_rms_envelope_nrmse": 0.05066376874870827, "全帯域_40ms_envelope_correlation": 0.9972858360320659, "全帯域_40ms_modulation_mae_db": 0.5298443820680977, "300–1000 Hz_10ms_log_rms_mae_db": 2.7517832232208153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1068724646256033, "300–1000 Hz_10ms_envelope_correlation": 0.9925521488191231, "300–1000 Hz_10ms_modulation_mae_db": 0.4806353561488957, "300–1000 Hz_40ms_log_rms_mae_db": 2.1163980223817056, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05983849663732887, "300–1000 Hz_40ms_envelope_correlation": 0.997636373919491, "300–1000 Hz_40ms_modulation_mae_db": 0.3167844278243199, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7664792745842863, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14597707496375859, "1000–3000 Hz_10ms_envelope_correlation": 0.9853575043746989, "1000–3000 Hz_10ms_modulation_mae_db": 0.594366170083855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1904510946538576, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06558203153247816, "1000–3000 Hz_40ms_envelope_correlation": 0.9969695439439401, "1000–3000 Hz_40ms_modulation_mae_db": 0.29599320978678095, "3000–6000 Hz_10ms_log_rms_mae_db": 1.677462527465704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16518596710861827, "3000–6000 Hz_10ms_envelope_correlation": 0.9813249587191896, "3000–6000 Hz_10ms_modulation_mae_db": 0.4317090312175826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.104223877717435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10230205538340262, "3000–6000 Hz_40ms_envelope_correlation": 0.9924897443257493, "3000–6000 Hz_40ms_modulation_mae_db": 0.32493224240391916, "mr_stft_db": 5.797438995355052, "band_mae_db": 0.16262332015705191, "rms_gain_error_db": 0.007030955621259331, "output_peak": 0.5679565644016944}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7819594767654654, "全帯域_10ms_rms_envelope_nrmse": 0.09328722846855517, "全帯域_10ms_envelope_correlation": 0.9913778548803596, "全帯域_10ms_modulation_mae_db": 0.12068237229174306, "全帯域_40ms_log_rms_mae_db": 0.8133054052147815, "全帯域_40ms_rms_envelope_nrmse": 0.04885670026114473, "全帯域_40ms_envelope_correlation": 0.9976142093339075, "全帯域_40ms_modulation_mae_db": 0.08360407658391687, "300–1000 Hz_10ms_log_rms_mae_db": 2.6720674392966552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09642282409341835, "300–1000 Hz_10ms_envelope_correlation": 0.9940259371896936, "300–1000 Hz_10ms_modulation_mae_db": 0.31464823943359127, "300–1000 Hz_40ms_log_rms_mae_db": 2.0105962600421203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061206261737546064, "300–1000 Hz_40ms_envelope_correlation": 0.9976544935415068, "300–1000 Hz_40ms_modulation_mae_db": 0.6479991307319098, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8375409646071492, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16780199389023312, "1000–3000 Hz_10ms_envelope_correlation": 0.9806561595674804, "1000–3000 Hz_10ms_modulation_mae_db": 0.14248702092303867, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2427857500109283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08857918266502236, "1000–3000 Hz_40ms_envelope_correlation": 0.9946360449093729, "1000–3000 Hz_40ms_modulation_mae_db": 0.24070941078281669, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6191639663830046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20034444303854468, "3000–6000 Hz_10ms_envelope_correlation": 0.9725838133050347, "3000–6000 Hz_10ms_modulation_mae_db": 0.6546765056221272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.038217462452247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10124915195810959, "3000–6000 Hz_40ms_envelope_correlation": 0.9925826898464, "3000–6000 Hz_40ms_modulation_mae_db": 0.7793388102351543, "mr_stft_db": 5.806808629743249, "band_mae_db": 0.1612872029601826, "rms_gain_error_db": -0.11167438637316049, "output_peak": 0.6029860802291611}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6901769943327243, "全帯域_10ms_rms_envelope_nrmse": 0.09580419736967212, "全帯域_10ms_envelope_correlation": 0.9912205819058248, "全帯域_10ms_modulation_mae_db": 0.28216721309318, "全帯域_40ms_log_rms_mae_db": 0.800561958350472, "全帯域_40ms_rms_envelope_nrmse": 0.05527724901197775, "全帯域_40ms_envelope_correlation": 0.9972968684135332, "全帯域_40ms_modulation_mae_db": 0.14929823671113626, "300–1000 Hz_10ms_log_rms_mae_db": 2.8612111103646933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11451268368098874, "300–1000 Hz_10ms_envelope_correlation": 0.9918025921193471, "300–1000 Hz_10ms_modulation_mae_db": 0.3105289998917294, "300–1000 Hz_40ms_log_rms_mae_db": 2.12217144298343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07767384882061387, "300–1000 Hz_40ms_envelope_correlation": 0.9965055871879911, "300–1000 Hz_40ms_modulation_mae_db": 0.6215195020173173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.760116376184791, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14673336121192257, "1000–3000 Hz_10ms_envelope_correlation": 0.9852679078791748, "1000–3000 Hz_10ms_modulation_mae_db": 0.8024290933971239, "1000–3000 Hz_40ms_log_rms_mae_db": 1.192160086571911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07534171418812254, "1000–3000 Hz_40ms_envelope_correlation": 0.9960774349061287, "1000–3000 Hz_40ms_modulation_mae_db": 0.6880347370226171, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6903448943308317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18044845418562955, "3000–6000 Hz_10ms_envelope_correlation": 0.9776983657182501, "3000–6000 Hz_10ms_modulation_mae_db": 0.9027363688055837, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1044534839943783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09709286793918458, "3000–6000 Hz_40ms_envelope_correlation": 0.993239930748238, "3000–6000 Hz_40ms_modulation_mae_db": 1.6643016274338454, "mr_stft_db": 5.811570523727132, "band_mae_db": 0.18874646823244687, "rms_gain_error_db": -0.20856252894343225, "output_peak": 0.5473028478336877}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7086660907121989, "全帯域_10ms_rms_envelope_nrmse": 0.08325707119369642, "全帯域_10ms_envelope_correlation": 0.993144037083265, "全帯域_10ms_modulation_mae_db": 0.21212440963392165, "全帯域_40ms_log_rms_mae_db": 0.7678300693064422, "全帯域_40ms_rms_envelope_nrmse": 0.03736040748353517, "全帯域_40ms_envelope_correlation": 0.9985982608226417, "全帯域_40ms_modulation_mae_db": 0.16331065419205537, "300–1000 Hz_10ms_log_rms_mae_db": 2.601475381052857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08556985249300084, "300–1000 Hz_10ms_envelope_correlation": 0.9951964847463711, "300–1000 Hz_10ms_modulation_mae_db": 0.18221975210767768, "300–1000 Hz_40ms_log_rms_mae_db": 1.9386811629960414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042408842384680866, "300–1000 Hz_40ms_envelope_correlation": 0.9988200238752485, "300–1000 Hz_40ms_modulation_mae_db": 0.6397986612834586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.846628630106399, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13771297676585484, "1000–3000 Hz_10ms_envelope_correlation": 0.9871046118221953, "1000–3000 Hz_10ms_modulation_mae_db": 0.49589940332993193, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2101599479573448, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0828440449780945, "1000–3000 Hz_40ms_envelope_correlation": 0.9953388885777716, "1000–3000 Hz_40ms_modulation_mae_db": 0.5150040786959055, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6318525788396885, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17738415989059975, "3000–6000 Hz_10ms_envelope_correlation": 0.9786020074349215, "3000–6000 Hz_10ms_modulation_mae_db": 0.8120714881062527, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0186945809506158, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08555647615459439, "3000–6000 Hz_40ms_envelope_correlation": 0.9947155661514832, "3000–6000 Hz_40ms_modulation_mae_db": 0.6312869516816906, "mr_stft_db": 5.804634642795978, "band_mae_db": 0.1547390698241038, "rms_gain_error_db": -0.07310177832555795, "output_peak": 0.602470952870654}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6947812701218374, "全帯域_10ms_rms_envelope_nrmse": 0.08735359521631031, "全帯域_10ms_envelope_correlation": 0.9924286317840537, "全帯域_10ms_modulation_mae_db": 0.16422466684954842, "全帯域_40ms_log_rms_mae_db": 0.7688248417360499, "全帯域_40ms_rms_envelope_nrmse": 0.042583904850535016, "全帯域_40ms_envelope_correlation": 0.9981559950607047, "全帯域_40ms_modulation_mae_db": 0.26799126773023635, "300–1000 Hz_10ms_log_rms_mae_db": 2.7445262586115247, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08830586633138933, "300–1000 Hz_10ms_envelope_correlation": 0.9950374244944024, "300–1000 Hz_10ms_modulation_mae_db": 0.3729005999293498, "300–1000 Hz_40ms_log_rms_mae_db": 2.0841702845283514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051127571738813, "300–1000 Hz_40ms_envelope_correlation": 0.9984635711786782, "300–1000 Hz_40ms_modulation_mae_db": 0.5089862960472165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.806437920540531, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15742884534696125, "1000–3000 Hz_10ms_envelope_correlation": 0.9837422602894362, "1000–3000 Hz_10ms_modulation_mae_db": 1.038131567181923, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2246258311443716, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10317026542450745, "1000–3000 Hz_40ms_envelope_correlation": 0.9935534147770171, "1000–3000 Hz_40ms_modulation_mae_db": 1.1543970418699667, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7468819001105431, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18570301311574047, "3000–6000 Hz_10ms_envelope_correlation": 0.9763874451789963, "3000–6000 Hz_10ms_modulation_mae_db": 1.2249912391184223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1593515407447306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10236964285725886, "3000–6000 Hz_40ms_envelope_correlation": 0.9926761250113733, "3000–6000 Hz_40ms_modulation_mae_db": 0.6778431368815525, "mr_stft_db": 5.813242929813704, "band_mae_db": 0.24112463357143843, "rms_gain_error_db": -0.08244029408728092, "output_peak": 0.5683344191320667}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7093522609405167, "全帯域_10ms_rms_envelope_nrmse": 0.08934524173971085, "全帯域_10ms_envelope_correlation": 0.9920209028124319, "全帯域_10ms_modulation_mae_db": 0.2331529647605196, "全帯域_40ms_log_rms_mae_db": 0.8263525738983158, "全帯域_40ms_rms_envelope_nrmse": 0.04548233977668087, "全帯域_40ms_envelope_correlation": 0.9978518622676266, "全帯域_40ms_modulation_mae_db": 0.13764020965645676, "300–1000 Hz_10ms_log_rms_mae_db": 2.759211577808702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09955035541939093, "300–1000 Hz_10ms_envelope_correlation": 0.9935125835562836, "300–1000 Hz_10ms_modulation_mae_db": 0.8020393212808167, "300–1000 Hz_40ms_log_rms_mae_db": 2.08851935421518, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05181535271315617, "300–1000 Hz_40ms_envelope_correlation": 0.9982740880784152, "300–1000 Hz_40ms_modulation_mae_db": 0.4096478977134832, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7074379950283738, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15333429376552615, "1000–3000 Hz_10ms_envelope_correlation": 0.9839108109859822, "1000–3000 Hz_10ms_modulation_mae_db": 1.2397669305554888, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1751469479748222, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08195655203256982, "1000–3000 Hz_40ms_envelope_correlation": 0.9954677999919184, "1000–3000 Hz_40ms_modulation_mae_db": 0.9896637315004523, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6858292389547604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16954505921286966, "3000–6000 Hz_10ms_envelope_correlation": 0.980437306958637, "3000–6000 Hz_10ms_modulation_mae_db": 0.6748748999557811, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0895350003472282, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08666436759005917, "3000–6000 Hz_40ms_envelope_correlation": 0.9945709874342925, "3000–6000 Hz_40ms_modulation_mae_db": 0.9407727310316423, "mr_stft_db": 5.8134180058874385, "band_mae_db": 0.16999353800841163, "rms_gain_error_db": -0.04570851071591632, "output_peak": 0.5294092440614436}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6867705558829875, "全帯域_10ms_rms_envelope_nrmse": 0.08562169690279689, "全帯域_10ms_envelope_correlation": 0.9927838160179164, "全帯域_10ms_modulation_mae_db": 0.30978535903602306, "全帯域_40ms_log_rms_mae_db": 0.7879633645754797, "全帯域_40ms_rms_envelope_nrmse": 0.0455355353139976, "全帯域_40ms_envelope_correlation": 0.9979460447232493, "全帯域_40ms_modulation_mae_db": 0.2842635964618875, "300–1000 Hz_10ms_log_rms_mae_db": 2.633924671249927, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09709724573038843, "300–1000 Hz_10ms_envelope_correlation": 0.9937927161964462, "300–1000 Hz_10ms_modulation_mae_db": 0.2771935069307308, "300–1000 Hz_40ms_log_rms_mae_db": 1.8843543568536794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05304454411706041, "300–1000 Hz_40ms_envelope_correlation": 0.9981307406464773, "300–1000 Hz_40ms_modulation_mae_db": 0.2676923740592169, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9100574594741395, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16277266820093947, "1000–3000 Hz_10ms_envelope_correlation": 0.9821824631451711, "1000–3000 Hz_10ms_modulation_mae_db": 1.231823083110917, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3320133411112816, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10045426389156151, "1000–3000 Hz_40ms_envelope_correlation": 0.9934758015850198, "1000–3000 Hz_40ms_modulation_mae_db": 1.4442751395315911, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7294355673121518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16238812871840144, "3000–6000 Hz_10ms_envelope_correlation": 0.9821507413091921, "3000–6000 Hz_10ms_modulation_mae_db": 0.552830197385138, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0834071983839113, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09069373935581056, "3000–6000 Hz_40ms_envelope_correlation": 0.9945069773785241, "3000–6000 Hz_40ms_modulation_mae_db": 0.9739891563337878, "mr_stft_db": 5.825133715747582, "band_mae_db": 0.2296545991134771, "rms_gain_error_db": -0.11115303852377084, "output_peak": 0.5457949289147628}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48320, "output_samples": 48320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.019534617635597, "spectral_convergence": 0.33093986652848617, "spectral_convergence_status": "ok", "amplitude_floor": 1.6806044167073843e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.615497335609449, "spectral_convergence": 0.2820163058073193, "spectral_convergence_status": "ok", "amplitude_floor": 1.0917469705551035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.774438453345385, "spectral_convergence": 0.2982382043229837, "spectral_convergence_status": "ok", "amplitude_floor": 1.0075473322757857e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0009427012142024427, "output_power": 0.0010184857324937206, "delta_db": 0.33580883089109287, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024186117972299723, "output_power": 0.002318536791389641, "delta_db": -0.1835217506401465, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0014180970240603914, "output_power": 0.0014734863812157929, "delta_db": 0.1664018057503276, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023968394187503082, "output_power": 0.00023289270238431125, "delta_db": -0.12483058270665127, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.504236566794763e-05, "output_power": 5.5476302524998456e-05, "delta_db": 0.03410416611763559, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00019111878985838765, "output_power": 0.00018552875589686657, "delta_db": -0.1289215451826219, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16226478021474594, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.01577246233019561, "pitch": null, "output_rms_dbfs": -22.77003195604891, "output_peak": 0.5367699696235401, "reference_rms_dbfs": -22.785804418379104, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.803156802196811}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00023057786100875665, "log_rms_mae_db": 1.7443089632526425, "rms_envelope_nrmse": 0.09490362461957506, "envelope_correlation": 0.991051791271481, "rms_gain_error_db": 0.01565793829887394, "reference_mean_power": 0.005273848963429562, "output_mean_power": 0.005292897476759953}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.9211277816293244e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014051118338624634, "output_power": 0.0013427302471646385, "reference_active": true, "delta_db": -0.19722119216391468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001364841166272218, "output_power": 0.00014479560823545197, "reference_active": true, "delta_db": 0.2567327637824451}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.767297964659028e-05, "output_power": 3.906096934924339e-05, "reference_active": true, "delta_db": 0.15713045716301421}], "active_band_count": 3, "active_band_mae_db": 0.20369480436979134}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00020448142520679275, "log_rms_mae_db": 0.8019779014229123, "rms_envelope_nrmse": 0.05024094882153537, "envelope_correlation": 0.997329739604791, "rms_gain_error_db": 0.014963357259126613, "reference_mean_power": 0.005325056781288949, "output_mean_power": 0.005343435590081144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.674469134143454e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013611842323655793, "output_power": 0.0012874176989270388, "reference_active": true, "delta_db": -0.2419743434018656}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.483051409804231e-05, "output_power": 7.182211534705349e-05, "reference_active": true, "delta_db": 0.44478726555501624}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6716298904493925e-06, "output_power": 2.0013357645732574e-06, "reference_active": true, "delta_db": 0.7817982826674262}], "active_band_count": 3, "active_band_mae_db": 0.48951996387476937}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00020294603022145288, "log_rms_mae_db": 2.6618194933495483, "rms_envelope_nrmse": 0.09203856555783603, "envelope_correlation": 0.9945275447829804, "rms_gain_error_db": 0.060215550729784836, "reference_mean_power": 0.002254450133685994, "output_mean_power": 0.002285926112279982}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.5930281169536517e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007344814515063544, "output_power": 0.0007595246844536758, "reference_active": true, "delta_db": 0.1456106029087641}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010026570767730039, "output_power": 9.327781310598408e-05, "reference_active": true, "delta_db": -0.31374067898629565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.614259763381322e-05, "output_power": 1.7895990132986272e-05, "reference_active": true, "delta_db": 0.44782309849076224}], "active_band_count": 3, "active_band_mae_db": 0.30239146012860735}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00017314576936579135, "log_rms_mae_db": 1.9968397284375174, "rms_envelope_nrmse": 0.05788582719679428, "envelope_correlation": 0.9978064340174805, "rms_gain_error_db": 0.06020281089571918, "reference_mean_power": 0.002277104155750591, "output_mean_power": 0.0023088896501928537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.4781762283606311e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007359452368517945, "output_power": 0.0007594877440449877, "reference_active": true, "delta_db": 0.1367527116338219}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.796294327084698e-05, "output_power": 4.786503617949265e-05, "reference_active": true, "delta_db": -0.008874344331057783}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4655779178142778e-06, "output_power": 1.5411067660421845e-06, "reference_active": true, "delta_db": 0.21823814403061081}], "active_band_count": 3, "active_band_mae_db": 0.12128839999849683}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 8.794013668187323e-05, "log_rms_mae_db": 1.723174374015662, "rms_envelope_nrmse": 0.14311141343234102, "envelope_correlation": 0.9859445663966246, "rms_gain_error_db": -0.07054008161001343, "reference_mean_power": 0.0002815001552852329, "output_mean_power": 0.00027696483403257655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.4039983283954285e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014419558923291274, "output_power": 0.00013755994729749887, "reference_active": true, "delta_db": -0.20459975287010423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9143053212884802e-05, "output_power": 1.6932886173578576e-05, "reference_active": true, "delta_db": -0.5328021760630357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.190405195331938e-06, "output_power": 5.252273423069163e-06, "reference_active": true, "delta_db": 0.9809130699941936}], "active_band_count": 3, "active_band_mae_db": 0.5727716663091112}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 7.034799916708663e-05, "log_rms_mae_db": 1.2090002564556361, "rms_envelope_nrmse": 0.07482593722376839, "envelope_correlation": 0.996019832581278, "rms_gain_error_db": -0.07057440538177841, "reference_mean_power": 0.00028432562712779955, "output_mean_power": 0.000279742573063542}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.2325770901894182e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001384543016382475, "output_power": 0.00013276967457004617, "reference_active": true, "delta_db": -0.18207562627778706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.792343358100607e-06, "output_power": 9.03659103951157e-06, "reference_active": true, "delta_db": -0.34882004873248007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0751348414140544e-07, "output_power": 3.154016459746039e-07, "reference_active": true, "delta_db": 0.10999791551752}], "active_band_count": 3, "active_band_mae_db": 0.2136311968425957}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 5.049105990445221e-05, "log_rms_mae_db": 1.7020181786494442, "rms_envelope_nrmse": 0.18720784186039832, "envelope_correlation": 0.9763970398566946, "rms_gain_error_db": 0.06417586499910706, "reference_mean_power": 0.00010138440655269997, "output_mean_power": 0.00010289369180263561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.3079247184162883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1617157093216335e-05, "output_power": 5.317965310356689e-05, "reference_active": true, "delta_db": 0.1295141873143006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0705364401401438e-05, "output_power": 2.046288172761543e-05, "reference_active": true, "delta_db": -0.05116084114456964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.974689061207133e-06, "output_power": 2.38372785799091e-06, "reference_active": true, "delta_db": -0.9618490443982387}], "active_band_count": 3, "active_band_mae_db": 0.3808413576190363}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 4.0590764349045424e-05, "log_rms_mae_db": 1.0671770850323228, "rms_envelope_nrmse": 0.10250230479948212, "envelope_correlation": 0.9925836361491259, "rms_gain_error_db": 0.06415794147934777, "reference_mean_power": 0.00010240265187942684, "output_mean_power": 0.0001039266665916768}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.2076196938321023e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.154744016518151e-05, "output_power": 6.388342244077282e-05, "reference_active": true, "delta_db": 0.16178179850737046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.945675894844695e-06, "output_power": 1.0824548691226513e-05, "reference_active": true, "delta_db": 0.3677549579226044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5378071346519667e-07, "output_power": 1.9628104759999745e-07, "reference_active": true, "delta_db": -1.115802467491606}], "active_band_count": 3, "active_band_mae_db": 0.5484464079738604}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7851681811516226, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0d4548d0363eb796a002d7cd011fa942d230cfa91ceb54a5bf4c21133a886f85", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9842707855575954, "standard_deviation": 0.002099076548819091, "minimum": 0.9806561595674804, "maximum": 0.9871046118221953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7947341243176664, "standard_deviation": 0.0683479050318399, "minimum": 1.7074379950283738, "maximum": 1.9100574594741395, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7647093668614238, "standard_deviation": 0.38643678470093246, "minimum": 0.14248702092303867, "maximum": 1.2397669305554888, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15185907844719213, "standard_deviation": 0.010309401644242103, "minimum": 0.13771297676585484, "maximum": 0.16780199389023312, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951923451590559, "standard_deviation": 0.0012343876307332746, "minimum": 0.9934758015850198, "maximum": 0.9969695439439401, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2220429069850192, "standard_deviation": 0.04918188603498761, "minimum": 1.1751469479748222, "maximum": 1.3320133411112816, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6927135682540907, "standard_deviation": 0.4612085209556828, "minimum": 0.2136311968425957, "maximum": 1.4442751395315911, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08409424899201559, "standard_deviation": 0.012889655557074626, "minimum": 0.06558203153247816, "maximum": 0.10317026542450745, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9781977098101144, "standard_deviation": 0.0031366507282395263, "minimum": 0.9725838133050347, "maximum": 0.9821507413091921, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.685373606505766, "standard_deviation": 0.043650383203826, "minimum": 1.6191639663830046, "maximum": 1.7468819001105431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7043413859787404, "standard_deviation": 0.27446410051730347, "minimum": 0.3808413576190363, "maximum": 1.2249912391184223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17852588339135028, "standard_deviation": 0.01268842354987819, "minimum": 0.16238812871840144, "maximum": 0.20034444303854468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9934207071306482, "standard_deviation": 0.001002605383552701, "minimum": 0.9924897443257493, "maximum": 0.9947155661514832, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0831325287028586, "standard_deviation": 0.0433820593112086, "minimum": 1.0186945809506158, "maximum": 1.1593515407447306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8176138829969315, "standard_deviation": 0.40132038521534025, "minimum": 0.32493224240391916, "maximum": 1.6643016274338454, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09605382575473773, "standard_deviation": 0.007325792240276586, "minimum": 0.08555647615459439, "maximum": 0.10250230479948212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.993805928988081, "standard_deviation": 0.0011766118977293944, "minimum": 0.9918025921193471, "maximum": 0.9951964847463711, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.71075239436934, "standard_deviation": 0.08413020498093673, "minimum": 2.601475381052857, "maximum": 2.8612111103646933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38031965448142485, "standard_deviation": 0.19015054451803742, "minimum": 0.18221975210767768, "maximum": 0.8020393212808167, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.097546232241502, "standard_deviation": 0.009557881233595814, "minimum": 0.08556985249300084, "maximum": 0.11451268368098874, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979114140556611, "standard_deviation": 0.0007009290885619441, "minimum": 0.9965055871879911, "maximum": 0.9988200238752485, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.0302163265547533, "standard_deviation": 0.08729804893339901, "minimum": 1.8843543568536794, "maximum": 2.12217144298343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4417145862094274, "standard_deviation": 0.19575436244601596, "minimum": 0.12128839999849683, "maximum": 0.6479991307319098, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056875093168249194, "standard_deviation": 0.010307633059886369, "minimum": 0.042408842384680866, "maximum": 0.07767384882061387, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1838042015102323, "standard_deviation": 0.0335248837796002, "minimum": 0.1547390698241038, "maximum": 0.24112463357143843, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.809425530658368, "standard_deviation": 0.008406079694736888, "minimum": 5.797438995355052, "maximum": 5.825133715747582, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5626281258833763, "standard_deviation": 0.02820214886225158, "minimum": 0.5294092440614436, "maximum": 0.6029860802291611, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07622963987720796, "standard_deviation": 0.07207592821663561, "minimum": -0.20856252894343225, "maximum": 0.01577246233019561, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918836082884525, "standard_deviation": 0.0008290986240929952, "minimum": 0.9910412505522871, "maximum": 0.993144037083265, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.713530475041789, "standard_deviation": 0.03324747414787532, "minimum": 1.6867705558829875, "maximum": 1.7819594767654654, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24636102441893787, "standard_deviation": 0.10035159160213195, "minimum": 0.12068237229174306, "maximum": 0.445056405316776, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0905684975984905, "standard_deviation": 0.004822229056874935, "minimum": 0.08325707119369642, "maximum": 0.09580419736967212, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997759852032315, "standard_deviation": 0.00047029560549886656, "minimum": 0.9972858360320659, "maximum": 0.9985982608226417, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8106178069546321, "standard_deviation": 0.047927512601763166, "minimum": 0.7678300693064422, "maximum": 0.9181263411326035, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.26318404840981957, "standard_deviation": 0.1663722296489507, "minimum": 0.08360407658391687, "maximum": 0.5298443820680977, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.047000106783514344, "standard_deviation": 0.0055130847899109055, "minimum": 0.03736040748353517, "maximum": 0.05527724901197775, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2961-960-0010", "language": "en", "style": "normal", "speaker": "librispeech_2961", "source_utterance_id": "2961-960-0010", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2961-960-0010.wav", "source_file_sha256": "e271404b2bebb5a4f3c062d3ab3ba2aa9e583ab728fb9fa8ceec955a3ca17da0", "decoded_pcm_sha256": "ab9593d38d133538cf3475c5a7bf30b265737c427a5c4a501ecf3384b1e2aee8", "original_source_sha256": "d93e37326bfa987a0cf66c0edd79dfa034ea013e5967c8f2ba9edf54cde8d2be", "corpus_entry_id": "librispeech_2961-960-0010"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.571021556854248, "predicted_mos": 3.571021556854248, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bc8aec5ef2edfa25d6eb1815bef006ce8ce4386f213753706e16be3e1f3c0ee6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.264085464001255, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2961-960-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "bc8aec5ef2edfa25d6eb1815bef006ce8ce4386f213753706e16be3e1f3c0ee6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "78b0dff6046aa18d21883b8e7e1a87010826b65f7bd7e20d432dff9e773dfeb2"}, "rows": [{"case": "librispeech_2961-960-0010", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.4985058307647705, "predicted_mos": 2.4985058307647705, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e72ad821660b21bbebc89664700b1578c0d697b147af86f2818fb544e427678e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.650998397999501, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2961-960-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "e72ad821660b21bbebc89664700b1578c0d697b147af86f2818fb544e427678e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7203330779907231, "全帯域_10ms_rms_envelope_nrmse": 0.0745379954838452, "全帯域_10ms_envelope_correlation": 0.9951212944172624, "全帯域_10ms_modulation_mae_db": 0.8642454403725877, "全帯域_40ms_log_rms_mae_db": 0.40240064991269076, "全帯域_40ms_rms_envelope_nrmse": 0.04417304168524292, "全帯域_40ms_envelope_correlation": 0.9985185150412497, "全帯域_40ms_modulation_mae_db": 0.871808756659122, "300–1000 Hz_10ms_log_rms_mae_db": 1.0835474459555865, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0895021535597851, "300–1000 Hz_10ms_envelope_correlation": 0.9935251849774269, "300–1000 Hz_10ms_modulation_mae_db": 0.9594772840225411, "300–1000 Hz_40ms_log_rms_mae_db": 0.5910340398664904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05333564863076636, "300–1000 Hz_40ms_envelope_correlation": 0.9979366998736852, "300–1000 Hz_40ms_modulation_mae_db": 0.9438226385825385, "1000–3000 Hz_10ms_log_rms_mae_db": 1.109176831937842, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10446529491396957, "1000–3000 Hz_10ms_envelope_correlation": 0.9924089658318996, "1000–3000 Hz_10ms_modulation_mae_db": 0.618825997079132, "1000–3000 Hz_40ms_log_rms_mae_db": 0.595623795013852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05863219219836173, "1000–3000 Hz_40ms_envelope_correlation": 0.9980697017220297, "1000–3000 Hz_40ms_modulation_mae_db": 0.6063200525925467, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2519931964221283, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15214504712065627, "3000–6000 Hz_10ms_envelope_correlation": 0.9798614810694827, "3000–6000 Hz_10ms_modulation_mae_db": 0.5336192245122922, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8147321552721335, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09146735050749616, "3000–6000 Hz_40ms_envelope_correlation": 0.9929402648324892, "3000–6000 Hz_40ms_modulation_mae_db": 0.8788939573094402, "mr_stft_db": 5.471937252170295, "band_mae_db": 0.2350618569174585, "rms_gain_error_db": -0.2039564967916298, "output_peak": 0.39241413588543844}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7685804984279044, "全帯域_10ms_rms_envelope_nrmse": 0.07109289009916518, "全帯域_10ms_envelope_correlation": 0.995269773821583, "全帯域_10ms_modulation_mae_db": 0.5992018520669705, "全帯域_40ms_log_rms_mae_db": 0.4119729171498961, "全帯域_40ms_rms_envelope_nrmse": 0.039236417216415825, "全帯域_40ms_envelope_correlation": 0.9985435409932978, "全帯域_40ms_modulation_mae_db": 0.7681484538619259, "300–1000 Hz_10ms_log_rms_mae_db": 1.186087396253821, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08622142252561008, "300–1000 Hz_10ms_envelope_correlation": 0.9936653028601437, "300–1000 Hz_10ms_modulation_mae_db": 0.7712999346927276, "300–1000 Hz_40ms_log_rms_mae_db": 0.6262914410401516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04740843181100921, "300–1000 Hz_40ms_envelope_correlation": 0.9980476777705913, "300–1000 Hz_40ms_modulation_mae_db": 0.9987819631156695, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1560114061583202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10039637340060087, "1000–3000 Hz_10ms_envelope_correlation": 0.9926534211504641, "1000–3000 Hz_10ms_modulation_mae_db": 0.48946838613216537, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6416234431347912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05755707063684321, "1000–3000 Hz_40ms_envelope_correlation": 0.9976690275918656, "1000–3000 Hz_40ms_modulation_mae_db": 0.49011450245434673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2491025935312126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14006368437160394, "3000–6000 Hz_10ms_envelope_correlation": 0.9829411962731742, "3000–6000 Hz_10ms_modulation_mae_db": 0.5023867628571769, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8470327350171795, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08385152602092567, "3000–6000 Hz_40ms_envelope_correlation": 0.9939792022705464, "3000–6000 Hz_40ms_modulation_mae_db": 0.8207776232219254, "mr_stft_db": 5.468098617602912, "band_mae_db": 0.19511127877952472, "rms_gain_error_db": -0.09397588455898186, "output_peak": 0.37865315964742097}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7306514529763344, "全帯域_10ms_rms_envelope_nrmse": 0.06888555378297614, "全帯域_10ms_envelope_correlation": 0.9958542033383677, "全帯域_10ms_modulation_mae_db": 0.6780934627606907, "全帯域_40ms_log_rms_mae_db": 0.40143404626021706, "全帯域_40ms_rms_envelope_nrmse": 0.04143026875414291, "全帯域_40ms_envelope_correlation": 0.9987073709771079, "全帯域_40ms_modulation_mae_db": 0.7189477796830688, "300–1000 Hz_10ms_log_rms_mae_db": 1.118565038028025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08376567453060198, "300–1000 Hz_10ms_envelope_correlation": 0.9944053964333314, "300–1000 Hz_10ms_modulation_mae_db": 0.8189761298566468, "300–1000 Hz_40ms_log_rms_mae_db": 0.6172823008189832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051342029897031634, "300–1000 Hz_40ms_envelope_correlation": 0.9981455412727948, "300–1000 Hz_40ms_modulation_mae_db": 0.968924348763483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1441438125367371, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10228610923713362, "1000–3000 Hz_10ms_envelope_correlation": 0.9923594893763048, "1000–3000 Hz_10ms_modulation_mae_db": 0.43053031788248264, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6473414668911767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057393131745849264, "1000–3000 Hz_40ms_envelope_correlation": 0.9976811471016939, "1000–3000 Hz_40ms_modulation_mae_db": 0.4993634755679142, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3021822107459045, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15205996378908493, "3000–6000 Hz_10ms_envelope_correlation": 0.9801796322775932, "3000–6000 Hz_10ms_modulation_mae_db": 0.7642566492582866, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8857782669101748, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0912617623460673, "3000–6000 Hz_40ms_envelope_correlation": 0.9933831062289502, "3000–6000 Hz_40ms_modulation_mae_db": 1.247927223487303, "mr_stft_db": 5.476958027978473, "band_mae_db": 0.22245530904488023, "rms_gain_error_db": -0.19144738333460934, "output_peak": 0.38970717244283626}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7443543572608466, "全帯域_10ms_rms_envelope_nrmse": 0.07413667517997828, "全帯域_10ms_envelope_correlation": 0.9947953697581524, "全帯域_10ms_modulation_mae_db": 0.20099120012104418, "全帯域_40ms_log_rms_mae_db": 0.41300622452717145, "全帯域_40ms_rms_envelope_nrmse": 0.04436888390684644, "全帯域_40ms_envelope_correlation": 0.9980162957290282, "全帯域_40ms_modulation_mae_db": 0.08411669835191588, "300–1000 Hz_10ms_log_rms_mae_db": 1.1776213392364492, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08850655173730124, "300–1000 Hz_10ms_envelope_correlation": 0.9932760055291624, "300–1000 Hz_10ms_modulation_mae_db": 0.25047873851854435, "300–1000 Hz_40ms_log_rms_mae_db": 0.6348234708158849, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05320431713450785, "300–1000 Hz_40ms_envelope_correlation": 0.9974059463005949, "300–1000 Hz_40ms_modulation_mae_db": 0.10639746878282026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1227007431125895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09938385618541494, "1000–3000 Hz_10ms_envelope_correlation": 0.9926988679570151, "1000–3000 Hz_10ms_modulation_mae_db": 0.3559975449385056, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6363804825862327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05744617483650394, "1000–3000 Hz_40ms_envelope_correlation": 0.9975341798046964, "1000–3000 Hz_40ms_modulation_mae_db": 0.29477330079778824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2493800253877887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13679732647145734, "3000–6000 Hz_10ms_envelope_correlation": 0.9837562487717069, "3000–6000 Hz_10ms_modulation_mae_db": 0.6156789175674953, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8164565309664955, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07692643496623458, "3000–6000 Hz_40ms_envelope_correlation": 0.9950977384262121, "3000–6000 Hz_40ms_modulation_mae_db": 1.0420978411875903, "mr_stft_db": 5.4622134012489, "band_mae_db": 0.14779259526502794, "rms_gain_error_db": -0.02343781209689089, "output_peak": 0.3930512821880918}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7763148790301871, "全帯域_10ms_rms_envelope_nrmse": 0.07115629136220673, "全帯域_10ms_envelope_correlation": 0.9952832198888042, "全帯域_10ms_modulation_mae_db": 0.4975633232290422, "全帯域_40ms_log_rms_mae_db": 0.42382018682957795, "全帯域_40ms_rms_envelope_nrmse": 0.04085815106468494, "全帯域_40ms_envelope_correlation": 0.9984437982793838, "全帯域_40ms_modulation_mae_db": 0.6713815450573045, "300–1000 Hz_10ms_log_rms_mae_db": 1.1811070751718875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08719407705487424, "300–1000 Hz_10ms_envelope_correlation": 0.9935472838299444, "300–1000 Hz_10ms_modulation_mae_db": 0.681154771282077, "300–1000 Hz_40ms_log_rms_mae_db": 0.6371731452796696, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05050083845077197, "300–1000 Hz_40ms_envelope_correlation": 0.9978064549758489, "300–1000 Hz_40ms_modulation_mae_db": 0.9190984489507787, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0637727374614143, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0930996951319965, "1000–3000 Hz_10ms_envelope_correlation": 0.9937642807731059, "1000–3000 Hz_10ms_modulation_mae_db": 0.36244190854926356, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5856000644216527, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053957372919676345, "1000–3000 Hz_40ms_envelope_correlation": 0.998028944347033, "1000–3000 Hz_40ms_modulation_mae_db": 0.4048897233479261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2753995083857557, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1515844649164949, "3000–6000 Hz_10ms_envelope_correlation": 0.979707179187538, "3000–6000 Hz_10ms_modulation_mae_db": 0.2930080852157881, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8566921476897228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08963470194795012, "3000–6000 Hz_40ms_envelope_correlation": 0.9927507206015513, "3000–6000 Hz_40ms_modulation_mae_db": 0.5102695425419833, "mr_stft_db": 5.468880402533803, "band_mae_db": 0.2169122742790628, "rms_gain_error_db": -0.1041818377260931, "output_peak": 0.37876236117141154}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7928418957289208, "全帯域_10ms_rms_envelope_nrmse": 0.07170497996903037, "全帯域_10ms_envelope_correlation": 0.9951294598754158, "全帯域_10ms_modulation_mae_db": 0.3607868053117535, "全帯域_40ms_log_rms_mae_db": 0.4282044604587602, "全帯域_40ms_rms_envelope_nrmse": 0.03902317047212196, "全帯域_40ms_envelope_correlation": 0.9984890862152898, "全帯域_40ms_modulation_mae_db": 0.5876805069283108, "300–1000 Hz_10ms_log_rms_mae_db": 1.168290619848296, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08511869571343776, "300–1000 Hz_10ms_envelope_correlation": 0.9937717069333356, "300–1000 Hz_10ms_modulation_mae_db": 0.3993101052973902, "300–1000 Hz_40ms_log_rms_mae_db": 0.6284189915816969, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048497565186318495, "300–1000 Hz_40ms_envelope_correlation": 0.9978669339322308, "300–1000 Hz_40ms_modulation_mae_db": 0.6082585877816077, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1205267533016852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10897808307922041, "1000–3000 Hz_10ms_envelope_correlation": 0.9912399212090719, "1000–3000 Hz_10ms_modulation_mae_db": 0.3845300820210175, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6465682283836282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059281316921493604, "1000–3000 Hz_40ms_envelope_correlation": 0.9974165371354028, "1000–3000 Hz_40ms_modulation_mae_db": 0.4774699263176765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2545798841372304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1532626611190778, "3000–6000 Hz_10ms_envelope_correlation": 0.979422273572831, "3000–6000 Hz_10ms_modulation_mae_db": 0.3534080492883464, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8223481596716471, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09365995792585531, "3000–6000 Hz_40ms_envelope_correlation": 0.9923755567959919, "3000–6000 Hz_40ms_modulation_mae_db": 0.054193561832416214, "mr_stft_db": 5.471008420795182, "band_mae_db": 0.16170990238918845, "rms_gain_error_db": -0.058722363660888226, "output_peak": 0.39576830616178377}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7601298683714521, "全帯域_10ms_rms_envelope_nrmse": 0.06760232399940451, "全帯域_10ms_envelope_correlation": 0.9957120495117086, "全帯域_10ms_modulation_mae_db": 0.31458529201064606, "全帯域_40ms_log_rms_mae_db": 0.41581956208016146, "全帯域_40ms_rms_envelope_nrmse": 0.03642745363927663, "全帯域_40ms_envelope_correlation": 0.9987396338275808, "全帯域_40ms_modulation_mae_db": 0.2035188394340571, "300–1000 Hz_10ms_log_rms_mae_db": 1.1377316054186166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08319648797794618, "300–1000 Hz_10ms_envelope_correlation": 0.994065317015764, "300–1000 Hz_10ms_modulation_mae_db": 0.4127531366583588, "300–1000 Hz_40ms_log_rms_mae_db": 0.5843513378269257, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044650331152217344, "300–1000 Hz_40ms_envelope_correlation": 0.9982204277567784, "300–1000 Hz_40ms_modulation_mae_db": 0.2817009062089979, "1000–3000 Hz_10ms_log_rms_mae_db": 1.103050030277761, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10028459761487396, "1000–3000 Hz_10ms_envelope_correlation": 0.9930266883043019, "1000–3000 Hz_10ms_modulation_mae_db": 0.6907546450397756, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6005208761207964, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062165315276519646, "1000–3000 Hz_40ms_envelope_correlation": 0.997670219199788, "1000–3000 Hz_40ms_modulation_mae_db": 0.5408336614646426, "3000–6000 Hz_10ms_log_rms_mae_db": 1.266348747689472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14968665618777816, "3000–6000 Hz_10ms_envelope_correlation": 0.9802538596395668, "3000–6000 Hz_10ms_modulation_mae_db": 0.21179238562125202, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8370353490080673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09023108972522383, "3000–6000 Hz_40ms_envelope_correlation": 0.9923085549099809, "3000–6000 Hz_40ms_modulation_mae_db": 0.24866409376703055, "mr_stft_db": 5.490915229198857, "band_mae_db": 0.19038425986038632, "rms_gain_error_db": -0.08703510110327105, "output_peak": 0.3922846874764865}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7595598408481132, "全帯域_10ms_rms_envelope_nrmse": 0.08290700366359217, "全帯域_10ms_envelope_correlation": 0.9936387080183795, "全帯域_10ms_modulation_mae_db": 0.13679358749181056, "全帯域_40ms_log_rms_mae_db": 0.4169740085378523, "全帯域_40ms_rms_envelope_nrmse": 0.04997524190652741, "全帯域_40ms_envelope_correlation": 0.9977139059821488, "全帯域_40ms_modulation_mae_db": 0.2566775434173197, "300–1000 Hz_10ms_log_rms_mae_db": 1.1486845089168354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09887827764687987, "300–1000 Hz_10ms_envelope_correlation": 0.9917756361483577, "300–1000 Hz_10ms_modulation_mae_db": 0.17540660726599855, "300–1000 Hz_40ms_log_rms_mae_db": 0.6014780581263978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060170194301659166, "300–1000 Hz_40ms_envelope_correlation": 0.9969681603027725, "300–1000 Hz_40ms_modulation_mae_db": 0.36293956326597004, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1485756362350499, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10960052819180506, "1000–3000 Hz_10ms_envelope_correlation": 0.9912068749658918, "1000–3000 Hz_10ms_modulation_mae_db": 0.4465922342904695, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6511905986916451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06626569736989141, "1000–3000 Hz_40ms_envelope_correlation": 0.9968433459780425, "1000–3000 Hz_40ms_modulation_mae_db": 0.373590919702103, "3000–6000 Hz_10ms_log_rms_mae_db": 1.292029385030312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14431860615882586, "3000–6000 Hz_10ms_envelope_correlation": 0.9818561710064181, "3000–6000 Hz_10ms_modulation_mae_db": 0.61129617249791, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8720058479765046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08643971137626606, "3000–6000 Hz_40ms_envelope_correlation": 0.9935363356193595, "3000–6000 Hz_40ms_modulation_mae_db": 0.9560447923196941, "mr_stft_db": 5.457910073816387, "band_mae_db": 0.18904135158615817, "rms_gain_error_db": -0.14595272029613046, "output_peak": 0.383218316355083}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.337391435584736, "spectral_convergence": 0.20758887756337663, "spectral_convergence_status": "ok", "amplitude_floor": 9.509362929412646e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.26060144928524, "spectral_convergence": 0.21050129126855427, "spectral_convergence_status": "ok", "amplitude_floor": 8.503880220761315e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.817818871640907, "spectral_convergence": 0.22165421673300803, "spectral_convergence_status": "ok", "amplitude_floor": 5.089992363719204e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.8030381557821282e-05, "output_power": 2.9690441765582545e-05, "delta_db": 0.24987649988543897, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003997973981863815, "output_power": 0.0003981919174933389, "delta_db": -0.017475230128125466, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008734907128440196, "output_power": 0.0008172064428788819, "delta_db": -0.2892650985979533, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001464675075572333, "output_power": 0.0001378618473985125, "delta_db": -0.2629719731774042, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.350074484130352e-05, "output_power": 5.935381475566605e-05, "delta_db": -0.2933018251949622, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.7271042012588545e-05, "output_power": 1.612762769267009e-05, "delta_db": -0.297480514520867, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2350618569174585, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2039564967916298, "pitch": null, "output_rms_dbfs": -28.361137878401177, "output_peak": 0.39241413588543844, "reference_rms_dbfs": -28.157181381609547, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.471937252170295}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012908667094909063, "log_rms_mae_db": 0.7203330779907231, "rms_envelope_nrmse": 0.0745379954838452, "envelope_correlation": 0.9951212944172624, "rms_gain_error_db": -0.20175840763106195, "reference_mean_power": 0.001527852496322364, "output_mean_power": 0.0014584971584929898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.179575522794606e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002748180655831064, "output_power": 0.0002567017506944514, "reference_active": true, "delta_db": -0.2961644778308211}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.543131824921543e-05, "output_power": 4.643419116512978e-05, "reference_active": true, "delta_db": -0.7691732196659786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2340545383743233e-05, "output_power": 8.681494434562827e-06, "reference_active": true, "delta_db": -1.5273986236209631}], "active_band_count": 3, "active_band_mae_db": 0.8642454403725877}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001146348873577052, "log_rms_mae_db": 0.40240064991269076, "rms_envelope_nrmse": 0.04417304168524292, "envelope_correlation": 0.9985185150412497, "rms_gain_error_db": -0.20031481456238784, "reference_mean_power": 0.0015234711554082644, "output_mean_power": 0.0014547981981336403}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.718780555754886e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002667517105928089, "output_power": 0.0002471478620219386, "reference_active": true, "delta_db": -0.3315035519222521}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.819324290787445e-05, "output_power": 2.3368948979963053e-05, "reference_active": true, "delta_db": -0.8150585275298551}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.654413380387e-07, "output_power": 5.457898988623707e-07, "reference_active": true, "delta_db": -1.4688641905252586}], "active_band_count": 3, "active_band_mae_db": 0.871808756659122}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001269978635547725, "log_rms_mae_db": 1.0835474459555865, "rms_envelope_nrmse": 0.0895021535597851, "envelope_correlation": 0.9935251849774269, "rms_gain_error_db": -0.22528652383820075, "reference_mean_power": 0.0011723709196259178, "output_mean_power": 0.001113105644261021}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.280903308109198e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025355561575728737, "output_power": 0.00023465855398146826, "reference_active": true, "delta_db": -0.33636843633242436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.608107242392719e-05, "output_power": 4.7660835797267436e-05, "reference_active": true, "delta_db": -0.7065465592091258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1840508111446508e-05, "output_power": 7.759230509842269e-06, "reference_active": true, "delta_db": -1.8355168565260729}], "active_band_count": 3, "active_band_mae_db": 0.9594772840225411}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.809451727166842e-05, "log_rms_mae_db": 0.5910340398664904, "rms_envelope_nrmse": 0.05333564863076636, "envelope_correlation": 0.9979366998736852, "rms_gain_error_db": -0.22343449095067192, "reference_mean_power": 0.0011689931508753605, "output_mean_power": 0.001110372040401897}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.891072559770159e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025025709822561247, "output_power": 0.00022942200321564375, "reference_active": true, "delta_db": -0.3775133702041092}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.836247458716607e-05, "output_power": 2.3626091995107943e-05, "reference_active": true, "delta_db": -0.7935222908093488}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.847292373572026e-07, "output_power": 5.353978320978137e-07, "reference_active": true, "delta_db": -1.6604322547341575}], "active_band_count": 3, "active_band_mae_db": 0.9438226385825385}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.9423530315000806e-05, "log_rms_mae_db": 1.109176831937842, "rms_envelope_nrmse": 0.10446529491396957, "envelope_correlation": 0.9924089658318996, "rms_gain_error_db": -0.27216764066519433, "reference_mean_power": 0.00015673633956673046, "output_mean_power": 0.00014721529580229186}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.197939100277272e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3248792094830553e-05, "output_power": 3.0435998675414154e-05, "reference_active": true, "delta_db": -0.3838831578559345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.648467501564817e-06, "output_power": 2.631383189089705e-06, "reference_active": true, "delta_db": -1.4192638634199253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1792733873632069e-06, "output_power": 1.1938440380416984e-06, "reference_active": true, "delta_db": 0.053330969961535996}], "active_band_count": 3, "active_band_mae_db": 0.618825997079132}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.3506304987563625e-05, "log_rms_mae_db": 0.595623795013852, "rms_envelope_nrmse": 0.05863219219836173, "envelope_correlation": 0.9980697017220297, "rms_gain_error_db": -0.2744553334142412, "reference_mean_power": 0.0001561782704543471, "output_mean_power": 0.00014661387615991215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.897719153255603e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.156452210778437e-05, "output_power": 2.8838776993868755e-05, "reference_active": true, "delta_db": -0.3922237971336078}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6348481818884734e-06, "output_power": 1.1787437592241127e-06, "reference_active": true, "delta_db": -1.4205802225036912}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.181225496757285e-08, "output_power": 7.171053297964033e-08, "reference_active": true, "delta_db": -0.006156138140341277}], "active_band_count": 3, "active_band_mae_db": 0.6063200525925467}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.766512572532955e-05, "log_rms_mae_db": 1.2519931964221283, "rms_envelope_nrmse": 0.15214504712065627, "envelope_correlation": 0.9798614810694827, "rms_gain_error_db": -0.2991656227349521, "reference_mean_power": 2.7841403311343547e-05, "output_mean_power": 2.5988101799126355e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1058323154967027e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.079425705940839e-06, "output_power": 4.714401063424249e-06, "reference_active": true, "delta_db": -0.32388086461649784}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8241793691525221e-06, "output_power": 1.5953447873230138e-06, "reference_active": true, "delta_db": -0.5821298205206297}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.319091479762654e-07, "output_power": 4.5326560063592206e-07, "reference_active": true, "delta_db": -0.694846988399749}], "active_band_count": 3, "active_band_mae_db": 0.5336192245122922}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5244936872690136e-05, "log_rms_mae_db": 0.8147321552721335, "rms_envelope_nrmse": 0.09146735050749616, "envelope_correlation": 0.9929402648324892, "rms_gain_error_db": -0.29926561384134664, "reference_mean_power": 2.7868742577273385e-05, "output_mean_power": 2.6013022263415744e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.800642728490222e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.984173950209188e-06, "output_power": 4.501345428204514e-06, "reference_active": true, "delta_db": -0.44250849524991137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0071402238290061e-06, "output_power": 8.4240425678288e-07, "reference_active": true, "delta_db": -0.7756938872100839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932914966264332e-08, "output_power": 2.8370475033919932e-08, "reference_active": true, "delta_db": -1.4184794894683252}], "active_band_count": 3, "active_band_mae_db": 0.8788939573094402}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8271151836952375, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e77b84ea053f94bff0b38c44955ff9304c27900bc24f791b60ecaee40b3329ad", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9924198136960068, "standard_deviation": 0.0008598551022787127, "minimum": 0.9912068749658918, "maximum": 0.9937642807731059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1209947438776748, "standard_deviation": 0.02990918372091408, "minimum": 1.0637727374614143, "maximum": 1.1560114061583202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4723926394916015, "standard_deviation": 0.12258594650274045, "minimum": 0.3559975449385056, "maximum": 0.6907546450397756, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10231181721937686, "standard_deviation": 0.00539069606891804, "minimum": 0.0930996951319965, "maximum": 0.10960052819180506, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997614137860069, "standard_deviation": 0.00038355676691836655, "minimum": 0.9968433459780425, "maximum": 0.9980697017220297, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6256061194054718, "standard_deviation": 0.026906547601508915, "minimum": 0.5856000644216527, "maximum": 0.6511905986916451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.460919445280618, "standard_deviation": 0.0989241948245019, "minimum": 0.29477330079778824, "maximum": 0.6063200525925467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05908728398814239, "standard_deviation": 0.003693035610566314, "minimum": 0.053957372919676345, "maximum": 0.06626569736989141, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9809972552247889, "standard_deviation": 0.0016381469979883387, "minimum": 0.979422273572831, "maximum": 0.9837562487717069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2676269439162255, "standard_deviation": 0.020502033225232842, "minimum": 1.2491025935312126, "maximum": 1.3021822107459045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.48568078085231847, "standard_deviation": 0.18612366742979178, "minimum": 0.21179238562125202, "maximum": 0.7642566492582866, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14748980126687242, "standard_deviation": 0.006290681998740414, "minimum": 0.13679732647145734, "maximum": 0.1532626611190778, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932964349606352, "standard_deviation": 0.0009277095080959692, "minimum": 0.9923085549099809, "maximum": 0.9950977384262121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8440101490639906, "standard_deviation": 0.026274910226822577, "minimum": 0.8147321552721335, "maximum": 0.8857782669101748, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7198585794584228, "standard_deviation": 0.4110990825973939, "minimum": 0.054193561832416214, "maximum": 1.247927223487303, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08793406685200239, "standard_deviation": 0.005406083223667619, "minimum": 0.07692643496623458, "maximum": 0.09365995792585531, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935039792159333, "standard_deviation": 0.0007807343895515269, "minimum": 0.9917756361483577, "maximum": 0.9944053964333314, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1502043786036897, "standard_deviation": 0.03570402954543704, "minimum": 1.0835474459555865, "maximum": 1.186087396253821, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5586070884492855, "standard_deviation": 0.2871484747198665, "minimum": 0.17540660726599855, "maximum": 0.9594772840225411, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08779791759330456, "standard_deviation": 0.004980552036156452, "minimum": 0.08319648797794618, "maximum": 0.09887827764687987, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997799730273162, "standard_deviation": 0.00041883277250337003, "minimum": 0.9969681603027725, "maximum": 0.9982204277567784, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.615106598169525, "standard_deviation": 0.020332426544941495, "minimum": 0.5843513378269257, "maximum": 0.6371731452796696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6487404906814832, "standard_deviation": 0.35804800148174076, "minimum": 0.10639746878282026, "maximum": 0.9987819631156695, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.051138669570535256, "standard_deviation": 0.004691921344484687, "minimum": 0.044650331152217344, "maximum": 0.060170194301659166, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1948086035152109, "standard_deviation": 0.02983811960099357, "minimum": 0.14779259526502794, "maximum": 0.2350618569174585, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.470990178168101, "standard_deviation": 0.009960302566882028, "minimum": 5.457910073816387, "maximum": 5.490915229198857, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.38798242766606905, "standard_deviation": 0.006784215255087667, "minimum": 0.37865315964742097, "maximum": 0.39576830616178377, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11358869994606184, "standard_deviation": 0.06278248815443017, "minimum": -0.2039564967916298, "maximum": -0.02343781209689089, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951005098287091, "standard_deviation": 0.0006798437903976847, "minimum": 0.9936387080183795, "maximum": 0.9958542033383677, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7565957338293102, "standard_deviation": 0.023920931591711773, "minimum": 0.7203330779907231, "maximum": 0.7928418957289208, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4565326204205682, "standard_deviation": 0.24905758096808092, "minimum": 0.13679358749181056, "maximum": 0.8642454403725877, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07275296419252482, "standard_deviation": 0.004722211181788691, "minimum": 0.06760232399940451, "maximum": 0.08290700366359217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983965183806358, "standard_deviation": 0.0003529851321641207, "minimum": 0.9977139059821488, "maximum": 0.9987396338275808, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41420400696954096, "standard_deviation": 0.009316422664919766, "minimum": 0.40143404626021706, "maximum": 0.4282044604587602, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5202850154241281, "standard_deviation": 0.2956631632154926, "minimum": 0.08411669835191588, "maximum": 0.871808756659122, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041936578580657385, "standard_deviation": 0.0041933044564582056, "minimum": 0.03642745363927663, "maximum": 0.04997524190652741, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0010", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.631831645965576, "predicted_mos": 2.631831645965576, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "adb01812722f6933ec8207bbcdf034c8523ec0d216051b3c86db2be1b07c6be1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.259578499943018, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7322714255151785, "全帯域_10ms_rms_envelope_nrmse": 0.07591075492518522, "全帯域_10ms_envelope_correlation": 0.9949579551160301, "全帯域_10ms_modulation_mae_db": 0.8383632069603046, "全帯域_40ms_log_rms_mae_db": 0.40830727411986173, "全帯域_40ms_rms_envelope_nrmse": 0.04603415845497645, "全帯域_40ms_envelope_correlation": 0.9983890839097904, "全帯域_40ms_modulation_mae_db": 0.8311821263683696, "300–1000 Hz_10ms_log_rms_mae_db": 1.0979461948251223, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09043056008411088, "300–1000 Hz_10ms_envelope_correlation": 0.9933834549811161, "300–1000 Hz_10ms_modulation_mae_db": 0.9231157091235098, "300–1000 Hz_40ms_log_rms_mae_db": 0.6044951228171523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054986886759008084, "300–1000 Hz_40ms_envelope_correlation": 0.9977758254880467, "300–1000 Hz_40ms_modulation_mae_db": 0.8870942993386121, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1129944090694446, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10465091073342613, "1000–3000 Hz_10ms_envelope_correlation": 0.9925112569717525, "1000–3000 Hz_10ms_modulation_mae_db": 0.6052780515624249, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5860503645143565, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059848488482039844, "1000–3000 Hz_40ms_envelope_correlation": 0.9981001075985948, "1000–3000 Hz_40ms_modulation_mae_db": 0.6162719929541389, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2408438258291445, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15307054266965736, "3000–6000 Hz_10ms_envelope_correlation": 0.9800438534216326, "3000–6000 Hz_10ms_modulation_mae_db": 0.5615733466898409, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8104471809519731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09259758038137042, "3000–6000 Hz_40ms_envelope_correlation": 0.9932856907304144, "3000–6000 Hz_40ms_modulation_mae_db": 0.897910883104084, "mr_stft_db": 5.438929669199365, "band_mae_db": 0.2564254077750615, "rms_gain_error_db": -0.212017807553011, "output_peak": 0.39394706845383676}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7838475019599943, "全帯域_10ms_rms_envelope_nrmse": 0.07488791278429238, "全帯域_10ms_envelope_correlation": 0.994762371202233, "全帯域_10ms_modulation_mae_db": 0.5440573414263055, "全帯域_40ms_log_rms_mae_db": 0.42177605104608135, "全帯域_40ms_rms_envelope_nrmse": 0.042086436969229916, "全帯域_40ms_envelope_correlation": 0.9983379337140874, "全帯域_40ms_modulation_mae_db": 0.6738390192599341, "300–1000 Hz_10ms_log_rms_mae_db": 1.1870253910262512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08963447663891988, "300–1000 Hz_10ms_envelope_correlation": 0.993152333072442, "300–1000 Hz_10ms_modulation_mae_db": 0.7169823426408707, "300–1000 Hz_40ms_log_rms_mae_db": 0.6224958793297173, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05061679352121119, "300–1000 Hz_40ms_envelope_correlation": 0.9977667802722736, "300–1000 Hz_40ms_modulation_mae_db": 0.8677473345357957, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1621015205940775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09820435742998944, "1000–3000 Hz_10ms_envelope_correlation": 0.9931238442098484, "1000–3000 Hz_10ms_modulation_mae_db": 0.4230970073091118, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6377097167691504, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05628492504305402, "1000–3000 Hz_40ms_envelope_correlation": 0.9979499784205401, "1000–3000 Hz_40ms_modulation_mae_db": 0.42021349388465085, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2694018963033173, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14336967639836865, "3000–6000 Hz_10ms_envelope_correlation": 0.9823223736315635, "3000–6000 Hz_10ms_modulation_mae_db": 0.5249880649467203, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8805328960160973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08857481892384665, "3000–6000 Hz_40ms_envelope_correlation": 0.9934337579595122, "3000–6000 Hz_40ms_modulation_mae_db": 0.8918429917827843, "mr_stft_db": 5.439556276944939, "band_mae_db": 0.2199019701852379, "rms_gain_error_db": -0.1060044685949091, "output_peak": 0.375152422030691}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7360487120712034, "全帯域_10ms_rms_envelope_nrmse": 0.07214277602945972, "全帯域_10ms_envelope_correlation": 0.99537934423983, "全帯域_10ms_modulation_mae_db": 0.6634120998056742, "全帯域_40ms_log_rms_mae_db": 0.4062144873556608, "全帯域_40ms_rms_envelope_nrmse": 0.044108684803556245, "全帯域_40ms_envelope_correlation": 0.998433986867111, "全帯域_40ms_modulation_mae_db": 0.6665480748090117, "300–1000 Hz_10ms_log_rms_mae_db": 1.1521371868325034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0875080486948371, "300–1000 Hz_10ms_envelope_correlation": 0.9937680062263299, "300–1000 Hz_10ms_modulation_mae_db": 0.777405316750588, "300–1000 Hz_40ms_log_rms_mae_db": 0.6461429860272404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053954237720719174, "300–1000 Hz_40ms_envelope_correlation": 0.9977971569613187, "300–1000 Hz_40ms_modulation_mae_db": 0.9019296638807353, "1000–3000 Hz_10ms_log_rms_mae_db": 1.15481658113075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09838279012694524, "1000–3000 Hz_10ms_envelope_correlation": 0.9930374622163518, "1000–3000 Hz_10ms_modulation_mae_db": 0.38973972469286117, "1000–3000 Hz_40ms_log_rms_mae_db": 0.640255807274107, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05501003806384924, "1000–3000 Hz_40ms_envelope_correlation": 0.9979892702887763, "1000–3000 Hz_40ms_modulation_mae_db": 0.4601259520839383, "3000–6000 Hz_10ms_log_rms_mae_db": 1.296412818716839, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15257561921605095, "3000–6000 Hz_10ms_envelope_correlation": 0.9804751034647573, "3000–6000 Hz_10ms_modulation_mae_db": 0.7758957457576893, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8894527007487341, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09067210938552757, "3000–6000 Hz_40ms_envelope_correlation": 0.9940312874240028, "3000–6000 Hz_40ms_modulation_mae_db": 1.2761709726126833, "mr_stft_db": 5.444205183285262, "band_mae_db": 0.2440024540331056, "rms_gain_error_db": -0.18508987217636375, "output_peak": 0.38756731686753954}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7645871146662286, "全帯域_10ms_rms_envelope_nrmse": 0.07727024039538512, "全帯域_10ms_envelope_correlation": 0.9943500886366371, "全帯域_10ms_modulation_mae_db": 0.2021007647585047, "全帯域_40ms_log_rms_mae_db": 0.4221561819861884, "全帯域_40ms_rms_envelope_nrmse": 0.04818010132842897, "全帯域_40ms_envelope_correlation": 0.9976571130232963, "全帯域_40ms_modulation_mae_db": 0.0979514474582512, "300–1000 Hz_10ms_log_rms_mae_db": 1.1930795273259434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09240257489229203, "300–1000 Hz_10ms_envelope_correlation": 0.9926817070638555, "300–1000 Hz_10ms_modulation_mae_db": 0.24957747271028663, "300–1000 Hz_40ms_log_rms_mae_db": 0.6344572421133596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05747157954919519, "300–1000 Hz_40ms_envelope_correlation": 0.9969720210012191, "300–1000 Hz_40ms_modulation_mae_db": 0.08646400176103719, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1233667007756742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0992075737159675, "1000–3000 Hz_10ms_envelope_correlation": 0.9927594345535931, "1000–3000 Hz_10ms_modulation_mae_db": 0.33880812215770933, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6165894839756726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05662845729312078, "1000–3000 Hz_40ms_envelope_correlation": 0.9976373817637717, "1000–3000 Hz_40ms_modulation_mae_db": 0.2838904164700204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2253599112681117, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13354366232862422, "3000–6000 Hz_10ms_envelope_correlation": 0.9846262549543514, "3000–6000 Hz_10ms_modulation_mae_db": 0.532980123383341, "3000–6000 Hz_40ms_log_rms_mae_db": 0.802317709342404, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07263800285669171, "3000–6000 Hz_40ms_envelope_correlation": 0.9957686438942744, "3000–6000 Hz_40ms_modulation_mae_db": 0.9824066234702933, "mr_stft_db": 5.429019212300279, "band_mae_db": 0.16626731209732332, "rms_gain_error_db": -0.012024005633546123, "output_peak": 0.39678067371939063}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7859340738136937, "全帯域_10ms_rms_envelope_nrmse": 0.07343626768989417, "全帯域_10ms_envelope_correlation": 0.9949851355735319, "全帯域_10ms_modulation_mae_db": 0.49847288892846725, "全帯域_40ms_log_rms_mae_db": 0.42641739147559493, "全帯域_40ms_rms_envelope_nrmse": 0.04121609024463846, "全帯域_40ms_envelope_correlation": 0.99843984489891, "全帯域_40ms_modulation_mae_db": 0.7179427054037012, "300–1000 Hz_10ms_log_rms_mae_db": 1.2064603885989063, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08997230387705799, "300–1000 Hz_10ms_envelope_correlation": 0.9931191529640875, "300–1000 Hz_10ms_modulation_mae_db": 0.65802782927874, "300–1000 Hz_40ms_log_rms_mae_db": 0.6596969508573044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05036420097978648, "300–1000 Hz_40ms_envelope_correlation": 0.9978241582379817, "300–1000 Hz_40ms_modulation_mae_db": 0.9601443111831589, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0549493688116496, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09033448685506196, "1000–3000 Hz_10ms_envelope_correlation": 0.9942783694182853, "1000–3000 Hz_10ms_modulation_mae_db": 0.3410478508893149, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5899799130056395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05252137763561638, "1000–3000 Hz_40ms_envelope_correlation": 0.998299021229021, "1000–3000 Hz_40ms_modulation_mae_db": 0.3775773079671261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2817467294603382, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1516117925149751, "3000–6000 Hz_10ms_envelope_correlation": 0.9799041289905259, "3000–6000 Hz_10ms_modulation_mae_db": 0.2945406143225467, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8600925142340353, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09049349563068503, "3000–6000 Hz_40ms_envelope_correlation": 0.9928611902513518, "3000–6000 Hz_40ms_modulation_mae_db": 0.43166261023286595, "mr_stft_db": 5.433005405792396, "band_mae_db": 0.23979941759352227, "rms_gain_error_db": -0.11337878117771175, "output_peak": 0.37777659641856387}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7814808380439027, "全帯域_10ms_rms_envelope_nrmse": 0.0722803138602669, "全帯域_10ms_envelope_correlation": 0.9950575359349131, "全帯域_10ms_modulation_mae_db": 0.34137825942513134, "全帯域_40ms_log_rms_mae_db": 0.43043051263355636, "全帯域_40ms_rms_envelope_nrmse": 0.03983526699912187, "全帯域_40ms_envelope_correlation": 0.9984342310671491, "全帯域_40ms_modulation_mae_db": 0.5612899511154593, "300–1000 Hz_10ms_log_rms_mae_db": 1.1643527788293557, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08571291749759463, "300–1000 Hz_10ms_envelope_correlation": 0.9936847212238535, "300–1000 Hz_10ms_modulation_mae_db": 0.39253996673725666, "300–1000 Hz_40ms_log_rms_mae_db": 0.6201602921288215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049210582088802544, "300–1000 Hz_40ms_envelope_correlation": 0.9978040782169965, "300–1000 Hz_40ms_modulation_mae_db": 0.6087574300569588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.129372931152426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1091445789708551, "1000–3000 Hz_10ms_envelope_correlation": 0.9913134473685428, "1000–3000 Hz_10ms_modulation_mae_db": 0.40081331708615425, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6514676979535058, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058708288685084986, "1000–3000 Hz_40ms_envelope_correlation": 0.9975919734568469, "1000–3000 Hz_40ms_modulation_mae_db": 0.5570036543636052, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2768399165281343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15487663791920692, "3000–6000 Hz_10ms_envelope_correlation": 0.9789959429399911, "3000–6000 Hz_10ms_modulation_mae_db": 0.3160835875711819, "3000–6000 Hz_40ms_log_rms_mae_db": 0.868009223392747, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0964687296468052, "3000–6000 Hz_40ms_envelope_correlation": 0.9917885308611897, "3000–6000 Hz_40ms_modulation_mae_db": 0.1763186570854979, "mr_stft_db": 5.43987257449373, "band_mae_db": 0.18010086983377724, "rms_gain_error_db": -0.06601364183628727, "output_peak": 0.39456275384981476}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7752672857071002, "全帯域_10ms_rms_envelope_nrmse": 0.06950243968915436, "全帯域_10ms_envelope_correlation": 0.9954704997010673, "全帯域_10ms_modulation_mae_db": 0.44048192808394876, "全帯域_40ms_log_rms_mae_db": 0.42545194301284384, "全帯域_40ms_rms_envelope_nrmse": 0.037465897243911576, "全帯域_40ms_envelope_correlation": 0.99867324192983, "全帯域_40ms_modulation_mae_db": 0.3626351296187367, "300–1000 Hz_10ms_log_rms_mae_db": 1.1556887761649697, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08488816276430833, "300–1000 Hz_10ms_envelope_correlation": 0.9938179496119046, "300–1000 Hz_10ms_modulation_mae_db": 0.5355245489522479, "300–1000 Hz_40ms_log_rms_mae_db": 0.5988574692685025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0457324954221239, "300–1000 Hz_40ms_envelope_correlation": 0.9981303196089546, "300–1000 Hz_40ms_modulation_mae_db": 0.43713830230750633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0986669964975997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09731531555522374, "1000–3000 Hz_10ms_envelope_correlation": 0.9935642614920824, "1000–3000 Hz_10ms_modulation_mae_db": 0.7360772183014882, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5940402654812565, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059670884391034844, "1000–3000 Hz_40ms_envelope_correlation": 0.9980092747142117, "1000–3000 Hz_40ms_modulation_mae_db": 0.6023991503276086, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2818817919626753, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15092398261731316, "3000–6000 Hz_10ms_envelope_correlation": 0.9800983577705941, "3000–6000 Hz_10ms_modulation_mae_db": 0.1542676589905905, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8568192119481998, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09154800015105276, "3000–6000 Hz_40ms_envelope_correlation": 0.9922511540677281, "3000–6000 Hz_40ms_modulation_mae_db": 0.30020885023478255, "mr_stft_db": 5.458263393815211, "band_mae_db": 0.20878809541067342, "rms_gain_error_db": -0.09211815374628564, "output_peak": 0.39219906391382187}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7729643294375768, "全帯域_10ms_rms_envelope_nrmse": 0.08336360548059545, "全帯域_10ms_envelope_correlation": 0.9935519443246249, "全帯域_10ms_modulation_mae_db": 0.14474565282389934, "全帯域_40ms_log_rms_mae_db": 0.42981746327682924, "全帯域_40ms_rms_envelope_nrmse": 0.04988986180783703, "全帯域_40ms_envelope_correlation": 0.9977050410294278, "全帯域_40ms_modulation_mae_db": 0.27652305811974065, "300–1000 Hz_10ms_log_rms_mae_db": 1.173849506258788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.099493405331882, "300–1000 Hz_10ms_envelope_correlation": 0.9916287729296416, "300–1000 Hz_10ms_modulation_mae_db": 0.15628778413550698, "300–1000 Hz_40ms_log_rms_mae_db": 0.6277698262041065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05974799710674001, "300–1000 Hz_40ms_envelope_correlation": 0.9969645013706737, "300–1000 Hz_40ms_modulation_mae_db": 0.3743067262068153, "1000–3000 Hz_10ms_log_rms_mae_db": 1.142398670030866, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10400301106090118, "1000–3000 Hz_10ms_envelope_correlation": 0.9922257254109499, "1000–3000 Hz_10ms_modulation_mae_db": 0.45453167088658386, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6437645316619168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06272905790425815, "1000–3000 Hz_40ms_envelope_correlation": 0.99732750880886, "1000–3000 Hz_40ms_modulation_mae_db": 0.3250745250484722, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3170759531796887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15035081712536857, "3000–6000 Hz_10ms_envelope_correlation": 0.9804215262869093, "3000–6000 Hz_10ms_modulation_mae_db": 0.523201288595701, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9306682363541283, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09205817819741734, "3000–6000 Hz_40ms_envelope_correlation": 0.9927097011635676, "3000–6000 Hz_40ms_modulation_mae_db": 0.9074680489885641, "mr_stft_db": 5.4247578775798715, "band_mae_db": 0.19658698541022554, "rms_gain_error_db": -0.140666152760911, "output_peak": 0.38367961758414804}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.307540596744883, "spectral_convergence": 0.20785684901664217, "spectral_convergence_status": "ok", "amplitude_floor": 9.509362929412646e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.230679918032438, "spectral_convergence": 0.21119226162093457, "spectral_convergence_status": "ok", "amplitude_floor": 8.503880220761315e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.778568492820773, "spectral_convergence": 0.2209091032840764, "spectral_convergence_status": "ok", "amplitude_floor": 5.089992363719204e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.8030381557821282e-05, "output_power": 2.9370717813095908e-05, "delta_db": 0.20285551157372209, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003997973981863815, "output_power": 0.00039980168956542356, "delta_db": 4.661641730925209e-05, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008734907128440196, "output_power": 0.0008149511243606595, "delta_db": -0.3012672855246649, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001464675075572333, "output_power": 0.00013738603677577072, "delta_db": -0.27798695753175473, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.350074484130352e-05, "output_power": 5.845502082390512e-05, "delta_db": -0.3595699988206068, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.7271042012588545e-05, "output_power": 1.576289319617558e-05, "delta_db": -0.3968260767823113, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2564254077750615, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.212017807553011, "pitch": null, "output_rms_dbfs": -28.369199189162554, "output_peak": 0.39394706845383676, "reference_rms_dbfs": -28.157181381609547, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.438929669199365}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012908667094909063, "log_rms_mae_db": 0.7322714255151785, "rms_envelope_nrmse": 0.07591075492518522, "envelope_correlation": 0.9949579551160301, "rms_gain_error_db": -0.20983726809553122, "reference_mean_power": 0.001527852496322364, "output_mean_power": 0.0014557865455804683}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.179575522794606e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002748180655831064, "output_power": 0.00025690687886565906, "reference_active": true, "delta_db": -0.2926954536576338}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.543131824921543e-05, "output_power": 4.5843756361864865e-05, "reference_active": true, "delta_db": -0.8247501117089309}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2340545383743233e-05, "output_power": 8.944785861048388e-06, "reference_active": true, "delta_db": -1.3976440555143492}], "active_band_count": 3, "active_band_mae_db": 0.8383632069603046}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001146348873577052, "log_rms_mae_db": 0.40830727411986173, "rms_envelope_nrmse": 0.04603415845497645, "envelope_correlation": 0.9983890839097904, "rms_gain_error_db": -0.20877638105190363, "reference_mean_power": 0.0015234711554082644, "output_mean_power": 0.0014519665048876634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.718780555754886e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002667517105928089, "output_power": 0.00024730601434917406, "reference_active": true, "delta_db": -0.3287253479827917}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.819324290787445e-05, "output_power": 2.3154729716071472e-05, "reference_active": true, "delta_db": -0.8550531729807398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.654413380387e-07, "output_power": 5.66154713015346e-07, "reference_active": true, "delta_db": -1.3097678581415775}], "active_band_count": 3, "active_band_mae_db": 0.8311821263683696}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001269978635547725, "log_rms_mae_db": 1.0979461948251223, "rms_envelope_nrmse": 0.09043056008411088, "envelope_correlation": 0.9933834549811161, "rms_gain_error_db": -0.22516390789179175, "reference_mean_power": 0.0011723709196259178, "output_mean_power": 0.0011131370714226446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.280903308109198e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025355561575728737, "output_power": 0.00023521611890522603, "reference_active": true, "delta_db": -0.3260615397356187}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.608107242392719e-05, "output_power": 4.730215925853088e-05, "reference_active": true, "delta_db": -0.7393534398707478}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1840508111446508e-05, "output_power": 7.99792127317041e-06, "reference_active": true, "delta_db": -1.7039321477641631}], "active_band_count": 3, "active_band_mae_db": 0.9231157091235098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.809451727166842e-05, "log_rms_mae_db": 0.6044951228171523, "rms_envelope_nrmse": 0.054986886759008084, "envelope_correlation": 0.9977758254880467, "rms_gain_error_db": -0.22404447102933356, "reference_mean_power": 0.0011689931508753605, "output_mean_power": 0.0011102160961543384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.891072559770159e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025025709822561247, "output_power": 0.00022989680086763558, "reference_active": true, "delta_db": -0.3685347665897009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.836247458716607e-05, "output_power": 2.363202416652457e-05, "reference_active": true, "delta_db": -0.792431976791471}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.847292373572026e-07, "output_power": 5.555053147138343e-07, "reference_active": true, "delta_db": -1.5003161546346644}], "active_band_count": 3, "active_band_mae_db": 0.8870942993386121}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.9423530315000806e-05, "log_rms_mae_db": 1.1129944090694446, "rms_envelope_nrmse": 0.10465091073342613, "envelope_correlation": 0.9925112569717525, "rms_gain_error_db": -0.30239125750530993, "reference_mean_power": 0.00015673633956673046, "output_mean_power": 0.00014619434514040637}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.197939100277272e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3248792094830553e-05, "output_power": 3.065402884270916e-05, "reference_active": true, "delta_db": -0.35288310614951035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.648467501564817e-06, "output_power": 2.625104052751162e-06, "reference_active": true, "delta_db": -1.4296395970023692}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1792733873632069e-06, "output_power": 1.1883534821298344e-06, "reference_active": true, "delta_db": 0.033311451535395054}], "active_band_count": 3, "active_band_mae_db": 0.6052780515624249}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.3506304987563625e-05, "log_rms_mae_db": 0.5860503645143565, "rms_envelope_nrmse": 0.059848488482039844, "envelope_correlation": 0.9981001075985948, "rms_gain_error_db": -0.3039673949986457, "reference_mean_power": 0.0001561782704543471, "output_mean_power": 0.00014562095321784676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.897719153255603e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.156452210778437e-05, "output_power": 2.9042214521461715e-05, "reference_active": true, "delta_db": -0.36169489389493725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6348481818884734e-06, "output_power": 1.168904937901591e-06, "reference_active": true, "delta_db": -1.4569823539705518}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.181225496757285e-08, "output_power": 7.23123436465818e-08, "reference_active": true, "delta_db": 0.030138730996927647}], "active_band_count": 3, "active_band_mae_db": 0.6162719929541389}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.766512572532955e-05, "log_rms_mae_db": 1.2408438258291445, "rms_envelope_nrmse": 0.15307054266965736, "envelope_correlation": 0.9800438534216326, "rms_gain_error_db": -0.3827727181642426, "reference_mean_power": 2.7841403311343547e-05, "output_mean_power": 2.5492583456282423e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1058323154967027e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.079425705940839e-06, "output_power": 4.772779816358791e-06, "reference_active": true, "delta_db": -0.2704321308129918}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8241793691525221e-06, "output_power": 1.5650390957262885e-06, "reference_active": true, "delta_db": -0.6654234871047975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.319091479762654e-07, "output_power": 4.4766281089005883e-07, "reference_active": true, "delta_db": -0.7488644221517335}], "active_band_count": 3, "active_band_mae_db": 0.5615733466898409}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5244936872690136e-05, "log_rms_mae_db": 0.8104471809519731, "rms_envelope_nrmse": 0.09259758038137042, "envelope_correlation": 0.9932856907304144, "rms_gain_error_db": -0.3823336989354792, "reference_mean_power": 2.7868742577273385e-05, "output_mean_power": 2.552019591777516e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.800642728490222e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.984173950209188e-06, "output_power": 4.563123023618278e-06, "reference_active": true, "delta_db": -0.3833101333111031}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0071402238290061e-06, "output_power": 8.295926617860049e-07, "reference_active": true, "delta_db": -0.8422503963400196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932914966264332e-08, "output_power": 2.804770588416649e-08, "reference_active": true, "delta_db": -1.4681721196611288}], "active_band_count": 3, "active_band_mae_db": 0.897910883104084}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8313093077993355, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "efc8826dad73dd560d9408443c83e607a28b20aaf5e67ade747166d65624b74e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9928517252051758, "standard_deviation": 0.0008891552402115317, "minimum": 0.9913134473685428, "maximum": 0.9942783694182853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.122333397257811, "standard_deviation": 0.03438816513245862, "minimum": 1.0549493688116496, "maximum": 1.1621015205940775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46117412036070604, "standard_deviation": 0.13937703362170312, "minimum": 0.33880812215770933, "maximum": 0.7360772183014882, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1001553780560463, "standard_deviation": 0.005709422892866971, "minimum": 0.09033448685506196, "maximum": 0.1091445789708551, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9978630645350779, "standard_deviation": 0.0003167581337406517, "minimum": 0.99732750880886, "maximum": 0.998299021229021, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6199822225794507, "standard_deviation": 0.02677477772689395, "minimum": 0.5860503645143565, "maximum": 0.6514676979535058, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4553195616374451, "standard_deviation": 0.12622477797642814, "minimum": 0.2838904164700204, "maximum": 0.6162719929541389, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05767518968725728, "standard_deviation": 0.0032073175000182476, "minimum": 0.05252137763561638, "maximum": 0.06272905790425815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9808609426825406, "standard_deviation": 0.0017847850083772114, "minimum": 0.9789959429399911, "maximum": 0.9846262549543514, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2736953554060313, "standard_deviation": 0.029212673329528944, "minimum": 1.2253599112681117, "maximum": 1.3170759531796887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4604413037822015, "standard_deviation": 0.19464644330621445, "minimum": 0.1542676589905905, "maximum": 0.7758957457576893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14879034134869562, "standard_deviation": 0.007035465096479711, "minimum": 0.13354366232862422, "maximum": 0.15487663791920692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993266244544005, "standard_deviation": 0.0012284477971767082, "minimum": 0.9917885308611897, "maximum": 0.9957686438942744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8622924591235399, "standard_deviation": 0.04157269062553181, "minimum": 0.802317709342404, "maximum": 0.9306682363541283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7329987046889443, "standard_deviation": 0.3832400747321357, "minimum": 0.1763186570854979, "maximum": 1.2761709726126833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08938136439667459, "standard_deviation": 0.007134703556072112, "minimum": 0.07263800285669171, "maximum": 0.0964687296468052, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931545122591539, "standard_deviation": 0.0007272983832452153, "minimum": 0.9916287729296416, "maximum": 0.9938179496119046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.16631746873273, "standard_deviation": 0.033393224362292154, "minimum": 1.0979461948251223, "maximum": 1.2064603885989063, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5511826212911258, "standard_deviation": 0.26745804503322085, "minimum": 0.15628778413550698, "maximum": 0.9231157091235098, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09000530622262534, "standard_deviation": 0.004581584561561604, "minimum": 0.08488816276430833, "maximum": 0.099493405331882, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976293551446831, "standard_deviation": 0.00042461027264384236, "minimum": 0.9969645013706737, "maximum": 0.9981303196089546, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6267594710932756, "standard_deviation": 0.020201927094250907, "minimum": 0.5988574692685025, "maximum": 0.6596969508573044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6404477586588274, "standard_deviation": 0.31693765716564887, "minimum": 0.08646400176103719, "maximum": 0.9601443111831589, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.052760596643448315, "standard_deviation": 0.004624989358933629, "minimum": 0.0457324954221239, "maximum": 0.05974799710674001, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21398406404236586, "standard_deviation": 0.03197865624907623, "minimum": 0.16626731209732332, "maximum": 0.2564254077750615, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.438451199176383, "standard_deviation": 0.01024853945078323, "minimum": 5.4247578775798715, "maximum": 5.458263393815211, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3877081891047258, "standard_deviation": 0.00811453423182788, "minimum": 0.375152422030691, "maximum": 0.39678067371939063, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1159141104348782, "standard_deviation": 0.06390613351309325, "minimum": -0.212017807553011, "maximum": -0.012024005633546123, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9948143593411085, "standard_deviation": 0.0006176671502074799, "minimum": 0.9935519443246249, "maximum": 0.9954704997010673, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7665501601518598, "standard_deviation": 0.021126194509768863, "minimum": 0.7322714255151785, "maximum": 0.7859340738136937, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45912651777652946, "standard_deviation": 0.23096823721305795, "minimum": 0.14474565282389934, "maximum": 0.8383632069603046, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07484928885677916, "standard_deviation": 0.004208716550235302, "minimum": 0.06950243968915436, "maximum": 0.08336360548059545, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982588095549503, "standard_deviation": 0.00036992627120645705, "minimum": 0.9976571130232963, "maximum": 0.99867324192983, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42132141311332705, "standard_deviation": 0.009232546386712328, "minimum": 0.4062144873556608, "maximum": 0.43043051263355636, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5234889390191506, "standard_deviation": 0.25211550508507147, "minimum": 0.0979514474582512, "maximum": 0.8311821263683696, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04360206223146257, "standard_deviation": 0.004251014556066949, "minimum": 0.037465897243911576, "maximum": 0.04988986180783703, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0010", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.4967901706695557, "predicted_mos": 2.4967901706695557, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bdba5d71644b5f034f80f797ce6491deeec8e0e39ee1784e56c63b11af5022c8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.451362200081348, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7402408028458084, "全帯域_10ms_rms_envelope_nrmse": 0.07524032159381232, "全帯域_10ms_envelope_correlation": 0.9947785959542357, "全帯域_10ms_modulation_mae_db": 0.6201282805787883, "全帯域_40ms_log_rms_mae_db": 0.4057738299483478, "全帯域_40ms_rms_envelope_nrmse": 0.04359195270849526, "全帯域_40ms_envelope_correlation": 0.9982896046737257, "全帯域_40ms_modulation_mae_db": 0.6548611088290942, "300–1000 Hz_10ms_log_rms_mae_db": 1.154936948723749, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09462002230920921, "300–1000 Hz_10ms_envelope_correlation": 0.9924476431297613, "300–1000 Hz_10ms_modulation_mae_db": 0.600027599843648, "300–1000 Hz_40ms_log_rms_mae_db": 0.6119540552891104, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05400425642376365, "300–1000 Hz_40ms_envelope_correlation": 0.9975696703519545, "300–1000 Hz_40ms_modulation_mae_db": 0.7309545493974875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.25216352452222, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11132464156529762, "1000–3000 Hz_10ms_envelope_correlation": 0.9909901556864176, "1000–3000 Hz_10ms_modulation_mae_db": 0.4388032114116835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6682017432652674, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058247095453518694, "1000–3000 Hz_40ms_envelope_correlation": 0.9977289655492505, "1000–3000 Hz_40ms_modulation_mae_db": 0.24614233542097752, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4676246971530034, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16253807084710162, "3000–6000 Hz_10ms_envelope_correlation": 0.9764926516048386, "3000–6000 Hz_10ms_modulation_mae_db": 0.6629176056343412, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9609925826180024, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08777599556745112, "3000–6000 Hz_40ms_envelope_correlation": 0.9929219551495057, "3000–6000 Hz_40ms_modulation_mae_db": 0.9580151985287165, "mr_stft_db": 5.644452218271353, "band_mae_db": 0.16991031175350602, "rms_gain_error_db": -0.13817270621359323, "output_peak": 0.38275859708221116}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7649504397428162, "全帯域_10ms_rms_envelope_nrmse": 0.07609727582578277, "全帯域_10ms_envelope_correlation": 0.9946768150683089, "全帯域_10ms_modulation_mae_db": 0.540024618140862, "全帯域_40ms_log_rms_mae_db": 0.4100050676056608, "全帯域_40ms_rms_envelope_nrmse": 0.041408729799092325, "全帯域_40ms_envelope_correlation": 0.9985116180884527, "全帯域_40ms_modulation_mae_db": 0.7986441905884408, "300–1000 Hz_10ms_log_rms_mae_db": 1.248925406967971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09525513957186373, "300–1000 Hz_10ms_envelope_correlation": 0.9923900920722002, "300–1000 Hz_10ms_modulation_mae_db": 0.5466993710514337, "300–1000 Hz_40ms_log_rms_mae_db": 0.6674185331957326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05104384662285427, "300–1000 Hz_40ms_envelope_correlation": 0.997922702745999, "300–1000 Hz_40ms_modulation_mae_db": 0.900429687277831, "1000–3000 Hz_10ms_log_rms_mae_db": 1.276766724683125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12010651703333128, "1000–3000 Hz_10ms_envelope_correlation": 0.9894165947899459, "1000–3000 Hz_10ms_modulation_mae_db": 0.5380039298901131, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7306808196732179, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06987498830958547, "1000–3000 Hz_40ms_envelope_correlation": 0.9965294805169909, "1000–3000 Hz_40ms_modulation_mae_db": 0.655577349423064, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5238512291602904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17376614892298253, "3000–6000 Hz_10ms_envelope_correlation": 0.9730274736440931, "3000–6000 Hz_10ms_modulation_mae_db": 0.4460463916981284, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9796832195514678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09620484743856057, "3000–6000 Hz_40ms_envelope_correlation": 0.991279583783703, "3000–6000 Hz_40ms_modulation_mae_db": 0.7837820389489702, "mr_stft_db": 5.645570465557891, "band_mae_db": 0.23128721703885233, "rms_gain_error_db": -0.14620644195661475, "output_peak": 0.3828633563916519}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7290609475399492, "全帯域_10ms_rms_envelope_nrmse": 0.06695764900880731, "全帯域_10ms_envelope_correlation": 0.9958322347933649, "全帯域_10ms_modulation_mae_db": 0.6738330353133595, "全帯域_40ms_log_rms_mae_db": 0.38403576412206947, "全帯域_40ms_rms_envelope_nrmse": 0.0377704805325825, "全帯域_40ms_envelope_correlation": 0.9986861435973415, "全帯域_40ms_modulation_mae_db": 0.947706576578916, "300–1000 Hz_10ms_log_rms_mae_db": 1.1529275089305935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0828323107036491, "300–1000 Hz_10ms_envelope_correlation": 0.9942247418109667, "300–1000 Hz_10ms_modulation_mae_db": 0.6849943704791123, "300–1000 Hz_40ms_log_rms_mae_db": 0.567810448676389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04722364216898476, "300–1000 Hz_40ms_envelope_correlation": 0.99814789714139, "300–1000 Hz_40ms_modulation_mae_db": 1.0058363988849086, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2615527544366454, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13034563758561535, "1000–3000 Hz_10ms_envelope_correlation": 0.9873908422676962, "1000–3000 Hz_10ms_modulation_mae_db": 0.3833679258123, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7088570767432841, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07657450550033872, "1000–3000 Hz_40ms_envelope_correlation": 0.9955721097110004, "1000–3000 Hz_40ms_modulation_mae_db": 0.3803070471441696, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4906795843860916, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16052190515002943, "3000–6000 Hz_10ms_envelope_correlation": 0.9772309111782753, "3000–6000 Hz_10ms_modulation_mae_db": 0.7774961632295944, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9848909103581617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09706846944166982, "3000–6000 Hz_40ms_envelope_correlation": 0.9914667585867837, "3000–6000 Hz_40ms_modulation_mae_db": 1.2510034283406386, "mr_stft_db": 5.646265072058154, "band_mae_db": 0.20212922622835192, "rms_gain_error_db": -0.10517708411702523, "output_peak": 0.39154486036907943}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7562413088143654, "全帯域_10ms_rms_envelope_nrmse": 0.07392352911455023, "全帯域_10ms_envelope_correlation": 0.9949504781112706, "全帯域_10ms_modulation_mae_db": 0.748412949160229, "全帯域_40ms_log_rms_mae_db": 0.38390838362819674, "全帯域_40ms_rms_envelope_nrmse": 0.039223311779677514, "全帯域_40ms_envelope_correlation": 0.9986466208124868, "全帯域_40ms_modulation_mae_db": 0.8361412626588614, "300–1000 Hz_10ms_log_rms_mae_db": 1.2065014856879854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08919973716382597, "300–1000 Hz_10ms_envelope_correlation": 0.9933194055617871, "300–1000 Hz_10ms_modulation_mae_db": 0.7780153815343143, "300–1000 Hz_40ms_log_rms_mae_db": 0.5879479445687159, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047311473013353, "300–1000 Hz_40ms_envelope_correlation": 0.9982055577927914, "300–1000 Hz_40ms_modulation_mae_db": 0.840985415078484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2902288221467286, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10513815082728566, "1000–3000 Hz_10ms_envelope_correlation": 0.991885592998071, "1000–3000 Hz_10ms_modulation_mae_db": 0.1740327383791412, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7238523443087962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05733180505848132, "1000–3000 Hz_40ms_envelope_correlation": 0.9976565885269372, "1000–3000 Hz_40ms_modulation_mae_db": 0.35881804946736945, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5372733164962937, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17976848083085872, "3000–6000 Hz_10ms_envelope_correlation": 0.9711079626643286, "3000–6000 Hz_10ms_modulation_mae_db": 0.2980133564928557, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9832030308106329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09890473730702919, "3000–6000 Hz_40ms_envelope_correlation": 0.9907659913809046, "3000–6000 Hz_40ms_modulation_mae_db": 0.6521052124837318, "mr_stft_db": 5.645130937692841, "band_mae_db": 0.12286623731524197, "rms_gain_error_db": -0.12788851831958634, "output_peak": 0.38291125701586803}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7133058141534592, "全帯域_10ms_rms_envelope_nrmse": 0.07223140436721726, "全帯域_10ms_envelope_correlation": 0.9951609476360209, "全帯域_10ms_modulation_mae_db": 0.4023672278372113, "全帯域_40ms_log_rms_mae_db": 0.3559017227524708, "全帯域_40ms_rms_envelope_nrmse": 0.03857393987654368, "全帯域_40ms_envelope_correlation": 0.9986651080730337, "全帯域_40ms_modulation_mae_db": 0.4778878850912112, "300–1000 Hz_10ms_log_rms_mae_db": 1.1581182558768819, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08615316444848932, "300–1000 Hz_10ms_envelope_correlation": 0.993771859941231, "300–1000 Hz_10ms_modulation_mae_db": 0.504141741845843, "300–1000 Hz_40ms_log_rms_mae_db": 0.599515014939076, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04632335538887707, "300–1000 Hz_40ms_envelope_correlation": 0.9982771498020879, "300–1000 Hz_40ms_modulation_mae_db": 0.5877431322839836, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2579896726405528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10689036231741471, "1000–3000 Hz_10ms_envelope_correlation": 0.991542214541399, "1000–3000 Hz_10ms_modulation_mae_db": 0.47032918933070594, "1000–3000 Hz_40ms_log_rms_mae_db": 0.662331049527442, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06141887575782807, "1000–3000 Hz_40ms_envelope_correlation": 0.9971784919819023, "1000–3000 Hz_40ms_modulation_mae_db": 0.38191599763363987, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4756519489864222, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1635565352017534, "3000–6000 Hz_10ms_envelope_correlation": 0.9762653491470865, "3000–6000 Hz_10ms_modulation_mae_db": 0.7446673447357867, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9365112908375317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0860299922782003, "3000–6000 Hz_40ms_envelope_correlation": 0.993386318067303, "3000–6000 Hz_40ms_modulation_mae_db": 1.1592916988861128, "mr_stft_db": 5.6344971094597085, "band_mae_db": 0.16734984940803221, "rms_gain_error_db": -0.12072888839087104, "output_peak": 0.38019360362454935}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7500503166628976, "全帯域_10ms_rms_envelope_nrmse": 0.0768793312560773, "全帯域_10ms_envelope_correlation": 0.9945282358101846, "全帯域_10ms_modulation_mae_db": 0.5471580145989751, "全帯域_40ms_log_rms_mae_db": 0.4256665077208668, "全帯域_40ms_rms_envelope_nrmse": 0.04622405618767616, "全帯域_40ms_envelope_correlation": 0.9980380591822986, "全帯域_40ms_modulation_mae_db": 0.6035349509027226, "300–1000 Hz_10ms_log_rms_mae_db": 1.1825583826121713, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09125505456371892, "300–1000 Hz_10ms_envelope_correlation": 0.9929989467106132, "300–1000 Hz_10ms_modulation_mae_db": 0.5865070805345095, "300–1000 Hz_40ms_log_rms_mae_db": 0.6693684188678451, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05503055880333403, "300–1000 Hz_40ms_envelope_correlation": 0.9974666935987532, "300–1000 Hz_40ms_modulation_mae_db": 0.6873659839358783, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2526129444253193, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1059290761172779, "1000–3000 Hz_10ms_envelope_correlation": 0.9917489457423997, "1000–3000 Hz_10ms_modulation_mae_db": 0.3653118718027497, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6989122559833776, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053517876865781355, "1000–3000 Hz_40ms_envelope_correlation": 0.9979698656752692, "1000–3000 Hz_40ms_modulation_mae_db": 0.5034463651869582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.47182039038755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16627065932161297, "3000–6000 Hz_10ms_envelope_correlation": 0.9754116867995328, "3000–6000 Hz_10ms_modulation_mae_db": 0.6306557937641275, "3000–6000 Hz_40ms_log_rms_mae_db": 0.920957646344683, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09080878042155495, "3000–6000 Hz_40ms_envelope_correlation": 0.9924427652786059, "3000–6000 Hz_40ms_modulation_mae_db": 1.0760959628429594, "mr_stft_db": 5.626583150560715, "band_mae_db": 0.169160101102211, "rms_gain_error_db": -0.13318667842246726, "output_peak": 0.39316462863080004}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7577763327856655, "全帯域_10ms_rms_envelope_nrmse": 0.07313025411135478, "全帯域_10ms_envelope_correlation": 0.9953169224939356, "全帯域_10ms_modulation_mae_db": 0.54213810070946, "全帯域_40ms_log_rms_mae_db": 0.39072440578485107, "全帯域_40ms_rms_envelope_nrmse": 0.04124097320392626, "全帯域_40ms_envelope_correlation": 0.9987761786000138, "全帯域_40ms_modulation_mae_db": 0.6695160415339022, "300–1000 Hz_10ms_log_rms_mae_db": 1.1905355587530047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09273218388518446, "300–1000 Hz_10ms_envelope_correlation": 0.993044924736458, "300–1000 Hz_10ms_modulation_mae_db": 0.5062257882931388, "300–1000 Hz_40ms_log_rms_mae_db": 0.6289267151426503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05266581961446421, "300–1000 Hz_40ms_envelope_correlation": 0.9980659358708699, "300–1000 Hz_40ms_modulation_mae_db": 0.7254701825006954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2798555335529893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10818583350413852, "1000–3000 Hz_10ms_envelope_correlation": 0.9914214250058531, "1000–3000 Hz_10ms_modulation_mae_db": 0.11264300567603143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7077917669896451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06349712277163842, "1000–3000 Hz_40ms_envelope_correlation": 0.9970986309134574, "1000–3000 Hz_40ms_modulation_mae_db": 0.18426733927859726, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4884689665985602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1674595004125446, "3000–6000 Hz_10ms_envelope_correlation": 0.9757097985863458, "3000–6000 Hz_10ms_modulation_mae_db": 1.08993012236862, "3000–6000 Hz_40ms_log_rms_mae_db": 1.015310428344623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09939934839020763, "3000–6000 Hz_40ms_envelope_correlation": 0.9919565328802006, "3000–6000 Hz_40ms_modulation_mae_db": 1.809666061389546, "mr_stft_db": 5.655656670852747, "band_mae_db": 0.2632096062014763, "rms_gain_error_db": -0.2025760975370404, "output_peak": 0.3927204357179986}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.73867536740024, "全帯域_10ms_rms_envelope_nrmse": 0.07167867555103426, "全帯域_10ms_envelope_correlation": 0.9951314753004905, "全帯域_10ms_modulation_mae_db": 0.06292128553970565, "全帯域_40ms_log_rms_mae_db": 0.38779728311483713, "全帯域_40ms_rms_envelope_nrmse": 0.039963225430264246, "全帯域_40ms_envelope_correlation": 0.9984069772266481, "全帯域_40ms_modulation_mae_db": 0.05734177778418103, "300–1000 Hz_10ms_log_rms_mae_db": 1.162549778886543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08774168507040253, "300–1000 Hz_10ms_envelope_correlation": 0.9933781265620879, "300–1000 Hz_10ms_modulation_mae_db": 0.1058834724097268, "300–1000 Hz_40ms_log_rms_mae_db": 0.5884650010788046, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04870743322139077, "300–1000 Hz_40ms_envelope_correlation": 0.9978314364076887, "300–1000 Hz_40ms_modulation_mae_db": 0.09644603105605332, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2643823639799976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1087724402677973, "1000–3000 Hz_10ms_envelope_correlation": 0.9913883769198334, "1000–3000 Hz_10ms_modulation_mae_db": 0.4078685382075305, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7088908966894848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06252493133587358, "1000–3000 Hz_40ms_envelope_correlation": 0.9973024586333012, "1000–3000 Hz_40ms_modulation_mae_db": 0.3428425915628848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5320443156444064, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16013241787994503, "3000–6000 Hz_10ms_envelope_correlation": 0.9776887846457112, "3000–6000 Hz_10ms_modulation_mae_db": 1.1515645888046597, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9871500854489986, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09074564371402111, "3000–6000 Hz_40ms_envelope_correlation": 0.9932221639044687, "3000–6000 Hz_40ms_modulation_mae_db": 1.6049936376019813, "mr_stft_db": 5.64270701690442, "band_mae_db": 0.15080319094532188, "rms_gain_error_db": -0.0564335675830279, "output_peak": 0.3917838327832537}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.531695548559955, "spectral_convergence": 0.21894775264518931, "spectral_convergence_status": "ok", "amplitude_floor": 9.509362929412646e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.492654163391992, "spectral_convergence": 0.2211034076888102, "spectral_convergence_status": "ok", "amplitude_floor": 8.503880220761315e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.909006942862112, "spectral_convergence": 0.22687637926603646, "spectral_convergence_status": "ok", "amplitude_floor": 5.089992363719204e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.8030381557821282e-05, "output_power": 3.0127597714182715e-05, "delta_db": 0.3133549414723309, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003997973981863815, "output_power": 0.00040089663473848994, "delta_db": 0.011924471750141212, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008734907128440196, "output_power": 0.0008316506817163413, "delta_db": -0.21317343969721986, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001464675075572333, "output_power": 0.00014178426106888428, "delta_db": -0.14113267072368843, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.350074484130352e-05, "output_power": 5.9060752239146347e-05, "delta_db": -0.3147984529941819, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.7271042012588545e-05, "output_power": 1.717159952075497e-05, "delta_db": -0.02507789388347384, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16991031175350602, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13817270621359323, "pitch": null, "output_rms_dbfs": -28.295354087823135, "output_peak": 0.38275859708221116, "reference_rms_dbfs": -28.157181381609547, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.644452218271353}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012908667094909063, "log_rms_mae_db": 0.7402408028458084, "rms_envelope_nrmse": 0.07524032159381232, "envelope_correlation": 0.9947785959542357, "rms_gain_error_db": -0.1355730437258696, "reference_mean_power": 0.001527852496322364, "output_mean_power": 0.001480894512169991}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.179575522794606e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002748180655831064, "output_power": 0.0002610539318102297, "reference_active": true, "delta_db": -0.22315039700333997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.543131824921543e-05, "output_power": 5.215183517743104e-05, "reference_active": true, "delta_db": -0.2648561148839928}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2340545383743233e-05, "output_power": 8.996975141803194e-06, "reference_active": true, "delta_db": -1.3723783298490322}], "active_band_count": 3, "active_band_mae_db": 0.6201282805787883}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001146348873577052, "log_rms_mae_db": 0.4057738299483478, "rms_envelope_nrmse": 0.04359195270849526, "envelope_correlation": 0.9982896046737257, "rms_gain_error_db": -0.1333406403808995, "reference_mean_power": 0.0015234711554082644, "output_mean_power": 0.001477407066361478}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.718780555754886e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002667517105928089, "output_power": 0.00025263548035151954, "reference_active": true, "delta_db": -0.2361286989019146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.819324290787445e-05, "output_power": 2.6838905256573244e-05, "reference_active": true, "delta_db": -0.21380235923755897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.654413380387e-07, "output_power": 5.400658121740752e-07, "reference_active": true, "delta_db": -1.514652268347809}], "active_band_count": 3, "active_band_mae_db": 0.6548611088290942}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001269978635547725, "log_rms_mae_db": 1.154936948723749, "rms_envelope_nrmse": 0.09462002230920921, "envelope_correlation": 0.9924476431297613, "rms_gain_error_db": -0.15737560656102154, "reference_mean_power": 0.0011723709196259178, "output_mean_power": 0.0011306481505441454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.280903308109198e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025355561575728737, "output_power": 0.00023534847572811158, "reference_active": true, "delta_db": -0.32361843883117825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.608107242392719e-05, "output_power": 5.2165379402052153e-05, "reference_active": true, "delta_db": -0.31433939559332374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1840508111446508e-05, "output_power": 9.060618467700416e-06, "reference_active": true, "delta_db": -1.1621249651064418}], "active_band_count": 3, "active_band_mae_db": 0.600027599843648}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.809451727166842e-05, "log_rms_mae_db": 0.6119540552891104, "rms_envelope_nrmse": 0.05400425642376365, "envelope_correlation": 0.9975696703519545, "rms_gain_error_db": -0.15518832119993833, "reference_mean_power": 0.0011689931508753605, "output_mean_power": 0.0011279585342824075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.891072559770159e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025025709822561247, "output_power": 0.00023224182406619755, "reference_active": true, "delta_db": -0.32445970680313124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.836247458716607e-05, "output_power": 2.643126960618028e-05, "reference_active": true, "delta_db": -0.3062609512250002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.847292373572026e-07, "output_power": 5.476530803876921e-07, "reference_active": true, "delta_db": -1.5621429901643311}], "active_band_count": 3, "active_band_mae_db": 0.7309545493974875}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.9423530315000806e-05, "log_rms_mae_db": 1.25216352452222, "rms_envelope_nrmse": 0.11132464156529762, "envelope_correlation": 0.9909901556864176, "rms_gain_error_db": -0.1821633998350714, "reference_mean_power": 0.00015673633956673046, "output_mean_power": 0.00015029805485432368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.197939100277272e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3248792094830553e-05, "output_power": 3.067082405214978e-05, "reference_active": true, "delta_db": -0.35050427713630644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.648467501564817e-06, "output_power": 3.466724054307322e-06, "reference_active": true, "delta_db": -0.22191208276559665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1792733873632069e-06, "output_power": 1.3996318778703787e-06, "reference_active": true, "delta_db": 0.7439932743331473}], "active_band_count": 3, "active_band_mae_db": 0.4388032114116835}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.3506304987563625e-05, "log_rms_mae_db": 0.6682017432652674, "rms_envelope_nrmse": 0.058247095453518694, "envelope_correlation": 0.9977289655492505, "rms_gain_error_db": -0.17827048450966707, "reference_mean_power": 0.0001561782704543471, "output_mean_power": 0.00014989721383399765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.897719153255603e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.156452210778437e-05, "output_power": 2.906593453745826e-05, "reference_active": true, "delta_db": -0.3581492732558447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6348481818884734e-06, "output_power": 1.6117668956601538e-06, "reference_active": true, "delta_db": -0.06175197176835938}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.181225496757285e-08, "output_power": 7.727715773570124e-08, "reference_active": true, "delta_db": 0.31852576123872844}], "active_band_count": 3, "active_band_mae_db": 0.24614233542097752}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.766512572532955e-05, "log_rms_mae_db": 1.4676246971530034, "rms_envelope_nrmse": 0.16253807084710162, "envelope_correlation": 0.9764926516048386, "rms_gain_error_db": -0.17126120773868803, "reference_mean_power": 2.7841403311343547e-05, "output_mean_power": 2.676486152504795e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1058323154967027e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.079425705940839e-06, "output_power": 4.622943230703635e-06, "reference_active": true, "delta_db": -0.4089605213573009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8241793691525221e-06, "output_power": 1.5637852190819663e-06, "reference_active": true, "delta_db": -0.668904358396873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.319091479762654e-07, "output_power": 4.3126942108459587e-07, "reference_active": true, "delta_db": -0.91088793714885}], "active_band_count": 3, "active_band_mae_db": 0.6629176056343412}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5244936872690136e-05, "log_rms_mae_db": 0.9609925826180024, "rms_envelope_nrmse": 0.08777599556745112, "envelope_correlation": 0.9929219551495057, "rms_gain_error_db": -0.17302306306998236, "reference_mean_power": 2.7868742577273385e-05, "output_mean_power": 2.6780277179999355e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.800642728490222e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.984173950209188e-06, "output_power": 4.448845836666934e-06, "reference_active": true, "delta_db": -0.49345834365935254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0071402238290061e-06, "output_power": 8.409109044742428e-07, "reference_active": true, "delta_db": -0.78339957242282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932914966264332e-08, "output_power": 2.7226749062556456e-08, "reference_active": true, "delta_db": -1.597187679503977}], "active_band_count": 3, "active_band_mae_db": 0.9580151985287165}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8063846167182096, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "26ecb55bb3ca5780b9567fcc2e0438914219de0a4c014f20435b4e2d2b6e642f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.990723018493952, "standard_deviation": 0.00155320707056889, "minimum": 0.9873908422676962, "maximum": 0.991885592998071, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2669440425484473, "standard_deviation": 0.013861692623908786, "minimum": 1.25216352452222, "maximum": 1.2902288221467286, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3612950513137819, "standard_deviation": 0.1457441181684724, "minimum": 0.11264300567603143, "maximum": 0.5380039298901131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11208658240226979, "standard_deviation": 0.008764283604819572, "minimum": 0.10513815082728566, "maximum": 0.13034563758561535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971295739385136, "standard_deviation": 0.0007710986022081383, "minimum": 0.9955721097110004, "maximum": 0.9979698656752692, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7011897441475643, "standard_deviation": 0.02434394242361711, "minimum": 0.662331049527442, "maximum": 0.7306808196732179, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3816646343897076, "standard_deviation": 0.14600487992154934, "minimum": 0.18426733927859726, "maximum": 0.655577349423064, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0628734001316307, "standard_deviation": 0.007358245262800666, "minimum": 0.053517876865781355, "maximum": 0.07657450550033872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9753668272837765, "standard_deviation": 0.0022255420426730717, "minimum": 0.9711079626643286, "maximum": 0.9776887846457112, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4984268061015773, "standard_deviation": 0.02833506915189164, "minimum": 1.4676246971530034, "maximum": 1.5372733164962937, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7251614208410142, "standard_deviation": 0.29060566016558464, "minimum": 0.2980133564928557, "maximum": 1.1515645888046597, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16675171482085355, "standard_deviation": 0.0068656541341870145, "minimum": 0.16013241787994503, "maximum": 0.17976848083085872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921802586289343, "standard_deviation": 0.0009658210414158826, "minimum": 0.9907659913809046, "maximum": 0.993386318067303, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9710873992892626, "standard_deviation": 0.030320442151989997, "minimum": 0.920957646344683, "maximum": 1.015310428344623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1618691548778322, "standard_deviation": 0.3920047961529555, "minimum": 0.6521052124837318, "maximum": 1.809666061389546, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09336722681983683, "standard_deviation": 0.005172965774750718, "minimum": 0.0860299922782003, "maximum": 0.09939934839020763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931969675656381, "standard_deviation": 0.0006221501014125816, "minimum": 0.9923900920722002, "maximum": 0.9942247418109667, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1821316658048624, "standard_deviation": 0.033120190327108966, "minimum": 1.1529275089305935, "maximum": 1.248925406967971, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5390618507489658, "standard_deviation": 0.19797634157431368, "minimum": 0.1058834724097268, "maximum": 0.7780153815343143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0899736622145429, "standard_deviation": 0.004307963720230969, "minimum": 0.0828323107036491, "maximum": 0.09525513957186373, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979358804639418, "standard_deviation": 0.00029674050158479755, "minimum": 0.9974666935987532, "maximum": 0.9982771498020879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6151757664697906, "standard_deviation": 0.037434075391429414, "minimum": 0.567810448676389, "maximum": 0.6693684188678451, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6969039225519151, "standard_deviation": 0.2757207942284691, "minimum": 0.09644603105605332, "maximum": 1.0058363988849086, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05028879815712772, "standard_deviation": 0.0033593016069013463, "minimum": 0.04632335538887707, "maximum": 0.05503055880333403, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1845894674991242, "standard_deviation": 0.045310777933188925, "minimum": 0.12286623731524197, "maximum": 0.2632096062014763, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6426078301697284, "standard_deviation": 0.008658412044305846, "minimum": 5.626583150560715, "maximum": 5.655656670852747, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.38724257145192653, "standard_deviation": 0.00550249664660413, "minimum": 0.38019360362454935, "maximum": 0.39316462863080004, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12879624781752827, "standard_deviation": 0.04090739294541901, "minimum": -0.2025760975370404, "maximum": -0.0564335675830279, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9950469631459764, "standard_deviation": 0.00041353568472689327, "minimum": 0.9945282358101846, "maximum": 0.9958322347933649, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7437876662431502, "standard_deviation": 0.016998298937612344, "minimum": 0.7133058141534592, "maximum": 0.7649504397428162, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5171229389848239, "standard_deviation": 0.21050401751165307, "minimum": 0.06292128553970565, "maximum": 0.748412949160229, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07326730510357952, "standard_deviation": 0.003135198067056926, "minimum": 0.06695764900880731, "maximum": 0.0768793312560773, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985025387817501, "standard_deviation": 0.00024671399340354326, "minimum": 0.9980380591822986, "maximum": 0.9987761786000138, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.39297662058466254, "standard_deviation": 0.02101354389640949, "minimum": 0.3559017227524708, "maximum": 0.4256665077208668, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6307042242459162, "standard_deviation": 0.2740761392139198, "minimum": 0.05734177778418103, "maximum": 0.947706576578916, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04099958368978224, "standard_deviation": 0.002797478326124986, "minimum": 0.0377704805325825, "maximum": 0.04622405618767616, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0010", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.580155611038208, "predicted_mos": 2.580155611038208, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3fff66edc6ea44345e2e7ac836123cbfba27e3b8573071d79e0c291febcf219c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.944582900032401, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7689959258980067, "全帯域_10ms_rms_envelope_nrmse": 0.07564324501332416, "全帯域_10ms_envelope_correlation": 0.9946232167172238, "全帯域_10ms_modulation_mae_db": 0.0844454551568699, "全帯域_40ms_log_rms_mae_db": 0.39943869377266555, "全帯域_40ms_rms_envelope_nrmse": 0.03841657989628656, "全帯域_40ms_envelope_correlation": 0.9986049100880423, "全帯域_40ms_modulation_mae_db": 0.20681729417741337, "300–1000 Hz_10ms_log_rms_mae_db": 1.1560192628118977, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09227551577796038, "300–1000 Hz_10ms_envelope_correlation": 0.9927711545258427, "300–1000 Hz_10ms_modulation_mae_db": 0.1468979687591993, "300–1000 Hz_40ms_log_rms_mae_db": 0.5981594671205527, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048030104587402964, "300–1000 Hz_40ms_envelope_correlation": 0.9980565228665024, "300–1000 Hz_40ms_modulation_mae_db": 0.3161651317017235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0645515151638347, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10623439089833282, "1000–3000 Hz_10ms_envelope_correlation": 0.9916386093309704, "1000–3000 Hz_10ms_modulation_mae_db": 0.6574940850193197, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6079536957351119, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06296667527723879, "1000–3000 Hz_40ms_envelope_correlation": 0.9969632214887244, "1000–3000 Hz_40ms_modulation_mae_db": 0.7876760387821576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2060713758598283, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14321261756863238, "3000–6000 Hz_10ms_envelope_correlation": 0.9819225418956665, "3000–6000 Hz_10ms_modulation_mae_db": 0.2459844082022149, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8060415519351711, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08447992910404925, "3000–6000 Hz_40ms_envelope_correlation": 0.993288103321392, "3000–6000 Hz_40ms_modulation_mae_db": 0.4868498779640942, "mr_stft_db": 5.408972974381309, "band_mae_db": 0.16947321251112926, "rms_gain_error_db": -0.09547375114254614, "output_peak": 0.374200027555859}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7342778730785653, "全帯域_10ms_rms_envelope_nrmse": 0.07643311902675715, "全帯域_10ms_envelope_correlation": 0.9945329366632875, "全帯域_10ms_modulation_mae_db": 0.2464527441948945, "全帯域_40ms_log_rms_mae_db": 0.38516799850231337, "全帯域_40ms_rms_envelope_nrmse": 0.04113620313421903, "全帯域_40ms_envelope_correlation": 0.9984105736780309, "全帯域_40ms_modulation_mae_db": 0.3633702367329099, "300–1000 Hz_10ms_log_rms_mae_db": 1.1186321660417637, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0935343101385162, "300–1000 Hz_10ms_envelope_correlation": 0.9925650142027489, "300–1000 Hz_10ms_modulation_mae_db": 0.19709534736400783, "300–1000 Hz_40ms_log_rms_mae_db": 0.5865406131308336, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050710488623874755, "300–1000 Hz_40ms_envelope_correlation": 0.9977897066627717, "300–1000 Hz_40ms_modulation_mae_db": 0.39893239132903985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.047399741435979, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0998090161741664, "1000–3000 Hz_10ms_envelope_correlation": 0.992772285731643, "1000–3000 Hz_10ms_modulation_mae_db": 0.3073952855503262, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6161269746665932, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05833618610611516, "1000–3000 Hz_40ms_envelope_correlation": 0.997639222917951, "1000–3000 Hz_40ms_modulation_mae_db": 0.2744679278340241, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1662424378215641, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13568619159255318, "3000–6000 Hz_10ms_envelope_correlation": 0.9843992158327248, "3000–6000 Hz_10ms_modulation_mae_db": 0.8694623805864907, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8425052970098836, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08250194005017626, "3000–6000 Hz_40ms_envelope_correlation": 0.9946566698540767, "3000–6000 Hz_40ms_modulation_mae_db": 1.17397554584881, "mr_stft_db": 5.4171645649692275, "band_mae_db": 0.2556000586703739, "rms_gain_error_db": -0.10875205861000999, "output_peak": 0.39691474268542715}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7232751018858052, "全帯域_10ms_rms_envelope_nrmse": 0.07134593836277511, "全帯域_10ms_envelope_correlation": 0.9952558558899585, "全帯域_10ms_modulation_mae_db": 0.13582398739314827, "全帯域_40ms_log_rms_mae_db": 0.41225476403852945, "全帯域_40ms_rms_envelope_nrmse": 0.04123987099428727, "全帯域_40ms_envelope_correlation": 0.9984045782629368, "全帯域_40ms_modulation_mae_db": 0.16690321069988, "300–1000 Hz_10ms_log_rms_mae_db": 1.132554337889043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08836302786242109, "300–1000 Hz_10ms_envelope_correlation": 0.9933971265900396, "300–1000 Hz_10ms_modulation_mae_db": 0.14192211109635525, "300–1000 Hz_40ms_log_rms_mae_db": 0.6474920663167614, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05118917688105731, "300–1000 Hz_40ms_envelope_correlation": 0.9977614602178864, "300–1000 Hz_40ms_modulation_mae_db": 0.17935350264259298, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0607270224375132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09704847788779417, "1000–3000 Hz_10ms_envelope_correlation": 0.9931084846341263, "1000–3000 Hz_10ms_modulation_mae_db": 0.16302933681717188, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6301662093509607, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054759891337385874, "1000–3000 Hz_40ms_envelope_correlation": 0.9978368550167797, "1000–3000 Hz_40ms_modulation_mae_db": 0.2724132725334534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1640720989257356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13090794347591622, "3000–6000 Hz_10ms_envelope_correlation": 0.984926676310385, "3000–6000 Hz_10ms_modulation_mae_db": 0.28609166177146234, "3000–6000 Hz_40ms_log_rms_mae_db": 0.807233761761362, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08162461449229384, "3000–6000 Hz_40ms_envelope_correlation": 0.9937982740125895, "3000–6000 Hz_40ms_modulation_mae_db": 0.15530424136275991, "mr_stft_db": 5.397248799085488, "band_mae_db": 0.16089916846560812, "rms_gain_error_db": -0.10815901816966961, "output_peak": 0.38357973656836997}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7643706768799862, "全帯域_10ms_rms_envelope_nrmse": 0.07488168415172256, "全帯域_10ms_envelope_correlation": 0.9947911014793968, "全帯域_10ms_modulation_mae_db": 0.13995899399030584, "全帯域_40ms_log_rms_mae_db": 0.42508761802061806, "全帯域_40ms_rms_envelope_nrmse": 0.04047581155261982, "全帯域_40ms_envelope_correlation": 0.9985176216415967, "全帯域_40ms_modulation_mae_db": 0.1467392788752863, "300–1000 Hz_10ms_log_rms_mae_db": 1.1412598304224817, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08787792630250964, "300–1000 Hz_10ms_envelope_correlation": 0.9934735718706824, "300–1000 Hz_10ms_modulation_mae_db": 0.11044711296672056, "300–1000 Hz_40ms_log_rms_mae_db": 0.6339294584484011, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048389146154054166, "300–1000 Hz_40ms_envelope_correlation": 0.9980430142794399, "300–1000 Hz_40ms_modulation_mae_db": 0.17231233986750816, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0327408308075396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0958458723497977, "1000–3000 Hz_10ms_envelope_correlation": 0.9933504042574026, "1000–3000 Hz_10ms_modulation_mae_db": 0.33933753501220904, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5598737029717756, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05083170745950652, "1000–3000 Hz_40ms_envelope_correlation": 0.9982575536970625, "1000–3000 Hz_40ms_modulation_mae_db": 0.38244989030822385, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1468954159959102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14226612653980206, "3000–6000 Hz_10ms_envelope_correlation": 0.9828658224422239, "3000–6000 Hz_10ms_modulation_mae_db": 0.23985554948476162, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8015851342291926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09304204661559037, "3000–6000 Hz_40ms_envelope_correlation": 0.993082145672888, "3000–6000 Hz_40ms_modulation_mae_db": 0.09009545905495085, "mr_stft_db": 5.410903552177795, "band_mae_db": 0.2435860635065242, "rms_gain_error_db": -0.12368116669446519, "output_peak": 0.3959526711017437}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7286646127427607, "全帯域_10ms_rms_envelope_nrmse": 0.06802123255153816, "全帯域_10ms_envelope_correlation": 0.9956417582084458, "全帯域_10ms_modulation_mae_db": 0.0917355527355966, "全帯域_40ms_log_rms_mae_db": 0.3773099562037902, "全帯域_40ms_rms_envelope_nrmse": 0.03652237557683984, "全帯域_40ms_envelope_correlation": 0.9987093242682215, "全帯域_40ms_modulation_mae_db": 0.07839909987700332, "300–1000 Hz_10ms_log_rms_mae_db": 1.0478750077679504, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.081808585766605, "300–1000 Hz_10ms_envelope_correlation": 0.9942525818994247, "300–1000 Hz_10ms_modulation_mae_db": 0.11816874346811627, "300–1000 Hz_40ms_log_rms_mae_db": 0.5639968849386819, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04267907881938801, "300–1000 Hz_40ms_envelope_correlation": 0.998361490484365, "300–1000 Hz_40ms_modulation_mae_db": 0.13198896230351764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0821768132098324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0987031394450816, "1000–3000 Hz_10ms_envelope_correlation": 0.9930688437853489, "1000–3000 Hz_10ms_modulation_mae_db": 0.3813705619557732, "1000–3000 Hz_40ms_log_rms_mae_db": 0.594461766978074, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0548727471717049, "1000–3000 Hz_40ms_envelope_correlation": 0.9980941016463869, "1000–3000 Hz_40ms_modulation_mae_db": 0.3624557102670092, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1798446573863892, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14246270951196113, "3000–6000 Hz_10ms_envelope_correlation": 0.9833984680165058, "3000–6000 Hz_10ms_modulation_mae_db": 0.46399616088792356, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8233646811300768, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08778025252531851, "3000–6000 Hz_40ms_envelope_correlation": 0.9947842194626458, "3000–6000 Hz_40ms_modulation_mae_db": 0.8071584903795198, "mr_stft_db": 5.408524504999909, "band_mae_db": 0.2820266220462053, "rms_gain_error_db": -0.07653708561039023, "output_peak": 0.3691420806231504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7741963214461444, "全帯域_10ms_rms_envelope_nrmse": 0.08652139290653128, "全帯域_10ms_envelope_correlation": 0.9928793479684446, "全帯域_10ms_modulation_mae_db": 0.41139844639031703, "全帯域_40ms_log_rms_mae_db": 0.4377628809963626, "全帯域_40ms_rms_envelope_nrmse": 0.051412420039639965, "全帯域_40ms_envelope_correlation": 0.9973348703442194, "全帯域_40ms_modulation_mae_db": 0.4325828909464226, "300–1000 Hz_10ms_log_rms_mae_db": 1.1383992482020011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10227660892758639, "300–1000 Hz_10ms_envelope_correlation": 0.9909851979323949, "300–1000 Hz_10ms_modulation_mae_db": 0.5069290611348519, "300–1000 Hz_40ms_log_rms_mae_db": 0.6292412134568087, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059311512935237745, "300–1000 Hz_40ms_envelope_correlation": 0.9967629413270933, "300–1000 Hz_40ms_modulation_mae_db": 0.4906900795869203, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0914040638079976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10020761556518942, "1000–3000 Hz_10ms_envelope_correlation": 0.993016748661539, "1000–3000 Hz_10ms_modulation_mae_db": 0.23505637384305755, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5734225082860188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05653519598917434, "1000–3000 Hz_40ms_envelope_correlation": 0.9981940146837399, "1000–3000 Hz_40ms_modulation_mae_db": 0.30836600528486163, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1720240496343226, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14556643542893186, "3000–6000 Hz_10ms_envelope_correlation": 0.9813559588476501, "3000–6000 Hz_10ms_modulation_mae_db": 0.5245463078871283, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8220986836058082, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09405207894112398, "3000–6000 Hz_40ms_envelope_correlation": 0.9917014697325459, "3000–6000 Hz_40ms_modulation_mae_db": 0.8077352527992421, "mr_stft_db": 5.393424295740394, "band_mae_db": 0.17409907827079107, "rms_gain_error_db": -0.04872281632493935, "output_peak": 0.39692425101311424}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7677016396995767, "全帯域_10ms_rms_envelope_nrmse": 0.0678957312592646, "全帯域_10ms_envelope_correlation": 0.9958563212531207, "全帯域_10ms_modulation_mae_db": 0.5120298300222467, "全帯域_40ms_log_rms_mae_db": 0.44456524564596117, "全帯域_40ms_rms_envelope_nrmse": 0.03950598829800083, "全帯域_40ms_envelope_correlation": 0.9987227575029454, "全帯域_40ms_modulation_mae_db": 0.6171345335884889, "300–1000 Hz_10ms_log_rms_mae_db": 1.154567472532968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08156621323171528, "300–1000 Hz_10ms_envelope_correlation": 0.9945865300679269, "300–1000 Hz_10ms_modulation_mae_db": 0.5751180750376211, "300–1000 Hz_40ms_log_rms_mae_db": 0.6676450349859553, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048140432546390645, "300–1000 Hz_40ms_envelope_correlation": 0.9982844704884054, "300–1000 Hz_40ms_modulation_mae_db": 0.6804500771847618, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0915736710981396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09580629626972208, "1000–3000 Hz_10ms_envelope_correlation": 0.9932738656778464, "1000–3000 Hz_10ms_modulation_mae_db": 0.428395313373452, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6196246586124632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053236233305674695, "1000–3000 Hz_40ms_envelope_correlation": 0.9979615681715056, "1000–3000 Hz_40ms_modulation_mae_db": 0.3868433885852029, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2322659500803257, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14123802287470802, "3000–6000 Hz_10ms_envelope_correlation": 0.9827018975789932, "3000–6000 Hz_10ms_modulation_mae_db": 0.22025565352873502, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8303225856999212, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07956358432348144, "3000–6000 Hz_40ms_envelope_correlation": 0.9947682662484678, "3000–6000 Hz_40ms_modulation_mae_db": 0.41355707020064053, "mr_stft_db": 5.410729077018634, "band_mae_db": 0.19694786578869528, "rms_gain_error_db": -0.1579062397223286, "output_peak": 0.407681611464551}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7529865043242674, "全帯域_10ms_rms_envelope_nrmse": 0.07615211207344194, "全帯域_10ms_envelope_correlation": 0.9945350751409465, "全帯域_10ms_modulation_mae_db": 0.29788683310692204, "全帯域_40ms_log_rms_mae_db": 0.3885814088199904, "全帯域_40ms_rms_envelope_nrmse": 0.04094138276287493, "全帯域_40ms_envelope_correlation": 0.9983845989198348, "全帯域_40ms_modulation_mae_db": 0.43996729735946155, "300–1000 Hz_10ms_log_rms_mae_db": 1.124179958497396, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0899162734779622, "300–1000 Hz_10ms_envelope_correlation": 0.9931120472350723, "300–1000 Hz_10ms_modulation_mae_db": 0.3180586616082061, "300–1000 Hz_40ms_log_rms_mae_db": 0.558513242621436, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0493411023698581, "300–1000 Hz_40ms_envelope_correlation": 0.9978984525013953, "300–1000 Hz_40ms_modulation_mae_db": 0.4664534346324362, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0949341135270014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10244327352050354, "1000–3000 Hz_10ms_envelope_correlation": 0.9922638951453908, "1000–3000 Hz_10ms_modulation_mae_db": 0.24174081997951605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6168677958348355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05517188609538624, "1000–3000 Hz_40ms_envelope_correlation": 0.9977308227471325, "1000–3000 Hz_40ms_modulation_mae_db": 0.16404494163531966, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1814839741835323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14881825914483016, "3000–6000 Hz_10ms_envelope_correlation": 0.9806234835661214, "3000–6000 Hz_10ms_modulation_mae_db": 0.09995094071779781, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8303479009670396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0837367058162224, "3000–6000 Hz_40ms_envelope_correlation": 0.9934185257870786, "3000–6000 Hz_40ms_modulation_mae_db": 0.3868490529646261, "mr_stft_db": 5.400997760078755, "band_mae_db": 0.16144896544357223, "rms_gain_error_db": -0.08693876300851963, "output_peak": 0.39401690880267054}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.234927779977039, "spectral_convergence": 0.21065897706173042, "spectral_convergence_status": "ok", "amplitude_floor": 9.509362929412646e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.183567016305108, "spectral_convergence": 0.21017831157601408, "spectral_convergence_status": "ok", "amplitude_floor": 8.503880220761315e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.808424126861781, "spectral_convergence": 0.22074537663552674, "spectral_convergence_status": "ok", "amplitude_floor": 5.089992363719204e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.8030381557821282e-05, "output_power": 2.9611399865491262e-05, "delta_db": 0.23829929404061423, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003997973981863815, "output_power": 0.00040482250367307086, "delta_db": 0.054246830167705665, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008734907128440196, "output_power": 0.0008349972508573941, "delta_db": -0.19573246207751094, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001464675075572333, "output_power": 0.00014938451715186023, "delta_db": 0.08564296462763978, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.350074484130352e-05, "output_power": 5.99986759161436e-05, "delta_db": -0.24637153203346357, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.7271042012588545e-05, "output_power": 1.650683922818962e-05, "delta_db": -0.19654619211984142, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16947321251112926, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09547375114254614, "pitch": null, "output_rms_dbfs": -28.252655132752093, "output_peak": 0.374200027555859, "reference_rms_dbfs": -28.157181381609547, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.408972974381309}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012908667094909063, "log_rms_mae_db": 0.7689959258980067, "rms_envelope_nrmse": 0.07564324501332416, "envelope_correlation": 0.9946232167172238, "rms_gain_error_db": -0.09433013318652161, "reference_mean_power": 0.001527852496322364, "output_mean_power": 0.0014950248614168638}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.179575522794606e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002748180655831064, "output_power": 0.00026343962718802105, "reference_active": true, "delta_db": -0.1836417522999215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.543131824921543e-05, "output_power": 5.509656276115122e-05, "reference_active": true, "delta_db": -0.026307010404618362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2340545383743233e-05, "output_power": 1.2217872627422534e-05, "reference_active": true, "delta_db": -0.04338760276606983}], "active_band_count": 3, "active_band_mae_db": 0.0844454551568699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001146348873577052, "log_rms_mae_db": 0.39943869377266555, "rms_envelope_nrmse": 0.03841657989628656, "envelope_correlation": 0.9986049100880423, "rms_gain_error_db": -0.09367468368619658, "reference_mean_power": 0.0015234711554082644, "output_mean_power": 0.0014909626618733937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.718780555754886e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002667517105928089, "output_power": 0.0002564813286850542, "reference_active": true, "delta_db": -0.1705145824550501}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.819324290787445e-05, "output_power": 2.8139766503281036e-05, "reference_active": true, "delta_db": -0.008245437141026497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.654413380387e-07, "output_power": 6.914212557109448e-07, "reference_active": true, "delta_db": -0.4416918629361635}], "active_band_count": 3, "active_band_mae_db": 0.20681729417741337}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001269978635547725, "log_rms_mae_db": 1.1560192628118977, "rms_envelope_nrmse": 0.09227551577796038, "envelope_correlation": 0.9927711545258427, "rms_gain_error_db": -0.13172371100510313, "reference_mean_power": 0.0011723709196259178, "output_mean_power": 0.0011373461615133763}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.280903308109198e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025355561575728737, "output_power": 0.00023996987555006973, "reference_active": true, "delta_db": -0.23916507477170135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.608107242392719e-05, "output_power": 5.421576850319472e-05, "reference_active": true, "delta_db": -0.14690691704964162}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1840508111446508e-05, "output_power": 1.169252072073586e-05, "reference_active": true, "delta_db": -0.054621914456254916}], "active_band_count": 3, "active_band_mae_db": 0.1468979687591993}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.809451727166842e-05, "log_rms_mae_db": 0.5981594671205527, "rms_envelope_nrmse": 0.048030104587402964, "envelope_correlation": 0.9980565228665024, "rms_gain_error_db": -0.13122337850752622, "reference_mean_power": 0.0011689931508753605, "output_mean_power": 0.0011341999630185712}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.891072559770159e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025025709822561247, "output_power": 0.00023795109001347044, "reference_active": true, "delta_db": -0.2189870597046556}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.836247458716607e-05, "output_power": 2.759787728174536e-05, "reference_active": true, "delta_db": -0.1186844060783234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.847292373572026e-07, "output_power": 6.817692996501262e-07, "reference_active": true, "delta_db": -0.6108239293221915}], "active_band_count": 3, "active_band_mae_db": 0.3161651317017235}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.9423530315000806e-05, "log_rms_mae_db": 1.0645515151638347, "rms_envelope_nrmse": 0.10623439089833282, "envelope_correlation": 0.9916386093309704, "rms_gain_error_db": -0.03496377074709493, "reference_mean_power": 0.00015673633956673046, "output_mean_power": 0.00015547956715983833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.197939100277272e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3248792094830553e-05, "output_power": 3.165150584036875e-05, "reference_active": true, "delta_db": -0.21381495634460063}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.648467501564817e-06, "output_power": 4.0447779647747776e-06, "reference_active": true, "delta_db": 0.4478420415200885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1792733873632069e-06, "output_power": 1.5947662042153117e-06, "reference_active": true, "delta_db": 1.3108252571932701}], "active_band_count": 3, "active_band_mae_db": 0.6574940850193197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.3506304987563625e-05, "log_rms_mae_db": 0.6079536957351119, "rms_envelope_nrmse": 0.06296667527723879, "envelope_correlation": 0.9969632214887244, "rms_gain_error_db": -0.03474577715473496, "reference_mean_power": 0.0001561782704543471, "output_mean_power": 0.00015493374954651497}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.897719153255603e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.156452210778437e-05, "output_power": 2.9991578207659105e-05, "reference_active": true, "delta_db": -0.2219989874878079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6348481818884734e-06, "output_power": 2.042722416113418e-06, "reference_active": true, "delta_db": 0.9673192605795096}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.181225496757285e-08, "output_power": 9.409565094133954e-08, "reference_active": true, "delta_db": 1.1737098682791554}], "active_band_count": 3, "active_band_mae_db": 0.7876760387821576}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.766512572532955e-05, "log_rms_mae_db": 1.2060713758598283, "rms_envelope_nrmse": 0.14321261756863238, "envelope_correlation": 0.9819225418956665, "rms_gain_error_db": -0.1302636524739016, "reference_mean_power": 2.7841403311343547e-05, "output_mean_power": 2.7018719107475523e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1058323154967027e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.079425705940839e-06, "output_power": 4.953104994747341e-06, "reference_active": true, "delta_db": -0.10937078360598351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8241793691525221e-06, "output_power": 2.050100589891853e-06, "reference_active": true, "delta_db": 0.5070763090288796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.319091479762654e-07, "output_power": 5.172337393810197e-07, "reference_active": true, "delta_db": -0.12150613197178159}], "active_band_count": 3, "active_band_mae_db": 0.2459844082022149}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5244936872690136e-05, "log_rms_mae_db": 0.8060415519351711, "rms_envelope_nrmse": 0.08447992910404925, "envelope_correlation": 0.993288103321392, "rms_gain_error_db": -0.13045382056781085, "reference_mean_power": 2.7868742577273385e-05, "output_mean_power": 2.7044066300149724e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.800642728490222e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.984173950209188e-06, "output_power": 4.9043084511612175e-06, "reference_active": true, "delta_db": -0.07015414164418787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0071402238290061e-06, "output_power": 1.1007396613877798e-06, "reference_active": true, "delta_db": 0.3859467362049435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932914966264332e-08, "output_power": 3.1208269020139175e-08, "reference_active": true, "delta_db": -1.0044487560431512}], "active_band_count": 3, "active_band_mae_db": 0.4868498779640942}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8362770799627172, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "91b030e9c78f1b16d0a3e9b8c49a62c60e4068f50ab73758c3697b6b05cb7b69", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9928116421530334, "standard_deviation": 0.0005828297624517042, "minimum": 0.9916386093309704, "maximum": 0.9933504042574026, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0706884714359797, "standard_deviation": 0.023004476465257488, "minimum": 1.0327408308075396, "maximum": 1.0949341135270014, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.34422741394385326, "standard_deviation": 0.15270650357410714, "minimum": 0.16302933681717188, "maximum": 0.6574940850193197, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09951226026382348, "standard_deviation": 0.0035506726043133398, "minimum": 0.09580629626972208, "maximum": 0.10623439089833282, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9978346700461602, "standard_deviation": 0.0004140010772871061, "minimum": 0.9969632214887244, "maximum": 0.9982575536970625, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6023121640544791, "standard_deviation": 0.02451639835882931, "minimum": 0.5598737029717756, "maximum": 0.6301662093509607, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3673396469037815, "standard_deviation": 0.1850706625356828, "minimum": 0.16404494163531966, "maximum": 0.7876760387821576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05583881534277331, "standard_deviation": 0.003623631090015897, "minimum": 0.05083170745950652, "maximum": 0.06296667527723879, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9827742580612838, "standard_deviation": 0.001465774989392091, "minimum": 0.9806234835661214, "maximum": 0.984926676310385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.181112494985951, "standard_deviation": 0.026742873932061845, "minimum": 1.1468954159959102, "maximum": 1.2322659500803257, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3687678828833143, "standard_deviation": 0.24403696506181588, "minimum": 0.09995094071779781, "maximum": 0.8694623805864907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14126978826716688, "standard_deviation": 0.005608365110783167, "minimum": 0.13090794347591622, "maximum": 0.14881825914483016, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9936872092614605, "standard_deviation": 0.0010599895278428338, "minimum": 0.9917014697325459, "maximum": 0.9947842194626458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8204374495423069, "standard_deviation": 0.014299052271930785, "minimum": 0.8015851342291926, "maximum": 0.8425052970098836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5401906238218304, "standard_deviation": 0.36599246905426686, "minimum": 0.09009545905495085, "maximum": 1.17397554584881, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08584764398353201, "standard_deviation": 0.005313742031263157, "minimum": 0.07956358432348144, "maximum": 0.09405207894112398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931429030405166, "standard_deviation": 0.0011095825169775038, "minimum": 0.9909851979323949, "maximum": 0.9945865300679269, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1266859105206877, "standard_deviation": 0.03444284687499229, "minimum": 1.0478750077679504, "maximum": 1.1560192628118977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2643296351793848, "standard_deviation": 0.18387411404125384, "minimum": 0.11044711296672056, "maximum": 0.5751180750376211, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08970230768565952, "standard_deviation": 0.006687871670237559, "minimum": 0.08156621323171528, "maximum": 0.10227660892758639, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978697573534825, "standard_deviation": 0.000496083127535543, "minimum": 0.9967629413270933, "maximum": 0.998361490484365, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6106897476274289, "standard_deviation": 0.039866484652239624, "minimum": 0.558513242621436, "maximum": 0.6676450349859553, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.35454323990606257, "standard_deviation": 0.19054371363011105, "minimum": 0.13198896230351764, "maximum": 0.6804500771847618, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04972388036465796, "standard_deviation": 0.004659500810814356, "minimum": 0.04267907881938801, "maximum": 0.059311512935237745, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20551012933786242, "standard_deviation": 0.04796382758169193, "minimum": 0.16089916846560812, "maximum": 0.2820266220462053, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.4059956910564395, "standard_deviation": 0.007982322587998403, "minimum": 5.393424295740394, "maximum": 5.4171645649692275, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.38980150372686073, "standard_deviation": 0.013014083900533225, "minimum": 0.3691420806231504, "maximum": 0.407681611464551, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10077136241035858, "standard_deviation": 0.03256483449590381, "minimum": -0.1579062397223286, "maximum": -0.04872281632493935, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947644516651031, "standard_deviation": 0.0009168120493314473, "minimum": 0.9928793479684446, "maximum": 0.9958563212531207, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7518085819943892, "standard_deviation": 0.020232672428842514, "minimum": 0.7232751018858052, "maximum": 0.7741963214461444, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.23996648037378762, "standard_deviation": 0.15757564054097253, "minimum": 0.0844454551568699, "maximum": 0.5120298300222467, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07461180691816938, "standard_deviation": 0.005950652960655464, "minimum": 0.0678957312592646, "maximum": 0.08652139290653128, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983861543382284, "standard_deviation": 0.00044545553323451073, "minimum": 0.9973348703442194, "maximum": 0.9987227575029454, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40877107075002883, "standard_deviation": 0.025189953520404634, "minimum": 0.3773099562037902, "maximum": 0.44456524564596117, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3064892302821082, "standard_deviation": 0.1853243740364335, "minimum": 0.07839909987700332, "maximum": 0.6171345335884889, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041206329031846026, "standard_deviation": 0.004430212305396787, "minimum": 0.03652237557683984, "maximum": 0.051412420039639965, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0010", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.5465826988220215, "predicted_mos": 2.5465826988220215, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b52af4cd48fdf5be5a751c427ff42449482b74a8de4a91a5ca2706d75fc49f71", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.237082100007683, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7206722010046341, "全帯域_10ms_rms_envelope_nrmse": 0.07453836877814486, "全帯域_10ms_envelope_correlation": 0.9951211912063921, "全帯域_10ms_modulation_mae_db": 0.8644087743267614, "全帯域_40ms_log_rms_mae_db": 0.4018799562062851, "全帯域_40ms_rms_envelope_nrmse": 0.04417247593347058, "全帯域_40ms_envelope_correlation": 0.9985185225778929, "全帯域_40ms_modulation_mae_db": 0.8722948822800136, "300–1000 Hz_10ms_log_rms_mae_db": 1.0838151049565894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08950655771241275, "300–1000 Hz_10ms_envelope_correlation": 0.9935244344350774, "300–1000 Hz_10ms_modulation_mae_db": 0.9594985449145478, "300–1000 Hz_40ms_log_rms_mae_db": 0.5909200377667753, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053335858428464916, "300–1000 Hz_40ms_envelope_correlation": 0.9979365588816433, "300–1000 Hz_40ms_modulation_mae_db": 0.9438321911261943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1093298371328528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10446798570054472, "1000–3000 Hz_10ms_envelope_correlation": 0.9924082747559347, "1000–3000 Hz_10ms_modulation_mae_db": 0.6190623138278156, "1000–3000 Hz_40ms_log_rms_mae_db": 0.596247340538068, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058629368156126814, "1000–3000 Hz_40ms_envelope_correlation": 0.9980697882279048, "1000–3000 Hz_40ms_modulation_mae_db": 0.6066022939716685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.252336684833084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.152147197804262, "3000–6000 Hz_10ms_envelope_correlation": 0.9798553439545821, "3000–6000 Hz_10ms_modulation_mae_db": 0.5323776074005425, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8158528910083909, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09141134636552023, "3000–6000 Hz_40ms_envelope_correlation": 0.9929432269606607, "3000–6000 Hz_40ms_modulation_mae_db": 0.8774555752562435, "mr_stft_db": 5.47038551441788, "band_mae_db": 0.23448938344584583, "rms_gain_error_db": -0.20395715463416192, "output_peak": 0.3924473128201879}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7688099616341636, "全帯域_10ms_rms_envelope_nrmse": 0.07109172390361751, "全帯域_10ms_envelope_correlation": 0.9952699001912753, "全帯域_10ms_modulation_mae_db": 0.5988476129638155, "全帯域_40ms_log_rms_mae_db": 0.41200505749216765, "全帯域_40ms_rms_envelope_nrmse": 0.03923985163869229, "全帯域_40ms_envelope_correlation": 0.9985432170944389, "全帯域_40ms_modulation_mae_db": 0.767872629849352, "300–1000 Hz_10ms_log_rms_mae_db": 1.1860471592762403, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08622858473863261, "300–1000 Hz_10ms_envelope_correlation": 0.993664176273813, "300–1000 Hz_10ms_modulation_mae_db": 0.7713286637082443, "300–1000 Hz_40ms_log_rms_mae_db": 0.6258761669512334, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047410471124120716, "300–1000 Hz_40ms_envelope_correlation": 0.9980473971070362, "300–1000 Hz_40ms_modulation_mae_db": 0.9989469615344454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.156881328361635, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1004050869387011, "1000–3000 Hz_10ms_envelope_correlation": 0.9926517339169868, "1000–3000 Hz_10ms_modulation_mae_db": 0.4896620308231192, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6432308172441547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057560578863866306, "1000–3000 Hz_40ms_envelope_correlation": 0.9976684712355693, "1000–3000 Hz_40ms_modulation_mae_db": 0.4903016557080741, "3000–6000 Hz_10ms_log_rms_mae_db": 1.248846989941106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.140067520818702, "3000–6000 Hz_10ms_envelope_correlation": 0.9829344264996133, "3000–6000 Hz_10ms_modulation_mae_db": 0.5012271930501975, "3000–6000 Hz_40ms_log_rms_mae_db": 0.846917620084612, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0838197799112209, "3000–6000 Hz_40ms_envelope_correlation": 0.9939769001675328, "3000–6000 Hz_40ms_modulation_mae_db": 0.8196851139773272, "mr_stft_db": 5.4665141376854605, "band_mae_db": 0.19495089170378424, "rms_gain_error_db": -0.09400338253828763, "output_peak": 0.3786677115220728}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7300869975252742, "全帯域_10ms_rms_envelope_nrmse": 0.06889330717418594, "全帯域_10ms_envelope_correlation": 0.9958532417282202, "全帯域_10ms_modulation_mae_db": 0.6776862951313706, "全帯域_40ms_log_rms_mae_db": 0.40148553069927617, "全帯域_40ms_rms_envelope_nrmse": 0.041431737205182805, "全帯域_40ms_envelope_correlation": 0.9987073439395387, "全帯域_40ms_modulation_mae_db": 0.7186787327596408, "300–1000 Hz_10ms_log_rms_mae_db": 1.118103424425608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08376893093367803, "300–1000 Hz_10ms_envelope_correlation": 0.9944046223884421, "300–1000 Hz_10ms_modulation_mae_db": 0.8190389004444336, "300–1000 Hz_40ms_log_rms_mae_db": 0.6163650088141518, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05133908408354425, "300–1000 Hz_40ms_envelope_correlation": 0.9981454739071328, "300–1000 Hz_40ms_modulation_mae_db": 0.9690101577418954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1439675703870515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10229203998627226, "1000–3000 Hz_10ms_envelope_correlation": 0.9923581439171423, "1000–3000 Hz_10ms_modulation_mae_db": 0.4306325381020278, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6475037434265511, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05739572754652895, "1000–3000 Hz_40ms_envelope_correlation": 0.9976806461339711, "1000–3000 Hz_40ms_modulation_mae_db": 0.499384975796523, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3022752432890745, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1520568570340752, "3000–6000 Hz_10ms_envelope_correlation": 0.9801727471942037, "3000–6000 Hz_10ms_modulation_mae_db": 0.7632330672676396, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8858564159161824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09121667074206156, "3000–6000 Hz_40ms_envelope_correlation": 0.9933810029530186, "3000–6000 Hz_40ms_modulation_mae_db": 1.2468090318505114, "mr_stft_db": 5.475980181912173, "band_mae_db": 0.22164928959861221, "rms_gain_error_db": -0.19146532310229358, "output_peak": 0.3896902579106566}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7450175323049604, "全帯域_10ms_rms_envelope_nrmse": 0.07414542785271357, "全帯域_10ms_envelope_correlation": 0.9947941081672963, "全帯域_10ms_modulation_mae_db": 0.2011804772017709, "全帯域_40ms_log_rms_mae_db": 0.41354478739119266, "全帯域_40ms_rms_envelope_nrmse": 0.04437264345762215, "全帯域_40ms_envelope_correlation": 0.9980159335131236, "全帯域_40ms_modulation_mae_db": 0.08431509493199725, "300–1000 Hz_10ms_log_rms_mae_db": 1.177856255888999, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0885127247396403, "300–1000 Hz_10ms_envelope_correlation": 0.9932747855752255, "300–1000 Hz_10ms_modulation_mae_db": 0.25069661864361764, "300–1000 Hz_40ms_log_rms_mae_db": 0.6344366673318008, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05320580289615998, "300–1000 Hz_40ms_envelope_correlation": 0.9974055683477105, "300–1000 Hz_40ms_modulation_mae_db": 0.10660968878144404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1230835542983804, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09939010397259497, "1000–3000 Hz_10ms_envelope_correlation": 0.9926978477559661, "1000–3000 Hz_10ms_modulation_mae_db": 0.3561624437860343, "1000–3000 Hz_40ms_log_rms_mae_db": 0.637605251339093, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0574501847976319, "1000–3000 Hz_40ms_envelope_correlation": 0.9975339699159005, "1000–3000 Hz_40ms_modulation_mae_db": 0.29505022905723327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2489736048794755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13679282407133078, "3000–6000 Hz_10ms_envelope_correlation": 0.983752907630061, "3000–6000 Hz_10ms_modulation_mae_db": 0.6149524769370399, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8165388038539747, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07688210773326803, "3000–6000 Hz_40ms_envelope_correlation": 0.9950983042880764, "3000–6000 Hz_40ms_modulation_mae_db": 1.0414356076979756, "mr_stft_db": 5.459836911113965, "band_mae_db": 0.14736054807054802, "rms_gain_error_db": -0.02345314152561916, "output_peak": 0.39302091208112716}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7769077748124754, "全帯域_10ms_rms_envelope_nrmse": 0.07116185362846787, "全帯域_10ms_envelope_correlation": 0.9952824419886911, "全帯域_10ms_modulation_mae_db": 0.4970285757450632, "全帯域_40ms_log_rms_mae_db": 0.4239224706893556, "全帯域_40ms_rms_envelope_nrmse": 0.04086266265741644, "全帯域_40ms_envelope_correlation": 0.9984434356463944, "全帯域_40ms_modulation_mae_db": 0.6702259362301378, "300–1000 Hz_10ms_log_rms_mae_db": 1.1814467089977352, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08720207962894759, "300–1000 Hz_10ms_envelope_correlation": 0.993546213368504, "300–1000 Hz_10ms_modulation_mae_db": 0.6811632485900834, "300–1000 Hz_40ms_log_rms_mae_db": 0.6371372932250492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05050550561876212, "300–1000 Hz_40ms_envelope_correlation": 0.997806152476541, "300–1000 Hz_40ms_modulation_mae_db": 0.9191367880036728, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0653465846755341, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09310504347160448, "1000–3000 Hz_10ms_envelope_correlation": 0.993762895647293, "1000–3000 Hz_10ms_modulation_mae_db": 0.3625897506243043, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5875763275343087, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053959775530172616, "1000–3000 Hz_40ms_envelope_correlation": 0.9980282298078366, "1000–3000 Hz_40ms_modulation_mae_db": 0.40498674682932045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2763665537224225, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15159367885686365, "3000–6000 Hz_10ms_envelope_correlation": 0.9797008056510174, "3000–6000 Hz_10ms_modulation_mae_db": 0.29188275315437656, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8581488017769744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08960028553747726, "3000–6000 Hz_40ms_envelope_correlation": 0.9927509920331665, "3000–6000 Hz_40ms_modulation_mae_db": 0.5091029844099888, "mr_stft_db": 5.466194639660863, "band_mae_db": 0.2159349907546555, "rms_gain_error_db": -0.10420137954145858, "output_peak": 0.37880968008325433}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7931049081996451, "全帯域_10ms_rms_envelope_nrmse": 0.07171998508650515, "全帯域_10ms_envelope_correlation": 0.9951273821936542, "全帯域_10ms_modulation_mae_db": 0.360501530208853, "全帯域_40ms_log_rms_mae_db": 0.42879830049429013, "全帯域_40ms_rms_envelope_nrmse": 0.0390289610440064, "全帯域_40ms_envelope_correlation": 0.9984885936676557, "全帯域_40ms_modulation_mae_db": 0.5873156751027734, "300–1000 Hz_10ms_log_rms_mae_db": 1.1688201807270842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08512802551447483, "300–1000 Hz_10ms_envelope_correlation": 0.9937702637410143, "300–1000 Hz_10ms_modulation_mae_db": 0.3993147449306825, "300–1000 Hz_40ms_log_rms_mae_db": 0.6290397346273721, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04850316745799837, "300–1000 Hz_40ms_envelope_correlation": 0.9978663827088269, "300–1000 Hz_40ms_modulation_mae_db": 0.6082829938773288, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1212246828571166, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10898781660393202, "1000–3000 Hz_10ms_envelope_correlation": 0.9912379157099463, "1000–3000 Hz_10ms_modulation_mae_db": 0.38475304026133256, "1000–3000 Hz_40ms_log_rms_mae_db": 0.647535917442693, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05928974799006322, "1000–3000 Hz_40ms_envelope_correlation": 0.997415641086874, "1000–3000 Hz_40ms_modulation_mae_db": 0.47755987913885506, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2545524653921645, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15328334933681625, "3000–6000 Hz_10ms_envelope_correlation": 0.9794117788715131, "3000–6000 Hz_10ms_modulation_mae_db": 0.3536218746391609, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8225198220410572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0936609657093279, "3000–6000 Hz_40ms_envelope_correlation": 0.9923688973086265, "3000–6000 Hz_40ms_modulation_mae_db": 0.053508633057036285, "mr_stft_db": 5.4696605237399085, "band_mae_db": 0.16106001739541462, "rms_gain_error_db": -0.05869586456928719, "output_peak": 0.39574570671984655}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7603837965794974, "全帯域_10ms_rms_envelope_nrmse": 0.06760072585452284, "全帯域_10ms_envelope_correlation": 0.9957122488904788, "全帯域_10ms_modulation_mae_db": 0.31469911184872895, "全帯域_40ms_log_rms_mae_db": 0.41660550605997626, "全帯域_40ms_rms_envelope_nrmse": 0.03642934274681886, "全帯域_40ms_envelope_correlation": 0.9987394864932553, "全帯域_40ms_modulation_mae_db": 0.20347636441490333, "300–1000 Hz_10ms_log_rms_mae_db": 1.1381940664536814, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08320253113729943, "300–1000 Hz_10ms_envelope_correlation": 0.9940642975295362, "300–1000 Hz_10ms_modulation_mae_db": 0.41277671577119396, "300–1000 Hz_40ms_log_rms_mae_db": 0.5844389249597705, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044653231990008584, "300–1000 Hz_40ms_envelope_correlation": 0.9982200343175253, "300–1000 Hz_40ms_modulation_mae_db": 0.2817169686171737, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1028359229321203, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10029216262958383, "1000–3000 Hz_10ms_envelope_correlation": 0.993025197242227, "1000–3000 Hz_10ms_modulation_mae_db": 0.6909922987732021, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6001403121346744, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06216885990140858, "1000–3000 Hz_40ms_envelope_correlation": 0.9976696312057202, "1000–3000 Hz_40ms_modulation_mae_db": 0.5410109705923016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2670863182532213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14967043392475596, "3000–6000 Hz_10ms_envelope_correlation": 0.9802553821842355, "3000–6000 Hz_10ms_modulation_mae_db": 0.21112111015284404, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8381211986745334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09017718199489283, "3000–6000 Hz_40ms_envelope_correlation": 0.992314053932017, "3000–6000 Hz_40ms_modulation_mae_db": 0.2477355811396308, "mr_stft_db": 5.490247603917219, "band_mae_db": 0.18985795241400535, "rms_gain_error_db": -0.08702334053271503, "output_peak": 0.39223885938898173}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7601357960402618, "全帯域_10ms_rms_envelope_nrmse": 0.08292073860022518, "全帯域_10ms_envelope_correlation": 0.993636528694616, "全帯域_10ms_modulation_mae_db": 0.13708341782691896, "全帯域_40ms_log_rms_mae_db": 0.41720065591668737, "全帯域_40ms_rms_envelope_nrmse": 0.049980794761547695, "全帯域_40ms_envelope_correlation": 0.9977133293401973, "全帯域_40ms_modulation_mae_db": 0.2565445551494892, "300–1000 Hz_10ms_log_rms_mae_db": 1.1486371959998498, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09888273733276878, "300–1000 Hz_10ms_envelope_correlation": 0.9917744167584114, "300–1000 Hz_10ms_modulation_mae_db": 0.17561409630401548, "300–1000 Hz_40ms_log_rms_mae_db": 0.6010089964666597, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06016919279822012, "300–1000 Hz_40ms_envelope_correlation": 0.9969678194352916, "300–1000 Hz_40ms_modulation_mae_db": 0.3629864340751862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1491122642523786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10961109332815618, "1000–3000 Hz_10ms_envelope_correlation": 0.9912047155923461, "1000–3000 Hz_10ms_modulation_mae_db": 0.4468501470680726, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6517277221434039, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06627322605520734, "1000–3000 Hz_40ms_envelope_correlation": 0.996842377672944, "1000–3000 Hz_40ms_modulation_mae_db": 0.37377255736821074, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2918211863033036, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14430055473919898, "3000–6000 Hz_10ms_envelope_correlation": 0.9818550393177983, "3000–6000 Hz_10ms_modulation_mae_db": 0.610446391761948, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8714866534777056, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08637592716121262, "3000–6000 Hz_40ms_envelope_correlation": 0.9935393990783011, "3000–6000 Hz_40ms_modulation_mae_db": 0.9551112117053577, "mr_stft_db": 5.456787103328032, "band_mae_db": 0.18822346913089816, "rms_gain_error_db": -0.14594745443589535, "output_peak": 0.38319421144009125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.336906213482562, "spectral_convergence": 0.20756144570653048, "spectral_convergence_status": "ok", "amplitude_floor": 9.509362929412646e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.259115971983505, "spectral_convergence": 0.21047000984989386, "spectral_convergence_status": "ok", "amplitude_floor": 8.503880220761315e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8151343577875725, "spectral_convergence": 0.22163003823016786, "spectral_convergence_status": "ok", "amplitude_floor": 5.089992363719204e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.8030381557821282e-05, "output_power": 2.966471142975672e-05, "delta_db": 0.24611118450334518, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003997973981863815, "output_power": 0.0003982016775885385, "delta_db": -0.017368781369259845, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008734907128440196, "output_power": 0.0008172176418608782, "delta_db": -0.2892055833701807, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001464675075572333, "output_power": 0.0001378644768189421, "delta_db": -0.2628891415662801, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.350074484130352e-05, "output_power": 5.935940541993597e-05, "delta_db": -0.29289277308757194, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.7271042012588545e-05, "output_power": 1.612395795230664e-05, "delta_db": -0.29846883677843705, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23448938344584583, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.20395715463416192, "pitch": null, "output_rms_dbfs": -28.361138536243708, "output_peak": 0.3924473128201879, "reference_rms_dbfs": -28.157181381609547, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.47038551441788}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012908667094909063, "log_rms_mae_db": 0.7206722010046341, "rms_envelope_nrmse": 0.07453836877814486, "envelope_correlation": 0.9951211912063921, "rms_gain_error_db": -0.2017557903219786, "reference_mean_power": 0.001527852496322364, "output_mean_power": 0.0014584980374677799}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.179575522794606e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002748180655831064, "output_power": 0.00025671991352280897, "reference_active": true, "delta_db": -0.29585720540020494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.543131824921543e-05, "output_power": 4.6435812694581064e-05, "reference_active": true, "delta_db": -0.769021562238663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2340545383743233e-05, "output_power": 8.679597738693336e-06, "reference_active": true, "delta_db": -1.5283475553414163}], "active_band_count": 3, "active_band_mae_db": 0.8644087743267614}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001146348873577052, "log_rms_mae_db": 0.4018799562062851, "rms_envelope_nrmse": 0.04417247593347058, "envelope_correlation": 0.9985185225778929, "rms_gain_error_db": -0.20031053180371133, "reference_mean_power": 0.0015234711554082644, "output_mean_power": 0.001454799632771412}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.718780555754886e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002667517105928089, "output_power": 0.0002471616591035683, "reference_active": true, "delta_db": -0.33126111287693305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.819324290787445e-05, "output_power": 2.33689490316168e-05, "reference_active": true, "delta_db": -0.8150585179303913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.654413380387e-07, "output_power": 5.455761932615536e-07, "reference_active": true, "delta_db": -1.4705650160327164}], "active_band_count": 3, "active_band_mae_db": 0.8722948822800136}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001269978635547725, "log_rms_mae_db": 1.0838151049565894, "rms_envelope_nrmse": 0.08950655771241275, "envelope_correlation": 0.9935244344350774, "rms_gain_error_db": -0.22521826704443765, "reference_mean_power": 0.0011723709196259178, "output_mean_power": 0.0011131231387544165}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.280903308109198e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025355561575728737, "output_power": 0.00023464663972882585, "reference_active": true, "delta_db": -0.3365889450439863}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.608107242392719e-05, "output_power": 4.765937122934695e-05, "reference_active": true, "delta_db": -0.7066800154375856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1840508111446508e-05, "output_power": 7.759748975167978e-06, "reference_active": true, "delta_db": -1.8352266742620718}], "active_band_count": 3, "active_band_mae_db": 0.9594985449145478}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.809451727166842e-05, "log_rms_mae_db": 0.5909200377667753, "rms_envelope_nrmse": 0.053335858428464916, "envelope_correlation": 0.9979365588816433, "rms_gain_error_db": -0.22336452045743183, "reference_mean_power": 0.0011689931508753605, "output_mean_power": 0.0011103899300846946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.891072559770159e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025025709822561247, "output_power": 0.00022941020891955295, "reference_active": true, "delta_db": -0.3777366412638046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.836247458716607e-05, "output_power": 2.3625393805114127e-05, "reference_active": true, "delta_db": -0.7936506338895806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.847292373572026e-07, "output_power": 5.354376476264991e-07, "reference_active": true, "delta_db": -1.6601092982251977}], "active_band_count": 3, "active_band_mae_db": 0.9438321911261943}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.9423530315000806e-05, "log_rms_mae_db": 1.1093298371328528, "rms_envelope_nrmse": 0.10446798570054472, "envelope_correlation": 0.9924082747559347, "rms_gain_error_db": -0.2720218897792076, "reference_mean_power": 0.00015673633956673046, "output_mean_power": 0.00014722023648672158}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.197939100277272e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3248792094830553e-05, "output_power": 3.0433716556064784e-05, "reference_active": true, "delta_db": -0.3842088080868158}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.648467501564817e-06, "output_power": 2.631208406535668e-06, "reference_active": true, "delta_db": -1.4195523414346056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1792733873632069e-06, "output_power": 1.1938701042065925e-06, "reference_active": true, "delta_db": 0.05342579196202561}], "active_band_count": 3, "active_band_mae_db": 0.6190623138278156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.3506304987563625e-05, "log_rms_mae_db": 0.596247340538068, "rms_envelope_nrmse": 0.058629368156126814, "envelope_correlation": 0.9980697882279048, "rms_gain_error_db": -0.2743121398440659, "reference_mean_power": 0.0001561782704543471, "output_mean_power": 0.0001466187103245972}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.897719153255603e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.156452210778437e-05, "output_power": 2.8836620468857496e-05, "reference_active": true, "delta_db": -0.3925485688574673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6348481818884734e-06, "output_power": 1.1786474364536693e-06, "reference_active": true, "delta_db": -1.4209351270930124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.181225496757285e-08, "output_power": 7.170777474562526e-08, "reference_active": true, "delta_db": -0.006323185964526118}], "active_band_count": 3, "active_band_mae_db": 0.6066022939716685}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.766512572532955e-05, "log_rms_mae_db": 1.252336684833084, "rms_envelope_nrmse": 0.152147197804262, "envelope_correlation": 0.9798553439545821, "rms_gain_error_db": -0.29779467159922546, "reference_mean_power": 2.7841403311343547e-05, "output_mean_power": 2.599630684045688e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1058323154967027e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.079425705940839e-06, "output_power": 4.715725309764177e-06, "reference_active": true, "delta_db": -0.32266112935711255}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8241793691525221e-06, "output_power": 1.5957751638407993e-06, "reference_active": true, "delta_db": -0.5809583813343341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.319091479762654e-07, "output_power": 4.534048155490058e-07, "reference_active": true, "delta_db": -0.693513311510181}], "active_band_count": 3, "active_band_mae_db": 0.5323776074005425}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5244936872690136e-05, "log_rms_mae_db": 0.8158528910083909, "rms_envelope_nrmse": 0.09141134636552023, "envelope_correlation": 0.9929432269606607, "rms_gain_error_db": -0.29789241827928853, "reference_mean_power": 2.7868742577273385e-05, "output_mean_power": 2.6021248620442458e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.800642728490222e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.984173950209188e-06, "output_power": 4.5026658347534515e-06, "reference_active": true, "delta_db": -0.44123474010803954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0071402238290061e-06, "output_power": 8.426539074603023e-07, "reference_active": true, "delta_db": -0.7744070246589634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932914966264332e-08, "output_power": 2.838193888259399e-08, "reference_active": true, "delta_db": -1.4167249610017276}], "active_band_count": 3, "active_band_mae_db": 0.8774555752562435}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8270297016337385, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c8e5605fb00a3169ca877f83b4571d71a2968247df76bf25a2481551f3b42b56", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9924183405672303, "standard_deviation": 0.0008601256961440887, "minimum": 0.9912047155923461, "maximum": 0.993762895647293, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1214727181121336, "standard_deviation": 0.029692017762880746, "minimum": 1.0653465846755341, "maximum": 1.156881328361635, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4725880704082386, "standard_deviation": 0.12261373022441117, "minimum": 0.3561624437860343, "maximum": 0.6909922987732021, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1023189165789237, "standard_deviation": 0.005391967121959824, "minimum": 0.09310504347160448, "maximum": 0.10961109332815618, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99761359441084, "standard_deviation": 0.00038377517013644507, "minimum": 0.996842377672944, "maximum": 0.9980697882279048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6264459289753684, "standard_deviation": 0.02685526635178843, "minimum": 0.5875763275343087, "maximum": 0.6517277221434039, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4610836635577733, "standard_deviation": 0.0989179950220212, "minimum": 0.29505022905723327, "maximum": 0.6066022939716685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05909093360512572, "standard_deviation": 0.0036945534043249565, "minimum": 0.053959775530172616, "maximum": 0.06627322605520734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9809923039128781, "standard_deviation": 0.0016392732434546027, "minimum": 0.9794117788715131, "maximum": 0.983752907630061, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2677823808267314, "standard_deviation": 0.020590266510297033, "minimum": 1.248846989941106, "maximum": 1.3022752432890745, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4848578092954686, "standard_deviation": 0.18597619282259603, "minimum": 0.21112111015284404, "maximum": 0.7632330672676396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1474890520732506, "standard_deviation": 0.006295104302839721, "minimum": 0.13679282407133078, "maximum": 0.15328334933681625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993296597090175, "standard_deviation": 0.0009276399800528429, "minimum": 0.992314053932017, "maximum": 0.9950983042880764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8444302758541788, "standard_deviation": 0.02606867595266888, "minimum": 0.8158528910083909, "maximum": 0.8858564159161824, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7188554673867589, "standard_deviation": 0.41102081334326657, "minimum": 0.053508633057036285, "maximum": 1.2468090318505114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08789303314437266, "standard_deviation": 0.005411095326882521, "minimum": 0.07688210773326803, "maximum": 0.0936609657093279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935029012587531, "standard_deviation": 0.0007808232052965126, "minimum": 0.9917744167584114, "maximum": 0.9944046223884421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1503650120907234, "standard_deviation": 0.03577003272840962, "minimum": 1.0838151049565894, "maximum": 1.1860471592762403, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5586789416633523, "standard_deviation": 0.2870893753269588, "minimum": 0.17561409630401548, "maximum": 0.9594985449145478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08780402146723179, "standard_deviation": 0.004979956874395637, "minimum": 0.08320253113729943, "maximum": 0.09888273733276878, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977994233977134, "standard_deviation": 0.00041887219035635734, "minimum": 0.9969678194352916, "maximum": 0.9982200343175253, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6149028537678516, "standard_deviation": 0.02033508144925341, "minimum": 0.5844389249597705, "maximum": 0.6371372932250492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6488152729696676, "standard_deviation": 0.3580332623087121, "minimum": 0.10660968878144404, "maximum": 0.9989469615344454, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.051140289299659886, "standard_deviation": 0.004690389963374754, "minimum": 0.044653231990008584, "maximum": 0.06016919279822012, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19419081781422048, "standard_deviation": 0.02975239704010033, "minimum": 0.14736054807054802, "maximum": 0.23448938344584583, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.469450826971938, "standard_deviation": 0.010337726988539183, "minimum": 5.456787103328032, "maximum": 5.490247603917219, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3879768314957773, "standard_deviation": 0.006766027750188222, "minimum": 0.3786677115220728, "maximum": 0.39574570671984655, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11359338010996481, "standard_deviation": 0.06278464539815418, "minimum": -0.20395715463416192, "maximum": -0.02345314152561916, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.995099630382578, "standard_deviation": 0.0006804293595198374, "minimum": 0.993636528694616, "maximum": 0.9958532417282202, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.756889871012614, "standard_deviation": 0.024047998345425328, "minimum": 0.7206722010046341, "maximum": 0.7931049081996451, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45642947440666026, "standard_deviation": 0.24892813755781767, "minimum": 0.13708341782691896, "maximum": 0.8644087743267614, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07275901635979787, "standard_deviation": 0.004725476213404495, "minimum": 0.06760072585452284, "maximum": 0.08292073860022518, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983962327840621, "standard_deviation": 0.00035313205406524326, "minimum": 0.9977133293401973, "maximum": 0.9987394864932553, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4144302831186539, "standard_deviation": 0.009566991251900166, "minimum": 0.40148553069927617, "maximum": 0.42879830049429013, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5200904838397884, "standard_deviation": 0.29557262822764, "minimum": 0.08431509493199725, "maximum": 0.8722948822800136, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04193980868059466, "standard_deviation": 0.00419365783478352, "minimum": 0.03642934274681886, "maximum": 0.049980794761547695, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0010", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.756009101867676, "predicted_mos": 2.756009101867676, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "34b31e2941c3f2e17cd5daad08e98d18c24384244b9801cc146080df2378db4f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.252629400230944, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7173671236913817, "全帯域_10ms_rms_envelope_nrmse": 0.07326075750504513, "全帯域_10ms_envelope_correlation": 0.9949549167830729, "全帯域_10ms_modulation_mae_db": 0.703363505027621, "全帯域_40ms_log_rms_mae_db": 0.3909250241836874, "全帯域_40ms_rms_envelope_nrmse": 0.040550899560334315, "全帯域_40ms_envelope_correlation": 0.9984219465252709, "全帯域_40ms_modulation_mae_db": 0.7704721798036241, "300–1000 Hz_10ms_log_rms_mae_db": 1.100312466601757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08868914988244629, "300–1000 Hz_10ms_envelope_correlation": 0.993279368346718, "300–1000 Hz_10ms_modulation_mae_db": 0.6698609090344064, "300–1000 Hz_40ms_log_rms_mae_db": 0.5800943480354311, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04988290889057304, "300–1000 Hz_40ms_envelope_correlation": 0.9978109157306161, "300–1000 Hz_40ms_modulation_mae_db": 0.7445282949110202, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1105957665837436, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1012976076644774, "1000–3000 Hz_10ms_envelope_correlation": 0.9925789117292428, "1000–3000 Hz_10ms_modulation_mae_db": 0.44972539456868815, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6023194488766578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06073686216443204, "1000–3000 Hz_40ms_envelope_correlation": 0.9974534298439662, "1000–3000 Hz_40ms_modulation_mae_db": 0.5779468812879925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2294890322204768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13450520975609995, "3000–6000 Hz_10ms_envelope_correlation": 0.9845963340411499, "3000–6000 Hz_10ms_modulation_mae_db": 0.8403235134940403, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8296202727639925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07901442390766819, "3000–6000 Hz_40ms_envelope_correlation": 0.9951362972087844, "3000–6000 Hz_40ms_modulation_mae_db": 1.2556188690623222, "mr_stft_db": 5.457014834192177, "band_mae_db": 0.24678800834403467, "rms_gain_error_db": -0.09148793014587984, "output_peak": 0.37409475901237726}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7257568325631255, "全帯域_10ms_rms_envelope_nrmse": 0.07140199005210872, "全帯域_10ms_envelope_correlation": 0.99526607145882, "全帯域_10ms_modulation_mae_db": 0.3775542261215734, "全帯域_40ms_log_rms_mae_db": 0.38817976017987627, "全帯域_40ms_rms_envelope_nrmse": 0.04356686082958625, "全帯域_40ms_envelope_correlation": 0.998224719963384, "全帯域_40ms_modulation_mae_db": 0.5081571804592464, "300–1000 Hz_10ms_log_rms_mae_db": 1.1003347688827323, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08663636491690528, "300–1000 Hz_10ms_envelope_correlation": 0.9936594215369007, "300–1000 Hz_10ms_modulation_mae_db": 0.42443739159548516, "300–1000 Hz_40ms_log_rms_mae_db": 0.5628926199515605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054143992504084894, "300–1000 Hz_40ms_envelope_correlation": 0.9974782236428296, "300–1000 Hz_40ms_modulation_mae_db": 0.5472411519452448, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1213838487315746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10308993444532923, "1000–3000 Hz_10ms_envelope_correlation": 0.9921635242857764, "1000–3000 Hz_10ms_modulation_mae_db": 0.2829528498624419, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6476832887389323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058172372073965924, "1000–3000 Hz_40ms_envelope_correlation": 0.9975024626158084, "1000–3000 Hz_40ms_modulation_mae_db": 0.2337387434569084, "3000–6000 Hz_10ms_log_rms_mae_db": 1.223254492441555, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13333534266909502, "3000–6000 Hz_10ms_envelope_correlation": 0.9843665140619938, "3000–6000 Hz_10ms_modulation_mae_db": 0.28910001344093567, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8210983963227165, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07757276974400287, "3000–6000 Hz_40ms_envelope_correlation": 0.9946263247415466, "3000–6000 Hz_40ms_modulation_mae_db": 0.5898129243944981, "mr_stft_db": 5.441169499393897, "band_mae_db": 0.15543707541770188, "rms_gain_error_db": -0.11615523903880196, "output_peak": 0.3866191279477625}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7516049113765395, "全帯域_10ms_rms_envelope_nrmse": 0.0775603449171045, "全帯域_10ms_envelope_correlation": 0.9944573800658394, "全帯域_10ms_modulation_mae_db": 0.11046935469172335, "全帯域_40ms_log_rms_mae_db": 0.3922386640593912, "全帯域_40ms_rms_envelope_nrmse": 0.04815525542021655, "全帯域_40ms_envelope_correlation": 0.9978921535061154, "全帯域_40ms_modulation_mae_db": 0.21258547324592714, "300–1000 Hz_10ms_log_rms_mae_db": 1.1003268401575494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0918184189343494, "300–1000 Hz_10ms_envelope_correlation": 0.9929139668540627, "300–1000 Hz_10ms_modulation_mae_db": 0.17321793579789857, "300–1000 Hz_40ms_log_rms_mae_db": 0.5656301183419933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05732020367039904, "300–1000 Hz_40ms_envelope_correlation": 0.9972424815355598, "300–1000 Hz_40ms_modulation_mae_db": 0.2671497102999893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1501835185349212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11491477306246602, "1000–3000 Hz_10ms_envelope_correlation": 0.9903599690119325, "1000–3000 Hz_10ms_modulation_mae_db": 0.687408837953314, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6760699014938776, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06424271495972694, "1000–3000 Hz_40ms_envelope_correlation": 0.9971067372970571, "1000–3000 Hz_40ms_modulation_mae_db": 0.5294715868171221, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2432621849075238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14417486460957205, "3000–6000 Hz_10ms_envelope_correlation": 0.9818975044580263, "3000–6000 Hz_10ms_modulation_mae_db": 0.6162283173921302, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8416877454086633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08157991858080658, "3000–6000 Hz_40ms_envelope_correlation": 0.9943982958705524, "3000–6000 Hz_40ms_modulation_mae_db": 0.9136511730477395, "mr_stft_db": 5.445422006137151, "band_mae_db": 0.21360445448992935, "rms_gain_error_db": -0.14496812873307705, "output_peak": 0.3901162872950848}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7490723144950442, "全帯域_10ms_rms_envelope_nrmse": 0.06836874793740001, "全帯域_10ms_envelope_correlation": 0.995716913072178, "全帯域_10ms_modulation_mae_db": 0.1564334102735271, "全帯域_40ms_log_rms_mae_db": 0.3875786010007174, "全帯域_40ms_rms_envelope_nrmse": 0.03735916141149175, "全帯域_40ms_envelope_correlation": 0.9987976802806949, "全帯域_40ms_modulation_mae_db": 0.2018317432649137, "300–1000 Hz_10ms_log_rms_mae_db": 1.1596972928394385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0833733104379126, "300–1000 Hz_10ms_envelope_correlation": 0.9942002418836597, "300–1000 Hz_10ms_modulation_mae_db": 0.1897445237490114, "300–1000 Hz_40ms_log_rms_mae_db": 0.5705669663881993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04620319861592745, "300–1000 Hz_40ms_envelope_correlation": 0.9983000532880497, "300–1000 Hz_40ms_modulation_mae_db": 0.22115764521485795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0977795272904871, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10304190561751829, "1000–3000 Hz_10ms_envelope_correlation": 0.9921875055525974, "1000–3000 Hz_10ms_modulation_mae_db": 0.3964865141485682, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6337818210602355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06121549918309939, "1000–3000 Hz_40ms_envelope_correlation": 0.9972434035632094, "1000–3000 Hz_40ms_modulation_mae_db": 0.686921692992498, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2762229963840115, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14176675325113508, "3000–6000 Hz_10ms_envelope_correlation": 0.9826036067887807, "3000–6000 Hz_10ms_modulation_mae_db": 0.5790788855054226, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8490639998912942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08618866082214127, "3000–6000 Hz_40ms_envelope_correlation": 0.9937656132240663, "3000–6000 Hz_40ms_modulation_mae_db": 0.8365297457589906, "mr_stft_db": 5.449059607943657, "band_mae_db": 0.2018296749004691, "rms_gain_error_db": -0.13366721117581315, "output_peak": 0.3821398139383391}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7206334392634053, "全帯域_10ms_rms_envelope_nrmse": 0.06410573947242075, "全帯域_10ms_envelope_correlation": 0.9962874833578567, "全帯域_10ms_modulation_mae_db": 0.4192047888209049, "全帯域_40ms_log_rms_mae_db": 0.37066664729155946, "全帯域_40ms_rms_envelope_nrmse": 0.0332836210082887, "全帯域_40ms_envelope_correlation": 0.9991204199471369, "全帯域_40ms_modulation_mae_db": 0.4741510248004446, "300–1000 Hz_10ms_log_rms_mae_db": 1.1108598383299255, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07942811590195938, "300–1000 Hz_10ms_envelope_correlation": 0.9947448286531541, "300–1000 Hz_10ms_modulation_mae_db": 0.5311047583492456, "300–1000 Hz_40ms_log_rms_mae_db": 0.5847039907793703, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040237247391521715, "300–1000 Hz_40ms_envelope_correlation": 0.9987603907122038, "300–1000 Hz_40ms_modulation_mae_db": 0.6012461434620632, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1247229473302747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10442668755551943, "1000–3000 Hz_10ms_envelope_correlation": 0.9921492681117108, "1000–3000 Hz_10ms_modulation_mae_db": 0.2679733701746111, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6739295297674067, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06552151538648089, "1000–3000 Hz_40ms_envelope_correlation": 0.9970600521745214, "1000–3000 Hz_40ms_modulation_mae_db": 0.4776591663166412, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2709766852501994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1525705498460092, "3000–6000 Hz_10ms_envelope_correlation": 0.9793802513945401, "3000–6000 Hz_10ms_modulation_mae_db": 0.6958430112535451, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8418906104487452, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08868249510211342, "3000–6000 Hz_40ms_envelope_correlation": 0.9927674730007447, "3000–6000 Hz_40ms_modulation_mae_db": 1.084227105550563, "mr_stft_db": 5.4512590948136905, "band_mae_db": 0.1840360528555264, "rms_gain_error_db": -0.14151645377592542, "output_peak": 0.383538907217562}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7542359907507119, "全帯域_10ms_rms_envelope_nrmse": 0.07566424627840454, "全帯域_10ms_envelope_correlation": 0.994598281466827, "全帯域_10ms_modulation_mae_db": 0.5001662267085897, "全帯域_40ms_log_rms_mae_db": 0.3819147499020607, "全帯域_40ms_rms_envelope_nrmse": 0.04463855174862366, "全帯域_40ms_envelope_correlation": 0.9980478504971274, "全帯域_40ms_modulation_mae_db": 0.6594048360353773, "300–1000 Hz_10ms_log_rms_mae_db": 1.1442219478483788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09300080423917775, "300–1000 Hz_10ms_envelope_correlation": 0.9925884416286697, "300–1000 Hz_10ms_modulation_mae_db": 0.6005484507184544, "300–1000 Hz_40ms_log_rms_mae_db": 0.5728855954714783, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05468045798795271, "300–1000 Hz_40ms_envelope_correlation": 0.9973312124529393, "300–1000 Hz_40ms_modulation_mae_db": 0.8012200049116952, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1320461418367358, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10850676083552356, "1000–3000 Hz_10ms_envelope_correlation": 0.9912705005461799, "1000–3000 Hz_10ms_modulation_mae_db": 0.47472808040843634, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6292079440257085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058544765784712204, "1000–3000 Hz_40ms_envelope_correlation": 0.9973906385252935, "1000–3000 Hz_40ms_modulation_mae_db": 0.36892005693871727, "3000–6000 Hz_10ms_log_rms_mae_db": 1.252063471601334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1406633111006976, "3000–6000 Hz_10ms_envelope_correlation": 0.9825771909882234, "3000–6000 Hz_10ms_modulation_mae_db": 0.5646762842498809, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8076486326038449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0792457762107922, "3000–6000 Hz_40ms_envelope_correlation": 0.9943801356387268, "3000–6000 Hz_40ms_modulation_mae_db": 0.9731224949330278, "mr_stft_db": 5.441334809188665, "band_mae_db": 0.1491468815475616, "rms_gain_error_db": -0.08323560012726408, "output_peak": 0.3833464598366404}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7796550510178939, "全帯域_10ms_rms_envelope_nrmse": 0.07859330192412949, "全帯域_10ms_envelope_correlation": 0.9942044013470797, "全帯域_10ms_modulation_mae_db": 0.158084183560543, "全帯域_40ms_log_rms_mae_db": 0.407408406038627, "全帯域_40ms_rms_envelope_nrmse": 0.04232850846164741, "全帯域_40ms_envelope_correlation": 0.998311858800004, "全帯域_40ms_modulation_mae_db": 0.1940464764104526, "300–1000 Hz_10ms_log_rms_mae_db": 1.208153413361804, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09479407006807648, "300–1000 Hz_10ms_envelope_correlation": 0.9923557584487844, "300–1000 Hz_10ms_modulation_mae_db": 0.14895983393880743, "300–1000 Hz_40ms_log_rms_mae_db": 0.5863570049151724, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05163576794316222, "300–1000 Hz_40ms_envelope_correlation": 0.9977297790358662, "300–1000 Hz_40ms_modulation_mae_db": 0.201675040586456, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1143015705361143, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09941435765765748, "1000–3000 Hz_10ms_envelope_correlation": 0.9926880174469982, "1000–3000 Hz_10ms_modulation_mae_db": 0.33608393712417106, "1000–3000 Hz_40ms_log_rms_mae_db": 0.636036475754746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059252449085153626, "1000–3000 Hz_40ms_envelope_correlation": 0.9973327754214637, "1000–3000 Hz_40ms_modulation_mae_db": 0.7038231885997949, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2446301178088979, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15521057246367143, "3000–6000 Hz_10ms_envelope_correlation": 0.9785672143616198, "3000–6000 Hz_10ms_modulation_mae_db": 0.6782929496308565, "3000–6000 Hz_40ms_log_rms_mae_db": 0.835942351960766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09451597716358053, "3000–6000 Hz_40ms_envelope_correlation": 0.9915005681500171, "3000–6000 Hz_40ms_modulation_mae_db": 1.2688153031011298, "mr_stft_db": 5.451081022127447, "band_mae_db": 0.17924504861604965, "rms_gain_error_db": -0.10693831005562954, "output_peak": 0.3858444878207676}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.73225377642692, "全帯域_10ms_rms_envelope_nrmse": 0.06343901804479718, "全帯域_10ms_envelope_correlation": 0.9963428729047872, "全帯域_10ms_modulation_mae_db": 0.1883758100145844, "全帯域_40ms_log_rms_mae_db": 0.3638060047489967, "全帯域_40ms_rms_envelope_nrmse": 0.03549330536786272, "全帯域_40ms_envelope_correlation": 0.9989400326282892, "全帯域_40ms_modulation_mae_db": 0.2645369842241792, "300–1000 Hz_10ms_log_rms_mae_db": 1.1179077136135993, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07974515178772952, "300–1000 Hz_10ms_envelope_correlation": 0.9947177781701216, "300–1000 Hz_10ms_modulation_mae_db": 0.24775107182862732, "300–1000 Hz_40ms_log_rms_mae_db": 0.5387902316563592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04514398284976031, "300–1000 Hz_40ms_envelope_correlation": 0.9983942124981988, "300–1000 Hz_40ms_modulation_mae_db": 0.32666700962722167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1080582713686888, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09857678866847276, "1000–3000 Hz_10ms_envelope_correlation": 0.9929255070831501, "1000–3000 Hz_10ms_modulation_mae_db": 0.4784765365290622, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6139697750990354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053552077426724055, "1000–3000 Hz_40ms_envelope_correlation": 0.9980421109434457, "1000–3000 Hz_40ms_modulation_mae_db": 0.3504190273687013, "3000–6000 Hz_10ms_log_rms_mae_db": 1.218922822176002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14496548264852946, "3000–6000 Hz_10ms_envelope_correlation": 0.9816380132059408, "3000–6000 Hz_10ms_modulation_mae_db": 0.17992109527000796, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8283132291837509, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08455937160239992, "3000–6000 Hz_40ms_envelope_correlation": 0.9932597642821752, "3000–6000 Hz_40ms_modulation_mae_db": 0.4331258833018046, "mr_stft_db": 5.447682188778387, "band_mae_db": 0.1469255823449432, "rms_gain_error_db": -0.13235637045043844, "output_peak": 0.38995541419781365}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.337619637183025, "spectral_convergence": 0.21792072368009383, "spectral_convergence_status": "ok", "amplitude_floor": 9.509362929412646e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.244913381246923, "spectral_convergence": 0.210890042316386, "spectral_convergence_status": "ok", "amplitude_floor": 8.503880220761315e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.788511484146582, "spectral_convergence": 0.223432035222807, "spectral_convergence_status": "ok", "amplitude_floor": 5.089992363719204e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.8030381557821282e-05, "output_power": 3.0831180360439654e-05, "delta_db": 0.41361142304746307, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003997973981863815, "output_power": 0.00040693323210834736, "delta_db": 0.07683194383526412, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008734907128440196, "output_power": 0.0008423615843833177, "delta_db": -0.1575973904848625, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001464675075572333, "output_power": 0.00014242247265807475, "delta_db": -0.12162769744355666, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.350074484130352e-05, "output_power": 5.8127710890267166e-05, "delta_db": -0.3839559896868182, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.7271042012588545e-05, "output_power": 1.601799596993902e-05, "delta_db": -0.3271036055662435, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24678800834403467, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09148793014587984, "pitch": null, "output_rms_dbfs": -28.248669311755425, "output_peak": 0.37409475901237726, "reference_rms_dbfs": -28.157181381609547, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.457014834192177}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012908667094909063, "log_rms_mae_db": 0.7173671236913817, "rms_envelope_nrmse": 0.07326075750504513, "envelope_correlation": 0.9949549167830729, "rms_gain_error_db": -0.09065264436790868, "reference_mean_power": 0.001527852496322364, "output_mean_power": 0.0014962913443795575}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.179575522794606e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002748180655831064, "output_power": 0.0002646304713376822, "reference_active": true, "delta_db": -0.1640542800632469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.543131824921543e-05, "output_power": 4.830108449116257e-05, "reference_active": true, "delta_db": -0.5979832500521018}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2340545383743233e-05, "output_power": 9.047509650821113e-06, "reference_active": true, "delta_db": -1.348052984967514}], "active_band_count": 3, "active_band_mae_db": 0.703363505027621}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001146348873577052, "log_rms_mae_db": 0.3909250241836874, "rms_envelope_nrmse": 0.040550899560334315, "envelope_correlation": 0.9984219465252709, "rms_gain_error_db": -0.08953378583197878, "reference_mean_power": 0.0015234711554082644, "output_mean_power": 0.001492384938376448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.718780555754886e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002667517105928089, "output_power": 0.00025436314043457176, "reference_active": true, "delta_db": -0.20653034836925308}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.819324290787445e-05, "output_power": 2.427791272158672e-05, "reference_active": true, "delta_db": -0.6493368735036807}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.654413380387e-07, "output_power": 5.474657835164546e-07, "reference_active": true, "delta_db": -1.4555493175379386}], "active_band_count": 3, "active_band_mae_db": 0.7704721798036241}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001269978635547725, "log_rms_mae_db": 1.100312466601757, "rms_envelope_nrmse": 0.08868914988244629, "envelope_correlation": 0.993279368346718, "rms_gain_error_db": -0.10330420420788083, "reference_mean_power": 0.0011723709196259178, "output_mean_power": 0.0011448131699480322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.280903308109198e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025355561575728737, "output_power": 0.00024376314974583404, "reference_active": true, "delta_db": -0.17105180841179263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.608107242392719e-05, "output_power": 4.778990109294012e-05, "reference_active": true, "delta_db": -0.6948017824366732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1840508111446508e-05, "output_power": 9.099078797884044e-06, "reference_active": true, "delta_db": -1.1437291362547537}], "active_band_count": 3, "active_band_mae_db": 0.6698609090344064}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.809451727166842e-05, "log_rms_mae_db": 0.5800943480354311, "rms_envelope_nrmse": 0.04988290889057304, "envelope_correlation": 0.9978109157306161, "rms_gain_error_db": -0.10222004605488016, "reference_mean_power": 0.0011689931508753605, "output_mean_power": 0.00114179979851718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.891072559770159e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025025709822561247, "output_power": 0.0002378451537132206, "reference_active": true, "delta_db": -0.22092097793114882}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.836247458716607e-05, "output_power": 2.369458432544808e-05, "reference_active": true, "delta_db": -0.780950254987056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.847292373572026e-07, "output_power": 5.909469134234338e-07, "reference_active": true, "delta_db": -1.2317136518148561}], "active_band_count": 3, "active_band_mae_db": 0.7445282949110202}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.9423530315000806e-05, "log_rms_mae_db": 1.1105957665837436, "rms_envelope_nrmse": 0.1012976076644774, "envelope_correlation": 0.9925789117292428, "rms_gain_error_db": -0.18268748199812807, "reference_mean_power": 0.00015673633956673046, "output_mean_power": 0.00015027991882439633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.197939100277272e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3248792094830553e-05, "output_power": 3.0698544840188146e-05, "reference_active": true, "delta_db": -0.34658082596997974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.648467501564817e-06, "output_power": 3.996833147883892e-06, "reference_active": true, "delta_db": 0.3960553638413544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1792733873632069e-06, "output_power": 1.0255586418537893e-06, "reference_active": true, "delta_db": -0.6065399938947303}], "active_band_count": 3, "active_band_mae_db": 0.44972539456868815}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.3506304987563625e-05, "log_rms_mae_db": 0.6023194488766578, "rms_envelope_nrmse": 0.06073686216443204, "envelope_correlation": 0.9974534298439662, "rms_gain_error_db": -0.18257769843784685, "reference_mean_power": 0.0001561782704543471, "output_mean_power": 0.00014974862357181558}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.897719153255603e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.156452210778437e-05, "output_power": 2.9084407932910976e-05, "reference_active": true, "delta_db": -0.3553899106670116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6348481818884734e-06, "output_power": 1.9149144201220865e-06, "reference_active": true, "delta_db": 0.6867194092414298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.181225496757285e-08, "output_power": 6.123862512905986e-08, "reference_active": true, "delta_db": -0.691731323955536}], "active_band_count": 3, "active_band_mae_db": 0.5779468812879925}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.766512572532955e-05, "log_rms_mae_db": 1.2294890322204768, "rms_envelope_nrmse": 0.13450520975609995, "envelope_correlation": 0.9845963340411499, "rms_gain_error_db": -0.3105104085219814, "reference_mean_power": 2.7841403311343547e-05, "output_mean_power": 2.5920303401125498e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1058323154967027e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.079425705940839e-06, "output_power": 4.648120254619562e-06, "reference_active": true, "delta_db": -0.3853725706241181}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8241793691525221e-06, "output_power": 1.6414507141099148e-06, "reference_active": true, "delta_db": -0.45839692479116856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.319091479762654e-07, "output_power": 3.6150753599647954e-07, "reference_active": true, "delta_db": -1.6772010450668344}], "active_band_count": 3, "active_band_mae_db": 0.8403235134940403}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5244936872690136e-05, "log_rms_mae_db": 0.8296202727639925, "rms_envelope_nrmse": 0.07901442390766819, "envelope_correlation": 0.9951362972087844, "rms_gain_error_db": -0.31345821680286223, "reference_mean_power": 2.7868742577273385e-05, "output_mean_power": 2.592815130276213e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.800642728490222e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.984173950209188e-06, "output_power": 4.454900692235726e-06, "reference_active": true, "delta_db": -0.4875516374002279}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0071402238290061e-06, "output_power": 8.817582977801051e-07, "reference_active": true, "delta_db": -0.5774038616901123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932914966264332e-08, "output_power": 2.1111760274719688e-08, "reference_active": true, "delta_db": -2.7019011080966266}], "active_band_count": 3, "active_band_mae_db": 1.2556188690623222}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8349459720194017, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1c2b1e91c9a0727f21dd0e8a6ab8fef221f11595effb4edba496b6683cdccdaf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9920404004709484, "standard_deviation": 0.0008416024818828594, "minimum": 0.9903599690119325, "maximum": 0.9929255070831501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1198839490265675, "standard_deviation": 0.016202000957148282, "minimum": 1.0977795272904871, "maximum": 1.1501835185349212, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.42172944009616165, "standard_deviation": 0.13536755955831675, "minimum": 0.2679733701746111, "maximum": 0.687408837953314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10415860193837052, "standard_deviation": 0.005330671563667611, "minimum": 0.09857678866847276, "maximum": 0.11491477306246602, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973914512980957, "standard_deviation": 0.0003060746685957327, "minimum": 0.9970600521745214, "maximum": 0.9980421109434457, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.639124773102075, "standard_deviation": 0.026113477144827115, "minimum": 0.6023194488766578, "maximum": 0.6760699014938776, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4911125429722969, "standard_deviation": 0.16636731974189206, "minimum": 0.2337387434569084, "maximum": 0.7038231885997949, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06015478200803688, "standard_deviation": 0.003739919309378601, "minimum": 0.053552077426724055, "maximum": 0.06552151538648089, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9819533286625343, "standard_deviation": 0.0021317345020745904, "minimum": 0.9785672143616198, "maximum": 0.9845963340411499, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.24485272534875, "standard_deviation": 0.021030644409971665, "minimum": 1.218922822176002, "maximum": 1.2762229963840115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5554330087796024, "standard_deviation": 0.2178647123815608, "minimum": 0.17992109527000796, "maximum": 0.8403235134940403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14339901079310124, "standard_deviation": 0.007720208466849534, "minimum": 0.13333534266909502, "maximum": 0.15521057246367143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937293090145767, "standard_deviation": 0.0011813779068441774, "minimum": 0.9915005681500171, "maximum": 0.9951362972087844, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8319081548229716, "standard_deviation": 0.013252604976815645, "minimum": 0.8076486326038449, "maximum": 0.8490639998912942, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9193629373937594, "standard_deviation": 0.29693167496383654, "minimum": 0.4331258833018046, "maximum": 1.2688153031011298, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08391992414168813, "standard_deviation": 0.005759316271113212, "minimum": 0.07757276974400287, "maximum": 0.09451597716358053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935574756902589, "standard_deviation": 0.0009299960749586195, "minimum": 0.9923557584487844, "maximum": 0.9947448286531541, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1302267852043981, "standard_deviation": 0.03844509766410599, "minimum": 1.100312466601757, "maximum": 1.208153413361804, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.37320310937649204, "standard_deviation": 0.20939505361404678, "minimum": 0.14895983393880743, "maximum": 0.6698609090344064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08718567327106959, "standard_deviation": 0.005925768553134483, "minimum": 0.07942811590195938, "maximum": 0.09479407006807648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978809086120329, "standard_deviation": 0.0005493244213794534, "minimum": 0.9972424815355598, "maximum": 0.9987603907122038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5702401094424455, "standard_deviation": 0.015294733293207307, "minimum": 0.5387902316563592, "maximum": 0.5863570049151724, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.46386062511981857, "standard_deviation": 0.24011190076904587, "minimum": 0.201675040586456, "maximum": 0.8012200049116952, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04990596998167267, "standard_deviation": 0.00571453533133964, "minimum": 0.040237247391521715, "maximum": 0.05732020367039904, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18462659731452696, "standard_deviation": 0.03497186188197625, "minimum": 0.1469255823449432, "maximum": 0.24678800834403467, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.448002882821884, "standard_deviation": 0.005349658170478767, "minimum": 5.441169499393897, "maximum": 5.457014834192177, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3844569071582934, "standard_deviation": 0.005122521621886569, "minimum": 0.37409475901237726, "maximum": 0.3901162872950848, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11879065543785369, "standard_deviation": 0.023188934645078683, "minimum": -0.14496812873307705, "maximum": -0.08323560012726408, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952285400570575, "standard_deviation": 0.0008210323496157439, "minimum": 0.9942044013470797, "maximum": 0.9963428729047872, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7413224299481278, "standard_deviation": 0.021140210019037913, "minimum": 0.7173671236913817, "maximum": 0.7796550510178939, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3267064381523833, "standard_deviation": 0.20922322132866922, "minimum": 0.11046935469172335, "maximum": 0.703363505027621, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0715492682664263, "standard_deviation": 0.00581787113893222, "minimum": 0.06343901804479718, "maximum": 0.07859330192412949, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984695827685028, "standard_deviation": 0.0004393089700543848, "minimum": 0.9978921535061154, "maximum": 0.9991204199471369, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3853397321756145, "standard_deviation": 0.013467562867753054, "minimum": 0.3638060047489967, "maximum": 0.407408406038627, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4106482372805207, "standard_deviation": 0.22551266065544603, "minimum": 0.1940464764104526, "maximum": 0.7704721798036241, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04067202047600642, "standard_deviation": 0.005003619704197335, "minimum": 0.0332836210082887, "maximum": 0.04815525542021655, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2961-960-0019", "language": "en", "style": "normal", "speaker": "librispeech_2961", "source_utterance_id": "2961-960-0019", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2961-960-0019.wav", "source_file_sha256": "539984a17681a297861f8b44ff396e54f93be623eb853a33ae9bcd4041536464", "decoded_pcm_sha256": "966508e2ccc362392d8ec80fc3d05a564460cb6f0dd69f613c7b86089cb21caa", "original_source_sha256": "f5052549b9cbc3b8433ae187d0e876b214f5250ddb4fbb637dea220e8a4726cf", "corpus_entry_id": "librispeech_2961-960-0019"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.4789912700653076, "predicted_mos": 3.4789912700653076, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5da349fbe2278a97551abe4e062b607a6d6f3ba031b4eb6cbb5164ca7a7666c0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.3658606839999266, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2961-960-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5da349fbe2278a97551abe4e062b607a6d6f3ba031b4eb6cbb5164ca7a7666c0", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "3aaa036a73b4b15888e1926e1d8bc85c1d6623633692de1b8c03961c9680dd7d"}, "rows": [{"case": "librispeech_2961-960-0019", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1068546772003174, "predicted_mos": 2.1068546772003174, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "af1288aac802f35373a632cb6f580cac869cc751a602d07b8fe2d1585a8cd995", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.378091234000749, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2961-960-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "af1288aac802f35373a632cb6f580cac869cc751a602d07b8fe2d1585a8cd995", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9025512198290396, "全帯域_10ms_rms_envelope_nrmse": 0.08885137828495722, "全帯域_10ms_envelope_correlation": 0.9932523022181998, "全帯域_10ms_modulation_mae_db": 0.282791282787632, "全帯域_40ms_log_rms_mae_db": 0.48386487716684823, "全帯域_40ms_rms_envelope_nrmse": 0.04254917065492387, "全帯域_40ms_envelope_correlation": 0.9985983783433907, "全帯域_40ms_modulation_mae_db": 0.2391994504902165, "300–1000 Hz_10ms_log_rms_mae_db": 1.2496191166258173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11520506915116892, "300–1000 Hz_10ms_envelope_correlation": 0.9891543155871927, "300–1000 Hz_10ms_modulation_mae_db": 0.24062021791829702, "300–1000 Hz_40ms_log_rms_mae_db": 0.652312122948713, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05715719204085468, "300–1000 Hz_40ms_envelope_correlation": 0.9975958261356579, "300–1000 Hz_40ms_modulation_mae_db": 0.16745697772754645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1677675774845568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08585715174599709, "1000–3000 Hz_10ms_envelope_correlation": 0.99511998232722, "1000–3000 Hz_10ms_modulation_mae_db": 0.3346519300711486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6581285424789979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05382240596996082, "1000–3000 Hz_40ms_envelope_correlation": 0.9981616419528111, "1000–3000 Hz_40ms_modulation_mae_db": 0.3660195680325895, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2817454687985792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14493226705692758, "3000–6000 Hz_10ms_envelope_correlation": 0.9858503243348866, "3000–6000 Hz_10ms_modulation_mae_db": 1.3522091061683306, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7567821855531888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07712639652386312, "3000–6000 Hz_40ms_envelope_correlation": 0.9960910049886843, "3000–6000 Hz_40ms_modulation_mae_db": 1.4644483563563935, "mr_stft_db": 5.701482098589643, "band_mae_db": 0.2575595261581776, "rms_gain_error_db": -0.15372051720817112, "output_peak": 0.6079365644609417}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8612052666995469, "全帯域_10ms_rms_envelope_nrmse": 0.0867091376804132, "全帯域_10ms_envelope_correlation": 0.9934424824168893, "全帯域_10ms_modulation_mae_db": 0.23669004556717868, "全帯域_40ms_log_rms_mae_db": 0.44719991659591674, "全帯域_40ms_rms_envelope_nrmse": 0.04002461391229455, "全帯域_40ms_envelope_correlation": 0.9986039395586667, "全帯域_40ms_modulation_mae_db": 0.16980152427280434, "300–1000 Hz_10ms_log_rms_mae_db": 1.2290606812012297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11213243895096768, "300–1000 Hz_10ms_envelope_correlation": 0.9894654497991424, "300–1000 Hz_10ms_modulation_mae_db": 0.12368390943007139, "300–1000 Hz_40ms_log_rms_mae_db": 0.6083518976030656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05030445037940505, "300–1000 Hz_40ms_envelope_correlation": 0.997806430292276, "300–1000 Hz_40ms_modulation_mae_db": 0.15276061076805836, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2080053696001214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09071174530248029, "1000–3000 Hz_10ms_envelope_correlation": 0.9946512423744204, "1000–3000 Hz_10ms_modulation_mae_db": 0.4111968894829244, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7073700113659374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058511283487505415, "1000–3000 Hz_40ms_envelope_correlation": 0.9979303350654599, "1000–3000 Hz_40ms_modulation_mae_db": 0.6549409889491246, "3000–6000 Hz_10ms_log_rms_mae_db": 1.258791219954486, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15382250029454375, "3000–6000 Hz_10ms_envelope_correlation": 0.9853223374561081, "3000–6000 Hz_10ms_modulation_mae_db": 1.123012873216498, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7490597884750839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09690764888685895, "3000–6000 Hz_40ms_envelope_correlation": 0.9953725208358815, "3000–6000 Hz_40ms_modulation_mae_db": 1.4326264651703713, "mr_stft_db": 5.712469009378846, "band_mae_db": 0.2864853034177351, "rms_gain_error_db": -0.08819882430097625, "output_peak": 0.6146770700182531}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8886020429370188, "全帯域_10ms_rms_envelope_nrmse": 0.07577678720056591, "全帯域_10ms_envelope_correlation": 0.995225506397118, "全帯域_10ms_modulation_mae_db": 0.3561477673625399, "全帯域_40ms_log_rms_mae_db": 0.5010556290942105, "全帯域_40ms_rms_envelope_nrmse": 0.04194521651621176, "全帯域_40ms_envelope_correlation": 0.9987153165313656, "全帯域_40ms_modulation_mae_db": 0.4029355243924552, "300–1000 Hz_10ms_log_rms_mae_db": 1.2235821417005572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1025058698019971, "300–1000 Hz_10ms_envelope_correlation": 0.9914262170962611, "300–1000 Hz_10ms_modulation_mae_db": 0.3897643297022779, "300–1000 Hz_40ms_log_rms_mae_db": 0.6925390792067825, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056236515519090646, "300–1000 Hz_40ms_envelope_correlation": 0.9975703338325937, "300–1000 Hz_40ms_modulation_mae_db": 0.4394829361915444, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1933046863058272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0909458572062152, "1000–3000 Hz_10ms_envelope_correlation": 0.9944630320019263, "1000–3000 Hz_10ms_modulation_mae_db": 0.9681703232047395, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6960754205991371, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05412522958770866, "1000–3000 Hz_40ms_envelope_correlation": 0.9980666818615977, "1000–3000 Hz_40ms_modulation_mae_db": 1.0939600267327543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2565506516672398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15381338529879662, "3000–6000 Hz_10ms_envelope_correlation": 0.985568480635665, "3000–6000 Hz_10ms_modulation_mae_db": 1.0273222689098023, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7774081375756317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09323907130619251, "3000–6000 Hz_40ms_envelope_correlation": 0.9961867037886816, "3000–6000 Hz_40ms_modulation_mae_db": 1.3865435551960232, "mr_stft_db": 5.690096537733587, "band_mae_db": 0.2571325685598634, "rms_gain_error_db": -0.17400547725458973, "output_peak": 0.6029106664350696}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8613868787010066, "全帯域_10ms_rms_envelope_nrmse": 0.08057485617752909, "全帯域_10ms_envelope_correlation": 0.9943636785217219, "全帯域_10ms_modulation_mae_db": 0.3384105776275792, "全帯域_40ms_log_rms_mae_db": 0.46084461015745587, "全帯域_40ms_rms_envelope_nrmse": 0.04421364217566599, "全帯域_40ms_envelope_correlation": 0.9982902658743278, "全帯域_40ms_modulation_mae_db": 0.47824402742907224, "300–1000 Hz_10ms_log_rms_mae_db": 1.218309540819286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10304912352337174, "300–1000 Hz_10ms_envelope_correlation": 0.9912133293427308, "300–1000 Hz_10ms_modulation_mae_db": 0.4085247870267829, "300–1000 Hz_40ms_log_rms_mae_db": 0.6365219103438634, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05720107532833485, "300–1000 Hz_40ms_envelope_correlation": 0.9973090103125783, "300–1000 Hz_40ms_modulation_mae_db": 0.49937563521118417, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2149785728479596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09307509812656087, "1000–3000 Hz_10ms_envelope_correlation": 0.994185554664643, "1000–3000 Hz_10ms_modulation_mae_db": 0.5932998455124112, "1000–3000 Hz_40ms_log_rms_mae_db": 0.707342185337481, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05379948720707064, "1000–3000 Hz_40ms_envelope_correlation": 0.9980870223127684, "1000–3000 Hz_40ms_modulation_mae_db": 0.4080568366226493, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2747726992236605, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1491470560611882, "3000–6000 Hz_10ms_envelope_correlation": 0.9854069063585278, "3000–6000 Hz_10ms_modulation_mae_db": 0.5440010628703793, "3000–6000 Hz_40ms_log_rms_mae_db": 0.775089858774609, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08335245226358934, "3000–6000 Hz_40ms_envelope_correlation": 0.9953813716828653, "3000–6000 Hz_40ms_modulation_mae_db": 0.6716964811901365, "mr_stft_db": 5.713037005870969, "band_mae_db": 0.21433955271786, "rms_gain_error_db": -0.09378630221823835, "output_peak": 0.6239382539438825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8846052114683419, "全帯域_10ms_rms_envelope_nrmse": 0.08221259817235702, "全帯域_10ms_envelope_correlation": 0.9941210809975819, "全帯域_10ms_modulation_mae_db": 0.4473183048100687, "全帯域_40ms_log_rms_mae_db": 0.46836584765380024, "全帯域_40ms_rms_envelope_nrmse": 0.048251710418600456, "全帯域_40ms_envelope_correlation": 0.9979356972610707, "全帯域_40ms_modulation_mae_db": 0.6128844776419817, "300–1000 Hz_10ms_log_rms_mae_db": 1.2194579971642892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10898296794062061, "300–1000 Hz_10ms_envelope_correlation": 0.9900728932513387, "300–1000 Hz_10ms_modulation_mae_db": 0.4160406842800648, "300–1000 Hz_40ms_log_rms_mae_db": 0.6639513191237356, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06559451353221965, "300–1000 Hz_40ms_envelope_correlation": 0.9962813619778648, "300–1000 Hz_40ms_modulation_mae_db": 0.5168568252000473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1750531164491487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07875295644265587, "1000–3000 Hz_10ms_envelope_correlation": 0.995819672385043, "1000–3000 Hz_10ms_modulation_mae_db": 0.3908083605637282, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6956706033956132, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0401814173651039, "1000–3000 Hz_40ms_envelope_correlation": 0.9989165776512808, "1000–3000 Hz_40ms_modulation_mae_db": 0.330587408233115, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3281220945280914, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14152101572157505, "3000–6000 Hz_10ms_envelope_correlation": 0.9876113297441377, "3000–6000 Hz_10ms_modulation_mae_db": 1.5091734389103042, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9010163376546112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09050149702257762, "3000–6000 Hz_40ms_envelope_correlation": 0.9958681490372133, "3000–6000 Hz_40ms_modulation_mae_db": 1.9553868865377335, "mr_stft_db": 5.703935689721288, "band_mae_db": 0.23352952815547146, "rms_gain_error_db": -0.08985841399599791, "output_peak": 0.6111213358637151}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8781964973453439, "全帯域_10ms_rms_envelope_nrmse": 0.09309497889245885, "全帯域_10ms_envelope_correlation": 0.9925177056707573, "全帯域_10ms_modulation_mae_db": 0.48947849757774103, "全帯域_40ms_log_rms_mae_db": 0.4584822137811987, "全帯域_40ms_rms_envelope_nrmse": 0.0501358902105112, "全帯域_40ms_envelope_correlation": 0.9978847471561509, "全帯域_40ms_modulation_mae_db": 0.3762023587746584, "300–1000 Hz_10ms_log_rms_mae_db": 1.2368401207927844, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11861552070281303, "300–1000 Hz_10ms_envelope_correlation": 0.9883361693503065, "300–1000 Hz_10ms_modulation_mae_db": 0.5456714565826815, "300–1000 Hz_40ms_log_rms_mae_db": 0.6431322413518926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06549127295329835, "300–1000 Hz_40ms_envelope_correlation": 0.9965015165681957, "300–1000 Hz_40ms_modulation_mae_db": 0.4272740191750781, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2282206436057892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08746338829551668, "1000–3000 Hz_10ms_envelope_correlation": 0.9948631992024707, "1000–3000 Hz_10ms_modulation_mae_db": 0.2908908500634179, "1000–3000 Hz_40ms_log_rms_mae_db": 0.70034046216055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048615584854630724, "1000–3000 Hz_40ms_envelope_correlation": 0.9984328729360249, "1000–3000 Hz_40ms_modulation_mae_db": 0.2702958661625918, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2402095278486787, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1292403591223701, "3000–6000 Hz_10ms_envelope_correlation": 0.9888650044769229, "3000–6000 Hz_10ms_modulation_mae_db": 1.2239588669669843, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7436549447511223, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06761136199508838, "3000–6000 Hz_40ms_envelope_correlation": 0.9971364547810172, "3000–6000 Hz_40ms_modulation_mae_db": 1.5371522201868384, "mr_stft_db": 5.696701075086618, "band_mae_db": 0.21128828581049028, "rms_gain_error_db": -0.1400586155692171, "output_peak": 0.6613138917000164}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8330557936277023, "全帯域_10ms_rms_envelope_nrmse": 0.08593086207235807, "全帯域_10ms_envelope_correlation": 0.9935817158778052, "全帯域_10ms_modulation_mae_db": 0.3948619367194725, "全帯域_40ms_log_rms_mae_db": 0.417104646388114, "全帯域_40ms_rms_envelope_nrmse": 0.04434198460312189, "全帯域_40ms_envelope_correlation": 0.998210649977472, "全帯域_40ms_modulation_mae_db": 0.30196020684662034, "300–1000 Hz_10ms_log_rms_mae_db": 1.1723153988031048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10993956217565336, "300–1000 Hz_10ms_envelope_correlation": 0.9900490836172025, "300–1000 Hz_10ms_modulation_mae_db": 0.46722434144530994, "300–1000 Hz_40ms_log_rms_mae_db": 0.5768110971478613, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05690773421884391, "300–1000 Hz_40ms_envelope_correlation": 0.9972163324305758, "300–1000 Hz_40ms_modulation_mae_db": 0.3931416371146054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2332358448096123, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0856935457451302, "1000–3000 Hz_10ms_envelope_correlation": 0.9950579789306258, "1000–3000 Hz_10ms_modulation_mae_db": 0.21742336546264243, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7512294980310099, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04544743809281768, "1000–3000 Hz_40ms_envelope_correlation": 0.9985790556742671, "1000–3000 Hz_40ms_modulation_mae_db": 0.1956208729596521, "3000–6000 Hz_10ms_log_rms_mae_db": 1.285950868366028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15621671880133525, "3000–6000 Hz_10ms_envelope_correlation": 0.9842730534772834, "3000–6000 Hz_10ms_modulation_mae_db": 1.0321620698143434, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8037127853018406, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08572137873466189, "3000–6000 Hz_40ms_envelope_correlation": 0.9962442461197597, "3000–6000 Hz_40ms_modulation_mae_db": 1.1715573942135606, "mr_stft_db": 5.704145674702102, "band_mae_db": 0.1696388780228453, "rms_gain_error_db": 0.021111918165140037, "output_peak": 0.6467421336494594}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8552163968010488, "全帯域_10ms_rms_envelope_nrmse": 0.08200305486829698, "全帯域_10ms_envelope_correlation": 0.9942766898873727, "全帯域_10ms_modulation_mae_db": 0.32011549481696516, "全帯域_40ms_log_rms_mae_db": 0.4557927781566433, "全帯域_40ms_rms_envelope_nrmse": 0.050999434301160766, "全帯域_40ms_envelope_correlation": 0.9978414762582614, "全帯域_40ms_modulation_mae_db": 0.3314381011099547, "300–1000 Hz_10ms_log_rms_mae_db": 1.2097599534338235, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1053681546947342, "300–1000 Hz_10ms_envelope_correlation": 0.9908195147536234, "300–1000 Hz_10ms_modulation_mae_db": 0.24803191611041908, "300–1000 Hz_40ms_log_rms_mae_db": 0.6709943932286643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06574211212718772, "300–1000 Hz_40ms_envelope_correlation": 0.996391126284385, "300–1000 Hz_40ms_modulation_mae_db": 0.19566415099894063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.176283499666327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08055462347752738, "1000–3000 Hz_10ms_envelope_correlation": 0.9958709616334007, "1000–3000 Hz_10ms_modulation_mae_db": 0.417071762444595, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6781765903898452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04646467865327237, "1000–3000 Hz_40ms_envelope_correlation": 0.9988393904799456, "1000–3000 Hz_40ms_modulation_mae_db": 0.2854762201367631, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3013806252539624, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19003343914521695, "3000–6000 Hz_10ms_envelope_correlation": 0.976487942828453, "3000–6000 Hz_10ms_modulation_mae_db": 0.6133023581934882, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8114914397868201, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12867825688993678, "3000–6000 Hz_40ms_envelope_correlation": 0.9901825072424622, "3000–6000 Hz_40ms_modulation_mae_db": 0.9304177847693982, "mr_stft_db": 5.692163444586988, "band_mae_db": 0.2699286355618716, "rms_gain_error_db": -0.15243296542797197, "output_peak": 0.6067713632510137}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.534560941532004, "spectral_convergence": 0.2506398441707466, "spectral_convergence_status": "ok", "amplitude_floor": 1.5171791511760096e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.514141319916119, "spectral_convergence": 0.2540151723007567, "spectral_convergence_status": "ok", "amplitude_floor": 1.2250623067014244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.055744034320804, "spectral_convergence": 0.26182171176273145, "spectral_convergence_status": "ok", "amplitude_floor": 7.883290232655995e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.642531119863568e-05, "output_power": 4.992129944011643e-05, "delta_db": 0.31531057819560343, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003916996423883643, "output_power": 0.0004387898489026153, "delta_db": 0.49303396870261146, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0020347588099002725, "output_power": 0.001895297469680759, "delta_db": -0.30835554722010666, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007257385551015356, "output_power": 0.0007043082792577657, "delta_db": -0.1301740210863342, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001922577058777525, "output_power": 0.0001843192297846541, "delta_db": -0.1831310881267759, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.752614415284284e-05, "output_power": 3.654254135897667e-05, "delta_db": -0.11535195361763412, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2575595261581776, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15372051720817112, "pitch": null, "output_rms_dbfs": -24.802797838778492, "output_peak": 0.6079365644609417, "reference_rms_dbfs": -24.64907732157032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.701482098589643}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019822260308892205, "log_rms_mae_db": 0.9025512198290396, "rms_envelope_nrmse": 0.08885137828495722, "envelope_correlation": 0.9932523022181998, "rms_gain_error_db": -0.15369910767373754, "reference_mean_power": 0.003431222806852645, "output_mean_power": 0.003311913681013872}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9226628348981265e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006393448264747715, "output_power": 0.0006266355678783275, "reference_active": true, "delta_db": -0.08720113421935835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011197882141156391, "output_power": 0.00010182818771706391, "reference_active": true, "delta_db": -0.4126787769122892}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.279682095777351e-05, "output_power": 3.949681866965491e-05, "reference_active": true, "delta_db": -0.3484939372312484}], "active_band_count": 3, "active_band_mae_db": 0.282791282787632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001679501427730559, "log_rms_mae_db": 0.48386487716684823, "rms_envelope_nrmse": 0.04254917065492387, "envelope_correlation": 0.9985983783433907, "rms_gain_error_db": -0.15399040480543572, "reference_mean_power": 0.003447933525092644, "output_mean_power": 0.0033278201242456337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8195786138583623e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005865504590263927, "output_power": 0.0005732806270579235, "reference_active": true, "delta_db": -0.09938113142725004}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.169773206547347e-05, "output_power": 5.56540081884881e-05, "reference_active": true, "delta_db": -0.4477275260030081}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.232527996839891e-06, "output_power": 2.1465842792215078e-06, "reference_active": true, "delta_db": -0.1704896940403913}], "active_band_count": 3, "active_band_mae_db": 0.2391994504902165}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001900450506992256, "log_rms_mae_db": 1.2496191166258173, "rms_envelope_nrmse": 0.11520506915116892, "envelope_correlation": 0.9891543155871927, "rms_gain_error_db": -0.2197019411052336, "reference_mean_power": 0.002194955532035717, "output_mean_power": 0.0020866784556890726}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4050180992423441e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048796679990628384, "output_power": 0.00046349121700108795, "reference_active": true, "delta_db": -0.22348765822615385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011744031611151056, "output_power": 0.00010989467399668499, "reference_active": true, "delta_db": -0.2884056653678714}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.011051692782722e-05, "output_power": 3.8217434155729885e-05, "reference_active": true, "delta_db": -0.20996733016086572}], "active_band_count": 3, "active_band_mae_db": 0.24062021791829702}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014767968623269652, "log_rms_mae_db": 0.652312122948713, "rms_envelope_nrmse": 0.05715719204085468, "envelope_correlation": 0.9975958261356579, "rms_gain_error_db": -0.21977379864115, "reference_mean_power": 0.002205773735941533, "output_mean_power": 0.002096928302328324}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315971715150661e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004578823491754882, "output_power": 0.0004326015313609391, "reference_active": true, "delta_db": -0.24665849850735264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0620858906342295e-05, "output_power": 5.746583144081291e-05, "reference_active": true, "delta_db": -0.23212390727480822}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.249804591930109e-06, "output_power": 2.2376179956399644e-06, "reference_active": true, "delta_db": -0.02358852740047856}], "active_band_count": 3, "active_band_mae_db": 0.16745697772754645}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012415579655882008, "log_rms_mae_db": 1.1677675774845568, "rms_envelope_nrmse": 0.08585715174599709, "envelope_correlation": 0.99511998232722, "rms_gain_error_db": -0.1371987184276643, "reference_mean_power": 0.0006561031712122296, "output_mean_power": 0.0006357000784721009}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1091236930101443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020492597233413883, "output_power": 0.00019781065029287175, "reference_active": true, "delta_db": -0.1534733373371715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1330216505908255e-06, "output_power": 7.251188045342712e-06, "reference_active": true, "delta_db": 0.07135625628810091}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.269143835584622e-06, "output_power": 5.108029619639661e-06, "reference_active": true, "delta_db": 0.7791261965881733}], "active_band_count": 3, "active_band_mae_db": 0.3346519300711486}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011294582885045651, "log_rms_mae_db": 0.6581285424789979, "rms_envelope_nrmse": 0.05382240596996082, "envelope_correlation": 0.9981616419528111, "rms_gain_error_db": -0.13825102884316548, "reference_mean_power": 0.000659140146325869, "output_mean_power": 0.0006384878850545499}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0026600561194904e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019159735065259618, "output_power": 0.00018467499720541667, "reference_active": true, "delta_db": -0.15981398376587172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0478497584910014e-06, "output_power": 3.2457018380987914e-06, "reference_active": true, "delta_db": 0.2731506649330479}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0860941276748228e-07, "output_power": 2.43132662555651e-07, "reference_active": true, "delta_db": 0.6650940553988488}], "active_band_count": 3, "active_band_mae_db": 0.3660195680325895}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.245162939338534e-05, "log_rms_mae_db": 1.2817454687985792, "rms_envelope_nrmse": 0.14493226705692758, "envelope_correlation": 0.9858503243348866, "rms_gain_error_db": -0.19225625272964558, "reference_mean_power": 8.990510589252764e-05, "output_mean_power": 8.601193768859902e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.06772172679864e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1343959051909258e-05, "output_power": 1.0558573159814363e-05, "reference_active": true, "delta_db": -0.3115941670297191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.893195477577164e-06, "output_power": 3.1207882131272495e-06, "reference_active": true, "delta_db": -0.9604191319296996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973785070466098e-07, "output_power": 4.199553548611097e-07, "reference_active": true, "delta_db": -2.7846140195455726}], "active_band_count": 3, "active_band_mae_db": 1.3522091061683306}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.719774514443707e-05, "log_rms_mae_db": 0.7567821855531888, "rms_envelope_nrmse": 0.07712639652386312, "envelope_correlation": 0.9960910049886843, "rms_gain_error_db": -0.19226420738165714, "reference_mean_power": 9.035589049479441e-05, "output_mean_power": 8.64430435966145e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9012156385948547e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1053455137857872e-05, "output_power": 1.0097543494144654e-05, "reference_active": true, "delta_db": -0.3928232040582333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0905012137138186e-06, "output_power": 1.5912024689293071e-06, "reference_active": true, "delta_db": -1.1852498005784642}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.495315305956309e-08, "output_power": 3.396823893516998e-08, "reference_active": true, "delta_db": -2.8152720644324836}], "active_band_count": 3, "active_band_mae_db": 1.4644483563563935}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8094938105726005, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "155fd40e2821b6f204bca7d509b576d62dfac932877bbf913640a1bbfc37907b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9950039529399688, "standard_deviation": 0.0006023847216177231, "minimum": 0.994185554664643, "maximum": 0.9958709616334007, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.199606163846168, "standard_deviation": 0.025232644703285068, "minimum": 1.1677675774845568, "maximum": 1.2332358448096123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4529391658507009, "standard_deviation": 0.235363733752929, "minimum": 0.21742336546264243, "maximum": 0.9681703232047395, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08663179579276045, "standard_deviation": 0.005042013831527132, "minimum": 0.07875295644265587, "maximum": 0.09307509812656087, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9983766972417695, "standard_deviation": 0.00037324247255631437, "minimum": 0.9979303350654599, "maximum": 0.9989165776512808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6992916642198215, "standard_deviation": 0.026705843073362826, "minimum": 0.6581285424789979, "maximum": 0.7512294980310099, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.45061972347865503, "standard_deviation": 0.2937014887933813, "minimum": 0.1956208729596521, "maximum": 1.0939600267327543, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05012094065225878, "standard_deviation": 0.005975364068493368, "minimum": 0.0401814173651039, "maximum": 0.058511283487505415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.984923172413998, "standard_deviation": 0.0037034606843157503, "minimum": 0.976487942828453, "maximum": 0.9888650044769229, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2784403944550906, "standard_deviation": 0.02778015297112404, "minimum": 1.2402095278486787, "maximum": 1.3281220945280914, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0531427556312662, "standard_deviation": 0.3349320546397136, "minimum": 0.5440010628703793, "maximum": 1.5091734389103042, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15234084268774417, "standard_deviation": 0.017549546989612835, "minimum": 0.1292403591223701, "maximum": 0.19003343914521695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9953078698095705, "standard_deviation": 0.0021447280880440137, "minimum": 0.9901825072424622, "maximum": 0.9971364547810172, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7897769347341135, "standard_deviation": 0.051105503197110755, "minimum": 0.7436549447511223, "maximum": 0.9010163376546112, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3187286429525569, "standard_deviation": 0.39332714604532376, "minimum": 0.6716964811901365, "maximum": 1.9553868865377335, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09039225795284607, "standard_deviation": 0.018058288915863502, "minimum": 0.06761136199508838, "maximum": 0.12867825688993678, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9900671215997248, "standard_deviation": 0.0010636061542519484, "minimum": 0.9883361693503065, "maximum": 0.9914262170962611, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2198681188176115, "standard_deviation": 0.022795680806843216, "minimum": 1.1723153988031048, "maximum": 1.2496191166258173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3549452053119881, "standard_deviation": 0.13876050745719282, "minimum": 0.12368390943007139, "maximum": 0.5456714565826815, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10947483836766583, "standard_deviation": 0.005739954565503863, "minimum": 0.1025058698019971, "maximum": 0.11861552070281303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970839922292658, "standard_deviation": 0.0006038119809327166, "minimum": 0.9962813619778648, "maximum": 0.997806430292276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6430767576193224, "standard_deviation": 0.03661942826994381, "minimum": 0.5768110971478613, "maximum": 0.6925390792067825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3490015990483756, "standard_deviation": 0.1521422561611287, "minimum": 0.15276061076805836, "maximum": 0.5168568252000473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05932935826240436, "standard_deviation": 0.0056628056048012325, "minimum": 0.05030445037940505, "maximum": 0.06574211212718772, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23748778480053934, "standard_deviation": 0.03795364522549582, "minimum": 0.1696388780228453, "maximum": 0.2864853034177351, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.701753816958755, "standard_deviation": 0.008502761612775456, "minimum": 5.690096537733587, "maximum": 5.713037005870969, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6219264099152939, "standard_deviation": 0.02113741993895266, "minimum": 0.6029106664350696, "maximum": 0.6613138917000164, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10886864972625279, "standard_deviation": 0.06214753859080476, "minimum": -0.17400547725458973, "maximum": 0.021111918165140037, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9938476452484308, "standard_deviation": 0.0008264974331871882, "minimum": 0.9925177056707573, "maximum": 0.995225506397118, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8706024134261312, "standard_deviation": 0.022097743806709383, "minimum": 0.8330557936277023, "maximum": 0.9025512198290396, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35822673840864716, "standard_deviation": 0.0835101483589373, "minimum": 0.23669004556717868, "maximum": 0.48947849757774103, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08439420666861705, "standard_deviation": 0.005380047935967486, "minimum": 0.07577678720056591, "maximum": 0.09309497889245885, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982600588700882, "standard_deviation": 0.00035144761301672333, "minimum": 0.9978414762582614, "maximum": 0.9987153165313656, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4615888148742735, "standard_deviation": 0.024869548337108808, "minimum": 0.417104646388114, "maximum": 0.5010556290942105, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36408320886972045, "standard_deviation": 0.13879966888110948, "minimum": 0.16980152427280434, "maximum": 0.6128844776419817, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04530770784906131, "standard_deviation": 0.0040236941820648384, "minimum": 0.04002461391229455, "maximum": 0.050999434301160766, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0019", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.9630200862884521, "predicted_mos": 1.9630200862884521, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a5db47ef1f0a75868b3a042510faf820f91853e30b019199723ff77f42222da1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.275642900262028, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9070331244899518, "全帯域_10ms_rms_envelope_nrmse": 0.08925065415112678, "全帯域_10ms_envelope_correlation": 0.9931947320826194, "全帯域_10ms_modulation_mae_db": 0.25891023298191257, "全帯域_40ms_log_rms_mae_db": 0.48893491411840206, "全帯域_40ms_rms_envelope_nrmse": 0.0432211549233168, "全帯域_40ms_envelope_correlation": 0.998549630111653, "全帯域_40ms_modulation_mae_db": 0.16836593579701406, "300–1000 Hz_10ms_log_rms_mae_db": 1.2352831795869204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11589731826017738, "300–1000 Hz_10ms_envelope_correlation": 0.9889677582061948, "300–1000 Hz_10ms_modulation_mae_db": 0.20248006653557074, "300–1000 Hz_40ms_log_rms_mae_db": 0.6341451098662704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05681970485624069, "300–1000 Hz_40ms_envelope_correlation": 0.9975643144872534, "300–1000 Hz_40ms_modulation_mae_db": 0.1784328522935524, "1000–3000 Hz_10ms_log_rms_mae_db": 1.156785012897676, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08226430365737418, "1000–3000 Hz_10ms_envelope_correlation": 0.9955871514056539, "1000–3000 Hz_10ms_modulation_mae_db": 0.22664654072082305, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6512486353772379, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05013326330841049, "1000–3000 Hz_40ms_envelope_correlation": 0.9984834324134292, "1000–3000 Hz_40ms_modulation_mae_db": 0.22273227207716428, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2851516759563157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1399692449192789, "3000–6000 Hz_10ms_envelope_correlation": 0.9870851442067952, "3000–6000 Hz_10ms_modulation_mae_db": 1.6220153280725196, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7767776464078457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07782324749614578, "3000–6000 Hz_40ms_envelope_correlation": 0.9963773572035138, "3000–6000 Hz_40ms_modulation_mae_db": 1.7182473330476125, "mr_stft_db": 5.665255992647704, "band_mae_db": 0.2697535604439791, "rms_gain_error_db": -0.15464584562842754, "output_peak": 0.6061673406405343}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8746174403075971, "全帯域_10ms_rms_envelope_nrmse": 0.08678842625078649, "全帯域_10ms_envelope_correlation": 0.9934212776572238, "全帯域_10ms_modulation_mae_db": 0.2263562126785482, "全帯域_40ms_log_rms_mae_db": 0.45085672728047077, "全帯域_40ms_rms_envelope_nrmse": 0.040492389949134296, "全帯域_40ms_envelope_correlation": 0.9985507624519148, "全帯域_40ms_modulation_mae_db": 0.3200265219825888, "300–1000 Hz_10ms_log_rms_mae_db": 1.2491347387410756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11208017699960031, "300–1000 Hz_10ms_envelope_correlation": 0.9894814721560681, "300–1000 Hz_10ms_modulation_mae_db": 0.13184151704615021, "300–1000 Hz_40ms_log_rms_mae_db": 0.6071836192493867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05099047324398023, "300–1000 Hz_40ms_envelope_correlation": 0.9977166372426227, "300–1000 Hz_40ms_modulation_mae_db": 0.2798569577503242, "1000–3000 Hz_10ms_log_rms_mae_db": 1.206700610329751, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08971857073346362, "1000–3000 Hz_10ms_envelope_correlation": 0.9948085642394329, "1000–3000 Hz_10ms_modulation_mae_db": 0.41834411898692686, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6949046808813008, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05737466371737402, "1000–3000 Hz_40ms_envelope_correlation": 0.9980575717940309, "1000–3000 Hz_40ms_modulation_mae_db": 0.609062444030957, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2615469663817265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15175526351383348, "3000–6000 Hz_10ms_envelope_correlation": 0.986198413693232, "3000–6000 Hz_10ms_modulation_mae_db": 1.3139623627822512, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7698353780731221, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09516588544740767, "3000–6000 Hz_40ms_envelope_correlation": 0.9961142277924352, "3000–6000 Hz_40ms_modulation_mae_db": 1.6219236826195462, "mr_stft_db": 5.674707721973708, "band_mae_db": 0.2969582022173487, "rms_gain_error_db": -0.07778768653939161, "output_peak": 0.613228676057835}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8919800309228109, "全帯域_10ms_rms_envelope_nrmse": 0.07731177680516833, "全帯域_10ms_envelope_correlation": 0.9950005331864339, "全帯域_10ms_modulation_mae_db": 0.32999324257387064, "全帯域_40ms_log_rms_mae_db": 0.5084250645285784, "全帯域_40ms_rms_envelope_nrmse": 0.042403538246872784, "全帯域_40ms_envelope_correlation": 0.998660396219997, "全帯域_40ms_modulation_mae_db": 0.350246579284216, "300–1000 Hz_10ms_log_rms_mae_db": 1.2265617637026967, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10265513302476051, "300–1000 Hz_10ms_envelope_correlation": 0.9913513009550458, "300–1000 Hz_10ms_modulation_mae_db": 0.3884926326335926, "300–1000 Hz_40ms_log_rms_mae_db": 0.6951042923656825, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05547585877796989, "300–1000 Hz_40ms_envelope_correlation": 0.9975861832728349, "300–1000 Hz_40ms_modulation_mae_db": 0.4141949946260299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1794240829484846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08781912171134772, "1000–3000 Hz_10ms_envelope_correlation": 0.9948817817048623, "1000–3000 Hz_10ms_modulation_mae_db": 0.8832492437437885, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6669657434282984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05232147447701416, "1000–3000 Hz_40ms_envelope_correlation": 0.9982331879834769, "1000–3000 Hz_40ms_modulation_mae_db": 1.0019959796897642, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2625603385211817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1525016552924046, "3000–6000 Hz_10ms_envelope_correlation": 0.9859633823884563, "3000–6000 Hz_10ms_modulation_mae_db": 1.091508100943998, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7600643902761844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09066408496507547, "3000–6000 Hz_40ms_envelope_correlation": 0.9966518724939841, "3000–6000 Hz_40ms_modulation_mae_db": 1.3853440374307053, "mr_stft_db": 5.654857500456543, "band_mae_db": 0.27226698990191583, "rms_gain_error_db": -0.16910446319146424, "output_peak": 0.6009120172813521}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.861756097775982, "全帯域_10ms_rms_envelope_nrmse": 0.08159988081793827, "全帯域_10ms_envelope_correlation": 0.9942142559434096, "全帯域_10ms_modulation_mae_db": 0.33775734728790113, "全帯域_40ms_log_rms_mae_db": 0.4728215966965786, "全帯域_40ms_rms_envelope_nrmse": 0.04594825620387173, "全帯域_40ms_envelope_correlation": 0.9981402670926413, "全帯域_40ms_modulation_mae_db": 0.4159669258550414, "300–1000 Hz_10ms_log_rms_mae_db": 1.2211012164372599, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1034710709336662, "300–1000 Hz_10ms_envelope_correlation": 0.9911064979290857, "300–1000 Hz_10ms_modulation_mae_db": 0.42757284210733837, "300–1000 Hz_40ms_log_rms_mae_db": 0.6430755715787726, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05808644127119337, "300–1000 Hz_40ms_envelope_correlation": 0.997171959802902, "300–1000 Hz_40ms_modulation_mae_db": 0.47546701631423743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1837485671469656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08925065508151786, "1000–3000 Hz_10ms_envelope_correlation": 0.9947069052961627, "1000–3000 Hz_10ms_modulation_mae_db": 0.5896001293406757, "1000–3000 Hz_40ms_log_rms_mae_db": 0.667879984553307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049687128903275334, "1000–3000 Hz_40ms_envelope_correlation": 0.9984348271509459, "1000–3000 Hz_40ms_modulation_mae_db": 0.39905903160496664, "3000–6000 Hz_10ms_log_rms_mae_db": 1.277634800339917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15035793614514315, "3000–6000 Hz_10ms_envelope_correlation": 0.9854958037405627, "3000–6000 Hz_10ms_modulation_mae_db": 0.7005845472134444, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7913869848472362, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08837538724749514, "3000–6000 Hz_40ms_envelope_correlation": 0.9950734731850619, "3000–6000 Hz_40ms_modulation_mae_db": 0.7727909150433862, "mr_stft_db": 5.676987389936642, "band_mae_db": 0.20622283749711548, "rms_gain_error_db": -0.09078044597792352, "output_peak": 0.6213044631713978}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9094491942294597, "全帯域_10ms_rms_envelope_nrmse": 0.08352046089790024, "全帯域_10ms_envelope_correlation": 0.9939258983052752, "全帯域_10ms_modulation_mae_db": 0.3786919790805483, "全帯域_40ms_log_rms_mae_db": 0.4860674618391242, "全帯域_40ms_rms_envelope_nrmse": 0.04829274089729578, "全帯域_40ms_envelope_correlation": 0.9979253909638518, "全帯域_40ms_modulation_mae_db": 0.5197920079742838, "300–1000 Hz_10ms_log_rms_mae_db": 1.2412895841773295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11032914503433457, "300–1000 Hz_10ms_envelope_correlation": 0.9898069342665815, "300–1000 Hz_10ms_modulation_mae_db": 0.33214271030973813, "300–1000 Hz_40ms_log_rms_mae_db": 0.6682442660482574, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06546047241483532, "300–1000 Hz_40ms_envelope_correlation": 0.9962690056552729, "300–1000 Hz_40ms_modulation_mae_db": 0.43095405107816065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1689581723936078, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07924032177937274, "1000–3000 Hz_10ms_envelope_correlation": 0.9957748711002509, "1000–3000 Hz_10ms_modulation_mae_db": 0.42350571843970347, "1000–3000 Hz_40ms_log_rms_mae_db": 0.678177849629571, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04002826804734172, "1000–3000 Hz_40ms_envelope_correlation": 0.9989296060796253, "1000–3000 Hz_40ms_modulation_mae_db": 0.38501005672479294, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3154555380015267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13789240139410522, "3000–6000 Hz_10ms_envelope_correlation": 0.9884980921496337, "3000–6000 Hz_10ms_modulation_mae_db": 1.7454531954014953, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8810526947366542, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08841042336216404, "3000–6000 Hz_40ms_envelope_correlation": 0.9963373727923229, "3000–6000 Hz_40ms_modulation_mae_db": 2.2009194967942016, "mr_stft_db": 5.670734102921212, "band_mae_db": 0.22355877256876, "rms_gain_error_db": -0.08740991049725753, "output_peak": 0.6094853910024556}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8827643048007756, "全帯域_10ms_rms_envelope_nrmse": 0.09303487376738646, "全帯域_10ms_envelope_correlation": 0.9925543425841898, "全帯域_10ms_modulation_mae_db": 0.4596095749761064, "全帯域_40ms_log_rms_mae_db": 0.46901040070364464, "全帯域_40ms_rms_envelope_nrmse": 0.05078602295310535, "全帯域_40ms_envelope_correlation": 0.9978582993342551, "全帯域_40ms_modulation_mae_db": 0.3078329595937182, "300–1000 Hz_10ms_log_rms_mae_db": 1.2272474494275603, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11878341403825064, "300–1000 Hz_10ms_envelope_correlation": 0.988295846848001, "300–1000 Hz_10ms_modulation_mae_db": 0.4951414102948856, "300–1000 Hz_40ms_log_rms_mae_db": 0.6402010189437547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06556940894880889, "300–1000 Hz_40ms_envelope_correlation": 0.996488340616964, "300–1000 Hz_40ms_modulation_mae_db": 0.35731023391828654, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2202058168075867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08302492740801255, "1000–3000 Hz_10ms_envelope_correlation": 0.9954505040008778, "1000–3000 Hz_10ms_modulation_mae_db": 0.4726418348217556, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6986869818536725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04465693148234409, "1000–3000 Hz_40ms_envelope_correlation": 0.9987742226538487, "1000–3000 Hz_40ms_modulation_mae_db": 0.22185218605817333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2307640877764272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1267022641061771, "3000–6000 Hz_10ms_envelope_correlation": 0.9893905582362928, "3000–6000 Hz_10ms_modulation_mae_db": 1.4387044883110007, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7464458636443945, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06472957904851073, "3000–6000 Hz_40ms_envelope_correlation": 0.9975157734268914, "3000–6000 Hz_40ms_modulation_mae_db": 1.7601684348216236, "mr_stft_db": 5.657952880572669, "band_mae_db": 0.22576486783551966, "rms_gain_error_db": -0.15091378734616678, "output_peak": 0.664692772082876}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8460279288923143, "全帯域_10ms_rms_envelope_nrmse": 0.08724960246335278, "全帯域_10ms_envelope_correlation": 0.9933872908025331, "全帯域_10ms_modulation_mae_db": 0.38954273865275396, "全帯域_40ms_log_rms_mae_db": 0.43005133437038084, "全帯域_40ms_rms_envelope_nrmse": 0.04615119480689045, "全帯域_40ms_envelope_correlation": 0.9980646003691154, "全帯域_40ms_modulation_mae_db": 0.307361551845172, "300–1000 Hz_10ms_log_rms_mae_db": 1.1993180076411454, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11133838730502252, "300–1000 Hz_10ms_envelope_correlation": 0.9898096445826556, "300–1000 Hz_10ms_modulation_mae_db": 0.4801573549639769, "300–1000 Hz_40ms_log_rms_mae_db": 0.5958674045274645, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05961865196503308, "300–1000 Hz_40ms_envelope_correlation": 0.9969512468447498, "300–1000 Hz_40ms_modulation_mae_db": 0.44103670891063995, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2059648300819241, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08334489886591688, "1000–3000 Hz_10ms_envelope_correlation": 0.995316437376609, "1000–3000 Hz_10ms_modulation_mae_db": 0.19292453328915618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7079791290827552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04298730457640802, "1000–3000 Hz_40ms_envelope_correlation": 0.9987229060457026, "1000–3000 Hz_40ms_modulation_mae_db": 0.22360359118547926, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2915287581062806, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1522565434656832, "3000–6000 Hz_10ms_envelope_correlation": 0.9852385789048169, "3000–6000 Hz_10ms_modulation_mae_db": 1.1498930769983084, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8243735854676634, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08181149606367596, "3000–6000 Hz_40ms_envelope_correlation": 0.9968175152901204, "3000–6000 Hz_40ms_modulation_mae_db": 1.3521330199488208, "mr_stft_db": 5.666933514451006, "band_mae_db": 0.1717697078900726, "rms_gain_error_db": 0.022627642057713324, "output_peak": 0.6476598544807259}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.856034310880367, "全帯域_10ms_rms_envelope_nrmse": 0.08377845326014549, "全帯域_10ms_envelope_correlation": 0.9940033391082473, "全帯域_10ms_modulation_mae_db": 0.3713810198381127, "全帯域_40ms_log_rms_mae_db": 0.4640355311677095, "全帯域_40ms_rms_envelope_nrmse": 0.05117596789778587, "全帯域_40ms_envelope_correlation": 0.9978108411792073, "全帯域_40ms_modulation_mae_db": 0.3203341238347772, "300–1000 Hz_10ms_log_rms_mae_db": 1.2116540867104209, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1058417731722331, "300–1000 Hz_10ms_envelope_correlation": 0.9906951171943715, "300–1000 Hz_10ms_modulation_mae_db": 0.2947626175214073, "300–1000 Hz_40ms_log_rms_mae_db": 0.6680379903900848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06403322735264746, "300–1000 Hz_40ms_envelope_correlation": 0.9965363113348339, "300–1000 Hz_40ms_modulation_mae_db": 0.19587223090903927, "1000–3000 Hz_10ms_log_rms_mae_db": 1.145928515536883, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07776564211826995, "1000–3000 Hz_10ms_envelope_correlation": 0.9962532600895813, "1000–3000 Hz_10ms_modulation_mae_db": 0.1919430778946527, "1000–3000 Hz_40ms_log_rms_mae_db": 0.633599916725777, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04458458007612675, "1000–3000 Hz_40ms_envelope_correlation": 0.9990436595619314, "1000–3000 Hz_40ms_modulation_mae_db": 0.2164820440817824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2946618996625157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19291039553015912, "3000–6000 Hz_10ms_envelope_correlation": 0.9757248026933545, "3000–6000 Hz_10ms_modulation_mae_db": 0.6728798592093238, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8099018864705331, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13035326166304712, "3000–6000 Hz_40ms_envelope_correlation": 0.9898639680973268, "3000–6000 Hz_40ms_modulation_mae_db": 0.9530385343513524, "mr_stft_db": 5.6592945015023615, "band_mae_db": 0.29055124667673304, "rms_gain_error_db": -0.14851299125437328, "output_peak": 0.6056579682913297}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.505054609142106, "spectral_convergence": 0.25012733114772245, "spectral_convergence_status": "ok", "amplitude_floor": 1.5171791511760096e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.483190269481903, "spectral_convergence": 0.2516097654476526, "spectral_convergence_status": "ok", "amplitude_floor": 1.2250623067014244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0075230993191004, "spectral_convergence": 0.2581213857621754, "spectral_convergence_status": "ok", "amplitude_floor": 7.883290232655995e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.642531119863568e-05, "output_power": 4.9204057639666266e-05, "delta_db": 0.2524609498846983, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003916996423883643, "output_power": 0.0004344633378948083, "delta_db": 0.4499995942612298, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0020347588099002725, "output_power": 0.0019052029368093282, "delta_db": -0.2857169523215587, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007257385551015356, "output_power": 0.0007032613294410533, "delta_db": -0.13663458416282417, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001922577058777525, "output_power": 0.00018070752337599044, "delta_db": -0.26907521901239917, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.752614415284284e-05, "output_power": 3.563448907000493e-05, "delta_db": -0.2246340630211644, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2697535604439791, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15464584562842754, "pitch": null, "output_rms_dbfs": -24.803723167198747, "output_peak": 0.6061673406405343, "reference_rms_dbfs": -24.64907732157032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.665255992647704}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019822260308892205, "log_rms_mae_db": 0.9070331244899518, "rms_envelope_nrmse": 0.08925065415112678, "envelope_correlation": 0.9931947320826194, "rms_gain_error_db": -0.15461907548811482, "reference_mean_power": 0.003431222806852645, "output_mean_power": 0.003311212191257017}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9226628348981265e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006393448264747715, "output_power": 0.000628916312328528, "reference_active": true, "delta_db": -0.07142295994457813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011197882141156391, "output_power": 0.00010332541916974413, "reference_active": true, "delta_db": -0.3492871638987197}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.279682095777351e-05, "output_power": 3.942842710357503e-05, "reference_active": true, "delta_db": -0.3560205751024399}], "active_band_count": 3, "active_band_mae_db": 0.25891023298191257}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001679501427730559, "log_rms_mae_db": 0.48893491411840206, "rms_envelope_nrmse": 0.0432211549233168, "envelope_correlation": 0.998549630111653, "rms_gain_error_db": -0.15495278249746672, "reference_mean_power": 0.003447933525092644, "output_mean_power": 0.0033270827754726252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8195786138583623e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005865504590263927, "output_power": 0.0005751341753191622, "reference_active": true, "delta_db": -0.08536204217252728}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.169773206547347e-05, "output_power": 5.645019051299032e-05, "reference_active": true, "delta_db": -0.3860378822503744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.232527996839891e-06, "output_power": 2.215272343176454e-06, "reference_active": true, "delta_db": -0.03369788296814047}], "active_band_count": 3, "active_band_mae_db": 0.16836593579701406}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001900450506992256, "log_rms_mae_db": 1.2352831795869204, "rms_envelope_nrmse": 0.11589731826017738, "envelope_correlation": 0.9889677582061948, "rms_gain_error_db": -0.20570918713897235, "reference_mean_power": 0.002194955532035717, "output_mean_power": 0.002093412473246642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4050180992423441e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048796679990628384, "output_power": 0.00046337442998088777, "reference_active": true, "delta_db": -0.22458209858791348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011744031611151056, "output_power": 0.00011117258415900257, "reference_active": true, "delta_db": -0.23819510787328785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.011051692782722e-05, "output_power": 3.879644746670198e-05, "reference_active": true, "delta_db": -0.1446629931455109}], "active_band_count": 3, "active_band_mae_db": 0.20248006653557074}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014767968623269652, "log_rms_mae_db": 0.6341451098662704, "rms_envelope_nrmse": 0.05681970485624069, "envelope_correlation": 0.9975643144872534, "rms_gain_error_db": -0.20584430788581007, "reference_mean_power": 0.002205773735941533, "output_mean_power": 0.0021036647535631614}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315971715150661e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004578823491754882, "output_power": 0.00043223380353155224, "reference_active": true, "delta_db": -0.2503517374472083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0620858906342295e-05, "output_power": 5.8182418809466565e-05, "reference_active": true, "delta_db": -0.17830313225026684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.249804591930109e-06, "output_power": 2.3057338084279567e-06, "reference_active": true, "delta_db": 0.10664368718318203}], "active_band_count": 3, "active_band_mae_db": 0.1784328522935524}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012415579655882008, "log_rms_mae_db": 1.156785012897676, "rms_envelope_nrmse": 0.08226430365737418, "envelope_correlation": 0.9955871514056539, "rms_gain_error_db": -0.16427380817709256, "reference_mean_power": 0.0006561031712122296, "output_mean_power": 0.0006317492806337821}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1091236930101443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020492597233413883, "output_power": 0.00019905690066681468, "reference_active": true, "delta_db": -0.1261976662361473}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1330216505908255e-06, "output_power": 6.891157771499729e-06, "reference_active": true, "delta_db": -0.14981349117776407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.269143835584622e-06, "output_power": 4.68525932527057e-06, "reference_active": true, "delta_db": 0.40392846474855776}], "active_band_count": 3, "active_band_mae_db": 0.22664654072082305}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011294582885045651, "log_rms_mae_db": 0.6512486353772379, "rms_envelope_nrmse": 0.05013326330841049, "envelope_correlation": 0.9984834324134292, "rms_gain_error_db": -0.16535778796303804, "reference_mean_power": 0.000659140146325869, "output_mean_power": 0.0006345151343496889}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0026600561194904e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019159735065259618, "output_power": 0.00018590510838186066, "reference_active": true, "delta_db": -0.1309817582813259}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0478497584910014e-06, "output_power": 2.9794679864152864e-06, "reference_active": true, "delta_db": -0.09854831560482397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0860941276748228e-07, "output_power": 2.307812784046386e-07, "reference_active": true, "delta_db": 0.438666742345343}], "active_band_count": 3, "active_band_mae_db": 0.22273227207716428}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.245162939338534e-05, "log_rms_mae_db": 1.2851516759563157, "rms_envelope_nrmse": 0.1399692449192789, "envelope_correlation": 0.9870851442067952, "rms_gain_error_db": -0.285653864220348, "reference_mean_power": 8.990510589252764e-05, "output_mean_power": 8.418194784809934e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.06772172679864e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1343959051909258e-05, "output_power": 1.0430743281955175e-05, "reference_active": true, "delta_db": -0.364493933560254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.893195477577164e-06, "output_power": 3.055497924147958e-06, "reference_active": true, "delta_db": -1.0522421721361903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973785070466098e-07, "output_power": 3.6035747686512677e-07, "reference_active": true, "delta_db": -3.4493098785211145}], "active_band_count": 3, "active_band_mae_db": 1.6220153280725196}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.719774514443707e-05, "log_rms_mae_db": 0.7767776464078457, "rms_envelope_nrmse": 0.07782324749614578, "envelope_correlation": 0.9963773572035138, "rms_gain_error_db": -0.2856528522172063, "reference_mean_power": 9.035589049479441e-05, "output_mean_power": 8.460405622501652e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9012156385948547e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1053455137857872e-05, "output_power": 9.989313453705224e-06, "reference_active": true, "delta_db": -0.4396241190386038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0905012137138186e-06, "output_power": 1.5566832333624426e-06, "reference_active": true, "delta_db": -1.280501762665771}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.495315305956309e-08, "output_power": 2.9453616404468526e-08, "reference_active": true, "delta_db": -3.4346161174384626}], "active_band_count": 3, "active_band_mae_db": 1.7182473330476125}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8094987767927498, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5ef9d5c255ca7ed71323770db968b44c0c5250c145f5c41dfd7996ac7dcfe17a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9953474344016788, "standard_deviation": 0.0005331525818605351, "minimum": 0.9947069052961627, "maximum": 0.9962532600895813, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1834644510178598, "standard_deviation": 0.02603209668347555, "minimum": 1.145928515536883, "maximum": 1.2202058168075867, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4248568996546853, "standard_deviation": 0.23510494076413918, "minimum": 0.1919430778946527, "maximum": 0.8832492437437885, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08405355516940943, "standard_deviation": 0.004484074936071775, "minimum": 0.07776564211826995, "maximum": 0.08971857073346362, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9985849267103739, "standard_deviation": 0.0003420036043507802, "minimum": 0.9980575717940309, "maximum": 0.9990436595619314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.67493036519149, "standard_deviation": 0.02520853153985896, "minimum": 0.633599916725777, "maximum": 0.7079791290827552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.409974700681635, "standard_deviation": 0.2760669990678974, "minimum": 0.2164820440817824, "maximum": 1.0019959796897642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04772170182353683, "standard_deviation": 0.005667626170418575, "minimum": 0.04002826804734172, "maximum": 0.05737466371737402, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.985449347001643, "standard_deviation": 0.004190848911281958, "minimum": 0.9757248026933545, "maximum": 0.9893905582362928, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2774130080932364, "standard_deviation": 0.02578040161620685, "minimum": 1.2307640877764272, "maximum": 1.3154555380015267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2168751198665428, "standard_deviation": 0.3936746035226781, "minimum": 0.6728798592093238, "maximum": 1.7454531954014953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1505432130458481, "standard_deviation": 0.01943301765559987, "minimum": 0.1267022641061771, "maximum": 0.19291039553015912, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995593945035207, "standard_deviation": 0.002415734118004316, "minimum": 0.9898639680973268, "maximum": 0.9975157734268914, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7949798037404542, "standard_deviation": 0.043171260250977096, "minimum": 0.7464458636443945, "maximum": 0.8810526947366542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4705706817571562, "standard_deviation": 0.4594336407456885, "minimum": 0.7727909150433862, "maximum": 2.2009194967942016, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08966667066169023, "standard_deviation": 0.01894996959803305, "minimum": 0.06472957904851073, "maximum": 0.13035326166304712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9899393215172505, "standard_deviation": 0.0010566879928444251, "minimum": 0.988295846848001, "maximum": 0.9913513009550458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.226448753303051, "standard_deviation": 0.01602300864042148, "minimum": 1.1993180076411454, "maximum": 1.2491347387410756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3440738939265825, "standard_deviation": 0.1298078527450575, "minimum": 0.13184151704615021, "maximum": 0.4951414102948856, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11004955234600565, "standard_deviation": 0.005756966844231726, "minimum": 0.10265513302476051, "maximum": 0.11878341403825064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970354999071791, "standard_deviation": 0.0005617130607789692, "minimum": 0.9962690056552729, "maximum": 0.9977166372426227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6439824091212092, "standard_deviation": 0.032867382797961094, "minimum": 0.5958674045274645, "maximum": 0.6951042923656825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3466406307250338, "standard_deviation": 0.1151260605043364, "minimum": 0.1784328522935524, "maximum": 0.47546701631423743, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05950677985383862, "standard_deviation": 0.005219697661028852, "minimum": 0.05099047324398023, "maximum": 0.06556940894880889, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24460577312893056, "standard_deviation": 0.04444196148308006, "minimum": 0.1717697078900726, "maximum": 0.2969582022173487, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.665840450557731, "standard_deviation": 0.008053884380397123, "minimum": 5.654857500456543, "maximum": 5.676987389936642, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6211385603760633, "standard_deviation": 0.022902394590113232, "minimum": 0.6009120172813521, "maximum": 0.664692772082876, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1070659360471614, "standard_deviation": 0.06333847365879762, "minimum": -0.16910446319146424, "maximum": 0.022627642057713324, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9937127087087416, "standard_deviation": 0.0007407426854488175, "minimum": 0.9925543425841898, "maximum": 0.9950005331864339, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8787078040374073, "standard_deviation": 0.023379224619542636, "minimum": 0.8460279288923143, "maximum": 0.9094491942294597, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3440302935087192, "standard_deviation": 0.07436248122875797, "minimum": 0.2263562126785482, "maximum": 0.4596095749761064, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08531676605172561, "standard_deviation": 0.004846714789477921, "minimum": 0.07731177680516833, "maximum": 0.09303487376738646, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981950234653294, "standard_deviation": 0.0003427946491971654, "minimum": 0.9978108411792073, "maximum": 0.998660396219997, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4712753788381111, "standard_deviation": 0.024163264728393613, "minimum": 0.43005133437038084, "maximum": 0.5084250645285784, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3387408257708514, "standard_deviation": 0.10037954649027328, "minimum": 0.16836593579701406, "maximum": 0.5197920079742838, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04605890823478413, "standard_deviation": 0.003890306297747727, "minimum": 0.040492389949134296, "maximum": 0.05117596789778587, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0019", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.8383374214172363, "predicted_mos": 1.8383374214172363, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6a6582ff75260409ed4bd76b3d5677dd71c2032516e3fd04656e104d2ca4b411", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.115340100135654, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8685489530688243, "全帯域_10ms_rms_envelope_nrmse": 0.08574296074449628, "全帯域_10ms_envelope_correlation": 0.9940051788295778, "全帯域_10ms_modulation_mae_db": 0.39453097361059264, "全帯域_40ms_log_rms_mae_db": 0.4556401103959666, "全帯域_40ms_rms_envelope_nrmse": 0.0487346942961726, "全帯域_40ms_envelope_correlation": 0.9984062228095923, "全帯域_40ms_modulation_mae_db": 0.3388238891981634, "300–1000 Hz_10ms_log_rms_mae_db": 1.23984721539924, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11125518016711038, "300–1000 Hz_10ms_envelope_correlation": 0.9903067083526057, "300–1000 Hz_10ms_modulation_mae_db": 0.3886793739540899, "300–1000 Hz_40ms_log_rms_mae_db": 0.6550702618203614, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06640910900878233, "300–1000 Hz_40ms_envelope_correlation": 0.9970763804401517, "300–1000 Hz_40ms_modulation_mae_db": 0.3300135378538806, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3191541959748487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09811048046515644, "1000–3000 Hz_10ms_envelope_correlation": 0.9935810887351139, "1000–3000 Hz_10ms_modulation_mae_db": 0.47821789019534244, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7533835533689082, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0550063762620815, "1000–3000 Hz_40ms_envelope_correlation": 0.9980695335719889, "1000–3000 Hz_40ms_modulation_mae_db": 0.48442618997178305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5063863067706893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15632479193828933, "3000–6000 Hz_10ms_envelope_correlation": 0.9836644893239529, "3000–6000 Hz_10ms_modulation_mae_db": 0.8560320650345236, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8985170281669503, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07704234553369677, "3000–6000 Hz_40ms_envelope_correlation": 0.9964337689366025, "3000–6000 Hz_40ms_modulation_mae_db": 1.185465200357353, "mr_stft_db": 5.853618522221008, "band_mae_db": 0.2133658149685583, "rms_gain_error_db": -0.2303943800960213, "output_peak": 0.6477984198569923}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8647188531899211, "全帯域_10ms_rms_envelope_nrmse": 0.08807478014543069, "全帯域_10ms_envelope_correlation": 0.993275821163071, "全帯域_10ms_modulation_mae_db": 0.719982140708406, "全帯域_40ms_log_rms_mae_db": 0.4988040166980194, "全帯域_40ms_rms_envelope_nrmse": 0.05541729039272092, "全帯域_40ms_envelope_correlation": 0.9972179963528223, "全帯域_40ms_modulation_mae_db": 0.6031404805629177, "300–1000 Hz_10ms_log_rms_mae_db": 1.2488699730260893, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11445596738741555, "300–1000 Hz_10ms_envelope_correlation": 0.9892334737020534, "300–1000 Hz_10ms_modulation_mae_db": 0.6352113744565115, "300–1000 Hz_40ms_log_rms_mae_db": 0.6794428613621347, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07100890275888926, "300–1000 Hz_40ms_envelope_correlation": 0.9956549143379839, "300–1000 Hz_40ms_modulation_mae_db": 0.5568590290237186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2943598235666927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09576931968841366, "1000–3000 Hz_10ms_envelope_correlation": 0.993824110343602, "1000–3000 Hz_10ms_modulation_mae_db": 0.19464652502838445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7537052586206745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055844230769179416, "1000–3000 Hz_40ms_envelope_correlation": 0.9979159549853109, "1000–3000 Hz_40ms_modulation_mae_db": 0.2581945865132191, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5097385355264041, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15136182408239246, "3000–6000 Hz_10ms_envelope_correlation": 0.9853457149591413, "3000–6000 Hz_10ms_modulation_mae_db": 0.9860004681040174, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9547897390994422, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09773543450076952, "3000–6000 Hz_40ms_envelope_correlation": 0.9946011776083958, "3000–6000 Hz_40ms_modulation_mae_db": 1.2423397700879184, "mr_stft_db": 5.844184900731424, "band_mae_db": 0.2518576813591336, "rms_gain_error_db": 0.00693179255422713, "output_peak": 0.6035081619997098}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8687760232065008, "全帯域_10ms_rms_envelope_nrmse": 0.07917157779218204, "全帯域_10ms_envelope_correlation": 0.9947509061824129, "全帯域_10ms_modulation_mae_db": 0.7489539332759282, "全帯域_40ms_log_rms_mae_db": 0.4743142577933262, "全帯域_40ms_rms_envelope_nrmse": 0.048148793113468216, "全帯域_40ms_envelope_correlation": 0.9981868571161318, "全帯域_40ms_modulation_mae_db": 1.0421892304617941, "300–1000 Hz_10ms_log_rms_mae_db": 1.2630739186334234, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10214178004652855, "300–1000 Hz_10ms_envelope_correlation": 0.9916469789179293, "300–1000 Hz_10ms_modulation_mae_db": 0.7873155049751229, "300–1000 Hz_40ms_log_rms_mae_db": 0.6718388394701443, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0627581115358013, "300–1000 Hz_40ms_envelope_correlation": 0.9970678575201147, "300–1000 Hz_40ms_modulation_mae_db": 1.1084839598120952, "1000–3000 Hz_10ms_log_rms_mae_db": 1.346323825403408, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09572825863212853, "1000–3000 Hz_10ms_envelope_correlation": 0.9938161804912835, "1000–3000 Hz_10ms_modulation_mae_db": 0.17613198406572764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8017709391589308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0587352402434939, "1000–3000 Hz_40ms_envelope_correlation": 0.997654890865095, "1000–3000 Hz_40ms_modulation_mae_db": 0.19143267401991956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5331054723248008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15514669853244603, "3000–6000 Hz_10ms_envelope_correlation": 0.9845629129927873, "3000–6000 Hz_10ms_modulation_mae_db": 0.9440366848428464, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9213873712037871, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09492324731257543, "3000–6000 Hz_40ms_envelope_correlation": 0.9950811281011909, "3000–6000 Hz_40ms_modulation_mae_db": 1.1541772917132769, "mr_stft_db": 5.851499458838362, "band_mae_db": 0.21943896382438258, "rms_gain_error_db": -0.17356809795556846, "output_peak": 0.6438126085480499}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8380087085402957, "全帯域_10ms_rms_envelope_nrmse": 0.08868563738718309, "全帯域_10ms_envelope_correlation": 0.9931643156611981, "全帯域_10ms_modulation_mae_db": 0.33228876818036684, "全帯域_40ms_log_rms_mae_db": 0.468923216683415, "全帯域_40ms_rms_envelope_nrmse": 0.05585741812545611, "全帯域_40ms_envelope_correlation": 0.9972265037122447, "全帯域_40ms_modulation_mae_db": 0.6313115063193804, "300–1000 Hz_10ms_log_rms_mae_db": 1.2041920058570428, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11890154778247589, "300–1000 Hz_10ms_envelope_correlation": 0.9882184200153824, "300–1000 Hz_10ms_modulation_mae_db": 0.2767029181968571, "300–1000 Hz_40ms_log_rms_mae_db": 0.6642121484379043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07746341294291477, "300–1000 Hz_40ms_envelope_correlation": 0.9948433273502648, "300–1000 Hz_40ms_modulation_mae_db": 0.5191362289832462, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3507581692056014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09671104079844178, "1000–3000 Hz_10ms_envelope_correlation": 0.9936725857658655, "1000–3000 Hz_10ms_modulation_mae_db": 0.32664941304704787, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7768666934003874, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04745319932274752, "1000–3000 Hz_40ms_envelope_correlation": 0.9984577845157714, "1000–3000 Hz_40ms_modulation_mae_db": 0.35175298770645164, "3000–6000 Hz_10ms_log_rms_mae_db": 1.516357628309328, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15332933534111254, "3000–6000 Hz_10ms_envelope_correlation": 0.9844861031125087, "3000–6000 Hz_10ms_modulation_mae_db": 0.6660709040473742, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9294026179723169, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08819810053872759, "3000–6000 Hz_40ms_envelope_correlation": 0.9953639626215122, "3000–6000 Hz_40ms_modulation_mae_db": 0.9957613205819166, "mr_stft_db": 5.8610976083755935, "band_mae_db": 0.226753956916785, "rms_gain_error_db": -0.1024030380717908, "output_peak": 0.6290520933255368}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8675785179019623, "全帯域_10ms_rms_envelope_nrmse": 0.08604551707617952, "全帯域_10ms_envelope_correlation": 0.9936041218123505, "全帯域_10ms_modulation_mae_db": 0.33011872435824935, "全帯域_40ms_log_rms_mae_db": 0.44891680916907384, "全帯域_40ms_rms_envelope_nrmse": 0.04823838136717556, "全帯域_40ms_envelope_correlation": 0.9980108880233601, "全帯域_40ms_modulation_mae_db": 0.23522301961617173, "300–1000 Hz_10ms_log_rms_mae_db": 1.2799157942421544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11594840189342545, "300–1000 Hz_10ms_envelope_correlation": 0.988917829234969, "300–1000 Hz_10ms_modulation_mae_db": 0.2843651896117135, "300–1000 Hz_40ms_log_rms_mae_db": 0.6916302030992738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06942742242894269, "300–1000 Hz_40ms_envelope_correlation": 0.9960804067391056, "300–1000 Hz_40ms_modulation_mae_db": 0.25281098509343164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3170354767192094, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09021577762264855, "1000–3000 Hz_10ms_envelope_correlation": 0.9945053083941748, "1000–3000 Hz_10ms_modulation_mae_db": 0.15948664844375035, "1000–3000 Hz_40ms_log_rms_mae_db": 0.785588594717578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05536358758086941, "1000–3000 Hz_40ms_envelope_correlation": 0.9979044283922194, "1000–3000 Hz_40ms_modulation_mae_db": 0.19282743733336336, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5405733759314555, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.154605645799384, "3000–6000 Hz_10ms_envelope_correlation": 0.9841146195170722, "3000–6000 Hz_10ms_modulation_mae_db": 0.9061419756634995, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9292543468782486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08716121590401567, "3000–6000 Hz_40ms_envelope_correlation": 0.9953386252940136, "3000–6000 Hz_40ms_modulation_mae_db": 1.1234879280363985, "mr_stft_db": 5.837887141708614, "band_mae_db": 0.22368804350427696, "rms_gain_error_db": -0.12385088522533845, "output_peak": 0.632121141526252}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8835820318157189, "全帯域_10ms_rms_envelope_nrmse": 0.10196686867241427, "全帯域_10ms_envelope_correlation": 0.9910407159967147, "全帯域_10ms_modulation_mae_db": 0.4883622945657488, "全帯域_40ms_log_rms_mae_db": 0.4653281302365624, "全帯域_40ms_rms_envelope_nrmse": 0.051071083200053984, "全帯域_40ms_envelope_correlation": 0.997856343875841, "全帯域_40ms_modulation_mae_db": 0.41818224178156327, "300–1000 Hz_10ms_log_rms_mae_db": 1.292671571519801, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13237545611939813, "300–1000 Hz_10ms_envelope_correlation": 0.9855008481065487, "300–1000 Hz_10ms_modulation_mae_db": 0.4783536551178907, "300–1000 Hz_40ms_log_rms_mae_db": 0.6746756989873666, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06495654589204203, "300–1000 Hz_40ms_envelope_correlation": 0.9966647491901051, "300–1000 Hz_40ms_modulation_mae_db": 0.35057989071380424, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3186476630938393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09763408420438727, "1000–3000 Hz_10ms_envelope_correlation": 0.9935852411226623, "1000–3000 Hz_10ms_modulation_mae_db": 0.39836761935405857, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7681514704785778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05513402743276056, "1000–3000 Hz_40ms_envelope_correlation": 0.997979601624995, "1000–3000 Hz_40ms_modulation_mae_db": 0.339606775333666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4843018627330515, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14906285930117252, "3000–6000 Hz_10ms_envelope_correlation": 0.9863215323141171, "3000–6000 Hz_10ms_modulation_mae_db": 0.8241757331871051, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8672924300645777, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09095599991574498, "3000–6000 Hz_40ms_envelope_correlation": 0.9961582890831786, "3000–6000 Hz_40ms_modulation_mae_db": 1.0893396588447228, "mr_stft_db": 5.854761331007665, "band_mae_db": 0.320473227477373, "rms_gain_error_db": -0.1579355292416159, "output_peak": 0.6418572973461475}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.85972207376501, "全帯域_10ms_rms_envelope_nrmse": 0.08671192709174239, "全帯域_10ms_envelope_correlation": 0.9934828513207105, "全帯域_10ms_modulation_mae_db": 0.43510319993399477, "全帯域_40ms_log_rms_mae_db": 0.4619222041796358, "全帯域_40ms_rms_envelope_nrmse": 0.04606101178448051, "全帯域_40ms_envelope_correlation": 0.9981714290926267, "全帯域_40ms_modulation_mae_db": 0.2542688754520533, "300–1000 Hz_10ms_log_rms_mae_db": 1.2308198663398788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11388194515180004, "300–1000 Hz_10ms_envelope_correlation": 0.9891764538331821, "300–1000 Hz_10ms_modulation_mae_db": 0.4295623113597234, "300–1000 Hz_40ms_log_rms_mae_db": 0.6235991986789621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06253281410891015, "300–1000 Hz_40ms_envelope_correlation": 0.9966116520685933, "300–1000 Hz_40ms_modulation_mae_db": 0.2493161093306012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2879771566318907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0984488316337018, "1000–3000 Hz_10ms_envelope_correlation": 0.9937437071321796, "1000–3000 Hz_10ms_modulation_mae_db": 0.6266437426300233, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7390143763016613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05909626604751818, "1000–3000 Hz_40ms_envelope_correlation": 0.9980063718940463, "1000–3000 Hz_40ms_modulation_mae_db": 0.802217483119192, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5616248416947647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16721833292708813, "3000–6000 Hz_10ms_envelope_correlation": 0.9815392165924075, "3000–6000 Hz_10ms_modulation_mae_db": 1.3022540308692592, "3000–6000 Hz_40ms_log_rms_mae_db": 0.985585599184533, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1046627132854579, "3000–6000 Hz_40ms_envelope_correlation": 0.9932861428563126, "3000–6000 Hz_40ms_modulation_mae_db": 1.6945860186452653, "mr_stft_db": 5.861211207889858, "band_mae_db": 0.26299837277932464, "rms_gain_error_db": -0.11192954122518387, "output_peak": 0.6219196837373917}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9030178905545975, "全帯域_10ms_rms_envelope_nrmse": 0.08134957111803273, "全帯域_10ms_envelope_correlation": 0.9942468943717547, "全帯域_10ms_modulation_mae_db": 0.3853927382411637, "全帯域_40ms_log_rms_mae_db": 0.48106506180917663, "全帯域_40ms_rms_envelope_nrmse": 0.042417604412587115, "全帯域_40ms_envelope_correlation": 0.9984118868267965, "全帯域_40ms_modulation_mae_db": 0.25251849251058983, "300–1000 Hz_10ms_log_rms_mae_db": 1.284113117094866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10571638794159656, "300–1000 Hz_10ms_envelope_correlation": 0.9906593012694226, "300–1000 Hz_10ms_modulation_mae_db": 0.42347580746352803, "300–1000 Hz_40ms_log_rms_mae_db": 0.6629663026248946, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055034852803843355, "300–1000 Hz_40ms_envelope_correlation": 0.9973597239369113, "300–1000 Hz_40ms_modulation_mae_db": 0.23038184539330117, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3281476425350156, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09412265246146842, "1000–3000 Hz_10ms_envelope_correlation": 0.9940511317939887, "1000–3000 Hz_10ms_modulation_mae_db": 0.4142258700787263, "1000–3000 Hz_40ms_log_rms_mae_db": 0.792048430619213, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053311583853647156, "1000–3000 Hz_40ms_envelope_correlation": 0.9981250376810805, "1000–3000 Hz_40ms_modulation_mae_db": 0.3777039574072234, "3000–6000 Hz_10ms_log_rms_mae_db": 1.497520527514055, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14649297283204718, "3000–6000 Hz_10ms_envelope_correlation": 0.9861877535877299, "3000–6000 Hz_10ms_modulation_mae_db": 0.6939034361918104, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9127063371332534, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08805982551211726, "3000–6000 Hz_40ms_envelope_correlation": 0.9957272203166557, "3000–6000 Hz_40ms_modulation_mae_db": 0.9087893389584737, "mr_stft_db": 5.841467363465917, "band_mae_db": 0.20986371070468202, "rms_gain_error_db": -0.08062286700277127, "output_peak": 0.6266511932825825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.701577487936311, "spectral_convergence": 0.2558768035866869, "spectral_convergence_status": "ok", "amplitude_floor": 1.5171791511760096e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.728170551324493, "spectral_convergence": 0.2462476314008898, "spectral_convergence_status": "ok", "amplitude_floor": 1.2250623067014244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.131107527402218, "spectral_convergence": 0.2556185052717966, "spectral_convergence_status": "ok", "amplitude_floor": 7.883290232655995e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.642531119863568e-05, "output_power": 4.828048294920087e-05, "delta_db": 0.17016782101586803, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003916996423883643, "output_power": 0.00040960123442790027, "delta_db": 0.1940808176802492, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0020347588099002725, "output_power": 0.0018713586113642661, "delta_db": -0.3635591752027972, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007257385551015356, "output_power": 0.0007049051280791671, "delta_db": -0.12649525726228178, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001922577058777525, "output_power": 0.00018098272385201045, "delta_db": -0.2624663556475305, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.752614415284284e-05, "output_power": 3.614027028533432e-05, "delta_db": -0.16342546300262295, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2133658149685583, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2303943800960213, "pitch": null, "output_rms_dbfs": -24.879471701666343, "output_peak": 0.6477984198569923, "reference_rms_dbfs": -24.64907732157032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.853618522221008}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019822260308892205, "log_rms_mae_db": 0.8685489530688243, "rms_envelope_nrmse": 0.08574296074449628, "envelope_correlation": 0.9940051788295778, "rms_gain_error_db": -0.23038866163166072, "reference_mean_power": 0.003431222806852645, "output_mean_power": 0.003253943844636611}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9226628348981265e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006393448264747715, "output_power": 0.0005990459465095258, "reference_active": true, "delta_db": -0.28275021426652897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011197882141156391, "output_power": 0.00010356456054351095, "reference_active": true, "delta_db": -0.33924725520719484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.279682095777351e-05, "output_power": 3.760555172220101e-05, "reference_active": true, "delta_db": -0.561595451358054}], "active_band_count": 3, "active_band_mae_db": 0.39453097361059264}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001679501427730559, "log_rms_mae_db": 0.4556401103959666, "rms_envelope_nrmse": 0.0487346942961726, "envelope_correlation": 0.9984062228095923, "rms_gain_error_db": -0.23043363545535211, "reference_mean_power": 0.003447933525092644, "output_mean_power": 0.0032697573195744393}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8195786138583623e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005865504590263927, "output_power": 0.0005497703600431923, "reference_active": true, "delta_db": -0.28124057126864194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.169773206547347e-05, "output_power": 5.665363938731454e-05, "reference_active": true, "delta_db": -0.3704138633568635}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.232527996839891e-06, "output_power": 2.05265135630039e-06, "reference_active": true, "delta_db": -0.36481723296898483}], "active_band_count": 3, "active_band_mae_db": 0.3388238891981634}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001900450506992256, "log_rms_mae_db": 1.23984721539924, "rms_envelope_nrmse": 0.11125518016711038, "envelope_correlation": 0.9903067083526057, "rms_gain_error_db": -0.30766319802351755, "reference_mean_power": 0.002194955532035717, "output_mean_power": 0.002044840350485597}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4050180992423441e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048796679990628384, "output_power": 0.0004475007307522405, "reference_active": true, "delta_db": -0.3759652581083529}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011744031611151056, "output_power": 0.00010622506359893182, "reference_active": true, "delta_db": -0.4359021177259712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.011051692782722e-05, "output_power": 3.6969296337716476e-05, "reference_active": true, "delta_db": -0.3541707460279456}], "active_band_count": 3, "active_band_mae_db": 0.3886793739540899}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014767968623269652, "log_rms_mae_db": 0.6550702618203614, "rms_envelope_nrmse": 0.06640910900878233, "envelope_correlation": 0.9970763804401517, "rms_gain_error_db": -0.30774706441545263, "reference_mean_power": 0.002205773735941533, "output_mean_power": 0.002054879004668052}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315971715150661e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004578823491754882, "output_power": 0.00041951821412558664, "reference_active": true, "delta_db": -0.38003081072868095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0620858906342295e-05, "output_power": 5.555881966287396e-05, "reference_active": true, "delta_db": -0.3786907471896695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.249804591930109e-06, "output_power": 2.133108239697325e-06, "reference_active": true, "delta_db": -0.23131905564329142}], "active_band_count": 3, "active_band_mae_db": 0.3300135378538806}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012415579655882008, "log_rms_mae_db": 1.3191541959748487, "rms_envelope_nrmse": 0.09811048046515644, "envelope_correlation": 0.9935810887351139, "rms_gain_error_db": -0.13545232840558616, "reference_mean_power": 0.0006561031712122296, "output_mean_power": 0.0006359557583307594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1091236930101443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020492597233413883, "output_power": 0.00019361099141729072, "reference_active": true, "delta_db": -0.24666995564846642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1330216505908255e-06, "output_power": 5.498681747061956e-06, "reference_active": true, "delta_db": -1.1301495792568723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.269143835584622e-06, "output_power": 4.326375403142548e-06, "reference_active": true, "delta_db": 0.05783413568068872}], "active_band_count": 3, "active_band_mae_db": 0.47821789019534244}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011294582885045651, "log_rms_mae_db": 0.7533835533689082, "rms_envelope_nrmse": 0.0550063762620815, "envelope_correlation": 0.9980695335719889, "rms_gain_error_db": -0.13533201853598908, "reference_mean_power": 0.000659140146325869, "output_mean_power": 0.0006389171742292332}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0026600561194904e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019159735065259618, "output_power": 0.00018026661204306944, "reference_active": true, "delta_db": -0.264742028817462}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0478497584910014e-06, "output_power": 2.5275281497884465e-06, "reference_active": true, "delta_db": -0.8129755381940897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0860941276748228e-07, "output_power": 2.2745212615295522e-07, "reference_active": true, "delta_db": 0.37556100290379735}], "active_band_count": 3, "active_band_mae_db": 0.48442618997178305}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.245162939338534e-05, "log_rms_mae_db": 1.5063863067706893, "rms_envelope_nrmse": 0.15632479193828933, "envelope_correlation": 0.9836644893239529, "rms_gain_error_db": -0.23436915842194708, "reference_mean_power": 8.990510589252764e-05, "output_mean_power": 8.518192317918664e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.06772172679864e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1343959051909258e-05, "output_power": 1.0563896710968144e-05, "reference_active": true, "delta_db": -0.3094050395713176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.893195477577164e-06, "output_power": 3.6038644480204195e-06, "reference_active": true, "delta_db": -0.33537762700629814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973785070466098e-07, "output_power": 5.120745269002457e-07, "reference_active": true, "delta_db": -1.923313528525955}], "active_band_count": 3, "active_band_mae_db": 0.8560320650345236}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.719774514443707e-05, "log_rms_mae_db": 0.8985170281669503, "rms_envelope_nrmse": 0.07704234553369677, "envelope_correlation": 0.9964337689366025, "rms_gain_error_db": -0.23439207410943494, "reference_mean_power": 9.035589049479441e-05, "output_mean_power": 8.560857400466978e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9012156385948547e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1053455137857872e-05, "output_power": 1.0194568220611525e-05, "reference_active": true, "delta_db": -0.3512921647844953}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0905012137138186e-06, "output_power": 1.91906677698536e-06, "reference_active": true, "delta_db": -0.37160337064724497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.495315305956309e-08, "output_power": 3.3825967835685056e-08, "reference_active": true, "delta_db": -2.833500065640319}], "active_band_count": 3, "active_band_mae_db": 1.185465200357353}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7829930305467129, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7967f0c7e13d34134b3b67452f1cb53829bd284e20746ac90e04f1fbcdf19e96", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9938474192223588, "standard_deviation": 0.00030632110974696393, "minimum": 0.9935810887351139, "maximum": 0.9945053083941748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3203004941413132, "standard_deviation": 0.022035309990185135, "minimum": 1.2879771566318907, "maximum": 1.3507581692056014, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3467962116053826, "standard_deviation": 0.16499317682608206, "minimum": 0.15948664844375035, "maximum": 0.6266437426300233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09584255568829331, "standard_deviation": 0.0026869293080722125, "minimum": 0.09021577762264855, "maximum": 0.0984488316337018, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9980142004413135, "standard_deviation": 0.0002280906111849134, "minimum": 0.997654890865095, "maximum": 0.9984577845157714, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7713161645832414, "standard_deviation": 0.02163652385033488, "minimum": 0.7390143763016613, "maximum": 0.8017709391589308, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3747702614256023, "standard_deviation": 0.19885966781709588, "minimum": 0.19143267401991956, "maximum": 0.802217483119192, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054993063939037204, "standard_deviation": 0.003613445237929722, "minimum": 0.04745319932274752, "maximum": 0.05909626604751818, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9845277927999646, "standard_deviation": 0.0015349514733964324, "minimum": 0.9815392165924075, "maximum": 0.9863215323141171, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5187010688505687, "standard_deviation": 0.025064852245820054, "minimum": 1.4843018627330515, "maximum": 1.5616248416947647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8973269122425545, "standard_deviation": 0.19856819283704943, "minimum": 0.6660709040473742, "maximum": 1.3022540308692592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15419280759424153, "standard_deviation": 0.006206608086506905, "minimum": 0.14649297283204718, "maximum": 0.16721833292708813, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9952487893522327, "standard_deviation": 0.0009842394072784473, "minimum": 0.9932861428563126, "maximum": 0.9964337689366025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9248669337128886, "standard_deviation": 0.035438237826837436, "minimum": 0.8672924300645777, "maximum": 0.985585599184533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1742433159031656, "standard_deviation": 0.2354414116675292, "minimum": 0.9087893389584737, "maximum": 1.6945860186452653, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09109236031288814, "standard_deviation": 0.008223737614328297, "minimum": 0.07704234553369677, "maximum": 0.1046627132854579, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9892075016790116, "standard_deviation": 0.0018543438748870374, "minimum": 0.9855008481065487, "maximum": 0.9916469789179293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2554379327640621, "standard_deviation": 0.030200489154578444, "minimum": 1.2041920058570428, "maximum": 1.292671571519801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.46295826689192965, "standard_deviation": 0.1731612871184701, "minimum": 0.2767029181968571, "maximum": 0.7873155049751229, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11433458331121882, "standard_deviation": 0.009114938669024799, "minimum": 0.10214178004652855, "maximum": 0.13237545611939813, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964198764479038, "standard_deviation": 0.0008465766848237029, "minimum": 0.9948433273502648, "maximum": 0.9973597239369113, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6654294393101302, "standard_deviation": 0.020262774376730137, "minimum": 0.6235991986789621, "maximum": 0.6916302030992738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44969769827550987, "standard_deviation": 0.2931284065575492, "minimum": 0.23038184539330117, "maximum": 1.1084839598120952, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06619889643501572, "standard_deviation": 0.006675665058648872, "minimum": 0.055034852803843355, "maximum": 0.07746341294291477, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2410549714418145, "standard_deviation": 0.03706834598003331, "minimum": 0.20986371070468202, "maximum": 0.320473227477373, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.850715941779805, "standard_deviation": 0.008751346569862276, "minimum": 5.837887141708614, "maximum": 5.861211207889858, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6308400749528327, "standard_deviation": 0.01427406671283512, "minimum": 0.6035081619997098, "maximum": 0.6477984198569923, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12172156828300786, "standard_deviation": 0.07022790199511425, "minimum": -0.2303943800960213, "maximum": 0.00693179255422713, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9934463506672238, "standard_deviation": 0.001106923051037021, "minimum": 0.9910407159967147, "maximum": 0.9947509061824129, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8692441315053538, "standard_deviation": 0.018656742281808907, "minimum": 0.8380087085402957, "maximum": 0.9030178905545975, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4793415966093063, "standard_deviation": 0.16585205196761696, "minimum": 0.33011872435824935, "maximum": 0.7489539332759282, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08721860500345763, "standard_deviation": 0.006797132700057377, "minimum": 0.07917157779218204, "maximum": 0.10196686867241427, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997936015976177, "standard_deviation": 0.00047761332907402996, "minimum": 0.9972179963528223, "maximum": 0.9984118868267965, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46936422587064697, "standard_deviation": 0.015613318952125848, "minimum": 0.44891680916907384, "maximum": 0.4988040166980194, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4719572169878292, "standard_deviation": 0.2777378687935621, "minimum": 0.23522301961617173, "maximum": 1.0421892304617941, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04949328458651438, "standard_deviation": 0.004533827366258316, "minimum": 0.042417604412587115, "maximum": 0.05585741812545611, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0019", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.824890375137329, "predicted_mos": 1.824890375137329, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9a3c49f3c29917ed1ba6ccbaf7519c5de71138cb1739198be70a4ae63f9ebade", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.192912200000137, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8661758970299637, "全帯域_10ms_rms_envelope_nrmse": 0.09273273697327664, "全帯域_10ms_envelope_correlation": 0.9924981387353223, "全帯域_10ms_modulation_mae_db": 0.11170075939123521, "全帯域_40ms_log_rms_mae_db": 0.48016090937890193, "全帯域_40ms_rms_envelope_nrmse": 0.04837230368536825, "全帯域_40ms_envelope_correlation": 0.9979164344733537, "全帯域_40ms_modulation_mae_db": 0.1533992257338231, "300–1000 Hz_10ms_log_rms_mae_db": 1.2194937964409203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12314435846399115, "300–1000 Hz_10ms_envelope_correlation": 0.9873081048439736, "300–1000 Hz_10ms_modulation_mae_db": 0.15838350245867405, "300–1000 Hz_40ms_log_rms_mae_db": 0.6966795010859076, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06463256946437604, "300–1000 Hz_40ms_envelope_correlation": 0.9963903428411837, "300–1000 Hz_40ms_modulation_mae_db": 0.1365857728444302, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1334888187517043, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0792094482174135, "1000–3000 Hz_10ms_envelope_correlation": 0.9957762336408832, "1000–3000 Hz_10ms_modulation_mae_db": 0.24690846549377296, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6712310576753001, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04517904898963157, "1000–3000 Hz_40ms_envelope_correlation": 0.9986150917959814, "1000–3000 Hz_40ms_modulation_mae_db": 0.27802001203574767, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2064482967926067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1246724014321839, "3000–6000 Hz_10ms_envelope_correlation": 0.9904740012911332, "3000–6000 Hz_10ms_modulation_mae_db": 0.442933167322209, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7348442584967989, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07511360018249798, "3000–6000 Hz_40ms_envelope_correlation": 0.9973927818592363, "3000–6000 Hz_40ms_modulation_mae_db": 0.6231930390645958, "mr_stft_db": 5.6441361324908, "band_mae_db": 0.24885556437782053, "rms_gain_error_db": -0.0848261556427091, "output_peak": 0.6431928069110918}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8859090541516671, "全帯域_10ms_rms_envelope_nrmse": 0.09819621418301473, "全帯域_10ms_envelope_correlation": 0.9916843326664214, "全帯域_10ms_modulation_mae_db": 0.628607466264171, "全帯域_40ms_log_rms_mae_db": 0.4878500948184168, "全帯域_40ms_rms_envelope_nrmse": 0.05468883301378379, "全帯域_40ms_envelope_correlation": 0.9974892247421551, "全帯域_40ms_modulation_mae_db": 0.6471696417390762, "300–1000 Hz_10ms_log_rms_mae_db": 1.2079416009309023, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1255037695308177, "300–1000 Hz_10ms_envelope_correlation": 0.9869331726301958, "300–1000 Hz_10ms_modulation_mae_db": 0.676145457665592, "300–1000 Hz_40ms_log_rms_mae_db": 0.6472138379098763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06876148470374804, "300–1000 Hz_40ms_envelope_correlation": 0.9961493494853921, "300–1000 Hz_40ms_modulation_mae_db": 0.7097113453651048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1175150277181498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0924236583471387, "1000–3000 Hz_10ms_envelope_correlation": 0.994372522264504, "1000–3000 Hz_10ms_modulation_mae_db": 0.4321223863821044, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6795022733346264, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060182223762016165, "1000–3000 Hz_40ms_envelope_correlation": 0.9977086487683348, "1000–3000 Hz_40ms_modulation_mae_db": 0.40599523874133264, "3000–6000 Hz_10ms_log_rms_mae_db": 1.244901062666548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1375714487561489, "3000–6000 Hz_10ms_envelope_correlation": 0.9875840328052176, "3000–6000 Hz_10ms_modulation_mae_db": 0.39854422147553725, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7592105842408298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08028074095945952, "3000–6000 Hz_40ms_envelope_correlation": 0.9961459393218343, "3000–6000 Hz_40ms_modulation_mae_db": 0.6716772745021727, "mr_stft_db": 5.634769470324792, "band_mae_db": 0.240367435928221, "rms_gain_error_db": -0.1545746348043944, "output_peak": 0.6139797563906257}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8871798144610432, "全帯域_10ms_rms_envelope_nrmse": 0.08537043461013415, "全帯域_10ms_envelope_correlation": 0.993655509129154, "全帯域_10ms_modulation_mae_db": 0.25060903962738607, "全帯域_40ms_log_rms_mae_db": 0.4726548021328606, "全帯域_40ms_rms_envelope_nrmse": 0.04380278409536015, "全帯域_40ms_envelope_correlation": 0.9983062338744034, "全帯域_40ms_modulation_mae_db": 0.3411364266884822, "300–1000 Hz_10ms_log_rms_mae_db": 1.2224511934076956, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10840753254948815, "300–1000 Hz_10ms_envelope_correlation": 0.9902071621903262, "300–1000 Hz_10ms_modulation_mae_db": 0.27586609736261725, "300–1000 Hz_40ms_log_rms_mae_db": 0.6529688774064054, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05593746533853069, "300–1000 Hz_40ms_envelope_correlation": 0.9973111030949849, "300–1000 Hz_40ms_modulation_mae_db": 0.35802307561737906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1110884061601323, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08036271819950165, "1000–3000 Hz_10ms_envelope_correlation": 0.9956524507063307, "1000–3000 Hz_10ms_modulation_mae_db": 0.44722386961237515, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6520054641088899, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.045173618682869406, "1000–3000 Hz_40ms_envelope_correlation": 0.9986290691113876, "1000–3000 Hz_40ms_modulation_mae_db": 0.38156589223282866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.244381794802706, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14567635722025274, "3000–6000 Hz_10ms_envelope_correlation": 0.987524519561057, "3000–6000 Hz_10ms_modulation_mae_db": 1.2257012229091346, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8063572463241673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09420498898647131, "3000–6000 Hz_40ms_envelope_correlation": 0.996283994062419, "3000–6000 Hz_40ms_modulation_mae_db": 1.4490209002406658, "mr_stft_db": 5.652614390991005, "band_mae_db": 0.256885727982219, "rms_gain_error_db": -0.08530485383275344, "output_peak": 0.613094045758018}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8880176014673954, "全帯域_10ms_rms_envelope_nrmse": 0.09016663926769754, "全帯域_10ms_envelope_correlation": 0.9931063915217945, "全帯域_10ms_modulation_mae_db": 0.5168477553672307, "全帯域_40ms_log_rms_mae_db": 0.4983869093978906, "全帯域_40ms_rms_envelope_nrmse": 0.04959365176220637, "全帯域_40ms_envelope_correlation": 0.9980724064308701, "全帯域_40ms_modulation_mae_db": 0.18559929105075978, "300–1000 Hz_10ms_log_rms_mae_db": 1.2213322430352147, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1169155514371164, "300–1000 Hz_10ms_envelope_correlation": 0.9886339784796424, "300–1000 Hz_10ms_modulation_mae_db": 0.5360757569149389, "300–1000 Hz_40ms_log_rms_mae_db": 0.7029926493140226, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06460836151134168, "300–1000 Hz_40ms_envelope_correlation": 0.9965033060706432, "300–1000 Hz_40ms_modulation_mae_db": 0.18862715146312162, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1684623576958615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0881993548200285, "1000–3000 Hz_10ms_envelope_correlation": 0.9953021233565947, "1000–3000 Hz_10ms_modulation_mae_db": 0.30463176629785416, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6996799278904781, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06130749137200632, "1000–3000 Hz_40ms_envelope_correlation": 0.9980727551958769, "1000–3000 Hz_40ms_modulation_mae_db": 0.35020737844524463, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2216928898179853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1372954357526695, "3000–6000 Hz_10ms_envelope_correlation": 0.9876475935977052, "3000–6000 Hz_10ms_modulation_mae_db": 0.33417113772143914, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7616760371008531, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09470057538384191, "3000–6000 Hz_40ms_envelope_correlation": 0.9943335884294956, "3000–6000 Hz_40ms_modulation_mae_db": 0.3312187850905724, "mr_stft_db": 5.646417680154258, "band_mae_db": 0.2286604499966237, "rms_gain_error_db": -0.17825351986869423, "output_peak": 0.5915463572080061}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8560786611003788, "全帯域_10ms_rms_envelope_nrmse": 0.08518192142037202, "全帯域_10ms_envelope_correlation": 0.9936526422124998, "全帯域_10ms_modulation_mae_db": 0.1897400357394928, "全帯域_40ms_log_rms_mae_db": 0.47576216678936706, "全帯域_40ms_rms_envelope_nrmse": 0.04356854096946552, "全帯域_40ms_envelope_correlation": 0.9982702861600986, "全帯域_40ms_modulation_mae_db": 0.28571993758365466, "300–1000 Hz_10ms_log_rms_mae_db": 1.185436696601106, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11110938963889637, "300–1000 Hz_10ms_envelope_correlation": 0.9896930953851344, "300–1000 Hz_10ms_modulation_mae_db": 0.2654868288003917, "300–1000 Hz_40ms_log_rms_mae_db": 0.6663379798833995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05745410639505641, "300–1000 Hz_40ms_envelope_correlation": 0.9971204693834291, "300–1000 Hz_40ms_modulation_mae_db": 0.22845715514889664, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1813940464940664, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08462324544849828, "1000–3000 Hz_10ms_envelope_correlation": 0.9951740616212434, "1000–3000 Hz_10ms_modulation_mae_db": 0.23525340262376945, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6893753570146418, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048466742755081615, "1000–3000 Hz_40ms_envelope_correlation": 0.9984192673931612, "1000–3000 Hz_40ms_modulation_mae_db": 0.23925113267653741, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2099449059008733, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15981649885019605, "3000–6000 Hz_10ms_envelope_correlation": 0.9828633860846875, "3000–6000 Hz_10ms_modulation_mae_db": 0.40766070796692194, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7363001866443025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08691099568046372, "3000–6000 Hz_40ms_envelope_correlation": 0.9947437359872909, "3000–6000 Hz_40ms_modulation_mae_db": 0.6093912241400353, "mr_stft_db": 5.636027887133115, "band_mae_db": 0.14575974680054524, "rms_gain_error_db": -0.03415491074502989, "output_peak": 0.6029000193087222}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8650395799326781, "全帯域_10ms_rms_envelope_nrmse": 0.08346391035064128, "全帯域_10ms_envelope_correlation": 0.9939481007026821, "全帯域_10ms_modulation_mae_db": 0.40723358370163637, "全帯域_40ms_log_rms_mae_db": 0.45024943581911403, "全帯域_40ms_rms_envelope_nrmse": 0.045771382730394355, "全帯域_40ms_envelope_correlation": 0.9981676148839455, "全帯域_40ms_modulation_mae_db": 0.34029978890699764, "300–1000 Hz_10ms_log_rms_mae_db": 1.2198467938605084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10977799265792397, "300–1000 Hz_10ms_envelope_correlation": 0.9899061016063533, "300–1000 Hz_10ms_modulation_mae_db": 0.5834846327956998, "300–1000 Hz_40ms_log_rms_mae_db": 0.6335969287536072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05942352728085626, "300–1000 Hz_40ms_envelope_correlation": 0.9969291663006253, "300–1000 Hz_40ms_modulation_mae_db": 0.28960807428332813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1239589287159215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08888094249034988, "1000–3000 Hz_10ms_envelope_correlation": 0.9948046185643385, "1000–3000 Hz_10ms_modulation_mae_db": 0.3329345207927235, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6770329100497569, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05143499882094686, "1000–3000 Hz_40ms_envelope_correlation": 0.9983901685762245, "1000–3000 Hz_40ms_modulation_mae_db": 0.303927299004097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2927451081085388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1550840170169211, "3000–6000 Hz_10ms_envelope_correlation": 0.984845335010488, "3000–6000 Hz_10ms_modulation_mae_db": 0.8787507016753443, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8035996407044238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09497391447817059, "3000–6000 Hz_40ms_envelope_correlation": 0.9953874513687375, "3000–6000 Hz_40ms_modulation_mae_db": 1.12039741068157, "mr_stft_db": 5.653190901258987, "band_mae_db": 0.24730999692370256, "rms_gain_error_db": -0.09998859935545165, "output_peak": 0.588729878662843}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8645468043689604, "全帯域_10ms_rms_envelope_nrmse": 0.0895947110689662, "全帯域_10ms_envelope_correlation": 0.9931407863924302, "全帯域_10ms_modulation_mae_db": 0.15598930051810345, "全帯域_40ms_log_rms_mae_db": 0.48223518604113635, "全帯域_40ms_rms_envelope_nrmse": 0.04637752962634533, "全帯域_40ms_envelope_correlation": 0.99830011041491, "全帯域_40ms_modulation_mae_db": 0.3440992233332902, "300–1000 Hz_10ms_log_rms_mae_db": 1.1323198039890303, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11065092003554301, "300–1000 Hz_10ms_envelope_correlation": 0.9898815751312254, "300–1000 Hz_10ms_modulation_mae_db": 0.16208899926900652, "300–1000 Hz_40ms_log_rms_mae_db": 0.6201774306314869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05602994753196465, "300–1000 Hz_40ms_envelope_correlation": 0.9975241804749604, "300–1000 Hz_40ms_modulation_mae_db": 0.29064858699989893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1549257206078605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08939791418040831, "1000–3000 Hz_10ms_envelope_correlation": 0.9947326452155176, "1000–3000 Hz_10ms_modulation_mae_db": 0.47323796818257313, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7151634606254328, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054523925008856715, "1000–3000 Hz_40ms_envelope_correlation": 0.9981374794075939, "1000–3000 Hz_40ms_modulation_mae_db": 0.433572447310073, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2089329015895325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13997245803437325, "3000–6000 Hz_10ms_envelope_correlation": 0.988847004808378, "3000–6000 Hz_10ms_modulation_mae_db": 1.2068214309150016, "3000–6000 Hz_40ms_log_rms_mae_db": 0.770524265626232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09589621844299802, "3000–6000 Hz_40ms_envelope_correlation": 0.996214743357573, "3000–6000 Hz_40ms_modulation_mae_db": 1.526410113997109, "mr_stft_db": 5.639140212204847, "band_mae_db": 0.32064425748853315, "rms_gain_error_db": -0.16178677973268776, "output_peak": 0.5934524867380149}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8682468391380181, "全帯域_10ms_rms_envelope_nrmse": 0.08819258928318409, "全帯域_10ms_envelope_correlation": 0.9932268350163733, "全帯域_10ms_modulation_mae_db": 0.586458285080222, "全帯域_40ms_log_rms_mae_db": 0.45659647916081986, "全帯域_40ms_rms_envelope_nrmse": 0.04956485698146646, "全帯域_40ms_envelope_correlation": 0.9978192986639806, "全帯域_40ms_modulation_mae_db": 0.7323290381859651, "300–1000 Hz_10ms_log_rms_mae_db": 1.2194056292461415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11477608733917101, "300–1000 Hz_10ms_envelope_correlation": 0.9890056971786556, "300–1000 Hz_10ms_modulation_mae_db": 0.6251710775940845, "300–1000 Hz_40ms_log_rms_mae_db": 0.6569750507513431, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06546275304982538, "300–1000 Hz_40ms_envelope_correlation": 0.9963209694932558, "300–1000 Hz_40ms_modulation_mae_db": 0.7176084370675929, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1079069600643074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08613169423690105, "1000–3000 Hz_10ms_envelope_correlation": 0.994989674711511, "1000–3000 Hz_10ms_modulation_mae_db": 0.18714008503133236, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6006358374744251, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04268474551536214, "1000–3000 Hz_40ms_envelope_correlation": 0.9987400415561851, "1000–3000 Hz_40ms_modulation_mae_db": 0.15496953433300445, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2325802453216155, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14572245188393151, "3000–6000 Hz_10ms_envelope_correlation": 0.9885676862108227, "3000–6000 Hz_10ms_modulation_mae_db": 1.3911065156273166, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7740492040697662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10199490591080355, "3000–6000 Hz_40ms_envelope_correlation": 0.996360343943369, "3000–6000 Hz_40ms_modulation_mae_db": 1.6465896078090558, "mr_stft_db": 5.643872144801065, "band_mae_db": 0.3100397628206675, "rms_gain_error_db": -0.08635616450726953, "output_peak": 0.634440601589322}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.416824120823677, "spectral_convergence": 0.25210616279318737, "spectral_convergence_status": "ok", "amplitude_floor": 1.5171791511760096e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.44859483299481, "spectral_convergence": 0.24823838430796108, "spectral_convergence_status": "ok", "amplitude_floor": 1.2250623067014244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.066989443653915, "spectral_convergence": 0.2579511362204596, "spectral_convergence_status": "ok", "amplitude_floor": 7.883290232655995e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.642531119863568e-05, "output_power": 4.9711502036805274e-05, "delta_db": 0.2970206198903845, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003916996423883643, "output_power": 0.00042206764064641974, "delta_db": 0.3242888161341002, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0020347588099002725, "output_power": 0.001954362144986402, "delta_db": -0.17507895542500898, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007257385551015356, "output_power": 0.0007233215611847999, "delta_db": -0.014487847432874689, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001922577058777525, "output_power": 0.0001779876813421645, "delta_db": -0.33493810262155543, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.752614415284284e-05, "output_power": 3.464192667071388e-05, "delta_db": -0.3473190447629996, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24885556437782053, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0848261556427091, "pitch": null, "output_rms_dbfs": -24.73390347721303, "output_peak": 0.6431928069110918, "reference_rms_dbfs": -24.64907732157032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6441361324908}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019822260308892205, "log_rms_mae_db": 0.8661758970299637, "rms_envelope_nrmse": 0.09273273697327664, "envelope_correlation": 0.9924981387353223, "rms_gain_error_db": -0.08481309291050147, "reference_mean_power": 0.003431222806852645, "output_mean_power": 0.0033648647353108087}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9226628348981265e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006393448264747715, "output_power": 0.0006297992620216591, "reference_active": true, "delta_db": -0.06533007843529938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011197882141156391, "output_power": 0.00011177920722647903, "reference_active": true, "delta_db": -0.0077486705274549495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.279682095777351e-05, "output_power": 4.5458370377068056e-05, "reference_active": true, "delta_db": 0.2620235292109513}], "active_band_count": 3, "active_band_mae_db": 0.11170075939123521}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001679501427730559, "log_rms_mae_db": 0.48016090937890193, "rms_envelope_nrmse": 0.04837230368536825, "envelope_correlation": 0.9979164344733537, "rms_gain_error_db": -0.08472437961011497, "reference_mean_power": 0.003447933525092644, "output_mean_power": 0.0033813213464366936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8195786138583623e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005865504590263927, "output_power": 0.0005839264500363312, "reference_active": true, "delta_db": -0.01947231124596684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.169773206547347e-05, "output_power": 6.06600070171002e-05, "reference_active": true, "delta_db": -0.07366743969636046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.232527996839891e-06, "output_power": 2.051592587188123e-06, "reference_active": true, "delta_db": -0.367057926259142}], "active_band_count": 3, "active_band_mae_db": 0.1533992257338231}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001900450506992256, "log_rms_mae_db": 1.2194937964409203, "rms_envelope_nrmse": 0.12314435846399115, "envelope_correlation": 0.9873081048439736, "rms_gain_error_db": -0.10795600026193272, "reference_mean_power": 0.002194955532035717, "output_mean_power": 0.002141066351197211}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4050180992423441e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048796679990628384, "output_power": 0.00048154044434485597, "reference_active": true, "delta_db": -0.057575054780895876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011744031611151056, "output_power": 0.00011975492215087169, "reference_active": true, "delta_db": 0.0847616121299112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.011051692782722e-05, "output_power": 4.3305159555400894e-05, "reference_active": true, "delta_db": 0.33281384046521506}], "active_band_count": 3, "active_band_mae_db": 0.15838350245867405}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014767968623269652, "log_rms_mae_db": 0.6966795010859076, "rms_envelope_nrmse": 0.06463256946437604, "envelope_correlation": 0.9963903428411837, "rms_gain_error_db": -0.10791165320456955, "reference_mean_power": 0.002205773735941533, "output_mean_power": 0.002151640924211623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315971715150661e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004578823491754882, "output_power": 0.00045774038098457477, "reference_active": true, "delta_db": -0.0013467556097652624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0620858906342295e-05, "output_power": 6.209923021779706e-05, "reference_active": true, "delta_db": 0.10464131320100441}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.249804591930109e-06, "output_power": 2.097818320705034e-06, "reference_active": true, "delta_db": -0.3037692497225209}], "active_band_count": 3, "active_band_mae_db": 0.1365857728444302}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012415579655882008, "log_rms_mae_db": 1.1334888187517043, "rms_envelope_nrmse": 0.0792094482174135, "envelope_correlation": 0.9957762336408832, "rms_gain_error_db": -0.06240630980378041, "reference_mean_power": 0.0006561031712122296, "output_mean_power": 0.0006467426560886661}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1091236930101443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020492597233413883, "output_power": 0.00019744562584569485, "reference_active": true, "delta_db": -0.1614948746438417}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1330216505908255e-06, "output_power": 7.199404726640506e-06, "reference_active": true, "delta_db": 0.04023046624467966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.269143835584622e-06, "output_power": 4.833267062998131e-06, "reference_active": true, "delta_db": 0.5390000555927975}], "active_band_count": 3, "active_band_mae_db": 0.24690846549377296}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011294582885045651, "log_rms_mae_db": 0.6712310576753001, "rms_envelope_nrmse": 0.04517904898963157, "envelope_correlation": 0.9986150917959814, "rms_gain_error_db": -0.06210972689322496, "reference_mean_power": 0.000659140146325869, "output_mean_power": 0.0006497806756693781}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0026600561194904e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019159735065259618, "output_power": 0.00018463479173899978, "reference_active": true, "delta_db": -0.1607595862274845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0478497584910014e-06, "output_power": 3.161169145418368e-06, "reference_active": true, "delta_db": 0.15854179389347006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0860941276748228e-07, "output_power": 2.3486038374132526e-07, "reference_active": true, "delta_db": 0.5147586559862884}], "active_band_count": 3, "active_band_mae_db": 0.27802001203574767}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.245162939338534e-05, "log_rms_mae_db": 1.2064482967926067, "rms_envelope_nrmse": 0.1246724014321839, "envelope_correlation": 0.9904740012911332, "rms_gain_error_db": -0.39194892761108036, "reference_mean_power": 8.990510589252764e-05, "output_mean_power": 8.214657563492911e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.06772172679864e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1343959051909258e-05, "output_power": 9.994065101320318e-06, "reference_active": true, "delta_db": -0.5502247606744027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.893195477577164e-06, "output_power": 3.724630703324329e-06, "reference_active": true, "delta_db": -0.1922299124299432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973785070466098e-07, "output_power": 6.966747050428437e-07, "reference_active": true, "delta_db": -0.586344828862281}], "active_band_count": 3, "active_band_mae_db": 0.442933167322209}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.719774514443707e-05, "log_rms_mae_db": 0.7348442584967989, "rms_envelope_nrmse": 0.07511360018249798, "envelope_correlation": 0.9973927818592363, "rms_gain_error_db": -0.391963836070811, "reference_mean_power": 9.035589049479441e-05, "output_mean_power": 8.255817552568374e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9012156385948547e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1053455137857872e-05, "output_power": 9.72272030176067e-06, "reference_active": true, "delta_db": -0.557102605610301}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0905012137138186e-06, "output_power": 2.042350266602456e-06, "reference_active": true, "delta_db": -0.10120197638214556}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.495315305956309e-08, "output_power": 4.9144258262255836e-08, "reference_active": true, "delta_db": -1.2112745352013408}], "active_band_count": 3, "active_band_mae_db": 0.6231930390645958}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8068807707378253, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "84e4e57b1ca42ad54c40a4fa802c80262d229d953f191bf530c8eb466e84f4c8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9951005412601155, "standard_deviation": 0.0004739538263296627, "minimum": 0.994372522264504, "maximum": 0.9957762336408832, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1373425332760005, "standard_deviation": 0.027674077322832395, "minimum": 1.1079069600643074, "maximum": 1.1813940464940664, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.33243155805206315, "standard_deviation": 0.10795665000303196, "minimum": 0.18714008503133236, "maximum": 0.47323796818257313, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08615362199252999, "standard_deviation": 0.004560086730689789, "minimum": 0.0792094482174135, "maximum": 0.0924236583471387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9983390652255932, "standard_deviation": 0.00034622487793085953, "minimum": 0.9977086487683348, "maximum": 0.9987400415561851, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.673078286021694, "standard_deviation": 0.034839827140270785, "minimum": 0.6006358374744251, "maximum": 0.7151634606254328, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3184386168473582, "standard_deviation": 0.09316138732130527, "minimum": 0.15496953433300445, "maximum": 0.433572447310073, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05111909936334635, "standard_deviation": 0.007035039294297135, "minimum": 0.04268474551536214, "maximum": 0.06130749137200632, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9872941949211861, "standard_deviation": 0.0023915056534096575, "minimum": 0.9828633860846875, "maximum": 0.9904740012911332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2327034006250508, "standard_deviation": 0.028744532863042467, "minimum": 1.2064482967926067, "maximum": 1.2927451081085388, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7857111382016131, "standard_deviation": 0.4408256110027327, "minimum": 0.33417113772143914, "maximum": 1.3911065156273166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14322638361833462, "standard_deviation": 0.011019857648867256, "minimum": 0.1246724014321839, "maximum": 0.15981649885019605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9958578222912444, "standard_deviation": 0.0009846232998587696, "minimum": 0.9943335884294956, "maximum": 0.9973927818592363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7683201779009218, "standard_deviation": 0.026731963577952732, "minimum": 0.7348442584967989, "maximum": 0.8063572463241673, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9972372944407221, "standard_deviation": 0.5015494280075737, "minimum": 0.3312187850905724, "maximum": 1.6465896078090558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09050949250308833, "standard_deviation": 0.00899757514089531, "minimum": 0.07511360018249798, "maximum": 0.10199490591080355, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9889461109306884, "standard_deviation": 0.0012410176211547325, "minimum": 0.9869331726301958, "maximum": 0.9902071621903262, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2035284696889397, "standard_deviation": 0.031318113032818, "minimum": 1.1323198039890303, "maximum": 1.2224511934076956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4103377941076256, "standard_deviation": 0.21605878344740095, "minimum": 0.15838350245867405, "maximum": 0.676145457665592, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11503570020661846, "standard_deviation": 0.006382457661236534, "minimum": 0.10840753254948815, "maximum": 0.1255037695308177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967811108930593, "standard_deviation": 0.0005086194901185758, "minimum": 0.9961493494853921, "maximum": 0.9975241804749604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.659617781967006, "standard_deviation": 0.028630121763086692, "minimum": 0.6201774306314869, "maximum": 0.7029926493140226, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.36490869984871904, "standard_deviation": 0.2256241174828082, "minimum": 0.1365857728444302, "maximum": 0.7176084370675929, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0615387769094624, "standard_deviation": 0.004919926034001466, "minimum": 0.05593746533853069, "maximum": 0.06876148470374804, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24981536778979158, "standard_deviation": 0.05353938963119008, "minimum": 0.14575974680054524, "maximum": 0.32064425748853315, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6437711024198585, "standard_deviation": 0.006933522096897942, "minimum": 5.634769470324792, "maximum": 5.653190901258987, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6101669940708304, "standard_deviation": 0.020147594098030214, "minimum": 0.588729878662843, "maximum": 0.6431928069110918, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11065570231112375, "standard_deviation": 0.049250353705823285, "minimum": -0.17825351986869423, "maximum": -0.03415491074502989, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931140920470847, "standard_deviation": 0.0007285830090161716, "minimum": 0.9916843326664214, "maximum": 0.9939481007026821, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8726492814562631, "standard_deviation": 0.01243596499101727, "minimum": 0.8560786611003788, "maximum": 0.8880176014673954, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3558982782111847, "standard_deviation": 0.20506398954804406, "minimum": 0.11170075939123521, "maximum": 0.628607466264171, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08911239464466084, "standard_deviation": 0.004768483110557981, "minimum": 0.08346391035064128, "maximum": 0.09819621418301473, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980427012054647, "standard_deviation": 0.000286899157591308, "minimum": 0.9974892247421551, "maximum": 0.9983062338744034, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4754869979423134, "standard_deviation": 0.015795654431923177, "minimum": 0.45024943581911403, "maximum": 0.4983869093978906, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3787190716527561, "standard_deviation": 0.20629585861103777, "minimum": 0.1533992257338231, "maximum": 0.7323290381859651, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.047717485358048775, "standard_deviation": 0.0036661127678462085, "minimum": 0.04356854096946552, "maximum": 0.05468883301378379, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0019", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.1381173133850098, "predicted_mos": 2.1381173133850098, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ab99c6609ffe6d84b044e131bc19f0b2b59e82ec8ee319ef2cd9596343da3f40", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.3022924000397325, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9026774404065938, "全帯域_10ms_rms_envelope_nrmse": 0.08885186999538726, "全帯域_10ms_envelope_correlation": 0.9932522803078283, "全帯域_10ms_modulation_mae_db": 0.28253719783926057, "全帯域_40ms_log_rms_mae_db": 0.48394816325501405, "全帯域_40ms_rms_envelope_nrmse": 0.04254832150031698, "全帯域_40ms_envelope_correlation": 0.9985984953623243, "全帯域_40ms_modulation_mae_db": 0.23896556116873227, "300–1000 Hz_10ms_log_rms_mae_db": 1.2492121179239646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11520649187624252, "300–1000 Hz_10ms_envelope_correlation": 0.9891534684842849, "300–1000 Hz_10ms_modulation_mae_db": 0.2406529394586242, "300–1000 Hz_40ms_log_rms_mae_db": 0.650914021474058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05715197967812182, "300–1000 Hz_40ms_envelope_correlation": 0.9975958928486753, "300–1000 Hz_40ms_modulation_mae_db": 0.1674647444733973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1680344313769033, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08586169640557512, "1000–3000 Hz_10ms_envelope_correlation": 0.9951195211546406, "1000–3000 Hz_10ms_modulation_mae_db": 0.3347440673348003, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6595357933529935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05382712128560935, "1000–3000 Hz_40ms_envelope_correlation": 0.9981614940531927, "1000–3000 Hz_40ms_modulation_mae_db": 0.36604567497589696, "3000–6000 Hz_10ms_log_rms_mae_db": 1.282620571150113, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14493438678601894, "3000–6000 Hz_10ms_envelope_correlation": 0.9858493239550953, "3000–6000 Hz_10ms_modulation_mae_db": 1.3518944684995298, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7581393048440517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0771166778785745, "3000–6000 Hz_40ms_envelope_correlation": 0.9960912621331058, "3000–6000 Hz_40ms_modulation_mae_db": 1.464143116782526, "mr_stft_db": 5.699866905951471, "band_mae_db": 0.25670075426640354, "rms_gain_error_db": -0.15372922005968373, "output_peak": 0.6079654891432077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.860844828275289, "全帯域_10ms_rms_envelope_nrmse": 0.08671705111446408, "全帯域_10ms_envelope_correlation": 0.9934412826896896, "全帯域_10ms_modulation_mae_db": 0.23650263828739462, "全帯域_40ms_log_rms_mae_db": 0.44682752346129756, "全帯域_40ms_rms_envelope_nrmse": 0.04003051130122454, "全帯域_40ms_envelope_correlation": 0.9986035053485395, "全帯域_40ms_modulation_mae_db": 0.16975435821774423, "300–1000 Hz_10ms_log_rms_mae_db": 1.2285151213575838, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11213574182071949, "300–1000 Hz_10ms_envelope_correlation": 0.9894643924054175, "300–1000 Hz_10ms_modulation_mae_db": 0.1236920499191424, "300–1000 Hz_40ms_log_rms_mae_db": 0.6075631651360397, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05030299220849998, "300–1000 Hz_40ms_envelope_correlation": 0.997806207593888, "300–1000 Hz_40ms_modulation_mae_db": 0.1527949135263898, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2087964453776028, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09071582761106407, "1000–3000 Hz_10ms_envelope_correlation": 0.9946508609150172, "1000–3000 Hz_10ms_modulation_mae_db": 0.4112225176409225, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7090739172077262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05851553632934421, "1000–3000 Hz_40ms_envelope_correlation": 0.9979302014240596, "1000–3000 Hz_40ms_modulation_mae_db": 0.6548852588775825, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2584805150042813, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1538238172714399, "3000–6000 Hz_10ms_envelope_correlation": 0.9853207665194682, "3000–6000 Hz_10ms_modulation_mae_db": 1.1226968595782016, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7490412505758862, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09689813975604808, "3000–6000 Hz_40ms_envelope_correlation": 0.995372336446848, "3000–6000 Hz_40ms_modulation_mae_db": 1.4324267539386992, "mr_stft_db": 5.710310287445068, "band_mae_db": 0.2855402173547848, "rms_gain_error_db": -0.0882059323199072, "output_peak": 0.6147046450387768}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8918389407679616, "全帯域_10ms_rms_envelope_nrmse": 0.0757816935848775, "全帯域_10ms_envelope_correlation": 0.9952247840863532, "全帯域_10ms_modulation_mae_db": 0.3564074923896126, "全帯域_40ms_log_rms_mae_db": 0.5031511049224838, "全帯域_40ms_rms_envelope_nrmse": 0.041946707341298115, "全帯域_40ms_envelope_correlation": 0.9987151325630997, "全帯域_40ms_modulation_mae_db": 0.4031918040511684, "300–1000 Hz_10ms_log_rms_mae_db": 1.2229364241315968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10250756255626023, "300–1000 Hz_10ms_envelope_correlation": 0.9914255008313372, "300–1000 Hz_10ms_modulation_mae_db": 0.3897864625007421, "300–1000 Hz_40ms_log_rms_mae_db": 0.6910925282953131, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056233996672427664, "300–1000 Hz_40ms_envelope_correlation": 0.9975702255587783, "300–1000 Hz_40ms_modulation_mae_db": 0.4395288277825135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1936349336331635, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09094957066569917, "1000–3000 Hz_10ms_envelope_correlation": 0.9944623853882606, "1000–3000 Hz_10ms_modulation_mae_db": 0.9682563818101405, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6980067100485314, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05412845735079996, "1000–3000 Hz_40ms_envelope_correlation": 0.9980663535195078, "1000–3000 Hz_40ms_modulation_mae_db": 1.0940043900284673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2585736438481536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15381209944664218, "3000–6000 Hz_10ms_envelope_correlation": 0.9855669228115023, "3000–6000 Hz_10ms_modulation_mae_db": 1.0268447587557554, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7815836315080065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0932234108737095, "3000–6000 Hz_40ms_envelope_correlation": 0.9961867227796017, "3000–6000 Hz_40ms_modulation_mae_db": 1.3861983971013265, "mr_stft_db": 5.687977564784571, "band_mae_db": 0.25648160411220783, "rms_gain_error_db": -0.17398988590857808, "output_peak": 0.6028870721890719}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8619376718387467, "全帯域_10ms_rms_envelope_nrmse": 0.0805883484666224, "全帯域_10ms_envelope_correlation": 0.9943617105900947, "全帯域_10ms_modulation_mae_db": 0.338254513472206, "全帯域_40ms_log_rms_mae_db": 0.46136287915909024, "全帯域_40ms_rms_envelope_nrmse": 0.0442173235127065, "全帯域_40ms_envelope_correlation": 0.9982899264040515, "全帯域_40ms_modulation_mae_db": 0.47801945287764697, "300–1000 Hz_10ms_log_rms_mae_db": 1.2182586996596316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10305309277588429, "300–1000 Hz_10ms_envelope_correlation": 0.9912122031977714, "300–1000 Hz_10ms_modulation_mae_db": 0.4085614007023321, "300–1000 Hz_40ms_log_rms_mae_db": 0.6352962093618574, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05719928283456804, "300–1000 Hz_40ms_envelope_correlation": 0.9973088161256686, "300–1000 Hz_40ms_modulation_mae_db": 0.4994471926615341, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2147038998973683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09307892823433349, "1000–3000 Hz_10ms_envelope_correlation": 0.9941850194684324, "1000–3000 Hz_10ms_modulation_mae_db": 0.5933483833688911, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7073584834479916, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05380297790868959, "1000–3000 Hz_40ms_envelope_correlation": 0.9980867980617896, "1000–3000 Hz_40ms_modulation_mae_db": 0.4080530334406072, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2748782694549505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14915544082516116, "3000–6000 Hz_10ms_envelope_correlation": 0.9854054958078734, "3000–6000 Hz_10ms_modulation_mae_db": 0.5436679462933807, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7755822853299286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08334946346752337, "3000–6000 Hz_40ms_envelope_correlation": 0.9953817598625877, "3000–6000 Hz_40ms_modulation_mae_db": 0.6714478022610405, "mr_stft_db": 5.711038241464094, "band_mae_db": 0.21345606495867972, "rms_gain_error_db": -0.09379888393552686, "output_peak": 0.6239736254441287}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8850898954624815, "全帯域_10ms_rms_envelope_nrmse": 0.08221054846977399, "全帯域_10ms_envelope_correlation": 0.9941213928221352, "全帯域_10ms_modulation_mae_db": 0.4473996837137088, "全帯域_40ms_log_rms_mae_db": 0.46957731601123925, "全帯域_40ms_rms_envelope_nrmse": 0.04825338255919227, "全帯域_40ms_envelope_correlation": 0.9979355673607296, "全帯域_40ms_modulation_mae_db": 0.6133005424591939, "300–1000 Hz_10ms_log_rms_mae_db": 1.2195491901061744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10898625145338191, "300–1000 Hz_10ms_envelope_correlation": 0.9900718999941004, "300–1000 Hz_10ms_modulation_mae_db": 0.41605381995341056, "300–1000 Hz_40ms_log_rms_mae_db": 0.6630344448537042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06559607252577057, "300–1000 Hz_40ms_envelope_correlation": 0.9962807641884813, "300–1000 Hz_40ms_modulation_mae_db": 0.516915323202313, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1764449534953603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07875683331305294, "1000–3000 Hz_10ms_envelope_correlation": 0.9958193227936774, "1000–3000 Hz_10ms_modulation_mae_db": 0.3908544611924624, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6987799063947746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04018549118391293, "1000–3000 Hz_40ms_envelope_correlation": 0.9989164944307112, "1000–3000 Hz_40ms_modulation_mae_db": 0.3306077481158803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.329137667633917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14151891429935715, "3000–6000 Hz_10ms_envelope_correlation": 0.9876101180086422, "3000–6000 Hz_10ms_modulation_mae_db": 1.508822872949615, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9027819339271504, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09048680521194433, "3000–6000 Hz_40ms_envelope_correlation": 0.9958681221030933, "3000–6000 Hz_40ms_modulation_mae_db": 1.9551021410043206, "mr_stft_db": 5.70060607061734, "band_mae_db": 0.23258230685875417, "rms_gain_error_db": -0.08985831499921346, "output_peak": 0.6109860935259485}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8786906751101378, "全帯域_10ms_rms_envelope_nrmse": 0.09310044857381675, "全帯域_10ms_envelope_correlation": 0.9925168168637246, "全帯域_10ms_modulation_mae_db": 0.48949307812830484, "全帯域_40ms_log_rms_mae_db": 0.4584134919913954, "全帯域_40ms_rms_envelope_nrmse": 0.05013871600803514, "全帯域_40ms_envelope_correlation": 0.9978844862850136, "全帯域_40ms_modulation_mae_db": 0.37628483737921786, "300–1000 Hz_10ms_log_rms_mae_db": 1.2371564441465976, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11862060237141267, "300–1000 Hz_10ms_envelope_correlation": 0.988334763449969, "300–1000 Hz_10ms_modulation_mae_db": 0.5457163108478134, "300–1000 Hz_40ms_log_rms_mae_db": 0.6432910188341023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06549213710663636, "300–1000 Hz_40ms_envelope_correlation": 0.9965011258246577, "300–1000 Hz_40ms_modulation_mae_db": 0.42734086460334736, "1000–3000 Hz_10ms_log_rms_mae_db": 1.228694766584557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08746681751995322, "1000–3000 Hz_10ms_envelope_correlation": 0.9948626886360135, "1000–3000 Hz_10ms_modulation_mae_db": 0.2909615504104885, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7007029776940474, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048618261393899985, "1000–3000 Hz_40ms_envelope_correlation": 0.9984326578565226, "1000–3000 Hz_40ms_modulation_mae_db": 0.27032384691774475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2411892802827105, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12924691527181573, "3000–6000 Hz_10ms_envelope_correlation": 0.9888632929383647, "3000–6000 Hz_10ms_modulation_mae_db": 1.2237605310936368, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7445222575805761, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06761068207198133, "3000–6000 Hz_40ms_envelope_correlation": 0.9971356449140722, "3000–6000 Hz_40ms_modulation_mae_db": 1.5368514810850618, "mr_stft_db": 5.693990108338919, "band_mae_db": 0.210609153132146, "rms_gain_error_db": -0.14005929854604737, "output_peak": 0.6613105310652706}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8330529045443262, "全帯域_10ms_rms_envelope_nrmse": 0.0859343187417978, "全帯域_10ms_envelope_correlation": 0.9935812404760521, "全帯域_10ms_modulation_mae_db": 0.3948086554429208, "全帯域_40ms_log_rms_mae_db": 0.4165150348918308, "全帯域_40ms_rms_envelope_nrmse": 0.04434669193722991, "全帯域_40ms_envelope_correlation": 0.9982103000762589, "全帯域_40ms_modulation_mae_db": 0.3020430439406238, "300–1000 Hz_10ms_log_rms_mae_db": 1.1720184195926764, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1099444600524552, "300–1000 Hz_10ms_envelope_correlation": 0.990047686264759, "300–1000 Hz_10ms_modulation_mae_db": 0.46715799869565205, "300–1000 Hz_40ms_log_rms_mae_db": 0.5757254584346972, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056909324616713365, "300–1000 Hz_40ms_envelope_correlation": 0.997215784117958, "300–1000 Hz_40ms_modulation_mae_db": 0.39312414936474327, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2338398139054236, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08569822912530513, "1000–3000 Hz_10ms_envelope_correlation": 0.9950574174536773, "1000–3000 Hz_10ms_modulation_mae_db": 0.21751810757021042, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7539291698261247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04545207806769141, "1000–3000 Hz_40ms_envelope_correlation": 0.998578865546514, "1000–3000 Hz_40ms_modulation_mae_db": 0.19562031556785522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2861665631213297, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15620795392626652, "3000–6000 Hz_10ms_envelope_correlation": 0.9842738887061722, "3000–6000 Hz_10ms_modulation_mae_db": 1.0317762549638203, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8036886606041083, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08569693297094617, "3000–6000 Hz_40ms_envelope_correlation": 0.9962460630674945, "3000–6000 Hz_40ms_modulation_mae_db": 1.1712726759517247, "mr_stft_db": 5.70356737126415, "band_mae_db": 0.16904194112359736, "rms_gain_error_db": 0.021104853098805526, "output_peak": 0.6467597618599559}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8568680825007172, "全帯域_10ms_rms_envelope_nrmse": 0.08200820446605082, "全帯域_10ms_envelope_correlation": 0.994275934955484, "全帯域_10ms_modulation_mae_db": 0.32013833147860354, "全帯域_40ms_log_rms_mae_db": 0.4580417710072023, "全帯域_40ms_rms_envelope_nrmse": 0.0510063185122683, "全帯域_40ms_envelope_correlation": 0.997840807730686, "全帯域_40ms_modulation_mae_db": 0.3317540472106339, "300–1000 Hz_10ms_log_rms_mae_db": 1.209095386992219, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10537291481458211, "300–1000 Hz_10ms_envelope_correlation": 0.9908182958059696, "300–1000 Hz_10ms_modulation_mae_db": 0.24806365857286558, "300–1000 Hz_40ms_log_rms_mae_db": 0.6686305937066606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06574398516349933, "300–1000 Hz_40ms_envelope_correlation": 0.996390548788466, "300–1000 Hz_40ms_modulation_mae_db": 0.19572957712304653, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1781367307637052, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0805584760642915, "1000–3000 Hz_10ms_envelope_correlation": 0.9958707724013212, "1000–3000 Hz_10ms_modulation_mae_db": 0.417089987512307, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6808718751270687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04646868842158909, "1000–3000 Hz_40ms_envelope_correlation": 0.9988394195901398, "1000–3000 Hz_40ms_modulation_mae_db": 0.2854697314649888, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3011123816127501, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19004119036129052, "3000–6000 Hz_10ms_envelope_correlation": 0.9764841232620299, "3000–6000 Hz_10ms_modulation_mae_db": 0.6127717635609901, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8103347822241519, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.128681021273417, "3000–6000 Hz_40ms_envelope_correlation": 0.9901798975212923, "3000–6000 Hz_40ms_modulation_mae_db": 0.9299450513320937, "mr_stft_db": 5.6915307959144945, "band_mae_db": 0.26916365274181503, "rms_gain_error_db": -0.1524341274273369, "output_peak": 0.6067403237239903}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5349417686853855, "spectral_convergence": 0.2506144768956535, "spectral_convergence_status": "ok", "amplitude_floor": 1.5171791511760096e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.511662665697262, "spectral_convergence": 0.2539872443313816, "spectral_convergence_status": "ok", "amplitude_floor": 1.2250623067014244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.052996283471769, "spectral_convergence": 0.26179695893295124, "spectral_convergence_status": "ok", "amplitude_floor": 7.883290232655995e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.642531119863568e-05, "output_power": 4.9865785488846626e-05, "delta_db": 0.31047840874009835, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003916996423883643, "output_power": 0.0004388088219136274, "delta_db": 0.4932217509831218, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0020347588099002725, "output_power": 0.0018953140481136545, "delta_db": -0.30831755903856195, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007257385551015356, "output_power": 0.0007043143276929663, "delta_db": -0.13013672504949894, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001922577058777525, "output_power": 0.00018432376946510433, "delta_db": -0.18302412512065602, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.752614415284284e-05, "output_power": 3.65452844756141e-05, "delta_db": -0.11502595666648423, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25670075426640354, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15372922005968373, "pitch": null, "output_rms_dbfs": -24.802806541630005, "output_peak": 0.6079654891432077, "reference_rms_dbfs": -24.64907732157032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.699866905951471}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019822260308892205, "log_rms_mae_db": 0.9026774404065938, "rms_envelope_nrmse": 0.08885186999538726, "envelope_correlation": 0.9932522803078283, "rms_gain_error_db": -0.1537078122427397, "reference_mean_power": 0.003431222806852645, "output_mean_power": 0.0033119070429483482}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9226628348981265e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006393448264747715, "output_power": 0.0006266220884238077, "reference_active": true, "delta_db": -0.08729455559504759}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011197882141156391, "output_power": 0.00010183294819123382, "reference_active": true, "delta_db": -0.41247574871504517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.279682095777351e-05, "output_power": 3.950275460913996e-05, "reference_active": true, "delta_db": -0.34784128920768903}], "active_band_count": 3, "active_band_mae_db": 0.28253719783926057}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001679501427730559, "log_rms_mae_db": 0.48394816325501405, "rms_envelope_nrmse": 0.04254832150031698, "envelope_correlation": 0.9985984953623243, "rms_gain_error_db": -0.15399924389809994, "reference_mean_power": 0.003447933525092644, "output_mean_power": 0.0033278133512190953}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8195786138583623e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005865504590263927, "output_power": 0.0005732660395911328, "reference_active": true, "delta_db": -0.09949164164923703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.169773206547347e-05, "output_power": 5.5657104871561894e-05, "reference_active": true, "delta_db": -0.4474858839360971}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.232527996839891e-06, "output_power": 2.146866296231769e-06, "reference_active": true, "delta_db": -0.16991915792086268}], "active_band_count": 3, "active_band_mae_db": 0.23896556116873227}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001900450506992256, "log_rms_mae_db": 1.2492121179239646, "rms_envelope_nrmse": 0.11520649187624252, "envelope_correlation": 0.9891534684842849, "rms_gain_error_db": -0.21964193041273364, "reference_mean_power": 0.002194955532035717, "output_mean_power": 0.0020867072895540046}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4050180992423441e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048796679990628384, "output_power": 0.0004634772492117762, "reference_active": true, "delta_db": -0.22361853935197484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011744031611151056, "output_power": 0.00010989222684915922, "reference_active": true, "delta_db": -0.28850237565067705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.011051692782722e-05, "output_power": 3.82185731141202e-05, "reference_active": true, "delta_db": -0.20983790337322075}], "active_band_count": 3, "active_band_mae_db": 0.2406529394586242}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014767968623269652, "log_rms_mae_db": 0.650914021474058, "rms_envelope_nrmse": 0.05715197967812182, "envelope_correlation": 0.9975958928486753, "rms_gain_error_db": -0.2197138925833509, "reference_mean_power": 0.002205773735941533, "output_mean_power": 0.00209695722730428}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315971715150661e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004578823491754882, "output_power": 0.00043258996319078824, "reference_active": true, "delta_db": -0.24677463447249232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0620858906342295e-05, "output_power": 5.746533160959986e-05, "reference_active": true, "delta_db": -0.23216168187517802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.249804591930109e-06, "output_power": 2.237685291087443e-06, "reference_active": true, "delta_db": -0.023457917072521504}], "active_band_count": 3, "active_band_mae_db": 0.1674647444733973}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012415579655882008, "log_rms_mae_db": 1.1680344313769033, "rms_envelope_nrmse": 0.08586169640557512, "envelope_correlation": 0.9951195211546406, "rms_gain_error_db": -0.13714717174393457, "reference_mean_power": 0.0006561031712122296, "output_mean_power": 0.0006357076236808756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1091236930101443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020492597233413883, "output_power": 0.000197807316846382, "reference_active": true, "delta_db": -0.15354652397359875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1330216505908255e-06, "output_power": 7.251316945265138e-06, "reference_active": true, "delta_db": 0.07143345746826246}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.269143835584622e-06, "output_power": 5.108177847066149e-06, "reference_active": true, "delta_db": 0.7792522205625395}], "active_band_count": 3, "active_band_mae_db": 0.3347440673348003}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011294582885045651, "log_rms_mae_db": 0.6595357933529935, "rms_envelope_nrmse": 0.05382712128560935, "envelope_correlation": 0.9981614940531927, "rms_gain_error_db": -0.13819966043183424, "reference_mean_power": 0.000659140146325869, "output_mean_power": 0.0006384954371427401}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0026600561194904e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019159735065259618, "output_power": 0.00018467218128682822, "reference_active": true, "delta_db": -0.15988020535767952}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0478497584910014e-06, "output_power": 3.2457215068225274e-06, "reference_active": true, "delta_db": 0.27317698279195496}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0860941276748228e-07, "output_power": 2.431318665507244e-07, "reference_active": true, "delta_db": 0.6650798367780565}], "active_band_count": 3, "active_band_mae_db": 0.36604567497589696}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.245162939338534e-05, "log_rms_mae_db": 1.282620571150113, "rms_envelope_nrmse": 0.14493438678601894, "envelope_correlation": 0.9858493239550953, "rms_gain_error_db": -0.19192090354367713, "reference_mean_power": 8.990510589252764e-05, "output_mean_power": 8.601857952913418e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.06772172679864e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1343959051909258e-05, "output_power": 1.0559232751894828e-05, "reference_active": true, "delta_db": -0.31132287255704183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.893195477577164e-06, "output_power": 3.120999861392616e-06, "reference_active": true, "delta_db": -0.9601246083967281}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973785070466098e-07, "output_power": 4.1999191759419113e-07, "reference_active": true, "delta_db": -2.7842359245448196}], "active_band_count": 3, "active_band_mae_db": 1.3518944684995298}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.719774514443707e-05, "log_rms_mae_db": 0.7581393048440517, "rms_envelope_nrmse": 0.0771166778785745, "envelope_correlation": 0.9960912621331058, "rms_gain_error_db": -0.19192889004752284, "reference_mean_power": 9.035589049479441e-05, "output_mean_power": 8.644971809310569e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9012156385948547e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1053455137857872e-05, "output_power": 1.0098130179729117e-05, "reference_active": true, "delta_db": -0.3925708784204456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0905012137138186e-06, "output_power": 1.5913201252182753e-06, "reference_active": true, "delta_db": -1.184928687526967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.495315305956309e-08, "output_power": 3.397091617574797e-08, "reference_active": true, "delta_db": -2.8149297844001664}], "active_band_count": 3, "active_band_mae_db": 1.464143116782526}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8094141120349648, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1b0c8e3f609ab637ba1b80f1b596917a2822be7c260aae09127518beb262f3df", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.99500349852638, "standard_deviation": 0.0006024940802285372, "minimum": 0.9941850194684324, "maximum": 0.9958707724013212, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2002857468792605, "standard_deviation": 0.02494739022286533, "minimum": 1.1680344313769033, "maximum": 1.2338398139054236, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.45299943210502785, "standard_deviation": 0.23536488667257874, "minimum": 0.21751810757021042, "maximum": 0.9682563818101405, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08663579736740933, "standard_deviation": 0.005041966623385101, "minimum": 0.07875683331305294, "maximum": 0.09307892823433349, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9983765355603047, "standard_deviation": 0.00037330991906316016, "minimum": 0.9979302014240596, "maximum": 0.9989164944307112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7010323541374073, "standard_deviation": 0.02684746798745626, "minimum": 0.6595357933529935, "maximum": 0.7539291698261247, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4506262499236279, "standard_deviation": 0.29370578713490153, "minimum": 0.19562031556785522, "maximum": 1.0940043900284673, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.050124826492692065, "standard_deviation": 0.005975317139211808, "minimum": 0.04018549118391293, "maximum": 0.05851553632934421, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9849217415011435, "standard_deviation": 0.0037041717427391393, "minimum": 0.9764841232620299, "maximum": 0.9888632929383647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2790198615135258, "standard_deviation": 0.02765122735922616, "minimum": 1.2411892802827105, "maximum": 1.329137667633917, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0527794319618664, "standard_deviation": 0.33498046892001987, "minimum": 0.5436679462933807, "maximum": 1.508822872949615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15234258977349902, "standard_deviation": 0.01755025657635521, "minimum": 0.12924691527181573, "maximum": 0.19004119036129052, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9953077261035119, "standard_deviation": 0.002145648501407612, "minimum": 0.9901798975212923, "maximum": 0.9971356449140722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7907092633242325, "standard_deviation": 0.051208644418910876, "minimum": 0.7445222575805761, "maximum": 0.9027819339271504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3184234274320992, "standard_deviation": 0.39334486571586785, "minimum": 0.6714478022610405, "maximum": 1.9551021410043206, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09038289168801802, "standard_deviation": 0.01806048494704967, "minimum": 0.06761068207198133, "maximum": 0.128681021273417, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9900660263042012, "standard_deviation": 0.001063697655429055, "minimum": 0.988334763449969, "maximum": 0.9914255008313372, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2195927254888055, "standard_deviation": 0.022840666450196292, "minimum": 1.1720184195926764, "maximum": 1.2492121179239646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3549605800813228, "standard_deviation": 0.13875600638513985, "minimum": 0.1236920499191424, "maximum": 0.5457163108478134, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10947838971511731, "standard_deviation": 0.005740133704979281, "minimum": 0.10250756255626023, "maximum": 0.11862060237141267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970836706308217, "standard_deviation": 0.000604004082697052, "minimum": 0.9962807641884813, "maximum": 0.997806207593888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6419434300120541, "standard_deviation": 0.036382815600459584, "minimum": 0.5757254584346972, "maximum": 0.6910925282953131, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.34904319909216064, "standard_deviation": 0.15215260496635116, "minimum": 0.1527949135263898, "maximum": 0.516915323202313, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059328721350779635, "standard_deviation": 0.005664302944912515, "minimum": 0.05030299220849998, "maximum": 0.06574398516349933, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23669696181854855, "standard_deviation": 0.03788361627302331, "minimum": 0.16904194112359736, "maximum": 0.2855402173547848, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.699860918222514, "standard_deviation": 0.008396312025247885, "minimum": 5.687977564784571, "maximum": 5.711038241464094, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6219159427487938, "standard_deviation": 0.021152034339144313, "minimum": 0.6028870721890719, "maximum": 0.6613105310652706, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10887135126218601, "standard_deviation": 0.062143386649567256, "minimum": -0.17398988590857808, "maximum": 0.021104853098805526, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9938469303489201, "standard_deviation": 0.0008264213649371159, "minimum": 0.9925168168637246, "maximum": 0.9952247840863532, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8713750548632817, "standard_deviation": 0.022421974049141608, "minimum": 0.8330529045443262, "maximum": 0.9026774404065938, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3581926988440015, "standard_deviation": 0.08359723403791931, "minimum": 0.23650263828739462, "maximum": 0.48949307812830484, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08439906042659884, "standard_deviation": 0.005379300026209128, "minimum": 0.0757816935848775, "maximum": 0.09310044857381675, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982597776413378, "standard_deviation": 0.0003515425614418782, "minimum": 0.997840807730686, "maximum": 0.9987151325630997, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4622296605874442, "standard_deviation": 0.02552352334413654, "minimum": 0.4165150348918308, "maximum": 0.5031511049224838, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36416420591312015, "standard_deviation": 0.13891484609090463, "minimum": 0.16975435821774423, "maximum": 0.6133005424591939, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045310996584033966, "standard_deviation": 0.004024239887724209, "minimum": 0.04003051130122454, "maximum": 0.0510063185122683, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-960-0019", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.110442638397217, "predicted_mos": 2.110442638397217, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d5bbf247f3e5500bdc27b8941bc7b0a9cc20766d2dfc121c643e4438c016da96", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.185159099753946, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8857664570780871, "全帯域_10ms_rms_envelope_nrmse": 0.08189955353364672, "全帯域_10ms_envelope_correlation": 0.9942189278947694, "全帯域_10ms_modulation_mae_db": 0.21612270068772663, "全帯域_40ms_log_rms_mae_db": 0.44855010085689156, "全帯域_40ms_rms_envelope_nrmse": 0.04074345421974594, "全帯域_40ms_envelope_correlation": 0.9986225403061124, "全帯域_40ms_modulation_mae_db": 0.1671265073874414, "300–1000 Hz_10ms_log_rms_mae_db": 1.2236206502390594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1063124620299864, "300–1000 Hz_10ms_envelope_correlation": 0.990660820510436, "300–1000 Hz_10ms_modulation_mae_db": 0.1958880502565128, "300–1000 Hz_40ms_log_rms_mae_db": 0.5855848624305763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053940065346804004, "300–1000 Hz_40ms_envelope_correlation": 0.9976605277237954, "300–1000 Hz_40ms_modulation_mae_db": 0.2085441823690887, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1329523885251565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08197080614991058, "1000–3000 Hz_10ms_envelope_correlation": 0.9954594010313811, "1000–3000 Hz_10ms_modulation_mae_db": 0.8300054615412166, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6740637052373548, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04942719546824061, "1000–3000 Hz_40ms_envelope_correlation": 0.9983120125363448, "1000–3000 Hz_40ms_modulation_mae_db": 0.9253486004107838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2803743199240032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14851697350553364, "3000–6000 Hz_10ms_envelope_correlation": 0.9866587399376033, "3000–6000 Hz_10ms_modulation_mae_db": 1.2715750792855471, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7729489498769289, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09033735158178763, "3000–6000 Hz_40ms_envelope_correlation": 0.9965105692026417, "3000–6000 Hz_40ms_modulation_mae_db": 1.4722067115114463, "mr_stft_db": 5.674376488581225, "band_mae_db": 0.24308192454179992, "rms_gain_error_db": -0.11887410232250208, "output_peak": 0.6432964750081189}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8331133672401365, "全帯域_10ms_rms_envelope_nrmse": 0.08389982313209546, "全帯域_10ms_envelope_correlation": 0.9939734564819921, "全帯域_10ms_modulation_mae_db": 0.6292440219806015, "全帯域_40ms_log_rms_mae_db": 0.3981814834733737, "全帯域_40ms_rms_envelope_nrmse": 0.051022019659470876, "全帯域_40ms_envelope_correlation": 0.9978102666688411, "全帯域_40ms_modulation_mae_db": 0.5964921487202638, "300–1000 Hz_10ms_log_rms_mae_db": 1.169182208865671, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10996264332560834, "300–1000 Hz_10ms_envelope_correlation": 0.9899488933992434, "300–1000 Hz_10ms_modulation_mae_db": 0.6149503140083318, "300–1000 Hz_40ms_log_rms_mae_db": 0.5997970878158306, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0667818978204859, "300–1000 Hz_40ms_envelope_correlation": 0.9961912291784486, "300–1000 Hz_40ms_modulation_mae_db": 0.600669621476018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.17987346336586, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0785551833187479, "1000–3000 Hz_10ms_envelope_correlation": 0.9960194394572334, "1000–3000 Hz_10ms_modulation_mae_db": 0.27765748702783394, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6766546004008009, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047737204791218034, "1000–3000 Hz_40ms_envelope_correlation": 0.998678487970548, "1000–3000 Hz_40ms_modulation_mae_db": 0.1742503692334962, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2874920317958432, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13577364279911247, "3000–6000 Hz_10ms_envelope_correlation": 0.9882534252461334, "3000–6000 Hz_10ms_modulation_mae_db": 0.5851053081748335, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7749306553525517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07969071085478587, "3000–6000 Hz_40ms_envelope_correlation": 0.9966641059255849, "3000–6000 Hz_40ms_modulation_mae_db": 0.9721130457217715, "mr_stft_db": 5.675914749007287, "band_mae_db": 0.22315033491126793, "rms_gain_error_db": -0.1428769952564438, "output_peak": 0.6251244698220846}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8630788558492276, "全帯域_10ms_rms_envelope_nrmse": 0.09004344766340198, "全帯域_10ms_envelope_correlation": 0.9929388842687902, "全帯域_10ms_modulation_mae_db": 0.4037238439606086, "全帯域_40ms_log_rms_mae_db": 0.442615434174274, "全帯域_40ms_rms_envelope_nrmse": 0.04778141190079506, "全帯域_40ms_envelope_correlation": 0.9979149537763385, "全帯域_40ms_modulation_mae_db": 0.2716753587648357, "300–1000 Hz_10ms_log_rms_mae_db": 1.2057170403675663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11600432376239479, "300–1000 Hz_10ms_envelope_correlation": 0.9888471345002001, "300–1000 Hz_10ms_modulation_mae_db": 0.2602537317304486, "300–1000 Hz_40ms_log_rms_mae_db": 0.5823116302624868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061407490395958775, "300–1000 Hz_40ms_envelope_correlation": 0.9967022231772561, "300–1000 Hz_40ms_modulation_mae_db": 0.30550543196076657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1650141416223752, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07933019469192726, "1000–3000 Hz_10ms_envelope_correlation": 0.995753781532249, "1000–3000 Hz_10ms_modulation_mae_db": 0.28827436897504444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6965481144728318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04894029402672098, "1000–3000 Hz_40ms_envelope_correlation": 0.9983507916162679, "1000–3000 Hz_40ms_modulation_mae_db": 0.13694971981371337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2668116767505566, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17566818819396268, "3000–6000 Hz_10ms_envelope_correlation": 0.9790612877043097, "3000–6000 Hz_10ms_modulation_mae_db": 0.7424664186478603, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7725998957010161, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10941896772276012, "3000–6000 Hz_40ms_envelope_correlation": 0.9916694348356985, "3000–6000 Hz_40ms_modulation_mae_db": 1.0757648697485658, "mr_stft_db": 5.682307599375467, "band_mae_db": 0.156607803509662, "rms_gain_error_db": -0.0017596290690461915, "output_peak": 0.6393412324885132}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8366085396018581, "全帯域_10ms_rms_envelope_nrmse": 0.08322098400154489, "全帯域_10ms_envelope_correlation": 0.9939843836354597, "全帯域_10ms_modulation_mae_db": 0.11254356372727602, "全帯域_40ms_log_rms_mae_db": 0.4185197203372781, "全帯域_40ms_rms_envelope_nrmse": 0.04378231084482386, "全帯域_40ms_envelope_correlation": 0.998297239971466, "全帯域_40ms_modulation_mae_db": 0.10978393626813114, "300–1000 Hz_10ms_log_rms_mae_db": 1.151012689494887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10655329717178243, "300–1000 Hz_10ms_envelope_correlation": 0.9905737800942807, "300–1000 Hz_10ms_modulation_mae_db": 0.11538166731620896, "300–1000 Hz_40ms_log_rms_mae_db": 0.5563444620636323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05667350426617266, "300–1000 Hz_40ms_envelope_correlation": 0.9972245725563648, "300–1000 Hz_40ms_modulation_mae_db": 0.08274439292005917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1975428321499157, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08463841419130763, "1000–3000 Hz_10ms_envelope_correlation": 0.9953280029878743, "1000–3000 Hz_10ms_modulation_mae_db": 0.34758046699128625, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7025812355499506, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05239881586390591, "1000–3000 Hz_40ms_envelope_correlation": 0.9983414994669784, "1000–3000 Hz_40ms_modulation_mae_db": 0.17725389374061384, "3000–6000 Hz_10ms_log_rms_mae_db": 1.280261876318656, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13474828386008347, "3000–6000 Hz_10ms_envelope_correlation": 0.9881931598125394, "3000–6000 Hz_10ms_modulation_mae_db": 0.9646153507397415, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7713096121944067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07866381410103267, "3000–6000 Hz_40ms_envelope_correlation": 0.9964411379410725, "3000–6000 Hz_40ms_modulation_mae_db": 1.3187661557027772, "mr_stft_db": 5.671275179344121, "band_mae_db": 0.16408740586075532, "rms_gain_error_db": -0.0698388228696476, "output_peak": 0.621557115470606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8662344978810548, "全帯域_10ms_rms_envelope_nrmse": 0.09103780715371265, "全帯域_10ms_envelope_correlation": 0.9928014411502846, "全帯域_10ms_modulation_mae_db": 0.20351622952964152, "全帯域_40ms_log_rms_mae_db": 0.4408969317674532, "全帯域_40ms_rms_envelope_nrmse": 0.04783092402321363, "全帯域_40ms_envelope_correlation": 0.9980238256948525, "全帯域_40ms_modulation_mae_db": 0.20081039520707286, "300–1000 Hz_10ms_log_rms_mae_db": 1.2511340897770968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11847999283693833, "300–1000 Hz_10ms_envelope_correlation": 0.9882619995112152, "300–1000 Hz_10ms_modulation_mae_db": 0.31629034033322273, "300–1000 Hz_40ms_log_rms_mae_db": 0.6444084258377475, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06388873392666175, "300–1000 Hz_40ms_envelope_correlation": 0.9965106309983537, "300–1000 Hz_40ms_modulation_mae_db": 0.3096460658162529, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2047111871311278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0834430716953415, "1000–3000 Hz_10ms_envelope_correlation": 0.9953906174778532, "1000–3000 Hz_10ms_modulation_mae_db": 0.251879112217436, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6670455316505425, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04875042590833007, "1000–3000 Hz_40ms_envelope_correlation": 0.9985051263988377, "1000–3000 Hz_40ms_modulation_mae_db": 0.37173557365316573, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2910467253831424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1524561966760392, "3000–6000 Hz_10ms_envelope_correlation": 0.9845339788509926, "3000–6000 Hz_10ms_modulation_mae_db": 1.1796212365443053, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8085137733817551, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07698914755046117, "3000–6000 Hz_40ms_envelope_correlation": 0.996484380200002, "3000–6000 Hz_40ms_modulation_mae_db": 1.3133735428391013, "mr_stft_db": 5.672697983009388, "band_mae_db": 0.2375099490771698, "rms_gain_error_db": -0.11547183171906078, "output_peak": 0.6479342641912005}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8939811162695491, "全帯域_10ms_rms_envelope_nrmse": 0.08431446780877697, "全帯域_10ms_envelope_correlation": 0.9937914773577816, "全帯域_10ms_modulation_mae_db": 0.46887042704900866, "全帯域_40ms_log_rms_mae_db": 0.46958832377628257, "全帯域_40ms_rms_envelope_nrmse": 0.05159134614626605, "全帯域_40ms_envelope_correlation": 0.9975900146836336, "全帯域_40ms_modulation_mae_db": 0.2569723633897258, "300–1000 Hz_10ms_log_rms_mae_db": 1.2252987343650763, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10773055600837089, "300–1000 Hz_10ms_envelope_correlation": 0.9903155275275143, "300–1000 Hz_10ms_modulation_mae_db": 0.38346305290260946, "300–1000 Hz_40ms_log_rms_mae_db": 0.6305838682405784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06703373711322921, "300–1000 Hz_40ms_envelope_correlation": 0.9960291717948815, "300–1000 Hz_40ms_modulation_mae_db": 0.24491549184660688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.181492910699394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08867892227067191, "1000–3000 Hz_10ms_envelope_correlation": 0.9949486617202696, "1000–3000 Hz_10ms_modulation_mae_db": 0.4080916840595426, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7056818006117841, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05259411340189248, "1000–3000 Hz_40ms_envelope_correlation": 0.9984409612223403, "1000–3000 Hz_40ms_modulation_mae_db": 0.40375523129681934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.266689585581699, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1543495102139471, "3000–6000 Hz_10ms_envelope_correlation": 0.9853447124442886, "3000–6000 Hz_10ms_modulation_mae_db": 1.2162370036718673, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8037507852338791, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10257179000882562, "3000–6000 Hz_40ms_envelope_correlation": 0.9946950641164028, "3000–6000 Hz_40ms_modulation_mae_db": 1.3669412448692047, "mr_stft_db": 5.684104084103399, "band_mae_db": 0.24601560950454449, "rms_gain_error_db": -0.06825622461657291, "output_peak": 0.6595643470191231}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.852183640252998, "全帯域_10ms_rms_envelope_nrmse": 0.08647411863583176, "全帯域_10ms_envelope_correlation": 0.9935016417653705, "全帯域_10ms_modulation_mae_db": 0.5958820060591412, "全帯域_40ms_log_rms_mae_db": 0.42657071217028286, "全帯域_40ms_rms_envelope_nrmse": 0.043760562627237515, "全帯域_40ms_envelope_correlation": 0.9983392398791059, "全帯域_40ms_modulation_mae_db": 0.6264751205767644, "300–1000 Hz_10ms_log_rms_mae_db": 1.196775694451451, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11133253026773529, "300–1000 Hz_10ms_envelope_correlation": 0.9896476922060783, "300–1000 Hz_10ms_modulation_mae_db": 0.6654391480308771, "300–1000 Hz_40ms_log_rms_mae_db": 0.5901927350432511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05742657981289221, "300–1000 Hz_40ms_envelope_correlation": 0.9971992959912237, "300–1000 Hz_40ms_modulation_mae_db": 0.6599040576589263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1802640368553878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09115380577003769, "1000–3000 Hz_10ms_envelope_correlation": 0.9944056969942672, "1000–3000 Hz_10ms_modulation_mae_db": 0.4164626804966391, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6746671361091843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04931912332781315, "1000–3000 Hz_40ms_envelope_correlation": 0.9983817699449055, "1000–3000 Hz_40ms_modulation_mae_db": 0.47720529678512635, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3015671694136648, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14775522322999332, "3000–6000 Hz_10ms_envelope_correlation": 0.9853021439640539, "3000–6000 Hz_10ms_modulation_mae_db": 0.3368671742124036, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8129414915962956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.089283285751015, "3000–6000 Hz_40ms_envelope_correlation": 0.9946716618404307, "3000–6000 Hz_40ms_modulation_mae_db": 0.7251541201690043, "mr_stft_db": 5.685261166328291, "band_mae_db": 0.18315444033556286, "rms_gain_error_db": -0.09795235972962588, "output_peak": 0.6293684192424557}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8704417856085839, "全帯域_10ms_rms_envelope_nrmse": 0.09109212289427139, "全帯域_10ms_envelope_correlation": 0.9930006748471004, "全帯域_10ms_modulation_mae_db": 0.29986687223894304, "全帯域_40ms_log_rms_mae_db": 0.47097154493427373, "全帯域_40ms_rms_envelope_nrmse": 0.050195051869480076, "全帯域_40ms_envelope_correlation": 0.9980827763854647, "全帯域_40ms_modulation_mae_db": 0.301577307359943, "300–1000 Hz_10ms_log_rms_mae_db": 1.2486499746146342, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1224410970943733, "300–1000 Hz_10ms_envelope_correlation": 0.9876635680948205, "300–1000 Hz_10ms_modulation_mae_db": 0.33739240678486143, "300–1000 Hz_40ms_log_rms_mae_db": 0.6803056716345756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06987343061772687, "300–1000 Hz_40ms_envelope_correlation": 0.9961267830891934, "300–1000 Hz_40ms_modulation_mae_db": 0.34096713239039844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1417203037449242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08647118575534311, "1000–3000 Hz_10ms_envelope_correlation": 0.9951612343509822, "1000–3000 Hz_10ms_modulation_mae_db": 0.3269344491068282, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6185103989083458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05113411293781616, "1000–3000 Hz_40ms_envelope_correlation": 0.9984892505789422, "1000–3000 Hz_40ms_modulation_mae_db": 0.5981941853523464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2393279850705934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12873983612132933, "3000–6000 Hz_10ms_envelope_correlation": 0.9893995111960312, "3000–6000 Hz_10ms_modulation_mae_db": 0.5979122415697964, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7563445206790178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07559384018767201, "3000–6000 Hz_40ms_envelope_correlation": 0.9969016890585515, "3000–6000 Hz_40ms_modulation_mae_db": 0.741911173874437, "mr_stft_db": 5.680316571861078, "band_mae_db": 0.2464431734734667, "rms_gain_error_db": -0.19301133941610332, "output_peak": 0.6231533302177765}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.513406446389917, "spectral_convergence": 0.24079918097465358, "spectral_convergence_status": "ok", "amplitude_floor": 1.5171791511760096e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.479303282737213, "spectral_convergence": 0.2398544482940321, "spectral_convergence_status": "ok", "amplitude_floor": 1.2250623067014244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.030419736616546, "spectral_convergence": 0.24770978790722772, "spectral_convergence_status": "ok", "amplitude_floor": 7.883290232655995e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.642531119863568e-05, "output_power": 4.9623323348040294e-05, "delta_db": 0.289310226710018, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003916996423883643, "output_power": 0.00040058107246070373, "delta_db": 0.09737249888985931, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0020347588099002725, "output_power": 0.0019443535205912147, "delta_db": -0.197377067253796, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007257385551015356, "output_power": 0.0007310432354753452, "delta_db": 0.03162867097907923, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001922577058777525, "output_power": 0.00017658653364691925, "delta_db": -0.36926174230343856, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.752614415284284e-05, "output_power": 3.3649592436454646e-05, "delta_db": -0.4735413411146086, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24308192454179992, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11887410232250208, "pitch": null, "output_rms_dbfs": -24.76795142389282, "output_peak": 0.6432964750081189, "reference_rms_dbfs": -24.64907732157032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.674376488581225}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019822260308892205, "log_rms_mae_db": 0.8857664570780871, "rms_envelope_nrmse": 0.08189955353364672, "envelope_correlation": 0.9942189278947694, "rms_gain_error_db": -0.11887459365528637, "reference_mean_power": 0.003431222806852645, "output_mean_power": 0.003338577487673694}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9226628348981265e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006393448264747715, "output_power": 0.0006349809967988414, "reference_active": true, "delta_db": -0.02974426982076228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011197882141156391, "output_power": 0.0001156062285441449, "reference_active": true, "delta_db": 0.13845341111864248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.279682095777351e-05, "output_power": 4.78000721312957e-05, "reference_active": true, "delta_db": 0.4801704211237751}], "active_band_count": 3, "active_band_mae_db": 0.21612270068772663}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001679501427730559, "log_rms_mae_db": 0.44855010085689156, "rms_envelope_nrmse": 0.04074345421974594, "envelope_correlation": 0.9986225403061124, "rms_gain_error_db": -0.11883433226974631, "reference_mean_power": 0.003447933525092644, "output_mean_power": 0.0033548681065531543}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8195786138583623e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005865504590263927, "output_power": 0.0005882766098675142, "reference_active": true, "delta_db": 0.012762019952238465}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.169773206547347e-05, "output_power": 5.929158233750229e-05, "reference_active": true, "delta_db": -0.17276159497860352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.232527996839891e-06, "output_power": 2.0759233941308914e-06, "reference_active": true, "delta_db": -0.31585590723148216}], "active_band_count": 3, "active_band_mae_db": 0.1671265073874414}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001900450506992256, "log_rms_mae_db": 1.2236206502390594, "rms_envelope_nrmse": 0.1063124620299864, "envelope_correlation": 0.990660820510436, "rms_gain_error_db": -0.15678096171704253, "reference_mean_power": 0.002194955532035717, "output_mean_power": 0.002117130506241997}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4050180992423441e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048796679990628384, "output_power": 0.0004925449644165628, "reference_active": true, "delta_db": 0.04055608724728256}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011744031611151056, "output_power": 0.00011950095069694648, "reference_active": true, "delta_db": 0.07554148799120639}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.011051692782722e-05, "output_power": 4.471105489681278e-05, "reference_active": true, "delta_db": 0.47156657553104947}], "active_band_count": 3, "active_band_mae_db": 0.1958880502565128}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014767968623269652, "log_rms_mae_db": 0.5855848624305763, "rms_envelope_nrmse": 0.053940065346804004, "envelope_correlation": 0.9976605277237954, "rms_gain_error_db": -0.1567805579242162, "reference_mean_power": 0.002205773735941533, "output_mean_power": 0.0021275653343606283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315971715150661e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004578823491754882, "output_power": 0.0004677150462143165, "reference_active": true, "delta_db": 0.09227438910441073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0620858906342295e-05, "output_power": 5.76950350912539e-05, "reference_active": true, "delta_db": -0.21483643540114922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.249804591930109e-06, "output_power": 2.090704368015353e-06, "reference_active": true, "delta_db": -0.3185217226017062}], "active_band_count": 3, "active_band_mae_db": 0.2085441823690887}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012415579655882008, "log_rms_mae_db": 1.1329523885251565, "rms_envelope_nrmse": 0.08197080614991058, "envelope_correlation": 0.9954594010313811, "rms_gain_error_db": -0.016119709544381893, "reference_mean_power": 0.0006561031712122296, "output_mean_power": 0.0006536724267694187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1091236930101443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020492597233413883, "output_power": 0.0001967640435947695, "reference_active": true, "delta_db": -0.17651265469580757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1330216505908255e-06, "output_power": 8.763210332651733e-06, "reference_active": true, "delta_db": 0.8938969343888032}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.269143835584622e-06, "output_power": 5.919724110060367e-06, "reference_active": true, "delta_db": 1.4196067955390388}], "active_band_count": 3, "active_band_mae_db": 0.8300054615412166}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011294582885045651, "log_rms_mae_db": 0.6740637052373548, "rms_envelope_nrmse": 0.04942719546824061, "envelope_correlation": 0.9983120125363448, "rms_gain_error_db": -0.015862516155737895, "reference_mean_power": 0.000659140146325869, "output_mean_power": 0.0006567370418861661}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0026600561194904e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019159735065259618, "output_power": 0.00018472382602409954, "reference_active": true, "delta_db": -0.15866584331160932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0478497584910014e-06, "output_power": 3.931065490040288e-06, "reference_active": true, "delta_db": 1.105167240909348}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0860941276748228e-07, "output_power": 2.9549842622761107e-07, "reference_active": true, "delta_db": 1.5122127170113944}], "active_band_count": 3, "active_band_mae_db": 0.9253486004107838}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.245162939338534e-05, "log_rms_mae_db": 1.2803743199240032, "rms_envelope_nrmse": 0.14851697350553364, "envelope_correlation": 0.9866587399376033, "rms_gain_error_db": -0.50521439524204, "reference_mean_power": 8.990510589252764e-05, "output_mean_power": 8.003186129582616e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.06772172679864e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1343959051909258e-05, "output_power": 1.0329602304370565e-05, "reference_active": true, "delta_db": -0.4068104892245358}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.893195477577164e-06, "output_power": 3.2116012106224767e-06, "reference_active": true, "delta_db": -0.8358459726536809}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973785070466098e-07, "output_power": 4.410193529040345e-07, "reference_active": true, "delta_db": -2.5720687759784244}], "active_band_count": 3, "active_band_mae_db": 1.2715750792855471}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.719774514443707e-05, "log_rms_mae_db": 0.7729489498769289, "rms_envelope_nrmse": 0.09033735158178763, "envelope_correlation": 0.9965105692026417, "rms_gain_error_db": -0.505217746504955, "reference_mean_power": 9.035589049479441e-05, "output_mean_power": 8.043307933885571e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9012156385948547e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1053455137857872e-05, "output_power": 9.977714483167923e-06, "reference_active": true, "delta_db": -0.4446698068421746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0905012137138186e-06, "output_power": 1.6346906551939757e-06, "reference_active": true, "delta_db": -1.0681484407068467}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.495315305956309e-08, "output_power": 3.32828148795012e-08, "reference_active": true, "delta_db": -2.9038018869853177}], "active_band_count": 3, "active_band_mae_db": 1.4722067115114463}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8144780104134627, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b8c9e743d7da3016340b1123ce74ce30a7423a2ba6e4145303ce8d5e4a015210", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9953083544440138, "standard_deviation": 0.0004927806204732454, "minimum": 0.9944056969942672, "maximum": 0.9960194394572334, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1729464080117675, "standard_deviation": 0.0251454960502676, "minimum": 1.1329523885251565, "maximum": 1.2047111871311278, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3933607138019784, "standard_deviation": 0.1860741165156328, "minimum": 0.251879112217436, "maximum": 0.8300054615412166, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08428019798041095, "standard_deviation": 0.004388888603006344, "minimum": 0.0785551833187479, "maximum": 0.09115380577003769, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9984374874668955, "standard_deviation": 0.00012007498949658995, "minimum": 0.9983120125363448, "maximum": 0.998678487970548, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6769690653675993, "standard_deviation": 0.02775781660865432, "minimum": 0.6185103989083458, "maximum": 0.7056818006117841, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4080866087857581, "standard_deviation": 0.2649249899471063, "minimum": 0.13694971981371337, "maximum": 0.9253486004107838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05003766071574218, "standard_deviation": 0.001787055031610869, "minimum": 0.047737204791218034, "maximum": 0.05259411340189248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.985843369894494, "standard_deviation": 0.0032269750811436734, "minimum": 0.9790612877043097, "maximum": 0.9893995111960312, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.27669642127977, "standard_deviation": 0.01912931525509069, "minimum": 1.2393279850705934, "maximum": 1.3015671694136648, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8617999766057944, "standard_deviation": 0.3466398329849567, "minimum": 0.3368671742124036, "maximum": 1.2715750792855471, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14725098182500015, "standard_deviation": 0.014724068061024073, "minimum": 0.12873983612132933, "maximum": 0.17566818819396268, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995504755390048, "standard_deviation": 0.001780970216717361, "minimum": 0.9916694348356985, "maximum": 0.9969016890585515, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7841674605019814, "standard_deviation": 0.021005498507165373, "minimum": 0.7563445206790178, "maximum": 0.8129414915962956, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1232788580545385, "standard_deviation": 0.2889702351819309, "minimum": 0.7251541201690043, "maximum": 1.4722067115114463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0878186134697925, "standard_deviation": 0.012588636569270686, "minimum": 0.07559384018767201, "maximum": 0.10941896772276012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9894899269804736, "standard_deviation": 0.0011160811770091626, "minimum": 0.9876635680948205, "maximum": 0.990660820510436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2089238852719304, "standard_deviation": 0.03572706848258074, "minimum": 1.151012689494887, "maximum": 1.2511340897770968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3611323389203841, "standard_deviation": 0.19197276944895653, "minimum": 0.11538166731620896, "maximum": 0.6654391480308771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11235211281214873, "standard_deviation": 0.005986755703267521, "minimum": 0.1063124620299864, "maximum": 0.1224410970943733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967055543136896, "standard_deviation": 0.0006000167826818448, "minimum": 0.9960291717948815, "maximum": 0.9976605277237954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6086910929160848, "standard_deviation": 0.04013353716377414, "minimum": 0.5563444620636323, "maximum": 0.6803056716345756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3441120470547646, "standard_deviation": 0.19448330856135904, "minimum": 0.08274439292005917, "maximum": 0.6599040576589263, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06212817991249142, "standard_deviation": 0.005710056093842775, "minimum": 0.053940065346804004, "maximum": 0.06987343061772687, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2125063301517786, "standard_deviation": 0.0383191389189236, "minimum": 0.156607803509662, "maximum": 0.2464431734734667, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6782817277012825, "standard_deviation": 0.005399524271389425, "minimum": 5.671275179344121, "maximum": 5.685261166328291, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6361674566824849, "standard_deviation": 0.013616020073023441, "minimum": 0.621557115470606, "maximum": 0.6595643470191231, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10100516312487531, "standard_deviation": 0.05688308770372945, "minimum": -0.19301133941610332, "maximum": -0.0017596290690461915, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9935263609251935, "standard_deviation": 0.0005486619257847125, "minimum": 0.9928014411502846, "maximum": 0.9942189278947694, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8626760324726869, "standard_deviation": 0.021546837683797827, "minimum": 0.8331133672401365, "maximum": 0.8939811162695491, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3662212081541184, "standard_deviation": 0.18959585201507786, "minimum": 0.11254356372727602, "maximum": 0.6292440219806015, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08649779060291023, "standard_deviation": 0.0037357245020848235, "minimum": 0.08189955353364672, "maximum": 0.09109212289427139, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980851071707268, "standard_deviation": 0.00032771613050408365, "minimum": 0.9975900146836336, "maximum": 0.9986225403061124, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4394867814362637, "standard_deviation": 0.024795979140696107, "minimum": 0.3981814834733737, "maximum": 0.47097154493427373, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3163641422092722, "standard_deviation": 0.19223629869733871, "minimum": 0.10978393626813114, "maximum": 0.6264751205767644, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04708838516137913, "standard_deviation": 0.003940555524895938, "minimum": 0.04074345421974594, "maximum": 0.05159134614626605, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_2961-961-0011", "language": "en", "style": "normal", "speaker": "librispeech_2961", "source_utterance_id": "2961-961-0011", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_2961-961-0011.wav", "source_file_sha256": "d500eb0932c1c8cf1731f1f45330a802d81748d48d1895d4ec1c514d69eb08d3", "decoded_pcm_sha256": "00bd4132c58a261e6169cd035b93bf1134d0bd1b082f44ceb99cf14faa9b7062", "original_source_sha256": "f859855c45fc07e8d561ed7a846fd14ca4ae1515749b60c4e8d4ce825df30180", "corpus_entry_id": "librispeech_2961-961-0011"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.471196174621582, "predicted_mos": 3.471196174621582, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7df238b38f176de7b24ec58cec1306675398b016a2e82f11b1512b374bffcffd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2303339099998993, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_2961-961-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7df238b38f176de7b24ec58cec1306675398b016a2e82f11b1512b374bffcffd", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "a0ddd73876a677f0ddbd97b071a56f85b42d159a0706fb433464b949cb2d2079"}, "rows": [{"case": "librispeech_2961-961-0011", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.8424274921417236, "predicted_mos": 2.8424274921417236, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9f766ca5b85802238bb53dc28be62baf2efec6178275d4ae14740e6ef05f6d7d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.598197757997696, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_2961-961-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9f766ca5b85802238bb53dc28be62baf2efec6178275d4ae14740e6ef05f6d7d", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.857236009940963, "全帯域_10ms_rms_envelope_nrmse": 0.08680204258280785, "全帯域_10ms_envelope_correlation": 0.9934930773298876, "全帯域_10ms_modulation_mae_db": 0.15781023829517088, "全帯域_40ms_log_rms_mae_db": 0.45811934445141705, "全帯域_40ms_rms_envelope_nrmse": 0.04090730715212322, "全帯域_40ms_envelope_correlation": 0.9985255856067436, "全帯域_40ms_modulation_mae_db": 0.3054232669026334, "300–1000 Hz_10ms_log_rms_mae_db": 1.4361397232296018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11078711440476954, "300–1000 Hz_10ms_envelope_correlation": 0.9904154571053392, "300–1000 Hz_10ms_modulation_mae_db": 0.28611107160421906, "300–1000 Hz_40ms_log_rms_mae_db": 0.787480253357002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05399168244942161, "300–1000 Hz_40ms_envelope_correlation": 0.9976863541129772, "300–1000 Hz_40ms_modulation_mae_db": 0.0750684500611052, "1000–3000 Hz_10ms_log_rms_mae_db": 1.380395205689077, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1365186092194447, "1000–3000 Hz_10ms_envelope_correlation": 0.9872419689591221, "1000–3000 Hz_10ms_modulation_mae_db": 1.0321033214349726, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8923028535841541, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07363902445366484, "1000–3000 Hz_40ms_envelope_correlation": 0.9961798669032674, "1000–3000 Hz_40ms_modulation_mae_db": 0.8798161015152219, "3000–6000 Hz_10ms_log_rms_mae_db": 1.341733270574663, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13118141241934184, "3000–6000 Hz_10ms_envelope_correlation": 0.9884364334099041, "3000–6000 Hz_10ms_modulation_mae_db": 0.22008566821183043, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8670978210700472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08019683660022382, "3000–6000 Hz_40ms_envelope_correlation": 0.9962000203478859, "3000–6000 Hz_40ms_modulation_mae_db": 0.19346764096418342, "mr_stft_db": 5.850271484019257, "band_mae_db": 0.1438678129602019, "rms_gain_error_db": -0.07272931899064802, "output_peak": 0.6080849366230011}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8696344174620917, "全帯域_10ms_rms_envelope_nrmse": 0.0937223824740253, "全帯域_10ms_envelope_correlation": 0.9924988001819761, "全帯域_10ms_modulation_mae_db": 0.5185759522173683, "全帯域_40ms_log_rms_mae_db": 0.471389008473773, "全帯域_40ms_rms_envelope_nrmse": 0.0482345969115324, "全帯域_40ms_envelope_correlation": 0.9980591427442613, "全帯域_40ms_modulation_mae_db": 0.25661675050414007, "300–1000 Hz_10ms_log_rms_mae_db": 1.4174089225468411, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12076656712578986, "300–1000 Hz_10ms_envelope_correlation": 0.9886832907407386, "300–1000 Hz_10ms_modulation_mae_db": 0.3796132750937676, "300–1000 Hz_40ms_log_rms_mae_db": 0.8001983719429822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06664650862175979, "300–1000 Hz_40ms_envelope_correlation": 0.9964919654672955, "300–1000 Hz_40ms_modulation_mae_db": 0.11318355008399937, "1000–3000 Hz_10ms_log_rms_mae_db": 1.454457755122123, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14373793543511276, "1000–3000 Hz_10ms_envelope_correlation": 0.9862939990124528, "1000–3000 Hz_10ms_modulation_mae_db": 0.6702655037135455, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9589879098899317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08436140553161274, "1000–3000 Hz_40ms_envelope_correlation": 0.9957746740295806, "1000–3000 Hz_40ms_modulation_mae_db": 0.7770937840776462, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3369371439887872, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13010916686261542, "3000–6000 Hz_10ms_envelope_correlation": 0.9882769090581038, "3000–6000 Hz_10ms_modulation_mae_db": 0.3135359565564163, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8252517922219236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07438403136814835, "3000–6000 Hz_40ms_envelope_correlation": 0.9963852541541877, "3000–6000 Hz_40ms_modulation_mae_db": 0.17296582095180432, "mr_stft_db": 5.846518716784764, "band_mae_db": 0.21236595372984346, "rms_gain_error_db": -0.13290589958590715, "output_peak": 0.5980781198758316}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8554566776103799, "全帯域_10ms_rms_envelope_nrmse": 0.08744368341522658, "全帯域_10ms_envelope_correlation": 0.9936442625728236, "全帯域_10ms_modulation_mae_db": 0.5622770985876183, "全帯域_40ms_log_rms_mae_db": 0.46143454843887616, "全帯域_40ms_rms_envelope_nrmse": 0.046332619776216034, "全帯域_40ms_envelope_correlation": 0.9984251091286773, "全帯域_40ms_modulation_mae_db": 0.38628413042404003, "300–1000 Hz_10ms_log_rms_mae_db": 1.4014200901192488, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11050862802703083, "300–1000 Hz_10ms_envelope_correlation": 0.9906723616107924, "300–1000 Hz_10ms_modulation_mae_db": 0.25191164437322255, "300–1000 Hz_40ms_log_rms_mae_db": 0.7573108269292526, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060256041738899126, "300–1000 Hz_40ms_envelope_correlation": 0.9974031176518454, "300–1000 Hz_40ms_modulation_mae_db": 0.31951970520973777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3565101769932117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15759348243348004, "1000–3000 Hz_10ms_envelope_correlation": 0.983815400732974, "1000–3000 Hz_10ms_modulation_mae_db": 1.4018373677245695, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8914313494898614, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0967925071212517, "1000–3000 Hz_40ms_envelope_correlation": 0.9947698208463688, "1000–3000 Hz_40ms_modulation_mae_db": 1.3060887769498817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3799986757630764, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1407690879411554, "3000–6000 Hz_10ms_envelope_correlation": 0.986190917597013, "3000–6000 Hz_10ms_modulation_mae_db": 0.48577153536955264, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8890155168886589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08558167841013045, "3000–6000 Hz_40ms_envelope_correlation": 0.9950371704415967, "3000–6000 Hz_40ms_modulation_mae_db": 0.653072359707934, "mr_stft_db": 5.850175665077759, "band_mae_db": 0.2308638108756712, "rms_gain_error_db": -0.18665807845503113, "output_peak": 0.6049670656453354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8866441762451885, "全帯域_10ms_rms_envelope_nrmse": 0.08949526732053639, "全帯域_10ms_envelope_correlation": 0.9931123278673187, "全帯域_10ms_modulation_mae_db": 0.4462940163919762, "全帯域_40ms_log_rms_mae_db": 0.49545886876238715, "全帯域_40ms_rms_envelope_nrmse": 0.04616174193623469, "全帯域_40ms_envelope_correlation": 0.9981609422799492, "全帯域_40ms_modulation_mae_db": 0.323721978062391, "300–1000 Hz_10ms_log_rms_mae_db": 1.4386810714040617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1113804885236246, "300–1000 Hz_10ms_envelope_correlation": 0.9903634400965904, "300–1000 Hz_10ms_modulation_mae_db": 0.4362056130115087, "300–1000 Hz_40ms_log_rms_mae_db": 0.8277951315258105, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05970380737679617, "300–1000 Hz_40ms_envelope_correlation": 0.9972196608328829, "300–1000 Hz_40ms_modulation_mae_db": 0.2743196410171911, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4378350077401807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1490621267764714, "1000–3000 Hz_10ms_envelope_correlation": 0.9847771018457959, "1000–3000 Hz_10ms_modulation_mae_db": 0.7782962112597355, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9138432605183452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08485072060632375, "1000–3000 Hz_40ms_envelope_correlation": 0.9950356455391954, "1000–3000 Hz_40ms_modulation_mae_db": 0.5314097318633677, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3477940056050104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14163890480428304, "3000–6000 Hz_10ms_envelope_correlation": 0.9858115156758789, "3000–6000 Hz_10ms_modulation_mae_db": 0.39556627229257896, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9044275854322712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08960775687083763, "3000–6000 Hz_40ms_envelope_correlation": 0.9941399038885225, "3000–6000 Hz_40ms_modulation_mae_db": 0.2992724581079416, "mr_stft_db": 5.850631290827658, "band_mae_db": 0.1482601864858565, "rms_gain_error_db": -0.10262082852696212, "output_peak": 0.6788528059550636}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9032893704783316, "全帯域_10ms_rms_envelope_nrmse": 0.108401549337408, "全帯域_10ms_envelope_correlation": 0.9898948620582905, "全帯域_10ms_modulation_mae_db": 0.5510578098921186, "全帯域_40ms_log_rms_mae_db": 0.5159155276201592, "全帯域_40ms_rms_envelope_nrmse": 0.061999677291615596, "全帯域_40ms_envelope_correlation": 0.9966850370655854, "全帯域_40ms_modulation_mae_db": 0.8561575720802654, "300–1000 Hz_10ms_log_rms_mae_db": 1.3966224671652232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12854036020094234, "300–1000 Hz_10ms_envelope_correlation": 0.9870651524570078, "300–1000 Hz_10ms_modulation_mae_db": 0.6977129425903182, "300–1000 Hz_40ms_log_rms_mae_db": 0.7924698275234642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07457268551715998, "300–1000 Hz_40ms_envelope_correlation": 0.9955383400253597, "300–1000 Hz_40ms_modulation_mae_db": 0.9637756997869961, "1000–3000 Hz_10ms_log_rms_mae_db": 1.419393455643684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14564916035301462, "1000–3000 Hz_10ms_envelope_correlation": 0.9861249531763672, "1000–3000 Hz_10ms_modulation_mae_db": 0.9442314471005503, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9233440592174382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09518828128152224, "1000–3000 Hz_40ms_envelope_correlation": 0.9946080298407383, "1000–3000 Hz_40ms_modulation_mae_db": 0.5298718467633269, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3326344100557603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14202827781578134, "3000–6000 Hz_10ms_envelope_correlation": 0.9867587486754978, "3000–6000 Hz_10ms_modulation_mae_db": 1.0427861851421314, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8983927281969657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0927907025188954, "3000–6000 Hz_40ms_envelope_correlation": 0.9951185362033471, "3000–6000 Hz_40ms_modulation_mae_db": 1.0438204193122178, "mr_stft_db": 5.8402866131273, "band_mae_db": 0.26081651983163295, "rms_gain_error_db": -0.14236411590282852, "output_peak": 0.5886377192685227}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.884754845969899, "全帯域_10ms_rms_envelope_nrmse": 0.08929753619379052, "全帯域_10ms_envelope_correlation": 0.9935872095840689, "全帯域_10ms_modulation_mae_db": 0.5403975770912304, "全帯域_40ms_log_rms_mae_db": 0.45854655305182146, "全帯域_40ms_rms_envelope_nrmse": 0.049228436205001634, "全帯域_40ms_envelope_correlation": 0.998453703645937, "全帯域_40ms_modulation_mae_db": 0.5183424125915795, "300–1000 Hz_10ms_log_rms_mae_db": 1.4380834797370567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11253488750509602, "300–1000 Hz_10ms_envelope_correlation": 0.990437289258272, "300–1000 Hz_10ms_modulation_mae_db": 0.43124244855925475, "300–1000 Hz_40ms_log_rms_mae_db": 0.75379608750214, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06116861931606962, "300–1000 Hz_40ms_envelope_correlation": 0.9974746543239266, "300–1000 Hz_40ms_modulation_mae_db": 0.45407781190465313, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4270124805943225, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15899697534025803, "1000–3000 Hz_10ms_envelope_correlation": 0.9836591511255299, "1000–3000 Hz_10ms_modulation_mae_db": 1.2208189916947099, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9326905957154049, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11266376169519807, "1000–3000 Hz_40ms_envelope_correlation": 0.9924597172808025, "1000–3000 Hz_40ms_modulation_mae_db": 0.9731040361136882, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3983041576550406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14706442172980136, "3000–6000 Hz_10ms_envelope_correlation": 0.9850358255612559, "3000–6000 Hz_10ms_modulation_mae_db": 0.30572141387610285, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9094435687857618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08760903997836332, "3000–6000 Hz_40ms_envelope_correlation": 0.9950395662775092, "3000–6000 Hz_40ms_modulation_mae_db": 0.2520474387410992, "mr_stft_db": 5.850971148533833, "band_mae_db": 0.23356456738614131, "rms_gain_error_db": -0.24251821416777436, "output_peak": 0.608178643505304}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8356833814042384, "全帯域_10ms_rms_envelope_nrmse": 0.08372606707597462, "全帯域_10ms_envelope_correlation": 0.9940475860992615, "全帯域_10ms_modulation_mae_db": 0.31113147667180524, "全帯域_40ms_log_rms_mae_db": 0.466825076571833, "全帯域_40ms_rms_envelope_nrmse": 0.04232648634939969, "全帯域_40ms_envelope_correlation": 0.9985575407476052, "全帯域_40ms_modulation_mae_db": 0.6392114044894361, "300–1000 Hz_10ms_log_rms_mae_db": 1.4036087197937306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1123246990977042, "300–1000 Hz_10ms_envelope_correlation": 0.9902129252689206, "300–1000 Hz_10ms_modulation_mae_db": 0.26705370410834234, "300–1000 Hz_40ms_log_rms_mae_db": 0.7920098263294992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05886397027702977, "300–1000 Hz_40ms_envelope_correlation": 0.9973571326711165, "300–1000 Hz_40ms_modulation_mae_db": 0.172108364367192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3924380073493254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14539594305094397, "1000–3000 Hz_10ms_envelope_correlation": 0.9856412525738297, "1000–3000 Hz_10ms_modulation_mae_db": 0.7992437295686855, "1000–3000 Hz_40ms_log_rms_mae_db": 0.919951465941995, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08238720785136522, "1000–3000 Hz_40ms_envelope_correlation": 0.9956172427993227, "1000–3000 Hz_40ms_modulation_mae_db": 0.42103533827896883, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3656463062559283, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14286497923981237, "3000–6000 Hz_10ms_envelope_correlation": 0.9859088974466116, "3000–6000 Hz_10ms_modulation_mae_db": 0.44546320132899114, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8671883392448174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08445378287096583, "3000–6000 Hz_40ms_envelope_correlation": 0.9954017551252974, "3000–6000 Hz_40ms_modulation_mae_db": 0.46453979661199235, "mr_stft_db": 5.84425110147999, "band_mae_db": 0.1701343885111155, "rms_gain_error_db": -0.13170688496623856, "output_peak": 0.6019765423973215}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8830880959344201, "全帯域_10ms_rms_envelope_nrmse": 0.10164745341417727, "全帯域_10ms_envelope_correlation": 0.9912046864455981, "全帯域_10ms_modulation_mae_db": 0.478612906715912, "全帯域_40ms_log_rms_mae_db": 0.4801226200224966, "全帯域_40ms_rms_envelope_nrmse": 0.05496936225164667, "全帯域_40ms_envelope_correlation": 0.9975163719485665, "全帯域_40ms_modulation_mae_db": 0.23618474373007878, "300–1000 Hz_10ms_log_rms_mae_db": 1.4705239647833706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13204845710572186, "300–1000 Hz_10ms_envelope_correlation": 0.9865226710692185, "300–1000 Hz_10ms_modulation_mae_db": 0.21136124609596948, "300–1000 Hz_40ms_log_rms_mae_db": 0.8271043766956474, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0720074394851741, "300–1000 Hz_40ms_envelope_correlation": 0.9960902441051819, "300–1000 Hz_40ms_modulation_mae_db": 0.42897867313369226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3996063552448719, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13949523089146437, "1000–3000 Hz_10ms_envelope_correlation": 0.9867457227297223, "1000–3000 Hz_10ms_modulation_mae_db": 1.1514013820928501, "1000–3000 Hz_40ms_log_rms_mae_db": 0.907428982775291, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07989000905368536, "1000–3000 Hz_40ms_envelope_correlation": 0.9957023166075952, "1000–3000 Hz_40ms_modulation_mae_db": 0.8252645903206416, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3848522207863152, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13554983925272712, "3000–6000 Hz_10ms_envelope_correlation": 0.9870713919808423, "3000–6000 Hz_10ms_modulation_mae_db": 0.41401564676197466, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9094527034451877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07639796332437765, "3000–6000 Hz_40ms_envelope_correlation": 0.9959038198945369, "3000–6000 Hz_40ms_modulation_mae_db": 0.23734814134325152, "mr_stft_db": 5.8555539198981705, "band_mae_db": 0.1996823089470027, "rms_gain_error_db": -0.16413317345990455, "output_peak": 0.6212319288175109}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.731725993047649, "spectral_convergence": 0.31233115880017354, "spectral_convergence_status": "ok", "amplitude_floor": 1.4064891899964367e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.617900716204186, "spectral_convergence": 0.29745837496058736, "spectral_convergence_status": "ok", "amplitude_floor": 1.2304331075199438e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.201187742805934, "spectral_convergence": 0.3137109185537112, "spectral_convergence_status": "ok", "amplitude_floor": 7.896502575224385e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.444718915891811e-05, "output_power": 5.654423413898074e-05, "delta_db": 0.16412861983255816, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025153246625132694, "output_power": 0.0002657801214545722, "delta_db": 0.23928446498726122, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010938925867664089, "output_power": 0.0010527253493227784, "delta_db": -0.16659598412478407, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031036369064007244, "output_power": 0.00030924870156545756, "delta_db": -0.015630227260097225, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001328835538454751, "output_power": 0.00012878341880691717, "delta_db": -0.13611284292654396, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6997042780401114e-05, "output_power": 3.5811419655981325e-05, "delta_db": -0.14145473862996677, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1438678129602019, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07272931899064802, "pitch": null, "output_rms_dbfs": -27.330881642707432, "output_peak": 0.6080849366230011, "reference_rms_dbfs": -27.258152323716782, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.850271484019257}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019413100530357063, "log_rms_mae_db": 0.857236009940963, "rms_envelope_nrmse": 0.08680204258280785, "envelope_correlation": 0.9934930773298876, "rms_gain_error_db": -0.07270787631678535, "reference_mean_power": 0.0018816420370153617, "output_mean_power": 0.0018504025543371158}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1507655155627412e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005726194534917747, "output_power": 0.0005650178625390801, "reference_active": true, "delta_db": -0.058039202683299}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.421513884012437e-05, "output_power": 8.587961079857588e-05, "reference_active": true, "delta_db": -0.40230624886997546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.679657413963232e-05, "output_power": 1.674604233969809e-05, "reference_active": true, "delta_db": -0.013085263332238172}], "active_band_count": 3, "active_band_mae_db": 0.15781023829517088}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016416501002274435, "log_rms_mae_db": 0.45811934445141705, "rms_envelope_nrmse": 0.04090730715212322, "envelope_correlation": 0.9985255856067436, "rms_gain_error_db": -0.07272201640753777, "reference_mean_power": 0.001890759403997248, "output_mean_power": 0.0018593624986918207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0781635397104837e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005383866955860211, "output_power": 0.0005303819515854988, "reference_active": true, "delta_db": -0.06505582220343178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8874025426333776e-05, "output_power": 3.641204137065125e-05, "reference_active": true, "delta_db": -0.28414486891700236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650199092025111e-06, "output_power": 1.373450322845301e-06, "reference_active": true, "delta_db": -0.5670691095874659}], "active_band_count": 3, "active_band_mae_db": 0.3054232669026334}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017990629858894173, "log_rms_mae_db": 1.4361397232296018, "rms_envelope_nrmse": 0.11078711440476954, "envelope_correlation": 0.9904154571053392, "rms_gain_error_db": -0.10032252080302978, "reference_mean_power": 0.0012781810161111924, "output_mean_power": 0.0012489933064298716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.355790711503392e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004472139003696596, "output_power": 0.0004371320782257785, "reference_active": true, "delta_db": -0.09902616242044046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.621773891011176e-05, "output_power": 7.186262959182713e-05, "reference_active": true, "delta_db": -0.2555295593509486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7010237353267386e-05, "output_power": 1.9102408271656602e-05, "reference_active": true, "delta_db": 0.5037774930412682}], "active_band_count": 3, "active_band_mae_db": 0.28611107160421906}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000149706333097475, "log_rms_mae_db": 0.787480253357002, "rms_envelope_nrmse": 0.05399168244942161, "envelope_correlation": 0.9976863541129772, "rms_gain_error_db": -0.10033639879241268, "reference_mean_power": 0.0012844845607787134, "output_mean_power": 0.0012551488965882416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.840745806614025e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004236215690036061, "output_power": 0.00041463857838092534, "reference_active": true, "delta_db": -0.09308357597982346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1600142098026355e-05, "output_power": 3.081354776614704e-05, "reference_active": true, "delta_db": -0.10947331165669041}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5445822829049627e-06, "output_power": 1.5365482513592384e-06, "reference_active": true, "delta_db": -0.02264846254680173}], "active_band_count": 3, "active_band_mae_db": 0.0750684500611052}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010005798255669499, "log_rms_mae_db": 1.380395205689077, "rms_envelope_nrmse": 0.1365186092194447, "envelope_correlation": 0.9872419689591221, "rms_gain_error_db": -0.050010220657927464, "reference_mean_power": 0.0003605696849433608, "output_mean_power": 0.00035644143904229876}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5037729366163507e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001406674030835393, "output_power": 0.00013533340049815584, "reference_active": true, "delta_db": -0.16788475456374627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0544824834133773e-05, "output_power": 2.4709782079026215e-05, "reference_active": true, "delta_db": 0.8016647247658127}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4458512826457347e-06, "output_power": 5.623058199788831e-06, "reference_active": true, "delta_db": 2.126760484975359}], "active_band_count": 3, "active_band_mae_db": 1.0321033214349726}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.18211380942645e-05, "log_rms_mae_db": 0.8923028535841541, "rms_envelope_nrmse": 0.07363902445366484, "envelope_correlation": 0.9961798669032674, "rms_gain_error_db": -0.05001035548492492, "reference_mean_power": 0.00036238111197119544, "output_mean_power": 0.00035823211550087356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3815388264892723e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001355192968806531, "output_power": 0.00013261460540285426, "reference_active": true, "delta_db": -0.09409782351600543}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.971075792205542e-06, "output_power": 1.0420487754976918e-05, "reference_active": true, "delta_db": 1.1637110899795258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.416322058192608e-07, "output_power": 3.3213814639915057e-07, "reference_active": true, "delta_db": 1.3816393910501346}], "active_band_count": 3, "active_band_mae_db": 0.8798161015152219}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.283283642007837e-05, "log_rms_mae_db": 1.341733270574663, "rms_envelope_nrmse": 0.13118141241934184, "envelope_correlation": 0.9884364334099041, "rms_gain_error_db": -0.3044363098831466, "reference_mean_power": 6.230386432542126e-05, "output_mean_power": 5.8085984299869825e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0221136919924677e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.16042156208526e-05, "output_power": 1.1013329422872069e-05, "reference_active": true, "delta_db": -0.2269716021532285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3772446197384082e-06, "output_power": 2.2418020830661278e-06, "reference_active": true, "delta_db": -0.2547660478918457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4509999616936e-07, "output_power": 4.2717482981870173e-07, "reference_active": true, "delta_db": -0.1785193545904171}], "active_band_count": 3, "active_band_mae_db": 0.22008566821183043}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.758960619357764e-05, "log_rms_mae_db": 0.8670978210700472, "rms_envelope_nrmse": 0.08019683660022382, "envelope_correlation": 0.9962000203478859, "rms_gain_error_db": -0.3044375456015043, "reference_mean_power": 6.26170059672204e-05, "output_mean_power": 5.837791010230887e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.846748801507289e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1232117533806203e-05, "output_power": 1.0634642404877415e-05, "reference_active": true, "delta_db": -0.23738748185834097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.067581168241283e-06, "output_power": 1.0708787208497973e-06, "reference_active": true, "delta_db": 0.013393845206783227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.579800269976513e-08, "output_power": 2.391244170471094e-08, "reference_active": true, "delta_db": -0.329621595827426}], "active_band_count": 3, "active_band_mae_db": 0.19346764096418342}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8445197101023576, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5dc0c977ea58446b86bb9afaeda8c27f401cff3d521ceb970280346f7939b24e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9855374437694742, "standard_deviation": 0.0013290812557878878, "minimum": 0.9836591511255299, "maximum": 0.9872419689591221, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4084560555470995, "standard_deviation": 0.03221341221149804, "minimum": 1.3565101769932117, "maximum": 1.454457755122123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9997747443237024, "standard_deviation": 0.2494214770829728, "minimum": 0.6702655037135455, "maximum": 1.4018373677245695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14705618293752376, "standard_deviation": 0.007942618727796584, "minimum": 0.1365186092194447, "maximum": 0.15899697534025803, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950184142308589, "standard_deviation": 0.0011664757867429693, "minimum": 0.9924597172808025, "maximum": 0.9961798669032674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9174975596415527, "standard_deviation": 0.02210827701977334, "minimum": 0.8914313494898614, "maximum": 0.9589879098899317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7804605257353429, "standard_deviation": 0.287673996911261, "minimum": 0.42103533827896883, "maximum": 1.3060887769498817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08872161469932799, "standard_deviation": 0.012304384967058955, "minimum": 0.07363902445366484, "maximum": 0.11266376169519807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9866863299256384, "standard_deviation": 0.0012003345192778587, "minimum": 0.9850358255612559, "maximum": 0.9884364334099041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3609875238355726, "standard_deviation": 0.024704823401626778, "minimum": 1.3326344100557603, "maximum": 1.3983041576550406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4528682349424473, "standard_deviation": 0.25331715928719123, "minimum": 0.22008566821183043, "maximum": 1.0427861851421314, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13890076125818973, "standard_deviation": 0.005990933242250177, "minimum": 0.13010916686261542, "maximum": 0.14706442172980136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954032532916104, "standard_deviation": 0.0007361223730242396, "minimum": 0.9941399038885225, "maximum": 0.9963852541541877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8837837569107041, "standard_deviation": 0.02918928620696186, "minimum": 0.8252517922219236, "maximum": 0.9094527034451877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.414566759467553, "standard_deviation": 0.30088203005148106, "minimum": 0.17296582095180432, "maximum": 1.0438204193122178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0838777239927428, "standard_deviation": 0.006426045637722798, "minimum": 0.07438403136814835, "maximum": 0.0927907025188954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9892965734508599, "standard_deviation": 0.0016679209084537394, "minimum": 0.9865226710692185, "maximum": 0.9906723616107924, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4253110548473917, "standard_deviation": 0.02515837765781917, "minimum": 1.3966224671652232, "maximum": 1.4705239647833706, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3701514931795753, "standard_deviation": 0.15692046712431548, "minimum": 0.21136124609596948, "maximum": 0.6977129425903182, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1173614002488349, "standard_deviation": 0.008675833072289318, "minimum": 0.11050862802703083, "maximum": 0.13204845710572186, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969076836488232, "standard_deviation": 0.0007735542690002281, "minimum": 0.9955383400253597, "maximum": 0.9976863541129772, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7922705877257248, "standard_deviation": 0.027422000728218265, "minimum": 0.75379608750214, "maximum": 0.8277951315258105, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3501289869455709, "standard_deviation": 0.28379109155599025, "minimum": 0.0750684500611052, "maximum": 0.9637756997869961, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06340134434778877, "standard_deviation": 0.007041449707892572, "minimum": 0.05399168244942161, "maximum": 0.07457268551715998, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1999444435909332, "standard_deviation": 0.04251148182904862, "minimum": 0.1438678129602019, "maximum": 0.26081651983163295, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.848582492468592, "standard_deviation": 0.004716788804070056, "minimum": 5.8402866131273, "maximum": 5.8555539198981705, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6137509702609864, "standard_deviation": 0.027901552083662615, "minimum": 0.5886377192685227, "maximum": 0.6788528059550636, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1469545642569118, "standard_deviation": 0.05200230889874906, "minimum": -0.24251821416777436, "maximum": -0.07272931899064802, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9926853515174031, "standard_deviation": 0.0014361293768420471, "minimum": 0.9898948620582905, "maximum": 0.9940475860992615, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.871973371880689, "standard_deviation": 0.021696782456218186, "minimum": 0.8356833814042384, "maximum": 0.9032893704783316, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.44576963448289997, "standard_deviation": 0.14194656912090828, "minimum": 0.15781023829517088, "maximum": 0.5622770985876183, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09256699772674332, "standard_deviation": 0.008386150424856645, "minimum": 0.08372606707597462, "maximum": 0.108401549337408, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980479291459157, "standard_deviation": 0.0006489225083985083, "minimum": 0.9966850370655854, "maximum": 0.9985575407476052, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47597644342409545, "standard_deviation": 0.020461887814130053, "minimum": 0.45811934445141705, "maximum": 0.5159155276201592, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44024278234807057, "standard_deviation": 0.21669475422094656, "minimum": 0.23618474373007878, "maximum": 0.8561575720802654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048770028484221246, "standard_deviation": 0.006866460028388014, "minimum": 0.04090730715212322, "maximum": 0.061999677291615596, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-961-0011", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.857811212539673, "predicted_mos": 2.857811212539673, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "70b503ed7cb72eeaf1b6aa5339d0710f7f2199e1c1d52145f8b8d3c0c0840875", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.184743399731815, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8624244289326648, "全帯域_10ms_rms_envelope_nrmse": 0.08688865107106608, "全帯域_10ms_envelope_correlation": 0.9934820194941633, "全帯域_10ms_modulation_mae_db": 0.23747749531405504, "全帯域_40ms_log_rms_mae_db": 0.45764618784768724, "全帯域_40ms_rms_envelope_nrmse": 0.041361669874356397, "全帯域_40ms_envelope_correlation": 0.9984954672060866, "全帯域_40ms_modulation_mae_db": 0.33681022843834046, "300–1000 Hz_10ms_log_rms_mae_db": 1.4027645346533963, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11073322688694137, "300–1000 Hz_10ms_envelope_correlation": 0.9904122353648455, "300–1000 Hz_10ms_modulation_mae_db": 0.2951102659326737, "300–1000 Hz_40ms_log_rms_mae_db": 0.7626331368518523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052724703370312605, "300–1000 Hz_40ms_envelope_correlation": 0.997779473834152, "300–1000 Hz_40ms_modulation_mae_db": 0.15159400023974576, "1000–3000 Hz_10ms_log_rms_mae_db": 1.388135977831511, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13443966812591812, "1000–3000 Hz_10ms_envelope_correlation": 0.9876358592114088, "1000–3000 Hz_10ms_modulation_mae_db": 1.1097197731182813, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8833304586620812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07628266966582166, "1000–3000 Hz_40ms_envelope_correlation": 0.9959160985985258, "1000–3000 Hz_40ms_modulation_mae_db": 0.9499824031148264, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2981729090565974, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12838436435178271, "3000–6000 Hz_10ms_envelope_correlation": 0.9889524859720563, "3000–6000 Hz_10ms_modulation_mae_db": 0.25431046263298956, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8336522880780662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07787427714491182, "3000–6000 Hz_40ms_envelope_correlation": 0.9964443790353688, "3000–6000 Hz_40ms_modulation_mae_db": 0.27635764959889425, "mr_stft_db": 5.792374127264874, "band_mae_db": 0.15147763901491815, "rms_gain_error_db": -0.07518787391861183, "output_peak": 0.6107767930922043}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8830262824898837, "全帯域_10ms_rms_envelope_nrmse": 0.0964342123827631, "全帯域_10ms_envelope_correlation": 0.9920472324856356, "全帯域_10ms_modulation_mae_db": 0.44209650836141917, "全帯域_40ms_log_rms_mae_db": 0.4744624361702084, "全帯域_40ms_rms_envelope_nrmse": 0.05066987767069125, "全帯域_40ms_envelope_correlation": 0.9978274430242812, "全帯域_40ms_modulation_mae_db": 0.1389376914240324, "300–1000 Hz_10ms_log_rms_mae_db": 1.4458782365697433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12304582681458677, "300–1000 Hz_10ms_envelope_correlation": 0.9882610132951241, "300–1000 Hz_10ms_modulation_mae_db": 0.3414803053089744, "300–1000 Hz_40ms_log_rms_mae_db": 0.7959531920863533, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06815724284588488, "300–1000 Hz_40ms_envelope_correlation": 0.996332433908156, "300–1000 Hz_40ms_modulation_mae_db": 0.06622492009001622, "1000–3000 Hz_10ms_log_rms_mae_db": 1.44458876779746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.141763139865096, "1000–3000 Hz_10ms_envelope_correlation": 0.9865437204049162, "1000–3000 Hz_10ms_modulation_mae_db": 0.7975692374901028, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9278044934475035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08523351812074224, "1000–3000 Hz_40ms_envelope_correlation": 0.9954378053183558, "1000–3000 Hz_40ms_modulation_mae_db": 0.6680769883617534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.333224966421044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12681923522133706, "3000–6000 Hz_10ms_envelope_correlation": 0.9889727451320856, "3000–6000 Hz_10ms_modulation_mae_db": 0.4668753726940844, "3000–6000 Hz_40ms_log_rms_mae_db": 0.843835896517163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07566920993843067, "3000–6000 Hz_40ms_envelope_correlation": 0.9963388188339114, "3000–6000 Hz_40ms_modulation_mae_db": 0.5077729839017172, "mr_stft_db": 5.800911694085321, "band_mae_db": 0.1955341593476875, "rms_gain_error_db": -0.12910307278306635, "output_peak": 0.5945532725070438}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8685090951155875, "全帯域_10ms_rms_envelope_nrmse": 0.08745941594426931, "全帯域_10ms_envelope_correlation": 0.99361137140813, "全帯域_10ms_modulation_mae_db": 0.6242539686778156, "全帯域_40ms_log_rms_mae_db": 0.4682042310636907, "全帯域_40ms_rms_envelope_nrmse": 0.04460650798367884, "全帯域_40ms_envelope_correlation": 0.9985336472420343, "全帯域_40ms_modulation_mae_db": 0.4356247701449421, "300–1000 Hz_10ms_log_rms_mae_db": 1.3908141033052472, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1092469751215189, "300–1000 Hz_10ms_envelope_correlation": 0.9908426845809366, "300–1000 Hz_10ms_modulation_mae_db": 0.2786026674077631, "300–1000 Hz_40ms_log_rms_mae_db": 0.7507335743161356, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05763113739442029, "300–1000 Hz_40ms_envelope_correlation": 0.9975957929782533, "300–1000 Hz_40ms_modulation_mae_db": 0.09835769049193416, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3895872751731122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15054664311322935, "1000–3000 Hz_10ms_envelope_correlation": 0.9853218285256033, "1000–3000 Hz_10ms_modulation_mae_db": 1.1115934041858921, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8889194544111725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0903114888605162, "1000–3000 Hz_40ms_envelope_correlation": 0.9955867537815141, "1000–3000 Hz_40ms_modulation_mae_db": 1.0487822254267238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.332131710036898, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13794054319849924, "3000–6000 Hz_10ms_envelope_correlation": 0.9867611200776112, "3000–6000 Hz_10ms_modulation_mae_db": 0.8031582578042329, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8458544825376502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08278160897318217, "3000–6000 Hz_40ms_envelope_correlation": 0.9953786676889782, "3000–6000 Hz_40ms_modulation_mae_db": 0.9170943610606921, "mr_stft_db": 5.799105904292453, "band_mae_db": 0.22154178613986752, "rms_gain_error_db": -0.17782406702766404, "output_peak": 0.6015331535334792}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8986216014853652, "全帯域_10ms_rms_envelope_nrmse": 0.09284027666468082, "全帯域_10ms_envelope_correlation": 0.9925813006878041, "全帯域_10ms_modulation_mae_db": 0.5362916376594625, "全帯域_40ms_log_rms_mae_db": 0.5186792229689648, "全帯域_40ms_rms_envelope_nrmse": 0.04819425968485145, "全帯域_40ms_envelope_correlation": 0.9979901227114232, "全帯域_40ms_modulation_mae_db": 0.3165711351498388, "300–1000 Hz_10ms_log_rms_mae_db": 1.3949855933836404, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1144654019092132, "300–1000 Hz_10ms_envelope_correlation": 0.9898011746387131, "300–1000 Hz_10ms_modulation_mae_db": 0.539701678708715, "300–1000 Hz_40ms_log_rms_mae_db": 0.7968037916383093, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06128992770725049, "300–1000 Hz_40ms_envelope_correlation": 0.9970415043442203, "300–1000 Hz_40ms_modulation_mae_db": 0.3274103529275128, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4419903780069885, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14919522825839301, "1000–3000 Hz_10ms_envelope_correlation": 0.9847738921448147, "1000–3000 Hz_10ms_modulation_mae_db": 0.7674568217861518, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8896082620695739, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08216305798991336, "1000–3000 Hz_40ms_envelope_correlation": 0.9953978145148947, "1000–3000 Hz_40ms_modulation_mae_db": 0.46241772051744645, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3341873668689543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1455110082562391, "3000–6000 Hz_10ms_envelope_correlation": 0.9850197218298012, "3000–6000 Hz_10ms_modulation_mae_db": 0.2685637566892209, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9019847088453775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09503557772009484, "3000–6000 Hz_40ms_envelope_correlation": 0.9933486400715801, "3000–6000 Hz_40ms_modulation_mae_db": 0.3268864965846849, "mr_stft_db": 5.797603925306866, "band_mae_db": 0.147594489392847, "rms_gain_error_db": -0.10513006795139022, "output_peak": 0.6799491838763909}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9136439128797046, "全帯域_10ms_rms_envelope_nrmse": 0.1109140892942601, "全帯域_10ms_envelope_correlation": 0.989432952301244, "全帯域_10ms_modulation_mae_db": 0.5282797300725249, "全帯域_40ms_log_rms_mae_db": 0.5291288605372838, "全帯域_40ms_rms_envelope_nrmse": 0.06423457831759724, "全帯域_40ms_envelope_correlation": 0.9964564921380631, "全帯域_40ms_modulation_mae_db": 0.8314717528388789, "300–1000 Hz_10ms_log_rms_mae_db": 1.399335676792107, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13183164268828315, "300–1000 Hz_10ms_envelope_correlation": 0.9864045115174698, "300–1000 Hz_10ms_modulation_mae_db": 0.6647599793151749, "300–1000 Hz_40ms_log_rms_mae_db": 0.8009415479498901, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07750028171729492, "300–1000 Hz_40ms_envelope_correlation": 0.9952015102304715, "300–1000 Hz_40ms_modulation_mae_db": 0.9053494406012477, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4625089166957423, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14999042766074377, "1000–3000 Hz_10ms_envelope_correlation": 0.9851867899959108, "1000–3000 Hz_10ms_modulation_mae_db": 0.978277451998122, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9341055426519121, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09593992878215804, "1000–3000 Hz_40ms_envelope_correlation": 0.994459327619639, "1000–3000 Hz_40ms_modulation_mae_db": 0.586724985695895, "3000–6000 Hz_10ms_log_rms_mae_db": 1.278064730370758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14198762310205662, "3000–6000 Hz_10ms_envelope_correlation": 0.9868921746870717, "3000–6000 Hz_10ms_modulation_mae_db": 1.3120830929820724, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8419345022109297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09360230978827258, "3000–6000 Hz_40ms_envelope_correlation": 0.9951455252491407, "3000–6000 Hz_40ms_modulation_mae_db": 1.3333947828428485, "mr_stft_db": 5.793005483674162, "band_mae_db": 0.2687166391604486, "rms_gain_error_db": -0.1535222706116165, "output_peak": 0.5899076915202265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8986072998601965, "全帯域_10ms_rms_envelope_nrmse": 0.09098070238962196, "全帯域_10ms_envelope_correlation": 0.9933190935576475, "全帯域_10ms_modulation_mae_db": 0.5438315848810863, "全帯域_40ms_log_rms_mae_db": 0.4774220440798729, "全帯域_40ms_rms_envelope_nrmse": 0.05103978629263168, "全帯域_40ms_envelope_correlation": 0.9982859993888374, "全帯域_40ms_modulation_mae_db": 0.4590487918305903, "300–1000 Hz_10ms_log_rms_mae_db": 1.4271996954284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11180879980918429, "300–1000 Hz_10ms_envelope_correlation": 0.9905507224192021, "300–1000 Hz_10ms_modulation_mae_db": 0.5447045188104637, "300–1000 Hz_40ms_log_rms_mae_db": 0.7667315272242667, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06135808257407186, "300–1000 Hz_40ms_envelope_correlation": 0.9974302278964398, "300–1000 Hz_40ms_modulation_mae_db": 0.5221691438590924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4632554319762099, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16114293531212753, "1000–3000 Hz_10ms_envelope_correlation": 0.9832502634259068, "1000–3000 Hz_10ms_modulation_mae_db": 1.263394551979552, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9346353664159717, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11150690415240164, "1000–3000 Hz_40ms_envelope_correlation": 0.9927523120930842, "1000–3000 Hz_40ms_modulation_mae_db": 1.1097553718184698, "3000–6000 Hz_10ms_log_rms_mae_db": 1.366684572240933, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1466559942901341, "3000–6000 Hz_10ms_envelope_correlation": 0.9850856779819643, "3000–6000 Hz_10ms_modulation_mae_db": 0.47803640775751166, "3000–6000 Hz_40ms_log_rms_mae_db": 0.889605986312594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08607829240213731, "3000–6000 Hz_40ms_envelope_correlation": 0.9951779138038724, "3000–6000 Hz_40ms_modulation_mae_db": 0.3712146462260007, "mr_stft_db": 5.79947387959037, "band_mae_db": 0.22917115790747042, "rms_gain_error_db": -0.2430524309762587, "output_peak": 0.6084968100770289}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8436690174179486, "全帯域_10ms_rms_envelope_nrmse": 0.08623739307664642, "全帯域_10ms_envelope_correlation": 0.9936805427989378, "全帯域_10ms_modulation_mae_db": 0.2857590581650414, "全帯域_40ms_log_rms_mae_db": 0.46818232545779875, "全帯域_40ms_rms_envelope_nrmse": 0.044264458204883735, "全帯域_40ms_envelope_correlation": 0.9984143477721226, "全帯域_40ms_modulation_mae_db": 0.5339240851084391, "300–1000 Hz_10ms_log_rms_mae_db": 1.3596840897610343, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11249218865233085, "300–1000 Hz_10ms_envelope_correlation": 0.9901605399551142, "300–1000 Hz_10ms_modulation_mae_db": 0.35843868929926037, "300–1000 Hz_40ms_log_rms_mae_db": 0.7408569340764554, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057166136796632135, "300–1000 Hz_40ms_envelope_correlation": 0.9974872229435113, "300–1000 Hz_40ms_modulation_mae_db": 0.12749815919185042, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4117646970288003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15054362114392492, "1000–3000 Hz_10ms_envelope_correlation": 0.9846965034956427, "1000–3000 Hz_10ms_modulation_mae_db": 0.6649646030288742, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9074679391525189, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08559750423588354, "1000–3000 Hz_40ms_envelope_correlation": 0.9954083540144726, "1000–3000 Hz_40ms_modulation_mae_db": 0.25577627435942923, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3273874831681938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14055827616127856, "3000–6000 Hz_10ms_envelope_correlation": 0.98643033913042, "3000–6000 Hz_10ms_modulation_mae_db": 0.4867574235971941, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8356447335836966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08121953747190687, "3000–6000 Hz_40ms_envelope_correlation": 0.9958615454988029, "3000–6000 Hz_40ms_modulation_mae_db": 0.5738713461588555, "mr_stft_db": 5.803713347800696, "band_mae_db": 0.1666124871951943, "rms_gain_error_db": -0.13549380807055073, "output_peak": 0.6023700125289643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9155926806252873, "全帯域_10ms_rms_envelope_nrmse": 0.1023066156739984, "全帯域_10ms_envelope_correlation": 0.9910617338161786, "全帯域_10ms_modulation_mae_db": 0.4643995720385488, "全帯域_40ms_log_rms_mae_db": 0.4866957949540017, "全帯域_40ms_rms_envelope_nrmse": 0.05433040855363654, "全帯域_40ms_envelope_correlation": 0.9975525837269776, "全帯域_40ms_modulation_mae_db": 0.30409434534505514, "300–1000 Hz_10ms_log_rms_mae_db": 1.4506684331772084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1323321391139146, "300–1000 Hz_10ms_envelope_correlation": 0.986424305842912, "300–1000 Hz_10ms_modulation_mae_db": 0.23008520839967173, "300–1000 Hz_40ms_log_rms_mae_db": 0.8077549513624551, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07190607923878059, "300–1000 Hz_40ms_envelope_correlation": 0.9960506129356762, "300–1000 Hz_40ms_modulation_mae_db": 0.44242331060080725, "1000–3000 Hz_10ms_log_rms_mae_db": 1.406329870484718, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1386657179595667, "1000–3000 Hz_10ms_envelope_correlation": 0.9868942652502781, "1000–3000 Hz_10ms_modulation_mae_db": 1.1557228935248285, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8973351038326166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0759716383817587, "1000–3000 Hz_40ms_envelope_correlation": 0.9961089571496039, "1000–3000 Hz_40ms_modulation_mae_db": 0.7994550005133451, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3596527941923109, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1393583883850101, "3000–6000 Hz_10ms_envelope_correlation": 0.986333817288307, "3000–6000 Hz_10ms_modulation_mae_db": 0.3808387343302919, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8969623909532127, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07872603041165396, "3000–6000 Hz_40ms_envelope_correlation": 0.9956450730588221, "3000–6000 Hz_40ms_modulation_mae_db": 0.43887854377068763, "mr_stft_db": 5.806532757353817, "band_mae_db": 0.19528417891881733, "rms_gain_error_db": -0.15564635323965176, "output_peak": 0.6194824151865354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.687828352929649, "spectral_convergence": 0.31183712266295616, "spectral_convergence_status": "ok", "amplitude_floor": 1.4064891899964367e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5529518432316625, "spectral_convergence": 0.2985976054666193, "spectral_convergence_status": "ok", "amplitude_floor": 1.2304331075199438e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.13634218563331, "spectral_convergence": 0.31504471109943644, "spectral_convergence_status": "ok", "amplitude_floor": 7.896502575224385e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.444718915891811e-05, "output_power": 5.643346719448993e-05, "delta_db": 0.15561269366029468, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025153246625132694, "output_power": 0.00026539248778716047, "delta_db": 0.23294576492204536, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010938925867664089, "output_power": 0.0010557373097015575, "delta_db": -0.1541880938888342, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031036369064007244, "output_power": 0.00030721228196229836, "delta_db": -0.04432333313910888, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001328835538454751, "output_power": 0.00012717421065177616, "delta_db": -0.19072183674094872, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6997042780401114e-05, "output_power": 3.589711948252351e-05, "delta_db": -0.13107411173827696, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15147763901491815, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07518787391861183, "pitch": null, "output_rms_dbfs": -27.333340197635394, "output_peak": 0.6107767930922043, "reference_rms_dbfs": -27.258152323716782, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.792374127264874}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019413100530357063, "log_rms_mae_db": 0.8624244289326648, "rms_envelope_nrmse": 0.08688865107106608, "envelope_correlation": 0.9934820194941633, "rms_gain_error_db": -0.07515165742784961, "reference_mean_power": 0.0018816420370153617, "output_mean_power": 0.0018493616231313936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1507655155627412e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005726194534917747, "output_power": 0.0005640123297736066, "reference_active": true, "delta_db": -0.0657750001819647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.421513884012437e-05, "output_power": 8.450327290377013e-05, "reference_active": true, "delta_db": -0.47247162484615945}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.679657413963232e-05, "output_power": 1.6136231844159857e-05, "reference_active": true, "delta_db": -0.17418586091404092}], "active_band_count": 3, "active_band_mae_db": 0.23747749531405504}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016416501002274435, "log_rms_mae_db": 0.45764618784768724, "rms_envelope_nrmse": 0.041361669874356397, "envelope_correlation": 0.9984954672060866, "rms_gain_error_db": -0.07514732846360987, "reference_mean_power": 0.001890759403997248, "output_mean_power": 0.0018583244299308678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0781635397104837e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005383866955860211, "output_power": 0.0005290408194906118, "reference_active": true, "delta_db": -0.07605136799200245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8874025426333776e-05, "output_power": 3.595909941632551e-05, "reference_active": true, "delta_db": -0.33850707951390613}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650199092025111e-06, "output_power": 1.3643715125438725e-06, "reference_active": true, "delta_db": -0.5958722378091128}], "active_band_count": 3, "active_band_mae_db": 0.33681022843834046}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017990629858894173, "log_rms_mae_db": 1.4027645346533963, "rms_envelope_nrmse": 0.11073322688694137, "envelope_correlation": 0.9904122353648455, "rms_gain_error_db": -0.08985999455937956, "reference_mean_power": 0.0012781810161111924, "output_mean_power": 0.0012520058656389046}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.355790711503392e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004472139003696596, "output_power": 0.0004368109213105092, "reference_active": true, "delta_db": -0.10221805668537351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.621773891011176e-05, "output_power": 6.94878561630302e-05, "reference_active": true, "delta_db": -0.40147147717254755}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7010237353267386e-05, "output_power": 1.8572676746731853e-05, "reference_active": true, "delta_db": 0.3816412639401001}], "active_band_count": 3, "active_band_mae_db": 0.2951102659326737}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000149706333097475, "log_rms_mae_db": 0.7626331368518523, "rms_envelope_nrmse": 0.052724703370312605, "envelope_correlation": 0.997779473834152, "rms_gain_error_db": -0.08986363662006876, "reference_mean_power": 0.0012844845607787134, "output_mean_power": 0.0012581792684298165}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.840745806614025e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004236215690036061, "output_power": 0.0004137198952235906, "reference_active": true, "delta_db": -0.1027165839622126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1600142098026355e-05, "output_power": 3.0046183932923878e-05, "reference_active": true, "delta_db": -0.2189971401136777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5445822829049627e-06, "output_power": 1.4979738131398954e-06, "reference_active": true, "delta_db": -0.13306827664334694}], "active_band_count": 3, "active_band_mae_db": 0.15159400023974576}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010005798255669499, "log_rms_mae_db": 1.388135977831511, "rms_envelope_nrmse": 0.13443966812591812, "envelope_correlation": 0.9876358592114088, "rms_gain_error_db": -0.0894978837286299, "reference_mean_power": 0.0003605696849433608, "output_mean_power": 0.0003532152306076366}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5037729366163507e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001406674030835393, "output_power": 0.0001361081210834738, "reference_active": true, "delta_db": -0.14309431079944895}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0544824834133773e-05, "output_power": 2.445063153409839e-05, "reference_active": true, "delta_db": 0.755876381698954}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4458512826457347e-06, "output_power": 6.0299733502065684e-06, "reference_active": true, "delta_db": 2.4301886268564408}], "active_band_count": 3, "active_band_mae_db": 1.1097197731182813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.18211380942645e-05, "log_rms_mae_db": 0.8833304586620812, "rms_envelope_nrmse": 0.07628266966582166, "envelope_correlation": 0.9959160985985258, "rms_gain_error_db": -0.08949758204249654, "reference_mean_power": 0.00036238111197119544, "output_mean_power": 0.0003549897350477045}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3815388264892723e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001355192968806531, "output_power": 0.00013365715980226145, "reference_active": true, "delta_db": -0.06008911544818643}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.971075792205542e-06, "output_power": 1.0254646470841986e-05, "reference_active": true, "delta_db": 1.0940375405406857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.416322058192608e-07, "output_power": 3.5705644269281287e-07, "reference_active": true, "delta_db": 1.6958205533556072}], "active_band_count": 3, "active_band_mae_db": 0.9499824031148264}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.283283642007837e-05, "log_rms_mae_db": 1.2981729090565974, "rms_envelope_nrmse": 0.12838436435178271, "envelope_correlation": 0.9889524859720563, "rms_gain_error_db": -0.31730088492246006, "reference_mean_power": 6.230386432542126e-05, "output_mean_power": 5.791417786877317e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0221136919924677e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.16042156208526e-05, "output_power": 1.1143939977919912e-05, "reference_active": true, "delta_db": -0.1757702566053657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3772446197384082e-06, "output_power": 2.2976253388318575e-06, "reference_active": true, "delta_db": -0.1479466105598337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4509999616936e-07, "output_power": 4.022871296846829e-07, "reference_active": true, "delta_db": -0.4392145207337692}], "active_band_count": 3, "active_band_mae_db": 0.25431046263298956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.758960619357764e-05, "log_rms_mae_db": 0.8336522880780662, "rms_envelope_nrmse": 0.07787427714491182, "envelope_correlation": 0.9964443790353688, "rms_gain_error_db": -0.3173006881587754, "reference_mean_power": 6.26170059672204e-05, "output_mean_power": 5.82052594130359e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.846748801507289e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1232117533806203e-05, "output_power": 1.0751748365704165e-05, "reference_active": true, "delta_db": -0.18982547717455772}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.067581168241283e-06, "output_power": 1.0833950927936346e-06, "reference_active": true, "delta_db": 0.06385959751343787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.579800269976513e-08, "output_power": 2.2596818454796665e-08, "reference_active": true, "delta_db": -0.5753878741086871}], "active_band_count": 3, "active_band_mae_db": 0.27635764959889425}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.845056441002114, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b80bb3186581a8c80927ee5617bf000a489f3ad8b747b0697cdf2bd3d18f9b82", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9855378903068102, "standard_deviation": 0.0014115713900052299, "minimum": 0.9832502634259068, "maximum": 0.9876358592114088, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4260201643743176, "standard_deviation": 0.03087122558577181, "minimum": 1.388135977831511, "maximum": 1.4632554319762099, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9810873421389756, "standard_deviation": 0.2148193416044752, "minimum": 0.6649646030288742, "maximum": 1.263394551979552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14703592267987492, "standard_deviation": 0.008396911052843688, "minimum": 0.13443966812591812, "maximum": 0.16114293531212753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951334278862612, "standard_deviation": 0.0010778277778741919, "minimum": 0.9927523120930842, "maximum": 0.9961089571496039, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9079008275804188, "standard_deviation": 0.021407655484910124, "minimum": 0.8833304586620812, "maximum": 0.9346353664159717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7351213712259862, "standard_deviation": 0.29753590041522754, "minimum": 0.25577627435942923, "maximum": 1.1097553718184698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08787583877364942, "standard_deviation": 0.011651543600133037, "minimum": 0.0759716383817587, "maximum": 0.11150690415240164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9868060102624148, "standard_deviation": 0.0015036147838818391, "minimum": 0.9850197218298012, "maximum": 0.9889727451320856, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3286883165444612, "standard_deviation": 0.029181260540145963, "minimum": 1.278064730370758, "maximum": 1.366684572240933, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5563279385609496, "standard_deviation": 0.3496695752023985, "minimum": 0.25431046263298956, "maximum": 1.3120830929820724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1384019291207922, "standard_deviation": 0.007288089354000359, "minimum": 0.12681923522133706, "maximum": 0.1466559942901341, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954175704050596, "standard_deviation": 0.0009684550267406288, "minimum": 0.9933486400715801, "maximum": 0.9964443790353688, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8611843736298362, "standard_deviation": 0.029445697386324774, "minimum": 0.8336522880780662, "maximum": 0.9019847088453775, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5931838512680476, "standard_deviation": 0.35953776073872956, "minimum": 0.27635764959889425, "maximum": 1.3333947828428485, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08387335548132377, "standard_deviation": 0.007192615904548577, "minimum": 0.07566920993843067, "maximum": 0.09503557772009484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9891071484517897, "standard_deviation": 0.0018364711844463205, "minimum": 0.9864045115174698, "maximum": 0.9908426845809366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4089162953838472, "standard_deviation": 0.03051675212819478, "minimum": 1.3596840897610343, "maximum": 1.4506684331772084, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4066104141478371, "standard_deviation": 0.15582330888139534, "minimum": 0.23008520839967173, "maximum": 0.6647599793151749, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11824452512449664, "standard_deviation": 0.009504046839525807, "minimum": 0.1092469751215189, "maximum": 0.1323321391139146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99686484738386, "standard_deviation": 0.0009119969830776826, "minimum": 0.9952015102304715, "maximum": 0.997779473834152, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7778010819382146, "standard_deviation": 0.02556068177413174, "minimum": 0.7408569340764554, "maximum": 0.8077549513624551, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.33012837725027583, "standard_deviation": 0.2870223193066693, "minimum": 0.06622492009001622, "maximum": 0.9053494406012477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06346669895558098, "standard_deviation": 0.008353713581395375, "minimum": 0.052724703370312605, "maximum": 0.07750028171729492, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19699156713465635, "standard_deviation": 0.041785578381745674, "minimum": 0.147594489392847, "maximum": 0.2687166391604486, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.79909013992107, "standard_deviation": 0.004848694538469044, "minimum": 5.792374127264874, "maximum": 5.806532757353817, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6133836665402341, "standard_deviation": 0.02845116207602932, "minimum": 0.5899076915202265, "maximum": 0.6799491838763909, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14686999307235127, "standard_deviation": 0.05022253612963991, "minimum": -0.2430524309762587, "maximum": -0.07518787391861183, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924020308187176, "standard_deviation": 0.001505908363368869, "minimum": 0.989432952301244, "maximum": 0.9936805427989378, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8855117898508298, "standard_deviation": 0.02568698695985871, "minimum": 0.8436690174179486, "maximum": 0.9155926806252873, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4577986943962442, "standard_deviation": 0.13345247016123815, "minimum": 0.23747749531405504, "maximum": 0.6242539686778156, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09425766956216328, "standard_deviation": 0.008656543582814244, "minimum": 0.08623739307664642, "maximum": 0.1109140892942601, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979445129012282, "standard_deviation": 0.0006940139587728268, "minimum": 0.9964564921380631, "maximum": 0.9985336472420343, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4850526378849385, "standard_deviation": 0.02554024466962595, "minimum": 0.45764618784768724, "maximum": 0.5291288605372838, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4195603500350147, "standard_deviation": 0.20497501801111187, "minimum": 0.1389376914240324, "maximum": 0.8314717528388789, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049837693322790894, "standard_deviation": 0.00719684894687073, "minimum": 0.041361669874356397, "maximum": 0.06423457831759724, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-961-0011", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.260782480239868, "predicted_mos": 2.260782480239868, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f418ff2bedd4dfe7cb4c3bd970a5e99527fffa58151dc0b9d9d88c3cc1b6fb04", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.116027799900621, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8863740319756273, "全帯域_10ms_rms_envelope_nrmse": 0.09590894343299353, "全帯域_10ms_envelope_correlation": 0.9920361609830213, "全帯域_10ms_modulation_mae_db": 0.18645305562861844, "全帯域_40ms_log_rms_mae_db": 0.5022612154215996, "全帯域_40ms_rms_envelope_nrmse": 0.04800102748034081, "全帯域_40ms_envelope_correlation": 0.9979354705463456, "全帯域_40ms_modulation_mae_db": 0.5132992323097768, "300–1000 Hz_10ms_log_rms_mae_db": 1.612947316586755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12387489881509751, "300–1000 Hz_10ms_envelope_correlation": 0.9879989035872501, "300–1000 Hz_10ms_modulation_mae_db": 0.13968419454876474, "300–1000 Hz_40ms_log_rms_mae_db": 0.8994229022355342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060755075038358225, "300–1000 Hz_40ms_envelope_correlation": 0.9969596201684904, "300–1000 Hz_40ms_modulation_mae_db": 0.6726314392487461, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5130815428149187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13045170978460824, "1000–3000 Hz_10ms_envelope_correlation": 0.9891408784625498, "1000–3000 Hz_10ms_modulation_mae_db": 0.7479618169071026, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9804994196208562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08435656514222166, "1000–3000 Hz_40ms_envelope_correlation": 0.9960890050302934, "1000–3000 Hz_40ms_modulation_mae_db": 0.768335340993504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6641543791445492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15766685236272787, "3000–6000 Hz_10ms_envelope_correlation": 0.9824571242613956, "3000–6000 Hz_10ms_modulation_mae_db": 0.6868681420429826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1105600374799731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10041892245749545, "3000–6000 Hz_40ms_envelope_correlation": 0.9927279318268011, "3000–6000 Hz_40ms_modulation_mae_db": 0.7587517970535641, "mr_stft_db": 6.064220403663168, "band_mae_db": 0.14590249580528367, "rms_gain_error_db": -0.06666637583417066, "output_peak": 0.6330495835380453}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8874342443598313, "全帯域_10ms_rms_envelope_nrmse": 0.09471549655629327, "全帯域_10ms_envelope_correlation": 0.9924362153980861, "全帯域_10ms_modulation_mae_db": 0.431535422453303, "全帯域_40ms_log_rms_mae_db": 0.47732213473014523, "全帯域_40ms_rms_envelope_nrmse": 0.04614673729185903, "全帯域_40ms_envelope_correlation": 0.9984022859393442, "全帯域_40ms_modulation_mae_db": 0.4561869401485565, "300–1000 Hz_10ms_log_rms_mae_db": 1.6153790918661899, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11608153966407013, "300–1000 Hz_10ms_envelope_correlation": 0.9895906727119576, "300–1000 Hz_10ms_modulation_mae_db": 0.2506885676610849, "300–1000 Hz_40ms_log_rms_mae_db": 0.9237415456279838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06065088161171827, "300–1000 Hz_40ms_envelope_correlation": 0.9972625176916202, "300–1000 Hz_40ms_modulation_mae_db": 0.7461564788748296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5538247283727038, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15352057631259797, "1000–3000 Hz_10ms_envelope_correlation": 0.9842789622417915, "1000–3000 Hz_10ms_modulation_mae_db": 1.1639703705883082, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0226445485902418, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08456670266120764, "1000–3000 Hz_40ms_envelope_correlation": 0.9958591738907184, "1000–3000 Hz_40ms_modulation_mae_db": 1.1149057312787114, "3000–6000 Hz_10ms_log_rms_mae_db": 1.643130439954983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15728201539154102, "3000–6000 Hz_10ms_envelope_correlation": 0.9828190327704468, "3000–6000 Hz_10ms_modulation_mae_db": 0.3368891620755958, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0859570317580454, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09957376950537329, "3000–6000 Hz_40ms_envelope_correlation": 0.9933363960400782, "3000–6000 Hz_40ms_modulation_mae_db": 0.3920223707112209, "mr_stft_db": 6.061617405633274, "band_mae_db": 0.17907418415427676, "rms_gain_error_db": -0.17559103613548382, "output_peak": 0.5903848618173276}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8641729017487051, "全帯域_10ms_rms_envelope_nrmse": 0.09356884544526095, "全帯域_10ms_envelope_correlation": 0.9926315420290277, "全帯域_10ms_modulation_mae_db": 0.41449784522156086, "全帯域_40ms_log_rms_mae_db": 0.46505449250493086, "全帯域_40ms_rms_envelope_nrmse": 0.05104798924367173, "全帯域_40ms_envelope_correlation": 0.9979582500420459, "全帯域_40ms_modulation_mae_db": 0.43925280470429073, "300–1000 Hz_10ms_log_rms_mae_db": 1.6175001959053599, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11029412672006585, "300–1000 Hz_10ms_envelope_correlation": 0.9906633869832574, "300–1000 Hz_10ms_modulation_mae_db": 0.5407516064962937, "300–1000 Hz_40ms_log_rms_mae_db": 0.893909579516342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05765300542809429, "300–1000 Hz_40ms_envelope_correlation": 0.9976095968140013, "300–1000 Hz_40ms_modulation_mae_db": 0.5119486745565397, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5969556502238624, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16158555526479562, "1000–3000 Hz_10ms_envelope_correlation": 0.9823479003429659, "1000–3000 Hz_10ms_modulation_mae_db": 1.2034316289355942, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9963323151836506, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09675773181789184, "1000–3000 Hz_40ms_envelope_correlation": 0.9939790293304394, "1000–3000 Hz_40ms_modulation_mae_db": 1.0373901966702295, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7134234488354527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15330102945343535, "3000–6000 Hz_10ms_envelope_correlation": 0.9834037594488392, "3000–6000 Hz_10ms_modulation_mae_db": 0.37117057655033725, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1315681293814592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08879804448373685, "3000–6000 Hz_40ms_envelope_correlation": 0.9943520519519764, "3000–6000 Hz_40ms_modulation_mae_db": 0.2689099397148047, "mr_stft_db": 6.057431352791699, "band_mae_db": 0.2599267833701338, "rms_gain_error_db": -0.17714192528794745, "output_peak": 0.5987156895083235}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8102000507054381, "全帯域_10ms_rms_envelope_nrmse": 0.09826391898952914, "全帯域_10ms_envelope_correlation": 0.9916513348594158, "全帯域_10ms_modulation_mae_db": 0.16961789394384774, "全帯域_40ms_log_rms_mae_db": 0.43273032110190096, "全帯域_40ms_rms_envelope_nrmse": 0.05184769133290448, "全帯域_40ms_envelope_correlation": 0.9976158580823297, "全帯域_40ms_modulation_mae_db": 0.5499181932809595, "300–1000 Hz_10ms_log_rms_mae_db": 1.5139367923135723, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11840052662635818, "300–1000 Hz_10ms_envelope_correlation": 0.9891294811487857, "300–1000 Hz_10ms_modulation_mae_db": 0.3528614327836383, "300–1000 Hz_40ms_log_rms_mae_db": 0.858525940020267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062281317131481016, "300–1000 Hz_40ms_envelope_correlation": 0.9970772796007357, "300–1000 Hz_40ms_modulation_mae_db": 0.6639462320100319, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5912906433758456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14151140176210195, "1000–3000 Hz_10ms_envelope_correlation": 0.9864683765265929, "1000–3000 Hz_10ms_modulation_mae_db": 0.9826210416059902, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9990784486187935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08027768308835585, "1000–3000 Hz_40ms_envelope_correlation": 0.9954862141469158, "1000–3000 Hz_40ms_modulation_mae_db": 0.7126978630872601, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6718781608036164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15838742831407626, "3000–6000 Hz_10ms_envelope_correlation": 0.9823350154383839, "3000–6000 Hz_10ms_modulation_mae_db": 0.36406035775497786, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0633577772400398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09960407613598223, "3000–6000 Hz_40ms_envelope_correlation": 0.9927548570731366, "3000–6000 Hz_40ms_modulation_mae_db": 0.1656670482942568, "mr_stft_db": 6.0593265931244185, "band_mae_db": 0.16625392067094655, "rms_gain_error_db": -0.0869419771662974, "output_peak": 0.6239381390890191}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8438250486490763, "全帯域_10ms_rms_envelope_nrmse": 0.0993457465614283, "全帯域_10ms_envelope_correlation": 0.9914753012398303, "全帯域_10ms_modulation_mae_db": 0.5891803796136948, "全帯域_40ms_log_rms_mae_db": 0.45479291058422444, "全帯域_40ms_rms_envelope_nrmse": 0.05294673847875906, "全帯域_40ms_envelope_correlation": 0.9974444342543451, "全帯域_40ms_modulation_mae_db": 0.28232601595577683, "300–1000 Hz_10ms_log_rms_mae_db": 1.5652176843074939, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11895681887687855, "300–1000 Hz_10ms_envelope_correlation": 0.9890070432238616, "300–1000 Hz_10ms_modulation_mae_db": 0.2834249924584609, "300–1000 Hz_40ms_log_rms_mae_db": 0.8432108843248081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06582552426530955, "300–1000 Hz_40ms_envelope_correlation": 0.9964593972736695, "300–1000 Hz_40ms_modulation_mae_db": 0.17298887571456822, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5225879990224958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1437821316245993, "1000–3000 Hz_10ms_envelope_correlation": 0.985861748994725, "1000–3000 Hz_10ms_modulation_mae_db": 1.4738279854609064, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0031901011850664, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08226017724603686, "1000–3000 Hz_40ms_envelope_correlation": 0.995378823906205, "1000–3000 Hz_40ms_modulation_mae_db": 0.8578944894592375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.675088334620918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15447051875133944, "3000–6000 Hz_10ms_envelope_correlation": 0.9832350794405442, "3000–6000 Hz_10ms_modulation_mae_db": 0.768718336510406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1118604675121997, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09446270382370169, "3000–6000 Hz_40ms_envelope_correlation": 0.9937367659978226, "3000–6000 Hz_40ms_modulation_mae_db": 0.5983021876302578, "mr_stft_db": 6.063946832410788, "band_mae_db": 0.15045624658285056, "rms_gain_error_db": -0.00046683782961941646, "output_peak": 0.6357987408241681}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8802853691573863, "全帯域_10ms_rms_envelope_nrmse": 0.08642608649548929, "全帯域_10ms_envelope_correlation": 0.9937611520122868, "全帯域_10ms_modulation_mae_db": 0.536368429716709, "全帯域_40ms_log_rms_mae_db": 0.501691316515392, "全帯域_40ms_rms_envelope_nrmse": 0.044157415688984514, "全帯域_40ms_envelope_correlation": 0.9985606625585196, "全帯域_40ms_modulation_mae_db": 0.7668820875764091, "300–1000 Hz_10ms_log_rms_mae_db": 1.6257078429423983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10969737149592053, "300–1000 Hz_10ms_envelope_correlation": 0.9906434927225026, "300–1000 Hz_10ms_modulation_mae_db": 0.448171497325413, "300–1000 Hz_40ms_log_rms_mae_db": 0.9650169473406863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05640532299179482, "300–1000 Hz_40ms_envelope_correlation": 0.9975337417340081, "300–1000 Hz_40ms_modulation_mae_db": 0.5594887127408429, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5370555125725456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13623235991338037, "1000–3000 Hz_10ms_envelope_correlation": 0.9885868621904008, "1000–3000 Hz_10ms_modulation_mae_db": 0.8766803178473562, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9664862782254039, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08345749801509723, "1000–3000 Hz_40ms_envelope_correlation": 0.9968720530172555, "1000–3000 Hz_40ms_modulation_mae_db": 0.5077919959005741, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5650035724403324, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1529666035322233, "3000–6000 Hz_10ms_envelope_correlation": 0.9844261980783332, "3000–6000 Hz_10ms_modulation_mae_db": 0.38827515668170914, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0334532624375572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09612694694945333, "3000–6000 Hz_40ms_envelope_correlation": 0.9947309255365596, "3000–6000 Hz_40ms_modulation_mae_db": 0.4644301652421272, "mr_stft_db": 6.05495891252608, "band_mae_db": 0.24614468763703112, "rms_gain_error_db": -0.17472453651141207, "output_peak": 0.608320288500637}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.864875736077281, "全帯域_10ms_rms_envelope_nrmse": 0.0926944884697515, "全帯域_10ms_envelope_correlation": 0.9928218337947743, "全帯域_10ms_modulation_mae_db": 0.5789663947773064, "全帯域_40ms_log_rms_mae_db": 0.46660270040071117, "全帯域_40ms_rms_envelope_nrmse": 0.05186590825286082, "全帯域_40ms_envelope_correlation": 0.9979340529010458, "全帯域_40ms_modulation_mae_db": 0.41421906225518507, "300–1000 Hz_10ms_log_rms_mae_db": 1.581851138917131, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10895447536697415, "300–1000 Hz_10ms_envelope_correlation": 0.9909240262088773, "300–1000 Hz_10ms_modulation_mae_db": 0.4172509443783117, "300–1000 Hz_40ms_log_rms_mae_db": 0.9240052808785887, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061068669583438484, "300–1000 Hz_40ms_envelope_correlation": 0.9972934032911587, "300–1000 Hz_40ms_modulation_mae_db": 0.3189434948917737, "1000–3000 Hz_10ms_log_rms_mae_db": 1.499714955310543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1455199617547287, "1000–3000 Hz_10ms_envelope_correlation": 0.9860748462305263, "1000–3000 Hz_10ms_modulation_mae_db": 1.4156541336491137, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9674348201936825, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09333942478876249, "1000–3000 Hz_40ms_envelope_correlation": 0.9947689369374942, "1000–3000 Hz_40ms_modulation_mae_db": 1.2864752951538472, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6320463182617466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14777045003694636, "3000–6000 Hz_10ms_envelope_correlation": 0.98464896844431, "3000–6000 Hz_10ms_modulation_mae_db": 0.34285217474635665, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1202942926660981, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09598236493466643, "3000–6000 Hz_40ms_envelope_correlation": 0.9934240520556553, "3000–6000 Hz_40ms_modulation_mae_db": 0.43937321768651255, "mr_stft_db": 6.050608413883006, "band_mae_db": 0.19246195288536774, "rms_gain_error_db": -0.19102069571606167, "output_peak": 0.6239365458407565}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8390642216730205, "全帯域_10ms_rms_envelope_nrmse": 0.09714538756711527, "全帯域_10ms_envelope_correlation": 0.9918257390030454, "全帯域_10ms_modulation_mae_db": 0.46669615703622297, "全帯域_40ms_log_rms_mae_db": 0.4655298236175835, "全帯域_40ms_rms_envelope_nrmse": 0.05589302925224366, "全帯域_40ms_envelope_correlation": 0.997163204233816, "全帯域_40ms_modulation_mae_db": 0.34826186247328356, "300–1000 Hz_10ms_log_rms_mae_db": 1.6035656612007885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12036118754213025, "300–1000 Hz_10ms_envelope_correlation": 0.9886539221456507, "300–1000 Hz_10ms_modulation_mae_db": 0.423934429397451, "300–1000 Hz_40ms_log_rms_mae_db": 0.9432831014621263, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07102719252662679, "300–1000 Hz_40ms_envelope_correlation": 0.9959005633994367, "300–1000 Hz_40ms_modulation_mae_db": 0.29184361477312065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5137148316772522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1459285500881606, "1000–3000 Hz_10ms_envelope_correlation": 0.985405522993052, "1000–3000 Hz_10ms_modulation_mae_db": 0.7486180557941476, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9949834041753154, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08493640779423156, "1000–3000 Hz_40ms_envelope_correlation": 0.9949126163068925, "1000–3000 Hz_40ms_modulation_mae_db": 0.15300665729522042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5919360494167503, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14394709912809325, "3000–6000 Hz_10ms_envelope_correlation": 0.9854376726503948, "3000–6000 Hz_10ms_modulation_mae_db": 0.05466373346289056, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0022635992611304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0822477992630973, "3000–6000 Hz_40ms_envelope_correlation": 0.9952775756217267, "3000–6000 Hz_40ms_modulation_mae_db": 0.30569399013643284, "mr_stft_db": 6.078014746718172, "band_mae_db": 0.1387038038956217, "rms_gain_error_db": -0.04838270039845499, "output_peak": 0.6067559351934477}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.977307185369825, "spectral_convergence": 0.31894298030855556, "spectral_convergence_status": "ok", "amplitude_floor": 1.4064891899964367e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.879383929965898, "spectral_convergence": 0.30279231094686915, "spectral_convergence_status": "ok", "amplitude_floor": 1.2304331075199438e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.335970095653785, "spectral_convergence": 0.323708165633513, "spectral_convergence_status": "ok", "amplitude_floor": 7.896502575224385e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.444718915891811e-05, "output_power": 5.6306086292642614e-05, "delta_db": 0.14579877394854115, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025153246625132694, "output_power": 0.0002607408984834725, "delta_db": 0.15615108673000583, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010938925867664089, "output_power": 0.001083410058708426, "delta_db": -0.04181815767389431, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031036369064007244, "output_power": 0.00028347711843224293, "delta_db": -0.3935289809645898, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001328835538454751, "output_power": 0.00013013844184695688, "delta_db": -0.09065631792630093, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6997042780401114e-05, "output_power": 3.740358049958642e-05, "delta_db": 0.047461657588369996, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14590249580528367, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06666637583417066, "pitch": null, "output_rms_dbfs": -27.32481869955095, "output_peak": 0.6330495835380453, "reference_rms_dbfs": -27.258152323716782, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.064220403663168}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019413100530357063, "log_rms_mae_db": 0.8863740319756273, "rms_envelope_nrmse": 0.09590894343299353, "envelope_correlation": 0.9920361609830213, "rms_gain_error_db": -0.06663435041818902, "reference_mean_power": 0.0018816420370153617, "output_mean_power": 0.0018529921174933964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1507655155627412e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005726194534917747, "output_power": 0.0005617498211620407, "reference_active": true, "delta_db": -0.08323155310283298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.421513884012437e-05, "output_power": 8.653654175644144e-05, "reference_active": true, "delta_db": -0.3692115687751317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.679657413963232e-05, "output_power": 1.721520964051774e-05, "reference_active": true, "delta_db": 0.10691604500789065}], "active_band_count": 3, "active_band_mae_db": 0.18645305562861844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016416501002274435, "log_rms_mae_db": 0.5022612154215996, "rms_envelope_nrmse": 0.04800102748034081, "envelope_correlation": 0.9979354705463456, "rms_gain_error_db": -0.06662805797950534, "reference_mean_power": 0.001890759403997248, "output_mean_power": 0.001861973361050207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0781635397104837e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005383866955860211, "output_power": 0.0005309262101579485, "reference_active": true, "delta_db": -0.060601535993563514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8874025426333776e-05, "output_power": 3.4773793210867006e-05, "reference_active": true, "delta_db": -0.4840744721082569}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650199092025111e-06, "output_power": 1.2445080147663253e-06, "reference_active": true, "delta_db": -0.9952216888275102}], "active_band_count": 3, "active_band_mae_db": 0.5132992323097768}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017990629858894173, "log_rms_mae_db": 1.612947316586755, "rms_envelope_nrmse": 0.12387489881509751, "envelope_correlation": 0.9879989035872501, "rms_gain_error_db": -0.017664863260300655, "reference_mean_power": 0.0012781810161111924, "output_mean_power": 0.0012729925929671252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.355790711503392e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004472139003696596, "output_power": 0.00043983511748490014, "reference_active": true, "delta_db": -0.07225392378054102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.621773891011176e-05, "output_power": 7.248873891475106e-05, "reference_active": true, "delta_db": -0.21785516359946067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7010237353267386e-05, "output_power": 1.6512621416473512e-05, "reference_active": true, "delta_db": -0.12894349626629253}], "active_band_count": 3, "active_band_mae_db": 0.13968419454876474}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000149706333097475, "log_rms_mae_db": 0.8994229022355342, "rms_envelope_nrmse": 0.060755075038358225, "envelope_correlation": 0.9969596201684904, "rms_gain_error_db": -0.017641997898335024, "reference_mean_power": 0.0012844845607787134, "output_mean_power": 0.0012792772854401605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.840745806614025e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004236215690036061, "output_power": 0.0004199399563641689, "reference_active": true, "delta_db": -0.037908656515584935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1600142098026355e-05, "output_power": 2.8768929461146276e-05, "reference_active": true, "delta_db": -0.40765334153711}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5445822829049627e-06, "output_power": 1.0754192770326428e-06, "reference_active": true, "delta_db": -1.5723323196935435}], "active_band_count": 3, "active_band_mae_db": 0.6726314392487461}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010005798255669499, "log_rms_mae_db": 1.5130815428149187, "rms_envelope_nrmse": 0.13045170978460824, "envelope_correlation": 0.9891408784625498, "rms_gain_error_db": -0.361632981841283, "reference_mean_power": 0.0003605696849433608, "output_mean_power": 0.0003317614436926681}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5037729366163507e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001406674030835393, "output_power": 0.00012882634922497233, "reference_active": true, "delta_db": -0.38188770237869607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0544824834133773e-05, "output_power": 1.7484818823559738e-05, "reference_active": true, "delta_db": -0.7004130626407898}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4458512826457347e-06, "output_power": 4.502513240724623e-06, "reference_active": true, "delta_db": 1.1615846857018222}], "active_band_count": 3, "active_band_mae_db": 0.7479618169071026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.18211380942645e-05, "log_rms_mae_db": 0.9804994196208562, "rms_envelope_nrmse": 0.08435656514222166, "envelope_correlation": 0.9960890050302934, "rms_gain_error_db": -0.36162696304129943, "reference_mean_power": 0.00036238111197119544, "output_mean_power": 0.0003334286062059194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3815388264892723e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001355192968806531, "output_power": 0.0001235830353363089, "reference_active": true, "delta_db": -0.4004228196440456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.971075792205542e-06, "output_power": 7.277660202512361e-06, "reference_active": true, "delta_db": -0.3952516429337149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.416322058192608e-07, "output_power": 3.4204872392965566e-07, "reference_active": true, "delta_db": 1.5093315604027513}], "active_band_count": 3, "active_band_mae_db": 0.768335340993504}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.283283642007837e-05, "log_rms_mae_db": 1.6641543791445492, "rms_envelope_nrmse": 0.15766685236272787, "envelope_correlation": 0.9824571242613956, "rms_gain_error_db": -0.1187745551684191, "reference_mean_power": 6.230386432542126e-05, "output_mean_power": 6.062301458144588e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0221136919924677e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.16042156208526e-05, "output_power": 1.0311675780537383e-05, "reference_active": true, "delta_db": -0.5128654045740388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3772446197384082e-06, "output_power": 1.8003929897173236e-06, "reference_active": true, "delta_db": -1.2070655992626826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4509999616936e-07, "output_power": 4.1151933804239384e-07, "reference_active": true, "delta_db": -0.3406734222922261}], "active_band_count": 3, "active_band_mae_db": 0.6868681420429826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.758960619357764e-05, "log_rms_mae_db": 1.1105600374799731, "rms_envelope_nrmse": 0.10041892245749545, "envelope_correlation": 0.9927279318268011, "rms_gain_error_db": -0.11877332096002184, "reference_mean_power": 6.26170059672204e-05, "output_mean_power": 6.092772552201673e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.846748801507289e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1232117533806203e-05, "output_power": 9.769921411130936e-06, "reference_active": true, "delta_db": -0.6057056901801189}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.067581168241283e-06, "output_power": 8.402220284493988e-07, "reference_active": true, "delta_db": -1.0400684095036299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.579800269976513e-08, "output_power": 2.231197192150979e-08, "reference_active": true, "delta_db": -0.6304812914769438}], "active_band_count": 3, "active_band_mae_db": 0.7587517970535641}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8255081364062556, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bfe3e7f05ff4b4005d0a4f14fcce12ba768697905fad8c4415d27c09d868c6bf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9860206372478255, "standard_deviation": 0.002186022611040974, "minimum": 0.9823479003429659, "maximum": 0.9891408784625498, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.541028232921271, "standard_deviation": 0.03665980130538064, "minimum": 1.499714955310543, "maximum": 1.5969556502238624, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0765956688485647, "standard_deviation": 0.2830554606502543, "minimum": 0.7479618169071026, "maximum": 1.4738279854609064, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1448165308131216, "standard_deviation": 0.009641590888420839, "minimum": 0.13045170978460824, "maximum": 0.16158555526479562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9954182315707767, "standard_deviation": 0.0008874258086421144, "minimum": 0.9939790293304394, "maximum": 0.9968720530172555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9913311669741263, "standard_deviation": 0.018993094680653022, "minimum": 0.9664862782254039, "maximum": 1.0226445485902418, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8048121962298229, "standard_deviation": 0.36014970474894126, "minimum": 0.15300665729522042, "maximum": 1.2864752951538472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08624402381922565, "standard_deviation": 0.005709659963426558, "minimum": 0.08027768308835585, "maximum": 0.09675773181789184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.983595356316581, "standard_deviation": 0.0011242613066066586, "minimum": 0.9823350154383839, "maximum": 0.9854376726503948, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6445825879347935, "standard_deviation": 0.04790095814115907, "minimum": 1.5650035724403324, "maximum": 1.7134234488354527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.414187204978157, "standard_deviation": 0.22213335203044313, "minimum": 0.05466373346289056, "maximum": 0.768718336510406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15322399962129785, "standard_deviation": 0.005075086353270279, "minimum": 0.14394709912809325, "maximum": 0.15838742831407626, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937925695129695, "standard_deviation": 0.0009224261229592131, "minimum": 0.9927279318268011, "maximum": 0.9952775756217267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0824143247170628, "standard_deviation": 0.04583709063134285, "minimum": 1.0022635992611304, "maximum": 1.1315681293814592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4241438395586471, "standard_deviation": 0.18902702312077502, "minimum": 0.1656670482942568, "maximum": 0.7587517970535641, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09465182844418832, "standard_deviation": 0.006257777230079604, "minimum": 0.0822477992630973, "maximum": 0.10041892245749545, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9895763660915179, "standard_deviation": 0.0010690377249964866, "minimum": 0.9879989035872501, "maximum": 0.9909240262088773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.592013215504961, "standard_deviation": 0.03745429597133065, "minimum": 1.5139367923135723, "maximum": 1.6257078429423983, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3570959581311773, "standard_deviation": 0.1276971331517758, "minimum": 0.13968419454876474, "maximum": 0.5407516064962937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1158276181384369, "standard_deviation": 0.005570685416267409, "minimum": 0.10895447536697415, "maximum": 0.12387489881509751, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99701201499664, "standard_deviation": 0.0005755679774475713, "minimum": 0.9959005633994367, "maximum": 0.9976095968140013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9063895226757921, "standard_deviation": 0.04122493071331355, "minimum": 0.8432108843248081, "maximum": 0.9650169473406863, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4922434403513066, "standard_deviation": 0.20823070004380587, "minimum": 0.17298887571456822, "maximum": 0.7461564788748296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06195837357210268, "standard_deviation": 0.004638722509401178, "minimum": 0.05640532299179482, "maximum": 0.07102719252662679, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18486550937518897, "standard_deviation": 0.04576334469427242, "minimum": 0.1387038038956217, "maximum": 0.2599267833701338, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.061265582593826, "standard_deviation": 0.008180016390661606, "minimum": 6.050608413883006, "maximum": 6.078014746718172, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6151124730389657, "standard_deviation": 0.01648242792630628, "minimum": 0.5903848618173276, "maximum": 0.6357987408241681, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11511701060993093, "standard_deviation": 0.07324913569314327, "minimum": -0.19102069571606167, "maximum": -0.00046683782961941646, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.992329909914936, "standard_deviation": 0.0007487153445088642, "minimum": 0.9914753012398303, "maximum": 0.9937611520122868, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8595289505432957, "standard_deviation": 0.026936711947436687, "minimum": 0.8102000507054381, "maximum": 0.8874342443598313, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4216644472989079, "standard_deviation": 0.16344112478182268, "minimum": 0.16961789394384774, "maximum": 0.5891803796136948, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09475861418973265, "standard_deviation": 0.004061279376478614, "minimum": 0.08642608649548929, "maximum": 0.0993457465614283, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978767773197239, "standard_deviation": 0.00046620158794898715, "minimum": 0.997163204233816, "maximum": 0.9985606625585196, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.470748114359561, "standard_deviation": 0.023208768412920373, "minimum": 0.43273032110190096, "maximum": 0.5022612154215996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47129327483802974, "standard_deviation": 0.14671485300862158, "minimum": 0.28232601595577683, "maximum": 0.7668820875764091, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05023831712770301, "standard_deviation": 0.0038531967320545066, "minimum": 0.044157415688984514, "maximum": 0.05589302925224366, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-961-0011", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.3339931964874268, "predicted_mos": 2.3339931964874268, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ec297d8c411f690446545cc4ee561aa6a9d908646cd2205da324675493bb896e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.112905099987984, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8903895747715466, "全帯域_10ms_rms_envelope_nrmse": 0.08999461467282908, "全帯域_10ms_envelope_correlation": 0.9931050594810209, "全帯域_10ms_modulation_mae_db": 0.41598048984046215, "全帯域_40ms_log_rms_mae_db": 0.4895715505137562, "全帯域_40ms_rms_envelope_nrmse": 0.04896942417634289, "全帯域_40ms_envelope_correlation": 0.9980081203317364, "全帯域_40ms_modulation_mae_db": 0.280977040488451, "300–1000 Hz_10ms_log_rms_mae_db": 1.3275674970099591, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11439006112815228, "300–1000 Hz_10ms_envelope_correlation": 0.9898879478056111, "300–1000 Hz_10ms_modulation_mae_db": 0.5164738509368499, "300–1000 Hz_40ms_log_rms_mae_db": 0.7349625006031024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05935453107103731, "300–1000 Hz_40ms_envelope_correlation": 0.9973779114938774, "300–1000 Hz_40ms_modulation_mae_db": 0.28648007765056854, "1000–3000 Hz_10ms_log_rms_mae_db": 1.375754811752196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14410737272413415, "1000–3000 Hz_10ms_envelope_correlation": 0.9858189916700842, "1000–3000 Hz_10ms_modulation_mae_db": 0.6902266940590357, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9220058807176477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09205942811258988, "1000–3000 Hz_40ms_envelope_correlation": 0.9941012132988988, "1000–3000 Hz_40ms_modulation_mae_db": 0.9484854114083993, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2785156751100426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13793908273108824, "3000–6000 Hz_10ms_envelope_correlation": 0.9866274094619502, "3000–6000 Hz_10ms_modulation_mae_db": 0.5797812749269756, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8536836545081438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08663840517912287, "3000–6000 Hz_40ms_envelope_correlation": 0.9946865475040926, "3000–6000 Hz_40ms_modulation_mae_db": 0.6948008823628905, "mr_stft_db": 5.7753267457901565, "band_mae_db": 0.173285099535372, "rms_gain_error_db": -0.13809721022638727, "output_peak": 0.5879714031370464}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8625629585540628, "全帯域_10ms_rms_envelope_nrmse": 0.0917989408050779, "全帯域_10ms_envelope_correlation": 0.9927572931246954, "全帯域_10ms_modulation_mae_db": 0.5071231222488296, "全帯域_40ms_log_rms_mae_db": 0.4694641870800739, "全帯域_40ms_rms_envelope_nrmse": 0.047786126706391784, "全帯域_40ms_envelope_correlation": 0.9980470194647786, "全帯域_40ms_modulation_mae_db": 0.5561796203324433, "300–1000 Hz_10ms_log_rms_mae_db": 1.3153637871025514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11393378495175056, "300–1000 Hz_10ms_envelope_correlation": 0.9898549606205699, "300–1000 Hz_10ms_modulation_mae_db": 0.4943569324662336, "300–1000 Hz_40ms_log_rms_mae_db": 0.758294313947806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05975138029360782, "300–1000 Hz_40ms_envelope_correlation": 0.9971720280702672, "300–1000 Hz_40ms_modulation_mae_db": 0.7105833827417358, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4430520007627983, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14153875195921006, "1000–3000 Hz_10ms_envelope_correlation": 0.9863536627298977, "1000–3000 Hz_10ms_modulation_mae_db": 0.6442687939262073, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9656137045209364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0844811747818893, "1000–3000 Hz_40ms_envelope_correlation": 0.995181193610077, "1000–3000 Hz_40ms_modulation_mae_db": 0.6027542873222274, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2387489420625728, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12720286587355123, "3000–6000 Hz_10ms_envelope_correlation": 0.988602977144892, "3000–6000 Hz_10ms_modulation_mae_db": 0.3181746091294156, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8335456917073689, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07588170827464938, "3000–6000 Hz_40ms_envelope_correlation": 0.9958807970616795, "3000–6000 Hz_40ms_modulation_mae_db": 0.35237317794562273, "mr_stft_db": 5.776943561085436, "band_mae_db": 0.18486689656351665, "rms_gain_error_db": -0.10664122166314753, "output_peak": 0.6111112289952493}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8864792774677054, "全帯域_10ms_rms_envelope_nrmse": 0.0921047982974151, "全帯域_10ms_envelope_correlation": 0.9926988417004948, "全帯域_10ms_modulation_mae_db": 0.0938331107874301, "全帯域_40ms_log_rms_mae_db": 0.48663460559074795, "全帯域_40ms_rms_envelope_nrmse": 0.04966986999868593, "全帯域_40ms_envelope_correlation": 0.9978314769665176, "全帯域_40ms_modulation_mae_db": 0.20752602802263695, "300–1000 Hz_10ms_log_rms_mae_db": 1.371923383842391, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1193862985706654, "300–1000 Hz_10ms_envelope_correlation": 0.9888946133425653, "300–1000 Hz_10ms_modulation_mae_db": 0.0958080943073123, "300–1000 Hz_40ms_log_rms_mae_db": 0.7941560944252422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06333031503815778, "300–1000 Hz_40ms_envelope_correlation": 0.9967701758102417, "300–1000 Hz_40ms_modulation_mae_db": 0.1629598203063263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3584977016620574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1431013873691995, "1000–3000 Hz_10ms_envelope_correlation": 0.9862193455139482, "1000–3000 Hz_10ms_modulation_mae_db": 0.5418101848734626, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9049431463627825, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09881645059699501, "1000–3000 Hz_40ms_envelope_correlation": 0.9935074355355675, "1000–3000 Hz_40ms_modulation_mae_db": 0.4299385539447367, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2432795931584135, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13279852894618202, "3000–6000 Hz_10ms_envelope_correlation": 0.9875799719035543, "3000–6000 Hz_10ms_modulation_mae_db": 0.2591366416697522, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8604953398453926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08606827150795957, "3000–6000 Hz_40ms_envelope_correlation": 0.994667881822148, "3000–6000 Hz_40ms_modulation_mae_db": 0.17096992619719228, "mr_stft_db": 5.761214738052904, "band_mae_db": 0.15438575641227906, "rms_gain_error_db": -0.0897669608391097, "output_peak": 0.6168377905021944}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9140872002361259, "全帯域_10ms_rms_envelope_nrmse": 0.09883332650153367, "全帯域_10ms_envelope_correlation": 0.9916499554136634, "全帯域_10ms_modulation_mae_db": 0.3469674335943746, "全帯域_40ms_log_rms_mae_db": 0.5317964371748446, "全帯域_40ms_rms_envelope_nrmse": 0.056370582925604006, "全帯域_40ms_envelope_correlation": 0.9973029379962238, "全帯域_40ms_modulation_mae_db": 0.6100839935332822, "300–1000 Hz_10ms_log_rms_mae_db": 1.3857367615471967, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12105000105599524, "300–1000 Hz_10ms_envelope_correlation": 0.988687137017807, "300–1000 Hz_10ms_modulation_mae_db": 0.36124170749952883, "300–1000 Hz_40ms_log_rms_mae_db": 0.8469842296415891, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07305134118607558, "300–1000 Hz_40ms_envelope_correlation": 0.9959044362166661, "300–1000 Hz_40ms_modulation_mae_db": 0.44539635170570296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3368372211237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1429372334865013, "1000–3000 Hz_10ms_envelope_correlation": 0.9861165253322898, "1000–3000 Hz_10ms_modulation_mae_db": 0.9592844704086702, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9060670966454945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07915926163433025, "1000–3000 Hz_40ms_envelope_correlation": 0.9958567448505039, "1000–3000 Hz_40ms_modulation_mae_db": 0.5467277737465012, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2839674435628732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1742646145332306, "3000–6000 Hz_10ms_envelope_correlation": 0.9786891486379578, "3000–6000 Hz_10ms_modulation_mae_db": 0.585006639584925, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8894114565644156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12514981094630737, "3000–6000 Hz_40ms_envelope_correlation": 0.9884603777781968, "3000–6000 Hz_40ms_modulation_mae_db": 0.7603212533312709, "mr_stft_db": 5.769564550838088, "band_mae_db": 0.1583626860901192, "rms_gain_error_db": -0.14391029360719496, "output_peak": 0.616763564179907}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8776852505498265, "全帯域_10ms_rms_envelope_nrmse": 0.09683801105403655, "全帯域_10ms_envelope_correlation": 0.9919523054276725, "全帯域_10ms_modulation_mae_db": 0.6214331503532479, "全帯域_40ms_log_rms_mae_db": 0.5307067901852023, "全帯域_40ms_rms_envelope_nrmse": 0.05842861998699424, "全帯域_40ms_envelope_correlation": 0.9970350494272997, "全帯域_40ms_modulation_mae_db": 0.28864736164986454, "300–1000 Hz_10ms_log_rms_mae_db": 1.3694926127740226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11234701331670403, "300–1000 Hz_10ms_envelope_correlation": 0.9902187296416467, "300–1000 Hz_10ms_modulation_mae_db": 0.43201291625558197, "300–1000 Hz_40ms_log_rms_mae_db": 0.8002431914062531, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06526071963477637, "300–1000 Hz_40ms_envelope_correlation": 0.9966978751551958, "300–1000 Hz_40ms_modulation_mae_db": 0.2677563524238069, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3745336944516513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1693942990095837, "1000–3000 Hz_10ms_envelope_correlation": 0.9802887263437231, "1000–3000 Hz_10ms_modulation_mae_db": 1.608748196877232, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9493545300685334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10418577234826924, "1000–3000 Hz_40ms_envelope_correlation": 0.9923853750816809, "1000–3000 Hz_40ms_modulation_mae_db": 1.4603462973477315, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2664475833327702, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1461121148941471, "3000–6000 Hz_10ms_envelope_correlation": 0.9853731526077207, "3000–6000 Hz_10ms_modulation_mae_db": 0.4844326146245637, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8255800390416963, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09064301667252805, "3000–6000 Hz_40ms_envelope_correlation": 0.9948000670652668, "3000–6000 Hz_40ms_modulation_mae_db": 0.4689259977051274, "mr_stft_db": 5.772548991764572, "band_mae_db": 0.22311526194800024, "rms_gain_error_db": -0.12642155609101213, "output_peak": 0.6437546320030427}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9295098711826043, "全帯域_10ms_rms_envelope_nrmse": 0.10305698826432595, "全帯域_10ms_envelope_correlation": 0.9909477201216271, "全帯域_10ms_modulation_mae_db": 0.5333543832838475, "全帯域_40ms_log_rms_mae_db": 0.5277332167759649, "全帯域_40ms_rms_envelope_nrmse": 0.05318614244144977, "全帯域_40ms_envelope_correlation": 0.9976912463974723, "全帯域_40ms_modulation_mae_db": 0.40125630658880596, "300–1000 Hz_10ms_log_rms_mae_db": 1.4461137166915687, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12363632944393409, "300–1000 Hz_10ms_envelope_correlation": 0.9881507223908951, "300–1000 Hz_10ms_modulation_mae_db": 0.35207273507519465, "300–1000 Hz_40ms_log_rms_mae_db": 0.8282445699201998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0703344339812093, "300–1000 Hz_40ms_envelope_correlation": 0.9961719241935966, "300–1000 Hz_40ms_modulation_mae_db": 0.36951507299483416, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3185586552647557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13673486348467762, "1000–3000 Hz_10ms_envelope_correlation": 0.9875067106510502, "1000–3000 Hz_10ms_modulation_mae_db": 1.1219534237172792, "1000–3000 Hz_40ms_log_rms_mae_db": 0.825899445907142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07263499904123681, "1000–3000 Hz_40ms_envelope_correlation": 0.9968671278509585, "1000–3000 Hz_40ms_modulation_mae_db": 0.3938333195419376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2468540845978253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13318476109196456, "3000–6000 Hz_10ms_envelope_correlation": 0.9876320183329856, "3000–6000 Hz_10ms_modulation_mae_db": 0.6263480593572721, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8153369599999274, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07721598561800723, "3000–6000 Hz_40ms_envelope_correlation": 0.9959718038623976, "3000–6000 Hz_40ms_modulation_mae_db": 0.7020316820341178, "mr_stft_db": 5.7561237536796765, "band_mae_db": 0.19260209806396397, "rms_gain_error_db": -0.1629137009951847, "output_peak": 0.5962578666611758}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.873505603118301, "全帯域_10ms_rms_envelope_nrmse": 0.08828479591589178, "全帯域_10ms_envelope_correlation": 0.9933794764533331, "全帯域_10ms_modulation_mae_db": 0.5630090656725466, "全帯域_40ms_log_rms_mae_db": 0.4613184747697832, "全帯域_40ms_rms_envelope_nrmse": 0.04516012857927475, "全帯域_40ms_envelope_correlation": 0.9983459028609736, "全帯域_40ms_modulation_mae_db": 0.19516176798400806, "300–1000 Hz_10ms_log_rms_mae_db": 1.4020906973161855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10870155133986457, "300–1000 Hz_10ms_envelope_correlation": 0.9908852492611321, "300–1000 Hz_10ms_modulation_mae_db": 0.5576591152400621, "300–1000 Hz_40ms_log_rms_mae_db": 0.7507649094475575, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057645305150305345, "300–1000 Hz_40ms_envelope_correlation": 0.9975011133096282, "300–1000 Hz_40ms_modulation_mae_db": 0.22433572146508396, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3516030777812502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14558185218393183, "1000–3000 Hz_10ms_envelope_correlation": 0.9856164192251373, "1000–3000 Hz_10ms_modulation_mae_db": 1.3717203280728503, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8564958707278469, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07994973032090609, "1000–3000 Hz_40ms_envelope_correlation": 0.9958316631679699, "1000–3000 Hz_40ms_modulation_mae_db": 0.9639395512216172, "3000–6000 Hz_10ms_log_rms_mae_db": 1.257758132423785, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13614794397280516, "3000–6000 Hz_10ms_envelope_correlation": 0.9871961046494644, "3000–6000 Hz_10ms_modulation_mae_db": 0.3400186700133712, "3000–6000 Hz_40ms_log_rms_mae_db": 0.797480067062882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07610741636162974, "3000–6000 Hz_40ms_envelope_correlation": 0.9963358854701448, "3000–6000 Hz_40ms_modulation_mae_db": 0.4276790845542628, "mr_stft_db": 5.764296539987938, "band_mae_db": 0.20133448412507712, "rms_gain_error_db": -0.13969620535868021, "output_peak": 0.6302923815134728}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9376876646553369, "全帯域_10ms_rms_envelope_nrmse": 0.10629605524164523, "全帯域_10ms_envelope_correlation": 0.9903540193572771, "全帯域_10ms_modulation_mae_db": 0.3979740437565698, "全帯域_40ms_log_rms_mae_db": 0.5178624106576403, "全帯域_40ms_rms_envelope_nrmse": 0.05574161349298758, "全帯域_40ms_envelope_correlation": 0.9974424330979124, "全帯域_40ms_modulation_mae_db": 0.3773822425898607, "300–1000 Hz_10ms_log_rms_mae_db": 1.4783088935167241, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12465955793682314, "300–1000 Hz_10ms_envelope_correlation": 0.9878613193155769, "300–1000 Hz_10ms_modulation_mae_db": 0.17548661562087828, "300–1000 Hz_40ms_log_rms_mae_db": 0.8766669578572753, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06416943540161509, "300–1000 Hz_40ms_envelope_correlation": 0.9966919416471045, "300–1000 Hz_40ms_modulation_mae_db": 0.34875962396716004, "1000–3000 Hz_10ms_log_rms_mae_db": 1.369983758650144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16146299556891203, "1000–3000 Hz_10ms_envelope_correlation": 0.9840237138462619, "1000–3000 Hz_10ms_modulation_mae_db": 1.2242555636383847, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9138285645594771, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11139248201937985, "1000–3000 Hz_40ms_envelope_correlation": 0.9938064615347135, "1000–3000 Hz_40ms_modulation_mae_db": 1.0192551308108493, "3000–6000 Hz_10ms_log_rms_mae_db": 1.236952034432792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12736187241969216, "3000–6000 Hz_10ms_envelope_correlation": 0.9886937691695209, "3000–6000 Hz_10ms_modulation_mae_db": 0.45266203085472273, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7739438178013678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07253815903092355, "3000–6000 Hz_40ms_envelope_correlation": 0.9964738091235035, "3000–6000 Hz_40ms_modulation_mae_db": 0.4858892264656463, "mr_stft_db": 5.762693827084195, "band_mae_db": 0.2826040642634811, "rms_gain_error_db": -0.16571457367766812, "output_peak": 0.5995781649628574}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.59898363128587, "spectral_convergence": 0.31104303789355214, "spectral_convergence_status": "ok", "amplitude_floor": 1.4064891899964367e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.532703244587193, "spectral_convergence": 0.30054171547015723, "spectral_convergence_status": "ok", "amplitude_floor": 1.2304331075199438e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.194293361497406, "spectral_convergence": 0.32633903381747703, "spectral_convergence_status": "ok", "amplitude_floor": 7.896502575224385e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.444718915891811e-05, "output_power": 5.627011719545015e-05, "delta_db": 0.1430235549453619, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025153246625132694, "output_power": 0.00026870346739145054, "delta_db": 0.2867922162387041, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010938925867664089, "output_power": 0.0010293545120816352, "delta_db": -0.2640970660754578, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031036369064007244, "output_power": 0.00030457863969439957, "delta_db": -0.0817146486716735, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001328835538454751, "output_power": 0.00012529496232006285, "delta_db": -0.25537624539242043, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6997042780401114e-05, "output_power": 3.707128996182355e-05, "delta_db": 0.008706865888614272, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.173285099535372, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13809721022638727, "pitch": null, "output_rms_dbfs": -27.39624953394317, "output_peak": 0.5879714031370464, "reference_rms_dbfs": -27.258152323716782, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7753267457901565}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019413100530357063, "log_rms_mae_db": 0.8903895747715466, "rms_envelope_nrmse": 0.08999461467282908, "envelope_correlation": 0.9931050594810209, "rms_gain_error_db": -0.13810010290833769, "reference_mean_power": 0.0018816420370153617, "output_mean_power": 0.0018227495370931792}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1507655155627412e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005726194534917747, "output_power": 0.0005458991287490833, "reference_active": true, "delta_db": -0.2075369696425504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.421513884012437e-05, "output_power": 8.237671561911945e-05, "reference_active": true, "delta_db": -0.5831621992555809}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.679657413963232e-05, "output_power": 1.8661430912349134e-05, "reference_active": true, "delta_db": 0.45724230062325505}], "active_band_count": 3, "active_band_mae_db": 0.41598048984046215}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016416501002274435, "log_rms_mae_db": 0.4895715505137562, "rms_envelope_nrmse": 0.04896942417634289, "envelope_correlation": 0.9980081203317364, "rms_gain_error_db": -0.13817183610795097, "reference_mean_power": 0.001890759403997248, "output_mean_power": 0.001831551292208609}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0781635397104837e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005383866955860211, "output_power": 0.0005122377715527962, "reference_active": true, "delta_db": -0.21622719764316323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8874025426333776e-05, "output_power": 3.592394184167537e-05, "reference_active": true, "delta_db": -0.3427552976243087}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650199092025111e-06, "output_power": 1.6707625397329443e-06, "reference_active": true, "delta_db": 0.283948626197881}], "active_band_count": 3, "active_band_mae_db": 0.280977040488451}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017990629858894173, "log_rms_mae_db": 1.3275674970099591, "rms_envelope_nrmse": 0.11439006112815228, "envelope_correlation": 0.9898879478056111, "rms_gain_error_db": -0.175452388127054, "reference_mean_power": 0.0012781810161111924, "output_mean_power": 0.0012275724246920319}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.355790711503392e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004472139003696596, "output_power": 0.0004257985934199251, "reference_active": true, "delta_db": -0.21311071435170215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.621773891011176e-05, "output_power": 6.620607922180858e-05, "reference_active": true, "delta_db": -0.6115819139661772}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7010237353267386e-05, "output_power": 2.0099407999278573e-05, "reference_active": true, "delta_db": 0.7247289244926707}], "active_band_count": 3, "active_band_mae_db": 0.5164738509368499}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000149706333097475, "log_rms_mae_db": 0.7349625006031024, "rms_envelope_nrmse": 0.05935453107103731, "envelope_correlation": 0.9973779114938774, "rms_gain_error_db": -0.17551186182032652, "reference_mean_power": 0.0012844845607787134, "output_mean_power": 0.0012336094918069384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.840745806614025e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004236215690036061, "output_power": 0.00040406261691168976, "reference_active": true, "delta_db": -0.2052939220171438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1600142098026355e-05, "output_power": 2.9215650462763923e-05, "reference_active": true, "delta_db": -0.3407347556934986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5445822829049627e-06, "output_power": 1.6601686344869338e-06, "reference_active": true, "delta_db": 0.31341155524106323}], "active_band_count": 3, "active_band_mae_db": 0.28648007765056854}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010005798255669499, "log_rms_mae_db": 1.375754811752196, "rms_envelope_nrmse": 0.14410737272413415, "envelope_correlation": 0.9858189916700842, "rms_gain_error_db": -0.1432596049557908, "reference_mean_power": 0.0003605696849433608, "output_mean_power": 0.0003488696985086181}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5037729366163507e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001406674030835393, "output_power": 0.00013143122074890992, "reference_active": true, "delta_db": -0.2949492815493775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0544824834133773e-05, "output_power": 1.9133640707983113e-05, "reference_active": true, "delta_db": -0.30904828294567194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4458512826457347e-06, "output_power": 4.83019631031164e-06, "reference_active": true, "delta_db": 1.4666825176820577}], "active_band_count": 3, "active_band_mae_db": 0.6902266940590357}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.18211380942645e-05, "log_rms_mae_db": 0.9220058807176477, "rms_envelope_nrmse": 0.09205942811258988, "envelope_correlation": 0.9941012132988988, "rms_gain_error_db": -0.1432637084732101, "reference_mean_power": 0.00036238111197119544, "output_mean_power": 0.00035062201594818806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3815388264892723e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001355192968806531, "output_power": 0.00012679226744651006, "reference_active": true, "delta_db": -0.2890837125861699}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.971075792205542e-06, "output_power": 7.870493424805515e-06, "reference_active": true, "delta_db": -0.0551497816238257}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.416322058192608e-07, "output_power": 4.2981057095010983e-07, "reference_active": true, "delta_db": 2.5012227400152023}], "active_band_count": 3, "active_band_mae_db": 0.9484854114083993}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.283283642007837e-05, "log_rms_mae_db": 1.2785156751100426, "rms_envelope_nrmse": 0.13793908273108824, "envelope_correlation": 0.9866274094619502, "rms_gain_error_db": -0.1790284742937538, "reference_mean_power": 6.230386432542126e-05, "output_mean_power": 5.978773992447202e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0221136919924677e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.16042156208526e-05, "output_power": 1.1824266968625326e-05, "reference_active": true, "delta_db": 0.08158436733490468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3772446197384082e-06, "output_power": 2.626302241043425e-06, "reference_active": true, "delta_db": 0.4327083109947768}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4509999616936e-07, "output_power": 5.901503991409319e-07, "reference_active": true, "delta_db": 1.2250511464512452}], "active_band_count": 3, "active_band_mae_db": 0.5797812749269756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.758960619357764e-05, "log_rms_mae_db": 0.8536836545081438, "rms_envelope_nrmse": 0.08663840517912287, "envelope_correlation": 0.9946865475040926, "rms_gain_error_db": -0.17902881007400565, "reference_mean_power": 6.26170059672204e-05, "output_mean_power": 6.008823078151562e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.846748801507289e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1232117533806203e-05, "output_power": 1.1651267631723926e-05, "reference_active": true, "delta_db": 0.15911538892974647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.067581168241283e-06, "output_power": 1.2887030492094403e-06, "reference_active": true, "delta_db": 0.8175195172659667}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.579800269976513e-08, "output_power": 3.3293761026156695e-08, "reference_active": true, "delta_db": 1.107767740892958}], "active_band_count": 3, "active_band_mae_db": 0.6948008823628905}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8485441661318673, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "133224d8f3430fd596b044f17a6f19735f10bd4bb9858a602960ca4618ef4fcd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9852430119140491, "standard_deviation": 0.0022225130302227854, "minimum": 0.9802887263437231, "maximum": 0.9875067106510502, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3661026151810691, "standard_deviation": 0.0368112597693485, "minimum": 1.3185586552647557, "maximum": 1.4430520007627983, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0202834569466401, "standard_deviation": 0.3788558091640175, "minimum": 0.5418101848734626, "maximum": 1.608748196877232, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1481073444732688, "standard_deviation": 0.01120047907465162, "minimum": 0.13673486348467762, "maximum": 0.1693942990095837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946921518662963, "standard_deviation": 0.0014872252619635574, "minimum": 0.9923853750816809, "maximum": 0.9968671278509585, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9055260299387327, "standard_deviation": 0.04569614789388359, "minimum": 0.825899445907142, "maximum": 0.9656137045209364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.795660040668, "standard_deviation": 0.36603532121125393, "minimum": 0.3938333195419376, "maximum": 1.4603462973477315, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09033491235694954, "standard_deviation": 0.01357770419302599, "minimum": 0.07263499904123681, "maximum": 0.11139248201937985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9862993189885056, "standard_deviation": 0.003254463677829267, "minimum": 0.9786891486379578, "maximum": 0.9886937691695209, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2565654360851344, "standard_deviation": 0.0181450478071113, "minimum": 1.236952034432792, "maximum": 1.2839674435628732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4556950675201248, "standard_deviation": 0.13784280864985132, "minimum": 0.2591366416697522, "maximum": 0.6263480593572721, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13937647305783263, "standard_deviation": 0.015349242716239631, "minimum": 0.12720286587355123, "maximum": 0.1742646145332306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9946596462109287, "standard_deviation": 0.002612206807946161, "minimum": 0.9884603777781968, "maximum": 0.9964738091235035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8311846283163993, "standard_deviation": 0.03675464982787833, "minimum": 0.7739438178013678, "maximum": 0.8894114565644156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5078739038245164, "standard_deviation": 0.2008854836450652, "minimum": 0.17096992619719228, "maximum": 0.7603212533312709, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08628034669889098, "standard_deviation": 0.016956549001884157, "minimum": 0.07253815903092355, "maximum": 0.12514981094630737, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9893050849244756, "standard_deviation": 0.0010651034119277972, "minimum": 0.9878613193155769, "maximum": 0.9908852492611321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.387074668725075, "standard_deviation": 0.05510873436354563, "minimum": 1.3153637871025514, "maximum": 1.4783088935167241, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3731389959252052, "standard_deviation": 0.16445438228872147, "minimum": 0.0958080943073123, "maximum": 0.5576591152400621, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11726307471798617, "standard_deviation": 0.005744325773151863, "minimum": 0.10870155133986457, "maximum": 0.12465955793682314, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967859257370721, "standard_deviation": 0.0005582972910305807, "minimum": 0.9959044362166661, "maximum": 0.9975011133096282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7987895959061282, "standard_deviation": 0.04971191357718098, "minimum": 0.7349625006031024, "maximum": 0.8766669578572753, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.35197330040690233, "standard_deviation": 0.16947246002160368, "minimum": 0.1629598203063263, "maximum": 0.7105833827417358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06411218271959807, "standard_deviation": 0.005398823275778608, "minimum": 0.057645305150305345, "maximum": 0.07305134118607558, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19631954337522617, "standard_deviation": 0.0415123257687826, "minimum": 0.15438575641227906, "maximum": 0.2826040642634811, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.767339088535371, "standard_deviation": 0.007389870440330091, "minimum": 5.7561237536796765, "maximum": 5.776943561085436, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6128208789943682, "standard_deviation": 0.018376821682231335, "minimum": 0.5879714031370464, "maximum": 0.6437546320030427, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1341452153072981, "standard_deviation": 0.0260537705332514, "minimum": -0.16571457367766812, "maximum": -0.0897669608391097, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.992105583884973, "standard_deviation": 0.0010715910660673966, "minimum": 0.9903540193572771, "maximum": 0.9933794764533331, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8964884250669387, "standard_deviation": 0.027431983155841253, "minimum": 0.8625629585540628, "maximum": 0.9376876646553369, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4349593499421635, "standard_deviation": 0.1655709597315712, "minimum": 0.0938331107874301, "maximum": 0.6214331503532479, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09590094134409441, "standard_deviation": 0.00646829475435645, "minimum": 0.08828479591589178, "maximum": 0.10629605524164523, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977130233178644, "standard_deviation": 0.0004333779817439072, "minimum": 0.9970350494272997, "maximum": 0.9983459028609736, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5018859590935016, "standard_deviation": 0.02861005134398061, "minimum": 0.4613184747697832, "maximum": 0.5317964371748446, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3646517951486691, "standard_deviation": 0.1533508306631273, "minimum": 0.19516176798400806, "maximum": 0.6100839935332822, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.051914063538466365, "standard_deviation": 0.004704574365115424, "minimum": 0.04516012857927475, "maximum": 0.05842861998699424, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-961-0011", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.8297817707061768, "predicted_mos": 2.8297817707061768, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b3ffcbd6f5265ab075a5d8369f722a62af2e9a806446c66b4d9ce92605d2cf52", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.1226881998591125, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8592205783357554, "全帯域_10ms_rms_envelope_nrmse": 0.08680853472409356, "全帯域_10ms_envelope_correlation": 0.9934920731018162, "全帯域_10ms_modulation_mae_db": 0.15803823195277752, "全帯域_40ms_log_rms_mae_db": 0.4583074000477415, "全帯域_40ms_rms_envelope_nrmse": 0.04091644126694648, "全帯域_40ms_envelope_correlation": 0.9985248473414563, "全帯域_40ms_modulation_mae_db": 0.3056498639847575, "300–1000 Hz_10ms_log_rms_mae_db": 1.436894235830512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11079474076061202, "300–1000 Hz_10ms_envelope_correlation": 0.9904137713064451, "300–1000 Hz_10ms_modulation_mae_db": 0.28624892481543, "300–1000 Hz_40ms_log_rms_mae_db": 0.7878975273641718, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05398845279438951, "300–1000 Hz_40ms_envelope_correlation": 0.9976862404591822, "300–1000 Hz_40ms_modulation_mae_db": 0.07541004081103672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.380317814945297, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13652521540730472, "1000–3000 Hz_10ms_envelope_correlation": 0.9872407416473183, "1000–3000 Hz_10ms_modulation_mae_db": 1.0320557334944844, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8930529305824555, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07364278335154713, "1000–3000 Hz_40ms_envelope_correlation": 0.9961795851484101, "1000–3000 Hz_40ms_modulation_mae_db": 0.8795914002244639, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3436965467405824, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13118619856564506, "3000–6000 Hz_10ms_envelope_correlation": 0.9884335404796162, "3000–6000 Hz_10ms_modulation_mae_db": 0.21922697151522585, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8700449138370119, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08018148248693269, "3000–6000 Hz_40ms_envelope_correlation": 0.9961994423173449, "3000–6000 Hz_40ms_modulation_mae_db": 0.1929171300782165, "mr_stft_db": 5.8505333730382345, "band_mae_db": 0.14286192778556628, "rms_gain_error_db": -0.07269182753774478, "output_peak": 0.6081810744641095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.86946566151716, "全帯域_10ms_rms_envelope_nrmse": 0.0937482150251464, "全帯域_10ms_envelope_correlation": 0.9924946246373793, "全帯域_10ms_modulation_mae_db": 0.5188733135645749, "全帯域_40ms_log_rms_mae_db": 0.4713481453432922, "全帯域_40ms_rms_envelope_nrmse": 0.04824625745898071, "全帯域_40ms_envelope_correlation": 0.9980581474692842, "全帯域_40ms_modulation_mae_db": 0.25613113331236537, "300–1000 Hz_10ms_log_rms_mae_db": 1.41824591515754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12077786514313213, "300–1000 Hz_10ms_envelope_correlation": 0.9886804724565011, "300–1000 Hz_10ms_modulation_mae_db": 0.37997823716312396, "300–1000 Hz_40ms_log_rms_mae_db": 0.8006377270375423, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0666479126753332, "300–1000 Hz_40ms_envelope_correlation": 0.9964912286128771, "300–1000 Hz_40ms_modulation_mae_db": 0.11258459009315795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4560283683471589, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14374228206035497, "1000–3000 Hz_10ms_envelope_correlation": 0.9862926324437767, "1000–3000 Hz_10ms_modulation_mae_db": 0.6704829950149828, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9616930823961616, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08436054517390024, "1000–3000 Hz_40ms_envelope_correlation": 0.9957743323682645, "1000–3000 Hz_40ms_modulation_mae_db": 0.7767593659631098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.337851364784485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13011173933795547, "3000–6000 Hz_10ms_envelope_correlation": 0.9882750863765991, "3000–6000 Hz_10ms_modulation_mae_db": 0.3140599384213363, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8268593061670426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07437603080808018, "3000–6000 Hz_40ms_envelope_correlation": 0.9963843103258435, "3000–6000 Hz_40ms_modulation_mae_db": 0.17199596393357033, "mr_stft_db": 5.846411281206863, "band_mae_db": 0.2114477470147218, "rms_gain_error_db": -0.1329170149520859, "output_peak": 0.5981752010897473}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8550978518637494, "全帯域_10ms_rms_envelope_nrmse": 0.087463680218153, "全帯域_10ms_envelope_correlation": 0.993641293733487, "全帯域_10ms_modulation_mae_db": 0.5618637790761672, "全帯域_40ms_log_rms_mae_db": 0.46171339786069254, "全帯域_40ms_rms_envelope_nrmse": 0.046352600936695756, "全帯域_40ms_envelope_correlation": 0.9984235294722054, "全帯域_40ms_modulation_mae_db": 0.386102070359904, "300–1000 Hz_10ms_log_rms_mae_db": 1.4018300542176496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11051567294394715, "300–1000 Hz_10ms_envelope_correlation": 0.9906705539923831, "300–1000 Hz_10ms_modulation_mae_db": 0.25204119498633376, "300–1000 Hz_40ms_log_rms_mae_db": 0.7566991871423567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06025163434558371, "300–1000 Hz_40ms_envelope_correlation": 0.9974029666679156, "300–1000 Hz_40ms_modulation_mae_db": 0.3199531543249412, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3577723411436684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15759928582265842, "1000–3000 Hz_10ms_envelope_correlation": 0.983813147812599, "1000–3000 Hz_10ms_modulation_mae_db": 1.4020565055077807, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8926528975674032, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09679016795994654, "1000–3000 Hz_40ms_envelope_correlation": 0.9947692385031736, "1000–3000 Hz_40ms_modulation_mae_db": 1.305725441153636, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3819680509624623, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14077550679414083, "3000–6000 Hz_10ms_envelope_correlation": 0.986188423968564, "3000–6000 Hz_10ms_modulation_mae_db": 0.48504576987385484, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8917298787741199, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08557640847550804, "3000–6000 Hz_40ms_envelope_correlation": 0.9950360593056764, "3000–6000 Hz_40ms_modulation_mae_db": 0.6522581316760802, "mr_stft_db": 5.850788922884424, "band_mae_db": 0.22974312292680746, "rms_gain_error_db": -0.18669552551568613, "output_peak": 0.6049679614046573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8860499908318393, "全帯域_10ms_rms_envelope_nrmse": 0.08949907858774815, "全帯域_10ms_envelope_correlation": 0.9931117442396178, "全帯域_10ms_modulation_mae_db": 0.44621488377172636, "全帯域_40ms_log_rms_mae_db": 0.4946680633155722, "全帯域_40ms_rms_envelope_nrmse": 0.04616038264935542, "全帯域_40ms_envelope_correlation": 0.9981611251559124, "全帯域_40ms_modulation_mae_db": 0.3249201397386166, "300–1000 Hz_10ms_log_rms_mae_db": 1.4386075666989873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1113867983566512, "300–1000 Hz_10ms_envelope_correlation": 0.9903618921764791, "300–1000 Hz_10ms_modulation_mae_db": 0.43637684803664456, "300–1000 Hz_40ms_log_rms_mae_db": 0.8273214891397515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059704087089180534, "300–1000 Hz_40ms_envelope_correlation": 0.9972192092217567, "300–1000 Hz_40ms_modulation_mae_db": 0.2740387832396774, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4390369058246513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14907253056537717, "1000–3000 Hz_10ms_envelope_correlation": 0.9847747259403757, "1000–3000 Hz_10ms_modulation_mae_db": 0.778625597821331, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9155785494134663, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08485753112419701, "1000–3000 Hz_40ms_envelope_correlation": 0.9950346230996852, "1000–3000 Hz_40ms_modulation_mae_db": 0.5314685828959949, "3000–6000 Hz_10ms_log_rms_mae_db": 1.347879186764985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14165521816502238, "3000–6000 Hz_10ms_envelope_correlation": 0.9858084838865222, "3000–6000 Hz_10ms_modulation_mae_db": 0.3962094227414598, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9054073447256977, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08961897070271249, "3000–6000 Hz_40ms_envelope_correlation": 0.9941382761956787, "3000–6000 Hz_40ms_modulation_mae_db": 0.2982902104444622, "mr_stft_db": 5.849568758539998, "band_mae_db": 0.14724984359229273, "rms_gain_error_db": -0.10264344741188391, "output_peak": 0.6789912325355518}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9021926171581918, "全帯域_10ms_rms_envelope_nrmse": 0.10839576292070462, "全帯域_10ms_envelope_correlation": 0.9898959621938664, "全帯域_10ms_modulation_mae_db": 0.5516500113110454, "全帯域_40ms_log_rms_mae_db": 0.5166346443400809, "全帯域_40ms_rms_envelope_nrmse": 0.06199950098680363, "全帯域_40ms_envelope_correlation": 0.9966850320323467, "全帯域_40ms_modulation_mae_db": 0.8567010699374755, "300–1000 Hz_10ms_log_rms_mae_db": 1.3967428095846057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12854346974240913, "300–1000 Hz_10ms_envelope_correlation": 0.9870642105638244, "300–1000 Hz_10ms_modulation_mae_db": 0.6981302444554046, "300–1000 Hz_40ms_log_rms_mae_db": 0.7921466250957072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07456819616571302, "300–1000 Hz_40ms_envelope_correlation": 0.9955385971509071, "300–1000 Hz_40ms_modulation_mae_db": 0.9644645717157886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4209939880568572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1456545495165074, "1000–3000 Hz_10ms_envelope_correlation": 0.986123063327018, "1000–3000 Hz_10ms_modulation_mae_db": 0.9445059273456916, "1000–3000 Hz_40ms_log_rms_mae_db": 0.92477819562911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09518854307509245, "1000–3000 Hz_40ms_envelope_correlation": 0.9946071985426251, "1000–3000 Hz_40ms_modulation_mae_db": 0.5298587201733661, "3000–6000 Hz_10ms_log_rms_mae_db": 1.334231955486401, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14202436253114448, "3000–6000 Hz_10ms_envelope_correlation": 0.986756470956502, "3000–6000 Hz_10ms_modulation_mae_db": 1.042299923676753, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9002456166858762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09276435233317172, "3000–6000 Hz_40ms_envelope_correlation": 0.9951184805389167, "3000–6000 Hz_40ms_modulation_mae_db": 1.0435520943033314, "mr_stft_db": 5.839860680728173, "band_mae_db": 0.2599546247700557, "rms_gain_error_db": -0.14236164003643328, "output_peak": 0.5887304222111513}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8855421026130085, "全帯域_10ms_rms_envelope_nrmse": 0.08925765216063461, "全帯域_10ms_envelope_correlation": 0.9935938630468831, "全帯域_10ms_modulation_mae_db": 0.5397878386010494, "全帯域_40ms_log_rms_mae_db": 0.45939593250328037, "全帯域_40ms_rms_envelope_nrmse": 0.049214226709600274, "全帯域_40ms_envelope_correlation": 0.9984553617368787, "全帯域_40ms_modulation_mae_db": 0.5169551389206015, "300–1000 Hz_10ms_log_rms_mae_db": 1.440012335434828, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11254806774778174, "300–1000 Hz_10ms_envelope_correlation": 0.9904342615108507, "300–1000 Hz_10ms_modulation_mae_db": 0.43133621906011504, "300–1000 Hz_40ms_log_rms_mae_db": 0.7559604477962697, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061164393301560684, "300–1000 Hz_40ms_envelope_correlation": 0.9974744986399314, "300–1000 Hz_40ms_modulation_mae_db": 0.45459391012668116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4275684719624395, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15899899319352365, "1000–3000 Hz_10ms_envelope_correlation": 0.9836578894362825, "1000–3000 Hz_10ms_modulation_mae_db": 1.2209394388449948, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9344771997845915, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11266205005911595, "1000–3000 Hz_40ms_envelope_correlation": 0.9924591995228202, "1000–3000 Hz_40ms_modulation_mae_db": 0.9727071537551071, "3000–6000 Hz_10ms_log_rms_mae_db": 1.399045555836025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14707024777303215, "3000–6000 Hz_10ms_envelope_correlation": 0.985033347529991, "3000–6000 Hz_10ms_modulation_mae_db": 0.30496223440239073, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9110515146550692, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08760589926458, "3000–6000 Hz_40ms_envelope_correlation": 0.9950381894904219, "3000–6000 Hz_40ms_modulation_mae_db": 0.25137776829783576, "mr_stft_db": 5.850685086929324, "band_mae_db": 0.2323509142353625, "rms_gain_error_db": -0.24259439982153636, "output_peak": 0.6080108750503134}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8350437855474186, "全帯域_10ms_rms_envelope_nrmse": 0.08374212854084502, "全帯域_10ms_envelope_correlation": 0.9940452172358683, "全帯域_10ms_modulation_mae_db": 0.3110388083768206, "全帯域_40ms_log_rms_mae_db": 0.4669589507004524, "全帯域_40ms_rms_envelope_nrmse": 0.042336026754633084, "全帯域_40ms_envelope_correlation": 0.998556792792993, "全帯域_40ms_modulation_mae_db": 0.638878657222708, "300–1000 Hz_10ms_log_rms_mae_db": 1.404692762636047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11233030378024672, "300–1000 Hz_10ms_envelope_correlation": 0.9902114854200911, "300–1000 Hz_10ms_modulation_mae_db": 0.267104446496356, "300–1000 Hz_40ms_log_rms_mae_db": 0.7930973160238971, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058861738911558874, "300–1000 Hz_40ms_envelope_correlation": 0.9973568927957382, "300–1000 Hz_40ms_modulation_mae_db": 0.17273438911811156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3933345708179277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14540471276124814, "1000–3000 Hz_10ms_envelope_correlation": 0.985639392594388, "1000–3000 Hz_10ms_modulation_mae_db": 0.7989515998626469, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9218808762368279, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08239574347261601, "1000–3000 Hz_40ms_envelope_correlation": 0.9956162248032353, "1000–3000 Hz_40ms_modulation_mae_db": 0.42047406964643375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3668528911718691, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1428646518065397, "3000–6000 Hz_10ms_envelope_correlation": 0.9859073210661968, "3000–6000 Hz_10ms_modulation_mae_db": 0.44587667970242345, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8689668091013211, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08443085282000332, "3000–6000 Hz_40ms_envelope_correlation": 0.995402451234258, "3000–6000 Hz_40ms_modulation_mae_db": 0.4640191694151741, "mr_stft_db": 5.842919979749893, "band_mae_db": 0.16907556488103337, "rms_gain_error_db": -0.13172191596462715, "output_peak": 0.6020387565548265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8830973705099879, "全帯域_10ms_rms_envelope_nrmse": 0.1016624540424518, "全帯域_10ms_envelope_correlation": 0.9912019299506566, "全帯域_10ms_modulation_mae_db": 0.4783919903870861, "全帯域_40ms_log_rms_mae_db": 0.48074053249428617, "全帯域_40ms_rms_envelope_nrmse": 0.05498241918058549, "全帯域_40ms_envelope_correlation": 0.9975149766672621, "全帯域_40ms_modulation_mae_db": 0.23595950929444767, "300–1000 Hz_10ms_log_rms_mae_db": 1.4711959509958175, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13205294379683175, "300–1000 Hz_10ms_envelope_correlation": 0.9865209166321742, "300–1000 Hz_10ms_modulation_mae_db": 0.21184556013418657, "300–1000 Hz_40ms_log_rms_mae_db": 0.8266467103800539, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07199947483867224, "300–1000 Hz_40ms_envelope_correlation": 0.9960904240759644, "300–1000 Hz_40ms_modulation_mae_db": 0.42874737780439426, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4006657540549876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13950509854409687, "1000–3000 Hz_10ms_envelope_correlation": 0.9867434009188394, "1000–3000 Hz_10ms_modulation_mae_db": 1.151918359495914, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9090669361314877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07988900643764425, "1000–3000 Hz_40ms_envelope_correlation": 0.9957019556619103, "1000–3000 Hz_40ms_modulation_mae_db": 0.8257155991191191, "3000–6000 Hz_10ms_log_rms_mae_db": 1.385615403962579, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1355612553833553, "3000–6000 Hz_10ms_envelope_correlation": 0.9870685016211376, "3000–6000 Hz_10ms_modulation_mae_db": 0.4145477520153438, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9098522749280504, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07639511283573382, "3000–6000 Hz_40ms_envelope_correlation": 0.9959029463099335, "3000–6000 Hz_40ms_modulation_mae_db": 0.2374401983880533, "mr_stft_db": 5.854859321577723, "band_mae_db": 0.19869102866698432, "rms_gain_error_db": -0.1641158006104136, "output_peak": 0.6212815809347643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.730810234189972, "spectral_convergence": 0.31231315015903616, "spectral_convergence_status": "ok", "amplitude_floor": 1.4064891899964367e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.619927221948608, "spectral_convergence": 0.2974340158664754, "spectral_convergence_status": "ok", "amplitude_floor": 1.2304331075199438e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.2008626629761245, "spectral_convergence": 0.3136983138155875, "spectral_convergence_status": "ok", "amplitude_floor": 7.896502575224385e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.444718915891811e-05, "output_power": 5.647892457267107e-05, "delta_db": 0.15910954449543643, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025153246625132694, "output_power": 0.00026579785009729195, "delta_db": 0.23957414786050635, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010938925867664089, "output_power": 0.001052761092864741, "delta_db": -0.1664485291449897, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031036369064007244, "output_power": 0.00030926085233727585, "delta_db": -0.01545959082873024, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001328835538454751, "output_power": 0.00012879380998954558, "delta_db": -0.13576243669696342, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6997042780401114e-05, "output_power": 3.581667614094748e-05, "delta_db": -0.14081731768677158, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14286192778556628, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07269182753774478, "pitch": null, "output_rms_dbfs": -27.330844151254524, "output_peak": 0.6081810744641095, "reference_rms_dbfs": -27.258152323716782, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8505333730382345}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019413100530357063, "log_rms_mae_db": 0.8592205783357554, "rms_envelope_nrmse": 0.08680853472409356, "envelope_correlation": 0.9934920731018162, "rms_gain_error_db": -0.07267025278861362, "reference_mean_power": 0.0018816420370153617, "output_mean_power": 0.001850418584698332}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1507655155627412e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005726194534917747, "output_power": 0.0005650128317124638, "reference_active": true, "delta_db": -0.058077871725653096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.421513884012437e-05, "output_power": 8.587501923040663e-05, "reference_active": true, "delta_db": -0.4025384513694747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.679657413963232e-05, "output_power": 1.6744449499057422e-05, "reference_active": true, "delta_db": -0.013498372763204747}], "active_band_count": 3, "active_band_mae_db": 0.15803823195277752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016416501002274435, "log_rms_mae_db": 0.4583074000477415, "rms_envelope_nrmse": 0.04091644126694648, "envelope_correlation": 0.9985248473414563, "rms_gain_error_db": -0.07268444952242095, "reference_mean_power": 0.001890759403997248, "output_mean_power": 0.0018593785824235736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0781635397104837e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005383866955860211, "output_power": 0.0005303796812757389, "reference_active": true, "delta_db": -0.06507441229889488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8874025426333776e-05, "output_power": 3.641257587298643e-05, "reference_active": true, "delta_db": -0.2840811181075856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650199092025111e-06, "output_power": 1.3732210769195547e-06, "reference_active": true, "delta_db": -0.5677940615477921}], "active_band_count": 3, "active_band_mae_db": 0.3056498639847575}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017990629858894173, "log_rms_mae_db": 1.436894235830512, "rms_envelope_nrmse": 0.11079474076061202, "envelope_correlation": 0.9904137713064451, "rms_gain_error_db": -0.10015315082602355, "reference_mean_power": 0.0012781810161111924, "output_mean_power": 0.0012490420167178004}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.355790711503392e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004472139003696596, "output_power": 0.0004371198153899213, "reference_active": true, "delta_db": -0.099147996458994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.621773891011176e-05, "output_power": 7.185821291842988e-05, "reference_active": true, "delta_db": -0.2557964847115152}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7010237353267386e-05, "output_power": 1.9102517355602468e-05, "reference_active": true, "delta_db": 0.5038022932757809}], "active_band_count": 3, "active_band_mae_db": 0.28624892481543}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000149706333097475, "log_rms_mae_db": 0.7878975273641718, "rms_envelope_nrmse": 0.05398845279438951, "envelope_correlation": 0.9976862404591822, "rms_gain_error_db": -0.10016708466125283, "reference_mean_power": 0.0012844845607787134, "output_mean_power": 0.0012551978308013982}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.840745806614025e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004236215690036061, "output_power": 0.0004146307858619718, "reference_active": true, "delta_db": -0.09316519597273978}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1600142098026355e-05, "output_power": 3.081294554919875e-05, "reference_active": true, "delta_db": -0.10955819057271834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5445822829049627e-06, "output_power": 1.5362446214138942e-06, "reference_active": true, "delta_db": -0.02350673588765204}], "active_band_count": 3, "active_band_mae_db": 0.07541004081103672}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010005798255669499, "log_rms_mae_db": 1.380317814945297, "rms_envelope_nrmse": 0.13652521540730472, "envelope_correlation": 0.9872407416473183, "rms_gain_error_db": -0.049811797826043044, "reference_mean_power": 0.0003605696849433608, "output_mean_power": 0.0003564577247052289}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5037729366163507e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001406674030835393, "output_power": 0.0001353319587407817, "reference_active": true, "delta_db": -0.16793102182702635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0544824834133773e-05, "output_power": 2.470744687366883e-05, "reference_active": true, "delta_db": 0.8012542740695389}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4458512826457347e-06, "output_power": 5.623344891688779e-06, "reference_active": true, "delta_db": 2.126981904586888}], "active_band_count": 3, "active_band_mae_db": 1.0320557334944844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.18211380942645e-05, "log_rms_mae_db": 0.8930529305824555, "rms_envelope_nrmse": 0.07364278335154713, "envelope_correlation": 0.9961795851484101, "rms_gain_error_db": -0.049812008331314286, "reference_mean_power": 0.00036238111197119544, "output_mean_power": 0.00035824847673639285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3815388264892723e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001355192968806531, "output_power": 0.00013261270818771872, "reference_active": true, "delta_db": -0.09415995513082329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.971075792205542e-06, "output_power": 1.0419978901259983e-05, "reference_active": true, "delta_db": 1.1634990099286806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.416322058192608e-07, "output_power": 3.3209806265021873e-07, "reference_active": true, "delta_db": 1.381115235613888}], "active_band_count": 3, "active_band_mae_db": 0.8795914002244639}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.283283642007837e-05, "log_rms_mae_db": 1.3436965467405824, "rms_envelope_nrmse": 0.13118619856564506, "envelope_correlation": 0.9884335404796162, "rms_gain_error_db": -0.30369713008889104, "reference_mean_power": 6.230386432542126e-05, "output_mean_power": 5.809587151737435e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0221136919924677e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.16042156208526e-05, "output_power": 1.1013928055739562e-05, "reference_active": true, "delta_db": -0.22673554648324376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3772446197384082e-06, "output_power": 2.241847775162532e-06, "reference_active": true, "delta_db": -0.25467753150023104}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4509999616936e-07, "output_power": 4.2739634798785933e-07, "reference_active": true, "delta_db": -0.1762678365622028}], "active_band_count": 3, "active_band_mae_db": 0.21922697151522585}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.758960619357764e-05, "log_rms_mae_db": 0.8700449138370119, "rms_envelope_nrmse": 0.08018148248693269, "envelope_correlation": 0.9961994423173449, "rms_gain_error_db": -0.3036983926299613, "reference_mean_power": 6.26170059672204e-05, "output_mean_power": 5.8387846649912615e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.846748801507289e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1232117533806203e-05, "output_power": 1.0635084482503845e-05, "reference_active": true, "delta_db": -0.23720695121516377}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.067581168241283e-06, "output_power": 1.07095186078026e-06, "reference_active": true, "delta_db": 0.013690453804040747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.579800269976513e-08, "output_power": 2.3922176225943687e-08, "reference_active": true, "delta_db": -0.3278539852154449}], "active_band_count": 3, "active_band_mae_db": 0.1929171300782165}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8444920144216925, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "62de862e1602bb6f5e99e221d2920fc255e128a1e47386c9f874da4e8c12b5f3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9855356242650748, "standard_deviation": 0.0013291696177913802, "minimum": 0.9836578894362825, "maximum": 0.9872407416473183, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4094647768941233, "standard_deviation": 0.03243147721250334, "minimum": 1.3577723411436684, "maximum": 1.4560283683471589, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9999420196734783, "standard_deviation": 0.24947331210776613, "minimum": 0.6704829950149828, "maximum": 1.4020565055077807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1470628334838839, "standard_deviation": 0.00794127608339119, "minimum": 0.13652521540730472, "maximum": 0.15899899319352365, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950177947062655, "standard_deviation": 0.0011665188114909383, "minimum": 0.9924591995228202, "maximum": 0.9961795851484101, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9191475834676879, "standard_deviation": 0.02261967551593387, "minimum": 0.8926528975674032, "maximum": 0.9616930823961616, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7802875416164039, "standard_deviation": 0.2876354114924978, "minimum": 0.42047406964643375, "maximum": 1.305725441153636, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08872329633175743, "standard_deviation": 0.012302264463487589, "minimum": 0.07364278335154713, "maximum": 0.11266205005911595, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9866838969856411, "standard_deviation": 0.0012003302527558444, "minimum": 0.985033347529991, "maximum": 0.9884335404796162, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3621426194636737, "standard_deviation": 0.02461360264891021, "minimum": 1.334231955486401, "maximum": 1.399045555836025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4527785865435984, "standard_deviation": 0.25324318508809374, "minimum": 0.21922697151522585, "maximum": 1.042299923676753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13890614754460442, "standard_deviation": 0.005990766443059624, "minimum": 0.13011173933795547, "maximum": 0.14707024777303215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954025194647591, "standard_deviation": 0.0007363466665245912, "minimum": 0.9941382761956787, "maximum": 0.9963843103258435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8855197073592735, "standard_deviation": 0.028907226299027897, "minimum": 0.8268593061670426, "maximum": 0.9110515146550692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.41398133331709047, "standard_deviation": 0.30096425732402976, "minimum": 0.17199596393357033, "maximum": 1.0435520943033314, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08386863871584029, "standard_deviation": 0.006424929487199576, "minimum": 0.07437603080808018, "maximum": 0.09276435233317172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9892946955073436, "standard_deviation": 0.0016677411007474537, "minimum": 0.9865209166321742, "maximum": 0.9906705539923831, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4260277038194984, "standard_deviation": 0.025272633898276133, "minimum": 1.3967428095846057, "maximum": 1.4711959509958175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3703827093934493, "standard_deviation": 0.1569643783930394, "minimum": 0.21184556013418657, "maximum": 0.6981302444554046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11736873278395149, "standard_deviation": 0.008674370299141356, "minimum": 0.11051567294394715, "maximum": 0.13205294379683175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996907507203034, "standard_deviation": 0.0007734263143840519, "minimum": 0.9955385971509071, "maximum": 0.9976862404591822, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7925508787474688, "standard_deviation": 0.02694766705774935, "minimum": 0.7559604477962697, "maximum": 0.8273214891397515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.35031585215422356, "standard_deviation": 0.28399413627662556, "minimum": 0.07541004081103672, "maximum": 0.9644645717157886, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06339823626524897, "standard_deviation": 0.0070404082198607726, "minimum": 0.05398845279438951, "maximum": 0.07456819616571302, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.198921846734103, "standard_deviation": 0.042515402303859834, "minimum": 0.14286192778556628, "maximum": 0.2599546247700557, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.848203425581829, "standard_deviation": 0.0048574707278395525, "minimum": 5.839860680728173, "maximum": 5.854859321577723, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6137971380306402, "standard_deviation": 0.02792824198790552, "minimum": 0.5887304222111513, "maximum": 0.6789912325355518, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1469676964813014, "standard_deviation": 0.05202944119931246, "minimum": -0.24259439982153636, "maximum": -0.07269182753774478, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9926845885174469, "standard_deviation": 0.001436198771988205, "minimum": 0.9898959621938664, "maximum": 0.9940452172358683, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8719637447971389, "standard_deviation": 0.02150317890465215, "minimum": 0.8350437855474186, "maximum": 0.9021926171581918, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.44573235713015597, "standard_deviation": 0.14186423889813476, "minimum": 0.15803823195277752, "maximum": 0.5618637790761672, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09257218827747216, "standard_deviation": 0.008384664144751398, "minimum": 0.08374212854084502, "maximum": 0.10839576292070462, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980474765835423, "standard_deviation": 0.0006489456390209301, "minimum": 0.9966850320323467, "maximum": 0.998556792792993, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4762208833256748, "standard_deviation": 0.02041695274855503, "minimum": 0.4583074000477415, "maximum": 0.5166346443400809, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4401621978463595, "standard_deviation": 0.21671343389562783, "minimum": 0.23595950929444767, "maximum": 0.8567010699374755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04877598199295011, "standard_deviation": 0.006864125407026335, "minimum": 0.04091644126694648, "maximum": 0.06199950098680363, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_2961-961-0011", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.2443196773529053, "predicted_mos": 2.2443196773529053, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0c40cd7b39ec3b8aff546f2b3e85c0ed37327c7163731e539d5daeb6e85b603f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.114059300161898, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8496965839373872, "全帯域_10ms_rms_envelope_nrmse": 0.09572912366365702, "全帯域_10ms_envelope_correlation": 0.9921873102378956, "全帯域_10ms_modulation_mae_db": 0.7182340821823572, "全帯域_40ms_log_rms_mae_db": 0.44191411109974316, "全帯域_40ms_rms_envelope_nrmse": 0.05581176832798326, "全帯域_40ms_envelope_correlation": 0.9973825036071993, "全帯域_40ms_modulation_mae_db": 0.9582127030909922, "300–1000 Hz_10ms_log_rms_mae_db": 1.4161946667476752, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11682323877248263, "300–1000 Hz_10ms_envelope_correlation": 0.9893982091872361, "300–1000 Hz_10ms_modulation_mae_db": 0.5383929552656811, "300–1000 Hz_40ms_log_rms_mae_db": 0.7595142304020968, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06939809130089966, "300–1000 Hz_40ms_envelope_correlation": 0.9962469277341277, "300–1000 Hz_40ms_modulation_mae_db": 0.7579555633573776, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4095546154180014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14401729107678746, "1000–3000 Hz_10ms_envelope_correlation": 0.9860392793517448, "1000–3000 Hz_10ms_modulation_mae_db": 0.9570235081648607, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9068394096832815, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08292616667575477, "1000–3000 Hz_40ms_envelope_correlation": 0.9956427828588001, "1000–3000 Hz_40ms_modulation_mae_db": 0.6744353102998945, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2917141958512786, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14129133939036478, "3000–6000 Hz_10ms_envelope_correlation": 0.9858924924750861, "3000–6000 Hz_10ms_modulation_mae_db": 0.13066518590730536, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8275699801856411, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07870285598561073, "3000–6000 Hz_40ms_envelope_correlation": 0.995506113434856, "3000–6000 Hz_40ms_modulation_mae_db": 0.13482430593673497, "mr_stft_db": 5.8189738524071375, "band_mae_db": 0.19781057202286223, "rms_gain_error_db": -0.14416262275488506, "output_peak": 0.5976494669600574}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8708955072195769, "全帯域_10ms_rms_envelope_nrmse": 0.08724061151465086, "全帯域_10ms_envelope_correlation": 0.9935161591190309, "全帯域_10ms_modulation_mae_db": 0.5103182880899403, "全帯域_40ms_log_rms_mae_db": 0.45494739213865787, "全帯域_40ms_rms_envelope_nrmse": 0.049554804925145066, "全帯域_40ms_envelope_correlation": 0.9979344743525317, "全帯域_40ms_modulation_mae_db": 0.8883429049414319, "300–1000 Hz_10ms_log_rms_mae_db": 1.456197465602543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1144110076468035, "300–1000 Hz_10ms_envelope_correlation": 0.9898935249713368, "300–1000 Hz_10ms_modulation_mae_db": 0.437813796948152, "300–1000 Hz_40ms_log_rms_mae_db": 0.821094258651779, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06440891673075327, "300–1000 Hz_40ms_envelope_correlation": 0.9968613002129832, "300–1000 Hz_40ms_modulation_mae_db": 0.8052351521321248, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3459291094294863, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12472945849557587, "1000–3000 Hz_10ms_envelope_correlation": 0.9893726600676419, "1000–3000 Hz_10ms_modulation_mae_db": 0.7236231974414817, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8371610067465566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07480825032952002, "1000–3000 Hz_40ms_envelope_correlation": 0.9961295096175284, "1000–3000 Hz_40ms_modulation_mae_db": 0.20123978256419275, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3272407305616518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13299151002930343, "3000–6000 Hz_10ms_envelope_correlation": 0.9876869036966798, "3000–6000 Hz_10ms_modulation_mae_db": 0.3498543122928382, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8903821479993496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07777155782237709, "3000–6000 Hz_40ms_envelope_correlation": 0.995951936391308, "3000–6000 Hz_40ms_modulation_mae_db": 0.2286085829415249, "mr_stft_db": 5.82036070190781, "band_mae_db": 0.1535946302448614, "rms_gain_error_db": -0.13122810129770593, "output_peak": 0.6112915598248643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9012504840363036, "全帯域_10ms_rms_envelope_nrmse": 0.10307831827989591, "全帯域_10ms_envelope_correlation": 0.9911952223219499, "全帯域_10ms_modulation_mae_db": 0.33647327258157117, "全帯域_40ms_log_rms_mae_db": 0.5085786337879824, "全帯域_40ms_rms_envelope_nrmse": 0.058433665849173574, "全帯域_40ms_envelope_correlation": 0.9974773558078138, "全帯域_40ms_modulation_mae_db": 0.35117825550509135, "300–1000 Hz_10ms_log_rms_mae_db": 1.4387761835747768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13139861645572345, "300–1000 Hz_10ms_envelope_correlation": 0.9867579619455962, "300–1000 Hz_10ms_modulation_mae_db": 0.3794403076011763, "300–1000 Hz_40ms_log_rms_mae_db": 0.8077531049714002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07743949585997856, "300–1000 Hz_40ms_envelope_correlation": 0.9955804157623634, "300–1000 Hz_40ms_modulation_mae_db": 0.2702819034175409, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4213426748648041, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14122054007336515, "1000–3000 Hz_10ms_envelope_correlation": 0.9876589447228655, "1000–3000 Hz_10ms_modulation_mae_db": 1.0543963312522202, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8999935991413829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08656062207732518, "1000–3000 Hz_40ms_envelope_correlation": 0.9965541864908839, "1000–3000 Hz_40ms_modulation_mae_db": 0.7153447770855643, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2810998841838983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13487166110022275, "3000–6000 Hz_10ms_envelope_correlation": 0.9875507908631083, "3000–6000 Hz_10ms_modulation_mae_db": 0.3090970745314415, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8280235005681512, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07803389606374894, "3000–6000 Hz_40ms_envelope_correlation": 0.9962296942534772, "3000–6000 Hz_40ms_modulation_mae_db": 0.1657651086401288, "mr_stft_db": 5.834177718920067, "band_mae_db": 0.28128012848608336, "rms_gain_error_db": -0.23653379744748734, "output_peak": 0.6214194402123866}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8280930989687321, "全帯域_10ms_rms_envelope_nrmse": 0.09501334484313743, "全帯域_10ms_envelope_correlation": 0.9925499469960468, "全帯域_10ms_modulation_mae_db": 0.6033783142351733, "全帯域_40ms_log_rms_mae_db": 0.42675693741739135, "全帯域_40ms_rms_envelope_nrmse": 0.055275785077230875, "全帯域_40ms_envelope_correlation": 0.9977557474819724, "全帯域_40ms_modulation_mae_db": 0.5843088225908102, "300–1000 Hz_10ms_log_rms_mae_db": 1.3734245818857964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11423534680989021, "300–1000 Hz_10ms_envelope_correlation": 0.9900009785106259, "300–1000 Hz_10ms_modulation_mae_db": 0.5132299018371698, "300–1000 Hz_40ms_log_rms_mae_db": 0.766281654844747, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06645649694403401, "300–1000 Hz_40ms_envelope_correlation": 0.9967675918976043, "300–1000 Hz_40ms_modulation_mae_db": 0.8292809544858613, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4042077350363766, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14900422964472143, "1000–3000 Hz_10ms_envelope_correlation": 0.9855273667064588, "1000–3000 Hz_10ms_modulation_mae_db": 1.1275578562605946, "1000–3000 Hz_40ms_log_rms_mae_db": 0.900454914162317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08783992945327221, "1000–3000 Hz_40ms_envelope_correlation": 0.9957582740158087, "1000–3000 Hz_40ms_modulation_mae_db": 0.9929268352907886, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3501405665525905, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13563138363064547, "3000–6000 Hz_10ms_envelope_correlation": 0.987077163170095, "3000–6000 Hz_10ms_modulation_mae_db": 0.33939437862400323, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8563799922202907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07809601050753433, "3000–6000 Hz_40ms_envelope_correlation": 0.9957471395834065, "3000–6000 Hz_40ms_modulation_mae_db": 0.24134909671920193, "mr_stft_db": 5.829817120546675, "band_mae_db": 0.24310211477911234, "rms_gain_error_db": -0.21438952911598838, "output_peak": 0.5720664213200506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8869131283274565, "全帯域_10ms_rms_envelope_nrmse": 0.09914918048193885, "全帯域_10ms_envelope_correlation": 0.991546469412269, "全帯域_10ms_modulation_mae_db": 0.4329679673040667, "全帯域_40ms_log_rms_mae_db": 0.49982325715439746, "全帯域_40ms_rms_envelope_nrmse": 0.054279218165897945, "全帯域_40ms_envelope_correlation": 0.997454119896107, "全帯域_40ms_modulation_mae_db": 0.6924378996212847, "300–1000 Hz_10ms_log_rms_mae_db": 1.4068660348676818, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1247043691604783, "300–1000 Hz_10ms_envelope_correlation": 0.9878609748122231, "300–1000 Hz_10ms_modulation_mae_db": 0.4949888417980158, "300–1000 Hz_40ms_log_rms_mae_db": 0.7947171125111528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07344965179919385, "300–1000 Hz_40ms_envelope_correlation": 0.9957017727645836, "300–1000 Hz_40ms_modulation_mae_db": 0.7526017724592377, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3463990213109336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14290344217625345, "1000–3000 Hz_10ms_envelope_correlation": 0.9860199240602802, "1000–3000 Hz_10ms_modulation_mae_db": 0.8387811502882334, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8695486030074934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08142024660463594, "1000–3000 Hz_40ms_envelope_correlation": 0.995386020898959, "1000–3000 Hz_40ms_modulation_mae_db": 0.740071772728356, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3534655539728893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13966329886208617, "3000–6000 Hz_10ms_envelope_correlation": 0.9869570369711671, "3000–6000 Hz_10ms_modulation_mae_db": 0.2128201417627903, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8590975206043907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09152758669269825, "3000–6000 Hz_40ms_envelope_correlation": 0.9949509530441477, "3000–6000 Hz_40ms_modulation_mae_db": 0.2916286759314302, "mr_stft_db": 5.818640360115307, "band_mae_db": 0.1694291699589041, "rms_gain_error_db": -0.12169010795568098, "output_peak": 0.5806004526377182}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8831418810694593, "全帯域_10ms_rms_envelope_nrmse": 0.09367947338606054, "全帯域_10ms_envelope_correlation": 0.9924154776768754, "全帯域_10ms_modulation_mae_db": 0.43205659828142934, "全帯域_40ms_log_rms_mae_db": 0.4658619244769783, "全帯域_40ms_rms_envelope_nrmse": 0.045335133306840095, "全帯域_40ms_envelope_correlation": 0.9981826254065694, "全帯域_40ms_modulation_mae_db": 0.21965195633678433, "300–1000 Hz_10ms_log_rms_mae_db": 1.4668886650533564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11776445300269456, "300–1000 Hz_10ms_envelope_correlation": 0.989157911838092, "300–1000 Hz_10ms_modulation_mae_db": 0.3271495001060021, "300–1000 Hz_40ms_log_rms_mae_db": 0.7675993590998279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05744962154683554, "300–1000 Hz_40ms_envelope_correlation": 0.997346179474212, "300–1000 Hz_40ms_modulation_mae_db": 0.16416033304521968, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3908970749437655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1355453239805719, "1000–3000 Hz_10ms_envelope_correlation": 0.9874789435090706, "1000–3000 Hz_10ms_modulation_mae_db": 0.8532118826643037, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8860843059408692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07465116534824945, "1000–3000 Hz_40ms_envelope_correlation": 0.9962684493873871, "1000–3000 Hz_40ms_modulation_mae_db": 0.6658021475795721, "3000–6000 Hz_10ms_log_rms_mae_db": 1.331435412916949, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1448735743006593, "3000–6000 Hz_10ms_envelope_correlation": 0.9854125712729556, "3000–6000 Hz_10ms_modulation_mae_db": 0.4236344767950197, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8699909908283345, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08013401926848215, "3000–6000 Hz_40ms_envelope_correlation": 0.9958613344692071, "3000–6000 Hz_40ms_modulation_mae_db": 0.39111111010806354, "mr_stft_db": 5.8314891140264615, "band_mae_db": 0.14034883616333746, "rms_gain_error_db": -0.07843390085870378, "output_peak": 0.6068645415055787}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8951953455881064, "全帯域_10ms_rms_envelope_nrmse": 0.10022186579238215, "全帯域_10ms_envelope_correlation": 0.9913033251095222, "全帯域_10ms_modulation_mae_db": 0.3750841050543148, "全帯域_40ms_log_rms_mae_db": 0.47606476462724817, "全帯域_40ms_rms_envelope_nrmse": 0.048361604759473216, "全帯域_40ms_envelope_correlation": 0.9979085545547028, "全帯域_40ms_modulation_mae_db": 0.6368752214061012, "300–1000 Hz_10ms_log_rms_mae_db": 1.4694477006320643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12536990453651614, "300–1000 Hz_10ms_envelope_correlation": 0.9876871564312176, "300–1000 Hz_10ms_modulation_mae_db": 0.5948642381901862, "300–1000 Hz_40ms_log_rms_mae_db": 0.7863952168743747, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06388629741532051, "300–1000 Hz_40ms_envelope_correlation": 0.9967051097165682, "300–1000 Hz_40ms_modulation_mae_db": 0.8448370769763183, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3479854136554594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1357045327952624, "1000–3000 Hz_10ms_envelope_correlation": 0.9874113729368476, "1000–3000 Hz_10ms_modulation_mae_db": 0.5540843951417264, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8460736835830147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07430012365206251, "1000–3000 Hz_40ms_envelope_correlation": 0.9962342360614974, "1000–3000 Hz_40ms_modulation_mae_db": 0.5226976915878154, "3000–6000 Hz_10ms_log_rms_mae_db": 1.381692340809212, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14238899578521405, "3000–6000 Hz_10ms_envelope_correlation": 0.9856659890591672, "3000–6000 Hz_10ms_modulation_mae_db": 0.3839534036242602, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8930365193269145, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08061321551771604, "3000–6000 Hz_40ms_envelope_correlation": 0.9952861773656795, "3000–6000 Hz_40ms_modulation_mae_db": 0.24040566592640408, "mr_stft_db": 5.831646648523613, "band_mae_db": 0.19573198678352835, "rms_gain_error_db": -0.07017392995459279, "output_peak": 0.5970307415387753}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8779021744227622, "全帯域_10ms_rms_envelope_nrmse": 0.09555221111823763, "全帯域_10ms_envelope_correlation": 0.9923026982672865, "全帯域_10ms_modulation_mae_db": 0.4930562870282224, "全帯域_40ms_log_rms_mae_db": 0.4668477412351029, "全帯域_40ms_rms_envelope_nrmse": 0.05138657803384026, "全帯域_40ms_envelope_correlation": 0.9979269825422505, "全帯域_40ms_modulation_mae_db": 0.9023929737473866, "300–1000 Hz_10ms_log_rms_mae_db": 1.3873145216675113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11697301973565966, "300–1000 Hz_10ms_envelope_correlation": 0.9895615098567996, "300–1000 Hz_10ms_modulation_mae_db": 0.48612863323599287, "300–1000 Hz_40ms_log_rms_mae_db": 0.7502099474397893, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06319092478380561, "300–1000 Hz_40ms_envelope_correlation": 0.9972080659101186, "300–1000 Hz_40ms_modulation_mae_db": 0.8626457300694464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.402655860275179, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12547708282300227, "1000–3000 Hz_10ms_envelope_correlation": 0.9892481869239318, "1000–3000 Hz_10ms_modulation_mae_db": 0.7088439316923988, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8850043968660887, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06644673384654604, "1000–3000 Hz_40ms_envelope_correlation": 0.9969401715622962, "1000–3000 Hz_40ms_modulation_mae_db": 0.5351029852310395, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3605838446708836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15024181169416248, "3000–6000 Hz_10ms_envelope_correlation": 0.9845764288155595, "3000–6000 Hz_10ms_modulation_mae_db": 0.301519123889955, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8968099890370577, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09605908717243168, "3000–6000 Hz_40ms_envelope_correlation": 0.9941364052120003, "3000–6000 Hz_40ms_modulation_mae_db": 0.2888225941436349, "mr_stft_db": 5.82420302466227, "band_mae_db": 0.13981374989018527, "rms_gain_error_db": -0.17818572071434874, "output_peak": 0.6353989862171155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7030889908553455, "spectral_convergence": 0.30803140689680236, "spectral_convergence_status": "ok", "amplitude_floor": 1.4064891899964367e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.565062683839144, "spectral_convergence": 0.2908968815944554, "spectral_convergence_status": "ok", "amplitude_floor": 1.2304331075199438e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.188769882526923, "spectral_convergence": 0.3107855859022919, "spectral_convergence_status": "ok", "amplitude_floor": 7.896502575224385e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.444718915891811e-05, "output_power": 5.894907217812129e-05, "delta_db": 0.3450150978033, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025153246625132694, "output_power": 0.0002646657283349859, "delta_db": 0.22103658936610993, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010938925867664089, "output_power": 0.0010357097282126444, "delta_db": -0.23736623652953634, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031036369064007244, "output_power": 0.00028972794842876254, "delta_db": -0.29880516394728523, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001328835538454751, "output_power": 0.00013215916797462427, "delta_db": -0.02373938568438251, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6997042780401114e-05, "output_power": 3.751950565653501e-05, "delta_db": 0.060900958806559186, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19781057202286223, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14416262275488506, "pitch": null, "output_rms_dbfs": -27.40231494647167, "output_peak": 0.5976494669600574, "reference_rms_dbfs": -27.258152323716782, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8189738524071375}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019413100530357063, "log_rms_mae_db": 0.8496965839373872, "rms_envelope_nrmse": 0.09572912366365702, "envelope_correlation": 0.9921873102378956, "rms_gain_error_db": -0.1441253752182485, "reference_mean_power": 0.0018816420370153617, "output_mean_power": 0.0018202224620661874}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1507655155627412e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005726194534917747, "output_power": 0.0005235747608981084, "reference_active": true, "delta_db": -0.38887395071989794}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.421513884012437e-05, "output_power": 7.388336488144092e-05, "reference_active": true, "delta_db": -1.0557402609927027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.679657413963232e-05, "output_power": 1.4263033681273535e-05, "reference_active": true, "delta_db": -0.7100880348344708}], "active_band_count": 3, "active_band_mae_db": 0.7182340821823572}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016416501002274435, "log_rms_mae_db": 0.44191411109974316, "rms_envelope_nrmse": 0.05581176832798326, "envelope_correlation": 0.9973825036071993, "rms_gain_error_db": -0.14419440451196114, "reference_mean_power": 0.001890759403997248, "output_mean_power": 0.0018290131530939338}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0781635397104837e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005383866955860211, "output_power": 0.00048740074779225556, "reference_active": true, "delta_db": -0.43208128059599477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8874025426333776e-05, "output_power": 3.207476256879097e-05, "reference_active": true, "delta_db": -0.8349606416402112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650199092025111e-06, "output_power": 1.0808371112020893e-06, "reference_active": true, "delta_db": -1.6075961870367703}], "active_band_count": 3, "active_band_mae_db": 0.9582127030909922}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017990629858894173, "log_rms_mae_db": 1.4161946667476752, "rms_envelope_nrmse": 0.11682323877248263, "envelope_correlation": 0.9893982091872361, "rms_gain_error_db": -0.1600554720380843, "reference_mean_power": 0.0012781810161111924, "output_mean_power": 0.0012319322253146934}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.355790711503392e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004472139003696596, "output_power": 0.00041164701739406157, "reference_active": true, "delta_db": -0.359903208346754}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.621773891011176e-05, "output_power": 6.219164552924906e-05, "reference_active": true, "delta_db": -0.8832401279810467}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7010237353267386e-05, "output_power": 1.561373559381237e-05, "reference_active": true, "delta_db": -0.37203552946924284}], "active_band_count": 3, "active_band_mae_db": 0.5383929552656811}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000149706333097475, "log_rms_mae_db": 0.7595142304020968, "rms_envelope_nrmse": 0.06939809130089966, "envelope_correlation": 0.9962469277341277, "rms_gain_error_db": -0.1601000204469549, "reference_mean_power": 0.0012844845607787134, "output_mean_power": 0.0012379949880304566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.840745806614025e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004236215690036061, "output_power": 0.0003855251804904187, "reference_active": true, "delta_db": -0.40925315261476436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1600142098026355e-05, "output_power": 2.7544436352058472e-05, "reference_active": true, "delta_db": -0.5965514580223918}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5445822829049627e-06, "output_power": 1.1534661152824764e-06, "reference_active": true, "delta_db": -1.2680620794349764}], "active_band_count": 3, "active_band_mae_db": 0.7579555633573776}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010005798255669499, "log_rms_mae_db": 1.4095546154180014, "rms_envelope_nrmse": 0.14401729107678746, "envelope_correlation": 0.9860392793517448, "rms_gain_error_db": -0.26168652496994105, "reference_mean_power": 0.0003605696849433608, "output_mean_power": 0.00033948497498473697}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5037729366163507e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001406674030835393, "output_power": 0.00012284438130645691, "reference_active": true, "delta_db": -0.5883817247291971}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0544824834133773e-05, "output_power": 1.722964318248751e-05, "reference_active": true, "delta_db": -0.7642615929022785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4458512826457347e-06, "output_power": 4.88809064352675e-06, "reference_active": true, "delta_db": 1.5184272068631064}], "active_band_count": 3, "active_band_mae_db": 0.9570235081648607}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.18211380942645e-05, "log_rms_mae_db": 0.9068394096832815, "rms_envelope_nrmse": 0.08292616667575477, "envelope_correlation": 0.9956427828588001, "rms_gain_error_db": -0.2616846200604942, "reference_mean_power": 0.00036238111197119544, "output_mean_power": 0.0003411906264730497}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3815388264892723e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001355192968806531, "output_power": 0.00011984677245978734, "reference_active": true, "delta_db": -0.5337479699243495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.971075792205542e-06, "output_power": 6.692703172174619e-06, "reference_active": true, "delta_db": -0.7591537454430687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.416322058192608e-07, "output_power": 2.8588754816396554e-07, "reference_active": true, "delta_db": 0.7304042155322649}], "active_band_count": 3, "active_band_mae_db": 0.6744353102998945}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.283283642007837e-05, "log_rms_mae_db": 1.2917141958512786, "rms_envelope_nrmse": 0.14129133939036478, "envelope_correlation": 0.9858924924750861, "rms_gain_error_db": -0.06998740468192972, "reference_mean_power": 6.230386432542126e-05, "output_mean_power": 6.130787223160517e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0221136919924677e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.16042156208526e-05, "output_power": 1.1434889292059487e-05, "reference_active": true, "delta_db": -0.0638382563988932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3772446197384082e-06, "output_power": 2.2657691739587653e-06, "reference_active": true, "delta_db": -0.2085820928531651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4509999616936e-07, "output_power": 4.575252967671229e-07, "reference_active": true, "delta_db": 0.11957520846985781}], "active_band_count": 3, "active_band_mae_db": 0.13066518590730536}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.758960619357764e-05, "log_rms_mae_db": 0.8275699801856411, "rms_envelope_nrmse": 0.07870285598561073, "envelope_correlation": 0.995506113434856, "rms_gain_error_db": -0.06998876405494327, "reference_mean_power": 6.26170059672204e-05, "output_mean_power": 6.161598869220803e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.846748801507289e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1232117533806203e-05, "output_power": 1.1024458708893855e-05, "reference_active": true, "delta_db": -0.08104364111873798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.067581168241283e-06, "output_power": 1.0854600756042974e-06, "reference_active": true, "delta_db": 0.07212949883236565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.579800269976513e-08, "output_power": 2.7334810134025208e-08, "reference_active": true, "delta_db": 0.25129977785910124}], "active_band_count": 3, "active_band_mae_db": 0.13482430593673497}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8476242688019808, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e66dfb5db722d312b05960f4c34a75bc90db92526c0ca7ab1d987fe9f573d030", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9873445847848552, "standard_deviation": 0.0014451374331372858, "minimum": 0.9855273667064588, "maximum": 0.9893726600676419, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3836214381167506, "standard_deviation": 0.031646727025280166, "minimum": 1.3459291094294863, "maximum": 1.4213426748648041, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8521902816132274, "standard_deviation": 0.19042442529991244, "minimum": 0.5540843951417264, "maximum": 1.1275578562605946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1373252376331925, "standard_deviation": 0.00872025050421846, "minimum": 0.12472945849557587, "maximum": 0.14900422964472143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961142038616451, "standard_deviation": 0.0005061641854115637, "minimum": 0.995386020898959, "maximum": 0.9969401715622962, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8788949898913755, "standard_deviation": 0.025873874876584782, "minimum": 0.8371610067465566, "maximum": 0.9068394096832815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.630952662795903, "standard_deviation": 0.22637484615481565, "minimum": 0.20123978256419275, "maximum": 0.9929268352907886, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07861915474842077, "standard_deviation": 0.007285757627202858, "minimum": 0.06644673384654604, "maximum": 0.08783992945327221, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9863524220404774, "standard_deviation": 0.0011230020282568848, "minimum": 0.9845764288155595, "maximum": 0.9876869036966798, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3346715661899191, "standard_deviation": 0.03434250626594639, "minimum": 1.2810998841838983, "maximum": 1.381692340809212, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.30636726217845167, "standard_deviation": 0.0944203084852416, "minimum": 0.13066518590730536, "maximum": 0.4236344767950197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1402441968490823, "standard_deviation": 0.005734123458932777, "minimum": 0.13299151002930343, "maximum": 0.15024181169416248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954587192192603, "standard_deviation": 0.0006677583773262693, "minimum": 0.9941364052120003, "maximum": 0.9962296942534772, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8651613300962663, "standard_deviation": 0.027630957577353468, "minimum": 0.8275699801856411, "maximum": 0.8968099890370577, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.24781439254339044, "standard_deviation": 0.07937227867565746, "minimum": 0.13482430593673497, "maximum": 0.39111111010806354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0826172786288249, "standard_deviation": 0.0070762341543096985, "minimum": 0.07777155782237709, "maximum": 0.09605908717243168, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9887897784441408, "standard_deviation": 0.0011948264254888184, "minimum": 0.9867579619455962, "maximum": 0.9900009785106259, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4268887275039257, "standard_deviation": 0.03655011965697245, "minimum": 1.3734245818857964, "maximum": 1.4694477006320643, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47150102187279697, "standard_deviation": 0.0867239661117274, "minimum": 0.3271495001060021, "maximum": 0.5948642381901862, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12020999451503106, "standard_deviation": 0.0062021055547307885, "minimum": 0.11423534680989021, "maximum": 0.13139861645572345, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965521704340702, "standard_deviation": 0.000653196943254698, "minimum": 0.9955804157623634, "maximum": 0.997346179474212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.781695610599396, "standard_deviation": 0.024907105248109955, "minimum": 0.7502099474397893, "maximum": 0.821094258651779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6608748107428908, "standard_deviation": 0.27797244373736635, "minimum": 0.16416033304521968, "maximum": 0.8626457300694464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06695993704760263, "standard_deviation": 0.006310333002020404, "minimum": 0.05744962154683554, "maximum": 0.07743949585997856, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19013889854110932, "standard_deviation": 0.05067827173430301, "minimum": 0.13981374989018527, "maximum": 0.28128012848608336, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.826163567638668, "standard_deviation": 0.006344945411617842, "minimum": 5.818640360115307, "maximum": 5.834177718920067, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6027902012770683, "standard_deviation": 0.020685360141728833, "minimum": 0.5720664213200506, "maximum": 0.6353989862171155, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14684971376242412, "standard_deviation": 0.059798431814564144, "minimum": -0.23653379744748734, "maximum": -0.07017392995459279, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9921270761426095, "standard_deviation": 0.0007662561562423799, "minimum": 0.9911952223219499, "maximum": 0.9935161591190309, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.874136025446223, "standard_deviation": 0.024392536444523615, "minimum": 0.8280930989687321, "maximum": 0.9012504840363036, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4876961143446344, "standard_deviation": 0.12462019930055099, "minimum": 0.33647327258157117, "maximum": 0.7182340821823572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09620801613499505, "standard_deviation": 0.004793701524216455, "minimum": 0.08724061151465086, "maximum": 0.10307831827989591, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977527954561434, "standard_deviation": 0.0002864965478008416, "minimum": 0.9973825036071993, "maximum": 0.9981826254065694, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4675993452421877, "standard_deviation": 0.027466433802324358, "minimum": 0.42675693741739135, "maximum": 0.5085786337879824, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6541750921549853, "standard_deviation": 0.2662296571217502, "minimum": 0.21965195633678433, "maximum": 0.9582127030909922, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05230481980569803, "standard_deviation": 0.0043934618611877994, "minimum": 0.045335133306840095, "maximum": 0.058433665849173574, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3570-5694-0001", "language": "en", "style": "normal", "speaker": "librispeech_3570", "source_utterance_id": "3570-5694-0001", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 90800, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3570-5694-0001.wav", "source_file_sha256": "617d0ed8ece557f0a17038f5cf5cdbaa783bb0ad496b59a55b4aaa08e74c3177", "decoded_pcm_sha256": "d7ebe73ddd2095245c2316a87024f8d15117a6bca41809d6c43a3d2cfe68c3a9", "original_source_sha256": "83d42febda01d467c35c0b90dc7bb5d5d80b18ed95d65748a831afca53f2e3cd", "corpus_entry_id": "librispeech_3570-5694-0001"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.253514289855957, "predicted_mos": 4.253514289855957, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "3309caddeb328bf21c4b7ef7b9f165e484371842cb10c51ada50dff7ee8991bc", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.303158465001616, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3570-5694-0001", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "3309caddeb328bf21c4b7ef7b9f165e484371842cb10c51ada50dff7ee8991bc", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "04fce52520379ca7b0df597b54831a312b6b920b3c63943448968dd0e39c4e24"}, "rows": [{"case": "librispeech_3570-5694-0001", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.1683268547058105, "predicted_mos": 3.1683268547058105, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "8aa787fc64ba9eda1a404b327bf2390fb9d4943337d1cba1d8f0ed613f10a697", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.793726011001127, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3570-5694-0001", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "8aa787fc64ba9eda1a404b327bf2390fb9d4943337d1cba1d8f0ed613f10a697", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0124148340769024, "全帯域_10ms_rms_envelope_nrmse": 0.0670814372448049, "全帯域_10ms_envelope_correlation": 0.9956701957902777, "全帯域_10ms_modulation_mae_db": 0.6116292552494023, "全帯域_40ms_log_rms_mae_db": 0.5385788945285266, "全帯域_40ms_rms_envelope_nrmse": 0.033413012996416365, "全帯域_40ms_envelope_correlation": 0.9988475506627414, "全帯域_40ms_modulation_mae_db": 0.7273433196665521, "300–1000 Hz_10ms_log_rms_mae_db": 1.634982758221589, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08958533282993314, "300–1000 Hz_10ms_envelope_correlation": 0.9938617777692327, "300–1000 Hz_10ms_modulation_mae_db": 0.9247277045458387, "300–1000 Hz_40ms_log_rms_mae_db": 1.0057660877903125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049793686855744326, "300–1000 Hz_40ms_envelope_correlation": 0.9980477213108356, "300–1000 Hz_40ms_modulation_mae_db": 1.1157549362367067, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8649367207959904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15757223687230132, "1000–3000 Hz_10ms_envelope_correlation": 0.9809942900897224, "1000–3000 Hz_10ms_modulation_mae_db": 0.7892737533098181, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0727804535133003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09450594889718439, "1000–3000 Hz_40ms_envelope_correlation": 0.993413623794148, "1000–3000 Hz_40ms_modulation_mae_db": 0.7107859473938106, "3000–6000 Hz_10ms_log_rms_mae_db": 1.745915961653821, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17794351172854297, "3000–6000 Hz_10ms_envelope_correlation": 0.9791876866047176, "3000–6000 Hz_10ms_modulation_mae_db": 1.369084490893248, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9904153707959601, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10031896704614932, "3000–6000 Hz_40ms_envelope_correlation": 0.9934719077933373, "3000–6000 Hz_40ms_modulation_mae_db": 1.5666245147078293, "mr_stft_db": 5.584284040456289, "band_mae_db": 0.18585436576034417, "rms_gain_error_db": -0.0653148976008988, "output_peak": 0.431569260663525}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0203290019836202, "全帯域_10ms_rms_envelope_nrmse": 0.06241063489114932, "全帯域_10ms_envelope_correlation": 0.9962539105236122, "全帯域_10ms_modulation_mae_db": 0.6326007576596168, "全帯域_40ms_log_rms_mae_db": 0.5409780358332051, "全帯域_40ms_rms_envelope_nrmse": 0.03625653060623051, "全帯域_40ms_envelope_correlation": 0.9986432956846916, "全帯域_40ms_modulation_mae_db": 0.7574314299992353, "300–1000 Hz_10ms_log_rms_mae_db": 1.6250886908076905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09444982328916884, "300–1000 Hz_10ms_envelope_correlation": 0.9931447904168615, "300–1000 Hz_10ms_modulation_mae_db": 0.7975628909780879, "300–1000 Hz_40ms_log_rms_mae_db": 1.0726202806539589, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05799827809357339, "300–1000 Hz_40ms_envelope_correlation": 0.997298520552086, "300–1000 Hz_40ms_modulation_mae_db": 0.9801600788409245, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7852287882399231, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1483912163336113, "1000–3000 Hz_10ms_envelope_correlation": 0.98284456922959, "1000–3000 Hz_10ms_modulation_mae_db": 0.9223697324893982, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9989503546349388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08333480182485617, "1000–3000 Hz_40ms_envelope_correlation": 0.9945626104043485, "1000–3000 Hz_40ms_modulation_mae_db": 1.2351864364584717, "3000–6000 Hz_10ms_log_rms_mae_db": 1.667421413990992, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.163906733128085, "3000–6000 Hz_10ms_envelope_correlation": 0.9821551822715628, "3000–6000 Hz_10ms_modulation_mae_db": 0.8050849768657669, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9557175340127358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09479031789204842, "3000–6000 Hz_40ms_envelope_correlation": 0.9935965234439035, "3000–6000 Hz_40ms_modulation_mae_db": 1.1402844894583426, "mr_stft_db": 5.566676075675453, "band_mae_db": 0.15395647165308968, "rms_gain_error_db": -0.0789819748380468, "output_peak": 0.42888409106005937}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9981139241897097, "全帯域_10ms_rms_envelope_nrmse": 0.0657826605690261, "全帯域_10ms_envelope_correlation": 0.9958311503428595, "全帯域_10ms_modulation_mae_db": 0.7695249215464403, "全帯域_40ms_log_rms_mae_db": 0.523392182572935, "全帯域_40ms_rms_envelope_nrmse": 0.03365365101302554, "全帯域_40ms_envelope_correlation": 0.9988381123666498, "全帯域_40ms_modulation_mae_db": 0.900704876860902, "300–1000 Hz_10ms_log_rms_mae_db": 1.6825141502943952, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0867023377917159, "300–1000 Hz_10ms_envelope_correlation": 0.9942471267496615, "300–1000 Hz_10ms_modulation_mae_db": 0.9165223256875533, "300–1000 Hz_40ms_log_rms_mae_db": 1.0203513607043335, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05036831298862538, "300–1000 Hz_40ms_envelope_correlation": 0.9979787294555057, "300–1000 Hz_40ms_modulation_mae_db": 1.122655050181253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8425507627750835, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15419441775851572, "1000–3000 Hz_10ms_envelope_correlation": 0.9817656616386226, "1000–3000 Hz_10ms_modulation_mae_db": 0.6460638818009067, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0770832203509173, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09219953238699084, "1000–3000 Hz_40ms_envelope_correlation": 0.9937191595452771, "1000–3000 Hz_40ms_modulation_mae_db": 0.7677732051280334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7727931321812882, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2020411075100499, "3000–6000 Hz_10ms_envelope_correlation": 0.9729367109597696, "3000–6000 Hz_10ms_modulation_mae_db": 1.580445418020803, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9842268572280726, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1056308396750136, "3000–6000 Hz_40ms_envelope_correlation": 0.9927789113692265, "3000–6000 Hz_40ms_modulation_mae_db": 1.8359614890712914, "mr_stft_db": 5.600015779818932, "band_mae_db": 0.19049604763081676, "rms_gain_error_db": -0.07646356589268885, "output_peak": 0.4219785124683386}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9573114981519882, "全帯域_10ms_rms_envelope_nrmse": 0.06471472617871812, "全帯域_10ms_envelope_correlation": 0.9959754713007403, "全帯域_10ms_modulation_mae_db": 0.5396985213315925, "全帯域_40ms_log_rms_mae_db": 0.5075616421271507, "全帯域_40ms_rms_envelope_nrmse": 0.03742856585618006, "全帯域_40ms_envelope_correlation": 0.998541047562407, "全帯域_40ms_modulation_mae_db": 0.550798601109081, "300–1000 Hz_10ms_log_rms_mae_db": 1.6438690886803617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0999285009567309, "300–1000 Hz_10ms_envelope_correlation": 0.9922891669040748, "300–1000 Hz_10ms_modulation_mae_db": 0.869370827331417, "300–1000 Hz_40ms_log_rms_mae_db": 1.1001229696510957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05848593730577601, "300–1000 Hz_40ms_envelope_correlation": 0.9972119809583113, "300–1000 Hz_40ms_modulation_mae_db": 0.9192970700834739, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8120954362924468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14849986653142386, "1000–3000 Hz_10ms_envelope_correlation": 0.9826342146152913, "1000–3000 Hz_10ms_modulation_mae_db": 0.5022149620755213, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0313333104655908, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08533173405769386, "1000–3000 Hz_40ms_envelope_correlation": 0.9940040628129965, "1000–3000 Hz_40ms_modulation_mae_db": 0.5968874384368505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7500371170242546, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18034015314977891, "3000–6000 Hz_10ms_envelope_correlation": 0.9784861208875643, "3000–6000 Hz_10ms_modulation_mae_db": 1.1508778566725777, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0594683206824005, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10609003906973936, "3000–6000 Hz_40ms_envelope_correlation": 0.9923292028924928, "3000–6000 Hz_40ms_modulation_mae_db": 1.3694123195071313, "mr_stft_db": 5.564217822922603, "band_mae_db": 0.18913927930211596, "rms_gain_error_db": -0.07347699013854624, "output_peak": 0.4203403366918643}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0327690981823932, "全帯域_10ms_rms_envelope_nrmse": 0.06991478389193753, "全帯域_10ms_envelope_correlation": 0.9952377080577077, "全帯域_10ms_modulation_mae_db": 0.41506187117401366, "全帯域_40ms_log_rms_mae_db": 0.5630515193562071, "全帯域_40ms_rms_envelope_nrmse": 0.03730341582417924, "全帯域_40ms_envelope_correlation": 0.9984870593495961, "全帯域_40ms_modulation_mae_db": 0.5793028496313647, "300–1000 Hz_10ms_log_rms_mae_db": 1.6765799067680323, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09439058240845326, "300–1000 Hz_10ms_envelope_correlation": 0.9931021962230439, "300–1000 Hz_10ms_modulation_mae_db": 0.5440412922231936, "300–1000 Hz_40ms_log_rms_mae_db": 1.0251401849352504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055094881384080814, "300–1000 Hz_40ms_envelope_correlation": 0.9974786508153056, "300–1000 Hz_40ms_modulation_mae_db": 0.7470161698895944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.758912505826384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15755180714083394, "1000–3000 Hz_10ms_envelope_correlation": 0.9802867483539298, "1000–3000 Hz_10ms_modulation_mae_db": 0.6425552736976062, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0337453647040942, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10484084136820111, "1000–3000 Hz_40ms_envelope_correlation": 0.990454715117096, "1000–3000 Hz_40ms_modulation_mae_db": 0.6300304014936077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7601541779430403, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20341484451931907, "3000–6000 Hz_10ms_envelope_correlation": 0.975120299838981, "3000–6000 Hz_10ms_modulation_mae_db": 1.8667442273353452, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9873895268916697, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.122993116387614, "3000–6000 Hz_40ms_envelope_correlation": 0.9933812492744696, "3000–6000 Hz_40ms_modulation_mae_db": 2.311948455971642, "mr_stft_db": 5.569973741776529, "band_mae_db": 0.2095590709720819, "rms_gain_error_db": -0.025764896830326785, "output_peak": 0.4397180773413571}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9831135562589306, "全帯域_10ms_rms_envelope_nrmse": 0.062254241722383676, "全帯域_10ms_envelope_correlation": 0.9962676864310053, "全帯域_10ms_modulation_mae_db": 0.40010275000084644, "全帯域_40ms_log_rms_mae_db": 0.5139857031547521, "全帯域_40ms_rms_envelope_nrmse": 0.03466254678458463, "全帯域_40ms_envelope_correlation": 0.9987403354709717, "全帯域_40ms_modulation_mae_db": 0.5883330929428257, "300–1000 Hz_10ms_log_rms_mae_db": 1.7113858176090608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09622706694383412, "300–1000 Hz_10ms_envelope_correlation": 0.9928664724638511, "300–1000 Hz_10ms_modulation_mae_db": 0.5205002117726218, "300–1000 Hz_40ms_log_rms_mae_db": 1.1389618233643184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05737931467195739, "300–1000 Hz_40ms_envelope_correlation": 0.9973405638919761, "300–1000 Hz_40ms_modulation_mae_db": 0.6846527623294284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8218990152222356, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14897669238879124, "1000–3000 Hz_10ms_envelope_correlation": 0.9827425290769671, "1000–3000 Hz_10ms_modulation_mae_db": 0.823151628585948, "1000–3000 Hz_40ms_log_rms_mae_db": 1.066264155947901, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09181062440703928, "1000–3000 Hz_40ms_envelope_correlation": 0.9932890845184797, "1000–3000 Hz_40ms_modulation_mae_db": 0.8382784478104858, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7091786878700812, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17663609903697156, "3000–6000 Hz_10ms_envelope_correlation": 0.9792207270095985, "3000–6000 Hz_10ms_modulation_mae_db": 0.2539116111042074, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9228567649851847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09375132059415459, "3000–6000 Hz_40ms_envelope_correlation": 0.9937840077065386, "3000–6000 Hz_40ms_modulation_mae_db": 0.3551815301038619, "mr_stft_db": 5.585661394481711, "band_mae_db": 0.1666472756752223, "rms_gain_error_db": -0.06239392608895854, "output_peak": 0.4389814703837134}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.995875825749391, "全帯域_10ms_rms_envelope_nrmse": 0.06674728450145555, "全帯域_10ms_envelope_correlation": 0.9958433828834643, "全帯域_10ms_modulation_mae_db": 0.7433878661192447, "全帯域_40ms_log_rms_mae_db": 0.5485010590241342, "全帯域_40ms_rms_envelope_nrmse": 0.03853398219831047, "全帯域_40ms_envelope_correlation": 0.998632306936115, "全帯域_40ms_modulation_mae_db": 0.967043925261137, "300–1000 Hz_10ms_log_rms_mae_db": 1.6118807423072612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0895310698093207, "300–1000 Hz_10ms_envelope_correlation": 0.9940177806884014, "300–1000 Hz_10ms_modulation_mae_db": 1.010198160920521, "300–1000 Hz_40ms_log_rms_mae_db": 1.054715459892259, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06010611663082528, "300–1000 Hz_40ms_envelope_correlation": 0.9972389429583189, "300–1000 Hz_40ms_modulation_mae_db": 1.3049741928854548, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8193395127178604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14374643179322968, "1000–3000 Hz_10ms_envelope_correlation": 0.9839034985329743, "1000–3000 Hz_10ms_modulation_mae_db": 0.6911822596590808, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0616073682662348, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08318351156021396, "1000–3000 Hz_40ms_envelope_correlation": 0.9945517389708342, "1000–3000 Hz_40ms_modulation_mae_db": 0.9889013987600023, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7868904907437577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1922140507611423, "3000–6000 Hz_10ms_envelope_correlation": 0.9768744879407603, "3000–6000 Hz_10ms_modulation_mae_db": 1.6071395499944083, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0115451809047316, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11787527691383164, "3000–6000 Hz_40ms_envelope_correlation": 0.9925486664240695, "3000–6000 Hz_40ms_modulation_mae_db": 1.797861777450248, "mr_stft_db": 5.589111215004557, "band_mae_db": 0.20803196197570317, "rms_gain_error_db": -0.13870665498295773, "output_peak": 0.41851224400591797}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9547341987900909, "全帯域_10ms_rms_envelope_nrmse": 0.0612277526154997, "全帯域_10ms_envelope_correlation": 0.9963627092289258, "全帯域_10ms_modulation_mae_db": 0.5327779361692871, "全帯域_40ms_log_rms_mae_db": 0.5028015984589284, "全帯域_40ms_rms_envelope_nrmse": 0.03125747395001395, "全帯域_40ms_envelope_correlation": 0.9989386936242355, "全帯域_40ms_modulation_mae_db": 0.587335105357829, "300–1000 Hz_10ms_log_rms_mae_db": 1.6842825784135669, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09316497169129809, "300–1000 Hz_10ms_envelope_correlation": 0.993214307061919, "300–1000 Hz_10ms_modulation_mae_db": 0.6797946213082048, "300–1000 Hz_40ms_log_rms_mae_db": 1.0369053862002602, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0567886064619842, "300–1000 Hz_40ms_envelope_correlation": 0.9972340147497635, "300–1000 Hz_40ms_modulation_mae_db": 0.8817990440824973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8377553079479179, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14652910392167393, "1000–3000 Hz_10ms_envelope_correlation": 0.9831855549763092, "1000–3000 Hz_10ms_modulation_mae_db": 0.9448801637687142, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0451218070985597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08635176558496978, "1000–3000 Hz_40ms_envelope_correlation": 0.9938769344299311, "1000–3000 Hz_40ms_modulation_mae_db": 0.9091513895580996, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6941598557501136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1814807266418449, "3000–6000 Hz_10ms_envelope_correlation": 0.9782302090156964, "3000–6000 Hz_10ms_modulation_mae_db": 1.0522298616617414, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9279378718404446, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09662418958323435, "3000–6000 Hz_40ms_envelope_correlation": 0.9937843551054729, "3000–6000 Hz_40ms_modulation_mae_db": 1.190470722960144, "mr_stft_db": 5.586073619526309, "band_mae_db": 0.15092388427608647, "rms_gain_error_db": -0.008586179037273664, "output_peak": 0.42067987328768447}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 90800, "output_samples": 90800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.897870085136182, "spectral_convergence": 0.17576333972160513, "spectral_convergence_status": "ok", "amplitude_floor": 1.4038593411999917e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.442906344052689, "spectral_convergence": 0.17225144019920643, "spectral_convergence_status": "ok", "amplitude_floor": 9.854110842773577e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.412075692179997, "spectral_convergence": 0.16601084035316818, "spectral_convergence_status": "ok", "amplitude_floor": 8.983114295610284e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0018717561089627995, "output_power": 0.0018823505802064511, "delta_db": 0.02451252985415112, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00201441754378434, "output_power": 0.0019918531316483355, "delta_db": -0.048921823294364813, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010569467904903444, "output_power": 0.0010109922850732758, "delta_db": -0.19305282836340573, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018132930505040446, "output_power": 0.00016875796727861835, "delta_db": -0.31203711525401795, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.873439242518033e-05, "output_power": 7.224616374948957e-05, "delta_db": -0.3734968935942539, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.113111558995142e-05, "output_power": 2.9983628174294578e-05, "delta_db": -0.16310500420187138, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18585436576034417, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0653148976008988, "pitch": null, "output_rms_dbfs": -22.876716136515615, "output_peak": 0.431569260663525, "reference_rms_dbfs": -22.81140123891472, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.584284040456289}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 90800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.0001853404348361204, "log_rms_mae_db": 1.0124148340769024, "rms_envelope_nrmse": 0.0670814372448049, "envelope_correlation": 0.9956701957902777, "rms_gain_error_db": -0.06531498504016522, "reference_mean_power": 0.005238930533969212, "output_mean_power": 0.005160730038633447}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 2.4623434734036186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012515931525511833, "output_power": 0.001222107081992633, "reference_active": true, "delta_db": -0.10353917660547812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003238300007472361, "output_power": 0.0002920897134488291, "reference_active": true, "delta_db": -0.4480081819261985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.578860153055394e-05, "output_power": 5.639879858907894e-05, "reference_active": true, "delta_db": -1.2833404072165302}], "active_band_count": 3, "active_band_mae_db": 0.6116292552494023}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00016148298564882675, "log_rms_mae_db": 0.5385788945285266, "rms_envelope_nrmse": 0.033413012996416365, "envelope_correlation": 0.9988475506627414, "rms_gain_error_db": -0.06531505140329334, "reference_mean_power": 0.005266794777733929, "output_mean_power": 0.005188178278981646}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.1218089587296202e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001164760758537026, "output_power": 0.0011338448820232582, "reference_active": true, "delta_db": -0.1168308651086534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015381285066774075, "output_power": 0.0001345254103339392, "reference_active": true, "delta_db": -0.58188295747046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.591871326154896e-06, "output_power": 3.9739788893769885e-06, "reference_active": true, "delta_db": -1.4833161364205427}], "active_band_count": 3, "active_band_mae_db": 0.7273433196665521}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.00014972999954250175, "log_rms_mae_db": 1.634982758221589, "rms_envelope_nrmse": 0.08958533282993314, "envelope_correlation": 0.9938617777692327, "rms_gain_error_db": -0.12506044950911188, "reference_mean_power": 0.0025001891623953804, "output_mean_power": 0.00242921986183327}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.4970716182960256e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007649864504896549, "output_power": 0.0007314420868806133, "reference_active": true, "delta_db": -0.19473797108855795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022996394368157333, "output_power": 0.00020145474395233485, "reference_active": true, "delta_db": -0.5748224899206114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.80930068112851e-05, "output_power": 3.031235406395081e-05, "reference_active": true, "delta_db": -2.004622652628347}], "active_band_count": 3, "active_band_mae_db": 0.9247277045458387}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00013636056906360855, "log_rms_mae_db": 1.0057660877903125, "rms_envelope_nrmse": 0.049793686855744326, "envelope_correlation": 0.9980477213108356, "rms_gain_error_db": -0.12506051212612607, "reference_mean_power": 0.0025134871346063777, "output_mean_power": 0.0024421403282798527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3161092414230216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215738253307795, "output_power": 0.0006884155327892643, "reference_active": true, "delta_db": -0.20430110244466898}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011225926953226224, "output_power": 9.373729079313434e-05, "reference_active": true, "delta_db": -0.783098146192188}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.841934016741924e-06, "output_power": 2.2313276431575593e-06, "reference_active": true, "delta_db": -2.3598655600732634}], "active_band_count": 3, "active_band_mae_db": 1.1157549362367067}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.2746042839695326e-05, "log_rms_mae_db": 1.8649367207959904, "rms_envelope_nrmse": 0.15757223687230132, "envelope_correlation": 0.9809942900897224, "rms_gain_error_db": -0.35796144846763756, "reference_mean_power": 0.00020292459461300377, "output_mean_power": 0.00018686955104198495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.5147206216944168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.937659859611645e-05, "output_power": 6.171014513412626e-05, "reference_active": true, "delta_db": -0.5085643556905007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3925889957061755e-05, "output_power": 2.058048778352445e-05, "reference_active": true, "delta_db": -0.6541243706668382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8985142545041605e-06, "output_power": 4.469195779062999e-06, "reference_active": true, "delta_db": -1.205132533572115}], "active_band_count": 3, "active_band_mae_db": 0.7892737533098181}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.512933910060942e-05, "log_rms_mae_db": 1.0727804535133003, "rms_envelope_nrmse": 0.09450594889718439, "envelope_correlation": 0.993413623794148, "rms_gain_error_db": -0.3579596119046664, "reference_mean_power": 0.00020400334148028423, "output_mean_power": 0.0001878630287639112}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3444198629453317e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.558296247449876e-05, "output_power": 5.8139664222840636e-05, "reference_active": true, "delta_db": -0.5231851123081998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3197503778987225e-05, "output_power": 1.1608194629646297e-05, "reference_active": true, "delta_db": -0.5572711390836159}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.991435827054748e-07, "output_power": 3.917737493075622e-07, "reference_active": true, "delta_db": -1.0519015907896165}], "active_band_count": 3, "active_band_mae_db": 0.7107859473938106}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.624711907970004e-05, "log_rms_mae_db": 1.745915961653821, "rms_envelope_nrmse": 0.17794351172854297, "envelope_correlation": 0.9791876866047176, "rms_gain_error_db": -0.3066259920307369, "reference_mean_power": 5.558522396873647e-05, "output_mean_power": 5.179606422372059e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 3.230968975305017e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3093182032279326e-05, "output_power": 1.1985143713206493e-05, "reference_active": true, "delta_db": -0.3840195983943979}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8826443460154885e-06, "output_power": 6.190196156450997e-06, "reference_active": true, "delta_db": -0.46050917118549023}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3224614880429907e-06, "output_power": 1.567427583743913e-06, "reference_active": true, "delta_db": -3.262724703099856}], "active_band_count": 3, "active_band_mae_db": 1.369084490893248}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.3621221947891856e-05, "log_rms_mae_db": 0.9904153707959601, "rms_envelope_nrmse": 0.10031896704614932, "envelope_correlation": 0.9934719077933373, "rms_gain_error_db": -0.3066268739952747, "reference_mean_power": 5.588085792782989e-05, "output_mean_power": 5.207153469307719e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.891660816172807e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3816996130233874e-05, "output_power": 1.2350650598237758e-05, "reference_active": true, "delta_db": -0.48723800322228095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.814539997745632e-06, "output_power": 3.1643772750168767e-06, "reference_active": true, "delta_db": -0.8115391612868088}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.118413559303936e-07, "output_power": 1.425030212953761e-07, "reference_active": true, "delta_db": -3.4010963796143976}], "active_band_count": 3, "active_band_mae_db": 1.5666245147078293}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9144876310602416, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c68a6f6e5105d7ed7ec4849d671a68963b8aff7101120ab960b2d68b2d05af65", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9822946333141758, "standard_deviation": 0.001195398907375146, "minimum": 0.9802867483539298, "maximum": 0.9839034985329743, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8178397562272302, "standard_deviation": 0.03340873977621629, "minimum": 1.758912505826384, "maximum": 1.8649367207959904, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7452114569233742, "standard_deviation": 0.15194161834118594, "minimum": 0.5022149620755213, "maximum": 0.9448801637687142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1506827215925476, "standard_deviation": 0.0051446617433066775, "minimum": 0.14374643179322968, "maximum": 0.15757223687230132, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934839911991389, "standard_deviation": 0.0013094787118809604, "minimum": 0.990454715117096, "maximum": 0.9945626104043485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.048360754372692, "standard_deviation": 0.026383783353902228, "minimum": 0.9989503546349388, "maximum": 1.0770832203509173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8346243331299201, "standard_deviation": 0.20976230539552806, "minimum": 0.5968874384368505, "maximum": 1.2351864364584717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09019484501089367, "standard_deviation": 0.0073129868574197425, "minimum": 0.08318351156021396, "maximum": 0.10484084136820111, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9777764280660813, "standard_deviation": 0.0028094761677942737, "minimum": 0.9729367109597696, "maximum": 0.9821551822715628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7358188546446685, "standard_deviation": 0.041395493700445964, "minimum": 1.667421413990992, "maximum": 1.7868904907437577, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2106897490685122, "standard_deviation": 0.5151996432697942, "minimum": 0.2539116111042074, "maximum": 1.8667442273353452, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18474715330946684, "standard_deviation": 0.01352147050141327, "minimum": 0.163906733128085, "maximum": 0.20341484451931907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932093530011887, "standard_deviation": 0.0005739201929613162, "minimum": 0.9923292028924928, "maximum": 0.9937843551054729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.97994467841765, "standard_deviation": 0.044811816005511755, "minimum": 0.9228567649851847, "maximum": 1.0594683206824005, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4459681624038114, "standard_deviation": 0.5849055429369214, "minimum": 0.3551815301038619, "maximum": 2.311948455971642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10475925839522315, "standard_deviation": 0.01078016797792918, "minimum": 0.09375132059415459, "maximum": 0.122993116387614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933429522846308, "standard_deviation": 0.0006539312512424615, "minimum": 0.9922891669040748, "maximum": 0.9942471267496615, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6588229666377448, "standard_deviation": 0.03468464879455998, "minimum": 1.6118807423072612, "maximum": 1.7113858176090608, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7828397543459298, "standard_deviation": 0.18278187751267314, "minimum": 0.5205002117726218, "maximum": 1.010198160920521, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09299746071505688, "standard_deviation": 0.004239860770741316, "minimum": 0.0867023377917159, "maximum": 0.0999285009567309, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974786405865129, "standard_deviation": 0.0003409763834399953, "minimum": 0.9972119809583113, "maximum": 0.9980477213108356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0568229441489736, "standard_deviation": 0.045060547114814435, "minimum": 1.0057660877903125, "maximum": 1.1389618233643184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9695386630661665, "standard_deviation": 0.20645038507876903, "minimum": 0.6846527623294284, "maximum": 1.3049741928854548, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05575189179907085, "standard_deviation": 0.0037818854376051206, "minimum": 0.049793686855744326, "maximum": 0.06010611663082528, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18182604465568256, "standard_deviation": 0.022560402155705286, "minimum": 0.15092388427608647, "maximum": 0.2095590709720819, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.580751711207798, "standard_deviation": 0.0124988473390328, "minimum": 5.564217822922603, "maximum": 5.600015779818932, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4275829832378075, "standard_deviation": 0.008523635761717592, "minimum": 0.41851224400591797, "maximum": 0.4397180773413571, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06621113567621217, "standard_deviation": 0.038819192261785596, "minimum": -0.13870665498295773, "maximum": -0.008586179037273664, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.995930276819824, "standard_deviation": 0.00037269186742107244, "minimum": 0.9952377080577077, "maximum": 0.9963627092289258, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9943327421728783, "standard_deviation": 0.028186335997876583, "minimum": 0.9547341987900909, "maximum": 1.0327690981823932, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5805979849063054, "standard_deviation": 0.13617290420730574, "minimum": 0.40010275000084644, "maximum": 0.7695249215464403, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06501669020187187, "standard_deviation": 0.0029448755872373686, "minimum": 0.0612277526154997, "maximum": 0.06991478389193753, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998708550207176, "standard_deviation": 0.00015913619517709493, "minimum": 0.9984870593495961, "maximum": 0.9989386936242355, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5298563293819799, "standard_deviation": 0.021287137611711703, "minimum": 0.5028015984589284, "maximum": 0.5630515193562071, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7072866501036159, "standard_deviation": 0.15910818686553807, "minimum": 0.550798601109081, "maximum": 0.967043925261137, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.035313647403617594, "standard_deviation": 0.002476637052265275, "minimum": 0.03125747395001395, "maximum": 0.03853398219831047, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0001", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.341106653213501, "predicted_mos": 3.341106653213501, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "d30e6fe7f1050c0283a5d9b4072e2720e3bd4957c2fa9cae54ffa42aa6315bf8", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.243517800234258, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0228838241260985, "全帯域_10ms_rms_envelope_nrmse": 0.06814671336339259, "全帯域_10ms_envelope_correlation": 0.9955252464496359, "全帯域_10ms_modulation_mae_db": 0.602014027506748, "全帯域_40ms_log_rms_mae_db": 0.5660030886407479, "全帯域_40ms_rms_envelope_nrmse": 0.03616310372359595, "全帯域_40ms_envelope_correlation": 0.9986368495234224, "全帯域_40ms_modulation_mae_db": 0.6895767553678186, "300–1000 Hz_10ms_log_rms_mae_db": 1.650814644248514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0961871434341305, "300–1000 Hz_10ms_envelope_correlation": 0.992849552877859, "300–1000 Hz_10ms_modulation_mae_db": 0.9706986679274442, "300–1000 Hz_40ms_log_rms_mae_db": 0.9787303924822924, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054502424282727516, "300–1000 Hz_40ms_envelope_correlation": 0.997569970327162, "300–1000 Hz_40ms_modulation_mae_db": 1.1931327480821752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.901864360339328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16676460690916226, "1000–3000 Hz_10ms_envelope_correlation": 0.9784733093752852, "1000–3000 Hz_10ms_modulation_mae_db": 0.7428635373974587, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0942786575835188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10369519250517865, "1000–3000 Hz_40ms_envelope_correlation": 0.9917376998493, "1000–3000 Hz_40ms_modulation_mae_db": 0.6463979560510991, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6509574799551552, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18223564509282306, "3000–6000 Hz_10ms_envelope_correlation": 0.9781749603226517, "3000–6000 Hz_10ms_modulation_mae_db": 1.387627131639274, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9715551746289597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10315326871874932, "3000–6000 Hz_40ms_envelope_correlation": 0.9932262910117984, "3000–6000 Hz_40ms_modulation_mae_db": 1.6244269483848257, "mr_stft_db": 5.463665072966125, "band_mae_db": 0.20412823897825716, "rms_gain_error_db": -0.06775452366821022, "output_peak": 0.4312950606373488}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0197169623566245, "全帯域_10ms_rms_envelope_nrmse": 0.06428117254447296, "全帯域_10ms_envelope_correlation": 0.996050699196798, "全帯域_10ms_modulation_mae_db": 0.65357149240315, "全帯域_40ms_log_rms_mae_db": 0.5458565314975449, "全帯域_40ms_rms_envelope_nrmse": 0.038387039488623764, "全帯域_40ms_envelope_correlation": 0.9985164455488558, "全帯域_40ms_modulation_mae_db": 0.7761415587279344, "300–1000 Hz_10ms_log_rms_mae_db": 1.6317846429799594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1021397959666903, "300–1000 Hz_10ms_envelope_correlation": 0.9919427011887981, "300–1000 Hz_10ms_modulation_mae_db": 0.8631012118139427, "300–1000 Hz_40ms_log_rms_mae_db": 1.0324682167413484, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061548160634433066, "300–1000 Hz_40ms_envelope_correlation": 0.9969374034141496, "300–1000 Hz_40ms_modulation_mae_db": 1.0052128306813806, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8125776081248757, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15065795673375043, "1000–3000 Hz_10ms_envelope_correlation": 0.9823930059803125, "1000–3000 Hz_10ms_modulation_mae_db": 0.8813449365305477, "1000–3000 Hz_40ms_log_rms_mae_db": 0.999071478797759, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0855391147847511, "1000–3000 Hz_40ms_envelope_correlation": 0.994440868070312, "1000–3000 Hz_40ms_modulation_mae_db": 1.181079585020702, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5842597453009908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1827930352706583, "3000–6000 Hz_10ms_envelope_correlation": 0.9777491449483334, "3000–6000 Hz_10ms_modulation_mae_db": 1.2265549807126457, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9041863729430056, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11677525537267214, "3000–6000 Hz_40ms_envelope_correlation": 0.9904481817975569, "3000–6000 Hz_40ms_modulation_mae_db": 1.6192867072755217, "mr_stft_db": 5.440475065069163, "band_mae_db": 0.16646790675453202, "rms_gain_error_db": -0.09799375018987919, "output_peak": 0.41888971558677185}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0161373390668957, "全帯域_10ms_rms_envelope_nrmse": 0.06858148503066866, "全帯域_10ms_envelope_correlation": 0.995446133429691, "全帯域_10ms_modulation_mae_db": 0.726581235909921, "全帯域_40ms_log_rms_mae_db": 0.538670890325515, "全帯域_40ms_rms_envelope_nrmse": 0.03600012132496553, "全帯域_40ms_envelope_correlation": 0.998647362303757, "全帯域_40ms_modulation_mae_db": 0.8819413035206042, "300–1000 Hz_10ms_log_rms_mae_db": 1.6915396420024038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08872216416385476, "300–1000 Hz_10ms_envelope_correlation": 0.9939334298965135, "300–1000 Hz_10ms_modulation_mae_db": 0.9646744964011779, "300–1000 Hz_40ms_log_rms_mae_db": 0.9790025889087145, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05054653240880001, "300–1000 Hz_40ms_envelope_correlation": 0.997922941215932, "300–1000 Hz_40ms_modulation_mae_db": 1.1433767788731977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.806284570465396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.155081132279759, "1000–3000 Hz_10ms_envelope_correlation": 0.98145790075106, "1000–3000 Hz_10ms_modulation_mae_db": 0.4815139597042406, "1000–3000 Hz_40ms_log_rms_mae_db": 1.031100271680053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08995609957886778, "1000–3000 Hz_40ms_envelope_correlation": 0.9940143525553218, "1000–3000 Hz_40ms_modulation_mae_db": 0.5044070021628332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.707481400951742, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2051162166589031, "3000–6000 Hz_10ms_envelope_correlation": 0.9721117013560873, "3000–6000 Hz_10ms_modulation_mae_db": 1.5976020092638274, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0022187592963132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10513277708066769, "3000–6000 Hz_40ms_envelope_correlation": 0.9930015384607213, "3000–6000 Hz_40ms_modulation_mae_db": 1.8202499308986235, "mr_stft_db": 5.470145721240965, "band_mae_db": 0.1800627764733167, "rms_gain_error_db": -0.07045357537869514, "output_peak": 0.4222430095268246}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.975204894145754, "全帯域_10ms_rms_envelope_nrmse": 0.06662990039896341, "全帯域_10ms_envelope_correlation": 0.995725803397522, "全帯域_10ms_modulation_mae_db": 0.5879064923788443, "全帯域_40ms_log_rms_mae_db": 0.5369291180779252, "全帯域_40ms_rms_envelope_nrmse": 0.038881013717123644, "全帯域_40ms_envelope_correlation": 0.9984141181683839, "全帯域_40ms_modulation_mae_db": 0.5978479237161921, "300–1000 Hz_10ms_log_rms_mae_db": 1.6680756497119948, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1056665479870289, "300–1000 Hz_10ms_envelope_correlation": 0.9913246726696208, "300–1000 Hz_10ms_modulation_mae_db": 0.9195122233943441, "300–1000 Hz_40ms_log_rms_mae_db": 1.0915247315752974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06179941570336841, "300–1000 Hz_40ms_envelope_correlation": 0.9968264272780372, "300–1000 Hz_40ms_modulation_mae_db": 1.0256892644050406, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7883419156553708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.147175935867516, "1000–3000 Hz_10ms_envelope_correlation": 0.9830978180921209, "1000–3000 Hz_10ms_modulation_mae_db": 0.6309680031586434, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0049495534788164, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08736770822954329, "1000–3000 Hz_40ms_envelope_correlation": 0.9939405356554709, "1000–3000 Hz_40ms_modulation_mae_db": 0.714795549547636, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6525594425369161, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19278599812827898, "3000–6000 Hz_10ms_envelope_correlation": 0.9752132271195405, "3000–6000 Hz_10ms_modulation_mae_db": 1.1603633959785589, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0131919019038178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11174367664168156, "3000–6000 Hz_40ms_envelope_correlation": 0.9912919865939638, "3000–6000 Hz_40ms_modulation_mae_db": 1.3839700700601292, "mr_stft_db": 5.4392372390163475, "band_mae_db": 0.1958148333581929, "rms_gain_error_db": -0.07053657261188512, "output_peak": 0.4193622408336199}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0357215213559032, "全帯域_10ms_rms_envelope_nrmse": 0.07119585575870617, "全帯域_10ms_envelope_correlation": 0.9950568420773178, "全帯域_10ms_modulation_mae_db": 0.36756534468658764, "全帯域_40ms_log_rms_mae_db": 0.595720442062001, "全帯域_40ms_rms_envelope_nrmse": 0.04101749526046705, "全帯域_40ms_envelope_correlation": 0.9981660438046527, "全帯域_40ms_modulation_mae_db": 0.4840243308387577, "300–1000 Hz_10ms_log_rms_mae_db": 1.6817764522028165, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0984543838938389, "300–1000 Hz_10ms_envelope_correlation": 0.9924633970443727, "300–1000 Hz_10ms_modulation_mae_db": 0.5473797963476433, "300–1000 Hz_40ms_log_rms_mae_db": 1.0093214925285066, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0584082914079142, "300–1000 Hz_40ms_envelope_correlation": 0.9971262603140023, "300–1000 Hz_40ms_modulation_mae_db": 0.7868374358580351, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7842030698308253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1605928527164445, "1000–3000 Hz_10ms_envelope_correlation": 0.9795442445855088, "1000–3000 Hz_10ms_modulation_mae_db": 0.6852523319095533, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0195209081837568, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10454691509618938, "1000–3000 Hz_40ms_envelope_correlation": 0.9906645495206975, "1000–3000 Hz_40ms_modulation_mae_db": 0.6344355138973174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6541371319470937, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20828954532386212, "3000–6000 Hz_10ms_envelope_correlation": 0.9737868926764389, "3000–6000 Hz_10ms_modulation_mae_db": 1.9057244450925594, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9565786028911809, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1238167525494245, "3000–6000 Hz_40ms_envelope_correlation": 0.9935413170882746, "3000–6000 Hz_40ms_modulation_mae_db": 2.2446351420106025, "mr_stft_db": 5.443442645919973, "band_mae_db": 0.24514354236340039, "rms_gain_error_db": -0.020755337673282692, "output_peak": 0.45052688345544245}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9922465460278885, "全帯域_10ms_rms_envelope_nrmse": 0.06631822294239702, "全帯域_10ms_envelope_correlation": 0.9957474647569144, "全帯域_10ms_modulation_mae_db": 0.381747688836289, "全帯域_40ms_log_rms_mae_db": 0.511429611337775, "全帯域_40ms_rms_envelope_nrmse": 0.03637042135514218, "全帯域_40ms_envelope_correlation": 0.9985939042016563, "全帯域_40ms_modulation_mae_db": 0.5095014137543449, "300–1000 Hz_10ms_log_rms_mae_db": 1.72328567871554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10273999334018538, "300–1000 Hz_10ms_envelope_correlation": 0.9917987257194337, "300–1000 Hz_10ms_modulation_mae_db": 0.49294208894448494, "300–1000 Hz_40ms_log_rms_mae_db": 1.1347156602890591, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06039801129981963, "300–1000 Hz_40ms_envelope_correlation": 0.9969818028497028, "300–1000 Hz_40ms_modulation_mae_db": 0.61429002271421, "1000–3000 Hz_10ms_log_rms_mae_db": 1.847775540354931, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14903916741549095, "1000–3000 Hz_10ms_envelope_correlation": 0.9827723571596149, "1000–3000 Hz_10ms_modulation_mae_db": 0.8401524389741354, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0609777767986612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08963507671434627, "1000–3000 Hz_40ms_envelope_correlation": 0.993757293745785, "1000–3000 Hz_40ms_modulation_mae_db": 0.8496894621967807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6293815392097317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17011001972587267, "3000–6000 Hz_10ms_envelope_correlation": 0.9808566605651491, "3000–6000 Hz_10ms_modulation_mae_db": 0.7650470331486501, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9204074790711165, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08468152382921804, "3000–6000 Hz_40ms_envelope_correlation": 0.9953261338235319, "3000–6000 Hz_40ms_modulation_mae_db": 0.964113990583091, "mr_stft_db": 5.449852414575619, "band_mae_db": 0.18551520051114015, "rms_gain_error_db": -0.0512133210759331, "output_peak": 0.42839610713890536}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0322015496206662, "全帯域_10ms_rms_envelope_nrmse": 0.07005160214320702, "全帯域_10ms_envelope_correlation": 0.9953692765052768, "全帯域_10ms_modulation_mae_db": 0.7784138770190651, "全帯域_40ms_log_rms_mae_db": 0.5957349437726805, "全帯域_40ms_rms_envelope_nrmse": 0.04206399642677659, "全帯域_40ms_envelope_correlation": 0.9982886691924935, "全帯域_40ms_modulation_mae_db": 1.008686315208781, "300–1000 Hz_10ms_log_rms_mae_db": 1.6538166316709366, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09238876436463032, "300–1000 Hz_10ms_envelope_correlation": 0.993568435966765, "300–1000 Hz_10ms_modulation_mae_db": 1.0609354526614239, "300–1000 Hz_40ms_log_rms_mae_db": 1.0482863518589927, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05958427416286908, "300–1000 Hz_40ms_envelope_correlation": 0.9972470877898091, "300–1000 Hz_40ms_modulation_mae_db": 1.3877819016870763, "1000–3000 Hz_10ms_log_rms_mae_db": 1.832188311262559, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15274085555764136, "1000–3000 Hz_10ms_envelope_correlation": 0.9815461374938615, "1000–3000 Hz_10ms_modulation_mae_db": 0.41503015477812194, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0566852375962263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09753827968522676, "1000–3000 Hz_40ms_envelope_correlation": 0.9919458927425815, "1000–3000 Hz_40ms_modulation_mae_db": 0.5846341251383984, "3000–6000 Hz_10ms_log_rms_mae_db": 1.699688423704991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20484823730734777, "3000–6000 Hz_10ms_envelope_correlation": 0.9745764367595943, "3000–6000 Hz_10ms_modulation_mae_db": 1.9166705118442284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.02379269172874, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1378008357572294, "3000–6000 Hz_40ms_envelope_correlation": 0.9904204104399656, "3000–6000 Hz_40ms_modulation_mae_db": 2.0879280813113597, "mr_stft_db": 5.4579274253487355, "band_mae_db": 0.21402817425316148, "rms_gain_error_db": -0.13092275004985307, "output_peak": 0.418496448152441}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9767338707490857, "全帯域_10ms_rms_envelope_nrmse": 0.06247697401125057, "全帯域_10ms_envelope_correlation": 0.9962044972970724, "全帯域_10ms_modulation_mae_db": 0.5862488733312784, "全帯域_40ms_log_rms_mae_db": 0.5142073608658552, "全帯域_40ms_rms_envelope_nrmse": 0.03219208907083214, "全帯域_40ms_envelope_correlation": 0.9988726232077261, "全帯域_40ms_modulation_mae_db": 0.6900185067785937, "300–1000 Hz_10ms_log_rms_mae_db": 1.6780564817845656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09819362229033655, "300–1000 Hz_10ms_envelope_correlation": 0.9924363570535706, "300–1000 Hz_10ms_modulation_mae_db": 0.7636572811933774, "300–1000 Hz_40ms_log_rms_mae_db": 1.0064143944875805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059407253957274364, "300–1000 Hz_40ms_envelope_correlation": 0.9969530343078512, "300–1000 Hz_40ms_modulation_mae_db": 1.0213492090737664, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8410920968950073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1433188248903873, "1000–3000 Hz_10ms_envelope_correlation": 0.9840661021323718, "1000–3000 Hz_10ms_modulation_mae_db": 1.0565559067092858, "1000–3000 Hz_40ms_log_rms_mae_db": 1.037344195384215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08348827424900682, "1000–3000 Hz_40ms_envelope_correlation": 0.9945580083085159, "1000–3000 Hz_40ms_modulation_mae_db": 1.0239920586377524, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6066997998703398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19730184988087301, "3000–6000 Hz_10ms_envelope_correlation": 0.9741308498009407, "3000–6000 Hz_10ms_modulation_mae_db": 1.0489991430376455, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9270919140507018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10689963143623137, "3000–6000 Hz_40ms_envelope_correlation": 0.992421356417828, "3000–6000 Hz_40ms_modulation_mae_db": 1.1485959452515637, "mr_stft_db": 5.453463972125718, "band_mae_db": 0.1610331787317759, "rms_gain_error_db": -0.015566049358865235, "output_peak": 0.41968324483888186}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 90800, "output_samples": 90800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.769697805864517, "spectral_convergence": 0.177187106655713, "spectral_convergence_status": "ok", "amplitude_floor": 1.4038593411999917e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.327561954560029, "spectral_convergence": 0.1726633994825634, "spectral_convergence_status": "ok", "amplitude_floor": 9.854110842773577e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.293735458473831, "spectral_convergence": 0.16563805032478465, "spectral_convergence_status": "ok", "amplitude_floor": 8.983114295610284e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0018717561089627995, "output_power": 0.001879662360203318, "delta_db": 0.01830585629558403, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00201441754378434, "output_power": 0.0019873307665928003, "delta_db": -0.05879339035238987, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010569467904903444, "output_power": 0.0010166664558901475, "delta_db": -0.16874629730962185, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018132930505040446, "output_power": 0.0001681832677301092, "delta_db": -0.3268521072331549, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.873439242518033e-05, "output_power": 7.224979901475945e-05, "delta_db": -0.37327837182837253, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.113111558995142e-05, "output_power": 2.9195461383370322e-05, "delta_db": -0.2787934108504198, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20412823897825716, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06775452366821022, "pitch": null, "output_rms_dbfs": -22.87915576258293, "output_peak": 0.4312950606373488, "reference_rms_dbfs": -22.81140123891472, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.463665072966125}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 90800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.0001853404348361204, "log_rms_mae_db": 1.0228838241260985, "rms_envelope_nrmse": 0.06814671336339259, "envelope_correlation": 0.9955252464496359, "rms_gain_error_db": -0.0677546194062078, "reference_mean_power": 0.005238930533969212, "output_mean_power": 0.00515783183032887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 2.4623434734036186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012515931525511833, "output_power": 0.0012202663109213408, "reference_active": true, "delta_db": -0.11008557014545257}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003238300007472361, "output_power": 0.00029473557295256936, "reference_active": true, "delta_db": -0.4088452489152704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.578860153055394e-05, "output_power": 5.6349850359504056e-05, "reference_active": true, "delta_db": -1.2871112634595212}], "active_band_count": 3, "active_band_mae_db": 0.602014027506748}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00016148298564882675, "log_rms_mae_db": 0.5660030886407479, "rms_envelope_nrmse": 0.03616310372359595, "envelope_correlation": 0.9986368495234224, "rms_gain_error_db": -0.06775473499150655, "reference_mean_power": 0.005266794777733929, "output_mean_power": 0.005185264597283223}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.1218089587296202e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001164760758537026, "output_power": 0.0011336903516891638, "reference_active": true, "delta_db": -0.1174228000915368}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015381285066774075, "output_power": 0.0001360244416004397, "reference_active": true, "delta_db": -0.5337566942589499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.591871326154896e-06, "output_power": 4.034614938313026e-06, "reference_active": true, "delta_db": -1.417550771752969}], "active_band_count": 3, "active_band_mae_db": 0.6895767553678186}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.00014972999954250175, "log_rms_mae_db": 1.650814644248514, "rms_envelope_nrmse": 0.0961871434341305, "envelope_correlation": 0.992849552877859, "rms_gain_error_db": -0.10847723149753914, "reference_mean_power": 0.0025001891623953804, "output_mean_power": 0.0024385133927282876}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.4970716182960256e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007649864504896549, "output_power": 0.0007315421316549087, "reference_active": true, "delta_db": -0.19414399484551975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022996394368157333, "output_power": 0.00020151828459285384, "reference_active": true, "delta_db": -0.5734529019914971}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.80930068112851e-05, "output_power": 2.9351616177879022e-05, "reference_active": true, "delta_db": -2.144499106945316}], "active_band_count": 3, "active_band_mae_db": 0.9706986679274442}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00013636056906360855, "log_rms_mae_db": 0.9787303924822924, "rms_envelope_nrmse": 0.054502424282727516, "envelope_correlation": 0.997569970327162, "rms_gain_error_db": -0.10847736514619245, "reference_mean_power": 0.0025134871346063777, "output_mean_power": 0.002451483249250678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3161092414230216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215738253307795, "output_power": 0.0006890004753344975, "reference_active": true, "delta_db": -0.20061249517717455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011225926953226224, "output_power": 9.360179891002265e-05, "reference_active": true, "delta_db": -0.7893801653405994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.841934016741924e-06, "output_power": 2.1164564838186796e-06, "reference_active": true, "delta_db": -2.5894055837287517}], "active_band_count": 3, "active_band_mae_db": 1.1931327480821752}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.2746042839695326e-05, "log_rms_mae_db": 1.901864360339328, "rms_envelope_nrmse": 0.16676460690916226, "envelope_correlation": 0.9784733093752852, "rms_gain_error_db": -0.3712201682044081, "reference_mean_power": 0.00020292459461300377, "output_mean_power": 0.0001862999207790586}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.5147206216944168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.937659859611645e-05, "output_power": 6.1002248469589796e-05, "reference_active": true, "delta_db": -0.5586716050938624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3925889957061755e-05, "output_power": 2.168743562092592e-05, "reference_active": true, "delta_db": -0.4265989800693562}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8985142545041605e-06, "output_power": 4.430070428106291e-06, "reference_active": true, "delta_db": -1.2433200270291573}], "active_band_count": 3, "active_band_mae_db": 0.7428635373974587}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.512933910060942e-05, "log_rms_mae_db": 1.0942786575835188, "rms_envelope_nrmse": 0.10369519250517865, "envelope_correlation": 0.9917376998493, "rms_gain_error_db": -0.37121810168828984, "reference_mean_power": 0.00020400334148028423, "output_mean_power": 0.00018729038002189438}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3444198629453317e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.558296247449876e-05, "output_power": 5.753955646085646e-05, "reference_active": true, "delta_db": -0.568245206256031}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3197503778987225e-05, "output_power": 1.2267632042887784e-05, "reference_active": true, "delta_db": -0.3173105381443716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.991435827054748e-07, "output_power": 3.91617129304039e-07, "reference_active": true, "delta_db": -1.0536381237528945}], "active_band_count": 3, "active_band_mae_db": 0.6463979560510991}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.624711907970004e-05, "log_rms_mae_db": 1.6509574799551552, "rms_envelope_nrmse": 0.18223564509282306, "envelope_correlation": 0.9781749603226517, "rms_gain_error_db": -0.336368050828025, "reference_mean_power": 5.558522396873647e-05, "output_mean_power": 5.144255787079408e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 3.230968975305017e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3093182032279326e-05, "output_power": 1.2170878423905693e-05, "reference_active": true, "delta_db": -0.31723281606226666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8826443460154885e-06, "output_power": 6.11609946639659e-06, "reference_active": true, "delta_db": -0.5128078863844904}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3224614880429907e-06, "output_power": 1.5423249576252302e-06, "reference_active": true, "delta_db": -3.3328406924710654}], "active_band_count": 3, "active_band_mae_db": 1.387627131639274}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.3621221947891856e-05, "log_rms_mae_db": 0.9715551746289597, "rms_envelope_nrmse": 0.10315326871874932, "envelope_correlation": 0.9932262910117984, "rms_gain_error_db": -0.3363688984215734, "reference_mean_power": 5.588085792782989e-05, "output_mean_power": 5.1716148672985345e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.891660816172807e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3816996130233874e-05, "output_power": 1.2586579392355102e-05, "reference_active": true, "delta_db": -0.405059164025684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.814539997745632e-06, "output_power": 3.145254159038746e-06, "reference_active": true, "delta_db": -0.8378642779347196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.118413559303936e-07, "output_power": 1.3517549538467563e-07, "reference_active": true, "delta_db": -3.6303574031940737}], "active_band_count": 3, "active_band_mae_db": 1.6244269483848257}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9155720978439438, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "784e954073e04bffa1ef245597edb64712c63e8874506c47a073f3438cb3c421", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.981668859446267, "standard_deviation": 0.001863785837067339, "minimum": 0.9784733093752852, "maximum": 0.9840661021323718, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8267909341160367, "standard_deviation": 0.038218564583319126, "minimum": 1.7842030698308253, "maximum": 1.901864360339328, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7167101586452483, "standard_deviation": 0.21152207827333258, "minimum": 0.41503015477812194, "maximum": 1.0565559067092858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15317141654626898, "standard_deviation": 0.007560208330768239, "minimum": 0.1433188248903873, "maximum": 0.16676460690916226, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9931324000559981, "standard_deviation": 0.0014642959035367556, "minimum": 0.9906645495206975, "maximum": 0.9945580083085159, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0379910099378757, "standard_deviation": 0.03171677772916545, "minimum": 0.999071478797759, "maximum": 1.0942786575835188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7674289065815649, "standard_deviation": 0.2334213822559156, "minimum": 0.5044070021628332, "maximum": 1.181079585020702, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09272083260538874, "standard_deviation": 0.008160351725840096, "minimum": 0.08348827424900682, "maximum": 0.10454691509618938, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.975824984193592, "standard_deviation": 0.002860770244479136, "minimum": 0.9721117013560873, "maximum": 0.9808566605651491, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.64814562043462, "standard_deviation": 0.042074651066056165, "minimum": 1.5842597453009908, "maximum": 1.707481400951742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3760735813396736, "standard_deviation": 0.4090409051209239, "minimum": 0.7650470331486501, "maximum": 1.9166705118442284, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19293506842357738, "standard_deviation": 0.013540818040557985, "minimum": 0.17011001972587267, "maximum": 0.20828954532386212, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992459651954205, "standard_deviation": 0.0016834285290149625, "minimum": 0.9904204104399656, "maximum": 0.9953261338235319, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9648778620642295, "standard_deviation": 0.04535866604017992, "minimum": 0.9041863729430056, "maximum": 1.02379269172874, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6116508519719646, "standard_deviation": 0.4402607333363337, "minimum": 0.964113990583091, "maximum": 2.2446351420106025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11125046517323425, "standard_deviation": 0.01568178842405673, "minimum": 0.08468152382921804, "maximum": 0.1378008357572294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925396590521167, "standard_deviation": 0.000886004941172524, "minimum": 0.9913246726696208, "maximum": 0.9939334298965135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6723937279145913, "standard_deviation": 0.02817082562396345, "minimum": 1.6317846429799594, "maximum": 1.72328567871554, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8228626523354798, "standard_deviation": 0.20627605781599123, "minimum": 0.49294208894448494, "maximum": 1.0609354526614239, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09806155193008695, "standard_deviation": 0.005599052251794552, "minimum": 0.08872216416385476, "maximum": 0.1056665479870289, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971956159370807, "standard_deviation": 0.00037495498412543977, "minimum": 0.9968264272780372, "maximum": 0.997922941215932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.035057978608974, "standard_deviation": 0.05490208132169479, "minimum": 0.9787303924822924, "maximum": 1.1347156602890591, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0222087739218604, "standard_deviation": 0.23893797498568412, "minimum": 0.61429002271421, "maximum": 1.3877819016870763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05827429548215078, "standard_deviation": 0.0038659274049956905, "minimum": 0.05054653240880001, "maximum": 0.06179941570336841, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1940242314279721, "standard_deviation": 0.027343647269575857, "minimum": 0.1610331787317759, "maximum": 0.24514354236340039, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.45227619453283, "standard_deviation": 0.011183923858069551, "minimum": 5.4392372390163475, "maximum": 5.470145721240965, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4261115887712795, "standard_deviation": 0.010955644643470798, "minimum": 0.418496448152441, "maximum": 0.45052688345544245, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06564948500082546, "standard_deviation": 0.03796687002616208, "minimum": -0.13092275004985307, "maximum": -0.015566049358865235, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956407453887786, "standard_deviation": 0.00037231984686916813, "minimum": 0.9950568420773178, "maximum": 0.9962044972970724, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0088558134311145, "standard_deviation": 0.02412654510299123, "minimum": 0.975204894145754, "maximum": 1.0357215213559032, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5855061290089854, "standard_deviation": 0.1468930564644763, "minimum": 0.36756534468658764, "maximum": 0.7784138770190651, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0672102407741323, "standard_deviation": 0.002899051536087406, "minimum": 0.06247697401125057, "maximum": 0.07119585575870617, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985170019938685, "standard_deviation": 0.0002234887494711837, "minimum": 0.9981660438046527, "maximum": 0.9988726232077261, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5505689983225056, "standard_deviation": 0.032777835688074665, "minimum": 0.511429611337775, "maximum": 0.5957349437726805, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7047172634891283, "standard_deviation": 0.180312394569135, "minimum": 0.4840243308387577, "maximum": 1.008686315208781, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037634410045940855, "standard_deviation": 0.0031450446235599928, "minimum": 0.03219208907083214, "maximum": 0.04206399642677659, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0001", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.152498245239258, "predicted_mos": 3.152498245239258, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "f29897f808c99e7393666449ffaedcc4e06afeb084438844c3ea753689bf099e", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.706440200097859, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9877692810456639, "全帯域_10ms_rms_envelope_nrmse": 0.06496028811727897, "全帯域_10ms_envelope_correlation": 0.9959168472591796, "全帯域_10ms_modulation_mae_db": 0.5672558785758671, "全帯域_40ms_log_rms_mae_db": 0.5175350893514982, "全帯域_40ms_rms_envelope_nrmse": 0.03231882054753892, "全帯域_40ms_envelope_correlation": 0.9989135565200024, "全帯域_40ms_modulation_mae_db": 0.8867669063890203, "300–1000 Hz_10ms_log_rms_mae_db": 1.8860752204713291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09221400444569795, "300–1000 Hz_10ms_envelope_correlation": 0.9934191909751878, "300–1000 Hz_10ms_modulation_mae_db": 0.6479262086551613, "300–1000 Hz_40ms_log_rms_mae_db": 1.2116201683865546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051003061213494856, "300–1000 Hz_40ms_envelope_correlation": 0.9978573357927735, "300–1000 Hz_40ms_modulation_mae_db": 0.9618240642492785, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1979750998160457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17571943121774122, "1000–3000 Hz_10ms_envelope_correlation": 0.9758064721625064, "1000–3000 Hz_10ms_modulation_mae_db": 1.3602360649063694, "1000–3000 Hz_40ms_log_rms_mae_db": 1.326671719290263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11284205131230551, "1000–3000 Hz_40ms_envelope_correlation": 0.9895277454203841, "1000–3000 Hz_40ms_modulation_mae_db": 1.651021180553898, "3000–6000 Hz_10ms_log_rms_mae_db": 2.408782698731656, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23399178079636299, "3000–6000 Hz_10ms_envelope_correlation": 0.9636095938780832, "3000–6000 Hz_10ms_modulation_mae_db": 1.6707262802801484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4761769902970072, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15040026063272544, "3000–6000 Hz_40ms_envelope_correlation": 0.9843573275586216, "3000–6000 Hz_40ms_modulation_mae_db": 1.997490848287453, "mr_stft_db": 6.130779673397015, "band_mae_db": 0.20990917445706817, "rms_gain_error_db": -0.06531039120598112, "output_peak": 0.4626668040552709}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9944002870505695, "全帯域_10ms_rms_envelope_nrmse": 0.06383955111317982, "全帯域_10ms_envelope_correlation": 0.9961064336737335, "全帯域_10ms_modulation_mae_db": 0.603615850774036, "全帯域_40ms_log_rms_mae_db": 0.5555779261780156, "全帯域_40ms_rms_envelope_nrmse": 0.03302809346841165, "全帯域_40ms_envelope_correlation": 0.998910882372305, "全帯域_40ms_modulation_mae_db": 0.6973406448808914, "300–1000 Hz_10ms_log_rms_mae_db": 1.8671274627812566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08480072370244035, "300–1000 Hz_10ms_envelope_correlation": 0.9945365014132208, "300–1000 Hz_10ms_modulation_mae_db": 0.5651784769347995, "300–1000 Hz_40ms_log_rms_mae_db": 1.2237191347388072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04937580584662438, "300–1000 Hz_40ms_envelope_correlation": 0.9980908502060736, "300–1000 Hz_40ms_modulation_mae_db": 0.5872051329925698, "1000–3000 Hz_10ms_log_rms_mae_db": 2.235269470011021, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17990967674096467, "1000–3000 Hz_10ms_envelope_correlation": 0.9744436027536327, "1000–3000 Hz_10ms_modulation_mae_db": 1.344451332738356, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3477456159273757, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11369988170017338, "1000–3000 Hz_40ms_envelope_correlation": 0.9891319697235362, "1000–3000 Hz_40ms_modulation_mae_db": 1.4933768659369708, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4257577064364715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23553136556356497, "3000–6000 Hz_10ms_envelope_correlation": 0.9632662514981796, "3000–6000 Hz_10ms_modulation_mae_db": 1.7635432960596367, "3000–6000 Hz_40ms_log_rms_mae_db": 1.442570295229962, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1385486810042561, "3000–6000 Hz_40ms_envelope_correlation": 0.986582738006171, "3000–6000 Hz_40ms_modulation_mae_db": 2.1229621511695447, "mr_stft_db": 6.100697845697174, "band_mae_db": 0.18087827278778354, "rms_gain_error_db": -0.0868892877683586, "output_peak": 0.4317468059658949}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9779546004981569, "全帯域_10ms_rms_envelope_nrmse": 0.06515224928366178, "全帯域_10ms_envelope_correlation": 0.9959475447983155, "全帯域_10ms_modulation_mae_db": 0.6982553155566534, "全帯域_40ms_log_rms_mae_db": 0.5425995392308872, "全帯域_40ms_rms_envelope_nrmse": 0.03619819046519988, "全帯域_40ms_envelope_correlation": 0.9986830141264903, "全帯域_40ms_modulation_mae_db": 0.7137062175243886, "300–1000 Hz_10ms_log_rms_mae_db": 1.8040893708507282, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09099576146736955, "300–1000 Hz_10ms_envelope_correlation": 0.9936731387267597, "300–1000 Hz_10ms_modulation_mae_db": 0.7514080809788991, "300–1000 Hz_40ms_log_rms_mae_db": 1.1554756272971611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05312932687085361, "300–1000 Hz_40ms_envelope_correlation": 0.997744703784561, "300–1000 Hz_40ms_modulation_mae_db": 0.8341417016431424, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1045353899216357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16045088765212573, "1000–3000 Hz_10ms_envelope_correlation": 0.9796658869788024, "1000–3000 Hz_10ms_modulation_mae_db": 1.0066491222759166, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2565112313000284, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0933978426378391, "1000–3000 Hz_40ms_envelope_correlation": 0.9925114603627737, "1000–3000 Hz_40ms_modulation_mae_db": 1.312810553398573, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4197623665266312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24309813908155836, "3000–6000 Hz_10ms_envelope_correlation": 0.9610585399984031, "3000–6000 Hz_10ms_modulation_mae_db": 1.3224782755792488, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4578552956123265, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1514330195013879, "3000–6000 Hz_40ms_envelope_correlation": 0.9837879136950937, "3000–6000 Hz_40ms_modulation_mae_db": 1.64105649011381, "mr_stft_db": 6.130239413841198, "band_mae_db": 0.14500690841865801, "rms_gain_error_db": -0.09362627628162085, "output_peak": 0.4503481031716475}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0843165000568216, "全帯域_10ms_rms_envelope_nrmse": 0.06940315728976107, "全帯域_10ms_envelope_correlation": 0.995407033404846, "全帯域_10ms_modulation_mae_db": 0.5997907512839661, "全帯域_40ms_log_rms_mae_db": 0.6167327882846747, "全帯域_40ms_rms_envelope_nrmse": 0.04002187027790642, "全帯域_40ms_envelope_correlation": 0.9983919744839439, "全帯域_40ms_modulation_mae_db": 0.6623298816256357, "300–1000 Hz_10ms_log_rms_mae_db": 1.8470048663738914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09441744429126107, "300–1000 Hz_10ms_envelope_correlation": 0.9934185323116336, "300–1000 Hz_10ms_modulation_mae_db": 0.8167226871218652, "300–1000 Hz_40ms_log_rms_mae_db": 1.2019900117192022, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05716785163791852, "300–1000 Hz_40ms_envelope_correlation": 0.9976992078139368, "300–1000 Hz_40ms_modulation_mae_db": 1.073301259025133, "1000–3000 Hz_10ms_log_rms_mae_db": 2.160276977831953, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1695840646892684, "1000–3000 Hz_10ms_envelope_correlation": 0.977504406845377, "1000–3000 Hz_10ms_modulation_mae_db": 1.1081889352926517, "1000–3000 Hz_40ms_log_rms_mae_db": 1.273873940654394, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10025582060445631, "1000–3000 Hz_40ms_envelope_correlation": 0.991887410365646, "1000–3000 Hz_40ms_modulation_mae_db": 1.3482323355031587, "3000–6000 Hz_10ms_log_rms_mae_db": 2.411385439488274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23907456784477588, "3000–6000 Hz_10ms_envelope_correlation": 0.962151815897504, "3000–6000 Hz_10ms_modulation_mae_db": 1.793114218939874, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4586139297308047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13568257878935933, "3000–6000 Hz_40ms_envelope_correlation": 0.9879713722150393, "3000–6000 Hz_40ms_modulation_mae_db": 2.3714045948214895, "mr_stft_db": 6.1137169034771235, "band_mae_db": 0.19739770404114518, "rms_gain_error_db": -0.10481854507959819, "output_peak": 0.4285812802276256}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9691478680693054, "全帯域_10ms_rms_envelope_nrmse": 0.061960723190561964, "全帯域_10ms_envelope_correlation": 0.9963344126902014, "全帯域_10ms_modulation_mae_db": 0.5348968428348196, "全帯域_40ms_log_rms_mae_db": 0.5189129693227467, "全帯域_40ms_rms_envelope_nrmse": 0.033629940793311504, "全帯域_40ms_envelope_correlation": 0.9988687079573305, "全帯域_40ms_modulation_mae_db": 0.6963670461338393, "300–1000 Hz_10ms_log_rms_mae_db": 1.8523023911050993, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09131548147724008, "300–1000 Hz_10ms_envelope_correlation": 0.9936551443669757, "300–1000 Hz_10ms_modulation_mae_db": 0.705250600637183, "300–1000 Hz_40ms_log_rms_mae_db": 1.1752889039270713, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05424800109571365, "300–1000 Hz_40ms_envelope_correlation": 0.9976864454882544, "300–1000 Hz_40ms_modulation_mae_db": 0.8762381022372884, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2020239879163945, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1663614437781136, "1000–3000 Hz_10ms_envelope_correlation": 0.9782611228893071, "1000–3000 Hz_10ms_modulation_mae_db": 1.106809223895053, "1000–3000 Hz_40ms_log_rms_mae_db": 1.265071716607009, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09794970007762574, "1000–3000 Hz_40ms_envelope_correlation": 0.9921168978536805, "1000–3000 Hz_40ms_modulation_mae_db": 1.3207501079350603, "3000–6000 Hz_10ms_log_rms_mae_db": 2.399685652603863, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24230941337981737, "3000–6000 Hz_10ms_envelope_correlation": 0.9618310459734948, "3000–6000 Hz_10ms_modulation_mae_db": 2.077139001683, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4869010551415174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1498516738881635, "3000–6000 Hz_40ms_envelope_correlation": 0.986243004068997, "3000–6000 Hz_40ms_modulation_mae_db": 2.509401895717931, "mr_stft_db": 6.121433439702767, "band_mae_db": 0.15036139552101016, "rms_gain_error_db": -0.08890233484829083, "output_peak": 0.4318761452563428}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0022619732664348, "全帯域_10ms_rms_envelope_nrmse": 0.0666313812773903, "全帯域_10ms_envelope_correlation": 0.9956919019433804, "全帯域_10ms_modulation_mae_db": 0.5019726429807481, "全帯域_40ms_log_rms_mae_db": 0.562355303483283, "全帯域_40ms_rms_envelope_nrmse": 0.03872996135996324, "全帯域_40ms_envelope_correlation": 0.9983993209089232, "全帯域_40ms_modulation_mae_db": 0.7875133761471185, "300–1000 Hz_10ms_log_rms_mae_db": 1.8106891572679737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0949433737427239, "300–1000 Hz_10ms_envelope_correlation": 0.993090123534695, "300–1000 Hz_10ms_modulation_mae_db": 0.9181426238953513, "300–1000 Hz_40ms_log_rms_mae_db": 1.1936174134233453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05733506668547977, "300–1000 Hz_40ms_envelope_correlation": 0.9973700822454284, "300–1000 Hz_40ms_modulation_mae_db": 0.9646874065167346, "1000–3000 Hz_10ms_log_rms_mae_db": 2.241636535251669, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1648957141283829, "1000–3000 Hz_10ms_envelope_correlation": 0.9785778630633202, "1000–3000 Hz_10ms_modulation_mae_db": 1.0302271782584915, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3027558297256956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10104111181874435, "1000–3000 Hz_40ms_envelope_correlation": 0.9914092821252058, "1000–3000 Hz_40ms_modulation_mae_db": 1.119297254184828, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4495434524688178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2427820798693949, "3000–6000 Hz_10ms_envelope_correlation": 0.9618862809505023, "3000–6000 Hz_10ms_modulation_mae_db": 1.7941872833331478, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4609974118804319, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15173381700921432, "3000–6000 Hz_40ms_envelope_correlation": 0.986159409196225, "3000–6000 Hz_40ms_modulation_mae_db": 2.179162656255927, "mr_stft_db": 6.12841244990189, "band_mae_db": 0.2123857187297594, "rms_gain_error_db": -0.05685986855482687, "output_peak": 0.45664233772908625}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.012560426504786, "全帯域_10ms_rms_envelope_nrmse": 0.07231392769955347, "全帯域_10ms_envelope_correlation": 0.9950048174346201, "全帯域_10ms_modulation_mae_db": 0.47567628186472133, "全帯域_40ms_log_rms_mae_db": 0.5268655795599354, "全帯域_40ms_rms_envelope_nrmse": 0.039353694843912074, "全帯域_40ms_envelope_correlation": 0.9984357665401975, "全帯域_40ms_modulation_mae_db": 0.7349772500895506, "300–1000 Hz_10ms_log_rms_mae_db": 1.889181471878662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10356264901086445, "300–1000 Hz_10ms_envelope_correlation": 0.9918331177512515, "300–1000 Hz_10ms_modulation_mae_db": 0.7650272557765027, "300–1000 Hz_40ms_log_rms_mae_db": 1.1595560306836823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06174905351349468, "300–1000 Hz_40ms_envelope_correlation": 0.9970243630256318, "300–1000 Hz_40ms_modulation_mae_db": 1.1192233056275427, "1000–3000 Hz_10ms_log_rms_mae_db": 2.243931950480954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17347178038895683, "1000–3000 Hz_10ms_envelope_correlation": 0.9762184128482447, "1000–3000 Hz_10ms_modulation_mae_db": 0.7206272675554604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3697753790334788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11479571410152672, "1000–3000 Hz_40ms_envelope_correlation": 0.9887539172770787, "1000–3000 Hz_40ms_modulation_mae_db": 0.9974868056288889, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4910937225102305, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2429758692204381, "3000–6000 Hz_10ms_envelope_correlation": 0.960788808416611, "3000–6000 Hz_10ms_modulation_mae_db": 1.1945466917176704, "3000–6000 Hz_40ms_log_rms_mae_db": 1.503232476966015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15058123322756203, "3000–6000 Hz_40ms_envelope_correlation": 0.9845095787509777, "3000–6000 Hz_40ms_modulation_mae_db": 1.5488483063546745, "mr_stft_db": 6.14412400491791, "band_mae_db": 0.1861435501452565, "rms_gain_error_db": -0.09470847548737302, "output_peak": 0.4450943083598887}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0234132231887547, "全帯域_10ms_rms_envelope_nrmse": 0.06358530013998172, "全帯域_10ms_envelope_correlation": 0.9960726708594133, "全帯域_10ms_modulation_mae_db": 0.5083867541041758, "全帯域_40ms_log_rms_mae_db": 0.538383945447762, "全帯域_40ms_rms_envelope_nrmse": 0.03511569576772459, "全帯域_40ms_envelope_correlation": 0.9986599126371131, "全帯域_40ms_modulation_mae_db": 0.7100521602090145, "300–1000 Hz_10ms_log_rms_mae_db": 1.9357053977542875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08951269741543734, "300–1000 Hz_10ms_envelope_correlation": 0.9937911249307213, "300–1000 Hz_10ms_modulation_mae_db": 0.4915066057195383, "300–1000 Hz_40ms_log_rms_mae_db": 1.2532904013494346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054314665354468564, "300–1000 Hz_40ms_envelope_correlation": 0.997503602127024, "300–1000 Hz_40ms_modulation_mae_db": 0.9456985469550426, "1000–3000 Hz_10ms_log_rms_mae_db": 2.188846215864241, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15644543612160883, "1000–3000 Hz_10ms_envelope_correlation": 0.9808289992124356, "1000–3000 Hz_10ms_modulation_mae_db": 0.7851185425610266, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2833023893692157, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08923455818954984, "1000–3000 Hz_40ms_envelope_correlation": 0.9933866209143678, "1000–3000 Hz_40ms_modulation_mae_db": 0.8588089756214822, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4471007892917203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2520779954784889, "3000–6000 Hz_10ms_envelope_correlation": 0.9578817407202013, "3000–6000 Hz_10ms_modulation_mae_db": 2.04007065983523, "3000–6000 Hz_40ms_log_rms_mae_db": 1.461050322107346, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16544307986111684, "3000–6000 Hz_40ms_envelope_correlation": 0.9815999675917758, "3000–6000 Hz_40ms_modulation_mae_db": 2.2737349933180115, "mr_stft_db": 6.101946715105991, "band_mae_db": 0.13226771543985713, "rms_gain_error_db": -0.021785671032116413, "output_peak": 0.4431788543811532}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 90800, "output_samples": 90800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.502644461727797, "spectral_convergence": 0.18022817775891742, "spectral_convergence_status": "ok", "amplitude_floor": 1.4038593411999917e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.0711753978145016, "spectral_convergence": 0.17686262673639955, "spectral_convergence_status": "ok", "amplitude_floor": 9.854110842773577e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.818519160648745, "spectral_convergence": 0.16989243886790634, "spectral_convergence_status": "ok", "amplitude_floor": 8.983114295610284e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0018717561089627995, "output_power": 0.0018883479567446272, "delta_db": 0.038327634441613224, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00201441754378434, "output_power": 0.0019593468630647705, "delta_db": -0.12038169170242977, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010569467904903444, "output_power": 0.0010352310437525781, "delta_db": -0.09015837493955652, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018132930505040446, "output_power": 0.0001721576990889633, "delta_db": -0.22541547487892752, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.873439242518033e-05, "output_power": 6.860436359871165e-05, "delta_db": -0.5981274048740761, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.113111558995142e-05, "output_power": 3.2501180124325373e-05, "delta_db": 0.18704446590580584, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20990917445706817, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06531039120598112, "pitch": null, "output_rms_dbfs": -22.876711630120703, "output_peak": 0.4626668040552709, "reference_rms_dbfs": -22.81140123891472, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.130779673397015}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 90800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.0001853404348361204, "log_rms_mae_db": 0.9877692810456639, "rms_envelope_nrmse": 0.06496028811727897, "envelope_correlation": 0.9959168472591796, "rms_gain_error_db": -0.06531017306464915, "reference_mean_power": 0.005238930533969212, "output_mean_power": 0.005160735756716771}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 2.4623434734036186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012515931525511833, "output_power": 0.0012192352450887639, "reference_active": true, "delta_db": -0.11375669900476912}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003238300007472361, "output_power": 0.00029777065170578725, "reference_active": true, "delta_db": -0.3643518941159754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.578860153055394e-05, "output_power": 5.7179188725558086e-05, "reference_active": true, "delta_db": -1.223659042606857}], "active_band_count": 3, "active_band_mae_db": 0.5672558785758671}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00016148298564882675, "log_rms_mae_db": 0.5175350893514982, "rms_envelope_nrmse": 0.03231882054753892, "envelope_correlation": 0.9989135565200024, "rms_gain_error_db": -0.0653105730602936, "reference_mean_power": 0.005266794777733929, "output_mean_power": 0.005188183628912355}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.1218089587296202e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001164760758537026, "output_power": 0.0011378822967545372, "reference_active": true, "delta_db": -0.10139389889671652}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015381285066774075, "output_power": 0.00014050637876121834, "reference_active": true, "delta_db": -0.3929658021502267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.591871326154896e-06, "output_power": 3.395963920444568e-06, "reference_active": true, "delta_db": -2.1659410181201175}], "active_band_count": 3, "active_band_mae_db": 0.8867669063890203}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.00014972999954250175, "log_rms_mae_db": 1.8860752204713291, "rms_envelope_nrmse": 0.09221400444569795, "envelope_correlation": 0.9934191909751878, "rms_gain_error_db": -0.09131713652446229, "reference_mean_power": 0.0025001891623953804, "output_mean_power": 0.002448167648673085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.4970716182960256e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007649864504896549, "output_power": 0.0007468307626081599, "reference_active": true, "delta_db": -0.10431544345322985}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022996394368157333, "output_power": 0.00021156275933712987, "reference_active": true, "delta_db": -0.36220525079418375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.80930068112851e-05, "output_power": 3.42259993031457e-05, "reference_active": true, "delta_db": -1.47725793171807}], "active_band_count": 3, "active_band_mae_db": 0.6479262086551613}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00013636056906360855, "log_rms_mae_db": 1.2116201683865546, "rms_envelope_nrmse": 0.051003061213494856, "envelope_correlation": 0.9978573357927735, "rms_gain_error_db": -0.09131823429826894, "reference_mean_power": 0.0025134871346063777, "output_mean_power": 0.002461188307443399}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3161092414230216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215738253307795, "output_power": 0.0007072987094529798, "reference_active": true, "delta_db": -0.08677905377397134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011225926953226224, "output_power": 0.00010177942673816685, "reference_active": true, "delta_db": -0.4256221152500547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.841934016741924e-06, "output_power": 2.2245532159161996e-06, "reference_active": true, "delta_db": -2.3730710237238095}], "active_band_count": 3, "active_band_mae_db": 0.9618240642492785}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.2746042839695326e-05, "log_rms_mae_db": 2.1979750998160457, "rms_envelope_nrmse": 0.17571943121774122, "envelope_correlation": 0.9758064721625064, "rms_gain_error_db": -0.2938728064076227, "reference_mean_power": 0.00020292459461300377, "output_mean_power": 0.00018964762416785643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.5147206216944168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.937659859611645e-05, "output_power": 6.483209324754584e-05, "reference_active": true, "delta_db": -0.29422959400677395}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3925889957061755e-05, "output_power": 1.9162394131903394e-05, "reference_active": true, "delta_db": -0.9641833245653065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8985142545041605e-06, "output_power": 3.0797332769430945e-06, "reference_active": true, "delta_db": -2.822295276147027}], "active_band_count": 3, "active_band_mae_db": 1.3602360649063694}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.512933910060942e-05, "log_rms_mae_db": 1.326671719290263, "rms_envelope_nrmse": 0.11284205131230551, "envelope_correlation": 0.9895277454203841, "rms_gain_error_db": -0.29386933496188755, "reference_mean_power": 0.00020400334148028423, "output_mean_power": 0.00019065594307535755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3444198629453317e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.558296247449876e-05, "output_power": 6.048180798989501e-05, "reference_active": true, "delta_db": -0.35166265364064186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3197503778987225e-05, "output_power": 1.0264727849180774e-05, "reference_active": true, "delta_db": -1.0914435570811984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.991435827054748e-07, "output_power": 2.2244864200829225e-07, "reference_active": true, "delta_db": -3.5099573309398533}], "active_band_count": 3, "active_band_mae_db": 1.651021180553898}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.624711907970004e-05, "log_rms_mae_db": 2.408782698731656, "rms_envelope_nrmse": 0.23399178079636299, "envelope_correlation": 0.9636095938780832, "rms_gain_error_db": -0.2736891671850012, "reference_mean_power": 5.558522396873647e-05, "output_mean_power": 5.219037810103361e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 3.230968975305017e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3093182032279326e-05, "output_power": 1.1399487904790302e-05, "reference_active": true, "delta_db": -0.6015986372635128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8826443460154885e-06, "output_power": 4.619279551676475e-06, "reference_active": true, "delta_db": -1.731810825122371}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3224614880429907e-06, "output_power": 1.7930112633324603e-06, "reference_active": true, "delta_db": -2.6787693784545614}], "active_band_count": 3, "active_band_mae_db": 1.6707262802801484}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.3621221947891856e-05, "log_rms_mae_db": 1.4761769902970072, "rms_envelope_nrmse": 0.15040026063272544, "envelope_correlation": 0.9843573275586216, "rms_gain_error_db": -0.27368953726152306, "reference_mean_power": 5.588085792782989e-05, "output_mean_power": 5.246795186044909e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.891660816172807e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3816996130233874e-05, "output_power": 1.1486735370794333e-05, "reference_active": true, "delta_db": -0.8021701996733244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.814539997745632e-06, "output_power": 2.3776174904103964e-06, "reference_active": true, "delta_db": -2.0530018621176676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.118413559303936e-07, "output_power": 1.5142711975088023e-07, "reference_active": true, "delta_db": -3.1373004830713667}], "active_band_count": 3, "active_band_mae_db": 1.997490848287453}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8916834392037112, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c892b8defc19670b413874505e8227b0f27d569c1ea56bbcf648930c4a7865bb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9776633458442032, "standard_deviation": 0.0021096048983667076, "minimum": 0.9744436027536327, "maximum": 0.9808289992124356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.196811953386739, "standard_deviation": 0.04728372967217803, "minimum": 2.1045353899216357, "maximum": 2.243931950480954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0577884584354156, "standard_deviation": 0.2299052007128401, "minimum": 0.7206272675554604, "maximum": 1.3602360649063694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16835480433964528, "standard_deviation": 0.007889333413220278, "minimum": 0.15644543612160883, "maximum": 0.17990967674096467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9910906630053341, "standard_deviation": 0.0017245762072653362, "minimum": 0.9887539172770787, "maximum": 0.9933866209143678, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3032134777384323, "standard_deviation": 0.04115693001019851, "minimum": 1.2565112313000284, "maximum": 1.3697753790334788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2627230098453577, "standard_deviation": 0.2594588214145658, "minimum": 0.8588089756214822, "maximum": 1.651021180553898, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10290208505527762, "standard_deviation": 0.009775901287770495, "minimum": 0.08923455818954984, "maximum": 0.11479571410152672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9615592596666224, "standard_deviation": 0.0017735390773105193, "minimum": 0.9578817407202013, "maximum": 0.9636095938780832, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.431638978507208, "standard_deviation": 0.029817966881456253, "minimum": 2.399685652603863, "maximum": 2.4910937225102305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7069757134284944, "standard_deviation": 0.3112037683765865, "minimum": 1.1945466917176704, "maximum": 2.077139001683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24148015140430018, "standard_deviation": 0.0055652317072152095, "minimum": 0.23399178079636299, "maximum": 0.2520779954784889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9851514138853626, "standard_deviation": 0.0019900791934992805, "minimum": 0.9815999675917758, "maximum": 0.9879713722150393, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4684247221206763, "standard_deviation": 0.01926883074512264, "minimum": 1.442570295229962, "maximum": 1.503232476966015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.080507742004855, "standard_deviation": 0.3380858041191587, "minimum": 1.5488483063546745, "maximum": 2.509401895717931, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1492092929892232, "standard_deviation": 0.009064750720976214, "minimum": 0.13568257878935933, "maximum": 0.16544307986111684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9934271092513056, "standard_deviation": 0.0007681418618035709, "minimum": 0.9918331177512515, "maximum": 0.9945365014132208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8615219173104036, "standard_deviation": 0.043228194971632466, "minimum": 1.8040893708507282, "maximum": 1.9357053977542875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7076453174649125, "standard_deviation": 0.13732115896260141, "minimum": 0.4915066057195383, "maximum": 0.9181426238953513, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09272026694412934, "standard_deviation": 0.005393000412832553, "minimum": 0.08480072370244035, "maximum": 0.10356264901086445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976220738104604, "standard_deviation": 0.0003240424262945718, "minimum": 0.9970243630256318, "maximum": 0.9980908502060736, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1968197114406574, "standard_deviation": 0.03323089978861117, "minimum": 1.1554756272971611, "maximum": 1.2532904013494346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9202899399058415, "standard_deviation": 0.16372290858503527, "minimum": 0.5872051329925698, "maximum": 1.1192233056275427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054790354027256, "standard_deviation": 0.003918671648640899, "minimum": 0.04937580584662438, "maximum": 0.06174905351349468, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17679380494256725, "standard_deviation": 0.030668003337515275, "minimum": 0.13226771543985713, "maximum": 0.2123857187297594, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.121418805755134, "standard_deviation": 0.015107737423084166, "minimum": 6.100697845697174, "maximum": 6.14412400491791, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4437668298933637, "standard_deviation": 0.012449205784183393, "minimum": 0.4285812802276256, "maximum": 0.4626668040552709, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07661260628227073, "standard_deviation": 0.02723718987354028, "minimum": -0.10481854507959819, "maximum": -0.021785671032116413, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9958102077579613, "standard_deviation": 0.00042876239296745996, "minimum": 0.9950048174346201, "maximum": 0.9963344126902014, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0064780199600616, "standard_deviation": 0.036062277586847304, "minimum": 0.9691478680693054, "maximum": 1.0843165000568216, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5612312897468734, "standard_deviation": 0.072123363600052, "minimum": 0.47567628186472133, "maximum": 0.6982553155566534, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06598082226392113, "standard_deviation": 0.003390251077784119, "minimum": 0.061960723190561964, "maximum": 0.07231392769955347, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986578919432882, "standard_deviation": 0.00022720598009121637, "minimum": 0.9983919744839439, "maximum": 0.9989135565200024, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5473703926073503, "standard_deviation": 0.03235293757547911, "minimum": 0.5175350893514982, "maximum": 0.6167327882846747, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7361316853749325, "standard_deviation": 0.07075925130007449, "minimum": 0.6623298816256357, "maximum": 0.8867669063890203, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03604953344049604, "standard_deviation": 0.003016088641272885, "minimum": 0.03231882054753892, "maximum": 0.04002187027790642, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0001", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.460015296936035, "predicted_mos": 3.460015296936035, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "a9cda117e77ec514f602be78d734a1c86d33c41a1f150b08ac49882543bb9905", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.506438400130719, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9865647808846679, "全帯域_10ms_rms_envelope_nrmse": 0.062469425175862625, "全帯域_10ms_envelope_correlation": 0.9962112531677124, "全帯域_10ms_modulation_mae_db": 0.483907491730788, "全帯域_40ms_log_rms_mae_db": 0.5449328811757416, "全帯域_40ms_rms_envelope_nrmse": 0.0332488234284923, "全帯域_40ms_envelope_correlation": 0.9988081924225176, "全帯域_40ms_modulation_mae_db": 0.649969247365701, "300–1000 Hz_10ms_log_rms_mae_db": 1.5581117807584426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0918134027663026, "300–1000 Hz_10ms_envelope_correlation": 0.9934541216823821, "300–1000 Hz_10ms_modulation_mae_db": 0.6009935084983118, "300–1000 Hz_40ms_log_rms_mae_db": 0.9938814179613495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054335218404331365, "300–1000 Hz_40ms_envelope_correlation": 0.9975330528726474, "300–1000 Hz_40ms_modulation_mae_db": 0.8428791899402754, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7172915251832515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15688602590352516, "1000–3000 Hz_10ms_envelope_correlation": 0.9810813204735672, "1000–3000 Hz_10ms_modulation_mae_db": 0.7514928642619284, "1000–3000 Hz_40ms_log_rms_mae_db": 1.093697146044595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09878428127566728, "1000–3000 Hz_40ms_envelope_correlation": 0.9925858923901214, "1000–3000 Hz_40ms_modulation_mae_db": 1.0151376131800405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5058248910942447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19622356708207872, "3000–6000 Hz_10ms_envelope_correlation": 0.976833074574683, "3000–6000 Hz_10ms_modulation_mae_db": 1.652693232142559, "3000–6000 Hz_40ms_log_rms_mae_db": 0.91793767032604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1319012054039873, "3000–6000 Hz_40ms_envelope_correlation": 0.9913771320826258, "3000–6000 Hz_40ms_modulation_mae_db": 1.868008593662226, "mr_stft_db": 5.407380542074144, "band_mae_db": 0.2745516867792714, "rms_gain_error_db": -0.034766091620660036, "output_peak": 0.46004723479279863}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0537716429506414, "全帯域_10ms_rms_envelope_nrmse": 0.06058024088222262, "全帯域_10ms_envelope_correlation": 0.9964801468385558, "全帯域_10ms_modulation_mae_db": 0.4571118541359292, "全帯域_40ms_log_rms_mae_db": 0.5752817922402211, "全帯域_40ms_rms_envelope_nrmse": 0.03269811599820355, "全帯域_40ms_envelope_correlation": 0.9989237837375506, "全帯域_40ms_modulation_mae_db": 0.4795085736610689, "300–1000 Hz_10ms_log_rms_mae_db": 1.515263944731233, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08105315105867128, "300–1000 Hz_10ms_envelope_correlation": 0.9949237114672844, "300–1000 Hz_10ms_modulation_mae_db": 0.5736680741347521, "300–1000 Hz_40ms_log_rms_mae_db": 0.9690147176237155, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04973347077247483, "300–1000 Hz_40ms_envelope_correlation": 0.9979587781821864, "300–1000 Hz_40ms_modulation_mae_db": 0.575178622797447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6514390833693484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15510888927750627, "1000–3000 Hz_10ms_envelope_correlation": 0.9818585843382919, "1000–3000 Hz_10ms_modulation_mae_db": 1.3055872675860563, "1000–3000 Hz_40ms_log_rms_mae_db": 1.004372288082591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10426394855481602, "1000–3000 Hz_40ms_envelope_correlation": 0.9919511998702146, "1000–3000 Hz_40ms_modulation_mae_db": 1.3945074573382332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4681293099219341, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17315419349456201, "3000–6000 Hz_10ms_envelope_correlation": 0.9800308699800705, "3000–6000 Hz_10ms_modulation_mae_db": 0.6936049484689201, "3000–6000 Hz_40ms_log_rms_mae_db": 0.829641199401054, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07831736363360656, "3000–6000 Hz_40ms_envelope_correlation": 0.9957434292027165, "3000–6000 Hz_40ms_modulation_mae_db": 0.7164081235926991, "mr_stft_db": 5.4011042266898555, "band_mae_db": 0.20195809310179444, "rms_gain_error_db": -0.08357404282787759, "output_peak": 0.4186596505083759}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0332787852030414, "全帯域_10ms_rms_envelope_nrmse": 0.06384938012720766, "全帯域_10ms_envelope_correlation": 0.9960529375094596, "全帯域_10ms_modulation_mae_db": 0.5164938602422606, "全帯域_40ms_log_rms_mae_db": 0.569318646771855, "全帯域_40ms_rms_envelope_nrmse": 0.035926541489300745, "全帯域_40ms_envelope_correlation": 0.9986219278685701, "全帯域_40ms_modulation_mae_db": 0.6042805228531671, "300–1000 Hz_10ms_log_rms_mae_db": 1.5547581603010088, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08424304795035883, "300–1000 Hz_10ms_envelope_correlation": 0.9945000470033657, "300–1000 Hz_10ms_modulation_mae_db": 0.7691068488036388, "300–1000 Hz_40ms_log_rms_mae_db": 1.0176040511272872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04827230308450857, "300–1000 Hz_40ms_envelope_correlation": 0.9980695810334217, "300–1000 Hz_40ms_modulation_mae_db": 0.8907443685774253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7139657852469632, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14366487496368774, "1000–3000 Hz_10ms_envelope_correlation": 0.9835781673598615, "1000–3000 Hz_10ms_modulation_mae_db": 0.1496363782725925, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0370381327018474, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09167236037742957, "1000–3000 Hz_40ms_envelope_correlation": 0.9926301326748795, "1000–3000 Hz_40ms_modulation_mae_db": 0.14431380949686853, "3000–6000 Hz_10ms_log_rms_mae_db": 1.488502164072802, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19228635109713085, "3000–6000 Hz_10ms_envelope_correlation": 0.9764022401521254, "3000–6000 Hz_10ms_modulation_mae_db": 0.6403099477200895, "3000–6000 Hz_40ms_log_rms_mae_db": 0.889155450398724, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1078894396795017, "3000–6000 Hz_40ms_envelope_correlation": 0.9923360718214761, "3000–6000 Hz_40ms_modulation_mae_db": 0.6067180803672866, "mr_stft_db": 5.386856993684151, "band_mae_db": 0.12402692279881106, "rms_gain_error_db": -0.04075719432488342, "output_peak": 0.4466064068896459}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0435823335000658, "全帯域_10ms_rms_envelope_nrmse": 0.06262096941178442, "全帯域_10ms_envelope_correlation": 0.996240150413539, "全帯域_10ms_modulation_mae_db": 0.48898064192599405, "全帯域_40ms_log_rms_mae_db": 0.5333071405285469, "全帯域_40ms_rms_envelope_nrmse": 0.032971253007105496, "全帯域_40ms_envelope_correlation": 0.9988925617766098, "全帯域_40ms_modulation_mae_db": 0.6482459723947965, "300–1000 Hz_10ms_log_rms_mae_db": 1.5721196595288853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08058596441888807, "300–1000 Hz_10ms_envelope_correlation": 0.9950312730900975, "300–1000 Hz_10ms_modulation_mae_db": 0.5344751312399889, "300–1000 Hz_40ms_log_rms_mae_db": 1.0156698383129275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047718120842336384, "300–1000 Hz_40ms_envelope_correlation": 0.998179556237875, "300–1000 Hz_40ms_modulation_mae_db": 0.7258715348166459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.663173859223671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1461662615091407, "1000–3000 Hz_10ms_envelope_correlation": 0.983695016933566, "1000–3000 Hz_10ms_modulation_mae_db": 0.9814341775364491, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9827438286174502, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09235077824839093, "1000–3000 Hz_40ms_envelope_correlation": 0.9936132276935491, "1000–3000 Hz_40ms_modulation_mae_db": 1.1546002719963024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4779528398487556, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19125859662981082, "3000–6000 Hz_10ms_envelope_correlation": 0.9759897055547466, "3000–6000 Hz_10ms_modulation_mae_db": 0.7055617712009253, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8371008347765491, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09876087333775034, "3000–6000 Hz_40ms_envelope_correlation": 0.9931821486778212, "3000–6000 Hz_40ms_modulation_mae_db": 0.8798250543539705, "mr_stft_db": 5.404647045635447, "band_mae_db": 0.27262182487985465, "rms_gain_error_db": -0.07685926877328464, "output_peak": 0.4385967307786476}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0268345148923292, "全帯域_10ms_rms_envelope_nrmse": 0.07132402565746843, "全帯域_10ms_envelope_correlation": 0.9950495316797477, "全帯域_10ms_modulation_mae_db": 0.7871262794826085, "全帯域_40ms_log_rms_mae_db": 0.5677960554729248, "全帯域_40ms_rms_envelope_nrmse": 0.040720483344057076, "全帯域_40ms_envelope_correlation": 0.9982202690799692, "全帯域_40ms_modulation_mae_db": 0.745052917946064, "300–1000 Hz_10ms_log_rms_mae_db": 1.605320023612551, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08736328114681811, "300–1000 Hz_10ms_envelope_correlation": 0.9940935790892681, "300–1000 Hz_10ms_modulation_mae_db": 0.8347075485108163, "300–1000 Hz_40ms_log_rms_mae_db": 1.0426335260989419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05120887520766284, "300–1000 Hz_40ms_envelope_correlation": 0.9978245808961468, "300–1000 Hz_40ms_modulation_mae_db": 0.8581787702477865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.692798539237564, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15755757761966024, "1000–3000 Hz_10ms_envelope_correlation": 0.980409743279932, "1000–3000 Hz_10ms_modulation_mae_db": 0.3323973653059847, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0221500683992786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09448744655339032, "1000–3000 Hz_40ms_envelope_correlation": 0.9926063560956702, "1000–3000 Hz_40ms_modulation_mae_db": 0.24007369408148382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.469829660522059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19018418558192135, "3000–6000 Hz_10ms_envelope_correlation": 0.9779537518397121, "3000–6000 Hz_10ms_modulation_mae_db": 1.5024534155816314, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8947057918261424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12086050616497356, "3000–6000 Hz_40ms_envelope_correlation": 0.9927151008522744, "3000–6000 Hz_40ms_modulation_mae_db": 1.6750961799647905, "mr_stft_db": 5.4014109511982875, "band_mae_db": 0.2411731939477617, "rms_gain_error_db": -0.05064883991079193, "output_peak": 0.4608903482488353}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0364172881038247, "全帯域_10ms_rms_envelope_nrmse": 0.06705576238848947, "全帯域_10ms_envelope_correlation": 0.9956685553679097, "全帯域_10ms_modulation_mae_db": 0.5484543760815116, "全帯域_40ms_log_rms_mae_db": 0.6060601770757048, "全帯域_40ms_rms_envelope_nrmse": 0.03891566632882633, "全帯域_40ms_envelope_correlation": 0.9984300012460648, "全帯域_40ms_modulation_mae_db": 0.7430089271342372, "300–1000 Hz_10ms_log_rms_mae_db": 1.7346576072628717, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09697730000574974, "300–1000 Hz_10ms_envelope_correlation": 0.9927883691711539, "300–1000 Hz_10ms_modulation_mae_db": 0.8353957564117688, "300–1000 Hz_40ms_log_rms_mae_db": 1.1842780808632858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059080312836409105, "300–1000 Hz_40ms_envelope_correlation": 0.9972161062889755, "300–1000 Hz_40ms_modulation_mae_db": 0.8913064475292517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6295866399679486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1502184412108341, "1000–3000 Hz_10ms_envelope_correlation": 0.9826748414302631, "1000–3000 Hz_10ms_modulation_mae_db": 0.7701240958071024, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0141500025206711, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10403386827462073, "1000–3000 Hz_40ms_envelope_correlation": 0.9914886298674573, "1000–3000 Hz_40ms_modulation_mae_db": 0.7384479297806372, "3000–6000 Hz_10ms_log_rms_mae_db": 1.489821458988055, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2056556046708626, "3000–6000 Hz_10ms_envelope_correlation": 0.971851347462812, "3000–6000 Hz_10ms_modulation_mae_db": 1.2826887860715968, "3000–6000 Hz_40ms_log_rms_mae_db": 0.868421547876412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09818437229460603, "3000–6000 Hz_40ms_envelope_correlation": 0.9938270970430173, "3000–6000 Hz_40ms_modulation_mae_db": 1.400136677507694, "mr_stft_db": 5.407603411668558, "band_mae_db": 0.17689964807894973, "rms_gain_error_db": -0.082332409917246, "output_peak": 0.4268300708936076}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9849294336693961, "全帯域_10ms_rms_envelope_nrmse": 0.060764452773198384, "全帯域_10ms_envelope_correlation": 0.9964548114831826, "全帯域_10ms_modulation_mae_db": 0.7870321352898557, "全帯域_40ms_log_rms_mae_db": 0.5269287987992388, "全帯域_40ms_rms_envelope_nrmse": 0.03306825014945008, "全帯域_40ms_envelope_correlation": 0.9988950454383223, "全帯域_40ms_modulation_mae_db": 0.9932976610136167, "300–1000 Hz_10ms_log_rms_mae_db": 1.5698557689760853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09188989123454888, "300–1000 Hz_10ms_envelope_correlation": 0.9934293616386696, "300–1000 Hz_10ms_modulation_mae_db": 0.8770299584149539, "300–1000 Hz_40ms_log_rms_mae_db": 0.9824522113463897, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05307099823098139, "300–1000 Hz_40ms_envelope_correlation": 0.9976464124502042, "300–1000 Hz_40ms_modulation_mae_db": 1.2141732630010185, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6743234813917578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14764144812128535, "1000–3000 Hz_10ms_envelope_correlation": 0.983203322676037, "1000–3000 Hz_10ms_modulation_mae_db": 0.9276432355417826, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9786500882536322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09551412648600374, "1000–3000 Hz_40ms_envelope_correlation": 0.992882906497728, "1000–3000 Hz_40ms_modulation_mae_db": 1.1330484937578815, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5008163994123558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1755780121966035, "3000–6000 Hz_10ms_envelope_correlation": 0.9829719315693409, "3000–6000 Hz_10ms_modulation_mae_db": 1.5046544993138689, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8724935024626512, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12030392463431107, "3000–6000 Hz_40ms_envelope_correlation": 0.9942690884074604, "3000–6000 Hz_40ms_modulation_mae_db": 1.6630382980254839, "mr_stft_db": 5.381491754336639, "band_mae_db": 0.2601414129984984, "rms_gain_error_db": -0.08139482760451458, "output_peak": 0.4299589872752298}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0307799050809816, "全帯域_10ms_rms_envelope_nrmse": 0.06277384659588829, "全帯域_10ms_envelope_correlation": 0.9962104907479757, "全帯域_10ms_modulation_mae_db": 0.5346959335394302, "全帯域_40ms_log_rms_mae_db": 0.555690701108304, "全帯域_40ms_rms_envelope_nrmse": 0.03492800874449073, "全帯域_40ms_envelope_correlation": 0.9987309346707065, "全帯域_40ms_modulation_mae_db": 0.6454993268489988, "300–1000 Hz_10ms_log_rms_mae_db": 1.672844250089054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09016897161564671, "300–1000 Hz_10ms_envelope_correlation": 0.9937209144499484, "300–1000 Hz_10ms_modulation_mae_db": 0.756344174456027, "300–1000 Hz_40ms_log_rms_mae_db": 1.0371554308717583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048902130223842113, "300–1000 Hz_40ms_envelope_correlation": 0.9980539755394914, "300–1000 Hz_40ms_modulation_mae_db": 1.0123556619057916, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6257154933645772, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14390676842084377, "1000–3000 Hz_10ms_envelope_correlation": 0.9837043007034112, "1000–3000 Hz_10ms_modulation_mae_db": 0.32577109238660773, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9774313268294085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08066272207407527, "1000–3000 Hz_40ms_envelope_correlation": 0.9947138186516231, "1000–3000 Hz_40ms_modulation_mae_db": 0.23011062785602565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4914465668477457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18310863829382826, "3000–6000 Hz_10ms_envelope_correlation": 0.977953410878069, "3000–6000 Hz_10ms_modulation_mae_db": 0.27259282009025754, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8872425070839296, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12528483232189072, "3000–6000 Hz_40ms_envelope_correlation": 0.9894338838192465, "3000–6000 Hz_40ms_modulation_mae_db": 0.4423227822870588, "mr_stft_db": 5.410881645400786, "band_mae_db": 0.1760978672713508, "rms_gain_error_db": -0.07076868099441147, "output_peak": 0.45920853946444945}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 90800, "output_samples": 90800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.624791161197132, "spectral_convergence": 0.17866321920758108, "spectral_convergence_status": "ok", "amplitude_floor": 1.4038593411999917e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.23503873773271, "spectral_convergence": 0.17431638618080633, "spectral_convergence_status": "ok", "amplitude_floor": 9.854110842773577e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.36231172729259, "spectral_convergence": 0.17086732683780226, "spectral_convergence_status": "ok", "amplitude_floor": 8.983114295610284e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0018717561089627995, "output_power": 0.0019104830309918113, "delta_db": 0.08893925380346923, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00201441754378434, "output_power": 0.00199453961745488, "delta_db": -0.04306827994551684, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010569467904903444, "output_power": 0.0010229150657853573, "delta_db": -0.14213549250753754, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018132930505040446, "output_power": 0.00016920646832445195, "delta_db": -0.30051036136173925, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.873439242518033e-05, "output_power": 6.675437837013224e-05, "delta_db": -0.7168472446466179, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.113111558995142e-05, "output_power": 2.8682284228524266e-05, "delta_db": -0.35580948841074744, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2745516867792714, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.034766091620660036, "pitch": null, "output_rms_dbfs": -22.84616733053538, "output_peak": 0.46004723479279863, "reference_rms_dbfs": -22.81140123891472, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.407380542074144}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 90800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.0001853404348361204, "log_rms_mae_db": 0.9865647808846679, "rms_envelope_nrmse": 0.062469425175862625, "envelope_correlation": 0.9962112531677124, "rms_gain_error_db": -0.03476596099516679, "reference_mean_power": 0.005238930533969212, "output_mean_power": 0.005197159480458916}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 2.4623434734036186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012515931525511833, "output_power": 0.001222004242641086, "reference_active": true, "delta_db": -0.10390464737986979}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003238300007472361, "output_power": 0.0003168324772302535, "reference_active": true, "delta_db": -0.09487387855561924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.578860153055394e-05, "output_power": 5.679492083233711e-05, "reference_active": true, "delta_db": -1.252943949256875}], "active_band_count": 3, "active_band_mae_db": 0.483907491730788}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00016148298564882675, "log_rms_mae_db": 0.5449328811757416, "rms_envelope_nrmse": 0.0332488234284923, "envelope_correlation": 0.9988081924225176, "rms_gain_error_db": -0.03476572993390076, "reference_mean_power": 0.005266794777733929, "output_mean_power": 0.00522480183494778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.1218089587296202e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001164760758537026, "output_power": 0.0011385720522832335, "reference_active": true, "delta_db": -0.0987621129417838}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015381285066774075, "output_power": 0.0001495437049527868, "reference_active": true, "delta_db": -0.12224485055750517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.591871326154896e-06, "output_power": 3.755494164603723e-06, "reference_active": true, "delta_db": -1.728900778597814}], "active_band_count": 3, "active_band_mae_db": 0.649969247365701}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.00014972999954250175, "log_rms_mae_db": 1.5581117807584426, "rms_envelope_nrmse": 0.0918134027663026, "envelope_correlation": 0.9934541216823821, "rms_gain_error_db": -0.08565181419621483, "reference_mean_power": 0.0025001891623953804, "output_mean_power": 0.0024513633395609508}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.4970716182960256e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007649864504896549, "output_power": 0.0007334640961359617, "reference_active": true, "delta_db": -0.18274883444859347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022996394368157333, "output_power": 0.00021249382160567435, "reference_active": true, "delta_db": -0.34313440631404124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.80930068112851e-05, "output_power": 3.5840346677821306e-05, "reference_active": true, "delta_db": -1.2770972847323008}], "active_band_count": 3, "active_band_mae_db": 0.6009935084983118}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00013636056906360855, "log_rms_mae_db": 0.9938814179613495, "rms_envelope_nrmse": 0.054335218404331365, "envelope_correlation": 0.9975330528726474, "rms_gain_error_db": -0.08565181653261988, "reference_mean_power": 0.0025134871346063777, "output_mean_power": 0.002464401616321809}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3161092414230216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215738253307795, "output_power": 0.0006924192419700667, "reference_active": true, "delta_db": -0.17911642852352222}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011225926953226224, "output_power": 0.00010121421026566108, "reference_active": true, "delta_db": -0.4498072112053925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.841934016741924e-06, "output_power": 2.480724345588417e-06, "reference_active": true, "delta_db": -1.8997139300919117}], "active_band_count": 3, "active_band_mae_db": 0.8428791899402754}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.2746042839695326e-05, "log_rms_mae_db": 1.7172915251832515, "rms_envelope_nrmse": 0.15688602590352516, "envelope_correlation": 0.9810813204735672, "rms_gain_error_db": -0.3639973861869732, "reference_mean_power": 0.00020292459461300377, "output_mean_power": 0.0001866100152738838}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.5147206216944168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.937659859611645e-05, "output_power": 6.318192555708732e-05, "reference_active": true, "delta_db": -0.40620145888607795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3925889957061755e-05, "output_power": 2.0850180017999712e-05, "reference_active": true, "delta_db": -0.5975829196880833}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8985142545041605e-06, "output_power": 4.422554686865044e-06, "reference_active": true, "delta_db": -1.2506942142116237}], "active_band_count": 3, "active_band_mae_db": 0.7514928642619284}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.512933910060942e-05, "log_rms_mae_db": 1.093697146044595, "rms_envelope_nrmse": 0.09878428127566728, "envelope_correlation": 0.9925858923901214, "rms_gain_error_db": -0.3639972860553676, "reference_mean_power": 0.00020400334148028423, "output_mean_power": 0.00018760203818510388}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3444198629453317e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.558296247449876e-05, "output_power": 5.9932804520674494e-05, "reference_active": true, "delta_db": -0.39126429738692214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3197503778987225e-05, "output_power": 1.1376926775671879e-05, "reference_active": true, "delta_db": -0.6446683214594604}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.991435827054748e-07, "output_power": 3.142515799589629e-07, "reference_active": true, "delta_db": -2.0094802206937388}], "active_band_count": 3, "active_band_mae_db": 1.0151376131800405}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.624711907970004e-05, "log_rms_mae_db": 1.5058248910942447, "rms_envelope_nrmse": 0.19622356708207872, "envelope_correlation": 0.976833074574683, "rms_gain_error_db": -0.7014498619106184, "reference_mean_power": 5.558522396873647e-05, "output_mean_power": 4.729490677040209e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 3.230968975305017e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3093182032279326e-05, "output_power": 1.0652347129874906e-05, "reference_active": true, "delta_db": -0.8959989541509018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8826443460154885e-06, "output_power": 5.550922186164137e-06, "reference_active": true, "delta_db": -0.9339018903099614}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3224614880429907e-06, "output_power": 1.6167470246314654e-06, "reference_active": true, "delta_db": -3.1281788519668137}], "active_band_count": 3, "active_band_mae_db": 1.652693232142559}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.3621221947891856e-05, "log_rms_mae_db": 0.91793767032604, "rms_envelope_nrmse": 0.1319012054039873, "envelope_correlation": 0.9913771320826258, "rms_gain_error_db": -0.7014501811104583, "reference_mean_power": 5.588085792782989e-05, "output_mean_power": 4.754644459444146e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.891660816172807e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3816996130233874e-05, "output_power": 1.0956114929263541e-05, "reference_active": true, "delta_db": -1.007570566171888}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.814539997745632e-06, "output_power": 2.8448950168237766e-06, "reference_active": true, "delta_db": -1.2737592853357238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.118413559303936e-07, "output_power": 1.4509889904968227e-07, "reference_active": true, "delta_db": -3.322695929479067}], "active_band_count": 3, "active_band_mae_db": 1.868008593662226}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9208065679050867, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bd512c8a2f25f832dccdbf6ed7082bc868e5e5734c01bb02e74958ed63a7f5f6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9825256621493663, "standard_deviation": 0.0012735806252056078, "minimum": 0.980409743279932, "maximum": 0.9837043007034112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.671036800873135, "standard_deviation": 0.03522338079405053, "minimum": 1.6257154933645772, "maximum": 1.7172915251832515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.693010809587313, "standard_deviation": 0.39334096072817776, "minimum": 0.1496363782725925, "maximum": 1.3055872675860563, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1501437858783104, "standard_deviation": 0.005707626495526187, "minimum": 0.14366487496368774, "maximum": 0.15755757761966024, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9928090204676554, "standard_deviation": 0.0009915796829022498, "minimum": 0.9914886298674573, "maximum": 0.9947138186516231, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0137791101811842, "standard_deviation": 0.0389229488447691, "minimum": 0.9774313268294085, "maximum": 1.093697146044595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.756279987185934, "standard_deviation": 0.49180409102945105, "minimum": 0.14431380949686853, "maximum": 1.3945074573382332, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09522119148054922, "standard_deviation": 0.0076114297041231, "minimum": 0.08066272207407527, "maximum": 0.10426394855481602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.977498291501445, "standard_deviation": 0.0032204018774062183, "minimum": 0.971851347462812, "maximum": 0.9829719315693409, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4865404113384941, "standard_deviation": 0.013660910535415674, "minimum": 1.4681293099219341, "maximum": 1.5058248910942447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.031819927573731, "standard_deviation": 0.5132566718668394, "minimum": 0.27259282009025754, "maximum": 1.652693232142559, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18843114363084978, "standard_deviation": 0.01075872701384494, "minimum": 0.17315419349456201, "maximum": 0.2056556046708626, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9928604939883298, "standard_deviation": 0.0019112977812328269, "minimum": 0.9894338838192465, "maximum": 0.9957434292027165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8745873130189378, "standard_deviation": 0.029578421464969328, "minimum": 0.829641199401054, "maximum": 0.91793767032604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1564442237201513, "standard_deviation": 0.5572464795143589, "minimum": 0.4423227822870588, "maximum": 1.868008593662226, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1101878146838284, "standard_deviation": 0.017773572265888795, "minimum": 0.07831736363360656, "maximum": 0.1319012054039873, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9939926721990212, "standard_deviation": 0.0007878261978437899, "minimum": 0.9927883691711539, "maximum": 0.9950312730900975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5978663994075166, "standard_deviation": 0.07181173271232862, "minimum": 1.515263944731233, "maximum": 1.7346576072628717, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7227151250587822, "standard_deviation": 0.13352413959496814, "minimum": 0.5344751312399889, "maximum": 0.8770299584149539, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08801187627462304, "standard_deviation": 0.005765850857581613, "minimum": 0.08058596441888807, "maximum": 0.09697730000574974, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978102554376186, "standard_deviation": 0.000325575436851284, "minimum": 0.9972161062889755, "maximum": 0.998179556237875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.030336159275707, "standard_deviation": 0.06724708204643676, "minimum": 0.9690147176237155, "maximum": 1.1842780808632858, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8763359823519552, "standard_deviation": 0.18816840646435856, "minimum": 0.575178622797447, "maximum": 1.2141732630010185, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05154017870031832, "standard_deviation": 0.003829018352153394, "minimum": 0.047718120842336384, "maximum": 0.059080312836409105, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21593383123203652, "standard_deviation": 0.05476402609879967, "minimum": 0.12402692279881106, "maximum": 0.2745516867792714, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.400172071335984, "standard_deviation": 0.010491877907213457, "minimum": 5.381491754336639, "maximum": 5.410881645400786, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4425997461064488, "standard_deviation": 0.01659815000336215, "minimum": 0.4186596505083759, "maximum": 0.4608903482488353, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06513766949670871, "standard_deviation": 0.019984725683883802, "minimum": -0.08357404282787759, "maximum": -0.034766091620660036, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9960459846510104, "standard_deviation": 0.0004755646562033752, "minimum": 0.9950495316797477, "maximum": 0.9964801468385558, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0245198355356187, "standard_deviation": 0.02532601132461617, "minimum": 0.9849294336693961, "maximum": 1.0537716429506414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5754753215535473, "standard_deviation": 0.13379100101673302, "minimum": 0.4571118541359292, "maximum": 0.7871262794826085, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06392976287651524, "standard_deviation": 0.003603046548815848, "minimum": 0.06058024088222262, "maximum": 0.07132402565746843, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986903395300388, "standard_deviation": 0.00025248881706006515, "minimum": 0.9982202690799692, "maximum": 0.9989237837375506, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5599145241465671, "standard_deviation": 0.02550751014944477, "minimum": 0.5269287987992388, "maximum": 0.6060601770757048, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6886078936522063, "standard_deviation": 0.1487065071297448, "minimum": 0.4795085736610689, "maximum": 0.9932976610136167, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.035309642811240785, "standard_deviation": 0.003032343887979853, "minimum": 0.03269811599820355, "maximum": 0.040720483344057076, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0001", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.1206793785095215, "predicted_mos": 3.1206793785095215, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "58577f334782c85133cb8d90d9822a1f453bbc8e2776ec1f48a9cea7fd72ad27", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.449006599839777, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0145786844510298, "全帯域_10ms_rms_envelope_nrmse": 0.06712911060172538, "全帯域_10ms_envelope_correlation": 0.9956639007088839, "全帯域_10ms_modulation_mae_db": 0.6116502511858665, "全帯域_40ms_log_rms_mae_db": 0.5403206998986344, "全帯域_40ms_rms_envelope_nrmse": 0.03345353286833441, "全帯域_40ms_envelope_correlation": 0.9988444546440659, "全帯域_40ms_modulation_mae_db": 0.7256132400471124, "300–1000 Hz_10ms_log_rms_mae_db": 1.6391478640982344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08962577416057972, "300–1000 Hz_10ms_envelope_correlation": 0.9938567145651187, "300–1000 Hz_10ms_modulation_mae_db": 0.9246864410550533, "300–1000 Hz_40ms_log_rms_mae_db": 1.0096871371267209, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04981755771920635, "300–1000 Hz_40ms_envelope_correlation": 0.998045890601042, "300–1000 Hz_40ms_modulation_mae_db": 1.1156054370249309, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8681745561535397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15763913043575412, "1000–3000 Hz_10ms_envelope_correlation": 0.9809755595311666, "1000–3000 Hz_10ms_modulation_mae_db": 0.7888968923800523, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0763329360330969, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09454815068544087, "1000–3000 Hz_40ms_envelope_correlation": 0.9934028739311822, "1000–3000 Hz_40ms_modulation_mae_db": 0.7109666743843771, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7484474433950077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17798534553852458, "3000–6000 Hz_10ms_envelope_correlation": 0.9791744511360989, "3000–6000 Hz_10ms_modulation_mae_db": 1.368336542630597, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9924312406671018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1002975948906067, "3000–6000 Hz_40ms_envelope_correlation": 0.9934673112435408, "3000–6000 Hz_40ms_modulation_mae_db": 1.56630341892136, "mr_stft_db": 5.584172832412148, "band_mae_db": 0.1852954140236187, "rms_gain_error_db": -0.065174510280683, "output_peak": 0.43161962546331417}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0216555285981526, "全帯域_10ms_rms_envelope_nrmse": 0.06247884539875602, "全帯域_10ms_envelope_correlation": 0.9962454723324, "全帯域_10ms_modulation_mae_db": 0.6334731299057902, "全帯域_40ms_log_rms_mae_db": 0.5406692324491043, "全帯域_40ms_rms_envelope_nrmse": 0.036269507618215825, "全帯域_40ms_envelope_correlation": 0.9986419415350883, "全帯域_40ms_modulation_mae_db": 0.7576124087472728, "300–1000 Hz_10ms_log_rms_mae_db": 1.6296828385697177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09449407856331005, "300–1000 Hz_10ms_envelope_correlation": 0.9931385254401054, "300–1000 Hz_10ms_modulation_mae_db": 0.7978733991076784, "300–1000 Hz_40ms_log_rms_mae_db": 1.075195418563358, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05801732228885567, "300–1000 Hz_40ms_envelope_correlation": 0.9972965634409949, "300–1000 Hz_40ms_modulation_mae_db": 0.9803107837756082, "1000–3000 Hz_10ms_log_rms_mae_db": 1.788972779637557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14843489350266809, "1000–3000 Hz_10ms_envelope_correlation": 0.9828336819322305, "1000–3000 Hz_10ms_modulation_mae_db": 0.9221299234071755, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0024113026296582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08331921012572761, "1000–3000 Hz_40ms_envelope_correlation": 0.9945624852995922, "1000–3000 Hz_40ms_modulation_mae_db": 1.2356919620573656, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6697250358334947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16397874243747376, "3000–6000 Hz_10ms_envelope_correlation": 0.9821418571940895, "3000–6000 Hz_10ms_modulation_mae_db": 0.8040988885664145, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9583113215695381, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09483762864284742, "3000–6000 Hz_40ms_envelope_correlation": 0.993589229373783, "3000–6000 Hz_40ms_modulation_mae_db": 1.1397800890010437, "mr_stft_db": 5.565925537894109, "band_mae_db": 0.15376522571960113, "rms_gain_error_db": -0.07885680577033627, "output_peak": 0.42853768333386494}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.996630966610435, "全帯域_10ms_rms_envelope_nrmse": 0.06574954290525384, "全帯域_10ms_envelope_correlation": 0.9958354432197426, "全帯域_10ms_modulation_mae_db": 0.7676634992602706, "全帯域_40ms_log_rms_mae_db": 0.5218222926610219, "全帯域_40ms_rms_envelope_nrmse": 0.033640290318738095, "全帯域_40ms_envelope_correlation": 0.9988388242253428, "全帯域_40ms_modulation_mae_db": 0.8973236644082867, "300–1000 Hz_10ms_log_rms_mae_db": 1.6859273987362102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0867437978725962, "300–1000 Hz_10ms_envelope_correlation": 0.9942419320427013, "300–1000 Hz_10ms_modulation_mae_db": 0.9167856851222712, "300–1000 Hz_40ms_log_rms_mae_db": 1.0228910139423275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05038814746060097, "300–1000 Hz_40ms_envelope_correlation": 0.9979769346069569, "300–1000 Hz_40ms_modulation_mae_db": 1.1228714655795844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8464651847043232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15422743630064012, "1000–3000 Hz_10ms_envelope_correlation": 0.981755418139836, "1000–3000 Hz_10ms_modulation_mae_db": 0.6454572756784088, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0811960932039453, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09221183795967362, "1000–3000 Hz_40ms_envelope_correlation": 0.9937134718928942, "1000–3000 Hz_40ms_modulation_mae_db": 0.7676248151102071, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7751022868373325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20207229528554946, "3000–6000 Hz_10ms_envelope_correlation": 0.9729256897339836, "3000–6000 Hz_10ms_modulation_mae_db": 1.5795388259635408, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9847364011794456, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10559034220257359, "3000–6000 Hz_40ms_envelope_correlation": 0.9927777801651114, "3000–6000 Hz_40ms_modulation_mae_db": 1.8356832303492447, "mr_stft_db": 5.600915579907678, "band_mae_db": 0.19023833513971677, "rms_gain_error_db": -0.07633741875089195, "output_peak": 0.4219930278704581}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9574788792819964, "全帯域_10ms_rms_envelope_nrmse": 0.0647407549697344, "全帯域_10ms_envelope_correlation": 0.9959723780441461, "全帯域_10ms_modulation_mae_db": 0.5384888541741908, "全帯域_40ms_log_rms_mae_db": 0.5074986207523042, "全帯域_40ms_rms_envelope_nrmse": 0.037461002566874936, "全帯域_40ms_envelope_correlation": 0.9985383248830692, "全帯域_40ms_modulation_mae_db": 0.5486275089717344, "300–1000 Hz_10ms_log_rms_mae_db": 1.649123990902243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09997416454938854, "300–1000 Hz_10ms_envelope_correlation": 0.9922826394249032, "300–1000 Hz_10ms_modulation_mae_db": 0.8694002645721305, "300–1000 Hz_40ms_log_rms_mae_db": 1.1050932292463276, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05851690327198733, "300–1000 Hz_40ms_envelope_correlation": 0.9972090577992276, "300–1000 Hz_40ms_modulation_mae_db": 0.9189679492650983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8157769112821045, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14861406730016608, "1000–3000 Hz_10ms_envelope_correlation": 0.982606995769957, "1000–3000 Hz_10ms_modulation_mae_db": 0.5007242596686609, "1000–3000 Hz_40ms_log_rms_mae_db": 1.035830735167664, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08543258961479828, "1000–3000 Hz_40ms_envelope_correlation": 0.9939876242847819, "1000–3000 Hz_40ms_modulation_mae_db": 0.5956297922548014, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7537667114705617, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18041180453336414, "3000–6000 Hz_10ms_envelope_correlation": 0.9784682176727691, "3000–6000 Hz_10ms_modulation_mae_db": 1.1504955983907286, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0609641673405374, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10611488325499509, "3000–6000 Hz_40ms_envelope_correlation": 0.9923203002914307, "3000–6000 Hz_40ms_modulation_mae_db": 1.369081962129344, "mr_stft_db": 5.564069416652063, "band_mae_db": 0.18878923553541163, "rms_gain_error_db": -0.0733337768650139, "output_peak": 0.42039690535972063}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0325333172288995, "全帯域_10ms_rms_envelope_nrmse": 0.06993711882586294, "全帯域_10ms_envelope_correlation": 0.9952345251740584, "全帯域_10ms_modulation_mae_db": 0.41538716375716006, "全帯域_40ms_log_rms_mae_db": 0.5641483417106046, "全帯域_40ms_rms_envelope_nrmse": 0.037298302376821205, "全帯域_40ms_envelope_correlation": 0.9984874383930238, "全帯域_40ms_modulation_mae_db": 0.5794194211315663, "300–1000 Hz_10ms_log_rms_mae_db": 1.680665260547864, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09444510315857385, "300–1000 Hz_10ms_envelope_correlation": 0.9930949337757545, "300–1000 Hz_10ms_modulation_mae_db": 0.5444195838768199, "300–1000 Hz_40ms_log_rms_mae_db": 1.0285405912518863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05512843402102493, "300–1000 Hz_40ms_envelope_correlation": 0.9974758797724059, "300–1000 Hz_40ms_modulation_mae_db": 0.7470836204889489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7630445860959303, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15758916496103506, "1000–3000 Hz_10ms_envelope_correlation": 0.9802781057745109, "1000–3000 Hz_10ms_modulation_mae_db": 0.6419425706296233, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0386858205309097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1048468013151132, "1000–3000 Hz_40ms_envelope_correlation": 0.9904520396617524, "1000–3000 Hz_40ms_modulation_mae_db": 0.6296064279712879, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7629083939832206, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20346482289224208, "3000–6000 Hz_10ms_envelope_correlation": 0.9750949433803298, "3000–6000 Hz_10ms_modulation_mae_db": 1.8660927789645807, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9900007398586192, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12296697619832504, "3000–6000 Hz_40ms_envelope_correlation": 0.9933694781391873, "3000–6000 Hz_40ms_modulation_mae_db": 2.3118985257093008, "mr_stft_db": 5.56984232528846, "band_mae_db": 0.20860909199246633, "rms_gain_error_db": -0.025754862479103698, "output_peak": 0.43970595689426595}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9837906072353247, "全帯域_10ms_rms_envelope_nrmse": 0.0622460532714265, "全帯域_10ms_envelope_correlation": 0.9962687237923307, "全帯域_10ms_modulation_mae_db": 0.40002464062567583, "全帯域_40ms_log_rms_mae_db": 0.5155308920841086, "全帯域_40ms_rms_envelope_nrmse": 0.03466383949135951, "全帯域_40ms_envelope_correlation": 0.998740080152011, "全帯域_40ms_modulation_mae_db": 0.5901057977668837, "300–1000 Hz_10ms_log_rms_mae_db": 1.7155502827647109, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0962690909374024, "300–1000 Hz_10ms_envelope_correlation": 0.992860475321156, "300–1000 Hz_10ms_modulation_mae_db": 0.5204297901912172, "300–1000 Hz_40ms_log_rms_mae_db": 1.1430368261866792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05740353681562264, "300–1000 Hz_40ms_envelope_correlation": 0.9973381247638631, "300–1000 Hz_40ms_modulation_mae_db": 0.6845493512318299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8260902419418572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.149010384518249, "1000–3000 Hz_10ms_envelope_correlation": 0.9827339478213452, "1000–3000 Hz_10ms_modulation_mae_db": 0.8223860963483572, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0693395358543734, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09180383802744788, "1000–3000 Hz_40ms_envelope_correlation": 0.9932882474121948, "1000–3000 Hz_40ms_modulation_mae_db": 0.8379215804767712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7109829434815642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1767047893420917, "3000–6000 Hz_10ms_envelope_correlation": 0.9792059208007604, "3000–6000 Hz_10ms_modulation_mae_db": 0.25275342515076876, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9234845301825987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09379134035573723, "3000–6000 Hz_40ms_envelope_correlation": 0.993776565757736, "3000–6000 Hz_40ms_modulation_mae_db": 0.3543378097435855, "mr_stft_db": 5.585298662540747, "band_mae_db": 0.16632547261438227, "rms_gain_error_db": -0.06220822342325239, "output_peak": 0.43865354323660943}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9947207141089203, "全帯域_10ms_rms_envelope_nrmse": 0.06679063655306176, "全帯域_10ms_envelope_correlation": 0.9958372854404254, "全帯域_10ms_modulation_mae_db": 0.7436280390703415, "全帯域_40ms_log_rms_mae_db": 0.5474477678360173, "全帯域_40ms_rms_envelope_nrmse": 0.03854132049722019, "全帯域_40ms_envelope_correlation": 0.9986307484974235, "全帯域_40ms_modulation_mae_db": 0.9667312586046662, "300–1000 Hz_10ms_log_rms_mae_db": 1.6166586989331004, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0895752575948844, "300–1000 Hz_10ms_envelope_correlation": 0.9940118646134054, "300–1000 Hz_10ms_modulation_mae_db": 1.0104025105719348, "300–1000 Hz_40ms_log_rms_mae_db": 1.0585856981087383, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060132802599721404, "300–1000 Hz_40ms_envelope_correlation": 0.9972359256325358, "300–1000 Hz_40ms_modulation_mae_db": 1.3054668536286906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8228677245396183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14377536364841673, "1000–3000 Hz_10ms_envelope_correlation": 0.9838958096256614, "1000–3000 Hz_10ms_modulation_mae_db": 0.6905887997369158, "1000–3000 Hz_40ms_log_rms_mae_db": 1.064113579557669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08318567433848564, "1000–3000 Hz_40ms_envelope_correlation": 0.99454892550202, "1000–3000 Hz_40ms_modulation_mae_db": 0.9885043889574264, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7892989058621076, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19225121521036503, "3000–6000 Hz_10ms_envelope_correlation": 0.9768560839079212, "3000–6000 Hz_10ms_modulation_mae_db": 1.6064062812186461, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0117050626228736, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11784285091900194, "3000–6000 Hz_40ms_envelope_correlation": 0.9925412570009675, "3000–6000 Hz_40ms_modulation_mae_db": 1.7978127644905741, "mr_stft_db": 5.589323100177961, "band_mae_db": 0.20750566613104193, "rms_gain_error_db": -0.13847893707842177, "output_peak": 0.4185160886560911}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9533999091295668, "全帯域_10ms_rms_envelope_nrmse": 0.06130205587186639, "全帯域_10ms_envelope_correlation": 0.9963538040581486, "全帯域_10ms_modulation_mae_db": 0.5331870890468341, "全帯域_40ms_log_rms_mae_db": 0.5023657216010641, "全帯域_40ms_rms_envelope_nrmse": 0.03130746349456764, "全帯域_40ms_envelope_correlation": 0.9989352563253909, "全帯域_40ms_modulation_mae_db": 0.5881063654759192, "300–1000 Hz_10ms_log_rms_mae_db": 1.6880461210355306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09320424483868116, "300–1000 Hz_10ms_envelope_correlation": 0.9932094223954131, "300–1000 Hz_10ms_modulation_mae_db": 0.6797241355236698, "300–1000 Hz_40ms_log_rms_mae_db": 1.0401580391770426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056814559334284094, "300–1000 Hz_40ms_envelope_correlation": 0.997232011665701, "300–1000 Hz_40ms_modulation_mae_db": 0.8816991584760845, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8418561466512275, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1465715113807441, "1000–3000 Hz_10ms_envelope_correlation": 0.9831770418215445, "1000–3000 Hz_10ms_modulation_mae_db": 0.9438211218654898, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0477847387984052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08635087951108082, "1000–3000 Hz_40ms_envelope_correlation": 0.9938768399497561, "1000–3000 Hz_40ms_modulation_mae_db": 0.9083923679643834, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6972196540365354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18154278230082785, "3000–6000 Hz_10ms_envelope_correlation": 0.9782149440259613, "3000–6000 Hz_10ms_modulation_mae_db": 1.0510730308896727, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9289820165927904, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0966396187807156, "3000–6000 Hz_40ms_envelope_correlation": 0.9937780635107276, "3000–6000 Hz_40ms_modulation_mae_db": 1.1899014870136895, "mr_stft_db": 5.583592564907701, "band_mae_db": 0.15066799323781832, "rms_gain_error_db": -0.00846785947624804, "output_peak": 0.42072943463032625}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 90800, "output_samples": 90800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.900908734439049, "spectral_convergence": 0.1757635482734265, "spectral_convergence_status": "ok", "amplitude_floor": 1.4038593411999917e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.441196937040846, "spectral_convergence": 0.17210989050673814, "spectral_convergence_status": "ok", "amplitude_floor": 9.854110842773577e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.4104128257565485, "spectral_convergence": 0.16587704147735782, "spectral_convergence_status": "ok", "amplitude_floor": 8.983114295610284e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0018717561089627995, "output_power": 0.0018822018207238856, "delta_db": 0.02416929955443948, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00201441754378434, "output_power": 0.0019920712318958456, "delta_db": -0.04844631359364476, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010569467904903444, "output_power": 0.0010110354966208419, "delta_db": -0.19286720740571978, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018132930505040446, "output_power": 0.00016878548173827976, "delta_db": -0.3113290951123283, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.873439242518033e-05, "output_power": 7.227228058799526e-05, "delta_db": -0.3719272117417743, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.113111558995142e-05, "output_power": 2.9984122831457414e-05, "delta_db": -0.1630333567338057, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1852954140236187, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.065174510280683, "pitch": null, "output_rms_dbfs": -22.876575749195403, "output_peak": 0.43161962546331417, "reference_rms_dbfs": -22.81140123891472, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.584172832412148}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 90800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.0001853404348361204, "log_rms_mae_db": 1.0145786844510298, "rms_envelope_nrmse": 0.06712911060172538, "envelope_correlation": 0.9956639007088839, "rms_gain_error_db": -0.06517460091100892, "reference_mean_power": 0.005238930533969212, "output_mean_power": 0.005160896860071883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 2.4623434734036186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012515931525511833, "output_power": 0.0012220984326749508, "reference_active": true, "delta_db": -0.10356991339024613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003238300007472361, "output_power": 0.00029214213723152195, "reference_active": true, "delta_db": -0.44722878725339527}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.578860153055394e-05, "output_power": 5.6388259296754644e-05, "reference_active": true, "delta_db": -1.284152052913958}], "active_band_count": 3, "active_band_mae_db": 0.6116502511858665}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00016148298564882675, "log_rms_mae_db": 0.5403206998986344, "rms_envelope_nrmse": 0.03345353286833441, "envelope_correlation": 0.9988444546440659, "rms_gain_error_db": -0.06517467831281702, "reference_mean_power": 0.005266794777733929, "output_mean_power": 0.005188345974501441}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.1218089587296202e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001164760758537026, "output_power": 0.0011338300339603612, "reference_active": true, "delta_db": -0.1168877377383039}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015381285066774075, "output_power": 0.00013456660860687496, "reference_active": true, "delta_db": -0.580553138562402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.591871326154896e-06, "output_power": 3.977564995469174e-06, "reference_active": true, "delta_db": -1.4793988438406311}], "active_band_count": 3, "active_band_mae_db": 0.7256132400471124}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.00014972999954250175, "log_rms_mae_db": 1.6391478640982344, "rms_envelope_nrmse": 0.08962577416057972, "envelope_correlation": 0.9938567145651187, "rms_gain_error_db": -0.1248406850720949, "reference_mean_power": 0.0025001891623953804, "output_mean_power": 0.002429342789861385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.4970716182960256e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007649864504896549, "output_power": 0.0007313884546111671, "reference_active": true, "delta_db": -0.1950564249083575}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022996394368157333, "output_power": 0.00020145292299271556, "reference_active": true, "delta_db": -0.5748617461958976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.80930068112851e-05, "output_power": 3.0315714968916072e-05, "reference_active": true, "delta_db": -2.004141152060905}], "active_band_count": 3, "active_band_mae_db": 0.9246864410550533}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00013636056906360855, "log_rms_mae_db": 1.0096871371267209, "rms_envelope_nrmse": 0.04981755771920635, "envelope_correlation": 0.998045890601042, "rms_gain_error_db": -0.12484075140533331, "reference_mean_power": 0.0025134871346063777, "output_mean_power": 0.0024422639080442887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3161092414230216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215738253307795, "output_power": 0.0006883689683661244, "reference_active": true, "delta_db": -0.20459486914546154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011225926953226224, "output_power": 9.374105975182795e-05, "reference_active": true, "delta_db": -0.7829235299809162}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.841934016741924e-06, "output_power": 2.2316193097076825e-06, "reference_active": true, "delta_db": -2.3592979119484148}], "active_band_count": 3, "active_band_mae_db": 1.1156054370249309}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.2746042839695326e-05, "log_rms_mae_db": 1.8681745561535397, "rms_envelope_nrmse": 0.15763913043575412, "envelope_correlation": 0.9809755595311666, "rms_gain_error_db": -0.3570253694702572, "reference_mean_power": 0.00020292459461300377, "output_mean_power": 0.00018690983327495565}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.5147206216944168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.937659859611645e-05, "output_power": 6.169924610542344e-05, "reference_active": true, "delta_db": -0.5093314590670757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3925889957061755e-05, "output_power": 2.05828890605183e-05, "reference_active": true, "delta_db": -0.6536176768875186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8985142545041605e-06, "output_power": 4.470627437355129e-06, "reference_active": true, "delta_db": -1.2037415411855628}], "active_band_count": 3, "active_band_mae_db": 0.7888968923800523}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.512933910060942e-05, "log_rms_mae_db": 1.0763329360330969, "rms_envelope_nrmse": 0.09454815068544087, "envelope_correlation": 0.9934028739311822, "rms_gain_error_db": -0.35702355733255065, "reference_mean_power": 0.00020400334148028423, "output_mean_power": 0.00018790352409750962}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3444198629453317e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.558296247449876e-05, "output_power": 5.813054001557584e-05, "reference_active": true, "delta_db": -0.5238667302516671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3197503778987225e-05, "output_power": 1.160949176938049e-05, "reference_active": true, "delta_db": -0.5567858705519267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.991435827054748e-07, "output_power": 3.917425533476411e-07, "reference_active": true, "delta_db": -1.0522474223495377}], "active_band_count": 3, "active_band_mae_db": 0.7109666743843771}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.624711907970004e-05, "log_rms_mae_db": 1.7484474433950077, "rms_envelope_nrmse": 0.17798534553852458, "envelope_correlation": 0.9791744511360989, "rms_gain_error_db": -0.30485286569672204, "reference_mean_power": 5.558522396873647e-05, "output_mean_power": 5.1817215705065486e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 3.230968975305017e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3093182032279326e-05, "output_power": 1.1988046330565492e-05, "reference_active": true, "delta_db": -0.3829679313397016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8826443460154885e-06, "output_power": 6.191833956733193e-06, "reference_active": true, "delta_db": -0.4593602681014827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3224614880429907e-06, "output_power": 1.567443202228531e-06, "reference_active": true, "delta_db": -3.2626814284506063}], "active_band_count": 3, "active_band_mae_db": 1.368336542630597}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.3621221947891856e-05, "log_rms_mae_db": 0.9924312406671018, "rms_envelope_nrmse": 0.1002975948906067, "envelope_correlation": 0.9934673112435408, "rms_gain_error_db": -0.3048537540009567, "reference_mean_power": 5.588085792782989e-05, "output_mean_power": 5.20927985897152e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.891660816172807e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3816996130233874e-05, "output_power": 1.2353312791983434e-05, "reference_active": true, "delta_db": -0.48630197847214907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.814539997745632e-06, "output_power": 3.164917136230812e-06, "reference_active": true, "delta_db": -0.8107982927743324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.118413559303936e-07, "output_power": 1.4247960800156365e-07, "reference_active": true, "delta_db": -3.4018099855175987}], "active_band_count": 3, "active_band_mae_db": 1.56630341892136}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9144050055909293, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c375415218be0059760d340c1f98637596a01719978319a82f99d526cd540911", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9822820700520316, "standard_deviation": 0.001196386260488462, "minimum": 0.9802781057745109, "maximum": 0.9838958096256614, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8216560163757696, "standard_deviation": 0.03326667751935671, "minimum": 1.7630445860959303, "maximum": 1.8681745561535397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7444933674643355, "standard_deviation": 0.1521179728579223, "minimum": 0.5007242596686609, "maximum": 0.9438211218654898, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15073274400595915, "standard_deviation": 0.005146121765675091, "minimum": 0.14377536364841673, "maximum": 0.15763913043575412, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934790634917717, "standard_deviation": 0.0013090511240009281, "minimum": 0.9904520396617524, "maximum": 0.9945624852995922, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0519618427219652, "standard_deviation": 0.026206351830710163, "minimum": 1.0024113026296582, "maximum": 1.0811960932039453, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8342922511470775, "standard_deviation": 0.21007376633522384, "minimum": 0.5956297922548014, "maximum": 1.2356919620573656, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09021237269722099, "standard_deviation": 0.007310889441410532, "minimum": 0.08318567433848564, "maximum": 0.1048468013151132, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9777602634814893, "standard_deviation": 0.0028104581239911794, "minimum": 0.9729256897339836, "maximum": 0.9821418571940895, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7384314218624781, "standard_deviation": 0.041471614180735805, "minimum": 1.6697250358334947, "maximum": 1.7892989058621076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2098494214718687, "standard_deviation": 0.5153501126529306, "minimum": 0.25275342515076876, "maximum": 1.8660927789645807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1848014746925548, "standard_deviation": 0.013509736345354186, "minimum": 0.16397874243747376, "maximum": 0.20346482289224208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932024981853105, "standard_deviation": 0.0005737466736903833, "minimum": 0.9923203002914307, "maximum": 0.9937780635107276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9813269350016882, "standard_deviation": 0.04486573160495176, "minimum": 0.9234845301825987, "maximum": 1.0609641673405374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4455999109197677, "standard_deviation": 0.5851589447766389, "minimum": 0.3543378097435855, "maximum": 2.3118985257093008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10476015440560033, "standard_deviation": 0.010755721037929293, "minimum": 0.09379134035573723, "maximum": 0.12296697619832504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933370634473198, "standard_deviation": 0.0006543768781094722, "minimum": 0.9922826394249032, "maximum": 0.9942419320427013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6631003069484513, "standard_deviation": 0.03432097316852273, "minimum": 1.6166586989331004, "maximum": 1.7155502827647109, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7829652262525969, "standard_deviation": 0.18279629849418066, "minimum": 0.5204297901912172, "maximum": 1.0104025105719348, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09304143895942704, "standard_deviation": 0.0042414422190405615, "minimum": 0.0867437978725962, "maximum": 0.09997416454938854, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974762985353409, "standard_deviation": 0.0003412877475216624, "minimum": 0.9972090577992276, "maximum": 0.998045890601042, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.060398494200385, "standard_deviation": 0.0454375604003789, "minimum": 1.0096871371267209, "maximum": 1.1430368261866792, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.969569327433847, "standard_deviation": 0.20660125865573015, "minimum": 0.6845493512318299, "maximum": 1.3054668536286906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05577740793891292, "standard_deviation": 0.003783358335385848, "minimum": 0.04981755771920635, "maximum": 0.060132802599721404, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18139955429925714, "standard_deviation": 0.022376997356740484, "minimum": 0.15066799323781832, "maximum": 0.20860909199246633, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.580392502472608, "standard_deviation": 0.012742484714375005, "minimum": 5.564069416652063, "maximum": 5.600915579907678, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4275190331805813, "standard_deviation": 0.008441090205179282, "minimum": 0.4185160886560911, "maximum": 0.43970595689426595, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06607654926549388, "standard_deviation": 0.03877364844481727, "minimum": -0.13847893707842177, "maximum": -0.00846785947624804, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.995926441596267, "standard_deviation": 0.00037178951820568764, "minimum": 0.9952345251740584, "maximum": 0.9963538040581486, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9943485758305406, "standard_deviation": 0.028700392639190858, "minimum": 0.9533999091295668, "maximum": 1.0325333172288995, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5804378333782662, "standard_deviation": 0.13588568949127663, "minimum": 0.40002464062567583, "maximum": 0.7676634992602706, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06504676479971092, "standard_deviation": 0.00293601757446133, "minimum": 0.06130205587186639, "maximum": 0.06993711882586294, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987071335819269, "standard_deviation": 0.0001586424547658495, "minimum": 0.9984874383930238, "maximum": 0.9989352563253909, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5299754461241074, "standard_deviation": 0.02150439749125784, "minimum": 0.5023657216010641, "maximum": 0.5641483417106046, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7066924581441801, "standard_deviation": 0.15845203725122042, "minimum": 0.5486275089717344, "maximum": 0.9667312586046662, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03532940740401648, "standard_deviation": 0.0024672556976122082, "minimum": 0.03130746349456764, "maximum": 0.03854132049722019, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0001", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.495435953140259, "predicted_mos": 3.495435953140259, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "8b42c007d47dcf65eb574d70976f2ffab3068c70251e92f1af887ce0f9ed135e", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.301932699978352, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9860000920946966, "全帯域_10ms_rms_envelope_nrmse": 0.057647125621210805, "全帯域_10ms_envelope_correlation": 0.9967814761054598, "全帯域_10ms_modulation_mae_db": 0.3587871514362346, "全帯域_40ms_log_rms_mae_db": 0.5217924216328674, "全帯域_40ms_rms_envelope_nrmse": 0.031442970512624874, "全帯域_40ms_envelope_correlation": 0.998950507660503, "全帯域_40ms_modulation_mae_db": 0.44703418572897585, "300–1000 Hz_10ms_log_rms_mae_db": 1.7535803593485972, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08517691570466325, "300–1000 Hz_10ms_envelope_correlation": 0.9943842247478852, "300–1000 Hz_10ms_modulation_mae_db": 0.5798914881102365, "300–1000 Hz_40ms_log_rms_mae_db": 1.0804286125278342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04688369300285979, "300–1000 Hz_40ms_envelope_correlation": 0.9981585365828524, "300–1000 Hz_40ms_modulation_mae_db": 0.6680293803880123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.828248501962905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16094333669199545, "1000–3000 Hz_10ms_envelope_correlation": 0.9795291209755669, "1000–3000 Hz_10ms_modulation_mae_db": 0.9362120130737615, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0786281932787227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10528362996879952, "1000–3000 Hz_40ms_envelope_correlation": 0.9905828469457546, "1000–3000 Hz_40ms_modulation_mae_db": 1.1879290580478674, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7375122454913734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.164315190175749, "3000–6000 Hz_10ms_envelope_correlation": 0.9821209807254557, "3000–6000 Hz_10ms_modulation_mae_db": 1.09461105857107, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9764038431116264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09098171458022214, "3000–6000 Hz_40ms_envelope_correlation": 0.9940950355589688, "3000–6000 Hz_40ms_modulation_mae_db": 1.3938784251898568, "mr_stft_db": 5.571427369185703, "band_mae_db": 0.19417943312961963, "rms_gain_error_db": -0.04835445624196401, "output_peak": 0.4378421353720049}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0084043042606587, "全帯域_10ms_rms_envelope_nrmse": 0.06538043033788132, "全帯域_10ms_envelope_correlation": 0.9958663532799484, "全帯域_10ms_modulation_mae_db": 0.418516175055974, "全帯域_40ms_log_rms_mae_db": 0.5036468354060039, "全帯域_40ms_rms_envelope_nrmse": 0.033160857860209524, "全帯域_40ms_envelope_correlation": 0.9988489496802666, "全帯域_40ms_modulation_mae_db": 0.625543308627058, "300–1000 Hz_10ms_log_rms_mae_db": 1.6588357943940633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08856518572974523, "300–1000 Hz_10ms_envelope_correlation": 0.9939178312272526, "300–1000 Hz_10ms_modulation_mae_db": 0.45946818462245304, "300–1000 Hz_40ms_log_rms_mae_db": 1.071436018070495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05356597910168933, "300–1000 Hz_40ms_envelope_correlation": 0.9976000748325303, "300–1000 Hz_40ms_modulation_mae_db": 0.6284182732068533, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8133345259022575, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14375452779087913, "1000–3000 Hz_10ms_envelope_correlation": 0.9840612265072817, "1000–3000 Hz_10ms_modulation_mae_db": 0.6204050679318699, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0215060700674878, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08339291127342446, "1000–3000 Hz_40ms_envelope_correlation": 0.9946946944885675, "1000–3000 Hz_40ms_modulation_mae_db": 0.8116619363676297, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6952979419073464, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2192336473625916, "3000–6000 Hz_10ms_envelope_correlation": 0.9678218087350337, "3000–6000 Hz_10ms_modulation_mae_db": 0.98287318560849, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9364566121861431, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11310634757119713, "3000–6000 Hz_40ms_envelope_correlation": 0.9912458852644627, "3000–6000 Hz_40ms_modulation_mae_db": 0.9856546354886229, "mr_stft_db": 5.548240807596258, "band_mae_db": 0.16402739716494166, "rms_gain_error_db": -0.062324295536325514, "output_peak": 0.4891803471027956}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.004783468425056, "全帯域_10ms_rms_envelope_nrmse": 0.060406114877277546, "全帯域_10ms_envelope_correlation": 0.9965450799184273, "全帯域_10ms_modulation_mae_db": 0.6265088670204854, "全帯域_40ms_log_rms_mae_db": 0.5062298366837998, "全帯域_40ms_rms_envelope_nrmse": 0.031171220125868718, "全帯域_40ms_envelope_correlation": 0.9990615517252794, "全帯域_40ms_modulation_mae_db": 0.8869849699112925, "300–1000 Hz_10ms_log_rms_mae_db": 1.6201307703309107, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08582311297709501, "300–1000 Hz_10ms_envelope_correlation": 0.9945139443420057, "300–1000 Hz_10ms_modulation_mae_db": 0.7410268206431624, "300–1000 Hz_40ms_log_rms_mae_db": 0.9920990763021423, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05072419597906962, "300–1000 Hz_40ms_envelope_correlation": 0.9981444428577276, "300–1000 Hz_40ms_modulation_mae_db": 1.1095745759044822, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7961748068518426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1358205088411211, "1000–3000 Hz_10ms_envelope_correlation": 0.9854882983861905, "1000–3000 Hz_10ms_modulation_mae_db": 0.7813975500614149, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0194694155366595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07890199160734124, "1000–3000 Hz_40ms_envelope_correlation": 0.9948167687608863, "1000–3000 Hz_40ms_modulation_mae_db": 0.7829156488543844, "3000–6000 Hz_10ms_log_rms_mae_db": 1.741388373161109, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1827085365162033, "3000–6000 Hz_10ms_envelope_correlation": 0.9779044670634138, "3000–6000 Hz_10ms_modulation_mae_db": 1.31950346502604, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9626422637195285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09847754617293296, "3000–6000 Hz_40ms_envelope_correlation": 0.9934854425721339, "3000–6000 Hz_40ms_modulation_mae_db": 1.6084704714902618, "mr_stft_db": 5.555391509690338, "band_mae_db": 0.1456382690129449, "rms_gain_error_db": -0.09319739386815046, "output_peak": 0.42252117872996575}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0105538747715759, "全帯域_10ms_rms_envelope_nrmse": 0.06320775835558581, "全帯域_10ms_envelope_correlation": 0.9961125054628435, "全帯域_10ms_modulation_mae_db": 0.2629118985472492, "全帯域_40ms_log_rms_mae_db": 0.534297832450421, "全帯域_40ms_rms_envelope_nrmse": 0.034526171364271795, "全帯域_40ms_envelope_correlation": 0.9987020181267414, "全帯域_40ms_modulation_mae_db": 0.2826969577725593, "300–1000 Hz_10ms_log_rms_mae_db": 1.5779901975673416, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08272891098989092, "300–1000 Hz_10ms_envelope_correlation": 0.9946662865727646, "300–1000 Hz_10ms_modulation_mae_db": 0.3657651331565317, "300–1000 Hz_40ms_log_rms_mae_db": 1.008341255886663, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05017238754331121, "300–1000 Hz_40ms_envelope_correlation": 0.997840539806783, "300–1000 Hz_40ms_modulation_mae_db": 0.5680165116435055, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8152413420137776, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14171404544221983, "1000–3000 Hz_10ms_envelope_correlation": 0.9841740706618253, "1000–3000 Hz_10ms_modulation_mae_db": 0.3904403971523194, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0816518817796184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08752969478830032, "1000–3000 Hz_40ms_envelope_correlation": 0.9935548354161238, "1000–3000 Hz_40ms_modulation_mae_db": 0.3731164509081904, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7049154998008136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18810235885950838, "3000–6000 Hz_10ms_envelope_correlation": 0.9771721381517654, "3000–6000 Hz_10ms_modulation_mae_db": 1.3518349715505824, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9716590909006939, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10970841407404601, "3000–6000 Hz_40ms_envelope_correlation": 0.9918470045668891, "3000–6000 Hz_40ms_modulation_mae_db": 1.4599135291532999, "mr_stft_db": 5.552624228069589, "band_mae_db": 0.12708993081703998, "rms_gain_error_db": -0.00709106113669685, "output_peak": 0.44776099866701774}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.949852638308489, "全帯域_10ms_rms_envelope_nrmse": 0.061876714355188274, "全帯域_10ms_envelope_correlation": 0.9963226516641766, "全帯域_10ms_modulation_mae_db": 0.6921228609648037, "全帯域_40ms_log_rms_mae_db": 0.4829819211205767, "全帯域_40ms_rms_envelope_nrmse": 0.03398410193642042, "全帯域_40ms_envelope_correlation": 0.9988157932500459, "全帯域_40ms_modulation_mae_db": 0.7914983471563827, "300–1000 Hz_10ms_log_rms_mae_db": 1.6358100452226465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08631966591077553, "300–1000 Hz_10ms_envelope_correlation": 0.9942983226700318, "300–1000 Hz_10ms_modulation_mae_db": 1.008533588410597, "300–1000 Hz_40ms_log_rms_mae_db": 1.0201123480243175, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05192991573892377, "300–1000 Hz_40ms_envelope_correlation": 0.9978331746254885, "300–1000 Hz_40ms_modulation_mae_db": 1.1163851895222832, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8379232485516415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1492747301323825, "1000–3000 Hz_10ms_envelope_correlation": 0.9826128954088231, "1000–3000 Hz_10ms_modulation_mae_db": 0.8964182731889082, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0827678330050143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08870035633314431, "1000–3000 Hz_40ms_envelope_correlation": 0.9937261205283403, "1000–3000 Hz_40ms_modulation_mae_db": 0.9819864260047315, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7273933675164728, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18615281513144566, "3000–6000 Hz_10ms_envelope_correlation": 0.9778808914480266, "3000–6000 Hz_10ms_modulation_mae_db": 1.6027922234644822, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9557034635079578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09810036831095534, "3000–6000 Hz_40ms_envelope_correlation": 0.9949050677806469, "3000–6000 Hz_40ms_modulation_mae_db": 2.1908867028151886, "mr_stft_db": 5.5612770079990055, "band_mae_db": 0.17136572879626752, "rms_gain_error_db": -0.07752875535722846, "output_peak": 0.42939994281678534}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.030151928613748, "全帯域_10ms_rms_envelope_nrmse": 0.060447215488811816, "全帯域_10ms_envelope_correlation": 0.9964480180700523, "全帯域_10ms_modulation_mae_db": 0.39999669763255313, "全帯域_40ms_log_rms_mae_db": 0.5451286393525627, "全帯域_40ms_rms_envelope_nrmse": 0.029574902786398568, "全帯域_40ms_envelope_correlation": 0.9990555588486941, "全帯域_40ms_modulation_mae_db": 0.5489771984211312, "300–1000 Hz_10ms_log_rms_mae_db": 1.6047900353982794, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08157459958212877, "300–1000 Hz_10ms_envelope_correlation": 0.9948485727520711, "300–1000 Hz_10ms_modulation_mae_db": 0.5772814671930856, "300–1000 Hz_40ms_log_rms_mae_db": 1.0052241957669779, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0448504683641676, "300–1000 Hz_40ms_envelope_correlation": 0.9983167248662786, "300–1000 Hz_40ms_modulation_mae_db": 0.7429363150567903, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7845878246653133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15074186635154938, "1000–3000 Hz_10ms_envelope_correlation": 0.9819395401778616, "1000–3000 Hz_10ms_modulation_mae_db": 0.34858340905795665, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0252168412641491, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08459497199087677, "1000–3000 Hz_40ms_envelope_correlation": 0.993872226784526, "1000–3000 Hz_40ms_modulation_mae_db": 0.2986573301134669, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7729387719204168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22134762719479067, "3000–6000 Hz_10ms_envelope_correlation": 0.9682298260683714, "3000–6000 Hz_10ms_modulation_mae_db": 0.6964920858258172, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0055823292138075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11711557663983331, "3000–6000 Hz_40ms_envelope_correlation": 0.990635255362587, "3000–6000 Hz_40ms_modulation_mae_db": 0.7719810390288426, "mr_stft_db": 5.564206741030303, "band_mae_db": 0.12711213318701586, "rms_gain_error_db": -0.023629981610632844, "output_peak": 0.47241829488778075}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.02026180564303, "全帯域_10ms_rms_envelope_nrmse": 0.06701257089962391, "全帯域_10ms_envelope_correlation": 0.9956776778253343, "全帯域_10ms_modulation_mae_db": 0.371231504279441, "全帯域_40ms_log_rms_mae_db": 0.535202687140183, "全帯域_40ms_rms_envelope_nrmse": 0.03452519529392226, "全帯域_40ms_envelope_correlation": 0.9987660179840612, "全帯域_40ms_modulation_mae_db": 0.5767816767022572, "300–1000 Hz_10ms_log_rms_mae_db": 1.6793368442173153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.093163904873124, "300–1000 Hz_10ms_envelope_correlation": 0.9934659073275547, "300–1000 Hz_10ms_modulation_mae_db": 0.5294244799862038, "300–1000 Hz_40ms_log_rms_mae_db": 1.0553953971565564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056207656502011705, "300–1000 Hz_40ms_envelope_correlation": 0.9976455554591512, "300–1000 Hz_40ms_modulation_mae_db": 0.7832665862418757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7550583263314992, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14886033962238854, "1000–3000 Hz_10ms_envelope_correlation": 0.9824605423610362, "1000–3000 Hz_10ms_modulation_mae_db": 0.7780150381563646, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0055223722060425, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09601337681488122, "1000–3000 Hz_40ms_envelope_correlation": 0.9920901331653327, "1000–3000 Hz_40ms_modulation_mae_db": 1.0582961045839572, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7678497929944383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18947949712613807, "3000–6000 Hz_10ms_envelope_correlation": 0.9760396433168721, "3000–6000 Hz_10ms_modulation_mae_db": 0.7755553943024035, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9750040808009279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10821050525569721, "3000–6000 Hz_40ms_envelope_correlation": 0.9917753308912489, "3000–6000 Hz_40ms_modulation_mae_db": 0.8193209213568525, "mr_stft_db": 5.56943706912579, "band_mae_db": 0.17599404136136707, "rms_gain_error_db": -0.06742536796115092, "output_peak": 0.4321041967178291}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.985148316874137, "全帯域_10ms_rms_envelope_nrmse": 0.059455852675066576, "全帯域_10ms_envelope_correlation": 0.9965687661143791, "全帯域_10ms_modulation_mae_db": 0.3972768399285099, "全帯域_40ms_log_rms_mae_db": 0.5063057682405695, "全帯域_40ms_rms_envelope_nrmse": 0.02926992402211109, "全帯域_40ms_envelope_correlation": 0.9990925769330237, "全帯域_40ms_modulation_mae_db": 0.49126909444267897, "300–1000 Hz_10ms_log_rms_mae_db": 1.6397003198468132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08820971634332897, "300–1000 Hz_10ms_envelope_correlation": 0.9939661273145606, "300–1000 Hz_10ms_modulation_mae_db": 0.6555415935542337, "300–1000 Hz_40ms_log_rms_mae_db": 1.061711190700565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048901946855981, "300–1000 Hz_40ms_envelope_correlation": 0.9980454888725114, "300–1000 Hz_40ms_modulation_mae_db": 0.9698348385602982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8216944716594774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14143241641998705, "1000–3000 Hz_10ms_envelope_correlation": 0.9841712302766273, "1000–3000 Hz_10ms_modulation_mae_db": 0.3567810469566181, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0185103245873717, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08067604706639678, "1000–3000 Hz_40ms_envelope_correlation": 0.9944453412069331, "1000–3000 Hz_40ms_modulation_mae_db": 0.34719484223500174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6947420501865105, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19313684312950813, "3000–6000 Hz_10ms_envelope_correlation": 0.976733079221622, "3000–6000 Hz_10ms_modulation_mae_db": 0.17924627156157372, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9422486374106113, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1168198497689154, "3000–6000 Hz_40ms_envelope_correlation": 0.9913440711541033, "3000–6000 Hz_40ms_modulation_mae_db": 0.33733306795788115, "mr_stft_db": 5.571605292372961, "band_mae_db": 0.19724112023488816, "rms_gain_error_db": -0.04655615538680636, "output_peak": 0.46779481483355057}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 90800, "output_samples": 90800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.884226736872538, "spectral_convergence": 0.17125859335786534, "spectral_convergence_status": "ok", "amplitude_floor": 1.4038593411999917e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.423680793837812, "spectral_convergence": 0.16844563731821532, "spectral_convergence_status": "ok", "amplitude_floor": 9.854110842773577e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.40637457684676, "spectral_convergence": 0.1641165025964664, "spectral_convergence_status": "ok", "amplitude_floor": 8.983114295610284e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0018717561089627995, "output_power": 0.001884423033389333, "delta_db": 0.02929144764681599, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00201441754378434, "output_power": 0.0019669063688809297, "delta_db": -0.10365808512505087, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010569467904903444, "output_power": 0.0010455604304777622, "delta_db": -0.04703985430334136, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018132930505040446, "output_power": 0.00017712450182281927, "delta_db": -0.10189355339820486, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.873439242518033e-05, "output_power": 6.893962479308353e-05, "delta_db": -0.5769556507651661, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.113111558995142e-05, "output_power": 3.3405539095286054e-05, "delta_db": 0.3062380075391387, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19417943312961963, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04835445624196401, "pitch": null, "output_rms_dbfs": -22.85975569515668, "output_peak": 0.4378421353720049, "reference_rms_dbfs": -22.81140123891472, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.571427369185703}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 90800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.0001853404348361204, "log_rms_mae_db": 0.9860000920946966, "rms_envelope_nrmse": 0.057647125621210805, "envelope_correlation": 0.9967814761054598, "rms_gain_error_db": -0.04835431459148093, "reference_mean_power": 0.005238930533969212, "output_mean_power": 0.005180923843501813}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 2.4623434734036186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012515931525511833, "output_power": 0.0012233510736533977, "reference_active": true, "delta_db": -0.09912070999497889}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003238300007472361, "output_power": 0.0003172200540786113, "reference_active": true, "delta_db": -0.08956446019646988}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.578860153055394e-05, "output_power": 6.177834433935858e-05, "reference_active": true, "delta_db": -0.887676284117255}], "active_band_count": 3, "active_band_mae_db": 0.3587871514362346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00016148298564882675, "log_rms_mae_db": 0.5217924216328674, "rms_envelope_nrmse": 0.031442970512624874, "envelope_correlation": 0.998950507660503, "rms_gain_error_db": -0.04835451734029137, "reference_mean_power": 0.005266794777733929, "output_mean_power": 0.0052084793245467395}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.1218089587296202e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001164760758537026, "output_power": 0.0011419160945010736, "reference_active": true, "delta_db": -0.08602536510609415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015381285066774075, "output_power": 0.00014865968635678015, "reference_active": true, "delta_db": -0.14799408976821088}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.591871326154896e-06, "output_power": 4.333600364680646e-06, "reference_active": true, "delta_db": -1.1070831023126224}], "active_band_count": 3, "active_band_mae_db": 0.44703418572897585}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.00014972999954250175, "log_rms_mae_db": 1.7535803593485972, "rms_envelope_nrmse": 0.08517691570466325, "envelope_correlation": 0.9943842247478852, "rms_gain_error_db": -0.054812906219789376, "reference_mean_power": 0.0025001891623953804, "output_mean_power": 0.002468832226114817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.4970716182960256e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007649864504896549, "output_power": 0.0007333532742159828, "reference_active": true, "delta_db": -0.1834050763146246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022996394368157333, "output_power": 0.00021790160035181253, "reference_active": true, "delta_db": -0.2339932792935075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.80930068112851e-05, "output_power": 3.546943899462866e-05, "reference_active": true, "delta_db": -1.3222761087225772}], "active_band_count": 3, "active_band_mae_db": 0.5798914881102365}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00013636056906360855, "log_rms_mae_db": 1.0804286125278342, "rms_envelope_nrmse": 0.04688369300285979, "envelope_correlation": 0.9981585365828524, "rms_gain_error_db": -0.05481339209150328, "reference_mean_power": 0.0025134871346063777, "output_mean_power": 0.0024819631398060047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3161092414230216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215738253307795, "output_power": 0.0006950053901428546, "reference_active": true, "delta_db": -0.16292598248970708}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011225926953226224, "output_power": 0.00010425720097620074, "reference_active": true, "delta_db": -0.32116150835963286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.841934016741924e-06, "output_power": 2.7073838676221196e-06, "reference_active": true, "delta_db": -1.520000650314697}], "active_band_count": 3, "active_band_mae_db": 0.6680293803880123}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.2746042839695326e-05, "log_rms_mae_db": 1.828248501962905, "rms_envelope_nrmse": 0.16094333669199545, "envelope_correlation": 0.9795291209755669, "rms_gain_error_db": -0.22594913237451383, "reference_mean_power": 0.00020292459461300377, "output_mean_power": 0.00019263702997804723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.5147206216944168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.937659859611645e-05, "output_power": 6.691714552352949e-05, "reference_active": true, "delta_db": -0.15675596348995094}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3925889957061755e-05, "output_power": 2.0404715897478485e-05, "reference_active": true, "delta_db": -0.6913754863596742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8985142545041605e-06, "output_power": 3.755711070895803e-06, "reference_active": true, "delta_db": -1.9605045893716593}], "active_band_count": 3, "active_band_mae_db": 0.9362120130737615}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.512933910060942e-05, "log_rms_mae_db": 1.0786281932787227, "rms_envelope_nrmse": 0.10528362996879952, "envelope_correlation": 0.9905828469457546, "rms_gain_error_db": -0.22595022687506877, "reference_mean_power": 0.00020400334148028423, "output_mean_power": 0.000193661039359719}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3444198629453317e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.558296247449876e-05, "output_power": 6.282905516019751e-05, "reference_active": true, "delta_db": -0.1863050173719644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3197503778987225e-05, "output_power": 1.1346922354216781e-05, "reference_active": true, "delta_db": -0.6561371198738029}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.991435827054748e-07, "output_power": 2.6674174517219264e-07, "reference_active": true, "delta_db": -2.721345036897835}], "active_band_count": 3, "active_band_mae_db": 1.1879290580478674}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.624711907970004e-05, "log_rms_mae_db": 1.7375122454913734, "rms_envelope_nrmse": 0.164315190175749, "envelope_correlation": 0.9821209807254557, "rms_gain_error_db": -0.026152957895734987, "reference_mean_power": 5.558522396873647e-05, "output_mean_power": 5.5251498868988115e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 3.230968975305017e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3093182032279326e-05, "output_power": 1.3042769935091255e-05, "reference_active": true, "delta_db": -0.016753721317400817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8826443460154885e-06, "output_power": 6.079400299275972e-06, "reference_active": true, "delta_db": -0.5389458785736002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3224614880429907e-06, "output_power": 1.7727463423577392e-06, "reference_active": true, "delta_db": -2.728133575822209}], "active_band_count": 3, "active_band_mae_db": 1.09461105857107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.3621221947891856e-05, "log_rms_mae_db": 0.9764038431116264, "rms_envelope_nrmse": 0.09098171458022214, "envelope_correlation": 0.9940950355589688, "rms_gain_error_db": -0.02615311627454252, "reference_mean_power": 5.588085792782989e-05, "output_mean_power": 5.5545355861819e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.891660816172807e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3816996130233874e-05, "output_power": 1.331294273181907e-05, "reference_active": true, "delta_db": -0.16139572147775366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.814539997745632e-06, "output_power": 3.137994163229162e-06, "reference_active": true, "delta_db": -0.84790041644199}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.118413559303936e-07, "output_power": 1.5021032867080506e-07, "reference_active": true, "delta_db": -3.1723391376498276}], "active_band_count": 3, "active_band_mae_db": 1.3938784251898568}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.910942902255947, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "18c2d8d9ed6472183428886019616862895267f9f0979e8b1fd697d3cc11e35f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9830546155944015, "standard_deviation": 0.0018383467214738004, "minimum": 0.9795291209755669, "maximum": 0.9854882983861905, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8065328809923393, "standard_deviation": 0.026852512100291003, "minimum": 1.7550583263314992, "maximum": 1.8379232485516415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6385315994474017, "standard_deviation": 0.24504128439600895, "minimum": 0.34858340905795665, "maximum": 0.9362120130737615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14656772141156538, "standard_deviation": 0.00765330889503566, "minimum": 0.1358205088411211, "maximum": 0.16094333669199545, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.993472870912058, "standard_deviation": 0.001451066288610349, "minimum": 0.9905828469457546, "maximum": 0.9948167687608863, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0416591164656333, "standard_deviation": 0.03309438421229577, "minimum": 1.0055223722060425, "maximum": 1.0827678330050143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7302197246394037, "standard_deviation": 0.3485490373744507, "minimum": 0.2986573301134669, "maximum": 1.1879290580478674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08813662248039558, "standard_deviation": 0.008723809446467834, "minimum": 0.07890199160734124, "maximum": 0.10528362996879952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9754878543413201, "standard_deviation": 0.004952692008960584, "minimum": 0.9678218087350337, "maximum": 0.9821209807254557, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.73025475537231, "standard_deviation": 0.030578557335400398, "minimum": 1.6947420501865105, "maximum": 1.7729387719204168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0003635819888075, "standard_deviation": 0.44953967192130934, "minimum": 0.17924627156157372, "maximum": 1.6027922234644822, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19305956443699185, "standard_deviation": 0.01891250481758541, "minimum": 0.164315190175749, "maximum": 0.22134762719479067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9924166366438801, "standard_deviation": 0.001539099324557635, "minimum": 0.990635255362587, "maximum": 0.9949050677806469, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.965712540106412, "standard_deviation": 0.021849061757835204, "minimum": 0.9364566121861431, "maximum": 1.0055823292138075, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1959298490601007, "standard_deviation": 0.5821108024190887, "minimum": 0.33733306795788115, "maximum": 2.1908867028151886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10656504029672495, "standard_deviation": 0.009649206202348281, "minimum": 0.09098171458022214, "maximum": 0.11711557663983331, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9942576521192659, "standard_deviation": 0.00045148240021418266, "minimum": 0.9934659073275547, "maximum": 0.9948485727520711, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6462717957907458, "standard_deviation": 0.05343270552162912, "minimum": 1.5779901975673416, "maximum": 1.7535803593485972, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.614616594459563, "standard_deviation": 0.1959156228486286, "minimum": 0.3657651331565317, "maximum": 1.008533588410597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08644525151384397, "standard_deviation": 0.003631089980664131, "minimum": 0.08157459958212877, "maximum": 0.093163904873124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979480672379154, "standard_deviation": 0.0002579132474262212, "minimum": 0.9976000748325303, "maximum": 0.9983167248662786, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.036843511804444, "standard_deviation": 0.034127978650892135, "minimum": 0.9920990763021423, "maximum": 1.0804286125278342, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8233077088155125, "standard_deviation": 0.2153545342047786, "minimum": 0.5680165116435055, "maximum": 1.1163851895222832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05040453038600175, "standard_deviation": 0.003618515197693114, "minimum": 0.0448504683641676, "maximum": 0.056207656502011705, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1628310067130106, "standard_deviation": 0.02742045827866118, "minimum": 0.12708993081703998, "maximum": 0.19724112023488816, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.561776253133743, "standard_deviation": 0.008962871587910527, "minimum": 5.548240807596258, "maximum": 5.571605292372961, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44987773864096625, "standard_deviation": 0.023924733005676067, "minimum": 0.42252117872996575, "maximum": 0.4891803471027956, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05326343338736943, "standard_deviation": 0.028145798270499983, "minimum": -0.09319739386815046, "maximum": -0.00709106113669685, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962903160550777, "standard_deviation": 0.0003772606934538561, "minimum": 0.9956776778253343, "maximum": 0.9967814761054598, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9993945536239239, "standard_deviation": 0.025221796172445285, "minimum": 0.949852638308489, "maximum": 1.030151928613748, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.44091899935815637, "standard_deviation": 0.14388321346169758, "minimum": 0.2629118985472492, "maximum": 0.6921228609648037, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06192922282633076, "standard_deviation": 0.003124986653990593, "minimum": 0.057647125621210805, "maximum": 0.06701257089962391, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998911621776077, "standard_deviation": 0.00014913982098945308, "minimum": 0.9987020181267414, "maximum": 0.9990925769330237, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.516948242753373, "standard_deviation": 0.0207385726163397, "minimum": 0.4829819211205767, "maximum": 0.5451286393525627, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.581348217345292, "standard_deviation": 0.19121637935971084, "minimum": 0.2826969577725593, "maximum": 0.8869849699112925, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0322069179877284, "standard_deviation": 0.002139140459580532, "minimum": 0.02926992402211109, "maximum": 0.034526171364271795, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3570-5694-0004", "language": "en", "style": "normal", "speaker": "librispeech_3570", "source_utterance_id": "3570-5694-0004", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 85280, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3570-5694-0004.wav", "source_file_sha256": "4f1814ba0708b2bb8d0a4d00cc5fdc294016356dffecbd88443a6fa51b760974", "decoded_pcm_sha256": "86a8946d8572399656832193f585d37dfcbd327696eb662dcf0e41d4b67d0bd1", "original_source_sha256": "e7c4c68060bd5da1df857aa90f0fb6601cfee83e0001d75b64164988b070e7d8", "corpus_entry_id": "librispeech_3570-5694-0004"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.262558460235596, "predicted_mos": 4.262558460235596, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "6f2bf49f577cb0bba65458633c45d569de3c13d910c149cbde33dce48bcdec04", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.125573397002881, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3570-5694-0004", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6f2bf49f577cb0bba65458633c45d569de3c13d910c149cbde33dce48bcdec04", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "2a7d7164eedcdf7707de438f724ea4169c4a77e4b474dae06130777a0dc42808"}, "rows": [{"case": "librispeech_3570-5694-0004", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.359541654586792, "predicted_mos": 3.359541654586792, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "0fc22309155e512934119db247aa4949138ee95ba0d13ec7da1838dcb1fda802", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.254667147000873, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3570-5694-0004", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "0fc22309155e512934119db247aa4949138ee95ba0d13ec7da1838dcb1fda802", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.900974052150036, "全帯域_10ms_rms_envelope_nrmse": 0.0634350808819122, "全帯域_10ms_envelope_correlation": 0.9962622427874954, "全帯域_10ms_modulation_mae_db": 0.5846739688947311, "全帯域_40ms_log_rms_mae_db": 0.4584284453695713, "全帯域_40ms_rms_envelope_nrmse": 0.03363136067740666, "全帯域_40ms_envelope_correlation": 0.9988762951155525, "全帯域_40ms_modulation_mae_db": 0.4993058858627633, "300–1000 Hz_10ms_log_rms_mae_db": 1.7475347358934208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1049218587306918, "300–1000 Hz_10ms_envelope_correlation": 0.9912571656802142, "300–1000 Hz_10ms_modulation_mae_db": 0.5309598885305203, "300–1000 Hz_40ms_log_rms_mae_db": 0.970246114127513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05768371578233098, "300–1000 Hz_40ms_envelope_correlation": 0.9971621790469066, "300–1000 Hz_40ms_modulation_mae_db": 0.6406404605413089, "1000–3000 Hz_10ms_log_rms_mae_db": 1.519712988467698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13682435831237447, "1000–3000 Hz_10ms_envelope_correlation": 0.9861342223405198, "1000–3000 Hz_10ms_modulation_mae_db": 0.965980787265673, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7984134810015721, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07984494454100839, "1000–3000 Hz_40ms_envelope_correlation": 0.9952368273050599, "1000–3000 Hz_40ms_modulation_mae_db": 1.1226153118126245, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6405122360589726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17163384496659373, "3000–6000 Hz_10ms_envelope_correlation": 0.9800855784567688, "3000–6000 Hz_10ms_modulation_mae_db": 0.9162112400737707, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8853376553269656, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08706746631364179, "3000–6000 Hz_40ms_envelope_correlation": 0.9945466251642854, "3000–6000 Hz_40ms_modulation_mae_db": 0.9994301516941656, "mr_stft_db": 5.510532554158478, "band_mae_db": 0.11900138025949636, "rms_gain_error_db": -0.04580766519206761, "output_peak": 0.5818326519878171}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8727459046769872, "全帯域_10ms_rms_envelope_nrmse": 0.06665788318722088, "全帯域_10ms_envelope_correlation": 0.9958571747730117, "全帯域_10ms_modulation_mae_db": 0.4762932866397094, "全帯域_40ms_log_rms_mae_db": 0.4732207387249143, "全帯域_40ms_rms_envelope_nrmse": 0.037558816579471195, "全帯域_40ms_envelope_correlation": 0.9985726975910421, "全帯域_40ms_modulation_mae_db": 0.376414813459723, "300–1000 Hz_10ms_log_rms_mae_db": 1.5611567404006774, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09757552684903487, "300–1000 Hz_10ms_envelope_correlation": 0.992476681959906, "300–1000 Hz_10ms_modulation_mae_db": 0.5453830084389081, "300–1000 Hz_40ms_log_rms_mae_db": 0.9433010240738071, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05526346043865184, "300–1000 Hz_40ms_envelope_correlation": 0.9974351790374658, "300–1000 Hz_40ms_modulation_mae_db": 0.5224328747177702, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5466145002764426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15313319489406796, "1000–3000 Hz_10ms_envelope_correlation": 0.9827377890613022, "1000–3000 Hz_10ms_modulation_mae_db": 0.7014938164243908, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8470601143596979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09043061960378738, "1000–3000 Hz_40ms_envelope_correlation": 0.9941309715380424, "1000–3000 Hz_40ms_modulation_mae_db": 0.6968001502558719, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6651819881239907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18151544179504797, "3000–6000 Hz_10ms_envelope_correlation": 0.9780353745803102, "3000–6000 Hz_10ms_modulation_mae_db": 1.1778854789319377, "3000–6000 Hz_40ms_log_rms_mae_db": 0.91380193622823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09936775579743579, "3000–6000 Hz_40ms_envelope_correlation": 0.9936853619972158, "3000–6000 Hz_40ms_modulation_mae_db": 1.5938521651156223, "mr_stft_db": 5.528867901155377, "band_mae_db": 0.1828937673064789, "rms_gain_error_db": -0.018598973261392968, "output_peak": 0.5270545346506824}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.908359578621244, "全帯域_10ms_rms_envelope_nrmse": 0.06387578288098172, "全帯域_10ms_envelope_correlation": 0.9962362812990884, "全帯域_10ms_modulation_mae_db": 0.398627578244972, "全帯域_40ms_log_rms_mae_db": 0.4763258678873479, "全帯域_40ms_rms_envelope_nrmse": 0.03296697506380035, "全帯域_40ms_envelope_correlation": 0.9989539550988886, "全帯域_40ms_modulation_mae_db": 0.3595099027327889, "300–1000 Hz_10ms_log_rms_mae_db": 1.6013144223754125, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10401663708889552, "300–1000 Hz_10ms_envelope_correlation": 0.991735010385447, "300–1000 Hz_10ms_modulation_mae_db": 0.8358583186482456, "300–1000 Hz_40ms_log_rms_mae_db": 0.9430695623041137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05843278719162179, "300–1000 Hz_40ms_envelope_correlation": 0.9975215673974459, "300–1000 Hz_40ms_modulation_mae_db": 0.7415541131541028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.606748876126176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15994582702884577, "1000–3000 Hz_10ms_envelope_correlation": 0.9807272887440407, "1000–3000 Hz_10ms_modulation_mae_db": 0.18263467987346957, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8833693861913494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08056865616502944, "1000–3000 Hz_40ms_envelope_correlation": 0.9946662787481306, "1000–3000 Hz_40ms_modulation_mae_db": 0.3867922510869479, "3000–6000 Hz_10ms_log_rms_mae_db": 1.600825425528921, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16478939653841285, "3000–6000 Hz_10ms_envelope_correlation": 0.9820277547487279, "3000–6000 Hz_10ms_modulation_mae_db": 0.3625340540500123, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8772230005993135, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08437318452464046, "3000–6000 Hz_40ms_envelope_correlation": 0.994980493793735, "3000–6000 Hz_40ms_modulation_mae_db": 0.22694902564709896, "mr_stft_db": 5.5209311801930285, "band_mae_db": 0.10985099073461645, "rms_gain_error_db": -0.06133110067256632, "output_peak": 0.6213617236907873}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.898694573859722, "全帯域_10ms_rms_envelope_nrmse": 0.06513720533859405, "全帯域_10ms_envelope_correlation": 0.9961045699586872, "全帯域_10ms_modulation_mae_db": 0.7531708720587414, "全帯域_40ms_log_rms_mae_db": 0.45639583247272664, "全帯域_40ms_rms_envelope_nrmse": 0.03262905401754598, "全帯域_40ms_envelope_correlation": 0.999004676704526, "全帯域_40ms_modulation_mae_db": 0.9171989535557531, "300–1000 Hz_10ms_log_rms_mae_db": 1.6642762671725004, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10764245156972356, "300–1000 Hz_10ms_envelope_correlation": 0.9909037025627234, "300–1000 Hz_10ms_modulation_mae_db": 0.8418210087846679, "300–1000 Hz_40ms_log_rms_mae_db": 0.9940948432160414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06360830258949898, "300–1000 Hz_40ms_envelope_correlation": 0.9966825593555512, "300–1000 Hz_40ms_modulation_mae_db": 1.0356396532057446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5927397141869024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15881063530668038, "1000–3000 Hz_10ms_envelope_correlation": 0.9813801354257543, "1000–3000 Hz_10ms_modulation_mae_db": 0.5416410517926471, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8443286745671734, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09010551015726798, "1000–3000 Hz_40ms_envelope_correlation": 0.9942574587162215, "1000–3000 Hz_40ms_modulation_mae_db": 0.6596768795979899, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6382921983189842, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17399830943502792, "3000–6000 Hz_10ms_envelope_correlation": 0.9795606702359052, "3000–6000 Hz_10ms_modulation_mae_db": 0.634154977968941, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8954459797991381, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11009040510483194, "3000–6000 Hz_40ms_envelope_correlation": 0.991217326719091, "3000–6000 Hz_40ms_modulation_mae_db": 0.6844579362954574, "mr_stft_db": 5.510571316687677, "band_mae_db": 0.2144456189545978, "rms_gain_error_db": -0.08322639762512964, "output_peak": 0.5510743060633021}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8527870039787582, "全帯域_10ms_rms_envelope_nrmse": 0.057032403634346945, "全帯域_10ms_envelope_correlation": 0.9969764242627212, "全帯域_10ms_modulation_mae_db": 0.3260004155490311, "全帯域_40ms_log_rms_mae_db": 0.44342842930637344, "全帯域_40ms_rms_envelope_nrmse": 0.0272471452993882, "全帯域_40ms_envelope_correlation": 0.9992654140862679, "全帯域_40ms_modulation_mae_db": 0.45753534237971566, "300–1000 Hz_10ms_log_rms_mae_db": 1.6534191732575767, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0975873645953518, "300–1000 Hz_10ms_envelope_correlation": 0.9924390669246078, "300–1000 Hz_10ms_modulation_mae_db": 0.28643704341364756, "300–1000 Hz_40ms_log_rms_mae_db": 0.9297625349978488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052033715424612985, "300–1000 Hz_40ms_envelope_correlation": 0.9976531001531153, "300–1000 Hz_40ms_modulation_mae_db": 0.40268551854104456, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5767037066623044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14725337790838497, "1000–3000 Hz_10ms_envelope_correlation": 0.9836694355142049, "1000–3000 Hz_10ms_modulation_mae_db": 1.1719802051617347, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8628682124657125, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08034487495688081, "1000–3000 Hz_40ms_envelope_correlation": 0.9948814630147284, "1000–3000 Hz_40ms_modulation_mae_db": 1.3592886816229424, "3000–6000 Hz_10ms_log_rms_mae_db": 1.724364962828988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17822051817501802, "3000–6000 Hz_10ms_envelope_correlation": 0.978522717128023, "3000–6000 Hz_10ms_modulation_mae_db": 0.39590543257662186, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9387059341213396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10401844947412542, "3000–6000 Hz_40ms_envelope_correlation": 0.9921716007017699, "3000–6000 Hz_40ms_modulation_mae_db": 0.23017637233620514, "mr_stft_db": 5.518898946368577, "band_mae_db": 0.05621306416681454, "rms_gain_error_db": -0.03938606420619398, "output_peak": 0.5522771135583289}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8909980348806822, "全帯域_10ms_rms_envelope_nrmse": 0.05844763404024739, "全帯域_10ms_envelope_correlation": 0.9968816081046733, "全帯域_10ms_modulation_mae_db": 0.47756993087222765, "全帯域_40ms_log_rms_mae_db": 0.4506668202143675, "全帯域_40ms_rms_envelope_nrmse": 0.029328619131946697, "全帯域_40ms_envelope_correlation": 0.9992182704954501, "全帯域_40ms_modulation_mae_db": 0.48755536823732054, "300–1000 Hz_10ms_log_rms_mae_db": 1.6750338915311995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09482334839319811, "300–1000 Hz_10ms_envelope_correlation": 0.9931019020380737, "300–1000 Hz_10ms_modulation_mae_db": 0.7560145042407784, "300–1000 Hz_40ms_log_rms_mae_db": 1.0051447463170613, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05267008930497546, "300–1000 Hz_40ms_envelope_correlation": 0.9979311782243276, "300–1000 Hz_40ms_modulation_mae_db": 0.8283754928549615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5604244663866167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14429148226985314, "1000–3000 Hz_10ms_envelope_correlation": 0.9850301675252396, "1000–3000 Hz_10ms_modulation_mae_db": 1.1078197263262954, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8596946607591552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0851247703302672, "1000–3000 Hz_40ms_envelope_correlation": 0.9952245208862286, "1000–3000 Hz_40ms_modulation_mae_db": 1.2447413687569238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6562344641123072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1908274351215163, "3000–6000 Hz_10ms_envelope_correlation": 0.9762518387660255, "3000–6000 Hz_10ms_modulation_mae_db": 0.9893130069104566, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8849390632350453, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10081618704959296, "3000–6000 Hz_40ms_envelope_correlation": 0.9944869966566687, "3000–6000 Hz_40ms_modulation_mae_db": 0.8964712356022758, "mr_stft_db": 5.5274602168898275, "band_mae_db": 0.19978732916807132, "rms_gain_error_db": -0.0855628098138842, "output_peak": 0.6024516744181848}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8770468149622815, "全帯域_10ms_rms_envelope_nrmse": 0.058614967942935144, "全帯域_10ms_envelope_correlation": 0.9968880993079695, "全帯域_10ms_modulation_mae_db": 0.6652760144022359, "全帯域_40ms_log_rms_mae_db": 0.44609140553820614, "全帯域_40ms_rms_envelope_nrmse": 0.033265138087779304, "全帯域_40ms_envelope_correlation": 0.9989895489303463, "全帯域_40ms_modulation_mae_db": 0.8443312525821111, "300–1000 Hz_10ms_log_rms_mae_db": 1.6900134793971897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10264864720956318, "300–1000 Hz_10ms_envelope_correlation": 0.9918902072182281, "300–1000 Hz_10ms_modulation_mae_db": 0.9981328439263821, "300–1000 Hz_40ms_log_rms_mae_db": 0.9760966333312008, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05936391366528178, "300–1000 Hz_40ms_envelope_correlation": 0.997321911015157, "300–1000 Hz_40ms_modulation_mae_db": 1.0532187193991527, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5888032500489528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.155617294438618, "1000–3000 Hz_10ms_envelope_correlation": 0.9832057243907344, "1000–3000 Hz_10ms_modulation_mae_db": 1.633667816736314, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8783777227871978, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10269230271079069, "1000–3000 Hz_40ms_envelope_correlation": 0.993454440546013, "1000–3000 Hz_40ms_modulation_mae_db": 1.9612573009892849, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6609853827670416, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19679349291284426, "3000–6000 Hz_10ms_envelope_correlation": 0.9737640195375571, "3000–6000 Hz_10ms_modulation_mae_db": 0.5738833190985467, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9336415775265778, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10390784036653566, "3000–6000 Hz_40ms_envelope_correlation": 0.9922162195936416, "3000–6000 Hz_40ms_modulation_mae_db": 0.7943456620412975, "mr_stft_db": 5.516817222507146, "band_mae_db": 0.21173740498273383, "rms_gain_error_db": -0.09423239039528165, "output_peak": 0.5670829249495991}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.935464269185758, "全帯域_10ms_rms_envelope_nrmse": 0.06552769561091797, "全帯域_10ms_envelope_correlation": 0.9959986733697446, "全帯域_10ms_modulation_mae_db": 0.33625476976735685, "全帯域_40ms_log_rms_mae_db": 0.5189087018347968, "全帯域_40ms_rms_envelope_nrmse": 0.03829694255850706, "全帯域_40ms_envelope_correlation": 0.9985137858974709, "全帯域_40ms_modulation_mae_db": 0.07298737842393314, "300–1000 Hz_10ms_log_rms_mae_db": 1.6506353778688532, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09776472918065561, "300–1000 Hz_10ms_envelope_correlation": 0.9923992170113043, "300–1000 Hz_10ms_modulation_mae_db": 0.5938779708029691, "300–1000 Hz_40ms_log_rms_mae_db": 0.9468669793869746, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05808076070346655, "300–1000 Hz_40ms_envelope_correlation": 0.9970511307231182, "300–1000 Hz_40ms_modulation_mae_db": 0.6494987150110858, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5634267975980447, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15928346292441492, "1000–3000 Hz_10ms_envelope_correlation": 0.9813196644814571, "1000–3000 Hz_10ms_modulation_mae_db": 0.6065193453710244, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8441719737113909, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10692420603938167, "1000–3000 Hz_40ms_envelope_correlation": 0.9915305780728608, "1000–3000 Hz_40ms_modulation_mae_db": 0.6470311975723623, "3000–6000 Hz_10ms_log_rms_mae_db": 1.630146237968339, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1744310259301604, "3000–6000 Hz_10ms_envelope_correlation": 0.9795640268926875, "3000–6000 Hz_10ms_modulation_mae_db": 1.0108281023225394, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8928916095785331, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0936520286921822, "3000–6000 Hz_40ms_envelope_correlation": 0.993690566603426, "3000–6000 Hz_40ms_modulation_mae_db": 1.1019180438575833, "mr_stft_db": 5.514030746038916, "band_mae_db": 0.13553786182160846, "rms_gain_error_db": 7.820856044590249e-05, "output_peak": 0.5876217731732474}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85280, "output_samples": 85280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.617976072018027, "spectral_convergence": 0.18396976620953312, "spectral_convergence_status": "ok", "amplitude_floor": 1.2882929606468862e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.302600878683929, "spectral_convergence": 0.1858719740329127, "spectral_convergence_status": "ok", "amplitude_floor": 9.933866418820512e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.611020711773478, "spectral_convergence": 0.18207497159208577, "spectral_convergence_status": "ok", "amplitude_floor": 8.99033450142914e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001234751326260816, "output_power": 0.0012458568997074143, "delta_db": 0.03888660401428644, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018448232014630782, "output_power": 0.0018218895142201835, "delta_db": -0.05432715607274722, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006193820215877917, "output_power": 0.0006047957756803921, "delta_db": -0.10349846125818989, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017436605640499536, "output_power": 0.00016222729008584318, "delta_db": -0.31338031724743926, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010535962705358282, "output_power": 0.00010181638156244087, "delta_db": -0.14856566016669792, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.506722876313705e-05, "output_power": 1.526048690035178e-05, "delta_db": 0.0553500827976174, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11900138025949636, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04580766519206761, "pitch": null, "output_rms_dbfs": -24.031999495714956, "output_peak": 0.5818326519878171, "reference_rms_dbfs": -23.986191830522884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.510532554158478}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.0001748275799785616, "log_rms_mae_db": 0.900974052150036, "rms_envelope_nrmse": 0.0634350808819122, "envelope_correlation": 0.9962622427874954, "rms_gain_error_db": -0.04580762608470082, "reference_mean_power": 0.003997498793754849, "output_mean_power": 0.003955556377351864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.9383161701154944e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012108392795166332, "output_power": 0.0011956478549148225, "reference_active": true, "delta_db": -0.05483211958472968}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002231683599850243, "output_power": 0.00020564343933268846, "reference_active": true, "delta_db": -0.3551776318831237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.764859371015312e-05, "output_power": 4.964294726369302e-05, "reference_active": true, "delta_db": -1.34401215521634}], "active_band_count": 3, "active_band_mae_db": 0.5846739688947311}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00017019095451879833, "log_rms_mae_db": 0.4584284453695713, "rms_envelope_nrmse": 0.03363136067740666, "envelope_correlation": 0.9988762951155525, "rms_gain_error_db": -0.04580809533326298, "reference_mean_power": 0.00402014109400192, "output_mean_power": 0.0039779606810391496}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.6741447857886242e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011176954288226882, "output_power": 0.0011082063987391676, "reference_active": true, "delta_db": -0.03702821316974024}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.194819962775497e-05, "output_power": 8.487153780551318e-05, "reference_active": true, "delta_db": -0.3478115872386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3053605179616555e-06, "output_power": 2.5580619009212644e-06, "reference_active": true, "delta_db": -1.1130778571799498}], "active_band_count": 3, "active_band_mae_db": 0.4993058858627633}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.00014662911909855806, "log_rms_mae_db": 1.7475347358934208, "rms_envelope_nrmse": 0.1049218587306918, "envelope_correlation": 0.9912571656802142, "rms_gain_error_db": -0.09190972912588745, "reference_mean_power": 0.0015423797799867744, "output_mean_power": 0.0015100813720068831}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.0119951767590758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000652626948773928, "output_power": 0.0006338676791649244, "reference_active": true, "delta_db": -0.12666395329637972}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016341376588393177, "output_power": 0.0001498213840478041, "reference_active": true, "delta_db": -0.3771483370111662}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.788084498799748e-05, "output_power": 3.726104825173356e-05, "reference_active": true, "delta_db": -1.0890673752840152}], "active_band_count": 3, "active_band_mae_db": 0.5309598885305203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00013732417437068112, "log_rms_mae_db": 0.970246114127513, "rms_envelope_nrmse": 0.05768371578233098, "envelope_correlation": 0.9971621790469066, "rms_gain_error_db": -0.09191120975168444, "reference_mean_power": 0.0015511172002359399, "output_mean_power": 0.0015186353074071708}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 8.798155816797883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006408046711327789, "output_power": 0.0006195246016712395, "reference_active": true, "delta_db": -0.14667111735360175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.486060187757687e-05, "output_power": 6.787376589326158e-05, "reference_active": true, "delta_db": -0.4255136834990513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5748897665955177e-06, "output_power": 2.6199258297500173e-06, "reference_active": true, "delta_db": -1.3497365807712736}], "active_band_count": 3, "active_band_mae_db": 0.6406404605413089}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 8.659636175169924e-05, "log_rms_mae_db": 1.519712988467698, "rms_envelope_nrmse": 0.13682435831237447, "envelope_correlation": 0.9861342223405198, "rms_gain_error_db": -0.2383737628175317, "reference_mean_power": 0.00020984803844209803, "output_mean_power": 0.00019864038016254146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 2.0976407038326697e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011684508155422031, "output_power": 0.00011726238283238158, "reference_active": true, "delta_db": 0.0154827899075318}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.122844755814475e-05, "output_power": 3.3636388820405094e-05, "reference_active": true, "delta_db": -0.883876178407541}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.871758765973707e-06, "output_power": 5.5995275398321146e-06, "reference_active": true, "delta_db": -1.9985833934819464}], "active_band_count": 3, "active_band_mae_db": 0.965980787265673}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 7.401841529112696e-05, "log_rms_mae_db": 0.7984134810015721, "rms_envelope_nrmse": 0.07984494454100839, "envelope_correlation": 0.9952368273050599, "rms_gain_error_db": -0.2383746660948978, "reference_mean_power": 0.000211036168686384, "output_mean_power": 0.0001997650126645632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.8857384691636124e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011739951182146869, "output_power": 0.00011701587131745604, "reference_active": true, "delta_db": -0.01421520212956739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97467891176206e-05, "output_power": 1.575978714614515e-05, "reference_active": true, "delta_db": -0.9794614067036943}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.379135569148535e-07, "output_power": 3.6927074211430404e-07, "reference_active": true, "delta_db": -2.3741693266046116}], "active_band_count": 3, "active_band_mae_db": 1.1226153118126245}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 4.122543101057287e-05, "log_rms_mae_db": 1.6405122360589726, "rms_envelope_nrmse": 0.17163384496659373, "envelope_correlation": 0.9800855784567688, "rms_gain_error_db": -0.09720647198465829, "reference_mean_power": 4.5035535336143536e-05, "output_mean_power": 4.403871947151587e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 3.018630059598249e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5795382838093414e-05, "output_power": 1.551911086112722e-05, "reference_active": true, "delta_db": -0.07663320980053936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.352602360298049e-06, "output_power": 3.566079583322089e-06, "reference_active": true, "delta_db": -0.8655796223683893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848555477734869e-06, "output_power": 1.219524136240405e-06, "reference_active": true, "delta_db": -1.8064208880523835}], "active_band_count": 3, "active_band_mae_db": 0.9162112400737707}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 3.1688229259308314e-05, "log_rms_mae_db": 0.8853376553269656, "rms_envelope_nrmse": 0.08706746631364179, "envelope_correlation": 0.9945466251642854, "rms_gain_error_db": -0.09720673606979495, "reference_mean_power": 4.529062388817587e-05, "output_mean_power": 4.428815920429019e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 2.7414916263445066e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5534436484200316e-05, "output_power": 1.4832849161395947e-05, "reference_active": true, "delta_db": -0.2007092342080639}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0764100435071083e-06, "output_power": 1.677368455556648e-06, "reference_active": true, "delta_db": -0.9268464942135534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4701562809614912e-07, "output_power": 9.556305632466181e-08, "reference_active": true, "delta_db": -1.8707347266608798}], "active_band_count": 3, "active_band_mae_db": 0.9994301516941656}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9371793364673878, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0589cf074b160b15279910d0f09c4a13534686092d00ed0d5aea032719abfaed", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9830255534354067, "standard_deviation": 0.001894036965682197, "minimum": 0.9807272887440407, "maximum": 0.9861342223405198, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5693967874691421, "standard_deviation": 0.027963569234595472, "minimum": 1.519712988467698, "maximum": 1.606748876126176, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8639671786189436, "standard_deviation": 0.4500741247457032, "minimum": 0.18263467987346957, "maximum": 1.633667816736314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15189495413540494, "standard_deviation": 0.00835893185919203, "minimum": 0.13682435831237447, "maximum": 0.15994582702884577, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941728173534106, "standard_deviation": 0.0012241783953320037, "minimum": 0.9915305780728608, "maximum": 0.9952368273050599, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8522855282304063, "standard_deviation": 0.02638327663739013, "minimum": 0.7984134810015721, "maximum": 0.8833693861913494, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0097753927118684, "standard_deviation": 0.512087190577841, "minimum": 0.3867922510869479, "maximum": 1.9612573009892849, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0895044855630517, "standard_deviation": 0.010384193270708471, "minimum": 0.07984494454100839, "maximum": 0.10692420603938167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9784764975432507, "standard_deviation": 0.0025329824596642568, "minimum": 0.9737640195375571, "maximum": 0.9820277547487279, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.652067861963443, "standard_deviation": 0.035707902563688665, "minimum": 1.600825425528921, "maximum": 1.724364962828988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7575894514916034, "standard_deviation": 0.3060119866523946, "minimum": 0.3625340540500123, "maximum": 1.1778854789319377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17902618310932766, "standard_deviation": 0.010457533783070334, "minimum": 0.16478939653841285, "maximum": 0.19679349291284426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933743989037291, "standard_deviation": 0.0013537910927687434, "minimum": 0.991217326719091, "maximum": 0.994980493793735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9027483445518929, "standard_deviation": 0.02327428844928781, "minimum": 0.8772230005993135, "maximum": 0.9387059341213396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8159500740737133, "standard_deviation": 0.45274842224839473, "minimum": 0.22694902564709896, "maximum": 1.5938521651156223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09791166466537327, "standard_deviation": 0.008873368623772424, "minimum": 0.08437318452464046, "maximum": 0.11009040510483194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920253692225631, "standard_deviation": 0.0007195597709277852, "minimum": 0.9909037025627234, "maximum": 0.9931019020380737, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6554230109871038, "standard_deviation": 0.05604187602548278, "minimum": 1.5611567404006774, "maximum": 1.7475347358934208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6735605733482649, "standard_deviation": 0.22670822338244737, "minimum": 0.28643704341364756, "maximum": 0.9981328439263821, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1008725704521393, "standard_deviation": 0.0045226804394712695, "minimum": 0.09482334839319811, "maximum": 0.10764245156972356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997344850619136, "standard_deviation": 0.00038488998475841616, "minimum": 0.9966825593555512, "maximum": 0.9979311782243276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.96357280471932, "standard_deviation": 0.027031060130652262, "minimum": 0.9297625349978488, "maximum": 1.0051447463170613, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7342556934281463, "standard_deviation": 0.23060069115940438, "minimum": 0.40268551854104456, "maximum": 1.0532187193991527, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05714209313755505, "standard_deviation": 0.0037647290951246526, "minimum": 0.052033715424612985, "maximum": 0.06360830258949898, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15368342717430222, "standard_deviation": 0.05731489722605907, "minimum": 0.05621306416681454, "maximum": 0.2144456189545978, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.518513760499879, "standard_deviation": 0.0069999727037158, "minimum": 5.510532554158478, "maximum": 5.528867901155377, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5738445878114936, "standard_deviation": 0.03058602493220343, "minimum": 0.5270545346506824, "maximum": 0.6213617236907873, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05350839907575881, "standard_deviation": 0.033762229504866494, "minimum": -0.09423239039528165, "maximum": 7.820856044590249e-05, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964006342329239, "standard_deviation": 0.0004458108305920271, "minimum": 0.9958571747730117, "maximum": 0.9969764242627212, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8921337790394337, "standard_deviation": 0.025113932347155615, "minimum": 0.8527870039787582, "maximum": 0.935464269185758, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5022333545536257, "standard_deviation": 0.1544442892569144, "minimum": 0.3260004155490311, "maximum": 0.7531708720587414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06234108168964453, "standard_deviation": 0.0037298918758578146, "minimum": 0.057032403634346945, "maximum": 0.06665788318722088, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998924330489943, "standard_deviation": 0.0002696095797183917, "minimum": 0.9985137858974709, "maximum": 0.9992654140862679, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.465433280168538, "standard_deviation": 0.02462874855308749, "minimum": 0.44342842930637344, "maximum": 0.5189087018347968, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5018548621542636, "standard_deviation": 0.27042310678756026, "minimum": 0.07298737842393314, "maximum": 0.9171989535557531, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03311550642698068, "standard_deviation": 0.003701653220609284, "minimum": 0.0272471452993882, "maximum": 0.03829694255850706, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0004", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.4053587913513184, "predicted_mos": 3.4053587913513184, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "3b1314ff00d73712b2f4c66943e675956d0c1b56dc779e2b501e8194d3f86ead", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.561496499925852, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9037275866890105, "全帯域_10ms_rms_envelope_nrmse": 0.06341598554784911, "全帯域_10ms_envelope_correlation": 0.9962544461255008, "全帯域_10ms_modulation_mae_db": 0.5195646844521585, "全帯域_40ms_log_rms_mae_db": 0.45912517939581465, "全帯域_40ms_rms_envelope_nrmse": 0.03261940564736586, "全帯域_40ms_envelope_correlation": 0.9989321974971193, "全帯域_40ms_modulation_mae_db": 0.42311065496137806, "300–1000 Hz_10ms_log_rms_mae_db": 1.7598865574283826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10984531725088373, "300–1000 Hz_10ms_envelope_correlation": 0.9903835118594005, "300–1000 Hz_10ms_modulation_mae_db": 0.5198497728977971, "300–1000 Hz_40ms_log_rms_mae_db": 0.9731040198276558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05886515899558727, "300–1000 Hz_40ms_envelope_correlation": 0.9969895901525458, "300–1000 Hz_40ms_modulation_mae_db": 0.6253171804844945, "1000–3000 Hz_10ms_log_rms_mae_db": 1.548290178449589, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1401259448988077, "1000–3000 Hz_10ms_envelope_correlation": 0.9854324572290014, "1000–3000 Hz_10ms_modulation_mae_db": 0.7851367401254231, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8339128165546071, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08162870166900164, "1000–3000 Hz_40ms_envelope_correlation": 0.9950726454299688, "1000–3000 Hz_40ms_modulation_mae_db": 0.9192552711548795, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6261650521399476, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17984503064241117, "3000–6000 Hz_10ms_envelope_correlation": 0.9781478495120138, "3000–6000 Hz_10ms_modulation_mae_db": 0.7925066670147727, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9535852366178823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0977787307188327, "3000–6000 Hz_40ms_envelope_correlation": 0.9932336980027135, "3000–6000 Hz_40ms_modulation_mae_db": 0.8543779018809473, "mr_stft_db": 5.411214482118843, "band_mae_db": 0.12416993770842123, "rms_gain_error_db": -0.03286690211216181, "output_peak": 0.5918225740302173}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8941144001077165, "全帯域_10ms_rms_envelope_nrmse": 0.06771633835762765, "全帯域_10ms_envelope_correlation": 0.9957271471249597, "全帯域_10ms_modulation_mae_db": 0.48136977183368307, "全帯域_40ms_log_rms_mae_db": 0.4829061934559478, "全帯域_40ms_rms_envelope_nrmse": 0.03803389871902868, "全帯域_40ms_envelope_correlation": 0.9985448888855809, "全帯域_40ms_modulation_mae_db": 0.4151721082688335, "300–1000 Hz_10ms_log_rms_mae_db": 1.5969885314134251, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10464907018221199, "300–1000 Hz_10ms_envelope_correlation": 0.9913067769436675, "300–1000 Hz_10ms_modulation_mae_db": 0.5108307305305708, "300–1000 Hz_40ms_log_rms_mae_db": 0.9476939149216964, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058855593640704224, "300–1000 Hz_40ms_envelope_correlation": 0.9970501177565443, "300–1000 Hz_40ms_modulation_mae_db": 0.5088098895110101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5542317758488746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16053322515998025, "1000–3000 Hz_10ms_envelope_correlation": 0.9815229768992073, "1000–3000 Hz_10ms_modulation_mae_db": 0.822409186244012, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8217023257423253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09946711879020742, "1000–3000 Hz_40ms_envelope_correlation": 0.9935249144351174, "1000–3000 Hz_40ms_modulation_mae_db": 0.8552285512687986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6736567399819156, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18737260217762214, "3000–6000 Hz_10ms_envelope_correlation": 0.977158924091929, "3000–6000 Hz_10ms_modulation_mae_db": 1.4067094931705324, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9599704183040906, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10169619989961995, "3000–6000 Hz_40ms_envelope_correlation": 0.9943543323430585, "3000–6000 Hz_40ms_modulation_mae_db": 1.8142902146080437, "mr_stft_db": 5.431147680255986, "band_mae_db": 0.2307743673454986, "rms_gain_error_db": -0.034421025009406254, "output_peak": 0.5260894139319328}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9126519304722034, "全帯域_10ms_rms_envelope_nrmse": 0.06122462015071417, "全帯域_10ms_envelope_correlation": 0.9965594328110493, "全帯域_10ms_modulation_mae_db": 0.45050219072964715, "全帯域_40ms_log_rms_mae_db": 0.47395629545002427, "全帯域_40ms_rms_envelope_nrmse": 0.030394072483224326, "全帯域_40ms_envelope_correlation": 0.9991347777794554, "全帯域_40ms_modulation_mae_db": 0.47057427113338807, "300–1000 Hz_10ms_log_rms_mae_db": 1.6221446344493313, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10202942453769841, "300–1000 Hz_10ms_envelope_correlation": 0.9920453316889354, "300–1000 Hz_10ms_modulation_mae_db": 0.9306795941631686, "300–1000 Hz_40ms_log_rms_mae_db": 0.9503652991676058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055053670126911175, "300–1000 Hz_40ms_envelope_correlation": 0.9978281008498074, "300–1000 Hz_40ms_modulation_mae_db": 0.8916438452014724, "1000–3000 Hz_10ms_log_rms_mae_db": 1.58839838484026, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15128840966427445, "1000–3000 Hz_10ms_envelope_correlation": 0.9826979507695628, "1000–3000 Hz_10ms_modulation_mae_db": 0.2812136472106809, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8490334755366377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07452623283040744, "1000–3000 Hz_40ms_envelope_correlation": 0.9956053599495517, "1000–3000 Hz_40ms_modulation_mae_db": 0.47305244191716733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.551054014578343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16256015980687386, "3000–6000 Hz_10ms_envelope_correlation": 0.9823622835096165, "3000–6000 Hz_10ms_modulation_mae_db": 0.2253829186055892, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8953856992590897, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08302889459551699, "3000–6000 Hz_40ms_envelope_correlation": 0.9950875140129888, "3000–6000 Hz_40ms_modulation_mae_db": 0.28442191898702085, "mr_stft_db": 5.426109666739035, "band_mae_db": 0.14064475764888454, "rms_gain_error_db": -0.07197829288218963, "output_peak": 0.6323229735098119}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8892085228436417, "全帯域_10ms_rms_envelope_nrmse": 0.06465542194088694, "全帯域_10ms_envelope_correlation": 0.9961674157298769, "全帯域_10ms_modulation_mae_db": 0.7275740516920234, "全帯域_40ms_log_rms_mae_db": 0.45322134956513804, "全帯域_40ms_rms_envelope_nrmse": 0.03221165470789509, "全帯域_40ms_envelope_correlation": 0.9990412876303917, "全帯域_40ms_modulation_mae_db": 0.8727800467676022, "300–1000 Hz_10ms_log_rms_mae_db": 1.6696164029437461, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11139730810812232, "300–1000 Hz_10ms_envelope_correlation": 0.9902256485231811, "300–1000 Hz_10ms_modulation_mae_db": 0.9033342879463341, "300–1000 Hz_40ms_log_rms_mae_db": 0.9716704928202128, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06520327503213086, "300–1000 Hz_40ms_envelope_correlation": 0.9964790831045635, "300–1000 Hz_40ms_modulation_mae_db": 1.1066397077850951, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5902420099503978, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16345713552650915, "1000–3000 Hz_10ms_envelope_correlation": 0.9804310962452332, "1000–3000 Hz_10ms_modulation_mae_db": 0.5586570467468047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8291441118733147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.095011132840018, "1000–3000 Hz_40ms_envelope_correlation": 0.9938515785425994, "1000–3000 Hz_40ms_modulation_mae_db": 0.6440292392375877, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6178897528495497, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1769871373445793, "3000–6000 Hz_10ms_envelope_correlation": 0.9789089502724878, "3000–6000 Hz_10ms_modulation_mae_db": 0.5878184758368682, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9519271742050799, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11228921214834967, "3000–6000 Hz_40ms_envelope_correlation": 0.9909760764481094, "3000–6000 Hz_40ms_modulation_mae_db": 0.673195327264846, "mr_stft_db": 5.417800086307937, "band_mae_db": 0.19583935532349236, "rms_gain_error_db": -0.08749445395429474, "output_peak": 0.5684215326239012}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.845470828002078, "全帯域_10ms_rms_envelope_nrmse": 0.058713227130392065, "全帯域_10ms_envelope_correlation": 0.9967856714143514, "全帯域_10ms_modulation_mae_db": 0.24586436791171715, "全帯域_40ms_log_rms_mae_db": 0.44616391904678815, "全帯域_40ms_rms_envelope_nrmse": 0.03027612741650888, "全帯域_40ms_envelope_correlation": 0.9990759029414166, "全帯域_40ms_modulation_mae_db": 0.34878383546107106, "300–1000 Hz_10ms_log_rms_mae_db": 1.6223065064273565, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10614584670497833, "300–1000 Hz_10ms_envelope_correlation": 0.9910802841234493, "300–1000 Hz_10ms_modulation_mae_db": 0.33795814324401224, "300–1000 Hz_40ms_log_rms_mae_db": 0.9024455914639299, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05884288944799774, "300–1000 Hz_40ms_envelope_correlation": 0.9969962776712143, "300–1000 Hz_40ms_modulation_mae_db": 0.4913485886474718, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5638430892425204, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14613695845759828, "1000–3000 Hz_10ms_envelope_correlation": 0.984021379220218, "1000–3000 Hz_10ms_modulation_mae_db": 1.0462075089059093, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8249719617796852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08079196740025282, "1000–3000 Hz_40ms_envelope_correlation": 0.9950441861620851, "1000–3000 Hz_40ms_modulation_mae_db": 1.2240165784661885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.687463234090053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1780791472112082, "3000–6000 Hz_10ms_envelope_correlation": 0.97859752918197, "3000–6000 Hz_10ms_modulation_mae_db": 0.612096331801234, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9253831896862954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10185681237400764, "3000–6000 Hz_40ms_envelope_correlation": 0.9926870364260806, "3000–6000 Hz_40ms_modulation_mae_db": 0.389542746109377, "mr_stft_db": 5.416399128361384, "band_mae_db": 0.08352371802627928, "rms_gain_error_db": -0.023146609884106106, "output_peak": 0.5663366115275611}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9083543920227889, "全帯域_10ms_rms_envelope_nrmse": 0.06034937683679185, "全帯域_10ms_envelope_correlation": 0.9966778562360801, "全帯域_10ms_modulation_mae_db": 0.5272906285527782, "全帯域_40ms_log_rms_mae_db": 0.45794099135957994, "全帯域_40ms_rms_envelope_nrmse": 0.030871504627678594, "全帯域_40ms_envelope_correlation": 0.9991337146063585, "全帯域_40ms_modulation_mae_db": 0.48324575000163694, "300–1000 Hz_10ms_log_rms_mae_db": 1.6720114153324808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10105272714590505, "300–1000 Hz_10ms_envelope_correlation": 0.9921172901668681, "300–1000 Hz_10ms_modulation_mae_db": 0.9346986982091009, "300–1000 Hz_40ms_log_rms_mae_db": 0.9842116129807239, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055551465274023354, "300–1000 Hz_40ms_envelope_correlation": 0.9976546403752885, "300–1000 Hz_40ms_modulation_mae_db": 0.9967092135997375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5930269010329914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14706040381584715, "1000–3000 Hz_10ms_envelope_correlation": 0.9841777588923193, "1000–3000 Hz_10ms_modulation_mae_db": 0.9606870249718701, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8731055498820356, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08192886667451343, "1000–3000 Hz_40ms_envelope_correlation": 0.995434820691847, "1000–3000 Hz_40ms_modulation_mae_db": 1.0380155125797925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6328854314113783, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20272552015167267, "3000–6000 Hz_10ms_envelope_correlation": 0.9727260506151931, "3000–6000 Hz_10ms_modulation_mae_db": 0.6491204413109837, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9339149438808376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10686010977778422, "3000–6000 Hz_40ms_envelope_correlation": 0.9932743636880298, "3000–6000 Hz_40ms_modulation_mae_db": 0.49667898295725427, "mr_stft_db": 5.431576575279341, "band_mae_db": 0.21096284811933638, "rms_gain_error_db": -0.0898917847727765, "output_peak": 0.6146717726142551}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8865584512366413, "全帯域_10ms_rms_envelope_nrmse": 0.060902523741731954, "全帯域_10ms_envelope_correlation": 0.9966262682181841, "全帯域_10ms_modulation_mae_db": 0.6197396263280577, "全帯域_40ms_log_rms_mae_db": 0.45507903580327713, "全帯域_40ms_rms_envelope_nrmse": 0.03483252489352904, "全帯域_40ms_envelope_correlation": 0.9988803812639045, "全帯域_40ms_modulation_mae_db": 0.825403296347349, "300–1000 Hz_10ms_log_rms_mae_db": 1.6861452749971015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10712692178756716, "300–1000 Hz_10ms_envelope_correlation": 0.9910937454572197, "300–1000 Hz_10ms_modulation_mae_db": 1.0584942615599597, "300–1000 Hz_40ms_log_rms_mae_db": 0.9512978301081869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06115049508530668, "300–1000 Hz_40ms_envelope_correlation": 0.9970861773039734, "300–1000 Hz_40ms_modulation_mae_db": 1.061849233926744, "1000–3000 Hz_10ms_log_rms_mae_db": 1.58203632779262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15612548779258978, "1000–3000 Hz_10ms_envelope_correlation": 0.9835050252780692, "1000–3000 Hz_10ms_modulation_mae_db": 1.6297307893673478, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8690729425022721, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10379535227910332, "1000–3000 Hz_40ms_envelope_correlation": 0.9938297018033335, "1000–3000 Hz_40ms_modulation_mae_db": 2.002580430448306, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6204469564226167, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19810261599021817, "3000–6000 Hz_10ms_envelope_correlation": 0.9734109984948408, "3000–6000 Hz_10ms_modulation_mae_db": 0.5094705945429115, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9429340361699766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10625755084063199, "3000–6000 Hz_40ms_envelope_correlation": 0.992039452332866, "3000–6000 Hz_40ms_modulation_mae_db": 0.7579377960162216, "mr_stft_db": 5.416385974351017, "band_mae_db": 0.2530580740635771, "rms_gain_error_db": -0.0949762791724415, "output_peak": 0.5804761724236043}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9536850784607236, "全帯域_10ms_rms_envelope_nrmse": 0.06651979075159821, "全帯域_10ms_envelope_correlation": 0.9958806595548386, "全帯域_10ms_modulation_mae_db": 0.322414412517898, "全帯域_40ms_log_rms_mae_db": 0.532223932818973, "全帯域_40ms_rms_envelope_nrmse": 0.038382864166706124, "全帯域_40ms_envelope_correlation": 0.9985092201767243, "全帯域_40ms_modulation_mae_db": 0.10346453622725332, "300–1000 Hz_10ms_log_rms_mae_db": 1.6860923487581714, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10169194916708513, "300–1000 Hz_10ms_envelope_correlation": 0.9917937278734008, "300–1000 Hz_10ms_modulation_mae_db": 0.6642562024124216, "300–1000 Hz_40ms_log_rms_mae_db": 0.9637706784684162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05898533464271913, "300–1000 Hz_40ms_envelope_correlation": 0.9969642515729603, "300–1000 Hz_40ms_modulation_mae_db": 0.661874865854749, "1000–3000 Hz_10ms_log_rms_mae_db": 1.592214916527709, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16478644247947233, "1000–3000 Hz_10ms_envelope_correlation": 0.979827139266041, "1000–3000 Hz_10ms_modulation_mae_db": 0.6890188769137158, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8624242425781993, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11608549955595797, "1000–3000 Hz_40ms_envelope_correlation": 0.9896628684189446, "1000–3000 Hz_40ms_modulation_mae_db": 0.8894579682515454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5761304281034725, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17544868168393685, "3000–6000 Hz_10ms_envelope_correlation": 0.9792106692933024, "3000–6000 Hz_10ms_modulation_mae_db": 0.9222838721119683, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8993631480680389, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09285037360318563, "3000–6000 Hz_40ms_envelope_correlation": 0.9937590411910097, "3000–6000 Hz_40ms_modulation_mae_db": 1.0327424355864083, "mr_stft_db": 5.421418873590348, "band_mae_db": 0.16083045488845663, "rms_gain_error_db": 0.005469285821508, "output_peak": 0.593802072872741}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85280, "output_samples": 85280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5175221912509524, "spectral_convergence": 0.1838846080399125, "spectral_convergence_status": "ok", "amplitude_floor": 1.2882929606468862e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.212017774878814, "spectral_convergence": 0.1850637548259887, "spectral_convergence_status": "ok", "amplitude_floor": 9.933866418820512e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.50410348022676, "spectral_convergence": 0.1791533909085866, "spectral_convergence_status": "ok", "amplitude_floor": 8.99033450142914e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001234751326260816, "output_power": 0.0012445844875478757, "delta_db": 0.03444882338226348, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018448232014630782, "output_power": 0.001829833697475779, "delta_db": -0.035431307763063864, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006193820215877917, "output_power": 0.0006113905016709583, "delta_db": -0.05639907801581392, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017436605640499536, "output_power": 0.00016314810362478616, "delta_db": -0.2887991525889997, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010535962705358282, "output_power": 9.996377040420367e-05, "delta_db": -0.22831596308522378, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.506722876313705e-05, "output_power": 1.4718747396759449e-05, "delta_db": -0.10162530141516266, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12416993770842123, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03286690211216181, "pitch": null, "output_rms_dbfs": -24.01905873263505, "output_peak": 0.5918225740302173, "reference_rms_dbfs": -23.986191830522884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.411214482118843}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.0001748275799785616, "log_rms_mae_db": 0.9037275866890105, "rms_envelope_nrmse": 0.06341598554784911, "envelope_correlation": 0.9962544461255008, "rms_gain_error_db": -0.03286677136948581, "reference_mean_power": 0.003997498793754849, "output_mean_power": 0.003967360492363928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.9383161701154944e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012108392795166332, "output_power": 0.0012057124289230187, "reference_active": true, "delta_db": -0.018427631862125017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002231683599850243, "output_power": 0.0002098777497235076, "reference_active": true, "delta_db": -0.26666222791989747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.764859371015312e-05, "output_power": 5.045431955360483e-05, "reference_active": true, "delta_db": -1.2736041935744533}], "active_band_count": 3, "active_band_mae_db": 0.5195646844521585}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00017019095451879833, "log_rms_mae_db": 0.45912517939581465, "rms_envelope_nrmse": 0.03261940564736586, "envelope_correlation": 0.9989321974971193, "rms_gain_error_db": -0.03286720593767009, "reference_mean_power": 0.00402014109400192, "output_mean_power": 0.003989831686515838}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.6741447857886242e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011176954288226882, "output_power": 0.001117666465193898, "reference_active": true, "delta_db": -0.00011254324585099304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.194819962775497e-05, "output_power": 8.640827820631453e-05, "reference_active": true, "delta_db": -0.2698787868885634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3053605179616555e-06, "output_power": 2.625939836993247e-06, "reference_active": true, "delta_db": -0.9993406347497198}], "active_band_count": 3, "active_band_mae_db": 0.42311065496137806}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.00014662911909855806, "log_rms_mae_db": 1.7598865574283826, "rms_envelope_nrmse": 0.10984531725088373, "envelope_correlation": 0.9903835118594005, "rms_gain_error_db": -0.05550317105027536, "reference_mean_power": 0.0015423797799867744, "output_mean_power": 0.0015227934708356636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.0119951767590758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000652626948773928, "output_power": 0.0006469829726522296, "reference_active": true, "delta_db": -0.03772152020466188}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016341376588393177, "output_power": 0.00015192972044732937, "reference_active": true, "delta_db": -0.3164589972829219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.788084498799748e-05, "output_power": 3.627646214698681e-05, "reference_active": true, "delta_db": -1.2053688012058077}], "active_band_count": 3, "active_band_mae_db": 0.5198497728977971}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00013732417437068112, "log_rms_mae_db": 0.9731040198276558, "rms_envelope_nrmse": 0.05886515899558727, "envelope_correlation": 0.9969895901525458, "rms_gain_error_db": -0.055504528258489574, "reference_mean_power": 0.0015511172002359399, "output_mean_power": 0.0015314194581085442}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 8.798155816797883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006408046711327789, "output_power": 0.00063359192580702, "reference_active": true, "delta_db": -0.04916034693435205}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.486060187757687e-05, "output_power": 6.783427682487333e-05, "reference_active": true, "delta_db": -0.42804115123597497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5748897665955177e-06, "output_power": 2.590524179918283e-06, "reference_active": true, "delta_db": -1.3987500432831563}], "active_band_count": 3, "active_band_mae_db": 0.6253171804844945}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 8.659636175169924e-05, "log_rms_mae_db": 1.548290178449589, "rms_envelope_nrmse": 0.1401259448988077, "envelope_correlation": 0.9854324572290014, "rms_gain_error_db": -0.2563274905532307, "reference_mean_power": 0.00020984803844209803, "output_mean_power": 0.00019782089615201504}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 2.0976407038326697e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011684508155422031, "output_power": 0.0001163805353001866, "reference_active": true, "delta_db": -0.01730085607614717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.122844755814475e-05, "output_power": 3.540074686642681e-05, "reference_active": true, "delta_db": -0.6618455727199579}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.871758765973707e-06, "output_power": 6.030916910748638e-06, "reference_active": true, "delta_db": -1.676263791580164}], "active_band_count": 3, "active_band_mae_db": 0.7851367401254231}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 7.401841529112696e-05, "log_rms_mae_db": 0.8339128165546071, "rms_envelope_nrmse": 0.08162870166900164, "envelope_correlation": 0.9950726454299688, "rms_gain_error_db": -0.25632799532418665, "reference_mean_power": 0.000211036168686384, "output_mean_power": 0.00019894090727630532}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.8857384691636124e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011739951182146869, "output_power": 0.00011656773435949521, "reference_active": true, "delta_db": -0.030879355095504657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97467891176206e-05, "output_power": 1.6953175618656473e-05, "reference_active": true, "delta_db": -0.6624542730685933}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.379135569148535e-07, "output_power": 3.965688640822189e-07, "reference_active": true, "delta_db": -2.0644321853005403}], "active_band_count": 3, "active_band_mae_db": 0.9192552711548795}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 4.122543101057287e-05, "log_rms_mae_db": 1.6261650521399476, "rms_envelope_nrmse": 0.17984503064241117, "envelope_correlation": 0.9781478495120138, "rms_gain_error_db": -0.14958361567924605, "reference_mean_power": 4.5035535336143536e-05, "output_mean_power": 4.351078993921615e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 3.018630059598249e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5795382838093414e-05, "output_power": 1.550284340171492e-05, "reference_active": true, "delta_db": -0.08118796379511595}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.352602360298049e-06, "output_power": 3.792278114338183e-06, "reference_active": true, "delta_db": -0.5984881353483991}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848555477734869e-06, "output_power": 1.2503975057588186e-06, "reference_active": true, "delta_db": -1.6978439019008031}], "active_band_count": 3, "active_band_mae_db": 0.7925066670147727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 3.1688229259308314e-05, "log_rms_mae_db": 0.9535852366178823, "rms_envelope_nrmse": 0.0977787307188327, "envelope_correlation": 0.9932336980027135, "rms_gain_error_db": -0.14958353292902649, "reference_mean_power": 4.529062388817587e-05, "output_mean_power": 4.375724292059675e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 2.7414916263445066e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5534436484200316e-05, "output_power": 1.4922822897689013e-05, "reference_active": true, "delta_db": -0.17444518868415448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0764100435071083e-06, "output_power": 1.7786135220938537e-06, "reference_active": true, "delta_db": -0.6723153111478047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4701562809614912e-07, "output_power": 9.902074354420561e-08, "reference_active": true, "delta_db": -1.7163732058108827}], "active_band_count": 3, "active_band_mae_db": 0.8543779018809473}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9399577328040583, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "44fc21ac5297c579a61eb08169fc120fd2c2bd1af8155d164d82b8a0b54c3091", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9827019729749564, "standard_deviation": 0.001957798037056902, "minimum": 0.979827139266041, "maximum": 0.9854324572290014, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5765354479606204, "standard_deviation": 0.018256183141017315, "minimum": 1.548290178449589, "maximum": 1.5930269010329914, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8466326025607205, "standard_deviation": 0.3960375959816668, "minimum": 0.2812136472106809, "maximum": 1.6297307893673478, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15368925097438488, "standard_deviation": 0.008960552502242994, "minimum": 0.1401259448988077, "maximum": 0.16478644247947233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994003259429181, "standard_deviation": 0.0019259189273852023, "minimum": 0.9896628684189446, "maximum": 0.9956053599495517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8454209283061347, "standard_deviation": 0.020723082264937236, "minimum": 0.8217023257423253, "maximum": 0.8731055498820356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.005704499165533, "standard_deviation": 0.4630061294766728, "minimum": 0.47305244191716733, "maximum": 2.002580430448306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09165435900493277, "standard_deviation": 0.014258104242150226, "minimum": 0.07452623283040744, "maximum": 0.11608549955595797, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9775654068714192, "standard_deviation": 0.003157083755328185, "minimum": 0.9727260506151931, "maximum": 0.9823622835096165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6232114511971596, "standard_deviation": 0.04509451933251748, "minimum": 1.551054014578343, "maximum": 1.687463234090053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7131735992993575, "standard_deviation": 0.34650930069285374, "minimum": 0.2253829186055892, "maximum": 1.4067094931705324, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1826401118760653, "standard_deviation": 0.012985713389678197, "minimum": 0.16256015980687386, "maximum": 0.20272552015167267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9931764393056071, "standard_deviation": 0.0012964667647186203, "minimum": 0.9909760764481094, "maximum": 0.9950875140129888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9328079807739114, "standard_deviation": 0.024532514304904857, "minimum": 0.8953856992590897, "maximum": 0.9599704183040906, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7878984154262649, "standard_deviation": 0.4823879971403402, "minimum": 0.28442191898702085, "maximum": 1.8142902146080437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1003272354947411, "standard_deviation": 0.009155642905854256, "minimum": 0.08302889459551699, "maximum": 0.11228921214834967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9912557895795153, "standard_deviation": 0.0007108000325108412, "minimum": 0.9902256485231811, "maximum": 0.9921172901668681, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6643989589687496, "standard_deviation": 0.051057287627588606, "minimum": 1.5969885314134251, "maximum": 1.7598865574283826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7325127113704206, "standard_deviation": 0.259200405938566, "minimum": 0.33795814324401224, "maximum": 1.0584942615599597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10549232061055652, "standard_deviation": 0.0038502691593692772, "minimum": 0.10105272714590505, "maximum": 0.11139730810812232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971310298483622, "standard_deviation": 0.00042420469604037244, "minimum": 0.9964790831045635, "maximum": 0.9978281008498074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9555699299698035, "standard_deviation": 0.02502523494000814, "minimum": 0.9024455914639299, "maximum": 0.9842116129807239, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7930240656263469, "standard_deviation": 0.2504130147766838, "minimum": 0.4913485886474718, "maximum": 1.1066397077850951, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059063485280672554, "standard_deviation": 0.003175865517666472, "minimum": 0.055053670126911175, "maximum": 0.06520327503213086, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17497543914049327, "standard_deviation": 0.05764931843302419, "minimum": 0.08352371802627928, "maximum": 0.2530580740635771, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.421506558375486, "standard_deviation": 0.007440466169606197, "minimum": 5.411214482118843, "maximum": 5.431576575279341, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5842428904417531, "standard_deviation": 0.03241121061059275, "minimum": 0.5260894139319328, "maximum": 0.6323229735098119, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.053663257745733564, "standard_deviation": 0.03728258439207213, "minimum": -0.0949762791724415, "maximum": 0.005469285821508, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963348621518552, "standard_deviation": 0.0003903013495106703, "minimum": 0.9957271471249597, "maximum": 0.9967856714143514, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8992213987293505, "standard_deviation": 0.030298991261704107, "minimum": 0.845470828002078, "maximum": 0.9536850784607236, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4867899667522454, "standard_deviation": 0.15342644653585558, "minimum": 0.24586436791171715, "maximum": 0.7275740516920234, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.062937160557199, "standard_deviation": 0.0031728368312616572, "minimum": 0.058713227130392065, "maximum": 0.06771633835762765, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989065463476189, "standard_deviation": 0.00025086668981322383, "minimum": 0.9985092201767243, "maximum": 0.9991347777794554, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47007711211194286, "standard_deviation": 0.027748850167153682, "minimum": 0.44616391904678815, "maximum": 0.532223932818973, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.492816812396064, "standard_deviation": 0.2503712975688716, "minimum": 0.10346453622725332, "maximum": 0.8727800467676022, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03345275658274208, "standard_deviation": 0.003285047841246295, "minimum": 0.03027612741650888, "maximum": 0.038382864166706124, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0004", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.966728448867798, "predicted_mos": 2.966728448867798, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "a8da03760d1aa83198c2e07f8ad9e83160a45dd315edafaebb7a4535d6951d43", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.583875699900091, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8879924773908462, "全帯域_10ms_rms_envelope_nrmse": 0.059528527600342, "全帯域_10ms_envelope_correlation": 0.9967859228025393, "全帯域_10ms_modulation_mae_db": 0.8201990215833869, "全帯域_40ms_log_rms_mae_db": 0.4787305151672562, "全帯域_40ms_rms_envelope_nrmse": 0.030765279516951287, "全帯域_40ms_envelope_correlation": 0.9991571151269304, "全帯域_40ms_modulation_mae_db": 0.8433101455385538, "300–1000 Hz_10ms_log_rms_mae_db": 1.9438432987754384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09268509254003646, "300–1000 Hz_10ms_envelope_correlation": 0.9934240618292164, "300–1000 Hz_10ms_modulation_mae_db": 1.1198219219319414, "300–1000 Hz_40ms_log_rms_mae_db": 1.1999862666462267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053133378741405224, "300–1000 Hz_40ms_envelope_correlation": 0.9978373503004824, "300–1000 Hz_40ms_modulation_mae_db": 1.2745860017910287, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8686960999190272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16640891554873685, "1000–3000 Hz_10ms_envelope_correlation": 0.980468809696541, "1000–3000 Hz_10ms_modulation_mae_db": 1.3912857900699105, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0221453047553355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10560636440024133, "1000–3000 Hz_40ms_envelope_correlation": 0.9928780460567106, "1000–3000 Hz_40ms_modulation_mae_db": 1.819142803356051, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1953145028875114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2011606418160522, "3000–6000 Hz_10ms_envelope_correlation": 0.9727932385828186, "3000–6000 Hz_10ms_modulation_mae_db": 0.9950681897430624, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3048368136216881, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10840162843506838, "3000–6000 Hz_40ms_envelope_correlation": 0.9919434036449277, "3000–6000 Hz_40ms_modulation_mae_db": 1.3643023180062153, "mr_stft_db": 5.930459395546255, "band_mae_db": 0.22677869524468106, "rms_gain_error_db": -0.09715647679025288, "output_peak": 0.5851839687745639}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8695487279817479, "全帯域_10ms_rms_envelope_nrmse": 0.05864007077654467, "全帯域_10ms_envelope_correlation": 0.9968559382229912, "全帯域_10ms_modulation_mae_db": 0.6433312998548697, "全帯域_40ms_log_rms_mae_db": 0.4679443688878667, "全帯域_40ms_rms_envelope_nrmse": 0.028801306863626606, "全帯域_40ms_envelope_correlation": 0.9992390667663879, "全帯域_40ms_modulation_mae_db": 0.647694572952164, "300–1000 Hz_10ms_log_rms_mae_db": 1.8623794517253642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09810354244424205, "300–1000 Hz_10ms_envelope_correlation": 0.9925619180385995, "300–1000 Hz_10ms_modulation_mae_db": 0.6954729053699916, "300–1000 Hz_40ms_log_rms_mae_db": 1.0693372480042624, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05417853023888826, "300–1000 Hz_40ms_envelope_correlation": 0.997762086171858, "300–1000 Hz_40ms_modulation_mae_db": 0.82173816409489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9291840103352849, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17509517997052693, "1000–3000 Hz_10ms_envelope_correlation": 0.9774770194606225, "1000–3000 Hz_10ms_modulation_mae_db": 1.7158409728799449, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0837564024987905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10062055703377346, "1000–3000 Hz_40ms_envelope_correlation": 0.9929615524812186, "1000–3000 Hz_40ms_modulation_mae_db": 2.028173303812243, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2574394779306974, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20630920079281287, "3000–6000 Hz_10ms_envelope_correlation": 0.9715513011284805, "3000–6000 Hz_10ms_modulation_mae_db": 0.8524025181981637, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3511729024053019, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11387008167292273, "3000–6000 Hz_40ms_envelope_correlation": 0.990864912134388, "3000–6000 Hz_40ms_modulation_mae_db": 0.9340970841903561, "mr_stft_db": 5.916868723187577, "band_mae_db": 0.23374473217327565, "rms_gain_error_db": -0.07905007117131678, "output_peak": 0.5725272408162597}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8639221430677945, "全帯域_10ms_rms_envelope_nrmse": 0.05607522793301722, "全帯域_10ms_envelope_correlation": 0.9971065862028851, "全帯域_10ms_modulation_mae_db": 0.46422352499966846, "全帯域_40ms_log_rms_mae_db": 0.43854843097917695, "全帯域_40ms_rms_envelope_nrmse": 0.027079974430515057, "全帯域_40ms_envelope_correlation": 0.9993078600196263, "全帯域_40ms_modulation_mae_db": 0.43169710644333, "300–1000 Hz_10ms_log_rms_mae_db": 1.8588990854723444, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09762389250892518, "300–1000 Hz_10ms_envelope_correlation": 0.9927242345056072, "300–1000 Hz_10ms_modulation_mae_db": 1.1335103923495573, "300–1000 Hz_40ms_log_rms_mae_db": 1.081499612721426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05832947568640378, "300–1000 Hz_40ms_envelope_correlation": 0.997403853013065, "300–1000 Hz_40ms_modulation_mae_db": 1.2305157717308475, "1000–3000 Hz_10ms_log_rms_mae_db": 1.931234035079306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16353096218022303, "1000–3000 Hz_10ms_envelope_correlation": 0.9797727385850563, "1000–3000 Hz_10ms_modulation_mae_db": 0.5713487617917558, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0504469253505875, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09305917194791324, "1000–3000 Hz_40ms_envelope_correlation": 0.9928639199018912, "1000–3000 Hz_40ms_modulation_mae_db": 0.6982178924741372, "3000–6000 Hz_10ms_log_rms_mae_db": 2.234227547840137, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20699860536904596, "3000–6000 Hz_10ms_envelope_correlation": 0.9712679205999527, "3000–6000 Hz_10ms_modulation_mae_db": 0.9820983101512969, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3464208719066029, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11892532988051151, "3000–6000 Hz_40ms_envelope_correlation": 0.9898985515590013, "3000–6000 Hz_40ms_modulation_mae_db": 1.1985718670570016, "mr_stft_db": 5.927327413691285, "band_mae_db": 0.18559863103821964, "rms_gain_error_db": -0.06435735566449895, "output_peak": 0.600836108263091}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.878536455573838, "全帯域_10ms_rms_envelope_nrmse": 0.05658346665814584, "全帯域_10ms_envelope_correlation": 0.9970329960102794, "全帯域_10ms_modulation_mae_db": 0.5325503463046668, "全帯域_40ms_log_rms_mae_db": 0.45582856350396755, "全帯域_40ms_rms_envelope_nrmse": 0.028429754936502138, "全帯域_40ms_envelope_correlation": 0.9992043126809954, "全帯域_40ms_modulation_mae_db": 0.5466867812766946, "300–1000 Hz_10ms_log_rms_mae_db": 1.820345641967846, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09596443001139333, "300–1000 Hz_10ms_envelope_correlation": 0.9928571858008035, "300–1000 Hz_10ms_modulation_mae_db": 0.5968963608448616, "300–1000 Hz_40ms_log_rms_mae_db": 1.035634186673031, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05590893185688141, "300–1000 Hz_40ms_envelope_correlation": 0.9975118840546695, "300–1000 Hz_40ms_modulation_mae_db": 0.6627115993144449, "1000–3000 Hz_10ms_log_rms_mae_db": 1.874174323961028, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18288991121960743, "1000–3000 Hz_10ms_envelope_correlation": 0.9746875635364862, "1000–3000 Hz_10ms_modulation_mae_db": 1.124192417743506, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0339489070124463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10095814939017142, "1000–3000 Hz_40ms_envelope_correlation": 0.9916685363528622, "1000–3000 Hz_40ms_modulation_mae_db": 1.0984637987673274, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2693202813999602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19892628184277047, "3000–6000 Hz_10ms_envelope_correlation": 0.9734695223377073, "3000–6000 Hz_10ms_modulation_mae_db": 1.2495428311335053, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3556309957943595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11127527360983906, "3000–6000 Hz_40ms_envelope_correlation": 0.9915858178898385, "3000–6000 Hz_40ms_modulation_mae_db": 1.5848375440261548, "mr_stft_db": 5.937133663727029, "band_mae_db": 0.09115720835019865, "rms_gain_error_db": -0.045232592899780294, "output_peak": 0.5751738554322345}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9040020662205102, "全帯域_10ms_rms_envelope_nrmse": 0.0634353680133803, "全帯域_10ms_envelope_correlation": 0.996249087585944, "全帯域_10ms_modulation_mae_db": 0.2441334350447614, "全帯域_40ms_log_rms_mae_db": 0.4561486033332767, "全帯域_40ms_rms_envelope_nrmse": 0.03492541527149329, "全帯域_40ms_envelope_correlation": 0.9987637718258598, "全帯域_40ms_modulation_mae_db": 0.10332143193182293, "300–1000 Hz_10ms_log_rms_mae_db": 1.912545399932665, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09852716346769301, "300–1000 Hz_10ms_envelope_correlation": 0.9922824835012037, "300–1000 Hz_10ms_modulation_mae_db": 0.30049844907424733, "300–1000 Hz_40ms_log_rms_mae_db": 1.1110564711588662, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05709039296606268, "300–1000 Hz_40ms_envelope_correlation": 0.997164488944436, "300–1000 Hz_40ms_modulation_mae_db": 0.40376641271714764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8964784063062596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16047953160501807, "1000–3000 Hz_10ms_envelope_correlation": 0.9805288876204328, "1000–3000 Hz_10ms_modulation_mae_db": 0.9585174820447936, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0336735656007465, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08873110688749246, "1000–3000 Hz_40ms_envelope_correlation": 0.9935991361065633, "1000–3000 Hz_40ms_modulation_mae_db": 1.2360021075883925, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2441168352075884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22096084252866932, "3000–6000 Hz_10ms_envelope_correlation": 0.9670533563528304, "3000–6000 Hz_10ms_modulation_mae_db": 0.8789297298239536, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3454455814130872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11530512884456354, "3000–6000 Hz_40ms_envelope_correlation": 0.9906398165519208, "3000–6000 Hz_40ms_modulation_mae_db": 1.1503318751545692, "mr_stft_db": 5.902464273267122, "band_mae_db": 0.15037487795861293, "rms_gain_error_db": -0.00837251682769571, "output_peak": 0.5876844107998143}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8919833177557608, "全帯域_10ms_rms_envelope_nrmse": 0.06490448986444457, "全帯域_10ms_envelope_correlation": 0.9961187492266221, "全帯域_10ms_modulation_mae_db": 0.5099498894398148, "全帯域_40ms_log_rms_mae_db": 0.4855243709644244, "全帯域_40ms_rms_envelope_nrmse": 0.0375728200316129, "全帯域_40ms_envelope_correlation": 0.9986414603942732, "全帯域_40ms_modulation_mae_db": 0.4469971939133231, "300–1000 Hz_10ms_log_rms_mae_db": 1.9312224549431123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11563193595690001, "300–1000 Hz_10ms_envelope_correlation": 0.9893831031417332, "300–1000 Hz_10ms_modulation_mae_db": 0.6123021503668311, "300–1000 Hz_40ms_log_rms_mae_db": 1.148731747941667, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07025254196873201, "300–1000 Hz_40ms_envelope_correlation": 0.9957811874651972, "300–1000 Hz_40ms_modulation_mae_db": 0.6270861376668777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.90197462999118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16818473493695454, "1000–3000 Hz_10ms_envelope_correlation": 0.9787030492961055, "1000–3000 Hz_10ms_modulation_mae_db": 0.5634097026825728, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0225851681094116, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08494840369930205, "1000–3000 Hz_40ms_envelope_correlation": 0.9941177975668388, "1000–3000 Hz_40ms_modulation_mae_db": 0.6086108444813402, "3000–6000 Hz_10ms_log_rms_mae_db": 2.259987751635576, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21704366847093293, "3000–6000 Hz_10ms_envelope_correlation": 0.9682868926619745, "3000–6000 Hz_10ms_modulation_mae_db": 1.4476105742638234, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3847866548534558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12393039784078443, "3000–6000 Hz_40ms_envelope_correlation": 0.9895010341220655, "3000–6000 Hz_40ms_modulation_mae_db": 1.7198357512547418, "mr_stft_db": 5.928409325371651, "band_mae_db": 0.11143586226422607, "rms_gain_error_db": -0.0794475799215947, "output_peak": 0.63630025520266}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8479247957016299, "全帯域_10ms_rms_envelope_nrmse": 0.0605567661927518, "全帯域_10ms_envelope_correlation": 0.9965927305673956, "全帯域_10ms_modulation_mae_db": 0.22491830803961646, "全帯域_40ms_log_rms_mae_db": 0.43750393310717484, "全帯域_40ms_rms_envelope_nrmse": 0.03255002647535784, "全帯域_40ms_envelope_correlation": 0.9989371469330776, "全帯域_40ms_modulation_mae_db": 0.12658938757092406, "300–1000 Hz_10ms_log_rms_mae_db": 1.8852732156666259, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10714846592353232, "300–1000 Hz_10ms_envelope_correlation": 0.9911128530063886, "300–1000 Hz_10ms_modulation_mae_db": 0.25297371858989876, "300–1000 Hz_40ms_log_rms_mae_db": 1.1347270098083397, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06449201284508639, "300–1000 Hz_40ms_envelope_correlation": 0.9965539058574464, "300–1000 Hz_40ms_modulation_mae_db": 0.3551275209433717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.894431358643337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18894071407788315, "1000–3000 Hz_10ms_envelope_correlation": 0.9736007016468026, "1000–3000 Hz_10ms_modulation_mae_db": 1.2207276129721112, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0288252300412082, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11669271984452566, "1000–3000 Hz_40ms_envelope_correlation": 0.9902057535882197, "1000–3000 Hz_40ms_modulation_mae_db": 1.4998189868401057, "3000–6000 Hz_10ms_log_rms_mae_db": 2.214522214021024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21648379628289474, "3000–6000 Hz_10ms_envelope_correlation": 0.9684323494031067, "3000–6000 Hz_10ms_modulation_mae_db": 1.7174481279641025, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3283598443693794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11615307722142394, "3000–6000 Hz_40ms_envelope_correlation": 0.9907698243977733, "3000–6000 Hz_40ms_modulation_mae_db": 2.011949291402859, "mr_stft_db": 5.927278769911052, "band_mae_db": 0.21466854635855936, "rms_gain_error_db": 0.016648775086266335, "output_peak": 0.6432700102928257}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.911711058144992, "全帯域_10ms_rms_envelope_nrmse": 0.06967403752596477, "全帯域_10ms_envelope_correlation": 0.9954707085784424, "全帯域_10ms_modulation_mae_db": 0.6188591656713305, "全帯域_40ms_log_rms_mae_db": 0.49097872649221785, "全帯域_40ms_rms_envelope_nrmse": 0.03432829388680278, "全帯域_40ms_envelope_correlation": 0.9988210907573781, "全帯域_40ms_modulation_mae_db": 0.6570564852812318, "300–1000 Hz_10ms_log_rms_mae_db": 1.9232384445359625, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10114126312636518, "300–1000 Hz_10ms_envelope_correlation": 0.9919331997473747, "300–1000 Hz_10ms_modulation_mae_db": 0.8758007987191451, "300–1000 Hz_40ms_log_rms_mae_db": 1.0901265007838572, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051999700891497615, "300–1000 Hz_40ms_envelope_correlation": 0.9977219339166302, "300–1000 Hz_40ms_modulation_mae_db": 0.9622598179753185, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9224397734545497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1915232706680235, "1000–3000 Hz_10ms_envelope_correlation": 0.9724501678907153, "1000–3000 Hz_10ms_modulation_mae_db": 1.139123531805508, "1000–3000 Hz_40ms_log_rms_mae_db": 1.024722525345984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12038255749931465, "1000–3000 Hz_40ms_envelope_correlation": 0.9889302226055551, "1000–3000 Hz_40ms_modulation_mae_db": 1.3978946863112622, "3000–6000 Hz_10ms_log_rms_mae_db": 2.239351969179721, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2086389755034261, "3000–6000 Hz_10ms_envelope_correlation": 0.9718792248404174, "3000–6000 Hz_10ms_modulation_mae_db": 1.6020426069244271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3281089214324289, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13037169798040282, "3000–6000 Hz_40ms_envelope_correlation": 0.990025714386948, "3000–6000 Hz_40ms_modulation_mae_db": 1.8663305806852037, "mr_stft_db": 5.93114018791107, "band_mae_db": 0.1530881967601003, "rms_gain_error_db": -0.031131126918448757, "output_peak": 0.6543568186110584}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85280, "output_samples": 85280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.076669051431423, "spectral_convergence": 0.1826341562098335, "spectral_convergence_status": "ok", "amplitude_floor": 1.2882929606468862e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.803165182011281, "spectral_convergence": 0.1860151407297869, "spectral_convergence_status": "ok", "amplitude_floor": 9.933866418820512e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.911543953196063, "spectral_convergence": 0.17833292561341785, "spectral_convergence_status": "ok", "amplitude_floor": 8.99033450142914e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001234751326260816, "output_power": 0.0012471518534713346, "delta_db": 0.04339835170753664, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018448232014630782, "output_power": 0.0018079685931492845, "delta_db": -0.08763869975394997, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006193820215877917, "output_power": 0.0005825752208103576, "delta_db": -0.2660658675976305, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017436605640499536, "output_power": 0.0001529770342076323, "delta_db": -0.568357084031429, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010535962705358282, "output_power": 9.918810075268555e-05, "delta_db": -0.2621465016510158, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.506722876313705e-05, "output_power": 1.5536026337775034e-05, "delta_db": 0.13306566672652445, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22677869524468106, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09715647679025288, "pitch": null, "output_rms_dbfs": -24.083348307313138, "output_peak": 0.5851839687745639, "reference_rms_dbfs": -23.986191830522884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.930459395546255}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.0001748275799785616, "log_rms_mae_db": 0.8879924773908462, "rms_envelope_nrmse": 0.059528527600342, "envelope_correlation": 0.9967859228025393, "rms_gain_error_db": -0.09715677764144705, "reference_mean_power": 0.003997498793754849, "output_mean_power": 0.003909062944699152}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.9383161701154944e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012108392795166332, "output_power": 0.0011806909903698294, "reference_active": true, "delta_db": -0.10950251753601908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002231683599850243, "output_power": 0.0001900832167483936, "reference_active": true, "delta_db": -0.6968884904503885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.764859371015312e-05, "output_power": 4.6220875466918626e-05, "reference_active": true, "delta_db": -1.654206056763753}], "active_band_count": 3, "active_band_mae_db": 0.8201990215833869}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00017019095451879833, "log_rms_mae_db": 0.4787305151672562, "rms_envelope_nrmse": 0.030765279516951287, "envelope_correlation": 0.9991571151269304, "rms_gain_error_db": -0.09715543217347061, "reference_mean_power": 0.00402014109400192, "output_mean_power": 0.0039312055518727915}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.6741447857886242e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011176954288226882, "output_power": 0.0010885216691205483, "reference_active": true, "delta_db": -0.1148639581559452}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.194819962775497e-05, "output_power": 7.612258113965487e-05, "reference_active": true, "delta_db": -0.8202972426594028}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3053605179616555e-06, "output_power": 2.2895066622672945e-06, "reference_active": true, "delta_db": -1.5947692358003132}], "active_band_count": 3, "active_band_mae_db": 0.8433101455385538}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.00014662911909855806, "log_rms_mae_db": 1.9438432987754384, "rms_envelope_nrmse": 0.09268509254003646, "envelope_correlation": 0.9934240618292164, "rms_gain_error_db": -0.16790822227361124, "reference_mean_power": 0.0015423797799867744, "output_mean_power": 0.0014838858753425295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.0119951767590758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000652626948773928, "output_power": 0.0006181352296834092, "reference_active": true, "delta_db": -0.23581506788924445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016341376588393177, "output_power": 0.00013655282937852415, "reference_active": true, "delta_db": -0.7798793529780917}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.788084498799748e-05, "output_power": 2.7911596079003e-05, "reference_active": true, "delta_db": -2.343771344928488}], "active_band_count": 3, "active_band_mae_db": 1.1198219219319414}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00013732417437068112, "log_rms_mae_db": 1.1999862666462267, "rms_envelope_nrmse": 0.053133378741405224, "envelope_correlation": 0.9978373503004824, "rms_gain_error_db": -0.16790980149106727, "reference_mean_power": 0.0015511172002359399, "output_mean_power": 0.0014922913910960592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 8.798155816797883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006408046711327789, "output_power": 0.000599227316269218, "reference_active": true, "delta_db": -0.2913406605540948}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.486060187757687e-05, "output_power": 6.027277785353687e-05, "reference_active": true, "delta_db": -0.9413210680456163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5748897665955177e-06, "output_power": 1.968579811712452e-06, "reference_active": true, "delta_db": -2.591096276773375}], "active_band_count": 3, "active_band_mae_db": 1.2745860017910287}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 8.659636175169924e-05, "log_rms_mae_db": 1.8686960999190272, "rms_envelope_nrmse": 0.16640891554873685, "envelope_correlation": 0.980468809696541, "rms_gain_error_db": -0.4341853760349881, "reference_mean_power": 0.00020984803844209803, "output_mean_power": 0.00018988312757317429}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 2.0976407038326697e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011684508155422031, "output_power": 9.922422580240435e-05, "reference_active": true, "delta_db": -0.7099271702939052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.122844755814475e-05, "output_power": 2.940079544550134e-05, "reference_active": true, "delta_db": -1.4683790140065471}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.871758765973707e-06, "output_power": 5.603438448022326e-06, "reference_active": true, "delta_db": -1.9955511859092785}], "active_band_count": 3, "active_band_mae_db": 1.3912857900699105}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 7.401841529112696e-05, "log_rms_mae_db": 1.0221453047553355, "rms_envelope_nrmse": 0.10560636440024133, "envelope_correlation": 0.9928780460567106, "rms_gain_error_db": -0.43418691158133027, "reference_mean_power": 0.000211036168686384, "output_mean_power": 0.000190958151766831}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.8857384691636124e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011739951182146869, "output_power": 9.85434508283106e-05, "reference_active": true, "delta_db": -0.7603852452479398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97467891176206e-05, "output_power": 1.3282220514407393e-05, "reference_active": true, "delta_db": -1.7222580200526978}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.379135569148535e-07, "output_power": 3.2157576938552805e-07, "reference_active": true, "delta_db": -2.974785144767515}], "active_band_count": 3, "active_band_mae_db": 1.819142803356051}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 4.122543101057287e-05, "log_rms_mae_db": 2.1953145028875114, "rms_envelope_nrmse": 0.2011606418160522, "envelope_correlation": 0.9727932385828186, "rms_gain_error_db": -0.1820134451205879, "reference_mean_power": 4.5035535336143536e-05, "output_mean_power": 4.318709430736077e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 3.018630059598249e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5795382838093414e-05, "output_power": 1.5022952977261561e-05, "reference_active": true, "delta_db": -0.21774848632785587}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.352602360298049e-06, "output_power": 3.0662012715593374e-06, "reference_active": true, "delta_db": -1.521483338186329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848555477734869e-06, "output_power": 1.3875071009366213e-06, "reference_active": true, "delta_db": -1.2459727447150024}], "active_band_count": 3, "active_band_mae_db": 0.9950681897430624}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 3.1688229259308314e-05, "log_rms_mae_db": 1.3048368136216881, "rms_envelope_nrmse": 0.10840162843506838, "envelope_correlation": 0.9919434036449277, "rms_gain_error_db": -0.18201362560116732, "reference_mean_power": 4.529062388817587e-05, "output_mean_power": 4.343171118571005e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 2.7414916263445066e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5534436484200316e-05, "output_power": 1.4403167528812267e-05, "reference_active": true, "delta_db": -0.32837491564023596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0764100435071083e-06, "output_power": 1.4263551301886254e-06, "reference_active": true, "delta_db": -1.630854523868801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4701562809614912e-07, "output_power": 8.994887837492033e-08, "reference_active": true, "delta_db": -2.1336775145096087}], "active_band_count": 3, "active_band_mae_db": 1.3643023180062153}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9204400401409736, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9a8ad81492428a3b008ff4fda1bac1da875c61370be775cb8f86cda45396de1c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9772111172165953, "standard_deviation": 0.0032193320594201542, "minimum": 0.9724501678907153, "maximum": 0.9805288876204328, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9023265797112465, "standard_deviation": 0.02384989161253908, "minimum": 1.8686960999190272, "maximum": 1.931234035079306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0855557839987628, "standard_deviation": 0.3903383320804919, "minimum": 0.5634097026825728, "maximum": 1.7158409728799449, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1746316525258717, "standard_deviation": 0.01189942197159176, "minimum": 0.16047953160501807, "maximum": 0.1915232706680235, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9921531205824824, "standard_deviation": 0.0017747982885353629, "minimum": 0.9889302226055551, "maximum": 0.9941177975668388, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0375130035893136, "standard_deviation": 0.020809553183190273, "minimum": 1.0221453047553355, "maximum": 1.0837564024987905, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2982905529538573, "standard_deviation": 0.497724316987523, "minimum": 0.6086108444813402, "maximum": 2.028173303812243, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10137487883784178, "standard_deviation": 0.012611906346068444, "minimum": 0.08494840369930205, "maximum": 0.12038255749931465, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9705917257384111, "standard_deviation": 0.0023500537989344854, "minimum": 0.9670533563528304, "maximum": 0.9734695223377073, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.239285072512777, "standard_deviation": 0.02465718164698479, "minimum": 2.1953145028875114, "maximum": 2.2693202813999602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2156428610252918, "standard_deviation": 0.3390295056168579, "minimum": 0.8524025181981637, "maximum": 1.7174481279641025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20956525157582556, "standard_deviation": 0.007887830973765404, "minimum": 0.19892628184277047, "maximum": 0.22096084252866932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906536343358578, "standard_deviation": 0.0008352690869843507, "minimum": 0.9895010341220655, "maximum": 0.9919434036449277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.343095323224538, "standard_deviation": 0.023551101356042246, "minimum": 1.3048368136216881, "maximum": 1.3847866548534558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4787820389721378, "standard_deviation": 0.37796455061297946, "minimum": 0.9340970841903561, "maximum": 2.011949291402859, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11727907693568956, "standard_deviation": 0.007071141408596566, "minimum": 0.10840162843506838, "maximum": 0.13037169798040282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920348799463659, "standard_deviation": 0.0012712587891962868, "minimum": 0.9893831031417332, "maximum": 0.9934240618292164, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8922183741274199, "standard_deviation": 0.04272226761663591, "minimum": 1.820345641967846, "maximum": 1.9438432987754384, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6984095871558093, "standard_deviation": 0.3322742295201348, "minimum": 0.25297371858989876, "maximum": 1.1335103923495573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10085322324738594, "standard_deviation": 0.007291305300215877, "minimum": 0.09268509254003646, "maximum": 0.11563193595690001, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972170862154731, "standard_deviation": 0.0007134214515668367, "minimum": 0.9957811874651972, "maximum": 0.9978373503004824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1088873804672095, "standard_deviation": 0.05153539129137663, "minimum": 1.035634186673031, "maximum": 1.1999862666462267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7922239282792407, "standard_deviation": 0.3465240968417912, "minimum": 0.3551275209433717, "maximum": 1.2745860017910287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05817312064936967, "standard_deviation": 0.006229319815734048, "minimum": 0.051999700891497615, "maximum": 0.07025254196873201, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17085584376848423, "standard_deviation": 0.05323404686177147, "minimum": 0.09115720835019865, "maximum": 0.23374473217327565, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.92513521907663, "standard_deviation": 0.01076234319568357, "minimum": 5.902464273267122, "maximum": 5.937133663727029, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6069165835240635, "standard_deviation": 0.03274344116174379, "minimum": 0.5725272408162597, "maximum": 0.6543568186110584, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04851236813841521, "standard_deviation": 0.03905472038853424, "minimum": -0.09715647679025288, "maximum": 0.016648775086266335, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9965265898996374, "standard_deviation": 0.0005517722400658056, "minimum": 0.9954707085784424, "maximum": 0.9971065862028851, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8819526302296399, "standard_deviation": 0.02123795502564556, "minimum": 0.8479247957016299, "maximum": 0.911711058144992, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5072706238672643, "standard_deviation": 0.20012567417765412, "minimum": 0.22491830803961646, "maximum": 0.8201990215833869, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06117474432057389, "standard_deviation": 0.00459597340358793, "minimum": 0.05607522793301722, "maximum": 0.06967403752596477, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999008978063066, "standard_deviation": 0.0002501508472100843, "minimum": 0.9986414603942732, "maximum": 0.9993078600196263, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4639009390544202, "standard_deviation": 0.020360531067458453, "minimum": 0.43750393310717484, "maximum": 0.49097872649221785, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47541913811350556, "standard_deviation": 0.2580011468457307, "minimum": 0.10332143193182293, "maximum": 0.8433101455385538, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031806608926607735, "standard_deviation": 0.0036595888171441614, "minimum": 0.027079974430515057, "maximum": 0.0375728200316129, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0004", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.5008692741394043, "predicted_mos": 3.5008692741394043, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "a5b0ee9628d6eb39734129d2f9fc65cf7649a20d62772de21b961b6120e519fe", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.539925800170749, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.880066206721599, "全帯域_10ms_rms_envelope_nrmse": 0.06132670536264165, "全帯域_10ms_envelope_correlation": 0.9965821838915471, "全帯域_10ms_modulation_mae_db": 0.5079393890822447, "全帯域_40ms_log_rms_mae_db": 0.46428433793877205, "全帯域_40ms_rms_envelope_nrmse": 0.032848411828192856, "全帯域_40ms_envelope_correlation": 0.9990277228916615, "全帯域_40ms_modulation_mae_db": 0.4295714919088461, "300–1000 Hz_10ms_log_rms_mae_db": 1.5167815131813118, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10046994697570802, "300–1000 Hz_10ms_envelope_correlation": 0.9921272882490937, "300–1000 Hz_10ms_modulation_mae_db": 0.6756543510014938, "300–1000 Hz_40ms_log_rms_mae_db": 0.8862431320384795, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058531867825829455, "300–1000 Hz_40ms_envelope_correlation": 0.9972721145979583, "300–1000 Hz_40ms_modulation_mae_db": 0.7166992079492625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3916371694818261, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15948496798111655, "1000–3000 Hz_10ms_envelope_correlation": 0.9813294616147479, "1000–3000 Hz_10ms_modulation_mae_db": 0.843998127956775, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7745865837689994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08751629203136667, "1000–3000 Hz_40ms_envelope_correlation": 0.9948462436183291, "1000–3000 Hz_40ms_modulation_mae_db": 1.0886195564126497, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4532662980728965, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18546767355302793, "3000–6000 Hz_10ms_envelope_correlation": 0.9777207375155758, "3000–6000 Hz_10ms_modulation_mae_db": 0.536775410538891, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8050657915007488, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10643244225815277, "3000–6000 Hz_40ms_envelope_correlation": 0.9936803645191641, "3000–6000 Hz_40ms_modulation_mae_db": 0.6177816775451156, "mr_stft_db": 5.390594564933614, "band_mae_db": 0.22468388411537013, "rms_gain_error_db": -0.09961793112028307, "output_peak": 0.5802541766748919}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8876664022715216, "全帯域_10ms_rms_envelope_nrmse": 0.06387587309267974, "全帯域_10ms_envelope_correlation": 0.9961935989898578, "全帯域_10ms_modulation_mae_db": 0.35907223509513714, "全帯域_40ms_log_rms_mae_db": 0.5019980096038371, "全帯域_40ms_rms_envelope_nrmse": 0.03256818795107038, "全帯域_40ms_envelope_correlation": 0.9989281054427027, "全帯域_40ms_modulation_mae_db": 0.27471871164862366, "300–1000 Hz_10ms_log_rms_mae_db": 1.6078655824104804, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10442794588033323, "300–1000 Hz_10ms_envelope_correlation": 0.9913328785754237, "300–1000 Hz_10ms_modulation_mae_db": 0.48768466269909977, "300–1000 Hz_40ms_log_rms_mae_db": 0.9344408844405802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06059666659050704, "300–1000 Hz_40ms_envelope_correlation": 0.9968595451305116, "300–1000 Hz_40ms_modulation_mae_db": 0.6203318972716162, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4093249632087657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1404241473920145, "1000–3000 Hz_10ms_envelope_correlation": 0.9850950217376315, "1000–3000 Hz_10ms_modulation_mae_db": 0.22292895198436033, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8160814089379582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07666254231359543, "1000–3000 Hz_40ms_envelope_correlation": 0.9952698621823416, "1000–3000 Hz_40ms_modulation_mae_db": 0.2875346602771201, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4300509814646518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18957442511053466, "3000–6000 Hz_10ms_envelope_correlation": 0.9758417931404824, "3000–6000 Hz_10ms_modulation_mae_db": 1.9564996528049736, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8495184377168439, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11926760226848461, "3000–6000 Hz_40ms_envelope_correlation": 0.9902612841173657, "3000–6000 Hz_40ms_modulation_mae_db": 2.1802740252995343, "mr_stft_db": 5.392574380487684, "band_mae_db": 0.15106669350641058, "rms_gain_error_db": -0.018352393607019714, "output_peak": 0.5574201129588657}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9278087686679911, "全帯域_10ms_rms_envelope_nrmse": 0.06778583265639389, "全帯域_10ms_envelope_correlation": 0.9957193166066224, "全帯域_10ms_modulation_mae_db": 0.7053888442259307, "全帯域_40ms_log_rms_mae_db": 0.5085352342687629, "全帯域_40ms_rms_envelope_nrmse": 0.03747373062499291, "全帯域_40ms_envelope_correlation": 0.9985842273642471, "全帯域_40ms_modulation_mae_db": 0.915010127179686, "300–1000 Hz_10ms_log_rms_mae_db": 1.5350037841100792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10411009327187254, "300–1000 Hz_10ms_envelope_correlation": 0.9913784836095288, "300–1000 Hz_10ms_modulation_mae_db": 0.6010519370995554, "300–1000 Hz_40ms_log_rms_mae_db": 0.9139575576787632, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060420989557321066, "300–1000 Hz_40ms_envelope_correlation": 0.9968393380249571, "300–1000 Hz_40ms_modulation_mae_db": 0.7617399420343253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4270283932701255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15023755780668918, "1000–3000 Hz_10ms_envelope_correlation": 0.9831096993552058, "1000–3000 Hz_10ms_modulation_mae_db": 0.4941493733987767, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8087232385665168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08464636244941506, "1000–3000 Hz_40ms_envelope_correlation": 0.994595885926501, "1000–3000 Hz_40ms_modulation_mae_db": 0.6169978420425996, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4425468948276192, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19387611043174122, "3000–6000 Hz_10ms_envelope_correlation": 0.9749944691197707, "3000–6000 Hz_10ms_modulation_mae_db": 1.1318636283837573, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8518434387376415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11909399471051009, "3000–6000 Hz_40ms_envelope_correlation": 0.9899307493850686, "3000–6000 Hz_40ms_modulation_mae_db": 1.3421832562271743, "mr_stft_db": 5.389102924064532, "band_mae_db": 0.1530202883189591, "rms_gain_error_db": -0.030300404449477274, "output_peak": 0.6051182293467673}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8535441912651028, "全帯域_10ms_rms_envelope_nrmse": 0.0651855971883784, "全帯域_10ms_envelope_correlation": 0.9960379964242694, "全帯域_10ms_modulation_mae_db": 0.7348878958511246, "全帯域_40ms_log_rms_mae_db": 0.47476482834544725, "全帯域_40ms_rms_envelope_nrmse": 0.03672298916802937, "全帯域_40ms_envelope_correlation": 0.9986446876115602, "全帯域_40ms_modulation_mae_db": 0.8298945924700197, "300–1000 Hz_10ms_log_rms_mae_db": 1.5678311143339179, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10223440838752727, "300–1000 Hz_10ms_envelope_correlation": 0.9916763983340818, "300–1000 Hz_10ms_modulation_mae_db": 0.7291290028791044, "300–1000 Hz_40ms_log_rms_mae_db": 0.9073081002434898, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060884903450760385, "300–1000 Hz_40ms_envelope_correlation": 0.9968080724579191, "300–1000 Hz_40ms_modulation_mae_db": 0.8033822274081638, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4134544501285582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13784534418970645, "1000–3000 Hz_10ms_envelope_correlation": 0.9858650883118283, "1000–3000 Hz_10ms_modulation_mae_db": 0.7408097819421192, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7881825822068864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07639344534845396, "1000–3000 Hz_40ms_envelope_correlation": 0.9956738178414611, "1000–3000 Hz_40ms_modulation_mae_db": 0.9402967108404529, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4910284396786033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15092944739277608, "3000–6000 Hz_10ms_envelope_correlation": 0.9848066952811473, "3000–6000 Hz_10ms_modulation_mae_db": 0.5274749619682181, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8580876562954322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08238865478770081, "3000–6000 Hz_40ms_envelope_correlation": 0.995512037269101, "3000–6000 Hz_40ms_modulation_mae_db": 0.8702770111762907, "mr_stft_db": 5.3787450249534325, "band_mae_db": 0.10324159173639946, "rms_gain_error_db": -0.032702322550643384, "output_peak": 0.6435174122006169}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.908751729714706, "全帯域_10ms_rms_envelope_nrmse": 0.06619585594401624, "全帯域_10ms_envelope_correlation": 0.9959418754639244, "全帯域_10ms_modulation_mae_db": 0.47624592813203015, "全帯域_40ms_log_rms_mae_db": 0.4954623852759317, "全帯域_40ms_rms_envelope_nrmse": 0.035333178524637444, "全帯域_40ms_envelope_correlation": 0.9987820913199182, "全帯域_40ms_modulation_mae_db": 0.2937090281588657, "300–1000 Hz_10ms_log_rms_mae_db": 1.6414892595912205, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10955663407289322, "300–1000 Hz_10ms_envelope_correlation": 0.9904625061031453, "300–1000 Hz_10ms_modulation_mae_db": 0.603009252363665, "300–1000 Hz_40ms_log_rms_mae_db": 0.9940078215923471, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06639377807962381, "300–1000 Hz_40ms_envelope_correlation": 0.9962470484570726, "300–1000 Hz_40ms_modulation_mae_db": 0.7392522102869837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.436739289091593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13337887845340674, "1000–3000 Hz_10ms_envelope_correlation": 0.9867036742050344, "1000–3000 Hz_10ms_modulation_mae_db": 0.44878693452017937, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8154304557146227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07227249045907372, "1000–3000 Hz_40ms_envelope_correlation": 0.9960433993196786, "1000–3000 Hz_40ms_modulation_mae_db": 0.3185565196298711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4343745195613693, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18681646749459027, "3000–6000 Hz_10ms_envelope_correlation": 0.9767960458968923, "3000–6000 Hz_10ms_modulation_mae_db": 0.6940913068741659, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8416602623142535, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11052145770852681, "3000–6000 Hz_40ms_envelope_correlation": 0.9921550640568262, "3000–6000 Hz_40ms_modulation_mae_db": 1.1616177813998652, "mr_stft_db": 5.379936014368876, "band_mae_db": 0.1290494536560793, "rms_gain_error_db": -0.06013018364069654, "output_peak": 0.642467255850341}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.910630828979479, "全帯域_10ms_rms_envelope_nrmse": 0.06256957131157359, "全帯域_10ms_envelope_correlation": 0.9964300434346728, "全帯域_10ms_modulation_mae_db": 0.481128596957561, "全帯域_40ms_log_rms_mae_db": 0.47245447871203183, "全帯域_40ms_rms_envelope_nrmse": 0.03358098753692311, "全帯域_40ms_envelope_correlation": 0.998964968177644, "全帯域_40ms_modulation_mae_db": 0.6437679230841175, "300–1000 Hz_10ms_log_rms_mae_db": 1.5667871101957624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09378430976751695, "300–1000 Hz_10ms_envelope_correlation": 0.9931445202388766, "300–1000 Hz_10ms_modulation_mae_db": 0.5720727250518899, "300–1000 Hz_40ms_log_rms_mae_db": 0.9591021662284106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05360208653013811, "300–1000 Hz_40ms_envelope_correlation": 0.9977134430156477, "300–1000 Hz_40ms_modulation_mae_db": 0.651532964837375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4468512629263794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1533226687753193, "1000–3000 Hz_10ms_envelope_correlation": 0.9833443872692054, "1000–3000 Hz_10ms_modulation_mae_db": 0.8072446325963677, "1000–3000 Hz_40ms_log_rms_mae_db": 0.817624095396018, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10101281647873113, "1000–3000 Hz_40ms_envelope_correlation": 0.9932968909275114, "1000–3000 Hz_40ms_modulation_mae_db": 0.7576363264398701, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3928112042170666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15374330949964654, "3000–6000 Hz_10ms_envelope_correlation": 0.9843144727680107, "3000–6000 Hz_10ms_modulation_mae_db": 0.5134813832477529, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8096713050820263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08342638581540372, "3000–6000 Hz_40ms_envelope_correlation": 0.9955492594674088, "3000–6000 Hz_40ms_modulation_mae_db": 0.631580427172134, "mr_stft_db": 5.3765840564410174, "band_mae_db": 0.21266614636466086, "rms_gain_error_db": -0.09360333680954473, "output_peak": 0.525686791016027}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.890465417743498, "全帯域_10ms_rms_envelope_nrmse": 0.06005013261211012, "全帯域_10ms_envelope_correlation": 0.9967345321617339, "全帯域_10ms_modulation_mae_db": 0.6710582523390257, "全帯域_40ms_log_rms_mae_db": 0.46840044040249157, "全帯域_40ms_rms_envelope_nrmse": 0.032158588668261504, "全帯域_40ms_envelope_correlation": 0.9990808703877481, "全帯域_40ms_modulation_mae_db": 0.4180995497655127, "300–1000 Hz_10ms_log_rms_mae_db": 1.5870775953391891, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09474951018394678, "300–1000 Hz_10ms_envelope_correlation": 0.9930802264848805, "300–1000 Hz_10ms_modulation_mae_db": 0.7733456381249745, "300–1000 Hz_40ms_log_rms_mae_db": 0.9753157987844371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055031956538824464, "300–1000 Hz_40ms_envelope_correlation": 0.9976870055398611, "300–1000 Hz_40ms_modulation_mae_db": 0.7531780078434611, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4636601052789893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1463926382426888, "1000–3000 Hz_10ms_envelope_correlation": 0.9841808633899498, "1000–3000 Hz_10ms_modulation_mae_db": 0.5293819373462653, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8382111434479297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0809942221259803, "1000–3000 Hz_40ms_envelope_correlation": 0.9953799037688191, "1000–3000 Hz_40ms_modulation_mae_db": 0.41892635095662545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3783714927107473, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17507525355993725, "3000–6000 Hz_10ms_envelope_correlation": 0.9793577548623094, "3000–6000 Hz_10ms_modulation_mae_db": 0.43300389391598254, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7460733657404228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09075495003421044, "3000–6000 Hz_40ms_envelope_correlation": 0.9940749377781007, "3000–6000 Hz_40ms_modulation_mae_db": 0.5681927844721785, "mr_stft_db": 5.380954535193094, "band_mae_db": 0.16673450697540834, "rms_gain_error_db": -0.10248258622785753, "output_peak": 0.5784710762006773}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.863577534579087, "全帯域_10ms_rms_envelope_nrmse": 0.06780344017876119, "全帯域_10ms_envelope_correlation": 0.9957232755982488, "全帯域_10ms_modulation_mae_db": 0.6470195764154273, "全帯域_40ms_log_rms_mae_db": 0.4860647351168409, "全帯域_40ms_rms_envelope_nrmse": 0.03476777879795351, "全帯域_40ms_envelope_correlation": 0.9987989308970706, "全帯域_40ms_modulation_mae_db": 0.7888049775504699, "300–1000 Hz_10ms_log_rms_mae_db": 1.5708835941461463, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10482915323394436, "300–1000 Hz_10ms_envelope_correlation": 0.9912719410863967, "300–1000 Hz_10ms_modulation_mae_db": 1.0541362518848179, "300–1000 Hz_40ms_log_rms_mae_db": 0.9470429222064916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06282582200198324, "300–1000 Hz_40ms_envelope_correlation": 0.9966279078457443, "300–1000 Hz_40ms_modulation_mae_db": 1.350977736985353, "1000–3000 Hz_10ms_log_rms_mae_db": 1.401589912738671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14242741337459158, "1000–3000 Hz_10ms_envelope_correlation": 0.9846750393983609, "1000–3000 Hz_10ms_modulation_mae_db": 0.3214650898347298, "1000–3000 Hz_40ms_log_rms_mae_db": 0.797043286248491, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08395519110583301, "1000–3000 Hz_40ms_envelope_correlation": 0.9943290950287781, "1000–3000 Hz_40ms_modulation_mae_db": 0.30665032563411976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4217777300078818, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1731979897858776, "3000–6000 Hz_10ms_envelope_correlation": 0.9806863194653412, "3000–6000 Hz_10ms_modulation_mae_db": 1.5978802329535042, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8072106866264878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11028940910129295, "3000–6000 Hz_40ms_envelope_correlation": 0.9927637814300768, "3000–6000 Hz_40ms_modulation_mae_db": 1.8701267665128147, "mr_stft_db": 5.39812619166324, "band_mae_db": 0.180857516535841, "rms_gain_error_db": -0.04763809769758431, "output_peak": 0.5461194132758378}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85280, "output_samples": 85280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.427114575037156, "spectral_convergence": 0.18465084926598288, "spectral_convergence_status": "ok", "amplitude_floor": 1.2882929606468862e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.19174986763863, "spectral_convergence": 0.1841757581768816, "spectral_convergence_status": "ok", "amplitude_floor": 9.933866418820512e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.552919252125053, "spectral_convergence": 0.17977423259533878, "spectral_convergence_status": "ok", "amplitude_floor": 8.99033450142914e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001234751326260816, "output_power": 0.0012394583594694515, "delta_db": 0.01652439586394932, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018448232014630782, "output_power": 0.0018064572933935547, "delta_db": -0.09127053066302618, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006193820215877917, "output_power": 0.0005890622863623394, "delta_db": -0.21797376496506718, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017436605640499536, "output_power": 0.0001580491381044348, "delta_db": -0.42669813501460163, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010535962705358282, "output_power": 9.463409945036037e-05, "delta_db": -0.4662657098184888, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.506722876313705e-05, "output_power": 1.552281419062901e-05, "delta_db": 0.12937076836708777, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22468388411537013, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09961793112028307, "pitch": null, "output_rms_dbfs": -24.08580976164317, "output_peak": 0.5802541766748919, "reference_rms_dbfs": -23.986191830522884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.390594564933614}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.0001748275799785616, "log_rms_mae_db": 0.880066206721599, "rms_envelope_nrmse": 0.06132670536264165, "envelope_correlation": 0.9965821838915471, "rms_gain_error_db": -0.09961816781947623, "reference_mean_power": 0.003997498793754849, "output_mean_power": 0.003906848087392253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.9383161701154944e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012108392795166332, "output_power": 0.001173492938704573, "reference_active": true, "delta_db": -0.1360602033791019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002231683599850243, "output_power": 0.00020715354839627938, "reference_active": true, "delta_db": -0.3234024504496868}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.764859371015312e-05, "output_power": 5.294478895664139e-05, "reference_active": true, "delta_db": -1.0643555134179452}], "active_band_count": 3, "active_band_mae_db": 0.5079393890822447}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00017019095451879833, "log_rms_mae_db": 0.46428433793877205, "rms_envelope_nrmse": 0.032848411828192856, "envelope_correlation": 0.9990277228916615, "rms_gain_error_db": -0.09961838250788901, "reference_mean_power": 0.00402014109400192, "output_mean_power": 0.003928976737222209}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.6741447857886242e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011176954288226882, "output_power": 0.0010864501633491908, "reference_active": true, "delta_db": -0.12313665193826664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.194819962775497e-05, "output_power": 8.335681258135017e-05, "reference_active": true, "delta_db": -0.4260213043432703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3053605179616555e-06, "output_power": 2.787810074908232e-06, "reference_active": true, "delta_db": -0.7395565194450013}], "active_band_count": 3, "active_band_mae_db": 0.4295714919088461}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.00014662911909855806, "log_rms_mae_db": 1.5167815131813118, "rms_envelope_nrmse": 0.10046994697570802, "envelope_correlation": 0.9921272882490937, "rms_gain_error_db": -0.16328571308746387, "reference_mean_power": 0.0015423797799867744, "output_mean_power": 0.0014854661228663585}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.0119951767590758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000652626948773928, "output_power": 0.000624389204186331, "reference_active": true, "delta_db": -0.1920961762667448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016341376588393177, "output_power": 0.00014628201296472906, "reference_active": true, "delta_db": -0.48097710492831103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.788084498799748e-05, "output_power": 3.505684503874899e-05, "reference_active": true, "delta_db": -1.3538897718094258}], "active_band_count": 3, "active_band_mae_db": 0.6756543510014938}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00013732417437068112, "log_rms_mae_db": 0.8862431320384795, "rms_envelope_nrmse": 0.058531867825829455, "envelope_correlation": 0.9972721145979583, "rms_gain_error_db": -0.16328729269175768, "reference_mean_power": 0.0015511172002359399, "output_mean_power": 0.0014938805898460068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 8.798155816797883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006408046711327789, "output_power": 0.0006091475718279559, "reference_active": true, "delta_db": -0.22003151495557285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.486060187757687e-05, "output_power": 6.600823302776297e-05, "reference_active": true, "delta_db": -0.5465520721806715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5748897665955177e-06, "output_power": 2.5996282693109712e-06, "reference_active": true, "delta_db": -1.3835140367115428}], "active_band_count": 3, "active_band_mae_db": 0.7166992079492625}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 8.659636175169924e-05, "log_rms_mae_db": 1.3916371694818261, "rms_envelope_nrmse": 0.15948496798111655, "envelope_correlation": 0.9813294616147479, "rms_gain_error_db": -0.35592678170660974, "reference_mean_power": 0.00020984803844209803, "output_mean_power": 0.00019333578041897252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 2.0976407038326697e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011684508155422031, "output_power": 0.000107170847142403, "reference_active": true, "delta_db": -0.3753377238747273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.122844755814475e-05, "output_power": 3.0211233002804103e-05, "reference_active": true, "delta_db": -1.3502853153498082}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.871758765973707e-06, "output_power": 7.368389751540427e-06, "reference_active": true, "delta_db": -0.8063713446457893}], "active_band_count": 3, "active_band_mae_db": 0.843998127956775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 7.401841529112696e-05, "log_rms_mae_db": 0.7745865837689994, "rms_envelope_nrmse": 0.08751629203136667, "envelope_correlation": 0.9948462436183291, "rms_gain_error_db": -0.35592394378152903, "reference_mean_power": 0.000211036168686384, "output_mean_power": 0.00019443054761958197}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.8857384691636124e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011739951182146869, "output_power": 0.00010698613364825087, "reference_active": true, "delta_db": -0.4033879808673099}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97467891176206e-05, "output_power": 1.394850875369234e-05, "reference_active": true, "delta_db": -1.5096870890396383}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.379135569148535e-07, "output_power": 4.6717917125651084e-07, "reference_active": true, "delta_db": -1.3527835993310011}], "active_band_count": 3, "active_band_mae_db": 1.0886195564126497}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 4.122543101057287e-05, "log_rms_mae_db": 1.4532662980728965, "rms_envelope_nrmse": 0.18546767355302793, "envelope_correlation": 0.9777207375155758, "rms_gain_error_db": -0.4885060697217266, "reference_mean_power": 4.5035535336143536e-05, "output_mean_power": 4.02443319520161e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 3.018630059598249e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5795382838093414e-05, "output_power": 1.4584058310368979e-05, "reference_active": true, "delta_db": -0.34651764454437517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.352602360298049e-06, "output_power": 3.343187741352178e-06, "reference_active": true, "delta_db": -1.1458822754023972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848555477734869e-06, "output_power": 1.7990360155176355e-06, "reference_active": true, "delta_db": -0.11792631166990086}], "active_band_count": 3, "active_band_mae_db": 0.536775410538891}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 3.1688229259308314e-05, "log_rms_mae_db": 0.8050657915007488, "rms_envelope_nrmse": 0.10643244225815277, "envelope_correlation": 0.9936803645191641, "rms_gain_error_db": -0.4885061884155659, "reference_mean_power": 4.529062388817587e-05, "output_mean_power": 4.047228124747573e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 2.7414916263445066e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5534436484200316e-05, "output_power": 1.4147456395782876e-05, "reference_active": true, "delta_db": -0.4061713966940725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0764100435071083e-06, "output_power": 1.641440271656829e-06, "reference_active": true, "delta_db": -1.020880365285972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4701562809614912e-07, "output_power": 1.332705458870324e-07, "reference_active": true, "delta_db": -0.42629327065530237}], "active_band_count": 3, "active_band_mae_db": 0.6177816775451156}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9422841185170915, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f094305b0215bb4c88a05d5ba53522c43f9f99564955959135610e95fdc813e7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9842879044102455, "standard_deviation": 0.0016973507904728396, "minimum": 0.9813294616147479, "maximum": 0.9867036742050344, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4237856932656134, "standard_deviation": 0.02432613978695435, "minimum": 1.3916371694818261, "maximum": 1.4636601052789893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5510956036974466, "standard_deviation": 0.22740794922162452, "minimum": 0.22292895198436033, "maximum": 0.843998127956775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14543920202694166, "standard_deviation": 0.008624761474523712, "minimum": 0.13337887845340674, "maximum": 0.15948496798111655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949293873266775, "standard_deviation": 0.0008666535887486099, "minimum": 0.9932968909275114, "maximum": 0.9960433993196786, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8069853492859278, "standard_deviation": 0.019811180815278086, "minimum": 0.7745865837689994, "maximum": 0.8382111434479297, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5919022865291635, "standard_deviation": 0.3104952715667293, "minimum": 0.2875346602771201, "maximum": 1.0886195564126497, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08293167028905617, "standard_deviation": 0.008874977175815828, "minimum": 0.07227249045907372, "maximum": 0.10101281647873113, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793147860061912, "standard_deviation": 0.0037158491689711364, "minimum": 0.9749944691197707, "maximum": 0.9848066952811473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4305284450676046, "standard_deviation": 0.03496168444157995, "minimum": 1.3783714927107473, "maximum": 1.4910284396786033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9238838088359056, "standard_deviation": 0.5773106353656129, "minimum": 0.43300389391598254, "maximum": 1.9564996528049736, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17608508460351643, "standard_deviation": 0.016226057240386393, "minimum": 0.15092944739277608, "maximum": 0.19387611043174122, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9929909347528889, "standard_deviation": 0.002142007227328853, "minimum": 0.9899307493850686, "maximum": 0.9955492594674088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8211413680017321, "standard_deviation": 0.03732709438365153, "minimum": 0.7460733657404228, "maximum": 0.8580876562954322, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1552542162256383, "standard_deviation": 0.6077658656721, "minimum": 0.5681927844721785, "maximum": 2.1802740252995343, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10277186208553528, "standard_deviation": 0.0151309513315248, "minimum": 0.08238865478770081, "maximum": 0.11926760226848461, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918092803351783, "standard_deviation": 0.0009282883735499282, "minimum": 0.9904625061031453, "maximum": 0.9931445202388766, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5742149441635136, "standard_deviation": 0.03921877233958475, "minimum": 1.5167815131813118, "maximum": 1.6414892595912205, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.687010477638075, "standard_deviation": 0.1738326657790674, "minimum": 0.48768466269909977, "maximum": 1.0541362518848179, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1017702502217178, "standard_deviation": 0.005311646768856268, "minimum": 0.09378430976751695, "maximum": 0.10955663407289322, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997006809383709, "standard_deviation": 0.0005130326612268774, "minimum": 0.9962470484570726, "maximum": 0.9977134430156477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9396772979016249, "standard_deviation": 0.03634239276300532, "minimum": 0.8862431320384795, "maximum": 0.9940078215923471, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7996367743270676, "standard_deviation": 0.23058787764474759, "minimum": 0.6203318972716162, "maximum": 1.350977736985353, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059786008821873446, "standard_deviation": 0.004098532705204472, "minimum": 0.05360208653013811, "maximum": 0.06639377807962381, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1651650101511411, "standard_deviation": 0.04058971077602883, "minimum": 0.10324159173639946, "maximum": 0.22468388411537013, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.385827211513186, "standard_deviation": 0.007785422400917775, "minimum": 5.3765840564410174, "maximum": 5.39812619166324, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5848818084405031, "standard_deviation": 0.04307296311728421, "minimum": 0.525686791016027, "maximum": 0.6435174122006169, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06060340701288832, "standard_deviation": 0.03383828750306808, "minimum": -0.10248258622785753, "maximum": -0.018352393607019714, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9961703528213596, "standard_deviation": 0.0003833898773926342, "minimum": 0.9957193166066224, "maximum": 0.9967345321617339, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.890313884992873, "standard_deviation": 0.02489259350524233, "minimum": 0.8535441912651028, "maximum": 0.9278087686679911, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5728425897623102, "standard_deviation": 0.13449816463401093, "minimum": 0.35907223509513714, "maximum": 0.7348878958511246, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06434912604331935, "standard_deviation": 0.002899765716754277, "minimum": 0.06005013261211012, "maximum": 0.06780344017876119, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988514505115691, "standard_deviation": 0.00017896359773743797, "minimum": 0.9985842273642471, "maximum": 0.9990808703877481, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48399555620801443, "standard_deviation": 0.01652788279203163, "minimum": 0.46428433793877205, "maximum": 0.5085352342687629, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5741970502207676, "standard_deviation": 0.2524381799631562, "minimum": 0.27471871164862366, "maximum": 0.915010127179686, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03443173163750764, "standard_deviation": 0.001974834324643643, "minimum": 0.032158588668261504, "maximum": 0.03747373062499291, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0004", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.347336769104004, "predicted_mos": 3.347336769104004, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "23c566cc3b64a1dd86eb5d3879790af6a1ab19fbc004fad9e1f2255e67cb5eeb", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.4048872999846935, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.904079591047336, "全帯域_10ms_rms_envelope_nrmse": 0.06351861451516237, "全帯域_10ms_envelope_correlation": 0.9962523866943708, "全帯域_10ms_modulation_mae_db": 0.5873656294916393, "全帯域_40ms_log_rms_mae_db": 0.45939132773294983, "全帯域_40ms_rms_envelope_nrmse": 0.03366398896153667, "全帯域_40ms_envelope_correlation": 0.9988740725505354, "全帯域_40ms_modulation_mae_db": 0.5016988793000264, "300–1000 Hz_10ms_log_rms_mae_db": 1.7510870357524186, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10495488857648767, "300–1000 Hz_10ms_envelope_correlation": 0.9912517084092689, "300–1000 Hz_10ms_modulation_mae_db": 0.5315562518599093, "300–1000 Hz_40ms_log_rms_mae_db": 0.9738853152057648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0576927264778925, "300–1000 Hz_40ms_envelope_correlation": 0.9971610409734996, "300–1000 Hz_40ms_modulation_mae_db": 0.6413466750475404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5215906988674446, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1368585143783422, "1000–3000 Hz_10ms_envelope_correlation": 0.9861263974100279, "1000–3000 Hz_10ms_modulation_mae_db": 0.9662396099640537, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8001930146550236, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07984268416688826, "1000–3000 Hz_40ms_envelope_correlation": 0.9952352196188545, "1000–3000 Hz_40ms_modulation_mae_db": 1.1226953810854017, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6434582744356894, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.171639379323786, "3000–6000 Hz_10ms_envelope_correlation": 0.9800852799449342, "3000–6000 Hz_10ms_modulation_mae_db": 0.9153282631540041, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8871761214812321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08703456465054907, "3000–6000 Hz_40ms_envelope_correlation": 0.9945496677204799, "3000–6000 Hz_40ms_modulation_mae_db": 0.998265592103397, "mr_stft_db": 5.508193904993118, "band_mae_db": 0.1188015432087937, "rms_gain_error_db": -0.04580590509833145, "output_peak": 0.5818677398100552}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8726642558681263, "全帯域_10ms_rms_envelope_nrmse": 0.06671221597303446, "全帯域_10ms_envelope_correlation": 0.9958503436940971, "全帯域_10ms_modulation_mae_db": 0.4739545634018545, "全帯域_40ms_log_rms_mae_db": 0.4734947467348608, "全帯域_40ms_rms_envelope_nrmse": 0.03758652080546397, "全帯域_40ms_envelope_correlation": 0.9985706002681589, "全帯域_40ms_modulation_mae_db": 0.3738998794090884, "300–1000 Hz_10ms_log_rms_mae_db": 1.5630405218925862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09760403655762998, "300–1000 Hz_10ms_envelope_correlation": 0.9924723136548163, "300–1000 Hz_10ms_modulation_mae_db": 0.5458188410331769, "300–1000 Hz_40ms_log_rms_mae_db": 0.9455790845728753, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05527799226465438, "300–1000 Hz_40ms_envelope_correlation": 0.997433637268996, "300–1000 Hz_40ms_modulation_mae_db": 0.5229640687929448, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5488371985210017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15316635435201037, "1000–3000 Hz_10ms_envelope_correlation": 0.982729678868545, "1000–3000 Hz_10ms_modulation_mae_db": 0.7016795682983915, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8489485633554342, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09043725841676123, "1000–3000 Hz_40ms_envelope_correlation": 0.9941288179518535, "1000–3000 Hz_40ms_modulation_mae_db": 0.6970368266339602, "3000–6000 Hz_10ms_log_rms_mae_db": 1.667199604831547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18151198409511018, "3000–6000 Hz_10ms_envelope_correlation": 0.9780336970548451, "3000–6000 Hz_10ms_modulation_mae_db": 1.1767553182612203, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9156924043043698, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09931702232598474, "3000–6000 Hz_40ms_envelope_correlation": 0.9936879492939261, "3000–6000 Hz_40ms_modulation_mae_db": 1.59238847823903, "mr_stft_db": 5.526447314508484, "band_mae_db": 0.18269379343550626, "rms_gain_error_db": -0.018661363257348105, "output_peak": 0.5271117623831943}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.907892954431139, "全帯域_10ms_rms_envelope_nrmse": 0.0638666718650486, "全帯域_10ms_envelope_correlation": 0.9962373074646559, "全帯域_10ms_modulation_mae_db": 0.39750487580954214, "全帯域_40ms_log_rms_mae_db": 0.4770663433623732, "全帯域_40ms_rms_envelope_nrmse": 0.032976731045887255, "全帯域_40ms_envelope_correlation": 0.9989533566951503, "全帯域_40ms_modulation_mae_db": 0.35916446577227323, "300–1000 Hz_10ms_log_rms_mae_db": 1.6032926498204934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10403827323703677, "300–1000 Hz_10ms_envelope_correlation": 0.9917313308284215, "300–1000 Hz_10ms_modulation_mae_db": 0.8362430824154007, "300–1000 Hz_40ms_log_rms_mae_db": 0.9445872221336797, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05843852093989422, "300–1000 Hz_40ms_envelope_correlation": 0.9975206359227845, "300–1000 Hz_40ms_modulation_mae_db": 0.7419959523214926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.608907021722672, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15998657136290886, "1000–3000 Hz_10ms_envelope_correlation": 0.9807181536910096, "1000–3000 Hz_10ms_modulation_mae_db": 0.1826329013632844, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8852259726528802, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0805938760330503, "1000–3000 Hz_40ms_envelope_correlation": 0.9946628724507232, "1000–3000 Hz_40ms_modulation_mae_db": 0.3867944770330661, "3000–6000 Hz_10ms_log_rms_mae_db": 1.604250788729958, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16483425194331575, "3000–6000 Hz_10ms_envelope_correlation": 0.9820222823093812, "3000–6000 Hz_10ms_modulation_mae_db": 0.3630105799384397, "3000–6000 Hz_40ms_log_rms_mae_db": 0.880632933896765, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08443368900156892, "3000–6000 Hz_40ms_envelope_correlation": 0.994975843164831, "3000–6000 Hz_40ms_modulation_mae_db": 0.22623419078698492, "mr_stft_db": 5.519958476297692, "band_mae_db": 0.10972970492396487, "rms_gain_error_db": -0.06146277573055259, "output_peak": 0.6213564021862095}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8990607418928274, "全帯域_10ms_rms_envelope_nrmse": 0.06512599202805103, "全帯域_10ms_envelope_correlation": 0.9961059814329374, "全帯域_10ms_modulation_mae_db": 0.7509068216022604, "全帯域_40ms_log_rms_mae_db": 0.4563281309061692, "全帯域_40ms_rms_envelope_nrmse": 0.032629193539745485, "全帯域_40ms_envelope_correlation": 0.9990047002965271, "全帯域_40ms_modulation_mae_db": 0.9160702924993789, "300–1000 Hz_10ms_log_rms_mae_db": 1.6670741735739285, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10766976417229257, "300–1000 Hz_10ms_envelope_correlation": 0.9908989642036846, "300–1000 Hz_10ms_modulation_mae_db": 0.8423515437031948, "300–1000 Hz_40ms_log_rms_mae_db": 0.9974525389898319, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06361737664730599, "300–1000 Hz_40ms_envelope_correlation": 0.9966811380103402, "300–1000 Hz_40ms_modulation_mae_db": 1.0361709942533377, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5938332840227551, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15884540700692557, "1000–3000 Hz_10ms_envelope_correlation": 0.9813710870381599, "1000–3000 Hz_10ms_modulation_mae_db": 0.5416514761033677, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8451866341447486, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09011892326479047, "1000–3000 Hz_40ms_envelope_correlation": 0.9942540715831687, "1000–3000 Hz_40ms_modulation_mae_db": 0.6594763514113275, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6401710122679567, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17399999299365307, "3000–6000 Hz_10ms_envelope_correlation": 0.9795612830601037, "3000–6000 Hz_10ms_modulation_mae_db": 0.634535672465384, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8954684831250537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11006087587646048, "3000–6000 Hz_40ms_envelope_correlation": 0.9912212205826202, "3000–6000 Hz_40ms_modulation_mae_db": 0.6853260126159157, "mr_stft_db": 5.508070598337292, "band_mae_db": 0.21392263982240542, "rms_gain_error_db": -0.08323879776771682, "output_peak": 0.5510706892713412}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8514400095528886, "全帯域_10ms_rms_envelope_nrmse": 0.057098404304323454, "全帯域_10ms_envelope_correlation": 0.996969353188452, "全帯域_10ms_modulation_mae_db": 0.3248884344235265, "全帯域_40ms_log_rms_mae_db": 0.4437159632930407, "全帯域_40ms_rms_envelope_nrmse": 0.027278884479909477, "全帯域_40ms_envelope_correlation": 0.9992636718741562, "全帯域_40ms_modulation_mae_db": 0.4582611915015918, "300–1000 Hz_10ms_log_rms_mae_db": 1.6547141298052461, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0976315653175905, "300–1000 Hz_10ms_envelope_correlation": 0.992432541492818, "300–1000 Hz_10ms_modulation_mae_db": 0.28664192433615293, "300–1000 Hz_40ms_log_rms_mae_db": 0.9313847537774691, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0520634828445921, "300–1000 Hz_40ms_envelope_correlation": 0.9976505352854623, "300–1000 Hz_40ms_modulation_mae_db": 0.4029681879647773, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5795103794033987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14727751491396707, "1000–3000 Hz_10ms_envelope_correlation": 0.9836644461526517, "1000–3000 Hz_10ms_modulation_mae_db": 1.1723391284454607, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8651681124951879, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08034043467810051, "1000–3000 Hz_40ms_envelope_correlation": 0.9948815168883094, "1000–3000 Hz_40ms_modulation_mae_db": 1.3596217943694382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.727026155101182, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17821693694571838, "3000–6000 Hz_10ms_envelope_correlation": 0.9785246386282023, "3000–6000 Hz_10ms_modulation_mae_db": 0.39590690176310367, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9412718521305969, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10397570427451781, "3000–6000 Hz_40ms_envelope_correlation": 0.9921768284289759, "3000–6000 Hz_40ms_modulation_mae_db": 0.23032983838452203, "mr_stft_db": 5.517279025458145, "band_mae_db": 0.05638983635225466, "rms_gain_error_db": -0.0394137669475842, "output_peak": 0.5522460034664297}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.89079536121656, "全帯域_10ms_rms_envelope_nrmse": 0.058494478437011874, "全帯域_10ms_envelope_correlation": 0.9968765253751343, "全帯域_10ms_modulation_mae_db": 0.4780250433810291, "全帯域_40ms_log_rms_mae_db": 0.45068206465061206, "全帯域_40ms_rms_envelope_nrmse": 0.0293525994954149, "全帯域_40ms_envelope_correlation": 0.9992168238646734, "全帯域_40ms_modulation_mae_db": 0.4866938181850717, "300–1000 Hz_10ms_log_rms_mae_db": 1.6771318513244988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09485289586011131, "300–1000 Hz_10ms_envelope_correlation": 0.9930972645889103, "300–1000 Hz_10ms_modulation_mae_db": 0.7563246178955568, "300–1000 Hz_40ms_log_rms_mae_db": 1.0078308808422005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05267982780257398, "300–1000 Hz_40ms_envelope_correlation": 0.9979297779337926, "300–1000 Hz_40ms_modulation_mae_db": 0.8287558999955026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5627715478060602, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14430604123021135, "1000–3000 Hz_10ms_envelope_correlation": 0.9850261289216883, "1000–3000 Hz_10ms_modulation_mae_db": 1.1079221075034678, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8606006344568577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08511344731838766, "1000–3000 Hz_40ms_envelope_correlation": 0.9952243388124045, "1000–3000 Hz_40ms_modulation_mae_db": 1.2449782145184531, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6587581851807354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19082061665733654, "3000–6000 Hz_10ms_envelope_correlation": 0.9762491857961387, "3000–6000 Hz_10ms_modulation_mae_db": 0.9883168218192253, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8861784127940077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1007686203936714, "3000–6000 Hz_40ms_envelope_correlation": 0.9944868540425348, "3000–6000 Hz_40ms_modulation_mae_db": 0.8952984022286317, "mr_stft_db": 5.5247546103640985, "band_mae_db": 0.19928267815985634, "rms_gain_error_db": -0.08555324591189163, "output_peak": 0.602620079169343}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8763483219684017, "全帯域_10ms_rms_envelope_nrmse": 0.058609789156402144, "全帯域_10ms_envelope_correlation": 0.9968886678308447, "全帯域_10ms_modulation_mae_db": 0.6666178166986195, "全帯域_40ms_log_rms_mae_db": 0.44688589637914, "全帯域_40ms_rms_envelope_nrmse": 0.03326839290214058, "全帯域_40ms_envelope_correlation": 0.9989893343920562, "全帯域_40ms_modulation_mae_db": 0.8447594892560941, "300–1000 Hz_10ms_log_rms_mae_db": 1.693125909848387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10267307383944384, "300–1000 Hz_10ms_envelope_correlation": 0.9918859453854125, "300–1000 Hz_10ms_modulation_mae_db": 0.9985426698028999, "300–1000 Hz_40ms_log_rms_mae_db": 0.9789820917500188, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05936148441637423, "300–1000 Hz_40ms_envelope_correlation": 0.9973216020327901, "300–1000 Hz_40ms_modulation_mae_db": 1.0536710819653357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5908192226588003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15563551208023962, "1000–3000 Hz_10ms_envelope_correlation": 0.983199780178791, "1000–3000 Hz_10ms_modulation_mae_db": 1.6340255693196157, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8802885622948059, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10267760845709997, "1000–3000 Hz_40ms_envelope_correlation": 0.9934543323312428, "1000–3000 Hz_40ms_modulation_mae_db": 1.961477565774338, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6626909113219854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19685665038483224, "3000–6000 Hz_10ms_envelope_correlation": 0.9737489327316857, "3000–6000 Hz_10ms_modulation_mae_db": 0.5725233867427596, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9348934850690648, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10398813736008561, "3000–6000 Hz_40ms_envelope_correlation": 0.9922033683781367, "3000–6000 Hz_40ms_modulation_mae_db": 0.7924754764346598, "mr_stft_db": 5.514409783563274, "band_mae_db": 0.21145039263356744, "rms_gain_error_db": -0.09424108578725233, "output_peak": 0.5670306941199313}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9334610802654872, "全帯域_10ms_rms_envelope_nrmse": 0.06552688597305495, "全帯域_10ms_envelope_correlation": 0.9959987343438266, "全帯域_10ms_modulation_mae_db": 0.3338914706612683, "全帯域_40ms_log_rms_mae_db": 0.5192895148990914, "全帯域_40ms_rms_envelope_nrmse": 0.03831634455192698, "全帯域_40ms_envelope_correlation": 0.998512274817502, "全帯域_40ms_modulation_mae_db": 0.07103266620210058, "300–1000 Hz_10ms_log_rms_mae_db": 1.6524163957675102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09781392373991758, "300–1000 Hz_10ms_envelope_correlation": 0.9923918518353877, "300–1000 Hz_10ms_modulation_mae_db": 0.5941299062248034, "300–1000 Hz_40ms_log_rms_mae_db": 0.9487696835852971, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058106219467137565, "300–1000 Hz_40ms_envelope_correlation": 0.9970485407535759, "300–1000 Hz_40ms_modulation_mae_db": 0.6497571709956865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.565110214642168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15932844830491266, "1000–3000 Hz_10ms_envelope_correlation": 0.9813074183223257, "1000–3000 Hz_10ms_modulation_mae_db": 0.6068160832087508, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8444446589732422, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1069460919254839, "1000–3000 Hz_40ms_envelope_correlation": 0.9915247801329871, "1000–3000 Hz_40ms_modulation_mae_db": 0.6474364897423562, "3000–6000 Hz_10ms_log_rms_mae_db": 1.633192514160681, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17448376226142537, "3000–6000 Hz_10ms_envelope_correlation": 0.9795550847449572, "3000–6000 Hz_10ms_modulation_mae_db": 1.0092019569650843, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8955424362796767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09370114853543914, "3000–6000 Hz_40ms_envelope_correlation": 0.9936852469817608, "3000–6000 Hz_40ms_modulation_mae_db": 1.0998434234703132, "mr_stft_db": 5.511474912225733, "band_mae_db": 0.13503571543088805, "rms_gain_error_db": 0.00011616188551943321, "output_peak": 0.5878074221160563}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85280, "output_samples": 85280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.617978704081326, "spectral_convergence": 0.18394847423431984, "spectral_convergence_status": "ok", "amplitude_floor": 1.2882929606468862e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.300696436636904, "spectral_convergence": 0.1857748441817338, "spectral_convergence_status": "ok", "amplitude_floor": 9.933866418820512e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.605906574261126, "spectral_convergence": 0.18198587987211273, "spectral_convergence_status": "ok", "amplitude_floor": 8.99033450142914e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001234751326260816, "output_power": 0.0012457246986543357, "delta_db": 0.03842573861099828, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018448232014630782, "output_power": 0.0018219605439716926, "delta_db": -0.054157841591401044, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006193820215877917, "output_power": 0.0006048235497377465, "delta_db": -0.10329902463217464, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017436605640499536, "output_power": 0.00016224344840988898, "delta_db": -0.31294776847535405, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010535962705358282, "output_power": 0.00010183257330376489, "delta_db": -0.1478750615894683, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.506722876313705e-05, "output_power": 1.5263135670214666e-05, "delta_db": 0.05610382435336597, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1188015432087937, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04580590509833145, "pitch": null, "output_rms_dbfs": -24.031997735621218, "output_peak": 0.5818677398100552, "reference_rms_dbfs": -23.986191830522884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.508193904993118}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.0001748275799785616, "log_rms_mae_db": 0.904079591047336, "rms_envelope_nrmse": 0.06351861451516237, "envelope_correlation": 0.9962523866943708, "rms_gain_error_db": -0.04580587882909613, "reference_mean_power": 0.003997498793754849, "output_mean_power": 0.003955557968753488}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.9383161701154944e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012108392795166332, "output_power": 0.0011956149358283564, "reference_active": true, "delta_db": -0.05495169303903791}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002231683599850243, "output_power": 0.00020563225697535937, "reference_active": true, "delta_db": -0.3554137963891951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.764859371015312e-05, "output_power": 4.955478918830806e-05, "reference_active": true, "delta_db": -1.3517313990466846}], "active_band_count": 3, "active_band_mae_db": 0.5873656294916393}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00017019095451879833, "log_rms_mae_db": 0.45939132773294983, "rms_envelope_nrmse": 0.03366398896153667, "envelope_correlation": 0.9988740725505354, "rms_gain_error_db": -0.04580636986016415, "reference_mean_power": 0.00402014109400192, "output_mean_power": 0.003977962261502589}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.6741447857886242e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011176954288226882, "output_power": 0.001108159689874437, "reference_active": true, "delta_db": -0.03721126417666288}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.194819962775497e-05, "output_power": 8.485988336556866e-05, "reference_active": true, "delta_db": -0.34840799524963434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3053605179616555e-06, "output_power": 2.554295259396617e-06, "reference_active": true, "delta_db": -1.119477378473782}], "active_band_count": 3, "active_band_mae_db": 0.5016988793000264}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.00014662911909855806, "log_rms_mae_db": 1.7510870357524186, "rms_envelope_nrmse": 0.10495488857648767, "envelope_correlation": 0.9912517084092689, "rms_gain_error_db": -0.09173459288168395, "reference_mean_power": 0.0015423797799867744, "output_mean_power": 0.001510142269698108}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.0119951767590758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000652626948773928, "output_power": 0.0006338326967372926, "reference_active": true, "delta_db": -0.12690364203634844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016341376588393177, "output_power": 0.00014981553704065037, "reference_active": true, "delta_db": -0.37731783033886457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.788084498799748e-05, "output_power": 3.7249210973080235e-05, "reference_active": true, "delta_db": -1.090447283204515}], "active_band_count": 3, "active_band_mae_db": 0.5315562518599093}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00013732417437068112, "log_rms_mae_db": 0.9738853152057648, "rms_envelope_nrmse": 0.0576927264778925, "envelope_correlation": 0.9971610409734996, "rms_gain_error_db": -0.09173609150134321, "reference_mean_power": 0.0015511172002359399, "output_mean_power": 0.0015186965437642482}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 8.798155816797883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006408046711327789, "output_power": 0.0006194872505120046, "reference_active": true, "delta_db": -0.14693296153733107}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.486060187757687e-05, "output_power": 6.787017003098513e-05, "reference_active": true, "delta_db": -0.4257437730620331}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5748897665955177e-06, "output_power": 2.6189446842238004e-06, "reference_active": true, "delta_db": -1.351363290543257}], "active_band_count": 3, "active_band_mae_db": 0.6413466750475404}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 8.659636175169924e-05, "log_rms_mae_db": 1.5215906988674446, "rms_envelope_nrmse": 0.1368585143783422, "envelope_correlation": 0.9861263974100279, "rms_gain_error_db": -0.23787591820126103, "reference_mean_power": 0.00020984803844209803, "output_mean_power": 0.00019866315220232536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 2.0976407038326697e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011684508155422031, "output_power": 0.00011725967273150017, "reference_active": true, "delta_db": 0.015382417099914437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.122844755814475e-05, "output_power": 3.3631699130116616e-05, "reference_active": true, "delta_db": -0.8844817275407103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.871758765973707e-06, "output_power": 5.599177763734959e-06, "reference_active": true, "delta_db": -1.998854685251536}], "active_band_count": 3, "active_band_mae_db": 0.9662396099640537}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 7.401841529112696e-05, "log_rms_mae_db": 0.8001930146550236, "rms_envelope_nrmse": 0.07984268416688826, "envelope_correlation": 0.9952352196188545, "rms_gain_error_db": -0.2378768797935279, "reference_mean_power": 0.000211036168686384, "output_mean_power": 0.0001997879109490357}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.8857384691636124e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011739951182146869, "output_power": 0.00011701247479225679, "reference_active": true, "delta_db": -0.014341263111179689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97467891176206e-05, "output_power": 1.5758084900944464e-05, "reference_active": true, "delta_db": -0.9799305219390876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.379135569148535e-07, "output_power": 3.693009255051758e-07, "reference_active": true, "delta_db": -2.3738143582059377}], "active_band_count": 3, "active_band_mae_db": 1.1226953810854017}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 4.122543101057287e-05, "log_rms_mae_db": 1.6434582744356894, "rms_envelope_nrmse": 0.171639379323786, "envelope_correlation": 0.9800852799449342, "rms_gain_error_db": -0.09609714979678045, "reference_mean_power": 4.5035535336143536e-05, "output_mean_power": 4.404996975686402e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 3.018630059598249e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5795382838093414e-05, "output_power": 1.5521922740314377e-05, "reference_active": true, "delta_db": -0.07584639088607852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.352602360298049e-06, "output_power": 3.56654525675157e-06, "reference_active": true, "delta_db": -0.8650125396867272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848555477734869e-06, "output_power": 1.2198878421774228e-06, "reference_active": true, "delta_db": -1.8051258588892063}], "active_band_count": 3, "active_band_mae_db": 0.9153282631540041}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 3.1688229259308314e-05, "log_rms_mae_db": 0.8871761214812321, "rms_envelope_nrmse": 0.08703456465054907, "envelope_correlation": 0.9945496677204799, "rms_gain_error_db": -0.09609743382746139, "reference_mean_power": 4.529062388817587e-05, "output_mean_power": 4.429947300893401e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 2.7414916263445066e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5534436484200316e-05, "output_power": 1.4835794102413639e-05, "reference_active": true, "delta_db": -0.19984706358142468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0764100435071083e-06, "output_power": 1.6777443332257561e-06, "reference_active": true, "delta_db": -0.9258734026454067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4701562809614912e-07, "output_power": 9.559955543456321e-08, "reference_active": true, "delta_db": -1.8690763100833596}], "active_band_count": 3, "active_band_mae_db": 0.998265592103397}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9371270386072421, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c3923044e2bacef82568d4f1c3653c151158a077cdcc77b8f568fd75a312d583", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9830178863228999, "standard_deviation": 0.0018957277864707607, "minimum": 0.9807181536910096, "maximum": 0.9861263974100279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5714224459555375, "standard_deviation": 0.027920723025143226, "minimum": 1.5215906988674446, "maximum": 1.608907021722672, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8641633055257989, "standard_deviation": 0.45017842464775165, "minimum": 0.1826329013632844, "maximum": 1.6340255693196157, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15192554545368972, "standard_deviation": 0.008363590396792613, "minimum": 0.1368585143783422, "maximum": 0.15998657136290886, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994170743721193, "standard_deviation": 0.0012255395678712885, "minimum": 0.9915247801329871, "maximum": 0.9952352196188545, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8537570191285225, "standard_deviation": 0.026521315723335853, "minimum": 0.8001930146550236, "maximum": 0.8852259726528802, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0099396375710428, "standard_deviation": 0.5121537349088146, "minimum": 0.3867944770330661, "maximum": 1.961477565774338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0895087905325703, "standard_deviation": 0.010385420163525292, "minimum": 0.07984268416688826, "maximum": 0.1069460919254839, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.978472548033781, "standard_deviation": 0.0025357425541618563, "minimum": 0.9737489327316857, "maximum": 0.9820222823093812, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.654593430753717, "standard_deviation": 0.035481802680945806, "minimum": 1.604250788729958, "maximum": 1.727026155101182, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7569473626386526, "standard_deviation": 0.30543193400777424, "minimum": 0.3630105799384397, "maximum": 1.1767553182612203, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1790454468256472, "standard_deviation": 0.010459017283628068, "minimum": 0.16483425194331575, "maximum": 0.19685665038483224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933733723241581, "standard_deviation": 0.0013533045961193954, "minimum": 0.9912212205826202, "maximum": 0.994975843164831, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9046070161350959, "standard_deviation": 0.023202068110306614, "minimum": 0.880632933896765, "maximum": 0.9412718521305969, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8150201767829318, "standard_deviation": 0.45218671910322955, "minimum": 0.22623419078698492, "maximum": 1.59238847823903, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09790997030228465, "standard_deviation": 0.008857057007754219, "minimum": 0.08443368900156892, "maximum": 0.11006087587646048, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920202400498399, "standard_deviation": 0.0007193096096714966, "minimum": 0.9908989642036846, "maximum": 0.9930972645889103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6577353334731337, "standard_deviation": 0.05656728479972886, "minimum": 1.5630405218925862, "maximum": 1.7510870357524186, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6739511046588869, "standard_deviation": 0.22675237271628038, "minimum": 0.28664192433615293, "maximum": 0.9985426698028999, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10090480266256377, "standard_deviation": 0.00451824189876654, "minimum": 0.09485289586011131, "maximum": 0.10766976417229257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973433635226552, "standard_deviation": 0.0003848913344386726, "minimum": 0.9966811380103402, "maximum": 0.9979297779337926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9660589463571421, "standard_deviation": 0.027620876740219152, "minimum": 0.9313847537774691, "maximum": 1.0078308808422005, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7347037539170771, "standard_deviation": 0.23063120895148537, "minimum": 0.4029681879647773, "maximum": 1.0536710819653357, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05715470385755312, "standard_deviation": 0.0037596760903710756, "minimum": 0.0520634828445921, "maximum": 0.06361737664730599, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1534132879959046, "standard_deviation": 0.05713211478099179, "minimum": 0.05638983635225466, "maximum": 0.21392263982240542, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.516323578218479, "standard_deviation": 0.007061698890554635, "minimum": 5.508070598337292, "maximum": 5.526447314508484, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.57388884906532, "standard_deviation": 0.030613379114327724, "minimum": 0.5271117623831943, "maximum": 0.6213564021862095, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.053532597326894714, "standard_deviation": 0.033766180285085885, "minimum": -0.09424108578725233, "maximum": 0.00011616188551943321, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963974125030399, "standard_deviation": 0.0004452645798369883, "minimum": 0.9958503436940971, "maximum": 0.996969353188452, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8919677895303457, "standard_deviation": 0.025164542739937658, "minimum": 0.8514400095528886, "maximum": 0.9334610802654872, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5016443319337175, "standard_deviation": 0.15503431438182538, "minimum": 0.3248884344235265, "maximum": 0.7509068216022604, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06236913153151111, "standard_deviation": 0.003721063034678771, "minimum": 0.057098404304323454, "maximum": 0.06671221597303446, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989231043448449, "standard_deviation": 0.0002698309161052026, "minimum": 0.998512274817502, "maximum": 0.9992636718741562, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46585674849477965, "standard_deviation": 0.02464615127218428, "minimum": 0.4437159632930407, "maximum": 0.5192895148990914, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5014475852657032, "standard_deviation": 0.2708789292253736, "minimum": 0.07103266620210058, "maximum": 0.9160702924993789, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033134081972753165, "standard_deviation": 0.0037002223271358167, "minimum": 0.027278884479909477, "maximum": 0.03831634455192698, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5694-0004", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.2831740379333496, "predicted_mos": 3.2831740379333496, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "bca059ecd48ac40d7de67482ea07d96e0589aacc32d04fa49c632d518166f491", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.406747500412166, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8972112094967184, "全帯域_10ms_rms_envelope_nrmse": 0.067024869120255, "全帯域_10ms_envelope_correlation": 0.9958599089579081, "全帯域_10ms_modulation_mae_db": 0.7331771382602236, "全帯域_40ms_log_rms_mae_db": 0.45899548236556276, "全帯域_40ms_rms_envelope_nrmse": 0.03392628319769184, "全帯域_40ms_envelope_correlation": 0.9988974363329745, "全帯域_40ms_modulation_mae_db": 0.9359268228072498, "300–1000 Hz_10ms_log_rms_mae_db": 1.604025603287772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10438373949677104, "300–1000 Hz_10ms_envelope_correlation": 0.9913627447319205, "300–1000 Hz_10ms_modulation_mae_db": 0.8582138663687761, "300–1000 Hz_40ms_log_rms_mae_db": 0.9377849858310677, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061698833644780356, "300–1000 Hz_40ms_envelope_correlation": 0.9967476055744787, "300–1000 Hz_40ms_modulation_mae_db": 1.1793987432401363, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5490324097597963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16364263668055207, "1000–3000 Hz_10ms_envelope_correlation": 0.981350697851245, "1000–3000 Hz_10ms_modulation_mae_db": 1.0116590587245053, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8126373147277494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10937606745233983, "1000–3000 Hz_40ms_envelope_correlation": 0.9924992758051273, "1000–3000 Hz_40ms_modulation_mae_db": 1.2090476354968265, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6328869984892662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18628329723183681, "3000–6000 Hz_10ms_envelope_correlation": 0.9770439468230627, "3000–6000 Hz_10ms_modulation_mae_db": 0.5783788819716166, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8835972305604785, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1134478305607272, "3000–6000 Hz_40ms_envelope_correlation": 0.990935472750538, "3000–6000 Hz_40ms_modulation_mae_db": 0.7591650340689208, "mr_stft_db": 5.49756799312956, "band_mae_db": 0.1883964005635345, "rms_gain_error_db": -0.07144187956057893, "output_peak": 0.577076788111629}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8788303906455505, "全帯域_10ms_rms_envelope_nrmse": 0.06531740228344221, "全帯域_10ms_envelope_correlation": 0.9960242096126186, "全帯域_10ms_modulation_mae_db": 0.3897074069996127, "全帯域_40ms_log_rms_mae_db": 0.4543916468126812, "全帯域_40ms_rms_envelope_nrmse": 0.036481833576436126, "全帯域_40ms_envelope_correlation": 0.9986542061350631, "全帯域_40ms_modulation_mae_db": 0.35572993549157655, "300–1000 Hz_10ms_log_rms_mae_db": 1.691317429057062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10648665612499397, "300–1000 Hz_10ms_envelope_correlation": 0.9909663769345161, "300–1000 Hz_10ms_modulation_mae_db": 0.5568678257080019, "300–1000 Hz_40ms_log_rms_mae_db": 0.9934341824189243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06358574235691804, "300–1000 Hz_40ms_envelope_correlation": 0.9965037838358279, "300–1000 Hz_40ms_modulation_mae_db": 0.6392492086508228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5388276845474087, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13584511905941105, "1000–3000 Hz_10ms_envelope_correlation": 0.986181617082262, "1000–3000 Hz_10ms_modulation_mae_db": 0.6260301525265405, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8370232788534839, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0797364694356198, "1000–3000 Hz_40ms_envelope_correlation": 0.9950057654571054, "1000–3000 Hz_40ms_modulation_mae_db": 0.6716192867940732, "3000–6000 Hz_10ms_log_rms_mae_db": 1.66369599298336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2106068310261882, "3000–6000 Hz_10ms_envelope_correlation": 0.972020584985728, "3000–6000 Hz_10ms_modulation_mae_db": 0.34401217666068457, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9502921934923346, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13237232566015758, "3000–6000 Hz_40ms_envelope_correlation": 0.9886703635126604, "3000–6000 Hz_40ms_modulation_mae_db": 0.37673897754928526, "mr_stft_db": 5.501060060067427, "band_mae_db": 0.15779966099660406, "rms_gain_error_db": -0.01827179294292993, "output_peak": 0.5919313498187777}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.854675390849428, "全帯域_10ms_rms_envelope_nrmse": 0.05860630831587098, "全帯域_10ms_envelope_correlation": 0.9968067973410443, "全帯域_10ms_modulation_mae_db": 0.33262150887037506, "全帯域_40ms_log_rms_mae_db": 0.4278648835906083, "全帯域_40ms_rms_envelope_nrmse": 0.030953217234314455, "全帯域_40ms_envelope_correlation": 0.9990399744013256, "全帯域_40ms_modulation_mae_db": 0.19388767841575752, "300–1000 Hz_10ms_log_rms_mae_db": 1.6160282957195966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10164975149468494, "300–1000 Hz_10ms_envelope_correlation": 0.9918310676037206, "300–1000 Hz_10ms_modulation_mae_db": 0.7611984895430677, "300–1000 Hz_40ms_log_rms_mae_db": 0.9170514438447329, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05862422249482339, "300–1000 Hz_40ms_envelope_correlation": 0.9970978034272201, "300–1000 Hz_40ms_modulation_mae_db": 0.5966492236730773, "1000–3000 Hz_10ms_log_rms_mae_db": 1.570592727644208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.148575884282762, "1000–3000 Hz_10ms_envelope_correlation": 0.9833235852928124, "1000–3000 Hz_10ms_modulation_mae_db": 0.2321898394492005, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7996271421300297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08201134103228866, "1000–3000 Hz_40ms_envelope_correlation": 0.9946234513189635, "1000–3000 Hz_40ms_modulation_mae_db": 0.3797855274381569, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6116499321238573, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15935329143140467, "3000–6000 Hz_10ms_envelope_correlation": 0.9829428967301271, "3000–6000 Hz_10ms_modulation_mae_db": 0.8872400894496599, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8403920902044664, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08463996353942882, "3000–6000 Hz_40ms_envelope_correlation": 0.9950937710136578, "3000–6000 Hz_40ms_modulation_mae_db": 1.0168742178296977, "mr_stft_db": 5.496490138178647, "band_mae_db": 0.14058062425370174, "rms_gain_error_db": -0.030063015174167575, "output_peak": 0.558470938707448}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8637243404288565, "全帯域_10ms_rms_envelope_nrmse": 0.06240717991706773, "全帯域_10ms_envelope_correlation": 0.9963953652157216, "全帯域_10ms_modulation_mae_db": 0.888258924769608, "全帯域_40ms_log_rms_mae_db": 0.45157044516706274, "全帯域_40ms_rms_envelope_nrmse": 0.03070579603445524, "全帯域_40ms_envelope_correlation": 0.9990817561523789, "全帯域_40ms_modulation_mae_db": 0.8099562971960307, "300–1000 Hz_10ms_log_rms_mae_db": 1.6179154587905364, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10310737138591577, "300–1000 Hz_10ms_envelope_correlation": 0.9915972760746302, "300–1000 Hz_10ms_modulation_mae_db": 1.1149939412198282, "300–1000 Hz_40ms_log_rms_mae_db": 0.900291241317826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053684631152265114, "300–1000 Hz_40ms_envelope_correlation": 0.9976242457545326, "300–1000 Hz_40ms_modulation_mae_db": 1.2653563438159756, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5504520355266578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13851763795162178, "1000–3000 Hz_10ms_envelope_correlation": 0.986168179428569, "1000–3000 Hz_10ms_modulation_mae_db": 1.0978048992889236, "1000–3000 Hz_40ms_log_rms_mae_db": 0.801603894645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0836261062321738, "1000–3000 Hz_40ms_envelope_correlation": 0.9952367401884422, "1000–3000 Hz_40ms_modulation_mae_db": 1.297590585327544, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6641387905754097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17371982236711522, "3000–6000 Hz_10ms_envelope_correlation": 0.9800219186701921, "3000–6000 Hz_10ms_modulation_mae_db": 0.3479303599252949, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9132145739795163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09686769359704497, "3000–6000 Hz_40ms_envelope_correlation": 0.9933719863154065, "3000–6000 Hz_40ms_modulation_mae_db": 0.7721345522603899, "mr_stft_db": 5.512146063510031, "band_mae_db": 0.12240442964481024, "rms_gain_error_db": -0.0568615883027213, "output_peak": 0.53056711593424}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8727905503644612, "全帯域_10ms_rms_envelope_nrmse": 0.06466143397113558, "全帯域_10ms_envelope_correlation": 0.9961004005677947, "全帯域_10ms_modulation_mae_db": 0.3664972553497259, "全帯域_40ms_log_rms_mae_db": 0.4448743915001172, "全帯域_40ms_rms_envelope_nrmse": 0.03284519361015411, "全帯域_40ms_envelope_correlation": 0.9989105966305769, "全帯域_40ms_modulation_mae_db": 0.2184584970464508, "300–1000 Hz_10ms_log_rms_mae_db": 1.624351893496987, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10994247910482165, "300–1000 Hz_10ms_envelope_correlation": 0.990397576251504, "300–1000 Hz_10ms_modulation_mae_db": 0.7193823279199337, "300–1000 Hz_40ms_log_rms_mae_db": 0.981574947627619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0615303546846197, "300–1000 Hz_40ms_envelope_correlation": 0.9967930872395809, "300–1000 Hz_40ms_modulation_mae_db": 0.6396371825640794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5947604074553767, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1541409914863765, "1000–3000 Hz_10ms_envelope_correlation": 0.9823992718751995, "1000–3000 Hz_10ms_modulation_mae_db": 0.7852136822069972, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8553045350370343, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08865735045623785, "1000–3000 Hz_40ms_envelope_correlation": 0.9937641703724354, "1000–3000 Hz_40ms_modulation_mae_db": 0.8538198954180166, "3000–6000 Hz_10ms_log_rms_mae_db": 1.638692724511836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17847746225447647, "3000–6000 Hz_10ms_envelope_correlation": 0.9784474883868278, "3000–6000 Hz_10ms_modulation_mae_db": 1.0217491228523574, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9002539316801529, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08811327833642936, "3000–6000 Hz_40ms_envelope_correlation": 0.9944378555228717, "3000–6000 Hz_40ms_modulation_mae_db": 1.3678160072072574, "mr_stft_db": 5.499614513364243, "band_mae_db": 0.0686056850889315, "rms_gain_error_db": -0.0074101917096471934, "output_peak": 0.6966594930927705}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8880520063899303, "全帯域_10ms_rms_envelope_nrmse": 0.06475503449288832, "全帯域_10ms_envelope_correlation": 0.9960961182233085, "全帯域_10ms_modulation_mae_db": 0.48963372388128595, "全帯域_40ms_log_rms_mae_db": 0.4304379279753719, "全帯域_40ms_rms_envelope_nrmse": 0.030427981747518058, "全帯域_40ms_envelope_correlation": 0.9990640917084487, "全帯域_40ms_modulation_mae_db": 0.30854999604221395, "300–1000 Hz_10ms_log_rms_mae_db": 1.6869163801123963, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10723603569903202, "300–1000 Hz_10ms_envelope_correlation": 0.9908549812998605, "300–1000 Hz_10ms_modulation_mae_db": 0.6320548529545545, "300–1000 Hz_40ms_log_rms_mae_db": 0.9667996317813051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05477202880785334, "300–1000 Hz_40ms_envelope_correlation": 0.9973996703862704, "300–1000 Hz_40ms_modulation_mae_db": 0.6596845793688476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6168472542407928, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1381319938047318, "1000–3000 Hz_10ms_envelope_correlation": 0.9859457948188115, "1000–3000 Hz_10ms_modulation_mae_db": 0.8368343480371303, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8536406976664379, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07054498212637206, "1000–3000 Hz_40ms_envelope_correlation": 0.9965619382493971, "1000–3000 Hz_40ms_modulation_mae_db": 1.0671452088919373, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6528297971218948, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17234433943186295, "3000–6000 Hz_10ms_envelope_correlation": 0.9801920421227712, "3000–6000 Hz_10ms_modulation_mae_db": 0.7409318696369812, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9021628072181928, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09805886404120294, "3000–6000 Hz_40ms_envelope_correlation": 0.993138728377451, "3000–6000 Hz_40ms_modulation_mae_db": 0.9449174035196051, "mr_stft_db": 5.494718789975818, "band_mae_db": 0.12437782896786136, "rms_gain_error_db": -0.006636507760371118, "output_peak": 0.6164738592163057}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8359069066268886, "全帯域_10ms_rms_envelope_nrmse": 0.06095421356503745, "全帯域_10ms_envelope_correlation": 0.9966280163593559, "全帯域_10ms_modulation_mae_db": 0.5891709047650936, "全帯域_40ms_log_rms_mae_db": 0.443182984796745, "全帯域_40ms_rms_envelope_nrmse": 0.03207201861791614, "全帯域_40ms_envelope_correlation": 0.9990712705079475, "全帯域_40ms_modulation_mae_db": 0.6092562503834944, "300–1000 Hz_10ms_log_rms_mae_db": 1.6018193864555126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10336805941436178, "300–1000 Hz_10ms_envelope_correlation": 0.9917418491287235, "300–1000 Hz_10ms_modulation_mae_db": 1.0162564444722573, "300–1000 Hz_40ms_log_rms_mae_db": 0.9031482407134356, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05738863759081924, "300–1000 Hz_40ms_envelope_correlation": 0.9974788279367097, "300–1000 Hz_40ms_modulation_mae_db": 1.0862613068168478, "1000–3000 Hz_10ms_log_rms_mae_db": 1.536050610403617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14396476305545827, "1000–3000 Hz_10ms_envelope_correlation": 0.9848124359979271, "1000–3000 Hz_10ms_modulation_mae_db": 0.8702725260650815, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8191926972503389, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07354757922772716, "1000–3000 Hz_40ms_envelope_correlation": 0.9963698422744068, "1000–3000 Hz_40ms_modulation_mae_db": 1.0208337516425348, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6231221111516771, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16872548531524648, "3000–6000 Hz_10ms_envelope_correlation": 0.9808438508616992, "3000–6000 Hz_10ms_modulation_mae_db": 0.3552017700198922, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8596116091468127, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08320301473404616, "3000–6000 Hz_40ms_envelope_correlation": 0.9950198761499895, "3000–6000 Hz_40ms_modulation_mae_db": 0.5525997733538799, "mr_stft_db": 5.511839056360965, "band_mae_db": 0.17904315021725212, "rms_gain_error_db": -0.095200904110989, "output_peak": 0.6830105033553103}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8923136041864607, "全帯域_10ms_rms_envelope_nrmse": 0.0630131898217918, "全帯域_10ms_envelope_correlation": 0.9963231325355006, "全帯域_10ms_modulation_mae_db": 0.7003108388316419, "全帯域_40ms_log_rms_mae_db": 0.448483597816697, "全帯域_40ms_rms_envelope_nrmse": 0.034239284715051714, "全帯域_40ms_envelope_correlation": 0.9988454965832835, "全帯域_40ms_modulation_mae_db": 0.6545752451288606, "300–1000 Hz_10ms_log_rms_mae_db": 1.655597702977453, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10250285019322761, "300–1000 Hz_10ms_envelope_correlation": 0.9918508594449725, "300–1000 Hz_10ms_modulation_mae_db": 0.7863166582113653, "300–1000 Hz_40ms_log_rms_mae_db": 0.9395118976275708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05961515428739188, "300–1000 Hz_40ms_envelope_correlation": 0.9972248791481592, "300–1000 Hz_40ms_modulation_mae_db": 0.7581636698943791, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5290317928111479, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14672095964286475, "1000–3000 Hz_10ms_envelope_correlation": 0.9840871290347525, "1000–3000 Hz_10ms_modulation_mae_db": 0.9165583319387545, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7941993602395249, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08111005472485758, "1000–3000 Hz_40ms_envelope_correlation": 0.9952957317776735, "1000–3000 Hz_40ms_modulation_mae_db": 0.9786417618260627, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6299652962221096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1915606477024352, "3000–6000 Hz_10ms_envelope_correlation": 0.9754013475651465, "3000–6000 Hz_10ms_modulation_mae_db": 1.758477076331484, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9022240197801924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1123330820059044, "3000–6000 Hz_40ms_envelope_correlation": 0.9915494987873109, "3000–6000 Hz_40ms_modulation_mae_db": 2.0132215189626907, "mr_stft_db": 5.502922793559432, "band_mae_db": 0.17702374445096614, "rms_gain_error_db": -0.054532827657798336, "output_peak": 0.6116164046146612}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85280, "output_samples": 85280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.614917392640109, "spectral_convergence": 0.18866201264984078, "spectral_convergence_status": "ok", "amplitude_floor": 1.2882929606468862e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3055042853363235, "spectral_convergence": 0.18948510559292417, "spectral_convergence_status": "ok", "amplitude_floor": 9.933866418820512e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.572282301412247, "spectral_convergence": 0.18320118801390795, "spectral_convergence_status": "ok", "amplitude_floor": 8.99033450142914e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001234751326260816, "output_power": 0.0012312568662147532, "delta_db": -0.012308359199150652, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018448232014630782, "output_power": 0.0018309991481990295, "delta_db": -0.032666096367327445, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006193820215877917, "output_power": 0.0005941820299193403, "delta_db": -0.18039082083508823, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017436605640499536, "output_power": 0.0001528004910017079, "delta_db": -0.5733719554558288, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010535962705358282, "output_power": 0.00010339164808569079, "delta_db": -0.08188766532606287, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.506722876313705e-05, "output_power": 1.5959112373689324e-05, "delta_db": 0.24975350619774905, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1883964005635345, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07144187956057893, "pitch": null, "output_rms_dbfs": -24.057633710083465, "output_peak": 0.577076788111629, "reference_rms_dbfs": -23.986191830522884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.49756799312956}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.0001748275799785616, "log_rms_mae_db": 0.8972112094967184, "rms_envelope_nrmse": 0.067024869120255, "envelope_correlation": 0.9958599089579081, "rms_gain_error_db": -0.0714423056287008, "reference_mean_power": 0.003997498793754849, "output_mean_power": 0.003932277069869112}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.9383161701154944e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012108392795166332, "output_power": 0.0011859609952052784, "reference_active": true, "delta_db": -0.09016095133599092}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002231683599850243, "output_power": 0.000191422365317987, "reference_active": true, "delta_db": -0.6663994360196113}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.764859371015312e-05, "output_power": 4.8524567153098985e-05, "reference_active": true, "delta_db": -1.4429710274250682}], "active_band_count": 3, "active_band_mae_db": 0.7331771382602236}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00017019095451879833, "log_rms_mae_db": 0.45899548236556276, "rms_envelope_nrmse": 0.03392628319769184, "envelope_correlation": 0.9988974363329745, "rms_gain_error_db": -0.0714433863692503, "reference_mean_power": 0.00402014109400192, "output_mean_power": 0.003954548962562924}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.6741447857886242e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011176954288226882, "output_power": 0.0010971146382286552, "reference_active": true, "delta_db": -0.08071465085743233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.194819962775497e-05, "output_power": 7.662120105102584e-05, "reference_active": true, "delta_db": -0.791942747021745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3053605179616555e-06, "output_power": 2.116930149167313e-06, "reference_active": true, "delta_db": -1.9351230705425724}], "active_band_count": 3, "active_band_mae_db": 0.9359268228072498}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.00014662911909855806, "log_rms_mae_db": 1.604025603287772, "rms_envelope_nrmse": 0.10438373949677104, "envelope_correlation": 0.9913627447319205, "rms_gain_error_db": -0.09603557148935477, "reference_mean_power": 0.0015423797799867744, "output_mean_power": 0.0015086474603544636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.0119951767590758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000652626948773928, "output_power": 0.0006429091457401328, "reference_active": true, "delta_db": -0.06515399157941211}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016341376588393177, "output_power": 0.00014086377592606144, "reference_active": true, "delta_db": -0.644893127612537}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.788084498799748e-05, "output_power": 3.116752798658174e-05, "reference_active": true, "delta_db": -1.864594479914379}], "active_band_count": 3, "active_band_mae_db": 0.8582138663687761}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00013732417437068112, "log_rms_mae_db": 0.9377849858310677, "rms_envelope_nrmse": 0.061698833644780356, "envelope_correlation": 0.9967476055744787, "rms_gain_error_db": -0.09603785090957119, "reference_mean_power": 0.0015511172002359399, "output_mean_power": 0.0015171929942305981}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 8.798155816797883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006408046711327789, "output_power": 0.0006246496298776994, "reference_active": true, "delta_db": -0.11089181888719638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.486060187757687e-05, "output_power": 6.280384392462566e-05, "reference_active": true, "delta_db": -0.7626708885978178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5748897665955177e-06, "output_power": 1.935527305175881e-06, "reference_active": true, "delta_db": -2.6646335222353947}], "active_band_count": 3, "active_band_mae_db": 1.1793987432401363}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 8.659636175169924e-05, "log_rms_mae_db": 1.5490324097597963, "rms_envelope_nrmse": 0.16364263668055207, "envelope_correlation": 0.981350697851245, "rms_gain_error_db": -0.47970982217316227, "reference_mean_power": 0.00020984803844209803, "output_mean_power": 0.00018790309423828125}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 2.0976407038326697e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011684508155422031, "output_power": 0.00010103711415774846, "reference_active": true, "delta_db": -0.6312950271799113}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.122844755814475e-05, "output_power": 2.8623646218815584e-05, "reference_active": true, "delta_db": -1.5847202642794662}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.871758765973707e-06, "output_power": 7.347059142141126e-06, "reference_active": true, "delta_db": -0.8189618847141384}], "active_band_count": 3, "active_band_mae_db": 1.0116590587245053}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 7.401841529112696e-05, "log_rms_mae_db": 0.8126373147277494, "rms_envelope_nrmse": 0.10937606745233983, "envelope_correlation": 0.9924992758051273, "rms_gain_error_db": -0.4797102104285489, "reference_mean_power": 0.000211036168686384, "output_mean_power": 0.00018896695838422988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.8857384691636124e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011739951182146869, "output_power": 9.962054953489415e-05, "reference_active": true, "delta_db": -0.7131735790972954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97467891176206e-05, "output_power": 1.3415725407633311e-05, "reference_active": true, "delta_db": -1.6788232765021984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.379135569148535e-07, "output_power": 4.800066197683095e-07, "reference_active": true, "delta_db": -1.2351460508909857}], "active_band_count": 3, "active_band_mae_db": 1.2090476354968265}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 4.122543101057287e-05, "log_rms_mae_db": 1.6328869984892662, "rms_envelope_nrmse": 0.18628329723183681, "envelope_correlation": 0.9770439468230627, "rms_gain_error_db": 0.09390503685826633, "reference_mean_power": 4.5035535336143536e-05, "output_mean_power": 4.601991723189585e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 3.018630059598249e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5795382838093414e-05, "output_power": 1.726064022026629e-05, "reference_active": true, "delta_db": 0.38526743729459}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.352602360298049e-06, "output_power": 4.564510535597957e-06, "reference_active": true, "delta_db": 0.2064522060640228}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848555477734869e-06, "output_power": 1.4206620803625382e-06, "reference_active": true, "delta_db": -1.143417002556237}], "active_band_count": 3, "active_band_mae_db": 0.5783788819716166}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 3.1688229259308314e-05, "log_rms_mae_db": 0.8835972305604785, "rms_envelope_nrmse": 0.1134478305607272, "envelope_correlation": 0.990935472750538, "rms_gain_error_db": 0.09390404093509583, "reference_mean_power": 4.529062388817587e-05, "output_mean_power": 4.6280570869056986e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 2.7414916263445066e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5534436484200316e-05, "output_power": 1.6938619856639378e-05, "reference_active": true, "delta_db": 0.375825177869774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0764100435071083e-06, "output_power": 2.1578358831804372e-06, "reference_active": true, "delta_db": 0.16705289934673004}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4701562809614912e-07, "output_power": 9.860565020996616e-08, "reference_active": true, "delta_db": -1.734617024990258}], "active_band_count": 3, "active_band_mae_db": 0.7591650340689208}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9397503711136932, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ab7f850058033b6055221a4d2e127525ff5fde2706c612dbd970518d1179174d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9842835889226974, "standard_deviation": 0.001824809350857463, "minimum": 0.981350697851245, "maximum": 0.986181617082262, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5606993652986256, "standard_deviation": 0.0310311155984128, "minimum": 1.5290317928111479, "maximum": 1.6168472542407928, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7970703547796417, "standard_deviation": 0.26891103571878944, "minimum": 0.2321898394492005, "maximum": 1.0978048992889236, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14619249824547229, "standard_deviation": 0.009331604892757198, "minimum": 0.13584511905941105, "maximum": 0.16364263668055207, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994919614430444, "standard_deviation": 0.0013272229843529178, "minimum": 0.9924992758051273, "maximum": 0.9965619382493971, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8216536150686999, "standard_deviation": 0.02426057097563214, "minimum": 0.7941993602395249, "maximum": 0.8553045350370343, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.934810456604394, "standard_deviation": 0.2970163209119085, "minimum": 0.3797855274381569, "maximum": 1.297590585327544, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0835762438359521, "standard_deviation": 0.011862990480071438, "minimum": 0.07054498212637206, "maximum": 0.10937606745233983, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9783642595181943, "standard_deviation": 0.0034635267231939767, "minimum": 0.972020584985728, "maximum": 0.9829428967301271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6396227053974264, "standard_deviation": 0.019083406024699507, "minimum": 1.6116499321238573, "maximum": 1.6641387905754097, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7542401683559964, "standard_deviation": 0.48128583709154943, "minimum": 0.34401217666068457, "maximum": 1.758477076331484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18013389709507074, "standard_deviation": 0.015872402935046943, "minimum": 0.15935329143140467, "maximum": 0.2106068310261882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927771940537358, "standard_deviation": 0.0022489255635240422, "minimum": 0.9886703635126604, "maximum": 0.9950937710136578, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8939685570077682, "standard_deviation": 0.03354308020338398, "minimum": 0.8403920902044664, "maximum": 0.9502921934923346, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9754334355939658, "standard_deviation": 0.5153282688336377, "minimum": 0.37673897754928526, "maximum": 2.0132215189626907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10112950655936769, "standard_deviation": 0.01708655772770315, "minimum": 0.08320301473404616, "maximum": 0.13237232566015758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913253414337311, "standard_deviation": 0.000533398357617058, "minimum": 0.990397576251504, "maximum": 0.9918508594449725, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6372465187371645, "standard_deviation": 0.036008748116937814, "minimum": 1.6018193864555126, "maximum": 1.691317429057062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8056605507997231, "standard_deviation": 0.1869420762221201, "minimum": 0.5568678257080019, "maximum": 1.1149939412198282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1048346178642261, "standard_deviation": 0.0028158800739183777, "minimum": 0.10164975149468494, "maximum": 0.10994247910482165, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971087379128474, "standard_deviation": 0.0003958690914324181, "minimum": 0.9965037838358279, "maximum": 0.9976242457545326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9424495713953102, "standard_deviation": 0.03529573754995255, "minimum": 0.900291241317826, "maximum": 0.9934341824189243, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8530500322530208, "standard_deviation": 0.2762706503355576, "minimum": 0.5966492236730773, "maximum": 1.2653563438159756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05886245062743388, "standard_deviation": 0.003460162484429381, "minimum": 0.053684631152265114, "maximum": 0.06358574235691804, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14477894052295773, "standard_deviation": 0.03968195023470704, "minimum": 0.0686056850889315, "maximum": 0.1883964005635345, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.502044926018265, "standard_deviation": 0.006656643847344371, "minimum": 5.494718789975818, "maximum": 5.512146063510031, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6082258066063928, "standard_deviation": 0.05761322280252439, "minimum": 0.53056711593424, "maximum": 0.6966594930927705, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04255233840240043, "standard_deviation": 0.032127445459786304, "minimum": -0.095200904110989, "maximum": -0.006636507760371118, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962792436016565, "standard_deviation": 0.00032110733990599436, "minimum": 0.9958599089579081, "maximum": 0.9968067973410443, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8729380498735367, "standard_deviation": 0.02076223985645677, "minimum": 0.8359069066268886, "maximum": 0.8972112094967184, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5611722127159459, "standard_deviation": 0.20048747516466897, "minimum": 0.33262150887037506, "maximum": 0.888258924769608, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06334245393593613, "standard_deviation": 0.0026809810165854104, "minimum": 0.05860630831587098, "maximum": 0.067024869120255, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989456035564999, "standard_deviation": 0.00014916646508774764, "minimum": 0.9986542061350631, "maximum": 0.9990817561523789, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44497517000310577, "standard_deviation": 0.011008836096964464, "minimum": 0.4278648835906083, "maximum": 0.45899548236556276, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5107925903139543, "standard_deviation": 0.28067292160679685, "minimum": 0.19388767841575752, "maximum": 0.9359268228072498, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.032706451091692214, "standard_deviation": 0.002096960671648911, "minimum": 0.030427981747518058, "maximum": 0.036481833576436126, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3570-5696-0010", "language": "en", "style": "normal", "speaker": "librispeech_3570", "source_utterance_id": "3570-5696-0010", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3570-5696-0010.wav", "source_file_sha256": "ac1b4fc670b304c84d517e25aa09870e11e69c812f83b1d48f4aefbe7a3d4ce1", "decoded_pcm_sha256": "0efec23fe6eeee5dfee017e6be4b06b64a152a981c62a71c533addc819701d30", "original_source_sha256": "426a986fec027f14754878775f6b605d58d61a0fd0d77bf122657dc1cef2ce9d", "corpus_entry_id": "librispeech_3570-5696-0010"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.8221397399902344, "predicted_mos": 3.8221397399902344, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "27cd80506539ced9831512ca48eba0acd53a1b411666ae1418d04a9c06b4fde3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.54960302699692, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3570-5696-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "27cd80506539ced9831512ca48eba0acd53a1b411666ae1418d04a9c06b4fde3", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "091793e4f5b935128df697e4a23a90f93fc6a9b25f5cc0dd77806c81e18fe35a"}, "rows": [{"case": "librispeech_3570-5696-0010", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.12797474861145, "predicted_mos": 3.12797474861145, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3e9ba24bb0b08804fbb836e703b734c130ae5a6583ad9639201306ce7fb0244b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.966598688002705, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3570-5696-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "3e9ba24bb0b08804fbb836e703b734c130ae5a6583ad9639201306ce7fb0244b", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7489628429663931, "全帯域_10ms_rms_envelope_nrmse": 0.05043446563250009, "全帯域_10ms_envelope_correlation": 0.9965236958059448, "全帯域_10ms_modulation_mae_db": 0.5422118846972271, "全帯域_40ms_log_rms_mae_db": 0.3620992133243598, "全帯域_40ms_rms_envelope_nrmse": 0.024420444980353038, "全帯域_40ms_envelope_correlation": 0.9991043692012126, "全帯域_40ms_modulation_mae_db": 0.6131082362445712, "300–1000 Hz_10ms_log_rms_mae_db": 1.4335953540226525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07985169650590192, "300–1000 Hz_10ms_envelope_correlation": 0.9935339608318633, "300–1000 Hz_10ms_modulation_mae_db": 0.907979454228018, "300–1000 Hz_40ms_log_rms_mae_db": 0.7129198902462478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03896419779065305, "300–1000 Hz_40ms_envelope_correlation": 0.9982896311348773, "300–1000 Hz_40ms_modulation_mae_db": 1.0717231517629966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7768805419677653, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1391819106063466, "1000–3000 Hz_10ms_envelope_correlation": 0.9870788607005622, "1000–3000 Hz_10ms_modulation_mae_db": 0.6649380345285154, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9883291755697202, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08431687663401354, "1000–3000 Hz_40ms_envelope_correlation": 0.995273171765221, "1000–3000 Hz_40ms_modulation_mae_db": 0.9055670200818288, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9049313042920286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2050934112663865, "3000–6000 Hz_10ms_envelope_correlation": 0.9700348509824557, "3000–6000 Hz_10ms_modulation_mae_db": 1.5497717904551849, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0519982450486907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1283377880392583, "3000–6000 Hz_40ms_envelope_correlation": 0.9888058023639907, "3000–6000 Hz_40ms_modulation_mae_db": 2.0250808082093283, "mr_stft_db": 5.448461127354335, "band_mae_db": 0.20275550339357218, "rms_gain_error_db": -0.0495522523014994, "output_peak": 0.7013798505551863}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7736821206348826, "全帯域_10ms_rms_envelope_nrmse": 0.05308937921062028, "全帯域_10ms_envelope_correlation": 0.9961719234681965, "全帯域_10ms_modulation_mae_db": 0.20414775010898945, "全帯域_40ms_log_rms_mae_db": 0.4129323673431361, "全帯域_40ms_rms_envelope_nrmse": 0.02928718487554599, "全帯域_40ms_envelope_correlation": 0.9987098169404355, "全帯域_40ms_modulation_mae_db": 0.20871512389312688, "300–1000 Hz_10ms_log_rms_mae_db": 1.5689582135163467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08072285428431387, "300–1000 Hz_10ms_envelope_correlation": 0.993385623941462, "300–1000 Hz_10ms_modulation_mae_db": 0.712773589463103, "300–1000 Hz_40ms_log_rms_mae_db": 0.8190305541600159, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044915611885155055, "300–1000 Hz_40ms_envelope_correlation": 0.9977057258998261, "300–1000 Hz_40ms_modulation_mae_db": 0.7876508181348258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.694791791657732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11803395379130453, "1000–3000 Hz_10ms_envelope_correlation": 0.9918466524121837, "1000–3000 Hz_10ms_modulation_mae_db": 0.6220365032974811, "1000–3000 Hz_40ms_log_rms_mae_db": 0.951009057129388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07201777152079905, "1000–3000 Hz_40ms_envelope_correlation": 0.9978841747537989, "1000–3000 Hz_40ms_modulation_mae_db": 0.7700682790319814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.852061348714087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16737339424155798, "3000–6000 Hz_10ms_envelope_correlation": 0.9790685850785169, "3000–6000 Hz_10ms_modulation_mae_db": 0.5875471601618915, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0142749146135064, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10183403736830898, "3000–6000 Hz_40ms_envelope_correlation": 0.9911862945247396, "3000–6000 Hz_40ms_modulation_mae_db": 0.4267275016651963, "mr_stft_db": 5.4462663505635005, "band_mae_db": 0.20061306552864885, "rms_gain_error_db": -0.057704656236802494, "output_peak": 0.6800214739702006}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7730556078487915, "全帯域_10ms_rms_envelope_nrmse": 0.05141805790477685, "全帯域_10ms_envelope_correlation": 0.9964527710630787, "全帯域_10ms_modulation_mae_db": 0.6119109359036335, "全帯域_40ms_log_rms_mae_db": 0.4065814404922272, "全帯域_40ms_rms_envelope_nrmse": 0.027690006639945274, "全帯域_40ms_envelope_correlation": 0.9989185668554228, "全帯域_40ms_modulation_mae_db": 0.7244598765487175, "300–1000 Hz_10ms_log_rms_mae_db": 1.4010600958530417, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08065480873221272, "300–1000 Hz_10ms_envelope_correlation": 0.9934225126531376, "300–1000 Hz_10ms_modulation_mae_db": 0.9131443097493271, "300–1000 Hz_40ms_log_rms_mae_db": 0.7237069836168736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04303204733894652, "300–1000 Hz_40ms_envelope_correlation": 0.9979287052842163, "300–1000 Hz_40ms_modulation_mae_db": 0.9461958088479484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7235931124884671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12324249229868021, "1000–3000 Hz_10ms_envelope_correlation": 0.9905011351626862, "1000–3000 Hz_10ms_modulation_mae_db": 1.055666767557784, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0064690309371638, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07828608591514298, "1000–3000 Hz_40ms_envelope_correlation": 0.9966486840988478, "1000–3000 Hz_40ms_modulation_mae_db": 1.501485725406928, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8997651486286453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22305410470101902, "3000–6000 Hz_10ms_envelope_correlation": 0.9626349290062832, "3000–6000 Hz_10ms_modulation_mae_db": 0.7783151314217235, "3000–6000 Hz_40ms_log_rms_mae_db": 1.079010585999856, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1541459523141871, "3000–6000 Hz_40ms_envelope_correlation": 0.9803756712532805, "3000–6000 Hz_40ms_modulation_mae_db": 1.0999370461181988, "mr_stft_db": 5.4722061031730815, "band_mae_db": 0.2258048359383448, "rms_gain_error_db": -0.0824454556658606, "output_peak": 0.6829822624234831}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7354343238093773, "全帯域_10ms_rms_envelope_nrmse": 0.04516295376701001, "全帯域_10ms_envelope_correlation": 0.9972320145882568, "全帯域_10ms_modulation_mae_db": 0.5662524797783307, "全帯域_40ms_log_rms_mae_db": 0.37878184022532363, "全帯域_40ms_rms_envelope_nrmse": 0.023174299623644156, "全帯域_40ms_envelope_correlation": 0.9992004898852039, "全帯域_40ms_modulation_mae_db": 0.6454681304976863, "300–1000 Hz_10ms_log_rms_mae_db": 1.458426436411189, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0742008000216234, "300–1000 Hz_10ms_envelope_correlation": 0.9945251971041602, "300–1000 Hz_10ms_modulation_mae_db": 0.7731448631321225, "300–1000 Hz_40ms_log_rms_mae_db": 0.8156835600984331, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04118552372231341, "300–1000 Hz_40ms_envelope_correlation": 0.9981785205144986, "300–1000 Hz_40ms_modulation_mae_db": 1.048692856088393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6577942423283503, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10837326706324882, "1000–3000 Hz_10ms_envelope_correlation": 0.9921044871431541, "1000–3000 Hz_10ms_modulation_mae_db": 0.4485367229600041, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8842302972448913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05662700502078218, "1000–3000 Hz_40ms_envelope_correlation": 0.997752346980637, "1000–3000 Hz_40ms_modulation_mae_db": 0.5736606990626104, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9352525303551802, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20288712811019755, "3000–6000 Hz_10ms_envelope_correlation": 0.9695014992587405, "3000–6000 Hz_10ms_modulation_mae_db": 0.8212760626025978, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1245776448560203, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11746409243225006, "3000–6000 Hz_40ms_envelope_correlation": 0.9894275717968796, "3000–6000 Hz_40ms_modulation_mae_db": 1.156951676284034, "mr_stft_db": 5.438977404990254, "band_mae_db": 0.16760970634968508, "rms_gain_error_db": -0.048979612650651254, "output_peak": 0.6782474631531616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7339393916568947, "全帯域_10ms_rms_envelope_nrmse": 0.048071126988006696, "全帯域_10ms_envelope_correlation": 0.9969120938420764, "全帯域_10ms_modulation_mae_db": 0.536037905518766, "全帯域_40ms_log_rms_mae_db": 0.3623003802404733, "全帯域_40ms_rms_envelope_nrmse": 0.025035163492699058, "全帯域_40ms_envelope_correlation": 0.999122945115367, "全帯域_40ms_modulation_mae_db": 0.5254946000939805, "300–1000 Hz_10ms_log_rms_mae_db": 1.3905781161582875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07533627320390553, "300–1000 Hz_10ms_envelope_correlation": 0.9942234575219122, "300–1000 Hz_10ms_modulation_mae_db": 0.7239957339966288, "300–1000 Hz_40ms_log_rms_mae_db": 0.7155536350741122, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040479715610460376, "300–1000 Hz_40ms_envelope_correlation": 0.9980982312745837, "300–1000 Hz_40ms_modulation_mae_db": 0.8012785164055405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7022365096054273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11725983969789189, "1000–3000 Hz_10ms_envelope_correlation": 0.9912901135414962, "1000–3000 Hz_10ms_modulation_mae_db": 0.9903609762869602, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9659268488805521, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0719101279204699, "1000–3000 Hz_40ms_envelope_correlation": 0.9970829164172665, "1000–3000 Hz_40ms_modulation_mae_db": 1.1922865137129484, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8986824006449343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1820383747534479, "3000–6000 Hz_10ms_envelope_correlation": 0.975193193128692, "3000–6000 Hz_10ms_modulation_mae_db": 0.4760309088724128, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0780129852707334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1057399167653372, "3000–6000 Hz_40ms_envelope_correlation": 0.9904947680419668, "3000–6000 Hz_40ms_modulation_mae_db": 0.421867316031527, "mr_stft_db": 5.454681009427404, "band_mae_db": 0.15330400404532268, "rms_gain_error_db": -0.07422396932492273, "output_peak": 0.6568432936287371}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7465727545771507, "全帯域_10ms_rms_envelope_nrmse": 0.049515944229083085, "全帯域_10ms_envelope_correlation": 0.9966926530793597, "全帯域_10ms_modulation_mae_db": 0.4159968736395987, "全帯域_40ms_log_rms_mae_db": 0.3967668415471195, "全帯域_40ms_rms_envelope_nrmse": 0.026343680225927806, "全帯域_40ms_envelope_correlation": 0.998983749767866, "全帯域_40ms_modulation_mae_db": 0.47545135596959837, "300–1000 Hz_10ms_log_rms_mae_db": 1.4707962938019061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0766953230179935, "300–1000 Hz_10ms_envelope_correlation": 0.9941530196209352, "300–1000 Hz_10ms_modulation_mae_db": 0.8443662234895813, "300–1000 Hz_40ms_log_rms_mae_db": 0.8507106445102051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04507212026937053, "300–1000 Hz_40ms_envelope_correlation": 0.9978183641419357, "300–1000 Hz_40ms_modulation_mae_db": 0.8411776075725409, "1000–3000 Hz_10ms_log_rms_mae_db": 1.725125013524786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1266190326823479, "1000–3000 Hz_10ms_envelope_correlation": 0.9898296904265931, "1000–3000 Hz_10ms_modulation_mae_db": 1.09629282262287, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0061276807580841, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0777473083179258, "1000–3000 Hz_40ms_envelope_correlation": 0.9966138663575864, "1000–3000 Hz_40ms_modulation_mae_db": 0.5628421187746276, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8875078981833802, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22404004254206694, "3000–6000 Hz_10ms_envelope_correlation": 0.9621369485996449, "3000–6000 Hz_10ms_modulation_mae_db": 0.9812042200848384, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0418899587728494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12875735150814566, "3000–6000 Hz_40ms_envelope_correlation": 0.9863988839794141, "3000–6000 Hz_40ms_modulation_mae_db": 1.3183032158680976, "mr_stft_db": 5.454005736103763, "band_mae_db": 0.21646879047640674, "rms_gain_error_db": -0.061343464048121576, "output_peak": 0.700178520953977}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7725957189230184, "全帯域_10ms_rms_envelope_nrmse": 0.05200648846733477, "全帯域_10ms_envelope_correlation": 0.9963055220100793, "全帯域_10ms_modulation_mae_db": 0.42696708507391695, "全帯域_40ms_log_rms_mae_db": 0.370347900052262, "全帯域_40ms_rms_envelope_nrmse": 0.024624993056970666, "全帯域_40ms_envelope_correlation": 0.9990881772444014, "全帯域_40ms_modulation_mae_db": 0.4492777526563753, "300–1000 Hz_10ms_log_rms_mae_db": 1.4546984282214372, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07934995293052217, "300–1000 Hz_10ms_envelope_correlation": 0.9936233275608841, "300–1000 Hz_10ms_modulation_mae_db": 0.6397589374404927, "300–1000 Hz_40ms_log_rms_mae_db": 0.7792110907848124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04300950656992878, "300–1000 Hz_40ms_envelope_correlation": 0.9978776981479212, "300–1000 Hz_40ms_modulation_mae_db": 0.5458632087024182, "1000–3000 Hz_10ms_log_rms_mae_db": 1.695617044620508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11766257011500132, "1000–3000 Hz_10ms_envelope_correlation": 0.991450158357381, "1000–3000 Hz_10ms_modulation_mae_db": 0.15968900356780966, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9654110995725215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0680147242913191, "1000–3000 Hz_40ms_envelope_correlation": 0.9977485109270173, "1000–3000 Hz_40ms_modulation_mae_db": 0.29551888160490436, "3000–6000 Hz_10ms_log_rms_mae_db": 1.85297653907583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1822610368899811, "3000–6000 Hz_10ms_envelope_correlation": 0.9765191535331866, "3000–6000 Hz_10ms_modulation_mae_db": 0.9328225696431797, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0363756074495933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11044356556364146, "3000–6000 Hz_40ms_envelope_correlation": 0.9920156073008455, "3000–6000 Hz_40ms_modulation_mae_db": 1.2027972625882275, "mr_stft_db": 5.440543219312761, "band_mae_db": 0.19677329489381779, "rms_gain_error_db": -0.05082260837701982, "output_peak": 0.6926894889402117}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7249747197216848, "全帯域_10ms_rms_envelope_nrmse": 0.04405251118543665, "全帯域_10ms_envelope_correlation": 0.9974833270715231, "全帯域_10ms_modulation_mae_db": 0.5601856422794426, "全帯域_40ms_log_rms_mae_db": 0.35311621152875894, "全帯域_40ms_rms_envelope_nrmse": 0.0238948414973771, "全帯域_40ms_envelope_correlation": 0.9992538937397778, "全帯域_40ms_modulation_mae_db": 0.681838027692352, "300–1000 Hz_10ms_log_rms_mae_db": 1.4374559457086051, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07541710894705754, "300–1000 Hz_10ms_envelope_correlation": 0.994354258947418, "300–1000 Hz_10ms_modulation_mae_db": 0.9543225132097669, "300–1000 Hz_40ms_log_rms_mae_db": 0.7285268762827274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04030179882846455, "300–1000 Hz_40ms_envelope_correlation": 0.9982741359675572, "300–1000 Hz_40ms_modulation_mae_db": 1.063380705136937, "1000–3000 Hz_10ms_log_rms_mae_db": 1.726372252932415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11519734891574801, "1000–3000 Hz_10ms_envelope_correlation": 0.9918664422703615, "1000–3000 Hz_10ms_modulation_mae_db": 0.5448537100940963, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9312761922155472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07193013095731919, "1000–3000 Hz_40ms_envelope_correlation": 0.9973709553855089, "1000–3000 Hz_40ms_modulation_mae_db": 0.7845360599513306, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8929195952002462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21233365718018543, "3000–6000 Hz_10ms_envelope_correlation": 0.9669757069173343, "3000–6000 Hz_10ms_modulation_mae_db": 1.0485462228530391, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0951256543077628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1335901937927053, "3000–6000 Hz_40ms_envelope_correlation": 0.9866799851439576, "3000–6000 Hz_40ms_modulation_mae_db": 1.4264688221883308, "mr_stft_db": 5.443154538611274, "band_mae_db": 0.21981956892603785, "rms_gain_error_db": -0.08466844246486269, "output_peak": 0.6791382406194592}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.564177400697128, "spectral_convergence": 0.14532610794128514, "spectral_convergence_status": "ok", "amplitude_floor": 1.1745223847485646e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.297687645669818, "spectral_convergence": 0.14182901011554752, "spectral_convergence_status": "ok", "amplitude_floor": 9.481908348017771e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.483518335696058, "spectral_convergence": 0.13690804573194063, "spectral_convergence_status": "ok", "amplitude_floor": 8.395402116778918e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004068927413639468, "output_power": 0.004047027719277139, "delta_db": -0.02343763449789541, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018176508169902323, "output_power": 0.0018069907722258016, "delta_db": -0.02554521241320378, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007805177117872849, "output_power": 0.0007531141812681578, "delta_db": -0.15521937158767304, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000310938544728363, "output_power": 0.0002999198213806739, "delta_db": -0.15669392856120523, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.08647023559392e-05, "output_power": 7.750507191821582e-05, "delta_db": -0.18428868760384814, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.353447039170603e-05, "output_power": 3.7299108092389934e-05, "delta_db": -0.6713481856976076, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20275550339357218, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0495522523014994, "pitch": null, "output_rms_dbfs": -21.53548039333164, "output_peak": 0.7013798505551863, "reference_rms_dbfs": -21.485928141030136, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.448461127354335}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019068326661124743, "log_rms_mae_db": 0.7489628429663931, "rms_envelope_nrmse": 0.05043446563250009, "envelope_correlation": 0.9965236958059448, "rms_gain_error_db": -0.048512125534140985, "reference_mean_power": 0.007104543911662494, "output_mean_power": 0.007025625407193316}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.7734542191089184e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001358301262786015, "output_power": 0.0013241840118257727, "reference_active": true, "delta_db": -0.11047764534408351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002660490997406596, "output_power": 0.0002539626488537831, "reference_active": true, "delta_db": -0.20191945536324088}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.883599774152297e-05, "output_power": 5.825064157314336e-05, "reference_active": true, "delta_db": -1.3142385533843572}], "active_band_count": 3, "active_band_mae_db": 0.5422118846972271}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001870958079240233, "log_rms_mae_db": 0.3620992133243598, "rms_envelope_nrmse": 0.024420444980353038, "envelope_correlation": 0.9991043692012126, "rms_gain_error_db": -0.048605745497706104, "reference_mean_power": 0.007117638239873263, "output_mean_power": 0.0070384225538828475}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4206220628648622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012103399895685932, "output_power": 0.0011903379178657763, "reference_active": true, "delta_db": -0.07237114449892443}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012555497766250794, "output_power": 0.00012089515040873913, "reference_active": true, "delta_db": -0.1642505520764995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0918852450489625e-06, "output_power": 4.211932596590843e-06, "reference_active": true, "delta_db": -1.6027030121582897}], "active_band_count": 3, "active_band_mae_db": 0.6131082362445712}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013521834879092463, "log_rms_mae_db": 1.4335953540226525, "rms_envelope_nrmse": 0.07985169650590192, "envelope_correlation": 0.9935339608318633, "rms_gain_error_db": -0.06529213000182651, "reference_mean_power": 0.0018937311393695897, "output_mean_power": 0.001865473600777834}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.102346197011796e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006150154421416123, "output_power": 0.0005995273709341271, "reference_active": true, "delta_db": -0.11077005165392478}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012427322589552618, "output_power": 0.00011070809281830098, "reference_active": true, "delta_db": -0.5019820274924416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6589172324725154e-05, "output_power": 1.6352576531557848e-05, "reference_active": true, "delta_db": -2.1111862835376876}], "active_band_count": 3, "active_band_mae_db": 0.907979454228018}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012428224153258587, "log_rms_mae_db": 0.7129198902462478, "rms_envelope_nrmse": 0.03896419779065305, "envelope_correlation": 0.9982896311348773, "rms_gain_error_db": -0.06677032991162866, "reference_mean_power": 0.0018973976433641784, "output_mean_power": 0.0018684493257849928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.796182335233462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005643605996811222, "output_power": 0.0005492347966784341, "reference_active": true, "delta_db": -0.11798642178347615}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.691592416494353e-05, "output_power": 4.9951393157537196e-05, "reference_active": true, "delta_db": -0.5668618672515914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2464097368217303e-06, "output_power": 1.2544601108418599e-06, "reference_active": true, "delta_db": -2.530321166253922}], "active_band_count": 3, "active_band_mae_db": 1.0717231517629966}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012748188769016831, "log_rms_mae_db": 1.7768805419677653, "rms_envelope_nrmse": 0.1391819106063466, "envelope_correlation": 0.9870788607005622, "rms_gain_error_db": -0.2072627802422381, "reference_mean_power": 0.00032033050063867626, "output_mean_power": 0.0003054020967942185}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0949985705721996e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.663152856065707e-05, "output_power": 6.392316960189949e-05, "reference_active": true, "delta_db": -0.1802147517714456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2062707324852824e-05, "output_power": 1.1243341013576183e-05, "reference_active": true, "delta_db": -0.3054940761395893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2084052775211794e-06, "output_power": 2.2666198848497445e-06, "reference_active": true, "delta_db": -1.5091052756745114}], "active_band_count": 3, "active_band_mae_db": 0.6649380345285154}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001243805164613915, "log_rms_mae_db": 0.9883291755697202, "rms_envelope_nrmse": 0.08431687663401354, "envelope_correlation": 0.995273171765221, "rms_gain_error_db": -0.20851604264277268, "reference_mean_power": 0.0003214084172642619, "output_mean_power": 0.0003063413641204899}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.993543951674123e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.454820101360055e-05, "output_power": 6.199497242271579e-05, "reference_active": true, "delta_db": -0.17527671645463144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.405216922862835e-06, "output_power": 5.311089678253679e-06, "reference_active": true, "delta_db": -0.07629493216620678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4345993241652877e-07, "output_power": 1.9469873909850863e-07, "reference_active": true, "delta_db": -2.4651294116246483}], "active_band_count": 3, "active_band_mae_db": 0.9055670200818288}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.885927766129993e-05, "log_rms_mae_db": 1.9049313042920286, "rms_envelope_nrmse": 0.2050934112663865, "envelope_correlation": 0.9700348509824557, "rms_gain_error_db": -0.5390913643024735, "reference_mean_power": 5.543232992418922e-05, "output_mean_power": 4.896141907990771e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.209399643069482e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1581941196726732e-05, "output_power": 1.980981126818765e-05, "reference_active": true, "delta_db": -0.3721016694983209}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7792321883200776e-05, "output_power": 1.3322407863406384e-05, "reference_active": true, "delta_db": -1.2564990146964554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1497984756130906e-05, "output_power": 5.735222349960396e-06, "reference_active": true, "delta_db": -3.0207146871707784}], "active_band_count": 3, "active_band_mae_db": 1.5497717904551849}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.78543565194846e-05, "log_rms_mae_db": 1.0519982450486907, "rms_envelope_nrmse": 0.1283377880392583, "envelope_correlation": 0.9888058023639907, "rms_gain_error_db": -0.539526095892294, "reference_mean_power": 5.5701179342093606e-05, "output_mean_power": 4.919395968092471e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.455779259683724e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.849693387286164e-05, "output_power": 2.3645599813487054e-05, "reference_active": true, "delta_db": -0.8104779945674258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0110098860775756e-05, "output_power": 6.511409806838197e-06, "reference_active": true, "delta_db": -1.9108037304345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1498909214748545e-06, "output_power": 5.312030686770688e-07, "reference_active": true, "delta_db": -3.3539606996260596}], "active_band_count": 3, "active_band_mae_db": 2.0250808082093283}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9244049692605704, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e7dbba627d5776dd336c47f171b4969f0e8708d6eeffd1ca1452ce3e5d1d4dda", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9907459425018023, "standard_deviation": 0.001666977808918444, "minimum": 0.9870788607005622, "maximum": 0.9921044871431541, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7128013136406814, "standard_deviation": 0.03446673211732947, "minimum": 1.6577942423283503, "maximum": 1.7768805419677653, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6977968176144401, "standard_deviation": 0.3282780370649121, "minimum": 0.15968900356780966, "maximum": 1.09629282262287, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12069630189632116, "standard_deviation": 0.009207730639314215, "minimum": 0.10837326706324882, "maximum": 0.1391819106063466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970468283357354, "standard_deviation": 0.000869273692975714, "minimum": 0.995273171765221, "maximum": 0.9978841747537989, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9623474227884836, "standard_deviation": 0.04098247427866605, "minimum": 0.8842302972448913, "maximum": 1.0064690309371638, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.823245662203395, "standard_deviation": 0.3809578193720107, "minimum": 0.29551888160490436, "maximum": 1.501485725406928, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07260625382222147, "standard_deviation": 0.008224842170276542, "minimum": 0.05662700502078218, "maximum": 0.08431687663401354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9702581083131068, "standard_deviation": 0.0062865253640974365, "minimum": 0.9621369485996449, "maximum": 0.9790685850785169, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8905120956367916, "standard_deviation": 0.027423118902467115, "minimum": 1.852061348714087, "maximum": 1.9352525303551802, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8969392582618584, "standard_deviation": 0.32738075129462296, "minimum": 0.4760309088724128, "maximum": 1.5497717904551849, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19988514371060528, "standard_deviation": 0.02069331671536491, "minimum": 0.16737339424155798, "maximum": 0.22404004254206694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9881730730506343, "standard_deviation": 0.0037299078384635466, "minimum": 0.9803756712532805, "maximum": 0.9920156073008455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0651581995398764, "standard_deviation": 0.03569312594224472, "minimum": 1.0142749146135064, "maximum": 1.1245776448560203, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1347667061191176, "standard_deviation": 0.5247194441484607, "minimum": 0.421867316031527, "maximum": 2.0250808082093283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12253911222297925, "standard_deviation": 0.017222829753672383, "minimum": 0.10183403736830898, "maximum": 0.1541459523141871, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9939026697727216, "standard_deviation": 0.0004580889640850503, "minimum": 0.993385623941462, "maximum": 0.9945251971041602, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.451946110461683, "standard_deviation": 0.0547707741340817, "minimum": 1.3905781161582875, "maximum": 1.5689582135163467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8086857030886301, "standard_deviation": 0.113026518621101, "minimum": 0.6397589374404927, "maximum": 0.9543225132097669, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07777860220544133, "standard_deviation": 0.002651787294691931, "minimum": 0.0742008000216234, "maximum": 0.08072285428431387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980213765456769, "standard_deviation": 0.00021926159759191966, "minimum": 0.9977057258998261, "maximum": 0.9982896311348773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7681679043466785, "standard_deviation": 0.05496595255142788, "minimum": 0.7129198902462478, "maximum": 0.8507106445102051, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8882453340814501, "standard_deviation": 0.18148653916741045, "minimum": 0.5458632087024182, "maximum": 1.0717231517629966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04212006525191153, "standard_deviation": 0.0022355003922557826, "minimum": 0.03896419779065305, "maximum": 0.04507212026937053, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1978935961939795, "standard_deviation": 0.02547395832473413, "minimum": 0.15330400404532268, "maximum": 0.2258048359383448, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.449786936192047, "standard_deviation": 0.010722489916793091, "minimum": 5.438977404990254, "maximum": 5.4722061031730815, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6839350742805521, "standard_deviation": 0.014394228929661482, "minimum": 0.6568432936287371, "maximum": 0.7013798505551863, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06371755763371757, "standard_deviation": 0.014764786021908465, "minimum": -0.08466844246486269, "maximum": -0.048979612650651254, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967217501160643, "standard_deviation": 0.0004574115186669301, "minimum": 0.9961719234681965, "maximum": 0.9974833270715231, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7511521850172742, "standard_deviation": 0.01964520510138331, "minimum": 0.7249747197216848, "maximum": 0.7736821206348826, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.48296381962498813, "standard_deviation": 0.13159101584856514, "minimum": 0.20414775010898945, "maximum": 0.6119109359036335, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04921886592309606, "standard_deviation": 0.003242861290810771, "minimum": 0.04405251118543665, "maximum": 0.05308937921062028, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990477510937109, "standard_deviation": 0.00017374126193268018, "minimum": 0.9987098169404355, "maximum": 0.9992538937397778, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3803657743442076, "standard_deviation": 0.0224317064455237, "minimum": 0.35311621152875894, "maximum": 0.4129323673431361, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.540476637949551, "standard_deviation": 0.1660323225056904, "minimum": 0.20871512389312688, "maximum": 0.7244598765487175, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.025558826799057886, "standard_deviation": 0.0020686633081820056, "minimum": 0.023174299623644156, "maximum": 0.02928718487554599, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5696-0010", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.911968231201172, "predicted_mos": 2.911968231201172, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6ae71aa339b51576b14ca0d6005de80044be39cf37c7c74e3d391d897ce392e6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.2763314000330865, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7483733366562493, "全帯域_10ms_rms_envelope_nrmse": 0.04954154956877914, "全帯域_10ms_envelope_correlation": 0.9966392944853716, "全帯域_10ms_modulation_mae_db": 0.5524166374815959, "全帯域_40ms_log_rms_mae_db": 0.36112749449659814, "全帯域_40ms_rms_envelope_nrmse": 0.02363713972919664, "全帯域_40ms_envelope_correlation": 0.9991517824926307, "全帯域_40ms_modulation_mae_db": 0.6499294261930302, "300–1000 Hz_10ms_log_rms_mae_db": 1.494413084911461, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08206793614298273, "300–1000 Hz_10ms_envelope_correlation": 0.9931642638680436, "300–1000 Hz_10ms_modulation_mae_db": 1.0518079025338762, "300–1000 Hz_40ms_log_rms_mae_db": 0.7490056397219887, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03965345384129995, "300–1000 Hz_40ms_envelope_correlation": 0.9982092784180189, "300–1000 Hz_40ms_modulation_mae_db": 1.2218904592923454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7416467766664068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.130602088537094, "1000–3000 Hz_10ms_envelope_correlation": 0.988749259953117, "1000–3000 Hz_10ms_modulation_mae_db": 0.857377970739845, "1000–3000 Hz_40ms_log_rms_mae_db": 0.939330769086352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07785204054010042, "1000–3000 Hz_40ms_envelope_correlation": 0.9961631664599249, "1000–3000 Hz_40ms_modulation_mae_db": 1.0972977823886914, "3000–6000 Hz_10ms_log_rms_mae_db": 1.849460902816375, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21749817665096954, "3000–6000 Hz_10ms_envelope_correlation": 0.9646435833387569, "3000–6000 Hz_10ms_modulation_mae_db": 1.2449219655270092, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0582539206532562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13051886150791844, "3000–6000 Hz_40ms_envelope_correlation": 0.9866511084526003, "3000–6000 Hz_40ms_modulation_mae_db": 1.7410313797994006, "mr_stft_db": 5.380332906101839, "band_mae_db": 0.18238497152045094, "rms_gain_error_db": -0.04191267933979285, "output_peak": 0.688843687291887}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.770157166542763, "全帯域_10ms_rms_envelope_nrmse": 0.05137214911167165, "全帯域_10ms_envelope_correlation": 0.9964218937282023, "全帯域_10ms_modulation_mae_db": 0.1979620397011246, "全帯域_40ms_log_rms_mae_db": 0.4096185563878564, "全帯域_40ms_rms_envelope_nrmse": 0.028502206709293865, "全帯域_40ms_envelope_correlation": 0.9987860628392695, "全帯域_40ms_modulation_mae_db": 0.19154215055297516, "300–1000 Hz_10ms_log_rms_mae_db": 1.600725954004648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08561589888353989, "300–1000 Hz_10ms_envelope_correlation": 0.9925349780515826, "300–1000 Hz_10ms_modulation_mae_db": 0.7534342000014348, "300–1000 Hz_40ms_log_rms_mae_db": 0.829487719819606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04742786420359861, "300–1000 Hz_40ms_envelope_correlation": 0.9974282161834414, "300–1000 Hz_40ms_modulation_mae_db": 0.8531273317885448, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7211382981886565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11992004714346866, "1000–3000 Hz_10ms_envelope_correlation": 0.9915524087994918, "1000–3000 Hz_10ms_modulation_mae_db": 0.752613092319257, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0254218676931823, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07270041764402377, "1000–3000 Hz_40ms_envelope_correlation": 0.9978690533575673, "1000–3000 Hz_40ms_modulation_mae_db": 0.5374739333704953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.801448973165022, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18222833805245145, "3000–6000 Hz_10ms_envelope_correlation": 0.9755141186783014, "3000–6000 Hz_10ms_modulation_mae_db": 0.8463301509178697, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0003252333445183, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11397440015402481, "3000–6000 Hz_40ms_envelope_correlation": 0.989355875237295, "3000–6000 Hz_40ms_modulation_mae_db": 0.6315808827771359, "mr_stft_db": 5.378293934322539, "band_mae_db": 0.25140056858505183, "rms_gain_error_db": -0.05875749688496356, "output_peak": 0.6821523488121497}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7672363564055026, "全帯域_10ms_rms_envelope_nrmse": 0.0521559276973548, "全帯域_10ms_envelope_correlation": 0.9963352456742314, "全帯域_10ms_modulation_mae_db": 0.5283549014819146, "全帯域_40ms_log_rms_mae_db": 0.40860007032308193, "全帯域_40ms_rms_envelope_nrmse": 0.027510648670106334, "全帯域_40ms_envelope_correlation": 0.9989237167263721, "全帯域_40ms_modulation_mae_db": 0.6360146789245638, "300–1000 Hz_10ms_log_rms_mae_db": 1.4667088511211028, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0840016510690768, "300–1000 Hz_10ms_envelope_correlation": 0.9928308682841331, "300–1000 Hz_10ms_modulation_mae_db": 0.8748006850783928, "300–1000 Hz_40ms_log_rms_mae_db": 0.7872495572434487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04484453531154981, "300–1000 Hz_40ms_envelope_correlation": 0.9977092187455241, "300–1000 Hz_40ms_modulation_mae_db": 0.8781489184938932, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7120949717287572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12070806943131485, "1000–3000 Hz_10ms_envelope_correlation": 0.9910618268849196, "1000–3000 Hz_10ms_modulation_mae_db": 1.2747366089664076, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0004500238357752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07760088306707807, "1000–3000 Hz_40ms_envelope_correlation": 0.9969260204558456, "1000–3000 Hz_40ms_modulation_mae_db": 1.5501671498774698, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8557416387094077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24075880532365132, "3000–6000 Hz_10ms_envelope_correlation": 0.9565436893819238, "3000–6000 Hz_10ms_modulation_mae_db": 0.8577614836144339, "3000–6000 Hz_40ms_log_rms_mae_db": 1.099207065659886, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16013287031019977, "3000–6000 Hz_40ms_envelope_correlation": 0.9783909803937333, "3000–6000 Hz_40ms_modulation_mae_db": 0.762501377980836, "mr_stft_db": 5.391731854287145, "band_mae_db": 0.26222276479295165, "rms_gain_error_db": -0.07917551056269337, "output_peak": 0.6812047823914172}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.734691367970375, "全帯域_10ms_rms_envelope_nrmse": 0.04620755810517348, "全帯域_10ms_envelope_correlation": 0.9971047362863664, "全帯域_10ms_modulation_mae_db": 0.5999384955807944, "全帯域_40ms_log_rms_mae_db": 0.3702734126553686, "全帯域_40ms_rms_envelope_nrmse": 0.023461175316999646, "全帯域_40ms_envelope_correlation": 0.9991851938805686, "全帯域_40ms_modulation_mae_db": 0.6707085379054262, "300–1000 Hz_10ms_log_rms_mae_db": 1.5332276491002141, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07696524709967896, "300–1000 Hz_10ms_envelope_correlation": 0.9940924093248303, "300–1000 Hz_10ms_modulation_mae_db": 0.8474115297026253, "300–1000 Hz_40ms_log_rms_mae_db": 0.8358467166399793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041591357567114455, "300–1000 Hz_40ms_envelope_correlation": 0.9981351709646962, "300–1000 Hz_40ms_modulation_mae_db": 1.098346354040535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6446603348678057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10122670442585895, "1000–3000 Hz_10ms_envelope_correlation": 0.9931671277814509, "1000–3000 Hz_10ms_modulation_mae_db": 0.5386653910586995, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8820760660597323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05305658848580217, "1000–3000 Hz_40ms_envelope_correlation": 0.9981544916418729, "1000–3000 Hz_40ms_modulation_mae_db": 0.4659900852494505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8561071221274839, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2090931216902264, "3000–6000 Hz_10ms_envelope_correlation": 0.9671144405345531, "3000–6000 Hz_10ms_modulation_mae_db": 0.8409457105360977, "3000–6000 Hz_40ms_log_rms_mae_db": 1.065998112656652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11878682965174221, "3000–6000 Hz_40ms_envelope_correlation": 0.9884933391422381, "3000–6000 Hz_40ms_modulation_mae_db": 1.01594437734986, "mr_stft_db": 5.3711832872318945, "band_mae_db": 0.1636963754165128, "rms_gain_error_db": -0.051242205228030906, "output_peak": 0.6784025770822887}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7532028289136481, "全帯域_10ms_rms_envelope_nrmse": 0.04972402200591653, "全帯域_10ms_envelope_correlation": 0.9967042643630871, "全帯域_10ms_modulation_mae_db": 0.5464279815716707, "全帯域_40ms_log_rms_mae_db": 0.3756084608884218, "全帯域_40ms_rms_envelope_nrmse": 0.025955068690458015, "全帯域_40ms_envelope_correlation": 0.9990701072881802, "全帯域_40ms_modulation_mae_db": 0.4985068070171091, "300–1000 Hz_10ms_log_rms_mae_db": 1.4716443058974262, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07829939725486294, "300–1000 Hz_10ms_envelope_correlation": 0.9937622030645371, "300–1000 Hz_10ms_modulation_mae_db": 0.8915480276589588, "300–1000 Hz_40ms_log_rms_mae_db": 0.7759914894631034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04269135479462222, "300–1000 Hz_40ms_envelope_correlation": 0.997893173766323, "300–1000 Hz_40ms_modulation_mae_db": 0.991577600347329, "1000–3000 Hz_10ms_log_rms_mae_db": 1.684760337996495, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11293980955533817, "1000–3000 Hz_10ms_envelope_correlation": 0.9921155381518032, "1000–3000 Hz_10ms_modulation_mae_db": 1.1727599379392963, "1000–3000 Hz_40ms_log_rms_mae_db": 0.950271696175294, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07063701832226377, "1000–3000 Hz_40ms_envelope_correlation": 0.9973964079331784, "1000–3000 Hz_40ms_modulation_mae_db": 1.4106816772533624, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8263817564789657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2129237968302478, "3000–6000 Hz_10ms_envelope_correlation": 0.9673938477121936, "3000–6000 Hz_10ms_modulation_mae_db": 0.7964280996836894, "3000–6000 Hz_40ms_log_rms_mae_db": 1.099061299719108, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1344899760526994, "3000–6000 Hz_40ms_envelope_correlation": 0.9857721973921475, "3000–6000 Hz_40ms_modulation_mae_db": 0.5826983084034268, "mr_stft_db": 5.377422313464638, "band_mae_db": 0.2417546381288466, "rms_gain_error_db": -0.08126718289440373, "output_peak": 0.6542672950877755}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7623832991422781, "全帯域_10ms_rms_envelope_nrmse": 0.05153634947943577, "全帯域_10ms_envelope_correlation": 0.9964234652536841, "全帯域_10ms_modulation_mae_db": 0.45367515133077924, "全帯域_40ms_log_rms_mae_db": 0.41785232499366515, "全帯域_40ms_rms_envelope_nrmse": 0.027096382072479795, "全帯域_40ms_envelope_correlation": 0.9989347753899751, "全帯域_40ms_modulation_mae_db": 0.5611407640755532, "300–1000 Hz_10ms_log_rms_mae_db": 1.5224527440168945, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08058405025806542, "300–1000 Hz_10ms_envelope_correlation": 0.9935051008488597, "300–1000 Hz_10ms_modulation_mae_db": 0.8781807174679397, "300–1000 Hz_40ms_log_rms_mae_db": 0.891390469871027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0467531741176989, "300–1000 Hz_40ms_envelope_correlation": 0.9976218169013753, "300–1000 Hz_40ms_modulation_mae_db": 0.871173679099242, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7220233109641958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12292734958877914, "1000–3000 Hz_10ms_envelope_correlation": 0.9906893747750151, "1000–3000 Hz_10ms_modulation_mae_db": 1.0834097064529769, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0072461039521245, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07506097379357665, "1000–3000 Hz_40ms_envelope_correlation": 0.9971867383870712, "1000–3000 Hz_40ms_modulation_mae_db": 0.6597405504322622, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8447472595285972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24804783669736247, "3000–6000 Hz_10ms_envelope_correlation": 0.9536271488418053, "3000–6000 Hz_10ms_modulation_mae_db": 1.1170495913695617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0752432110689767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14178442897481708, "3000–6000 Hz_40ms_envelope_correlation": 0.9830412195747202, "3000–6000 Hz_40ms_modulation_mae_db": 1.2362620023059911, "mr_stft_db": 5.388568317494923, "band_mae_db": 0.21585489757849474, "rms_gain_error_db": -0.06673307238513262, "output_peak": 0.7011917864870979}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7675934915562191, "全帯域_10ms_rms_envelope_nrmse": 0.05135519118993612, "全帯域_10ms_envelope_correlation": 0.9963989639988315, "全帯域_10ms_modulation_mae_db": 0.47904644223755327, "全帯域_40ms_log_rms_mae_db": 0.37965272699588815, "全帯域_40ms_rms_envelope_nrmse": 0.02473474027498048, "全帯域_40ms_envelope_correlation": 0.9990848552730303, "全帯域_40ms_modulation_mae_db": 0.48256394952655146, "300–1000 Hz_10ms_log_rms_mae_db": 1.5272328836678013, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08332441491634898, "300–1000 Hz_10ms_envelope_correlation": 0.9929582860975247, "300–1000 Hz_10ms_modulation_mae_db": 0.7514020870916727, "300–1000 Hz_40ms_log_rms_mae_db": 0.8103630749135378, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0453301098973697, "300–1000 Hz_40ms_envelope_correlation": 0.9976374350271002, "300–1000 Hz_40ms_modulation_mae_db": 0.6516820127573335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6755049858506794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11053759076760754, "1000–3000 Hz_10ms_envelope_correlation": 0.9925715163391854, "1000–3000 Hz_10ms_modulation_mae_db": 0.4884593268393673, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9230765723755952, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06170513119475963, "1000–3000 Hz_40ms_envelope_correlation": 0.9983658493723558, "1000–3000 Hz_40ms_modulation_mae_db": 0.7091199845116242, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8296906279017873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19515659582685652, "3000–6000 Hz_10ms_envelope_correlation": 0.9724645719443583, "3000–6000 Hz_10ms_modulation_mae_db": 1.06471912320707, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1101476354427917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11523359009937655, "3000–6000 Hz_40ms_envelope_correlation": 0.9909557294046184, "3000–6000 Hz_40ms_modulation_mae_db": 1.3734295939602579, "mr_stft_db": 5.375586752284044, "band_mae_db": 0.2185505765873955, "rms_gain_error_db": -0.05299779771336782, "output_peak": 0.6980326895914054}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7184563445131148, "全帯域_10ms_rms_envelope_nrmse": 0.04367927172891506, "全帯域_10ms_envelope_correlation": 0.9975172099696596, "全帯域_10ms_modulation_mae_db": 0.5832401297267702, "全帯域_40ms_log_rms_mae_db": 0.3483839730485046, "全帯域_40ms_rms_envelope_nrmse": 0.024086477153490146, "全帯域_40ms_envelope_correlation": 0.9992250063881194, "全帯域_40ms_modulation_mae_db": 0.6996480721556319, "300–1000 Hz_10ms_log_rms_mae_db": 1.5389700051014816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0779785692638682, "300–1000 Hz_10ms_envelope_correlation": 0.9939490972758187, "300–1000 Hz_10ms_modulation_mae_db": 1.0725475888240006, "300–1000 Hz_40ms_log_rms_mae_db": 0.777511959952972, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04115963084973862, "300–1000 Hz_40ms_envelope_correlation": 0.9981695977932163, "300–1000 Hz_40ms_modulation_mae_db": 1.2181302820831261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.722142359064039, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11117255213550313, "1000–3000 Hz_10ms_envelope_correlation": 0.9926287372248381, "1000–3000 Hz_10ms_modulation_mae_db": 0.7563494361197832, "1000–3000 Hz_40ms_log_rms_mae_db": 0.951338585004414, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0712456358984433, "1000–3000 Hz_40ms_envelope_correlation": 0.9976089075188421, "1000–3000 Hz_40ms_modulation_mae_db": 1.0776173237826514, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8385815166571755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21804241808170916, "3000–6000 Hz_10ms_envelope_correlation": 0.964562357901632, "3000–6000 Hz_10ms_modulation_mae_db": 1.0209070406351306, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0748534730090138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12596310385897708, "3000–6000 Hz_40ms_envelope_correlation": 0.9879051634601766, "3000–6000 Hz_40ms_modulation_mae_db": 1.4419817465308842, "mr_stft_db": 5.36788678539389, "band_mae_db": 0.2280712186750586, "rms_gain_error_db": -0.08044302949289961, "output_peak": 0.6793902957353408}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.508059760835341, "spectral_convergence": 0.14455942499931224, "spectral_convergence_status": "ok", "amplitude_floor": 1.1745223847485646e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.229801201401083, "spectral_convergence": 0.1411140914510885, "spectral_convergence_status": "ok", "amplitude_floor": 9.481908348017771e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.403137756069093, "spectral_convergence": 0.13566955049977195, "spectral_convergence_status": "ok", "amplitude_floor": 8.395402116778918e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004068927413639468, "output_power": 0.004053446892870896, "delta_db": -0.016554550759767007, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018176508169902323, "output_power": 0.001813333798374841, "delta_db": -0.01032699636579354, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007805177117872849, "output_power": 0.0007537297009929897, "delta_db": -0.15167133535097918, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000310938544728363, "output_power": 0.00029936782098044176, "delta_db": -0.16469445380753828, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.08647023559392e-05, "output_power": 7.469662596375161e-05, "delta_db": -0.34458007113980477, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.353447039170603e-05, "output_power": 3.9644689735089175e-05, "delta_db": -0.4064824216988227, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18238497152045094, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04191267933979285, "pitch": null, "output_rms_dbfs": -21.52784082036993, "output_peak": 0.688843687291887, "reference_rms_dbfs": -21.485928141030136, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.380332906101839}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019068326661124743, "log_rms_mae_db": 0.7483733366562493, "rms_envelope_nrmse": 0.04954154956877914, "envelope_correlation": 0.9966392944853716, "rms_gain_error_db": -0.04140200957149194, "reference_mean_power": 0.007104543911662494, "output_mean_power": 0.007037136933684217}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.7734542191089184e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001358301262786015, "output_power": 0.0013242893951292381, "reference_active": true, "delta_db": -0.11013203200257357}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002660490997406596, "output_power": 0.00025558659344607324, "reference_active": true, "delta_db": -0.17423724117664774}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.883599774152297e-05, "output_power": 5.7469379023884945e-05, "reference_active": true, "delta_db": -1.3728806392655661}], "active_band_count": 3, "active_band_mae_db": 0.5524166374815959}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001870958079240233, "log_rms_mae_db": 0.36112749449659814, "rms_envelope_nrmse": 0.02363713972919664, "envelope_correlation": 0.9991517824926307, "rms_gain_error_db": -0.04175433259870232, "reference_mean_power": 0.007117638239873263, "output_mean_power": 0.007049535105254052}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4206220628648622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012103399895685932, "output_power": 0.0011921577300311819, "reference_active": true, "delta_db": -0.06573663470718237}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012555497766250794, "output_power": 0.00012068372538326823, "reference_active": true, "delta_db": -0.17185227180802065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0918852450489625e-06, "output_power": 4.1070668947651134e-06, "reference_active": true, "delta_db": -1.7121993720638875}], "active_band_count": 3, "active_band_mae_db": 0.6499294261930302}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013521834879092463, "log_rms_mae_db": 1.494413084911461, "rms_envelope_nrmse": 0.08206793614298273, "envelope_correlation": 0.9931642638680436, "rms_gain_error_db": -0.05369706272319971, "reference_mean_power": 0.0018937311393695897, "output_mean_power": 0.0018704608141737237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.102346197011796e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006150154421416123, "output_power": 0.0006022606008170283, "reference_active": true, "delta_db": -0.09101567350338952}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012427322589552618, "output_power": 0.00010879648323895706, "reference_active": true, "delta_db": -0.5776271457717856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6589172324725154e-05, "output_power": 1.4997771406382535e-05, "reference_active": true, "delta_db": -2.4867808883264533}], "active_band_count": 3, "active_band_mae_db": 1.0518079025338762}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012428224153258587, "log_rms_mae_db": 0.7490056397219887, "rms_envelope_nrmse": 0.03965345384129995, "envelope_correlation": 0.9982092784180189, "rms_gain_error_db": -0.055036171654347314, "reference_mean_power": 0.0018973976433641784, "output_mean_power": 0.0018735044961193455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.796182335233462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005643605996811222, "output_power": 0.0005520779737665242, "reference_active": true, "delta_db": -0.09556265813061433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.691592416494353e-05, "output_power": 4.889229178230257e-05, "reference_active": true, "delta_db": -0.6599339713870782}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2464097368217303e-06, "output_power": 1.1494008078007276e-06, "reference_active": true, "delta_db": -2.910174748359344}], "active_band_count": 3, "active_band_mae_db": 1.2218904592923454}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012748188769016831, "log_rms_mae_db": 1.7416467766664068, "rms_envelope_nrmse": 0.130602088537094, "envelope_correlation": 0.988749259953117, "rms_gain_error_db": -0.23924586188084354, "reference_mean_power": 0.00032033050063867626, "output_mean_power": 0.0003031612620113638}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0949985705721996e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.663152856065707e-05, "output_power": 6.402384032438212e-05, "reference_active": true, "delta_db": -0.17338055559744942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2062707324852824e-05, "output_power": 1.1082610904161445e-05, "reference_active": true, "delta_db": -0.368027047503609}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2084052775211794e-06, "output_power": 2.0100950362789587e-06, "reference_active": true, "delta_db": -2.0307263091184766}], "active_band_count": 3, "active_band_mae_db": 0.857377970739845}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001243805164613915, "log_rms_mae_db": 0.939330769086352, "rms_envelope_nrmse": 0.07785204054010042, "envelope_correlation": 0.9961631664599249, "rms_gain_error_db": -0.23991056853646914, "reference_mean_power": 0.0003214084172642619, "output_mean_power": 0.000304134851179693}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.993543951674123e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.454820101360055e-05, "output_power": 6.214869759977611e-05, "reference_active": true, "delta_db": -0.16452110882067464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.405216922862835e-06, "output_power": 5.1225460537951985e-06, "reference_active": true, "delta_db": -0.23327256224505966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4345993241652877e-07, "output_power": 1.763868126574568e-07, "reference_active": true, "delta_db": -2.89409967610034}], "active_band_count": 3, "active_band_mae_db": 1.0972977823886914}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.885927766129993e-05, "log_rms_mae_db": 1.849460902816375, "rms_envelope_nrmse": 0.21749817665096954, "envelope_correlation": 0.9646435833387569, "rms_gain_error_db": -0.4001121379094033, "reference_mean_power": 5.543232992418922e-05, "output_mean_power": 5.055358039553069e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.209399643069482e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1581941196726732e-05, "output_power": 2.151946488305538e-05, "reference_active": true, "delta_db": -0.01259037221667849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7792321883200776e-05, "output_power": 1.5770986545531855e-05, "reference_active": true, "delta_db": -0.5237376561575631}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1497984756130906e-05, "output_power": 5.505261359498273e-06, "reference_active": true, "delta_db": -3.1984378682067858}], "active_band_count": 3, "active_band_mae_db": 1.2449219655270092}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.78543565194846e-05, "log_rms_mae_db": 1.0582539206532562, "rms_envelope_nrmse": 0.13051886150791844, "envelope_correlation": 0.9866511084526003, "rms_gain_error_db": -0.40037995346580957, "reference_mean_power": 5.5701179342093606e-05, "output_mean_power": 5.0795635146417886e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.455779259683724e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.849693387286164e-05, "output_power": 2.59575631600599e-05, "reference_active": true, "delta_db": -0.4053421520002205}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0110098860775756e-05, "output_power": 7.3244752949563945e-06, "reference_active": true, "delta_db": -1.3997888383348465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1498909214748545e-06, "output_power": 5.234320730125209e-07, "reference_active": true, "delta_db": -3.417963149063135}], "active_band_count": 3, "active_band_mae_db": 1.7410313797994006}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9232887204089153, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "665ce7a0bab7c11e461eb75dcfbdd6216a17f0fd2a22bb56ce7beaab28d6dd91", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9915669737387276, "standard_deviation": 0.0014123404998393372, "minimum": 0.988749259953117, "maximum": 0.9931671277814509, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7029964219158793, "standard_deviation": 0.03190562755465885, "minimum": 1.6446603348678057, "maximum": 1.7416467766664068, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8655464338044541, "standard_deviation": 0.2886103879424524, "minimum": 0.4884593268393673, "maximum": 1.2747366089664076, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11625427644812056, "standard_deviation": 0.009094596452493374, "minimum": 0.10122670442585895, "maximum": 0.130602088537094, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974588293908323, "standard_deviation": 0.0007109346068209109, "minimum": 0.9961631664599249, "maximum": 0.9983658493723558, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9599014605228087, "standard_deviation": 0.04807835562772294, "minimum": 0.8820760660597323, "maximum": 1.0254218676931823, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.938511060858251, "standard_deviation": 0.40639040886899414, "minimum": 0.4659900852494505, "maximum": 1.5501671498774698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06998233611825597, "standard_deviation": 0.008537083999506798, "minimum": 0.05305658848580217, "maximum": 0.07785204054010042, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9652329697916906, "standard_deviation": 0.007340719975637213, "minimum": 0.9536271488418053, "maximum": 0.9755141186783014, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8377699746731018, "standard_deviation": 0.01835924484376722, "minimum": 1.801448973165022, "maximum": 1.8561071221274839, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9736328956863578, "standard_deviation": 0.16183644981932763, "minimum": 0.7964280996836894, "maximum": 1.2449219655270092, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21546863614418432, "standard_deviation": 0.02163449674117463, "minimum": 0.18222833805245145, "maximum": 0.24804783669736247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9863207016321911, "standard_deviation": 0.003999502493202644, "minimum": 0.9783909803937333, "maximum": 0.9909557294046184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0728862439442755, "standard_deviation": 0.03443919544027217, "minimum": 1.0003252333445183, "maximum": 1.1101476354427917, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.098178708638474, "standard_deviation": 0.41904064944793307, "minimum": 0.5826983084034268, "maximum": 1.7410313797994006, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13011050757621942, "standard_deviation": 0.015510791071775202, "minimum": 0.11397440015402481, "maximum": 0.16013287031019977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933496508519162, "standard_deviation": 0.0005638834761736328, "minimum": 0.9925349780515826, "maximum": 0.9940924093248303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5194219347276285, "standard_deviation": 0.0429951469896697, "minimum": 1.4667088511211028, "maximum": 1.600725954004648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8901415922948627, "standard_deviation": 0.1192546777268833, "minimum": 0.7514020870916727, "maximum": 1.0725475888240006, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08110464561105299, "standard_deviation": 0.0031542887438224744, "minimum": 0.07696524709967896, "maximum": 0.08561589888353989, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978504884749619, "standard_deviation": 0.0002950497825184506, "minimum": 0.9974282161834414, "maximum": 0.9982092784180189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8071058284532079, "standard_deviation": 0.044809598348208864, "minimum": 0.7490056397219887, "maximum": 0.891390469871027, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9730095797377938, "standard_deviation": 0.19823225328370006, "minimum": 0.6516820127573335, "maximum": 1.2218904592923454, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04368143507287403, "standard_deviation": 0.002815139050419529, "minimum": 0.03965345384129995, "maximum": 0.04742786420359861, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22049200141059533, "standard_deviation": 0.03361441254072903, "minimum": 0.1636963754165128, "maximum": 0.26222276479295165, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.378875768822613, "standard_deviation": 0.008058255799944179, "minimum": 5.36788678539389, "maximum": 5.391731854287145, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6829356828099202, "standard_deviation": 0.014415836293245343, "minimum": 0.6542672950877755, "maximum": 0.7011917864870979, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06406612181266055, "standard_deviation": 0.015144084220856238, "minimum": -0.08126718289440373, "maximum": -0.04191267933979285, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9966931342199292, "standard_deviation": 0.00041618210048065496, "minimum": 0.9963352456742314, "maximum": 0.9975172099696596, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7527617739625188, "standard_deviation": 0.01831284399920093, "minimum": 0.7184563445131148, "maximum": 0.770157166542763, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.49263272238902533, "standard_deviation": 0.12873086116006616, "minimum": 0.1979620397011246, "maximum": 0.5999384955807944, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.049446502360897815, "standard_deviation": 0.0029975565112467647, "minimum": 0.04367927172891506, "maximum": 0.0521559276973548, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990451875347682, "standard_deviation": 0.00015098969327315858, "minimum": 0.9987860628392695, "maximum": 0.9992250063881194, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3838896274736731, "standard_deviation": 0.02528902962945043, "minimum": 0.3483839730485046, "maximum": 0.41785232499366515, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5487567982938552, "standard_deviation": 0.16481698235595985, "minimum": 0.19154215055297516, "maximum": 0.6996480721556319, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.025622979827125615, "standard_deviation": 0.0019235832594462158, "minimum": 0.023461175316999646, "maximum": 0.028502206709293865, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5696-0010", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.998669147491455, "predicted_mos": 2.998669147491455, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "033b59d6e7e8fa9c6dddfa1a62c5b428c836575ee90f79782dd5a767de0a2d96", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.391876100096852, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7471117565350721, "全帯域_10ms_rms_envelope_nrmse": 0.05057439509326536, "全帯域_10ms_envelope_correlation": 0.9965071235321737, "全帯域_10ms_modulation_mae_db": 0.5987093170828547, "全帯域_40ms_log_rms_mae_db": 0.38272206664838326, "全帯域_40ms_rms_envelope_nrmse": 0.026958805423573585, "全帯域_40ms_envelope_correlation": 0.9989012038188132, "全帯域_40ms_modulation_mae_db": 0.6887652301334194, "300–1000 Hz_10ms_log_rms_mae_db": 1.5774535479271747, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07894767157418467, "300–1000 Hz_10ms_envelope_correlation": 0.9937712513082513, "300–1000 Hz_10ms_modulation_mae_db": 0.915706488379017, "300–1000 Hz_40ms_log_rms_mae_db": 0.8216816948851121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04016877430752059, "300–1000 Hz_40ms_envelope_correlation": 0.9982470516247314, "300–1000 Hz_40ms_modulation_mae_db": 0.9325138952893375, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0715777169956504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13092033007079956, "1000–3000 Hz_10ms_envelope_correlation": 0.9885335610320757, "1000–3000 Hz_10ms_modulation_mae_db": 0.24899653628553456, "1000–3000 Hz_40ms_log_rms_mae_db": 1.161330798356849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07262924744822251, "1000–3000 Hz_40ms_envelope_correlation": 0.9963001801733163, "1000–3000 Hz_40ms_modulation_mae_db": 0.48183953982833505, "3000–6000 Hz_10ms_log_rms_mae_db": 2.545460029670473, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2776604291662159, "3000–6000 Hz_10ms_envelope_correlation": 0.9474205454145516, "3000–6000 Hz_10ms_modulation_mae_db": 2.962777815598804, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5124764701310789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1882008646798481, "3000–6000 Hz_40ms_envelope_correlation": 0.9796640583501028, "3000–6000 Hz_40ms_modulation_mae_db": 3.7813475627221727, "mr_stft_db": 5.822821274748783, "band_mae_db": 0.28323260662521865, "rms_gain_error_db": -0.0505332809659493, "output_peak": 0.6865223851118339}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7024081212300757, "全帯域_10ms_rms_envelope_nrmse": 0.04752004663272372, "全帯域_10ms_envelope_correlation": 0.9969296505326751, "全帯域_10ms_modulation_mae_db": 0.6103181579955145, "全帯域_40ms_log_rms_mae_db": 0.35055409251918557, "全帯域_40ms_rms_envelope_nrmse": 0.024149734346371046, "全帯域_40ms_envelope_correlation": 0.999122448286972, "全帯域_40ms_modulation_mae_db": 0.5872320769077717, "300–1000 Hz_10ms_log_rms_mae_db": 1.5957246704904704, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08199154392224997, "300–1000 Hz_10ms_envelope_correlation": 0.993237512850882, "300–1000 Hz_10ms_modulation_mae_db": 0.7678537371928327, "300–1000 Hz_40ms_log_rms_mae_db": 0.8453044617749351, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04643436735685151, "300–1000 Hz_40ms_envelope_correlation": 0.9975647825497923, "300–1000 Hz_40ms_modulation_mae_db": 0.7464968952799933, "1000–3000 Hz_10ms_log_rms_mae_db": 2.097326776046737, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12854971274441132, "1000–3000 Hz_10ms_envelope_correlation": 0.9891787609270928, "1000–3000 Hz_10ms_modulation_mae_db": 0.5679560700393157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1455963873720563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06912114753459446, "1000–3000 Hz_40ms_envelope_correlation": 0.9970751809564715, "1000–3000 Hz_40ms_modulation_mae_db": 0.7789049410783501, "3000–6000 Hz_10ms_log_rms_mae_db": 2.504350194971692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24220691349833867, "3000–6000 Hz_10ms_envelope_correlation": 0.957859994133758, "3000–6000 Hz_10ms_modulation_mae_db": 1.4698918399885954, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4746601565633992, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15075907443125605, "3000–6000 Hz_40ms_envelope_correlation": 0.984520545268149, "3000–6000 Hz_40ms_modulation_mae_db": 1.8916156788364404, "mr_stft_db": 5.812051123613994, "band_mae_db": 0.18529292834721556, "rms_gain_error_db": -0.04698760728404479, "output_peak": 0.7139989141519327}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7898265828164982, "全帯域_10ms_rms_envelope_nrmse": 0.05027386809397689, "全帯域_10ms_envelope_correlation": 0.9966380891037799, "全帯域_10ms_modulation_mae_db": 0.5926219009024005, "全帯域_40ms_log_rms_mae_db": 0.3953298840695134, "全帯域_40ms_rms_envelope_nrmse": 0.027428837475077236, "全帯域_40ms_envelope_correlation": 0.9989581437805704, "全帯域_40ms_modulation_mae_db": 0.677650780455422, "300–1000 Hz_10ms_log_rms_mae_db": 1.5802814895721438, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08223918878692567, "300–1000 Hz_10ms_envelope_correlation": 0.9932486445457775, "300–1000 Hz_10ms_modulation_mae_db": 0.8110908318859127, "300–1000 Hz_40ms_log_rms_mae_db": 0.8313549374984557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04565212552136573, "300–1000 Hz_40ms_envelope_correlation": 0.9977535837285609, "300–1000 Hz_40ms_modulation_mae_db": 1.017720191771023, "1000–3000 Hz_10ms_log_rms_mae_db": 2.120259748864915, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13625458031462567, "1000–3000 Hz_10ms_envelope_correlation": 0.9875770342405249, "1000–3000 Hz_10ms_modulation_mae_db": 1.0792570944546094, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2056041200558287, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0742963736063036, "1000–3000 Hz_40ms_envelope_correlation": 0.9962449862704792, "1000–3000 Hz_40ms_modulation_mae_db": 0.7883615345772883, "3000–6000 Hz_10ms_log_rms_mae_db": 2.57245088740499, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2405914722838476, "3000–6000 Hz_10ms_envelope_correlation": 0.9567208506991379, "3000–6000 Hz_10ms_modulation_mae_db": 0.9595138933199627, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5094082400805109, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14197744530149836, "3000–6000 Hz_40ms_envelope_correlation": 0.9835651906392301, "3000–6000 Hz_40ms_modulation_mae_db": 1.3760657271569727, "mr_stft_db": 5.847565561370391, "band_mae_db": 0.13522115267173582, "rms_gain_error_db": -0.0872781326297, "output_peak": 0.7102723292403151}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7732588302986616, "全帯域_10ms_rms_envelope_nrmse": 0.048868081112006204, "全帯域_10ms_envelope_correlation": 0.9967751389943501, "全帯域_10ms_modulation_mae_db": 0.35953647613514655, "全帯域_40ms_log_rms_mae_db": 0.3798424176420655, "全帯域_40ms_rms_envelope_nrmse": 0.027259693178136292, "全帯域_40ms_envelope_correlation": 0.9989242546228625, "全帯域_40ms_modulation_mae_db": 0.2757952191700201, "300–1000 Hz_10ms_log_rms_mae_db": 1.5522173774123913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08006102167826097, "300–1000 Hz_10ms_envelope_correlation": 0.993554187635662, "300–1000 Hz_10ms_modulation_mae_db": 0.7945479867937603, "300–1000 Hz_40ms_log_rms_mae_db": 0.7945014143473398, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04428954787684245, "300–1000 Hz_40ms_envelope_correlation": 0.9978644850632845, "300–1000 Hz_40ms_modulation_mae_db": 0.8916604988156283, "1000–3000 Hz_10ms_log_rms_mae_db": 2.123345376131897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12742296212649218, "1000–3000 Hz_10ms_envelope_correlation": 0.9891763050170496, "1000–3000 Hz_10ms_modulation_mae_db": 0.5838503941133982, "1000–3000 Hz_40ms_log_rms_mae_db": 1.210413055937063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07172035818857277, "1000–3000 Hz_40ms_envelope_correlation": 0.9964008879138417, "1000–3000 Hz_40ms_modulation_mae_db": 1.0038596884179267, "3000–6000 Hz_10ms_log_rms_mae_db": 2.513679015437753, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2315957505110772, "3000–6000 Hz_10ms_envelope_correlation": 0.9610254330473802, "3000–6000 Hz_10ms_modulation_mae_db": 1.3925359683180911, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4736641878573775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1364783649074094, "3000–6000 Hz_40ms_envelope_correlation": 0.986838060737082, "3000–6000 Hz_40ms_modulation_mae_db": 1.7419215079356605, "mr_stft_db": 5.85234815318314, "band_mae_db": 0.177428929547115, "rms_gain_error_db": -0.07201332537360253, "output_peak": 0.6954070761396671}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7194277675995115, "全帯域_10ms_rms_envelope_nrmse": 0.04501038645833446, "全帯域_10ms_envelope_correlation": 0.9972877233186701, "全帯域_10ms_modulation_mae_db": 0.5633095080963617, "全帯域_40ms_log_rms_mae_db": 0.3637364964708168, "全帯域_40ms_rms_envelope_nrmse": 0.023628275464102438, "全帯域_40ms_envelope_correlation": 0.9992255362127179, "全帯域_40ms_modulation_mae_db": 0.5442175544939231, "300–1000 Hz_10ms_log_rms_mae_db": 1.56720806844513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07960213160366941, "300–1000 Hz_10ms_envelope_correlation": 0.9936607378043372, "300–1000 Hz_10ms_modulation_mae_db": 0.9359337932000602, "300–1000 Hz_40ms_log_rms_mae_db": 0.8608191978473706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04275028901763747, "300–1000 Hz_40ms_envelope_correlation": 0.9980198895108225, "300–1000 Hz_40ms_modulation_mae_db": 1.0621637626952298, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0816095751888173, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12575171725450782, "1000–3000 Hz_10ms_envelope_correlation": 0.9894382295818837, "1000–3000 Hz_10ms_modulation_mae_db": 0.7889790311910762, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1083101138942046, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06670038517536088, "1000–3000 Hz_40ms_envelope_correlation": 0.9969190211551117, "1000–3000 Hz_40ms_modulation_mae_db": 1.4067793122929644, "3000–6000 Hz_10ms_log_rms_mae_db": 2.513437135906166, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2649964912733424, "3000–6000 Hz_10ms_envelope_correlation": 0.9485791224852477, "3000–6000 Hz_10ms_modulation_mae_db": 1.9580625312992794, "3000–6000 Hz_40ms_log_rms_mae_db": 1.526408333294242, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16249866298441146, "3000–6000 Hz_40ms_envelope_correlation": 0.9812377321112993, "3000–6000 Hz_40ms_modulation_mae_db": 2.4442973077151517, "mr_stft_db": 5.844844098062441, "band_mae_db": 0.1823961843937991, "rms_gain_error_db": -0.07291666301843883, "output_peak": 0.6576110628016456}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7268146915345833, "全帯域_10ms_rms_envelope_nrmse": 0.0469484319710251, "全帯域_10ms_envelope_correlation": 0.9971282188813254, "全帯域_10ms_modulation_mae_db": 0.4799355286841795, "全帯域_40ms_log_rms_mae_db": 0.34761185434231345, "全帯域_40ms_rms_envelope_nrmse": 0.025342507808192728, "全帯域_40ms_envelope_correlation": 0.9991584466996897, "全帯域_40ms_modulation_mae_db": 0.5081185549838655, "300–1000 Hz_10ms_log_rms_mae_db": 1.5577349572885295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08429032342907766, "300–1000 Hz_10ms_envelope_correlation": 0.992960095924113, "300–1000 Hz_10ms_modulation_mae_db": 1.0219346115311962, "300–1000 Hz_40ms_log_rms_mae_db": 0.8255420771430328, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044071563851182806, "300–1000 Hz_40ms_envelope_correlation": 0.9979725135415038, "300–1000 Hz_40ms_modulation_mae_db": 1.128521654009388, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1186656830716624, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15131633114939988, "1000–3000 Hz_10ms_envelope_correlation": 0.9861186936825984, "1000–3000 Hz_10ms_modulation_mae_db": 0.6803238062745777, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2421031195232044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09732361003243048, "1000–3000 Hz_40ms_envelope_correlation": 0.9954233885660425, "1000–3000 Hz_40ms_modulation_mae_db": 0.6649360699357257, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5451783483346455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22823950290881337, "3000–6000 Hz_10ms_envelope_correlation": 0.9621038270934102, "3000–6000 Hz_10ms_modulation_mae_db": 1.0302414400083781, "3000–6000 Hz_40ms_log_rms_mae_db": 1.51172998905922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13348467336576697, "3000–6000 Hz_40ms_envelope_correlation": 0.9872976260041006, "3000–6000 Hz_40ms_modulation_mae_db": 1.2541358991437102, "mr_stft_db": 5.830139032460413, "band_mae_db": 0.24080785915704664, "rms_gain_error_db": -0.09011796274973873, "output_peak": 0.6674663033158611}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.75119407888588, "全帯域_10ms_rms_envelope_nrmse": 0.05029504836981849, "全帯域_10ms_envelope_correlation": 0.9965436219380931, "全帯域_10ms_modulation_mae_db": 0.333046472610107, "全帯域_40ms_log_rms_mae_db": 0.37858920371117927, "全帯域_40ms_rms_envelope_nrmse": 0.02759033428697757, "全帯域_40ms_envelope_correlation": 0.9988304682126061, "全帯域_40ms_modulation_mae_db": 0.45293535868583, "300–1000 Hz_10ms_log_rms_mae_db": 1.593451779748448, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08651959162097445, "300–1000 Hz_10ms_envelope_correlation": 0.9923858302182358, "300–1000 Hz_10ms_modulation_mae_db": 0.6761802701864972, "300–1000 Hz_40ms_log_rms_mae_db": 0.8501698892010162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048390578962423846, "300–1000 Hz_40ms_envelope_correlation": 0.9973006054406977, "300–1000 Hz_40ms_modulation_mae_db": 0.7710308825719325, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1632886153635176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15090887693212837, "1000–3000 Hz_10ms_envelope_correlation": 0.9852103850101254, "1000–3000 Hz_10ms_modulation_mae_db": 0.7474196644193926, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2558047623006205, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09236562518184137, "1000–3000 Hz_40ms_envelope_correlation": 0.994879224422501, "1000–3000 Hz_40ms_modulation_mae_db": 0.5431697641833005, "3000–6000 Hz_10ms_log_rms_mae_db": 2.534302242249267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2150524374758936, "3000–6000 Hz_10ms_envelope_correlation": 0.9666962906112154, "3000–6000 Hz_10ms_modulation_mae_db": 1.0712001550024939, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4901813259511083, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12095687677886263, "3000–6000 Hz_40ms_envelope_correlation": 0.990068702894842, "3000–6000 Hz_40ms_modulation_mae_db": 1.4546830882521178, "mr_stft_db": 5.831101334375179, "band_mae_db": 0.19515094326362123, "rms_gain_error_db": -0.041401736174962206, "output_peak": 0.6539427532910372}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7552385344037387, "全帯域_10ms_rms_envelope_nrmse": 0.048947475133690466, "全帯域_10ms_envelope_correlation": 0.9969248999623156, "全帯域_10ms_modulation_mae_db": 0.6490131064283766, "全帯域_40ms_log_rms_mae_db": 0.3811589427705679, "全帯域_40ms_rms_envelope_nrmse": 0.02753039081117262, "全帯域_40ms_envelope_correlation": 0.9990571996423694, "全帯域_40ms_modulation_mae_db": 0.7072718878366037, "300–1000 Hz_10ms_log_rms_mae_db": 1.5684287943996558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08172965504902197, "300–1000 Hz_10ms_envelope_correlation": 0.9933578685406905, "300–1000 Hz_10ms_modulation_mae_db": 0.8625546105299051, "300–1000 Hz_40ms_log_rms_mae_db": 0.8301016326917586, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04681165886817345, "300–1000 Hz_40ms_envelope_correlation": 0.9976502563147629, "300–1000 Hz_40ms_modulation_mae_db": 0.9772922493734203, "1000–3000 Hz_10ms_log_rms_mae_db": 2.123483853517578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13843174645277068, "1000–3000 Hz_10ms_envelope_correlation": 0.9882149394532485, "1000–3000 Hz_10ms_modulation_mae_db": 1.067564544146129, "1000–3000 Hz_40ms_log_rms_mae_db": 1.206153928334138, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0856332832202145, "1000–3000 Hz_40ms_envelope_correlation": 0.9963495708404493, "1000–3000 Hz_40ms_modulation_mae_db": 0.3315171561242947, "3000–6000 Hz_10ms_log_rms_mae_db": 2.559984280584638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2645897386741263, "3000–6000 Hz_10ms_envelope_correlation": 0.9503891480486226, "3000–6000 Hz_10ms_modulation_mae_db": 2.10782565014568, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5553622182338767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1748892377935993, "3000–6000 Hz_40ms_envelope_correlation": 0.9799023005044996, "3000–6000 Hz_40ms_modulation_mae_db": 2.8008712396697297, "mr_stft_db": 5.84454492934713, "band_mae_db": 0.30518471421803844, "rms_gain_error_db": -0.11202093972557495, "output_peak": 0.6598450147178457}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.001039446259472, "spectral_convergence": 0.144063613094613, "spectral_convergence_status": "ok", "amplitude_floor": 1.1745223847485646e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.754541504538568, "spectral_convergence": 0.14187832167475364, "spectral_convergence_status": "ok", "amplitude_floor": 9.481908348017771e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7128828734483115, "spectral_convergence": 0.1360707255740012, "spectral_convergence_status": "ok", "amplitude_floor": 8.395402116778918e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004068927413639468, "output_power": 0.0040507791963347905, "delta_db": -0.019413715630331597, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018176508169902323, "output_power": 0.0018001624426707786, "delta_db": -0.041987593492879616, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007805177117872849, "output_power": 0.0007466147594195272, "delta_db": -0.192861916704518, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000310938544728363, "output_power": 0.00031408332883669956, "delta_db": 0.04370323561714222, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.08647023559392e-05, "output_power": 7.433308474058596e-05, "delta_db": -0.365768364345075, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.353447039170603e-05, "output_power": 3.429787270303054e-05, "delta_db": -1.0356608139613654, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.28323260662521865, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0505332809659493, "pitch": null, "output_rms_dbfs": -21.536461421996087, "output_peak": 0.6865223851118339, "reference_rms_dbfs": -21.485928141030136, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.822821274748783}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019068326661124743, "log_rms_mae_db": 0.7471117565350721, "rms_envelope_nrmse": 0.05057439509326536, "envelope_correlation": 0.9965071235321737, "rms_gain_error_db": -0.050790865437481056, "reference_mean_power": 0.007104543911662494, "output_mean_power": 0.007021940033729633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.7734542191089184e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001358301262786015, "output_power": 0.0013454775680210589, "reference_active": true, "delta_db": -0.04119642902085929}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002660490997406596, "output_power": 0.00024522121076724876, "reference_active": true, "delta_db": -0.3540376128621358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.883599774152297e-05, "output_power": 5.7099877621319215e-05, "reference_active": true, "delta_db": -1.400893909365569}], "active_band_count": 3, "active_band_mae_db": 0.5987093170828547}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001870958079240233, "log_rms_mae_db": 0.38272206664838326, "rms_envelope_nrmse": 0.026958805423573585, "envelope_correlation": 0.9989012038188132, "rms_gain_error_db": -0.05020519508119847, "reference_mean_power": 0.007117638239873263, "output_mean_power": 0.0070358308724933335}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4206220628648622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012103399895685932, "output_power": 0.001209362768167091, "reference_active": true, "delta_db": -0.003507884462148489}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012555497766250794, "output_power": 0.00011398702082476865, "reference_active": true, "delta_db": -0.41978532071269414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0918852450489625e-06, "output_power": 4.173029600160256e-06, "reference_active": true, "delta_db": -1.6430024852254153}], "active_band_count": 3, "active_band_mae_db": 0.6887652301334194}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013521834879092463, "log_rms_mae_db": 1.5774535479271747, "rms_envelope_nrmse": 0.07894767157418467, "envelope_correlation": 0.9937712513082513, "rms_gain_error_db": -0.0805709722968518, "reference_mean_power": 0.0018937311393695897, "output_mean_power": 0.0018589222398496333}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.102346197011796e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006150154421416123, "output_power": 0.0005984788668455147, "reference_active": true, "delta_db": -0.11837200967123995}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012427322589552618, "output_power": 0.00010955251111251529, "reference_active": true, "delta_db": -0.5475523522736383}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6589172324725154e-05, "output_power": 1.6465893737969295e-05, "reference_active": true, "delta_db": -2.081195103192173}], "active_band_count": 3, "active_band_mae_db": 0.915706488379017}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012428224153258587, "log_rms_mae_db": 0.8216816948851121, "rms_envelope_nrmse": 0.04016877430752059, "envelope_correlation": 0.9982470516247314, "rms_gain_error_db": -0.08052882854457627, "reference_mean_power": 0.0018973976433641784, "output_mean_power": 0.0018625394233097825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.796182335233462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005643605996811222, "output_power": 0.0005484550386400601, "reference_active": true, "delta_db": -0.12415655581134612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.691592416494353e-05, "output_power": 4.862848766695019e-05, "reference_active": true, "delta_db": -0.6834302870861847}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2464097368217303e-06, "output_power": 1.4206709123744334e-06, "reference_active": true, "delta_db": -1.9899548429704816}], "active_band_count": 3, "active_band_mae_db": 0.9325138952893375}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012748188769016831, "log_rms_mae_db": 2.0715777169956504, "rms_envelope_nrmse": 0.13092033007079956, "envelope_correlation": 0.9885335610320757, "rms_gain_error_db": 0.015551611168526639, "reference_mean_power": 0.00032033050063867626, "output_mean_power": 0.0003214796254001643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0949985705721996e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.663152856065707e-05, "output_power": 6.385470299713112e-05, "reference_active": true, "delta_db": -0.18486887081200234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2062707324852824e-05, "output_power": 1.1368637528317944e-05, "reference_active": true, "delta_db": -0.25736370860325875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2084052775211794e-06, "output_power": 3.441635913913595e-06, "reference_active": true, "delta_db": 0.30475702944134264}], "active_band_count": 3, "active_band_mae_db": 0.24899653628553456}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001243805164613915, "log_rms_mae_db": 1.161330798356849, "rms_envelope_nrmse": 0.07262924744822251, "envelope_correlation": 0.9963001801733163, "rms_gain_error_db": 0.01668655137511753, "reference_mean_power": 0.0003214084172642619, "output_mean_power": 0.0003226457147189029}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.993543951674123e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.454820101360055e-05, "output_power": 6.1788063123059e-05, "reference_active": true, "delta_db": -0.1897956124230081}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.405216922862835e-06, "output_power": 5.277429814372464e-06, "reference_active": true, "delta_db": -0.10390661555030982}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4345993241652877e-07, "output_power": 2.634477179616807e-07, "reference_active": true, "delta_db": -1.1518163915116872}], "active_band_count": 3, "active_band_mae_db": 0.48183953982833505}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.885927766129993e-05, "log_rms_mae_db": 2.545460029670473, "rms_envelope_nrmse": 0.2776604291662159, "envelope_correlation": 0.9474205454145516, "rms_gain_error_db": -0.9564539798916223, "reference_mean_power": 5.543232992418922e-05, "output_mean_power": 4.447518212569855e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.209399643069482e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1581941196726732e-05, "output_power": 1.6880978623143213e-05, "reference_active": true, "delta_db": -1.0669288498784062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7792321883200776e-05, "output_power": 9.270109552111716e-06, "reference_active": true, "delta_db": -2.8314776027948336}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1497984756130906e-05, "output_power": 3.644425100448964e-06, "reference_active": true, "delta_db": -4.989926994123173}], "active_band_count": 3, "active_band_mae_db": 2.962777815598804}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.78543565194846e-05, "log_rms_mae_db": 1.5124764701310789, "rms_envelope_nrmse": 0.1882008646798481, "envelope_correlation": 0.9796640583501028, "rms_gain_error_db": -0.9577458297656358, "reference_mean_power": 5.5701179342093606e-05, "output_mean_power": 4.467759711015114e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.455779259683724e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.849693387286164e-05, "output_power": 1.8715769013201137e-05, "reference_active": true, "delta_db": -1.8259045804893463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0110098860775756e-05, "output_power": 4.3960429792517594e-06, "reference_active": true, "delta_db": -3.6169347254052218}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1498909214748545e-06, "output_power": 2.95485590366696e-07, "reference_active": true, "delta_db": -5.90120338227195}], "active_band_count": 3, "active_band_mae_db": 3.7813475627221727}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9098774389550371, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "28dae8ca1fd8254732563f40598e5bb0aa141ca9da7e47ff6d3ec1b47452c44c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9879309886180749, "standard_deviation": 0.0015415496011077952, "minimum": 0.9852103850101254, "maximum": 0.9894382295818837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.112444668147597, "standard_deviation": 0.028722897401093413, "minimum": 2.0715777169956504, "maximum": 2.1632886153635176, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7205433926155042, "standard_deviation": 0.2726711493462558, "minimum": 0.24899653628553456, "maximum": 1.0792570944546094, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13619453213064195, "standard_deviation": 0.010159040842656345, "minimum": 0.12575171725450782, "maximum": 0.15131633114939988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961990550372766, "standard_deviation": 0.0007275483932653745, "minimum": 0.994879224422501, "maximum": 0.9970751809564715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1919145357217453, "standard_deviation": 0.04986216784372206, "minimum": 1.1083101138942046, "maximum": 1.2558047623006205, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7499210008047732, "standard_deviation": 0.33703823461574783, "minimum": 0.3315171561242947, "maximum": 1.4067793122929644, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07872375379844257, "standard_deviation": 0.011478928501441903, "minimum": 0.06670038517536088, "maximum": 0.09732361003243048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9563494014416654, "standard_deviation": 0.006970171057555475, "minimum": 0.9474205454145516, "maximum": 0.9666962906112154, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.536105266819953, "standard_deviation": 0.02417040760375054, "minimum": 2.504350194971692, "maximum": 2.57245088740499, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.6190061617101605, "standard_deviation": 0.6881460779138484, "minimum": 0.9595138933199627, "maximum": 2.962777815598804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24561659197395688, "standard_deviation": 0.021487407573666798, "minimum": 0.2150524374758936, "maximum": 0.2776604291662159, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9841367770636631, "standard_deviation": 0.003766204961934227, "minimum": 0.9796640583501028, "maximum": 0.990068702894842, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5067363651463517, "standard_deviation": 0.027289991128172868, "minimum": 1.4736641878573775, "maximum": 1.5553622182338767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.0931172514289944, "standard_deviation": 0.8671757802061173, "minimum": 1.2541358991437102, "maximum": 3.7813475627221727, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15115565003033152, "standard_deviation": 0.022641125017558023, "minimum": 0.12095687677886263, "maximum": 0.1882008646798481, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932720161034937, "standard_deviation": 0.00044266854588327035, "minimum": 0.9923858302182358, "maximum": 0.9937712513082513, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.574062585660493, "standard_deviation": 0.015665652781074006, "minimum": 1.5522173774123913, "maximum": 1.5957246704904704, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8482252912123976, "standard_deviation": 0.10885511188477347, "minimum": 0.6761802701864972, "maximum": 1.0219346115311962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08192264095804559, "standard_deviation": 0.002523628403426986, "minimum": 0.07894767157418467, "maximum": 0.08651959162097445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977966459717695, "standard_deviation": 0.00029552268894197535, "minimum": 0.9973006054406977, "maximum": 0.9982470516247314, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8324344131736277, "standard_deviation": 0.020334790879913488, "minimum": 0.7945014143473398, "maximum": 0.8608191978473706, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9409250037257442, "standard_deviation": 0.13431065125436387, "minimum": 0.7464968952799933, "maximum": 1.128521654009388, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04482111322024973, "standard_deviation": 0.0025849422271214803, "minimum": 0.04016877430752059, "maximum": 0.048390578962423846, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2130894147779738, "standard_deviation": 0.05799207822773111, "minimum": 0.13522115267173582, "maximum": 0.30518471421803844, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.835676938395184, "standard_deviation": 0.013923251902566826, "minimum": 5.812051123613994, "maximum": 5.85234815318314, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6806332298462674, "standard_deviation": 0.024179723922496156, "minimum": 0.6539427532910372, "maximum": 0.7139989141519327, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07165870599025143, "standard_deviation": 0.024453920771777566, "minimum": -0.11202093972557495, "maximum": -0.041401736174962206, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968418082829229, "standard_deviation": 0.00027835409479525934, "minimum": 0.9965071235321737, "maximum": 0.9972877233186701, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7456600454130027, "standard_deviation": 0.02866273116285442, "minimum": 0.7024081212300757, "maximum": 0.7898265828164982, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5233113084918677, "standard_deviation": 0.11974029606785211, "minimum": 0.333046472610107, "maximum": 0.6490131064283766, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04855471660810509, "standard_deviation": 0.0019467468369856761, "minimum": 0.04501038645833446, "maximum": 0.05057439509326536, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990222126595751, "standard_deviation": 0.000139530396899621, "minimum": 0.9988304682126061, "maximum": 0.9992255362127179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37244311977175315, "standard_deviation": 0.016773605125825805, "minimum": 0.34761185434231345, "maximum": 0.3953298840695134, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.555248332833357, "standard_deviation": 0.14533120973207234, "minimum": 0.2757952191700201, "maximum": 0.7072718878366037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02623607234920044, "standard_deviation": 0.0016236636468083651, "minimum": 0.023628275464102438, "maximum": 0.02759033428697757, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5696-0010", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.2653579711914062, "predicted_mos": 3.2653579711914062, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "40e1ec59fa861ffc833f2d710841436c7e1135201d790fa96ab3b50d1f02a573", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.296217099763453, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7476324890830811, "全帯域_10ms_rms_envelope_nrmse": 0.04903644164635436, "全帯域_10ms_envelope_correlation": 0.9967382762279056, "全帯域_10ms_modulation_mae_db": 0.581498910688351, "全帯域_40ms_log_rms_mae_db": 0.3639538702298398, "全帯域_40ms_rms_envelope_nrmse": 0.025394014623708962, "全帯域_40ms_envelope_correlation": 0.9990457070600017, "全帯域_40ms_modulation_mae_db": 0.5940631769700256, "300–1000 Hz_10ms_log_rms_mae_db": 1.4026518331438795, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07683041973036694, "300–1000 Hz_10ms_envelope_correlation": 0.9939763325081655, "300–1000 Hz_10ms_modulation_mae_db": 0.42986000831385446, "300–1000 Hz_40ms_log_rms_mae_db": 0.8144698551254183, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03956039742141606, "300–1000 Hz_40ms_envelope_correlation": 0.9981762638719152, "300–1000 Hz_40ms_modulation_mae_db": 0.617148737007238, "1000–3000 Hz_10ms_log_rms_mae_db": 1.543906026063719, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1019116684111141, "1000–3000 Hz_10ms_envelope_correlation": 0.9938419168192341, "1000–3000 Hz_10ms_modulation_mae_db": 0.5523669526711679, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8901082916447777, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06166767656173028, "1000–3000 Hz_40ms_envelope_correlation": 0.9983367375732819, "1000–3000 Hz_40ms_modulation_mae_db": 0.9653833052042827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6375081384111116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21895275853889815, "3000–6000 Hz_10ms_envelope_correlation": 0.9671133776417582, "3000–6000 Hz_10ms_modulation_mae_db": 2.028060033129967, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9876574785759429, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1432140670971656, "3000–6000 Hz_40ms_envelope_correlation": 0.9878911674705585, "3000–6000 Hz_40ms_modulation_mae_db": 2.5413620471630187, "mr_stft_db": 5.319794571905142, "band_mae_db": 0.2493281894183926, "rms_gain_error_db": -0.05989442932235897, "output_peak": 0.65185343014128}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7781666519510176, "全帯域_10ms_rms_envelope_nrmse": 0.05068407247893678, "全帯域_10ms_envelope_correlation": 0.9965679761030617, "全帯域_10ms_modulation_mae_db": 0.5715335479910723, "全帯域_40ms_log_rms_mae_db": 0.36620028345902833, "全帯域_40ms_rms_envelope_nrmse": 0.02583948493451167, "全帯域_40ms_envelope_correlation": 0.9990854778120218, "全帯域_40ms_modulation_mae_db": 0.5360679563365455, "300–1000 Hz_10ms_log_rms_mae_db": 1.3732685397199011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07817280796692978, "300–1000 Hz_10ms_envelope_correlation": 0.9938747422403402, "300–1000 Hz_10ms_modulation_mae_db": 0.6485568943991552, "300–1000 Hz_40ms_log_rms_mae_db": 0.6972594633263236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0409219644923689, "300–1000 Hz_40ms_envelope_correlation": 0.9982152160316871, "300–1000 Hz_40ms_modulation_mae_db": 0.7530695212853279, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5975270931099392, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11461831519283017, "1000–3000 Hz_10ms_envelope_correlation": 0.9919685089311366, "1000–3000 Hz_10ms_modulation_mae_db": 0.8399383073272029, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9710071650007582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06888054515663747, "1000–3000 Hz_40ms_envelope_correlation": 0.9977179516536885, "1000–3000 Hz_40ms_modulation_mae_db": 1.266344659112589, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6075419316069126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17637783038335608, "3000–6000 Hz_10ms_envelope_correlation": 0.9770050166664621, "3000–6000 Hz_10ms_modulation_mae_db": 0.6431353392637328, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9294006750826762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0965980260375152, "3000–6000 Hz_40ms_envelope_correlation": 0.9929580436181465, "3000–6000 Hz_40ms_modulation_mae_db": 0.8610394752108421, "mr_stft_db": 5.323783167462969, "band_mae_db": 0.1992580759238812, "rms_gain_error_db": -0.08539177517060034, "output_peak": 0.6728083385277819}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7357211689612598, "全帯域_10ms_rms_envelope_nrmse": 0.04523351839691215, "全帯域_10ms_envelope_correlation": 0.9972333014674144, "全帯域_10ms_modulation_mae_db": 0.4345830298244735, "全帯域_40ms_log_rms_mae_db": 0.37781281291635976, "全帯域_40ms_rms_envelope_nrmse": 0.024700424876175946, "全帯域_40ms_envelope_correlation": 0.999093202908769, "全帯域_40ms_modulation_mae_db": 0.43872496699383606, "300–1000 Hz_10ms_log_rms_mae_db": 1.3805631085565029, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07508642831797409, "300–1000 Hz_10ms_envelope_correlation": 0.9943510328472868, "300–1000 Hz_10ms_modulation_mae_db": 0.7027868905509121, "300–1000 Hz_40ms_log_rms_mae_db": 0.7511073811855212, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04457312319529847, "300–1000 Hz_40ms_envelope_correlation": 0.997807653813499, "300–1000 Hz_40ms_modulation_mae_db": 0.7873488898334965, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5218097628773128, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11363932907608043, "1000–3000 Hz_10ms_envelope_correlation": 0.9918966544623185, "1000–3000 Hz_10ms_modulation_mae_db": 0.37167970188801847, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8970573772519321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06661841114722143, "1000–3000 Hz_40ms_envelope_correlation": 0.9976722440176748, "1000–3000 Hz_40ms_modulation_mae_db": 0.637225250669823, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6088230917826254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18674670616197764, "3000–6000 Hz_10ms_envelope_correlation": 0.9738850676631664, "3000–6000 Hz_10ms_modulation_mae_db": 0.4709322426668287, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9661906809663392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12829424601398778, "3000–6000 Hz_40ms_envelope_correlation": 0.9862783428303612, "3000–6000 Hz_40ms_modulation_mae_db": 0.3657806053041421, "mr_stft_db": 5.299070559510903, "band_mae_db": 0.1760811353332403, "rms_gain_error_db": -0.05450745226187323, "output_peak": 0.6542336821636237}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7914003682091025, "全帯域_10ms_rms_envelope_nrmse": 0.05412457120942455, "全帯域_10ms_envelope_correlation": 0.9961442119573727, "全帯域_10ms_modulation_mae_db": 0.5306226031373705, "全帯域_40ms_log_rms_mae_db": 0.3922360577190418, "全帯域_40ms_rms_envelope_nrmse": 0.031059642432094543, "全帯域_40ms_envelope_correlation": 0.9986984946303297, "全帯域_40ms_modulation_mae_db": 0.6835107976435436, "300–1000 Hz_10ms_log_rms_mae_db": 1.4219207987425655, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08325698918038092, "300–1000 Hz_10ms_envelope_correlation": 0.9930848226550527, "300–1000 Hz_10ms_modulation_mae_db": 0.7991602001516472, "300–1000 Hz_40ms_log_rms_mae_db": 0.8197089084006774, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05005377744410147, "300–1000 Hz_40ms_envelope_correlation": 0.9972990040965941, "300–1000 Hz_40ms_modulation_mae_db": 0.9602700971583661, "1000–3000 Hz_10ms_log_rms_mae_db": 1.512363090799601, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12031882010549048, "1000–3000 Hz_10ms_envelope_correlation": 0.9912929665159902, "1000–3000 Hz_10ms_modulation_mae_db": 0.9081003052230318, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8751655448980188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07962003572960302, "1000–3000 Hz_40ms_envelope_correlation": 0.9968576983412305, "1000–3000 Hz_40ms_modulation_mae_db": 1.2662505245258104, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6044369709879922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20845658516503288, "3000–6000 Hz_10ms_envelope_correlation": 0.9692095901379028, "3000–6000 Hz_10ms_modulation_mae_db": 0.5827195506018552, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9769103639645511, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1449443022463227, "3000–6000 Hz_40ms_envelope_correlation": 0.9837107260890399, "3000–6000 Hz_40ms_modulation_mae_db": 0.7081888442114841, "mr_stft_db": 5.319679277315372, "band_mae_db": 0.2623481174344274, "rms_gain_error_db": -0.10529355553736942, "output_peak": 0.6805912706303023}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.766960694991264, "全帯域_10ms_rms_envelope_nrmse": 0.04597797830143477, "全帯域_10ms_envelope_correlation": 0.9971934407856408, "全帯域_10ms_modulation_mae_db": 0.47511764368923, "全帯域_40ms_log_rms_mae_db": 0.3989407532912519, "全帯域_40ms_rms_envelope_nrmse": 0.025363004874764578, "全帯域_40ms_envelope_correlation": 0.9990990226748077, "全帯域_40ms_modulation_mae_db": 0.5264663008524708, "300–1000 Hz_10ms_log_rms_mae_db": 1.4235234505136354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07214287451875673, "300–1000 Hz_10ms_envelope_correlation": 0.9947533289998496, "300–1000 Hz_10ms_modulation_mae_db": 0.7604596618634288, "300–1000 Hz_40ms_log_rms_mae_db": 0.7735578302431253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04052856209684211, "300–1000 Hz_40ms_envelope_correlation": 0.998162341234422, "300–1000 Hz_40ms_modulation_mae_db": 1.0300307650487448, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5607662041851134, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1261209384401456, "1000–3000 Hz_10ms_envelope_correlation": 0.9901604845554041, "1000–3000 Hz_10ms_modulation_mae_db": 0.2927141799154307, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9212390574526745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08373417114989062, "1000–3000 Hz_40ms_envelope_correlation": 0.9962070330920245, "1000–3000 Hz_40ms_modulation_mae_db": 0.5168671843482963, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6326868914089498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2020548102567931, "3000–6000 Hz_10ms_envelope_correlation": 0.9694794674750449, "3000–6000 Hz_10ms_modulation_mae_db": 0.3739644545000335, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9690422573837445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14357706694461564, "3000–6000 Hz_40ms_envelope_correlation": 0.9826146395754612, "3000–6000 Hz_40ms_modulation_mae_db": 0.33932595493880574, "mr_stft_db": 5.307862616650378, "band_mae_db": 0.20733948278202222, "rms_gain_error_db": -0.06949775036055596, "output_peak": 0.6741411529843585}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.757668263096474, "全帯域_10ms_rms_envelope_nrmse": 0.04525097713401814, "全帯域_10ms_envelope_correlation": 0.9972414124214736, "全帯域_10ms_modulation_mae_db": 0.37267187562000864, "全帯域_40ms_log_rms_mae_db": 0.3596719151138258, "全帯域_40ms_rms_envelope_nrmse": 0.022832551997092037, "全帯域_40ms_envelope_correlation": 0.9992594546209529, "全帯域_40ms_modulation_mae_db": 0.3643297819426054, "300–1000 Hz_10ms_log_rms_mae_db": 1.3783563928185292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07922043107914457, "300–1000 Hz_10ms_envelope_correlation": 0.9936083794653929, "300–1000 Hz_10ms_modulation_mae_db": 0.2874794398014728, "300–1000 Hz_40ms_log_rms_mae_db": 0.7530853325451901, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04379316865537345, "300–1000 Hz_40ms_envelope_correlation": 0.9978048138405241, "300–1000 Hz_40ms_modulation_mae_db": 0.3092927690864434, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5781472772062535, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13098033562082603, "1000–3000 Hz_10ms_envelope_correlation": 0.9886927365340863, "1000–3000 Hz_10ms_modulation_mae_db": 0.2451736960444663, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9180648391433511, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07071364462305561, "1000–3000 Hz_40ms_envelope_correlation": 0.9969342425347203, "1000–3000 Hz_40ms_modulation_mae_db": 0.3939554622252405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6155234800308025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1994689474888386, "3000–6000 Hz_10ms_envelope_correlation": 0.9701539212160017, "3000–6000 Hz_10ms_modulation_mae_db": 0.537927440216838, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9677884940296245, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11485822651240207, "3000–6000 Hz_40ms_envelope_correlation": 0.9893074638851781, "3000–6000 Hz_40ms_modulation_mae_db": 0.7725392632280768, "mr_stft_db": 5.307715133862005, "band_mae_db": 0.157873761534946, "rms_gain_error_db": -0.06687243418430788, "output_peak": 0.6781335213182408}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7541011880468468, "全帯域_10ms_rms_envelope_nrmse": 0.04893292987930361, "全帯域_10ms_envelope_correlation": 0.9967491547046397, "全帯域_10ms_modulation_mae_db": 0.524489373825017, "全帯域_40ms_log_rms_mae_db": 0.38654127557329093, "全帯域_40ms_rms_envelope_nrmse": 0.024602750477647328, "全帯域_40ms_envelope_correlation": 0.9990937561026743, "全帯域_40ms_modulation_mae_db": 0.6776914333398375, "300–1000 Hz_10ms_log_rms_mae_db": 1.3704107835659052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07819983951230323, "300–1000 Hz_10ms_envelope_correlation": 0.9937842680072813, "300–1000 Hz_10ms_modulation_mae_db": 0.5950998192447677, "300–1000 Hz_40ms_log_rms_mae_db": 0.7388330996843382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045288971906810714, "300–1000 Hz_40ms_envelope_correlation": 0.9976752020419676, "300–1000 Hz_40ms_modulation_mae_db": 0.7589649518270954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5159752936192716, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1161052888133778, "1000–3000 Hz_10ms_envelope_correlation": 0.9910073425466318, "1000–3000 Hz_10ms_modulation_mae_db": 1.0816721414407986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9251490489618321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05894332143163818, "1000–3000 Hz_40ms_envelope_correlation": 0.9977214181509573, "1000–3000 Hz_40ms_modulation_mae_db": 1.3294107287688044, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6308827018311145, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1906476874625212, "3000–6000 Hz_10ms_envelope_correlation": 0.9727221378819497, "3000–6000 Hz_10ms_modulation_mae_db": 0.5864591775981394, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9384576444682419, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12096251301178544, "3000–6000 Hz_40ms_envelope_correlation": 0.9878329705966361, "3000–6000 Hz_40ms_modulation_mae_db": 0.7543408177718492, "mr_stft_db": 5.30666832759347, "band_mae_db": 0.14866409076864692, "rms_gain_error_db": -0.04720129623898249, "output_peak": 0.6762177896602592}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7456404019199606, "全帯域_10ms_rms_envelope_nrmse": 0.05096305324119439, "全帯域_10ms_envelope_correlation": 0.9964542840971647, "全帯域_10ms_modulation_mae_db": 0.32907803624169385, "全帯域_40ms_log_rms_mae_db": 0.38339217392739444, "全帯域_40ms_rms_envelope_nrmse": 0.02943212339541453, "全帯域_40ms_envelope_correlation": 0.9986636372040528, "全帯域_40ms_modulation_mae_db": 0.3096262845090307, "300–1000 Hz_10ms_log_rms_mae_db": 1.4122054869301164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07973213357580504, "300–1000 Hz_10ms_envelope_correlation": 0.993506749816318, "300–1000 Hz_10ms_modulation_mae_db": 0.6298803892592638, "300–1000 Hz_40ms_log_rms_mae_db": 0.8060893094622514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045558577457993236, "300–1000 Hz_40ms_envelope_correlation": 0.9975846636856477, "300–1000 Hz_40ms_modulation_mae_db": 0.591640510389418, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4863075919304214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11792668464820423, "1000–3000 Hz_10ms_envelope_correlation": 0.991692555368671, "1000–3000 Hz_10ms_modulation_mae_db": 0.9616645541521255, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8717934615933639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06936853998220927, "1000–3000 Hz_40ms_envelope_correlation": 0.9979882155444681, "1000–3000 Hz_40ms_modulation_mae_db": 1.3214302085499023, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6303249578932675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21098870001229056, "3000–6000 Hz_10ms_envelope_correlation": 0.9710882595827115, "3000–6000 Hz_10ms_modulation_mae_db": 0.9290535544823507, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9563167903932066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15818076508651718, "3000–6000 Hz_40ms_envelope_correlation": 0.9832488310118959, "3000–6000 Hz_40ms_modulation_mae_db": 0.925588016251452, "mr_stft_db": 5.321570746416017, "band_mae_db": 0.2891579867557916, "rms_gain_error_db": -0.03746253197570409, "output_peak": 0.6680140729616847}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.359306284916451, "spectral_convergence": 0.14001671687507086, "spectral_convergence_status": "ok", "amplitude_floor": 1.1745223847485646e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1505540741210485, "spectral_convergence": 0.13854414770598344, "spectral_convergence_status": "ok", "amplitude_floor": 9.481908348017771e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.449523356677925, "spectral_convergence": 0.13433694691714304, "spectral_convergence_status": "ok", "amplitude_floor": 8.395402116778918e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004068927413639468, "output_power": 0.0040315733146141725, "delta_db": -0.040053805603551575, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018176508169902323, "output_power": 0.0018051655332026683, "delta_db": -0.029934232357754012, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007805177117872849, "output_power": 0.0007693551111020434, "delta_db": -0.06255919371001763, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000310938544728363, "output_power": 0.0002863324314633802, "delta_db": -0.3580402041938812, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.08647023559392e-05, "output_power": 7.596113107604639e-05, "delta_db": -0.2716756950821798, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.353447039170603e-05, "output_power": 3.676737841721198e-05, "delta_db": -0.7337060055629713, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2493281894183926, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05989442932235897, "pitch": null, "output_rms_dbfs": -21.5458225703525, "output_peak": 0.65185343014128, "reference_rms_dbfs": -21.485928141030136, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.319794571905142}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019068326661124743, "log_rms_mae_db": 0.7476324890830811, "rms_envelope_nrmse": 0.04903644164635436, "envelope_correlation": 0.9967382762279056, "rms_gain_error_db": -0.0584365870061081, "reference_mean_power": 0.007104543911662494, "output_mean_power": 0.007009588836594204}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.7734542191089184e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001358301262786015, "output_power": 0.0013721892332425032, "reference_active": true, "delta_db": 0.04417902904266928}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002660490997406596, "output_power": 0.0002428828889376493, "reference_active": true, "delta_db": -0.3956487379060461}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.883599774152297e-05, "output_power": 5.837913716279688e-05, "reference_active": true, "delta_db": -1.3046689651163377}], "active_band_count": 3, "active_band_mae_db": 0.581498910688351}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001870958079240233, "log_rms_mae_db": 0.3639538702298398, "rms_envelope_nrmse": 0.025394014623708962, "envelope_correlation": 0.9990457070600017, "rms_gain_error_db": -0.05743435561911896, "reference_mean_power": 0.007117638239873263, "output_mean_power": 0.007024128941267339}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4206220628648622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012103399895685932, "output_power": 0.001230904946187669, "reference_active": true, "delta_db": 0.07317134238820062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012555497766250794, "output_power": 0.00011493896360440501, "reference_active": true, "delta_db": -0.3836665830135141}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0918852450489625e-06, "output_power": 4.48969204652638e-06, "reference_active": true, "delta_db": -1.3253516055083623}], "active_band_count": 3, "active_band_mae_db": 0.5940631769700256}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013521834879092463, "log_rms_mae_db": 1.4026518331438795, "rms_envelope_nrmse": 0.07683041973036694, "envelope_correlation": 0.9939763325081655, "rms_gain_error_db": -0.026004448139198943, "reference_mean_power": 0.0018937311393695897, "output_mean_power": 0.001882425839736166}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.102346197011796e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006150154421416123, "output_power": 0.0006181737628163917, "reference_active": true, "delta_db": 0.022245479195017903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012427322589552618, "output_power": 0.00011478133189303485, "reference_active": true, "delta_db": -0.34506312055932215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6589172324725154e-05, "output_power": 2.1501941511641006e-05, "reference_active": true, "delta_db": -0.9222714251872233}], "active_band_count": 3, "active_band_mae_db": 0.42986000831385446}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012428224153258587, "log_rms_mae_db": 0.8144698551254183, "rms_envelope_nrmse": 0.03956039742141606, "envelope_correlation": 0.9981762638719152, "rms_gain_error_db": -0.02443553088202433, "reference_mean_power": 0.0018973976433641784, "output_mean_power": 0.0018867519336252647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.796182335233462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005643605996811222, "output_power": 0.000565238182752062, "reference_active": true, "delta_db": 0.0067480519229922604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.691592416494353e-05, "output_power": 5.278233198888309e-05, "reference_active": true, "delta_db": -0.3274521806709074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2464097368217303e-06, "output_power": 1.5840335472985169e-06, "reference_active": true, "delta_db": -1.5172459784278143}], "active_band_count": 3, "active_band_mae_db": 0.617148737007238}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012748188769016831, "log_rms_mae_db": 1.543906026063719, "rms_envelope_nrmse": 0.1019116684111141, "envelope_correlation": 0.9938419168192341, "rms_gain_error_db": -0.3404584181844688, "reference_mean_power": 0.00032033050063867626, "output_mean_power": 0.00029617776439581537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0949985705721996e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.663152856065707e-05, "output_power": 6.313776495661444e-05, "reference_active": true, "delta_db": -0.23390572281440353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2062707324852824e-05, "output_power": 1.1131582080424165e-05, "reference_active": true, "delta_db": -0.3488789775369943}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2084052775211794e-06, "output_power": 2.505287700490502e-06, "reference_active": true, "delta_db": -1.0743161576621056}], "active_band_count": 3, "active_band_mae_db": 0.5523669526711679}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001243805164613915, "log_rms_mae_db": 0.8901082916447777, "rms_envelope_nrmse": 0.06166767656173028, "envelope_correlation": 0.9983367375732819, "rms_gain_error_db": -0.3400310655325573, "reference_mean_power": 0.0003214084172642619, "output_mean_power": 0.0002972036505966445}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.993543951674123e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.454820101360055e-05, "output_power": 6.074164169302095e-05, "reference_active": true, "delta_db": -0.2639761717242894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.405216922862835e-06, "output_power": 5.0229499751076184e-06, "reference_active": true, "delta_db": -0.3185427487420795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4345993241652877e-07, "output_power": 2.0161042290368517e-07, "reference_active": true, "delta_db": -2.3136309951464793}], "active_band_count": 3, "active_band_mae_db": 0.9653833052042827}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.885927766129993e-05, "log_rms_mae_db": 1.6375081384111116, "rms_envelope_nrmse": 0.21895275853889815, "envelope_correlation": 0.9671133776417582, "rms_gain_error_db": -0.7523337925447033, "reference_mean_power": 5.543232992418922e-05, "output_mean_power": 4.6615436358725635e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.209399643069482e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1581941196726732e-05, "output_power": 1.8272815744345412e-05, "reference_active": true, "delta_db": -0.7228502990861287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7792321883200776e-05, "output_power": 1.3009041662488797e-05, "reference_active": true, "delta_db": -1.3598732229708934}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1497984756130906e-05, "output_power": 4.575895214530805e-06, "reference_active": true, "delta_db": -4.001456577332879}], "active_band_count": 3, "active_band_mae_db": 2.028060033129967}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.78543565194846e-05, "log_rms_mae_db": 0.9876574785759429, "rms_envelope_nrmse": 0.1432140670971656, "envelope_correlation": 0.9878911674705585, "rms_gain_error_db": -0.7527490740489367, "reference_mean_power": 5.5701179342093606e-05, "output_mean_power": 4.683704455576365e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.455779259683724e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.849693387286164e-05, "output_power": 2.1719675477412533e-05, "reference_active": true, "delta_db": -1.179448026212805}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0110098860775756e-05, "output_power": 5.934204239326798e-06, "reference_active": true, "delta_db": -2.313929128404814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1498909214748545e-06, "output_power": 4.442073563861469e-07, "reference_active": true, "delta_db": -4.130708986871437}], "active_band_count": 3, "active_band_mae_db": 2.5413620471630187}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.934270327644802, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fd88fe736d9a538079ed4d4ca56c8ffe2d17c416fc59431dd84726b9650fc184", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9913191457166841, "standard_deviation": 0.00149271685134967, "minimum": 0.9886927365340863, "maximum": 0.9938419168192341, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.539600292473954, "standard_deviation": 0.037348819114376014, "minimum": 1.4863075919304214, "maximum": 1.5975270931099392, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6566637298327802, "standard_deviation": 0.3304422602847922, "minimum": 0.2451736960444663, "maximum": 1.0816721414407986, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11770267253850861, "standard_deviation": 0.008717711148509406, "minimum": 0.1019116684111141, "maximum": 0.13098033562082603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974294426135057, "standard_deviation": 0.0006999235370618207, "minimum": 0.9962070330920245, "maximum": 0.9983367375732819, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9086980982433386, "standard_deviation": 0.03243793349811324, "minimum": 0.8717934615933639, "maximum": 0.9710071650007582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9621084154255936, "standard_deviation": 0.3920063014486851, "minimum": 0.3939554622252405, "maximum": 1.3294107287688044, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06994329322274824, "standard_deviation": 0.008335988474305642, "minimum": 0.05894332143163818, "maximum": 0.08373417114989062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9713321047831247, "standard_deviation": 0.0031125164331450785, "minimum": 0.9671133776417582, "maximum": 0.9770050166664621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.620966020494097, "standard_deviation": 0.01324191593052355, "minimum": 1.6044369709879922, "maximum": 1.6375081384111116, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7690314740574682, "standard_deviation": 0.533586887816909, "minimum": 0.3739644545000335, "maximum": 2.028060033129967, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19921175318371354, "standard_deviation": 0.014000655388814747, "minimum": 0.17637783038335608, "maximum": 0.21895275853889815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9867302731346597, "standard_deviation": 0.0035126595321113355, "minimum": 0.9826146395754612, "maximum": 0.9929580436181465, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9614705481080409, "standard_deviation": 0.01937709187091956, "minimum": 0.9294006750826762, "maximum": 0.9876574785759429, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9085206280099589, "standard_deviation": 0.6938156746640243, "minimum": 0.33932595493880574, "maximum": 2.5413620471630187, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13132865161878896, "standard_deviation": 0.019970923228634807, "minimum": 0.0965980260375152, "maximum": 0.15818076508651718, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938674570674608, "standard_deviation": 0.0005136966195973447, "minimum": 0.9930848226550527, "maximum": 0.9947533289998496, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3953625492488793, "standard_deviation": 0.022213779040282627, "minimum": 1.3704107835659052, "maximum": 1.4235234505136354, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6066604129480628, "standard_deviation": 0.1713449666687304, "minimum": 0.2874794398014728, "maximum": 0.7991602001516472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07783024048520767, "standard_deviation": 0.0032977841678417615, "minimum": 0.07214287451875673, "maximum": 0.08325698918038092, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997840644827032, "standard_deviation": 0.0003262507616815035, "minimum": 0.9972990040965941, "maximum": 0.9982152160316871, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7692638974966057, "standard_deviation": 0.04253553740988908, "minimum": 0.6972594633263236, "maximum": 0.8197089084006774, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7259707802045163, "standard_deviation": 0.22565659662508225, "minimum": 0.3092927690864434, "maximum": 1.0300307650487448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04378481783377555, "standard_deviation": 0.0034220902312169367, "minimum": 0.03956039742141606, "maximum": 0.05005377744410147, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21125635499391854, "standard_deviation": 0.05111972849369041, "minimum": 0.14866409076864692, "maximum": 0.2891579867557916, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.313268050089532, "standard_deviation": 0.009010949262411925, "minimum": 5.299070559510903, "maximum": 5.323783167462969, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6694991572984414, "standard_deviation": 0.010835378158390044, "minimum": 0.65185343014128, "maximum": 0.6805912706303023, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06576515313146905, "standard_deviation": 0.021612541388072056, "minimum": -0.10529355553736942, "maximum": -0.03746253197570409, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967902572205841, "standard_deviation": 0.0004046937050717202, "minimum": 0.9961442119573727, "maximum": 0.9972414124214736, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7596614032823757, "standard_deviation": 0.018331832029850427, "minimum": 0.7357211689612598, "maximum": 0.7914003682091025, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4774493776271521, "standard_deviation": 0.09223698531906895, "minimum": 0.32907803624169385, "maximum": 0.581498910688351, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.048775442785947344, "standard_deviation": 0.00316063959035434, "minimum": 0.04523351839691215, "maximum": 0.05412457120942455, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990048441267012, "standard_deviation": 0.00020971152467243114, "minimum": 0.9986636372040528, "maximum": 0.9992594546209529, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37859364277875407, "standard_deviation": 0.014206994899110939, "minimum": 0.3596719151138258, "maximum": 0.3989407532912519, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5163100873234869, "standard_deviation": 0.13751609768128664, "minimum": 0.3096262845090307, "maximum": 0.6835107976435436, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.026152999701426198, "standard_deviation": 0.002716743218485667, "minimum": 0.022832551997092037, "maximum": 0.031059642432094543, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5696-0010", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.090153932571411, "predicted_mos": 3.090153932571411, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ddc438d929100fd7b5cf15542783d9a9027a03078f00d12654f25421557ec287", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.333842699881643, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7461338154697337, "全帯域_10ms_rms_envelope_nrmse": 0.05051545739344884, "全帯域_10ms_envelope_correlation": 0.996512423192733, "全帯域_10ms_modulation_mae_db": 0.5418683181466409, "全帯域_40ms_log_rms_mae_db": 0.3612960967290375, "全帯域_40ms_rms_envelope_nrmse": 0.024477555326449643, "全帯域_40ms_envelope_correlation": 0.9990999040953192, "全帯域_40ms_modulation_mae_db": 0.6150531917230094, "300–1000 Hz_10ms_log_rms_mae_db": 1.44048438617934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07995083033713503, "300–1000 Hz_10ms_envelope_correlation": 0.9935201550173742, "300–1000 Hz_10ms_modulation_mae_db": 0.9084956116587124, "300–1000 Hz_40ms_log_rms_mae_db": 0.7164706802373805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03900265351872532, "300–1000 Hz_40ms_envelope_correlation": 0.9982865097042127, "300–1000 Hz_40ms_modulation_mae_db": 1.0730108885714813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7904695123572798, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1392832520925139, "1000–3000 Hz_10ms_envelope_correlation": 0.9870592141189853, "1000–3000 Hz_10ms_modulation_mae_db": 0.6655125952950275, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9944469148362639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08434745960464925, "1000–3000 Hz_40ms_envelope_correlation": 0.995266047238917, "1000–3000 Hz_40ms_modulation_mae_db": 0.9071040748172604, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9206325068827688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20514053954941255, "3000–6000 Hz_10ms_envelope_correlation": 0.9700091306979541, "3000–6000 Hz_10ms_modulation_mae_db": 1.5494215451410567, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0551986993074518, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1281834812441995, "3000–6000 Hz_40ms_envelope_correlation": 0.9888060653187053, "3000–6000 Hz_40ms_modulation_mae_db": 2.025561313844627, "mr_stft_db": 5.456089226750502, "band_mae_db": 0.2005662686761752, "rms_gain_error_db": -0.04955083561231998, "output_peak": 0.700664360259545}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7641597868162662, "全帯域_10ms_rms_envelope_nrmse": 0.05315232361039563, "全帯域_10ms_envelope_correlation": 0.9961629469588494, "全帯域_10ms_modulation_mae_db": 0.20483807396522724, "全帯域_40ms_log_rms_mae_db": 0.40852488606904713, "全帯域_40ms_rms_envelope_nrmse": 0.029349936458013506, "全帯域_40ms_envelope_correlation": 0.9987040808610184, "全帯域_40ms_modulation_mae_db": 0.20880720970980157, "300–1000 Hz_10ms_log_rms_mae_db": 1.5818629203343633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08082403074110144, "300–1000 Hz_10ms_envelope_correlation": 0.9933711362641124, "300–1000 Hz_10ms_modulation_mae_db": 0.7132964995158909, "300–1000 Hz_40ms_log_rms_mae_db": 0.8263543359748697, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04495713984999482, "300–1000 Hz_40ms_envelope_correlation": 0.997701760527664, "300–1000 Hz_40ms_modulation_mae_db": 0.7887863572245609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7076725424889168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1180956667497749, "1000–3000 Hz_10ms_envelope_correlation": 0.9918319267598205, "1000–3000 Hz_10ms_modulation_mae_db": 0.6238515452295449, "1000–3000 Hz_40ms_log_rms_mae_db": 0.952862161155456, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07199725108635736, "1000–3000 Hz_40ms_envelope_correlation": 0.9978783994843565, "1000–3000 Hz_40ms_modulation_mae_db": 0.7730419032109858, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8617766083503087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16757260885661057, "3000–6000 Hz_10ms_envelope_correlation": 0.979029150614041, "3000–6000 Hz_10ms_modulation_mae_db": 0.5870963014602216, "3000–6000 Hz_40ms_log_rms_mae_db": 1.012996232724691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10191239215408733, "3000–6000 Hz_40ms_envelope_correlation": 0.9911692239972052, "3000–6000 Hz_40ms_modulation_mae_db": 0.42654855481549503, "mr_stft_db": 5.455163439896839, "band_mae_db": 0.20046516846629334, "rms_gain_error_db": -0.057718164883817015, "output_peak": 0.6801046106128623}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7618792717519388, "全帯域_10ms_rms_envelope_nrmse": 0.051447819880843036, "全帯域_10ms_envelope_correlation": 0.9964485797178794, "全帯域_10ms_modulation_mae_db": 0.6092151630831552, "全帯域_40ms_log_rms_mae_db": 0.40180086525432807, "全帯域_40ms_rms_envelope_nrmse": 0.02773324769940306, "全帯域_40ms_envelope_correlation": 0.998915025929325, "全帯域_40ms_modulation_mae_db": 0.7193708859514251, "300–1000 Hz_10ms_log_rms_mae_db": 1.412201290234869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0807494992384864, "300–1000 Hz_10ms_envelope_correlation": 0.9934090244999074, "300–1000 Hz_10ms_modulation_mae_db": 0.9136034846104811, "300–1000 Hz_40ms_log_rms_mae_db": 0.7295971159523725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043060407378790674, "300–1000 Hz_40ms_envelope_correlation": 0.9979261513030444, "300–1000 Hz_40ms_modulation_mae_db": 0.9474002867546186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7316675568941393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12332172828403004, "1000–3000 Hz_10ms_envelope_correlation": 0.9904827171186589, "1000–3000 Hz_10ms_modulation_mae_db": 1.0561294439267046, "1000–3000 Hz_40ms_log_rms_mae_db": 1.006053229706257, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07826947836928104, "1000–3000 Hz_40ms_envelope_correlation": 0.9966420585036924, "1000–3000 Hz_40ms_modulation_mae_db": 1.5041391515882754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.915937785578424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2231956340296111, "3000–6000 Hz_10ms_envelope_correlation": 0.9625823370132327, "3000–6000 Hz_10ms_modulation_mae_db": 0.7778020082898486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0850832016242802, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15417629341093675, "3000–6000 Hz_40ms_envelope_correlation": 0.9803427972813431, "3000–6000 Hz_40ms_modulation_mae_db": 1.0998500958104869, "mr_stft_db": 5.481156120168742, "band_mae_db": 0.22378330428016527, "rms_gain_error_db": -0.08251448055833747, "output_peak": 0.6833990857792048}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7265099080860203, "全帯域_10ms_rms_envelope_nrmse": 0.04513560268458888, "全帯域_10ms_envelope_correlation": 0.9972350215636608, "全帯域_10ms_modulation_mae_db": 0.5632159331926676, "全帯域_40ms_log_rms_mae_db": 0.3751571140861873, "全帯域_40ms_rms_envelope_nrmse": 0.023190069370942194, "全帯域_40ms_envelope_correlation": 0.9991992441255593, "全帯域_40ms_modulation_mae_db": 0.6410777956505563, "300–1000 Hz_10ms_log_rms_mae_db": 1.4662711672205446, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07431182417187324, "300–1000 Hz_10ms_envelope_correlation": 0.9945115337657413, "300–1000 Hz_10ms_modulation_mae_db": 0.7737060057565945, "300–1000 Hz_40ms_log_rms_mae_db": 0.8221443312778435, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04123863059225844, "300–1000 Hz_40ms_envelope_correlation": 0.9981747761081406, "300–1000 Hz_40ms_modulation_mae_db": 1.050399176913323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6724186679767064, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10849893976777245, "1000–3000 Hz_10ms_envelope_correlation": 0.9920883423652431, "1000–3000 Hz_10ms_modulation_mae_db": 0.4481058471674129, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8835792256482569, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056681903450580204, "1000–3000 Hz_40ms_envelope_correlation": 0.9977472586777286, "1000–3000 Hz_40ms_modulation_mae_db": 0.5753908698442816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.948903254930827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20300873619462742, "3000–6000 Hz_10ms_envelope_correlation": 0.9694591289547617, "3000–6000 Hz_10ms_modulation_mae_db": 0.8213414662827333, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1310986370953926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1173926327200806, "3000–6000 Hz_40ms_envelope_correlation": 0.9894161828007714, "3000–6000 Hz_40ms_modulation_mae_db": 1.1575264931391736, "mr_stft_db": 5.448471868487478, "band_mae_db": 0.16550079209453372, "rms_gain_error_db": -0.04897580530021796, "output_peak": 0.678229030494646}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7230659620251769, "全帯域_10ms_rms_envelope_nrmse": 0.048125423065612044, "全帯域_10ms_envelope_correlation": 0.9969042272328024, "全帯域_10ms_modulation_mae_db": 0.5328720060738193, "全帯域_40ms_log_rms_mae_db": 0.3588458622012025, "全帯域_40ms_rms_envelope_nrmse": 0.02507290406095715, "全帯域_40ms_envelope_correlation": 0.999119498969636, "全帯域_40ms_modulation_mae_db": 0.5214336574995028, "300–1000 Hz_10ms_log_rms_mae_db": 1.3988675023609376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07542856637244795, "300–1000 Hz_10ms_envelope_correlation": 0.9942117879183071, "300–1000 Hz_10ms_modulation_mae_db": 0.7238723854113948, "300–1000 Hz_40ms_log_rms_mae_db": 0.722051068475376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04051341173667469, "300–1000 Hz_40ms_envelope_correlation": 0.998095492341313, "300–1000 Hz_40ms_modulation_mae_db": 0.8018541493851453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7174270201779043, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11737648914544466, "1000–3000 Hz_10ms_envelope_correlation": 0.9912735260470019, "1000–3000 Hz_10ms_modulation_mae_db": 0.9920189806375102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9678680466681808, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07195963941101755, "1000–3000 Hz_40ms_envelope_correlation": 0.997077846275561, "1000–3000 Hz_40ms_modulation_mae_db": 1.1951486027538207, "3000–6000 Hz_10ms_log_rms_mae_db": 1.910899768763466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18222399812021378, "3000–6000 Hz_10ms_envelope_correlation": 0.9751537134745936, "3000–6000 Hz_10ms_modulation_mae_db": 0.4753258852221731, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0843874301876215, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10581642962497714, "3000–6000 Hz_40ms_envelope_correlation": 0.9904761047090116, "3000–6000 Hz_40ms_modulation_mae_db": 0.42179243121104326, "mr_stft_db": 5.4637452454899025, "band_mae_db": 0.15339230929468337, "rms_gain_error_db": -0.0742067465802621, "output_peak": 0.6569928981205709}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7422243269266813, "全帯域_10ms_rms_envelope_nrmse": 0.04959750594534769, "全帯域_10ms_envelope_correlation": 0.9966813731104455, "全帯域_10ms_modulation_mae_db": 0.4173193138489548, "全帯域_40ms_log_rms_mae_db": 0.3959282342882793, "全帯域_40ms_rms_envelope_nrmse": 0.02638537234632251, "全帯域_40ms_envelope_correlation": 0.9989801977667141, "全帯域_40ms_modulation_mae_db": 0.4781088415527175, "300–1000 Hz_10ms_log_rms_mae_db": 1.4812769572519255, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07678829351183077, "300–1000 Hz_10ms_envelope_correlation": 0.9941407721475232, "300–1000 Hz_10ms_modulation_mae_db": 0.8449779911374545, "300–1000 Hz_40ms_log_rms_mae_db": 0.8601527045170019, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045110423991544316, "300–1000 Hz_40ms_envelope_correlation": 0.9978144880153053, "300–1000 Hz_40ms_modulation_mae_db": 0.842499560343307, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7341110790838197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12673545011068243, "1000–3000 Hz_10ms_envelope_correlation": 0.9898086418188032, "1000–3000 Hz_10ms_modulation_mae_db": 1.0985572592370652, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0071806224814948, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07779585880934727, "1000–3000 Hz_40ms_envelope_correlation": 0.9966059012784276, "1000–3000 Hz_40ms_modulation_mae_db": 0.5632864410033491, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9059729875804952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22416246474670012, "3000–6000 Hz_10ms_envelope_correlation": 0.9620949556486268, "3000–6000 Hz_10ms_modulation_mae_db": 0.9806677230028685, "3000–6000 Hz_40ms_log_rms_mae_db": 1.050507853882176, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12875969622973396, "3000–6000 Hz_40ms_envelope_correlation": 0.9863763543773642, "3000–6000 Hz_40ms_modulation_mae_db": 1.318259010771063, "mr_stft_db": 5.463472383460903, "band_mae_db": 0.21463852592703847, "rms_gain_error_db": -0.061291235703681314, "output_peak": 0.7001405055768726}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7704915443916417, "全帯域_10ms_rms_envelope_nrmse": 0.05204207854781294, "全帯域_10ms_envelope_correlation": 0.9962999780890053, "全帯域_10ms_modulation_mae_db": 0.4252059286052572, "全帯域_40ms_log_rms_mae_db": 0.36854433958495286, "全帯域_40ms_rms_envelope_nrmse": 0.024652494889074807, "全帯域_40ms_envelope_correlation": 0.9990859633699543, "全帯域_40ms_modulation_mae_db": 0.44748161323701985, "300–1000 Hz_10ms_log_rms_mae_db": 1.4615084443957744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07946077915757423, "300–1000 Hz_10ms_envelope_correlation": 0.9936092472280406, "300–1000 Hz_10ms_modulation_mae_db": 0.6402025292358493, "300–1000 Hz_40ms_log_rms_mae_db": 0.7881786977025292, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04307360994419999, "300–1000 Hz_40ms_envelope_correlation": 0.9978738041966981, "300–1000 Hz_40ms_modulation_mae_db": 0.5470589520141752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7092720365146334, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.117767575558651, "1000–3000 Hz_10ms_envelope_correlation": 0.9914341612928775, "1000–3000 Hz_10ms_modulation_mae_db": 0.1598383089239152, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9686195489178814, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06805921306297615, "1000–3000 Hz_40ms_envelope_correlation": 0.9977423245673867, "1000–3000 Hz_40ms_modulation_mae_db": 0.29644760377041085, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8673729952205789, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18225170874536215, "3000–6000 Hz_10ms_envelope_correlation": 0.9765019197404115, "3000–6000 Hz_10ms_modulation_mae_db": 0.932719606566386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0389791215006623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1102023349981173, "3000–6000 Hz_40ms_envelope_correlation": 0.9920198217932712, "3000–6000 Hz_40ms_modulation_mae_db": 1.203088789877053, "mr_stft_db": 5.450760607130692, "band_mae_db": 0.1956099914631225, "rms_gain_error_db": -0.05079697481622626, "output_peak": 0.6926709374125314}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7164230348876626, "全帯域_10ms_rms_envelope_nrmse": 0.04411334640613438, "全帯域_10ms_envelope_correlation": 0.9974769773071693, "全帯域_10ms_modulation_mae_db": 0.5571891515065951, "全帯域_40ms_log_rms_mae_db": 0.34963542625450944, "全帯域_40ms_rms_envelope_nrmse": 0.02393404755733118, "全帯域_40ms_envelope_correlation": 0.9992515824255149, "全帯域_40ms_modulation_mae_db": 0.680523851753383, "300–1000 Hz_10ms_log_rms_mae_db": 1.4469734103751541, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07552169297013965, "300–1000 Hz_10ms_envelope_correlation": 0.9943413118517993, "300–1000 Hz_10ms_modulation_mae_db": 0.9548871148805471, "300–1000 Hz_40ms_log_rms_mae_db": 0.7356380219095722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040356931294369035, "300–1000 Hz_40ms_envelope_correlation": 0.9982700911025256, "300–1000 Hz_40ms_modulation_mae_db": 1.064942324572834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7385320478563133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11533914224892812, "1000–3000 Hz_10ms_envelope_correlation": 0.9918483741151707, "1000–3000 Hz_10ms_modulation_mae_db": 0.5451298214301962, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9400863750883686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07199822278073434, "1000–3000 Hz_40ms_envelope_correlation": 0.9973663632632215, "1000–3000 Hz_40ms_modulation_mae_db": 0.7857804008263368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9063893575694517, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21244974525897817, "3000–6000 Hz_10ms_envelope_correlation": 0.9669292083496587, "3000–6000 Hz_10ms_modulation_mae_db": 1.0481443661688612, "3000–6000 Hz_40ms_log_rms_mae_db": 1.101011033677686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13354033287778955, "3000–6000 Hz_40ms_envelope_correlation": 0.9866600970188986, "3000–6000 Hz_40ms_modulation_mae_db": 1.426826817695578, "mr_stft_db": 5.450638302158982, "band_mae_db": 0.21742283207043303, "rms_gain_error_db": -0.08460011005295176, "output_peak": 0.6789079644032302}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.571279791085918, "spectral_convergence": 0.14507158638387552, "spectral_convergence_status": "ok", "amplitude_floor": 1.1745223847485646e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.308229939874916, "spectral_convergence": 0.14170745649644753, "spectral_convergence_status": "ok", "amplitude_floor": 9.481908348017771e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.488757949290674, "spectral_convergence": 0.13692676817944635, "spectral_convergence_status": "ok", "amplitude_floor": 8.395402116778918e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004068927413639468, "output_power": 0.004046411624191875, "delta_db": -0.024098828546604296, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018176508169902323, "output_power": 0.0018073059743346324, "delta_db": -0.02478771768327129, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007805177117872849, "output_power": 0.0007531922246988053, "delta_db": -0.154769345866809, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000310938544728363, "output_power": 0.0003000036052270723, "delta_db": -0.15548087833558688, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.08647023559392e-05, "output_power": 7.758909656607793e-05, "delta_db": -0.17958297302576892, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.353447039170603e-05, "output_power": 3.7356439707210075e-05, "delta_db": -0.6646778685990106, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2005662686761752, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04955083561231998, "pitch": null, "output_rms_dbfs": -21.53547897664246, "output_peak": 0.700664360259545, "reference_rms_dbfs": -21.485928141030136, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.456089226750502}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019068326661124743, "log_rms_mae_db": 0.7461338154697337, "rms_envelope_nrmse": 0.05051545739344884, "envelope_correlation": 0.996512423192733, "rms_gain_error_db": -0.048513254681385114, "reference_mean_power": 0.007104543911662494, "output_mean_power": 0.007025623580560728}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.7734542191089184e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001358301262786015, "output_power": 0.0013241716087227624, "reference_active": true, "delta_db": -0.11051832416496472}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002660490997406596, "output_power": 0.000254116434164037, "reference_active": true, "delta_db": -0.19929041134576386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.883599774152297e-05, "output_power": 5.8229752809574594e-05, "reference_active": true, "delta_db": -1.315796218929194}], "active_band_count": 3, "active_band_mae_db": 0.5418683181466409}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001870958079240233, "log_rms_mae_db": 0.3612960967290375, "rms_envelope_nrmse": 0.024477555326449643, "envelope_correlation": 0.9990999040953192, "rms_gain_error_db": -0.04861158636136098, "reference_mean_power": 0.007117638239873263, "output_mean_power": 0.007038413087854484}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4206220628648622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012103399895685932, "output_power": 0.0011903223999947552, "reference_active": true, "delta_db": -0.07242776178031396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012555497766250794, "output_power": 0.00012099742073662431, "reference_active": true, "delta_db": -0.160578224191511}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0918852450489625e-06, "output_power": 4.2027770772410075e-06, "reference_active": true, "delta_db": -1.6121535891972032}], "active_band_count": 3, "active_band_mae_db": 0.6150531917230094}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013521834879092463, "log_rms_mae_db": 1.44048438617934, "rms_envelope_nrmse": 0.07995083033713503, "envelope_correlation": 0.9935201550173742, "rms_gain_error_db": -0.06478522529763453, "reference_mean_power": 0.0018937311393695897, "output_mean_power": 0.001865691349925278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.102346197011796e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006150154421416123, "output_power": 0.000599399638628897, "reference_active": true, "delta_db": -0.11169543635357115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012427322589552618, "output_power": 0.00011066023173774747, "reference_active": true, "delta_db": -0.503859965860352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6589172324725154e-05, "output_power": 1.635730212882277e-05, "reference_active": true, "delta_db": -2.109931432762214}], "active_band_count": 3, "active_band_mae_db": 0.9084956116587124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012428224153258587, "log_rms_mae_db": 0.7164706802373805, "rms_envelope_nrmse": 0.03900265351872532, "envelope_correlation": 0.9982865097042127, "rms_gain_error_db": -0.0662656889104909, "reference_mean_power": 0.0018973976433641784, "output_mean_power": 0.0018686664482585728}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.796182335233462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005643605996811222, "output_power": 0.0005491277818726537, "reference_active": true, "delta_db": -0.11883269860839737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.691592416494353e-05, "output_power": 4.993275853087897e-05, "reference_active": true, "delta_db": -0.5684823276523827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2464097368217303e-06, "output_power": 1.254056804242444e-06, "reference_active": true, "delta_db": -2.531717639453664}], "active_band_count": 3, "active_band_mae_db": 1.0730108885714813}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012748188769016831, "log_rms_mae_db": 1.7904695123572798, "rms_envelope_nrmse": 0.1392832520925139, "envelope_correlation": 0.9870592141189853, "rms_gain_error_db": -0.2054917640755487, "reference_mean_power": 0.00032033050063867626, "output_mean_power": 0.0003055266625830604}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0949985705721996e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.663152856065707e-05, "output_power": 6.390633612400886e-05, "reference_active": true, "delta_db": -0.18135857010351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2062707324852824e-05, "output_power": 1.1236030133752858e-05, "reference_active": true, "delta_db": -0.3083189548750871}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2084052775211794e-06, "output_power": 2.2677918800711964e-06, "reference_active": true, "delta_db": -1.5068602609064854}], "active_band_count": 3, "active_band_mae_db": 0.6655125952950275}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001243805164613915, "log_rms_mae_db": 0.9944469148362639, "rms_envelope_nrmse": 0.08434745960464925, "envelope_correlation": 0.995266047238917, "rms_gain_error_db": -0.2067576498192879, "reference_mean_power": 0.0003214084172642619, "output_mean_power": 0.00030646542222915584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.993543951674123e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.454820101360055e-05, "output_power": 6.19761067376702e-05, "reference_active": true, "delta_db": -0.17659851877035893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.405216922862835e-06, "output_power": 5.308963195882891e-06, "reference_active": true, "delta_db": -0.07803413174041773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4345993241652877e-07, "output_power": 1.9462925610878467e-07, "reference_active": true, "delta_db": -2.4666795739410046}], "active_band_count": 3, "active_band_mae_db": 0.9071040748172604}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.885927766129993e-05, "log_rms_mae_db": 1.9206325068827688, "rms_envelope_nrmse": 0.20514053954941255, "envelope_correlation": 0.9700091306979541, "rms_gain_error_db": -0.5330375418918334, "reference_mean_power": 5.543232992418922e-05, "output_mean_power": 4.902971615242145e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.209399643069482e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1581941196726732e-05, "output_power": 1.9808656830831066e-05, "reference_active": true, "delta_db": -0.37235476649972793}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7792321883200776e-05, "output_power": 1.3321808580720417e-05, "reference_active": true, "delta_db": -1.2566943780423472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1497984756130906e-05, "output_power": 5.737202505974757e-06, "reference_active": true, "delta_db": -3.019215490881095}], "active_band_count": 3, "active_band_mae_db": 1.5494215451410567}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.78543565194846e-05, "log_rms_mae_db": 1.0551986993074518, "rms_envelope_nrmse": 0.1281834812441995, "envelope_correlation": 0.9888060653187053, "rms_gain_error_db": -0.5334808943841727, "reference_mean_power": 5.5701179342093606e-05, "output_mean_power": 4.9262483340162365e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.455779259683724e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.849693387286164e-05, "output_power": 2.3644141343378878e-05, "reference_active": true, "delta_db": -0.8107458774125029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0110098860775756e-05, "output_power": 6.51085338079184e-06, "reference_active": true, "delta_db": -1.9111748683207266}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1498909214748545e-06, "output_power": 5.311049212067213e-07, "reference_active": true, "delta_db": -3.3547631958006505}], "active_band_count": 3, "active_band_mae_db": 2.025561313844627}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9241571485810357, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "082053dee07fe162aab9c85cd7739672905a5ca6490dafe79c2274d2a9441668", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9907283629545701, "standard_deviation": 0.0016684491594309216, "minimum": 0.9870592141189853, "maximum": 0.9920883423652431, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.725196307918714, "standard_deviation": 0.03373486807873442, "minimum": 1.6724186679767064, "maximum": 1.7904695123572798, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.698642975230922, "standard_deviation": 0.32888002151018153, "minimum": 0.1598383089239152, "maximum": 1.0985572592370652, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12080228049472469, "standard_deviation": 0.009200820933648116, "minimum": 0.10849893976777245, "maximum": 0.1392832520925139, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970407749111614, "standard_deviation": 0.0008699783560452934, "minimum": 0.995266047238917, "maximum": 0.9978783994843565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.96508701556277, "standard_deviation": 0.0409709269083418, "minimum": 0.8835792256482569, "maximum": 1.0071806224814948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8250423809768401, "standard_deviation": 0.3816103475001275, "minimum": 0.29644760377041085, "maximum": 1.5041391515882754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0726386283218679, "standard_deviation": 0.008213845899749519, "minimum": 0.056681903450580204, "maximum": 0.08434745960464925, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.97021994306166, "standard_deviation": 0.006292945264301714, "minimum": 0.9620949556486268, "maximum": 0.979029150614041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.90473565810954, "standard_deviation": 0.02831628650583392, "minimum": 1.8617766083503087, "maximum": 1.948903254930827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8965648627667686, "standard_deviation": 0.32744790028198306, "minimum": 0.4753258852221731, "maximum": 1.5494215451410567, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20000067943768948, "standard_deviation": 0.020684223511775027, "minimum": 0.16757260885661057, "maximum": 0.22416246474670012, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9881583309120713, "standard_deviation": 0.0037388525112634265, "minimum": 0.9803427972813431, "maximum": 0.9920198217932712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0699077762499951, "standard_deviation": 0.03768271181944806, "minimum": 1.012996232724691, "maximum": 1.1310986370953926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.134931688395565, "standard_deviation": 0.5249209057789365, "minimum": 0.42179243121104326, "maximum": 2.025561313844627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12249794915749027, "standard_deviation": 0.01722237381559305, "minimum": 0.10191239215408733, "maximum": 0.15417629341093675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938893710866007, "standard_deviation": 0.00045865985184498484, "minimum": 0.9933711362641124, "maximum": 0.9945115337657413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4611807597941135, "standard_deviation": 0.0559415253360202, "minimum": 1.3988675023609376, "maximum": 1.5818629203343633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8091302027758656, "standard_deviation": 0.11311354546954532, "minimum": 0.6402025292358493, "maximum": 0.9548871148805471, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07787943956257358, "standard_deviation": 0.00265069753827295, "minimum": 0.07431182417187324, "maximum": 0.08082403074110144, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998017884162363, "standard_deviation": 0.00021937711611615113, "minimum": 0.997701760527664, "maximum": 0.9982865097042127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7750733695058682, "standard_deviation": 0.05621963524629109, "minimum": 0.7164706802373805, "maximum": 0.8601527045170019, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8894939619724307, "standard_deviation": 0.18165791601895837, "minimum": 0.5470589520141752, "maximum": 1.0730108885714813, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.042164151038319664, "standard_deviation": 0.002234608941619607, "minimum": 0.03900265351872532, "maximum": 0.045110423991544316, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19642239903405562, "standard_deviation": 0.02496676261306036, "minimum": 0.15339230929468337, "maximum": 0.22378330428016527, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.458687149193006, "standard_deviation": 0.010720268903959742, "minimum": 5.448471868487478, "maximum": 5.481156120168742, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6838886740824329, "standard_deviation": 0.014230020855943727, "minimum": 0.6569928981205709, "maximum": 0.700664360259545, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06370679418847673, "standard_deviation": 0.014766099936929027, "minimum": -0.08460011005295176, "maximum": -0.04897580530021796, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967151908965681, "standard_deviation": 0.0004593461411200952, "minimum": 0.9961629469588494, "maximum": 0.9974769773071693, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7438609562943902, "standard_deviation": 0.020494128326755977, "minimum": 0.7164230348876626, "maximum": 0.7704915443916417, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4814654860527897, "standard_deviation": 0.1302913854488038, "minimum": 0.20483807396522724, "maximum": 0.6092151630831552, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04926619469177293, "standard_deviation": 0.0032547204785002633, "minimum": 0.04411334640613438, "maximum": 0.05315232361039563, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990444371928802, "standard_deviation": 0.0001748574997103368, "minimum": 0.9987040808610184, "maximum": 0.9992515824255149, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37746660305844304, "standard_deviation": 0.021926296280195268, "minimum": 0.34963542625450944, "maximum": 0.40852488606904713, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.538982130884677, "standard_deviation": 0.16483143872147044, "minimum": 0.20880720970980157, "maximum": 0.7193708859514251, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02559945346356176, "standard_deviation": 0.002078743363750952, "minimum": 0.023190069370942194, "maximum": 0.029349936458013506, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3570-5696-0010", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.015476942062378, "predicted_mos": 3.015476942062378, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "51f1633abd67c7e0316a0f639c28ea408df5aec89c7659e2d7021c44fe72195e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.068868199829012, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7386435489162115, "全帯域_10ms_rms_envelope_nrmse": 0.04816710575145329, "全帯域_10ms_envelope_correlation": 0.9968245052463924, "全帯域_10ms_modulation_mae_db": 0.5854341892425966, "全帯域_40ms_log_rms_mae_db": 0.36217412688139244, "全帯域_40ms_rms_envelope_nrmse": 0.025222487134068484, "全帯域_40ms_envelope_correlation": 0.9990151112412234, "全帯域_40ms_modulation_mae_db": 0.666886971354628, "300–1000 Hz_10ms_log_rms_mae_db": 1.4385779499287983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07220481016850401, "300–1000 Hz_10ms_envelope_correlation": 0.9947688190117482, "300–1000 Hz_10ms_modulation_mae_db": 0.8082130991317925, "300–1000 Hz_40ms_log_rms_mae_db": 0.7431474468222474, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03793767900273273, "300–1000 Hz_40ms_envelope_correlation": 0.998391281363847, "300–1000 Hz_40ms_modulation_mae_db": 1.0410219316636298, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6795133347559081, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11402944676360581, "1000–3000 Hz_10ms_envelope_correlation": 0.9916018277590692, "1000–3000 Hz_10ms_modulation_mae_db": 1.0618495084363762, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9423407546787137, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06885362695162145, "1000–3000 Hz_40ms_envelope_correlation": 0.997181652697962, "1000–3000 Hz_40ms_modulation_mae_db": 1.533785523855219, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9166403896840312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20687893717637806, "3000–6000 Hz_10ms_envelope_correlation": 0.9696779735047537, "3000–6000 Hz_10ms_modulation_mae_db": 1.5408403807136486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0871452395957166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1296179307836062, "3000–6000 Hz_40ms_envelope_correlation": 0.9889395731914679, "3000–6000 Hz_40ms_modulation_mae_db": 1.9540296403753858, "mr_stft_db": 5.452721157278909, "band_mae_db": 0.21371672965276614, "rms_gain_error_db": -0.03291898484927279, "output_peak": 0.661632624614443}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7446523439451418, "全帯域_10ms_rms_envelope_nrmse": 0.047362319668094816, "全帯域_10ms_envelope_correlation": 0.996966412975777, "全帯域_10ms_modulation_mae_db": 0.48879701926293045, "全帯域_40ms_log_rms_mae_db": 0.361538882674839, "全帯域_40ms_rms_envelope_nrmse": 0.024005228608182893, "全帯域_40ms_envelope_correlation": 0.9991415678660174, "全帯域_40ms_modulation_mae_db": 0.539310645937349, "300–1000 Hz_10ms_log_rms_mae_db": 1.47110229001604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07665868422201667, "300–1000 Hz_10ms_envelope_correlation": 0.9940990833998586, "300–1000 Hz_10ms_modulation_mae_db": 0.656681044252207, "300–1000 Hz_40ms_log_rms_mae_db": 0.7863465419727023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04354619947264369, "300–1000 Hz_40ms_envelope_correlation": 0.9979061352353412, "300–1000 Hz_40ms_modulation_mae_db": 0.7780736788991826, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7016422401999878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12178925813515579, "1000–3000 Hz_10ms_envelope_correlation": 0.990032893667851, "1000–3000 Hz_10ms_modulation_mae_db": 0.5015571643583673, "1000–3000 Hz_40ms_log_rms_mae_db": 0.926627055922753, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07315328477081628, "1000–3000 Hz_40ms_envelope_correlation": 0.9962872742412122, "1000–3000 Hz_40ms_modulation_mae_db": 0.9869210202696731, "3000–6000 Hz_10ms_log_rms_mae_db": 1.883789689051238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18520175917673912, "3000–6000 Hz_10ms_envelope_correlation": 0.974254902579925, "3000–6000 Hz_10ms_modulation_mae_db": 0.7249968595082485, "3000–6000 Hz_40ms_log_rms_mae_db": 1.023410623379706, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1114996043285064, "3000–6000 Hz_40ms_envelope_correlation": 0.9895919985521417, "3000–6000 Hz_40ms_modulation_mae_db": 0.7091230412168602, "mr_stft_db": 5.440089848906738, "band_mae_db": 0.10854345243317104, "rms_gain_error_db": -0.047940398460559655, "output_peak": 0.6835421325501861}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.774750243292384, "全帯域_10ms_rms_envelope_nrmse": 0.04955203570689028, "全帯域_10ms_envelope_correlation": 0.9966902651523646, "全帯域_10ms_modulation_mae_db": 0.3912065716678299, "全帯域_40ms_log_rms_mae_db": 0.3907352257853242, "全帯域_40ms_rms_envelope_nrmse": 0.026018143693398214, "全帯域_40ms_envelope_correlation": 0.999016528428031, "全帯域_40ms_modulation_mae_db": 0.5062867082340593, "300–1000 Hz_10ms_log_rms_mae_db": 1.472053886481572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08139498731633381, "300–1000 Hz_10ms_envelope_correlation": 0.9933259189004182, "300–1000 Hz_10ms_modulation_mae_db": 0.9018021115695709, "300–1000 Hz_40ms_log_rms_mae_db": 0.8079140586296463, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045956279497149186, "300–1000 Hz_40ms_envelope_correlation": 0.997606064440794, "300–1000 Hz_40ms_modulation_mae_db": 1.134686068396088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.749777785848341, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1158259867169939, "1000–3000 Hz_10ms_envelope_correlation": 0.9915946707362738, "1000–3000 Hz_10ms_modulation_mae_db": 0.5299726357621014, "1000–3000 Hz_40ms_log_rms_mae_db": 1.009731363835021, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0719716271891765, "1000–3000 Hz_40ms_envelope_correlation": 0.9971596497491245, "1000–3000 Hz_40ms_modulation_mae_db": 0.6460384281757583, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8449469099532223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2065987881649968, "3000–6000 Hz_10ms_envelope_correlation": 0.9682846830999752, "3000–6000 Hz_10ms_modulation_mae_db": 0.9007752382617841, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0205844900425862, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12639845918840942, "3000–6000 Hz_40ms_envelope_correlation": 0.9874924493320442, "3000–6000 Hz_40ms_modulation_mae_db": 1.355822384326282, "mr_stft_db": 5.458455922296231, "band_mae_db": 0.19452474862586444, "rms_gain_error_db": -0.06342400501282408, "output_peak": 0.6742861598108026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7705470672887058, "全帯域_10ms_rms_envelope_nrmse": 0.05042876119666652, "全帯域_10ms_envelope_correlation": 0.9965340255318283, "全帯域_10ms_modulation_mae_db": 0.4264889641652759, "全帯域_40ms_log_rms_mae_db": 0.37303456301311755, "全帯域_40ms_rms_envelope_nrmse": 0.025149293644675635, "全帯域_40ms_envelope_correlation": 0.9990403693540588, "全帯域_40ms_modulation_mae_db": 0.6028675464355325, "300–1000 Hz_10ms_log_rms_mae_db": 1.4683145753969031, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07656111297870691, "300–1000 Hz_10ms_envelope_correlation": 0.9941378782272134, "300–1000 Hz_10ms_modulation_mae_db": 0.5647911431589366, "300–1000 Hz_40ms_log_rms_mae_db": 0.8026115033078236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04436615880374387, "300–1000 Hz_40ms_envelope_correlation": 0.9978010632235695, "300–1000 Hz_40ms_modulation_mae_db": 0.7768358377633717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6629717745187202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12203539666875152, "1000–3000 Hz_10ms_envelope_correlation": 0.9901997759735018, "1000–3000 Hz_10ms_modulation_mae_db": 0.2368366031974737, "1000–3000 Hz_40ms_log_rms_mae_db": 0.947160871949694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06913065108475656, "1000–3000 Hz_40ms_envelope_correlation": 0.9970057030212314, "1000–3000 Hz_40ms_modulation_mae_db": 0.6275465706616182, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8769996376948141, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2057309977626757, "3000–6000 Hz_10ms_envelope_correlation": 0.969019343396565, "3000–6000 Hz_10ms_modulation_mae_db": 1.5132640524389223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.036957034032952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1283052140672624, "3000–6000 Hz_40ms_envelope_correlation": 0.9876103330371694, "3000–6000 Hz_40ms_modulation_mae_db": 1.951207646616372, "mr_stft_db": 5.433662614250675, "band_mae_db": 0.1613559882544035, "rms_gain_error_db": -0.042971436955439736, "output_peak": 0.6588278169431228}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7249146323888899, "全帯域_10ms_rms_envelope_nrmse": 0.047180384790579406, "全帯域_10ms_envelope_correlation": 0.9969989222276393, "全帯域_10ms_modulation_mae_db": 0.6343850156082899, "全帯域_40ms_log_rms_mae_db": 0.36126279405765716, "全帯域_40ms_rms_envelope_nrmse": 0.025015052383224858, "全帯域_40ms_envelope_correlation": 0.9990748640722543, "全帯域_40ms_modulation_mae_db": 0.709091868322402, "300–1000 Hz_10ms_log_rms_mae_db": 1.4391227439891419, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08457807742178804, "300–1000 Hz_10ms_envelope_correlation": 0.9927499041412391, "300–1000 Hz_10ms_modulation_mae_db": 0.6595574258146031, "300–1000 Hz_40ms_log_rms_mae_db": 0.7806648487344516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04686326995696935, "300–1000 Hz_40ms_envelope_correlation": 0.9974875970292052, "300–1000 Hz_40ms_modulation_mae_db": 0.687324954053735, "1000–3000 Hz_10ms_log_rms_mae_db": 1.706519034215612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12407935676198503, "1000–3000 Hz_10ms_envelope_correlation": 0.9907838192428212, "1000–3000 Hz_10ms_modulation_mae_db": 0.941187483797498, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9496693484737294, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08250193231872124, "1000–3000 Hz_40ms_envelope_correlation": 0.996628998568524, "1000–3000 Hz_40ms_modulation_mae_db": 1.1425711241555248, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8256930427281488, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.177749019557663, "3000–6000 Hz_10ms_envelope_correlation": 0.9764670194559131, "3000–6000 Hz_10ms_modulation_mae_db": 0.3497579184969232, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9995078906605233, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11909808208241604, "3000–6000 Hz_40ms_envelope_correlation": 0.9879521739276941, "3000–6000 Hz_40ms_modulation_mae_db": 0.5014913146803468, "mr_stft_db": 5.430593113579078, "band_mae_db": 0.15294690466074984, "rms_gain_error_db": -0.052406303493477135, "output_peak": 0.6552583199968018}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7341413154195662, "全帯域_10ms_rms_envelope_nrmse": 0.046937448185305666, "全帯域_10ms_envelope_correlation": 0.9970226559340514, "全帯域_10ms_modulation_mae_db": 0.3193290272956871, "全帯域_40ms_log_rms_mae_db": 0.3641958773279942, "全帯域_40ms_rms_envelope_nrmse": 0.026057887548764102, "全帯域_40ms_envelope_correlation": 0.9989907271782197, "全帯域_40ms_modulation_mae_db": 0.3804604903802198, "300–1000 Hz_10ms_log_rms_mae_db": 1.4772477232137275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08541981831227688, "300–1000 Hz_10ms_envelope_correlation": 0.9925587864728406, "300–1000 Hz_10ms_modulation_mae_db": 0.6341886582519668, "300–1000 Hz_40ms_log_rms_mae_db": 0.7932967687136215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04883339133120589, "300–1000 Hz_40ms_envelope_correlation": 0.9972379757945765, "300–1000 Hz_40ms_modulation_mae_db": 0.675848643212884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7027850332587642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12164690281195899, "1000–3000 Hz_10ms_envelope_correlation": 0.9925270170925244, "1000–3000 Hz_10ms_modulation_mae_db": 0.5673893105792277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.949557228545218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08532708098152167, "1000–3000 Hz_40ms_envelope_correlation": 0.9977931070244082, "1000–3000 Hz_40ms_modulation_mae_db": 0.8938593226992383, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8359739816062561, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18694421713873613, "3000–6000 Hz_10ms_envelope_correlation": 0.973929939378435, "3000–6000 Hz_10ms_modulation_mae_db": 0.44227764843319967, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0121117897109277, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11866735763428334, "3000–6000 Hz_40ms_envelope_correlation": 0.9881268659948866, "3000–6000 Hz_40ms_modulation_mae_db": 0.41601249166454185, "mr_stft_db": 5.440718564293355, "band_mae_db": 0.1757427955563885, "rms_gain_error_db": -0.0570369096753303, "output_peak": 0.6824879303834215}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7565208703101122, "全帯域_10ms_rms_envelope_nrmse": 0.05292949381870996, "全帯域_10ms_envelope_correlation": 0.9962566150873068, "全帯域_10ms_modulation_mae_db": 0.62062507804604, "全帯域_40ms_log_rms_mae_db": 0.39276148764181434, "全帯域_40ms_rms_envelope_nrmse": 0.028993905790303455, "全帯域_40ms_envelope_correlation": 0.9988213961664523, "全帯域_40ms_modulation_mae_db": 0.5727048869095505, "300–1000 Hz_10ms_log_rms_mae_db": 1.4508092918812503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07708286742648568, "300–1000 Hz_10ms_envelope_correlation": 0.9940046293398977, "300–1000 Hz_10ms_modulation_mae_db": 0.7357364032195921, "300–1000 Hz_40ms_log_rms_mae_db": 0.8201009828973208, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040987609904209535, "300–1000 Hz_40ms_envelope_correlation": 0.9981304890431543, "300–1000 Hz_40ms_modulation_mae_db": 0.7477897712220073, "1000–3000 Hz_10ms_log_rms_mae_db": 1.727936267859191, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12371251848932659, "1000–3000 Hz_10ms_envelope_correlation": 0.991303443966157, "1000–3000 Hz_10ms_modulation_mae_db": 0.8289098004078728, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9678913738995308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0778118769816888, "1000–3000 Hz_40ms_envelope_correlation": 0.9977338657851014, "1000–3000 Hz_40ms_modulation_mae_db": 1.3365591660529565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8721319966014993, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.187615025707086, "3000–6000 Hz_10ms_envelope_correlation": 0.9744520216933543, "3000–6000 Hz_10ms_modulation_mae_db": 0.9727836497158712, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0546182762983738, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10741821556619638, "3000–6000 Hz_40ms_envelope_correlation": 0.9917942441538623, "3000–6000 Hz_40ms_modulation_mae_db": 1.307406325692099, "mr_stft_db": 5.445578188383565, "band_mae_db": 0.23286324710470183, "rms_gain_error_db": -0.08836510780700177, "output_peak": 0.6575499324156289}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.739109287447697, "全帯域_10ms_rms_envelope_nrmse": 0.049801997553273476, "全帯域_10ms_envelope_correlation": 0.9966358666535029, "全帯域_10ms_modulation_mae_db": 0.6122634588311126, "全帯域_40ms_log_rms_mae_db": 0.3588206599874649, "全帯域_40ms_rms_envelope_nrmse": 0.024811674701671803, "全帯域_40ms_envelope_correlation": 0.9991111833700184, "全帯域_40ms_modulation_mae_db": 0.6082591898630326, "300–1000 Hz_10ms_log_rms_mae_db": 1.4124168962920658, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07745934484704192, "300–1000 Hz_10ms_envelope_correlation": 0.9939659762178727, "300–1000 Hz_10ms_modulation_mae_db": 0.8708345966661449, "300–1000 Hz_40ms_log_rms_mae_db": 0.7672099665510674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04226326358057361, "300–1000 Hz_40ms_envelope_correlation": 0.9980307305591707, "300–1000 Hz_40ms_modulation_mae_db": 0.9362361774263803, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7164906953088162, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1112396985291885, "1000–3000 Hz_10ms_envelope_correlation": 0.9916877735706614, "1000–3000 Hz_10ms_modulation_mae_db": 0.28792641287999227, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9981370207736242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06485882675765484, "1000–3000 Hz_40ms_envelope_correlation": 0.9970749228227798, "1000–3000 Hz_40ms_modulation_mae_db": 0.42886915902502337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9185813002298626, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2420979914619302, "3000–6000 Hz_10ms_envelope_correlation": 0.9598203820150829, "3000–6000 Hz_10ms_modulation_mae_db": 2.1597412553777726, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1056846193064451, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15504300391317413, "3000–6000 Hz_40ms_envelope_correlation": 0.9865402248998935, "3000–6000 Hz_40ms_modulation_mae_db": 2.6903135809257406, "mr_stft_db": 5.43785872772395, "band_mae_db": 0.24032775918590152, "rms_gain_error_db": -0.06544079306386227, "output_peak": 0.6873399361864241}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.600564561119487, "spectral_convergence": 0.14230810733265406, "spectral_convergence_status": "ok", "amplitude_floor": 1.1745223847485646e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.297744476633102, "spectral_convergence": 0.13731413515675508, "spectral_convergence_status": "ok", "amplitude_floor": 9.481908348017771e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.459854434084138, "spectral_convergence": 0.13459337709990055, "spectral_convergence_status": "ok", "amplitude_floor": 8.395402116778918e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004068927413639468, "output_power": 0.004067145476441796, "delta_db": -0.0019023563699008096, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018176508169902323, "output_power": 0.0018177012380843072, "delta_db": 0.00012047032913389186, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007805177117872849, "output_power": 0.0007573405376275267, "delta_db": -0.1309155905791233, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000310938544728363, "output_power": 0.0002940597376861672, "delta_db": -0.2423899613536843, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.08647023559392e-05, "output_power": 7.389081294992783e-05, "delta_db": -0.3916854763905201, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.353447039170603e-05, "output_power": 3.866380690955918e-05, "delta_db": -0.5152865228942345, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21371672965276614, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03291898484927279, "pitch": null, "output_rms_dbfs": -21.51884712587941, "output_peak": 0.661632624614443, "reference_rms_dbfs": -21.485928141030136, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.452721157278909}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019068326661124743, "log_rms_mae_db": 0.7386435489162115, "rms_envelope_nrmse": 0.04816710575145329, "envelope_correlation": 0.9968245052463924, "rms_gain_error_db": -0.03204953594259602, "reference_mean_power": 0.007104543911662494, "output_mean_power": 0.007052307643124137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.7734542191089184e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001358301262786015, "output_power": 0.0013557665156948943, "reference_active": true, "delta_db": -0.00811200813390218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002660490997406596, "output_power": 0.0002401151252514218, "reference_active": true, "delta_db": -0.44542275842637113}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.883599774152297e-05, "output_power": 5.840469876032668e-05, "reference_active": true, "delta_db": -1.3027678011675166}], "active_band_count": 3, "active_band_mae_db": 0.5854341892425966}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001870958079240233, "log_rms_mae_db": 0.36217412688139244, "rms_envelope_nrmse": 0.025222487134068484, "envelope_correlation": 0.9990151112412234, "rms_gain_error_db": -0.03141221436914286, "reference_mean_power": 0.007117638239873263, "output_mean_power": 0.007066342595700105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4206220628648622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012103399895685932, "output_power": 0.0012171002943630854, "reference_active": true, "delta_db": 0.024189848690440593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012555497766250794, "output_power": 0.00011338469129807432, "reference_active": true, "delta_db": -0.4427951313182612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0918852450489625e-06, "output_power": 4.279412199258288e-06, "reference_active": true, "delta_db": -1.533675934055182}], "active_band_count": 3, "active_band_mae_db": 0.666886971354628}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013521834879092463, "log_rms_mae_db": 1.4385779499287983, "rms_envelope_nrmse": 0.07220481016850401, "envelope_correlation": 0.9947688190117482, "rms_gain_error_db": -0.06290793518457576, "reference_mean_power": 0.0018937311393695897, "output_mean_power": 0.0018664979917701649}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.102346197011796e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006150154421416123, "output_power": 0.00060155828830593, "reference_active": true, "delta_db": -0.09608305507551719}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012427322589552618, "output_power": 0.00010809520664387805, "reference_active": true, "delta_db": -0.6057113584052205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6589172324725154e-05, "output_power": 1.788217457243248e-05, "reference_active": true, "delta_db": -1.7228448839146402}], "active_band_count": 3, "active_band_mae_db": 0.8082130991317925}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012428224153258587, "log_rms_mae_db": 0.7431474468222474, "rms_envelope_nrmse": 0.03793767900273273, "envelope_correlation": 0.998391281363847, "rms_gain_error_db": -0.06237899930238388, "reference_mean_power": 0.0018973976433641784, "output_mean_power": 0.0018703395474324873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.796182335233462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005643605996811222, "output_power": 0.0005523227983773218, "reference_active": true, "delta_db": -0.0936371618447184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.691592416494353e-05, "output_power": 4.936378161109852e-05, "reference_active": true, "delta_db": -0.6182536974785175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2464097368217303e-06, "output_power": 1.2893519436811042e-06, "reference_active": true, "delta_db": -2.4111749356676535}], "active_band_count": 3, "active_band_mae_db": 1.0410219316636298}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012748188769016831, "log_rms_mae_db": 1.6795133347559081, "rms_envelope_nrmse": 0.11402944676360581, "envelope_correlation": 0.9916018277590692, "rms_gain_error_db": -0.2566108269612447, "reference_mean_power": 0.00032033050063867626, "output_mean_power": 0.00030195151279249435}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0949985705721996e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.663152856065707e-05, "output_power": 6.305912163856336e-05, "reference_active": true, "delta_db": -0.2393185922458058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2062707324852824e-05, "output_power": 9.102644533416773e-06, "reference_active": true, "delta_db": -1.2227720728080718}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2084052775211794e-06, "output_power": 2.1574633955152616e-06, "reference_active": true, "delta_db": -1.7234578602552508}], "active_band_count": 3, "active_band_mae_db": 1.0618495084363762}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001243805164613915, "log_rms_mae_db": 0.9423407546787137, "rms_envelope_nrmse": 0.06885362695162145, "envelope_correlation": 0.997181652697962, "rms_gain_error_db": -0.25839272594951196, "reference_mean_power": 0.0003214084172642619, "output_mean_power": 0.00030284330255081693}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.993543951674123e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.454820101360055e-05, "output_power": 6.036697426950124e-05, "reference_active": true, "delta_db": -0.29084734226702813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.405216922862835e-06, "output_power": 4.165180906683368e-06, "reference_active": true, "delta_db": -1.1317925889683043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4345993241652877e-07, "output_power": 1.651978596180199e-07, "reference_active": true, "delta_db": -3.1787166403303244}], "active_band_count": 3, "active_band_mae_db": 1.533785523855219}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.885927766129993e-05, "log_rms_mae_db": 1.9166403896840312, "rms_envelope_nrmse": 0.20687893717637806, "envelope_correlation": 0.9696779735047537, "rms_gain_error_db": -0.5881797933439055, "reference_mean_power": 5.543232992418922e-05, "output_mean_power": 4.841112263185128e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.209399643069482e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1581941196726732e-05, "output_power": 1.9491137063627965e-05, "reference_active": true, "delta_db": -0.4425332940449571}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7792321883200776e-05, "output_power": 1.3994946126724384e-05, "reference_active": true, "delta_db": -1.0426139585362957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1497984756130906e-05, "output_power": 5.583214838881855e-06, "reference_active": true, "delta_db": -3.137373889559693}], "active_band_count": 3, "active_band_mae_db": 1.5408403807136486}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.78543565194846e-05, "log_rms_mae_db": 1.0871452395957166, "rms_envelope_nrmse": 0.1296179307836062, "envelope_correlation": 0.9889395731914679, "rms_gain_error_db": -0.5888386242357111, "reference_mean_power": 5.5701179342093606e-05, "output_mean_power": 4.863853976761618e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.455779259683724e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.849693387286164e-05, "output_power": 2.3560980662809873e-05, "reference_active": true, "delta_db": -0.8260477177681019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0110098860775756e-05, "output_power": 6.874634454619407e-06, "reference_active": true, "delta_db": -1.675057919770958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1498909214748545e-06, "output_power": 5.303448022543349e-07, "reference_active": true, "delta_db": -3.3609832835870974}], "active_band_count": 3, "active_band_mae_db": 1.9540296403753858}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9314902238793407, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "219e10e63d653a560adc20645050f8ce359a79f159b0c13ec4a7ae51e4160de9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9912164027511075, "standard_deviation": 0.0008332782014149161, "minimum": 0.990032893667851, "maximum": 0.9925270170925244, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7059545207456674, "standard_deviation": 0.02696638019167694, "minimum": 1.6629717745187202, "maximum": 1.749777785848341, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6194536149273637, "standard_deviation": 0.29857324453059714, "minimum": 0.2368366031974737, "maximum": 1.0618495084363762, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11929482060962077, "standard_deviation": 0.004874350787313333, "minimum": 0.1112396985291885, "maximum": 0.12407935676198503, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997108146738793, "standard_deviation": 0.0005048003529132384, "minimum": 0.9962872742412122, "maximum": 0.9977931070244082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9613893772597855, "standard_deviation": 0.02873696255558542, "minimum": 0.926627055922753, "maximum": 1.009731363835021, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9495187893618764, "standard_deviation": 0.37792222444578294, "minimum": 0.42886915902502337, "maximum": 1.533785523855219, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07420111337949467, "standard_deviation": 0.007105133312344854, "minimum": 0.06485882675765484, "maximum": 0.08532708098152167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9707382831405005, "standard_deviation": 0.005324650010544795, "minimum": 0.9598203820150829, "maximum": 0.9764670194559131, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8718446184436341, "standard_deviation": 0.03484758101277182, "minimum": 1.8256930427281488, "maximum": 1.9185813002298626, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0755546253682962, "standard_deviation": 0.618218878182101, "minimum": 0.3497579184969232, "maximum": 2.1597412553777726, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1998520920182756, "standard_deviation": 0.020471963771754628, "minimum": 0.177749019557663, "maximum": 0.2420979914619302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9885059828861449, "standard_deviation": 0.0016179427017003532, "minimum": 0.9865402248998935, "maximum": 0.9917942441538623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.042502495378404, "standard_deviation": 0.037377474584634425, "minimum": 0.9995078906605233, "maximum": 1.1056846193064451, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3606758031872035, "standard_deviation": 0.8041590288854066, "minimum": 0.41601249166454185, "maximum": 2.6903135809257406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12450598344548178, "standard_deviation": 0.01464411095547449, "minimum": 0.10741821556619638, "maximum": 0.15504300391317413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937013744638861, "standard_deviation": 0.0007561078320912334, "minimum": 0.9925587864728406, "maximum": 0.9947688190117482, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4537056696499373, "standard_deviation": 0.022556624964593725, "minimum": 1.4124168962920658, "maximum": 1.4772477232137275, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7289755602581017, "standard_deviation": 0.12089177784720129, "minimum": 0.5647911431589366, "maximum": 0.9018021115695709, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07891996283664424, "standard_deviation": 0.004498686815966577, "minimum": 0.07220481016850401, "maximum": 0.08541981831227688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978239170862073, "standard_deviation": 0.0003722212623666976, "minimum": 0.9972379757945765, "maximum": 0.998391281363847, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7876615147036101, "standard_deviation": 0.02440197131480395, "minimum": 0.7431474468222474, "maximum": 0.8201009828973208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8472271328296598, "standard_deviation": 0.1701578639666429, "minimum": 0.675848643212884, "maximum": 1.134686068396088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.043844231443653486, "standard_deviation": 0.0034718335187771086, "minimum": 0.03793767900273273, "maximum": 0.04883339133120589, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18500270318424333, "standard_deviation": 0.04437689619613816, "minimum": 0.10854345243317104, "maximum": 0.24032775918590152, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.442459767089063, "standard_deviation": 0.009403060957582695, "minimum": 5.430593113579078, "maximum": 5.458455922296231, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6701156066126038, "standard_deviation": 0.013231063828847572, "minimum": 0.6552583199968018, "maximum": 0.6873399361864241, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05631299241472097, "standard_deviation": 0.016785983277191587, "minimum": -0.08836510780700177, "maximum": -0.03291898484927279, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9967411586011079, "standard_deviation": 0.00026584960697894635, "minimum": 0.9962566150873068, "maximum": 0.9970226559340514, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7479099136260885, "standard_deviation": 0.017725196551799507, "minimum": 0.7249146323888899, "maximum": 0.774750243292384, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5098161655149702, "standard_deviation": 0.12056317816437204, "minimum": 0.3193290272956871, "maximum": 0.6343850156082899, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.049044943333871674, "standard_deviation": 0.002047817433857225, "minimum": 0.046937448185305666, "maximum": 0.05292949381870996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990264684595345, "standard_deviation": 9.748240489162819e-05, "minimum": 0.9988213961664523, "maximum": 0.9991415678660174, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3705654521712005, "standard_deviation": 0.013745257045350252, "minimum": 0.3588206599874649, "maximum": 0.39276148764181434, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5732335384295968, "standard_deviation": 0.10150847271018867, "minimum": 0.3804604903802198, "maximum": 0.709091868322402, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02565920918803618, "standard_deviation": 0.0014989341122802191, "minimum": 0.024005228608182893, "maximum": 0.028993905790303455, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3575-170457-0034", "language": "en", "style": "normal", "speaker": "librispeech_3575", "source_utterance_id": "3575-170457-0034", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 55920, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3575-170457-0034.wav", "source_file_sha256": "0748e6c5564072599215a83c5c2c77d2850662a8bb7944e508d06ef834b647a2", "decoded_pcm_sha256": "b1c60dfc008f07f56156adc48b00db15f16da435fb212895c0516ecbd9e5de30", "original_source_sha256": "7379f6afffbfc11aded5aae60cf94ff2f5aa5a2dc317f61b2cd70c9c4d4847df", "corpus_entry_id": "librispeech_3575-170457-0034"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.7620341777801514, "predicted_mos": 3.7620341777801514, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "ac8ff148d90e17b8dd51e29bb3180d903d9284bfe0d3b6105fcd645c44734f25", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.38017998199939, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3575-170457-0034", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ac8ff148d90e17b8dd51e29bb3180d903d9284bfe0d3b6105fcd645c44734f25", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "f057399e7bc52496d756298eb590948a7652ca6ab292cde23ed3a5d6e4f624f5"}, "rows": [{"case": "librispeech_3575-170457-0034", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.3260498046875, "predicted_mos": 2.3260498046875, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "dffbad99064a1b21aa95335b83933a3521fbe6a32fb43e815f3c625adaa767a5", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.335413324999536, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3575-170457-0034", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "dffbad99064a1b21aa95335b83933a3521fbe6a32fb43e815f3c625adaa767a5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7635987659098031, "全帯域_10ms_rms_envelope_nrmse": 0.16710946559272283, "全帯域_10ms_envelope_correlation": 0.9678291446947767, "全帯域_10ms_modulation_mae_db": 1.2042405780231726, "全帯域_40ms_log_rms_mae_db": 0.4524390782980965, "全帯域_40ms_rms_envelope_nrmse": 0.08252258239400814, "全帯域_40ms_envelope_correlation": 0.9918604959098112, "全帯域_40ms_modulation_mae_db": 1.685252811647891, "300–1000 Hz_10ms_log_rms_mae_db": 1.3861700366617402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11152628162117094, "300–1000 Hz_10ms_envelope_correlation": 0.9883838085134776, "300–1000 Hz_10ms_modulation_mae_db": 1.0967717038496572, "300–1000 Hz_40ms_log_rms_mae_db": 0.8689853017396794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06672118963612272, "300–1000 Hz_40ms_envelope_correlation": 0.9959043043376116, "300–1000 Hz_40ms_modulation_mae_db": 1.406576089277788, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2633688354028567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10168973958289189, "1000–3000 Hz_10ms_envelope_correlation": 0.9932317975675976, "1000–3000 Hz_10ms_modulation_mae_db": 0.884540256198452, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7996163852631227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057494147437707636, "1000–3000 Hz_40ms_envelope_correlation": 0.9978409352784784, "1000–3000 Hz_40ms_modulation_mae_db": 0.16972545605426678, "3000–6000 Hz_10ms_log_rms_mae_db": 1.487508817413126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1741089109710799, "3000–6000 Hz_10ms_envelope_correlation": 0.9746356185010852, "3000–6000 Hz_10ms_modulation_mae_db": 0.7551734539672155, "3000–6000 Hz_40ms_log_rms_mae_db": 0.975538199597204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10692335083654769, "3000–6000 Hz_40ms_envelope_correlation": 0.9896528876812744, "3000–6000 Hz_40ms_modulation_mae_db": 0.5712398159663322, "mr_stft_db": 5.433391186847909, "band_mae_db": 0.3662988408259837, "rms_gain_error_db": -0.18298065313257159, "output_peak": 0.8566661798711086}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7621416992667439, "全帯域_10ms_rms_envelope_nrmse": 0.1070749406854656, "全帯域_10ms_envelope_correlation": 0.9870851786599614, "全帯域_10ms_modulation_mae_db": 0.3091322143640934, "全帯域_40ms_log_rms_mae_db": 0.42370305289284593, "全帯域_40ms_rms_envelope_nrmse": 0.05350418629953443, "全帯域_40ms_envelope_correlation": 0.9966839521856665, "全帯域_40ms_modulation_mae_db": 0.25984264200614743, "300–1000 Hz_10ms_log_rms_mae_db": 1.3973273858958626, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11933937951787758, "300–1000 Hz_10ms_envelope_correlation": 0.9869081211076575, "300–1000 Hz_10ms_modulation_mae_db": 0.9613947748986872, "300–1000 Hz_40ms_log_rms_mae_db": 0.9870220839666878, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07970516577920853, "300–1000 Hz_40ms_envelope_correlation": 0.9943122751502937, "300–1000 Hz_40ms_modulation_mae_db": 1.1188292666261146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2335072088324042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10109863108046886, "1000–3000 Hz_10ms_envelope_correlation": 0.9935620530511315, "1000–3000 Hz_10ms_modulation_mae_db": 0.37174472378709217, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8160530545400638, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054929117404581465, "1000–3000 Hz_40ms_envelope_correlation": 0.9983958638953813, "1000–3000 Hz_40ms_modulation_mae_db": 0.15469732753543522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5050902482090835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1876757514654069, "3000–6000 Hz_10ms_envelope_correlation": 0.9706375435310302, "3000–6000 Hz_10ms_modulation_mae_db": 0.5421662875643095, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0197708800078615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12845532790123448, "3000–6000 Hz_40ms_envelope_correlation": 0.9854246743180044, "3000–6000 Hz_40ms_modulation_mae_db": 0.8541682860570082, "mr_stft_db": 5.431533574995964, "band_mae_db": 0.23572677324694521, "rms_gain_error_db": -0.1368693068538583, "output_peak": 1.2383544772656818}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 29, "at_or_above_full_scale_fraction": 0.0005185979971387696, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0005185979971387696, "longest_full_scale_run_samples": 29, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7566928405356753, "全帯域_10ms_rms_envelope_nrmse": 0.1496018006573612, "全帯域_10ms_envelope_correlation": 0.9752841242422988, "全帯域_10ms_modulation_mae_db": 1.0687303198881246, "全帯域_40ms_log_rms_mae_db": 0.417034618020036, "全帯域_40ms_rms_envelope_nrmse": 0.08231970234953638, "全帯域_40ms_envelope_correlation": 0.9918934033101497, "全帯域_40ms_modulation_mae_db": 0.8155959784358636, "300–1000 Hz_10ms_log_rms_mae_db": 1.3053346793765277, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10383763780373821, "300–1000 Hz_10ms_envelope_correlation": 0.9896453326348308, "300–1000 Hz_10ms_modulation_mae_db": 0.24746390434883647, "300–1000 Hz_40ms_log_rms_mae_db": 0.8112936970901521, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05680918784184934, "300–1000 Hz_40ms_envelope_correlation": 0.9966438868896615, "300–1000 Hz_40ms_modulation_mae_db": 0.1303292865277947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2799317687430982, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0904197365712525, "1000–3000 Hz_10ms_envelope_correlation": 0.9946595626169559, "1000–3000 Hz_10ms_modulation_mae_db": 0.1994381164447494, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8159950114613088, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0520253363642725, "1000–3000 Hz_40ms_envelope_correlation": 0.9982590533475635, "1000–3000 Hz_40ms_modulation_mae_db": 0.6213151611324396, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4582799005657991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1917245523369575, "3000–6000 Hz_10ms_envelope_correlation": 0.969114182137609, "3000–6000 Hz_10ms_modulation_mae_db": 0.740468598793179, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0033735407798756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12148764762551507, "3000–6000 Hz_40ms_envelope_correlation": 0.986533079553379, "3000–6000 Hz_40ms_modulation_mae_db": 0.6167174648705274, "mr_stft_db": 5.415842617926454, "band_mae_db": 0.18963029722279245, "rms_gain_error_db": 0.06927814417717487, "output_peak": 1.0661123380729942}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 12, "at_or_above_full_scale_fraction": 0.00021459227467811158, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.00021459227467811158, "longest_full_scale_run_samples": 12, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7496434409398535, "全帯域_10ms_rms_envelope_nrmse": 0.22182657453224439, "全帯域_10ms_envelope_correlation": 0.9425850268710135, "全帯域_10ms_modulation_mae_db": 1.8271623632093343, "全帯域_40ms_log_rms_mae_db": 0.4436006296175179, "全帯域_40ms_rms_envelope_nrmse": 0.12471008797770426, "全帯域_40ms_envelope_correlation": 0.9811400711474307, "全帯域_40ms_modulation_mae_db": 2.337518752198104, "300–1000 Hz_10ms_log_rms_mae_db": 1.3768833633264608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10233831147619304, "300–1000 Hz_10ms_envelope_correlation": 0.990002563112551, "300–1000 Hz_10ms_modulation_mae_db": 0.44656822055141226, "300–1000 Hz_40ms_log_rms_mae_db": 0.8543271445536582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057769508254319625, "300–1000 Hz_40ms_envelope_correlation": 0.9965482156948423, "300–1000 Hz_40ms_modulation_mae_db": 0.5750710382515472, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2564197848035732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1291808204021943, "1000–3000 Hz_10ms_envelope_correlation": 0.9890643753525278, "1000–3000 Hz_10ms_modulation_mae_db": 1.160924280965873, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8212796863932589, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06573220825932936, "1000–3000 Hz_40ms_envelope_correlation": 0.9970817766789232, "1000–3000 Hz_40ms_modulation_mae_db": 0.7920139159860827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.516012103210518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20270033460976142, "3000–6000 Hz_10ms_envelope_correlation": 0.9656796652634827, "3000–6000 Hz_10ms_modulation_mae_db": 1.0322092411832717, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0489109359614819, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13120595638084895, "3000–6000 Hz_40ms_envelope_correlation": 0.9850662275001987, "3000–6000 Hz_40ms_modulation_mae_db": 0.8257945554197978, "mr_stft_db": 5.43755006912588, "band_mae_db": 0.40551396016058533, "rms_gain_error_db": -0.27996820889593715, "output_peak": 1.0505617584740041}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 8, "at_or_above_full_scale_fraction": 0.0001430615164520744, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0001430615164520744, "longest_full_scale_run_samples": 7, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8140867163649441, "全帯域_10ms_rms_envelope_nrmse": 0.1537388846588363, "全帯域_10ms_envelope_correlation": 0.9744487080372923, "全帯域_10ms_modulation_mae_db": 3.0265258539935913, "全帯域_40ms_log_rms_mae_db": 0.47822779947172106, "全帯域_40ms_rms_envelope_nrmse": 0.12231480741000195, "全帯域_40ms_envelope_correlation": 0.9832018374720621, "全帯域_40ms_modulation_mae_db": 3.640062933158339, "300–1000 Hz_10ms_log_rms_mae_db": 1.3264312941276886, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10970648985396664, "300–1000 Hz_10ms_envelope_correlation": 0.9885546685297908, "300–1000 Hz_10ms_modulation_mae_db": 0.4233431641421109, "300–1000 Hz_40ms_log_rms_mae_db": 0.7506561370508528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056144999025758494, "300–1000 Hz_40ms_envelope_correlation": 0.996761422560423, "300–1000 Hz_40ms_modulation_mae_db": 0.46044756430515427, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2182312341231996, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1084238270171102, "1000–3000 Hz_10ms_envelope_correlation": 0.992256285706743, "1000–3000 Hz_10ms_modulation_mae_db": 0.4469943230309834, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7585684167242349, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0546488151406061, "1000–3000 Hz_40ms_envelope_correlation": 0.9980511435021141, "1000–3000 Hz_40ms_modulation_mae_db": 0.3361364940676847, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5121540923650945, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19783997732233075, "3000–6000 Hz_10ms_envelope_correlation": 0.9674838678118839, "3000–6000 Hz_10ms_modulation_mae_db": 0.725519990777606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0739641183006579, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13942230651152276, "3000–6000 Hz_40ms_envelope_correlation": 0.9830193759934979, "3000–6000 Hz_40ms_modulation_mae_db": 0.7921834482712694, "mr_stft_db": 5.435640494383873, "band_mae_db": 0.5408015680530299, "rms_gain_error_db": -0.377959834915546, "output_peak": 0.7027624570942617}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7148867685184976, "全帯域_10ms_rms_envelope_nrmse": 0.1168416763550334, "全帯域_10ms_envelope_correlation": 0.984547657945058, "全帯域_10ms_modulation_mae_db": 0.4115263279455519, "全帯域_40ms_log_rms_mae_db": 0.3968676022242449, "全帯域_40ms_rms_envelope_nrmse": 0.057862575903998445, "全帯域_40ms_envelope_correlation": 0.995900982713249, "全帯域_40ms_modulation_mae_db": 0.46351192198103514, "300–1000 Hz_10ms_log_rms_mae_db": 1.3269402167368138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1218049776207567, "300–1000 Hz_10ms_envelope_correlation": 0.9859269525712199, "300–1000 Hz_10ms_modulation_mae_db": 1.2287856136522153, "300–1000 Hz_40ms_log_rms_mae_db": 0.8711620273527733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07140316293683759, "300–1000 Hz_40ms_envelope_correlation": 0.9950582747139153, "300–1000 Hz_40ms_modulation_mae_db": 1.5641630455979563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3085127668299796, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10916022169939199, "1000–3000 Hz_10ms_envelope_correlation": 0.9924804906659035, "1000–3000 Hz_10ms_modulation_mae_db": 0.6825638185314311, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8151673871166872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061010873313053, "1000–3000 Hz_40ms_envelope_correlation": 0.9979783750075202, "1000–3000 Hz_40ms_modulation_mae_db": 0.7585346171296656, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4865524436661293, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1933649066496913, "3000–6000 Hz_10ms_envelope_correlation": 0.9685740919386224, "3000–6000 Hz_10ms_modulation_mae_db": 0.7255029566419343, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0217075019542137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12204442711317702, "3000–6000 Hz_40ms_envelope_correlation": 0.9864349849523906, "3000–6000 Hz_40ms_modulation_mae_db": 0.42059731495065905, "mr_stft_db": 5.421189150889806, "band_mae_db": 0.20938741723639945, "rms_gain_error_db": -0.0666234340246703, "output_peak": 0.8952720116761691}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7516553026766564, "全帯域_10ms_rms_envelope_nrmse": 0.09375843446984519, "全帯域_10ms_envelope_correlation": 0.9905264206897096, "全帯域_10ms_modulation_mae_db": 1.1522615489422001, "全帯域_40ms_log_rms_mae_db": 0.40763976769155646, "全帯域_40ms_rms_envelope_nrmse": 0.06128667995319648, "全帯域_40ms_envelope_correlation": 0.9960384580488407, "全帯域_40ms_modulation_mae_db": 1.347285576661945, "300–1000 Hz_10ms_log_rms_mae_db": 1.3085740513386799, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10146207795443657, "300–1000 Hz_10ms_envelope_correlation": 0.990348031733985, "300–1000 Hz_10ms_modulation_mae_db": 0.48414855490722336, "300–1000 Hz_40ms_log_rms_mae_db": 0.8291607209483938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06000328650832699, "300–1000 Hz_40ms_envelope_correlation": 0.9963992478205815, "300–1000 Hz_40ms_modulation_mae_db": 0.5018709601118562, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3061490842440615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10038718389658126, "1000–3000 Hz_10ms_envelope_correlation": 0.9938115641084527, "1000–3000 Hz_10ms_modulation_mae_db": 0.5744251531972591, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8002851548064236, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06367741458427206, "1000–3000 Hz_40ms_envelope_correlation": 0.9978341760248058, "1000–3000 Hz_40ms_modulation_mae_db": 0.9078115276140831, "3000–6000 Hz_10ms_log_rms_mae_db": 1.539337227540521, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18993445637348816, "3000–6000 Hz_10ms_envelope_correlation": 0.9701212633585383, "3000–6000 Hz_10ms_modulation_mae_db": 1.0122305047327322, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0829820373411583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1401691923993885, "3000–6000 Hz_40ms_envelope_correlation": 0.9827689342433144, "3000–6000 Hz_40ms_modulation_mae_db": 0.7075533726899144, "mr_stft_db": 5.41920662883103, "band_mae_db": 0.3504111416000717, "rms_gain_error_db": -0.21673117653023224, "output_peak": 1.0037023925858537}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.78826895565093e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 3.57653791130186e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7618577500017589, "全帯域_10ms_rms_envelope_nrmse": 0.25794256758148754, "全帯域_10ms_envelope_correlation": 0.9341818477335804, "全帯域_10ms_modulation_mae_db": 1.769360226707744, "全帯域_40ms_log_rms_mae_db": 0.4421098824944352, "全帯域_40ms_rms_envelope_nrmse": 0.17221293139990104, "全帯域_40ms_envelope_correlation": 0.9694919538124483, "全帯域_40ms_modulation_mae_db": 2.24767093513035, "300–1000 Hz_10ms_log_rms_mae_db": 1.3203862416286558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11368036915724403, "300–1000 Hz_10ms_envelope_correlation": 0.9878624892358664, "300–1000 Hz_10ms_modulation_mae_db": 0.5757120612381141, "300–1000 Hz_40ms_log_rms_mae_db": 0.7934191360362431, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06710992866017616, "300–1000 Hz_40ms_envelope_correlation": 0.9957982122686198, "300–1000 Hz_40ms_modulation_mae_db": 0.7390336020660673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2850844346368666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12030650361481446, "1000–3000 Hz_10ms_envelope_correlation": 0.9916345248251343, "1000–3000 Hz_10ms_modulation_mae_db": 0.5736986312491344, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7715604795964947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0780346446622668, "1000–3000 Hz_40ms_envelope_correlation": 0.9973964718638312, "1000–3000 Hz_40ms_modulation_mae_db": 0.35350743556872805, "3000–6000 Hz_10ms_log_rms_mae_db": 1.581102055581683, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1864492612091798, "3000–6000 Hz_10ms_envelope_correlation": 0.9709339331595552, "3000–6000 Hz_10ms_modulation_mae_db": 0.4038395805408616, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0869705279560877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12151642946281982, "3000–6000 Hz_40ms_envelope_correlation": 0.9866542213055851, "3000–6000 Hz_40ms_modulation_mae_db": 0.2660241851185453, "mr_stft_db": 5.421193138554465, "band_mae_db": 0.49762021102987064, "rms_gain_error_db": 0.33768003223513193, "output_peak": 1.290081306516348}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 59, "at_or_above_full_scale_fraction": 0.0010550786838340487, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0010550786838340487, "longest_full_scale_run_samples": 56, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 55920, "output_samples": 55920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.352246013062165, "spectral_convergence": 0.38824763222825465, "spectral_convergence_status": "ok", "amplitude_floor": 7.545858591823444e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.28996016682028, "spectral_convergence": 0.37273260930339386, "spectral_convergence_status": "ok", "amplitude_floor": 3.9963548992246115e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.657967380661283, "spectral_convergence": 0.33422959418851933, "spectral_convergence_status": "ok", "amplitude_floor": 1.7986227023732554e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001975478876015244, "output_power": 0.0014868007641465537, "delta_db": -1.2341961445887606, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005013278749394044, "output_power": 0.005198199203193711, "delta_db": 0.15731065314742085, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015170067046631896, "output_power": 0.00143796034129152, "delta_db": -0.23240591785572445, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0018605244766058998, "output_power": 0.0018202022744738702, "delta_db": -0.09515735153383256, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002538671104412511, "output_power": 0.00023940361408744448, "delta_db": -0.25475737508657453, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.356486516629775e-05, "output_power": 6.036991575827152e-05, "delta_db": -0.22396560274358868, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3662988408259837, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.18298065313257159, "pitch": null, "output_rms_dbfs": -19.89575536047884, "output_peak": 0.8566661798711086, "reference_rms_dbfs": -19.712774707346266, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.433391186847909}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 55920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.0006292775937517623, "log_rms_mae_db": 0.7635987659098031, "rms_envelope_nrmse": 0.16710946559272283, "envelope_correlation": 0.9678291446947767, "rms_gain_error_db": -0.1829814552887588, "reference_mean_power": 0.010699017072596153, "output_mean_power": 0.010257599420435172}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 4.048774453403744e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014613887448305467, "output_power": 0.0013850430579101086, "reference_active": true, "delta_db": -0.23302483358255552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0008261048603288555, "output_power": 0.0007272599664783157, "reference_active": true, "delta_db": -0.5534549571463581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00047302858802070836, "output_power": 0.00024675342771416183, "reference_active": true, "delta_db": -2.8262419433406043}], "active_band_count": 3, "active_band_mae_db": 1.2042405780231726}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00037792225909995443, "log_rms_mae_db": 0.4524390782980965, "rms_envelope_nrmse": 0.08252258239400814, "envelope_correlation": 0.9918604959098112, "rms_gain_error_db": -0.1829858889270232, "reference_mean_power": 0.01079168973829664, "output_mean_power": 0.010346438055524959}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 3.4667222612961444e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016650159461876711, "output_power": 0.0014907575434280342, "reference_active": true, "delta_db": -0.4801138157638889}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005334276556675961, "output_power": 0.00041736665813684164, "reference_active": true, "delta_db": -1.0655777607377301}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.906803616368198e-05, "output_power": 1.7410646394784514e-05, "reference_active": true, "delta_db": -3.5100668584420545}], "active_band_count": 3, "active_band_mae_db": 1.685252811647891}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00018013905998130842, "log_rms_mae_db": 1.3861700366617402, "rms_envelope_nrmse": 0.11152628162117094, "envelope_correlation": 0.9883838085134776, "rms_gain_error_db": -0.17447108781997903, "reference_mean_power": 0.002043548171012202, "output_mean_power": 0.001963078970429383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.1637162379955263e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006991910792272859, "output_power": 0.0006397721533973265, "reference_active": true, "delta_db": -0.3857054548716764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001302332633123258, "output_power": 0.00010735424341139579, "reference_active": true, "delta_db": -0.8390270745179245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.350140195021887e-05, "output_power": 2.082115049972357e-05, "reference_active": true, "delta_db": -2.0655825821593705}], "active_band_count": 3, "active_band_mae_db": 1.0967717038496572}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00014142899773968785, "log_rms_mae_db": 0.8689853017396794, "rms_envelope_nrmse": 0.06672118963612272, "envelope_correlation": 0.9959043043376116, "rms_gain_error_db": -0.1744755973992182, "reference_mean_power": 0.002061256155357224, "output_mean_power": 0.0019800876079092305}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.0272195933213497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006464643086637218, "output_power": 0.0005869754147264847, "reference_active": true, "delta_db": -0.41924641105826177}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.880640785962183e-05, "output_power": 5.494972391566108e-05, "reference_active": true, "delta_db": -0.9766337071614575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.140839173745677e-06, "output_power": 1.1173757146605098e-06, "reference_active": true, "delta_db": -2.823848149613645}], "active_band_count": 3, "active_band_mae_db": 1.406576089277788}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00022296985761207562, "log_rms_mae_db": 1.2633688354028567, "rms_envelope_nrmse": 0.10168973958289189, "envelope_correlation": 0.9932317975675976, "rms_gain_error_db": -0.10565410725266643, "reference_mean_power": 0.0018916055464765064, "output_mean_power": 0.0018461422778594792}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.550199090959336e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001029129669204316, "output_power": 0.0009467721696941429, "reference_active": true, "delta_db": -0.3622461544941285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.51753044024429e-05, "output_power": 5.244062608479391e-05, "reference_active": true, "delta_db": -0.22076869786220668}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2165950116970791e-05, "output_power": 1.959776849208526e-05, "reference_active": true, "delta_db": 2.0706059162390207}], "active_band_count": 3, "active_band_mae_db": 0.884540256198452}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00021618630891631995, "log_rms_mae_db": 0.7996163852631227, "rms_envelope_nrmse": 0.057494147437707636, "envelope_correlation": 0.9978409352784784, "rms_gain_error_db": -0.10566020329187739, "reference_mean_power": 0.0019079864948334626, "output_mean_power": 0.0018621269090545781}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.466045308126094e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009690551708671025, "output_power": 0.0008936044773260133, "reference_active": true, "delta_db": -0.3520316717809236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151940036741328e-05, "output_power": 2.1568585639131852e-05, "reference_active": true, "delta_db": 0.009915014998883375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.641667021707976e-07, "output_power": 7.905167526385657e-07, "reference_active": true, "delta_db": 0.1472296813829933}], "active_band_count": 3, "active_band_mae_db": 0.16972545605426678}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 3.608035305023674e-05, "log_rms_mae_db": 1.487508817413126, "rms_envelope_nrmse": 0.1741089109710799, "envelope_correlation": 0.9746356185010852, "rms_gain_error_db": -0.10876567754532582, "reference_mean_power": 6.674409144257214e-05, "output_mean_power": 6.50932952834946e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 5.317396933477194e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.505970371058983e-05, "output_power": 3.2905550624459626e-05, "reference_active": true, "delta_db": -0.27539079096439634}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.310128917388118e-06, "output_power": 2.3520476437115134e-06, "reference_active": true, "delta_db": 0.07809898186347705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.893247149634755e-07, "output_power": 1.5365330902595648e-06, "reference_active": true, "delta_db": 1.912030589073773}], "active_band_count": 3, "active_band_mae_db": 0.7551734539672155}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 2.62382889528218e-05, "log_rms_mae_db": 0.975538199597204, "rms_envelope_nrmse": 0.10692335083654769, "envelope_correlation": 0.9896528876812744, "rms_gain_error_db": -0.10880535605450893, "reference_mean_power": 6.730512481108481e-05, "output_mean_power": 6.563985278236863e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 4.920971050071525e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.323536847986801e-05, "output_power": 3.153507281725144e-05, "reference_active": true, "delta_db": -0.22806659996538217}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2494288246616935e-06, "output_power": 1.1223876019914547e-06, "reference_active": true, "delta_db": -0.4656866023159174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.101103941251684e-08, "output_power": 1.0245680977005848e-07, "reference_active": true, "delta_db": 1.019966245617697}], "active_band_count": 3, "active_band_mae_db": 0.5712398159663322}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9122204001216223, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ad82944d0b0688b54a572d0bad9e3f9d43b0a042abc7f64b28170021d6d272e7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9925875817368057, "standard_deviation": 0.0017154080003228358, "minimum": 0.9890643753525278, "maximum": 0.9946595626169559, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.268900639702005, "standard_deviation": 0.032396261845754594, "minimum": 1.2182312341231996, "maximum": 1.3085127668299796, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6117911629256219, "standard_deviation": 0.30204318611942277, "minimum": 0.1994381164447494, "maximum": 1.160924280965873, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10758333298308818, "standard_deviation": 0.012274729833416684, "minimum": 0.0904197365712525, "maximum": 0.1291808204021943, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9978547244498273, "standard_deviation": 0.0004339584788783964, "minimum": 0.9970817766789232, "maximum": 0.9983958638953813, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7998156969876993, "standard_deviation": 0.02305817110484402, "minimum": 0.7585684167242349, "maximum": 0.8212796863932589, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5117177418860482, "standard_deviation": 0.29487720736410655, "minimum": 0.15469732753543522, "maximum": 0.9078115276140831, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060944069645761115, "standard_deviation": 0.008364425600607598, "minimum": 0.0520253363642725, "maximum": 0.0780346446622668, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9696475207127259, "standard_deviation": 0.0026619682276468274, "minimum": 0.9656796652634827, "maximum": 0.9746356185010852, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5107546110689942, "standard_deviation": 0.03723884447346216, "minimum": 1.4582799005657991, "maximum": 1.581102055581683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7421388267751388, "standard_deviation": 0.21120449985415657, "minimum": 0.4038395805408616, "maximum": 1.0322092411832717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19047476886723697, "standard_deviation": 0.008500403864552232, "minimum": 0.1741089109710799, "maximum": 0.20270033460976142, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9856942981934556, "standard_deviation": 0.0022046764260621856, "minimum": 0.9827689342433144, "maximum": 0.9896528876812744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0391522177373176, "standard_deviation": 0.0405701443192378, "minimum": 0.975538199597204, "maximum": 1.0869705279560877, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6317848054180067, "standard_deviation": 0.20730808817626992, "minimum": 0.2660241851185453, "maximum": 0.8541682860570082, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1264030797788818, "standard_deviation": 0.01090729758791068, "minimum": 0.10692335083654769, "maximum": 0.1401691923993885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9884539959299223, "standard_deviation": 0.0015374066107956396, "minimum": 0.9859269525712199, "maximum": 0.990348031733985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3435059086365537, "standard_deviation": 0.03704976330153933, "minimum": 1.3053346793765277, "maximum": 1.3973273858958626, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.683023499698532, "standard_deviation": 0.36068556817432573, "minimum": 0.24746390434883647, "maximum": 1.2287856136522153, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11046194062567297, "standard_deviation": 0.007656999467609007, "minimum": 0.10146207795443657, "maximum": 0.1218049776207567, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959282299294936, "standard_deviation": 0.000861744410189389, "minimum": 0.9943122751502937, "maximum": 0.996761422560423, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8457532810923051, "standard_deviation": 0.07016391267741194, "minimum": 0.7506561370508528, "maximum": 0.9870220839666878, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8120401065955349, "standard_deviation": 0.501329330775026, "minimum": 0.1303292865277947, "maximum": 1.5641630455979563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06445830358032494, "standard_deviation": 0.008324291574162082, "minimum": 0.056144999025758494, "maximum": 0.07970516577920853, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3494237761719598, "standard_deviation": 0.130980173079773, "minimum": 0.18963029722279245, "maximum": 0.5408015680530299, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.426943357694423, "standard_deviation": 0.00845193488128825, "minimum": 5.415842617926454, "maximum": 5.43755006912588, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 1.0129391151945526, "standard_deviation": 0.19509132968460263, "minimum": 0.7027624570942617, "maximum": 1.290081306516348, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10677180474256359, "standard_deviation": 0.22440001443524352, "minimum": -0.377959834915546, "maximum": 0.33768003223513193, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9695610136092114, "standard_deviation": 0.020737393154322447, "minimum": 0.9341818477335804, "maximum": 0.9905264206897096, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7593204105267416, "standard_deviation": 0.02716536825399728, "minimum": 0.7148867685184976, "maximum": 0.8140867163649441, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.3461174291342266, "standard_deviation": 0.871596464807175, "minimum": 0.3091322143640934, "maximum": 3.0265258539935913, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15848679306662455, "standard_deviation": 0.05687890636397755, "minimum": 0.09375843446984519, "maximum": 0.25794256758148754, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9882763943249573, "standard_deviation": 0.009595631370655411, "minimum": 0.9694919538124483, "maximum": 0.9966839521856665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4327028038388067, "standard_deviation": 0.02647938959174832, "minimum": 0.3968676022242449, "maximum": 0.47822779947172106, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.5995926939024594, "standard_deviation": 1.1275330961556402, "minimum": 0.25984264200614743, "maximum": 3.640062933158339, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09459169421098515, "standard_deviation": 0.04165066586376055, "minimum": 0.05350418629953443, "maximum": 0.17221293139990104, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0034", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.310403823852539, "predicted_mos": 2.310403823852539, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "481128373e580e434c193f691341a51d526aee11f6df4cd0601d2b0ad545a646", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.48390890005976, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7981476160415488, "全帯域_10ms_rms_envelope_nrmse": 0.17503418753145664, "全帯域_10ms_envelope_correlation": 0.9650139488613522, "全帯域_10ms_modulation_mae_db": 1.1491024217948447, "全帯域_40ms_log_rms_mae_db": 0.4758129264285665, "全帯域_40ms_rms_envelope_nrmse": 0.07464306498314568, "全帯域_40ms_envelope_correlation": 0.9930124029331292, "全帯域_40ms_modulation_mae_db": 1.168331718219065, "300–1000 Hz_10ms_log_rms_mae_db": 1.3130512889868344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11244678512510299, "300–1000 Hz_10ms_envelope_correlation": 0.9881616932625481, "300–1000 Hz_10ms_modulation_mae_db": 1.1134956898189612, "300–1000 Hz_40ms_log_rms_mae_db": 0.7810527449281041, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06460427547071745, "300–1000 Hz_40ms_envelope_correlation": 0.9961883991744449, "300–1000 Hz_40ms_modulation_mae_db": 1.471649707907562, "1000–3000 Hz_10ms_log_rms_mae_db": 1.216982207978929, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10816848946127303, "1000–3000 Hz_10ms_envelope_correlation": 0.9923499696906795, "1000–3000 Hz_10ms_modulation_mae_db": 0.8796314751604043, "1000–3000 Hz_40ms_log_rms_mae_db": 0.734314467688715, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06268200752613484, "1000–3000 Hz_40ms_envelope_correlation": 0.9973930668663835, "1000–3000 Hz_40ms_modulation_mae_db": 0.24908699727503836, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4396084857114246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17193038110979392, "3000–6000 Hz_10ms_envelope_correlation": 0.9755318552773758, "3000–6000 Hz_10ms_modulation_mae_db": 0.6733512525058433, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9105053174284593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09990942906708423, "3000–6000 Hz_40ms_envelope_correlation": 0.9915973454797387, "3000–6000 Hz_40ms_modulation_mae_db": 0.5054453880353266, "mr_stft_db": 5.384777057418769, "band_mae_db": 0.22250017402142927, "rms_gain_error_db": -0.04107824871385227, "output_peak": 0.7346132278660205}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8005285589783944, "全帯域_10ms_rms_envelope_nrmse": 0.14794485604625182, "全帯域_10ms_envelope_correlation": 0.9749598898885967, "全帯域_10ms_modulation_mae_db": 0.7650925374091756, "全帯域_40ms_log_rms_mae_db": 0.44257455844529153, "全帯域_40ms_rms_envelope_nrmse": 0.06813800120652644, "全帯域_40ms_envelope_correlation": 0.994471414541364, "全帯域_40ms_modulation_mae_db": 1.2051573517035261, "300–1000 Hz_10ms_log_rms_mae_db": 1.3394691321367842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11801055221282253, "300–1000 Hz_10ms_envelope_correlation": 0.9871619791752779, "300–1000 Hz_10ms_modulation_mae_db": 0.9391014059433044, "300–1000 Hz_40ms_log_rms_mae_db": 0.8821988577747568, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07516992545359025, "300–1000 Hz_40ms_envelope_correlation": 0.9949893135537311, "300–1000 Hz_40ms_modulation_mae_db": 1.1059846498810355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.229651830396807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1022781606236674, "1000–3000 Hz_10ms_envelope_correlation": 0.9935224408282717, "1000–3000 Hz_10ms_modulation_mae_db": 0.3251849803662461, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7902280857162277, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060966648389615304, "1000–3000 Hz_40ms_envelope_correlation": 0.99804835689426, "1000–3000 Hz_40ms_modulation_mae_db": 0.40284393297979454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.459378265800292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18550210141759227, "3000–6000 Hz_10ms_envelope_correlation": 0.9721634145661092, "3000–6000 Hz_10ms_modulation_mae_db": 0.7057989859033148, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9711441182434856, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12074637157881464, "3000–6000 Hz_40ms_envelope_correlation": 0.9884952091410981, "3000–6000 Hz_40ms_modulation_mae_db": 1.2867729924079991, "mr_stft_db": 5.3754492518071535, "band_mae_db": 0.27555308271242, "rms_gain_error_db": -0.1565042008735748, "output_peak": 1.080187075202333}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 16, "at_or_above_full_scale_fraction": 0.0002861230329041488, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0002861230329041488, "longest_full_scale_run_samples": 16, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7624464831518752, "全帯域_10ms_rms_envelope_nrmse": 0.18558447922223617, "全帯域_10ms_envelope_correlation": 0.9643844023057179, "全帯域_10ms_modulation_mae_db": 1.1206795636751905, "全帯域_40ms_log_rms_mae_db": 0.42574364457875385, "全帯域_40ms_rms_envelope_nrmse": 0.1179746639183957, "全帯域_40ms_envelope_correlation": 0.985058620677269, "全帯域_40ms_modulation_mae_db": 0.9291171306219805, "300–1000 Hz_10ms_log_rms_mae_db": 1.2583283262873066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10035631359917038, "300–1000 Hz_10ms_envelope_correlation": 0.9903242448729767, "300–1000 Hz_10ms_modulation_mae_db": 0.22072400354627372, "300–1000 Hz_40ms_log_rms_mae_db": 0.7325459613562583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052800513890598214, "300–1000 Hz_40ms_envelope_correlation": 0.9971062497441847, "300–1000 Hz_40ms_modulation_mae_db": 0.09311485205344144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.21743222898605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08842305301680181, "1000–3000 Hz_10ms_envelope_correlation": 0.9948851594849084, "1000–3000 Hz_10ms_modulation_mae_db": 0.16263102400318843, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7193377574290617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04977068184702944, "1000–3000 Hz_40ms_envelope_correlation": 0.9983981547173173, "1000–3000 Hz_40ms_modulation_mae_db": 0.6901324493364847, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4600156504160435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19042687877013287, "3000–6000 Hz_10ms_envelope_correlation": 0.9697215241731825, "3000–6000 Hz_10ms_modulation_mae_db": 0.7766886505939773, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9427217123653587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11401310406238045, "3000–6000 Hz_40ms_envelope_correlation": 0.9885935816560777, "3000–6000 Hz_40ms_modulation_mae_db": 0.6014631627592567, "mr_stft_db": 5.366615857728303, "band_mae_db": 0.32771459399008424, "rms_gain_error_db": 0.2317024948655942, "output_peak": 1.2074954044216741}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 30, "at_or_above_full_scale_fraction": 0.000536480686695279, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.000536480686695279, "longest_full_scale_run_samples": 30, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7704612983423852, "全帯域_10ms_rms_envelope_nrmse": 0.22827242472120363, "全帯域_10ms_envelope_correlation": 0.9393443000869979, "全帯域_10ms_modulation_mae_db": 2.9116846633702758, "全帯域_40ms_log_rms_mae_db": 0.4767608249665898, "全帯域_40ms_rms_envelope_nrmse": 0.15019805972545647, "全帯域_40ms_envelope_correlation": 0.9730186159018706, "全帯域_40ms_modulation_mae_db": 3.3790703471852033, "300–1000 Hz_10ms_log_rms_mae_db": 1.3058878496364525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10025796352239913, "300–1000 Hz_10ms_envelope_correlation": 0.9903749412921986, "300–1000 Hz_10ms_modulation_mae_db": 0.42215411052716867, "300–1000 Hz_40ms_log_rms_mae_db": 0.7426487311024804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05461475945022255, "300–1000 Hz_40ms_envelope_correlation": 0.9968868098222273, "300–1000 Hz_40ms_modulation_mae_db": 0.5460661401053316, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2106423245698001, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13823536054531033, "1000–3000 Hz_10ms_envelope_correlation": 0.9876391953145015, "1000–3000 Hz_10ms_modulation_mae_db": 1.1763292655789377, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7420001714880923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07295761406903944, "1000–3000 Hz_40ms_envelope_correlation": 0.9965011191018467, "1000–3000 Hz_40ms_modulation_mae_db": 0.9118501523067217, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4682021456136518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1986474931390426, "3000–6000 Hz_10ms_envelope_correlation": 0.9674795314811593, "3000–6000 Hz_10ms_modulation_mae_db": 0.88528871031652, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9780192880609426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12212385609420788, "3000–6000 Hz_40ms_envelope_correlation": 0.9878418786133292, "3000–6000 Hz_40ms_modulation_mae_db": 0.624471244389995, "mr_stft_db": 5.3878754507468685, "band_mae_db": 0.6128720101254576, "rms_gain_error_db": -0.3950367270360615, "output_peak": 0.7111930929265787}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8347788387629704, "全帯域_10ms_rms_envelope_nrmse": 0.19272623475220685, "全帯域_10ms_envelope_correlation": 0.9569929978117119, "全帯域_10ms_modulation_mae_db": 2.3332153777342373, "全帯域_40ms_log_rms_mae_db": 0.4719280112218176, "全帯域_40ms_rms_envelope_nrmse": 0.10556498345018099, "全帯域_40ms_envelope_correlation": 0.9868097567322396, "全帯域_40ms_modulation_mae_db": 2.7754144589935836, "300–1000 Hz_10ms_log_rms_mae_db": 1.2680652065691977, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10824875074526075, "300–1000 Hz_10ms_envelope_correlation": 0.9888814995074335, "300–1000 Hz_10ms_modulation_mae_db": 0.5119561542620307, "300–1000 Hz_40ms_log_rms_mae_db": 0.694032114758708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05558239652360773, "300–1000 Hz_40ms_envelope_correlation": 0.9968372308602518, "300–1000 Hz_40ms_modulation_mae_db": 0.611508972173976, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1993355932376515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10557327352164558, "1000–3000 Hz_10ms_envelope_correlation": 0.992659710703513, "1000–3000 Hz_10ms_modulation_mae_db": 0.5010825483483324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6900756301246966, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054259116013084155, "1000–3000 Hz_40ms_envelope_correlation": 0.998066766242031, "1000–3000 Hz_40ms_modulation_mae_db": 0.38873571655833206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4576771217795617, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18781701619776905, "3000–6000 Hz_10ms_envelope_correlation": 0.9715934367543048, "3000–6000 Hz_10ms_modulation_mae_db": 0.765836532228252, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9622624277570065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12320506845291168, "3000–6000 Hz_40ms_envelope_correlation": 0.9882177283606228, "3000–6000 Hz_40ms_modulation_mae_db": 0.8671307072077231, "mr_stft_db": 5.383779845086494, "band_mae_db": 0.3871831441233206, "rms_gain_error_db": -0.25069038391718934, "output_peak": 0.8754026945383254}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7233644266697451, "全帯域_10ms_rms_envelope_nrmse": 0.11219743348871057, "全帯域_10ms_envelope_correlation": 0.9858235601848371, "全帯域_10ms_modulation_mae_db": 0.31781818944421164, "全帯域_40ms_log_rms_mae_db": 0.39887261253363254, "全帯域_40ms_rms_envelope_nrmse": 0.059239573769561224, "全帯域_40ms_envelope_correlation": 0.9956940826603508, "全帯域_40ms_modulation_mae_db": 0.3673329447602082, "300–1000 Hz_10ms_log_rms_mae_db": 1.2783275367520541, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1205283247736933, "300–1000 Hz_10ms_envelope_correlation": 0.9862077841610518, "300–1000 Hz_10ms_modulation_mae_db": 1.1704239332417117, "300–1000 Hz_40ms_log_rms_mae_db": 0.8014739353108686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06786888649382407, "300–1000 Hz_40ms_envelope_correlation": 0.995545918914583, "300–1000 Hz_40ms_modulation_mae_db": 1.4758946950481517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2600576717256808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11129294458046655, "1000–3000 Hz_10ms_envelope_correlation": 0.992049372414881, "1000–3000 Hz_10ms_modulation_mae_db": 0.7585369984134517, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7607913767943784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06074470765970177, "1000–3000 Hz_40ms_envelope_correlation": 0.9978553686686868, "1000–3000 Hz_40ms_modulation_mae_db": 0.7075333243230671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4885753025791286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18948637159703433, "3000–6000 Hz_10ms_envelope_correlation": 0.9700724000111902, "3000–6000 Hz_10ms_modulation_mae_db": 0.7204481803316827, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0089264537830593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11500223343386393, "3000–6000 Hz_40ms_envelope_correlation": 0.9884783738707127, "3000–6000 Hz_40ms_modulation_mae_db": 0.34712366418987645, "mr_stft_db": 5.378284898420556, "band_mae_db": 0.24347235074825038, "rms_gain_error_db": -0.03458989241045264, "output_peak": 0.9151871464579479}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7669622308175226, "全帯域_10ms_rms_envelope_nrmse": 0.10081063034701306, "全帯域_10ms_envelope_correlation": 0.9890346839311625, "全帯域_10ms_modulation_mae_db": 0.8514817192940386, "全帯域_40ms_log_rms_mae_db": 0.4247950143859068, "全帯域_40ms_rms_envelope_nrmse": 0.06780484699333582, "全帯域_40ms_envelope_correlation": 0.9951368485869984, "全帯域_40ms_modulation_mae_db": 0.858405603219118, "300–1000 Hz_10ms_log_rms_mae_db": 1.2825858527810357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10214248782210168, "300–1000 Hz_10ms_envelope_correlation": 0.99023764891901, "300–1000 Hz_10ms_modulation_mae_db": 0.570314167978922, "300–1000 Hz_40ms_log_rms_mae_db": 0.7766210158251484, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06097241859427354, "300–1000 Hz_40ms_envelope_correlation": 0.9962853634256521, "300–1000 Hz_40ms_modulation_mae_db": 0.5667435064503613, "1000–3000 Hz_10ms_log_rms_mae_db": 1.270315075558623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10145086928864787, "1000–3000 Hz_10ms_envelope_correlation": 0.993910875364673, "1000–3000 Hz_10ms_modulation_mae_db": 0.7659243483897593, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7451329060916598, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06884564026210035, "1000–3000 Hz_40ms_envelope_correlation": 0.9976297916475301, "1000–3000 Hz_40ms_modulation_mae_db": 1.0203139110197885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4943694335745825, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18032645067940758, "3000–6000 Hz_10ms_envelope_correlation": 0.9741217581819858, "3000–6000 Hz_10ms_modulation_mae_db": 1.0547620566489682, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9794645442019773, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12707702940573332, "3000–6000 Hz_40ms_envelope_correlation": 0.9873714062917105, "3000–6000 Hz_40ms_modulation_mae_db": 0.6918542642130063, "mr_stft_db": 5.372226256703628, "band_mae_db": 0.40248656521646997, "rms_gain_error_db": -0.24239891287176218, "output_peak": 1.0226586823277302}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 7, "at_or_above_full_scale_fraction": 0.00012517882689556508, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.00012517882689556508, "longest_full_scale_run_samples": 7, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7877598067735548, "全帯域_10ms_rms_envelope_nrmse": 0.2815537631504134, "全帯域_10ms_envelope_correlation": 0.9219867650569056, "全帯域_10ms_modulation_mae_db": 1.9205949433530616, "全帯域_40ms_log_rms_mae_db": 0.46512605647837363, "全帯域_40ms_rms_envelope_nrmse": 0.18593990324457707, "全帯域_40ms_envelope_correlation": 0.9646328953219544, "全帯域_40ms_modulation_mae_db": 2.366620122191471, "300–1000 Hz_10ms_log_rms_mae_db": 1.2765076448697856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11544323617178537, "300–1000 Hz_10ms_envelope_correlation": 0.9875140050865395, "300–1000 Hz_10ms_modulation_mae_db": 0.5820717169586622, "300–1000 Hz_40ms_log_rms_mae_db": 0.7467231732470166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06797893976082202, "300–1000 Hz_40ms_envelope_correlation": 0.9957473295283631, "300–1000 Hz_40ms_modulation_mae_db": 0.750615285262901, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2454829316083003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11790086069002971, "1000–3000 Hz_10ms_envelope_correlation": 0.991976043301789, "1000–3000 Hz_10ms_modulation_mae_db": 0.5558948457987322, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7197996748537556, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07747317549789802, "1000–3000 Hz_40ms_envelope_correlation": 0.9973842570142195, "1000–3000 Hz_40ms_modulation_mae_db": 0.3362919632811899, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4948926774939244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17930150205251072, "3000–6000 Hz_10ms_envelope_correlation": 0.9732090872382819, "3000–6000 Hz_10ms_modulation_mae_db": 0.34338153677708516, "3000–6000 Hz_40ms_log_rms_mae_db": 0.924455057358481, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10358068219598483, "3000–6000 Hz_40ms_envelope_correlation": 0.9907619582275797, "3000–6000 Hz_40ms_modulation_mae_db": 0.6394264723394056, "mr_stft_db": 5.365471325243699, "band_mae_db": 0.5477037264781067, "rms_gain_error_db": 0.37090525526522844, "output_peak": 1.2381936520145476}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 66, "at_or_above_full_scale_fraction": 0.0011802575107296136, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.001198140200286123, "longest_full_scale_run_samples": 66, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 55920, "output_samples": 55920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.319893455786316, "spectral_convergence": 0.3851214636012479, "spectral_convergence_status": "ok", "amplitude_floor": 7.545858591823444e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.241185209361245, "spectral_convergence": 0.36584608616499464, "spectral_convergence_status": "ok", "amplitude_floor": 3.9963548992246115e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.593252507108743, "spectral_convergence": 0.34980311659967195, "spectral_convergence_status": "ok", "amplitude_floor": 1.7986227023732554e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001975478876015244, "output_power": 0.0018388841627100371, "delta_db": -0.3111801761101978, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005013278749394044, "output_power": 0.005172077742681214, "delta_db": 0.13543191218112263, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015170067046631896, "output_power": 0.0014335255358092908, "delta_db": -0.24582066661666593, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0018605244766058998, "output_power": 0.0018417865598707999, "delta_db": -0.043960884132770175, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002538671104412511, "output_power": 0.0002377304152641021, "delta_db": -0.28521690847183945, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.356486516629775e-05, "output_power": 5.913955882919302e-05, "delta_db": -0.3133904966159796, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22250017402142927, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04107824871385227, "pitch": null, "output_rms_dbfs": -19.753852956060122, "output_peak": 0.7346132278660205, "reference_rms_dbfs": -19.712774707346266, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.384777057418769}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 55920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.0006292775937517623, "log_rms_mae_db": 0.7981476160415488, "rms_envelope_nrmse": 0.17503418753145664, "envelope_correlation": 0.9650139488613522, "rms_gain_error_db": -0.041078974757461625, "reference_mean_power": 0.010699017072596153, "output_mean_power": 0.010598294494027246}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 4.048774453403744e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014613887448305467, "output_power": 0.0015489518980349609, "reference_active": true, "delta_db": 0.2527217294116514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0008261048603288555, "output_power": 0.0008767647909859916, "reference_active": true, "delta_db": 0.25847923834971803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00047302858802070836, "output_power": 0.00024058954944830934, "reference_active": true, "delta_db": -2.9361062976231644}], "active_band_count": 3, "active_band_mae_db": 1.1491024217948447}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00037792225909995443, "log_rms_mae_db": 0.4758129264285665, "rms_envelope_nrmse": 0.07464306498314568, "envelope_correlation": 0.9930124029331292, "rms_gain_error_db": -0.04108183282397405, "reference_mean_power": 0.01079168973829664, "output_mean_power": 0.010690087686567066}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 3.4667222612961444e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016650159461876711, "output_power": 0.0017018585439255278, "reference_active": true, "delta_db": 0.09505062119198773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005334276556675961, "output_power": 0.00048185269090762615, "reference_active": true, "delta_db": -0.44161239379804995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.906803616368198e-05, "output_power": 1.9723699039649544e-05, "reference_active": true, "delta_db": -2.968332139667157}], "active_band_count": 3, "active_band_mae_db": 1.168331718219065}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00018013905998130842, "log_rms_mae_db": 1.3130512889868344, "rms_envelope_nrmse": 0.11244678512510299, "envelope_correlation": 0.9881616932625481, "rms_gain_error_db": -0.19077264368099628, "reference_mean_power": 0.002043548171012202, "output_mean_power": 0.001955724224186647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.1637162379955263e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006991910792272859, "output_power": 0.0006330945615373436, "reference_active": true, "delta_db": -0.43127295802357474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001302332633123258, "output_power": 0.00010724621174861388, "reference_active": true, "delta_db": -0.8433996243287173}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.350140195021887e-05, "output_power": 2.0820038719801766e-05, "reference_active": true, "delta_db": -2.065814487104592}], "active_band_count": 3, "active_band_mae_db": 1.1134956898189612}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00014142899773968785, "log_rms_mae_db": 0.7810527449281041, "rms_envelope_nrmse": 0.06460427547071745, "envelope_correlation": 0.9961883991744449, "rms_gain_error_db": -0.19077608015209105, "reference_mean_power": 0.002061256155357224, "output_mean_power": 0.0019726696256234313}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.0272195933213497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006464643086637218, "output_power": 0.0005819627738565719, "reference_active": true, "delta_db": -0.456493472276415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.880640785962183e-05, "output_power": 5.517162555211957e-05, "reference_active": true, "delta_db": -0.9591310544710353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.140839173745677e-06, "output_power": 1.0731281401381277e-06, "reference_active": true, "delta_db": -2.999324596975236}], "active_band_count": 3, "active_band_mae_db": 1.471649707907562}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00022296985761207562, "log_rms_mae_db": 1.216982207978929, "rms_envelope_nrmse": 0.10816848946127303, "envelope_correlation": 0.9923499696906795, "rms_gain_error_db": -0.05183478990721457, "reference_mean_power": 0.0018916055464765064, "output_mean_power": 0.0018691626736574462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.550199090959336e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001029129669204316, "output_power": 0.0009502924181731791, "reference_active": true, "delta_db": -0.34612834539137854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.51753044024429e-05, "output_power": 5.418366640206016e-05, "reference_active": true, "delta_db": -0.07876349102499249}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2165950116970791e-05, "output_power": 2.0255654951534906e-05, "reference_active": true, "delta_db": 2.214002589064842}], "active_band_count": 3, "active_band_mae_db": 0.8796314751604043}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00021618630891631995, "log_rms_mae_db": 0.734314467688715, "rms_envelope_nrmse": 0.06268200752613484, "envelope_correlation": 0.9973930668663835, "rms_gain_error_db": -0.05183996354467247, "reference_mean_power": 0.0019079864948334626, "output_mean_power": 0.001885347024989365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.466045308126094e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009690551708671025, "output_power": 0.0008976063224648987, "reference_active": true, "delta_db": -0.3326260032449532}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151940036741328e-05, "output_power": 2.2326388694245375e-05, "reference_active": true, "delta_db": 0.15988315720469487}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.641667021707976e-07, "output_power": 8.103325661178302e-07, "reference_active": true, "delta_db": 0.25475183137546703}], "active_band_count": 3, "active_band_mae_db": 0.24908699727503836}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 3.608035305023674e-05, "log_rms_mae_db": 1.4396084857114246, "rms_envelope_nrmse": 0.17193038110979392, "envelope_correlation": 0.9755318552773758, "rms_gain_error_db": -0.2945301323869986, "reference_mean_power": 6.674409144257214e-05, "output_mean_power": 6.2367712053585e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 5.317396933477194e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.505970371058983e-05, "output_power": 3.209973474772832e-05, "reference_active": true, "delta_db": -0.3830679787259552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.310128917388118e-06, "output_power": 2.2083467508452475e-06, "reference_active": true, "delta_db": -0.19568949901797608}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.893247149634755e-07, "output_power": 1.378695907959389e-06, "reference_active": true, "delta_db": 1.4412962797735986}], "active_band_count": 3, "active_band_mae_db": 0.6733512525058433}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 2.62382889528218e-05, "log_rms_mae_db": 0.9105053174284593, "rms_envelope_nrmse": 0.09990942906708423, "envelope_correlation": 0.9915973454797387, "rms_gain_error_db": -0.2946155261740867, "reference_mean_power": 6.730512481108481e-05, "output_mean_power": 6.289072211189987e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 4.920971050071525e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.323536847986801e-05, "output_power": 3.072151303775703e-05, "reference_active": true, "delta_db": -0.34157897215884825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2494288246616935e-06, "output_power": 1.0605887189715002e-06, "reference_active": true, "delta_db": -0.7116451761451764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.101103941251684e-08, "output_power": 9.012710184959066e-08, "reference_active": true, "delta_db": 0.4631120158019548}], "active_band_count": 3, "active_band_mae_db": 0.5054453880353266}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9122008648772083, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "506c1e589b8fd8af90f991bb1c83b3d3ba0498fb9a1a8a37fb362e5ddf0368b8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9923740958879022, "standard_deviation": 0.0021630388736361295, "minimum": 0.9876391953145015, "maximum": 0.9948851594849084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2312374830077304, "standard_deviation": 0.02506790105773138, "minimum": 1.1993355932376515, "maximum": 1.270315075558623, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6406519357573816, "standard_deviation": 0.3225652220914298, "minimum": 0.16263102400318843, "maximum": 1.1763292655789377, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10916537646598029, "standard_deviation": 0.014525777763606558, "minimum": 0.08842305301680181, "maximum": 0.13823536054531033, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9976596101440344, "standard_deviation": 0.0005843830584677822, "minimum": 0.9965011191018467, "maximum": 0.9983981547173173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7377100087733234, "standard_deviation": 0.030007823421061175, "minimum": 0.6900756301246966, "maximum": 0.7902280857162277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5883485558850521, "standard_deviation": 0.2850314525718836, "minimum": 0.24908699727503836, "maximum": 1.0203139110197885, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06346244890807543, "standard_deviation": 0.009272289116844499, "minimum": 0.04977068184702944, "maximum": 0.07747317549789802, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9717366259604487, "standard_deviation": 0.002602651169988538, "minimum": 0.9674795314811593, "maximum": 0.9755318552773758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.470339885371076, "standard_deviation": 0.020173782837250406, "minimum": 1.4396084857114246, "maximum": 1.4948926774939244, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7406944881632054, "standard_deviation": 0.20195258656790646, "minimum": 0.34338153677708516, "maximum": 1.0547620566489682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18542977437041042, "standard_deviation": 0.008173760546789244, "minimum": 0.17193038110979392, "maximum": 0.1986474931390426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9889196852051088, "standard_deviation": 0.0014685605993434686, "minimum": 0.9873714062917105, "maximum": 0.9915973454797387, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9596873648998463, "standard_deviation": 0.03215497051549245, "minimum": 0.9105053174284593, "maximum": 1.0089264537830593, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6954609869428237, "standard_deviation": 0.28122009902987133, "minimum": 0.34712366418987645, "maximum": 1.2867729924079991, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11570722178637262, "standard_deviation": 0.009651400468963199, "minimum": 0.09990942906708423, "maximum": 0.12707702940573332, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9886079745346295, "standard_deviation": 0.0016056223734269399, "minimum": 0.9862077841610518, "maximum": 0.9903749412921986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2902778547524314, "standard_deviation": 0.026959898264746398, "minimum": 1.2583283262873066, "maximum": 1.3394691321367842, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6912801477846293, "standard_deviation": 0.3426427460969319, "minimum": 0.22072400354627372, "maximum": 1.1704239332417117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10967930174654202, "standard_deviation": 0.008128632827527526, "minimum": 0.10025796352239913, "maximum": 0.1205283247736933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961983268779298, "standard_deviation": 0.0007372684035647514, "minimum": 0.9949893135537311, "maximum": 0.9971062497441847, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7696620667879176, "standard_deviation": 0.05632571969335737, "minimum": 0.694032114758708, "maximum": 0.8821988577747568, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.827697226110345, "standard_deviation": 0.4857956376925513, "minimum": 0.09311485205344144, "maximum": 1.4758946950481517, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06244901445470698, "standard_deviation": 0.007839658385942026, "minimum": 0.052800513890598214, "maximum": 0.07516992545359025, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.37743570592694237, "standard_deviation": 0.14139482094096353, "minimum": 0.22250017402142927, "maximum": 0.6128720101254576, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.376809992894434, "standard_deviation": 0.008383833849922645, "minimum": 5.365471325243699, "maximum": 5.3878754507468685, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.9731163719693946, "standard_deviation": 0.1991085070411305, "minimum": 0.7111930929265787, "maximum": 1.2381936520145476, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06471132696150876, "standard_deviation": 0.25712439266852766, "minimum": -0.3950367270360615, "maximum": 0.37090525526522844, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9621925685159102, "standard_deviation": 0.022741646133921534, "minimum": 0.9219867650569056, "maximum": 0.9890346839311625, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7805561574422496, "standard_deviation": 0.03291960632695393, "minimum": 0.7233644266697451, "maximum": 0.8347788387629704, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.4212086770093795, "standard_deviation": 0.881522129713897, "minimum": 0.31781818944421164, "maximum": 2.9116846633702758, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17801550115743653, "standard_deviation": 0.059432275035540376, "minimum": 0.10081063034701306, "maximum": 0.2815537631504134, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.985979329669397, "standard_deviation": 0.011495364769428345, "minimum": 0.9646328953219544, "maximum": 0.9956940826603508, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44770170612986654, "standard_deviation": 0.02913671213010714, "minimum": 0.39887261253363254, "maximum": 0.4767608249665898, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.6311812096117695, "standard_deviation": 1.0682676955156085, "minimum": 0.3673329447602082, "maximum": 3.3790703471852033, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10368788716139742, "standard_deviation": 0.04552372537498609, "minimum": 0.059239573769561224, "maximum": 0.18593990324457707, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0034", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.9766786098480225, "predicted_mos": 1.9766786098480225, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "89b563afe1956183348c9941dfa93ee75897c53adf82c677db6a5e68961effdc", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.275958599988371, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7562464698039572, "全帯域_10ms_rms_envelope_nrmse": 0.21742799010439345, "全帯域_10ms_envelope_correlation": 0.965187926435954, "全帯域_10ms_modulation_mae_db": 2.9081060038188347, "全帯域_40ms_log_rms_mae_db": 0.4116944255177821, "全帯域_40ms_rms_envelope_nrmse": 0.20066363471729165, "全帯域_40ms_envelope_correlation": 0.9686671202915008, "全帯域_40ms_modulation_mae_db": 3.323247246240456, "300–1000 Hz_10ms_log_rms_mae_db": 1.4939963615688017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10583318061997571, "300–1000 Hz_10ms_envelope_correlation": 0.9894779545169128, "300–1000 Hz_10ms_modulation_mae_db": 0.6552333649221681, "300–1000 Hz_40ms_log_rms_mae_db": 0.9926823077895404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.068409485301313, "300–1000 Hz_40ms_envelope_correlation": 0.995472701528684, "300–1000 Hz_40ms_modulation_mae_db": 1.0950036832707415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6401108198118048, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12077674855544815, "1000–3000 Hz_10ms_envelope_correlation": 0.9906691267000388, "1000–3000 Hz_10ms_modulation_mae_db": 0.9838979829568114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0425215100558063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05956868454718641, "1000–3000 Hz_40ms_envelope_correlation": 0.9981458199734056, "1000–3000 Hz_40ms_modulation_mae_db": 0.7380007794234681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.962046875306592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2698624343634184, "3000–6000 Hz_10ms_envelope_correlation": 0.9401084724080269, "3000–6000 Hz_10ms_modulation_mae_db": 1.5537376513079062, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4262169132489442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21800020040854037, "3000–6000 Hz_40ms_envelope_correlation": 0.9569135829567283, "3000–6000 Hz_40ms_modulation_mae_db": 1.2100638794911591, "mr_stft_db": 5.7236215062253395, "band_mae_db": 0.6263801957490915, "rms_gain_error_db": 0.6865355588557621, "output_peak": 1.3431069748737272}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 117, "at_or_above_full_scale_fraction": 0.002092274678111588, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0021101573676680974, "longest_full_scale_run_samples": 60, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7588508280745871, "全帯域_10ms_rms_envelope_nrmse": 0.10755471062234441, "全帯域_10ms_envelope_correlation": 0.9868270098810059, "全帯域_10ms_modulation_mae_db": 0.5587240671454134, "全帯域_40ms_log_rms_mae_db": 0.3706908235018443, "全帯域_40ms_rms_envelope_nrmse": 0.049568069194767896, "全帯域_40ms_envelope_correlation": 0.9970914089377957, "全帯域_40ms_modulation_mae_db": 0.6087316184984781, "300–1000 Hz_10ms_log_rms_mae_db": 1.4616003203255192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10946319034825934, "300–1000 Hz_10ms_envelope_correlation": 0.9888355515670298, "300–1000 Hz_10ms_modulation_mae_db": 0.44698073223347096, "300–1000 Hz_40ms_log_rms_mae_db": 0.932902134870725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06900536028289694, "300–1000 Hz_40ms_envelope_correlation": 0.9952773599715052, "300–1000 Hz_40ms_modulation_mae_db": 0.4985830124892698, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5857509242837542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11234871210649326, "1000–3000 Hz_10ms_envelope_correlation": 0.9916871395549008, "1000–3000 Hz_10ms_modulation_mae_db": 1.0086542351163605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.082200894717342, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0603752976098471, "1000–3000 Hz_40ms_envelope_correlation": 0.9976280885995754, "1000–3000 Hz_40ms_modulation_mae_db": 0.4466462672751071, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9495091983679549, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26638899453524806, "3000–6000 Hz_10ms_envelope_correlation": 0.9408477385447973, "3000–6000 Hz_10ms_modulation_mae_db": 1.1838567293888105, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4530668773905286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22233569866485114, "3000–6000 Hz_40ms_envelope_correlation": 0.9544634314184729, "3000–6000 Hz_40ms_modulation_mae_db": 0.9469723667774496, "mr_stft_db": 5.735111934618356, "band_mae_db": 0.21197753181853465, "rms_gain_error_db": -0.11036086995970369, "output_peak": 1.0050239471402562}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.78826895565093e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 5.3648068669527894e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7957339184772149, "全帯域_10ms_rms_envelope_nrmse": 0.16883776867120473, "全帯域_10ms_envelope_correlation": 0.96867021267468, "全帯域_10ms_modulation_mae_db": 2.681502773947924, "全帯域_40ms_log_rms_mae_db": 0.4865313966139817, "全帯域_40ms_rms_envelope_nrmse": 0.11497527915392675, "全帯域_40ms_envelope_correlation": 0.9859533613148264, "全帯域_40ms_modulation_mae_db": 3.2180213141609557, "300–1000 Hz_10ms_log_rms_mae_db": 1.466803858055701, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12774963743648446, "300–1000 Hz_10ms_envelope_correlation": 0.9852821436485585, "300–1000 Hz_10ms_modulation_mae_db": 1.4780264427057723, "300–1000 Hz_40ms_log_rms_mae_db": 0.9741307073426416, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08502617669229129, "300–1000 Hz_40ms_envelope_correlation": 0.9938206316406357, "300–1000 Hz_40ms_modulation_mae_db": 1.925005357098028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5818025443459387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11691692478514673, "1000–3000 Hz_10ms_envelope_correlation": 0.9910549166818081, "1000–3000 Hz_10ms_modulation_mae_db": 0.9764188538192542, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0433853526717063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08181143468807024, "1000–3000 Hz_40ms_envelope_correlation": 0.995597731342876, "1000–3000 Hz_40ms_modulation_mae_db": 0.6059775467184237, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0504340941634336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29706745709249693, "3000–6000 Hz_10ms_envelope_correlation": 0.9273783205560023, "3000–6000 Hz_10ms_modulation_mae_db": 1.562555447025641, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4966650360835536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2509037804671591, "3000–6000 Hz_40ms_envelope_correlation": 0.9427313853808642, "3000–6000 Hz_40ms_modulation_mae_db": 1.1890341064034087, "mr_stft_db": 5.739231210316493, "band_mae_db": 0.47631636746355194, "rms_gain_error_db": -0.43959360167547046, "output_peak": 0.906524166557904}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7693843821136419, "全帯域_10ms_rms_envelope_nrmse": 0.11128268192997244, "全帯域_10ms_envelope_correlation": 0.9865211817309324, "全帯域_10ms_modulation_mae_db": 0.5203696206043972, "全帯域_40ms_log_rms_mae_db": 0.44515039218381885, "全帯域_40ms_rms_envelope_nrmse": 0.06900907830456854, "全帯域_40ms_envelope_correlation": 0.9944235849232548, "全帯域_40ms_modulation_mae_db": 0.5200253881637317, "300–1000 Hz_10ms_log_rms_mae_db": 1.5601781718225476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1062258230386998, "300–1000 Hz_10ms_envelope_correlation": 0.9893128710311384, "300–1000 Hz_10ms_modulation_mae_db": 0.7486613539080452, "300–1000 Hz_40ms_log_rms_mae_db": 1.0000471550151047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06264141681003516, "300–1000 Hz_40ms_envelope_correlation": 0.9961147884231668, "300–1000 Hz_40ms_modulation_mae_db": 1.1199571615891228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6043451558030788, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11175115071967315, "1000–3000 Hz_10ms_envelope_correlation": 0.9919908840521261, "1000–3000 Hz_10ms_modulation_mae_db": 0.8730797818220039, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1020882581081834, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07236924324001213, "1000–3000 Hz_40ms_envelope_correlation": 0.9968146958779489, "1000–3000 Hz_40ms_modulation_mae_db": 0.9460253837132218, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0382832789671523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28389196110363857, "3000–6000 Hz_10ms_envelope_correlation": 0.9318325551409531, "3000–6000 Hz_10ms_modulation_mae_db": 1.8250948678568595, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4558827617718586, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22479290045621367, "3000–6000 Hz_40ms_envelope_correlation": 0.9530151873603537, "3000–6000 Hz_40ms_modulation_mae_db": 1.4610554352033283, "mr_stft_db": 5.744955078974523, "band_mae_db": 0.22264434739327763, "rms_gain_error_db": 0.08308870517986032, "output_peak": 1.21347211554204}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 26, "at_or_above_full_scale_fraction": 0.0004649499284692418, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0004649499284692418, "longest_full_scale_run_samples": 26, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.715023074546193, "全帯域_10ms_rms_envelope_nrmse": 0.1975725911788458, "全帯域_10ms_envelope_correlation": 0.9558959157800789, "全帯域_10ms_modulation_mae_db": 3.0663096089556947, "全帯域_40ms_log_rms_mae_db": 0.41616689624245834, "全帯域_40ms_rms_envelope_nrmse": 0.1552916350541837, "全帯域_40ms_envelope_correlation": 0.9714631458177756, "全帯域_40ms_modulation_mae_db": 3.617107628364949, "300–1000 Hz_10ms_log_rms_mae_db": 1.5107965489107058, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09990107671898138, "300–1000 Hz_10ms_envelope_correlation": 0.9905572666574342, "300–1000 Hz_10ms_modulation_mae_db": 0.800650504091004, "300–1000 Hz_40ms_log_rms_mae_db": 0.9734369028711173, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05891619034053261, "300–1000 Hz_40ms_envelope_correlation": 0.9966189340954058, "300–1000 Hz_40ms_modulation_mae_db": 1.2176473655879734, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6383663658424061, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1060632251711753, "1000–3000 Hz_10ms_envelope_correlation": 0.9926509655703857, "1000–3000 Hz_10ms_modulation_mae_db": 0.41269866967789604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1217799331822618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06479461968290776, "1000–3000 Hz_40ms_envelope_correlation": 0.9971794125929948, "1000–3000 Hz_40ms_modulation_mae_db": 0.46684133967006486, "3000–6000 Hz_10ms_log_rms_mae_db": 2.02873082162551, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2716407990846656, "3000–6000 Hz_10ms_envelope_correlation": 0.9387372587939212, "3000–6000 Hz_10ms_modulation_mae_db": 1.0595501794254873, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4786258476787066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2107980592210054, "3000–6000 Hz_40ms_envelope_correlation": 0.9593906249083184, "3000–6000 Hz_40ms_modulation_mae_db": 0.7207775138061908, "mr_stft_db": 5.732455420254873, "band_mae_db": 0.6352398501659485, "rms_gain_error_db": -0.4278326412865947, "output_peak": 0.8663329958760217}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7606783761563715, "全帯域_10ms_rms_envelope_nrmse": 0.12213270337182663, "全帯域_10ms_envelope_correlation": 0.9841104663517833, "全帯域_10ms_modulation_mae_db": 1.190288494223836, "全帯域_40ms_log_rms_mae_db": 0.46012103443815283, "全帯域_40ms_rms_envelope_nrmse": 0.07675400227034147, "全帯域_40ms_envelope_correlation": 0.9932119618729679, "全帯域_40ms_modulation_mae_db": 1.6268255975901067, "300–1000 Hz_10ms_log_rms_mae_db": 1.6193846933716198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.142921404602903, "300–1000 Hz_10ms_envelope_correlation": 0.980571786835761, "300–1000 Hz_10ms_modulation_mae_db": 0.9528914155989548, "300–1000 Hz_40ms_log_rms_mae_db": 1.1136438204251748, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09415123980082092, "300–1000 Hz_40ms_envelope_correlation": 0.9912734526885945, "300–1000 Hz_40ms_modulation_mae_db": 1.2432062482386457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6352513808192328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1140229377220969, "1000–3000 Hz_10ms_envelope_correlation": 0.991985611521579, "1000–3000 Hz_10ms_modulation_mae_db": 0.36833660811290664, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0834465644643367, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06704557166069665, "1000–3000 Hz_40ms_envelope_correlation": 0.9977519540966163, "1000–3000 Hz_40ms_modulation_mae_db": 0.46853737176039556, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9845491892606548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35114206883471616, "3000–6000 Hz_10ms_envelope_correlation": 0.8999971788590043, "3000–6000 Hz_10ms_modulation_mae_db": 2.7208436789269954, "3000–6000 Hz_40ms_log_rms_mae_db": 1.51311507073789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3026484133359492, "3000–6000 Hz_40ms_envelope_correlation": 0.9177527624639307, "3000–6000 Hz_40ms_modulation_mae_db": 3.0040168154268607, "mr_stft_db": 5.754013596332727, "band_mae_db": 0.23465872920490508, "rms_gain_error_db": 0.05195190329322254, "output_peak": 1.3008837090770864}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 50, "at_or_above_full_scale_fraction": 0.000894134477825465, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.000894134477825465, "longest_full_scale_run_samples": 47, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7455624278143508, "全帯域_10ms_rms_envelope_nrmse": 0.19020843708286828, "全帯域_10ms_envelope_correlation": 0.9580784368465683, "全帯域_10ms_modulation_mae_db": 1.4983414723886594, "全帯域_40ms_log_rms_mae_db": 0.3932086097231586, "全帯域_40ms_rms_envelope_nrmse": 0.09204273089179667, "全帯域_40ms_envelope_correlation": 0.989932704371648, "全帯域_40ms_modulation_mae_db": 1.6018981208377292, "300–1000 Hz_10ms_log_rms_mae_db": 1.6283762015159169, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11924203810559579, "300–1000 Hz_10ms_envelope_correlation": 0.9864763829020059, "300–1000 Hz_10ms_modulation_mae_db": 0.3136955743642127, "300–1000 Hz_40ms_log_rms_mae_db": 1.0772770483997327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08166476822064976, "300–1000 Hz_40ms_envelope_correlation": 0.9933515930837238, "300–1000 Hz_40ms_modulation_mae_db": 0.5266604747405034, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6595061600105483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11231007929471934, "1000–3000 Hz_10ms_envelope_correlation": 0.9917006530956376, "1000–3000 Hz_10ms_modulation_mae_db": 0.19367655454080737, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1312710795789431, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06054953015244968, "1000–3000 Hz_40ms_envelope_correlation": 0.997506521609974, "1000–3000 Hz_40ms_modulation_mae_db": 0.12259422557755928, "3000–6000 Hz_10ms_log_rms_mae_db": 2.115594268857613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29299318197646995, "3000–6000 Hz_10ms_envelope_correlation": 0.9303285992355772, "3000–6000 Hz_10ms_modulation_mae_db": 1.6125524744677702, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5752403018493275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2230419978390829, "3000–6000 Hz_40ms_envelope_correlation": 0.9556509800191995, "3000–6000 Hz_40ms_modulation_mae_db": 1.1823993358818286, "mr_stft_db": 5.745298128865275, "band_mae_db": 0.23519115327650775, "rms_gain_error_db": -0.20775507835574314, "output_peak": 1.0084561573032707}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.78826895565093e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.78826895565093e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7616851920285478, "全帯域_10ms_rms_envelope_nrmse": 0.09842678131105388, "全帯域_10ms_envelope_correlation": 0.9892082049032491, "全帯域_10ms_modulation_mae_db": 0.3081066546986538, "全帯域_40ms_log_rms_mae_db": 0.4482508166923661, "全帯域_40ms_rms_envelope_nrmse": 0.05682440923026054, "全帯域_40ms_envelope_correlation": 0.9961100411714771, "全帯域_40ms_modulation_mae_db": 0.5597776254515532, "300–1000 Hz_10ms_log_rms_mae_db": 1.5796245097883834, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11188573842885635, "300–1000 Hz_10ms_envelope_correlation": 0.9881896867464199, "300–1000 Hz_10ms_modulation_mae_db": 0.8226253885056417, "300–1000 Hz_40ms_log_rms_mae_db": 1.0333937959582717, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06377603663859475, "300–1000 Hz_40ms_envelope_correlation": 0.9960920448819893, "300–1000 Hz_40ms_modulation_mae_db": 1.1785243115327813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6463592881072329, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10903357221016463, "1000–3000 Hz_10ms_envelope_correlation": 0.9922693189232463, "1000–3000 Hz_10ms_modulation_mae_db": 0.49660134404402045, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0820611474838684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06283481475145192, "1000–3000 Hz_40ms_envelope_correlation": 0.9975342109691165, "1000–3000 Hz_40ms_modulation_mae_db": 0.257710394981846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9604633872542896, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29065880662420224, "3000–6000 Hz_10ms_envelope_correlation": 0.9334967749366515, "3000–6000 Hz_10ms_modulation_mae_db": 1.920493393108763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4552603083763758, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2448049613441503, "3000–6000 Hz_40ms_envelope_correlation": 0.9482330406323262, "3000–6000 Hz_40ms_modulation_mae_db": 1.9391809804796498, "mr_stft_db": 5.733756691822701, "band_mae_db": 0.22945129571433232, "rms_gain_error_db": 0.04572966587637169, "output_peak": 1.044773630615583}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 7, "at_or_above_full_scale_fraction": 0.00012517882689556508, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.00012517882689556508, "longest_full_scale_run_samples": 4, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 55920, "output_samples": 55920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.669668770047079, "spectral_convergence": 0.44890222344091996, "spectral_convergence_status": "ok", "amplitude_floor": 7.545858591823444e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.644023596517655, "spectral_convergence": 0.4421481771400921, "spectral_convergence_status": "ok", "amplitude_floor": 3.9963548992246115e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.857172152111287, "spectral_convergence": 0.38063389063137304, "spectral_convergence_status": "ok", "amplitude_floor": 1.7986227023732554e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001975478876015244, "output_power": 0.004003547819253038, "delta_db": 3.067726300928718, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005013278749394044, "output_power": 0.004991581274900583, "delta_db": -0.018837061575878705, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015170067046631896, "output_power": 0.0014437532764833907, "delta_db": -0.21494517393424428, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0018605244766058998, "output_power": 0.0017555792498935537, "delta_db": -0.25214948816972604, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002538671104412511, "output_power": 0.0002526485639846981, "delta_db": -0.020896057109185275, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.356486516629775e-05, "output_power": 6.63116507632824e-05, "delta_db": 0.18372709277679683, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6263801957490915, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6865355588557621, "pitch": null, "output_rms_dbfs": -19.026239148490507, "output_peak": 1.3431069748737272, "reference_rms_dbfs": -19.712774707346266, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7236215062253395}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 55920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.0006292775937517623, "log_rms_mae_db": 0.7562464698039572, "rms_envelope_nrmse": 0.21742799010439345, "envelope_correlation": 0.965187926435954, "rms_gain_error_db": 0.6865362591548474, "reference_mean_power": 0.010699017072596153, "output_mean_power": 0.012531339799011195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 4.048774453403744e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014613887448305467, "output_power": 0.0021767551728871896, "reference_active": true, "delta_db": 1.7304382698068046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0008261048603288555, "output_power": 0.0016940923836732285, "reference_active": true, "delta_db": 3.1190191264264415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00047302858802070836, "output_power": 0.001154445539669676, "reference_active": true, "delta_db": 3.8748606152232563}], "active_band_count": 3, "active_band_mae_db": 2.9081060038188347}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00037792225909995443, "log_rms_mae_db": 0.4116944255177821, "rms_envelope_nrmse": 0.20066363471729165, "envelope_correlation": 0.9686671202915008, "rms_gain_error_db": 0.6865385747162605, "reference_mean_power": 0.01079168973829664, "output_mean_power": 0.012639890403243717}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 3.4667222612961444e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016650159461876711, "output_power": 0.002899001047877913, "reference_active": true, "delta_db": 2.4082997514928444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005334276556675961, "output_power": 0.0011799477235386344, "reference_active": true, "delta_db": 3.447872387033125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.906803616368198e-05, "output_power": 0.00010073457096220728, "reference_active": true, "delta_db": 4.113569600195397}], "active_band_count": 3, "active_band_mae_db": 3.323247246240456}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00018013905998130842, "log_rms_mae_db": 1.4939963615688017, "rms_envelope_nrmse": 0.10583318061997571, "envelope_correlation": 0.9894779545169128, "rms_gain_error_db": -0.09425672190352477, "reference_mean_power": 0.002043548171012202, "output_mean_power": 0.0019996740339767714}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.1637162379955263e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006991910792272859, "output_power": 0.0006483972901858442, "reference_active": true, "delta_db": -0.3275468747131615}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001302332633123258, "output_power": 9.869993830679698e-05, "reference_active": true, "delta_db": -1.2040504179270344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.350140195021887e-05, "output_power": 3.0314659976487413e-05, "reference_active": true, "delta_db": -0.43410280212630825}], "active_band_count": 3, "active_band_mae_db": 0.6552333649221681}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00014142899773968785, "log_rms_mae_db": 0.9926823077895404, "rms_envelope_nrmse": 0.068409485301313, "envelope_correlation": 0.995472701528684, "rms_gain_error_db": -0.09426723891832103, "reference_mean_power": 0.002061256155357224, "output_mean_power": 0.0020169969507652244}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.0272195933213497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006464643086637218, "output_power": 0.0005931487065559208, "reference_active": true, "delta_db": -0.3738096475311806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.880640785962183e-05, "output_power": 5.19226328336337e-05, "reference_active": true, "delta_db": -1.222721793920717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.140839173745677e-06, "output_power": 1.4512294472512138e-06, "reference_active": true, "delta_db": -1.6884796083603273}], "active_band_count": 3, "active_band_mae_db": 1.0950036832707415}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00022296985761207562, "log_rms_mae_db": 1.6401108198118048, "rms_envelope_nrmse": 0.12077674855544815, "envelope_correlation": 0.9906691267000388, "rms_gain_error_db": -0.24484974714587704, "reference_mean_power": 0.0018916055464765064, "output_mean_power": 0.0017879097966262592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.550199090959336e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001029129669204316, "output_power": 0.0009469682179548218, "reference_active": true, "delta_db": -0.36134695332950184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.51753044024429e-05, "output_power": 5.887145321832699e-05, "reference_active": true, "delta_db": 0.2816001848521382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2165950116970791e-05, "output_power": 2.070240080925354e-05, "reference_active": true, "delta_db": 2.3087468106887945}], "active_band_count": 3, "active_band_mae_db": 0.9838979829568114}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00021618630891631995, "log_rms_mae_db": 1.0425215100558063, "rms_envelope_nrmse": 0.05956868454718641, "envelope_correlation": 0.9981458199734056, "rms_gain_error_db": -0.24485368083797793, "reference_mean_power": 0.0019079864948334626, "output_mean_power": 0.0018033911258360073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.466045308126094e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009690551708671025, "output_power": 0.0008989709696503477, "reference_active": true, "delta_db": -0.3260283593921961}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151940036741328e-05, "output_power": 2.4370681836066945e-05, "reference_active": true, "delta_db": 0.540375142649959}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.641667021707976e-07, "output_power": 1.0421926198037283e-06, "reference_active": true, "delta_db": 1.3475988362282494}], "active_band_count": 3, "active_band_mae_db": 0.7380007794234681}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 3.608035305023674e-05, "log_rms_mae_db": 1.962046875306592, "rms_envelope_nrmse": 0.2698624343634184, "envelope_correlation": 0.9401084724080269, "rms_gain_error_db": 0.12997172641230975, "reference_mean_power": 6.674409144257214e-05, "output_mean_power": 6.877173772233923e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 5.317396933477194e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.505970371058983e-05, "output_power": 3.580898239864551e-05, "reference_active": true, "delta_db": 0.09183738076278535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.310128917388118e-06, "output_power": 3.221707317750967e-06, "reference_active": true, "delta_db": 1.4444986705779572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.893247149634755e-07, "output_power": 2.0315454780663053e-06, "reference_active": true, "delta_db": 3.1248769025829763}], "active_band_count": 3, "active_band_mae_db": 1.5537376513079062}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 2.62382889528218e-05, "log_rms_mae_db": 1.4262169132489442, "rms_envelope_nrmse": 0.21800020040854037, "envelope_correlation": 0.9569135829567283, "rms_gain_error_db": 0.13002361293586906, "reference_mean_power": 6.730512481108481e-05, "output_mean_power": 6.935064350386348e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 4.920971050071525e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.323536847986801e-05, "output_power": 3.4810383152853216e-05, "reference_active": true, "delta_db": 0.2010830537603459}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2494288246616935e-06, "output_power": 1.660000878222795e-06, "reference_active": true, "delta_db": 1.2339679678642705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.101103941251684e-08, "output_power": 1.3429451221637483e-07, "reference_active": true, "delta_db": 2.195140616848861}], "active_band_count": 3, "active_band_mae_db": 1.2100638794911591}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9057243482551227, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d13ebadea26ff5b601ad074dfabc4e805523469c0fea6a8a8e9338ec8fcd52f9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9917510770124653, "standard_deviation": 0.0006383879475915014, "minimum": 0.9906691267000388, "maximum": 0.9926509655703857, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6239365798779994, "standard_deviation": 0.0292356619230605, "minimum": 1.5818025443459387, "maximum": 1.6595061600105483, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6641705037612575, "standard_deviation": 0.32998627187635876, "minimum": 0.19367655454080737, "maximum": 1.0086542351163605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11290291882061468, "standard_deviation": 0.004521414236242814, "minimum": 0.1060632251711753, "maximum": 0.12077674855544815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972698043828134, "standard_deviation": 0.0007805957255957753, "minimum": 0.995597731342876, "maximum": 0.9981458199734056, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.086094342532806, "standard_deviation": 0.03236243447819274, "minimum": 1.0425215100558063, "maximum": 1.1312710795789431, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5065416636400109, "standard_deviation": 0.25982062307327186, "minimum": 0.12259422557755928, "maximum": 0.9460253837132218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06616864954157774, "standard_deviation": 0.007612137502109131, "minimum": 0.05956868454718641, "maximum": 0.08181143468807024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9303408623093667, "standard_deviation": 0.013190911322172918, "minimum": 0.8999971788590043, "maximum": 0.9408477385447973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0112013892254, "standard_deviation": 0.057301914035947504, "minimum": 1.9495091983679549, "maximum": 2.115594268857613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.679835552688529, "standard_deviation": 0.5109718524077465, "minimum": 1.0595501794254873, "maximum": 2.7208436789269954, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29045571295185696, "standard_deviation": 0.027080311317932888, "minimum": 0.26638899453524806, "maximum": 0.35114206883471616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9485188743925242, "standard_deviation": 0.013497184804704243, "minimum": 0.9177527624639307, "maximum": 0.9593906249083184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4817591396421481, "standard_deviation": 0.046650501736244535, "minimum": 1.4262169132489442, "maximum": 1.5752403018493275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4566875541837345, "standard_deviation": 0.7207253904455737, "minimum": 0.7207775138061908, "maximum": 3.0040168154268607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23716575146711902, "standard_deviation": 0.029689204003055067, "minimum": 0.2107980592210054, "maximum": 0.3026484133359492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9873379554881576, "standard_deviation": 0.0032196228633420212, "minimum": 0.980571786835761, "maximum": 0.9905572666574342, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5400950831698994, "standard_deviation": 0.06607661340845056, "minimum": 1.4616003203255192, "maximum": 1.6283762015159169, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7773455970411588, "standard_deviation": 0.3513666848519628, "minimum": 0.3136955743642127, "maximum": 1.4780264427057723, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11540276116246947, "standard_deviation": 0.01408243269371754, "minimum": 0.09990107671898138, "maximum": 0.142921404602903, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947526882892131, "standard_deviation": 0.0018072332224183733, "minimum": 0.9912734526885945, "maximum": 0.9966189340954058, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0121892340840386, "standard_deviation": 0.05945427906950513, "minimum": 0.932902134870725, "maximum": 1.1136438204251748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1005734518183834, "standard_deviation": 0.4491245553775282, "minimum": 0.4985830124892698, "maximum": 1.925005357098028, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07294883426089181, "standard_deviation": 0.012506150188659462, "minimum": 0.05891619034053261, "maximum": 0.09415123980082092, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.35898243384826867, "standard_deviation": 0.18873430704292996, "minimum": 0.21197753181853465, "maximum": 0.6352398501659485, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.738555445926286, "standard_deviation": 0.009418438050099634, "minimum": 5.7236215062253395, "maximum": 5.754013596332727, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 1.0860717121232362, "standard_deviation": 0.17859093472244564, "minimum": 0.8663329958760217, "maximum": 1.3431069748737272, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03977954475903692, "standard_deviation": 0.3589649225668679, "minimum": -0.43959360167547046, "maximum": 0.6865355588557621, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9743124193255315, "standard_deviation": 0.013842768437493666, "minimum": 0.9558959157800789, "maximum": 0.9892082049032491, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.757895583626858, "standard_deviation": 0.022613611503255603, "minimum": 0.715023074546193, "maximum": 0.7957339184772149, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.5914685869729266, "standard_deviation": 1.1418853376778306, "minimum": 0.3081066546986538, "maximum": 3.0663096089556947, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15168045803406371, "standard_deviation": 0.04704965481078725, "minimum": 0.09842678131105388, "maximum": 0.21742799010439345, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9871066660876557, "standard_deviation": 0.011120258029127075, "minimum": 0.9686671202915008, "maximum": 0.9970914089377957, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42897679936419536, "standard_deviation": 0.03790017023664718, "minimum": 0.3706908235018443, "maximum": 0.4865313966139817, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.884454317413495, "standard_deviation": 1.3223748728295428, "minimum": 0.5200253881637317, "maximum": 3.617107628364949, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10189110485214214, "standard_deviation": 0.05258683170230354, "minimum": 0.049568069194767896, "maximum": 0.20066363471729165, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0034", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.9142537117004395, "predicted_mos": 1.9142537117004395, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "9acd4cde565bbd0f6b2ed86fd670fe2bbf2f4e3ae93265721a2c55984da532d4", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.461118400096893, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7471694409519061, "全帯域_10ms_rms_envelope_nrmse": 0.27614824677088046, "全帯域_10ms_envelope_correlation": 0.9434451578828447, "全帯域_10ms_modulation_mae_db": 3.265611469377634, "全帯域_40ms_log_rms_mae_db": 0.4879499697708083, "全帯域_40ms_rms_envelope_nrmse": 0.23349148455896088, "全帯域_40ms_envelope_correlation": 0.9596444099726095, "全帯域_40ms_modulation_mae_db": 3.9228453677186828, "300–1000 Hz_10ms_log_rms_mae_db": 1.2190053259465345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10571891811788428, "300–1000 Hz_10ms_envelope_correlation": 0.9893968661598029, "300–1000 Hz_10ms_modulation_mae_db": 0.19744157567658807, "300–1000 Hz_40ms_log_rms_mae_db": 0.6682703630588026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06358902633185534, "300–1000 Hz_40ms_envelope_correlation": 0.9958282955566272, "300–1000 Hz_40ms_modulation_mae_db": 0.24362908840296513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1592211323367017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10242433725738329, "1000–3000 Hz_10ms_envelope_correlation": 0.9933912189021664, "1000–3000 Hz_10ms_modulation_mae_db": 0.8075003007359003, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7204962728593698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06185137904842591, "1000–3000 Hz_40ms_envelope_correlation": 0.9978359316046121, "1000–3000 Hz_40ms_modulation_mae_db": 0.5338195348236742, "3000–6000 Hz_10ms_log_rms_mae_db": 1.318046121130842, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21012305167639442, "3000–6000 Hz_10ms_envelope_correlation": 0.9628733958269572, "3000–6000 Hz_10ms_modulation_mae_db": 1.1461461981898273, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9083131150529571, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12858840975628494, "3000–6000 Hz_40ms_envelope_correlation": 0.9848692513779025, "3000–6000 Hz_40ms_modulation_mae_db": 0.6873682771743367, "mr_stft_db": 5.319628418793717, "band_mae_db": 0.6657009745292984, "rms_gain_error_db": 0.8192788565177511, "output_peak": 1.5550985392193348}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 86, "at_or_above_full_scale_fraction": 0.0015379113018597996, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0015379113018597996, "longest_full_scale_run_samples": 86, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.807707335023411, "全帯域_10ms_rms_envelope_nrmse": 0.15923220247166894, "全帯域_10ms_envelope_correlation": 0.9721050593797559, "全帯域_10ms_modulation_mae_db": 2.69950755033736, "全帯域_40ms_log_rms_mae_db": 0.4276890182615515, "全帯域_40ms_rms_envelope_nrmse": 0.10590189441698393, "全帯域_40ms_envelope_correlation": 0.988025707770983, "全帯域_40ms_modulation_mae_db": 3.5524948893302857, "300–1000 Hz_10ms_log_rms_mae_db": 1.2362482326335038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10901442232854132, "300–1000 Hz_10ms_envelope_correlation": 0.9886396025799211, "300–1000 Hz_10ms_modulation_mae_db": 0.6791797204004643, "300–1000 Hz_40ms_log_rms_mae_db": 0.7411816122665376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06515883488386107, "300–1000 Hz_40ms_envelope_correlation": 0.9956267769600425, "300–1000 Hz_40ms_modulation_mae_db": 0.9152690017308102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1266881522271204, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10101321698378257, "1000–3000 Hz_10ms_envelope_correlation": 0.993363734969372, "1000–3000 Hz_10ms_modulation_mae_db": 1.1638395919113638, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7066458510074671, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056465321281595586, "1000–3000 Hz_40ms_envelope_correlation": 0.9979971305610504, "1000–3000 Hz_40ms_modulation_mae_db": 0.6113364333841512, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3186996302933853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1721614685763964, "3000–6000 Hz_10ms_envelope_correlation": 0.9752380098097276, "3000–6000 Hz_10ms_modulation_mae_db": 0.8334515123137806, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9117195546769494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11295097192585433, "3000–6000 Hz_40ms_envelope_correlation": 0.9882545062803808, "3000–6000 Hz_40ms_modulation_mae_db": 0.6165516150322695, "mr_stft_db": 5.317845952060936, "band_mae_db": 0.4272692713152077, "rms_gain_error_db": -0.362547928370942, "output_peak": 0.7365971138089961}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8012373690549017, "全帯域_10ms_rms_envelope_nrmse": 0.24743196632269066, "全帯域_10ms_envelope_correlation": 0.9551186725104335, "全帯域_10ms_modulation_mae_db": 2.852409870604376, "全帯域_40ms_log_rms_mae_db": 0.4911558420144777, "全帯域_40ms_rms_envelope_nrmse": 0.22092065217310825, "全帯域_40ms_envelope_correlation": 0.9631693217658062, "全帯域_40ms_modulation_mae_db": 3.068424337853522, "300–1000 Hz_10ms_log_rms_mae_db": 1.2594384343910772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1031225641290871, "300–1000 Hz_10ms_envelope_correlation": 0.9898448752707646, "300–1000 Hz_10ms_modulation_mae_db": 1.1559277072330774, "300–1000 Hz_40ms_log_rms_mae_db": 0.7905532442150507, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0566394049721059, "300–1000 Hz_40ms_envelope_correlation": 0.9967895956776857, "300–1000 Hz_40ms_modulation_mae_db": 1.4699144825043973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1364141540402377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1132406434064551, "1000–3000 Hz_10ms_envelope_correlation": 0.9925852753726713, "1000–3000 Hz_10ms_modulation_mae_db": 0.8391504441741257, "1000–3000 Hz_40ms_log_rms_mae_db": 0.737074109331317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06831860312807482, "1000–3000 Hz_40ms_envelope_correlation": 0.9981645152551022, "1000–3000 Hz_40ms_modulation_mae_db": 0.5552224422803125, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3190734243913023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17486768385200302, "3000–6000 Hz_10ms_envelope_correlation": 0.9745650749016623, "3000–6000 Hz_10ms_modulation_mae_db": 0.9245507305600102, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8593417586696087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11287954515374272, "3000–6000 Hz_40ms_envelope_correlation": 0.9883110250909022, "3000–6000 Hz_40ms_modulation_mae_db": 0.4663787367034826, "mr_stft_db": 5.312281200083711, "band_mae_db": 0.7481648945833833, "rms_gain_error_db": 0.7297798742047095, "output_peak": 1.549800780268235}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 68, "at_or_above_full_scale_fraction": 0.0012160228898426323, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0012160228898426323, "longest_full_scale_run_samples": 68, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7753618126211107, "全帯域_10ms_rms_envelope_nrmse": 0.09965388303435796, "全帯域_10ms_envelope_correlation": 0.9888033549445613, "全帯域_10ms_modulation_mae_db": 0.7884655889592632, "全帯域_40ms_log_rms_mae_db": 0.45556260056497017, "全帯域_40ms_rms_envelope_nrmse": 0.04716386325989474, "全帯域_40ms_envelope_correlation": 0.9975005345985446, "全帯域_40ms_modulation_mae_db": 1.006562249338705, "300–1000 Hz_10ms_log_rms_mae_db": 1.313842887048298, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11670926109721066, "300–1000 Hz_10ms_envelope_correlation": 0.9869848937993793, "300–1000 Hz_10ms_modulation_mae_db": 0.48969560125993644, "300–1000 Hz_40ms_log_rms_mae_db": 0.805060940418346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07123809815952283, "300–1000 Hz_40ms_envelope_correlation": 0.9949170035636873, "300–1000 Hz_40ms_modulation_mae_db": 0.44326328066959436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1365205860371794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10955693083788245, "1000–3000 Hz_10ms_envelope_correlation": 0.9921208219254378, "1000–3000 Hz_10ms_modulation_mae_db": 1.2332628853392242, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6986019747226283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04604358631594332, "1000–3000 Hz_40ms_envelope_correlation": 0.9986650373379848, "1000–3000 Hz_40ms_modulation_mae_db": 1.145253024482714, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3065764322655016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1741972317861575, "3000–6000 Hz_10ms_envelope_correlation": 0.9757360647522073, "3000–6000 Hz_10ms_modulation_mae_db": 0.6871251531588652, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8883575145299342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1133554906543103, "3000–6000 Hz_40ms_envelope_correlation": 0.9902109732146619, "3000–6000 Hz_40ms_modulation_mae_db": 0.5858309835090819, "mr_stft_db": 5.323086603854146, "band_mae_db": 0.26010440948643404, "rms_gain_error_db": -0.14203020858504156, "output_peak": 0.9004867539236318}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8165740814689404, "全帯域_10ms_rms_envelope_nrmse": 0.15719965344593623, "全帯域_10ms_envelope_correlation": 0.9718814644154216, "全帯域_10ms_modulation_mae_db": 1.6591937781464559, "全帯域_40ms_log_rms_mae_db": 0.5088200997248965, "全帯域_40ms_rms_envelope_nrmse": 0.08176203194663692, "全帯域_40ms_envelope_correlation": 0.9926078633454404, "全帯域_40ms_modulation_mae_db": 2.1307520756041245, "300–1000 Hz_10ms_log_rms_mae_db": 1.3035394255019506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11591344837465434, "300–1000 Hz_10ms_envelope_correlation": 0.9871285877523241, "300–1000 Hz_10ms_modulation_mae_db": 0.7779254309224988, "300–1000 Hz_40ms_log_rms_mae_db": 0.8674457450540972, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0706987347380008, "300–1000 Hz_40ms_envelope_correlation": 0.9949086393778062, "300–1000 Hz_40ms_modulation_mae_db": 0.8186199320076191, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1201466283763044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12019028864890562, "1000–3000 Hz_10ms_envelope_correlation": 0.9911916721538679, "1000–3000 Hz_10ms_modulation_mae_db": 0.8766880388143689, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6826868180341065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08035031047289498, "1000–3000 Hz_40ms_envelope_correlation": 0.9965851351118292, "1000–3000 Hz_40ms_modulation_mae_db": 0.4040061769887988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3309860003887757, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18623977558516275, "3000–6000 Hz_10ms_envelope_correlation": 0.9709400448723916, "3000–6000 Hz_10ms_modulation_mae_db": 1.1769722241202574, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8921298063571774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11124970544459188, "3000–6000 Hz_40ms_envelope_correlation": 0.9886287779691009, "3000–6000 Hz_40ms_modulation_mae_db": 0.9142911997934365, "mr_stft_db": 5.320312922208353, "band_mae_db": 0.33625451700381026, "rms_gain_error_db": -0.26054232229835744, "output_peak": 0.7125471198086123}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8024775472114064, "全帯域_10ms_rms_envelope_nrmse": 0.12150877316475577, "全帯域_10ms_envelope_correlation": 0.9840885808700117, "全帯域_10ms_modulation_mae_db": 1.5365320107036806, "全帯域_40ms_log_rms_mae_db": 0.42062868253331964, "全帯域_40ms_rms_envelope_nrmse": 0.08555206562151674, "全帯域_40ms_envelope_correlation": 0.9922430829368519, "全帯域_40ms_modulation_mae_db": 1.948652768985976, "300–1000 Hz_10ms_log_rms_mae_db": 1.350187470312952, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1108290513659328, "300–1000 Hz_10ms_envelope_correlation": 0.9881883125285543, "300–1000 Hz_10ms_modulation_mae_db": 0.17789592310423644, "300–1000 Hz_40ms_log_rms_mae_db": 0.8141003281364395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06240438487933973, "300–1000 Hz_40ms_envelope_correlation": 0.9959397383762315, "300–1000 Hz_40ms_modulation_mae_db": 0.14574518799347935, "1000–3000 Hz_10ms_log_rms_mae_db": 1.130276816284693, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09454912217399575, "1000–3000 Hz_10ms_envelope_correlation": 0.9943068405156091, "1000–3000 Hz_10ms_modulation_mae_db": 0.49100581171270535, "1000–3000 Hz_40ms_log_rms_mae_db": 0.680350443136408, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04570727567798066, "1000–3000 Hz_40ms_envelope_correlation": 0.9988894058258654, "1000–3000 Hz_40ms_modulation_mae_db": 0.4560776520975842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2897237781446913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16618736300753714, "3000–6000 Hz_10ms_envelope_correlation": 0.9770840979442922, "3000–6000 Hz_10ms_modulation_mae_db": 0.9585523495159584, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8340770302372017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10436381123438657, "3000–6000 Hz_40ms_envelope_correlation": 0.990455300346821, "3000–6000 Hz_40ms_modulation_mae_db": 0.7935433066916379, "mr_stft_db": 5.30866903635312, "band_mae_db": 0.4070343548146116, "rms_gain_error_db": -0.2922430487032852, "output_peak": 0.9081943658570473}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7751782536133942, "全帯域_10ms_rms_envelope_nrmse": 0.17810207643731793, "全帯域_10ms_envelope_correlation": 0.9695192426839633, "全帯域_10ms_modulation_mae_db": 1.2398680303498948, "全帯域_40ms_log_rms_mae_db": 0.4251669180455824, "全帯域_40ms_rms_envelope_nrmse": 0.11944849073797266, "全帯域_40ms_envelope_correlation": 0.9864958252503907, "全帯域_40ms_modulation_mae_db": 1.4917277121949228, "300–1000 Hz_10ms_log_rms_mae_db": 1.2530159120737692, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09356061409066374, "300–1000 Hz_10ms_envelope_correlation": 0.9916929807100703, "300–1000 Hz_10ms_modulation_mae_db": 0.6433713835982348, "300–1000 Hz_40ms_log_rms_mae_db": 0.7754899133999013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05172100329354153, "300–1000 Hz_40ms_envelope_correlation": 0.9973398561900092, "300–1000 Hz_40ms_modulation_mae_db": 0.7234376694395476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1334944782241085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0999836775981284, "1000–3000 Hz_10ms_envelope_correlation": 0.9934555367730242, "1000–3000 Hz_10ms_modulation_mae_db": 0.6580715916844718, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7279891270589881, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048934720806217656, "1000–3000 Hz_40ms_envelope_correlation": 0.9984316586180607, "1000–3000 Hz_40ms_modulation_mae_db": 0.5097954595544522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2947795006872447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17637711348940044, "3000–6000 Hz_10ms_envelope_correlation": 0.9739261818833211, "3000–6000 Hz_10ms_modulation_mae_db": 0.7506279474069126, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8860953066785733, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10831683324132554, "3000–6000 Hz_40ms_envelope_correlation": 0.9893031522029442, "3000–6000 Hz_40ms_modulation_mae_db": 0.6359338937470113, "mr_stft_db": 5.31301732664588, "band_mae_db": 0.3900015302504956, "rms_gain_error_db": 0.33814523300863686, "output_peak": 1.303391095642953}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 45, "at_or_above_full_scale_fraction": 0.0008047210300429185, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0008047210300429185, "longest_full_scale_run_samples": 45, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.738241705913482, "全帯域_10ms_rms_envelope_nrmse": 0.24858535377916158, "全帯域_10ms_envelope_correlation": 0.9425426865429974, "全帯域_10ms_modulation_mae_db": 2.108812758832635, "全帯域_40ms_log_rms_mae_db": 0.4384021774858769, "全帯域_40ms_rms_envelope_nrmse": 0.17197369495785692, "全帯域_40ms_envelope_correlation": 0.9725015925629961, "全帯域_40ms_modulation_mae_db": 2.6245986859807178, "300–1000 Hz_10ms_log_rms_mae_db": 1.3310629643751524, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10629519158272105, "300–1000 Hz_10ms_envelope_correlation": 0.9893656401135876, "300–1000 Hz_10ms_modulation_mae_db": 0.6716820804318631, "300–1000 Hz_40ms_log_rms_mae_db": 0.8456592174272504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06274604301773234, "300–1000 Hz_40ms_envelope_correlation": 0.9963121905677328, "300–1000 Hz_40ms_modulation_mae_db": 0.7975427492066268, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1386760254795203, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08676328264599407, "1000–3000 Hz_10ms_envelope_correlation": 0.9950651179443639, "1000–3000 Hz_10ms_modulation_mae_db": 0.6746452662129704, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6659806185075605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04462609086761944, "1000–3000 Hz_40ms_envelope_correlation": 0.998702081713148, "1000–3000 Hz_40ms_modulation_mae_db": 0.8435283229960037, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3462556157310053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18957750823431344, "3000–6000 Hz_10ms_envelope_correlation": 0.9698271112584366, "3000–6000 Hz_10ms_modulation_mae_db": 0.64305058843229, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8992707769783835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13424232763300237, "3000–6000 Hz_40ms_envelope_correlation": 0.9835286399389216, "3000–6000 Hz_40ms_modulation_mae_db": 0.7405660738669666, "mr_stft_db": 5.3324880128046805, "band_mae_db": 0.4877298710807429, "rms_gain_error_db": 0.43907953693086, "output_peak": 1.5508097634050872}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 55, "at_or_above_full_scale_fraction": 0.0009835479256080114, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0009835479256080114, "longest_full_scale_run_samples": 55, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 55920, "output_samples": 55920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.190645244620426, "spectral_convergence": 0.425206904285382, "spectral_convergence_status": "ok", "amplitude_floor": 7.545858591823444e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.137495808451118, "spectral_convergence": 0.4733023718375903, "spectral_convergence_status": "ok", "amplitude_floor": 3.9963548992246115e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.630744203309608, "spectral_convergence": 0.4699117751814026, "spectral_convergence_status": "ok", "amplitude_floor": 1.7986227023732554e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001975478876015244, "output_power": 0.004240769710578806, "delta_db": 3.31772299219604, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005013278749394044, "output_power": 0.0050521347306620975, "delta_db": 0.03353070812802345, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015170067046631896, "output_power": 0.0015407357403088308, "delta_db": 0.06740656747017021, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0018605244766058998, "output_power": 0.001758669638524467, "delta_db": -0.244511218269522, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002538671104412511, "output_power": 0.00024958251884895445, "delta_db": -0.07392276420624179, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.356486516629775e-05, "output_power": 5.991091747372545e-05, "delta_db": -0.257111596905793, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6657009745292984, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8192788565177511, "pitch": null, "output_rms_dbfs": -18.893495850828515, "output_peak": 1.5550985392193348, "reference_rms_dbfs": -19.712774707346266, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.319628418793717}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 55920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.0006292775937517623, "log_rms_mae_db": 0.7471694409519061, "rms_envelope_nrmse": 0.27614824677088046, "envelope_correlation": 0.9434451578828447, "rms_gain_error_db": 0.8192793056802128, "reference_mean_power": 0.010699017072596153, "output_mean_power": 0.012920276596199057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 4.048774453403744e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014613887448305467, "output_power": 0.002186290473527507, "reference_active": true, "delta_db": 1.749421041842919}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0008261048603288555, "output_power": 0.0019184174367177678, "reference_active": true, "delta_db": 3.6590793581501435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00047302858802070836, "output_power": 0.0012993339597139764, "reference_active": true, "delta_db": 4.3883340081398385}], "active_band_count": 3, "active_band_mae_db": 3.265611469377634}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00037792225909995443, "log_rms_mae_db": 0.4879499697708083, "rms_envelope_nrmse": 0.23349148455896088, "envelope_correlation": 0.9596444099726095, "rms_gain_error_db": 0.8192821376995951, "reference_mean_power": 0.01079168973829664, "output_mean_power": 0.013032197849201977}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 3.4667222612961444e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016650159461876711, "output_power": 0.003050587521436973, "reference_active": true, "delta_db": 2.629650922407893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005334276556675961, "output_power": 0.0013615140680597116, "reference_active": true, "delta_db": 4.069466050905079}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.906803616368198e-05, "output_power": 0.0001255346147265624, "reference_active": true, "delta_db": 5.069419129843076}], "active_band_count": 3, "active_band_mae_db": 3.9228453677186828}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00018013905998130842, "log_rms_mae_db": 1.2190053259465345, "rms_envelope_nrmse": 0.10571891811788428, "envelope_correlation": 0.9893968661598029, "rms_gain_error_db": 0.050899158819804904, "reference_mean_power": 0.002043548171012202, "output_mean_power": 0.0020676393812012975}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.1637162379955263e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006991910792272859, "output_power": 0.0006680894666552665, "reference_active": true, "delta_db": -0.1976125406568789}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001302332633123258, "output_power": 0.00013190063731090588, "reference_active": true, "delta_db": 0.055249709506928}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.350140195021887e-05, "output_power": 3.6225069656942784e-05, "reference_active": true, "delta_db": 0.3394624768659573}], "active_band_count": 3, "active_band_mae_db": 0.19744157567658807}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00014142899773968785, "log_rms_mae_db": 0.6682703630588026, "rms_envelope_nrmse": 0.06358902633185534, "envelope_correlation": 0.9958282955566272, "rms_gain_error_db": 0.050903857289458156, "reference_mean_power": 0.002061256155357224, "output_mean_power": 0.002085558379707377}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.0272195933213497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006464643086637218, "output_power": 0.000619157227070907, "reference_active": true, "delta_db": -0.18743605889274692}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.880640785962183e-05, "output_power": 7.149084007753128e-05, "reference_active": true, "delta_db": 0.16621514969843776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.140839173745677e-06, "output_power": 2.3351116398647002e-06, "reference_active": true, "delta_db": 0.3772360566177107}], "active_band_count": 3, "active_band_mae_db": 0.24362908840296513}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00022296985761207562, "log_rms_mae_db": 1.1592211323367017, "rms_envelope_nrmse": 0.10242433725738329, "envelope_correlation": 0.9933912189021664, "rms_gain_error_db": -0.2390514821030504, "reference_mean_power": 0.0018916055464765064, "output_mean_power": 0.0017902984289313998}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.550199090959336e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001029129669204316, "output_power": 0.0009321910297823009, "reference_active": true, "delta_db": -0.42965179315848473}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.51753044024429e-05, "output_power": 5.2657435432371896e-05, "reference_active": true, "delta_db": -0.20285033942840563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2165950116970791e-05, "output_power": 1.8371554623429462e-05, "reference_active": true, "delta_db": 1.7899987696208106}], "active_band_count": 3, "active_band_mae_db": 0.8075003007359003}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00021618630891631995, "log_rms_mae_db": 0.7204962728593698, "rms_envelope_nrmse": 0.06185137904842591, "envelope_correlation": 0.9978359316046121, "rms_gain_error_db": -0.2390611135996398, "reference_mean_power": 0.0019079864948334626, "output_mean_power": 0.0018057980719147217}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.466045308126094e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009690551708671025, "output_power": 0.000880044660283691, "reference_active": true, "delta_db": -0.41843791109229694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151940036741328e-05, "output_power": 2.282381092230577e-05, "reference_active": true, "delta_db": 0.2555799521033786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.641667021707976e-07, "output_power": 9.460894158142279e-07, "reference_active": true, "delta_db": 0.927440741275347}], "active_band_count": 3, "active_band_mae_db": 0.5338195348236742}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 3.608035305023674e-05, "log_rms_mae_db": 1.318046121130842, "rms_envelope_nrmse": 0.21012305167639442, "envelope_correlation": 0.9628733958269572, "rms_gain_error_db": -0.17577438889900074, "reference_mean_power": 6.674409144257214e-05, "output_mean_power": 6.409665810289594e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 5.317396933477194e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.505970371058983e-05, "output_power": 3.160996244220053e-05, "reference_active": true, "delta_db": -0.4498426172533368}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.310128917388118e-06, "output_power": 2.640341961013684e-06, "reference_active": true, "delta_db": 0.5802396119638226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.893247149634755e-07, "output_power": 1.722560543409364e-06, "reference_active": true, "delta_db": 2.408356365352323}], "active_band_count": 3, "active_band_mae_db": 1.1461461981898273}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 2.62382889528218e-05, "log_rms_mae_db": 0.9083131150529571, "rms_envelope_nrmse": 0.12858840975628494, "envelope_correlation": 0.9848692513779025, "rms_gain_error_db": -0.175779551253243, "reference_mean_power": 6.730512481108481e-05, "output_mean_power": 6.463536100893098e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 4.920971050071525e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.323536847986801e-05, "output_power": 3.0446293556925106e-05, "reference_active": true, "delta_db": -0.3806606770932681}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2494288246616935e-06, "output_power": 1.3334333698528218e-06, "reference_active": true, "delta_db": 0.28259798350308224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.101103941251684e-08, "output_power": 1.1179665809808718e-07, "reference_active": true, "delta_db": 1.3988461709266597}], "active_band_count": 3, "active_band_mae_db": 0.6873682771743367}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9195660263429128, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7c1ea9ea726b5e80dbae6135177a1d5090f0eddb5ef225c8e3913f52ba2c541a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9931850273195642, "standard_deviation": 0.001219281562280775, "minimum": 0.9911916721538679, "maximum": 0.9950651179443639, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1351797466257332, "standard_deviation": 0.011454970723699802, "minimum": 1.1201466283763044, "maximum": 1.1592211323367017, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8430204913231413, "standard_deviation": 0.2518771902897171, "minimum": 0.49100581171270535, "maximum": 1.2332628853392242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10346518744406591, "standard_deviation": 0.010623855703095974, "minimum": 0.08676328264599407, "maximum": 0.12019028864890562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9981588620034566, "standard_deviation": 0.0007335337933910884, "minimum": 0.9965851351118292, "maximum": 0.9988894058258654, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7024781518322307, "standard_deviation": 0.025122953669244646, "minimum": 0.6659806185075605, "maximum": 0.737074109331317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6323798808259613, "standard_deviation": 0.2455305420977823, "minimum": 0.4040061769887988, "maximum": 1.145253024482714, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05653716094984404, "standard_deviation": 0.012879589363371254, "minimum": 0.04462609086761944, "maximum": 0.08035031047289498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9725237476561245, "standard_deviation": 0.0045907684759012145, "minimum": 0.9628733958269572, "maximum": 0.9770840979442922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3155175628790934, "standard_deviation": 0.01847235051220544, "minimum": 1.2897237781446913, "maximum": 1.3462556157310053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8900595879622377, "standard_deviation": 0.19950710596806787, "minimum": 0.64305058843229, "maximum": 1.1769722241202574, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18121639952592064, "standard_deviation": 0.013895499237402706, "minimum": 0.16618736300753714, "maximum": 0.21012305167639442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9879452033027044, "standard_deviation": 0.002476824179345609, "minimum": 0.9835286399389216, "maximum": 0.990455300346821, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8849131078975981, "standard_deviation": 0.026120138771604542, "minimum": 0.8340770302372017, "maximum": 0.9117195546769494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6800580108147778, "standard_deviation": 0.13730899968415283, "minimum": 0.4663787367034826, "maximum": 0.9142911997934365, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11574338688043734, "standard_deviation": 0.010238221486416933, "minimum": 0.10436381123438657, "maximum": 0.13424232763300237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9889052198643005, "standard_deviation": 0.0015370123258483528, "minimum": 0.9869848937993793, "maximum": 0.9916929807100703, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2832925815354046, "standard_deviation": 0.04770730780076438, "minimum": 1.2190053259465345, "maximum": 1.350187470312952, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5991399278283625, "standard_deviation": 0.3176674315887609, "minimum": 0.17789592310423644, "maximum": 1.1559277072330774, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10764543388583692, "standard_deviation": 0.00743443430167144, "minimum": 0.09356061409066374, "maximum": 0.11670926109721066, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959577620337279, "standard_deviation": 0.0008483787171354126, "minimum": 0.9949086393778062, "maximum": 0.9973398561900092, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7884701704970531, "standard_deviation": 0.062446451129561216, "minimum": 0.6682703630588026, "maximum": 0.8674457450540972, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.69467767399438, "standard_deviation": 0.4216300994645122, "minimum": 0.14574518799347935, "maximum": 1.4699144825043973, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06302444128449494, "standard_deviation": 0.006549064341571458, "minimum": 0.05172100329354153, "maximum": 0.07123809815952283, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.465282477882998, "standard_deviation": 0.1647271528351296, "minimum": 0.26010440948643404, "maximum": 0.7481648945833833, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.318416184100568, "standard_deviation": 0.007435571422426517, "minimum": 5.30866903635312, "maximum": 5.3324880128046805, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 1.1521156914917372, "standard_deviation": 0.37627512146031905, "minimum": 0.7125471198086123, "maximum": 1.5550985392193348, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": 0.1586149990880414, "standard_deviation": 0.4802620841261494, "minimum": -0.362547928370942, "maximum": 0.8192788565177511, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9659380274037487, "standard_deviation": 0.017356666969582737, "minimum": 0.9425426865429974, "maximum": 0.9888033549445613, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7829934432323191, "standard_deviation": 0.028924292696958603, "minimum": 0.738241705913482, "maximum": 0.8165740814689404, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 2.0188001321639124, "standard_deviation": 0.8621622441785632, "minimum": 0.7884655889592632, "maximum": 3.265611469377634, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18598276942834618, "standard_deviation": 0.06441011390320946, "minimum": 0.09965388303435796, "maximum": 0.27614824677088046, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9815235422754528, "standard_deviation": 0.014427187533247123, "minimum": 0.9596444099726095, "maximum": 0.9975005345985446, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4569219135501854, "standard_deviation": 0.03454415960339543, "minimum": 0.42062868253331964, "maximum": 0.5088200997248965, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 2.468257260875867, "standard_deviation": 1.0106646894349032, "minimum": 1.006562249338705, "maximum": 3.9228453677186828, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.13327677220911638, "standard_deviation": 0.0681483094541249, "minimum": 0.04716386325989474, "maximum": 0.23349148455896088, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0034", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.3577494621276855, "predicted_mos": 2.3577494621276855, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "181dca288f51fb8cd51e46cc092b8a28d47fd473fd18bd0f2bacb9bf84dbe882", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.156978300306946, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.753580170221588, "全帯域_10ms_rms_envelope_nrmse": 0.15983825167634927, "全帯域_10ms_envelope_correlation": 0.9706033859974049, "全帯域_10ms_modulation_mae_db": 0.9352859086622402, "全帯域_40ms_log_rms_mae_db": 0.4373356814770962, "全帯域_40ms_rms_envelope_nrmse": 0.07548096905332055, "全帯域_40ms_envelope_correlation": 0.9929866329268198, "全帯域_40ms_modulation_mae_db": 1.434517390982407, "300–1000 Hz_10ms_log_rms_mae_db": 1.419780767221803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11283938079581109, "300–1000 Hz_10ms_envelope_correlation": 0.9881031865504842, "300–1000 Hz_10ms_modulation_mae_db": 1.080939546081203, "300–1000 Hz_40ms_log_rms_mae_db": 0.9110228099050117, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06853530281388753, "300–1000 Hz_40ms_envelope_correlation": 0.9956545738694321, "300–1000 Hz_40ms_modulation_mae_db": 1.3850515873330878, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2832124550151136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10180036270421668, "1000–3000 Hz_10ms_envelope_correlation": 0.9932103819215884, "1000–3000 Hz_10ms_modulation_mae_db": 0.8868350784412083, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8340222637215705, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05760997284999653, "1000–3000 Hz_40ms_envelope_correlation": 0.9978259160658797, "1000–3000 Hz_40ms_modulation_mae_db": 0.1655326050176185, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5327435358556132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17667520107737983, "3000–6000 Hz_10ms_envelope_correlation": 0.9739132147048357, "3000–6000 Hz_10ms_modulation_mae_db": 0.7708030154143534, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0278746256328783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11031590407090383, "3000–6000 Hz_40ms_envelope_correlation": 0.9889999857735682, "3000–6000 Hz_40ms_modulation_mae_db": 0.5669768746518162, "mr_stft_db": 5.450010214302484, "band_mae_db": 0.3075211655210028, "rms_gain_error_db": -0.1202526087006569, "output_peak": 0.8706827940478361}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7574056643795587, "全帯域_10ms_rms_envelope_nrmse": 0.10964253369724804, "全帯域_10ms_envelope_correlation": 0.9865281488515077, "全帯域_10ms_modulation_mae_db": 0.219596115102224, "全帯域_40ms_log_rms_mae_db": 0.4254904671464468, "全帯域_40ms_rms_envelope_nrmse": 0.05736231306916912, "全帯域_40ms_envelope_correlation": 0.9960488564852485, "全帯域_40ms_modulation_mae_db": 0.2493567121531168, "300–1000 Hz_10ms_log_rms_mae_db": 1.4340226556673923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12076487285438396, "300–1000 Hz_10ms_envelope_correlation": 0.9865703222729193, "300–1000 Hz_10ms_modulation_mae_db": 0.9393923001346557, "300–1000 Hz_40ms_log_rms_mae_db": 1.028405558015399, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08161322516637803, "300–1000 Hz_40ms_envelope_correlation": 0.993988717888434, "300–1000 Hz_40ms_modulation_mae_db": 1.0884764052055584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2605866363294995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1012091026016467, "1000–3000 Hz_10ms_envelope_correlation": 0.9935439686247024, "1000–3000 Hz_10ms_modulation_mae_db": 0.3753543296166859, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8576651272569894, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05502655071888685, "1000–3000 Hz_40ms_envelope_correlation": 0.9983864826753659, "1000–3000 Hz_40ms_modulation_mae_db": 0.15388816153508936, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5323727576029658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19028834822351515, "3000–6000 Hz_10ms_envelope_correlation": 0.9697678800242973, "3000–6000 Hz_10ms_modulation_mae_db": 0.5405396756859, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0574321681597334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13155197499415996, "3000–6000 Hz_40ms_envelope_correlation": 0.9846226162966827, "3000–6000 Hz_40ms_modulation_mae_db": 0.8238712225972761, "mr_stft_db": 5.450001595435089, "band_mae_db": 0.2572158997033938, "rms_gain_error_db": -0.05757159768949416, "output_peak": 1.2823865369691436}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 30, "at_or_above_full_scale_fraction": 0.000536480686695279, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0005543633762517883, "longest_full_scale_run_samples": 30, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7361845669214552, "全帯域_10ms_rms_envelope_nrmse": 0.16016124863474895, "全帯域_10ms_envelope_correlation": 0.9730323585574242, "全帯域_10ms_modulation_mae_db": 1.1489129824369302, "全帯域_40ms_log_rms_mae_db": 0.411329287673566, "全帯域_40ms_rms_envelope_nrmse": 0.0988669818025983, "全帯域_40ms_envelope_correlation": 0.9892801990053732, "全帯域_40ms_modulation_mae_db": 0.9218718559659318, "300–1000 Hz_10ms_log_rms_mae_db": 1.3374991952727802, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1044771737496441, "300–1000 Hz_10ms_envelope_correlation": 0.989522427008461, "300–1000 Hz_10ms_modulation_mae_db": 0.25520807016794933, "300–1000 Hz_40ms_log_rms_mae_db": 0.851721241842162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05762362154920231, "300–1000 Hz_40ms_envelope_correlation": 0.9965554754979489, "300–1000 Hz_40ms_modulation_mae_db": 0.1385423802872818, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3092257752402126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09063184001344836, "1000–3000 Hz_10ms_envelope_correlation": 0.9946346055201382, "1000–3000 Hz_10ms_modulation_mae_db": 0.1963702088115715, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8567130172541945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05225831162859219, "1000–3000 Hz_40ms_envelope_correlation": 0.998243245309586, "1000–3000 Hz_40ms_modulation_mae_db": 0.6192248161657002, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4879004356868228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1935723307461146, "3000–6000 Hz_10ms_envelope_correlation": 0.9685238703984416, "3000–6000 Hz_10ms_modulation_mae_db": 0.7455298367090754, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0487792467137187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12404036110604963, "3000–6000 Hz_40ms_envelope_correlation": 0.9859499943467457, "3000–6000 Hz_40ms_modulation_mae_db": 0.62334225745717, "mr_stft_db": 5.43413911308166, "band_mae_db": 0.26361790702511034, "rms_gain_error_db": 0.17820035213438765, "output_peak": 1.1406567496386306}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 24, "at_or_above_full_scale_fraction": 0.00042918454935622315, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.00042918454935622315, "longest_full_scale_run_samples": 23, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7378791498781185, "全帯域_10ms_rms_envelope_nrmse": 0.2113683197711662, "全帯域_10ms_envelope_correlation": 0.9479928135802297, "全帯域_10ms_modulation_mae_db": 1.6932033367218846, "全帯域_40ms_log_rms_mae_db": 0.42264765502834595, "全帯域_40ms_rms_envelope_nrmse": 0.11274901502324204, "全帯域_40ms_envelope_correlation": 0.9845593024557464, "全帯域_40ms_modulation_mae_db": 2.215268028822779, "300–1000 Hz_10ms_log_rms_mae_db": 1.4108405192084768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10314798348885416, "300–1000 Hz_10ms_envelope_correlation": 0.9898566550571822, "300–1000 Hz_10ms_modulation_mae_db": 0.44238007228843307, "300–1000 Hz_40ms_log_rms_mae_db": 0.8948361264389064, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058834540922586456, "300–1000 Hz_40ms_envelope_correlation": 0.9964327443677988, "300–1000 Hz_40ms_modulation_mae_db": 0.5701633790808223, "1000–3000 Hz_10ms_log_rms_mae_db": 1.291889596807108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12929404944465142, "1000–3000 Hz_10ms_envelope_correlation": 0.9890405737346545, "1000–3000 Hz_10ms_modulation_mae_db": 1.163991107384862, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8627387121707554, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0658738565075224, "1000–3000 Hz_40ms_envelope_correlation": 0.9970658732189347, "1000–3000 Hz_40ms_modulation_mae_db": 0.7941806591200983, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5524352481717334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2050519482692881, "3000–6000 Hz_10ms_envelope_correlation": 0.9648299395823363, "3000–6000 Hz_10ms_modulation_mae_db": 1.030250394243724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0944489446413668, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13414893629252897, "3000–6000 Hz_40ms_envelope_correlation": 0.9842849009882945, "3000–6000 Hz_40ms_modulation_mae_db": 0.8232016883810572, "mr_stft_db": 5.452453951721492, "band_mae_db": 0.35571420869821835, "rms_gain_error_db": -0.2380840570918642, "output_peak": 1.0857241713000114}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 14, "at_or_above_full_scale_fraction": 0.00025035765379113016, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.00025035765379113016, "longest_full_scale_run_samples": 14, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7992729310336588, "全帯域_10ms_rms_envelope_nrmse": 0.13166064657155685, "全帯域_10ms_envelope_correlation": 0.9813739823238677, "全帯域_10ms_modulation_mae_db": 2.4159780422023114, "全帯域_40ms_log_rms_mae_db": 0.47253654464658756, "全帯域_40ms_rms_envelope_nrmse": 0.0995577730079718, "全帯域_40ms_envelope_correlation": 0.9891151865689564, "全帯域_40ms_modulation_mae_db": 2.9789537737975826, "300–1000 Hz_10ms_log_rms_mae_db": 1.3420370932385695, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11007552979323283, "300–1000 Hz_10ms_envelope_correlation": 0.9884850467131255, "300–1000 Hz_10ms_modulation_mae_db": 0.4204282611513266, "300–1000 Hz_40ms_log_rms_mae_db": 0.7785442231648468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056505953789653625, "300–1000 Hz_40ms_envelope_correlation": 0.9967297142434997, "300–1000 Hz_40ms_modulation_mae_db": 0.4567927436936669, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2360696829973392, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10854351387920048, "1000–3000 Hz_10ms_envelope_correlation": 0.9922385090804605, "1000–3000 Hz_10ms_modulation_mae_db": 0.45157030774284984, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7759425108085096, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05477151261305081, "1000–3000 Hz_40ms_envelope_correlation": 0.9980418090003778, "1000–3000 Hz_40ms_modulation_mae_db": 0.34083560313094535, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5539037946721934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20015175605287788, "3000–6000 Hz_10ms_envelope_correlation": 0.9666687035870781, "3000–6000 Hz_10ms_modulation_mae_db": 0.7248380697573005, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1286275160463684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1421898357389422, "3000–6000 Hz_40ms_envelope_correlation": 0.9822354673949842, "3000–6000 Hz_40ms_modulation_mae_db": 0.7720304087230597, "mr_stft_db": 5.450250862792017, "band_mae_db": 0.45332692881288605, "rms_gain_error_db": -0.31309492727199134, "output_peak": 0.7985225348598501}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.708880440197045, "全帯域_10ms_rms_envelope_nrmse": 0.10867373106270857, "全帯域_10ms_envelope_correlation": 0.9866217688155632, "全帯域_10ms_modulation_mae_db": 0.5983429557170655, "全帯域_40ms_log_rms_mae_db": 0.39230818109921484, "全帯域_40ms_rms_envelope_nrmse": 0.05256624089873422, "全帯域_40ms_envelope_correlation": 0.9967952691597186, "全帯域_40ms_modulation_mae_db": 0.9434474240384795, "300–1000 Hz_10ms_log_rms_mae_db": 1.359177027137853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12197151779139477, "300–1000 Hz_10ms_envelope_correlation": 0.9858939741897833, "300–1000 Hz_10ms_modulation_mae_db": 1.2276391675707423, "300–1000 Hz_40ms_log_rms_mae_db": 0.9139100764682221, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07154982091332403, "300–1000 Hz_40ms_envelope_correlation": 0.9950427407518898, "300–1000 Hz_40ms_modulation_mae_db": 1.564723148603053, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3416460302307487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10929091080853255, "1000–3000 Hz_10ms_envelope_correlation": 0.9924574981502178, "1000–3000 Hz_10ms_modulation_mae_db": 0.6858722494073101, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8547171772352449, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06110843715826638, "1000–3000 Hz_40ms_envelope_correlation": 0.9979664926430383, "1000–3000 Hz_40ms_modulation_mae_db": 0.7621399325876569, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5195293009450759, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1953202365012351, "3000–6000 Hz_10ms_envelope_correlation": 0.9679405617392606, "3000–6000 Hz_10ms_modulation_mae_db": 0.7358063972737195, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0699603493798975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12446835662511527, "3000–6000 Hz_40ms_envelope_correlation": 0.9858789072607108, "3000–6000 Hz_40ms_modulation_mae_db": 0.4186160858518461, "mr_stft_db": 5.439375218728604, "band_mae_db": 0.1647048139388331, "rms_gain_error_db": -0.13849586875727513, "output_peak": 0.8618359710737723}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7390445688561361, "全帯域_10ms_rms_envelope_nrmse": 0.09069882771411915, "全帯域_10ms_envelope_correlation": 0.9909249712551387, "全帯域_10ms_modulation_mae_db": 0.9465270504660984, "全帯域_40ms_log_rms_mae_db": 0.39353034302449863, "全帯域_40ms_rms_envelope_nrmse": 0.05169131122472804, "全帯域_40ms_envelope_correlation": 0.9971190214911, "全帯域_40ms_modulation_mae_db": 1.1185607446570176, "300–1000 Hz_10ms_log_rms_mae_db": 1.3358802231920206, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1021573729291279, "300–1000 Hz_10ms_envelope_correlation": 0.9902265154378692, "300–1000 Hz_10ms_modulation_mae_db": 0.49138463905301927, "300–1000 Hz_40ms_log_rms_mae_db": 0.8617403805483888, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06079263276507794, "300–1000 Hz_40ms_envelope_correlation": 0.9963164341733215, "300–1000 Hz_40ms_modulation_mae_db": 0.511272636611831, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3343657727107223, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10051791662645713, "1000–3000 Hz_10ms_envelope_correlation": 0.9937866080101033, "1000–3000 Hz_10ms_modulation_mae_db": 0.5722206222124956, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8341545071928232, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06377403705681309, "1000–3000 Hz_40ms_envelope_correlation": 0.9978186915275435, "1000–3000 Hz_40ms_modulation_mae_db": 0.9063766064039651, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5762975371254508, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19220477468066985, "3000–6000 Hz_10ms_envelope_correlation": 0.9693451668627366, "3000–6000 Hz_10ms_modulation_mae_db": 1.0087098783778228, "3000–6000 Hz_40ms_log_rms_mae_db": 1.126255582869259, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1426956772598395, "3000–6000 Hz_40ms_envelope_correlation": 0.9820459749458303, "3000–6000 Hz_40ms_modulation_mae_db": 0.702180936213281, "mr_stft_db": 5.4370362619372985, "band_mae_db": 0.3020061723566787, "rms_gain_error_db": -0.17288880406112278, "output_peak": 1.028235248695319}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 5, "at_or_above_full_scale_fraction": 8.94134477825465e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 8.94134477825465e-05, "longest_full_scale_run_samples": 3, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7494521730924694, "全帯域_10ms_rms_envelope_nrmse": 0.24884540094632415, "全帯域_10ms_envelope_correlation": 0.9373011828983203, "全帯域_10ms_modulation_mae_db": 1.5133040287898833, "全帯域_40ms_log_rms_mae_db": 0.43181628748573714, "全帯域_40ms_rms_envelope_nrmse": 0.16081333829762326, "全帯域_40ms_envelope_correlation": 0.9725596572667093, "全帯域_40ms_modulation_mae_db": 1.946092540077803, "300–1000 Hz_10ms_log_rms_mae_db": 1.3475135423630928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11395906954445148, "300–1000 Hz_10ms_envelope_correlation": 0.9878073339471044, "300–1000 Hz_10ms_modulation_mae_db": 0.5809023527396039, "300–1000 Hz_40ms_log_rms_mae_db": 0.8282970269582547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06740944256397693, "300–1000 Hz_40ms_envelope_correlation": 0.9957648076737853, "300–1000 Hz_40ms_modulation_mae_db": 0.7358925806001717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3098810332233797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1203803333280784, "1000–3000 Hz_10ms_envelope_correlation": 0.9916220326221431, "1000–3000 Hz_10ms_modulation_mae_db": 0.5742440151269324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8025723097443278, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07806489533172345, "1000–3000 Hz_40ms_envelope_correlation": 0.9973947034883671, "1000–3000 Hz_40ms_modulation_mae_db": 0.35367598012167917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6171939900107295, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1898285986205145, "3000–6000 Hz_10ms_envelope_correlation": 0.9699234456921629, "3000–6000 Hz_10ms_modulation_mae_db": 0.4337182127549866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.135282694235478, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12601304259207835, "3000–6000 Hz_40ms_envelope_correlation": 0.9856747808647072, "3000–6000 Hz_40ms_modulation_mae_db": 0.2590700639731441, "mr_stft_db": 5.437132364797801, "band_mae_db": 0.45705632175900734, "rms_gain_error_db": 0.2877611556504815, "output_peak": 1.2614091766964044}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 55, "at_or_above_full_scale_fraction": 0.0009835479256080114, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0010014306151645207, "longest_full_scale_run_samples": 53, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 55920, "output_samples": 55920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.367333589961615, "spectral_convergence": 0.3906912597907201, "spectral_convergence_status": "ok", "amplitude_floor": 7.545858591823444e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.30783973295878, "spectral_convergence": 0.380346165847611, "spectral_convergence_status": "ok", "amplitude_floor": 3.9963548992246115e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.674857319987057, "spectral_convergence": 0.3407181249624228, "spectral_convergence_status": "ok", "amplitude_floor": 1.7986227023732554e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001975478876015244, "output_power": 0.001606872908363853, "delta_db": -0.8969086141544834, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005013278749394044, "output_power": 0.005225438482398725, "delta_db": 0.18000886261160193, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015170067046631896, "output_power": 0.0014387024809128463, "delta_db": -0.23016507720518634, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0018605244766058998, "output_power": 0.0018205404030462746, "delta_db": -0.0943506624234047, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002538671104412511, "output_power": 0.00023958935221933325, "delta_db": -0.2513892650943416, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.356486516629775e-05, "output_power": 6.081163483148688e-05, "delta_db": -0.19230451163699863, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3075211655210028, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1202526087006569, "pitch": null, "output_rms_dbfs": -19.833027316046923, "output_peak": 0.8706827940478361, "reference_rms_dbfs": -19.712774707346266, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.450010214302484}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 55920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.0006292775937517623, "log_rms_mae_db": 0.753580170221588, "rms_envelope_nrmse": 0.15983825167634927, "envelope_correlation": 0.9706033859974049, "rms_gain_error_db": -0.12025334316815238, "reference_mean_power": 0.010699017072596153, "output_mean_power": 0.010406832060288416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 4.048774453403744e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014613887448305467, "output_power": 0.0014585541620709007, "reference_active": true, "delta_db": -0.00843197334310704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0008261048603288555, "output_power": 0.000802503772772303, "reference_active": true, "delta_db": -0.12588094501007652}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00047302858802070836, "output_power": 0.0002557012834193602, "reference_active": true, "delta_db": -2.671544807633537}], "active_band_count": 3, "active_band_mae_db": 0.9352859086622402}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00037792225909995443, "log_rms_mae_db": 0.4373356814770962, "rms_envelope_nrmse": 0.07548096905332055, "envelope_correlation": 0.9929866329268198, "rms_gain_error_db": -0.12025748158870511, "reference_mean_power": 0.01079168973829664, "output_mean_power": 0.010496963877414034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 3.4667222612961444e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016650159461876711, "output_power": 0.0015846490728681378, "reference_active": true, "delta_db": -0.21485296305428847}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005334276556675961, "output_power": 0.0004552625013451403, "reference_active": true, "delta_db": -0.6881364782362932}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.906803616368198e-05, "output_power": 1.7855224203570637e-05, "reference_active": true, "delta_db": -3.4005627316566396}], "active_band_count": 3, "active_band_mae_db": 1.434517390982407}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00018013905998130842, "log_rms_mae_db": 1.419780767221803, "rms_envelope_nrmse": 0.11283938079581109, "envelope_correlation": 0.9881031865504842, "rms_gain_error_db": -0.17029407058686702, "reference_mean_power": 0.002043548171012202, "output_mean_power": 0.0019649679557978603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.1637162379955263e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006991910792272859, "output_power": 0.0006385776149557964, "reference_active": true, "delta_db": -0.3938218814931904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001302332633123258, "output_power": 0.00010780559740893775, "reference_active": true, "delta_db": -0.8208061242414201}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.350140195021887e-05, "output_power": 2.1001190689143482e-05, "reference_active": true, "delta_db": -2.028190632508999}], "active_band_count": 3, "active_band_mae_db": 1.080939546081203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00014142899773968785, "log_rms_mae_db": 0.9110228099050117, "rms_envelope_nrmse": 0.06853530281388753, "envelope_correlation": 0.9956545738694321, "rms_gain_error_db": -0.170299337182741, "reference_mean_power": 0.002061256155357224, "output_mean_power": 0.00198199261446833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.0272195933213497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006464643086637218, "output_power": 0.0005856949197885987, "reference_active": true, "delta_db": -0.42873095399053923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.880640785962183e-05, "output_power": 5.5305586160928064e-05, "reference_active": true, "delta_db": -0.9485988585893247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.140839173745677e-06, "output_power": 1.1292797696038652e-06, "reference_active": true, "delta_db": -2.777824949419399}], "active_band_count": 3, "active_band_mae_db": 1.3850515873330878}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00022296985761207562, "log_rms_mae_db": 1.2832124550151136, "rms_envelope_nrmse": 0.10180036270421668, "envelope_correlation": 0.9932103819215884, "rms_gain_error_db": -0.10468619485761602, "reference_mean_power": 0.0018916055464765064, "output_mean_power": 0.0018465537735626789}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.550199090959336e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001029129669204316, "output_power": 0.0009462226973348881, "reference_active": true, "delta_db": -0.3647673744344574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.51753044024429e-05, "output_power": 5.2405264369690944e-05, "reference_active": true, "delta_db": -0.2236982162735818}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2165950116970791e-05, "output_power": 1.9604239335357753e-05, "reference_active": true, "delta_db": 2.072039644615586}], "active_band_count": 3, "active_band_mae_db": 0.8868350784412083}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00021618630891631995, "log_rms_mae_db": 0.8340222637215705, "rms_envelope_nrmse": 0.05760997284999653, "envelope_correlation": 0.9978259160658797, "rms_gain_error_db": -0.10469362041590123, "reference_mean_power": 0.0019079864948334626, "output_mean_power": 0.001862541397464969}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.466045308126094e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009690551708671025, "output_power": 0.0008929867746141456, "reference_active": true, "delta_db": -0.3550347637262002}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151940036741328e-05, "output_power": 2.1539058151274376e-05, "reference_active": true, "delta_db": 0.003965431157617998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.641667021707976e-07, "output_power": 7.887654367549007e-07, "reference_active": true, "delta_db": 0.13759762016903734}], "active_band_count": 3, "active_band_mae_db": 0.1655326050176185}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 3.608035305023674e-05, "log_rms_mae_db": 1.5327435358556132, "rms_envelope_nrmse": 0.17667520107737983, "envelope_correlation": 0.9739132147048357, "rms_gain_error_db": -0.08997029041814758, "reference_mean_power": 6.674409144257214e-05, "output_mean_power": 6.537561637868212e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 5.317396933477194e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.505970371058983e-05, "output_power": 3.2970238642406635e-05, "reference_active": true, "delta_db": -0.26686150924126106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.310128917388118e-06, "output_power": 2.366211114654676e-06, "reference_active": true, "delta_db": 0.10417273519646111}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.893247149634755e-07, "output_power": 1.5469502207743258e-06, "reference_active": true, "delta_db": 1.941374801805338}], "active_band_count": 3, "active_band_mae_db": 0.7708030154143534}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 2.62382889528218e-05, "log_rms_mae_db": 1.0278746256328783, "rms_envelope_nrmse": 0.11031590407090383, "envelope_correlation": 0.9889999857735682, "rms_gain_error_db": -0.09003643966432151, "reference_mean_power": 6.730512481108481e-05, "output_mean_power": 6.592414257748061e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 4.920971050071525e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.323536847986801e-05, "output_power": 3.159207667542483e-05, "reference_active": true, "delta_db": -0.22022323368645155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2494288246616935e-06, "output_power": 1.128623270661516e-06, "reference_active": true, "delta_db": -0.44162520435018754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.101103941251684e-08, "output_power": 1.0290877843988233e-07, "reference_active": true, "delta_db": 1.0390821859188095}], "active_band_count": 3, "active_band_mae_db": 0.5669768746518162}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9105567129805101, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3477ab112929b30b8ff6f03be4bd0f9ffad64b8f90978d6b0d61e622f15b8db6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9925667722080009, "standard_deviation": 0.0017146172563427423, "minimum": 0.9890405737346545, "maximum": 0.9946346055201382, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2958596228192656, "standard_deviation": 0.035755329479263076, "minimum": 1.2360696829973392, "maximum": 1.3416460302307487, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6133072398429895, "standard_deviation": 0.30311875025222973, "minimum": 0.1963702088115715, "maximum": 1.163991107384862, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10770850367577897, "standard_deviation": 0.012248484659996063, "minimum": 0.09063184001344836, "maximum": 0.12929404944465142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9978429017411367, "standard_deviation": 0.00043361003865802726, "minimum": 0.9970658732189347, "maximum": 0.9983864826753659, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8348157031730519, "standard_deviation": 0.030901011475699354, "minimum": 0.7759425108085096, "maximum": 0.8627387121707554, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5119817955103441, "standard_deviation": 0.29565265990111567, "minimum": 0.15388816153508936, "maximum": 0.9063766064039651, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.061060946733106455, "standard_deviation": 0.00832404439147078, "minimum": 0.05225831162859219, "maximum": 0.07806489533172345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9688640978238936, "standard_deviation": 0.0026653038165855222, "minimum": 0.9648299395823363, "maximum": 0.9739132147048357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.546547075008823, "standard_deviation": 0.0387772132812963, "minimum": 1.4879004356868228, "maximum": 1.6171939900107295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7487744350271102, "standard_deviation": 0.2037695455567194, "minimum": 0.4337182127549866, "maximum": 1.030250394243724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19288664927144938, "standard_deviation": 0.008337074082228647, "minimum": 0.17667520107737983, "maximum": 0.2050519482692881, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9849615784839404, "standard_deviation": 0.0022405695236976876, "minimum": 0.9820459749458303, "maximum": 0.9889999857735682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0860826409598374, "standard_deviation": 0.04101405000632046, "minimum": 1.0278746256328783, "maximum": 1.135282694235478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6236611922310813, "standard_deviation": 0.2022242886689958, "minimum": 0.2590700639731441, "maximum": 0.8238712225972761, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12942801108495222, "standard_deviation": 0.010665385136107328, "minimum": 0.11031590407090383, "maximum": 0.1426956772598395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9883081826471162, "standard_deviation": 0.0015457255228216752, "minimum": 0.9858939741897833, "maximum": 0.9902265154378692, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3733438779127485, "standard_deviation": 0.04102383319081498, "minimum": 1.3358802231920206, "maximum": 1.4340226556673923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6797843011483666, "standard_deviation": 0.35408436646274916, "minimum": 0.25520807016794933, "maximum": 1.2276391675707423, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11117411261836255, "standard_deviation": 0.0076613311172821505, "minimum": 0.1021573729291279, "maximum": 0.12197151779139477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958106510582638, "standard_deviation": 0.0009241092737170686, "minimum": 0.993988717888434, "maximum": 0.9967297142434997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8835596804176489, "standard_deviation": 0.07402203500184155, "minimum": 0.7785442231648468, "maximum": 1.028405558015399, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8063643576769342, "standard_deviation": 0.4936159549959007, "minimum": 0.1385423802872818, "maximum": 1.564723148603053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06535806756051085, "standard_deviation": 0.00860607833049995, "minimum": 0.056505953789653625, "maximum": 0.08161322516637803, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3201454272268913, "standard_deviation": 0.09968000115108767, "minimum": 0.1647048139388331, "maximum": 0.45705632175900734, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.443799947849556, "standard_deviation": 0.007527549362935155, "minimum": 5.43413911308166, "maximum": 5.452453951721492, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 1.041181647910121, "standard_deviation": 0.18471796283691483, "minimum": 0.7985225348598501, "maximum": 1.2823865369691436, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07180329447344191, "standard_deviation": 0.2052439030384962, "minimum": -0.31309492727199134, "maximum": 0.2877611556504815, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.971797326534932, "standard_deviation": 0.019471333891531654, "minimum": 0.9373011828983203, "maximum": 0.9909249712551387, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7477124580725036, "standard_deviation": 0.025641878176384192, "minimum": 0.708880440197045, "maximum": 0.7992729310336588, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1838938025123298, "standard_deviation": 0.6843193922501449, "minimum": 0.219596115102224, "maximum": 2.4159780422023114, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15261112000927765, "standard_deviation": 0.05460445959982892, "minimum": 0.09069882771411915, "maximum": 0.24884540094632415, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.989808015669959, "standard_deviation": 0.008256059937152015, "minimum": 0.9725596572667093, "maximum": 0.9971190214911, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4233743059476866, "standard_deviation": 0.025881509212764144, "minimum": 0.39230818109921484, "maximum": 0.47253654464658756, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4760085588118896, "standard_deviation": 0.866122878956844, "minimum": 0.2493567121531168, "maximum": 2.9789537737975826, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08863599279717341, "standard_deviation": 0.03747761170791315, "minimum": 0.05169131122472804, "maximum": 0.16081333829762326, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0034", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.3212900161743164, "predicted_mos": 2.3212900161743164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "c73a3490365ed9179c8dedaa1f8d68d6f58cad80d0317ad1fcaa6ab050f0d5f3", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.229903200175613, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7098590698123592, "全帯域_10ms_rms_envelope_nrmse": 0.16699862543729277, "全帯域_10ms_envelope_correlation": 0.9716193377821672, "全帯域_10ms_modulation_mae_db": 1.0979769368613546, "全帯域_40ms_log_rms_mae_db": 0.4253183811145946, "全帯域_40ms_rms_envelope_nrmse": 0.10407402048380066, "全帯域_40ms_envelope_correlation": 0.9887492063463453, "全帯域_40ms_modulation_mae_db": 1.2054218384905215, "300–1000 Hz_10ms_log_rms_mae_db": 1.2516717385184928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0952330968057073, "300–1000 Hz_10ms_envelope_correlation": 0.9915542488059845, "300–1000 Hz_10ms_modulation_mae_db": 0.38594808258791335, "300–1000 Hz_40ms_log_rms_mae_db": 0.7777219013326341, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052758628504952956, "300–1000 Hz_40ms_envelope_correlation": 0.9973083522841594, "300–1000 Hz_40ms_modulation_mae_db": 0.37122825489221895, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2326915003027001, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09164636583227484, "1000–3000 Hz_10ms_envelope_correlation": 0.994602433600362, "1000–3000 Hz_10ms_modulation_mae_db": 0.5445403387714184, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7666335635388898, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04891520849940495, "1000–3000 Hz_40ms_envelope_correlation": 0.9985920702611181, "1000–3000 Hz_40ms_modulation_mae_db": 0.765822381931764, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4919106715283033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17889740318126124, "3000–6000 Hz_10ms_envelope_correlation": 0.9732753495723158, "3000–6000 Hz_10ms_modulation_mae_db": 0.7004961925160393, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9782744913469981, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11814378849355786, "3000–6000 Hz_40ms_envelope_correlation": 0.9871966590983695, "3000–6000 Hz_40ms_modulation_mae_db": 0.3918973385856411, "mr_stft_db": 5.404805209563172, "band_mae_db": 0.3035425140223593, "rms_gain_error_db": 0.22336181571920327, "output_peak": 1.2379669514458258}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 51, "at_or_above_full_scale_fraction": 0.0009120171673819743, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0009298998569384836, "longest_full_scale_run_samples": 27, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.749903451492366, "全帯域_10ms_rms_envelope_nrmse": 0.2326905162511927, "全帯域_10ms_envelope_correlation": 0.9383976669146403, "全帯域_10ms_modulation_mae_db": 4.031128846484671, "全帯域_40ms_log_rms_mae_db": 0.43462755514016255, "全帯域_40ms_rms_envelope_nrmse": 0.20482455951039055, "全帯域_40ms_envelope_correlation": 0.9486853536701537, "全帯域_40ms_modulation_mae_db": 5.375817711477743, "300–1000 Hz_10ms_log_rms_mae_db": 1.3714265699687767, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12148986171050087, "300–1000 Hz_10ms_envelope_correlation": 0.9859510875770513, "300–1000 Hz_10ms_modulation_mae_db": 0.3199665676091126, "300–1000 Hz_40ms_log_rms_mae_db": 0.8651575269260189, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06309634095029021, "300–1000 Hz_40ms_envelope_correlation": 0.9959025708496095, "300–1000 Hz_40ms_modulation_mae_db": 0.25250317571080144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2451441925158495, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09754407416465588, "1000–3000 Hz_10ms_envelope_correlation": 0.9938752612995364, "1000–3000 Hz_10ms_modulation_mae_db": 0.9194684685096908, "1000–3000 Hz_40ms_log_rms_mae_db": 0.780524072602128, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0489807027282929, "1000–3000 Hz_40ms_envelope_correlation": 0.9986211567541174, "1000–3000 Hz_40ms_modulation_mae_db": 0.38018771093459175, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4239722740769145, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1861454290774382, "3000–6000 Hz_10ms_envelope_correlation": 0.9711620539233121, "3000–6000 Hz_10ms_modulation_mae_db": 0.7414993990809359, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9915381460416933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12106452992093113, "3000–6000 Hz_40ms_envelope_correlation": 0.9865449971057128, "3000–6000 Hz_40ms_modulation_mae_db": 0.5814893895512773, "mr_stft_db": 5.406271624251889, "band_mae_db": 0.8429304662217035, "rms_gain_error_db": -0.5693703826680439, "output_peak": 0.6185886211301832}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7365944456366268, "全帯域_10ms_rms_envelope_nrmse": 0.1375307699094022, "全帯域_10ms_envelope_correlation": 0.9799584148021168, "全帯域_10ms_modulation_mae_db": 2.0950886318677777, "全帯域_40ms_log_rms_mae_db": 0.3945635593601064, "全帯域_40ms_rms_envelope_nrmse": 0.1084858484311921, "全帯域_40ms_envelope_correlation": 0.9873470597268478, "全帯域_40ms_modulation_mae_db": 2.7771552129931982, "300–1000 Hz_10ms_log_rms_mae_db": 1.3355785901019002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09189551978562181, "300–1000 Hz_10ms_envelope_correlation": 0.9920159299716016, "300–1000 Hz_10ms_modulation_mae_db": 0.3323284092858878, "300–1000 Hz_40ms_log_rms_mae_db": 0.8360048368734382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050265607344941, "300–1000 Hz_40ms_envelope_correlation": 0.9974926186875561, "300–1000 Hz_40ms_modulation_mae_db": 0.4598078333973498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2623288572391667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09510909251231371, "1000–3000 Hz_10ms_envelope_correlation": 0.9945584917604025, "1000–3000 Hz_10ms_modulation_mae_db": 0.48318269104353445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7895319593325668, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05277127021650758, "1000–3000 Hz_40ms_envelope_correlation": 0.9987894874217188, "1000–3000 Hz_40ms_modulation_mae_db": 0.1507783248498772, "3000–6000 Hz_10ms_log_rms_mae_db": 1.501286595357626, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19757655879659092, "3000–6000 Hz_10ms_envelope_correlation": 0.9671830891951378, "3000–6000 Hz_10ms_modulation_mae_db": 0.7865993204727474, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0456082833605527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1265768060591616, "3000–6000 Hz_40ms_envelope_correlation": 0.9853096763621095, "3000–6000 Hz_40ms_modulation_mae_db": 0.6898429310175004, "mr_stft_db": 5.421516423041072, "band_mae_db": 0.5385506246941569, "rms_gain_error_db": -0.36588431258667053, "output_peak": 0.8116683236990309}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7322486804901601, "全帯域_10ms_rms_envelope_nrmse": 0.12676904751896845, "全帯域_10ms_envelope_correlation": 0.9820421102777667, "全帯域_10ms_modulation_mae_db": 1.1531214228722317, "全帯域_40ms_log_rms_mae_db": 0.42956720955589156, "全帯域_40ms_rms_envelope_nrmse": 0.06878010403951108, "全帯域_40ms_envelope_correlation": 0.9943096168896225, "全帯域_40ms_modulation_mae_db": 1.0106600892904778, "300–1000 Hz_10ms_log_rms_mae_db": 1.360169664374466, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1290998962157534, "300–1000 Hz_10ms_envelope_correlation": 0.9849522888426852, "300–1000 Hz_10ms_modulation_mae_db": 1.2585040312223261, "300–1000 Hz_40ms_log_rms_mae_db": 0.8881184404162906, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0886764767588529, "300–1000 Hz_40ms_envelope_correlation": 0.9932217729874635, "300–1000 Hz_40ms_modulation_mae_db": 1.6612859080799758, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2373633717157126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10088243291795441, "1000–3000 Hz_10ms_envelope_correlation": 0.993585316375055, "1000–3000 Hz_10ms_modulation_mae_db": 0.28231365004701986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8146350444964662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05207265751910076, "1000–3000 Hz_40ms_envelope_correlation": 0.9985989497209331, "1000–3000 Hz_40ms_modulation_mae_db": 0.3945725358258619, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4386518735869378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1893363957315266, "3000–6000 Hz_10ms_envelope_correlation": 0.9699627622330088, "3000–6000 Hz_10ms_modulation_mae_db": 0.9625650734888427, "3000–6000 Hz_40ms_log_rms_mae_db": 0.97286808314863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1166620676509173, "3000–6000 Hz_40ms_envelope_correlation": 0.9878308051712339, "3000–6000 Hz_40ms_modulation_mae_db": 0.6610258647516402, "mr_stft_db": 5.387433123694556, "band_mae_db": 0.2979968015094785, "rms_gain_error_db": -0.019405261122595812, "output_peak": 0.8640804357139844}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7834695310347768, "全帯域_10ms_rms_envelope_nrmse": 0.13888372196072218, "全帯域_10ms_envelope_correlation": 0.9779026851382732, "全帯域_10ms_modulation_mae_db": 0.8062671759691429, "全帯域_40ms_log_rms_mae_db": 0.45922720166811243, "全帯域_40ms_rms_envelope_nrmse": 0.06155095410982342, "全帯域_40ms_envelope_correlation": 0.9952661762629285, "全帯域_40ms_modulation_mae_db": 1.0713920991515085, "300–1000 Hz_10ms_log_rms_mae_db": 1.2867377228887558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11410176725396465, "300–1000 Hz_10ms_envelope_correlation": 0.9876023135279426, "300–1000 Hz_10ms_modulation_mae_db": 0.6142534753524449, "300–1000 Hz_40ms_log_rms_mae_db": 0.814921745034749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06799452145000948, "300–1000 Hz_40ms_envelope_correlation": 0.9952256638909285, "300–1000 Hz_40ms_modulation_mae_db": 0.9618599083384277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.277280627205833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09635836947752884, "1000–3000 Hz_10ms_envelope_correlation": 0.9939079900074967, "1000–3000 Hz_10ms_modulation_mae_db": 1.1273100402807221, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7891842552155934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04770184182149679, "1000–3000 Hz_40ms_envelope_correlation": 0.998459399802708, "1000–3000 Hz_40ms_modulation_mae_db": 0.9472775381474863, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4645050188154671, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19906876495835082, "3000–6000 Hz_10ms_envelope_correlation": 0.9668479164036454, "3000–6000 Hz_10ms_modulation_mae_db": 0.6655478300244422, "3000–6000 Hz_40ms_log_rms_mae_db": 1.013880833829412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1386455103782068, "3000–6000 Hz_40ms_envelope_correlation": 0.9822655635810604, "3000–6000 Hz_40ms_modulation_mae_db": 0.5353318842949771, "mr_stft_db": 5.406356440461534, "band_mae_db": 0.08693703577530203, "rms_gain_error_db": -0.056129087496402205, "output_peak": 0.8477421986476055}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7900325255951787, "全帯域_10ms_rms_envelope_nrmse": 0.12798549916903793, "全帯域_10ms_envelope_correlation": 0.9812359669291252, "全帯域_10ms_modulation_mae_db": 0.7377771560309886, "全帯域_40ms_log_rms_mae_db": 0.4610635907940421, "全帯域_40ms_rms_envelope_nrmse": 0.06534746596785616, "全帯域_40ms_envelope_correlation": 0.994680982038607, "全帯域_40ms_modulation_mae_db": 1.4020730290267442, "300–1000 Hz_10ms_log_rms_mae_db": 1.3353332044304171, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10191768740031219, "300–1000 Hz_10ms_envelope_correlation": 0.9900328804554583, "300–1000 Hz_10ms_modulation_mae_db": 0.5422780506882255, "300–1000 Hz_40ms_log_rms_mae_db": 0.8241735597816221, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05819436037988875, "300–1000 Hz_40ms_envelope_correlation": 0.9965179560770114, "300–1000 Hz_40ms_modulation_mae_db": 0.583623205533241, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3082015349214733, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09031055789226643, "1000–3000 Hz_10ms_envelope_correlation": 0.994710935907988, "1000–3000 Hz_10ms_modulation_mae_db": 0.27012410068777, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8100033751221671, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048239497903215425, "1000–3000 Hz_40ms_envelope_correlation": 0.9985702738934199, "1000–3000 Hz_40ms_modulation_mae_db": 0.44534997619716116, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4646243464674287, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1995811547779231, "3000–6000 Hz_10ms_envelope_correlation": 0.9671055237818039, "3000–6000 Hz_10ms_modulation_mae_db": 0.5640191443142178, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0025210589551279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13390002343646287, "3000–6000 Hz_40ms_envelope_correlation": 0.9837153325333237, "3000–6000 Hz_40ms_modulation_mae_db": 0.5649032815030006, "mr_stft_db": 5.410379404244236, "band_mae_db": 0.11965355119558257, "rms_gain_error_db": -0.07155154367964602, "output_peak": 0.9796247353704423}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8043678134496224, "全帯域_10ms_rms_envelope_nrmse": 0.2942131911437847, "全帯域_10ms_envelope_correlation": 0.9389297042422364, "全帯域_10ms_modulation_mae_db": 3.3234269218611323, "全帯域_40ms_log_rms_mae_db": 0.4998916465986943, "全帯域_40ms_rms_envelope_nrmse": 0.25640309581386733, "全帯域_40ms_envelope_correlation": 0.9530152537135975, "全帯域_40ms_modulation_mae_db": 3.743450607921561, "300–1000 Hz_10ms_log_rms_mae_db": 1.3582230565829891, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10641526698669468, "300–1000 Hz_10ms_envelope_correlation": 0.9891200919641451, "300–1000 Hz_10ms_modulation_mae_db": 0.33144590154514547, "300–1000 Hz_40ms_log_rms_mae_db": 0.8124789305594629, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06153120383974126, "300–1000 Hz_40ms_envelope_correlation": 0.9960729587894521, "300–1000 Hz_40ms_modulation_mae_db": 0.4901157061613904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2519840166041523, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10218240768636966, "1000–3000 Hz_10ms_envelope_correlation": 0.993121444901021, "1000–3000 Hz_10ms_modulation_mae_db": 0.5045557841987048, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7807965354489476, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05767657109018994, "1000–3000 Hz_40ms_envelope_correlation": 0.9977565083243654, "1000–3000 Hz_40ms_modulation_mae_db": 0.19068505227475438, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5287625193455319, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19528188018131926, "3000–6000 Hz_10ms_envelope_correlation": 0.9680256966015787, "3000–6000 Hz_10ms_modulation_mae_db": 0.6793828950330401, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0595088431628328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12479255791815183, "3000–6000 Hz_40ms_envelope_correlation": 0.9857402746498153, "3000–6000 Hz_40ms_modulation_mae_db": 0.4572500778132061, "mr_stft_db": 5.399346007856444, "band_mae_db": 0.7278815344900068, "rms_gain_error_db": 0.9196657938344095, "output_peak": 1.9380101624496704}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 67, "at_or_above_full_scale_fraction": 0.001198140200286123, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.001198140200286123, "longest_full_scale_run_samples": 67, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7568091614042526, "全帯域_10ms_rms_envelope_nrmse": 0.15506224237476124, "全帯域_10ms_envelope_correlation": 0.9772521161907671, "全帯域_10ms_modulation_mae_db": 1.1455039830861895, "全帯域_40ms_log_rms_mae_db": 0.4495557540001546, "全帯域_40ms_rms_envelope_nrmse": 0.11980366198966758, "全帯域_40ms_envelope_correlation": 0.9862695867602902, "全帯域_40ms_modulation_mae_db": 1.043610640057184, "300–1000 Hz_10ms_log_rms_mae_db": 1.3392851863733286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12321417527799766, "300–1000 Hz_10ms_envelope_correlation": 0.9855016123562469, "300–1000 Hz_10ms_modulation_mae_db": 0.42241455127188177, "300–1000 Hz_40ms_log_rms_mae_db": 0.9040683994623576, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07611676603057695, "300–1000 Hz_40ms_envelope_correlation": 0.9939933611773606, "300–1000 Hz_40ms_modulation_mae_db": 0.5885875543952669, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2585427839472727, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11662390826590234, "1000–3000 Hz_10ms_envelope_correlation": 0.991382734263413, "1000–3000 Hz_10ms_modulation_mae_db": 0.53090888770709, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7304581377554142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056032736091452255, "1000–3000 Hz_40ms_envelope_correlation": 0.9984557972934703, "1000–3000 Hz_40ms_modulation_mae_db": 0.2590633461669676, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4547117141545838, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1675614206151302, "3000–6000 Hz_10ms_envelope_correlation": 0.9765016683479275, "3000–6000 Hz_10ms_modulation_mae_db": 0.3867916167202969, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0152094611883085, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11057057880167948, "3000–6000 Hz_40ms_envelope_correlation": 0.9888441925540666, "3000–6000 Hz_40ms_modulation_mae_db": 0.3921773975855673, "mr_stft_db": 5.390237488678683, "band_mae_db": 0.398638045323446, "rms_gain_error_db": 0.32579051997403025, "output_peak": 1.1576553618007248}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 25, "at_or_above_full_scale_fraction": 0.0004470672389127325, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0004470672389127325, "longest_full_scale_run_samples": 25, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 55920, "output_samples": 55920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.342300535186682, "spectral_convergence": 0.3919562044246267, "spectral_convergence_status": "ok", "amplitude_floor": 7.545858591823444e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.253520858067511, "spectral_convergence": 0.3849593965333858, "spectral_convergence_status": "ok", "amplitude_floor": 3.9963548992246115e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.618594235435322, "spectral_convergence": 0.3469408866113913, "spectral_convergence_status": "ok", "amplitude_floor": 1.7986227023732554e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001975478876015244, "output_power": 0.002637717658779143, "delta_db": 1.255559167397171, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005013278749394044, "output_power": 0.005001277017681544, "delta_db": -0.01040942498619862, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015170067046631896, "output_power": 0.0015146459250557078, "delta_db": -0.00676379446449118, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0018605244766058998, "output_power": 0.0017877938638109579, "delta_db": -0.173179455385125, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002538671104412511, "output_power": 0.0002459593290283345, "delta_db": -0.13743140201234139, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.356486516629775e-05, "output_power": 6.0176364419755556e-05, "delta_db": -0.23791183988882864, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3035425140223593, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.22336181571920327, "pitch": null, "output_rms_dbfs": -19.48941289162707, "output_peak": 1.2379669514458258, "reference_rms_dbfs": -19.712774707346266, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.404805209563172}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 55920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.0006292775937517623, "log_rms_mae_db": 0.7098590698123592, "rms_envelope_nrmse": 0.16699862543729277, "envelope_correlation": 0.9716193377821672, "rms_gain_error_db": 0.2233631146765077, "reference_mean_power": 0.010699017072596153, "output_mean_power": 0.011263677102652029}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 4.048774453403744e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014613887448305467, "output_power": 0.0017194836728739997, "reference_active": true, "delta_db": 0.7063229817403951}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0008261048603288555, "output_power": 0.0011968592843940727, "reference_active": true, "delta_db": 1.610079157830966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00047302858802070836, "output_power": 0.0005924343830921127, "reference_active": true, "delta_db": 0.9775286710127028}], "active_band_count": 3, "active_band_mae_db": 1.0979769368613546}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00037792225909995443, "log_rms_mae_db": 0.4253183811145946, "rms_envelope_nrmse": 0.10407402048380066, "envelope_correlation": 0.9887492063463453, "rms_gain_error_db": 0.22337079224018688, "reference_mean_power": 0.01079168973829664, "output_mean_power": 0.011361260821054686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 3.4667222612961444e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016650159461876711, "output_power": 0.002100426867715157, "reference_active": true, "delta_db": 1.0088916776352759}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005334276556675961, "output_power": 0.0007650580807451601, "reference_active": true, "delta_db": 1.56618878618171}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.906803616368198e-05, "output_power": 4.965238180001713e-05, "reference_active": true, "delta_db": 1.0411850516545789}], "active_band_count": 3, "active_band_mae_db": 1.2054218384905215}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00018013905998130842, "log_rms_mae_db": 1.2516717385184928, "rms_envelope_nrmse": 0.0952330968057073, "envelope_correlation": 0.9915542488059845, "rms_gain_error_db": 0.10200720947714262, "reference_mean_power": 0.002043548171012202, "output_mean_power": 0.002092115227042586}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.1637162379955263e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006991910792272859, "output_power": 0.0006867426106344194, "reference_active": true, "delta_db": -0.07801883526947584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001302332633123258, "output_power": 0.00013200936279642226, "reference_active": true, "delta_db": 0.05882811737735756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.350140195021887e-05, "output_power": 4.238017162018409e-05, "reference_active": true, "delta_db": 1.0209972951169066}], "active_band_count": 3, "active_band_mae_db": 0.38594808258791335}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00014142899773968785, "log_rms_mae_db": 0.7777219013326341, "rms_envelope_nrmse": 0.052758628504952956, "envelope_correlation": 0.9973083522841594, "rms_gain_error_db": 0.10201104982770216, "reference_mean_power": 0.002061256155357224, "output_mean_power": 0.0021102459261586512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.0272195933213497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006464643086637218, "output_power": 0.0006383926782547983, "reference_active": true, "delta_db": -0.05456655017354567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.880640785962183e-05, "output_power": 7.282469548618365e-05, "reference_active": true, "delta_db": 0.24649791865030593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.140839173745677e-06, "output_power": 2.5813451199236852e-06, "reference_active": true, "delta_db": 0.8126202958528053}], "active_band_count": 3, "active_band_mae_db": 0.37122825489221895}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00022296985761207562, "log_rms_mae_db": 1.2326915003027001, "rms_envelope_nrmse": 0.09164636583227484, "envelope_correlation": 0.994602433600362, "rms_gain_error_db": -0.1679299324793526, "reference_mean_power": 0.0018916055464765064, "output_mean_power": 0.0018198583493371208}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.550199090959336e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001029129669204316, "output_power": 0.0009362048542131089, "reference_active": true, "delta_db": -0.4109921022631324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.51753044024429e-05, "output_power": 4.7967436785349524e-05, "reference_active": true, "delta_db": -0.6079822596095908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2165950116970791e-05, "output_power": 1.4015567807801529e-05, "reference_active": true, "delta_db": 0.6146466544415318}], "active_band_count": 3, "active_band_mae_db": 0.5445403387714184}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00021618630891631995, "log_rms_mae_db": 0.7666335635388898, "rms_envelope_nrmse": 0.04891520849940495, "envelope_correlation": 0.9985920702611181, "rms_gain_error_db": -0.16793026329392421, "reference_mean_power": 0.0019079864948334626, "output_mean_power": 0.0018356178406330745}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.466045308126094e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009690551708671025, "output_power": 0.0008833636999975237, "reference_active": true, "delta_db": -0.40208954428693106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151940036741328e-05, "output_power": 1.8287287669513054e-05, "reference_active": true, "delta_db": -0.7068086901265407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.641667021707976e-07, "output_power": 5.812075117640064e-07, "reference_active": true, "delta_db": -1.1885689113818203}], "active_band_count": 3, "active_band_mae_db": 0.765822381931764}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 3.608035305023674e-05, "log_rms_mae_db": 1.4919106715283033, "rms_envelope_nrmse": 0.17889740318126124, "envelope_correlation": 0.9732753495723158, "rms_gain_error_db": -0.04313115671079411, "reference_mean_power": 6.674409144257214e-05, "output_mean_power": 6.608451543944865e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 5.317396933477194e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.505970371058983e-05, "output_power": 3.4296027596390876e-05, "reference_active": true, "delta_db": -0.09564421593484998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.310128917388118e-06, "output_power": 2.4241170958043227e-06, "reference_active": true, "delta_db": 0.20917377892894215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.893247149634755e-07, "output_power": 1.4962560411062083e-06, "reference_active": true, "delta_db": 1.796670582684326}], "active_band_count": 3, "active_band_mae_db": 0.7004961925160393}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 2.62382889528218e-05, "log_rms_mae_db": 0.9782744913469981, "rms_envelope_nrmse": 0.11814378849355786, "envelope_correlation": 0.9871966590983695, "rms_gain_error_db": -0.04321239268078646, "reference_mean_power": 6.730512481108481e-05, "output_mean_power": 6.663875807681892e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 4.920971050071525e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.323536847986801e-05, "output_power": 3.288595797996377e-05, "reference_active": true, "delta_db": -0.045900006147388955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2494288246616935e-06, "output_power": 1.1575644670015523e-06, "reference_active": true, "delta_db": -0.3316633395474854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.101103941251684e-08, "output_power": 9.735473331185805e-08, "reference_active": true, "delta_db": 0.7981286700620489}], "active_band_count": 3, "active_band_mae_db": 0.3918973385856411}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9218718198247879, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f691cc440235ccf1d3bf9664e443cc294d77287701f6fe87fa780e7b28a331ea", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9937180760144093, "standard_deviation": 0.0010929872025198042, "minimum": 0.991382734263413, "maximum": 0.994710935907988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.25919211055652, "standard_deviation": 0.024399811822640036, "minimum": 1.2326915003027001, "maximum": 1.3082015349214733, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5828004951557437, "standard_deviation": 0.29715252479261944, "minimum": 0.27012410068777, "maximum": 1.1273100402807221, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09883215109365827, "standard_deviation": 0.008262371342800513, "minimum": 0.09031055789226643, "maximum": 0.11662390826590234, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9984804554339813, "standard_deviation": 0.000310532844660136, "minimum": 0.9977565083243654, "maximum": 0.9987894874217188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7827208679390216, "standard_deviation": 0.026321472763256307, "minimum": 0.7304581377554142, "maximum": 0.8146350444964662, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.44171710829105804, "standard_deviation": 0.2797146805217323, "minimum": 0.1507783248498772, "maximum": 0.9472775381474863, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.051548810733707574, "standard_deviation": 0.003755103516159841, "minimum": 0.04770184182149679, "maximum": 0.05767657109018994, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9700080075073412, "standard_deviation": 0.003474341316927651, "minimum": 0.9668479164036454, "maximum": 0.9765016683479275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4710531266665992, "standard_deviation": 0.034428332157420215, "minimum": 1.4239722740769145, "maximum": 1.5287625193455319, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6858626839563203, "standard_deviation": 0.16680429007725958, "minimum": 0.3867916167202969, "maximum": 0.9625650734888427, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18918112591494254, "standard_deviation": 0.011313927426038607, "minimum": 0.1675614206151302, "maximum": 0.1995811547779231, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9859309376319614, "standard_deviation": 0.002166031974440436, "minimum": 0.9822655635810604, "maximum": 0.9888441925540666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0099261501291945, "standard_deviation": 0.03057378167399389, "minimum": 0.97286808314863, "maximum": 1.0595088431628328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5342397706378512, "standard_deviation": 0.11328316346979125, "minimum": 0.3918973385856411, "maximum": 0.6898429310175004, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1237944828323836, "standard_deviation": 0.009231455697205404, "minimum": 0.11057057880167948, "maximum": 0.1386455103782068, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9883413066876394, "standard_deviation": 0.0027542783807225314, "minimum": 0.9849522888426852, "maximum": 0.9920159299716016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3298032166548908, "standard_deviation": 0.04066469172535612, "minimum": 1.2516717385184928, "maximum": 1.3714265699687767, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5258923836953672, "standard_deviation": 0.3145904267569622, "minimum": 0.3199665676091126, "maximum": 1.2585040312223261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11042090892956907, "standard_deviation": 0.013684466101410518, "minimum": 0.09189551978562181, "maximum": 0.1290998962157534, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9957169068429426, "standard_deviation": 0.0015083764872426612, "minimum": 0.9932217729874635, "maximum": 0.9974926186875561, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8403306675483216, "standard_deviation": 0.042398614865930453, "minimum": 0.7777219013326341, "maximum": 0.9040683994623576, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.671126443313584, "standard_deviation": 0.45071625594568765, "minimum": 0.25250317571080144, "maximum": 1.6612859080799758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06482923815740668, "standard_deviation": 0.012659376077841468, "minimum": 0.050265607344941, "maximum": 0.0886764767588529, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4145163216540045, "standard_deviation": 0.27190263235354384, "minimum": 0.08693703577530203, "maximum": 0.8429304662217035, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.403293215223949, "standard_deviation": 0.010957154994409541, "minimum": 5.387433123694556, "maximum": 5.421516423041072, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 1.0569170987821834, "standard_deviation": 0.40680730372365065, "minimum": 0.6185886211301832, "maximum": 1.9380101624496704, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": 0.048309692746785565, "standard_deviation": 0.45510519185145104, "minimum": -0.5693703826680439, "maximum": 0.9196657938344095, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9684172502846367, "standard_deviation": 0.018639923498686466, "minimum": 0.9383976669146403, "maximum": 0.9820421102777667, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7579105848644179, "standard_deviation": 0.032360777490294645, "minimum": 0.7098590698123592, "maximum": 0.8043678134496224, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.7987863843791858, "standard_deviation": 1.2444910973858987, "minimum": 0.7377771560309886, "maximum": 4.031128846484671, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17251670172064526, "standard_deviation": 0.05999654683858283, "minimum": 0.12676904751896845, "maximum": 0.2942131911437847, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.981040404426049, "standard_deviation": 0.018986131025600927, "minimum": 0.9486853536701537, "maximum": 0.9952661762629285, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44422686227896985, "standard_deviation": 0.03109082855459376, "minimum": 0.3945635593601064, "maximum": 0.4998916465986943, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 2.2036976535511172, "standard_deviation": 1.6235961590419923, "minimum": 1.0106600892904778, "maximum": 5.375817711477743, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12365871379326362, "standard_deviation": 0.07081852077477918, "minimum": 0.06155095410982342, "maximum": 0.25640309581386733, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3575-170457-0041", "language": "en", "style": "normal", "speaker": "librispeech_3575", "source_utterance_id": "3575-170457-0041", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3575-170457-0041.wav", "source_file_sha256": "53144f21f4f1aa43bfeac6d1741a13db593ed0f1668f1715b0c44e4ec00c513c", "decoded_pcm_sha256": "ce9155c8469c627681fdbab9602fc52b17f46ce8849d53a2be4d450d58cfb2c4", "original_source_sha256": "90d1dd17384a0960d3acaf50b3f7f40b4699288aa74a04c25a221270e104ce55", "corpus_entry_id": "librispeech_3575-170457-0041"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.734901189804077, "predicted_mos": 3.734901189804077, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d6912402c8c323aee6a7b6a1d12627253745e0f237fc5cb375d9b1184a1e8f0a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.478819369000121, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3575-170457-0041", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d6912402c8c323aee6a7b6a1d12627253745e0f237fc5cb375d9b1184a1e8f0a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "6be0695d1829e2c748580d3856118cd6eabdb573ad8adcd1eb7f3b1824b606fa"}, "rows": [{"case": "librispeech_3575-170457-0041", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.2309367656707764, "predicted_mos": 2.2309367656707764, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "777b28e804d870ba26683f6092d7324e8781232321f473c07f47f7bd3aae2d8a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7930271059994993, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3575-170457-0041", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "777b28e804d870ba26683f6092d7324e8781232321f473c07f47f7bd3aae2d8a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8895092229720947, "全帯域_10ms_rms_envelope_nrmse": 0.07383972395095457, "全帯域_10ms_envelope_correlation": 0.992856438350343, "全帯域_10ms_modulation_mae_db": 0.20955158123644044, "全帯域_40ms_log_rms_mae_db": 0.5061251420947285, "全帯域_40ms_rms_envelope_nrmse": 0.041816403373567054, "全帯域_40ms_envelope_correlation": 0.9975923823640591, "全帯域_40ms_modulation_mae_db": 0.3337718639983514, "300–1000 Hz_10ms_log_rms_mae_db": 1.63248010459893, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.123971073755053, "300–1000 Hz_10ms_envelope_correlation": 0.9871845589904612, "300–1000 Hz_10ms_modulation_mae_db": 0.8664147177253589, "300–1000 Hz_40ms_log_rms_mae_db": 1.0114322761031664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07361996952037156, "300–1000 Hz_40ms_envelope_correlation": 0.9952437398302604, "300–1000 Hz_40ms_modulation_mae_db": 0.4780380510299574, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5026551454193369, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1370624297193433, "1000–3000 Hz_10ms_envelope_correlation": 0.9864909119114786, "1000–3000 Hz_10ms_modulation_mae_db": 0.6921579405863144, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8842040059152881, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07870304099526022, "1000–3000 Hz_40ms_envelope_correlation": 0.9960887051477185, "1000–3000 Hz_40ms_modulation_mae_db": 0.648632066919541, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6872072004928267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16111258695923839, "3000–6000 Hz_10ms_envelope_correlation": 0.9830590421823715, "3000–6000 Hz_10ms_modulation_mae_db": 0.1459308455655535, "3000–6000 Hz_40ms_log_rms_mae_db": 1.052180826603865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08980011307110725, "3000–6000 Hz_40ms_envelope_correlation": 0.9946476000678196, "3000–6000 Hz_40ms_modulation_mae_db": 0.33620621263706324, "mr_stft_db": 5.660604952047091, "band_mae_db": 0.1608183100771152, "rms_gain_error_db": -0.09558426560534639, "output_peak": 0.6559213389659905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8878804738171402, "全帯域_10ms_rms_envelope_nrmse": 0.066407767353469, "全帯域_10ms_envelope_correlation": 0.9943450302509489, "全帯域_10ms_modulation_mae_db": 0.16466121553492868, "全帯域_40ms_log_rms_mae_db": 0.4796258276859586, "全帯域_40ms_rms_envelope_nrmse": 0.03666179823586662, "全帯域_40ms_envelope_correlation": 0.9983013818727284, "全帯域_40ms_modulation_mae_db": 0.26715576435502414, "300–1000 Hz_10ms_log_rms_mae_db": 1.5837166501648186, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10422757709468215, "300–1000 Hz_10ms_envelope_correlation": 0.9910946402277366, "300–1000 Hz_10ms_modulation_mae_db": 0.3470422098216548, "300–1000 Hz_40ms_log_rms_mae_db": 0.9468745001937557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060866528206438474, "300–1000 Hz_40ms_envelope_correlation": 0.9969318415616629, "300–1000 Hz_40ms_modulation_mae_db": 0.48870023821616604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5051050013712226, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14338765174179952, "1000–3000 Hz_10ms_envelope_correlation": 0.9846205713487859, "1000–3000 Hz_10ms_modulation_mae_db": 0.8631658655227709, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8987549961046365, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08000366255542606, "1000–3000 Hz_40ms_envelope_correlation": 0.9951954663979333, "1000–3000 Hz_40ms_modulation_mae_db": 0.6382642239242675, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6962309757594487, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1724936126683452, "3000–6000 Hz_10ms_envelope_correlation": 0.9809941960190465, "3000–6000 Hz_10ms_modulation_mae_db": 0.5119615397135958, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0446356251599578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08835329127934911, "3000–6000 Hz_40ms_envelope_correlation": 0.9951862444062071, "3000–6000 Hz_40ms_modulation_mae_db": 0.47294598198085386, "mr_stft_db": 5.664587752332337, "band_mae_db": 0.11945204724651266, "rms_gain_error_db": -0.12423218756033386, "output_peak": 0.608326896776574}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8832284442322516, "全帯域_10ms_rms_envelope_nrmse": 0.06869653749452505, "全帯域_10ms_envelope_correlation": 0.9938506344755369, "全帯域_10ms_modulation_mae_db": 0.23139101441812984, "全帯域_40ms_log_rms_mae_db": 0.4731415803016304, "全帯域_40ms_rms_envelope_nrmse": 0.03552709649271113, "全帯域_40ms_envelope_correlation": 0.9983247286803986, "全帯域_40ms_modulation_mae_db": 0.4534274987646209, "300–1000 Hz_10ms_log_rms_mae_db": 1.5698510879388583, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10558662711380598, "300–1000 Hz_10ms_envelope_correlation": 0.9908053685089518, "300–1000 Hz_10ms_modulation_mae_db": 0.6460113844481845, "300–1000 Hz_40ms_log_rms_mae_db": 0.8974742030405908, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05329535439903644, "300–1000 Hz_40ms_envelope_correlation": 0.997651037197835, "300–1000 Hz_40ms_modulation_mae_db": 0.5532081518096672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.474611661967091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14600632884318437, "1000–3000 Hz_10ms_envelope_correlation": 0.9847649370313082, "1000–3000 Hz_10ms_modulation_mae_db": 0.9105921124659165, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8858844955686817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08573913471611944, "1000–3000 Hz_40ms_envelope_correlation": 0.995478893955032, "1000–3000 Hz_40ms_modulation_mae_db": 1.1563133916918227, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7016496764917994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16048182703312672, "3000–6000 Hz_10ms_envelope_correlation": 0.9834402549657899, "3000–6000 Hz_10ms_modulation_mae_db": 0.3334862833432101, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0160158342588117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0877934440215647, "3000–6000 Hz_40ms_envelope_correlation": 0.9950942156196169, "3000–6000 Hz_40ms_modulation_mae_db": 0.33894750558099024, "mr_stft_db": 5.665704497516295, "band_mae_db": 0.1736596730047156, "rms_gain_error_db": -0.09864242340014934, "output_peak": 0.5347022058246758}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8961704421183929, "全帯域_10ms_rms_envelope_nrmse": 0.06846226943510784, "全帯域_10ms_envelope_correlation": 0.9939968107321621, "全帯域_10ms_modulation_mae_db": 0.32386584424420195, "全帯域_40ms_log_rms_mae_db": 0.49357152438655516, "全帯域_40ms_rms_envelope_nrmse": 0.040149241034362994, "全帯域_40ms_envelope_correlation": 0.9979350033586349, "全帯域_40ms_modulation_mae_db": 0.31585723666642557, "300–1000 Hz_10ms_log_rms_mae_db": 1.6244635521020423, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10518418883038168, "300–1000 Hz_10ms_envelope_correlation": 0.9909580682677195, "300–1000 Hz_10ms_modulation_mae_db": 0.6729113892272686, "300–1000 Hz_40ms_log_rms_mae_db": 1.009325076861392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06709284572754544, "300–1000 Hz_40ms_envelope_correlation": 0.9962315145968018, "300–1000 Hz_40ms_modulation_mae_db": 0.6764295566026344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.508392630624575, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14867531584222765, "1000–3000 Hz_10ms_envelope_correlation": 0.983508190352638, "1000–3000 Hz_10ms_modulation_mae_db": 0.0563162376438272, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8471365017755652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09094500115563657, "1000–3000 Hz_40ms_envelope_correlation": 0.9938261006357122, "1000–3000 Hz_40ms_modulation_mae_db": 0.15239664946947937, "3000–6000 Hz_10ms_log_rms_mae_db": 1.77353788204526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15227523952008984, "3000–6000 Hz_10ms_envelope_correlation": 0.9844476653012002, "3000–6000 Hz_10ms_modulation_mae_db": 0.25412046590702614, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0905310170900504, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08973707843138955, "3000–6000 Hz_40ms_envelope_correlation": 0.9943054207926607, "3000–6000 Hz_40ms_modulation_mae_db": 0.31656144131219305, "mr_stft_db": 5.655262589817002, "band_mae_db": 0.14865082745260896, "rms_gain_error_db": -0.13078401180976998, "output_peak": 0.6439260329274719}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8516606931787553, "全帯域_10ms_rms_envelope_nrmse": 0.07450385222037194, "全帯域_10ms_envelope_correlation": 0.9927126280773969, "全帯域_10ms_modulation_mae_db": 0.3228352649193369, "全帯域_40ms_log_rms_mae_db": 0.46758039048528915, "全帯域_40ms_rms_envelope_nrmse": 0.040956169201908324, "全帯域_40ms_envelope_correlation": 0.9976893456945106, "全帯域_40ms_modulation_mae_db": 0.5686126492691448, "300–1000 Hz_10ms_log_rms_mae_db": 1.553554457926985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11663323870794612, "300–1000 Hz_10ms_envelope_correlation": 0.9886733166878114, "300–1000 Hz_10ms_modulation_mae_db": 0.7882962339903102, "300–1000 Hz_40ms_log_rms_mae_db": 0.9441994036367132, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06981352426156691, "300–1000 Hz_40ms_envelope_correlation": 0.9957124591529817, "300–1000 Hz_40ms_modulation_mae_db": 0.3166322153795503, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5456060329566732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14992320639797488, "1000–3000 Hz_10ms_envelope_correlation": 0.9834056390219822, "1000–3000 Hz_10ms_modulation_mae_db": 0.6749083059398467, "1000–3000 Hz_40ms_log_rms_mae_db": 0.949504558592864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09101202837864876, "1000–3000 Hz_40ms_envelope_correlation": 0.9940883851384892, "1000–3000 Hz_40ms_modulation_mae_db": 0.6374973076889895, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7279071083290993, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16717626075457964, "3000–6000 Hz_10ms_envelope_correlation": 0.9811130112213196, "3000–6000 Hz_10ms_modulation_mae_db": 0.40871325516966567, "3000–6000 Hz_40ms_log_rms_mae_db": 1.054982299435174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09856336573949086, "3000–6000 Hz_40ms_envelope_correlation": 0.9930601613032871, "3000–6000 Hz_40ms_modulation_mae_db": 0.23671159039009768, "mr_stft_db": 5.645066736397134, "band_mae_db": 0.16266612123006266, "rms_gain_error_db": -0.09147088512037327, "output_peak": 0.6210638395022321}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8800312912122499, "全帯域_10ms_rms_envelope_nrmse": 0.07235621815174337, "全帯域_10ms_envelope_correlation": 0.9931797589482758, "全帯域_10ms_modulation_mae_db": 0.21948858662189372, "全帯域_40ms_log_rms_mae_db": 0.5002674408748735, "全帯域_40ms_rms_envelope_nrmse": 0.0416982132437503, "全帯域_40ms_envelope_correlation": 0.99765524177113, "全帯域_40ms_modulation_mae_db": 0.19144492480476896, "300–1000 Hz_10ms_log_rms_mae_db": 1.6065260427517174, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10822055589980546, "300–1000 Hz_10ms_envelope_correlation": 0.9903146470529163, "300–1000 Hz_10ms_modulation_mae_db": 0.4544543293351407, "300–1000 Hz_40ms_log_rms_mae_db": 0.999111066637694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06624477266049442, "300–1000 Hz_40ms_envelope_correlation": 0.996242089657437, "300–1000 Hz_40ms_modulation_mae_db": 0.2979828265705522, "1000–3000 Hz_10ms_log_rms_mae_db": 1.452580697862892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13699468528614103, "1000–3000 Hz_10ms_envelope_correlation": 0.9864428869345638, "1000–3000 Hz_10ms_modulation_mae_db": 0.5508035460269334, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8508985781714121, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0836337060791925, "1000–3000 Hz_40ms_envelope_correlation": 0.9953900712475698, "1000–3000 Hz_40ms_modulation_mae_db": 0.30537156406065463, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6794388831945715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.178773780494508, "3000–6000 Hz_10ms_envelope_correlation": 0.9808902211183885, "3000–6000 Hz_10ms_modulation_mae_db": 1.2628691360969624, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0392568904341442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10239427874358051, "3000–6000 Hz_40ms_envelope_correlation": 0.9944419992516976, "3000–6000 Hz_40ms_modulation_mae_db": 1.1274636341161164, "mr_stft_db": 5.657401607475502, "band_mae_db": 0.21699463288907173, "rms_gain_error_db": -0.10408939526607279, "output_peak": 0.5925037693270897}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8300685505702663, "全帯域_10ms_rms_envelope_nrmse": 0.06686790928671074, "全帯域_10ms_envelope_correlation": 0.9941783812811558, "全帯域_10ms_modulation_mae_db": 0.23240372202850643, "全帯域_40ms_log_rms_mae_db": 0.42487453496644745, "全帯域_40ms_rms_envelope_nrmse": 0.031353799450433095, "全帯域_40ms_envelope_correlation": 0.9987288887152685, "全帯域_40ms_modulation_mae_db": 0.3689846616097577, "300–1000 Hz_10ms_log_rms_mae_db": 1.5654491305202365, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09719500961109664, "300–1000 Hz_10ms_envelope_correlation": 0.9922114953335471, "300–1000 Hz_10ms_modulation_mae_db": 0.5314884524550417, "300–1000 Hz_40ms_log_rms_mae_db": 0.9134340034738445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0484455691707141, "300–1000 Hz_40ms_envelope_correlation": 0.9980606033729502, "300–1000 Hz_40ms_modulation_mae_db": 0.33209056001444964, "1000–3000 Hz_10ms_log_rms_mae_db": 1.511517277405329, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1528698748262375, "1000–3000 Hz_10ms_envelope_correlation": 0.982426479331896, "1000–3000 Hz_10ms_modulation_mae_db": 0.6412839501752944, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8613170291379165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07250893959574978, "1000–3000 Hz_40ms_envelope_correlation": 0.995999373694836, "1000–3000 Hz_40ms_modulation_mae_db": 0.6687169387824691, "3000–6000 Hz_10ms_log_rms_mae_db": 1.635065966519795, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14312006346967787, "3000–6000 Hz_10ms_envelope_correlation": 0.9860688276465124, "3000–6000 Hz_10ms_modulation_mae_db": 0.3331104630600697, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9867704108903068, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08331360537767986, "3000–6000 Hz_40ms_envelope_correlation": 0.995103337895908, "3000–6000 Hz_40ms_modulation_mae_db": 0.32204237635856703, "mr_stft_db": 5.662667847550114, "band_mae_db": 0.1189164214367836, "rms_gain_error_db": -0.09798835032704163, "output_peak": 0.7264204836776311}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8083187988145603, "全帯域_10ms_rms_envelope_nrmse": 0.06282626308611822, "全帯域_10ms_envelope_correlation": 0.9949101452908385, "全帯域_10ms_modulation_mae_db": 0.492212079980424, "全帯域_40ms_log_rms_mae_db": 0.4315507631621175, "全帯域_40ms_rms_envelope_nrmse": 0.036619578473219574, "全帯域_40ms_envelope_correlation": 0.998237790708162, "全帯域_40ms_modulation_mae_db": 0.5283890959603785, "300–1000 Hz_10ms_log_rms_mae_db": 1.5464258546704865, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09546201742738214, "300–1000 Hz_10ms_envelope_correlation": 0.9925172328246027, "300–1000 Hz_10ms_modulation_mae_db": 0.3561181755744858, "300–1000 Hz_40ms_log_rms_mae_db": 0.9156506165235934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052430278700043624, "300–1000 Hz_40ms_envelope_correlation": 0.9976970225467916, "300–1000 Hz_40ms_modulation_mae_db": 0.41284172911533706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5157052046285155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14335578125263754, "1000–3000 Hz_10ms_envelope_correlation": 0.9851241754530133, "1000–3000 Hz_10ms_modulation_mae_db": 0.45874570906251483, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9065182159666487, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08105213741461116, "1000–3000 Hz_40ms_envelope_correlation": 0.9958109070210239, "1000–3000 Hz_40ms_modulation_mae_db": 0.6435672959310246, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7208307119323036, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.174278886831876, "3000–6000 Hz_10ms_envelope_correlation": 0.979712127019429, "3000–6000 Hz_10ms_modulation_mae_db": 0.4936636198954007, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0320127922129279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10056068786330324, "3000–6000 Hz_40ms_envelope_correlation": 0.9929113028013714, "3000–6000 Hz_40ms_modulation_mae_db": 0.37657717611267055, "mr_stft_db": 5.66106047960216, "band_mae_db": 0.2038334564353396, "rms_gain_error_db": -0.10604570946608011, "output_peak": 0.6247746940855792}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.692127836699606, "spectral_convergence": 0.23464044086107994, "spectral_convergence_status": "ok", "amplitude_floor": 1.6781522163747868e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.525712373932618, "spectral_convergence": 0.2320135585472816, "spectral_convergence_status": "ok", "amplitude_floor": 1.6368745556803986e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.763974645509051, "spectral_convergence": 0.23829658521108474, "spectral_convergence_status": "ok", "amplitude_floor": 1.5514964797831996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006090282805006308, "output_power": 0.0006462438092058745, "delta_db": 0.2575893597004276, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005054165098664251, "output_power": 0.0049386840615038225, "delta_db": -0.10038180143926817, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0022275244303280044, "output_power": 0.0021546787835973816, "delta_db": -0.1443994052294203, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006076310340369887, "output_power": 0.0005619117941737145, "delta_db": -0.33971799077149056, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001299593081283461, "output_power": 0.0001331994252726266, "delta_db": 0.10694960348186452, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003180661511978585, "output_power": 0.0003169058700371502, "delta_db": -0.015871699840220025, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1608183100771152, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09558426560534639, "pitch": null, "output_rms_dbfs": -20.5791136211532, "output_peak": 0.6559213389659905, "reference_rms_dbfs": -20.48352935554785, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.660604952047091}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002515751859318019, "log_rms_mae_db": 0.8895092229720947, "rms_envelope_nrmse": 0.07383972395095457, "envelope_correlation": 0.992856438350343, "rms_gain_error_db": -0.09558155973243823, "reference_mean_power": 0.008953808913712473, "output_mean_power": 0.008758901970383225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.40642035802602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010131590544257714, "output_power": 0.0009702279349973344, "reference_active": true, "delta_db": -0.18803855233140343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021577628078402175, "output_power": 0.00019917187266678567, "reference_active": true, "delta_db": -0.34775696395233024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.554717092269403e-05, "output_power": 4.6531529649151996e-05, "reference_active": true, "delta_db": 0.0928592274255876}], "active_band_count": 3, "active_band_mae_db": 0.20955158123644044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002371340103862925, "log_rms_mae_db": 0.5061251420947285, "rms_envelope_nrmse": 0.041816403373567054, "envelope_correlation": 0.9975923823640591, "rms_gain_error_db": -0.09553429629363916, "reference_mean_power": 0.008997361160684636, "output_mean_power": 0.008801601955578813}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.083478820328128e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000878683378715502, "output_power": 0.0008366854096733728, "reference_active": true, "delta_db": -0.2127021559291091}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.731034174913729e-05, "output_power": 9.247721793629648e-05, "reference_active": true, "delta_db": -0.22124241724274293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4907380289326384e-06, "output_power": 2.1857020006452106e-06, "reference_active": true, "delta_db": -0.5673710188232023}], "active_band_count": 3, "active_band_mae_db": 0.3337718639983514}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019737503490519328, "log_rms_mae_db": 1.63248010459893, "rms_envelope_nrmse": 0.123971073755053, "envelope_correlation": 0.9871845589904612, "rms_gain_error_db": -0.08393232474787776, "reference_mean_power": 0.0030829546180243575, "output_mean_power": 0.003023945079910918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6914511369831456e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008052884179451131, "output_power": 0.0007801805470321357, "reference_active": true, "delta_db": -0.13756335457917793}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019776686451848026, "output_power": 0.00016996587321634667, "reference_active": true, "delta_db": -0.6579179825993432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9868777263914452e-05, "output_power": 3.0098678135251175e-05, "reference_active": true, "delta_db": 1.8037628159975558}], "active_band_count": 3, "active_band_mae_db": 0.8664147177253589}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017438355068165795, "log_rms_mae_db": 1.0114322761031664, "rms_envelope_nrmse": 0.07361996952037156, "envelope_correlation": 0.9952437398302604, "rms_gain_error_db": -0.08405353837418952, "reference_mean_power": 0.0030981947388459344, "output_mean_power": 0.003038808680228782}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5303812378070619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007441051144552582, "output_power": 0.0007177425820420012, "reference_active": true, "delta_db": -0.1566557687788927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.887152997212749e-05, "output_power": 8.860123367487467e-05, "reference_active": true, "delta_db": -0.4763148562775966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5498600424852708e-06, "output_power": 1.863832303314725e-06, "reference_active": true, "delta_db": 0.801143528033383}], "active_band_count": 3, "active_band_mae_db": 0.4780380510299574}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011259893106980884, "log_rms_mae_db": 1.5026551454193369, "rms_envelope_nrmse": 0.1370624297193433, "envelope_correlation": 0.9864909119114786, "rms_gain_error_db": -0.3191682827265212, "reference_mean_power": 0.0006356182176644616, "output_mean_power": 0.0005905810503707316}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.9581273226652876e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002027625859711201, "output_power": 0.00018642840984671154, "reference_active": true, "delta_db": -0.3647572619675089}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.693176540437241e-05, "output_power": 2.2221799132157304e-05, "reference_active": true, "delta_db": -0.834856051954054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.128022173574573e-06, "output_power": 5.824826464677753e-06, "reference_active": true, "delta_db": -0.8768605078373802}], "active_band_count": 3, "active_band_mae_db": 0.6921579405863144}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001021231893168278, "log_rms_mae_db": 0.8842040059152881, "rms_envelope_nrmse": 0.07870304099526022, "envelope_correlation": 0.9960887051477185, "rms_gain_error_db": -0.31922651906937416, "reference_mean_power": 0.0006388062286465308, "output_mean_power": 0.0005935352136229531}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.69069653778161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018976599367877858, "output_power": 0.00017433367025771848, "reference_active": true, "delta_db": -0.36837115424908523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3827689740852905e-05, "output_power": 1.1232133937258737e-05, "reference_active": true, "delta_db": -0.9028735293729937}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.22079217642337e-07, "output_power": 3.6135063076916357e-07, "reference_active": true, "delta_db": -0.674651517136544}], "active_band_count": 3, "active_band_mae_db": 0.648632066919541}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.113996831641667e-05, "log_rms_mae_db": 1.6872072004928267, "rms_envelope_nrmse": 0.16111258695923839, "envelope_correlation": 0.9830590421823715, "rms_gain_error_db": 0.18878750021140872, "reference_mean_power": 0.0001805297653048898, "output_mean_power": 0.00018855044753952088}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6198199988921157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.959931264840548e-05, "output_power": 0.0001054694965825718, "reference_active": true, "delta_db": 0.24870531785603792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361167220741e-06, "output_power": 9.3056539356242e-06, "reference_active": true, "delta_db": -0.02583105529469197}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4041286689847167e-06, "output_power": 2.3154323196700716e-06, "reference_active": true, "delta_db": -0.16325616354593064}], "active_band_count": 3, "active_band_mae_db": 0.1459308455655535}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.456026972891915e-05, "log_rms_mae_db": 1.052180826603865, "rms_envelope_nrmse": 0.08980011307110725, "envelope_correlation": 0.9946476000678196, "rms_gain_error_db": 0.18877263178104325, "reference_mean_power": 0.00018143628465930542, "output_mean_power": 0.00018949659350883855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.558011664744216e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.473019444861243e-05, "output_power": 0.0001005880415459896, "reference_active": true, "delta_db": 0.2605792373254701}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179317052925175e-06, "output_power": 4.382365789449055e-06, "reference_active": true, "delta_db": 0.2060330563662054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8116784694632823e-07, "output_power": 1.59911791471037e-07, "reference_active": true, "delta_db": -0.5420063442195141}], "active_band_count": 3, "active_band_mae_db": 0.33620621263706324}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.883287758838418, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ca23604e3b09f0daf4526a94c6bf03c7325c8215f59dfb5335ee936259044632", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9845979739232082, "standard_deviation": 0.001445034509101012, "minimum": 0.982426479331896, "maximum": 0.9864909119114786, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5020217065294543, "standard_deviation": 0.027834249209028845, "minimum": 1.452580697862892, "maximum": 1.5456060329566732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6059967084279273, "standard_deviation": 0.26704998277533226, "minimum": 0.0563162376438272, "maximum": 0.9105921124659165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14478440923869323, "standard_deviation": 0.005765846587706504, "minimum": 0.13699468528614103, "maximum": 0.1528698748262375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9952347379047894, "standard_deviation": 0.0008476492949037937, "minimum": 0.9938261006357122, "maximum": 0.9960887051477185, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8855272976541266, "standard_deviation": 0.03374208434015941, "minimum": 0.8471365017755652, "maximum": 0.949504558592864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.606344929808531, "standard_deviation": 0.2947571889760311, "minimum": 0.15239664946947937, "maximum": 1.1563133916918227, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08294970636133056, "standard_deviation": 0.00628488277783428, "minimum": 0.07250893959574978, "maximum": 0.09101202837864876, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9824656681842572, "standard_deviation": 0.0021471548075972916, "minimum": 0.979712127019429, "maximum": 0.9860688276465124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.702733550595638, "standard_deviation": 0.040321100260478686, "minimum": 1.635065966519795, "maximum": 1.77353788204526, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4679819510939355, "standard_deviation": 0.34302961552128597, "minimum": 0.1459308455655535, "maximum": 1.2628691360969624, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16371403221643022, "standard_deviation": 0.011950208757721107, "minimum": 0.14312006346967787, "maximum": 0.178773780494508, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943437852673211, "standard_deviation": 0.0008986435062178944, "minimum": 0.9929113028013714, "maximum": 0.9951862444062071, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0395482120106547, "standard_deviation": 0.03028214633350577, "minimum": 0.9867704108903068, "maximum": 1.0905310170900504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.44093198981106896, "standard_deviation": 0.28514888463543464, "minimum": 0.23671159039009768, "maximum": 1.1274636341161164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09256448306593312, "standard_deviation": 0.00695148919114177, "minimum": 0.08331360537767986, "maximum": 0.10239427874358051, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904699159867183, "standard_deviation": 0.0017732598855905877, "minimum": 0.9871845589904612, "maximum": 0.9925172328246027, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5853083600842592, "standard_deviation": 0.03243565483333507, "minimum": 1.5464258546704865, "maximum": 1.63248010459893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5828421115721807, "standard_deviation": 0.1930409316123734, "minimum": 0.3470422098216548, "maximum": 0.8664147177253589, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10706003605501914, "standard_deviation": 0.009445494426616038, "minimum": 0.09546201742738214, "maximum": 0.123971073755053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967212884895901, "standard_deviation": 0.0010229497419366573, "minimum": 0.9952437398302604, "maximum": 0.9980606033729502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9546876433088438, "standard_deviation": 0.04605714062891244, "minimum": 0.8974742030405908, "maximum": 1.0114322761031664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44449041609228923, "standard_deviation": 0.13100862755285575, "minimum": 0.2979828265705522, "maximum": 0.6764295566026344, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06147610533077637, "standard_deviation": 0.00918221942357463, "minimum": 0.0484455691707141, "maximum": 0.07361996952037156, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16312393622152624, "standard_deviation": 0.035348981353398666, "minimum": 0.1189164214367836, "maximum": 0.21699463288907173, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.659044557842204, "standard_deviation": 0.0066234524979011675, "minimum": 5.645066736397134, "maximum": 5.665704497516295, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6259549076359056, "standard_deviation": 0.054897796964223415, "minimum": 0.5347022058246758, "maximum": 0.7264204836776311, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10610465356939591, "standard_deviation": 0.014082045060187974, "minimum": -0.13078401180976998, "maximum": -0.09147088512037327, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9937537284258322, "standard_deviation": 0.0007703624404842344, "minimum": 0.9927126280773969, "maximum": 0.9949101452908385, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8658584896144639, "standard_deviation": 0.03220256477695315, "minimum": 0.8083187988145603, "maximum": 0.8961704421183929, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2745511636229827, "standard_deviation": 0.10358668266025622, "minimum": 0.16466121553492868, "maximum": 0.492212079980424, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06924506762237509, "standard_deviation": 0.004040169805286581, "minimum": 0.06282626308611822, "maximum": 0.07450385222037194, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980580953956115, "standard_deviation": 0.000404058581979342, "minimum": 0.9975923823640591, "maximum": 0.9987288887152685, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4720921504947, "standard_deviation": 0.030153833881778985, "minimum": 0.42487453496644745, "maximum": 0.5061251420947285, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.378455461928559, "standard_deviation": 0.12962918447994137, "minimum": 0.19144492480476896, "maximum": 0.5686126492691448, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038097787438227386, "standard_deviation": 0.003693758660034867, "minimum": 0.031353799450433095, "maximum": 0.041816403373567054, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0041", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.307793140411377, "predicted_mos": 2.307793140411377, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c0817566336c51bbc6248abd54a07d738e48c6166b66d7069d8116c7424e865c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.22112900018692, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9022524549058175, "全帯域_10ms_rms_envelope_nrmse": 0.07561175191378762, "全帯域_10ms_envelope_correlation": 0.9924990555607607, "全帯域_10ms_modulation_mae_db": 0.20887376338072408, "全帯域_40ms_log_rms_mae_db": 0.5206659300149995, "全帯域_40ms_rms_envelope_nrmse": 0.042946909405619446, "全帯域_40ms_envelope_correlation": 0.9974493350561565, "全帯域_40ms_modulation_mae_db": 0.41151669154393405, "300–1000 Hz_10ms_log_rms_mae_db": 1.6235933547587291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.127981468042503, "300–1000 Hz_10ms_envelope_correlation": 0.9863302248130434, "300–1000 Hz_10ms_modulation_mae_db": 0.831875771168939, "300–1000 Hz_40ms_log_rms_mae_db": 0.9517554464907524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07708274751551587, "300–1000 Hz_40ms_envelope_correlation": 0.994742158396504, "300–1000 Hz_40ms_modulation_mae_db": 0.4667544394429668, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4452217422025573, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13256428924830949, "1000–3000 Hz_10ms_envelope_correlation": 0.9881747057566388, "1000–3000 Hz_10ms_modulation_mae_db": 0.8827736396435778, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8216496957522053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08029665329206516, "1000–3000 Hz_40ms_envelope_correlation": 0.9967750557726137, "1000–3000 Hz_40ms_modulation_mae_db": 0.7963023462281814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6338700980475103, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1621299828114573, "3000–6000 Hz_10ms_envelope_correlation": 0.9829549097711543, "3000–6000 Hz_10ms_modulation_mae_db": 0.19292208321405782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0072098053014698, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09180749082326806, "3000–6000 Hz_40ms_envelope_correlation": 0.9945011610977508, "3000–6000 Hz_40ms_modulation_mae_db": 0.32848842270550715, "mr_stft_db": 5.608020666298572, "band_mae_db": 0.17250365146818106, "rms_gain_error_db": -0.095019984473801, "output_peak": 0.6541495416344958}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8818493046708644, "全帯域_10ms_rms_envelope_nrmse": 0.06616090720821624, "全帯域_10ms_envelope_correlation": 0.9944213650449183, "全帯域_10ms_modulation_mae_db": 0.1963459843361163, "全帯域_40ms_log_rms_mae_db": 0.4843248825409873, "全帯域_40ms_rms_envelope_nrmse": 0.03757185910253029, "全帯域_40ms_envelope_correlation": 0.9982402037236096, "全帯域_40ms_modulation_mae_db": 0.2263747491307813, "300–1000 Hz_10ms_log_rms_mae_db": 1.5396963793399876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10374072859115785, "300–1000 Hz_10ms_envelope_correlation": 0.9911962149919609, "300–1000 Hz_10ms_modulation_mae_db": 0.44360646389086034, "300–1000 Hz_40ms_log_rms_mae_db": 0.8694333555592905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05998069085103145, "300–1000 Hz_40ms_envelope_correlation": 0.9970412251414612, "300–1000 Hz_40ms_modulation_mae_db": 0.6517696772850141, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4745178929730898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14309270699767357, "1000–3000 Hz_10ms_envelope_correlation": 0.9849408234730763, "1000–3000 Hz_10ms_modulation_mae_db": 0.7573090278552082, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8525245389510707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08276749823808745, "1000–3000 Hz_40ms_envelope_correlation": 0.9952062221239892, "1000–3000 Hz_40ms_modulation_mae_db": 0.4922061769178036, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6271365069596526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16740832217876786, "3000–6000 Hz_10ms_envelope_correlation": 0.9818887355779057, "3000–6000 Hz_10ms_modulation_mae_db": 0.5308748884524749, "3000–6000 Hz_40ms_log_rms_mae_db": 0.965780142558592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08475601757647262, "3000–6000 Hz_40ms_envelope_correlation": 0.9954217044384093, "3000–6000 Hz_40ms_modulation_mae_db": 0.47053256441437274, "mr_stft_db": 5.610797303580103, "band_mae_db": 0.1434691219151036, "rms_gain_error_db": -0.13246009677286982, "output_peak": 0.5704590440068156}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9198388329709352, "全帯域_10ms_rms_envelope_nrmse": 0.07127775830324806, "全帯域_10ms_envelope_correlation": 0.9933363659043524, "全帯域_10ms_modulation_mae_db": 0.18265737663660733, "全帯域_40ms_log_rms_mae_db": 0.5070801876082277, "全帯域_40ms_rms_envelope_nrmse": 0.03740889848382097, "全帯域_40ms_envelope_correlation": 0.9980831654657609, "全帯域_40ms_modulation_mae_db": 0.4219025329611094, "300–1000 Hz_10ms_log_rms_mae_db": 1.5597548872835898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11178262197469192, "300–1000 Hz_10ms_envelope_correlation": 0.9896573518209231, "300–1000 Hz_10ms_modulation_mae_db": 0.6566238656539755, "300–1000 Hz_40ms_log_rms_mae_db": 0.9108288675629913, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057605816741503994, "300–1000 Hz_40ms_envelope_correlation": 0.9972026968156238, "300–1000 Hz_40ms_modulation_mae_db": 0.6068214536460982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.449973058141298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14482486493551208, "1000–3000 Hz_10ms_envelope_correlation": 0.9851070467876459, "1000–3000 Hz_10ms_modulation_mae_db": 0.8022024094509425, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8281737649525257, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0865668556950877, "1000–3000 Hz_40ms_envelope_correlation": 0.9954393019202767, "1000–3000 Hz_40ms_modulation_mae_db": 1.0956729869963928, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6779932913546476, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16029672623290062, "3000–6000 Hz_10ms_envelope_correlation": 0.9833434861073331, "3000–6000 Hz_10ms_modulation_mae_db": 0.3300565458734846, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9814381259772412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08696187054010895, "3000–6000 Hz_40ms_envelope_correlation": 0.9950941622690096, "3000–6000 Hz_40ms_modulation_mae_db": 0.2520763444552035, "mr_stft_db": 5.614025690441256, "band_mae_db": 0.1599708526961137, "rms_gain_error_db": -0.08752643511028255, "output_peak": 0.5357275475373294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8961151420171647, "全帯域_10ms_rms_envelope_nrmse": 0.06753640679996967, "全帯域_10ms_envelope_correlation": 0.9941276884158518, "全帯域_10ms_modulation_mae_db": 0.2807111571439802, "全帯域_40ms_log_rms_mae_db": 0.49439275166524427, "全帯域_40ms_rms_envelope_nrmse": 0.03861762604753001, "全帯域_40ms_envelope_correlation": 0.998065363908062, "全帯域_40ms_modulation_mae_db": 0.31336757769178997, "300–1000 Hz_10ms_log_rms_mae_db": 1.6200955231986716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10805051656013463, "300–1000 Hz_10ms_envelope_correlation": 0.9903688751746494, "300–1000 Hz_10ms_modulation_mae_db": 0.6082323746762065, "300–1000 Hz_40ms_log_rms_mae_db": 0.9380379582288985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06666908939295843, "300–1000 Hz_40ms_envelope_correlation": 0.9961871592132067, "300–1000 Hz_40ms_modulation_mae_db": 0.7484549064308, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4629162398907865, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13319298450909614, "1000–3000 Hz_10ms_envelope_correlation": 0.987175319492098, "1000–3000 Hz_10ms_modulation_mae_db": 0.34617010351960226, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7863563219111331, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07651017245003104, "1000–3000 Hz_40ms_envelope_correlation": 0.996193794180109, "1000–3000 Hz_40ms_modulation_mae_db": 0.2112904791617072, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7287352315115951, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15016428194772194, "3000–6000 Hz_10ms_envelope_correlation": 0.9848797036229597, "3000–6000 Hz_10ms_modulation_mae_db": 0.2488791330741121, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0384102337444268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08560174068389904, "3000–6000 Hz_40ms_envelope_correlation": 0.9948180948088231, "3000–6000 Hz_40ms_modulation_mae_db": 0.3523659854116243, "mr_stft_db": 5.607591138729929, "band_mae_db": 0.14299969968269646, "rms_gain_error_db": -0.11961622513555342, "output_peak": 0.6676077086103053}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8568787998862829, "全帯域_10ms_rms_envelope_nrmse": 0.07354883725474, "全帯域_10ms_envelope_correlation": 0.9929160031012227, "全帯域_10ms_modulation_mae_db": 0.3561405035543839, "全帯域_40ms_log_rms_mae_db": 0.474913986198221, "全帯域_40ms_rms_envelope_nrmse": 0.04012793920624182, "全帯域_40ms_envelope_correlation": 0.997803061588827, "全帯域_40ms_modulation_mae_db": 0.48384568032821146, "300–1000 Hz_10ms_log_rms_mae_db": 1.5760612232491629, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1189450806778653, "300–1000 Hz_10ms_envelope_correlation": 0.9882144199641016, "300–1000 Hz_10ms_modulation_mae_db": 0.84780829013312, "300–1000 Hz_40ms_log_rms_mae_db": 0.9072096941630771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06898482031352754, "300–1000 Hz_40ms_envelope_correlation": 0.9958335002889954, "300–1000 Hz_40ms_modulation_mae_db": 0.3304278201588782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5018761810908972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14563170934808947, "1000–3000 Hz_10ms_envelope_correlation": 0.9844314734713517, "1000–3000 Hz_10ms_modulation_mae_db": 0.5714826974387401, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9032427228445629, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0890505315564685, "1000–3000 Hz_40ms_envelope_correlation": 0.9944163238020827, "1000–3000 Hz_40ms_modulation_mae_db": 0.7112059180388712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.687152934562123, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16492834069608547, "3000–6000 Hz_10ms_envelope_correlation": 0.9814718147631806, "3000–6000 Hz_10ms_modulation_mae_db": 0.3874832493494799, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9962213237646529, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09752247808079906, "3000–6000 Hz_40ms_envelope_correlation": 0.9931134412697168, "3000–6000 Hz_40ms_modulation_mae_db": 0.19097024020861883, "mr_stft_db": 5.595877726340514, "band_mae_db": 0.14719575383725136, "rms_gain_error_db": -0.09715989692391286, "output_peak": 0.6156586449038403}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9071307686115946, "全帯域_10ms_rms_envelope_nrmse": 0.0729895061353568, "全帯域_10ms_envelope_correlation": 0.9930625950845466, "全帯域_10ms_modulation_mae_db": 0.18464155182483588, "全帯域_40ms_log_rms_mae_db": 0.5289606352798308, "全帯域_40ms_rms_envelope_nrmse": 0.04184156873009611, "全帯域_40ms_envelope_correlation": 0.9976452939723076, "全帯域_40ms_modulation_mae_db": 0.12956755282158078, "300–1000 Hz_10ms_log_rms_mae_db": 1.5488053109176385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11135530744524645, "300–1000 Hz_10ms_envelope_correlation": 0.9897106149095738, "300–1000 Hz_10ms_modulation_mae_db": 0.47354343692944667, "300–1000 Hz_40ms_log_rms_mae_db": 0.8997202971812335, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06804020243414997, "300–1000 Hz_40ms_envelope_correlation": 0.995988749034228, "300–1000 Hz_40ms_modulation_mae_db": 0.34250715874453047, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4222100527545387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1325314853673616, "1000–3000 Hz_10ms_envelope_correlation": 0.9875204768762182, "1000–3000 Hz_10ms_modulation_mae_db": 0.4345015716001403, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8013490353346157, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08222847073691564, "1000–3000 Hz_40ms_envelope_correlation": 0.9957402254095103, "1000–3000 Hz_40ms_modulation_mae_db": 0.20456386914376132, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6333861375629048, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17770992915373413, "3000–6000 Hz_10ms_envelope_correlation": 0.9809961759438726, "3000–6000 Hz_10ms_modulation_mae_db": 1.310602686427677, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9944355844102365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10098135355496857, "3000–6000 Hz_40ms_envelope_correlation": 0.9945228295643365, "3000–6000 Hz_40ms_modulation_mae_db": 1.1530874301892462, "mr_stft_db": 5.6192443530900755, "band_mae_db": 0.20972864221536583, "rms_gain_error_db": -0.10751535051747091, "output_peak": 0.5984544746371825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8509525773388866, "全帯域_10ms_rms_envelope_nrmse": 0.06905711296265947, "全帯域_10ms_envelope_correlation": 0.9937642344132707, "全帯域_10ms_modulation_mae_db": 0.258448008986763, "全帯域_40ms_log_rms_mae_db": 0.43482570683308136, "全帯域_40ms_rms_envelope_nrmse": 0.034021638514438055, "全帯域_40ms_envelope_correlation": 0.9984507184285397, "全帯域_40ms_modulation_mae_db": 0.2705024193076954, "300–1000 Hz_10ms_log_rms_mae_db": 1.5608607332385052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10242292911063537, "300–1000 Hz_10ms_envelope_correlation": 0.9913036885948625, "300–1000 Hz_10ms_modulation_mae_db": 0.5293909247303993, "300–1000 Hz_40ms_log_rms_mae_db": 0.8919511615610126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054055415948789716, "300–1000 Hz_40ms_envelope_correlation": 0.9974920181873822, "300–1000 Hz_40ms_modulation_mae_db": 0.2768040059852657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.509640863215581, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1458688272708886, "1000–3000 Hz_10ms_envelope_correlation": 0.9841543340080973, "1000–3000 Hz_10ms_modulation_mae_db": 0.3684180625182223, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8534354121262053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07028241466030262, "1000–3000 Hz_40ms_envelope_correlation": 0.9964845317383246, "1000–3000 Hz_40ms_modulation_mae_db": 0.4368205974020256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.615672633033662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13977474303519566, "3000–6000 Hz_10ms_envelope_correlation": 0.9866552625876432, "3000–6000 Hz_10ms_modulation_mae_db": 0.4669449943408062, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9659832792340731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07787529642915345, "3000–6000 Hz_40ms_envelope_correlation": 0.9956568560336504, "3000–6000 Hz_40ms_modulation_mae_db": 0.32823014439882175, "mr_stft_db": 5.614071078754765, "band_mae_db": 0.14289642653922724, "rms_gain_error_db": -0.09232636717930198, "output_peak": 0.714995092518207}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8227585727376423, "全帯域_10ms_rms_envelope_nrmse": 0.06219542841076936, "全帯域_10ms_envelope_correlation": 0.9950534512903062, "全帯域_10ms_modulation_mae_db": 0.3993900608572154, "全帯域_40ms_log_rms_mae_db": 0.45370997950498637, "全帯域_40ms_rms_envelope_nrmse": 0.03736301987722666, "全帯域_40ms_envelope_correlation": 0.9982025355293614, "全帯域_40ms_modulation_mae_db": 0.4434599574702747, "300–1000 Hz_10ms_log_rms_mae_db": 1.5150501235727174, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09718138956775373, "300–1000 Hz_10ms_envelope_correlation": 0.9922194315046432, "300–1000 Hz_10ms_modulation_mae_db": 0.35465118565069514, "300–1000 Hz_40ms_log_rms_mae_db": 0.8425498006321466, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05461740491775392, "300–1000 Hz_40ms_envelope_correlation": 0.9974606214729329, "300–1000 Hz_40ms_modulation_mae_db": 0.3482366665792893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4748981573799356, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13946831262627857, "1000–3000 Hz_10ms_envelope_correlation": 0.986397011432298, "1000–3000 Hz_10ms_modulation_mae_db": 0.400986048335016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8419444769738397, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08412585510713722, "1000–3000 Hz_40ms_envelope_correlation": 0.9959173333464735, "1000–3000 Hz_40ms_modulation_mae_db": 0.6289692493355866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6970320695682442, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16742790173274735, "3000–6000 Hz_10ms_envelope_correlation": 0.9809583514337792, "3000–6000 Hz_10ms_modulation_mae_db": 0.5448892774883237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0274047819143917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09530158680180265, "3000–6000 Hz_40ms_envelope_correlation": 0.9934403332508616, "3000–6000 Hz_40ms_modulation_mae_db": 0.44776235833314826, "mr_stft_db": 5.614327376808298, "band_mae_db": 0.1992287976353049, "rms_gain_error_db": -0.11816236213678291, "output_peak": 0.643323378578037}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.643038039050674, "spectral_convergence": 0.2352932464425811, "spectral_convergence_status": "ok", "amplitude_floor": 1.6781522163747868e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.475034188324294, "spectral_convergence": 0.23263414771892604, "spectral_convergence_status": "ok", "amplitude_floor": 1.6368745556803986e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7059897715207475, "spectral_convergence": 0.2387613354363354, "spectral_convergence_status": "ok", "amplitude_floor": 1.5514964797831996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006090282805006308, "output_power": 0.0006489230946240894, "delta_db": 0.27555770847442274, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005054165098664251, "output_power": 0.004945347716140452, "delta_db": -0.09452591400069237, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0022275244303280044, "output_power": 0.0021659681272515707, "delta_db": -0.1217041426349274, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006076310340369887, "output_power": 0.0005452406234815637, "delta_db": -0.47051741094178245, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001299593081283461, "output_power": 0.00013129110394770967, "delta_db": 0.04427909461528731, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003180661511978585, "output_power": 0.00031599025535102704, "delta_db": -0.02843763814197413, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17250365146818106, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.095019984473801, "pitch": null, "output_rms_dbfs": -20.578549340021652, "output_peak": 0.6541495416344958, "reference_rms_dbfs": -20.48352935554785, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.608020666298572}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002515751859318019, "log_rms_mae_db": 0.9022524549058175, "rms_envelope_nrmse": 0.07561175191378762, "envelope_correlation": 0.9924990555607607, "rms_gain_error_db": -0.0950164430319082, "reference_mean_power": 0.008953808913712473, "output_mean_power": 0.008760041778518801}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.40642035802602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010131590544257714, "output_power": 0.0009627128011411637, "reference_active": true, "delta_db": -0.22180883352744538}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021577628078402175, "output_power": 0.00019682333872477858, "reference_active": true, "delta_db": -0.3992710863914263}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.554717092269403e-05, "output_power": 4.5489092177722016e-05, "reference_active": true, "delta_db": -0.0055413702233005424}], "active_band_count": 3, "active_band_mae_db": 0.20887376338072408}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002371340103862925, "log_rms_mae_db": 0.5206659300149995, "rms_envelope_nrmse": 0.042946909405619446, "envelope_correlation": 0.9974493350561565, "rms_gain_error_db": -0.09498110517027267, "reference_mean_power": 0.008997361160684636, "output_mean_power": 0.008802723148314729}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.083478820328128e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000878683378715502, "output_power": 0.0008285946054039409, "reference_active": true, "delta_db": -0.25490309829247915}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.731034174913729e-05, "output_power": 9.1630367508683e-05, "reference_active": true, "delta_db": -0.26119569354843086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4907380289326384e-06, "output_power": 2.11097418996456e-06, "reference_active": true, "delta_db": -0.718451282790892}], "active_band_count": 3, "active_band_mae_db": 0.41151669154393405}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019737503490519328, "log_rms_mae_db": 1.6235933547587291, "rms_envelope_nrmse": 0.127981468042503, "envelope_correlation": 0.9863302248130434, "rms_gain_error_db": -0.06138145378345258, "reference_mean_power": 0.0030829546180243575, "output_mean_power": 0.003039687842260612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6914511369831456e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008052884179451131, "output_power": 0.0007816103084085878, "reference_active": true, "delta_db": -0.12961174380450907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019776686451848026, "output_power": 0.00016561104982473364, "reference_active": true, "delta_db": -0.7706421788562372}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9868777263914452e-05, "output_power": 2.8688542201616477e-05, "reference_active": true, "delta_db": 1.5953733908460708}], "active_band_count": 3, "active_band_mae_db": 0.831875771168939}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017438355068165795, "log_rms_mae_db": 0.9517554464907524, "rms_envelope_nrmse": 0.07708274751551587, "envelope_correlation": 0.994742158396504, "rms_gain_error_db": -0.0614246270359627, "reference_mean_power": 0.0030981947388459344, "output_mean_power": 0.003054683713518261}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5303812378070619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007441051144552582, "output_power": 0.0007180129428065805, "reference_active": true, "delta_db": -0.15502016734765167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.887152997212749e-05, "output_power": 8.586402191441061e-05, "reference_active": true, "delta_db": -0.6126002743657342}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5498600424852708e-06, "output_power": 1.7929028986781496e-06, "reference_active": true, "delta_db": 0.6326428766155147}], "active_band_count": 3, "active_band_mae_db": 0.4667544394429668}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011259893106980884, "log_rms_mae_db": 1.4452217422025573, "rms_envelope_nrmse": 0.13256428924830949, "envelope_correlation": 0.9881747057566388, "rms_gain_error_db": -0.44554649662476364, "reference_mean_power": 0.0006356182176644616, "output_mean_power": 0.0005736429843167613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.9581273226652876e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002027625859711201, "output_power": 0.00018017733353114553, "reference_active": true, "delta_db": -0.5128766599667792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.693176540437241e-05, "output_power": 2.148750458666621e-05, "reference_active": true, "delta_db": -0.9807884038724718}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.128022173574573e-06, "output_power": 5.46390898537372e-06, "reference_active": true, "delta_db": -1.1546558550914823}], "active_band_count": 3, "active_band_mae_db": 0.8827736396435778}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001021231893168278, "log_rms_mae_db": 0.8216496957522053, "rms_envelope_nrmse": 0.08029665329206516, "envelope_correlation": 0.9967750557726137, "rms_gain_error_db": -0.44556875455277056, "reference_mean_power": 0.0006388062286465308, "output_mean_power": 0.0005765171972013189}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.69069653778161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018976599367877858, "output_power": 0.00016821716325535663, "reference_active": true, "delta_db": -0.5234808396980788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3827689740852905e-05, "output_power": 1.1032477527138116e-05, "reference_active": true, "delta_db": -0.9807657501211247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.22079217642337e-07, "output_power": 3.4429278678885466e-07, "reference_active": true, "delta_db": -0.8846604488653405}], "active_band_count": 3, "active_band_mae_db": 0.7963023462281814}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.113996831641667e-05, "log_rms_mae_db": 1.6338700980475103, "rms_envelope_nrmse": 0.1621299828114573, "envelope_correlation": 0.9829549097711543, "rms_gain_error_db": 0.19088400396373287, "reference_mean_power": 0.0001805297653048898, "output_mean_power": 0.00018864148994621492}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6198199988921157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.959931264840548e-05, "output_power": 0.00010711445871678285, "reference_active": true, "delta_db": 0.31591756213867567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361167220741e-06, "output_power": 9.301299718227822e-06, "reference_active": true, "delta_db": -0.02786364231344918}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4041286689847167e-06, "output_power": 2.2775042856845364e-06, "reference_active": true, "delta_db": -0.23498504519004867}], "active_band_count": 3, "active_band_mae_db": 0.19292208321405782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.456026972891915e-05, "log_rms_mae_db": 1.0072098053014698, "rms_envelope_nrmse": 0.09180749082326806, "envelope_correlation": 0.9945011610977508, "rms_gain_error_db": 0.19088723437105884, "reference_mean_power": 0.00018143628465930542, "output_mean_power": 0.00018958888285946715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.558011664744216e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.473019444861243e-05, "output_power": 0.00010221723444156051, "reference_active": true, "delta_db": 0.33035697874828934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179317052925175e-06, "output_power": 4.3605203520687e-06, "reference_active": true, "delta_db": 0.18432998829559347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8116784694632823e-07, "output_power": 1.62556108124389e-07, "reference_active": true, "delta_db": -0.47077830107263857}], "active_band_count": 3, "active_band_mae_db": 0.32848842270550715}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8828945919720372, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "25e402daa469b22a10b436f578d269c24bc6e04d81d511344f4c7b2b8f4b7117", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.985987648912178, "standard_deviation": 0.0015290028230081985, "minimum": 0.9841543340080973, "maximum": 0.9881747057566388, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4676567734560855, "standard_deviation": 0.029123053582193928, "minimum": 1.4222100527545387, "maximum": 1.509640863215581, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5704804450451813, "standard_deviation": 0.21530409469018422, "minimum": 0.34617010351960226, "maximum": 0.8827736396435778, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1396468975379012, "standard_deviation": 0.0060438002201060196, "minimum": 0.1325314853673616, "maximum": 0.1458688272708886, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957715985366724, "standard_deviation": 0.0007542748058440322, "minimum": 0.9944163238020827, "maximum": 0.9967750557726137, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8360844961057698, "standard_deviation": 0.03598865448665635, "minimum": 0.7863563219111331, "maximum": 0.9032427228445629, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5721289529030411, "standard_deviation": 0.301645436033788, "minimum": 0.20456386914376132, "maximum": 1.0956729869963928, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08147855646701191, "standard_deviation": 0.005902697378466769, "minimum": 0.07028241466030262, "maximum": 0.0890505315564685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9828935549759785, "standard_deviation": 0.0020240518144857412, "minimum": 0.9809583514337792, "maximum": 0.9866552625876432, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6626223628250423, "standard_deviation": 0.04059895566765132, "minimum": 1.615672633033662, "maximum": 1.7287352315115951, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5015816072775521, "standard_deviation": 0.3505132922527914, "minimum": 0.19292208321405782, "maximum": 1.310602686427677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1612300284735763, "standard_deviation": 0.011627004134875916, "minimum": 0.13977474303519566, "maximum": 0.17770992915373413, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945710728415698, "standard_deviation": 0.0008982219374274235, "minimum": 0.9931134412697168, "maximum": 0.9956568560336504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9971104096131356, "standard_deviation": 0.02653696014696797, "minimum": 0.965780142558592, "maximum": 1.0384102337444268, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4404391862645678, "standard_deviation": 0.30223519845446645, "minimum": 0.19097024020861883, "maximum": 1.1530874301892462, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09010097931130905, "standard_deviation": 0.007667675914083929, "minimum": 0.07787529642915345, "maximum": 0.10098135355496857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9898751027217199, "standard_deviation": 0.0018869621167086064, "minimum": 0.9863302248130434, "maximum": 0.9922194315046432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5679896919448753, "standard_deviation": 0.037724032808949856, "minimum": 1.5150501235727174, "maximum": 1.6235933547587291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5932165391042052, "standard_deviation": 0.17877084934465956, "minimum": 0.35465118565069514, "maximum": 0.84780829013312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11018250524624854, "standard_deviation": 0.009803815119501857, "minimum": 0.09718138956775373, "maximum": 0.127981468042503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964935160687918, "standard_deviation": 0.0009700324524407731, "minimum": 0.994742158396504, "maximum": 0.9974920181873822, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9014358226724253, "standard_deviation": 0.03500606288300821, "minimum": 0.8425498006321466, "maximum": 0.9517554464907524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4714720160341053, "standard_deviation": 0.17613974935231175, "minimum": 0.2768040059852657, "maximum": 0.7484549064308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06337952351440387, "standard_deviation": 0.008109450654841577, "minimum": 0.054055415948789716, "maximum": 0.07708274751551587, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1647491182486555, "standard_deviation": 0.02674350832589757, "minimum": 0.14289642653922724, "maximum": 0.20972864221536583, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.610494416755439, "standard_deviation": 0.007018376056153854, "minimum": 5.595877726340514, "maximum": 5.6192443530900755, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6250469290532765, "standard_deviation": 0.05711663549624832, "minimum": 0.5357275475373294, "maximum": 0.714995092518207, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10622333978124693, "standard_deviation": 0.015866890956809362, "minimum": -0.13246009677286982, "maximum": -0.08752643511028255, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936475948519037, "standard_deviation": 0.0008546264374384608, "minimum": 0.9924990555607607, "maximum": 0.9950534512903062, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8797220566423984, "standard_deviation": 0.03324327285370227, "minimum": 0.8227585727376423, "maximum": 0.9198388329709352, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.25840105084007825, "standard_deviation": 0.08229736670806934, "minimum": 0.18265737663660733, "maximum": 0.3993900608572154, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06979721362359341, "standard_deviation": 0.004422913368834563, "minimum": 0.06219542841076936, "maximum": 0.07561175191378762, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979924597090781, "standard_deviation": 0.00033390174834706677, "minimum": 0.9974493350561565, "maximum": 0.9984507184285397, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4873592574556973, "standard_deviation": 0.032369593880973144, "minimum": 0.43482570683308136, "maximum": 0.5289606352798308, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33756714515692215, "standard_deviation": 0.12299974823600847, "minimum": 0.12956755282158078, "maximum": 0.48384568032821146, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03873743242093792, "standard_deviation": 0.002842396046322046, "minimum": 0.034021638514438055, "maximum": 0.042946909405619446, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0041", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.487549066543579, "predicted_mos": 2.487549066543579, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0a497c6714f0e5e0b2ff9cba8b084870aec471bf15db8e6cd9de22d5ffc4ed5a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.413560600019991, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8394496727449801, "全帯域_10ms_rms_envelope_nrmse": 0.0666522500068029, "全帯域_10ms_envelope_correlation": 0.9942844219188246, "全帯域_10ms_modulation_mae_db": 0.303906001635523, "全帯域_40ms_log_rms_mae_db": 0.4329710988823158, "全帯域_40ms_rms_envelope_nrmse": 0.03421652518166452, "全帯域_40ms_envelope_correlation": 0.9985352134191572, "全帯域_40ms_modulation_mae_db": 0.32575522358318554, "300–1000 Hz_10ms_log_rms_mae_db": 1.8205218267065242, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10473019094197178, "300–1000 Hz_10ms_envelope_correlation": 0.990956912062306, "300–1000 Hz_10ms_modulation_mae_db": 0.4732999404739752, "300–1000 Hz_40ms_log_rms_mae_db": 1.2051742253104867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057680045577639165, "300–1000 Hz_40ms_envelope_correlation": 0.997166094570302, "300–1000 Hz_40ms_modulation_mae_db": 0.7069763818581395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8761517807068484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16588003363826942, "1000–3000 Hz_10ms_envelope_correlation": 0.9794447873673179, "1000–3000 Hz_10ms_modulation_mae_db": 0.4783771207477518, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1042298850816175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08935307392700688, "1000–3000 Hz_40ms_envelope_correlation": 0.9942337805918405, "1000–3000 Hz_40ms_modulation_mae_db": 0.8288257345487002, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3673281581315178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1957237625342072, "3000–6000 Hz_10ms_envelope_correlation": 0.9748774657398978, "3000–6000 Hz_10ms_modulation_mae_db": 0.18376585682617844, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4891912622531824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11388804921519198, "3000–6000 Hz_40ms_envelope_correlation": 0.9912906874369181, "3000–6000 Hz_40ms_modulation_mae_db": 0.23947478527089036, "mr_stft_db": 5.996790785372377, "band_mae_db": 0.18615353124808712, "rms_gain_error_db": -0.11864841331146361, "output_peak": 0.634947263674331}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7993087667203043, "全帯域_10ms_rms_envelope_nrmse": 0.060627122607906675, "全帯域_10ms_envelope_correlation": 0.9953493641836144, "全帯域_10ms_modulation_mae_db": 0.33642836453862746, "全帯域_40ms_log_rms_mae_db": 0.4060419243786254, "全帯域_40ms_rms_envelope_nrmse": 0.03245656987200885, "全帯域_40ms_envelope_correlation": 0.9987541462974936, "全帯域_40ms_modulation_mae_db": 0.3615970936837091, "300–1000 Hz_10ms_log_rms_mae_db": 1.739643205402163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10120693991494735, "300–1000 Hz_10ms_envelope_correlation": 0.991850078023969, "300–1000 Hz_10ms_modulation_mae_db": 0.622872838829929, "300–1000 Hz_40ms_log_rms_mae_db": 1.1088594087769212, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06517014278730064, "300–1000 Hz_40ms_envelope_correlation": 0.996698724187842, "300–1000 Hz_40ms_modulation_mae_db": 0.9292736156040515, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8658561664807174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19593695466766584, "1000–3000 Hz_10ms_envelope_correlation": 0.9710178075549899, "1000–3000 Hz_10ms_modulation_mae_db": 0.6232646896072702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0930348782572157, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08793751569555597, "1000–3000 Hz_40ms_envelope_correlation": 0.9939573302906046, "1000–3000 Hz_40ms_modulation_mae_db": 0.32613875854114754, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3442326406218723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18981227438185896, "3000–6000 Hz_10ms_envelope_correlation": 0.975644757353274, "3000–6000 Hz_10ms_modulation_mae_db": 0.444185291809675, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4272796825559915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10498697914877064, "3000–6000 Hz_40ms_envelope_correlation": 0.992193514813582, "3000–6000 Hz_40ms_modulation_mae_db": 0.16419361592540052, "mr_stft_db": 5.976716982033861, "band_mae_db": 0.12863689069776818, "rms_gain_error_db": -0.12832185820521552, "output_peak": 0.7151296937583158}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8323953459881666, "全帯域_10ms_rms_envelope_nrmse": 0.06297917476030528, "全帯域_10ms_envelope_correlation": 0.9948624596840908, "全帯域_10ms_modulation_mae_db": 0.46337370502832226, "全帯域_40ms_log_rms_mae_db": 0.44556144742152826, "全帯域_40ms_rms_envelope_nrmse": 0.03505653844472722, "全帯域_40ms_envelope_correlation": 0.9983753161591674, "全帯域_40ms_modulation_mae_db": 0.4877108578302436, "300–1000 Hz_10ms_log_rms_mae_db": 1.7700625217642112, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0918426539079026, "300–1000 Hz_10ms_envelope_correlation": 0.9932612037456432, "300–1000 Hz_10ms_modulation_mae_db": 0.3979306780736722, "300–1000 Hz_40ms_log_rms_mae_db": 1.1018363523079138, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053894354536630495, "300–1000 Hz_40ms_envelope_correlation": 0.9977630641560543, "300–1000 Hz_40ms_modulation_mae_db": 0.6194245458324, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8289213832924898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1510934551486692, "1000–3000 Hz_10ms_envelope_correlation": 0.9828612844281707, "1000–3000 Hz_10ms_modulation_mae_db": 0.501715577892296, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0972119199283576, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07646826235873123, "1000–3000 Hz_40ms_envelope_correlation": 0.9954545088634232, "1000–3000 Hz_40ms_modulation_mae_db": 0.7018863535398768, "3000–6000 Hz_10ms_log_rms_mae_db": 2.404316175049094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19971553723249888, "3000–6000 Hz_10ms_envelope_correlation": 0.9729319184563746, "3000–6000 Hz_10ms_modulation_mae_db": 0.3680858427813469, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5478870581964352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12368985354789572, "3000–6000 Hz_40ms_envelope_correlation": 0.989516482967816, "3000–6000 Hz_40ms_modulation_mae_db": 0.37936277491705866, "mr_stft_db": 5.991767705115116, "band_mae_db": 0.15512612370311596, "rms_gain_error_db": -0.09969544721204088, "output_peak": 0.6207115472976119}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8565199299515883, "全帯域_10ms_rms_envelope_nrmse": 0.07073470680727548, "全帯域_10ms_envelope_correlation": 0.9934632884330095, "全帯域_10ms_modulation_mae_db": 0.32716844238305715, "全帯域_40ms_log_rms_mae_db": 0.4539519766820418, "全帯域_40ms_rms_envelope_nrmse": 0.036832488835383266, "全帯域_40ms_envelope_correlation": 0.9981808443019208, "全帯域_40ms_modulation_mae_db": 0.2974355869513026, "300–1000 Hz_10ms_log_rms_mae_db": 1.805252449510731, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11009104174309903, "300–1000 Hz_10ms_envelope_correlation": 0.9900528654164498, "300–1000 Hz_10ms_modulation_mae_db": 1.267594057034369, "300–1000 Hz_40ms_log_rms_mae_db": 1.1460948144906544, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06386152644821849, "300–1000 Hz_40ms_envelope_correlation": 0.9966472736949166, "300–1000 Hz_40ms_modulation_mae_db": 0.7880688557799364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8903656092717083, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1553829401512338, "1000–3000 Hz_10ms_envelope_correlation": 0.9819339502932116, "1000–3000 Hz_10ms_modulation_mae_db": 0.8132196841700754, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1027108360264333, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0689490114444635, "1000–3000 Hz_40ms_envelope_correlation": 0.9966360001861748, "1000–3000 Hz_40ms_modulation_mae_db": 1.0739963543519966, "3000–6000 Hz_10ms_log_rms_mae_db": 2.346308139496286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19104980875854974, "3000–6000 Hz_10ms_envelope_correlation": 0.9752659752889816, "3000–6000 Hz_10ms_modulation_mae_db": 0.5260965723200278, "3000–6000 Hz_40ms_log_rms_mae_db": 1.474309431165823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10768969611490999, "3000–6000 Hz_40ms_envelope_correlation": 0.9918929257271859, "3000–6000 Hz_40ms_modulation_mae_db": 0.44370596160900827, "mr_stft_db": 5.999850757975367, "band_mae_db": 0.19937038035897628, "rms_gain_error_db": -0.09571115230825178, "output_peak": 0.5895374555797368}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8657628250558932, "全帯域_10ms_rms_envelope_nrmse": 0.06419587627567133, "全帯域_10ms_envelope_correlation": 0.9946221152258387, "全帯域_10ms_modulation_mae_db": 0.10338824355057692, "全帯域_40ms_log_rms_mae_db": 0.40455799996797215, "全帯域_40ms_rms_envelope_nrmse": 0.031175171412654766, "全帯域_40ms_envelope_correlation": 0.9987117840007745, "全帯域_40ms_modulation_mae_db": 0.18133691927523146, "300–1000 Hz_10ms_log_rms_mae_db": 1.854438098457301, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11029010967726931, "300–1000 Hz_10ms_envelope_correlation": 0.9899674336245842, "300–1000 Hz_10ms_modulation_mae_db": 0.42313939616474044, "300–1000 Hz_40ms_log_rms_mae_db": 1.1752318235218127, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05739926287212194, "300–1000 Hz_40ms_envelope_correlation": 0.9972392924657187, "300–1000 Hz_40ms_modulation_mae_db": 0.5534848943631165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8238812996785452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14668955396982974, "1000–3000 Hz_10ms_envelope_correlation": 0.9844003375576726, "1000–3000 Hz_10ms_modulation_mae_db": 0.7042822244888488, "1000–3000 Hz_40ms_log_rms_mae_db": 1.098593553722802, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07668229710945758, "1000–3000 Hz_40ms_envelope_correlation": 0.9963720414511933, "1000–3000 Hz_40ms_modulation_mae_db": 0.7235331558731608, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3520975124576307, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17501688691411438, "3000–6000 Hz_10ms_envelope_correlation": 0.9795601029561344, "3000–6000 Hz_10ms_modulation_mae_db": 0.5922305187437192, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4327377638401846, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09401891306806677, "3000–6000 Hz_40ms_envelope_correlation": 0.9939053964175775, "3000–6000 Hz_40ms_modulation_mae_db": 0.5722045862683188, "mr_stft_db": 5.9850666681017834, "band_mae_db": 0.16228591662914169, "rms_gain_error_db": -0.08823609240956581, "output_peak": 0.6456741539367485}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8372844022322236, "全帯域_10ms_rms_envelope_nrmse": 0.0645326884501305, "全帯域_10ms_envelope_correlation": 0.99464718172645, "全帯域_10ms_modulation_mae_db": 0.3416720887146784, "全帯域_40ms_log_rms_mae_db": 0.439559145468005, "全帯域_40ms_rms_envelope_nrmse": 0.03623193518461352, "全帯域_40ms_envelope_correlation": 0.9983151077119948, "全帯域_40ms_modulation_mae_db": 0.3589208597213995, "300–1000 Hz_10ms_log_rms_mae_db": 1.8252779090302276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10452726894426635, "300–1000 Hz_10ms_envelope_correlation": 0.9910636065321805, "300–1000 Hz_10ms_modulation_mae_db": 0.29478500241262334, "300–1000 Hz_40ms_log_rms_mae_db": 1.1615557367056681, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06451669019437152, "300–1000 Hz_40ms_envelope_correlation": 0.9965360186360858, "300–1000 Hz_40ms_modulation_mae_db": 0.28406697456152025, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8870325455216939, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1615238853660583, "1000–3000 Hz_10ms_envelope_correlation": 0.9803956934906103, "1000–3000 Hz_10ms_modulation_mae_db": 0.6304257549793667, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1162622746809603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09124969318324885, "1000–3000 Hz_40ms_envelope_correlation": 0.9934860037054534, "1000–3000 Hz_40ms_modulation_mae_db": 0.38629301345669537, "3000–6000 Hz_10ms_log_rms_mae_db": 2.348896360882823, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20428461091155628, "3000–6000 Hz_10ms_envelope_correlation": 0.9742933869286449, "3000–6000 Hz_10ms_modulation_mae_db": 0.49454958994061055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.453219056858867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12100464151002309, "3000–6000 Hz_40ms_envelope_correlation": 0.9914069633506796, "3000–6000 Hz_40ms_modulation_mae_db": 0.3535134100697293, "mr_stft_db": 6.003605268771277, "band_mae_db": 0.20372017299614953, "rms_gain_error_db": -0.11529752423041496, "output_peak": 0.5615675773484008}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8643219645682059, "全帯域_10ms_rms_envelope_nrmse": 0.06692339875531828, "全帯域_10ms_envelope_correlation": 0.9942133717484608, "全帯域_10ms_modulation_mae_db": 0.4360727803031194, "全帯域_40ms_log_rms_mae_db": 0.47967878568086886, "全帯域_40ms_rms_envelope_nrmse": 0.0357684019227772, "全帯域_40ms_envelope_correlation": 0.9983474823586443, "全帯域_40ms_modulation_mae_db": 0.42134298380599083, "300–1000 Hz_10ms_log_rms_mae_db": 1.8107278507297808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09904862778817683, "300–1000 Hz_10ms_envelope_correlation": 0.9920109681666114, "300–1000 Hz_10ms_modulation_mae_db": 0.4321758736832986, "300–1000 Hz_40ms_log_rms_mae_db": 1.167494531328345, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05670081676505802, "300–1000 Hz_40ms_envelope_correlation": 0.9973694756529727, "300–1000 Hz_40ms_modulation_mae_db": 0.44122473527257905, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8906959864494006, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15998183411446204, "1000–3000 Hz_10ms_envelope_correlation": 0.9808012483847204, "1000–3000 Hz_10ms_modulation_mae_db": 0.9286069645982078, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0933010974679302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08172653973222295, "1000–3000 Hz_40ms_envelope_correlation": 0.9950528232959951, "1000–3000 Hz_40ms_modulation_mae_db": 1.1686043971379438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.41174837260641, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2077292584042095, "3000–6000 Hz_10ms_envelope_correlation": 0.9707206744579309, "3000–6000 Hz_10ms_modulation_mae_db": 0.9200934901801365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5162672451045263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12476592313068584, "3000–6000 Hz_40ms_envelope_correlation": 0.9894555195738435, "3000–6000 Hz_40ms_modulation_mae_db": 0.654729192963589, "mr_stft_db": 5.989510071751994, "band_mae_db": 0.14510958095755636, "rms_gain_error_db": -0.11186233847124472, "output_peak": 0.6238208144287767}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8551041960190366, "全帯域_10ms_rms_envelope_nrmse": 0.06518323108171115, "全帯域_10ms_envelope_correlation": 0.9944496015938635, "全帯域_10ms_modulation_mae_db": 0.49521609853278115, "全帯域_40ms_log_rms_mae_db": 0.43583661430730336, "全帯域_40ms_rms_envelope_nrmse": 0.03220900198486937, "全帯域_40ms_envelope_correlation": 0.9986119598150334, "全帯域_40ms_modulation_mae_db": 0.5888277434424762, "300–1000 Hz_10ms_log_rms_mae_db": 1.8290099808603488, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09255633248003521, "300–1000 Hz_10ms_envelope_correlation": 0.9931111593844372, "300–1000 Hz_10ms_modulation_mae_db": 0.4365028192501197, "300–1000 Hz_40ms_log_rms_mae_db": 1.1338994823317403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052811786627095895, "300–1000 Hz_40ms_envelope_correlation": 0.9978095133712067, "300–1000 Hz_40ms_modulation_mae_db": 0.6583135872178533, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8298985836135793, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17405711183587705, "1000–3000 Hz_10ms_envelope_correlation": 0.9774522330602905, "1000–3000 Hz_10ms_modulation_mae_db": 0.35940097347276456, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0790200044235527, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09272833806210354, "1000–3000 Hz_40ms_envelope_correlation": 0.9940370668037483, "1000–3000 Hz_40ms_modulation_mae_db": 0.5625312253451275, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3541481895502288, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18966100090264004, "3000–6000 Hz_10ms_envelope_correlation": 0.9760048587663157, "3000–6000 Hz_10ms_modulation_mae_db": 0.7509645069107674, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4482521086498523, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09996493943224294, "3000–6000 Hz_40ms_envelope_correlation": 0.9930732644070183, "3000–6000 Hz_40ms_modulation_mae_db": 0.38552682609794964, "mr_stft_db": 6.005482673930691, "band_mae_db": 0.16049098112114743, "rms_gain_error_db": -0.08700239723508028, "output_peak": 0.6632929968356494}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.107670250496585, "spectral_convergence": 0.23561698905195202, "spectral_convergence_status": "ok", "amplitude_floor": 1.6781522163747868e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.954914601152896, "spectral_convergence": 0.2353842357850911, "spectral_convergence_status": "ok", "amplitude_floor": 1.6368745556803986e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.927787504467651, "spectral_convergence": 0.2369726305249138, "spectral_convergence_status": "ok", "amplitude_floor": 1.5514964797831996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006090282805006308, "output_power": 0.0006339834550419832, "delta_db": 0.17440464561002889, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005054165098664251, "output_power": 0.0049436094297401045, "delta_db": -0.0960527245395132, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0022275244303280044, "output_power": 0.002106886358319098, "delta_db": -0.24181364678318912, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006076310340369887, "output_power": 0.0005710470813885418, "delta_db": -0.2696803063027307, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001299593081283461, "output_power": 0.00013771490253587174, "delta_db": 0.25173548485774805, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003180661511978585, "output_power": 0.0003120283221025229, "delta_db": -0.08323437939531303, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18615353124808712, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11864841331146361, "pitch": null, "output_rms_dbfs": -20.602177768859317, "output_peak": 0.634947263674331, "reference_rms_dbfs": -20.48352935554785, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.996790785372377}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002515751859318019, "log_rms_mae_db": 0.8394496727449801, "rms_envelope_nrmse": 0.0666522500068029, "envelope_correlation": 0.9942844219188246, "rms_gain_error_db": -0.11864327296198775, "reference_mean_power": 0.008953808913712473, "output_mean_power": 0.008712514109455788}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.40642035802602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010131590544257714, "output_power": 0.000986456834685855, "reference_active": true, "delta_db": -0.11599543822813724}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021577628078402175, "output_power": 0.0002023923436557088, "reference_active": true, "delta_db": -0.27809623675508677}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.554717092269403e-05, "output_power": 4.042953755423345e-05, "reference_active": true, "delta_db": -0.517626329923345}], "active_band_count": 3, "active_band_mae_db": 0.303906001635523}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002371340103862925, "log_rms_mae_db": 0.4329710988823158, "rms_envelope_nrmse": 0.03421652518166452, "envelope_correlation": 0.9985352134191572, "rms_gain_error_db": -0.1188431650893221, "reference_mean_power": 0.008997361160684636, "output_mean_power": 0.008754489722721304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.083478820328128e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000878683378715502, "output_power": 0.0008548746777597004, "reference_active": true, "delta_db": -0.11929958274111976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.731034174913729e-05, "output_power": 9.091277258860645e-05, "reference_active": true, "delta_db": -0.2953409503059468}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4907380289326384e-06, "output_power": 2.1880917965333933e-06, "reference_active": true, "delta_db": -0.56262513770249}], "active_band_count": 3, "active_band_mae_db": 0.32575522358318554}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019737503490519328, "log_rms_mae_db": 1.8205218267065242, "rms_envelope_nrmse": 0.10473019094197178, "envelope_correlation": 0.990956912062306, "rms_gain_error_db": -0.10169345175174635, "reference_mean_power": 0.0030829546180243575, "output_mean_power": 0.003011603454686728}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6914511369831456e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008052884179451131, "output_power": 0.0007586304776227359, "reference_active": true, "delta_db": -0.2592116660621479}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019776686451848026, "output_power": 0.00015887148189814307, "reference_active": true, "delta_db": -0.951075815889967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9868777263914452e-05, "output_power": 1.8932584648417282e-05, "reference_active": true, "delta_db": -0.2096123394698106}], "active_band_count": 3, "active_band_mae_db": 0.4732999404739752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017438355068165795, "log_rms_mae_db": 1.2051742253104867, "rms_envelope_nrmse": 0.057680045577639165, "envelope_correlation": 0.997166094570302, "rms_gain_error_db": -0.10219219984884335, "reference_mean_power": 0.0030981947388459344, "output_mean_power": 0.003026143316541551}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5303812378070619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007441051144552582, "output_power": 0.0006981993620041383, "reference_active": true, "delta_db": -0.2765484195097422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.887152997212749e-05, "output_power": 7.893775752656003e-05, "reference_active": true, "delta_db": -0.9778646988647262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5498600424852708e-06, "output_power": 1.2695238932479749e-06, "reference_active": true, "delta_db": -0.8665160271999504}], "active_band_count": 3, "active_band_mae_db": 0.7069763818581395}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011259893106980884, "log_rms_mae_db": 1.8761517807068484, "rms_envelope_nrmse": 0.16588003363826942, "envelope_correlation": 0.9794447873673179, "rms_gain_error_db": -0.22963649840382072, "reference_mean_power": 0.0006356182176644616, "output_mean_power": 0.00060288251199741}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.9581273226652876e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002027625859711201, "output_power": 0.00019174723137926948, "reference_active": true, "delta_db": -0.24258719515779684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.693176540437241e-05, "output_power": 2.16932616471599e-05, "reference_active": true, "delta_db": -0.9393996838013118}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.128022173574573e-06, "output_power": 7.555852830604036e-06, "reference_active": true, "delta_db": 0.2531444832841467}], "active_band_count": 3, "active_band_mae_db": 0.4783771207477518}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001021231893168278, "log_rms_mae_db": 1.1042298850816175, "rms_envelope_nrmse": 0.08935307392700688, "envelope_correlation": 0.9942337805918405, "rms_gain_error_db": -0.22975740562091646, "reference_mean_power": 0.0006388062286465308, "output_mean_power": 0.0006058894654089088}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.69069653778161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018976599367877858, "output_power": 0.00018098751653470687, "reference_active": true, "delta_db": -0.2056976811158543}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3827689740852905e-05, "output_power": 1.1168228508481841e-05, "reference_active": true, "delta_db": -0.9276533519710227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.22079217642337e-07, "output_power": 3.090874359231505e-07, "reference_active": true, "delta_db": -1.3531261705592235}], "active_band_count": 3, "active_band_mae_db": 0.8288257345487002}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.113996831641667e-05, "log_rms_mae_db": 2.3673281581315178, "rms_envelope_nrmse": 0.1957237625342072, "envelope_correlation": 0.9748774657398978, "rms_gain_error_db": 0.18613305201894909, "reference_mean_power": 0.0001805297653048898, "output_mean_power": 0.00018843523896737956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6198199988921157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.959931264840548e-05, "output_power": 0.00010633568071171992, "reference_active": true, "delta_db": 0.2842267427366502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361167220741e-06, "output_power": 9.040910913023442e-06, "reference_active": true, "delta_db": -0.15117813386618714}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4041286689847167e-06, "output_power": 2.340822256194867e-06, "reference_active": true, "delta_db": -0.11589269387569798}], "active_band_count": 3, "active_band_mae_db": 0.18376585682617844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.456026972891915e-05, "log_rms_mae_db": 1.4891912622531824, "rms_envelope_nrmse": 0.11388804921519198, "envelope_correlation": 0.9912906874369181, "rms_gain_error_db": 0.1860754667037136, "reference_mean_power": 0.00018143628465930542, "output_mean_power": 0.00018937894409370523}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.558011664744216e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.473019444861243e-05, "output_power": 0.00010132655163438689, "reference_active": true, "delta_db": 0.2923483414185383}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179317052925175e-06, "output_power": 3.9729257176805305e-06, "reference_active": true, "delta_db": -0.21994873935335169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8116784694632823e-07, "output_power": 1.899738615356806e-07, "reference_active": true, "delta_db": 0.20612727504078104}], "active_band_count": 3, "active_band_mae_db": 0.23947478527089036}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8628644254242537, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2c6ee9dc718258687661439f9d21f7692e9b563290559932b851ebb3405a2752", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.979788417767123, "standard_deviation": 0.004126736474072934, "minimum": 0.9710178075549899, "maximum": 0.9844003375576726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.861600419376873, "standard_deviation": 0.029389727297027258, "minimum": 1.8238812996785452, "maximum": 1.8906959864494006, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6299116237445728, "standard_deviation": 0.18553749242410542, "minimum": 0.35940097347276456, "maximum": 0.9286069645982078, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16381822111150818, "standard_deviation": 0.015524845123932731, "minimum": 0.14668955396982974, "maximum": 0.19593695466766584, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949036943985542, "standard_deviation": 0.001170338304826145, "minimum": 0.9934860037054534, "maximum": 0.9966360001861748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0980455561986087, "standard_deviation": 0.010706761848829006, "minimum": 1.0790200044235527, "maximum": 1.1162622746809603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.721476124099331, "standard_deviation": 0.3000021346943929, "minimum": 0.32613875854114754, "maximum": 1.1686043971379438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08313684143909882, "standard_deviation": 0.008526578202199068, "minimum": 0.0689490114444635, "maximum": 0.09272833806210354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9749123924934442, "standard_deviation": 0.0025448334060460486, "minimum": 0.9707206744579309, "maximum": 0.9795601029561344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3661344435994827, "standard_deviation": 0.02686082813356514, "minimum": 2.3442326406218723, "maximum": 2.41174837260641, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5349964586890577, "standard_deviation": 0.22659011885304486, "minimum": 0.18376585682617844, "maximum": 0.9200934901801365, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19412414250495438, "standard_deviation": 0.010248396111323702, "minimum": 0.17501688691411438, "maximum": 0.2077292584042095, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9915918443368277, "standard_deviation": 0.0015581890029889847, "minimum": 0.9894555195738435, "maximum": 0.9939053964175775, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4736429510781077, "standard_deviation": 0.04218298472890153, "minimum": 1.4272796825559915, "maximum": 1.5478870581964352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.39908889414024307, "standard_deviation": 0.16063015090050065, "minimum": 0.16419361592540052, "maximum": 0.654729192963589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11125112439597337, "standard_deviation": 0.01143744852738031, "minimum": 0.09401891306806677, "maximum": 0.12476592313068584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915342783695227, "standard_deviation": 0.0012535810531509523, "minimum": 0.9899674336245842, "maximum": 0.9932612037456432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8068667303076609, "standard_deviation": 0.03619364189017481, "minimum": 1.739643205402163, "maximum": 1.854438098457301, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5435375757403409, "standard_deviation": 0.3063420120186746, "minimum": 0.29478500241262334, "maximum": 1.267594057034369, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10178664567470855, "standard_deviation": 0.007064618313363216, "minimum": 0.0918426539079026, "maximum": 0.11029010967726931, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971536820918874, "standard_deviation": 0.0004928714027301837, "minimum": 0.9965360186360858, "maximum": 0.9978095133712067, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1500182968466928, "standard_deviation": 0.034655845526509126, "minimum": 1.1018363523079138, "maximum": 1.2051742253104867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6226041988111995, "standard_deviation": 0.20075289004538502, "minimum": 0.28406697456152025, "maximum": 0.9292736156040515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05900432822605452, "standard_deviation": 0.0048710706534818545, "minimum": 0.052811786627095895, "maximum": 0.06517014278730064, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16761169721399283, "standard_deviation": 0.02650222214177972, "minimum": 0.12863689069776818, "maximum": 0.20372017299614953, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.993598864131558, "standard_deviation": 0.0097723756851013, "minimum": 5.976716982033861, "maximum": 6.005482673930691, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6318351878574464, "standard_deviation": 0.046357879990136804, "minimum": 0.5615675773484008, "maximum": 0.7151296937583158, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10559690292290969, "standard_deviation": 0.015117784609736985, "minimum": -0.12832185820521552, "maximum": -0.08700239723508028, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944864755642691, "standard_deviation": 0.0005469197357177007, "minimum": 0.9934632884330095, "maximum": 0.9953493641836144, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8437683879100498, "standard_deviation": 0.021921954157438653, "minimum": 0.7993087667203043, "maximum": 0.8657628250558932, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3509032155858357, "standard_deviation": 0.122509561090359, "minimum": 0.10338824355057692, "maximum": 0.49521609853278115, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06522855609314018, "standard_deviation": 0.003000064569169483, "minimum": 0.060627122607906675, "maximum": 0.07073470680727548, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984789817580233, "standard_deviation": 0.0002051260467852299, "minimum": 0.9981808443019208, "maximum": 0.9987541462974936, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.43726987409858253, "standard_deviation": 0.024543211265643978, "minimum": 0.40455799996797215, "maximum": 0.47967878568086886, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37786590853669233, "standard_deviation": 0.12357349328245869, "minimum": 0.18133691927523146, "maximum": 0.5888277434424762, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.034243329104837336, "standard_deviation": 0.002083635704478324, "minimum": 0.031175171412654766, "maximum": 0.036832488835383266, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0041", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.4048993587493896, "predicted_mos": 2.4048993587493896, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3cea4f866b9047a9f77568852aa0480aba144d6f913d5dc082ed91a06770c332", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8906696997582912, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.857517269874557, "全帯域_10ms_rms_envelope_nrmse": 0.06662021740880876, "全帯域_10ms_envelope_correlation": 0.9942323802664286, "全帯域_10ms_modulation_mae_db": 0.4124860765701246, "全帯域_40ms_log_rms_mae_db": 0.4740840480990695, "全帯域_40ms_rms_envelope_nrmse": 0.03735914514978987, "全帯域_40ms_envelope_correlation": 0.9981222804801059, "全帯域_40ms_modulation_mae_db": 0.4597794057872934, "300–1000 Hz_10ms_log_rms_mae_db": 1.4860575773176234, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1080787650072381, "300–1000 Hz_10ms_envelope_correlation": 0.9902661029092903, "300–1000 Hz_10ms_modulation_mae_db": 0.6017329024589823, "300–1000 Hz_40ms_log_rms_mae_db": 0.9101377485661529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06356102996403479, "300–1000 Hz_40ms_envelope_correlation": 0.996430617471258, "300–1000 Hz_40ms_modulation_mae_db": 0.3679886584225744, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3212815857442417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.128522425499727, "1000–3000 Hz_10ms_envelope_correlation": 0.9877652919723288, "1000–3000 Hz_10ms_modulation_mae_db": 0.19485672701440357, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7643200463535319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06717889344189872, "1000–3000 Hz_40ms_envelope_correlation": 0.996782121759698, "1000–3000 Hz_40ms_modulation_mae_db": 0.2166062538705608, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4332313826230614, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17369720713568068, "3000–6000 Hz_10ms_envelope_correlation": 0.9794287471507026, "3000–6000 Hz_10ms_modulation_mae_db": 1.2144607207549052, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9212792354477856, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10539566332167909, "3000–6000 Hz_40ms_envelope_correlation": 0.9919731290631572, "3000–6000 Hz_40ms_modulation_mae_db": 0.48066912591724487, "mr_stft_db": 5.53963928126053, "band_mae_db": 0.13989956946546775, "rms_gain_error_db": -0.09401475402372156, "output_peak": 0.7173018088852271}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9080813431200723, "全帯域_10ms_rms_envelope_nrmse": 0.07316067538639343, "全帯域_10ms_envelope_correlation": 0.9929621041578901, "全帯域_10ms_modulation_mae_db": 0.27400550248192396, "全帯域_40ms_log_rms_mae_db": 0.517974920864872, "全帯域_40ms_rms_envelope_nrmse": 0.04056528044526774, "全帯域_40ms_envelope_correlation": 0.9977229334895776, "全帯域_40ms_modulation_mae_db": 0.19136800124937806, "300–1000 Hz_10ms_log_rms_mae_db": 1.4944879584624486, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11251189055548341, "300–1000 Hz_10ms_envelope_correlation": 0.9894648747351013, "300–1000 Hz_10ms_modulation_mae_db": 0.8743336119250973, "300–1000 Hz_40ms_log_rms_mae_db": 0.8966399924901851, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06845710042447015, "300–1000 Hz_40ms_envelope_correlation": 0.9958613218947535, "300–1000 Hz_40ms_modulation_mae_db": 0.6269950413519437, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3582248068748377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1357250767324043, "1000–3000 Hz_10ms_envelope_correlation": 0.986936267180171, "1000–3000 Hz_10ms_modulation_mae_db": 0.7908685597209019, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7964947616884669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0735003365882918, "1000–3000 Hz_40ms_envelope_correlation": 0.99693618994079, "1000–3000 Hz_40ms_modulation_mae_db": 1.3996935086558995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4544247565992496, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15714736496322762, "3000–6000 Hz_10ms_envelope_correlation": 0.9831655021387478, "3000–6000 Hz_10ms_modulation_mae_db": 0.3604417288666877, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8976133929567548, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09120310880674967, "3000–6000 Hz_40ms_envelope_correlation": 0.9939776599052845, "3000–6000 Hz_40ms_modulation_mae_db": 0.16786602896425926, "mr_stft_db": 5.542056392116564, "band_mae_db": 0.19227554672015654, "rms_gain_error_db": -0.08038379601003938, "output_peak": 0.5769654807722765}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8215266487335546, "全帯域_10ms_rms_envelope_nrmse": 0.06623531524384346, "全帯域_10ms_envelope_correlation": 0.9945786899046548, "全帯域_10ms_modulation_mae_db": 0.4680415272641419, "全帯域_40ms_log_rms_mae_db": 0.42906260214855024, "全帯域_40ms_rms_envelope_nrmse": 0.03827226051257523, "全帯域_40ms_envelope_correlation": 0.9983687994660768, "全帯域_40ms_modulation_mae_db": 0.5000974432905715, "300–1000 Hz_10ms_log_rms_mae_db": 1.5388229504428461, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11066959674634771, "300–1000 Hz_10ms_envelope_correlation": 0.9901087252617472, "300–1000 Hz_10ms_modulation_mae_db": 0.5491340334323408, "300–1000 Hz_40ms_log_rms_mae_db": 0.9200634663161937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06627039009545033, "300–1000 Hz_40ms_envelope_correlation": 0.9965532120984582, "300–1000 Hz_40ms_modulation_mae_db": 0.3960839850069248, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3099979406691256, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12370668163166386, "1000–3000 Hz_10ms_envelope_correlation": 0.9888100249793184, "1000–3000 Hz_10ms_modulation_mae_db": 0.4389219021982495, "1000–3000 Hz_40ms_log_rms_mae_db": 0.755165847116636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06714876634531228, "1000–3000 Hz_40ms_envelope_correlation": 0.9969563540933878, "1000–3000 Hz_40ms_modulation_mae_db": 0.4905311879733652, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4472801094790737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1479830514832509, "3000–6000 Hz_10ms_envelope_correlation": 0.985313708944741, "3000–6000 Hz_10ms_modulation_mae_db": 0.6285109753304629, "3000–6000 Hz_40ms_log_rms_mae_db": 0.896522668211547, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09198836052833757, "3000–6000 Hz_40ms_envelope_correlation": 0.9944475212202257, "3000–6000 Hz_40ms_modulation_mae_db": 0.46095608243408764, "mr_stft_db": 5.521235670483463, "band_mae_db": 0.21223267671923016, "rms_gain_error_db": -0.1679454394393699, "output_peak": 0.6019655853367558}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9133318506878664, "全帯域_10ms_rms_envelope_nrmse": 0.07136829416712304, "全帯域_10ms_envelope_correlation": 0.9934625400899407, "全帯域_10ms_modulation_mae_db": 0.30370702199990146, "全帯域_40ms_log_rms_mae_db": 0.49335153478718785, "全帯域_40ms_rms_envelope_nrmse": 0.03709260859535265, "全帯域_40ms_envelope_correlation": 0.9983085465760183, "全帯域_40ms_modulation_mae_db": 0.2572304710868502, "300–1000 Hz_10ms_log_rms_mae_db": 1.5316245115080511, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10814675663462255, "300–1000 Hz_10ms_envelope_correlation": 0.9905508924110427, "300–1000 Hz_10ms_modulation_mae_db": 0.5344807552089845, "300–1000 Hz_40ms_log_rms_mae_db": 0.8977934217129826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0645745106942214, "300–1000 Hz_40ms_envelope_correlation": 0.9967045835896338, "300–1000 Hz_40ms_modulation_mae_db": 0.2945682996811616, "1000–3000 Hz_10ms_log_rms_mae_db": 1.337718574466504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13460433629298568, "1000–3000 Hz_10ms_envelope_correlation": 0.9873295697721944, "1000–3000 Hz_10ms_modulation_mae_db": 0.4934006688801153, "1000–3000 Hz_40ms_log_rms_mae_db": 0.758163837153557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07475599583700264, "1000–3000 Hz_40ms_envelope_correlation": 0.9969955996201019, "1000–3000 Hz_40ms_modulation_mae_db": 0.48772733150280395, "3000–6000 Hz_10ms_log_rms_mae_db": 1.455089476513538, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16639359342474258, "3000–6000 Hz_10ms_envelope_correlation": 0.9813358121763092, "3000–6000 Hz_10ms_modulation_mae_db": 0.433855829155623, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9130072526903575, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08602286611307584, "3000–6000 Hz_40ms_envelope_correlation": 0.9947156479876528, "3000–6000 Hz_40ms_modulation_mae_db": 0.1444359769951224, "mr_stft_db": 5.528861460881668, "band_mae_db": 0.19946209990136451, "rms_gain_error_db": -0.13075082781273412, "output_peak": 0.5702027249171582}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8878954709200314, "全帯域_10ms_rms_envelope_nrmse": 0.0674689492727612, "全帯域_10ms_envelope_correlation": 0.9941032855839208, "全帯域_10ms_modulation_mae_db": 0.117875874593459, "全帯域_40ms_log_rms_mae_db": 0.4645300579897696, "全帯域_40ms_rms_envelope_nrmse": 0.03339636312374603, "全帯域_40ms_envelope_correlation": 0.9985828299257199, "全帯域_40ms_modulation_mae_db": 0.17207474638056178, "300–1000 Hz_10ms_log_rms_mae_db": 1.499450993305668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.103814052339057, "300–1000 Hz_10ms_envelope_correlation": 0.991089993079627, "300–1000 Hz_10ms_modulation_mae_db": 0.4070155821449557, "300–1000 Hz_40ms_log_rms_mae_db": 0.8710793046954053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05797827191053387, "300–1000 Hz_40ms_envelope_correlation": 0.9971340528897755, "300–1000 Hz_40ms_modulation_mae_db": 0.2503951180590301, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4137180530039475, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14975283407947454, "1000–3000 Hz_10ms_envelope_correlation": 0.9844324315972418, "1000–3000 Hz_10ms_modulation_mae_db": 0.8113188420886739, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8678805006920117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08692432387429007, "1000–3000 Hz_40ms_envelope_correlation": 0.9959766171623501, "1000–3000 Hz_40ms_modulation_mae_db": 0.8940932653448632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4268787441216024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16560782764541299, "3000–6000 Hz_10ms_envelope_correlation": 0.9814672012148501, "3000–6000 Hz_10ms_modulation_mae_db": 0.46338197735960335, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8831332152501555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09054274230193667, "3000–6000 Hz_40ms_envelope_correlation": 0.9947224979130904, "3000–6000 Hz_40ms_modulation_mae_db": 0.366059779121303, "mr_stft_db": 5.5336211977581415, "band_mae_db": 0.22118287391104716, "rms_gain_error_db": -0.10892646823574731, "output_peak": 0.6255132566087341}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8591819128852899, "全帯域_10ms_rms_envelope_nrmse": 0.06675385337756291, "全帯域_10ms_envelope_correlation": 0.9941480285131794, "全帯域_10ms_modulation_mae_db": 0.34060494017307136, "全帯域_40ms_log_rms_mae_db": 0.49626142753317753, "全帯域_40ms_rms_envelope_nrmse": 0.03845310357394577, "全帯域_40ms_envelope_correlation": 0.997926574968896, "全帯域_40ms_modulation_mae_db": 0.1860286938877136, "300–1000 Hz_10ms_log_rms_mae_db": 1.4848460171666484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10193881374123004, "300–1000 Hz_10ms_envelope_correlation": 0.9913729387905018, "300–1000 Hz_10ms_modulation_mae_db": 0.5462443511980523, "300–1000 Hz_40ms_log_rms_mae_db": 0.8959790620783065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05942228615280468, "300–1000 Hz_40ms_envelope_correlation": 0.9969068609722511, "300–1000 Hz_40ms_modulation_mae_db": 0.6824458538081236, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3699325925712647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.144269014131382, "1000–3000 Hz_10ms_envelope_correlation": 0.9849052647550752, "1000–3000 Hz_10ms_modulation_mae_db": 0.3772204065818041, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8194005723446712, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0884868258769164, "1000–3000 Hz_40ms_envelope_correlation": 0.9947575795710435, "1000–3000 Hz_40ms_modulation_mae_db": 0.5499598693121729, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4343887166814038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1568058441396091, "3000–6000 Hz_10ms_envelope_correlation": 0.9833144799229718, "3000–6000 Hz_10ms_modulation_mae_db": 0.3295255401408779, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8856899813793169, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09447541706936756, "3000–6000 Hz_40ms_envelope_correlation": 0.9939427061979488, "3000–6000 Hz_40ms_modulation_mae_db": 0.2185101042141715, "mr_stft_db": 5.526324404612191, "band_mae_db": 0.18268154356166455, "rms_gain_error_db": -0.0768937638949686, "output_peak": 0.6105570053277057}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.857362468555824, "全帯域_10ms_rms_envelope_nrmse": 0.06684109792968602, "全帯域_10ms_envelope_correlation": 0.994348727727653, "全帯域_10ms_modulation_mae_db": 0.4839587438880509, "全帯域_40ms_log_rms_mae_db": 0.46241850378014615, "全帯域_40ms_rms_envelope_nrmse": 0.037597770762571765, "全帯域_40ms_envelope_correlation": 0.998297716626564, "全帯域_40ms_modulation_mae_db": 0.5037307514496131, "300–1000 Hz_10ms_log_rms_mae_db": 1.4862915132186876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10124722088769426, "300–1000 Hz_10ms_envelope_correlation": 0.99189677996967, "300–1000 Hz_10ms_modulation_mae_db": 0.4434499038207756, "300–1000 Hz_40ms_log_rms_mae_db": 0.892357172167579, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06098458097984339, "300–1000 Hz_40ms_envelope_correlation": 0.9972759887325999, "300–1000 Hz_40ms_modulation_mae_db": 0.4174147203552083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3643390337938728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14769224305241732, "1000–3000 Hz_10ms_envelope_correlation": 0.9845405848572901, "1000–3000 Hz_10ms_modulation_mae_db": 0.5427000282212752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7961330075774973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08261928721760134, "1000–3000 Hz_40ms_envelope_correlation": 0.996109874297866, "1000–3000 Hz_40ms_modulation_mae_db": 0.7158421284067898, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4473499011934061, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15082789139647487, "3000–6000 Hz_10ms_envelope_correlation": 0.9846799063578556, "3000–6000 Hz_10ms_modulation_mae_db": 1.040995854435989, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9006439064652515, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09256648262239418, "3000–6000 Hz_40ms_envelope_correlation": 0.9939184248573014, "3000–6000 Hz_40ms_modulation_mae_db": 0.9511343467365321, "mr_stft_db": 5.525609853421149, "band_mae_db": 0.16482463474807013, "rms_gain_error_db": -0.14373689163118542, "output_peak": 0.6291236248519615}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8796667117902449, "全帯域_10ms_rms_envelope_nrmse": 0.06612262566073908, "全帯域_10ms_envelope_correlation": 0.9942194881742042, "全帯域_10ms_modulation_mae_db": 0.1782915340828746, "全帯域_40ms_log_rms_mae_db": 0.5276139180077442, "全帯域_40ms_rms_envelope_nrmse": 0.039191278209592915, "全帯域_40ms_envelope_correlation": 0.9977870106945913, "全帯域_40ms_modulation_mae_db": 0.22225089180794963, "300–1000 Hz_10ms_log_rms_mae_db": 1.5031507295958098, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10460865973738401, "300–1000 Hz_10ms_envelope_correlation": 0.9908871587532471, "300–1000 Hz_10ms_modulation_mae_db": 0.4597427439179558, "300–1000 Hz_40ms_log_rms_mae_db": 0.9455526319261129, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06130983386557422, "300–1000 Hz_40ms_envelope_correlation": 0.9966605924797607, "300–1000 Hz_40ms_modulation_mae_db": 0.21438081642826287, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3577469671306155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.136453731637697, "1000–3000 Hz_10ms_envelope_correlation": 0.9862083708139413, "1000–3000 Hz_10ms_modulation_mae_db": 0.5930823992978637, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7618832856557103, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06921691605915217, "1000–3000 Hz_40ms_envelope_correlation": 0.9966434852043076, "1000–3000 Hz_40ms_modulation_mae_db": 0.8915610277059872, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4371579484384276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1861956601465046, "3000–6000 Hz_10ms_envelope_correlation": 0.9797016196412816, "3000–6000 Hz_10ms_modulation_mae_db": 1.587511972988682, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8774360996291313, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12905387388277315, "3000–6000 Hz_40ms_envelope_correlation": 0.9907275452734887, "3000–6000 Hz_40ms_modulation_mae_db": 1.3803845843868767, "mr_stft_db": 5.5364846994470716, "band_mae_db": 0.15668526409924252, "rms_gain_error_db": -0.05106297672859769, "output_peak": 0.574889523737685}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.4853421190767335, "spectral_convergence": 0.226256463270167, "spectral_convergence_status": "ok", "amplitude_floor": 1.6781522163747868e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3930313010521385, "spectral_convergence": 0.23199519564459897, "spectral_convergence_status": "ok", "amplitude_floor": 1.6368745556803986e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.740544423652717, "spectral_convergence": 0.2394394518417768, "spectral_convergence_status": "ok", "amplitude_floor": 1.5514964797831996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006090282805006308, "output_power": 0.0006325055491163517, "delta_db": 0.16426880270089878, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005054165098664251, "output_power": 0.004956353534381962, "delta_db": -0.08487147588228555, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0022275244303280044, "output_power": 0.002150625426074453, "delta_db": -0.15257699844827122, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006076310340369887, "output_power": 0.0005793445984557777, "delta_db": -0.2070298467821471, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001299593081283461, "output_power": 0.00012678024898862383, "delta_db": -0.10755790236732463, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003180661511978585, "output_power": 0.0003091777373810125, "delta_db": -0.12309239061187922, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13989956946546775, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09401475402372156, "pitch": null, "output_rms_dbfs": -20.577544109571573, "output_peak": 0.7173018088852271, "reference_rms_dbfs": -20.48352935554785, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.53963928126053}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002515751859318019, "log_rms_mae_db": 0.857517269874557, "rms_envelope_nrmse": 0.06662021740880876, "envelope_correlation": 0.9942323802664286, "rms_gain_error_db": -0.09401206385066065, "reference_mean_power": 0.008953808913712473, "output_mean_power": 0.008762067920094128}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.40642035802602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010131590544257714, "output_power": 0.0010084356848662275, "reference_active": true, "delta_db": -0.020294246206786666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021577628078402175, "output_power": 0.00020101362057696928, "reference_active": true, "delta_db": -0.307782171502517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.554717092269403e-05, "output_power": 3.694224001049182e-05, "reference_active": true, "delta_db": -0.90938181200107}], "active_band_count": 3, "active_band_mae_db": 0.4124860765701246}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002371340103862925, "log_rms_mae_db": 0.4740840480990695, "rms_envelope_nrmse": 0.03735914514978987, "envelope_correlation": 0.9981222804801059, "rms_gain_error_db": -0.0941124458853354, "reference_mean_power": 0.008997361160684636, "output_mean_power": 0.008804484011573138}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.083478820328128e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000878683378715502, "output_power": 0.0008749857589559425, "reference_active": true, "delta_db": -0.018314266962113822}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.731034174913729e-05, "output_power": 9.03848783870076e-05, "reference_active": true, "delta_db": -0.3206321976519193}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4907380289326384e-06, "output_power": 1.9601480569432562e-06, "reference_active": true, "delta_db": -1.0403917527478472}], "active_band_count": 3, "active_band_mae_db": 0.4597794057872934}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019737503490519328, "log_rms_mae_db": 1.4860575773176234, "rms_envelope_nrmse": 0.1080787650072381, "envelope_correlation": 0.9902661029092903, "rms_gain_error_db": -0.07574834157508048, "reference_mean_power": 0.0030829546180243575, "output_mean_power": 0.003029648870590913}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6914511369831456e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008052884179451131, "output_power": 0.0007668213328731845, "reference_active": true, "delta_db": -0.2125726650804149}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019776686451848026, "output_power": 0.00016510851677564923, "reference_active": true, "delta_db": -0.7838405209082164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9868777263914452e-05, "output_power": 2.3935896227673418e-05, "reference_active": true, "delta_db": 0.8087855213883155}], "active_band_count": 3, "active_band_mae_db": 0.6017329024589823}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017438355068165795, "log_rms_mae_db": 0.9101377485661529, "rms_envelope_nrmse": 0.06356102996403479, "envelope_correlation": 0.996430617471258, "rms_gain_error_db": -0.075793638762929, "reference_mean_power": 0.0030981947388459344, "output_mean_power": 0.0030445937270260405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5303812378070619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007441051144552582, "output_power": 0.0007072021878234691, "reference_active": true, "delta_db": -0.2209069402816541}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.887152997212749e-05, "output_power": 8.131211229270582e-05, "reference_active": true, "delta_db": -0.8491601149562266}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5498600424852708e-06, "output_power": 1.5378096778281835e-06, "reference_active": true, "delta_db": -0.033898920029842614}], "active_band_count": 3, "active_band_mae_db": 0.3679886584225744}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011259893106980884, "log_rms_mae_db": 1.3212815857442417, "rms_envelope_nrmse": 0.128522425499727, "envelope_correlation": 0.9877652919723288, "rms_gain_error_db": -0.1983632755557878, "reference_mean_power": 0.0006356182176644616, "output_mean_power": 0.0006072394924757576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.9581273226652876e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002027625859711201, "output_power": 0.00019003383893303138, "reference_active": true, "delta_db": -0.2815687968377857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.693176540437241e-05, "output_power": 2.5707618512941954e-05, "reference_active": true, "delta_db": -0.2020297619313316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.128022173574573e-06, "output_power": 6.964210342455044e-06, "reference_active": true, "delta_db": -0.10097162227409337}], "active_band_count": 3, "active_band_mae_db": 0.19485672701440357}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001021231893168278, "log_rms_mae_db": 0.7643200463535319, "rms_envelope_nrmse": 0.06717889344189872, "envelope_correlation": 0.996782121759698, "rms_gain_error_db": -0.19842553833476984, "reference_mean_power": 0.0006388062286465308, "output_mean_power": 0.0006102764176223}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.69069653778161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018976599367877858, "output_power": 0.00018086717968770902, "reference_active": true, "delta_db": -0.20858622345580366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3827689740852905e-05, "output_power": 1.298152297737562e-05, "reference_active": true, "delta_db": -0.27423980102516554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.22079217642337e-07, "output_power": 4.0615770381985104e-07, "reference_active": true, "delta_db": -0.1669927371307131}], "active_band_count": 3, "active_band_mae_db": 0.2166062538705608}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.113996831641667e-05, "log_rms_mae_db": 1.4332313826230614, "rms_envelope_nrmse": 0.17369720713568068, "envelope_correlation": 0.9794287471507026, "rms_gain_error_db": -0.019328752739238002, "reference_mean_power": 0.0001805297653048898, "output_mean_power": 0.00017972808307431027}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6198199988921157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.959931264840548e-05, "output_power": 8.901335269353267e-05, "reference_active": true, "delta_db": -0.4880118222126697}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361167220741e-06, "output_power": 1.1221004695984109e-05, "reference_active": true, "delta_db": 0.787017409034273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4041286689847167e-06, "output_power": 4.147563085702564e-06, "reference_active": true, "delta_db": 2.368352931017773}], "active_band_count": 3, "active_band_mae_db": 1.2144607207549052}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.456026972891915e-05, "log_rms_mae_db": 0.9212792354477856, "rms_envelope_nrmse": 0.10539566332167909, "envelope_correlation": 0.9919731290631572, "rms_gain_error_db": -0.01932957322588013, "reference_mean_power": 0.00018143628465930542, "output_mean_power": 0.00018063054270418694}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.558011664744216e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.473019444861243e-05, "output_power": 8.686307144557148e-05, "reference_active": true, "delta_db": -0.37653246976354776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179317052925175e-06, "output_power": 4.607648064722944e-06, "reference_active": true, "delta_db": 0.4237398097049053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8116784694632823e-07, "output_power": 2.1001708821323445e-07, "reference_active": true, "delta_db": 0.6417350982832815}], "active_band_count": 3, "active_band_mae_db": 0.48066912591724487}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.882534072769823, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "32528cb4f5bcd39110c5d04080fea4b8e18bebc55e16a4dd9495b230cf0ca860", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9863659757409451, "standard_deviation": 0.0016219198988833836, "minimum": 0.9844324315972418, "maximum": 0.9888100249793184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3541199442818013, "standard_deviation": 0.03213925796402878, "minimum": 1.3099979406691256, "maximum": 1.4137180530039475, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5302961917504109, "standard_deviation": 0.20585363138359034, "minimum": 0.19485672701440357, "maximum": 0.8113188420886739, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13759079288221895, "standard_deviation": 0.009128055190017184, "minimum": 0.12370668163166386, "maximum": 0.14975283407947454, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963947277061931, "standard_deviation": 0.0007669809981166137, "minimum": 0.9947575795710435, "maximum": 0.9969955996201019, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7899302323227604, "standard_deviation": 0.03907226259496252, "minimum": 0.755165847116636, "maximum": 0.8678805006920117, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7057518215965553, "standard_deviation": 0.36036559968987053, "minimum": 0.2166062538705608, "maximum": 1.3996935086558995, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07622891815505817, "standard_deviation": 0.008690058006410224, "minimum": 0.06714876634531228, "maximum": 0.0884868258769164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823008721934324, "standard_deviation": 0.0021759575444048112, "minimum": 0.9794287471507026, "maximum": 0.985313708944741, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4419751294562202, "standard_deviation": 0.01048211503747271, "minimum": 1.4268787441216024, "maximum": 1.455089476513538, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7573355748791039, "standard_deviation": 0.4670284783042919, "minimum": 0.3295255401408779, "maximum": 1.587511972988682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1630823050418629, "standard_deviation": 0.012632281848187215, "minimum": 0.1479830514832509, "maximum": 0.1861956601465046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935531415522687, "standard_deviation": 0.001437486528429248, "minimum": 0.9907275452734887, "maximum": 0.9947224979130904, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8969157190037875, "standard_deviation": 0.014940305865269036, "minimum": 0.8774360996291313, "maximum": 0.9212792354477856, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5212520035961996, "standard_deviation": 0.4327320909124555, "minimum": 0.1444359769951224, "maximum": 1.3803845843868767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09765606433078922, "standard_deviation": 0.013843195072034391, "minimum": 0.08602286611307584, "maximum": 0.12905387388277315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907046832387785, "standard_deviation": 0.0007706876748661082, "minimum": 0.9894648747351013, "maximum": 0.99189677996967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5030915313772228, "standard_deviation": 0.02098788338272728, "minimum": 1.4848460171666484, "maximum": 1.5388229504428461, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.552016735513393, "standard_deviation": 0.14542331706321243, "minimum": 0.4070155821449557, "maximum": 0.8743336119250973, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10637696945613213, "standard_deviation": 0.0041022497423174745, "minimum": 0.10124722088769426, "maximum": 0.11251189055548341, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966909037660614, "standard_deviation": 0.0004409993557277348, "minimum": 0.9958613218947535, "maximum": 0.9972759887325999, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9037003499941147, "standard_deviation": 0.02203232753259899, "minimum": 0.8710793046954053, "maximum": 0.9455526319261129, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.40628406163915365, "standard_deviation": 0.16914987976723356, "minimum": 0.21438081642826287, "maximum": 0.6824458538081236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0628197505108666, "standard_deviation": 0.0035445265794095747, "minimum": 0.05797827191053387, "maximum": 0.06845710042447015, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1836555261407804, "standard_deviation": 0.0281491887387823, "minimum": 0.13989956946546775, "maximum": 0.22118287391104716, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.531729119997597, "standard_deviation": 0.007371867548987704, "minimum": 5.521235670483463, "maximum": 5.542056392116564, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6133148763046881, "standard_deviation": 0.04781639780110493, "minimum": 0.5702027249171582, "maximum": 0.7173018088852271, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10671436472204548, "standard_deviation": 0.038809941932289364, "minimum": -0.1679454394393699, "maximum": -0.05106297672859769, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940069055522339, "standard_deviation": 0.0005287395353859468, "minimum": 0.9929621041578901, "maximum": 0.9945786899046548, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8730704595709301, "standard_deviation": 0.030351116577588756, "minimum": 0.8215266487335546, "maximum": 0.9133318506878664, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3223714026316935, "standard_deviation": 0.13150144992431578, "minimum": 0.117875874593459, "maximum": 0.4839587438880509, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06807137855586473, "standard_deviation": 0.002663446044922002, "minimum": 0.06612262566073908, "maximum": 0.07316067538639343, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981395865284437, "standard_deviation": 0.00030381827341787425, "minimum": 0.9977229334895776, "maximum": 0.9985828299257199, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48316212665131464, "standard_deviation": 0.03218235996753443, "minimum": 0.42906260214855024, "maximum": 0.5276139180077442, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31157005061749143, "standard_deviation": 0.14882066511277578, "minimum": 0.17207474638056178, "maximum": 0.5037307514496131, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03774097629660524, "standard_deviation": 0.0020816135529818716, "minimum": 0.03339636312374603, "maximum": 0.04056528044526774, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0041", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2282419204711914, "predicted_mos": 2.2282419204711914, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b4d67f57c4aa254608e758b2ca498ea13c0e4df09f600514fa6f12707ec2c4ec", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.093756599817425, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8740676111321698, "全帯域_10ms_rms_envelope_nrmse": 0.07376187201923592, "全帯域_10ms_envelope_correlation": 0.9928706067032245, "全帯域_10ms_modulation_mae_db": 0.21061268065269956, "全帯域_40ms_log_rms_mae_db": 0.4853076951025381, "全帯域_40ms_rms_envelope_nrmse": 0.04151829943866356, "全帯域_40ms_envelope_correlation": 0.9976276893352103, "全帯域_40ms_modulation_mae_db": 0.3351727963156615, "300–1000 Hz_10ms_log_rms_mae_db": 1.6512966946431509, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12403163732352904, "300–1000 Hz_10ms_envelope_correlation": 0.9871742571509818, "300–1000 Hz_10ms_modulation_mae_db": 0.8670848027126867, "300–1000 Hz_40ms_log_rms_mae_db": 1.0350183544800626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07367461988967866, "300–1000 Hz_40ms_envelope_correlation": 0.9952398776578475, "300–1000 Hz_40ms_modulation_mae_db": 0.47849401079967907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5225101740463969, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13712859311422157, "1000–3000 Hz_10ms_envelope_correlation": 0.986477475091154, "1000–3000 Hz_10ms_modulation_mae_db": 0.6920651593688475, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9156634040725341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07875355135042326, "1000–3000 Hz_40ms_envelope_correlation": 0.9960828449525717, "1000–3000 Hz_40ms_modulation_mae_db": 0.6485861409963544, "3000–6000 Hz_10ms_log_rms_mae_db": 1.716435296133741, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16139648370546453, "3000–6000 Hz_10ms_envelope_correlation": 0.9830998513500664, "3000–6000 Hz_10ms_modulation_mae_db": 0.14239432290526655, "3000–6000 Hz_40ms_log_rms_mae_db": 1.088554727131276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08977153256211638, "3000–6000 Hz_40ms_envelope_correlation": 0.9947322294163734, "3000–6000 Hz_40ms_modulation_mae_db": 0.33885238588924677, "mr_stft_db": 5.66907586606911, "band_mae_db": 0.1596146069463211, "rms_gain_error_db": -0.09505419606591264, "output_peak": 0.6557666248446633}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.875745816695895, "全帯域_10ms_rms_envelope_nrmse": 0.06642920207405131, "全帯域_10ms_envelope_correlation": 0.9943385179323485, "全帯域_10ms_modulation_mae_db": 0.16647633935228695, "全帯域_40ms_log_rms_mae_db": 0.46595571477592224, "全帯域_40ms_rms_envelope_nrmse": 0.03661044273320062, "全帯域_40ms_envelope_correlation": 0.9983042329647823, "全帯域_40ms_modulation_mae_db": 0.26424052754533656, "300–1000 Hz_10ms_log_rms_mae_db": 1.5982851432908667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10426662041257753, "300–1000 Hz_10ms_envelope_correlation": 0.9910905850311938, "300–1000 Hz_10ms_modulation_mae_db": 0.34715933738393373, "300–1000 Hz_40ms_log_rms_mae_db": 0.9767600220969116, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06088952301682021, "300–1000 Hz_40ms_envelope_correlation": 0.9969320777207694, "300–1000 Hz_40ms_modulation_mae_db": 0.4892671724945485, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5279600459018254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14342711457326116, "1000–3000 Hz_10ms_envelope_correlation": 0.9846088753911095, "1000–3000 Hz_10ms_modulation_mae_db": 0.8635578705481022, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9305439811215936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08002375379644845, "1000–3000 Hz_40ms_envelope_correlation": 0.9951905011308996, "1000–3000 Hz_40ms_modulation_mae_db": 0.6381886589106779, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7250530237877655, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17318632520096816, "3000–6000 Hz_10ms_envelope_correlation": 0.9809775042344688, "3000–6000 Hz_10ms_modulation_mae_db": 0.5123668791925438, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0852731945937308, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08921584041898285, "3000–6000 Hz_40ms_envelope_correlation": 0.9951949457209581, "3000–6000 Hz_40ms_modulation_mae_db": 0.4730653028446494, "mr_stft_db": 5.671614624575992, "band_mae_db": 0.12183556706510414, "rms_gain_error_db": -0.1235516248231851, "output_peak": 0.6086717258747768}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8730122797415856, "全帯域_10ms_rms_envelope_nrmse": 0.06874589999205778, "全帯域_10ms_envelope_correlation": 0.9938400051906021, "全帯域_10ms_modulation_mae_db": 0.23063547731390877, "全帯域_40ms_log_rms_mae_db": 0.4656855536533808, "全帯域_40ms_rms_envelope_nrmse": 0.03569331408871034, "全帯域_40ms_envelope_correlation": 0.9983052899546859, "全帯域_40ms_modulation_mae_db": 0.45295969103441847, "300–1000 Hz_10ms_log_rms_mae_db": 1.5844617197935018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10562855241253302, "300–1000 Hz_10ms_envelope_correlation": 0.9908006383868071, "300–1000 Hz_10ms_modulation_mae_db": 0.6466175877836874, "300–1000 Hz_40ms_log_rms_mae_db": 0.9129081812507971, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05332027657153356, "300–1000 Hz_40ms_envelope_correlation": 0.9976514100489027, "300–1000 Hz_40ms_modulation_mae_db": 0.5533812496230844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4911935427900607, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14604797397073124, "1000–3000 Hz_10ms_envelope_correlation": 0.9847527212203943, "1000–3000 Hz_10ms_modulation_mae_db": 0.910777566850275, "1000–3000 Hz_40ms_log_rms_mae_db": 0.913279620016309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08574922527093568, "1000–3000 Hz_40ms_envelope_correlation": 0.9954749749076445, "1000–3000 Hz_40ms_modulation_mae_db": 1.1568326906383053, "3000–6000 Hz_10ms_log_rms_mae_db": 1.72414624582952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16086966552487147, "3000–6000 Hz_10ms_envelope_correlation": 0.9834823825380569, "3000–6000 Hz_10ms_modulation_mae_db": 0.3404030632818227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0473044125498174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08813557322050602, "3000–6000 Hz_40ms_envelope_correlation": 0.9951556351102542, "3000–6000 Hz_40ms_modulation_mae_db": 0.34265574928962966, "mr_stft_db": 5.674125744350377, "band_mae_db": 0.17676582118625894, "rms_gain_error_db": -0.0977770651158129, "output_peak": 0.5348040351937746}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.887656021921718, "全帯域_10ms_rms_envelope_nrmse": 0.06846807485695582, "全帯域_10ms_envelope_correlation": 0.993996637708633, "全帯域_10ms_modulation_mae_db": 0.32681394727266605, "全帯域_40ms_log_rms_mae_db": 0.47698786568338397, "全帯域_40ms_rms_envelope_nrmse": 0.04011503780606506, "全帯域_40ms_envelope_correlation": 0.9979401212354155, "全帯域_40ms_modulation_mae_db": 0.3225350846680685, "300–1000 Hz_10ms_log_rms_mae_db": 1.6382424900423898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10523218685366714, "300–1000 Hz_10ms_envelope_correlation": 0.9909521773125999, "300–1000 Hz_10ms_modulation_mae_db": 0.6726908494600959, "300–1000 Hz_40ms_log_rms_mae_db": 1.0392725935083758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06711462849929226, "300–1000 Hz_40ms_envelope_correlation": 0.9962309330768425, "300–1000 Hz_40ms_modulation_mae_db": 0.6767130221733401, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5260237794113547, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14874632870085616, "1000–3000 Hz_10ms_envelope_correlation": 0.9834914567810239, "1000–3000 Hz_10ms_modulation_mae_db": 0.056533082367502784, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8789521741036337, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0909966242585361, "1000–3000 Hz_40ms_envelope_correlation": 0.993818199727102, "1000–3000 Hz_40ms_modulation_mae_db": 0.1529384246305758, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7955415570658062, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15285774812127037, "3000–6000 Hz_10ms_envelope_correlation": 0.9844019150137328, "3000–6000 Hz_10ms_modulation_mae_db": 0.25785298887118774, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1198900060169839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09048037400058256, "3000–6000 Hz_40ms_envelope_correlation": 0.9942626247676971, "3000–6000 Hz_40ms_modulation_mae_db": 0.32048049287777053, "mr_stft_db": 5.663686316240667, "band_mae_db": 0.1439667546518158, "rms_gain_error_db": -0.1310749169482331, "output_peak": 0.6437192497788555}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8391243985953201, "全帯域_10ms_rms_envelope_nrmse": 0.07450988109129505, "全帯域_10ms_envelope_correlation": 0.9927113544487729, "全帯域_10ms_modulation_mae_db": 0.3219618694799382, "全帯域_40ms_log_rms_mae_db": 0.4628529440357325, "全帯域_40ms_rms_envelope_nrmse": 0.04095468143792141, "全帯域_40ms_envelope_correlation": 0.9976900501618148, "全帯域_40ms_modulation_mae_db": 0.5697915942215178, "300–1000 Hz_10ms_log_rms_mae_db": 1.5707253467498932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11669463185122526, "300–1000 Hz_10ms_envelope_correlation": 0.9886638977760893, "300–1000 Hz_10ms_modulation_mae_db": 0.7892000749563554, "300–1000 Hz_40ms_log_rms_mae_db": 0.9768910985995047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06986445644520443, "300–1000 Hz_40ms_envelope_correlation": 0.9957093755262576, "300–1000 Hz_40ms_modulation_mae_db": 0.31738308674153254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5703413872048586, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1499882834282282, "1000–3000 Hz_10ms_envelope_correlation": 0.9833874043718392, "1000–3000 Hz_10ms_modulation_mae_db": 0.6757314993305662, "1000–3000 Hz_40ms_log_rms_mae_db": 0.98072938908217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09105289871092968, "1000–3000 Hz_40ms_envelope_correlation": 0.9940787487165295, "1000–3000 Hz_40ms_modulation_mae_db": 0.6373436841238994, "3000–6000 Hz_10ms_log_rms_mae_db": 1.747336039226269, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16746310671693546, "3000–6000 Hz_10ms_envelope_correlation": 0.9811101363216291, "3000–6000 Hz_10ms_modulation_mae_db": 0.414536630199374, "3000–6000 Hz_40ms_log_rms_mae_db": 1.087478472020569, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09887889042819359, "3000–6000 Hz_40ms_envelope_correlation": 0.9930535620630824, "3000–6000 Hz_40ms_modulation_mae_db": 0.24234559093655994, "mr_stft_db": 5.652943395975329, "band_mae_db": 0.15881174604512976, "rms_gain_error_db": -0.09136356889986831, "output_peak": 0.6216493770411278}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8696554234601936, "全帯域_10ms_rms_envelope_nrmse": 0.07215266527171241, "全帯域_10ms_envelope_correlation": 0.9932194966923826, "全帯域_10ms_modulation_mae_db": 0.21967228439650585, "全帯域_40ms_log_rms_mae_db": 0.4850460453838561, "全帯域_40ms_rms_envelope_nrmse": 0.041442888603829636, "全帯域_40ms_envelope_correlation": 0.9976869300524904, "全帯域_40ms_modulation_mae_db": 0.1912816684886773, "300–1000 Hz_10ms_log_rms_mae_db": 1.6244299354780178, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1082719068514876, "300–1000 Hz_10ms_envelope_correlation": 0.9903061247920785, "300–1000 Hz_10ms_modulation_mae_db": 0.4553277387712928, "300–1000 Hz_40ms_log_rms_mae_db": 1.0275323144430024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06627627117796653, "300–1000 Hz_40ms_envelope_correlation": 0.9962394319980191, "300–1000 Hz_40ms_modulation_mae_db": 0.29812409541183066, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4708012034794273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13706906039562483, "1000–3000 Hz_10ms_envelope_correlation": 0.9864290869311813, "1000–3000 Hz_10ms_modulation_mae_db": 0.5515206841775689, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8811863440583622, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08370262734735824, "1000–3000 Hz_40ms_envelope_correlation": 0.9953841278036684, "1000–3000 Hz_40ms_modulation_mae_db": 0.30586806348903806, "3000–6000 Hz_10ms_log_rms_mae_db": 1.706669474421294, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17897824818370925, "3000–6000 Hz_10ms_envelope_correlation": 0.9810282247807953, "3000–6000 Hz_10ms_modulation_mae_db": 1.2691929222355778, "3000–6000 Hz_40ms_log_rms_mae_db": 1.070506752243503, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10237693293471563, "3000–6000 Hz_40ms_envelope_correlation": 0.9946101128703844, "3000–6000 Hz_40ms_modulation_mae_db": 1.1346413594840057, "mr_stft_db": 5.666669176220052, "band_mae_db": 0.21883879474280143, "rms_gain_error_db": -0.1037348536459985, "output_peak": 0.5927191072523328}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8231007657784102, "全帯域_10ms_rms_envelope_nrmse": 0.06684788589788113, "全帯域_10ms_envelope_correlation": 0.9941805645553087, "全帯域_10ms_modulation_mae_db": 0.23005074862855937, "全帯域_40ms_log_rms_mae_db": 0.4182111320612712, "全帯域_40ms_rms_envelope_nrmse": 0.03131744877589237, "全帯域_40ms_envelope_correlation": 0.9987305151468959, "全帯域_40ms_modulation_mae_db": 0.36479286520651316, "300–1000 Hz_10ms_log_rms_mae_db": 1.5839639240439876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09724383275674849, "300–1000 Hz_10ms_envelope_correlation": 0.9922055986058606, "300–1000 Hz_10ms_modulation_mae_db": 0.5322998313871478, "300–1000 Hz_40ms_log_rms_mae_db": 0.9407881715973713, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04847232150322589, "300–1000 Hz_40ms_envelope_correlation": 0.998060454857417, "300–1000 Hz_40ms_modulation_mae_db": 0.33255749602714596, "1000–3000 Hz_10ms_log_rms_mae_db": 1.531614304314188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1529342222812925, "1000–3000 Hz_10ms_envelope_correlation": 0.9824093521049846, "1000–3000 Hz_10ms_modulation_mae_db": 0.6419024991939195, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8939284402987367, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07256340829953116, "1000–3000 Hz_40ms_envelope_correlation": 0.9959916241545371, "1000–3000 Hz_40ms_modulation_mae_db": 0.6688179264803855, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6544389792854184, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14366946549235074, "3000–6000 Hz_10ms_envelope_correlation": 0.9859651304862155, "3000–6000 Hz_10ms_modulation_mae_db": 0.32793949674446876, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0138009584811019, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08400596106567755, "3000–6000 Hz_40ms_envelope_correlation": 0.9949981125165229, "3000–6000 Hz_40ms_modulation_mae_db": 0.31348737475894756, "mr_stft_db": 5.67133615009778, "band_mae_db": 0.11373179448261596, "rms_gain_error_db": -0.09750372423865, "output_peak": 0.7262741811185043}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7973936338922032, "全帯域_10ms_rms_envelope_nrmse": 0.06271317990470533, "全帯域_10ms_envelope_correlation": 0.9949278869307823, "全帯域_10ms_modulation_mae_db": 0.49127833798021686, "全帯域_40ms_log_rms_mae_db": 0.4135234604282359, "全帯域_40ms_rms_envelope_nrmse": 0.036545441883882285, "全帯域_40ms_envelope_correlation": 0.9982439301464106, "全帯域_40ms_modulation_mae_db": 0.5279622860420249, "300–1000 Hz_10ms_log_rms_mae_db": 1.5658826952206124, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09553386954655553, "300–1000 Hz_10ms_envelope_correlation": 0.9925095824762129, "300–1000 Hz_10ms_modulation_mae_db": 0.35647611664742046, "300–1000 Hz_40ms_log_rms_mae_db": 0.9511000019290327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05249170435098953, "300–1000 Hz_40ms_envelope_correlation": 0.9976949630098868, "300–1000 Hz_40ms_modulation_mae_db": 0.4131268155261376, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5393617227666991, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14341451747926942, "1000–3000 Hz_10ms_envelope_correlation": 0.9851123550003733, "1000–3000 Hz_10ms_modulation_mae_db": 0.45865775890305466, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9393258435156727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08108253706418556, "1000–3000 Hz_40ms_envelope_correlation": 0.9958087534235364, "1000–3000 Hz_40ms_modulation_mae_db": 0.6436631563516998, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7434379373445466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17405448157092382, "3000–6000 Hz_10ms_envelope_correlation": 0.9798498793833245, "3000–6000 Hz_10ms_modulation_mae_db": 0.4931871671148178, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0629312070345904, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09984275288270293, "3000–6000 Hz_40ms_envelope_correlation": 0.9930750872045508, "3000–6000 Hz_40ms_modulation_mae_db": 0.37570068422792796, "mr_stft_db": 5.668489156274223, "band_mae_db": 0.199648518971844, "rms_gain_error_db": -0.1055109458075547, "output_peak": 0.6246671394847036}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.698837967678366, "spectral_convergence": 0.23456845365969137, "spectral_convergence_status": "ok", "amplitude_floor": 1.6781522163747868e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.536794704636035, "spectral_convergence": 0.23201783449866747, "spectral_convergence_status": "ok", "amplitude_floor": 1.6368745556803986e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.771594925892925, "spectral_convergence": 0.23836608716943136, "spectral_convergence_status": "ok", "amplitude_floor": 1.5514964797831996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006090282805006308, "output_power": 0.0006459646016386351, "delta_db": 0.255712598852455, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005054165098664251, "output_power": 0.004939059183095345, "delta_db": -0.10005194221275195, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0022275244303280044, "output_power": 0.0021548346337730093, "delta_db": -0.14408528683714195, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006076310340369887, "output_power": 0.0005620146509231325, "delta_db": -0.33892309678143856, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001299593081283461, "output_power": 0.0001333261251605981, "delta_db": 0.11107866850090248, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003180661511978585, "output_power": 0.0003174927765138304, "delta_db": -0.007836048493236785, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1596146069463211, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09505419606591264, "pitch": null, "output_rms_dbfs": -20.578583551613768, "output_peak": 0.6557666248446633, "reference_rms_dbfs": -20.48352935554785, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.66907586606911}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002515751859318019, "log_rms_mae_db": 0.8740676111321698, "rms_envelope_nrmse": 0.07376187201923592, "envelope_correlation": 0.9928706067032245, "rms_gain_error_db": -0.09505316656948823, "reference_mean_power": 0.008953808913712473, "output_mean_power": 0.008759967704733217}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.40642035802602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010131590544257714, "output_power": 0.0009703041944138535, "reference_active": true, "delta_db": -0.1876972125174164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021577628078402175, "output_power": 0.00019910274738026203, "reference_active": true, "delta_db": -0.34926450318719793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.554717092269403e-05, "output_power": 4.655314643171158e-05, "reference_active": true, "delta_db": 0.09487632625348438}], "active_band_count": 3, "active_band_mae_db": 0.21061268065269956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002371340103862925, "log_rms_mae_db": 0.4853076951025381, "rms_envelope_nrmse": 0.04151829943866356, "envelope_correlation": 0.9976276893352103, "rms_gain_error_db": -0.09500778495450499, "reference_mean_power": 0.008997361160684636, "output_mean_power": 0.008802669071175218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.083478820328128e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000878683378715502, "output_power": 0.0008367479293645393, "reference_active": true, "delta_db": -0.2123776499417697}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.731034174913729e-05, "output_power": 9.246398219270314e-05, "reference_active": true, "delta_db": -0.22186404297159903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4907380289326384e-06, "output_power": 2.1837372487476416e-06, "reference_active": true, "delta_db": -0.5712766960336157}], "active_band_count": 3, "active_band_mae_db": 0.3351727963156615}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019737503490519328, "log_rms_mae_db": 1.6512966946431509, "rms_envelope_nrmse": 0.12403163732352904, "envelope_correlation": 0.9871742571509818, "rms_gain_error_db": -0.08353111542012943, "reference_mean_power": 0.0030829546180243575, "output_mean_power": 0.0030242244504913407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6914511369831456e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008052884179451131, "output_power": 0.0007800977588148514, "reference_active": true, "delta_db": -0.1380242270508667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019776686451848026, "output_power": 0.00016994762992717487, "reference_active": true, "delta_db": -0.6583841576558453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9868777263914452e-05, "output_power": 3.0106186215481812e-05, "reference_active": true, "delta_db": 1.804846023431348}], "active_band_count": 3, "active_band_mae_db": 0.8670848027126867}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017438355068165795, "log_rms_mae_db": 1.0350183544800626, "rms_envelope_nrmse": 0.07367461988967866, "envelope_correlation": 0.9952398776578475, "rms_gain_error_db": -0.0836546322600472, "reference_mean_power": 0.0030981947388459344, "output_mean_power": 0.0030390878122660636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5303812378070619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007441051144552582, "output_power": 0.0007176657025676444, "reference_active": true, "delta_db": -0.1571209790173274}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.887152997212749e-05, "output_power": 8.859922242667259e-05, "reference_active": true, "delta_db": -0.47641344225363247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5498600424852708e-06, "output_power": 1.8641774181746016e-06, "reference_active": true, "delta_db": 0.8019476111280771}], "active_band_count": 3, "active_band_mae_db": 0.47849401079967907}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011259893106980884, "log_rms_mae_db": 1.5225101740463969, "rms_envelope_nrmse": 0.13712859311422157, "envelope_correlation": 0.986477475091154, "rms_gain_error_db": -0.31797693405062827, "reference_mean_power": 0.0006356182176644616, "output_mean_power": 0.000590743079706584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.9581273226652876e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002027625859711201, "output_power": 0.0001864261805695153, "reference_active": true, "delta_db": -0.3648091944264277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.693176540437241e-05, "output_power": 2.2216230162418193e-05, "reference_active": true, "delta_db": -0.8359445668493535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.128022173574573e-06, "output_power": 5.826729680538623e-06, "reference_active": true, "delta_db": -0.8754417168307614}], "active_band_count": 3, "active_band_mae_db": 0.6920651593688475}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001021231893168278, "log_rms_mae_db": 0.9156634040725341, "rms_envelope_nrmse": 0.07875355135042326, "envelope_correlation": 0.9960828449525717, "rms_gain_error_db": -0.3180382096335773, "reference_mean_power": 0.0006388062286465308, "output_mean_power": 0.0005936976379744042}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.69069653778161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018976599367877858, "output_power": 0.0001743290670358012, "reference_active": true, "delta_db": -0.36848582976005495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3827689740852905e-05, "output_power": 1.12316034888357e-05, "reference_active": true, "delta_db": -0.9030786339989361}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.22079217642337e-07, "output_power": 3.613887034491265e-07, "reference_active": true, "delta_db": -0.6741939592300723}], "active_band_count": 3, "active_band_mae_db": 0.6485861409963544}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.113996831641667e-05, "log_rms_mae_db": 1.716435296133741, "rms_envelope_nrmse": 0.16139648370546453, "envelope_correlation": 0.9830998513500664, "rms_gain_error_db": 0.20742472649267785, "reference_mean_power": 0.0001805297653048898, "output_mean_power": 0.00018936132780117678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6198199988921157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.959931264840548e-05, "output_power": 0.00010558826269886436, "reference_active": true, "delta_db": 0.25359302959549723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361167220741e-06, "output_power": 9.323807130775654e-06, "reference_active": true, "delta_db": -0.017367219491407272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4041286689847167e-06, "output_power": 2.3191852243011473e-06, "reference_active": true, "delta_db": -0.15622271962889508}], "active_band_count": 3, "active_band_mae_db": 0.14239432290526655}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.456026972891915e-05, "log_rms_mae_db": 1.088554727131276, "rms_envelope_nrmse": 0.08977153256211638, "envelope_correlation": 0.9947322294163734, "rms_gain_error_db": 0.207407647668847, "reference_mean_power": 0.00018143628465930542, "output_mean_power": 0.00019031144590558814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.558011664744216e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.473019444861243e-05, "output_power": 0.00010070434089261649, "reference_active": true, "delta_db": 0.26559762594973574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179317052925175e-06, "output_power": 4.392056923140422e-06, "reference_active": true, "delta_db": 0.21562641135901325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8116784694632823e-07, "output_power": 1.6015769546083844e-07, "reference_active": true, "delta_db": -0.5353331203589913}], "active_band_count": 3, "active_band_mae_db": 0.33885238588924677}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8830212611533288, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6e55987b65de94d7e178070d364d7c0459a98cd2948a5fec953ded4c2f605204", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9845835908615075, "standard_deviation": 0.001446790648146779, "minimum": 0.9824093521049846, "maximum": 0.986477475091154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5224757699893514, "standard_deviation": 0.030121826262090645, "minimum": 1.4708012034794273, "maximum": 1.5703413872048586, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6063432650924796, "standard_deviation": 0.2670940946638224, "minimum": 0.056533082367502784, "maximum": 0.910777566850275, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14484451174293564, "standard_deviation": 0.005764686126032736, "minimum": 0.13706906039562483, "maximum": 0.1529342222812925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9952287218520611, "standard_deviation": 0.0008490537671016158, "minimum": 0.993818199727102, "maximum": 0.9960828449525717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9167011495336266, "standard_deviation": 0.03385005347870121, "minimum": 0.8789521741036337, "maximum": 0.98072938908217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.606529843202617, "standard_deviation": 0.2947044763174702, "minimum": 0.1529384246305758, "maximum": 1.1568326906383053, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08299057826229352, "standard_deviation": 0.006283036911709643, "minimum": 0.07256340829953116, "maximum": 0.09105289871092968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9824893780135362, "standard_deviation": 0.0020835197028707605, "minimum": 0.9798498793833245, "maximum": 0.9859651304862155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7266323191367952, "standard_deviation": 0.03996892257257964, "minimum": 1.6544389792854184, "maximum": 1.7955415570658062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.46973418381813237, "standard_deviation": 0.3450519486425209, "minimum": 0.14239432290526655, "maximum": 1.2691929222355778, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16405944056456173, "standard_deviation": 0.011807090079486746, "minimum": 0.14366946549235074, "maximum": 0.17897824818370925, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943852887087279, "standard_deviation": 0.0008701900544613277, "minimum": 0.9930535620630824, "maximum": 0.9951949457209581, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0719674662589465, "standard_deviation": 0.03176866663395976, "minimum": 1.0138009584811019, "maximum": 1.1198900060169839, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4426536175385922, "standard_deviation": 0.2870446372493872, "minimum": 0.24234559093655994, "maximum": 1.1346413594840057, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0928384821891847, "standard_deviation": 0.006598155067583024, "minimum": 0.08400596106567755, "maximum": 0.10237693293471563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990462857691478, "standard_deviation": 0.00177480440917426, "minimum": 0.9871742571509818, "maximum": 0.9925095824762129, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6021609936578025, "standard_deviation": 0.03201747725941854, "minimum": 1.5658826952206124, "maximum": 1.6512966946431509, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5833570423878276, "standard_deviation": 0.19313868940642073, "minimum": 0.34715933738393373, "maximum": 0.8670848027126867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10711290475104046, "standard_deviation": 0.009446924416032178, "minimum": 0.09553386954655553, "maximum": 0.12403163732352904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967198154869928, "standard_deviation": 0.0010241462992885049, "minimum": 0.9952398776578475, "maximum": 0.998060454857417, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9825338422381322, "standard_deviation": 0.04727196387212887, "minimum": 0.9129081812507971, "maximum": 1.0392725935083758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44488086859966236, "standard_deviation": 0.1309506789284496, "minimum": 0.29812409541183066, "maximum": 0.6767130221733401, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06151297518183889, "standard_deviation": 0.009185946788578397, "minimum": 0.04847232150322589, "maximum": 0.07367461988967866, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1616517005114864, "standard_deviation": 0.03615279653589718, "minimum": 0.11373179448261596, "maximum": 0.21883879474280143, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6672425537254405, "standard_deviation": 0.006607435197234205, "minimum": 5.652943395975329, "maximum": 5.674125744350377, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6260339300735923, "standard_deviation": 0.054772544343230756, "minimum": 0.5348040351937746, "maximum": 0.7262741811185043, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1056963619431519, "standard_deviation": 0.014219114832713074, "minimum": -0.1310749169482331, "maximum": -0.09136356889986831, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9937606337702568, "standard_deviation": 0.0007672524452019143, "minimum": 0.9927113544487729, "maximum": 0.9949278869307823, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8549694939021869, "standard_deviation": 0.03159494526877766, "minimum": 0.7973936338922032, "maximum": 0.887656021921718, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27468771063459774, "standard_deviation": 0.10326011233659142, "minimum": 0.16647633935228695, "maximum": 0.49127833798021686, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06920358263848685, "standard_deviation": 0.0040312026054755284, "minimum": 0.06271317990470533, "maximum": 0.07450988109129505, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980660948747132, "standard_deviation": 0.0003928373130275225, "minimum": 0.9976276893352103, "maximum": 0.9987305151468959, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4591963013905401, "standard_deviation": 0.028115876329133798, "minimum": 0.4135234604282359, "maximum": 0.4853076951025381, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3785920641902773, "standard_deviation": 0.1297132889557915, "minimum": 0.1912816684886773, "maximum": 0.5697915942215178, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03802469434602066, "standard_deviation": 0.0036145432665513613, "minimum": 0.03131744877589237, "maximum": 0.04151829943866356, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0041", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.681070566177368, "predicted_mos": 2.681070566177368, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3917e50440a60f1e4a979fd304c07ac3f7eb56435ef374992a46fe4fdb60824a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9587680003605783, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8457009566148384, "全帯域_10ms_rms_envelope_nrmse": 0.06361692060216093, "全帯域_10ms_envelope_correlation": 0.9947656512476758, "全帯域_10ms_modulation_mae_db": 0.5873020056166546, "全帯域_40ms_log_rms_mae_db": 0.43667762284295225, "全帯域_40ms_rms_envelope_nrmse": 0.03329594163769708, "全帯域_40ms_envelope_correlation": 0.998563130705634, "全帯域_40ms_modulation_mae_db": 0.42988256097025346, "300–1000 Hz_10ms_log_rms_mae_db": 1.5388171595786584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09957721333595722, "300–1000 Hz_10ms_envelope_correlation": 0.9919721272729511, "300–1000 Hz_10ms_modulation_mae_db": 0.40740060476510936, "300–1000 Hz_40ms_log_rms_mae_db": 0.9140273293569848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05668372538730297, "300–1000 Hz_40ms_envelope_correlation": 0.997470712960114, "300–1000 Hz_40ms_modulation_mae_db": 0.5704405877846512, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4790537841766318, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14250049071316934, "1000–3000 Hz_10ms_envelope_correlation": 0.9847787891464272, "1000–3000 Hz_10ms_modulation_mae_db": 0.4694355839895348, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8529569462698543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0716465008623806, "1000–3000 Hz_40ms_envelope_correlation": 0.9961520075484697, "1000–3000 Hz_40ms_modulation_mae_db": 0.12287857592166447, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7054536073866458, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16786756079183415, "3000–6000 Hz_10ms_envelope_correlation": 0.9812740591426155, "3000–6000 Hz_10ms_modulation_mae_db": 0.7663142871208719, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0197968885006143, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09619568706445102, "3000–6000 Hz_40ms_envelope_correlation": 0.9935922736848037, "3000–6000 Hz_40ms_modulation_mae_db": 0.3217916813325523, "mr_stft_db": 5.644264531233683, "band_mae_db": 0.14186232883888136, "rms_gain_error_db": -0.1028870819984276, "output_peak": 0.6865870779354848}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8194103755058842, "全帯域_10ms_rms_envelope_nrmse": 0.06716834135082711, "全帯域_10ms_envelope_correlation": 0.9942147874898868, "全帯域_10ms_modulation_mae_db": 0.4162022477463554, "全帯域_40ms_log_rms_mae_db": 0.4819548084010292, "全帯域_40ms_rms_envelope_nrmse": 0.03579610939061846, "全帯域_40ms_envelope_correlation": 0.9984022487911288, "全帯域_40ms_modulation_mae_db": 0.6798853553490528, "300–1000 Hz_10ms_log_rms_mae_db": 1.5423815495957485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10174071824808756, "300–1000 Hz_10ms_envelope_correlation": 0.9917402623302459, "300–1000 Hz_10ms_modulation_mae_db": 0.518738818704703, "300–1000 Hz_40ms_log_rms_mae_db": 0.917787722223846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05801060033586833, "300–1000 Hz_40ms_envelope_correlation": 0.9975165679917077, "300–1000 Hz_40ms_modulation_mae_db": 0.819985155916329, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4527521777409527, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14016304689355785, "1000–3000 Hz_10ms_envelope_correlation": 0.9853742605425684, "1000–3000 Hz_10ms_modulation_mae_db": 0.5514499747684416, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8672101245901441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08089577278362077, "1000–3000 Hz_40ms_envelope_correlation": 0.9951747705567422, "1000–3000 Hz_40ms_modulation_mae_db": 0.344988418026378, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6669477328269648, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15994538900818223, "3000–6000 Hz_10ms_envelope_correlation": 0.9826588117374254, "3000–6000 Hz_10ms_modulation_mae_db": 1.0033855695973612, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0065808282436568, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08697589716620245, "3000–6000 Hz_40ms_envelope_correlation": 0.9949793485553542, "3000–6000 Hz_40ms_modulation_mae_db": 0.7471914035479855, "mr_stft_db": 5.628009190054141, "band_mae_db": 0.19805465657978114, "rms_gain_error_db": -0.12595252632881138, "output_peak": 0.5784269027474491}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8682284700288948, "全帯域_10ms_rms_envelope_nrmse": 0.07028157992944445, "全帯域_10ms_envelope_correlation": 0.9934905245191227, "全帯域_10ms_modulation_mae_db": 0.25837583296364663, "全帯域_40ms_log_rms_mae_db": 0.45120950770178603, "全帯域_40ms_rms_envelope_nrmse": 0.03574102789962732, "全帯域_40ms_envelope_correlation": 0.9982173513150379, "全帯域_40ms_modulation_mae_db": 0.323952899507691, "300–1000 Hz_10ms_log_rms_mae_db": 1.560579069618383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11141321340983786, "300–1000 Hz_10ms_envelope_correlation": 0.9896521394357578, "300–1000 Hz_10ms_modulation_mae_db": 0.5670151340830475, "300–1000 Hz_40ms_log_rms_mae_db": 0.8742494510520812, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055210466197532816, "300–1000 Hz_40ms_envelope_correlation": 0.9973042770494774, "300–1000 Hz_40ms_modulation_mae_db": 0.3104616781103951, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4567559539766946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14894895640118624, "1000–3000 Hz_10ms_envelope_correlation": 0.9844441728760788, "1000–3000 Hz_10ms_modulation_mae_db": 0.6515550755907008, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8738689064702613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08528856803503966, "1000–3000 Hz_40ms_envelope_correlation": 0.9960438436456329, "1000–3000 Hz_40ms_modulation_mae_db": 0.4031164233215326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6744838764233418, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15889576154315177, "3000–6000 Hz_10ms_envelope_correlation": 0.9829576291382055, "3000–6000 Hz_10ms_modulation_mae_db": 0.47682982307917987, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0211190101463494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09708359918169025, "3000–6000 Hz_40ms_envelope_correlation": 0.993648033253342, "3000–6000 Hz_40ms_modulation_mae_db": 0.24840604181018042, "mr_stft_db": 5.6288076034273615, "band_mae_db": 0.20391521641852114, "rms_gain_error_db": -0.07087325400422292, "output_peak": 0.6939818701619229}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8408587248381059, "全帯域_10ms_rms_envelope_nrmse": 0.06780024991334539, "全帯域_10ms_envelope_correlation": 0.9941480753897297, "全帯域_10ms_modulation_mae_db": 0.583242279996223, "全帯域_40ms_log_rms_mae_db": 0.447945968222262, "全帯域_40ms_rms_envelope_nrmse": 0.036566280594902266, "全帯域_40ms_envelope_correlation": 0.9983790441230856, "全帯域_40ms_modulation_mae_db": 0.5442453248176305, "300–1000 Hz_10ms_log_rms_mae_db": 1.5088139330957202, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10688430824497387, "300–1000 Hz_10ms_envelope_correlation": 0.9908199175745459, "300–1000 Hz_10ms_modulation_mae_db": 0.6981074119248167, "300–1000 Hz_40ms_log_rms_mae_db": 0.8767935169552614, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061608680779952206, "300–1000 Hz_40ms_envelope_correlation": 0.9971086067124229, "300–1000 Hz_40ms_modulation_mae_db": 0.7795894855329534, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4750177856979123, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14945907642330047, "1000–3000 Hz_10ms_envelope_correlation": 0.9843393874773432, "1000–3000 Hz_10ms_modulation_mae_db": 0.888734985619577, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8627297073228852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08919865160767647, "1000–3000 Hz_40ms_envelope_correlation": 0.9954578753387601, "1000–3000 Hz_40ms_modulation_mae_db": 1.2676290364307043, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7120625183375322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15955121749428378, "3000–6000 Hz_10ms_envelope_correlation": 0.9826731200125886, "3000–6000 Hz_10ms_modulation_mae_db": 0.31117996291279365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0353246359690715, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09319170980605454, "3000–6000 Hz_40ms_envelope_correlation": 0.993965969219942, "3000–6000 Hz_40ms_modulation_mae_db": 0.1864138147749562, "mr_stft_db": 5.635633893038857, "band_mae_db": 0.22757849046056802, "rms_gain_error_db": -0.13634011996907686, "output_peak": 0.5852820146433303}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8120996369806791, "全帯域_10ms_rms_envelope_nrmse": 0.06469001120140566, "全帯域_10ms_envelope_correlation": 0.9946020192021436, "全帯域_10ms_modulation_mae_db": 0.5624143755286463, "全帯域_40ms_log_rms_mae_db": 0.4453022343553329, "全帯域_40ms_rms_envelope_nrmse": 0.03374063151342936, "全帯域_40ms_envelope_correlation": 0.9985469081950462, "全帯域_40ms_modulation_mae_db": 0.4105589530324077, "300–1000 Hz_10ms_log_rms_mae_db": 1.5264906589850904, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09854354854141477, "300–1000 Hz_10ms_envelope_correlation": 0.9919760842106802, "300–1000 Hz_10ms_modulation_mae_db": 0.3707273785772387, "300–1000 Hz_40ms_log_rms_mae_db": 0.9448240171074296, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05367348147834232, "300–1000 Hz_40ms_envelope_correlation": 0.9975338793072911, "300–1000 Hz_40ms_modulation_mae_db": 0.6714581548040269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4864869626490433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13449902662014535, "1000–3000 Hz_10ms_envelope_correlation": 0.9864850824083125, "1000–3000 Hz_10ms_modulation_mae_db": 0.4193177218728432, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8575272406738392, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07223705561391099, "1000–3000 Hz_40ms_envelope_correlation": 0.996112225928955, "1000–3000 Hz_40ms_modulation_mae_db": 0.5847775994968115, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6991590944589134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17976433719511634, "3000–6000 Hz_10ms_envelope_correlation": 0.9791233718755291, "3000–6000 Hz_10ms_modulation_mae_db": 0.5757038431984495, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0063785373315055, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10393639739695697, "3000–6000 Hz_40ms_envelope_correlation": 0.9929315020744246, "3000–6000 Hz_40ms_modulation_mae_db": 0.6931916060402615, "mr_stft_db": 5.633310352472716, "band_mae_db": 0.19494305724501101, "rms_gain_error_db": -0.11021060182280155, "output_peak": 0.7068934039993662}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8619769562520492, "全帯域_10ms_rms_envelope_nrmse": 0.07035077878660198, "全帯域_10ms_envelope_correlation": 0.993737638132327, "全帯域_10ms_modulation_mae_db": 0.7910290410243229, "全帯域_40ms_log_rms_mae_db": 0.46012217861376636, "全帯域_40ms_rms_envelope_nrmse": 0.040509289204746474, "全帯域_40ms_envelope_correlation": 0.9980009041963316, "全帯域_40ms_modulation_mae_db": 0.9534255889640534, "300–1000 Hz_10ms_log_rms_mae_db": 1.534124743769609, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11476776041276682, "300–1000 Hz_10ms_envelope_correlation": 0.9894701994394126, "300–1000 Hz_10ms_modulation_mae_db": 0.4324525126768779, "300–1000 Hz_40ms_log_rms_mae_db": 0.9033736642787484, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07381064575566162, "300–1000 Hz_40ms_envelope_correlation": 0.9957717102505662, "300–1000 Hz_40ms_modulation_mae_db": 0.6136743097313202, "1000–3000 Hz_10ms_log_rms_mae_db": 1.462398263222829, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1372220133787812, "1000–3000 Hz_10ms_envelope_correlation": 0.9874310192699792, "1000–3000 Hz_10ms_modulation_mae_db": 0.9148360935903992, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8451101611682145, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07976190627050334, "1000–3000 Hz_40ms_envelope_correlation": 0.997147213442049, "1000–3000 Hz_40ms_modulation_mae_db": 1.08470024177587, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6968860161434423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18074277822953422, "3000–6000 Hz_10ms_envelope_correlation": 0.9776254095044535, "3000–6000 Hz_10ms_modulation_mae_db": 0.3526401473252026, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0530770312962137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09528542014831116, "3000–6000 Hz_40ms_envelope_correlation": 0.9936866350621774, "3000–6000 Hz_40ms_modulation_mae_db": 0.47043920501540476, "mr_stft_db": 5.63377796444417, "band_mae_db": 0.2633892931556508, "rms_gain_error_db": -0.14891279485592604, "output_peak": 0.5602917357646373}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8781375007022528, "全帯域_10ms_rms_envelope_nrmse": 0.07079733903788578, "全帯域_10ms_envelope_correlation": 0.9935314243626249, "全帯域_10ms_modulation_mae_db": 0.15943175443747679, "全帯域_40ms_log_rms_mae_db": 0.48035884558292175, "全帯域_40ms_rms_envelope_nrmse": 0.03861303918862256, "全帯域_40ms_envelope_correlation": 0.9980798305762797, "全帯域_40ms_modulation_mae_db": 0.25269554925202314, "300–1000 Hz_10ms_log_rms_mae_db": 1.5490896083279175, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10970538443168745, "300–1000 Hz_10ms_envelope_correlation": 0.9900251068179745, "300–1000 Hz_10ms_modulation_mae_db": 0.9551144899473308, "300–1000 Hz_40ms_log_rms_mae_db": 0.8725542884372803, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06366927594253348, "300–1000 Hz_40ms_envelope_correlation": 0.9965233838469265, "300–1000 Hz_40ms_modulation_mae_db": 0.4076626772016844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4843587253460104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14860231357854115, "1000–3000 Hz_10ms_envelope_correlation": 0.983725146477345, "1000–3000 Hz_10ms_modulation_mae_db": 0.36938140186483603, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8601138921566289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08443234375709409, "1000–3000 Hz_40ms_envelope_correlation": 0.9950584435533187, "1000–3000 Hz_40ms_modulation_mae_db": 0.2606220628641091, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6638138756219056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17271176610334424, "3000–6000 Hz_10ms_envelope_correlation": 0.9795926198412767, "3000–6000 Hz_10ms_modulation_mae_db": 0.9609188970481263, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0222974354474375, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08875648097395858, "3000–6000 Hz_40ms_envelope_correlation": 0.994338634431466, "3000–6000 Hz_40ms_modulation_mae_db": 0.4036268707966892, "mr_stft_db": 5.6428704751726455, "band_mae_db": 0.20645494736785852, "rms_gain_error_db": -0.1230242337877483, "output_peak": 0.6115344873308971}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8505340289276478, "全帯域_10ms_rms_envelope_nrmse": 0.06851110304870764, "全帯域_10ms_envelope_correlation": 0.9940121278094851, "全帯域_10ms_modulation_mae_db": 0.449124252617231, "全帯域_40ms_log_rms_mae_db": 0.4409774344199389, "全帯域_40ms_rms_envelope_nrmse": 0.03694715098657091, "全帯域_40ms_envelope_correlation": 0.9983304145498967, "全帯域_40ms_modulation_mae_db": 0.43901840913312323, "300–1000 Hz_10ms_log_rms_mae_db": 1.558184487790236, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10138631836538495, "300–1000 Hz_10ms_envelope_correlation": 0.9916390118001379, "300–1000 Hz_10ms_modulation_mae_db": 0.30164255211982566, "300–1000 Hz_40ms_log_rms_mae_db": 0.9158195854872287, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05521292722254018, "300–1000 Hz_40ms_envelope_correlation": 0.9975996611488995, "300–1000 Hz_40ms_modulation_mae_db": 0.43899714638432624, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4584013881538627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15447681846086167, "1000–3000 Hz_10ms_envelope_correlation": 0.9832774643331355, "1000–3000 Hz_10ms_modulation_mae_db": 0.48041980287000796, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8573510417189973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09336740732850426, "1000–3000 Hz_40ms_envelope_correlation": 0.9950054718491724, "1000–3000 Hz_40ms_modulation_mae_db": 0.7945236375886376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6532509769589212, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17239231786064116, "3000–6000 Hz_10ms_envelope_correlation": 0.9797119396241883, "3000–6000 Hz_10ms_modulation_mae_db": 0.6116557076400322, "3000–6000 Hz_40ms_log_rms_mae_db": 1.025964051507937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09923639007510013, "3000–6000 Hz_40ms_envelope_correlation": 0.9928620255865238, "3000–6000 Hz_40ms_modulation_mae_db": 0.9607988380557764, "mr_stft_db": 5.6405693290274606, "band_mae_db": 0.23382295102121473, "rms_gain_error_db": -0.1365772001695059, "output_peak": 0.6265017102631525}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6873868898564615, "spectral_convergence": 0.22567034023890198, "spectral_convergence_status": "ok", "amplitude_floor": 1.6781522163747868e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.489863744912664, "spectral_convergence": 0.22481490298871845, "spectral_convergence_status": "ok", "amplitude_floor": 1.6368745556803986e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.755542958931926, "spectral_convergence": 0.23390545511439814, "spectral_convergence_status": "ok", "amplitude_floor": 1.5514964797831996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006090282805006308, "output_power": 0.0006407536330138212, "delta_db": 0.2205361751315091, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005054165098664251, "output_power": 0.004983199287444834, "delta_db": -0.06141168653951959, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0022275244303280044, "output_power": 0.0020806354135885063, "delta_db": -0.29626489698969444, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006076310340369887, "output_power": 0.0005878356981609031, "delta_db": -0.1438399040676009, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001299593081283461, "output_power": 0.0001262311128840297, "delta_db": -0.12640979524281673, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003180661511978585, "output_power": 0.00031826479761959273, "delta_db": 0.002711515062147636, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14186232883888136, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1028870819984276, "pitch": null, "output_rms_dbfs": -20.58641643754628, "output_peak": 0.6865870779354848, "reference_rms_dbfs": -20.48352935554785, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.644264531233683}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002515751859318019, "log_rms_mae_db": 0.8457009566148384, "rms_envelope_nrmse": 0.06361692060216093, "envelope_correlation": 0.9947656512476758, "rms_gain_error_db": -0.10288268256701998, "reference_mean_power": 0.008953808913712473, "output_mean_power": 0.008744189350841826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.40642035802602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010131590544257714, "output_power": 0.0009645539106382293, "reference_active": true, "delta_db": -0.21351123845667536}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021577628078402175, "output_power": 0.00018795970865985364, "reference_active": true, "delta_db": -0.5993893994396006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.554717092269403e-05, "output_power": 3.66067243386205e-05, "reference_active": true, "delta_db": -0.949005378953688}], "active_band_count": 3, "active_band_mae_db": 0.5873020056166546}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002371340103862925, "log_rms_mae_db": 0.43667762284295225, "rms_envelope_nrmse": 0.03329594163769708, "envelope_correlation": 0.998563130705634, "rms_gain_error_db": -0.10277939215397994, "reference_mean_power": 0.008997361160684636, "output_mean_power": 0.00878693096816917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.083478820328128e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000878683378715502, "output_power": 0.0008365965624056352, "reference_active": true, "delta_db": -0.2131633558797541}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.731034174913729e-05, "output_power": 8.714122535162523e-05, "reference_active": true, "delta_db": -0.47935335229598836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4907380289326384e-06, "output_power": 2.1707757141268947e-06, "reference_active": true, "delta_db": -0.597130974735018}], "active_band_count": 3, "active_band_mae_db": 0.42988256097025346}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019737503490519328, "log_rms_mae_db": 1.5388171595786584, "rms_envelope_nrmse": 0.09957721333595722, "envelope_correlation": 0.9919721272729511, "rms_gain_error_db": -0.16617535429504265, "reference_mean_power": 0.0030829546180243575, "output_mean_power": 0.0029672189669080534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6914511369831456e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008052884179451131, "output_power": 0.0007488224140340779, "reference_active": true, "delta_db": -0.3157261750973237}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019776686451848026, "output_power": 0.00016889483090936174, "reference_active": true, "delta_db": -0.6853717007269157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9868777263914452e-05, "output_power": 2.0906509737119213e-05, "reference_active": true, "delta_db": 0.22110393847108867}], "active_band_count": 3, "active_band_mae_db": 0.40740060476510936}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017438355068165795, "log_rms_mae_db": 0.9140273293569848, "rms_envelope_nrmse": 0.05668372538730297, "envelope_correlation": 0.997470712960114, "rms_gain_error_db": -0.16664219312962744, "reference_mean_power": 0.0030981947388459344, "output_mean_power": 0.00298156644942729}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5303812378070619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007441051144552582, "output_power": 0.0006888815439022017, "reference_active": true, "delta_db": -0.3348974002981752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.887152997212749e-05, "output_power": 8.359268634338224e-05, "reference_active": true, "delta_db": -0.7290297264203702}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5498600424852708e-06, "output_power": 1.3352205814093202e-06, "reference_active": true, "delta_db": -0.6473946366354082}], "active_band_count": 3, "active_band_mae_db": 0.5704405877846512}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011259893106980884, "log_rms_mae_db": 1.4790537841766318, "rms_envelope_nrmse": 0.14250049071316934, "envelope_correlation": 0.9847787891464272, "rms_gain_error_db": -0.15599175986318983, "reference_mean_power": 0.0006356182176644616, "output_mean_power": 0.0006131929600862611}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.9581273226652876e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002027625859711201, "output_power": 0.0001911571638414401, "reference_active": true, "delta_db": -0.25597243082937776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.693176540437241e-05, "output_power": 2.5537473172816124e-05, "reference_active": true, "delta_db": -0.23086899372494896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.128022173574573e-06, "output_power": 8.812833695663055e-06, "reference_active": true, "delta_db": 0.9214653274142776}], "active_band_count": 3, "active_band_mae_db": 0.4694355839895348}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001021231893168278, "log_rms_mae_db": 0.8529569462698543, "rms_envelope_nrmse": 0.0716465008623806, "envelope_correlation": 0.9961520075484697, "rms_gain_error_db": -0.1560371184210704, "reference_mean_power": 0.0006388062286465308, "output_mean_power": 0.0006162620584024367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.69069653778161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018976599367877858, "output_power": 0.0001811377224676907, "reference_active": true, "delta_db": -0.2020948606239781}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3827689740852905e-05, "output_power": 1.3358087723246823e-05, "reference_active": true, "delta_db": -0.15005335313785118}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.22079217642337e-07, "output_power": 4.236846406086993e-07, "reference_active": true, "delta_db": 0.0164875140031642}], "active_band_count": 3, "active_band_mae_db": 0.12287857592166447}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.113996831641667e-05, "log_rms_mae_db": 1.7054536073866458, "rms_envelope_nrmse": 0.16786756079183415, "envelope_correlation": 0.9812740591426155, "rms_gain_error_db": 0.13146048973514673, "reference_mean_power": 0.0001805297653048898, "output_mean_power": 0.00018607793015839257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6198199988921157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.959931264840548e-05, "output_power": 9.7886697709731e-05, "reference_active": true, "delta_db": -0.07532663830639745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361167220741e-06, "output_power": 9.716848844698395e-06, "reference_active": true, "delta_db": 0.16195443680374957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4041286689847167e-06, "output_power": 3.864772070879549e-06, "reference_active": true, "delta_db": 2.061661786252469}], "active_band_count": 3, "active_band_mae_db": 0.7663142871208719}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.456026972891915e-05, "log_rms_mae_db": 1.0197968885006143, "rms_envelope_nrmse": 0.09619568706445102, "envelope_correlation": 0.9935922736848037, "rms_gain_error_db": 0.1314323316935997, "reference_mean_power": 0.00018143628465930542, "output_mean_power": 0.00018701109677436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.558011664744216e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.473019444861243e-05, "output_power": 9.476546300326913e-05, "reference_active": true, "delta_db": 0.0016166005126374743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179317052925175e-06, "output_power": 4.245690585205348e-06, "reference_active": true, "delta_db": 0.068430217353669}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8116784694632823e-07, "output_power": 2.2264551223836357e-07, "reference_active": true, "delta_db": 0.8953282261313504}], "active_band_count": 3, "active_band_mae_db": 0.3217916813325523}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8877657438028939, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4fcc16914a5df372755b7184b9c1058ebdf5ead8fdac53b7f591a3be3208f58a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9849819153163987, "standard_deviation": 0.0013958840785030248, "minimum": 0.9832774643331355, "maximum": 0.9874310192699792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.469403130120492, "standard_deviation": 0.013349683835152254, "minimum": 1.4527521777409527, "maximum": 1.4864869626490433, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5931413300207926, "standard_deviation": 0.20850004158784355, "minimum": 0.36938140186483603, "maximum": 0.9148360935903992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1444839678086929, "standard_deviation": 0.00693382224543475, "minimum": 0.13449902662014535, "maximum": 0.15447681846086167, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957689814828875, "standard_deviation": 0.0007345827801480979, "minimum": 0.9950054718491724, "maximum": 0.997147213442049, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8596085025463531, "standard_deviation": 0.008742457971387927, "minimum": 0.8451101611682145, "maximum": 0.8738689064702613, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6079044994282135, "standard_deviation": 0.4079303967357027, "minimum": 0.12287857592166447, "maximum": 1.2676290364307043, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08210352578234127, "standard_deviation": 0.00761973170018545, "minimum": 0.0716465008623806, "maximum": 0.09336740732850426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9807021201095353, "standard_deviation": 0.001974767991909533, "minimum": 0.9776254095044535, "maximum": 0.9829576291382055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6840072122697083, "standard_deviation": 0.021967273727488364, "minimum": 1.6532509769589212, "maximum": 1.7120625183375322, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6323285297402522, "standard_deviation": 0.25988533218709875, "minimum": 0.31117996291279365, "maximum": 1.0033855695973612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.168983891028261, "standard_deviation": 0.008893739951597595, "minimum": 0.15889576154315177, "maximum": 0.18074277822953422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937505527335042, "standard_deviation": 0.000697152981852491, "minimum": 0.9928620255865238, "maximum": 0.9949793485553542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.023817302305348, "standard_deviation": 0.015213340319322362, "minimum": 1.0063785373315055, "maximum": 1.0530770312962137, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5039824326717258, "standard_deviation": 0.2710273888274994, "minimum": 0.1864138147749562, "maximum": 0.9607988380557764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09508269772659064, "standard_deviation": 0.005480376359617704, "minimum": 0.08697589716620245, "maximum": 0.10393639739695697, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9909118561102133, "standard_deviation": 0.0010644787932617052, "minimum": 0.9894701994394126, "maximum": 0.9919760842106802, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5398101513451703, "standard_deviation": 0.017042673043778624, "minimum": 1.5088139330957202, "maximum": 1.560579069618383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5313998628498687, "standard_deviation": 0.21129691528640235, "minimum": 0.30164255211982566, "maximum": 0.9551144899473308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10550230812376382, "standard_deviation": 0.00603543341394658, "minimum": 0.09854354854141477, "maximum": 0.11476776041276682, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971035999084257, "standard_deviation": 0.0006422514936511146, "minimum": 0.9957717102505662, "maximum": 0.9975996611488995, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9024286968623576, "standard_deviation": 0.02588908373956018, "minimum": 0.8725542884372803, "maximum": 0.9448240171074296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5765336494332108, "standard_deviation": 0.18093231299795834, "minimum": 0.3104616781103951, "maximum": 0.819985155916329, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059734975387466735, "standard_deviation": 0.0066231131693448705, "minimum": 0.05367348147834232, "maximum": 0.07381064575566162, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20875261763593583, "standard_deviation": 0.035408316656254, "minimum": 0.14186232883888136, "maximum": 0.2633892931556508, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.635905417358879, "standard_deviation": 0.0061435494458123435, "minimum": 5.628009190054141, "maximum": 5.644264531233683, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.63118740035578, "standard_deviation": 0.05740710656227596, "minimum": 0.5602917357646373, "maximum": 0.7068934039993662, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11934722661706507, "standard_deviation": 0.024556186007838182, "minimum": -0.14891279485592604, "maximum": -0.07087325400422292, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940627810191245, "standard_deviation": 0.00046801808812766023, "minimum": 0.9934905245191227, "maximum": 0.9947656512476758, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.847118331231294, "standard_deviation": 0.02291444941354894, "minimum": 0.8120996369806791, "maximum": 0.8781375007022528, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4758902237413196, "standard_deviation": 0.20088144933266489, "minimum": 0.15943175443747679, "maximum": 0.7910290410243229, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06790204048379736, "standard_deviation": 0.0026629816378098306, "minimum": 0.06361692060216093, "maximum": 0.07079733903788578, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983149790565551, "standard_deviation": 0.00020381993506616466, "minimum": 0.9980009041963316, "maximum": 0.998563130705634, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4555685750174987, "standard_deviation": 0.017253566684672407, "minimum": 0.43667762284295225, "maximum": 0.4819548084010292, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5042080801282794, "standard_deviation": 0.22314520403048752, "minimum": 0.25269554925202314, "maximum": 0.9534255889640534, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0364011838020268, "standard_deviation": 0.002378990170128922, "minimum": 0.03329594163769708, "maximum": 0.040509289204746474, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3575-170457-0054", "language": "en", "style": "normal", "speaker": "librispeech_3575", "source_utterance_id": "3575-170457-0054", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3575-170457-0054.wav", "source_file_sha256": "c7b404d20513822ba1dcab36e7ebd3c0056fd109b1c1ddce7f83804c6e21ebdf", "decoded_pcm_sha256": "5ff0a2aa3e7e6a57cda2e60dda3bc719388bba59551820dff2445caca9cf7c6a", "original_source_sha256": "638053737df40babfd92ea956f1083964cfe38bf547b96334228a8f67b2cb78b", "corpus_entry_id": "librispeech_3575-170457-0054"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.132341146469116, "predicted_mos": 3.132341146469116, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3d1434c601d1d2f47ecd28e2a73e2f1f94013f7309d07bc8576f743d4f9d11bd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.953017054001975, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3575-170457-0054", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "3d1434c601d1d2f47ecd28e2a73e2f1f94013f7309d07bc8576f743d4f9d11bd", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "bf86c0c00709c2e5eee207cba7f35d17746c67a2a9f9c9ec302b8a3636061627"}, "rows": [{"case": "librispeech_3575-170457-0054", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.544389247894287, "predicted_mos": 2.544389247894287, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "07018f71ae85fe25e7f9dd045de5ccce7d4e0fe5be8ba0c0375a5fb880199f37", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.198664941999596, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3575-170457-0054", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "07018f71ae85fe25e7f9dd045de5ccce7d4e0fe5be8ba0c0375a5fb880199f37", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1517907577706763, "全帯域_10ms_rms_envelope_nrmse": 0.073248064698933, "全帯域_10ms_envelope_correlation": 0.9951823404750403, "全帯域_10ms_modulation_mae_db": 0.440126597089809, "全帯域_40ms_log_rms_mae_db": 0.6411578852469176, "全帯域_40ms_rms_envelope_nrmse": 0.0386748129105127, "全帯域_40ms_envelope_correlation": 0.9986893485739035, "全帯域_40ms_modulation_mae_db": 0.9067113877665699, "300–1000 Hz_10ms_log_rms_mae_db": 1.3957524393701477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10794650023904469, "300–1000 Hz_10ms_envelope_correlation": 0.9917522602521612, "300–1000 Hz_10ms_modulation_mae_db": 0.6304220209292737, "300–1000 Hz_40ms_log_rms_mae_db": 0.9481284733410728, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06353259002072684, "300–1000 Hz_40ms_envelope_correlation": 0.9970825380217583, "300–1000 Hz_40ms_modulation_mae_db": 0.5311974672029867, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4418942471423175, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14315528144819673, "1000–3000 Hz_10ms_envelope_correlation": 0.987326270335721, "1000–3000 Hz_10ms_modulation_mae_db": 0.39093400435500114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8554285001479415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08363549747008604, "1000–3000 Hz_40ms_envelope_correlation": 0.9961131548586558, "1000–3000 Hz_40ms_modulation_mae_db": 0.9499674233555296, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8871659549871487, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19514072110011868, "3000–6000 Hz_10ms_envelope_correlation": 0.978064713105579, "3000–6000 Hz_10ms_modulation_mae_db": 0.7028570692153563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1853596399188415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12258247791825526, "3000–6000 Hz_40ms_envelope_correlation": 0.99158103858218, "3000–6000 Hz_40ms_modulation_mae_db": 0.9648434801140473, "mr_stft_db": 5.674554506460542, "band_mae_db": 0.12009878225119974, "rms_gain_error_db": -0.11105861231556219, "output_peak": 0.6700941159509076}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1324995047831203, "全帯域_10ms_rms_envelope_nrmse": 0.09298036499503316, "全帯域_10ms_envelope_correlation": 0.9921706350619515, "全帯域_10ms_modulation_mae_db": 0.547203808238597, "全帯域_40ms_log_rms_mae_db": 0.6132016243115619, "全帯域_40ms_rms_envelope_nrmse": 0.050662134436654606, "全帯域_40ms_envelope_correlation": 0.997683850151602, "全帯域_40ms_modulation_mae_db": 1.1636233176220288, "300–1000 Hz_10ms_log_rms_mae_db": 1.3411909134526496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13123297005034631, "300–1000 Hz_10ms_envelope_correlation": 0.9877703574335062, "300–1000 Hz_10ms_modulation_mae_db": 0.5910052919414603, "300–1000 Hz_40ms_log_rms_mae_db": 0.8654509249280269, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07474880513985287, "300–1000 Hz_40ms_envelope_correlation": 0.9959608294696036, "300–1000 Hz_40ms_modulation_mae_db": 1.060324281947669, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4294430854746256, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17043553804582054, "1000–3000 Hz_10ms_envelope_correlation": 0.9817826386102758, "1000–3000 Hz_10ms_modulation_mae_db": 1.1265430395258316, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8769543925716122, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09483284577386752, "1000–3000 Hz_40ms_envelope_correlation": 0.9949703322724217, "1000–3000 Hz_40ms_modulation_mae_db": 0.42918651177805095, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8543003588242644, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20220483422481075, "3000–6000 Hz_10ms_envelope_correlation": 0.9742791984520399, "3000–6000 Hz_10ms_modulation_mae_db": 0.5368305917300003, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2340256267762832, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1257160169921495, "3000–6000 Hz_40ms_envelope_correlation": 0.9895063539167922, "3000–6000 Hz_40ms_modulation_mae_db": 0.48223838208268416, "mr_stft_db": 5.667206379270745, "band_mae_db": 0.18339330849830246, "rms_gain_error_db": -0.12706263090305542, "output_peak": 0.6444215532602156}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1074474880986356, "全帯域_10ms_rms_envelope_nrmse": 0.0790366063520544, "全帯域_10ms_envelope_correlation": 0.9943116383987871, "全帯域_10ms_modulation_mae_db": 0.5050947548790057, "全帯域_40ms_log_rms_mae_db": 0.6314315665215244, "全帯域_40ms_rms_envelope_nrmse": 0.04325835190359285, "全帯域_40ms_envelope_correlation": 0.9982306690240114, "全帯域_40ms_modulation_mae_db": 0.6156803461720528, "300–1000 Hz_10ms_log_rms_mae_db": 1.3599236093855287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11634727869473493, "300–1000 Hz_10ms_envelope_correlation": 0.9903875354487647, "300–1000 Hz_10ms_modulation_mae_db": 0.40044212235541554, "300–1000 Hz_40ms_log_rms_mae_db": 0.865787751116424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07487960298154739, "300–1000 Hz_40ms_envelope_correlation": 0.9957896420740534, "300–1000 Hz_40ms_modulation_mae_db": 0.3397638695130942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4544821567751536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1431884787319541, "1000–3000 Hz_10ms_envelope_correlation": 0.987318968851316, "1000–3000 Hz_10ms_modulation_mae_db": 1.1578853155880797, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8967182609264346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08129994181114973, "1000–3000 Hz_40ms_envelope_correlation": 0.9963921245153428, "1000–3000 Hz_40ms_modulation_mae_db": 0.6887868446437347, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9237800271781582, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1898551017511495, "3000–6000 Hz_10ms_envelope_correlation": 0.97712685582594, "3000–6000 Hz_10ms_modulation_mae_db": 0.76457009520021, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2365510515321891, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11530493372206302, "3000–6000 Hz_40ms_envelope_correlation": 0.9911421092295645, "3000–6000 Hz_40ms_modulation_mae_db": 0.9545585095739733, "mr_stft_db": 5.693632591032433, "band_mae_db": 0.11509207539543724, "rms_gain_error_db": -0.05189720845533038, "output_peak": 0.800810371493647}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1243270160942196, "全帯域_10ms_rms_envelope_nrmse": 0.08631038952795707, "全帯域_10ms_envelope_correlation": 0.993214942826415, "全帯域_10ms_modulation_mae_db": 0.49191918456543965, "全帯域_40ms_log_rms_mae_db": 0.6269979369852497, "全帯域_40ms_rms_envelope_nrmse": 0.047141265686412887, "全帯域_40ms_envelope_correlation": 0.9979086303706357, "全帯域_40ms_modulation_mae_db": 0.5389343274784464, "300–1000 Hz_10ms_log_rms_mae_db": 1.324489381628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13634567453664845, "300–1000 Hz_10ms_envelope_correlation": 0.9867317455708106, "300–1000 Hz_10ms_modulation_mae_db": 0.6694484689838665, "300–1000 Hz_40ms_log_rms_mae_db": 0.9055994042509706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07767739784939588, "300–1000 Hz_40ms_envelope_correlation": 0.9954577813257726, "300–1000 Hz_40ms_modulation_mae_db": 1.2260823292256895, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4030167176657624, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1408599385247344, "1000–3000 Hz_10ms_envelope_correlation": 0.9873360087378673, "1000–3000 Hz_10ms_modulation_mae_db": 1.3374989392692953, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8676324477792196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08521888614429271, "1000–3000 Hz_40ms_envelope_correlation": 0.9953477045603217, "1000–3000 Hz_40ms_modulation_mae_db": 0.8532752804962201, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8283943139464298, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1952053471529709, "3000–6000 Hz_10ms_envelope_correlation": 0.9757460950908818, "3000–6000 Hz_10ms_modulation_mae_db": 0.6837472623490823, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2386836929760998, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11716729222094137, "3000–6000 Hz_40ms_envelope_correlation": 0.9909711502537388, "3000–6000 Hz_40ms_modulation_mae_db": 1.117874723976183, "mr_stft_db": 5.662466689826439, "band_mae_db": 0.11709157114738622, "rms_gain_error_db": -0.0708228138532531, "output_peak": 0.796998861829532}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1201656179555426, "全帯域_10ms_rms_envelope_nrmse": 0.0748398928176494, "全帯域_10ms_envelope_correlation": 0.9949376114683945, "全帯域_10ms_modulation_mae_db": 0.5298549386361702, "全帯域_40ms_log_rms_mae_db": 0.5872974614409625, "全帯域_40ms_rms_envelope_nrmse": 0.03958944915944326, "全帯域_40ms_envelope_correlation": 0.9985862085235854, "全帯域_40ms_modulation_mae_db": 0.6020236695605785, "300–1000 Hz_10ms_log_rms_mae_db": 1.2959642718423054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12797056221919226, "300–1000 Hz_10ms_envelope_correlation": 0.988314416191858, "300–1000 Hz_10ms_modulation_mae_db": 1.0259759534550528, "300–1000 Hz_40ms_log_rms_mae_db": 0.8143723371829683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07196173600116114, "300–1000 Hz_40ms_envelope_correlation": 0.9961244242834816, "300–1000 Hz_40ms_modulation_mae_db": 0.40449520021138813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4435130203812148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15727506851976758, "1000–3000 Hz_10ms_envelope_correlation": 0.9848726247249191, "1000–3000 Hz_10ms_modulation_mae_db": 0.9367728103951637, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8604933121045165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09227420940406428, "1000–3000 Hz_40ms_envelope_correlation": 0.9955789120116312, "1000–3000 Hz_40ms_modulation_mae_db": 0.5307922333222885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9537897218277618, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20910176998548263, "3000–6000 Hz_10ms_envelope_correlation": 0.9721533358392856, "3000–6000 Hz_10ms_modulation_mae_db": 0.35804077208816776, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2746601027482587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1271796341033861, "3000–6000 Hz_40ms_envelope_correlation": 0.989325230228032, "3000–6000 Hz_40ms_modulation_mae_db": 0.5627577988101954, "mr_stft_db": 5.690286280258468, "band_mae_db": 0.19729911399277353, "rms_gain_error_db": -0.0976577790936621, "output_peak": 0.6768136336275622}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1667551087765609, "全帯域_10ms_rms_envelope_nrmse": 0.07612145476366365, "全帯域_10ms_envelope_correlation": 0.9947028854430209, "全帯域_10ms_modulation_mae_db": 0.27432335731922225, "全帯域_40ms_log_rms_mae_db": 0.6028448027112879, "全帯域_40ms_rms_envelope_nrmse": 0.042634564620460995, "全帯域_40ms_envelope_correlation": 0.9982531146609485, "全帯域_40ms_modulation_mae_db": 0.5118843103025849, "300–1000 Hz_10ms_log_rms_mae_db": 1.3640482645487737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1133701868976212, "300–1000 Hz_10ms_envelope_correlation": 0.99087458995403, "300–1000 Hz_10ms_modulation_mae_db": 0.4103187052654112, "300–1000 Hz_40ms_log_rms_mae_db": 0.8826787835620212, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06660299697890645, "300–1000 Hz_40ms_envelope_correlation": 0.9966644585571601, "300–1000 Hz_40ms_modulation_mae_db": 0.29619767249948065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3845669423572076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15191477777804052, "1000–3000 Hz_10ms_envelope_correlation": 0.985276254816812, "1000–3000 Hz_10ms_modulation_mae_db": 0.20509974294804664, "1000–3000 Hz_40ms_log_rms_mae_db": 0.881200437085686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09148794201708448, "1000–3000 Hz_40ms_envelope_correlation": 0.9946835010216641, "1000–3000 Hz_40ms_modulation_mae_db": 1.0184915948736777, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9388123857547281, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20102077028181417, "3000–6000 Hz_10ms_envelope_correlation": 0.9751834567819739, "3000–6000 Hz_10ms_modulation_mae_db": 0.44728925365475874, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2460614039850069, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12318099705947821, "3000–6000 Hz_40ms_envelope_correlation": 0.9903073738465771, "3000–6000 Hz_40ms_modulation_mae_db": 0.21508987678485955, "mr_stft_db": 5.686288912767959, "band_mae_db": 0.09803952477223872, "rms_gain_error_db": -0.041467350081293326, "output_peak": 0.7056957805896491}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1582770945853662, "全帯域_10ms_rms_envelope_nrmse": 0.07243705813923416, "全帯域_10ms_envelope_correlation": 0.995466438821193, "全帯域_10ms_modulation_mae_db": 0.5080612989641514, "全帯域_40ms_log_rms_mae_db": 0.6047186555912888, "全帯域_40ms_rms_envelope_nrmse": 0.04324820718383222, "全帯域_40ms_envelope_correlation": 0.9985249603354512, "全帯域_40ms_modulation_mae_db": 0.5556476624955816, "300–1000 Hz_10ms_log_rms_mae_db": 1.3466468749009515, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10873763622758623, "300–1000 Hz_10ms_envelope_correlation": 0.9919605637206056, "300–1000 Hz_10ms_modulation_mae_db": 0.5092172707601492, "300–1000 Hz_40ms_log_rms_mae_db": 0.9258576340711445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06984896142190204, "300–1000 Hz_40ms_envelope_correlation": 0.996830419144969, "300–1000 Hz_40ms_modulation_mae_db": 0.6936492792785649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.410100671612587, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13391221781617446, "1000–3000 Hz_10ms_envelope_correlation": 0.988784076094417, "1000–3000 Hz_10ms_modulation_mae_db": 0.2875871421454229, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8558226103533789, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07850059129249161, "1000–3000 Hz_40ms_envelope_correlation": 0.9963847957642714, "1000–3000 Hz_40ms_modulation_mae_db": 0.5397880108512026, "3000–6000 Hz_10ms_log_rms_mae_db": 1.922180582976039, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19333991422130353, "3000–6000 Hz_10ms_envelope_correlation": 0.9761938118768403, "3000–6000 Hz_10ms_modulation_mae_db": 1.0521829644934846, "3000–6000 Hz_40ms_log_rms_mae_db": 1.246249408459198, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11674134964067735, "3000–6000 Hz_40ms_envelope_correlation": 0.9909869725659549, "3000–6000 Hz_40ms_modulation_mae_db": 1.3156326702335452, "mr_stft_db": 5.68048920399291, "band_mae_db": 0.15558137335136554, "rms_gain_error_db": -0.16801647315309892, "output_peak": 0.6822889555408858}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1539736541629995, "全帯域_10ms_rms_envelope_nrmse": 0.08002599676854497, "全帯域_10ms_envelope_correlation": 0.9941964589370379, "全帯域_10ms_modulation_mae_db": 0.7208827768286459, "全帯域_40ms_log_rms_mae_db": 0.6137394291025627, "全帯域_40ms_rms_envelope_nrmse": 0.05148278694375228, "全帯域_40ms_envelope_correlation": 0.9975178045280559, "全帯域_40ms_modulation_mae_db": 0.8793510147373719, "300–1000 Hz_10ms_log_rms_mae_db": 1.345732768298757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11362577743784044, "300–1000 Hz_10ms_envelope_correlation": 0.9908101472081264, "300–1000 Hz_10ms_modulation_mae_db": 0.4450327668299077, "300–1000 Hz_40ms_log_rms_mae_db": 0.8893665398664179, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07468382263379411, "300–1000 Hz_40ms_envelope_correlation": 0.9958125441641981, "300–1000 Hz_40ms_modulation_mae_db": 0.33656644967976646, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3981618667128068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13239643041899746, "1000–3000 Hz_10ms_envelope_correlation": 0.9897749018939473, "1000–3000 Hz_10ms_modulation_mae_db": 0.8157849454047582, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8976316100909217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08119653368437874, "1000–3000 Hz_40ms_envelope_correlation": 0.9969797176562663, "1000–3000 Hz_40ms_modulation_mae_db": 0.9999283667282417, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8729197953372068, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2018342046963904, "3000–6000 Hz_10ms_envelope_correlation": 0.9741014203867274, "3000–6000 Hz_10ms_modulation_mae_db": 0.3254225157824336, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1781923051293164, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12157317692797631, "3000–6000 Hz_40ms_envelope_correlation": 0.9902147434183498, "3000–6000 Hz_40ms_modulation_mae_db": 0.48768597307932965, "mr_stft_db": 5.675809696625986, "band_mae_db": 0.15477991090377793, "rms_gain_error_db": -0.090551749958999, "output_peak": 0.7444195807837876}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.788349573504722, "spectral_convergence": 0.2425575034651658, "spectral_convergence_status": "ok", "amplitude_floor": 1.8779397470586322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.546395858819146, "spectral_convergence": 0.2435305096982845, "spectral_convergence_status": "ok", "amplitude_floor": 1.7710051963816244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.688918087057758, "spectral_convergence": 0.25028572194715726, "spectral_convergence_status": "ok", "amplitude_floor": 1.1208016909489659e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004732055872254276, "output_power": 0.00047577927246106555, "delta_db": 0.023556540970626802, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016454515092618504, "output_power": 0.0016203572923235191, "delta_db": -0.06674300365690158, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001811442607675158, "output_power": 0.0017539946064316077, "delta_db": -0.1399632517538455, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004356719801952226, "output_power": 0.00040180019955479636, "delta_db": -0.3514948151611333, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011106135183657726, "output_power": 0.00010931105485991721, "delta_db": -0.0689887009536429, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.173475364573245e-05, "output_power": 6.27356435677574e-05, "delta_db": 0.06984638101104836, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12009878225119974, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11105861231556219, "pitch": null, "output_rms_dbfs": -23.541870346078817, "output_peak": 0.6700941159509076, "reference_rms_dbfs": -23.430811733763257, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.674554506460542}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027737662806153396, "log_rms_mae_db": 1.1517907577706763, "rms_envelope_nrmse": 0.073248064698933, "envelope_correlation": 0.9951823404750403, "rms_gain_error_db": -0.11045350119937128, "reference_mean_power": 0.004541375985905238, "output_mean_power": 0.004427332189956933}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.847489317868867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007100074703458764, "output_power": 0.00067697949156837, "reference_active": true, "delta_db": -0.20687405816077378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018446902300009587, "output_power": 0.000147992628041175, "reference_active": true, "delta_db": -0.956833652077721}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9341715429184444e-05, "output_power": 2.83020759881662e-05, "reference_active": true, "delta_db": -0.1566720810309324}], "active_band_count": 3, "active_band_mae_db": 0.440126597089809}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002314838901045675, "log_rms_mae_db": 0.6411578852469176, "rms_envelope_nrmse": 0.0386748129105127, "envelope_correlation": 0.9986893485739035, "rms_gain_error_db": -0.11028596850074221, "reference_mean_power": 0.0045632674272876395, "output_mean_power": 0.004448845504384542}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7204098415154255e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006818437378247786, "output_power": 0.0006436343047261464, "reference_active": true, "delta_db": -0.2504567293934056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.591872232322897e-05, "output_power": 6.197963022428117e-05, "reference_active": true, "delta_db": -0.8809990963332669}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0593810422650415e-06, "output_power": 1.4284626084711638e-06, "reference_active": true, "delta_db": -1.5886783375730371}], "active_band_count": 3, "active_band_mae_db": 0.9067113877665699}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002609159645197948, "log_rms_mae_db": 1.3957524393701477, "rms_envelope_nrmse": 0.10794650023904469, "envelope_correlation": 0.9917522602521612, "rms_gain_error_db": -0.12743819542037052, "reference_mean_power": 0.0022069594688974205, "output_mean_power": 0.0021431399787186067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0859718062152698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048225083122535774, "output_power": 0.00044086061068092245, "reference_active": true, "delta_db": -0.3897168698188124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016124782840735576, "output_power": 0.00013636069425280501, "reference_active": true, "delta_db": -0.7280467083466855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2038625150435634e-05, "output_power": 1.84431139568001e-05, "reference_active": true, "delta_db": -0.7735024846223232}], "active_band_count": 3, "active_band_mae_db": 0.6304220209292737}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001981315811908706, "log_rms_mae_db": 0.9481284733410728, "rms_envelope_nrmse": 0.06353259002072684, "envelope_correlation": 0.9970825380217583, "rms_gain_error_db": -0.12751927596347606, "reference_mean_power": 0.002217812362061406, "output_mean_power": 0.0021536388269213086}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0257217759681144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004916402889453352, "output_power": 0.00044410160629976503, "reference_active": true, "delta_db": -0.44165121238813504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.925750439584527e-05, "output_power": 6.132330924474819e-05, "reference_active": true, "delta_db": -0.5284125522076061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6222165494821603e-06, "output_power": 1.4052576686292456e-06, "reference_active": true, "delta_db": -0.6235286370132188}], "active_band_count": 3, "active_band_mae_db": 0.5311974672029867}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001406989062134312, "log_rms_mae_db": 1.4418942471423175, "rms_envelope_nrmse": 0.14315528144819673, "envelope_correlation": 0.987326270335721, "rms_gain_error_db": -0.32598442696933455, "reference_mean_power": 0.0005036178738924359, "output_mean_power": 0.0004671998477200621}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.692987420204184e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020569637799863847, "output_power": 0.0001946070762650591, "reference_active": true, "delta_db": -0.2406801654282838}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.72045822558081e-05, "output_power": 5.666212001497427e-05, "reference_active": true, "delta_db": -0.7410606622997896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.876518957296052e-06, "output_power": 7.1857940049198496e-06, "reference_active": true, "delta_db": 0.19106118533692995}], "active_band_count": 3, "active_band_mae_db": 0.39093400435500114}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012225910578362228, "log_rms_mae_db": 0.8554285001479415, "rms_envelope_nrmse": 0.08363549747008604, "envelope_correlation": 0.9961131548586558, "rms_gain_error_db": -0.32614380384191594, "reference_mean_power": 0.0005061264034494051, "output_mean_power": 0.0004695097481140555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5254708519662464e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002159112955909936, "output_power": 0.00020402575551009138, "reference_active": true, "delta_db": -0.24590368700516038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9510380576050556e-05, "output_power": 2.519008855935787e-05, "reference_active": true, "delta_db": -0.6874511602242699}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.547001199869126e-07, "output_power": 5.497421323299658e-07, "reference_active": true, "delta_db": -1.9165474228371584}], "active_band_count": 3, "active_band_mae_db": 0.9499674233555296}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3844040204199326e-05, "log_rms_mae_db": 1.8871659549871487, "rms_envelope_nrmse": 0.19514072110011868, "envelope_correlation": 0.978064713105579, "rms_gain_error_db": 0.36642168678923903, "reference_mean_power": 4.7417249737341345e-05, "output_mean_power": 5.159154447947812e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.272212516285404e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.356579821454215e-05, "output_power": 3.705844348768073e-05, "reference_active": true, "delta_db": 0.42990194898102563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1452306980819945e-06, "output_power": 6.101763946521862e-06, "reference_active": true, "delta_db": -0.6856085295079594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.623091261016208e-06, "output_power": 1.2913288877022457e-06, "reference_active": true, "delta_db": -0.9930607291570837}], "active_band_count": 3, "active_band_mae_db": 0.7028570692153563}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.685962479131491e-05, "log_rms_mae_db": 1.1853596399188415, "rms_envelope_nrmse": 0.12258247791825526, "envelope_correlation": 0.99158103858218, "rms_gain_error_db": 0.3659492183590374, "reference_mean_power": 4.7645706194829226e-05, "output_mean_power": 5.18344733293888e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.965193836309832e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1852605209893166e-05, "output_power": 3.503057014752418e-05, "reference_active": true, "delta_db": 0.4130224695829936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.594578726064393e-06, "output_power": 2.8085728026241908e-06, "reference_active": true, "delta_db": -1.0716231372555458}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8710953646003778e-07, "output_power": 1.3524070919378665e-07, "reference_active": true, "delta_db": -1.4098848335036025}], "active_band_count": 3, "active_band_mae_db": 0.9648434801140473}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9200835439706927, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "005e799b6359a7861f1c79af026b67654d9fc10b572cad9635aba003c61e10ef", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9865589680081595, "standard_deviation": 0.0025175359980307947, "minimum": 0.9817826386102758, "maximum": 0.9897749018939473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4206473385152094, "standard_deviation": 0.0251449882198944, "minimum": 1.3845669423572076, "maximum": 1.4544821567751536, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7822632424539498, "standard_deviation": 0.43488337231936086, "minimum": 0.20509974294804664, "maximum": 1.3374989392692953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1466422164104607, "standard_deviation": 0.012702624258974565, "minimum": 0.13239643041899746, "maximum": 0.17043553804582054, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9958062803325718, "standard_deviation": 0.0007901583269132774, "minimum": 0.9946835010216641, "maximum": 0.9969797176562663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8739851963824639, "standard_deviation": 0.01704380822568194, "minimum": 0.8554285001479415, "maximum": 0.8976316100909217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7512770332561183, "standard_deviation": 0.23424684856374156, "minimum": 0.42918651177805095, "maximum": 1.0184915948736777, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08605580594967688, "standard_deviation": 0.006036596000554703, "minimum": 0.07850059129249161, "maximum": 0.09483284577386752, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9753561109199085, "standard_deviation": 0.001864995608390822, "minimum": 0.9721533358392856, "maximum": 0.978064713105579, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8976678926039672, "standard_deviation": 0.04396033161766212, "minimum": 1.8283943139464298, "maximum": 1.9537897218277618, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6088675655641866, "standard_deviation": 0.24194594127615132, "minimum": 0.3254225157824336, "maximum": 1.0521829644934846, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19846283292675507, "standard_deviation": 0.00617695855793308, "minimum": 0.1898551017511495, "maximum": 0.20910176998548263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905043715051487, "standard_deviation": 0.0008040800731401735, "minimum": 0.989325230228032, "maximum": 0.99158103858218, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.229972903940649, "standard_deviation": 0.032354859083007365, "minimum": 1.1781923051293164, "maximum": 1.2746601027482587, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7625851768318522, "standard_deviation": 0.37866312886429593, "minimum": 0.21508987678485955, "maximum": 1.3156326702335452, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1211807348231159, "standard_deviation": 0.0043591680292455245, "minimum": 0.11530493372206302, "maximum": 0.1271796341033861, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9898252019724829, "standard_deviation": 0.001954306850579379, "minimum": 0.9867317455708106, "maximum": 0.9919605637206056, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.346718565428389, "standard_deviation": 0.0292298263837779, "minimum": 1.2959642718423054, "maximum": 1.3957524393701477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5852328250650671, "standard_deviation": 0.2047992531511854, "minimum": 0.40044212235541554, "maximum": 1.0259759534550528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11944707328787682, "standard_deviation": 0.010852912268248409, "minimum": 0.10794650023904469, "maximum": 0.13634567453664845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962153296301246, "standard_deviation": 0.0005760037044805322, "minimum": 0.9954577813257726, "maximum": 0.9970825380217583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8871552310398808, "standard_deviation": 0.041067408022083904, "minimum": 0.8143723371829683, "maximum": 0.9481284733410728, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.61103456869483, "standard_deviation": 0.3553761528051192, "minimum": 0.29619767249948065, "maximum": 1.2260823292256895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07174198912841084, "standard_deviation": 0.004783966389099017, "minimum": 0.06353259002072684, "maximum": 0.07767739784939588, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1426719575390602, "standard_deviation": 0.035601401581009655, "minimum": 0.09803952477223872, "maximum": 0.19729911399277353, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.678841782529435, "standard_deviation": 0.010948051631950107, "minimum": 5.662466689826439, "maximum": 5.693632591032433, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7151928566345234, "standard_deviation": 0.059239323875833895, "minimum": 0.6444215532602156, "maximum": 0.800810371493647, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09481682722678181, "standard_deviation": 0.04132106097016385, "minimum": -0.16801647315309892, "maximum": -0.041467350081293326, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9942728689289799, "standard_deviation": 0.0010970402294729017, "minimum": 0.9921706350619515, "maximum": 0.995466438821193, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1394045302783902, "standard_deviation": 0.021167039516563492, "minimum": 1.1074474880986356, "maximum": 1.1667551087765609, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5021833395651301, "standard_deviation": 0.12337028715765937, "minimum": 0.27432335731922225, "maximum": 0.7208827768286459, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07937497850788372, "standard_deviation": 0.007076833622154794, "minimum": 0.07243705813923416, "maximum": 0.09298036499503316, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981743232710243, "standard_deviation": 0.00043250157058350255, "minimum": 0.9975178045280559, "maximum": 0.9986893485739035, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6151736702389194, "standard_deviation": 0.017417762857523052, "minimum": 0.5872974614409625, "maximum": 0.6411578852469176, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7217320045169018, "standard_deviation": 0.23450286354211577, "minimum": 0.5118843103025849, "maximum": 1.1636233176220288, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.044586446605582725, "standard_deviation": 0.004755185939115141, "minimum": 0.0386748129105127, "maximum": 0.05148278694375228, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0054", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.4345510005950928, "predicted_mos": 2.4345510005950928, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f56212e05691553b4c7312abc572ae296d3790ee27f677f5a8be09e6db65350d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.510559199843556, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1527023895586184, "全帯域_10ms_rms_envelope_nrmse": 0.07490026340308292, "全帯域_10ms_envelope_correlation": 0.9949369267300838, "全帯域_10ms_modulation_mae_db": 0.5039052870065724, "全帯域_40ms_log_rms_mae_db": 0.6429695606154887, "全帯域_40ms_rms_envelope_nrmse": 0.03953765307454789, "全帯域_40ms_envelope_correlation": 0.9985976201916004, "全帯域_40ms_modulation_mae_db": 0.9285031292740801, "300–1000 Hz_10ms_log_rms_mae_db": 1.40120622852765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11093602334011995, "300–1000 Hz_10ms_envelope_correlation": 0.9912375404195681, "300–1000 Hz_10ms_modulation_mae_db": 0.7114952120182502, "300–1000 Hz_40ms_log_rms_mae_db": 0.880777764494841, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06507435529601635, "300–1000 Hz_40ms_envelope_correlation": 0.9968584363194524, "300–1000 Hz_40ms_modulation_mae_db": 0.5330950039487461, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4485279636613346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15396207985561372, "1000–3000 Hz_10ms_envelope_correlation": 0.9855689582605863, "1000–3000 Hz_10ms_modulation_mae_db": 0.6801686299165133, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8250507400835009, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08627284578170194, "1000–3000 Hz_40ms_envelope_correlation": 0.9963136075881827, "1000–3000 Hz_40ms_modulation_mae_db": 1.21088856383456, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8857889986387255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19215493239441897, "3000–6000 Hz_10ms_envelope_correlation": 0.9781420539597443, "3000–6000 Hz_10ms_modulation_mae_db": 0.8542956338234431, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1757180192831527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12209159251022707, "3000–6000 Hz_40ms_envelope_correlation": 0.9911573181839793, "3000–6000 Hz_40ms_modulation_mae_db": 1.207845687382931, "mr_stft_db": 5.627426023363564, "band_mae_db": 0.12488728440939988, "rms_gain_error_db": -0.09934292892951292, "output_peak": 0.6776455206268582}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1558968204513635, "全帯域_10ms_rms_envelope_nrmse": 0.09509998039607376, "全帯域_10ms_envelope_correlation": 0.9917781979128262, "全帯域_10ms_modulation_mae_db": 0.5546246829978595, "全帯域_40ms_log_rms_mae_db": 0.6311912019318286, "全帯域_40ms_rms_envelope_nrmse": 0.05074026111106371, "全帯域_40ms_envelope_correlation": 0.997644505160243, "全帯域_40ms_modulation_mae_db": 1.1093732597137966, "300–1000 Hz_10ms_log_rms_mae_db": 1.3396077039081196, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13505820607145777, "300–1000 Hz_10ms_envelope_correlation": 0.9870018959351713, "300–1000 Hz_10ms_modulation_mae_db": 0.5828042633539362, "300–1000 Hz_40ms_log_rms_mae_db": 0.8391525405329676, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07457847272233989, "300–1000 Hz_40ms_envelope_correlation": 0.995914792852388, "300–1000 Hz_40ms_modulation_mae_db": 0.8924694163186521, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4608685030363089, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16897553175303867, "1000–3000 Hz_10ms_envelope_correlation": 0.9820054630116785, "1000–3000 Hz_10ms_modulation_mae_db": 0.9070010272615248, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8716282233440112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09345941552618978, "1000–3000 Hz_40ms_envelope_correlation": 0.9949834786689211, "1000–3000 Hz_40ms_modulation_mae_db": 0.5650025330618287, "3000–6000 Hz_10ms_log_rms_mae_db": 1.822879574263743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20474387629658491, "3000–6000 Hz_10ms_envelope_correlation": 0.9734588143767561, "3000–6000 Hz_10ms_modulation_mae_db": 0.590016833658443, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1975297368412932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1277387535533093, "3000–6000 Hz_40ms_envelope_correlation": 0.9890858716589187, "3000–6000 Hz_40ms_modulation_mae_db": 0.5941858219738637, "mr_stft_db": 5.617124009433297, "band_mae_db": 0.19531311840131335, "rms_gain_error_db": -0.11626443676239626, "output_peak": 0.6504568732256897}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1248261576991234, "全帯域_10ms_rms_envelope_nrmse": 0.08089979658064846, "全帯域_10ms_envelope_correlation": 0.9940473820713367, "全帯域_10ms_modulation_mae_db": 0.45312951685596126, "全帯域_40ms_log_rms_mae_db": 0.6506690821896219, "全帯域_40ms_rms_envelope_nrmse": 0.04750860197228195, "全帯域_40ms_envelope_correlation": 0.9978801550920597, "全帯域_40ms_modulation_mae_db": 0.7319093531222265, "300–1000 Hz_10ms_log_rms_mae_db": 1.401421405100286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1167376960844113, "300–1000 Hz_10ms_envelope_correlation": 0.9903071071653672, "300–1000 Hz_10ms_modulation_mae_db": 0.6592605479643344, "300–1000 Hz_40ms_log_rms_mae_db": 0.873104465528999, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0757263303558358, "300–1000 Hz_40ms_envelope_correlation": 0.9956858755686993, "300–1000 Hz_40ms_modulation_mae_db": 0.649225406320746, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4839218415404922, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14694356191352212, "1000–3000 Hz_10ms_envelope_correlation": 0.9876915458669037, "1000–3000 Hz_10ms_modulation_mae_db": 1.1032046101231059, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8927294402637684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09206729985375692, "1000–3000 Hz_40ms_envelope_correlation": 0.9964417293598582, "1000–3000 Hz_40ms_modulation_mae_db": 0.8069491563149226, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8584623903104616, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18715017590924646, "3000–6000 Hz_10ms_envelope_correlation": 0.9777141072893892, "3000–6000 Hz_10ms_modulation_mae_db": 0.9496250718121804, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1757580968641432, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11696387056032635, "3000–6000 Hz_40ms_envelope_correlation": 0.9910403781120565, "3000–6000 Hz_40ms_modulation_mae_db": 1.2572554777281943, "mr_stft_db": 5.632558036344165, "band_mae_db": 0.1810668518082479, "rms_gain_error_db": -0.07362531080506411, "output_peak": 0.78255306721124}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1323903364463768, "全帯域_10ms_rms_envelope_nrmse": 0.09016879210844557, "全帯域_10ms_envelope_correlation": 0.992584207285687, "全帯域_10ms_modulation_mae_db": 0.6141167512970909, "全帯域_40ms_log_rms_mae_db": 0.6256341243384987, "全帯域_40ms_rms_envelope_nrmse": 0.04997423133469121, "全帯域_40ms_envelope_correlation": 0.9976247925222592, "全帯域_40ms_modulation_mae_db": 0.8105471491365804, "300–1000 Hz_10ms_log_rms_mae_db": 1.2798552261178877, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1450926182234516, "300–1000 Hz_10ms_envelope_correlation": 0.9849924019186744, "300–1000 Hz_10ms_modulation_mae_db": 0.9775319424498078, "300–1000 Hz_40ms_log_rms_mae_db": 0.8491156031737439, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08363132240718123, "300–1000 Hz_40ms_envelope_correlation": 0.9947264447912227, "300–1000 Hz_40ms_modulation_mae_db": 1.650892969504518, "1000–3000 Hz_10ms_log_rms_mae_db": 1.436594037159068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13555107170474603, "1000–3000 Hz_10ms_envelope_correlation": 0.9883473347974857, "1000–3000 Hz_10ms_modulation_mae_db": 1.0287397252481707, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8524807426121543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08056812501223037, "1000–3000 Hz_40ms_envelope_correlation": 0.995954190875004, "1000–3000 Hz_40ms_modulation_mae_db": 0.610513385306435, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8079779579930384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19984901985174064, "3000–6000 Hz_10ms_envelope_correlation": 0.9745053877489886, "3000–6000 Hz_10ms_modulation_mae_db": 0.5109700129395297, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1966135527350958, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12128645203181332, "3000–6000 Hz_40ms_envelope_correlation": 0.9902825106164425, "3000–6000 Hz_40ms_modulation_mae_db": 0.8142470074527299, "mr_stft_db": 5.610860371974617, "band_mae_db": 0.12084378776395459, "rms_gain_error_db": -0.056638437591740504, "output_peak": 0.8075248169651521}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1402964259125188, "全帯域_10ms_rms_envelope_nrmse": 0.07436104715944425, "全帯域_10ms_envelope_correlation": 0.9950295095236484, "全帯域_10ms_modulation_mae_db": 0.749650750806118, "全帯域_40ms_log_rms_mae_db": 0.6019336363919353, "全帯域_40ms_rms_envelope_nrmse": 0.04204567284144507, "全帯域_40ms_envelope_correlation": 0.9984223153522197, "全帯域_40ms_modulation_mae_db": 0.4565824127393247, "300–1000 Hz_10ms_log_rms_mae_db": 1.2952121480631993, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12133505575942015, "300–1000 Hz_10ms_envelope_correlation": 0.9895037668374835, "300–1000 Hz_10ms_modulation_mae_db": 0.8979456780099664, "300–1000 Hz_40ms_log_rms_mae_db": 0.7724183309547218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06991585356586721, "300–1000 Hz_40ms_envelope_correlation": 0.9963376408106857, "300–1000 Hz_40ms_modulation_mae_db": 0.48712926727258327, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4743640618689846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1480079043102542, "1000–3000 Hz_10ms_envelope_correlation": 0.9872769963113678, "1000–3000 Hz_10ms_modulation_mae_db": 0.9216452223980384, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8455612634632266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09044616282256367, "1000–3000 Hz_40ms_envelope_correlation": 0.996426451710958, "1000–3000 Hz_40ms_modulation_mae_db": 0.6168127267485016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8995175370009425, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20892190593266047, "3000–6000 Hz_10ms_envelope_correlation": 0.9721155019425853, "3000–6000 Hz_10ms_modulation_mae_db": 0.3064325503726569, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2257097881858485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13057128347319175, "3000–6000 Hz_40ms_envelope_correlation": 0.9887710252736578, "3000–6000 Hz_40ms_modulation_mae_db": 0.7500080583237186, "mr_stft_db": 5.636101706861961, "band_mae_db": 0.22058615914307303, "rms_gain_error_db": -0.11121899310599197, "output_peak": 0.6948079606663599}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1874605204712794, "全帯域_10ms_rms_envelope_nrmse": 0.0774221510813533, "全帯域_10ms_envelope_correlation": 0.9945231367838875, "全帯域_10ms_modulation_mae_db": 0.2445225042549495, "全帯域_40ms_log_rms_mae_db": 0.6151389163896215, "全帯域_40ms_rms_envelope_nrmse": 0.04445891377094183, "全帯域_40ms_envelope_correlation": 0.9980949210517335, "全帯域_40ms_modulation_mae_db": 0.4942021483644421, "300–1000 Hz_10ms_log_rms_mae_db": 1.377377233258001, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11494118241152361, "300–1000 Hz_10ms_envelope_correlation": 0.990625193499103, "300–1000 Hz_10ms_modulation_mae_db": 0.4258420528454083, "300–1000 Hz_40ms_log_rms_mae_db": 0.8536705056781162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06725080480987622, "300–1000 Hz_40ms_envelope_correlation": 0.9966031959150936, "300–1000 Hz_40ms_modulation_mae_db": 0.332071928510421, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4137816006001724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15685083710456277, "1000–3000 Hz_10ms_envelope_correlation": 0.984216097449134, "1000–3000 Hz_10ms_modulation_mae_db": 0.24450743634115701, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8451980003466641, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09434000053628544, "1000–3000 Hz_40ms_envelope_correlation": 0.9941551307155986, "1000–3000 Hz_40ms_modulation_mae_db": 0.9493010979550931, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8728018985443475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19823685846357242, "3000–6000 Hz_10ms_envelope_correlation": 0.9756756035786066, "3000–6000 Hz_10ms_modulation_mae_db": 0.258287831205738, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1700246031863208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12245092704699992, "3000–6000 Hz_40ms_envelope_correlation": 0.9903110858831118, "3000–6000 Hz_40ms_modulation_mae_db": 0.34666449469461513, "mr_stft_db": 5.627571611524995, "band_mae_db": 0.08519808742946955, "rms_gain_error_db": -0.02949247917237182, "output_peak": 0.7108468038701996}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.177695657049788, "全帯域_10ms_rms_envelope_nrmse": 0.07496319800878917, "全帯域_10ms_envelope_correlation": 0.9950744323500929, "全帯域_10ms_modulation_mae_db": 0.7173087962313881, "全帯域_40ms_log_rms_mae_db": 0.6144290824450599, "全帯域_40ms_rms_envelope_nrmse": 0.04349380814190923, "全帯域_40ms_envelope_correlation": 0.9984544044721265, "全帯域_40ms_modulation_mae_db": 0.46982170581008714, "300–1000 Hz_10ms_log_rms_mae_db": 1.3463918541008517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10728080538118678, "300–1000 Hz_10ms_envelope_correlation": 0.9920607721471986, "300–1000 Hz_10ms_modulation_mae_db": 0.58434261127211, "300–1000 Hz_40ms_log_rms_mae_db": 0.8835337763587868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06708520804678442, "300–1000 Hz_40ms_envelope_correlation": 0.9969623425385765, "300–1000 Hz_40ms_modulation_mae_db": 0.594596537509921, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4320738986885155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15649258157203771, "1000–3000 Hz_10ms_envelope_correlation": 0.9844645012598787, "1000–3000 Hz_10ms_modulation_mae_db": 0.6690814579431809, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8595274757054309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09420828397949452, "1000–3000 Hz_40ms_envelope_correlation": 0.9945242193808955, "1000–3000 Hz_40ms_modulation_mae_db": 0.36745081008002867, "3000–6000 Hz_10ms_log_rms_mae_db": 1.878398665825371, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19529658839066913, "3000–6000 Hz_10ms_envelope_correlation": 0.9756863948962621, "3000–6000 Hz_10ms_modulation_mae_db": 1.1271688186430926, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1911871269584577, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11763730431967508, "3000–6000 Hz_40ms_envelope_correlation": 0.9909626835812897, "3000–6000 Hz_40ms_modulation_mae_db": 1.4344813072570488, "mr_stft_db": 5.627517926682947, "band_mae_db": 0.16750021026214626, "rms_gain_error_db": -0.15547409322028652, "output_peak": 0.6733116460179989}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.148213516551692, "全帯域_10ms_rms_envelope_nrmse": 0.07665746907236305, "全帯域_10ms_envelope_correlation": 0.994688871281625, "全帯域_10ms_modulation_mae_db": 0.7374763679584064, "全帯域_40ms_log_rms_mae_db": 0.6169262522224466, "全帯域_40ms_rms_envelope_nrmse": 0.04838066803406858, "全帯域_40ms_envelope_correlation": 0.9978285351848953, "全帯域_40ms_modulation_mae_db": 0.8468495481745905, "300–1000 Hz_10ms_log_rms_mae_db": 1.331404402860005, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11357564813022127, "300–1000 Hz_10ms_envelope_correlation": 0.9908111266250901, "300–1000 Hz_10ms_modulation_mae_db": 0.3561891414951485, "300–1000 Hz_40ms_log_rms_mae_db": 0.8473283791159063, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07502572620898668, "300–1000 Hz_40ms_envelope_correlation": 0.9957753857936221, "300–1000 Hz_40ms_modulation_mae_db": 0.3489620219802594, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4031931345433604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13927197911285655, "1000–3000 Hz_10ms_envelope_correlation": 0.9883203188348936, "1000–3000 Hz_10ms_modulation_mae_db": 0.9148587251517274, "1000–3000 Hz_40ms_log_rms_mae_db": 0.845428859398866, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07871848656008838, "1000–3000 Hz_40ms_envelope_correlation": 0.9970133953279502, "1000–3000 Hz_40ms_modulation_mae_db": 0.7400377407664681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8387090910435835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19807441607128218, "3000–6000 Hz_10ms_envelope_correlation": 0.9749693131152148, "3000–6000 Hz_10ms_modulation_mae_db": 0.6516233189222982, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1690484489879491, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12363285244826887, "3000–6000 Hz_40ms_envelope_correlation": 0.9898443293279315, "3000–6000 Hz_40ms_modulation_mae_db": 0.7043218656727465, "mr_stft_db": 5.62179927035739, "band_mae_db": 0.1570322781791169, "rms_gain_error_db": -0.09465442989585604, "output_peak": 0.738457536200197}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.750761698853653, "spectral_convergence": 0.24405604917264687, "spectral_convergence_status": "ok", "amplitude_floor": 1.8779397470586322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.504255579191722, "spectral_convergence": 0.24344251399363862, "spectral_convergence_status": "ok", "amplitude_floor": 1.7710051963816244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.627260792045316, "spectral_convergence": 0.24988692561927153, "spectral_convergence_status": "ok", "amplitude_floor": 1.1208016909489659e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004732055872254276, "output_power": 0.0004763151391367386, "delta_db": 0.02844521507730379, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016454515092618504, "output_power": 0.0016180009542842199, "delta_db": -0.07306314930256536, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001811442607675158, "output_power": 0.0017793702869575807, "delta_db": -0.0775824457244838, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004356719801952226, "output_power": 0.0003921438534036068, "delta_db": -0.4571421750178747, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011106135183657726, "output_power": 0.00010830649884987516, "delta_db": -0.10908438472140548, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.173475364573245e-05, "output_power": 6.17917298057833e-05, "delta_db": 0.004006336612766104, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12488728440939988, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09934292892951292, "pitch": null, "output_rms_dbfs": -23.530154662692766, "output_peak": 0.6776455206268582, "reference_rms_dbfs": -23.430811733763257, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.627426023363564}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027737662806153396, "log_rms_mae_db": 1.1527023895586184, "rms_envelope_nrmse": 0.07490026340308292, "envelope_correlation": 0.9949369267300838, "rms_gain_error_db": -0.09875320708476051, "reference_mean_power": 0.004541375985905238, "output_mean_power": 0.004439275912958061}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.847489317868867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007100074703458764, "output_power": 0.0006731661894781939, "reference_active": true, "delta_db": -0.23140623246519806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018446902300009587, "output_power": 0.00015177584130443458, "reference_active": true, "delta_db": -0.8472079874720695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9341715429184444e-05, "output_power": 2.6556775158390433e-05, "reference_active": true, "delta_db": -0.43310164108244975}], "active_band_count": 3, "active_band_mae_db": 0.5039052870065724}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002314838901045675, "log_rms_mae_db": 0.6429695606154887, "rms_envelope_nrmse": 0.03953765307454789, "envelope_correlation": 0.9985976201916004, "rms_gain_error_db": -0.0985945827067776, "reference_mean_power": 0.0045632674272876395, "output_mean_power": 0.004460838114228704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7204098415154255e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006818437378247786, "output_power": 0.0006407080381979817, "reference_active": true, "delta_db": -0.27024683616453243}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.591872232322897e-05, "output_power": 6.416673198701494e-05, "reference_active": true, "delta_db": -0.7303896927919817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0593810422650415e-06, "output_power": 1.3653670709900427e-06, "reference_active": true, "delta_db": -1.784872858865726}], "active_band_count": 3, "active_band_mae_db": 0.9285031292740801}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002609159645197948, "log_rms_mae_db": 1.40120622852765, "rms_envelope_nrmse": 0.11093602334011995, "envelope_correlation": 0.9912375404195681, "rms_gain_error_db": -0.07981514707516484, "reference_mean_power": 0.0022069594688974205, "output_mean_power": 0.0021667701432711395}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0859718062152698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048225083122535774, "output_power": 0.0004457290534330644, "reference_active": true, "delta_db": -0.34202041766958813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016124782840735576, "output_power": 0.00014132169641139795, "reference_active": true, "delta_db": -0.5728503249810031}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2038625150435634e-05, "output_power": 1.6642670267416013e-05, "reference_active": true, "delta_db": -1.2196148934041593}], "active_band_count": 3, "active_band_mae_db": 0.7114952120182502}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001981315811908706, "log_rms_mae_db": 0.880777764494841, "rms_envelope_nrmse": 0.06507435529601635, "envelope_correlation": 0.9968584363194524, "rms_gain_error_db": -0.0796629951064241, "reference_mean_power": 0.002217812362061406, "output_mean_power": 0.002177501688210578}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0257217759681144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004916402889453352, "output_power": 0.00044971292097988877, "reference_active": true, "delta_db": -0.3871209951268651}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.925750439584527e-05, "output_power": 6.362385741379962e-05, "reference_active": true, "delta_db": -0.36846842101397637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6222165494821603e-06, "output_power": 1.3357933223503564e-06, "reference_active": true, "delta_db": -0.843695595705397}], "active_band_count": 3, "active_band_mae_db": 0.5330950039487461}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001406989062134312, "log_rms_mae_db": 1.4485279636613346, "rms_envelope_nrmse": 0.15396207985561372, "envelope_correlation": 0.9855689582605863, "rms_gain_error_db": -0.41160312958554707, "reference_mean_power": 0.0005036178738924359, "output_mean_power": 0.00045807946376726365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.692987420204184e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020569637799863847, "output_power": 0.00018541533262743444, "reference_active": true, "delta_db": -0.4508099990647119}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.72045822558081e-05, "output_power": 5.268961820686907e-05, "reference_active": true, "delta_db": -1.0567383420776315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.876518957296052e-06, "output_power": 7.774355564648747e-06, "reference_active": true, "delta_db": 0.5329575486071964}], "active_band_count": 3, "active_band_mae_db": 0.6801686299165133}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012225910578362228, "log_rms_mae_db": 0.8250507400835009, "rms_envelope_nrmse": 0.08627284578170194, "envelope_correlation": 0.9963136075881827, "rms_gain_error_db": -0.41164476142636414, "reference_mean_power": 0.0005061264034494051, "output_mean_power": 0.00046035675265579525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5254708519662464e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002159112955909936, "output_power": 0.0001948009278085743, "reference_active": true, "delta_db": -0.4468434240054641}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9510380576050556e-05, "output_power": 2.3637739443623324e-05, "reference_active": true, "delta_db": -0.9636886919176881}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.547001199869126e-07, "output_power": 5.123897192874437e-07, "reference_active": true, "delta_db": -2.222133575580527}], "active_band_count": 3, "active_band_mae_db": 1.21088856383456}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3844040204199326e-05, "log_rms_mae_db": 1.8857889986387255, "rms_envelope_nrmse": 0.19215493239441897, "envelope_correlation": 0.9781420539597443, "rms_gain_error_db": 0.2833001408090404, "reference_mean_power": 4.7417249737341345e-05, "output_mean_power": 5.0613500505479053e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.272212516285404e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.356579821454215e-05, "output_power": 3.704019556675302e-05, "reference_active": true, "delta_db": 0.42776291603077254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1452306980819945e-06, "output_power": 6.009754264792125e-06, "reference_active": true, "delta_db": -0.7515954109266092}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.623091261016208e-06, "output_power": 1.1802936809144459e-06, "reference_active": true, "delta_db": -1.3835285745129475}], "active_band_count": 3, "active_band_mae_db": 0.8542956338234431}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.685962479131491e-05, "log_rms_mae_db": 1.1757180192831527, "rms_envelope_nrmse": 0.12209159251022707, "envelope_correlation": 0.9911573181839793, "rms_gain_error_db": 0.2832911199788581, "reference_mean_power": 4.7645706194829226e-05, "output_mean_power": 5.085725087229047e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.965193836309832e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1852605209893166e-05, "output_power": 3.5019203157156426e-05, "reference_active": true, "delta_db": 0.41161300856606137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.594578726064393e-06, "output_power": 2.8095185959847933e-06, "reference_active": true, "delta_db": -1.0701608867784085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8710953646003778e-07, "output_power": 1.1426667920409385e-07, "reference_active": true, "delta_db": -2.1417631668043233}], "active_band_count": 3, "active_band_mae_db": 1.207845687382931}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9219205579000619, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a3307b940edc0ab2eaee7d08dbd393969ae64e089ec01e09f855ab513ac5a1df", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.985986401973991, "standard_deviation": 0.0023011368407727236, "minimum": 0.9820054630116785, "maximum": 0.9883473347974857, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4441656301372796, "standard_deviation": 0.02827414562763724, "minimum": 1.4031931345433604, "maximum": 1.4839218415404922, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8086508542979273, "standard_deviation": 0.2733827030183286, "minimum": 0.24450743634115701, "maximum": 1.1032046101231059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15075694341582896, "standard_deviation": 0.010673364578422542, "minimum": 0.13555107170474603, "maximum": 0.16897553175303867, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995726525453421, "standard_deviation": 0.001036538489598067, "minimum": 0.9941551307155986, "maximum": 0.9970133953279502, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8547005931522028, "standard_deviation": 0.0203535530739136, "minimum": 0.8250507400835009, "maximum": 0.8927294402637684, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7333695017584798, "standard_deviation": 0.25940381428970194, "minimum": 0.36745081008002867, "maximum": 1.21088856383456, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08876007750903889, "standard_deviation": 0.006219908497461084, "minimum": 0.07871848656008838, "maximum": 0.09434000053628544, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9752833971134434, "standard_deviation": 0.00201654277537104, "minimum": 0.9721155019425853, "maximum": 0.9781420539597443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8580670142025266, "standard_deviation": 0.03215831847131904, "minimum": 1.8079779579930384, "maximum": 1.8995175370009425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6560525089221727, "standard_deviation": 0.3055009648253775, "minimum": 0.258287831205738, "maximum": 1.1271688186430926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1980534716637719, "standard_deviation": 0.006839146453574107, "minimum": 0.18715017590924646, "maximum": 0.20892190593266047, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9901819003296735, "standard_deviation": 0.0008968722928990489, "minimum": 0.9887710252736578, "maximum": 0.9911573181839793, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1876986716302826, "standard_deviation": 0.01919619049652671, "minimum": 1.1690484489879491, "maximum": 1.2257097881858485, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.888626215060731, "standard_deviation": 0.3731869296075379, "minimum": 0.34666449469461513, "maximum": 1.4344813072570488, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12279662949297646, "standard_deviation": 0.0046152857808657455, "minimum": 0.11696387056032635, "maximum": 0.13057128347319175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.989567475568457, "standard_deviation": 0.0023825463888504836, "minimum": 0.9849924019186744, "maximum": 0.9920607721471986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3465595252420002, "standard_deviation": 0.04521357176398392, "minimum": 1.2798552261178877, "maximum": 1.401421405100286, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6494264311761202, "standard_deviation": 0.2131454182124966, "minimum": 0.3561891414951485, "maximum": 0.9775319424498078, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12061965442522406, "standard_deviation": 0.012964455827076452, "minimum": 0.10728080538118678, "maximum": 0.1450926182234516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961080143237175, "standard_deviation": 0.0007392906619537679, "minimum": 0.9947264447912227, "maximum": 0.9969623425385765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8498876707297602, "standard_deviation": 0.03538255301539325, "minimum": 0.7724183309547218, "maximum": 0.8835337763587868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6860553189207309, "standard_deviation": 0.4284764014600601, "minimum": 0.332071928510421, "maximum": 1.650892969504518, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07228600917661097, "standard_deviation": 0.006133521306970018, "minimum": 0.06507435529601635, "maximum": 0.08363132240718123, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15655347217459017, "standard_deviation": 0.04422827348370766, "minimum": 0.08519808742946955, "maximum": 0.22058615914307303, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.625119869567867, "standard_deviation": 0.008205403857450482, "minimum": 5.610860371974617, "maximum": 5.636101706861961, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7169505280979619, "standard_deviation": 0.05525025773836832, "minimum": 0.6504568732256897, "maximum": 0.8075248169651521, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09208888868540252, "standard_deviation": 0.03880952377642625, "minimum": -0.15547409322028652, "maximum": -0.02949247917237182, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940828329923985, "standard_deviation": 0.0012380035954930929, "minimum": 0.9917781979128262, "maximum": 0.9950744323500929, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1524352280175951, "standard_deviation": 0.021401422482839282, "minimum": 1.1248261576991234, "maximum": 1.1874605204712794, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5718418321760432, "standard_deviation": 0.1724367287861029, "minimum": 0.2445225042549495, "maximum": 0.749650750806118, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08055908722627506, "standard_deviation": 0.00784415197478498, "minimum": 0.07436104715944425, "maximum": 0.09509998039607376, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980684061283922, "standard_deviation": 0.00038256252719399996, "minimum": 0.9976247925222592, "maximum": 0.9985976201916004, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6248614820655627, "standard_deviation": 0.01615378897464959, "minimum": 0.6019336363919353, "maximum": 0.6506690821896219, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7309735882918911, "standard_deviation": 0.23949823558047512, "minimum": 0.4565824127393247, "maximum": 1.1093732597137966, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04576747628511868, "standard_deviation": 0.003996964697202731, "minimum": 0.03953765307454789, "maximum": 0.05074026111106371, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0054", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.1364400386810303, "predicted_mos": 2.1364400386810303, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "759b9cfc7a0e64907e471b8481fee1babcd712dbbff6fabe71a600bdbf738ee5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.97688930016011, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0940875441526603, "全帯域_10ms_rms_envelope_nrmse": 0.08520379009010708, "全帯域_10ms_envelope_correlation": 0.9938859095959184, "全帯域_10ms_modulation_mae_db": 0.5485664516662129, "全帯域_40ms_log_rms_mae_db": 0.5512479871105689, "全帯域_40ms_rms_envelope_nrmse": 0.050106569928542476, "全帯域_40ms_envelope_correlation": 0.9982455692626523, "全帯域_40ms_modulation_mae_db": 0.5298692150830626, "300–1000 Hz_10ms_log_rms_mae_db": 1.5420723251889101, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13730194469077306, "300–1000 Hz_10ms_envelope_correlation": 0.9875917606015671, "300–1000 Hz_10ms_modulation_mae_db": 1.2200247199835237, "300–1000 Hz_40ms_log_rms_mae_db": 1.085771881982542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08110540984369383, "300–1000 Hz_40ms_envelope_correlation": 0.9965123643615652, "300–1000 Hz_40ms_modulation_mae_db": 0.5827300947846298, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7208209627980675, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16914795638910982, "1000–3000 Hz_10ms_envelope_correlation": 0.9816116373704475, "1000–3000 Hz_10ms_modulation_mae_db": 0.6808154620708753, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0879924772252692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11101775535185437, "1000–3000 Hz_40ms_envelope_correlation": 0.9919074178413483, "1000–3000 Hz_40ms_modulation_mae_db": 0.4580844586997834, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5827743944090407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23114535786437393, "3000–6000 Hz_10ms_envelope_correlation": 0.9667980261777944, "3000–6000 Hz_10ms_modulation_mae_db": 0.7506901455898811, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6714772319012674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14082164380898365, "3000–6000 Hz_40ms_envelope_correlation": 0.987092972858754, "3000–6000 Hz_40ms_modulation_mae_db": 1.1508163231364537, "mr_stft_db": 6.0547334741306615, "band_mae_db": 0.20657085512459503, "rms_gain_error_db": -0.23450461409249157, "output_peak": 0.6211840625278036}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1377868083215839, "全帯域_10ms_rms_envelope_nrmse": 0.08165136858380465, "全帯域_10ms_envelope_correlation": 0.9940581206739018, "全帯域_10ms_modulation_mae_db": 0.5856418954190564, "全帯域_40ms_log_rms_mae_db": 0.6006993660382925, "全帯域_40ms_rms_envelope_nrmse": 0.04486493250515332, "全帯域_40ms_envelope_correlation": 0.9982866509946018, "全帯域_40ms_modulation_mae_db": 0.8916872353743993, "300–1000 Hz_10ms_log_rms_mae_db": 1.5159817231201493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13664447860731016, "300–1000 Hz_10ms_envelope_correlation": 0.9869316138465265, "300–1000 Hz_10ms_modulation_mae_db": 0.845881250423074, "300–1000 Hz_40ms_log_rms_mae_db": 1.0310329490494532, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07149263511313968, "300–1000 Hz_40ms_envelope_correlation": 0.9966682546233756, "300–1000 Hz_40ms_modulation_mae_db": 0.5092985845733691, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6826931304892705, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13177539036816593, "1000–3000 Hz_10ms_envelope_correlation": 0.9890210245595462, "1000–3000 Hz_10ms_modulation_mae_db": 0.7376460045222112, "1000–3000 Hz_40ms_log_rms_mae_db": 1.067789261359315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07288633978944416, "1000–3000 Hz_40ms_envelope_correlation": 0.9967793990153611, "1000–3000 Hz_40ms_modulation_mae_db": 0.9523213012284811, "3000–6000 Hz_10ms_log_rms_mae_db": 2.635426087050946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23797488242787046, "3000–6000 Hz_10ms_envelope_correlation": 0.9650248057646968, "3000–6000 Hz_10ms_modulation_mae_db": 0.3019222889295691, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7418076967572484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1342695694902007, "3000–6000 Hz_40ms_envelope_correlation": 0.9884428153799816, "3000–6000 Hz_40ms_modulation_mae_db": 0.7134069266644611, "mr_stft_db": 6.061840118253549, "band_mae_db": 0.1374409201218421, "rms_gain_error_db": -0.14545778720796254, "output_peak": 0.7062234511019084}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0516959843570957, "全帯域_10ms_rms_envelope_nrmse": 0.07787190681482169, "全帯域_10ms_envelope_correlation": 0.9945138529370843, "全帯域_10ms_modulation_mae_db": 0.39383381231929815, "全帯域_40ms_log_rms_mae_db": 0.5592683357916267, "全帯域_40ms_rms_envelope_nrmse": 0.03947057566112985, "全帯域_40ms_envelope_correlation": 0.998600304352738, "全帯域_40ms_modulation_mae_db": 0.5898941435284069, "300–1000 Hz_10ms_log_rms_mae_db": 1.5449957812177892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12088822597007369, "300–1000 Hz_10ms_envelope_correlation": 0.9896397858770913, "300–1000 Hz_10ms_modulation_mae_db": 0.37306874799693973, "300–1000 Hz_40ms_log_rms_mae_db": 1.068679632322995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06325700127276083, "300–1000 Hz_40ms_envelope_correlation": 0.9971385743823947, "300–1000 Hz_40ms_modulation_mae_db": 0.2660287485388729, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6999822905972461, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13279879844552328, "1000–3000 Hz_10ms_envelope_correlation": 0.9902650104130227, "1000–3000 Hz_10ms_modulation_mae_db": 1.5564321444646536, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0547096123377917, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08895724604676979, "1000–3000 Hz_40ms_envelope_correlation": 0.9966826059371428, "1000–3000 Hz_40ms_modulation_mae_db": 2.094289415365596, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5182541553416242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23630742023854034, "3000–6000 Hz_10ms_envelope_correlation": 0.9656770565637821, "3000–6000 Hz_10ms_modulation_mae_db": 0.3946656683895322, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6286807711752633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13768868386069674, "3000–6000 Hz_40ms_envelope_correlation": 0.9879078537782489, "3000–6000 Hz_40ms_modulation_mae_db": 0.8801845859030025, "mr_stft_db": 6.0527254255608804, "band_mae_db": 0.18979787452658367, "rms_gain_error_db": -0.09961382199593942, "output_peak": 0.6687971066844055}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1441843221973536, "全帯域_10ms_rms_envelope_nrmse": 0.07380745473893993, "全帯域_10ms_envelope_correlation": 0.9953332906637233, "全帯域_10ms_modulation_mae_db": 0.5856282930750026, "全帯域_40ms_log_rms_mae_db": 0.6020571635922098, "全帯域_40ms_rms_envelope_nrmse": 0.0419324624822435, "全帯域_40ms_envelope_correlation": 0.9987045181175167, "全帯域_40ms_modulation_mae_db": 0.7585112490704501, "300–1000 Hz_10ms_log_rms_mae_db": 1.5266706913475905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11781158068414023, "300–1000 Hz_10ms_envelope_correlation": 0.9903940599329177, "300–1000 Hz_10ms_modulation_mae_db": 0.6366357700414436, "300–1000 Hz_40ms_log_rms_mae_db": 1.063212547971934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06877045356164772, "300–1000 Hz_40ms_envelope_correlation": 0.9968648581813756, "300–1000 Hz_40ms_modulation_mae_db": 0.6593077483950235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.652320915440781, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15245217618658097, "1000–3000 Hz_10ms_envelope_correlation": 0.986543016306774, "1000–3000 Hz_10ms_modulation_mae_db": 1.167666314381454, "1000–3000 Hz_40ms_log_rms_mae_db": 1.027393126725667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10828013313921035, "1000–3000 Hz_40ms_envelope_correlation": 0.9940814455682591, "1000–3000 Hz_40ms_modulation_mae_db": 1.6748715165811843, "3000–6000 Hz_10ms_log_rms_mae_db": 2.54711598810655, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23013843971004336, "3000–6000 Hz_10ms_envelope_correlation": 0.9664747019879946, "3000–6000 Hz_10ms_modulation_mae_db": 0.8240261094715761, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6453244138794065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1345482804494305, "3000–6000 Hz_40ms_envelope_correlation": 0.9881760908920477, "3000–6000 Hz_40ms_modulation_mae_db": 0.8244890632479063, "mr_stft_db": 6.052032795874961, "band_mae_db": 0.2546249959730991, "rms_gain_error_db": -0.18171387123411498, "output_peak": 0.6300633820162203}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1027505408577334, "全帯域_10ms_rms_envelope_nrmse": 0.07164236785413622, "全帯域_10ms_envelope_correlation": 0.9953367536136554, "全帯域_10ms_modulation_mae_db": 0.3698995054593035, "全帯域_40ms_log_rms_mae_db": 0.5789301570637339, "全帯域_40ms_rms_envelope_nrmse": 0.036345675230674335, "全帯域_40ms_envelope_correlation": 0.9987762197186173, "全帯域_40ms_modulation_mae_db": 0.45535773749117875, "300–1000 Hz_10ms_log_rms_mae_db": 1.5243349823587304, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11007942293612406, "300–1000 Hz_10ms_envelope_correlation": 0.9913867414074791, "300–1000 Hz_10ms_modulation_mae_db": 0.7441170022981113, "300–1000 Hz_40ms_log_rms_mae_db": 1.090568299090097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05891588279044918, "300–1000 Hz_40ms_envelope_correlation": 0.997450418737854, "300–1000 Hz_40ms_modulation_mae_db": 0.663182448770944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.751678011506853, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1693850658615023, "1000–3000 Hz_10ms_envelope_correlation": 0.9815432179050018, "1000–3000 Hz_10ms_modulation_mae_db": 0.1530715720750477, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1102710443418704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09131561825447652, "1000–3000 Hz_40ms_envelope_correlation": 0.994596634595232, "1000–3000 Hz_40ms_modulation_mae_db": 0.3025760908411624, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5834017993631875, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23356752401937697, "3000–6000 Hz_10ms_envelope_correlation": 0.9654445587397414, "3000–6000 Hz_10ms_modulation_mae_db": 0.7963950704033284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6818057309349317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1330799605346357, "3000–6000 Hz_40ms_envelope_correlation": 0.988603795228283, "3000–6000 Hz_40ms_modulation_mae_db": 0.8155500323011483, "mr_stft_db": 6.056184390709004, "band_mae_db": 0.12425585881406266, "rms_gain_error_db": -0.0682764173403918, "output_peak": 0.7608202493189461}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1035669830189174, "全帯域_10ms_rms_envelope_nrmse": 0.07490409953719994, "全帯域_10ms_envelope_correlation": 0.9950466346301542, "全帯域_10ms_modulation_mae_db": 1.011486796940103, "全帯域_40ms_log_rms_mae_db": 0.5930699968173748, "全帯域_40ms_rms_envelope_nrmse": 0.043108770534598745, "全帯域_40ms_envelope_correlation": 0.9984443410084307, "全帯域_40ms_modulation_mae_db": 1.1562080083930912, "300–1000 Hz_10ms_log_rms_mae_db": 1.538616910673232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09989338708531416, "300–1000 Hz_10ms_envelope_correlation": 0.9931756548532193, "300–1000 Hz_10ms_modulation_mae_db": 1.1704171098497034, "300–1000 Hz_40ms_log_rms_mae_db": 1.0980179829715297, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06022764963305362, "300–1000 Hz_40ms_envelope_correlation": 0.9976230521200572, "300–1000 Hz_40ms_modulation_mae_db": 1.2447424731629877, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6985928099865861, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15283777368663645, "1000–3000 Hz_10ms_envelope_correlation": 0.986026914421147, "1000–3000 Hz_10ms_modulation_mae_db": 1.2647292202834206, "1000–3000 Hz_40ms_log_rms_mae_db": 1.061291604046706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09495864342479869, "1000–3000 Hz_40ms_envelope_correlation": 0.9954711015339651, "1000–3000 Hz_40ms_modulation_mae_db": 1.9528493762056314, "3000–6000 Hz_10ms_log_rms_mae_db": 2.555906665335675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24173847937814966, "3000–6000 Hz_10ms_envelope_correlation": 0.9649790842164162, "3000–6000 Hz_10ms_modulation_mae_db": 0.4833011791237312, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6532048847260412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13681396589359807, "3000–6000 Hz_40ms_envelope_correlation": 0.9886702098916186, "3000–6000 Hz_40ms_modulation_mae_db": 1.0512985861245623, "mr_stft_db": 6.054957472569181, "band_mae_db": 0.2023194408213832, "rms_gain_error_db": -0.14636501893341902, "output_peak": 0.7799723324324631}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1204483349132013, "全帯域_10ms_rms_envelope_nrmse": 0.07040870982326218, "全帯域_10ms_envelope_correlation": 0.9955320960158558, "全帯域_10ms_modulation_mae_db": 0.2328330347830965, "全帯域_40ms_log_rms_mae_db": 0.6196011554308173, "全帯域_40ms_rms_envelope_nrmse": 0.036471108644576825, "全帯域_40ms_envelope_correlation": 0.998816369935621, "全帯域_40ms_modulation_mae_db": 0.26185203067029883, "300–1000 Hz_10ms_log_rms_mae_db": 1.50333437381708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10502238468453533, "300–1000 Hz_10ms_envelope_correlation": 0.9922454215269906, "300–1000 Hz_10ms_modulation_mae_db": 0.23561211431458182, "300–1000 Hz_40ms_log_rms_mae_db": 1.0024287731492387, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059969871600381607, "300–1000 Hz_40ms_envelope_correlation": 0.9974308846986735, "300–1000 Hz_40ms_modulation_mae_db": 0.3080110106109945, "1000–3000 Hz_10ms_log_rms_mae_db": 1.664164153408698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13554421204091704, "1000–3000 Hz_10ms_envelope_correlation": 0.9891041399246412, "1000–3000 Hz_10ms_modulation_mae_db": 1.0718242095981119, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0165361009844094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08143043805634488, "1000–3000 Hz_40ms_envelope_correlation": 0.9968047395878616, "1000–3000 Hz_40ms_modulation_mae_db": 1.2373489578363601, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6098402781201178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26770853740120104, "3000–6000 Hz_10ms_envelope_correlation": 0.9568819187038179, "3000–6000 Hz_10ms_modulation_mae_db": 0.34022245358886366, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6893468653050754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.158077399008337, "3000–6000 Hz_40ms_envelope_correlation": 0.9846385926163609, "3000–6000 Hz_40ms_modulation_mae_db": 0.6044061273551524, "mr_stft_db": 6.051562156571184, "band_mae_db": 0.2252643678319359, "rms_gain_error_db": -0.09467344138982313, "output_peak": 0.6304265725068166}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1260578429479244, "全帯域_10ms_rms_envelope_nrmse": 0.07832176514221509, "全帯域_10ms_envelope_correlation": 0.9945023792795988, "全帯域_10ms_modulation_mae_db": 0.4391078434659996, "全帯域_40ms_log_rms_mae_db": 0.6381119282539158, "全帯域_40ms_rms_envelope_nrmse": 0.03853117253261058, "全帯域_40ms_envelope_correlation": 0.9987426338785762, "全帯域_40ms_modulation_mae_db": 0.2660903821678839, "300–1000 Hz_10ms_log_rms_mae_db": 1.532079985118344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11418186303099959, "300–1000 Hz_10ms_envelope_correlation": 0.9908893898322763, "300–1000 Hz_10ms_modulation_mae_db": 0.5432693926723665, "300–1000 Hz_40ms_log_rms_mae_db": 1.0246673222295544, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057076559913390085, "300–1000 Hz_40ms_envelope_correlation": 0.9978474541977604, "300–1000 Hz_40ms_modulation_mae_db": 0.5853367892894809, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7343238061869524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16000779559631284, "1000–3000 Hz_10ms_envelope_correlation": 0.983529828002615, "1000–3000 Hz_10ms_modulation_mae_db": 0.3101287098967836, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1117908703718227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09442769321506547, "1000–3000 Hz_40ms_envelope_correlation": 0.9940694173335954, "1000–3000 Hz_40ms_modulation_mae_db": 0.19099873510619014, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5677714418162543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22428114083999562, "3000–6000 Hz_10ms_envelope_correlation": 0.9690563732566408, "3000–6000 Hz_10ms_modulation_mae_db": 0.7834266608643587, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7170278532081094, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.135781204774554, "3000–6000 Hz_40ms_envelope_correlation": 0.9881736249928299, "3000–6000 Hz_40ms_modulation_mae_db": 1.234847904891473, "mr_stft_db": 6.0551146819114114, "band_mae_db": 0.12512633014385902, "rms_gain_error_db": -0.12629558145391875, "output_peak": 0.6640589661104125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.222673679341983, "spectral_convergence": 0.2606137808617202, "spectral_convergence_status": "ok", "amplitude_floor": 1.8779397470586322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.029846592302323, "spectral_convergence": 0.2504054372992801, "spectral_convergence_status": "ok", "amplitude_floor": 1.7710051963816244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.91168015074768, "spectral_convergence": 0.25812325538344505, "spectral_convergence_status": "ok", "amplitude_floor": 1.1208016909489659e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004732055872254276, "output_power": 0.00047799267497287983, "delta_db": 0.043713775105332996, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016454515092618504, "output_power": 0.0016310595360717966, "delta_db": -0.038152746331621544, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001811442607675158, "output_power": 0.0016046466580194284, "delta_db": -0.5264516279142701, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004356719801952226, "output_power": 0.00042154921790805576, "delta_db": -0.14311341721251633, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011106135183657726, "output_power": 0.00010157493716961719, "delta_db": -0.38776393032034706, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.173475364573245e-05, "output_power": 6.317608114238073e-05, "delta_db": 0.10022963386348199, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20657085512459503, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.23450461409249157, "pitch": null, "output_rms_dbfs": -23.665316347855743, "output_peak": 0.6211840625278036, "reference_rms_dbfs": -23.430811733763257, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0547334741306615}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027737662806153396, "log_rms_mae_db": 1.0940875441526603, "rms_envelope_nrmse": 0.08520379009010708, "envelope_correlation": 0.9938859095959184, "rms_gain_error_db": -0.23418794259733824, "reference_mean_power": 0.004541375985905238, "output_mean_power": 0.0043029734347139765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.847489317868867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007100074703458764, "output_power": 0.0006853366869203033, "reference_active": true, "delta_db": -0.15358937413583046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018446902300009587, "output_power": 0.00014437209510197573, "reference_active": true, "delta_db": -1.0644018870083114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9341715429184444e-05, "output_power": 3.2378469048750524e-05, "reference_active": true, "delta_db": 0.4277080938544967}], "active_band_count": 3, "active_band_mae_db": 0.5485664516662129}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002314838901045675, "log_rms_mae_db": 0.5512479871105689, "rms_envelope_nrmse": 0.050106569928542476, "envelope_correlation": 0.9982455692626523, "rms_gain_error_db": -0.23422462485916923, "reference_mean_power": 0.0045632674272876395, "output_mean_power": 0.00432367915063854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7204098415154255e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006818437378247786, "output_power": 0.0006549153557140044, "reference_active": true, "delta_db": -0.17499682822645385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.591872232322897e-05, "output_power": 6.237980693953949e-05, "reference_active": true, "delta_db": -0.853048641898035}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0593810422650415e-06, "output_power": 1.8095912723747438e-06, "reference_active": true, "delta_db": -0.5615621751246992}], "active_band_count": 3, "active_band_mae_db": 0.5298692150830626}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002609159645197948, "log_rms_mae_db": 1.5420723251889101, "rms_envelope_nrmse": 0.13730194469077306, "envelope_correlation": 0.9875917606015671, "rms_gain_error_db": -0.3860819037814138, "reference_mean_power": 0.0022069594688974205, "output_mean_power": 0.002019231752803723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0859718062152698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048225083122535774, "output_power": 0.00042399009524061873, "reference_active": true, "delta_db": -0.5591727364973281}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016124782840735576, "output_power": 0.00011490918179521398, "reference_active": true, "delta_db": -1.4713914223077431}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2038625150435634e-05, "output_power": 3.2072698247329063e-05, "reference_active": true, "delta_db": 1.6295100011454997}], "active_band_count": 3, "active_band_mae_db": 1.2200247199835237}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001981315811908706, "log_rms_mae_db": 1.085771881982542, "rms_envelope_nrmse": 0.08110540984369383, "envelope_correlation": 0.9965123643615652, "rms_gain_error_db": -0.3863318475495537, "reference_mean_power": 0.002217812362061406, "output_mean_power": 0.0020290447021747257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0257217759681144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004916402889453352, "output_power": 0.0004298461177548267, "reference_active": true, "delta_db": -0.5833445657259178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.925750439584527e-05, "output_power": 5.513781448595216e-05, "reference_active": true, "delta_db": -0.9901729040758986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6222165494821603e-06, "output_power": 1.5582658254752033e-06, "reference_active": true, "delta_db": -0.174672814552073}], "active_band_count": 3, "active_band_mae_db": 0.5827300947846298}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001406989062134312, "log_rms_mae_db": 1.7208209627980675, "rms_envelope_nrmse": 0.16914795638910982, "envelope_correlation": 0.9816116373704475, "rms_gain_error_db": -0.18891589882228133, "reference_mean_power": 0.0005036178738924359, "output_mean_power": 0.0004821803916335858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.692987420204184e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020569637799863847, "output_power": 0.00021894024894316078, "reference_active": true, "delta_db": 0.27098963062804926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.72045822558081e-05, "output_power": 7.167544985514177e-05, "reference_active": true, "delta_db": 0.27971541520670434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.876518957296052e-06, "output_power": 9.694887606451596e-06, "reference_active": true, "delta_db": 1.4917413403778723}], "active_band_count": 3, "active_band_mae_db": 0.6808154620708753}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012225910578362228, "log_rms_mae_db": 1.0879924772252692, "rms_envelope_nrmse": 0.11101775535185437, "envelope_correlation": 0.9919074178413483, "rms_gain_error_db": -0.18904158867671472, "reference_mean_power": 0.0005061264034494051, "output_mean_power": 0.00048456811654581423}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5254708519662464e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002159112955909936, "output_power": 0.00023401841686872767, "reference_active": true, "delta_db": 0.34974673594638966}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9510380576050556e-05, "output_power": 3.393528596836742e-05, "reference_active": true, "delta_db": 0.6067670294006726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.547001199869126e-07, "output_power": 9.409958887974328e-07, "reference_active": true, "delta_db": 0.4177396107522879}], "active_band_count": 3, "active_band_mae_db": 0.4580844586997834}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3844040204199326e-05, "log_rms_mae_db": 2.5827743944090407, "rms_envelope_nrmse": 0.23114535786437393, "envelope_correlation": 0.9667980261777944, "rms_gain_error_db": 0.11076669055231171, "reference_mean_power": 4.7417249737341345e-05, "output_mean_power": 4.8642179918385806e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.272212516285404e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.356579821454215e-05, "output_power": 3.2792356538224455e-05, "reference_active": true, "delta_db": -0.10124352178544736}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1452306980819945e-06, "output_power": 5.7630924048938365e-06, "reference_active": true, "delta_db": -0.9336067234543579}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.623091261016208e-06, "output_power": 1.2263683119707778e-06, "reference_active": true, "delta_db": -1.2172201915298382}], "active_band_count": 3, "active_band_mae_db": 0.7506901455898811}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.685962479131491e-05, "log_rms_mae_db": 1.6714772319012674, "rms_envelope_nrmse": 0.14082164380898365, "envelope_correlation": 0.987092972858754, "rms_gain_error_db": 0.10936538100336372, "reference_mean_power": 4.7645706194829226e-05, "output_mean_power": 4.8860769965053815e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.965193836309832e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1852605209893166e-05, "output_power": 3.1156850981351884e-05, "reference_active": true, "delta_db": -0.09591401734544053}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.594578726064393e-06, "output_power": 2.994575699118892e-06, "reference_active": true, "delta_db": -0.7931270370067962}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8710953646003778e-07, "output_power": 1.0369435901386181e-07, "reference_active": true, "delta_db": -2.563407915057124}], "active_band_count": 3, "active_band_mae_db": 1.1508163231364537}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9089088288986455, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3a1fe41cca9732049da170fc57e9527c6c81404d2225b5d81321032bdca687c9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9859555986128994, "standard_deviation": 0.0034320920321871153, "minimum": 0.9815432179050018, "maximum": 0.9902650104130227, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7005720100518067, "standard_deviation": 0.03409785646225578, "minimum": 1.652320915440781, "maximum": 1.751678011506853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8677892046615697, "standard_deviation": 0.4835951662035669, "minimum": 0.1530715720750477, "maximum": 1.5564321444646536, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15049364607184357, "standard_deviation": 0.015545716716890581, "minimum": 0.13177539036816593, "maximum": 0.1693850658615023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950490951765956, "standard_deviation": 0.0017276597276089868, "minimum": 0.9919074178413483, "maximum": 0.9968047395878616, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0672217621741065, "standard_deviation": 0.03508859782368283, "minimum": 1.0165361009844094, "maximum": 1.1117908703718227, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1079174814830486, "standard_deviation": 0.7522192647097662, "minimum": 0.19099873510619014, "maximum": 2.094289415365596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09290923340974552, "standard_deviation": 0.012657723515064131, "minimum": 0.07288633978944416, "maximum": 0.11101775535185437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9650420656763605, "standard_deviation": 0.0035538306637974007, "minimum": 0.9568819187038179, "maximum": 0.9690563732566408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.5750613511929243, "standard_deviation": 0.0366656218748328, "minimum": 2.5182541553416242, "maximum": 2.635426087050946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.584331197045105, "standard_deviation": 0.22528172463997756, "minimum": 0.3019222889295691, "maximum": 0.8240261094715761, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23785772273494393, "standard_deviation": 0.013179644077843727, "minimum": 0.22428114083999562, "maximum": 0.26770853740120104, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9877132444547656, "standard_deviation": 0.0013386244483375041, "minimum": 0.9846385926163609, "maximum": 0.9886702098916186, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.678584430985918, "standard_deviation": 0.03759278513679895, "minimum": 1.6286807711752633, "maximum": 1.7418076967572484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.90937494370302, "standard_deviation": 0.21799054854381825, "minimum": 0.6044061273551524, "maximum": 1.234847904891473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13888508847755454, "standard_deviation": 0.008117533815297756, "minimum": 0.1330799605346357, "maximum": 0.158077399008337, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9902818034847585, "standard_deviation": 0.0021615176541862576, "minimum": 0.9869316138465265, "maximum": 0.9931756548532193, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5285108466052282, "standard_deviation": 0.01405192301106657, "minimum": 1.50333437381708, "maximum": 1.5449957812177892, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7211282634474679, "standard_deviation": 0.351017713928027, "minimum": 0.23561211431458182, "maximum": 1.2200247199835237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11772791096115878, "standard_deviation": 0.01363581620153202, "minimum": 0.09989338708531416, "maximum": 0.13730194469077306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997191982662882, "standard_deviation": 0.0004758994736545641, "minimum": 0.9965123643615652, "maximum": 0.9978474541977604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.058047423595918, "standard_deviation": 0.0348487921588906, "minimum": 1.0024287731492387, "maximum": 1.0980179829715297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6023297372657879, "standard_deviation": 0.2996887049229328, "minimum": 0.2660287485388729, "maximum": 1.2447424731629877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06510193296606456, "standard_deviation": 0.008164248576218915, "minimum": 0.057076559913390085, "maximum": 0.08110540984369383, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18317508041967007, "standard_deviation": 0.04894176460684517, "minimum": 0.12425585881406266, "maximum": 0.2546249959730991, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.054893814447604, "standard_deviation": 0.003250989943821766, "minimum": 6.051562156571184, "maximum": 6.061840118253549, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6826932653373721, "standard_deviation": 0.060881789822761746, "minimum": 0.6211840625278036, "maximum": 0.7799723324324631, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13711256920600765, "standard_deviation": 0.053039602022652836, "minimum": -0.23450461409249157, "maximum": -0.0682764173403918, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947761296762365, "standard_deviation": 0.000623485901812277, "minimum": 0.9938859095959184, "maximum": 0.9955320960158558, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1100722950958088, "standard_deviation": 0.029408836541955946, "minimum": 1.0516959843570957, "maximum": 1.1441843221973536, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5208747041410091, "standard_deviation": 0.23251469889958776, "minimum": 0.2328330347830965, "maximum": 1.011486796940103, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07672643282306084, "standard_deviation": 0.005036728622623726, "minimum": 0.07040870982326218, "maximum": 0.08520379009010708, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985770759085942, "standard_deviation": 0.00022455424027892465, "minimum": 0.9982455692626523, "maximum": 0.998816369935621, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5928732612623175, "standard_deviation": 0.029186377985772973, "minimum": 0.5512479871105689, "maximum": 0.6381119282539158, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6136837502223464, "standard_deviation": 0.3093445072767379, "minimum": 0.26185203067029883, "maximum": 1.1562080083930912, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0413539084399412, "standard_deviation": 0.0046719459535457175, "minimum": 0.036345675230674335, "maximum": 0.050106569928542476, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0054", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.0475821495056152, "predicted_mos": 2.0475821495056152, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "17f3c179235172f2f95833c8ebb8ce9703429026c0336d31525e20b91ff3e458", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.997767999768257, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1702270212189174, "全帯域_10ms_rms_envelope_nrmse": 0.08147336139641108, "全帯域_10ms_envelope_correlation": 0.9939749680580027, "全帯域_10ms_modulation_mae_db": 0.4441458146785295, "全帯域_40ms_log_rms_mae_db": 0.6515845834680245, "全帯域_40ms_rms_envelope_nrmse": 0.04113243572190011, "全帯域_40ms_envelope_correlation": 0.9984617386080193, "全帯域_40ms_modulation_mae_db": 0.8176733125444281, "300–1000 Hz_10ms_log_rms_mae_db": 1.2840399513042011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12121325456470378, "300–1000 Hz_10ms_envelope_correlation": 0.9896181601653077, "300–1000 Hz_10ms_modulation_mae_db": 0.4386175357015654, "300–1000 Hz_40ms_log_rms_mae_db": 0.8476929220570956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05872531768486522, "300–1000 Hz_40ms_envelope_correlation": 0.9976257244895996, "300–1000 Hz_40ms_modulation_mae_db": 0.3000483026951153, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3008789837067898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1416946884962127, "1000–3000 Hz_10ms_envelope_correlation": 0.9876468595044557, "1000–3000 Hz_10ms_modulation_mae_db": 0.6236919300288782, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7941314478463168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0818492520131199, "1000–3000 Hz_40ms_envelope_correlation": 0.9963825824877988, "1000–3000 Hz_40ms_modulation_mae_db": 1.0420337229107266, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6346015985982953, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1948406396859962, "3000–6000 Hz_10ms_envelope_correlation": 0.9764945229955075, "3000–6000 Hz_10ms_modulation_mae_db": 0.1362122919020493, "3000–6000 Hz_40ms_log_rms_mae_db": 1.057978478873492, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11194618306160534, "3000–6000 Hz_40ms_envelope_correlation": 0.991894995386348, "3000–6000 Hz_40ms_modulation_mae_db": 0.4133731531794423, "mr_stft_db": 5.554871065987261, "band_mae_db": 0.12624627569430313, "rms_gain_error_db": -0.09473755808890731, "output_peak": 0.622048478758226}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2121803627282923, "全帯域_10ms_rms_envelope_nrmse": 0.0799732368213536, "全帯域_10ms_envelope_correlation": 0.9943716597054131, "全帯域_10ms_modulation_mae_db": 0.9101421467801153, "全帯域_40ms_log_rms_mae_db": 0.663197391337446, "全帯域_40ms_rms_envelope_nrmse": 0.04631772864517656, "全帯域_40ms_envelope_correlation": 0.9982313924923284, "全帯域_40ms_modulation_mae_db": 1.1547322961711144, "300–1000 Hz_10ms_log_rms_mae_db": 1.2783001108749612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10955183988183019, "300–1000 Hz_10ms_envelope_correlation": 0.9915787119621677, "300–1000 Hz_10ms_modulation_mae_db": 0.7593368108968569, "300–1000 Hz_40ms_log_rms_mae_db": 0.7716031729397796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06845518209529482, "300–1000 Hz_40ms_envelope_correlation": 0.996680302899533, "300–1000 Hz_40ms_modulation_mae_db": 0.9980276056677085, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3531519687250932, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1448877744596097, "1000–3000 Hz_10ms_envelope_correlation": 0.9882744639751535, "1000–3000 Hz_10ms_modulation_mae_db": 0.801184755417271, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9121003650959588, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0908309240351916, "1000–3000 Hz_40ms_envelope_correlation": 0.996818142635228, "1000–3000 Hz_40ms_modulation_mae_db": 1.2797686396801131, "3000–6000 Hz_10ms_log_rms_mae_db": 1.620164089852445, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2017753147535893, "3000–6000 Hz_10ms_envelope_correlation": 0.9741442776967617, "3000–6000 Hz_10ms_modulation_mae_db": 0.6837293525816145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0839870215264147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13019469103520487, "3000–6000 Hz_40ms_envelope_correlation": 0.9890904474347323, "3000–6000 Hz_40ms_modulation_mae_db": 0.5396088070205427, "mr_stft_db": 5.5589645210633165, "band_mae_db": 0.293482545185124, "rms_gain_error_db": -0.16463278819823174, "output_peak": 0.6333292259517038}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.103490106151203, "全帯域_10ms_rms_envelope_nrmse": 0.07183752089637353, "全帯域_10ms_envelope_correlation": 0.9954725714078565, "全帯域_10ms_modulation_mae_db": 0.6581358316096282, "全帯域_40ms_log_rms_mae_db": 0.5546321284722185, "全帯域_40ms_rms_envelope_nrmse": 0.04400516955102905, "全帯域_40ms_envelope_correlation": 0.9983768927847764, "全帯域_40ms_modulation_mae_db": 0.8869062391950422, "300–1000 Hz_10ms_log_rms_mae_db": 1.2933392215035227, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10765193680606977, "300–1000 Hz_10ms_envelope_correlation": 0.9918975842067481, "300–1000 Hz_10ms_modulation_mae_db": 0.8620784098440811, "300–1000 Hz_40ms_log_rms_mae_db": 0.8650519607406938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06394751426006383, "300–1000 Hz_40ms_envelope_correlation": 0.9971587411027223, "300–1000 Hz_40ms_modulation_mae_db": 1.0979738267524966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2781978940452825, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14106431463665198, "1000–3000 Hz_10ms_envelope_correlation": 0.987934350152296, "1000–3000 Hz_10ms_modulation_mae_db": 0.5729704170982456, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8043213342650188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08420835383103281, "1000–3000 Hz_40ms_envelope_correlation": 0.9963371469758494, "1000–3000 Hz_40ms_modulation_mae_db": 1.0695074177656285, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6204329657832495, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20914545494565964, "3000–6000 Hz_10ms_envelope_correlation": 0.9731166735540174, "3000–6000 Hz_10ms_modulation_mae_db": 0.34693144043008245, "3000–6000 Hz_40ms_log_rms_mae_db": 1.101446412520411, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.125601343983729, "3000–6000 Hz_40ms_envelope_correlation": 0.989918123495174, "3000–6000 Hz_40ms_modulation_mae_db": 0.3566885164914592, "mr_stft_db": 5.5467666180537165, "band_mae_db": 0.1634570249114005, "rms_gain_error_db": -0.1481738464932702, "output_peak": 0.6798838988703811}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1833283342863863, "全帯域_10ms_rms_envelope_nrmse": 0.07971309817372065, "全帯域_10ms_envelope_correlation": 0.994376841636306, "全帯域_10ms_modulation_mae_db": 0.4511616533802549, "全帯域_40ms_log_rms_mae_db": 0.6685693877410038, "全帯域_40ms_rms_envelope_nrmse": 0.043931415968127475, "全帯域_40ms_envelope_correlation": 0.998400418893032, "全帯域_40ms_modulation_mae_db": 0.7707336656078815, "300–1000 Hz_10ms_log_rms_mae_db": 1.2300661706077052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10228442285707845, "300–1000 Hz_10ms_envelope_correlation": 0.9927103809830533, "300–1000 Hz_10ms_modulation_mae_db": 0.3987496213319523, "300–1000 Hz_40ms_log_rms_mae_db": 0.8253190619965969, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060922760418992174, "300–1000 Hz_40ms_envelope_correlation": 0.997452096700286, "300–1000 Hz_40ms_modulation_mae_db": 0.4834008344844885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.30143280272487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17002469109396326, "1000–3000 Hz_10ms_envelope_correlation": 0.9825069084394584, "1000–3000 Hz_10ms_modulation_mae_db": 1.1279201257250588, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8063624076784993, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11334456047164695, "1000–3000 Hz_40ms_envelope_correlation": 0.993086213853672, "1000–3000 Hz_40ms_modulation_mae_db": 0.9914016468488464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5787481207078684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1930724069614527, "3000–6000 Hz_10ms_envelope_correlation": 0.978569131274636, "3000–6000 Hz_10ms_modulation_mae_db": 0.41586299100359864, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0484793999511042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12155872932195455, "3000–6000 Hz_40ms_envelope_correlation": 0.9917515719428152, "3000–6000 Hz_40ms_modulation_mae_db": 0.40950119680580177, "mr_stft_db": 5.56203102653943, "band_mae_db": 0.16294833518122345, "rms_gain_error_db": -0.15523661332231298, "output_peak": 0.7855547115425792}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.206233482997294, "全帯域_10ms_rms_envelope_nrmse": 0.0839120051442929, "全帯域_10ms_envelope_correlation": 0.9936837374567529, "全帯域_10ms_modulation_mae_db": 0.3322689805404279, "全帯域_40ms_log_rms_mae_db": 0.6708958422881264, "全帯域_40ms_rms_envelope_nrmse": 0.04411915044217651, "全帯域_40ms_envelope_correlation": 0.9983254798232206, "全帯域_40ms_modulation_mae_db": 0.6050568374340495, "300–1000 Hz_10ms_log_rms_mae_db": 1.2712975946956158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11650807540631937, "300–1000 Hz_10ms_envelope_correlation": 0.9904953614655161, "300–1000 Hz_10ms_modulation_mae_db": 0.5513366657249695, "300–1000 Hz_40ms_log_rms_mae_db": 0.8177420198272267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06597795550211032, "300–1000 Hz_40ms_envelope_correlation": 0.9970004035037192, "300–1000 Hz_40ms_modulation_mae_db": 0.6488322842958598, "1000–3000 Hz_10ms_log_rms_mae_db": 1.347848093791082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16599019463832723, "1000–3000 Hz_10ms_envelope_correlation": 0.9825309633329935, "1000–3000 Hz_10ms_modulation_mae_db": 0.43926399108055825, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8679404145318423, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10012390395641561, "1000–3000 Hz_40ms_envelope_correlation": 0.9938298560633515, "1000–3000 Hz_40ms_modulation_mae_db": 0.5971799055640095, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6389104406071984, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24842330465754567, "3000–6000 Hz_10ms_envelope_correlation": 0.9654851488395851, "3000–6000 Hz_10ms_modulation_mae_db": 1.666991956702125, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0722373026471284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1572439817653383, "3000–6000 Hz_40ms_envelope_correlation": 0.9867953126165175, "3000–6000 Hz_40ms_modulation_mae_db": 1.0695500163583833, "mr_stft_db": 5.535376382593186, "band_mae_db": 0.13164671655813384, "rms_gain_error_db": -0.13444705766959658, "output_peak": 0.5950245172280947}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.14176266152031, "全帯域_10ms_rms_envelope_nrmse": 0.07658459750451498, "全帯域_10ms_envelope_correlation": 0.9946746378268765, "全帯域_10ms_modulation_mae_db": 0.44476256994250846, "全帯域_40ms_log_rms_mae_db": 0.6586809685277074, "全帯域_40ms_rms_envelope_nrmse": 0.04043686567026426, "全帯域_40ms_envelope_correlation": 0.9984913063166414, "全帯域_40ms_modulation_mae_db": 0.38986170110331014, "300–1000 Hz_10ms_log_rms_mae_db": 1.2037365716728528, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11187685418247652, "300–1000 Hz_10ms_envelope_correlation": 0.9911500386808462, "300–1000 Hz_10ms_modulation_mae_db": 0.6870461456070854, "300–1000 Hz_40ms_log_rms_mae_db": 0.8021547250442612, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06433181277985062, "300–1000 Hz_40ms_envelope_correlation": 0.9970133536367842, "300–1000 Hz_40ms_modulation_mae_db": 0.5278303422752032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.281870206696663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14417097689232955, "1000–3000 Hz_10ms_envelope_correlation": 0.9867929949503965, "1000–3000 Hz_10ms_modulation_mae_db": 0.6569592265684486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7820775428468318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0834822771971265, "1000–3000 Hz_40ms_envelope_correlation": 0.995660276791431, "1000–3000 Hz_40ms_modulation_mae_db": 0.5214869830092911, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5708237324057654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18374596498094756, "3000–6000 Hz_10ms_envelope_correlation": 0.9785275526881803, "3000–6000 Hz_10ms_modulation_mae_db": 0.4229026986821027, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0202193907559856, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1067136491828164, "3000–6000 Hz_40ms_envelope_correlation": 0.9925154788498661, "3000–6000 Hz_40ms_modulation_mae_db": 0.49170933745745243, "mr_stft_db": 5.5292060706023785, "band_mae_db": 0.15925230815251223, "rms_gain_error_db": -0.08373602512384663, "output_peak": 0.6367839258652463}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.201204394362242, "全帯域_10ms_rms_envelope_nrmse": 0.09977008453385582, "全帯域_10ms_envelope_correlation": 0.9908889814013935, "全帯域_10ms_modulation_mae_db": 0.9580279873450159, "全帯域_40ms_log_rms_mae_db": 0.622360532590724, "全帯域_40ms_rms_envelope_nrmse": 0.06320265027121201, "全帯域_40ms_envelope_correlation": 0.9961659896397511, "全帯域_40ms_modulation_mae_db": 1.1171121477056405, "300–1000 Hz_10ms_log_rms_mae_db": 1.2351521729258708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14346597145247497, "300–1000 Hz_10ms_envelope_correlation": 0.9853146058220761, "300–1000 Hz_10ms_modulation_mae_db": 1.1704483431966315, "300–1000 Hz_40ms_log_rms_mae_db": 0.7983554406561004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09363954886435964, "300–1000 Hz_40ms_envelope_correlation": 0.993395944665159, "300–1000 Hz_40ms_modulation_mae_db": 0.7289084674805949, "1000–3000 Hz_10ms_log_rms_mae_db": 1.263292386470015, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15188314958656737, "1000–3000 Hz_10ms_envelope_correlation": 0.986173850621482, "1000–3000 Hz_10ms_modulation_mae_db": 1.030824943884612, "1000–3000 Hz_40ms_log_rms_mae_db": 0.75379252836175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09705008332246708, "1000–3000 Hz_40ms_envelope_correlation": 0.9951573453913798, "1000–3000 Hz_40ms_modulation_mae_db": 1.7469827824127366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6472306417245484, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2049895709948284, "3000–6000 Hz_10ms_envelope_correlation": 0.9737799194150552, "3000–6000 Hz_10ms_modulation_mae_db": 0.3349753935905042, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1382529969637247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12884512288058894, "3000–6000 Hz_40ms_envelope_correlation": 0.9890983908875339, "3000–6000 Hz_40ms_modulation_mae_db": 0.11477546851817377, "mr_stft_db": 5.556199266420595, "band_mae_db": 0.14662655884806233, "rms_gain_error_db": -0.07217858845830072, "output_peak": 0.7938814652345876}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1448280681390548, "全帯域_10ms_rms_envelope_nrmse": 0.08679135949475421, "全帯域_10ms_envelope_correlation": 0.9931448423747864, "全帯域_10ms_modulation_mae_db": 0.5961015528490584, "全帯域_40ms_log_rms_mae_db": 0.6153514400123813, "全帯域_40ms_rms_envelope_nrmse": 0.05456060178045582, "全帯域_40ms_envelope_correlation": 0.9971958926009316, "全帯域_40ms_modulation_mae_db": 0.9123885502487689, "300–1000 Hz_10ms_log_rms_mae_db": 1.3296658689782228, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13413989804506485, "300–1000 Hz_10ms_envelope_correlation": 0.987221728056736, "300–1000 Hz_10ms_modulation_mae_db": 0.9751063537072606, "300–1000 Hz_40ms_log_rms_mae_db": 0.8752434838895371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09015342776656064, "300–1000 Hz_40ms_envelope_correlation": 0.9940401762487774, "300–1000 Hz_40ms_modulation_mae_db": 0.5791409274582378, "1000–3000 Hz_10ms_log_rms_mae_db": 1.338925964515056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16183176539828847, "1000–3000 Hz_10ms_envelope_correlation": 0.9833771966911893, "1000–3000 Hz_10ms_modulation_mae_db": 1.3403039010258535, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7951833033106274, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09136854388739346, "1000–3000 Hz_40ms_envelope_correlation": 0.9949525323680339, "1000–3000 Hz_40ms_modulation_mae_db": 0.7518078900749564, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6433894352878229, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18815659761992817, "3000–6000 Hz_10ms_envelope_correlation": 0.977608327591663, "3000–6000 Hz_10ms_modulation_mae_db": 0.42254040695194245, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0632892011371742, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10884164024341271, "3000–6000 Hz_40ms_envelope_correlation": 0.9920895851515762, "3000–6000 Hz_40ms_modulation_mae_db": 0.19812261547341023, "mr_stft_db": 5.546999692085495, "band_mae_db": 0.19046054661895037, "rms_gain_error_db": -0.09446791495626107, "output_peak": 0.6673339472832485}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.595360829715579, "spectral_convergence": 0.24904415031760685, "spectral_convergence_status": "ok", "amplitude_floor": 1.8779397470586322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.39593793717033, "spectral_convergence": 0.24454499750007455, "spectral_convergence_status": "ok", "amplitude_floor": 1.7710051963816244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6733144310758785, "spectral_convergence": 0.2506247345747105, "spectral_convergence_status": "ok", "amplitude_floor": 1.1208016909489659e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004732055872254276, "output_power": 0.0004801653849511167, "delta_db": 0.06340984845918148, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016454515092618504, "output_power": 0.001661025391675956, "delta_db": 0.0409118310222674, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001811442607675158, "output_power": 0.001723865031359809, "delta_db": -0.21521318664234887, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004356719801952226, "output_power": 0.0004053167454613957, "delta_db": -0.31365082846906306, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011106135183657726, "output_power": 0.00010930589524830877, "delta_db": -0.06919369794492067, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.173475364573245e-05, "output_power": 6.095648182766274e-05, "delta_db": -0.05509826162803721, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12624627569430313, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09473755808890731, "pitch": null, "output_rms_dbfs": -23.52554929185216, "output_peak": 0.622048478758226, "reference_rms_dbfs": -23.430811733763257, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.554871065987261}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027737662806153396, "log_rms_mae_db": 1.1702270212189174, "rms_envelope_nrmse": 0.08147336139641108, "envelope_correlation": 0.9939749680580027, "rms_gain_error_db": -0.0940505283148351, "reference_mean_power": 0.004541375985905238, "output_mean_power": 0.004444085505620134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.847489317868867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007100074703458764, "output_power": 0.0006742797691839402, "reference_active": true, "delta_db": -0.2242278869685131}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018446902300009587, "output_power": 0.00016333100059098062, "reference_active": true, "delta_db": -0.528548249954419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9341715429184444e-05, "output_power": 2.5675527215070375e-05, "reference_active": true, "delta_db": -0.5796613071126564}], "active_band_count": 3, "active_band_mae_db": 0.4441458146785295}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002314838901045675, "log_rms_mae_db": 0.6515845834680245, "rms_envelope_nrmse": 0.04113243572190011, "envelope_correlation": 0.9984617386080193, "rms_gain_error_db": -0.09367267768919696, "reference_mean_power": 0.0045632674272876395, "output_mean_power": 0.004465896494773566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7204098415154255e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006818437378247786, "output_power": 0.0006397034079163104, "reference_active": true, "delta_db": -0.27706191859041684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.591872232322897e-05, "output_power": 6.904470297936616e-05, "reference_active": true, "delta_db": -0.41218524826999287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0593810422650415e-06, "output_power": 1.3720168132608881e-06, "reference_active": true, "delta_db": -1.7637727707728748}], "active_band_count": 3, "active_band_mae_db": 0.8176733125444281}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002609159645197948, "log_rms_mae_db": 1.2840399513042011, "rms_envelope_nrmse": 0.12121325456470378, "envelope_correlation": 0.9896181601653077, "rms_gain_error_db": -0.14570986552687465, "reference_mean_power": 0.0022069594688974205, "output_mean_power": 0.002134142284935817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0859718062152698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048225083122535774, "output_power": 0.0004593544415090686, "reference_active": true, "delta_db": -0.21125064947296543}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016124782840735576, "output_power": 0.0001424523299518625, "reference_active": true, "delta_db": -0.5382431756050973}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2038625150435634e-05, "output_power": 2.5108477333164935e-05, "reference_active": true, "delta_db": 0.5663587820266335}], "active_band_count": 3, "active_band_mae_db": 0.4386175357015654}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001981315811908706, "log_rms_mae_db": 0.8476929220570956, "rms_envelope_nrmse": 0.05872531768486522, "envelope_correlation": 0.9976257244895996, "rms_gain_error_db": -0.1455536895824222, "reference_mean_power": 0.002217812362061406, "output_mean_power": 0.0021447142183278886}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0257217759681144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004916402889453352, "output_power": 0.0004643881271243771, "reference_active": true, "delta_db": -0.24766357514784845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.925750439584527e-05, "output_power": 6.36542273848595e-05, "reference_active": true, "delta_db": -0.36639587079953073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6222165494821603e-06, "output_power": 1.5187989262965188e-06, "reference_active": true, "delta_db": -0.2860854621379667}], "active_band_count": 3, "active_band_mae_db": 0.3000483026951153}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001406989062134312, "log_rms_mae_db": 1.3008789837067898, "rms_envelope_nrmse": 0.1416946884962127, "envelope_correlation": 0.9876468595044557, "rms_gain_error_db": -0.334133250120337, "reference_mean_power": 0.0005036178738924359, "output_mean_power": 0.000466324045792956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.692987420204184e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020569637799863847, "output_power": 0.00018591864867650366, "reference_active": true, "delta_db": -0.4390369038118468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.72045822558081e-05, "output_power": 5.097624234515103e-05, "reference_active": true, "delta_db": -1.2003106706307038}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.876518957296052e-06, "output_power": 7.253397297638769e-06, "reference_active": true, "delta_db": 0.23172821564408405}], "active_band_count": 3, "active_band_mae_db": 0.6236919300288782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012225910578362228, "log_rms_mae_db": 0.7941314478463168, "rms_envelope_nrmse": 0.0818492520131199, "envelope_correlation": 0.9963825824877988, "rms_gain_error_db": -0.33428836944776025, "reference_mean_power": 0.0005061264034494051, "output_mean_power": 0.0004686300755163079}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5254708519662464e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002159112955909936, "output_power": 0.00019241470762700318, "reference_active": true, "delta_db": -0.5003709823767122}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9510380576050556e-05, "output_power": 2.303280853768782e-05, "reference_active": true, "delta_db": -1.0762791285002709}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.547001199869126e-07, "output_power": 5.982306486928215e-07, "reference_active": true, "delta_db": -1.5494510578551965}], "active_band_count": 3, "active_band_mae_db": 1.0420337229107266}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3844040204199326e-05, "log_rms_mae_db": 1.6346015985982953, "rms_envelope_nrmse": 0.1948406396859962, "envelope_correlation": 0.9764945229955075, "rms_gain_error_db": 0.1269589426605756, "reference_mean_power": 4.7417249737341345e-05, "output_mean_power": 4.8823876116342563e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.272212516285404e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.356579821454215e-05, "output_power": 3.388855380377729e-05, "reference_active": true, "delta_db": 0.04156056264267717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1452306980819945e-06, "output_power": 6.845615226778805e-06, "reference_active": true, "delta_db": -0.18603770432014413}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.623091261016208e-06, "output_power": 1.6921809450037874e-06, "reference_active": true, "delta_db": 0.18103860874332664}], "active_band_count": 3, "active_band_mae_db": 0.1362122919020493}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.685962479131491e-05, "log_rms_mae_db": 1.057978478873492, "rms_envelope_nrmse": 0.11194618306160534, "envelope_correlation": 0.991894995386348, "rms_gain_error_db": 0.1265955731646472, "reference_mean_power": 4.7645706194829226e-05, "output_mean_power": 4.9055005153187024e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.965193836309832e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1852605209893166e-05, "output_power": 3.2391376989735384e-05, "reference_active": true, "delta_db": 0.07284451824615559}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.594578726064393e-06, "output_power": 3.0356666563684493e-06, "reference_active": true, "delta_db": -0.7339391929675054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8710953646003778e-07, "output_power": 2.0674237341612557e-07, "reference_active": true, "delta_db": 0.43333574832466587}], "active_band_count": 3, "active_band_mae_db": 0.4133731531794423}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9246165755969153, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7b3cc8d9f1d2e26953d2deffcb954398836b511c4b071fec73db7d090f339143", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9856546984584281, "standard_deviation": 0.002462497110962612, "minimum": 0.9825069084394584, "maximum": 0.9882744639751535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3081997875843565, "standard_deviation": 0.03432243351584401, "minimum": 1.263292386470015, "maximum": 1.3531519687250932, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8241399113536159, "standard_deviation": 0.3119747444373511, "minimum": 0.43926399108055825, "maximum": 1.3403039010258535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1526934444002438, "standard_deviation": 0.01165645984815408, "minimum": 0.14106431463665198, "maximum": 0.17002469109396326, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9952780120708431, "standard_deviation": 0.0013028322858996179, "minimum": 0.993086213853672, "maximum": 0.996818142635228, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8144886679921057, "standard_deviation": 0.050816419334986436, "minimum": 0.75379252836175, "maximum": 0.9121003650959588, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0000211235332885, "standard_deviation": 0.395648449908143, "minimum": 0.5214869830092911, "maximum": 1.7469827824127366, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09278223733929924, "standard_deviation": 0.01055773764077818, "minimum": 0.0818492520131199, "maximum": 0.11334456047164695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9747156942569257, "standard_deviation": 0.004307349668874091, "minimum": 0.9654851488395851, "maximum": 0.978569131274636, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6192876281208992, "standard_deviation": 0.029217319746847114, "minimum": 1.5708237324057654, "maximum": 1.6472306417245484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5537683164805024, "standard_deviation": 0.47427027663129045, "minimum": 0.1362122919020493, "maximum": 1.666991956702125, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20301865682499348, "standard_deviation": 0.020222601950664026, "minimum": 0.18374596498094756, "maximum": 0.24842330465754567, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9903942382205704, "standard_deviation": 0.0020009473251809567, "minimum": 0.9867953126165175, "maximum": 0.9925154788498661, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0732362755469294, "standard_deviation": 0.035647079601212515, "minimum": 1.0202193907559856, "maximum": 1.1382529969637247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4491661389130832, "standard_deviation": 0.288125092517585, "minimum": 0.11477546851817377, "maximum": 1.0695500163583833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12386816768433127, "standard_deviation": 0.016253725292705194, "minimum": 0.1067136491828164, "maximum": 0.1572439817653383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9899983214178063, "standard_deviation": 0.002530521883354515, "minimum": 0.9853146058220761, "maximum": 0.9927103809830533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.265699707820369, "standard_deviation": 0.040398752843583755, "minimum": 1.2037365716728528, "maximum": 1.3296658689782228, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7303399857513003, "standard_deviation": 0.2674676141250454, "minimum": 0.3987496213319523, "maximum": 1.1704483431966315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11833653164950224, "standard_deviation": 0.014058972362825434, "minimum": 0.10228442285707845, "maximum": 0.14346597145247497, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962958429058226, "standard_deviation": 0.0016259757425760295, "minimum": 0.993395944665159, "maximum": 0.9976257244895996, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8253953483939114, "standard_deviation": 0.03541211634551923, "minimum": 0.7716031729397796, "maximum": 0.8752434838895371, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6705203238887131, "standard_deviation": 0.2658761553401726, "minimum": 0.3000483026951153, "maximum": 1.0979738267524966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07076918992151217, "standard_deviation": 0.013400931045396509, "minimum": 0.05872531768486522, "maximum": 0.09363954886435964, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17176503889371375, "standard_deviation": 0.05315930224660452, "minimum": 0.12624627569430313, "maximum": 0.293482545185124, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5488018304181725, "standard_deviation": 0.011597264671125205, "minimum": 5.5292060706023785, "maximum": 5.56203102653943, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6767300213417584, "standard_deviation": 0.07446061973644522, "minimum": 0.5950245172280947, "maximum": 0.7938814652345876, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1184512990388409, "standard_deviation": 0.036073420406212746, "minimum": -0.16463278819823174, "maximum": -0.07217858845830072, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9938235299834235, "standard_deviation": 0.0013713908398506733, "minimum": 0.9908889814013935, "maximum": 0.9954725714078565, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1704068039254625, "standard_deviation": 0.03797870465958529, "minimum": 1.103490106151203, "maximum": 1.2121803627282923, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5993433171406923, "standard_deviation": 0.2299023559371756, "minimum": 0.3322689805404279, "maximum": 0.9580279873450159, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0825069079956596, "standard_deviation": 0.008302759876567235, "minimum": 0.07183752089637353, "maximum": 0.09977008453385582, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979561388948376, "standard_deviation": 0.000837619060979976, "minimum": 0.9961659896397511, "maximum": 0.9984913063166414, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.638159034304704, "standard_deviation": 0.039589547109436296, "minimum": 0.5546321284722185, "maximum": 0.6708958422881264, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8318080937512795, "standard_deviation": 0.2523801224176311, "minimum": 0.38986170110331014, "maximum": 1.1547322961711144, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04721325225629272, "standard_deviation": 0.007782561196476274, "minimum": 0.04043686567026426, "maximum": 0.06320265027121201, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0054", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.4917962551116943, "predicted_mos": 2.4917962551116943, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "92f16c9b9bb50ba909be1a1bb02902e2e3fe2a01b9b1d4e99ea73b43276e60c0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9966750000603497, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1440308714940977, "全帯域_10ms_rms_envelope_nrmse": 0.07323972570461441, "全帯域_10ms_envelope_correlation": 0.9951832900138401, "全帯域_10ms_modulation_mae_db": 0.4385143710475477, "全帯域_40ms_log_rms_mae_db": 0.6224895606717094, "全帯域_40ms_rms_envelope_nrmse": 0.03867431757475578, "全帯域_40ms_envelope_correlation": 0.9986892480120755, "全帯域_40ms_modulation_mae_db": 0.9053527238445441, "300–1000 Hz_10ms_log_rms_mae_db": 1.4175379344300767, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10799193643439657, "300–1000 Hz_10ms_envelope_correlation": 0.9917448751872027, "300–1000 Hz_10ms_modulation_mae_db": 0.6306151936672729, "300–1000 Hz_40ms_log_rms_mae_db": 0.9833239715124532, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06354510669584791, "300–1000 Hz_40ms_envelope_correlation": 0.9970802952631012, "300–1000 Hz_40ms_modulation_mae_db": 0.5315006097911895, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4628885508074538, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14318081648103204, "1000–3000 Hz_10ms_envelope_correlation": 0.98732106656083, "1000–3000 Hz_10ms_modulation_mae_db": 0.3913224893106791, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8769994950370398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08364176105359274, "1000–3000 Hz_40ms_envelope_correlation": 0.9961109239938555, "1000–3000 Hz_40ms_modulation_mae_db": 0.9500714535069514, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9400085865565513, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1953610622878312, "3000–6000 Hz_10ms_envelope_correlation": 0.978055436966507, "3000–6000 Hz_10ms_modulation_mae_db": 0.7012816560594534, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2499901901094699, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12286790290961196, "3000–6000 Hz_40ms_envelope_correlation": 0.9915712716541395, "3000–6000 Hz_40ms_modulation_mae_db": 0.9643355908836572, "mr_stft_db": 5.6952391640382745, "band_mae_db": 0.12011252017067432, "rms_gain_error_db": -0.11094938347458229, "output_peak": 0.6702436918505661}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1149026116975749, "全帯域_10ms_rms_envelope_nrmse": 0.09300290673021844, "全帯域_10ms_envelope_correlation": 0.992166858902323, "全帯域_10ms_modulation_mae_db": 0.5485271698151167, "全帯域_40ms_log_rms_mae_db": 0.6059302112091789, "全帯域_40ms_rms_envelope_nrmse": 0.050658642452762324, "全帯域_40ms_envelope_correlation": 0.9976845422882592, "全帯域_40ms_modulation_mae_db": 1.1636007956377064, "300–1000 Hz_10ms_log_rms_mae_db": 1.3639791468579052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1312771532955198, "300–1000 Hz_10ms_envelope_correlation": 0.9877625728145983, "300–1000 Hz_10ms_modulation_mae_db": 0.5912804316708361, "300–1000 Hz_40ms_log_rms_mae_db": 0.8984738709559568, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07476760055853347, "300–1000 Hz_40ms_envelope_correlation": 0.9959588228125742, "300–1000 Hz_40ms_modulation_mae_db": 1.060781242439685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4526647875737415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1704695678204084, "1000–3000 Hz_10ms_envelope_correlation": 0.9817724108374586, "1000–3000 Hz_10ms_modulation_mae_db": 1.1272036264280914, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9075166636457767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09483628343169712, "1000–3000 Hz_40ms_envelope_correlation": 0.9949673544288561, "1000–3000 Hz_40ms_modulation_mae_db": 0.42940678925342807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9119577292989138, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20231188142171805, "3000–6000 Hz_10ms_envelope_correlation": 0.97427281772236, "3000–6000 Hz_10ms_modulation_mae_db": 0.5366212261862716, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2980402166255476, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12576299030305882, "3000–6000 Hz_40ms_envelope_correlation": 0.9895114638113215, "3000–6000 Hz_40ms_modulation_mae_db": 0.48361361818385684, "mr_stft_db": 5.687064352259043, "band_mae_db": 0.18120080476779019, "rms_gain_error_db": -0.12702106970219412, "output_peak": 0.6446577873927267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0880495759362112, "全帯域_10ms_rms_envelope_nrmse": 0.07905598160522363, "全帯域_10ms_envelope_correlation": 0.9943078711673482, "全帯域_10ms_modulation_mae_db": 0.5049313689255501, "全帯域_40ms_log_rms_mae_db": 0.6185903960470893, "全帯域_40ms_rms_envelope_nrmse": 0.043253645693337464, "全帯域_40ms_envelope_correlation": 0.9982298839632141, "全帯域_40ms_modulation_mae_db": 0.615001012958164, "300–1000 Hz_10ms_log_rms_mae_db": 1.3838266414300084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11639279937402011, "300–1000 Hz_10ms_envelope_correlation": 0.9903795686741841, "300–1000 Hz_10ms_modulation_mae_db": 0.40082896528487194, "300–1000 Hz_40ms_log_rms_mae_db": 0.9003042117313462, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07489590606987452, "300–1000 Hz_40ms_envelope_correlation": 0.9957870760420321, "300–1000 Hz_40ms_modulation_mae_db": 0.3401235566962966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4819045111019085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14321047215178495, "1000–3000 Hz_10ms_envelope_correlation": 0.9873123044427707, "1000–3000 Hz_10ms_modulation_mae_db": 1.158442534683937, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9226048637140758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08128542542007298, "1000–3000 Hz_40ms_envelope_correlation": 0.9963906304766121, "1000–3000 Hz_40ms_modulation_mae_db": 0.6889408818367379, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9856266874837956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19004245173272555, "3000–6000 Hz_10ms_envelope_correlation": 0.9770972402081346, "3000–6000 Hz_10ms_modulation_mae_db": 0.7638304959417382, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3008068046620307, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11542670384422359, "3000–6000 Hz_40ms_envelope_correlation": 0.9911302536756071, "3000–6000 Hz_40ms_modulation_mae_db": 0.9553040708742092, "mr_stft_db": 5.714048564085819, "band_mae_db": 0.1130758126372507, "rms_gain_error_db": -0.05188441476479819, "output_peak": 0.8005826803016234}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1139839046856652, "全帯域_10ms_rms_envelope_nrmse": 0.08629899544008539, "全帯域_10ms_envelope_correlation": 0.9932157881403676, "全帯域_10ms_modulation_mae_db": 0.48953720170275483, "全帯域_40ms_log_rms_mae_db": 0.6093507497499937, "全帯域_40ms_rms_envelope_nrmse": 0.047103312646701916, "全帯域_40ms_envelope_correlation": 0.9979108703097339, "全帯域_40ms_modulation_mae_db": 0.5344013603359588, "300–1000 Hz_10ms_log_rms_mae_db": 1.3500409719866007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13638433729587957, "300–1000 Hz_10ms_envelope_correlation": 0.9867243760580764, "300–1000 Hz_10ms_modulation_mae_db": 0.6692360164179219, "300–1000 Hz_40ms_log_rms_mae_db": 0.934926708879492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07770099004645284, "300–1000 Hz_40ms_envelope_correlation": 0.9954552209136938, "300–1000 Hz_40ms_modulation_mae_db": 1.226034023431635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4268805699569558, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1408961152534085, "1000–3000 Hz_10ms_envelope_correlation": 0.9873275808583379, "1000–3000 Hz_10ms_modulation_mae_db": 1.3381589011615702, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8964375694726816, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08523459934587566, "1000–3000 Hz_40ms_envelope_correlation": 0.9953433802598048, "1000–3000 Hz_40ms_modulation_mae_db": 0.8534362897438291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8908576670799193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19535572850316138, "3000–6000 Hz_10ms_envelope_correlation": 0.9757182525505046, "3000–6000 Hz_10ms_modulation_mae_db": 0.6815272734593436, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3111510572067144, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11730483610188312, "3000–6000 Hz_40ms_envelope_correlation": 0.9909519332057861, "3000–6000 Hz_40ms_modulation_mae_db": 1.1161045174350066, "mr_stft_db": 5.682694258264376, "band_mae_db": 0.11592636449250042, "rms_gain_error_db": -0.07086500842473617, "output_peak": 0.7999740787568608}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.107195072647745, "全帯域_10ms_rms_envelope_nrmse": 0.0748935506143653, "全帯域_10ms_envelope_correlation": 0.9949303553470966, "全帯域_10ms_modulation_mae_db": 0.5323569376830094, "全帯域_40ms_log_rms_mae_db": 0.5888277903976635, "全帯域_40ms_rms_envelope_nrmse": 0.039616706926639725, "全帯域_40ms_envelope_correlation": 0.99858428663249, "全帯域_40ms_modulation_mae_db": 0.5987744633237208, "300–1000 Hz_10ms_log_rms_mae_db": 1.3194597480738732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12800477479006922, "300–1000 Hz_10ms_envelope_correlation": 0.9883087067666065, "300–1000 Hz_10ms_modulation_mae_db": 1.0259595740991456, "300–1000 Hz_40ms_log_rms_mae_db": 0.850848702657516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07198199757773269, "300–1000 Hz_40ms_envelope_correlation": 0.9961226761496559, "300–1000 Hz_40ms_modulation_mae_db": 0.40474907757979345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4656635148629067, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15730417588952456, "1000–3000 Hz_10ms_envelope_correlation": 0.9848641067932965, "1000–3000 Hz_10ms_modulation_mae_db": 0.9376401366889614, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8886335074458811, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09226641604807462, "1000–3000 Hz_40ms_envelope_correlation": 0.9955775042461839, "1000–3000 Hz_40ms_modulation_mae_db": 0.531046683758639, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0078419102238474, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20929184550168628, "3000–6000 Hz_10ms_envelope_correlation": 0.9721174543830374, "3000–6000 Hz_10ms_modulation_mae_db": 0.3582895954989005, "3000–6000 Hz_40ms_log_rms_mae_db": 1.341876828024242, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12734594073652625, "3000–6000 Hz_40ms_envelope_correlation": 0.9893029331757028, "3000–6000 Hz_40ms_modulation_mae_db": 0.5623156282235265, "mr_stft_db": 5.711643250681745, "band_mae_db": 0.1959206363779936, "rms_gain_error_db": -0.09770982465215795, "output_peak": 0.6767416525570651}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.131770927478391, "全帯域_10ms_rms_envelope_nrmse": 0.07613007893462889, "全帯域_10ms_envelope_correlation": 0.9947014582967554, "全帯域_10ms_modulation_mae_db": 0.27067860444624564, "全帯域_40ms_log_rms_mae_db": 0.596267732534323, "全帯域_40ms_rms_envelope_nrmse": 0.04264780408326272, "全帯域_40ms_envelope_correlation": 0.9982518228288962, "全帯域_40ms_modulation_mae_db": 0.507259502616384, "300–1000 Hz_10ms_log_rms_mae_db": 1.3887920441629829, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1134349986464347, "300–1000 Hz_10ms_envelope_correlation": 0.9908649017997818, "300–1000 Hz_10ms_modulation_mae_db": 0.4102115076204782, "300–1000 Hz_40ms_log_rms_mae_db": 0.9218594282457525, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06664179136018429, "300–1000 Hz_40ms_envelope_correlation": 0.9966608312519869, "300–1000 Hz_40ms_modulation_mae_db": 0.2963802435271739, "1000–3000 Hz_10ms_log_rms_mae_db": 1.410817420583147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1519448158400384, "1000–3000 Hz_10ms_envelope_correlation": 0.9852674796000209, "1000–3000 Hz_10ms_modulation_mae_db": 0.20461675217412437, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9109612367515103, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0914958706416591, "1000–3000 Hz_40ms_envelope_correlation": 0.9946791393433125, "1000–3000 Hz_40ms_modulation_mae_db": 1.0183295342517695, "3000–6000 Hz_10ms_log_rms_mae_db": 2.001440236479224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20120327338450436, "3000–6000 Hz_10ms_envelope_correlation": 0.9751622538216078, "3000–6000 Hz_10ms_modulation_mae_db": 0.4474883259474242, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3180423018333252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12335228984396894, "3000–6000 Hz_40ms_envelope_correlation": 0.9902965143372701, "3000–6000 Hz_40ms_modulation_mae_db": 0.2139320862854263, "mr_stft_db": 5.706577455450876, "band_mae_db": 0.09767315746230203, "rms_gain_error_db": -0.04131675963679345, "output_peak": 0.7058684562673853}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1438107212761495, "全帯域_10ms_rms_envelope_nrmse": 0.0724571706113292, "全帯域_10ms_envelope_correlation": 0.9954634131474172, "全帯域_10ms_modulation_mae_db": 0.5097577023998755, "全帯域_40ms_log_rms_mae_db": 0.5971766825476317, "全帯域_40ms_rms_envelope_nrmse": 0.04326930225570728, "全帯域_40ms_envelope_correlation": 0.9985228370874225, "全帯域_40ms_modulation_mae_db": 0.5548816698284242, "300–1000 Hz_10ms_log_rms_mae_db": 1.3689341205086378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10878798527252684, "300–1000 Hz_10ms_envelope_correlation": 0.9919537821435035, "300–1000 Hz_10ms_modulation_mae_db": 0.5092780353956395, "300–1000 Hz_40ms_log_rms_mae_db": 0.9578652841695128, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0698674990950493, "300–1000 Hz_40ms_envelope_correlation": 0.9968288469438084, "300–1000 Hz_40ms_modulation_mae_db": 0.6945649270401847, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4289013795418153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13393237473468922, "1000–3000 Hz_10ms_envelope_correlation": 0.988779086614791, "1000–3000 Hz_10ms_modulation_mae_db": 0.2881528899670645, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8795551934505043, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07849988095369884, "1000–3000 Hz_40ms_envelope_correlation": 0.9963828287901255, "1000–3000 Hz_40ms_modulation_mae_db": 0.5398741333454469, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9785413788417183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19348527507396382, "3000–6000 Hz_10ms_envelope_correlation": 0.9761668422615783, "3000–6000 Hz_10ms_modulation_mae_db": 1.051100686511849, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3169787937299864, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11682411001895451, "3000–6000 Hz_40ms_envelope_correlation": 0.9909755442842809, "3000–6000 Hz_40ms_modulation_mae_db": 1.31623073841149, "mr_stft_db": 5.701521362471836, "band_mae_db": 0.15583397898552848, "rms_gain_error_db": -0.16797597200240894, "output_peak": 0.6819686283957773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.127271494417758, "全帯域_10ms_rms_envelope_nrmse": 0.0800670928116123, "全帯域_10ms_envelope_correlation": 0.9941904284595431, "全帯域_10ms_modulation_mae_db": 0.7165603912842585, "全帯域_40ms_log_rms_mae_db": 0.5958753794176195, "全帯域_40ms_rms_envelope_nrmse": 0.051518640524634444, "全帯域_40ms_envelope_correlation": 0.9975142332709781, "全帯域_40ms_modulation_mae_db": 0.8750481262798586, "300–1000 Hz_10ms_log_rms_mae_db": 1.3673174035138371, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11365469510693249, "300–1000 Hz_10ms_envelope_correlation": 0.9908045415720095, "300–1000 Hz_10ms_modulation_mae_db": 0.4453968008393925, "300–1000 Hz_40ms_log_rms_mae_db": 0.9246089485652256, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07468633168028943, "300–1000 Hz_40ms_envelope_correlation": 0.9958110905817117, "300–1000 Hz_40ms_modulation_mae_db": 0.33668381732813213, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4299564204344584, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13241213322861786, "1000–3000 Hz_10ms_envelope_correlation": 0.9897718890710642, "1000–3000 Hz_10ms_modulation_mae_db": 0.8162309388926746, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9321412720976459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08118099416634861, "1000–3000 Hz_40ms_envelope_correlation": 0.9969805295310719, "1000–3000 Hz_40ms_modulation_mae_db": 1.0002009883336545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9283934767443902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2019618478884153, "3000–6000 Hz_10ms_envelope_correlation": 0.9740843788686177, "3000–6000 Hz_10ms_modulation_mae_db": 0.32413238472560707, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2449894800197558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12166067322773101, "3000–6000 Hz_40ms_envelope_correlation": 0.990209660143034, "3000–6000 Hz_40ms_modulation_mae_db": 0.48813965091424044, "mr_stft_db": 5.697412530966543, "band_mae_db": 0.15321991639713878, "rms_gain_error_db": -0.09068076143007837, "output_peak": 0.7440696561973585}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.806190018121504, "spectral_convergence": 0.24241643917133163, "spectral_convergence_status": "ok", "amplitude_floor": 1.8779397470586322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.569995114676284, "spectral_convergence": 0.24354450583983486, "spectral_convergence_status": "ok", "amplitude_floor": 1.7710051963816244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.709532359317036, "spectral_convergence": 0.2503437993447695, "spectral_convergence_status": "ok", "amplitude_floor": 1.1208016909489659e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004732055872254276, "output_power": 0.00047543696712107136, "delta_db": 0.020430830527163255, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016454515092618504, "output_power": 0.0016205068533372497, "delta_db": -0.0663421628928474, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001811442607675158, "output_power": 0.0017540894955832696, "delta_db": -0.1397283096386002, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004356719801952226, "output_power": 0.000401855976129113, "delta_db": -0.3508919837708694, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011106135183657726, "output_power": 0.00010936763056532685, "delta_db": -0.06674152106323747, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.173475364573245e-05, "output_power": 6.283241475960256e-05, "delta_db": 0.07654031313132824, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12011252017067432, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11094938347458229, "pitch": null, "output_rms_dbfs": -23.54176111723784, "output_peak": 0.6702436918505661, "reference_rms_dbfs": -23.430811733763257, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6952391640382745}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027737662806153396, "log_rms_mae_db": 1.1440308714940977, "rms_envelope_nrmse": 0.07323972570461441, "envelope_correlation": 0.9951832900138401, "rms_gain_error_db": -0.11034522756921666, "reference_mean_power": 0.004541375985905238, "output_mean_power": 0.004427442568818193}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.847489317868867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007100074703458764, "output_power": 0.0006767975584592802, "reference_active": true, "delta_db": -0.20804134853295542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018446902300009587, "output_power": 0.0001480452701848729, "reference_active": true, "delta_db": -0.9552891070784809}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9341715429184444e-05, "output_power": 2.833115205709454e-05, "reference_active": true, "delta_db": -0.15221265753120677}], "active_band_count": 3, "active_band_mae_db": 0.4385143710475477}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002314838901045675, "log_rms_mae_db": 0.6224895606717094, "rms_envelope_nrmse": 0.03867431757475578, "envelope_correlation": 0.9986892480120755, "rms_gain_error_db": -0.11018587554240579, "reference_mean_power": 0.0045632674272876395, "output_mean_power": 0.004448948039244587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7204098415154255e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006818437378247786, "output_power": 0.0006434833293481004, "reference_active": true, "delta_db": -0.25147556049560693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.591872232322897e-05, "output_power": 6.199078489309905e-05, "reference_active": true, "delta_db": -0.8802175532330757}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0593810422650415e-06, "output_power": 1.4298820185201215e-06, "reference_active": true, "delta_db": -1.5843650578049497}], "active_band_count": 3, "active_band_mae_db": 0.9053527238445441}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002609159645197948, "log_rms_mae_db": 1.4175379344300767, "rms_envelope_nrmse": 0.10799193643439657, "envelope_correlation": 0.9917448751872027, "rms_gain_error_db": -0.12715394079334885, "reference_mean_power": 0.0022069594688974205, "output_mean_power": 0.0021432802562072087}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0859718062152698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048225083122535774, "output_power": 0.00044080706608393605, "reference_active": true, "delta_db": -0.3902443729506239}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016124782840735576, "output_power": 0.00013632778308408526, "reference_active": true, "delta_db": -0.7290950209983211}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2038625150435634e-05, "output_power": 1.8447345403015327e-05, "reference_active": true, "delta_db": -0.7725061870528737}], "active_band_count": 3, "active_band_mae_db": 0.6306151936672729}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001981315811908706, "log_rms_mae_db": 0.9833239715124532, "rms_envelope_nrmse": 0.06354510669584791, "envelope_correlation": 0.9970802952631012, "rms_gain_error_db": -0.1272381878824504, "reference_mean_power": 0.002217812362061406, "output_mean_power": 0.0021537782212311916}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0257217759681144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004916402889453352, "output_power": 0.0004440472532579412, "reference_active": true, "delta_db": -0.4421827725182689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.925750439584527e-05, "output_power": 6.130883250785328e-05, "reference_active": true, "delta_db": -0.5294379223738245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6222165494821603e-06, "output_power": 1.4054671982847934e-06, "reference_active": true, "delta_db": -0.6228811344814752}], "active_band_count": 3, "active_band_mae_db": 0.5315006097911895}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001406989062134312, "log_rms_mae_db": 1.4628885508074538, "rms_envelope_nrmse": 0.14318081648103204, "envelope_correlation": 0.98732106656083, "rms_gain_error_db": -0.3252017725146613, "reference_mean_power": 0.0005036178738924359, "output_mean_power": 0.00046728405072222226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.692987420204184e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020569637799863847, "output_power": 0.00019458375080747467, "reference_active": true, "delta_db": -0.24120073872160167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.72045822558081e-05, "output_power": 5.666056470138546e-05, "reference_active": true, "delta_db": -0.7411798730493379}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.876518957296052e-06, "output_power": 7.18666382736287e-06, "reference_active": true, "delta_db": 0.1915868561610978}], "active_band_count": 3, "active_band_mae_db": 0.3913224893106791}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012225910578362228, "log_rms_mae_db": 0.8769994950370398, "rms_envelope_nrmse": 0.08364176105359274, "envelope_correlation": 0.9961109239938555, "rms_gain_error_db": -0.3253633303998761, "reference_mean_power": 0.0005061264034494051, "output_mean_power": 0.0004695941315987648}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5254708519662464e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002159112955909936, "output_power": 0.00020399826647143448, "reference_active": true, "delta_db": -0.24648886518960117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9510380576050556e-05, "output_power": 2.5192126140681168e-05, "reference_active": true, "delta_db": -0.6870998813728384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.547001199869126e-07, "output_power": 5.497322347676968e-07, "reference_active": true, "delta_db": -1.9166256139584148}], "active_band_count": 3, "active_band_mae_db": 0.9500714535069514}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3844040204199326e-05, "log_rms_mae_db": 1.9400085865565513, "rms_envelope_nrmse": 0.1953610622878312, "envelope_correlation": 0.978055436966507, "rms_gain_error_db": 0.37155953612153136, "reference_mean_power": 4.7417249737341345e-05, "output_mean_power": 5.165261512355905e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.272212516285404e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.356579821454215e-05, "output_power": 3.7059783752022285e-05, "reference_active": true, "delta_db": 0.4300590140997071}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1452306980819945e-06, "output_power": 6.10455625841466e-06, "reference_active": true, "delta_db": -0.6836215495703719}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.623091261016208e-06, "output_power": 1.2921903665525671e-06, "reference_active": true, "delta_db": -0.9901644045082812}], "active_band_count": 3, "active_band_mae_db": 0.7012816560594534}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.685962479131491e-05, "log_rms_mae_db": 1.2499901901094699, "rms_envelope_nrmse": 0.12286790290961196, "envelope_correlation": 0.9915712716541395, "rms_gain_error_db": 0.37108061265460723, "reference_mean_power": 4.7645706194829226e-05, "output_mean_power": 5.1895754402383774e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.965193836309832e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1852605209893166e-05, "output_power": 3.503314988858455e-05, "reference_active": true, "delta_db": 0.4133422834059498}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.594578726064393e-06, "output_power": 2.8103038244189746e-06, "reference_active": true, "delta_db": -1.0689472527782322}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8710953646003778e-07, "output_power": 1.3521479037971615e-07, "reference_active": true, "delta_db": -1.4107172364667893}], "active_band_count": 3, "active_band_mae_db": 0.9643355908836572}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9199766004739945, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4843d313c953c29ffe3b6134f5d2fce73410ba8d5d662bfac99edeca5eab8800", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9865519905973212, "standard_deviation": 0.0025196967222668974, "minimum": 0.9817724108374586, "maximum": 0.9897718890710642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4449596443577986, "standard_deviation": 0.024348356716491843, "minimum": 1.410817420583147, "maximum": 1.4819045111019085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7827210336633879, "standard_deviation": 0.4351459524418287, "minimum": 0.20461675217412437, "maximum": 1.3381589011615702, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.146668808924938, "standard_deviation": 0.012707383331765982, "minimum": 0.13241213322861786, "maximum": 0.1704695678204084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9958040363837278, "standard_deviation": 0.0007915951643448265, "minimum": 0.9946791393433125, "maximum": 0.9969805295310719, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9018562252018893, "standard_deviation": 0.019935707414136812, "minimum": 0.8769994950370398, "maximum": 0.9321412720976459, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.751413344253807, "standard_deviation": 0.23418997262342142, "minimum": 0.42940678925342807, "maximum": 1.0183295342517695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08605515388262747, "standard_deviation": 0.0060400695238357624, "minimum": 0.07849988095369884, "maximum": 0.09483628343169712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9753343345977934, "standard_deviation": 0.0018677654260799035, "minimum": 0.9721174543830374, "maximum": 0.978055436966507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.955583459088545, "standard_deviation": 0.04365114242426862, "minimum": 1.8908576670799193, "maximum": 2.0078419102238474, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6080339555413234, "standard_deviation": 0.24157936167302815, "minimum": 0.32413238472560707, "maximum": 1.051100686511849, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19862667072425075, "standard_deviation": 0.006171073596155072, "minimum": 0.19004245173272555, "maximum": 0.20929184550168628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9904936967858927, "standard_deviation": 0.0008027441527089081, "minimum": 0.9893029331757028, "maximum": 0.9915712716541395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.297734459026384, "standard_deviation": 0.03376893662475234, "minimum": 1.2449894800197558, "maximum": 1.341876828024242, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7624969876514267, "standard_deviation": 0.3786473396194608, "minimum": 0.2139320862854263, "maximum": 1.31623073841149, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12131818087324478, "standard_deviation": 0.004371333062858784, "minimum": 0.11542670384422359, "maximum": 0.12734594073652625, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9898179156269953, "standard_deviation": 0.0019542004720611416, "minimum": 0.9867243760580764, "maximum": 0.9919537821435035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3699860013704903, "standard_deviation": 0.028788473732170618, "minimum": 1.3194597480738732, "maximum": 1.4175379344300767, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5853508156244449, "standard_deviation": 0.20471342821120303, "minimum": 0.40082896528487194, "maximum": 1.0259595740991456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11949108502697242, "standard_deviation": 0.010848976627684921, "minimum": 0.10799193643439657, "maximum": 0.13638433729587957, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962131074948205, "standard_deviation": 0.0005759411244814318, "minimum": 0.9954552209136938, "maximum": 0.9970802952631012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9215263908396569, "standard_deviation": 0.0401884363448866, "minimum": 0.850848702657516, "maximum": 0.9833239715124532, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6113521872292613, "standard_deviation": 0.3553711789085315, "minimum": 0.2963802435271739, "maximum": 1.226034023431635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07176090288549555, "standard_deviation": 0.00478165714836113, "minimum": 0.06354510669584791, "maximum": 0.07770099004645284, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14162039891139733, "standard_deviation": 0.03529594203770591, "minimum": 0.09767315746230203, "maximum": 0.1959206363779936, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6995251172773145, "standard_deviation": 0.011166794072732331, "minimum": 5.682694258264376, "maximum": 5.714048564085819, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7155133289649205, "standard_deviation": 0.05973421827481777, "minimum": 0.6446577873927267, "maximum": 0.8005826803016234, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09480039926096868, "standard_deviation": 0.041324925512029334, "minimum": -0.16797597200240894, "maximum": -0.04131675963679345, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9942699329343364, "standard_deviation": 0.0010969366420822233, "minimum": 0.992166858902323, "maximum": 0.9954634131474172, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.121376897454199, "standard_deviation": 0.01915227833190617, "minimum": 1.0880495759362112, "maximum": 1.1440308714940977, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5013579684130448, "standard_deviation": 0.12356796533788036, "minimum": 0.27067860444624564, "maximum": 0.7165603912842585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0793931878065097, "standard_deviation": 0.007074607803105955, "minimum": 0.0724571706113292, "maximum": 0.09300290673021844, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981734655491337, "standard_deviation": 0.00043239834403270127, "minimum": 0.9975142332709781, "maximum": 0.9986892480120755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6043135628219011, "standard_deviation": 0.011885550655781544, "minimum": 0.5888277903976635, "maximum": 0.6224895606717094, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7192899568530952, "standard_deviation": 0.23539117641941507, "minimum": 0.507259502616384, "maximum": 1.1636007956377064, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04459279651972521, "standard_deviation": 0.004753679950891429, "minimum": 0.03867431757475578, "maximum": 0.051518640524634444, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3575-170457-0054", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.421778917312622, "predicted_mos": 2.421778917312622, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d65fe4ad82985d00636fec83ae6ea8fef5c9891108288bd6330488a51555d923", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.811796700116247, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1176443962038174, "全帯域_10ms_rms_envelope_nrmse": 0.08671696263527612, "全帯域_10ms_envelope_correlation": 0.9931459739172929, "全帯域_10ms_modulation_mae_db": 0.8067232638814007, "全帯域_40ms_log_rms_mae_db": 0.649467156657808, "全帯域_40ms_rms_envelope_nrmse": 0.04774291667323639, "全帯域_40ms_envelope_correlation": 0.9978604152819239, "全帯域_40ms_modulation_mae_db": 0.6225289856741758, "300–1000 Hz_10ms_log_rms_mae_db": 1.3121500850264218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11880883329775221, "300–1000 Hz_10ms_envelope_correlation": 0.9899334084969577, "300–1000 Hz_10ms_modulation_mae_db": 0.7417250497753202, "300–1000 Hz_40ms_log_rms_mae_db": 0.8810056540686819, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07578593532913058, "300–1000 Hz_40ms_envelope_correlation": 0.9956855317076645, "300–1000 Hz_40ms_modulation_mae_db": 0.47209032130964257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3737808090460184, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15267181059171403, "1000–3000 Hz_10ms_envelope_correlation": 0.9856190861654461, "1000–3000 Hz_10ms_modulation_mae_db": 1.122945235475946, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8090992147770979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07635432938294022, "1000–3000 Hz_40ms_envelope_correlation": 0.9971754424113465, "1000–3000 Hz_40ms_modulation_mae_db": 0.73003687561825, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8548970825890694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18725795045757224, "3000–6000 Hz_10ms_envelope_correlation": 0.9777261471213807, "3000–6000 Hz_10ms_modulation_mae_db": 0.6256286566990618, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1657506275199234, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10561709239113021, "3000–6000 Hz_40ms_envelope_correlation": 0.9926533226516613, "3000–6000 Hz_40ms_modulation_mae_db": 0.8978206816586866, "mr_stft_db": 5.66116804068935, "band_mae_db": 0.16784329198284767, "rms_gain_error_db": -0.08458335183689464, "output_peak": 0.7220703114258734}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1051526830638148, "全帯域_10ms_rms_envelope_nrmse": 0.07542897430354743, "全帯域_10ms_envelope_correlation": 0.9950551933650156, "全帯域_10ms_modulation_mae_db": 0.4748332069571586, "全帯域_40ms_log_rms_mae_db": 0.5917937697715355, "全帯域_40ms_rms_envelope_nrmse": 0.043419173026841375, "全帯域_40ms_envelope_correlation": 0.998517253775949, "全帯域_40ms_modulation_mae_db": 0.6585446867832657, "300–1000 Hz_10ms_log_rms_mae_db": 1.3504021446240175, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11237104150647942, "300–1000 Hz_10ms_envelope_correlation": 0.9912487803981458, "300–1000 Hz_10ms_modulation_mae_db": 0.17277939457412084, "300–1000 Hz_40ms_log_rms_mae_db": 0.8583824669990271, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06336506141335035, "300–1000 Hz_40ms_envelope_correlation": 0.9973643670426067, "300–1000 Hz_40ms_modulation_mae_db": 0.188428866292782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3852977990632498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1594123389323163, "1000–3000 Hz_10ms_envelope_correlation": 0.9854926623558696, "1000–3000 Hz_10ms_modulation_mae_db": 0.8313558315147155, "1000–3000 Hz_40ms_log_rms_mae_db": 0.912244772516843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10788425948754671, "1000–3000 Hz_40ms_envelope_correlation": 0.9946805797133719, "1000–3000 Hz_40ms_modulation_mae_db": 1.2490354546700584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8846220973249777, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19707607853815154, "3000–6000 Hz_10ms_envelope_correlation": 0.9754433678800177, "3000–6000 Hz_10ms_modulation_mae_db": 0.27210828849562224, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2248195910049735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11797390131155858, "3000–6000 Hz_40ms_envelope_correlation": 0.9907204788537524, "3000–6000 Hz_40ms_modulation_mae_db": 0.48190150487683486, "mr_stft_db": 5.664364307967496, "band_mae_db": 0.2402826622547268, "rms_gain_error_db": -0.1694906429484938, "output_peak": 0.6559649226045557}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1275462436545394, "全帯域_10ms_rms_envelope_nrmse": 0.07160102693417515, "全帯域_10ms_envelope_correlation": 0.9955298419118784, "全帯域_10ms_modulation_mae_db": 0.5024966449281653, "全帯域_40ms_log_rms_mae_db": 0.5880852224491633, "全帯域_40ms_rms_envelope_nrmse": 0.0387390437955847, "全帯域_40ms_envelope_correlation": 0.9988382413439051, "全帯域_40ms_modulation_mae_db": 0.4524783331199793, "300–1000 Hz_10ms_log_rms_mae_db": 1.3009665268849837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10303528646686158, "300–1000 Hz_10ms_envelope_correlation": 0.9926317715628216, "300–1000 Hz_10ms_modulation_mae_db": 0.6029294796455472, "300–1000 Hz_40ms_log_rms_mae_db": 0.8621860720812846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058927679444821884, "300–1000 Hz_40ms_envelope_correlation": 0.9976823030143379, "300–1000 Hz_40ms_modulation_mae_db": 0.39669270459507494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4018348720263833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1406747482827905, "1000–3000 Hz_10ms_envelope_correlation": 0.9880121001097271, "1000–3000 Hz_10ms_modulation_mae_db": 1.2106832350494268, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8499969658165898, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08213918457601416, "1000–3000 Hz_40ms_envelope_correlation": 0.996571832655531, "1000–3000 Hz_40ms_modulation_mae_db": 0.7353075756927963, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8775887634917798, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19054749217984826, "3000–6000 Hz_10ms_envelope_correlation": 0.9769366662708293, "3000–6000 Hz_10ms_modulation_mae_db": 0.6680606587162314, "3000–6000 Hz_40ms_log_rms_mae_db": 1.211633875016415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11782332044143036, "3000–6000 Hz_40ms_envelope_correlation": 0.9909342602895085, "3000–6000 Hz_40ms_modulation_mae_db": 0.7332740054866326, "mr_stft_db": 5.68537019827216, "band_mae_db": 0.19219203695200346, "rms_gain_error_db": -0.15815220164024635, "output_peak": 0.6920136448942794}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0940893332796005, "全帯域_10ms_rms_envelope_nrmse": 0.0736168206910295, "全帯域_10ms_envelope_correlation": 0.9951912403476266, "全帯域_10ms_modulation_mae_db": 0.4573224434957304, "全帯域_40ms_log_rms_mae_db": 0.5654435307527323, "全帯域_40ms_rms_envelope_nrmse": 0.04924736696569666, "全帯域_40ms_envelope_correlation": 0.997848192876882, "全帯域_40ms_modulation_mae_db": 0.6399051110961466, "300–1000 Hz_10ms_log_rms_mae_db": 1.2963703100151043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11840935770600355, "300–1000 Hz_10ms_envelope_correlation": 0.9901213174993928, "300–1000 Hz_10ms_modulation_mae_db": 0.5072322332587541, "300–1000 Hz_40ms_log_rms_mae_db": 0.8280842201740563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08345958282210852, "300–1000 Hz_40ms_envelope_correlation": 0.9949466025822908, "300–1000 Hz_40ms_modulation_mae_db": 0.5096187177945983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4372809566267462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1313563633044118, "1000–3000 Hz_10ms_envelope_correlation": 0.98918161202669, "1000–3000 Hz_10ms_modulation_mae_db": 0.7036760269573804, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8536020159150072, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07538885773861712, "1000–3000 Hz_40ms_envelope_correlation": 0.9966410660381774, "1000–3000 Hz_40ms_modulation_mae_db": 0.9245433078073867, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8580300826101477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1873437581263256, "3000–6000 Hz_10ms_envelope_correlation": 0.977953968795339, "3000–6000 Hz_10ms_modulation_mae_db": 0.17859754989779827, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2230792045430827, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11414259375134825, "3000–6000 Hz_40ms_envelope_correlation": 0.9913677465235253, "3000–6000 Hz_40ms_modulation_mae_db": 0.6012516818581458, "mr_stft_db": 5.655909092291883, "band_mae_db": 0.15985021949624184, "rms_gain_error_db": -0.13570535545264029, "output_peak": 0.6518759974020673}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.115671786756339, "全帯域_10ms_rms_envelope_nrmse": 0.08343025312209043, "全帯域_10ms_envelope_correlation": 0.9940480163606009, "全帯域_10ms_modulation_mae_db": 0.5357319975815524, "全帯域_40ms_log_rms_mae_db": 0.5832774796753969, "全帯域_40ms_rms_envelope_nrmse": 0.04886780590734878, "全帯域_40ms_envelope_correlation": 0.9982290395970143, "全帯域_40ms_modulation_mae_db": 0.6716788381423516, "300–1000 Hz_10ms_log_rms_mae_db": 1.3814000697644209, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12714344354920842, "300–1000 Hz_10ms_envelope_correlation": 0.9890099084357852, "300–1000 Hz_10ms_modulation_mae_db": 0.7070550611310885, "300–1000 Hz_40ms_log_rms_mae_db": 0.8925259727940282, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08194769197873038, "300–1000 Hz_40ms_envelope_correlation": 0.995696563693985, "300–1000 Hz_40ms_modulation_mae_db": 0.5731723982157625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4993086922802341, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16484473105668845, "1000–3000 Hz_10ms_envelope_correlation": 0.9837205790819894, "1000–3000 Hz_10ms_modulation_mae_db": 0.7204522039025184, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9260190930694918, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0948169129161272, "1000–3000 Hz_40ms_envelope_correlation": 0.9958808400703559, "1000–3000 Hz_40ms_modulation_mae_db": 1.274580628736288, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8587850905306627, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18956861820865625, "3000–6000 Hz_10ms_envelope_correlation": 0.9780989871721619, "3000–6000 Hz_10ms_modulation_mae_db": 0.2986752637716316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1797206283440071, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11260003867171862, "3000–6000 Hz_40ms_envelope_correlation": 0.9920543875493725, "3000–6000 Hz_40ms_modulation_mae_db": 0.14973458347147237, "mr_stft_db": 5.655259303563828, "band_mae_db": 0.26180735552497075, "rms_gain_error_db": -0.20783208233605072, "output_peak": 0.7549056410797826}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1175705143992094, "全帯域_10ms_rms_envelope_nrmse": 0.08505658449875758, "全帯域_10ms_envelope_correlation": 0.9936684853948088, "全帯域_10ms_modulation_mae_db": 0.7751758538737424, "全帯域_40ms_log_rms_mae_db": 0.6319082865313788, "全帯域_40ms_rms_envelope_nrmse": 0.05127587473289705, "全帯域_40ms_envelope_correlation": 0.9978487283174159, "全帯域_40ms_modulation_mae_db": 0.48739198234026, "300–1000 Hz_10ms_log_rms_mae_db": 1.3568497963847639, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12278341096464067, "300–1000 Hz_10ms_envelope_correlation": 0.9897029976707761, "300–1000 Hz_10ms_modulation_mae_db": 0.7775754959699785, "300–1000 Hz_40ms_log_rms_mae_db": 0.906752708113909, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08110859647759386, "300–1000 Hz_40ms_envelope_correlation": 0.9956920523521042, "300–1000 Hz_40ms_modulation_mae_db": 0.3309765095644889, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4326375133266336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13991620906536506, "1000–3000 Hz_10ms_envelope_correlation": 0.9879178264986728, "1000–3000 Hz_10ms_modulation_mae_db": 0.7368116351035018, "1000–3000 Hz_40ms_log_rms_mae_db": 0.887505433356681, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08074478395987443, "1000–3000 Hz_40ms_envelope_correlation": 0.996438710748087, "1000–3000 Hz_40ms_modulation_mae_db": 0.8379969428405126, "3000–6000 Hz_10ms_log_rms_mae_db": 1.880059862107245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20913914396168407, "3000–6000 Hz_10ms_envelope_correlation": 0.9721576065961925, "3000–6000 Hz_10ms_modulation_mae_db": 0.6074344823935789, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1899414699859983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11771840293918506, "3000–6000 Hz_40ms_envelope_correlation": 0.9909204144488077, "3000–6000 Hz_40ms_modulation_mae_db": 0.9435263931988405, "mr_stft_db": 5.66113871918548, "band_mae_db": 0.1912131097824921, "rms_gain_error_db": -0.1879975052801551, "output_peak": 0.5941583384080196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1254922393456084, "全帯域_10ms_rms_envelope_nrmse": 0.07603315898252841, "全帯域_10ms_envelope_correlation": 0.9948623306374057, "全帯域_10ms_modulation_mae_db": 1.0350724029423757, "全帯域_40ms_log_rms_mae_db": 0.5794624890878692, "全帯域_40ms_rms_envelope_nrmse": 0.045804599295571624, "全帯域_40ms_envelope_correlation": 0.9981807739469875, "全帯域_40ms_modulation_mae_db": 1.3204170470643124, "300–1000 Hz_10ms_log_rms_mae_db": 1.3062762229190479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11192639096809397, "300–1000 Hz_10ms_envelope_correlation": 0.9912547625305593, "300–1000 Hz_10ms_modulation_mae_db": 0.7507320480347713, "300–1000 Hz_40ms_log_rms_mae_db": 0.8462315752058002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06907011435107953, "300–1000 Hz_40ms_envelope_correlation": 0.9967009973274775, "300–1000 Hz_40ms_modulation_mae_db": 0.71666410594552, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4195189845634397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15576177806583155, "1000–3000 Hz_10ms_envelope_correlation": 0.9848366420829983, "1000–3000 Hz_10ms_modulation_mae_db": 1.0847865552011737, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8364859536224821, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09208968784675106, "1000–3000 Hz_40ms_envelope_correlation": 0.99511655580325, "1000–3000 Hz_40ms_modulation_mae_db": 0.4409213282684234, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8931871222612078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19891250815246875, "3000–6000 Hz_10ms_envelope_correlation": 0.9747975472585084, "3000–6000 Hz_10ms_modulation_mae_db": 1.374524553020635, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2261672092722888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11887963722145034, "3000–6000 Hz_40ms_envelope_correlation": 0.990670255458736, "3000–6000 Hz_40ms_modulation_mae_db": 1.6749297080633554, "mr_stft_db": 5.6593643687042805, "band_mae_db": 0.1544826912730976, "rms_gain_error_db": -0.1386024853572414, "output_peak": 0.5854376397215448}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1153240251128875, "全帯域_10ms_rms_envelope_nrmse": 0.07643589788071783, "全帯域_10ms_envelope_correlation": 0.9946936198619367, "全帯域_10ms_modulation_mae_db": 0.6272338689256445, "全帯域_40ms_log_rms_mae_db": 0.5635980880331011, "全帯域_40ms_rms_envelope_nrmse": 0.03850401938725692, "全帯域_40ms_envelope_correlation": 0.9986395736790327, "全帯域_40ms_modulation_mae_db": 0.9292501602143081, "300–1000 Hz_10ms_log_rms_mae_db": 1.31076593620971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11144216880586476, "300–1000 Hz_10ms_envelope_correlation": 0.9911654518313, "300–1000 Hz_10ms_modulation_mae_db": 0.3751150007327757, "300–1000 Hz_40ms_log_rms_mae_db": 0.8404473766758491, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05870983360197372, "300–1000 Hz_40ms_envelope_correlation": 0.9974634560458453, "300–1000 Hz_40ms_modulation_mae_db": 0.29725533642565, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4269711629127362, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14093160035018218, "1000–3000 Hz_10ms_envelope_correlation": 0.9877975096432293, "1000–3000 Hz_10ms_modulation_mae_db": 0.9405375668746013, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8325379646061725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0857455028879136, "1000–3000 Hz_40ms_envelope_correlation": 0.9959395513375993, "1000–3000 Hz_40ms_modulation_mae_db": 1.3882250033325354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.853448093841657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1869306349199255, "3000–6000 Hz_10ms_envelope_correlation": 0.9777722072618176, "3000–6000 Hz_10ms_modulation_mae_db": 1.1590214693198113, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1732958722873286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12034061249740123, "3000–6000 Hz_40ms_envelope_correlation": 0.9903154682164552, "3000–6000 Hz_40ms_modulation_mae_db": 1.1449240928853393, "mr_stft_db": 5.6444040759551575, "band_mae_db": 0.1600801490052068, "rms_gain_error_db": -0.08388132484855211, "output_peak": 0.7293137325031696}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.77656126220608, "spectral_convergence": 0.24129726352935865, "spectral_convergence_status": "ok", "amplitude_floor": 1.8779397470586322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.526280792135027, "spectral_convergence": 0.24191105895890133, "spectral_convergence_status": "ok", "amplitude_floor": 1.7710051963816244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.68066206772694, "spectral_convergence": 0.24833830732923753, "spectral_convergence_status": "ok", "amplitude_floor": 1.1208016909489659e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004732055872254276, "output_power": 0.0004862040464429921, "delta_db": 0.1176870516593715, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016454515092618504, "output_power": 0.0016389680837072362, "delta_db": -0.017145918904669655, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001811442607675158, "output_power": 0.00176517192589254, "delta_db": -0.11237567132688872, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004356719801952226, "output_power": 0.00039498771562275253, "delta_db": -0.425760406285279, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011106135183657726, "output_power": 0.00010554847143371708, "delta_db": -0.22111007210836295, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.173475364573245e-05, "output_power": 6.014944940530591e-05, "delta_db": -0.11298063161251433, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16784329198284767, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08458335183689464, "pitch": null, "output_rms_dbfs": -23.515395085600147, "output_peak": 0.7220703114258734, "reference_rms_dbfs": -23.430811733763257, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.66116804068935}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027737662806153396, "log_rms_mae_db": 1.1176443962038174, "rms_envelope_nrmse": 0.08671696263527612, "envelope_correlation": 0.9931459739172929, "rms_gain_error_db": -0.0839340015860986, "reference_mean_power": 0.004541375985905238, "output_mean_power": 0.004454449697692066}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.847489317868867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007100074703458764, "output_power": 0.0006482722867512805, "reference_active": true, "delta_db": -0.39505462000966096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018446902300009587, "output_power": 0.00014352891695452736, "reference_active": true, "delta_db": -1.0898403987425367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9341715429184444e-05, "output_power": 3.639259111939473e-05, "reference_active": true, "delta_db": 0.9352747728920048}], "active_band_count": 3, "active_band_mae_db": 0.8067232638814007}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002314838901045675, "log_rms_mae_db": 0.649467156657808, "rms_envelope_nrmse": 0.04774291667323639, "envelope_correlation": 0.9978604152819239, "rms_gain_error_db": -0.08362861594023732, "reference_mean_power": 0.0045632674272876395, "output_mean_power": 0.004476236863270643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7204098415154255e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006818437378247786, "output_power": 0.0006162583674252897, "reference_active": true, "delta_db": -0.43922027132688857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.591872232322897e-05, "output_power": 6.06229514542716e-05, "reference_active": true, "delta_db": -0.9771181414811907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0593810422650415e-06, "output_power": 1.85614481309776e-06, "reference_active": true, "delta_db": -0.45124854421444804}], "active_band_count": 3, "active_band_mae_db": 0.6225289856741758}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002609159645197948, "log_rms_mae_db": 1.3121500850264218, "rms_envelope_nrmse": 0.11880883329775221, "envelope_correlation": 0.9899334084969577, "rms_gain_error_db": -0.06212474922921979, "reference_mean_power": 0.0022069594688974205, "output_mean_power": 0.0021756141884308926}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0859718062152698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048225083122535774, "output_power": 0.0004414924351847409, "reference_active": true, "delta_db": -0.3834971838614787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016124782840735576, "output_power": 0.00012545427370327843, "reference_active": true, "delta_db": -1.0900841401712156}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2038625150435634e-05, "output_power": 2.620256944570936e-05, "reference_active": true, "delta_db": 0.7515938252932666}], "active_band_count": 3, "active_band_mae_db": 0.7417250497753202}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001981315811908706, "log_rms_mae_db": 0.8810056540686819, "rms_envelope_nrmse": 0.07578593532913058, "envelope_correlation": 0.9956855317076645, "rms_gain_error_db": -0.062094478373540665, "reference_mean_power": 0.002217812362061406, "output_mean_power": 0.002186328177685281}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0257217759681144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004916402889453352, "output_power": 0.0004435743944885667, "reference_active": true, "delta_db": -0.4468099686712035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.925750439584527e-05, "output_power": 5.6107886031308534e-05, "reference_active": true, "delta_db": -0.9144293209675785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6222165494821603e-06, "output_power": 1.6017903031865083e-06, "reference_active": true, "delta_db": -0.055031674290145854}], "active_band_count": 3, "active_band_mae_db": 0.47209032130964257}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001406989062134312, "log_rms_mae_db": 1.3737808090460184, "rms_envelope_nrmse": 0.15267181059171403, "envelope_correlation": 0.9856190861654461, "rms_gain_error_db": -0.36945877157273743, "reference_mean_power": 0.0005036178738924359, "output_mean_power": 0.00046254634983523244}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.692987420204184e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020569637799863847, "output_power": 0.00018564378668636, "reference_active": true, "delta_db": -0.44546226095334734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.72045822558081e-05, "output_power": 5.290518393577045e-05, "reference_active": true, "delta_db": -1.0390065721984258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.876518957296052e-06, "output_power": 1.0612197696296207e-05, "reference_active": true, "delta_db": 1.8843668732760643}], "active_band_count": 3, "active_band_mae_db": 1.122945235475946}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012225910578362228, "log_rms_mae_db": 0.8090992147770979, "rms_envelope_nrmse": 0.07635432938294022, "envelope_correlation": 0.9971754424113465, "rms_gain_error_db": -0.3696191325708706, "reference_mean_power": 0.0005061264034494051, "output_mean_power": 0.000464833137362036}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5254708519662464e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002159112955909936, "output_power": 0.00019508177154125848, "reference_active": true, "delta_db": -0.440586726558733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9510380576050556e-05, "output_power": 2.4200167254860927e-05, "reference_active": true, "delta_db": -0.8615644281291533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.547001199869126e-07, "output_power": 6.966550950504452e-07, "reference_active": true, "delta_db": -0.8879594721668636}], "active_band_count": 3, "active_band_mae_db": 0.73003687561825}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3844040204199326e-05, "log_rms_mae_db": 1.8548970825890694, "rms_envelope_nrmse": 0.18725795045757224, "envelope_correlation": 0.9777261471213807, "rms_gain_error_db": -0.1139383894416225, "reference_mean_power": 4.7417249737341345e-05, "output_mean_power": 4.618942137506208e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.272212516285404e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.356579821454215e-05, "output_power": 3.138266754526791e-05, "reference_active": true, "delta_db": -0.2920712331622746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1452306980819945e-06, "output_power": 6.27470820451543e-06, "reference_active": true, "delta_db": -0.5642272098293146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.623091261016208e-06, "output_power": 1.283169968405239e-06, "reference_active": true, "delta_db": -1.020587527105596}], "active_band_count": 3, "active_band_mae_db": 0.6256286566990618}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.685962479131491e-05, "log_rms_mae_db": 1.1657506275199234, "rms_envelope_nrmse": 0.10561709239113021, "envelope_correlation": 0.9926533226516613, "rms_gain_error_db": -0.11443657951729651, "reference_mean_power": 4.7645706194829226e-05, "output_mean_power": 4.640663842461451e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.965193836309832e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1852605209893166e-05, "output_power": 2.9915295897118698e-05, "reference_active": true, "delta_db": -0.2725165599192892}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.594578726064393e-06, "output_power": 2.955020764207702e-06, "reference_active": true, "delta_db": -0.8508746277059589}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8710953646003778e-07, "output_power": 1.3034333750592735e-07, "reference_active": true, "delta_db": -1.5700708573508118}], "active_band_count": 3, "active_band_mae_db": 0.8978206816586866}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9257748858021619, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "361a14a4144533ff56b207d715ce8a0460653f29eeab11b08d3b95ef81fd4aa6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9865722522455778, "standard_deviation": 0.001905337183317511, "minimum": 0.9837205790819894, "maximum": 0.98918161202669, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4220788487306801, "standard_deviation": 0.038605866883260065, "minimum": 1.3737808090460184, "maximum": 1.4993086922802341, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.918906036259908, "standard_deviation": 0.2004571518149291, "minimum": 0.7036760269573804, "maximum": 1.2106832350494268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14819619745616247, "standard_deviation": 0.011598826383681524, "minimum": 0.1313563633044118, "maximum": 0.16484473105668845, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960555723472149, "standard_deviation": 0.000830368302026387, "minimum": 0.9946805797133719, "maximum": 0.9971754424113465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8634364267100456, "standard_deviation": 0.04103843518228768, "minimum": 0.8090992147770979, "maximum": 0.9260190930694918, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9475808896207814, "standard_deviation": 0.32811138344673035, "minimum": 0.4409213282684234, "maximum": 1.3882250033325354, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08689543984947307, "standard_deviation": 0.010925408305270912, "minimum": 0.07538885773861712, "maximum": 0.10788425948754671, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9763608122945309, "standard_deviation": 0.0020941968203149414, "minimum": 0.9721576065961925, "maximum": 0.9780989871721619, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8700772743445935, "standard_deviation": 0.015497961081634095, "minimum": 1.853448093841657, "maximum": 1.8931871222612078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6480063652892962, "standard_deviation": 0.42681383470214185, "minimum": 0.17859754989779827, "maximum": 1.374524553020635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19334702306807905, "standard_deviation": 0.00784375374606373, "minimum": 0.1869306349199255, "maximum": 0.20913914396168407, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9912045417489774, "standard_deviation": 0.0007848430825517476, "minimum": 0.9903154682164552, "maximum": 0.9926533226516613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1993010597467522, "standard_deviation": 0.0249667791336242, "minimum": 1.1657506275199234, "maximum": 1.2261672092722888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8284203314374134, "standard_deviation": 0.45921775571352963, "minimum": 0.14973458347147237, "maximum": 1.6749297080633554, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11563694990315285, "standard_deviation": 0.004766953838656418, "minimum": 0.10561709239113021, "maximum": 0.12034061249740123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9906335498032173, "standard_deviation": 0.0011523232587517835, "minimum": 0.9890099084357852, "maximum": 0.9926317715628216, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.326897636478559, "standard_deviation": 0.03145839395282233, "minimum": 1.2963703100151043, "maximum": 1.3814000697644209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5793929703902946, "standard_deviation": 0.2151769818278109, "minimum": 0.17277939457412084, "maximum": 0.7775754959699785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11573999165811308, "standard_deviation": 0.0075778433162546075, "minimum": 0.10303528646686158, "maximum": 0.12714344354920842, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996403984220789, "standard_deviation": 0.0010291853558765302, "minimum": 0.9949466025822908, "maximum": 0.9976823030143379, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8644520057640795, "standard_deviation": 0.027056333318968246, "minimum": 0.8280842201740563, "maximum": 0.906752708113909, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4356123700179399, "standard_deviation": 0.16793969026968178, "minimum": 0.188428866292782, "maximum": 0.71666410594552, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0715468119273486, "standard_deviation": 0.010396399766474301, "minimum": 0.05870983360197372, "maximum": 0.08345958282210852, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1909689395339484, "standard_deviation": 0.04007138989893462, "minimum": 0.1544826912730976, "maximum": 0.26180735552497075, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.660872263328704, "standard_deviation": 0.011601200682043143, "minimum": 5.6444040759551575, "maximum": 5.68537019827216, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6732175285049116, "standard_deviation": 0.06241382402191231, "minimum": 0.5854376397215448, "maximum": 0.7549056410797826, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14578061871253428, "standard_deviation": 0.04484502766643208, "minimum": -0.20783208233605072, "maximum": -0.08388132484855211, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945243377245707, "standard_deviation": 0.0008228654792707758, "minimum": 0.9931459739172929, "maximum": 0.9955298419118784, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.114811402726977, "standard_deviation": 0.01079413242517735, "minimum": 1.0940893332796005, "maximum": 1.1275462436545394, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6518237103232213, "standard_deviation": 0.20410994341682806, "minimum": 0.4573224434957304, "maximum": 1.0350724029423757, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07853995988101531, "standard_deviation": 0.005682305709947833, "minimum": 0.07160102693417515, "maximum": 0.08671696263527612, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982452773523888, "standard_deviation": 0.0003870248143630617, "minimum": 0.997848192876882, "maximum": 0.9988382413439051, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5941295028698732, "standard_deviation": 0.030751123987686358, "minimum": 0.5635980880331011, "maximum": 0.649467156657808, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.72277439305435, "standard_deviation": 0.2807964423102122, "minimum": 0.4524783331199793, "maximum": 1.3204170470643124, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04545009997305419, "standard_deviation": 0.004823058528596035, "minimum": 0.03850401938725692, "maximum": 0.05127587473289705, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3729-6852-0000", "language": "en", "style": "normal", "speaker": "librispeech_3729", "source_utterance_id": "3729-6852-0000", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3729-6852-0000.wav", "source_file_sha256": "fcbc1a0c8ca22a7a621df25a9361397eb07ebd4b77f0ef0daa791547eed2c96d", "decoded_pcm_sha256": "e1524c8eb759928cd93fd86e98694c9a5efccdfa2a0fd948935f558dbffb8843", "original_source_sha256": "3fa96f8e0276d79ec027bb1595215a8bbeb488cefcc1702ff798e91ec8504403", "corpus_entry_id": "librispeech_3729-6852-0000"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.227080821990967, "predicted_mos": 4.227080821990967, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b69e34f93bbf40e5b9b1ae373611ffd5a03d9a210266ecf32682443739b5e402", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.936250390001078, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3729-6852-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "b69e34f93bbf40e5b9b1ae373611ffd5a03d9a210266ecf32682443739b5e402", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "1b783a7c2c383d87de40953142aeba194d79926cf87591d0c55408731d8625ce"}, "rows": [{"case": "librispeech_3729-6852-0000", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.185030221939087, "predicted_mos": 3.185030221939087, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9d277cc6368d957ef432da51505f38a6fff83bdf3f3379e13a6d1158a00acbfb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.714422908000415, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3729-6852-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9d277cc6368d957ef432da51505f38a6fff83bdf3f3379e13a6d1158a00acbfb", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.653890948526216, "全帯域_10ms_rms_envelope_nrmse": 0.06783201868733375, "全帯域_10ms_envelope_correlation": 0.9956602180634839, "全帯域_10ms_modulation_mae_db": 0.2867238215013853, "全帯域_40ms_log_rms_mae_db": 0.3465624266626724, "全帯域_40ms_rms_envelope_nrmse": 0.041283354528556485, "全帯域_40ms_envelope_correlation": 0.9984311685347484, "全帯域_40ms_modulation_mae_db": 0.13961902979286547, "300–1000 Hz_10ms_log_rms_mae_db": 1.2129274393937608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08490729890547986, "300–1000 Hz_10ms_envelope_correlation": 0.9942972394523183, "300–1000 Hz_10ms_modulation_mae_db": 0.21390608133089437, "300–1000 Hz_40ms_log_rms_mae_db": 0.6679553541898389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05114328911115993, "300–1000 Hz_40ms_envelope_correlation": 0.997958497961206, "300–1000 Hz_40ms_modulation_mae_db": 0.19399673539880613, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2484881353278596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11691770097648238, "1000–3000 Hz_10ms_envelope_correlation": 0.9898623969759441, "1000–3000 Hz_10ms_modulation_mae_db": 0.7107927207019439, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7293082984672715, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06929414371742168, "1000–3000 Hz_40ms_envelope_correlation": 0.996672520473826, "1000–3000 Hz_40ms_modulation_mae_db": 0.6888072727872238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4781967841501418, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20774640893409071, "3000–6000 Hz_10ms_envelope_correlation": 0.9690891627247848, "3000–6000 Hz_10ms_modulation_mae_db": 0.3814451189102613, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9212964505026143, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13889988904634057, "3000–6000 Hz_40ms_envelope_correlation": 0.985721951519045, "3000–6000 Hz_40ms_modulation_mae_db": 0.8843176111431843, "mr_stft_db": 5.532902264122277, "band_mae_db": 0.12095574015170468, "rms_gain_error_db": -0.14029928294283397, "output_peak": 0.37418067991798043}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6605565021238262, "全帯域_10ms_rms_envelope_nrmse": 0.06766805807535015, "全帯域_10ms_envelope_correlation": 0.9955480991924187, "全帯域_10ms_modulation_mae_db": 0.17015987193276458, "全帯域_40ms_log_rms_mae_db": 0.35260088706620146, "全帯域_40ms_rms_envelope_nrmse": 0.04163899868372229, "全帯域_40ms_envelope_correlation": 0.998235937511572, "全帯域_40ms_modulation_mae_db": 0.17788265103953493, "300–1000 Hz_10ms_log_rms_mae_db": 1.2242212817367808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08461729131624875, "300–1000 Hz_10ms_envelope_correlation": 0.9941833630245762, "300–1000 Hz_10ms_modulation_mae_db": 0.037707165266943087, "300–1000 Hz_40ms_log_rms_mae_db": 0.6881101912286266, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05464232195376903, "300–1000 Hz_40ms_envelope_correlation": 0.9974213204866318, "300–1000 Hz_40ms_modulation_mae_db": 0.1971003732633302, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2090445058261872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10329437667198886, "1000–3000 Hz_10ms_envelope_correlation": 0.992045072276082, "1000–3000 Hz_10ms_modulation_mae_db": 0.6757571072793493, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7012384654692033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0567273368348576, "1000–3000 Hz_40ms_envelope_correlation": 0.9977644453194454, "1000–3000 Hz_40ms_modulation_mae_db": 0.711436822197815, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4930985970218196, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20139356371400946, "3000–6000 Hz_10ms_envelope_correlation": 0.9688742010755911, "3000–6000 Hz_10ms_modulation_mae_db": 0.3080774230590884, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9539872690627453, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12763189027139385, "3000–6000 Hz_40ms_envelope_correlation": 0.9875758245347906, "3000–6000 Hz_40ms_modulation_mae_db": 0.6697467823977282, "mr_stft_db": 5.544236992874553, "band_mae_db": 0.12040665156686935, "rms_gain_error_db": -0.08154426295367458, "output_peak": 0.38652908183199575}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6742912530513322, "全帯域_10ms_rms_envelope_nrmse": 0.06704930804765018, "全帯域_10ms_envelope_correlation": 0.9956120312392169, "全帯域_10ms_modulation_mae_db": 0.5012735283027988, "全帯域_40ms_log_rms_mae_db": 0.3513853353660039, "全帯域_40ms_rms_envelope_nrmse": 0.03591898933857179, "全帯域_40ms_envelope_correlation": 0.9986858983873939, "全帯域_40ms_modulation_mae_db": 0.46401996360514114, "300–1000 Hz_10ms_log_rms_mae_db": 1.2221259612592152, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08488238205531252, "300–1000 Hz_10ms_envelope_correlation": 0.9941667773109298, "300–1000 Hz_10ms_modulation_mae_db": 0.45272006960419203, "300–1000 Hz_40ms_log_rms_mae_db": 0.6512478734657914, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04356458013598958, "300–1000 Hz_40ms_envelope_correlation": 0.9984150499689429, "300–1000 Hz_40ms_modulation_mae_db": 0.40844572573962007, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2185053774470458, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12382321759587485, "1000–3000 Hz_10ms_envelope_correlation": 0.9882940342745378, "1000–3000 Hz_10ms_modulation_mae_db": 0.5224308983561404, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7514667344865164, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07322344976860336, "1000–3000 Hz_40ms_envelope_correlation": 0.9958023430854732, "1000–3000 Hz_40ms_modulation_mae_db": 0.6062429257741777, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5445061779790485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24599307709951818, "3000–6000 Hz_10ms_envelope_correlation": 0.9546220747597283, "3000–6000 Hz_10ms_modulation_mae_db": 0.4287543092651706, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9546049426900592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16724707199645436, "3000–6000 Hz_40ms_envelope_correlation": 0.9776501573007902, "3000–6000 Hz_40ms_modulation_mae_db": 0.43860679728734703, "mr_stft_db": 5.548631181628841, "band_mae_db": 0.09554881396792604, "rms_gain_error_db": -0.0710637725890907, "output_peak": 0.40081082677938396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6302202137159695, "全帯域_10ms_rms_envelope_nrmse": 0.06978244991240111, "全帯域_10ms_envelope_correlation": 0.9952856136218755, "全帯域_10ms_modulation_mae_db": 0.3851622866075633, "全帯域_40ms_log_rms_mae_db": 0.3314940423836221, "全帯域_40ms_rms_envelope_nrmse": 0.03982095794017512, "全帯域_40ms_envelope_correlation": 0.998427376909239, "全帯域_40ms_modulation_mae_db": 0.423202672272176, "300–1000 Hz_10ms_log_rms_mae_db": 1.18304435521055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08484065407141445, "300–1000 Hz_10ms_envelope_correlation": 0.9941444865543912, "300–1000 Hz_10ms_modulation_mae_db": 0.2731534029551769, "300–1000 Hz_40ms_log_rms_mae_db": 0.6262516310726187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049401254962194656, "300–1000 Hz_40ms_envelope_correlation": 0.9978972018012662, "300–1000 Hz_40ms_modulation_mae_db": 0.35402736279752883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2228101452492877, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11594218708444501, "1000–3000 Hz_10ms_envelope_correlation": 0.9905243575691804, "1000–3000 Hz_10ms_modulation_mae_db": 0.691720075187975, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7248836161260164, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07174156343130335, "1000–3000 Hz_40ms_envelope_correlation": 0.9969738711412156, "1000–3000 Hz_40ms_modulation_mae_db": 0.6582811769801777, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5104962913953377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20693926732741572, "3000–6000 Hz_10ms_envelope_correlation": 0.9668011844083354, "3000–6000 Hz_10ms_modulation_mae_db": 0.47271044521081396, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9435096058701847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12417626124791398, "3000–6000 Hz_40ms_envelope_correlation": 0.9872928661924437, "3000–6000 Hz_40ms_modulation_mae_db": 0.43908283783932883, "mr_stft_db": 5.52312125221529, "band_mae_db": 0.1278741746769049, "rms_gain_error_db": -0.09836696929448668, "output_peak": 0.39989346437158885}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6401487956701547, "全帯域_10ms_rms_envelope_nrmse": 0.06831882069310187, "全帯域_10ms_envelope_correlation": 0.9954182376268554, "全帯域_10ms_modulation_mae_db": 0.18515530818885262, "全帯域_40ms_log_rms_mae_db": 0.34783371820935055, "全帯域_40ms_rms_envelope_nrmse": 0.039486914554726804, "全帯域_40ms_envelope_correlation": 0.9983551208970316, "全帯域_40ms_modulation_mae_db": 0.289839484672715, "300–1000 Hz_10ms_log_rms_mae_db": 1.2611485479725173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08744831621153173, "300–1000 Hz_10ms_envelope_correlation": 0.9937870676544689, "300–1000 Hz_10ms_modulation_mae_db": 0.3559928155440877, "300–1000 Hz_40ms_log_rms_mae_db": 0.6939184013815539, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05020398527038191, "300–1000 Hz_40ms_envelope_correlation": 0.997807572718071, "300–1000 Hz_40ms_modulation_mae_db": 0.409500272545392, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2305280466540565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11918108943195048, "1000–3000 Hz_10ms_envelope_correlation": 0.9891122139238182, "1000–3000 Hz_10ms_modulation_mae_db": 0.15581818638073863, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7120366986245382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06222927053547803, "1000–3000 Hz_40ms_envelope_correlation": 0.996934877158938, "1000–3000 Hz_40ms_modulation_mae_db": 0.1866954115015218, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5544338269559732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22115445335444892, "3000–6000 Hz_10ms_envelope_correlation": 0.9619220565741927, "3000–6000 Hz_10ms_modulation_mae_db": 0.5788959663576637, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9445743044256227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12397715214409386, "3000–6000 Hz_40ms_envelope_correlation": 0.9875393085628815, "3000–6000 Hz_40ms_modulation_mae_db": 0.691852747690508, "mr_stft_db": 5.5406805613392, "band_mae_db": 0.12904712458236356, "rms_gain_error_db": -0.029613638659116998, "output_peak": 0.40541158209773137}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6307303614623885, "全帯域_10ms_rms_envelope_nrmse": 0.06873413352508574, "全帯域_10ms_envelope_correlation": 0.9953678649936498, "全帯域_10ms_modulation_mae_db": 0.571015818611455, "全帯域_40ms_log_rms_mae_db": 0.3149500777432831, "全帯域_40ms_rms_envelope_nrmse": 0.03367357526748635, "全帯域_40ms_envelope_correlation": 0.9988296768398923, "全帯域_40ms_modulation_mae_db": 0.4526643370259585, "300–1000 Hz_10ms_log_rms_mae_db": 1.2277858274747064, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08428092158807612, "300–1000 Hz_10ms_envelope_correlation": 0.9942145082573726, "300–1000 Hz_10ms_modulation_mae_db": 0.6851353129871307, "300–1000 Hz_40ms_log_rms_mae_db": 0.6319815387846006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04521891197362661, "300–1000 Hz_40ms_envelope_correlation": 0.9982260196733833, "300–1000 Hz_40ms_modulation_mae_db": 0.5537975786299137, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1963105680339057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10702847522228116, "1000–3000 Hz_10ms_envelope_correlation": 0.9914325894416226, "1000–3000 Hz_10ms_modulation_mae_db": 0.6438371167098104, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7138026092735658, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05558212225169411, "1000–3000 Hz_40ms_envelope_correlation": 0.9978680474883275, "1000–3000 Hz_40ms_modulation_mae_db": 0.47332154211905336, "3000–6000 Hz_10ms_log_rms_mae_db": 1.572131218515638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22459112332791292, "3000–6000 Hz_10ms_envelope_correlation": 0.9624517700858392, "3000–6000 Hz_10ms_modulation_mae_db": 0.2591545310954791, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9783737920149684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11802136319139034, "3000–6000 Hz_40ms_envelope_correlation": 0.9891021347912091, "3000–6000 Hz_40ms_modulation_mae_db": 0.7352940153269252, "mr_stft_db": 5.531353542370464, "band_mae_db": 0.07260347678189372, "rms_gain_error_db": -0.053557048870337604, "output_peak": 0.3861713603627825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6482490484792124, "全帯域_10ms_rms_envelope_nrmse": 0.06562567466067455, "全帯域_10ms_envelope_correlation": 0.9957987576361168, "全帯域_10ms_modulation_mae_db": 0.42556401782929537, "全帯域_40ms_log_rms_mae_db": 0.3484665927176608, "全帯域_40ms_rms_envelope_nrmse": 0.03944376977528662, "全帯域_40ms_envelope_correlation": 0.9984026047867682, "全帯域_40ms_modulation_mae_db": 0.3726933108988546, "300–1000 Hz_10ms_log_rms_mae_db": 1.2315784191598325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0765864158178592, "300–1000 Hz_10ms_envelope_correlation": 0.9952385181401155, "300–1000 Hz_10ms_modulation_mae_db": 0.4167515141871572, "300–1000 Hz_40ms_log_rms_mae_db": 0.637103346774467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045194071175953844, "300–1000 Hz_40ms_envelope_correlation": 0.9982508748849087, "300–1000 Hz_40ms_modulation_mae_db": 0.47597941109892544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2119841036906855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10775177432347842, "1000–3000 Hz_10ms_envelope_correlation": 0.9913138382602994, "1000–3000 Hz_10ms_modulation_mae_db": 0.9189441291449869, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6880783639375005, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0674197782944414, "1000–3000 Hz_40ms_envelope_correlation": 0.996671505481189, "1000–3000 Hz_40ms_modulation_mae_db": 0.963957720909649, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5532569540924817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2754267085619791, "3000–6000 Hz_10ms_envelope_correlation": 0.9447188191981796, "3000–6000 Hz_10ms_modulation_mae_db": 0.6155243055395364, "3000–6000 Hz_40ms_log_rms_mae_db": 0.952872300488928, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.186346965736897, "3000–6000 Hz_40ms_envelope_correlation": 0.9731941108742337, "3000–6000 Hz_40ms_modulation_mae_db": 0.9849491595890804, "mr_stft_db": 5.52434547830845, "band_mae_db": 0.1356620435525526, "rms_gain_error_db": -0.07124396068392833, "output_peak": 0.3972620470148738}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.648241483204672, "全帯域_10ms_rms_envelope_nrmse": 0.06164154842509124, "全帯域_10ms_envelope_correlation": 0.9963142023726774, "全帯域_10ms_modulation_mae_db": 0.4340242584105735, "全帯域_40ms_log_rms_mae_db": 0.3492928845616936, "全帯域_40ms_rms_envelope_nrmse": 0.03372575017099461, "全帯域_40ms_envelope_correlation": 0.998866362919606, "全帯域_40ms_modulation_mae_db": 0.5039317750298579, "300–1000 Hz_10ms_log_rms_mae_db": 1.2149278078217722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0811349890572859, "300–1000 Hz_10ms_envelope_correlation": 0.9946551090685049, "300–1000 Hz_10ms_modulation_mae_db": 0.46291131980376005, "300–1000 Hz_40ms_log_rms_mae_db": 0.6354466372628002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046616932008542916, "300–1000 Hz_40ms_envelope_correlation": 0.9981370770364095, "300–1000 Hz_40ms_modulation_mae_db": 0.4380352471918856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2457533422199076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10917225040069625, "1000–3000 Hz_10ms_envelope_correlation": 0.9914484008372928, "1000–3000 Hz_10ms_modulation_mae_db": 0.7881843434515954, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7009708230532978, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06412843992619921, "1000–3000 Hz_40ms_envelope_correlation": 0.9974871415441311, "1000–3000 Hz_40ms_modulation_mae_db": 0.9999212996082439, "3000–6000 Hz_10ms_log_rms_mae_db": 1.534463491903306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.206378042432024, "3000–6000 Hz_10ms_envelope_correlation": 0.9670156336666273, "3000–6000 Hz_10ms_modulation_mae_db": 0.4113627102264504, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9255319697468342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11529070625948454, "3000–6000 Hz_40ms_envelope_correlation": 0.9890596136127319, "3000–6000 Hz_40ms_modulation_mae_db": 0.3150088136459781, "mr_stft_db": 5.5215620610388845, "band_mae_db": 0.15832763054250398, "rms_gain_error_db": -0.07982069120187771, "output_peak": 0.36068081766419174}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.331049346081365, "spectral_convergence": 0.2359433423661233, "spectral_convergence_status": "ok", "amplitude_floor": 1.174323222508143e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.326528273202316, "spectral_convergence": 0.23772966364359052, "spectral_convergence_status": "ok", "amplitude_floor": 1.1752576893073956e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.94112917308315, "spectral_convergence": 0.2454369227475397, "spectral_convergence_status": "ok", "amplitude_floor": 9.275961087692859e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.278493760532488e-05, "output_power": 9.5378431867101e-05, "delta_db": 0.1197269781227081, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009129851997232178, "output_power": 0.0009022942170720323, "delta_db": -0.051155633801893974, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007449145920476126, "output_power": 0.0007050033690399904, "delta_db": -0.23915289318712862, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00040544152038629675, "output_power": 0.0003804457739646481, "delta_db": -0.27635457519542833, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011224761509504284, "output_power": 0.00011180750203059115, "delta_db": -0.017061776905067445, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.007390871044891e-05, "output_power": 4.9817649740809476e-05, "delta_db": -0.022282583698001658, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12095574015170468, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14029928294283397, "pitch": null, "output_rms_dbfs": -26.488326110716116, "output_peak": 0.37418067991798043, "reference_rms_dbfs": -26.348026827773285, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.532902264122277}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001795198121335632, "log_rms_mae_db": 0.653890948526216, "rms_envelope_nrmse": 0.06783201868733375, "envelope_correlation": 0.9956602180634839, "rms_gain_error_db": -0.13967409813378667, "reference_mean_power": 0.002319729266015563, "output_mean_power": 0.0022463110414290316}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2030894540936354e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042099620393065603, "output_power": 0.00041872307567868876, "reference_active": true, "delta_db": -0.02351284511040496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010212932714007922, "output_power": 0.00010030428596566833, "reference_active": true, "delta_db": -0.07830979987175535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7753962605616007e-05, "output_power": 1.4909408664784532e-05, "reference_active": true, "delta_db": -0.7583488195219956}], "active_band_count": 3, "active_band_mae_db": 0.2867238215013853}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017377061993537507, "log_rms_mae_db": 0.3465624266626724, "rms_envelope_nrmse": 0.041283354528556485, "envelope_correlation": 0.9984311685347484, "rms_gain_error_db": -0.13946285690364474, "reference_mean_power": 0.002329667623369792, "output_mean_power": 0.0022560445864874966}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1345793651623815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000381470345516612, "output_power": 0.0003791454704488422, "reference_active": true, "delta_db": -0.026549107610529712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.449258478788481e-05, "output_power": 5.3604232751365284e-05, "reference_active": true, "delta_db": -0.07138324365826625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.451644818904224e-07, "output_power": 8.778390965112407e-07, "reference_active": true, "delta_db": -0.32092473810980043}], "active_band_count": 3, "active_band_mae_db": 0.13961902979286547}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001705289920117241, "log_rms_mae_db": 1.2129274393937608, "rms_envelope_nrmse": 0.08490729890547986, "envelope_correlation": 0.9942972394523183, "rms_gain_error_db": -0.15728325582191713, "reference_mean_power": 0.0014327712363456494, "output_mean_power": 0.0013818106279225224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.476686730813627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004044339102676846, "output_power": 0.00040143081102327294, "reference_active": true, "delta_db": -0.03236859570641601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.628475293024379e-05, "output_power": 9.246480766337001e-05, "reference_active": true, "delta_db": -0.17581049705234775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2048085119397813e-05, "output_power": 1.0903454409933254e-05, "reference_active": true, "delta_db": -0.43353915123391934}], "active_band_count": 3, "active_band_mae_db": 0.21390608133089437}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015393777861692204, "log_rms_mae_db": 0.6679553541898389, "rms_envelope_nrmse": 0.05114328911115993, "envelope_correlation": 0.997958497961206, "rms_gain_error_db": -0.1565237321389172, "reference_mean_power": 0.0014385306131629134, "output_mean_power": 0.0013876078090826285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.958921932585308e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037465347734378165, "output_power": 0.00036947733859189763, "reference_active": true, "delta_db": -0.06041961189117494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.015607336171049e-05, "output_power": 4.797626841978735e-05, "reference_active": true, "delta_db": -0.192970632903838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018660953269709e-07, "output_power": 7.434330339824772e-07, "reference_active": true, "delta_db": -0.32859996140140546}], "active_band_count": 3, "active_band_mae_db": 0.19399673539880613}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.828766409891504e-05, "log_rms_mae_db": 1.2484881353278596, "rms_envelope_nrmse": 0.11691770097648238, "envelope_correlation": 0.9898623969759441, "rms_gain_error_db": -0.25450568568191323, "reference_mean_power": 0.00046178379763570874, "output_mean_power": 0.0004354999629729212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.102877053654468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.713589283902264e-05, "output_power": 8.088921205226917e-05, "reference_active": true, "delta_db": -0.3230648061600832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3909625300627975e-05, "output_power": 1.2040901729424137e-05, "reference_active": true, "delta_db": -0.6265641910690334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.457858491278464e-06, "output_power": 3.3950905500072757e-06, "reference_active": true, "delta_db": -1.182749164876715}], "active_band_count": 3, "active_band_mae_db": 0.7107927207019439}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.324411735838957e-05, "log_rms_mae_db": 0.7293082984672715, "rms_envelope_nrmse": 0.06929414371742168, "envelope_correlation": 0.996672520473826, "rms_gain_error_db": -0.2554573089638195, "reference_mean_power": 0.0004639006426222396, "output_mean_power": 0.0004374004678532171}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.97726981805197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.049771362210352e-05, "output_power": 7.516286763345819e-05, "reference_active": true, "delta_db": -0.2978020420800293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.625262232859704e-06, "output_power": 5.856956162207862e-06, "reference_active": true, "delta_db": -0.5353109919213149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8383435618714896e-07, "output_power": 2.1366534294074986e-07, "reference_active": true, "delta_db": -1.2333087843603272}], "active_band_count": 3, "active_band_mae_db": 0.6888072727872238}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.433462630283978e-05, "log_rms_mae_db": 1.4781967841501418, "rms_envelope_nrmse": 0.20774640893409071, "envelope_correlation": 0.9690891627247848, "rms_gain_error_db": 0.26982792718655163, "reference_mean_power": 4.2981917565965076e-05, "output_mean_power": 4.573709529186441e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5620745474492675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.382029340071467e-06, "output_power": 5.811860618764961e-06, "reference_active": true, "delta_db": -0.4064360603051398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1967382600492452e-06, "output_power": 1.3245524125511107e-06, "reference_active": true, "delta_db": 0.440699722451634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059624559400323e-07, "output_power": 5.417990626891858e-07, "reference_active": true, "delta_db": 0.29719957397401}], "active_band_count": 3, "active_band_mae_db": 0.3814451189102613}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.1780456996026785e-05, "log_rms_mae_db": 0.9212964505026143, "rms_envelope_nrmse": 0.13889988904634057, "envelope_correlation": 0.985721951519045, "rms_gain_error_db": 0.26956479357365215, "reference_mean_power": 4.317499760725885e-05, "output_mean_power": 4.5939768411533745e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4490976093596605e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.797984044318825e-06, "output_power": 5.374431919155912e-06, "reference_active": true, "delta_db": -0.32944450057109304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.94257411489492e-07, "output_power": 6.580584084982169e-07, "reference_active": true, "delta_db": 0.44289835907601705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.104930954274029e-08, "output_power": 3.2456288275572345e-08, "reference_active": true, "delta_db": 1.880609973782443}], "active_band_count": 3, "active_band_mae_db": 0.8843176111431843}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8832157523219137, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b26c1d6215929d2424e83d7fded3544fb09c07cbec79e8aa3124f7ef73fbca74", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9905041129448472, "standard_deviation": 0.0013102888895825924, "minimum": 0.9882940342745378, "maximum": 0.992045072276082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.222928028056117, "standard_deviation": 0.018018478277008063, "minimum": 1.1963105680339057, "maximum": 1.2484881353278596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6384355721515675, "standard_deviation": 0.22582649449811118, "minimum": 0.15581818638073863, "maximum": 0.9189441291449869, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11288888396339967, "standard_deviation": 0.007084265758793791, "minimum": 0.10329437667198886, "maximum": 0.12382321759587485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970218439615682, "standard_deviation": 0.0006788467088411544, "minimum": 0.9958023430854732, "maximum": 0.9978680474883275, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7152232011797387, "standard_deviation": 0.019831520628431708, "minimum": 0.6880783639375005, "maximum": 0.7514667344865164, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6610830214847327, "standard_deviation": 0.25998952333269065, "minimum": 0.1866954115015218, "maximum": 0.9999212996082439, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06504326309499985, "standard_deviation": 0.006578360229308041, "minimum": 0.05558212225169411, "maximum": 0.07322344976860336, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9619368628116598, "standard_deviation": 0.008442833758153415, "minimum": 0.9447188191981796, "maximum": 0.9690891627247848, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5300729177517183, "standard_deviation": 0.03288275313032634, "minimum": 1.4781967841501418, "maximum": 1.572131218515638, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.431990601208058, "standard_deviation": 0.12255699824660388, "minimum": 0.2591545310954791, "maximum": 0.6155243055395364, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22370283059392487, "standard_deviation": 0.025391151583125864, "minimum": 0.20139356371400946, "maximum": 0.2754267085619791, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9846419959235158, "standard_deviation": 0.005910562606972149, "minimum": 0.9731941108742337, "maximum": 0.9891021347912091, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9468438293502446, "standard_deviation": 0.01798760023876244, "minimum": 0.9212964505026143, "maximum": 0.9783737920149684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.64485734561501, "standard_deviation": 0.23228461439274053, "minimum": 0.3150088136459781, "maximum": 0.9849491595890804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13769891248674604, "standard_deviation": 0.025643145986082246, "minimum": 0.11529070625948454, "maximum": 0.186346965736897, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943358836828347, "standard_deviation": 0.0004346707441545857, "minimum": 0.9937870676544689, "maximum": 0.9952385181401155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.222219955003642, "standard_deviation": 0.021752711350904647, "minimum": 1.18304435521055, "maximum": 1.2611485479725173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.36228471020991776, "standard_deviation": 0.1930411470930043, "minimum": 0.037707165266943087, "maximum": 0.6851353129871307, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08358728362790108, "standard_deviation": 0.00330510813661122, "minimum": 0.0765864158178592, "maximum": 0.08744831621153173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980142018163525, "standard_deviation": 0.00031342674821412933, "minimum": 0.9974213204866318, "maximum": 0.9984150499689429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6540018717700372, "standard_deviation": 0.026305400773590257, "minimum": 0.6262516310726187, "maximum": 0.6939184013815539, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.37886033833317523, "standard_deviation": 0.12715504380676057, "minimum": 0.19399673539880613, "maximum": 0.5537975786299137, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.048248168323952303, "standard_deviation": 0.003732532410996772, "minimum": 0.04356458013598958, "maximum": 0.05464232195376903, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12005320697783986, "standard_deviation": 0.02592160441934219, "minimum": 0.07260347678189372, "maximum": 0.15832763054250398, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5333541667372454, "standard_deviation": 0.010244361683028363, "minimum": 5.5215620610388845, "maximum": 5.548631181628841, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3888674825050661, "standard_deviation": 0.015265830802410873, "minimum": 0.36068081766419174, "maximum": 0.40541158209773137, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07818870339941833, "standard_deviation": 0.03239060507453475, "minimum": -0.14029928294283397, "maximum": -0.029613638659116998, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956256280932869, "standard_deviation": 0.00032438136939139936, "minimum": 0.9952856136218755, "maximum": 0.9963142023726774, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6482910757792215, "standard_deviation": 0.01492382644583124, "minimum": 0.6302202137159695, "maximum": 0.6742912530513322, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36988486392308606, "standard_deviation": 0.14446695715833516, "minimum": 0.17015987193276458, "maximum": 0.571015818611455, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06708150150333607, "standard_deviation": 0.0025122032461186343, "minimum": 0.06164154842509124, "maximum": 0.06978244991240111, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985292683482814, "standard_deviation": 0.00023324738311891687, "minimum": 0.998235937511572, "maximum": 0.998866362919606, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.342823245588811, "standard_deviation": 0.013021772380259197, "minimum": 0.3149500777432831, "maximum": 0.35260088706620146, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35298165304213797, "standard_deviation": 0.13656072806771377, "minimum": 0.13961902979286547, "maximum": 0.5039317750298579, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038124038782440015, "standard_deviation": 0.0032255383878563378, "minimum": 0.03367357526748635, "maximum": 0.04163899868372229, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0000", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.3046321868896484, "predicted_mos": 3.3046321868896484, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ff97d88d05ce88192fd8f0f586b590ec96f1ea5ed8ee81e7f3123747627c884a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.827160899993032, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.667855043329185, "全帯域_10ms_rms_envelope_nrmse": 0.07050611301728811, "全帯域_10ms_envelope_correlation": 0.9952685375312317, "全帯域_10ms_modulation_mae_db": 0.2700377536327434, "全帯域_40ms_log_rms_mae_db": 0.36041937289632087, "全帯域_40ms_rms_envelope_nrmse": 0.043414785702623776, "全帯域_40ms_envelope_correlation": 0.9982076741027835, "全帯域_40ms_modulation_mae_db": 0.1263363306202235, "300–1000 Hz_10ms_log_rms_mae_db": 1.2268129955338218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08936051966928178, "300–1000 Hz_10ms_envelope_correlation": 0.9936213393091878, "300–1000 Hz_10ms_modulation_mae_db": 0.13804800114595728, "300–1000 Hz_40ms_log_rms_mae_db": 0.6681283619373815, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05381015656039849, "300–1000 Hz_40ms_envelope_correlation": 0.997664837491487, "300–1000 Hz_40ms_modulation_mae_db": 0.16357786445905123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2832285690647145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11779307074311922, "1000–3000 Hz_10ms_envelope_correlation": 0.9896864668586938, "1000–3000 Hz_10ms_modulation_mae_db": 0.6856992487526021, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7323854082779657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06738557540185922, "1000–3000 Hz_40ms_envelope_correlation": 0.9968748174781193, "1000–3000 Hz_40ms_modulation_mae_db": 0.6661965040300242, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4809069111233388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2062881769404953, "3000–6000 Hz_10ms_envelope_correlation": 0.9695500471201909, "3000–6000 Hz_10ms_modulation_mae_db": 0.4067623609441695, "3000–6000 Hz_40ms_log_rms_mae_db": 0.953402082279869, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1350070093374557, "3000–6000 Hz_40ms_envelope_correlation": 0.9866117369614097, "3000–6000 Hz_40ms_modulation_mae_db": 0.8799690668103848, "mr_stft_db": 5.498517796541209, "band_mae_db": 0.12306791969772425, "rms_gain_error_db": -0.13223750646470112, "output_peak": 0.37504155226934444}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6674669456197987, "全帯域_10ms_rms_envelope_nrmse": 0.06830421817955785, "全帯域_10ms_envelope_correlation": 0.9954572768185319, "全帯域_10ms_modulation_mae_db": 0.16152575867307953, "全帯域_40ms_log_rms_mae_db": 0.36134120850097745, "全帯域_40ms_rms_envelope_nrmse": 0.041590703583825744, "全帯域_40ms_envelope_correlation": 0.9982340278598582, "全帯域_40ms_modulation_mae_db": 0.18094103248746388, "300–1000 Hz_10ms_log_rms_mae_db": 1.2381619449378607, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0870086106536169, "300–1000 Hz_10ms_envelope_correlation": 0.9938414483423041, "300–1000 Hz_10ms_modulation_mae_db": 0.030731693609819827, "300–1000 Hz_40ms_log_rms_mae_db": 0.6830867836806124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05521809493101474, "300–1000 Hz_40ms_envelope_correlation": 0.9973559637113482, "300–1000 Hz_40ms_modulation_mae_db": 0.14521080374303966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.225628712780005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10648282157703731, "1000–3000 Hz_10ms_envelope_correlation": 0.9915663120718253, "1000–3000 Hz_10ms_modulation_mae_db": 0.6867885639082468, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6960970264672985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05952661133121208, "1000–3000 Hz_40ms_envelope_correlation": 0.9975520194198744, "1000–3000 Hz_40ms_modulation_mae_db": 0.7015190275971449, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5273161629664567, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19787871128591739, "3000–6000 Hz_10ms_envelope_correlation": 0.970012817787726, "3000–6000 Hz_10ms_modulation_mae_db": 0.3433609686158617, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9961724553697537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12321852495771066, "3000–6000 Hz_40ms_envelope_correlation": 0.9885445785023431, "3000–6000 Hz_40ms_modulation_mae_db": 0.6935666800998002, "mr_stft_db": 5.51849453741728, "band_mae_db": 0.12391237108665065, "rms_gain_error_db": -0.07858135127376975, "output_peak": 0.38656693326032704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6776904631970817, "全帯域_10ms_rms_envelope_nrmse": 0.06932196526521851, "全帯域_10ms_envelope_correlation": 0.9953180223027374, "全帯域_10ms_modulation_mae_db": 0.47873813349786726, "全帯域_40ms_log_rms_mae_db": 0.3553539699200982, "全帯域_40ms_rms_envelope_nrmse": 0.03740993114744887, "全帯域_40ms_envelope_correlation": 0.9985864135714431, "全帯域_40ms_modulation_mae_db": 0.457391667682334, "300–1000 Hz_10ms_log_rms_mae_db": 1.2213578925999555, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08712809070958526, "300–1000 Hz_10ms_envelope_correlation": 0.9938478638977158, "300–1000 Hz_10ms_modulation_mae_db": 0.4105454810035179, "300–1000 Hz_40ms_log_rms_mae_db": 0.6699786620827504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0451188996406775, "300–1000 Hz_40ms_envelope_correlation": 0.9982931988179593, "300–1000 Hz_40ms_modulation_mae_db": 0.40151025574185156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.251825350834983, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12108736763320387, "1000–3000 Hz_10ms_envelope_correlation": 0.9888779604731768, "1000–3000 Hz_10ms_modulation_mae_db": 0.6195218055980183, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7569979118292116, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07074617642377876, "1000–3000 Hz_40ms_envelope_correlation": 0.9961893009494421, "1000–3000 Hz_40ms_modulation_mae_db": 0.6925116391663337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5578298789519265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26517589126916147, "3000–6000 Hz_10ms_envelope_correlation": 0.9466892333602758, "3000–6000 Hz_10ms_modulation_mae_db": 0.4011211361701575, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0041014262119117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18067557924485325, "3000–6000 Hz_40ms_envelope_correlation": 0.9735631437848492, "3000–6000 Hz_40ms_modulation_mae_db": 0.5013538706842549, "mr_stft_db": 5.520303214375786, "band_mae_db": 0.11066560884501521, "rms_gain_error_db": -0.08023987879988959, "output_peak": 0.4017020770148636}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6357562177042323, "全帯域_10ms_rms_envelope_nrmse": 0.07166718058069764, "全帯域_10ms_envelope_correlation": 0.9950045487230501, "全帯域_10ms_modulation_mae_db": 0.3950833012694281, "全帯域_40ms_log_rms_mae_db": 0.33649428358365857, "全帯域_40ms_rms_envelope_nrmse": 0.040552022029041755, "全帯域_40ms_envelope_correlation": 0.9983442322195597, "全帯域_40ms_modulation_mae_db": 0.4390525549560454, "300–1000 Hz_10ms_log_rms_mae_db": 1.1775772042172865, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08836185098925736, "300–1000 Hz_10ms_envelope_correlation": 0.9936375701602893, "300–1000 Hz_10ms_modulation_mae_db": 0.3074494254087346, "300–1000 Hz_40ms_log_rms_mae_db": 0.6140759138207327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05132033863915636, "300–1000 Hz_40ms_envelope_correlation": 0.9977101859599417, "300–1000 Hz_40ms_modulation_mae_db": 0.4876141923897362, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2347352003613112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11591733354267192, "1000–3000 Hz_10ms_envelope_correlation": 0.9905982292638614, "1000–3000 Hz_10ms_modulation_mae_db": 0.7348919969231041, "1000–3000 Hz_40ms_log_rms_mae_db": 0.720167193125686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07212690714481271, "1000–3000 Hz_40ms_envelope_correlation": 0.9970073973309167, "1000–3000 Hz_40ms_modulation_mae_db": 0.6793572506512519, "3000–6000 Hz_10ms_log_rms_mae_db": 1.515999769620045, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19454366117572666, "3000–6000 Hz_10ms_envelope_correlation": 0.9707051808102124, "3000–6000 Hz_10ms_modulation_mae_db": 0.501405774635782, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9790819810168909, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11777889313470294, "3000–6000 Hz_40ms_envelope_correlation": 0.9887184971986737, "3000–6000 Hz_40ms_modulation_mae_db": 0.36250651820900925, "mr_stft_db": 5.491076015253225, "band_mae_db": 0.13089028118553422, "rms_gain_error_db": -0.09047819021857724, "output_peak": 0.3970457756018128}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6556093789851191, "全帯域_10ms_rms_envelope_nrmse": 0.07015923771008208, "全帯域_10ms_envelope_correlation": 0.9951685951464833, "全帯域_10ms_modulation_mae_db": 0.2573591910735146, "全帯域_40ms_log_rms_mae_db": 0.3553822118716622, "全帯域_40ms_rms_envelope_nrmse": 0.04021632152806323, "全帯域_40ms_envelope_correlation": 0.9982982823528661, "全帯域_40ms_modulation_mae_db": 0.345343196229649, "300–1000 Hz_10ms_log_rms_mae_db": 1.283585068437172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08961932857421197, "300–1000 Hz_10ms_envelope_correlation": 0.993468933468159, "300–1000 Hz_10ms_modulation_mae_db": 0.43790962216932733, "300–1000 Hz_40ms_log_rms_mae_db": 0.6863475727799492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0522642477930051, "300–1000 Hz_40ms_envelope_correlation": 0.997622856638528, "300–1000 Hz_40ms_modulation_mae_db": 0.49930988221895806, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2365302734937615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1166821787901459, "1000–3000 Hz_10ms_envelope_correlation": 0.9896048761485436, "1000–3000 Hz_10ms_modulation_mae_db": 0.16406561650323126, "1000–3000 Hz_40ms_log_rms_mae_db": 0.695426562712479, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059004255148918945, "1000–3000 Hz_40ms_envelope_correlation": 0.9973096159351413, "1000–3000 Hz_40ms_modulation_mae_db": 0.23100456868798902, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5607675672857266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21615435000731983, "3000–6000 Hz_10ms_envelope_correlation": 0.9636880629841262, "3000–6000 Hz_10ms_modulation_mae_db": 0.5004223394493713, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9729919172676246, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11889987418240995, "3000–6000 Hz_40ms_envelope_correlation": 0.9886835594547951, "3000–6000 Hz_40ms_modulation_mae_db": 0.5924614788019845, "mr_stft_db": 5.501020737373751, "band_mae_db": 0.1283342159455428, "rms_gain_error_db": -0.03989941235862783, "output_peak": 0.40674393893869965}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6383165601242917, "全帯域_10ms_rms_envelope_nrmse": 0.06713425889972088, "全帯域_10ms_envelope_correlation": 0.9955721115442865, "全帯域_10ms_modulation_mae_db": 0.538692135737382, "全帯域_40ms_log_rms_mae_db": 0.32355325751023345, "全帯域_40ms_rms_envelope_nrmse": 0.032348721640690865, "全帯域_40ms_envelope_correlation": 0.9989038633386754, "全帯域_40ms_modulation_mae_db": 0.4108615302519969, "300–1000 Hz_10ms_log_rms_mae_db": 1.2406759512227157, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08479844119275079, "300–1000 Hz_10ms_envelope_correlation": 0.9941508138468188, "300–1000 Hz_10ms_modulation_mae_db": 0.6495345994813255, "300–1000 Hz_40ms_log_rms_mae_db": 0.630063489647966, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04410329727065294, "300–1000 Hz_40ms_envelope_correlation": 0.9983071238096941, "300–1000 Hz_40ms_modulation_mae_db": 0.5460957547173376, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2111353426281124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10511282632767181, "1000–3000 Hz_10ms_envelope_correlation": 0.9917147307283447, "1000–3000 Hz_10ms_modulation_mae_db": 0.6047832189714296, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7140513958477187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05588196281994626, "1000–3000 Hz_40ms_envelope_correlation": 0.9977838694056017, "1000–3000 Hz_40ms_modulation_mae_db": 0.44070964503941173, "3000–6000 Hz_10ms_log_rms_mae_db": 1.584181608640673, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2229680003972908, "3000–6000 Hz_10ms_envelope_correlation": 0.9633720664458406, "3000–6000 Hz_10ms_modulation_mae_db": 0.25009145229013996, "3000–6000 Hz_40ms_log_rms_mae_db": 1.02365902151557, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12006622467739485, "3000–6000 Hz_40ms_envelope_correlation": 0.9889901723458936, "3000–6000 Hz_40ms_modulation_mae_db": 0.6749671405711436, "mr_stft_db": 5.5063379197980415, "band_mae_db": 0.07077047495535918, "rms_gain_error_db": -0.03250747591549959, "output_peak": 0.3878872333006618}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6506911700024957, "全帯域_10ms_rms_envelope_nrmse": 0.06708453930074798, "全帯域_10ms_envelope_correlation": 0.9956070968231123, "全帯域_10ms_modulation_mae_db": 0.43978678806614774, "全帯域_40ms_log_rms_mae_db": 0.3431983315758041, "全帯域_40ms_rms_envelope_nrmse": 0.03953840770617116, "全帯域_40ms_envelope_correlation": 0.9983943518710867, "全帯域_40ms_modulation_mae_db": 0.4099194645155299, "300–1000 Hz_10ms_log_rms_mae_db": 1.2371965135867766, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0770584069455065, "300–1000 Hz_10ms_envelope_correlation": 0.995177756427454, "300–1000 Hz_10ms_modulation_mae_db": 0.45500592204501605, "300–1000 Hz_40ms_log_rms_mae_db": 0.6310376297157541, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043391244998839215, "300–1000 Hz_40ms_envelope_correlation": 0.9983880378774781, "300–1000 Hz_40ms_modulation_mae_db": 0.5211992977640754, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2421718783510936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10983738857666031, "1000–3000 Hz_10ms_envelope_correlation": 0.9909974701143835, "1000–3000 Hz_10ms_modulation_mae_db": 0.9230938764554265, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6843091179007794, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0653875479385214, "1000–3000 Hz_40ms_envelope_correlation": 0.996941896753001, "1000–3000 Hz_40ms_modulation_mae_db": 0.9635291605106616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5794734153943795, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2832349765696572, "3000–6000 Hz_10ms_envelope_correlation": 0.9423274226007626, "3000–6000 Hz_10ms_modulation_mae_db": 0.655767611380592, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0093652698958955, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19404250435233913, "3000–6000 Hz_40ms_envelope_correlation": 0.9715097505625164, "3000–6000 Hz_40ms_modulation_mae_db": 0.9685508332537337, "mr_stft_db": 5.496535149506975, "band_mae_db": 0.14003782234696333, "rms_gain_error_db": -0.0714150234316316, "output_peak": 0.3939272444841173}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6479397088792633, "全帯域_10ms_rms_envelope_nrmse": 0.0609584103735142, "全帯域_10ms_envelope_correlation": 0.9963981127416891, "全帯域_10ms_modulation_mae_db": 0.41457524391529316, "全帯域_40ms_log_rms_mae_db": 0.35116328094965404, "全帯域_40ms_rms_envelope_nrmse": 0.03317514346559654, "全帯域_40ms_envelope_correlation": 0.9989066117610893, "全帯域_40ms_modulation_mae_db": 0.4869880139167179, "300–1000 Hz_10ms_log_rms_mae_db": 1.2151376396483595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07900114431866012, "300–1000 Hz_10ms_envelope_correlation": 0.9949307593535219, "300–1000 Hz_10ms_modulation_mae_db": 0.45762772304475074, "300–1000 Hz_40ms_log_rms_mae_db": 0.6431520075601987, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04360159087864067, "300–1000 Hz_40ms_envelope_correlation": 0.9983723298989378, "300–1000 Hz_40ms_modulation_mae_db": 0.4502685681195439, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2656364226343646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10739962385880522, "1000–3000 Hz_10ms_envelope_correlation": 0.991741569135483, "1000–3000 Hz_10ms_modulation_mae_db": 0.6636308135195389, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6989607587518872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062332001458993036, "1000–3000 Hz_40ms_envelope_correlation": 0.9976691068230769, "1000–3000 Hz_40ms_modulation_mae_db": 0.8399841861128897, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5586266559082695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20562273477260945, "3000–6000 Hz_10ms_envelope_correlation": 0.9672834058993286, "3000–6000 Hz_10ms_modulation_mae_db": 0.4438075954341314, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9696529930456103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11286808179160247, "3000–6000 Hz_40ms_envelope_correlation": 0.9895250045816343, "3000–6000 Hz_40ms_modulation_mae_db": 0.29054999252789243, "mr_stft_db": 5.491019504232535, "band_mae_db": 0.15186868763878894, "rms_gain_error_db": -0.08029330642726172, "output_peak": 0.3593879760031926}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.296064865048752, "spectral_convergence": 0.23801059090052604, "spectral_convergence_status": "ok", "amplitude_floor": 1.174323222508143e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.294581682420186, "spectral_convergence": 0.23988760467144776, "spectral_convergence_status": "ok", "amplitude_floor": 1.1752576893073956e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.904906842154687, "spectral_convergence": 0.24733750723641243, "spectral_convergence_status": "ok", "amplitude_floor": 9.275961087692859e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.278493760532488e-05, "output_power": 9.528829042844559e-05, "delta_db": 0.11562055285413447, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009129851997232178, "output_power": 0.0009040350513855508, "delta_db": -0.042784679695809597, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007449145920476126, "output_power": 0.0007076315114865725, "delta_db": -0.2229931749919377, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00040544152038629675, "output_power": 0.0003812026044084008, "delta_db": -0.2677236267690419, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011224761509504284, "output_power": 0.00011127775348600276, "delta_db": -0.037687730119634145, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.007390871044891e-05, "output_power": 4.948250963114017e-05, "delta_db": -0.05159775375578783, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12306791969772425, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13223750646470112, "pitch": null, "output_rms_dbfs": -26.480264334237983, "output_peak": 0.37504155226934444, "reference_rms_dbfs": -26.348026827773285, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.498517796541209}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001795198121335632, "log_rms_mae_db": 0.667855043329185, "rms_envelope_nrmse": 0.07050611301728811, "envelope_correlation": 0.9952685375312317, "rms_gain_error_db": -0.13156567340970976, "reference_mean_power": 0.002319729266015563, "output_mean_power": 0.00225050889759587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2030894540936354e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042099620393065603, "output_power": 0.0004219464393754413, "reference_active": true, "delta_db": 0.00979146536870943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010212932714007922, "output_power": 0.00010087993286156293, "reference_active": true, "delta_db": -0.0534568611946808}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7753962605616007e-05, "output_power": 1.4948885221979523e-05, "reference_active": true, "delta_db": -0.74686493433484}], "active_band_count": 3, "active_band_mae_db": 0.2700377536327434}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017377061993537507, "log_rms_mae_db": 0.36041937289632087, "rms_envelope_nrmse": 0.043414785702623776, "envelope_correlation": 0.9982076741027835, "rms_gain_error_db": -0.13125652672961108, "reference_mean_power": 0.002329667623369792, "output_mean_power": 0.0022603115873893087}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1345793651623815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000381470345516612, "output_power": 0.00038216687832666343, "reference_active": true, "delta_db": 0.007922621640631485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.449258478788481e-05, "output_power": 5.388898399730585e-05, "reference_active": true, "delta_db": -0.048374129625755685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.451644818904224e-07, "output_power": 8.774778631185956e-07, "reference_active": true, "delta_db": -0.32271224059428333}], "active_band_count": 3, "active_band_mae_db": 0.1263363306202235}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001705289920117241, "log_rms_mae_db": 1.2268129955338218, "rms_envelope_nrmse": 0.08936051966928178, "envelope_correlation": 0.9936213393091878, "rms_gain_error_db": -0.14307246805325685, "reference_mean_power": 0.0014327712363456494, "output_mean_power": 0.0013863395318082455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.476686730813627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004044339102676846, "output_power": 0.00040584868598710164, "reference_active": true, "delta_db": 0.015165817834423627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.628475293024379e-05, "output_power": 9.341621204432038e-05, "reference_active": true, "delta_db": -0.13135267326877326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2048085119397813e-05, "output_power": 1.1328058311977147e-05, "reference_active": true, "delta_db": -0.2676255123346749}], "active_band_count": 3, "active_band_mae_db": 0.13804800114595728}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015393777861692204, "log_rms_mae_db": 0.6681283619373815, "rms_envelope_nrmse": 0.05381015656039849, "envelope_correlation": 0.997664837491487, "rms_gain_error_db": -0.1421971325260839, "reference_mean_power": 0.0014385306131629134, "output_mean_power": 0.001392192837953319}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.958921932585308e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037465347734378165, "output_power": 0.00037357147168271486, "reference_active": true, "delta_db": -0.012560645462530934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.015607336171049e-05, "output_power": 4.839082087506918e-05, "reference_active": true, "delta_db": -0.15560539635197415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018660953269709e-07, "output_power": 7.444663901562445e-07, "reference_active": true, "delta_db": -0.3225675515626486}], "active_band_count": 3, "active_band_mae_db": 0.16357786445905123}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.828766409891504e-05, "log_rms_mae_db": 1.2832285690647145, "rms_envelope_nrmse": 0.11779307074311922, "envelope_correlation": 0.9896864668586938, "rms_gain_error_db": -0.2506082694564376, "reference_mean_power": 0.00046178379763570874, "output_mean_power": 0.00043589096182768914}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.102877053654468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.713589283902264e-05, "output_power": 8.122768552552316e-05, "reference_active": true, "delta_db": -0.304930067895109}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3909625300627975e-05, "output_power": 1.1806723496595733e-05, "reference_active": true, "delta_db": -0.7118603851387768}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.457858491278464e-06, "output_power": 3.508290512145289e-06, "reference_active": true, "delta_db": -1.0403072932239206}], "active_band_count": 3, "active_band_mae_db": 0.6856992487526021}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.324411735838957e-05, "log_rms_mae_db": 0.7323854082779657, "rms_envelope_nrmse": 0.06738557540185922, "envelope_correlation": 0.9968748174781193, "rms_gain_error_db": -0.25149066799909237, "reference_mean_power": 0.0004639006426222396, "output_mean_power": 0.00043780015130898787}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.97726981805197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.049771362210352e-05, "output_power": 7.533904309947794e-05, "reference_active": true, "delta_db": -0.2876344545306132}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.625262232859704e-06, "output_power": 5.78519794655196e-06, "reference_active": true, "delta_db": -0.5888484927268763}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8383435618714896e-07, "output_power": 2.1920694330802884e-07, "reference_active": true, "delta_db": -1.122106564832583}], "active_band_count": 3, "active_band_mae_db": 0.6661965040300242}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.433462630283978e-05, "log_rms_mae_db": 1.4809069111233388, "rms_envelope_nrmse": 0.2062881769404953, "envelope_correlation": 0.9695500471201909, "rms_gain_error_db": 0.2351773853083106, "reference_mean_power": 4.2981917565965076e-05, "output_mean_power": 4.537363001806105e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5620745474492675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.382029340071467e-06, "output_power": 5.672940026093883e-06, "reference_active": true, "delta_db": -0.5115060418238028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1967382600492452e-06, "output_power": 1.343435436210999e-06, "reference_active": true, "delta_db": 0.502176238685713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059624559400323e-07, "output_power": 5.306140827783721e-07, "reference_active": true, "delta_db": 0.20660480232299275}], "active_band_count": 3, "active_band_mae_db": 0.4067623609441695}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.1780456996026785e-05, "log_rms_mae_db": 0.953402082279869, "rms_envelope_nrmse": 0.1350070093374557, "envelope_correlation": 0.9866117369614097, "rms_gain_error_db": 0.23527379566292034, "reference_mean_power": 4.317499760725885e-05, "output_mean_power": 4.5578465723812654e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4490976093596605e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.797984044318825e-06, "output_power": 5.233861057554808e-06, "reference_active": true, "delta_db": -0.44454826556994587}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.94257411489492e-07, "output_power": 6.599566972468419e-07, "reference_active": true, "delta_db": 0.4554083355556333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.104930954274029e-08, "output_power": 3.142193418655778e-08, "reference_active": true, "delta_db": 1.739950599305575}], "active_band_count": 3, "active_band_mae_db": 0.8799690668103848}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8830020338687826, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f6982902545f00beba53e9f717efc523d1aba44916b8442ad8bb20ee5f1b9f6b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.990598451849289, "standard_deviation": 0.001097254012333375, "minimum": 0.9888779604731768, "maximum": 0.991741569135483, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2438614687685432, "standard_deviation": 0.022774243776779148, "minimum": 1.2111353426281124, "maximum": 1.2832285690647145, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6353093925789497, "standard_deviation": 0.2144693589456678, "minimum": 0.16406561650323126, "maximum": 0.9230938764554265, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11253907638116445, "standard_deviation": 0.0060331957941972296, "minimum": 0.10511282632767181, "maximum": 0.12108736763320387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971660030118967, "standard_deviation": 0.0005235348228893419, "minimum": 0.9961893009494421, "maximum": 0.9977838694056017, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7122994218641283, "standard_deviation": 0.023829636814473622, "minimum": 0.6843091179007794, "maximum": 0.7569979118292116, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6518514977244634, "standard_deviation": 0.22656199083269962, "minimum": 0.23100456868798902, "maximum": 0.9635291605106616, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0640488797085053, "standard_deviation": 0.005838043115706987, "minimum": 0.05588196281994626, "maximum": 0.07212690714481271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9617035296260579, "standard_deviation": 0.01102180859793245, "minimum": 0.9423274226007626, "maximum": 0.9707051808102124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.545637746236352, "standard_deviation": 0.035012088283646124, "minimum": 1.4809069111233388, "maximum": 1.584181608640673, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4378424048650257, "standard_deviation": 0.12069744465027167, "minimum": 0.25009145229013996, "maximum": 0.655767611380592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22398331280227227, "standard_deviation": 0.03266339949206724, "minimum": 0.19454366117572666, "maximum": 0.2832349765696572, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9845183054240143, "standard_deviation": 0.007463118640876943, "minimum": 0.9715097505625164, "maximum": 0.9895250045816343, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9885533933253907, "standard_deviation": 0.02357370737937321, "minimum": 0.953402082279869, "maximum": 1.02365902151557, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6204906976197754, "standard_deviation": 0.23518339059732488, "minimum": 0.29054999252789243, "maximum": 0.9685508332537337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13781958645980863, "standard_deviation": 0.03143269122243484, "minimum": 0.11286808179160247, "maximum": 0.19404250435233913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9940845606006814, "standard_deviation": 0.0006351624113975315, "minimum": 0.993468933468159, "maximum": 0.995177756427454, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2300631512729936, "standard_deviation": 0.029658997750412284, "minimum": 1.1775772042172865, "maximum": 1.283585068437172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.36085655848855613, "standard_deviation": 0.19692215192900914, "minimum": 0.030731693609819827, "maximum": 0.6495345994813255, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08529204913160884, "standard_deviation": 0.004760418818340261, "minimum": 0.0770584069455065, "maximum": 0.08961932857421197, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979643167756718, "standard_deviation": 0.00041621648393204154, "minimum": 0.9973559637113482, "maximum": 0.9983880378774781, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6532338026531681, "standard_deviation": 0.02713584032171234, "minimum": 0.6140759138207327, "maximum": 0.6863475727799492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4018483273941992, "standard_deviation": 0.1589639576140898, "minimum": 0.14521080374303966, "maximum": 0.5460957547173376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.048603483839048126, "standard_deviation": 0.0050174749847385295, "minimum": 0.043391244998839215, "maximum": 0.05521809493101474, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12244342271269733, "standard_deviation": 0.024180329114731794, "minimum": 0.07077047495535918, "maximum": 0.15186868763878894, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.502913109312351, "standard_deviation": 0.011352022082416067, "minimum": 5.491019504232535, "maximum": 5.520303214375786, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3885378413591274, "standard_deviation": 0.015317412778748395, "minimum": 0.3593879760031926, "maximum": 0.40674393893869965, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0757065181112448, "standard_deviation": 0.03077692221816495, "minimum": -0.13223750646470112, "maximum": -0.03250747591549959, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954742877038902, "standard_deviation": 0.00042456776263873064, "minimum": 0.9950045487230501, "maximum": 0.9963981127416891, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6551656859801834, "standard_deviation": 0.014893717611155252, "minimum": 0.6357562177042323, "maximum": 0.6776904631970817, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36947478823318197, "standard_deviation": 0.12756345398581928, "minimum": 0.16152575867307953, "maximum": 0.538692135737382, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0681419904158534, "standard_deviation": 0.003321822529137348, "minimum": 0.0609584103735142, "maximum": 0.07166718058069764, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984844321346702, "standard_deviation": 0.00028442370448500024, "minimum": 0.9982076741027835, "maximum": 0.9989066117610893, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3483632396010511, "standard_deviation": 0.01310854761314517, "minimum": 0.32355325751023345, "maximum": 0.36134120850097745, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35710422383249507, "standard_deviation": 0.13295961214335006, "minimum": 0.1263363306202235, "maximum": 0.4869880139167179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038530754600432746, "standard_deviation": 0.0039517702653511415, "minimum": 0.032348721640690865, "maximum": 0.043414785702623776, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0000", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.0685527324676514, "predicted_mos": 3.0685527324676514, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "05c00adb844adfa44beb17ccfddea9bff1768720754cf2f96988feef3d42a72a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8953036996535957, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.643660752254516, "全帯域_10ms_rms_envelope_nrmse": 0.06587883891847013, "全帯域_10ms_envelope_correlation": 0.9958306738114677, "全帯域_10ms_modulation_mae_db": 0.3766182175641, "全帯域_40ms_log_rms_mae_db": 0.3383763907715841, "全帯域_40ms_rms_envelope_nrmse": 0.036424981709817214, "全帯域_40ms_envelope_correlation": 0.9987322084675586, "全帯域_40ms_modulation_mae_db": 0.40216771381605615, "300–1000 Hz_10ms_log_rms_mae_db": 1.3800444574816613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0818238232557155, "300–1000 Hz_10ms_envelope_correlation": 0.9945698740958738, "300–1000 Hz_10ms_modulation_mae_db": 0.3986862982835939, "300–1000 Hz_40ms_log_rms_mae_db": 0.7530054809187184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043466647682219264, "300–1000 Hz_40ms_envelope_correlation": 0.9984049061118984, "300–1000 Hz_40ms_modulation_mae_db": 0.5414127201999732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.366032943567076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12291122228970648, "1000–3000 Hz_10ms_envelope_correlation": 0.9889797775935759, "1000–3000 Hz_10ms_modulation_mae_db": 0.8598225779781891, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7897475846465645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07297894434084819, "1000–3000 Hz_40ms_envelope_correlation": 0.9965438346367679, "1000–3000 Hz_40ms_modulation_mae_db": 0.8485918532554292, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7568104915605816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20636620137146866, "3000–6000 Hz_10ms_envelope_correlation": 0.9676229969628386, "3000–6000 Hz_10ms_modulation_mae_db": 0.37342043672801695, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0420911494144105, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1271584428928323, "3000–6000 Hz_40ms_envelope_correlation": 0.9868841684259793, "3000–6000 Hz_40ms_modulation_mae_db": 0.588166089073396, "mr_stft_db": 5.746893517311752, "band_mae_db": 0.11017987474078927, "rms_gain_error_db": -0.10677432845016652, "output_peak": 0.3826377382935182}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6393507083407193, "全帯域_10ms_rms_envelope_nrmse": 0.0652156753999633, "全帯域_10ms_envelope_correlation": 0.9960834073313737, "全帯域_10ms_modulation_mae_db": 0.6676842872215015, "全帯域_40ms_log_rms_mae_db": 0.3590758690808248, "全帯域_40ms_rms_envelope_nrmse": 0.040021395443173395, "全帯域_40ms_envelope_correlation": 0.998628799341874, "全帯域_40ms_modulation_mae_db": 0.5807885808877654, "300–1000 Hz_10ms_log_rms_mae_db": 1.3761493304905514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08795792563768998, "300–1000 Hz_10ms_envelope_correlation": 0.993924064827855, "300–1000 Hz_10ms_modulation_mae_db": 0.6871263899098544, "300–1000 Hz_40ms_log_rms_mae_db": 0.8055733325513869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05155299015875455, "300–1000 Hz_40ms_envelope_correlation": 0.9980003030422254, "300–1000 Hz_40ms_modulation_mae_db": 0.8084865399044764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4074733143513232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11957995148459066, "1000–3000 Hz_10ms_envelope_correlation": 0.9896568907947701, "1000–3000 Hz_10ms_modulation_mae_db": 0.8723705951001529, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8167751042935609, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06800929678476934, "1000–3000 Hz_40ms_envelope_correlation": 0.9971887238673902, "1000–3000 Hz_40ms_modulation_mae_db": 0.8762102284976808, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8605118613063925, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21914787287519977, "3000–6000 Hz_10ms_envelope_correlation": 0.9710143845471602, "3000–6000 Hz_10ms_modulation_mae_db": 0.5772970581754815, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0863043278839748, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15047378615903464, "3000–6000 Hz_40ms_envelope_correlation": 0.9881298207829947, "3000–6000 Hz_40ms_modulation_mae_db": 0.8940263203104942, "mr_stft_db": 5.728051455051116, "band_mae_db": 0.17387289163785702, "rms_gain_error_db": -0.16077581279025696, "output_peak": 0.3861123687370104}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6271180769313204, "全帯域_10ms_rms_envelope_nrmse": 0.06308494954579544, "全帯域_10ms_envelope_correlation": 0.9961022910409866, "全帯域_10ms_modulation_mae_db": 0.36036073952871045, "全帯域_40ms_log_rms_mae_db": 0.3290352197558284, "全帯域_40ms_rms_envelope_nrmse": 0.03492244546159312, "全帯域_40ms_envelope_correlation": 0.9987324154117966, "全帯域_40ms_modulation_mae_db": 0.1304162763266198, "300–1000 Hz_10ms_log_rms_mae_db": 1.3564111255891884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08225650382698833, "300–1000 Hz_10ms_envelope_correlation": 0.9944959445806719, "300–1000 Hz_10ms_modulation_mae_db": 0.27175249533208384, "300–1000 Hz_40ms_log_rms_mae_db": 0.7851222149116378, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0489543508661651, "300–1000 Hz_40ms_envelope_correlation": 0.9979173074695212, "300–1000 Hz_40ms_modulation_mae_db": 0.2300979261157771, "1000–3000 Hz_10ms_log_rms_mae_db": 1.348135921197326, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12275803615529472, "1000–3000 Hz_10ms_envelope_correlation": 0.98885513815257, "1000–3000 Hz_10ms_modulation_mae_db": 0.5847037791131783, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7789485405616146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06815101343447966, "1000–3000 Hz_40ms_envelope_correlation": 0.9969403098924892, "1000–3000 Hz_40ms_modulation_mae_db": 0.5019484570911877, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8579813634405726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24642251360037984, "3000–6000 Hz_10ms_envelope_correlation": 0.9570268179772148, "3000–6000 Hz_10ms_modulation_mae_db": 0.3395689012923057, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0917200343193116, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14904400298982304, "3000–6000 Hz_40ms_envelope_correlation": 0.9841076521869239, "3000–6000 Hz_40ms_modulation_mae_db": 0.6710323516846423, "mr_stft_db": 5.7463857361373725, "band_mae_db": 0.10782508966439446, "rms_gain_error_db": -0.051624133990681983, "output_peak": 0.4041283336376955}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6459616463116331, "全帯域_10ms_rms_envelope_nrmse": 0.06572632059551926, "全帯域_10ms_envelope_correlation": 0.9957574376715126, "全帯域_10ms_modulation_mae_db": 0.11266548745777294, "全帯域_40ms_log_rms_mae_db": 0.3428913910170045, "全帯域_40ms_rms_envelope_nrmse": 0.03827416035998451, "全帯域_40ms_envelope_correlation": 0.9984581791427187, "全帯域_40ms_modulation_mae_db": 0.18179874849424507, "300–1000 Hz_10ms_log_rms_mae_db": 1.4478542683664157, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07923743250001858, "300–1000 Hz_10ms_envelope_correlation": 0.9948893741851246, "300–1000 Hz_10ms_modulation_mae_db": 0.40387903622615323, "300–1000 Hz_40ms_log_rms_mae_db": 0.830829135040896, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04756630777181341, "300–1000 Hz_40ms_envelope_correlation": 0.9980366596384791, "300–1000 Hz_40ms_modulation_mae_db": 0.3091645918345651, "1000–3000 Hz_10ms_log_rms_mae_db": 1.335968202939294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12214601847188607, "1000–3000 Hz_10ms_envelope_correlation": 0.9886443364383573, "1000–3000 Hz_10ms_modulation_mae_db": 0.6459049266019261, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7954952648209539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0653237038662853, "1000–3000 Hz_40ms_envelope_correlation": 0.9968006069828941, "1000–3000 Hz_40ms_modulation_mae_db": 0.6510833305527693, "3000–6000 Hz_10ms_log_rms_mae_db": 1.852644849662578, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22953885768478813, "3000–6000 Hz_10ms_envelope_correlation": 0.9608624531765138, "3000–6000 Hz_10ms_modulation_mae_db": 0.44748707852217134, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0782276215237987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12429883316634253, "3000–6000 Hz_40ms_envelope_correlation": 0.98794777783182, "3000–6000 Hz_40ms_modulation_mae_db": 0.6918718191102098, "mr_stft_db": 5.751863565677073, "band_mae_db": 0.0737160157405836, "rms_gain_error_db": -0.04215131246916845, "output_peak": 0.36440210852347477}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6533160184453277, "全帯域_10ms_rms_envelope_nrmse": 0.06332019426995783, "全帯域_10ms_envelope_correlation": 0.9960914545170956, "全帯域_10ms_modulation_mae_db": 0.3102880553002691, "全帯域_40ms_log_rms_mae_db": 0.35384078091202803, "全帯域_40ms_rms_envelope_nrmse": 0.033765062855847565, "全帯域_40ms_envelope_correlation": 0.9988447939405023, "全帯域_40ms_modulation_mae_db": 0.24680877732346315, "300–1000 Hz_10ms_log_rms_mae_db": 1.4045452413072474, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08483646543479066, "300–1000 Hz_10ms_envelope_correlation": 0.9941575242121412, "300–1000 Hz_10ms_modulation_mae_db": 0.2844565619829457, "300–1000 Hz_40ms_log_rms_mae_db": 0.8168325057606557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04780961863357073, "300–1000 Hz_40ms_envelope_correlation": 0.9980539884169465, "300–1000 Hz_40ms_modulation_mae_db": 0.23770784519640645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3694060206878553, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11735223573752634, "1000–3000 Hz_10ms_envelope_correlation": 0.9895370456154514, "1000–3000 Hz_10ms_modulation_mae_db": 0.7776658253226518, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7934055502720131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06346483954164955, "1000–3000 Hz_40ms_envelope_correlation": 0.9969850326870844, "1000–3000 Hz_40ms_modulation_mae_db": 0.8223365757615545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8027260203356577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21515676975554573, "3000–6000 Hz_10ms_envelope_correlation": 0.9652756909650482, "3000–6000 Hz_10ms_modulation_mae_db": 0.30080802040625154, "3000–6000 Hz_40ms_log_rms_mae_db": 1.064471254923309, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12410715388664159, "3000–6000 Hz_40ms_envelope_correlation": 0.98777287799049, "3000–6000 Hz_40ms_modulation_mae_db": 0.39779218107097253, "mr_stft_db": 5.744297089550375, "band_mae_db": 0.09419104264207306, "rms_gain_error_db": -0.06800869984083296, "output_peak": 0.3998517453448594}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.605942498002597, "全帯域_10ms_rms_envelope_nrmse": 0.06018612912104554, "全帯域_10ms_envelope_correlation": 0.9965207189100453, "全帯域_10ms_modulation_mae_db": 0.37723956620063026, "全帯域_40ms_log_rms_mae_db": 0.3217177787030861, "全帯域_40ms_rms_envelope_nrmse": 0.03191923313788511, "全帯域_40ms_envelope_correlation": 0.999024665039055, "全帯域_40ms_modulation_mae_db": 0.41341187697912507, "300–1000 Hz_10ms_log_rms_mae_db": 1.3222456989119575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0786028413944275, "300–1000 Hz_10ms_envelope_correlation": 0.9950717285271485, "300–1000 Hz_10ms_modulation_mae_db": 0.3388526985654285, "300–1000 Hz_40ms_log_rms_mae_db": 0.7480888513595048, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042275734328024395, "300–1000 Hz_40ms_envelope_correlation": 0.9985862048621785, "300–1000 Hz_40ms_modulation_mae_db": 0.5288503449972097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3521385153407797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11813360698922595, "1000–3000 Hz_10ms_envelope_correlation": 0.9894439787117854, "1000–3000 Hz_10ms_modulation_mae_db": 0.7403856354657058, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7902758434124084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06535127018298184, "1000–3000 Hz_40ms_envelope_correlation": 0.9968564640888224, "1000–3000 Hz_40ms_modulation_mae_db": 0.821310525379913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8204386786594717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19656047164017132, "3000–6000 Hz_10ms_envelope_correlation": 0.9704361973117899, "3000–6000 Hz_10ms_modulation_mae_db": 0.41053070671583636, "3000–6000 Hz_40ms_log_rms_mae_db": 1.077227921641739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12014446251178167, "3000–6000 Hz_40ms_envelope_correlation": 0.9881993245800811, "3000–6000 Hz_40ms_modulation_mae_db": 0.4174530925764268, "mr_stft_db": 5.720682137751884, "band_mae_db": 0.1364428704266232, "rms_gain_error_db": -0.09005562672629, "output_peak": 0.38464239439954123}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6549944701599963, "全帯域_10ms_rms_envelope_nrmse": 0.06495776357350055, "全帯域_10ms_envelope_correlation": 0.9958602791996544, "全帯域_10ms_modulation_mae_db": 0.24348033868655047, "全帯域_40ms_log_rms_mae_db": 0.33496689681077196, "全帯域_40ms_rms_envelope_nrmse": 0.034146293829040164, "全帯域_40ms_envelope_correlation": 0.9987645440017443, "全帯域_40ms_modulation_mae_db": 0.09334047971239656, "300–1000 Hz_10ms_log_rms_mae_db": 1.3552095942578577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08173973876588297, "300–1000 Hz_10ms_envelope_correlation": 0.9945941606601771, "300–1000 Hz_10ms_modulation_mae_db": 0.18699191855421202, "300–1000 Hz_40ms_log_rms_mae_db": 0.7499291412744721, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0454120016202676, "300–1000 Hz_40ms_envelope_correlation": 0.9982167721640521, "300–1000 Hz_40ms_modulation_mae_db": 0.12609334196150285, "1000–3000 Hz_10ms_log_rms_mae_db": 1.38647597071174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11700446635367102, "1000–3000 Hz_10ms_envelope_correlation": 0.9896129563225148, "1000–3000 Hz_10ms_modulation_mae_db": 0.7504395091738001, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8419546169093938, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0629521641096851, "1000–3000 Hz_40ms_envelope_correlation": 0.9970527500761991, "1000–3000 Hz_40ms_modulation_mae_db": 0.691002556831477, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8862248439448959, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24262636336914775, "3000–6000 Hz_10ms_envelope_correlation": 0.9559864531100283, "3000–6000 Hz_10ms_modulation_mae_db": 0.6412446584960461, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1626078507258288, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1350490850719951, "3000–6000 Hz_40ms_envelope_correlation": 0.9856045478498155, "3000–6000 Hz_40ms_modulation_mae_db": 0.8195636213380886, "mr_stft_db": 5.742979076836374, "band_mae_db": 0.10800491527409296, "rms_gain_error_db": 0.0032225988626459314, "output_peak": 0.37844909098324997}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.633968482483547, "全帯域_10ms_rms_envelope_nrmse": 0.06722980241311584, "全帯域_10ms_envelope_correlation": 0.995572448732173, "全帯域_10ms_modulation_mae_db": 0.38553203826361554, "全帯域_40ms_log_rms_mae_db": 0.3418882539241765, "全帯域_40ms_rms_envelope_nrmse": 0.036481978101923815, "全帯域_40ms_envelope_correlation": 0.9986217689183571, "全帯域_40ms_modulation_mae_db": 0.42249945108025355, "300–1000 Hz_10ms_log_rms_mae_db": 1.3841201191879104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09660785558331855, "300–1000 Hz_10ms_envelope_correlation": 0.9923957749077615, "300–1000 Hz_10ms_modulation_mae_db": 0.5444285343545593, "300–1000 Hz_40ms_log_rms_mae_db": 0.8048704758265548, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056193476736282694, "300–1000 Hz_40ms_envelope_correlation": 0.9972687863261127, "300–1000 Hz_40ms_modulation_mae_db": 0.5801635415633813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3766317202854919, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1169033382978017, "1000–3000 Hz_10ms_envelope_correlation": 0.9899069450287626, "1000–3000 Hz_10ms_modulation_mae_db": 0.6683064502914737, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7825322834080147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0618799373997612, "1000–3000 Hz_40ms_envelope_correlation": 0.9975371055736955, "1000–3000 Hz_40ms_modulation_mae_db": 0.6966110482059459, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8579363474214445, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21175852571943685, "3000–6000 Hz_10ms_envelope_correlation": 0.9700682117950576, "3000–6000 Hz_10ms_modulation_mae_db": 0.7182035163541026, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0773270511298068, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1367884480989662, "3000–6000 Hz_40ms_envelope_correlation": 0.9881540440601155, "3000–6000 Hz_40ms_modulation_mae_db": 0.748915785638549, "mr_stft_db": 5.738130302860713, "band_mae_db": 0.1389161782303731, "rms_gain_error_db": -0.05797215226486383, "output_peak": 0.4031370823852255}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.570008134008727, "spectral_convergence": 0.23810293850309686, "spectral_convergence_status": "ok", "amplitude_floor": 1.174323222508143e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.600832609615054, "spectral_convergence": 0.243152359060047, "spectral_convergence_status": "ok", "amplitude_floor": 1.1752576893073956e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0698398083114755, "spectral_convergence": 0.25471564002627883, "spectral_convergence_status": "ok", "amplitude_floor": 9.275961087692859e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.278493760532488e-05, "output_power": 9.275961348663916e-05, "delta_db": -0.001185496961008819, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009129851997232178, "output_power": 0.0009156221255372597, "delta_db": 0.012525413282736732, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007449145920476126, "output_power": 0.0007177151411270267, "delta_db": -0.16154373371532715, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00040544152038629675, "output_power": 0.0003732169173247435, "delta_db": -0.3596690039641934, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011224761509504284, "output_power": 0.0001115835002854409, "delta_db": -0.02577141636699854, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.007390871044891e-05, "output_power": 5.1244812880769586e-05, "delta_db": 0.10038418415447108, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11017987474078927, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10677432845016652, "pitch": null, "output_rms_dbfs": -26.45480115622345, "output_peak": 0.3826377382935182, "reference_rms_dbfs": -26.348026827773285, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.746893517311752}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001795198121335632, "log_rms_mae_db": 0.643660752254516, "rms_envelope_nrmse": 0.06587883891847013, "envelope_correlation": 0.9958306738114677, "rms_gain_error_db": -0.10645089707170542, "reference_mean_power": 0.002319729266015563, "output_mean_power": 0.002263561048365768}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2030894540936354e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042099620393065603, "output_power": 0.000415440874503384, "reference_active": true, "delta_db": -0.05768956096889136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010212932714007922, "output_power": 9.960314510803182e-05, "reference_active": true, "delta_db": -0.10877418556022132}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7753962605616007e-05, "output_power": 1.4221853812893464e-05, "reference_active": true, "delta_db": -0.9633909061631876}], "active_band_count": 3, "active_band_mae_db": 0.3766182175641}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017377061993537507, "log_rms_mae_db": 0.3383763907715841, "rms_envelope_nrmse": 0.036424981709817214, "envelope_correlation": 0.9987322084675586, "rms_gain_error_db": -0.10709893018406952, "reference_mean_power": 0.002329667623369792, "output_mean_power": 0.0022729195863319825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1345793651623815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000381470345516612, "output_power": 0.0003776148989887943, "reference_active": true, "delta_db": -0.04411661437960025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.449258478788481e-05, "output_power": 5.337180174991113e-05, "reference_active": true, "delta_db": -0.09025544445919378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.451644818904224e-07, "output_power": 7.384044150652676e-07, "reference_active": true, "delta_db": -1.0721310826093744}], "active_band_count": 3, "active_band_mae_db": 0.40216771381605615}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001705289920117241, "log_rms_mae_db": 1.3800444574816613, "rms_envelope_nrmse": 0.0818238232557155, "envelope_correlation": 0.9945698740958738, "rms_gain_error_db": -0.07527527736360189, "reference_mean_power": 0.0014327712363456494, "output_mean_power": 0.001408151319608912}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.476686730813627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004044339102676846, "output_power": 0.0003987070305267873, "reference_active": true, "delta_db": -0.06193668628891866}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.628475293024379e-05, "output_power": 8.995553328026163e-05, "reference_active": true, "delta_db": -0.2952963781387941}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2048085119397813e-05, "output_power": 9.931971029735776e-06, "reference_active": true, "delta_db": -0.8388258304230691}], "active_band_count": 3, "active_band_mae_db": 0.3986862982835939}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015393777861692204, "log_rms_mae_db": 0.7530054809187184, "rms_envelope_nrmse": 0.043466647682219264, "envelope_correlation": 0.9984049061118984, "rms_gain_error_db": -0.07643971668066027, "reference_mean_power": 0.0014385306131629134, "output_mean_power": 0.0014134327073069798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.958921932585308e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037465347734378165, "output_power": 0.0003694976940029621, "reference_active": true, "delta_db": -0.060180355020118204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.015607336171049e-05, "output_power": 4.683512586534941e-05, "reference_active": true, "delta_db": -0.2975183724073282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018660953269709e-07, "output_power": 5.990290824717767e-07, "reference_active": true, "delta_db": -1.2665394331724733}], "active_band_count": 3, "active_band_mae_db": 0.5414127201999732}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.828766409891504e-05, "log_rms_mae_db": 1.366032943567076, "rms_envelope_nrmse": 0.12291122228970648, "envelope_correlation": 0.9889797775935759, "rms_gain_error_db": -0.30504225153166703, "reference_mean_power": 0.00046178379763570874, "output_mean_power": 0.0004304616497914109}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.102877053654468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.713589283902264e-05, "output_power": 8.06951132472228e-05, "reference_active": true, "delta_db": -0.33349850142898546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3909625300627975e-05, "output_power": 1.190578665389827e-05, "reference_active": true, "delta_db": -0.6755733513354795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.457858491278464e-06, "output_power": 3.105179355627907e-06, "reference_active": true, "delta_db": -1.5703958811701022}], "active_band_count": 3, "active_band_mae_db": 0.8598225779781891}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.324411735838957e-05, "log_rms_mae_db": 0.7897475846465645, "rms_envelope_nrmse": 0.07297894434084819, "envelope_correlation": 0.9965438346367679, "rms_gain_error_db": -0.3047027665099635, "reference_mean_power": 0.0004639006426222396, "output_mean_power": 0.00043246871661117224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.97726981805197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.049771362210352e-05, "output_power": 7.472519636871095e-05, "reference_active": true, "delta_db": -0.3231648028386631}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.625262232859704e-06, "output_power": 5.539910153051968e-06, "reference_active": true, "delta_db": -0.7770035130719454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8383435618714896e-07, "output_power": 2.034718052488862e-07, "reference_active": true, "delta_db": -1.4456072438556788}], "active_band_count": 3, "active_band_mae_db": 0.8485918532554292}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.433462630283978e-05, "log_rms_mae_db": 1.7568104915605816, "rms_envelope_nrmse": 0.20636620137146866, "envelope_correlation": 0.9676229969628386, "rms_gain_error_db": 0.07518534205429225, "reference_mean_power": 4.2981917565965076e-05, "output_mean_power": 4.373250164119628e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5620745474492675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.382029340071467e-06, "output_power": 5.899296262590788e-06, "reference_active": true, "delta_db": -0.3415858946119177}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1967382600492452e-06, "output_power": 1.217663702652818e-06, "reference_active": true, "delta_db": 0.07528184577432162}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059624559400323e-07, "output_power": 5.949187393038284e-07, "reference_active": true, "delta_db": 0.7033935697978115}], "active_band_count": 3, "active_band_mae_db": 0.37342043672801695}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.1780456996026785e-05, "log_rms_mae_db": 1.0420911494144105, "rms_envelope_nrmse": 0.1271584428928323, "envelope_correlation": 0.9868841684259793, "rms_gain_error_db": 0.07459439625891194, "reference_mean_power": 4.317499760725885e-05, "output_mean_power": 4.392297637902311e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4490976093596605e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.797984044318825e-06, "output_power": 5.442144071629558e-06, "reference_active": true, "delta_db": -0.27506981056628566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.94257411489492e-07, "output_power": 5.978957309934511e-07, "reference_active": true, "delta_db": 0.02650845656309315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.104930954274029e-08, "output_power": 2.94801612529324e-08, "reference_active": true, "delta_db": 1.4629200000908091}], "active_band_count": 3, "active_band_mae_db": 0.588166089073396}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8797630980783425, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ef3ce2f93641684e247d150f22c4bc0026f5a3c12c68727b6fdb3d2768699e5c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9893296335822235, "standard_deviation": 0.00044624295524863243, "minimum": 0.9886443364383573, "maximum": 0.9899069450287626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3677828261351108, "standard_deviation": 0.0228400771489892, "minimum": 1.335968202939294, "maximum": 1.4074733143513232, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7374499123808846, "standard_deviation": 0.1010417252275031, "minimum": 0.5847037791131783, "maximum": 0.8723705951001529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11959860947246287, "standard_deviation": 0.0026357843460085976, "minimum": 0.1169033382978017, "maximum": 0.12291122228970648, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969881034756678, "standard_deviation": 0.0002922775952261843, "minimum": 0.9965438346367679, "maximum": 0.9975371055736955, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7986418485405654, "standard_deviation": 0.020834631597316425, "minimum": 0.7789485405616146, "maximum": 0.8419546169093938, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7386368219469946, "standard_deviation": 0.1268178254217976, "minimum": 0.5019484570911877, "maximum": 0.8762102284976808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06601389620755752, "standard_deviation": 0.003604317578966036, "minimum": 0.0618799373997612, "maximum": 0.07297894434084819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9647866507307064, "standard_deviation": 0.006093776234882393, "minimum": 0.9559864531100283, "maximum": 0.9710143845471602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8369093070414495, "standard_deviation": 0.0413667861389538, "minimum": 1.7568104915605816, "maximum": 1.8862248439448959, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4760700470862765, "standard_deviation": 0.15173136576669397, "minimum": 0.30080802040625154, "maximum": 0.7182035163541026, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22094719700201726, "standard_deviation": 0.01740817027330198, "minimum": 0.19656047164017132, "maximum": 0.24642251360037984, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9871000267135275, "standard_deviation": 0.0015018394498025472, "minimum": 0.9841076521869239, "maximum": 0.9881993245800811, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0849971514452723, "standard_deviation": 0.03484164554799888, "minimum": 1.0420911494144105, "maximum": 1.1626078507258288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6536026576003474, "standard_deviation": 0.17795948811635437, "minimum": 0.39779218107097253, "maximum": 0.8940263203104942, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13338302684717712, "standard_deviation": 0.011552997734570827, "minimum": 0.12014446251178167, "maximum": 0.15047378615903464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9942623057495943, "standard_deviation": 0.0008381115339690245, "minimum": 0.9923957749077615, "maximum": 0.9950717285271485, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3783224794490987, "standard_deviation": 0.03726837327516957, "minimum": 1.3222456989119575, "maximum": 1.4478542683664157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38952174165110387, "standard_deviation": 0.16095639632468323, "minimum": 0.18699191855421202, "maximum": 0.6871263899098544, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08413282329985401, "standard_deviation": 0.005853904815648818, "minimum": 0.0786028413944275, "maximum": 0.09660785558331855, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980606160039267, "standard_deviation": 0.00039114322059371906, "minimum": 0.9972687863261127, "maximum": 0.9985862048621785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7867813922054783, "standard_deviation": 0.03278150874972567, "minimum": 0.7480888513595048, "maximum": 0.830829135040896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4202471064716615, "standard_deviation": 0.23025512191715952, "minimum": 0.12609334196150285, "maximum": 0.8084865399044764, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.047903890974637216, "standard_deviation": 0.004485440899834704, "minimum": 0.042275734328024395, "maximum": 0.056193476736282694, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11789360979459833, "standard_deviation": 0.03091887551261586, "minimum": 0.0737160157405836, "maximum": 0.17387289163785702, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.739910360147082, "standard_deviation": 0.010527252308164398, "minimum": 5.720682137751884, "maximum": 5.751863565677073, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.38792010778807184, "standard_deviation": 0.013739089581021085, "minimum": 0.36440210852347477, "maximum": 0.4041283336376955, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07176743345870185, "standard_deviation": 0.0486874026626939, "minimum": -0.16077581279025696, "maximum": 0.0032225988626459314, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9959773389017885, "standard_deviation": 0.0002885069890180017, "minimum": 0.995572448732173, "maximum": 0.9965207189100453, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6380390816162071, "standard_deviation": 0.015959723308676954, "minimum": 0.605942498002597, "maximum": 0.6549944701599963, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3542335912778938, "standard_deviation": 0.15714027480849602, "minimum": 0.11266548745777294, "maximum": 0.6676842872215015, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06444995922967099, "standard_deviation": 0.0021896633018208696, "minimum": 0.06018612912104554, "maximum": 0.06722980241311584, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987259217829507, "standard_deviation": 0.00016772349587272685, "minimum": 0.9984581791427187, "maximum": 0.999024665039055, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.34022407262191307, "standard_deviation": 0.012244031957372444, "minimum": 0.3217177787030861, "maximum": 0.3590758690808248, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3089039880774906, "standard_deviation": 0.17103977353509955, "minimum": 0.09334047971239656, "maximum": 0.5807885808877654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03574444386240811, "standard_deviation": 0.002604359533407943, "minimum": 0.03191923313788511, "maximum": 0.040021395443173395, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0000", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.734154462814331, "predicted_mos": 2.734154462814331, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "36995c08552eab5b52e3e8e31034ccf56922ded92eb55f4ac8dda5fb36a9f528", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.89174780016765, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6540273534162858, "全帯域_10ms_rms_envelope_nrmse": 0.07288485998929015, "全帯域_10ms_envelope_correlation": 0.9947831917864893, "全帯域_10ms_modulation_mae_db": 0.4277384793483074, "全帯域_40ms_log_rms_mae_db": 0.34570620380180117, "全帯域_40ms_rms_envelope_nrmse": 0.039403256383249055, "全帯域_40ms_envelope_correlation": 0.9983754702046702, "全帯域_40ms_modulation_mae_db": 0.398226619371038, "300–1000 Hz_10ms_log_rms_mae_db": 1.2002845315338946, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0888893709128524, "300–1000 Hz_10ms_envelope_correlation": 0.9935687850227193, "300–1000 Hz_10ms_modulation_mae_db": 0.4376061888217309, "300–1000 Hz_40ms_log_rms_mae_db": 0.6591805780560304, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05122240012883999, "300–1000 Hz_40ms_envelope_correlation": 0.9977204221132194, "300–1000 Hz_40ms_modulation_mae_db": 0.3984305466617862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1778536572358687, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11498869460908019, "1000–3000 Hz_10ms_envelope_correlation": 0.9899610302444461, "1000–3000 Hz_10ms_modulation_mae_db": 0.7022485392671799, "1000–3000 Hz_40ms_log_rms_mae_db": 0.728409669990757, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06642344367996406, "1000–3000 Hz_40ms_envelope_correlation": 0.9966330844432003, "1000–3000 Hz_40ms_modulation_mae_db": 0.7407415229482393, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4189733122641035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20295177467075426, "3000–6000 Hz_10ms_envelope_correlation": 0.9681052385128258, "3000–6000 Hz_10ms_modulation_mae_db": 0.22482701491289522, "3000–6000 Hz_40ms_log_rms_mae_db": 0.91531899172681, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11619791688605444, "3000–6000 Hz_40ms_envelope_correlation": 0.9889584788469326, "3000–6000 Hz_40ms_modulation_mae_db": 0.7191251676946843, "mr_stft_db": 5.461165612405874, "band_mae_db": 0.09762516786046178, "rms_gain_error_db": -0.03631186401656027, "output_peak": 0.38307355017841954}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6649357496823485, "全帯域_10ms_rms_envelope_nrmse": 0.06708390313585412, "全帯域_10ms_envelope_correlation": 0.9956597446690768, "全帯域_10ms_modulation_mae_db": 0.33846618187008937, "全帯域_40ms_log_rms_mae_db": 0.35909508749683705, "全帯域_40ms_rms_envelope_nrmse": 0.03572932744881165, "全帯域_40ms_envelope_correlation": 0.9987711366346443, "全帯域_40ms_modulation_mae_db": 0.41223083803196275, "300–1000 Hz_10ms_log_rms_mae_db": 1.1518218681115342, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08892678190643519, "300–1000 Hz_10ms_envelope_correlation": 0.9936193363409365, "300–1000 Hz_10ms_modulation_mae_db": 0.41050071515257813, "300–1000 Hz_40ms_log_rms_mae_db": 0.6224905163920477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05023369434418839, "300–1000 Hz_40ms_envelope_correlation": 0.9979110939117707, "300–1000 Hz_40ms_modulation_mae_db": 0.4949371561800186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1852370598671542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10745003866365842, "1000–3000 Hz_10ms_envelope_correlation": 0.991507888375965, "1000–3000 Hz_10ms_modulation_mae_db": 0.187855686788313, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7018250140115186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059496795118709006, "1000–3000 Hz_40ms_envelope_correlation": 0.9976911202874954, "1000–3000 Hz_40ms_modulation_mae_db": 0.08750275592353134, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4198559910743487, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23512911285215027, "3000–6000 Hz_10ms_envelope_correlation": 0.9585792424095846, "3000–6000 Hz_10ms_modulation_mae_db": 0.47854259439392705, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9502130823295694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12224012742064902, "3000–6000 Hz_40ms_envelope_correlation": 0.9881068613520387, "3000–6000 Hz_40ms_modulation_mae_db": 1.0131094200215751, "mr_stft_db": 5.475205752080071, "band_mae_db": 0.11154145889139107, "rms_gain_error_db": -0.10131710259427655, "output_peak": 0.36862748655389777}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6513675495922817, "全帯域_10ms_rms_envelope_nrmse": 0.061568920088442315, "全帯域_10ms_envelope_correlation": 0.9962981479132286, "全帯域_10ms_modulation_mae_db": 0.303258572265118, "全帯域_40ms_log_rms_mae_db": 0.35166300633614084, "全帯域_40ms_rms_envelope_nrmse": 0.034211013705355696, "全帯域_40ms_envelope_correlation": 0.998797179038153, "全帯域_40ms_modulation_mae_db": 0.23793343482051887, "300–1000 Hz_10ms_log_rms_mae_db": 1.1715912525256422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08341846139159902, "300–1000 Hz_10ms_envelope_correlation": 0.9943320552815916, "300–1000 Hz_10ms_modulation_mae_db": 0.40665111232679085, "300–1000 Hz_40ms_log_rms_mae_db": 0.6547300329800511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04574043345900274, "300–1000 Hz_40ms_envelope_correlation": 0.9981812570825178, "300–1000 Hz_40ms_modulation_mae_db": 0.38954806435296213, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1800854551830486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10903858338394896, "1000–3000 Hz_10ms_envelope_correlation": 0.9911582323387435, "1000–3000 Hz_10ms_modulation_mae_db": 0.6220236924934065, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6715991388736224, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061916367888665565, "1000–3000 Hz_40ms_envelope_correlation": 0.9973359393182077, "1000–3000 Hz_40ms_modulation_mae_db": 0.6708455533368017, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3865932075742053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20650982696001827, "3000–6000 Hz_10ms_envelope_correlation": 0.9669280599110792, "3000–6000 Hz_10ms_modulation_mae_db": 0.33263546852668546, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9191982670251145, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1290690671786883, "3000–6000 Hz_40ms_envelope_correlation": 0.9862840550661001, "3000–6000 Hz_40ms_modulation_mae_db": 0.6196055721044712, "mr_stft_db": 5.4619328864684595, "band_mae_db": 0.11412431958521879, "rms_gain_error_db": -0.059323105359988214, "output_peak": 0.37244419761551617}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.693063467050103, "全帯域_10ms_rms_envelope_nrmse": 0.07259238752944917, "全帯域_10ms_envelope_correlation": 0.9949274972734924, "全帯域_10ms_modulation_mae_db": 0.6165874463634667, "全帯域_40ms_log_rms_mae_db": 0.3518114265919279, "全帯域_40ms_rms_envelope_nrmse": 0.036912356882501204, "全帯域_40ms_envelope_correlation": 0.9987194442835755, "全帯域_40ms_modulation_mae_db": 0.503014779098867, "300–1000 Hz_10ms_log_rms_mae_db": 1.1974812773690093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08504669926575702, "300–1000 Hz_10ms_envelope_correlation": 0.9941452369707167, "300–1000 Hz_10ms_modulation_mae_db": 0.5364031889940746, "300–1000 Hz_40ms_log_rms_mae_db": 0.6851452067609907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04484256680724548, "300–1000 Hz_40ms_envelope_correlation": 0.9983196029390199, "300–1000 Hz_40ms_modulation_mae_db": 0.682922288285812, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2025438251701706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11105568324052105, "1000–3000 Hz_10ms_envelope_correlation": 0.9913500574345024, "1000–3000 Hz_10ms_modulation_mae_db": 0.914948353392138, "1000–3000 Hz_40ms_log_rms_mae_db": 0.685884989728198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06937878135113834, "1000–3000 Hz_40ms_envelope_correlation": 0.997177934813216, "1000–3000 Hz_40ms_modulation_mae_db": 0.9598699779410462, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4363623825867855, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24550781660151513, "3000–6000 Hz_10ms_envelope_correlation": 0.9530576689595073, "3000–6000 Hz_10ms_modulation_mae_db": 0.43873355684577175, "3000–6000 Hz_40ms_log_rms_mae_db": 0.957910602351561, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14236924807300175, "3000–6000 Hz_40ms_envelope_correlation": 0.9832745155289134, "3000–6000 Hz_40ms_modulation_mae_db": 0.6917714315870276, "mr_stft_db": 5.472045906377783, "band_mae_db": 0.161618062972806, "rms_gain_error_db": -0.11737521675041612, "output_peak": 0.39040064164413285}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6405319091709617, "全帯域_10ms_rms_envelope_nrmse": 0.06325758754808111, "全帯域_10ms_envelope_correlation": 0.9961206431299142, "全帯域_10ms_modulation_mae_db": 0.5272869407459697, "全帯域_40ms_log_rms_mae_db": 0.3343790129940388, "全帯域_40ms_rms_envelope_nrmse": 0.0367464705503798, "全帯域_40ms_envelope_correlation": 0.9986436169077386, "全帯域_40ms_modulation_mae_db": 0.513341777587554, "300–1000 Hz_10ms_log_rms_mae_db": 1.1682421203502016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0796377356741135, "300–1000 Hz_10ms_envelope_correlation": 0.9948491871377138, "300–1000 Hz_10ms_modulation_mae_db": 0.4599116212484886, "300–1000 Hz_40ms_log_rms_mae_db": 0.6781938547381282, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04743646686043628, "300–1000 Hz_40ms_envelope_correlation": 0.9980710087277597, "300–1000 Hz_40ms_modulation_mae_db": 0.6358163206115172, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1678611711172184, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10535490383162122, "1000–3000 Hz_10ms_envelope_correlation": 0.9917477053242477, "1000–3000 Hz_10ms_modulation_mae_db": 0.5268405937701163, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6986950685624749, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06219370228287683, "1000–3000 Hz_40ms_envelope_correlation": 0.9972713395527971, "1000–3000 Hz_40ms_modulation_mae_db": 0.3922974459125265, "3000–6000 Hz_10ms_log_rms_mae_db": 1.411344789274068, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2049103447620642, "3000–6000 Hz_10ms_envelope_correlation": 0.9676205662832005, "3000–6000 Hz_10ms_modulation_mae_db": 0.32591272754300527, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9072845841169853, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12309819608375315, "3000–6000 Hz_40ms_envelope_correlation": 0.9875049610201903, "3000–6000 Hz_40ms_modulation_mae_db": 0.715824664631024, "mr_stft_db": 5.470110297082958, "band_mae_db": 0.1202298188444436, "rms_gain_error_db": -0.08235080930575074, "output_peak": 0.37198151726571427}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6594591586877901, "全帯域_10ms_rms_envelope_nrmse": 0.07506253065136025, "全帯域_10ms_envelope_correlation": 0.9946214060503973, "全帯域_10ms_modulation_mae_db": 0.17033956449811663, "全帯域_40ms_log_rms_mae_db": 0.3452002783905569, "全帯域_40ms_rms_envelope_nrmse": 0.04129253271725388, "全帯域_40ms_envelope_correlation": 0.9984235419792098, "全帯域_40ms_modulation_mae_db": 0.19158403917540323, "300–1000 Hz_10ms_log_rms_mae_db": 1.2043855167567512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09015511365633537, "300–1000 Hz_10ms_envelope_correlation": 0.9934490657297043, "300–1000 Hz_10ms_modulation_mae_db": 0.05580182128113909, "300–1000 Hz_40ms_log_rms_mae_db": 0.6491097563630525, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05136154009475752, "300–1000 Hz_40ms_envelope_correlation": 0.997827047946665, "300–1000 Hz_40ms_modulation_mae_db": 0.23506507413664027, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1914064552788675, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11170380656104269, "1000–3000 Hz_10ms_envelope_correlation": 0.991184338456041, "1000–3000 Hz_10ms_modulation_mae_db": 0.44365791158067536, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7020191811683802, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06914497443261287, "1000–3000 Hz_40ms_envelope_correlation": 0.9971393561429859, "1000–3000 Hz_40ms_modulation_mae_db": 0.3948481142672005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.419628996491649, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22802614689502573, "3000–6000 Hz_10ms_envelope_correlation": 0.9595958501736823, "3000–6000 Hz_10ms_modulation_mae_db": 0.2588759142489972, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9345508163351348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1376447910050608, "3000–6000 Hz_40ms_envelope_correlation": 0.9843498659749801, "3000–6000 Hz_40ms_modulation_mae_db": 0.766919662541738, "mr_stft_db": 5.474769835518672, "band_mae_db": 0.17613857916710188, "rms_gain_error_db": -0.13853698597992742, "output_peak": 0.4145991855070935}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6396889134187631, "全帯域_10ms_rms_envelope_nrmse": 0.06791191674945671, "全帯域_10ms_envelope_correlation": 0.9954730640068898, "全帯域_10ms_modulation_mae_db": 0.08908155079529893, "全帯域_40ms_log_rms_mae_db": 0.35566783991142564, "全帯域_40ms_rms_envelope_nrmse": 0.03547199704625611, "全帯域_40ms_envelope_correlation": 0.9986861957720117, "全帯域_40ms_modulation_mae_db": 0.04797623957873733, "300–1000 Hz_10ms_log_rms_mae_db": 1.1541784129282775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0884733599819837, "300–1000 Hz_10ms_envelope_correlation": 0.9936338489533227, "300–1000 Hz_10ms_modulation_mae_db": 0.10952194249931423, "300–1000 Hz_40ms_log_rms_mae_db": 0.6510056431786722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047309022702025576, "300–1000 Hz_40ms_envelope_correlation": 0.9980522117223948, "300–1000 Hz_40ms_modulation_mae_db": 0.09942035489543061, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1865029820021098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11919081671964774, "1000–3000 Hz_10ms_envelope_correlation": 0.9892902744252715, "1000–3000 Hz_10ms_modulation_mae_db": 0.2746758150981821, "1000–3000 Hz_40ms_log_rms_mae_db": 0.733037494170233, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06821192296968602, "1000–3000 Hz_40ms_envelope_correlation": 0.9965988250416271, "1000–3000 Hz_40ms_modulation_mae_db": 0.2668456436478093, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4084211271668134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21196188323284873, "3000–6000 Hz_10ms_envelope_correlation": 0.9660496449989898, "3000–6000 Hz_10ms_modulation_mae_db": 0.3247558367806642, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9289603852984036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1350743704067854, "3000–6000 Hz_40ms_envelope_correlation": 0.9852579903382854, "3000–6000 Hz_40ms_modulation_mae_db": 0.3227058555856586, "mr_stft_db": 5.455504922117849, "band_mae_db": 0.10098801662092871, "rms_gain_error_db": -0.048518320607903, "output_peak": 0.36132916113215374}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6544742577843946, "全帯域_10ms_rms_envelope_nrmse": 0.06840474947689208, "全帯域_10ms_envelope_correlation": 0.9954911491197411, "全帯域_10ms_modulation_mae_db": 0.509770900805557, "全帯域_40ms_log_rms_mae_db": 0.3368962605290776, "全帯域_40ms_rms_envelope_nrmse": 0.0351660580010996, "全帯域_40ms_envelope_correlation": 0.9988181916264989, "全帯域_40ms_modulation_mae_db": 0.6334673921540969, "300–1000 Hz_10ms_log_rms_mae_db": 1.195602503774227, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08371270387703299, "300–1000 Hz_10ms_envelope_correlation": 0.9943193163837808, "300–1000 Hz_10ms_modulation_mae_db": 0.3401141589763246, "300–1000 Hz_40ms_log_rms_mae_db": 0.61747124283018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04107429922287318, "300–1000 Hz_40ms_envelope_correlation": 0.9985909771265176, "300–1000 Hz_40ms_modulation_mae_db": 0.6577850098642459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1803913815470568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11833290112422075, "1000–3000 Hz_10ms_envelope_correlation": 0.9895374994376477, "1000–3000 Hz_10ms_modulation_mae_db": 0.6596362819279299, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7018144446046348, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06929477285968526, "1000–3000 Hz_40ms_envelope_correlation": 0.9965855523422776, "1000–3000 Hz_40ms_modulation_mae_db": 0.6682507539696186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4322433453866596, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2135574970104569, "3000–6000 Hz_10ms_envelope_correlation": 0.9650388473257907, "3000–6000 Hz_10ms_modulation_mae_db": 0.3206248073342411, "3000–6000 Hz_40ms_log_rms_mae_db": 0.91704067333445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14459549975547803, "3000–6000 Hz_40ms_envelope_correlation": 0.9841326761139614, "3000–6000 Hz_40ms_modulation_mae_db": 0.5080046809671527, "mr_stft_db": 5.473426553537698, "band_mae_db": 0.12344677164598515, "rms_gain_error_db": -0.10257515809151405, "output_peak": 0.3567529374916397}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.192185148907391, "spectral_convergence": 0.2360479419937724, "spectral_convergence_status": "ok", "amplitude_floor": 1.174323222508143e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.242201658528246, "spectral_convergence": 0.23297176411040055, "spectral_convergence_status": "ok", "amplitude_floor": 1.1752576893073956e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.949110029781984, "spectral_convergence": 0.24341622526335246, "spectral_convergence_status": "ok", "amplitude_floor": 9.275961087692859e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.278493760532488e-05, "output_power": 9.365282655259128e-05, "delta_db": 0.04043409392566206, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009129851997232178, "output_power": 0.0009222963863690853, "delta_db": 0.0440676973704576, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007449145920476126, "output_power": 0.0007352420545257503, "delta_db": -0.05676141749683221, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00040544152038629675, "output_power": 0.0003928606263288686, "delta_db": -0.13689717093593662, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011224761509504284, "output_power": 0.00010542088434129766, "delta_db": -0.27250467347880836, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.007390871044891e-05, "output_power": 4.9670999352509346e-05, "delta_db": -0.03508595395507387, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09762516786046178, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03631186401656027, "pitch": null, "output_rms_dbfs": -26.384338691789843, "output_peak": 0.38307355017841954, "reference_rms_dbfs": -26.348026827773285, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.461165612405874}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001795198121335632, "log_rms_mae_db": 0.6540273534162858, "rms_envelope_nrmse": 0.07288485998929015, "envelope_correlation": 0.9947831917864893, "rms_gain_error_db": -0.03548856216966816, "reference_mean_power": 0.002319729266015563, "output_mean_power": 0.0023008507360037073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2030894540936354e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042099620393065603, "output_power": 0.00041806934881020606, "reference_active": true, "delta_db": -0.03029851909845435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010212932714007922, "output_power": 9.17021490711538e-05, "reference_active": true, "delta_db": -0.4677095702072235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7753962605616007e-05, "output_power": 1.4817487376995004e-05, "reference_active": true, "delta_db": -0.7852073487392444}], "active_band_count": 3, "active_band_mae_db": 0.4277384793483074}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017377061993537507, "log_rms_mae_db": 0.34570620380180117, "rms_envelope_nrmse": 0.039403256383249055, "envelope_correlation": 0.9983754702046702, "rms_gain_error_db": -0.03584814605214351, "reference_mean_power": 0.002329667623369792, "output_mean_power": 0.0023105169001873707}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1345793651623815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000381470345516612, "output_power": 0.0003786316105256865, "reference_active": true, "delta_db": -0.032439139624105724}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.449258478788481e-05, "output_power": 4.902206703492011e-05, "reference_active": true, "delta_db": -0.4594578910909877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.451644818904224e-07, "output_power": 8.039501309394555e-07, "reference_active": true, "delta_db": -0.7027828273980206}], "active_band_count": 3, "active_band_mae_db": 0.398226619371038}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001705289920117241, "log_rms_mae_db": 1.2002845315338946, "rms_envelope_nrmse": 0.0888893709128524, "envelope_correlation": 0.9935687850227193, "rms_gain_error_db": -0.020320268562490695, "reference_mean_power": 0.0014327712363456494, "output_mean_power": 0.0014260830807342607}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.476686730813627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004044339102676846, "output_power": 0.0004104950783094022, "reference_active": true, "delta_db": 0.06460392039312855}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.628475293024379e-05, "output_power": 8.669662078555768e-05, "reference_active": true, "delta_db": -0.455553502194217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2048085119397813e-05, "output_power": 1.0038109128248678e-05, "reference_active": true, "delta_db": -0.7926611438778473}], "active_band_count": 3, "active_band_mae_db": 0.4376061888217309}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015393777861692204, "log_rms_mae_db": 0.6591805780560304, "rms_envelope_nrmse": 0.05122240012883999, "envelope_correlation": 0.9977204221132194, "rms_gain_error_db": -0.02059538753682403, "reference_mean_power": 0.0014385306131629134, "output_mean_power": 0.001431724872388714}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.958921932585308e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037465347734378165, "output_power": 0.0003775299534030618, "reference_active": true, "delta_db": 0.033216466053026736}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.015607336171049e-05, "output_power": 4.4859378452773845e-05, "reference_active": true, "delta_db": -0.48470276997473466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018660953269709e-07, "output_power": 6.860639675376235e-07, "reference_active": true, "delta_db": -0.6773724039575973}], "active_band_count": 3, "active_band_mae_db": 0.3984305466617862}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.828766409891504e-05, "log_rms_mae_db": 1.1778536572358687, "rms_envelope_nrmse": 0.11498869460908019, "envelope_correlation": 0.9899610302444461, "rms_gain_error_db": -0.15805303567572354, "reference_mean_power": 0.00046178379763570874, "output_mean_power": 0.00044528020344909323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.102877053654468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.713589283902264e-05, "output_power": 8.028401619979393e-05, "reference_active": true, "delta_db": -0.3556799566215705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3909625300627975e-05, "output_power": 1.1739387126123984e-05, "reference_active": true, "delta_db": -0.7367000661522889}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.457858491278464e-06, "output_power": 3.5293092856679303e-06, "reference_active": true, "delta_db": -1.0143655950276806}], "active_band_count": 3, "active_band_mae_db": 0.7022485392671799}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.324411735838957e-05, "log_rms_mae_db": 0.728409669990757, "rms_envelope_nrmse": 0.06642344367996406, "envelope_correlation": 0.9966330844432003, "rms_gain_error_db": -0.15857761683061822, "reference_mean_power": 0.0004639006426222396, "output_mean_power": 0.0004472673665933276}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.97726981805197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.049771362210352e-05, "output_power": 7.400028126729087e-05, "reference_active": true, "delta_db": -0.36550174827874843}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.625262232859704e-06, "output_power": 5.369832961702314e-06, "reference_active": true, "delta_db": -0.912422962512359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8383435618714896e-07, "output_power": 2.2836785718017292e-07, "reference_active": true, "delta_db": -0.9442998580536102}], "active_band_count": 3, "active_band_mae_db": 0.7407415229482393}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.433462630283978e-05, "log_rms_mae_db": 1.4189733122641035, "rms_envelope_nrmse": 0.20295177467075426, "envelope_correlation": 0.9681052385128258, "rms_gain_error_db": -0.15930556275024438, "reference_mean_power": 4.2981917565965076e-05, "output_mean_power": 4.143384438670517e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5620745474492675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.382029340071467e-06, "output_power": 5.7215812081613284e-06, "reference_active": true, "delta_db": -0.4744273004726344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1967382600492452e-06, "output_power": 1.2349250526774909e-06, "reference_active": true, "delta_db": 0.13641425396715973}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059624559400323e-07, "output_power": 5.134311820491953e-07, "reference_active": true, "delta_db": 0.06363949029889156}], "active_band_count": 3, "active_band_mae_db": 0.22482701491289522}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.1780456996026785e-05, "log_rms_mae_db": 0.91531899172681, "rms_envelope_nrmse": 0.11619791688605444, "envelope_correlation": 0.9889584788469326, "rms_gain_error_db": -0.1584703886403641, "reference_mean_power": 4.317499760725885e-05, "output_mean_power": 4.162797483093213e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4490976093596605e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.797984044318825e-06, "output_power": 5.2633304604160745e-06, "reference_active": true, "delta_db": -0.42016377635024593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.94257411489492e-07, "output_power": 6.106937598012695e-07, "reference_active": true, "delta_db": 0.11848875415377867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.104930954274029e-08, "output_power": 3.0556960442924706e-08, "reference_active": true, "delta_db": 1.6187229725800278}], "active_band_count": 3, "active_band_mae_db": 0.7191251676946843}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8979873733209942, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fafa454e23d9293b9ee8dd3075a4e50e259f80f09e7d69e87cfd072561797da7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9907171282546081, "standard_deviation": 0.0009636217689520412, "minimum": 0.9892902744252715, "maximum": 0.9917477053242477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.183985248425187, "standard_deviation": 0.010233751178126128, "minimum": 1.1678611711172184, "maximum": 1.2025438251701706, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5414858592897427, "standard_deviation": 0.23658022754642716, "minimum": 0.187855686788313, "maximum": 0.914948353392138, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11213942851671763, "standard_deviation": 0.005002844320118839, "minimum": 0.10535490383162122, "maximum": 0.11919081671964774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970541439927258, "standard_deviation": 0.0004069207049838937, "minimum": 0.9965855523422776, "maximum": 0.9976911202874954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7029106251387274, "standard_deviation": 0.02017444179490523, "minimum": 0.6715991388736224, "maximum": 0.733037494170233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5226502209933467, "standard_deviation": 0.285357126168915, "minimum": 0.08750275592353134, "maximum": 0.9598699779410462, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06575759507291724, "standard_deviation": 0.003968087154180165, "minimum": 0.059496795118709006, "maximum": 0.06937878135113834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9631218898218326, "standard_deviation": 0.005426948681687826, "minimum": 0.9530576689595073, "maximum": 0.9681052385128258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.416677893977329, "standard_deviation": 0.015378766352274236, "minimum": 1.3865932075742053, "maximum": 1.4363623825867855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3381134900732734, "standard_deviation": 0.08419390740274238, "minimum": 0.22482701491289522, "maximum": 0.47854259439392705, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21856930037310418, "standard_deviation": 0.015739051535812537, "minimum": 0.20295177467075426, "maximum": 0.24550781660151513, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9859836755301753, "standard_deviation": 0.0020606382377076835, "minimum": 0.9832745155289134, "maximum": 0.9889584788469326, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9288096753147536, "standard_deviation": 0.01778510561285688, "minimum": 0.9072845841169853, "maximum": 0.957910602351561, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6696333068916664, "standard_deviation": 0.2005346073100411, "minimum": 0.3227058555856586, "maximum": 1.0131094200215751, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13128615210118386, "standard_deviation": 0.010255313129363291, "minimum": 0.11619791688605444, "maximum": 0.14459549975547803, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9939896039775606, "standard_deviation": 0.0004959864563785563, "minimum": 0.9934490657297043, "maximum": 0.9948491871377138, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1804484354186922, "standard_deviation": 0.021458452608622374, "minimum": 1.1518218681115342, "maximum": 1.2043855167567512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3445638436625551, "standard_deviation": 0.1713871235045216, "minimum": 0.05580182128113909, "maximum": 0.5364031889940746, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08603252833326364, "standard_deviation": 0.003656145232752635, "minimum": 0.0796377356741135, "maximum": 0.09015511365633537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980842026962331, "standard_deviation": 0.000280632495741771, "minimum": 0.9977204221132194, "maximum": 0.9985909771265176, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.652165853912394, "standard_deviation": 0.023651950770252957, "minimum": 0.61747124283018, "maximum": 0.6851452067609907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44924060187355164, "standard_deviation": 0.21023347469972087, "minimum": 0.09942035489543061, "maximum": 0.682922288285812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04740255295242114, "standard_deviation": 0.0035386243659867108, "minimum": 0.04107429922287318, "maximum": 0.05136154009475752, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12571402444854213, "standard_deviation": 0.028290570988792133, "minimum": 0.09762516786046178, "maximum": 0.17613857916710188, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.468020220698671, "standard_deviation": 0.007441833032085604, "minimum": 5.455504922117849, "maximum": 5.475205752080071, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3774010846735709, "standard_deviation": 0.01851808083423756, "minimum": 0.3567529374916397, "maximum": 0.4145991855070935, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08578857033829204, "standard_deviation": 0.035561398509250754, "minimum": -0.13853698597992742, "maximum": -0.03631186401656027, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954218554936537, "standard_deviation": 0.000610995491603911, "minimum": 0.9946214060503973, "maximum": 0.9962981479132286, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6571935448503661, "standard_deviation": 0.01684852599104727, "minimum": 0.6396889134187631, "maximum": 0.693063467050103, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37281620458649045, "standard_deviation": 0.18227883018953858, "minimum": 0.08908155079529893, "maximum": 0.6165874463634667, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06859585689610323, "standard_deviation": 0.004733632476642397, "minimum": 0.061568920088442315, "maximum": 0.07506253065136025, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986543470558128, "standard_deviation": 0.00016788441244699673, "minimum": 0.9983754702046702, "maximum": 0.9988181916264989, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3475523895064757, "standard_deviation": 0.008702731440404357, "minimum": 0.3343790129940388, "maximum": 0.35909508749683705, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36722188997727223, "standard_deviation": 0.19388746402410642, "minimum": 0.04797623957873733, "maximum": 0.6334673921540969, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03686662659186338, "standard_deviation": 0.0023671054870648073, "minimum": 0.034211013705355696, "maximum": 0.04129253271725388, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0000", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.206979990005493, "predicted_mos": 3.206979990005493, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5430892f0b5c7024553716f792d20b94ec0de08cb996349237f6a67cdb361c91", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.875820599962026, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6531896422686224, "全帯域_10ms_rms_envelope_nrmse": 0.06782100705162901, "全帯域_10ms_envelope_correlation": 0.9956620940934883, "全帯域_10ms_modulation_mae_db": 0.28627168836927347, "全帯域_40ms_log_rms_mae_db": 0.3474597777998758, "全帯域_40ms_rms_envelope_nrmse": 0.04132474916026353, "全帯域_40ms_envelope_correlation": 0.998427926479604, "全帯域_40ms_modulation_mae_db": 0.14060651637883184, "300–1000 Hz_10ms_log_rms_mae_db": 1.2137616678139613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08491279322068739, "300–1000 Hz_10ms_envelope_correlation": 0.9942961460015718, "300–1000 Hz_10ms_modulation_mae_db": 0.213902751331951, "300–1000 Hz_40ms_log_rms_mae_db": 0.6694512482868947, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05114357418313003, "300–1000 Hz_40ms_envelope_correlation": 0.9979581504557067, "300–1000 Hz_40ms_modulation_mae_db": 0.19405489367389717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2492292819618347, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11692761346850593, "1000–3000 Hz_10ms_envelope_correlation": 0.9898595408788325, "1000–3000 Hz_10ms_modulation_mae_db": 0.71069354864761, "1000–3000 Hz_40ms_log_rms_mae_db": 0.730085237236461, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06929764467213305, "1000–3000 Hz_40ms_envelope_correlation": 0.9966707488478093, "1000–3000 Hz_40ms_modulation_mae_db": 0.688895477027283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.478754944546487, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20784450623162548, "3000–6000 Hz_10ms_envelope_correlation": 0.9690889433258996, "3000–6000 Hz_10ms_modulation_mae_db": 0.3813557777169336, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9207573995250138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13895309668989062, "3000–6000 Hz_40ms_envelope_correlation": 0.9857348589261969, "3000–6000 Hz_40ms_modulation_mae_db": 0.8840207334746918, "mr_stft_db": 5.530663773761802, "band_mae_db": 0.12158462374127514, "rms_gain_error_db": -0.14040633531804167, "output_peak": 0.3743638405003378}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6602971615927558, "全帯域_10ms_rms_envelope_nrmse": 0.06767747486788744, "全帯域_10ms_envelope_correlation": 0.9955467787771459, "全帯域_10ms_modulation_mae_db": 0.17193780118919408, "全帯域_40ms_log_rms_mae_db": 0.3521252905432537, "全帯域_40ms_rms_envelope_nrmse": 0.041633585515324326, "全帯域_40ms_envelope_correlation": 0.9982364038398164, "全帯域_40ms_modulation_mae_db": 0.1798134369829261, "300–1000 Hz_10ms_log_rms_mae_db": 1.2248029862297376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08462749580683804, "300–1000 Hz_10ms_envelope_correlation": 0.9941817182391973, "300–1000 Hz_10ms_modulation_mae_db": 0.037663082977748105, "300–1000 Hz_40ms_log_rms_mae_db": 0.6885509877226477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05464913184605577, "300–1000 Hz_40ms_envelope_correlation": 0.9974204493082061, "300–1000 Hz_40ms_modulation_mae_db": 0.1971515339357346, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2100521776415891, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10330080295579726, "1000–3000 Hz_10ms_envelope_correlation": 0.9920430036771944, "1000–3000 Hz_10ms_modulation_mae_db": 0.6757325370084072, "1000–3000 Hz_40ms_log_rms_mae_db": 0.701406583045064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05672252408636405, "1000–3000 Hz_40ms_envelope_correlation": 0.997763535510736, "1000–3000 Hz_40ms_modulation_mae_db": 0.7116805794006665, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4933791278350665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20144135010335776, "3000–6000 Hz_10ms_envelope_correlation": 0.9688483740530592, "3000–6000 Hz_10ms_modulation_mae_db": 0.30713413459602806, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9538133266719224, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12756935253531887, "3000–6000 Hz_40ms_envelope_correlation": 0.9875696143546908, "3000–6000 Hz_40ms_modulation_mae_db": 0.6711347191289493, "mr_stft_db": 5.542091238583112, "band_mae_db": 0.11951195299728812, "rms_gain_error_db": -0.08154423033066308, "output_peak": 0.38651985380769716}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6743640286914772, "全帯域_10ms_rms_envelope_nrmse": 0.0670952772422313, "全帯域_10ms_envelope_correlation": 0.9956059334136591, "全帯域_10ms_modulation_mae_db": 0.5001385858626023, "全帯域_40ms_log_rms_mae_db": 0.35162299898480104, "全帯域_40ms_rms_envelope_nrmse": 0.035916663135374165, "全帯域_40ms_envelope_correlation": 0.9986860729486614, "全帯域_40ms_modulation_mae_db": 0.4631231950646906, "300–1000 Hz_10ms_log_rms_mae_db": 1.2224663230555932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08489326067099146, "300–1000 Hz_10ms_envelope_correlation": 0.9941650532923397, "300–1000 Hz_10ms_modulation_mae_db": 0.452696104196151, "300–1000 Hz_40ms_log_rms_mae_db": 0.651319123072183, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043571425944201075, "300–1000 Hz_40ms_envelope_correlation": 0.9984143055122359, "300–1000 Hz_40ms_modulation_mae_db": 0.4085544118372269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2191922848275785, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12383179024762256, "1000–3000 Hz_10ms_envelope_correlation": 0.9882911022928643, "1000–3000 Hz_10ms_modulation_mae_db": 0.5222575794068424, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7516097503690989, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07322305693078009, "1000–3000 Hz_40ms_envelope_correlation": 0.995800878796476, "1000–3000 Hz_40ms_modulation_mae_db": 0.6064426927798839, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5442680218093248, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24608002515110924, "3000–6000 Hz_10ms_envelope_correlation": 0.9546178830450076, "3000–6000 Hz_10ms_modulation_mae_db": 0.4270575749163695, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9531445565527116, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1672787083142708, "3000–6000 Hz_40ms_envelope_correlation": 0.9776597913791492, "3000–6000 Hz_40ms_modulation_mae_db": 0.4385936870716923, "mr_stft_db": 5.545836459301818, "band_mae_db": 0.09404808700020635, "rms_gain_error_db": -0.07108960795943664, "output_peak": 0.40087035417335165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6295746143802797, "全帯域_10ms_rms_envelope_nrmse": 0.06979915442323258, "全帯域_10ms_envelope_correlation": 0.9952830177207672, "全帯域_10ms_modulation_mae_db": 0.3846158789456913, "全帯域_40ms_log_rms_mae_db": 0.33239378181748686, "全帯域_40ms_rms_envelope_nrmse": 0.039834566067038994, "全帯域_40ms_envelope_correlation": 0.9984257685410758, "全帯域_40ms_modulation_mae_db": 0.42358198452335394, "300–1000 Hz_10ms_log_rms_mae_db": 1.1839221671990436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08484764675648067, "300–1000 Hz_10ms_envelope_correlation": 0.9941434778893999, "300–1000 Hz_10ms_modulation_mae_db": 0.27317624878663244, "300–1000 Hz_40ms_log_rms_mae_db": 0.6266032718395658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049402942011046704, "300–1000 Hz_40ms_envelope_correlation": 0.9978969522813655, "300–1000 Hz_40ms_modulation_mae_db": 0.3541016871752545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.223421712200367, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11594709082038764, "1000–3000 Hz_10ms_envelope_correlation": 0.9905222133427609, "1000–3000 Hz_10ms_modulation_mae_db": 0.6915957591024169, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7257362595412133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07173930072368967, "1000–3000 Hz_40ms_envelope_correlation": 0.9969727247569599, "1000–3000 Hz_40ms_modulation_mae_db": 0.6583520789424823, "3000–6000 Hz_10ms_log_rms_mae_db": 1.510746134523672, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20699330908552946, "3000–6000 Hz_10ms_envelope_correlation": 0.9667897531833863, "3000–6000 Hz_10ms_modulation_mae_db": 0.46779764030471815, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9433093582811449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12418228718117273, "3000–6000 Hz_40ms_envelope_correlation": 0.9872879746297416, "3000–6000 Hz_40ms_modulation_mae_db": 0.43694289282619475, "mr_stft_db": 5.522098515431357, "band_mae_db": 0.1279185698893255, "rms_gain_error_db": -0.09818587624895221, "output_peak": 0.3999978797815068}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6409728526138634, "全帯域_10ms_rms_envelope_nrmse": 0.06829757370013101, "全帯域_10ms_envelope_correlation": 0.9954210639790881, "全帯域_10ms_modulation_mae_db": 0.1852154033550327, "全帯域_40ms_log_rms_mae_db": 0.3489143735692815, "全帯域_40ms_rms_envelope_nrmse": 0.03946174113087146, "全帯域_40ms_envelope_correlation": 0.9983571972792918, "全帯域_40ms_modulation_mae_db": 0.2908500544341256, "300–1000 Hz_10ms_log_rms_mae_db": 1.2623658758034393, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08745998383389693, "300–1000 Hz_10ms_envelope_correlation": 0.9937852339209668, "300–1000 Hz_10ms_modulation_mae_db": 0.355833548329931, "300–1000 Hz_40ms_log_rms_mae_db": 0.695604409130189, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050212305163690775, "300–1000 Hz_40ms_envelope_correlation": 0.9978066561871767, "300–1000 Hz_40ms_modulation_mae_db": 0.4094370766939012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.231944233511056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.119201300346275, "1000–3000 Hz_10ms_envelope_correlation": 0.9891075559768371, "1000–3000 Hz_10ms_modulation_mae_db": 0.15573538734707087, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7133428592557418, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06224997191998632, "1000–3000 Hz_40ms_envelope_correlation": 0.996931692026345, "1000–3000 Hz_40ms_modulation_mae_db": 0.1867877203198334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5547498381572586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22119267919713453, "3000–6000 Hz_10ms_envelope_correlation": 0.9619103118893854, "3000–6000 Hz_10ms_modulation_mae_db": 0.5806521495462046, "3000–6000 Hz_40ms_log_rms_mae_db": 0.945148811640114, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12396850721611802, "3000–6000 Hz_40ms_envelope_correlation": 0.9875314221580146, "3000–6000 Hz_40ms_modulation_mae_db": 0.6935736334929219, "mr_stft_db": 5.539344896145921, "band_mae_db": 0.12838617677059116, "rms_gain_error_db": -0.029521512633780018, "output_peak": 0.40544294220653493}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6315735671436207, "全帯域_10ms_rms_envelope_nrmse": 0.06875676091524875, "全帯域_10ms_envelope_correlation": 0.995364680097609, "全帯域_10ms_modulation_mae_db": 0.5696075138161939, "全帯域_40ms_log_rms_mae_db": 0.31513322173162306, "全帯域_40ms_rms_envelope_nrmse": 0.03366441965784252, "全帯域_40ms_envelope_correlation": 0.998830118491311, "全帯域_40ms_modulation_mae_db": 0.4526465574044698, "300–1000 Hz_10ms_log_rms_mae_db": 1.228713793719136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08429028744025951, "300–1000 Hz_10ms_envelope_correlation": 0.9942131181426186, "300–1000 Hz_10ms_modulation_mae_db": 0.6849859922067436, "300–1000 Hz_40ms_log_rms_mae_db": 0.6332414301109415, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04522145942186218, "300–1000 Hz_40ms_envelope_correlation": 0.9982256387079782, "300–1000 Hz_40ms_modulation_mae_db": 0.5537607216997845, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1970762961557282, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10703011231868523, "1000–3000 Hz_10ms_envelope_correlation": 0.9914312797647427, "1000–3000 Hz_10ms_modulation_mae_db": 0.6438980348669892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7141550263415067, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055575373209530245, "1000–3000 Hz_40ms_envelope_correlation": 0.997867485928331, "1000–3000 Hz_40ms_modulation_mae_db": 0.47352508757429995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5732328335448187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22483420352180555, "3000–6000 Hz_10ms_envelope_correlation": 0.9624009371415269, "3000–6000 Hz_10ms_modulation_mae_db": 0.2609701919502182, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9787000808690411, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11816334538053419, "3000–6000 Hz_40ms_envelope_correlation": 0.989093872872549, "3000–6000 Hz_40ms_modulation_mae_db": 0.7364262882332225, "mr_stft_db": 5.530383145292145, "band_mae_db": 0.07208792970154511, "rms_gain_error_db": -0.053456956219854745, "output_peak": 0.38631632042021175}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6490707908411928, "全帯域_10ms_rms_envelope_nrmse": 0.0656219925090325, "全帯域_10ms_envelope_correlation": 0.9957991936326759, "全帯域_10ms_modulation_mae_db": 0.42432594439452104, "全帯域_40ms_log_rms_mae_db": 0.3486825660616022, "全帯域_40ms_rms_envelope_nrmse": 0.0394141521439083, "全帯域_40ms_envelope_correlation": 0.9984050473960431, "全帯域_40ms_modulation_mae_db": 0.3713761669301874, "300–1000 Hz_10ms_log_rms_mae_db": 1.2321086406459543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07659250599746933, "300–1000 Hz_10ms_envelope_correlation": 0.9952377358302084, "300–1000 Hz_10ms_modulation_mae_db": 0.41680639374924205, "300–1000 Hz_40ms_log_rms_mae_db": 0.6373984308250517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04519361772088168, "300–1000 Hz_40ms_envelope_correlation": 0.9982507943295946, "300–1000 Hz_40ms_modulation_mae_db": 0.4761749323540592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.212651805964625, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10775341909826085, "1000–3000 Hz_10ms_envelope_correlation": 0.9913122194255549, "1000–3000 Hz_10ms_modulation_mae_db": 0.9189813411734834, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6888378790072368, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06741360566121218, "1000–3000 Hz_40ms_envelope_correlation": 0.9966704287168036, "1000–3000 Hz_40ms_modulation_mae_db": 0.9641586602587765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5532180815331653, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27546169887736593, "3000–6000 Hz_10ms_envelope_correlation": 0.9447375972101764, "3000–6000 Hz_10ms_modulation_mae_db": 0.6172016184976764, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9531606895355501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18638121324382062, "3000–6000 Hz_40ms_envelope_correlation": 0.9732077148253846, "3000–6000 Hz_40ms_modulation_mae_db": 0.9865202227987737, "mr_stft_db": 5.522655813873864, "band_mae_db": 0.13493935015998226, "rms_gain_error_db": -0.07120944850121555, "output_peak": 0.3971694709585282}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6481579088577172, "全帯域_10ms_rms_envelope_nrmse": 0.06164572886838216, "全帯域_10ms_envelope_correlation": 0.9963132204189044, "全帯域_10ms_modulation_mae_db": 0.43568035364151064, "全帯域_40ms_log_rms_mae_db": 0.3498154430590705, "全帯域_40ms_rms_envelope_nrmse": 0.03373646621731622, "全帯域_40ms_envelope_correlation": 0.9988649235585315, "全帯域_40ms_modulation_mae_db": 0.505507100812104, "300–1000 Hz_10ms_log_rms_mae_db": 1.2155557698909036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08114271868633131, "300–1000 Hz_10ms_envelope_correlation": 0.9946538175872977, "300–1000 Hz_10ms_modulation_mae_db": 0.46293212575594617, "300–1000 Hz_40ms_log_rms_mae_db": 0.6365784783370768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04661918509569958, "300–1000 Hz_40ms_envelope_correlation": 0.9981365429286863, "300–1000 Hz_40ms_modulation_mae_db": 0.43814996103434495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2460264137195607, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10917657729290954, "1000–3000 Hz_10ms_envelope_correlation": 0.9914468658198958, "1000–3000 Hz_10ms_modulation_mae_db": 0.7882483923999585, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7017544406712531, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06412449006959477, "1000–3000 Hz_40ms_envelope_correlation": 0.997486487580726, "1000–3000 Hz_40ms_modulation_mae_db": 1.0001972051212797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5345793725099754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2064009296090947, "3000–6000 Hz_10ms_envelope_correlation": 0.9670156521475859, "3000–6000 Hz_10ms_modulation_mae_db": 0.4101922194413517, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9247015427667624, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11523607523833213, "3000–6000 Hz_40ms_envelope_correlation": 0.9890685737081056, "3000–6000 Hz_40ms_modulation_mae_db": 0.31562451403725905, "mr_stft_db": 5.519260582955861, "band_mae_db": 0.1596928282994905, "rms_gain_error_db": -0.07958173820091252, "output_peak": 0.3605487822765336}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.332949753891908, "spectral_convergence": 0.23573699228656467, "spectral_convergence_status": "ok", "amplitude_floor": 1.174323222508143e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.32261813127996, "spectral_convergence": 0.23761654506468333, "spectral_convergence_status": "ok", "amplitude_floor": 1.1752576893073956e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.936423436113536, "spectral_convergence": 0.24534750569136324, "spectral_convergence_status": "ok", "amplitude_floor": 9.275961087692859e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.278493760532488e-05, "output_power": 9.532766533807288e-05, "delta_db": 0.11741476846961364, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009129851997232178, "output_power": 0.0009023069583168293, "delta_db": -0.051094307760623126, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007449145920476126, "output_power": 0.0007050219927189853, "delta_db": -0.2390381695624325, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00040544152038629675, "output_power": 0.000380466826806481, "delta_db": -0.2761142550054493, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011224761509504284, "output_power": 0.00011183909985212125, "delta_db": -0.015834594421053096, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.007390871044891e-05, "output_power": 4.972906898026212e-05, "delta_db": -0.030011647228479207, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12158462374127514, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14040633531804167, "pitch": null, "output_rms_dbfs": -26.488433163091326, "output_peak": 0.3743638405003378, "reference_rms_dbfs": -26.348026827773285, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.530663773761802}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001795198121335632, "log_rms_mae_db": 0.6531896422686224, "rms_envelope_nrmse": 0.06782100705162901, "envelope_correlation": 0.9956620940934883, "rms_gain_error_db": -0.139781218299291, "reference_mean_power": 0.002319729266015563, "output_mean_power": 0.0022462556361100526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2030894540936354e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042099620393065603, "output_power": 0.0004188180254102238, "reference_active": true, "delta_db": -0.022528149682636688}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010212932714007922, "output_power": 0.00010027499231244952, "reference_active": true, "delta_db": -0.0795783329074091}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7753962605616007e-05, "output_power": 1.4915040691775674e-05, "reference_active": true, "delta_db": -0.7567085825177745}], "active_band_count": 3, "active_band_mae_db": 0.28627168836927347}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017377061993537507, "log_rms_mae_db": 0.3474597777998758, "rms_envelope_nrmse": 0.04132474916026353, "envelope_correlation": 0.998427926479604, "rms_gain_error_db": -0.1395705215455305, "reference_mean_power": 0.002329667623369792, "output_mean_power": 0.0022559886582563435}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1345793651623815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000381470345516612, "output_power": 0.000379258434872819, "reference_active": true, "delta_db": -0.025255342480812626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.449258478788481e-05, "output_power": 5.3594066154138024e-05, "reference_active": true, "delta_db": -0.07220700618693503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.451644818904224e-07, "output_power": 8.771455672659494e-07, "reference_active": true, "delta_db": -0.32435720046874783}], "active_band_count": 3, "active_band_mae_db": 0.14060651637883184}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001705289920117241, "log_rms_mae_db": 1.2137616678139613, "rms_envelope_nrmse": 0.08491279322068739, "envelope_correlation": 0.9942961460015718, "rms_gain_error_db": -0.15719315226755493, "reference_mean_power": 0.0014327712363456494, "output_mean_power": 0.0013818392967971682}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.476686730813627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004044339102676846, "output_power": 0.00040141464917471496, "reference_active": true, "delta_db": -0.032543448825602735}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.628475293024379e-05, "output_power": 9.246479570210472e-05, "reference_active": true, "delta_db": -0.17581105885656007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2048085119397813e-05, "output_power": 1.0903919899901718e-05, "reference_active": true, "delta_db": -0.4333537463136902}], "active_band_count": 3, "active_band_mae_db": 0.213902751331951}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015393777861692204, "log_rms_mae_db": 0.6694512482868947, "rms_envelope_nrmse": 0.05114357418313003, "envelope_correlation": 0.9979581504557067, "rms_gain_error_db": -0.15643423109172452, "reference_mean_power": 0.0014385306131629134, "output_mean_power": 0.001387636405723132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.958921932585308e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037465347734378165, "output_power": 0.0003694622663950405, "reference_active": true, "delta_db": -0.06059677851973148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.015607336171049e-05, "output_power": 4.797692680644007e-05, "reference_active": true, "delta_db": -0.19291103432776863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018660953269709e-07, "output_power": 7.434232926435161e-07, "reference_active": true, "delta_db": -0.3286568681741914}], "active_band_count": 3, "active_band_mae_db": 0.19405489367389717}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.828766409891504e-05, "log_rms_mae_db": 1.2492292819618347, "rms_envelope_nrmse": 0.11692761346850593, "envelope_correlation": 0.9898595408788325, "rms_gain_error_db": -0.2542202351190167, "reference_mean_power": 0.00046178379763570874, "output_mean_power": 0.0004355285882030937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.102877053654468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.713589283902264e-05, "output_power": 8.088306610322712e-05, "reference_active": true, "delta_db": -0.3233947949303148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3909625300627975e-05, "output_power": 1.2040478013932361e-05, "reference_active": true, "delta_db": -0.6267170206011906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.457858491278464e-06, "output_power": 3.3957006301040376e-06, "reference_active": true, "delta_db": -1.1819688304113245}], "active_band_count": 3, "active_band_mae_db": 0.71069354864761}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.324411735838957e-05, "log_rms_mae_db": 0.730085237236461, "rms_envelope_nrmse": 0.06929764467213305, "envelope_correlation": 0.9966707488478093, "rms_gain_error_db": -0.2551723916129301, "reference_mean_power": 0.0004639006426222396, "output_mean_power": 0.0004374291642967098}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.97726981805197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.049771362210352e-05, "output_power": 7.51570995581537e-05, "reference_active": true, "delta_db": -0.29813533689428084}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.625262232859704e-06, "output_power": 5.856653894791937e-06, "reference_active": true, "delta_db": -0.5355351296042081}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8383435618714896e-07, "output_power": 2.136797496509278e-07, "reference_active": true, "delta_db": -1.23301596458336}], "active_band_count": 3, "active_band_mae_db": 0.688895477027283}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.433462630283978e-05, "log_rms_mae_db": 1.478754944546487, "rms_envelope_nrmse": 0.20784450623162548, "envelope_correlation": 0.9690889433258996, "rms_gain_error_db": 0.27359577048309847, "reference_mean_power": 4.2981917565965076e-05, "output_mean_power": 4.577679300658406e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5620745474492675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.382029340071467e-06, "output_power": 5.820051220761989e-06, "reference_active": true, "delta_db": -0.40031989662068945}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1967382600492452e-06, "output_power": 1.3259237951148398e-06, "reference_active": true, "delta_db": 0.44519388792598763}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059624559400323e-07, "output_power": 5.41968002563e-07, "reference_active": true, "delta_db": 0.29855354860412353}], "active_band_count": 3, "active_band_mae_db": 0.3813557777169336}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.1780456996026785e-05, "log_rms_mae_db": 0.9207573995250138, "rms_envelope_nrmse": 0.13895309668989062, "envelope_correlation": 0.9857348589261969, "rms_gain_error_db": 0.2733342451141761, "reference_mean_power": 4.317499760725885e-05, "output_mean_power": 4.59796590640965e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4490976093596605e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.797984044318825e-06, "output_power": 5.3825042841556495e-06, "reference_active": true, "delta_db": -0.322926317015611}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.94257411489492e-07, "output_power": 6.58641586155142e-07, "reference_active": true, "delta_db": 0.4467454139246766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.104930954274029e-08, "output_power": 3.2469597247166166e-08, "reference_active": true, "delta_db": 1.882390469483788}], "active_band_count": 3, "active_band_mae_db": 0.8840207334746918}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8831956814106386, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c3100316c5c2458ce3e5725d4a8790c5690b82dc8e5497f0f9b688c1a86fde4b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9905017226473354, "standard_deviation": 0.0013111164621254416, "minimum": 0.9882911022928643, "maximum": 0.9920430036771944, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2236992757477922, "standard_deviation": 0.017951221853300794, "minimum": 1.1970762961557282, "maximum": 1.2492292819618347, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6383928224940973, "standard_deviation": 0.22586807532131906, "minimum": 0.15573538734707087, "maximum": 0.9189813411734834, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1128960883185555, "standard_deviation": 0.007087897847738707, "minimum": 0.10330080295579726, "maximum": 0.12383179024762256, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970204977705233, "standard_deviation": 0.0006791961993259334, "minimum": 0.995800878796476, "maximum": 0.997867485928331, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.715866004433447, "standard_deviation": 0.01973122348420771, "minimum": 0.6888378790072368, "maximum": 0.7516097503690989, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6612549376780632, "standard_deviation": 0.26003123887023233, "minimum": 0.1867877203198334, "maximum": 1.0001972051212797, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06504324590916129, "standard_deviation": 0.0065790412848813805, "minimum": 0.055575373209530245, "maximum": 0.07322305693078009, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9619261814995035, "standard_deviation": 0.008433464367519837, "minimum": 0.9447375972101764, "maximum": 0.9690889433258996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.530366044307471, "standard_deviation": 0.03291122736999381, "minimum": 1.478754944546487, "maximum": 1.5732328335448187, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.43154516337118753, "standard_deviation": 0.12281486694165031, "minimum": 0.2609701919502182, "maximum": 0.6172016184976764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22378108772212785, "standard_deviation": 0.025390872953779252, "minimum": 0.20144135010335776, "maximum": 0.27546169887736593, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.984644227856729, "standard_deviation": 0.005904272187277381, "minimum": 0.9732077148253846, "maximum": 0.989093872872549, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9465919707302826, "standard_deviation": 0.018239037116104327, "minimum": 0.9207573995250138, "maximum": 0.9787000808690411, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6453545863829632, "standard_deviation": 0.23285393704129523, "minimum": 0.31562451403725905, "maximum": 0.9865202227987737, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13771657322493225, "standard_deviation": 0.02565304165694001, "minimum": 0.11523607523833213, "maximum": 0.18638121324382062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.99433453761295, "standard_deviation": 0.0004349450515695915, "minimum": 0.9937852339209668, "maximum": 0.9952377358302084, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.222962153044721, "standard_deviation": 0.021832647316934312, "minimum": 1.1839221671990436, "maximum": 1.2623658758034393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3622495309167931, "standard_deviation": 0.19301783258492536, "minimum": 0.037663082977748105, "maximum": 0.6849859922067436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08359583655161933, "standard_deviation": 0.003306430016739189, "minimum": 0.07659250599746933, "maximum": 0.08745998383389693, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980136862138688, "standard_deviation": 0.0003135593364960421, "minimum": 0.9974204493082061, "maximum": 0.9984143055122359, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6548434224155688, "standard_deviation": 0.026526416346452378, "minimum": 0.6266032718395658, "maximum": 0.695604409130189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.37892315230052537, "standard_deviation": 0.12715362464076518, "minimum": 0.19405489367389717, "maximum": 0.5537607216997845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.048251705173320975, "standard_deviation": 0.0037333190256135405, "minimum": 0.043571425944201075, "maximum": 0.05464913184605577, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11977118981996301, "standard_deviation": 0.026465975919655337, "minimum": 0.07208792970154511, "maximum": 0.1596928282994905, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.531541803168235, "standard_deviation": 0.0099828018125947, "minimum": 5.519260582955861, "maximum": 5.545836459301818, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3889036805155877, "standard_deviation": 0.015287423722805414, "minimum": 0.3605487822765336, "maximum": 0.40544294220653493, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07812446317660705, "standard_deviation": 0.03243311726043314, "minimum": -0.14040633531804167, "maximum": -0.029521512633780018, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956244977666673, "standard_deviation": 0.00032473176765311044, "minimum": 0.9952830177207672, "maximum": 0.9963132204189044, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6484000707986911, "standard_deviation": 0.014792898808201739, "minimum": 0.6295746143802797, "maximum": 0.6743640286914772, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3697241461967524, "standard_deviation": 0.14374672928918414, "minimum": 0.17193780118919408, "maximum": 0.5696075138161939, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06708937119722184, "standard_deviation": 0.002514259355383849, "minimum": 0.06164572886838216, "maximum": 0.06979915442323258, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985291823167919, "standard_deviation": 0.00023285622625291994, "minimum": 0.9982364038398164, "maximum": 0.9988649235585315, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.34326843169587434, "standard_deviation": 0.012981916232420164, "minimum": 0.31513322173162306, "maximum": 0.3521252905432537, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35343812656633616, "standard_deviation": 0.13606727744096248, "minimum": 0.14060651637883184, "maximum": 0.505507100812104, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038123292878492435, "standard_deviation": 0.0032282836525617535, "minimum": 0.03366441965784252, "maximum": 0.041633585515324326, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0000", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.8998730182647705, "predicted_mos": 2.8998730182647705, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0720f721cb7269b9c15ab96e3b5e5c42b6506a7a46e11cf64b0ebef343918ae5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7789830998517573, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6352536332658129, "全帯域_10ms_rms_envelope_nrmse": 0.07256790157397382, "全帯域_10ms_envelope_correlation": 0.9948668303262039, "全帯域_10ms_modulation_mae_db": 0.2563068475793229, "全帯域_40ms_log_rms_mae_db": 0.3349159940295125, "全帯域_40ms_rms_envelope_nrmse": 0.03947501949762536, "全帯域_40ms_envelope_correlation": 0.9984229980631415, "全帯域_40ms_modulation_mae_db": 0.2706724194956102, "300–1000 Hz_10ms_log_rms_mae_db": 1.2694184656830665, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09657812450482742, "300–1000 Hz_10ms_envelope_correlation": 0.9924107573707172, "300–1000 Hz_10ms_modulation_mae_db": 0.2268107985660677, "300–1000 Hz_40ms_log_rms_mae_db": 0.6744008128256574, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05462205041504942, "300–1000 Hz_40ms_envelope_correlation": 0.9974410647307749, "300–1000 Hz_40ms_modulation_mae_db": 0.531032085654324, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2448085586349682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10722362486184918, "1000–3000 Hz_10ms_envelope_correlation": 0.9916113220032973, "1000–3000 Hz_10ms_modulation_mae_db": 0.6600378505229343, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7310324441430806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061293855981478455, "1000–3000 Hz_40ms_envelope_correlation": 0.9975861858229731, "1000–3000 Hz_40ms_modulation_mae_db": 0.4872215869561661, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5377209743474043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18742574915513183, "3000–6000 Hz_10ms_envelope_correlation": 0.9730013990943907, "3000–6000 Hz_10ms_modulation_mae_db": 0.6657851237982213, "3000–6000 Hz_40ms_log_rms_mae_db": 0.946436256150849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11381948398960808, "3000–6000 Hz_40ms_envelope_correlation": 0.9893210421917826, "3000–6000 Hz_40ms_modulation_mae_db": 0.8308167784730927, "mr_stft_db": 5.524818518833437, "band_mae_db": 0.10577760478116555, "rms_gain_error_db": -0.08519767527235311, "output_peak": 0.38961865369137666}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.632588234216454, "全帯域_10ms_rms_envelope_nrmse": 0.06810859753047833, "全帯域_10ms_envelope_correlation": 0.9954845567037935, "全帯域_10ms_modulation_mae_db": 0.6259650743185019, "全帯域_40ms_log_rms_mae_db": 0.3159999107141965, "全帯域_40ms_rms_envelope_nrmse": 0.03494799998242743, "全帯域_40ms_envelope_correlation": 0.9987798617554678, "全帯域_40ms_modulation_mae_db": 0.5370760369868796, "300–1000 Hz_10ms_log_rms_mae_db": 1.2206779642460637, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.091084061522883, "300–1000 Hz_10ms_envelope_correlation": 0.9933562046586601, "300–1000 Hz_10ms_modulation_mae_db": 0.7176970403906008, "300–1000 Hz_40ms_log_rms_mae_db": 0.6808056472860138, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053841801542062535, "300–1000 Hz_40ms_envelope_correlation": 0.9976565901750438, "300–1000 Hz_40ms_modulation_mae_db": 0.8208491559026668, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2098108683910367, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11784810797223404, "1000–3000 Hz_10ms_envelope_correlation": 0.9893869080216771, "1000–3000 Hz_10ms_modulation_mae_db": 0.06327771185294344, "1000–3000 Hz_40ms_log_rms_mae_db": 0.686377066553387, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06124141888802602, "1000–3000 Hz_40ms_envelope_correlation": 0.9969672124744838, "1000–3000 Hz_40ms_modulation_mae_db": 0.3656262076266268, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4890614391184147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20519212574717816, "3000–6000 Hz_10ms_envelope_correlation": 0.9673232869373514, "3000–6000 Hz_10ms_modulation_mae_db": 0.19578199684093134, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9174465426027344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11838327081833762, "3000–6000 Hz_40ms_envelope_correlation": 0.9885369737510658, "3000–6000 Hz_40ms_modulation_mae_db": 0.7098121885069094, "mr_stft_db": 5.5279858159588064, "band_mae_db": 0.06405169968641246, "rms_gain_error_db": -0.0817981809604312, "output_peak": 0.39466804684314416}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6269108439597285, "全帯域_10ms_rms_envelope_nrmse": 0.06861185192270566, "全帯域_10ms_envelope_correlation": 0.9955791775672271, "全帯域_10ms_modulation_mae_db": 0.922258003782341, "全帯域_40ms_log_rms_mae_db": 0.32272566074047504, "全帯域_40ms_rms_envelope_nrmse": 0.037428967025737496, "全帯域_40ms_envelope_correlation": 0.9987887880526831, "全帯域_40ms_modulation_mae_db": 0.9172311166553907, "300–1000 Hz_10ms_log_rms_mae_db": 1.1485863485122478, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08891640098106499, "300–1000 Hz_10ms_envelope_correlation": 0.9938017777342557, "300–1000 Hz_10ms_modulation_mae_db": 0.905595743604512, "300–1000 Hz_40ms_log_rms_mae_db": 0.6779673788787463, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05136626126798524, "300–1000 Hz_40ms_envelope_correlation": 0.9980502803649782, "300–1000 Hz_40ms_modulation_mae_db": 0.897133603220052, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1986373877661758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11150608903057183, "1000–3000 Hz_10ms_envelope_correlation": 0.9905323307854302, "1000–3000 Hz_10ms_modulation_mae_db": 0.6546183845292494, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6698649506428496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06485242246326753, "1000–3000 Hz_40ms_envelope_correlation": 0.9967558029441966, "1000–3000 Hz_40ms_modulation_mae_db": 0.540699957468629, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4833524515376306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20402290035359516, "3000–6000 Hz_10ms_envelope_correlation": 0.968084175307725, "3000–6000 Hz_10ms_modulation_mae_db": 0.15989042360838698, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9146437569448983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11742615652285349, "3000–6000 Hz_40ms_envelope_correlation": 0.9886524976070215, "3000–6000 Hz_40ms_modulation_mae_db": 0.4916973037059833, "mr_stft_db": 5.521765777328739, "band_mae_db": 0.13995382227618813, "rms_gain_error_db": -0.14593413208580858, "output_peak": 0.37312910504245556}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6559899305852959, "全帯域_10ms_rms_envelope_nrmse": 0.0641904454768823, "全帯域_10ms_envelope_correlation": 0.9959995465332819, "全帯域_10ms_modulation_mae_db": 0.04315231386731879, "全帯域_40ms_log_rms_mae_db": 0.33304457678584504, "全帯域_40ms_rms_envelope_nrmse": 0.03280162730628937, "全帯域_40ms_envelope_correlation": 0.9989342829530221, "全帯域_40ms_modulation_mae_db": 0.13561495787550434, "300–1000 Hz_10ms_log_rms_mae_db": 1.2132127466938947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08308827804212653, "300–1000 Hz_10ms_envelope_correlation": 0.9944064420908801, "300–1000 Hz_10ms_modulation_mae_db": 0.18496026209970876, "300–1000 Hz_40ms_log_rms_mae_db": 0.6738142038580279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04510739966931124, "300–1000 Hz_40ms_envelope_correlation": 0.9982834954320301, "300–1000 Hz_40ms_modulation_mae_db": 0.04737347887473692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1373169841317299, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10486961465274038, "1000–3000 Hz_10ms_envelope_correlation": 0.9918339614366398, "1000–3000 Hz_10ms_modulation_mae_db": 0.7791796278338362, "1000–3000 Hz_40ms_log_rms_mae_db": 0.64163974679196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06504766710893543, "1000–3000 Hz_40ms_envelope_correlation": 0.996978701950074, "1000–3000 Hz_40ms_modulation_mae_db": 0.7786797411180477, "3000–6000 Hz_10ms_log_rms_mae_db": 1.555570127311784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2021663834280648, "3000–6000 Hz_10ms_envelope_correlation": 0.9692411690639292, "3000–6000 Hz_10ms_modulation_mae_db": 0.2895274323930154, "3000–6000 Hz_40ms_log_rms_mae_db": 0.930641181437604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11211618970424143, "3000–6000 Hz_40ms_envelope_correlation": 0.9899262535281098, "3000–6000 Hz_40ms_modulation_mae_db": 0.5815919246242012, "mr_stft_db": 5.5259478615674595, "band_mae_db": 0.10307551197002957, "rms_gain_error_db": -0.08185421251815901, "output_peak": 0.4035092894786945}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6189569132173701, "全帯域_10ms_rms_envelope_nrmse": 0.07301434094275232, "全帯域_10ms_envelope_correlation": 0.9948335044254268, "全帯域_10ms_modulation_mae_db": 0.9175719452801268, "全帯域_40ms_log_rms_mae_db": 0.3224839332812416, "全帯域_40ms_rms_envelope_nrmse": 0.038602863792483046, "全帯域_40ms_envelope_correlation": 0.9985407271524588, "全帯域_40ms_modulation_mae_db": 0.7844349063832929, "300–1000 Hz_10ms_log_rms_mae_db": 1.177459406195914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09208883076576041, "300–1000 Hz_10ms_envelope_correlation": 0.9931522819131233, "300–1000 Hz_10ms_modulation_mae_db": 1.0642368329564107, "300–1000 Hz_40ms_log_rms_mae_db": 0.6373341824189862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05058805389973433, "300–1000 Hz_40ms_envelope_correlation": 0.9978918694903777, "300–1000 Hz_40ms_modulation_mae_db": 0.9023158045132721, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1932711529476328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11893930937601888, "1000–3000 Hz_10ms_envelope_correlation": 0.9891729425166742, "1000–3000 Hz_10ms_modulation_mae_db": 0.8936021253972447, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6702871848534507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0634735988625865, "1000–3000 Hz_40ms_envelope_correlation": 0.9968725842366674, "1000–3000 Hz_40ms_modulation_mae_db": 0.822709223116385, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5296596279460424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21733048693170598, "3000–6000 Hz_10ms_envelope_correlation": 0.9635365069017728, "3000–6000 Hz_10ms_modulation_mae_db": 0.23984129847854552, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9540076984311167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12036660176023567, "3000–6000 Hz_40ms_envelope_correlation": 0.988031540939802, "3000–6000 Hz_40ms_modulation_mae_db": 0.4934405395377775, "mr_stft_db": 5.517149020989038, "band_mae_db": 0.10374714896994053, "rms_gain_error_db": -0.09956258169146152, "output_peak": 0.3859916728580368}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6423161178438895, "全帯域_10ms_rms_envelope_nrmse": 0.06957075121085149, "全帯域_10ms_envelope_correlation": 0.995414175684946, "全帯域_10ms_modulation_mae_db": 0.4985442194984695, "全帯域_40ms_log_rms_mae_db": 0.3573000730106652, "全帯域_40ms_rms_envelope_nrmse": 0.03639884988076132, "全帯域_40ms_envelope_correlation": 0.998827868592095, "全帯域_40ms_modulation_mae_db": 0.3478629845357637, "300–1000 Hz_10ms_log_rms_mae_db": 1.2109392934231404, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09076212343434219, "300–1000 Hz_10ms_envelope_correlation": 0.9933375800875938, "300–1000 Hz_10ms_modulation_mae_db": 0.39895520235337295, "300–1000 Hz_40ms_log_rms_mae_db": 0.6830117123373216, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051671909685656735, "300–1000 Hz_40ms_envelope_correlation": 0.9977653046968897, "300–1000 Hz_40ms_modulation_mae_db": 0.2469323522223987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1875327463855723, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11745800873954404, "1000–3000 Hz_10ms_envelope_correlation": 0.9902197730207133, "1000–3000 Hz_10ms_modulation_mae_db": 0.6892496241649173, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6817301452061986, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07091935102492566, "1000–3000 Hz_40ms_envelope_correlation": 0.997043078333884, "1000–3000 Hz_40ms_modulation_mae_db": 0.6084577905542313, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5375385223588334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1979175435259294, "3000–6000 Hz_10ms_envelope_correlation": 0.9698322840529348, "3000–6000 Hz_10ms_modulation_mae_db": 0.3739331581635213, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9491113767743348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12901615628928487, "3000–6000 Hz_40ms_envelope_correlation": 0.9862588419527696, "3000–6000 Hz_40ms_modulation_mae_db": 0.7607929994773602, "mr_stft_db": 5.518195241146366, "band_mae_db": 0.13317171270074915, "rms_gain_error_db": -0.13841475524201913, "output_peak": 0.37986038684159057}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6658373718239913, "全帯域_10ms_rms_envelope_nrmse": 0.07319265629282717, "全帯域_10ms_envelope_correlation": 0.9947933730959377, "全帯域_10ms_modulation_mae_db": 0.48049733195131816, "全帯域_40ms_log_rms_mae_db": 0.34951543362095283, "全帯域_40ms_rms_envelope_nrmse": 0.03904595182184749, "全帯域_40ms_envelope_correlation": 0.998483739940884, "全帯域_40ms_modulation_mae_db": 0.39953445332693577, "300–1000 Hz_10ms_log_rms_mae_db": 1.2229038850741518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09191941731980754, "300–1000 Hz_10ms_envelope_correlation": 0.9931650305414694, "300–1000 Hz_10ms_modulation_mae_db": 0.5142098212609341, "300–1000 Hz_40ms_log_rms_mae_db": 0.6431076074497876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05265844656162422, "300–1000 Hz_40ms_envelope_correlation": 0.9976858232813225, "300–1000 Hz_40ms_modulation_mae_db": 0.46510162163645674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.200593241961612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11097724376919739, "1000–3000 Hz_10ms_envelope_correlation": 0.9906238827926811, "1000–3000 Hz_10ms_modulation_mae_db": 0.4374242351901698, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6827585567524438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059290946550205066, "1000–3000 Hz_40ms_envelope_correlation": 0.9973256752337513, "1000–3000 Hz_40ms_modulation_mae_db": 0.34729542908675354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5078264106406913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20094540456065899, "3000–6000 Hz_10ms_envelope_correlation": 0.9686701121375574, "3000–6000 Hz_10ms_modulation_mae_db": 0.3566558108392766, "3000–6000 Hz_40ms_log_rms_mae_db": 0.941186915926683, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12587991270755217, "3000–6000 Hz_40ms_envelope_correlation": 0.9870297968932618, "3000–6000 Hz_40ms_modulation_mae_db": 0.6813513264470287, "mr_stft_db": 5.524411686707385, "band_mae_db": 0.10482672625679419, "rms_gain_error_db": -0.09294873441138332, "output_peak": 0.3700354796496056}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6813044900457121, "全帯域_10ms_rms_envelope_nrmse": 0.073097018236771, "全帯域_10ms_envelope_correlation": 0.9947608885028398, "全帯域_10ms_modulation_mae_db": 0.27326652056934625, "全帯域_40ms_log_rms_mae_db": 0.3791882605952575, "全帯域_40ms_rms_envelope_nrmse": 0.04339841510624332, "全帯域_40ms_envelope_correlation": 0.9980338923163088, "全帯域_40ms_modulation_mae_db": 0.18288127640835797, "300–1000 Hz_10ms_log_rms_mae_db": 1.2956282200955493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08909273762551703, "300–1000 Hz_10ms_envelope_correlation": 0.9935529724130441, "300–1000 Hz_10ms_modulation_mae_db": 0.1971453544063829, "300–1000 Hz_40ms_log_rms_mae_db": 0.6936781490666011, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053621315267037555, "300–1000 Hz_40ms_envelope_correlation": 0.9974973057150872, "300–1000 Hz_40ms_modulation_mae_db": 0.2769431874427977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2013041664852107, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11215025276724, "1000–3000 Hz_10ms_envelope_correlation": 0.9903980936815686, "1000–3000 Hz_10ms_modulation_mae_db": 0.5436237564100169, "1000–3000 Hz_40ms_log_rms_mae_db": 0.709408050595395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06293586900345592, "1000–3000 Hz_40ms_envelope_correlation": 0.9968885390669232, "1000–3000 Hz_40ms_modulation_mae_db": 0.4076336114023838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5409440286573368, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2536984754033133, "3000–6000 Hz_10ms_envelope_correlation": 0.952116118828356, "3000–6000 Hz_10ms_modulation_mae_db": 0.23880186389733424, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9751912263291621, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17974327578887675, "3000–6000 Hz_40ms_envelope_correlation": 0.9781554259314157, "3000–6000 Hz_40ms_modulation_mae_db": 0.4558152001109907, "mr_stft_db": 5.526037992842853, "band_mae_db": 0.15187518447413761, "rms_gain_error_db": -0.035421155963028216, "output_peak": 0.3832735055239592}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.331071153790594, "spectral_convergence": 0.23420526984602844, "spectral_convergence_status": "ok", "amplitude_floor": 1.174323222508143e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.307355884778254, "spectral_convergence": 0.24017412490340312, "spectral_convergence_status": "ok", "amplitude_floor": 1.1752576893073956e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.936028517931465, "spectral_convergence": 0.25544260974164434, "spectral_convergence_status": "ok", "amplitude_floor": 9.275961087692859e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.278493760532488e-05, "output_power": 9.274815127118876e-05, "delta_db": -0.001722183598716957, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009129851997232178, "output_power": 0.0009188509638950119, "delta_db": 0.027813379487809056, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007449145920476126, "output_power": 0.0007205676300881406, "delta_db": -0.14431733200177913, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00040544152038629675, "output_power": 0.0003802055569974059, "delta_db": -0.27909761665357136, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011224761509504284, "output_power": 0.00010991028356744777, "delta_db": -0.09138794052422781, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.007390871044891e-05, "output_power": 5.112628196736829e-05, "delta_db": 0.09032717642088892, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10577760478116555, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08519767527235311, "pitch": null, "output_rms_dbfs": -26.433224503045636, "output_peak": 0.38961865369137666, "reference_rms_dbfs": -26.348026827773285, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.524818518833437}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001795198121335632, "log_rms_mae_db": 0.6352536332658129, "rms_envelope_nrmse": 0.07256790157397382, "envelope_correlation": 0.9948668303262039, "rms_gain_error_db": -0.08434345025586834, "reference_mean_power": 0.002319729266015563, "output_mean_power": 0.0022751129193737773}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2030894540936354e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042099620393065603, "output_power": 0.0003982102780982411, "reference_active": true, "delta_db": -0.24165714581131742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010212932714007922, "output_power": 9.276937401745365e-05, "reference_active": true, "delta_db": -0.41745844543258587}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7753962605616007e-05, "output_power": 1.7310707002611117e-05, "reference_active": true, "delta_db": -0.10980495149406547}], "active_band_count": 3, "active_band_mae_db": 0.2563068475793229}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017377061993537507, "log_rms_mae_db": 0.3349159940295125, "rms_envelope_nrmse": 0.03947501949762536, "envelope_correlation": 0.9984229980631415, "rms_gain_error_db": -0.08319509387615284, "reference_mean_power": 0.002329667623369792, "output_mean_power": 0.002285464367964065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1345793651623815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000381470345516612, "output_power": 0.00035876609499892805, "reference_active": true, "delta_db": -0.2664938917299005}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.449258478788481e-05, "output_power": 5.0175260837082497e-05, "reference_active": true, "delta_db": -0.3584776975668351}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.451644818904224e-07, "output_power": 9.867609625318216e-07, "reference_active": true, "delta_db": 0.1870456691900951}], "active_band_count": 3, "active_band_mae_db": 0.2706724194956102}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001705289920117241, "log_rms_mae_db": 1.2694184656830665, "rms_envelope_nrmse": 0.09657812450482742, "envelope_correlation": 0.9924107573707172, "rms_gain_error_db": -0.07774884012195413, "reference_mean_power": 0.0014327712363456494, "output_mean_power": 0.0014073495228866376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.476686730813627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004044339102676846, "output_power": 0.0003929937262062458, "reference_active": true, "delta_db": -0.12461945103000012}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.628475293024379e-05, "output_power": 8.536756666869008e-05, "reference_active": true, "delta_db": -0.5226461793185544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2048085119397813e-05, "output_power": 1.1956425183080931e-05, "reference_active": true, "delta_db": -0.03316676534964853}], "active_band_count": 3, "active_band_mae_db": 0.2268107985660677}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015393777861692204, "log_rms_mae_db": 0.6744008128256574, "rms_envelope_nrmse": 0.05462205041504942, "envelope_correlation": 0.9974410647307749, "rms_gain_error_db": -0.07556062283628445, "reference_mean_power": 0.0014385306131629134, "output_mean_power": 0.0014137188416718982}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.958921932585308e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037465347734378165, "output_power": 0.00036073981887703563, "reference_active": true, "delta_db": -0.16435684964391795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.015607336171049e-05, "output_power": 4.435750356752667e-05, "reference_active": true, "delta_db": -0.5335643278914328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018660953269709e-07, "output_power": 9.85415431223675e-07, "reference_active": true, "delta_db": 0.8951750794276213}], "active_band_count": 3, "active_band_mae_db": 0.531032085654324}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.828766409891504e-05, "log_rms_mae_db": 1.2448085586349682, "rms_envelope_nrmse": 0.10722362486184918, "envelope_correlation": 0.9916113220032973, "rms_gain_error_db": -0.26141153213470947, "reference_mean_power": 0.00046178379763570874, "output_mean_power": 0.00043480801174810706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.102877053654468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.713589283902264e-05, "output_power": 7.823383549655148e-05, "reference_active": true, "delta_db": -0.46802463044185816}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3909625300627975e-05, "output_power": 1.2344149472970604e-05, "reference_active": true, "delta_db": -0.5185425918934105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.457858491278464e-06, "output_power": 3.546268752557395e-06, "reference_active": true, "delta_db": -0.9935463292335344}], "active_band_count": 3, "active_band_mae_db": 0.6600378505229343}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.324411735838957e-05, "log_rms_mae_db": 0.7310324441430806, "rms_envelope_nrmse": 0.061293855981478455, "envelope_correlation": 0.9975861858229731, "rms_gain_error_db": -0.2620972814214148, "reference_mean_power": 0.0004639006426222396, "output_mean_power": 0.0004367322327998337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.97726981805197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.049771362210352e-05, "output_power": 7.220537761760272e-05, "reference_active": true, "delta_db": -0.4721400167953447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.625262232859704e-06, "output_power": 5.848771052873894e-06, "reference_active": true, "delta_db": -0.5413845118856031}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8383435618714896e-07, "output_power": 2.560064689049703e-07, "reference_active": true, "delta_db": -0.4481402321875504}], "active_band_count": 3, "active_band_mae_db": 0.4872215869561661}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.433462630283978e-05, "log_rms_mae_db": 1.5377209743474043, "rms_envelope_nrmse": 0.18742574915513183, "envelope_correlation": 0.9730013990943907, "rms_gain_error_db": -0.020387618892269456, "reference_mean_power": 4.2981917565965076e-05, "output_mean_power": 4.278061514399788e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5620745474492675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.382029340071467e-06, "output_power": 5.75621626600223e-06, "reference_active": true, "delta_db": -0.44821694053141026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1967382600492452e-06, "output_power": 1.3754284902511556e-06, "reference_active": true, "delta_db": 0.604388401920612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059624559400323e-07, "output_power": 6.289169431483924e-07, "reference_active": true, "delta_db": 0.9447500289426416}], "active_band_count": 3, "active_band_mae_db": 0.6657851237982213}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.1780456996026785e-05, "log_rms_mae_db": 0.946436256150849, "rms_envelope_nrmse": 0.11381948398960808, "envelope_correlation": 0.9893210421917826, "rms_gain_error_db": -0.02120479528114539, "reference_mean_power": 4.317499760725885e-05, "output_mean_power": 4.296470583239205e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4490976093596605e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.797984044318825e-06, "output_power": 5.371458043148933e-06, "reference_active": true, "delta_db": -0.33184828095075103}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.94257411489492e-07, "output_power": 6.77040686991777e-07, "reference_active": true, "delta_db": 0.5664016165513236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.104930954274029e-08, "output_power": 3.0384906096592375e-08, "reference_active": true, "delta_db": 1.5942004379172032}], "active_band_count": 3, "active_band_mae_db": 0.8308167784730927}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8878799243160352, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "274144969b5728b10be60031e92d8bbc52d13023cf71d9bbc14b805943bedb4e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9904724017823352, "standard_deviation": 0.0009345553577146359, "minimum": 0.9891729425166742, "maximum": 0.9918339614366398, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1966593883379923, "standard_deviation": 0.02963724092457046, "minimum": 1.1373169841317299, "maximum": 1.2448085586349682, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5901266644876639, "standard_deviation": 0.2534788992659509, "minimum": 0.06327771185294344, "maximum": 0.8936021253972447, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11262153139617447, "standard_deviation": 0.005127273566341021, "minimum": 0.10486961465274038, "maximum": 0.11893930937601888, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970522225078692, "standard_deviation": 0.0002724323098967585, "minimum": 0.9967558029441966, "maximum": 0.9975861858229731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6841372681923457, "standard_deviation": 0.026864860042229242, "minimum": 0.64163974679196, "maximum": 0.7310324441430806, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5447904434161528, "standard_deviation": 0.18096708016717042, "minimum": 0.34729542908675354, "maximum": 0.822709223116385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06363189123536007, "standard_deviation": 0.003528157330897318, "minimum": 0.059290946550205066, "maximum": 0.07091935102492566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9664756315405022, "standard_deviation": 0.006374689828152877, "minimum": 0.952116118828356, "maximum": 0.9730013990943907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.522709197739767, "standard_deviation": 0.02620943946709045, "minimum": 1.4833524515376306, "maximum": 1.555570127311784, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3150271385024041, "standard_deviation": 0.15974724317763286, "minimum": 0.15989042360838698, "maximum": 0.6657851237982213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2085873836381972, "standard_deviation": 0.020018220590180557, "minimum": 0.18742574915513183, "maximum": 0.2536984754033133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9869890465994036, "standard_deviation": 0.003758476964991351, "minimum": 0.9781554259314157, "maximum": 0.9899262535281098, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9410831193246728, "standard_deviation": 0.019965288387668315, "minimum": 0.9146437569448983, "maximum": 0.9751912263291621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.625664782610418, "standard_deviation": 0.1397705194902905, "minimum": 0.4558152001109907, "maximum": 0.8308167784730927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12709388094762375, "standard_deviation": 0.02201384109324687, "minimum": 0.11211618970424143, "maximum": 0.17974327578887675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933978808512179, "standard_deviation": 0.0005730502118170774, "minimum": 0.9924107573707172, "maximum": 0.9944064420908801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2198532912405036, "standard_deviation": 0.04661132200408998, "minimum": 1.1485863485122478, "maximum": 1.2956282200955493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5262013819547487, "standard_deviation": 0.3382277963791859, "minimum": 0.18496026209970876, "maximum": 1.0642368329564107, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09044124677454113, "standard_deviation": 0.0038038841591897005, "minimum": 0.08308827804212653, "maximum": 0.09657812450482742, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997783966735813, "standard_deviation": 0.00028235145154281884, "minimum": 0.9974410647307749, "maximum": 0.9982834954320301, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6705149617651427, "standard_deviation": 0.019756556870308015, "minimum": 0.6373341824189862, "maximum": 0.6936781490666011, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5234601611833382, "standard_deviation": 0.32490954426192853, "minimum": 0.04737347887473692, "maximum": 0.9023158045132721, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05168465478855766, "standard_deviation": 0.002989143563285058, "minimum": 0.04510739966931124, "maximum": 0.05462205041504942, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11330992638942715, "standard_deviation": 0.02762744566123202, "minimum": 0.06405169968641246, "maximum": 0.15187518447413761, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.52328898942176, "standard_deviation": 0.003896012040902848, "minimum": 5.517149020989038, "maximum": 5.5279858159588064, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3850107674911079, "standard_deviation": 0.011036503064434953, "minimum": 0.3700354796496056, "maximum": 0.4035092894786945, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09514142851808052, "standard_deviation": 0.03484007460911072, "minimum": -0.14593413208580858, "maximum": -0.035421155963028216, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9952165066049571, "standard_deviation": 0.0004646349823516251, "minimum": 0.9947608885028398, "maximum": 0.9959995465332819, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6448946918697818, "standard_deviation": 0.02116131467388215, "minimum": 0.6189569132173701, "maximum": 0.6813044900457121, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5021952821058432, "standard_deviation": 0.31359288940491825, "minimum": 0.04315231386731879, "maximum": 0.922258003782341, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07029419539840526, "standard_deviation": 0.003256029027216173, "minimum": 0.0641904454768823, "maximum": 0.07319265629282717, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986015198532576, "standard_deviation": 0.0002928869004572552, "minimum": 0.9980338923163088, "maximum": 0.9989342829530221, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.33939673034726825, "standard_deviation": 0.021307574113215915, "minimum": 0.3159999107141965, "maximum": 0.3791882605952575, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44691351895846687, "standard_deviation": 0.28106560950821935, "minimum": 0.13561495787550434, "maximum": 0.9172311166553907, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037762461801676854, "standard_deviation": 0.0031971455150385034, "minimum": 0.03280162730628937, "maximum": 0.04339841510624332, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3729-6852-0007", "language": "en", "style": "normal", "speaker": "librispeech_3729", "source_utterance_id": "3729-6852-0007", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3729-6852-0007.wav", "source_file_sha256": "4cdf6f8d8dc77324819abdfc00843cd3fad449023658f1c20d9cf4d717bfed19", "decoded_pcm_sha256": "d081e3d6cdc89188a9f3d229a815f9784efd7e72dfcfa5317b4ac90fa9d86ff1", "original_source_sha256": "9d9bbbeeff1bfb647e2698db36f186cb439a8b2894772ca3cdfe9425c3e55a1a", "corpus_entry_id": "librispeech_3729-6852-0007"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.101187705993652, "predicted_mos": 4.101187705993652, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "27dac7420766d0f0fd74833f8ada599c85fe69b6ef7dc16f938fbfed928f23bb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.80583998200018, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3729-6852-0007", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "27dac7420766d0f0fd74833f8ada599c85fe69b6ef7dc16f938fbfed928f23bb", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "5ab238d03e046aeef6b327b51b75630087f055df4633f0a25c71d45f3af209b7"}, "rows": [{"case": "librispeech_3729-6852-0007", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.7882821559906006, "predicted_mos": 2.7882821559906006, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "05cc385fd36f23e275ac4928905a510caf088f7bcda30df1c4f6b8d4185b1256", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5029451890004566, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3729-6852-0007", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "05cc385fd36f23e275ac4928905a510caf088f7bcda30df1c4f6b8d4185b1256", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6004257103929255, "全帯域_10ms_rms_envelope_nrmse": 0.057450006407651605, "全帯域_10ms_envelope_correlation": 0.9963584003756585, "全帯域_10ms_modulation_mae_db": 0.11146754737985637, "全帯域_40ms_log_rms_mae_db": 0.32675356100071745, "全帯域_40ms_rms_envelope_nrmse": 0.032580191329230235, "全帯域_40ms_envelope_correlation": 0.9988052389834511, "全帯域_40ms_modulation_mae_db": 0.32126112340003715, "300–1000 Hz_10ms_log_rms_mae_db": 1.068843951426529, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07287288766470316, "300–1000 Hz_10ms_envelope_correlation": 0.9951130980483839, "300–1000 Hz_10ms_modulation_mae_db": 0.3454361114296849, "300–1000 Hz_40ms_log_rms_mae_db": 0.5991329236489847, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03938135864291475, "300–1000 Hz_40ms_envelope_correlation": 0.9985034995472414, "300–1000 Hz_40ms_modulation_mae_db": 0.563964770419073, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0901586049135499, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.102735899577425, "1000–3000 Hz_10ms_envelope_correlation": 0.9900141167882631, "1000–3000 Hz_10ms_modulation_mae_db": 0.07742792887247148, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6046748790603182, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05734386965948882, "1000–3000 Hz_40ms_envelope_correlation": 0.9971272294834065, "1000–3000 Hz_40ms_modulation_mae_db": 0.08733447602955573, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6233510798263344, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2605875020165354, "3000–6000 Hz_10ms_envelope_correlation": 0.9396859978614147, "3000–6000 Hz_10ms_modulation_mae_db": 0.5644976178888551, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0372125708233935, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19203025931244108, "3000–6000 Hz_40ms_envelope_correlation": 0.9641279833793891, "3000–6000 Hz_40ms_modulation_mae_db": 0.45463462577169517, "mr_stft_db": 5.476312053152767, "band_mae_db": 0.18501531994966705, "rms_gain_error_db": -0.07854288520422589, "output_peak": 0.32395386491502803}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5876513280581466, "全帯域_10ms_rms_envelope_nrmse": 0.06144155904890396, "全帯域_10ms_envelope_correlation": 0.9958379240109742, "全帯域_10ms_modulation_mae_db": 0.13526526053058938, "全帯域_40ms_log_rms_mae_db": 0.31191967465216525, "全帯域_40ms_rms_envelope_nrmse": 0.032205538199665304, "全帯域_40ms_envelope_correlation": 0.9988544388999733, "全帯域_40ms_modulation_mae_db": 0.28617453088567896, "300–1000 Hz_10ms_log_rms_mae_db": 1.0379113720159812, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07265706736760622, "300–1000 Hz_10ms_envelope_correlation": 0.9951773530018769, "300–1000 Hz_10ms_modulation_mae_db": 0.3137731384227999, "300–1000 Hz_40ms_log_rms_mae_db": 0.5649479204073992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03881016607799224, "300–1000 Hz_40ms_envelope_correlation": 0.9985978829026462, "300–1000 Hz_40ms_modulation_mae_db": 0.4731214991586155, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0868725386773623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11744359938050701, "1000–3000 Hz_10ms_envelope_correlation": 0.9866725069771232, "1000–3000 Hz_10ms_modulation_mae_db": 0.4609097033485296, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6257630090075568, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06462979012351806, "1000–3000 Hz_40ms_envelope_correlation": 0.9960860024143532, "1000–3000 Hz_40ms_modulation_mae_db": 0.3142160087730594, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6563801619575516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27567017015079726, "3000–6000 Hz_10ms_envelope_correlation": 0.9328901448132957, "3000–6000 Hz_10ms_modulation_mae_db": 0.46334699496790926, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0511423762528003, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1962002636909126, "3000–6000 Hz_40ms_envelope_correlation": 0.9627069916647036, "3000–6000 Hz_40ms_modulation_mae_db": 0.5032709761223574, "mr_stft_db": 5.460299199955325, "band_mae_db": 0.14635480944275633, "rms_gain_error_db": -0.0892835061523645, "output_peak": 0.30988486124186193}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6030757041348778, "全帯域_10ms_rms_envelope_nrmse": 0.05721960020138235, "全帯域_10ms_envelope_correlation": 0.9963990921399049, "全帯域_10ms_modulation_mae_db": 0.32898070322144657, "全帯域_40ms_log_rms_mae_db": 0.33903942775626633, "全帯域_40ms_rms_envelope_nrmse": 0.03313741448013441, "全帯域_40ms_envelope_correlation": 0.9987694382710113, "全帯域_40ms_modulation_mae_db": 0.4459577798461381, "300–1000 Hz_10ms_log_rms_mae_db": 1.0114806151901787, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07664221128713677, "300–1000 Hz_10ms_envelope_correlation": 0.9945972710195159, "300–1000 Hz_10ms_modulation_mae_db": 0.35105795305231213, "300–1000 Hz_40ms_log_rms_mae_db": 0.6054649229215381, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045523145091622176, "300–1000 Hz_40ms_envelope_correlation": 0.9979904934811706, "300–1000 Hz_40ms_modulation_mae_db": 0.5207159207102554, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1002763477998692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09656225202464788, "1000–3000 Hz_10ms_envelope_correlation": 0.9910896069509579, "1000–3000 Hz_10ms_modulation_mae_db": 0.44536731440196586, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6404191378198626, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05459256311479155, "1000–3000 Hz_40ms_envelope_correlation": 0.9972498172143919, "1000–3000 Hz_40ms_modulation_mae_db": 0.45363725631942126, "3000–6000 Hz_10ms_log_rms_mae_db": 1.585973887599866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2568881203612135, "3000–6000 Hz_10ms_envelope_correlation": 0.944727115131788, "3000–6000 Hz_10ms_modulation_mae_db": 1.8707508440737968, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0249868690163266, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18968436763056307, "3000–6000 Hz_40ms_envelope_correlation": 0.9672626695099897, "3000–6000 Hz_40ms_modulation_mae_db": 1.7396449107543204, "mr_stft_db": 5.456769839696773, "band_mae_db": 0.1124305651436585, "rms_gain_error_db": -0.08529885095481746, "output_peak": 0.31285554044225405}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5778502800278575, "全帯域_10ms_rms_envelope_nrmse": 0.057860735547472196, "全帯域_10ms_envelope_correlation": 0.9964045542137173, "全帯域_10ms_modulation_mae_db": 0.2773317845211516, "全帯域_40ms_log_rms_mae_db": 0.3292200423585897, "全帯域_40ms_rms_envelope_nrmse": 0.03327676708274629, "全帯域_40ms_envelope_correlation": 0.9988681399619758, "全帯域_40ms_modulation_mae_db": 0.3271388836420884, "300–1000 Hz_10ms_log_rms_mae_db": 1.057346254568323, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07459017137730829, "300–1000 Hz_10ms_envelope_correlation": 0.9949714492904246, "300–1000 Hz_10ms_modulation_mae_db": 0.41399499438830456, "300–1000 Hz_40ms_log_rms_mae_db": 0.6138251766089211, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04256037352869056, "300–1000 Hz_40ms_envelope_correlation": 0.9983633050583958, "300–1000 Hz_40ms_modulation_mae_db": 0.42781781034111893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0894705858999276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0998360938423338, "1000–3000 Hz_10ms_envelope_correlation": 0.9905774199770357, "1000–3000 Hz_10ms_modulation_mae_db": 0.4553118325140118, "1000–3000 Hz_40ms_log_rms_mae_db": 0.653008497473464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05883177135706195, "1000–3000 Hz_40ms_envelope_correlation": 0.9969370432862886, "1000–3000 Hz_40ms_modulation_mae_db": 0.4282854342979517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6322640278038716, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2513490572467764, "3000–6000 Hz_10ms_envelope_correlation": 0.9478448199290559, "3000–6000 Hz_10ms_modulation_mae_db": 1.3931383202315188, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0288788599881036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18579290847735555, "3000–6000 Hz_40ms_envelope_correlation": 0.9691927524950897, "3000–6000 Hz_40ms_modulation_mae_db": 1.4745431564744134, "mr_stft_db": 5.450502250586987, "band_mae_db": 0.18629523326469663, "rms_gain_error_db": -0.11892828541055947, "output_peak": 0.3233468357232606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5916664589595676, "全帯域_10ms_rms_envelope_nrmse": 0.056591150133782006, "全帯域_10ms_envelope_correlation": 0.9966761408041568, "全帯域_10ms_modulation_mae_db": 0.7657619238804102, "全帯域_40ms_log_rms_mae_db": 0.3415738455020351, "全帯域_40ms_rms_envelope_nrmse": 0.03421932231660754, "全帯域_40ms_envelope_correlation": 0.9989034097768561, "全帯域_40ms_modulation_mae_db": 0.769902564971173, "300–1000 Hz_10ms_log_rms_mae_db": 1.0455371487781517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08123484716699399, "300–1000 Hz_10ms_envelope_correlation": 0.994142929056598, "300–1000 Hz_10ms_modulation_mae_db": 0.7359621328134193, "300–1000 Hz_40ms_log_rms_mae_db": 0.6224053125773685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05001699517410892, "300–1000 Hz_40ms_envelope_correlation": 0.9978425012344884, "300–1000 Hz_40ms_modulation_mae_db": 0.7883858689334798, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1032990447292457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09119686314236804, "1000–3000 Hz_10ms_envelope_correlation": 0.9923491217699312, "1000–3000 Hz_10ms_modulation_mae_db": 0.6225947019206113, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6374929197615357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054437047327194295, "1000–3000 Hz_40ms_envelope_correlation": 0.9975867481887668, "1000–3000 Hz_40ms_modulation_mae_db": 0.5831800355641082, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6333490455694895, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25351446436538216, "3000–6000 Hz_10ms_envelope_correlation": 0.9435228656192438, "3000–6000 Hz_10ms_modulation_mae_db": 0.6522819041840207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0117024088086266, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18750118094517843, "3000–6000 Hz_40ms_envelope_correlation": 0.9661011921335911, "3000–6000 Hz_40ms_modulation_mae_db": 0.45029567195888553, "mr_stft_db": 5.456068281507124, "band_mae_db": 0.17048832011987802, "rms_gain_error_db": -0.14739377239689316, "output_peak": 0.3218205369116824}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5517799931798515, "全帯域_10ms_rms_envelope_nrmse": 0.0569618224169214, "全帯域_10ms_envelope_correlation": 0.9964227101704919, "全帯域_10ms_modulation_mae_db": 0.42970019133090975, "全帯域_40ms_log_rms_mae_db": 0.30182917098520684, "全帯域_40ms_rms_envelope_nrmse": 0.03129812752944175, "全帯域_40ms_envelope_correlation": 0.9989049501067009, "全帯域_40ms_modulation_mae_db": 0.6576712992864687, "300–1000 Hz_10ms_log_rms_mae_db": 0.9743016691713721, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07524262456284224, "300–1000 Hz_10ms_envelope_correlation": 0.9948004723935057, "300–1000 Hz_10ms_modulation_mae_db": 0.2159259245441303, "300–1000 Hz_40ms_log_rms_mae_db": 0.5457800805992009, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04456861987174663, "300–1000 Hz_40ms_envelope_correlation": 0.9980831895138649, "300–1000 Hz_40ms_modulation_mae_db": 0.43120095385702467, "1000–3000 Hz_10ms_log_rms_mae_db": 1.123785928801932, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09830589304617011, "1000–3000 Hz_10ms_envelope_correlation": 0.9907473865768142, "1000–3000 Hz_10ms_modulation_mae_db": 0.503700719890824, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6413069379636633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05216330019464309, "1000–3000 Hz_40ms_envelope_correlation": 0.9975349569295309, "1000–3000 Hz_40ms_modulation_mae_db": 0.5481604136531563, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5847475377245759, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25541592663148366, "3000–6000 Hz_10ms_envelope_correlation": 0.9442395110741791, "3000–6000 Hz_10ms_modulation_mae_db": 1.209190316424088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0065362393952524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18052367354125887, "3000–6000 Hz_40ms_envelope_correlation": 0.9696188729144803, "3000–6000 Hz_40ms_modulation_mae_db": 0.8826307187218504, "mr_stft_db": 5.452019467015606, "band_mae_db": 0.19411438892236302, "rms_gain_error_db": -0.08088209199512203, "output_peak": 0.31887853029717417}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6197190080775862, "全帯域_10ms_rms_envelope_nrmse": 0.06521208075516392, "全帯域_10ms_envelope_correlation": 0.9952904224070452, "全帯域_10ms_modulation_mae_db": 0.2222221895887765, "全帯域_40ms_log_rms_mae_db": 0.3530482109025647, "全帯域_40ms_rms_envelope_nrmse": 0.038076223538763374, "全帯域_40ms_envelope_correlation": 0.998340810070544, "全帯域_40ms_modulation_mae_db": 0.36765701209272755, "300–1000 Hz_10ms_log_rms_mae_db": 1.0905807874515547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07910901081791674, "300–1000 Hz_10ms_envelope_correlation": 0.9942420395614117, "300–1000 Hz_10ms_modulation_mae_db": 0.12789762092199103, "300–1000 Hz_40ms_log_rms_mae_db": 0.6165398823359538, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04565155018712933, "300–1000 Hz_40ms_envelope_correlation": 0.9979836100310361, "300–1000 Hz_40ms_modulation_mae_db": 0.13644312168951567, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1392391588357882, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11284899907480574, "1000–3000 Hz_10ms_envelope_correlation": 0.9877234440448641, "1000–3000 Hz_10ms_modulation_mae_db": 0.5583011354570783, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6415102607694039, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06537101378727533, "1000–3000 Hz_40ms_envelope_correlation": 0.9959118350126541, "1000–3000 Hz_40ms_modulation_mae_db": 0.48337615315933724, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6515193639489598, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2739642173050508, "3000–6000 Hz_10ms_envelope_correlation": 0.9350927393845965, "3000–6000 Hz_10ms_modulation_mae_db": 0.7492021669945027, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0543480741018356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19732362835542142, "3000–6000 Hz_40ms_envelope_correlation": 0.9630748860448206, "3000–6000 Hz_40ms_modulation_mae_db": 0.45000572901816316, "mr_stft_db": 5.4538966825973825, "band_mae_db": 0.1568675489895436, "rms_gain_error_db": -0.0862412414044984, "output_peak": 0.3099781854039948}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6026881529010302, "全帯域_10ms_rms_envelope_nrmse": 0.06041858559116413, "全帯域_10ms_envelope_correlation": 0.9959513876384092, "全帯域_10ms_modulation_mae_db": 0.29854092218116995, "全帯域_40ms_log_rms_mae_db": 0.32705053861237987, "全帯域_40ms_rms_envelope_nrmse": 0.032419954983258725, "全帯域_40ms_envelope_correlation": 0.998801859917281, "全帯域_40ms_modulation_mae_db": 0.3775667663221707, "300–1000 Hz_10ms_log_rms_mae_db": 1.0486866748960708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07508189373919891, "300–1000 Hz_10ms_envelope_correlation": 0.9948311095895805, "300–1000 Hz_10ms_modulation_mae_db": 0.5665852026565799, "300–1000 Hz_40ms_log_rms_mae_db": 0.5785561308617179, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042152609351562884, "300–1000 Hz_40ms_envelope_correlation": 0.9983076273578668, "300–1000 Hz_40ms_modulation_mae_db": 0.698658702633625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1498891537091525, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1052723579816425, "1000–3000 Hz_10ms_envelope_correlation": 0.9891586573260068, "1000–3000 Hz_10ms_modulation_mae_db": 0.15207154026092976, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6739230538356082, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05540145200808101, "1000–3000 Hz_40ms_envelope_correlation": 0.9969078560153573, "1000–3000 Hz_40ms_modulation_mae_db": 0.30401288075611826, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6299411737570182, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2545125505537683, "3000–6000 Hz_10ms_envelope_correlation": 0.9450920515761871, "3000–6000 Hz_10ms_modulation_mae_db": 1.110288144342751, "3000–6000 Hz_40ms_log_rms_mae_db": 1.002110686564979, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18326166802918892, "3000–6000 Hz_40ms_envelope_correlation": 0.9690310720008616, "3000–6000 Hz_40ms_modulation_mae_db": 0.9827356697062135, "mr_stft_db": 5.4468990281624245, "band_mae_db": 0.1025232460236097, "rms_gain_error_db": -0.07390364429430181, "output_peak": 0.3281639589362129}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.159680862457142, "spectral_convergence": 0.20982179715002017, "spectral_convergence_status": "ok", "amplitude_floor": 1.0845908458793198e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.268553065708653, "spectral_convergence": 0.20331108643831586, "spectral_convergence_status": "ok", "amplitude_floor": 9.938577161724507e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.000702231292503, "spectral_convergence": 0.21431483721119718, "spectral_convergence_status": "ok", "amplitude_floor": 7.780261052298239e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.1610975793667e-05, "output_power": 8.320182351684084e-05, "delta_db": 0.0838427414272101, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005078113602376632, "output_power": 0.0005089737891457308, "delta_db": 0.009930056130829404, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000931370997501075, "output_power": 0.0009136366820455895, "delta_db": -0.08349182107606451, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000354698831360673, "output_power": 0.00033923407358186597, "delta_db": -0.1936029010797946, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.667116250090391e-05, "output_power": 9.004835143626823e-05, "delta_db": -0.308211747859533, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.4087248189368409e-05, "output_power": 1.555705658265559e-05, "delta_db": 0.43101265212457074, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18501531994966705, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07854288520422589, "pitch": null, "output_rms_dbfs": -27.098202524534386, "output_peak": 0.32395386491502803, "reference_rms_dbfs": -27.01965963933016, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.476312053152767}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001511941088699898, "log_rms_mae_db": 0.6004257103929255, "rms_envelope_nrmse": 0.057450006407651605, "envelope_correlation": 0.9963584003756585, "rms_gain_error_db": -0.0785454030610126, "reference_mean_power": 0.0019872702200913737, "output_mean_power": 0.0019516520146375313}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.92641346613152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000630498462430925, "output_power": 0.0006260176607684332, "reference_active": true, "delta_db": -0.030974464227232344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.948331074534803e-05, "output_power": 7.476873347273711e-05, "reference_active": true, "delta_db": -0.2655592454974277}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3123506088131581e-05, "output_power": 1.300957124549565e-05, "reference_active": true, "delta_db": -0.03786893241490907}], "active_band_count": 3, "active_band_mae_db": 0.11146754737985637}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013307215895407113, "log_rms_mae_db": 0.32675356100071745, "rms_envelope_nrmse": 0.032580191329230235, "envelope_correlation": 0.9988052389834511, "rms_gain_error_db": -0.07914489678121868, "reference_mean_power": 0.0019932283004473585, "output_mean_power": 0.0019572331150052905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.959638607230672e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005794424432114469, "output_power": 0.0005768722673768991, "reference_active": true, "delta_db": -0.019306422559150323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.108494970400102e-05, "output_power": 3.882636773496014e-05, "reference_active": true, "delta_db": -0.2455599610369122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.934091733283841e-07, "output_power": 6.754691500797694e-07, "reference_active": true, "delta_db": -0.698916986604049}], "active_band_count": 3, "active_band_mae_db": 0.32126112340003715}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014956359127228178, "log_rms_mae_db": 1.068843951426529, "rms_envelope_nrmse": 0.07287288766470316, "envelope_correlation": 0.9951130980483839, "rms_gain_error_db": -0.07083944212387282, "reference_mean_power": 0.001265981682968443, "output_mean_power": 0.0012454992715381806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.65319874591582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005127917421071942, "output_power": 0.000506163568972254, "reference_active": true, "delta_db": -0.056501390946367275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.49871434628785e-05, "output_power": 6.939054536339665e-05, "reference_active": true, "delta_db": -0.33686509266450626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1228212970044594e-05, "output_power": 9.683145472192173e-06, "reference_active": true, "delta_db": -0.642941850678181}], "active_band_count": 3, "active_band_mae_db": 0.3454361114296849}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012969159209010741, "log_rms_mae_db": 0.5991329236489847, "rms_envelope_nrmse": 0.03938135864291475, "envelope_correlation": 0.9985034995472414, "rms_gain_error_db": -0.07091093607085582, "reference_mean_power": 0.0012704534730935044, "output_mean_power": 0.0012498781364642453}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.90499086389905e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004798268169104879, "output_power": 0.00047266892723985264, "reference_active": true, "delta_db": -0.0652746329931638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7762549535694356e-05, "output_power": 3.523751381077107e-05, "reference_active": true, "delta_db": -0.3005604888084385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.708330714477498e-07, "output_power": 5.680079577624259e-07, "reference_active": true, "delta_db": -1.3260591894556168}], "active_band_count": 3, "active_band_mae_db": 0.563964770419073}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.983632889928613e-05, "log_rms_mae_db": 1.0901586049135499, "rms_envelope_nrmse": 0.102735899577425, "envelope_correlation": 0.9900141167882631, "rms_gain_error_db": -0.2153071388644729, "reference_mean_power": 0.00040681650183919396, "output_mean_power": 0.000387139822996116}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8313223654566366e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016437493457690703, "output_power": 0.00015980877489045836, "reference_active": true, "delta_db": -0.12234970806020451}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7703924674415462e-05, "output_power": 2.7283358056472146e-05, "reference_active": true, "delta_db": -0.06643475148474103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.687130809571553e-06, "output_power": 4.734313432824502e-06, "reference_active": true, "delta_db": 0.04349932707246887}], "active_band_count": 3, "active_band_mae_db": 0.07742792887247148}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.811887222434066e-05, "log_rms_mae_db": 0.6046748790603182, "rms_envelope_nrmse": 0.05734386965948882, "envelope_correlation": 0.9971272294834065, "rms_gain_error_db": -0.2183417081521487, "reference_mean_power": 0.00040698573595692556, "output_mean_power": 0.0003870303453847225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.593242154160197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015388869283670126, "output_power": 0.0001509419293133722, "reference_active": true, "delta_db": -0.0839681390893304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4161267354242513e-05, "output_power": 1.3836788862255708e-05, "reference_active": true, "delta_db": -0.10066807999130756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7920568451213675e-07, "output_power": 2.7427582824676007e-07, "reference_active": true, "delta_db": -0.07736720900802922}], "active_band_count": 3, "active_band_mae_db": 0.08733447602955573}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4474245023095566e-05, "log_rms_mae_db": 1.6233510798263344, "rms_envelope_nrmse": 0.2605875020165354, "envelope_correlation": 0.9396859978614147, "rms_gain_error_db": -0.05360238945001522, "reference_mean_power": 1.6116938837494647e-05, "output_mean_power": 1.5919239582186177e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.1253846819517355e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.226560638885928e-06, "output_power": 2.583368538779889e-06, "reference_active": true, "delta_db": 0.6455183892722223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.434590024496585e-06, "output_power": 1.2768778853484443e-06, "reference_active": true, "delta_db": -0.5057844126989463}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.807607987621042e-07, "output_power": 5.446883082549154e-07, "reference_active": true, "delta_db": 0.5421900516953971}], "active_band_count": 3, "active_band_mae_db": 0.5644976178888551}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.920336454108668e-05, "log_rms_mae_db": 1.0372125708233935, "rms_envelope_nrmse": 0.19203025931244108, "envelope_correlation": 0.9641279833793891, "rms_gain_error_db": -0.05358694109101347, "reference_mean_power": 1.619697385840346e-05, "output_mean_power": 1.5998349757300377e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.254945521069739e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4889493401597257e-06, "output_power": 2.886466117371619e-06, "reference_active": true, "delta_db": 0.6435040681238438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.64830026091652e-07, "output_power": 6.877542831425637e-07, "reference_active": true, "delta_db": -0.46131625624003947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8106205193126903e-08, "output_power": 4.5320296667454696e-08, "reference_active": true, "delta_db": -0.25908355295120233}], "active_band_count": 3, "active_band_mae_db": 0.45463462577169517}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8932322417842404, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2dd9dc8cc9d0d66af6f044daaf58df03078b1b694e598b95b31e368410c5f1b2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897915325513745, "standard_deviation": 0.0018598356731743255, "minimum": 0.9866725069771232, "maximum": 0.9923491217699312, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1103739204208534, "standard_deviation": 0.024273665263550983, "minimum": 1.0868725386773623, "maximum": 1.1498891537091525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.40946060958330277, "standard_deviation": 0.19235172815173088, "minimum": 0.07742792887247148, "maximum": 0.6225947019206113, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1030252447587375, "standard_deviation": 0.00864913032977427, "minimum": 0.09119686314236804, "maximum": 0.11744359938050701, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969176860680936, "standard_deviation": 0.0006195381178636673, "minimum": 0.9959118350126541, "maximum": 0.9975867481887668, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6397623369614267, "standard_deviation": 0.01993041072790207, "minimum": 0.6046748790603182, "maximum": 0.6739230538356082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4002753323190885, "standard_deviation": 0.16056444242816767, "minimum": 0.08733447602955573, "maximum": 0.5831800355641082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05784635094650677, "standard_deviation": 0.004845383986768759, "minimum": 0.05216330019464309, "maximum": 0.06537101378727533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.94163690567372, "standard_deviation": 0.005257566060335678, "minimum": 0.9328901448132957, "maximum": 0.9478448199290559, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6246907847734584, "standard_deviation": 0.02667296995163106, "minimum": 1.5847475377245759, "maximum": 1.6563801619575516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0015870386384305, "standard_deviation": 0.4825727013854785, "minimum": 0.46334699496790926, "maximum": 1.8707508440737968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2602377510788759, "standard_deviation": 0.009397242448563827, "minimum": 0.2513490572467764, "maximum": 0.27567017015079726, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9663895525178657, "standard_deviation": 0.002822676382983822, "minimum": 0.9627069916647036, "maximum": 0.9696188729144803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0271147606189148, "standard_deviation": 0.01969338271698213, "minimum": 1.002110686564979, "maximum": 1.0543480741018356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8672201823159873, "standard_deviation": 0.506264643911759, "minimum": 0.45000572901816316, "maximum": 1.7396449107543204, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18903974374778998, "standard_deviation": 0.005953069303002355, "minimum": 0.18052367354125887, "maximum": 0.19732362835542142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9947344652451622, "standard_deviation": 0.0003818360245149575, "minimum": 0.994142929056598, "maximum": 0.9951773530018769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0418360591872702, "standard_deviation": 0.0356659499918473, "minimum": 0.9743016691713721, "maximum": 1.0905807874515547, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38382913477865277, "standard_deviation": 0.19252296810962988, "minimum": 0.12789762092199103, "maximum": 0.7359621328134193, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07592883924796329, "standard_deviation": 0.0029712835415637287, "minimum": 0.07265706736760622, "maximum": 0.08123484716699399, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982090136408388, "standard_deviation": 0.00027260454377232206, "minimum": 0.9978425012344884, "maximum": 0.9985978829026462, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5933315437451355, "standard_deviation": 0.02743787644899474, "minimum": 0.5457800805992009, "maximum": 0.6224053125773685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5050385809678385, "standard_deviation": 0.19631929793846206, "minimum": 0.13644312168951567, "maximum": 0.7883858689334798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04358310224072094, "standard_deviation": 0.0036632600478891985, "minimum": 0.03881016607799224, "maximum": 0.05001699517410892, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1567611789820216, "standard_deviation": 0.034370905631093886, "minimum": 0.1025232460236097, "maximum": 0.19411438892236302, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.456595850334299, "standard_deviation": 0.008961354824428796, "minimum": 5.4468990281624245, "maximum": 5.476312053152767, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31861028923393364, "standard_deviation": 0.006932350309927865, "minimum": 0.30988486124186193, "maximum": 0.3281639589362129, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09505928472659783, "standard_deviation": 0.025174694595889564, "minimum": -0.14739377239689316, "maximum": -0.07390364429430181, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9961675789700447, "standard_deviation": 0.00044636791539476154, "minimum": 0.9952904224070452, "maximum": 0.9966761408041568, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5918570794664804, "standard_deviation": 0.020399461302078654, "minimum": 0.5517799931798515, "maximum": 0.6197190080775862, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3211588153292888, "standard_deviation": 0.20711005065306845, "minimum": 0.11146754737985637, "maximum": 0.7657619238804102, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05914444251280519, "standard_deviation": 0.003005318772527712, "minimum": 0.056591150133782006, "maximum": 0.06521208075516392, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987810357484742, "standard_deviation": 0.00018449077763994036, "minimum": 0.998340810070544, "maximum": 0.9989049501067009, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3288043089712407, "standard_deviation": 0.016369782955296376, "minimum": 0.30182917098520684, "maximum": 0.3530482109025647, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44416624505581037, "standard_deviation": 0.1755576320439821, "minimum": 0.28617453088567896, "maximum": 0.769902564971173, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03340169243248096, "standard_deviation": 0.002073632134408135, "minimum": 0.03129812752944175, "maximum": 0.038076223538763374, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0007", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.728482723236084, "predicted_mos": 2.728482723236084, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5b72cae023b6c7e12c4afa53c807458c7c17c3c0b05dd7244a30467074df99a6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8447066000662744, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6069015540201149, "全帯域_10ms_rms_envelope_nrmse": 0.05822931656088282, "全帯域_10ms_envelope_correlation": 0.9962523336184302, "全帯域_10ms_modulation_mae_db": 0.12739459066914122, "全帯域_40ms_log_rms_mae_db": 0.3336161045642072, "全帯域_40ms_rms_envelope_nrmse": 0.03333489040158514, "全帯域_40ms_envelope_correlation": 0.9987382949846351, "全帯域_40ms_modulation_mae_db": 0.22629979809482606, "300–1000 Hz_10ms_log_rms_mae_db": 1.073711635094464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0742692549118412, "300–1000 Hz_10ms_envelope_correlation": 0.9949159661989996, "300–1000 Hz_10ms_modulation_mae_db": 0.2956570713959504, "300–1000 Hz_40ms_log_rms_mae_db": 0.5905770595420468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04063660191954761, "300–1000 Hz_40ms_envelope_correlation": 0.9983925915505723, "300–1000 Hz_40ms_modulation_mae_db": 0.4626047984834791, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0748998256276006, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10518751267280489, "1000–3000 Hz_10ms_envelope_correlation": 0.9895283207094847, "1000–3000 Hz_10ms_modulation_mae_db": 0.09570581418806795, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6165078103282027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05914079635455778, "1000–3000 Hz_40ms_envelope_correlation": 0.996924655895771, "1000–3000 Hz_40ms_modulation_mae_db": 0.08601763752055624, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5936038056508401, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2587625898732065, "3000–6000 Hz_10ms_envelope_correlation": 0.940406808200491, "3000–6000 Hz_10ms_modulation_mae_db": 0.5032505483824069, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0665858322872326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18973305337433843, "3000–6000 Hz_40ms_envelope_correlation": 0.9649525987210793, "3000–6000 Hz_40ms_modulation_mae_db": 0.37776383845036143, "mr_stft_db": 5.435091985454059, "band_mae_db": 0.1800735674676187, "rms_gain_error_db": -0.0769055812226758, "output_peak": 0.32429128012812647}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6096497834803487, "全帯域_10ms_rms_envelope_nrmse": 0.062189486507367046, "全帯域_10ms_envelope_correlation": 0.9957287109584565, "全帯域_10ms_modulation_mae_db": 0.15834601995812378, "全帯域_40ms_log_rms_mae_db": 0.32155933383443747, "全帯域_40ms_rms_envelope_nrmse": 0.03241641905400119, "全帯域_40ms_envelope_correlation": 0.9988326946273395, "全帯域_40ms_modulation_mae_db": 0.32171160405639687, "300–1000 Hz_10ms_log_rms_mae_db": 1.0651650660003686, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07340470769080026, "300–1000 Hz_10ms_envelope_correlation": 0.995052626179284, "300–1000 Hz_10ms_modulation_mae_db": 0.2600074328713051, "300–1000 Hz_40ms_log_rms_mae_db": 0.5717560255055073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03901770296089131, "300–1000 Hz_40ms_envelope_correlation": 0.998549685761817, "300–1000 Hz_40ms_modulation_mae_db": 0.40288811798868235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0587764907450188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11688485434724478, "1000–3000 Hz_10ms_envelope_correlation": 0.9869114258265993, "1000–3000 Hz_10ms_modulation_mae_db": 0.4738186583498405, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6248009860765812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06451010335811871, "1000–3000 Hz_40ms_envelope_correlation": 0.9962443781831964, "1000–3000 Hz_40ms_modulation_mae_db": 0.3699782303920999, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6203598904090017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26905648966420687, "3000–6000 Hz_10ms_envelope_correlation": 0.935844134017148, "3000–6000 Hz_10ms_modulation_mae_db": 0.46727352730661703, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0644702768281196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19038661055539022, "3000–6000 Hz_40ms_envelope_correlation": 0.9647993361947571, "3000–6000 Hz_40ms_modulation_mae_db": 0.5380329191022956, "mr_stft_db": 5.418540347717545, "band_mae_db": 0.15529591734301576, "rms_gain_error_db": -0.08722527886019632, "output_peak": 0.31086627020093466}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6049687398284017, "全帯域_10ms_rms_envelope_nrmse": 0.0564594300451528, "全帯域_10ms_envelope_correlation": 0.9965031223138487, "全帯域_10ms_modulation_mae_db": 0.36625011364692694, "全帯域_40ms_log_rms_mae_db": 0.33227760202385975, "全帯域_40ms_rms_envelope_nrmse": 0.032584482144740766, "全帯域_40ms_envelope_correlation": 0.9988212193907782, "全帯域_40ms_modulation_mae_db": 0.4844096966869094, "300–1000 Hz_10ms_log_rms_mae_db": 1.0075645039361962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07713841751813208, "300–1000 Hz_10ms_envelope_correlation": 0.9945231479349981, "300–1000 Hz_10ms_modulation_mae_db": 0.39348809078394514, "300–1000 Hz_40ms_log_rms_mae_db": 0.6030237321758392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04574426637761558, "300–1000 Hz_40ms_envelope_correlation": 0.9979667145851217, "300–1000 Hz_40ms_modulation_mae_db": 0.5782680224131156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.06735783835947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09469199089165402, "1000–3000 Hz_10ms_envelope_correlation": 0.9915124299753717, "1000–3000 Hz_10ms_modulation_mae_db": 0.45573493862217745, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6185747011717612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05344085399609724, "1000–3000 Hz_40ms_envelope_correlation": 0.9974561535686746, "1000–3000 Hz_40ms_modulation_mae_db": 0.4483597195209339, "3000–6000 Hz_10ms_log_rms_mae_db": 1.552148924629526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24306363570685383, "3000–6000 Hz_10ms_envelope_correlation": 0.9497816652942613, "3000–6000 Hz_10ms_modulation_mae_db": 1.621291423482406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0218059851546297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1770031485848473, "3000–6000 Hz_40ms_envelope_correlation": 0.9709054839110846, "3000–6000 Hz_40ms_modulation_mae_db": 1.3963081394337558, "mr_stft_db": 5.418118300198478, "band_mae_db": 0.11332738628598991, "rms_gain_error_db": -0.08760477189140056, "output_peak": 0.31226319227639904}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5911596562108251, "全帯域_10ms_rms_envelope_nrmse": 0.05775338803333161, "全帯域_10ms_envelope_correlation": 0.9964127718761722, "全帯域_10ms_modulation_mae_db": 0.2803309822923164, "全帯域_40ms_log_rms_mae_db": 0.33221018825772, "全帯域_40ms_rms_envelope_nrmse": 0.03205793731349016, "全帯域_40ms_envelope_correlation": 0.9989587930686105, "全帯域_40ms_modulation_mae_db": 0.36495892178053585, "300–1000 Hz_10ms_log_rms_mae_db": 1.060402972496103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07533732921200793, "300–1000 Hz_10ms_envelope_correlation": 0.9948457185890964, "300–1000 Hz_10ms_modulation_mae_db": 0.4337307640906249, "300–1000 Hz_40ms_log_rms_mae_db": 0.6145413773124121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042793846457241186, "300–1000 Hz_40ms_envelope_correlation": 0.9983165252072064, "300–1000 Hz_40ms_modulation_mae_db": 0.47305205348307683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.064431902035904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09951321946974309, "1000–3000 Hz_10ms_envelope_correlation": 0.9906660651216078, "1000–3000 Hz_10ms_modulation_mae_db": 0.40108961569532764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6396975221078097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058503554595887394, "1000–3000 Hz_40ms_envelope_correlation": 0.996999716861165, "1000–3000 Hz_40ms_modulation_mae_db": 0.4429736369552472, "3000–6000 Hz_10ms_log_rms_mae_db": 1.587422203483739, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25186804250686484, "3000–6000 Hz_10ms_envelope_correlation": 0.9468295628221232, "3000–6000 Hz_10ms_modulation_mae_db": 1.3988762400328258, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0466790750281731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18543345023041719, "3000–6000 Hz_40ms_envelope_correlation": 0.9686003456518322, "3000–6000 Hz_40ms_modulation_mae_db": 1.458057779166471, "mr_stft_db": 5.403370442402918, "band_mae_db": 0.17154854515597098, "rms_gain_error_db": -0.1171548243822488, "output_peak": 0.32509695363530144}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6030205158208347, "全帯域_10ms_rms_envelope_nrmse": 0.05911339812307963, "全帯域_10ms_envelope_correlation": 0.9963314152871195, "全帯域_10ms_modulation_mae_db": 0.7361264901165585, "全帯域_40ms_log_rms_mae_db": 0.3554592069772115, "全帯域_40ms_rms_envelope_nrmse": 0.03526436798424868, "全帯域_40ms_envelope_correlation": 0.998802842554102, "全帯域_40ms_modulation_mae_db": 0.7308721380435035, "300–1000 Hz_10ms_log_rms_mae_db": 1.0604573826992614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08297635780089826, "300–1000 Hz_10ms_envelope_correlation": 0.9938539626878994, "300–1000 Hz_10ms_modulation_mae_db": 0.7434141526082184, "300–1000 Hz_40ms_log_rms_mae_db": 0.6285129069143304, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05030535000969354, "300–1000 Hz_40ms_envelope_correlation": 0.9977901753779903, "300–1000 Hz_40ms_modulation_mae_db": 0.8081926524205089, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0838977743020797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09073160536684431, "1000–3000 Hz_10ms_envelope_correlation": 0.9924028699366272, "1000–3000 Hz_10ms_modulation_mae_db": 0.5913353799690279, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6367086629677985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05356295516902102, "1000–3000 Hz_40ms_envelope_correlation": 0.9976474843128462, "1000–3000 Hz_40ms_modulation_mae_db": 0.5652385993241368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6039937197167078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2509440408731859, "3000–6000 Hz_10ms_envelope_correlation": 0.9447392919163528, "3000–6000 Hz_10ms_modulation_mae_db": 0.6619588708128145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0092652623484537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18411626360423455, "3000–6000 Hz_40ms_envelope_correlation": 0.9673338942001101, "3000–6000 Hz_40ms_modulation_mae_db": 0.4921511051178888, "mr_stft_db": 5.4099811127619155, "band_mae_db": 0.16818056821227587, "rms_gain_error_db": -0.14535537523928052, "output_peak": 0.32396537739501635}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5584624081402753, "全帯域_10ms_rms_envelope_nrmse": 0.058966633440325475, "全帯域_10ms_envelope_correlation": 0.9961562275970665, "全帯域_10ms_modulation_mae_db": 0.42366435595418067, "全帯域_40ms_log_rms_mae_db": 0.31042741977142935, "全帯域_40ms_rms_envelope_nrmse": 0.032898871293125576, "全帯域_40ms_envelope_correlation": 0.9987755205906096, "全帯域_40ms_modulation_mae_db": 0.662201321122884, "300–1000 Hz_10ms_log_rms_mae_db": 0.9808374543878564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07696763923117361, "300–1000 Hz_10ms_envelope_correlation": 0.9945508363325012, "300–1000 Hz_10ms_modulation_mae_db": 0.22992324765984465, "300–1000 Hz_40ms_log_rms_mae_db": 0.5371318834502784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04495064717506258, "300–1000 Hz_40ms_envelope_correlation": 0.9980422307621235, "300–1000 Hz_40ms_modulation_mae_db": 0.45899014094410573, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1250544065504742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10027637134078957, "1000–3000 Hz_10ms_envelope_correlation": 0.9903696087372197, "1000–3000 Hz_10ms_modulation_mae_db": 0.5108610652027494, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6527116992907115, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054695813416601506, "1000–3000 Hz_40ms_envelope_correlation": 0.997245895367917, "1000–3000 Hz_40ms_modulation_mae_db": 0.56396017799383, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5477928941702934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24810978109778073, "3000–6000 Hz_10ms_envelope_correlation": 0.9472236254749731, "3000–6000 Hz_10ms_modulation_mae_db": 1.1598370221863021, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0055746985261895, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17407278404138382, "3000–6000 Hz_40ms_envelope_correlation": 0.9715771272810879, "3000–6000 Hz_40ms_modulation_mae_db": 0.8214061431355675, "mr_stft_db": 5.405343626318987, "band_mae_db": 0.19520648039869412, "rms_gain_error_db": -0.07936256450195646, "output_peak": 0.31834642658102524}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6181858683304394, "全帯域_10ms_rms_envelope_nrmse": 0.06501139239990421, "全帯域_10ms_envelope_correlation": 0.9953308878614385, "全帯域_10ms_modulation_mae_db": 0.30477555221956554, "全帯域_40ms_log_rms_mae_db": 0.350424753689529, "全帯域_40ms_rms_envelope_nrmse": 0.0375223926655088, "全帯域_40ms_envelope_correlation": 0.9984060526468869, "全帯域_40ms_modulation_mae_db": 0.4634938569110652, "300–1000 Hz_10ms_log_rms_mae_db": 1.08432639165273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07739408486506515, "300–1000 Hz_10ms_envelope_correlation": 0.9944968630519483, "300–1000 Hz_10ms_modulation_mae_db": 0.2739690819880026, "300–1000 Hz_40ms_log_rms_mae_db": 0.6170437252305272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04438700865606631, "300–1000 Hz_40ms_envelope_correlation": 0.9981046738375567, "300–1000 Hz_40ms_modulation_mae_db": 0.3182635886259521, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1323880811893556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11572314095765011, "1000–3000 Hz_10ms_envelope_correlation": 0.987126970719717, "1000–3000 Hz_10ms_modulation_mae_db": 0.5475284540119921, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6465191080455545, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06806129591762435, "1000–3000 Hz_40ms_envelope_correlation": 0.9955720366628723, "1000–3000 Hz_40ms_modulation_mae_db": 0.4484473731443992, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6380917279571459, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25958523985701754, "3000–6000 Hz_10ms_envelope_correlation": 0.9407037725363434, "3000–6000 Hz_10ms_modulation_mae_db": 0.5817334698478248, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0773308313201517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1847014781736325, "3000–6000 Hz_40ms_envelope_correlation": 0.9670759959959798, "3000–6000 Hz_40ms_modulation_mae_db": 0.5372488946553703, "mr_stft_db": 5.414902678374964, "band_mae_db": 0.16623105693160356, "rms_gain_error_db": -0.09142343039189087, "output_peak": 0.3109500700126255}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6000794865402099, "全帯域_10ms_rms_envelope_nrmse": 0.059856677328645795, "全帯域_10ms_envelope_correlation": 0.9960219106495249, "全帯域_10ms_modulation_mae_db": 0.20601409389801914, "全帯域_40ms_log_rms_mae_db": 0.3308921074839333, "全帯域_40ms_rms_envelope_nrmse": 0.03228787553566491, "全帯域_40ms_envelope_correlation": 0.9988045168737889, "全帯域_40ms_modulation_mae_db": 0.22966269496294353, "300–1000 Hz_10ms_log_rms_mae_db": 1.047381384455016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07531323987586948, "300–1000 Hz_10ms_envelope_correlation": 0.9947945820608595, "300–1000 Hz_10ms_modulation_mae_db": 0.5130869875904301, "300–1000 Hz_40ms_log_rms_mae_db": 0.5793576787920734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041919687279735135, "300–1000 Hz_40ms_envelope_correlation": 0.9983207643799048, "300–1000 Hz_40ms_modulation_mae_db": 0.5830776917587269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1186219047324764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10643710276837591, "1000–3000 Hz_10ms_envelope_correlation": 0.9889453999621965, "1000–3000 Hz_10ms_modulation_mae_db": 0.1143094563115085, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6525551502350838, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058875719424308866, "1000–3000 Hz_40ms_envelope_correlation": 0.9964898238679393, "1000–3000 Hz_40ms_modulation_mae_db": 0.38338662631412584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5979475949297333, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2452459131192848, "3000–6000 Hz_10ms_envelope_correlation": 0.9484225297944808, "3000–6000 Hz_10ms_modulation_mae_db": 1.115318188438612, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0155949366384085, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17333397506897805, "3000–6000 Hz_40ms_envelope_correlation": 0.9718064365015473, "3000–6000 Hz_40ms_modulation_mae_db": 0.9795572756130314, "mr_stft_db": 5.408743599537099, "band_mae_db": 0.11163131358081844, "rms_gain_error_db": -0.07016656040773915, "output_peak": 0.32875177378942627}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.120257096926774, "spectral_convergence": 0.20827443681542002, "spectral_convergence_status": "ok", "amplitude_floor": 1.0845908458793198e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.229900885396703, "spectral_convergence": 0.20411673047463974, "spectral_convergence_status": "ok", "amplitude_floor": 9.938577161724507e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.955117974038699, "spectral_convergence": 0.2142702836295256, "spectral_convergence_status": "ok", "amplitude_floor": 7.780261052298239e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.1610975793667e-05, "output_power": 8.253550594478805e-05, "delta_db": 0.0489224727388157, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005078113602376632, "output_power": 0.0005106161278610102, "delta_db": 0.023921156796794026, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000931370997501075, "output_power": 0.0009137303805504708, "delta_db": -0.0830464508546013, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000354698831360673, "output_power": 0.00033853241968682593, "delta_db": -0.2025949219810674, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.667116250090391e-05, "output_power": 9.04145459424355e-05, "delta_db": -0.2905863551692093, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.4087248189368409e-05, "output_power": 1.5558336876428472e-05, "delta_db": 0.4313700472652246, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1800735674676187, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0769055812226758, "pitch": null, "output_rms_dbfs": -27.096565220552836, "output_peak": 0.32429128012812647, "reference_rms_dbfs": -27.01965963933016, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.435091985454059}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001511941088699898, "log_rms_mae_db": 0.6069015540201149, "rms_envelope_nrmse": 0.05822931656088282, "envelope_correlation": 0.9962523336184302, "rms_gain_error_db": -0.07692739713567794, "reference_mean_power": 0.0019872702200913737, "output_mean_power": 0.0019523792568568225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.92641346613152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000630498462430925, "output_power": 0.0006254328556683186, "reference_active": true, "delta_db": -0.035033396532105715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.948331074534803e-05, "output_power": 7.510748262123754e-05, "reference_active": true, "delta_db": -0.24592742661486003}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3123506088131581e-05, "output_power": 1.3432973931589112e-05, "reference_active": true, "delta_db": 0.10122294886045789}], "active_band_count": 3, "active_band_mae_db": 0.12739459066914122}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013307215895407113, "log_rms_mae_db": 0.3336161045642072, "rms_envelope_nrmse": 0.03333489040158514, "envelope_correlation": 0.9987382949846351, "rms_gain_error_db": -0.07744620275888892, "reference_mean_power": 0.0019932283004473585, "output_mean_power": 0.0019579988144634883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.959638607230672e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005794424432114469, "output_power": 0.0005757826502865372, "reference_active": true, "delta_db": -0.02751728989571308}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.108494970400102e-05, "output_power": 3.913618887890323e-05, "reference_active": true, "delta_db": -0.21104228297365604}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.934091733283841e-07, "output_power": 7.16907684017495e-07, "reference_active": true, "delta_db": -0.4403398214151091}], "active_band_count": 3, "active_band_mae_db": 0.22629979809482606}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014956359127228178, "log_rms_mae_db": 1.073711635094464, "rms_envelope_nrmse": 0.0742692549118412, "envelope_correlation": 0.9949159661989996, "rms_gain_error_db": -0.0656476219496249, "reference_mean_power": 0.001265981682968443, "output_mean_power": 0.00124698910740393}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.65319874591582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005127917421071942, "output_power": 0.0005078385166810478, "reference_active": true, "delta_db": -0.0421538620943265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.49871434628785e-05, "output_power": 6.92596191049967e-05, "reference_active": true, "delta_db": -0.34506711228388065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1228212970044594e-05, "output_power": 1.0007730865257237e-05, "reference_active": true, "delta_db": -0.49975023980964395}], "active_band_count": 3, "active_band_mae_db": 0.2956570713959504}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012969159209010741, "log_rms_mae_db": 0.5905770595420468, "rms_envelope_nrmse": 0.04063660191954761, "envelope_correlation": 0.9983925915505723, "rms_gain_error_db": -0.06549385767849064, "reference_mean_power": 0.0012704534730935044, "output_mean_power": 0.0012514381176567905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.90499086389905e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004798268169104879, "output_power": 0.00047402480728933073, "reference_active": true, "delta_db": -0.05283446137766219}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7762549535694356e-05, "output_power": 3.52487496739299e-05, "reference_active": true, "delta_db": -0.29917591451853043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.708330714477498e-07, "output_power": 6.072673576536112e-07, "reference_active": true, "delta_db": -1.0358040195542446}], "active_band_count": 3, "active_band_mae_db": 0.4626047984834791}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.983632889928613e-05, "log_rms_mae_db": 1.0748998256276006, "rms_envelope_nrmse": 0.10518751267280489, "envelope_correlation": 0.9895283207094847, "rms_gain_error_db": -0.21760699365362418, "reference_mean_power": 0.00040681650183919396, "output_mean_power": 0.0003869348630661055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8313223654566366e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016437493457690703, "output_power": 0.00015960447975930624, "reference_active": true, "delta_db": -0.1279051606875442}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7703924674415462e-05, "output_power": 2.7445133272766282e-05, "reference_active": true, "delta_db": -0.04075953673966005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.687130809571553e-06, "output_power": 4.816730508140315e-06, "reference_active": true, "delta_db": 0.11845274513699962}], "active_band_count": 3, "active_band_mae_db": 0.09570581418806795}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.811887222434066e-05, "log_rms_mae_db": 0.6165078103282027, "rms_envelope_nrmse": 0.05914079635455778, "envelope_correlation": 0.996924655895771, "rms_gain_error_db": -0.22088885863773872, "reference_mean_power": 0.00040698573595692556, "output_mean_power": 0.0003868034174509812}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.593242154160197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015388869283670126, "output_power": 0.00015058768216401832, "reference_active": true, "delta_db": -0.09417261965244966}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4161267354242513e-05, "output_power": 1.3859525298137161e-05, "reference_active": true, "delta_db": -0.09353766487368165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7920568451213675e-07, "output_power": 2.7471982002016695e-07, "reference_active": true, "delta_db": -0.07034262803553741}], "active_band_count": 3, "active_band_mae_db": 0.08601763752055624}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4474245023095566e-05, "log_rms_mae_db": 1.5936038056508401, "rms_envelope_nrmse": 0.2587625898732065, "envelope_correlation": 0.940406808200491, "rms_gain_error_db": -0.11848554001636882, "reference_mean_power": 1.6116938837494647e-05, "output_mean_power": 1.568317555110615e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.1253846819517355e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.226560638885928e-06, "output_power": 2.613523190915809e-06, "reference_active": true, "delta_db": 0.6959183078537851}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.434590024496585e-06, "output_power": 1.3456785036986818e-06, "reference_active": true, "delta_db": -0.277864917135343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.807607987621042e-07, "output_power": 5.439085553796089e-07, "reference_active": true, "delta_db": 0.5359684201580927}], "active_band_count": 3, "active_band_mae_db": 0.5032505483824069}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.920336454108668e-05, "log_rms_mae_db": 1.0665858322872326, "rms_envelope_nrmse": 0.18973305337433843, "envelope_correlation": 0.9649525987210793, "rms_gain_error_db": -0.11847971961935828, "reference_mean_power": 1.619697385840346e-05, "output_mean_power": 1.576107767213475e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.254945521069739e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4889493401597257e-06, "output_power": 2.9372242969942202e-06, "reference_active": true, "delta_db": 0.7192105496562545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.64830026091652e-07, "output_power": 7.186575070602849e-07, "reference_active": true, "delta_db": -0.2704296280359587}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8106205193126903e-08, "output_power": 4.6541027341950144e-08, "reference_active": true, "delta_db": -0.14365133765887103}], "active_band_count": 3, "active_band_mae_db": 0.37776383845036143}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8930663763378828, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bad2241c152b49a357ef76d0b9f84db4e1930c358c7b1a5b39341c240e529e9b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.989682886373603, "standard_deviation": 0.0019626965260661597, "minimum": 0.9869114258265993, "maximum": 0.9924028699366272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0906785279427973, "standard_deviation": 0.029875087917818392, "minimum": 1.0587764907450188, "maximum": 1.1323880811893556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3987979227938364, "standard_deviation": 0.19027171979273547, "minimum": 0.09570581418806795, "maximum": 0.5913353799690279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10368072472688833, "standard_deviation": 0.009309531273567907, "minimum": 0.09073160536684431, "maximum": 0.11688485434724478, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968225180900476, "standard_deviation": 0.0006878307558468033, "minimum": 0.9955720366628723, "maximum": 0.9976474843128462, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6360094550279378, "standard_deviation": 0.014575826830917824, "minimum": 0.6165078103282027, "maximum": 0.6527116992907115, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4135452501456661, "standard_deviation": 0.15071909678793113, "minimum": 0.08601763752055624, "maximum": 0.5652385993241368, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05884888652902711, "standard_deviation": 0.005231210604504157, "minimum": 0.05344085399609724, "maximum": 0.06806129591762435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9442439237570217, "standard_deviation": 0.00480698101276088, "minimum": 0.935844134017148, "maximum": 0.9497816652942613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5926700951183732, "standard_deviation": 0.030869001038878473, "minimum": 1.5477928941702934, "maximum": 1.6380917279571459, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9386924113112263, "standard_deviation": 0.44298640582389576, "minimum": 0.46727352730661703, "maximum": 1.621291423482406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2533294665873001, "standard_deviation": 0.008633779472262642, "minimum": 0.24306363570685383, "maximum": 0.26905648966420687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9683814023071848, "standard_deviation": 0.0028221494343927903, "minimum": 0.9647993361947571, "maximum": 0.9718064365015473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0384133622664198, "standard_deviation": 0.028735287804129938, "minimum": 1.0055746985261895, "maximum": 1.0773308313201517, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8250657618343427, "standard_deviation": 0.4186739866649537, "minimum": 0.37776383845036143, "maximum": 1.458057779166471, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18234759545415274, "standard_deviation": 0.006715048896480707, "minimum": 0.17333397506897805, "maximum": 0.19038661055539022, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9946292128794483, "standard_deviation": 0.0003725683003751112, "minimum": 0.9938539626878994, "maximum": 0.995052626179284, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0474808488402494, "standard_deviation": 0.03531170976925895, "minimum": 0.9808374543878564, "maximum": 1.08432639165273, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.39290960362354016, "standard_deviation": 0.171878356568968, "minimum": 0.22992324765984465, "maximum": 0.7434141526082184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0766001288882235, "standard_deviation": 0.0029424459265899846, "minimum": 0.07340470769080026, "maximum": 0.08297635780089826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981854201827866, "standard_deviation": 0.0002513363550090019, "minimum": 0.9977901753779903, "maximum": 0.998549685761817, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5927430486153769, "standard_deviation": 0.029683525250650697, "minimum": 0.5371318834502784, "maximum": 0.6285129069143304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5106671332647059, "standard_deviation": 0.1480180672632, "minimum": 0.3182635886259521, "maximum": 0.8081926524205089, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04371938885448166, "standard_deviation": 0.003487563754430914, "minimum": 0.03901770296089131, "maximum": 0.05030535000969354, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15768685442199842, "standard_deviation": 0.030186910563683042, "minimum": 0.11163131358081844, "maximum": 0.19520648039869412, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.414261511595746, "standard_deviation": 0.010102448264229381, "minimum": 5.403370442402918, "maximum": 5.435091985454059, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3193164180023569, "standard_deviation": 0.007180253285298438, "minimum": 0.31086627020093466, "maximum": 0.32875177378942627, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09439979836217356, "standard_deviation": 0.02490032347012309, "minimum": -0.14535537523928052, "maximum": -0.07016656040773915, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9960921725202572, "standard_deviation": 0.00039192761371258187, "minimum": 0.9953308878614385, "maximum": 0.9965031223138487, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5990535015464312, "standard_deviation": 0.018133040587729025, "minimum": 0.5584624081402753, "maximum": 0.6181858683304394, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.325362774844354, "standard_deviation": 0.19414413172585882, "minimum": 0.12739459066914122, "maximum": 0.7361264901165585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.059697465304836175, "standard_deviation": 0.002719463143342334, "minimum": 0.0564594300451528, "maximum": 0.06501139239990421, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987674918420939, "standard_deviation": 0.00015942304318553, "minimum": 0.9984060526468869, "maximum": 0.9989587930686105, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.33335833957529093, "standard_deviation": 0.01440927002602755, "minimum": 0.31042741977142935, "maximum": 0.3554592069772115, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43545125395738304, "standard_deviation": 0.1874153324138157, "minimum": 0.22629979809482606, "maximum": 0.7308721380435035, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03354590454904566, "standard_deviation": 0.0018983266066575236, "minimum": 0.03205793731349016, "maximum": 0.0375223926655088, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0007", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.7952802181243896, "predicted_mos": 2.7952802181243896, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0e990e551ad7c0c584bf5c7904230a15d0da28a13253f12fb170bfc8cc53c033", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.852994400076568, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5856857688717183, "全帯域_10ms_rms_envelope_nrmse": 0.05986184154643562, "全帯域_10ms_envelope_correlation": 0.996172488767677, "全帯域_10ms_modulation_mae_db": 0.25753284897087964, "全帯域_40ms_log_rms_mae_db": 0.3174113669738127, "全帯域_40ms_rms_envelope_nrmse": 0.03340266827714624, "全帯域_40ms_envelope_correlation": 0.99889524997713, "全帯域_40ms_modulation_mae_db": 0.26935985043888305, "300–1000 Hz_10ms_log_rms_mae_db": 1.1219006463248298, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07574745890877747, "300–1000 Hz_10ms_envelope_correlation": 0.9948725578371342, "300–1000 Hz_10ms_modulation_mae_db": 0.4187903077531454, "300–1000 Hz_40ms_log_rms_mae_db": 0.6018710189206484, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04491727327258837, "300–1000 Hz_40ms_envelope_correlation": 0.9982334206136555, "300–1000 Hz_40ms_modulation_mae_db": 0.5202434302196712, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2061212286051302, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10058996365821066, "1000–3000 Hz_10ms_envelope_correlation": 0.9903058842189663, "1000–3000 Hz_10ms_modulation_mae_db": 0.32356903885776295, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7154453416328677, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05702505795530962, "1000–3000 Hz_40ms_envelope_correlation": 0.9969925462917142, "1000–3000 Hz_40ms_modulation_mae_db": 0.4054733158458876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.893006446135197, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29002597965539245, "3000–6000 Hz_10ms_envelope_correlation": 0.925733714689876, "3000–6000 Hz_10ms_modulation_mae_db": 0.5221028974870744, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1858991708347681, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2169867216528057, "3000–6000 Hz_40ms_envelope_correlation": 0.9544828924103614, "3000–6000 Hz_40ms_modulation_mae_db": 0.5634605026984355, "mr_stft_db": 5.666462923348476, "band_mae_db": 0.13944887496466216, "rms_gain_error_db": -0.13059532636230364, "output_peak": 0.3210694561637506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5848264401984038, "全帯域_10ms_rms_envelope_nrmse": 0.05411301564600171, "全帯域_10ms_envelope_correlation": 0.9967634055597698, "全帯域_10ms_modulation_mae_db": 0.10650621818245838, "全帯域_40ms_log_rms_mae_db": 0.3245002484285437, "全帯域_40ms_rms_envelope_nrmse": 0.027752836689434755, "全帯域_40ms_envelope_correlation": 0.999137053944072, "全帯域_40ms_modulation_mae_db": 0.09937414956904674, "300–1000 Hz_10ms_log_rms_mae_db": 1.1278901383593172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07343069189160918, "300–1000 Hz_10ms_envelope_correlation": 0.9950573076229616, "300–1000 Hz_10ms_modulation_mae_db": 0.1950314578533955, "300–1000 Hz_40ms_log_rms_mae_db": 0.6369295512582734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04077630929557102, "300–1000 Hz_40ms_envelope_correlation": 0.9984198746734495, "300–1000 Hz_40ms_modulation_mae_db": 0.3140980123049875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.213165698562428, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10387284206676206, "1000–3000 Hz_10ms_envelope_correlation": 0.9896255341565973, "1000–3000 Hz_10ms_modulation_mae_db": 0.3408688564738209, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7129731158211867, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059637748060551665, "1000–3000 Hz_40ms_envelope_correlation": 0.9966666651314522, "1000–3000 Hz_40ms_modulation_mae_db": 0.28001376453273724, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9267451655322885, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28818522005100194, "3000–6000 Hz_10ms_envelope_correlation": 0.9287647299676886, "3000–6000 Hz_10ms_modulation_mae_db": 0.5737316256476236, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2070419765961389, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2053614989551312, "3000–6000 Hz_40ms_envelope_correlation": 0.9606583441639513, "3000–6000 Hz_40ms_modulation_mae_db": 0.6753841708319452, "mr_stft_db": 5.634575962835242, "band_mae_db": 0.1712864789451017, "rms_gain_error_db": -0.06980999544501174, "output_peak": 0.3148607381988732}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5546844090365485, "全帯域_10ms_rms_envelope_nrmse": 0.05552830909927886, "全帯域_10ms_envelope_correlation": 0.9965927607352488, "全帯域_10ms_modulation_mae_db": 0.33204989519625583, "全帯域_40ms_log_rms_mae_db": 0.2780931218260938, "全帯域_40ms_rms_envelope_nrmse": 0.02904945268312042, "全帯域_40ms_envelope_correlation": 0.9990540022278908, "全帯域_40ms_modulation_mae_db": 0.33525306840891816, "300–1000 Hz_10ms_log_rms_mae_db": 1.043040852312232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07723795075325673, "300–1000 Hz_10ms_envelope_correlation": 0.9945085290949129, "300–1000 Hz_10ms_modulation_mae_db": 0.2871897486781507, "300–1000 Hz_40ms_log_rms_mae_db": 0.5586455295523597, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042030707561055605, "300–1000 Hz_40ms_envelope_correlation": 0.9982972276617776, "300–1000 Hz_40ms_modulation_mae_db": 0.28041524960111935, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1580260081224831, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10929903565090862, "1000–3000 Hz_10ms_envelope_correlation": 0.9884128196159715, "1000–3000 Hz_10ms_modulation_mae_db": 0.3490877904376732, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6738204399355969, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05443905479611113, "1000–3000 Hz_40ms_envelope_correlation": 0.9972477518080338, "1000–3000 Hz_40ms_modulation_mae_db": 0.5328292025080393, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8773584891890318, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2774544797241317, "3000–6000 Hz_10ms_envelope_correlation": 0.9367560535568523, "3000–6000 Hz_10ms_modulation_mae_db": 0.7259437535556033, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1722546096321538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19772137166758902, "3000–6000 Hz_40ms_envelope_correlation": 0.9657008557937741, "3000–6000 Hz_40ms_modulation_mae_db": 0.5541820662302879, "mr_stft_db": 5.620322937561608, "band_mae_db": 0.14595066072439952, "rms_gain_error_db": -0.07359098889351598, "output_peak": 0.3183535337404563}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6053959510620258, "全帯域_10ms_rms_envelope_nrmse": 0.05728938338751179, "全帯域_10ms_envelope_correlation": 0.9964288745167188, "全帯域_10ms_modulation_mae_db": 0.525618914155603, "全帯域_40ms_log_rms_mae_db": 0.31407462408530873, "全帯域_40ms_rms_envelope_nrmse": 0.03242655825065978, "全帯域_40ms_envelope_correlation": 0.9988730498748599, "全帯域_40ms_modulation_mae_db": 0.6295418194151025, "300–1000 Hz_10ms_log_rms_mae_db": 1.1462168833474748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08147926002306494, "300–1000 Hz_10ms_envelope_correlation": 0.9940102668879205, "300–1000 Hz_10ms_modulation_mae_db": 0.7198743112051448, "300–1000 Hz_40ms_log_rms_mae_db": 0.6369290515292034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04797394428878852, "300–1000 Hz_40ms_envelope_correlation": 0.9979350959723348, "300–1000 Hz_40ms_modulation_mae_db": 1.0290532241702184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2026319483138597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10935032265858055, "1000–3000 Hz_10ms_envelope_correlation": 0.9883497982899984, "1000–3000 Hz_10ms_modulation_mae_db": 0.5125399131293671, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7119386985942876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05627992640095599, "1000–3000 Hz_40ms_envelope_correlation": 0.9969161811492119, "1000–3000 Hz_40ms_modulation_mae_db": 0.44737474921101256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8500169594868912, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2872386196874038, "3000–6000 Hz_10ms_envelope_correlation": 0.9277444427913911, "3000–6000 Hz_10ms_modulation_mae_db": 0.543196880139, "3000–6000 Hz_40ms_log_rms_mae_db": 1.180569421166764, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20530756949190876, "3000–6000 Hz_40ms_envelope_correlation": 0.9596193824575322, "3000–6000 Hz_40ms_modulation_mae_db": 0.800229237687403, "mr_stft_db": 5.6442091321735015, "band_mae_db": 0.18750690224253833, "rms_gain_error_db": -0.10133523613631673, "output_peak": 0.3271738422294197}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5736079572257131, "全帯域_10ms_rms_envelope_nrmse": 0.060639206352362474, "全帯域_10ms_envelope_correlation": 0.9959543452299445, "全帯域_10ms_modulation_mae_db": 0.37370125867671117, "全帯域_40ms_log_rms_mae_db": 0.3150409667860738, "全帯域_40ms_rms_envelope_nrmse": 0.03345776332850025, "全帯域_40ms_envelope_correlation": 0.9987575515337564, "全帯域_40ms_modulation_mae_db": 0.45670958265338335, "300–1000 Hz_10ms_log_rms_mae_db": 1.086147630391911, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07553846553474848, "300–1000 Hz_10ms_envelope_correlation": 0.9947799791359679, "300–1000 Hz_10ms_modulation_mae_db": 0.5072564038213905, "300–1000 Hz_40ms_log_rms_mae_db": 0.6427022694598317, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04187414993558019, "300–1000 Hz_40ms_envelope_correlation": 0.9983470872688084, "300–1000 Hz_40ms_modulation_mae_db": 0.6356630059303464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2396896578851646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11036668789728694, "1000–3000 Hz_10ms_envelope_correlation": 0.9884262312701141, "1000–3000 Hz_10ms_modulation_mae_db": 0.757177888949576, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7398833475703924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0667766260412669, "1000–3000 Hz_40ms_envelope_correlation": 0.995953327849999, "1000–3000 Hz_40ms_modulation_mae_db": 0.9038302462396194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.837274388669772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2630862429174505, "3000–6000 Hz_10ms_envelope_correlation": 0.9426523406195549, "3000–6000 Hz_10ms_modulation_mae_db": 0.5130629661462166, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1303470703942238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18717182007600525, "3000–6000 Hz_40ms_envelope_correlation": 0.9689341174489692, "3000–6000 Hz_40ms_modulation_mae_db": 0.7326264111325433, "mr_stft_db": 5.64772038586212, "band_mae_db": 0.10283595935285676, "rms_gain_error_db": -0.08995539284135176, "output_peak": 0.31667066992261916}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5539027847545206, "全帯域_10ms_rms_envelope_nrmse": 0.05248194398151487, "全帯域_10ms_envelope_correlation": 0.9970668822242067, "全帯域_10ms_modulation_mae_db": 0.3464738087224024, "全帯域_40ms_log_rms_mae_db": 0.2876109154480804, "全帯域_40ms_rms_envelope_nrmse": 0.028829670239504803, "全帯域_40ms_envelope_correlation": 0.9991883647662089, "全帯域_40ms_modulation_mae_db": 0.37667192419481016, "300–1000 Hz_10ms_log_rms_mae_db": 1.0641258522377337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07262949666970593, "300–1000 Hz_10ms_envelope_correlation": 0.9952580395081667, "300–1000 Hz_10ms_modulation_mae_db": 0.31846530382635935, "300–1000 Hz_40ms_log_rms_mae_db": 0.611142501758001, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04195564128950215, "300–1000 Hz_40ms_envelope_correlation": 0.9984312004130124, "300–1000 Hz_40ms_modulation_mae_db": 0.4366304430603712, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2305101035107493, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10840645270036427, "1000–3000 Hz_10ms_envelope_correlation": 0.9886948418562115, "1000–3000 Hz_10ms_modulation_mae_db": 0.540164310730074, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7541419059449079, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06099172896156769, "1000–3000 Hz_40ms_envelope_correlation": 0.9965406403180435, "1000–3000 Hz_40ms_modulation_mae_db": 0.7801906917974312, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8980741958617016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28303659857749974, "3000–6000 Hz_10ms_envelope_correlation": 0.9290521660552284, "3000–6000 Hz_10ms_modulation_mae_db": 0.6316558323236419, "3000–6000 Hz_40ms_log_rms_mae_db": 1.19425959647394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19808335109293818, "3000–6000 Hz_40ms_envelope_correlation": 0.9619395741163371, "3000–6000 Hz_40ms_modulation_mae_db": 0.9689498480323984, "mr_stft_db": 5.64170437996911, "band_mae_db": 0.13113681503721575, "rms_gain_error_db": -0.11403115815198364, "output_peak": 0.33002117306006906}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5788524693352254, "全帯域_10ms_rms_envelope_nrmse": 0.056509122089161506, "全帯域_10ms_envelope_correlation": 0.9965154676600104, "全帯域_10ms_modulation_mae_db": 0.6846086055523811, "全帯域_40ms_log_rms_mae_db": 0.3083648791815619, "全帯域_40ms_rms_envelope_nrmse": 0.03254435989984418, "全帯域_40ms_envelope_correlation": 0.998842500095178, "全帯域_40ms_modulation_mae_db": 0.8819192515905735, "300–1000 Hz_10ms_log_rms_mae_db": 1.0867890467088275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07325532207905272, "300–1000 Hz_10ms_envelope_correlation": 0.9951453271165832, "300–1000 Hz_10ms_modulation_mae_db": 0.7309187833382217, "300–1000 Hz_40ms_log_rms_mae_db": 0.5549151283846027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04256218664705781, "300–1000 Hz_40ms_envelope_correlation": 0.9983401679861393, "300–1000 Hz_40ms_modulation_mae_db": 0.9473924954255675, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2233212575193317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10072663524990122, "1000–3000 Hz_10ms_envelope_correlation": 0.9902835819718057, "1000–3000 Hz_10ms_modulation_mae_db": 0.3579378196825438, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7376323810107119, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055932761126313305, "1000–3000 Hz_40ms_envelope_correlation": 0.9971361605391957, "1000–3000 Hz_40ms_modulation_mae_db": 0.4535321812337852, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8814035943364766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28075609493177656, "3000–6000 Hz_10ms_envelope_correlation": 0.9364540184015844, "3000–6000 Hz_10ms_modulation_mae_db": 0.4694696270706153, "3000–6000 Hz_40ms_log_rms_mae_db": 1.173383168705635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20015290965419663, "3000–6000 Hz_40ms_envelope_correlation": 0.9657657366347723, "3000–6000 Hz_40ms_modulation_mae_db": 0.5780761742856061, "mr_stft_db": 5.661257318678463, "band_mae_db": 0.18174314160663255, "rms_gain_error_db": -0.09369516449711388, "output_peak": 0.3179117275485853}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5796266438102836, "全帯域_10ms_rms_envelope_nrmse": 0.058709035430694896, "全帯域_10ms_envelope_correlation": 0.9962689310490385, "全帯域_10ms_modulation_mae_db": 0.3677044750865757, "全帯域_40ms_log_rms_mae_db": 0.3240139857076392, "全帯域_40ms_rms_envelope_nrmse": 0.03392381418443145, "全帯域_40ms_envelope_correlation": 0.9987840860197748, "全帯域_40ms_modulation_mae_db": 0.47720215908444424, "300–1000 Hz_10ms_log_rms_mae_db": 1.1412717894709141, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08413984394355734, "300–1000 Hz_10ms_envelope_correlation": 0.9935097272259766, "300–1000 Hz_10ms_modulation_mae_db": 0.5576430949437529, "300–1000 Hz_40ms_log_rms_mae_db": 0.6362073362475369, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047495199433244206, "300–1000 Hz_40ms_envelope_correlation": 0.9978619068173413, "300–1000 Hz_40ms_modulation_mae_db": 0.7699579524397198, "1000–3000 Hz_10ms_log_rms_mae_db": 1.221075497555399, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10274415341594012, "1000–3000 Hz_10ms_envelope_correlation": 0.9901633131734018, "1000–3000 Hz_10ms_modulation_mae_db": 0.6452979265341097, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7310487237814183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06103398830174727, "1000–3000 Hz_40ms_envelope_correlation": 0.9968693580958118, "1000–3000 Hz_40ms_modulation_mae_db": 0.7556478532081793, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8854215218946846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2659534204909101, "3000–6000 Hz_10ms_envelope_correlation": 0.9380690023038434, "3000–6000 Hz_10ms_modulation_mae_db": 1.1338672888255366, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1443505513946897, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19225486819592078, "3000–6000 Hz_40ms_envelope_correlation": 0.9645653550164629, "3000–6000 Hz_40ms_modulation_mae_db": 0.7706102048970752, "mr_stft_db": 5.647180766377967, "band_mae_db": 0.11456853092172664, "rms_gain_error_db": -0.11212832314057253, "output_peak": 0.3241945024421669}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.379667997802834, "spectral_convergence": 0.20864402700652687, "spectral_convergence_status": "ok", "amplitude_floor": 1.0845908458793198e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.511858606189778, "spectral_convergence": 0.210465513230916, "spectral_convergence_status": "ok", "amplitude_floor": 9.938577161724507e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.107862166052818, "spectral_convergence": 0.22149059695643233, "spectral_convergence_status": "ok", "amplitude_floor": 7.780261052298239e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.1610975793667e-05, "output_power": 8.517318141252922e-05, "delta_db": 0.18554298844631292, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005078113602376632, "output_power": 0.0004993343460755993, "delta_db": -0.07310972765355472, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000931370997501075, "output_power": 0.000897996923518515, "delta_db": -0.15847861256765838, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000354698831360673, "output_power": 0.00033816967715463025, "delta_db": -0.20725094701508504, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.667116250090391e-05, "output_power": 9.275629262912223e-05, "delta_db": -0.17953559129104593, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.4087248189368409e-05, "output_power": 1.3981334550534543e-05, "delta_db": -0.03277538281431594, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13944887496466216, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13059532636230364, "pitch": null, "output_rms_dbfs": -27.150254965692465, "output_peak": 0.3210694561637506, "reference_rms_dbfs": -27.01965963933016, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.666462923348476}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001511941088699898, "log_rms_mae_db": 0.5856857688717183, "rms_envelope_nrmse": 0.05986184154643562, "envelope_correlation": 0.996172488767677, "rms_gain_error_db": -0.13029239798561015, "reference_mean_power": 0.0019872702200913737, "output_mean_power": 0.001928535709074368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.92641346613152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000630498462430925, "output_power": 0.0006183199313069181, "reference_active": true, "delta_db": -0.08470785774071671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.948331074534803e-05, "output_power": 7.232024730070224e-05, "reference_active": true, "delta_db": -0.4101604608061609}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3123506088131581e-05, "output_power": 1.2310532930831278e-05, "reference_active": true, "delta_db": -0.2777302283657613}], "active_band_count": 3, "active_band_mae_db": 0.25753284897087964}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013307215895407113, "log_rms_mae_db": 0.3174113669738127, "rms_envelope_nrmse": 0.03340266827714624, "envelope_correlation": 0.99889524997713, "rms_gain_error_db": -0.129476814640786, "reference_mean_power": 0.0019932283004473585, "output_mean_power": 0.0019346809854605095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.959638607230672e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005794424432114469, "output_power": 0.0005699421470312189, "reference_active": true, "delta_db": -0.0717952930039013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.108494970400102e-05, "output_power": 3.7668313152008196e-05, "reference_active": true, "delta_db": -0.3770658741939476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.934091733283841e-07, "output_power": 7.304245792061151e-07, "reference_active": true, "delta_db": -0.35921838411880025}], "active_band_count": 3, "active_band_mae_db": 0.26935985043888305}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014956359127228178, "log_rms_mae_db": 1.1219006463248298, "rms_envelope_nrmse": 0.07574745890877747, "envelope_correlation": 0.9948725578371342, "rms_gain_error_db": -0.14743675304362092, "reference_mean_power": 0.001265981682968443, "output_mean_power": 0.001223724758401811}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.65319874591582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005127917421071942, "output_power": 0.0005005453604006873, "reference_active": true, "delta_db": -0.10497582562927704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.49871434628785e-05, "output_power": 6.415757116113776e-05, "reference_active": true, "delta_db": -0.6773889570785503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1228212970044594e-05, "output_power": 1.006723084915901e-05, "reference_active": true, "delta_db": -0.47400614055160895}], "active_band_count": 3, "active_band_mae_db": 0.4187903077531454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012969159209010741, "log_rms_mae_db": 0.6018710189206484, "rms_envelope_nrmse": 0.04491727327258837, "envelope_correlation": 0.9982334206136555, "rms_gain_error_db": -0.14682115910618562, "reference_mean_power": 0.0012704534730935044, "output_mean_power": 0.001228221368437983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.90499086389905e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004798268169104879, "output_power": 0.00046690768430303434, "reference_active": true, "delta_db": -0.11853494927288666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7762549535694356e-05, "output_power": 3.279580235602227e-05, "reference_active": true, "delta_db": -0.6124304750371681}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.708330714477498e-07, "output_power": 6.367718446698035e-07, "reference_active": true, "delta_db": -0.8297648663489589}], "active_band_count": 3, "active_band_mae_db": 0.5202434302196712}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.983632889928613e-05, "log_rms_mae_db": 1.2061212286051302, "rms_envelope_nrmse": 0.10058996365821066, "envelope_correlation": 0.9903058842189663, "rms_gain_error_db": -0.19392651377689385, "reference_mean_power": 0.00040681650183919396, "output_mean_power": 0.0003890504389676708}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8313223654566366e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016437493457690703, "output_power": 0.00015909325652460632, "reference_active": true, "delta_db": -0.14183821354638654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7703924674415462e-05, "output_power": 2.457061966782743e-05, "reference_active": true, "delta_db": -0.5212518824959099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.687130809571553e-06, "output_power": 4.366619722843797e-06, "reference_active": true, "delta_db": -0.3076170205309924}], "active_band_count": 3, "active_band_mae_db": 0.32356903885776295}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.811887222434066e-05, "log_rms_mae_db": 0.7154453416328677, "rms_envelope_nrmse": 0.05702505795530962, "envelope_correlation": 0.9969925462917142, "rms_gain_error_db": -0.1924224448960264, "reference_mean_power": 0.00040698573595692556, "output_mean_power": 0.00038934709962548203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.593242154160197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015388869283670126, "output_power": 0.00015001389438230934, "reference_active": true, "delta_db": -0.11075225126244086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4161267354242513e-05, "output_power": 1.2817268348293718e-05, "reference_active": true, "delta_db": -0.4330664501824386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7920568451213675e-07, "output_power": 2.39146556331013e-07, "reference_active": true, "delta_db": -0.6726012460927833}], "active_band_count": 3, "active_band_mae_db": 0.4054733158458876}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4474245023095566e-05, "log_rms_mae_db": 1.893006446135197, "rms_envelope_nrmse": 0.29002597965539245, "envelope_correlation": 0.925733714689876, "rms_gain_error_db": 0.044862998331544536, "reference_mean_power": 1.6116938837494647e-05, "output_mean_power": 1.6284291113377323e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.1253846819517355e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.226560638885928e-06, "output_power": 2.7854235994970006e-06, "reference_active": true, "delta_db": 0.972567234855374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.434590024496585e-06, "output_power": 1.368786172729075e-06, "reference_active": true, "delta_db": -0.20392197240488819}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.807607987621042e-07, "output_power": 5.259094765421248e-07, "reference_active": true, "delta_db": 0.38981948520096094}], "active_band_count": 3, "active_band_mae_db": 0.5221028974870744}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.920336454108668e-05, "log_rms_mae_db": 1.1858991708347681, "rms_envelope_nrmse": 0.2169867216528057, "envelope_correlation": 0.9544828924103614, "rms_gain_error_db": 0.044859937531074964, "reference_mean_power": 1.619697385840346e-05, "output_mean_power": 1.6365145654308063e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.254945521069739e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4889493401597257e-06, "output_power": 3.0954612773055755e-06, "reference_active": true, "delta_db": 0.9470931846345412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.64830026091652e-07, "output_power": 7.766379665644264e-07, "reference_active": true, "delta_db": 0.06653688200503244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8106205193126903e-08, "output_power": 4.116479483658204e-08, "reference_active": true, "delta_db": -0.676751441455733}], "active_band_count": 3, "active_band_mae_db": 0.5634605026984355}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8860424474294685, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7df6e05a8611789f47ddffbe39c8d917c45b8a99a3b11a3ec4633b7edeabf82b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9892827505691333, "standard_deviation": 0.0008982362744051123, "minimum": 0.9883497982899984, "maximum": 0.9903058842189663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2118176750093184, "standard_deviation": 0.024947034384572832, "minimum": 1.1580260081224831, "maximum": 1.2396896578851646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.478330443099366, "standard_deviation": 0.16242474331573795, "minimum": 0.32356903885776295, "maximum": 0.757177888949576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10566951166224431, "standard_deviation": 0.004110886287160752, "minimum": 0.10058996365821066, "maximum": 0.11036668789728694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967903288979327, "standard_deviation": 0.0004088344737894073, "minimum": 0.995953327849999, "maximum": 0.9972477518080338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7221104942864212, "standard_deviation": 0.024570648234358606, "minimum": 0.6738204399355969, "maximum": 0.7541419059449079, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5698615005720864, "standard_deviation": 0.21752902986937928, "minimum": 0.28001376453273724, "maximum": 0.9038302462396194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05901461145547794, "standard_deviation": 0.003973023170811302, "minimum": 0.05443905479611113, "maximum": 0.0667766260412669, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9331533085482524, "standard_deviation": 0.006078713515540921, "minimum": 0.925733714689876, "maximum": 0.9426523406195549, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8811625951382553, "standard_deviation": 0.027859193958560482, "minimum": 1.837274388669772, "maximum": 1.9267451655322885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.639128858899414, "standard_deviation": 0.21521101560922914, "minimum": 0.4694696270706153, "maximum": 1.1338672888255366, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2794670820044458, "standard_deviation": 0.010122799980291454, "minimum": 0.2630862429174505, "maximum": 0.29002597965539245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9627082822552702, "standard_deviation": 0.004511964494665436, "minimum": 0.9544828924103614, "maximum": 0.9689341174489692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.173513195649789, "standard_deviation": 0.025278930057694278, "minimum": 1.1303470703942238, "maximum": 1.2070419765961389, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7054398269744617, "standard_deviation": 0.14321349362684518, "minimum": 0.5541820662302879, "maximum": 0.9689498480323984, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20038001384831192, "standard_deviation": 0.00908489902110275, "minimum": 0.18717182007600525, "maximum": 0.2169867216528057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.994642716803703, "standard_deviation": 0.0006066838637289218, "minimum": 0.9935097272259766, "maximum": 0.9952580395081667, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.102172854894155, "standard_deviation": 0.03771472820702957, "minimum": 1.043040852312232, "maximum": 1.1462168833474748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.46689617642744513, "standard_deviation": 0.1978890557620792, "minimum": 0.1950314578533955, "maximum": 0.7309187833382217, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07668231122547159, "standard_deviation": 0.004137963057153148, "minimum": 0.07262949666970593, "maximum": 0.08413984394355734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982332476758149, "standard_deviation": 0.00021688979033935953, "minimum": 0.9978619068173413, "maximum": 0.9984312004130124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6099177983888071, "standard_deviation": 0.035748314147879094, "minimum": 0.5549151283846027, "maximum": 0.6427022694598317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6166817266440001, "standard_deviation": 0.28021028135032555, "minimum": 0.28041524960111935, "maximum": 1.0290532241702184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04369817646542348, "standard_deviation": 0.0027557085092674376, "minimum": 0.04077630929557102, "maximum": 0.04797394428878852, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1468096704743917, "standard_deviation": 0.031051713058668183, "minimum": 0.10283595935285676, "maximum": 0.18750690224253833, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.645429225850811, "standard_deviation": 0.014465448472477384, "minimum": 5.620322937561608, "maximum": 5.666462923348476, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.32128195541324256, "standard_deviation": 0.005372501614750537, "minimum": 0.3148607381988732, "maximum": 0.33002117306006906, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09814269818352124, "standard_deviation": 0.020714891945121404, "minimum": -0.13059532636230364, "maximum": -0.06980999544501174, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964703944678268, "standard_deviation": 0.00034969870237917225, "minimum": 0.9959543452299445, "maximum": 0.9970668822242067, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5770728030368049, "standard_deviation": 0.016887388312848188, "minimum": 0.5539027847545206, "maximum": 0.6053959510620258, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37427450306790844, "standard_deviation": 0.1720216531706504, "minimum": 0.10650621818245838, "maximum": 0.6846086055523811, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05689148219162021, "standard_deviation": 0.0028157374045443082, "minimum": 0.05248194398151487, "maximum": 0.060639206352362474, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989414823048588, "standard_deviation": 0.00016348087324496958, "minimum": 0.9987575515337564, "maximum": 0.9991883647662089, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.30863876355463926, "standard_deviation": 0.016948498258740712, "minimum": 0.2780931218260938, "maximum": 0.3245002484285437, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4407539756693952, "standard_deviation": 0.23694598078738155, "minimum": 0.09937414956904674, "maximum": 0.8819192515905735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03142339044408023, "standard_deviation": 0.00246133473319518, "minimum": 0.027752836689434755, "maximum": 0.03392381418443145, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0007", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8548200130462646, "predicted_mos": 2.8548200130462646, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "55f7c3d57b731237606748dcf314fe393b4e48a4ab679ad41805b89479ecf89e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8453265000134706, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5936630494503533, "全帯域_10ms_rms_envelope_nrmse": 0.05940924664405841, "全帯域_10ms_envelope_correlation": 0.9960928840679696, "全帯域_10ms_modulation_mae_db": 0.07215392260529617, "全帯域_40ms_log_rms_mae_db": 0.3171796565340536, "全帯域_40ms_rms_envelope_nrmse": 0.030852044546254915, "全帯域_40ms_envelope_correlation": 0.9989263090551648, "全帯域_40ms_modulation_mae_db": 0.17133748097439724, "300–1000 Hz_10ms_log_rms_mae_db": 0.9998188721183835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07584399367411593, "300–1000 Hz_10ms_envelope_correlation": 0.9947267027117024, "300–1000 Hz_10ms_modulation_mae_db": 0.4528412813851261, "300–1000 Hz_40ms_log_rms_mae_db": 0.555927640898717, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040981551636164786, "300–1000 Hz_40ms_envelope_correlation": 0.9984099397449955, "300–1000 Hz_40ms_modulation_mae_db": 0.3882392430241544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0457241649859668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09456343359264802, "1000–3000 Hz_10ms_envelope_correlation": 0.9913398407174705, "1000–3000 Hz_10ms_modulation_mae_db": 0.19262519504384445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6292964935637718, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0510624008537289, "1000–3000 Hz_40ms_envelope_correlation": 0.9974438132397101, "1000–3000 Hz_40ms_modulation_mae_db": 0.34798963884626827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5028682041929935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2516802732781047, "3000–6000 Hz_10ms_envelope_correlation": 0.9432518165020259, "3000–6000 Hz_10ms_modulation_mae_db": 0.6900402564786626, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9986567672673398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18283197399825848, "3000–6000 Hz_40ms_envelope_correlation": 0.9672330488943574, "3000–6000 Hz_40ms_modulation_mae_db": 0.4138954098013808, "mr_stft_db": 5.402825597892633, "band_mae_db": 0.10146580670866996, "rms_gain_error_db": -0.07564891530024204, "output_peak": 0.3277445370120416}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5834912277814728, "全帯域_10ms_rms_envelope_nrmse": 0.05431908671979257, "全帯域_10ms_envelope_correlation": 0.9967967270950876, "全帯域_10ms_modulation_mae_db": 0.39702958614676603, "全帯域_40ms_log_rms_mae_db": 0.3154025083538866, "全帯域_40ms_rms_envelope_nrmse": 0.029046982093643994, "全帯域_40ms_envelope_correlation": 0.9991198664983091, "全帯域_40ms_modulation_mae_db": 0.3903602109355948, "300–1000 Hz_10ms_log_rms_mae_db": 0.9917553450182839, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.069253196040876, "300–1000 Hz_10ms_envelope_correlation": 0.9956576863050444, "300–1000 Hz_10ms_modulation_mae_db": 0.6171743809716922, "300–1000 Hz_40ms_log_rms_mae_db": 0.5779797952790121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03970824356182062, "300–1000 Hz_40ms_envelope_correlation": 0.998558253271742, "300–1000 Hz_40ms_modulation_mae_db": 0.6840534338931471, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0553600249818471, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10034970825679189, "1000–3000 Hz_10ms_envelope_correlation": 0.9903143751152497, "1000–3000 Hz_10ms_modulation_mae_db": 0.3201003215635551, "1000–3000 Hz_40ms_log_rms_mae_db": 0.601435210381375, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054720555790843436, "1000–3000 Hz_40ms_envelope_correlation": 0.9972105918592133, "1000–3000 Hz_40ms_modulation_mae_db": 0.4208017365146155, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4909387577679736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2539492707939407, "3000–6000 Hz_10ms_envelope_correlation": 0.9437232900423347, "3000–6000 Hz_10ms_modulation_mae_db": 1.008942357867876, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9789342111284656, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18647980393065047, "3000–6000 Hz_40ms_envelope_correlation": 0.9666751747367801, "3000–6000 Hz_40ms_modulation_mae_db": 0.8606253525570017, "mr_stft_db": 5.396819637611601, "band_mae_db": 0.10296516909916759, "rms_gain_error_db": -0.0983525788302573, "output_peak": 0.3158543537370337}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.612118542607921, "全帯域_10ms_rms_envelope_nrmse": 0.06211068090229433, "全帯域_10ms_envelope_correlation": 0.9958191287327204, "全帯域_10ms_modulation_mae_db": 0.3265346053725915, "全帯域_40ms_log_rms_mae_db": 0.35004632176396494, "全帯域_40ms_rms_envelope_nrmse": 0.036535579541532304, "全帯域_40ms_envelope_correlation": 0.9985790588232927, "全帯域_40ms_modulation_mae_db": 0.3243423258996429, "300–1000 Hz_10ms_log_rms_mae_db": 0.98928220231896, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07059932820139489, "300–1000 Hz_10ms_envelope_correlation": 0.9954777886448679, "300–1000 Hz_10ms_modulation_mae_db": 0.6219336623338002, "300–1000 Hz_40ms_log_rms_mae_db": 0.5567267957268757, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0432845303855042, "300–1000 Hz_40ms_envelope_correlation": 0.9982496552232455, "300–1000 Hz_40ms_modulation_mae_db": 0.6094167739059796, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1101449397552834, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11141563094639702, "1000–3000 Hz_10ms_envelope_correlation": 0.9883870421991638, "1000–3000 Hz_10ms_modulation_mae_db": 0.39523985983678056, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6788006371329027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06381925797066912, "1000–3000 Hz_40ms_envelope_correlation": 0.9966028525586917, "1000–3000 Hz_40ms_modulation_mae_db": 0.3629595913244346, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5493841059810423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24660495028588345, "3000–6000 Hz_10ms_envelope_correlation": 0.9500830828782765, "3000–6000 Hz_10ms_modulation_mae_db": 1.279739103498487, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0155484517942284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16865782296073056, "3000–6000 Hz_40ms_envelope_correlation": 0.9749521530457135, "3000–6000 Hz_40ms_modulation_mae_db": 1.128642538442541, "mr_stft_db": 5.388022674474111, "band_mae_db": 0.21895157750597347, "rms_gain_error_db": -0.11677428896378565, "output_peak": 0.32483859770251416}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5967386662451578, "全帯域_10ms_rms_envelope_nrmse": 0.0595788932496066, "全帯域_10ms_envelope_correlation": 0.9961964690651411, "全帯域_10ms_modulation_mae_db": 0.5600482472297205, "全帯域_40ms_log_rms_mae_db": 0.3251834665233741, "全帯域_40ms_rms_envelope_nrmse": 0.034990092290038334, "全帯域_40ms_envelope_correlation": 0.9987404784956555, "全帯域_40ms_modulation_mae_db": 0.6439025051930161, "300–1000 Hz_10ms_log_rms_mae_db": 0.9838287024289726, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07632642618652612, "300–1000 Hz_10ms_envelope_correlation": 0.9947087474914529, "300–1000 Hz_10ms_modulation_mae_db": 0.609026055406417, "300–1000 Hz_40ms_log_rms_mae_db": 0.5170017961721316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04439312858514874, "300–1000 Hz_40ms_envelope_correlation": 0.9981747039257837, "300–1000 Hz_40ms_modulation_mae_db": 0.6724001724726515, "1000–3000 Hz_10ms_log_rms_mae_db": 1.021069034005683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10315273477009532, "1000–3000 Hz_10ms_envelope_correlation": 0.9902198095920186, "1000–3000 Hz_10ms_modulation_mae_db": 0.6175486817949967, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5930376532595476, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06321528342298147, "1000–3000 Hz_40ms_envelope_correlation": 0.9967420246658203, "1000–3000 Hz_40ms_modulation_mae_db": 0.700703379141932, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4998717899665008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24488234120907396, "3000–6000 Hz_10ms_envelope_correlation": 0.9489905025813551, "3000–6000 Hz_10ms_modulation_mae_db": 0.8818526662743023, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0312805904295552, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17941257390732862, "3000–6000 Hz_40ms_envelope_correlation": 0.9701427455665973, "3000–6000 Hz_40ms_modulation_mae_db": 0.6387100266526134, "mr_stft_db": 5.380765980336361, "band_mae_db": 0.20457108537431812, "rms_gain_error_db": -0.12247305071488689, "output_peak": 0.31864545547482587}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5950573156460736, "全帯域_10ms_rms_envelope_nrmse": 0.0566369678126118, "全帯域_10ms_envelope_correlation": 0.9964880671804779, "全帯域_10ms_modulation_mae_db": 0.3417312070275094, "全帯域_40ms_log_rms_mae_db": 0.33983411187322005, "全帯域_40ms_rms_envelope_nrmse": 0.03295129062701137, "全帯域_40ms_envelope_correlation": 0.9987987610047274, "全帯域_40ms_modulation_mae_db": 0.3062335618162602, "300–1000 Hz_10ms_log_rms_mae_db": 1.0284966082728637, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07102834160694689, "300–1000 Hz_10ms_envelope_correlation": 0.995397293706706, "300–1000 Hz_10ms_modulation_mae_db": 0.33786630756665303, "300–1000 Hz_40ms_log_rms_mae_db": 0.5967567992256095, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0412775205465734, "300–1000 Hz_40ms_envelope_correlation": 0.998392378277869, "300–1000 Hz_40ms_modulation_mae_db": 0.3091807144331738, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0331342735502342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10525881434879421, "1000–3000 Hz_10ms_envelope_correlation": 0.9893486461733088, "1000–3000 Hz_10ms_modulation_mae_db": 0.2502802022270918, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5955093140393101, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05783439794221196, "1000–3000 Hz_40ms_envelope_correlation": 0.9969080444005363, "1000–3000 Hz_40ms_modulation_mae_db": 0.2879987293562168, "3000–6000 Hz_10ms_log_rms_mae_db": 1.504234148496868, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25640762297068137, "3000–6000 Hz_10ms_envelope_correlation": 0.9447189034780061, "3000–6000 Hz_10ms_modulation_mae_db": 1.8704577815002026, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9901008252222004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1861918040171061, "3000–6000 Hz_40ms_envelope_correlation": 0.9681833600224373, "3000–6000 Hz_40ms_modulation_mae_db": 1.8088746274021315, "mr_stft_db": 5.396087496059606, "band_mae_db": 0.09868247477125985, "rms_gain_error_db": -0.08955846093156475, "output_peak": 0.3233338788355641}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6183631918823046, "全帯域_10ms_rms_envelope_nrmse": 0.06600987810032452, "全帯域_10ms_envelope_correlation": 0.9952122599401299, "全帯域_10ms_modulation_mae_db": 0.2120370051678547, "全帯域_40ms_log_rms_mae_db": 0.3612803904569225, "全帯域_40ms_rms_envelope_nrmse": 0.03626734225558165, "全帯域_40ms_envelope_correlation": 0.9985561172027356, "全帯域_40ms_modulation_mae_db": 0.3262537963293464, "300–1000 Hz_10ms_log_rms_mae_db": 1.0248249802580043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07548361820760725, "300–1000 Hz_10ms_envelope_correlation": 0.9947851902832296, "300–1000 Hz_10ms_modulation_mae_db": 0.26599139521512966, "300–1000 Hz_40ms_log_rms_mae_db": 0.5716938123682119, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04299553923514633, "300–1000 Hz_40ms_envelope_correlation": 0.9982481937834171, "300–1000 Hz_40ms_modulation_mae_db": 0.3814768536334263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0892255325117974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11330614779442237, "1000–3000 Hz_10ms_envelope_correlation": 0.9877733193919378, "1000–3000 Hz_10ms_modulation_mae_db": 0.7366398706214906, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6081015743693888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06566884295594112, "1000–3000 Hz_40ms_envelope_correlation": 0.996107606473235, "1000–3000 Hz_40ms_modulation_mae_db": 0.6833506231986757, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4883464003464382, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2603764806142565, "3000–6000 Hz_10ms_envelope_correlation": 0.9397022507557532, "3000–6000 Hz_10ms_modulation_mae_db": 0.4756600328566198, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9917299530249033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18303762890337588, "3000–6000 Hz_40ms_envelope_correlation": 0.9673019827516512, "3000–6000 Hz_40ms_modulation_mae_db": 0.308669624593512, "mr_stft_db": 5.373074954257575, "band_mae_db": 0.19632790336357844, "rms_gain_error_db": -0.10254935638730735, "output_peak": 0.31767337919555083}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5813438027930531, "全帯域_10ms_rms_envelope_nrmse": 0.060475329849430604, "全帯域_10ms_envelope_correlation": 0.9960111500617099, "全帯域_10ms_modulation_mae_db": 0.3364717532268748, "全帯域_40ms_log_rms_mae_db": 0.3169801053205686, "全帯域_40ms_rms_envelope_nrmse": 0.03582212007406784, "全帯域_40ms_envelope_correlation": 0.998598975681531, "全帯域_40ms_modulation_mae_db": 0.3972862978813197, "300–1000 Hz_10ms_log_rms_mae_db": 1.0125527164829067, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07163077190906317, "300–1000 Hz_10ms_envelope_correlation": 0.9954162913411754, "300–1000 Hz_10ms_modulation_mae_db": 0.42371838403265477, "300–1000 Hz_40ms_log_rms_mae_db": 0.5955770870194191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04522413621451015, "300–1000 Hz_40ms_envelope_correlation": 0.9981722071747328, "300–1000 Hz_40ms_modulation_mae_db": 0.4748694857381596, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0297774889460118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10438695603434199, "1000–3000 Hz_10ms_envelope_correlation": 0.9895293277090541, "1000–3000 Hz_10ms_modulation_mae_db": 0.2898699332870383, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5873725149989603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058674463071462415, "1000–3000 Hz_40ms_envelope_correlation": 0.9967939989574631, "1000–3000 Hz_40ms_modulation_mae_db": 0.46487980586263244, "3000–6000 Hz_10ms_log_rms_mae_db": 1.537581380280064, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2627378339179931, "3000–6000 Hz_10ms_envelope_correlation": 0.9437362676387123, "3000–6000 Hz_10ms_modulation_mae_db": 1.1588892617550992, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0478407456372287, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19240319323252086, "3000–6000 Hz_40ms_envelope_correlation": 0.9675039106731943, "3000–6000 Hz_40ms_modulation_mae_db": 0.9992177379837152, "mr_stft_db": 5.402014128393733, "band_mae_db": 0.21035794205731664, "rms_gain_error_db": -0.10462369762712809, "output_peak": 0.32177898368980307}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5840440940984767, "全帯域_10ms_rms_envelope_nrmse": 0.06534625539605556, "全帯域_10ms_envelope_correlation": 0.9952412750710833, "全帯域_10ms_modulation_mae_db": 0.1135925485636291, "全帯域_40ms_log_rms_mae_db": 0.3236753246112792, "全帯域_40ms_rms_envelope_nrmse": 0.0390793883688805, "全帯域_40ms_envelope_correlation": 0.9982046034296276, "全帯域_40ms_modulation_mae_db": 0.03616240011501376, "300–1000 Hz_10ms_log_rms_mae_db": 1.022514129252883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08489365460954429, "300–1000 Hz_10ms_envelope_correlation": 0.9933331601847651, "300–1000 Hz_10ms_modulation_mae_db": 0.18234523090419316, "300–1000 Hz_40ms_log_rms_mae_db": 0.5871350196901995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05298666526611168, "300–1000 Hz_40ms_envelope_correlation": 0.9972123232491698, "300–1000 Hz_40ms_modulation_mae_db": 0.15845302168534262, "1000–3000 Hz_10ms_log_rms_mae_db": 1.061696065140106, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10521701740128016, "1000–3000 Hz_10ms_envelope_correlation": 0.9894370236785945, "1000–3000 Hz_10ms_modulation_mae_db": 0.24319866649089752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6324763055079965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06188753663879887, "1000–3000 Hz_40ms_envelope_correlation": 0.9964491744640667, "1000–3000 Hz_40ms_modulation_mae_db": 0.47028803425336957, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5186823204490203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2742440119458416, "3000–6000 Hz_10ms_envelope_correlation": 0.9377803365091798, "3000–6000 Hz_10ms_modulation_mae_db": 0.8702768996434654, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0403388694536284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20247818094270073, "3000–6000 Hz_40ms_envelope_correlation": 0.9632494928321104, "3000–6000 Hz_40ms_modulation_mae_db": 0.4616268535729804, "mr_stft_db": 5.389931647638615, "band_mae_db": 0.1445047560448931, "rms_gain_error_db": -0.06792324243174949, "output_peak": 0.31869126346341603}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.033357605751498, "spectral_convergence": 0.20856278542251913, "spectral_convergence_status": "ok", "amplitude_floor": 1.0845908458793198e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.181263567472094, "spectral_convergence": 0.20739957313332275, "spectral_convergence_status": "ok", "amplitude_floor": 9.938577161724507e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.993855620454308, "spectral_convergence": 0.2193738805077718, "spectral_convergence_status": "ok", "amplitude_floor": 7.780261052298239e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.1610975793667e-05, "output_power": 8.165984914270503e-05, "delta_db": 0.002600026850060214, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005078113602376632, "output_power": 0.0005030575520259045, "delta_db": -0.04084739116965399, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000931370997501075, "output_power": 0.0009143086241249443, "delta_db": -0.08029893834175711, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000354698831360673, "output_power": 0.0003472311318715951, "delta_db": -0.09241101585161739, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.667116250090391e-05, "output_power": 9.115464435304283e-05, "delta_db": -0.2551814017849805, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.4087248189368409e-05, "output_power": 1.4540246618726142e-05, "delta_db": 0.1374560662539506, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10146580670866996, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07564891530024204, "pitch": null, "output_rms_dbfs": -27.095308554630403, "output_peak": 0.3277445370120416, "reference_rms_dbfs": -27.01965963933016, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.402825597892633}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001511941088699898, "log_rms_mae_db": 0.5936630494503533, "rms_envelope_nrmse": 0.05940924664405841, "envelope_correlation": 0.9960928840679696, "rms_gain_error_db": -0.07532672344585993, "reference_mean_power": 0.0019872702200913737, "output_mean_power": 0.0019530989754397562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.92641346613152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000630498462430925, "output_power": 0.0006170880165182608, "reference_active": true, "delta_db": -0.09336919059293988}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.948331074534803e-05, "output_power": 7.745140696280573e-05, "reference_active": true, "delta_db": -0.11246637178789612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3123506088131581e-05, "output_power": 1.3091435082974291e-05, "reference_active": true, "delta_db": -0.01062620543505248}], "active_band_count": 3, "active_band_mae_db": 0.07215392260529617}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013307215895407113, "log_rms_mae_db": 0.3171796565340536, "rms_envelope_nrmse": 0.030852044546254915, "envelope_correlation": 0.9989263090551648, "rms_gain_error_db": -0.07442741315609627, "reference_mean_power": 0.0019932283004473585, "output_mean_power": 0.001959360296472945}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.959638607230672e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005794424432114469, "output_power": 0.0005690518838191003, "reference_active": true, "delta_db": -0.07858438001803401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.108494970400102e-05, "output_power": 4.0097524695428686e-05, "reference_active": true, "delta_db": -0.10565196113271283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.934091733283841e-07, "output_power": 8.56002095076012e-07, "reference_active": true, "delta_db": 0.3297761017724449}], "active_band_count": 3, "active_band_mae_db": 0.17133748097439724}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014956359127228178, "log_rms_mae_db": 0.9998188721183835, "rms_envelope_nrmse": 0.07584399367411593, "envelope_correlation": 0.9947267027117024, "rms_gain_error_db": -0.08964808796256243, "reference_mean_power": 0.001265981682968443, "output_mean_power": 0.00124011686369583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.65319874591582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005127917421071942, "output_power": 0.0005006051907567435, "reference_active": true, "delta_db": -0.10445674299007879}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.49871434628785e-05, "output_power": 6.771063040073856e-05, "reference_active": true, "delta_db": -0.44329952842453885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1228212970044594e-05, "output_power": 9.316094477511953e-06, "reference_active": true, "delta_db": -0.8107675727407606}], "active_band_count": 3, "active_band_mae_db": 0.4528412813851261}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012969159209010741, "log_rms_mae_db": 0.555927640898717, "rms_envelope_nrmse": 0.040981551636164786, "envelope_correlation": 0.9984099397449955, "rms_gain_error_db": -0.08875541449813408, "reference_mean_power": 0.0012704534730935044, "output_mean_power": 0.0012447531195851822}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.90499086389905e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004798268169104879, "output_power": 0.000468154654395316, "reference_active": true, "delta_db": -0.10695170976473191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7762549535694356e-05, "output_power": 3.4172766243010574e-05, "reference_active": true, "delta_db": -0.4338117215421805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.708330714477498e-07, "output_power": 6.676746837241631e-07, "reference_active": true, "delta_db": -0.6239542977655509}], "active_band_count": 3, "active_band_mae_db": 0.3882392430241544}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.983632889928613e-05, "log_rms_mae_db": 1.0457241649859668, "rms_envelope_nrmse": 0.09456343359264802, "envelope_correlation": 0.9913398407174705, "rms_gain_error_db": -0.13226640024555666, "reference_mean_power": 0.00040681650183919396, "output_mean_power": 0.00039461348395123324}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8313223654566366e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016437493457690703, "output_power": 0.0001603157730415618, "reference_active": true, "delta_db": -0.10859339447051038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7703924674415462e-05, "output_power": 2.8010033347207355e-05, "reference_active": true, "delta_db": 0.04772328152983167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.687130809571553e-06, "output_power": 5.164912459140558e-06, "reference_active": true, "delta_db": 0.4215589091311914}], "active_band_count": 3, "active_band_mae_db": 0.19262519504384445}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.811887222434066e-05, "log_rms_mae_db": 0.6292964935637718, "rms_envelope_nrmse": 0.0510624008537289, "envelope_correlation": 0.9974438132397101, "rms_gain_error_db": -0.13116260765453577, "reference_mean_power": 0.00040698573595692556, "output_mean_power": 0.00039487799016393855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.593242154160197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015388869283670126, "output_power": 0.0001502269454952132, "reference_active": true, "delta_db": -0.10458873678084213}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4161267354242513e-05, "output_power": 1.4396403124298366e-05, "reference_active": true, "delta_db": 0.07151877130412776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7920568451213675e-07, "output_power": 3.4096546785576997e-07, "reference_active": true, "delta_db": 0.8678614084538351}], "active_band_count": 3, "active_band_mae_db": 0.34798963884626827}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4474245023095566e-05, "log_rms_mae_db": 1.5028682041929935, "rms_envelope_nrmse": 0.2516802732781047, "envelope_correlation": 0.9432518165020259, "rms_gain_error_db": -0.11851697480009166, "reference_mean_power": 1.6116938837494647e-05, "output_mean_power": 1.568306203470935e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.1253846819517355e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.226560638885928e-06, "output_power": 2.586122700030652e-06, "reference_active": true, "delta_db": 0.650145990292877}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.434590024496585e-06, "output_power": 1.295083609277998e-06, "reference_active": true, "delta_db": -0.44429999686998617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.807607987621042e-07, "output_power": 6.01861448885844e-07, "reference_active": true, "delta_db": 0.975674782273125}], "active_band_count": 3, "active_band_mae_db": 0.6900402564786626}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.920336454108668e-05, "log_rms_mae_db": 0.9986567672673398, "rms_envelope_nrmse": 0.18283197399825848, "envelope_correlation": 0.9672330488943574, "rms_gain_error_db": -0.11851678756297342, "reference_mean_power": 1.619697385840346e-05, "output_mean_power": 1.576094314860992e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.254945521069739e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4889493401597257e-06, "output_power": 2.910284819891888e-06, "reference_active": true, "delta_db": 0.6791943691216191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.64830026091652e-07, "output_power": 6.803442009641592e-07, "reference_active": true, "delta_db": -0.5083624200728133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8106205193126903e-08, "output_power": 4.870954170334182e-08, "reference_active": true, "delta_db": 0.05412944020971016}], "active_band_count": 3, "active_band_mae_db": 0.4138954098013808}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8954431034305158, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9c544e5ba44ebcc1d449b6f4e3c8abd2d4d4cfce538f15c289253a9241411924", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895436730720998, "standard_deviation": 0.0011211552043749692, "minimum": 0.9877733193919378, "maximum": 0.9913398407174705, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.055766440484616, "standard_deviation": 0.030715876630140245, "minimum": 1.021069034005683, "maximum": 1.1101449397552834, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3806878413582119, "standard_deviation": 0.1950136957023308, "minimum": 0.19262519504384445, "maximum": 0.7366398706214906, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10470630539309637, "standard_deviation": 0.00590292644458447, "minimum": 0.09456343359264802, "maximum": 0.11330614779442237, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996782263327342, "standard_deviation": 0.0004203722256316976, "minimum": 0.996107606473235, "maximum": 0.9974438132397101, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6157537129066566, "standard_deviation": 0.030314750889516847, "minimum": 0.5873725149989603, "maximum": 0.6788006371329027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4673714423122681, "standard_deviation": 0.15147025703256917, "minimum": 0.2879987293562168, "maximum": 0.700703379141932, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059610342330829666, "standard_deviation": 0.004980085760076823, "minimum": 0.0510624008537289, "maximum": 0.06566884295594112, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9439983062982054, "standard_deviation": 0.004145334327225924, "minimum": 0.9377803365091798, "maximum": 0.9500830828782765, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5114883884351125, "standard_deviation": 0.022001292677104, "minimum": 1.4883464003464382, "maximum": 1.5493841059810423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0294822949843394, "standard_deviation": 0.4236098528305257, "minimum": 0.4756600328566198, "maximum": 1.8704577815002026, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25636034812697195, "standard_deviation": 0.00949630105439089, "minimum": 0.24488234120907396, "maximum": 0.2742440119458416, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9681552335653552, "standard_deviation": 0.003346824908591223, "minimum": 0.9632494928321104, "maximum": 0.9749521530457135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0118038017446938, "standard_deviation": 0.02572800644912501, "minimum": 0.9789342111284656, "maximum": 1.0478407456372287, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8275327713757346, "standard_deviation": 0.49195321764916594, "minimum": 0.308669624593512, "maximum": 1.8088746274021315, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18518662273658398, "standard_deviation": 0.009786747519073878, "minimum": 0.16865782296073056, "maximum": 0.20247818094270073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949378575836179, "standard_deviation": 0.0007507232904168411, "minimum": 0.9933331601847651, "maximum": 0.9956576863050444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0066341945189072, "standard_deviation": 0.017667896370071, "minimum": 0.9838287024289726, "maximum": 1.0284966082728637, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.43886208722695824, "standard_deviation": 0.16927421651547386, "minimum": 0.18234523090419316, "maximum": 0.6219336623338002, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07438241630450931, "standard_deviation": 0.005029288209080278, "minimum": 0.069253196040876, "maximum": 0.08489365460954429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981772068313695, "standard_deviation": 0.0004117869170443002, "minimum": 0.9972123232491698, "maximum": 0.998558253271742, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5698498432975221, "standard_deviation": 0.026492875573485074, "minimum": 0.5170017961721316, "maximum": 0.5967567992256095, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45976121234825434, "standard_deviation": 0.18626107639800757, "minimum": 0.15845302168534262, "maximum": 0.6840534338931471, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04385641442887249, "standard_deviation": 0.004118099666678917, "minimum": 0.03970824356182062, "maximum": 0.05298666526611168, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15972833936564715, "standard_deviation": 0.05344701494068126, "minimum": 0.09868247477125985, "maximum": 0.21895157750597347, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.39119276458303, "standard_deviation": 0.010396823620716112, "minimum": 5.373074954257575, "maximum": 5.402825597892633, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.32107005613884365, "standard_deviation": 0.004047157392055036, "minimum": 0.3158543537370337, "maximum": 0.3277445370120416, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09723794889836519, "standard_deviation": 0.018848624818659152, "minimum": -0.12247305071488689, "maximum": -0.06792324243174949, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9959822451517899, "standard_deviation": 0.00055378257236083, "minimum": 0.9952122599401299, "maximum": 0.9967967270950876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5956024863131016, "standard_deviation": 0.01352124484079105, "minimum": 0.5813438027930531, "maximum": 0.6183631918823046, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29494985941753027, "standard_deviation": 0.1581186847260768, "minimum": 0.07215392260529617, "maximum": 0.5600482472297205, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0604857923342718, "standard_deviation": 0.003992906602514274, "minimum": 0.05431908671979257, "maximum": 0.06600987810032452, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986905212738805, "standard_deviation": 0.00027493631521220853, "minimum": 0.9982046034296276, "maximum": 0.9991198664983091, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3311977356796587, "standard_deviation": 0.01722007621345815, "minimum": 0.3154025083538866, "maximum": 0.3612803904569225, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3244848223930739, "standard_deviation": 0.17671100830482503, "minimum": 0.03616240011501376, "maximum": 0.6439025051930161, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03444310497462636, "standard_deviation": 0.00328724070448471, "minimum": 0.029046982093643994, "maximum": 0.0390793883688805, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0007", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.744166374206543, "predicted_mos": 2.744166374206543, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "41750e5427621f7b83c67d160b6a47c8ff8af3113dbfcdc744766abc7f7fb565", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.836048100143671, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5998489795795409, "全帯域_10ms_rms_envelope_nrmse": 0.057457615376175705, "全帯域_10ms_envelope_correlation": 0.9963570647183968, "全帯域_10ms_modulation_mae_db": 0.11089882701662773, "全帯域_40ms_log_rms_mae_db": 0.3269225080754121, "全帯域_40ms_rms_envelope_nrmse": 0.0325825035544438, "全帯域_40ms_envelope_correlation": 0.9988045893300052, "全帯域_40ms_modulation_mae_db": 0.3206228241798482, "300–1000 Hz_10ms_log_rms_mae_db": 1.0688761836341432, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07288094730894852, "300–1000 Hz_10ms_envelope_correlation": 0.9951118942795885, "300–1000 Hz_10ms_modulation_mae_db": 0.3453725341264173, "300–1000 Hz_40ms_log_rms_mae_db": 0.5990506164938947, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0393803400276235, "300–1000 Hz_40ms_envelope_correlation": 0.998503390764297, "300–1000 Hz_40ms_modulation_mae_db": 0.563996521484499, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0902529716790244, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10273700845489646, "1000–3000 Hz_10ms_envelope_correlation": 0.9900121927556493, "1000–3000 Hz_10ms_modulation_mae_db": 0.07756780680487692, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6042571851136054, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057332852935908035, "1000–3000 Hz_40ms_envelope_correlation": 0.9971267835930099, "1000–3000 Hz_40ms_modulation_mae_db": 0.08747122552258844, "3000–6000 Hz_10ms_log_rms_mae_db": 1.623543398293226, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26056330266295, "3000–6000 Hz_10ms_envelope_correlation": 0.9397077287721113, "3000–6000 Hz_10ms_modulation_mae_db": 0.56445517991146, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0372546115394223, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19198159168546247, "3000–6000 Hz_40ms_envelope_correlation": 0.9641512421179012, "3000–6000 Hz_40ms_modulation_mae_db": 0.45401315479951715, "mr_stft_db": 5.473983747303102, "band_mae_db": 0.1862540014819359, "rms_gain_error_db": -0.07840093198598651, "output_peak": 0.3240045070165769}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5878816828312414, "全帯域_10ms_rms_envelope_nrmse": 0.06145279726338691, "全帯域_10ms_envelope_correlation": 0.9958362375304095, "全帯域_10ms_modulation_mae_db": 0.13496138276230596, "全帯域_40ms_log_rms_mae_db": 0.3119013925776308, "全帯域_40ms_rms_envelope_nrmse": 0.03220572041148468, "全帯域_40ms_envelope_correlation": 0.9988542676844185, "全帯域_40ms_modulation_mae_db": 0.2852214955041173, "300–1000 Hz_10ms_log_rms_mae_db": 1.038150511373704, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07266303693057441, "300–1000 Hz_10ms_envelope_correlation": 0.9951763539300714, "300–1000 Hz_10ms_modulation_mae_db": 0.3137084001011475, "300–1000 Hz_40ms_log_rms_mae_db": 0.5649199361603225, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03880916967921058, "300–1000 Hz_40ms_envelope_correlation": 0.9985977043233205, "300–1000 Hz_40ms_modulation_mae_db": 0.47312990664539206, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0870520892639273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11745305228806582, "1000–3000 Hz_10ms_envelope_correlation": 0.9866690331710997, "1000–3000 Hz_10ms_modulation_mae_db": 0.46122215903818287, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6257548548001879, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06463107167241468, "1000–3000 Hz_40ms_envelope_correlation": 0.9960844526448142, "1000–3000 Hz_40ms_modulation_mae_db": 0.31440987593820485, "3000–6000 Hz_10ms_log_rms_mae_db": 1.656808848170496, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2757220136443973, "3000–6000 Hz_10ms_envelope_correlation": 0.9328815404856577, "3000–6000 Hz_10ms_modulation_mae_db": 0.463172020482995, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0511280373541336, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19624112754504097, "3000–6000 Hz_40ms_envelope_correlation": 0.9627001042959565, "3000–6000 Hz_40ms_modulation_mae_db": 0.5026969015893644, "mr_stft_db": 5.456295016549529, "band_mae_db": 0.14641658061238258, "rms_gain_error_db": -0.0892290804086879, "output_peak": 0.30993330294659427}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6027942537860922, "全帯域_10ms_rms_envelope_nrmse": 0.05718897440448797, "全帯域_10ms_envelope_correlation": 0.9964030777348585, "全帯域_10ms_modulation_mae_db": 0.3285534848619999, "全帯域_40ms_log_rms_mae_db": 0.33912364553615465, "全帯域_40ms_rms_envelope_nrmse": 0.03312096587051584, "全帯域_40ms_envelope_correlation": 0.998770808108258, "全帯域_40ms_modulation_mae_db": 0.44538490191803093, "300–1000 Hz_10ms_log_rms_mae_db": 1.011390979097712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0766523421472437, "300–1000 Hz_10ms_envelope_correlation": 0.9945957286764394, "300–1000 Hz_10ms_modulation_mae_db": 0.3509826957169075, "300–1000 Hz_40ms_log_rms_mae_db": 0.6049578884451798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045526417974021474, "300–1000 Hz_40ms_envelope_correlation": 0.9979900691562326, "300–1000 Hz_40ms_modulation_mae_db": 0.5206726059131498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1006895923422046, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09657382933818852, "1000–3000 Hz_10ms_envelope_correlation": 0.9910860410719385, "1000–3000 Hz_10ms_modulation_mae_db": 0.4454352229932457, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6404746658960022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054593679420401, "1000–3000 Hz_40ms_envelope_correlation": 0.9972482027431208, "1000–3000 Hz_40ms_modulation_mae_db": 0.4537698737371067, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5863968146118619, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2568865331900443, "3000–6000 Hz_10ms_envelope_correlation": 0.9447510727705001, "3000–6000 Hz_10ms_modulation_mae_db": 1.870966825235902, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0249672439618356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18966623302571287, "3000–6000 Hz_40ms_envelope_correlation": 0.9672889179133234, "3000–6000 Hz_40ms_modulation_mae_db": 1.7398118246026186, "mr_stft_db": 5.453755091740781, "band_mae_db": 0.11383528302719058, "rms_gain_error_db": -0.08529669244677598, "output_peak": 0.3128937731575474}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5782688025195183, "全帯域_10ms_rms_envelope_nrmse": 0.05787164383775368, "全帯域_10ms_envelope_correlation": 0.9964028296954607, "全帯域_10ms_modulation_mae_db": 0.2768710965762738, "全帯域_40ms_log_rms_mae_db": 0.3286504744739779, "全帯域_40ms_rms_envelope_nrmse": 0.033284077363984414, "全帯域_40ms_envelope_correlation": 0.9988672305629628, "全帯域_40ms_modulation_mae_db": 0.32628557273765685, "300–1000 Hz_10ms_log_rms_mae_db": 1.0577617259708774, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07460002994989057, "300–1000 Hz_10ms_envelope_correlation": 0.9949698666462364, "300–1000 Hz_10ms_modulation_mae_db": 0.41393912282104345, "300–1000 Hz_40ms_log_rms_mae_db": 0.6136452847724518, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04256250888955883, "300–1000 Hz_40ms_envelope_correlation": 0.9983628658627767, "300–1000 Hz_40ms_modulation_mae_db": 0.4278125212673025, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0900157704793045, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09984659669745201, "1000–3000 Hz_10ms_envelope_correlation": 0.9905741659931885, "1000–3000 Hz_10ms_modulation_mae_db": 0.4554445740023166, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6528519524489239, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0588348099664625, "1000–3000 Hz_40ms_envelope_correlation": 0.9969354186608753, "1000–3000 Hz_40ms_modulation_mae_db": 0.4284756916044367, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6324212525800874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2513949025028585, "3000–6000 Hz_10ms_envelope_correlation": 0.9478499743070158, "3000–6000 Hz_10ms_modulation_mae_db": 1.3933585447152979, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0287458317933649, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18583478291324365, "3000–6000 Hz_40ms_envelope_correlation": 0.9691996614936644, "3000–6000 Hz_40ms_modulation_mae_db": 1.4746912971639379, "mr_stft_db": 5.4475220747856214, "band_mae_db": 0.18657815414706289, "rms_gain_error_db": -0.11881549244691642, "output_peak": 0.3234742692676792}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5918601741039348, "全帯域_10ms_rms_envelope_nrmse": 0.05658059804596779, "全帯域_10ms_envelope_correlation": 0.996677343866671, "全帯域_10ms_modulation_mae_db": 0.7657245805530609, "全帯域_40ms_log_rms_mae_db": 0.34167586357605867, "全帯域_40ms_rms_envelope_nrmse": 0.03422541825559847, "全帯域_40ms_envelope_correlation": 0.9989026210970865, "全帯域_40ms_modulation_mae_db": 0.769977721354289, "300–1000 Hz_10ms_log_rms_mae_db": 1.0457382995962965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0812424162572254, "300–1000 Hz_10ms_envelope_correlation": 0.9941415439198532, "300–1000 Hz_10ms_modulation_mae_db": 0.7358986423484005, "300–1000 Hz_40ms_log_rms_mae_db": 0.6224049031401729, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05001442099990939, "300–1000 Hz_40ms_envelope_correlation": 0.9978424187350871, "300–1000 Hz_40ms_modulation_mae_db": 0.7883965893142054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1040332496008671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0912106837233661, "1000–3000 Hz_10ms_envelope_correlation": 0.9923456314798115, "1000–3000 Hz_10ms_modulation_mae_db": 0.6227029160479977, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6379410970462412, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054442090361897746, "1000–3000 Hz_40ms_envelope_correlation": 0.9975849993860935, "1000–3000 Hz_40ms_modulation_mae_db": 0.5833461002276671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6334725230531362, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2535276133670844, "3000–6000 Hz_10ms_envelope_correlation": 0.9435324738125144, "3000–6000 Hz_10ms_modulation_mae_db": 0.6524036197534552, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0116477354769131, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18749479050680556, "3000–6000 Hz_40ms_envelope_correlation": 0.9661139835519593, "3000–6000 Hz_40ms_modulation_mae_db": 0.45031029470917017, "mr_stft_db": 5.454664293550568, "band_mae_db": 0.17054348413730278, "rms_gain_error_db": -0.14732207528406469, "output_peak": 0.3218187069781079}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5526206803483086, "全帯域_10ms_rms_envelope_nrmse": 0.056968830722262, "全帯域_10ms_envelope_correlation": 0.9964219583876422, "全帯域_10ms_modulation_mae_db": 0.4292721471267897, "全帯域_40ms_log_rms_mae_db": 0.3024470843113217, "全帯域_40ms_rms_envelope_nrmse": 0.031306426209876594, "全帯域_40ms_envelope_correlation": 0.9989044885969812, "全帯域_40ms_modulation_mae_db": 0.6577929348365142, "300–1000 Hz_10ms_log_rms_mae_db": 0.9744022399945768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07524762518325723, "300–1000 Hz_10ms_envelope_correlation": 0.9947996165203216, "300–1000 Hz_10ms_modulation_mae_db": 0.2159405212200821, "300–1000 Hz_40ms_log_rms_mae_db": 0.5455439890982429, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044568173514078775, "300–1000 Hz_40ms_envelope_correlation": 0.9980830759469428, "300–1000 Hz_40ms_modulation_mae_db": 0.43129329135343575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.123914888752998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09831127236818586, "1000–3000 Hz_10ms_envelope_correlation": 0.9907451200528532, "1000–3000 Hz_10ms_modulation_mae_db": 0.5037463090724484, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6410725765894507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0521571815905647, "1000–3000 Hz_40ms_envelope_correlation": 0.9975343864567268, "1000–3000 Hz_40ms_modulation_mae_db": 0.548232890556215, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5853057541334477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25543866599675996, "3000–6000 Hz_10ms_envelope_correlation": 0.9442517603099019, "3000–6000 Hz_10ms_modulation_mae_db": 1.2097483810555871, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0068320559702157, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18053587595670434, "3000–6000 Hz_40ms_envelope_correlation": 0.9696318380776521, "3000–6000 Hz_40ms_modulation_mae_db": 0.8830424935589702, "mr_stft_db": 5.449200183814092, "band_mae_db": 0.1907932380331958, "rms_gain_error_db": -0.08097083593732908, "output_peak": 0.31876863655732923}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6199975490447863, "全帯域_10ms_rms_envelope_nrmse": 0.06521409189878542, "全帯域_10ms_envelope_correlation": 0.9952903164541398, "全帯域_10ms_modulation_mae_db": 0.22174006457677875, "全帯域_40ms_log_rms_mae_db": 0.35315221672278657, "全帯域_40ms_rms_envelope_nrmse": 0.03807772137127982, "全帯域_40ms_envelope_correlation": 0.9983409059479832, "全帯域_40ms_modulation_mae_db": 0.36725118479224045, "300–1000 Hz_10ms_log_rms_mae_db": 1.090927177277491, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07911835305566876, "300–1000 Hz_10ms_envelope_correlation": 0.9942405535922324, "300–1000 Hz_10ms_modulation_mae_db": 0.12807690626969923, "300–1000 Hz_40ms_log_rms_mae_db": 0.6166004705422371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0456541627344496, "300–1000 Hz_40ms_envelope_correlation": 0.9979832445220969, "300–1000 Hz_40ms_modulation_mae_db": 0.13647077285011708, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1393084728424225, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11285714011369759, "1000–3000 Hz_10ms_envelope_correlation": 0.9877200488402109, "1000–3000 Hz_10ms_modulation_mae_db": 0.5585874049443298, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6412237466116447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06536942379356378, "1000–3000 Hz_40ms_envelope_correlation": 0.9959104654102267, "1000–3000 Hz_40ms_modulation_mae_db": 0.4836220435419185, "3000–6000 Hz_10ms_log_rms_mae_db": 1.651434734342262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27397897672716615, "3000–6000 Hz_10ms_envelope_correlation": 0.9351040191704789, "3000–6000 Hz_10ms_modulation_mae_db": 0.7489109963272268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.053697276288687, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19730796155367994, "3000–6000 Hz_40ms_envelope_correlation": 0.9630934111512889, "3000–6000 Hz_40ms_modulation_mae_db": 0.4498439912286278, "mr_stft_db": 5.451910767898936, "band_mae_db": 0.1532008214475529, "rms_gain_error_db": -0.08633246624631484, "output_peak": 0.30985451898277633}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6030420640001426, "全帯域_10ms_rms_envelope_nrmse": 0.060423076332823566, "全帯域_10ms_envelope_correlation": 0.9959507215088715, "全帯域_10ms_modulation_mae_db": 0.29786361564418645, "全帯域_40ms_log_rms_mae_db": 0.3272544870000846, "全帯域_40ms_rms_envelope_nrmse": 0.03242237474335805, "全帯域_40ms_envelope_correlation": 0.9988015806375046, "全帯域_40ms_modulation_mae_db": 0.3768562559130147, "300–1000 Hz_10ms_log_rms_mae_db": 1.0488717969520331, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07509273755971964, "300–1000 Hz_10ms_envelope_correlation": 0.9948294431522384, "300–1000 Hz_10ms_modulation_mae_db": 0.5665994836757979, "300–1000 Hz_40ms_log_rms_mae_db": 0.5783875639773172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042156988786862276, "300–1000 Hz_40ms_envelope_correlation": 0.9983070207706991, "300–1000 Hz_40ms_modulation_mae_db": 0.6986780451151052, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1501082346250167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10528616873114333, "1000–3000 Hz_10ms_envelope_correlation": 0.9891547095678032, "1000–3000 Hz_10ms_modulation_mae_db": 0.15218576861591476, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6738024982961053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05540619009937157, "1000–3000 Hz_40ms_envelope_correlation": 0.9969060744311173, "1000–3000 Hz_40ms_modulation_mae_db": 0.3040136053411718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6301726484706216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2545020473995852, "3000–6000 Hz_10ms_envelope_correlation": 0.945117557574615, "3000–6000 Hz_10ms_modulation_mae_db": 1.1103510991895291, "3000–6000 Hz_40ms_log_rms_mae_db": 1.00209491965941, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18323689634262785, "3000–6000 Hz_40ms_envelope_correlation": 0.969056727717429, "3000–6000 Hz_40ms_modulation_mae_db": 0.9826777973268745, "mr_stft_db": 5.445121428509509, "band_mae_db": 0.10242336277712792, "rms_gain_error_db": -0.07386494919842183, "output_peak": 0.3281189995869226}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.158712445959529, "spectral_convergence": 0.20982438053505484, "spectral_convergence_status": "ok", "amplitude_floor": 1.0845908458793198e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.267313981677056, "spectral_convergence": 0.20327886023876762, "spectral_convergence_status": "ok", "amplitude_floor": 9.938577161724507e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.995924814272718, "spectral_convergence": 0.21429830227954474, "spectral_convergence_status": "ok", "amplitude_floor": 7.780261052298239e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.1610975793667e-05, "output_power": 8.317493445625315e-05, "delta_db": 0.08243896714037716, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005078113602376632, "output_power": 0.0005089884823197461, "delta_db": 0.01005542746483369, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000931370997501075, "output_power": 0.0009136533469938374, "delta_db": -0.08341260546222744, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000354698831360673, "output_power": 0.00033924898443862746, "delta_db": -0.19341201340068562, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.667116250090391e-05, "output_power": 9.005873137783903e-05, "delta_db": -0.30771116217670114, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.4087248189368409e-05, "output_power": 1.5591056644610874e-05, "delta_db": 0.4404938332467905, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1862540014819359, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07840093198598651, "pitch": null, "output_rms_dbfs": -27.09806057131615, "output_peak": 0.3240045070165769, "reference_rms_dbfs": -27.01965963933016, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.473983747303102}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001511941088699898, "log_rms_mae_db": 0.5998489795795409, "rms_envelope_nrmse": 0.057457615376175705, "envelope_correlation": 0.9963570647183968, "rms_gain_error_db": -0.07840316094556406, "reference_mean_power": 0.0019872702200913737, "output_mean_power": 0.0019517159370839258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.92641346613152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000630498462430925, "output_power": 0.0006259514124866816, "reference_active": true, "delta_db": -0.031434080424498106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.948331074534803e-05, "output_power": 7.47818328024574e-05, "reference_active": true, "delta_db": -0.26479843706046996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3123506088131581e-05, "output_power": 1.3013780601119722e-05, "reference_active": true, "delta_db": -0.03646396356491514}], "active_band_count": 3, "active_band_mae_db": 0.11089882701662773}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013307215895407113, "log_rms_mae_db": 0.3269225080754121, "rms_envelope_nrmse": 0.0325825035544438, "envelope_correlation": 0.9988045893300052, "rms_gain_error_db": -0.07900295199130483, "reference_mean_power": 0.0019932283004473585, "output_mean_power": 0.001957297086249473}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.959638607230672e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005794424432114469, "output_power": 0.000576810586781457, "reference_active": true, "delta_db": -0.019770805680907295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.108494970400102e-05, "output_power": 3.8831962310233026e-05, "reference_active": true, "delta_db": -0.24493422180909352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.934091733283841e-07, "output_power": 6.757419378836897e-07, "reference_active": true, "delta_db": -0.6971634450495439}], "active_band_count": 3, "active_band_mae_db": 0.3206228241798482}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014956359127228178, "log_rms_mae_db": 1.0688761836341432, "rms_envelope_nrmse": 0.07288094730894852, "envelope_correlation": 0.9951118942795885, "rms_gain_error_db": -0.07075187196705386, "reference_mean_power": 0.001265981682968443, "output_mean_power": 0.0012455243857569182}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.65319874591582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005127917421071942, "output_power": 0.0005061472551298819, "reference_active": true, "delta_db": -0.056641367948543936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.49871434628785e-05, "output_power": 6.938837850430987e-05, "reference_active": true, "delta_db": -0.3370007119528395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1228212970044594e-05, "output_power": 9.684185265794851e-06, "reference_active": true, "delta_db": -0.6424755224778683}], "active_band_count": 3, "active_band_mae_db": 0.3453725341264173}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012969159209010741, "log_rms_mae_db": 0.5990506164938947, "rms_envelope_nrmse": 0.0393803400276235, "envelope_correlation": 0.998503390764297, "rms_gain_error_db": -0.07082345908155321, "reference_mean_power": 0.0012704534730935044, "output_mean_power": 0.001249903312164622}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.90499086389905e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004798268169104879, "output_power": 0.0004726542339883456, "reference_active": true, "delta_db": -0.06540963863595244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7762549535694356e-05, "output_power": 3.523710891748888e-05, "reference_active": true, "delta_db": -0.300610391287206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.708330714477498e-07, "output_power": 5.680196837315035e-07, "reference_active": true, "delta_db": -1.3259695345303384}], "active_band_count": 3, "active_band_mae_db": 0.563996521484499}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.983632889928613e-05, "log_rms_mae_db": 1.0902529716790244, "rms_envelope_nrmse": 0.10273700845489646, "envelope_correlation": 0.9900121927556493, "rms_gain_error_db": -0.21507962546854612, "reference_mean_power": 0.00040681650183919396, "output_mean_power": 0.00038716010458076527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8313223654566366e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016437493457690703, "output_power": 0.00015979911253239884, "reference_active": true, "delta_db": -0.1226122991259343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7703924674415462e-05, "output_power": 2.7282509295751206e-05, "reference_active": true, "delta_db": -0.06656985869204018}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.687130809571553e-06, "output_power": 4.734337345149758e-06, "reference_active": true, "delta_db": 0.043521262596656264}], "active_band_count": 3, "active_band_mae_db": 0.07756780680487692}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.811887222434066e-05, "log_rms_mae_db": 0.6042571851136054, "rms_envelope_nrmse": 0.057332852935908035, "envelope_correlation": 0.9971267835930099, "rms_gain_error_db": -0.21811456829763215, "reference_mean_power": 0.00040698573595692556, "output_mean_power": 0.00038705058794338566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.593242154160197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015388869283670126, "output_power": 0.00015093334982462057, "reference_active": true, "delta_db": -0.08421499763528517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4161267354242513e-05, "output_power": 1.383642681694154e-05, "reference_active": true, "delta_db": -0.10078171642877372}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7920568451213675e-07, "output_power": 2.742726861154097e-07, "reference_active": true, "delta_db": -0.07741696250370643}], "active_band_count": 3, "active_band_mae_db": 0.08747122552258844}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4474245023095566e-05, "log_rms_mae_db": 1.623543398293226, "rms_envelope_nrmse": 0.26056330266295, "envelope_correlation": 0.9397077287721113, "rms_gain_error_db": -0.05157130652966919, "reference_mean_power": 1.6116938837494647e-05, "output_mean_power": 1.5926686339828642e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.1253846819517355e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.226560638885928e-06, "output_power": 2.5828065893922526e-06, "reference_active": true, "delta_db": 0.6445735838288646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.434590024496585e-06, "output_power": 1.2769757070734323e-06, "reference_active": true, "delta_db": -0.5054517120655241}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.807607987621042e-07, "output_power": 5.448325834432836e-07, "reference_active": true, "delta_db": 0.5433402438399911}], "active_band_count": 3, "active_band_mae_db": 0.56445517991146}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.920336454108668e-05, "log_rms_mae_db": 1.0372546115394223, "rms_envelope_nrmse": 0.19198159168546247, "envelope_correlation": 0.9641512421179012, "rms_gain_error_db": -0.05155577581585646, "reference_mean_power": 1.619697385840346e-05, "output_mean_power": 1.600583382489457e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.254945521069739e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4889493401597257e-06, "output_power": 2.885710322829124e-06, "reference_active": true, "delta_db": 0.6423667591571673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.64830026091652e-07, "output_power": 6.877504516353707e-07, "reference_active": true, "delta_db": -0.46134045103034754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8106205193126903e-08, "output_power": 4.532813743644996e-08, "reference_active": true, "delta_db": -0.25833225421103645}], "active_band_count": 3, "active_band_mae_db": 0.45401315479951715}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8931926134520461, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "88d983320e2bed88e04623e3af8d10152089eadf0e2b0d462d2bbfb1bd1baa4f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897883678665693, "standard_deviation": 0.0018599623396735202, "minimum": 0.9866690331710997, "maximum": 0.9923456314798115, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1106719086982206, "standard_deviation": 0.02418936150159716, "minimum": 1.0870520892639273, "maximum": 1.1501082346250167, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4096115201899141, "standard_deviation": 0.19236565828935132, "minimum": 0.07756780680487692, "maximum": 0.6227029160479977, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10303446896437446, "standard_deviation": 0.00864830227473027, "minimum": 0.0912106837233661, "maximum": 0.11745305228806582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996916347915748, "standard_deviation": 0.000619653799906245, "minimum": 0.9959104654102267, "maximum": 0.9975849993860935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6396723221002703, "standard_deviation": 0.019980375849182646, "minimum": 0.6042571851136054, "maximum": 0.6738024982961053, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4004176633086637, "standard_deviation": 0.16057725752656873, "minimum": 0.08747122552258844, "maximum": 0.5833461002276671, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.057845912480073, "standard_deviation": 0.004845611938339275, "minimum": 0.0521571815905647, "maximum": 0.06536942379356378, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9416495159003494, "standard_deviation": 0.005263096814570939, "minimum": 0.9328815404856577, "maximum": 0.9478499743070158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6249444967068924, "standard_deviation": 0.0265441654908965, "minimum": 1.5853057541334477, "maximum": 1.656808848170496, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0016708333339317, "standard_deviation": 0.48273270799269696, "minimum": 0.463172020482995, "maximum": 1.870966825235902, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2602517569363557, "standard_deviation": 0.009404179760288971, "minimum": 0.2513949025028585, "maximum": 0.2757220136443973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9664044857898968, "standard_deviation": 0.0028257128945766053, "minimum": 0.9627001042959565, "maximum": 0.9696318380776521, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0270459640054979, "standard_deviation": 0.01952993288340087, "minimum": 1.00209491965941, "maximum": 1.053697276288687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.867135969372385, "standard_deviation": 0.5064797398679894, "minimum": 0.4498439912286278, "maximum": 1.7398118246026186, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1890374074411597, "standard_deviation": 0.005951203102875447, "minimum": 0.18053587595670434, "maximum": 0.19730796155367994, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9947331250896226, "standard_deviation": 0.0003819378190352735, "minimum": 0.9941415439198532, "maximum": 0.9951763539300714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.042014864237104, "standard_deviation": 0.03575114887134798, "minimum": 0.9744022399945768, "maximum": 1.090927177277491, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.383814788284937, "standard_deviation": 0.19247821895481254, "minimum": 0.12807690626969923, "maximum": 0.7358986423484005, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07593718604906602, "standard_deviation": 0.0029716268004799833, "minimum": 0.07266303693057441, "maximum": 0.0812424162572254, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982087237601815, "standard_deviation": 0.0002725996747980152, "minimum": 0.9978424187350871, "maximum": 0.9985977043233205, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5931888315787274, "standard_deviation": 0.027467540223224517, "minimum": 0.5455439890982429, "maximum": 0.6224049031401729, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.505056281742901, "standard_deviation": 0.19631282858291613, "minimum": 0.13647077285011708, "maximum": 0.7883965893142054, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0435840228257143, "standard_deviation": 0.0036630791091151085, "minimum": 0.03880916967921058, "maximum": 0.05001442099990939, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15625561570796892, "standard_deviation": 0.033848157760613465, "minimum": 0.10242336277712792, "maximum": 0.1907932380331958, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.454056575519017, "standard_deviation": 0.008890513321868382, "minimum": 5.445121428509509, "maximum": 5.473983747303102, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3186083393116917, "standard_deviation": 0.006950052491929141, "minimum": 0.30985451898277633, "maximum": 0.3281189995869226, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09502906549431216, "standard_deviation": 0.025146403297166935, "minimum": -0.14732207528406469, "maximum": -0.07386494919842183, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9961674437370562, "standard_deviation": 0.00044684293219843555, "minimum": 0.9952903164541398, "maximum": 0.996677343866671, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5920392732766957, "standard_deviation": 0.020142906139017053, "minimum": 0.5526206803483086, "maximum": 0.6199975490447863, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3207356498897529, "standard_deviation": 0.2072429091148218, "minimum": 0.11089882701662773, "maximum": 0.7657245805530609, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05914470348520538, "standard_deviation": 0.003009506179618107, "minimum": 0.05658059804596779, "maximum": 0.06521409189878542, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.99878081149565, "standard_deviation": 0.00018424028276247065, "minimum": 0.9983409059479832, "maximum": 0.9989044885969812, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3288909590341784, "standard_deviation": 0.01626259877881272, "minimum": 0.3024470843113217, "maximum": 0.35315221672278657, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44367411140446394, "standard_deviation": 0.17592930906900162, "minimum": 0.2852214955041173, "maximum": 0.769977721354289, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033403150972567706, "standard_deviation": 0.0020731969008925483, "minimum": 0.031306426209876594, "maximum": 0.03807772137127982, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0007", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.59824275970459, "predicted_mos": 2.59824275970459, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f063cce2b9994846560e210c69c5c4ba1dd2ef6a5dcd11a0586b393c429f67dd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.73776250006631, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5743346283713556, "全帯域_10ms_rms_envelope_nrmse": 0.05762923026748647, "全帯域_10ms_envelope_correlation": 0.9963307514845517, "全帯域_10ms_modulation_mae_db": 0.16725412864003145, "全帯域_40ms_log_rms_mae_db": 0.32240295648993283, "全帯域_40ms_rms_envelope_nrmse": 0.0320116530915173, "全帯域_40ms_envelope_correlation": 0.9988434736339786, "全帯域_40ms_modulation_mae_db": 0.2543279855930918, "300–1000 Hz_10ms_log_rms_mae_db": 0.9922038999780942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07450288614873297, "300–1000 Hz_10ms_envelope_correlation": 0.9949025051637034, "300–1000 Hz_10ms_modulation_mae_db": 0.3096296621323837, "300–1000 Hz_40ms_log_rms_mae_db": 0.5356739274861236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042780007112213685, "300–1000 Hz_40ms_envelope_correlation": 0.9982405102027629, "300–1000 Hz_40ms_modulation_mae_db": 0.4576501691607411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.100207615041316, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10055226387641268, "1000–3000 Hz_10ms_envelope_correlation": 0.9902479169761258, "1000–3000 Hz_10ms_modulation_mae_db": 0.29442467434028324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6437002023919715, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060766510935664333, "1000–3000 Hz_40ms_envelope_correlation": 0.9964317196552944, "1000–3000 Hz_40ms_modulation_mae_db": 0.244258283116818, "3000–6000 Hz_10ms_log_rms_mae_db": 1.669933178573462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26680554510008747, "3000–6000 Hz_10ms_envelope_correlation": 0.9410964649893946, "3000–6000 Hz_10ms_modulation_mae_db": 0.6612897534743859, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0884631988352065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19544054626641238, "3000–6000 Hz_40ms_envelope_correlation": 0.9658944928900115, "3000–6000 Hz_40ms_modulation_mae_db": 0.8514324273861753, "mr_stft_db": 5.45356079134915, "band_mae_db": 0.1639567345569313, "rms_gain_error_db": -0.0769221414934308, "output_peak": 0.316236971300656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5583535095039598, "全帯域_10ms_rms_envelope_nrmse": 0.06014059561573567, "全帯域_10ms_envelope_correlation": 0.9959992806435621, "全帯域_10ms_modulation_mae_db": 0.10502287364977018, "全帯域_40ms_log_rms_mae_db": 0.3179962844472152, "全帯域_40ms_rms_envelope_nrmse": 0.03683990315024952, "全帯域_40ms_envelope_correlation": 0.9984384133571593, "全帯域_40ms_modulation_mae_db": 0.1334276662268224, "300–1000 Hz_10ms_log_rms_mae_db": 1.023253022624351, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07944063204859127, "300–1000 Hz_10ms_envelope_correlation": 0.9941625926228892, "300–1000 Hz_10ms_modulation_mae_db": 0.24505951020171982, "300–1000 Hz_40ms_log_rms_mae_db": 0.5947327638839836, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05121451998934017, "300–1000 Hz_40ms_envelope_correlation": 0.9973926786934006, "300–1000 Hz_40ms_modulation_mae_db": 0.13761863050709622, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1071854961917797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09938141764435068, "1000–3000 Hz_10ms_envelope_correlation": 0.9908948619427875, "1000–3000 Hz_10ms_modulation_mae_db": 0.6127474886876599, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6337483001913593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05833420167902842, "1000–3000 Hz_40ms_envelope_correlation": 0.9972623865538847, "1000–3000 Hz_40ms_modulation_mae_db": 0.6688154695991866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6013071886664112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2518768984885068, "3000–6000 Hz_10ms_envelope_correlation": 0.9455100027341132, "3000–6000 Hz_10ms_modulation_mae_db": 1.6320366204471932, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0374144840505235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18382616506160898, "3000–6000 Hz_40ms_envelope_correlation": 0.9681484522159666, "3000–6000 Hz_40ms_modulation_mae_db": 1.621494038661078, "mr_stft_db": 5.445998397014914, "band_mae_db": 0.14061748957396178, "rms_gain_error_db": -0.08010287333562016, "output_peak": 0.3368034709612734}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5489206988233224, "全帯域_10ms_rms_envelope_nrmse": 0.05003768062499307, "全帯域_10ms_envelope_correlation": 0.9972362782505667, "全帯域_10ms_modulation_mae_db": 0.31252968933108716, "全帯域_40ms_log_rms_mae_db": 0.2935324482688658, "全帯域_40ms_rms_envelope_nrmse": 0.027234435518149053, "全帯域_40ms_envelope_correlation": 0.9991634312160259, "全帯域_40ms_modulation_mae_db": 0.2796605320976768, "300–1000 Hz_10ms_log_rms_mae_db": 0.9836968697944006, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06860988814752969, "300–1000 Hz_10ms_envelope_correlation": 0.9956666942527799, "300–1000 Hz_10ms_modulation_mae_db": 0.421481248935895, "300–1000 Hz_40ms_log_rms_mae_db": 0.527917824605543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03756436734049417, "300–1000 Hz_40ms_envelope_correlation": 0.9986307561326491, "300–1000 Hz_40ms_modulation_mae_db": 0.47395320721697837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1106336102039074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09118750799358068, "1000–3000 Hz_10ms_envelope_correlation": 0.9921053116842964, "1000–3000 Hz_10ms_modulation_mae_db": 0.5177753865848121, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6198641137507094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04995170233272323, "1000–3000 Hz_40ms_envelope_correlation": 0.9977960483159486, "1000–3000 Hz_40ms_modulation_mae_db": 0.5496627226937386, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6230873320491255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2658825522882939, "3000–6000 Hz_10ms_envelope_correlation": 0.9398716245405915, "3000–6000 Hz_10ms_modulation_mae_db": 2.072158608090699, "3000–6000 Hz_40ms_log_rms_mae_db": 1.054763740885034, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2031168515415913, "3000–6000 Hz_40ms_envelope_correlation": 0.9616284826385765, "3000–6000 Hz_40ms_modulation_mae_db": 2.17759703553209, "mr_stft_db": 5.448316773690476, "band_mae_db": 0.08389564732917217, "rms_gain_error_db": -0.06711439030647603, "output_peak": 0.3292089250205143}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.563888874917064, "全帯域_10ms_rms_envelope_nrmse": 0.05669527547898273, "全帯域_10ms_envelope_correlation": 0.9964819772303322, "全帯域_10ms_modulation_mae_db": 0.37616071674893653, "全帯域_40ms_log_rms_mae_db": 0.3113543342028758, "全帯域_40ms_rms_envelope_nrmse": 0.029839244359633487, "全帯域_40ms_envelope_correlation": 0.9990439453828296, "全帯域_40ms_modulation_mae_db": 0.4078948250264296, "300–1000 Hz_10ms_log_rms_mae_db": 1.0313926589422535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0731870839326559, "300–1000 Hz_10ms_envelope_correlation": 0.995117312122355, "300–1000 Hz_10ms_modulation_mae_db": 0.7418514696534696, "300–1000 Hz_40ms_log_rms_mae_db": 0.5884229919332571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03895754849075941, "300–1000 Hz_40ms_envelope_correlation": 0.9985975068403147, "300–1000 Hz_40ms_modulation_mae_db": 0.756796399477523, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1631693104626837, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10390382880894633, "1000–3000 Hz_10ms_envelope_correlation": 0.9896567442391807, "1000–3000 Hz_10ms_modulation_mae_db": 0.5889953038713333, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6606840048790815, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05981038869447549, "1000–3000 Hz_40ms_envelope_correlation": 0.996677858719264, "1000–3000 Hz_40ms_modulation_mae_db": 0.3793531161448303, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5937116820889439, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2651834342892944, "3000–6000 Hz_10ms_envelope_correlation": 0.9456378995240137, "3000–6000 Hz_10ms_modulation_mae_db": 1.5726358913177985, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0323809054371444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1938077238177875, "3000–6000 Hz_40ms_envelope_correlation": 0.969697740344506, "3000–6000 Hz_40ms_modulation_mae_db": 1.5079437787232213, "mr_stft_db": 5.446488835769446, "band_mae_db": 0.19262319421194798, "rms_gain_error_db": -0.09152021827914006, "output_peak": 0.32862875129803026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6010990783946321, "全帯域_10ms_rms_envelope_nrmse": 0.06360668629211708, "全帯域_10ms_envelope_correlation": 0.9957057401083546, "全帯域_10ms_modulation_mae_db": 0.3563123611206704, "全帯域_40ms_log_rms_mae_db": 0.35943501745162126, "全帯域_40ms_rms_envelope_nrmse": 0.03661260352104948, "全帯域_40ms_envelope_correlation": 0.9986905677701892, "全帯域_40ms_modulation_mae_db": 0.40007324950401685, "300–1000 Hz_10ms_log_rms_mae_db": 1.008084082073538, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08334286302022978, "300–1000 Hz_10ms_envelope_correlation": 0.993912177086026, "300–1000 Hz_10ms_modulation_mae_db": 0.40572970639650596, "300–1000 Hz_40ms_log_rms_mae_db": 0.5751726661736366, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049810003696053345, "300–1000 Hz_40ms_envelope_correlation": 0.997981422671096, "300–1000 Hz_40ms_modulation_mae_db": 0.6825483437903088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.134450645767968, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1051835351100704, "1000–3000 Hz_10ms_envelope_correlation": 0.9892788125565105, "1000–3000 Hz_10ms_modulation_mae_db": 0.672001153929739, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6415792050334295, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056372247549064426, "1000–3000 Hz_40ms_envelope_correlation": 0.996954976837437, "1000–3000 Hz_40ms_modulation_mae_db": 0.4134545693792584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.641445349040652, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26306733846597885, "3000–6000 Hz_10ms_envelope_correlation": 0.9419024581140922, "3000–6000 Hz_10ms_modulation_mae_db": 1.0982756632165884, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0670936206268729, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18551431267325072, "3000–6000 Hz_40ms_envelope_correlation": 0.9686806824255972, "3000–6000 Hz_40ms_modulation_mae_db": 0.9733543862382193, "mr_stft_db": 5.4443442599124054, "band_mae_db": 0.1223637327295718, "rms_gain_error_db": -0.1468240702209917, "output_peak": 0.3210546509346415}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5633085855802249, "全帯域_10ms_rms_envelope_nrmse": 0.05574187653972411, "全帯域_10ms_envelope_correlation": 0.9966787810509717, "全帯域_10ms_modulation_mae_db": 0.5888773065478672, "全帯域_40ms_log_rms_mae_db": 0.30217747832801445, "全帯域_40ms_rms_envelope_nrmse": 0.029840307438841272, "全帯域_40ms_envelope_correlation": 0.9991279382740561, "全帯域_40ms_modulation_mae_db": 0.7114267374574063, "300–1000 Hz_10ms_log_rms_mae_db": 1.0611924239742503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07624979364593094, "300–1000 Hz_10ms_envelope_correlation": 0.9947688528973597, "300–1000 Hz_10ms_modulation_mae_db": 0.7161479294962249, "300–1000 Hz_40ms_log_rms_mae_db": 0.5434446379498121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04133552792651304, "300–1000 Hz_40ms_envelope_correlation": 0.9985098241369379, "300–1000 Hz_40ms_modulation_mae_db": 0.7863183307256781, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0697071176608641, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09607576924463487, "1000–3000 Hz_10ms_envelope_correlation": 0.9911817126056435, "1000–3000 Hz_10ms_modulation_mae_db": 0.5958202230858259, "1000–3000 Hz_40ms_log_rms_mae_db": 0.59540490567312, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.051105605926855126, "1000–3000 Hz_40ms_envelope_correlation": 0.9976750513710605, "1000–3000 Hz_40ms_modulation_mae_db": 0.6578976370069651, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6003760799046307, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2697174054600152, "3000–6000 Hz_10ms_envelope_correlation": 0.9361577339066878, "3000–6000 Hz_10ms_modulation_mae_db": 0.5204209847380575, "3000–6000 Hz_40ms_log_rms_mae_db": 1.012696784073019, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20073510718320406, "3000–6000 Hz_40ms_envelope_correlation": 0.961181309105124, "3000–6000 Hz_40ms_modulation_mae_db": 0.46447556172877347, "mr_stft_db": 5.445914594089333, "band_mae_db": 0.11134293997845507, "rms_gain_error_db": -0.11723726569384114, "output_peak": 0.3211620752508}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5775663621970876, "全帯域_10ms_rms_envelope_nrmse": 0.056493646745616834, "全帯域_10ms_envelope_correlation": 0.996499287791836, "全帯域_10ms_modulation_mae_db": 0.5171228494566068, "全帯域_40ms_log_rms_mae_db": 0.3018355836461175, "全帯域_40ms_rms_envelope_nrmse": 0.029552469817447562, "全帯域_40ms_envelope_correlation": 0.9990497230381947, "全帯域_40ms_modulation_mae_db": 0.551943360364387, "300–1000 Hz_10ms_log_rms_mae_db": 1.0670917229356423, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07464490173418885, "300–1000 Hz_10ms_envelope_correlation": 0.9948802411782851, "300–1000 Hz_10ms_modulation_mae_db": 0.5679019841160124, "300–1000 Hz_40ms_log_rms_mae_db": 0.5683520924932896, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0396478152100876, "300–1000 Hz_40ms_envelope_correlation": 0.9984879454178801, "300–1000 Hz_40ms_modulation_mae_db": 0.562938320603668, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1324510784785726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10426145467910956, "1000–3000 Hz_10ms_envelope_correlation": 0.9898496550611318, "1000–3000 Hz_10ms_modulation_mae_db": 0.5465167400406264, "1000–3000 Hz_40ms_log_rms_mae_db": 0.657974290133718, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062444048751939506, "1000–3000 Hz_40ms_envelope_correlation": 0.9966783063791694, "1000–3000 Hz_40ms_modulation_mae_db": 0.5962032941669896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6621069130879984, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27131289684342025, "3000–6000 Hz_10ms_envelope_correlation": 0.9363352293514584, "3000–6000 Hz_10ms_modulation_mae_db": 0.6387178650769206, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0647971731224026, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19419496033265607, "3000–6000 Hz_40ms_envelope_correlation": 0.9643105180643405, "3000–6000 Hz_40ms_modulation_mae_db": 0.5422961473281559, "mr_stft_db": 5.44212943568477, "band_mae_db": 0.16242223603341355, "rms_gain_error_db": -0.08555394000429392, "output_peak": 0.31583052496820074}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5732286362777411, "全帯域_10ms_rms_envelope_nrmse": 0.05416808269014812, "全帯域_10ms_envelope_correlation": 0.9967755874050866, "全帯域_10ms_modulation_mae_db": 0.24727313419202468, "全帯域_40ms_log_rms_mae_db": 0.29231768896210103, "全帯域_40ms_rms_envelope_nrmse": 0.027248667577073303, "全帯域_40ms_envelope_correlation": 0.9991952318623225, "全帯域_40ms_modulation_mae_db": 0.261080602765425, "300–1000 Hz_10ms_log_rms_mae_db": 0.9612720542452955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06771960702454867, "300–1000 Hz_10ms_envelope_correlation": 0.9958060027552899, "300–1000 Hz_10ms_modulation_mae_db": 0.4034445059301382, "300–1000 Hz_40ms_log_rms_mae_db": 0.5102952369260266, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03495827226625857, "300–1000 Hz_40ms_envelope_correlation": 0.9988546661867311, "300–1000 Hz_40ms_modulation_mae_db": 0.4461550920044474, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0998384996229322, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1190281932616702, "1000–3000 Hz_10ms_envelope_correlation": 0.986171109409872, "1000–3000 Hz_10ms_modulation_mae_db": 0.41449716739583, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6021907291291829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06737501902326981, "1000–3000 Hz_40ms_envelope_correlation": 0.9954843998237998, "1000–3000 Hz_40ms_modulation_mae_db": 0.480942552465017, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6162538772117676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2628159070054391, "3000–6000 Hz_10ms_envelope_correlation": 0.9406585112585075, "3000–6000 Hz_10ms_modulation_mae_db": 0.7676971501313417, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0404980946653617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19047650218026457, "3000–6000 Hz_40ms_envelope_correlation": 0.9659507804059041, "3000–6000 Hz_40ms_modulation_mae_db": 0.5418970706721377, "mr_stft_db": 5.438556442053479, "band_mae_db": 0.11972544365898075, "rms_gain_error_db": -0.0808677312687538, "output_peak": 0.3223694676515195}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.132311750122033, "spectral_convergence": 0.2061470792289983, "spectral_convergence_status": "ok", "amplitude_floor": 1.0845908458793198e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.247591177806376, "spectral_convergence": 0.20650460439348658, "spectral_convergence_status": "ok", "amplitude_floor": 9.938577161724507e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.98077944611904, "spectral_convergence": 0.21505023855325767, "spectral_convergence_status": "ok", "amplitude_floor": 7.780261052298239e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.1610975793667e-05, "output_power": 8.346470146342512e-05, "delta_db": 0.09754273742854788, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005078113602376632, "output_power": 0.0005123345767263853, "delta_db": 0.03851254538203733, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000931370997501075, "output_power": 0.0009071974750880338, "delta_db": -0.1142087723866119, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000354698831360673, "output_power": 0.0003413334360994407, "delta_db": -0.16680924913956158, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.667116250090391e-05, "output_power": 9.180669262882916e-05, "delta_db": -0.2242259910199559, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.4087248189368409e-05, "output_power": 1.5242993950640517e-05, "delta_db": 0.34244111198487315, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1639567345569313, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0769221414934308, "pitch": null, "output_rms_dbfs": -27.09658178082359, "output_peak": 0.316236971300656, "reference_rms_dbfs": -27.01965963933016, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.45356079134915}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001511941088699898, "log_rms_mae_db": 0.5743346283713556, "rms_envelope_nrmse": 0.05762923026748647, "envelope_correlation": 0.9963307514845517, "rms_gain_error_db": -0.0774901429275479, "reference_mean_power": 0.0019872702200913737, "output_mean_power": 0.0019521262897856083}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.92641346613152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000630498462430925, "output_power": 0.0006114343643559404, "reference_active": true, "delta_db": -0.13334188150290102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.948331074534803e-05, "output_power": 7.595815163406928e-05, "reference_active": true, "delta_db": -0.19701560560050482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3123506088131581e-05, "output_power": 1.2615642896481956e-05, "reference_active": true, "delta_db": -0.17140489881668858}], "active_band_count": 3, "active_band_mae_db": 0.16725412864003145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013307215895407113, "log_rms_mae_db": 0.32240295648993283, "rms_envelope_nrmse": 0.0320116530915173, "envelope_correlation": 0.9988434736339786, "rms_gain_error_db": -0.07776237437775885, "reference_mean_power": 0.0019932283004473585, "output_mean_power": 0.0019578562749777324}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.959638607230672e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005794424432114469, "output_power": 0.0005648620742094615, "reference_active": true, "delta_db": -0.11067886940907663}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.108494970400102e-05, "output_power": 3.903380137156542e-05, "reference_active": true, "delta_db": -0.22241911818596005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.934091733283841e-07, "output_power": 7.186354227856519e-07, "reference_active": true, "delta_db": -0.4298859691842386}], "active_band_count": 3, "active_band_mae_db": 0.2543279855930918}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014956359127228178, "log_rms_mae_db": 0.9922038999780942, "rms_envelope_nrmse": 0.07450288614873297, "envelope_correlation": 0.9949025051637034, "rms_gain_error_db": -0.08130928908952664, "reference_mean_power": 0.001265981682968443, "output_mean_power": 0.0012425002739729829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.65319874591582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005127917421071942, "output_power": 0.0004972336198577706, "reference_active": true, "delta_db": -0.13380537512568194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.49871434628785e-05, "output_power": 6.932969494488713e-05, "reference_active": true, "delta_db": -0.34067520755710956}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1228212970044594e-05, "output_power": 1.0112762342341947e-05, "reference_active": true, "delta_db": -0.45440840371435964}], "active_band_count": 3, "active_band_mae_db": 0.3096296621323837}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012969159209010741, "log_rms_mae_db": 0.5356739274861236, "rms_envelope_nrmse": 0.042780007112213685, "envelope_correlation": 0.9982405102027629, "rms_gain_error_db": -0.08152482721417059, "reference_mean_power": 0.0012704534730935044, "output_mean_power": 0.0012468272404706218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.90499086389905e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004798268169104879, "output_power": 0.00046471647565063964, "reference_active": true, "delta_db": -0.13896446628557638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7762549535694356e-05, "output_power": 3.4496936546589375e-05, "reference_active": true, "delta_db": -0.39280778118857484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.708330714477498e-07, "output_power": 6.351005855380175e-07, "reference_active": true, "delta_db": -0.8411782600080722}], "active_band_count": 3, "active_band_mae_db": 0.4576501691607411}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.983632889928613e-05, "log_rms_mae_db": 1.100207615041316, "rms_envelope_nrmse": 0.10055226387641268, "envelope_correlation": 0.9902479169761258, "rms_gain_error_db": -0.17189074458305387, "reference_mean_power": 0.00040681650183919396, "output_mean_power": 0.00039102946755434337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8313223654566366e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016437493457690703, "output_power": 0.0001562929717730794, "reference_active": true, "delta_db": -0.21896143999546708}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7703924674415462e-05, "output_power": 2.4355315163421483e-05, "reference_active": true, "delta_db": -0.559475439026763}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.687130809571553e-06, "output_power": 4.575339863506974e-06, "reference_active": true, "delta_db": -0.10483714399861971}], "active_band_count": 3, "active_band_mae_db": 0.29442467434028324}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.811887222434066e-05, "log_rms_mae_db": 0.6437002023919715, "rms_envelope_nrmse": 0.060766510935664333, "envelope_correlation": 0.9964317196552944, "rms_gain_error_db": -0.17330983493271276, "reference_mean_power": 0.00040698573595692556, "output_mean_power": 0.000391064330193382}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.593242154160197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015388869283670126, "output_power": 0.00014688788399778652, "reference_active": true, "delta_db": -0.202207360559962}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4161267354242513e-05, "output_power": 1.272218879572176e-05, "reference_active": true, "delta_db": -0.46540285855528357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7920568451213675e-07, "output_power": 2.750475577711971e-07, "reference_active": true, "delta_db": -0.06516463023520844}], "active_band_count": 3, "active_band_mae_db": 0.244258283116818}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4474245023095566e-05, "log_rms_mae_db": 1.669933178573462, "rms_envelope_nrmse": 0.26680554510008747, "envelope_correlation": 0.9410964649893946, "rms_gain_error_db": 0.2824016326427122, "reference_mean_power": 1.6116938837494647e-05, "output_mean_power": 1.7199773232630003e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.1253846819517355e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.226560638885928e-06, "output_power": 2.8329280997209296e-06, "reference_active": true, "delta_db": 1.0460102483262035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.434590024496585e-06, "output_power": 1.275499114085524e-06, "reference_active": true, "delta_db": -0.510476453174705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.807607987621042e-07, "output_power": 5.30477910123875e-07, "reference_active": true, "delta_db": 0.42738255892224913}], "active_band_count": 3, "active_band_mae_db": 0.6612897534743859}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.920336454108668e-05, "log_rms_mae_db": 1.0884631988352065, "rms_envelope_nrmse": 0.19544054626641238, "envelope_correlation": 0.9658944928900115, "rms_gain_error_db": 0.2824243120373538, "reference_mean_power": 1.619697385840346e-05, "output_mean_power": 1.7285275760716997e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.254945521069739e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4889493401597257e-06, "output_power": 3.11962277221381e-06, "reference_active": true, "delta_db": 0.980860247772128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.64830026091652e-07, "output_power": 6.611459509610061e-07, "reference_active": true, "delta_db": -0.6326758681767348}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8106205193126903e-08, "output_power": 3.873691073091754e-08, "reference_active": true, "delta_db": -0.9407611662096629}], "active_band_count": 3, "active_band_mae_db": 0.8514324273861753}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8868019980709333, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c2e5f642d3bae4935f73fa8f554b07e06f3a836bd12b5d06f96230f17757c0e2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9899232655594435, "standard_deviation": 0.0017713941199328672, "minimum": 0.986171109409872, "maximum": 0.9921053116842964, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.114705421678753, "standard_deviation": 0.02824059587307537, "minimum": 1.0697071176608641, "maximum": 1.1631693104626837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5303472672420138, "standard_deviation": 0.12193800518830437, "minimum": 0.29442467434028324, "maximum": 0.672001153929739, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10244674632734693, "standard_deviation": 0.008178331576042457, "minimum": 0.09118750799358068, "maximum": 0.1190281932616702, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968700934569823, "standard_deviation": 0.0007418609437552083, "minimum": 0.9954843998237998, "maximum": 0.9977960483159486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6318932188978215, "standard_deviation": 0.024213722256389087, "minimum": 0.59540490567312, "maximum": 0.6606840048790815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4988234555716004, "standard_deviation": 0.147819149413916, "minimum": 0.244258283116818, "maximum": 0.6688154695991866, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05826996561162755, "standard_deviation": 0.005770090656721949, "minimum": 0.04995170233272323, "maximum": 0.06737501902326981, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9408962405523574, "standard_deviation": 0.0035694200504227955, "minimum": 0.9361577339066878, "maximum": 0.9456378995240137, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.626027700077874, "standard_deviation": 0.028994364602886645, "minimum": 1.5937116820889439, "maximum": 1.669933178573462, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.120404067061623, "standard_deviation": 0.5731862429348249, "minimum": 0.5204209847380575, "maximum": 2.072158608090699, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26458274724262953, "standard_deviation": 0.005922204678821923, "minimum": 0.2518768984885068, "maximum": 0.27131289684342025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9656865572612534, "standard_deviation": 0.0031597257465529947, "minimum": 0.961181309105124, "maximum": 0.969697740344506, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0497635002119456, "standard_deviation": 0.02379512393514074, "minimum": 1.012696784073019, "maximum": 1.0884631988352065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0850613057837313, "standard_deviation": 0.621292769905279, "minimum": 0.46447556172877347, "maximum": 2.17759703553209, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19338902113209694, "standard_deviation": 0.006714284200103955, "minimum": 0.18382616506160898, "maximum": 0.2031168515415913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.994902047259836, "standard_deviation": 0.0006538907834514754, "minimum": 0.993912177086026, "maximum": 0.9958060027552899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0160233418209783, "standard_deviation": 0.03705172381850361, "minimum": 0.9612720542452955, "maximum": 1.0670917229356423, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47640575210779373, "standard_deviation": 0.18165107715279574, "minimum": 0.24505951020171982, "maximum": 0.7418514696534696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07471220696280101, "standard_deviation": 0.005178520125830788, "minimum": 0.06771960702454867, "maximum": 0.08334286302022978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983369137852216, "standard_deviation": 0.0004632610572462966, "minimum": 0.9973926786934006, "maximum": 0.9988546661867311, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5555015176814591, "standard_deviation": 0.030514575268628417, "minimum": 0.5102952369260266, "maximum": 0.5947327638839836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5379973116858051, "standard_deviation": 0.2105600750418359, "minimum": 0.13761863050709622, "maximum": 0.7863183307256781, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.042033507753965, "standard_deviation": 0.005744494749378398, "minimum": 0.03495827226625857, "maximum": 0.05121451998934017, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1371184272590543, "standard_deviation": 0.034786814016379064, "minimum": 0.08389564732917217, "maximum": 0.19262319421194798, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.445663691195497, "standard_deviation": 0.004391611039734372, "minimum": 5.438556442053479, "maximum": 5.45356079134915, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3239118546732045, "standard_deviation": 0.0071595831975236664, "minimum": 0.31583052496820074, "maximum": 0.3368034709612734, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09326782882531845, "standard_deviation": 0.026141443376804473, "minimum": -0.1468240702209917, "maximum": -0.06711439030647603, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9964634604956577, "standard_deviation": 0.0004705429535733138, "minimum": 0.9957057401083546, "maximum": 0.9972362782505667, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5700875467581734, "standard_deviation": 0.015655454774768082, "minimum": 0.5489206988233224, "maximum": 0.6010990783946321, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3338191324608743, "standard_deviation": 0.16431188057103505, "minimum": 0.10502287364977018, "maximum": 0.5888773065478672, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.056814134281850504, "standard_deviation": 0.004001017349459775, "minimum": 0.05003768062499307, "maximum": 0.06360668629211708, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989440905668445, "standard_deviation": 0.0002662486671735632, "minimum": 0.9984384133571593, "maximum": 0.9991952318623225, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.312631473974593, "standard_deviation": 0.02177720219260189, "minimum": 0.29231768896210103, "maximum": 0.35943501745162126, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37497936987940694, "standard_deviation": 0.1856220457636964, "minimum": 0.1334276662268224, "maximum": 0.7114267374574063, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031147410559245123, "standard_deviation": 0.003770088006158865, "minimum": 0.027234435518149053, "maximum": 0.03683990315024952, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_3729-6852-0017", "language": "en", "style": "normal", "speaker": "librispeech_3729", "source_utterance_id": "3729-6852-0017", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_3729-6852-0017.wav", "source_file_sha256": "2c5246b5e1f8cbc975b887a30ce5bfb90f4e35c3cc744304e2e58e3aca134f5e", "decoded_pcm_sha256": "ba156f183286a1f4044b0b7794b3982603585d0e122bdb35978c337419392c08", "original_source_sha256": "12a9659e29190ce17dd89b5209d4ccaebc746bfb35093185d5970a91878e42b9", "corpus_entry_id": "librispeech_3729-6852-0017"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.0289225578308105, "predicted_mos": 4.0289225578308105, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a606fe2f723e6d1773dd1b66dda6e2ab056e9def5a98f8cebfb3877194ede967", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.256136348001746, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_3729-6852-0017", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "a606fe2f723e6d1773dd1b66dda6e2ab056e9def5a98f8cebfb3877194ede967", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "fa4d86d7b8c74df263aaabe4276525584eed7ed36d8b9421d2158a8fad60fd8c"}, "rows": [{"case": "librispeech_3729-6852-0017", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.705315113067627, "predicted_mos": 2.705315113067627, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "107d61e8999105ad2a6d3e09a026839919807d2a13723c8924950b3394d34151", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.483205255000939, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_3729-6852-0017", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "107d61e8999105ad2a6d3e09a026839919807d2a13723c8924950b3394d34151", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5934959057065748, "全帯域_10ms_rms_envelope_nrmse": 0.05539296028815742, "全帯域_10ms_envelope_correlation": 0.9971129890823357, "全帯域_10ms_modulation_mae_db": 0.27528642727161384, "全帯域_40ms_log_rms_mae_db": 0.3270179622098544, "全帯域_40ms_rms_envelope_nrmse": 0.03277696498928793, "全帯域_40ms_envelope_correlation": 0.9991964399177337, "全帯域_40ms_modulation_mae_db": 0.44972136151422526, "300–1000 Hz_10ms_log_rms_mae_db": 1.1137864227107923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06525614996722517, "300–1000 Hz_10ms_envelope_correlation": 0.9966260300918929, "300–1000 Hz_10ms_modulation_mae_db": 0.401208031490238, "300–1000 Hz_40ms_log_rms_mae_db": 0.5693942229267708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03700670876875101, "300–1000 Hz_40ms_envelope_correlation": 0.9991077464062602, "300–1000 Hz_40ms_modulation_mae_db": 0.6992935021665883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2316982597951505, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11208759310006758, "1000–3000 Hz_10ms_envelope_correlation": 0.990951635094597, "1000–3000 Hz_10ms_modulation_mae_db": 0.816616014413667, "1000–3000 Hz_40ms_log_rms_mae_db": 0.809218082592798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06750073867206921, "1000–3000 Hz_40ms_envelope_correlation": 0.9971650581295857, "1000–3000 Hz_40ms_modulation_mae_db": 0.9326908906848453, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3561029323538352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.227269691773838, "3000–6000 Hz_10ms_envelope_correlation": 0.9609963994135732, "3000–6000 Hz_10ms_modulation_mae_db": 1.2014818223269639, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9118827734454659, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15895049905927972, "3000–6000 Hz_40ms_envelope_correlation": 0.9798917549804526, "3000–6000 Hz_40ms_modulation_mae_db": 1.2980792948932314, "mr_stft_db": 5.38461899651103, "band_mae_db": 0.22205936932244008, "rms_gain_error_db": -0.16416001010780795, "output_peak": 0.3468273187406886}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6053377298434366, "全帯域_10ms_rms_envelope_nrmse": 0.05416440650747002, "全帯域_10ms_envelope_correlation": 0.9969627078492003, "全帯域_10ms_modulation_mae_db": 0.22124725081492683, "全帯域_40ms_log_rms_mae_db": 0.30351339036763086, "全帯域_40ms_rms_envelope_nrmse": 0.026889081732291643, "全帯域_40ms_envelope_correlation": 0.9992520183119474, "全帯域_40ms_modulation_mae_db": 0.3252895498921183, "300–1000 Hz_10ms_log_rms_mae_db": 1.1489110879789632, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06597010873213399, "300–1000 Hz_10ms_envelope_correlation": 0.996293574540801, "300–1000 Hz_10ms_modulation_mae_db": 0.32491677274458036, "300–1000 Hz_40ms_log_rms_mae_db": 0.634807510580036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.032315088264241364, "300–1000 Hz_40ms_envelope_correlation": 0.9991135656837951, "300–1000 Hz_40ms_modulation_mae_db": 0.36416492235466663, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2238191572341994, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10367629864094807, "1000–3000 Hz_10ms_envelope_correlation": 0.9918981399990272, "1000–3000 Hz_10ms_modulation_mae_db": 0.5386035261388068, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8173627421635419, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06125269751928594, "1000–3000 Hz_40ms_envelope_correlation": 0.9972215804935685, "1000–3000 Hz_40ms_modulation_mae_db": 0.6549040094776932, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3601074758383336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21889509816532823, "3000–6000 Hz_10ms_envelope_correlation": 0.9630401839933357, "3000–6000 Hz_10ms_modulation_mae_db": 1.208786587947856, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9417832059637143, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14660791335495954, "3000–6000 Hz_40ms_envelope_correlation": 0.9823566844105747, "3000–6000 Hz_40ms_modulation_mae_db": 1.3280563945044446, "mr_stft_db": 5.392873559038808, "band_mae_db": 0.13231225196898302, "rms_gain_error_db": -0.061946545049092555, "output_peak": 0.3711175717442075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5900151603009406, "全帯域_10ms_rms_envelope_nrmse": 0.056411852850753806, "全帯域_10ms_envelope_correlation": 0.9968192745509972, "全帯域_10ms_modulation_mae_db": 0.35843520173544313, "全帯域_40ms_log_rms_mae_db": 0.3460955813647482, "全帯域_40ms_rms_envelope_nrmse": 0.03330693069079309, "全帯域_40ms_envelope_correlation": 0.9989648626664207, "全帯域_40ms_modulation_mae_db": 0.32373814131660994, "300–1000 Hz_10ms_log_rms_mae_db": 1.0902962659086577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06486256040039626, "300–1000 Hz_10ms_envelope_correlation": 0.9964516148224042, "300–1000 Hz_10ms_modulation_mae_db": 0.3998727310415551, "300–1000 Hz_40ms_log_rms_mae_db": 0.5979678303877407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03834015645291115, "300–1000 Hz_40ms_envelope_correlation": 0.9987643923568174, "300–1000 Hz_40ms_modulation_mae_db": 0.3495821624559829, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1985599256012724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09632967104033864, "1000–3000 Hz_10ms_envelope_correlation": 0.9933309493487067, "1000–3000 Hz_10ms_modulation_mae_db": 0.3457026068321805, "1000–3000 Hz_40ms_log_rms_mae_db": 0.806491594867171, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05677313046271407, "1000–3000 Hz_40ms_envelope_correlation": 0.9980027560448488, "1000–3000 Hz_40ms_modulation_mae_db": 0.248274231167388, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3178269041680513, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25490510541655687, "3000–6000 Hz_10ms_envelope_correlation": 0.947160743378465, "3000–6000 Hz_10ms_modulation_mae_db": 1.591012781849276, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8997760558317791, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16913489876740875, "3000–6000 Hz_40ms_envelope_correlation": 0.9752353011003059, "3000–6000 Hz_40ms_modulation_mae_db": 1.6351765978477424, "mr_stft_db": 5.3912387967596125, "band_mae_db": 0.24950450603992821, "rms_gain_error_db": -0.12057829246583913, "output_peak": 0.33377405453882913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5439212152830166, "全帯域_10ms_rms_envelope_nrmse": 0.05180368559449946, "全帯域_10ms_envelope_correlation": 0.9971883566169927, "全帯域_10ms_modulation_mae_db": 0.06993106634797405, "全帯域_40ms_log_rms_mae_db": 0.2816807728322364, "全帯域_40ms_rms_envelope_nrmse": 0.02589671702673865, "全帯域_40ms_envelope_correlation": 0.9992571111336838, "全帯域_40ms_modulation_mae_db": 0.09651829330841379, "300–1000 Hz_10ms_log_rms_mae_db": 1.1266168269460195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06772889635729401, "300–1000 Hz_10ms_envelope_correlation": 0.9960890861417, "300–1000 Hz_10ms_modulation_mae_db": 0.1286094960582643, "300–1000 Hz_40ms_log_rms_mae_db": 0.6151474828376546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034342961291174055, "300–1000 Hz_40ms_envelope_correlation": 0.9989517455720683, "300–1000 Hz_40ms_modulation_mae_db": 0.13260954407143052, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2047794156425968, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09743340888094522, "1000–3000 Hz_10ms_envelope_correlation": 0.992972482516134, "1000–3000 Hz_10ms_modulation_mae_db": 0.4069917568489681, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7742146770203974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05927788191409775, "1000–3000 Hz_40ms_envelope_correlation": 0.9975237138637107, "1000–3000 Hz_40ms_modulation_mae_db": 0.24769165549137942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3365362871020199, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2294835182598659, "3000–6000 Hz_10ms_envelope_correlation": 0.95840451355373, "3000–6000 Hz_10ms_modulation_mae_db": 1.2649470474269158, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9183023255952462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15679761074638038, "3000–6000 Hz_40ms_envelope_correlation": 0.9791131948195844, "3000–6000 Hz_40ms_modulation_mae_db": 1.4455751972508246, "mr_stft_db": 5.374889546603786, "band_mae_db": 0.08311045598062479, "rms_gain_error_db": -0.006377595224073973, "output_peak": 0.35570848983864356}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5667475959254888, "全帯域_10ms_rms_envelope_nrmse": 0.05467883054682784, "全帯域_10ms_envelope_correlation": 0.996891007769369, "全帯域_10ms_modulation_mae_db": 0.20710256549437958, "全帯域_40ms_log_rms_mae_db": 0.30505772992627195, "全帯域_40ms_rms_envelope_nrmse": 0.02842569439308292, "全帯域_40ms_envelope_correlation": 0.9991445375156938, "全帯域_40ms_modulation_mae_db": 0.12547912619110813, "300–1000 Hz_10ms_log_rms_mae_db": 1.140525376943339, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06917121696483211, "300–1000 Hz_10ms_envelope_correlation": 0.995904780249337, "300–1000 Hz_10ms_modulation_mae_db": 0.13275470656478605, "300–1000 Hz_40ms_log_rms_mae_db": 0.6315655843026173, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03754195285125451, "300–1000 Hz_40ms_envelope_correlation": 0.9987670041963899, "300–1000 Hz_40ms_modulation_mae_db": 0.11380960450080789, "1000–3000 Hz_10ms_log_rms_mae_db": 1.184436701640035, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09976926508579557, "1000–3000 Hz_10ms_envelope_correlation": 0.9929197659372345, "1000–3000 Hz_10ms_modulation_mae_db": 0.6096776674226537, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7883784041405293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0586983437257031, "1000–3000 Hz_40ms_envelope_correlation": 0.997944029069974, "1000–3000 Hz_40ms_modulation_mae_db": 0.41715495902711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3394854604151536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25974965955438306, "3000–6000 Hz_10ms_envelope_correlation": 0.9457849824925356, "3000–6000 Hz_10ms_modulation_mae_db": 1.227261404759874, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8981461483329923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1737140901685955, "3000–6000 Hz_40ms_envelope_correlation": 0.9739342888893727, "3000–6000 Hz_40ms_modulation_mae_db": 1.5755708794932872, "mr_stft_db": 5.372801703293619, "band_mae_db": 0.13952497603978453, "rms_gain_error_db": -0.0570212361243918, "output_peak": 0.37626407313177457}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5425578558823764, "全帯域_10ms_rms_envelope_nrmse": 0.048002780998037876, "全帯域_10ms_envelope_correlation": 0.9976157386351762, "全帯域_10ms_modulation_mae_db": 0.1945353755990077, "全帯域_40ms_log_rms_mae_db": 0.2903599903101211, "全帯域_40ms_rms_envelope_nrmse": 0.0253917184578731, "全帯域_40ms_envelope_correlation": 0.9993274981456053, "全帯域_40ms_modulation_mae_db": 0.32817507863503104, "300–1000 Hz_10ms_log_rms_mae_db": 1.1391062961493996, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05965811801276405, "300–1000 Hz_10ms_envelope_correlation": 0.9969453067220776, "300–1000 Hz_10ms_modulation_mae_db": 0.22039034806726907, "300–1000 Hz_40ms_log_rms_mae_db": 0.6190286108035757, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030339939365747977, "300–1000 Hz_40ms_envelope_correlation": 0.9991794709685127, "300–1000 Hz_40ms_modulation_mae_db": 0.45811140524011323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1951480574438762, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10456560224037399, "1000–3000 Hz_10ms_envelope_correlation": 0.991922113751168, "1000–3000 Hz_10ms_modulation_mae_db": 0.6231115859087691, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8048432558517388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06048703253208696, "1000–3000 Hz_40ms_envelope_correlation": 0.9975084051594255, "1000–3000 Hz_40ms_modulation_mae_db": 0.7977306353747308, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3211771008325204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23021449186488557, "3000–6000 Hz_10ms_envelope_correlation": 0.9584865758473924, "3000–6000 Hz_10ms_modulation_mae_db": 1.0252511714036143, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9265143302013266, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1595772834086324, "3000–6000 Hz_40ms_envelope_correlation": 0.9786341267652595, "3000–6000 Hz_40ms_modulation_mae_db": 1.124745394363323, "mr_stft_db": 5.387878956038201, "band_mae_db": 0.15852162606608608, "rms_gain_error_db": -0.05745760653357923, "output_peak": 0.3541620133231308}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5783716210628066, "全帯域_10ms_rms_envelope_nrmse": 0.058010968487931615, "全帯域_10ms_envelope_correlation": 0.9966052814873483, "全帯域_10ms_modulation_mae_db": 0.5720338297362416, "全帯域_40ms_log_rms_mae_db": 0.32514119327982344, "全帯域_40ms_rms_envelope_nrmse": 0.029173571331127784, "全帯域_40ms_envelope_correlation": 0.9992279056864676, "全帯域_40ms_modulation_mae_db": 0.5771162327957047, "300–1000 Hz_10ms_log_rms_mae_db": 1.1143659421571435, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0682492543097093, "300–1000 Hz_10ms_envelope_correlation": 0.9962375784357935, "300–1000 Hz_10ms_modulation_mae_db": 0.6462772715302973, "300–1000 Hz_40ms_log_rms_mae_db": 0.6072281626281427, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03831419037981438, "300–1000 Hz_40ms_envelope_correlation": 0.9989708897876287, "300–1000 Hz_40ms_modulation_mae_db": 0.6759024831347885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2216796734624875, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10572361895655415, "1000–3000 Hz_10ms_envelope_correlation": 0.9915774178989019, "1000–3000 Hz_10ms_modulation_mae_db": 0.7219722067837878, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8057147726229288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06298033123218585, "1000–3000 Hz_40ms_envelope_correlation": 0.9970642395208879, "1000–3000 Hz_40ms_modulation_mae_db": 0.5683293901721789, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3743913381316744, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26428143417432803, "3000–6000 Hz_10ms_envelope_correlation": 0.9452626654765294, "3000–6000 Hz_10ms_modulation_mae_db": 1.4041718207391636, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9414125816207943, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1695435274911794, "3000–6000 Hz_40ms_envelope_correlation": 0.9757817191737388, "3000–6000 Hz_40ms_modulation_mae_db": 1.5516969367285842, "mr_stft_db": 5.383413307756677, "band_mae_db": 0.14245952469734072, "rms_gain_error_db": -0.10650018542659491, "output_peak": 0.33564624950997374}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5688123012528017, "全帯域_10ms_rms_envelope_nrmse": 0.051249517250355925, "全帯域_10ms_envelope_correlation": 0.9973111228630918, "全帯域_10ms_modulation_mae_db": 0.448142932137986, "全帯域_40ms_log_rms_mae_db": 0.2974324817602053, "全帯域_40ms_rms_envelope_nrmse": 0.02683647064056383, "全帯域_40ms_envelope_correlation": 0.9992823195774914, "全帯域_40ms_modulation_mae_db": 0.4345708123083929, "300–1000 Hz_10ms_log_rms_mae_db": 1.1673240985482873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06594304070095115, "300–1000 Hz_10ms_envelope_correlation": 0.9962963468361837, "300–1000 Hz_10ms_modulation_mae_db": 0.2600589765152967, "300–1000 Hz_40ms_log_rms_mae_db": 0.659691691895243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03374631333626722, "300–1000 Hz_40ms_envelope_correlation": 0.9990257555337764, "300–1000 Hz_40ms_modulation_mae_db": 0.40575971646161113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.223156858974444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10523805750595709, "1000–3000 Hz_10ms_envelope_correlation": 0.9920151591238507, "1000–3000 Hz_10ms_modulation_mae_db": 0.40886024508552427, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7597100660231646, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0611094992064968, "1000–3000 Hz_40ms_envelope_correlation": 0.9977029721554209, "1000–3000 Hz_40ms_modulation_mae_db": 0.45265285593698684, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3403831005236821, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2527496699380374, "3000–6000 Hz_10ms_envelope_correlation": 0.9548857404871274, "3000–6000 Hz_10ms_modulation_mae_db": 2.0636208951139436, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8987092767712485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17410896716623603, "3000–6000 Hz_40ms_envelope_correlation": 0.9781768222327113, "3000–6000 Hz_40ms_modulation_mae_db": 2.278898156084818, "mr_stft_db": 5.376415267001561, "band_mae_db": 0.12855004118669136, "rms_gain_error_db": -0.07666016645541987, "output_peak": 0.3507786669526911}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.12508497243231, "spectral_convergence": 0.20111466448496376, "spectral_convergence_status": "ok", "amplitude_floor": 1.1114078093309192e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.193979978360382, "spectral_convergence": 0.19589028347266393, "spectral_convergence_status": "ok", "amplitude_floor": 9.683415126458649e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8347920387403995, "spectral_convergence": 0.20605162968585994, "spectral_convergence_status": "ok", "amplitude_floor": 7.055286564887976e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011089954817882694, "output_power": 0.0001123150407354222, "delta_db": 0.05508142186843117, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007963949709957278, "output_power": 0.0007687649798139307, "delta_db": -0.1533491758392698, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005473467336544822, "output_power": 0.0005281127175465641, "delta_db": -0.15535904820341886, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020656259415557, "output_power": 0.00019464647682457606, "delta_db": -0.2580513189666817, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.567254948983353e-05, "output_power": 9.015519076926704e-05, "delta_db": -0.25796610542797915, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.697083983250908e-05, "output_power": 3.331228332627822e-05, "delta_db": -0.4525491456288597, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22205936932244008, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16416001010780795, "pitch": null, "output_rms_dbfs": -27.626305451643496, "output_peak": 0.3468273187406886, "reference_rms_dbfs": -27.46214544153569, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.38461899651103}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015211892717126925, "log_rms_mae_db": 0.5934959057065748, "rms_envelope_nrmse": 0.05539296028815742, "envelope_correlation": 0.9971129890823357, "rms_gain_error_db": -0.1634937658752771, "reference_mean_power": 0.0017946561941472319, "output_mean_power": 0.001728350762759701}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.098233390179555e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033819477659287355, "output_power": 0.00032104420128433424, "reference_active": true, "delta_db": -0.22602065573907326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.872738107665891e-05, "output_power": 3.46644508181028e-05, "reference_active": true, "delta_db": -0.48133804842483574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.329707857377202e-06, "output_power": 7.132415044521933e-06, "reference_active": true, "delta_db": -0.11850057765093255}], "active_band_count": 3, "active_band_mae_db": 0.27528642727161384}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012855597503212083, "log_rms_mae_db": 0.3270179622098544, "rms_envelope_nrmse": 0.03277696498928793, "envelope_correlation": 0.9991964399177337, "rms_gain_error_db": -0.16364264088605515, "reference_mean_power": 0.0017996791281982027, "output_mean_power": 0.0017331287070947068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.521732580259768e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030590519204748624, "output_power": 0.0002898505680554844, "reference_active": true, "delta_db": -0.23412692364547485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7673272930744677e-05, "output_power": 1.621361898436453e-05, "reference_active": true, "delta_db": -0.3743702137688748}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.996266898550172e-07, "output_power": 3.369673255543785e-07, "reference_active": true, "delta_db": -0.7406669471283261}], "active_band_count": 3, "active_band_mae_db": 0.44972136151422526}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013968182745477764, "log_rms_mae_db": 1.1137864227107923, "rms_envelope_nrmse": 0.06525614996722517, "envelope_correlation": 0.9966260300918929, "rms_gain_error_db": -0.17019553816760583, "reference_mean_power": 0.0010910330804496425, "output_mean_power": 0.0010491035700375692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.057825806205699e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031975947498638787, "output_power": 0.00030680295440088537, "reference_active": true, "delta_db": -0.17963884788790102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9465579420304694e-05, "output_power": 3.4414600819727255e-05, "reference_active": true, "delta_db": -0.5947574688289694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.299925023694004e-06, "output_power": 5.7070621639253784e-06, "reference_active": true, "delta_db": -0.42922777775384346}], "active_band_count": 3, "active_band_mae_db": 0.401208031490238}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012087571887341512, "log_rms_mae_db": 0.5693942229267708, "rms_envelope_nrmse": 0.03700670876875101, "envelope_correlation": 0.9991077464062602, "rms_gain_error_db": -0.17022347558480586, "reference_mean_power": 0.0010942126232004053, "output_mean_power": 0.0010521541513796421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.627912926828161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002965811192999727, "output_power": 0.0002836383373727334, "reference_active": true, "delta_db": -0.19378568910431102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8971998770340482e-05, "output_power": 1.672867934534956e-05, "reference_active": true, "delta_db": -0.546514311286775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.113014123121411e-07, "output_power": 3.0088611732990857e-07, "reference_active": true, "delta_db": -1.3575805061086785}], "active_band_count": 3, "active_band_mae_db": 0.6992935021665883}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.190065013136077e-05, "log_rms_mae_db": 1.2316982597951505, "rms_envelope_nrmse": 0.11208759310006758, "envelope_correlation": 0.990951635094597, "rms_gain_error_db": -0.2932288501601733, "reference_mean_power": 0.00025666306625944256, "output_mean_power": 0.00023990565948852269}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2531384738435257e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.158137778593675e-05, "output_power": 6.420326724448013e-05, "reference_active": true, "delta_db": -0.47242923886290344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.324550913609082e-06, "output_power": 5.646295858604594e-06, "reference_active": true, "delta_db": 0.2548064674069805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7772043437261862e-06, "output_power": 2.6424002376422596e-06, "reference_active": true, "delta_db": 1.722612336971117}], "active_band_count": 3, "active_band_mae_db": 0.816616014413667}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.863962912697811e-05, "log_rms_mae_db": 0.809218082592798, "rms_envelope_nrmse": 0.06750073867206921, "envelope_correlation": 0.9971650581295857, "rms_gain_error_db": -0.2934984845577545, "reference_mean_power": 0.00025765686496881694, "output_mean_power": 0.00024081962166515962}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.1689575138838273e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607374483640592e-05, "output_power": 5.898523052850863e-05, "reference_active": true, "delta_db": -0.49285640764707267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.394936823991012e-06, "output_power": 2.6516831014312947e-06, "reference_active": true, "delta_db": 0.44227559315440823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.833675392620637e-08, "output_power": 1.5101147006612893e-07, "reference_active": true, "delta_db": 1.862940671253055}], "active_band_count": 3, "active_band_mae_db": 0.9326908906848453}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.534983553210222e-05, "log_rms_mae_db": 1.3561029323538352, "rms_envelope_nrmse": 0.227269691773838, "envelope_correlation": 0.9609963994135732, "rms_gain_error_db": 0.2161914731795893, "reference_mean_power": 1.9026430507347487e-05, "output_mean_power": 1.9997535044504692e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1975009508047096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.298926861031976e-06, "output_power": 5.697865143171249e-06, "reference_active": true, "delta_db": 0.31524240999183273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588034222302733e-06, "output_power": 1.6277466847148014e-06, "reference_active": true, "delta_db": 0.10726962194783646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.193680029927605e-07, "output_power": 8.725466089591835e-07, "reference_active": true, "delta_db": 3.1819334350412225}], "active_band_count": 3, "active_band_mae_db": 1.2014818223269639}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3831266720127666e-05, "log_rms_mae_db": 0.9118827734454659, "rms_envelope_nrmse": 0.15895049905927972, "envelope_correlation": 0.9798917549804526, "rms_gain_error_db": 0.21627836896629699, "reference_mean_power": 1.9104126133492904e-05, "output_mean_power": 2.007959799563377e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1095182205605233e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.38789012101126e-06, "output_power": 5.912997950781979e-06, "reference_active": true, "delta_db": 0.4038899821736109}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.477294263383285e-07, "output_power": 7.881321123877913e-07, "reference_active": true, "delta_db": 0.22854550843529015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.443035434026362e-08, "output_power": 5.1773795679851575e-08, "reference_active": true, "delta_db": 3.261802394070793}], "active_band_count": 3, "active_band_mae_db": 1.2980792948932314}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9219961841386896, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ce59cc5514544ac37c382dc0d473973acb809a26304e05136bac21dfb9c6a60e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9921984579587024, "standard_deviation": 0.0008055476783767625, "minimum": 0.990951635094597, "maximum": 0.9933309493487067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2104097562242577, "standard_deviation": 0.016912169404521962, "minimum": 1.184436701640035, "maximum": 1.2316982597951505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5589419511792947, "standard_deviation": 0.16525508057976573, "minimum": 0.3457026068321805, "maximum": 0.816616014413667, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10310293943137254, "standard_deviation": 0.005123516387210834, "minimum": 0.09632967104033864, "maximum": 0.11208759310006758, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975165943046778, "standard_deviation": 0.00035236828364240317, "minimum": 0.9970642395208879, "maximum": 0.9980027560448488, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7957416994102837, "standard_deviation": 0.01986688766862361, "minimum": 0.7597100660231646, "maximum": 0.8173627421635419, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5399285784165391, "standard_deviation": 0.24724342577095412, "minimum": 0.24769165549137942, "maximum": 0.9326908906848453, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06100995690807996, "standard_deviation": 0.003221562222613882, "minimum": 0.05677313046271407, "maximum": 0.06750073867206921, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9542527255803361, "standard_deviation": 0.00718212231550457, "minimum": 0.9452626654765294, "maximum": 0.9630401839933357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3432513249206588, "standard_deviation": 0.019347080914657325, "minimum": 1.3178269041680513, "maximum": 1.3743913381316744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.373316691445951, "standard_deviation": 0.3241883558190549, "minimum": 1.0252511714036143, "maximum": 2.0636208951139436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24219358364340288, "standard_deviation": 0.017483416886198082, "minimum": 0.21889509816532823, "maximum": 0.26428143417432803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9778904865465, "standard_deviation": 0.00275759762462563, "minimum": 0.9739342888893727, "maximum": 0.9823566844105747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9170658372203209, "standard_deviation": 0.01818646998373022, "minimum": 0.8981461483329923, "maximum": 0.9417832059637143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5297248563957817, "standard_deviation": 0.34648947569627964, "minimum": 1.124745394363323, "maximum": 2.278898156084818, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16355434877033398, "standard_deviation": 0.009647211128076342, "minimum": 0.14660791335495954, "maximum": 0.17410896716623603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963555397300237, "standard_deviation": 0.0003219717242706771, "minimum": 0.995904780249337, "maximum": 0.9969453067220776, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1301165396678252, "standard_deviation": 0.024022172294578707, "minimum": 1.0902962659086577, "maximum": 1.1673240985482873, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.31426104175153585, "standard_deviation": 0.1708792718964712, "minimum": 0.1286094960582643, "maximum": 0.6462772715302973, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06585491818066326, "standard_deviation": 0.002927874446744826, "minimum": 0.05965811801276405, "maximum": 0.06917121696483211, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989850713131561, "standard_deviation": 0.00015518219864545747, "minimum": 0.9987643923568174, "maximum": 0.9991794709685127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6168538870452226, "standard_deviation": 0.026947681644771477, "minimum": 0.5693942229267708, "maximum": 0.659691691895243, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3999041675482486, "standard_deviation": 0.21591583229551226, "minimum": 0.11380960450080789, "maximum": 0.6992935021665883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03524341383877021, "standard_deviation": 0.0030019815186534808, "minimum": 0.030339939365747977, "maximum": 0.03834015645291115, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15700534391273485, "standard_deviation": 0.05370218398179381, "minimum": 0.08311045598062479, "maximum": 0.24950450603992821, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.383016266625411, "standard_deviation": 0.007607350083078894, "minimum": 5.372801703293619, "maximum": 5.392873559038808, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3530348047224924, "standard_deviation": 0.015080573857917307, "minimum": 0.33377405453882913, "maximum": 0.37626407313177457, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08133770467334994, "standard_deviation": 0.04812331276343951, "minimum": -0.16416001010780795, "maximum": -0.006377595224073973, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9970633098568139, "standard_deviation": 0.00031483454836675836, "minimum": 0.9966052814873483, "maximum": 0.9976157386351762, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5736574231571803, "standard_deviation": 0.02270825057742952, "minimum": 0.5425578558823764, "maximum": 0.6053377298434366, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29333933114219657, "standard_deviation": 0.15968973503952633, "minimum": 0.06993106634797405, "maximum": 0.5720338297362416, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05371437531550424, "standard_deviation": 0.003207936977438178, "minimum": 0.048002780998037876, "maximum": 0.058010968487931615, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992065866193804, "standard_deviation": 0.0001120011876416082, "minimum": 0.9989648626664207, "maximum": 0.9993274981456053, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.30953738775636147, "standard_deviation": 0.02148819923911184, "minimum": 0.2816807728322364, "maximum": 0.3460955813647482, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3325760744952005, "standard_deviation": 0.16155000427019747, "minimum": 0.09651829330841379, "maximum": 0.5771162327957047, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028587143657719867, "standard_deviation": 0.003014213085872502, "minimum": 0.0253917184578731, "maximum": 0.03330693069079309, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0017", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.827402114868164, "predicted_mos": 2.827402114868164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "dcb60cfe358462c0594b112a5e823068007296f44214f947797af9c465149494", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.821757299825549, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6008830797100233, "全帯域_10ms_rms_envelope_nrmse": 0.05635062607691683, "全帯域_10ms_envelope_correlation": 0.9970038998889468, "全帯域_10ms_modulation_mae_db": 0.3114186819375629, "全帯域_40ms_log_rms_mae_db": 0.3392548841105275, "全帯域_40ms_rms_envelope_nrmse": 0.03342602392695288, "全帯域_40ms_envelope_correlation": 0.999152338319421, "全帯域_40ms_modulation_mae_db": 0.5043364626091433, "300–1000 Hz_10ms_log_rms_mae_db": 1.1262596561301175, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06675917425033157, "300–1000 Hz_10ms_envelope_correlation": 0.996450334579776, "300–1000 Hz_10ms_modulation_mae_db": 0.4802389459516146, "300–1000 Hz_40ms_log_rms_mae_db": 0.5799748384581929, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03810144043542077, "300–1000 Hz_40ms_envelope_correlation": 0.99902833708323, "300–1000 Hz_40ms_modulation_mae_db": 0.7619337904067041, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2362102214623814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11229139686507361, "1000–3000 Hz_10ms_envelope_correlation": 0.990920814820747, "1000–3000 Hz_10ms_modulation_mae_db": 0.8457551498284426, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8166648402704031, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0661985996601192, "1000–3000 Hz_40ms_envelope_correlation": 0.9973062428182156, "1000–3000 Hz_40ms_modulation_mae_db": 0.964814623836331, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3586321658740246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2288949057662453, "3000–6000 Hz_10ms_envelope_correlation": 0.9610486529409196, "3000–6000 Hz_10ms_modulation_mae_db": 1.2635236449660356, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9346369879777614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1613821102924361, "3000–6000 Hz_40ms_envelope_correlation": 0.9797710013216037, "3000–6000 Hz_40ms_modulation_mae_db": 1.3448094228907852, "mr_stft_db": 5.361520508867493, "band_mae_db": 0.22550626368539683, "rms_gain_error_db": -0.16553761406024628, "output_peak": 0.34644950844564315}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6089376415692107, "全帯域_10ms_rms_envelope_nrmse": 0.05518248427960378, "全帯域_10ms_envelope_correlation": 0.9968408291043528, "全帯域_10ms_modulation_mae_db": 0.2607438420900148, "全帯域_40ms_log_rms_mae_db": 0.31207256752100787, "全帯域_40ms_rms_envelope_nrmse": 0.02771679449976189, "全帯域_40ms_envelope_correlation": 0.9991961116089667, "全帯域_40ms_modulation_mae_db": 0.2860065695199037, "300–1000 Hz_10ms_log_rms_mae_db": 1.1610283799615504, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06701150076155533, "300–1000 Hz_10ms_envelope_correlation": 0.9961641717582503, "300–1000 Hz_10ms_modulation_mae_db": 0.33853398347424485, "300–1000 Hz_40ms_log_rms_mae_db": 0.6454176579739044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03385916348428153, "300–1000 Hz_40ms_envelope_correlation": 0.999007553556928, "300–1000 Hz_40ms_modulation_mae_db": 0.33960306688047665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2219618999547572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10201452127506423, "1000–3000 Hz_10ms_envelope_correlation": 0.9922030753534963, "1000–3000 Hz_10ms_modulation_mae_db": 0.4887120589501946, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8254571845377612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059823631211335423, "1000–3000 Hz_40ms_envelope_correlation": 0.997404167279197, "1000–3000 Hz_40ms_modulation_mae_db": 0.6016020885335411, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3497349898161168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2207769489315195, "3000–6000 Hz_10ms_envelope_correlation": 0.9623222208962469, "3000–6000 Hz_10ms_modulation_mae_db": 1.2364941363774182, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9417443102819498, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14566230107030084, "3000–6000 Hz_40ms_envelope_correlation": 0.982521817092727, "3000–6000 Hz_40ms_modulation_mae_db": 1.3496159011153237, "mr_stft_db": 5.367999991940944, "band_mae_db": 0.1265570327400464, "rms_gain_error_db": -0.058980288493363855, "output_peak": 0.3732351880864657}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.593367021226858, "全帯域_10ms_rms_envelope_nrmse": 0.05721369720333366, "全帯域_10ms_envelope_correlation": 0.9967202528058581, "全帯域_10ms_modulation_mae_db": 0.23804403347499647, "全帯域_40ms_log_rms_mae_db": 0.34869183696895817, "全帯域_40ms_rms_envelope_nrmse": 0.03402018082241928, "全帯域_40ms_envelope_correlation": 0.9989085396047761, "全帯域_40ms_modulation_mae_db": 0.27924757455563687, "300–1000 Hz_10ms_log_rms_mae_db": 1.09720391047941, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0661347651744873, "300–1000 Hz_10ms_envelope_correlation": 0.9963012840301345, "300–1000 Hz_10ms_modulation_mae_db": 0.3261100220310235, "300–1000 Hz_40ms_log_rms_mae_db": 0.5986484660009948, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03939487522654562, "300–1000 Hz_40ms_envelope_correlation": 0.9986835933895754, "300–1000 Hz_40ms_modulation_mae_db": 0.2085955649550305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.210454520356508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.096805709005671, "1000–3000 Hz_10ms_envelope_correlation": 0.9933092632067632, "1000–3000 Hz_10ms_modulation_mae_db": 0.47163158476461353, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8235980916126864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0567710163630094, "1000–3000 Hz_40ms_envelope_correlation": 0.9980600720463982, "1000–3000 Hz_40ms_modulation_mae_db": 0.328320265002088, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3073598361040157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25036555960744006, "3000–6000 Hz_10ms_envelope_correlation": 0.9491623459803009, "3000–6000 Hz_10ms_modulation_mae_db": 1.5318790564138938, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8897838418256009, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16128458395494674, "3000–6000 Hz_40ms_envelope_correlation": 0.9775815311801799, "3000–6000 Hz_40ms_modulation_mae_db": 1.5993486071676664, "mr_stft_db": 5.361396395845021, "band_mae_db": 0.2373127817797219, "rms_gain_error_db": -0.11966716262836083, "output_peak": 0.3312770584100516}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5528638789246835, "全帯域_10ms_rms_envelope_nrmse": 0.05283248946995029, "全帯域_10ms_envelope_correlation": 0.9970766674992557, "全帯域_10ms_modulation_mae_db": 0.17361126470737762, "全帯域_40ms_log_rms_mae_db": 0.28669935633293275, "全帯域_40ms_rms_envelope_nrmse": 0.02704076700041408, "全帯域_40ms_envelope_correlation": 0.999190274564139, "全帯域_40ms_modulation_mae_db": 0.21543962170723954, "300–1000 Hz_10ms_log_rms_mae_db": 1.10985463518338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06726514192672862, "300–1000 Hz_10ms_envelope_correlation": 0.9961501455051126, "300–1000 Hz_10ms_modulation_mae_db": 0.33168028737699645, "300–1000 Hz_40ms_log_rms_mae_db": 0.6161033106441018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03510899204356759, "300–1000 Hz_40ms_envelope_correlation": 0.9989104577386217, "300–1000 Hz_40ms_modulation_mae_db": 0.30823255870198923, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2153575151810527, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09816297307305871, "1000–3000 Hz_10ms_envelope_correlation": 0.9928861103904362, "1000–3000 Hz_10ms_modulation_mae_db": 0.3624613301405069, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7867097831686194, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059867896331100075, "1000–3000 Hz_40ms_envelope_correlation": 0.9974948312617334, "1000–3000 Hz_40ms_modulation_mae_db": 0.18295226891110114, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3101019898690052, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22952707552922627, "3000–6000 Hz_10ms_envelope_correlation": 0.9588337178198638, "3000–6000 Hz_10ms_modulation_mae_db": 1.3372171634630503, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8996829070695217, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15803010648671373, "3000–6000 Hz_40ms_envelope_correlation": 0.979075817661093, "3000–6000 Hz_40ms_modulation_mae_db": 1.5379955232977869, "mr_stft_db": 5.343445972442652, "band_mae_db": 0.08342218439514194, "rms_gain_error_db": -0.001023797975474682, "output_peak": 0.35807530417517885}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5587771538327829, "全帯域_10ms_rms_envelope_nrmse": 0.05480755790028076, "全帯域_10ms_envelope_correlation": 0.9968791479189519, "全帯域_10ms_modulation_mae_db": 0.14191782525337313, "全帯域_40ms_log_rms_mae_db": 0.3047590067208292, "全帯域_40ms_rms_envelope_nrmse": 0.029335632623332348, "全帯域_40ms_envelope_correlation": 0.9990897039094582, "全帯域_40ms_modulation_mae_db": 0.13599473936930812, "300–1000 Hz_10ms_log_rms_mae_db": 1.1462231491509074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07008007050706955, "300–1000 Hz_10ms_envelope_correlation": 0.995795621922568, "300–1000 Hz_10ms_modulation_mae_db": 0.08148395470912599, "300–1000 Hz_40ms_log_rms_mae_db": 0.6311523636484543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03827564025605373, "300–1000 Hz_40ms_envelope_correlation": 0.9987170692519379, "300–1000 Hz_40ms_modulation_mae_db": 0.10464259974600297, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2130065263823557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10329568490316986, "1000–3000 Hz_10ms_envelope_correlation": 0.9924113061795725, "1000–3000 Hz_10ms_modulation_mae_db": 0.7052783060859263, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8214552662688951, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06222669009494093, "1000–3000 Hz_40ms_envelope_correlation": 0.9976531065686681, "1000–3000 Hz_40ms_modulation_mae_db": 0.48610546235885455, "3000–6000 Hz_10ms_log_rms_mae_db": 1.320102415256966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2518247066112478, "3000–6000 Hz_10ms_envelope_correlation": 0.9487960256881164, "3000–6000 Hz_10ms_modulation_mae_db": 1.2047090934267743, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8868666827297168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17081600073307993, "3000–6000 Hz_40ms_envelope_correlation": 0.9747684494570564, "3000–6000 Hz_40ms_modulation_mae_db": 1.5516359902675612, "mr_stft_db": 5.352076646669989, "band_mae_db": 0.13839181903539913, "rms_gain_error_db": -0.05962177269554912, "output_peak": 0.3782456555931453}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5475280489581728, "全帯域_10ms_rms_envelope_nrmse": 0.04875956483460422, "全帯域_10ms_envelope_correlation": 0.9975353911891811, "全帯域_10ms_modulation_mae_db": 0.20005131512664487, "全帯域_40ms_log_rms_mae_db": 0.3019116837726274, "全帯域_40ms_rms_envelope_nrmse": 0.02615171720049917, "全帯域_40ms_envelope_correlation": 0.9992803579818932, "全帯域_40ms_modulation_mae_db": 0.38520263408022554, "300–1000 Hz_10ms_log_rms_mae_db": 1.1252298987445308, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05909213058287633, "300–1000 Hz_10ms_envelope_correlation": 0.9969992080897935, "300–1000 Hz_10ms_modulation_mae_db": 0.1907672250161966, "300–1000 Hz_40ms_log_rms_mae_db": 0.6181706447692895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0307875259032639, "300–1000 Hz_40ms_envelope_correlation": 0.9991479459679542, "300–1000 Hz_40ms_modulation_mae_db": 0.5329516744883342, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1836015344775759, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1048871791564186, "1000–3000 Hz_10ms_envelope_correlation": 0.9919466820042254, "1000–3000 Hz_10ms_modulation_mae_db": 0.5991466162620102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7814559418113013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06081267432408007, "1000–3000 Hz_40ms_envelope_correlation": 0.9975640544585335, "1000–3000 Hz_40ms_modulation_mae_db": 0.7630877750829894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3334584414765662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23426832052756774, "3000–6000 Hz_10ms_envelope_correlation": 0.9571803122802671, "3000–6000 Hz_10ms_modulation_mae_db": 1.0567352047598675, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9503471793952956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16027552493872405, "3000–6000 Hz_40ms_envelope_correlation": 0.9785736428332329, "3000–6000 Hz_40ms_modulation_mae_db": 1.1737476487062453, "mr_stft_db": 5.356092837034371, "band_mae_db": 0.14861145660636832, "rms_gain_error_db": -0.054523093494747404, "output_peak": 0.3581489295530481}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5918450018099838, "全帯域_10ms_rms_envelope_nrmse": 0.058971032042712805, "全帯域_10ms_envelope_correlation": 0.9964875209298478, "全帯域_10ms_modulation_mae_db": 0.5753380141367997, "全帯域_40ms_log_rms_mae_db": 0.3344694664266347, "全帯域_40ms_rms_envelope_nrmse": 0.03009202451705961, "全帯域_40ms_envelope_correlation": 0.9991690589591932, "全帯域_40ms_modulation_mae_db": 0.6078163944970821, "300–1000 Hz_10ms_log_rms_mae_db": 1.1381217344137493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06958239856225894, "300–1000 Hz_10ms_envelope_correlation": 0.9960702715404622, "300–1000 Hz_10ms_modulation_mae_db": 0.6370004106688206, "300–1000 Hz_40ms_log_rms_mae_db": 0.6217608336451963, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03902284140695989, "300–1000 Hz_40ms_envelope_correlation": 0.9989143392377019, "300–1000 Hz_40ms_modulation_mae_db": 0.6810877999432566, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2458112486913424, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10645067181956908, "1000–3000 Hz_10ms_envelope_correlation": 0.9915191762383292, "1000–3000 Hz_10ms_modulation_mae_db": 0.7646790125096788, "1000–3000 Hz_40ms_log_rms_mae_db": 0.828484078096459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06392716338511031, "1000–3000 Hz_40ms_envelope_correlation": 0.9970399139352187, "1000–3000 Hz_40ms_modulation_mae_db": 0.6203239672026616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3532994001865308, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.259024503739175, "3000–6000 Hz_10ms_envelope_correlation": 0.9480064510880027, "3000–6000 Hz_10ms_modulation_mae_db": 1.3774176250260048, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9056271187972508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16589296760554062, "3000–6000 Hz_40ms_envelope_correlation": 0.977158725346322, "3000–6000 Hz_40ms_modulation_mae_db": 1.5837481563914826, "mr_stft_db": 5.357126389470195, "band_mae_db": 0.14638618853356009, "rms_gain_error_db": -0.1084864098347974, "output_peak": 0.33385972755247895}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5706135489202202, "全帯域_10ms_rms_envelope_nrmse": 0.05151351954844255, "全帯域_10ms_envelope_correlation": 0.997276535512425, "全帯域_10ms_modulation_mae_db": 0.4267162514120959, "全帯域_40ms_log_rms_mae_db": 0.2961273341871031, "全帯域_40ms_rms_envelope_nrmse": 0.02777753313613861, "全帯域_40ms_envelope_correlation": 0.9992193609237151, "全帯域_40ms_modulation_mae_db": 0.4013651395068962, "300–1000 Hz_10ms_log_rms_mae_db": 1.1322510293966148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06579083945856905, "300–1000 Hz_10ms_envelope_correlation": 0.9963083627208228, "300–1000 Hz_10ms_modulation_mae_db": 0.31547514083658307, "300–1000 Hz_40ms_log_rms_mae_db": 0.6417461958780816, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03457212633666416, "300–1000 Hz_40ms_envelope_correlation": 0.9989669924003333, "300–1000 Hz_40ms_modulation_mae_db": 0.43066138590268416, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2279317650803583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10315463581837, "1000–3000 Hz_10ms_envelope_correlation": 0.9924024035871767, "1000–3000 Hz_10ms_modulation_mae_db": 0.39762024008688307, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7666727384644165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06053574705947612, "1000–3000 Hz_40ms_envelope_correlation": 0.9978106650379004, "1000–3000 Hz_40ms_modulation_mae_db": 0.3850308405199548, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3377567743609002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25199363571366207, "3000–6000 Hz_10ms_envelope_correlation": 0.9547454672218867, "3000–6000 Hz_10ms_modulation_mae_db": 2.033478219803248, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9164908790040824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17149925510051042, "3000–6000 Hz_40ms_envelope_correlation": 0.9785390513502581, "3000–6000 Hz_40ms_modulation_mae_db": 2.239967234299356, "mr_stft_db": 5.346940405985947, "band_mae_db": 0.13274751403464768, "rms_gain_error_db": -0.07578750706399279, "output_peak": 0.3530575874887741}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.107751413878713, "spectral_convergence": 0.20148408554819805, "spectral_convergence_status": "ok", "amplitude_floor": 1.1114078093309192e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1745525077353385, "spectral_convergence": 0.1970234158467014, "spectral_convergence_status": "ok", "amplitude_floor": 9.683415126458649e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.802257604988426, "spectral_convergence": 0.20620082436577553, "spectral_convergence_status": "ok", "amplitude_floor": 7.055286564887976e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011089954817882694, "output_power": 0.00011209265269671315, "delta_db": 0.04647370185735806, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007963949709957278, "output_power": 0.0007684397295011737, "delta_db": -0.15518698465314648, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005473467336544822, "output_power": 0.000528291099759705, "delta_db": -0.15389236642313864, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020656259415557, "output_power": 0.0001951888554210331, "delta_db": -0.24596661734794023, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.567254948983353e-05, "output_power": 8.950906710110629e-05, "delta_db": -0.28920316468169865, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.697083983250908e-05, "output_power": 3.323746105266876e-05, "delta_db": -0.4623147471490988, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22550626368539683, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16553761406024628, "pitch": null, "output_rms_dbfs": -27.627683055595934, "output_peak": 0.34644950844564315, "reference_rms_dbfs": -27.46214544153569, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.361520508867493}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015211892717126925, "log_rms_mae_db": 0.6008830797100233, "rms_envelope_nrmse": 0.05635062607691683, "envelope_correlation": 0.9970038998889468, "rms_gain_error_db": -0.16488096149026482, "reference_mean_power": 0.0017946561941472319, "output_mean_power": 0.001727798792188305}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.098233390179555e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033819477659287355, "output_power": 0.00032068381970382624, "reference_active": true, "delta_db": -0.2308984775928341}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.872738107665891e-05, "output_power": 3.379967186164717e-05, "reference_active": true, "delta_db": -0.5910564496779027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.329707857377202e-06, "output_power": 7.142603681902153e-06, "reference_active": true, "delta_db": -0.11230111854195189}], "active_band_count": 3, "active_band_mae_db": 0.3114186819375629}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012855597503212083, "log_rms_mae_db": 0.3392548841105275, "rms_envelope_nrmse": 0.03342602392695288, "envelope_correlation": 0.999152338319421, "rms_gain_error_db": -0.16478706193144738, "reference_mean_power": 0.0017996791281982027, "output_mean_power": 0.0017326720658655368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.521732580259768e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030590519204748624, "output_power": 0.0002895228730200716, "reference_active": true, "delta_db": -0.23903968459422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7673272930744677e-05, "output_power": 1.5828739245906302e-05, "reference_active": true, "delta_db": -0.478706595546461}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.996266898550172e-07, "output_power": 3.327577468498942e-07, "reference_active": true, "delta_db": -0.795263107686749}], "active_band_count": 3, "active_band_mae_db": 0.5043364626091433}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013968182745477764, "log_rms_mae_db": 1.1262596561301175, "rms_envelope_nrmse": 0.06675917425033157, "envelope_correlation": 0.996450334579776, "rms_gain_error_db": -0.16997204123193926, "reference_mean_power": 0.0010910330804496425, "output_mean_power": 0.001049157560469446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.057825806205699e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031975947498638787, "output_power": 0.0003061090359533823, "reference_active": true, "delta_db": -0.18947272570878293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9465579420304694e-05, "output_power": 3.3549926388686194e-05, "reference_active": true, "delta_db": -0.7052691172502262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.299925023694004e-06, "output_power": 5.555688422668395e-06, "reference_active": true, "delta_db": -0.5459749948958346}], "active_band_count": 3, "active_band_mae_db": 0.4802389459516146}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012087571887341512, "log_rms_mae_db": 0.5799748384581929, "rms_envelope_nrmse": 0.03810144043542077, "envelope_correlation": 0.99902833708323, "rms_gain_error_db": -0.16969854081566527, "reference_mean_power": 0.0010942126232004053, "output_mean_power": 0.001052281333671866}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.627912926828161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002965811192999727, "output_power": 0.00028295519155323214, "reference_active": true, "delta_db": -0.2042583326614825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8971998770340482e-05, "output_power": 1.6379859354229016e-05, "reference_active": true, "delta_db": -0.6380291946332005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.113014123121411e-07, "output_power": 2.9499103414564844e-07, "reference_active": true, "delta_db": -1.4435138439254291}], "active_band_count": 3, "active_band_mae_db": 0.7619337904067041}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.190065013136077e-05, "log_rms_mae_db": 1.2362102214623814, "rms_envelope_nrmse": 0.11229139686507361, "envelope_correlation": 0.990920814820747, "rms_gain_error_db": -0.294056157349755, "reference_mean_power": 0.00025666306625944256, "output_mean_power": 0.00023985996312761513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2531384738435257e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.158137778593675e-05, "output_power": 6.488007712180425e-05, "reference_active": true, "delta_db": -0.42688695917648045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.324550913609082e-06, "output_power": 5.631626775171116e-06, "reference_active": true, "delta_db": 0.24350880951508663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7772043437261862e-06, "output_power": 2.7316454756947213e-06, "reference_active": true, "delta_db": 1.8668696807937604}], "active_band_count": 3, "active_band_mae_db": 0.8457551498284426}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.863962912697811e-05, "log_rms_mae_db": 0.8166648402704031, "rms_envelope_nrmse": 0.0661985996601192, "envelope_correlation": 0.9973062428182156, "rms_gain_error_db": -0.2938914817941729, "reference_mean_power": 0.00025765686496881694, "output_mean_power": 0.00024079783065289378}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.1689575138838273e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607374483640592e-05, "output_power": 5.955601342681154e-05, "reference_active": true, "delta_db": -0.4510330269240758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.394936823991012e-06, "output_power": 2.6183672915982583e-06, "reference_active": true, "delta_db": 0.38736505514489955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.833675392620637e-08, "output_power": 1.5787757505919458e-07, "reference_active": true, "delta_db": 2.0560457894400175}], "active_band_count": 3, "active_band_mae_db": 0.964814623836331}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.534983553210222e-05, "log_rms_mae_db": 1.3586321658740246, "rms_envelope_nrmse": 0.2288949057662453, "envelope_correlation": 0.9610486529409196, "rms_gain_error_db": 0.23332918576917008, "reference_mean_power": 1.9026430507347487e-05, "output_mean_power": 2.0076603303797862e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1975009508047096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.298926861031976e-06, "output_power": 5.842927942592531e-06, "reference_active": true, "delta_db": 0.4244260528119014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588034222302733e-06, "output_power": 1.6766892955808018e-06, "reference_active": true, "delta_db": 0.23592734412641977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.193680029927605e-07, "output_power": 8.622179209944666e-07, "reference_active": true, "delta_db": 3.130217537959786}], "active_band_count": 3, "active_band_mae_db": 1.2635236449660356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3831266720127666e-05, "log_rms_mae_db": 0.9346369879777614, "rms_envelope_nrmse": 0.1613821102924361, "envelope_correlation": 0.9797710013216037, "rms_gain_error_db": 0.23394920442055045, "reference_mean_power": 1.9104126133492904e-05, "output_mean_power": 2.016146551414581e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1095182205605233e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.38789012101126e-06, "output_power": 6.061870182238394e-06, "reference_active": true, "delta_db": 0.5118790124144276}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.477294263383285e-07, "output_power": 8.063212075440674e-07, "reference_active": true, "delta_db": 0.3276361027165694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.443035434026362e-08, "output_power": 5.098249454305657e-08, "reference_active": true, "delta_db": 3.194913153541359}], "active_band_count": 3, "active_band_mae_db": 1.3448094228907852}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9226797699101724, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a043abb0308e4a599831004a22c389619d1163086aa56fe148d2b60234b201e2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9921998539725934, "standard_deviation": 0.0007512698635203247, "minimum": 0.990920814820747, "maximum": 0.9933092632067632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2192919039482915, "standard_deviation": 0.01881552702267442, "minimum": 1.1836015344775759, "maximum": 1.2458112486913424, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.579410537328532, "standard_deviation": 0.17786562174612486, "minimum": 0.3624613301405069, "maximum": 0.8457551498284426, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10338284648954939, "standard_deviation": 0.004832548069860298, "minimum": 0.096805709005671, "maximum": 0.11229139686507361, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975416316757331, "standard_deviation": 0.00031220509405875544, "minimum": 0.9970399139352187, "maximum": 0.9980600720463982, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8063122405288177, "standard_deviation": 0.024103317804537436, "minimum": 0.7666727384644165, "maximum": 0.828484078096459, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5415296614309402, "standard_deviation": 0.250620690111806, "minimum": 0.18295226891110114, "maximum": 0.964814623836331, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.061270427303646444, "standard_deviation": 0.0028614150254113115, "minimum": 0.0567710163630094, "maximum": 0.0661985996601192, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9550118992394505, "standard_deviation": 0.005746963888071818, "minimum": 0.9480064510880027, "maximum": 0.9623222208962469, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3338057516180157, "standard_deviation": 0.01969358564825565, "minimum": 1.3073598361040157, "maximum": 1.3586321658740246, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3801817680295365, "standard_deviation": 0.29792748660540286, "minimum": 1.0567352047598675, "maximum": 2.033478219803248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24083445705326045, "standard_deviation": 0.014055299513242676, "minimum": 0.2207769489315195, "maximum": 0.259024503739175, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9784987545303092, "standard_deviation": 0.0022263368714134394, "minimum": 0.9747684494570564, "maximum": 0.982521817092727, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9156474883851474, "standard_deviation": 0.024198476110878362, "minimum": 0.8868666827297168, "maximum": 0.9503471793952956, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.547608560517026, "standard_deviation": 0.31702925314165165, "minimum": 1.1737476487062453, "maximum": 2.239967234299356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16185535627278158, "standard_deviation": 0.008189657428514505, "minimum": 0.14566230107030084, "maximum": 0.17149925510051042, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.996279925018365, "standard_deviation": 0.00034977912917628353, "minimum": 0.995795621922568, "maximum": 0.9969992080897935, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1295215491825326, "standard_deviation": 0.02004748848842457, "minimum": 1.09720391047941, "maximum": 1.1610283799615504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3376612462580757, "standard_deviation": 0.16809826274970924, "minimum": 0.08148395470912599, "maximum": 0.6370004106688206, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06646450265298459, "standard_deviation": 0.003353784348221633, "minimum": 0.05909213058287633, "maximum": 0.07008007050706955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989220360782853, "standard_deviation": 0.00015617257256657696, "minimum": 0.9986835933895754, "maximum": 0.9991479459679542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6191217888772769, "standard_deviation": 0.021767250541536123, "minimum": 0.5799748384581929, "maximum": 0.6454176579739044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.42096355512805983, "standard_deviation": 0.22710841575690663, "minimum": 0.10464259974600297, "maximum": 0.7619337904067041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03614032563659465, "standard_deviation": 0.0030398682714130173, "minimum": 0.0307875259032639, "maximum": 0.03939487522654562, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15486690510128528, "standard_deviation": 0.05149866525836371, "minimum": 0.08342218439514194, "maximum": 0.2373127817797219, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.355824893532077, "standard_deviation": 0.008103804570075335, "minimum": 5.343445972442652, "maximum": 5.367999991940944, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.35404361991309824, "standard_deviation": 0.016782267374733366, "minimum": 0.3312770584100516, "maximum": 0.3782456555931453, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08045345578081654, "standard_deviation": 0.049944478525294565, "minimum": -0.16553761406024628, "maximum": -0.001023797975474682, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9969775306061024, "standard_deviation": 0.00032640982781356027, "minimum": 0.9964875209298478, "maximum": 0.9975353911891811, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5781019218689919, "standard_deviation": 0.023582563692504815, "minimum": 0.5475280489581728, "maximum": 0.6089376415692107, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2909801535173582, "standard_deviation": 0.1452490321406966, "minimum": 0.14191782525337313, "maximum": 0.5753380141367997, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05445387141948061, "standard_deviation": 0.0032944694683404024, "minimum": 0.04875956483460422, "maximum": 0.058971032042712805, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991507182339453, "standard_deviation": 0.00011203047286947757, "minimum": 0.9989085396047761, "maximum": 0.9992803579818932, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.31549826700507755, "standard_deviation": 0.022495615500471668, "minimum": 0.28669935633293275, "maximum": 0.34869183696895817, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3519261419806794, "standard_deviation": 0.15452358655662624, "minimum": 0.13599473936930812, "maximum": 0.6078163944970821, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.029445084215822235, "standard_deviation": 0.002918038011691525, "minimum": 0.02615171720049917, "maximum": 0.03402018082241928, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0017", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5426416397094727, "predicted_mos": 2.5426416397094727, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9efd7f87803e3e3c9afae964e5fcf0240d7f9e0ae95534f46a91fdb7d44373dc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.83354360004887, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5690976536274484, "全帯域_10ms_rms_envelope_nrmse": 0.059604726513707795, "全帯域_10ms_envelope_correlation": 0.9963380100865994, "全帯域_10ms_modulation_mae_db": 0.21774092716187274, "全帯域_40ms_log_rms_mae_db": 0.31454784210920284, "全帯域_40ms_rms_envelope_nrmse": 0.036386927619671716, "全帯域_40ms_envelope_correlation": 0.9986193963423889, "全帯域_40ms_modulation_mae_db": 0.35645372584240215, "300–1000 Hz_10ms_log_rms_mae_db": 1.1580070620273815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07495212075552024, "300–1000 Hz_10ms_envelope_correlation": 0.9951719781096463, "300–1000 Hz_10ms_modulation_mae_db": 0.1438726029231791, "300–1000 Hz_40ms_log_rms_mae_db": 0.6332682166959751, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041766875657414285, "300–1000 Hz_40ms_envelope_correlation": 0.9984392465935079, "300–1000 Hz_40ms_modulation_mae_db": 0.4424386216453131, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3031956648253689, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09962740673219726, "1000–3000 Hz_10ms_envelope_correlation": 0.9929531191687849, "1000–3000 Hz_10ms_modulation_mae_db": 0.8339725568936084, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8310357274804072, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06396391873046266, "1000–3000 Hz_40ms_envelope_correlation": 0.9974479145820375, "1000–3000 Hz_40ms_modulation_mae_db": 0.6634873965615836, "3000–6000 Hz_10ms_log_rms_mae_db": 1.616552787015298, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27184950964503485, "3000–6000 Hz_10ms_envelope_correlation": 0.9394956249138855, "3000–6000 Hz_10ms_modulation_mae_db": 1.4570674889270858, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1290308845908865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18776437979370664, "3000–6000 Hz_40ms_envelope_correlation": 0.9694212568035159, "3000–6000 Hz_40ms_modulation_mae_db": 1.4015255225561587, "mr_stft_db": 5.570189631430492, "band_mae_db": 0.1770778149010018, "rms_gain_error_db": -0.08359666457217454, "output_peak": 0.3582976150740998}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.558382860388747, "全帯域_10ms_rms_envelope_nrmse": 0.055257967717101716, "全帯域_10ms_envelope_correlation": 0.996902354027613, "全帯域_10ms_modulation_mae_db": 0.2768503818217827, "全帯域_40ms_log_rms_mae_db": 0.29712667269827897, "全帯域_40ms_rms_envelope_nrmse": 0.03203137008124266, "全帯域_40ms_envelope_correlation": 0.9989958749089985, "全帯域_40ms_modulation_mae_db": 0.30471773584308753, "300–1000 Hz_10ms_log_rms_mae_db": 1.1807079688475275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0687838295888032, "300–1000 Hz_10ms_envelope_correlation": 0.9959553700632892, "300–1000 Hz_10ms_modulation_mae_db": 0.33024678752355247, "300–1000 Hz_40ms_log_rms_mae_db": 0.6186106595348855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03651424422375067, "300–1000 Hz_40ms_envelope_correlation": 0.9988382977869509, "300–1000 Hz_40ms_modulation_mae_db": 0.5114889174274352, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3171443553934057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09673357563291414, "1000–3000 Hz_10ms_envelope_correlation": 0.9933565026053102, "1000–3000 Hz_10ms_modulation_mae_db": 0.45994315049598206, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8361000404438894, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058964893303686386, "1000–3000 Hz_40ms_envelope_correlation": 0.997908830464196, "1000–3000 Hz_40ms_modulation_mae_db": 0.4078876518892738, "3000–6000 Hz_10ms_log_rms_mae_db": 1.593596685961318, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.265156989025707, "3000–6000 Hz_10ms_envelope_correlation": 0.9437963975898217, "3000–6000 Hz_10ms_modulation_mae_db": 1.568542944350982, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0826888742754328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16771347862938593, "3000–6000 Hz_40ms_envelope_correlation": 0.9758074959287659, "3000–6000 Hz_40ms_modulation_mae_db": 1.7506647835288074, "mr_stft_db": 5.565195566131007, "band_mae_db": 0.1550363802037413, "rms_gain_error_db": -0.09947936561890254, "output_peak": 0.34154271630631583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5978922489050813, "全帯域_10ms_rms_envelope_nrmse": 0.0524318089952937, "全帯域_10ms_envelope_correlation": 0.9972605055185251, "全帯域_10ms_modulation_mae_db": 0.5726668509591684, "全帯域_40ms_log_rms_mae_db": 0.3263780364368797, "全帯域_40ms_rms_envelope_nrmse": 0.0276663516945239, "全帯域_40ms_envelope_correlation": 0.9993227231383304, "全帯域_40ms_modulation_mae_db": 0.6336353864889507, "300–1000 Hz_10ms_log_rms_mae_db": 1.209962250691798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06751422427404154, "300–1000 Hz_10ms_envelope_correlation": 0.9962296287454195, "300–1000 Hz_10ms_modulation_mae_db": 0.43859606239423315, "300–1000 Hz_40ms_log_rms_mae_db": 0.6485765610649953, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035558274313797475, "300–1000 Hz_40ms_envelope_correlation": 0.999051293906054, "300–1000 Hz_40ms_modulation_mae_db": 0.6433727605056797, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3266363393735856, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10383634261807098, "1000–3000 Hz_10ms_envelope_correlation": 0.991965444731155, "1000–3000 Hz_10ms_modulation_mae_db": 0.2201266946429524, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8424263509760636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06117853947731777, "1000–3000 Hz_40ms_envelope_correlation": 0.9973410214370594, "1000–3000 Hz_40ms_modulation_mae_db": 0.18952304952733526, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6015032485621217, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2748488451842592, "3000–6000 Hz_10ms_envelope_correlation": 0.9390103154010306, "3000–6000 Hz_10ms_modulation_mae_db": 1.4318819644232093, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0801176316403982, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16690133536941407, "3000–6000 Hz_40ms_envelope_correlation": 0.9759012726641575, "3000–6000 Hz_40ms_modulation_mae_db": 1.476088297169978, "mr_stft_db": 5.555820981230924, "band_mae_db": 0.19143946552270008, "rms_gain_error_db": -0.11223069730698469, "output_peak": 0.3532097712634009}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.590632855572886, "全帯域_10ms_rms_envelope_nrmse": 0.050045038919254894, "全帯域_10ms_envelope_correlation": 0.9974128462986973, "全帯域_10ms_modulation_mae_db": 0.17469774605774804, "全帯域_40ms_log_rms_mae_db": 0.30877901297867877, "全帯域_40ms_rms_envelope_nrmse": 0.023271528878091204, "全帯域_40ms_envelope_correlation": 0.9994532131413696, "全帯域_40ms_modulation_mae_db": 0.3863481169565827, "300–1000 Hz_10ms_log_rms_mae_db": 1.1453535119622786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.059832394715252354, "300–1000 Hz_10ms_envelope_correlation": 0.9969435818944313, "300–1000 Hz_10ms_modulation_mae_db": 0.173257025091347, "300–1000 Hz_40ms_log_rms_mae_db": 0.6302547322398193, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03122172263868969, "300–1000 Hz_40ms_envelope_correlation": 0.9991482984119303, "300–1000 Hz_40ms_modulation_mae_db": 0.41414303161543703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3566351073706124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11306790964576113, "1000–3000 Hz_10ms_envelope_correlation": 0.9904024233634946, "1000–3000 Hz_10ms_modulation_mae_db": 0.31403500383114563, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9039392095014998, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06255230389415341, "1000–3000 Hz_40ms_envelope_correlation": 0.997220265632266, "1000–3000 Hz_40ms_modulation_mae_db": 0.23115029729103673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6208578599151116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2693745998851891, "3000–6000 Hz_10ms_envelope_correlation": 0.9406169939407724, "3000–6000 Hz_10ms_modulation_mae_db": 1.637588812222876, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0986364230630103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1653448381288208, "3000–6000 Hz_40ms_envelope_correlation": 0.9764560180487366, "3000–6000 Hz_40ms_modulation_mae_db": 1.632039355886743, "mr_stft_db": 5.557165433876115, "band_mae_db": 0.10058347419199302, "rms_gain_error_db": -0.06390899858959256, "output_peak": 0.3446098021582987}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5639419729483318, "全帯域_10ms_rms_envelope_nrmse": 0.048754109213226204, "全帯域_10ms_envelope_correlation": 0.997621396331392, "全帯域_10ms_modulation_mae_db": 0.36757324067575253, "全帯域_40ms_log_rms_mae_db": 0.31366156596052536, "全帯域_40ms_rms_envelope_nrmse": 0.028473706332189107, "全帯域_40ms_envelope_correlation": 0.9992361084697561, "全帯域_40ms_modulation_mae_db": 0.44458029963481827, "300–1000 Hz_10ms_log_rms_mae_db": 1.1712099596012473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06170908920911659, "300–1000 Hz_10ms_envelope_correlation": 0.9968414853976116, "300–1000 Hz_10ms_modulation_mae_db": 0.26836498425421634, "300–1000 Hz_40ms_log_rms_mae_db": 0.6025178033142126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03454134035733515, "300–1000 Hz_40ms_envelope_correlation": 0.9990651429933118, "300–1000 Hz_40ms_modulation_mae_db": 0.515939236469792, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3084440039703962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10680489450976823, "1000–3000 Hz_10ms_envelope_correlation": 0.9915413221199489, "1000–3000 Hz_10ms_modulation_mae_db": 0.3982879084648898, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8574330762508192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07000070217542939, "1000–3000 Hz_40ms_envelope_correlation": 0.9964869428295398, "1000–3000 Hz_40ms_modulation_mae_db": 0.4119921939110913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5782936079371874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25587901566710414, "3000–6000 Hz_10ms_envelope_correlation": 0.9477714833103431, "3000–6000 Hz_10ms_modulation_mae_db": 1.1106391530116595, "3000–6000 Hz_40ms_log_rms_mae_db": 1.114499986636222, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.166396773400177, "3000–6000 Hz_40ms_envelope_correlation": 0.9762832668704141, "3000–6000 Hz_40ms_modulation_mae_db": 1.241970452713706, "mr_stft_db": 5.561290387635733, "band_mae_db": 0.16862142956166296, "rms_gain_error_db": -0.10070011037064941, "output_peak": 0.3819226430463125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.561065085091957, "全帯域_10ms_rms_envelope_nrmse": 0.05186943620667234, "全帯域_10ms_envelope_correlation": 0.9972552283975087, "全帯域_10ms_modulation_mae_db": 0.21439948621787974, "全帯域_40ms_log_rms_mae_db": 0.2963791640131692, "全帯域_40ms_rms_envelope_nrmse": 0.029763594823412877, "全帯域_40ms_envelope_correlation": 0.9991150981000232, "全帯域_40ms_modulation_mae_db": 0.22992054856997854, "300–1000 Hz_10ms_log_rms_mae_db": 1.1886624629670306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0642548951751155, "300–1000 Hz_10ms_envelope_correlation": 0.9965929240220102, "300–1000 Hz_10ms_modulation_mae_db": 0.3642597429815555, "300–1000 Hz_40ms_log_rms_mae_db": 0.6506568650885545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03839373476952483, "300–1000 Hz_40ms_envelope_correlation": 0.9988416532433226, "300–1000 Hz_40ms_modulation_mae_db": 0.304137146639532, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3184318825583852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10863071367563024, "1000–3000 Hz_10ms_envelope_correlation": 0.9910232542496175, "1000–3000 Hz_10ms_modulation_mae_db": 0.7617996227495304, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8702397269252063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06059114321177296, "1000–3000 Hz_40ms_envelope_correlation": 0.9971941412588765, "1000–3000 Hz_40ms_modulation_mae_db": 0.5330286662293382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5475115475251986, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2578843519941891, "3000–6000 Hz_10ms_envelope_correlation": 0.9473355221996369, "3000–6000 Hz_10ms_modulation_mae_db": 1.3631297706160452, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0742496363660365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16554409365200967, "3000–6000 Hz_40ms_envelope_correlation": 0.9767038030047897, "3000–6000 Hz_40ms_modulation_mae_db": 1.620516753670021, "mr_stft_db": 5.560129865507843, "band_mae_db": 0.11457929456707654, "rms_gain_error_db": -0.08378669318834273, "output_peak": 0.3223040680518008}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5797220506421308, "全帯域_10ms_rms_envelope_nrmse": 0.049337584502934746, "全帯域_10ms_envelope_correlation": 0.9974909502884062, "全帯域_10ms_modulation_mae_db": 0.35453938789767275, "全帯域_40ms_log_rms_mae_db": 0.3274090896898417, "全帯域_40ms_rms_envelope_nrmse": 0.027794471736862413, "全帯域_40ms_envelope_correlation": 0.9991970851940126, "全帯域_40ms_modulation_mae_db": 0.24269999525244537, "300–1000 Hz_10ms_log_rms_mae_db": 1.1886852176282494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.062392539599074634, "300–1000 Hz_10ms_envelope_correlation": 0.9966827827321836, "300–1000 Hz_10ms_modulation_mae_db": 0.4502535176094362, "300–1000 Hz_40ms_log_rms_mae_db": 0.6530543224984051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034446085171174264, "300–1000 Hz_40ms_envelope_correlation": 0.9989667047758966, "300–1000 Hz_40ms_modulation_mae_db": 0.2039566189442453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2823761356052292, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10161250321769846, "1000–3000 Hz_10ms_envelope_correlation": 0.9925940687077237, "1000–3000 Hz_10ms_modulation_mae_db": 0.6523161241290016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8486146863521986, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0571812341561281, "1000–3000 Hz_40ms_envelope_correlation": 0.9980645987753219, "1000–3000 Hz_40ms_modulation_mae_db": 0.5977766779632376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6324072633209292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28340670094380843, "3000–6000 Hz_10ms_envelope_correlation": 0.9413049844273949, "3000–6000 Hz_10ms_modulation_mae_db": 1.5561800256781335, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0884768236834645, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18136173327548658, "3000–6000 Hz_40ms_envelope_correlation": 0.9752850604969098, "3000–6000 Hz_40ms_modulation_mae_db": 1.8011062146961139, "mr_stft_db": 5.558700316168121, "band_mae_db": 0.12636085785890652, "rms_gain_error_db": -0.06547420446140584, "output_peak": 0.33803031826896324}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5780458142462561, "全帯域_10ms_rms_envelope_nrmse": 0.047615476474997705, "全帯域_10ms_envelope_correlation": 0.9976858136832726, "全帯域_10ms_modulation_mae_db": 0.1945391551992853, "全帯域_40ms_log_rms_mae_db": 0.3016321293425626, "全帯域_40ms_rms_envelope_nrmse": 0.026097638176880253, "全帯域_40ms_envelope_correlation": 0.9993242880025193, "全帯域_40ms_modulation_mae_db": 0.26652728461420777, "300–1000 Hz_10ms_log_rms_mae_db": 1.204350904598411, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.059916836277491926, "300–1000 Hz_10ms_envelope_correlation": 0.9969664665039039, "300–1000 Hz_10ms_modulation_mae_db": 0.22471052691197835, "300–1000 Hz_40ms_log_rms_mae_db": 0.6453740739223846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03527860981945467, "300–1000 Hz_40ms_envelope_correlation": 0.9989455287339156, "300–1000 Hz_40ms_modulation_mae_db": 0.17271887604986058, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3324221650631023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0995947295734567, "1000–3000 Hz_10ms_envelope_correlation": 0.9924531344522209, "1000–3000 Hz_10ms_modulation_mae_db": 0.25242251032096336, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8269638311494493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055091476253044155, "1000–3000 Hz_40ms_envelope_correlation": 0.9976703749025614, "1000–3000 Hz_40ms_modulation_mae_db": 0.1723126361623005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5459686408328, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2555435213486735, "3000–6000 Hz_10ms_envelope_correlation": 0.9470106539184056, "3000–6000 Hz_10ms_modulation_mae_db": 1.4237820427317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0314025814636734, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16608779940457158, "3000–6000 Hz_40ms_envelope_correlation": 0.9761027710854869, "3000–6000 Hz_40ms_modulation_mae_db": 1.4557227881604542, "mr_stft_db": 5.553712869068508, "band_mae_db": 0.16381154269359802, "rms_gain_error_db": -0.07648412469905443, "output_peak": 0.33766308172386317}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.322195771694457, "spectral_convergence": 0.20289365080716454, "spectral_convergence_status": "ok", "amplitude_floor": 1.1114078093309192e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.436149305593608, "spectral_convergence": 0.20178261576358028, "spectral_convergence_status": "ok", "amplitude_floor": 9.683415126458649e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.952223817003412, "spectral_convergence": 0.2100399390082538, "spectral_convergence_status": "ok", "amplitude_floor": 7.055286564887976e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011089954817882694, "output_power": 0.00011309798297199234, "delta_db": 0.08525082865819765, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007963949709957278, "output_power": 0.0007834186015008722, "delta_db": -0.07134629692014827, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005473467336544822, "output_power": 0.000542412384265716, "delta_db": -0.03932934024306486, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020656259415557, "output_power": 0.0001976827337070563, "delta_db": -0.19082940936672496, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.567254948983353e-05, "output_power": 8.903248318622084e-05, "delta_db": -0.31238860870378843, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.697083983250908e-05, "output_power": 3.400377044118015e-05, "delta_db": -0.3633224055140865, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1770778149010018, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08359666457217454, "pitch": null, "output_rms_dbfs": -27.545742106107866, "output_peak": 0.3582976150740998, "reference_rms_dbfs": -27.46214544153569, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.570189631430492}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015211892717126925, "log_rms_mae_db": 0.5690976536274484, "rms_envelope_nrmse": 0.059604726513707795, "envelope_correlation": 0.9963380100865994, "rms_gain_error_db": -0.08363162708563932, "reference_mean_power": 0.0017946561941472319, "output_mean_power": 0.0017604273190905833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.098233390179555e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033819477659287355, "output_power": 0.00032674185496777894, "reference_active": true, "delta_db": -0.14962125369924623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.872738107665891e-05, "output_power": 3.6089108041887516e-05, "reference_active": true, "delta_db": -0.3064198057815654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.329707857377202e-06, "output_power": 7.004360544130812e-06, "reference_active": true, "delta_db": -0.1971817220048065}], "active_band_count": 3, "active_band_mae_db": 0.21774092716187274}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012855597503212083, "log_rms_mae_db": 0.31454784210920284, "rms_envelope_nrmse": 0.036386927619671716, "envelope_correlation": 0.9986193963423889, "rms_gain_error_db": -0.08381368770799576, "reference_mean_power": 0.0017996791281982027, "output_mean_power": 0.0017652804485207097}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.521732580259768e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030590519204748624, "output_power": 0.00029590100767610584, "reference_active": true, "delta_db": -0.14440404080675823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7673272930744677e-05, "output_power": 1.695016707296056e-05, "reference_active": true, "delta_db": -0.18143001098886608}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.996266898550172e-07, "output_power": 3.367454785176795e-07, "reference_active": true, "delta_db": -0.7435271257315822}], "active_band_count": 3, "active_band_mae_db": 0.35645372584240215}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013968182745477764, "log_rms_mae_db": 1.1580070620273815, "rms_envelope_nrmse": 0.07495212075552024, "envelope_correlation": 0.9951719781096463, "rms_gain_error_db": -0.04972163362535634, "reference_mean_power": 0.0010910330804496425, "output_mean_power": 0.0010786132608758235}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.057825806205699e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031975947498638787, "output_power": 0.0003143428779932262, "reference_active": true, "delta_db": -0.07419797113570287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9465579420304694e-05, "output_power": 3.671310682308521e-05, "reference_active": true, "delta_db": -0.31397345402335713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.299925023694004e-06, "output_power": 6.2372154001181706e-06, "reference_active": true, "delta_db": -0.04344638361047731}], "active_band_count": 3, "active_band_mae_db": 0.1438726029231791}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012087571887341512, "log_rms_mae_db": 0.6332682166959751, "rms_envelope_nrmse": 0.041766875657414285, "envelope_correlation": 0.9984392465935079, "rms_gain_error_db": -0.04969459514740543, "reference_mean_power": 0.0010942126232004053, "output_mean_power": 0.0010817633440361205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.627912926828161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002965811192999727, "output_power": 0.00029173331832532353, "reference_active": true, "delta_db": -0.07157467951302013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8971998770340482e-05, "output_power": 1.7907016871322214e-05, "reference_active": true, "delta_db": -0.25089845043716924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.113014123121411e-07, "output_power": 3.263442215960001e-07, "reference_active": true, "delta_db": -1.00484273498575}], "active_band_count": 3, "active_band_mae_db": 0.4424386216453131}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.190065013136077e-05, "log_rms_mae_db": 1.3031956648253689, "rms_envelope_nrmse": 0.09962740673219726, "envelope_correlation": 0.9929531191687849, "rms_gain_error_db": -0.2753349172312679, "reference_mean_power": 0.00025666306625944256, "output_mean_power": 0.00024089616521570377}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2531384738435257e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.158137778593675e-05, "output_power": 6.236726622997239e-05, "reference_active": true, "delta_db": -0.5984334549952995}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.324550913609082e-06, "output_power": 4.704527973770518e-06, "reference_active": true, "delta_db": -0.5376692888499124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7772043437261862e-06, "output_power": 2.4339903873830176e-06, "reference_active": true, "delta_db": 1.3658149268356135}], "active_band_count": 3, "active_band_mae_db": 0.8339725568936084}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.863962912697811e-05, "log_rms_mae_db": 0.8310357274804072, "rms_envelope_nrmse": 0.06396391873046266, "envelope_correlation": 0.9974479145820375, "rms_gain_error_db": -0.2755056083932171, "reference_mean_power": 0.00025765686496881694, "output_mean_power": 0.00024181941008878598}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.1689575138838273e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607374483640592e-05, "output_power": 5.757922144478724e-05, "reference_active": true, "delta_db": -0.5976313356264241}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.394936823991012e-06, "output_power": 2.1632499209722288e-06, "reference_active": true, "delta_db": -0.4418736513145182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.833675392620637e-08, "output_power": 1.2240850349558788e-07, "reference_active": true, "delta_db": 0.9509572027438087}], "active_band_count": 3, "active_band_mae_db": 0.6634873965615836}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.534983553210222e-05, "log_rms_mae_db": 1.616552787015298, "rms_envelope_nrmse": 0.27184950964503485, "envelope_correlation": 0.9394956249138855, "rms_gain_error_db": -0.20435273973477333, "reference_mean_power": 1.9026430507347487e-05, "output_mean_power": 1.8151898115758334e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1975009508047096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.298926861031976e-06, "output_power": 4.993888195348706e-06, "reference_active": true, "delta_db": -0.2574911015227825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588034222302733e-06, "output_power": 1.4252697778711176e-06, "reference_active": true, "delta_db": -0.46962781042781776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.193680029927605e-07, "output_power": 9.70518117178262e-07, "reference_active": true, "delta_db": 3.6440835548306567}], "active_band_count": 3, "active_band_mae_db": 1.4570674889270858}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3831266720127666e-05, "log_rms_mae_db": 1.1290308845908865, "rms_envelope_nrmse": 0.18776437979370664, "envelope_correlation": 0.9694212568035159, "rms_gain_error_db": -0.2053789625976748, "reference_mean_power": 1.9104126133492904e-05, "output_mean_power": 1.8221716296467108e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1095182205605233e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.38789012101126e-06, "output_power": 5.225433067736935e-06, "reference_active": true, "delta_db": -0.1329644101338182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.477294263383285e-07, "output_power": 7.224542803862587e-07, "reference_active": true, "delta_db": -0.14934103804831933}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.443035434026362e-08, "output_power": 6.027772849684979e-08, "reference_active": true, "delta_db": 3.9222711194863384}], "active_band_count": 3, "active_band_mae_db": 1.4015255225561587}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.909632543799793, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a2eaae5f5134952bd850245bc62db9a82a0555c1c45c9691825dfc4fb690bd02", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9920361586747819, "standard_deviation": 0.00100101285605845, "minimum": 0.9904024233634946, "maximum": 0.9933565026053102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3181607067700107, "standard_deviation": 0.021894271513764635, "minimum": 1.2823761356052292, "maximum": 1.3566351073706124, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4866129464410092, "standard_deviation": 0.23543266197716278, "minimum": 0.2201266946429524, "maximum": 0.8339725568936084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10373850945068713, "standard_deviation": 0.005451667325095468, "minimum": 0.09673357563291414, "maximum": 0.11306790964576113, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974167612352323, "standard_deviation": 0.0004907115807194465, "minimum": 0.9964869428295398, "maximum": 0.9980645987753219, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8520940811349418, "standard_deviation": 0.02530401131502638, "minimum": 0.8269638311494493, "maximum": 0.9039392095014998, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4008948211918996, "standard_deviation": 0.18939368487573885, "minimum": 0.1723126361623005, "maximum": 0.6634873965615836, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.061190526400249357, "standard_deviation": 0.004563066383163838, "minimum": 0.055091476253044155, "maximum": 0.07000070217542939, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9432927469626613, "standard_deviation": 0.003671274925274916, "minimum": 0.9390103154010306, "maximum": 0.9477714833103431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5920864551337455, "standard_deviation": 0.03261712371333937, "minimum": 1.5459686408328, "maximum": 1.6324072633209292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.4436015252452115, "standard_deviation": 0.16211392875037897, "minimum": 1.1106391530116595, "maximum": 1.637588812222876, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2667429417117456, "standard_deviation": 0.010005933717604073, "minimum": 0.2555435213486735, "maximum": 0.28340670094380843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.975245118112847, "standard_deviation": 0.00239264125662855, "minimum": 0.9694212568035159, "maximum": 0.9767038030047897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0873878552148906, "standard_deviation": 0.029225297116877835, "minimum": 1.0314025814636734, "maximum": 1.1290308845908865, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.547454271047748, "standard_deviation": 0.18755948158386235, "minimum": 1.241970452713706, "maximum": 1.8011062146961139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17088930395669652, "standard_deviation": 0.008643572108697598, "minimum": 0.1653448381288208, "maximum": 0.18776437979370664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9964230271835619, "standard_deviation": 0.0006163230348404983, "minimum": 0.9951719781096463, "maximum": 0.9969664665039039, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1808674172904905, "standard_deviation": 0.022029167266713053, "minimum": 1.1453535119622786, "maximum": 1.209962250691798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2991951562111873, "standard_deviation": 0.11584029238958844, "minimum": 0.1438726029231791, "maximum": 0.4502535176094362, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.064919491199302, "standard_deviation": 0.005214791693396326, "minimum": 0.059832394715252354, "maximum": 0.07495212075552024, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989120208056113, "standard_deviation": 0.00021931248382133794, "minimum": 0.9984392465935079, "maximum": 0.9991482984119303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.635289154294904, "standard_deviation": 0.017774748323191918, "minimum": 0.6025178033142126, "maximum": 0.6530543224984051, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.40102440116216187, "standard_deviation": 0.16317127539631338, "minimum": 0.17271887604986058, "maximum": 0.6433727605056797, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03596511086889263, "standard_deviation": 0.0031014127948691766, "minimum": 0.03122172263868969, "maximum": 0.041766875657414285, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14968878243758504, "standard_deviation": 0.03222238003351661, "minimum": 0.10058347419199302, "maximum": 0.19143946552270008, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.560275631381092, "standard_deviation": 0.005326655832979654, "minimum": 5.553712869068508, "maximum": 5.570189631430492, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3471975019866319, "standard_deviation": 0.017722034652802476, "minimum": 0.3223040680518008, "maximum": 0.3819226430463125, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08570760735088834, "standard_deviation": 0.017302763332935556, "minimum": -0.11223069730698469, "maximum": -0.06390899858959256, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9972458880790018, "standard_deviation": 0.0004415929805259435, "minimum": 0.9963380100865994, "maximum": 0.9976858136832726, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5748475676778548, "standard_deviation": 0.014279150296914187, "minimum": 0.558382860388747, "maximum": 0.5978922489050813, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2966258969988953, "standard_deviation": 0.13269157320967315, "minimum": 0.17469774605774804, "maximum": 0.5726668509591684, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.051864518567898636, "standard_deviation": 0.003949196406615273, "minimum": 0.047615476474997705, "maximum": 0.059604726513707795, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991579734121748, "standard_deviation": 0.0002585077530889698, "minimum": 0.9986193963423889, "maximum": 0.9994532131413696, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3107391891536424, "standard_deviation": 0.012093664262712965, "minimum": 0.2963791640131692, "maximum": 0.3274090896898417, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35811038665030914, "standard_deviation": 0.1337276646739082, "minimum": 0.22992054856997854, "maximum": 0.6336353864889507, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028935698667859264, "standard_deviation": 0.003947040315388064, "minimum": 0.023271528878091204, "maximum": 0.036386927619671716, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0017", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.6656975746154785, "predicted_mos": 2.6656975746154785, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8f0bb98f34f9995d4eaae7d32652a385dad2c258aa0ce439a7db77783007b66a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.853550999891013, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5849909857348047, "全帯域_10ms_rms_envelope_nrmse": 0.051605022096425875, "全帯域_10ms_envelope_correlation": 0.9972666595646027, "全帯域_10ms_modulation_mae_db": 0.06487424940324528, "全帯域_40ms_log_rms_mae_db": 0.298951080568641, "全帯域_40ms_rms_envelope_nrmse": 0.02802756132592753, "全帯域_40ms_envelope_correlation": 0.9992055955420194, "全帯域_40ms_modulation_mae_db": 0.17345365371827745, "300–1000 Hz_10ms_log_rms_mae_db": 1.0690971358894026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06348069723712403, "300–1000 Hz_10ms_envelope_correlation": 0.9966332968191313, "300–1000 Hz_10ms_modulation_mae_db": 0.23615224598535756, "300–1000 Hz_40ms_log_rms_mae_db": 0.5899273889504009, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036228058524046466, "300–1000 Hz_40ms_envelope_correlation": 0.9989405770016395, "300–1000 Hz_40ms_modulation_mae_db": 0.2861250265191058, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1876341628493805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09639380826710024, "1000–3000 Hz_10ms_envelope_correlation": 0.9930480932140402, "1000–3000 Hz_10ms_modulation_mae_db": 0.7523739041040282, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7882436654498315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0520803724917533, "1000–3000 Hz_40ms_envelope_correlation": 0.9980709939297792, "1000–3000 Hz_40ms_modulation_mae_db": 0.49659347677498983, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2354970770270526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2714626946473786, "3000–6000 Hz_10ms_envelope_correlation": 0.945166964774997, "3000–6000 Hz_10ms_modulation_mae_db": 1.7378265550852419, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8512913214049628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1762862357869115, "3000–6000 Hz_40ms_envelope_correlation": 0.9757719945723747, "3000–6000 Hz_40ms_modulation_mae_db": 2.066716707924795, "mr_stft_db": 5.338254074687477, "band_mae_db": 0.11961884519649958, "rms_gain_error_db": -0.07673639146226659, "output_peak": 0.3300410933821262}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5693031279927364, "全帯域_10ms_rms_envelope_nrmse": 0.05869749866523727, "全帯域_10ms_envelope_correlation": 0.9964707834371415, "全帯域_10ms_modulation_mae_db": 0.4419455159984884, "全帯域_40ms_log_rms_mae_db": 0.3028515386138085, "全帯域_40ms_rms_envelope_nrmse": 0.03344111111740807, "全帯域_40ms_envelope_correlation": 0.9988687177202288, "全帯域_40ms_modulation_mae_db": 0.3249735359432851, "300–1000 Hz_10ms_log_rms_mae_db": 1.0894445312837604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0672262909879079, "300–1000 Hz_10ms_envelope_correlation": 0.9961904631684033, "300–1000 Hz_10ms_modulation_mae_db": 0.38658603357480753, "300–1000 Hz_40ms_log_rms_mae_db": 0.5601517474832647, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03536088126538468, "300–1000 Hz_40ms_envelope_correlation": 0.9989771696284308, "300–1000 Hz_40ms_modulation_mae_db": 0.23374514820396142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1548569875508636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11416854847240514, "1000–3000 Hz_10ms_envelope_correlation": 0.9901821952341198, "1000–3000 Hz_10ms_modulation_mae_db": 0.8559848174489533, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7683349165159598, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06673851452006684, "1000–3000 Hz_40ms_envelope_correlation": 0.9967158327966275, "1000–3000 Hz_40ms_modulation_mae_db": 0.6200205011859454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.215619560168866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28890498306704965, "3000–6000 Hz_10ms_envelope_correlation": 0.9449350339179342, "3000–6000 Hz_10ms_modulation_mae_db": 2.5597641805743074, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8579069022067709, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22099372796951125, "3000–6000 Hz_40ms_envelope_correlation": 0.9672754840917724, "3000–6000 Hz_40ms_modulation_mae_db": 3.215045344507056, "mr_stft_db": 5.3227798193985265, "band_mae_db": 0.15796179600554094, "rms_gain_error_db": -0.092037163869762, "output_peak": 0.33609124820921027}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5931800234969435, "全帯域_10ms_rms_envelope_nrmse": 0.05559761550057001, "全帯域_10ms_envelope_correlation": 0.9968137303599349, "全帯域_10ms_modulation_mae_db": 0.13423008955304988, "全帯域_40ms_log_rms_mae_db": 0.3329680803222327, "全帯域_40ms_rms_envelope_nrmse": 0.031341059365156416, "全帯域_40ms_envelope_correlation": 0.9989860208388708, "全帯域_40ms_modulation_mae_db": 0.345201911280439, "300–1000 Hz_10ms_log_rms_mae_db": 1.0748217074291535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.060735460035111453, "300–1000 Hz_10ms_envelope_correlation": 0.9968435769696994, "300–1000 Hz_10ms_modulation_mae_db": 0.4525337661223537, "300–1000 Hz_40ms_log_rms_mae_db": 0.6229822671794026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034130331619183024, "300–1000 Hz_40ms_envelope_correlation": 0.9989718660965335, "300–1000 Hz_40ms_modulation_mae_db": 0.49672864711226444, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1551705217867019, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1075338707571878, "1000–3000 Hz_10ms_envelope_correlation": 0.9914344361801887, "1000–3000 Hz_10ms_modulation_mae_db": 0.8639801944427244, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7853454913785883, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0644088825076892, "1000–3000 Hz_40ms_envelope_correlation": 0.997094824971612, "1000–3000 Hz_40ms_modulation_mae_db": 0.7977763954523316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.218403987663642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22957657625296563, "3000–6000 Hz_10ms_envelope_correlation": 0.9578202089762243, "3000–6000 Hz_10ms_modulation_mae_db": 1.3919250599764224, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8551015099754581, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15870909026497804, "3000–6000 Hz_40ms_envelope_correlation": 0.978301916871036, "3000–6000 Hz_40ms_modulation_mae_db": 1.598675393410297, "mr_stft_db": 5.313283153191726, "band_mae_db": 0.11168479743689459, "rms_gain_error_db": -0.07685582691823101, "output_peak": 0.36543156443718466}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5915808845601157, "全帯域_10ms_rms_envelope_nrmse": 0.05613023622748144, "全帯域_10ms_envelope_correlation": 0.9968239455997205, "全帯域_10ms_modulation_mae_db": 0.3920902740621224, "全帯域_40ms_log_rms_mae_db": 0.33590558157220385, "全帯域_40ms_rms_envelope_nrmse": 0.03212882759334957, "全帯域_40ms_envelope_correlation": 0.9990158343588047, "全帯域_40ms_modulation_mae_db": 0.5385829440207653, "300–1000 Hz_10ms_log_rms_mae_db": 1.091938681941802, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0685608717776225, "300–1000 Hz_10ms_envelope_correlation": 0.9960334790951381, "300–1000 Hz_10ms_modulation_mae_db": 0.3680766513119979, "300–1000 Hz_40ms_log_rms_mae_db": 0.6223316121623284, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03811315111760083, "300–1000 Hz_40ms_envelope_correlation": 0.9987933912268965, "300–1000 Hz_40ms_modulation_mae_db": 0.6116999774279156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1788240843603381, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09527187512368239, "1000–3000 Hz_10ms_envelope_correlation": 0.9934309176322331, "1000–3000 Hz_10ms_modulation_mae_db": 0.21243285254082225, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7755990615775225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060763880194609064, "1000–3000 Hz_40ms_envelope_correlation": 0.997537971684949, "1000–3000 Hz_40ms_modulation_mae_db": 0.2736193178114571, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2527092182658446, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2512850873667794, "3000–6000 Hz_10ms_envelope_correlation": 0.9485921642633183, "3000–6000 Hz_10ms_modulation_mae_db": 1.3944855863543546, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8939311355544308, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17165926492866607, "3000–6000 Hz_40ms_envelope_correlation": 0.9745720994825235, "3000–6000 Hz_40ms_modulation_mae_db": 1.5283927702273925, "mr_stft_db": 5.3183256318014225, "band_mae_db": 0.1664232827734526, "rms_gain_error_db": -0.1106339946800419, "output_peak": 0.3434630642411127}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5631882281745288, "全帯域_10ms_rms_envelope_nrmse": 0.047603372660318194, "全帯域_10ms_envelope_correlation": 0.9977266432953411, "全帯域_10ms_modulation_mae_db": 0.21095304916769134, "全帯域_40ms_log_rms_mae_db": 0.303245439351033, "全帯域_40ms_rms_envelope_nrmse": 0.02584699630868631, "全帯域_40ms_envelope_correlation": 0.9993832491282997, "全帯域_40ms_modulation_mae_db": 0.2259220452801348, "300–1000 Hz_10ms_log_rms_mae_db": 1.040880483294718, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05866592721991697, "300–1000 Hz_10ms_envelope_correlation": 0.9971419735983629, "300–1000 Hz_10ms_modulation_mae_db": 0.2091337322114731, "300–1000 Hz_40ms_log_rms_mae_db": 0.5870142050429049, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03354016355353817, "300–1000 Hz_40ms_envelope_correlation": 0.9991128140835298, "300–1000 Hz_40ms_modulation_mae_db": 0.3601959542456334, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1602029468969839, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10016372977004803, "1000–3000 Hz_10ms_envelope_correlation": 0.992715411823057, "1000–3000 Hz_10ms_modulation_mae_db": 0.8247862533876469, "1000–3000 Hz_40ms_log_rms_mae_db": 0.735670544280925, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05823767114504867, "1000–3000 Hz_40ms_envelope_correlation": 0.9978162047344629, "1000–3000 Hz_40ms_modulation_mae_db": 0.5595620044160464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.260351833912531, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22663706986260945, "3000–6000 Hz_10ms_envelope_correlation": 0.9591663777803682, "3000–6000 Hz_10ms_modulation_mae_db": 1.5067278292629354, "3000–6000 Hz_40ms_log_rms_mae_db": 0.894677935441472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16641370083386894, "3000–6000 Hz_40ms_envelope_correlation": 0.9762901480507926, "3000–6000 Hz_40ms_modulation_mae_db": 1.78517518510788, "mr_stft_db": 5.318917999707484, "band_mae_db": 0.14504032218032942, "rms_gain_error_db": -0.0928698279896194, "output_peak": 0.37581882937555244}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5925535831575192, "全帯域_10ms_rms_envelope_nrmse": 0.05208085627963523, "全帯域_10ms_envelope_correlation": 0.9971720995162905, "全帯域_10ms_modulation_mae_db": 0.3709837816227391, "全帯域_40ms_log_rms_mae_db": 0.3192351073499033, "全帯域_40ms_rms_envelope_nrmse": 0.028320487651629443, "全帯域_40ms_envelope_correlation": 0.9991367790794182, "全帯域_40ms_modulation_mae_db": 0.5462863606180685, "300–1000 Hz_10ms_log_rms_mae_db": 1.0865276939882078, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06153233241430172, "300–1000 Hz_10ms_envelope_correlation": 0.996746374320439, "300–1000 Hz_10ms_modulation_mae_db": 0.48994277995471824, "300–1000 Hz_40ms_log_rms_mae_db": 0.6308649318893061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035749420879764744, "300–1000 Hz_40ms_envelope_correlation": 0.9988498940721131, "300–1000 Hz_40ms_modulation_mae_db": 0.5321628803342272, "1000–3000 Hz_10ms_log_rms_mae_db": 1.199827576689775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10452982241450069, "1000–3000 Hz_10ms_envelope_correlation": 0.9919492737891527, "1000–3000 Hz_10ms_modulation_mae_db": 0.5176428999631744, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8100080448534673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06295481727080274, "1000–3000 Hz_40ms_envelope_correlation": 0.9972774958441019, "1000–3000 Hz_40ms_modulation_mae_db": 0.3588169662712502, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2784703337683516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2488311471555735, "3000–6000 Hz_10ms_envelope_correlation": 0.9514307627115076, "3000–6000 Hz_10ms_modulation_mae_db": 1.4654654693729128, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9065741652646934, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16085622775035843, "3000–6000 Hz_40ms_envelope_correlation": 0.978131525313167, "3000–6000 Hz_40ms_modulation_mae_db": 1.653763110726941, "mr_stft_db": 5.332781432028527, "band_mae_db": 0.09529046700920346, "rms_gain_error_db": -0.044093090363986474, "output_peak": 0.3439870332166889}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.571055981208849, "全帯域_10ms_rms_envelope_nrmse": 0.051575885017265156, "全帯域_10ms_envelope_correlation": 0.9973320859866854, "全帯域_10ms_modulation_mae_db": 0.17929740190461432, "全帯域_40ms_log_rms_mae_db": 0.2750545952106062, "全帯域_40ms_rms_envelope_nrmse": 0.02620296354715931, "全帯域_40ms_envelope_correlation": 0.9993892222048231, "全帯域_40ms_modulation_mae_db": 0.15513784502414546, "300–1000 Hz_10ms_log_rms_mae_db": 1.0806986361122586, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06361354290689426, "300–1000 Hz_10ms_envelope_correlation": 0.9965735127265646, "300–1000 Hz_10ms_modulation_mae_db": 0.2899190199121699, "300–1000 Hz_40ms_log_rms_mae_db": 0.5653597982119172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03264499419881413, "300–1000 Hz_40ms_envelope_correlation": 0.9991140215022947, "300–1000 Hz_40ms_modulation_mae_db": 0.42826430575747504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.122511757833132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09741895800766513, "1000–3000 Hz_10ms_envelope_correlation": 0.9933946064227033, "1000–3000 Hz_10ms_modulation_mae_db": 0.37634580347758173, "1000–3000 Hz_40ms_log_rms_mae_db": 0.775006499278827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061368558887875645, "1000–3000 Hz_40ms_envelope_correlation": 0.9978261572488022, "1000–3000 Hz_40ms_modulation_mae_db": 0.2870110318908672, "3000–6000 Hz_10ms_log_rms_mae_db": 1.265238212827146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23370456740404186, "3000–6000 Hz_10ms_envelope_correlation": 0.9562031797317223, "3000–6000 Hz_10ms_modulation_mae_db": 1.2882632189478476, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8672850706739423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15401481459699673, "3000–6000 Hz_40ms_envelope_correlation": 0.9795455773502104, "3000–6000 Hz_40ms_modulation_mae_db": 1.2795303245389626, "mr_stft_db": 5.3322152795442985, "band_mae_db": 0.15006936776154464, "rms_gain_error_db": -0.10424098043279093, "output_peak": 0.34774256483905325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6128942413320468, "全帯域_10ms_rms_envelope_nrmse": 0.048068725728426985, "全帯域_10ms_envelope_correlation": 0.997621639938559, "全帯域_10ms_modulation_mae_db": 0.11117396100593256, "全帯域_40ms_log_rms_mae_db": 0.348803228295378, "全帯域_40ms_rms_envelope_nrmse": 0.02601475318507023, "全帯域_40ms_envelope_correlation": 0.9993066047691289, "全帯域_40ms_modulation_mae_db": 0.08248367138529959, "300–1000 Hz_10ms_log_rms_mae_db": 1.0964982943522006, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05802942669731262, "300–1000 Hz_10ms_envelope_correlation": 0.9971175808051277, "300–1000 Hz_10ms_modulation_mae_db": 0.15380153364753932, "300–1000 Hz_40ms_log_rms_mae_db": 0.5951332578391829, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03143960759527628, "300–1000 Hz_40ms_envelope_correlation": 0.9991266180756369, "300–1000 Hz_40ms_modulation_mae_db": 0.1528927558217907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1612143631890812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09121894169267956, "1000–3000 Hz_10ms_envelope_correlation": 0.9942268336684923, "1000–3000 Hz_10ms_modulation_mae_db": 0.3080533359860031, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8201968869589679, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05629711607466662, "1000–3000 Hz_40ms_envelope_correlation": 0.9982039263206872, "1000–3000 Hz_40ms_modulation_mae_db": 0.36565574603999157, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2248756366973832, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22047598203386512, "3000–6000 Hz_10ms_envelope_correlation": 0.9618737478491384, "3000–6000 Hz_10ms_modulation_mae_db": 1.1104411165211776, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8215552575752765, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1489525773547636, "3000–6000 Hz_40ms_envelope_correlation": 0.9813536399703737, "3000–6000 Hz_40ms_modulation_mae_db": 1.1973209378213834, "mr_stft_db": 5.326504196817603, "band_mae_db": 0.13626373260560967, "rms_gain_error_db": -0.06656149303257719, "output_peak": 0.35187460969772416}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.023603997909685, "spectral_convergence": 0.19432268245323658, "spectral_convergence_status": "ok", "amplitude_floor": 1.1114078093309192e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.136304754840084, "spectral_convergence": 0.19094071458979187, "spectral_convergence_status": "ok", "amplitude_floor": 9.683415126458649e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.854853471312664, "spectral_convergence": 0.20471135379759045, "spectral_convergence_status": "ok", "amplitude_floor": 7.055286564887976e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011089954817882694, "output_power": 0.00011351193759469172, "delta_db": 0.10111760165002948, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007963949709957278, "output_power": 0.0007803548329956512, "delta_db": -0.08836384471553699, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005473467336544822, "output_power": 0.0005411586009299945, "delta_db": -0.049379654152369966, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020656259415557, "output_power": 0.00019997352978754035, "delta_db": -0.14079166602438345, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.567254948983353e-05, "output_power": 9.177696222715185e-05, "delta_db": -0.1805366857104622, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.697083983250908e-05, "output_power": 3.565389279151525e-05, "delta_db": -0.1575236189262155, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11961884519649958, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07673639146226659, "pitch": null, "output_rms_dbfs": -27.53888183299796, "output_peak": 0.3300410933821262, "reference_rms_dbfs": -27.46214544153569, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.338254074687477}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015211892717126925, "log_rms_mae_db": 0.5849909857348047, "rms_envelope_nrmse": 0.051605022096425875, "envelope_correlation": 0.9972666595646027, "rms_gain_error_db": -0.07611440183850354, "reference_mean_power": 0.0017946561941472319, "output_mean_power": 0.0017634770903500886}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.098233390179555e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033819477659287355, "output_power": 0.00033388907675991583, "reference_active": true, "delta_db": -0.05564684383491575}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.872738107665891e-05, "output_power": 3.783781401936263e-05, "reference_active": true, "delta_db": -0.10092090764089517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.329707857377202e-06, "output_power": 7.265761960579093e-06, "reference_active": true, "delta_db": -0.03805499673392494}], "active_band_count": 3, "active_band_mae_db": 0.06487424940324528}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012855597503212083, "log_rms_mae_db": 0.298951080568641, "rms_envelope_nrmse": 0.02802756132592753, "envelope_correlation": 0.9992055955420194, "rms_gain_error_db": -0.07570332998737882, "reference_mean_power": 0.0017996791281982027, "output_mean_power": 0.001768580152578034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.521732580259768e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030590519204748624, "output_power": 0.00030236139683073766, "reference_active": true, "delta_db": -0.0506050521077671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7673272930744677e-05, "output_power": 1.7821368756768194e-05, "reference_active": true, "delta_db": 0.036240722656632265}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.996266898550172e-07, "output_power": 3.616620728382318e-07, "reference_active": true, "delta_db": -0.4335151863904329}], "active_band_count": 3, "active_band_mae_db": 0.17345365371827745}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013968182745477764, "log_rms_mae_db": 1.0690971358894026, "rms_envelope_nrmse": 0.06348069723712403, "envelope_correlation": 0.9966332968191313, "rms_gain_error_db": -0.11095527640532976, "reference_mean_power": 0.0010910330804496425, "output_mean_power": 0.0010635119914638451}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.057825806205699e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031975947498638787, "output_power": 0.00031802712988647774, "reference_active": true, "delta_db": -0.023592523332714322}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9465579420304694e-05, "output_power": 3.7643888518451406e-05, "reference_active": true, "delta_db": -0.20524004759349793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.299925023694004e-06, "output_power": 5.64121905468295e-06, "reference_active": true, "delta_db": -0.47962416702986044}], "active_band_count": 3, "active_band_mae_db": 0.23615224598535756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012087571887341512, "log_rms_mae_db": 0.5899273889504009, "rms_envelope_nrmse": 0.036228058524046466, "envelope_correlation": 0.9989405770016395, "rms_gain_error_db": -0.11040627285327573, "reference_mean_power": 0.0010942126232004053, "output_mean_power": 0.0010667461726742158}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.627912926828161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002965811192999727, "output_power": 0.00029484203539819373, "reference_active": true, "delta_db": -0.025540992535325355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8971998770340482e-05, "output_power": 1.8723933370283955e-05, "reference_active": true, "delta_db": -0.057160008239367587}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.113014123121411e-07, "output_power": 3.4402720960145245e-07, "reference_active": true, "delta_db": -0.7756740787826246}], "active_band_count": 3, "active_band_mae_db": 0.2861250265191058}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.190065013136077e-05, "log_rms_mae_db": 1.1876341628493805, "rms_envelope_nrmse": 0.09639380826710024, "envelope_correlation": 0.9930480932140402, "rms_gain_error_db": -0.15845094500429913, "reference_mean_power": 0.00025666306625944256, "output_mean_power": 0.0002474675643098181}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2531384738435257e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.158137778593675e-05, "output_power": 6.851562598576891e-05, "reference_active": true, "delta_db": -0.1901042339266507}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.324550913609082e-06, "output_power": 6.644005491470966e-06, "reference_active": true, "delta_db": 0.9614699860108454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7772043437261862e-06, "output_power": 2.2924090551706786e-06, "reference_active": true, "delta_db": 1.1055474923745887}], "active_band_count": 3, "active_band_mae_db": 0.7523739041040282}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.863962912697811e-05, "log_rms_mae_db": 0.7882436654498315, "rms_envelope_nrmse": 0.0520803724917533, "envelope_correlation": 0.9980709939297792, "rms_gain_error_db": -0.158738933884925, "reference_mean_power": 0.00025765686496881694, "output_mean_power": 0.00024840928502467704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.1689575138838273e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607374483640592e-05, "output_power": 6.347648140359806e-05, "reference_active": true, "delta_db": -0.17416076691748314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.394936823991012e-06, "output_power": 3.010345810722164e-06, "reference_active": true, "delta_db": 0.9932232598812186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.833675392620637e-08, "output_power": 1.0591452151947299e-07, "reference_active": true, "delta_db": 0.3223964035262678}], "active_band_count": 3, "active_band_mae_db": 0.49659347677498983}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.534983553210222e-05, "log_rms_mae_db": 1.2354970770270526, "rms_envelope_nrmse": 0.2714626946473786, "envelope_correlation": 0.945166964774997, "rms_gain_error_db": 0.22729657019412675, "reference_mean_power": 1.9026430507347487e-05, "output_mean_power": 2.0048735035585236e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1975009508047096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.298926861031976e-06, "output_power": 5.5519296651701965e-06, "reference_active": true, "delta_db": 0.2025603044299667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588034222302733e-06, "output_power": 1.869577543243184e-06, "reference_active": true, "delta_db": 0.7088362552836752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.193680029927605e-07, "output_power": 1.129285110562729e-06, "reference_active": true, "delta_db": 4.302083105542083}], "active_band_count": 3, "active_band_mae_db": 1.7378265550852419}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3831266720127666e-05, "log_rms_mae_db": 0.8512913214049628, "rms_envelope_nrmse": 0.1762862357869115, "envelope_correlation": 0.9757719945723747, "rms_gain_error_db": 0.22761726524763773, "reference_mean_power": 1.9104126133492904e-05, "output_mean_power": 2.013209186103764e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1095182205605233e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.38789012101126e-06, "output_power": 5.9794705995637265e-06, "reference_active": true, "delta_db": 0.4524400457498572}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.477294263383285e-07, "output_power": 9.728954858231293e-07, "reference_active": true, "delta_db": 1.1432171597588927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.443035434026362e-08, "output_power": 7.053084027774941e-08, "reference_active": true, "delta_db": 4.604492918265635}], "active_band_count": 3, "active_band_mae_db": 2.066716707924795}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9222951451904348, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b9c663579c2068c5b8ade1e0f66e361505ce46947481a5ecbd3e71a575c0e156", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9925477209954984, "standard_deviation": 0.0012977132322349707, "minimum": 0.9901821952341198, "maximum": 0.9942268336684923, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1650303001445321, "standard_deviation": 0.023780112399238468, "minimum": 1.122511757833132, "maximum": 1.199827576689775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5889500076688667, "standard_deviation": 0.2673223462171564, "minimum": 0.21243285254082225, "maximum": 0.8639801944427244, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10083744431315862, "standard_deviation": 0.007479219932085429, "minimum": 0.09121894169267956, "maximum": 0.11416854847240514, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975679259413777, "standard_deviation": 0.000510046749183927, "minimum": 0.9967158327966275, "maximum": 0.9982039263206872, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7823006387867611, "standard_deviation": 0.02595583994325666, "minimum": 0.735670544280925, "maximum": 0.8201968869589679, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.46988192998035994, "standard_deviation": 0.18282376190499214, "minimum": 0.2736193178114571, "maximum": 0.7977763954523316, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06035622663656401, "standard_deviation": 0.004698870795703026, "minimum": 0.0520803724917533, "maximum": 0.06673851452006684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9531485550006512, "standard_deviation": 0.0065288095871072205, "minimum": 0.9449350339179342, "maximum": 0.9618737478491384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2438957325413522, "standard_deviation": 0.023546482172867977, "minimum": 1.215619560168866, "maximum": 1.2784703337683516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5568623770119001, "standard_deviation": 0.44311154983244855, "minimum": 1.1104411165211776, "maximum": 2.5597641805743074, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2463597634737829, "standard_deviation": 0.02381897071742055, "minimum": 0.22047598203386512, "maximum": 0.28890498306704965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9764052982127813, "standard_deviation": 0.004277063713700941, "minimum": 0.9672754840917724, "maximum": 0.9813536399703737, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8685404122621259, "standard_deviation": 0.028220592087375696, "minimum": 0.8215552575752765, "maximum": 0.9065741652646934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.7905774717830885, "standard_deviation": 0.6372708192485981, "minimum": 1.1973209378213834, "maximum": 3.215045344507056, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16973570493575682, "standard_deviation": 0.02256592344137858, "minimum": 0.1489525773547636, "maximum": 0.22099372796951125, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9966600321878583, "standard_deviation": 0.0003971372828986177, "minimum": 0.9960334790951381, "maximum": 0.9971419735983629, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0787383955364378, "standard_deviation": 0.017769614819640702, "minimum": 1.040880483294718, "maximum": 1.0964982943522006, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.32326822034005215, "standard_deviation": 0.12009159156005891, "minimum": 0.15380153364753932, "maximum": 0.48994277995471824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06273056865952394, "standard_deviation": 0.003771785564486763, "minimum": 0.05802942669731262, "maximum": 0.0685608717776225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989857939608844, "standard_deviation": 0.00012547531058332103, "minimum": 0.9987933912268965, "maximum": 0.9991266180756369, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5967206510948384, "standard_deviation": 0.026659484629334426, "minimum": 0.5601517474832647, "maximum": 0.6308649318893061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3877268369277967, "standard_deviation": 0.15788273351434287, "minimum": 0.1528927558217907, "maximum": 0.6116999774279156, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.034650826094201045, "standard_deviation": 0.0021389656102409095, "minimum": 0.03143960759527628, "maximum": 0.03811315111760083, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1352940763711344, "standard_deviation": 0.0244963305330031, "minimum": 0.09529046700920346, "maximum": 0.1664232827734526, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.325382698397133, "standard_deviation": 0.008561742163660237, "minimum": 5.313283153191726, "maximum": 5.338254074687477, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3493062509248316, "standard_deviation": 0.015024758769183046, "minimum": 0.3300410933821262, "maximum": 0.37581882937555244, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08300359609365944, "standard_deviation": 0.021563383073882156, "minimum": -0.1106339946800419, "maximum": -0.044093090363986474, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9971534484622844, "standard_deviation": 0.0004280927165566589, "minimum": 0.9964707834371415, "maximum": 0.9977266432953411, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.584843381957193, "standard_deviation": 0.01630841514856348, "minimum": 0.5631882281745288, "maximum": 0.6128942413320468, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.23819354033973542, "standard_deviation": 0.14343991730809175, "minimum": 0.06487424940324528, "maximum": 0.4419455159984884, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05266990152192002, "standard_deviation": 0.003898638969952913, "minimum": 0.047603372660318194, "maximum": 0.05869749866523727, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991615029551992, "standard_deviation": 0.00019360264591208217, "minimum": 0.9988687177202288, "maximum": 0.9993892222048231, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3146268314104758, "standard_deviation": 0.02407824619630185, "minimum": 0.2750545952106062, "maximum": 0.348803228295378, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2990052459088019, "standard_deviation": 0.17322062104660774, "minimum": 0.08248367138529959, "maximum": 0.5462863606180685, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02891547001179836, "standard_deviation": 0.003000484346959443, "minimum": 0.02584699630868631, "maximum": 0.03344111111740807, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0017", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.715669870376587, "predicted_mos": 2.715669870376587, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "17cd3a386ac1fb2fce09fc4b23e34d29a0317cdb782637c2fa38ec409c006a69", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8499819999560714, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5941722791980908, "全帯域_10ms_rms_envelope_nrmse": 0.05532990164405451, "全帯域_10ms_envelope_correlation": 0.997120548635358, "全帯域_10ms_modulation_mae_db": 0.2787485911942544, "全帯域_40ms_log_rms_mae_db": 0.32749102552678433, "全帯域_40ms_rms_envelope_nrmse": 0.03270677103435187, "全帯域_40ms_envelope_correlation": 0.9992017168725866, "全帯域_40ms_modulation_mae_db": 0.4454948738102609, "300–1000 Hz_10ms_log_rms_mae_db": 1.1143245298694686, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06526423851217492, "300–1000 Hz_10ms_envelope_correlation": 0.9966252012999552, "300–1000 Hz_10ms_modulation_mae_db": 0.4012617902112559, "300–1000 Hz_40ms_log_rms_mae_db": 0.5695764003676124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03700873245761768, "300–1000 Hz_40ms_envelope_correlation": 0.999107660334863, "300–1000 Hz_40ms_modulation_mae_db": 0.6994368845546673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2330589880830414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11210462473557012, "1000–3000 Hz_10ms_envelope_correlation": 0.9909473097882568, "1000–3000 Hz_10ms_modulation_mae_db": 0.817371086531259, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8110895534898745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06752051876866484, "1000–3000 Hz_40ms_envelope_correlation": 0.9971616636618689, "1000–3000 Hz_40ms_modulation_mae_db": 0.9334777163583351, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3568891830835028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22751531341761366, "3000–6000 Hz_10ms_envelope_correlation": 0.9610502270784637, "3000–6000 Hz_10ms_modulation_mae_db": 1.2178637774154037, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9133073623729057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15913045362770264, "3000–6000 Hz_40ms_envelope_correlation": 0.9799568270221515, "3000–6000 Hz_40ms_modulation_mae_db": 1.3142633230991512, "mr_stft_db": 5.38269674587696, "band_mae_db": 0.22248298174183745, "rms_gain_error_db": -0.16417752799868446, "output_peak": 0.3468111449304205}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6047488624977794, "全帯域_10ms_rms_envelope_nrmse": 0.0541813348924106, "全帯域_10ms_envelope_correlation": 0.99696097470376, "全帯域_10ms_modulation_mae_db": 0.22502985902441278, "全帯域_40ms_log_rms_mae_db": 0.3033864604742534, "全帯域_40ms_rms_envelope_nrmse": 0.02688681348248397, "全帯域_40ms_envelope_correlation": 0.9992523403211306, "全帯域_40ms_modulation_mae_db": 0.32628874648715805, "300–1000 Hz_10ms_log_rms_mae_db": 1.1503470809680176, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0659830220061096, "300–1000 Hz_10ms_envelope_correlation": 0.9962919095043873, "300–1000 Hz_10ms_modulation_mae_db": 0.32516146981275657, "300–1000 Hz_40ms_log_rms_mae_db": 0.6358109239849875, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03232167719732162, "300–1000 Hz_40ms_envelope_correlation": 0.9991129996155169, "300–1000 Hz_40ms_modulation_mae_db": 0.3643144495281922, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2249089989560624, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10369980395641182, "1000–3000 Hz_10ms_envelope_correlation": 0.9918936649381368, "1000–3000 Hz_10ms_modulation_mae_db": 0.5393971549803226, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8189472927146534, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06128337767617827, "1000–3000 Hz_40ms_envelope_correlation": 0.9972179335180231, "1000–3000 Hz_40ms_modulation_mae_db": 0.6559723443248675, "3000–6000 Hz_10ms_log_rms_mae_db": 1.360305969599351, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21899685720947273, "3000–6000 Hz_10ms_envelope_correlation": 0.963121488332643, "3000–6000 Hz_10ms_modulation_mae_db": 1.223650439244219, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9427568890618027, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1467277946531647, "3000–6000 Hz_40ms_envelope_correlation": 0.9824139002292883, "3000–6000 Hz_40ms_modulation_mae_db": 1.3432608368448093, "mr_stft_db": 5.3902935781687225, "band_mae_db": 0.1331019483040291, "rms_gain_error_db": -0.06197617957163719, "output_peak": 0.3709955788999374}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5903190973447922, "全帯域_10ms_rms_envelope_nrmse": 0.05645526099766935, "全帯域_10ms_envelope_correlation": 0.9968151321578465, "全帯域_10ms_modulation_mae_db": 0.35755179361247125, "全帯域_40ms_log_rms_mae_db": 0.34849256009126856, "全帯域_40ms_rms_envelope_nrmse": 0.03338946910125934, "全帯域_40ms_envelope_correlation": 0.9989600547387204, "全帯域_40ms_modulation_mae_db": 0.32837664737445155, "300–1000 Hz_10ms_log_rms_mae_db": 1.0907721683945986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06487296249245453, "300–1000 Hz_10ms_envelope_correlation": 0.9964503154268282, "300–1000 Hz_10ms_modulation_mae_db": 0.40005063274523467, "300–1000 Hz_40ms_log_rms_mae_db": 0.5975406984731864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03834610097406258, "300–1000 Hz_40ms_envelope_correlation": 0.9987638162396196, "300–1000 Hz_40ms_modulation_mae_db": 0.3500094826090645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1987395190394632, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09633669547456065, "1000–3000 Hz_10ms_envelope_correlation": 0.9933289211563799, "1000–3000 Hz_10ms_modulation_mae_db": 0.3458209023367657, "1000–3000 Hz_40ms_log_rms_mae_db": 0.807042043527158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05677551362151056, "1000–3000 Hz_40ms_envelope_correlation": 0.9980016072987155, "1000–3000 Hz_40ms_modulation_mae_db": 0.24897374657398222, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3168820487707782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25469653875596143, "3000–6000 Hz_10ms_envelope_correlation": 0.9472837949505214, "3000–6000 Hz_10ms_modulation_mae_db": 1.5785850642905512, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8973383523883581, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1686743229643234, "3000–6000 Hz_40ms_envelope_correlation": 0.9753596473134186, "3000–6000 Hz_40ms_modulation_mae_db": 1.6235055040472244, "mr_stft_db": 5.389580990991715, "band_mae_db": 0.25317216590056163, "rms_gain_error_db": -0.12092621737645319, "output_peak": 0.3338047264737063}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.543187309208285, "全帯域_10ms_rms_envelope_nrmse": 0.051791953413219946, "全帯域_10ms_envelope_correlation": 0.9971895690822213, "全帯域_10ms_modulation_mae_db": 0.06662186120487673, "全帯域_40ms_log_rms_mae_db": 0.28145968184863884, "全帯域_40ms_rms_envelope_nrmse": 0.02589975615164312, "全帯域_40ms_envelope_correlation": 0.9992567940336218, "全帯域_40ms_modulation_mae_db": 0.09452338803390588, "300–1000 Hz_10ms_log_rms_mae_db": 1.126809967978437, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06774458445515552, "300–1000 Hz_10ms_envelope_correlation": 0.9960872676041123, "300–1000 Hz_10ms_modulation_mae_db": 0.12861560588134632, "300–1000 Hz_40ms_log_rms_mae_db": 0.61467666255935, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03435304805861507, "300–1000 Hz_40ms_envelope_correlation": 0.9989511029618117, "300–1000 Hz_40ms_modulation_mae_db": 0.13230292441879002, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2051296502914992, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09743638644458086, "1000–3000 Hz_10ms_envelope_correlation": 0.9929706290143343, "1000–3000 Hz_10ms_modulation_mae_db": 0.4076061368245174, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7741369871719208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05927520563332067, "1000–3000 Hz_40ms_envelope_correlation": 0.9975225201749073, "1000–3000 Hz_40ms_modulation_mae_db": 0.248460844402662, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3357962591522592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22960416318223797, "3000–6000 Hz_10ms_envelope_correlation": 0.9584551411284467, "3000–6000 Hz_10ms_modulation_mae_db": 1.2839323733825172, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9180641042271732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15683997514765088, "3000–6000 Hz_40ms_envelope_correlation": 0.9791566542402237, "3000–6000 Hz_40ms_modulation_mae_db": 1.4648820392386313, "mr_stft_db": 5.372666855345365, "band_mae_db": 0.08092758929254958, "rms_gain_error_db": -0.006042711893799695, "output_peak": 0.3556795119153543}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5670292462351365, "全帯域_10ms_rms_envelope_nrmse": 0.054677888494803005, "全帯域_10ms_envelope_correlation": 0.9968906132875766, "全帯域_10ms_modulation_mae_db": 0.2057231734174789, "全帯域_40ms_log_rms_mae_db": 0.3063679238776039, "全帯域_40ms_rms_envelope_nrmse": 0.02842246493257247, "全帯域_40ms_envelope_correlation": 0.999144189411862, "全帯域_40ms_modulation_mae_db": 0.12418027706834729, "300–1000 Hz_10ms_log_rms_mae_db": 1.1419479500666394, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06918085061022693, "300–1000 Hz_10ms_envelope_correlation": 0.9959036624827603, "300–1000 Hz_10ms_modulation_mae_db": 0.13302153033511174, "300–1000 Hz_40ms_log_rms_mae_db": 0.6330896663831755, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0375441686783971, "300–1000 Hz_40ms_envelope_correlation": 0.9987669198620446, "300–1000 Hz_40ms_modulation_mae_db": 0.1140274361002015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1850365053371315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0997668207306177, "1000–3000 Hz_10ms_envelope_correlation": 0.992918729724962, "1000–3000 Hz_10ms_modulation_mae_db": 0.6097118632509009, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7891535848058652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05868711755394281, "1000–3000 Hz_40ms_envelope_correlation": 0.9979437670010953, "1000–3000 Hz_40ms_modulation_mae_db": 0.41693434591871964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.339971078822867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2596146815897172, "3000–6000 Hz_10ms_envelope_correlation": 0.9459175537983161, "3000–6000 Hz_10ms_modulation_mae_db": 1.2313212832967544, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8977249256072786, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17336225825509619, "3000–6000 Hz_40ms_envelope_correlation": 0.9740670539138231, "3000–6000 Hz_40ms_modulation_mae_db": 1.5890494132087818, "mr_stft_db": 5.372309869950482, "band_mae_db": 0.13611244394165298, "rms_gain_error_db": -0.05656047764621883, "output_peak": 0.3761845689181075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5402736774967672, "全帯域_10ms_rms_envelope_nrmse": 0.04787168279868813, "全帯域_10ms_envelope_correlation": 0.9976283270983191, "全帯域_10ms_modulation_mae_db": 0.19604449203875007, "全帯域_40ms_log_rms_mae_db": 0.2894706293804721, "全帯域_40ms_rms_envelope_nrmse": 0.025318979126987627, "全帯域_40ms_envelope_correlation": 0.9993307906889707, "全帯域_40ms_modulation_mae_db": 0.33100313160300904, "300–1000 Hz_10ms_log_rms_mae_db": 1.1401761911851929, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0596744362465963, "300–1000 Hz_10ms_envelope_correlation": 0.9969436177294166, "300–1000 Hz_10ms_modulation_mae_db": 0.22047802412250775, "300–1000 Hz_40ms_log_rms_mae_db": 0.6198896705624815, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030351097083197436, "300–1000 Hz_40ms_envelope_correlation": 0.9991788224175601, "300–1000 Hz_40ms_modulation_mae_db": 0.4581805762547175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1961024669519396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10457987443552838, "1000–3000 Hz_10ms_envelope_correlation": 0.9919186408458236, "1000–3000 Hz_10ms_modulation_mae_db": 0.6238401956103415, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8042934504467895, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06050195279247936, "1000–3000 Hz_40ms_envelope_correlation": 0.9975060239159289, "1000–3000 Hz_40ms_modulation_mae_db": 0.798549155093256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.320582524170401, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23016650050715817, "3000–6000 Hz_10ms_envelope_correlation": 0.9586023796987664, "3000–6000 Hz_10ms_modulation_mae_db": 1.0436551747629235, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9261375328286393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15938787972828244, "3000–6000 Hz_40ms_envelope_correlation": 0.9787488847603268, "3000–6000 Hz_40ms_modulation_mae_db": 1.1431284749884931, "mr_stft_db": 5.384348328414362, "band_mae_db": 0.15495894271928382, "rms_gain_error_db": -0.05699694317949726, "output_peak": 0.35419067508589747}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5783278466660503, "全帯域_10ms_rms_envelope_nrmse": 0.057988967033912146, "全帯域_10ms_envelope_correlation": 0.9966075891961533, "全帯域_10ms_modulation_mae_db": 0.5714348273557129, "全帯域_40ms_log_rms_mae_db": 0.32388929701512753, "全帯域_40ms_rms_envelope_nrmse": 0.029096434536860397, "全帯域_40ms_envelope_correlation": 0.9992324976573576, "全帯域_40ms_modulation_mae_db": 0.57594957697368, "300–1000 Hz_10ms_log_rms_mae_db": 1.1154006426823506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06825795436324406, "300–1000 Hz_10ms_envelope_correlation": 0.996236655863493, "300–1000 Hz_10ms_modulation_mae_db": 0.6464751879263427, "300–1000 Hz_40ms_log_rms_mae_db": 0.6085356989189961, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03831696479936444, "300–1000 Hz_40ms_envelope_correlation": 0.9989707748178952, "300–1000 Hz_40ms_modulation_mae_db": 0.6761924205941741, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2225270557156964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10574067955188751, "1000–3000 Hz_10ms_envelope_correlation": 0.9915737010400143, "1000–3000 Hz_10ms_modulation_mae_db": 0.7221538013008729, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8070330915704319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06299644506577308, "1000–3000 Hz_40ms_envelope_correlation": 0.9970617644800679, "1000–3000 Hz_40ms_modulation_mae_db": 0.568362585105018, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3745579482869845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2644349035169545, "3000–6000 Hz_10ms_envelope_correlation": 0.9453243788286642, "3000–6000 Hz_10ms_modulation_mae_db": 1.4059786484308343, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9419787009109253, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16959260816996505, "3000–6000 Hz_40ms_envelope_correlation": 0.9758398629452463, "3000–6000 Hz_40ms_modulation_mae_db": 1.5544001166827208, "mr_stft_db": 5.380986375752475, "band_mae_db": 0.13972187206044223, "rms_gain_error_db": -0.10610701524523113, "output_peak": 0.33567824429916426}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5688155471341, "全帯域_10ms_rms_envelope_nrmse": 0.05128418176203756, "全帯域_10ms_envelope_correlation": 0.9973074251497737, "全帯域_10ms_modulation_mae_db": 0.4444915324974445, "全帯域_40ms_log_rms_mae_db": 0.2982217053928293, "全帯域_40ms_rms_envelope_nrmse": 0.026868752352414507, "全帯域_40ms_envelope_correlation": 0.9992804489120182, "全帯域_40ms_modulation_mae_db": 0.4334878052987477, "300–1000 Hz_10ms_log_rms_mae_db": 1.167813517540168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06595762711360934, "300–1000 Hz_10ms_envelope_correlation": 0.9962946734647864, "300–1000 Hz_10ms_modulation_mae_db": 0.260159450133624, "300–1000 Hz_40ms_log_rms_mae_db": 0.6591120177103997, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03375684502000979, "300–1000 Hz_40ms_envelope_correlation": 0.9990250462127219, "300–1000 Hz_40ms_modulation_mae_db": 0.40607100200120394, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2242893084959425, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10524314277151045, "1000–3000 Hz_10ms_envelope_correlation": 0.9920133270830459, "1000–3000 Hz_10ms_modulation_mae_db": 0.4096213196769038, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7614924131322927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061111074332725016, "1000–3000 Hz_40ms_envelope_correlation": 0.9977018381382065, "1000–3000 Hz_40ms_modulation_mae_db": 0.4533586316471718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3403011640950324, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2529250233098136, "3000–6000 Hz_10ms_envelope_correlation": 0.9550180864444613, "3000–6000 Hz_10ms_modulation_mae_db": 2.0791378127882503, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8986892763486732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1742510414915612, "3000–6000 Hz_40ms_envelope_correlation": 0.9783063351462988, "3000–6000 Hz_40ms_modulation_mae_db": 2.2940718832762643, "mr_stft_db": 5.3749943239663205, "band_mae_db": 0.129545250141121, "rms_gain_error_db": -0.07670719173636986, "output_peak": 0.3507772928034293}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.1250195027860315, "spectral_convergence": 0.2010668470983575, "spectral_convergence_status": "ok", "amplitude_floor": 1.1114078093309192e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.192911625718703, "spectral_convergence": 0.1958882577531419, "spectral_convergence_status": "ok", "amplitude_floor": 9.683415126458649e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.830159109126147, "spectral_convergence": 0.20610565251795224, "spectral_convergence_status": "ok", "amplitude_floor": 7.055286564887976e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011089954817882694, "output_power": 0.00011229293342314664, "delta_db": 0.05422650264323832, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007963949709957278, "output_power": 0.0007687742493421619, "delta_db": -0.153296810281574, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005473467336544822, "output_power": 0.0005281223211821759, "delta_db": -0.15528007324425555, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020656259415557, "output_power": 0.0001946535653070001, "delta_db": -0.2578931638950654, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.567254948983353e-05, "output_power": 9.01828330351234e-05, "delta_db": -0.2566347296737723, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.697083983250908e-05, "output_power": 3.327381940118908e-05, "delta_db": -0.4575666107131191, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22248298174183745, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16417752799868446, "pitch": null, "output_rms_dbfs": -27.626322969534375, "output_peak": 0.3468111449304205, "reference_rms_dbfs": -27.46214544153569, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.38269674587696}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015211892717126925, "log_rms_mae_db": 0.5941722791980908, "rms_envelope_nrmse": 0.05532990164405451, "envelope_correlation": 0.997120548635358, "rms_gain_error_db": -0.16351106359119935, "reference_mean_power": 0.0017946561941472319, "output_mean_power": 0.0017283438788451646}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.098233390179555e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033819477659287355, "output_power": 0.00032106286653521473, "reference_active": true, "delta_db": -0.22576816777034422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.872738107665891e-05, "output_power": 3.467040416449983e-05, "reference_active": true, "delta_db": -0.4805922458999457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.329707857377202e-06, "output_power": 7.1137423105327445e-06, "reference_active": true, "delta_db": -0.12988535991247313}], "active_band_count": 3, "active_band_mae_db": 0.2787485911942544}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012855597503212083, "log_rms_mae_db": 0.32749102552678433, "rms_envelope_nrmse": 0.03270677103435187, "envelope_correlation": 0.9992017168725866, "rms_gain_error_db": -0.16365986200733026, "reference_mean_power": 0.0017996791281982027, "output_mean_power": 0.0017331218347162358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.521732580259768e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030590519204748624, "output_power": 0.0002898555566695783, "reference_active": true, "delta_db": -0.23405217792584307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7673272930744677e-05, "output_power": 1.621361208591529e-05, "reference_active": true, "delta_db": -0.3743720615729206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.996266898550172e-07, "output_power": 3.3794688485984e-07, "reference_active": true, "delta_db": -0.7280603819320193}], "active_band_count": 3, "active_band_mae_db": 0.4454948738102609}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013968182745477764, "log_rms_mae_db": 1.1143245298694686, "rms_envelope_nrmse": 0.06526423851217492, "envelope_correlation": 0.9966252012999552, "rms_gain_error_db": -0.17013705411885682, "reference_mean_power": 0.0010910330804496425, "output_mean_power": 0.0010491176978333423}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.057825806205699e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031975947498638787, "output_power": 0.00030678458876460364, "reference_active": true, "delta_db": -0.1798988301703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9465579420304694e-05, "output_power": 3.441166375201209e-05, "reference_active": true, "delta_db": -0.5951281274170482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.299925023694004e-06, "output_power": 5.707678989243189e-06, "reference_active": true, "delta_db": -0.4287584130464195}], "active_band_count": 3, "active_band_mae_db": 0.4012617902112559}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012087571887341512, "log_rms_mae_db": 0.5695764003676124, "rms_envelope_nrmse": 0.03700873245761768, "envelope_correlation": 0.999107660334863, "rms_gain_error_db": -0.17016485250752753, "reference_mean_power": 0.0010942126232004053, "output_mean_power": 0.0010521683539387336}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.627912926828161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002965811192999727, "output_power": 0.00028362118029371503, "reference_active": true, "delta_db": -0.1940483986583495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8971998770340482e-05, "output_power": 1.67270972108784e-05, "reference_active": true, "delta_db": -0.5469250698114267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.113014123121411e-07, "output_power": 3.0090297546170955e-07, "reference_active": true, "delta_db": -1.357337185194226}], "active_band_count": 3, "active_band_mae_db": 0.6994368845546673}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.190065013136077e-05, "log_rms_mae_db": 1.2330589880830414, "rms_envelope_nrmse": 0.11210462473557012, "envelope_correlation": 0.9909473097882568, "rms_gain_error_db": -0.2929494646365088, "reference_mean_power": 0.00025666306625944256, "output_mean_power": 0.00023992109333055206}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2531384738435257e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.158137778593675e-05, "output_power": 6.419952047180427e-05, "reference_active": true, "delta_db": -0.4726826917243992}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.324550913609082e-06, "output_power": 5.647600569153612e-06, "reference_active": true, "delta_db": 0.25580989186383585}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7772043437261862e-06, "output_power": 2.643013817770871e-06, "reference_active": true, "delta_db": 1.723620676005542}], "active_band_count": 3, "active_band_mae_db": 0.817371086531259}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.863962912697811e-05, "log_rms_mae_db": 0.8110895534898745, "rms_envelope_nrmse": 0.06752051876866484, "envelope_correlation": 0.9971616636618689, "rms_gain_error_db": -0.2932193771323792, "reference_mean_power": 0.00025765686496881694, "output_mean_power": 0.00024083509888332927}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.1689575138838273e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607374483640592e-05, "output_power": 5.8982194714383545e-05, "reference_active": true, "delta_db": -0.4930799333059884}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.394936823991012e-06, "output_power": 2.6524922374452405e-06, "reference_active": true, "delta_db": 0.44360059953613895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.833675392620637e-08, "output_power": 1.5103970539320177e-07, "reference_active": true, "delta_db": 1.8637526162328777}], "active_band_count": 3, "active_band_mae_db": 0.9334777163583351}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.534983553210222e-05, "log_rms_mae_db": 1.3568891830835028, "rms_envelope_nrmse": 0.22751531341761366, "envelope_correlation": 0.9610502270784637, "rms_gain_error_db": 0.22973363677612071, "reference_mean_power": 1.9026430507347487e-05, "output_mean_power": 2.0059988647090967e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1975009508047096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.298926861031976e-06, "output_power": 5.729957291899251e-06, "reference_active": true, "delta_db": 0.339634598131633}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588034222302733e-06, "output_power": 1.6365715810811702e-06, "reference_active": true, "delta_db": 0.13075148137601772}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.193680029927605e-07, "output_power": 8.728021689006912e-07, "reference_active": true, "delta_db": 3.183205252738561}], "active_band_count": 3, "active_band_mae_db": 1.2178637774154037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3831266720127666e-05, "log_rms_mae_db": 0.9133073623729057, "rms_envelope_nrmse": 0.15913045362770264, "envelope_correlation": 0.9799568270221515, "rms_gain_error_db": 0.22983187003925845, "reference_mean_power": 1.9104126133492904e-05, "output_mean_power": 2.0142360468732793e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1095182205605233e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.38789012101126e-06, "output_power": 5.9463310727207134e-06, "reference_active": true, "delta_db": 0.4283035539965422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.477294263383285e-07, "output_power": 7.921183744063716e-07, "reference_active": true, "delta_db": 0.250456152154403}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.443035434026362e-08, "output_power": 5.180036171590087e-08, "reference_active": true, "delta_db": 3.2640302631465086}], "active_band_count": 3, "active_band_mae_db": 1.3142633230991512}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9219748130932653, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c5d1d49b20338a093792ece2c5bd86a719f089925db1d20fe10e2b7e8c0ab60b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9921956154488691, "standard_deviation": 0.0008065875874728652, "minimum": 0.9909473097882568, "maximum": 0.9933289211563799, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.211224061608847, "standard_deviation": 0.017196163431145963, "minimum": 1.1850365053371315, "maximum": 1.2330589880830414, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5594403075639854, "standard_deviation": 0.16527590156422384, "minimum": 0.3458209023367657, "maximum": 0.817371086531259, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10311350351258344, "standard_deviation": 0.005128725002882048, "minimum": 0.09633669547456065, "maximum": 0.11210462473557012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975146397736017, "standard_deviation": 0.0003533896834086116, "minimum": 0.9970617644800679, "maximum": 0.9980016072987155, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7966485521073733, "standard_deviation": 0.019924799143286192, "minimum": 0.7614924131322927, "maximum": 0.8189472927146534, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5405111711780015, "standard_deviation": 0.24734809212304243, "minimum": 0.248460844402662, "maximum": 0.9334777163583351, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06101890068057433, "standard_deviation": 0.0032295824020555316, "minimum": 0.05677551362151056, "maximum": 0.06752051876866484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9543466312825353, "standard_deviation": 0.007168507079529626, "minimum": 0.9453243788286642, "maximum": 0.963121488332643, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.343160771997647, "standard_deviation": 0.019788690382662625, "minimum": 1.3168820487707782, "maximum": 1.3745579482869845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3830155717014319, "standard_deviation": 0.3216092084665019, "minimum": 1.0436551747629235, "maximum": 2.0791378127882503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24224424768611616, "standard_deviation": 0.017428720479457444, "minimum": 0.21899685720947273, "maximum": 0.2644349035169545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9779811456963472, "standard_deviation": 0.002736188206501124, "minimum": 0.9740670539138231, "maximum": 0.9824139002292883, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9169996429682195, "standard_deviation": 0.01881870919078011, "minimum": 0.8973383523883581, "maximum": 0.9427568890618027, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5408201989232595, "standard_deviation": 0.3445532224804327, "minimum": 1.1431284749884931, "maximum": 2.2940718832762643, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16349579175471832, "standard_deviation": 0.009549926354837592, "minimum": 0.1467277946531647, "maximum": 0.1742510414915612, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963541629219674, "standard_deviation": 0.0003219516878762354, "minimum": 0.9959036624827603, "maximum": 0.9969436177294166, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.130949006085609, "standard_deviation": 0.0241744551268775, "minimum": 1.0907721683945986, "maximum": 1.167813517540168, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3144029613960225, "standard_deviation": 0.17090018233109147, "minimum": 0.12861560588134632, "maximum": 0.6464751879263427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0658669594749464, "standard_deviation": 0.002926346923196834, "minimum": 0.0596744362465963, "maximum": 0.06918085061022693, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989846428077541, "standard_deviation": 0.00015511831187998946, "minimum": 0.9987638162396196, "maximum": 0.9991788224175601, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6172789673700236, "standard_deviation": 0.02698831065056742, "minimum": 0.5695764003676124, "maximum": 0.6591120177103997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4000668970076263, "standard_deviation": 0.21599637684199133, "minimum": 0.1140274361002015, "maximum": 0.6994368845546673, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03524982928357322, "standard_deviation": 0.0029989717922686634, "minimum": 0.030351097083197436, "maximum": 0.03834610097406258, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1562528992626847, "standard_deviation": 0.05527014509175145, "minimum": 0.08092758929254958, "maximum": 0.25317216590056163, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.3809846335583, "standard_deviation": 0.007122603382927693, "minimum": 5.372309869950482, "maximum": 5.3902935781687225, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.35301521791575213, "standard_deviation": 0.01503189387892094, "minimum": 0.3338047264737063, "maximum": 0.3761845689181075, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08118678308098645, "standard_deviation": 0.04827755036609712, "minimum": -0.16417752799868446, "maximum": -0.006042711893799695, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.997065022413876, "standard_deviation": 0.0003179400315885524, "minimum": 0.9966075891961533, "maximum": 0.9976283270983191, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5733592332226252, "standard_deviation": 0.02328865031466069, "minimum": 0.5402736774967672, "maximum": 0.6047488624977794, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2932057662931752, "standard_deviation": 0.15934237722919206, "minimum": 0.06662186120487673, "maximum": 0.5714348273557129, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05369764637959941, "standard_deviation": 0.0032354572774304924, "minimum": 0.04787168279868813, "maximum": 0.057988967033912146, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992073540795334, "standard_deviation": 0.0001139286988146501, "minimum": 0.9989600547387204, "maximum": 0.9993307906889707, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3098474104508722, "standard_deviation": 0.022077795514480796, "minimum": 0.28145968184863884, "maximum": 0.34849256009126856, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.332413055831195, "standard_deviation": 0.1613876484511152, "minimum": 0.09452338803390588, "maximum": 0.57594957697368, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02857368008982166, "standard_deviation": 0.003025266172929263, "minimum": 0.025318979126987627, "maximum": 0.03338946910125934, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_3729-6852-0017", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6962802410125732, "predicted_mos": 2.6962802410125732, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3101a21fa16c99bf3e48fa9fab9539e364ef136d799faba517a6244de7909583", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.810576300136745, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5946620224912368, "全帯域_10ms_rms_envelope_nrmse": 0.05683616829006302, "全帯域_10ms_envelope_correlation": 0.9968697506487229, "全帯域_10ms_modulation_mae_db": 0.3444892710176592, "全帯域_40ms_log_rms_mae_db": 0.318826484809904, "全帯域_40ms_rms_envelope_nrmse": 0.03085507239875818, "全帯域_40ms_envelope_correlation": 0.9992545663617458, "全帯域_40ms_modulation_mae_db": 0.2605286028647394, "300–1000 Hz_10ms_log_rms_mae_db": 1.124749166862816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06990201745262498, "300–1000 Hz_10ms_envelope_correlation": 0.9960087886125855, "300–1000 Hz_10ms_modulation_mae_db": 0.3222803444179723, "300–1000 Hz_40ms_log_rms_mae_db": 0.6083755445998568, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03950682779617806, "300–1000 Hz_40ms_envelope_correlation": 0.9988540851493616, "300–1000 Hz_40ms_modulation_mae_db": 0.3933785086756177, "1000–3000 Hz_10ms_log_rms_mae_db": 1.173425344553183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1025044750659972, "1000–3000 Hz_10ms_envelope_correlation": 0.9926305886536915, "1000–3000 Hz_10ms_modulation_mae_db": 0.5740481790610121, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7641698703591592, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06140459909171687, "1000–3000 Hz_40ms_envelope_correlation": 0.9978674211753367, "1000–3000 Hz_40ms_modulation_mae_db": 0.31935682121812364, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2771515038211358, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24645005483564453, "3000–6000 Hz_10ms_envelope_correlation": 0.950759152162105, "3000–6000 Hz_10ms_modulation_mae_db": 1.2435764447067568, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8955761536990176, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1648075707623831, "3000–6000 Hz_40ms_envelope_correlation": 0.9764985157189582, "3000–6000 Hz_40ms_modulation_mae_db": 1.189354318729955, "mr_stft_db": 5.369626239703815, "band_mae_db": 0.16142016575160095, "rms_gain_error_db": -0.14956405508956977, "output_peak": 0.3447426176366877}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5783884388870906, "全帯域_10ms_rms_envelope_nrmse": 0.05262250788303527, "全帯域_10ms_envelope_correlation": 0.9971101797456217, "全帯域_10ms_modulation_mae_db": 0.3449235102423283, "全帯域_40ms_log_rms_mae_db": 0.2918079739724332, "全帯域_40ms_rms_envelope_nrmse": 0.026413910783468816, "全帯域_40ms_envelope_correlation": 0.999249361497395, "全帯域_40ms_modulation_mae_db": 0.33328058981684944, "300–1000 Hz_10ms_log_rms_mae_db": 1.1349174906277202, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06457754351986497, "300–1000 Hz_10ms_envelope_correlation": 0.996411093413686, "300–1000 Hz_10ms_modulation_mae_db": 0.4462900871957703, "300–1000 Hz_40ms_log_rms_mae_db": 0.6276448713765425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03347912943321233, "300–1000 Hz_40ms_envelope_correlation": 0.998985435152893, "300–1000 Hz_40ms_modulation_mae_db": 0.429423298004832, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2138988732394274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1034457369082634, "1000–3000 Hz_10ms_envelope_correlation": 0.9919744777742608, "1000–3000 Hz_10ms_modulation_mae_db": 0.26594689574769687, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7713188641810301, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053848733063435156, "1000–3000 Hz_40ms_envelope_correlation": 0.9979458066081903, "1000–3000 Hz_40ms_modulation_mae_db": 0.21354738791349262, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3079688863916725, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2373007578067196, "3000–6000 Hz_10ms_envelope_correlation": 0.9542081785119076, "3000–6000 Hz_10ms_modulation_mae_db": 1.313914282164996, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9339551659484562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16458206860584412, "3000–6000 Hz_40ms_envelope_correlation": 0.9766745911679281, "3000–6000 Hz_40ms_modulation_mae_db": 1.4305212354656291, "mr_stft_db": 5.3752119673502206, "band_mae_db": 0.12466024394146787, "rms_gain_error_db": -0.04011635726722202, "output_peak": 0.3365724533158196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5597503220659711, "全帯域_10ms_rms_envelope_nrmse": 0.053722008831663205, "全帯域_10ms_envelope_correlation": 0.9971589140752213, "全帯域_10ms_modulation_mae_db": 0.2741713919979841, "全帯域_40ms_log_rms_mae_db": 0.3127704106276437, "全帯域_40ms_rms_envelope_nrmse": 0.03234126995683756, "全帯域_40ms_envelope_correlation": 0.9990655863253888, "全帯域_40ms_modulation_mae_db": 0.16141471427431467, "300–1000 Hz_10ms_log_rms_mae_db": 1.0834373828558166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0667861202055159, "300–1000 Hz_10ms_envelope_correlation": 0.9963373387805987, "300–1000 Hz_10ms_modulation_mae_db": 0.2363559539518104, "300–1000 Hz_40ms_log_rms_mae_db": 0.5837164338412743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038872011577548604, "300–1000 Hz_40ms_envelope_correlation": 0.998844327059737, "300–1000 Hz_40ms_modulation_mae_db": 0.23203006205892995, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1644565426802627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09619927078790243, "1000–3000 Hz_10ms_envelope_correlation": 0.9936790127584688, "1000–3000 Hz_10ms_modulation_mae_db": 0.20046200988822005, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7552748192784067, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06034415844296742, "1000–3000 Hz_40ms_envelope_correlation": 0.9980487215233635, "1000–3000 Hz_40ms_modulation_mae_db": 0.3270900458565542, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2813351814141047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24286504106353363, "3000–6000 Hz_10ms_envelope_correlation": 0.9549402893082822, "3000–6000 Hz_10ms_modulation_mae_db": 1.4992351002816287, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8635589068063004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16647014301639623, "3000–6000 Hz_40ms_envelope_correlation": 0.9775457554879027, "3000–6000 Hz_40ms_modulation_mae_db": 1.6369667340978833, "mr_stft_db": 5.365196308156072, "band_mae_db": 0.13733201477699117, "rms_gain_error_db": -0.11762549077487001, "output_peak": 0.33355731594876475}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5823198875947072, "全帯域_10ms_rms_envelope_nrmse": 0.05178765949251317, "全帯域_10ms_envelope_correlation": 0.9974868469922588, "全帯域_10ms_modulation_mae_db": 0.5265521325402466, "全帯域_40ms_log_rms_mae_db": 0.3244329474627996, "全帯域_40ms_rms_envelope_nrmse": 0.03275622547958077, "全帯域_40ms_envelope_correlation": 0.9991531326241337, "全帯域_40ms_modulation_mae_db": 0.6956975664148816, "300–1000 Hz_10ms_log_rms_mae_db": 1.0663784444780497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05933253100597554, "300–1000 Hz_10ms_envelope_correlation": 0.9971747328733382, "300–1000 Hz_10ms_modulation_mae_db": 0.5642185342384911, "300–1000 Hz_40ms_log_rms_mae_db": 0.571993398723362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03541626288887946, "300–1000 Hz_40ms_envelope_correlation": 0.9991047493442892, "300–1000 Hz_40ms_modulation_mae_db": 0.8519498736708883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2507762272977276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10246358628411917, "1000–3000 Hz_10ms_envelope_correlation": 0.9926537863496274, "1000–3000 Hz_10ms_modulation_mae_db": 0.3477448882812273, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8248857067611115, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06410749376896069, "1000–3000 Hz_40ms_envelope_correlation": 0.9976059501381835, "1000–3000 Hz_40ms_modulation_mae_db": 0.3844013774392321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3053771743300246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23263602869519662, "3000–6000 Hz_10ms_envelope_correlation": 0.9560342715676431, "3000–6000 Hz_10ms_modulation_mae_db": 1.1750581905526387, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8782503544110355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16418322639053468, "3000–6000 Hz_40ms_envelope_correlation": 0.9767537138325314, "3000–6000 Hz_40ms_modulation_mae_db": 1.1899122343380615, "mr_stft_db": 5.382215340484282, "band_mae_db": 0.23129816203824205, "rms_gain_error_db": -0.1592206115505418, "output_peak": 0.3419230810451049}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6080622966125925, "全帯域_10ms_rms_envelope_nrmse": 0.0527878700496882, "全帯域_10ms_envelope_correlation": 0.9970811181945506, "全帯域_10ms_modulation_mae_db": 0.28635527434129515, "全帯域_40ms_log_rms_mae_db": 0.3369576058714818, "全帯域_40ms_rms_envelope_nrmse": 0.028627310625166463, "全帯域_40ms_envelope_correlation": 0.9991012606590454, "全帯域_40ms_modulation_mae_db": 0.36897833897653326, "300–1000 Hz_10ms_log_rms_mae_db": 1.1631680484831206, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06748477755146655, "300–1000 Hz_10ms_envelope_correlation": 0.9960897589591802, "300–1000 Hz_10ms_modulation_mae_db": 0.3057432383151228, "300–1000 Hz_40ms_log_rms_mae_db": 0.6738815210705962, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036919767358962546, "300–1000 Hz_40ms_envelope_correlation": 0.9987660418478842, "300–1000 Hz_40ms_modulation_mae_db": 0.5646078313423942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2412239113007468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09672874925459045, "1000–3000 Hz_10ms_envelope_correlation": 0.9930540173354455, "1000–3000 Hz_10ms_modulation_mae_db": 0.2203149458996618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7580033226365255, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055609823425685344, "1000–3000 Hz_40ms_envelope_correlation": 0.9978421410738206, "1000–3000 Hz_40ms_modulation_mae_db": 0.40399136523385804, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3544712197767443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2511612965313319, "3000–6000 Hz_10ms_envelope_correlation": 0.9489305704951688, "3000–6000 Hz_10ms_modulation_mae_db": 1.2312211220393687, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9366295085197534, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16672755079696217, "3000–6000 Hz_40ms_envelope_correlation": 0.9759592078768693, "3000–6000 Hz_40ms_modulation_mae_db": 1.195798707482181, "mr_stft_db": 5.376550133937786, "band_mae_db": 0.15475170230604865, "rms_gain_error_db": -0.019312698517825147, "output_peak": 0.37389532333426984}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5448242970694748, "全帯域_10ms_rms_envelope_nrmse": 0.054034073956186326, "全帯域_10ms_envelope_correlation": 0.9970168818515156, "全帯域_10ms_modulation_mae_db": 0.31380299553508256, "全帯域_40ms_log_rms_mae_db": 0.2799379607390981, "全帯域_40ms_rms_envelope_nrmse": 0.027156319124589674, "全帯域_40ms_envelope_correlation": 0.9992869393879711, "全帯域_40ms_modulation_mae_db": 0.13209530638298028, "300–1000 Hz_10ms_log_rms_mae_db": 1.0357750129495984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06945184169502784, "300–1000 Hz_10ms_envelope_correlation": 0.9958727395122127, "300–1000 Hz_10ms_modulation_mae_db": 0.5389219869773342, "300–1000 Hz_40ms_log_rms_mae_db": 0.5285020594226799, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03211217387221711, "300–1000 Hz_40ms_envelope_correlation": 0.9991108950504174, "300–1000 Hz_40ms_modulation_mae_db": 0.372251132054366, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1950069232966039, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10321619352652213, "1000–3000 Hz_10ms_envelope_correlation": 0.9927964388201113, "1000–3000 Hz_10ms_modulation_mae_db": 0.620896984307338, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7950017276772973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0635653827647639, "1000–3000 Hz_40ms_envelope_correlation": 0.9979759498583263, "1000–3000 Hz_40ms_modulation_mae_db": 0.5394150674022904, "3000–6000 Hz_10ms_log_rms_mae_db": 1.288217859475294, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24011131185038687, "3000–6000 Hz_10ms_envelope_correlation": 0.9568245088351086, "3000–6000 Hz_10ms_modulation_mae_db": 1.4734635130770852, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8468513792953276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16299479004492431, "3000–6000 Hz_40ms_envelope_correlation": 0.979046773542698, "3000–6000 Hz_40ms_modulation_mae_db": 1.6885420087844096, "mr_stft_db": 5.348180546754777, "band_mae_db": 0.1406532714790452, "rms_gain_error_db": -0.0859031081613386, "output_peak": 0.33772425377618853}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.5740556790972533, "全帯域_10ms_rms_envelope_nrmse": 0.05050369812899721, "全帯域_10ms_envelope_correlation": 0.9974332078849941, "全帯域_10ms_modulation_mae_db": 0.22124425212659235, "全帯域_40ms_log_rms_mae_db": 0.31108961039713084, "全帯域_40ms_rms_envelope_nrmse": 0.02929762077029422, "全帯域_40ms_envelope_correlation": 0.9991768473007367, "全帯域_40ms_modulation_mae_db": 0.2754826897996956, "300–1000 Hz_10ms_log_rms_mae_db": 1.125723938444343, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06640497934093285, "300–1000 Hz_10ms_envelope_correlation": 0.9963609622744695, "300–1000 Hz_10ms_modulation_mae_db": 0.3039553799572562, "300–1000 Hz_40ms_log_rms_mae_db": 0.6213393406203405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03992725224236333, "300–1000 Hz_40ms_envelope_correlation": 0.9987492449160685, "300–1000 Hz_40ms_modulation_mae_db": 0.35128381239638173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1732523250400786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0947670321667833, "1000–3000 Hz_10ms_envelope_correlation": 0.9932579679402044, "1000–3000 Hz_10ms_modulation_mae_db": 0.24564589429970884, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7482612129182168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04781861526959844, "1000–3000 Hz_40ms_envelope_correlation": 0.9983775551097546, "1000–3000 Hz_40ms_modulation_mae_db": 0.24280076823805782, "3000–6000 Hz_10ms_log_rms_mae_db": 1.306193273413887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22921840168013027, "3000–6000 Hz_10ms_envelope_correlation": 0.9574031122144243, "3000–6000 Hz_10ms_modulation_mae_db": 1.169728222469203, "3000–6000 Hz_40ms_log_rms_mae_db": 0.885047935130834, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15738936708587337, "3000–6000 Hz_40ms_envelope_correlation": 0.9785824526456053, "3000–6000 Hz_40ms_modulation_mae_db": 0.976524875737546, "mr_stft_db": 5.362891866023716, "band_mae_db": 0.12899855234233812, "rms_gain_error_db": -0.09717950088507606, "output_peak": 0.35801059693451515}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.549798601967563, "全帯域_10ms_rms_envelope_nrmse": 0.05087616848053216, "全帯域_10ms_envelope_correlation": 0.9972904556395684, "全帯域_10ms_modulation_mae_db": 0.08825844146173185, "全帯域_40ms_log_rms_mae_db": 0.3016753554770105, "全帯域_40ms_rms_envelope_nrmse": 0.028902624673903286, "全帯域_40ms_envelope_correlation": 0.9990812557234308, "全帯域_40ms_modulation_mae_db": 0.18603096358230586, "300–1000 Hz_10ms_log_rms_mae_db": 1.0718881672404046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06485256174860371, "300–1000 Hz_10ms_envelope_correlation": 0.9963830362914694, "300–1000 Hz_10ms_modulation_mae_db": 0.17988789931833515, "300–1000 Hz_40ms_log_rms_mae_db": 0.5938793518644689, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037220867866178155, "300–1000 Hz_40ms_envelope_correlation": 0.9987409583426898, "300–1000 Hz_40ms_modulation_mae_db": 0.22632025282147397, "1000–3000 Hz_10ms_log_rms_mae_db": 1.169825991064571, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09176278425297833, "1000–3000 Hz_10ms_envelope_correlation": 0.993636975245371, "1000–3000 Hz_10ms_modulation_mae_db": 0.2860814060164058, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7606524925994117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052415193551960684, "1000–3000 Hz_40ms_envelope_correlation": 0.9979410468199615, "1000–3000 Hz_40ms_modulation_mae_db": 0.20798741405993837, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3372260670402563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2439615702817758, "3000–6000 Hz_10ms_envelope_correlation": 0.9522544601911507, "3000–6000 Hz_10ms_modulation_mae_db": 1.5278491653696398, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9181772249199046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16926753593087845, "3000–6000 Hz_40ms_envelope_correlation": 0.9752801152586792, "3000–6000 Hz_40ms_modulation_mae_db": 1.5586696808762597, "mr_stft_db": 5.353422305179595, "band_mae_db": 0.11414179547561831, "rms_gain_error_db": -0.028247717179511172, "output_peak": 0.3837020293707197}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.101123733316089, "spectral_convergence": 0.19997289608550783, "spectral_convergence_status": "ok", "amplitude_floor": 1.1114078093309192e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.177552423680966, "spectral_convergence": 0.19370271964077573, "spectral_convergence_status": "ok", "amplitude_floor": 9.683415126458649e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.830202562114391, "spectral_convergence": 0.2016166626452733, "spectral_convergence_status": "ok", "amplitude_floor": 7.055286564887976e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011089954817882694, "output_power": 0.00010961845148443593, "delta_db": -0.050461140166502, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007963949709957278, "output_power": 0.0007721635325617458, "delta_db": -0.13419221837898043, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005473467336544822, "output_power": 0.0005339909346947416, "delta_db": -0.10728646448753756, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020656259415557, "output_power": 0.00019008280783143169, "delta_db": -0.3610884054381863, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.567254948983353e-05, "output_power": 9.11992645123987e-05, "delta_db": -0.207960113096141, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.697083983250908e-05, "output_power": 3.606667038779247e-05, "delta_db": -0.10753265294225833, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16142016575160095, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14956405508956977, "pitch": null, "output_rms_dbfs": -27.61170949662526, "output_peak": 0.3447426176366877, "reference_rms_dbfs": -27.46214544153569, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.369626239703815}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015211892717126925, "log_rms_mae_db": 0.5946620224912368, "rms_envelope_nrmse": 0.05683616829006302, "envelope_correlation": 0.9968697506487229, "rms_gain_error_db": -0.1485234925351318, "reference_mean_power": 0.0017946561941472319, "output_mean_power": 0.0017343187245395526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.098233390179555e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033819477659287355, "output_power": 0.00031880386134846097, "reference_active": true, "delta_db": -0.25643322724257167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.872738107665891e-05, "output_power": 3.392980868373686e-05, "reference_active": true, "delta_db": -0.5743671814824898}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.329707857377202e-06, "output_power": 7.67986181194824e-06, "reference_active": true, "delta_db": 0.202667404327916}], "active_band_count": 3, "active_band_mae_db": 0.3444892710176592}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012855597503212083, "log_rms_mae_db": 0.318826484809904, "rms_envelope_nrmse": 0.03085507239875818, "envelope_correlation": 0.9992545663617458, "rms_gain_error_db": -0.14829926477840025, "reference_mean_power": 0.0017996791281982027, "output_mean_power": 0.0017392625808055773}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.521732580259768e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030590519204748624, "output_power": 0.0002882737212063294, "reference_active": true, "delta_db": -0.2578179394468646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7673272930744677e-05, "output_power": 1.614794604338034e-05, "reference_active": true, "delta_db": -0.39199694783336725}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.996266898550172e-07, "output_power": 3.8768356783310763e-07, "reference_active": true, "delta_db": -0.1317709213139863}], "active_band_count": 3, "active_band_mae_db": 0.2605286028647394}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013968182745477764, "log_rms_mae_db": 1.124749166862816, "rms_envelope_nrmse": 0.06990201745262498, "envelope_correlation": 0.9960087886125855, "rms_gain_error_db": -0.1520572053111569, "reference_mean_power": 0.0010910330804496425, "output_mean_power": 0.0010534943194639918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.057825806205699e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031975947498638787, "output_power": 0.00030060273716053077, "reference_active": true, "delta_db": -0.2683049141427678}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9465579420304694e-05, "output_power": 3.40269346114873e-05, "reference_active": true, "delta_db": -0.6439565683379489}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.299925023694004e-06, "output_power": 6.3795983340220576e-06, "reference_active": true, "delta_db": 0.05457955077320037}], "active_band_count": 3, "active_band_mae_db": 0.3222803444179723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012087571887341512, "log_rms_mae_db": 0.6083755445998568, "rms_envelope_nrmse": 0.03950682779617806, "envelope_correlation": 0.9988540851493616, "rms_gain_error_db": -0.15152691096844592, "reference_mean_power": 0.0010942126232004053, "output_mean_power": 0.0010566934843478945}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.627912926828161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002965811192999727, "output_power": 0.0002783804769807977, "reference_active": true, "delta_db": -0.27504725250895745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8971998770340482e-05, "output_power": 1.6781358874135023e-05, "reference_active": true, "delta_db": -0.5328596282584702}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.113014123121411e-07, "output_power": 3.775177154871238e-07, "reference_active": true, "delta_db": -0.3722286452594254}], "active_band_count": 3, "active_band_mae_db": 0.3933785086756177}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.190065013136077e-05, "log_rms_mae_db": 1.173425344553183, "rms_envelope_nrmse": 0.1025044750659972, "envelope_correlation": 0.9926305886536915, "rms_gain_error_db": -0.3036443285086327, "reference_mean_power": 0.00025666306625944256, "output_mean_power": 0.00023933099450784047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2531384738435257e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.158137778593675e-05, "output_power": 6.572893067256723e-05, "reference_active": true, "delta_db": -0.3704348644904949}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.324550913609082e-06, "output_power": 4.685288227492631e-06, "reference_active": true, "delta_db": -0.5554667135410883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7772043437261862e-06, "output_power": 2.134821961982526e-06, "reference_active": true, "delta_db": 0.7962429591514532}], "active_band_count": 3, "active_band_mae_db": 0.5740481790610121}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.863962912697811e-05, "log_rms_mae_db": 0.7641698703591592, "rms_envelope_nrmse": 0.06140459909171687, "envelope_correlation": 0.9978674211753367, "rms_gain_error_db": -0.30338559974123414, "reference_mean_power": 0.00025765686496881694, "output_mean_power": 0.0002402719971393224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.1689575138838273e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607374483640592e-05, "output_power": 6.0642291253498634e-05, "reference_active": true, "delta_db": -0.37253319924387346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.394936823991012e-06, "output_power": 2.1025827400520663e-06, "reference_active": true, "delta_db": -0.5654096665213846}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.833675392620637e-08, "output_power": 9.879355832046257e-08, "reference_active": true, "delta_db": 0.020127597889112905}], "active_band_count": 3, "active_band_mae_db": 0.31935682121812364}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.534983553210222e-05, "log_rms_mae_db": 1.2771515038211358, "rms_envelope_nrmse": 0.24645005483564453, "envelope_correlation": 0.950759152162105, "rms_gain_error_db": -0.16251338580031582, "reference_mean_power": 1.9026430507347487e-05, "output_mean_power": 1.8327616155936388e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1975009508047096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.298926861031976e-06, "output_power": 5.0007422766961415e-06, "reference_active": true, "delta_db": -0.2515345227724863}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588034222302733e-06, "output_power": 1.4393710136573354e-06, "reference_active": true, "delta_db": -0.42687104753195426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.193680029927605e-07, "output_power": 8.468912931044862e-07, "reference_active": true, "delta_db": 3.05232376381583}], "active_band_count": 3, "active_band_mae_db": 1.2435764447067568}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3831266720127666e-05, "log_rms_mae_db": 0.8955761536990176, "rms_envelope_nrmse": 0.1648075707623831, "envelope_correlation": 0.9764985157189582, "rms_gain_error_db": -0.16216291437465086, "reference_mean_power": 1.9104126133492904e-05, "output_mean_power": 1.8403943250114907e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1095182205605233e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.38789012101126e-06, "output_power": 5.184834796353489e-06, "reference_active": true, "delta_db": -0.1668380718917568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.477294263383285e-07, "output_power": 6.957447973162576e-07, "reference_active": true, "delta_db": -0.3129450489707803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.443035434026362e-08, "output_power": 4.974595235793e-08, "reference_active": true, "delta_db": 3.0882798353273277}], "active_band_count": 3, "active_band_mae_db": 1.189354318729955}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9124171920159616, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8ce238f5562d404127cabf6dd5af338b242c5cccbc5d87c5a5d7505c56623d51", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9929604081096476, "standard_deviation": 0.000569859957954061, "minimum": 0.9919744777742608, "maximum": 0.9936790127584688, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.197733267309075, "standard_deviation": 0.03391185970051341, "minimum": 1.1644565426802627, "maximum": 1.2507762272977276, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.34514265043765885, "standard_deviation": 0.16240127177965147, "minimum": 0.20046200988822005, "maximum": 0.620896984307338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09888597853089455, "standard_deviation": 0.004551492773914749, "minimum": 0.09176278425297833, "maximum": 0.1034457369082634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9979505740383672, "standard_deviation": 0.0002171704175007587, "minimum": 0.9976059501381835, "maximum": 0.9983775551097546, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7721960020513948, "standard_deviation": 0.02551016486824046, "minimum": 0.7482612129182168, "maximum": 0.8248857067611115, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3298237809201934, "standard_deviation": 0.11241656405871704, "minimum": 0.20798741405993837, "maximum": 0.5394150674022904, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05738924992238606, "standard_deviation": 0.005857520603029072, "minimum": 0.04781861526959844, "maximum": 0.06410749376896069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9539193179107238, "standard_deviation": 0.0030195034745907857, "minimum": 0.9489305704951688, "maximum": 0.9574031122144243, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3072426457078898, "standard_deviation": 0.026921229997249906, "minimum": 1.2771515038211358, "maximum": 1.3544712197767443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3292557550826647, "standard_deviation": 0.1490643268294865, "minimum": 1.169728222469203, "maximum": 1.5278491653696398, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2404630578430899, "standard_deviation": 0.0072322106970231885, "minimum": 0.22921840168013027, "maximum": 0.2511612965313319, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9770426406913966, "standard_deviation": 0.0012781698730245504, "minimum": 0.9752801152586792, "maximum": 0.979046773542698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8947558285913286, "standard_deviation": 0.032675866856617004, "minimum": 0.8468513792953276, "maximum": 0.9366295085197534, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3582862244389906, "standard_deviation": 0.25666399542348955, "minimum": 0.976524875737546, "maximum": 1.6885420087844096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16455278157922454, "standard_deviation": 0.0034760198040572413, "minimum": 0.15738936708587337, "maximum": 0.16926753593087845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963298063396925, "standard_deviation": 0.00039532166001720806, "minimum": 0.9958727395122127, "maximum": 0.9971747328733382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1007547064927337, "standard_deviation": 0.04276008235389196, "minimum": 1.0357750129495984, "maximum": 1.1631680484831206, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3622066780465115, "standard_deviation": 0.13957133242861974, "minimum": 0.17988789931833515, "maximum": 0.5642185342384911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06609904656500154, "standard_deviation": 0.0033347556538925556, "minimum": 0.05933253100597554, "maximum": 0.06990201745262498, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9988944671079176, "standard_deviation": 0.00015335628864322803, "minimum": 0.9987409583426898, "maximum": 0.9991108950504174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6011665651898901, "standard_deviation": 0.04298885297735196, "minimum": 0.5285020594226799, "maximum": 0.6738815210705962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.42765559637811046, "standard_deviation": 0.20271978063108448, "minimum": 0.22632025282147397, "maximum": 0.8519498736708883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03668178662944245, "standard_deviation": 0.002838464312531954, "minimum": 0.03211217387221711, "maximum": 0.03992725224236333, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14915698851391904, "standard_deviation": 0.036593361460127684, "minimum": 0.11414179547561831, "maximum": 0.23129816203824205, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.366661838448783, "standard_deviation": 0.011678955721832428, "minimum": 5.348180546754777, "maximum": 5.382215340484282, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3512659589202588, "standard_deviation": 0.018717422027669368, "minimum": 0.33355731594876475, "maximum": 0.3837020293707197, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08714619242824431, "standard_deviation": 0.05398958935864435, "minimum": -0.1592206115505418, "maximum": -0.019312698517825147, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9971809193790566, "standard_deviation": 0.00020990551679204457, "minimum": 0.9968697506487229, "maximum": 0.9974868469922588, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.5739826932232361, "standard_deviation": 0.02176373616431978, "minimum": 0.5448242970694748, "maximum": 0.6080622966125925, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.299974658657865, "standard_deviation": 0.12386862526767105, "minimum": 0.08825844146173185, "maximum": 0.5265521325402466, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.052896269389084816, "standard_deviation": 0.0020220274410657803, "minimum": 0.05050369812899721, "maximum": 0.05683616829006302, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999171118734981, "standard_deviation": 8.531352068657628e-05, "minimum": 0.9990655863253888, "maximum": 0.9992869393879711, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3096872936696877, "standard_deviation": 0.018224728830835955, "minimum": 0.2799379607390981, "maximum": 0.3369576058714818, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.30168859651403745, "standard_deviation": 0.17927045036529143, "minimum": 0.13209530638298028, "maximum": 0.6956975664148816, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.029543794226574868, "standard_deviation": 0.0022888826078950658, "minimum": 0.026413910783468816, "maximum": 0.03275622547958077, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4077-13751-0001", "language": "en", "style": "normal", "speaker": "librispeech_4077", "source_utterance_id": "4077-13751-0001", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4077-13751-0001.wav", "source_file_sha256": "50263d9104646909a2b9c5fd3ea8c9abe695b6c8b655c372c5dc775acf1968a0", "decoded_pcm_sha256": "ff5a7ca9328a329a2ca9ba953b0550653b86fe7bbf7bda2cae73908a1fbfde2a", "original_source_sha256": "1f85c5ad8dfb8fcf3c8d86e54afc055f756f147e7b426ff8b7f48780d8d43c0e", "corpus_entry_id": "librispeech_4077-13751-0001"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.011975288391113, "predicted_mos": 4.011975288391113, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "935015400615ea7003cc3bbd6d27f3f289fa242297b7b2ed9e3a5b447bbd4f4c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.57266622500174, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4077-13751-0001", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "935015400615ea7003cc3bbd6d27f3f289fa242297b7b2ed9e3a5b447bbd4f4c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "e251b8c76c67e0e5339a7ad31f65d12566654050606a6b0e7f395fe9bfdd03bf"}, "rows": [{"case": "librispeech_4077-13751-0001", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.0205321311950684, "predicted_mos": 2.0205321311950684, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "864e3a76833ac698410c4a698649f6ac3cb033af06793fba60fd28011583d4da", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.760329567998269, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4077-13751-0001", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "864e3a76833ac698410c4a698649f6ac3cb033af06793fba60fd28011583d4da", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8472024342411928, "全帯域_10ms_rms_envelope_nrmse": 0.12802923908380948, "全帯域_10ms_envelope_correlation": 0.9847378607935444, "全帯域_10ms_modulation_mae_db": 0.5714630614535585, "全帯域_40ms_log_rms_mae_db": 0.9846100177789018, "全帯域_40ms_rms_envelope_nrmse": 0.07209031281043878, "全帯域_40ms_envelope_correlation": 0.9946001378301392, "全帯域_40ms_modulation_mae_db": 0.41563927682946167, "300–1000 Hz_10ms_log_rms_mae_db": 1.8853064172920349, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12795748654401015, "300–1000 Hz_10ms_envelope_correlation": 0.9894537875570923, "300–1000 Hz_10ms_modulation_mae_db": 0.49185472061707197, "300–1000 Hz_40ms_log_rms_mae_db": 1.191814140788571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0712687094919613, "300–1000 Hz_40ms_envelope_correlation": 0.9966676892716085, "300–1000 Hz_40ms_modulation_mae_db": 0.44665977924659184, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0014026402512886, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18578565751585036, "1000–3000 Hz_10ms_envelope_correlation": 0.976149376533207, "1000–3000 Hz_10ms_modulation_mae_db": 0.2899358551826933, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2088414306364839, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12121224986081304, "1000–3000 Hz_40ms_envelope_correlation": 0.9893937290366404, "1000–3000 Hz_40ms_modulation_mae_db": 0.48558334365678063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9522097542362906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17428061968819575, "3000–6000 Hz_10ms_envelope_correlation": 0.9806038399430326, "3000–6000 Hz_10ms_modulation_mae_db": 0.8230281077450273, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1905968771923285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08866772590032422, "3000–6000 Hz_40ms_envelope_correlation": 0.9948354026746279, "3000–6000 Hz_40ms_modulation_mae_db": 0.9932550628065333, "mr_stft_db": 5.9837988612907465, "band_mae_db": 0.14907444223033797, "rms_gain_error_db": 0.00564947358504323, "output_peak": 0.7104650905361378}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6807562419322393, "全帯域_10ms_rms_envelope_nrmse": 0.10945119491081917, "全帯域_10ms_envelope_correlation": 0.9887552471312708, "全帯域_10ms_modulation_mae_db": 0.39299865016083557, "全帯域_40ms_log_rms_mae_db": 0.925877335955307, "全帯域_40ms_rms_envelope_nrmse": 0.06371522076993842, "全帯域_40ms_envelope_correlation": 0.9958625180841283, "全帯域_40ms_modulation_mae_db": 0.24324877598584685, "300–1000 Hz_10ms_log_rms_mae_db": 1.9277976417428806, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11538069261058029, "300–1000 Hz_10ms_envelope_correlation": 0.9917448932902626, "300–1000 Hz_10ms_modulation_mae_db": 0.5401554249240346, "300–1000 Hz_40ms_log_rms_mae_db": 1.2360166477602808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07530693610218243, "300–1000 Hz_40ms_envelope_correlation": 0.9966334079472868, "300–1000 Hz_40ms_modulation_mae_db": 0.7781344739024526, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0066245027740512, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16089852578785618, "1000–3000 Hz_10ms_envelope_correlation": 0.9820863531531023, "1000–3000 Hz_10ms_modulation_mae_db": 1.0249501035531516, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2275060259124635, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10328014694242205, "1000–3000 Hz_40ms_envelope_correlation": 0.9921137891836921, "1000–3000 Hz_40ms_modulation_mae_db": 1.1451005151162565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.919450050597443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17836579337027594, "3000–6000 Hz_10ms_envelope_correlation": 0.979664630214172, "3000–6000 Hz_10ms_modulation_mae_db": 0.636786174483913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1528261314282628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09167169421856651, "3000–6000 Hz_40ms_envelope_correlation": 0.9945858579989151, "3000–6000 Hz_40ms_modulation_mae_db": 0.76943291287385, "mr_stft_db": 5.997825555533235, "band_mae_db": 0.17720156092796366, "rms_gain_error_db": -0.11500821571932038, "output_peak": 0.7220608936665389}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7763854291681735, "全帯域_10ms_rms_envelope_nrmse": 0.11665388319053915, "全帯域_10ms_envelope_correlation": 0.9872552356847524, "全帯域_10ms_modulation_mae_db": 0.5286849058317048, "全帯域_40ms_log_rms_mae_db": 1.0175572869036105, "全帯域_40ms_rms_envelope_nrmse": 0.07160367376566273, "全帯域_40ms_envelope_correlation": 0.9946699004723127, "全帯域_40ms_modulation_mae_db": 0.46763189504739583, "300–1000 Hz_10ms_log_rms_mae_db": 1.8424484101716516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11472396211900499, "300–1000 Hz_10ms_envelope_correlation": 0.9915294483035638, "300–1000 Hz_10ms_modulation_mae_db": 0.37383338587102993, "300–1000 Hz_40ms_log_rms_mae_db": 1.1715074726008732, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06919827185002195, "300–1000 Hz_40ms_envelope_correlation": 0.9967918942278963, "300–1000 Hz_40ms_modulation_mae_db": 0.7094071186922735, "1000–3000 Hz_10ms_log_rms_mae_db": 2.019000315955206, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18025264069356164, "1000–3000 Hz_10ms_envelope_correlation": 0.97791041160509, "1000–3000 Hz_10ms_modulation_mae_db": 1.295665779922493, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2156832636602444, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11619958706466138, "1000–3000 Hz_40ms_envelope_correlation": 0.9908039133934875, "1000–3000 Hz_40ms_modulation_mae_db": 1.5547289786024499, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9210969495690902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17653602288459982, "3000–6000 Hz_10ms_envelope_correlation": 0.9800478461900158, "3000–6000 Hz_10ms_modulation_mae_db": 0.9334467200906721, "3000–6000 Hz_40ms_log_rms_mae_db": 1.18011750113155, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09260939193119373, "3000–6000 Hz_40ms_envelope_correlation": 0.9943155460710835, "3000–6000 Hz_40ms_modulation_mae_db": 1.174662861317523, "mr_stft_db": 5.987038125892859, "band_mae_db": 0.16481213088892013, "rms_gain_error_db": -0.016382477917921894, "output_peak": 0.70092152994628}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6367104609376044, "全帯域_10ms_rms_envelope_nrmse": 0.12060417755455356, "全帯域_10ms_envelope_correlation": 0.9863573447508455, "全帯域_10ms_modulation_mae_db": 0.19966017395564997, "全帯域_40ms_log_rms_mae_db": 0.923204927121198, "全帯域_40ms_rms_envelope_nrmse": 0.0732149193370035, "全帯域_40ms_envelope_correlation": 0.9944163628294103, "全帯域_40ms_modulation_mae_db": 0.06932711130872732, "300–1000 Hz_10ms_log_rms_mae_db": 1.953973064888238, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11662508106029526, "300–1000 Hz_10ms_envelope_correlation": 0.991238729282024, "300–1000 Hz_10ms_modulation_mae_db": 0.7536982172450415, "300–1000 Hz_40ms_log_rms_mae_db": 1.2171553800220933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06313081329039102, "300–1000 Hz_40ms_envelope_correlation": 0.9973035660620256, "300–1000 Hz_40ms_modulation_mae_db": 1.0131148757690172, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9406523676359395, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1553266819290495, "1000–3000 Hz_10ms_envelope_correlation": 0.9833757013959894, "1000–3000 Hz_10ms_modulation_mae_db": 1.1320357441967868, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1988076853571525, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1054869856188217, "1000–3000 Hz_40ms_envelope_correlation": 0.9919458390867486, "1000–3000 Hz_40ms_modulation_mae_db": 1.298104563316685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9469541375553223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.171399282052242, "3000–6000 Hz_10ms_envelope_correlation": 0.9814778306726034, "3000–6000 Hz_10ms_modulation_mae_db": 0.8236565111092803, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1616621366401434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08956348793503695, "3000–6000 Hz_40ms_envelope_correlation": 0.9952313375614833, "3000–6000 Hz_40ms_modulation_mae_db": 1.1369250939455224, "mr_stft_db": 5.970988162850037, "band_mae_db": 0.13130913585298995, "rms_gain_error_db": -0.042143960289640875, "output_peak": 0.7612388281581857}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6287605246259491, "全帯域_10ms_rms_envelope_nrmse": 0.1134736331539424, "全帯域_10ms_envelope_correlation": 0.9878228286587597, "全帯域_10ms_modulation_mae_db": 0.17220049395044448, "全帯域_40ms_log_rms_mae_db": 0.8450523930664224, "全帯域_40ms_rms_envelope_nrmse": 0.06315737323221864, "全帯域_40ms_envelope_correlation": 0.9958815868174972, "全帯域_40ms_modulation_mae_db": 0.21701652446615805, "300–1000 Hz_10ms_log_rms_mae_db": 1.7943677816278436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11051014403537227, "300–1000 Hz_10ms_envelope_correlation": 0.9923886814186325, "300–1000 Hz_10ms_modulation_mae_db": 0.28509249963140076, "300–1000 Hz_40ms_log_rms_mae_db": 1.130471454726399, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06990712786923464, "300–1000 Hz_40ms_envelope_correlation": 0.997035293879552, "300–1000 Hz_40ms_modulation_mae_db": 0.5990827528470414, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9574578062119647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16202870103572092, "1000–3000 Hz_10ms_envelope_correlation": 0.9820405699852137, "1000–3000 Hz_10ms_modulation_mae_db": 1.5069557131906788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1986699444080509, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10413823866247492, "1000–3000 Hz_40ms_envelope_correlation": 0.9925540791463701, "1000–3000 Hz_40ms_modulation_mae_db": 1.9092999970041962, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9845959200563505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2046948569992342, "3000–6000 Hz_10ms_envelope_correlation": 0.973331234080205, "3000–6000 Hz_10ms_modulation_mae_db": 1.3053092783936722, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2143982541273384, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10957855066994732, "3000–6000 Hz_40ms_envelope_correlation": 0.9926958947151254, "3000–6000 Hz_40ms_modulation_mae_db": 1.8158656390118126, "mr_stft_db": 5.975528164660375, "band_mae_db": 0.16974395183226668, "rms_gain_error_db": -0.10056085173121221, "output_peak": 0.7099994118040914}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7983837975078112, "全帯域_10ms_rms_envelope_nrmse": 0.1210895111761768, "全帯域_10ms_envelope_correlation": 0.986115486157221, "全帯域_10ms_modulation_mae_db": 0.21100772895301723, "全帯域_40ms_log_rms_mae_db": 0.9789443950727162, "全帯域_40ms_rms_envelope_nrmse": 0.06615531094664509, "全帯域_40ms_envelope_correlation": 0.9954992227218633, "全帯域_40ms_modulation_mae_db": 0.2496609108879714, "300–1000 Hz_10ms_log_rms_mae_db": 1.8917776109956306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10955423983618022, "300–1000 Hz_10ms_envelope_correlation": 0.992439544956538, "300–1000 Hz_10ms_modulation_mae_db": 0.6044752346127044, "300–1000 Hz_40ms_log_rms_mae_db": 1.1616494120189285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06535977774364939, "300–1000 Hz_40ms_envelope_correlation": 0.9974467883811546, "300–1000 Hz_40ms_modulation_mae_db": 0.7996426170898027, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9514611774392558, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18202069102847857, "1000–3000 Hz_10ms_envelope_correlation": 0.9769767019985796, "1000–3000 Hz_10ms_modulation_mae_db": 0.5149990959677746, "1000–3000 Hz_40ms_log_rms_mae_db": 1.173165510029729, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10993947332947727, "1000–3000 Hz_40ms_envelope_correlation": 0.9909708974923318, "1000–3000 Hz_40ms_modulation_mae_db": 0.873456236470563, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9809859809639152, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1679948427819057, "3000–6000 Hz_10ms_envelope_correlation": 0.9819625893189854, "3000–6000 Hz_10ms_modulation_mae_db": 0.6713178494305599, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2134208856748105, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08449209530601598, "3000–6000 Hz_40ms_envelope_correlation": 0.9953369333502499, "3000–6000 Hz_40ms_modulation_mae_db": 0.7489939998321685, "mr_stft_db": 5.995534781918955, "band_mae_db": 0.14597012494094772, "rms_gain_error_db": -0.0993030391613983, "output_peak": 0.7233325218860086}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7598672056786635, "全帯域_10ms_rms_envelope_nrmse": 0.12023246453549201, "全帯域_10ms_envelope_correlation": 0.986449682382484, "全帯域_10ms_modulation_mae_db": 0.5286013590256561, "全帯域_40ms_log_rms_mae_db": 0.8937470166993641, "全帯域_40ms_rms_envelope_nrmse": 0.062303347170719954, "全帯域_40ms_envelope_correlation": 0.9959994351534978, "全帯域_40ms_modulation_mae_db": 0.36002047328518233, "300–1000 Hz_10ms_log_rms_mae_db": 1.9049069877505458, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10226264608439939, "300–1000 Hz_10ms_envelope_correlation": 0.993241992768027, "300–1000 Hz_10ms_modulation_mae_db": 0.41923158490458307, "300–1000 Hz_40ms_log_rms_mae_db": 1.1823646753223485, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0649792242896816, "300–1000 Hz_40ms_envelope_correlation": 0.9971262510048647, "300–1000 Hz_40ms_modulation_mae_db": 0.6091029938447813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9865225463653124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17646284033803866, "1000–3000 Hz_10ms_envelope_correlation": 0.978748736015658, "1000–3000 Hz_10ms_modulation_mae_db": 1.154344667104539, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2309354324777808, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10873755500066645, "1000–3000 Hz_40ms_envelope_correlation": 0.9920619199040754, "1000–3000 Hz_40ms_modulation_mae_db": 1.4549799601423004, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9649484088739158, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19455190788237503, "3000–6000 Hz_10ms_envelope_correlation": 0.9762731598664862, "3000–6000 Hz_10ms_modulation_mae_db": 1.053032820950287, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1986613638812305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11097165641267376, "3000–6000 Hz_40ms_envelope_correlation": 0.9929370028589614, "3000–6000 Hz_40ms_modulation_mae_db": 1.2217012558827987, "mr_stft_db": 5.990316373457912, "band_mae_db": 0.1429301266331434, "rms_gain_error_db": -0.06945065790523317, "output_peak": 0.7389709200724898}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8292649323125636, "全帯域_10ms_rms_envelope_nrmse": 0.11975821055138595, "全帯域_10ms_envelope_correlation": 0.9866789198652894, "全帯域_10ms_modulation_mae_db": 0.47051239160852426, "全帯域_40ms_log_rms_mae_db": 1.0300115603809883, "全帯域_40ms_rms_envelope_nrmse": 0.06646128343109867, "全帯域_40ms_envelope_correlation": 0.9957032484172108, "全帯域_40ms_modulation_mae_db": 0.24941129743920523, "300–1000 Hz_10ms_log_rms_mae_db": 1.8317115690958747, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11462313034280473, "300–1000 Hz_10ms_envelope_correlation": 0.9916768709255185, "300–1000 Hz_10ms_modulation_mae_db": 0.49147214286246926, "300–1000 Hz_40ms_log_rms_mae_db": 1.1254190424565804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07112323950291063, "300–1000 Hz_40ms_envelope_correlation": 0.9968389124329168, "300–1000 Hz_40ms_modulation_mae_db": 0.7438089585937454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9090679831150925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15764450609370986, "1000–3000 Hz_10ms_envelope_correlation": 0.9828538361737822, "1000–3000 Hz_10ms_modulation_mae_db": 0.6681140275530542, "1000–3000 Hz_40ms_log_rms_mae_db": 1.127933681228873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0931354410723769, "1000–3000 Hz_40ms_envelope_correlation": 0.9935589969349755, "1000–3000 Hz_40ms_modulation_mae_db": 0.7846401608553948, "3000–6000 Hz_10ms_log_rms_mae_db": 1.998631069815017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1821834959821583, "3000–6000 Hz_10ms_envelope_correlation": 0.9788780539440028, "3000–6000 Hz_10ms_modulation_mae_db": 0.8371988084840071, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1645918351659388, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09834419418983381, "3000–6000 Hz_40ms_envelope_correlation": 0.9939179353995784, "3000–6000 Hz_40ms_modulation_mae_db": 0.9847523244907417, "mr_stft_db": 5.977317941696156, "band_mae_db": 0.15385441176655104, "rms_gain_error_db": -0.1685891481127576, "output_peak": 0.699872331938542}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.319825205145359, "spectral_convergence": 0.4260231813174762, "spectral_convergence_status": "ok", "amplitude_floor": 1.963677544918102e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.792661797325743, "spectral_convergence": 0.37257226228043727, "spectral_convergence_status": "ok", "amplitude_floor": 1.0999213683671372e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.838909581401138, "spectral_convergence": 0.3895708370015359, "spectral_convergence_status": "ok", "amplitude_floor": 9.527550859021228e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.000972009721466654, "output_power": 0.001022432317870367, "delta_db": 0.2196396005900108, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00078771634185205, "output_power": 0.000784835078429161, "delta_db": -0.015914496525508387, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008277001014055567, "output_power": 0.0007900069257950007, "delta_db": -0.2024210974711552, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015338014805248517, "output_power": 0.00014293938705842534, "delta_db": -0.30617237181784557, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000153079755450355, "output_power": 0.0001507692123913032, "delta_db": -0.06605093642582596, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000364753212040036, "output_power": 0.0003718980832813, "delta_db": 0.08424815055168204, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14907444223033797, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.00564947358504323, "pitch": null, "output_rms_dbfs": -24.863987643641234, "output_peak": 0.7104650905361378, "reference_rms_dbfs": -24.869637117226276, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9837988612907465}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002718271425028404, "log_rms_mae_db": 1.8472024342411928, "rms_envelope_nrmse": 0.12802923908380948, "envelope_correlation": 0.9847378607935444, "rms_gain_error_db": 0.00902141571748293, "reference_mean_power": 0.0032374876277477364, "output_mean_power": 0.0032442197136820904}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2735378968683332e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004666640237447372, "output_power": 0.0004810447300133604, "reference_active": true, "delta_db": 0.1318113981982884}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002202584731702016, "output_power": 0.0002468236361173809, "reference_active": true, "delta_db": 0.494541215228675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.478672520125922e-05, "output_power": 4.469070477638954e-05, "reference_active": true, "delta_db": 1.088036570933712}], "active_band_count": 3, "active_band_mae_db": 0.5714630614535585}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002399603663054526, "log_rms_mae_db": 0.9846100177789018, "rms_envelope_nrmse": 0.07209031281043878, "envelope_correlation": 0.9946001378301392, "rms_gain_error_db": 0.01927962839134429, "reference_mean_power": 0.0031195469748277875, "output_mean_power": 0.0031334263595848015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1403707959359127e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046404851798205753, "output_power": 0.0004821225313139068, "reference_active": true, "delta_db": 0.16594038078249967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.861370772885771e-05, "output_power": 0.00010710312963347605, "reference_active": true, "delta_db": 0.35864873491174387}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9585409286275983e-06, "output_power": 2.312946157907038e-06, "reference_active": true, "delta_db": 0.7223287147941415}], "active_band_count": 3, "active_band_mae_db": 0.41563927682946167}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022360065406936124, "log_rms_mae_db": 1.8853064172920349, "rms_envelope_nrmse": 0.12795748654401015, "envelope_correlation": 0.9894537875570923, "rms_gain_error_db": -0.1290047294522082, "reference_mean_power": 0.0012597240435666534, "output_mean_power": 0.0012228549492241233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2605416603296333e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001978275618555469, "output_power": 0.00019078168617048247, "reference_active": true, "delta_db": -0.15750115666074732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.107555923758901e-05, "output_power": 7.377165289690823e-05, "reference_active": true, "delta_db": -0.4100043881432712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1829892137914497e-05, "output_power": 1.458097772692961e-05, "reference_active": true, "delta_db": 0.9080586170471973}], "active_band_count": 3, "active_band_mae_db": 0.49185472061707197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019906826982741464, "log_rms_mae_db": 1.191814140788571, "rms_envelope_nrmse": 0.0712687094919613, "envelope_correlation": 0.9966676892716085, "rms_gain_error_db": -0.11738249133925162, "reference_mean_power": 0.0011746644654148715, "output_mean_power": 0.001143340485458283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9316122263011187e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214336491155288, "output_power": 0.0002035090962840081, "reference_active": true, "delta_db": -0.18045627394850525}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6617395760512454e-05, "output_power": 3.2822988819518506e-05, "reference_active": true, "delta_db": -0.47509329098673414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0023494047314195e-06, "output_power": 8.562018453359272e-07, "reference_active": true, "delta_db": -0.6844297728045361}], "active_band_count": 3, "active_band_mae_db": 0.44665977924659184}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.946276793614836e-05, "log_rms_mae_db": 2.0014026402512886, "rms_envelope_nrmse": 0.18578565751585036, "envelope_correlation": 0.976149376533207, "rms_gain_error_db": -0.26497670166308795, "reference_mean_power": 0.00019501605476176557, "output_mean_power": 0.0001834732255358282}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.789505238315189e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.753576467028983e-05, "output_power": 3.5425064669129095e-05, "reference_active": true, "delta_db": -0.2513461591052816}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.624079451059313e-05, "output_power": 1.4424914771677049e-05, "reference_active": true, "delta_db": -0.5149401555795454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.016250052718367e-06, "output_power": 3.9216480864752024e-06, "reference_active": true, "delta_db": -0.10352125086325295}], "active_band_count": 3, "active_band_mae_db": 0.2899358551826933}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.236492865510781e-05, "log_rms_mae_db": 1.2088414306364839, "rms_envelope_nrmse": 0.12121224986081304, "envelope_correlation": 0.9893937290366404, "rms_gain_error_db": -0.25657139359425535, "reference_mean_power": 0.00017673870709398627, "output_mean_power": 0.00016659982247331498}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.939463723397018e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.781437846450251e-05, "output_power": 3.534235017188273e-05, "reference_active": true, "delta_db": -0.29361540961689886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.77351369725072e-06, "output_power": 6.941079683204843e-06, "reference_active": true, "delta_db": -0.49190338036923054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.249207604003898e-07, "output_power": 2.783904524482758e-07, "reference_active": true, "delta_db": -0.6712312409842124}], "active_band_count": 3, "active_band_mae_db": 0.48558334365678063}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014584429481341953, "log_rms_mae_db": 1.9522097542362906, "rms_envelope_nrmse": 0.17428061968819575, "envelope_correlation": 0.9806038399430326, "rms_gain_error_db": -0.1527008548500185, "reference_mean_power": 0.00024652309966206675, "output_mean_power": 0.00023800579651556047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9096739107120075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.093194518056527e-05, "output_power": 7.537156665355109e-05, "reference_active": true, "delta_db": 0.26365672742323454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.832881668922571e-05, "output_power": 2.4962607501253474e-05, "reference_active": true, "delta_db": -0.5493848600373361}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4378297018808954e-05, "output_power": 9.819812087978707e-06, "reference_active": true, "delta_db": -1.656042735774511}], "active_band_count": 3, "active_band_mae_db": 0.8230281077450273}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.656067210642841e-05, "log_rms_mae_db": 1.1905968771923285, "rms_envelope_nrmse": 0.08866772590032422, "envelope_correlation": 0.9948354026746279, "rms_gain_error_db": -0.15154883885282822, "reference_mean_power": 0.0002460055124763468, "output_mean_power": 0.00023756910139578603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8079454178991016e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.471033966073099e-05, "output_power": 7.644334550295243e-05, "reference_active": true, "delta_db": 0.09958974592926079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5789672835999803e-05, "output_power": 1.2947874417496552e-05, "reference_active": true, "delta_db": -0.8617465297251015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1060211769697704e-06, "output_power": 6.948971831237371e-07, "reference_active": true, "delta_db": -2.0184289127652377}], "active_band_count": 3, "active_band_mae_db": 0.9932550628065333}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7420857499258204, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0e37dbbe072bf77ca6a347d09818fec65813442815ebc24a660d210b326119f0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9800177108575778, "standard_deviation": 0.002877342520872403, "minimum": 0.976149376533207, "maximum": 0.9833757013959894, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9715236674685137, "standard_deviation": 0.03790575115403144, "minimum": 1.9090679831150925, "maximum": 2.019000315955206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9483751233338964, "standard_deviation": 0.41654099652088955, "minimum": 0.2899358551826933, "maximum": 1.5069557131906788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17005253055278322, "standard_deviation": 0.012275933480107727, "minimum": 0.1553266819290495, "maximum": 0.18578565751585036, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9916753955222901, "standard_deviation": 0.001264717333833954, "minimum": 0.9893937290366404, "maximum": 0.9935589969349755, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1976928717138473, "standard_deviation": 0.033619658761466845, "minimum": 1.127933681228873, "maximum": 1.2309354324777808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1882367193955783, "standard_deviation": 0.46214270508437366, "minimum": 0.48558334365678063, "maximum": 1.9092999970041962, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10776620969396422, "standard_deviation": 0.008536690745905733, "minimum": 0.0931354410723769, "maximum": 0.12121224986081304, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.979029898028688, "standard_deviation": 0.0028935119962780866, "minimum": 0.973331234080205, "maximum": 0.9819625893189854, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.958609033958418, "standard_deviation": 0.029118886553982638, "minimum": 1.919450050597443, "maximum": 1.998631069815017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8854720338359274, "standard_deviation": 0.21527353967815469, "minimum": 0.636786174483913, "maximum": 1.3053092783936722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18125085270512334, "standard_deviation": 0.012417279486829712, "minimum": 0.1679948427819057, "maximum": 0.2046948569992342, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9942319888287532, "standard_deviation": 0.000989308351553967, "minimum": 0.9926958947151254, "maximum": 0.9953369333502499, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1845343731552003, "standard_deviation": 0.023629198617020463, "minimum": 1.1528261314282628, "maximum": 1.2143982541273384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1056986437701188, "standard_deviation": 0.3361779582876794, "minimum": 0.7489939998321685, "maximum": 1.8158656390118126, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09573734957044903, "standard_deviation": 0.009793030713908172, "minimum": 0.08449209530601598, "maximum": 0.11097165641267376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9917142435627073, "standard_deviation": 0.0011140150524642733, "minimum": 0.9894537875570923, "maximum": 0.993241992768027, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8790361854455875, "standard_deviation": 0.052930138433843664, "minimum": 1.7943677816278436, "maximum": 1.953973064888238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4949766513335419, "standard_deviation": 0.14414987290662887, "minimum": 0.28509249963140076, "maximum": 0.7536982172450415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11395467282908092, "standard_deviation": 0.0073115771648888245, "minimum": 0.10226264608439939, "maximum": 0.12795748654401015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969804754009132, "standard_deviation": 0.0002976058937405825, "minimum": 0.9966334079472868, "maximum": 0.9974467883811546, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1770497782120093, "standard_deviation": 0.03860967136094797, "minimum": 1.1254190424565804, "maximum": 1.2360166477602808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7123691962482133, "standard_deviation": 0.16796519160707438, "minimum": 0.44665977924659184, "maximum": 1.0131148757690172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06878426251750411, "standard_deviation": 0.004031796678016233, "minimum": 0.06313081329039102, "maximum": 0.07530693610218243, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15436198563414005, "standard_deviation": 0.015251308961114441, "minimum": 0.13130913585298995, "maximum": 0.17720156092796366, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.984793495912534, "standard_deviation": 0.009666267829124938, "minimum": 5.970988162850037, "maximum": 5.997825555533235, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7208576910010343, "standard_deviation": 0.02080536323089982, "minimum": 0.699872331938542, "maximum": 0.7612388281581857, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07572360965655514, "standard_deviation": 0.05688341204787874, "minimum": -0.1685891481127576, "maximum": 0.00564947358504323, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9867715756780209, "standard_deviation": 0.0012026310211413378, "minimum": 0.9847378607935444, "maximum": 0.9887552471312708, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7446663783005247, "standard_deviation": 0.08532830305550618, "minimum": 1.6287605246259491, "maximum": 1.8472024342411928, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38439109561742385, "standard_deviation": 0.1661824936836069, "minimum": 0.17220049395044448, "maximum": 0.5714630614535585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1186615392695898, "standard_deviation": 0.0055660794228425525, "minimum": 0.10945119491081917, "maximum": 0.12802923908380948, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9953290515407575, "standard_deviation": 0.0006554250004794756, "minimum": 0.9944163628294103, "maximum": 0.9959994351534978, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9498756166223135, "standard_deviation": 0.0637849305511008, "minimum": 0.8450523930664224, "maximum": 1.0300115603809883, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28399453315624357, "standard_deviation": 0.12617229318927156, "minimum": 0.06932711130872732, "maximum": 0.46763189504739583, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06733768018296571, "standard_deviation": 0.004366036348426188, "minimum": 0.062303347170719954, "maximum": 0.0732149193370035, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13751-0001", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.0192127227783203, "predicted_mos": 2.0192127227783203, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b17dd76adeb41cd6cfe109585f63123dd583a7e63c247984c2f53eb9c991bd48", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.870262899901718, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8508736050222574, "全帯域_10ms_rms_envelope_nrmse": 0.1308296734100937, "全帯域_10ms_envelope_correlation": 0.9840877390196118, "全帯域_10ms_modulation_mae_db": 0.5852499179542967, "全帯域_40ms_log_rms_mae_db": 0.9892348661367201, "全帯域_40ms_rms_envelope_nrmse": 0.07542868221273279, "全帯域_40ms_envelope_correlation": 0.9941060207110775, "全帯域_40ms_modulation_mae_db": 0.482000889536989, "300–1000 Hz_10ms_log_rms_mae_db": 1.8120470178740335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1297465269603388, "300–1000 Hz_10ms_envelope_correlation": 0.9891444597069285, "300–1000 Hz_10ms_modulation_mae_db": 0.4500012236848789, "300–1000 Hz_40ms_log_rms_mae_db": 1.0948731782051404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07134615182643281, "300–1000 Hz_40ms_envelope_correlation": 0.9966461707492336, "300–1000 Hz_40ms_modulation_mae_db": 0.5149761764187809, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9199002601800208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1848376125156913, "1000–3000 Hz_10ms_envelope_correlation": 0.9764743483111908, "1000–3000 Hz_10ms_modulation_mae_db": 0.3225276242875063, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1270208669094004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11710903528954203, "1000–3000 Hz_40ms_envelope_correlation": 0.990287162382598, "1000–3000 Hz_40ms_modulation_mae_db": 0.466914692012162, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8889173588146273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1806800806266662, "3000–6000 Hz_10ms_envelope_correlation": 0.9791228028650222, "3000–6000 Hz_10ms_modulation_mae_db": 1.0052449030921082, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1239196014576527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09213164883122357, "3000–6000 Hz_40ms_envelope_correlation": 0.9945318677589562, "3000–6000 Hz_40ms_modulation_mae_db": 1.2270831186536035, "mr_stft_db": 5.9262610116834695, "band_mae_db": 0.1586916885313521, "rms_gain_error_db": 0.011798477457287092, "output_peak": 0.7084639259765522}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7398260324012174, "全帯域_10ms_rms_envelope_nrmse": 0.1108814847731318, "全帯域_10ms_envelope_correlation": 0.9885054066445307, "全帯域_10ms_modulation_mae_db": 0.38158823354157506, "全帯域_40ms_log_rms_mae_db": 0.9760455285846872, "全帯域_40ms_rms_envelope_nrmse": 0.06546295296980102, "全帯域_40ms_envelope_correlation": 0.9957109771575425, "全帯域_40ms_modulation_mae_db": 0.2967393047822668, "300–1000 Hz_10ms_log_rms_mae_db": 1.8680822716522192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11972665315525215, "300–1000 Hz_10ms_envelope_correlation": 0.9910422812756056, "300–1000 Hz_10ms_modulation_mae_db": 0.6387199706624801, "300–1000 Hz_40ms_log_rms_mae_db": 1.1388768229860025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07530653432884947, "300–1000 Hz_40ms_envelope_correlation": 0.9965857279393306, "300–1000 Hz_40ms_modulation_mae_db": 0.86422369840569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9041245125735682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1566848202731973, "1000–3000 Hz_10ms_envelope_correlation": 0.983019261141946, "1000–3000 Hz_10ms_modulation_mae_db": 0.9559340207880535, "1000–3000 Hz_40ms_log_rms_mae_db": 1.08632762339351, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09737916705879368, "1000–3000 Hz_40ms_envelope_correlation": 0.9931030911036348, "1000–3000 Hz_40ms_modulation_mae_db": 1.0893066533137468, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8456826793978953, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18411652514748536, "3000–6000 Hz_10ms_envelope_correlation": 0.9782690322704056, "3000–6000 Hz_10ms_modulation_mae_db": 0.7348605616839272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0508763628016944, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09389997583006844, "3000–6000 Hz_40ms_envelope_correlation": 0.9942680884613512, "3000–6000 Hz_40ms_modulation_mae_db": 0.9204155797840871, "mr_stft_db": 5.937492492218813, "band_mae_db": 0.1831814759142968, "rms_gain_error_db": -0.14281080696739917, "output_peak": 0.7212233340719095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.779373452244168, "全帯域_10ms_rms_envelope_nrmse": 0.11898722212517156, "全帯域_10ms_envelope_correlation": 0.986776065615306, "全帯域_10ms_modulation_mae_db": 0.662381123109561, "全帯域_40ms_log_rms_mae_db": 1.014763011764805, "全帯域_40ms_rms_envelope_nrmse": 0.0748387401044168, "全帯域_40ms_envelope_correlation": 0.9942033741887217, "全帯域_40ms_modulation_mae_db": 0.45359848982650036, "300–1000 Hz_10ms_log_rms_mae_db": 1.7701453818484396, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11599194878014037, "300–1000 Hz_10ms_envelope_correlation": 0.9913359962403967, "300–1000 Hz_10ms_modulation_mae_db": 0.3492917782853313, "300–1000 Hz_40ms_log_rms_mae_db": 1.1172859895541454, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07069359198277084, "300–1000 Hz_40ms_envelope_correlation": 0.9966208231252555, "300–1000 Hz_40ms_modulation_mae_db": 0.6519775516507974, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9087318783063627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18045954698772604, "1000–3000 Hz_10ms_envelope_correlation": 0.9781432729924018, "1000–3000 Hz_10ms_modulation_mae_db": 1.488239125572332, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1287641664674466, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11752207382108949, "1000–3000 Hz_40ms_envelope_correlation": 0.9909724551896936, "1000–3000 Hz_40ms_modulation_mae_db": 1.7557797185576487, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8615476788713397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1843618685823513, "3000–6000 Hz_10ms_envelope_correlation": 0.9781906230290995, "3000–6000 Hz_10ms_modulation_mae_db": 1.037413531063229, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1111591875248608, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09590278228016405, "3000–6000 Hz_40ms_envelope_correlation": 0.993824992961749, "3000–6000 Hz_40ms_modulation_mae_db": 1.3620877202698594, "mr_stft_db": 5.9283772497698, "band_mae_db": 0.16982598714065922, "rms_gain_error_db": -0.00626299273803254, "output_peak": 0.7024711118580971}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6403071080981162, "全帯域_10ms_rms_envelope_nrmse": 0.1193921775937128, "全帯域_10ms_envelope_correlation": 0.9866197430024017, "全帯域_10ms_modulation_mae_db": 0.20616884635122545, "全帯域_40ms_log_rms_mae_db": 0.9092147325314084, "全帯域_40ms_rms_envelope_nrmse": 0.0725958882055075, "全帯域_40ms_envelope_correlation": 0.9945122387093179, "全帯域_40ms_modulation_mae_db": 0.06752057902772038, "300–1000 Hz_10ms_log_rms_mae_db": 1.9321396473171901, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11870001600948514, "300–1000 Hz_10ms_envelope_correlation": 0.9909428419284759, "300–1000 Hz_10ms_modulation_mae_db": 0.6610933737868393, "300–1000 Hz_40ms_log_rms_mae_db": 1.1650372198920707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062497852253196796, "300–1000 Hz_40ms_envelope_correlation": 0.9973517570836342, "300–1000 Hz_40ms_modulation_mae_db": 0.8817734126399331, "1000–3000 Hz_10ms_log_rms_mae_db": 1.839178883514076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15269878248787402, "1000–3000 Hz_10ms_envelope_correlation": 0.9840015097996065, "1000–3000 Hz_10ms_modulation_mae_db": 1.1231840894254066, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0798254418664923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09918127204714425, "1000–3000 Hz_40ms_envelope_correlation": 0.9930871648942617, "1000–3000 Hz_40ms_modulation_mae_db": 1.2396255194545895, "3000–6000 Hz_10ms_log_rms_mae_db": 1.859609934545014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1790379413418958, "3000–6000 Hz_10ms_envelope_correlation": 0.9796228087721097, "3000–6000 Hz_10ms_modulation_mae_db": 0.9801576266165736, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0769065615622184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09296590608403604, "3000–6000 Hz_40ms_envelope_correlation": 0.9946888108254791, "3000–6000 Hz_40ms_modulation_mae_db": 1.3543409288936779, "mr_stft_db": 5.912591484032611, "band_mae_db": 0.14781750567207563, "rms_gain_error_db": -0.04383127323142187, "output_peak": 0.7608444832590888}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.674099249470791, "全帯域_10ms_rms_envelope_nrmse": 0.11333348805067311, "全帯域_10ms_envelope_correlation": 0.9878488297715586, "全帯域_10ms_modulation_mae_db": 0.2475646391359679, "全帯域_40ms_log_rms_mae_db": 0.8725006961019063, "全帯域_40ms_rms_envelope_nrmse": 0.06394733493522033, "全帯域_40ms_envelope_correlation": 0.9957784284618212, "全帯域_40ms_modulation_mae_db": 0.2608312761094253, "300–1000 Hz_10ms_log_rms_mae_db": 1.7419346890797773, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11188689174892054, "300–1000 Hz_10ms_envelope_correlation": 0.9921530050293292, "300–1000 Hz_10ms_modulation_mae_db": 0.3751732154412874, "300–1000 Hz_40ms_log_rms_mae_db": 1.0695789052460911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07103897168538185, "300–1000 Hz_40ms_envelope_correlation": 0.9968841196939342, "300–1000 Hz_40ms_modulation_mae_db": 0.6052868325840698, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8786906984990388, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1650881355281786, "1000–3000 Hz_10ms_envelope_correlation": 0.9815184468432598, "1000–3000 Hz_10ms_modulation_mae_db": 1.4674649970777385, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1114578239829684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10274783753517783, "1000–3000 Hz_40ms_envelope_correlation": 0.9932726560622803, "1000–3000 Hz_40ms_modulation_mae_db": 2.0529259692490616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8984437430546186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20912906958440347, "3000–6000 Hz_10ms_envelope_correlation": 0.9718525967161951, "3000–6000 Hz_10ms_modulation_mae_db": 1.2267881712507747, "3000–6000 Hz_40ms_log_rms_mae_db": 1.09144411349891, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1066679169847937, "3000–6000 Hz_40ms_envelope_correlation": 0.9925988007487667, "3000–6000 Hz_40ms_modulation_mae_db": 1.7802589362084078, "mr_stft_db": 5.919808880194655, "band_mae_db": 0.160520336051469, "rms_gain_error_db": -0.10021923048491735, "output_peak": 0.7105970086945317}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8078044417167456, "全帯域_10ms_rms_envelope_nrmse": 0.12212217540872716, "全帯域_10ms_envelope_correlation": 0.9858786507423122, "全帯域_10ms_modulation_mae_db": 0.24704581254184177, "全帯域_40ms_log_rms_mae_db": 0.9956195333283975, "全帯域_40ms_rms_envelope_nrmse": 0.0664796803818132, "全帯域_40ms_envelope_correlation": 0.9954646145429223, "全帯域_40ms_modulation_mae_db": 0.2579351221505689, "300–1000 Hz_10ms_log_rms_mae_db": 1.8115762496787053, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11032900430279914, "300–1000 Hz_10ms_envelope_correlation": 0.9923464840758759, "300–1000 Hz_10ms_modulation_mae_db": 0.7706809805700011, "300–1000 Hz_40ms_log_rms_mae_db": 1.0710835321377352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0662137158491267, "300–1000 Hz_40ms_envelope_correlation": 0.9973818811687049, "300–1000 Hz_40ms_modulation_mae_db": 0.9168029536949144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8963484831345412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18332064599673398, "1000–3000 Hz_10ms_envelope_correlation": 0.9766361874691942, "1000–3000 Hz_10ms_modulation_mae_db": 0.5072471457878435, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1411217884484053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1110884453493293, "1000–3000 Hz_40ms_envelope_correlation": 0.9908997037041483, "1000–3000 Hz_40ms_modulation_mae_db": 0.8365226826422333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.881385090532634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16577185918506163, "3000–6000 Hz_10ms_envelope_correlation": 0.9824147903505724, "3000–6000 Hz_10ms_modulation_mae_db": 0.7031772242521318, "3000–6000 Hz_40ms_log_rms_mae_db": 1.11374613196059, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08140974520507736, "3000–6000 Hz_40ms_envelope_correlation": 0.9956624883311292, "3000–6000 Hz_40ms_modulation_mae_db": 0.7884646982524638, "mr_stft_db": 5.938247112240191, "band_mae_db": 0.1618629707461525, "rms_gain_error_db": -0.10605890340876864, "output_peak": 0.7265336451429438}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7857803756554, "全帯域_10ms_rms_envelope_nrmse": 0.12135027843695302, "全帯域_10ms_envelope_correlation": 0.9862375194900611, "全帯域_10ms_modulation_mae_db": 0.5772686042404516, "全帯域_40ms_log_rms_mae_db": 0.906051548615257, "全帯域_40ms_rms_envelope_nrmse": 0.06169890797775545, "全帯域_40ms_envelope_correlation": 0.9960940518910928, "全帯域_40ms_modulation_mae_db": 0.41177348724651597, "300–1000 Hz_10ms_log_rms_mae_db": 1.8489169717056577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10297939360188584, "300–1000 Hz_10ms_envelope_correlation": 0.9931519361280913, "300–1000 Hz_10ms_modulation_mae_db": 0.5396231800967399, "300–1000 Hz_40ms_log_rms_mae_db": 1.080232669506316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06211776671720647, "300–1000 Hz_40ms_envelope_correlation": 0.997370043624266, "300–1000 Hz_40ms_modulation_mae_db": 0.7009548671009306, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8939498440861957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17487092058715045, "1000–3000 Hz_10ms_envelope_correlation": 0.979242329571412, "1000–3000 Hz_10ms_modulation_mae_db": 1.3295041024152487, "1000–3000 Hz_40ms_log_rms_mae_db": 1.09301806386489, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10652572697437256, "1000–3000 Hz_40ms_envelope_correlation": 0.9927272154412433, "1000–3000 Hz_40ms_modulation_mae_db": 1.5953801884263565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8959730262503034, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18947643787700716, "3000–6000 Hz_10ms_envelope_correlation": 0.9773427729072502, "3000–6000 Hz_10ms_modulation_mae_db": 1.020128041366941, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1186948525578169, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10148738337896013, "3000–6000 Hz_40ms_envelope_correlation": 0.9940247090054539, "3000–6000 Hz_40ms_modulation_mae_db": 1.2800647212529428, "mr_stft_db": 5.933448082909877, "band_mae_db": 0.14542088670765155, "rms_gain_error_db": -0.06608519809521357, "output_peak": 0.7623040664568955}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.82430803446348, "全帯域_10ms_rms_envelope_nrmse": 0.11858445121995675, "全帯域_10ms_envelope_correlation": 0.9868919529043627, "全帯域_10ms_modulation_mae_db": 0.4861988261014177, "全帯域_40ms_log_rms_mae_db": 1.0196252067338694, "全帯域_40ms_rms_envelope_nrmse": 0.06545054562650825, "全帯域_40ms_envelope_correlation": 0.9957606727373255, "全帯域_40ms_modulation_mae_db": 0.22769344023914684, "300–1000 Hz_10ms_log_rms_mae_db": 1.7644029556890826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11386192819692345, "300–1000 Hz_10ms_envelope_correlation": 0.9917454017080307, "300–1000 Hz_10ms_modulation_mae_db": 0.5267511996159079, "300–1000 Hz_40ms_log_rms_mae_db": 1.053675629252413, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06993947491015629, "300–1000 Hz_40ms_envelope_correlation": 0.9968891502646485, "300–1000 Hz_40ms_modulation_mae_db": 0.7895956988014697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.78173089436969, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1550508533007541, "1000–3000 Hz_10ms_envelope_correlation": 0.9833927205219364, "1000–3000 Hz_10ms_modulation_mae_db": 0.9223616420183389, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0106680347174337, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0912903363674359, "1000–3000 Hz_40ms_envelope_correlation": 0.9938909026877324, "1000–3000 Hz_40ms_modulation_mae_db": 1.121649032892804, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9037169115530836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18647998012655478, "3000–6000 Hz_10ms_envelope_correlation": 0.9777458015270082, "3000–6000 Hz_10ms_modulation_mae_db": 0.8394765396009567, "3000–6000 Hz_40ms_log_rms_mae_db": 1.093988520435319, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09475012183451566, "3000–6000 Hz_40ms_envelope_correlation": 0.9942774334042365, "3000–6000 Hz_40ms_modulation_mae_db": 1.1308905611125977, "mr_stft_db": 5.913851550631056, "band_mae_db": 0.1526702195138203, "rms_gain_error_db": -0.14892857189262107, "output_peak": 0.6980140664255167}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.247034433524915, "spectral_convergence": 0.4248991778285463, "spectral_convergence_status": "ok", "amplitude_floor": 1.963677544918102e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.740596479179723, "spectral_convergence": 0.37574441418947324, "spectral_convergence_status": "ok", "amplitude_floor": 1.0999213683671372e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.791152122345773, "spectral_convergence": 0.3897834951502303, "spectral_convergence_status": "ok", "amplitude_floor": 9.527550859021228e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.000972009721466654, "output_power": 0.0010317427489221877, "delta_db": 0.25900816838184226, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00078771634185205, "output_power": 0.0007837874865148276, "delta_db": -0.021715298649473, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008277001014055567, "output_power": 0.0007942838269019874, "delta_db": -0.17897288811449785, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015338014805248517, "output_power": 0.00014186405386310702, "delta_db": -0.33896786689115854, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000153079755450355, "output_power": 0.00014871807093144978, "delta_db": -0.12554016323144657, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000364753212040036, "output_power": 0.0003671078739386853, "delta_db": 0.027945745919694218, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1586916885313521, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.011798477457287092, "pitch": null, "output_rms_dbfs": -24.85783863976899, "output_peak": 0.7084639259765522, "reference_rms_dbfs": -24.869637117226276, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9262610116834695}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002718271425028404, "log_rms_mae_db": 1.8508736050222574, "rms_envelope_nrmse": 0.1308296734100937, "envelope_correlation": 0.9840877390196118, "rms_gain_error_db": 0.01501892154751924, "reference_mean_power": 0.0032374876277477364, "output_mean_power": 0.003248703000629885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2735378968683332e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004666640237447372, "output_power": 0.0004871512879074063, "reference_active": true, "delta_db": 0.1865953368800345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002202584731702016, "output_power": 0.00025089956404794817, "reference_active": true, "delta_db": 0.5656728233173116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.478672520125922e-05, "output_power": 4.382901452457489e-05, "reference_active": true, "delta_db": 1.003481593665544}], "active_band_count": 3, "active_band_mae_db": 0.5852499179542967}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002399603663054526, "log_rms_mae_db": 0.9892348661367201, "rms_envelope_nrmse": 0.07542868221273279, "envelope_correlation": 0.9941060207110775, "rms_gain_error_db": 0.025988370214369404, "reference_mean_power": 0.0031195469748277875, "output_mean_power": 0.0031382704444208784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1403707959359127e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046404851798205753, "output_power": 0.000489022726604364, "reference_active": true, "delta_db": 0.22765652766402572}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.861370772885771e-05, "output_power": 0.00010934681292379546, "reference_active": true, "delta_db": 0.44868841461334863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9585409286275983e-06, "output_power": 2.338290279638601e-06, "reference_active": true, "delta_db": 0.7696577263335926}], "active_band_count": 3, "active_band_mae_db": 0.482000889536989}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022360065406936124, "log_rms_mae_db": 1.8120470178740335, "rms_envelope_nrmse": 0.1297465269603388, "envelope_correlation": 0.9891444597069285, "rms_gain_error_db": -0.1300887940558207, "reference_mean_power": 0.0012597240435666534, "output_mean_power": 0.0012225497442576304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2605416603296333e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001978275618555469, "output_power": 0.00019288726947492855, "reference_active": true, "delta_db": -0.10983233292437405}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.107555923758901e-05, "output_power": 7.308148339767204e-05, "reference_active": true, "delta_db": -0.4508259898459206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1829892137914497e-05, "output_power": 1.4187808556670641e-05, "reference_active": true, "delta_db": 0.789345348284342}], "active_band_count": 3, "active_band_mae_db": 0.4500012236848789}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019906826982741464, "log_rms_mae_db": 1.0948731782051404, "rms_envelope_nrmse": 0.07134615182643281, "envelope_correlation": 0.9966461707492336, "rms_gain_error_db": -0.11709404551607967, "reference_mean_power": 0.0011746644654148715, "output_mean_power": 0.0011434164253454763}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9316122263011187e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214336491155288, "output_power": 0.00020523983751579783, "reference_active": true, "delta_db": -0.1436779096995095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6617395760512454e-05, "output_power": 3.239198237446483e-05, "reference_active": true, "delta_db": -0.5324992647163541}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0023494047314195e-06, "output_power": 8.20623597719547e-07, "reference_active": true, "delta_db": -0.8687513548404792}], "active_band_count": 3, "active_band_mae_db": 0.5149761764187809}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.946276793614836e-05, "log_rms_mae_db": 1.9199002601800208, "rms_envelope_nrmse": 0.1848376125156913, "envelope_correlation": 0.9764743483111908, "rms_gain_error_db": -0.31063488358526603, "reference_mean_power": 0.00019501605476176557, "output_mean_power": 0.00018155444152896403}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.789505238315189e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.753576467028983e-05, "output_power": 3.52865713931592e-05, "reference_active": true, "delta_db": -0.2683580529861943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.624079451059313e-05, "output_power": 1.4421223230108072e-05, "reference_active": true, "delta_db": -0.5160517193817958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.016250052718367e-06, "output_power": 4.18926713129894e-06, "reference_active": true, "delta_db": 0.18317310049452867}], "active_band_count": 3, "active_band_mae_db": 0.3225276242875063}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.236492865510781e-05, "log_rms_mae_db": 1.1270208669094004, "rms_envelope_nrmse": 0.11710903528954203, "envelope_correlation": 0.990287162382598, "rms_gain_error_db": -0.3007348991492738, "reference_mean_power": 0.00017673870709398627, "output_mean_power": 0.00016491424989275422}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.939463723397018e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.781437846450251e-05, "output_power": 3.535169413606735e-05, "reference_active": true, "delta_db": -0.29246735474244917}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.77351369725072e-06, "output_power": 6.883861093323856e-06, "reference_active": true, "delta_db": -0.5278526985464057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.249207604003898e-07, "output_power": 2.8427263731483143e-07, "reference_active": true, "delta_db": -0.5804240227476313}], "active_band_count": 3, "active_band_mae_db": 0.466914692012162}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014584429481341953, "log_rms_mae_db": 1.8889173588146273, "rms_envelope_nrmse": 0.1806800806266662, "envelope_correlation": 0.9791228028650222, "rms_gain_error_db": -0.21440574354704217, "reference_mean_power": 0.00024652309966206675, "output_mean_power": 0.00023464810181917414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9096739107120075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.093194518056527e-05, "output_power": 7.419621702386914e-05, "reference_active": true, "delta_db": 0.19539893105227815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.832881668922571e-05, "output_power": 2.458925662349414e-05, "reference_active": true, "delta_db": -0.6148303472134059}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4378297018808954e-05, "output_power": 8.652804823352993e-06, "reference_active": true, "delta_db": -2.2055054310106406}], "active_band_count": 3, "active_band_mae_db": 1.0052449030921082}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.656067210642841e-05, "log_rms_mae_db": 1.1239196014576527, "rms_envelope_nrmse": 0.09213164883122357, "envelope_correlation": 0.9945318677589562, "rms_gain_error_db": -0.21278858046886742, "reference_mean_power": 0.0002460055124763468, "output_mean_power": 0.0002342426544260197}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8079454178991016e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.471033966073099e-05, "output_power": 7.489354775534893e-05, "reference_active": true, "delta_db": 0.01063693005335189}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5789672835999803e-05, "output_power": 1.2501644789147416e-05, "reference_active": true, "delta_db": -1.0140597638869895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1060211769697704e-06, "output_power": 5.999404938717513e-07, "reference_active": true, "delta_db": -2.656552662020469}], "active_band_count": 3, "active_band_mae_db": 1.2270831186536035}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7443256163157115, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c172ce313cec9b49070fc2ce9715ffae3dbcd3350db33820f324698f26812499", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9803035095813685, "standard_deviation": 0.0030699991060919377, "minimum": 0.9764743483111908, "maximum": 0.9840015097996065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8778319318329366, "standard_deviation": 0.045901611452461716, "minimum": 1.78173089436969, "maximum": 1.9199002601800208, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0145578434215585, "standard_deviation": 0.42906817881970877, "minimum": 0.3225276242875063, "maximum": 1.488239125572332, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16912641470966322, "standard_deviation": 0.013356895884036855, "minimum": 0.15269878248787402, "maximum": 0.1848376125156913, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.992280043933199, "standard_deviation": 0.0013467241927207826, "minimum": 0.990287162382598, "maximum": 0.9938909026877324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0972754762063184, "standard_deviation": 0.041305996847390176, "minimum": 1.0106680347174337, "maximum": 1.1411217884484053, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2697630570685754, "standard_deviation": 0.5131493971805884, "minimum": 0.466914692012162, "maximum": 2.0529259692490616, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10535548680536064, "standard_deviation": 0.009462549591786464, "minimum": 0.0912903363674359, "maximum": 0.11752207382108949, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9780701535547078, "standard_deviation": 0.0029698735712564966, "minimum": 0.9718525967161951, "maximum": 0.9824147903505724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8794095528774395, "standard_deviation": 0.021281686554580886, "minimum": 1.8456826793978953, "maximum": 1.9037169115530836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9434058248658302, "standard_deviation": 0.17411255464041092, "minimum": 0.7031772242521318, "maximum": 1.2267881712507747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18488172030892822, "standard_deviation": 0.012126982318798859, "minimum": 0.16577185918506163, "maximum": 0.20912906958440347, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9942346489371403, "standard_deviation": 0.0008638922043194478, "minimum": 0.9925988007487667, "maximum": 0.9956624883311292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0975919164748826, "standard_deviation": 0.024624370149680573, "minimum": 1.0508763628016944, "maximum": 1.1239196014576527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.230450783053455, "standard_deviation": 0.3020938134789447, "minimum": 0.7884646982524638, "maximum": 1.7802589362084078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09490193505360486, "standard_deviation": 0.007341632807826216, "minimum": 0.08140974520507736, "maximum": 0.1066679169847937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9914828007615917, "standard_deviation": 0.0011967275719941192, "minimum": 0.9891444597069285, "maximum": 0.9931519361280913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8186556481056382, "standard_deviation": 0.06269410418230376, "minimum": 1.7419346890797773, "maximum": 1.9321396473171901, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5389168652679333, "standard_deviation": 0.14610491060156866, "minimum": 0.3492917782853313, "maximum": 0.7706809805700011, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11540279534446818, "standard_deviation": 0.007841986491884292, "minimum": 0.10297939360188584, "maximum": 0.1297465269603388, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996966209206126, "standard_deviation": 0.00035125723280903003, "minimum": 0.9965857279393306, "maximum": 0.9973818811687049, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0988304933474893, "standard_deviation": 0.03845905969160448, "minimum": 1.053675629252413, "maximum": 1.1650372198920707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7406988989120733, "standard_deviation": 0.14506263196354505, "minimum": 0.5149761764187809, "maximum": 0.9168029536949144, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06864425744414016, "standard_deviation": 0.00462260120957574, "minimum": 0.06211776671720647, "maximum": 0.07530653432884947, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15999888378468463, "standard_deviation": 0.012271181814136294, "minimum": 0.14542088670765155, "maximum": 0.1831814759142968, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.926259732960059, "standard_deviation": 0.01006000508023136, "minimum": 5.912591484032611, "maximum": 5.938247112240191, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.723806455235692, "standard_deviation": 0.02506653588496121, "minimum": 0.6980140664255167, "maximum": 0.7623040664568955, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07529981242013589, "standard_deviation": 0.059729629317176985, "minimum": -0.14892857189262107, "maximum": 0.011798477457287092, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9866057383987681, "standard_deviation": 0.0013239391621880166, "minimum": 0.9840877390196118, "maximum": 0.9885054066445307, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.762796537384022, "standard_deviation": 0.07345407677707796, "minimum": 1.6403071080981162, "maximum": 1.8508736050222574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42418325037204213, "standard_deviation": 0.17803618280729222, "minimum": 0.20616884635122545, "maximum": 0.662381123109561, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11943511887730249, "standard_deviation": 0.00600053126210994, "minimum": 0.1108814847731318, "maximum": 0.1308296734100937, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9952037972999777, "standard_deviation": 0.0007966299220563881, "minimum": 0.9941060207110775, "maximum": 0.9960940518910928, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9603818904746313, "standard_deviation": 0.056162494519255736, "minimum": 0.8725006961019063, "maximum": 1.0196252067338694, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3072615736148917, "standard_deviation": 0.13699712378765513, "minimum": 0.06752057902772038, "maximum": 0.482000889536989, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06823784155171941, "standard_deviation": 0.005264420697728974, "minimum": 0.06169890797775545, "maximum": 0.07542868221273279, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13751-0001", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.56133770942688, "predicted_mos": 2.56133770942688, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "24d471690bf66f5573ae82c9efd942883ed7e79e1ce94d28877b2144da2c782f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9577802000567317, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7249129667924288, "全帯域_10ms_rms_envelope_nrmse": 0.10931027869072489, "全帯域_10ms_envelope_correlation": 0.988763680707961, "全帯域_10ms_modulation_mae_db": 0.535458837238341, "全帯域_40ms_log_rms_mae_db": 0.9337873944320029, "全帯域_40ms_rms_envelope_nrmse": 0.06007391997163566, "全帯域_40ms_envelope_correlation": 0.996321558348865, "全帯域_40ms_modulation_mae_db": 0.7019002443969665, "300–1000 Hz_10ms_log_rms_mae_db": 2.357149986791226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12095214835797458, "300–1000 Hz_10ms_envelope_correlation": 0.9907998172869538, "300–1000 Hz_10ms_modulation_mae_db": 0.898697942637774, "300–1000 Hz_40ms_log_rms_mae_db": 1.469969101729349, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06845765387021369, "300–1000 Hz_40ms_envelope_correlation": 0.9972339940389426, "300–1000 Hz_40ms_modulation_mae_db": 1.3496855087303843, "1000–3000 Hz_10ms_log_rms_mae_db": 2.7056951594608396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2008616668315551, "1000–3000 Hz_10ms_envelope_correlation": 0.9720568242433284, "1000–3000 Hz_10ms_modulation_mae_db": 1.1712559186947233, "1000–3000 Hz_40ms_log_rms_mae_db": 1.684251198395865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12498961399812084, "1000–3000 Hz_40ms_envelope_correlation": 0.9885758812849599, "1000–3000 Hz_40ms_modulation_mae_db": 1.537309737108423, "3000–6000 Hz_10ms_log_rms_mae_db": 2.666166491370586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20506532507090783, "3000–6000 Hz_10ms_envelope_correlation": 0.9732017818448547, "3000–6000 Hz_10ms_modulation_mae_db": 1.4159185727709904, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6399601008391418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11663410954053445, "3000–6000 Hz_40ms_envelope_correlation": 0.9909586539305565, "3000–6000 Hz_40ms_modulation_mae_db": 1.6039408109518023, "mr_stft_db": 6.427942429165756, "band_mae_db": 0.14144378579091957, "rms_gain_error_db": -0.10239113560445308, "output_peak": 0.7307252323628555}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.657799508599202, "全帯域_10ms_rms_envelope_nrmse": 0.1027870541536675, "全帯域_10ms_envelope_correlation": 0.9900720501147757, "全帯域_10ms_modulation_mae_db": 0.14771948121409398, "全帯域_40ms_log_rms_mae_db": 0.7844232146716533, "全帯域_40ms_rms_envelope_nrmse": 0.05332968373042016, "全帯域_40ms_envelope_correlation": 0.9971465948270727, "全帯域_40ms_modulation_mae_db": 0.13983206354648478, "300–1000 Hz_10ms_log_rms_mae_db": 2.369710266468788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10523048143395462, "300–1000 Hz_10ms_envelope_correlation": 0.9929443274164057, "300–1000 Hz_10ms_modulation_mae_db": 0.7318549747496231, "300–1000 Hz_40ms_log_rms_mae_db": 1.4720355359039194, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05633912265550192, "300–1000 Hz_40ms_envelope_correlation": 0.9979128887061984, "300–1000 Hz_40ms_modulation_mae_db": 0.9802568111306229, "1000–3000 Hz_10ms_log_rms_mae_db": 2.7906114417010364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18387989841965796, "1000–3000 Hz_10ms_envelope_correlation": 0.976813510910963, "1000–3000 Hz_10ms_modulation_mae_db": 0.6783016404464085, "1000–3000 Hz_40ms_log_rms_mae_db": 1.784598496003305, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11800471561256803, "1000–3000 Hz_40ms_envelope_correlation": 0.9896981357776348, "1000–3000 Hz_40ms_modulation_mae_db": 0.672714603097444, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7165544100165784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19893995859760288, "3000–6000 Hz_10ms_envelope_correlation": 0.9750991483519064, "3000–6000 Hz_10ms_modulation_mae_db": 1.6638200807100247, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6780866589348287, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10959090722815139, "3000–6000 Hz_40ms_envelope_correlation": 0.9926280981559875, "3000–6000 Hz_40ms_modulation_mae_db": 1.8049260329753078, "mr_stft_db": 6.414946908155241, "band_mae_db": 0.12679534973663784, "rms_gain_error_db": -0.1141767959440202, "output_peak": 0.7213946000175125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6732448713178614, "全帯域_10ms_rms_envelope_nrmse": 0.12416097356392457, "全帯域_10ms_envelope_correlation": 0.9858143342697647, "全帯域_10ms_modulation_mae_db": 1.069948768838252, "全帯域_40ms_log_rms_mae_db": 0.8678331863577996, "全帯域_40ms_rms_envelope_nrmse": 0.07855727636358585, "全帯域_40ms_envelope_correlation": 0.9937329316046821, "全帯域_40ms_modulation_mae_db": 1.5392341834939405, "300–1000 Hz_10ms_log_rms_mae_db": 2.4263303119338877, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12472567278442034, "300–1000 Hz_10ms_envelope_correlation": 0.9900280372085432, "300–1000 Hz_10ms_modulation_mae_db": 0.7460977032359232, "300–1000 Hz_40ms_log_rms_mae_db": 1.618935727129187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07738862203934413, "300–1000 Hz_40ms_envelope_correlation": 0.9959786201915114, "300–1000 Hz_40ms_modulation_mae_db": 1.0281293176527517, "1000–3000 Hz_10ms_log_rms_mae_db": 2.620974315333456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.188830918969756, "1000–3000 Hz_10ms_envelope_correlation": 0.9755376794387142, "1000–3000 Hz_10ms_modulation_mae_db": 1.268311385578447, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5901669134898058, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11601803050051837, "1000–3000 Hz_40ms_envelope_correlation": 0.9904528429176449, "1000–3000 Hz_40ms_modulation_mae_db": 1.3510823780007126, "3000–6000 Hz_10ms_log_rms_mae_db": 2.698068703252357, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21415040347805073, "3000–6000 Hz_10ms_envelope_correlation": 0.9709372442834396, "3000–6000 Hz_10ms_modulation_mae_db": 1.6589528709895174, "3000–6000 Hz_40ms_log_rms_mae_db": 1.697383687512715, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10723514483919178, "3000–6000 Hz_40ms_envelope_correlation": 0.9929775520490615, "3000–6000 Hz_40ms_modulation_mae_db": 1.8488606809765622, "mr_stft_db": 6.439930544972866, "band_mae_db": 0.12156773725286003, "rms_gain_error_db": 0.06844451095937411, "output_peak": 0.7801546462727471}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8059017281425926, "全帯域_10ms_rms_envelope_nrmse": 0.1180450468994083, "全帯域_10ms_envelope_correlation": 0.9868305373029952, "全帯域_10ms_modulation_mae_db": 0.35708796585895614, "全帯域_40ms_log_rms_mae_db": 0.9941056036151297, "全帯域_40ms_rms_envelope_nrmse": 0.06138193150909674, "全帯域_40ms_envelope_correlation": 0.996060865008841, "全帯域_40ms_modulation_mae_db": 0.5040171503364367, "300–1000 Hz_10ms_log_rms_mae_db": 2.323478302966423, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11647771111546636, "300–1000 Hz_10ms_envelope_correlation": 0.9912833342549209, "300–1000 Hz_10ms_modulation_mae_db": 0.5082593825710937, "300–1000 Hz_40ms_log_rms_mae_db": 1.4717289599851715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06653955819432893, "300–1000 Hz_40ms_envelope_correlation": 0.9970906145280746, "300–1000 Hz_40ms_modulation_mae_db": 0.966778290714544, "1000–3000 Hz_10ms_log_rms_mae_db": 2.687410924330149, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20290407939794255, "1000–3000 Hz_10ms_envelope_correlation": 0.9715557392420073, "1000–3000 Hz_10ms_modulation_mae_db": 1.2144686868159849, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6699884336927602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12938217050220732, "1000–3000 Hz_40ms_envelope_correlation": 0.9876682696938531, "1000–3000 Hz_40ms_modulation_mae_db": 1.617578636450143, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7556225803055714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22695868221512647, "3000–6000 Hz_10ms_envelope_correlation": 0.9672336554860352, "3000–6000 Hz_10ms_modulation_mae_db": 1.6272690942429895, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7400531942689073, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12422685675537594, "3000–6000 Hz_40ms_envelope_correlation": 0.9905516488114398, "3000–6000 Hz_40ms_modulation_mae_db": 1.9435237096426352, "mr_stft_db": 6.4215859597977065, "band_mae_db": 0.11319768981325866, "rms_gain_error_db": -0.06038376117239658, "output_peak": 0.7099081247087033}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6992992263816145, "全帯域_10ms_rms_envelope_nrmse": 0.11055594333936808, "全帯域_10ms_envelope_correlation": 0.9885759790740365, "全帯域_10ms_modulation_mae_db": 0.2641777476060356, "全帯域_40ms_log_rms_mae_db": 0.9133027717264226, "全帯域_40ms_rms_envelope_nrmse": 0.05819223880731149, "全帯域_40ms_envelope_correlation": 0.996660665878288, "全帯域_40ms_modulation_mae_db": 0.6322550456739213, "300–1000 Hz_10ms_log_rms_mae_db": 2.361903182546151, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1214581716086184, "300–1000 Hz_10ms_envelope_correlation": 0.9904817566461936, "300–1000 Hz_10ms_modulation_mae_db": 0.4873596218997682, "300–1000 Hz_40ms_log_rms_mae_db": 1.5375694955081847, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07989450473021535, "300–1000 Hz_40ms_envelope_correlation": 0.9956949410992719, "300–1000 Hz_40ms_modulation_mae_db": 0.66149905978119, "1000–3000 Hz_10ms_log_rms_mae_db": 2.6564755700841958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19612084680177022, "1000–3000 Hz_10ms_envelope_correlation": 0.9734830264919216, "1000–3000 Hz_10ms_modulation_mae_db": 0.9030881308712649, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6780314604599722, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12169114160019447, "1000–3000 Hz_40ms_envelope_correlation": 0.9893199284732254, "1000–3000 Hz_40ms_modulation_mae_db": 1.0838459623677739, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7588234677001675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22197028144729283, "3000–6000 Hz_10ms_envelope_correlation": 0.9685338622813512, "3000–6000 Hz_10ms_modulation_mae_db": 1.394919816541818, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7376087910040434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1201229860474934, "3000–6000 Hz_40ms_envelope_correlation": 0.9905561351168651, "3000–6000 Hz_40ms_modulation_mae_db": 1.7148598238095925, "mr_stft_db": 6.427898640703276, "band_mae_db": 0.19114832552783337, "rms_gain_error_db": -0.13065680128678775, "output_peak": 0.7182412075072797}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6766836289753795, "全帯域_10ms_rms_envelope_nrmse": 0.11463783691122709, "全帯域_10ms_envelope_correlation": 0.9877289897305562, "全帯域_10ms_modulation_mae_db": 0.3061389056729303, "全帯域_40ms_log_rms_mae_db": 0.9082092847472791, "全帯域_40ms_rms_envelope_nrmse": 0.06517958478935829, "全帯域_40ms_envelope_correlation": 0.9957770866563546, "全帯域_40ms_modulation_mae_db": 0.3383332384027518, "300–1000 Hz_10ms_log_rms_mae_db": 2.3317449827619554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1075195404469519, "300–1000 Hz_10ms_envelope_correlation": 0.9927116037058069, "300–1000 Hz_10ms_modulation_mae_db": 0.8791036326167472, "300–1000 Hz_40ms_log_rms_mae_db": 1.473827509143437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06767163012080192, "300–1000 Hz_40ms_envelope_correlation": 0.9971120801218767, "300–1000 Hz_40ms_modulation_mae_db": 1.2370686806663727, "1000–3000 Hz_10ms_log_rms_mae_db": 2.656263549113277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1891393484068926, "1000–3000 Hz_10ms_envelope_correlation": 0.9753381696059629, "1000–3000 Hz_10ms_modulation_mae_db": 1.1484282520071318, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6704195577563128, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12378409080172287, "1000–3000 Hz_40ms_envelope_correlation": 0.9888049243257182, "1000–3000 Hz_40ms_modulation_mae_db": 1.173029346092743, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6876678964354648, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.201872065362092, "3000–6000 Hz_10ms_envelope_correlation": 0.9748305057454381, "3000–6000 Hz_10ms_modulation_mae_db": 1.1768200279347167, "3000–6000 Hz_40ms_log_rms_mae_db": 1.652360828079835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10817490557568109, "3000–6000 Hz_40ms_envelope_correlation": 0.9938619365481038, "3000–6000 Hz_40ms_modulation_mae_db": 1.3183149049978828, "mr_stft_db": 6.435077657331687, "band_mae_db": 0.16504706554315485, "rms_gain_error_db": -0.15451097679543702, "output_peak": 0.723118230944389}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6473059981716163, "全帯域_10ms_rms_envelope_nrmse": 0.11821514768243925, "全帯域_10ms_envelope_correlation": 0.9867854549478076, "全帯域_10ms_modulation_mae_db": 0.40236049148795566, "全帯域_40ms_log_rms_mae_db": 0.9800825220444175, "全帯域_40ms_rms_envelope_nrmse": 0.06930715284511837, "全帯域_40ms_envelope_correlation": 0.9949482285475717, "全帯域_40ms_modulation_mae_db": 0.47292466704962116, "300–1000 Hz_10ms_log_rms_mae_db": 2.422579782406306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12430850182861042, "300–1000 Hz_10ms_envelope_correlation": 0.990116549737832, "300–1000 Hz_10ms_modulation_mae_db": 0.5616520685544952, "300–1000 Hz_40ms_log_rms_mae_db": 1.5496781923911949, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07550167547854365, "300–1000 Hz_40ms_envelope_correlation": 0.9963390291443366, "300–1000 Hz_40ms_modulation_mae_db": 1.1997748031758457, "1000–3000 Hz_10ms_log_rms_mae_db": 2.7310146241218582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1906905714257453, "1000–3000 Hz_10ms_envelope_correlation": 0.9748991764338538, "1000–3000 Hz_10ms_modulation_mae_db": 0.9230103363959671, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7072875731413013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12335201510124072, "1000–3000 Hz_40ms_envelope_correlation": 0.9887854513895162, "1000–3000 Hz_40ms_modulation_mae_db": 1.1937551250703493, "3000–6000 Hz_10ms_log_rms_mae_db": 2.760198084101029, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22898290449029857, "3000–6000 Hz_10ms_envelope_correlation": 0.9668515849822966, "3000–6000 Hz_10ms_modulation_mae_db": 1.145644889527113, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7595994252918326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12588495198224064, "3000–6000 Hz_40ms_envelope_correlation": 0.9893988561976368, "3000–6000 Hz_40ms_modulation_mae_db": 1.270964225177101, "mr_stft_db": 6.439234818326226, "band_mae_db": 0.10970205702990721, "rms_gain_error_db": -0.021904396959697012, "output_peak": 0.709609830896334}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.661879159394768, "全帯域_10ms_rms_envelope_nrmse": 0.10494576067060442, "全帯域_10ms_envelope_correlation": 0.9897900365851786, "全帯域_10ms_modulation_mae_db": 0.14773332618122334, "全帯域_40ms_log_rms_mae_db": 0.8431641300133487, "全帯域_40ms_rms_envelope_nrmse": 0.05551585407303539, "全帯域_40ms_envelope_correlation": 0.9969017443136279, "全帯域_40ms_modulation_mae_db": 0.2517407879588416, "300–1000 Hz_10ms_log_rms_mae_db": 2.3618672968069547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11057771487062969, "300–1000 Hz_10ms_envelope_correlation": 0.9922461598142938, "300–1000 Hz_10ms_modulation_mae_db": 0.5834922347930722, "300–1000 Hz_40ms_log_rms_mae_db": 1.5174470509388078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06988977164633997, "300–1000 Hz_40ms_envelope_correlation": 0.9967708038971077, "300–1000 Hz_40ms_modulation_mae_db": 1.074045805924497, "1000–3000 Hz_10ms_log_rms_mae_db": 2.619671112579692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.182621471123265, "1000–3000 Hz_10ms_envelope_correlation": 0.9770740860941057, "1000–3000 Hz_10ms_modulation_mae_db": 0.6225753931601331, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6079539561286955, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11514335955787881, "1000–3000 Hz_40ms_envelope_correlation": 0.9901518142396276, "1000–3000 Hz_40ms_modulation_mae_db": 0.7879573509604705, "3000–6000 Hz_10ms_log_rms_mae_db": 2.697160138551512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21065629314563805, "3000–6000 Hz_10ms_envelope_correlation": 0.9716484439257063, "3000–6000 Hz_10ms_modulation_mae_db": 0.9815598663294267, "3000–6000 Hz_40ms_log_rms_mae_db": 1.700341603849235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10096187080394571, "3000–6000 Hz_40ms_envelope_correlation": 0.9934444812193814, "3000–6000 Hz_40ms_modulation_mae_db": 1.0602680175564938, "mr_stft_db": 6.435069313439892, "band_mae_db": 0.1131194045510449, "rms_gain_error_db": 0.04619960069772357, "output_peak": 0.7187022599494866}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.844445755926524, "spectral_convergence": 0.4175098400426137, "spectral_convergence_status": "ok", "amplitude_floor": 1.963677544918102e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.357384735512666, "spectral_convergence": 0.3884690142861977, "spectral_convergence_status": "ok", "amplitude_floor": 1.0999213683671372e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.081996796058077, "spectral_convergence": 0.39622531044500925, "spectral_convergence_status": "ok", "amplitude_floor": 9.527550859021228e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.000972009721466654, "output_power": 0.0009591933602189413, "delta_db": -0.05764444700643612, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00078771634185205, "output_power": 0.0007647085371686403, "delta_db": -0.12873916716490383, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008277001014055567, "output_power": 0.0007915264405368347, "delta_db": -0.19407581580140004, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015338014805248517, "output_power": 0.0001520404579009222, "delta_db": -0.0380998374341327, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000153079755450355, "output_power": 0.00014190774724137553, "delta_db": -0.329116539912773, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000364753212040036, "output_power": 0.0003733342340559741, "delta_db": 0.10098690742587173, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14144378579091957, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10239113560445308, "pitch": null, "output_rms_dbfs": -24.972028252830732, "output_peak": 0.7307252323628555, "reference_rms_dbfs": -24.869637117226276, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.427942429165756}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002718271425028404, "log_rms_mae_db": 1.7249129667924288, "rms_envelope_nrmse": 0.10931027869072489, "envelope_correlation": 0.988763680707961, "rms_gain_error_db": -0.09826795819283468, "reference_mean_power": 0.0032374876277477364, "output_mean_power": 0.00316505543961219}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2735378968683332e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004666640237447372, "output_power": 0.00043743889050534184, "reference_active": true, "delta_db": -0.28086929996683474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002202584731702016, "output_power": 0.0002050593141885398, "reference_active": true, "delta_db": -0.31053123810156263}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.478672520125922e-05, "output_power": 2.753695717175315e-05, "reference_active": true, "delta_db": -1.0149759736466255}], "active_band_count": 3, "active_band_mae_db": 0.535458837238341}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002399603663054526, "log_rms_mae_db": 0.9337873944320029, "rms_envelope_nrmse": 0.06007391997163566, "envelope_correlation": 0.996321558348865, "rms_gain_error_db": -0.10379388284245124, "reference_mean_power": 0.0031195469748277875, "output_mean_power": 0.0030458754567613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1403707959359127e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046404851798205753, "output_power": 0.00042769160506044414, "reference_active": true, "delta_db": -0.3543266422033358}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.861370772885771e-05, "output_power": 9.358310981752388e-05, "reference_active": true, "delta_db": -0.2273981499809793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9585409286275983e-06, "output_power": 1.3789072579994589e-06, "reference_active": true, "delta_db": -1.5239759410065843}], "active_band_count": 3, "active_band_mae_db": 0.7019002443969665}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022360065406936124, "log_rms_mae_db": 2.357149986791226, "rms_envelope_nrmse": 0.12095214835797458, "envelope_correlation": 0.9907998172869538, "rms_gain_error_db": -0.19445469410178387, "reference_mean_power": 0.0012597240435666534, "output_mean_power": 0.0012045641929739215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2605416603296333e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001978275618555469, "output_power": 0.0001781459684630409, "reference_active": true, "delta_db": -0.45510800038815297}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.107555923758901e-05, "output_power": 6.43074671036655e-05, "reference_active": true, "delta_db": -1.0062854876242486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1829892137914497e-05, "output_power": 8.902474756175369e-06, "reference_active": true, "delta_db": -1.2347003399009209}], "active_band_count": 3, "active_band_mae_db": 0.898697942637774}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019906826982741464, "log_rms_mae_db": 1.469969101729349, "rms_envelope_nrmse": 0.06845765387021369, "envelope_correlation": 0.9972339940389426, "rms_gain_error_db": -0.2069490607498039, "reference_mean_power": 0.0011746644654148715, "output_mean_power": 0.0011200023272845848}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9316122263011187e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214336491155288, "output_power": 0.0001857470090593582, "reference_active": true, "delta_db": -0.5770762375107095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6617395760512454e-05, "output_power": 2.890471643812799e-05, "reference_active": true, "delta_db": -1.0271874071430656}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0023494047314195e-06, "output_power": 5.70873454200226e-07, "reference_active": true, "delta_db": -2.444792881537378}], "active_band_count": 3, "active_band_mae_db": 1.3496855087303843}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.946276793614836e-05, "log_rms_mae_db": 2.7056951594608396, "rms_envelope_nrmse": 0.2008616668315551, "envelope_correlation": 0.9720568242433284, "rms_gain_error_db": -0.11657993522228186, "reference_mean_power": 0.00019501605476176557, "output_mean_power": 0.00018985077430669258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.789505238315189e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.753576467028983e-05, "output_power": 3.588875754425386e-05, "reference_active": true, "delta_db": -0.194868441672591}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.624079451059313e-05, "output_power": 1.0696399245969046e-05, "reference_active": true, "delta_db": -1.813696666886647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.016250052718367e-06, "output_power": 2.8398832062328802e-06, "reference_active": true, "delta_db": -1.505202647524932}], "active_band_count": 3, "active_band_mae_db": 1.1712559186947233}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.236492865510781e-05, "log_rms_mae_db": 1.684251198395865, "rms_envelope_nrmse": 0.12498961399812084, "envelope_correlation": 0.9885758812849599, "rms_gain_error_db": -0.1524303010880015, "reference_mean_power": 0.00017673870709398627, "output_mean_power": 0.00017064306518266357}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.939463723397018e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.781437846450251e-05, "output_power": 3.4886002684631534e-05, "reference_active": true, "delta_db": -0.3500575663939399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.77351369725072e-06, "output_power": 5.175022443726635e-06, "reference_active": true, "delta_db": -1.7670513050495986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.249207604003898e-07, "output_power": 1.8293441964500265e-07, "reference_active": true, "delta_db": -2.4948203398817306}], "active_band_count": 3, "active_band_mae_db": 1.537309737108423}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014584429481341953, "log_rms_mae_db": 2.666166491370586, "rms_envelope_nrmse": 0.20506532507090783, "envelope_correlation": 0.9732017818448547, "rms_gain_error_db": -0.13894711783778596, "reference_mean_power": 0.00024652309966206675, "output_mean_power": 0.00023876073542098764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9096739107120075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.093194518056527e-05, "output_power": 7.440859624179813e-05, "reference_active": true, "delta_db": 0.20781241661668876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.832881668922571e-05, "output_power": 2.5149232358615783e-05, "reference_active": true, "delta_db": -0.5170370074442758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4378297018808954e-05, "output_power": 6.388763642543696e-06, "reference_active": true, "delta_db": -3.5229062942520066}], "active_band_count": 3, "active_band_mae_db": 1.4159185727709904}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.656067210642841e-05, "log_rms_mae_db": 1.6399601008391418, "rms_envelope_nrmse": 0.11663410954053445, "envelope_correlation": 0.9909586539305565, "rms_gain_error_db": -0.13924071253541676, "reference_mean_power": 0.0002460055124763468, "output_mean_power": 0.00023824333926286418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8079454178991016e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.471033966073099e-05, "output_power": 7.465116135921734e-05, "reference_active": true, "delta_db": -0.003441423789279836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5789672835999803e-05, "output_power": 1.2969463767746014e-05, "reference_active": true, "delta_db": -0.8545111124280697}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1060211769697704e-06, "output_power": 4.450168474491616e-07, "reference_active": true, "delta_db": -3.9538698966380577}], "active_band_count": 3, "active_band_mae_db": 1.6039408109518023}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.715011514048197, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a5ecefe4eb27a75be2e314613451421076048db7bfb997bcca5152e297d9ec46", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9745947765576071, "standard_deviation": 0.002053465858509516, "minimum": 0.9715557392420073, "maximum": 0.9770740860941057, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.6835145870905635, "standard_deviation": 0.05817971094630848, "minimum": 2.619671112579692, "maximum": 2.7906114417010364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9911799679962576, "standard_deviation": 0.24784312327665087, "minimum": 0.6225753931601331, "maximum": 1.268311385578447, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1918811001720731, "standard_deviation": 0.00744628600072768, "minimum": 0.182621471123265, "maximum": 0.20290407939794255, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9891821560127725, "standard_deviation": 0.0009101960378915876, "minimum": 0.9876682696938531, "maximum": 0.9904528429176449, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.6740871986335022, "standard_deviation": 0.05961824659900995, "minimum": 1.5901669134898058, "maximum": 1.784598496003305, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1771591423935073, "standard_deviation": 0.33141886156515493, "minimum": 0.672714603097444, "maximum": 1.617578636450143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12154564220930641, "standard_deviation": 0.0048645774751378806, "minimum": 0.11514335955787881, "maximum": 0.12938217050220732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9710420283626285, "standard_deviation": 0.0032559934826200303, "minimum": 0.9668515849822966, "maximum": 0.9750991483519064, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.7175327214666583, "standard_deviation": 0.036452201268063214, "minimum": 2.666166491370586, "maximum": 2.760198084101029, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3831131523808247, "standard_deviation": 0.2607419954781511, "minimum": 0.9815598663294267, "maximum": 1.6638200807100247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21357448922587619, "standard_deviation": 0.011460583637356404, "minimum": 0.19893995859760288, "maximum": 0.22898290449029857, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991797170253629, "standard_deviation": 0.00163038765939964, "minimum": 0.9893988561976368, "maximum": 0.9938619365481038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7006742862225672, "standard_deviation": 0.042976023589755157, "minimum": 1.6399601008391418, "maximum": 1.7595994252918326, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5707072757609222, "standard_deviation": 0.3178499877335584, "minimum": 1.0602680175564938, "maximum": 1.9435237096426352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11410396659657679, "standard_deviation": 0.008938772717785332, "minimum": 0.10096187080394571, "maximum": 0.12588495198224064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913264482588687, "standard_deviation": 0.0011664651443578154, "minimum": 0.9900280372085432, "maximum": 0.9929443274164057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3693455140852113, "standard_deviation": 0.037528905971304684, "minimum": 2.323478302966423, "maximum": 2.4263303119338877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6745646951323121, "standard_deviation": 0.16222457435587684, "minimum": 0.4873596218997682, "maximum": 0.898697942637774, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11640624280582829, "standard_deviation": 0.007706155437759092, "minimum": 0.10523048143395462, "maximum": 0.12472567278442034, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996766621465915, "standard_deviation": 0.0007279105516462167, "minimum": 0.9956949410992719, "maximum": 0.9979128887061984, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5138989465911563, "standard_deviation": 0.05338829181402365, "minimum": 1.469969101729349, "maximum": 1.618935727129187, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.062154784722026, "standard_deviation": 0.2103860907041264, "minimum": 0.66149905978119, "maximum": 1.3496855087303843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07021031734191119, "standard_deviation": 0.007452842611796875, "minimum": 0.05633912265550192, "maximum": 0.07989450473021535, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13525267690570206, "standard_deviation": 0.029126083772243256, "minimum": 0.10970205702990721, "maximum": 0.19114832552783337, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.430210783986581, "standard_deviation": 0.008791746550347623, "minimum": 6.414946908155241, "maximum": 6.439930544972866, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7264817665824135, "standard_deviation": 0.022752560319997398, "minimum": 0.709609830896334, "maximum": 0.7801546462727471, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05867246951321174, "standard_deviation": 0.0826960985688365, "minimum": -0.15451097679543702, "maximum": 0.06844451095937411, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9880451328416344, "standard_deviation": 0.0015160866408839501, "minimum": 0.9858143342697647, "maximum": 0.9900720501147757, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6933783859719327, "standard_deviation": 0.051762899983789426, "minimum": 1.6473059981716163, "maximum": 1.8059017281425926, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4038281905122235, "standard_deviation": 0.29843681979354086, "minimum": 0.14771948121409398, "maximum": 1.069948768838252, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11283225523892051, "standard_deviation": 0.00725492431885431, "minimum": 0.1027870541536675, "maximum": 0.12416097356392457, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959437093981629, "standard_deviation": 0.001130206670703335, "minimum": 0.9937329316046821, "maximum": 0.9971465948270727, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9031135134510067, "standard_deviation": 0.06989837590461803, "minimum": 0.7844232146716533, "maximum": 0.9941056036151297, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5725296726073705, "standard_deviation": 0.4333288586935934, "minimum": 0.13983206354648478, "maximum": 1.5392341834939405, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06269220526119525, "standard_deviation": 0.008190794733620669, "minimum": 0.05332968373042016, "maximum": 0.07855727636358585, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13751-0001", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.6853766441345215, "predicted_mos": 2.6853766441345215, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "56af7aeec60a1c10f69d44f79bae1fbeae8d7b9549892734e4ee88b2a055fd24", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8819673997350037, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7386464861432795, "全帯域_10ms_rms_envelope_nrmse": 0.12246857049038991, "全帯域_10ms_envelope_correlation": 0.9858993167550576, "全帯域_10ms_modulation_mae_db": 0.28648312999483533, "全帯域_40ms_log_rms_mae_db": 0.9092155886950972, "全帯域_40ms_rms_envelope_nrmse": 0.06511626541457638, "全帯域_40ms_envelope_correlation": 0.9957303822229784, "全帯域_40ms_modulation_mae_db": 0.12438175685274633, "300–1000 Hz_10ms_log_rms_mae_db": 1.6164639035199524, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10363290847781843, "300–1000 Hz_10ms_envelope_correlation": 0.9931325361904082, "300–1000 Hz_10ms_modulation_mae_db": 0.386196643817767, "300–1000 Hz_40ms_log_rms_mae_db": 1.009553821543034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05979175275012858, "300–1000 Hz_40ms_envelope_correlation": 0.9977093361928817, "300–1000 Hz_40ms_modulation_mae_db": 0.2976793279139135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6685355518066711, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17494152193062593, "1000–3000 Hz_10ms_envelope_correlation": 0.9788722851966937, "1000–3000 Hz_10ms_modulation_mae_db": 0.9525819247721478, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0738347348974187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11215950029502679, "1000–3000 Hz_40ms_envelope_correlation": 0.9908794324515019, "1000–3000 Hz_40ms_modulation_mae_db": 0.9282075069595966, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5566010726214272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16693251095140335, "3000–6000 Hz_10ms_envelope_correlation": 0.9826059141412572, "3000–6000 Hz_10ms_modulation_mae_db": 0.7301619775122677, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9588212217761442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09514941868523859, "3000–6000 Hz_40ms_envelope_correlation": 0.9947628678177446, "3000–6000 Hz_40ms_modulation_mae_db": 0.8847623969130979, "mr_stft_db": 5.817109160475707, "band_mae_db": 0.18595644737031128, "rms_gain_error_db": -0.1370406738075978, "output_peak": 0.7280380073085854}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8337585046344715, "全帯域_10ms_rms_envelope_nrmse": 0.13106885854467826, "全帯域_10ms_envelope_correlation": 0.9837476590030494, "全帯域_10ms_modulation_mae_db": 0.24974142239154062, "全帯域_40ms_log_rms_mae_db": 0.9725193494424028, "全帯域_40ms_rms_envelope_nrmse": 0.07984450808347549, "全帯域_40ms_envelope_correlation": 0.993292477424116, "全帯域_40ms_modulation_mae_db": 0.1817142566281963, "300–1000 Hz_10ms_log_rms_mae_db": 1.653743579111012, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1183407340640661, "300–1000 Hz_10ms_envelope_correlation": 0.9909756174667456, "300–1000 Hz_10ms_modulation_mae_db": 0.3034504754424201, "300–1000 Hz_40ms_log_rms_mae_db": 1.0657027369648464, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07324544354635097, "300–1000 Hz_40ms_envelope_correlation": 0.9964197288805108, "300–1000 Hz_40ms_modulation_mae_db": 0.3912527099838747, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6202432481192508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1626895984270923, "1000–3000 Hz_10ms_envelope_correlation": 0.981684544672708, "1000–3000 Hz_10ms_modulation_mae_db": 0.4471569613201936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9714262614845508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09621573462134982, "1000–3000 Hz_40ms_envelope_correlation": 0.9932298225403969, "1000–3000 Hz_40ms_modulation_mae_db": 0.6095411083793876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6060959144745028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1923463870991241, "3000–6000 Hz_10ms_envelope_correlation": 0.9765541343837033, "3000–6000 Hz_10ms_modulation_mae_db": 0.5987759536394712, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0189162723704404, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11612621298802799, "3000–6000 Hz_40ms_envelope_correlation": 0.9909555329442097, "3000–6000 Hz_40ms_modulation_mae_db": 1.0185085055552912, "mr_stft_db": 5.827670213133664, "band_mae_db": 0.11206047871114512, "rms_gain_error_db": -0.04824228189108669, "output_peak": 0.7054475861463159}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7333128950103753, "全帯域_10ms_rms_envelope_nrmse": 0.10789310459957102, "全帯域_10ms_envelope_correlation": 0.9891491861434131, "全帯域_10ms_modulation_mae_db": 0.32599294855040734, "全帯域_40ms_log_rms_mae_db": 0.9035403539671465, "全帯域_40ms_rms_envelope_nrmse": 0.058414653372432686, "全帯域_40ms_envelope_correlation": 0.9966077246158201, "全帯域_40ms_modulation_mae_db": 0.20302365431377575, "300–1000 Hz_10ms_log_rms_mae_db": 1.6779231010646112, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09896937976751174, "300–1000 Hz_10ms_envelope_correlation": 0.9938206757287317, "300–1000 Hz_10ms_modulation_mae_db": 0.6920785723959132, "300–1000 Hz_40ms_log_rms_mae_db": 1.0846959488617225, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0623321560262503, "300–1000 Hz_40ms_envelope_correlation": 0.9975871554510088, "300–1000 Hz_40ms_modulation_mae_db": 0.9224536967704889, "1000–3000 Hz_10ms_log_rms_mae_db": 1.595093410370189, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1634943552868556, "1000–3000 Hz_10ms_envelope_correlation": 0.9816930240381786, "1000–3000 Hz_10ms_modulation_mae_db": 0.6402057810811015, "1000–3000 Hz_40ms_log_rms_mae_db": 0.994708773832635, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10523546557221446, "1000–3000 Hz_40ms_envelope_correlation": 0.9923385081230853, "1000–3000 Hz_40ms_modulation_mae_db": 0.8909700318640117, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5124116992677943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17292594339281306, "3000–6000 Hz_10ms_envelope_correlation": 0.9808974146033734, "3000–6000 Hz_10ms_modulation_mae_db": 0.9920878935159267, "3000–6000 Hz_40ms_log_rms_mae_db": 0.950081530924236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10043478929981285, "3000–6000 Hz_40ms_envelope_correlation": 0.9934565256124502, "3000–6000 Hz_40ms_modulation_mae_db": 1.2158684838788603, "mr_stft_db": 5.8083952553083025, "band_mae_db": 0.18117764933933042, "rms_gain_error_db": -0.13030688916172878, "output_peak": 0.7035766337065137}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7618860808564176, "全帯域_10ms_rms_envelope_nrmse": 0.11426687112259834, "全帯域_10ms_envelope_correlation": 0.9877660849367735, "全帯域_10ms_modulation_mae_db": 0.4344089459867328, "全帯域_40ms_log_rms_mae_db": 0.939418104113448, "全帯域_40ms_rms_envelope_nrmse": 0.06503522714575008, "全帯域_40ms_envelope_correlation": 0.9956225618029031, "全帯域_40ms_modulation_mae_db": 0.48524857650461745, "300–1000 Hz_10ms_log_rms_mae_db": 1.6215366733415055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11067060399603383, "300–1000 Hz_10ms_envelope_correlation": 0.9921887239679927, "300–1000 Hz_10ms_modulation_mae_db": 0.39626847284251737, "300–1000 Hz_40ms_log_rms_mae_db": 1.0571879856359014, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0676653902319398, "300–1000 Hz_40ms_envelope_correlation": 0.9970908408905156, "300–1000 Hz_40ms_modulation_mae_db": 0.48010400707601536, "1000–3000 Hz_10ms_log_rms_mae_db": 1.65334849490277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16730642661345743, "1000–3000 Hz_10ms_envelope_correlation": 0.980585789895602, "1000–3000 Hz_10ms_modulation_mae_db": 0.9324842962206142, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0003122539971827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11187523146361229, "1000–3000 Hz_40ms_envelope_correlation": 0.9906474266018082, "1000–3000 Hz_40ms_modulation_mae_db": 1.3899491332144354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5563169302766535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1832071062763881, "3000–6000 Hz_10ms_envelope_correlation": 0.9784970530707338, "3000–6000 Hz_10ms_modulation_mae_db": 0.45048115269144096, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9141807114915956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08786948749751448, "3000–6000 Hz_40ms_envelope_correlation": 0.9948045674150634, "3000–6000 Hz_40ms_modulation_mae_db": 0.49966784471076436, "mr_stft_db": 5.817916070908349, "band_mae_db": 0.1594665197624542, "rms_gain_error_db": -0.058806152267925016, "output_peak": 0.7074419403283165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7728253950630106, "全帯域_10ms_rms_envelope_nrmse": 0.12177413482615587, "全帯域_10ms_envelope_correlation": 0.9860442334516485, "全帯域_10ms_modulation_mae_db": 0.0515702530853037, "全帯域_40ms_log_rms_mae_db": 1.0180341173756091, "全帯域_40ms_rms_envelope_nrmse": 0.06618957182913915, "全帯域_40ms_envelope_correlation": 0.9954817677853222, "全帯域_40ms_modulation_mae_db": 0.09298759962999541, "300–1000 Hz_10ms_log_rms_mae_db": 1.6642839960833518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10837526360445436, "300–1000 Hz_10ms_envelope_correlation": 0.9926152815079873, "300–1000 Hz_10ms_modulation_mae_db": 0.23611155649021998, "300–1000 Hz_40ms_log_rms_mae_db": 1.0827114383879954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06668209342050777, "300–1000 Hz_40ms_envelope_correlation": 0.997279815371415, "300–1000 Hz_40ms_modulation_mae_db": 0.4232275886424599, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6917534776109988, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1667332886903624, "1000–3000 Hz_10ms_envelope_correlation": 0.9809482860934835, "1000–3000 Hz_10ms_modulation_mae_db": 0.6337837182254605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.06500162035496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10808066710002173, "1000–3000 Hz_40ms_envelope_correlation": 0.991785543947824, "1000–3000 Hz_40ms_modulation_mae_db": 0.5965337674067307, "3000–6000 Hz_10ms_log_rms_mae_db": 1.577328163893038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17020240170720347, "3000–6000 Hz_10ms_envelope_correlation": 0.9821029539565622, "3000–6000 Hz_10ms_modulation_mae_db": 1.1375704429561269, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0073423888652915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09503627508989472, "3000–6000 Hz_40ms_envelope_correlation": 0.9950883251143569, "3000–6000 Hz_40ms_modulation_mae_db": 1.6218106227400841, "mr_stft_db": 5.810300308513021, "band_mae_db": 0.17727080540083862, "rms_gain_error_db": -0.10423352754146198, "output_peak": 0.7100126653910689}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.708788917936674, "全帯域_10ms_rms_envelope_nrmse": 0.11707427723541322, "全帯域_10ms_envelope_correlation": 0.9871419938699977, "全帯域_10ms_modulation_mae_db": 0.09495943332756747, "全帯域_40ms_log_rms_mae_db": 0.971491063303461, "全帯域_40ms_rms_envelope_nrmse": 0.06619981961526916, "全帯域_40ms_envelope_correlation": 0.9955445005902738, "全帯域_40ms_modulation_mae_db": 0.08750320684817069, "300–1000 Hz_10ms_log_rms_mae_db": 1.677493167171123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10377449069059172, "300–1000 Hz_10ms_envelope_correlation": 0.9930631359570987, "300–1000 Hz_10ms_modulation_mae_db": 0.577265595470576, "300–1000 Hz_40ms_log_rms_mae_db": 0.9848043502395726, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06167771875426759, "300–1000 Hz_40ms_envelope_correlation": 0.9974838717638672, "300–1000 Hz_40ms_modulation_mae_db": 0.8188073638165503, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6662751334426136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17273615815284746, "1000–3000 Hz_10ms_envelope_correlation": 0.9793544199678461, "1000–3000 Hz_10ms_modulation_mae_db": 0.9452234995000529, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0344285788737073, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11620892228507661, "1000–3000 Hz_40ms_envelope_correlation": 0.9902019733340689, "1000–3000 Hz_40ms_modulation_mae_db": 1.056018648028903, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5815311787425812, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18612996415493147, "3000–6000 Hz_10ms_envelope_correlation": 0.9782312720097031, "3000–6000 Hz_10ms_modulation_mae_db": 0.9401876659852363, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9971893826066636, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10775665991165378, "3000–6000 Hz_40ms_envelope_correlation": 0.9931737237510174, "3000–6000 Hz_40ms_modulation_mae_db": 1.3075538810447014, "mr_stft_db": 5.83086984193474, "band_mae_db": 0.15907632628325213, "rms_gain_error_db": -0.11565482387650294, "output_peak": 0.7306096980924199}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.693314433064194, "全帯域_10ms_rms_envelope_nrmse": 0.11519694434164392, "全帯域_10ms_envelope_correlation": 0.987604420284399, "全帯域_10ms_modulation_mae_db": 0.2623702088532908, "全帯域_40ms_log_rms_mae_db": 0.8724784517364809, "全帯域_40ms_rms_envelope_nrmse": 0.05895675027741403, "全帯域_40ms_envelope_correlation": 0.9964594960464758, "全帯域_40ms_modulation_mae_db": 0.49164139690562375, "300–1000 Hz_10ms_log_rms_mae_db": 1.6095216027769654, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10417704224863109, "300–1000 Hz_10ms_envelope_correlation": 0.9930369879857092, "300–1000 Hz_10ms_modulation_mae_db": 0.45162783336388673, "300–1000 Hz_40ms_log_rms_mae_db": 1.0287144975131362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06509692246783941, "300–1000 Hz_40ms_envelope_correlation": 0.9971867792472656, "300–1000 Hz_40ms_modulation_mae_db": 0.74410144246588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6637986038785582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16548511852290954, "1000–3000 Hz_10ms_envelope_correlation": 0.9811760373534568, "1000–3000 Hz_10ms_modulation_mae_db": 0.7226699735646442, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0433179924106177, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1086158396108431, "1000–3000 Hz_40ms_envelope_correlation": 0.9916975032550522, "1000–3000 Hz_40ms_modulation_mae_db": 0.7477265752451553, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6002825265823535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17960592040890322, "3000–6000 Hz_10ms_envelope_correlation": 0.9800416004806669, "3000–6000 Hz_10ms_modulation_mae_db": 1.2181762632023896, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9901216583399594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10508572562491536, "3000–6000 Hz_40ms_envelope_correlation": 0.9938578033541198, "3000–6000 Hz_40ms_modulation_mae_db": 1.5272657078877232, "mr_stft_db": 5.804399827711343, "band_mae_db": 0.15151855942131523, "rms_gain_error_db": -0.054955379549257805, "output_peak": 0.6885904887950047}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7390241764114693, "全帯域_10ms_rms_envelope_nrmse": 0.1183931920939102, "全帯域_10ms_envelope_correlation": 0.9867899075327212, "全帯域_10ms_modulation_mae_db": 0.3796954289175593, "全帯域_40ms_log_rms_mae_db": 0.8505391061516492, "全帯域_40ms_rms_envelope_nrmse": 0.06169834301155372, "全帯域_40ms_envelope_correlation": 0.9960293209120101, "全帯域_40ms_modulation_mae_db": 0.31323357973229743, "300–1000 Hz_10ms_log_rms_mae_db": 1.7325808699347625, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1055379569250252, "300–1000 Hz_10ms_envelope_correlation": 0.9929137542986043, "300–1000 Hz_10ms_modulation_mae_db": 0.5229583470949718, "300–1000 Hz_40ms_log_rms_mae_db": 1.1369818857138174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07360439929956995, "300–1000 Hz_40ms_envelope_correlation": 0.9965157845999628, "300–1000 Hz_40ms_modulation_mae_db": 0.6676698692787241, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6765434107100743, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16902862760881013, "1000–3000 Hz_10ms_envelope_correlation": 0.9802116229181188, "1000–3000 Hz_10ms_modulation_mae_db": 0.7562395251901726, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0675865419217347, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11237635450505457, "1000–3000 Hz_40ms_envelope_correlation": 0.9908798135334682, "1000–3000 Hz_40ms_modulation_mae_db": 0.8754813061895964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5889076483309907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17254447855133326, "3000–6000 Hz_10ms_envelope_correlation": 0.9809156744483672, "3000–6000 Hz_10ms_modulation_mae_db": 0.7995445307110169, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0014647910720194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08379947343429497, "3000–6000 Hz_40ms_envelope_correlation": 0.9953439539581119, "3000–6000 Hz_40ms_modulation_mae_db": 1.0086191627253396, "mr_stft_db": 5.8110394204735485, "band_mae_db": 0.15007758887624573, "rms_gain_error_db": -0.0613523832282932, "output_peak": 0.6941832403687058}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.044648431842907, "spectral_convergence": 0.4169598973135819, "spectral_convergence_status": "ok", "amplitude_floor": 1.963677544918102e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.592208229858784, "spectral_convergence": 0.36868556490260734, "spectral_convergence_status": "ok", "amplitude_floor": 1.0999213683671372e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.814470819725431, "spectral_convergence": 0.39105203819047224, "spectral_convergence_status": "ok", "amplitude_floor": 9.527550859021228e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.000972009721466654, "output_power": 0.0009630374032478798, "delta_db": -0.040274535621755314, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00078771634185205, "output_power": 0.0007666948879764261, "delta_db": -0.11747287815750676, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008277001014055567, "output_power": 0.0008023155042613445, "delta_db": -0.13527823644878245, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015338014805248517, "output_power": 0.00014741320835220584, "delta_db": -0.17232754211752657, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000153079755450355, "output_power": 0.00014389746388289513, "delta_db": -0.2686461996187711, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000364753212040036, "output_power": 0.0003340605530592979, "delta_db": -0.38173929225752556, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18595644737031128, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1370406738075978, "pitch": null, "output_rms_dbfs": -25.006677791033876, "output_peak": 0.7280380073085854, "reference_rms_dbfs": -24.869637117226276, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.817109160475707}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002718271425028404, "log_rms_mae_db": 1.7386464861432795, "rms_envelope_nrmse": 0.12246857049038991, "envelope_correlation": 0.9858993167550576, "rms_gain_error_db": -0.13356794703597688, "reference_mean_power": 0.0032374876277477364, "output_mean_power": 0.0031394337494836007}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2735378968683332e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004666640237447372, "output_power": 0.00044976530187016225, "reference_active": true, "delta_db": -0.16018373574034844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002202584731702016, "output_power": 0.00020893005302752738, "reference_active": true, "delta_db": -0.22931709881610435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.478672520125922e-05, "output_power": 3.121897866579132e-05, "reference_active": true, "delta_db": -0.4699485554280531}], "active_band_count": 3, "active_band_mae_db": 0.28648312999483533}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002399603663054526, "log_rms_mae_db": 0.9092155886950972, "rms_envelope_nrmse": 0.06511626541457638, "envelope_correlation": 0.9957303822229784, "rms_gain_error_db": -0.13186204261678314, "reference_mean_power": 0.0031195469748277875, "output_mean_power": 0.0030262536444147204}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1403707959359127e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046404851798205753, "output_power": 0.00043890908115728296, "reference_active": true, "delta_db": -0.2418882339895328}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.861370772885771e-05, "output_power": 9.637683377691429e-05, "reference_active": true, "delta_db": -0.09964633409495802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9585409286275983e-06, "output_power": 1.9728484360959086e-06, "reference_active": true, "delta_db": 0.0316107024737482}], "active_band_count": 3, "active_band_mae_db": 0.12438175685274633}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022360065406936124, "log_rms_mae_db": 1.6164639035199524, "rms_envelope_nrmse": 0.10363290847781843, "envelope_correlation": 0.9931325361904082, "rms_gain_error_db": -0.12743607709209365, "reference_mean_power": 0.0012597240435666534, "output_mean_power": 0.0012232967187728256}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2605416603296333e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001978275618555469, "output_power": 0.00018755056235636612, "reference_active": true, "delta_db": -0.2316842783623223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.107555923758901e-05, "output_power": 7.06501753305121e-05, "reference_active": true, "delta_db": -0.5977670904081053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1829892137914497e-05, "output_power": 1.276129185147838e-05, "reference_active": true, "delta_db": 0.32913856268287345}], "active_band_count": 3, "active_band_mae_db": 0.386196643817767}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019906826982741464, "log_rms_mae_db": 1.009553821543034, "rms_envelope_nrmse": 0.05979175275012858, "envelope_correlation": 0.9977093361928817, "rms_gain_error_db": -0.13159980646072963, "reference_mean_power": 0.0011746644654148715, "output_mean_power": 0.0011396036997918411}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9316122263011187e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214336491155288, "output_power": 0.0001988148739117222, "reference_active": true, "delta_db": -0.28180581037107577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6617395760512454e-05, "output_power": 3.259365720264717e-05, "reference_active": true, "delta_db": -0.5055436028903678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0023494047314195e-06, "output_power": 1.027041502871443e-06, "reference_active": true, "delta_db": 0.10568857048029685}], "active_band_count": 3, "active_band_mae_db": 0.2976793279139135}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.946276793614836e-05, "log_rms_mae_db": 1.6685355518066711, "rms_envelope_nrmse": 0.17494152193062593, "envelope_correlation": 0.9788722851966937, "rms_gain_error_db": -0.2274898916316008, "reference_mean_power": 0.00019501605476176557, "output_mean_power": 0.00018506375807704261}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.789505238315189e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.753576467028983e-05, "output_power": 3.382584423119625e-05, "reference_active": true, "delta_db": -0.4519662304282054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.624079451059313e-05, "output_power": 1.2257921681410658e-05, "reference_active": true, "delta_db": -1.2219042918220648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.016250052718367e-06, "output_power": 3.0579699899447395e-06, "reference_active": true, "delta_db": -1.183875252066173}], "active_band_count": 3, "active_band_mae_db": 0.9525819247721478}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.236492865510781e-05, "log_rms_mae_db": 1.0738347348974187, "rms_envelope_nrmse": 0.11215950029502679, "envelope_correlation": 0.9908794324515019, "rms_gain_error_db": -0.2136411791881769, "reference_mean_power": 0.00017673870709398627, "output_mean_power": 0.00016825483604459765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.939463723397018e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.781437846450251e-05, "output_power": 3.321763127826077e-05, "reference_active": true, "delta_db": -0.562883064755031}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.77351369725072e-06, "output_power": 5.940750405759851e-06, "reference_active": true, "delta_db": -1.1677606179983357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.249207604003898e-07, "output_power": 2.549057177177782e-07, "reference_active": true, "delta_db": -1.0539788381254231}], "active_band_count": 3, "active_band_mae_db": 0.9282075069595966}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014584429481341953, "log_rms_mae_db": 1.5566010726214272, "rms_envelope_nrmse": 0.16693251095140335, "envelope_correlation": 0.9826059141412572, "rms_gain_error_db": -0.35158255268660366, "reference_mean_power": 0.00024652309966206675, "output_mean_power": 0.00022735230646391862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9096739107120075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.093194518056527e-05, "output_power": 6.826147753546235e-05, "reference_active": true, "delta_db": -0.16666185305283368}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.832881668922571e-05, "output_power": 2.6178163603404826e-05, "reference_active": true, "delta_db": -0.3428925662034801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4378297018808954e-05, "output_power": 9.76369713715401e-06, "reference_active": true, "delta_db": -1.6809315132804894}], "active_band_count": 3, "active_band_mae_db": 0.7301619775122677}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.656067210642841e-05, "log_rms_mae_db": 0.9588212217761442, "rms_envelope_nrmse": 0.09514941868523859, "envelope_correlation": 0.9947628678177446, "rms_gain_error_db": -0.354773069920479, "reference_mean_power": 0.0002460055124763468, "output_mean_power": 0.00022670835821781367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8079454178991016e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.471033966073099e-05, "output_power": 7.051227491081527e-05, "reference_active": true, "delta_db": -0.25115984494930216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5789672835999803e-05, "output_power": 1.3941260222975674e-05, "reference_active": true, "delta_db": -0.5407109779334756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1060211769697704e-06, "output_power": 7.203139169043832e-07, "reference_active": true, "delta_db": -1.862416367856516}], "active_band_count": 3, "active_band_mae_db": 0.8847623969130979}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7410176265308404, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bda67f9fbb8ae735ac45fa10192d16914b867a1a8463d17baebcd5d13945f35a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.980565751267011, "standard_deviation": 0.0010352490292401151, "minimum": 0.9788722851966937, "maximum": 0.9816930240381786, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6544489163551406, "standard_deviation": 0.0316223724929686, "minimum": 1.595093410370189, "maximum": 1.6917534776109988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7537932099842984, "standard_deviation": 0.1814220905538767, "minimum": 0.4471569613201936, "maximum": 0.9525819247721478, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16780188690412012, "standard_deviation": 0.004280711467978873, "minimum": 0.1626895984270923, "maximum": 0.17494152193062593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9914575029734007, "standard_deviation": 0.0009987825988080998, "minimum": 0.9902019733340689, "maximum": 0.9932298225403969, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.031327094721601, "standard_deviation": 0.03835911715974122, "minimum": 0.9714262614845508, "maximum": 1.0738347348974187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8868035096609771, "standard_deviation": 0.25749410003342854, "minimum": 0.5965337674067307, "maximum": 1.3899491332144354, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10884596443164993, "standard_deviation": 0.006094843553652281, "minimum": 0.09621573462134982, "maximum": 0.11620892228507661, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9799807521367959, "standard_deviation": 0.002074230366744676, "minimum": 0.9765541343837033, "maximum": 0.9826059141412572, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5724343917736676, "standard_deviation": 0.03023686401756769, "minimum": 1.5124116992677943, "maximum": 1.6060959144745028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8583732350267347, "standard_deviation": 0.26339762265666256, "minimum": 0.45048115269144096, "maximum": 1.2181762632023896, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1779868390677625, "standard_deviation": 0.008786412465660899, "minimum": 0.16693251095140335, "maximum": 0.1923463870991241, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9939304124958842, "standard_deviation": 0.001436678552993065, "minimum": 0.9909555329442097, "maximum": 0.9953439539581119, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9797647446807937, "standard_deviation": 0.035448421282415515, "minimum": 0.9141807114915956, "maximum": 1.0189162723704404, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1355070756819827, "standard_deviation": 0.36311699303020206, "minimum": 0.49966784471076436, "maximum": 1.6218106227400841, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09890725531641908, "standard_deviation": 0.010653601244199616, "minimum": 0.08379947343429497, "maximum": 0.11612621298802799, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927183391379097, "standard_deviation": 0.0008429121377662819, "minimum": 0.9909756174667456, "maximum": 0.9938206757287317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6566933616254107, "standard_deviation": 0.04103681879302403, "minimum": 1.6095216027769654, "maximum": 1.7325808699347625, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.445744687114784, "standard_deviation": 0.14840208433263197, "minimum": 0.23611155649021998, "maximum": 0.6920785723959132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10668479747176655, "standard_deviation": 0.005849552474704361, "minimum": 0.09896937976751174, "maximum": 0.1183407340640661, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971591640496784, "standard_deviation": 0.0004738881734867522, "minimum": 0.9964197288805108, "maximum": 0.9977093361928817, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0562940831075034, "standard_deviation": 0.048048269561689796, "minimum": 0.9848043502395726, "maximum": 1.1369818857138174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5931620007434883, "standard_deviation": 0.22596465414991662, "minimum": 0.2976793279139135, "maximum": 0.9224536967704889, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06626198456210679, "standard_deviation": 0.005134005776486155, "minimum": 0.05979175275012858, "maximum": 0.07360439929956995, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1595755468956116, "standard_deviation": 0.02356993841988566, "minimum": 0.11206047871114512, "maximum": 0.18595644737031128, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.815962512307334, "standard_deviation": 0.009347928472253516, "minimum": 5.804399827711343, "maximum": 5.83086984193474, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7084875325171164, "standard_deviation": 0.014681834701589213, "minimum": 0.6885904887950047, "maximum": 0.7306096980924199, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08882401391548178, "standard_deviation": 0.03674768132945709, "minimum": -0.1370406738075978, "maximum": -0.04824228189108669, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9867678502471325, "standard_deviation": 0.0015994230831569089, "minimum": 0.9837476590030494, "maximum": 0.9891491861434131, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7476946111399865, "standard_deviation": 0.0432181895751638, "minimum": 1.693314433064194, "maximum": 1.8337585046344715, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2606527213884047, "standard_deviation": 0.13125259048655197, "minimum": 0.0515702530853037, "maximum": 0.4344089459867328, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11851699415679509, "standard_deviation": 0.006843383514054665, "minimum": 0.10789310459957102, "maximum": 0.13106885854467826, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9955960289249874, "standard_deviation": 0.0010210401337805044, "minimum": 0.993292477424116, "maximum": 0.9966077246158201, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9296545168481618, "standard_deviation": 0.0562164462929693, "minimum": 0.8505391061516492, "maximum": 1.0180341173756091, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24746675342692787, "standard_deviation": 0.1653513008346446, "minimum": 0.08750320684817069, "maximum": 0.49164139690562375, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06518189234370134, "standard_deviation": 0.0066949100577676154, "minimum": 0.058414653372432686, "maximum": 0.07984450808347549, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13751-0001", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.0906763076782227, "predicted_mos": 2.0906763076782227, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bce5045ded15b2a17e66e33484467ebd8cb8c69e8f9004f634fc240c46d236ab", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7758240001276135, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8067970295157896, "全帯域_10ms_rms_envelope_nrmse": 0.1272943030820075, "全帯域_10ms_envelope_correlation": 0.9848807276137699, "全帯域_10ms_modulation_mae_db": 0.5766048737109334, "全帯域_40ms_log_rms_mae_db": 0.9529140826416242, "全帯域_40ms_rms_envelope_nrmse": 0.07251643017738445, "全帯域_40ms_envelope_correlation": 0.9945203274507387, "全帯域_40ms_modulation_mae_db": 0.46450744381539816, "300–1000 Hz_10ms_log_rms_mae_db": 1.9798483079559228, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1294003833598693, "300–1000 Hz_10ms_envelope_correlation": 0.9892189267026659, "300–1000 Hz_10ms_modulation_mae_db": 0.4937893933241317, "300–1000 Hz_40ms_log_rms_mae_db": 1.2999100786034257, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07251224334620658, "300–1000 Hz_40ms_envelope_correlation": 0.9965522506472522, "300–1000 Hz_40ms_modulation_mae_db": 0.4560847594519621, "1000–3000 Hz_10ms_log_rms_mae_db": 2.119558776953767, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18689428612226813, "1000–3000 Hz_10ms_envelope_correlation": 0.9758545786347638, "1000–3000 Hz_10ms_modulation_mae_db": 0.27841332711102185, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3599609586523338, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.122025430085845, "1000–3000 Hz_40ms_envelope_correlation": 0.9891916299597906, "1000–3000 Hz_40ms_modulation_mae_db": 0.4824725710813982, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0909228359748164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17606628856544138, "3000–6000 Hz_10ms_envelope_correlation": 0.98024807438627, "3000–6000 Hz_10ms_modulation_mae_db": 0.8221830366737546, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3589617231429187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09015322778038574, "3000–6000 Hz_40ms_envelope_correlation": 0.9945727830525799, "3000–6000 Hz_40ms_modulation_mae_db": 0.9995848929633103, "mr_stft_db": 6.044240205111007, "band_mae_db": 0.1333593538044961, "rms_gain_error_db": 0.0004013588204409619, "output_peak": 0.713193763566102}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6322666144656857, "全帯域_10ms_rms_envelope_nrmse": 0.10700871567881565, "全帯域_10ms_envelope_correlation": 0.98925549546383, "全帯域_10ms_modulation_mae_db": 0.38174599345212373, "全帯域_40ms_log_rms_mae_db": 0.8507976617506873, "全帯域_40ms_rms_envelope_nrmse": 0.06048815475789655, "全帯域_40ms_envelope_correlation": 0.9962882836838876, "全帯域_40ms_modulation_mae_db": 0.22099152473294179, "300–1000 Hz_10ms_log_rms_mae_db": 2.0281279118636237, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11658612726822833, "300–1000 Hz_10ms_envelope_correlation": 0.9915681574853086, "300–1000 Hz_10ms_modulation_mae_db": 0.5300111588128317, "300–1000 Hz_40ms_log_rms_mae_db": 1.3793512465950923, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0759918557011752, "300–1000 Hz_40ms_envelope_correlation": 0.9965624535883938, "300–1000 Hz_40ms_modulation_mae_db": 0.7750512220233107, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1402263371482526, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16209688238321482, "1000–3000 Hz_10ms_envelope_correlation": 0.9818796592702268, "1000–3000 Hz_10ms_modulation_mae_db": 1.0064435956408329, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3957563437173302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10384206402009909, "1000–3000 Hz_40ms_envelope_correlation": 0.9920472152949922, "1000–3000 Hz_40ms_modulation_mae_db": 1.1388050178034983, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0566193162154462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17896389368324817, "3000–6000 Hz_10ms_envelope_correlation": 0.9795066611601368, "3000–6000 Hz_10ms_modulation_mae_db": 0.5694556570358427, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3264958782420901, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09110966023250629, "3000–6000 Hz_40ms_envelope_correlation": 0.9945046968675929, "3000–6000 Hz_40ms_modulation_mae_db": 0.7081098583289059, "mr_stft_db": 6.0603876680896605, "band_mae_db": 0.1625976485424154, "rms_gain_error_db": -0.11456818951613626, "output_peak": 0.7251479854685448}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.687062706325655, "全帯域_10ms_rms_envelope_nrmse": 0.11188461324871823, "全帯域_10ms_envelope_correlation": 0.9882433665857852, "全帯域_10ms_modulation_mae_db": 0.5174240630390788, "全帯域_40ms_log_rms_mae_db": 0.968149666673215, "全帯域_40ms_rms_envelope_nrmse": 0.06820782534360283, "全帯域_40ms_envelope_correlation": 0.9951469217390208, "全帯域_40ms_modulation_mae_db": 0.48317027063397777, "300–1000 Hz_10ms_log_rms_mae_db": 1.9231604555705424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11643051274382253, "300–1000 Hz_10ms_envelope_correlation": 0.9912933682023184, "300–1000 Hz_10ms_modulation_mae_db": 0.3756175801515825, "300–1000 Hz_40ms_log_rms_mae_db": 1.2707652354516492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07063332615238656, "300–1000 Hz_40ms_envelope_correlation": 0.9966699471908519, "300–1000 Hz_40ms_modulation_mae_db": 0.7153692976093105, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1310339027002168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18117441675066195, "1000–3000 Hz_10ms_envelope_correlation": 0.9776238719441371, "1000–3000 Hz_10ms_modulation_mae_db": 1.276799156364645, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3683353621387282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11647068229196937, "1000–3000 Hz_40ms_envelope_correlation": 0.9906457005401996, "1000–3000 Hz_40ms_modulation_mae_db": 1.5430115383568273, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0642819249914637, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1784191511873225, "3000–6000 Hz_10ms_envelope_correlation": 0.9796596905190711, "3000–6000 Hz_10ms_modulation_mae_db": 0.9226595099126312, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3670657055090178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09395004082695368, "3000–6000 Hz_40ms_envelope_correlation": 0.9940599417804598, "3000–6000 Hz_40ms_modulation_mae_db": 1.1660138811673855, "mr_stft_db": 6.049160026917278, "band_mae_db": 0.15607238813645027, "rms_gain_error_db": -0.02121043076348772, "output_peak": 0.7007242738840274}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.590703766393316, "全帯域_10ms_rms_envelope_nrmse": 0.11794722387307198, "全帯域_10ms_envelope_correlation": 0.9869345737917338, "全帯域_10ms_modulation_mae_db": 0.1678608488416969, "全帯域_40ms_log_rms_mae_db": 0.8737426576190362, "全帯域_40ms_rms_envelope_nrmse": 0.06950732947160937, "全帯域_40ms_envelope_correlation": 0.9949535827479371, "全帯域_40ms_modulation_mae_db": 0.11540269979711908, "300–1000 Hz_10ms_log_rms_mae_db": 2.0550597876053405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11766847059889989, "300–1000 Hz_10ms_envelope_correlation": 0.9911019195713943, "300–1000 Hz_10ms_modulation_mae_db": 0.7541067107098268, "300–1000 Hz_40ms_log_rms_mae_db": 1.3294481604156088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06392825951571168, "300–1000 Hz_40ms_envelope_correlation": 0.9972491051867444, "300–1000 Hz_40ms_modulation_mae_db": 1.0198898226907842, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0765462796045013, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15661415980222462, "1000–3000 Hz_10ms_envelope_correlation": 0.983124598613751, "1000–3000 Hz_10ms_modulation_mae_db": 1.1212879240495452, "1000–3000 Hz_40ms_log_rms_mae_db": 1.365160472019072, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10630557132540305, "1000–3000 Hz_40ms_envelope_correlation": 0.9917802277040174, "1000–3000 Hz_40ms_modulation_mae_db": 1.2960580859384945, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0755804709240193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17286333475324173, "3000–6000 Hz_10ms_envelope_correlation": 0.9810218362853945, "3000–6000 Hz_10ms_modulation_mae_db": 0.8191220439044463, "3000–6000 Hz_40ms_log_rms_mae_db": 1.330407623795415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09043578517369509, "3000–6000 Hz_40ms_envelope_correlation": 0.9948656769233057, "3000–6000 Hz_40ms_modulation_mae_db": 1.1085357076014979, "mr_stft_db": 6.0325529805061455, "band_mae_db": 0.11697414471549461, "rms_gain_error_db": -0.032166594346967364, "output_peak": 0.7643690584667285}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6119795802219887, "全帯域_10ms_rms_envelope_nrmse": 0.11201566799857827, "全帯域_10ms_envelope_correlation": 0.9881312786299433, "全帯域_10ms_modulation_mae_db": 0.23530738116358893, "全帯域_40ms_log_rms_mae_db": 0.8277934483867173, "全帯域_40ms_rms_envelope_nrmse": 0.062350316632323544, "全帯域_40ms_envelope_correlation": 0.9959908946071203, "全帯域_40ms_modulation_mae_db": 0.21921438912316052, "300–1000 Hz_10ms_log_rms_mae_db": 1.9016969297095998, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11155596075765578, "300–1000 Hz_10ms_envelope_correlation": 0.9922475764147121, "300–1000 Hz_10ms_modulation_mae_db": 0.27935127238775753, "300–1000 Hz_40ms_log_rms_mae_db": 1.2600050593713312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07059067199835829, "300–1000 Hz_40ms_envelope_correlation": 0.9969782346557962, "300–1000 Hz_40ms_modulation_mae_db": 0.596652465510577, "1000–3000 Hz_10ms_log_rms_mae_db": 2.103662293312995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1628503825604437, "1000–3000 Hz_10ms_envelope_correlation": 0.9818160505782261, "1000–3000 Hz_10ms_modulation_mae_db": 1.4929767771766052, "1000–3000 Hz_40ms_log_rms_mae_db": 1.386196330393485, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10419332272469979, "1000–3000 Hz_40ms_envelope_correlation": 0.9924320432777893, "1000–3000 Hz_40ms_modulation_mae_db": 1.9026755701542737, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1131595392586697, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2044489822645467, "3000–6000 Hz_10ms_envelope_correlation": 0.9732953248997429, "3000–6000 Hz_10ms_modulation_mae_db": 1.2516350243148098, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3661905109112555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1072855593412792, "3000–6000 Hz_40ms_envelope_correlation": 0.9927733349294431, "3000–6000 Hz_40ms_modulation_mae_db": 1.7676912055746037, "mr_stft_db": 6.039868506387291, "band_mae_db": 0.14443731328161788, "rms_gain_error_db": -0.10068063598902534, "output_peak": 0.7067206514225176}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7496367795604502, "全帯域_10ms_rms_envelope_nrmse": 0.11841187456302785, "全帯域_10ms_envelope_correlation": 0.9867139866558191, "全帯域_10ms_modulation_mae_db": 0.24932895137015607, "全帯域_40ms_log_rms_mae_db": 0.9281339835757682, "全帯域_40ms_rms_envelope_nrmse": 0.06342724777760293, "全帯域_40ms_envelope_correlation": 0.9958628803954356, "全帯域_40ms_modulation_mae_db": 0.19803244082936058, "300–1000 Hz_10ms_log_rms_mae_db": 2.005423784696663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11070279903942915, "300–1000 Hz_10ms_envelope_correlation": 0.9922787441687533, "300–1000 Hz_10ms_modulation_mae_db": 0.6087866610766536, "300–1000 Hz_40ms_log_rms_mae_db": 1.302589605022553, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06600940504528224, "300–1000 Hz_40ms_envelope_correlation": 0.9973858755782183, "300–1000 Hz_40ms_modulation_mae_db": 0.8085639155291927, "1000–3000 Hz_10ms_log_rms_mae_db": 2.069260581361091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18264991937625885, "1000–3000 Hz_10ms_envelope_correlation": 0.9768601906687066, "1000–3000 Hz_10ms_modulation_mae_db": 0.5071641875686206, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3111278919383071, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10989487376264419, "1000–3000 Hz_40ms_envelope_correlation": 0.9909714832955434, "1000–3000 Hz_40ms_modulation_mae_db": 0.8726092294256418, "3000–6000 Hz_10ms_log_rms_mae_db": 2.111402482995221, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1702647649837517, "3000–6000 Hz_10ms_envelope_correlation": 0.98147929077045, "3000–6000 Hz_10ms_modulation_mae_db": 0.7088178078558179, "3000–6000 Hz_40ms_log_rms_mae_db": 1.372874870994751, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08659639999751809, "3000–6000 Hz_40ms_envelope_correlation": 0.994984851379401, "3000–6000 Hz_40ms_modulation_mae_db": 0.7428847689248288, "mr_stft_db": 6.057730694302873, "band_mae_db": 0.11485702194386171, "rms_gain_error_db": -0.09145239955141843, "output_peak": 0.7248600389489075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7239949973606574, "全帯域_10ms_rms_envelope_nrmse": 0.11922883425544643, "全帯域_10ms_envelope_correlation": 0.9866908527667605, "全帯域_10ms_modulation_mae_db": 0.5208680926560211, "全帯域_40ms_log_rms_mae_db": 0.8831979804465078, "全帯域_40ms_rms_envelope_nrmse": 0.062406365128299335, "全帯域_40ms_envelope_correlation": 0.9959943053774796, "全帯域_40ms_modulation_mae_db": 0.35218672253930183, "300–1000 Hz_10ms_log_rms_mae_db": 2.0097287189963824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10327341479980977, "300–1000 Hz_10ms_envelope_correlation": 0.9931135143568638, "300–1000 Hz_10ms_modulation_mae_db": 0.4190758714372094, "300–1000 Hz_40ms_log_rms_mae_db": 1.3216450759123506, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0655957037965156, "300–1000 Hz_40ms_envelope_correlation": 0.9970707257572512, "300–1000 Hz_40ms_modulation_mae_db": 0.6158821558544285, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1237444277258435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17726829941061484, "1000–3000 Hz_10ms_envelope_correlation": 0.9785112029376278, "1000–3000 Hz_10ms_modulation_mae_db": 1.1494765851940811, "1000–3000 Hz_40ms_log_rms_mae_db": 1.400239278381642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10868910285059973, "1000–3000 Hz_40ms_envelope_correlation": 0.9919689108043638, "1000–3000 Hz_40ms_modulation_mae_db": 1.4585027625389195, "3000–6000 Hz_10ms_log_rms_mae_db": 2.089747813808196, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19472193205428423, "3000–6000 Hz_10ms_envelope_correlation": 0.97599968917011, "3000–6000 Hz_10ms_modulation_mae_db": 0.9885335941565517, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3350220728417763, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10941651155829114, "3000–6000 Hz_40ms_envelope_correlation": 0.9927746608055616, "3000–6000 Hz_40ms_modulation_mae_db": 1.1646395222852446, "mr_stft_db": 6.055409804136801, "band_mae_db": 0.12124580424594024, "rms_gain_error_db": -0.06727797766018075, "output_peak": 0.7385945547660457}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.77556601771262, "全帯域_10ms_rms_envelope_nrmse": 0.11820221784971023, "全帯域_10ms_envelope_correlation": 0.9869945554431542, "全帯域_10ms_modulation_mae_db": 0.446414323967469, "全帯域_40ms_log_rms_mae_db": 0.9918811637943273, "全帯域_40ms_rms_envelope_nrmse": 0.06499568340724134, "全帯域_40ms_envelope_correlation": 0.9958653158195122, "全帯域_40ms_modulation_mae_db": 0.20918904580274736, "300–1000 Hz_10ms_log_rms_mae_db": 1.9347342001695842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11542535430655362, "300–1000 Hz_10ms_envelope_correlation": 0.9915636448784911, "300–1000 Hz_10ms_modulation_mae_db": 0.495121306541966, "300–1000 Hz_40ms_log_rms_mae_db": 1.253939285963569, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07163071430621856, "300–1000 Hz_40ms_envelope_correlation": 0.9967884567291533, "300–1000 Hz_40ms_modulation_mae_db": 0.7502652704705707, "1000–3000 Hz_10ms_log_rms_mae_db": 2.059329085538628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1590344532028959, "1000–3000 Hz_10ms_envelope_correlation": 0.9826376910241784, "1000–3000 Hz_10ms_modulation_mae_db": 0.657414767064788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3022628542034753, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09423459833365966, "1000–3000 Hz_40ms_envelope_correlation": 0.9934591176801765, "1000–3000 Hz_40ms_modulation_mae_db": 0.7760068124798494, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1469920568980245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18437489636311535, "3000–6000 Hz_10ms_envelope_correlation": 0.9783009986999333, "3000–6000 Hz_10ms_modulation_mae_db": 0.8815962332813277, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3357242465259467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10001577817685701, "3000–6000 Hz_40ms_envelope_correlation": 0.9934971691307463, "3000–6000 Hz_40ms_modulation_mae_db": 0.9753911197342636, "mr_stft_db": 6.038329768895305, "band_mae_db": 0.13128168365092557, "rms_gain_error_db": -0.15906109840541224, "output_peak": 0.6999391245273577}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.385718691451072, "spectral_convergence": 0.4243951466922347, "spectral_convergence_status": "ok", "amplitude_floor": 1.963677544918102e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.855418885888174, "spectral_convergence": 0.3752578671284085, "spectral_convergence_status": "ok", "amplitude_floor": 1.0999213683671372e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.891583037993778, "spectral_convergence": 0.3933863940854157, "spectral_convergence_status": "ok", "amplitude_floor": 9.527550859021228e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.000972009721466654, "output_power": 0.0010073929438901862, "delta_db": 0.15528296087005636, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00078771634185205, "output_power": 0.000787577250914993, "delta_db": -0.0007669227740267845, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008277001014055567, "output_power": 0.000790921290137599, "delta_db": -0.19739742304095073, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015338014805248517, "output_power": 0.0001435551493116425, "delta_db": -0.28750377423060824, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000153079755450355, "output_power": 0.00015251562809259474, "delta_db": -0.016034121590337132, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000364753212040036, "output_power": 0.0003769781831683578, "delta_db": 0.14317092032099732, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1333593538044961, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.0004013588204409619, "pitch": null, "output_rms_dbfs": -24.869235758405836, "output_peak": 0.713193763566102, "reference_rms_dbfs": -24.869637117226276, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.044240205111007}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002718271425028404, "log_rms_mae_db": 1.8067970295157896, "rms_envelope_nrmse": 0.1272943030820075, "envelope_correlation": 0.9848807276137699, "rms_gain_error_db": 0.0037213258762048076, "reference_mean_power": 0.0032374876277477364, "output_mean_power": 0.003240262912748526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2735378968683332e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004666640237447372, "output_power": 0.0004849715309775801, "reference_active": true, "delta_db": 0.16711923829153694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002202584731702016, "output_power": 0.0002484465093478996, "reference_active": true, "delta_db": 0.523002749517397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.478672520125922e-05, "output_power": 4.419598421761226e-05, "reference_active": true, "delta_db": 1.0396926333238663}], "active_band_count": 3, "active_band_mae_db": 0.5766048737109334}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002399603663054526, "log_rms_mae_db": 0.9529140826416242, "rms_envelope_nrmse": 0.07251643017738445, "envelope_correlation": 0.9945203274507387, "rms_gain_error_db": 0.013763606159769827, "reference_mean_power": 0.0031195469748277875, "output_mean_power": 0.003129449086445343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1403707959359127e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046404851798205753, "output_power": 0.0004854266780611265, "reference_active": true, "delta_db": 0.195602505748582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.861370772885771e-05, "output_power": 0.00010864992432691745, "reference_active": true, "delta_db": 0.42092140241531234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9585409286275983e-06, "output_power": 2.34224593543944e-06, "reference_active": true, "delta_db": 0.7769984232823}], "active_band_count": 3, "active_band_mae_db": 0.46450744381539816}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022360065406936124, "log_rms_mae_db": 1.9798483079559228, "rms_envelope_nrmse": 0.1294003833598693, "envelope_correlation": 0.9892189267026659, "rms_gain_error_db": -0.12167262855296222, "reference_mean_power": 0.0012597240435666534, "output_mean_power": 0.001224921212817583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2605416603296333e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001978275618555469, "output_power": 0.00019089316996653691, "reference_active": true, "delta_db": -0.1549640862697326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.107555923758901e-05, "output_power": 7.359567282398697e-05, "reference_active": true, "delta_db": -0.42037672957643574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1829892137914497e-05, "output_power": 1.4574159604750489e-05, "reference_active": true, "delta_db": 0.9060273641262266}], "active_band_count": 3, "active_band_mae_db": 0.4937893933241317}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019906826982741464, "log_rms_mae_db": 1.2999100786034257, "rms_envelope_nrmse": 0.07251224334620658, "envelope_correlation": 0.9965522506472522, "rms_gain_error_db": -0.10949763503206122, "reference_mean_power": 0.0011746644654148715, "output_mean_power": 0.0011454181687936072}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9316122263011187e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214336491155288, "output_power": 0.00020347900809078055, "reference_active": true, "delta_db": -0.18109841243719882}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6617395760512454e-05, "output_power": 3.2740130088221535e-05, "reference_active": true, "delta_db": -0.48607053231926134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0023494047314195e-06, "output_power": 8.529245272371856e-07, "reference_active": true, "delta_db": -0.7010853335994262}], "active_band_count": 3, "active_band_mae_db": 0.4560847594519621}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.946276793614836e-05, "log_rms_mae_db": 2.119558776953767, "rms_envelope_nrmse": 0.18689428612226813, "envelope_correlation": 0.9758545786347638, "rms_gain_error_db": -0.23526776657120058, "reference_mean_power": 0.00019501605476176557, "output_mean_power": 0.0001847326199515879}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.789505238315189e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.753576467028983e-05, "output_power": 3.5652836047653556e-05, "reference_active": true, "delta_db": -0.2235118560556407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.624079451059313e-05, "output_power": 1.436923303672477e-05, "reference_active": true, "delta_db": -0.5317368328328513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.016250052718367e-06, "output_power": 3.942953133478839e-06, "reference_active": true, "delta_db": -0.07999129244457354}], "active_band_count": 3, "active_band_mae_db": 0.27841332711102185}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.236492865510781e-05, "log_rms_mae_db": 1.3599609586523338, "rms_envelope_nrmse": 0.122025430085845, "envelope_correlation": 0.9891916299597906, "rms_gain_error_db": -0.2238436452126047, "reference_mean_power": 0.00017673870709398627, "output_mean_power": 0.00016786003492264438}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.939463723397018e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.781437846450251e-05, "output_power": 3.5523968647049844e-05, "reference_active": true, "delta_db": -0.2713548842839938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.77351369725072e-06, "output_power": 6.913264721809276e-06, "reference_active": true, "delta_db": -0.5093418099702894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.249207604003898e-07, "output_power": 2.7867971584031315e-07, "reference_active": true, "delta_db": -0.6667210189899114}], "active_band_count": 3, "active_band_mae_db": 0.4824725710813982}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014584429481341953, "log_rms_mae_db": 2.0909228359748164, "rms_envelope_nrmse": 0.17606628856544138, "envelope_correlation": 0.98024807438627, "rms_gain_error_db": -0.08472913375339053, "reference_mean_power": 0.00024652309966206675, "output_mean_power": 0.00024176014436269616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9096739107120075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.093194518056527e-05, "output_power": 7.69513249548759e-05, "reference_active": true, "delta_db": 0.3537423219945484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.832881668922571e-05, "output_power": 2.532381851911542e-05, "reference_active": true, "delta_db": -0.4869924164399936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4378297018808954e-05, "output_power": 9.888399719608928e-06, "reference_active": true, "delta_db": -1.625814371586722}], "active_band_count": 3, "active_band_mae_db": 0.8221830366737546}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.656067210642841e-05, "log_rms_mae_db": 1.3589617231429187, "rms_envelope_nrmse": 0.09015322778038574, "envelope_correlation": 0.9945727830525799, "rms_gain_error_db": -0.08314198817126081, "reference_mean_power": 0.0002460055124763468, "output_mean_power": 0.00024134074000261465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8079454178991016e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.471033966073099e-05, "output_power": 7.803724581031986e-05, "reference_active": true, "delta_db": 0.18921222443666014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5789672835999803e-05, "output_power": 1.3128275718639597e-05, "reference_active": true, "delta_db": -0.8016544231844105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1060211769697704e-06, "output_power": 6.96585835756569e-07, "reference_active": true, "delta_db": -2.0078880312688603}], "active_band_count": 3, "active_band_mae_db": 0.9995848929633103}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7395292426520407, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7df2077f01a500035c3f935044db299106788112582bc334b5d3f61d9e1376c4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9797884804589522, "standard_deviation": 0.0028809519047894917, "minimum": 0.9758545786347638, "maximum": 0.983124598613751, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1029202105431617, "standard_deviation": 0.03075690031353762, "minimum": 2.059329085538628, "maximum": 2.1402263371482526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9362470400212675, "standard_deviation": 0.4149290804890954, "minimum": 0.27841332711102185, "maximum": 1.4929767771766052, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17107284995107286, "standard_deviation": 0.012112126778346687, "minimum": 0.15661415980222462, "maximum": 0.18689428612226813, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9915620410696091, "standard_deviation": 0.0012879459397466386, "minimum": 0.9891916299597906, "maximum": 0.9934591176801765, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3611299364305467, "standard_deviation": 0.03662926333038786, "minimum": 1.3022628542034753, "maximum": 1.400239278381642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1837676984723626, "standard_deviation": 0.46150347195199753, "minimum": 0.4824725710813982, "maximum": 1.9026755701542737, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10820695567436499, "standard_deviation": 0.008420421661909474, "minimum": 0.09423459833365966, "maximum": 0.122025430085845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9786889457363885, "standard_deviation": 0.002770869882206671, "minimum": 0.9732953248997429, "maximum": 0.98147929077045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.093588305133232, "standard_deviation": 0.02957809298311045, "minimum": 2.0566193162154462, "maximum": 2.1469920568980245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8705003633918977, "standard_deviation": 0.20134640653793048, "minimum": 0.5694556570358427, "maximum": 1.2516350243148098, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18251540548186895, "standard_deviation": 0.011635924016656083, "minimum": 0.1702647649837517, "maximum": 0.2044489822645467, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940041393586363, "standard_deviation": 0.0008915945820417913, "minimum": 0.9927733349294431, "maximum": 0.994984851379401, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3490928289953965, "standard_deviation": 0.018954079331425205, "minimum": 1.3264958782420901, "maximum": 1.372874870994751, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.079106369572505, "standard_deviation": 0.3289915093306426, "minimum": 0.7081098583289059, "maximum": 1.7676912055746037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09612037038593578, "standard_deviation": 0.008494265921761758, "minimum": 0.08659639999751809, "maximum": 0.10941651155829114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915482314725634, "standard_deviation": 0.0011441333329559398, "minimum": 0.9892189267026659, "maximum": 0.9931135143568638, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9797225120709574, "standard_deviation": 0.05462984199316699, "minimum": 1.9016969297095998, "maximum": 2.0550597876053405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4944824943052449, "standard_deviation": 0.14532221341830442, "minimum": 0.27935127238775753, "maximum": 0.7541067107098268, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11513037785928354, "standard_deviation": 0.007434850605876177, "minimum": 0.10327341479980977, "maximum": 0.1294003833598693, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969071311667077, "standard_deviation": 0.0003146526462991368, "minimum": 0.9965522506472522, "maximum": 0.9973858755782183, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3022067184169475, "standard_deviation": 0.041706349639218106, "minimum": 1.253939285963569, "maximum": 1.3793512465950923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.717219863642517, "standard_deviation": 0.16792434546428503, "minimum": 0.4560847594519621, "maximum": 1.0198898226907842, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06961152248273184, "standard_deviation": 0.004081391802971159, "minimum": 0.06392825951571168, "maximum": 0.0759918557011752, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13510316979015025, "standard_deviation": 0.01785265085452141, "minimum": 0.11485702194386171, "maximum": 0.1625976485424154, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.047209956793296, "standard_deviation": 0.010083786652734064, "minimum": 6.0325529805061455, "maximum": 6.0603876680896605, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7216936813812789, "standard_deviation": 0.021837720940851808, "minimum": 0.6999391245273577, "maximum": 0.7643690584667285, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0732519959265234, "standard_deviation": 0.05346648509120302, "minimum": -0.15906109840541224, "maximum": 0.0004013588204409619, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9872306046188495, "standard_deviation": 0.001318038856707045, "minimum": 0.9848807276137699, "maximum": 0.98925549546383, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6972509364445205, "standard_deviation": 0.07977055094673102, "minimum": 1.590703766393316, "maximum": 1.8067970295157896, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3869443160251335, "standard_deviation": 0.15326200165928225, "minimum": 0.1678608488416969, "maximum": 0.5766048737109334, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11649918131867203, "standard_deviation": 0.006142431851798002, "minimum": 0.10700871567881565, "maximum": 0.1272943030820075, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9955778139776414, "standard_deviation": 0.0006217944735933754, "minimum": 0.9945203274507387, "maximum": 0.9962882836838876, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9095763306109854, "standard_deviation": 0.05922199122528956, "minimum": 0.8277934483867173, "maximum": 0.9918811637943273, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2828368171592509, "standard_deviation": 0.1344139300916552, "minimum": 0.11540269979711908, "maximum": 0.48317027063397777, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06548741908699504, "standard_deviation": 0.0041709941526934455, "minimum": 0.06048815475789655, "maximum": 0.07251643017738445, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13751-0001", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.1809709072113037, "predicted_mos": 2.1809709072113037, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "28026d30b0a5f94315884642499d3ab511fdd7d98f3da6c9e16ce836ca029b45", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.721433499827981, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.679893186988231, "全帯域_10ms_rms_envelope_nrmse": 0.11752545700540894, "全帯域_10ms_envelope_correlation": 0.9869401801542707, "全帯域_10ms_modulation_mae_db": 0.4050202003226013, "全帯域_40ms_log_rms_mae_db": 0.8783443507434626, "全帯域_40ms_rms_envelope_nrmse": 0.06377226905316809, "全帯域_40ms_envelope_correlation": 0.9957255185307347, "全帯域_40ms_modulation_mae_db": 0.174904069789806, "300–1000 Hz_10ms_log_rms_mae_db": 1.8084218837120942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1083081691841343, "300–1000 Hz_10ms_envelope_correlation": 0.9924878161399591, "300–1000 Hz_10ms_modulation_mae_db": 0.3682255262847323, "300–1000 Hz_40ms_log_rms_mae_db": 1.099553006754695, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058888503047819264, "300–1000 Hz_40ms_envelope_correlation": 0.9977663633356463, "300–1000 Hz_40ms_modulation_mae_db": 0.6747697019922895, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9169283794585283, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.164646845898444, "1000–3000 Hz_10ms_envelope_correlation": 0.9814358172967462, "1000–3000 Hz_10ms_modulation_mae_db": 0.9534984606346018, "1000–3000 Hz_40ms_log_rms_mae_db": 1.175619545911369, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10969287021772293, "1000–3000 Hz_40ms_envelope_correlation": 0.9913190080945858, "1000–3000 Hz_40ms_modulation_mae_db": 1.0116885743901232, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9976938502115007, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18866466949059027, "3000–6000 Hz_10ms_envelope_correlation": 0.9775328692562565, "3000–6000 Hz_10ms_modulation_mae_db": 1.3256794322682584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2065923793473448, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10841920579627831, "3000–6000 Hz_40ms_envelope_correlation": 0.9928363498949375, "3000–6000 Hz_40ms_modulation_mae_db": 1.6558495820848815, "mr_stft_db": 5.955812285667012, "band_mae_db": 0.17359440760979214, "rms_gain_error_db": -0.027622585621789805, "output_peak": 0.6900757528729458}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6372283063537414, "全帯域_10ms_rms_envelope_nrmse": 0.10488796461743968, "全帯域_10ms_envelope_correlation": 0.9896371189339627, "全帯域_10ms_modulation_mae_db": 0.3311119846037785, "全帯域_40ms_log_rms_mae_db": 0.9339180422243532, "全帯域_40ms_rms_envelope_nrmse": 0.05705201943215151, "全帯域_40ms_envelope_correlation": 0.9966627666855689, "全帯域_40ms_modulation_mae_db": 0.5193340699830116, "300–1000 Hz_10ms_log_rms_mae_db": 1.8263157689583465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1094992348274974, "300–1000 Hz_10ms_envelope_correlation": 0.9923150299396858, "300–1000 Hz_10ms_modulation_mae_db": 0.5980636877254913, "300–1000 Hz_40ms_log_rms_mae_db": 1.1926890996006763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0639264681951259, "300–1000 Hz_40ms_envelope_correlation": 0.9973221405303303, "300–1000 Hz_40ms_modulation_mae_db": 0.673718661771896, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9514682807440653, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17027654098025605, "1000–3000 Hz_10ms_envelope_correlation": 0.9799216750911443, "1000–3000 Hz_10ms_modulation_mae_db": 0.8460875702185735, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2005163971077428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10687585976288713, "1000–3000 Hz_40ms_envelope_correlation": 0.9916503290157586, "1000–3000 Hz_40ms_modulation_mae_db": 1.1553675596633133, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8815247916670426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17320218855723074, "3000–6000 Hz_10ms_envelope_correlation": 0.9809919196105619, "3000–6000 Hz_10ms_modulation_mae_db": 0.8344811036688659, "3000–6000 Hz_40ms_log_rms_mae_db": 1.154573795856132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10272764052354753, "3000–6000 Hz_40ms_envelope_correlation": 0.9932776548204115, "3000–6000 Hz_40ms_modulation_mae_db": 1.1807054465323021, "mr_stft_db": 5.956733040080731, "band_mae_db": 0.138706817129134, "rms_gain_error_db": -0.09866405891553384, "output_peak": 0.7119471055674749}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.634110842009811, "全帯域_10ms_rms_envelope_nrmse": 0.12454133759957202, "全帯域_10ms_envelope_correlation": 0.9853883400785832, "全帯域_10ms_modulation_mae_db": 0.11483326742436754, "全帯域_40ms_log_rms_mae_db": 0.8637250809837703, "全帯域_40ms_rms_envelope_nrmse": 0.07083555630358351, "全帯域_40ms_envelope_correlation": 0.9947335927220199, "全帯域_40ms_modulation_mae_db": 0.47395910784737233, "300–1000 Hz_10ms_log_rms_mae_db": 1.8096931748951874, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12202454689881329, "300–1000 Hz_10ms_envelope_correlation": 0.9904322685043343, "300–1000 Hz_10ms_modulation_mae_db": 0.5549146865106553, "300–1000 Hz_40ms_log_rms_mae_db": 1.1775394170203262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08243587500317907, "300–1000 Hz_40ms_envelope_correlation": 0.9954666142026574, "300–1000 Hz_40ms_modulation_mae_db": 0.9408214161603295, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9753473535806692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16818755909127833, "1000–3000 Hz_10ms_envelope_correlation": 0.9807623485837544, "1000–3000 Hz_10ms_modulation_mae_db": 1.0601497531675594, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1907316535525694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10320610398138992, "1000–3000 Hz_40ms_envelope_correlation": 0.9930889919225959, "1000–3000 Hz_40ms_modulation_mae_db": 1.2634377588075696, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9363014416844249, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1770632951071578, "3000–6000 Hz_10ms_envelope_correlation": 0.9802835356244822, "3000–6000 Hz_10ms_modulation_mae_db": 0.8981095192587693, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1710153044508265, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09597607812404782, "3000–6000 Hz_40ms_envelope_correlation": 0.9945702739940043, "3000–6000 Hz_40ms_modulation_mae_db": 1.2034206886111105, "mr_stft_db": 5.971540354514256, "band_mae_db": 0.21974459537309113, "rms_gain_error_db": -0.022883431415448904, "output_peak": 0.7183631358029988}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6004681049568283, "全帯域_10ms_rms_envelope_nrmse": 0.10116724632171124, "全帯域_10ms_envelope_correlation": 0.9903623997559292, "全帯域_10ms_modulation_mae_db": 0.2510620119373981, "全帯域_40ms_log_rms_mae_db": 0.894907774622173, "全帯域_40ms_rms_envelope_nrmse": 0.056962296561556505, "全帯域_40ms_envelope_correlation": 0.9966965114668989, "全帯域_40ms_modulation_mae_db": 0.19443011250538386, "300–1000 Hz_10ms_log_rms_mae_db": 1.8561516930533626, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10493476996251706, "300–1000 Hz_10ms_envelope_correlation": 0.9929919053712005, "300–1000 Hz_10ms_modulation_mae_db": 0.7348439465564253, "300–1000 Hz_40ms_log_rms_mae_db": 1.136191898171721, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06018282983665146, "300–1000 Hz_40ms_envelope_correlation": 0.99770490471807, "300–1000 Hz_40ms_modulation_mae_db": 1.0982087826287292, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9672436482899018, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1701521030361751, "1000–3000 Hz_10ms_envelope_correlation": 0.9799868461521914, "1000–3000 Hz_10ms_modulation_mae_db": 0.4224370801961031, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1988421571893288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1031975656749323, "1000–3000 Hz_40ms_envelope_correlation": 0.9923040503279004, "1000–3000 Hz_40ms_modulation_mae_db": 0.8008508058916387, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9886028841492973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17753267076590304, "3000–6000 Hz_10ms_envelope_correlation": 0.9798967695150577, "3000–6000 Hz_10ms_modulation_mae_db": 0.37729773506910824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2148385469447411, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09288775644624507, "3000–6000 Hz_40ms_envelope_correlation": 0.9945027311738177, "3000–6000 Hz_40ms_modulation_mae_db": 0.4450504714354211, "mr_stft_db": 5.966976031204635, "band_mae_db": 0.11632529335455956, "rms_gain_error_db": -0.09713402372384423, "output_peak": 0.6882066549780737}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6748980568914795, "全帯域_10ms_rms_envelope_nrmse": 0.13052271283963934, "全帯域_10ms_envelope_correlation": 0.9839964278889568, "全帯域_10ms_modulation_mae_db": 0.5838221104802342, "全帯域_40ms_log_rms_mae_db": 0.8965778651001257, "全帯域_40ms_rms_envelope_nrmse": 0.07586329267213641, "全帯域_40ms_envelope_correlation": 0.9940349408405956, "全帯域_40ms_modulation_mae_db": 1.1881711183636812, "300–1000 Hz_10ms_log_rms_mae_db": 1.873839563368476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11358228937446603, "300–1000 Hz_10ms_envelope_correlation": 0.991728732126962, "300–1000 Hz_10ms_modulation_mae_db": 0.6410022206179136, "300–1000 Hz_40ms_log_rms_mae_db": 1.1546647867028, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07391407413027452, "300–1000 Hz_40ms_envelope_correlation": 0.9963667519659388, "300–1000 Hz_40ms_modulation_mae_db": 0.9036237540507516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.991276422945695, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17622405859108758, "1000–3000 Hz_10ms_envelope_correlation": 0.9786416233036433, "1000–3000 Hz_10ms_modulation_mae_db": 0.6783334801393259, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1916146479940606, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11246715236867721, "1000–3000 Hz_40ms_envelope_correlation": 0.9909835713230568, "1000–3000 Hz_40ms_modulation_mae_db": 0.897297132287911, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9350662779761862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1900553407709796, "3000–6000 Hz_10ms_envelope_correlation": 0.9770626111701203, "3000–6000 Hz_10ms_modulation_mae_db": 1.0728698750271892, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1462487471853062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09478213755334162, "3000–6000 Hz_40ms_envelope_correlation": 0.9946463825788631, "3000–6000 Hz_40ms_modulation_mae_db": 1.346434225024115, "mr_stft_db": 5.963181870997951, "band_mae_db": 0.145015468772184, "rms_gain_error_db": -0.08415764372386543, "output_peak": 0.7381671213019168}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7608995747440583, "全帯域_10ms_rms_envelope_nrmse": 0.1169593798073532, "全帯域_10ms_envelope_correlation": 0.9873045638408273, "全帯域_10ms_modulation_mae_db": 0.13056112601302136, "全帯域_40ms_log_rms_mae_db": 1.0117766907333279, "全帯域_40ms_rms_envelope_nrmse": 0.06431531176122138, "全帯域_40ms_envelope_correlation": 0.9957751803579583, "全帯域_40ms_modulation_mae_db": 0.1972072159845851, "300–1000 Hz_10ms_log_rms_mae_db": 1.9226122451025995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11460154684876735, "300–1000 Hz_10ms_envelope_correlation": 0.9916024365444805, "300–1000 Hz_10ms_modulation_mae_db": 0.7568573495384899, "300–1000 Hz_40ms_log_rms_mae_db": 1.1476002860374828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06531169540386847, "300–1000 Hz_40ms_envelope_correlation": 0.9971879173769126, "300–1000 Hz_40ms_modulation_mae_db": 1.1283806593710288, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9251993077174907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15885193665662486, "1000–3000 Hz_10ms_envelope_correlation": 0.9825745367114351, "1000–3000 Hz_10ms_modulation_mae_db": 0.29445643791025367, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1135326006795532, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09117262764687922, "1000–3000 Hz_40ms_envelope_correlation": 0.9937827112989981, "1000–3000 Hz_40ms_modulation_mae_db": 0.42479077221550404, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8975191105327682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1814252931791741, "3000–6000 Hz_10ms_envelope_correlation": 0.9790539472927224, "3000–6000 Hz_10ms_modulation_mae_db": 0.6444921020579361, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1756486365418362, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09609550236569502, "3000–6000 Hz_40ms_envelope_correlation": 0.9937689197660776, "3000–6000 Hz_40ms_modulation_mae_db": 0.8834931048893462, "mr_stft_db": 5.974975613039548, "band_mae_db": 0.08187479124620277, "rms_gain_error_db": 0.025560832038973268, "output_peak": 0.6825657470802229}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7745033267916757, "全帯域_10ms_rms_envelope_nrmse": 0.1203219531708347, "全帯域_10ms_envelope_correlation": 0.98635166964071, "全帯域_10ms_modulation_mae_db": 0.23946357060586146, "全帯域_40ms_log_rms_mae_db": 1.0159184850449843, "全帯域_40ms_rms_envelope_nrmse": 0.0651702912136752, "全帯域_40ms_envelope_correlation": 0.9955722545542348, "全帯域_40ms_modulation_mae_db": 0.7854057515256195, "300–1000 Hz_10ms_log_rms_mae_db": 1.915378091015934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11601948239796213, "300–1000 Hz_10ms_envelope_correlation": 0.991366827167273, "300–1000 Hz_10ms_modulation_mae_db": 0.3418406552139506, "300–1000 Hz_40ms_log_rms_mae_db": 1.1642061986754944, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06118482818170194, "300–1000 Hz_40ms_envelope_correlation": 0.9975871891070026, "300–1000 Hz_40ms_modulation_mae_db": 0.7438060781077548, "1000–3000 Hz_10ms_log_rms_mae_db": 1.954311319348097, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17376254985178913, "1000–3000 Hz_10ms_envelope_correlation": 0.9791115557685509, "1000–3000 Hz_10ms_modulation_mae_db": 0.831796502682704, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1730396727358199, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10867392634018029, "1000–3000 Hz_40ms_envelope_correlation": 0.9913100957550085, "1000–3000 Hz_40ms_modulation_mae_db": 0.914364499953883, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9653728569548712, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17907494378220212, "3000–6000 Hz_10ms_envelope_correlation": 0.9794913909579183, "3000–6000 Hz_10ms_modulation_mae_db": 0.6536922561707093, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1481451991684932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08421446671427253, "3000–6000 Hz_40ms_envelope_correlation": 0.9954499332312972, "3000–6000 Hz_40ms_modulation_mae_db": 0.9909091497692089, "mr_stft_db": 5.972423415735349, "band_mae_db": 0.18475055718620137, "rms_gain_error_db": -0.05074947299460264, "output_peak": 0.7130902186217112}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6462948708962357, "全帯域_10ms_rms_envelope_nrmse": 0.11569607853415706, "全帯域_10ms_envelope_correlation": 0.9874286991302768, "全帯域_10ms_modulation_mae_db": 0.6329922372753134, "全帯域_40ms_log_rms_mae_db": 0.8656892047139357, "全帯域_40ms_rms_envelope_nrmse": 0.06556949551566553, "全帯域_40ms_envelope_correlation": 0.9956523946950757, "全帯域_40ms_modulation_mae_db": 0.40112314005289224, "300–1000 Hz_10ms_log_rms_mae_db": 1.8271383744617153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10047414687919227, "300–1000 Hz_10ms_envelope_correlation": 0.9936143276230395, "300–1000 Hz_10ms_modulation_mae_db": 0.46075020459888716, "300–1000 Hz_40ms_log_rms_mae_db": 1.1338889615281162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057502611799198876, "300–1000 Hz_40ms_envelope_correlation": 0.9979668825237519, "300–1000 Hz_40ms_modulation_mae_db": 0.3811646592690449, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9929016818583243, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16544186237955508, "1000–3000 Hz_10ms_envelope_correlation": 0.9811569351231785, "1000–3000 Hz_10ms_modulation_mae_db": 1.1426524101058202, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1608869848265913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10211225061047631, "1000–3000 Hz_40ms_envelope_correlation": 0.9926057323742649, "1000–3000 Hz_40ms_modulation_mae_db": 1.2933218312991857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.920766032083342, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17788589167458038, "3000–6000 Hz_10ms_envelope_correlation": 0.9801015155687386, "3000–6000 Hz_10ms_modulation_mae_db": 0.42889621427705343, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1628051062653721, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10530494233314618, "3000–6000 Hz_40ms_envelope_correlation": 0.9932449153450524, "3000–6000 Hz_40ms_modulation_mae_db": 0.6127045604504243, "mr_stft_db": 5.9512713926179766, "band_mae_db": 0.1773994214264084, "rms_gain_error_db": -0.1170517041325813, "output_peak": 0.6974664075977103}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.301405638694034, "spectral_convergence": 0.42412253445557674, "spectral_convergence_status": "ok", "amplitude_floor": 1.963677544918102e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.757760349346631, "spectral_convergence": 0.366196545605118, "spectral_convergence_status": "ok", "amplitude_floor": 1.0999213683671372e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.80827086896037, "spectral_convergence": 0.38730894642481856, "spectral_convergence_status": "ok", "amplitude_floor": 9.527550859021228e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.000972009721466654, "output_power": 0.0010017767021613406, "delta_db": 0.13100318788497883, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00078771634185205, "output_power": 0.0007948690404067918, "delta_db": 0.03925726472006472, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008277001014055567, "output_power": 0.0008102376264776914, "delta_db": -0.09260600968086644, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015338014805248517, "output_power": 0.00014709021671098613, "delta_db": -0.18185364821927713, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000153079755450355, "output_power": 0.0001425111357909314, "delta_db": -0.31068958401912267, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000364753212040036, "output_power": 0.0003414942927468325, "delta_db": -0.286156751134443, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17359440760979214, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.027622585621789805, "pitch": null, "output_rms_dbfs": -24.897259702848068, "output_peak": 0.6900757528729458, "reference_rms_dbfs": -24.869637117226276, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.955812285667012}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002718271425028404, "log_rms_mae_db": 1.679893186988231, "rms_envelope_nrmse": 0.11752545700540894, "envelope_correlation": 0.9869401801542707, "rms_gain_error_db": -0.026778362995807662, "reference_mean_power": 0.0032374876277477364, "output_mean_power": 0.0032175868705845998}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2735378968683332e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004666640237447372, "output_power": 0.00046740100557848797, "reference_active": true, "delta_db": 0.00685321043032358}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002202584731702016, "output_power": 0.0002033403214835638, "reference_active": true, "delta_db": -0.3470911852602352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.478672520125922e-05, "output_power": 4.241555626675406e-05, "reference_active": true, "delta_db": 0.861116205277245}], "active_band_count": 3, "active_band_mae_db": 0.4050202003226013}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002399603663054526, "log_rms_mae_db": 0.8783443507434626, "rms_envelope_nrmse": 0.06377226905316809, "envelope_correlation": 0.9957255185307347, "rms_gain_error_db": -0.019167578615200496, "reference_mean_power": 0.0031195469748277875, "output_mean_power": 0.0031058091983724063}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1403707959359127e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046404851798205753, "output_power": 0.00045884211353201034, "reference_active": true, "delta_db": -0.049001184614477444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.861370772885771e-05, "output_power": 9.021419320283496e-05, "reference_active": true, "delta_db": -0.3866241841160951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9585409286275983e-06, "output_power": 1.999131377511532e-06, "reference_active": true, "delta_db": 0.0890868406388454}], "active_band_count": 3, "active_band_mae_db": 0.174904069789806}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022360065406936124, "log_rms_mae_db": 1.8084218837120942, "rms_envelope_nrmse": 0.1083081691841343, "envelope_correlation": 0.9924878161399591, "rms_gain_error_db": -0.10905461749901388, "reference_mean_power": 0.0012597240435666534, "output_mean_power": 0.0012284852793382406}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2605416603296333e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001978275618555469, "output_power": 0.0001817356678699321, "reference_active": true, "delta_db": -0.36846627193890097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.107555923758901e-05, "output_power": 7.012697565915168e-05, "reference_active": true, "delta_db": -0.6300484336209654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1829892137914497e-05, "output_power": 1.2122633408438869e-05, "reference_active": true, "delta_db": 0.10616187329433058}], "active_band_count": 3, "active_band_mae_db": 0.3682255262847323}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019906826982741464, "log_rms_mae_db": 1.099553006754695, "rms_envelope_nrmse": 0.058888503047819264, "envelope_correlation": 0.9977663633356463, "rms_gain_error_db": -0.11224149143847198, "reference_mean_power": 0.0011746644654148715, "output_mean_power": 0.0011446947264069642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9316122263011187e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214336491155288, "output_power": 0.00019272213995850072, "reference_active": true, "delta_db": -0.41697843771760357}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6617395760512454e-05, "output_power": 3.1810520154095376e-05, "reference_active": true, "delta_db": -0.6111668332687418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0023494047314195e-06, "output_power": 7.968980301538766e-07, "reference_active": true, "delta_db": -0.9961638349905232}], "active_band_count": 3, "active_band_mae_db": 0.6747697019922895}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.946276793614836e-05, "log_rms_mae_db": 1.9169283794585283, "rms_envelope_nrmse": 0.164646845898444, "envelope_correlation": 0.9814358172967462, "rms_gain_error_db": -0.257309954856156, "reference_mean_power": 0.00019501605476176557, "output_mean_power": 0.00018379740305881468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.789505238315189e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.753576467028983e-05, "output_power": 3.433082124312031e-05, "reference_active": true, "delta_db": -0.3876107513861768}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.624079451059313e-05, "output_power": 1.3821950650550134e-05, "reference_active": true, "delta_db": -0.7003793335846392}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.016250052718367e-06, "output_power": 2.670362463123097e-06, "reference_active": true, "delta_db": -1.7725052969329893}], "active_band_count": 3, "active_band_mae_db": 0.9534984606346018}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.236492865510781e-05, "log_rms_mae_db": 1.175619545911369, "rms_envelope_nrmse": 0.10969287021772293, "envelope_correlation": 0.9913190080945858, "rms_gain_error_db": -0.22873291654158856, "reference_mean_power": 0.00017673870709398627, "output_mean_power": 0.0001676711650462035}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.939463723397018e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.781437846450251e-05, "output_power": 3.37625792228294e-05, "reference_active": true, "delta_db": -0.4922135026063896}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.77351369725072e-06, "output_power": 6.627569419674552e-06, "reference_active": true, "delta_db": -0.692630827501602}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.249207604003898e-07, "output_power": 2.1220485782909985e-07, "reference_active": true, "delta_db": -1.8502213930623777}], "active_band_count": 3, "active_band_mae_db": 1.0116885743901232}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014584429481341953, "log_rms_mae_db": 1.9976938502115007, "rms_envelope_nrmse": 0.18866466949059027, "envelope_correlation": 0.9775328692562565, "rms_gain_error_db": -0.33171346769196897, "reference_mean_power": 0.00024652309966206675, "output_mean_power": 0.00022839483212513253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9096739107120075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.093194518056527e-05, "output_power": 7.088496168291918e-05, "reference_active": true, "delta_db": -0.002877608173245407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.832881668922571e-05, "output_power": 2.2176848331513885e-05, "reference_active": true, "delta_db": -1.0632860784519524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4378297018808954e-05, "output_power": 7.3556319368570315e-06, "reference_active": true, "delta_db": -2.9108746101795773}], "active_band_count": 3, "active_band_mae_db": 1.3256794322682584}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.656067210642841e-05, "log_rms_mae_db": 1.2065923793473448, "rms_envelope_nrmse": 0.10841920579627831, "envelope_correlation": 0.9928363498949375, "rms_gain_error_db": -0.33295584437759873, "reference_mean_power": 0.0002460055124763468, "output_mean_power": 0.00022785011620875414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8079454178991016e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.471033966073099e-05, "output_power": 7.07015154229385e-05, "reference_active": true, "delta_db": -0.23951988270880173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5789672835999803e-05, "output_power": 1.1331544123606554e-05, "reference_active": true, "delta_db": -1.4408403725350079}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1060211769697704e-06, "output_power": 5.188533120750063e-07, "reference_active": true, "delta_db": -3.287188491010835}], "active_band_count": 3, "active_band_mae_db": 1.6558495820848815}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7461452160175378, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0830a7a47b5866e9e057087fb45a715e00c8af97772d6972885abba50602e9c7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9804489172538305, "standard_deviation": 0.0012894129029980333, "minimum": 0.9786416233036433, "maximum": 0.9825745367114351, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9593345492428464, "standard_deviation": 0.028072885765990672, "minimum": 1.9169283794585283, "maximum": 1.9929016818583243, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7786764618818677, "standard_deviation": 0.29786830137936615, "minimum": 0.29445643791025367, "maximum": 1.1426524101058202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16844293206065128, "standard_deviation": 0.005485002255440555, "minimum": 0.15885193665662486, "maximum": 0.17622405859108758, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9921305612640211, "standard_deviation": 0.0009843102182247, "minimum": 0.9909835713230568, "maximum": 0.9937827112989981, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1755979574996294, "standard_deviation": 0.028590335788558022, "minimum": 1.1135326006795532, "maximum": 1.2005163971077428, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9701398668136412, "standard_deviation": 0.2830919097509031, "minimum": 0.42479077221550404, "maximum": 1.2933218312991857, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10467479457539317, "standard_deviation": 0.006546035513251874, "minimum": 0.09117262764687922, "maximum": 0.11246715236867721, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793018198744823, "standard_deviation": 0.0013656134745863421, "minimum": 0.9770626111701203, "maximum": 0.9809919196105619, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.940355905657429, "standard_deviation": 0.04134442724093102, "minimum": 1.8815247916670426, "maximum": 1.9976938502115007, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7794397797247362, "standard_deviation": 0.32076100671614605, "minimum": 0.37729773506910824, "maximum": 1.3256794322682584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18061303666597728, "standard_deviation": 0.0058731677632121795, "minimum": 0.17320218855723074, "maximum": 0.1900553407709796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940371451005576, "standard_deviation": 0.0008933440261186596, "minimum": 0.9928363498949375, "maximum": 0.9954499332312972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1724834644700064, "standard_deviation": 0.02581412484316152, "minimum": 1.1462487471853062, "maximum": 1.2148385469447411, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.039820903599601, "standard_deviation": 0.3933605005151511, "minimum": 0.4450504714354211, "maximum": 1.6558495820848815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09755096623207175, "standard_deviation": 0.007717452744889477, "minimum": 0.08421446671427253, "maximum": 0.10841920579627831, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920674179271168, "standard_deviation": 0.0009986497893775495, "minimum": 0.9904322685043343, "maximum": 0.9936143276230395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8549438493209645, "standard_deviation": 0.04534698681161679, "minimum": 1.8084218837120942, "maximum": 1.9226122451025995, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5570622846308182, "standard_deviation": 0.15653010445142287, "minimum": 0.3418406552139506, "maximum": 0.7568573495384899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11118052329666873, "standard_deviation": 0.006794886052576118, "minimum": 0.10047414687919227, "maximum": 0.12202454689881329, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971710954700388, "standard_deviation": 0.0008469297107720566, "minimum": 0.9954666142026574, "maximum": 0.9979668825237519, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.150791706811414, "standard_deviation": 0.02878548151973391, "minimum": 1.099553006754695, "maximum": 1.1926890996006763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8180617141689781, "standard_deviation": 0.2492650592715899, "minimum": 0.3811646592690449, "maximum": 1.1283806593710288, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06541836069972744, "standard_deviation": 0.00857439513154126, "minimum": 0.057502611799198876, "maximum": 0.08243587500317907, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15467641901219667, "standard_deviation": 0.043312476726963596, "minimum": 0.08187479124620277, "maximum": 0.21974459537309113, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.964114250482182, "standard_deviation": 0.008779533425702578, "minimum": 5.9512713926179766, "maximum": 5.974975613039548, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7049852679778817, "standard_deviation": 0.01873554230670975, "minimum": 0.6825657470802229, "maximum": 0.7381671213019168, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05908776106108661, "standard_deviation": 0.04860605676515655, "minimum": -0.1170517041325813, "maximum": 0.025560832038973268, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9871761749279395, "standard_deviation": 0.002081591884776047, "minimum": 0.9839964278889568, "maximum": 0.9903623997559292, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6760370337040076, "standard_deviation": 0.0618438252392241, "minimum": 1.6004681049568283, "maximum": 1.7745033267916757, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.336108313582822, "standard_deviation": 0.1934873096232163, "minimum": 0.11483326742436754, "maximum": 0.6329922372753134, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11645276623701452, "standard_deviation": 0.009619404839563347, "minimum": 0.10116724632171124, "maximum": 0.13052271283963934, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9956066449816359, "standard_deviation": 0.0008929475265118632, "minimum": 0.9940349408405956, "maximum": 0.9966965114668989, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9201071867707666, "standard_deviation": 0.06192953946700265, "minimum": 0.8637250809837703, "maximum": 1.0159184850449843, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49181682325654397, "standard_deviation": 0.34970856721697124, "minimum": 0.174904069789806, "maximum": 1.1881711183636812, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06494256656414477, "standard_deviation": 0.00634871160641011, "minimum": 0.056962296561556505, "maximum": 0.07586329267213641, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4077-13754-0002", "language": "en", "style": "normal", "speaker": "librispeech_4077", "source_utterance_id": "4077-13754-0002", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4077-13754-0002.wav", "source_file_sha256": "0ae3803b1891dafb40d50e63bd4819e9392fb1d6fc2beb1cbb07f48451430055", "decoded_pcm_sha256": "e4ef84f14b8958b7a78512946198f223e31d6190beadfe364d4772e4a9963ced", "original_source_sha256": "58ff360d46e4fc6bcaf04da1a6ead97dc191c54426528cfa13efe7e780349ff5", "corpus_entry_id": "librispeech_4077-13754-0002"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.470141887664795, "predicted_mos": 3.470141887664795, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d5268ab53285e3bc6ed8483a2a804f99f23eece1836d96e4a04451974aa8bb05", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.059448212999996, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4077-13754-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d5268ab53285e3bc6ed8483a2a804f99f23eece1836d96e4a04451974aa8bb05", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "5ebee7ab93321e8f9c22488b77a9d83b935967cc2b1755b26ca51d20e4dbcdd6"}, "rows": [{"case": "librispeech_4077-13754-0002", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.155151605606079, "predicted_mos": 2.155151605606079, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "85add4d1266554b03849ddb2f494323cbf7ede4258648cf43460d817c09b4f46", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.753954468997108, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4077-13754-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "85add4d1266554b03849ddb2f494323cbf7ede4258648cf43460d817c09b4f46", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2149777306883278, "全帯域_10ms_rms_envelope_nrmse": 0.12691294716721774, "全帯域_10ms_envelope_correlation": 0.9857701705204579, "全帯域_10ms_modulation_mae_db": 0.8597462945458837, "全帯域_40ms_log_rms_mae_db": 0.6602544225615469, "全帯域_40ms_rms_envelope_nrmse": 0.0803653122203358, "全帯域_40ms_envelope_correlation": 0.9944691219054574, "全帯域_40ms_modulation_mae_db": 0.903327865826788, "300–1000 Hz_10ms_log_rms_mae_db": 1.8659807113466869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18615559108162108, "300–1000 Hz_10ms_envelope_correlation": 0.9774965925485424, "300–1000 Hz_10ms_modulation_mae_db": 1.0434826683812726, "300–1000 Hz_40ms_log_rms_mae_db": 1.1097144047574665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1248442328043865, "300–1000 Hz_40ms_envelope_correlation": 0.9914863499274812, "300–1000 Hz_40ms_modulation_mae_db": 1.2823922961354215, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8434114729815192, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1739628454004172, "1000–3000 Hz_10ms_envelope_correlation": 0.975665524277458, "1000–3000 Hz_10ms_modulation_mae_db": 0.3682313035787745, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0759044356572216, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10231051717854255, "1000–3000 Hz_40ms_envelope_correlation": 0.9908350134853728, "1000–3000 Hz_40ms_modulation_mae_db": 0.49525232484935383, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9558069928978055, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16597080541910147, "3000–6000 Hz_10ms_envelope_correlation": 0.9798945720151674, "3000–6000 Hz_10ms_modulation_mae_db": 0.7907700402048374, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1166361873142627, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09117475680006205, "3000–6000 Hz_40ms_envelope_correlation": 0.9935095615290434, "3000–6000 Hz_40ms_modulation_mae_db": 0.8966288993943209, "mr_stft_db": 6.0406959369757205, "band_mae_db": 0.3495260428579687, "rms_gain_error_db": -0.29789730835478156, "output_peak": 0.49564394274605506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1907404065884766, "全帯域_10ms_rms_envelope_nrmse": 0.1124749698910268, "全帯域_10ms_envelope_correlation": 0.9881538498581668, "全帯域_10ms_modulation_mae_db": 0.7795730398393679, "全帯域_40ms_log_rms_mae_db": 0.6109595457058572, "全帯域_40ms_rms_envelope_nrmse": 0.06562649672269039, "全帯域_40ms_envelope_correlation": 0.9954794663508248, "全帯域_40ms_modulation_mae_db": 0.7011857939962268, "300–1000 Hz_10ms_log_rms_mae_db": 1.7528846233577529, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1446650789751502, "300–1000 Hz_10ms_envelope_correlation": 0.9851509637890816, "300–1000 Hz_10ms_modulation_mae_db": 1.1102614299324884, "300–1000 Hz_40ms_log_rms_mae_db": 1.0289192466294552, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08907263359267457, "300–1000 Hz_40ms_envelope_correlation": 0.9939790087591547, "300–1000 Hz_40ms_modulation_mae_db": 1.0736177471060981, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8823662350096708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19517491676902926, "1000–3000 Hz_10ms_envelope_correlation": 0.9716513247419526, "1000–3000 Hz_10ms_modulation_mae_db": 1.3515557031758678, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1241881438360917, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12993352256962037, "1000–3000 Hz_40ms_envelope_correlation": 0.9887109011925324, "1000–3000 Hz_40ms_modulation_mae_db": 1.3275991918747325, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0074559591096373, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16859976108220293, "3000–6000 Hz_10ms_envelope_correlation": 0.9792775730026723, "3000–6000 Hz_10ms_modulation_mae_db": 0.5387958453057503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1653712382968633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08508883082604576, "3000–6000 Hz_40ms_envelope_correlation": 0.9945549006887078, "3000–6000 Hz_40ms_modulation_mae_db": 0.6645999180905838, "mr_stft_db": 6.033906461415653, "band_mae_db": 0.3210005310793204, "rms_gain_error_db": -0.07013973244554284, "output_peak": 0.6664639462291506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.181748781299876, "全帯域_10ms_rms_envelope_nrmse": 0.11179511873130588, "全帯域_10ms_envelope_correlation": 0.988359003833563, "全帯域_10ms_modulation_mae_db": 0.6013562786787254, "全帯域_40ms_log_rms_mae_db": 0.6196039965550788, "全帯域_40ms_rms_envelope_nrmse": 0.06378398324431422, "全帯域_40ms_envelope_correlation": 0.9958477828730748, "全帯域_40ms_modulation_mae_db": 0.5527106753711832, "300–1000 Hz_10ms_log_rms_mae_db": 1.779737057279032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15904943908471533, "300–1000 Hz_10ms_envelope_correlation": 0.982076325293601, "300–1000 Hz_10ms_modulation_mae_db": 0.2389740564047409, "300–1000 Hz_40ms_log_rms_mae_db": 1.102328530366524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09629324550015768, "300–1000 Hz_40ms_envelope_correlation": 0.9931885284339059, "300–1000 Hz_40ms_modulation_mae_db": 0.2956352939644756, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8066293438394987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18352947112266768, "1000–3000 Hz_10ms_envelope_correlation": 0.9736781605062615, "1000–3000 Hz_10ms_modulation_mae_db": 0.857916718548935, "1000–3000 Hz_40ms_log_rms_mae_db": 1.085848067126847, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10055123674744305, "1000–3000 Hz_40ms_envelope_correlation": 0.9926985814461633, "1000–3000 Hz_40ms_modulation_mae_db": 0.9498264803257666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9866339492060803, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1689772578267079, "3000–6000 Hz_10ms_envelope_correlation": 0.9791043759489314, "3000–6000 Hz_10ms_modulation_mae_db": 1.3498069623649913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1221614152835826, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08945377210621512, "3000–6000 Hz_40ms_envelope_correlation": 0.9937566511121525, "3000–6000 Hz_40ms_modulation_mae_db": 1.6628558649837972, "mr_stft_db": 6.042915463844832, "band_mae_db": 0.28277867903718096, "rms_gain_error_db": -0.12933773907416424, "output_peak": 0.6338079486358299}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1632917549832025, "全帯域_10ms_rms_envelope_nrmse": 0.11098148265500964, "全帯域_10ms_envelope_correlation": 0.9885005243173351, "全帯域_10ms_modulation_mae_db": 0.3575582171625104, "全帯域_40ms_log_rms_mae_db": 0.566483844299372, "全帯域_40ms_rms_envelope_nrmse": 0.05678812845264417, "全帯域_40ms_envelope_correlation": 0.9967043600723816, "全帯域_40ms_modulation_mae_db": 0.47152142976765354, "300–1000 Hz_10ms_log_rms_mae_db": 1.8258049413612063, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15664867980312275, "300–1000 Hz_10ms_envelope_correlation": 0.9827946817574653, "300–1000 Hz_10ms_modulation_mae_db": 0.6003892327879311, "300–1000 Hz_40ms_log_rms_mae_db": 1.0589771711997014, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08232036972190714, "300–1000 Hz_40ms_envelope_correlation": 0.9954652027481998, "300–1000 Hz_40ms_modulation_mae_db": 0.7275264221789173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8974540836306535, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1807963116159641, "1000–3000 Hz_10ms_envelope_correlation": 0.9744757299563153, "1000–3000 Hz_10ms_modulation_mae_db": 0.6103828776578518, "1000–3000 Hz_40ms_log_rms_mae_db": 1.120752970009094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1064120582991542, "1000–3000 Hz_40ms_envelope_correlation": 0.9914807306705504, "1000–3000 Hz_40ms_modulation_mae_db": 0.8285216645242809, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8963637443274068, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15503638663822022, "3000–6000 Hz_10ms_envelope_correlation": 0.9832164600912402, "3000–6000 Hz_10ms_modulation_mae_db": 0.29087542892605395, "3000–6000 Hz_40ms_log_rms_mae_db": 1.083024294213124, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08685114600487111, "3000–6000 Hz_40ms_envelope_correlation": 0.9945269579370872, "3000–6000 Hz_40ms_modulation_mae_db": 0.31707861487438577, "mr_stft_db": 6.038677279555269, "band_mae_db": 0.20369347992229916, "rms_gain_error_db": -0.09019570377404168, "output_peak": 0.5801902368437833}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1691584562943662, "全帯域_10ms_rms_envelope_nrmse": 0.11088080964286239, "全帯域_10ms_envelope_correlation": 0.9884940024054482, "全帯域_10ms_modulation_mae_db": 0.7493173873816223, "全帯域_40ms_log_rms_mae_db": 0.6419130475514729, "全帯域_40ms_rms_envelope_nrmse": 0.05411361964426635, "全帯域_40ms_envelope_correlation": 0.9969002762569855, "全帯域_40ms_modulation_mae_db": 0.8066276542541001, "300–1000 Hz_10ms_log_rms_mae_db": 1.8152424221904047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14255992581588206, "300–1000 Hz_10ms_envelope_correlation": 0.9857754306239408, "300–1000 Hz_10ms_modulation_mae_db": 0.4698284967309368, "300–1000 Hz_40ms_log_rms_mae_db": 1.0507636538816592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07489215947400696, "300–1000 Hz_40ms_envelope_correlation": 0.9957985437491966, "300–1000 Hz_40ms_modulation_mae_db": 0.5573150668341569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8723323238693792, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19078308910862848, "1000–3000 Hz_10ms_envelope_correlation": 0.970783128583531, "1000–3000 Hz_10ms_modulation_mae_db": 0.7080530299708038, "1000–3000 Hz_40ms_log_rms_mae_db": 1.083790130251186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.102294955432069, "1000–3000 Hz_40ms_envelope_correlation": 0.991284808343914, "1000–3000 Hz_40ms_modulation_mae_db": 0.7684443924239467, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9981995178153453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17397442279023487, "3000–6000 Hz_10ms_envelope_correlation": 0.9783184297652192, "3000–6000 Hz_10ms_modulation_mae_db": 0.9567595969649646, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1264237141624331, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09005062696089708, "3000–6000 Hz_40ms_envelope_correlation": 0.9944774595021202, "3000–6000 Hz_40ms_modulation_mae_db": 1.2953720651841, "mr_stft_db": 6.038194877629199, "band_mae_db": 0.28209062223198805, "rms_gain_error_db": -0.02038365986376555, "output_peak": 0.7874090514976657}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2093722373342832, "全帯域_10ms_rms_envelope_nrmse": 0.1123561028305612, "全帯域_10ms_envelope_correlation": 0.9881897791182845, "全帯域_10ms_modulation_mae_db": 0.3905576351990068, "全帯域_40ms_log_rms_mae_db": 0.6286448319337617, "全帯域_40ms_rms_envelope_nrmse": 0.0599415712692108, "全帯域_40ms_envelope_correlation": 0.9962752050061107, "全帯域_40ms_modulation_mae_db": 0.5401424643671842, "300–1000 Hz_10ms_log_rms_mae_db": 1.8703770570383018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18469074808232708, "300–1000 Hz_10ms_envelope_correlation": 0.9762234786668924, "300–1000 Hz_10ms_modulation_mae_db": 0.5566104325198661, "300–1000 Hz_40ms_log_rms_mae_db": 1.150835352256843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11261690679756085, "300–1000 Hz_40ms_envelope_correlation": 0.9915609995294571, "300–1000 Hz_40ms_modulation_mae_db": 0.5843527041115296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.855304347158561, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23498849198488628, "1000–3000 Hz_10ms_envelope_correlation": 0.9607147817259842, "1000–3000 Hz_10ms_modulation_mae_db": 0.43784200061580747, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0847283842098252, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15967749748305707, "1000–3000 Hz_40ms_envelope_correlation": 0.9815942409002846, "1000–3000 Hz_40ms_modulation_mae_db": 0.47357412943466093, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9946784075363553, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17155432440190965, "3000–6000 Hz_10ms_envelope_correlation": 0.9788879092052513, "3000–6000 Hz_10ms_modulation_mae_db": 1.0967121505199695, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1527771175694315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09427686591544757, "3000–6000 Hz_40ms_envelope_correlation": 0.9937309357279038, "3000–6000 Hz_40ms_modulation_mae_db": 1.307018816027945, "mr_stft_db": 6.045107346955213, "band_mae_db": 0.4424711813302506, "rms_gain_error_db": -0.09763619577395106, "output_peak": 0.896065756667075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1628642479465165, "全帯域_10ms_rms_envelope_nrmse": 0.12333907815375762, "全帯域_10ms_envelope_correlation": 0.9860755207893767, "全帯域_10ms_modulation_mae_db": 0.7238412716379662, "全帯域_40ms_log_rms_mae_db": 0.6241908243530994, "全帯域_40ms_rms_envelope_nrmse": 0.07867899877015577, "全帯域_40ms_envelope_correlation": 0.9939690584559361, "全帯域_40ms_modulation_mae_db": 0.6298369254242101, "300–1000 Hz_10ms_log_rms_mae_db": 1.8475061835656887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18623916025333367, "300–1000 Hz_10ms_envelope_correlation": 0.9761855991401206, "300–1000 Hz_10ms_modulation_mae_db": 0.79035570815795, "300–1000 Hz_40ms_log_rms_mae_db": 1.092038322221683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13057577320662278, "300–1000 Hz_40ms_envelope_correlation": 0.9885774232811779, "300–1000 Hz_40ms_modulation_mae_db": 0.975735822654765, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8503113389205001, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.188262148987439, "1000–3000 Hz_10ms_envelope_correlation": 0.9716790521706291, "1000–3000 Hz_10ms_modulation_mae_db": 0.8596877165004702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0980729025429294, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11014218136754111, "1000–3000 Hz_40ms_envelope_correlation": 0.9896190008318803, "1000–3000 Hz_40ms_modulation_mae_db": 0.9076536380790793, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9624861343453373, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15687908355941285, "3000–6000 Hz_10ms_envelope_correlation": 0.9820577187225769, "3000–6000 Hz_10ms_modulation_mae_db": 0.5895233027434913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0974291837413952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08231351651487862, "3000–6000 Hz_40ms_envelope_correlation": 0.994468238014401, "3000–6000 Hz_40ms_modulation_mae_db": 0.5732892731550409, "mr_stft_db": 6.027183264181929, "band_mae_db": 0.30826827312469995, "rms_gain_error_db": -0.20060349876516917, "output_peak": 0.5751918377705286}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1886401440829484, "全帯域_10ms_rms_envelope_nrmse": 0.12479455135637389, "全帯域_10ms_envelope_correlation": 0.9854276628975362, "全帯域_10ms_modulation_mae_db": 0.47054426964215756, "全帯域_40ms_log_rms_mae_db": 0.599565220745379, "全帯域_40ms_rms_envelope_nrmse": 0.06601674538030163, "全帯域_40ms_envelope_correlation": 0.9953663722165048, "全帯域_40ms_modulation_mae_db": 0.6411234514103138, "300–1000 Hz_10ms_log_rms_mae_db": 1.7500330645343058, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1923761228673344, "300–1000 Hz_10ms_envelope_correlation": 0.9735476050446152, "300–1000 Hz_10ms_modulation_mae_db": 0.37916843338065803, "300–1000 Hz_40ms_log_rms_mae_db": 1.004362147894508, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09993197301808386, "300–1000 Hz_40ms_envelope_correlation": 0.9925762776779199, "300–1000 Hz_40ms_modulation_mae_db": 0.6433464335591367, "1000–3000 Hz_10ms_log_rms_mae_db": 1.807771031190175, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18194738858444948, "1000–3000 Hz_10ms_envelope_correlation": 0.9739183535554726, "1000–3000 Hz_10ms_modulation_mae_db": 0.20994797679879595, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0718197981629431, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09770521962970882, "1000–3000 Hz_40ms_envelope_correlation": 0.9917556744350318, "1000–3000 Hz_40ms_modulation_mae_db": 0.3948712371078053, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9991947389408136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18382201101796647, "3000–6000 Hz_10ms_envelope_correlation": 0.9756963441658316, "3000–6000 Hz_10ms_modulation_mae_db": 1.3061356649787839, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1207591807210628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11289789407249698, "3000–6000 Hz_40ms_envelope_correlation": 0.9899250200605859, "3000–6000 Hz_40ms_modulation_mae_db": 1.6176442090734116, "mr_stft_db": 6.040787602352431, "band_mae_db": 0.19821335258582543, "rms_gain_error_db": -0.006947041794542282, "output_peak": 0.5705936443855231}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.4149333148427745, "spectral_convergence": 0.45887871775807443, "spectral_convergence_status": "ok", "amplitude_floor": 1.602383167696498e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7928821109684066, "spectral_convergence": 0.42650693718158084, "spectral_convergence_status": "ok", "amplitude_floor": 1.3018377539140758e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.914272385115982, "spectral_convergence": 0.4550735594012467, "spectral_convergence_status": "ok", "amplitude_floor": 8.15121073207966e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00018979827012669627, "output_power": 0.00020668297894027576, "delta_db": 0.3701246261325948, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007146011617002843, "output_power": 0.0006970822428489026, "delta_db": -0.10779698022867332, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006180173350849692, "output_power": 0.0004989695078207223, "delta_db": -0.92926650442537, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00021309861976674557, "output_power": 0.000193232551960277, "delta_db": -0.42500347309969555, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003708463313924785, "output_power": 0.0003684273560539414, "delta_db": -0.028421176311874703, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001752804045595543, "output_power": 0.0001659888886658615, "delta_db": -0.23654349694960358, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3495260428579687, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.29789730835478156, "pitch": null, "output_rms_dbfs": -26.71542204844096, "output_peak": 0.49564394274605506, "reference_rms_dbfs": -26.41752474008618, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0406959369757205}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002801281523487693, "log_rms_mae_db": 1.2149777306883278, "rms_envelope_nrmse": 0.12691294716721774, "envelope_correlation": 0.9857701705204579, "rms_gain_error_db": -0.29789495541047306, "reference_mean_power": 0.0022835195719053173, "output_mean_power": 0.0021321376675858426}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.817352658591746e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000450774313706133, "output_power": 0.0004193543097067565, "reference_active": true, "delta_db": -0.3137805022556468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021888119814572043, "output_power": 0.00021114292819774522, "reference_active": true, "delta_db": -0.15631917159209774}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.753791744185585e-05, "output_power": 2.309703613746986e-05, "reference_active": true, "delta_db": -2.1091392097899067}], "active_band_count": 3, "active_band_mae_db": 0.8597462945458837}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022259441477048914, "log_rms_mae_db": 0.6602544225615469, "rms_envelope_nrmse": 0.0803653122203358, "envelope_correlation": 0.9944691219054574, "rms_gain_error_db": -0.29796069844357376, "reference_mean_power": 0.002294705859570105, "output_mean_power": 0.0021425499460937182}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.266680044598684e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042660839170957063, "output_power": 0.0003936373066803357, "reference_active": true, "delta_db": -0.34933142454724564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012272337901795565, "output_power": 0.00011831206109175951, "reference_active": true, "delta_db": -0.1589828427883673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.241371364832561e-06, "output_power": 1.952368797080908e-06, "reference_active": true, "delta_db": -2.2016693301447514}], "active_band_count": 3, "active_band_mae_db": 0.903327865826788}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002244694917839089, "log_rms_mae_db": 1.8659807113466869, "rms_envelope_nrmse": 0.18615559108162108, "envelope_correlation": 0.9774965925485424, "rms_gain_error_db": -0.6236877158652293, "reference_mean_power": 0.001082154267593255, "output_mean_power": 0.0009373901914400368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.600154760198564e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003310843525581766, "output_power": 0.00029086995392226054, "reference_active": true, "delta_db": -0.562397938400536}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.047815482126218e-05, "output_power": 7.897274122044279e-05, "reference_active": true, "delta_db": -0.5906652210586796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6938491072881828e-05, "output_power": 1.0743263678280194e-05, "reference_active": true, "delta_db": -1.9773848456846022}], "active_band_count": 3, "active_band_mae_db": 1.0434826683812726}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018235367957614062, "log_rms_mae_db": 1.1097144047574665, "rms_envelope_nrmse": 0.1248442328043865, "envelope_correlation": 0.9914863499274812, "rms_gain_error_db": -0.6236970702863405, "reference_mean_power": 0.0010875823794426604, "output_mean_power": 0.0009420901339964058}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.047574386178106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00032394555601725244, "output_power": 0.0002840297619544375, "reference_active": true, "delta_db": -0.5710817672019862}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.774463594481584e-05, "output_power": 4.173021681947669e-05, "reference_active": true, "delta_db": -0.5847394497261398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5157059923891315e-06, "output_power": 8.156041003300858e-07, "reference_active": true, "delta_db": -2.6913556714781386}], "active_band_count": 3, "active_band_mae_db": 1.2823922961354215}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014040871531833862, "log_rms_mae_db": 1.8434114729815192, "rms_envelope_nrmse": 0.1739628454004172, "envelope_correlation": 0.975665524277458, "rms_gain_error_db": -0.18117241849856577, "reference_mean_power": 0.00036159326722347827, "output_mean_power": 0.0003468191692146779}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2796709554971252e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010222442860363171, "output_power": 0.00010005928784489871, "reference_active": true, "delta_db": -0.09297284148561602}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.551583786781791e-05, "output_power": 4.466094081996383e-05, "reference_active": true, "delta_db": -0.08234674143506021}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1818852015306123e-05, "output_power": 9.541965941216947e-06, "reference_active": true, "delta_db": -0.9293743278156472}], "active_band_count": 3, "active_band_mae_db": 0.3682313035787745}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010845211232047094, "log_rms_mae_db": 1.0759044356572216, "rms_envelope_nrmse": 0.10231051717854255, "envelope_correlation": 0.9908350134853728, "rms_gain_error_db": -0.18116981331350468, "reference_mean_power": 0.00036337479954251316, "output_mean_power": 0.00034852812016364956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.941433739824422e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.608975832570782e-05, "output_power": 9.535611522343367e-05, "reference_active": true, "delta_db": -0.033285513524559426}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4640366344588194e-05, "output_power": 2.37135981344259e-05, "reference_active": true, "delta_db": -0.16649704717973865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.520284367741145e-07, "output_power": 7.08031336549551e-07, "reference_active": true, "delta_db": -1.2859744138437634}], "active_band_count": 3, "active_band_mae_db": 0.49525232484935383}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012084003773539908, "log_rms_mae_db": 1.9558069928978055, "rms_envelope_nrmse": 0.16597080541910147, "envelope_correlation": 0.9798945720151674, "rms_gain_error_db": -0.20369534512269533, "reference_mean_power": 0.00030833628390941923, "output_mean_power": 0.00029420842116833297}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2964446771205807e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7472950448048514e-05, "output_power": 4.96097809053542e-05, "reference_active": true, "delta_db": 0.1912108504226885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.978240331431912e-05, "output_power": 4.675703010799359e-05, "reference_active": true, "delta_db": -0.27228940900691617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3740568823166797e-05, "output_power": 8.853678052463428e-06, "reference_active": true, "delta_db": -1.9088098611849076}], "active_band_count": 3, "active_band_mae_db": 0.7907700402048374}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010527157253957595, "log_rms_mae_db": 1.1166361873142627, "rms_envelope_nrmse": 0.09117475680006205, "envelope_correlation": 0.9935095615290434, "rms_gain_error_db": -0.2036608600938078, "reference_mean_power": 0.0003098592063705908, "output_mean_power": 0.0002956639115463322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.056454468513196e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.721882315895364e-05, "output_power": 5.795544898519256e-05, "reference_active": true, "delta_db": 0.05555353427241804}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8898944201462896e-05, "output_power": 2.5342244446241483e-05, "reference_active": true, "delta_db": -0.5703690074484744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2532553850736713e-06, "output_power": 7.79189629523796e-07, "reference_active": true, "delta_db": -2.06396415646207}], "active_band_count": 3, "active_band_mae_db": 0.8966288993943209}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8051645633137787, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "757795df8ea59edb32ed90f7f257ffa1639d86df28cc35f3d0f9f72caf983b96", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9715707569397005, "standard_deviation": 0.004685069335306886, "minimum": 0.9607147817259842, "maximum": 0.975665524277458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8519475220749948, "standard_deviation": 0.03274096279887501, "minimum": 1.8066293438394987, "maximum": 1.8974540836306535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6754521658559134, "standard_deviation": 0.3580076359457846, "minimum": 0.20994797679879595, "maximum": 1.3515557031758678, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1911805829466852, "standard_deviation": 0.01886734829830187, "minimum": 0.1739628454004172, "maximum": 0.23498849198488628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9897473689132161, "standard_deviation": 0.003522548190471324, "minimum": 0.9815942409002846, "maximum": 0.9926985814461633, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0931381039745172, "standard_deviation": 0.019689591743802345, "minimum": 1.0718197981629431, "maximum": 1.1241881438360917, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7682178823274532, "standard_deviation": 0.3093035757861475, "minimum": 0.3948712371078053, "maximum": 1.3275991918747325, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11362839858839202, "standard_deviation": 0.021151518074814895, "minimum": 0.09770521962970882, "maximum": 0.15967749748305707, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9795566728646113, "standard_deviation": 0.0022968471080577245, "minimum": 0.9756963441658316, "maximum": 0.9832164600912402, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9751024305223477, "standard_deviation": 0.0366608509052127, "minimum": 1.8963637443274068, "maximum": 2.0074559591096373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8649223740011053, "standard_deviation": 0.37969431936651704, "minimum": 0.29087542892605395, "maximum": 1.3498069623649913, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16810175659196955, "standard_deviation": 0.009225295796930447, "minimum": 0.15503638663822022, "maximum": 0.18382201101796647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9936187155715003, "standard_deviation": 0.0015513771300229718, "minimum": 0.9899250200605859, "maximum": 0.9945549006887078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1230727914127696, "standard_deviation": 0.026704264774585, "minimum": 1.083024294213124, "maximum": 1.1653712382968633, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0418109575979482, "standard_deviation": 0.5014300534729664, "minimum": 0.31707861487438577, "maximum": 1.6628558649837972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09151342615011429, "standard_deviation": 0.009404249338650078, "minimum": 0.08231351651487862, "maximum": 0.11289789407249698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9799063346080324, "standard_deviation": 0.004608244643820066, "minimum": 0.9735476050446152, "maximum": 0.9857754306239408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8134457575841725, "standard_deviation": 0.04802124732811834, "minimum": 1.7500330645343058, "maximum": 1.8703770570383018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6486338072869805, "standard_deviation": 0.30990566792074536, "minimum": 0.2389740564047409, "maximum": 1.1102614299324884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16904809324543582, "standard_deviation": 0.020448408323987134, "minimum": 0.14255992581588206, "maximum": 0.1923761228673344, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9928290417633117, "standard_deviation": 0.0023534122161422886, "minimum": 0.9885774232811779, "maximum": 0.9957985437491966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0747423536509801, "standard_deviation": 0.04773436316860414, "minimum": 1.004362147894508, "maximum": 1.150835352256843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7674902233180626, "standard_deviation": 0.32061842980000455, "minimum": 0.2956352939644756, "maximum": 1.2823922961354215, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10131841176442505, "standard_deviation": 0.01990520193013517, "minimum": 0.07489215947400696, "maximum": 0.13057577320662278, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.29850527027119167, "standard_deviation": 0.07882636111844006, "minimum": 0.19821335258582543, "maximum": 0.4424711813302506, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.038433529113781, "standard_deviation": 0.005637462770859908, "minimum": 6.027183264181929, "maximum": 6.045107346955213, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6506707955969514, "standard_deviation": 0.1312359240058123, "minimum": 0.49564394274605506, "maximum": 0.896065756667075, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1141426099807448, "standard_deviation": 0.09601783036507137, "minimum": -0.29789730835478156, "maximum": -0.006947041794542282, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9873713142175211, "standard_deviation": 0.0013529778806067993, "minimum": 0.9854276628975362, "maximum": 0.9885005243173351, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1850992199022496, "standard_deviation": 0.019862235525993702, "minimum": 1.1628642479465165, "maximum": 1.2149777306883278, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.616561799260905, "standard_deviation": 0.19069697697730204, "minimum": 0.3575582171625104, "maximum": 0.8597462945458837, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1166918825535144, "standard_deviation": 0.00698207380600483, "minimum": 0.11088080964286239, "maximum": 0.12691294716721774, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9956264553921595, "standard_deviation": 0.0010302561713959977, "minimum": 0.9939690584559361, "maximum": 0.9969002762569855, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.618951966713196, "standard_deviation": 0.028151516283060722, "minimum": 0.566483844299372, "maximum": 0.6602544225615469, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6558095325522074, "standard_deviation": 0.14389035676323791, "minimum": 0.47152142976765354, "maximum": 0.903327865826788, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06566435696298989, "standard_deviation": 0.009517785189802772, "minimum": 0.05411361964426635, "maximum": 0.0803653122203358, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0002", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.3910162448883057, "predicted_mos": 2.3910162448883057, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9e243c137c898aed8bc7ae4747910c7b7a700dd2de19722122180f402aab1f4c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7939561000093818, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2229401578361991, "全帯域_10ms_rms_envelope_nrmse": 0.1194411734011058, "全帯域_10ms_envelope_correlation": 0.9874409449513187, "全帯域_10ms_modulation_mae_db": 0.7346432009320502, "全帯域_40ms_log_rms_mae_db": 0.6517345909752644, "全帯域_40ms_rms_envelope_nrmse": 0.07456280250215969, "全帯域_40ms_envelope_correlation": 0.9952869003944983, "全帯域_40ms_modulation_mae_db": 0.7866527983577537, "300–1000 Hz_10ms_log_rms_mae_db": 1.8589959334044677, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1638857180342968, "300–1000 Hz_10ms_envelope_correlation": 0.9826718994060892, "300–1000 Hz_10ms_modulation_mae_db": 1.0106319992159871, "300–1000 Hz_40ms_log_rms_mae_db": 1.1053819996762648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10865664847565734, "300–1000 Hz_40ms_envelope_correlation": 0.9935647502584035, "300–1000 Hz_40ms_modulation_mae_db": 1.154509590524537, "1000–3000 Hz_10ms_log_rms_mae_db": 1.894614161548354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1743293012018293, "1000–3000 Hz_10ms_envelope_correlation": 0.9756154729436551, "1000–3000 Hz_10ms_modulation_mae_db": 0.46867440273189437, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0688373899722767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10751620910071645, "1000–3000 Hz_40ms_envelope_correlation": 0.9899606217088296, "1000–3000 Hz_40ms_modulation_mae_db": 0.5596377858324022, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0874462391908035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17565716671567796, "3000–6000 Hz_10ms_envelope_correlation": 0.977361021304946, "3000–6000 Hz_10ms_modulation_mae_db": 0.8808091436947777, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1728055444511887, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09616623207875592, "3000–6000 Hz_40ms_envelope_correlation": 0.9925067786554703, "3000–6000 Hz_40ms_modulation_mae_db": 1.0284685057013916, "mr_stft_db": 6.047616174188416, "band_mae_db": 0.3022946738272633, "rms_gain_error_db": -0.27910337812729047, "output_peak": 0.5001176627590027}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1860084247849523, "全帯域_10ms_rms_envelope_nrmse": 0.12260113995522388, "全帯域_10ms_envelope_correlation": 0.9859439228080152, "全帯域_10ms_modulation_mae_db": 0.7246087464205089, "全帯域_40ms_log_rms_mae_db": 0.6225082280032466, "全帯域_40ms_rms_envelope_nrmse": 0.06964011993286949, "全帯域_40ms_envelope_correlation": 0.9950016068214733, "全帯域_40ms_modulation_mae_db": 0.7071921518109329, "300–1000 Hz_10ms_log_rms_mae_db": 1.74242549815611, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15297301286053297, "300–1000 Hz_10ms_envelope_correlation": 0.983382620757058, "300–1000 Hz_10ms_modulation_mae_db": 0.9646084980201355, "300–1000 Hz_40ms_log_rms_mae_db": 0.9840486316084238, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08943794444456342, "300–1000 Hz_40ms_envelope_correlation": 0.9940228203311681, "300–1000 Hz_40ms_modulation_mae_db": 1.0514406526480924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9421373575380643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2012153422228958, "1000–3000 Hz_10ms_envelope_correlation": 0.970166245976332, "1000–3000 Hz_10ms_modulation_mae_db": 1.4905395186822499, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1130643007101901, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1326493151190357, "1000–3000 Hz_40ms_envelope_correlation": 0.9888328054209574, "1000–3000 Hz_40ms_modulation_mae_db": 1.490728161212374, "3000–6000 Hz_10ms_log_rms_mae_db": 2.112037427907023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17652518484747537, "3000–6000 Hz_10ms_envelope_correlation": 0.977171887500249, "3000–6000 Hz_10ms_modulation_mae_db": 0.6286448716207601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1797499769429276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09036701830804264, "3000–6000 Hz_40ms_envelope_correlation": 0.993642701172596, "3000–6000 Hz_40ms_modulation_mae_db": 0.7914328674386525, "mr_stft_db": 6.0375815178910806, "band_mae_db": 0.3188784732548115, "rms_gain_error_db": -0.1282725404456337, "output_peak": 0.601870310076845}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1891667601865252, "全帯域_10ms_rms_envelope_nrmse": 0.11225335046431091, "全帯域_10ms_envelope_correlation": 0.9882821404084489, "全帯域_10ms_modulation_mae_db": 0.6342174917930404, "全帯域_40ms_log_rms_mae_db": 0.6297677722164648, "全帯域_40ms_rms_envelope_nrmse": 0.06359565002424383, "全帯域_40ms_envelope_correlation": 0.9959113945377834, "全帯域_40ms_modulation_mae_db": 0.6272051098402317, "300–1000 Hz_10ms_log_rms_mae_db": 1.7672858305172148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15393418300802084, "300–1000 Hz_10ms_envelope_correlation": 0.983176967893001, "300–1000 Hz_10ms_modulation_mae_db": 0.3475562045494927, "300–1000 Hz_40ms_log_rms_mae_db": 1.0697111439743487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09364459647884807, "300–1000 Hz_40ms_envelope_correlation": 0.9934302406767559, "300–1000 Hz_40ms_modulation_mae_db": 0.4374823523760731, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8684863191417664, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17918202647320108, "1000–3000 Hz_10ms_envelope_correlation": 0.9758509152650816, "1000–3000 Hz_10ms_modulation_mae_db": 1.0669296877114791, "1000–3000 Hz_40ms_log_rms_mae_db": 1.073668463874191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10165578127726736, "1000–3000 Hz_40ms_envelope_correlation": 0.993703260798964, "1000–3000 Hz_40ms_modulation_mae_db": 1.19281193476491, "3000–6000 Hz_10ms_log_rms_mae_db": 2.093823740494994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16959510497403332, "3000–6000 Hz_10ms_envelope_correlation": 0.9789857750834183, "3000–6000 Hz_10ms_modulation_mae_db": 1.4204302152113062, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1584681596974675, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0877834084437057, "3000–6000 Hz_40ms_envelope_correlation": 0.9940720100379954, "3000–6000 Hz_40ms_modulation_mae_db": 1.8030806965141222, "mr_stft_db": 6.053563474679168, "band_mae_db": 0.287737038535079, "rms_gain_error_db": -0.1408185304528456, "output_peak": 0.6816537492077002}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1815484958469782, "全帯域_10ms_rms_envelope_nrmse": 0.11389401141182225, "全帯域_10ms_envelope_correlation": 0.9878614794124906, "全帯域_10ms_modulation_mae_db": 0.4240266876787551, "全帯域_40ms_log_rms_mae_db": 0.5852334650247393, "全帯域_40ms_rms_envelope_nrmse": 0.05971778990073179, "全帯域_40ms_envelope_correlation": 0.9963007342529001, "全帯域_40ms_modulation_mae_db": 0.535897135668752, "300–1000 Hz_10ms_log_rms_mae_db": 1.7864578146080763, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15538700191149954, "300–1000 Hz_10ms_envelope_correlation": 0.982893886675654, "300–1000 Hz_10ms_modulation_mae_db": 0.6838430409862708, "300–1000 Hz_40ms_log_rms_mae_db": 0.99699093495105, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08317671521127495, "300–1000 Hz_40ms_envelope_correlation": 0.9950049827633681, "300–1000 Hz_40ms_modulation_mae_db": 0.7626066195533584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.936877704423256, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18543629586601176, "1000–3000 Hz_10ms_envelope_correlation": 0.9732376870405642, "1000–3000 Hz_10ms_modulation_mae_db": 0.590576829863608, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1056384809751976, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11672537752232663, "1000–3000 Hz_40ms_envelope_correlation": 0.9895963469047345, "1000–3000 Hz_40ms_modulation_mae_db": 0.8195453366401928, "3000–6000 Hz_10ms_log_rms_mae_db": 2.023257503269364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15784233260734376, "3000–6000 Hz_10ms_envelope_correlation": 0.9825222212257869, "3000–6000 Hz_10ms_modulation_mae_db": 0.5118081990222824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1701247960639571, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08815046616320706, "3000–6000 Hz_40ms_envelope_correlation": 0.9942780409313693, "3000–6000 Hz_40ms_modulation_mae_db": 0.6219314329440194, "mr_stft_db": 6.056281798443014, "band_mae_db": 0.19773387136535148, "rms_gain_error_db": -0.07793056925714939, "output_peak": 0.5431462214921661}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1810519927630627, "全帯域_10ms_rms_envelope_nrmse": 0.1161446446377736, "全帯域_10ms_envelope_correlation": 0.987424356721475, "全帯域_10ms_modulation_mae_db": 0.7784296950968533, "全帯域_40ms_log_rms_mae_db": 0.6739443032983125, "全帯域_40ms_rms_envelope_nrmse": 0.06074763353757128, "全帯域_40ms_envelope_correlation": 0.9961307204596191, "全帯域_40ms_modulation_mae_db": 0.8326366134401201, "300–1000 Hz_10ms_log_rms_mae_db": 1.8142575090212716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15436937199738893, "300–1000 Hz_10ms_envelope_correlation": 0.9836683920628146, "300–1000 Hz_10ms_modulation_mae_db": 0.561037498330667, "300–1000 Hz_40ms_log_rms_mae_db": 1.0731107384444643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08725236605372284, "300–1000 Hz_40ms_envelope_correlation": 0.9945568664637648, "300–1000 Hz_40ms_modulation_mae_db": 0.6443615798715451, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9284110776276162, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18831791020079988, "1000–3000 Hz_10ms_envelope_correlation": 0.9716445910109602, "1000–3000 Hz_10ms_modulation_mae_db": 0.6925542303388005, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0834164474708914, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10347637791362134, "1000–3000 Hz_40ms_envelope_correlation": 0.9912302495069732, "1000–3000 Hz_40ms_modulation_mae_db": 0.7960784720747505, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0898767869767716, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17777799443868914, "3000–6000 Hz_10ms_envelope_correlation": 0.9772578207575074, "3000–6000 Hz_10ms_modulation_mae_db": 1.059053516816806, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1692049031444371, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09254574656539134, "3000–6000 Hz_40ms_envelope_correlation": 0.9940104929855842, "3000–6000 Hz_40ms_modulation_mae_db": 1.4282929959005815, "mr_stft_db": 6.036314257148347, "band_mae_db": 0.2910842937558166, "rms_gain_error_db": 0.0033803979053294735, "output_peak": 0.7417569345561927}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2114116684927876, "全帯域_10ms_rms_envelope_nrmse": 0.11223698511388182, "全帯域_10ms_envelope_correlation": 0.9882083376627103, "全帯域_10ms_modulation_mae_db": 0.39618752676315677, "全帯域_40ms_log_rms_mae_db": 0.6300993001403086, "全帯域_40ms_rms_envelope_nrmse": 0.058519829666943704, "全帯域_40ms_envelope_correlation": 0.9964357321321576, "全帯域_40ms_modulation_mae_db": 0.5159162357152199, "300–1000 Hz_10ms_log_rms_mae_db": 1.8625282274979804, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17971025541073005, "300–1000 Hz_10ms_envelope_correlation": 0.9772850332536644, "300–1000 Hz_10ms_modulation_mae_db": 0.38246071134333154, "300–1000 Hz_40ms_log_rms_mae_db": 1.1197764791345826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10511009889677782, "300–1000 Hz_40ms_envelope_correlation": 0.9923982163385792, "300–1000 Hz_40ms_modulation_mae_db": 0.6616498371624656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.897712941186159, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22051737433308655, "1000–3000 Hz_10ms_envelope_correlation": 0.9646829929346258, "1000–3000 Hz_10ms_modulation_mae_db": 0.35566793831813515, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0676473882380755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14526317402049477, "1000–3000 Hz_40ms_envelope_correlation": 0.9843900669877471, "1000–3000 Hz_40ms_modulation_mae_db": 0.36348305136648057, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0961746841051476, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18312454813687282, "3000–6000 Hz_10ms_envelope_correlation": 0.9756286081659368, "3000–6000 Hz_10ms_modulation_mae_db": 1.1356652200958226, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1966294268015525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09823271826695237, "3000–6000 Hz_40ms_envelope_correlation": 0.9928687292751938, "3000–6000 Hz_40ms_modulation_mae_db": 1.3953105185715795, "mr_stft_db": 6.057365984361357, "band_mae_db": 0.37271811205243943, "rms_gain_error_db": -0.08747739172494011, "output_peak": 0.8354527782436852}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1800474580837175, "全帯域_10ms_rms_envelope_nrmse": 0.11578390506355903, "全帯域_10ms_envelope_correlation": 0.9876292027981491, "全帯域_10ms_modulation_mae_db": 0.6808614934377935, "全帯域_40ms_log_rms_mae_db": 0.6315987410004058, "全帯域_40ms_rms_envelope_nrmse": 0.07128398487922817, "全帯域_40ms_envelope_correlation": 0.9949348307653358, "全帯域_40ms_modulation_mae_db": 0.5826362161804542, "300–1000 Hz_10ms_log_rms_mae_db": 1.856037594796503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17745362950381668, "300–1000 Hz_10ms_envelope_correlation": 0.9781043717247931, "300–1000 Hz_10ms_modulation_mae_db": 0.7629701375238677, "300–1000 Hz_40ms_log_rms_mae_db": 1.0527860611019313, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12138405063972411, "300–1000 Hz_40ms_envelope_correlation": 0.9897874870861303, "300–1000 Hz_40ms_modulation_mae_db": 0.9409917399543044, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9305880333481507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19165701735891552, "1000–3000 Hz_10ms_envelope_correlation": 0.9708755476323381, "1000–3000 Hz_10ms_modulation_mae_db": 0.7957929638431182, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1110185203797311, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11036502865784016, "1000–3000 Hz_40ms_envelope_correlation": 0.989752235175295, "1000–3000 Hz_40ms_modulation_mae_db": 0.9217748376462538, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0895313030375298, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16232327573131786, "3000–6000 Hz_10ms_envelope_correlation": 0.9809430715801724, "3000–6000 Hz_10ms_modulation_mae_db": 0.8239480264490887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1634456767157235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08424405735905324, "3000–6000 Hz_40ms_envelope_correlation": 0.9942715447014087, "3000–6000 Hz_40ms_modulation_mae_db": 0.7939291381944061, "mr_stft_db": 6.033279097792235, "band_mae_db": 0.22374866908499602, "rms_gain_error_db": -0.15988053561544244, "output_peak": 0.5741649679610984}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1866925652378457, "全帯域_10ms_rms_envelope_nrmse": 0.12369906358868814, "全帯域_10ms_envelope_correlation": 0.9857317805174551, "全帯域_10ms_modulation_mae_db": 0.49450232218702306, "全帯域_40ms_log_rms_mae_db": 0.5986820052039651, "全帯域_40ms_rms_envelope_nrmse": 0.0639381008491, "全帯域_40ms_envelope_correlation": 0.9956706353047964, "全帯域_40ms_modulation_mae_db": 0.6310856288479885, "300–1000 Hz_10ms_log_rms_mae_db": 1.7474303241972673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18488372837160635, "300–1000 Hz_10ms_envelope_correlation": 0.9756035526898613, "300–1000 Hz_10ms_modulation_mae_db": 0.21335251414346765, "300–1000 Hz_40ms_log_rms_mae_db": 1.0059173276164588, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0962359175716846, "300–1000 Hz_40ms_envelope_correlation": 0.9929986254525308, "300–1000 Hz_40ms_modulation_mae_db": 0.7056019318686818, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8716322234028906, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18743842008279843, "1000–3000 Hz_10ms_envelope_correlation": 0.9728937870877051, "1000–3000 Hz_10ms_modulation_mae_db": 0.2532959660435203, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0813644262325104, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1022583901943209, "1000–3000 Hz_40ms_envelope_correlation": 0.9914256674443045, "1000–3000 Hz_40ms_modulation_mae_db": 0.3953834137400409, "3000–6000 Hz_10ms_log_rms_mae_db": 2.102595751693274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19088762905734857, "3000–6000 Hz_10ms_envelope_correlation": 0.973706536147797, "3000–6000 Hz_10ms_modulation_mae_db": 1.3786150063742435, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1716925492191785, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11149039754612648, "3000–6000 Hz_40ms_envelope_correlation": 0.9901074321921088, "3000–6000 Hz_40ms_modulation_mae_db": 1.646005707691223, "mr_stft_db": 6.051074382540393, "band_mae_db": 0.17092772490572253, "rms_gain_error_db": 0.01844168727463201, "output_peak": 0.5596768572417506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.446892611613976, "spectral_convergence": 0.4565049406727822, "spectral_convergence_status": "ok", "amplitude_floor": 1.602383167696498e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.81011914109805, "spectral_convergence": 0.421692603404277, "spectral_convergence_status": "ok", "amplitude_floor": 1.3018377539140758e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.885836769853222, "spectral_convergence": 0.45149549588639715, "spectral_convergence_status": "ok", "amplitude_floor": 8.15121073207966e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00018979827012669627, "output_power": 0.00020051984722880368, "delta_db": 0.23865115232019407, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007146011617002843, "output_power": 0.0006947258713104843, "delta_db": -0.12250245756864839, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006180173350849692, "output_power": 0.0005166348065081506, "delta_db": -0.7781699507561256, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00021309861976674557, "output_power": 0.00019016262619399587, "delta_db": -0.4945547033296144, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003708463313924785, "output_power": 0.0003682045287116227, "delta_db": -0.03104861316708599, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001752804045595543, "output_power": 0.00016937497743281583, "delta_db": -0.14884116582191162, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3022946738272633, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.27910337812729047, "pitch": null, "output_rms_dbfs": -26.69662811821347, "output_peak": 0.5001176627590027, "reference_rms_dbfs": -26.41752474008618, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.047616174188416}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002801281523487693, "log_rms_mae_db": 1.2229401578361991, "rms_envelope_nrmse": 0.1194411734011058, "envelope_correlation": 0.9874409449513187, "rms_gain_error_db": -0.2790966225588202, "reference_mean_power": 0.0022835195719053173, "output_mean_power": 0.002141386576922441}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.817352658591746e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000450774313706133, "output_power": 0.00042362386658573, "reference_active": true, "delta_db": -0.26978740965702114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021888119814572043, "output_power": 0.000210563908297114, "reference_active": true, "delta_db": -0.16824524364112245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.753791744185585e-05, "output_power": 2.499657348168695e-05, "reference_active": true, "delta_db": -1.7658969494980068}], "active_band_count": 3, "active_band_mae_db": 0.7346432009320502}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022259441477048914, "log_rms_mae_db": 0.6517345909752644, "rms_envelope_nrmse": 0.07456280250215969, "envelope_correlation": 0.9952869003944983, "rms_gain_error_db": -0.2791618339075783, "reference_mean_power": 0.002294705859570105, "output_mean_power": 0.002151844285849745}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.266680044598684e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042660839170957063, "output_power": 0.00039889199011467803, "reference_active": true, "delta_db": -0.2917407848007654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012272337901795565, "output_power": 0.00011803975123433876, "reference_active": true, "delta_db": -0.16899018900754545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.241371364832561e-06, "output_power": 2.0931774577148844e-06, "reference_active": true, "delta_db": -1.8992274212649503}], "active_band_count": 3, "active_band_mae_db": 0.7866527983577537}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002244694917839089, "log_rms_mae_db": 1.8589959334044677, "rms_envelope_nrmse": 0.1638857180342968, "envelope_correlation": 0.9826718994060892, "rms_gain_error_db": -0.5420515708561603, "reference_mean_power": 0.001082154267593255, "output_mean_power": 0.0009551773584656278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.600154760198564e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003310843525581766, "output_power": 0.00029992558563624337, "reference_active": true, "delta_db": -0.42925140429614433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.047815482126218e-05, "output_power": 7.992567845546346e-05, "reference_active": true, "delta_db": -0.538574036607107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6938491072881828e-05, "output_power": 1.0530952743010811e-05, "reference_active": true, "delta_db": -2.06407055674471}], "active_band_count": 3, "active_band_mae_db": 1.0106319992159871}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018235367957614062, "log_rms_mae_db": 1.1053819996762648, "rms_envelope_nrmse": 0.10865664847565734, "envelope_correlation": 0.9935647502584035, "rms_gain_error_db": -0.5420649000078754, "reference_mean_power": 0.0010875823794426604, "output_mean_power": 0.0009599656047994817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.047574386178106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00032394555601725244, "output_power": 0.0002929806748110151, "reference_active": true, "delta_db": -0.43633051577107196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.774463594481584e-05, "output_power": 4.224253160952163e-05, "reference_active": true, "delta_db": -0.531746484485155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5157059923891315e-06, "output_power": 8.532372169572646e-07, "reference_active": true, "delta_db": -2.4954517713173843}], "active_band_count": 3, "active_band_mae_db": 1.154509590524537}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014040871531833862, "log_rms_mae_db": 1.894614161548354, "rms_envelope_nrmse": 0.1743293012018293, "envelope_correlation": 0.9756154729436551, "rms_gain_error_db": -0.21819550408321392, "reference_mean_power": 0.00036159326722347827, "output_mean_power": 0.0003438751438050506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2796709554971252e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010222442860363171, "output_power": 9.885049235758076e-05, "reference_active": true, "delta_db": -0.14575854886815234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.551583786781791e-05, "output_power": 4.209397622479648e-05, "reference_active": true, "delta_db": -0.33942590321422156}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1818852015306123e-05, "output_power": 9.560738047922905e-06, "reference_active": true, "delta_db": -0.9208387561133093}], "active_band_count": 3, "active_band_mae_db": 0.46867440273189437}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010845211232047094, "log_rms_mae_db": 1.0688373899722767, "rms_envelope_nrmse": 0.10751620910071645, "envelope_correlation": 0.9899606217088296, "rms_gain_error_db": -0.218213330631173, "reference_mean_power": 0.00036337479954251316, "output_mean_power": 0.00034556796231704053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.941433739824422e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.608975832570782e-05, "output_power": 9.36054444732063e-05, "reference_active": true, "delta_db": -0.11375991290132385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4640366344588194e-05, "output_power": 2.2342214936191895e-05, "reference_active": true, "delta_db": -0.42520934994313164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.520284367741145e-07, "output_power": 7.322434766181411e-07, "reference_active": true, "delta_db": -1.1399440946527513}], "active_band_count": 3, "active_band_mae_db": 0.5596377858324022}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012084003773539908, "log_rms_mae_db": 2.0874462391908035, "rms_envelope_nrmse": 0.17565716671567796, "envelope_correlation": 0.977361021304946, "rms_gain_error_db": -0.12301131547713307, "reference_mean_power": 0.00030833628390941923, "output_mean_power": 0.00029972536821599804}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2964446771205807e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7472950448048514e-05, "output_power": 5.191653613888799e-05, "reference_active": true, "delta_db": 0.38859484718798215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.978240331431912e-05, "output_power": 4.821876963089249e-05, "reference_active": true, "delta_db": -0.13859734391132303}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3740568823166797e-05, "output_power": 8.442696700292824e-06, "reference_active": true, "delta_db": -2.115235239985028}], "active_band_count": 3, "active_band_mae_db": 0.8808091436947777}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010527157253957595, "log_rms_mae_db": 1.1728055444511887, "rms_envelope_nrmse": 0.09616623207875592, "envelope_correlation": 0.9925067786554703, "rms_gain_error_db": -0.12295712486930581, "reference_mean_power": 0.0003098592063705908, "output_mean_power": 0.0003012095184039494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.056454468513196e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.721882315895364e-05, "output_power": 6.0623916278344e-05, "reference_active": true, "delta_db": 0.25105066907230233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8898944201462896e-05, "output_power": 2.5866369650144742e-05, "reference_active": true, "delta_db": -0.48146496782839854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2532553850736713e-06, "output_power": 7.290387143188511e-07, "reference_active": true, "delta_db": -2.352889880203474}], "active_band_count": 3, "active_band_mae_db": 1.0284685057013916}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8024883280334996, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c84102eabb553a0a5402f39efe4da8650f38d2dd0bb4dad4ab1bc7a4c988916b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9718709049864078, "standard_deviation": 0.0035534801505823707, "minimum": 0.9646829929346258, "maximum": 0.9758509152650816, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.908807477277032, "standard_deviation": 0.029495390865473734, "minimum": 1.8684863191417664, "maximum": 1.9421373575380643, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7142539421916007, "standard_deviation": 0.4054610221808028, "minimum": 0.2532959660435203, "maximum": 1.4905395186822499, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1910117109674423, "standard_deviation": 0.01436498290225166, "minimum": 0.1743293012018293, "maximum": 0.22051737433308655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9898614067434757, "standard_deviation": 0.0026770474886943737, "minimum": 0.9843900669877471, "maximum": 0.993703260798964, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0880819272316329, "standard_deviation": 0.018977433645733677, "minimum": 1.0676473882380755, "maximum": 1.1130643007101901, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8174303741596756, "standard_deviation": 0.38844712771472867, "minimum": 0.36348305136648057, "maximum": 1.490728161212374, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11498870672570291, "standard_deviation": 0.015950972556881868, "minimum": 0.10165578127726736, "maximum": 0.14526317402049477, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9779471177207267, "standard_deviation": 0.0028224548728816717, "minimum": 0.973706536147797, "maximum": 0.9825222212257869, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0868429295843636, "standard_deviation": 0.02693545082540383, "minimum": 2.023257503269364, "maximum": 2.112037427907023, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9798717749106359, "standard_deviation": 0.3295448633403049, "minimum": 0.5118081990222824, "maximum": 1.4204302152113062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17421665456359486, "standard_deviation": 0.01074569074895913, "minimum": 0.15784233260734376, "maximum": 0.19088762905734857, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932197162439658, "standard_deviation": 0.001417987857136669, "minimum": 0.9901074321921088, "maximum": 0.9942780409313693, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.172765129129554, "standard_deviation": 0.011528517185125622, "minimum": 1.1584681596974675, "maximum": 1.1966294268015525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.188556482869497, "standard_deviation": 0.4387076221520186, "minimum": 0.6219314329440194, "maximum": 1.8030806965141222, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09362250559140435, "standard_deviation": 0.00854180758693038, "minimum": 0.08424405735905324, "maximum": 0.11149039754612648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.980848340557867, "standard_deviation": 0.003274178661650864, "minimum": 0.9756035526898613, "maximum": 0.9836683920628146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8044273415248613, "standard_deviation": 0.050580881353198395, "minimum": 1.74242549815611, "maximum": 1.8625282274979804, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6158075755141525, "standard_deviation": 0.29159988441407747, "minimum": 0.21335251414346765, "maximum": 1.0106319992159871, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16532461263723652, "standard_deviation": 0.013308249336088093, "minimum": 0.15297301286053297, "maximum": 0.18488372837160635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9932204986713375, "standard_deviation": 0.0016163197953274783, "minimum": 0.9897874870861303, "maximum": 0.9950049827633681, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0509654145634406, "standard_deviation": 0.050655263815571006, "minimum": 0.9840486316084238, "maximum": 1.1197764791345826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7948305379948821, "standard_deviation": 0.23714273158287982, "minimum": 0.4374823523760731, "maximum": 1.154509590524537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09811229222153164, "standard_deviation": 0.012769731736660347, "minimum": 0.08317671521127495, "maximum": 0.12138405063972411, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27064035709768497, "standard_deviation": 0.06752335661725728, "minimum": 0.17092772490572253, "maximum": 0.37271811205243943, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.0466345858805015, "standard_deviation": 0.009589821395547876, "minimum": 6.033279097792235, "maximum": 6.057365984361357, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6297299351923051, "standard_deviation": 0.11378875030403762, "minimum": 0.5001176627590027, "maximum": 0.8354527782436852, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10645760755541753, "standard_deviation": 0.09498526010305201, "minimum": -0.27910337812729047, "maximum": 0.01844168727463201, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9873152706600079, "standard_deviation": 0.0009669564127746653, "minimum": 0.9857317805174551, "maximum": 0.9882821404084489, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1923584404040084, "standard_deviation": 0.01593347576353859, "minimum": 1.1800474580837175, "maximum": 1.2229401578361991, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6084346455386477, "standard_deviation": 0.14941287483103954, "minimum": 0.39618752676315677, "maximum": 0.7784296950968533, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11700678420454569, "standard_deviation": 0.004460683140413211, "minimum": 0.11223698511388182, "maximum": 0.12369906358868814, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9957090693335705, "standard_deviation": 0.0005826642888152134, "minimum": 0.9949348307653358, "maximum": 0.9964357321321576, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6279460507328383, "standard_deviation": 0.027808915422767033, "minimum": 0.5852334650247393, "maximum": 0.6739443032983125, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6524027362326816, "standard_deviation": 0.1144352819486973, "minimum": 0.5159162357152199, "maximum": 0.8326366134401201, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06525073891160599, "standard_deviation": 0.005892370501696686, "minimum": 0.058519829666943704, "maximum": 0.07456280250215969, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0002", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.8296236991882324, "predicted_mos": 1.8296236991882324, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0317f27032a79f73b8ea5612550336382b54bf8998c280a769bb9ee33e0b36cf", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7589022000320256, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.131843456463189, "全帯域_10ms_rms_envelope_nrmse": 0.11301748064195555, "全帯域_10ms_envelope_correlation": 0.9880991408298391, "全帯域_10ms_modulation_mae_db": 0.8046705317604738, "全帯域_40ms_log_rms_mae_db": 0.5831172210299612, "全帯域_40ms_rms_envelope_nrmse": 0.0634570362872321, "全帯域_40ms_envelope_correlation": 0.9958228646082682, "全帯域_40ms_modulation_mae_db": 0.7671856347304878, "300–1000 Hz_10ms_log_rms_mae_db": 2.1358944856013866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15024353959170567, "300–1000 Hz_10ms_envelope_correlation": 0.9840888612355891, "300–1000 Hz_10ms_modulation_mae_db": 0.8373047050908077, "300–1000 Hz_40ms_log_rms_mae_db": 1.3025257021097796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08156332671700695, "300–1000 Hz_40ms_envelope_correlation": 0.9951369815086287, "300–1000 Hz_40ms_modulation_mae_db": 0.8973615145839274, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1785341294755445, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19382711708640502, "1000–3000 Hz_10ms_envelope_correlation": 0.971592420697943, "1000–3000 Hz_10ms_modulation_mae_db": 1.2159424958015208, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2753258682093458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12549488714275642, "1000–3000 Hz_40ms_envelope_correlation": 0.9890232613787485, "1000–3000 Hz_40ms_modulation_mae_db": 1.2380577119892993, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4660142578499302, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19756014676567799, "3000–6000 Hz_10ms_envelope_correlation": 0.9731422339850966, "3000–6000 Hz_10ms_modulation_mae_db": 1.089553237224326, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3562973572001216, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11102791170309917, "3000–6000 Hz_40ms_envelope_correlation": 0.9914173667242481, "3000–6000 Hz_40ms_modulation_mae_db": 1.401653126654818, "mr_stft_db": 6.3484035808071555, "band_mae_db": 0.2883873591596074, "rms_gain_error_db": -0.0731467692890856, "output_peak": 0.7127733806040093}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1517707624663207, "全帯域_10ms_rms_envelope_nrmse": 0.11630652947124012, "全帯域_10ms_envelope_correlation": 0.9873282252062483, "全帯域_10ms_modulation_mae_db": 0.45182886607444733, "全帯域_40ms_log_rms_mae_db": 0.5834134467622749, "全帯域_40ms_rms_envelope_nrmse": 0.05683458018006382, "全帯域_40ms_envelope_correlation": 0.996650583555301, "全帯域_40ms_modulation_mae_db": 0.655429694191569, "300–1000 Hz_10ms_log_rms_mae_db": 2.1092763426175405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16818909659277076, "300–1000 Hz_10ms_envelope_correlation": 0.979843979600127, "300–1000 Hz_10ms_modulation_mae_db": 0.2908412811727104, "300–1000 Hz_40ms_log_rms_mae_db": 1.2800300059365428, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08396759671441373, "300–1000 Hz_40ms_envelope_correlation": 0.9947025385459113, "300–1000 Hz_40ms_modulation_mae_db": 0.38697935155731117, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2602723336114807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2094348975818637, "1000–3000 Hz_10ms_envelope_correlation": 0.9656311652615635, "1000–3000 Hz_10ms_modulation_mae_db": 1.2959896083913456, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3052560576663537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12454951051905712, "1000–3000 Hz_40ms_envelope_correlation": 0.988465949787939, "1000–3000 Hz_40ms_modulation_mae_db": 1.5459392430566041, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3585703232598862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18503477823075382, "3000–6000 Hz_10ms_envelope_correlation": 0.9749637725038934, "3000–6000 Hz_10ms_modulation_mae_db": 1.0936513435965391, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3173278463725757, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10600526704457491, "3000–6000 Hz_40ms_envelope_correlation": 0.9912481023769978, "3000–6000 Hz_40ms_modulation_mae_db": 1.2656551320816467, "mr_stft_db": 6.35065867802102, "band_mae_db": 0.29449544554495377, "rms_gain_error_db": -0.08719174447969952, "output_peak": 0.6117152171396587}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.148247439601518, "全帯域_10ms_rms_envelope_nrmse": 0.13688654994832392, "全帯域_10ms_envelope_correlation": 0.9833179771184892, "全帯域_10ms_modulation_mae_db": 0.8780500852439846, "全帯域_40ms_log_rms_mae_db": 0.6172489426759166, "全帯域_40ms_rms_envelope_nrmse": 0.08636387210298296, "全帯域_40ms_envelope_correlation": 0.9935572161918883, "全帯域_40ms_modulation_mae_db": 0.9367133065891631, "300–1000 Hz_10ms_log_rms_mae_db": 2.12514100757966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19721097964768294, "300–1000 Hz_10ms_envelope_correlation": 0.9740607433328726, "300–1000 Hz_10ms_modulation_mae_db": 0.9400052519201211, "300–1000 Hz_40ms_log_rms_mae_db": 1.28060924825525, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13453768290400264, "300–1000 Hz_40ms_envelope_correlation": 0.9891863337676776, "300–1000 Hz_40ms_modulation_mae_db": 1.1183009117196046, "1000–3000 Hz_10ms_log_rms_mae_db": 2.170394289300218, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19075188178236208, "1000–3000 Hz_10ms_envelope_correlation": 0.9718260810576005, "1000–3000 Hz_10ms_modulation_mae_db": 0.8328076284057553, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3284744418812446, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11904569518466503, "1000–3000 Hz_40ms_envelope_correlation": 0.9894648955423508, "1000–3000 Hz_40ms_modulation_mae_db": 0.8840791306523269, "3000–6000 Hz_10ms_log_rms_mae_db": 2.386510075181756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17859495748511042, "3000–6000 Hz_10ms_envelope_correlation": 0.9765916234862507, "3000–6000 Hz_10ms_modulation_mae_db": 1.0989251355764484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2792572049790658, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0899308345974763, "3000–6000 Hz_40ms_envelope_correlation": 0.9935171112661331, "3000–6000 Hz_40ms_modulation_mae_db": 1.2931057671991502, "mr_stft_db": 6.332017510927316, "band_mae_db": 0.42805519685765675, "rms_gain_error_db": -0.3255612679850689, "output_peak": 0.6392792940433446}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1584507123775991, "全帯域_10ms_rms_envelope_nrmse": 0.12174935226953706, "全帯域_10ms_envelope_correlation": 0.9861097252162544, "全帯域_10ms_modulation_mae_db": 0.6324787276467593, "全帯域_40ms_log_rms_mae_db": 0.600107997102427, "全帯域_40ms_rms_envelope_nrmse": 0.06016930842804313, "全帯域_40ms_envelope_correlation": 0.9962603029657502, "全帯域_40ms_modulation_mae_db": 0.8394252615410718, "300–1000 Hz_10ms_log_rms_mae_db": 2.0873547629721165, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1699463617744681, "300–1000 Hz_10ms_envelope_correlation": 0.9794393254946735, "300–1000 Hz_10ms_modulation_mae_db": 0.3941945820226646, "300–1000 Hz_40ms_log_rms_mae_db": 1.2988638762482032, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08928205629680222, "300–1000 Hz_40ms_envelope_correlation": 0.99411786222655, "300–1000 Hz_40ms_modulation_mae_db": 0.7920015720979121, "1000–3000 Hz_10ms_log_rms_mae_db": 2.213261743919854, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19796263298215597, "1000–3000 Hz_10ms_envelope_correlation": 0.9684885947993033, "1000–3000 Hz_10ms_modulation_mae_db": 1.1022444344445521, "1000–3000 Hz_40ms_log_rms_mae_db": 1.353355516637426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11089733289560899, "1000–3000 Hz_40ms_envelope_correlation": 0.9896498219405601, "1000–3000 Hz_40ms_modulation_mae_db": 1.2576587901640577, "3000–6000 Hz_10ms_log_rms_mae_db": 2.380731198957737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18871255434968234, "3000–6000 Hz_10ms_envelope_correlation": 0.9743259876237473, "3000–6000 Hz_10ms_modulation_mae_db": 1.5600729691248258, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3587898586833456, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11227158328106464, "3000–6000 Hz_40ms_envelope_correlation": 0.9907956764422935, "3000–6000 Hz_40ms_modulation_mae_db": 1.8466178160030753, "mr_stft_db": 6.3303023865270065, "band_mae_db": 0.24722133513319947, "rms_gain_error_db": -0.10365130280701627, "output_peak": 0.6654965009721195}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1585200718902808, "全帯域_10ms_rms_envelope_nrmse": 0.11715952589049922, "全帯域_10ms_envelope_correlation": 0.9871298873752429, "全帯域_10ms_modulation_mae_db": 0.2928387454154826, "全帯域_40ms_log_rms_mae_db": 0.6064237930214654, "全帯域_40ms_rms_envelope_nrmse": 0.0625115048782642, "全帯域_40ms_envelope_correlation": 0.9958814528544307, "全帯域_40ms_modulation_mae_db": 0.44081689220134906, "300–1000 Hz_10ms_log_rms_mae_db": 2.129626055961593, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15800321137494552, "300–1000 Hz_10ms_envelope_correlation": 0.9822393099784056, "300–1000 Hz_10ms_modulation_mae_db": 0.3235669498484613, "300–1000 Hz_40ms_log_rms_mae_db": 1.3126298328426305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0950978550616902, "300–1000 Hz_40ms_envelope_correlation": 0.9931073326898386, "300–1000 Hz_40ms_modulation_mae_db": 0.6471785714719188, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2153019203751927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19906577700119643, "1000–3000 Hz_10ms_envelope_correlation": 0.9681068000409291, "1000–3000 Hz_10ms_modulation_mae_db": 0.4349666113482626, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2684489124987033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10455709874103732, "1000–3000 Hz_40ms_envelope_correlation": 0.9908897204065844, "1000–3000 Hz_40ms_modulation_mae_db": 0.557089842149756, "3000–6000 Hz_10ms_log_rms_mae_db": 2.366825501228731, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16988671014883955, "3000–6000 Hz_10ms_envelope_correlation": 0.9798831518864237, "3000–6000 Hz_10ms_modulation_mae_db": 1.3915431265575586, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3274179153873937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09988712421214271, "3000–6000 Hz_40ms_envelope_correlation": 0.9935058297297074, "3000–6000 Hz_40ms_modulation_mae_db": 1.6351045649676408, "mr_stft_db": 6.340432166962112, "band_mae_db": 0.36265881402005645, "rms_gain_error_db": -0.06472773216145722, "output_peak": 0.7074180457141602}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1597317756166452, "全帯域_10ms_rms_envelope_nrmse": 0.11648817366617209, "全帯域_10ms_envelope_correlation": 0.9874011111504869, "全帯域_10ms_modulation_mae_db": 0.6033260780921409, "全帯域_40ms_log_rms_mae_db": 0.6142891714544699, "全帯域_40ms_rms_envelope_nrmse": 0.06812326898358222, "全帯域_40ms_envelope_correlation": 0.9951960835700647, "全帯域_40ms_modulation_mae_db": 0.6777309004433759, "300–1000 Hz_10ms_log_rms_mae_db": 2.0282965700515296, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15642931832155502, "300–1000 Hz_10ms_envelope_correlation": 0.9829394173376003, "300–1000 Hz_10ms_modulation_mae_db": 0.3610907965484835, "300–1000 Hz_40ms_log_rms_mae_db": 1.2225574095406966, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09228418786545274, "300–1000 Hz_40ms_envelope_correlation": 0.9936500484885351, "300–1000 Hz_40ms_modulation_mae_db": 0.5432598152185588, "1000–3000 Hz_10ms_log_rms_mae_db": 2.213211280003284, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22100123241700534, "1000–3000 Hz_10ms_envelope_correlation": 0.9604894350263182, "1000–3000 Hz_10ms_modulation_mae_db": 0.5020807898551518, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3069388365595238, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12540901646283784, "1000–3000 Hz_40ms_envelope_correlation": 0.9860827024739607, "1000–3000 Hz_40ms_modulation_mae_db": 0.6239836445613471, "3000–6000 Hz_10ms_log_rms_mae_db": 2.351920335920878, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1802224255638659, "3000–6000 Hz_10ms_envelope_correlation": 0.9772058117141924, "3000–6000 Hz_10ms_modulation_mae_db": 1.4532158293203221, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2943076091693473, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09813626920356416, "3000–6000 Hz_40ms_envelope_correlation": 0.9939650153876173, "3000–6000 Hz_40ms_modulation_mae_db": 1.7529885274274113, "mr_stft_db": 6.33357459062988, "band_mae_db": 0.26216783680270633, "rms_gain_error_db": 0.016098155404116733, "output_peak": 0.802712928867062}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1728003001207628, "全帯域_10ms_rms_envelope_nrmse": 0.12379076942673721, "全帯域_10ms_envelope_correlation": 0.9856276511416516, "全帯域_10ms_modulation_mae_db": 0.1821925491058726, "全帯域_40ms_log_rms_mae_db": 0.5965659080505987, "全帯域_40ms_rms_envelope_nrmse": 0.06086343076636436, "全帯域_40ms_envelope_correlation": 0.9961213480425994, "全帯域_40ms_modulation_mae_db": 0.31028866283021533, "300–1000 Hz_10ms_log_rms_mae_db": 2.1709346930631, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17015456148552596, "300–1000 Hz_10ms_envelope_correlation": 0.9794588746461127, "300–1000 Hz_10ms_modulation_mae_db": 0.39193179295563585, "300–1000 Hz_40ms_log_rms_mae_db": 1.2971513093294718, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08929239673812621, "300–1000 Hz_40ms_envelope_correlation": 0.9942527604428323, "300–1000 Hz_40ms_modulation_mae_db": 0.5647101782608325, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1892960911138224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2045236301293767, "1000–3000 Hz_10ms_envelope_correlation": 0.9665909648718091, "1000–3000 Hz_10ms_modulation_mae_db": 0.6105395480463388, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2921862321304203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11802265233051379, "1000–3000 Hz_40ms_envelope_correlation": 0.9887447624847506, "1000–3000 Hz_40ms_modulation_mae_db": 0.7830607311657994, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4008620986410243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17281785445114947, "3000–6000 Hz_10ms_envelope_correlation": 0.9781476672243847, "3000–6000 Hz_10ms_modulation_mae_db": 0.9737377767677439, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2955883028849486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09800964965610486, "3000–6000 Hz_40ms_envelope_correlation": 0.9922202223089188, "3000–6000 Hz_40ms_modulation_mae_db": 1.1062061891158843, "mr_stft_db": 6.348658272720972, "band_mae_db": 0.34716159773892286, "rms_gain_error_db": -0.06893540652853675, "output_peak": 0.7838511933440218}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1587189693288904, "全帯域_10ms_rms_envelope_nrmse": 0.1153918373723978, "全帯域_10ms_envelope_correlation": 0.9875222553278222, "全帯域_10ms_modulation_mae_db": 0.597388845153349, "全帯域_40ms_log_rms_mae_db": 0.6047885112675714, "全帯域_40ms_rms_envelope_nrmse": 0.062094636318940584, "全帯域_40ms_envelope_correlation": 0.9959602709759395, "全帯域_40ms_modulation_mae_db": 0.7399365613845438, "300–1000 Hz_10ms_log_rms_mae_db": 2.0837386713744754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18871999049536164, "300–1000 Hz_10ms_envelope_correlation": 0.9745598357003415, "300–1000 Hz_10ms_modulation_mae_db": 0.32539031977264593, "300–1000 Hz_40ms_log_rms_mae_db": 1.292814798908957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10456777186717342, "300–1000 Hz_40ms_envelope_correlation": 0.9917286478875904, "300–1000 Hz_40ms_modulation_mae_db": 0.6741044663629682, "1000–3000 Hz_10ms_log_rms_mae_db": 2.190793750973296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20396634680952438, "1000–3000 Hz_10ms_envelope_correlation": 0.9667770213579291, "1000–3000 Hz_10ms_modulation_mae_db": 0.3553356935221565, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2860199983957687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11569940556138457, "1000–3000 Hz_40ms_envelope_correlation": 0.988221653585471, "1000–3000 Hz_40ms_modulation_mae_db": 0.5852432863774969, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4009857377935626, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17356239067869272, "3000–6000 Hz_10ms_envelope_correlation": 0.9786791374769432, "3000–6000 Hz_10ms_modulation_mae_db": 1.0535852069627352, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3173491478960755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10875727377680156, "3000–6000 Hz_40ms_envelope_correlation": 0.9915037478130255, "3000–6000 Hz_40ms_modulation_mae_db": 1.1911437650648085, "mr_stft_db": 6.330763878512947, "band_mae_db": 0.18914237764000982, "rms_gain_error_db": -0.08260240982798236, "output_peak": 0.682603202139034}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.807023250199886, "spectral_convergence": 0.45292244951261995, "spectral_convergence_status": "ok", "amplitude_floor": 1.602383167696498e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.194148783380935, "spectral_convergence": 0.4291038560242612, "spectral_convergence_status": "ok", "amplitude_floor": 1.3018377539140758e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.044038708840647, "spectral_convergence": 0.4566067738412843, "spectral_convergence_status": "ok", "amplitude_floor": 8.15121073207966e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00018979827012669627, "output_power": 0.00020877618110064374, "delta_db": 0.41388699448641764, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007146011617002843, "output_power": 0.0006948474310842701, "delta_db": -0.12174261655816973, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006180173350849692, "output_power": 0.0006120255322027779, "delta_db": -0.042311167850207505, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00021309861976674557, "output_power": 0.00017385477092764467, "delta_db": -0.8839402370669013, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003708463313924785, "output_power": 0.0003686691676312965, "delta_db": -0.025571686886852345, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001752804045595543, "output_power": 0.00018536191695676815, "delta_db": 0.24287145210909575, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2883873591596074, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0731467692890856, "pitch": null, "output_rms_dbfs": -26.490671509375264, "output_peak": 0.7127733806040093, "reference_rms_dbfs": -26.41752474008618, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.3484035808071555}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002801281523487693, "log_rms_mae_db": 1.131843456463189, "rms_envelope_nrmse": 0.11301748064195555, "envelope_correlation": 0.9880991408298391, "rms_gain_error_db": -0.07316088324631882, "reference_mean_power": 0.0022835195719053173, "output_mean_power": 0.002245373795571612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.817352658591746e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000450774313706133, "output_power": 0.00043988211096717057, "reference_active": true, "delta_db": -0.10622860309296833}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021888119814572043, "output_power": 0.00021159955414820197, "reference_active": true, "delta_db": -0.14693709072805158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.753791744185585e-05, "output_power": 2.2823675534098723e-05, "reference_active": true, "delta_db": -2.1608459014604016}], "active_band_count": 3, "active_band_mae_db": 0.8046705317604738}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022259441477048914, "log_rms_mae_db": 0.5831172210299612, "rms_envelope_nrmse": 0.0634570362872321, "envelope_correlation": 0.9958228646082682, "rms_gain_error_db": -0.07316281967502566, "reference_mean_power": 0.002294705859570105, "output_mean_power": 0.002256372212279548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.266680044598684e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042660839170957063, "output_power": 0.0004129570693084159, "reference_active": true, "delta_db": -0.14124489007334973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012272337901795565, "output_power": 0.00011719526535554791, "reference_active": true, "delta_db": -0.20017237762026885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.241371364832561e-06, "output_power": 2.064024402129923e-06, "reference_active": true, "delta_db": -1.960139636497845}], "active_band_count": 3, "active_band_mae_db": 0.7671856347304878}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002244694917839089, "log_rms_mae_db": 2.1358944856013866, "rms_envelope_nrmse": 0.15024353959170567, "envelope_correlation": 0.9840888612355891, "rms_gain_error_db": -0.0927767761528521, "reference_mean_power": 0.001082154267593255, "output_mean_power": 0.001059281771510849}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.600154760198564e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003310843525581766, "output_power": 0.0003260646201768385, "reference_active": true, "delta_db": -0.06634977963530247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.047815482126218e-05, "output_power": 8.460397622158796e-05, "reference_active": true, "delta_db": -0.2915296065242643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6938491072881828e-05, "output_power": 1.0315047843435457e-05, "reference_active": true, "delta_db": -2.1540347291128565}], "active_band_count": 3, "active_band_mae_db": 0.8373047050908077}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018235367957614062, "log_rms_mae_db": 1.3025257021097796, "rms_envelope_nrmse": 0.08156332671700695, "envelope_correlation": 0.9951369815086287, "rms_gain_error_db": -0.09276203551517578, "reference_mean_power": 0.0010875823794426604, "output_mean_power": 0.0010645987677789564}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.047574386178106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00032394555601725244, "output_power": 0.0003153106693204876, "reference_active": true, "delta_db": -0.11733360109480373}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.774463594481584e-05, "output_power": 4.470882633469009e-05, "reference_active": true, "delta_db": -0.28531316954534486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5157059923891315e-06, "output_power": 8.946871304159179e-07, "reference_active": true, "delta_db": -2.2894377731116338}], "active_band_count": 3, "active_band_mae_db": 0.8973615145839274}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014040871531833862, "log_rms_mae_db": 2.1785341294755445, "rms_envelope_nrmse": 0.19382711708640502, "envelope_correlation": 0.971592420697943, "rms_gain_error_db": -0.5600124387951684, "reference_mean_power": 0.00036159326722347827, "output_mean_power": 0.00031784771346272874}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2796709554971252e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010222442860363171, "output_power": 8.805063787595896e-05, "reference_active": true, "delta_db": -0.648221851621455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.551583786781791e-05, "output_power": 4.120909795039294e-05, "reference_active": true, "delta_db": -0.43169433643385685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1818852015306123e-05, "output_power": 6.54310914696968e-06, "reference_active": true, "delta_db": -2.5679112993492503}], "active_band_count": 3, "active_band_mae_db": 1.2159424958015208}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010845211232047094, "log_rms_mae_db": 1.2753258682093458, "rms_envelope_nrmse": 0.12549488714275642, "envelope_correlation": 0.9890232613787485, "rms_gain_error_db": -0.5600044887651, "reference_mean_power": 0.00036337479954251316, "output_mean_power": 0.00031941430070753696}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.941433739824422e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.608975832570782e-05, "output_power": 8.35482393457026e-05, "reference_active": true, "delta_db": -0.6073979887698676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4640366344588194e-05, "output_power": 2.1699585116736672e-05, "reference_active": true, "delta_db": -0.5519573000939539}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.520284367741145e-07, "output_power": 5.286498777349309e-07, "reference_active": true, "delta_db": -2.554817847104076}], "active_band_count": 3, "active_band_mae_db": 1.2380577119892993}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012084003773539908, "log_rms_mae_db": 2.4660142578499302, "rms_envelope_nrmse": 0.19756014676567799, "envelope_correlation": 0.9731422339850966, "rms_gain_error_db": 0.2323952296437854, "reference_mean_power": 0.00030833628390941923, "output_mean_power": 0.00032528508985879}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2964446771205807e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7472950448048514e-05, "output_power": 4.687631211487697e-05, "reference_active": true, "delta_db": -0.054927867544452155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.978240331431912e-05, "output_power": 4.082223116396316e-05, "reference_active": true, "delta_db": -0.8617912128622575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3740568823166797e-05, "output_power": 7.994856051331242e-06, "reference_active": true, "delta_db": -2.3519406312662685}], "active_band_count": 3, "active_band_mae_db": 1.089553237224326}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010527157253957595, "log_rms_mae_db": 1.3562973572001216, "rms_envelope_nrmse": 0.11102791170309917, "envelope_correlation": 0.9914173667242481, "rms_gain_error_db": 0.23248983300179699, "reference_mean_power": 0.0003098592063705908, "output_mean_power": 0.0003268988460311107}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.056454468513196e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.721882315895364e-05, "output_power": 5.3949209982458445e-05, "reference_active": true, "delta_db": -0.2555383188787292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8898944201462896e-05, "output_power": 2.2181854777947045e-05, "reference_active": true, "delta_db": -1.148841187739878}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2532553850736713e-06, "output_power": 6.576299625621354e-07, "reference_active": true, "delta_db": -2.800579873345847}], "active_band_count": 3, "active_band_mae_db": 1.401653126654818}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.766514626922162, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2773725031b6cf5ed8dfa368b1b765cd103aff8743e99ba315ef07d1ec0b59d6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9674378103891745, "standard_deviation": 0.003600933957861946, "minimum": 0.9604894350263182, "maximum": 0.9718260810576005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2038831923465865, "standard_deviation": 0.028368220813478104, "minimum": 2.170394289300218, "maximum": 2.2602723336114807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7937383512268855, "standard_deviation": 0.37162866701493086, "minimum": 0.3553356935221565, "maximum": 1.2959896083913456, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20256668947373618, "standard_deviation": 0.009578374781220964, "minimum": 0.19075188178236208, "maximum": 0.22100123241700534, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9888178459500456, "standard_deviation": 0.0013841626534197445, "minimum": 0.9860827024739607, "maximum": 0.9908897204065844, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3020007329973482, "standard_deviation": 0.028188484087818703, "minimum": 1.2684489124987033, "maximum": 1.353355516637426, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.934389047514586, "standard_deviation": 0.36970102594608795, "minimum": 0.557089842149756, "maximum": 1.5459392430566041, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11795944985473264, "standard_deviation": 0.007478502450611008, "minimum": 0.10455709874103732, "maximum": 0.12549488714275642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9766174232376166, "standard_deviation": 0.0023196779567228655, "minimum": 0.9731422339850966, "maximum": 0.9798831518864237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.389052441104188, "standard_deviation": 0.03602120231397078, "minimum": 2.351920335920878, "maximum": 2.4660142578499302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2142855781413124, "standard_deviation": 0.21882265628598294, "minimum": 0.9737377767677439, "maximum": 1.5600729691248258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18079897720922153, "standard_deviation": 0.009272368402846154, "minimum": 0.16988671014883955, "maximum": 0.19756014676567799, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922716340061177, "standard_deviation": 0.0012241688653119448, "minimum": 0.9907956764422935, "maximum": 0.9939650153876173, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3182919053216091, "standard_deviation": 0.028745207591314135, "minimum": 1.2792572049790658, "maximum": 1.3587898586833456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4365593610643044, "standard_deviation": 0.27465906633467446, "minimum": 1.1062061891158843, "maximum": 1.8466178160030753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10300323918435354, "standard_deviation": 0.007761074337140719, "minimum": 0.0899308345974763, "maximum": 0.11227158328106464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9795787934157153, "standard_deviation": 0.0036733756770798262, "minimum": 0.9740607433328726, "maximum": 0.9840888612355891, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1087828236526756, "standard_deviation": 0.04286953551496893, "minimum": 2.0282965700515296, "maximum": 2.1709346930631, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4830407099164413, "standard_deviation": 0.2542810502348039, "minimum": 0.2908412811727104, "maximum": 0.9400052519201211, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16986213241050194, "standard_deviation": 0.01609096286568491, "minimum": 0.15024353959170567, "maximum": 0.19721097964768294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9932353131946954, "standard_deviation": 0.0019433345210695412, "minimum": 0.9891863337676776, "maximum": 0.9951369815086287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2858977728964414, "standard_deviation": 0.027795423667884366, "minimum": 1.2225574095406966, "maximum": 1.3126298328426305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7029870476591292, "standard_deviation": 0.22885852134332424, "minimum": 0.38697935155731117, "maximum": 1.1183009117196046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09632410927058352, "standard_deviation": 0.016969129292238868, "minimum": 0.08156332671700695, "maximum": 0.13453768290400264, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3024112453621391, "standard_deviation": 0.07480774396890925, "minimum": 0.18914237764000982, "maximum": 0.42805519685765675, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.339351383138551, "standard_deviation": 0.008786109786253788, "minimum": 6.3303023865270065, "maximum": 6.35065867802102, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7007312203529262, "standard_deviation": 0.06629934777513397, "minimum": 0.6117152171396587, "maximum": 0.802712928867062, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09871480970934124, "standard_deviation": 0.09838245415701126, "minimum": -0.3255612679850689, "maximum": 0.016098155404116733, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9865669966707544, "standard_deviation": 0.0015352822856780043, "minimum": 0.9833179771184892, "maximum": 0.9880991408298391, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1550104359831508, "standard_deviation": 0.011769325287426104, "minimum": 1.131843456463189, "maximum": 1.1728003001207628, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5553468035615639, "standard_deviation": 0.23760385663115033, "minimum": 0.1821925491058726, "maximum": 0.8780500852439846, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12009877733585786, "standard_deviation": 0.007620848855706386, "minimum": 0.11301748064195555, "maximum": 0.13688654994832392, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9956812653455303, "standard_deviation": 0.0009527022343794808, "minimum": 0.9935572161918883, "maximum": 0.996650583555301, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6007443739205857, "standard_deviation": 0.012724640247036699, "minimum": 0.5831172210299612, "maximum": 0.6172489426759166, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.670940864238972, "standard_deviation": 0.205759637578086, "minimum": 0.31028866283021533, "maximum": 0.9367133065891631, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06505220474318417, "standard_deviation": 0.009182929720236472, "minimum": 0.05683458018006382, "maximum": 0.08636387210298296, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0002", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.1380350589752197, "predicted_mos": 2.1380350589752197, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9bdb807af09731618b9cbe4bb8c71e49291e987780a42afc6a33cc0932a3949e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.776877900119871, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1854386101872576, "全帯域_10ms_rms_envelope_nrmse": 0.12764862422649065, "全帯域_10ms_envelope_correlation": 0.9848301286487229, "全帯域_10ms_modulation_mae_db": 0.8459452151085172, "全帯域_40ms_log_rms_mae_db": 0.5960499382229251, "全帯域_40ms_rms_envelope_nrmse": 0.06541106112221914, "全帯域_40ms_envelope_correlation": 0.9957803812827353, "全帯域_40ms_modulation_mae_db": 0.8729026537748984, "300–1000 Hz_10ms_log_rms_mae_db": 1.738526875234294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17135275765938485, "300–1000 Hz_10ms_envelope_correlation": 0.9791747197675943, "300–1000 Hz_10ms_modulation_mae_db": 0.6808595760742632, "300–1000 Hz_40ms_log_rms_mae_db": 1.023993662745842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09192850389821546, "300–1000 Hz_40ms_envelope_correlation": 0.9938990883907949, "300–1000 Hz_40ms_modulation_mae_db": 0.8012069391342899, "1000–3000 Hz_10ms_log_rms_mae_db": 1.688085725136372, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.174035148282334, "1000–3000 Hz_10ms_envelope_correlation": 0.9762181304289438, "1000–3000 Hz_10ms_modulation_mae_db": 1.2055285566551814, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0348851777352064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1015312530650686, "1000–3000 Hz_40ms_envelope_correlation": 0.9920116611105982, "1000–3000 Hz_40ms_modulation_mae_db": 1.4076055238041398, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7964584246724558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17115236703015743, "3000–6000 Hz_10ms_envelope_correlation": 0.9797595946842257, "3000–6000 Hz_10ms_modulation_mae_db": 0.825995857674601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0810712071026949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09978691972702063, "3000–6000 Hz_40ms_envelope_correlation": 0.993974530237765, "3000–6000 Hz_40ms_modulation_mae_db": 0.9817639380088714, "mr_stft_db": 5.926170835359734, "band_mae_db": 0.30424229059295, "rms_gain_error_db": -0.17128274563757762, "output_peak": 0.6751659225258044}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.23759269588862, "全帯域_10ms_rms_envelope_nrmse": 0.11614311131052606, "全帯域_10ms_envelope_correlation": 0.9877166597798611, "全帯域_10ms_modulation_mae_db": 0.13864567885361578, "全帯域_40ms_log_rms_mae_db": 0.6706634622790644, "全帯域_40ms_rms_envelope_nrmse": 0.06412540130138458, "全帯域_40ms_envelope_correlation": 0.9958958408986018, "全帯域_40ms_modulation_mae_db": 0.3207804311060303, "300–1000 Hz_10ms_log_rms_mae_db": 1.7390473373838748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15124487227728875, "300–1000 Hz_10ms_envelope_correlation": 0.9839160995876526, "300–1000 Hz_10ms_modulation_mae_db": 0.1913406045688064, "300–1000 Hz_40ms_log_rms_mae_db": 1.0321557327993973, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0822856375833148, "300–1000 Hz_40ms_envelope_correlation": 0.994931297970599, "300–1000 Hz_40ms_modulation_mae_db": 0.1627101715766508, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6652638563724427, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19059203308033695, "1000–3000 Hz_10ms_envelope_correlation": 0.9721749755730745, "1000–3000 Hz_10ms_modulation_mae_db": 0.241272824099785, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0262733473969277, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11739737874281815, "1000–3000 Hz_40ms_envelope_correlation": 0.988796332318404, "1000–3000 Hz_40ms_modulation_mae_db": 0.44024977425104783, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7581681407982053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16319373303373638, "3000–6000 Hz_10ms_envelope_correlation": 0.9807919584076124, "3000–6000 Hz_10ms_modulation_mae_db": 0.7288069104808801, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0778166303188907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09248711081058578, "3000–6000 Hz_40ms_envelope_correlation": 0.9931897304024196, "3000–6000 Hz_40ms_modulation_mae_db": 0.8546249126959365, "mr_stft_db": 5.927073556714544, "band_mae_db": 0.19348047391979506, "rms_gain_error_db": 0.08684041011475734, "output_peak": 0.8757533500085376}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.233269282080088, "全帯域_10ms_rms_envelope_nrmse": 0.1362627624435625, "全帯域_10ms_envelope_correlation": 0.9833683489589992, "全帯域_10ms_modulation_mae_db": 0.7089048897774092, "全帯域_40ms_log_rms_mae_db": 0.6083186702419164, "全帯域_40ms_rms_envelope_nrmse": 0.08620048106512504, "全帯域_40ms_envelope_correlation": 0.9934080414872176, "全帯域_40ms_modulation_mae_db": 0.7396779138993796, "300–1000 Hz_10ms_log_rms_mae_db": 1.706419406266526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19214090649548352, "300–1000 Hz_10ms_envelope_correlation": 0.9751432816298362, "300–1000 Hz_10ms_modulation_mae_db": 0.9167505797711956, "300–1000 Hz_40ms_log_rms_mae_db": 1.0159472784129198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1279134145122915, "300–1000 Hz_40ms_envelope_correlation": 0.990023883286696, "300–1000 Hz_40ms_modulation_mae_db": 1.0859180518984657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7011213328897843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2043292815062326, "1000–3000 Hz_10ms_envelope_correlation": 0.9675065249024636, "1000–3000 Hz_10ms_modulation_mae_db": 0.9040624907304435, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0146621154080424, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11535659822056335, "1000–3000 Hz_40ms_envelope_correlation": 0.9908603555857601, "1000–3000 Hz_40ms_modulation_mae_db": 1.0969522815177715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8121842883759263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15495121620543004, "3000–6000 Hz_10ms_envelope_correlation": 0.9824930925994345, "3000–6000 Hz_10ms_modulation_mae_db": 0.1469635872909109, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0430454903200441, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08522524460554572, "3000–6000 Hz_40ms_envelope_correlation": 0.9941154797923557, "3000–6000 Hz_40ms_modulation_mae_db": 0.17492439888291325, "mr_stft_db": 5.917460758127756, "band_mae_db": 0.3555334798355609, "rms_gain_error_db": -0.30816881847793187, "output_peak": 0.6196926859457612}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2094509983008115, "全帯域_10ms_rms_envelope_nrmse": 0.12189794641391742, "全帯域_10ms_envelope_correlation": 0.986085211364982, "全帯域_10ms_modulation_mae_db": 0.38051632872663227, "全帯域_40ms_log_rms_mae_db": 0.6238571958574938, "全帯域_40ms_rms_envelope_nrmse": 0.06607396379494433, "全帯域_40ms_envelope_correlation": 0.9953575964918744, "全帯域_40ms_modulation_mae_db": 0.47468755669439516, "300–1000 Hz_10ms_log_rms_mae_db": 1.5865288781774196, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1581312766508349, "300–1000 Hz_10ms_envelope_correlation": 0.9822202652944956, "300–1000 Hz_10ms_modulation_mae_db": 0.24827760751425457, "300–1000 Hz_40ms_log_rms_mae_db": 0.9609183070686738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08421132732837082, "300–1000 Hz_40ms_envelope_correlation": 0.9947429872254756, "300–1000 Hz_40ms_modulation_mae_db": 0.41215639319383196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6853989378512702, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17920455245219954, "1000–3000 Hz_10ms_envelope_correlation": 0.9743375199327109, "1000–3000 Hz_10ms_modulation_mae_db": 0.42738745883330864, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0579745352085013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10773044059875003, "1000–3000 Hz_40ms_envelope_correlation": 0.9901867689352979, "1000–3000 Hz_40ms_modulation_mae_db": 0.4166056510253551, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7808210538213691, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17766071171703182, "3000–6000 Hz_10ms_envelope_correlation": 0.9793503945226375, "3000–6000 Hz_10ms_modulation_mae_db": 0.5679298592070021, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0730213883403137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1187475974447065, "3000–6000 Hz_40ms_envelope_correlation": 0.9908030191593254, "3000–6000 Hz_40ms_modulation_mae_db": 0.7187860722649778, "mr_stft_db": 5.926383481473771, "band_mae_db": 0.19477926308602492, "rms_gain_error_db": -0.01911036652743314, "output_peak": 0.69947907805075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.198755791315488, "全帯域_10ms_rms_envelope_nrmse": 0.10703702377274489, "全帯域_10ms_envelope_correlation": 0.989554330211541, "全帯域_10ms_modulation_mae_db": 0.11782932592528216, "全帯域_40ms_log_rms_mae_db": 0.6121742652554055, "全帯域_40ms_rms_envelope_nrmse": 0.05775170935838145, "全帯域_40ms_envelope_correlation": 0.9966735741125007, "全帯域_40ms_modulation_mae_db": 0.1659726882077855, "300–1000 Hz_10ms_log_rms_mae_db": 1.656289787217537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1568468970633894, "300–1000 Hz_10ms_envelope_correlation": 0.9838604529921084, "300–1000 Hz_10ms_modulation_mae_db": 0.3496343335255143, "300–1000 Hz_40ms_log_rms_mae_db": 0.965701405269856, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09861793339358557, "300–1000 Hz_40ms_envelope_correlation": 0.9936537740665479, "300–1000 Hz_40ms_modulation_mae_db": 0.19206330987257947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6715501793476113, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19115794138630116, "1000–3000 Hz_10ms_envelope_correlation": 0.9721164264329966, "1000–3000 Hz_10ms_modulation_mae_db": 0.7825995799235487, "1000–3000 Hz_40ms_log_rms_mae_db": 1.014580817070327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11221071103719732, "1000–3000 Hz_40ms_envelope_correlation": 0.9917586383614819, "1000–3000 Hz_40ms_modulation_mae_db": 0.9299275797298909, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7598450379423067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15997460837091357, "3000–6000 Hz_10ms_envelope_correlation": 0.9816334806393098, "3000–6000 Hz_10ms_modulation_mae_db": 0.3478104040217316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0652136646521386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09275232908052042, "3000–6000 Hz_40ms_envelope_correlation": 0.9931622471479286, "3000–6000 Hz_40ms_modulation_mae_db": 0.33149115562763015, "mr_stft_db": 5.916006368741567, "band_mae_db": 0.3603973587882878, "rms_gain_error_db": 0.08641709609776654, "output_peak": 0.7485696270432497}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1831852500755193, "全帯域_10ms_rms_envelope_nrmse": 0.11790877219919699, "全帯域_10ms_envelope_correlation": 0.9872920702832247, "全帯域_10ms_modulation_mae_db": 0.6463576614246945, "全帯域_40ms_log_rms_mae_db": 0.6053849027672789, "全帯域_40ms_rms_envelope_nrmse": 0.06141714232868349, "全帯域_40ms_envelope_correlation": 0.996605861655733, "全帯域_40ms_modulation_mae_db": 0.7564595627268164, "300–1000 Hz_10ms_log_rms_mae_db": 1.7852337095834758, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1670733559440253, "300–1000 Hz_10ms_envelope_correlation": 0.980771075591877, "300–1000 Hz_10ms_modulation_mae_db": 0.43606207858940893, "300–1000 Hz_40ms_log_rms_mae_db": 1.1081943870057778, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09474647141935207, "300–1000 Hz_40ms_envelope_correlation": 0.9944505322778855, "300–1000 Hz_40ms_modulation_mae_db": 0.715482305223713, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6987917597933444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16969720120622006, "1000–3000 Hz_10ms_envelope_correlation": 0.9773763367689119, "1000–3000 Hz_10ms_modulation_mae_db": 0.7441293282781382, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0351024543800915, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09747865398396167, "1000–3000 Hz_40ms_envelope_correlation": 0.9926311651961112, "1000–3000 Hz_40ms_modulation_mae_db": 0.6705762585511152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7496426132493492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15540546480539139, "3000–6000 Hz_10ms_envelope_correlation": 0.9823474876877684, "3000–6000 Hz_10ms_modulation_mae_db": 0.6866999622901014, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0506355168105472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08700115702015648, "3000–6000 Hz_40ms_envelope_correlation": 0.9939996633314134, "3000–6000 Hz_40ms_modulation_mae_db": 0.882277381733585, "mr_stft_db": 5.930395295428087, "band_mae_db": 0.29774712792036795, "rms_gain_error_db": -0.21274035610271724, "output_peak": 0.6692146956771972}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2041414406044375, "全帯域_10ms_rms_envelope_nrmse": 0.11523102432932963, "全帯域_10ms_envelope_correlation": 0.9876269594137757, "全帯域_10ms_modulation_mae_db": 0.6472521591523557, "全帯域_40ms_log_rms_mae_db": 0.5964609915099568, "全帯域_40ms_rms_envelope_nrmse": 0.05607965911047341, "全帯域_40ms_envelope_correlation": 0.9968703971760711, "全帯域_40ms_modulation_mae_db": 0.7417359441474988, "300–1000 Hz_10ms_log_rms_mae_db": 1.7051931633771509, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15421802290258207, "300–1000 Hz_10ms_envelope_correlation": 0.9831423476372338, "300–1000 Hz_10ms_modulation_mae_db": 0.12452411897474118, "300–1000 Hz_40ms_log_rms_mae_db": 1.0907004551532788, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08765101211889652, "300–1000 Hz_40ms_envelope_correlation": 0.9943161899857841, "300–1000 Hz_40ms_modulation_mae_db": 0.3310842265569447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7811164621265163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18461190372319766, "1000–3000 Hz_10ms_envelope_correlation": 0.9730711747735282, "1000–3000 Hz_10ms_modulation_mae_db": 1.0611916227381748, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0728476568061815, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10412806098108147, "1000–3000 Hz_40ms_envelope_correlation": 0.9915703635808659, "1000–3000 Hz_40ms_modulation_mae_db": 1.0929530954848679, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7661722075550639, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16759118119978236, "3000–6000 Hz_10ms_envelope_correlation": 0.979606189350022, "3000–6000 Hz_10ms_modulation_mae_db": 1.0837575599644464, "3000–6000 Hz_40ms_log_rms_mae_db": 1.044440566000013, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08627843634316824, "3000–6000 Hz_40ms_envelope_correlation": 0.994500255552617, "3000–6000 Hz_40ms_modulation_mae_db": 1.3327127356219144, "mr_stft_db": 5.9354478842607294, "band_mae_db": 0.21027212291861483, "rms_gain_error_db": -0.13713177938299906, "output_peak": 0.6814953842481377}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.192795038827077, "全帯域_10ms_rms_envelope_nrmse": 0.11616741053956943, "全帯域_10ms_envelope_correlation": 0.9873486063241892, "全帯域_10ms_modulation_mae_db": 0.1114037114301556, "全帯域_40ms_log_rms_mae_db": 0.590777677011631, "全帯域_40ms_rms_envelope_nrmse": 0.059698090797280695, "全帯域_40ms_envelope_correlation": 0.9962491329989167, "全帯域_40ms_modulation_mae_db": 0.1904989722436732, "300–1000 Hz_10ms_log_rms_mae_db": 1.772553857463916, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1563316039765927, "300–1000 Hz_10ms_envelope_correlation": 0.9829374533349147, "300–1000 Hz_10ms_modulation_mae_db": 0.3307148917886323, "300–1000 Hz_40ms_log_rms_mae_db": 1.0211998111031964, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08633779185356114, "300–1000 Hz_40ms_envelope_correlation": 0.9950404549119096, "300–1000 Hz_40ms_modulation_mae_db": 0.4422776864236491, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7017343845838317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20869871861329017, "1000–3000 Hz_10ms_envelope_correlation": 0.9671662329000815, "1000–3000 Hz_10ms_modulation_mae_db": 0.5343590254758874, "1000–3000 Hz_40ms_log_rms_mae_db": 1.020115747056106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11859027611621044, "1000–3000 Hz_40ms_envelope_correlation": 0.9888748491328696, "1000–3000 Hz_40ms_modulation_mae_db": 0.5199136895980097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7684936242894758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16995540782215035, "3000–6000 Hz_10ms_envelope_correlation": 0.9794344487335614, "3000–6000 Hz_10ms_modulation_mae_db": 0.3677998321887143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0756932317950527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10346786821356636, "3000–6000 Hz_40ms_envelope_correlation": 0.9916541371294642, "3000–6000 Hz_40ms_modulation_mae_db": 0.3920129601572719, "mr_stft_db": 5.9092380305219665, "band_mae_db": 0.2611758782778882, "rms_gain_error_db": -0.06166375662578842, "output_peak": 0.8583161302004639}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.232382801883882, "spectral_convergence": 0.4600047007299081, "spectral_convergence_status": "ok", "amplitude_floor": 1.602383167696498e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.662327132062108, "spectral_convergence": 0.43032264476683846, "spectral_convergence_status": "ok", "amplitude_floor": 1.3018377539140758e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.883802572133215, "spectral_convergence": 0.4455292853029788, "spectral_convergence_status": "ok", "amplitude_floor": 8.15121073207966e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00018979827012669627, "output_power": 0.00020928943532477488, "delta_db": 0.42455056405943153, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007146011617002843, "output_power": 0.0007077800974215158, "delta_db": -0.04165371649463644, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006180173350849692, "output_power": 0.0005861886511639971, "delta_db": -0.22963250931274898, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00021309861976674557, "output_power": 0.00019372264772705197, "delta_db": -0.4140024062528006, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003708463313924785, "output_power": 0.0003287275753062541, "delta_db": -0.5235785070510386, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001752804045595543, "output_power": 0.0001676987258437291, "delta_db": -0.192036040387044, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.30424229059295, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.17128274563757762, "pitch": null, "output_rms_dbfs": -26.588807485723756, "output_peak": 0.6751659225258044, "reference_rms_dbfs": -26.41752474008618, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.926170835359734}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002801281523487693, "log_rms_mae_db": 1.1854386101872576, "rms_envelope_nrmse": 0.12764862422649065, "envelope_correlation": 0.9848301286487229, "rms_gain_error_db": -0.17128140700924419, "reference_mean_power": 0.0022835195719053173, "output_mean_power": 0.0021952126545299634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.817352658591746e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000450774313706133, "output_power": 0.00042419409928101676, "reference_active": true, "delta_db": -0.2639453784598149}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021888119814572043, "output_power": 0.0002032961163669433, "reference_active": true, "delta_db": -0.3207937510883762}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.753791744185585e-05, "output_power": 2.394200533523207e-05, "reference_active": true, "delta_db": -1.9530965157773605}], "active_band_count": 3, "active_band_mae_db": 0.8459452151085172}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022259441477048914, "log_rms_mae_db": 0.5960499382229251, "rms_envelope_nrmse": 0.06541106112221914, "envelope_correlation": 0.9957803812827353, "rms_gain_error_db": -0.17125488782142045, "reference_mean_power": 0.002294705859570105, "output_mean_power": 0.002205979822963146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.266680044598684e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042660839170957063, "output_power": 0.0003999553720340112, "reference_active": true, "delta_db": -0.28017859671881773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012272337901795565, "output_power": 0.0001127464205675009, "reference_active": true, "delta_db": -0.36824541520436527}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.241371364832561e-06, "output_power": 2.05920885058418e-06, "reference_active": true, "delta_db": -1.970283949401512}], "active_band_count": 3, "active_band_mae_db": 0.8729026537748984}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002244694917839089, "log_rms_mae_db": 1.738526875234294, "rms_envelope_nrmse": 0.17135275765938485, "envelope_correlation": 0.9791747197675943, "rms_gain_error_db": -0.22566903280482317, "reference_mean_power": 0.001082154267593255, "output_mean_power": 0.0010273591025427174}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.600154760198564e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003310843525581766, "output_power": 0.0003143442550450354, "reference_active": true, "delta_db": -0.22533128367637348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.047815482126218e-05, "output_power": 8.71913821508481e-05, "reference_active": true, "delta_db": -0.16070173043607097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6938491072881828e-05, "output_power": 1.156698435478155e-05, "reference_active": true, "delta_db": -1.656545714110345}], "active_band_count": 3, "active_band_mae_db": 0.6808595760742632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018235367957614062, "log_rms_mae_db": 1.023993662745842, "rms_envelope_nrmse": 0.09192850389821546, "envelope_correlation": 0.9938990883907949, "rms_gain_error_db": -0.22567293106248804, "reference_mean_power": 0.0010875823794426604, "output_mean_power": 0.0010325114337356221}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.047574386178106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00032394555601725244, "output_power": 0.0003074573152295637, "reference_active": true, "delta_db": -0.22687196012000946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.774463594481584e-05, "output_power": 4.591706383366008e-05, "reference_active": true, "delta_db": -0.1695047685260092}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5157059923891315e-06, "output_power": 9.54751960287666e-07, "reference_active": true, "delta_db": -2.007244088756851}], "active_band_count": 3, "active_band_mae_db": 0.8012069391342899}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014040871531833862, "log_rms_mae_db": 1.688085725136372, "rms_envelope_nrmse": 0.174035148282334, "envelope_correlation": 0.9762181304289438, "rms_gain_error_db": -0.3733418108937439, "reference_mean_power": 0.00036159326722347827, "output_mean_power": 0.00033180746636330457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2796709554971252e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010222442860363171, "output_power": 8.842256035865697e-05, "reference_active": true, "delta_db": -0.6299160553086072}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.551583786781791e-05, "output_power": 3.592426943932809e-05, "reference_active": true, "delta_db": -1.027745967244191}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1818852015306123e-05, "output_power": 7.52805754326628e-06, "reference_active": true, "delta_db": -1.958923647412746}], "active_band_count": 3, "active_band_mae_db": 1.2055285566551814}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010845211232047094, "log_rms_mae_db": 1.0348851777352064, "rms_envelope_nrmse": 0.1015312530650686, "envelope_correlation": 0.9920116611105982, "rms_gain_error_db": -0.3733585089616349, "reference_mean_power": 0.00036337479954251316, "output_mean_power": 0.00033344096510733027}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.941433739824422e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.608975832570782e-05, "output_power": 8.251114329140642e-05, "reference_active": true, "delta_db": -0.6616449626489916}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4640366344588194e-05, "output_power": 1.896625134032564e-05, "reference_active": true, "delta_db": -1.136656589612464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.520284367741145e-07, "output_power": 5.447514800865873e-07, "reference_active": true, "delta_db": -2.4245150191509643}], "active_band_count": 3, "active_band_mae_db": 1.4076055238041398}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012084003773539908, "log_rms_mae_db": 1.7964584246724558, "rms_envelope_nrmse": 0.17115236703015743, "envelope_correlation": 0.9797595946842257, "rms_gain_error_db": -0.4814289407083825, "reference_mean_power": 0.00030833628390941923, "output_mean_power": 0.00027598262422073465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2964446771205807e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7472950448048514e-05, "output_power": 4.4100504888766504e-05, "reference_active": true, "delta_db": -0.3200266244001837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.978240331431912e-05, "output_power": 4.3384782944469745e-05, "reference_active": true, "delta_db": -0.5973842994164296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3740568823166797e-05, "output_power": 9.592836596853991e-06, "reference_active": true, "delta_db": -1.5605766492071902}], "active_band_count": 3, "active_band_mae_db": 0.825995857674601}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010527157253957595, "log_rms_mae_db": 1.0810712071026949, "rms_envelope_nrmse": 0.09978691972702063, "envelope_correlation": 0.993974530237765, "rms_gain_error_db": -0.481420878549344, "reference_mean_power": 0.0003098592063705908, "output_mean_power": 0.00027734626161668983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.056454468513196e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.721882315895364e-05, "output_power": 5.1439617639685686e-05, "reference_active": true, "delta_db": -0.46241189527303267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8898944201462896e-05, "output_power": 2.423665030975612e-05, "reference_active": true, "delta_db": -0.7640937963779528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2532553850736713e-06, "output_power": 8.436473897799166e-07, "reference_active": true, "delta_db": -1.7187861223756289}], "active_band_count": 3, "active_band_mae_db": 0.9817639380088714}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8109282126839231, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "20fcd643d14877b5e4e1da75ee7d9ebdeac30a2c7c6cbeebf5aac25ce87da5f6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9724959152140888, "standard_deviation": 0.0036828245254538023, "minimum": 0.9671662329000815, "maximum": 0.9773763367689119, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6991328297626467, "standard_deviation": 0.03575856102600565, "minimum": 1.6652638563724427, "maximum": 1.7811164621265163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7375663608418084, "standard_deviation": 0.32454922551746745, "minimum": 0.241272824099785, "maximum": 1.2055285566551814, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18779084753126402, "standard_deviation": 0.013785459622983642, "minimum": 0.16969720120622006, "maximum": 0.20869871861329017, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9908362667776736, "standard_deviation": 0.001435459066383839, "minimum": 0.988796332318404, "maximum": 0.9926311651961112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0345552313826731, "standard_deviation": 0.020999944062434845, "minimum": 1.014580817070327, "maximum": 1.0728476568061815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8218479817452747, "standard_deviation": 0.36407602142000767, "minimum": 0.4166056510253551, "maximum": 1.4076055238041398, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10930292159320638, "standard_deviation": 0.007804224549939851, "minimum": 0.09747865398396167, "maximum": 0.11859027611621044, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9806770808280715, "standard_deviation": 0.001326050836404841, "minimum": 0.9793503945226375, "maximum": 0.9824930925994345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.773973173838019, "standard_deviation": 0.021195785462552845, "minimum": 1.7496426132493492, "maximum": 1.8121842883759263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5944704966397985, "standard_deviation": 0.3006530987035577, "minimum": 0.1469635872909109, "maximum": 1.0837575599644464, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16498558627307416, "standard_deviation": 0.008026422336214674, "minimum": 0.15495121620543004, "maximum": 0.17766071171703182, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9931748828441611, "standard_deviation": 0.0013038305773231287, "minimum": 0.9908030191593254, "maximum": 0.994500255552617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.063867211917462, "standard_deviation": 0.015593252975240331, "minimum": 1.0430454903200441, "maximum": 1.0810712071026949, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7085741943741375, "standard_deviation": 0.386000404166396, "minimum": 0.17492439888291325, "maximum": 1.3327127356219144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09571833290565876, "standard_deviation": 0.01134774616352518, "minimum": 0.08522524460554572, "maximum": 0.1187475974447065, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9813957119794641, "standard_deviation": 0.002998827269621575, "minimum": 0.9751432816298362, "maximum": 0.9839160995876526, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7112241268380242, "standard_deviation": 0.06479028794059298, "minimum": 1.5865288781774196, "maximum": 1.7852337095834758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4097704738508521, "standard_deviation": 0.26625244591599295, "minimum": 0.12452411897474118, "maximum": 0.9167505797711956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1634174616211977, "standard_deviation": 0.0133924169628153, "minimum": 0.15124487227728875, "maximum": 0.19214090649548352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9938822760144617, "standard_deviation": 0.0016315433519488577, "minimum": 0.990023883286696, "maximum": 0.9950404549119096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0273513799448677, "standard_deviation": 0.05205082827901262, "minimum": 0.9609183070686738, "maximum": 1.1081943870057778, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5178623854850155, "standard_deviation": 0.3220638858118539, "minimum": 0.1627101715766508, "maximum": 1.0859180518984657, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09421151151344848, "standard_deviation": 0.014677479793998518, "minimum": 0.0822856375833148, "maximum": 0.1279134145122915, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27220349941743616, "standard_deviation": 0.06819083987744061, "minimum": 0.19348047391979506, "maximum": 0.3603973587882878, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.923522026328519, "standard_deviation": 0.008573564151480223, "minimum": 5.9092380305219665, "maximum": 5.9354478842607294, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7284608592124877, "standard_deviation": 0.09272733065725484, "minimum": 0.6196926859457612, "maximum": 0.8757533500085376, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09210503956774044, "standard_deviation": 0.14125937312687153, "minimum": -0.30816881847793187, "maximum": 0.08684041011475734, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9867277893731621, "standard_deviation": 0.0019190545962730456, "minimum": 0.9833683489589992, "maximum": 0.989554330211541, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2055786384099123, "standard_deviation": 0.020444309618528382, "minimum": 1.1831852500755193, "maximum": 1.23759269588862, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4496068712998328, "standard_deviation": 0.2995287159202177, "minimum": 0.1114037114301556, "maximum": 0.8459452151085172, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11978708440441721, "standard_deviation": 0.008871607027100104, "minimum": 0.10703702377274489, "maximum": 0.1362627624435625, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9958551032629563, "standard_deviation": 0.0011124285601169595, "minimum": 0.9934080414872176, "maximum": 0.9968703971760711, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6129608878932089, "standard_deviation": 0.025573990059539, "minimum": 0.590777677011631, "maximum": 0.6706634622790644, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5328394653500597, "standard_deviation": 0.2808679726950029, "minimum": 0.1659726882077855, "maximum": 0.8729026537748984, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06459468860981152, "standard_deviation": 0.00943618249285158, "minimum": 0.05607965911047341, "maximum": 0.08620048106512504, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0002", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2594192028045654, "predicted_mos": 2.2594192028045654, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "989a394a8852f11fccc1a5921eaaf72e0b2fe039a3f1671512a17779c46e63c3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7294010003097355, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.197601279503507, "全帯域_10ms_rms_envelope_nrmse": 0.12712499649552378, "全帯域_10ms_envelope_correlation": 0.9857328224282565, "全帯域_10ms_modulation_mae_db": 0.8564581797950929, "全帯域_40ms_log_rms_mae_db": 0.6543159854009365, "全帯域_40ms_rms_envelope_nrmse": 0.08047924236761271, "全帯域_40ms_envelope_correlation": 0.9944687547187092, "全帯域_40ms_modulation_mae_db": 0.9014168243821494, "300–1000 Hz_10ms_log_rms_mae_db": 1.8888667316114296, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18621891413230446, "300–1000 Hz_10ms_envelope_correlation": 0.9774653527147029, "300–1000 Hz_10ms_modulation_mae_db": 1.0461368727594014, "300–1000 Hz_40ms_log_rms_mae_db": 1.1224716257113219, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12474510949679482, "300–1000 Hz_40ms_envelope_correlation": 0.9914887355696211, "300–1000 Hz_40ms_modulation_mae_db": 1.2867418695354265, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8789556618530119, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.174173355678136, "1000–3000 Hz_10ms_envelope_correlation": 0.9756048760718277, "1000–3000 Hz_10ms_modulation_mae_db": 0.3698023847106276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1129482162172506, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10237019038166303, "1000–3000 Hz_40ms_envelope_correlation": 0.9908093487783466, "1000–3000 Hz_40ms_modulation_mae_db": 0.4981022985321575, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9847987247358192, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16618748827943214, "3000–6000 Hz_10ms_envelope_correlation": 0.9798353815007047, "3000–6000 Hz_10ms_modulation_mae_db": 0.7928898347159145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1329385178482878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09118101972560608, "3000–6000 Hz_40ms_envelope_correlation": 0.993483762127266, "3000–6000 Hz_40ms_modulation_mae_db": 0.9008315477446968, "mr_stft_db": 6.053907709950939, "band_mae_db": 0.3344282562046709, "rms_gain_error_db": -0.29969898992070615, "output_peak": 0.4947182436254951}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1740035092233494, "全帯域_10ms_rms_envelope_nrmse": 0.1124489293887229, "全帯域_10ms_envelope_correlation": 0.9881575547533722, "全帯域_10ms_modulation_mae_db": 0.7823755989968665, "全帯域_40ms_log_rms_mae_db": 0.5982016957122351, "全帯域_40ms_rms_envelope_nrmse": 0.06556958062413314, "全帯域_40ms_envelope_correlation": 0.9954860646574031, "全帯域_40ms_modulation_mae_db": 0.7104004129415312, "300–1000 Hz_10ms_log_rms_mae_db": 1.7740392260138982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14474510641670627, "300–1000 Hz_10ms_envelope_correlation": 0.9851321614827029, "300–1000 Hz_10ms_modulation_mae_db": 1.1120485683109653, "300–1000 Hz_40ms_log_rms_mae_db": 1.0413569969552268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08905092571005872, "300–1000 Hz_40ms_envelope_correlation": 0.9939753632010511, "300–1000 Hz_40ms_modulation_mae_db": 1.0768753315474222, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9106992585504958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19532753299125874, "1000–3000 Hz_10ms_envelope_correlation": 0.9715542545661288, "1000–3000 Hz_10ms_modulation_mae_db": 1.350221054067626, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1537632380372687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12987829204877108, "1000–3000 Hz_40ms_envelope_correlation": 0.9886553210834549, "1000–3000 Hz_40ms_modulation_mae_db": 1.328140128775689, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0346111419408777, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16872489344224198, "3000–6000 Hz_10ms_envelope_correlation": 0.9792294881770408, "3000–6000 Hz_10ms_modulation_mae_db": 0.5363851884870661, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1853694823685192, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0850243867066873, "3000–6000 Hz_40ms_envelope_correlation": 0.9945261670930942, "3000–6000 Hz_40ms_modulation_mae_db": 0.6641124474719424, "mr_stft_db": 6.047506497951844, "band_mae_db": 0.30880002642403376, "rms_gain_error_db": -0.07087197232016282, "output_peak": 0.6624431285562334}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1638178198309452, "全帯域_10ms_rms_envelope_nrmse": 0.11201868895917583, "全帯域_10ms_envelope_correlation": 0.988316692248962, "全帯域_10ms_modulation_mae_db": 0.6101752432617883, "全帯域_40ms_log_rms_mae_db": 0.6115342465236688, "全帯域_40ms_rms_envelope_nrmse": 0.064033151914687, "全帯域_40ms_envelope_correlation": 0.9958201119448808, "全帯域_40ms_modulation_mae_db": 0.564567401400279, "300–1000 Hz_10ms_log_rms_mae_db": 1.8014361738135038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15917691855837646, "300–1000 Hz_10ms_envelope_correlation": 0.9820410993581639, "300–1000 Hz_10ms_modulation_mae_db": 0.2396172890163298, "300–1000 Hz_40ms_log_rms_mae_db": 1.1154062432035525, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09635220252537148, "300–1000 Hz_40ms_envelope_correlation": 0.9931708821316755, "300–1000 Hz_40ms_modulation_mae_db": 0.2969893863149163, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8383300791007988, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18356938136902148, "1000–3000 Hz_10ms_envelope_correlation": 0.9736251159153575, "1000–3000 Hz_10ms_modulation_mae_db": 0.860033964063739, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1101633611656505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10030544529980898, "1000–3000 Hz_40ms_envelope_correlation": 0.9926826410163123, "1000–3000 Hz_40ms_modulation_mae_db": 0.9527485046535625, "3000–6000 Hz_10ms_log_rms_mae_db": 2.013405485881407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16906094582385658, "3000–6000 Hz_10ms_envelope_correlation": 0.979073978971473, "3000–6000 Hz_10ms_modulation_mae_db": 1.3472265471858176, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1416709240097798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0892633328310946, "3000–6000 Hz_40ms_envelope_correlation": 0.9937523023489279, "3000–6000 Hz_40ms_modulation_mae_db": 1.662346503624742, "mr_stft_db": 6.057230578412368, "band_mae_db": 0.26769828624131364, "rms_gain_error_db": -0.13133357305435356, "output_peak": 0.6311914718906977}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1542294109625064, "全帯域_10ms_rms_envelope_nrmse": 0.11123678009760188, "全帯域_10ms_envelope_correlation": 0.9884481812745497, "全帯域_10ms_modulation_mae_db": 0.3607264401227535, "全帯域_40ms_log_rms_mae_db": 0.5593468451700134, "全帯域_40ms_rms_envelope_nrmse": 0.056789778637823, "全帯域_40ms_envelope_correlation": 0.9967052149119506, "全帯域_40ms_modulation_mae_db": 0.4695622745478396, "300–1000 Hz_10ms_log_rms_mae_db": 1.8443195240885326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1568108161010522, "300–1000 Hz_10ms_envelope_correlation": 0.9827570583116951, "300–1000 Hz_10ms_modulation_mae_db": 0.6012472192064598, "300–1000 Hz_40ms_log_rms_mae_db": 1.0679792917090372, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0823868803250373, "300–1000 Hz_40ms_envelope_correlation": 0.995452836448539, "300–1000 Hz_40ms_modulation_mae_db": 0.7301374396716259, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9326956236887027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18094398494046968, "1000–3000 Hz_10ms_envelope_correlation": 0.9743983677483347, "1000–3000 Hz_10ms_modulation_mae_db": 0.6122000955497886, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1506306723611386, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10627964196820511, "1000–3000 Hz_40ms_envelope_correlation": 0.9914542360527381, "1000–3000 Hz_40ms_modulation_mae_db": 0.8329611261742316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9259305322064932, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15528684232387976, "3000–6000 Hz_10ms_envelope_correlation": 0.9831868924589837, "3000–6000 Hz_10ms_modulation_mae_db": 0.2954505341902045, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1037976662459006, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08703883789833579, "3000–6000 Hz_40ms_envelope_correlation": 0.9945139417479318, "3000–6000 Hz_40ms_modulation_mae_db": 0.3217100943211864, "mr_stft_db": 6.054365525199966, "band_mae_db": 0.19757828357614304, "rms_gain_error_db": -0.08844703712263315, "output_peak": 0.58199240940651}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1516228834605415, "全帯域_10ms_rms_envelope_nrmse": 0.11087305646288148, "全帯域_10ms_envelope_correlation": 0.9884951527804959, "全帯域_10ms_modulation_mae_db": 0.7550425674342621, "全帯域_40ms_log_rms_mae_db": 0.6326613560514318, "全帯域_40ms_rms_envelope_nrmse": 0.05414887917013127, "全帯域_40ms_envelope_correlation": 0.99689630918172, "全帯域_40ms_modulation_mae_db": 0.813315549256402, "300–1000 Hz_10ms_log_rms_mae_db": 1.8364956318294676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14275004330269814, "300–1000 Hz_10ms_envelope_correlation": 0.9857424341968013, "300–1000 Hz_10ms_modulation_mae_db": 0.46842838527588243, "300–1000 Hz_40ms_log_rms_mae_db": 1.0706426560096143, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07497983189595583, "300–1000 Hz_40ms_envelope_correlation": 0.995790542253463, "300–1000 Hz_40ms_modulation_mae_db": 0.559371826331725, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9082099163161255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19094161811384883, "1000–3000 Hz_10ms_envelope_correlation": 0.9707118568678628, "1000–3000 Hz_10ms_modulation_mae_db": 0.7072224897036286, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1153356988648495, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10218277036641776, "1000–3000 Hz_40ms_envelope_correlation": 0.9912671033041447, "1000–3000 Hz_40ms_modulation_mae_db": 0.7696896518936059, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0258108558016787, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17413778802496702, "3000–6000 Hz_10ms_envelope_correlation": 0.9782598036021246, "3000–6000 Hz_10ms_modulation_mae_db": 0.9529294922726423, "3000–6000 Hz_40ms_log_rms_mae_db": 1.147375945136098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08987477842153065, "3000–6000 Hz_40ms_envelope_correlation": 0.9944665399720605, "3000–6000 Hz_40ms_modulation_mae_db": 1.2938022200399788, "mr_stft_db": 6.052779638717844, "band_mae_db": 0.2707171161849603, "rms_gain_error_db": -0.020744863235892846, "output_peak": 0.784593605266972}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1966147808591148, "全帯域_10ms_rms_envelope_nrmse": 0.11223033257429588, "全帯域_10ms_envelope_correlation": 0.9882168371540003, "全帯域_10ms_modulation_mae_db": 0.39535548352121186, "全帯域_40ms_log_rms_mae_db": 0.6266235148821998, "全帯域_40ms_rms_envelope_nrmse": 0.05990679722684484, "全帯域_40ms_envelope_correlation": 0.9962806507703068, "全帯域_40ms_modulation_mae_db": 0.5436228084590115, "300–1000 Hz_10ms_log_rms_mae_db": 1.8941621626027139, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1848020988279743, "300–1000 Hz_10ms_envelope_correlation": 0.9761811048508633, "300–1000 Hz_10ms_modulation_mae_db": 0.5598162254769128, "300–1000 Hz_40ms_log_rms_mae_db": 1.1699411687495007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11260501432882294, "300–1000 Hz_40ms_envelope_correlation": 0.9915471468010324, "300–1000 Hz_40ms_modulation_mae_db": 0.5849977441934203, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8805855789280987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23525107516962987, "1000–3000 Hz_10ms_envelope_correlation": 0.9606466796975898, "1000–3000 Hz_10ms_modulation_mae_db": 0.43679839375906365, "1000–3000 Hz_40ms_log_rms_mae_db": 1.107460241605078, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15983493536943852, "1000–3000 Hz_40ms_envelope_correlation": 0.9815677660869174, "1000–3000 Hz_40ms_modulation_mae_db": 0.4718011090845331, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0252268549795045, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1717044597882782, "3000–6000 Hz_10ms_envelope_correlation": 0.9788272420340834, "3000–6000 Hz_10ms_modulation_mae_db": 1.0938979062626326, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1806109849201343, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09424750262444827, "3000–6000 Hz_40ms_envelope_correlation": 0.993690400100618, "3000–6000 Hz_40ms_modulation_mae_db": 1.3059233247812296, "mr_stft_db": 6.060164703441498, "band_mae_db": 0.43080351745993095, "rms_gain_error_db": -0.09807907512046093, "output_peak": 0.8968446309234513}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.145037918504134, "全帯域_10ms_rms_envelope_nrmse": 0.12325866512624967, "全帯域_10ms_envelope_correlation": 0.9860920819542527, "全帯域_10ms_modulation_mae_db": 0.7208330980832839, "全帯域_40ms_log_rms_mae_db": 0.6015045336069934, "全帯域_40ms_rms_envelope_nrmse": 0.07852247694010474, "全帯域_40ms_envelope_correlation": 0.9939920002956237, "全帯域_40ms_modulation_mae_db": 0.6350264210548903, "300–1000 Hz_10ms_log_rms_mae_db": 1.869844736550502, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18639634422839582, "300–1000 Hz_10ms_envelope_correlation": 0.9761359770461928, "300–1000 Hz_10ms_modulation_mae_db": 0.791138827851621, "300–1000 Hz_40ms_log_rms_mae_db": 1.118660975532529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13064345580918707, "300–1000 Hz_40ms_envelope_correlation": 0.988553120712597, "300–1000 Hz_40ms_modulation_mae_db": 0.9783315603128345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8803671666940054, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18835314181171886, "1000–3000 Hz_10ms_envelope_correlation": 0.9716281521601887, "1000–3000 Hz_10ms_modulation_mae_db": 0.8590124357089604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1228875030501169, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10999479335982204, "1000–3000 Hz_40ms_envelope_correlation": 0.9896060667018022, "1000–3000 Hz_40ms_modulation_mae_db": 0.909120886073398, "3000–6000 Hz_10ms_log_rms_mae_db": 1.987125050727128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15708019308788665, "3000–6000 Hz_10ms_envelope_correlation": 0.9820266329670287, "3000–6000 Hz_10ms_modulation_mae_db": 0.5925988098699586, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1076053575557903, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08236712282805117, "3000–6000 Hz_40ms_envelope_correlation": 0.994461503328019, "3000–6000 Hz_40ms_modulation_mae_db": 0.5765484628450145, "mr_stft_db": 6.040664827266656, "band_mae_db": 0.29977300355374276, "rms_gain_error_db": -0.20017999097620415, "output_peak": 0.5747858331689614}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1800218147926658, "全帯域_10ms_rms_envelope_nrmse": 0.12489649008825834, "全帯域_10ms_envelope_correlation": 0.9854044262998204, "全帯域_10ms_modulation_mae_db": 0.47433848816693946, "全帯域_40ms_log_rms_mae_db": 0.5909717775767618, "全帯域_40ms_rms_envelope_nrmse": 0.06623918123701404, "全帯域_40ms_envelope_correlation": 0.9953353773567556, "全帯域_40ms_modulation_mae_db": 0.642225650623257, "300–1000 Hz_10ms_log_rms_mae_db": 1.765434885326452, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19255657265726114, "300–1000 Hz_10ms_envelope_correlation": 0.9734957435062755, "300–1000 Hz_10ms_modulation_mae_db": 0.3824162126596522, "300–1000 Hz_40ms_log_rms_mae_db": 1.0140291551654772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10007159556643504, "300–1000 Hz_40ms_envelope_correlation": 0.9925507660110965, "300–1000 Hz_40ms_modulation_mae_db": 0.6456569449330039, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8359751095200079, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1821872573407053, "1000–3000 Hz_10ms_envelope_correlation": 0.9738595311979119, "1000–3000 Hz_10ms_modulation_mae_db": 0.20868771217093438, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0928554013598188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09780587083526096, "1000–3000 Hz_40ms_envelope_correlation": 0.9917349543340079, "1000–3000 Hz_40ms_modulation_mae_db": 0.39497108492510047, "3000–6000 Hz_10ms_log_rms_mae_db": 2.030752445681892, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18397179879451275, "3000–6000 Hz_10ms_envelope_correlation": 0.9756767429753959, "3000–6000 Hz_10ms_modulation_mae_db": 1.304221054604848, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1481812854032547, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11289915808188213, "3000–6000 Hz_40ms_envelope_correlation": 0.989927458195055, "3000–6000 Hz_40ms_modulation_mae_db": 1.6193643085662381, "mr_stft_db": 6.055093886702674, "band_mae_db": 0.1898689282107148, "rms_gain_error_db": -0.006513724740710415, "output_peak": 0.573610751655163}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.4288026563432865, "spectral_convergence": 0.45831952191120223, "spectral_convergence_status": "ok", "amplitude_floor": 1.602383167696498e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.808833978669372, "spectral_convergence": 0.4258408311916586, "spectral_convergence_status": "ok", "amplitude_floor": 1.3018377539140758e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.924086494840156, "spectral_convergence": 0.45472203977835834, "spectral_convergence_status": "ok", "amplitude_floor": 8.15121073207966e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00018979827012669627, "output_power": 0.00020370281181047114, "delta_db": 0.3070477399194983, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007146011617002843, "output_power": 0.0006975476529620124, "delta_db": -0.10489836100279785, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006180173350849692, "output_power": 0.0004994016893355777, "delta_db": -0.925506498941619, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00021309861976674557, "output_power": 0.00019354103847406002, "delta_db": -0.4180756981739964, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003708463313924785, "output_power": 0.0003690786925802653, "delta_db": -0.020750135964136683, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001752804045595543, "output_power": 0.00016622802946796814, "delta_db": -0.2302911032259774, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3344282562046709, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.29969898992070615, "pitch": null, "output_rms_dbfs": -26.717223730006886, "output_peak": 0.4947182436254951, "reference_rms_dbfs": -26.41752474008618, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.053907709950939}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002801281523487693, "log_rms_mae_db": 1.197601279503507, "rms_envelope_nrmse": 0.12712499649552378, "envelope_correlation": 0.9857328224282565, "rms_gain_error_db": -0.2996964833180394, "reference_mean_power": 0.0022835195719053173, "output_mean_power": 0.002131253403773403}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.817352658591746e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000450774313706133, "output_power": 0.0004186594936167504, "reference_active": true, "delta_db": -0.32098217033741505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021888119814572043, "output_power": 0.0002110502122396679, "reference_active": true, "delta_db": -0.15822664122000032}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.753791744185585e-05, "output_power": 2.319816334167713e-05, "reference_active": true, "delta_db": -2.090165727827863}], "active_band_count": 3, "active_band_mae_db": 0.8564581797950929}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022259441477048914, "log_rms_mae_db": 0.6543159854009365, "rms_envelope_nrmse": 0.08047924236761271, "envelope_correlation": 0.9944687547187092, "rms_gain_error_db": -0.29976306402531605, "reference_mean_power": 0.002294705859570105, "output_mean_power": 0.002141660950898454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.266680044598684e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042660839170957063, "output_power": 0.00039289077880414125, "reference_active": true, "delta_db": -0.35757558112753596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012272337901795565, "output_power": 0.0001182689541874878, "reference_active": true, "delta_db": -0.16056547957282963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.241371364832561e-06, "output_power": 1.9593762965013963e-06, "reference_active": true, "delta_db": -2.1861094124460827}], "active_band_count": 3, "active_band_mae_db": 0.9014168243821494}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002244694917839089, "log_rms_mae_db": 1.8888667316114296, "rms_envelope_nrmse": 0.18621891413230446, "envelope_correlation": 0.9774653527147029, "rms_gain_error_db": -0.6206813589919506, "reference_mean_power": 0.001082154267593255, "output_mean_power": 0.0009380393143733161}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.600154760198564e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003310843525581766, "output_power": 0.00029071679625644354, "reference_active": true, "delta_db": -0.5646853196258712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.047815482126218e-05, "output_power": 7.890575904825386e-05, "reference_active": true, "delta_db": -0.5943503320096243}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6938491072881828e-05, "output_power": 1.0738341788377573e-05, "reference_active": true, "delta_db": -1.9793749666427083}], "active_band_count": 3, "active_band_mae_db": 1.0461368727594014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018235367957614062, "log_rms_mae_db": 1.1224716257113219, "rms_envelope_nrmse": 0.12474510949679482, "envelope_correlation": 0.9914887355696211, "rms_gain_error_db": -0.620690657824221, "reference_mean_power": 0.0010875823794426604, "output_mean_power": 0.000942742523607679}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.047574386178106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00032394555601725244, "output_power": 0.0002838882725544803, "reference_active": true, "delta_db": -0.5732457438037757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.774463594481584e-05, "output_power": 4.169931452342972e-05, "reference_active": true, "delta_db": -0.5879567031492552}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5157059923891315e-06, "output_power": 8.141654177032026e-07, "reference_active": true, "delta_db": -2.6990231616532485}], "active_band_count": 3, "active_band_mae_db": 1.2867418695354265}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014040871531833862, "log_rms_mae_db": 1.8789556618530119, "rms_envelope_nrmse": 0.174173355678136, "envelope_correlation": 0.9756048760718277, "rms_gain_error_db": -0.17405439277655518, "reference_mean_power": 0.00036159326722347827, "output_mean_power": 0.00034738806705518927}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2796709554971252e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010222442860363171, "output_power": 0.00010002469377681775, "reference_active": true, "delta_db": -0.09447461218230853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.551583786781791e-05, "output_power": 4.4644259063998845e-05, "reference_active": true, "delta_db": -0.08396922134837431}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1818852015306123e-05, "output_power": 9.538475373811664e-06, "reference_active": true, "delta_db": -0.9309633206012}], "active_band_count": 3, "active_band_mae_db": 0.3698023847106276}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010845211232047094, "log_rms_mae_db": 1.1129482162172506, "rms_envelope_nrmse": 0.10237019038166303, "envelope_correlation": 0.9908093487783466, "rms_gain_error_db": -0.17405250086062557, "reference_mean_power": 0.00036337479954251316, "output_mean_power": 0.00034909976391258414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.941433739824422e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.608975832570782e-05, "output_power": 9.533551605324458e-05, "reference_active": true, "delta_db": -0.03422379339765237}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4640366344588194e-05, "output_power": 2.37024232380646e-05, "reference_active": true, "delta_db": -0.16854411725222873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.520284367741145e-07, "output_power": 7.071247241273308e-07, "reference_active": true, "delta_db": -1.2915389849465915}], "active_band_count": 3, "active_band_mae_db": 0.4981022985321575}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012084003773539908, "log_rms_mae_db": 1.9847987247358192, "rms_envelope_nrmse": 0.16618748827943214, "envelope_correlation": 0.9798353815007047, "rms_gain_error_db": -0.19566020018397862, "reference_mean_power": 0.00030833628390941923, "output_mean_power": 0.0002947532578288636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2964446771205807e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7472950448048514e-05, "output_power": 4.968897561937294e-05, "reference_active": true, "delta_db": 0.1981381948910904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.978240331431912e-05, "output_power": 4.6781205960503296e-05, "reference_active": true, "delta_db": -0.2700444576340092}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3740568823166797e-05, "output_power": 8.850259941527826e-06, "reference_active": true, "delta_db": -1.9104868516226439}], "active_band_count": 3, "active_band_mae_db": 0.7928898347159145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010527157253957595, "log_rms_mae_db": 1.1329385178482878, "rms_envelope_nrmse": 0.09118101972560608, "envelope_correlation": 0.993483762127266, "rms_gain_error_db": -0.19562632287792672, "reference_mean_power": 0.0003098592063705908, "output_mean_power": 0.0002962114021404811}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.056454468513196e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.721882315895364e-05, "output_power": 5.803901238453204e-05, "reference_active": true, "delta_db": 0.061810924336029194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8898944201462896e-05, "output_power": 2.5347317297451998e-05, "reference_active": true, "delta_db": -0.5694997510494006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2532553850736713e-06, "output_power": 7.778953628640381e-07, "reference_active": true, "delta_db": -2.0711839678486603}], "active_band_count": 3, "active_band_mae_db": 0.9008315477446968}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.804417025238634, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "19de1bc019946fd4de2e3d3ae10eabafb059f1c9dac0cd3a2937b592432598fe", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9715036042781502, "standard_deviation": 0.0046868972692612415, "minimum": 0.9606466796975898, "maximum": 0.9756048760718277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8832272993314059, "standard_deviation": 0.034033367173170515, "minimum": 1.8359751095200079, "maximum": 1.9326956236887027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6754973162167961, "standard_deviation": 0.35783765150407054, "minimum": 0.20868771217093438, "maximum": 1.350221054067626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1913434184268486, "standard_deviation": 0.018898559646079574, "minimum": 0.174173355678136, "maximum": 0.23525107516962987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9897221796697155, "standard_deviation": 0.003526035179278716, "minimum": 0.9815677660869174, "maximum": 0.9926826410163123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1207555415826465, "standard_deviation": 0.021193944082318963, "minimum": 1.0928554013598188, "maximum": 1.1537632380372687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7696918487640347, "standard_deviation": 0.30977713326594536, "minimum": 0.39497108492510047, "maximum": 1.328140128775689, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11358149245367344, "standard_deviation": 0.02121958635101742, "minimum": 0.09780587083526096, "maximum": 0.15983493536943852, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9795145203358544, "standard_deviation": 0.0022975297109621255, "minimum": 0.9756767429753959, "maximum": 0.9831868924589837, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.00345763649435, "standard_deviation": 0.03667118906013893, "minimum": 1.9259305322064932, "maximum": 2.0346111419408777, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8644499209486355, "standard_deviation": 0.37746477161971315, "minimum": 0.2954505341902045, "maximum": 1.3472265471858176, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16826930119563188, "standard_deviation": 0.009194070078042857, "minimum": 0.15528684232387976, "maximum": 0.18397179879451275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9936027593641215, "standard_deviation": 0.001545432733360452, "minimum": 0.989927458195055, "maximum": 0.9945261670930942, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1434437704359706, "standard_deviation": 0.029632949641289697, "minimum": 1.1037976662459006, "maximum": 1.1853694823685192, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0430798636743785, "standard_deviation": 0.4999171496833032, "minimum": 0.3217100943211864, "maximum": 1.662346503624742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0914870173897045, "standard_deviation": 0.009399560345175436, "minimum": 0.08236712282805117, "maximum": 0.11289915808188213, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9798688664334247, "standard_deviation": 0.0046165686694699904, "minimum": 0.9734957435062755, "maximum": 0.9857424341968013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8343248839795625, "standard_deviation": 0.04983320420173833, "minimum": 1.765434885326452, "maximum": 1.8941621626027139, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6501062000696531, "standard_deviation": 0.31025948374018214, "minimum": 0.2396172890163298, "maximum": 1.1120485683109653, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1691821017780961, "standard_deviation": 0.02044472049612244, "minimum": 0.14275004330269814, "maximum": 0.19255657265726114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9928161741411344, "standard_deviation": 0.0023569041554920905, "minimum": 0.988553120712597, "maximum": 0.995790542253463, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0900610141295326, "standard_deviation": 0.050581229208722635, "minimum": 1.0140291551654772, "maximum": 1.1699411687495007, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7698877628550468, "standard_deviation": 0.3215977720699964, "minimum": 0.2969893863149163, "maximum": 1.2867418695354265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10135437695720789, "standard_deviation": 0.01987452818996844, "minimum": 0.07497983189595583, "maximum": 0.13064345580918707, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2874584272319388, "standard_deviation": 0.07712140483178964, "minimum": 0.1898689282107148, "maximum": 0.43080351745993095, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.052714170955474, "standard_deviation": 0.006071405477739543, "minimum": 6.040664827266656, "maximum": 6.060164703441498, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6500225093116854, "standard_deviation": 0.130814068113546, "minimum": 0.4947182436254951, "maximum": 0.8968446309234513, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11448365331139049, "standard_deviation": 0.0965297611600009, "minimum": -0.29969898992070615, "maximum": -0.006513724740710415, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9873579686117138, "standard_deviation": 0.001354241610587059, "minimum": 0.9854044262998204, "maximum": 0.9884951527804959, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1703686771420956, "standard_deviation": 0.020118610505527268, "minimum": 1.145037918504134, "maximum": 1.197601279503507, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6194131374227748, "standard_deviation": 0.18886166474746996, "minimum": 0.3607264401227535, "maximum": 0.8564581797950929, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11676099239908871, "standard_deviation": 0.00699605575961039, "minimum": 0.11087305646288148, "maximum": 0.12712499649552378, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9956230604796688, "standard_deviation": 0.0010252374840671841, "minimum": 0.9939920002956237, "maximum": 0.99689630918172, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6093949943655301, "standard_deviation": 0.02901329449249622, "minimum": 0.5593468451700134, "maximum": 0.6543159854009365, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.66001716783317, "standard_deviation": 0.14351958851286004, "minimum": 0.4695622745478396, "maximum": 0.9014168243821494, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06571113601479384, "standard_deviation": 0.009504199608843066, "minimum": 0.05414887917013127, "maximum": 0.08047924236761271, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0002", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.9397056102752686, "predicted_mos": 1.9397056102752686, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3784539310c2ac675c555e088e77e7ee6146db7a36ed77c6a7760a5f6d28f948", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.949279199820012, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1736377336869228, "全帯域_10ms_rms_envelope_nrmse": 0.13396618375928343, "全帯域_10ms_envelope_correlation": 0.9841574093437943, "全帯域_10ms_modulation_mae_db": 0.8006954484238834, "全帯域_40ms_log_rms_mae_db": 0.6015127383161037, "全帯域_40ms_rms_envelope_nrmse": 0.08080946961149879, "全帯域_40ms_envelope_correlation": 0.9946704727368311, "全帯域_40ms_modulation_mae_db": 0.9642809828254574, "300–1000 Hz_10ms_log_rms_mae_db": 1.7283853273750973, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17757317383740195, "300–1000 Hz_10ms_envelope_correlation": 0.9792647773149978, "300–1000 Hz_10ms_modulation_mae_db": 0.6249336022402862, "300–1000 Hz_40ms_log_rms_mae_db": 1.0086765536762115, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12065583846618042, "300–1000 Hz_40ms_envelope_correlation": 0.9915759204432383, "300–1000 Hz_40ms_modulation_mae_db": 0.8297659023949894, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8589581154039458, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20427195699333467, "1000–3000 Hz_10ms_envelope_correlation": 0.9676193690675348, "1000–3000 Hz_10ms_modulation_mae_db": 1.1668287951186616, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0904260022109065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10823259573377161, "1000–3000 Hz_40ms_envelope_correlation": 0.9924569284696215, "1000–3000 Hz_40ms_modulation_mae_db": 1.2457421311336567, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9267688321010303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1601462680822538, "3000–6000 Hz_10ms_envelope_correlation": 0.9813418985522128, "3000–6000 Hz_10ms_modulation_mae_db": 0.7080003242153836, "3000–6000 Hz_40ms_log_rms_mae_db": 1.050306236642325, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08334829987034167, "3000–6000 Hz_40ms_envelope_correlation": 0.9946334352783891, "3000–6000 Hz_40ms_modulation_mae_db": 0.9443520392708046, "mr_stft_db": 6.0239693299671275, "band_mae_db": 0.35431377478383946, "rms_gain_error_db": -0.32642749919486025, "output_peak": 0.57294640495255}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1681081303823, "全帯域_10ms_rms_envelope_nrmse": 0.124844221706577, "全帯域_10ms_envelope_correlation": 0.9853852680043154, "全帯域_10ms_modulation_mae_db": 0.529843129977441, "全帯域_40ms_log_rms_mae_db": 0.6190156638488387, "全帯域_40ms_rms_envelope_nrmse": 0.06351344244270328, "全帯域_40ms_envelope_correlation": 0.995752870582223, "全帯域_40ms_modulation_mae_db": 0.5956470192256338, "300–1000 Hz_10ms_log_rms_mae_db": 1.8715899614693867, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1891462939240159, "300–1000 Hz_10ms_envelope_correlation": 0.9744546519872987, "300–1000 Hz_10ms_modulation_mae_db": 0.23057025404666756, "300–1000 Hz_40ms_log_rms_mae_db": 1.156575702316433, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10069397501061013, "300–1000 Hz_40ms_envelope_correlation": 0.992514057790343, "300–1000 Hz_40ms_modulation_mae_db": 0.5063399142550621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.795404575528411, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17174642627995085, "1000–3000 Hz_10ms_envelope_correlation": 0.9765516595170539, "1000–3000 Hz_10ms_modulation_mae_db": 0.5203209060329334, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0155041022704236, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09167463799149057, "1000–3000 Hz_40ms_envelope_correlation": 0.9932644025540845, "1000–3000 Hz_40ms_modulation_mae_db": 0.5459433531299663, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9189869933919435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16041892094306484, "3000–6000 Hz_10ms_envelope_correlation": 0.9812327324368102, "3000–6000 Hz_10ms_modulation_mae_db": 1.2430095145168503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.095423381754638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08809417138200139, "3000–6000 Hz_40ms_envelope_correlation": 0.993721487270245, "3000–6000 Hz_40ms_modulation_mae_db": 1.5886702176733973, "mr_stft_db": 6.026499371566835, "band_mae_db": 0.2824207302182103, "rms_gain_error_db": -0.05919630600536169, "output_peak": 0.6215571063455367}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1699119866822303, "全帯域_10ms_rms_envelope_nrmse": 0.1089678571157291, "全帯域_10ms_envelope_correlation": 0.9889412423249315, "全帯域_10ms_modulation_mae_db": 0.6157016600820485, "全帯域_40ms_log_rms_mae_db": 0.6000635852718593, "全帯域_40ms_rms_envelope_nrmse": 0.06442660203627297, "全帯域_40ms_envelope_correlation": 0.9957181032562201, "全帯域_40ms_modulation_mae_db": 0.6741075211669031, "300–1000 Hz_10ms_log_rms_mae_db": 1.7419989861495275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1690368650924642, "300–1000 Hz_10ms_envelope_correlation": 0.9805924447962844, "300–1000 Hz_10ms_modulation_mae_db": 0.9186072010556011, "300–1000 Hz_40ms_log_rms_mae_db": 1.013317576430573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1198373546458743, "300–1000 Hz_40ms_envelope_correlation": 0.9905275721835355, "300–1000 Hz_40ms_modulation_mae_db": 1.127909240872538, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8695790478985148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18524907291978876, "1000–3000 Hz_10ms_envelope_correlation": 0.9735069421425909, "1000–3000 Hz_10ms_modulation_mae_db": 0.22779549163646542, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0846272878689147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10555351421991975, "1000–3000 Hz_40ms_envelope_correlation": 0.9907241884341151, "1000–3000 Hz_40ms_modulation_mae_db": 0.4359742945251554, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9209421074077624, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15995243229691058, "3000–6000 Hz_10ms_envelope_correlation": 0.9812700332939067, "3000–6000 Hz_10ms_modulation_mae_db": 0.9237311880575488, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0878601143292443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08230626552897746, "3000–6000 Hz_40ms_envelope_correlation": 0.9946204003492588, "3000–6000 Hz_40ms_modulation_mae_db": 1.2110126728513217, "mr_stft_db": 6.0078962988286575, "band_mae_db": 0.29979472372821053, "rms_gain_error_db": -0.09852111793544799, "output_peak": 0.6856092140139531}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1320206148689822, "全帯域_10ms_rms_envelope_nrmse": 0.1096363278231586, "全帯域_10ms_envelope_correlation": 0.9888264128281586, "全帯域_10ms_modulation_mae_db": 0.7238187260419457, "全帯域_40ms_log_rms_mae_db": 0.5571736044749863, "全帯域_40ms_rms_envelope_nrmse": 0.05522398289917306, "全帯域_40ms_envelope_correlation": 0.9969630893385566, "全帯域_40ms_modulation_mae_db": 0.8056696212011394, "300–1000 Hz_10ms_log_rms_mae_db": 1.764142248381339, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16210466558909756, "300–1000 Hz_10ms_envelope_correlation": 0.9814050384585992, "300–1000 Hz_10ms_modulation_mae_db": 0.41257246780812973, "300–1000 Hz_40ms_log_rms_mae_db": 1.0346823131960976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0814656120233339, "300–1000 Hz_40ms_envelope_correlation": 0.9953635214412493, "300–1000 Hz_40ms_modulation_mae_db": 0.7098368603020414, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8875546380200312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1881848159951027, "1000–3000 Hz_10ms_envelope_correlation": 0.971705272158689, "1000–3000 Hz_10ms_modulation_mae_db": 0.6846712152143489, "1000–3000 Hz_40ms_log_rms_mae_db": 1.123360037667984, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10376015129729803, "1000–3000 Hz_40ms_envelope_correlation": 0.9912062263950435, "1000–3000 Hz_40ms_modulation_mae_db": 1.003797481062853, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9531112092848262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13788602237915484, "3000–6000 Hz_10ms_envelope_correlation": 0.9861299003986606, "3000–6000 Hz_10ms_modulation_mae_db": 0.7806915808567405, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0715880892066956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0686115381564989, "3000–6000 Hz_40ms_envelope_correlation": 0.9961793162233622, "3000–6000 Hz_40ms_modulation_mae_db": 0.9684425066427759, "mr_stft_db": 6.035177103768585, "band_mae_db": 0.18696612250362932, "rms_gain_error_db": -0.1272204012444859, "output_peak": 0.6895419986053196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1760092663111998, "全帯域_10ms_rms_envelope_nrmse": 0.11245108701036857, "全帯域_10ms_envelope_correlation": 0.9881536631646524, "全帯域_10ms_modulation_mae_db": 0.5632925905693056, "全帯域_40ms_log_rms_mae_db": 0.592931007605468, "全帯域_40ms_rms_envelope_nrmse": 0.06219899243548821, "全帯域_40ms_envelope_correlation": 0.9959432069286237, "全帯域_40ms_modulation_mae_db": 0.7347532264251863, "300–1000 Hz_10ms_log_rms_mae_db": 1.806796433032643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15027136687359027, "300–1000 Hz_10ms_envelope_correlation": 0.9839893612858291, "300–1000 Hz_10ms_modulation_mae_db": 0.4234307679046106, "300–1000 Hz_40ms_log_rms_mae_db": 1.128363818193965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09433612632767452, "300–1000 Hz_40ms_envelope_correlation": 0.993368959051368, "300–1000 Hz_40ms_modulation_mae_db": 0.6920300120670936, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8147300036441476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1824699474845257, "1000–3000 Hz_10ms_envelope_correlation": 0.9735801201807546, "1000–3000 Hz_10ms_modulation_mae_db": 0.5473884055682815, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9777535155314313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08633589181170796, "1000–3000 Hz_40ms_envelope_correlation": 0.9934555072532345, "1000–3000 Hz_40ms_modulation_mae_db": 0.7353359778619204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9233346299697982, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15087945905345415, "3000–6000 Hz_10ms_envelope_correlation": 0.9833666752750492, "3000–6000 Hz_10ms_modulation_mae_db": 0.5225630433310603, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0826675019165999, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07484923246878204, "3000–6000 Hz_40ms_envelope_correlation": 0.9955884620593768, "3000–6000 Hz_40ms_modulation_mae_db": 0.7121170781370152, "mr_stft_db": 6.029997827795541, "band_mae_db": 0.12387764855894072, "rms_gain_error_db": -0.07293274641453551, "output_peak": 0.6358760659474776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1811432726234787, "全帯域_10ms_rms_envelope_nrmse": 0.11894204732763267, "全帯域_10ms_envelope_correlation": 0.9868628747396367, "全帯域_10ms_modulation_mae_db": 0.24855011942923397, "全帯域_40ms_log_rms_mae_db": 0.6356349348334183, "全帯域_40ms_rms_envelope_nrmse": 0.06789485521729251, "全帯域_40ms_envelope_correlation": 0.9953789722809516, "全帯域_40ms_modulation_mae_db": 0.31517466106394526, "300–1000 Hz_10ms_log_rms_mae_db": 1.8134671509922444, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15811938787824817, "300–1000 Hz_10ms_envelope_correlation": 0.9823533251761516, "300–1000 Hz_10ms_modulation_mae_db": 0.20124942061960635, "300–1000 Hz_40ms_log_rms_mae_db": 1.0515920054228316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09344388305260419, "300–1000 Hz_40ms_envelope_correlation": 0.9937396420521963, "300–1000 Hz_40ms_modulation_mae_db": 0.296965252738916, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8581566962194185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1938903423197165, "1000–3000 Hz_10ms_envelope_correlation": 0.9700993987939238, "1000–3000 Hz_10ms_modulation_mae_db": 0.3986316332749619, "1000–3000 Hz_40ms_log_rms_mae_db": 1.106188255778881, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11596886579072917, "1000–3000 Hz_40ms_envelope_correlation": 0.98909137181642, "1000–3000 Hz_40ms_modulation_mae_db": 0.5640754738301985, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9646758400140023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14635305411619504, "3000–6000 Hz_10ms_envelope_correlation": 0.9843469855327196, "3000–6000 Hz_10ms_modulation_mae_db": 0.525586180437525, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1108506691335147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07311614672329925, "3000–6000 Hz_40ms_envelope_correlation": 0.9957392568299046, "3000–6000 Hz_40ms_modulation_mae_db": 0.6393660097186231, "mr_stft_db": 6.0276894369370035, "band_mae_db": 0.29677188910249647, "rms_gain_error_db": -0.1280013673833095, "output_peak": 0.6860002707680022}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1741542041839383, "全帯域_10ms_rms_envelope_nrmse": 0.1113365420995681, "全帯域_10ms_envelope_correlation": 0.9884509251999679, "全帯域_10ms_modulation_mae_db": 0.3511184775759529, "全帯域_40ms_log_rms_mae_db": 0.5773515138273595, "全帯域_40ms_rms_envelope_nrmse": 0.056920106322292684, "全帯域_40ms_envelope_correlation": 0.9967429747881494, "全帯域_40ms_modulation_mae_db": 0.4674481033110061, "300–1000 Hz_10ms_log_rms_mae_db": 1.7694499614834993, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16186576459358168, "300–1000 Hz_10ms_envelope_correlation": 0.9815544690447158, "300–1000 Hz_10ms_modulation_mae_db": 0.2697223786600289, "300–1000 Hz_40ms_log_rms_mae_db": 1.0310257218202998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08441475580231225, "300–1000 Hz_40ms_envelope_correlation": 0.9951898936905476, "300–1000 Hz_40ms_modulation_mae_db": 0.4309175929115321, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8181699372553397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18397945449276035, "1000–3000 Hz_10ms_envelope_correlation": 0.9729752322853562, "1000–3000 Hz_10ms_modulation_mae_db": 0.20545316750302992, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0481480994772734, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09678342905694298, "1000–3000 Hz_40ms_envelope_correlation": 0.9917355421451685, "1000–3000 Hz_40ms_modulation_mae_db": 0.38399225286509603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9432211006778304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16830683119304424, "3000–6000 Hz_10ms_envelope_correlation": 0.9794832934608912, "3000–6000 Hz_10ms_modulation_mae_db": 0.8164238777662943, "3000–6000 Hz_40ms_log_rms_mae_db": 1.106538782067876, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0930744664549428, "3000–6000 Hz_40ms_envelope_correlation": 0.9935322042611311, "3000–6000 Hz_40ms_modulation_mae_db": 0.9114903709410007, "mr_stft_db": 6.017078531232165, "band_mae_db": 0.19847716625936562, "rms_gain_error_db": -0.10933899462797608, "output_peak": 0.6363557712306146}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.161313579324371, "全帯域_10ms_rms_envelope_nrmse": 0.10742754876342131, "全帯域_10ms_envelope_correlation": 0.9891968161550231, "全帯域_10ms_modulation_mae_db": 0.5843131556116817, "全帯域_40ms_log_rms_mae_db": 0.5793756337225534, "全帯域_40ms_rms_envelope_nrmse": 0.05809486026940518, "全帯域_40ms_envelope_correlation": 0.9964448224983765, "全帯域_40ms_modulation_mae_db": 0.6348369612740473, "300–1000 Hz_10ms_log_rms_mae_db": 1.754472096085289, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15013864875663502, "300–1000 Hz_10ms_envelope_correlation": 0.9839792144837486, "300–1000 Hz_10ms_modulation_mae_db": 0.4524417579200839, "300–1000 Hz_40ms_log_rms_mae_db": 0.9969798922212466, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07503450068994869, "300–1000 Hz_40ms_envelope_correlation": 0.9958062757245808, "300–1000 Hz_40ms_modulation_mae_db": 0.58665398904226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8926033771607456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20610683417380987, "1000–3000 Hz_10ms_envelope_correlation": 0.9667272792288281, "1000–3000 Hz_10ms_modulation_mae_db": 0.14926979016461497, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1074842151434785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1266733737752765, "1000–3000 Hz_40ms_envelope_correlation": 0.9863737627889521, "1000–3000 Hz_40ms_modulation_mae_db": 0.1812766505911891, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9413975790699203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15455675101608132, "3000–6000 Hz_10ms_envelope_correlation": 0.9825174340958581, "3000–6000 Hz_10ms_modulation_mae_db": 0.925296797258305, "3000–6000 Hz_40ms_log_rms_mae_db": 1.138536280133022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08550861399592292, "3000–6000 Hz_40ms_envelope_correlation": 0.9940980417180509, "3000–6000 Hz_40ms_modulation_mae_db": 1.188216193423793, "mr_stft_db": 6.021452970714429, "band_mae_db": 0.22356964247586544, "rms_gain_error_db": -0.033063268830632236, "output_peak": 0.891508988087913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.415804370548301, "spectral_convergence": 0.45499589553078634, "spectral_convergence_status": "ok", "amplitude_floor": 1.602383167696498e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.767593065178093, "spectral_convergence": 0.4355927045640709, "spectral_convergence_status": "ok", "amplitude_floor": 1.3018377539140758e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.88851055417499, "spectral_convergence": 0.4689152656943462, "spectral_convergence_status": "ok", "amplitude_floor": 8.15121073207966e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00018979827012669627, "output_power": 0.00020222832226455715, "delta_db": 0.2754972902982752, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007146011617002843, "output_power": 0.0006845223131012189, "delta_db": -0.1867610865822851, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006180173350849692, "output_power": 0.0005241447526911256, "delta_db": -0.7154941465523189, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00021309861976674557, "output_power": 0.00018713678813130465, "delta_db": -0.5642146549548177, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003708463313924785, "output_power": 0.00034187848257371547, "delta_db": -0.3532221942192034, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001752804045595543, "output_power": 0.00017652356696836594, "delta_db": 0.030693276096136597, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.35431377478383946, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.32642749919486025, "pitch": null, "output_rms_dbfs": -26.743952239281043, "output_peak": 0.57294640495255, "reference_rms_dbfs": -26.41752474008618, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0239693299671275}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002801281523487693, "log_rms_mae_db": 1.1736377336869228, "rms_envelope_nrmse": 0.13396618375928343, "envelope_correlation": 0.9841574093437943, "rms_gain_error_db": -0.3264088298843498, "reference_mean_power": 0.0022835195719053173, "output_mean_power": 0.002118184839247806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.817352658591746e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000450774313706133, "output_power": 0.0004058747917575706, "reference_active": true, "delta_db": -0.4556708203859522}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021888119814572043, "output_power": 0.00019457436400087803, "reference_active": true, "delta_db": -0.5112283779638943}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.753791744185585e-05, "output_power": 2.6974379305639262e-05, "reference_active": true, "delta_db": -1.4351871469218038}], "active_band_count": 3, "active_band_mae_db": 0.8006954484238834}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022259441477048914, "log_rms_mae_db": 0.6015127383161037, "rms_envelope_nrmse": 0.08080946961149879, "envelope_correlation": 0.9946704727368311, "rms_gain_error_db": -0.32635950870920777, "reference_mean_power": 0.002294705859570105, "output_mean_power": 0.0021285853743077585}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.266680044598684e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042660839170957063, "output_power": 0.00038101993842236447, "reference_active": true, "delta_db": -0.490816915978017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012272337901795565, "output_power": 0.00010739820843870793, "reference_active": true, "delta_db": -0.5793026771714478}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.241371364832561e-06, "output_power": 2.1303769552152255e-06, "reference_active": true, "delta_db": -1.8227233553269073}], "active_band_count": 3, "active_band_mae_db": 0.9642809828254574}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002244694917839089, "log_rms_mae_db": 1.7283853273750973, "rms_envelope_nrmse": 0.17757317383740195, "envelope_correlation": 0.9792647773149978, "rms_gain_error_db": -0.5476918147642544, "reference_mean_power": 0.001082154267593255, "output_mean_power": 0.0009539376612906412}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.600154760198564e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003310843525581766, "output_power": 0.0002948412944518738, "reference_active": true, "delta_db": -0.5035034673141702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.047815482126218e-05, "output_power": 7.69371805251011e-05, "reference_active": true, "delta_db": -0.7040746825388908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6938491072881828e-05, "output_power": 1.4526214535075996e-05, "reference_active": true, "delta_db": -0.6672226568677977}], "active_band_count": 3, "active_band_mae_db": 0.6249336022402862}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018235367957614062, "log_rms_mae_db": 1.0086765536762115, "rms_envelope_nrmse": 0.12065583846618042, "envelope_correlation": 0.9915759204432383, "rms_gain_error_db": -0.5477196458720458, "reference_mean_power": 0.0010875823794426604, "output_mean_power": 0.0009587164917472349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.047574386178106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00032394555601725244, "output_power": 0.00028679740518956743, "reference_active": true, "delta_db": -0.5289680882195459}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.774463594481584e-05, "output_power": 4.0334735261193595e-05, "reference_active": true, "delta_db": -0.7324537539521346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5157059923891315e-06, "output_power": 1.1424241655655302e-06, "reference_active": true, "delta_db": -1.2278758650132875}], "active_band_count": 3, "active_band_mae_db": 0.8297659023949894}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014040871531833862, "log_rms_mae_db": 1.8589581154039458, "rms_envelope_nrmse": 0.20427195699333467, "envelope_correlation": 0.9676193690675348, "rms_gain_error_db": -0.537130379711571, "reference_mean_power": 0.00036159326722347827, "output_mean_power": 0.00031952680544628274}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2796709554971252e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010222442860363171, "output_power": 8.448681122560725e-05, "reference_active": true, "delta_db": -0.8276577278673654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.551583786781791e-05, "output_power": 3.8613198338284684e-05, "reference_active": true, "delta_db": -0.7142676594537867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1818852015306123e-05, "output_power": 7.528686185693374e-06, "reference_active": true, "delta_db": -1.9585609980348324}], "active_band_count": 3, "active_band_mae_db": 1.1668287951186616}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010845211232047094, "log_rms_mae_db": 1.0904260022109065, "rms_envelope_nrmse": 0.10823259573377161, "envelope_correlation": 0.9924569284696215, "rms_gain_error_db": -0.5370148452972114, "reference_mean_power": 0.00036337479954251316, "output_mean_power": 0.0003211096229907026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.941433739824422e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.608975832570782e-05, "output_power": 7.91483703680941e-05, "reference_active": true, "delta_db": -0.842351237472527}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4640366344588194e-05, "output_power": 2.067516016917703e-05, "reference_active": true, "delta_db": -0.7619827779317356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.520284367741145e-07, "output_power": 5.825868572325337e-07, "reference_active": true, "delta_db": -2.1328923779967077}], "active_band_count": 3, "active_band_mae_db": 1.2457421311336567}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012084003773539908, "log_rms_mae_db": 1.9267688321010303, "rms_envelope_nrmse": 0.1601462680822538, "envelope_correlation": 0.9813418985522128, "rms_gain_error_db": -0.18218287137488903, "reference_mean_power": 0.00030833628390941923, "output_mean_power": 0.00029566937741328533}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2964446771205807e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7472950448048514e-05, "output_power": 4.757340825962562e-05, "reference_active": true, "delta_db": 0.00918042312637973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.978240331431912e-05, "output_power": 4.5357378380443026e-05, "reference_active": true, "delta_db": -0.40427914202337545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3740568823166797e-05, "output_power": 9.267243427026718e-06, "reference_active": true, "delta_db": -1.7105414074963956}], "active_band_count": 3, "active_band_mae_db": 0.7080003242153836}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010527157253957595, "log_rms_mae_db": 1.050306236642325, "rms_envelope_nrmse": 0.08334829987034167, "envelope_correlation": 0.9946334352783891, "rms_gain_error_db": -0.18215079065129836, "reference_mean_power": 0.0003098592063705908, "output_mean_power": 0.0002971319308512295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.056454468513196e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.721882315895364e-05, "output_power": 5.5698208994838126e-05, "reference_active": true, "delta_db": -0.11697690832123854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8898944201462896e-05, "output_power": 2.4915515961091033e-05, "reference_active": true, "delta_db": -0.6441209130914431}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2532553850736713e-06, "output_power": 7.777566797999916e-07, "reference_active": true, "delta_db": -2.071958296399732}], "active_band_count": 3, "active_band_mae_db": 0.9443520392708046}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8087539865449151, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f7c4c39cb314f93694a38773b5bc259d17a7ab156d03be7a1f56a95b84d957dd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9715956591718414, "standard_deviation": 0.0032910933024063267, "minimum": 0.9667272792288281, "maximum": 0.9765516595170539, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8493945488913193, "standard_deviation": 0.03583767085440114, "minimum": 1.795404575528411, "maximum": 1.8926033771607456, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4875449255641622, "standard_deviation": 0.33259199064103023, "minimum": 0.14926979016461497, "maximum": 1.1668287951186616, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18948735633237368, "standard_deviation": 0.011514953088323617, "minimum": 0.17174642627995085, "maximum": 0.20610683417380987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99103849123208, "standard_deviation": 0.002359551368765355, "minimum": 0.9863737627889521, "maximum": 0.9934555072532345, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0691864394936617, "standard_deviation": 0.05091246865935228, "minimum": 0.9777535155314313, "maximum": 1.123360037667984, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6370172018750044, "standard_deviation": 0.3465121439905826, "minimum": 0.1812766505911891, "maximum": 1.2457421311336567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10437280745964207, "standard_deviation": 0.013052770405637784, "minimum": 0.08633589181170796, "maximum": 0.1266733737752765, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9824611191307635, "standard_deviation": 0.0021005907722655487, "minimum": 0.9794832934608912, "maximum": 0.9861299003986606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9365547864896393, "standard_deviation": 0.01671492795619931, "minimum": 1.9189869933919435, "maximum": 1.9646758400140023, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8056628133049635, "standard_deviation": 0.2355511145818673, "minimum": 0.5225630433310603, "maximum": 1.2430095145168503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15481246738501986, "standard_deviation": 0.009576750149949954, "minimum": 0.13788602237915484, "maximum": 0.16830683119304424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9947640754987148, "standard_deviation": 0.0009797452665766069, "minimum": 0.9935322042611311, "maximum": 0.9961793162233622, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0929713818979894, "standard_deviation": 0.026714817554321566, "minimum": 1.050306236642325, "maximum": 1.138536280133022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0204583860823413, "standard_deviation": 0.3042545741552628, "minimum": 0.6393660097186231, "maximum": 1.5886702176733973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08111359182259581, "standard_deviation": 0.008255096223272782, "minimum": 0.0686115381564989, "maximum": 0.0930744664549428, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.980949160318453, "standard_deviation": 0.00307322499702849, "minimum": 0.9744546519872987, "maximum": 0.9839893612858291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7812877706211283, "standard_deviation": 0.04680682214715898, "minimum": 1.7283853273750973, "maximum": 1.8715899614693867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4416909812818767, "standard_deviation": 0.2373307357996847, "minimum": 0.20124942061960635, "maximum": 0.9186072010556011, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16478202081812932, "standard_deviation": 0.013418462662904106, "minimum": 0.15013864875663502, "maximum": 0.1891462939240159, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9935107302971323, "standard_deviation": 0.0018985156156455343, "minimum": 0.9905275721835355, "maximum": 0.9958062757245808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0526516979097074, "standard_deviation": 0.05843746938681537, "minimum": 0.9969798922212466, "maximum": 1.156575702316433, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6475523455730541, "standard_deviation": 0.2572640769583831, "minimum": 0.296965252738916, "maximum": 1.127909240872538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09623525575231731, "standard_deviation": 0.016869419390356327, "minimum": 0.07503450068994869, "maximum": 0.12065583846618042, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24577396220381975, "standard_deviation": 0.07529940215665802, "minimum": 0.12387764855894072, "maximum": 0.35431377478383946, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.023720108851292, "standard_deviation": 0.008399471903224523, "minimum": 6.0078962988286575, "maximum": 6.035177103768585, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6774244774939209, "standard_deviation": 0.09529599457442514, "minimum": 0.57294640495255, "maximum": 0.891508988087913, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11933771270457613, "standard_deviation": 0.09005583165257369, "minimum": -0.32642749919486025, "maximum": -0.033063268830632236, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.98749682647006, "standard_deviation": 0.0018554214576348763, "minimum": 0.9841574093437943, "maximum": 0.9891968161550231, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.167037348507928, "standard_deviation": 0.01531586670421864, "minimum": 1.1320206148689822, "maximum": 1.1811432726234787, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5521666634639366, "standard_deviation": 0.18107874591511017, "minimum": 0.24855011942923397, "maximum": 0.8006954484238834, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11594647695071734, "standard_deviation": 0.00931522288308401, "minimum": 0.10742754876342131, "maximum": 0.13396618375928343, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959518140512416, "standard_deviation": 0.0007517528970198091, "minimum": 0.9946704727368311, "maximum": 0.9969630893385566, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5953823352375733, "standard_deviation": 0.02474308869455405, "minimum": 0.5571736044749863, "maximum": 0.6356349348334183, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6489897620616649, "standard_deviation": 0.19977810951590882, "minimum": 0.31517466106394526, "maximum": 0.9642809828254574, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06363528890426584, "standard_deviation": 0.00812703120981895, "minimum": 0.05522398289917306, "maximum": 0.08080946961149879, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4077-13754-0012", "language": "en", "style": "normal", "speaker": "librispeech_4077", "source_utterance_id": "4077-13754-0012", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4077-13754-0012.wav", "source_file_sha256": "506f0d6fa315cfcff14b01d28c1d0c758f37eb9e2e00d4b6516dece565e8354e", "decoded_pcm_sha256": "43e800e9f69ae92ed54ca36d54a78bd7c56d8ff0657f75ee376d88adab6b3764", "original_source_sha256": "65e48d9b9257be487e80c6938e34856d742870d390ad5d3d1b04d5f95f955d4a", "corpus_entry_id": "librispeech_4077-13754-0012"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 2.722717761993408, "predicted_mos": 2.722717761993408, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8b4ad8956f301639d9d0d59dd594d8020fe66ec2a233200c1a9cacf1c7e627c3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.29340639899965, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4077-13754-0012", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "8b4ad8956f301639d9d0d59dd594d8020fe66ec2a233200c1a9cacf1c7e627c3", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "da6d5022702fb704c7663cfd01a247727508d36f33ba6cdf57c29ec0bca2540e"}, "rows": [{"case": "librispeech_4077-13754-0012", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.063739061355591, "predicted_mos": 2.063739061355591, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e8032b98b485cbabead414bb251f6f522790a8f78ef1a17f0c418abb1718099b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.383063584999036, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4077-13754-0012", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "e8032b98b485cbabead414bb251f6f522790a8f78ef1a17f0c418abb1718099b", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.283493407293791, "全帯域_10ms_rms_envelope_nrmse": 0.10287594761589834, "全帯域_10ms_envelope_correlation": 0.9905912688926223, "全帯域_10ms_modulation_mae_db": 0.13145605012447734, "全帯域_40ms_log_rms_mae_db": 0.7031002782052236, "全帯域_40ms_rms_envelope_nrmse": 0.05474205851488686, "全帯域_40ms_envelope_correlation": 0.9971346055611914, "全帯域_40ms_modulation_mae_db": 0.35943643907115796, "300–1000 Hz_10ms_log_rms_mae_db": 1.7464868674582552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11511407231414306, "300–1000 Hz_10ms_envelope_correlation": 0.9907119036321371, "300–1000 Hz_10ms_modulation_mae_db": 0.4977342320319805, "300–1000 Hz_40ms_log_rms_mae_db": 0.9620822083649117, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06418614217828546, "300–1000 Hz_40ms_envelope_correlation": 0.9969552385759067, "300–1000 Hz_40ms_modulation_mae_db": 0.4922146558126559, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7351043929346333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.171407842510078, "1000–3000 Hz_10ms_envelope_correlation": 0.9765924285235618, "1000–3000 Hz_10ms_modulation_mae_db": 1.0441616322407288, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9773100224813905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09660877072722807, "1000–3000 Hz_40ms_envelope_correlation": 0.9918444555223159, "1000–3000 Hz_40ms_modulation_mae_db": 0.9077435956039794, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9291378342849121, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1541547135449272, "3000–6000 Hz_10ms_envelope_correlation": 0.9856977750809905, "3000–6000 Hz_10ms_modulation_mae_db": 0.39385492181501475, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0332859338506102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08818205181033617, "3000–6000 Hz_40ms_envelope_correlation": 0.9957279034862377, "3000–6000 Hz_40ms_modulation_mae_db": 0.3388066723540077, "mr_stft_db": 5.936614131284778, "band_mae_db": 0.13202100382589618, "rms_gain_error_db": -0.07245707229875212, "output_peak": 0.4964670621558838}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1968085943771318, "全帯域_10ms_rms_envelope_nrmse": 0.10041551027691616, "全帯域_10ms_envelope_correlation": 0.9910318678320574, "全帯域_10ms_modulation_mae_db": 0.5584467101799525, "全帯域_40ms_log_rms_mae_db": 0.6361620352948059, "全帯域_40ms_rms_envelope_nrmse": 0.05136520497835408, "全帯域_40ms_envelope_correlation": 0.9974804727380454, "全帯域_40ms_modulation_mae_db": 0.37160923966642795, "300–1000 Hz_10ms_log_rms_mae_db": 1.7039322514271822, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11424523768307016, "300–1000 Hz_10ms_envelope_correlation": 0.9908950613300112, "300–1000 Hz_10ms_modulation_mae_db": 0.9505764096756208, "300–1000 Hz_40ms_log_rms_mae_db": 1.0036793866539278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05758323890505187, "300–1000 Hz_40ms_envelope_correlation": 0.997745424209695, "300–1000 Hz_40ms_modulation_mae_db": 0.5668622251237964, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6993163313709008, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1654631029745897, "1000–3000 Hz_10ms_envelope_correlation": 0.9779323294094847, "1000–3000 Hz_10ms_modulation_mae_db": 0.2630462213245048, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9072170590617387, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0802370568120809, "1000–3000 Hz_40ms_envelope_correlation": 0.9944821643101789, "1000–3000 Hz_40ms_modulation_mae_db": 0.40287963146364053, "3000–6000 Hz_10ms_log_rms_mae_db": 1.840291074562493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14358741821603685, "3000–6000 Hz_10ms_envelope_correlation": 0.9872343000929014, "3000–6000 Hz_10ms_modulation_mae_db": 0.4176151711579073, "3000–6000 Hz_40ms_log_rms_mae_db": 1.013928929047032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08263689000725696, "3000–6000 Hz_40ms_envelope_correlation": 0.9956484639418923, "3000–6000 Hz_40ms_modulation_mae_db": 0.5151737647515676, "mr_stft_db": 5.91059185669657, "band_mae_db": 0.14881420615594018, "rms_gain_error_db": -0.06989687996831141, "output_peak": 0.5066495103001892}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2518997159333989, "全帯域_10ms_rms_envelope_nrmse": 0.11084595067992443, "全帯域_10ms_envelope_correlation": 0.9891188785212712, "全帯域_10ms_modulation_mae_db": 0.4051382611321476, "全帯域_40ms_log_rms_mae_db": 0.6851173188276658, "全帯域_40ms_rms_envelope_nrmse": 0.0567748902588034, "全帯域_40ms_envelope_correlation": 0.9970400433711947, "全帯域_40ms_modulation_mae_db": 0.5863421265874021, "300–1000 Hz_10ms_log_rms_mae_db": 1.8103481339402197, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13048522931682444, "300–1000 Hz_10ms_envelope_correlation": 0.988065609580969, "300–1000 Hz_10ms_modulation_mae_db": 0.40516405577589465, "300–1000 Hz_40ms_log_rms_mae_db": 1.022478824472529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08049319886459055, "300–1000 Hz_40ms_envelope_correlation": 0.9953426189705521, "300–1000 Hz_40ms_modulation_mae_db": 0.16124046521379096, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7805213885822506, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17732843765973821, "1000–3000 Hz_10ms_envelope_correlation": 0.9745963346515009, "1000–3000 Hz_10ms_modulation_mae_db": 0.2733420243023746, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9904390976734834, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0891295086803207, "1000–3000 Hz_40ms_envelope_correlation": 0.9930845264582393, "1000–3000 Hz_40ms_modulation_mae_db": 0.4188471076276616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8569584220754138, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14028952075464732, "3000–6000 Hz_10ms_envelope_correlation": 0.9879175245228915, "3000–6000 Hz_10ms_modulation_mae_db": 0.7223492902667811, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0292709376806506, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06972551282793427, "3000–6000 Hz_40ms_envelope_correlation": 0.9971056160263685, "3000–6000 Hz_40ms_modulation_mae_db": 0.8737171986199376, "mr_stft_db": 5.942080153059229, "band_mae_db": 0.1712054538923614, "rms_gain_error_db": -0.1296057944441399, "output_peak": 0.43057957637461536}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2986240905729862, "全帯域_10ms_rms_envelope_nrmse": 0.108690870397627, "全帯域_10ms_envelope_correlation": 0.9895870090261425, "全帯域_10ms_modulation_mae_db": 0.26835778025802914, "全帯域_40ms_log_rms_mae_db": 0.7464512774214881, "全帯域_40ms_rms_envelope_nrmse": 0.05966851276756901, "全帯域_40ms_envelope_correlation": 0.9966062944514744, "全帯域_40ms_modulation_mae_db": 0.12222384737611253, "300–1000 Hz_10ms_log_rms_mae_db": 1.7890852384834943, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14112341954800575, "300–1000 Hz_10ms_envelope_correlation": 0.9859919183839224, "300–1000 Hz_10ms_modulation_mae_db": 0.7315286735517647, "300–1000 Hz_40ms_log_rms_mae_db": 1.0640050667663958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07837308172435362, "300–1000 Hz_40ms_envelope_correlation": 0.9956198054595348, "300–1000 Hz_40ms_modulation_mae_db": 0.3509025320574361, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7280018915441653, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17427331626402284, "1000–3000 Hz_10ms_envelope_correlation": 0.9754858340575634, "1000–3000 Hz_10ms_modulation_mae_db": 0.5522846522826722, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9759726339499045, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08478077343445496, "1000–3000 Hz_40ms_envelope_correlation": 0.9937415646240253, "1000–3000 Hz_40ms_modulation_mae_db": 0.5484947959186637, "3000–6000 Hz_10ms_log_rms_mae_db": 1.834196972291095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1537475105932664, "3000–6000 Hz_10ms_envelope_correlation": 0.9860350442858457, "3000–6000 Hz_10ms_modulation_mae_db": 0.6898293423108227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0633625007218928, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09414648489453588, "3000–6000 Hz_40ms_envelope_correlation": 0.9947179276038312, "3000–6000 Hz_40ms_modulation_mae_db": 0.4504950742227964, "mr_stft_db": 5.938991818880781, "band_mae_db": 0.20073018199311568, "rms_gain_error_db": 0.016470178913726024, "output_peak": 0.44270137297353845}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3036631775969474, "全帯域_10ms_rms_envelope_nrmse": 0.10453079077769259, "全帯域_10ms_envelope_correlation": 0.990287414389467, "全帯域_10ms_modulation_mae_db": 0.29877343659786265, "全帯域_40ms_log_rms_mae_db": 0.704997874935293, "全帯域_40ms_rms_envelope_nrmse": 0.050309412848847546, "全帯域_40ms_envelope_correlation": 0.9976196092608498, "全帯域_40ms_modulation_mae_db": 0.47316617062747574, "300–1000 Hz_10ms_log_rms_mae_db": 1.803159914841262, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11989518355811306, "300–1000 Hz_10ms_envelope_correlation": 0.9898903667691242, "300–1000 Hz_10ms_modulation_mae_db": 0.48702723016257005, "300–1000 Hz_40ms_log_rms_mae_db": 1.07141652067051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06890281471503944, "300–1000 Hz_40ms_envelope_correlation": 0.9965498278505311, "300–1000 Hz_40ms_modulation_mae_db": 0.8370404417301506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7534556281421732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18949406460224263, "1000–3000 Hz_10ms_envelope_correlation": 0.9712999766678401, "1000–3000 Hz_10ms_modulation_mae_db": 0.5543708795315162, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9771385314818091, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11303735889597441, "1000–3000 Hz_40ms_envelope_correlation": 0.9896261189012596, "1000–3000 Hz_40ms_modulation_mae_db": 0.4954379181851633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8817321598323395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.146167339826843, "3000–6000 Hz_10ms_envelope_correlation": 0.9867462515197083, "3000–6000 Hz_10ms_modulation_mae_db": 0.23576051302766354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0390008296324935, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07021856776785441, "3000–6000 Hz_40ms_envelope_correlation": 0.9968197377032866, "3000–6000 Hz_40ms_modulation_mae_db": 0.6255454543258562, "mr_stft_db": 5.923687226170173, "band_mae_db": 0.1603157954050076, "rms_gain_error_db": -0.08502087166324926, "output_peak": 0.5026940234953363}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2540304705816454, "全帯域_10ms_rms_envelope_nrmse": 0.1132568595015097, "全帯域_10ms_envelope_correlation": 0.9885710894201172, "全帯域_10ms_modulation_mae_db": 0.2636837728516321, "全帯域_40ms_log_rms_mae_db": 0.6319175741000829, "全帯域_40ms_rms_envelope_nrmse": 0.057344253975339154, "全帯域_40ms_envelope_correlation": 0.9968549059247219, "全帯域_40ms_modulation_mae_db": 0.2236556067316546, "300–1000 Hz_10ms_log_rms_mae_db": 1.7160661675414097, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10927969725777946, "300–1000 Hz_10ms_envelope_correlation": 0.991601377762053, "300–1000 Hz_10ms_modulation_mae_db": 0.6635108962250589, "300–1000 Hz_40ms_log_rms_mae_db": 1.0081919842698701, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05724409417481801, "300–1000 Hz_40ms_envelope_correlation": 0.9976021054292546, "300–1000 Hz_40ms_modulation_mae_db": 0.4892865661838883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.707501069813869, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.170901424354832, "1000–3000 Hz_10ms_envelope_correlation": 0.9766680668710741, "1000–3000 Hz_10ms_modulation_mae_db": 0.6497317316939858, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9287731338145618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08414643121172169, "1000–3000 Hz_40ms_envelope_correlation": 0.9944799489667836, "1000–3000 Hz_40ms_modulation_mae_db": 0.6962583517663111, "3000–6000 Hz_10ms_log_rms_mae_db": 1.90200150628992, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16987801438314562, "3000–6000 Hz_10ms_envelope_correlation": 0.9821162230100482, "3000–6000 Hz_10ms_modulation_mae_db": 0.24479002699283078, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0174204728952332, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09657338590948573, "3000–6000 Hz_40ms_envelope_correlation": 0.9939175487054589, "3000–6000 Hz_40ms_modulation_mae_db": 0.4315241423576843, "mr_stft_db": 5.947429934420879, "band_mae_db": 0.16352585633601457, "rms_gain_error_db": -0.0729368052299961, "output_peak": 0.44953733079410974}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2595934285667572, "全帯域_10ms_rms_envelope_nrmse": 0.10126207016379461, "全帯域_10ms_envelope_correlation": 0.9909013669521258, "全帯域_10ms_modulation_mae_db": 0.4087900328823899, "全帯域_40ms_log_rms_mae_db": 0.6874971832518143, "全帯域_40ms_rms_envelope_nrmse": 0.04999261290314487, "全帯域_40ms_envelope_correlation": 0.9976643608848715, "全帯域_40ms_modulation_mae_db": 0.49985788766797173, "300–1000 Hz_10ms_log_rms_mae_db": 1.7306814846660024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11623639025847968, "300–1000 Hz_10ms_envelope_correlation": 0.9905063562847506, "300–1000 Hz_10ms_modulation_mae_db": 0.5958232136920638, "300–1000 Hz_40ms_log_rms_mae_db": 1.0413041511206522, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0654264514146988, "300–1000 Hz_40ms_envelope_correlation": 0.9969062987861059, "300–1000 Hz_40ms_modulation_mae_db": 0.19225866619879395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.707021713587535, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16806225030031655, "1000–3000 Hz_10ms_envelope_correlation": 0.9773572683341886, "1000–3000 Hz_10ms_modulation_mae_db": 1.0995254571464579, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9546040097434779, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09267372906754308, "1000–3000 Hz_40ms_envelope_correlation": 0.9924863127439797, "1000–3000 Hz_40ms_modulation_mae_db": 0.935433203966114, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8398243186559613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13388113195098292, "3000–6000 Hz_10ms_envelope_correlation": 0.9890705332146721, "3000–6000 Hz_10ms_modulation_mae_db": 0.49254967442928566, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9913400112000263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07267491061238833, "3000–6000 Hz_40ms_envelope_correlation": 0.9968942306897745, "3000–6000 Hz_40ms_modulation_mae_db": 0.6351406641309566, "mr_stft_db": 5.925285729669263, "band_mae_db": 0.16522135121466217, "rms_gain_error_db": -0.09528265325078575, "output_peak": 0.5175156638034353}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2330590276889428, "全帯域_10ms_rms_envelope_nrmse": 0.10362518932799984, "全帯域_10ms_envelope_correlation": 0.9905157013559556, "全帯域_10ms_modulation_mae_db": 0.21011831335232778, "全帯域_40ms_log_rms_mae_db": 0.5816675304435766, "全帯域_40ms_rms_envelope_nrmse": 0.05382899541478297, "全帯域_40ms_envelope_correlation": 0.9972169599716642, "全帯域_40ms_modulation_mae_db": 0.3434259035989277, "300–1000 Hz_10ms_log_rms_mae_db": 1.6962358484754292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11406997651522349, "300–1000 Hz_10ms_envelope_correlation": 0.9909176581347421, "300–1000 Hz_10ms_modulation_mae_db": 0.16535113940990218, "300–1000 Hz_40ms_log_rms_mae_db": 0.9816462953165058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06759777693530257, "300–1000 Hz_40ms_envelope_correlation": 0.9967747634005509, "300–1000 Hz_40ms_modulation_mae_db": 0.15916515230682246, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7127001923953442, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1792316348454924, "1000–3000 Hz_10ms_envelope_correlation": 0.974971278389386, "1000–3000 Hz_10ms_modulation_mae_db": 0.276407645785741, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9374716657788197, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09674201167836496, "1000–3000 Hz_40ms_envelope_correlation": 0.9921648219453105, "1000–3000 Hz_40ms_modulation_mae_db": 0.37603762699666676, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8578585355199662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14493171027246685, "3000–6000 Hz_10ms_envelope_correlation": 0.9872183047812816, "3000–6000 Hz_10ms_modulation_mae_db": 0.36133682989372273, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0136936161544254, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08138323805849289, "3000–6000 Hz_40ms_envelope_correlation": 0.9958040184472738, "3000–6000 Hz_40ms_modulation_mae_db": 0.6402298228572052, "mr_stft_db": 5.925742153858171, "band_mae_db": 0.11483296037956704, "rms_gain_error_db": 0.025087256719413767, "output_peak": 0.46986140638024687}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.32015776302339, "spectral_convergence": 0.46301543804643874, "spectral_convergence_status": "ok", "amplitude_floor": 1.32832143223519e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.653395701908779, "spectral_convergence": 0.43343377579784276, "spectral_convergence_status": "ok", "amplitude_floor": 1.1133586355586426e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8362889289221656, "spectral_convergence": 0.45941623928680153, "spectral_convergence_status": "ok", "amplitude_floor": 6.710028542715064e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00015990772541983682, "output_power": 0.00016164460444288898, "delta_db": 0.046917670417728104, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005346827040387907, "output_power": 0.0005256510737328835, "delta_db": -0.07398579861019067, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025601253900142847, "output_power": 0.00026573004921110675, "delta_db": 0.16179431213759005, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015659557404440672, "output_power": 0.000152617105103383, "delta_db": -0.11176271726024639, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003554436518358531, "output_power": 0.00035899121132464595, "delta_db": 0.04313054282634194, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004235024710497043, "output_power": 0.00039030350425993545, "delta_db": -0.35453498170327996, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13202100382589618, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07245707229875212, "pitch": null, "output_rms_dbfs": -27.316707076031427, "output_peak": 0.4964670621558838, "reference_rms_dbfs": -27.244250003732674, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.936614131284778}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018816516597127566, "log_rms_mae_db": 1.283493407293791, "rms_envelope_nrmse": 0.10287594761589834, "envelope_correlation": 0.9905912688926223, "rms_gain_error_db": -0.07216765264949417, "reference_mean_power": 0.0018869472812584292, "output_mean_power": 0.0018558505565418521}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.448172186664129e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039739520206635815, "output_power": 0.0003931646011051135, "reference_active": true, "delta_db": -0.046482105068591975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001684948117890412, "output_power": 0.00016315421372395888, "reference_active": true, "delta_db": -0.13988238194236707}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9559595301379592e-05, "output_power": 1.864487595883855e-05, "reference_active": true, "delta_db": -0.20800366336247292}], "active_band_count": 3, "active_band_mae_db": 0.13145605012447734}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016656172387320294, "log_rms_mae_db": 0.7031002782052236, "rms_envelope_nrmse": 0.05474205851488686, "envelope_correlation": 0.9971346055611914, "rms_gain_error_db": -0.07180417850329596, "reference_mean_power": 0.0018945898208991615, "output_mean_power": 0.0018635231051445447}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.514104383617561e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003829793712831679, "output_power": 0.00038106604831152584, "reference_active": true, "delta_db": -0.021751255209125693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.831616093630877e-05, "output_power": 7.564454304273638e-05, "reference_active": true, "delta_db": -0.15073786455614657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3812315351816923e-06, "output_power": 1.9329454844170998e-06, "reference_active": true, "delta_db": -0.9058201974482016}], "active_band_count": 3, "active_band_mae_db": 0.35943643907115796}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015092951870698086, "log_rms_mae_db": 1.7464868674582552, "rms_envelope_nrmse": 0.11511407231414306, "envelope_correlation": 0.9907119036321371, "rms_gain_error_db": -0.012932666325362425, "reference_mean_power": 0.000631981580682194, "output_mean_power": 0.0006301024295534602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.167139764996573e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018037000728302603, "output_power": 0.00017363574943816458, "reference_active": true, "delta_db": -0.16525176944471542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6944061726753776e-05, "output_power": 2.762178569289084e-05, "reference_active": true, "delta_db": 0.10788686659331306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.521461134259169e-06, "output_power": 4.663686035276419e-06, "reference_active": true, "delta_db": 1.2200640600579131}], "active_band_count": 3, "active_band_mae_db": 0.4977342320319805}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013588907795973108, "log_rms_mae_db": 0.9620822083649117, "rms_envelope_nrmse": 0.06418614217828546, "envelope_correlation": 0.9969552385759067, "rms_gain_error_db": -0.012460154220591, "reference_mean_power": 0.0006345375227010602, "output_mean_power": 0.0006327196077712843}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.990151794989553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001715162379513309, "output_power": 0.00016612983596592533, "reference_active": true, "delta_db": -0.13857606018084517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2552077755571312e-05, "output_power": 1.2373555541202913e-05, "reference_active": true, "delta_db": -0.06221108801667012}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.057253626713498e-07, "output_power": 4.1012585689057254e-07, "reference_active": true, "delta_db": 1.2758568192404525}], "active_band_count": 3, "active_band_mae_db": 0.4922146558126559}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.324369547066356e-05, "log_rms_mae_db": 1.7351043929346333, "rms_envelope_nrmse": 0.171407842510078, "envelope_correlation": 0.9765924285235618, "rms_gain_error_db": 0.015865294968815585, "reference_mean_power": 0.00026216016454011753, "output_mean_power": 0.0002631196182950146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5456728460340458e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.502085853898132e-05, "output_power": 7.47092472922577e-05, "reference_active": true, "delta_db": -0.018076690667129827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151594827944158e-05, "output_power": 2.2457441372480677e-05, "reference_active": true, "delta_db": 0.1859978293955054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.513878290774259e-06, "output_power": 4.337968557271127e-06, "reference_active": true, "delta_db": -2.928410376659551}], "active_band_count": 3, "active_band_mae_db": 1.0441616322407288}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.873751028776232e-05, "log_rms_mae_db": 0.9773100224813905, "rms_envelope_nrmse": 0.09660877072722807, "envelope_correlation": 0.9918444555223159, "rms_gain_error_db": 0.01661190361596214, "reference_mean_power": 0.0002632887650860277, "output_mean_power": 0.00026429778161079863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3635149275752304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.257955032157826e-05, "output_power": 7.247736650502371e-05, "reference_active": true, "delta_db": -0.006118684442069431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.405048683009261e-06, "output_power": 9.620127817111807e-06, "reference_active": true, "delta_db": 0.09819794382590077}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.677553554297748e-07, "output_power": 4.200794519320969e-07, "reference_active": true, "delta_db": -2.618914158543968}], "active_band_count": 3, "active_band_mae_db": 0.9077435956039794}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015682729984940356, "log_rms_mae_db": 1.9291378342849121, "rms_envelope_nrmse": 0.1541547135449272, "envelope_correlation": 0.9856977750809905, "rms_gain_error_db": -0.3343031667143803, "reference_mean_power": 0.0005797089728212459, "output_mean_power": 0.0005367594581726894}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2924452919576597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020605682430557354, "output_power": 0.00020515178161379272, "reference_active": true, "delta_db": -0.019117094143361293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012645863441832237, "output_power": 0.00011463559185830693, "reference_active": true, "delta_db": -0.4262901015127164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.26990748699409e-06, "output_power": 1.098224573018624e-05, "reference_active": true, "delta_db": 0.7361575697889666}], "active_band_count": 3, "active_band_mae_db": 0.39385492181501475}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012704534221320322, "log_rms_mae_db": 1.0332859338506102, "rms_envelope_nrmse": 0.08818205181033617, "envelope_correlation": 0.9957279034862377, "rms_gain_error_db": -0.33410616772888435, "reference_mean_power": 0.0005824704905358117, "output_mean_power": 0.000539340844709513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.040463159930301e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022945963060392102, "output_power": 0.00022701259460143242, "reference_active": true, "delta_db": -0.04656337689663118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.01414803619856e-05, "output_power": 5.499210137864853e-05, "reference_active": true, "delta_db": -0.3887379848494738}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3285241811094974e-06, "output_power": 1.1621377215879414e-06, "reference_active": true, "delta_db": -0.5811186553159179}], "active_band_count": 3, "active_band_mae_db": 0.3388066723540077}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8103903057462521, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8d8fe746d3a5c78490d3b2867c402f4cfec61ef15a1bd762caaef94e2eeef710", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9756129396130749, "standard_deviation": 0.0020893595647068883, "minimum": 0.9712999766678401, "maximum": 0.9779323294094847, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.727952826046359, "standard_deviation": 0.02770784174859295, "minimum": 1.6993163313709008, "maximum": 1.7805213885822506, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5891087805384976, "standard_deviation": 0.33330457939881447, "minimum": 0.2630462213245048, "maximum": 1.0995254571464579, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17452025918891403, "standard_deviation": 0.007567598382018687, "minimum": 0.1654631029745897, "maximum": 0.18949406460224263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927387391840117, "standard_deviation": 0.0016070300369229965, "minimum": 0.9896261189012596, "maximum": 0.9944821643101789, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9561157692481482, "standard_deviation": 0.029159978525733304, "minimum": 0.9072170590617387, "maximum": 0.9904390976734834, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5976415289410251, "standard_deviation": 0.22420038734071548, "minimum": 0.37603762699666676, "maximum": 0.935433203966114, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09216945506346111, "standard_deviation": 0.010329738139231012, "minimum": 0.0802370568120809, "maximum": 0.11303735889597441, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9865044945635424, "standard_deviation": 0.002063627274712476, "minimum": 0.9821162230100482, "maximum": 0.9890705332146721, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8677501029390124, "standard_deviation": 0.03382955983755856, "minimum": 1.834196972291095, "maximum": 1.9291378342849121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4447607212367536, "standard_deviation": 0.18258266983832078, "minimum": 0.23576051302766354, "maximum": 0.7223492902667811, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1483296699427895, "standard_deviation": 0.010952692300203836, "minimum": 0.13388113195098292, "maximum": 0.16987801438314562, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9958294308255153, "standard_deviation": 0.00111339829770499, "minimum": 0.9939175487054589, "maximum": 0.9971056160263685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0251629038977956, "standard_deviation": 0.021364981374402812, "minimum": 0.9913400112000263, "maximum": 1.0633625007218928, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5638290992025015, "standard_deviation": 0.1665248113203288, "minimum": 0.3388066723540077, "maximum": 0.8737171986199376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08194263023603557, "standard_deviation": 0.010524261789284497, "minimum": 0.06972551282793427, "maximum": 0.09657338590948573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9898225314847138, "standard_deviation": 0.0018722594764630928, "minimum": 0.9859919183839224, "maximum": 0.991601377762053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7494994883541568, "standard_deviation": 0.04558100976137566, "minimum": 1.6962358484754292, "maximum": 1.8103481339402197, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.562089481315607, "standard_deviation": 0.23386774108822053, "minimum": 0.16535113940990218, "maximum": 0.9505764096756208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12005615080645488, "standard_deviation": 0.01053815555009875, "minimum": 0.10927969725777946, "maximum": 0.14112341954800575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966870103352664, "standard_deviation": 0.0008494121326172525, "minimum": 0.9953426189705521, "maximum": 0.997745424209695, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0193505547044128, "standard_deviation": 0.038299590817446126, "minimum": 0.9620822083649117, "maximum": 1.07141652067051, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4061213380784168, "standard_deviation": 0.23772666769456713, "minimum": 0.15916515230682246, "maximum": 0.8370404417301506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06747584986401754, "standard_deviation": 0.00851743521798465, "minimum": 0.05724409417481801, "maximum": 0.08049319886459055, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1570833511503206, "standard_deviation": 0.025923670306388903, "minimum": 0.11483296037956704, "maximum": 0.20073018199311568, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.93130287550498, "standard_deviation": 0.012057203614837402, "minimum": 5.91059185669657, "maximum": 5.947429934420879, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47700074328466935, "standard_deviation": 0.03315800072952617, "minimum": 0.43057957637461536, "maximum": 0.5175156638034353, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06045533015276184, "standard_deviation": 0.05374840195242321, "minimum": -0.1296057944441399, "maximum": 0.025087256719413767, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900755745487199, "standard_deviation": 0.0008878865258226331, "minimum": 0.9885710894201172, "maximum": 0.9910318678320574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2601464890764502, "standard_deviation": 0.035396623894204074, "minimum": 1.1968085943771318, "maximum": 1.3036631775969474, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31809554467235235, "standard_deviation": 0.1340958190633603, "minimum": 0.13145605012447734, "maximum": 0.5584467101799525, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10568789859267033, "standard_deviation": 0.004687358110179888, "minimum": 0.10041551027691616, "maximum": 0.1132568595015097, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972021565205016, "standard_deviation": 0.00037264634916800427, "minimum": 0.9966062944514744, "maximum": 0.9976643608848715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6721138840599938, "standard_deviation": 0.05218390335576694, "minimum": 0.5816675304435766, "maximum": 0.7464512774214881, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3724646526658913, "standard_deviation": 0.15018641450224812, "minimum": 0.12222384737611253, "maximum": 0.5863421265874021, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05425324270771599, "standard_deviation": 0.0035396661610553053, "minimum": 0.04999261290314487, "maximum": 0.05966851276756901, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0012", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.3467466831207275, "predicted_mos": 2.3467466831207275, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "da429c76ca224e94b0db04064b639e5d8830e010b3c159532573c69732844152", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.735290599986911, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2906199709437864, "全帯域_10ms_rms_envelope_nrmse": 0.10653515830466742, "全帯域_10ms_envelope_correlation": 0.9899077881918248, "全帯域_10ms_modulation_mae_db": 0.021334740529741495, "全帯域_40ms_log_rms_mae_db": 0.7072979727113958, "全帯域_40ms_rms_envelope_nrmse": 0.05792056729441475, "全帯域_40ms_envelope_correlation": 0.9967798930899348, "全帯域_40ms_modulation_mae_db": 0.2540908591595062, "300–1000 Hz_10ms_log_rms_mae_db": 1.7315946675979448, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11367691746223023, "300–1000 Hz_10ms_envelope_correlation": 0.9909563886953534, "300–1000 Hz_10ms_modulation_mae_db": 0.5366183505359982, "300–1000 Hz_40ms_log_rms_mae_db": 0.9471412732280549, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06196344316297622, "300–1000 Hz_40ms_envelope_correlation": 0.9971746149267005, "300–1000 Hz_40ms_modulation_mae_db": 0.4958853237223791, "1000–3000 Hz_10ms_log_rms_mae_db": 1.772672767034512, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17770412984137482, "1000–3000 Hz_10ms_envelope_correlation": 0.9748270860384447, "1000–3000 Hz_10ms_modulation_mae_db": 1.0492933647125875, "1000–3000 Hz_40ms_log_rms_mae_db": 0.944800025431722, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09981833629595428, "1000–3000 Hz_40ms_envelope_correlation": 0.9913187511519602, "1000–3000 Hz_40ms_modulation_mae_db": 0.9303664302447077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9983705526541047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16075902257345398, "3000–6000 Hz_10ms_envelope_correlation": 0.9841473712415797, "3000–6000 Hz_10ms_modulation_mae_db": 0.4206241725095432, "3000–6000 Hz_40ms_log_rms_mae_db": 1.040769996872589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0910153344361166, "3000–6000 Hz_40ms_envelope_correlation": 0.9950655311319442, "3000–6000 Hz_40ms_modulation_mae_db": 0.31687443807497, "mr_stft_db": 5.926039235434011, "band_mae_db": 0.13448768681541082, "rms_gain_error_db": -0.06844352289206435, "output_peak": 0.4978483618956231}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2240311245204705, "全帯域_10ms_rms_envelope_nrmse": 0.10151365070498151, "全帯域_10ms_envelope_correlation": 0.9908272285100591, "全帯域_10ms_modulation_mae_db": 0.5296800352194758, "全帯域_40ms_log_rms_mae_db": 0.6386638157164154, "全帯域_40ms_rms_envelope_nrmse": 0.05053438414234161, "全帯域_40ms_envelope_correlation": 0.9975462037240259, "全帯域_40ms_modulation_mae_db": 0.38923176436452583, "300–1000 Hz_10ms_log_rms_mae_db": 1.704335276379277, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1172898337008198, "300–1000 Hz_10ms_envelope_correlation": 0.9903779058565204, "300–1000 Hz_10ms_modulation_mae_db": 0.9012500800655848, "300–1000 Hz_40ms_log_rms_mae_db": 0.9333021513655062, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05866007709779376, "300–1000 Hz_40ms_envelope_correlation": 0.9976195764609184, "300–1000 Hz_40ms_modulation_mae_db": 0.5693022077014119, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7364724446541109, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1661282484774389, "1000–3000 Hz_10ms_envelope_correlation": 0.9777482695488177, "1000–3000 Hz_10ms_modulation_mae_db": 0.3289173966694873, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9253549123807069, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08510220892540593, "1000–3000 Hz_40ms_envelope_correlation": 0.9937397882301773, "1000–3000 Hz_40ms_modulation_mae_db": 0.5217024938816041, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8973292196012455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14573477507557728, "3000–6000 Hz_10ms_envelope_correlation": 0.9868270598310983, "3000–6000 Hz_10ms_modulation_mae_db": 0.4677831412854812, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9975937236041839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07834441161615173, "3000–6000 Hz_40ms_envelope_correlation": 0.996020454718298, "3000–6000 Hz_40ms_modulation_mae_db": 0.5776917032808893, "mr_stft_db": 5.90164999488118, "band_mae_db": 0.11100140432972798, "rms_gain_error_db": -0.05463372865080117, "output_peak": 0.4966993264338022}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.248173090246287, "全帯域_10ms_rms_envelope_nrmse": 0.11175315650594417, "全帯域_10ms_envelope_correlation": 0.9889621515428784, "全帯域_10ms_modulation_mae_db": 0.3832063444355369, "全帯域_40ms_log_rms_mae_db": 0.6785600871363656, "全帯域_40ms_rms_envelope_nrmse": 0.05900131298059612, "全帯域_40ms_envelope_correlation": 0.9968300022903159, "全帯域_40ms_modulation_mae_db": 0.5348604178463994, "300–1000 Hz_10ms_log_rms_mae_db": 1.7706679281948197, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1316872517652592, "300–1000 Hz_10ms_envelope_correlation": 0.9878732683824522, "300–1000 Hz_10ms_modulation_mae_db": 0.43081079823203283, "300–1000 Hz_40ms_log_rms_mae_db": 0.9985106575624103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08116353653198102, "300–1000 Hz_40ms_envelope_correlation": 0.995279697585384, "300–1000 Hz_40ms_modulation_mae_db": 0.13545928082540173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8563360550506605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17782200038005563, "1000–3000 Hz_10ms_envelope_correlation": 0.9744997766134529, "1000–3000 Hz_10ms_modulation_mae_db": 0.31109837694430054, "1000–3000 Hz_40ms_log_rms_mae_db": 0.985599769085111, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09239841908543585, "1000–3000 Hz_40ms_envelope_correlation": 0.9927549420610875, "1000–3000 Hz_40ms_modulation_mae_db": 0.3893638767460952, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8759898780698805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15020880262754488, "3000–6000 Hz_10ms_envelope_correlation": 0.9860588760481893, "3000–6000 Hz_10ms_modulation_mae_db": 0.6784675703049793, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9888432758283856, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07667174060444676, "3000–6000 Hz_40ms_envelope_correlation": 0.996344270756172, "3000–6000 Hz_40ms_modulation_mae_db": 0.8306576715966042, "mr_stft_db": 5.926257768428363, "band_mae_db": 0.15005592461394693, "rms_gain_error_db": -0.1430262080590797, "output_peak": 0.4194679067226995}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3294237587494089, "全帯域_10ms_rms_envelope_nrmse": 0.10762961028193065, "全帯域_10ms_envelope_correlation": 0.9897701751350935, "全帯域_10ms_modulation_mae_db": 0.14395391946660221, "全帯域_40ms_log_rms_mae_db": 0.761377643384558, "全帯域_40ms_rms_envelope_nrmse": 0.059075446333433354, "全帯域_40ms_envelope_correlation": 0.9966582582821129, "全帯域_40ms_modulation_mae_db": 0.04255434749720186, "300–1000 Hz_10ms_log_rms_mae_db": 1.7331800980566499, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13606635585999027, "300–1000 Hz_10ms_envelope_correlation": 0.9870150146526249, "300–1000 Hz_10ms_modulation_mae_db": 0.7603346558553956, "300–1000 Hz_40ms_log_rms_mae_db": 0.9689392767043762, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07573096654362439, "300–1000 Hz_40ms_envelope_correlation": 0.995952244419574, "300–1000 Hz_40ms_modulation_mae_db": 0.3899935722370582, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7843939763039927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17841478836667915, "1000–3000 Hz_10ms_envelope_correlation": 0.9743480096511536, "1000–3000 Hz_10ms_modulation_mae_db": 0.9615793418313275, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9786926905385595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.088739242739505, "1000–3000 Hz_40ms_envelope_correlation": 0.9934114628539743, "1000–3000 Hz_40ms_modulation_mae_db": 1.060508078105108, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8764326713753046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1557344351466874, "3000–6000 Hz_10ms_envelope_correlation": 0.9859670887829973, "3000–6000 Hz_10ms_modulation_mae_db": 0.601089354051764, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0292838076776851, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.095929782212648, "3000–6000 Hz_40ms_envelope_correlation": 0.9947621335795681, "3000–6000 Hz_40ms_modulation_mae_db": 0.335105183469041, "mr_stft_db": 5.926127787505595, "band_mae_db": 0.20967324660610523, "rms_gain_error_db": 0.008126796626566832, "output_peak": 0.4367563371380234}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.315947303828151, "全帯域_10ms_rms_envelope_nrmse": 0.10236794132217986, "全帯域_10ms_envelope_correlation": 0.990690590301127, "全帯域_10ms_modulation_mae_db": 0.2859832260737372, "全帯域_40ms_log_rms_mae_db": 0.7078184057156856, "全帯域_40ms_rms_envelope_nrmse": 0.048468425283630476, "全帯域_40ms_envelope_correlation": 0.9977982970818432, "全帯域_40ms_modulation_mae_db": 0.42674665860636224, "300–1000 Hz_10ms_log_rms_mae_db": 1.772777345685823, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12135905565445272, "300–1000 Hz_10ms_envelope_correlation": 0.9896559638995153, "300–1000 Hz_10ms_modulation_mae_db": 0.5306336973225759, "300–1000 Hz_40ms_log_rms_mae_db": 0.9786365834902312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07070161102350857, "300–1000 Hz_40ms_envelope_correlation": 0.9963730704802531, "300–1000 Hz_40ms_modulation_mae_db": 0.7096041827217526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7998973377365912, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18684656401138341, "1000–3000 Hz_10ms_envelope_correlation": 0.9722656151418944, "1000–3000 Hz_10ms_modulation_mae_db": 0.5984086319785283, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9594943906124106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10678875613748408, "1000–3000 Hz_40ms_envelope_correlation": 0.9911080578225177, "1000–3000 Hz_40ms_modulation_mae_db": 0.47666560065805497, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9185371921235095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1430877166416602, "3000–6000 Hz_10ms_envelope_correlation": 0.9873340444146278, "3000–6000 Hz_10ms_modulation_mae_db": 0.23621867505414204, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0075056299289553, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06587670163108436, "3000–6000 Hz_40ms_envelope_correlation": 0.9971741397653497, "3000–6000 Hz_40ms_modulation_mae_db": 0.6960870044062105, "mr_stft_db": 5.908263474216487, "band_mae_db": 0.15423099319110015, "rms_gain_error_db": -0.0877768258464651, "output_peak": 0.49739457992114067}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2749365356782818, "全帯域_10ms_rms_envelope_nrmse": 0.11418290195860616, "全帯域_10ms_envelope_correlation": 0.9883838938328698, "全帯域_10ms_modulation_mae_db": 0.12940435866600153, "全帯域_40ms_log_rms_mae_db": 0.6424427745805301, "全帯域_40ms_rms_envelope_nrmse": 0.05826928791804883, "全帯域_40ms_envelope_correlation": 0.9967607651618305, "全帯域_40ms_modulation_mae_db": 0.40106912992615884, "300–1000 Hz_10ms_log_rms_mae_db": 1.6786484470132925, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10951688629210836, "300–1000 Hz_10ms_envelope_correlation": 0.9915741714234051, "300–1000 Hz_10ms_modulation_mae_db": 0.5278367627793491, "300–1000 Hz_40ms_log_rms_mae_db": 0.9480361802610519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0581209319565605, "300–1000 Hz_40ms_envelope_correlation": 0.9975222775406979, "300–1000 Hz_40ms_modulation_mae_db": 0.3376590208802632, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7575071657532921, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1701717473892206, "1000–3000 Hz_10ms_envelope_correlation": 0.9769472510393177, "1000–3000 Hz_10ms_modulation_mae_db": 0.911063088208345, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9046655386872064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08213127693943255, "1000–3000 Hz_40ms_envelope_correlation": 0.994898491685709, "1000–3000 Hz_40ms_modulation_mae_db": 0.9849058247038233, "3000–6000 Hz_10ms_log_rms_mae_db": 1.940503129326325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17124010129520478, "3000–6000 Hz_10ms_envelope_correlation": 0.9818489364481687, "3000–6000 Hz_10ms_modulation_mae_db": 0.2366708408355922, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0132557474263082, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09674124572858049, "3000–6000 Hz_40ms_envelope_correlation": 0.9938920938654762, "3000–6000 Hz_40ms_modulation_mae_db": 0.7716733382810427, "mr_stft_db": 5.929327800217302, "band_mae_db": 0.16683657337422916, "rms_gain_error_db": -0.07772315323883205, "output_peak": 0.45252485206414017}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2906960333511415, "全帯域_10ms_rms_envelope_nrmse": 0.10493870413006401, "全帯域_10ms_envelope_correlation": 0.9902144843672392, "全帯域_10ms_modulation_mae_db": 0.33564757387717736, "全帯域_40ms_log_rms_mae_db": 0.7299860590984067, "全帯域_40ms_rms_envelope_nrmse": 0.05231361418815035, "全帯域_40ms_envelope_correlation": 0.9974213281825923, "全帯域_40ms_modulation_mae_db": 0.4428845436617392, "300–1000 Hz_10ms_log_rms_mae_db": 1.677102108813065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1122664214090841, "300–1000 Hz_10ms_envelope_correlation": 0.9911616959908851, "300–1000 Hz_10ms_modulation_mae_db": 0.3290493823721434, "300–1000 Hz_40ms_log_rms_mae_db": 0.9742630715999809, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061706531700492076, "300–1000 Hz_40ms_envelope_correlation": 0.9972667251788259, "300–1000 Hz_40ms_modulation_mae_db": 0.33958072468130324, "1000–3000 Hz_10ms_log_rms_mae_db": 1.738354151416679, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1650751574503974, "1000–3000 Hz_10ms_envelope_correlation": 0.9780897868373236, "1000–3000 Hz_10ms_modulation_mae_db": 1.276708530897813, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9328269595673973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08486799734980546, "1000–3000 Hz_40ms_envelope_correlation": 0.9937576380220424, "1000–3000 Hz_40ms_modulation_mae_db": 1.329397131259493, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8862559419885987, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14445300294957633, "3000–6000 Hz_10ms_envelope_correlation": 0.9870734358058368, "3000–6000 Hz_10ms_modulation_mae_db": 0.3072807207758687, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9814588852683178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07800828031852883, "3000–6000 Hz_40ms_envelope_correlation": 0.996114481833926, "3000–6000 Hz_40ms_modulation_mae_db": 0.4475719725374076, "mr_stft_db": 5.914966911309937, "band_mae_db": 0.15177490277596953, "rms_gain_error_db": -0.09139303073716237, "output_peak": 0.5180782526561716}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.239882120988017, "全帯域_10ms_rms_envelope_nrmse": 0.1008560724765197, "全帯域_10ms_envelope_correlation": 0.9909943376877368, "全帯域_10ms_modulation_mae_db": 0.27106861975775, "全帯域_40ms_log_rms_mae_db": 0.5861859256729354, "全帯域_40ms_rms_envelope_nrmse": 0.05098920412628466, "全帯域_40ms_envelope_correlation": 0.9974937465647208, "全帯域_40ms_modulation_mae_db": 0.4759717407701229, "300–1000 Hz_10ms_log_rms_mae_db": 1.6297472220478688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1159076334482527, "300–1000 Hz_10ms_envelope_correlation": 0.9906181624155531, "300–1000 Hz_10ms_modulation_mae_db": 0.08982053064720578, "300–1000 Hz_40ms_log_rms_mae_db": 0.9014173957082448, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06854975258657908, "300–1000 Hz_40ms_envelope_correlation": 0.9966742523528878, "300–1000 Hz_40ms_modulation_mae_db": 0.08364795740832888, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7886751405700743, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18208646530067849, "1000–3000 Hz_10ms_envelope_correlation": 0.9737992654290059, "1000–3000 Hz_10ms_modulation_mae_db": 0.24737129891521106, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9310932170925986, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09027485405633878, "1000–3000 Hz_40ms_envelope_correlation": 0.9930030478425534, "1000–3000 Hz_40ms_modulation_mae_db": 0.5279672453295675, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9044388299148314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14166492084202728, "3000–6000 Hz_10ms_envelope_correlation": 0.9878594819845651, "3000–6000 Hz_10ms_modulation_mae_db": 0.5344756170290471, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9752450823067919, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07802588702548709, "3000–6000 Hz_40ms_envelope_correlation": 0.996195925874078, "3000–6000 Hz_40ms_modulation_mae_db": 0.8510081338835764, "mr_stft_db": 5.9089506521379, "band_mae_db": 0.0959450783302428, "rms_gain_error_db": 0.006843300721007706, "output_peak": 0.45315908836458285}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.3097646400895915, "spectral_convergence": 0.46449820186794644, "spectral_convergence_status": "ok", "amplitude_floor": 1.32832143223519e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.660346595329801, "spectral_convergence": 0.4340932789292474, "spectral_convergence_status": "ok", "amplitude_floor": 1.1133586355586426e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.808006470882643, "spectral_convergence": 0.45899370531325295, "spectral_convergence_status": "ok", "amplitude_floor": 6.710028542715064e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00015990772541983682, "output_power": 0.00015765560718529706, "delta_db": -0.061600242808904505, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005346827040387907, "output_power": 0.0005237408466639444, "delta_db": -0.08979689788415801, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025601253900142847, "output_power": 0.00026634539302654867, "delta_db": 0.1718395245811258, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015659557404440672, "output_power": 0.00015211147462982255, "delta_db": -0.1261750658670286, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003554436518358531, "output_power": 0.0003608409083843655, "delta_db": 0.06545005239873371, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004235024710497043, "output_power": 0.000395958358318906, "delta_db": -0.29206433735251425, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13448768681541082, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06844352289206435, "pitch": null, "output_rms_dbfs": -27.312693526624734, "output_peak": 0.4978483618956231, "reference_rms_dbfs": -27.244250003732674, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.926039235434011}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018816516597127566, "log_rms_mae_db": 1.2906199709437864, "rms_envelope_nrmse": 0.10653515830466742, "envelope_correlation": 0.9899077881918248, "rms_gain_error_db": -0.06818527282190555, "reference_mean_power": 0.0018869472812584292, "output_mean_power": 0.0018575531090067546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.448172186664129e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039739520206635815, "output_power": 0.00039493250388494415, "reference_active": true, "delta_db": -0.026997406930401882}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001684948117890412, "output_power": 0.00016729382667357107, "reference_active": true, "delta_db": -0.031066175136156927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9559595301379592e-05, "output_power": 1.9586368842741152e-05, "reference_active": true, "delta_db": 0.0059406395226656835}], "active_band_count": 3, "active_band_mae_db": 0.021334740529741495}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016656172387320294, "log_rms_mae_db": 0.7072979727113958, "rms_envelope_nrmse": 0.05792056729441475, "envelope_correlation": 0.9967798930899348, "rms_gain_error_db": -0.06792315554757034, "reference_mean_power": 0.0018945898208991615, "output_mean_power": 0.0018651891655664338}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.514104383617561e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003829793712831679, "output_power": 0.0003835696899106363, "reference_active": true, "delta_db": 0.006688995816019789}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.831616093630877e-05, "output_power": 7.794693045099875e-05, "reference_active": true, "delta_db": -0.020523728198161735}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3812315351816923e-06, "output_power": 2.0104609406843288e-06, "reference_active": true, "delta_db": -0.7350598534643371}], "active_band_count": 3, "active_band_mae_db": 0.2540908591595062}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015092951870698086, "log_rms_mae_db": 1.7315946675979448, "rms_envelope_nrmse": 0.11367691746223023, "envelope_correlation": 0.9909563886953534, "rms_gain_error_db": -0.02141226203328747, "reference_mean_power": 0.000631981580682194, "output_mean_power": 0.0006288733554210467}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.167139764996573e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018037000728302603, "output_power": 0.00017443126530589922, "reference_active": true, "delta_db": -0.14539991686705267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6944061726753776e-05, "output_power": 2.81078042252253e-05, "reference_active": true, "delta_db": 0.183638551984685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.521461134259169e-06, "output_power": 4.729383782489129e-06, "reference_active": true, "delta_db": 1.2808165827562568}], "active_band_count": 3, "active_band_mae_db": 0.5366183505359982}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013588907795973108, "log_rms_mae_db": 0.9471412732280549, "rms_envelope_nrmse": 0.06196344316297622, "envelope_correlation": 0.9971746149267005, "rms_gain_error_db": -0.020935207923954615, "reference_mean_power": 0.0006345375227010602, "output_mean_power": 0.0006314860890163034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.990151794989553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001715162379513309, "output_power": 0.00016698140186989057, "reference_active": true, "delta_db": -0.11637139478968378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2552077755571312e-05, "output_power": 1.256313786875545e-05, "reference_active": true, "delta_db": 0.003825048932447895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.057253626713498e-07, "output_power": 4.1886823002965656e-07, "reference_active": true, "delta_db": 1.3674595274450057}], "active_band_count": 3, "active_band_mae_db": 0.4958853237223791}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.324369547066356e-05, "log_rms_mae_db": 1.772672767034512, "rms_envelope_nrmse": 0.17770412984137482, "envelope_correlation": 0.9748270860384447, "rms_gain_error_db": 2.3269479440040146e-05, "reference_mean_power": 0.00026216016454011753, "output_mean_power": 0.0002621615691969013}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5456728460340458e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.502085853898132e-05, "output_power": 7.438049192891515e-05, "reference_active": true, "delta_db": -0.0372298359277497}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151594827944158e-05, "output_power": 2.290856502706766e-05, "reference_active": true, "delta_db": 0.272373945627871}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.513878290774259e-06, "output_power": 4.428940095789886e-06, "reference_active": true, "delta_db": -2.8382763125821415}], "active_band_count": 3, "active_band_mae_db": 1.0492933647125875}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.873751028776232e-05, "log_rms_mae_db": 0.944800025431722, "rms_envelope_nrmse": 0.09981833629595428, "envelope_correlation": 0.9913187511519602, "rms_gain_error_db": 0.0006357106262508103, "reference_mean_power": 0.0002632887650860277, "output_mean_power": 0.0002633273075320718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3635149275752304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.257955032157826e-05, "output_power": 7.240126095948499e-05, "reference_active": true, "delta_db": -0.010681430802709772}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.405048683009261e-06, "output_power": 9.971127295426524e-06, "reference_active": true, "delta_db": 0.2538321275461269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.677553554297748e-07, "output_power": 4.2910569781191127e-07, "reference_active": true, "delta_db": -2.5265857323852865}], "active_band_count": 3, "active_band_mae_db": 0.9303664302447077}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015682729984940356, "log_rms_mae_db": 1.9983705526541047, "rms_envelope_nrmse": 0.16075902257345398, "envelope_correlation": 0.9841473712415797, "rms_gain_error_db": -0.26476182362114736, "reference_mean_power": 0.0005797089728212459, "output_mean_power": 0.000545423492724968}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2924452919576597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020605682430557354, "output_power": 0.00021050967662538796, "reference_active": true, "delta_db": 0.09285061799943727}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012645863441832237, "output_power": 0.00011715100162236975, "reference_active": true, "delta_db": -0.33202481613976353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.26990748699409e-06, "output_power": 1.1240227679111776e-05, "reference_active": true, "delta_db": 0.8369970833894287}], "active_band_count": 3, "active_band_mae_db": 0.4206241725095432}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012704534221320322, "log_rms_mae_db": 1.040769996872589, "rms_envelope_nrmse": 0.0910153344361166, "envelope_correlation": 0.9950655311319442, "rms_gain_error_db": -0.2645085790794894, "reference_mean_power": 0.0005824704905358117, "output_mean_power": 0.00054805364419099}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.040463159930301e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022945963060392102, "output_power": 0.00023251124998897006, "reference_active": true, "delta_db": 0.05737680846706314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.01414803619856e-05, "output_power": 5.6418014679088065e-05, "reference_active": true, "delta_db": -0.2775631441747507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3285241811094974e-06, "output_power": 1.1529251883391078e-06, "reference_active": true, "delta_db": -0.6156833615830962}], "active_band_count": 3, "active_band_mae_db": 0.31687443807497}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8130264345379201, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "96558e35dc5bf743d4d09a5dedac0b6648630aae9da7791ecbe8aebc0cd50ee9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9753156325374264, "standard_deviation": 0.0020604577087031373, "minimum": 0.9722656151418944, "maximum": 0.9780897868373236, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.779288629814989, "standard_deviation": 0.038733991670376944, "minimum": 1.7364724446541109, "maximum": 1.8563360550506605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7105550037697, "standard_deviation": 0.39111909342569573, "minimum": 0.24737129891521106, "maximum": 1.276708530897813, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17553113765215356, "standard_deviation": 0.007706782667684582, "minimum": 0.1650751574503974, "maximum": 0.18684656401138341, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929990224587528, "standard_deviation": 0.0012739686186371028, "minimum": 0.9911080578225177, "maximum": 0.994898491685709, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.945315937924464, "standard_deviation": 0.027638604527358272, "minimum": 0.9046655386872064, "maximum": 0.985599769085111, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7776095851160567, "standard_deviation": 0.3422445214218005, "minimum": 0.3893638767460952, "maximum": 1.329397131259493, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09126513644117024, "standard_deviation": 0.00832481297510718, "minimum": 0.08213127693943255, "maximum": 0.10678875613748408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9858895368196329, "standard_deviation": 0.0019872498834886338, "minimum": 0.9818489364481687, "maximum": 0.9878594819845651, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.912232176881725, "standard_deviation": 0.04107750120994629, "minimum": 1.8759898780698805, "maximum": 1.9983705526541047, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4353262614808022, "standard_deviation": 0.16613194485392097, "minimum": 0.23621867505414204, "maximum": 0.6784675703049793, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1516103471439665, "standard_deviation": 0.010318461936131864, "minimum": 0.14166492084202728, "maximum": 0.17124010129520478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9956961289406016, "standard_deviation": 0.0010460231233654059, "minimum": 0.9938920938654762, "maximum": 0.9971741397653497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.004244518614152, "standard_deviation": 0.022973579226249884, "minimum": 0.9752450823067919, "maximum": 1.040769996872589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6033336806912177, "standard_deviation": 0.21675690666793723, "minimum": 0.31687443807497, "maximum": 0.8510081338835764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08257667294663049, "standard_deviation": 0.010845676060593818, "minimum": 0.06587670163108436, "maximum": 0.09674124572858049, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9899040714145387, "standard_deviation": 0.0016364175845856625, "minimum": 0.9870150146526249, "maximum": 0.9915741714234051, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7122566367235925, "standard_deviation": 0.04945746883496622, "minimum": 1.6297472220478688, "maximum": 1.772777345685823, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5132942822262857, "standard_deviation": 0.2486174700876614, "minimum": 0.08982053064720578, "maximum": 0.9012500800655848, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11972129444902468, "standard_deviation": 0.00948474723979899, "minimum": 0.10951688629210836, "maximum": 0.13606635585999027, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967328073681552, "standard_deviation": 0.0008231933118192109, "minimum": 0.995279697585384, "maximum": 0.9976195764609184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.956280823739982, "standard_deviation": 0.030369762164205442, "minimum": 0.9014173957082448, "maximum": 0.9985106575624103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3826415337722373, "standard_deviation": 0.210101633248576, "minimum": 0.08364795740832888, "maximum": 0.7096041827217526, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06707460632543946, "standard_deviation": 0.00840641376775149, "minimum": 0.0581209319565605, "maximum": 0.08116353653198102, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14675072625459157, "standard_deviation": 0.034773585863496, "minimum": 0.0959450783302428, "maximum": 0.20967324660610523, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.917697953016347, "standard_deviation": 0.010553828687028434, "minimum": 5.90164999488118, "maximum": 5.929327800217302, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4714910881495229, "standard_deviation": 0.035416321097192796, "minimum": 0.4194679067226995, "maximum": 0.5180782526561716, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06350329650960378, "standard_deviation": 0.050815813545154125, "minimum": -0.1430262080590797, "maximum": 0.008126796626566832, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899688311961037, "standard_deviation": 0.0009210562296857615, "minimum": 0.9883838938328698, "maximum": 0.9909943376877368, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2767137422881931, "standard_deviation": 0.03714115866723478, "minimum": 1.2240311245204705, "maximum": 1.3294237587494089, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2625348522532528, "standard_deviation": 0.16118265554166045, "minimum": 0.021334740529741495, "maximum": 0.5296800352194758, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10622214946061169, "standard_deviation": 0.004833091805336549, "minimum": 0.1008560724765197, "maximum": 0.11418290195860616, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997161061797172, "standard_deviation": 0.00044733490318739073, "minimum": 0.9966582582821129, "maximum": 0.9977982970818432, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6815415855020366, "standard_deviation": 0.056749244518674066, "minimum": 0.5861859256729354, "maximum": 0.761377643384558, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3709261827290021, "standard_deviation": 0.15534532064994167, "minimum": 0.04255434749720186, "maximum": 0.5348604178463994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054571530283362514, "standard_deviation": 0.004412245627592939, "minimum": 0.048468425283630476, "maximum": 0.059075446333433354, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0012", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.953908920288086, "predicted_mos": 1.953908920288086, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fd645fafb8f6982f6c3d61254b7b57f2a61308fc98c6a73599a0393812c0cbc1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.803871300071478, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.187240062239824, "全帯域_10ms_rms_envelope_nrmse": 0.10493265443058092, "全帯域_10ms_envelope_correlation": 0.9902001766164129, "全帯域_10ms_modulation_mae_db": 0.1399374048707561, "全帯域_40ms_log_rms_mae_db": 0.5715183002900752, "全帯域_40ms_rms_envelope_nrmse": 0.05108482924417836, "全帯域_40ms_envelope_correlation": 0.9974740648620569, "全帯域_40ms_modulation_mae_db": 0.2740485097520136, "300–1000 Hz_10ms_log_rms_mae_db": 2.1331134857969882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13756725351708668, "300–1000 Hz_10ms_envelope_correlation": 0.9866753562987194, "300–1000 Hz_10ms_modulation_mae_db": 0.9569189283231312, "300–1000 Hz_40ms_log_rms_mae_db": 1.407343206118933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08011614510754497, "300–1000 Hz_40ms_envelope_correlation": 0.9953168655432864, "300–1000 Hz_40ms_modulation_mae_db": 0.7150220779798754, "1000–3000 Hz_10ms_log_rms_mae_db": 2.153453805944763, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19461852537074437, "1000–3000 Hz_10ms_envelope_correlation": 0.9697582729961625, "1000–3000 Hz_10ms_modulation_mae_db": 0.7861254326329453, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2178758526486995, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1107812213474879, "1000–3000 Hz_40ms_envelope_correlation": 0.9892010611430918, "1000–3000 Hz_40ms_modulation_mae_db": 0.730371413100079, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4070487659547983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14067716614600578, "3000–6000 Hz_10ms_envelope_correlation": 0.9880434732189098, "3000–6000 Hz_10ms_modulation_mae_db": 0.3157516999156486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3730193583193249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07360914628019255, "3000–6000 Hz_40ms_envelope_correlation": 0.9970157332531359, "3000–6000 Hz_40ms_modulation_mae_db": 0.49816770725021464, "mr_stft_db": 6.258620097405537, "band_mae_db": 0.1633367256514997, "rms_gain_error_db": -0.03786869474066756, "output_peak": 0.5143129868934739}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1883441351641568, "全帯域_10ms_rms_envelope_nrmse": 0.09995472092245254, "全帯域_10ms_envelope_correlation": 0.9911341315750387, "全帯域_10ms_modulation_mae_db": 0.03061061730321947, "全帯域_40ms_log_rms_mae_db": 0.6569833022331474, "全帯域_40ms_rms_envelope_nrmse": 0.05041436141332846, "全帯域_40ms_envelope_correlation": 0.9975345771799525, "全帯域_40ms_modulation_mae_db": 0.148386562006339, "300–1000 Hz_10ms_log_rms_mae_db": 2.0567883414163783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12184941261999291, "300–1000 Hz_10ms_envelope_correlation": 0.9896612726962392, "300–1000 Hz_10ms_modulation_mae_db": 0.37495574438556906, "300–1000 Hz_40ms_log_rms_mae_db": 1.3311588312681975, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06719780907829152, "300–1000 Hz_40ms_envelope_correlation": 0.9969160612001658, "300–1000 Hz_40ms_modulation_mae_db": 0.431770488032142, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1877596876540513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17956220570451173, "1000–3000 Hz_10ms_envelope_correlation": 0.9740701289097614, "1000–3000 Hz_10ms_modulation_mae_db": 0.2982596289786268, "1000–3000 Hz_40ms_log_rms_mae_db": 1.209536676125496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09523778214122743, "1000–3000 Hz_40ms_envelope_correlation": 0.9920046542511166, "1000–3000 Hz_40ms_modulation_mae_db": 0.5607778417791655, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4196381354955507, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16063165543096014, "3000–6000 Hz_10ms_envelope_correlation": 0.9840528057013841, "3000–6000 Hz_10ms_modulation_mae_db": 0.13341834146125034, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3596298500081294, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08888455551920625, "3000–6000 Hz_40ms_envelope_correlation": 0.9948498565302297, "3000–6000 Hz_40ms_modulation_mae_db": 0.16611655284090987, "mr_stft_db": 6.2457636130434535, "band_mae_db": 0.1488255697644753, "rms_gain_error_db": -0.005528311820065909, "output_peak": 0.4498715425262905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1907366207829853, "全帯域_10ms_rms_envelope_nrmse": 0.10089542824809128, "全帯域_10ms_envelope_correlation": 0.9909389572113797, "全帯域_10ms_modulation_mae_db": 0.20664061281616278, "全帯域_40ms_log_rms_mae_db": 0.5971258400711923, "全帯域_40ms_rms_envelope_nrmse": 0.05127465891377538, "全帯域_40ms_envelope_correlation": 0.9974569919312282, "全帯域_40ms_modulation_mae_db": 0.3961934973828731, "300–1000 Hz_10ms_log_rms_mae_db": 2.090483580043283, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12237227922298301, "300–1000 Hz_10ms_envelope_correlation": 0.9895006010945707, "300–1000 Hz_10ms_modulation_mae_db": 0.6461937572032581, "300–1000 Hz_40ms_log_rms_mae_db": 1.371680584537304, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06337357580332731, "300–1000 Hz_40ms_envelope_correlation": 0.9971358941691684, "300–1000 Hz_40ms_modulation_mae_db": 0.5973287495442068, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0895883756881872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19244840495097845, "1000–3000 Hz_10ms_envelope_correlation": 0.9702904375022854, "1000–3000 Hz_10ms_modulation_mae_db": 0.6780765601963514, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2033741029214544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09730257768416194, "1000–3000 Hz_40ms_envelope_correlation": 0.9916411337637382, "1000–3000 Hz_40ms_modulation_mae_db": 0.8890732664654707, "3000–6000 Hz_10ms_log_rms_mae_db": 2.36493363942778, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13964581061871242, "3000–6000 Hz_10ms_envelope_correlation": 0.988122664518056, "3000–6000 Hz_10ms_modulation_mae_db": 0.35961737003451133, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3868720847499356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07970895997906083, "3000–6000 Hz_40ms_envelope_correlation": 0.996239678949481, "3000–6000 Hz_40ms_modulation_mae_db": 0.17636120060526192, "mr_stft_db": 6.249529219658609, "band_mae_db": 0.12022494548202046, "rms_gain_error_db": -0.04134877832327661, "output_peak": 0.41891969939898155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2308483454412587, "全帯域_10ms_rms_envelope_nrmse": 0.10414384421153218, "全帯域_10ms_envelope_correlation": 0.9903718152230855, "全帯域_10ms_modulation_mae_db": 0.30489978138727464, "全帯域_40ms_log_rms_mae_db": 0.6489917665202866, "全帯域_40ms_rms_envelope_nrmse": 0.05325894105252728, "全帯域_40ms_envelope_correlation": 0.9972600587406337, "全帯域_40ms_modulation_mae_db": 0.35076322622905537, "300–1000 Hz_10ms_log_rms_mae_db": 2.0858286989205514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12021850802875496, "300–1000 Hz_10ms_envelope_correlation": 0.9899328016253389, "300–1000 Hz_10ms_modulation_mae_db": 0.6025619208463339, "300–1000 Hz_40ms_log_rms_mae_db": 1.238667922893675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06142080400393253, "300–1000 Hz_40ms_envelope_correlation": 0.9973350991630493, "300–1000 Hz_40ms_modulation_mae_db": 0.2610278949028484, "1000–3000 Hz_10ms_log_rms_mae_db": 2.19324692747752, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20370831474677242, "1000–3000 Hz_10ms_envelope_correlation": 0.9664745551914666, "1000–3000 Hz_10ms_modulation_mae_db": 0.44066440584417493, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2654061277843263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11069944552060595, "1000–3000 Hz_40ms_envelope_correlation": 0.9891942895442898, "1000–3000 Hz_40ms_modulation_mae_db": 0.5684979869982778, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3584445159956338, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14422353100313734, "3000–6000 Hz_10ms_envelope_correlation": 0.9871585410438998, "3000–6000 Hz_10ms_modulation_mae_db": 0.3665384881178089, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3417766442590067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08285108312097442, "3000–6000 Hz_40ms_envelope_correlation": 0.9957035797984306, "3000–6000 Hz_40ms_modulation_mae_db": 0.7000423973995109, "mr_stft_db": 6.255594546993463, "band_mae_db": 0.21347708168728566, "rms_gain_error_db": -0.0010328404091384706, "output_peak": 0.43186664980021955}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2599351217821713, "全帯域_10ms_rms_envelope_nrmse": 0.10180013032314483, "全帯域_10ms_envelope_correlation": 0.991088678833518, "全帯域_10ms_modulation_mae_db": 0.252206692297207, "全帯域_40ms_log_rms_mae_db": 0.6687301700519106, "全帯域_40ms_rms_envelope_nrmse": 0.055201622002461266, "全帯域_40ms_envelope_correlation": 0.9972549219085293, "全帯域_40ms_modulation_mae_db": 0.13355150170896637, "300–1000 Hz_10ms_log_rms_mae_db": 2.0869397836360792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12171806507467559, "300–1000 Hz_10ms_envelope_correlation": 0.9896422778645639, "300–1000 Hz_10ms_modulation_mae_db": 0.7401445619219871, "300–1000 Hz_40ms_log_rms_mae_db": 1.2871729449185252, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06508222221404435, "300–1000 Hz_40ms_envelope_correlation": 0.9969284993453114, "300–1000 Hz_40ms_modulation_mae_db": 0.09853629633706158, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1330024863086754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19053630470629715, "1000–3000 Hz_10ms_envelope_correlation": 0.9715227025993729, "1000–3000 Hz_10ms_modulation_mae_db": 0.18916978998941333, "1000–3000 Hz_40ms_log_rms_mae_db": 1.216975242337698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10953477073481334, "1000–3000 Hz_40ms_envelope_correlation": 0.9897856597939372, "1000–3000 Hz_40ms_modulation_mae_db": 0.5797243630186727, "3000–6000 Hz_10ms_log_rms_mae_db": 2.384143921065827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14130005399663254, "3000–6000 Hz_10ms_envelope_correlation": 0.9881613725950553, "3000–6000 Hz_10ms_modulation_mae_db": 0.2324002565558001, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3413048817346036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07916965839438216, "3000–6000 Hz_40ms_envelope_correlation": 0.9962509937868831, "3000–6000 Hz_40ms_modulation_mae_db": 0.37711011013177576, "mr_stft_db": 6.240108560513199, "band_mae_db": 0.1458183586032642, "rms_gain_error_db": 0.1087773409783927, "output_peak": 0.48171092166326385}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1712784310992574, "全帯域_10ms_rms_envelope_nrmse": 0.10268770035212418, "全帯域_10ms_envelope_correlation": 0.9906999571395094, "全帯域_10ms_modulation_mae_db": 0.5532756927749355, "全帯域_40ms_log_rms_mae_db": 0.646360478032311, "全帯域_40ms_rms_envelope_nrmse": 0.05174311833078175, "全帯域_40ms_envelope_correlation": 0.9974389418102314, "全帯域_40ms_modulation_mae_db": 0.7170924019265907, "300–1000 Hz_10ms_log_rms_mae_db": 2.1132817881271664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1252971834412242, "300–1000 Hz_10ms_envelope_correlation": 0.98896880083931, "300–1000 Hz_10ms_modulation_mae_db": 0.8105780658543255, "300–1000 Hz_40ms_log_rms_mae_db": 1.3150263425708901, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06531171896136947, "300–1000 Hz_40ms_envelope_correlation": 0.9969337574090384, "300–1000 Hz_40ms_modulation_mae_db": 0.6202239193779723, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1965825427739314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19832660277728595, "1000–3000 Hz_10ms_envelope_correlation": 0.9684320051616893, "1000–3000 Hz_10ms_modulation_mae_db": 0.5348379919484678, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2786052965415893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10788879212441652, "1000–3000 Hz_40ms_envelope_correlation": 0.9896957419682763, "1000–3000 Hz_40ms_modulation_mae_db": 0.7038143192395032, "3000–6000 Hz_10ms_log_rms_mae_db": 2.357859030223826, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1432231486727931, "3000–6000 Hz_10ms_envelope_correlation": 0.9873697784182613, "3000–6000 Hz_10ms_modulation_mae_db": 0.7466211504568383, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3783665340945523, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07640299296487227, "3000–6000 Hz_40ms_envelope_correlation": 0.996207651454002, "3000–6000 Hz_40ms_modulation_mae_db": 0.9137380548719031, "mr_stft_db": 6.271633208392983, "band_mae_db": 0.1211949195542678, "rms_gain_error_db": 0.03599489199028824, "output_peak": 0.4758526137327783}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2599092531020397, "全帯域_10ms_rms_envelope_nrmse": 0.11466103768460449, "全帯域_10ms_envelope_correlation": 0.9882850095475955, "全帯域_10ms_modulation_mae_db": 0.19943916251600202, "全帯域_40ms_log_rms_mae_db": 0.6466534689994723, "全帯域_40ms_rms_envelope_nrmse": 0.05955769018321466, "全帯域_40ms_envelope_correlation": 0.9965731758630211, "全帯域_40ms_modulation_mae_db": 0.49642956246121256, "300–1000 Hz_10ms_log_rms_mae_db": 2.1340415544276317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12867329961497664, "300–1000 Hz_10ms_envelope_correlation": 0.9884495156124284, "300–1000 Hz_10ms_modulation_mae_db": 0.4813911432086147, "300–1000 Hz_40ms_log_rms_mae_db": 1.3438196585458801, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0807963756919112, "300–1000 Hz_40ms_envelope_correlation": 0.9953955283047019, "300–1000 Hz_40ms_modulation_mae_db": 0.5185559712315791, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1079251548796725, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1877392530170984, "1000–3000 Hz_10ms_envelope_correlation": 0.9715446975172387, "1000–3000 Hz_10ms_modulation_mae_db": 0.754945922524248, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2238969019499266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0982556775935104, "1000–3000 Hz_40ms_envelope_correlation": 0.9914469746377085, "1000–3000 Hz_40ms_modulation_mae_db": 0.8707485167039491, "3000–6000 Hz_10ms_log_rms_mae_db": 2.394812243017509, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17847370433330936, "3000–6000 Hz_10ms_envelope_correlation": 0.9801864989343196, "3000–6000 Hz_10ms_modulation_mae_db": 0.47263185279187586, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3050576109325995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09715521977008534, "3000–6000 Hz_40ms_envelope_correlation": 0.9939825956962146, "3000–6000 Hz_40ms_modulation_mae_db": 0.8701461686799113, "mr_stft_db": 6.232548931383174, "band_mae_db": 0.13583847731373702, "rms_gain_error_db": -0.05651855983386528, "output_peak": 0.5127267466775451}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2289219234402597, "全帯域_10ms_rms_envelope_nrmse": 0.09855871707616164, "全帯域_10ms_envelope_correlation": 0.9913986310584603, "全帯域_10ms_modulation_mae_db": 0.4801736698805443, "全帯域_40ms_log_rms_mae_db": 0.6506070671311225, "全帯域_40ms_rms_envelope_nrmse": 0.05639714132432074, "全帯域_40ms_envelope_correlation": 0.9970122245316413, "全帯域_40ms_modulation_mae_db": 0.35705493850875375, "300–1000 Hz_10ms_log_rms_mae_db": 2.1281731209471655, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1245046328028086, "300–1000 Hz_10ms_envelope_correlation": 0.9892031640095822, "300–1000 Hz_10ms_modulation_mae_db": 0.8102865074528465, "300–1000 Hz_40ms_log_rms_mae_db": 1.3757159281897595, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07006051265422848, "300–1000 Hz_40ms_envelope_correlation": 0.9966069346943791, "300–1000 Hz_40ms_modulation_mae_db": 0.6636095812747741, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1756038336897965, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18937029812309897, "1000–3000 Hz_10ms_envelope_correlation": 0.97116030647513, "1000–3000 Hz_10ms_modulation_mae_db": 0.4354128731759889, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2655718922308807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1152639354950805, "1000–3000 Hz_40ms_envelope_correlation": 0.9882119612728479, "1000–3000 Hz_40ms_modulation_mae_db": 0.04467425461580673, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3504470873140213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1377397076918623, "3000–6000 Hz_10ms_envelope_correlation": 0.9888867318507732, "3000–6000 Hz_10ms_modulation_mae_db": 0.5493368270596931, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3566749111668222, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09113680665603008, "3000–6000 Hz_40ms_envelope_correlation": 0.9954732783305129, "3000–6000 Hz_40ms_modulation_mae_db": 0.8007911473920921, "mr_stft_db": 6.244339311648303, "band_mae_db": 0.2323179474870448, "rms_gain_error_db": -0.10150379989779681, "output_peak": 0.4835456160810527}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.718621649384091, "spectral_convergence": 0.47503132061479025, "spectral_convergence_status": "ok", "amplitude_floor": 1.32832143223519e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.079202294236256, "spectral_convergence": 0.44348353708564053, "spectral_convergence_status": "ok", "amplitude_floor": 1.1133586355586426e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.978036348596263, "spectral_convergence": 0.462241145980303, "spectral_convergence_status": "ok", "amplitude_floor": 6.710028542715064e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00015990772541983682, "output_power": 0.00017615029827456157, "delta_db": 0.42013937133415885, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005346827040387907, "output_power": 0.0005383942687284828, "delta_db": 0.030042923906908988, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025601253900142847, "output_power": 0.0002563862390716677, "delta_db": 0.0063347499686537, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015659557404440672, "output_power": 0.00015008580502966305, "delta_db": -0.18439864133947634, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003554436518358531, "output_power": 0.0003438293605597805, "delta_db": -0.14427802640305606, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004235024710497043, "output_power": 0.00040492378207336696, "delta_db": -0.19482664095674423, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1633367256514997, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03786869474066756, "pitch": null, "output_rms_dbfs": -27.28211869847334, "output_peak": 0.5143129868934739, "reference_rms_dbfs": -27.244250003732674, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.258620097405537}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018816516597127566, "log_rms_mae_db": 1.187240062239824, "rms_envelope_nrmse": 0.10493265443058092, "envelope_correlation": 0.9902001766164129, "rms_gain_error_db": -0.03733454904486584, "reference_mean_power": 0.0018869472812584292, "output_mean_power": 0.0018707954795133073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.448172186664129e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039739520206635815, "output_power": 0.00039234021643195886, "reference_active": true, "delta_db": -0.0555979204772784}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001684948117890412, "output_power": 0.00016151009935383328, "reference_active": true, "delta_db": -0.18386848534522995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9559595301379592e-05, "output_power": 2.0388930506956255e-05, "reference_active": true, "delta_db": 0.18034580878975992}], "active_band_count": 3, "active_band_mae_db": 0.1399374048707561}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016656172387320294, "log_rms_mae_db": 0.5715183002900752, "rms_envelope_nrmse": 0.05108482924417836, "envelope_correlation": 0.9974740648620569, "rms_gain_error_db": -0.03703212848104241, "reference_mean_power": 0.0018945898208991615, "output_mean_power": 0.0018785034057639764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.514104383617561e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003829793712831679, "output_power": 0.0003791844859204993, "reference_active": true, "delta_db": -0.043248216945072096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.831616093630877e-05, "output_power": 7.68681602422668e-05, "reference_active": true, "delta_db": -0.0810490335423518}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3812315351816923e-06, "output_power": 2.0277611470867646e-06, "reference_active": true, "delta_db": -0.6978482787686169}], "active_band_count": 3, "active_band_mae_db": 0.2740485097520136}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015092951870698086, "log_rms_mae_db": 2.1331134857969882, "rms_envelope_nrmse": 0.13756725351708668, "envelope_correlation": 0.9866753562987194, "rms_gain_error_db": -0.06101695382382328, "reference_mean_power": 0.000631981580682194, "output_mean_power": 0.0006231645296158578}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.167139764996573e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018037000728302603, "output_power": 0.000175918529401964, "reference_active": true, "delta_db": -0.10852736949982861}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6944061726753776e-05, "output_power": 2.4208914508594905e-05, "reference_active": true, "delta_db": -0.4648774799032849}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.521461134259169e-06, "output_power": 5.976653706354234e-06, "reference_active": true, "delta_db": 2.2973519355662804}], "active_band_count": 3, "active_band_mae_db": 0.9569189283231312}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013588907795973108, "log_rms_mae_db": 1.407343206118933, "rms_envelope_nrmse": 0.08011614510754497, "envelope_correlation": 0.9953168655432864, "rms_gain_error_db": -0.06116810426674766, "reference_mean_power": 0.0006345375227010602, "output_mean_power": 0.0006256630368188987}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.990151794989553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001715162379513309, "output_power": 0.00016731677379439665, "reference_active": true, "delta_db": -0.10765760358987825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2552077755571312e-05, "output_power": 1.1580146893226635e-05, "reference_active": true, "delta_db": -0.35001552467356256}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.057253626713498e-07, "output_power": 4.508901826744156e-07, "reference_active": true, "delta_db": 1.6873931056761853}], "active_band_count": 3, "active_band_mae_db": 0.7150220779798754}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.324369547066356e-05, "log_rms_mae_db": 2.153453805944763, "rms_envelope_nrmse": 0.19461852537074437, "envelope_correlation": 0.9697582729961625, "rms_gain_error_db": 0.02747539829316229, "reference_mean_power": 0.00026216016454011753, "output_mean_power": 0.00026382396362330815}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5456728460340458e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.502085853898132e-05, "output_power": 7.809451247018567e-05, "reference_active": true, "delta_db": 0.17438488119555895}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151594827944158e-05, "output_power": 2.496524732116654e-05, "reference_active": true, "delta_db": 0.6457538115848617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.513878290774259e-06, "output_power": 5.974529047527017e-06, "reference_active": true, "delta_db": -1.5382376051184157}], "active_band_count": 3, "active_band_mae_db": 0.7861254326329453}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.873751028776232e-05, "log_rms_mae_db": 1.2178758526486995, "rms_envelope_nrmse": 0.1107812213474879, "envelope_correlation": 0.9892010611430918, "rms_gain_error_db": 0.02817965001140285, "reference_mean_power": 0.0002632887650860277, "output_mean_power": 0.00026500269617098574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3635149275752304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.257955032157826e-05, "output_power": 7.655292388237561e-05, "reference_active": true, "delta_db": 0.23147509777275074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.405048683009261e-06, "output_power": 1.1028368301563204e-05, "reference_active": true, "delta_db": 0.6915021338774053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.677553554297748e-07, "output_power": 5.733359228994102e-07, "reference_active": true, "delta_db": -1.268137007650081}], "active_band_count": 3, "active_band_mae_db": 0.730371413100079}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015682729984940356, "log_rms_mae_db": 2.4070487659547983, "rms_envelope_nrmse": 0.14067716614600578, "envelope_correlation": 0.9880434732189098, "rms_gain_error_db": -0.2739282954844866, "reference_mean_power": 0.0005797089728212459, "output_mean_power": 0.0005442735042298132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2924452919576597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020605682430557354, "output_power": 0.00019044098571545742, "reference_active": true, "delta_db": -0.3422658160685019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012645863441832237, "output_power": 0.00011234198294453687, "reference_active": true, "delta_db": -0.5140640238637803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.26990748699409e-06, "output_power": 9.077847317340487e-06, "reference_active": true, "delta_db": -0.09092525981466372}], "active_band_count": 3, "active_band_mae_db": 0.3157516999156486}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012704534221320322, "log_rms_mae_db": 1.3730193583193249, "rms_envelope_nrmse": 0.07360914628019255, "envelope_correlation": 0.9970157332531359, "rms_gain_error_db": -0.27364734540698954, "reference_mean_power": 0.0005824704905358117, "output_mean_power": 0.0005469015991118813}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.040463159930301e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022945963060392102, "output_power": 0.00021139039142716872, "reference_active": true, "delta_db": -0.35621047170548004}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.01414803619856e-05, "output_power": 5.574480323591792e-05, "reference_active": true, "delta_db": -0.3296972705000007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3285241811094974e-06, "output_power": 1.1028334384812786e-06, "reference_active": true, "delta_db": -0.8085953795451633}], "active_band_count": 3, "active_band_mae_db": 0.49816770725021464}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8010887383891598, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "71c51c959f8ed98982a83454f6a72d306f8128ddb898f12ba383c6d8366fcb3f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9704066382941383, "standard_deviation": 0.0022779781531696887, "minimum": 0.9664745551914666, "maximum": 0.9740701289097614, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.154645351802075, "standard_deviation": 0.04083004353071222, "minimum": 2.0895883756881872, "maximum": 2.1965825427739314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5146865756612771, "standard_deviation": 0.21483261789739383, "minimum": 0.18916978998941333, "maximum": 0.7861254326329453, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19203873867459842, "standard_deviation": 0.007231544996081286, "minimum": 0.17956220570451173, "maximum": 0.20370831474677242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9901476845468757, "standard_deviation": 0.00137603354896826, "minimum": 0.9882119612728479, "maximum": 0.9920046542511166, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2351552615675088, "standard_deviation": 0.029641639546476056, "minimum": 1.2033741029214544, "maximum": 1.2786052965415893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6184602452401156, "standard_deviation": 0.265449625336992, "minimum": 0.04467425461580673, "maximum": 0.8890732664654707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.105620525330163, "standard_deviation": 0.007532031750025785, "minimum": 0.09523778214122743, "maximum": 0.1152639354950805, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9864977332850824, "standard_deviation": 0.0029401458390376535, "minimum": 0.9801864989343196, "maximum": 0.9888867318507732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3796659173118684, "standard_deviation": 0.02561732271814676, "minimum": 2.3504470873140213, "maximum": 2.4196381354955507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3970394982991783, "standard_deviation": 0.19148168120401104, "minimum": 0.13341834146125034, "maximum": 0.7466211504568383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14823934723667664, "standard_deviation": 0.014134084209948553, "minimum": 0.1377397076918623, "maximum": 0.17847370433330936, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9957154209748612, "standard_deviation": 0.0009490932555681431, "minimum": 0.9939825956962146, "maximum": 0.9970157332531359, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3553377344081219, "standard_deviation": 0.026094624311000812, "minimum": 1.3050576109325995, "maximum": 1.3868720847499356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5628091673964475, "standard_deviation": 0.30200407243261007, "minimum": 0.16611655284090987, "maximum": 0.9137380548719031, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0836148028356005, "standard_deviation": 0.008065936095020725, "minimum": 0.07360914628019255, "maximum": 0.09715521977008534, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9890042237550941, "standard_deviation": 0.0010501449180704758, "minimum": 0.9866753562987194, "maximum": 0.9899328016253389, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.103581294164406, "standard_deviation": 0.027908166695677736, "minimum": 2.0567883414163783, "maximum": 2.1340415544276317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6778788286495083, "standard_deviation": 0.19046096922621938, "minimum": 0.37495574438556906, "maximum": 0.9569189283231312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12527507929031284, "standard_deviation": 0.0056267502470711385, "minimum": 0.12021850802875496, "maximum": 0.13756725351708668, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965710799786376, "standard_deviation": 0.0007780283779211363, "minimum": 0.9953168655432864, "maximum": 0.9973350991630493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3338231773803955, "standard_deviation": 0.053831080364768545, "minimum": 1.238667922893675, "maximum": 1.407343206118933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48825937233505745, "standard_deviation": 0.2135104177530064, "minimum": 0.09853629633706158, "maximum": 0.7150220779798754, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06916989543933122, "standard_deviation": 0.007415409496737089, "minimum": 0.06142080400393253, "maximum": 0.0807963756919112, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16012925319294938, "standard_deviation": 0.04157323853174037, "minimum": 0.12022494548202046, "maximum": 0.2323179474870448, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.249767186129841, "standard_deviation": 0.012105249714264181, "minimum": 6.232548931383174, "maximum": 6.271633208392983, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4711008470967007, "standard_deviation": 0.03505690796073009, "minimum": 0.41891969939898155, "maximum": 0.5143129868934739, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.012378594007016214, "standard_deviation": 0.06389171972224779, "minimum": -0.10150379989779681, "maximum": 0.1087773409783927, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9905146696506251, "standard_deviation": 0.0009855858451566118, "minimum": 0.9882850095475955, "maximum": 0.9913986310584603, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.214651736631494, "standard_deviation": 0.034767518344897365, "minimum": 1.1712784310992574, "maximum": 1.2599351217821713, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2708979542307627, "standard_deviation": 0.1728595774131258, "minimum": 0.03061061730321947, "maximum": 0.5532756927749355, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10345427915608651, "standard_deviation": 0.0049905272472238705, "minimum": 0.09855871707616164, "maximum": 0.11466103768460449, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972506196034118, "standard_deviation": 0.000321710677236435, "minimum": 0.9965731758630211, "maximum": 0.9975345771799525, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6358712991661897, "standard_deviation": 0.03334561165156843, "minimum": 0.5715183002900752, "maximum": 0.6687301700519106, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3591900249969756, "standard_deviation": 0.18934984604253177, "minimum": 0.13355150170896637, "maximum": 0.7170924019265907, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.053616545308073485, "standard_deviation": 0.0031921106768015802, "minimum": 0.05041436141332846, "maximum": 0.05955769018321466, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0012", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.110881805419922, "predicted_mos": 2.110881805419922, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "356734d5773ee47ea118841215919ab4a288a568fbe2fbc7073be1d4615f73f8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7169675999321043, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2724906108957652, "全帯域_10ms_rms_envelope_nrmse": 0.10433291920985033, "全帯域_10ms_envelope_correlation": 0.9905409262606536, "全帯域_10ms_modulation_mae_db": 0.2862767066883311, "全帯域_40ms_log_rms_mae_db": 0.7052613173738217, "全帯域_40ms_rms_envelope_nrmse": 0.054138765488526495, "全帯域_40ms_envelope_correlation": 0.9975571893845752, "全帯域_40ms_modulation_mae_db": 0.3326341364778162, "300–1000 Hz_10ms_log_rms_mae_db": 1.564857325232049, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10700795573944553, "300–1000 Hz_10ms_envelope_correlation": 0.9921526892156765, "300–1000 Hz_10ms_modulation_mae_db": 0.29741171899024965, "300–1000 Hz_40ms_log_rms_mae_db": 0.9136986528714797, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06381828108431195, "300–1000 Hz_40ms_envelope_correlation": 0.9973207744507664, "300–1000 Hz_40ms_modulation_mae_db": 0.22923878569749545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.573254528240287, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17591516442784852, "1000–3000 Hz_10ms_envelope_correlation": 0.9750897436400111, "1000–3000 Hz_10ms_modulation_mae_db": 0.9423682610188896, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8771558803337726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09080061967450666, "1000–3000 Hz_40ms_envelope_correlation": 0.9930769096749009, "1000–3000 Hz_40ms_modulation_mae_db": 1.0525067222715367, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6191435074552394, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15286185998205068, "3000–6000 Hz_10ms_envelope_correlation": 0.9858534858706584, "3000–6000 Hz_10ms_modulation_mae_db": 0.7269710357753439, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8910231183587589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08580953660427196, "3000–6000 Hz_40ms_envelope_correlation": 0.9958740962900824, "3000–6000 Hz_40ms_modulation_mae_db": 0.3325220935413417, "mr_stft_db": 5.8036566423853175, "band_mae_db": 0.20960982192845432, "rms_gain_error_db": -0.18480572029145284, "output_peak": 0.4150892216167527}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.288904351770732, "全帯域_10ms_rms_envelope_nrmse": 0.11622851370652114, "全帯域_10ms_envelope_correlation": 0.9880043762712015, "全帯域_10ms_modulation_mae_db": 0.2317742086224496, "全帯域_40ms_log_rms_mae_db": 0.740206173512806, "全帯域_40ms_rms_envelope_nrmse": 0.06676532240252865, "全帯域_40ms_envelope_correlation": 0.9956721807845071, "全帯域_40ms_modulation_mae_db": 0.5311353092998743, "300–1000 Hz_10ms_log_rms_mae_db": 1.5927711799919095, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11019954234325834, "300–1000 Hz_10ms_envelope_correlation": 0.9918087618977192, "300–1000 Hz_10ms_modulation_mae_db": 1.0985997497960294, "300–1000 Hz_40ms_log_rms_mae_db": 0.8981257865218061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06469817232408805, "300–1000 Hz_40ms_envelope_correlation": 0.9974291426567735, "300–1000 Hz_40ms_modulation_mae_db": 0.5910646637109789, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6048394214096473, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18690418416082638, "1000–3000 Hz_10ms_envelope_correlation": 0.9717459397403986, "1000–3000 Hz_10ms_modulation_mae_db": 0.3206937761059872, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9309041045416454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09905794645229728, "1000–3000 Hz_40ms_envelope_correlation": 0.9914431301559112, "1000–3000 Hz_40ms_modulation_mae_db": 0.23457992942720474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6938574870648644, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17111376681937587, "3000–6000 Hz_10ms_envelope_correlation": 0.9831927917926521, "3000–6000 Hz_10ms_modulation_mae_db": 0.4868241787686948, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9421229395287068, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1067471690446149, "3000–6000 Hz_40ms_envelope_correlation": 0.993584811934662, "3000–6000 Hz_40ms_modulation_mae_db": 0.7646524930757437, "mr_stft_db": 5.801093045864, "band_mae_db": 0.22652508288818263, "rms_gain_error_db": -0.0036698133245457202, "output_peak": 0.45701301263196575}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3162157887078572, "全帯域_10ms_rms_envelope_nrmse": 0.11448036727970155, "全帯域_10ms_envelope_correlation": 0.9883252397402178, "全帯域_10ms_modulation_mae_db": 0.5090232293937207, "全帯域_40ms_log_rms_mae_db": 0.7258695243144728, "全帯域_40ms_rms_envelope_nrmse": 0.06588350533105161, "全帯域_40ms_envelope_correlation": 0.9958258187342144, "全帯域_40ms_modulation_mae_db": 0.6441890504668456, "300–1000 Hz_10ms_log_rms_mae_db": 1.6123396580504628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11833822906087126, "300–1000 Hz_10ms_envelope_correlation": 0.9902349123245839, "300–1000 Hz_10ms_modulation_mae_db": 0.36217173519620954, "300–1000 Hz_40ms_log_rms_mae_db": 0.9301961038371752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06807374116212651, "300–1000 Hz_40ms_envelope_correlation": 0.9967412355546369, "300–1000 Hz_40ms_modulation_mae_db": 0.6444165779671768, "1000–3000 Hz_10ms_log_rms_mae_db": 1.555732228837308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17700357290182392, "1000–3000 Hz_10ms_envelope_correlation": 0.9752564373217024, "1000–3000 Hz_10ms_modulation_mae_db": 0.5980423784232795, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8881436003537931, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10078990151277234, "1000–3000 Hz_40ms_envelope_correlation": 0.9922532207966257, "1000–3000 Hz_40ms_modulation_mae_db": 0.711672279409629, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6623760058006398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15698433113653126, "3000–6000 Hz_10ms_envelope_correlation": 0.9848902023558738, "3000–6000 Hz_10ms_modulation_mae_db": 0.41723798786935795, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9302655669619142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09166199940505951, "3000–6000 Hz_40ms_envelope_correlation": 0.9945907467497392, "3000–6000 Hz_40ms_modulation_mae_db": 0.6270195867165985, "mr_stft_db": 5.790993366336703, "band_mae_db": 0.1619539810896576, "rms_gain_error_db": -0.07822568282207654, "output_peak": 0.46376362159028284}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2888239787944902, "全帯域_10ms_rms_envelope_nrmse": 0.11591457562189016, "全帯域_10ms_envelope_correlation": 0.9880263800118185, "全帯域_10ms_modulation_mae_db": 0.6610336567913287, "全帯域_40ms_log_rms_mae_db": 0.666983754747872, "全帯域_40ms_rms_envelope_nrmse": 0.06531414792181578, "全帯域_40ms_envelope_correlation": 0.9958973196678128, "全帯域_40ms_modulation_mae_db": 0.763281031539735, "300–1000 Hz_10ms_log_rms_mae_db": 1.6322964442909857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1317829688516577, "300–1000 Hz_10ms_envelope_correlation": 0.9878132339594535, "300–1000 Hz_10ms_modulation_mae_db": 1.0058358762935176, "300–1000 Hz_40ms_log_rms_mae_db": 0.9275429240233559, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07365662396012411, "300–1000 Hz_40ms_envelope_correlation": 0.996047500055912, "300–1000 Hz_40ms_modulation_mae_db": 0.5194309153698321, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5886105010585008, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1589203447387581, "1000–3000 Hz_10ms_envelope_correlation": 0.9798092966648443, "1000–3000 Hz_10ms_modulation_mae_db": 0.7001205416589924, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8622417804606939, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08811571207493044, "1000–3000 Hz_40ms_envelope_correlation": 0.9935975345178294, "1000–3000 Hz_40ms_modulation_mae_db": 0.7668292618328086, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6336766844861075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15842613827161503, "3000–6000 Hz_10ms_envelope_correlation": 0.9846788242103451, "3000–6000 Hz_10ms_modulation_mae_db": 0.4636961424782445, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9383926148438235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09299144766110833, "3000–6000 Hz_40ms_envelope_correlation": 0.9949089274274645, "3000–6000 Hz_40ms_modulation_mae_db": 0.6498525078007792, "mr_stft_db": 5.815542163292458, "band_mae_db": 0.2762930576669039, "rms_gain_error_db": -0.07879832351685254, "output_peak": 0.45794295352101205}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3117190877597007, "全帯域_10ms_rms_envelope_nrmse": 0.10225328020287561, "全帯域_10ms_envelope_correlation": 0.9907306984945153, "全帯域_10ms_modulation_mae_db": 0.29419290764011263, "全帯域_40ms_log_rms_mae_db": 0.705846056912592, "全帯域_40ms_rms_envelope_nrmse": 0.04974383906545413, "全帯域_40ms_envelope_correlation": 0.997609027253735, "全帯域_40ms_modulation_mae_db": 0.4532574092832011, "300–1000 Hz_10ms_log_rms_mae_db": 1.6043661444010056, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10429570321622647, "300–1000 Hz_10ms_envelope_correlation": 0.9924200066579839, "300–1000 Hz_10ms_modulation_mae_db": 0.35263647237638307, "300–1000 Hz_40ms_log_rms_mae_db": 0.8886205259745811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05909546914781199, "300–1000 Hz_40ms_envelope_correlation": 0.9974433443526642, "300–1000 Hz_40ms_modulation_mae_db": 0.42305117693294697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6005300112124146, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17556002645055183, "1000–3000 Hz_10ms_envelope_correlation": 0.9756150508496855, "1000–3000 Hz_10ms_modulation_mae_db": 0.9627131474377687, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9615190887080279, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09761874140425081, "1000–3000 Hz_40ms_envelope_correlation": 0.992698942173813, "1000–3000 Hz_40ms_modulation_mae_db": 1.100217173342035, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6505554358973389, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14541641326880633, "3000–6000 Hz_10ms_envelope_correlation": 0.9870285317066623, "3000–6000 Hz_10ms_modulation_mae_db": 0.7558918349546541, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9056905984152189, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07642355545977329, "3000–6000 Hz_40ms_envelope_correlation": 0.9962463889845996, "3000–6000 Hz_40ms_modulation_mae_db": 0.23727264241944965, "mr_stft_db": 5.806656630247949, "band_mae_db": 0.22659014690060364, "rms_gain_error_db": -1.5963968945731426e-05, "output_peak": 0.506070399647035}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2879914755839796, "全帯域_10ms_rms_envelope_nrmse": 0.10311885939043419, "全帯域_10ms_envelope_correlation": 0.9905632499086217, "全帯域_10ms_modulation_mae_db": 0.18174792556142053, "全帯域_40ms_log_rms_mae_db": 0.6904752871804669, "全帯域_40ms_rms_envelope_nrmse": 0.051787341746003995, "全帯域_40ms_envelope_correlation": 0.9974030873108362, "全帯域_40ms_modulation_mae_db": 0.2843246503015333, "300–1000 Hz_10ms_log_rms_mae_db": 1.593999812500721, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11539153385922621, "300–1000 Hz_10ms_envelope_correlation": 0.9906656244229108, "300–1000 Hz_10ms_modulation_mae_db": 0.7070235532669534, "300–1000 Hz_40ms_log_rms_mae_db": 0.9406009858108768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06331750577861538, "300–1000 Hz_40ms_envelope_correlation": 0.9971211740074571, "300–1000 Hz_40ms_modulation_mae_db": 0.567866214834504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5442285615307219, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16168707337100374, "1000–3000 Hz_10ms_envelope_correlation": 0.9789891095758987, "1000–3000 Hz_10ms_modulation_mae_db": 0.33404260105686445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8816177094469497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08802603463082001, "1000–3000 Hz_40ms_envelope_correlation": 0.9932915006894479, "1000–3000 Hz_40ms_modulation_mae_db": 0.29954734224530905, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6514564717634326, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14177427107854468, "3000–6000 Hz_10ms_envelope_correlation": 0.9876485721683178, "3000–6000 Hz_10ms_modulation_mae_db": 0.13364216218934186, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9024880675817447, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07427799458312194, "3000–6000 Hz_40ms_envelope_correlation": 0.9964581824744448, "3000–6000 Hz_40ms_modulation_mae_db": 0.5477099114889313, "mr_stft_db": 5.790383265056107, "band_mae_db": 0.15563884928148983, "rms_gain_error_db": -0.015049169638133382, "output_peak": 0.40644518661555323}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2480152200016807, "全帯域_10ms_rms_envelope_nrmse": 0.10263463268430068, "全帯域_10ms_envelope_correlation": 0.9906973673100506, "全帯域_10ms_modulation_mae_db": 0.20749705301758956, "全帯域_40ms_log_rms_mae_db": 0.611505847670654, "全帯域_40ms_rms_envelope_nrmse": 0.04916061194030399, "全帯域_40ms_envelope_correlation": 0.9978257450459133, "全帯域_40ms_modulation_mae_db": 0.33933894011306837, "300–1000 Hz_10ms_log_rms_mae_db": 1.5982032760929337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10903645647724375, "300–1000 Hz_10ms_envelope_correlation": 0.9917635981369456, "300–1000 Hz_10ms_modulation_mae_db": 0.21819334501794288, "300–1000 Hz_40ms_log_rms_mae_db": 0.9409683712136383, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06245846735019788, "300–1000 Hz_40ms_envelope_correlation": 0.9973354414836992, "300–1000 Hz_40ms_modulation_mae_db": 0.3889414333106987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5745231237645183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1813052778040633, "1000–3000 Hz_10ms_envelope_correlation": 0.9735291158798165, "1000–3000 Hz_10ms_modulation_mae_db": 0.5959671002195908, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8532006083331243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0959606498789374, "1000–3000 Hz_40ms_envelope_correlation": 0.9921741354608371, "1000–3000 Hz_40ms_modulation_mae_db": 0.6602949827490151, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6892169014559606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15407769019310683, "3000–6000 Hz_10ms_envelope_correlation": 0.985437080025037, "3000–6000 Hz_10ms_modulation_mae_db": 0.5740990286850854, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9943341136571421, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08774658630533093, "3000–6000 Hz_40ms_envelope_correlation": 0.995382990814084, "3000–6000 Hz_40ms_modulation_mae_db": 0.1954475934983966, "mr_stft_db": 5.810469938580211, "band_mae_db": 0.21053371036072818, "rms_gain_error_db": -0.12679643188683212, "output_peak": 0.4609894972895697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3426791472161514, "全帯域_10ms_rms_envelope_nrmse": 0.12542631120088674, "全帯域_10ms_envelope_correlation": 0.9860924650057049, "全帯域_10ms_modulation_mae_db": 0.5708531986390413, "全帯域_40ms_log_rms_mae_db": 0.7241043498768308, "全帯域_40ms_rms_envelope_nrmse": 0.0672796863305728, "全帯域_40ms_envelope_correlation": 0.9956372846252207, "全帯域_40ms_modulation_mae_db": 0.3600500611314823, "300–1000 Hz_10ms_log_rms_mae_db": 1.5716980549100303, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11606934699751091, "300–1000 Hz_10ms_envelope_correlation": 0.9906221095570359, "300–1000 Hz_10ms_modulation_mae_db": 0.4716339874640485, "300–1000 Hz_40ms_log_rms_mae_db": 0.872715900964323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05924103532569916, "300–1000 Hz_40ms_envelope_correlation": 0.9974373664475313, "300–1000 Hz_40ms_modulation_mae_db": 0.7287953204300187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6010785254945132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18612339564084643, "1000–3000 Hz_10ms_envelope_correlation": 0.9720578090924976, "1000–3000 Hz_10ms_modulation_mae_db": 0.8229144077977676, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8746236473010072, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10370771701657241, "1000–3000 Hz_40ms_envelope_correlation": 0.9909112645675562, "1000–3000 Hz_40ms_modulation_mae_db": 1.1048470112940423, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6459976408979171, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17383942035514313, "3000–6000 Hz_10ms_envelope_correlation": 0.9814174200226179, "3000–6000 Hz_10ms_modulation_mae_db": 0.9655131480976141, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9151438241155915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0998982179699049, "3000–6000 Hz_40ms_envelope_correlation": 0.9935264121362023, "3000–6000 Hz_40ms_modulation_mae_db": 0.6040049257707835, "mr_stft_db": 5.804679105815752, "band_mae_db": 0.20738137677166268, "rms_gain_error_db": 0.03382773482360744, "output_peak": 0.44500361560420315}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0952290561336975, "spectral_convergence": 0.4576368793177469, "spectral_convergence_status": "ok", "amplitude_floor": 1.32832143223519e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.508478172458581, "spectral_convergence": 0.4275075222659662, "spectral_convergence_status": "ok", "amplitude_floor": 1.1133586355586426e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.807262698563677, "spectral_convergence": 0.4595839972533862, "spectral_convergence_status": "ok", "amplitude_floor": 6.710028542715064e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00015990772541983682, "output_power": 0.00016704872474933272, "delta_db": 0.18973718807410095, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005346827040387907, "output_power": 0.0005082346198147488, "delta_db": -0.22031890834464124, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025601253900142847, "output_power": 0.0002527477975895256, "delta_db": -0.055738568450860115, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015659557404440672, "output_power": 0.00014911716419183205, "delta_db": -0.21251847194069834, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003554436518358531, "output_power": 0.00032464282981045016, "delta_db": -0.39364946891216934, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004235024710497043, "output_power": 0.00040577596206926457, "delta_db": -0.18569632584825574, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20960982192845432, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.18480572029145284, "pitch": null, "output_rms_dbfs": -27.429055724024124, "output_peak": 0.4150892216167527, "reference_rms_dbfs": -27.244250003732674, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8036566423853175}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018816516597127566, "log_rms_mae_db": 1.2724906108957652, "rms_envelope_nrmse": 0.10433291920985033, "envelope_correlation": 0.9905409262606536, "rms_gain_error_db": -0.18446066036049216, "reference_mean_power": 0.0018869472812584292, "output_mean_power": 0.0018084799602080175}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.448172186664129e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039739520206635815, "output_power": 0.00039214757223163217, "reference_active": true, "delta_db": -0.05773088713560734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001684948117890412, "output_power": 0.00015172154006491044, "reference_active": true, "delta_db": -0.4553929042366273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9559595301379592e-05, "output_power": 2.1180221501725616e-05, "reference_active": true, "delta_db": 0.3457063286927587}], "active_band_count": 3, "active_band_mae_db": 0.2862767066883311}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016656172387320294, "log_rms_mae_db": 0.7052613173738217, "rms_envelope_nrmse": 0.054138765488526495, "envelope_correlation": 0.9975571893845752, "rms_gain_error_db": -0.18418972822025367, "reference_mean_power": 0.0018945898208991615, "output_mean_power": 0.0018159179719091276}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.514104383617561e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003829793712831679, "output_power": 0.00037935838021042065, "reference_active": true, "delta_db": -0.041256995743413426}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.831616093630877e-05, "output_power": 7.041623144444758e-05, "reference_active": true, "delta_db": -0.46178611508307943}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3812315351816923e-06, "output_power": 2.124788443028006e-06, "reference_active": true, "delta_db": -0.4948592986069556}], "active_band_count": 3, "active_band_mae_db": 0.3326341364778162}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015092951870698086, "log_rms_mae_db": 1.564857325232049, "rms_envelope_nrmse": 0.10700795573944553, "envelope_correlation": 0.9921526892156765, "rms_gain_error_db": -0.19699697771974845, "reference_mean_power": 0.000631981580682194, "output_mean_power": 0.0006039551974677487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.167139764996573e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018037000728302603, "output_power": 0.00017432166572326776, "reference_active": true, "delta_db": -0.14812955675178305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6944061726753776e-05, "output_power": 2.546902759899963e-05, "reference_active": true, "delta_db": -0.24450700728434446}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.521461134259169e-06, "output_power": 3.950779201650051e-06, "reference_active": true, "delta_db": 0.4995985929346214}], "active_band_count": 3, "active_band_mae_db": 0.29741171899024965}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013588907795973108, "log_rms_mae_db": 0.9136986528714797, "rms_envelope_nrmse": 0.06381828108431195, "envelope_correlation": 0.9973207744507664, "rms_gain_error_db": -0.19730184468007603, "reference_mean_power": 0.0006345375227010602, "output_mean_power": 0.0006063552249997356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.990151794989553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001715162379513309, "output_power": 0.00016620565327877648, "reference_active": true, "delta_db": -0.13659450586354674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2552077755571312e-05, "output_power": 1.1751481754853147e-05, "reference_active": true, "delta_db": -0.28622990243679747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.057253626713498e-07, "output_power": 2.8763535978658555e-07, "reference_active": true, "delta_db": -0.2648919487921422}], "active_band_count": 3, "active_band_mae_db": 0.22923878569749545}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.324369547066356e-05, "log_rms_mae_db": 1.573254528240287, "rms_envelope_nrmse": 0.17591516442784852, "envelope_correlation": 0.9750897436400111, "rms_gain_error_db": -0.21243549342992174, "reference_mean_power": 0.00026216016454011753, "output_mean_power": 0.00024964516059082205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5456728460340458e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.502085853898132e-05, "output_power": 6.962777988015915e-05, "reference_active": true, "delta_db": -0.32399482231516963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151594827944158e-05, "output_power": 2.047736448228361e-05, "reference_active": true, "delta_db": -0.21486431158735791}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.513878290774259e-06, "output_power": 5.0269304032796305e-06, "reference_active": true, "delta_db": -2.288245649154141}], "active_band_count": 3, "active_band_mae_db": 0.9423682610188896}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.873751028776232e-05, "log_rms_mae_db": 0.8771558803337726, "rms_envelope_nrmse": 0.09080061967450666, "envelope_correlation": 0.9930769096749009, "rms_gain_error_db": -0.21201709025056809, "reference_mean_power": 0.0002632887650860277, "output_mean_power": 0.0002507440397361309}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3635149275752304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.257955032157826e-05, "output_power": 6.680994820254672e-05, "reference_active": true, "delta_db": -0.3597313808343019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.405048683009261e-06, "output_power": 8.950487183949316e-06, "reference_active": true, "delta_db": -0.21514372873519777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.677553554297748e-07, "output_power": 4.236023383314616e-07, "reference_active": true, "delta_db": -2.5826450572451103}], "active_band_count": 3, "active_band_mae_db": 1.0525067222715367}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015682729984940356, "log_rms_mae_db": 1.6191435074552394, "rms_envelope_nrmse": 0.15286185998205068, "envelope_correlation": 0.9858534858706584, "rms_gain_error_db": -0.31297864535874936, "reference_mean_power": 0.0005797089728212459, "output_mean_power": 0.0005394015100982289}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2924452919576597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020605682430557354, "output_power": 0.00021135704598538162, "reference_active": true, "delta_db": 0.1102972812402864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012645863441832237, "output_power": 0.00011006838195468219, "reference_active": true, "delta_db": -0.6028590547593237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.26990748699409e-06, "output_power": 1.2997238162851895e-05, "reference_active": true, "delta_db": 1.4677567713264217}], "active_band_count": 3, "active_band_mae_db": 0.7269710357753439}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012704534221320322, "log_rms_mae_db": 0.8910231183587589, "rms_envelope_nrmse": 0.08580953660427196, "envelope_correlation": 0.9958740962900824, "rms_gain_error_db": -0.3128326997186784, "reference_mean_power": 0.0005824704905358117, "output_mean_power": 0.0005419892314316075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.040463159930301e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022945963060392102, "output_power": 0.00023251550324037464, "reference_active": true, "delta_db": 0.05745625179381258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.01414803619856e-05, "output_power": 5.282626520514503e-05, "reference_active": true, "delta_db": -0.5632420657225669}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3285241811094974e-06, "output_power": 1.4489595443617479e-06, "reference_active": true, "delta_db": 0.3768679631076458}], "active_band_count": 3, "active_band_mae_db": 0.3325220935413417}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8189425882968607, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8c33f84595fe0c992ed53a947ce4d6328103860d5b4b3b30f8a8a14bb1d2a1e0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9752615628456068, "standard_deviation": 0.0029363730255025154, "minimum": 0.9717459397403986, "maximum": 0.9798092966648443, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5803496126934888, "standard_deviation": 0.022360562986432846, "minimum": 1.5442285615307219, "maximum": 1.6048394214096473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6596077767148925, "standard_deviation": 0.2475111871946391, "minimum": 0.3206937761059872, "maximum": 0.9627131474377687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17542737993696528, "standard_deviation": 0.010314273183295792, "minimum": 0.1589203447387581, "maximum": 0.18690418416082638, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9924308297546152, "standard_deviation": 0.0009237596898942857, "minimum": 0.9909112645675562, "maximum": 0.9935975345178294, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8911758024348767, "standard_deviation": 0.036599946044036255, "minimum": 0.8532006083331243, "maximum": 0.9615190887080279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7413118378214476, "standard_deviation": 0.3418102205314191, "minimum": 0.23457992942720474, "maximum": 1.1048470112940423, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09550966533063592, "standard_deviation": 0.005918267333798724, "minimum": 0.08802603463082001, "maximum": 0.10370771701657241, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9850183635190205, "standard_deviation": 0.002009361837369415, "minimum": 0.9814174200226179, "maximum": 0.9876485721683178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6557850168526875, "standard_deviation": 0.02558990077044463, "minimum": 1.6191435074552394, "maximum": 1.6938574870648644, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.565484439852292, "standard_deviation": 0.25282493457285005, "minimum": 0.13364216218934186, "maximum": 0.9655131480976141, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15681173638814672, "standard_deviation": 0.011187885426886224, "minimum": 0.14177427107854468, "maximum": 0.17383942035514313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9950715696014099, "standard_deviation": 0.0011275467138802092, "minimum": 0.9935264121362023, "maximum": 0.9964581824744448, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9274326054328625, "standard_deviation": 0.03249477964418384, "minimum": 0.8910231183587589, "maximum": 0.9943341136571421, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.49481021928900304, "standard_deviation": 0.21089440930144504, "minimum": 0.1954475934983966, "maximum": 0.7646524930757437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08944456337914822, "standard_deviation": 0.010966341165027337, "minimum": 0.07427799458312194, "maximum": 0.1067471690446149, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9909351170215386, "standard_deviation": 0.001490160874016316, "minimum": 0.9878132339594535, "maximum": 0.9924200066579839, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.596316486933762, "standard_deviation": 0.021491986994346467, "minimum": 1.564857325232049, "maximum": 1.6322964442909857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5641883048001668, "standard_deviation": 0.335104143926753, "minimum": 0.21819334501794288, "maximum": 1.0985997497960294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11401521706818002, "standard_deviation": 0.008647156636555457, "minimum": 0.10429570321622647, "maximum": 0.1317829688516577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99710949737618, "standard_deviation": 0.0004899570243852581, "minimum": 0.996047500055912, "maximum": 0.9974433443526642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9140586564021546, "standard_deviation": 0.02530154861630733, "minimum": 0.872715900964323, "maximum": 0.9409683712136383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5116006360317065, "standard_deviation": 0.1590012430021599, "minimum": 0.22923878569749545, "maximum": 0.7287953204300187, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06429491201662188, "standard_deviation": 0.004769167700152438, "minimum": 0.05909546914781199, "maximum": 0.07365662396012411, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20931575336096037, "standard_deviation": 0.038191592567007784, "minimum": 0.15563884928148983, "maximum": 0.2762930576669039, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.802934269697312, "standard_deviation": 0.008755981838100927, "minimum": 5.790383265056107, "maximum": 5.815542163292458, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45153968856454685, "standard_deviation": 0.030910358526486686, "minimum": 0.40644518661555323, "maximum": 0.506070399647035, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.056691671328153934, "standard_deviation": 0.07390304341757994, "minimum": -0.18480572029145284, "maximum": 0.03382773482360744, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.989122587875348, "standard_deviation": 0.0017493924833753003, "minimum": 0.9860924650057049, "maximum": 0.9907306984945153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2946049575912948, "standard_deviation": 0.028884874752633315, "minimum": 1.2480152200016807, "maximum": 1.3426791472161514, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36779986079424926, "standard_deviation": 0.18440640522075444, "minimum": 0.18174792556142053, "maximum": 0.6610336567913287, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11054868241205755, "standard_deviation": 0.008644892306803584, "minimum": 0.10225328020287561, "maximum": 0.12542631120088674, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9966784566008519, "standard_deviation": 0.0009937871163528842, "minimum": 0.9956372846252207, "maximum": 0.9978257450459133, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6962815389486895, "standard_deviation": 0.04110967592451202, "minimum": 0.611505847670654, "maximum": 0.740206173512806, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46352632357669454, "standard_deviation": 0.17010256520867253, "minimum": 0.2843246503015333, "maximum": 0.763281031539735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05875915252828218, "standard_deviation": 0.008226999106009695, "minimum": 0.04916061194030399, "maximum": 0.0672796863305728, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0012", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.9884307384490967, "predicted_mos": 1.9884307384490967, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9600364ca18b72f6b834bdefc49d6c8fcdcd1ad22d0446f9a6711c60efd001e9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9008448999375105, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2545947307400207, "全帯域_10ms_rms_envelope_nrmse": 0.10304980810183802, "全帯域_10ms_envelope_correlation": 0.990558182628565, "全帯域_10ms_modulation_mae_db": 0.13383401903287556, "全帯域_40ms_log_rms_mae_db": 0.6830695230796254, "全帯域_40ms_rms_envelope_nrmse": 0.05482109317889505, "全帯域_40ms_envelope_correlation": 0.997125510015818, "全帯域_40ms_modulation_mae_db": 0.36663133598243475, "300–1000 Hz_10ms_log_rms_mae_db": 1.785738057154186, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11530047104794924, "300–1000 Hz_10ms_envelope_correlation": 0.9906789746790577, "300–1000 Hz_10ms_modulation_mae_db": 0.5020084269152557, "300–1000 Hz_40ms_log_rms_mae_db": 1.0024727017610324, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06426515699198986, "300–1000 Hz_40ms_envelope_correlation": 0.9969455670765086, "300–1000 Hz_40ms_modulation_mae_db": 0.49743395728756384, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7889130330232592, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1715763143747759, "1000–3000 Hz_10ms_envelope_correlation": 0.9765512030536304, "1000–3000 Hz_10ms_modulation_mae_db": 1.0432627245328387, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0319938457556062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09661383058317201, "1000–3000 Hz_40ms_envelope_correlation": 0.9918353119701087, "1000–3000 Hz_40ms_modulation_mae_db": 0.9139328635164574, "3000–6000 Hz_10ms_log_rms_mae_db": 1.983325660370545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1542250626239364, "3000–6000 Hz_10ms_envelope_correlation": 0.9856753142205583, "3000–6000 Hz_10ms_modulation_mae_db": 0.39341053726772107, "3000–6000 Hz_40ms_log_rms_mae_db": 1.096258653143301, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08809227087645692, "3000–6000 Hz_40ms_envelope_correlation": 0.9957245832544525, "3000–6000 Hz_40ms_modulation_mae_db": 0.3400780369264267, "mr_stft_db": 5.961025130511122, "band_mae_db": 0.12469981217062297, "rms_gain_error_db": -0.07227914949145582, "output_peak": 0.4960893481116174}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1756545000416205, "全帯域_10ms_rms_envelope_nrmse": 0.10050938822364255, "全帯域_10ms_envelope_correlation": 0.991015289460073, "全帯域_10ms_modulation_mae_db": 0.5605512948952514, "全帯域_40ms_log_rms_mae_db": 0.611010959831365, "全帯域_40ms_rms_envelope_nrmse": 0.051352633671045936, "全帯域_40ms_envelope_correlation": 0.9974826112228125, "全帯域_40ms_modulation_mae_db": 0.37867393221872886, "300–1000 Hz_10ms_log_rms_mae_db": 1.7387447426727711, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1144668873978758, "300–1000 Hz_10ms_envelope_correlation": 0.9908557468717222, "300–1000 Hz_10ms_modulation_mae_db": 0.9602808597274501, "300–1000 Hz_40ms_log_rms_mae_db": 1.0471251116210611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057843114485131276, "300–1000 Hz_40ms_envelope_correlation": 0.9977219634048214, "300–1000 Hz_40ms_modulation_mae_db": 0.5755518209625174, "1000–3000 Hz_10ms_log_rms_mae_db": 1.75180360226213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16571087243758867, "1000–3000 Hz_10ms_envelope_correlation": 0.9778676726483805, "1000–3000 Hz_10ms_modulation_mae_db": 0.2607389849175356, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9486530796886242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08027067884701716, "1000–3000 Hz_40ms_envelope_correlation": 0.9944664985236574, "1000–3000 Hz_40ms_modulation_mae_db": 0.40344715016819843, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8917317767745596, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14374101683683504, "3000–6000 Hz_10ms_envelope_correlation": 0.9872054282215919, "3000–6000 Hz_10ms_modulation_mae_db": 0.41743816395278577, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0685593116658103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08265965611988486, "3000–6000 Hz_40ms_envelope_correlation": 0.9956381000064181, "3000–6000 Hz_40ms_modulation_mae_db": 0.5155046992388138, "mr_stft_db": 5.933211032380245, "band_mae_db": 0.13731188559258925, "rms_gain_error_db": -0.07042595155647399, "output_peak": 0.5071037725587864}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2113924555140245, "全帯域_10ms_rms_envelope_nrmse": 0.11058999713330801, "全帯域_10ms_envelope_correlation": 0.9891701000053132, "全帯域_10ms_modulation_mae_db": 0.40657478996636925, "全帯域_40ms_log_rms_mae_db": 0.6600010490239985, "全帯域_40ms_rms_envelope_nrmse": 0.056597279070574255, "全帯域_40ms_envelope_correlation": 0.9970604950974582, "全帯域_40ms_modulation_mae_db": 0.5938724569315528, "300–1000 Hz_10ms_log_rms_mae_db": 1.8365091748994247, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1306201253299264, "300–1000 Hz_10ms_envelope_correlation": 0.988032961423641, "300–1000 Hz_10ms_modulation_mae_db": 0.41405059227321583, "300–1000 Hz_40ms_log_rms_mae_db": 1.0504721501111365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08051275397938451, "300–1000 Hz_40ms_envelope_correlation": 0.9953329988257759, "300–1000 Hz_40ms_modulation_mae_db": 0.15591079464326296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8348831837063595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17760185350671157, "1000–3000 Hz_10ms_envelope_correlation": 0.9745152383437665, "1000–3000 Hz_10ms_modulation_mae_db": 0.26834622735800945, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0409984066936866, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08923147059602843, "1000–3000 Hz_40ms_envelope_correlation": 0.9930499750440153, "1000–3000 Hz_40ms_modulation_mae_db": 0.4165806218896944, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9122894569648574, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14042581000593046, "3000–6000 Hz_10ms_envelope_correlation": 0.9878934656633961, "3000–6000 Hz_10ms_modulation_mae_db": 0.7239119328406245, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0818341483536902, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0697574412309313, "3000–6000 Hz_40ms_envelope_correlation": 0.9970971183759293, "3000–6000 Hz_40ms_modulation_mae_db": 0.8752615725840984, "mr_stft_db": 5.964638719450747, "band_mae_db": 0.1609534663055158, "rms_gain_error_db": -0.12961864595453682, "output_peak": 0.4313575155279817}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2575784605777636, "全帯域_10ms_rms_envelope_nrmse": 0.10866775348233126, "全帯域_10ms_envelope_correlation": 0.9895896009437183, "全帯域_10ms_modulation_mae_db": 0.2678784120158029, "全帯域_40ms_log_rms_mae_db": 0.7047944678437519, "全帯域_40ms_rms_envelope_nrmse": 0.05969838193133377, "全帯域_40ms_envelope_correlation": 0.9966005188165996, "全帯域_40ms_modulation_mae_db": 0.11938054079280386, "300–1000 Hz_10ms_log_rms_mae_db": 1.819849920138898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14142933266435884, "300–1000 Hz_10ms_envelope_correlation": 0.9859274715818255, "300–1000 Hz_10ms_modulation_mae_db": 0.7391598318297161, "300–1000 Hz_40ms_log_rms_mae_db": 1.1040221732079816, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07858661732544346, "300–1000 Hz_40ms_envelope_correlation": 0.9955926840402527, "300–1000 Hz_40ms_modulation_mae_db": 0.35660983978781435, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7767904842250606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17452073446056457, "1000–3000 Hz_10ms_envelope_correlation": 0.9754102271803246, "1000–3000 Hz_10ms_modulation_mae_db": 0.5509363992823912, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0250320107203243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08480831913500547, "1000–3000 Hz_40ms_envelope_correlation": 0.9937186402343269, "1000–3000 Hz_40ms_modulation_mae_db": 0.5500990073178531, "3000–6000 Hz_10ms_log_rms_mae_db": 1.890366415172659, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15389402308559447, "3000–6000 Hz_10ms_envelope_correlation": 0.9860165958082395, "3000–6000 Hz_10ms_modulation_mae_db": 0.6903317016919887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1204995712370729, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09421127713081656, "3000–6000 Hz_40ms_envelope_correlation": 0.9947137664147052, "3000–6000 Hz_40ms_modulation_mae_db": 0.4493230113882511, "mr_stft_db": 5.961969390341792, "band_mae_db": 0.19222569605865827, "rms_gain_error_db": 0.01722514387435386, "output_peak": 0.4432858972118653}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2800416355337068, "全帯域_10ms_rms_envelope_nrmse": 0.10428741099700892, "全帯域_10ms_envelope_correlation": 0.9903335791941772, "全帯域_10ms_modulation_mae_db": 0.29154955979105857, "全帯域_40ms_log_rms_mae_db": 0.6820329283711606, "全帯域_40ms_rms_envelope_nrmse": 0.05024912034802801, "全帯域_40ms_envelope_correlation": 0.9976256267553004, "全帯域_40ms_modulation_mae_db": 0.47511083586641467, "300–1000 Hz_10ms_log_rms_mae_db": 1.8390294727304513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12013127149904111, "300–1000 Hz_10ms_envelope_correlation": 0.9898501793717535, "300–1000 Hz_10ms_modulation_mae_db": 0.4970698429306828, "300–1000 Hz_40ms_log_rms_mae_db": 1.112013619398982, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0690793126771551, "300–1000 Hz_40ms_envelope_correlation": 0.9965320851354326, "300–1000 Hz_40ms_modulation_mae_db": 0.8316347981480812, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8002471822661168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18955764543176934, "1000–3000 Hz_10ms_envelope_correlation": 0.9712521712455939, "1000–3000 Hz_10ms_modulation_mae_db": 0.5551901330448953, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0292627650781354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11289537675071792, "1000–3000 Hz_40ms_envelope_correlation": 0.9896105728609855, "1000–3000 Hz_40ms_modulation_mae_db": 0.4995841587688621, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9351282561913652, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1462375566967907, "3000–6000 Hz_10ms_envelope_correlation": 0.9867338738185772, "3000–6000 Hz_10ms_modulation_mae_db": 0.2353542874439577, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1025907426256356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07020025576522022, "3000–6000 Hz_40ms_envelope_correlation": 0.9968168450426987, "3000–6000 Hz_40ms_modulation_mae_db": 0.6253432699612312, "mr_stft_db": 5.948451615254872, "band_mae_db": 0.1478475452143377, "rms_gain_error_db": -0.0859403790172296, "output_peak": 0.5034733266765564}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2247409963009896, "全帯域_10ms_rms_envelope_nrmse": 0.11317415924765435, "全帯域_10ms_envelope_correlation": 0.9885875832299104, "全帯域_10ms_modulation_mae_db": 0.26459313269867607, "全帯域_40ms_log_rms_mae_db": 0.6226832362004938, "全帯域_40ms_rms_envelope_nrmse": 0.05734705390757133, "全帯域_40ms_envelope_correlation": 0.9968546135093651, "全帯域_40ms_modulation_mae_db": 0.23004557157595032, "300–1000 Hz_10ms_log_rms_mae_db": 1.7610732089839785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10946656037372106, "300–1000 Hz_10ms_envelope_correlation": 0.9915713713425375, "300–1000 Hz_10ms_modulation_mae_db": 0.6690227516240834, "300–1000 Hz_40ms_log_rms_mae_db": 1.0582224794592447, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05734459076874554, "300–1000 Hz_40ms_envelope_correlation": 0.9975939688501293, "300–1000 Hz_40ms_modulation_mae_db": 0.49273832874329315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7550858426588098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17107094387412491, "1000–3000 Hz_10ms_envelope_correlation": 0.9766063466791283, "1000–3000 Hz_10ms_modulation_mae_db": 0.6495840253020465, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9861652893545354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08402512502665582, "1000–3000 Hz_40ms_envelope_correlation": 0.9944724511485458, "1000–3000 Hz_40ms_modulation_mae_db": 0.6989721297478121, "3000–6000 Hz_10ms_log_rms_mae_db": 1.944155781107858, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17005780653425415, "3000–6000 Hz_10ms_envelope_correlation": 0.9820809511524435, "3000–6000 Hz_10ms_modulation_mae_db": 0.2462608768516783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0682401647694253, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09670268294257663, "3000–6000 Hz_40ms_envelope_correlation": 0.9938980777482048, "3000–6000 Hz_40ms_modulation_mae_db": 0.4317708788602341, "mr_stft_db": 5.971306686037505, "band_mae_db": 0.15531161726820725, "rms_gain_error_db": -0.07249089198286701, "output_peak": 0.44744100079677906}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.221758645500903, "全帯域_10ms_rms_envelope_nrmse": 0.10137245866690582, "全帯域_10ms_envelope_correlation": 0.9908810750756324, "全帯域_10ms_modulation_mae_db": 0.4057431294088705, "全帯域_40ms_log_rms_mae_db": 0.6786386427414245, "全帯域_40ms_rms_envelope_nrmse": 0.050073549418460185, "全帯域_40ms_envelope_correlation": 0.9976560913901118, "全帯域_40ms_modulation_mae_db": 0.4975513776389624, "300–1000 Hz_10ms_log_rms_mae_db": 1.7648182096841163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11657617230854468, "300–1000 Hz_10ms_envelope_correlation": 0.9904515999338739, "300–1000 Hz_10ms_modulation_mae_db": 0.607703494396599, "300–1000 Hz_40ms_log_rms_mae_db": 1.0928600450620258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06577536230099244, "300–1000 Hz_40ms_envelope_correlation": 0.9968749494055162, "300–1000 Hz_40ms_modulation_mae_db": 0.18096229644534365, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7548196192408259, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16821171452194175, "1000–3000 Hz_10ms_envelope_correlation": 0.9773078415989619, "1000–3000 Hz_10ms_modulation_mae_db": 1.0991550276332416, "1000–3000 Hz_40ms_log_rms_mae_db": 0.997976146906726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09267654481631844, "1000–3000 Hz_40ms_envelope_correlation": 0.9924628506683041, "1000–3000 Hz_40ms_modulation_mae_db": 0.9410832786446015, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8957459629676485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.133963492758983, "3000–6000 Hz_10ms_envelope_correlation": 0.9890530827715115, "3000–6000 Hz_10ms_modulation_mae_db": 0.49460203969919, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0495139216283311, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07267197470054455, "3000–6000 Hz_40ms_envelope_correlation": 0.9968848943135573, "3000–6000 Hz_40ms_modulation_mae_db": 0.6368882152479197, "mr_stft_db": 5.950024070462152, "band_mae_db": 0.1572252871131378, "rms_gain_error_db": -0.0947009775307777, "output_peak": 0.5171791584572033}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.222184900225048, "全帯域_10ms_rms_envelope_nrmse": 0.10363861670462267, "全帯域_10ms_envelope_correlation": 0.9905134842246917, "全帯域_10ms_modulation_mae_db": 0.1959862773130535, "全帯域_40ms_log_rms_mae_db": 0.5682612064020807, "全帯域_40ms_rms_envelope_nrmse": 0.05384600784700106, "全帯域_40ms_envelope_correlation": 0.9972146397402771, "全帯域_40ms_modulation_mae_db": 0.33406007592771997, "300–1000 Hz_10ms_log_rms_mae_db": 1.7278507847425038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11421601083502307, "300–1000 Hz_10ms_envelope_correlation": 0.9908917488524848, "300–1000 Hz_10ms_modulation_mae_db": 0.17272669820974676, "300–1000 Hz_40ms_log_rms_mae_db": 1.011096866403489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0676489454620733, "300–1000 Hz_40ms_envelope_correlation": 0.9967664984940479, "300–1000 Hz_40ms_modulation_mae_db": 0.16032813749788621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7605475966182058, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1795974783630259, "1000–3000 Hz_10ms_envelope_correlation": 0.9748916518464054, "1000–3000 Hz_10ms_modulation_mae_db": 0.2740708745786507, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9878956905609516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09701762569881289, "1000–3000 Hz_40ms_envelope_correlation": 0.9921309909830703, "1000–3000 Hz_40ms_modulation_mae_db": 0.37981059646696047, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9164100740272147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14510757244733144, "3000–6000 Hz_10ms_envelope_correlation": 0.9871953214221457, "3000–6000 Hz_10ms_modulation_mae_db": 0.36162639429613647, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0745519486388353, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08146026147845653, "3000–6000 Hz_40ms_envelope_correlation": 0.9957983427557985, "3000–6000 Hz_40ms_modulation_mae_db": 0.6405144273653002, "mr_stft_db": 5.9493570252930015, "band_mae_db": 0.10783160882905389, "rms_gain_error_db": 0.025436801909934848, "output_peak": 0.46934054467574693}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.350497272247688, "spectral_convergence": 0.46280386364194087, "spectral_convergence_status": "ok", "amplitude_floor": 1.32832143223519e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6779523827130856, "spectral_convergence": 0.43356510213863914, "spectral_convergence_status": "ok", "amplitude_floor": 1.1133586355586426e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.854625736572594, "spectral_convergence": 0.4595515686527461, "spectral_convergence_status": "ok", "amplitude_floor": 6.710028542715064e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00015990772541983682, "output_power": 0.00016012583649373824, "delta_db": 0.005919657296989483, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005346827040387907, "output_power": 0.0005259677220362943, "delta_db": -0.07137042856194512, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025601253900142847, "output_power": 0.00026594999522689723, "delta_db": 0.16538750121521703, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015659557404440672, "output_power": 0.00015287127176252083, "delta_db": -0.10453604604284393, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003554436518358531, "output_power": 0.0003595195075774369, "delta_db": 0.049516980292256096, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004235024710497043, "output_power": 0.0003905792100619946, "delta_db": -0.35146825961448613, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12469981217062297, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07227914949145582, "pitch": null, "output_rms_dbfs": -27.31652915322413, "output_peak": 0.4960893481116174, "reference_rms_dbfs": -27.244250003732674, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.961025130511122}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018816516597127566, "log_rms_mae_db": 1.2545947307400207, "rms_envelope_nrmse": 0.10304980810183802, "envelope_correlation": 0.990558182628565, "rms_gain_error_db": -0.07199027324572298, "reference_mean_power": 0.0018869472812584292, "output_mean_power": 0.0018559263568113887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.448172186664129e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039739520206635815, "output_power": 0.0003928703738063458, "reference_active": true, "delta_db": -0.049733392920318566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001684948117890412, "output_power": 0.00016285104016090256, "reference_active": true, "delta_db": -0.1479599599495899}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9559595301379592e-05, "output_power": 1.866289421207357e-05, "reference_active": true, "delta_db": -0.20380870422871822}], "active_band_count": 3, "active_band_mae_db": 0.13383401903287556}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016656172387320294, "log_rms_mae_db": 0.6830695230796254, "rms_envelope_nrmse": 0.05482109317889505, "envelope_correlation": 0.997125510015818, "rms_gain_error_db": -0.07163317657572706, "reference_mean_power": 0.0018945898208991615, "output_mean_power": 0.001863596482157169}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.514104383617561e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003829793712831679, "output_power": 0.0003807108076597475, "reference_active": true, "delta_db": -0.025801760676419774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.831616093630877e-05, "output_power": 7.552443801957087e-05, "reference_active": true, "delta_db": -0.1576388781660183}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3812315351816923e-06, "output_power": 1.9282186915750182e-06, "reference_active": true, "delta_db": -0.9164533691048662}], "active_band_count": 3, "active_band_mae_db": 0.36663133598243475}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015092951870698086, "log_rms_mae_db": 1.785738057154186, "rms_envelope_nrmse": 0.11530047104794924, "envelope_correlation": 0.9906789746790577, "rms_gain_error_db": -0.010080655992640807, "reference_mean_power": 0.000631981580682194, "output_mean_power": 0.0006305163534939522}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.167139764996573e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018037000728302603, "output_power": 0.00017354982878835048, "reference_active": true, "delta_db": -0.16740133251402275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6944061726753776e-05, "output_power": 2.7589027962443355e-05, "reference_active": true, "delta_db": 0.10273334600540293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.521461134259169e-06, "output_power": 4.680712426214552e-06, "reference_active": true, "delta_db": 1.2358906022263414}], "active_band_count": 3, "active_band_mae_db": 0.5020084269152557}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013588907795973108, "log_rms_mae_db": 1.0024727017610324, "rms_envelope_nrmse": 0.06426515699198986, "envelope_correlation": 0.9969455670765086, "rms_gain_error_db": -0.009607191887888236, "reference_mean_power": 0.0006345375227010602, "output_mean_power": 0.0006331353897633705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.990151794989553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001715162379513309, "output_power": 0.00016604665685833936, "reference_active": true, "delta_db": -0.14075106226375123}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2552077755571312e-05, "output_power": 1.2356949688684101e-05, "reference_active": true, "delta_db": -0.06804342434223996}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.057253626713498e-07, "output_power": 4.1084897458874204e-07, "reference_active": true, "delta_db": 1.2835073852567003}], "active_band_count": 3, "active_band_mae_db": 0.49743395728756384}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.324369547066356e-05, "log_rms_mae_db": 1.7889130330232592, "rms_envelope_nrmse": 0.1715763143747759, "envelope_correlation": 0.9765512030536304, "rms_gain_error_db": 0.023471868149380366, "reference_mean_power": 0.00026216016454011753, "output_mean_power": 0.0002635808703894079}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5456728460340458e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.502085853898132e-05, "output_power": 7.466316745565131e-05, "reference_active": true, "delta_db": -0.02075619737976563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151594827944158e-05, "output_power": 2.244832716540655e-05, "reference_active": true, "delta_db": 0.18423491562186436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.513878290774259e-06, "output_power": 4.34157923492787e-06, "reference_active": true, "delta_db": -2.9247970605968865}], "active_band_count": 3, "active_band_mae_db": 1.0432627245328387}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.873751028776232e-05, "log_rms_mae_db": 1.0319938457556062, "rms_envelope_nrmse": 0.09661383058317201, "envelope_correlation": 0.9918353119701087, "rms_gain_error_db": 0.024211175368073545, "reference_mean_power": 0.0002632887650860277, "output_mean_power": 0.00026476065392049437}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3635149275752304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.257955032157826e-05, "output_power": 7.241666089699348e-05, "reference_active": true, "delta_db": -0.009757773313212882}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.405048683009261e-06, "output_power": 9.623608936850026e-06, "reference_active": true, "delta_db": 0.0997691886719238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.677553554297748e-07, "output_power": 4.1878941057587074e-07, "reference_active": true, "delta_db": -2.6322716285642356}], "active_band_count": 3, "active_band_mae_db": 0.9139328635164574}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015682729984940356, "log_rms_mae_db": 1.983325660370545, "rms_envelope_nrmse": 0.1542250626239364, "envelope_correlation": 0.9856753142205583, "rms_gain_error_db": -0.33029171847516636, "reference_mean_power": 0.0005797089728212459, "output_mean_power": 0.0005372554758737291}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2924452919576597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020605682430557354, "output_power": 0.00020514960895760893, "reference_active": true, "delta_db": -0.01916308826443345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012645863441832237, "output_power": 0.0001146128068035741, "reference_active": true, "delta_db": -0.4271533942232151}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.26990748699409e-06, "output_power": 1.0976576610162481e-05, "reference_active": true, "delta_db": 0.7339151293155148}], "active_band_count": 3, "active_band_mae_db": 0.39341053726772107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012704534221320322, "log_rms_mae_db": 1.096258653143301, "rms_envelope_nrmse": 0.08809227087645692, "envelope_correlation": 0.9957245832544525, "rms_gain_error_db": -0.33010180825591917, "reference_mean_power": 0.0005824704905358117, "output_mean_power": 0.0005398383667103892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.040463159930301e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022945963060392102, "output_power": 0.00022698006611127428, "reference_active": true, "delta_db": -0.04718571928350269}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.01414803619856e-05, "output_power": 5.499865576038603e-05, "reference_active": true, "delta_db": -0.38822039011728066}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3285241811094974e-06, "output_power": 1.1611455536575978e-06, "reference_active": true, "delta_db": -0.5848280013784967}], "active_band_count": 3, "active_band_mae_db": 0.3400780369264267}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8090642696946961, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1563600812a31b8c9dfee3e688be4c0582b511657cac263aaac80424ef84dcff", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.975550294074524, "standard_deviation": 0.002089938863809508, "minimum": 0.9712521712455939, "maximum": 0.9778676726483805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.777886318000096, "standard_deviation": 0.02904261360186755, "minimum": 1.75180360226213, "maximum": 1.8348831837063595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5876605495812011, "standard_deviation": 0.3343683343252625, "minimum": 0.2607389849175356, "maximum": 1.0991550276332416, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17473094462131283, "standard_deviation": 0.007549927856532221, "minimum": 0.16571087243758867, "maximum": 0.18955764543176934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927184114291268, "standard_deviation": 0.0016076493587087136, "minimum": 0.9896105728609855, "maximum": 0.9944724511485458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0059971543448238, "standard_deviation": 0.031346394982827, "minimum": 0.9486530796886242, "maximum": 1.0409984066936866, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.600438725815055, "standard_deviation": 0.22615358343489733, "minimum": 0.37981059646696047, "maximum": 0.9410832786446015, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09219237143171602, "standard_deviation": 0.010308131761447042, "minimum": 0.08027067884701716, "maximum": 0.11289537675071792, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.986481754134808, "standard_deviation": 0.0020679490075201073, "minimum": 0.9820809511524435, "maximum": 0.9890530827715115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9211441729470886, "standard_deviation": 0.03195951066275941, "minimum": 1.890366415172659, "maximum": 1.983325660370545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.44536699175551026, "standard_deviation": 0.1829362279503698, "minimum": 0.2353542874439577, "maximum": 0.7239119328406245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14845654262370694, "standard_deviation": 0.0109699164164801, "minimum": 0.133963492758983, "maximum": 0.17005780653425415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9958214659889706, "standard_deviation": 0.0011160461445752356, "minimum": 0.9938980777482048, "maximum": 0.9970971183759293, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0827560577577628, "standard_deviation": 0.022632982577256357, "minimum": 1.0495139216283311, "maximum": 1.1204995712370729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5643355139465345, "standard_deviation": 0.16687942791743707, "minimum": 0.3400780369264267, "maximum": 0.8752615725840984, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08196947753061094, "standard_deviation": 0.010550864659315175, "minimum": 0.0697574412309313, "maximum": 0.09670268294257663, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.989782506757112, "standard_deviation": 0.0018807454429541912, "minimum": 0.9859274715818255, "maximum": 0.9915713713425375, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7842016963757912, "standard_deviation": 0.0433716826621498, "minimum": 1.7278507847425038, "maximum": 1.8390294727304513, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5702528122383437, "standard_deviation": 0.23429185145189863, "minimum": 0.17272669820974676, "maximum": 0.9602808597274501, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12027585393205503, "standard_deviation": 0.010557618874238091, "minimum": 0.10946656037372106, "maximum": 0.14142933266435884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966700894040605, "standard_deviation": 0.0008497731744286823, "minimum": 0.9953329988257759, "maximum": 0.9977219634048214, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.059785643378119, "standard_deviation": 0.04085408194346644, "minimum": 1.0024727017610324, "maximum": 1.112013619398982, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4063962466894704, "standard_deviation": 0.23957578047199424, "minimum": 0.15591079464326296, "maximum": 0.8316347981480812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06763198174886444, "standard_deviation": 0.008489054033251786, "minimum": 0.05734459076874554, "maximum": 0.08051275397938451, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14792586481901537, "standard_deviation": 0.025447098112950813, "minimum": 0.10783160882905389, "maximum": 0.19222569605865827, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.9549979587164295, "standard_deviation": 0.01205896808702956, "minimum": 5.933211032380245, "maximum": 5.971306686037505, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47690882050206707, "standard_deviation": 0.0332481029082675, "minimum": 0.4313575155279817, "maximum": 0.5171791584572033, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06034925621863153, "standard_deviation": 0.05398612448460834, "minimum": -0.12961864595453682, "maximum": 0.025436801909934848, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900811118452602, "standard_deviation": 0.0008695700430786936, "minimum": 0.9885875832299104, "maximum": 0.991015289460073, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2309932905542595, "standard_deviation": 0.032333245720603546, "minimum": 1.1756545000416205, "maximum": 1.2800416355337068, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31583882689024473, "standard_deviation": 0.1358605721168816, "minimum": 0.13383401903287556, "maximum": 0.5605512948952514, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10566119906966395, "standard_deviation": 0.004590501614962326, "minimum": 0.10050938822364255, "maximum": 0.11317415924765435, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972025133184678, "standard_deviation": 0.0003728060152027939, "minimum": 0.9966005188165996, "maximum": 0.9976560913901118, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6513115016867376, "standard_deviation": 0.046254187958459175, "minimum": 0.5682612064020807, "maximum": 0.7047944678437519, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37441576586682096, "standard_deviation": 0.1516839502432176, "minimum": 0.11938054079280386, "maximum": 0.5938724569315528, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054248139921613696, "standard_deviation": 0.0035278517563716136, "minimum": 0.050073549418460185, "maximum": 0.05969838193133377, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4077-13754-0012", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.0850601196289062, "predicted_mos": 2.0850601196289062, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8dc8f446901c015d26940230790ca2085a4d1f0f766dc0f9e69b9eb439e65a7d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.038056999910623, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2122206317108506, "全帯域_10ms_rms_envelope_nrmse": 0.11495059184279953, "全帯域_10ms_envelope_correlation": 0.9883506237624421, "全帯域_10ms_modulation_mae_db": 0.5255283434002981, "全帯域_40ms_log_rms_mae_db": 0.6269799888285947, "全帯域_40ms_rms_envelope_nrmse": 0.058953206320184, "全帯域_40ms_envelope_correlation": 0.9966686537190473, "全帯域_40ms_modulation_mae_db": 0.2227228191758667, "300–1000 Hz_10ms_log_rms_mae_db": 1.7121597604072651, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11808257422919145, "300–1000 Hz_10ms_envelope_correlation": 0.9902337296535445, "300–1000 Hz_10ms_modulation_mae_db": 0.6562154137289057, "300–1000 Hz_40ms_log_rms_mae_db": 0.9505734803638501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06534627264362956, "300–1000 Hz_40ms_envelope_correlation": 0.9969684493153762, "300–1000 Hz_40ms_modulation_mae_db": 0.3161245827904237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7039742571615202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18244594732932798, "1000–3000 Hz_10ms_envelope_correlation": 0.9732853910089589, "1000–3000 Hz_10ms_modulation_mae_db": 0.9896426718341892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9203787426181712, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10185225974833946, "1000–3000 Hz_40ms_envelope_correlation": 0.9908585650040967, "1000–3000 Hz_40ms_modulation_mae_db": 0.88594344193756, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8670483921562726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1721270437277129, "3000–6000 Hz_10ms_envelope_correlation": 0.9823276204016168, "3000–6000 Hz_10ms_modulation_mae_db": 1.0491735729818379, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0265814644195204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09512457691202857, "3000–6000 Hz_40ms_envelope_correlation": 0.9944910887709969, "3000–6000 Hz_40ms_modulation_mae_db": 0.5240866410193805, "mr_stft_db": 5.9123950752791705, "band_mae_db": 0.18947975002861686, "rms_gain_error_db": 0.03209186299159666, "output_peak": 0.5297423230437182}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2427340983002009, "全帯域_10ms_rms_envelope_nrmse": 0.09806108404350607, "全帯域_10ms_envelope_correlation": 0.9914521198761257, "全帯域_10ms_modulation_mae_db": 0.3924871470342735, "全帯域_40ms_log_rms_mae_db": 0.6143621486564701, "全帯域_40ms_rms_envelope_nrmse": 0.052881797898530335, "全帯域_40ms_envelope_correlation": 0.9972830882805074, "全帯域_40ms_modulation_mae_db": 0.44575954167646553, "300–1000 Hz_10ms_log_rms_mae_db": 1.6691266297067797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1174115453011591, "300–1000 Hz_10ms_envelope_correlation": 0.9903509649028958, "300–1000 Hz_10ms_modulation_mae_db": 0.4588952069988827, "300–1000 Hz_40ms_log_rms_mae_db": 0.9687008721210131, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06993740420323707, "300–1000 Hz_40ms_envelope_correlation": 0.9965147209354416, "300–1000 Hz_40ms_modulation_mae_db": 0.3027508218554609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7437362001850316, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18341689307181697, "1000–3000 Hz_10ms_envelope_correlation": 0.9728593194528464, "1000–3000 Hz_10ms_modulation_mae_db": 0.7116858270659546, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9524600949322469, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11068238469008468, "1000–3000 Hz_40ms_envelope_correlation": 0.9892383434248958, "1000–3000 Hz_40ms_modulation_mae_db": 0.41543841113448804, "3000–6000 Hz_10ms_log_rms_mae_db": 1.834078607800364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.133182233888025, "3000–6000 Hz_10ms_envelope_correlation": 0.9890431400545606, "3000–6000 Hz_10ms_modulation_mae_db": 0.3937100691077388, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0066592527539735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07615708825932598, "3000–6000 Hz_40ms_envelope_correlation": 0.9962182544482051, "3000–6000 Hz_40ms_modulation_mae_db": 0.5947688957571898, "mr_stft_db": 5.909107948327371, "band_mae_db": 0.22906491066140558, "rms_gain_error_db": -0.01412855987130926, "output_peak": 0.4324361967810148}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2983109772616168, "全帯域_10ms_rms_envelope_nrmse": 0.10455309388472174, "全帯域_10ms_envelope_correlation": 0.9903505396992143, "全帯域_10ms_modulation_mae_db": 0.262221748276996, "全帯域_40ms_log_rms_mae_db": 0.6566203664271303, "全帯域_40ms_rms_envelope_nrmse": 0.05298184202054456, "全帯域_40ms_envelope_correlation": 0.9974733593972687, "全帯域_40ms_modulation_mae_db": 0.35322884744095645, "300–1000 Hz_10ms_log_rms_mae_db": 1.7287307097505824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12233855500439744, "300–1000 Hz_10ms_envelope_correlation": 0.9895805279629981, "300–1000 Hz_10ms_modulation_mae_db": 0.1458998802558029, "300–1000 Hz_40ms_log_rms_mae_db": 0.9998755611792425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0619204148331962, "300–1000 Hz_40ms_envelope_correlation": 0.9974390808995003, "300–1000 Hz_40ms_modulation_mae_db": 0.3300392454068235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7380142451633287, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1703452470561614, "1000–3000 Hz_10ms_envelope_correlation": 0.9766821051818444, "1000–3000 Hz_10ms_modulation_mae_db": 1.3194078403935425, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9374674378000885, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08479296977929923, "1000–3000 Hz_40ms_envelope_correlation": 0.9940699096266065, "1000–3000 Hz_40ms_modulation_mae_db": 1.5238495782722576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7826921434794027, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14552079308408697, "3000–6000 Hz_10ms_envelope_correlation": 0.98687121167164, "3000–6000 Hz_10ms_modulation_mae_db": 0.2670286468574555, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9679779972655739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08179250989423992, "3000–6000 Hz_40ms_envelope_correlation": 0.9957001048947738, "3000–6000 Hz_40ms_modulation_mae_db": 0.1835890397617972, "mr_stft_db": 5.918510323127941, "band_mae_db": 0.15817436723254727, "rms_gain_error_db": -0.12875043679142326, "output_peak": 0.4598089314308381}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2395349714829194, "全帯域_10ms_rms_envelope_nrmse": 0.1162715258113068, "全帯域_10ms_envelope_correlation": 0.9880889684108698, "全帯域_10ms_modulation_mae_db": 0.5403784238297734, "全帯域_40ms_log_rms_mae_db": 0.6733633142563545, "全帯域_40ms_rms_envelope_nrmse": 0.05534018347164545, "全帯域_40ms_envelope_correlation": 0.9970816755303519, "全帯域_40ms_modulation_mae_db": 0.6149789945825197, "300–1000 Hz_10ms_log_rms_mae_db": 1.6989236540203807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12190774323525075, "300–1000 Hz_10ms_envelope_correlation": 0.9895365757270472, "300–1000 Hz_10ms_modulation_mae_db": 0.4571877967967041, "300–1000 Hz_40ms_log_rms_mae_db": 0.9733547859036672, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07515379437791647, "300–1000 Hz_40ms_envelope_correlation": 0.9958414026960202, "300–1000 Hz_40ms_modulation_mae_db": 0.7715787908389643, "1000–3000 Hz_10ms_log_rms_mae_db": 1.673629883108352, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16885979264540676, "1000–3000 Hz_10ms_envelope_correlation": 0.9770105237916382, "1000–3000 Hz_10ms_modulation_mae_db": 1.0233562082815126, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9512686026330706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08915686712673895, "1000–3000 Hz_40ms_envelope_correlation": 0.9930037251518938, "1000–3000 Hz_40ms_modulation_mae_db": 1.2189896265377937, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8738544490165447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17279756373718733, "3000–6000 Hz_10ms_envelope_correlation": 0.9816830260768666, "3000–6000 Hz_10ms_modulation_mae_db": 0.4789190404040588, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0018364670207998, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0787215966545259, "3000–6000 Hz_40ms_envelope_correlation": 0.9960107670056332, "3000–6000 Hz_40ms_modulation_mae_db": 0.51607271867354, "mr_stft_db": 5.906610114999295, "band_mae_db": 0.19025756133986058, "rms_gain_error_db": 0.027612040243297827, "output_peak": 0.49310139628323435}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2527357376567, "全帯域_10ms_rms_envelope_nrmse": 0.10320431372684971, "全帯域_10ms_envelope_correlation": 0.9905272323561476, "全帯域_10ms_modulation_mae_db": 0.28206442238449103, "全帯域_40ms_log_rms_mae_db": 0.7072144720296503, "全帯域_40ms_rms_envelope_nrmse": 0.054413317781851746, "全帯域_40ms_envelope_correlation": 0.9971725140474473, "全帯域_40ms_modulation_mae_db": 0.7311320994426996, "300–1000 Hz_10ms_log_rms_mae_db": 1.7187790486739807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12190316621420189, "300–1000 Hz_10ms_envelope_correlation": 0.9895725736863252, "300–1000 Hz_10ms_modulation_mae_db": 0.5459421777388319, "300–1000 Hz_40ms_log_rms_mae_db": 1.0125713321657448, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06595970891795241, "300–1000 Hz_40ms_envelope_correlation": 0.9967826752616712, "300–1000 Hz_40ms_modulation_mae_db": 0.10558316889953111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7733385806791069, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21190994515960107, "1000–3000 Hz_10ms_envelope_correlation": 0.9636252167906024, "1000–3000 Hz_10ms_modulation_mae_db": 0.7369887351692093, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9872201657915395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1117771157513212, "1000–3000 Hz_40ms_envelope_correlation": 0.9896365529692026, "1000–3000 Hz_40ms_modulation_mae_db": 0.9828351836273371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8691841355263281, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13289163314986474, "3000–6000 Hz_10ms_envelope_correlation": 0.9891217057838008, "3000–6000 Hz_10ms_modulation_mae_db": 0.49025436290316654, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0067903761871915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07641889138173827, "3000–6000 Hz_40ms_envelope_correlation": 0.9963487597843679, "3000–6000 Hz_40ms_modulation_mae_db": 0.7315910454600717, "mr_stft_db": 5.920310703130661, "band_mae_db": 0.15395963021549997, "rms_gain_error_db": -0.07274174926255633, "output_peak": 0.5871240559564298}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2640161592598202, "全帯域_10ms_rms_envelope_nrmse": 0.0988156919778713, "全帯域_10ms_envelope_correlation": 0.9913673831484814, "全帯域_10ms_modulation_mae_db": 0.2832302005140899, "全帯域_40ms_log_rms_mae_db": 0.7145038968438715, "全帯域_40ms_rms_envelope_nrmse": 0.05134256443754843, "全帯域_40ms_envelope_correlation": 0.9974593565219787, "全帯域_40ms_modulation_mae_db": 0.45937222615929185, "300–1000 Hz_10ms_log_rms_mae_db": 1.6908573018255124, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10993411251589294, "300–1000 Hz_10ms_envelope_correlation": 0.9915550793020242, "300–1000 Hz_10ms_modulation_mae_db": 0.4397199126959361, "300–1000 Hz_40ms_log_rms_mae_db": 1.0122972710018505, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06237076702909107, "300–1000 Hz_40ms_envelope_correlation": 0.9972618319782358, "300–1000 Hz_40ms_modulation_mae_db": 0.20980165692303698, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7479103904043827, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1737648766641518, "1000–3000 Hz_10ms_envelope_correlation": 0.9760023489332039, "1000–3000 Hz_10ms_modulation_mae_db": 0.5936439258975804, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9699380051640142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08884970325813346, "1000–3000 Hz_40ms_envelope_correlation": 0.9931431076596664, "1000–3000 Hz_40ms_modulation_mae_db": 0.7791047459634465, "3000–6000 Hz_10ms_log_rms_mae_db": 1.837626695505746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1291089955319899, "3000–6000 Hz_10ms_envelope_correlation": 0.9896869009105826, "3000–6000 Hz_10ms_modulation_mae_db": 0.5963671322450181, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9994352036107618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06420205657058899, "3000–6000 Hz_40ms_envelope_correlation": 0.9973606569011955, "3000–6000 Hz_40ms_modulation_mae_db": 0.7461004746747496, "mr_stft_db": 5.91181696850137, "band_mae_db": 0.12367014305900215, "rms_gain_error_db": 0.014698757208317616, "output_peak": 0.45346586676216455}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2537174605990298, "全帯域_10ms_rms_envelope_nrmse": 0.11133821443955494, "全帯域_10ms_envelope_correlation": 0.9891616874206539, "全帯域_10ms_modulation_mae_db": 0.5629281462808167, "全帯域_40ms_log_rms_mae_db": 0.6129524618009401, "全帯域_40ms_rms_envelope_nrmse": 0.05967049307545353, "全帯域_40ms_envelope_correlation": 0.996934616577667, "全帯域_40ms_modulation_mae_db": 0.6815089717768835, "300–1000 Hz_10ms_log_rms_mae_db": 1.7697434110131511, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12134221198642373, "300–1000 Hz_10ms_envelope_correlation": 0.9899402916284952, "300–1000 Hz_10ms_modulation_mae_db": 0.382909490177103, "300–1000 Hz_40ms_log_rms_mae_db": 0.9870304562852686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06980439308148369, "300–1000 Hz_40ms_envelope_correlation": 0.9969185847440483, "300–1000 Hz_40ms_modulation_mae_db": 0.19548784828091734, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7370413354411744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17209402344199481, "1000–3000 Hz_10ms_envelope_correlation": 0.9768138611245315, "1000–3000 Hz_10ms_modulation_mae_db": 0.965226903637837, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9921772378469705, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09958266197054762, "1000–3000 Hz_40ms_envelope_correlation": 0.9926347177774327, "1000–3000 Hz_40ms_modulation_mae_db": 1.0232097058470517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8615578692773065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16002941499185774, "3000–6000 Hz_10ms_envelope_correlation": 0.9842264855617406, "3000–6000 Hz_10ms_modulation_mae_db": 0.6897962840935644, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9997320638079777, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08708916997642266, "3000–6000 Hz_40ms_envelope_correlation": 0.9954110086756371, "3000–6000 Hz_40ms_modulation_mae_db": 0.8014641820448833, "mr_stft_db": 5.919219615119253, "band_mae_db": 0.23132968440584098, "rms_gain_error_db": -0.1849461941630895, "output_peak": 0.4778648605258099}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1765574849678857, "全帯域_10ms_rms_envelope_nrmse": 0.09222981804677446, "全帯域_10ms_envelope_correlation": 0.9924478852158685, "全帯域_10ms_modulation_mae_db": 0.15104601436593781, "全帯域_40ms_log_rms_mae_db": 0.6278767612832512, "全帯域_40ms_rms_envelope_nrmse": 0.048212412888483394, "全帯域_40ms_envelope_correlation": 0.9977877353290915, "全帯域_40ms_modulation_mae_db": 0.3064199095258826, "300–1000 Hz_10ms_log_rms_mae_db": 1.7145019907572396, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11411811123782777, "300–1000 Hz_10ms_envelope_correlation": 0.9908644828123687, "300–1000 Hz_10ms_modulation_mae_db": 0.5407078240243443, "300–1000 Hz_40ms_log_rms_mae_db": 0.9794772667273883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06003553707767129, "300–1000 Hz_40ms_envelope_correlation": 0.9974289047132888, "300–1000 Hz_40ms_modulation_mae_db": 0.8356259955079509, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7100499891135348, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18314050455890785, "1000–3000 Hz_10ms_envelope_correlation": 0.9729052411767684, "1000–3000 Hz_10ms_modulation_mae_db": 0.10323845804520088, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9465273798783406, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09808146099776598, "1000–3000 Hz_40ms_envelope_correlation": 0.9915205288326682, "1000–3000 Hz_40ms_modulation_mae_db": 0.25940210554489584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8386939166815783, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12263476893804663, "3000–6000 Hz_10ms_envelope_correlation": 0.9907012259797622, "3000–6000 Hz_10ms_modulation_mae_db": 0.12086351769401908, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0052119814436344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06747151185036053, "3000–6000 Hz_40ms_envelope_correlation": 0.9970592001357608, "3000–6000 Hz_40ms_modulation_mae_db": 0.27683398432102063, "mr_stft_db": 5.908849751823138, "band_mae_db": 0.127295545785992, "rms_gain_error_db": -0.06142412247787861, "output_peak": 0.5109528709567026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.317205272432316, "spectral_convergence": 0.46435972712858076, "spectral_convergence_status": "ok", "amplitude_floor": 1.32832143223519e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.615730669854437, "spectral_convergence": 0.4327018524593207, "spectral_convergence_status": "ok", "amplitude_floor": 1.1133586355586426e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.804249283550757, "spectral_convergence": 0.45870797637243044, "spectral_convergence_status": "ok", "amplitude_floor": 6.710028542715064e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00015990772541983682, "output_power": 0.00016754200884853444, "delta_db": 0.2025427260639287, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005346827040387907, "output_power": 0.0005264806283446048, "delta_db": -0.06713739618582819, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025601253900142847, "output_power": 0.0002484376291643235, "delta_db": -0.13043860611512664, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015659557404440672, "output_power": 0.0001590114948101492, "delta_db": 0.06649037078786396, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003554436518358531, "output_power": 0.0003341246685902276, "delta_db": -0.2686222120760029, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004235024710497043, "output_power": 0.0004645373815096884, "delta_db": 0.4016471889429508, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18947975002861686, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.03209186299159666, "pitch": null, "output_rms_dbfs": -27.212158140741074, "output_peak": 0.5297423230437182, "reference_rms_dbfs": -27.244250003732674, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9123950752791705}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018816516597127566, "log_rms_mae_db": 1.2122206317108506, "rms_envelope_nrmse": 0.11495059184279953, "envelope_correlation": 0.9883506237624421, "rms_gain_error_db": 0.0328201965384588, "reference_mean_power": 0.0018869472812584292, "output_mean_power": 0.001901261204241419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.448172186664129e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039739520206635815, "output_power": 0.00040990894240514075, "reference_active": true, "delta_db": 0.13464773439941913}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001684948117890412, "output_power": 0.00015186273094652648, "reference_active": true, "delta_db": -0.45135327300028893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9559595301379592e-05, "output_power": 2.457074173458255e-05, "reference_active": true, "delta_db": 0.9905840228011864}], "active_band_count": 3, "active_band_mae_db": 0.5255283434002981}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016656172387320294, "log_rms_mae_db": 0.6269799888285947, "rms_envelope_nrmse": 0.058953206320184, "envelope_correlation": 0.9966686537190473, "rms_gain_error_db": 0.033595702429840106, "reference_mean_power": 0.0018945898208991615, "output_mean_power": 0.001909302626008531}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.514104383617561e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003829793712831679, "output_power": 0.00039731110939054986, "reference_active": true, "delta_db": 0.15955326868105454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.831616093630877e-05, "output_power": 7.041932391596735e-05, "reference_active": true, "delta_db": -0.4615953900473086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3812315351816923e-06, "output_power": 2.3555896936625718e-06, "reference_active": true, "delta_db": -0.04701979879923697}], "active_band_count": 3, "active_band_mae_db": 0.2227228191758667}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015092951870698086, "log_rms_mae_db": 1.7121597604072651, "rms_envelope_nrmse": 0.11808257422919145, "envelope_correlation": 0.9902337296535445, "rms_gain_error_db": -0.13390887198813, "reference_mean_power": 0.000631981580682194, "output_mean_power": 0.0006127926302416488}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.167139764996573e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018037000728302603, "output_power": 0.00017623221101260874, "reference_active": true, "delta_db": -0.10079032891045486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6944061726753776e-05, "output_power": 2.3782625141470616e-05, "reference_active": true, "delta_db": -0.5420327412263181}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.521461134259169e-06, "output_power": 4.778649931031267e-06, "reference_active": true, "delta_db": 1.3258231710499442}], "active_band_count": 3, "active_band_mae_db": 0.6562154137289057}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013588907795973108, "log_rms_mae_db": 0.9505734803638501, "rms_envelope_nrmse": 0.06534627264362956, "envelope_correlation": 0.9969684493153762, "rms_gain_error_db": -0.13314012628076818, "reference_mean_power": 0.0006345375227010602, "output_mean_power": 0.0006153798847153106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.990151794989553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001715162379513309, "output_power": 0.00016764712459452276, "reference_active": true, "delta_db": -0.09909133261009898}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2552077755571312e-05, "output_power": 1.1297868647132714e-05, "reference_active": true, "delta_db": -0.457190997099309}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.057253626713498e-07, "output_power": 3.3461130577584e-07, "reference_active": true, "delta_db": 0.39209141866186326}], "active_band_count": 3, "active_band_mae_db": 0.3161245827904237}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.324369547066356e-05, "log_rms_mae_db": 1.7039742571615202, "rms_envelope_nrmse": 0.18244594732932798, "envelope_correlation": 0.9732853910089589, "rms_gain_error_db": -0.043124462107534116, "reference_mean_power": 0.00026216016454011753, "output_mean_power": 0.00025956985513796464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5456728460340458e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.502085853898132e-05, "output_power": 7.61543930963518e-05, "reference_active": true, "delta_db": 0.06512931503525199}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151594827944158e-05, "output_power": 2.271382139333094e-05, "reference_active": true, "delta_db": 0.2352971480974263}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.513878290774259e-06, "output_power": 4.605505255607858e-06, "reference_active": true, "delta_db": -2.6685015523698894}], "active_band_count": 3, "active_band_mae_db": 0.9896426718341892}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.873751028776232e-05, "log_rms_mae_db": 0.9203787426181712, "rms_envelope_nrmse": 0.10185225974833946, "envelope_correlation": 0.9908585650040967, "rms_gain_error_db": -0.04221398115549553, "reference_mean_power": 0.0002632887650860277, "output_mean_power": 0.0002607419621680038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3635149275752304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.257955032157826e-05, "output_power": 7.378956969633796e-05, "reference_active": true, "delta_db": 0.07180704634624976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.405048683009261e-06, "output_power": 9.852919364488659e-06, "reference_active": true, "delta_db": 0.20203880657797657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.677553554297748e-07, "output_power": 4.4342932153967473e-07, "reference_active": true, "delta_db": -2.3839844728884536}], "active_band_count": 3, "active_band_mae_db": 0.88594344193756}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015682729984940356, "log_rms_mae_db": 1.8670483921562726, "rms_envelope_nrmse": 0.1721270437277129, "envelope_correlation": 0.9823276204016168, "rms_gain_error_db": 0.16766209042501892, "reference_mean_power": 0.0005797089728212459, "output_mean_power": 0.0006025266095517686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2924452919576597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020605682430557354, "output_power": 0.0002164498290113313, "reference_active": true, "delta_db": 0.21370244762883073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012645863441832237, "output_power": 0.00011319658411926178, "reference_active": true, "delta_db": -0.4811516616420907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.26990748699409e-06, "output_power": 1.630579811358693e-05, "reference_active": true, "delta_db": 2.4526666096745924}], "active_band_count": 3, "active_band_mae_db": 1.0491735729818379}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012704534221320322, "log_rms_mae_db": 1.0265814644195204, "rms_envelope_nrmse": 0.09512457691202857, "envelope_correlation": 0.9944910887709969, "rms_gain_error_db": 0.16798392619465086, "reference_mean_power": 0.0005824704905358117, "output_mean_power": 0.0006054416868433259}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.040463159930301e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022945963060392102, "output_power": 0.00023927525192828504, "reference_active": true, "delta_db": 0.18191492062763215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.01414803619856e-05, "output_power": 5.463504567990201e-05, "reference_active": true, "delta_db": -0.417028033835099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3285241811094974e-06, "output_power": 1.6622684441456527e-06, "reference_active": true, "delta_db": 0.9733169685954104}], "active_band_count": 3, "active_band_mae_db": 0.5240866410193805}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8243728701708789, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5265604cc45d5f9f48f1ff2c42f9fa8ec373f089d276215f73c560b2a488ed55", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9736480009325492, "standard_deviation": 0.00443749241095258, "minimum": 0.9636252167906024, "maximum": 0.9770105237916382, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7284618601570538, "standard_deviation": 0.031019606001007886, "minimum": 1.673629883108352, "maximum": 1.7733385806791069, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8053988212906282, "standard_deviation": 0.36302949176508337, "minimum": 0.10323845804520088, "maximum": 1.3194078403935425, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18074715374092107, "standard_deviation": 0.013938635814237588, "minimum": 0.16885979264540676, "maximum": 0.21190994515960107, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9917631813058079, "standard_deviation": 0.0017443083945246842, "minimum": 0.9892383434248958, "maximum": 0.9940699096266065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9571797083330552, "standard_deviation": 0.024505401749377556, "minimum": 0.9203787426181712, "maximum": 0.9921772378469705, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8860965998581039, "standard_deviation": 0.4090786487748062, "minimum": 0.25940210554489584, "maximum": 1.5238495782722576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09809692791527883, "standard_deviation": 0.010030543187052182, "minimum": 0.08479296977929923, "maximum": 0.1117771157513212, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9867076645550713, "standard_deviation": 0.0035197645711985007, "minimum": 0.9816830260768666, "maximum": 0.9907012259797622, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.845592026180443, "standard_deviation": 0.029929966847989292, "minimum": 1.7826921434794027, "maximum": 1.8738544490165447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5107640782858573, "standard_deviation": 0.28178791446079526, "minimum": 0.12086351769401908, "maximum": 1.0491735729818379, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1460365558810964, "standard_deviation": 0.01987510910631956, "minimum": 0.12263476893804663, "maximum": 0.17279756373718733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9960749800770712, "standard_deviation": 0.0009111500821798557, "minimum": 0.9944910887709969, "maximum": 0.9973606569011955, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.001778100813679, "standard_deviation": 0.016160236792100973, "minimum": 0.9679779972655739, "maximum": 1.0265814644195204, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.546813372714079, "standard_deviation": 0.22273853428453563, "minimum": 0.1835890397617972, "maximum": 0.8014641820448833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07837217518740384, "standard_deviation": 0.009977033606452971, "minimum": 0.06420205657058899, "maximum": 0.09512457691202857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9902042782094624, "standard_deviation": 0.0007157961179617981, "minimum": 0.9895365757270472, "maximum": 0.9915550793020242, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7128528132693615, "standard_deviation": 0.029560521967771612, "minimum": 1.6691266297067797, "maximum": 1.7697434110131511, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4534347128020638, "standard_deviation": 0.14979877851575119, "minimum": 0.1458998802558029, "maximum": 0.6562154137289057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11837975246554314, "standard_deviation": 0.004469349703417894, "minimum": 0.10993411251589294, "maximum": 0.12233855500439744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968944563179478, "standard_deviation": 0.0005330356378362852, "minimum": 0.9958414026960202, "maximum": 0.9974390808995003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9854851282185031, "standard_deviation": 0.02186255787408036, "minimum": 0.9505734803638501, "maximum": 1.0125713321657448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3833740138128886, "standard_deviation": 0.2703696713286239, "minimum": 0.10558316889953111, "maximum": 0.8356259955079509, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06631603652052223, "standard_deviation": 0.005053819281391993, "minimum": 0.06003553707767129, "maximum": 0.07515379437791647, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1754039490910957, "standard_deviation": 0.04170211898385645, "minimum": 0.12367014305900215, "maximum": 0.23132968440584098, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.913352562538525, "standard_deviation": 0.00529781632428288, "minimum": 5.906610114999295, "maximum": 5.920310703130661, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4930620627174891, "standard_deviation": 0.04941553064973747, "minimum": 0.4324361967810148, "maximum": 0.5871240559564298, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0484485502653806, "standard_deviation": 0.07862745843441153, "minimum": -0.1849461941630895, "maximum": 0.03209186299159666, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9902183049862254, "standard_deviation": 0.0015613811665056055, "minimum": 0.9880889684108698, "maximum": 0.9924478852158685, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2424784401548779, "standard_deviation": 0.036020847039526356, "minimum": 1.1765574849678857, "maximum": 1.2983109772616168, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3749855557608346, "standard_deviation": 0.15379236070085983, "minimum": 0.15104601436593781, "maximum": 0.5629281462808167, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10492804172167307, "standard_deviation": 0.008611363690064763, "minimum": 0.09222981804677446, "maximum": 0.1162715258113068, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.99723262492542, "standard_deviation": 0.00034914711343813704, "minimum": 0.9966686537190473, "maximum": 0.9977877353290915, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6542341762657828, "standard_deviation": 0.040606880792358274, "minimum": 0.6129524618009401, "maximum": 0.7145038968438715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4768904262225707, "standard_deviation": 0.18353886517409349, "minimum": 0.2227228191758667, "maximum": 0.7311320994426996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05422447723678018, "standard_deviation": 0.0038003326472770644, "minimum": 0.048212412888483394, "maximum": 0.05967049307545353, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4446-2271-0003", "language": "en", "style": "normal", "speaker": "librispeech_4446", "source_utterance_id": "4446-2271-0003", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 59200, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4446-2271-0003.wav", "source_file_sha256": "74827694544d99377da5076577fe75292a5490f14ec9e15a1567d5fc674155b3", "decoded_pcm_sha256": "257baf7b0200414f4ef5d6ab5db2362889681a1c0de114633c54f50f9b628549", "original_source_sha256": "b8fecefeca8a7ae57ad53900dbba581245a09ad79d46ae1fc06140e0e5a5b281", "corpus_entry_id": "librispeech_4446-2271-0003"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.479140281677246, "predicted_mos": 3.479140281677246, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "44460e50fae0ac58d10f8bde235e44f7351a8d9e2530826d2c629443558482d0", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.0848485580027045, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4446-2271-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "44460e50fae0ac58d10f8bde235e44f7351a8d9e2530826d2c629443558482d0", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "8fd653ce4da1222d377403b051d957dbbd98eca247e0a76171fa28bf4a523f5f"}, "rows": [{"case": "librispeech_4446-2271-0003", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.5726349353790283, "predicted_mos": 2.5726349353790283, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "89f99d2dc65278fab6243a67e870b214dd27014143acee06cebe6192a1d07c1c", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.470347286001925, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4446-2271-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "89f99d2dc65278fab6243a67e870b214dd27014143acee06cebe6192a1d07c1c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7765968866169247, "全帯域_10ms_rms_envelope_nrmse": 0.07161194134556598, "全帯域_10ms_envelope_correlation": 0.9930957525591753, "全帯域_10ms_modulation_mae_db": 0.3295771905717984, "全帯域_40ms_log_rms_mae_db": 0.41816405240323185, "全帯域_40ms_rms_envelope_nrmse": 0.03487575486881722, "全帯域_40ms_envelope_correlation": 0.9981580615982151, "全帯域_40ms_modulation_mae_db": 0.3697518645002402, "300–1000 Hz_10ms_log_rms_mae_db": 1.1084645167347933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09747130879955733, "300–1000 Hz_10ms_envelope_correlation": 0.9894548227695171, "300–1000 Hz_10ms_modulation_mae_db": 0.3700452858824666, "300–1000 Hz_40ms_log_rms_mae_db": 0.5905902492746401, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049169354381971356, "300–1000 Hz_40ms_envelope_correlation": 0.9970610696843432, "300–1000 Hz_40ms_modulation_mae_db": 0.3154194605014304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5599743567954019, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1326653627546369, "1000–3000 Hz_10ms_envelope_correlation": 0.98643734371885, "1000–3000 Hz_10ms_modulation_mae_db": 0.22906553693197218, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0375329641356692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08275423961950074, "1000–3000 Hz_40ms_envelope_correlation": 0.9944773447462374, "1000–3000 Hz_40ms_modulation_mae_db": 0.2996841469487643, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2024089962685474, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5909408348564645, "3000–6000 Hz_10ms_envelope_correlation": 0.8116110890571903, "3000–6000 Hz_10ms_modulation_mae_db": 0.871408570699948, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9795766518264941, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5016115868748389, "3000–6000 Hz_40ms_envelope_correlation": 0.8526235337236809, "3000–6000 Hz_40ms_modulation_mae_db": 0.7653146125138653, "mr_stft_db": 6.029910768437212, "band_mae_db": 0.6620648209454681, "rms_gain_error_db": -0.06361395745852891, "output_peak": 0.4773840079363525}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7940057316509641, "全帯域_10ms_rms_envelope_nrmse": 0.07486780722769702, "全帯域_10ms_envelope_correlation": 0.9924677632951587, "全帯域_10ms_modulation_mae_db": 0.412897123573909, "全帯域_40ms_log_rms_mae_db": 0.4688586579405995, "全帯域_40ms_rms_envelope_nrmse": 0.043667616449653224, "全帯域_40ms_envelope_correlation": 0.9971017217420592, "全帯域_40ms_modulation_mae_db": 0.4814626388833679, "300–1000 Hz_10ms_log_rms_mae_db": 1.0495513746016332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09903779842823664, "300–1000 Hz_10ms_envelope_correlation": 0.9891796278261479, "300–1000 Hz_10ms_modulation_mae_db": 0.5964043186092577, "300–1000 Hz_40ms_log_rms_mae_db": 0.6256795649027586, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06105014784115388, "300–1000 Hz_40ms_envelope_correlation": 0.9954661637286206, "300–1000 Hz_40ms_modulation_mae_db": 0.567811133168135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.580441156915629, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13554910656725452, "1000–3000 Hz_10ms_envelope_correlation": 0.985889520738258, "1000–3000 Hz_10ms_modulation_mae_db": 0.36161463551661605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.063955983886842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08575323430776775, "1000–3000 Hz_40ms_envelope_correlation": 0.9942547215746397, "1000–3000 Hz_40ms_modulation_mae_db": 0.4761954157308172, "3000–6000 Hz_10ms_log_rms_mae_db": 2.220450563614042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5820858907991652, "3000–6000 Hz_10ms_envelope_correlation": 0.8080916837095111, "3000–6000 Hz_10ms_modulation_mae_db": 0.3690737026988071, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9778141029327918, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49851270177605556, "3000–6000 Hz_40ms_envelope_correlation": 0.8452551562925474, "3000–6000 Hz_40ms_modulation_mae_db": 0.3598795223063302, "mr_stft_db": 6.019245899174142, "band_mae_db": 0.6439613860416509, "rms_gain_error_db": -0.07626365288763938, "output_peak": 0.4714283739647763}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8477643480309679, "全帯域_10ms_rms_envelope_nrmse": 0.07232417960049038, "全帯域_10ms_envelope_correlation": 0.9929533934993389, "全帯域_10ms_modulation_mae_db": 0.4513568568937271, "全帯域_40ms_log_rms_mae_db": 0.4737269276293024, "全帯域_40ms_rms_envelope_nrmse": 0.03964780724858432, "全帯域_40ms_envelope_correlation": 0.9975886449187085, "全帯域_40ms_modulation_mae_db": 0.4732018754728274, "300–1000 Hz_10ms_log_rms_mae_db": 1.171336820011832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09754970484109707, "300–1000 Hz_10ms_envelope_correlation": 0.9893998384682798, "300–1000 Hz_10ms_modulation_mae_db": 0.6696707648973472, "300–1000 Hz_40ms_log_rms_mae_db": 0.6595226687011794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05558945890834277, "300–1000 Hz_40ms_envelope_correlation": 0.9961284259016921, "300–1000 Hz_40ms_modulation_mae_db": 0.6272604823730289, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4802131593937007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1302623448274755, "1000–3000 Hz_10ms_envelope_correlation": 0.9876256396490597, "1000–3000 Hz_10ms_modulation_mae_db": 0.46851848412176467, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0334519377842497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09126541197939288, "1000–3000 Hz_40ms_envelope_correlation": 0.9943403766596982, "1000–3000 Hz_40ms_modulation_mae_db": 0.3564641925934498, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2094648620546846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5665440080478281, "3000–6000 Hz_10ms_envelope_correlation": 0.8138543040788212, "3000–6000 Hz_10ms_modulation_mae_db": 0.614157721075027, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9963109269697514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49040710906660123, "3000–6000 Hz_40ms_envelope_correlation": 0.8456933398505515, "3000–6000 Hz_40ms_modulation_mae_db": 1.0181683745167633, "mr_stft_db": 6.010815037405675, "band_mae_db": 0.6668015556550704, "rms_gain_error_db": -0.058480319158228454, "output_peak": 0.5424104418265422}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7960942593001294, "全帯域_10ms_rms_envelope_nrmse": 0.07287037217191006, "全帯域_10ms_envelope_correlation": 0.9928468314308825, "全帯域_10ms_modulation_mae_db": 0.18591934131333723, "全帯域_40ms_log_rms_mae_db": 0.43218602765502445, "全帯域_40ms_rms_envelope_nrmse": 0.03743465205013986, "全帯域_40ms_envelope_correlation": 0.9978266152158729, "全帯域_40ms_modulation_mae_db": 0.3240106655169562, "300–1000 Hz_10ms_log_rms_mae_db": 1.1477823149207411, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10166949705744734, "300–1000 Hz_10ms_envelope_correlation": 0.9884291432884145, "300–1000 Hz_10ms_modulation_mae_db": 0.3472777555673177, "300–1000 Hz_40ms_log_rms_mae_db": 0.6489613202881264, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053103819849709015, "300–1000 Hz_40ms_envelope_correlation": 0.9963544091134126, "300–1000 Hz_40ms_modulation_mae_db": 0.5196072195350672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.530743239860176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12060888382374338, "1000–3000 Hz_10ms_envelope_correlation": 0.9888245297663285, "1000–3000 Hz_10ms_modulation_mae_db": 0.2967629653613097, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0687679856158094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0752631932938098, "1000–3000 Hz_40ms_envelope_correlation": 0.9955048499223562, "1000–3000 Hz_40ms_modulation_mae_db": 0.38657747097254624, "3000–6000 Hz_10ms_log_rms_mae_db": 2.268545594567096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6018425229990325, "3000–6000 Hz_10ms_envelope_correlation": 0.78232592674088, "3000–6000 Hz_10ms_modulation_mae_db": 0.8914345225462231, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0626722018925046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5149869180063459, "3000–6000 Hz_40ms_envelope_correlation": 0.8221925376664508, "3000–6000 Hz_40ms_modulation_mae_db": 1.2742258860800095, "mr_stft_db": 5.995702403089678, "band_mae_db": 0.5582681884525796, "rms_gain_error_db": -0.03330338005883764, "output_peak": 0.5320714631112232}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8131420064401718, "全帯域_10ms_rms_envelope_nrmse": 0.08192626892412025, "全帯域_10ms_envelope_correlation": 0.9911787545259813, "全帯域_10ms_modulation_mae_db": 0.16511610250038047, "全帯域_40ms_log_rms_mae_db": 0.4403079533653575, "全帯域_40ms_rms_envelope_nrmse": 0.048836883558129277, "全帯域_40ms_envelope_correlation": 0.9966625526726826, "全帯域_40ms_modulation_mae_db": 0.18455055811156154, "300–1000 Hz_10ms_log_rms_mae_db": 1.168221782747402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10952004641370156, "300–1000 Hz_10ms_envelope_correlation": 0.9870047147063965, "300–1000 Hz_10ms_modulation_mae_db": 0.3692090497831157, "300–1000 Hz_40ms_log_rms_mae_db": 0.6216379827984779, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07013722773726597, "300–1000 Hz_40ms_envelope_correlation": 0.9943368707705482, "300–1000 Hz_40ms_modulation_mae_db": 0.3295166937283571, "1000–3000 Hz_10ms_log_rms_mae_db": 1.497056262180185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12978675561081143, "1000–3000 Hz_10ms_envelope_correlation": 0.9879368319118671, "1000–3000 Hz_10ms_modulation_mae_db": 0.9828884220555748, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0081140790142702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08628427702942228, "1000–3000 Hz_40ms_envelope_correlation": 0.995311988293137, "1000–3000 Hz_40ms_modulation_mae_db": 0.9795339281210822, "3000–6000 Hz_10ms_log_rms_mae_db": 2.210889437473118, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5921107590456426, "3000–6000 Hz_10ms_envelope_correlation": 0.8034846111587124, "3000–6000 Hz_10ms_modulation_mae_db": 0.5684219005077845, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9609492957880106, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5027133327078505, "3000–6000 Hz_40ms_envelope_correlation": 0.8446659184144153, "3000–6000 Hz_40ms_modulation_mae_db": 0.8984916910579179, "mr_stft_db": 6.014189439309074, "band_mae_db": 0.6973603940223235, "rms_gain_error_db": -0.14667947603813136, "output_peak": 0.5457506770145948}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8122476614747692, "全帯域_10ms_rms_envelope_nrmse": 0.07696295352971985, "全帯域_10ms_envelope_correlation": 0.9924550327364831, "全帯域_10ms_modulation_mae_db": 0.4156976553550617, "全帯域_40ms_log_rms_mae_db": 0.47626556778783535, "全帯域_40ms_rms_envelope_nrmse": 0.04930316277119051, "全帯域_40ms_envelope_correlation": 0.9968035848050932, "全帯域_40ms_modulation_mae_db": 0.48093916169896306, "300–1000 Hz_10ms_log_rms_mae_db": 1.1784540407263544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10826703452694558, "300–1000 Hz_10ms_envelope_correlation": 0.9877215588507884, "300–1000 Hz_10ms_modulation_mae_db": 0.6060777460212998, "300–1000 Hz_40ms_log_rms_mae_db": 0.6887230966988943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06595864959320348, "300–1000 Hz_40ms_envelope_correlation": 0.9956324885463553, "300–1000 Hz_40ms_modulation_mae_db": 0.6729508850430252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4776967985638256, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11840585292675784, "1000–3000 Hz_10ms_envelope_correlation": 0.9894442471582726, "1000–3000 Hz_10ms_modulation_mae_db": 0.2686824669416112, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0270489505144385, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07899136237415891, "1000–3000 Hz_40ms_envelope_correlation": 0.9953788331241071, "1000–3000 Hz_40ms_modulation_mae_db": 0.2004757495002927, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1998278519151215, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5636498941600807, "3000–6000 Hz_10ms_envelope_correlation": 0.8111353004134996, "3000–6000 Hz_10ms_modulation_mae_db": 0.4427115738436818, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9574657195006808, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.487804884349398, "3000–6000 Hz_40ms_envelope_correlation": 0.8426407224372063, "3000–6000 Hz_40ms_modulation_mae_db": 0.8879109353015817, "mr_stft_db": 5.989710504916037, "band_mae_db": 0.6358596325753659, "rms_gain_error_db": -0.18019654254306808, "output_peak": 0.49156908677192407}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.822844586040355, "全帯域_10ms_rms_envelope_nrmse": 0.07152796219909721, "全帯域_10ms_envelope_correlation": 0.9932092644268772, "全帯域_10ms_modulation_mae_db": 0.4807644613521556, "全帯域_40ms_log_rms_mae_db": 0.47042055472112093, "全帯域_40ms_rms_envelope_nrmse": 0.03846734006065042, "全帯域_40ms_envelope_correlation": 0.9978828192097801, "全帯域_40ms_modulation_mae_db": 0.35261259660439787, "300–1000 Hz_10ms_log_rms_mae_db": 1.2864503532914198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1103617915742639, "300–1000 Hz_10ms_envelope_correlation": 0.9867696328431251, "300–1000 Hz_10ms_modulation_mae_db": 0.777469086946673, "300–1000 Hz_40ms_log_rms_mae_db": 0.7738981646017477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06738248021968671, "300–1000 Hz_40ms_envelope_correlation": 0.9948316397893998, "300–1000 Hz_40ms_modulation_mae_db": 0.5132113011156463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.512706049796335, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13160155250541578, "1000–3000 Hz_10ms_envelope_correlation": 0.9866482424380093, "1000–3000 Hz_10ms_modulation_mae_db": 0.38562099567450714, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0994786728967563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0823918485105905, "1000–3000 Hz_40ms_envelope_correlation": 0.9945005553281125, "1000–3000 Hz_40ms_modulation_mae_db": 0.42438493153749324, "3000–6000 Hz_10ms_log_rms_mae_db": 2.228580134601372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5879955003799194, "3000–6000 Hz_10ms_envelope_correlation": 0.802401884871559, "3000–6000 Hz_10ms_modulation_mae_db": 0.46043026547796106, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9951189872218333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5031689562643827, "3000–6000 Hz_40ms_envelope_correlation": 0.8399846767586073, "3000–6000 Hz_40ms_modulation_mae_db": 0.525541316413435, "mr_stft_db": 6.011786078105108, "band_mae_db": 0.6368691181524906, "rms_gain_error_db": -0.10690554667246344, "output_peak": 0.49896310624475504}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9019592378318904, "全帯域_10ms_rms_envelope_nrmse": 0.08479614572545613, "全帯域_10ms_envelope_correlation": 0.9903574810154991, "全帯域_10ms_modulation_mae_db": 0.4672213719113986, "全帯域_40ms_log_rms_mae_db": 0.49180721791377235, "全帯域_40ms_rms_envelope_nrmse": 0.0440648361933248, "全帯域_40ms_envelope_correlation": 0.9971316645462364, "全帯域_40ms_modulation_mae_db": 0.538726040454966, "300–1000 Hz_10ms_log_rms_mae_db": 1.232386960307351, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11452056065147116, "300–1000 Hz_10ms_envelope_correlation": 0.9854227180142211, "300–1000 Hz_10ms_modulation_mae_db": 0.5601711856451521, "300–1000 Hz_40ms_log_rms_mae_db": 0.6816833674448388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062133629312981754, "300–1000 Hz_40ms_envelope_correlation": 0.995310876970621, "300–1000 Hz_40ms_modulation_mae_db": 0.7110394698853003, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5890769868526011, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1574430403499874, "1000–3000 Hz_10ms_envelope_correlation": 0.9817055422048829, "1000–3000 Hz_10ms_modulation_mae_db": 0.8906986612345286, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0796553821355868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1150807468755306, "1000–3000 Hz_40ms_envelope_correlation": 0.9905445456468299, "1000–3000 Hz_40ms_modulation_mae_db": 0.8452327605193526, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2737829541909567, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5857238663878448, "3000–6000 Hz_10ms_envelope_correlation": 0.803949318058767, "3000–6000 Hz_10ms_modulation_mae_db": 0.43096059808366505, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0117416936657557, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4986365515935258, "3000–6000 Hz_40ms_envelope_correlation": 0.8431461377870256, "3000–6000 Hz_40ms_modulation_mae_db": 0.9401473787603067, "mr_stft_db": 5.9846951064974325, "band_mae_db": 0.6445393080861946, "rms_gain_error_db": -0.10083771692402126, "output_peak": 0.4923521920326539}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59200, "output_samples": 59200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.713822320545142, "spectral_convergence": 0.20617611399535293, "spectral_convergence_status": "ok", "amplitude_floor": 2.0296349328313604e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7056967759547, "spectral_convergence": 0.20595916355616556, "spectral_convergence_status": "ok", "amplitude_floor": 1.6850709272784723e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.670213208811794, "spectral_convergence": 0.21711512282401965, "spectral_convergence_status": "ok", "amplitude_floor": 8.273401155700601e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185911287346583, "output_power": 0.0008024795206566276, "delta_db": -0.08633075795480652, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003085801603155809, "output_power": 0.00301772120923491, "delta_db": -0.096888851324716, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013392891870332754, "output_power": 0.001340324187530089, "delta_db": 0.0033549244913721136, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027936558999729056, "output_power": 0.00027491015004163727, "delta_db": -0.06982137318116763, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.040755904577448e-06, "output_power": 9.894428442140934e-06, "delta_db": 1.477714239166892, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.309787748977687e-06, "output_power": 1.0564346288134419e-05, "delta_db": 2.2382787795538546, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6620648209454681, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06361395745852891, "pitch": null, "output_rms_dbfs": -22.631340885345107, "output_peak": 0.4773840079363525, "reference_rms_dbfs": -22.567726927886582, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.029910768437212}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002672933594000982, "log_rms_mae_db": 0.7765968866169247, "rms_envelope_nrmse": 0.07161194134556598, "envelope_correlation": 0.9930957525591753, "rms_gain_error_db": -0.06361401285669864, "reference_mean_power": 0.0055438895852693145, "output_mean_power": 0.00546327637157158}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.3303693041440347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007860073564586987, "output_power": 0.0007534813440326913, "reference_active": true, "delta_db": -0.18354106954340502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024364364087793404, "output_power": 0.00023330323499105493, "reference_active": true, "delta_db": -0.18834319959766788}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.963914121364001e-05, "output_power": 4.306649512648689e-05, "reference_active": true, "delta_db": -0.6168473025743224}], "active_band_count": 3, "active_band_mae_db": 0.3295771905717984}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00019127079187884833, "log_rms_mae_db": 0.41816405240323185, "rms_envelope_nrmse": 0.03487575486881722, "envelope_correlation": 0.9981580615982151, "rms_gain_error_db": -0.0636141242704274, "reference_mean_power": 0.005589268172913711, "output_mean_power": 0.005507994971839096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.0930617045795464e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007030365316665854, "output_power": 0.0006753747080221686, "reference_active": true, "delta_db": -0.1743309992078304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013359627830049334, "output_power": 0.00012650567859931325, "reference_active": true, "delta_db": -0.23684339212928446}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7787329335262405e-06, "output_power": 2.3661304694748653e-06, "reference_active": true, "delta_db": -0.6980812021636058}], "active_band_count": 3, "active_band_mae_db": 0.3697518645002402}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002516933783769556, "log_rms_mae_db": 1.1084645167347933, "rms_envelope_nrmse": 0.09747130879955733, "envelope_correlation": 0.9894548227695171, "rms_gain_error_db": -0.12010525331419998, "reference_mean_power": 0.003609667084718048, "output_mean_power": 0.0035112085385938883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.0389022511376814e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005770551731550911, "output_power": 0.000546421298560565, "reference_active": true, "delta_db": -0.23689719667321119}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021321841152816334, "output_power": 0.0001980767779912897, "reference_active": true, "delta_db": -0.3199114058122323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0182675609825566e-05, "output_power": 4.417952961098165e-05, "reference_active": true, "delta_db": -0.5533272551619564}], "active_band_count": 3, "active_band_mae_db": 0.3700452858824666}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00017703062295034008, "log_rms_mae_db": 0.5905902492746401, "rms_envelope_nrmse": 0.049169354381971356, "envelope_correlation": 0.9970610696843432, "rms_gain_error_db": -0.12010529174011782, "reference_mean_power": 0.003639213964400563, "output_mean_power": 0.003539949455896353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 8.593451292498866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005240710435115444, "output_power": 0.0004964821892162882, "reference_active": true, "delta_db": -0.2348649105322076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012044100626957096, "output_power": 0.00010998587634894168, "reference_active": true, "delta_db": -0.3943745578778066}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.706489326091672e-06, "output_power": 2.5159640271309566e-06, "reference_active": true, "delta_db": -0.3170189130942768}], "active_band_count": 3, "active_band_mae_db": 0.3154194605014304}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 7.943457135434651e-05, "log_rms_mae_db": 1.5599743567954019, "rms_envelope_nrmse": 0.1326653627546369, "envelope_correlation": 0.98643734371885, "rms_gain_error_db": -0.05503347931154762, "reference_mean_power": 0.00024301128670702193, "output_mean_power": 0.00023995129439068595}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.5901124046586967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.178189042095757e-05, "output_power": 9.113483427477752e-05, "reference_active": true, "delta_db": -0.030725903975258963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7997292458206206e-05, "output_power": 1.5733757025422217e-05, "reference_active": true, "delta_db": -0.583747349271889}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3356967361336817e-06, "output_power": 2.3751376805832028e-06, "reference_active": true, "delta_db": 0.07272335754876862}], "active_band_count": 3, "active_band_mae_db": 0.22906553693197218}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 6.195280015885064e-05, "log_rms_mae_db": 1.0375329641356692, "rms_envelope_nrmse": 0.08275423961950074, "envelope_correlation": 0.9944773447462374, "rms_gain_error_db": -0.05503412130044117, "reference_mean_power": 0.00024500023514433585, "output_mean_power": 0.00024191516227607018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.4888167929489627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.774520748189054e-05, "output_power": 8.740540043300629e-05, "reference_active": true, "delta_db": -0.01685138443481808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.213495209539377e-06, "output_power": 7.928633174262146e-06, "reference_active": true, "delta_db": -0.6522608910254859}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5741557318464897e-07, "output_power": 1.6597463093436264e-07, "reference_active": true, "delta_db": 0.22994016538598883}], "active_band_count": 3, "active_band_mae_db": 0.2996841469487643}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 1.5353574631917574e-05, "log_rms_mae_db": 2.2024089962685474, "rms_envelope_nrmse": 0.5909408348564645, "envelope_correlation": 0.8116110890571903, "rms_gain_error_db": 1.6315325752862488, "reference_mean_power": 7.123238480443002e-06, "output_mean_power": 1.0371241375641232e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 4.409187309389394e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6167395840190636e-06, "output_power": 2.73023225679907e-06, "reference_active": true, "delta_db": 0.18439089272247117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.009142614437233e-07, "output_power": 9.179490456663456e-07, "reference_active": true, "delta_db": 1.1715367803831576}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.642480310119981e-07, "output_power": 3.5305434756976804e-07, "reference_active": true, "delta_db": 1.2582980389942158}], "active_band_count": 3, "active_band_mae_db": 0.871408570699948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 1.282880604964887e-05, "log_rms_mae_db": 1.9795766518264941, "rms_envelope_nrmse": 0.5016115868748389, "envelope_correlation": 0.8526235337236809, "rms_gain_error_db": 1.6315488105116827, "reference_mean_power": 7.1814170088864185e-06, "output_mean_power": 1.0455986817313766e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 3.966042949113477e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6674623865712373e-06, "output_power": 2.810315864427158e-06, "reference_active": true, "delta_db": 0.22656830790574148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.255436860296439e-07, "output_power": 5.119790390446556e-07, "reference_active": true, "delta_db": 1.966429043155039}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7820276752795505e-08, "output_power": 2.8487615214177958e-08, "reference_active": true, "delta_db": 0.1029464864808152}], "active_band_count": 3, "active_band_mae_db": 0.7653146125138653}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9161885646635092, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6dd6f814a2dac9b2e015bfa482430ed66853c57dd0a25c6536bc6d795be268c5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9868139871981909, "standard_deviation": 0.0023904658526615456, "minimum": 0.9817055422048829, "maximum": 0.9894442471582726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5284885012947318, "standard_deviation": 0.04393804580459486, "minimum": 1.4776967985638256, "maximum": 1.5890769868526011, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4854815209797355, "standard_deviation": 0.2892312743374039, "minimum": 0.22906553693197218, "maximum": 0.9828884220555748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13204036242076034, "standard_deviation": 0.01185031134788526, "minimum": 0.11840585292675784, "maximum": 0.1574430403499874, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942891519118897, "standard_deviation": 0.0015954181946276087, "minimum": 0.9905445456468299, "maximum": 0.9955048499223562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0522507444979528, "standard_deviation": 0.030576618333170706, "minimum": 1.0081140790142702, "maximum": 1.0994786728967563, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4960685744904748, "standard_deviation": 0.2721888815913002, "minimum": 0.2004757495002927, "maximum": 0.9795339281210822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08722303924877169, "standard_deviation": 0.012244784581258853, "minimum": 0.0752631932938098, "maximum": 0.1150807468755306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8046067647611175, "standard_deviation": 0.009945082655719084, "minimum": 0.78232592674088, "maximum": 0.8138543040788212, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.226743799335617, "standard_deviation": 0.028957166511482547, "minimum": 2.1998278519151215, "maximum": 2.2737829541909567, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5810748568666373, "standard_deviation": 0.20108621330042253, "minimum": 0.3690737026988071, "maximum": 0.8914345225462231, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5838616595844972, "standard_deviation": 0.012948249405006195, "minimum": 0.5636498941600807, "maximum": 0.6018425229990325, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8420252528663106, "standard_deviation": 0.008801944093564835, "minimum": 0.8221925376664508, "maximum": 0.8526235337236809, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.9927061974747278, "standard_deviation": 0.03363875116561924, "minimum": 1.9574657195006808, "maximum": 2.0626722018925046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8337099646187762, "standard_deviation": 0.28553299795888826, "minimum": 0.3598795223063302, "maximum": 1.2742258860800095, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.4997302550798748, "standard_deviation": 0.008356714434285673, "minimum": 0.487804884349398, "maximum": 0.5149869180063459, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9879227570958613, "standard_deviation": 0.0014554488327096558, "minimum": 0.9854227180142211, "maximum": 0.9894548227695171, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1678310204176907, "standard_deviation": 0.07199177964871481, "minimum": 1.0495513746016332, "maximum": 1.2864503532914198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5370406491690788, "standard_deviation": 0.15869405511365772, "minimum": 0.3472777555673177, "maximum": 0.777469086946673, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10479971778659007, "standard_deviation": 0.006644025254160709, "minimum": 0.09747130879955733, "maximum": 0.11452056065147116, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956402430631242, "standard_deviation": 0.0008667318151931511, "minimum": 0.9943368707705482, "maximum": 0.9970610696843432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6613370518388328, "standard_deviation": 0.05586080323466219, "minimum": 0.5905902492746401, "maximum": 0.7738981646017477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5321020806687489, "standard_deviation": 0.14672248478473737, "minimum": 0.3154194605014304, "maximum": 0.7110394698853003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06056559598053937, "standard_deviation": 0.007369618688367109, "minimum": 0.049169354381971356, "maximum": 0.07013722773726597, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.643215550491393, "standard_deviation": 0.039921366229182305, "minimum": 0.5582681884525796, "maximum": 0.6973603940223235, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.007006904616794, "standard_deviation": 0.015529164624565816, "minimum": 5.9846951064974325, "maximum": 6.029910768437212, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5064911686128528, "standard_deviation": 0.02937864125558619, "minimum": 0.4714283739647763, "maximum": 0.5457506770145948, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0957850739676148, "standard_deviation": 0.04866269284828514, "minimum": -0.18019654254306808, "maximum": -0.03330338005883764, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9923205341861745, "standard_deviation": 0.001018774437429633, "minimum": 0.9903574810154991, "maximum": 0.9932092644268772, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8205818396732716, "standard_deviation": 0.0391548328764855, "minimum": 0.7765968866169247, "maximum": 0.9019592378318904, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.363568762933971, "standard_deviation": 0.12503391284657964, "minimum": 0.16511610250038047, "maximum": 0.4807644613521556, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07586095384050712, "standard_deviation": 0.005030020151649686, "minimum": 0.07152796219909721, "maximum": 0.08479614572545613, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997394458088581, "standard_deviation": 0.000545844954663666, "minimum": 0.9966625526726826, "maximum": 0.9981580615982151, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45896711992703054, "standard_deviation": 0.025508725873556702, "minimum": 0.41816405240323185, "maximum": 0.49180721791377235, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40065692515541, "standard_deviation": 0.11535079575250413, "minimum": 0.18455055811156154, "maximum": 0.538726040454966, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04203725665006121, "standard_deviation": 0.005301244052025105, "minimum": 0.03487575486881722, "maximum": 0.04930316277119051, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0003", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.4930763244628906, "predicted_mos": 2.4930763244628906, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "0642b5c921084279b0476874ee92b99d217cd0536d8c925128e2381c6ff54cc4", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.0582727999426425, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7809974601379254, "全帯域_10ms_rms_envelope_nrmse": 0.07238116594507885, "全帯域_10ms_envelope_correlation": 0.9929470365304177, "全帯域_10ms_modulation_mae_db": 0.24936720701848616, "全帯域_40ms_log_rms_mae_db": 0.4204183579760375, "全帯域_40ms_rms_envelope_nrmse": 0.0345255629031099, "全帯域_40ms_envelope_correlation": 0.9981987728610316, "全帯域_40ms_modulation_mae_db": 0.27704032236109155, "300–1000 Hz_10ms_log_rms_mae_db": 1.1169224456639635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09811173128615787, "300–1000 Hz_10ms_envelope_correlation": 0.9892815617017366, "300–1000 Hz_10ms_modulation_mae_db": 0.29600296154692723, "300–1000 Hz_40ms_log_rms_mae_db": 0.581567060083901, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046737842685567695, "300–1000 Hz_40ms_envelope_correlation": 0.9973272419881843, "300–1000 Hz_40ms_modulation_mae_db": 0.22801409581628299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.472993214232719, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1282959478270891, "1000–3000 Hz_10ms_envelope_correlation": 0.9873104105597321, "1000–3000 Hz_10ms_modulation_mae_db": 0.2515596648833632, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0078741143252783, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08179705224687712, "1000–3000 Hz_40ms_envelope_correlation": 0.9945731707275812, "1000–3000 Hz_40ms_modulation_mae_db": 0.2334641551555595, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0612757681978553, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5649318668920861, "3000–6000 Hz_10ms_envelope_correlation": 0.8080028399602868, "3000–6000 Hz_10ms_modulation_mae_db": 0.21015464815539278, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8814155591481767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4820269456410877, "3000–6000 Hz_40ms_envelope_correlation": 0.8439165451050638, "3000–6000 Hz_40ms_modulation_mae_db": 0.8118259003977227, "mr_stft_db": 5.8797184271167895, "band_mae_db": 0.5385196764702814, "rms_gain_error_db": -0.0644672806144368, "output_peak": 0.47542442966991927}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8057517615748979, "全帯域_10ms_rms_envelope_nrmse": 0.07392653979640054, "全帯域_10ms_envelope_correlation": 0.9926522872185374, "全帯域_10ms_modulation_mae_db": 0.38720985072907704, "全帯域_40ms_log_rms_mae_db": 0.46306041044674134, "全帯域_40ms_rms_envelope_nrmse": 0.04389231546507862, "全帯域_40ms_envelope_correlation": 0.997062289163242, "全帯域_40ms_modulation_mae_db": 0.4766593203464376, "300–1000 Hz_10ms_log_rms_mae_db": 1.087667182754276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09816694120001868, "300–1000 Hz_10ms_envelope_correlation": 0.9893069942166386, "300–1000 Hz_10ms_modulation_mae_db": 0.5803977737319937, "300–1000 Hz_40ms_log_rms_mae_db": 0.6257065355794017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0592284231167231, "300–1000 Hz_40ms_envelope_correlation": 0.9956585797483442, "300–1000 Hz_40ms_modulation_mae_db": 0.5739274428021904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5626577594735311, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1398175290871089, "1000–3000 Hz_10ms_envelope_correlation": 0.9850227366919412, "1000–3000 Hz_10ms_modulation_mae_db": 0.37171520060508123, "1000–3000 Hz_40ms_log_rms_mae_db": 1.093905881288613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0921129440426266, "1000–3000 Hz_40ms_envelope_correlation": 0.9933579262592708, "1000–3000 Hz_40ms_modulation_mae_db": 0.48379686696724683, "3000–6000 Hz_10ms_log_rms_mae_db": 2.058629401796386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5716408641448925, "3000–6000 Hz_10ms_envelope_correlation": 0.8026953711657853, "3000–6000 Hz_10ms_modulation_mae_db": 0.14042921938408434, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9065318625512295, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4940401631491815, "3000–6000 Hz_40ms_envelope_correlation": 0.8347826305585289, "3000–6000 Hz_40ms_modulation_mae_db": 0.18025268990140933, "mr_stft_db": 5.8833110864150635, "band_mae_db": 0.5542898767481937, "rms_gain_error_db": -0.07428171305878362, "output_peak": 0.4925346226713177}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8439081395436285, "全帯域_10ms_rms_envelope_nrmse": 0.07426161523606654, "全帯域_10ms_envelope_correlation": 0.992553637748693, "全帯域_10ms_modulation_mae_db": 0.3804333794322121, "全帯域_40ms_log_rms_mae_db": 0.47680101416460957, "全帯域_40ms_rms_envelope_nrmse": 0.0407880530651881, "全帯域_40ms_envelope_correlation": 0.9974188087273284, "全帯域_40ms_modulation_mae_db": 0.39990125253840064, "300–1000 Hz_10ms_log_rms_mae_db": 1.1966503356502083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09984042702211499, "300–1000 Hz_10ms_envelope_correlation": 0.9888506705980441, "300–1000 Hz_10ms_modulation_mae_db": 0.5160983830040997, "300–1000 Hz_40ms_log_rms_mae_db": 0.6780825977372635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05614826767436144, "300–1000 Hz_40ms_envelope_correlation": 0.9959709774889668, "300–1000 Hz_40ms_modulation_mae_db": 0.424909245305012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4463900610256724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.130147651184401, "1000–3000 Hz_10ms_envelope_correlation": 0.9877547195056533, "1000–3000 Hz_10ms_modulation_mae_db": 0.505749820207913, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0109041179197298, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09046198867252751, "1000–3000 Hz_40ms_envelope_correlation": 0.9945284310169522, "1000–3000 Hz_40ms_modulation_mae_db": 0.4252215730857367, "3000–6000 Hz_10ms_log_rms_mae_db": 2.08431311735789, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5572230494892123, "3000–6000 Hz_10ms_envelope_correlation": 0.805133661525394, "3000–6000 Hz_10ms_modulation_mae_db": 0.8740961926863505, "3000–6000 Hz_40ms_log_rms_mae_db": 1.907085411381386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4844975532320563, "3000–6000 Hz_40ms_envelope_correlation": 0.8339254886659688, "3000–6000 Hz_40ms_modulation_mae_db": 1.2453058006443778, "mr_stft_db": 5.86905997082604, "band_mae_db": 0.563372310658089, "rms_gain_error_db": -0.04616140632306135, "output_peak": 0.5414708299515537}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7972924136904438, "全帯域_10ms_rms_envelope_nrmse": 0.0724560050897026, "全帯域_10ms_envelope_correlation": 0.9929282806323324, "全帯域_10ms_modulation_mae_db": 0.1424864557280833, "全帯域_40ms_log_rms_mae_db": 0.42561316417509454, "全帯域_40ms_rms_envelope_nrmse": 0.03550974736814511, "全帯域_40ms_envelope_correlation": 0.9980757906474399, "全帯域_40ms_modulation_mae_db": 0.2678727205965326, "300–1000 Hz_10ms_log_rms_mae_db": 1.1847445935800076, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10254277506093305, "300–1000 Hz_10ms_envelope_correlation": 0.9882109095228772, "300–1000 Hz_10ms_modulation_mae_db": 0.3256020315687999, "300–1000 Hz_40ms_log_rms_mae_db": 0.6420178427317558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051265532134480915, "300–1000 Hz_40ms_envelope_correlation": 0.9966438535253334, "300–1000 Hz_40ms_modulation_mae_db": 0.5120685167935627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5009657798371303, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13014903171090672, "1000–3000 Hz_10ms_envelope_correlation": 0.9870555941851734, "1000–3000 Hz_10ms_modulation_mae_db": 0.19730524205731045, "1000–3000 Hz_40ms_log_rms_mae_db": 1.064093564430746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0781004708102333, "1000–3000 Hz_40ms_envelope_correlation": 0.9952162679916858, "1000–3000 Hz_40ms_modulation_mae_db": 0.31438717232961144, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1426609967707186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5858457837111041, "3000–6000 Hz_10ms_envelope_correlation": 0.7752067933724639, "3000–6000 Hz_10ms_modulation_mae_db": 1.4811806484354062, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0695014922078694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5080437576099616, "3000–6000 Hz_40ms_envelope_correlation": 0.8081742076351424, "3000–6000 Hz_40ms_modulation_mae_db": 1.8546151527461505, "mr_stft_db": 5.8462892831423225, "band_mae_db": 0.4496446302882553, "rms_gain_error_db": -0.0576450509545219, "output_peak": 0.5198176304305061}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8443055308777757, "全帯域_10ms_rms_envelope_nrmse": 0.08280071406239368, "全帯域_10ms_envelope_correlation": 0.9909412548299545, "全帯域_10ms_modulation_mae_db": 0.23328533346344416, "全帯域_40ms_log_rms_mae_db": 0.4660887140924183, "全帯域_40ms_rms_envelope_nrmse": 0.049645737864980714, "全帯域_40ms_envelope_correlation": 0.9964802229668019, "全帯域_40ms_modulation_mae_db": 0.2912413431815873, "300–1000 Hz_10ms_log_rms_mae_db": 1.2241621532006222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1108096867403369, "300–1000 Hz_10ms_envelope_correlation": 0.9866012341295824, "300–1000 Hz_10ms_modulation_mae_db": 0.4474038951478136, "300–1000 Hz_40ms_log_rms_mae_db": 0.6614038863479457, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07132953178659478, "300–1000 Hz_40ms_envelope_correlation": 0.9940038412845568, "300–1000 Hz_40ms_modulation_mae_db": 0.4672668019112369, "1000–3000 Hz_10ms_log_rms_mae_db": 1.410863516244714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1269236442430639, "1000–3000 Hz_10ms_envelope_correlation": 0.9887493001148886, "1000–3000 Hz_10ms_modulation_mae_db": 0.9702908863370333, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9791123370512281, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0868243119049255, "1000–3000 Hz_40ms_envelope_correlation": 0.9954655124861075, "1000–3000 Hz_40ms_modulation_mae_db": 0.9824572428495243, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1192197420199292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5813066476873139, "3000–6000 Hz_10ms_envelope_correlation": 0.7934954990432908, "3000–6000 Hz_10ms_modulation_mae_db": 0.9360968401011848, "3000–6000 Hz_40ms_log_rms_mae_db": 1.972349965731548, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4989128383103294, "3000–6000 Hz_40ms_envelope_correlation": 0.8292096992809668, "3000–6000 Hz_40ms_modulation_mae_db": 1.317324619161873, "mr_stft_db": 5.872179418934938, "band_mae_db": 0.5825747162444315, "rms_gain_error_db": -0.1386274601531439, "output_peak": 0.5373071727501876}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8002542025938963, "全帯域_10ms_rms_envelope_nrmse": 0.07534549438343417, "全帯域_10ms_envelope_correlation": 0.9927312827864353, "全帯域_10ms_modulation_mae_db": 0.3049705478205806, "全帯域_40ms_log_rms_mae_db": 0.4610641371147117, "全帯域_40ms_rms_envelope_nrmse": 0.04785269616204465, "全帯域_40ms_envelope_correlation": 0.9969556239587769, "全帯域_40ms_modulation_mae_db": 0.3922203617484648, "300–1000 Hz_10ms_log_rms_mae_db": 1.2078721423283905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11085882087009176, "300–1000 Hz_10ms_envelope_correlation": 0.986977158954206, "300–1000 Hz_10ms_modulation_mae_db": 0.5322405988696653, "300–1000 Hz_40ms_log_rms_mae_db": 0.6875548252049147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06598210536634275, "300–1000 Hz_40ms_envelope_correlation": 0.995516236675496, "300–1000 Hz_40ms_modulation_mae_db": 0.5919117263500383, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4459759670354768, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11776398173392379, "1000–3000 Hz_10ms_envelope_correlation": 0.9895288766761244, "1000–3000 Hz_10ms_modulation_mae_db": 0.2886541782003855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0276625386458251, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0769924900575917, "1000–3000 Hz_40ms_envelope_correlation": 0.9955530489076749, "1000–3000 Hz_40ms_modulation_mae_db": 0.33709569916852117, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0199411488460113, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5577389657511892, "3000–6000 Hz_10ms_envelope_correlation": 0.8028852211042941, "3000–6000 Hz_10ms_modulation_mae_db": 0.8743742836841006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.875640385145077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4861222659010811, "3000–6000 Hz_40ms_envelope_correlation": 0.8308200516564528, "3000–6000 Hz_40ms_modulation_mae_db": 1.36731655409008, "mr_stft_db": 5.851935245859419, "band_mae_db": 0.532721548419628, "rms_gain_error_db": -0.16794317854676985, "output_peak": 0.4934148263386739}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8324628120559717, "全帯域_10ms_rms_envelope_nrmse": 0.07110784813785, "全帯域_10ms_envelope_correlation": 0.9933077328059279, "全帯域_10ms_modulation_mae_db": 0.49529897120830607, "全帯域_40ms_log_rms_mae_db": 0.48762787388683515, "全帯域_40ms_rms_envelope_nrmse": 0.040165401872025705, "全帯域_40ms_envelope_correlation": 0.9976967919833792, "全帯域_40ms_modulation_mae_db": 0.3858054619190312, "300–1000 Hz_10ms_log_rms_mae_db": 1.3269691875823726, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11142988043354074, "300–1000 Hz_10ms_envelope_correlation": 0.9865114840291485, "300–1000 Hz_10ms_modulation_mae_db": 0.8469382407853571, "300–1000 Hz_40ms_log_rms_mae_db": 0.791778379274642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06845184371050285, "300–1000 Hz_40ms_envelope_correlation": 0.9946633180676873, "300–1000 Hz_40ms_modulation_mae_db": 0.6588012574930924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4598954959319297, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13131349677486756, "1000–3000 Hz_10ms_envelope_correlation": 0.986710300599672, "1000–3000 Hz_10ms_modulation_mae_db": 0.40554145017668874, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1062036470217353, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08296807490862587, "1000–3000 Hz_40ms_envelope_correlation": 0.994412536344787, "1000–3000 Hz_40ms_modulation_mae_db": 0.5172146831452237, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0727760786053966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5734363633989157, "3000–6000 Hz_10ms_envelope_correlation": 0.7973650221306399, "3000–6000 Hz_10ms_modulation_mae_db": 0.5282282384463418, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8956632899581805, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49655002477548943, "3000–6000 Hz_40ms_envelope_correlation": 0.8285400338393851, "3000–6000 Hz_40ms_modulation_mae_db": 1.0253174516362815, "mr_stft_db": 5.867955197189363, "band_mae_db": 0.544872676290043, "rms_gain_error_db": -0.11244762942542492, "output_peak": 0.4887118772688523}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8982162062982741, "全帯域_10ms_rms_envelope_nrmse": 0.08723705128475483, "全帯域_10ms_envelope_correlation": 0.9897473057511031, "全帯域_10ms_modulation_mae_db": 0.46229729978643386, "全帯域_40ms_log_rms_mae_db": 0.4857908236877004, "全帯域_40ms_rms_envelope_nrmse": 0.045229754607624205, "全帯域_40ms_envelope_correlation": 0.9969024641806002, "全帯域_40ms_modulation_mae_db": 0.5650422332768429, "300–1000 Hz_10ms_log_rms_mae_db": 1.245408309805364, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11883401303788098, "300–1000 Hz_10ms_envelope_correlation": 0.984199511275114, "300–1000 Hz_10ms_modulation_mae_db": 0.5745269112265495, "300–1000 Hz_40ms_log_rms_mae_db": 0.6615403874318246, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06388889994310712, "300–1000 Hz_40ms_envelope_correlation": 0.9948769188965099, "300–1000 Hz_40ms_modulation_mae_db": 0.7263076475337967, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5329865158504394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1548617539838866, "1000–3000 Hz_10ms_envelope_correlation": 0.9824393672574431, "1000–3000 Hz_10ms_modulation_mae_db": 0.9129774380931878, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1014534537399674, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11254908532755488, "1000–3000 Hz_40ms_envelope_correlation": 0.9910987574071055, "1000–3000 Hz_40ms_modulation_mae_db": 0.8840404083694228, "3000–6000 Hz_10ms_log_rms_mae_db": 2.153463183438024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5756024873967963, "3000–6000 Hz_10ms_envelope_correlation": 0.7964098956566638, "3000–6000 Hz_10ms_modulation_mae_db": 0.8051386899601582, "3000–6000 Hz_40ms_log_rms_mae_db": 1.934195666162177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49375170218288855, "3000–6000 Hz_40ms_envelope_correlation": 0.8315014747603229, "3000–6000 Hz_40ms_modulation_mae_db": 1.1281069625804836, "mr_stft_db": 5.852502219891984, "band_mae_db": 0.5525292202830668, "rms_gain_error_db": -0.08106217911737394, "output_peak": 0.4999742066982459}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59200, "output_samples": 59200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5814147769429665, "spectral_convergence": 0.2067489232853595, "spectral_convergence_status": "ok", "amplitude_floor": 2.0296349328313604e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.560190838937529, "spectral_convergence": 0.20711467780983114, "spectral_convergence_status": "ok", "amplitude_floor": 1.6850709272784723e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.497549665469876, "spectral_convergence": 0.2162171621262655, "spectral_convergence_status": "ok", "amplitude_floor": 8.273401155700601e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185911287346583, "output_power": 0.00079876064152772, "delta_db": -0.10650376668432675, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003085801603155809, "output_power": 0.0030189959701996654, "delta_db": -0.09505467011492197, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013392891870332754, "output_power": 0.0013437614310720714, "delta_db": 0.014478092164685895, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027936558999729056, "output_power": 0.00027459234288865013, "delta_db": -0.0748448957317147, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.040755904577448e-06, "output_power": 9.037680922281483e-06, "delta_db": 1.084377163267372, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.309787748977687e-06, "output_power": 9.673880038980827e-06, "delta_db": 1.855859470858667, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5385196764702814, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0644672806144368, "pitch": null, "output_rms_dbfs": -22.63219420850102, "output_peak": 0.47542442966991927, "reference_rms_dbfs": -22.567726927886582, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8797184271167895}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002672933594000982, "log_rms_mae_db": 0.7809974601379254, "rms_envelope_nrmse": 0.07238116594507885, "envelope_correlation": 0.9929470365304177, "rms_gain_error_db": -0.06446735721138704, "reference_mean_power": 0.0055438895852693145, "output_mean_power": 0.005462202998952125}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.3303693041440347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007860073564586987, "output_power": 0.0007548869031298526, "reference_active": true, "delta_db": -0.17544720071902717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024364364087793404, "output_power": 0.00023978377086250724, "reference_active": true, "delta_db": -0.06935295066995326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.963914121364001e-05, "output_power": 4.4207312375672616e-05, "reference_active": true, "delta_db": -0.5033014696664782}], "active_band_count": 3, "active_band_mae_db": 0.24936720701848616}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00019127079187884833, "log_rms_mae_db": 0.4204183579760375, "rms_envelope_nrmse": 0.0345255629031099, "envelope_correlation": 0.9981987728610316, "rms_gain_error_db": -0.06446749051503173, "reference_mean_power": 0.005589268172913711, "output_mean_power": 0.005506912785577349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.0930617045795464e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007030365316665854, "output_power": 0.0006766447881970815, "reference_active": true, "delta_db": -0.16617151666583913}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013359627830049334, "output_power": 0.000130882910683151, "reference_active": true, "delta_db": -0.08911415178673616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7787329335262405e-06, "output_power": 2.4336789095641656e-06, "reference_active": true, "delta_db": -0.5758352986306994}], "active_band_count": 3, "active_band_mae_db": 0.27704032236109155}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002516933783769556, "log_rms_mae_db": 1.1169224456639635, "rms_envelope_nrmse": 0.09811173128615787, "envelope_correlation": 0.9892815617017366, "rms_gain_error_db": -0.11067866741313057, "reference_mean_power": 0.003609667084718048, "output_mean_power": 0.003518838075122242}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.0389022511376814e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005770551731550911, "output_power": 0.0005457914714966898, "reference_active": true, "delta_db": -0.24190593592643989}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021321841152816334, "output_power": 0.00020451263657137115, "reference_active": true, "delta_db": -0.1810455587639066}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0182675609825566e-05, "output_power": 4.5086661027817094e-05, "reference_active": true, "delta_db": -0.4650573899504352}], "active_band_count": 3, "active_band_mae_db": 0.29600296154692723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00017703062295034008, "log_rms_mae_db": 0.581567060083901, "rms_envelope_nrmse": 0.046737842685567695, "envelope_correlation": 0.9973272419881843, "rms_gain_error_db": -0.11067871073905622, "reference_mean_power": 0.003639213964400563, "output_mean_power": 0.0035476414398190213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 8.593451292498866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005240710435115444, "output_power": 0.000496067013301953, "reference_active": true, "delta_db": -0.23849815347217607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012044100626957096, "output_power": 0.00011383501474207132, "reference_active": true, "delta_db": -0.2449850720713584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.706489326091672e-06, "output_power": 2.5843445396796794e-06, "reference_active": true, "delta_db": -0.2005590619053145}], "active_band_count": 3, "active_band_mae_db": 0.22801409581628299}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 7.943457135434651e-05, "log_rms_mae_db": 1.472993214232719, "rms_envelope_nrmse": 0.1282959478270891, "envelope_correlation": 0.9873104105597321, "rms_gain_error_db": -0.08426672089414951, "reference_mean_power": 0.00024301128670702193, "output_mean_power": 0.00023834155741794257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.5901124046586967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.178189042095757e-05, "output_power": 9.162458670537495e-05, "reference_active": true, "delta_db": -0.007449698994517379}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7997292458206206e-05, "output_power": 1.5907283322152003e-05, "reference_active": true, "delta_db": -0.5361115775901707}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3356967361336817e-06, "output_power": 2.2248702151102114e-06, "reference_active": true, "delta_db": -0.21111771806540147}], "active_band_count": 3, "active_band_mae_db": 0.2515596648833632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 6.195280015885064e-05, "log_rms_mae_db": 1.0078741143252783, "rms_envelope_nrmse": 0.08179705224687712, "envelope_correlation": 0.9945731707275812, "rms_gain_error_db": -0.08426729097638938, "reference_mean_power": 0.00024500023514433585, "output_mean_power": 0.00024029225448007655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.4888167929489627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.774520748189054e-05, "output_power": 8.774800490317323e-05, "reference_active": true, "delta_db": 0.00013845602462084496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.213495209539377e-06, "output_power": 8.118887914275577e-06, "reference_active": true, "delta_db": -0.5492786860037205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5741557318464897e-07, "output_power": 1.5203730499843883e-07, "reference_active": true, "delta_db": -0.15097532343833703}], "active_band_count": 3, "active_band_mae_db": 0.2334641551555595}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 1.5353574631917574e-05, "log_rms_mae_db": 2.0612757681978553, "rms_envelope_nrmse": 0.5649318668920861, "envelope_correlation": 0.8080028399602868, "rms_gain_error_db": 1.1841440331870636, "reference_mean_power": 7.123238480443002e-06, "output_mean_power": 9.35603607613778e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 4.409187309389394e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6167395840190636e-06, "output_power": 2.5847420422749616e-06, "reference_active": true, "delta_db": -0.05343297201442947}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.009142614437233e-07, "output_power": 7.007014962793415e-07, "reference_active": true, "delta_db": -0.0013185173901315993}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.642480310119981e-07, "output_power": 2.3144111580624893e-07, "reference_active": true, "delta_db": -0.5757124550616173}], "active_band_count": 3, "active_band_mae_db": 0.21015464815539278}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 1.282880604964887e-05, "log_rms_mae_db": 1.8814155591481767, "rms_envelope_nrmse": 0.4820269456410877, "envelope_correlation": 0.8439165451050638, "rms_gain_error_db": 1.184156150182444, "reference_mean_power": 7.1814170088864185e-06, "output_mean_power": 9.432477132334162e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 3.966042949113477e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6674623865712373e-06, "output_power": 2.5831519586910795e-06, "reference_active": true, "delta_db": -0.1394834912695944}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.255436860296439e-07, "output_power": 3.55702263241419e-07, "reference_active": true, "delta_db": 0.38477351942267113}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7820276752795505e-08, "output_power": 1.7915930825446273e-08, "reference_active": true, "delta_db": -1.9112206905009024}], "active_band_count": 3, "active_band_mae_db": 0.8118259003977227}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9174371722200022, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1a078a0b4a8c45ab5b7e845c2f651a3cdf2fbac706750b7e99713b07a1404597", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9868214131988285, "standard_deviation": 0.002225245927861021, "minimum": 0.9824393672574431, "maximum": 0.9895288766761244, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4790910387039515, "standard_deviation": 0.05013301255647796, "minimum": 1.410863516244714, "maximum": 1.5626577594735311, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4879742350701204, "standard_deviation": 0.29616432957359545, "minimum": 0.19730524205731045, "maximum": 0.9702908863370333, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13240912956815593, "standard_deviation": 0.010901616085843877, "minimum": 0.11776398173392379, "maximum": 0.1548617539838866, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942757063926456, "standard_deviation": 0.0014657178150928513, "minimum": 0.9910987574071055, "maximum": 0.9955530489076749, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0489012068028902, "standard_deviation": 0.0489169926889129, "minimum": 0.9791123370512281, "maximum": 1.1062036470217353, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5222097251338558, "standard_deviation": 0.271105630677184, "minimum": 0.2334641551555595, "maximum": 0.9824572428495243, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0877258022463703, "standard_deviation": 0.011397862198284441, "minimum": 0.0769924900575917, "maximum": 0.11254908532755488, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.7976492879948524, "standard_deviation": 0.010272197318975072, "minimum": 0.7752067933724639, "maximum": 0.8080028399602868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0890349296290265, "standard_deviation": 0.04581568870688439, "minimum": 2.0199411488460113, "maximum": 2.153463183438024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7312123451066274, "standard_deviation": 0.43301842983776895, "minimum": 0.14042921938408434, "maximum": 1.4811806484354062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5709657535589387, "standard_deviation": 0.010396606560653294, "minimum": 0.5572230494892123, "maximum": 0.5858457837111041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8301087664377289, "standard_deviation": 0.010098845781496505, "minimum": 0.8081742076351424, "maximum": 0.8439165451050638, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.9302979540357055, "standard_deviation": 0.06418871335011365, "minimum": 1.875640385145077, "maximum": 2.0695014922078694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1162581413947974, "standard_deviation": 0.4843962704582918, "minimum": 0.18025268990140933, "maximum": 1.8546151527461505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.4929931563502594, "standard_deviation": 0.008581777588281351, "minimum": 0.4820269456410877, "maximum": 0.5080437576099616, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9874924405534184, "standard_deviation": 0.0017614916140694993, "minimum": 0.984199511275114, "maximum": 0.9893069942166386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1987995438206505, "standard_deviation": 0.07424075723871223, "minimum": 1.087667182754276, "maximum": 1.3269691875823726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5149013494851508, "standard_deviation": 0.17183971401630493, "minimum": 0.29600296154692723, "maximum": 0.8469382407853571, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10632428445638437, "standard_deviation": 0.007686442356630113, "minimum": 0.09811173128615787, "maximum": 0.11883401303788098, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955826209593848, "standard_deviation": 0.001080118576022056, "minimum": 0.9940038412845568, "maximum": 0.9973272419881843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6662064392989562, "standard_deviation": 0.060709025338984986, "minimum": 0.581567060083901, "maximum": 0.791778379274642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5229008417506515, "standard_deviation": 0.15444948291107477, "minimum": 0.22801409581628299, "maximum": 0.7263076475337967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060379055802210084, "standard_deviation": 0.008595335378680298, "minimum": 0.046737842685567695, "maximum": 0.07132953178659478, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.5398155819252486, "standard_deviation": 0.03957413342088126, "minimum": 0.4496446302882553, "maximum": 0.5825747162444315, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.86536885617199, "standard_deviation": 0.013651652893002846, "minimum": 5.8462892831423225, "maximum": 5.8833110864150635, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.506081949472407, "standard_deviation": 0.024020675273694312, "minimum": 0.47542442966991927, "maximum": 0.5414708299515537, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09282948727418953, "standard_deviation": 0.04280712726519045, "minimum": -0.16794317854676985, "maximum": -0.04616140632306135, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9922261022879253, "standard_deviation": 0.0012260337246152906, "minimum": 0.9897473057511031, "maximum": 0.9933077328059279, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8253985658466017, "standard_deviation": 0.03747768837063333, "minimum": 0.7809974601379254, "maximum": 0.8982162062982741, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33191863064832794, "standard_deviation": 0.12086388201803114, "minimum": 0.1424864557280833, "maximum": 0.49529897120830607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07618955424196015, "standard_deviation": 0.0057267816059857896, "minimum": 0.07110784813785, "maximum": 0.08723705128475483, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973488455610751, "standard_deviation": 0.000605603179482746, "minimum": 0.9964802229668019, "maximum": 0.9981987728610316, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4608080619430186, "standard_deviation": 0.025360960348786498, "minimum": 0.4204183579760375, "maximum": 0.48762787388683515, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38197287699604854, "standard_deviation": 0.10363599485346936, "minimum": 0.2678727205965326, "maximum": 0.5650422332768429, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04220115866352463, "standard_deviation": 0.005467109040844717, "minimum": 0.0345255629031099, "maximum": 0.049645737864980714, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0003", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.2992329597473145, "predicted_mos": 2.2992329597473145, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "2983295f81df7292a6dd35c1fabf3b05a8b2a2d8b5fd763bce57003ea9082f2b", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.881620700005442, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8080695232865087, "全帯域_10ms_rms_envelope_nrmse": 0.07253690695679355, "全帯域_10ms_envelope_correlation": 0.992937622386385, "全帯域_10ms_modulation_mae_db": 0.57383788527206, "全帯域_40ms_log_rms_mae_db": 0.47613605852063773, "全帯域_40ms_rms_envelope_nrmse": 0.04258612522856553, "全帯域_40ms_envelope_correlation": 0.99724544284693, "全帯域_40ms_modulation_mae_db": 0.5422135016277002, "300–1000 Hz_10ms_log_rms_mae_db": 1.1498795177977792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09370565062316487, "300–1000 Hz_10ms_envelope_correlation": 0.9904309727409658, "300–1000 Hz_10ms_modulation_mae_db": 0.7290001867030637, "300–1000 Hz_40ms_log_rms_mae_db": 0.6582573007425482, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053262351404961654, "300–1000 Hz_40ms_envelope_correlation": 0.9967411092487602, "300–1000 Hz_40ms_modulation_mae_db": 0.6777174688670756, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7641734377682097, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14089662904615607, "1000–3000 Hz_10ms_envelope_correlation": 0.9847823941601684, "1000–3000 Hz_10ms_modulation_mae_db": 0.4206368275825387, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1263294672112591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08325601729941452, "1000–3000 Hz_40ms_envelope_correlation": 0.9947408072783779, "1000–3000 Hz_40ms_modulation_mae_db": 0.21297122475576447, "3000–6000 Hz_10ms_log_rms_mae_db": 3.097535650196166, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6545923605024206, "3000–6000 Hz_10ms_envelope_correlation": 0.7666942280226705, "3000–6000 Hz_10ms_modulation_mae_db": 0.6418404168382207, "3000–6000 Hz_40ms_log_rms_mae_db": 2.6262890777897647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5413649394171971, "3000–6000 Hz_40ms_envelope_correlation": 0.8289002989636107, "3000–6000 Hz_40ms_modulation_mae_db": 0.1409921424737175, "mr_stft_db": 6.428331002197351, "band_mae_db": 0.7144438187847729, "rms_gain_error_db": -0.07356003204547859, "output_peak": 0.4738975457765095}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.784068921025318, "全帯域_10ms_rms_envelope_nrmse": 0.07555404790538578, "全帯域_10ms_envelope_correlation": 0.9923248110370747, "全帯域_10ms_modulation_mae_db": 0.24853226585814825, "全帯域_40ms_log_rms_mae_db": 0.43022758047545195, "全帯域_40ms_rms_envelope_nrmse": 0.04640638025921522, "全帯域_40ms_envelope_correlation": 0.9967086565110438, "全帯域_40ms_modulation_mae_db": 0.11903221521661754, "300–1000 Hz_10ms_log_rms_mae_db": 1.269725323266559, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.104557689540347, "300–1000 Hz_10ms_envelope_correlation": 0.9879039686796934, "300–1000 Hz_10ms_modulation_mae_db": 0.4312779428050497, "300–1000 Hz_40ms_log_rms_mae_db": 0.680818953831422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06318108743182742, "300–1000 Hz_40ms_envelope_correlation": 0.9951316957576463, "300–1000 Hz_40ms_modulation_mae_db": 0.3192482626936916, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7899881380830431, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13247139894687643, "1000–3000 Hz_10ms_envelope_correlation": 0.9869412549987614, "1000–3000 Hz_10ms_modulation_mae_db": 0.6397289190077925, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3165227269662658, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08635813938185194, "1000–3000 Hz_40ms_envelope_correlation": 0.99487976455211, "1000–3000 Hz_40ms_modulation_mae_db": 0.7862394071409651, "3000–6000 Hz_10ms_log_rms_mae_db": 2.953158806511072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6531658215475509, "3000–6000 Hz_10ms_envelope_correlation": 0.7717565005433066, "3000–6000 Hz_10ms_modulation_mae_db": 0.10038988445767227, "3000–6000 Hz_40ms_log_rms_mae_db": 2.589897196707363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5425794183983343, "3000–6000 Hz_40ms_envelope_correlation": 0.8315955716734056, "3000–6000 Hz_40ms_modulation_mae_db": 0.5764782022399867, "mr_stft_db": 6.42398196537967, "band_mae_db": 0.7981950628446599, "rms_gain_error_db": -0.07645276604828655, "output_peak": 0.5120627479207538}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7718839747221883, "全帯域_10ms_rms_envelope_nrmse": 0.07443938979394396, "全帯域_10ms_envelope_correlation": 0.9925996118659863, "全帯域_10ms_modulation_mae_db": 0.23889016400391572, "全帯域_40ms_log_rms_mae_db": 0.441958691369726, "全帯域_40ms_rms_envelope_nrmse": 0.04641438198987176, "全帯域_40ms_envelope_correlation": 0.9967714812348246, "全帯域_40ms_modulation_mae_db": 0.33474826021636317, "300–1000 Hz_10ms_log_rms_mae_db": 1.1050879248160883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09597761337161266, "300–1000 Hz_10ms_envelope_correlation": 0.9897963119350738, "300–1000 Hz_10ms_modulation_mae_db": 0.5932283143134334, "300–1000 Hz_40ms_log_rms_mae_db": 0.6302348732271656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06027911612482628, "300–1000 Hz_40ms_envelope_correlation": 0.9954977609558863, "300–1000 Hz_40ms_modulation_mae_db": 0.6319527845278455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7926829976310439, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14611116740156951, "1000–3000 Hz_10ms_envelope_correlation": 0.9835459456482506, "1000–3000 Hz_10ms_modulation_mae_db": 0.3139683601176249, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2419739182078762, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10105198216876216, "1000–3000 Hz_40ms_envelope_correlation": 0.9916861740682043, "1000–3000 Hz_40ms_modulation_mae_db": 0.3886851158117124, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9435673183654822, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6343073495638704, "3000–6000 Hz_10ms_envelope_correlation": 0.7907287936798566, "3000–6000 Hz_10ms_modulation_mae_db": 0.29721205303561943, "3000–6000 Hz_40ms_log_rms_mae_db": 2.523658536403992, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.525715162049897, "3000–6000 Hz_40ms_envelope_correlation": 0.8482039842759884, "3000–6000 Hz_40ms_modulation_mae_db": 0.5064780134878953, "mr_stft_db": 6.429699107834087, "band_mae_db": 0.7493989610080058, "rms_gain_error_db": -0.09749770252464077, "output_peak": 0.5226357546619675}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8281810229007294, "全帯域_10ms_rms_envelope_nrmse": 0.06944492984604571, "全帯域_10ms_envelope_correlation": 0.9935172528081377, "全帯域_10ms_modulation_mae_db": 0.40733092896591305, "全帯域_40ms_log_rms_mae_db": 0.4316076755739116, "全帯域_40ms_rms_envelope_nrmse": 0.036317294681508905, "全帯域_40ms_envelope_correlation": 0.9980014708617985, "全帯域_40ms_modulation_mae_db": 0.5079968793772186, "300–1000 Hz_10ms_log_rms_mae_db": 1.1514354704071252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09984010308872722, "300–1000 Hz_10ms_envelope_correlation": 0.988961529412655, "300–1000 Hz_10ms_modulation_mae_db": 0.6403787477582831, "300–1000 Hz_40ms_log_rms_mae_db": 0.611944657027556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053128786204030054, "300–1000 Hz_40ms_envelope_correlation": 0.9966142302016965, "300–1000 Hz_40ms_modulation_mae_db": 0.6877664242346925, "1000–3000 Hz_10ms_log_rms_mae_db": 1.79969653830972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13073361857961308, "1000–3000 Hz_10ms_envelope_correlation": 0.9868780598359072, "1000–3000 Hz_10ms_modulation_mae_db": 0.3359800733193381, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2570165514494496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07733542415193477, "1000–3000 Hz_40ms_envelope_correlation": 0.9953657541539553, "1000–3000 Hz_40ms_modulation_mae_db": 0.38913371374789857, "3000–6000 Hz_10ms_log_rms_mae_db": 2.928909631521083, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6101458816644222, "3000–6000 Hz_10ms_envelope_correlation": 0.7869613865963574, "3000–6000 Hz_10ms_modulation_mae_db": 0.30671184398361867, "3000–6000 Hz_40ms_log_rms_mae_db": 2.4058749246997166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5045140223987843, "3000–6000 Hz_40ms_envelope_correlation": 0.8422663399931635, "3000–6000 Hz_40ms_modulation_mae_db": 0.6713730249504123, "mr_stft_db": 6.422144220727059, "band_mae_db": 0.7215196518809951, "rms_gain_error_db": -0.06341001348639566, "output_peak": 0.5626773229958485}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8127469418029378, "全帯域_10ms_rms_envelope_nrmse": 0.0706088307468978, "全帯域_10ms_envelope_correlation": 0.9932969145101718, "全帯域_10ms_modulation_mae_db": 0.3969466614677876, "全帯域_40ms_log_rms_mae_db": 0.4409763025029898, "全帯域_40ms_rms_envelope_nrmse": 0.03632032113609372, "全帯域_40ms_envelope_correlation": 0.9980090547811514, "全帯域_40ms_modulation_mae_db": 0.5935412360872214, "300–1000 Hz_10ms_log_rms_mae_db": 1.206044095719421, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09761872080386014, "300–1000 Hz_10ms_envelope_correlation": 0.9895127215349919, "300–1000 Hz_10ms_modulation_mae_db": 0.6047896555850181, "300–1000 Hz_40ms_log_rms_mae_db": 0.6450125382670905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05407347192414424, "300–1000 Hz_40ms_envelope_correlation": 0.9965458671635332, "300–1000 Hz_40ms_modulation_mae_db": 0.6644791943634473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8128009571369135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13609695991444662, "1000–3000 Hz_10ms_envelope_correlation": 0.9861373188373674, "1000–3000 Hz_10ms_modulation_mae_db": 0.4195471049924209, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2661617591445533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0836781761783957, "1000–3000 Hz_40ms_envelope_correlation": 0.9951845562803834, "1000–3000 Hz_40ms_modulation_mae_db": 0.45549744234767164, "3000–6000 Hz_10ms_log_rms_mae_db": 3.0904356800241484, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6702296994539516, "3000–6000 Hz_10ms_envelope_correlation": 0.7657402858495738, "3000–6000 Hz_10ms_modulation_mae_db": 0.9478156255829537, "3000–6000 Hz_40ms_log_rms_mae_db": 2.618574090290689, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5507481430558281, "3000–6000 Hz_40ms_envelope_correlation": 0.8327217127504688, "3000–6000 Hz_40ms_modulation_mae_db": 0.5000718734499395, "mr_stft_db": 6.451014053867342, "band_mae_db": 0.7539803817584074, "rms_gain_error_db": -0.06629952369610516, "output_peak": 0.5072285891927973}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.804982963818945, "全帯域_10ms_rms_envelope_nrmse": 0.07861568039809524, "全帯域_10ms_envelope_correlation": 0.991679649508346, "全帯域_10ms_modulation_mae_db": 0.5667681449009921, "全帯域_40ms_log_rms_mae_db": 0.4719280046129302, "全帯域_40ms_rms_envelope_nrmse": 0.0425798091118311, "全帯域_40ms_envelope_correlation": 0.9972460900689396, "全帯域_40ms_modulation_mae_db": 0.7559997492079952, "300–1000 Hz_10ms_log_rms_mae_db": 1.214979450560383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11044716937972478, "300–1000 Hz_10ms_envelope_correlation": 0.9864628669974629, "300–1000 Hz_10ms_modulation_mae_db": 0.9418588354493256, "300–1000 Hz_40ms_log_rms_mae_db": 0.6734258771710813, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06462324349307981, "300–1000 Hz_40ms_envelope_correlation": 0.9948900094400708, "300–1000 Hz_40ms_modulation_mae_db": 1.220753008536141, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7468716950501066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13555165690572846, "1000–3000 Hz_10ms_envelope_correlation": 0.9859532225972283, "1000–3000 Hz_10ms_modulation_mae_db": 0.5183237820803949, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1981207135497118, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07819611393507381, "1000–3000 Hz_40ms_envelope_correlation": 0.9953914084410792, "1000–3000 Hz_40ms_modulation_mae_db": 0.6016084811388864, "3000–6000 Hz_10ms_log_rms_mae_db": 2.930190113818348, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6271076993758857, "3000–6000 Hz_10ms_envelope_correlation": 0.7825947846448101, "3000–6000 Hz_10ms_modulation_mae_db": 0.4554208427429089, "3000–6000 Hz_40ms_log_rms_mae_db": 2.489148718342413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.525533451375488, "3000–6000 Hz_40ms_envelope_correlation": 0.8346352047922359, "3000–6000 Hz_40ms_modulation_mae_db": 0.33098945036896, "mr_stft_db": 6.43592734852639, "band_mae_db": 0.7210789096753206, "rms_gain_error_db": -0.07601898714968208, "output_peak": 0.54487211478611}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8024523823161757, "全帯域_10ms_rms_envelope_nrmse": 0.07304691808888013, "全帯域_10ms_envelope_correlation": 0.9928641407123644, "全帯域_10ms_modulation_mae_db": 0.41448301300347534, "全帯域_40ms_log_rms_mae_db": 0.4574479696356079, "全帯域_40ms_rms_envelope_nrmse": 0.04411055587187361, "全帯域_40ms_envelope_correlation": 0.9970805356679204, "全帯域_40ms_modulation_mae_db": 0.47846166842395355, "300–1000 Hz_10ms_log_rms_mae_db": 1.2486749392850014, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10535910491682529, "300–1000 Hz_10ms_envelope_correlation": 0.9877068501337802, "300–1000 Hz_10ms_modulation_mae_db": 0.6996207432142492, "300–1000 Hz_40ms_log_rms_mae_db": 0.6656870873728394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06476241698266114, "300–1000 Hz_40ms_envelope_correlation": 0.9948515850485365, "300–1000 Hz_40ms_modulation_mae_db": 0.7363097838759419, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8314253429300136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13574731233521395, "1000–3000 Hz_10ms_envelope_correlation": 0.9858525610489194, "1000–3000 Hz_10ms_modulation_mae_db": 0.4043844298766886, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2761764362822599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08185970152591397, "1000–3000 Hz_40ms_envelope_correlation": 0.994773500974493, "1000–3000 Hz_40ms_modulation_mae_db": 0.4038271744286634, "3000–6000 Hz_10ms_log_rms_mae_db": 2.967171837110877, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6261502564804122, "3000–6000 Hz_10ms_envelope_correlation": 0.7815289335931043, "3000–6000 Hz_10ms_modulation_mae_db": 0.3974739085217394, "3000–6000 Hz_40ms_log_rms_mae_db": 2.5857450485129494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5229180949699844, "3000–6000 Hz_40ms_envelope_correlation": 0.8352782098915672, "3000–6000 Hz_40ms_modulation_mae_db": 0.8446937331702515, "mr_stft_db": 6.419959478750261, "band_mae_db": 0.7304372350684235, "rms_gain_error_db": -0.09073460092849615, "output_peak": 0.5248677357860683}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.829860131160505, "全帯域_10ms_rms_envelope_nrmse": 0.0696485948519619, "全帯域_10ms_envelope_correlation": 0.9935349785501728, "全帯域_10ms_modulation_mae_db": 0.26806302280352934, "全帯域_40ms_log_rms_mae_db": 0.424061035266538, "全帯域_40ms_rms_envelope_nrmse": 0.032727980462905135, "全帯域_40ms_envelope_correlation": 0.9984820279486301, "全帯域_40ms_modulation_mae_db": 0.3364615819679447, "300–1000 Hz_10ms_log_rms_mae_db": 1.203666506908597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09195955965090426, "300–1000 Hz_10ms_envelope_correlation": 0.9909257365854459, "300–1000 Hz_10ms_modulation_mae_db": 0.43332625758504734, "300–1000 Hz_40ms_log_rms_mae_db": 0.6442328863260145, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04731646538085978, "300–1000 Hz_40ms_envelope_correlation": 0.9977241435761369, "300–1000 Hz_40ms_modulation_mae_db": 0.33777280084427375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7229700559806318, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12224393817621701, "1000–3000 Hz_10ms_envelope_correlation": 0.9886530782678747, "1000–3000 Hz_10ms_modulation_mae_db": 0.565881602846617, "1000–3000 Hz_40ms_log_rms_mae_db": 1.175398631943208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08402202006901391, "1000–3000 Hz_40ms_envelope_correlation": 0.9945711281240395, "1000–3000 Hz_40ms_modulation_mae_db": 0.6357153493428845, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9521152202987575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6447879799241488, "3000–6000 Hz_10ms_envelope_correlation": 0.779277548359557, "3000–6000 Hz_10ms_modulation_mae_db": 0.4545654296371142, "3000–6000 Hz_40ms_log_rms_mae_db": 2.5081412800870133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.528920104320643, "3000–6000 Hz_40ms_envelope_correlation": 0.8425064113485772, "3000–6000 Hz_40ms_modulation_mae_db": 1.0876757808507642, "mr_stft_db": 6.428342416132352, "band_mae_db": 0.7487242188373813, "rms_gain_error_db": -0.09082121779085667, "output_peak": 0.4888020552463003}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59200, "output_samples": 59200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.152589761099605, "spectral_convergence": 0.21389743622431504, "spectral_convergence_status": "ok", "amplitude_floor": 2.0296349328313604e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.2006650454649135, "spectral_convergence": 0.2130281256895258, "spectral_convergence_status": "ok", "amplitude_floor": 1.6850709272784723e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.931738200027535, "spectral_convergence": 0.22608925523977833, "spectral_convergence_status": "ok", "amplitude_floor": 8.273401155700601e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185911287346583, "output_power": 0.0008166602628874822, "delta_db": -0.01025609611755619, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003085801603155809, "output_power": 0.002968206385570689, "delta_db": -0.16873905275758008, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013392891870332754, "output_power": 0.001361607337729351, "delta_db": 0.07177520665474113, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027936558999729056, "output_power": 0.0002757997560944668, "delta_db": -0.05579034408354625, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.040755904577448e-06, "output_power": 9.771446042819708e-06, "delta_db": 1.4233955016971227, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.309787748977687e-06, "output_power": 1.136803551401574e-05, "delta_db": 2.556706711398091, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7144438187847729, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07356003204547859, "pitch": null, "output_rms_dbfs": -22.64128695993206, "output_peak": 0.4738975457765095, "reference_rms_dbfs": -22.567726927886582, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.428331002197351}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002672933594000982, "log_rms_mae_db": 0.8080695232865087, "rms_envelope_nrmse": 0.07253690695679355, "envelope_correlation": 0.992937622386385, "rms_gain_error_db": -0.07355999962425626, "reference_mean_power": 0.0055438895852693145, "output_mean_power": 0.005450778975548574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.3303693041440347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007860073564586987, "output_power": 0.0007808510938419561, "reference_active": true, "delta_db": -0.028583877311813437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024364364087793404, "output_power": 0.00022411004728503124, "reference_active": true, "delta_db": -0.36293753508643767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.963914121364001e-05, "output_power": 3.654475334459169e-05, "reference_active": true, "delta_db": -1.329992243417929}], "active_band_count": 3, "active_band_mae_db": 0.57383788527206}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00019127079187884833, "log_rms_mae_db": 0.47613605852063773, "rms_envelope_nrmse": 0.04258612522856553, "envelope_correlation": 0.99724544284693, "rms_gain_error_db": -0.07356002882198477, "reference_mean_power": 0.005589268172913711, "output_mean_power": 0.005495395384812317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.0930617045795464e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007030365316665854, "output_power": 0.0007022498040602205, "reference_active": true, "delta_db": -0.004862660159909459}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013359627830049334, "output_power": 0.00011764403847046141, "reference_active": true, "delta_db": -0.5522443531575199}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7787329335262405e-06, "output_power": 2.1721682387994737e-06, "reference_active": true, "delta_db": -1.0695334915656713}], "active_band_count": 3, "active_band_mae_db": 0.5422135016277002}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002516933783769556, "log_rms_mae_db": 1.1498795177977792, "rms_envelope_nrmse": 0.09370565062316487, "envelope_correlation": 0.9904309727409658, "rms_gain_error_db": -0.1716371496850213, "reference_mean_power": 0.003609667084718048, "output_mean_power": 0.00346979194367353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.0389022511376814e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005770551731550911, "output_power": 0.0005764191662555975, "reference_active": true, "delta_db": -0.004789258078086191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021321841152816334, "output_power": 0.00018523737554652803, "reference_active": true, "delta_db": -0.6109608428630411}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0182675609825566e-05, "output_power": 3.494851824286081e-05, "reference_active": true, "delta_db": -1.5712504591680636}], "active_band_count": 3, "active_band_mae_db": 0.7290001867030637}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00017703062295034008, "log_rms_mae_db": 0.6582573007425482, "rms_envelope_nrmse": 0.053262351404961654, "envelope_correlation": 0.9967411092487602, "rms_gain_error_db": -0.1716371750527917, "reference_mean_power": 0.003639213964400563, "output_mean_power": 0.003498193856902353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 8.593451292498866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005240710435115444, "output_power": 0.000526823011913863, "reference_active": true, "delta_db": 0.02274572621868909}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012044100626957096, "output_power": 0.0001001763322131037, "reference_active": true, "delta_db": -0.8000924864506286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.706489326091672e-06, "output_power": 2.048211967061726e-06, "reference_active": true, "delta_db": -1.2103141939319089}], "active_band_count": 3, "active_band_mae_db": 0.6777174688670756}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 7.943457135434651e-05, "log_rms_mae_db": 1.7641734377682097, "rms_envelope_nrmse": 0.14089662904615607, "envelope_correlation": 0.9847823941601684, "rms_gain_error_db": -0.07749676994520957, "reference_mean_power": 0.00024301128670702193, "output_mean_power": 0.00023871338322147673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.5901124046586967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.178189042095757e-05, "output_power": 8.529305757343954e-05, "reference_active": true, "delta_db": -0.3184331534766616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7997292458206206e-05, "output_power": 1.7408824706280907e-05, "reference_active": true, "delta_db": -0.14437721764249925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3356967361336817e-06, "output_power": 2.8075433131299345e-06, "reference_active": true, "delta_db": 0.7991001116284552}], "active_band_count": 3, "active_band_mae_db": 0.4206368275825387}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 6.195280015885064e-05, "log_rms_mae_db": 1.1263294672112591, "rms_envelope_nrmse": 0.08325601729941452, "envelope_correlation": 0.9947408072783779, "rms_gain_error_db": -0.07749725974021605, "reference_mean_power": 0.00024500023514433585, "output_mean_power": 0.00024066712792638202}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.4888167929489627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.774520748189054e-05, "output_power": 8.185447612138593e-05, "reference_active": true, "delta_db": -0.30180971930167905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.213495209539377e-06, "output_power": 8.848452434682101e-06, "reference_active": true, "delta_db": -0.17557093713931965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5741557318464897e-07, "output_power": 1.633807919447508e-07, "reference_active": true, "delta_db": 0.16153301782629464}], "active_band_count": 3, "active_band_mae_db": 0.21297122475576447}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 1.5353574631917574e-05, "log_rms_mae_db": 3.097535650196166, "rms_envelope_nrmse": 0.6545923605024206, "envelope_correlation": 0.7666942280226705, "rms_gain_error_db": 1.7738450932948955, "reference_mean_power": 7.123238480443002e-06, "output_mean_power": 1.0716722711419248e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 4.409187309389394e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6167395840190636e-06, "output_power": 2.7497062874078364e-06, "reference_active": true, "delta_db": 0.21525802587822065}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.009142614437233e-07, "output_power": 6.389712466586671e-07, "reference_active": true, "delta_db": -0.4018358103252513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.642480310119981e-07, "output_power": 3.5715316358998823e-07, "reference_active": true, "delta_db": 1.3084274143111903}], "active_band_count": 3, "active_band_mae_db": 0.6418404168382207}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 1.282880604964887e-05, "log_rms_mae_db": 2.6262890777897647, "rms_envelope_nrmse": 0.5413649394171971, "envelope_correlation": 0.8289002989636107, "rms_gain_error_db": 1.773863072086521, "reference_mean_power": 7.1814170088864185e-06, "output_mean_power": 1.0804295486265083e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 3.966042949113477e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6674623865712373e-06, "output_power": 2.709169018496908e-06, "reference_active": true, "delta_db": 0.06737796231985275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.255436860296439e-07, "output_power": 3.063236244843644e-07, "reference_active": true, "delta_db": -0.26428784523664384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7820276752795505e-08, "output_power": 2.8411391719888475e-08, "reference_active": true, "delta_db": 0.09131061986465586}], "active_band_count": 3, "active_band_mae_db": 0.1409921424737175}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9038100776766215, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "aa29950a3c7c43d5e1225ab8f93614072a74d44c26a79a5d84fb73aea90167aa", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9860929794243096, "standard_deviation": 0.0015200724856856585, "minimum": 0.9835459456482506, "maximum": 0.9886530782678747, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7825761453612103, "standard_deviation": 0.03568979568775097, "minimum": 1.7229700559806318, "maximum": 1.8314253429300136, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.45230638747792695, "standard_deviation": 0.1130276186691527, "minimum": 0.3139683601176249, "maximum": 0.6397289190077925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13498158516322764, "standard_deviation": 0.007046829392085528, "minimum": 0.12224393817621701, "maximum": 0.14611116740156951, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945741367340803, "standard_deviation": 0.0012052548130163214, "minimum": 0.9916861740682043, "maximum": 0.9953914084410792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2322125255943228, "standard_deviation": 0.061534430146541134, "minimum": 1.1263294672112591, "maximum": 1.3165227269662658, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4842097385893058, "standard_deviation": 0.18007094774284446, "minimum": 0.21297122475576447, "maximum": 0.7862394071409651, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0844696968387951, "standard_deviation": 0.007347556486214859, "minimum": 0.07733542415193477, "maximum": 0.10105198216876216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.7781603076611545, "standard_deviation": 0.009215072703154198, "minimum": 0.7657402858495738, "maximum": 0.7907287936798566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.9828855322307417, "standard_deviation": 0.06971904975709284, "minimum": 2.928909631521083, "maximum": 3.097535650196166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4501787505999809, "standard_deviation": 0.2545401992513661, "minimum": 0.10038988445767227, "maximum": 0.9478156255829537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.6400608810640828, "standard_deviation": 0.019289920137495555, "minimum": 0.6101458816644222, "maximum": 0.6702296994539516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8370134667111272, "standard_deviation": 0.006603638363587409, "minimum": 0.8289002989636107, "maximum": 0.8482039842759884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.5434161091042373, "standard_deviation": 0.07556084023554042, "minimum": 2.4058749246997166, "maximum": 2.6262890777897647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5823440276239908, "standard_deviation": 0.2934141608662526, "minimum": 0.1409921424737175, "maximum": 1.0876757808507642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.5302866669982695, "standard_deviation": 0.014424166885645875, "minimum": 0.5045140223987843, "maximum": 0.5507481430558281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9889626197525087, "standard_deviation": 0.001509683376422648, "minimum": 0.9864628669974629, "maximum": 0.9909257365854459, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1936866535951192, "standard_deviation": 0.054857269693977785, "minimum": 1.1050879248160883, "maximum": 1.269725323266559, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6341850854266837, "standard_deviation": 0.16562882439384338, "minimum": 0.4312779428050497, "maximum": 0.9418588354493256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09993320142189577, "standard_deviation": 0.006377901110302202, "minimum": 0.09195955965090426, "maximum": 0.11044716937972478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959995501740333, "standard_deviation": 0.0010525000393948733, "minimum": 0.9948515850485365, "maximum": 0.9977241435761369, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6512017717457147, "standard_deviation": 0.022985389954198465, "minimum": 0.611944657027556, "maximum": 0.680818953831422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6594999659928886, "standard_deviation": 0.27828526635045236, "minimum": 0.3192482626936916, "maximum": 1.220753008536141, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0575783673682988, "standard_deviation": 0.006502121140044955, "minimum": 0.04731646538085978, "maximum": 0.06476241698266114, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.7422222799822459, "standard_deviation": 0.027127098856303752, "minimum": 0.7144438187847729, "maximum": 0.7981950628446599, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.429924949176814, "standard_deviation": 0.009863675127648484, "minimum": 6.419959478750261, "maximum": 6.451014053867342, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5171304832957944, "standard_deviation": 0.028609693622667996, "minimum": 0.4738975457765095, "maximum": 0.5626773229958485, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0793493554587427, "standard_deviation": 0.012351683062947365, "minimum": -0.09749770252464077, "maximum": -0.06341001348639566, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9928443726723298, "standard_deviation": 0.0006358365411153989, "minimum": 0.991679649508346, "maximum": 0.9935349785501728, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8052807326291636, "standard_deviation": 0.019876275813186368, "minimum": 0.7718839747221883, "maximum": 0.829860131160505, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3893565107844777, "standard_deviation": 0.13293607290830806, "minimum": 0.23889016400391572, "maximum": 0.57383788527206, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0729869123235005, "standard_deviation": 0.0031607930029797903, "minimum": 0.06944492984604571, "maximum": 0.07861568039809524, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974430949901548, "standard_deviation": 0.0006448994692481647, "minimum": 0.9967086565110438, "maximum": 0.9984820279486301, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44679291474472416, "standard_deviation": 0.01958594922902215, "minimum": 0.424061035266538, "maximum": 0.47613605852063773, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4585568865156268, "standard_deviation": 0.19348119710942815, "minimum": 0.11903221521661754, "maximum": 0.7559997492079952, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04093285609273312, "standard_deviation": 0.005147107258298093, "minimum": 0.032727980462905135, "maximum": 0.04641438198987176, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0003", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.0324032306671143, "predicted_mos": 2.0324032306671143, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "ae81451a07025279bc06855e2457da86a8000b489580b6c05cdd8d2f6dc43c7b", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.146073500160128, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8240798998469089, "全帯域_10ms_rms_envelope_nrmse": 0.07036158994243329, "全帯域_10ms_envelope_correlation": 0.9933150663856891, "全帯域_10ms_modulation_mae_db": 0.3386038564630067, "全帯域_40ms_log_rms_mae_db": 0.4420775224040768, "全帯域_40ms_rms_envelope_nrmse": 0.03780715292912129, "全帯域_40ms_envelope_correlation": 0.9977537304558024, "全帯域_40ms_modulation_mae_db": 0.44283219409193436, "300–1000 Hz_10ms_log_rms_mae_db": 1.0801876470250973, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09626951172595821, "300–1000 Hz_10ms_envelope_correlation": 0.9896179453469487, "300–1000 Hz_10ms_modulation_mae_db": 0.5085891963070718, "300–1000 Hz_40ms_log_rms_mae_db": 0.6168273598238375, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0528208067276695, "300–1000 Hz_40ms_envelope_correlation": 0.9963869771727515, "300–1000 Hz_40ms_modulation_mae_db": 0.5976649504775041, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4423761237795834, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13766928131584913, "1000–3000 Hz_10ms_envelope_correlation": 0.9854015379261907, "1000–3000 Hz_10ms_modulation_mae_db": 0.629242455844845, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0692488193576646, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07985577358608749, "1000–3000 Hz_40ms_envelope_correlation": 0.9949545035274451, "1000–3000 Hz_40ms_modulation_mae_db": 0.5998445537954474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9184926534109599, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5803528346246264, "3000–6000 Hz_10ms_envelope_correlation": 0.8183135408570886, "3000–6000 Hz_10ms_modulation_mae_db": 0.5581210581695203, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8541587950116896, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5036719414907045, "3000–6000 Hz_40ms_envelope_correlation": 0.8488482785149388, "3000–6000 Hz_40ms_modulation_mae_db": 0.6056542981428163, "mr_stft_db": 5.837232182833499, "band_mae_db": 0.604753376136855, "rms_gain_error_db": -0.006222034197654425, "output_peak": 0.5739100865521474}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8310883036813825, "全帯域_10ms_rms_envelope_nrmse": 0.07757455979643411, "全帯域_10ms_envelope_correlation": 0.9921632116519371, "全帯域_10ms_modulation_mae_db": 0.25437513026011455, "全帯域_40ms_log_rms_mae_db": 0.491518417517482, "全帯域_40ms_rms_envelope_nrmse": 0.04642704296075937, "全帯域_40ms_envelope_correlation": 0.9970635045935797, "全帯域_40ms_modulation_mae_db": 0.13290347123609123, "300–1000 Hz_10ms_log_rms_mae_db": 1.1115286869941274, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1021674337494409, "300–1000 Hz_10ms_envelope_correlation": 0.9891656160896576, "300–1000 Hz_10ms_modulation_mae_db": 0.48184852545623635, "300–1000 Hz_40ms_log_rms_mae_db": 0.6735250038106951, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0693482256259658, "300–1000 Hz_40ms_envelope_correlation": 0.9949744334746183, "300–1000 Hz_40ms_modulation_mae_db": 0.3090060582368941, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4671238557092758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1276309360173712, "1000–3000 Hz_10ms_envelope_correlation": 0.9879300485109803, "1000–3000 Hz_10ms_modulation_mae_db": 0.43045287569830587, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0563559677691712, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08876496471766723, "1000–3000 Hz_40ms_envelope_correlation": 0.9943672597041769, "1000–3000 Hz_40ms_modulation_mae_db": 0.6091351083737887, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9541874172947726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5687126153865084, "3000–6000 Hz_10ms_envelope_correlation": 0.8043098591757613, "3000–6000 Hz_10ms_modulation_mae_db": 0.7894786439367234, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8476085593342917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4957293499712173, "3000–6000 Hz_40ms_envelope_correlation": 0.8327739062981531, "3000–6000 Hz_40ms_modulation_mae_db": 1.0253920378710906, "mr_stft_db": 5.821612957367523, "band_mae_db": 0.5856502630597856, "rms_gain_error_db": -0.1463879680592831, "output_peak": 0.4461613974595096}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8202275585484318, "全帯域_10ms_rms_envelope_nrmse": 0.07906749117132464, "全帯域_10ms_envelope_correlation": 0.9917897760226528, "全帯域_10ms_modulation_mae_db": 0.43187305499525763, "全帯域_40ms_log_rms_mae_db": 0.46346888555450205, "全帯域_40ms_rms_envelope_nrmse": 0.04441317116972012, "全帯域_40ms_envelope_correlation": 0.9972841451504161, "全帯域_40ms_modulation_mae_db": 0.5737959581811726, "300–1000 Hz_10ms_log_rms_mae_db": 1.0988341817518166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10586489434351654, "300–1000 Hz_10ms_envelope_correlation": 0.9879022595239386, "300–1000 Hz_10ms_modulation_mae_db": 0.6378253162739843, "300–1000 Hz_40ms_log_rms_mae_db": 0.6494301762829359, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06281789550102758, "300–1000 Hz_40ms_envelope_correlation": 0.9956321433249626, "300–1000 Hz_40ms_modulation_mae_db": 0.8387206615734569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.441892027049113, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12396965667693256, "1000–3000 Hz_10ms_envelope_correlation": 0.9884896503830265, "1000–3000 Hz_10ms_modulation_mae_db": 0.38767239338951587, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0502442098994111, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07319535591453413, "1000–3000 Hz_40ms_envelope_correlation": 0.9962142749665736, "1000–3000 Hz_40ms_modulation_mae_db": 0.4477072459900538, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9280349266998216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5742339230744702, "3000–6000 Hz_10ms_envelope_correlation": 0.8092969082365462, "3000–6000 Hz_10ms_modulation_mae_db": 0.25476797072580104, "3000–6000 Hz_40ms_log_rms_mae_db": 1.789749245953019, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5007033149053803, "3000–6000 Hz_40ms_envelope_correlation": 0.8381050035137394, "3000–6000 Hz_40ms_modulation_mae_db": 0.6247332284194044, "mr_stft_db": 5.836215367957384, "band_mae_db": 0.5957489145439165, "rms_gain_error_db": -0.140369017418258, "output_peak": 0.48534373531786074}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8131085944172005, "全帯域_10ms_rms_envelope_nrmse": 0.06583473241668511, "全帯域_10ms_envelope_correlation": 0.9941439494159672, "全帯域_10ms_modulation_mae_db": 0.35098886434689786, "全帯域_40ms_log_rms_mae_db": 0.4302991151992177, "全帯域_40ms_rms_envelope_nrmse": 0.03661595419128376, "全帯域_40ms_envelope_correlation": 0.9979024132937091, "全帯域_40ms_modulation_mae_db": 0.3158662376428739, "300–1000 Hz_10ms_log_rms_mae_db": 1.082381590246719, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09033235307566262, "300–1000 Hz_10ms_envelope_correlation": 0.9908553134068028, "300–1000 Hz_10ms_modulation_mae_db": 0.4574741262808119, "300–1000 Hz_40ms_log_rms_mae_db": 0.5703526263708396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05005395944331576, "300–1000 Hz_40ms_envelope_correlation": 0.9967625799904531, "300–1000 Hz_40ms_modulation_mae_db": 0.4759807582648857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4418168902805066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11011070764119249, "1000–3000 Hz_10ms_envelope_correlation": 0.9908426349166196, "1000–3000 Hz_10ms_modulation_mae_db": 0.13480152975316081, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0171671913556433, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0700791007194052, "1000–3000 Hz_40ms_envelope_correlation": 0.9963199900875973, "1000–3000 Hz_40ms_modulation_mae_db": 0.30618460127708375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8878535419498899, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5766394712103845, "3000–6000 Hz_10ms_envelope_correlation": 0.8053907958191326, "3000–6000 Hz_10ms_modulation_mae_db": 0.5489603318460367, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8015989998595507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5041962115469151, "3000–6000 Hz_40ms_envelope_correlation": 0.8338720970176952, "3000–6000 Hz_40ms_modulation_mae_db": 0.7665033487513487, "mr_stft_db": 5.837300732225671, "band_mae_db": 0.5851024582729693, "rms_gain_error_db": -0.028047698834344117, "output_peak": 0.49677203123779334}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8208992904752908, "全帯域_10ms_rms_envelope_nrmse": 0.07203040846734324, "全帯域_10ms_envelope_correlation": 0.9930196868666399, "全帯域_10ms_modulation_mae_db": 0.7129412063902075, "全帯域_40ms_log_rms_mae_db": 0.46925955460571095, "全帯域_40ms_rms_envelope_nrmse": 0.03899928198876962, "全帯域_40ms_envelope_correlation": 0.997694458006373, "全帯域_40ms_modulation_mae_db": 0.7027686213553662, "300–1000 Hz_10ms_log_rms_mae_db": 1.1141133305036406, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09811917301534823, "300–1000 Hz_10ms_envelope_correlation": 0.9893670361345543, "300–1000 Hz_10ms_modulation_mae_db": 0.7003989997774679, "300–1000 Hz_40ms_log_rms_mae_db": 0.6410305620857075, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0579910913693181, "300–1000 Hz_40ms_envelope_correlation": 0.9959327832204531, "300–1000 Hz_40ms_modulation_mae_db": 0.6847623111967679, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4533640008075435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12052781081190414, "1000–3000 Hz_10ms_envelope_correlation": 0.9888235213119027, "1000–3000 Hz_10ms_modulation_mae_db": 0.3412558385748732, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0539018550296657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07440288378538966, "1000–3000 Hz_40ms_envelope_correlation": 0.9954948336095312, "1000–3000 Hz_40ms_modulation_mae_db": 0.5006065276820573, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9086702234212163, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5646759561501467, "3000–6000 Hz_10ms_envelope_correlation": 0.8040219897590289, "3000–6000 Hz_10ms_modulation_mae_db": 0.5867907556107536, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7790653239455028, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.491109893561236, "3000–6000 Hz_40ms_envelope_correlation": 0.8330585469597278, "3000–6000 Hz_40ms_modulation_mae_db": 0.8118387260910586, "mr_stft_db": 5.847851581965631, "band_mae_db": 0.5335607309715181, "rms_gain_error_db": -0.06738987851110648, "output_peak": 0.5096987855425421}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8311279763984802, "全帯域_10ms_rms_envelope_nrmse": 0.07095002811431174, "全帯域_10ms_envelope_correlation": 0.9932135673891996, "全帯域_10ms_modulation_mae_db": 0.19111375515211396, "全帯域_40ms_log_rms_mae_db": 0.46986370588541493, "全帯域_40ms_rms_envelope_nrmse": 0.0391791739165817, "全帯域_40ms_envelope_correlation": 0.9976396911437183, "全帯域_40ms_modulation_mae_db": 0.32744406683572486, "300–1000 Hz_10ms_log_rms_mae_db": 1.1199283220555682, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09798727760101777, "300–1000 Hz_10ms_envelope_correlation": 0.989411994328309, "300–1000 Hz_10ms_modulation_mae_db": 0.2813101605890409, "300–1000 Hz_40ms_log_rms_mae_db": 0.6693056847727943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0544258439164532, "300–1000 Hz_40ms_envelope_correlation": 0.9964366131148497, "300–1000 Hz_40ms_modulation_mae_db": 0.2010735458806506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.448868057396688, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12956796431511805, "1000–3000 Hz_10ms_envelope_correlation": 0.987398084912509, "1000–3000 Hz_10ms_modulation_mae_db": 0.7125511357640196, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0667332482115401, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08221246506943204, "1000–3000 Hz_40ms_envelope_correlation": 0.9951719458738558, "1000–3000 Hz_40ms_modulation_mae_db": 0.6406520903996817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9068750729674047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5618668553294609, "3000–6000 Hz_10ms_envelope_correlation": 0.8052529042676688, "3000–6000 Hz_10ms_modulation_mae_db": 0.42414933183041253, "3000–6000 Hz_40ms_log_rms_mae_db": 1.777592615298059, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4880446651681476, "3000–6000 Hz_40ms_envelope_correlation": 0.8344271247610804, "3000–6000 Hz_40ms_modulation_mae_db": 0.8327640793813803, "mr_stft_db": 5.848455330424336, "band_mae_db": 0.5315717472040669, "rms_gain_error_db": -0.05668534664774247, "output_peak": 0.49521219518119186}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8178609996263251, "全帯域_10ms_rms_envelope_nrmse": 0.08066217205284804, "全帯域_10ms_envelope_correlation": 0.9912195231967247, "全帯域_10ms_modulation_mae_db": 0.5478365881126227, "全帯域_40ms_log_rms_mae_db": 0.44510146225154007, "全帯域_40ms_rms_envelope_nrmse": 0.046661036688494785, "全帯域_40ms_envelope_correlation": 0.9966268307075694, "全帯域_40ms_modulation_mae_db": 0.6073706436569632, "300–1000 Hz_10ms_log_rms_mae_db": 1.1324865238218311, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10785596462768912, "300–1000 Hz_10ms_envelope_correlation": 0.9870591702889862, "300–1000 Hz_10ms_modulation_mae_db": 0.5559122058478513, "300–1000 Hz_40ms_log_rms_mae_db": 0.6289013412442904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06233304353467188, "300–1000 Hz_40ms_envelope_correlation": 0.9951430481910326, "300–1000 Hz_40ms_modulation_mae_db": 0.507608532201253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4618228064514356, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13228113204276712, "1000–3000 Hz_10ms_envelope_correlation": 0.9865656405882097, "1000–3000 Hz_10ms_modulation_mae_db": 0.4460432769668993, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0189179165397502, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07289731794328737, "1000–3000 Hz_40ms_envelope_correlation": 0.9959027634578658, "1000–3000 Hz_40ms_modulation_mae_db": 0.5642909778192063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9535010231836891, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5744893184361389, "3000–6000 Hz_10ms_envelope_correlation": 0.8190096996748905, "3000–6000 Hz_10ms_modulation_mae_db": 0.6603774587483934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8028292040337348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49900100770307737, "3000–6000 Hz_40ms_envelope_correlation": 0.8494075361408895, "3000–6000 Hz_40ms_modulation_mae_db": 0.46046971379692403, "mr_stft_db": 5.838448679452855, "band_mae_db": 0.630277633293304, "rms_gain_error_db": -0.05888676147287834, "output_peak": 0.5143628150345888}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8207631805585448, "全帯域_10ms_rms_envelope_nrmse": 0.06983560121162921, "全帯域_10ms_envelope_correlation": 0.9934305719213973, "全帯域_10ms_modulation_mae_db": 0.286710657776598, "全帯域_40ms_log_rms_mae_db": 0.4166599890490331, "全帯域_40ms_rms_envelope_nrmse": 0.03651028105761166, "全帯域_40ms_envelope_correlation": 0.9979272392697309, "全帯域_40ms_modulation_mae_db": 0.29992713868297277, "300–1000 Hz_10ms_log_rms_mae_db": 1.1453001668157614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0922460083094804, "300–1000 Hz_10ms_envelope_correlation": 0.9905369679860156, "300–1000 Hz_10ms_modulation_mae_db": 0.5612277965167128, "300–1000 Hz_40ms_log_rms_mae_db": 0.6342544207264739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050089069037138875, "300–1000 Hz_40ms_envelope_correlation": 0.9968729100885306, "300–1000 Hz_40ms_modulation_mae_db": 0.644339904280593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3648486979719374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11984201727927936, "1000–3000 Hz_10ms_envelope_correlation": 0.989124022503882, "1000–3000 Hz_10ms_modulation_mae_db": 0.43020055416922026, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9745586903696697, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07882655309316733, "1000–3000 Hz_40ms_envelope_correlation": 0.9953103318035204, "1000–3000 Hz_40ms_modulation_mae_db": 0.3841928320317725, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9513935916774345, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.571540281047153, "3000–6000 Hz_10ms_envelope_correlation": 0.807828033057296, "3000–6000 Hz_10ms_modulation_mae_db": 0.452335817419716, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8774668648576565, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4975307986537468, "3000–6000 Hz_40ms_envelope_correlation": 0.8374492861040871, "3000–6000 Hz_40ms_modulation_mae_db": 0.7804686277321258, "mr_stft_db": 5.844466208249425, "band_mae_db": 0.5435422298485791, "rms_gain_error_db": -0.021320465423879678, "output_peak": 0.5373469751755524}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59200, "output_samples": 59200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.454155329723112, "spectral_convergence": 0.20893279071862947, "spectral_convergence_status": "ok", "amplitude_floor": 2.0296349328313604e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.509898188575514, "spectral_convergence": 0.20244285618013866, "spectral_convergence_status": "ok", "amplitude_floor": 1.6850709272784723e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.547643030201868, "spectral_convergence": 0.21129317513869086, "spectral_convergence_status": "ok", "amplitude_floor": 8.273401155700601e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185911287346583, "output_power": 0.0008098865327257773, "delta_db": -0.04642856463415403, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003085801603155809, "output_power": 0.0030553884562612025, "delta_db": -0.0430156687128136, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013392891870332754, "output_power": 0.0013734364721928187, "delta_db": 0.10934213583946742, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027936558999729056, "output_power": 0.00027023423761193664, "delta_db": -0.14432540607186659, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.040755904577448e-06, "output_power": 8.953165151190143e-06, "delta_db": 1.0435730755904675, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.309787748977687e-06, "output_power": 1.0573001434384031e-05, "delta_db": 2.241835405972361, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.604753376136855, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.006222034197654425, "pitch": null, "output_rms_dbfs": -22.573948962084238, "output_peak": 0.5739100865521474, "reference_rms_dbfs": -22.567726927886582, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.837232182833499}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002672933594000982, "log_rms_mae_db": 0.8240798998469089, "rms_envelope_nrmse": 0.07036158994243329, "envelope_correlation": 0.9933150663856891, "rms_gain_error_db": -0.006222047038338797, "reference_mean_power": 0.0055438895852693145, "output_mean_power": 0.005535952656447272}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.3303693041440347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007860073564586987, "output_power": 0.0007647122083812682, "reference_active": true, "delta_db": -0.11928587125561191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024364364087793404, "output_power": 0.00023806744417455042, "reference_active": true, "delta_db": -0.10055071230189783}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.963914121364001e-05, "output_power": 4.1326322525020184e-05, "reference_active": true, "delta_db": -0.7959749858315104}], "active_band_count": 3, "active_band_mae_db": 0.3386038564630067}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00019127079187884833, "log_rms_mae_db": 0.4420775224040768, "rms_envelope_nrmse": 0.03780715292912129, "envelope_correlation": 0.9977537304558024, "rms_gain_error_db": -0.006221991290960983, "reference_mean_power": 0.005589268172913711, "output_mean_power": 0.005581266349321566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.0930617045795464e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007030365316665854, "output_power": 0.0006802874323887168, "reference_active": true, "delta_db": -0.14285444823607055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013359627830049334, "output_power": 0.00012992641773332334, "reference_active": true, "delta_db": -0.1209689535287027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7787329335262405e-06, "output_power": 2.1746005338196573e-06, "reference_active": true, "delta_db": -1.0646731805110299}], "active_band_count": 3, "active_band_mae_db": 0.44283219409193436}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002516933783769556, "log_rms_mae_db": 1.0801876470250973, "rms_envelope_nrmse": 0.09626951172595821, "envelope_correlation": 0.9896179453469487, "rms_gain_error_db": -0.03815889065113608, "reference_mean_power": 0.003609667084718048, "output_mean_power": 0.00357809000009782}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.0389022511376814e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005770551731550911, "output_power": 0.0005515576988856744, "reference_active": true, "delta_db": -0.1962638768038484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021321841152816334, "output_power": 0.00020398021958820323, "reference_active": true, "delta_db": -0.19236648569362622}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0182675609825566e-05, "output_power": 3.862247408165316e-05, "reference_active": true, "delta_db": -1.137137226423741}], "active_band_count": 3, "active_band_mae_db": 0.5085891963070718}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00017703062295034008, "log_rms_mae_db": 0.6168273598238375, "rms_envelope_nrmse": 0.0528208067276695, "envelope_correlation": 0.9963869771727515, "rms_gain_error_db": -0.0381589570410842, "reference_mean_power": 0.003639213964400563, "output_mean_power": 0.003607378350848213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 8.593451292498866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005240710435115444, "output_power": 0.0004980789971569267, "reference_active": true, "delta_db": -0.2209193537849668}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012044100626957096, "output_power": 0.00011528672620647399, "reference_active": true, "delta_db": -0.18995068619641348}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.706489326091672e-06, "output_power": 1.968764807220952e-06, "reference_active": true, "delta_db": -1.382124811451132}], "active_band_count": 3, "active_band_mae_db": 0.5976649504775041}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 7.943457135434651e-05, "log_rms_mae_db": 1.4423761237795834, "rms_envelope_nrmse": 0.13766928131584913, "envelope_correlation": 0.9854015379261907, "rms_gain_error_db": -0.1387256085340795, "reference_mean_power": 0.00024301128670702193, "output_mean_power": 0.00023537150489923677}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.5901124046586967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.178189042095757e-05, "output_power": 8.502796293485429e-05, "reference_active": true, "delta_db": -0.3319522401217967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7997292458206206e-05, "output_power": 1.6226072571439157e-05, "reference_active": true, "delta_db": -0.44993760015900075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3356967361336817e-06, "output_power": 1.810642462808299e-06, "reference_active": true, "delta_db": -1.1058375272537373}], "active_band_count": 3, "active_band_mae_db": 0.629242455844845}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 6.195280015885064e-05, "log_rms_mae_db": 1.0692488193576646, "rms_envelope_nrmse": 0.07985577358608749, "envelope_correlation": 0.9949545035274451, "rms_gain_error_db": -0.13872649246746313, "reference_mean_power": 0.00024500023514433585, "output_mean_power": 0.00023729787653506754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.4888167929489627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.774520748189054e-05, "output_power": 8.086706139345442e-05, "reference_active": true, "delta_db": -0.3545174355415573}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.213495209539377e-06, "output_power": 8.364350132237775e-06, "reference_active": true, "delta_db": -0.41992210251121487}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5741557318464897e-07, "output_power": 1.2431922174945875e-07, "reference_active": true, "delta_db": -1.0250941233335702}], "active_band_count": 3, "active_band_mae_db": 0.5998445537954474}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 1.5353574631917574e-05, "log_rms_mae_db": 1.9184926534109599, "rms_envelope_nrmse": 0.5803528346246264, "envelope_correlation": 0.8183135408570886, "rms_gain_error_db": 1.5616813175153645, "reference_mean_power": 7.123238480443002e-06, "output_mean_power": 1.0205766227135419e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 4.409187309389394e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6167395840190636e-06, "output_power": 2.721718599270275e-06, "reference_active": true, "delta_db": 0.17082716997946254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.009142614437233e-07, "output_power": 6.968034067182294e-07, "reference_active": true, "delta_db": -0.025546313274396666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.642480310119981e-07, "output_power": 3.7137334470220564e-07, "reference_active": true, "delta_db": 1.4779896912547017}], "active_band_count": 3, "active_band_mae_db": 0.5581210581695203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 1.282880604964887e-05, "log_rms_mae_db": 1.8541587950116896, "rms_envelope_nrmse": 0.5036719414907045, "envelope_correlation": 0.8488482785149388, "rms_gain_error_db": 1.561708149996316, "reference_mean_power": 7.1814170088864185e-06, "output_mean_power": 1.0289184645699917e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 3.966042949113477e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6674623865712373e-06, "output_power": 2.732726293034819e-06, "reference_active": true, "delta_db": 0.10497831168700128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.255436860296439e-07, "output_power": 3.747170823533627e-07, "reference_active": true, "delta_db": 0.6109421531768908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7820276752795505e-08, "output_power": 3.584806257089647e-08, "reference_active": true, "delta_db": 1.101042429564557}], "active_band_count": 3, "active_band_mae_db": 0.6056542981428163}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9199584515884465, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b951f8f43fff871a5021722599528bea93b7744f11eb290dfd0fd2654247743e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9880718926316651, "standard_deviation": 0.0016646399592642209, "minimum": 0.9854015379261907, "maximum": 0.9908426349166196, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4402640574307604, "standard_deviation": 0.031912524626570876, "minimum": 1.3648486979719374, "maximum": 1.4671238557092758, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.439027507520105, "standard_deviation": 0.17564072695425026, "minimum": 0.13480152975316081, "maximum": 0.7125511357640196, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12519993826255177, "standard_deviation": 0.008538582927898322, "minimum": 0.11011070764119249, "maximum": 0.13766928131584913, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9954669878788207, "standard_deviation": 0.0006612066544089962, "minimum": 0.9943672597041769, "maximum": 0.9963199900875973, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0383909873165644, "standard_deviation": 0.03243351025092098, "minimum": 0.9745586903696697, "maximum": 1.0692488193576646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5065767421711365, "standard_deviation": 0.11902000299898802, "minimum": 0.30618460127708375, "maximum": 0.6406520903996817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07752930185362131, "standard_deviation": 0.006103696987740351, "minimum": 0.0700791007194052, "maximum": 0.08876496471766723, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8091779663559266, "standard_deviation": 0.006117766911218451, "minimum": 0.8040219897590289, "maximum": 0.8190096996748905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9261260563256486, "standard_deviation": 0.025006517944970514, "minimum": 1.8878535419498899, "maximum": 1.9541874172947726, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5343726710359196, "standard_deviation": 0.16104068854012996, "minimum": 0.25476797072580104, "maximum": 0.7894786439367234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5715639069073611, "standard_deviation": 0.006191015596528276, "minimum": 0.5618668553294609, "maximum": 0.5803528346246264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8384927224137889, "standard_deviation": 0.006842348908441334, "minimum": 0.8327739062981531, "maximum": 0.8494075361408895, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.816258701036688, "standard_deviation": 0.038064030602414346, "minimum": 1.777592615298059, "maximum": 1.8774668648576565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7384780075232686, "standard_deviation": 0.17196701305346923, "minimum": 0.46046971379692403, "maximum": 1.0253920378710906, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.4974983978750531, "standard_deviation": 0.0057173833673117075, "minimum": 0.4880446651681476, "maximum": 0.5041962115469151, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9892395378881516, "standard_deviation": 0.0012553329290984354, "minimum": 0.9870591702889862, "maximum": 0.9908553134068028, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.11059505615182, "standard_deviation": 0.022812807152277004, "minimum": 1.0801876470250973, "maximum": 1.1453001668157614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5230732908811472, "standard_deviation": 0.1263829367148539, "minimum": 0.2813101605890409, "maximum": 0.7003989997774679, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09885532705601421, "standard_deviation": 0.006157708459616232, "minimum": 0.09033235307566262, "maximum": 0.10785596462768912, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960176860722064, "standard_deviation": 0.0007178708560025556, "minimum": 0.9949744334746183, "maximum": 0.9968729100885306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6354533968896967, "standard_deviation": 0.03262620843573927, "minimum": 0.5703526263708396, "maximum": 0.6735250038106951, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5323945902640006, "standard_deviation": 0.20609169159326682, "minimum": 0.2010735458806506, "maximum": 0.8387206615734569, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05748499189444509, "standard_deviation": 0.006907230628145583, "minimum": 0.05005395944331576, "maximum": 0.0693482256259658, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.5762759191663744, "standard_deviation": 0.03618736768319192, "minimum": 0.5315717472040669, "maximum": 0.630277633293304, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.838947880059541, "standard_deviation": 0.008568976830181284, "minimum": 5.821612957367523, "maximum": 5.848455330424336, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5073510026876482, "standard_deviation": 0.03755434252382118, "minimum": 0.4461613974595096, "maximum": 0.5739100865521474, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06566364632064334, "standard_deviation": 0.05229231977334231, "minimum": -0.1463879680592831, "maximum": -0.006222034197654425, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9927869191062759, "standard_deviation": 0.0009718174872199638, "minimum": 0.9912195231967247, "maximum": 0.9941439494159672, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8223944754440706, "standard_deviation": 0.00622287902043191, "minimum": 0.8131085944172005, "maximum": 0.8311279763984802, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38930538918710234, "standard_deviation": 0.17060182060952767, "minimum": 0.19111375515211396, "maximum": 0.7129412063902075, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07328957289662617, "standard_deviation": 0.0051994132450490765, "minimum": 0.06583473241668511, "maximum": 0.08066217205284804, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974865015776123, "standard_deviation": 0.0004573627879382601, "minimum": 0.9966268307075694, "maximum": 0.9979272392697309, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4535310815583722, "standard_deviation": 0.024370302026527308, "minimum": 0.4166599890490331, "maximum": 0.491518417517482, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4253635414603874, "standard_deviation": 0.19096094089453466, "minimum": 0.13290347123609123, "maximum": 0.7027686213553662, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04082663686279278, "standard_deviation": 0.004306132128232645, "minimum": 0.03651028105761166, "maximum": 0.046661036688494785, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0003", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.535574197769165, "predicted_mos": 2.535574197769165, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "17e1c82af984aeb37c52de8bdf2256edbcfac296bbf0ae4d7e7e16fb57cdb5ad", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.169315000064671, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7763457457878513, "全帯域_10ms_rms_envelope_nrmse": 0.071704047803339, "全帯域_10ms_envelope_correlation": 0.993077943670271, "全帯域_10ms_modulation_mae_db": 0.32905172287059353, "全帯域_40ms_log_rms_mae_db": 0.4177490667786885, "全帯域_40ms_rms_envelope_nrmse": 0.03494300466246296, "全帯域_40ms_envelope_correlation": 0.9981507215221594, "全帯域_40ms_modulation_mae_db": 0.3686809109059687, "300–1000 Hz_10ms_log_rms_mae_db": 1.1108103452669154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09749867777615648, "300–1000 Hz_10ms_envelope_correlation": 0.9894479187885842, "300–1000 Hz_10ms_modulation_mae_db": 0.3701559335433831, "300–1000 Hz_40ms_log_rms_mae_db": 0.5954287697980275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049163725506490306, "300–1000 Hz_40ms_envelope_correlation": 0.9970608833172564, "300–1000 Hz_40ms_modulation_mae_db": 0.3154740192309562, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5645424359940623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13272576115089105, "1000–3000 Hz_10ms_envelope_correlation": 0.9864257171348991, "1000–3000 Hz_10ms_modulation_mae_db": 0.22936346227931806, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0450160766214207, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08279693494971717, "1000–3000 Hz_40ms_envelope_correlation": 0.9944727273441385, "1000–3000 Hz_40ms_modulation_mae_db": 0.29983591420665684, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2099356249469695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5912392169150643, "3000–6000 Hz_10ms_envelope_correlation": 0.8115806845191474, "3000–6000 Hz_10ms_modulation_mae_db": 0.875350154421219, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9872417358023946, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5017912274650987, "3000–6000 Hz_40ms_envelope_correlation": 0.8526964151192938, "3000–6000 Hz_40ms_modulation_mae_db": 0.7694454895179891, "mr_stft_db": 6.029643205175375, "band_mae_db": 0.663341722670246, "rms_gain_error_db": -0.06365766583005214, "output_peak": 0.47694135518242387}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7904859096017186, "全帯域_10ms_rms_envelope_nrmse": 0.07486558312259288, "全帯域_10ms_envelope_correlation": 0.992468096964175, "全帯域_10ms_modulation_mae_db": 0.4121273976182393, "全帯域_40ms_log_rms_mae_db": 0.4635782126484974, "全帯域_40ms_rms_envelope_nrmse": 0.043664917213890375, "全帯域_40ms_envelope_correlation": 0.9971019996367798, "全帯域_40ms_modulation_mae_db": 0.4815928571712553, "300–1000 Hz_10ms_log_rms_mae_db": 1.0498195420656355, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09907838905799593, "300–1000 Hz_10ms_envelope_correlation": 0.9891695186969244, "300–1000 Hz_10ms_modulation_mae_db": 0.5965961847961575, "300–1000 Hz_40ms_log_rms_mae_db": 0.6284507109072736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061058166752632474, "300–1000 Hz_40ms_envelope_correlation": 0.9954635306435995, "300–1000 Hz_40ms_modulation_mae_db": 0.5678740905707492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5841587188068127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13556953442329928, "1000–3000 Hz_10ms_envelope_correlation": 0.9858856046947868, "1000–3000 Hz_10ms_modulation_mae_db": 0.3618082643716744, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0677329894040735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08575821156795049, "1000–3000 Hz_40ms_envelope_correlation": 0.9942549409209759, "1000–3000 Hz_40ms_modulation_mae_db": 0.47652796870909037, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2213252738819476, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5822815360900742, "3000–6000 Hz_10ms_envelope_correlation": 0.8081054177118726, "3000–6000 Hz_10ms_modulation_mae_db": 0.37011596606642533, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9824072510039628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4986067910762137, "3000–6000 Hz_40ms_envelope_correlation": 0.8453565068275293, "3000–6000 Hz_40ms_modulation_mae_db": 0.35832770967431343, "mr_stft_db": 6.0202158386523905, "band_mae_db": 0.6451081630643887, "rms_gain_error_db": -0.07622001844973231, "output_peak": 0.4715567828333044}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8450044437883435, "全帯域_10ms_rms_envelope_nrmse": 0.07242233965968917, "全帯域_10ms_envelope_correlation": 0.9929340131362221, "全帯域_10ms_modulation_mae_db": 0.4506042159327898, "全帯域_40ms_log_rms_mae_db": 0.4700550699853854, "全帯域_40ms_rms_envelope_nrmse": 0.039698188719228644, "全帯域_40ms_envelope_correlation": 0.9975822720317419, "全帯域_40ms_modulation_mae_db": 0.4728916942228256, "300–1000 Hz_10ms_log_rms_mae_db": 1.1725647017914966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0975613076825587, "300–1000 Hz_10ms_envelope_correlation": 0.9893964678524261, "300–1000 Hz_10ms_modulation_mae_db": 0.6699098040565121, "300–1000 Hz_40ms_log_rms_mae_db": 0.6625005289648582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055570658289922414, "300–1000 Hz_40ms_envelope_correlation": 0.9961300047382788, "300–1000 Hz_40ms_modulation_mae_db": 0.6275606589404602, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4834709196325278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13028361933763175, "1000–3000 Hz_10ms_envelope_correlation": 0.987621245201671, "1000–3000 Hz_10ms_modulation_mae_db": 0.4688235273310042, "1000–3000 Hz_40ms_log_rms_mae_db": 1.036441375144053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09126878809160786, "1000–3000 Hz_40ms_envelope_correlation": 0.9943405284909503, "1000–3000 Hz_40ms_modulation_mae_db": 0.35687662640123824, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2132248580106033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.566771144088213, "3000–6000 Hz_10ms_envelope_correlation": 0.8138510831528422, "3000–6000 Hz_10ms_modulation_mae_db": 0.6104330312398115, "3000–6000 Hz_40ms_log_rms_mae_db": 2.00372242384567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4905253027425834, "3000–6000 Hz_40ms_envelope_correlation": 0.8457836387670643, "3000–6000 Hz_40ms_modulation_mae_db": 1.0146552775499138, "mr_stft_db": 6.011223239048434, "band_mae_db": 0.6675121173870425, "rms_gain_error_db": -0.058428383873099614, "output_peak": 0.5422052157772522}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7888743725459136, "全帯域_10ms_rms_envelope_nrmse": 0.07284859013070652, "全帯域_10ms_envelope_correlation": 0.992851046508888, "全帯域_10ms_modulation_mae_db": 0.1852488948940482, "全帯域_40ms_log_rms_mae_db": 0.4250455020190895, "全帯域_40ms_rms_envelope_nrmse": 0.03743492350170326, "全帯域_40ms_envelope_correlation": 0.9978264211948187, "全帯域_40ms_modulation_mae_db": 0.3241667852284393, "300–1000 Hz_10ms_log_rms_mae_db": 1.1490833183309008, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1017131023552628, "300–1000 Hz_10ms_envelope_correlation": 0.9884187126744459, "300–1000 Hz_10ms_modulation_mae_db": 0.3474874215324264, "300–1000 Hz_40ms_log_rms_mae_db": 0.6507054598118543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053130915607069534, "300–1000 Hz_40ms_envelope_correlation": 0.9963499229122964, "300–1000 Hz_40ms_modulation_mae_db": 0.5197670405271794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5356714063038117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1206501137851319, "1000–3000 Hz_10ms_envelope_correlation": 0.9888162568683924, "1000–3000 Hz_10ms_modulation_mae_db": 0.29680937095698856, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0735485809850351, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0752755024537581, "1000–3000 Hz_40ms_envelope_correlation": 0.9955032784174025, "1000–3000 Hz_40ms_modulation_mae_db": 0.3865563487071843, "3000–6000 Hz_10ms_log_rms_mae_db": 2.270957779742404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.6021021791127754, "3000–6000 Hz_10ms_envelope_correlation": 0.7822881428445572, "3000–6000 Hz_10ms_modulation_mae_db": 0.8878186536313324, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0628007802957438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5151159713488496, "3000–6000 Hz_40ms_envelope_correlation": 0.8222673856295314, "3000–6000 Hz_40ms_modulation_mae_db": 1.2706849202824015, "mr_stft_db": 5.994918378542994, "band_mae_db": 0.5594545076697678, "rms_gain_error_db": -0.033289944705564414, "output_peak": 0.5320879533605916}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8087721533973469, "全帯域_10ms_rms_envelope_nrmse": 0.08198408154360431, "全帯域_10ms_envelope_correlation": 0.9911655321192752, "全帯域_10ms_modulation_mae_db": 0.16299409712925178, "全帯域_40ms_log_rms_mae_db": 0.43926784763919086, "全帯域_40ms_rms_envelope_nrmse": 0.04885791250659208, "全帯域_40ms_envelope_correlation": 0.9966591817225886, "全帯域_40ms_modulation_mae_db": 0.1827527564541173, "300–1000 Hz_10ms_log_rms_mae_db": 1.1692047545227116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10953713850712415, "300–1000 Hz_10ms_envelope_correlation": 0.9869995651688582, "300–1000 Hz_10ms_modulation_mae_db": 0.3692311654126468, "300–1000 Hz_40ms_log_rms_mae_db": 0.623711722644828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07012844483629052, "300–1000 Hz_40ms_envelope_correlation": 0.9943373082104939, "300–1000 Hz_40ms_modulation_mae_db": 0.3294637075231347, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4998063525870315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12980652542042348, "1000–3000 Hz_10ms_envelope_correlation": 0.9879328357700224, "1000–3000 Hz_10ms_modulation_mae_db": 0.9830066915594861, "1000–3000 Hz_40ms_log_rms_mae_db": 1.009187214267884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08628616268150949, "1000–3000 Hz_40ms_envelope_correlation": 0.9953120814288473, "1000–3000 Hz_40ms_modulation_mae_db": 0.979716655934014, "3000–6000 Hz_10ms_log_rms_mae_db": 2.218226874469908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5923886845383638, "3000–6000 Hz_10ms_envelope_correlation": 0.8034455098047993, "3000–6000 Hz_10ms_modulation_mae_db": 0.5665568103779068, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9710058480819128, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5028716853806218, "3000–6000 Hz_40ms_envelope_correlation": 0.844730987002115, "3000–6000 Hz_40ms_modulation_mae_db": 0.895215153191392, "mr_stft_db": 6.014218870013693, "band_mae_db": 0.6984574871224599, "rms_gain_error_db": -0.14667241416431473, "output_peak": 0.5457560406482406}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8096590897087851, "全帯域_10ms_rms_envelope_nrmse": 0.07698396926856964, "全帯域_10ms_envelope_correlation": 0.9924506092257022, "全帯域_10ms_modulation_mae_db": 0.41557454523989806, "全帯域_40ms_log_rms_mae_db": 0.4749047662572805, "全帯域_40ms_rms_envelope_nrmse": 0.04932323638193839, "全帯域_40ms_envelope_correlation": 0.99680064711745, "全帯域_40ms_modulation_mae_db": 0.4806542529120712, "300–1000 Hz_10ms_log_rms_mae_db": 1.1799271085430059, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10828166231728435, "300–1000 Hz_10ms_envelope_correlation": 0.9877170075125982, "300–1000 Hz_10ms_modulation_mae_db": 0.6063799383683045, "300–1000 Hz_40ms_log_rms_mae_db": 0.6931533277863327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06595725162782054, "300–1000 Hz_40ms_envelope_correlation": 0.9956314473920068, "300–1000 Hz_40ms_modulation_mae_db": 0.6731099074779908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4811722137254821, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.118432997403769, "1000–3000 Hz_10ms_envelope_correlation": 0.9894397733310472, "1000–3000 Hz_10ms_modulation_mae_db": 0.2688719655893748, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0297664408176872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07900265789590004, "1000–3000 Hz_40ms_envelope_correlation": 0.9953783144416728, "1000–3000 Hz_40ms_modulation_mae_db": 0.20051986680826325, "3000–6000 Hz_10ms_log_rms_mae_db": 2.207926898818341, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5637959967517826, "3000–6000 Hz_10ms_envelope_correlation": 0.8111578098019339, "3000–6000 Hz_10ms_modulation_mae_db": 0.43886201053093793, "3000–6000 Hz_40ms_log_rms_mae_db": 1.966832964115201, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.48787285154641924, "3000–6000 Hz_40ms_envelope_correlation": 0.8427330387278692, "3000–6000 Hz_40ms_modulation_mae_db": 0.8840901898383774, "mr_stft_db": 5.98961614322931, "band_mae_db": 0.6369351416285913, "rms_gain_error_db": -0.1803171636388899, "output_peak": 0.49141612038457155}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8199852619046496, "全帯域_10ms_rms_envelope_nrmse": 0.07155816764642899, "全帯域_10ms_envelope_correlation": 0.9932029824106603, "全帯域_10ms_modulation_mae_db": 0.47993737723035707, "全帯域_40ms_log_rms_mae_db": 0.4678373757318712, "全帯域_40ms_rms_envelope_nrmse": 0.03847161485331468, "全帯域_40ms_envelope_correlation": 0.9978818124287562, "全帯域_40ms_modulation_mae_db": 0.3521040704362422, "300–1000 Hz_10ms_log_rms_mae_db": 1.2873371437918666, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11037046862879757, "300–1000 Hz_10ms_envelope_correlation": 0.9867657283762722, "300–1000 Hz_10ms_modulation_mae_db": 0.7775479532301862, "300–1000 Hz_40ms_log_rms_mae_db": 0.7756374253192692, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0673599374778343, "300–1000 Hz_40ms_envelope_correlation": 0.9948333160152809, "300–1000 Hz_40ms_modulation_mae_db": 0.5131853907088567, "1000–3000 Hz_10ms_log_rms_mae_db": 1.515669380390568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1316342336048928, "1000–3000 Hz_10ms_envelope_correlation": 0.986642085789567, "1000–3000 Hz_10ms_modulation_mae_db": 0.38558048112402094, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1012281887340356, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08240799041288245, "1000–3000 Hz_40ms_envelope_correlation": 0.9944991924469305, "1000–3000 Hz_40ms_modulation_mae_db": 0.42442391397609897, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2346428824035383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.588253515069758, "3000–6000 Hz_10ms_envelope_correlation": 0.8023784955134284, "3000–6000 Hz_10ms_modulation_mae_db": 0.4598469320857004, "3000–6000 Hz_40ms_log_rms_mae_db": 2.001886255453844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.503306471054003, "3000–6000 Hz_40ms_envelope_correlation": 0.840067311822231, "3000–6000 Hz_40ms_modulation_mae_db": 0.5217772091886119, "mr_stft_db": 6.012463965515775, "band_mae_db": 0.6375833295217436, "rms_gain_error_db": -0.10679050080689051, "output_peak": 0.4991621948600976}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9022236356613923, "全帯域_10ms_rms_envelope_nrmse": 0.08479297133918674, "全帯域_10ms_envelope_correlation": 0.9903586720157964, "全帯域_10ms_modulation_mae_db": 0.468533472862509, "全帯域_40ms_log_rms_mae_db": 0.4865270819289785, "全帯域_40ms_rms_envelope_nrmse": 0.04405576695448607, "全帯域_40ms_envelope_correlation": 0.9971335433411481, "全帯域_40ms_modulation_mae_db": 0.5407806624393546, "300–1000 Hz_10ms_log_rms_mae_db": 1.232607323485329, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11455182779647623, "300–1000 Hz_10ms_envelope_correlation": 0.9854134063897512, "300–1000 Hz_10ms_modulation_mae_db": 0.5602150881452111, "300–1000 Hz_40ms_log_rms_mae_db": 0.6825857436980057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062137613734782875, "300–1000 Hz_40ms_envelope_correlation": 0.9953085532613078, "300–1000 Hz_40ms_modulation_mae_db": 0.7112034708592313, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5918673889275285, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15745753039998042, "1000–3000 Hz_10ms_envelope_correlation": 0.9817013219645644, "1000–3000 Hz_10ms_modulation_mae_db": 0.8906744418366195, "1000–3000 Hz_40ms_log_rms_mae_db": 1.077703985371418, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11507042455058267, "1000–3000 Hz_40ms_envelope_correlation": 0.9905464578814613, "1000–3000 Hz_40ms_modulation_mae_db": 0.8452233453241234, "3000–6000 Hz_10ms_log_rms_mae_db": 2.278475183740784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5859521899815263, "3000–6000 Hz_10ms_envelope_correlation": 0.8039455141923628, "3000–6000 Hz_10ms_modulation_mae_db": 0.4300785096746313, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0217414125110404, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49876662115484904, "3000–6000 Hz_40ms_envelope_correlation": 0.8432270034281961, "3000–6000 Hz_40ms_modulation_mae_db": 0.93954380271383, "mr_stft_db": 5.984428101974271, "band_mae_db": 0.6458143925279852, "rms_gain_error_db": -0.10096003502447787, "output_peak": 0.4925283782498864}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59200, "output_samples": 59200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.714529464098036, "spectral_convergence": 0.2061692093046626, "spectral_convergence_status": "ok", "amplitude_floor": 2.0296349328313604e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.704598786244845, "spectral_convergence": 0.2058146143347425, "spectral_convergence_status": "ok", "amplitude_floor": 1.6850709272784723e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.669801365183247, "spectral_convergence": 0.21701748375376198, "spectral_convergence_status": "ok", "amplitude_floor": 8.273401155700601e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185911287346583, "output_power": 0.0008021596660065693, "delta_db": -0.08806212675753468, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003085801603155809, "output_power": 0.0030178586208366265, "delta_db": -0.09669110031465368, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013392891870332754, "output_power": 0.0013404132131204288, "delta_db": 0.003643377382263509, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027936558999729056, "output_power": 0.00027493398561358504, "delta_db": -0.06944484257166772, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.040755904577448e-06, "output_power": 9.901117616443461e-06, "delta_db": 1.4806493151582811, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.309787748977687e-06, "output_power": 1.0572329935879647e-05, "delta_db": 2.2415595738370757, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.663341722670246, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06365766583005214, "pitch": null, "output_rms_dbfs": -22.631384593716632, "output_peak": 0.47694135518242387, "reference_rms_dbfs": -22.567726927886582, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.029643205175375}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002672933594000982, "log_rms_mae_db": 0.7763457457878513, "rms_envelope_nrmse": 0.071704047803339, "envelope_correlation": 0.993077943670271, "rms_gain_error_db": -0.0636577381858812, "reference_mean_power": 0.0055438895852693145, "output_mean_power": 0.0054632213668765715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.3303693041440347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007860073564586987, "output_power": 0.0007534880041042703, "reference_active": true, "delta_db": -0.183502682135732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024364364087793404, "output_power": 0.00023332477022347796, "reference_active": true, "delta_db": -0.1879423392759577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.963914121364001e-05, "output_power": 4.307777312115632e-05, "reference_active": true, "delta_db": -0.615710147200091}], "active_band_count": 3, "active_band_mae_db": 0.32905172287059353}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00019127079187884833, "log_rms_mae_db": 0.4177490667786885, "rms_envelope_nrmse": 0.03494300466246296, "envelope_correlation": 0.9981507215221594, "rms_gain_error_db": -0.06365787404608836, "reference_mean_power": 0.005589268172913711, "output_mean_power": 0.005507939485909467}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.0930617045795464e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007030365316665854, "output_power": 0.0006753768767674797, "reference_active": true, "delta_db": -0.1743170532812539}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013359627830049334, "output_power": 0.00012651662165434467, "reference_active": true, "delta_db": -0.23646773287604814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7787329335262405e-06, "output_power": 2.367669140394977e-06, "reference_active": true, "delta_db": -0.695257946560604}], "active_band_count": 3, "active_band_mae_db": 0.3686809109059687}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002516933783769556, "log_rms_mae_db": 1.1108103452669154, "rms_envelope_nrmse": 0.09749867777615648, "envelope_correlation": 0.9894479187885842, "rms_gain_error_db": -0.11994263664923657, "reference_mean_power": 0.003609667084718048, "output_mean_power": 0.00351134001429444}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.0389022511376814e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005770551731550911, "output_power": 0.0005463735967026142, "reference_active": true, "delta_db": -0.23727634657201654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021321841152816334, "output_power": 0.00019806397107799965, "reference_active": true, "delta_db": -0.32019221367090844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0182675609825566e-05, "output_power": 4.418286653703773e-05, "reference_active": true, "delta_db": -0.5529992403872244}], "active_band_count": 3, "active_band_mae_db": 0.3701559335433831}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00017703062295034008, "log_rms_mae_db": 0.5954287697980275, "rms_envelope_nrmse": 0.049163725506490306, "envelope_correlation": 0.9970608833172564, "rms_gain_error_db": -0.11994268274415074, "reference_mean_power": 0.003639213964400563, "output_mean_power": 0.0035400820015369892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 8.593451292498866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005240710435115444, "output_power": 0.0004964381465551949, "reference_active": true, "delta_db": -0.2352501878601981}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012044100626957096, "output_power": 0.00010998033631457977, "reference_active": true, "delta_db": -0.3945933193252407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.706489326091672e-06, "output_power": 2.5162191518553807e-06, "reference_active": true, "delta_db": -0.31657855050742983}], "active_band_count": 3, "active_band_mae_db": 0.3154740192309562}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 7.943457135434651e-05, "log_rms_mae_db": 1.5645424359940623, "rms_envelope_nrmse": 0.13272576115089105, "envelope_correlation": 0.9864257171348991, "rms_gain_error_db": -0.054599705863996977, "reference_mean_power": 0.00024301128670702193, "output_mean_power": 0.00023997526192946525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.5901124046586967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.178189042095757e-05, "output_power": 9.112485382440384e-05, "reference_active": true, "delta_db": -0.03120153899622627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7997292458206206e-05, "output_power": 1.5732354615806965e-05, "reference_active": true, "delta_db": -0.5841344697172872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3356967361336817e-06, "output_power": 2.37515464566205e-06, "reference_active": true, "delta_db": 0.07275437812444069}], "active_band_count": 3, "active_band_mae_db": 0.22936346227931806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 6.195280015885064e-05, "log_rms_mae_db": 1.0450160766214207, "rms_envelope_nrmse": 0.08279693494971717, "envelope_correlation": 0.9944727273441385, "rms_gain_error_db": -0.05460047809131528, "reference_mean_power": 0.00024500023514433585, "output_mean_power": 0.0002419393187204251}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.4888167929489627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.774520748189054e-05, "output_power": 8.739574954601831e-05, "reference_active": true, "delta_db": -0.017330938141313708}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.213495209539377e-06, "output_power": 7.928014907621965e-06, "reference_active": true, "delta_db": -0.6525995625895197}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5741557318464897e-07, "output_power": 1.6596076164078973e-07, "reference_active": true, "delta_db": 0.22957724188913703}], "active_band_count": 3, "active_band_mae_db": 0.29983591420665684}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 1.5353574631917574e-05, "log_rms_mae_db": 2.2099356249469695, "rms_envelope_nrmse": 0.5912392169150643, "envelope_correlation": 0.8115806845191474, "rms_gain_error_db": 1.6353164601309997, "reference_mean_power": 7.123238480443002e-06, "output_mean_power": 1.0380281482199337e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 4.409187309389394e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6167395840190636e-06, "output_power": 2.731723629261094e-06, "reference_active": true, "delta_db": 0.18676255174106193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.009142614437233e-07, "output_power": 9.188991717732822e-07, "reference_active": true, "delta_db": 1.176029634591123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.642480310119981e-07, "output_power": 3.5345781436405476e-07, "reference_active": true, "delta_db": 1.2632582769314717}], "active_band_count": 3, "active_band_mae_db": 0.875350154421219}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 1.282880604964887e-05, "log_rms_mae_db": 1.9872417358023946, "rms_envelope_nrmse": 0.5017912274650987, "envelope_correlation": 0.8526964151192938, "rms_gain_error_db": 1.6353311058697348, "reference_mean_power": 7.1814170088864185e-06, "output_mean_power": 1.0465096962199374e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 3.966042949113477e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6674623865712373e-06, "output_power": 2.8119635484333296e-06, "reference_active": true, "delta_db": 0.22911382384239204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.255436860296439e-07, "output_power": 5.125696923791896e-07, "reference_active": true, "delta_db": 1.9714364674864817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7820276752795505e-08, "output_power": 2.8519378937285056e-08, "reference_active": true, "delta_db": 0.1077861772250937}], "active_band_count": 3, "active_band_mae_db": 0.7694454895179891}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9161391592501423, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "27fc6e5d011a50c5bf5c39829ab9e5895852b42fb700c1e79e113b64b994864d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9868081050943688, "standard_deviation": 0.0023901162872341495, "minimum": 0.9817013219645644, "maximum": 0.9894397733310472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5320448520459782, "standard_deviation": 0.0441082516495405, "minimum": 1.4811722137254821, "maximum": 1.5918673889275285, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.48561727563106083, "standard_deviation": 0.289180757023335, "minimum": 0.22936346227931806, "maximum": 0.9830066915594861, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13207003944075246, "standard_deviation": 0.011844766596198162, "minimum": 0.118432997403769, "maximum": 0.15745753039998042, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942884401715475, "standard_deviation": 0.0015944611471145871, "minimum": 0.9905464578814613, "maximum": 0.9955032784174025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0550781064182009, "standard_deviation": 0.030091032324110156, "minimum": 1.009187214267884, "maximum": 1.1012281887340356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4962100800083337, "standard_deviation": 0.272177141563442, "minimum": 0.20051986680826325, "maximum": 0.979716655934014, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08723333407548853, "standard_deviation": 0.012235551020512242, "minimum": 0.0752755024537581, "maximum": 0.11507042455058267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.804594082192618, "standard_deviation": 0.009957913505047971, "minimum": 0.7822881428445572, "maximum": 0.8138510831528422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.231839422001812, "standard_deviation": 0.02779243324441416, "minimum": 2.207926898818341, "maximum": 2.278475183740784, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5798827585034956, "standard_deviation": 0.20141473515941175, "minimum": 0.37011596606642533, "maximum": 0.8878186536313324, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.5840980578184447, "standard_deviation": 0.012985046528537119, "minimum": 0.5637959967517826, "maximum": 0.6021021791127754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8421077859154787, "standard_deviation": 0.008803529657876112, "minimum": 0.8222673856295314, "maximum": 0.8526964151192938, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.9997048338887211, "standard_deviation": 0.03127375822770273, "minimum": 1.966832964115201, "maximum": 2.0628007802957438, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8317174689946036, "standard_deviation": 0.2850059321250432, "minimum": 0.35832770967431343, "maximum": 1.2706849202824015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.49985711522107973, "standard_deviation": 0.008375277881468364, "minimum": 0.48787285154641924, "maximum": 0.5151159713488496, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9879160406824825, "standard_deviation": 0.0014554400304873565, "minimum": 0.9854134063897512, "maximum": 0.9894479187885842, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1689192797247325, "standard_deviation": 0.07188121300369464, "minimum": 1.0498195420656355, "maximum": 1.2873371437918666, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5371904361356035, "standard_deviation": 0.1587138516171445, "minimum": 0.3474874215324264, "maximum": 0.7775479532301862, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10482407176520703, "standard_deviation": 0.006640350956393138, "minimum": 0.09749867777615648, "maximum": 0.11455182779647623, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956393708113151, "standard_deviation": 0.000866175327868754, "minimum": 0.9943373082104939, "maximum": 0.9970608833172564, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6640217111163061, "standard_deviation": 0.05532522212416452, "minimum": 0.5954287697980275, "maximum": 0.7756374253192692, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5322047857298198, "standard_deviation": 0.14680036459122076, "minimum": 0.3154740192309562, "maximum": 0.7112034708592313, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060563339229105374, "standard_deviation": 0.007364213229009407, "minimum": 0.049163725506490306, "maximum": 0.07012844483629052, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.6442758576990282, "standard_deviation": 0.03988461433869598, "minimum": 0.5594545076697678, "maximum": 0.6984574871224599, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.007090967769031, "standard_deviation": 0.01578771817426763, "minimum": 5.984428101974271, "maximum": 6.029643205175375, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.506456755162046, "standard_deviation": 0.029379153176541512, "minimum": 0.4715567828333044, "maximum": 0.5457560406482406, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09579201581162768, "standard_deviation": 0.048696171835725686, "minimum": -0.1803171636388899, "maximum": -0.033289944705564414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9923136120063738, "standard_deviation": 0.0010163961298259392, "minimum": 0.9903586720157964, "maximum": 0.9932029824106603, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8176688265495001, "standard_deviation": 0.04021558479487027, "minimum": 0.7763457457878513, "maximum": 0.9022236356613923, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36300896547221084, "standard_deviation": 0.12559233835117153, "minimum": 0.16299409712925178, "maximum": 0.47993737723035707, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07589496881426466, "standard_deviation": 0.0050172351417577905, "minimum": 0.07155816764642899, "maximum": 0.08479297133918674, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973920748744303, "standard_deviation": 0.0005448617981856291, "minimum": 0.9966591817225886, "maximum": 0.9981507215221594, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45562061537362275, "standard_deviation": 0.02502718230934773, "minimum": 0.4177490667786885, "maximum": 0.4865270819289785, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40045299872128426, "standard_deviation": 0.1161988455169529, "minimum": 0.1827527564541173, "maximum": 0.5407806624393546, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.042056195599202054, "standard_deviation": 0.005291807173574783, "minimum": 0.03494300466246296, "maximum": 0.04932323638193839, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0003", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.2231929302215576, "predicted_mos": 2.2231929302215576, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "29ca5b0c96d5067b94ac7d3977619cd75e89a7c7a441712d0fcccc06ba5f568b", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.99063039990142, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7955658369618125, "全帯域_10ms_rms_envelope_nrmse": 0.06827531639218551, "全帯域_10ms_envelope_correlation": 0.9937031687436745, "全帯域_10ms_modulation_mae_db": 0.18668882657993646, "全帯域_40ms_log_rms_mae_db": 0.42583109082022763, "全帯域_40ms_rms_envelope_nrmse": 0.03863446981871037, "全帯域_40ms_envelope_correlation": 0.9976622471311417, "全帯域_40ms_modulation_mae_db": 0.17103114832603697, "300–1000 Hz_10ms_log_rms_mae_db": 1.0976317973779801, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09687936079190695, "300–1000 Hz_10ms_envelope_correlation": 0.9895016746022558, "300–1000 Hz_10ms_modulation_mae_db": 0.3073943752685216, "300–1000 Hz_40ms_log_rms_mae_db": 0.6321822150569644, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05617821176673089, "300–1000 Hz_40ms_envelope_correlation": 0.9959332619602037, "300–1000 Hz_40ms_modulation_mae_db": 0.12654630368395195, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4672791263391303, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1313043232920218, "1000–3000 Hz_10ms_envelope_correlation": 0.9869255434978259, "1000–3000 Hz_10ms_modulation_mae_db": 0.7477192021533607, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1238804816093628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09389166657386558, "1000–3000 Hz_40ms_envelope_correlation": 0.993283316175649, "1000–3000 Hz_40ms_modulation_mae_db": 0.7856465677493453, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1645122532603067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.583658698964231, "3000–6000 Hz_10ms_envelope_correlation": 0.8174735365385285, "3000–6000 Hz_10ms_modulation_mae_db": 0.4553505327638782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9573831721412978, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4980812201778194, "3000–6000 Hz_40ms_envelope_correlation": 0.8557973120326602, "3000–6000 Hz_40ms_modulation_mae_db": 0.8399777203851716, "mr_stft_db": 5.983436498103015, "band_mae_db": 0.6628312045109687, "rms_gain_error_db": -0.028465172021143074, "output_peak": 0.5338230199929951}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.801021861681728, "全帯域_10ms_rms_envelope_nrmse": 0.07534312836684602, "全帯域_10ms_envelope_correlation": 0.9923227187424051, "全帯域_10ms_modulation_mae_db": 0.4154543782916684, "全帯域_40ms_log_rms_mae_db": 0.44444845969245517, "全帯域_40ms_rms_envelope_nrmse": 0.03822862376486618, "全帯域_40ms_envelope_correlation": 0.9977196248493881, "全帯域_40ms_modulation_mae_db": 0.532764883210924, "300–1000 Hz_10ms_log_rms_mae_db": 1.1525725963039868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10588515612045257, "300–1000 Hz_10ms_envelope_correlation": 0.9874267173638144, "300–1000 Hz_10ms_modulation_mae_db": 0.6300951532993398, "300–1000 Hz_40ms_log_rms_mae_db": 0.6405432227695844, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056768436754424605, "300–1000 Hz_40ms_envelope_correlation": 0.9958674118311347, "300–1000 Hz_40ms_modulation_mae_db": 0.6193091450408251, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5426032732159913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12622250775743024, "1000–3000 Hz_10ms_envelope_correlation": 0.9878664316117898, "1000–3000 Hz_10ms_modulation_mae_db": 0.2692239855592223, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0252569103588014, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08101115007420276, "1000–3000 Hz_40ms_envelope_correlation": 0.9950048376373979, "1000–3000 Hz_40ms_modulation_mae_db": 0.2722570147593289, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1526084559820813, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5862132970764831, "3000–6000 Hz_10ms_envelope_correlation": 0.8030872487763909, "3000–6000 Hz_10ms_modulation_mae_db": 0.4265446972764792, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9009538284907403, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4988736899555159, "3000–6000 Hz_40ms_envelope_correlation": 0.8428727664872204, "3000–6000 Hz_40ms_modulation_mae_db": 0.7532832013562, "mr_stft_db": 5.9956544506652, "band_mae_db": 0.6261184410956021, "rms_gain_error_db": -0.03556632779901752, "output_peak": 0.5425267861507407}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8269318462581431, "全帯域_10ms_rms_envelope_nrmse": 0.07178259604470064, "全帯域_10ms_envelope_correlation": 0.993082374931101, "全帯域_10ms_modulation_mae_db": 0.40926881494854533, "全帯域_40ms_log_rms_mae_db": 0.44732115937125133, "全帯域_40ms_rms_envelope_nrmse": 0.0328284454797898, "全帯域_40ms_envelope_correlation": 0.9984167446505776, "全帯域_40ms_modulation_mae_db": 0.5522803594980527, "300–1000 Hz_10ms_log_rms_mae_db": 1.142117557357296, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09401292352247478, "300–1000 Hz_10ms_envelope_correlation": 0.9904429644039101, "300–1000 Hz_10ms_modulation_mae_db": 0.5525538752372242, "300–1000 Hz_40ms_log_rms_mae_db": 0.640312874095028, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04737768882189459, "300–1000 Hz_40ms_envelope_correlation": 0.9976558480855755, "300–1000 Hz_40ms_modulation_mae_db": 0.683634801099489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6050107658573451, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12393969753887467, "1000–3000 Hz_10ms_envelope_correlation": 0.9882453104075892, "1000–3000 Hz_10ms_modulation_mae_db": 0.41317471971757147, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1324201787385415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07752754332291525, "1000–3000 Hz_40ms_envelope_correlation": 0.9953324530171661, "1000–3000 Hz_40ms_modulation_mae_db": 0.2308945642784538, "3000–6000 Hz_10ms_log_rms_mae_db": 2.202035868725704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5747831101764243, "3000–6000 Hz_10ms_envelope_correlation": 0.8062370203604012, "3000–6000 Hz_10ms_modulation_mae_db": 0.10345409339092322, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8932026685237917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4906763857397548, "3000–6000 Hz_40ms_envelope_correlation": 0.8434100020098444, "3000–6000 Hz_40ms_modulation_mae_db": 0.6820710758455023, "mr_stft_db": 5.990303835342232, "band_mae_db": 0.6096495628118513, "rms_gain_error_db": -0.07569315320662487, "output_peak": 0.5150220653688828}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8382509507495267, "全帯域_10ms_rms_envelope_nrmse": 0.07730248671557337, "全帯域_10ms_envelope_correlation": 0.9919367486778966, "全帯域_10ms_modulation_mae_db": 0.4512533630460089, "全帯域_40ms_log_rms_mae_db": 0.47914256588799137, "全帯域_40ms_rms_envelope_nrmse": 0.04103574363571166, "全帯域_40ms_envelope_correlation": 0.9974083132780408, "全帯域_40ms_modulation_mae_db": 0.44473793229463143, "300–1000 Hz_10ms_log_rms_mae_db": 1.1210483764090104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10135764329383447, "300–1000 Hz_10ms_envelope_correlation": 0.9886490881708243, "300–1000 Hz_10ms_modulation_mae_db": 0.5825265800348967, "300–1000 Hz_40ms_log_rms_mae_db": 0.6599804727960756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05682189711119598, "300–1000 Hz_40ms_envelope_correlation": 0.9961405970059423, "300–1000 Hz_40ms_modulation_mae_db": 0.5834912692596047, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5190358200860403, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13014488212329892, "1000–3000 Hz_10ms_envelope_correlation": 0.9870601927536583, "1000–3000 Hz_10ms_modulation_mae_db": 0.26332088120048586, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0873751879238112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07978976234253755, "1000–3000 Hz_40ms_envelope_correlation": 0.9951742621785088, "1000–3000 Hz_40ms_modulation_mae_db": 0.18837114149611486, "3000–6000 Hz_10ms_log_rms_mae_db": 2.212228502625377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5870828405158409, "3000–6000 Hz_10ms_envelope_correlation": 0.8113522462074496, "3000–6000 Hz_10ms_modulation_mae_db": 0.4193860342989287, "3000–6000 Hz_40ms_log_rms_mae_db": 1.978722442218226, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5062391513340875, "3000–6000 Hz_40ms_envelope_correlation": 0.8465821849027256, "3000–6000 Hz_40ms_modulation_mae_db": 0.7950155838563532, "mr_stft_db": 6.0018831452146655, "band_mae_db": 0.640867625218028, "rms_gain_error_db": -0.05632010383528383, "output_peak": 0.49596099625116774}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7962212733954426, "全帯域_10ms_rms_envelope_nrmse": 0.07332873254949902, "全帯域_10ms_envelope_correlation": 0.9927624682086879, "全帯域_10ms_modulation_mae_db": 0.3477059819357267, "全帯域_40ms_log_rms_mae_db": 0.4576640801072438, "全帯域_40ms_rms_envelope_nrmse": 0.041435472373368656, "全帯域_40ms_envelope_correlation": 0.9973752950558167, "全帯域_40ms_modulation_mae_db": 0.3752212247259305, "300–1000 Hz_10ms_log_rms_mae_db": 1.1133794076788572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09361622628808651, "300–1000 Hz_10ms_envelope_correlation": 0.990249173203512, "300–1000 Hz_10ms_modulation_mae_db": 0.5869466733895874, "300–1000 Hz_40ms_log_rms_mae_db": 0.6152820509645182, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053783021976931564, "300–1000 Hz_40ms_envelope_correlation": 0.9963919756110726, "300–1000 Hz_40ms_modulation_mae_db": 0.6413374213357548, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5406956920396315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12147805624775625, "1000–3000 Hz_10ms_envelope_correlation": 0.9890410915820192, "1000–3000 Hz_10ms_modulation_mae_db": 0.5782494769167666, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0595729512548646, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08164863286979177, "1000–3000 Hz_40ms_envelope_correlation": 0.9952587277985996, "1000–3000 Hz_40ms_modulation_mae_db": 0.34418200142875577, "3000–6000 Hz_10ms_log_rms_mae_db": 2.20542494289277, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5832754432366052, "3000–6000 Hz_10ms_envelope_correlation": 0.7866983300629009, "3000–6000 Hz_10ms_modulation_mae_db": 0.5551389614855884, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9577757740548891, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5016064661940706, "3000–6000 Hz_40ms_envelope_correlation": 0.8225862291405155, "3000–6000 Hz_40ms_modulation_mae_db": 0.7964460620945356, "mr_stft_db": 5.999063901277926, "band_mae_db": 0.5388927869694288, "rms_gain_error_db": -0.06435942418061495, "output_peak": 0.5194024529027258}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8001890401789076, "全帯域_10ms_rms_envelope_nrmse": 0.08191491981376146, "全帯域_10ms_envelope_correlation": 0.9909654949911499, "全帯域_10ms_modulation_mae_db": 0.34419142968111976, "全帯域_40ms_log_rms_mae_db": 0.42568166814092856, "全帯域_40ms_rms_envelope_nrmse": 0.05114589551251116, "全帯域_40ms_envelope_correlation": 0.9959866135484564, "全帯域_40ms_modulation_mae_db": 0.2857667028381971, "300–1000 Hz_10ms_log_rms_mae_db": 1.198280671012424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11623742211514225, "300–1000 Hz_10ms_envelope_correlation": 0.9850502379043797, "300–1000 Hz_10ms_modulation_mae_db": 0.483655189977986, "300–1000 Hz_40ms_log_rms_mae_db": 0.697711110187621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07521724226024723, "300–1000 Hz_40ms_envelope_correlation": 0.9930515657850055, "300–1000 Hz_40ms_modulation_mae_db": 0.4782527973934188, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5096761587218073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11719984805380569, "1000–3000 Hz_10ms_envelope_correlation": 0.9894232421885912, "1000–3000 Hz_10ms_modulation_mae_db": 0.4691084601427393, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0557767243504406, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06989091429491108, "1000–3000 Hz_40ms_envelope_correlation": 0.9960513141321456, "1000–3000 Hz_40ms_modulation_mae_db": 0.3837824179166372, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1790249419921888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5821639639390322, "3000–6000 Hz_10ms_envelope_correlation": 0.805261137727971, "3000–6000 Hz_10ms_modulation_mae_db": 0.20703200644902794, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9814344715436185, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49370484233787837, "3000–6000 Hz_40ms_envelope_correlation": 0.8456985376305751, "3000–6000 Hz_40ms_modulation_mae_db": 0.3172705259797484, "mr_stft_db": 5.987439192659559, "band_mae_db": 0.5807954170648498, "rms_gain_error_db": -0.08234888970807876, "output_peak": 0.500353586617196}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8354793172580852, "全帯域_10ms_rms_envelope_nrmse": 0.07018110734484391, "全帯域_10ms_envelope_correlation": 0.9934477124547075, "全帯域_10ms_modulation_mae_db": 0.504318565852982, "全帯域_40ms_log_rms_mae_db": 0.4538985231419265, "全帯域_40ms_rms_envelope_nrmse": 0.033252108941353616, "全帯域_40ms_envelope_correlation": 0.9984478078064698, "全帯域_40ms_modulation_mae_db": 0.580348448965332, "300–1000 Hz_10ms_log_rms_mae_db": 1.1391973940352738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09526846324394947, "300–1000 Hz_10ms_envelope_correlation": 0.9901325021517545, "300–1000 Hz_10ms_modulation_mae_db": 0.557867797034788, "300–1000 Hz_40ms_log_rms_mae_db": 0.6148412177434762, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04994151526697041, "300–1000 Hz_40ms_envelope_correlation": 0.9972700181170587, "300–1000 Hz_40ms_modulation_mae_db": 0.6407352889612529, "1000–3000 Hz_10ms_log_rms_mae_db": 1.479986623173135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11397363757485686, "1000–3000 Hz_10ms_envelope_correlation": 0.9900572106912426, "1000–3000 Hz_10ms_modulation_mae_db": 0.32193370847626807, "1000–3000 Hz_40ms_log_rms_mae_db": 1.022784585491781, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07107972578397004, "1000–3000 Hz_40ms_envelope_correlation": 0.9960482540727704, "1000–3000 Hz_40ms_modulation_mae_db": 0.13350575346545343, "3000–6000 Hz_10ms_log_rms_mae_db": 2.217737103676756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5834511549070722, "3000–6000 Hz_10ms_envelope_correlation": 0.7950200429558563, "3000–6000 Hz_10ms_modulation_mae_db": 0.19346038218863348, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9713213706646053, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.49515094970804546, "3000–6000 Hz_40ms_envelope_correlation": 0.8357986360006784, "3000–6000 Hz_40ms_modulation_mae_db": 0.562613878395002, "mr_stft_db": 5.97968218491063, "band_mae_db": 0.5751663849327608, "rms_gain_error_db": -0.09863980279004105, "output_peak": 0.5177614115020022}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8379534973582627, "全帯域_10ms_rms_envelope_nrmse": 0.08296329542991093, "全帯域_10ms_envelope_correlation": 0.9906948745334497, "全帯域_10ms_modulation_mae_db": 0.2506028169259978, "全帯域_40ms_log_rms_mae_db": 0.46458932927057744, "全帯域_40ms_rms_envelope_nrmse": 0.0457706446028437, "全帯域_40ms_envelope_correlation": 0.9967037340298591, "全帯域_40ms_modulation_mae_db": 0.25138633918288394, "300–1000 Hz_10ms_log_rms_mae_db": 1.1773613934121376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11027146958428467, "300–1000 Hz_10ms_envelope_correlation": 0.9863452334966133, "300–1000 Hz_10ms_modulation_mae_db": 0.7847872709978577, "300–1000 Hz_40ms_log_rms_mae_db": 0.6554598749370224, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05765688065250228, "300–1000 Hz_40ms_envelope_correlation": 0.9956753543674534, "300–1000 Hz_40ms_modulation_mae_db": 0.9953559560156676, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5458135310886851, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12020093143458592, "1000–3000 Hz_10ms_envelope_correlation": 0.9891024814339548, "1000–3000 Hz_10ms_modulation_mae_db": 0.2626743730264492, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0554596301231687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07155060857368623, "1000–3000 Hz_40ms_envelope_correlation": 0.9963337486253191, "1000–3000 Hz_40ms_modulation_mae_db": 0.3757175812163616, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1774026459392397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.5936461805265105, "3000–6000 Hz_10ms_envelope_correlation": 0.7833749157779787, "3000–6000 Hz_10ms_modulation_mae_db": 0.9969479499624597, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9762891350382628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.5041205623167891, "3000–6000 Hz_40ms_envelope_correlation": 0.8251286284956625, "3000–6000 Hz_40ms_modulation_mae_db": 1.3755703969821915, "mr_stft_db": 6.000683714766513, "band_mae_db": 0.5777942976173348, "rms_gain_error_db": -0.003530676304690011, "output_peak": 0.46269122237660604}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59200, "output_samples": 59200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.668475735357064, "spectral_convergence": 0.20337043908986152, "spectral_convergence_status": "ok", "amplitude_floor": 2.0296349328313604e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.660156509563807, "spectral_convergence": 0.20232419288144, "spectral_convergence_status": "ok", "amplitude_floor": 1.6850709272784723e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.621677249388172, "spectral_convergence": 0.21674232016896358, "spectral_convergence_status": "ok", "amplitude_floor": 8.273401155700601e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185911287346583, "output_power": 0.0008126193786689177, "delta_db": -0.031798589765847735, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003085801603155809, "output_power": 0.0030635924116162016, "delta_db": -0.03137015207441571, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013392891870332754, "output_power": 0.0013365043562745468, "delta_db": -0.009039838452755411, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027936558999729056, "output_power": 0.00026703171088782597, "delta_db": -0.1961007386156508, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.040755904577448e-06, "output_power": 9.519966311882384e-06, "delta_db": 1.310161235004469, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.309787748977687e-06, "output_power": 1.0961409885464436e-05, "delta_db": 2.3985166731526735, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6628312045109687, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.028465172021143074, "pitch": null, "output_rms_dbfs": -22.596192099907725, "output_peak": 0.5338230199929951, "reference_rms_dbfs": -22.567726927886582, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.983436498103015}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002672933594000982, "log_rms_mae_db": 0.7955658369618125, "rms_envelope_nrmse": 0.06827531639218551, "envelope_correlation": 0.9937031687436745, "rms_gain_error_db": -0.02846513584306301, "reference_mean_power": 0.0055438895852693145, "output_mean_power": 0.005507671870672453}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.3303693041440347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007860073564586987, "output_power": 0.0007432067969750672, "reference_active": true, "delta_db": -0.243169379090791}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024364364087793404, "output_power": 0.00022758084613970852, "reference_active": true, "delta_db": -0.2961937295308636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.963914121364001e-05, "output_power": 4.9403068246904216e-05, "reference_active": true, "delta_db": -0.02070337111815482}], "active_band_count": 3, "active_band_mae_db": 0.18668882657993646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00019127079187884833, "log_rms_mae_db": 0.42583109082022763, "rms_envelope_nrmse": 0.03863446981871037, "envelope_correlation": 0.9976622471311417, "rms_gain_error_db": -0.028465040376898003, "reference_mean_power": 0.005589268172913711, "output_mean_power": 0.005552754126287989}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.0930617045795464e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007030365316665854, "output_power": 0.0006590390554063522, "reference_active": true, "delta_db": -0.280667405708495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013359627830049334, "output_power": 0.00012768839436921788, "reference_active": true, "delta_db": -0.19642933894864362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7787329335262405e-06, "output_power": 2.7557964625559997e-06, "reference_active": true, "delta_db": -0.03599670032097231}], "active_band_count": 3, "active_band_mae_db": 0.17103114832603697}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002516933783769556, "log_rms_mae_db": 1.0976317973779801, "rms_envelope_nrmse": 0.09687936079190695, "envelope_correlation": 0.9895016746022558, "rms_gain_error_db": -0.06445949130938522, "reference_mean_power": 0.003609667084718048, "output_mean_power": 0.0035564867919730576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.0389022511376814e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005770551731550911, "output_power": 0.0005552225712830781, "reference_active": true, "delta_db": -0.16750225682836223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021321841152816334, "output_power": 0.00020151908204407433, "reference_active": true, "delta_db": -0.24508527338776367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0182675609825566e-05, "output_power": 4.4626646580184165e-05, "reference_active": true, "delta_db": -0.5095955955894389}], "active_band_count": 3, "active_band_mae_db": 0.3073943752685216}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00017703062295034008, "log_rms_mae_db": 0.6321822150569644, "rms_envelope_nrmse": 0.05617821176673089, "envelope_correlation": 0.9959332619602037, "rms_gain_error_db": -0.06445953622926327, "reference_mean_power": 0.003639213964400563, "output_mean_power": 0.003585598328022969}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 8.593451292498866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005240710435115444, "output_power": 0.000497213360447954, "reference_active": true, "delta_db": -0.2284737439690897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012044100626957096, "output_power": 0.00011720859529739156, "reference_active": true, "delta_db": -0.11814914218657992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.706489326091672e-06, "output_power": 2.685992008263287e-06, "reference_active": true, "delta_db": -0.03301602489618625}], "active_band_count": 3, "active_band_mae_db": 0.12654630368395195}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 7.943457135434651e-05, "log_rms_mae_db": 1.4672791263391303, "rms_envelope_nrmse": 0.1313043232920218, "envelope_correlation": 0.9869255434978259, "rms_gain_error_db": -0.18205038760257602, "reference_mean_power": 0.00024301128670702193, "output_mean_power": 0.00023303513559664208}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.5901124046586967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.178189042095757e-05, "output_power": 7.839949488863759e-05, "reference_active": true, "delta_db": -0.6844373393789484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7997292458206206e-05, "output_power": 1.4949115391656391e-05, "reference_active": true, "delta_db": -0.8059167984662936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3356967361336817e-06, "output_power": 1.9639756863546187e-06, "reference_active": true, "delta_db": -0.7528034686148406}], "active_band_count": 3, "active_band_mae_db": 0.7477192021533607}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 6.195280015885064e-05, "log_rms_mae_db": 1.1238804816093628, "rms_envelope_nrmse": 0.09389166657386558, "envelope_correlation": 0.993283316175649, "rms_gain_error_db": -0.1820512977277308, "reference_mean_power": 0.00024500023514433585, "output_mean_power": 0.00023494238406351805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.4888167929489627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.774520748189054e-05, "output_power": 7.464533483374236e-05, "reference_active": true, "delta_db": -0.7022073476365415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.213495209539377e-06, "output_power": 7.647451563220484e-06, "reference_active": true, "delta_db": -0.8090767940447535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5741557318464897e-07, "output_power": 1.2956334113387573e-07, "reference_active": true, "delta_db": -0.845655561566741}], "active_band_count": 3, "active_band_mae_db": 0.7856465677493453}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 1.5353574631917574e-05, "log_rms_mae_db": 2.1645122532603067, "rms_envelope_nrmse": 0.583658698964231, "envelope_correlation": 0.8174735365385285, "rms_gain_error_db": 1.6460324558410668, "reference_mean_power": 7.123238480443002e-06, "output_mean_power": 1.0405925924584604e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 4.409187309389394e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6167395840190636e-06, "output_power": 2.9007176595562197e-06, "reference_active": true, "delta_db": 0.4474495481665324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.009142614437233e-07, "output_power": 8.073027459374374e-07, "reference_active": true, "delta_db": 0.6137153305641664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.642480310119981e-07, "output_power": 2.8346564524809194e-07, "reference_active": true, "delta_db": 0.3048867195609359}], "active_band_count": 3, "active_band_mae_db": 0.4553505327638782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 1.282880604964887e-05, "log_rms_mae_db": 1.9573831721412978, "rms_envelope_nrmse": 0.4980812201778194, "envelope_correlation": 0.8557973120326602, "rms_gain_error_db": 1.646059097056016, "reference_mean_power": 7.1814170088864185e-06, "output_mean_power": 1.0490979917557418e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 3.966042949113477e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6674623865712373e-06, "output_power": 2.9122061404445873e-06, "reference_active": true, "delta_db": 0.38123809026151567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.255436860296439e-07, "output_power": 3.968633255508562e-07, "reference_active": true, "delta_db": 0.860316907142995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7820276752795505e-08, "output_power": 2.0726388271199847e-08, "reference_active": true, "delta_db": -1.278378163751004}], "active_band_count": 3, "active_band_mae_db": 0.8399777203851716}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9157534942834199, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8d55cfc06c86a3e51e8ae7a555961f7516e093cb7bae3af93bb53f3c8120969e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9884651880208339, "standard_deviation": 0.0011302075749826924, "minimum": 0.9869255434978259, "maximum": 0.9900572106912426, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5262626238152208, "standard_deviation": 0.043105678873281095, "minimum": 1.4672791263391303, "maximum": 1.6050107658573451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.41567560089910793, "standard_deviation": 0.17582519627930634, "minimum": 0.2626743730264492, "maximum": 0.7477192021533607, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12305798550282879, "standard_deviation": 0.006053569123194673, "minimum": 0.11397363757485686, "maximum": 0.1313043232920218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953108642046946, "standard_deviation": 0.0009535460222438862, "minimum": 0.993283316175649, "maximum": 0.9963337486253191, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0703158312313463, "standard_deviation": 0.041151585427499755, "minimum": 1.022784585491781, "maximum": 1.1324201787385415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.33929463028880635, "standard_deviation": 0.20139914547067547, "minimum": 0.13350575346545343, "maximum": 0.7856465677493453, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07829875047948504, "standard_deviation": 0.007863833841551334, "minimum": 0.06989091429491108, "maximum": 0.09389166657386558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8010630598009346, "standard_deviation": 0.011828513509055448, "minimum": 0.7833749157779787, "maximum": 0.8174735365385285, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.188871839386803, "standard_deviation": 0.02379363751430026, "minimum": 2.1526084559820813, "maximum": 2.217737103676756, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.41966433222698984, "standard_deviation": 0.28012760367469264, "minimum": 0.10345409339092322, "maximum": 0.9969479499624597, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.584284336167775, "standard_deviation": 0.005292439326435503, "minimum": 0.5747831101764243, "maximum": 0.5936461805265105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.8397342870874853, "standard_deviation": 0.011254955710274302, "minimum": 0.8225862291405155, "maximum": 0.8557973120326602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.952135357834429, "standard_deviation": 0.03519087508997925, "minimum": 1.8932026685237917, "maximum": 1.9814344715436185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7652810556118381, "standard_deviation": 0.2997006459641328, "minimum": 0.3172705259797484, "maximum": 1.3755703969821915, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.49855665847049513, "standard_deviation": 0.005302873487399695, "minimum": 0.4906763857397548, "maximum": 0.5062391513340875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.988474698912133, "standard_deviation": 0.0020077013079983757, "minimum": 0.9850502379043797, "maximum": 0.9904429644039101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1426986491983708, "standard_deviation": 0.03329592776557054, "minimum": 1.0976317973779801, "maximum": 1.198280671012424, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5607283644050252, "standard_deviation": 0.13436097240676992, "minimum": 0.3073943752685216, "maximum": 0.7847872709978577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10169108312001646, "standard_deviation": 0.008379231178376615, "minimum": 0.09361622628808651, "maximum": 0.11623742211514225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959982540954307, "standard_deviation": 0.0013809344459363947, "minimum": 0.9930515657850055, "maximum": 0.9976558480855755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6445391298187864, "standard_deviation": 0.027016983391300393, "minimum": 0.6148412177434762, "maximum": 0.697711110187621, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5960828728487456, "standard_deviation": 0.24088309914361555, "minimum": 0.12654630368395195, "maximum": 0.9953559560156676, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05671811182636219, "standard_deviation": 0.008325386520691013, "minimum": 0.04737768882189459, "maximum": 0.07521724226024723, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.601514465027603, "standard_deviation": 0.04068620403267285, "minimum": 0.5388927869694288, "maximum": 0.6628312045109687, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.992268365367467, "standard_deviation": 0.008319521656096335, "minimum": 5.97968218491063, "maximum": 6.0018831452146655, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5109426926452895, "standard_deviation": 0.02484394974000403, "minimum": 0.46269122237660604, "maximum": 0.5425267861507407, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.055615443730686755, "standard_deviation": 0.031391008870671464, "minimum": -0.09863980279004105, "maximum": -0.003530676304690011, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.992364445160384, "standard_deviation": 0.0011063897140495852, "minimum": 0.9906948745334497, "maximum": 0.9937031687436745, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8164517029802385, "standard_deviation": 0.019847496270402977, "minimum": 0.7955658369618125, "maximum": 0.8382509507495267, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3636855221577482, "standard_deviation": 0.10484990946182711, "minimum": 0.18668882657993646, "maximum": 0.504318565852982, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07513644783216511, "standard_deviation": 0.005319617968948297, "minimum": 0.06827531639218551, "maximum": 0.08296329542991093, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974650475437188, "standard_deviation": 0.0008239176475670541, "minimum": 0.9959866135484564, "maximum": 0.9984478078064698, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4498221095540752, "standard_deviation": 0.018319912444590237, "minimum": 0.42568166814092856, "maximum": 0.47914256588799137, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3991921298802486, "standard_deviation": 0.15283681101232546, "minimum": 0.17103114832603697, "maximum": 0.580348448965332, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0402914255161444, "standard_deviation": 0.00611733970527577, "minimum": 0.0328284454797898, "maximum": 0.05114589551251116, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4446-2271-0022", "language": "en", "style": "normal", "speaker": "librispeech_4446", "source_utterance_id": "4446-2271-0022", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 37200, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4446-2271-0022.wav", "source_file_sha256": "834518b042180d4f98fd54a9808776f4f376d987ab5e5d7ec8b42e65e6c096c9", "decoded_pcm_sha256": "36c74fdf4f12eb94aa845b16c467769326a4db667ab611a8845b7f72c2cae5b1", "original_source_sha256": "ed740f49d11cc5c58f71f1925854cc9c1f30170b81143873aa8c58cd35430583", "corpus_entry_id": "librispeech_4446-2271-0022"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.7196924686431885, "predicted_mos": 3.7196924686431885, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "0fd3f224cd1563164ea2a06e6f6a8357cc009c3410083ad514c7e2f6e55ae0b6", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.855251627002872, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4446-2271-0022", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "0fd3f224cd1563164ea2a06e6f6a8357cc009c3410083ad514c7e2f6e55ae0b6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0e102c1dba2f0649f1df1e8dfcc4595e3cf32e9e0dcf6815b2f748a1174aa58c"}, "rows": [{"case": "librispeech_4446-2271-0022", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.0299246311187744, "predicted_mos": 3.0299246311187744, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "9791e89e5e4532b566d6c1582579a65c9c805c96012def3e030fce3ed3dad9ac", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7392716989998007, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4446-2271-0022", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9791e89e5e4532b566d6c1582579a65c9c805c96012def3e030fce3ed3dad9ac", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7359298799676555, "全帯域_10ms_rms_envelope_nrmse": 0.05516130943865477, "全帯域_10ms_envelope_correlation": 0.997191427843096, "全帯域_10ms_modulation_mae_db": 0.07734458947759389, "全帯域_40ms_log_rms_mae_db": 0.40001114994991743, "全帯域_40ms_rms_envelope_nrmse": 0.02847597787058268, "全帯域_40ms_envelope_correlation": 0.9992292829360641, "全帯域_40ms_modulation_mae_db": 0.22683047704440307, "300–1000 Hz_10ms_log_rms_mae_db": 1.0183101189976795, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06497727242989321, "300–1000 Hz_10ms_envelope_correlation": 0.9966468712342602, "300–1000 Hz_10ms_modulation_mae_db": 0.6818547073855862, "300–1000 Hz_40ms_log_rms_mae_db": 0.5021395102823685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03156821769394453, "300–1000 Hz_40ms_envelope_correlation": 0.9991982979954368, "300–1000 Hz_40ms_modulation_mae_db": 0.5235313282630037, "1000–3000 Hz_10ms_log_rms_mae_db": 1.491993647550379, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11803683419533217, "1000–3000 Hz_10ms_envelope_correlation": 0.9910280061872757, "1000–3000 Hz_10ms_modulation_mae_db": 1.3784245393584886, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8912044975204294, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08062330328888419, "1000–3000 Hz_40ms_envelope_correlation": 0.9964638032916401, "1000–3000 Hz_40ms_modulation_mae_db": 1.7126806760801347, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7374920090694965, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2086463485149492, "3000–6000 Hz_10ms_envelope_correlation": 0.9703666142586442, "3000–6000 Hz_10ms_modulation_mae_db": 1.3036520331365187, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4398195188695568, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13921703305259447, "3000–6000 Hz_40ms_envelope_correlation": 0.9863286612967364, "3000–6000 Hz_40ms_modulation_mae_db": 1.0645000004906193, "mr_stft_db": 5.618919722533548, "band_mae_db": 0.14636389430646762, "rms_gain_error_db": -0.04752704002277342, "output_peak": 0.47487486004943225}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7751444354412257, "全帯域_10ms_rms_envelope_nrmse": 0.06506169482225956, "全帯域_10ms_envelope_correlation": 0.9961061659792301, "全帯域_10ms_modulation_mae_db": 0.2437067796490537, "全帯域_40ms_log_rms_mae_db": 0.45687659436665795, "全帯域_40ms_rms_envelope_nrmse": 0.03089256590541334, "全帯域_40ms_envelope_correlation": 0.9991228142144425, "全帯域_40ms_modulation_mae_db": 0.1515670805721467, "300–1000 Hz_10ms_log_rms_mae_db": 1.016316313876683, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07342634737709036, "300–1000 Hz_10ms_envelope_correlation": 0.9956969859962135, "300–1000 Hz_10ms_modulation_mae_db": 0.47412654555960093, "300–1000 Hz_40ms_log_rms_mae_db": 0.5362273316322215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036763328852080834, "300–1000 Hz_40ms_envelope_correlation": 0.9988901960029029, "300–1000 Hz_40ms_modulation_mae_db": 0.4321162836401649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.404206275150699, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.123142719246937, "1000–3000 Hz_10ms_envelope_correlation": 0.9892269572050152, "1000–3000 Hz_10ms_modulation_mae_db": 0.4769359725107414, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8614721207802652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07409742066811285, "1000–3000 Hz_40ms_envelope_correlation": 0.9961704199403085, "1000–3000 Hz_40ms_modulation_mae_db": 0.456219748354522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6316330563263581, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15888041262977082, "3000–6000 Hz_10ms_envelope_correlation": 0.9830650399099564, "3000–6000 Hz_10ms_modulation_mae_db": 0.9385154573386659, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2927835165050487, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11532663106622694, "3000–6000 Hz_40ms_envelope_correlation": 0.9913612357633962, "3000–6000 Hz_40ms_modulation_mae_db": 0.6140719245401424, "mr_stft_db": 5.604630633400073, "band_mae_db": 0.12038684063432335, "rms_gain_error_db": -0.0700371743503836, "output_peak": 0.47037361173936176}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.806217902247935, "全帯域_10ms_rms_envelope_nrmse": 0.057214604657742134, "全帯域_10ms_envelope_correlation": 0.9969735093108795, "全帯域_10ms_modulation_mae_db": 0.2922430804118582, "全帯域_40ms_log_rms_mae_db": 0.4532374736740613, "全帯域_40ms_rms_envelope_nrmse": 0.031748166106916234, "全帯域_40ms_envelope_correlation": 0.9990332141846426, "全帯域_40ms_modulation_mae_db": 0.2662957227516984, "300–1000 Hz_10ms_log_rms_mae_db": 1.1029243586210618, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0701631852768688, "300–1000 Hz_10ms_envelope_correlation": 0.9960770130133373, "300–1000 Hz_10ms_modulation_mae_db": 0.3547754674126802, "300–1000 Hz_40ms_log_rms_mae_db": 0.48146607782256345, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04140584980393843, "300–1000 Hz_40ms_envelope_correlation": 0.9985863270318072, "300–1000 Hz_40ms_modulation_mae_db": 0.4027919215848501, "1000–3000 Hz_10ms_log_rms_mae_db": 1.469396197576896, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12099865081846102, "1000–3000 Hz_10ms_envelope_correlation": 0.9894366700803529, "1000–3000 Hz_10ms_modulation_mae_db": 0.1234827523486689, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8580418023300362, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06513932376011285, "1000–3000 Hz_40ms_envelope_correlation": 0.9968098655069523, "1000–3000 Hz_40ms_modulation_mae_db": 0.19991997328459374, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7342881062910867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20370160265105502, "3000–6000 Hz_10ms_envelope_correlation": 0.9713752451439708, "3000–6000 Hz_10ms_modulation_mae_db": 2.1789866072429036, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3377573537613296, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14800523930876877, "3000–6000 Hz_40ms_envelope_correlation": 0.9849733092269353, "3000–6000 Hz_40ms_modulation_mae_db": 1.818391756742666, "mr_stft_db": 5.636139413729357, "band_mae_db": 0.139331734547916, "rms_gain_error_db": -0.04092671372888181, "output_peak": 0.4708750921211354}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7345167266909484, "全帯域_10ms_rms_envelope_nrmse": 0.04692143034993003, "全帯域_10ms_envelope_correlation": 0.9979883139869636, "全帯域_10ms_modulation_mae_db": 0.3995800655600042, "全帯域_40ms_log_rms_mae_db": 0.37170941234050137, "全帯域_40ms_rms_envelope_nrmse": 0.02664289937745548, "全帯域_40ms_envelope_correlation": 0.9993443188765134, "全帯域_40ms_modulation_mae_db": 0.4366084322080268, "300–1000 Hz_10ms_log_rms_mae_db": 0.9992127317648769, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06190319279221664, "300–1000 Hz_10ms_envelope_correlation": 0.9970200068017178, "300–1000 Hz_10ms_modulation_mae_db": 0.8261140026496463, "300–1000 Hz_40ms_log_rms_mae_db": 0.520435975876543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03569776988864613, "300–1000 Hz_40ms_envelope_correlation": 0.9990283211255309, "300–1000 Hz_40ms_modulation_mae_db": 0.8981924476294344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.413917504113741, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10138824562016488, "1000–3000 Hz_10ms_envelope_correlation": 0.9927134516053173, "1000–3000 Hz_10ms_modulation_mae_db": 0.608068582927622, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7713326249341597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052256003318808524, "1000–3000 Hz_40ms_envelope_correlation": 0.998148248102593, "1000–3000 Hz_40ms_modulation_mae_db": 0.6348723383754252, "3000–6000 Hz_10ms_log_rms_mae_db": 1.747196610663819, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1687278683220968, "3000–6000 Hz_10ms_envelope_correlation": 0.9805821031299816, "3000–6000 Hz_10ms_modulation_mae_db": 1.848967901069182, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3220141623386872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12273064632491805, "3000–6000 Hz_40ms_envelope_correlation": 0.9898861463167253, "3000–6000 Hz_40ms_modulation_mae_db": 2.2398359342373224, "mr_stft_db": 5.628937315381002, "band_mae_db": 0.12197184259236221, "rms_gain_error_db": -0.05689888101785999, "output_peak": 0.4758267374522612}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7601249576973231, "全帯域_10ms_rms_envelope_nrmse": 0.047337499431963755, "全帯域_10ms_envelope_correlation": 0.9980872703785477, "全帯域_10ms_modulation_mae_db": 0.05530746272434767, "全帯域_40ms_log_rms_mae_db": 0.387444510583894, "全帯域_40ms_rms_envelope_nrmse": 0.024279153382967628, "全帯域_40ms_envelope_correlation": 0.9996196092778753, "全帯域_40ms_modulation_mae_db": 0.2593754264512398, "300–1000 Hz_10ms_log_rms_mae_db": 1.0448855500752556, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05777928794953354, "300–1000 Hz_10ms_envelope_correlation": 0.9975237398421594, "300–1000 Hz_10ms_modulation_mae_db": 0.34421216293343954, "300–1000 Hz_40ms_log_rms_mae_db": 0.5275363629927202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03346364863733841, "300–1000 Hz_40ms_envelope_correlation": 0.9992859332761695, "300–1000 Hz_40ms_modulation_mae_db": 0.25246343007355443, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4438266657395986, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10860541957049287, "1000–3000 Hz_10ms_envelope_correlation": 0.9923173713133995, "1000–3000 Hz_10ms_modulation_mae_db": 1.5527159817022322, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7874926095198587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0637520433108491, "1000–3000 Hz_40ms_envelope_correlation": 0.9979961679116897, "1000–3000 Hz_40ms_modulation_mae_db": 1.39476622401648, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6960778133096406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1852479888913811, "3000–6000 Hz_10ms_envelope_correlation": 0.9763836342056172, "3000–6000 Hz_10ms_modulation_mae_db": 1.4304046709782512, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3922860896447076, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12893871453793895, "3000–6000 Hz_40ms_envelope_correlation": 0.9886471825425984, "3000–6000 Hz_40ms_modulation_mae_db": 1.1416970846308265, "mr_stft_db": 5.632814633807153, "band_mae_db": 0.1798503522486673, "rms_gain_error_db": -0.12357806350414158, "output_peak": 0.46704589192374085}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8312354497756312, "全帯域_10ms_rms_envelope_nrmse": 0.05408655030665181, "全帯域_10ms_envelope_correlation": 0.9974363762316061, "全帯域_10ms_modulation_mae_db": 0.4442472375202507, "全帯域_40ms_log_rms_mae_db": 0.4204721236906446, "全帯域_40ms_rms_envelope_nrmse": 0.03227969509761724, "全帯域_40ms_envelope_correlation": 0.9991593236050665, "全帯域_40ms_modulation_mae_db": 0.5651687947060702, "300–1000 Hz_10ms_log_rms_mae_db": 1.0509721439164488, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06712199267476884, "300–1000 Hz_10ms_envelope_correlation": 0.996663106388833, "300–1000 Hz_10ms_modulation_mae_db": 0.5955409987698712, "300–1000 Hz_40ms_log_rms_mae_db": 0.5357009453078585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04203591047468284, "300–1000 Hz_40ms_envelope_correlation": 0.9988156750435752, "300–1000 Hz_40ms_modulation_mae_db": 0.9012829571253359, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4864625511831642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10807815535840536, "1000–3000 Hz_10ms_envelope_correlation": 0.991660526748643, "1000–3000 Hz_10ms_modulation_mae_db": 0.6318436720279149, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8285258638223255, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055428573589888334, "1000–3000 Hz_40ms_envelope_correlation": 0.9978530889278504, "1000–3000 Hz_40ms_modulation_mae_db": 0.5759637416189486, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7102074194247316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18789003433467388, "3000–6000 Hz_10ms_envelope_correlation": 0.9759104051859594, "3000–6000 Hz_10ms_modulation_mae_db": 1.107137392303671, "3000–6000 Hz_40ms_log_rms_mae_db": 1.298946650321319, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12474142268463465, "3000–6000 Hz_40ms_envelope_correlation": 0.9898743093101394, "3000–6000 Hz_40ms_modulation_mae_db": 1.2889868724138525, "mr_stft_db": 5.631021023692593, "band_mae_db": 0.15456460285612236, "rms_gain_error_db": -0.12310568311418196, "output_peak": 0.4663279074491576}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7811935634394744, "全帯域_10ms_rms_envelope_nrmse": 0.0488838546856656, "全帯域_10ms_envelope_correlation": 0.9978407901717355, "全帯域_10ms_modulation_mae_db": 0.48226560854311634, "全帯域_40ms_log_rms_mae_db": 0.4173318295748794, "全帯域_40ms_rms_envelope_nrmse": 0.02460236630180772, "全帯域_40ms_envelope_correlation": 0.9994806770266038, "全帯域_40ms_modulation_mae_db": 0.39764609281975033, "300–1000 Hz_10ms_log_rms_mae_db": 1.031906416843602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.054506331796428706, "300–1000 Hz_10ms_envelope_correlation": 0.9977162502145192, "300–1000 Hz_10ms_modulation_mae_db": 0.6118431941285479, "300–1000 Hz_40ms_log_rms_mae_db": 0.48977039698962616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0281800669725451, "300–1000 Hz_40ms_envelope_correlation": 0.9994342922764541, "300–1000 Hz_40ms_modulation_mae_db": 0.6001032404896133, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4890285336143452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11935757475749664, "1000–3000 Hz_10ms_envelope_correlation": 0.9898079294944709, "1000–3000 Hz_10ms_modulation_mae_db": 0.5053706083737052, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8106487416087117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06422036415955476, "1000–3000 Hz_40ms_envelope_correlation": 0.9970775818800908, "1000–3000 Hz_40ms_modulation_mae_db": 0.6938200816035774, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6596906819719257, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16765982233367566, "3000–6000 Hz_10ms_envelope_correlation": 0.9809523394845626, "3000–6000 Hz_10ms_modulation_mae_db": 1.4373172154401208, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3234115361350485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12047377388702642, "3000–6000 Hz_40ms_envelope_correlation": 0.9904504693433592, "3000–6000 Hz_40ms_modulation_mae_db": 1.4060241500140511, "mr_stft_db": 5.623003685402632, "band_mae_db": 0.06257672535572065, "rms_gain_error_db": -0.07800414419674258, "output_peak": 0.47252662580606425}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7900841765534641, "全帯域_10ms_rms_envelope_nrmse": 0.05433956529872974, "全帯域_10ms_envelope_correlation": 0.9974402793794582, "全帯域_10ms_modulation_mae_db": 0.8468566464309454, "全帯域_40ms_log_rms_mae_db": 0.4758982759844522, "全帯域_40ms_rms_envelope_nrmse": 0.03190642428442514, "全帯域_40ms_envelope_correlation": 0.9992189505078519, "全帯域_40ms_modulation_mae_db": 0.9451139198304688, "300–1000 Hz_10ms_log_rms_mae_db": 0.99491461031702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06674984568897445, "300–1000 Hz_10ms_envelope_correlation": 0.9966207993520355, "300–1000 Hz_10ms_modulation_mae_db": 0.9924869166118088, "300–1000 Hz_40ms_log_rms_mae_db": 0.5238190781269242, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0393809756887674, "300–1000 Hz_40ms_envelope_correlation": 0.9989147239484153, "300–1000 Hz_40ms_modulation_mae_db": 1.2140782114309963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4708770598083616, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12484958586536991, "1000–3000 Hz_10ms_envelope_correlation": 0.9896887072419759, "1000–3000 Hz_10ms_modulation_mae_db": 1.0807784708050976, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8660107509008719, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07689191317036294, "1000–3000 Hz_40ms_envelope_correlation": 0.9968104695755458, "1000–3000 Hz_40ms_modulation_mae_db": 1.3413947256575807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.696028221081471, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17109534748250482, "3000–6000 Hz_10ms_envelope_correlation": 0.9798578803724669, "3000–6000 Hz_10ms_modulation_mae_db": 0.7853206639824903, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3082880805443358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11180508494064588, "3000–6000 Hz_40ms_envelope_correlation": 0.9914937234517744, "3000–6000 Hz_40ms_modulation_mae_db": 1.0772943121783949, "mr_stft_db": 5.623899622106649, "band_mae_db": 0.1779151747686948, "rms_gain_error_db": -0.13459177872693293, "output_peak": 0.4620961067924141}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37200, "output_samples": 37200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.521377316057349, "spectral_convergence": 0.15064019759934874, "spectral_convergence_status": "ok", "amplitude_floor": 1.3482074212160965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.455610829677773, "spectral_convergence": 0.1484420521058112, "spectral_convergence_status": "ok", "amplitude_floor": 1.080252897815134e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.879771021865523, "spectral_convergence": 0.14763432906477267, "spectral_convergence_status": "ok", "amplitude_floor": 8.596818112459857e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004510568589627121, "output_power": 0.0004474205076572331, "delta_db": -0.0351540503466792, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019377343288488795, "output_power": 0.0019080329514415467, "delta_db": -0.06708362647144128, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013786956670426643, "output_power": 0.001396813632792903, "delta_db": 0.05670054369485747, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031975863120548915, "output_power": 0.00028887979355290483, "delta_db": -0.44105111114955187, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.4830953182024825e-05, "output_power": 1.5092278574669723e-05, "delta_db": 0.07585748916509964, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.03312086795025e-05, "output_power": 2.1300845997192303e-05, "delta_db": 0.2023365450111764, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14636389430646762, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04752704002277342, "pitch": null, "output_rms_dbfs": -23.89601768842165, "output_peak": 0.47487486004943225, "reference_rms_dbfs": -23.848490648398872, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.618919722533548}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00019668187909197959, "log_rms_mae_db": 0.7359298799676555, "rms_envelope_nrmse": 0.05516130943865477, "envelope_correlation": 0.997191427843096, "rms_gain_error_db": -0.0475270425656379, "reference_mean_power": 0.004131292024738224, "output_mean_power": 0.004086327688277753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.0026327800860698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011612269501705267, "output_power": 0.001183883924208247, "reference_active": true, "delta_db": 0.08392016839226488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017095886275955477, "output_power": 0.00016898098355013435, "reference_active": true, "delta_db": -0.05053786598094922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.217923048678998e-05, "output_power": 4.1242129239489223e-05, "reference_active": true, "delta_db": -0.09757573405956754}], "active_band_count": 3, "active_band_mae_db": 0.07734458947759389}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00018115923356648065, "log_rms_mae_db": 0.40001114994991743, "rms_envelope_nrmse": 0.02847597787058268, "envelope_correlation": 0.9992292829360641, "rms_gain_error_db": -0.04752694573916304, "reference_mean_power": 0.004185412442455258, "output_mean_power": 0.004139859160128859}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.7728315536386115e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066252574844359, "output_power": 0.0010841771990899032, "reference_active": true, "delta_db": 0.07240176627200887}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.440983264620606e-05, "output_power": 8.643005100033999e-05, "reference_active": true, "delta_db": 0.10271730043526386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.940091323355611e-06, "output_power": 3.3029200158480994e-06, "reference_active": true, "delta_db": 0.5053723644259365}], "active_band_count": 3, "active_band_mae_db": 0.22683047704440307}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.000176021246464811, "log_rms_mae_db": 1.0183101189976795, "rms_envelope_nrmse": 0.06497727242989321, "envelope_correlation": 0.9966468712342602, "rms_gain_error_db": -0.061591185896671785, "reference_mean_power": 0.0026395436088586166, "output_mean_power": 0.0026023740696949588}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.515899729654865e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009091516678957249, "output_power": 0.0009086731917869532, "reference_active": true, "delta_db": -0.0022862437759513976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001833276731517994, "output_power": 0.00016851539508271366, "reference_active": true, "delta_db": -0.3658844336558598}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7517429693718724e-05, "output_power": 2.5497272935398698e-05, "reference_active": true, "delta_db": -1.6773934447249474}], "active_band_count": 3, "active_band_mae_db": 0.6818547073855862}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00016136314602147107, "log_rms_mae_db": 0.5021395102823685, "rms_envelope_nrmse": 0.03156821769394453, "envelope_correlation": 0.9991982979954368, "rms_gain_error_db": -0.06159117539717099, "reference_mean_power": 0.002674122622482191, "output_mean_power": 0.0026364661547553635}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.348456388983372e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008380618007305507, "output_power": 0.0008339350811560258, "reference_active": true, "delta_db": -0.021438020587436418}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.026333057603631e-05, "output_power": 8.744004020109835e-05, "reference_active": true, "delta_db": -0.1380100565715414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498114091574834e-06, "output_power": 1.8050850878087215e-06, "reference_active": true, "delta_db": -1.4111459076300337}], "active_band_count": 3, "active_band_mae_db": 0.5235313282630037}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 6.667691377903701e-05, "log_rms_mae_db": 1.491993647550379, "rms_envelope_nrmse": 0.11803683419533217, "envelope_correlation": 0.9910280061872757, "rms_gain_error_db": -0.3987896420352344, "reference_mean_power": 0.0003102146198149334, "output_mean_power": 0.0002829979549860604}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.247013787334141e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018435630971779633, "output_power": 0.00016583451238194754, "reference_active": true, "delta_db": -0.4598308809936105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0328682781544803e-05, "output_power": 1.452524179352524e-05, "reference_active": true, "delta_db": -1.459858684020421}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.580926194982421e-06, "output_power": 2.7503938576154533e-06, "reference_active": true, "delta_db": -2.215584053061434}], "active_band_count": 3, "active_band_mae_db": 1.3784245393584886}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 6.131102329922278e-05, "log_rms_mae_db": 0.8912044975204294, "rms_envelope_nrmse": 0.08062330328888419, "envelope_correlation": 0.9964638032916401, "rms_gain_error_db": -0.3987893660758828, "reference_mean_power": 0.0003142782197045969, "output_mean_power": 0.00028670505335161463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 3.052275331975406e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017569545994603763, "output_power": 0.00015777094679361118, "reference_active": true, "delta_db": -0.46733507494934223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.589536328721284e-06, "output_power": 5.880948666082084e-06, "reference_active": true, "delta_db": -1.6452233231065212}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8479520444652144e-07, "output_power": 1.4190062525736478e-07, "reference_active": true, "delta_db": -3.0254836301845414}], "active_band_count": 3, "active_band_mae_db": 1.7126806760801347}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 2.0872644535346135e-05, "log_rms_mae_db": 1.7374920090694965, "rms_envelope_nrmse": 0.2086463485149492, "envelope_correlation": 0.9703666142586442, "rms_gain_error_db": 0.06638084693810016, "reference_mean_power": 1.984318614204444e-05, "output_mean_power": 2.0148814152375194e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.6220002221112295e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.352365504647456e-06, "output_power": 9.196155218717084e-06, "reference_active": true, "delta_db": -0.07315178408211973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.791249366368675e-07, "output_power": 8.392059746820158e-07, "reference_active": true, "delta_db": -0.6696954431288777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495952213974179e-07, "output_power": 5.176630342122497e-07, "reference_active": true, "delta_db": 3.1681088721985584}], "active_band_count": 3, "active_band_mae_db": 1.3036520331365187}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 1.9522833601394496e-05, "log_rms_mae_db": 1.4398195188695568, "rms_envelope_nrmse": 0.13921703305259447, "envelope_correlation": 0.9863286612967364, "rms_gain_error_db": 0.06638059298543642, "reference_mean_power": 2.010291979320635e-05, "output_mean_power": 2.0412547070250182e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.5131001811961146e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.175181026763155e-06, "output_power": 8.252011644608016e-06, "reference_active": true, "delta_db": 0.0406245392507386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.283226898133603e-07, "output_power": 3.5518705146870964e-07, "reference_active": true, "delta_db": -0.813139562061375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.754857048422832e-08, "output_power": 3.007567162340804e-08, "reference_active": true, "delta_db": 2.339735900159744}], "active_band_count": 3, "active_band_mae_db": 1.0645000004906193}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9426581018490227, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "33e85349250927c25e840598398ca808b8b405a5bd92c0d5bed3ae5871b26c19", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9907349524845563, "standard_deviation": 0.0013770913450197582, "minimum": 0.9892269572050152, "maximum": 0.9927134516053173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.458713554342148, "standard_deviation": 0.03434694451537923, "minimum": 1.404206275150699, "maximum": 1.491993647550379, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7947025725068089, "standard_deviation": 0.4918964749879765, "minimum": 0.1234827523486689, "maximum": 1.5527159817022322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11555714817908248, "standard_deviation": 0.008444034127746186, "minimum": 0.10138824562016488, "maximum": 0.12484958586536991, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971662056420838, "standard_deviation": 0.0007436773733450688, "minimum": 0.9961704199403085, "maximum": 0.998148248102593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8343411264270824, "standard_deviation": 0.04190844127993097, "minimum": 0.7713326249341597, "maximum": 0.8912044975204294, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8762046886239079, "standard_deviation": 0.5345757135740549, "minimum": 0.19991997328459374, "maximum": 1.7126806760801347, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0665511181583217, "standard_deviation": 0.010037265001541245, "minimum": 0.052256003318808524, "maximum": 0.08062330328888419, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9773116577113949, "standard_deviation": 0.004626544415041978, "minimum": 0.9703666142586442, "maximum": 0.9830650399099564, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7015767397673163, "standard_deviation": 0.04004138263558436, "minimum": 1.6316330563263581, "maximum": 1.747196610663819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3787877426864754, "standard_deviation": 0.4618224391480813, "minimum": 0.7853206639824903, "maximum": 2.1789866072429036, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1814811781450134, "standard_deviation": 0.017965414852047636, "minimum": 0.15888041262977082, "maximum": 0.2086463485149492, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9891268796564581, "standard_deviation": 0.002354133325349305, "minimum": 0.9849733092269353, "maximum": 0.9914937234517744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3394133635150043, "standard_deviation": 0.05101699296693954, "minimum": 1.2927835165050487, "maximum": 1.4398195188695568, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3313502544059843, "standard_deviation": 0.5008758770355352, "minimum": 0.6140719245401424, "maximum": 2.2398359342373224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12640481822534427, "standard_deviation": 0.01209766107159579, "minimum": 0.11180508494064588, "maximum": 0.14800523930876877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9967455966053844, "standard_deviation": 0.0006763019401876231, "minimum": 0.9956969859962135, "maximum": 0.9977162502145192, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0324302805515786, "standard_deviation": 0.034720298420485095, "minimum": 0.99491461031702, "maximum": 1.1029243586210618, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6101192494313976, "standard_deviation": 0.2244425732470129, "minimum": 0.34421216293343954, "maximum": 0.9924869166118088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06457843199822182, "standard_deviation": 0.006275250192345186, "minimum": 0.054506331796428706, "maximum": 0.07342634737709036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9990192208375365, "standard_deviation": 0.0002756722207665061, "minimum": 0.9985863270318072, "maximum": 0.9994342922764541, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5146369598788532, "standard_deviation": 0.020938719192961627, "minimum": 0.48146607782256345, "maximum": 0.5362273316322215, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6530699775296192, "standard_deviation": 0.32248054868800136, "minimum": 0.25246343007355443, "maximum": 1.2140782114309963, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03606197100149296, "standard_deviation": 0.004854377110293712, "minimum": 0.0281800669725451, "maximum": 0.04203591047468284, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1378701459137843, "standard_deviation": 0.037695107956435686, "minimum": 0.06257672535572065, "maximum": 0.1798503522486673, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.624920756256627, "standard_deviation": 0.009956598351661204, "minimum": 5.604630633400073, "maximum": 5.636139413729357, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46999335416669596, "standard_deviation": 0.004624261060406728, "minimum": 0.4620961067924141, "maximum": 0.4758267374522612, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08433368483273723, "standard_deviation": 0.037426654194746695, "minimum": -0.13459177872693293, "maximum": -0.04092671372888181, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973830166601896, "standard_deviation": 0.0006450682366801062, "minimum": 0.9961061659792301, "maximum": 0.9980872703785477, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7768058864767071, "standard_deviation": 0.03330260280932643, "minimum": 0.7345167266909484, "maximum": 0.8312354497756312, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35519393378964625, "standard_deviation": 0.2539134842161003, "minimum": 0.05530746272434767, "maximum": 0.8468566464309454, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05362581362394967, "standard_deviation": 0.006020395004145258, "minimum": 0.04692143034993003, "maximum": 0.06506169482225956, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992760238286325, "standard_deviation": 0.0001949684341463722, "minimum": 0.9990332141846426, "maximum": 0.9996196092778753, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42287267127062605, "standard_deviation": 0.036492561106265496, "minimum": 0.37170941234050137, "maximum": 0.4758982759844522, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4060757432979755, "standard_deviation": 0.25490587081977956, "minimum": 0.1515670805721467, "maximum": 0.9451139198304688, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02885340604089818, "standard_deviation": 0.0033299957928576342, "minimum": 0.024279153382967628, "maximum": 0.03227969509761724, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0022", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.1984798908233643, "predicted_mos": 3.1984798908233643, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "f2cbbcc07ebaee96e735846eb16d2da33000ccb91fa6d5259bbf21754f7e2698", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8686807001940906, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7487609590083476, "全帯域_10ms_rms_envelope_nrmse": 0.052562506181304675, "全帯域_10ms_envelope_correlation": 0.997473971403217, "全帯域_10ms_modulation_mae_db": 0.05841456831359362, "全帯域_40ms_log_rms_mae_db": 0.4066552197712254, "全帯域_40ms_rms_envelope_nrmse": 0.02760974633741073, "全帯域_40ms_envelope_correlation": 0.999303379537847, "全帯域_40ms_modulation_mae_db": 0.2620463846051006, "300–1000 Hz_10ms_log_rms_mae_db": 1.0490158599319606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06641220685702405, "300–1000 Hz_10ms_envelope_correlation": 0.9965046834651662, "300–1000 Hz_10ms_modulation_mae_db": 0.643347666500701, "300–1000 Hz_40ms_log_rms_mae_db": 0.5167595284588804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.032948640150701515, "300–1000 Hz_40ms_envelope_correlation": 0.9991357494296184, "300–1000 Hz_40ms_modulation_mae_db": 0.4607408951388299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4836162796381729, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11831093375699525, "1000–3000 Hz_10ms_envelope_correlation": 0.9910572114722442, "1000–3000 Hz_10ms_modulation_mae_db": 1.3135682160708109, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8627566371500606, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08302496041079543, "1000–3000 Hz_40ms_envelope_correlation": 0.9962502275527846, "1000–3000 Hz_40ms_modulation_mae_db": 1.7328728854735207, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5763904202965713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1670743257333871, "3000–6000 Hz_10ms_envelope_correlation": 0.9807777766191854, "3000–6000 Hz_10ms_modulation_mae_db": 1.1631134714824223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3128421586071575, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1165354185691929, "3000–6000 Hz_40ms_envelope_correlation": 0.9903140951360877, "3000–6000 Hz_40ms_modulation_mae_db": 0.8734333956875314, "mr_stft_db": 5.5322417778132085, "band_mae_db": 0.12261744680131122, "rms_gain_error_db": -0.06417381197133798, "output_peak": 0.4747125109098952}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8142225116225141, "全帯域_10ms_rms_envelope_nrmse": 0.06902463336568361, "全帯域_10ms_envelope_correlation": 0.9955976205167093, "全帯域_10ms_modulation_mae_db": 0.25881306687918754, "全帯域_40ms_log_rms_mae_db": 0.463754141573952, "全帯域_40ms_rms_envelope_nrmse": 0.031492431485087964, "全帯域_40ms_envelope_correlation": 0.9990718583244634, "全帯域_40ms_modulation_mae_db": 0.14334184713227527, "300–1000 Hz_10ms_log_rms_mae_db": 1.04435564717698, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07654778714252752, "300–1000 Hz_10ms_envelope_correlation": 0.9953220953914427, "300–1000 Hz_10ms_modulation_mae_db": 0.5092595110379091, "300–1000 Hz_40ms_log_rms_mae_db": 0.5470464537547542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03684755522798656, "300–1000 Hz_40ms_envelope_correlation": 0.9988869111434003, "300–1000 Hz_40ms_modulation_mae_db": 0.4566014397024422, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4018463508613863, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12555502967487692, "1000–3000 Hz_10ms_envelope_correlation": 0.9887752539874443, "1000–3000 Hz_10ms_modulation_mae_db": 0.4398431082891973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8429644521152118, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07671058711455507, "1000–3000 Hz_40ms_envelope_correlation": 0.9958568316285477, "1000–3000 Hz_40ms_modulation_mae_db": 0.4577395206867452, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5581659240005243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15425951153081788, "3000–6000 Hz_10ms_envelope_correlation": 0.9846769063391542, "3000–6000 Hz_10ms_modulation_mae_db": 0.7588417298585645, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3057398272986123, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1200085950607258, "3000–6000 Hz_40ms_envelope_correlation": 0.9911919904196228, "3000–6000 Hz_40ms_modulation_mae_db": 0.649894701301085, "mr_stft_db": 5.538203743700174, "band_mae_db": 0.15222005087997634, "rms_gain_error_db": -0.06357521812527796, "output_peak": 0.4691690921610525}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8438400358441719, "全帯域_10ms_rms_envelope_nrmse": 0.060940538248688426, "全帯域_10ms_envelope_correlation": 0.9965600930370423, "全帯域_10ms_modulation_mae_db": 0.3376723025494961, "全帯域_40ms_log_rms_mae_db": 0.47560932488978186, "全帯域_40ms_rms_envelope_nrmse": 0.0359132123167789, "全帯域_40ms_envelope_correlation": 0.9987483030142864, "全帯域_40ms_modulation_mae_db": 0.2904784237513196, "300–1000 Hz_10ms_log_rms_mae_db": 1.1250998480697216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07546378774514197, "300–1000 Hz_10ms_envelope_correlation": 0.9954475243887073, "300–1000 Hz_10ms_modulation_mae_db": 0.3088945477754141, "300–1000 Hz_40ms_log_rms_mae_db": 0.5088327381451965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04650826688470546, "300–1000 Hz_40ms_envelope_correlation": 0.9981947887910276, "300–1000 Hz_40ms_modulation_mae_db": 0.3855883640516478, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4109483812550825, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10881845280433279, "1000–3000 Hz_10ms_envelope_correlation": 0.9914725838548337, "1000–3000 Hz_10ms_modulation_mae_db": 0.19125315865621517, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8225517306519257, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05675873484337507, "1000–3000 Hz_40ms_envelope_correlation": 0.9976090237226928, "1000–3000 Hz_40ms_modulation_mae_db": 0.12813544299759538, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6067619056875662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1769719275243732, "3000–6000 Hz_10ms_envelope_correlation": 0.9794200564801648, "3000–6000 Hz_10ms_modulation_mae_db": 1.92993525862273, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2607997361254348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13407865296902005, "3000–6000 Hz_40ms_envelope_correlation": 0.9888051540995926, "3000–6000 Hz_40ms_modulation_mae_db": 1.6871488159237622, "mr_stft_db": 5.564048151104938, "band_mae_db": 0.11145305613909758, "rms_gain_error_db": -0.03214311698425731, "output_peak": 0.4696103169544389}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7424353818337026, "全帯域_10ms_rms_envelope_nrmse": 0.04883174190552702, "全帯域_10ms_envelope_correlation": 0.9978203917429833, "全帯域_10ms_modulation_mae_db": 0.348219604465095, "全帯域_40ms_log_rms_mae_db": 0.3776042824083426, "全帯域_40ms_rms_envelope_nrmse": 0.02738071593016111, "全帯域_40ms_envelope_correlation": 0.9993085599247507, "全帯域_40ms_modulation_mae_db": 0.4269715277339345, "300–1000 Hz_10ms_log_rms_mae_db": 1.044001636075665, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06324010616603264, "300–1000 Hz_10ms_envelope_correlation": 0.9968749328820435, "300–1000 Hz_10ms_modulation_mae_db": 0.7518964873212323, "300–1000 Hz_40ms_log_rms_mae_db": 0.5277820236856843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035552724150556235, "300–1000 Hz_40ms_envelope_correlation": 0.9990281756619577, "300–1000 Hz_40ms_modulation_mae_db": 0.8357705248561219, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3891277016447507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09736583895423329, "1000–3000 Hz_10ms_envelope_correlation": 0.9933259945772629, "1000–3000 Hz_10ms_modulation_mae_db": 0.5702511837402721, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7721974845771143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05294510466452595, "1000–3000 Hz_40ms_envelope_correlation": 0.998133070324286, "1000–3000 Hz_40ms_modulation_mae_db": 0.5812315667131163, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5685070048857905, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1637821270132526, "3000–6000 Hz_10ms_envelope_correlation": 0.9819157485788284, "3000–6000 Hz_10ms_modulation_mae_db": 1.8733762084484222, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2274385792152556, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11859175541886172, "3000–6000 Hz_40ms_envelope_correlation": 0.9907775739007261, "3000–6000 Hz_40ms_modulation_mae_db": 2.3627694516950286, "mr_stft_db": 5.5543179948031245, "band_mae_db": 0.07587426550192934, "rms_gain_error_db": -0.05738906999957793, "output_peak": 0.4755511038908124}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7812993460827222, "全帯域_10ms_rms_envelope_nrmse": 0.04893036313067437, "全帯域_10ms_envelope_correlation": 0.9979499220593808, "全帯域_10ms_modulation_mae_db": 0.09842138236300126, "全帯域_40ms_log_rms_mae_db": 0.40406214852668826, "全帯域_40ms_rms_envelope_nrmse": 0.025343544649362046, "全帯域_40ms_envelope_correlation": 0.9995752944319996, "全帯域_40ms_modulation_mae_db": 0.26597867979958667, "300–1000 Hz_10ms_log_rms_mae_db": 1.0231571198458829, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05792631271526937, "300–1000 Hz_10ms_envelope_correlation": 0.9975054457142984, "300–1000 Hz_10ms_modulation_mae_db": 0.31968432065095115, "300–1000 Hz_40ms_log_rms_mae_db": 0.5180132989545483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03272664789759545, "300–1000 Hz_40ms_envelope_correlation": 0.9993223200097573, "300–1000 Hz_40ms_modulation_mae_db": 0.29120184928191367, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3865703589047271, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10491784158227127, "1000–3000 Hz_10ms_envelope_correlation": 0.9928061242715545, "1000–3000 Hz_10ms_modulation_mae_db": 1.3968699362859989, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7670334344424335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06285718241736862, "1000–3000 Hz_40ms_envelope_correlation": 0.9979785608083215, "1000–3000 Hz_40ms_modulation_mae_db": 1.323390335276632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5684340695483021, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1699212013038108, "3000–6000 Hz_10ms_envelope_correlation": 0.980876744242438, "3000–6000 Hz_10ms_modulation_mae_db": 0.8815160787531408, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3101180787092572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12973768318197873, "3000–6000 Hz_40ms_envelope_correlation": 0.9891909802448622, "3000–6000 Hz_40ms_modulation_mae_db": 0.46106329753084546, "mr_stft_db": 5.546598210262708, "band_mae_db": 0.23175881472063486, "rms_gain_error_db": -0.1251354593995665, "output_peak": 0.4710121905187557}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8368835377184018, "全帯域_10ms_rms_envelope_nrmse": 0.05310311630965209, "全帯域_10ms_envelope_correlation": 0.9975558532796414, "全帯域_10ms_modulation_mae_db": 0.3629473127340992, "全帯域_40ms_log_rms_mae_db": 0.42833610846567816, "全帯域_40ms_rms_envelope_nrmse": 0.03320279020659758, "全帯域_40ms_envelope_correlation": 0.9991221766968156, "全帯域_40ms_modulation_mae_db": 0.4991964182773494, "300–1000 Hz_10ms_log_rms_mae_db": 1.0625809859238726, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06902305684996252, "300–1000 Hz_10ms_envelope_correlation": 0.9964599527887245, "300–1000 Hz_10ms_modulation_mae_db": 0.517027201581968, "300–1000 Hz_40ms_log_rms_mae_db": 0.566281614217262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043365236991730435, "300–1000 Hz_40ms_envelope_correlation": 0.9987279950577711, "300–1000 Hz_40ms_modulation_mae_db": 0.8444564070813875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.467928089432974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10506316867791454, "1000–3000 Hz_10ms_envelope_correlation": 0.9921586759553813, "1000–3000 Hz_10ms_modulation_mae_db": 0.7019244193003095, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7921917238281443, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05614107771055402, "1000–3000 Hz_40ms_envelope_correlation": 0.9978324254129541, "1000–3000 Hz_40ms_modulation_mae_db": 0.7657762807357368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.555535273552747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18555778793788993, "3000–6000 Hz_10ms_envelope_correlation": 0.976972870831057, "3000–6000 Hz_10ms_modulation_mae_db": 1.2764411049284783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2685493812377968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1354140934298831, "3000–6000 Hz_40ms_envelope_correlation": 0.988330715329931, "3000–6000 Hz_40ms_modulation_mae_db": 1.582894473231594, "mr_stft_db": 5.54376891378418, "band_mae_db": 0.19133202336494545, "rms_gain_error_db": -0.13044512653740373, "output_peak": 0.4634876816894676}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7578976312905605, "全帯域_10ms_rms_envelope_nrmse": 0.046350892564809724, "全帯域_10ms_envelope_correlation": 0.9980350770799121, "全帯域_10ms_modulation_mae_db": 0.43931904965455043, "全帯域_40ms_log_rms_mae_db": 0.4022722467358314, "全帯域_40ms_rms_envelope_nrmse": 0.021883139803291707, "全帯域_40ms_envelope_correlation": 0.9995693107058153, "全帯域_40ms_modulation_mae_db": 0.3570810498364048, "300–1000 Hz_10ms_log_rms_mae_db": 1.0373870295194159, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05414337834764288, "300–1000 Hz_10ms_envelope_correlation": 0.9977093715473705, "300–1000 Hz_10ms_modulation_mae_db": 0.5835260038222841, "300–1000 Hz_40ms_log_rms_mae_db": 0.4760618555654064, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.026124080249248727, "300–1000 Hz_40ms_envelope_correlation": 0.999483424003149, "300–1000 Hz_40ms_modulation_mae_db": 0.5528822511379171, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4349394672148537, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12343567635162019, "1000–3000 Hz_10ms_envelope_correlation": 0.9890101383786764, "1000–3000 Hz_10ms_modulation_mae_db": 0.19762508533390444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8020840614361956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0643231920893239, "1000–3000 Hz_40ms_envelope_correlation": 0.9969391555026828, "1000–3000 Hz_40ms_modulation_mae_db": 0.4744750570004967, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5172913216026631, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16548982893146755, "3000–6000 Hz_10ms_envelope_correlation": 0.9816527619354841, "3000–6000 Hz_10ms_modulation_mae_db": 1.4287712593473263, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2822527958117285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12178586840057719, "3000–6000 Hz_40ms_envelope_correlation": 0.9904193546487935, "3000–6000 Hz_40ms_modulation_mae_db": 1.470773066667576, "mr_stft_db": 5.541348263875793, "band_mae_db": 0.06590500597422043, "rms_gain_error_db": -0.057274474268720704, "output_peak": 0.4701211214471054}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7999897686149674, "全帯域_10ms_rms_envelope_nrmse": 0.05407437833998409, "全帯域_10ms_envelope_correlation": 0.9974402630494592, "全帯域_10ms_modulation_mae_db": 0.7265066139275499, "全帯域_40ms_log_rms_mae_db": 0.4689857547176353, "全帯域_40ms_rms_envelope_nrmse": 0.030938393584909626, "全帯域_40ms_envelope_correlation": 0.9992491127851097, "全帯域_40ms_modulation_mae_db": 0.8336618342946868, "300–1000 Hz_10ms_log_rms_mae_db": 1.0666458301225727, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06689043123407003, "300–1000 Hz_10ms_envelope_correlation": 0.9965770850591849, "300–1000 Hz_10ms_modulation_mae_db": 0.9038701119727893, "300–1000 Hz_40ms_log_rms_mae_db": 0.5783859648604758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03821106007597169, "300–1000 Hz_40ms_envelope_correlation": 0.9989578299122854, "300–1000 Hz_40ms_modulation_mae_db": 1.1400122259199759, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4532033945441882, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1259622807313844, "1000–3000 Hz_10ms_envelope_correlation": 0.9895516886290462, "1000–3000 Hz_10ms_modulation_mae_db": 0.9528081836500472, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8614308702885447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07822370772241988, "1000–3000 Hz_40ms_envelope_correlation": 0.9967390210135352, "1000–3000 Hz_40ms_modulation_mae_db": 1.1368361015912052, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5129188562629163, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16634818941588608, "3000–6000 Hz_10ms_envelope_correlation": 0.9810957303290032, "3000–6000 Hz_10ms_modulation_mae_db": 0.5949129351925588, "3000–6000 Hz_40ms_log_rms_mae_db": 1.183386584136792, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11346180571049126, "3000–6000 Hz_40ms_envelope_correlation": 0.9913696797721147, "3000–6000 Hz_40ms_modulation_mae_db": 0.8550783545512703, "mr_stft_db": 5.543066092131649, "band_mae_db": 0.19724890578871135, "rms_gain_error_db": -0.12508427094071756, "output_peak": 0.4667353075104027}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37200, "output_samples": 37200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.439515834013957, "spectral_convergence": 0.1474292032982251, "spectral_convergence_status": "ok", "amplitude_floor": 1.3482074212160965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.367336677854577, "spectral_convergence": 0.14656834531234572, "spectral_convergence_status": "ok", "amplitude_floor": 1.080252897815134e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.789872821571092, "spectral_convergence": 0.14673939020723967, "spectral_convergence_status": "ok", "amplitude_floor": 8.596818112459857e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004510568589627121, "output_power": 0.00044398100108588895, "delta_db": -0.06866905167892506, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019377343288488795, "output_power": 0.0018992158622508676, "delta_db": -0.08719904427069712, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013786956670426643, "output_power": 0.0013954685747705777, "delta_db": 0.0525165015480035, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031975863120548915, "output_power": 0.00028837633436773903, "delta_db": -0.4486265907623218, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.4830953182024825e-05, "output_power": 1.4565826989766635e-05, "delta_db": -0.07833916755698041, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.03312086795025e-05, "output_power": 2.0332867496159733e-05, "delta_db": 0.0003543249909393365, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12261744680131122, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06417381197133798, "pitch": null, "output_rms_dbfs": -23.91266446037021, "output_peak": 0.4747125109098952, "reference_rms_dbfs": -23.848490648398872, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5322417778132085}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00019668187909197959, "log_rms_mae_db": 0.7487609590083476, "rms_envelope_nrmse": 0.052562506181304675, "envelope_correlation": 0.997473971403217, "rms_gain_error_db": -0.06417383517673206, "reference_mean_power": 0.004131292024738224, "output_mean_power": 0.004070694506622619}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.0026327800860698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011612269501705267, "output_power": 0.0011803524657031063, "reference_active": true, "delta_db": 0.07094605049926676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017095886275955477, "output_power": 0.00016941197515241212, "reference_active": true, "delta_db": -0.03947514313089665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.217923048678998e-05, "output_power": 4.281351681446265e-05, "reference_active": true, "delta_db": 0.06482251131061745}], "active_band_count": 3, "active_band_mae_db": 0.05841456831359362}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00018115923356648065, "log_rms_mae_db": 0.4066552197712254, "rms_envelope_nrmse": 0.02760974633741073, "envelope_correlation": 0.999303379537847, "rms_gain_error_db": -0.06417377959534913, "reference_mean_power": 0.004185412442455258, "output_mean_power": 0.004124021142412464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.7728315536386115e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066252574844359, "output_power": 0.0010828711806318513, "reference_active": true, "delta_db": 0.06716702685802696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.440983264620606e-05, "output_power": 8.696430232873886e-05, "reference_active": true, "delta_db": 0.12947977956436557}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.940091323355611e-06, "output_power": 3.367519003144831e-06, "reference_active": true, "delta_db": 0.5894923473929093}], "active_band_count": 3, "active_band_mae_db": 0.2620463846051006}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.000176021246464811, "log_rms_mae_db": 1.0490158599319606, "rms_envelope_nrmse": 0.06641220685702405, "envelope_correlation": 0.9965046834651662, "rms_gain_error_db": -0.07113954756258282, "reference_mean_power": 0.0026395436088586166, "output_mean_power": 0.0025966587972015343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.515899729654865e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009091516678957249, "output_power": 0.0009091733159767849, "reference_active": true, "delta_db": 0.00010340992102012446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001833276731517994, "output_power": 0.00017103994093966595, "reference_active": true, "delta_db": -0.3013048838023913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7517429693718724e-05, "output_power": 2.5785146711337968e-05, "reference_active": true, "delta_db": -1.6286347057786914}], "active_band_count": 3, "active_band_mae_db": 0.643347666500701}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00016136314602147107, "log_rms_mae_db": 0.5167595284588804, "rms_envelope_nrmse": 0.032948640150701515, "envelope_correlation": 0.9991357494296184, "rms_gain_error_db": -0.07113955449932707, "reference_mean_power": 0.002674122622482191, "output_mean_power": 0.002630675999467843}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.348456388983372e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008380618007305507, "output_power": 0.0008360095211603545, "reference_active": true, "delta_db": -0.01064822182314168}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.026333057603631e-05, "output_power": 8.934557534959836e-05, "reference_active": true, "delta_db": -0.044383044086735296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498114091574834e-06, "output_power": 1.840319073353218e-06, "reference_active": true, "delta_db": -1.3271914195066126}], "active_band_count": 3, "active_band_mae_db": 0.4607408951388299}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 6.667691377903701e-05, "log_rms_mae_db": 1.4836162796381729, "rms_envelope_nrmse": 0.11831093375699525, "envelope_correlation": 0.9910572114722442, "rms_gain_error_db": -0.4096928793965, "reference_mean_power": 0.0003102146198149334, "output_mean_power": 0.0002822883618517849}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.247013787334141e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018435630971779633, "output_power": 0.0001668309470659363, "reference_active": true, "delta_db": -0.4338139100861466}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0328682781544803e-05, "output_power": 1.421202930336575e-05, "reference_active": true, "delta_db": -1.554531447031515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.580926194982421e-06, "output_power": 2.922249988807211e-06, "reference_active": true, "delta_db": -1.9523592910947705}], "active_band_count": 3, "active_band_mae_db": 1.3135682160708109}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 6.131102329922278e-05, "log_rms_mae_db": 0.8627566371500606, "rms_envelope_nrmse": 0.08302496041079543, "envelope_correlation": 0.9962502275527846, "rms_gain_error_db": -0.4096924804006391, "reference_mean_power": 0.0003142782197045969, "output_mean_power": 0.0002859861730877347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 3.052275331975406e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017569545994603763, "output_power": 0.0001586353897646434, "reference_active": true, "delta_db": -0.44360459278503045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.589536328721284e-06, "output_power": 5.735041245735454e-06, "reference_active": true, "delta_db": -1.754331751950608}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8479520444652144e-07, "output_power": 1.427132974945416e-07, "reference_active": true, "delta_db": -3.0006823116849235}], "active_band_count": 3, "active_band_mae_db": 1.7328728854735207}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 2.0872644535346135e-05, "log_rms_mae_db": 1.5763904202965713, "rms_envelope_nrmse": 0.1670743257333871, "envelope_correlation": 0.9807777766191854, "rms_gain_error_db": -0.12210604645062867, "reference_mean_power": 1.984318614204444e-05, "output_mean_power": 1.9293046086153885e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.6220002221112295e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.352365504647456e-06, "output_power": 9.809790434488086e-06, "reference_active": true, "delta_db": 0.20738258318747374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.791249366368675e-07, "output_power": 7.320523308467807e-07, "reference_active": true, "delta_db": -1.2629598368561357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495952213974179e-07, "output_power": 3.9731601028847067e-07, "reference_active": true, "delta_db": 2.018997994403658}], "active_band_count": 3, "active_band_mae_db": 1.1631134714824223}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 1.9522833601394496e-05, "log_rms_mae_db": 1.3128421586071575, "rms_envelope_nrmse": 0.1165354185691929, "envelope_correlation": 0.9903140951360877, "rms_gain_error_db": -0.12210811047627251, "reference_mean_power": 2.010291979320635e-05, "output_mean_power": 1.9545569493353578e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.5131001811961146e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.175181026763155e-06, "output_power": 8.810770163912754e-06, "reference_active": true, "delta_db": 0.32516494246704936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.283226898133603e-07, "output_power": 2.9300443607743664e-07, "reference_active": true, "delta_db": -1.648968855589701}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.754857048422832e-08, "output_power": 2.0363779380795403e-08, "reference_active": true, "delta_db": 0.6461663890058436}], "active_band_count": 3, "active_band_mae_db": 0.8734333956875314}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.940933907974905, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b973b66362a591c6f26f6383e87773d57bfbe328a6f5f76160e5b19d6fb14fd8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9910197088908055, "standard_deviation": 0.0017421258227627897, "minimum": 0.9887752539874443, "maximum": 0.9933259945772629, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.428522502937017, "standard_deviation": 0.03697653294334491, "minimum": 1.3865703589047271, "maximum": 1.4836162796381729, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7205179114158444, "standard_deviation": 0.46573470905950204, "minimum": 0.19125315865621517, "maximum": 1.3968699362859989, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11367865281670358, "standard_deviation": 0.011014932706475722, "minimum": 0.09736583895423329, "maximum": 0.1259622807313844, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971672894957255, "standard_deviation": 0.0008470275849426351, "minimum": 0.9958568316285477, "maximum": 0.998133070324286, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8154012993112038, "standard_deviation": 0.037973973540741085, "minimum": 0.7670334344424335, "maximum": 0.8627566371500606, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8250571488093811, "standard_deviation": 0.5313280577584917, "minimum": 0.12813544299759538, "maximum": 1.7328728854735207, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06637306837161475, "standard_deviation": 0.01145544258347563, "minimum": 0.05294510466452595, "maximum": 0.08302496041079543, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9809235744194144, "standard_deviation": 0.002188454681598641, "minimum": 0.976972870831057, "maximum": 0.9846769063391542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.558000596979635, "standard_deviation": 0.030777557653636732, "minimum": 1.5129188562629163, "maximum": 1.6067619056875662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2383635058292053, "standard_deviation": 0.4922528476686572, "minimum": 0.5949129351925588, "maximum": 1.92993525862273, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16867561242386064, "standard_deviation": 0.009298913790615565, "minimum": 0.15425951153081788, "maximum": 0.18555778793788993, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900499429439663, "standard_deviation": 0.0011354499236688074, "minimum": 0.988330715329931, "maximum": 0.9913696797721147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2688908926427542, "standard_deviation": 0.045138902445366696, "minimum": 1.183386584136792, "maximum": 1.3128421586071575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2428819445735866, "standard_deviation": 0.640181010899651, "minimum": 0.46106329753084546, "maximum": 2.3627694516950286, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12370173409259136, "standard_deviation": 0.008288773808845669, "minimum": 0.11346180571049126, "maximum": 0.1354140934298831, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9965501364046172, "standard_deviation": 0.0008532875932061888, "minimum": 0.9953220953914427, "maximum": 0.9977093715473705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.056530494583259, "standard_deviation": 0.030897545050216513, "minimum": 1.0231571198458829, "maximum": 1.1250998480697216, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5671882313329062, "standard_deviation": 0.20262022507485675, "minimum": 0.3088945477754141, "maximum": 0.9038701119727893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06620588338220887, "standard_deviation": 0.007782421223499575, "minimum": 0.05414337834764288, "maximum": 0.07654778714252752, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989671492511208, "standard_deviation": 0.0003937179089689637, "minimum": 0.9981947887910276, "maximum": 0.999483424003149, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.529895434705276, "standard_deviation": 0.033013716082621325, "minimum": 0.4760618555654064, "maximum": 0.5783859648604758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6209067446462795, "standard_deviation": 0.28957311410330927, "minimum": 0.29120184928191367, "maximum": 1.1400122259199759, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.036535526453562006, "standard_deviation": 0.006389083259232997, "minimum": 0.026124080249248727, "maximum": 0.04650826688470546, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1435511961463533, "standard_deviation": 0.05988058918825769, "minimum": 0.06590500597422043, "maximum": 0.23175881472063486, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.545449143434472, "standard_deviation": 0.009842688423458272, "minimum": 5.5322417778132085, "maximum": 5.564048151104938, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4700499156352413, "standard_deviation": 0.003927059032454917, "minimum": 0.4634876816894676, "maximum": 0.4755511038908124, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08190256852835745, "standard_deviation": 0.03858900741177118, "minimum": -0.13044512653740373, "maximum": -0.03214311698425731, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973041490210433, "standard_deviation": 0.0008275925273853725, "minimum": 0.9955976205167093, "maximum": 0.9980350770799121, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7906661465019236, "standard_deviation": 0.03936841729872098, "minimum": 0.7424353818337026, "maximum": 0.8438400358441719, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3287892376108216, "standard_deviation": 0.20808228038410218, "minimum": 0.05841456831359362, "maximum": 0.7265066139275499, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05422727125579049, "standard_deviation": 0.00743826723685702, "minimum": 0.046350892564809724, "maximum": 0.06902463336568361, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999243499427636, "standard_deviation": 0.0002708426745974217, "minimum": 0.9987483030142864, "maximum": 0.9995752944319996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4284099033861419, "standard_deviation": 0.03674987534344486, "minimum": 0.3776042824083426, "maximum": 0.47560932488978186, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38484452067883224, "standard_deviation": 0.21162493927193574, "minimum": 0.14334184713227527, "maximum": 0.8336618342946868, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02922049678919996, "standard_deviation": 0.004529440752595486, "minimum": 0.021883139803291707, "maximum": 0.0359132123167789, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0022", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.578738212585449, "predicted_mos": 2.578738212585449, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "c50c09a9a80e0b085212c5bcef4f7d976564b89a7879ba5621629de2d937f61b", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.087062099948525, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7250175061115008, "全帯域_10ms_rms_envelope_nrmse": 0.04832743141841603, "全帯域_10ms_envelope_correlation": 0.997927110206046, "全帯域_10ms_modulation_mae_db": 0.16561131535125723, "全帯域_40ms_log_rms_mae_db": 0.40026668416059424, "全帯域_40ms_rms_envelope_nrmse": 0.025832593467841655, "全帯域_40ms_envelope_correlation": 0.9994631723236366, "全帯域_40ms_modulation_mae_db": 0.20450363122775728, "300–1000 Hz_10ms_log_rms_mae_db": 1.201393909720821, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06044386858861699, "300–1000 Hz_10ms_envelope_correlation": 0.9972316779175872, "300–1000 Hz_10ms_modulation_mae_db": 0.8037186287810067, "300–1000 Hz_40ms_log_rms_mae_db": 0.6294380217863079, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03373599566692499, "300–1000 Hz_40ms_envelope_correlation": 0.9992243650525341, "300–1000 Hz_40ms_modulation_mae_db": 0.7417415626456229, "1000–3000 Hz_10ms_log_rms_mae_db": 1.642620876811913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12113786886790094, "1000–3000 Hz_10ms_envelope_correlation": 0.9899829475098899, "1000–3000 Hz_10ms_modulation_mae_db": 0.8535055947144962, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9726950353027253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07033142172349185, "1000–3000 Hz_40ms_envelope_correlation": 0.9970900020114225, "1000–3000 Hz_40ms_modulation_mae_db": 0.8722689143431435, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1677817617285666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20316954827241165, "3000–6000 Hz_10ms_envelope_correlation": 0.9716635720785268, "3000–6000 Hz_10ms_modulation_mae_db": 1.6914170722930197, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6190547526207186, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12920034390820725, "3000–6000 Hz_40ms_envelope_correlation": 0.9885866553990532, "3000–6000 Hz_40ms_modulation_mae_db": 1.6096538015509267, "mr_stft_db": 5.9590638119986545, "band_mae_db": 0.10305535700125383, "rms_gain_error_db": -0.09750767213619246, "output_peak": 0.4642731388569612}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8058030451619212, "全帯域_10ms_rms_envelope_nrmse": 0.05167111767503357, "全帯域_10ms_envelope_correlation": 0.9975879900866225, "全帯域_10ms_modulation_mae_db": 0.3670953357406908, "全帯域_40ms_log_rms_mae_db": 0.43225438925832926, "全帯域_40ms_rms_envelope_nrmse": 0.027853816721725023, "全帯域_40ms_envelope_correlation": 0.9993289412909598, "全帯域_40ms_modulation_mae_db": 0.39269468738169927, "300–1000 Hz_10ms_log_rms_mae_db": 1.1884776233205607, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05587932322812986, "300–1000 Hz_10ms_envelope_correlation": 0.9975178826539086, "300–1000 Hz_10ms_modulation_mae_db": 0.4415204752593243, "300–1000 Hz_40ms_log_rms_mae_db": 0.5480673994884664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.029510466689703696, "300–1000 Hz_40ms_envelope_correlation": 0.9992983641218864, "300–1000 Hz_40ms_modulation_mae_db": 0.37876054260315933, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8357082549094412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12975924533420852, "1000–3000 Hz_10ms_envelope_correlation": 0.9886549447475877, "1000–3000 Hz_10ms_modulation_mae_db": 1.4923451669251848, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0811063939900178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07284724564986045, "1000–3000 Hz_40ms_envelope_correlation": 0.997171266214936, "1000–3000 Hz_40ms_modulation_mae_db": 1.6532370630268565, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2142649535641854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22938729562504984, "3000–6000 Hz_10ms_envelope_correlation": 0.963624008167324, "3000–6000 Hz_10ms_modulation_mae_db": 2.347826827569208, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6674881322433746, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13386102650192122, "3000–6000 Hz_40ms_envelope_correlation": 0.9879195555698986, "3000–6000 Hz_40ms_modulation_mae_db": 1.4351898462989316, "mr_stft_db": 5.966998577468396, "band_mae_db": 0.1549435166527563, "rms_gain_error_db": -0.07002722127133489, "output_peak": 0.45805666907020937}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7274897883571867, "全帯域_10ms_rms_envelope_nrmse": 0.04772587026031922, "全帯域_10ms_envelope_correlation": 0.997906034322869, "全帯域_10ms_modulation_mae_db": 0.4642434854889917, "全帯域_40ms_log_rms_mae_db": 0.3657345119396929, "全帯域_40ms_rms_envelope_nrmse": 0.02433163089275748, "全帯域_40ms_envelope_correlation": 0.999444337619632, "全帯域_40ms_modulation_mae_db": 0.41248111904252854, "300–1000 Hz_10ms_log_rms_mae_db": 1.2348331034570321, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06384283109509432, "300–1000 Hz_10ms_envelope_correlation": 0.9967905932821655, "300–1000 Hz_10ms_modulation_mae_db": 0.5672350618255662, "300–1000 Hz_40ms_log_rms_mae_db": 0.5651165981629765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.032705267370405244, "300–1000 Hz_40ms_envelope_correlation": 0.9991634967668367, "300–1000 Hz_40ms_modulation_mae_db": 0.48109660376530905, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7992816819029676, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12483747543855633, "1000–3000 Hz_10ms_envelope_correlation": 0.9887991717943068, "1000–3000 Hz_10ms_modulation_mae_db": 0.651073850315916, "1000–3000 Hz_40ms_log_rms_mae_db": 1.038968828642775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06578022496366144, "1000–3000 Hz_40ms_envelope_correlation": 0.9968855399759636, "1000–3000 Hz_40ms_modulation_mae_db": 0.8814981738512819, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1560855341929748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2260795872554968, "3000–6000 Hz_10ms_envelope_correlation": 0.9645337334963344, "3000–6000 Hz_10ms_modulation_mae_db": 1.8301756331738765, "3000–6000 Hz_40ms_log_rms_mae_db": 1.573539958296979, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13973569515211648, "3000–6000 Hz_40ms_envelope_correlation": 0.986737377111148, "3000–6000 Hz_40ms_modulation_mae_db": 1.4489434052953174, "mr_stft_db": 5.979330172554342, "band_mae_db": 0.12431181172783913, "rms_gain_error_db": -0.04358099416671979, "output_peak": 0.4635314932761297}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7895803451064677, "全帯域_10ms_rms_envelope_nrmse": 0.0493204690026911, "全帯域_10ms_envelope_correlation": 0.9977458741168973, "全帯域_10ms_modulation_mae_db": 0.2335792052621962, "全帯域_40ms_log_rms_mae_db": 0.4140239637287666, "全帯域_40ms_rms_envelope_nrmse": 0.024340070061354455, "全帯域_40ms_envelope_correlation": 0.999425260273858, "全帯域_40ms_modulation_mae_db": 0.17391428663930572, "300–1000 Hz_10ms_log_rms_mae_db": 1.2037522564707264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06287883411071152, "300–1000 Hz_10ms_envelope_correlation": 0.9968427631867086, "300–1000 Hz_10ms_modulation_mae_db": 0.38112277383129367, "300–1000 Hz_40ms_log_rms_mae_db": 0.5341700583086176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03349090921473307, "300–1000 Hz_40ms_envelope_correlation": 0.9990654845961446, "300–1000 Hz_40ms_modulation_mae_db": 0.17488719306472075, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7439268932328074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1037499952485044, "1000–3000 Hz_10ms_envelope_correlation": 0.9923703925290482, "1000–3000 Hz_10ms_modulation_mae_db": 0.19476113889138677, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0732994166613472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05539587874837817, "1000–3000 Hz_40ms_envelope_correlation": 0.9979176990944202, "1000–3000 Hz_40ms_modulation_mae_db": 0.309391220639246, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2657704610172873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23169391188240904, "3000–6000 Hz_10ms_envelope_correlation": 0.9627525367729557, "3000–6000 Hz_10ms_modulation_mae_db": 1.8911466492498157, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7191106411041612, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15411189204002995, "3000–6000 Hz_40ms_envelope_correlation": 0.9838383335326113, "3000–6000 Hz_40ms_modulation_mae_db": 1.7288639100588075, "mr_stft_db": 5.9733717583964845, "band_mae_db": 0.07624004321037489, "rms_gain_error_db": -0.027415870130589843, "output_peak": 0.46652960078842975}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7083916859901842, "全帯域_10ms_rms_envelope_nrmse": 0.052370303277258175, "全帯域_10ms_envelope_correlation": 0.9974901322785692, "全帯域_10ms_modulation_mae_db": 0.19356041836267235, "全帯域_40ms_log_rms_mae_db": 0.4008475658547334, "全帯域_40ms_rms_envelope_nrmse": 0.029135127364160247, "全帯域_40ms_envelope_correlation": 0.9992168606162817, "全帯域_40ms_modulation_mae_db": 0.26890006698001706, "300–1000 Hz_10ms_log_rms_mae_db": 1.1684019734088678, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05994709757520123, "300–1000 Hz_10ms_envelope_correlation": 0.9971591452050331, "300–1000 Hz_10ms_modulation_mae_db": 0.2708423838068315, "300–1000 Hz_40ms_log_rms_mae_db": 0.585343731328693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03448461786807884, "300–1000 Hz_40ms_envelope_correlation": 0.9990485242522527, "300–1000 Hz_40ms_modulation_mae_db": 0.3120173408815227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7933048765363944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14763644341157617, "1000–3000 Hz_10ms_envelope_correlation": 0.9847372845125902, "1000–3000 Hz_10ms_modulation_mae_db": 0.7121307598427133, "1000–3000 Hz_40ms_log_rms_mae_db": 1.052552759361546, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08343223227289027, "1000–3000 Hz_40ms_envelope_correlation": 0.9956222291553373, "1000–3000 Hz_40ms_modulation_mae_db": 0.8717672821367967, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1957057104390043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20324454523438115, "3000–6000 Hz_10ms_envelope_correlation": 0.9723685009447435, "3000–6000 Hz_10ms_modulation_mae_db": 1.7306718245602113, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6609126286185663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1337430561068106, "3000–6000 Hz_40ms_envelope_correlation": 0.9879901976265271, "3000–6000 Hz_40ms_modulation_mae_db": 1.5009541542110043, "mr_stft_db": 5.955841531615911, "band_mae_db": 0.17010887618768447, "rms_gain_error_db": -0.06350071250235559, "output_peak": 0.4610232116060946}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7347914550071751, "全帯域_10ms_rms_envelope_nrmse": 0.0432345646110573, "全帯域_10ms_envelope_correlation": 0.9983005926678886, "全帯域_10ms_modulation_mae_db": 0.3179084471935913, "全帯域_40ms_log_rms_mae_db": 0.38757009987208474, "全帯域_40ms_rms_envelope_nrmse": 0.02440321815323825, "全帯域_40ms_envelope_correlation": 0.9994602523765497, "全帯域_40ms_modulation_mae_db": 0.33876607652446994, "300–1000 Hz_10ms_log_rms_mae_db": 1.0670723753983673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.050710188839500486, "300–1000 Hz_10ms_envelope_correlation": 0.9980353940429305, "300–1000 Hz_10ms_modulation_mae_db": 0.32505451947975544, "300–1000 Hz_40ms_log_rms_mae_db": 0.5114426761023546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03155835645581005, "300–1000 Hz_40ms_envelope_correlation": 0.9992702560954575, "300–1000 Hz_40ms_modulation_mae_db": 0.3109343484013485, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7602470635580927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12285684334958323, "1000–3000 Hz_10ms_envelope_correlation": 0.9892099249216749, "1000–3000 Hz_10ms_modulation_mae_db": 0.38224567641428936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9689208288067447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05867422125501382, "1000–3000 Hz_40ms_envelope_correlation": 0.9976537252581358, "1000–3000 Hz_40ms_modulation_mae_db": 0.22867695660901646, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1834970198407464, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22219340907131044, "3000–6000 Hz_10ms_envelope_correlation": 0.9666059858855099, "3000–6000 Hz_10ms_modulation_mae_db": 1.4788441655369529, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7098063722294163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14382527036215442, "3000–6000 Hz_40ms_envelope_correlation": 0.9861209276042758, "3000–6000 Hz_40ms_modulation_mae_db": 0.9727409899534507, "mr_stft_db": 5.947227892443858, "band_mae_db": 0.2226799441084096, "rms_gain_error_db": -0.06036839948026512, "output_peak": 0.4663239689621831}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.733701799223396, "全帯域_10ms_rms_envelope_nrmse": 0.04518390639152931, "全帯域_10ms_envelope_correlation": 0.9981986439397138, "全帯域_10ms_modulation_mae_db": 0.46429241156323453, "全帯域_40ms_log_rms_mae_db": 0.38574887803881025, "全帯域_40ms_rms_envelope_nrmse": 0.02482008682010977, "全帯域_40ms_envelope_correlation": 0.9995047480256386, "全帯域_40ms_modulation_mae_db": 0.5438955412513237, "300–1000 Hz_10ms_log_rms_mae_db": 1.1719234005666648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.058155928173083335, "300–1000 Hz_10ms_envelope_correlation": 0.9974165154113669, "300–1000 Hz_10ms_modulation_mae_db": 0.6577271827680546, "300–1000 Hz_40ms_log_rms_mae_db": 0.557797861180649, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031152607391063448, "300–1000 Hz_40ms_envelope_correlation": 0.9993135301578816, "300–1000 Hz_40ms_modulation_mae_db": 0.7756821850559681, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7411029617679885, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11727737856326477, "1000–3000 Hz_10ms_envelope_correlation": 0.9904295788329546, "1000–3000 Hz_10ms_modulation_mae_db": 0.6794183940577746, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0097901207227844, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06449393525223447, "1000–3000 Hz_40ms_envelope_correlation": 0.9974025262786995, "1000–3000 Hz_40ms_modulation_mae_db": 0.8100838824035576, "3000–6000 Hz_10ms_log_rms_mae_db": 2.284192694378314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24224435059038213, "3000–6000 Hz_10ms_envelope_correlation": 0.9599822510844397, "3000–6000 Hz_10ms_modulation_mae_db": 2.137354522654424, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7773513278343422, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15570289965161638, "3000–6000 Hz_40ms_envelope_correlation": 0.983164166306049, "3000–6000 Hz_40ms_modulation_mae_db": 1.90195875935284, "mr_stft_db": 5.964246569374858, "band_mae_db": 0.1907741183826995, "rms_gain_error_db": -0.09402838123465837, "output_peak": 0.4887904077515099}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7372433471089098, "全帯域_10ms_rms_envelope_nrmse": 0.05159817474445658, "全帯域_10ms_envelope_correlation": 0.9976034774686641, "全帯域_10ms_modulation_mae_db": 0.40832748145755837, "全帯域_40ms_log_rms_mae_db": 0.4033443250453033, "全帯域_40ms_rms_envelope_nrmse": 0.02774039281892018, "全帯域_40ms_envelope_correlation": 0.9993316504145139, "全帯域_40ms_modulation_mae_db": 0.5956776066765753, "300–1000 Hz_10ms_log_rms_mae_db": 1.1588059559779054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06025815292076462, "300–1000 Hz_10ms_envelope_correlation": 0.9972212916532138, "300–1000 Hz_10ms_modulation_mae_db": 0.5865648850756512, "300–1000 Hz_40ms_log_rms_mae_db": 0.5595640312861248, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031774345795898946, "300–1000 Hz_40ms_envelope_correlation": 0.9992692410984234, "300–1000 Hz_40ms_modulation_mae_db": 0.7850131553798013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6970028509461412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12604938845361954, "1000–3000 Hz_10ms_envelope_correlation": 0.9895418178513172, "1000–3000 Hz_10ms_modulation_mae_db": 0.8229181170278791, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8981000540965374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07185769094532833, "1000–3000 Hz_40ms_envelope_correlation": 0.9974961947746661, "1000–3000 Hz_40ms_modulation_mae_db": 1.2410388230034428, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1923064438309803, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2083010110792813, "3000–6000 Hz_10ms_envelope_correlation": 0.9706873669908985, "3000–6000 Hz_10ms_modulation_mae_db": 1.7947266955074763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.622171440788958, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12841072862103997, "3000–6000 Hz_40ms_envelope_correlation": 0.9887182165784643, "3000–6000 Hz_40ms_modulation_mae_db": 1.291965276144648, "mr_stft_db": 5.949748480088185, "band_mae_db": 0.19605476149725928, "rms_gain_error_db": -0.055127996930173014, "output_peak": 0.463951511711585}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37200, "output_samples": 37200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.917470508648344, "spectral_convergence": 0.15031907666539326, "spectral_convergence_status": "ok", "amplitude_floor": 1.3482074212160965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.862283018309352, "spectral_convergence": 0.14582138874645129, "spectral_convergence_status": "ok", "amplitude_floor": 1.080252897815134e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.097437909038271, "spectral_convergence": 0.14714246396051037, "spectral_convergence_status": "ok", "amplitude_floor": 8.596818112459857e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004510568589627121, "output_power": 0.0004493449129144132, "delta_db": -0.016514619167094846, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019377343288488795, "output_power": 0.0018895332461023968, "delta_db": -0.10939695602603858, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013786956670426643, "output_power": 0.001359612278549512, "delta_db": -0.06053332685821437, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031975863120548915, "output_power": 0.000296735751913302, "delta_db": -0.32452401156568894, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.4830953182024825e-05, "output_power": 1.4866937781746476e-05, "delta_db": 0.010524599716477229, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.03312086795025e-05, "output_power": 2.078964439601152e-05, "delta_db": 0.09683862867400896, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10305535700125383, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09750767213619246, "pitch": null, "output_rms_dbfs": -23.945998320535065, "output_peak": 0.4642731388569612, "reference_rms_dbfs": -23.848490648398872, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9590638119986545}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00019668187909197959, "log_rms_mae_db": 0.7250175061115008, "rms_envelope_nrmse": 0.04832743141841603, "envelope_correlation": 0.997927110206046, "rms_gain_error_db": -0.09750770046877416, "reference_mean_power": 0.004131292024738224, "output_mean_power": 0.004039569872950454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.0026327800860698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011612269501705267, "output_power": 0.0011280884453853859, "reference_active": true, "delta_db": -0.12573955617055826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017095886275955477, "output_power": 0.00015898893360809737, "reference_active": true, "delta_db": -0.3152472378792665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.217923048678998e-05, "output_power": 4.16403083708244e-05, "reference_active": true, "delta_db": -0.0558471520039469}], "active_band_count": 3, "active_band_mae_db": 0.16561131535125723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00018115923356648065, "log_rms_mae_db": 0.40026668416059424, "rms_envelope_nrmse": 0.025832593467841655, "envelope_correlation": 0.9994631723236366, "rms_gain_error_db": -0.0975076152232183, "reference_mean_power": 0.004185412442455258, "output_mean_power": 0.00409248879988097}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.7728315536386115e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066252574844359, "output_power": 0.0010384615834804337, "reference_active": true, "delta_db": -0.11469657919139739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.440983264620606e-05, "output_power": 7.90437131779902e-05, "reference_active": true, "delta_db": -0.28525705634476134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.940091323355611e-06, "output_power": 2.7990141539901017e-06, "reference_active": true, "delta_db": -0.21355725814711318}], "active_band_count": 3, "active_band_mae_db": 0.20450363122775728}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.000176021246464811, "log_rms_mae_db": 1.201393909720821, "rms_envelope_nrmse": 0.06044386858861699, "envelope_correlation": 0.9972316779175872, "rms_gain_error_db": -0.12694169933009922, "reference_mean_power": 0.0026395436088586166, "output_mean_power": 0.002563507967585466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.515899729654865e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009091516678957249, "output_power": 0.0008756939362150917, "reference_active": true, "delta_db": -0.16283997381876236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001833276731517994, "output_power": 0.0001696927907204541, "reference_active": true, "delta_db": -0.33564634270227606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7517429693718724e-05, "output_power": 2.415272409970968e-05, "reference_active": true, "delta_db": -1.912669569821982}], "active_band_count": 3, "active_band_mae_db": 0.8037186287810067}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00016136314602147107, "log_rms_mae_db": 0.6294380217863079, "rms_envelope_nrmse": 0.03373599566692499, "envelope_correlation": 0.9992243650525341, "rms_gain_error_db": -0.12694172435011108, "reference_mean_power": 0.002674122622482191, "output_mean_power": 0.002597090870760419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.348456388983372e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008380618007305507, "output_power": 0.0008087013774638543, "reference_active": true, "delta_db": -0.15487862903806138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.026333057603631e-05, "output_power": 8.557803992236428e-05, "reference_active": true, "delta_db": -0.23149019078316108}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498114091574834e-06, "output_power": 1.635786736171698e-06, "reference_active": true, "delta_db": -1.8388558681156464}], "active_band_count": 3, "active_band_mae_db": 0.7417415626456229}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 6.667691377903701e-05, "log_rms_mae_db": 1.642620876811913, "rms_envelope_nrmse": 0.12113786886790094, "envelope_correlation": 0.9899829475098899, "rms_gain_error_db": -0.30581226948211493, "reference_mean_power": 0.0003102146198149334, "output_mean_power": 0.00028912193010743577}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.247013787334141e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018435630971779633, "output_power": 0.00017339514853565618, "reference_active": true, "delta_db": -0.26621064134161887}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0328682781544803e-05, "output_power": 1.7229658040017723e-05, "reference_active": true, "delta_db": -0.7183258096400766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.580926194982421e-06, "output_power": 3.186803323632676e-06, "reference_active": true, "delta_db": -1.5759803331617932}], "active_band_count": 3, "active_band_mae_db": 0.8535055947144962}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 6.131102329922278e-05, "log_rms_mae_db": 0.9726950353027253, "rms_envelope_nrmse": 0.07033142172349185, "envelope_correlation": 0.9970900020114225, "rms_gain_error_db": -0.3058130473815509, "reference_mean_power": 0.0003142782197045969, "output_mean_power": 0.0002929091776861655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 3.052275331975406e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017569545994603763, "output_power": 0.0001660621459553105, "reference_active": true, "delta_db": -0.24489893426029768}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.589536328721284e-06, "output_power": 7.366793410540152e-06, "reference_active": true, "delta_db": -0.6669122987815894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8479520444652144e-07, "output_power": 1.923238397528074e-07, "reference_active": true, "delta_db": -1.7049955099875431}], "active_band_count": 3, "active_band_mae_db": 0.8722689143431435}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 2.0872644535346135e-05, "log_rms_mae_db": 2.1677817617285666, "rms_envelope_nrmse": 0.20316954827241165, "envelope_correlation": 0.9716635720785268, "rms_gain_error_db": -0.052657515539405696, "reference_mean_power": 1.984318614204444e-05, "output_mean_power": 1.960404337823715e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.6220002221112295e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.352365504647456e-06, "output_power": 7.606295244085485e-06, "reference_active": true, "delta_db": -0.8974829250283777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.791249366368675e-07, "output_power": 1.0019447287576586e-06, "reference_active": true, "delta_db": 0.100056533413801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495952213974179e-07, "output_power": 6.381274725424363e-07, "reference_active": true, "delta_db": 4.07671175843688}], "active_band_count": 3, "active_band_mae_db": 1.6914170722930197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 1.9522833601394496e-05, "log_rms_mae_db": 1.6190547526207186, "rms_envelope_nrmse": 0.12920034390820725, "envelope_correlation": 0.9885866553990532, "rms_gain_error_db": -0.05265686963500872, "reference_mean_power": 2.010291979320635e-05, "output_mean_power": 1.9860649768965713e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.5131001811961146e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.175181026763155e-06, "output_power": 6.760190830170004e-06, "reference_active": true, "delta_db": -0.8253842261527919}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.283226898133603e-07, "output_power": 5.049722420111958e-07, "reference_active": true, "delta_db": 0.7149642472861424}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.754857048422832e-08, "output_power": 3.7419935752246575e-08, "reference_active": true, "delta_db": 3.2886129312138457}], "active_band_count": 3, "active_band_mae_db": 1.6096538015509267}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9346619504335701, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9ef8b53e1709e3eaa914fb6b4742400cdc72312f9f4c7d3663fc867d265cee8d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9892157578374212, "standard_deviation": 0.0021601848191497033, "minimum": 0.9847372845125902, "maximum": 0.9923703925290482, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7516494324582184, "standard_deviation": 0.061165060724842936, "minimum": 1.642620876811913, "maximum": 1.8357082549094412, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.723549837273705, "standard_deviation": 0.3819584081432462, "minimum": 0.19476113889138677, "maximum": 1.4923451669251848, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12416307983340175, "standard_deviation": 0.012313139874701392, "minimum": 0.1037499952485044, "maximum": 0.14763644341157617, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971548978454476, "standard_deviation": 0.0007004696726608402, "minimum": 0.9956222291553373, "maximum": 0.9979176990944202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0119291796980598, "standard_deviation": 0.06243155399908642, "minimum": 0.8981000540965374, "maximum": 1.0811063939900178, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8584952895016676, "standard_deviation": 0.4597171658522123, "minimum": 0.22867695660901646, "maximum": 1.6532370630268565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06785160635135734, "standard_deviation": 0.008822411700977414, "minimum": 0.05539587874837817, "maximum": 0.08343223227289027, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9665272444275916, "standard_deviation": 0.004587298952413625, "minimum": 0.9599822510844397, "maximum": 0.9723685009447435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2074505723740074, "standard_deviation": 0.04549731431260527, "minimum": 2.1560855341929748, "maximum": 2.284192694378314, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.862770423818123, "standard_deviation": 0.2702869122315217, "minimum": 1.4788441655369529, "maximum": 2.347826827569208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22078920737634028, "standard_deviation": 0.014428866786838345, "minimum": 0.20316954827241165, "maximum": 0.24224435059038213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9866344287160034, "standard_deviation": 0.002130666310249838, "minimum": 0.983164166306049, "maximum": 0.9887182165784643, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6686794067170645, "standard_deviation": 0.06522836471627233, "minimum": 1.573539958296979, "maximum": 1.7773513278343422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4862837678582408, "standard_deviation": 0.28095677396298363, "minimum": 0.9727409899534507, "maximum": 1.90195875935284, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13982386404298702, "standard_deviation": 0.010609706062310351, "minimum": 0.12841072862103997, "maximum": 0.15570289965161638, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9972769079191143, "standard_deviation": 0.0003958446355415034, "minimum": 0.9967905932821655, "maximum": 0.9980353940429305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1743325747901183, "standard_deviation": 0.04963442397571151, "minimum": 1.0670723753983673, "maximum": 1.2348331034570321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5042232388534355, "standard_deviation": 0.18113333702684375, "minimum": 0.2708423838068315, "maximum": 0.8037186287810067, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.059014528066387795, "standard_deviation": 0.004178746009557173, "minimum": 0.050710188839500486, "maximum": 0.06384283109509432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9992066577676771, "standard_deviation": 0.00010344472420831248, "minimum": 0.9990485242522527, "maximum": 0.9993135301578816, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5613675472055237, "standard_deviation": 0.03514782600895724, "minimum": 0.5114426761023546, "maximum": 0.6294380217863079, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4950166164746816, "standard_deviation": 0.24122054476360325, "minimum": 0.17488719306472075, "maximum": 0.7850131553798013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.032301570806577286, "standard_deviation": 0.001618706662827024, "minimum": 0.029510466689703696, "maximum": 0.03448461786807884, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15477105359603463, "standard_deviation": 0.0501666643011761, "minimum": 0.07624004321037489, "maximum": 0.2226799441084096, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.961978599242586, "standard_deviation": 0.011177424895951518, "minimum": 5.947227892443858, "maximum": 5.979330172554342, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4665600002528878, "standard_deviation": 0.009399210637043288, "minimum": 0.45805666907020937, "maximum": 0.4887904077515099, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06394465598153613, "standard_deviation": 0.02362884215307279, "minimum": -0.09750767213619246, "maximum": -0.027415870130589843, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9978449818859088, "standard_deviation": 0.00029355949899267513, "minimum": 0.9974901322785692, "maximum": 0.9983005926678886, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7452523715083428, "standard_deviation": 0.033852650379257596, "minimum": 0.7083916859901842, "maximum": 0.8058030451619212, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32682726255252403, "standard_deviation": 0.11866463413856039, "minimum": 0.16561131535125723, "maximum": 0.46429241156323453, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04867897967259516, "standard_deviation": 0.003260174503158021, "minimum": 0.0432345646110573, "maximum": 0.052370303277258175, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9993969028676337, "standard_deviation": 9.59028140380253e-05, "minimum": 0.9992168606162817, "maximum": 0.9995047480256386, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3987238022372893, "standard_deviation": 0.019884244674541307, "minimum": 0.3657345119396929, "maximum": 0.43225438925832926, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36635412696545955, "standard_deviation": 0.15138352717737216, "minimum": 0.17391428663930572, "maximum": 0.5956776066765753, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.026057117037513382, "standard_deviation": 0.0019189760494320559, "minimum": 0.02433163089275748, "maximum": 0.029135127364160247, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0022", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8938686847686768, "predicted_mos": 2.8938686847686768, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "d2780bf1d6f12dbfc413f472e980c4bec8276a35a2d5f4b64be510cbc0efbd56", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.262422100175172, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7843687756511099, "全帯域_10ms_rms_envelope_nrmse": 0.04801872160598107, "全帯域_10ms_envelope_correlation": 0.9979578091167112, "全帯域_10ms_modulation_mae_db": 0.29518315697515135, "全帯域_40ms_log_rms_mae_db": 0.45440212020501203, "全帯域_40ms_rms_envelope_nrmse": 0.027953575998766352, "全帯域_40ms_envelope_correlation": 0.999352492073329, "全帯域_40ms_modulation_mae_db": 0.4917517392038091, "300–1000 Hz_10ms_log_rms_mae_db": 0.9357537075035861, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0633584876255258, "300–1000 Hz_10ms_envelope_correlation": 0.9969482709744235, "300–1000 Hz_10ms_modulation_mae_db": 0.5336568627146175, "300–1000 Hz_40ms_log_rms_mae_db": 0.4760778372572405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03784316876478425, "300–1000 Hz_40ms_envelope_correlation": 0.9989887099767392, "300–1000 Hz_40ms_modulation_mae_db": 0.5659586244845168, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3458885740956112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10688638243481553, "1000–3000 Hz_10ms_envelope_correlation": 0.9920060754480444, "1000–3000 Hz_10ms_modulation_mae_db": 0.8332463268692427, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7420258604680805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05905870043246649, "1000–3000 Hz_40ms_envelope_correlation": 0.99774473089138, "1000–3000 Hz_40ms_modulation_mae_db": 0.5409613681768976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4927578020635988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1827947901559848, "3000–6000 Hz_10ms_envelope_correlation": 0.97693279499881, "3000–6000 Hz_10ms_modulation_mae_db": 1.5405296344471218, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3129945424027885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11374380238585154, "3000–6000 Hz_40ms_envelope_correlation": 0.9908068757939648, "3000–6000 Hz_40ms_modulation_mae_db": 1.1839323753097981, "mr_stft_db": 5.494214487639812, "band_mae_db": 0.17470720459584413, "rms_gain_error_db": -0.0999259264158839, "output_peak": 0.4629269323437658}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8062654348707272, "全帯域_10ms_rms_envelope_nrmse": 0.05152440134674353, "全帯域_10ms_envelope_correlation": 0.9976054147760416, "全帯域_10ms_modulation_mae_db": 0.60887172441722, "全帯域_40ms_log_rms_mae_db": 0.38420285735447623, "全帯域_40ms_rms_envelope_nrmse": 0.02521717301302306, "全帯域_40ms_envelope_correlation": 0.9994672971214213, "全帯域_40ms_modulation_mae_db": 0.4461739101044886, "300–1000 Hz_10ms_log_rms_mae_db": 1.0053121827261076, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0676883361754409, "300–1000 Hz_10ms_envelope_correlation": 0.9964555225853445, "300–1000 Hz_10ms_modulation_mae_db": 0.9540984625946763, "300–1000 Hz_40ms_log_rms_mae_db": 0.5192076449451739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035863689539708685, "300–1000 Hz_40ms_envelope_correlation": 0.999078670076771, "300–1000 Hz_40ms_modulation_mae_db": 0.768389247497871, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3948453536273124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10053387511119219, "1000–3000 Hz_10ms_envelope_correlation": 0.9927675088077896, "1000–3000 Hz_10ms_modulation_mae_db": 1.0709206125205408, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8831384467412801, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05337524637204808, "1000–3000 Hz_40ms_envelope_correlation": 0.9979575885724299, "1000–3000 Hz_40ms_modulation_mae_db": 1.1678381159583795, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5359623657013919, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18819616957098426, "3000–6000 Hz_10ms_envelope_correlation": 0.9772311035625105, "3000–6000 Hz_10ms_modulation_mae_db": 1.6469828026644728, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2627782592799368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13189094811733962, "3000–6000 Hz_40ms_envelope_correlation": 0.9886538900941243, "3000–6000 Hz_40ms_modulation_mae_db": 1.7689430892713587, "mr_stft_db": 5.530147622574392, "band_mae_db": 0.1930218487913374, "rms_gain_error_db": -0.08389798728989167, "output_peak": 0.46341039072444024}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.774127636030457, "全帯域_10ms_rms_envelope_nrmse": 0.04657462089483094, "全帯域_10ms_envelope_correlation": 0.9980644310893095, "全帯域_10ms_modulation_mae_db": 0.3714212302887297, "全帯域_40ms_log_rms_mae_db": 0.42531927259831953, "全帯域_40ms_rms_envelope_nrmse": 0.02292830371595785, "全帯域_40ms_envelope_correlation": 0.9995812047645226, "全帯域_40ms_modulation_mae_db": 0.4691292015579105, "300–1000 Hz_10ms_log_rms_mae_db": 0.9411922701652492, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05895110008551333, "300–1000 Hz_10ms_envelope_correlation": 0.9973389360363253, "300–1000 Hz_10ms_modulation_mae_db": 0.7119352407304866, "300–1000 Hz_40ms_log_rms_mae_db": 0.4513840023611826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02953938613001904, "300–1000 Hz_40ms_envelope_correlation": 0.9994026356273643, "300–1000 Hz_40ms_modulation_mae_db": 0.7231896077315428, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4488917515487312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10321246468305802, "1000–3000 Hz_10ms_envelope_correlation": 0.992486427995709, "1000–3000 Hz_10ms_modulation_mae_db": 0.8318336337252341, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7782458813661767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05639041632345948, "1000–3000 Hz_40ms_envelope_correlation": 0.9978782258244315, "1000–3000 Hz_40ms_modulation_mae_db": 1.125896508500097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5288571079952553, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16783918391090225, "3000–6000 Hz_10ms_envelope_correlation": 0.9806058307383955, "3000–6000 Hz_10ms_modulation_mae_db": 0.6097611018354783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2852158972513625, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11818300917084597, "3000–6000 Hz_40ms_envelope_correlation": 0.9903012119469375, "3000–6000 Hz_40ms_modulation_mae_db": 0.4259255152437005, "mr_stft_db": 5.52083043653107, "band_mae_db": 0.12655816110015564, "rms_gain_error_db": -0.0892334700772005, "output_peak": 0.4697463434383246}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8294193335411598, "全帯域_10ms_rms_envelope_nrmse": 0.04954869613651806, "全帯域_10ms_envelope_correlation": 0.9977205905611549, "全帯域_10ms_modulation_mae_db": 0.31301398736966424, "全帯域_40ms_log_rms_mae_db": 0.42949040166514757, "全帯域_40ms_rms_envelope_nrmse": 0.023280586754336578, "全帯域_40ms_envelope_correlation": 0.9994679936630871, "全帯域_40ms_modulation_mae_db": 0.24357274855737077, "300–1000 Hz_10ms_log_rms_mae_db": 1.0592128059222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06107403457185481, "300–1000 Hz_10ms_envelope_correlation": 0.9970222203952821, "300–1000 Hz_10ms_modulation_mae_db": 0.33149688581683684, "300–1000 Hz_40ms_log_rms_mae_db": 0.5385691952709947, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02989018059399976, "300–1000 Hz_40ms_envelope_correlation": 0.999249950542352, "300–1000 Hz_40ms_modulation_mae_db": 0.2664259139093502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.364434771105872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10294792519934574, "1000–3000 Hz_10ms_envelope_correlation": 0.9927059641448005, "1000–3000 Hz_10ms_modulation_mae_db": 1.3034483689683753, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7957162698363129, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06432083703955553, "1000–3000 Hz_40ms_envelope_correlation": 0.9973527022294915, "1000–3000 Hz_40ms_modulation_mae_db": 1.4426160795011895, "3000–6000 Hz_10ms_log_rms_mae_db": 1.522203576690107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16897821098206856, "3000–6000 Hz_10ms_envelope_correlation": 0.9802846660868925, "3000–6000 Hz_10ms_modulation_mae_db": 2.089952822185167, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2875119473781214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11746499507529591, "3000–6000 Hz_40ms_envelope_correlation": 0.990125592883113, "3000–6000 Hz_40ms_modulation_mae_db": 1.9680705679259207, "mr_stft_db": 5.530810202804315, "band_mae_db": 0.15435959486111714, "rms_gain_error_db": -0.012570039044973304, "output_peak": 0.45389162518641607}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7941702759330629, "全帯域_10ms_rms_envelope_nrmse": 0.042069288467090696, "全帯域_10ms_envelope_correlation": 0.9984173820273743, "全帯域_10ms_modulation_mae_db": 0.32197910297055093, "全帯域_40ms_log_rms_mae_db": 0.43621442170387703, "全帯域_40ms_rms_envelope_nrmse": 0.022228303719940862, "全帯域_40ms_envelope_correlation": 0.9995895859104496, "全帯域_40ms_modulation_mae_db": 0.5410049410568861, "300–1000 Hz_10ms_log_rms_mae_db": 0.9535523297340177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05644972396814097, "300–1000 Hz_10ms_envelope_correlation": 0.9975043576055451, "300–1000 Hz_10ms_modulation_mae_db": 0.374029373138901, "300–1000 Hz_40ms_log_rms_mae_db": 0.48304020493637845, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03199644724974075, "300–1000 Hz_40ms_envelope_correlation": 0.9992098603296506, "300–1000 Hz_40ms_modulation_mae_db": 0.763731563933591, "1000–3000 Hz_10ms_log_rms_mae_db": 1.301578412606918, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09774039213194498, "1000–3000 Hz_10ms_envelope_correlation": 0.9940053659826046, "1000–3000 Hz_10ms_modulation_mae_db": 1.5551645930904758, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7778516734270543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06023180903055212, "1000–3000 Hz_40ms_envelope_correlation": 0.9983566221604986, "1000–3000 Hz_40ms_modulation_mae_db": 1.7169880579842973, "3000–6000 Hz_10ms_log_rms_mae_db": 1.572284904898318, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17136952543911727, "3000–6000 Hz_10ms_envelope_correlation": 0.9797658788437964, "3000–6000 Hz_10ms_modulation_mae_db": 1.284149131083036, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3180800040260885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11125546570645257, "3000–6000 Hz_40ms_envelope_correlation": 0.9914288342852889, "3000–6000 Hz_40ms_modulation_mae_db": 0.8533493040999728, "mr_stft_db": 5.489633204756569, "band_mae_db": 0.19737355492886932, "rms_gain_error_db": -0.07611397336631866, "output_peak": 0.46787932821851275}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8469467288694854, "全帯域_10ms_rms_envelope_nrmse": 0.046460138482529216, "全帯域_10ms_envelope_correlation": 0.9980594268491515, "全帯域_10ms_modulation_mae_db": 0.6777447772046464, "全帯域_40ms_log_rms_mae_db": 0.45358294368102287, "全帯域_40ms_rms_envelope_nrmse": 0.023515934545377475, "全帯域_40ms_envelope_correlation": 0.9995338398079214, "全帯域_40ms_modulation_mae_db": 0.7883287047312054, "300–1000 Hz_10ms_log_rms_mae_db": 0.9931697841650378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.056566159714248596, "300–1000 Hz_10ms_envelope_correlation": 0.9975491803580834, "300–1000 Hz_10ms_modulation_mae_db": 0.8888741947847416, "300–1000 Hz_40ms_log_rms_mae_db": 0.4694208412197246, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030457274277065162, "300–1000 Hz_40ms_envelope_correlation": 0.9993334970248078, "300–1000 Hz_40ms_modulation_mae_db": 1.0099331076482707, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4425769055121118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1168386419124345, "1000–3000 Hz_10ms_envelope_correlation": 0.9907070874297521, "1000–3000 Hz_10ms_modulation_mae_db": 1.1946735451712096, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8445410041276671, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06351575861595328, "1000–3000 Hz_40ms_envelope_correlation": 0.9977636855304967, "1000–3000 Hz_40ms_modulation_mae_db": 1.6396200483975376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.503521041571975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1746806273254002, "3000–6000 Hz_10ms_envelope_correlation": 0.9795646918398255, "3000–6000 Hz_10ms_modulation_mae_db": 1.4497280391666554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2480574063060885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12432363201636161, "3000–6000 Hz_40ms_envelope_correlation": 0.9902738151887046, "3000–6000 Hz_40ms_modulation_mae_db": 1.4341821732258395, "mr_stft_db": 5.518678887754113, "band_mae_db": 0.14579265046670614, "rms_gain_error_db": -0.07926548587776557, "output_peak": 0.46584969407503457}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7535902143925345, "全帯域_10ms_rms_envelope_nrmse": 0.05195013381783767, "全帯域_10ms_envelope_correlation": 0.9975199642703151, "全帯域_10ms_modulation_mae_db": 0.6941330937250805, "全帯域_40ms_log_rms_mae_db": 0.41653370810908336, "全帯域_40ms_rms_envelope_nrmse": 0.029353067688698463, "全帯域_40ms_envelope_correlation": 0.9991873773654075, "全帯域_40ms_modulation_mae_db": 0.6981243151398046, "300–1000 Hz_10ms_log_rms_mae_db": 0.9193158427173967, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.058584471685526586, "300–1000 Hz_10ms_envelope_correlation": 0.9973194268108103, "300–1000 Hz_10ms_modulation_mae_db": 0.6361037381391662, "300–1000 Hz_40ms_log_rms_mae_db": 0.44387666615107935, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033175669297431175, "300–1000 Hz_40ms_envelope_correlation": 0.9991357442181146, "300–1000 Hz_40ms_modulation_mae_db": 0.6298694559664506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4185336498980157, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12625387126537824, "1000–3000 Hz_10ms_envelope_correlation": 0.9896156337122962, "1000–3000 Hz_10ms_modulation_mae_db": 1.9851873997993252, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8501255537968857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09252057070082663, "1000–3000 Hz_40ms_envelope_correlation": 0.994949494548537, "1000–3000 Hz_40ms_modulation_mae_db": 2.1084871519188337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5061003566342408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1671627251028529, "3000–6000 Hz_10ms_envelope_correlation": 0.9813087296480021, "3000–6000 Hz_10ms_modulation_mae_db": 1.2277059748962806, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2035943378295497, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12048899391989328, "3000–6000 Hz_40ms_envelope_correlation": 0.9907409730372834, "3000–6000 Hz_40ms_modulation_mae_db": 1.3391929557095053, "mr_stft_db": 5.48849434082111, "band_mae_db": 0.15769878122672135, "rms_gain_error_db": -0.0510310312548684, "output_peak": 0.47231837084730943}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7331837410446194, "全帯域_10ms_rms_envelope_nrmse": 0.04869759437996548, "全帯域_10ms_envelope_correlation": 0.9978423736726081, "全帯域_10ms_modulation_mae_db": 0.7782040028115866, "全帯域_40ms_log_rms_mae_db": 0.4214304878302204, "全帯域_40ms_rms_envelope_nrmse": 0.024313771212149747, "全帯域_40ms_envelope_correlation": 0.9994780040315092, "全帯域_40ms_modulation_mae_db": 0.8793314906612446, "300–1000 Hz_10ms_log_rms_mae_db": 0.990312722532374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06172387994701173, "300–1000 Hz_10ms_envelope_correlation": 0.9970644128759847, "300–1000 Hz_10ms_modulation_mae_db": 1.2698026048004853, "300–1000 Hz_40ms_log_rms_mae_db": 0.5212897787150433, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030131884408400457, "300–1000 Hz_40ms_envelope_correlation": 0.9993573601388693, "300–1000 Hz_40ms_modulation_mae_db": 1.3864507225970801, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4785289170066083, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1170171789772551, "1000–3000 Hz_10ms_envelope_correlation": 0.9905295322120131, "1000–3000 Hz_10ms_modulation_mae_db": 0.8866015134586331, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8390194761984034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06468000331815121, "1000–3000 Hz_40ms_envelope_correlation": 0.9974755967127553, "1000–3000 Hz_40ms_modulation_mae_db": 0.9103153088411667, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4417158428600183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1606873167008553, "3000–6000 Hz_10ms_envelope_correlation": 0.9824350079359753, "3000–6000 Hz_10ms_modulation_mae_db": 0.42685733437885026, "3000–6000 Hz_40ms_log_rms_mae_db": 1.227841045196368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11357630137135749, "3000–6000 Hz_40ms_envelope_correlation": 0.991316863486161, "3000–6000 Hz_40ms_modulation_mae_db": 0.2582549339159196, "mr_stft_db": 5.512252417318258, "band_mae_db": 0.14608354827325218, "rms_gain_error_db": -0.07029454305680811, "output_peak": 0.4589002763339723}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37200, "output_samples": 37200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.33129051525633, "spectral_convergence": 0.14763102928519012, "spectral_convergence_status": "ok", "amplitude_floor": 1.3482074212160965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3217358687024605, "spectral_convergence": 0.1444062802598789, "spectral_convergence_status": "ok", "amplitude_floor": 1.080252897815134e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.829617078960644, "spectral_convergence": 0.144802250224761, "spectral_convergence_status": "ok", "amplitude_floor": 8.596818112459857e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004510568589627121, "output_power": 0.00044667758264908255, "delta_db": -0.042371341363429765, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019377343288488795, "output_power": 0.0019084238160867934, "delta_db": -0.06619405601625118, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013786956670426643, "output_power": 0.0013329491031885252, "delta_db": -0.14654843894785138, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031975863120548915, "output_power": 0.0003051404716898163, "delta_db": -0.20322462932910773, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.4830953182024825e-05, "output_power": 1.6136529965942522e-05, "delta_db": 0.36641084874104385, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.03312086795025e-05, "output_power": 1.9311402094918148e-05, "delta_db": -0.22349391317738088, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17470720459584413, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0999259264158839, "pitch": null, "output_rms_dbfs": -23.948416574814754, "output_peak": 0.4629269323437658, "reference_rms_dbfs": -23.848490648398872, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.494214487639812}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00019668187909197959, "log_rms_mae_db": 0.7843687756511099, "rms_envelope_nrmse": 0.04801872160598107, "envelope_correlation": 0.9979578091167112, "rms_gain_error_db": -0.0999259532265797, "reference_mean_power": 0.004131292024738224, "output_mean_power": 0.004037321172546179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.0026327800860698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011612269501705267, "output_power": 0.0011319008785701553, "reference_active": true, "delta_db": -0.1110870952249117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017095886275955477, "output_power": 0.00016115054331988278, "reference_active": true, "delta_db": -0.25659846105718354}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.217923048678998e-05, "output_power": 3.743796731450316e-05, "reference_active": true, "delta_db": -0.5178639146433588}], "active_band_count": 3, "active_band_mae_db": 0.29518315697515135}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00018115923356648065, "log_rms_mae_db": 0.45440212020501203, "rms_envelope_nrmse": 0.027953575998766352, "envelope_correlation": 0.999352492073329, "rms_gain_error_db": -0.0999259553321045, "reference_mean_power": 0.004185412442455258, "output_mean_power": 0.004090210558921023}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.7728315536386115e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066252574844359, "output_power": 0.0010412376763627824, "reference_active": true, "delta_db": -0.103102186103711}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.440983264620606e-05, "output_power": 7.944868615074671e-05, "reference_active": true, "delta_db": -0.26306319583251936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.940091323355611e-06, "output_power": 2.2774656574728265e-06, "reference_active": true, "delta_db": -1.109089835675197}], "active_band_count": 3, "active_band_mae_db": 0.4917517392038091}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.000176021246464811, "log_rms_mae_db": 0.9357537075035861, "rms_envelope_nrmse": 0.0633584876255258, "envelope_correlation": 0.9969482709744235, "rms_gain_error_db": -0.13445149461826078, "reference_mean_power": 0.0026395436088586166, "output_mean_power": 0.0025590789946823163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.515899729654865e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009091516678957249, "output_power": 0.0008640468672440648, "reference_active": true, "delta_db": -0.22099039881859223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001833276731517994, "output_power": 0.00016398455824086144, "reference_active": true, "delta_db": -0.48425072055050056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7517429693718724e-05, "output_power": 3.052531332895676e-05, "reference_active": true, "delta_db": -0.8957294687747599}], "active_band_count": 3, "active_band_mae_db": 0.5336568627146175}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00016136314602147107, "log_rms_mae_db": 0.4760778372572405, "rms_envelope_nrmse": 0.03784316876478425, "envelope_correlation": 0.9989887099767392, "rms_gain_error_db": -0.13445151576888892, "reference_mean_power": 0.002674122622482191, "output_mean_power": 0.0025926038789828992}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.348456388983372e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008380618007305507, "output_power": 0.0007986121308227355, "reference_active": true, "delta_db": -0.20940142960694907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.026333057603631e-05, "output_power": 8.246610903634107e-05, "reference_active": true, "delta_db": -0.39235850355350016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498114091574834e-06, "output_power": 1.9408888949261544e-06, "reference_active": true, "delta_db": -1.096115940293101}], "active_band_count": 3, "active_band_mae_db": 0.5659586244845168}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 6.667691377903701e-05, "log_rms_mae_db": 1.3458885740956112, "rms_envelope_nrmse": 0.10688638243481553, "envelope_correlation": 0.9920060754480444, "rms_gain_error_db": -0.20894410715653766, "reference_mean_power": 0.0003102146198149334, "output_mean_power": 0.00029564317027147146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.247013787334141e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018435630971779633, "output_power": 0.00017425849097392362, "reference_active": true, "delta_db": -0.24464057360273322}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0328682781544803e-05, "output_power": 1.6901578619818023e-05, "reference_active": true, "delta_db": -0.8018196904969206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.580926194982421e-06, "output_power": 3.2781243450229712e-06, "reference_active": true, "delta_db": -1.4532787165080743}], "active_band_count": 3, "active_band_mae_db": 0.8332463268692427}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 6.131102329922278e-05, "log_rms_mae_db": 0.7420258604680805, "rms_envelope_nrmse": 0.05905870043246649, "envelope_correlation": 0.99774473089138, "rms_gain_error_db": -0.20894479408391828, "reference_mean_power": 0.0003142782197045969, "output_mean_power": 0.00029951584673814166}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 3.052275331975406e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017569545994603763, "output_power": 0.00016641905722488274, "reference_active": true, "delta_db": -0.23557481891411605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.589536328721284e-06, "output_power": 7.3407538313449415e-06, "reference_active": true, "delta_db": -0.6822906037919019}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8479520444652144e-07, "output_power": 2.421200773204506e-07, "reference_active": true, "delta_db": -0.7050186818246749}], "active_band_count": 3, "active_band_mae_db": 0.5409613681768976}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 2.0872644535346135e-05, "log_rms_mae_db": 1.4927578020635988, "rms_envelope_nrmse": 0.1827947901559848, "envelope_correlation": 0.97693279499881, "rms_gain_error_db": -0.14408171380731336, "reference_mean_power": 1.984318614204444e-05, "output_mean_power": 1.9195668222381693e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.6220002221112295e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.352365504647456e-06, "output_power": 9.11632793944574e-06, "reference_active": true, "delta_db": -0.1110153181435864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.791249366368675e-07, "output_power": 9.146512742396057e-07, "reference_active": true, "delta_db": -0.29582567697407597}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495952213974179e-07, "output_power": 6.587354825183867e-07, "reference_active": true, "delta_db": 4.2147479082237025}], "active_band_count": 3, "active_band_mae_db": 1.5405296344471218}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 1.9522833601394496e-05, "log_rms_mae_db": 1.3129945424027885, "rms_envelope_nrmse": 0.11374380238585154, "envelope_correlation": 0.9908068757939648, "rms_gain_error_db": -0.1440780820524462, "reference_mean_power": 2.010291979320635e-05, "output_mean_power": 1.9446942571928393e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.5131001811961146e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.175181026763155e-06, "output_power": 8.302960378843884e-06, "reference_active": true, "delta_db": 0.06735587285138556}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.283226898133603e-07, "output_power": 3.430634413806762e-07, "reference_active": true, "delta_db": -0.9639664131703751}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.754857048422832e-08, "output_power": 3.1353731028408116e-08, "reference_active": true, "delta_db": 2.5204748399076338}], "active_band_count": 3, "active_band_mae_db": 1.1839323753097981}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9466998305706352, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "48e872b59abb028cd76edc9056d8bc9391c83be78f4c66d48ff38e320d9aa73a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9918529494666262, "standard_deviation": 0.0014487288577464228, "minimum": 0.9896156337122962, "maximum": 0.9940053659826046, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3994097919251476, "standard_deviation": 0.059301900278622206, "minimum": 1.301578412606918, "maximum": 1.4785289170066083, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2076344992003796, "standard_deviation": 0.40317660472393546, "minimum": 0.8318336337252341, "maximum": 1.9851873997993252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10892884146442804, "standard_deviation": 0.009974393105448726, "minimum": 0.09774039213194498, "maximum": 0.12625387126537824, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974348308087525, "standard_deviation": 0.0010494257541012275, "minimum": 0.994949494548537, "maximum": 0.9983566221604986, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8138330207452326, "standard_deviation": 0.04742409662380742, "minimum": 0.7420258604680805, "maximum": 0.8831384467412801, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3315903299097998, "standard_deviation": 0.49727839854897443, "minimum": 0.5409613681768976, "maximum": 2.1084871519188337, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06426166772912661, "standard_deviation": 0.012089439793928038, "minimum": 0.05337524637204808, "maximum": 0.09252057070082663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.979766087956776, "standard_deviation": 0.0018887344123843886, "minimum": 0.97693279499881, "maximum": 0.9824350079359753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5129253748018632, "standard_deviation": 0.03780913098676091, "minimum": 1.4417158428600183, "maximum": 1.572284904898318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.284458355082133, "standard_deviation": 0.5434083840180336, "minimum": 0.42685733437885026, "maximum": 2.089952822185167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1727135686485207, "standard_deviation": 0.008945652206820242, "minimum": 0.1606873167008553, "maximum": 0.18819616957098426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9904560070894473, "standard_deviation": 0.0008706102593163567, "minimum": 0.9886538900941243, "maximum": 0.9914288342852889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.268259179958788, "standard_deviation": 0.04034650994617831, "minimum": 1.2035943378295497, "maximum": 1.3180800040260885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.153981364337752, "standard_deviation": 0.6067951922859603, "minimum": 0.2582549339159196, "maximum": 1.9680705679259207, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11886589347042474, "standard_deviation": 0.006725606434089813, "minimum": 0.11125546570645257, "maximum": 0.13189094811733962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9971502909552248, "standard_deviation": 0.00035776331483598463, "minimum": 0.9964555225853445, "maximum": 0.9975491803580834, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9747277056832461, "standard_deviation": 0.04601026978638985, "minimum": 0.9193158427173967, "maximum": 1.0592128059222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.712499670339989, "standard_deviation": 0.3156877256379084, "minimum": 0.33149688581683684, "maximum": 1.2698026048004853, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06054952422165784, "standard_deviation": 0.0037720604467825665, "minimum": 0.05644972396814097, "maximum": 0.0676883361754409, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9992195534918336, "standard_deviation": 0.00014475184013626802, "minimum": 0.9989887099767392, "maximum": 0.9994026356273643, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.48785827135710214, "standard_deviation": 0.034720787988426184, "minimum": 0.44387666615107935, "maximum": 0.5385691952709947, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7642435304710842, "standard_deviation": 0.32828639655007047, "minimum": 0.2664259139093502, "maximum": 1.3864507225970801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03236221253264367, "standard_deviation": 0.0030654564494515396, "minimum": 0.02953938613001904, "maximum": 0.03784316876478425, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16194941803050042, "standard_deviation": 0.02456509691115925, "minimum": 0.12655816110015564, "maximum": 0.19737355492886932, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.510632700024955, "standard_deviation": 0.01756606191797122, "minimum": 5.48849434082111, "maximum": 5.530810202804315, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.464365370145972, "standard_deviation": 0.005967817086537631, "minimum": 0.45389162518641607, "maximum": 0.47231837084730943, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07029155704796376, "standard_deviation": 0.027358206642115444, "minimum": -0.0999259264158839, "maximum": -0.012570039044973304, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9978984240453332, "standard_deviation": 0.00029036352606665283, "minimum": 0.9975199642703151, "maximum": 0.9984173820273743, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7902590175416444, "standard_deviation": 0.037636674806205146, "minimum": 0.7331837410446194, "maximum": 0.8469467288694854, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5075688844703287, "standard_deviation": 0.20114341978848907, "minimum": 0.29518315697515135, "maximum": 0.7782040028115866, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.048105449391437086, "standard_deviation": 0.003171247432620078, "minimum": 0.042069288467090696, "maximum": 0.05195013381783767, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999457224342206, "standard_deviation": 0.00013266290962155256, "minimum": 0.9991873773654075, "maximum": 0.9995895859104496, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4276470266433949, "standard_deviation": 0.022425338352045847, "minimum": 0.38420285735447623, "maximum": 0.45440212020501203, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5696771313765899, "standard_deviation": 0.20660330289278755, "minimum": 0.24357274855737077, "maximum": 0.8793314906612446, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.024848839581031297, "standard_deviation": 0.0025394359811583395, "minimum": 0.022228303719940862, "maximum": 0.029353067688698463, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0022", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.0285823345184326, "predicted_mos": 3.0285823345184326, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "821e24a6ae0a575960ea180801faba1a6078869712a643d6bc3d102c73572b91", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.071183699648827, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7326557279496707, "全帯域_10ms_rms_envelope_nrmse": 0.055201717712743945, "全帯域_10ms_envelope_correlation": 0.9971871981810784, "全帯域_10ms_modulation_mae_db": 0.07797082806008812, "全帯域_40ms_log_rms_mae_db": 0.39386748298173085, "全帯域_40ms_rms_envelope_nrmse": 0.028502511278406573, "全帯域_40ms_envelope_correlation": 0.9992277164657467, "全帯域_40ms_modulation_mae_db": 0.22640303232274325, "300–1000 Hz_10ms_log_rms_mae_db": 1.0215626299136287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06499308557326841, "300–1000 Hz_10ms_envelope_correlation": 0.9966452892819535, "300–1000 Hz_10ms_modulation_mae_db": 0.6820359971337843, "300–1000 Hz_40ms_log_rms_mae_db": 0.5054944184225035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03157713521807611, "300–1000 Hz_40ms_envelope_correlation": 0.9991977847838344, "300–1000 Hz_40ms_modulation_mae_db": 0.5237338386904872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5042440093668281, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1180386212090109, "1000–3000 Hz_10ms_envelope_correlation": 0.9910267425448434, "1000–3000 Hz_10ms_modulation_mae_db": 1.3785190578000976, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9182687612591769, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08061301924960332, "1000–3000 Hz_40ms_envelope_correlation": 0.9964638059004324, "1000–3000 Hz_40ms_modulation_mae_db": 1.7128174074893956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7475725381998575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20880223710020132, "3000–6000 Hz_10ms_envelope_correlation": 0.9703265632291882, "3000–6000 Hz_10ms_modulation_mae_db": 1.3044972314236625, "3000–6000 Hz_40ms_log_rms_mae_db": 1.460605439186536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13933814018728125, "3000–6000 Hz_40ms_envelope_correlation": 0.9863098645335207, "3000–6000 Hz_40ms_modulation_mae_db": 1.0649648496839814, "mr_stft_db": 5.619920651152963, "band_mae_db": 0.14682431700712698, "rms_gain_error_db": -0.04749336940765911, "output_peak": 0.4749742704410223}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7579806127262718, "全帯域_10ms_rms_envelope_nrmse": 0.06507469787703854, "全帯域_10ms_envelope_correlation": 0.9961046061638759, "全帯域_10ms_modulation_mae_db": 0.24345675604336767, "全帯域_40ms_log_rms_mae_db": 0.4365519328323944, "全帯域_40ms_rms_envelope_nrmse": 0.030897857072802033, "全帯域_40ms_envelope_correlation": 0.9991225463460565, "全帯域_40ms_modulation_mae_db": 0.15164554537403466, "300–1000 Hz_10ms_log_rms_mae_db": 1.0196636184443577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07344145166177585, "300–1000 Hz_10ms_envelope_correlation": 0.9956952232944787, "300–1000 Hz_10ms_modulation_mae_db": 0.4740931096631658, "300–1000 Hz_40ms_log_rms_mae_db": 0.5377881288522103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03677167262973985, "300–1000 Hz_40ms_envelope_correlation": 0.998889665740315, "300–1000 Hz_40ms_modulation_mae_db": 0.43213320778296954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4129685514637231, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12315165020357616, "1000–3000 Hz_10ms_envelope_correlation": 0.9892250581977667, "1000–3000 Hz_10ms_modulation_mae_db": 0.47691979217464797, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8763102430918944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07409194573815313, "1000–3000 Hz_40ms_envelope_correlation": 0.996170499263896, "1000–3000 Hz_40ms_modulation_mae_db": 0.45620381430104917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6453280412541975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1589212405148946, "3000–6000 Hz_10ms_envelope_correlation": 0.983054886236812, "3000–6000 Hz_10ms_modulation_mae_db": 0.9382292357179011, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3143022369318058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11532680808420312, "3000–6000 Hz_40ms_envelope_correlation": 0.9913607859176187, "3000–6000 Hz_40ms_modulation_mae_db": 0.6136635323026302, "mr_stft_db": 5.604746365361816, "band_mae_db": 0.12037282599838323, "rms_gain_error_db": -0.07005347922291895, "output_peak": 0.470295530276792}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7956235717913238, "全帯域_10ms_rms_envelope_nrmse": 0.057241867268474765, "全帯域_10ms_envelope_correlation": 0.9969705581341635, "全帯域_10ms_modulation_mae_db": 0.2926238342400811, "全帯域_40ms_log_rms_mae_db": 0.44853737290008067, "全帯域_40ms_rms_envelope_nrmse": 0.031756336180379416, "全帯域_40ms_envelope_correlation": 0.9990326516944152, "全帯域_40ms_modulation_mae_db": 0.26642543546181063, "300–1000 Hz_10ms_log_rms_mae_db": 1.105477997948779, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07018011503983343, "300–1000 Hz_10ms_envelope_correlation": 0.9960751126867909, "300–1000 Hz_10ms_modulation_mae_db": 0.35484150995261743, "300–1000 Hz_40ms_log_rms_mae_db": 0.48354217130181865, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04141484425691928, "300–1000 Hz_40ms_envelope_correlation": 0.9985856491338285, "300–1000 Hz_40ms_modulation_mae_db": 0.40296598542373446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4804025028884886, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12101794600686597, "1000–3000 Hz_10ms_envelope_correlation": 0.989433163010687, "1000–3000 Hz_10ms_modulation_mae_db": 0.12331042473123764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8739343708878357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06515189590894507, "1000–3000 Hz_40ms_envelope_correlation": 0.9968083777818255, "1000–3000 Hz_40ms_modulation_mae_db": 0.1999797003872179, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7374580072411634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20375317275757646, "3000–6000 Hz_10ms_envelope_correlation": 0.9713597785158492, "3000–6000 Hz_10ms_modulation_mae_db": 2.179898319858401, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3452427295000275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1480240761834903, "3000–6000 Hz_40ms_envelope_correlation": 0.9849696639129387, "3000–6000 Hz_40ms_modulation_mae_db": 1.8191147392689662, "mr_stft_db": 5.636451601169033, "band_mae_db": 0.13927886176520918, "rms_gain_error_db": -0.0408904458615644, "output_peak": 0.47083677272355595}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7293745822427187, "全帯域_10ms_rms_envelope_nrmse": 0.04693602278170269, "全帯域_10ms_envelope_correlation": 0.997987032377939, "全帯域_10ms_modulation_mae_db": 0.3990005949708512, "全帯域_40ms_log_rms_mae_db": 0.36891012622939356, "全帯域_40ms_rms_envelope_nrmse": 0.02665053266885918, "全帯域_40ms_envelope_correlation": 0.9993439218489687, "全帯域_40ms_modulation_mae_db": 0.43590320945968514, "300–1000 Hz_10ms_log_rms_mae_db": 1.003440430062521, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06192060615256715, "300–1000 Hz_10ms_envelope_correlation": 0.9970183883035127, "300–1000 Hz_10ms_modulation_mae_db": 0.8262897323951846, "300–1000 Hz_40ms_log_rms_mae_db": 0.5262229285744887, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035708767669202356, "300–1000 Hz_40ms_envelope_correlation": 0.9990276924718794, "300–1000 Hz_40ms_modulation_mae_db": 0.8984432030024493, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4226380411564155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10140954758148522, "1000–3000 Hz_10ms_envelope_correlation": 0.9927103108693306, "1000–3000 Hz_10ms_modulation_mae_db": 0.6079962474714213, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7892003910860029, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05226676376010166, "1000–3000 Hz_40ms_envelope_correlation": 0.9981472353755771, "1000–3000 Hz_40ms_modulation_mae_db": 0.6347494906281254, "3000–6000 Hz_10ms_log_rms_mae_db": 1.761701079787678, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16877843123360892, "3000–6000 Hz_10ms_envelope_correlation": 0.9805713545756403, "3000–6000 Hz_10ms_modulation_mae_db": 1.8496619961235936, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3462733343999538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12275069527484297, "3000–6000 Hz_40ms_envelope_correlation": 0.989885145749262, "3000–6000 Hz_40ms_modulation_mae_db": 2.240161623783412, "mr_stft_db": 5.630814057207832, "band_mae_db": 0.12218793101255572, "rms_gain_error_db": -0.05686471292435002, "output_peak": 0.4756299341821797}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7532142692005834, "全帯域_10ms_rms_envelope_nrmse": 0.04737672673710702, "全帯域_10ms_envelope_correlation": 0.9980839171480904, "全帯域_10ms_modulation_mae_db": 0.0549935851488057, "全帯域_40ms_log_rms_mae_db": 0.385758420912306, "全帯域_40ms_rms_envelope_nrmse": 0.024301818053758455, "全帯域_40ms_envelope_correlation": 0.9996186726765169, "全帯域_40ms_modulation_mae_db": 0.2587521236898233, "300–1000 Hz_10ms_log_rms_mae_db": 1.0485266643921904, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05779026269277327, "300–1000 Hz_10ms_envelope_correlation": 0.9975227471343365, "300–1000 Hz_10ms_modulation_mae_db": 0.3442102133336878, "300–1000 Hz_40ms_log_rms_mae_db": 0.5291158568900403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03346864356867745, "300–1000 Hz_40ms_envelope_correlation": 0.9992856090312426, "300–1000 Hz_40ms_modulation_mae_db": 0.2524959739087094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4489520361082326, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10861882178308518, "1000–3000 Hz_10ms_envelope_correlation": 0.9923150402150445, "1000–3000 Hz_10ms_modulation_mae_db": 1.5528499917435188, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8029926468440751, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06374995242311685, "1000–3000 Hz_40ms_envelope_correlation": 0.9979959653226756, "1000–3000 Hz_40ms_modulation_mae_db": 1.395032953260683, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7030724930791923, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18532359705814372, "3000–6000 Hz_10ms_envelope_correlation": 0.9763623885079525, "3000–6000 Hz_10ms_modulation_mae_db": 1.4314639294131677, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4084966321279329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12896938113334244, "3000–6000 Hz_40ms_envelope_correlation": 0.9886408461045627, "3000–6000 Hz_40ms_modulation_mae_db": 1.1427879239860452, "mr_stft_db": 5.6334409424194805, "band_mae_db": 0.17962162794300504, "rms_gain_error_db": -0.12363682365851053, "output_peak": 0.46731353379064877}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8181473675535634, "全帯域_10ms_rms_envelope_nrmse": 0.05412273441865757, "全帯域_10ms_envelope_correlation": 0.997432595275709, "全帯域_10ms_modulation_mae_db": 0.4442255263338595, "全帯域_40ms_log_rms_mae_db": 0.4121923307965157, "全帯域_40ms_rms_envelope_nrmse": 0.03229009904163046, "全帯域_40ms_envelope_correlation": 0.9991585714814681, "全帯域_40ms_modulation_mae_db": 0.5645585595987628, "300–1000 Hz_10ms_log_rms_mae_db": 1.054164256107286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06713937807037322, "300–1000 Hz_10ms_envelope_correlation": 0.9966614038551936, "300–1000 Hz_10ms_modulation_mae_db": 0.5955915018544946, "300–1000 Hz_40ms_log_rms_mae_db": 0.538177983419954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0420436902504598, "300–1000 Hz_40ms_envelope_correlation": 0.99881520061984, "300–1000 Hz_40ms_modulation_mae_db": 0.90142064732785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4964825600858513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10808823409579263, "1000–3000 Hz_10ms_envelope_correlation": 0.9916588954352288, "1000–3000 Hz_10ms_modulation_mae_db": 0.6318787128145024, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8404464979960246, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055430710643359915, "1000–3000 Hz_40ms_envelope_correlation": 0.9978527486446545, "1000–3000 Hz_40ms_modulation_mae_db": 0.5760633117352723, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7233510913100747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18794967167129453, "3000–6000 Hz_10ms_envelope_correlation": 0.9758957563603602, "3000–6000 Hz_10ms_modulation_mae_db": 1.107678909950331, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3198062468642717, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12475556721392693, "3000–6000 Hz_40ms_envelope_correlation": 0.9898739674216976, "3000–6000 Hz_40ms_modulation_mae_db": 1.2891771398702458, "mr_stft_db": 5.630739569202073, "band_mae_db": 0.1547256441804221, "rms_gain_error_db": -0.12308868840644135, "output_peak": 0.4664731763640246}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7680570465930776, "全帯域_10ms_rms_envelope_nrmse": 0.04884833294492951, "全帯域_10ms_envelope_correlation": 0.9978439306908207, "全帯域_10ms_modulation_mae_db": 0.4827927236313599, "全帯域_40ms_log_rms_mae_db": 0.40271298972059755, "全帯域_40ms_rms_envelope_nrmse": 0.024600270670144652, "全帯域_40ms_envelope_correlation": 0.9994806760158288, "全帯域_40ms_modulation_mae_db": 0.39851794285184233, "300–1000 Hz_10ms_log_rms_mae_db": 1.0364010547021014, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.054518968609507265, "300–1000 Hz_10ms_envelope_correlation": 0.99771530574048, "300–1000 Hz_10ms_modulation_mae_db": 0.6118583532581302, "300–1000 Hz_40ms_log_rms_mae_db": 0.4968352733805412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.028188210824538405, "300–1000 Hz_40ms_envelope_correlation": 0.9994339701319311, "300–1000 Hz_40ms_modulation_mae_db": 0.6002782199025356, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4945346939230855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1193720235162638, "1000–3000 Hz_10ms_envelope_correlation": 0.9898055525031205, "1000–3000 Hz_10ms_modulation_mae_db": 0.5055027641758668, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8240368108879191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06422263584426087, "1000–3000 Hz_40ms_envelope_correlation": 0.9970772523521365, "1000–3000 Hz_40ms_modulation_mae_db": 0.6937689820305449, "3000–6000 Hz_10ms_log_rms_mae_db": 1.668340341162083, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16767529651593788, "3000–6000 Hz_10ms_envelope_correlation": 0.9809495084323385, "3000–6000 Hz_10ms_modulation_mae_db": 1.4378746473103945, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3383887987720309, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12046619894529159, "3000–6000 Hz_40ms_envelope_correlation": 0.9904536975117574, "3000–6000 Hz_40ms_modulation_mae_db": 1.4060832086703037, "mr_stft_db": 5.625055849552555, "band_mae_db": 0.06275416745285257, "rms_gain_error_db": -0.07799641291381804, "output_peak": 0.47253933813588994}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7760693286884871, "全帯域_10ms_rms_envelope_nrmse": 0.05433699103597503, "全帯域_10ms_envelope_correlation": 0.9974405957363466, "全帯域_10ms_modulation_mae_db": 0.845715138704327, "全帯域_40ms_log_rms_mae_db": 0.47127606080478796, "全帯域_40ms_rms_envelope_nrmse": 0.031918039002397966, "全帯域_40ms_envelope_correlation": 0.99921830567588, "全帯域_40ms_modulation_mae_db": 0.9438043235444177, "300–1000 Hz_10ms_log_rms_mae_db": 0.9974804796225148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06675934308065591, "300–1000 Hz_10ms_envelope_correlation": 0.9966198367321621, "300–1000 Hz_10ms_modulation_mae_db": 0.9926620983353963, "300–1000 Hz_40ms_log_rms_mae_db": 0.5254521289762664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03938448549554212, "300–1000 Hz_40ms_envelope_correlation": 0.9989144192559294, "300–1000 Hz_40ms_modulation_mae_db": 1.214424536651256, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4784869328006591, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12485936413362535, "1000–3000 Hz_10ms_envelope_correlation": 0.9896865700511622, "1000–3000 Hz_10ms_modulation_mae_db": 1.08067787113164, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8797189761973317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07689088793186374, "1000–3000 Hz_40ms_envelope_correlation": 0.996810106373358, "1000–3000 Hz_40ms_modulation_mae_db": 1.3413273560200576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7013392850156899, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17114132595666015, "3000–6000 Hz_10ms_envelope_correlation": 0.9798494262094327, "3000–6000 Hz_10ms_modulation_mae_db": 0.7862792613368267, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3118809894703167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1118223300799275, "3000–6000 Hz_40ms_envelope_correlation": 0.9914945983931764, "3000–6000 Hz_40ms_modulation_mae_db": 1.0782052798280242, "mr_stft_db": 5.62601547077234, "band_mae_db": 0.17813110478644167, "rms_gain_error_db": -0.1346252869616873, "output_peak": 0.4620900736469932}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37200, "output_samples": 37200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.524011675500204, "spectral_convergence": 0.1506016082297362, "spectral_convergence_status": "ok", "amplitude_floor": 1.3482074212160965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.454950368919396, "spectral_convergence": 0.1483837693120489, "spectral_convergence_status": "ok", "amplitude_floor": 1.080252897815134e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.880799909039289, "spectral_convergence": 0.14759244450813339, "spectral_convergence_status": "ok", "amplitude_floor": 8.596818112459857e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004510568589627121, "output_power": 0.00044736615184847074, "delta_db": -0.0356816940611476, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019377343288488795, "output_power": 0.0019080801881227202, "delta_db": -0.0669761106341072, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013786956670426643, "output_power": 0.0013968328962610525, "delta_db": 0.056760436869120814, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031975863120548915, "output_power": 0.00028888859323123627, "delta_db": -0.44091882105443464, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.4830953182024825e-05, "output_power": 1.5095138897030074e-05, "delta_db": 0.07668049578529067, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.03312086795025e-05, "output_power": 2.130865472454593e-05, "delta_db": 0.20392834363866086, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14682431700712698, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04749336940765911, "pitch": null, "output_rms_dbfs": -23.89598401780653, "output_peak": 0.4749742704410223, "reference_rms_dbfs": -23.848490648398872, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.619920651152963}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00019668187909197959, "log_rms_mae_db": 0.7326557279496707, "rms_envelope_nrmse": 0.055201717712743945, "envelope_correlation": 0.9971871981810784, "rms_gain_error_db": -0.04749337496788711, "reference_mean_power": 0.004131292024738224, "output_mean_power": 0.004086359366637918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.0026327800860698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011612269501705267, "output_power": 0.0011839969653338381, "reference_active": true, "delta_db": 0.08433482723195941}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017095886275955477, "output_power": 0.00016897564655039113, "reference_active": true, "delta_db": -0.050675033258176246}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.217923048678998e-05, "output_power": 4.122953055427629e-05, "reference_active": true, "delta_db": -0.0989026236901287}], "active_band_count": 3, "active_band_mae_db": 0.07797082806008812}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00018115923356648065, "log_rms_mae_db": 0.39386748298173085, "rms_envelope_nrmse": 0.028502511278406573, "envelope_correlation": 0.9992277164657467, "rms_gain_error_db": -0.04749330670172974, "reference_mean_power": 0.004185412442455258, "output_mean_power": 0.0041398912262550465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.7728315536386115e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066252574844359, "output_power": 0.0010842565372187133, "reference_active": true, "delta_db": 0.07271956350035176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.440983264620606e-05, "output_power": 8.64239578925622e-05, "reference_active": true, "delta_db": 0.10241112262618897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.940091323355611e-06, "output_power": 3.3019360778165635e-06, "reference_active": true, "delta_db": 0.504078410841689}], "active_band_count": 3, "active_band_mae_db": 0.22640303232274325}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.000176021246464811, "log_rms_mae_db": 1.0215626299136287, "rms_envelope_nrmse": 0.06499308557326841, "envelope_correlation": 0.9966452892819535, "rms_gain_error_db": -0.061522363247085356, "reference_mean_power": 0.0026395436088586166, "output_mean_power": 0.002602415309845428}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.515899729654865e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009091516678957249, "output_power": 0.0009086558466169987, "reference_active": true, "delta_db": -0.00236914468646316}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001833276731517994, "output_power": 0.00016850585117149548, "reference_active": true, "delta_db": -0.36613040437877464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7517429693718724e-05, "output_power": 2.5496010723394308e-05, "reference_active": true, "delta_db": -1.6776084423361148}], "active_band_count": 3, "active_band_mae_db": 0.6820359971337843}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00016136314602147107, "log_rms_mae_db": 0.5054944184225035, "rms_envelope_nrmse": 0.03157713521807611, "envelope_correlation": 0.9991977847838344, "rms_gain_error_db": -0.06152236205058713, "reference_mean_power": 0.002674122622482191, "output_mean_power": 0.0026365079295197573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.348456388983372e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008380618007305507, "output_power": 0.0008339183954957641, "reference_active": true, "delta_db": -0.021524916592154268}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.026333057603631e-05, "output_power": 8.743309389312543e-05, "reference_active": true, "delta_db": -0.13835507734782673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498114091574834e-06, "output_power": 1.80501209754017e-06, "reference_active": true, "delta_db": -1.4113215221314803}], "active_band_count": 3, "active_band_mae_db": 0.5237338386904872}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 6.667691377903701e-05, "log_rms_mae_db": 1.5042440093668281, "rms_envelope_nrmse": 0.1180386212090109, "envelope_correlation": 0.9910267425448434, "rms_gain_error_db": -0.398644234573399, "reference_mean_power": 0.0003102146198149334, "output_mean_power": 0.0002830074302856416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.247013787334141e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018435630971779633, "output_power": 0.00016583051917234582, "reference_active": true, "delta_db": -0.4599354581259702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0328682781544803e-05, "output_power": 1.4524254063104385e-05, "reference_active": true, "delta_db": -1.4601540184893347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.580926194982421e-06, "output_power": 2.7504675472174337e-06, "reference_active": true, "delta_db": -2.215467696784988}], "active_band_count": 3, "active_band_mae_db": 1.3785190578000976}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 6.131102329922278e-05, "log_rms_mae_db": 0.9182687612591769, "rms_envelope_nrmse": 0.08061301924960332, "envelope_correlation": 0.9964638059004324, "rms_gain_error_db": -0.39864410732492883, "reference_mean_power": 0.0003142782197045969, "output_mean_power": 0.00028671464295409874}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 3.052275331975406e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017569545994603763, "output_power": 0.00015776735909199104, "reference_active": true, "delta_db": -0.4674338343701381}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.589536328721284e-06, "output_power": 5.880756184484062e-06, "reference_active": true, "delta_db": -1.6453654686491244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8479520444652144e-07, "output_power": 1.4189509403712705e-07, "reference_active": true, "delta_db": -3.025652919448925}], "active_band_count": 3, "active_band_mae_db": 1.7128174074893956}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 2.0872644535346135e-05, "log_rms_mae_db": 1.7475725381998575, "rms_envelope_nrmse": 0.20880223710020132, "envelope_correlation": 0.9703265632291882, "rms_gain_error_db": 0.06766671236258515, "reference_mean_power": 1.984318614204444e-05, "output_mean_power": 2.0154780725854395e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.6220002221112295e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.352365504647456e-06, "output_power": 9.195092271129386e-06, "reference_active": true, "delta_db": -0.07365379708875076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.791249366368675e-07, "output_power": 8.392974319524862e-07, "reference_active": true, "delta_db": -0.6692221716683906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495952213974179e-07, "output_power": 5.179619281547703e-07, "reference_active": true, "delta_db": 3.170615725513846}], "active_band_count": 3, "active_band_mae_db": 1.3044972314236625}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 1.9522833601394496e-05, "log_rms_mae_db": 1.460605439186536, "rms_envelope_nrmse": 0.13933814018728125, "envelope_correlation": 0.9863098645335207, "rms_gain_error_db": 0.06766496385861408, "reference_mean_power": 2.010291979320635e-05, "output_mean_power": 2.0418584715003554e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.5131001811961146e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.175181026763155e-06, "output_power": 8.251087617457438e-06, "reference_active": true, "delta_db": 0.04013820648781644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.283226898133603e-07, "output_power": 3.552627665257655e-07, "reference_active": true, "delta_db": -0.812213877213044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.754857048422832e-08, "output_power": 3.00951138720742e-08, "reference_active": true, "delta_db": 2.3425424653510833}], "active_band_count": 3, "active_band_mae_db": 1.0649648496839814}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9426332355611222, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dd31f41bf99b6207e264139ca1ff7a027b7ed6770d92173137f54637f97f37fb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.990732666603398, "standard_deviation": 0.0013770991686146725, "minimum": 0.9892250581977667, "maximum": 0.9927103108693306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4673386659741605, "standard_deviation": 0.034924952046359044, "minimum": 1.4129685514637231, "maximum": 1.5042440093668281, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7947068577553665, "standard_deviation": 0.4919598954146064, "minimum": 0.12331042473123764, "maximum": 1.5528499917435188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11556952606621315, "standard_deviation": 0.008441541883115784, "minimum": 0.10140954758148522, "maximum": 0.12485936413362535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971657488768195, "standard_deviation": 0.0007435260427784973, "minimum": 0.996170499263896, "maximum": 0.9981472353755771, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8506135872812826, "standard_deviation": 0.04387897752560456, "minimum": 0.7892003910860029, "maximum": 0.9182687612591769, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8762428769815432, "standard_deviation": 0.5346282965842585, "minimum": 0.1999797003872179, "maximum": 1.7128174074893956, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06655222643742557, "standard_deviation": 0.010031695471085803, "minimum": 0.05226676376010166, "maximum": 0.08061301924960332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9772962077584467, "standard_deviation": 0.0046353442512205014, "minimum": 0.9703265632291882, "maximum": 0.983054886236812, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.711020359631242, "standard_deviation": 0.03973043052387731, "minimum": 1.6453280412541975, "maximum": 1.761701079787678, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3794479413917848, "standard_deviation": 0.4619740953218432, "minimum": 0.7862792613368267, "maximum": 2.179898319858401, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1815431216010397, "standard_deviation": 0.01799556444527192, "minimum": 0.1589212405148946, "maximum": 0.20880223710020132, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9891235711930668, "standard_deviation": 0.0023586963834354054, "minimum": 0.9849696639129387, "maximum": 0.9914945983931764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3556245509066094, "standard_deviation": 0.052389647557053665, "minimum": 1.3118809894703167, "maximum": 1.460605439186536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.331769787174201, "standard_deviation": 0.5009830366851514, "minimum": 0.6136635323026302, "maximum": 2.240161623783412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12643164963778825, "standard_deviation": 0.012118141457756713, "minimum": 0.1118223300799275, "maximum": 0.1480240761834903, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9967441633786135, "standard_deviation": 0.0006765982063847242, "minimum": 0.9956952232944787, "maximum": 0.99771530574048, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0358396413991724, "standard_deviation": 0.03450299967806286, "minimum": 0.9974804796225148, "maximum": 1.105477997948779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6101978144908076, "standard_deviation": 0.22450969070925542, "minimum": 0.3442102133336878, "maximum": 0.9926620983353963, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06459290136009432, "standard_deviation": 0.006276408612428462, "minimum": 0.054518968609507265, "maximum": 0.07344145166177585, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9990187488961001, "standard_deviation": 0.00027576161621216486, "minimum": 0.9985856491338285, "maximum": 0.9994339701319311, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5178286112272279, "standard_deviation": 0.0201312702833787, "minimum": 0.48354217130181865, "maximum": 0.538177983419954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.653236951586249, "standard_deviation": 0.32256653686278963, "minimum": 0.2524959739087094, "maximum": 1.214424536651256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03606968123914442, "standard_deviation": 0.004854106828838721, "minimum": 0.028188210824538405, "maximum": 0.0420436902504598, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13798706001824956, "standard_deviation": 0.03765404451891774, "minimum": 0.06275416745285257, "maximum": 0.17962162794300504, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.625898063354762, "standard_deviation": 0.009998044470370405, "minimum": 5.604746365361816, "maximum": 5.636451601169033, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47001907869513826, "standard_deviation": 0.0045653179956566155, "minimum": 0.4620900736469932, "maximum": 0.4756299341821797, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08433115241961872, "standard_deviation": 0.03745299105966984, "minimum": -0.1346252869616873, "maximum": -0.0408904458615644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973813042135029, "standard_deviation": 0.0006455438973385546, "minimum": 0.9961046061638759, "maximum": 0.9980839171480904, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7663903133432121, "standard_deviation": 0.03019544715468792, "minimum": 0.7293745822427187, "maximum": 0.8181473675535634, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35509737339159253, "standard_deviation": 0.25357759900114824, "minimum": 0.0549935851488057, "maximum": 0.845715138704327, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05364238634707864, "standard_deviation": 0.00602394608108686, "minimum": 0.04693602278170269, "maximum": 0.06507469787703854, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992753827756101, "standard_deviation": 0.00019498826995723782, "minimum": 0.9990326516944152, "maximum": 0.9996186726765169, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41497583964722584, "standard_deviation": 0.034519176564827586, "minimum": 0.36891012622939356, "maximum": 0.47127606080478796, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4057512715378899, "standard_deviation": 0.25451305073590974, "minimum": 0.15164554537403466, "maximum": 0.9438043235444177, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02886468299604734, "standard_deviation": 0.003329317646584805, "minimum": 0.024301818053758455, "maximum": 0.03229009904163046, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2271-0022", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.981858491897583, "predicted_mos": 2.981858491897583, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "0021001d859b6748c8840f83214d288d79ab0c3af2c569b93e784e2e68b93dac", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8499963004142046, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7858569074517784, "全帯域_10ms_rms_envelope_nrmse": 0.051970518471083055, "全帯域_10ms_envelope_correlation": 0.9976026340104348, "全帯域_10ms_modulation_mae_db": 0.723151227206214, "全帯域_40ms_log_rms_mae_db": 0.4376184563854207, "全帯域_40ms_rms_envelope_nrmse": 0.026318044159305144, "全帯域_40ms_envelope_correlation": 0.999458920179794, "全帯域_40ms_modulation_mae_db": 0.7669055167316561, "300–1000 Hz_10ms_log_rms_mae_db": 1.0501792617358077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.058063247801251876, "300–1000 Hz_10ms_envelope_correlation": 0.9974275949812623, "300–1000 Hz_10ms_modulation_mae_db": 0.9451124845365254, "300–1000 Hz_40ms_log_rms_mae_db": 0.5168905191991018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.028288581058276325, "300–1000 Hz_40ms_envelope_correlation": 0.9994723456039036, "300–1000 Hz_40ms_modulation_mae_db": 0.9481213569696306, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5626478228227565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12090767846147722, "1000–3000 Hz_10ms_envelope_correlation": 0.9901145117544511, "1000–3000 Hz_10ms_modulation_mae_db": 0.964286464123691, "1000–3000 Hz_40ms_log_rms_mae_db": 0.897085657469048, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0695861420023903, "1000–3000 Hz_40ms_envelope_correlation": 0.9972868681639699, "1000–3000 Hz_40ms_modulation_mae_db": 1.0158151133135032, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7608570468580766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1879878947329024, "3000–6000 Hz_10ms_envelope_correlation": 0.9755928328708828, "3000–6000 Hz_10ms_modulation_mae_db": 1.2163290648334166, "3000–6000 Hz_40ms_log_rms_mae_db": 1.441536187653387, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12275318945609592, "3000–6000 Hz_40ms_envelope_correlation": 0.9896602578984617, "3000–6000 Hz_40ms_modulation_mae_db": 1.3209143651401731, "mr_stft_db": 5.643736828452742, "band_mae_db": 0.1789720068559656, "rms_gain_error_db": -0.10685199226947299, "output_peak": 0.477311873167415}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7812478926443104, "全帯域_10ms_rms_envelope_nrmse": 0.058768797975542936, "全帯域_10ms_envelope_correlation": 0.9968044500744009, "全帯域_10ms_modulation_mae_db": 0.494280800668425, "全帯域_40ms_log_rms_mae_db": 0.4368332436025361, "全帯域_40ms_rms_envelope_nrmse": 0.026551089540223345, "全帯域_40ms_envelope_correlation": 0.9993336663076227, "全帯域_40ms_modulation_mae_db": 0.31203326554739935, "300–1000 Hz_10ms_log_rms_mae_db": 1.0480877563433415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07557068627374282, "300–1000 Hz_10ms_envelope_correlation": 0.9954769863379485, "300–1000 Hz_10ms_modulation_mae_db": 0.8842823994137875, "300–1000 Hz_40ms_log_rms_mae_db": 0.49516995724273616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033765083438849916, "300–1000 Hz_40ms_envelope_correlation": 0.9991162690893214, "300–1000 Hz_40ms_modulation_mae_db": 0.7651778688004452, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4508705052105408, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10215796720096068, "1000–3000 Hz_10ms_envelope_correlation": 0.9928174681780941, "1000–3000 Hz_10ms_modulation_mae_db": 0.43126084896090333, "1000–3000 Hz_40ms_log_rms_mae_db": 0.835139934674171, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05601011945746432, "1000–3000 Hz_40ms_envelope_correlation": 0.9981163326456927, "1000–3000 Hz_40ms_modulation_mae_db": 0.5208966280195851, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7360929193234784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1835715653588228, "3000–6000 Hz_10ms_envelope_correlation": 0.978300335494924, "3000–6000 Hz_10ms_modulation_mae_db": 1.1790624207439775, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3738865864164937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12284550302550813, "3000–6000 Hz_40ms_envelope_correlation": 0.9902719258062793, "3000–6000 Hz_40ms_modulation_mae_db": 0.944923001432012, "mr_stft_db": 5.646486063456042, "band_mae_db": 0.21303299324089, "rms_gain_error_db": -0.04708415978218486, "output_peak": 0.4759594360750711}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.729202591968515, "全帯域_10ms_rms_envelope_nrmse": 0.04855112763700382, "全帯域_10ms_envelope_correlation": 0.9978349416964757, "全帯域_10ms_modulation_mae_db": 0.534288361145448, "全帯域_40ms_log_rms_mae_db": 0.36591184556004647, "全帯域_40ms_rms_envelope_nrmse": 0.019974750776647357, "全帯域_40ms_envelope_correlation": 0.9996424301816743, "全帯域_40ms_modulation_mae_db": 0.5430159686601067, "300–1000 Hz_10ms_log_rms_mae_db": 0.996924141259376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06347477568626772, "300–1000 Hz_10ms_envelope_correlation": 0.9967935344217376, "300–1000 Hz_10ms_modulation_mae_db": 1.055054927178499, "300–1000 Hz_40ms_log_rms_mae_db": 0.5155125872727295, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.029012734335103436, "300–1000 Hz_40ms_envelope_correlation": 0.9993228127149347, "300–1000 Hz_40ms_modulation_mae_db": 0.9469930270064465, "1000–3000 Hz_10ms_log_rms_mae_db": 1.36668505105608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09461278736576646, "1000–3000 Hz_10ms_envelope_correlation": 0.9939075546876437, "1000–3000 Hz_10ms_modulation_mae_db": 0.9935103977458036, "1000–3000 Hz_40ms_log_rms_mae_db": 0.770652459076434, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05828049595981224, "1000–3000 Hz_40ms_envelope_correlation": 0.9979096851050007, "1000–3000 Hz_40ms_modulation_mae_db": 1.1118509378535488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.77854258356558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18946965312838987, "3000–6000 Hz_10ms_envelope_correlation": 0.975441565571526, "3000–6000 Hz_10ms_modulation_mae_db": 1.3832528746105759, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3767676059035872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12589195589528762, "3000–6000 Hz_40ms_envelope_correlation": 0.989619066650176, "3000–6000 Hz_40ms_modulation_mae_db": 1.505564264845618, "mr_stft_db": 5.613996496893331, "band_mae_db": 0.13388671063512583, "rms_gain_error_db": -0.05091491841444819, "output_peak": 0.4648975299632566}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7494317271611168, "全帯域_10ms_rms_envelope_nrmse": 0.05249141241880815, "全帯域_10ms_envelope_correlation": 0.9975066149330424, "全帯域_10ms_modulation_mae_db": 0.7084672883775026, "全帯域_40ms_log_rms_mae_db": 0.39098993696767886, "全帯域_40ms_rms_envelope_nrmse": 0.030589231056509746, "全帯域_40ms_envelope_correlation": 0.9991616842306829, "全帯域_40ms_modulation_mae_db": 0.6959092116812945, "300–1000 Hz_10ms_log_rms_mae_db": 1.0706874763384595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06798906652460997, "300–1000 Hz_10ms_envelope_correlation": 0.9964652267187375, "300–1000 Hz_10ms_modulation_mae_db": 1.2131353577584219, "300–1000 Hz_40ms_log_rms_mae_db": 0.5187990586799643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041324031313467686, "300–1000 Hz_40ms_envelope_correlation": 0.9987472798200017, "300–1000 Hz_40ms_modulation_mae_db": 1.1550452149452208, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4565705689681598, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09521969999243272, "1000–3000 Hz_10ms_envelope_correlation": 0.9936795152650404, "1000–3000 Hz_10ms_modulation_mae_db": 0.7944689675486214, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7433007529874598, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04781369489216808, "1000–3000 Hz_40ms_envelope_correlation": 0.9985673352167206, "1000–3000 Hz_40ms_modulation_mae_db": 0.747080663896341, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6500903722791997, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16489144334874092, "3000–6000 Hz_10ms_envelope_correlation": 0.9823342476519402, "3000–6000 Hz_10ms_modulation_mae_db": 1.1878126534647906, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2858856565288597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12349928467094472, "3000–6000 Hz_40ms_envelope_correlation": 0.9908365747883965, "3000–6000 Hz_40ms_modulation_mae_db": 1.2951009095899568, "mr_stft_db": 5.605913312467847, "band_mae_db": 0.1120459556413414, "rms_gain_error_db": -0.08300181281503423, "output_peak": 0.4625911649726622}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7787882044557777, "全帯域_10ms_rms_envelope_nrmse": 0.0530154839323769, "全帯域_10ms_envelope_correlation": 0.9974070311751687, "全帯域_10ms_modulation_mae_db": 0.540835123129691, "全帯域_40ms_log_rms_mae_db": 0.41356214158395044, "全帯域_40ms_rms_envelope_nrmse": 0.024020774084452686, "全帯域_40ms_envelope_correlation": 0.9994592369251286, "全帯域_40ms_modulation_mae_db": 0.5880956873324048, "300–1000 Hz_10ms_log_rms_mae_db": 1.0237846440336584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06534188593496873, "300–1000 Hz_10ms_envelope_correlation": 0.9966235972574156, "300–1000 Hz_10ms_modulation_mae_db": 0.856328752716609, "300–1000 Hz_40ms_log_rms_mae_db": 0.4955567448458876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.029478183316721907, "300–1000 Hz_40ms_envelope_correlation": 0.9993145485492512, "300–1000 Hz_40ms_modulation_mae_db": 0.6574387425203086, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4280582651702782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09996214466552729, "1000–3000 Hz_10ms_envelope_correlation": 0.9929306452737455, "1000–3000 Hz_10ms_modulation_mae_db": 0.8066088861513045, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8167500076561073, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056847447156230455, "1000–3000 Hz_40ms_envelope_correlation": 0.9977800685724248, "1000–3000 Hz_40ms_modulation_mae_db": 1.1284648789086384, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7389705857103663, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17750256382889212, "3000–6000 Hz_10ms_envelope_correlation": 0.9784569331887834, "3000–6000 Hz_10ms_modulation_mae_db": 1.9166741821122661, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3870388999053218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1263989502823454, "3000–6000 Hz_40ms_envelope_correlation": 0.9892220961645571, "3000–6000 Hz_40ms_modulation_mae_db": 1.6145257579921992, "mr_stft_db": 5.628996824922638, "band_mae_db": 0.0683205987482794, "rms_gain_error_db": -0.046024706112025277, "output_peak": 0.4759925721670739}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7343218052052084, "全帯域_10ms_rms_envelope_nrmse": 0.05319389129216231, "全帯域_10ms_envelope_correlation": 0.9973848236663233, "全帯域_10ms_modulation_mae_db": 0.4652305665552762, "全帯域_40ms_log_rms_mae_db": 0.4221648333415999, "全帯域_40ms_rms_envelope_nrmse": 0.026142399207341977, "全帯域_40ms_envelope_correlation": 0.9993515436980897, "全帯域_40ms_modulation_mae_db": 0.4463317041991364, "300–1000 Hz_10ms_log_rms_mae_db": 1.0357165413504914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06894813898726522, "300–1000 Hz_10ms_envelope_correlation": 0.996274781813937, "300–1000 Hz_10ms_modulation_mae_db": 0.8648063062983019, "300–1000 Hz_40ms_log_rms_mae_db": 0.5281452034006453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03558597496882616, "300–1000 Hz_40ms_envelope_correlation": 0.999031897462586, "300–1000 Hz_40ms_modulation_mae_db": 0.7501107645212902, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4310254992083578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11260060815411659, "1000–3000 Hz_10ms_envelope_correlation": 0.9909881692383932, "1000–3000 Hz_10ms_modulation_mae_db": 0.2156411959986153, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7793201406895245, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056171491690719194, "1000–3000 Hz_40ms_envelope_correlation": 0.9978751878838237, "1000–3000 Hz_40ms_modulation_mae_db": 0.6529164495015393, "3000–6000 Hz_10ms_log_rms_mae_db": 1.661761508296005, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1816823886823503, "3000–6000 Hz_10ms_envelope_correlation": 0.9775215000467682, "3000–6000 Hz_10ms_modulation_mae_db": 1.7402543369905035, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3289253827824412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12739129933162147, "3000–6000 Hz_40ms_envelope_correlation": 0.9893126787802745, "3000–6000 Hz_40ms_modulation_mae_db": 2.266430731769185, "mr_stft_db": 5.607286360086122, "band_mae_db": 0.15312433118388316, "rms_gain_error_db": -0.04493435474213902, "output_peak": 0.46275939096493107}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8156570343995516, "全帯域_10ms_rms_envelope_nrmse": 0.050367233773769185, "全帯域_10ms_envelope_correlation": 0.9976836531689794, "全帯域_10ms_modulation_mae_db": 0.19317157780455943, "全帯域_40ms_log_rms_mae_db": 0.4644877190148534, "全帯域_40ms_rms_envelope_nrmse": 0.02633099163976274, "全帯域_40ms_envelope_correlation": 0.9993745293437795, "全帯域_40ms_modulation_mae_db": 0.2419811628800863, "300–1000 Hz_10ms_log_rms_mae_db": 1.0132928501714291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.059904384376292065, "300–1000 Hz_10ms_envelope_correlation": 0.9971865522923489, "300–1000 Hz_10ms_modulation_mae_db": 0.14367433319021508, "300–1000 Hz_40ms_log_rms_mae_db": 0.5158563769915699, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031153250546813904, "300–1000 Hz_40ms_envelope_correlation": 0.9992617051618926, "300–1000 Hz_40ms_modulation_mae_db": 0.1270322231884505, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4533442777889376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12148458477411501, "1000–3000 Hz_10ms_envelope_correlation": 0.9896484252424036, "1000–3000 Hz_10ms_modulation_mae_db": 0.7160128201292744, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8716884541736839, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06905969067315063, "1000–3000 Hz_40ms_envelope_correlation": 0.9968990252624612, "1000–3000 Hz_40ms_modulation_mae_db": 0.5534348842625453, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6520922283711854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17046668306224505, "3000–6000 Hz_10ms_envelope_correlation": 0.9799092691522563, "3000–6000 Hz_10ms_modulation_mae_db": 1.0654332895608032, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3118279091652987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1157717322904381, "3000–6000 Hz_40ms_envelope_correlation": 0.9904901001694837, "3000–6000 Hz_40ms_modulation_mae_db": 1.3668084472228745, "mr_stft_db": 5.6211761388585115, "band_mae_db": 0.12052930717751209, "rms_gain_error_db": -0.0668563433383188, "output_peak": 0.46703251222699904}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7510070584261008, "全帯域_10ms_rms_envelope_nrmse": 0.05130768565545476, "全帯域_10ms_envelope_correlation": 0.9975571073327243, "全帯域_10ms_modulation_mae_db": 0.17582004294270623, "全帯域_40ms_log_rms_mae_db": 0.39655486545394547, "全帯域_40ms_rms_envelope_nrmse": 0.026389968991915872, "全帯域_40ms_envelope_correlation": 0.9993179078701979, "全帯域_40ms_modulation_mae_db": 0.14567642942812667, "300–1000 Hz_10ms_log_rms_mae_db": 1.0506406930096768, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.069082388566301, "300–1000 Hz_10ms_envelope_correlation": 0.9961934175354628, "300–1000 Hz_10ms_modulation_mae_db": 0.47852457851577307, "300–1000 Hz_40ms_log_rms_mae_db": 0.4538572404947015, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03531975674230638, "300–1000 Hz_40ms_envelope_correlation": 0.9989529662491313, "300–1000 Hz_40ms_modulation_mae_db": 0.4526839262852649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.441444607117877, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09841905831956013, "1000–3000 Hz_10ms_envelope_correlation": 0.9931323971268246, "1000–3000 Hz_10ms_modulation_mae_db": 0.7092114244995961, "1000–3000 Hz_40ms_log_rms_mae_db": 0.753077464966912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05497390730698339, "1000–3000 Hz_40ms_envelope_correlation": 0.9979041689536006, "1000–3000 Hz_40ms_modulation_mae_db": 1.000238062992538, "3000–6000 Hz_10ms_log_rms_mae_db": 1.641269471477979, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16807953304780324, "3000–6000 Hz_10ms_envelope_correlation": 0.9805043454172462, "3000–6000 Hz_10ms_modulation_mae_db": 1.1183330789967307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.317153213963486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11406036004935635, "3000–6000 Hz_40ms_envelope_correlation": 0.9909649176065631, "3000–6000 Hz_40ms_modulation_mae_db": 1.2372868808408208, "mr_stft_db": 5.600291427628715, "band_mae_db": 0.05800439294965281, "rms_gain_error_db": -0.009915474518774543, "output_peak": 0.4641699385004667}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37200, "output_samples": 37200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.562530887830007, "spectral_convergence": 0.1486181607447843, "spectral_convergence_status": "ok", "amplitude_floor": 1.3482074212160965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.464400991793894, "spectral_convergence": 0.14241629543654444, "spectral_convergence_status": "ok", "amplitude_floor": 1.080252897815134e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.904278605734325, "spectral_convergence": 0.1445764619188947, "spectral_convergence_status": "ok", "amplitude_floor": 8.596818112459857e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004510568589627121, "output_power": 0.000458801472236558, "delta_db": 0.07393511554877853, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019377343288488795, "output_power": 0.0018808108302842091, "delta_db": -0.12949116355657095, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013786956670426643, "output_power": 0.001355496799049746, "delta_db": -0.07369914344152169, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031975863120548915, "output_power": 0.0002932568895337236, "delta_db": -0.37574052305398964, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.4830953182024825e-05, "output_power": 1.3485212509498506e-05, "delta_db": -0.4131126917215936, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.03312086795025e-05, "output_power": 2.0368007131113946e-05, "delta_db": 0.00785340381333932, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1789720068559656, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10685199226947299, "pitch": null, "output_rms_dbfs": -23.95534264066835, "output_peak": 0.477311873167415, "reference_rms_dbfs": -23.848490648398872, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.643736828452742}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00019668187909197959, "log_rms_mae_db": 0.7858569074517784, "rms_envelope_nrmse": 0.051970518471083055, "envelope_correlation": 0.9976026340104348, "rms_gain_error_db": -0.10685198517385674, "reference_mean_power": 0.004131292024738224, "output_mean_power": 0.00403088767376691}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.0026327800860698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011612269501705267, "output_power": 0.001123689677781948, "reference_active": true, "delta_db": -0.14270715086997515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017095886275955477, "output_power": 0.00015810745523487513, "reference_active": true, "delta_db": -0.3393927149625208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.217923048678998e-05, "output_power": 2.859981808222482e-05, "reference_active": true, "delta_db": -1.687353815786146}], "active_band_count": 3, "active_band_mae_db": 0.723151227206214}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00018115923356648065, "log_rms_mae_db": 0.4376184563854207, "rms_envelope_nrmse": 0.026318044159305144, "envelope_correlation": 0.999458920179794, "rms_gain_error_db": -0.10685190398848718, "reference_mean_power": 0.004185412442455258, "output_mean_power": 0.004083692858863279}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.7728315536386115e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066252574844359, "output_power": 0.0010275155776388668, "reference_active": true, "delta_db": -0.1607167821979859}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.440983264620606e-05, "output_power": 7.754612091645898e-05, "reference_active": true, "delta_db": -0.36832961209509024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.940091323355611e-06, "output_power": 1.955211778493965e-06, "reference_active": true, "delta_db": -1.771670155901892}], "active_band_count": 3, "active_band_mae_db": 0.7669055167316561}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.000176021246464811, "log_rms_mae_db": 1.0501792617358077, "rms_envelope_nrmse": 0.058063247801251876, "envelope_correlation": 0.9974275949812623, "rms_gain_error_db": -0.11532527258880029, "reference_mean_power": 0.0026395436088586166, "output_mean_power": 0.0025703739686907343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.515899729654865e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009091516678957249, "output_power": 0.0008656008623799626, "reference_active": true, "delta_db": -0.21318659365690582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001833276731517994, "output_power": 0.00016271491123239832, "reference_active": true, "delta_db": -0.5180067267434287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7517429693718724e-05, "output_power": 2.3110996087343095e-05, "reference_active": true, "delta_db": -2.1041441332092417}], "active_band_count": 3, "active_band_mae_db": 0.9451124845365254}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00016136314602147107, "log_rms_mae_db": 0.5168905191991018, "rms_envelope_nrmse": 0.028288581058276325, "envelope_correlation": 0.9994723456039036, "rms_gain_error_db": -0.11532530441002038, "reference_mean_power": 0.002674122622482191, "output_mean_power": 0.0026040468149433633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.348456388983372e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008380618007305507, "output_power": 0.0007947852557080523, "reference_active": true, "delta_db": -0.23026243960816373}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.026333057603631e-05, "output_power": 8.25004026348097e-05, "reference_active": true, "delta_db": -0.3905528618334684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498114091574834e-06, "output_power": 1.4971226892361706e-06, "reference_active": true, "delta_db": -2.2235487694672598}], "active_band_count": 3, "active_band_mae_db": 0.9481213569696306}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 6.667691377903701e-05, "log_rms_mae_db": 1.5626478228227565, "rms_envelope_nrmse": 0.12090767846147722, "envelope_correlation": 0.9901145117544511, "rms_gain_error_db": -0.3271944924807976, "reference_mean_power": 0.0003102146198149334, "output_mean_power": 0.0002877019544308296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.247013787334141e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018435630971779633, "output_power": 0.00016797362551051236, "reference_active": true, "delta_db": -0.40416910170532566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0328682781544803e-05, "output_power": 1.6782558195922122e-05, "reference_active": true, "delta_db": -0.8325107714256756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.580926194982421e-06, "output_power": 3.128494118335324e-06, "reference_active": true, "delta_db": -1.6561795192400717}], "active_band_count": 3, "active_band_mae_db": 0.964286464123691}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 6.131102329922278e-05, "log_rms_mae_db": 0.897085657469048, "rms_envelope_nrmse": 0.0695861420023903, "envelope_correlation": 0.9972868681639699, "rms_gain_error_db": -0.3271944945632983, "reference_mean_power": 0.0003142782197045969, "output_mean_power": 0.00029147065362244626}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 3.052275331975406e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017569545994603763, "output_power": 0.00016103029927227867, "reference_active": true, "delta_db": -0.378529392039474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.589536328721284e-06, "output_power": 6.946307125631461e-06, "reference_active": true, "delta_db": -0.9221573944188146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8479520444652144e-07, "output_power": 1.9048326071537813e-07, "reference_active": true, "delta_db": -1.7467585534822208}], "active_band_count": 3, "active_band_mae_db": 1.0158151133135032}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 2.0872644535346135e-05, "log_rms_mae_db": 1.7608570468580766, "rms_envelope_nrmse": 0.1879878947329024, "envelope_correlation": 0.9755928328708828, "rms_gain_error_db": -0.17589690346699155, "reference_mean_power": 1.984318614204444e-05, "output_mean_power": 1.9055559990692597e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.6220002221112295e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.352365504647456e-06, "output_power": 8.022418157120064e-06, "reference_active": true, "delta_db": -0.6661617615646597}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.791249366368675e-07, "output_power": 9.731773242162534e-07, "reference_active": true, "delta_db": -0.026461304660992998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495952213974179e-07, "output_power": 4.930292428416338e-07, "reference_active": true, "delta_db": 2.956364128274597}], "active_band_count": 3, "active_band_mae_db": 1.2163290648334166}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 1.9522833601394496e-05, "log_rms_mae_db": 1.441536187653387, "rms_envelope_nrmse": 0.12275318945609592, "envelope_correlation": 0.9896602578984617, "rms_gain_error_db": -0.17590235487015699, "reference_mean_power": 2.010291979320635e-05, "output_mean_power": 1.930495992533408e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.5131001811961146e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.175181026763155e-06, "output_power": 7.156597951288984e-06, "reference_active": true, "delta_db": -0.5779075844054747}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.283226898133603e-07, "output_power": 4.593044321058439e-07, "reference_active": true, "delta_db": 0.3032955504097897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.754857048422832e-08, "output_power": 3.567760890171428e-08, "reference_active": true, "delta_db": 3.081539960605255}], "active_band_count": 3, "active_band_mae_db": 1.3209143651401731}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9454691888130307, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3c6cd6f482dd5c8a88faac233d4fce3a6f8de9965450806e65aac8e49bde31a8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9921523358458245, "standard_deviation": 0.0016561287082424668, "minimum": 0.9896484252424036, "maximum": 0.9939075546876437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4488308246678736, "standard_deviation": 0.05423461920396149, "minimum": 1.36668505105608, "maximum": 1.5626478228227565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7038751256447262, "standard_deviation": 0.26276845848296565, "minimum": 0.2156411959986153, "maximum": 0.9935103977458036, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10567056611674452, "standard_deviation": 0.011079467025189242, "minimum": 0.09461278736576646, "maximum": 0.12148458477411501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9977923339754617, "standard_deviation": 0.0005060197293945529, "minimum": 0.9968990252624612, "maximum": 0.9985673352167206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8083768589616676, "standard_deviation": 0.056346284742483065, "minimum": 0.7433007529874598, "maximum": 0.897085657469048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8413372023435299, "standard_deviation": 0.2510744312668361, "minimum": 0.5208966280195851, "maximum": 1.1284648789086384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.058592873642364825, "standard_deviation": 0.007322876318450232, "minimum": 0.04781369489216808, "maximum": 0.0695861420023903, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9785076286742909, "standard_deviation": 0.0023757481309311375, "minimum": 0.975441565571526, "maximum": 0.9823342476519402, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7024595894852337, "standard_deviation": 0.05649666117542488, "minimum": 1.641269471477979, "maximum": 1.77854258356558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.350893987664133, "standard_deviation": 0.31226305792998393, "minimum": 1.0654332895608032, "maximum": 1.9166741821122661, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17795646564876832, "standard_deviation": 0.009282444089913959, "minimum": 0.16489144334874092, "maximum": 0.18946965312838987, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.990047202233024, "standard_deviation": 0.0006829892831947993, "minimum": 0.9892220961645571, "maximum": 0.9909649176065631, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3528776802898594, "standard_deviation": 0.05079705776515145, "minimum": 1.2858856565288597, "maximum": 1.441536187653387, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4439442948541048, "standard_deviation": 0.3862547621832667, "minimum": 0.944923001432012, "maximum": 2.266430731769185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12232653437519972, "standard_deviation": 0.004900006131633651, "minimum": 0.11406036004935635, "maximum": 0.12739129933162147, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9965552114198563, "standard_deviation": 0.0006096042045982136, "minimum": 0.9954769863379485, "maximum": 0.9974275949812623, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.03616417053028, "standard_deviation": 0.023773794899498674, "minimum": 0.996924141259376, "maximum": 1.0706874763384595, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8051148924510165, "standard_deviation": 0.33915194908553936, "minimum": 0.14367433319021508, "maximum": 1.2131353577584219, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06604682176883743, "standard_deviation": 0.005614877726450316, "minimum": 0.058063247801251876, "maximum": 0.07557068627374282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9991524780813779, "standard_deviation": 0.00023584737497942711, "minimum": 0.9987472798200017, "maximum": 0.9994723456039036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.504973461015917, "standard_deviation": 0.02359556660559985, "minimum": 0.4538572404947015, "maximum": 0.5281452034006453, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7253253905296322, "standard_deviation": 0.3212771222786477, "minimum": 0.1270322231884505, "maximum": 1.1550452149452208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.032990949465045716, "standard_deviation": 0.004405388985286015, "minimum": 0.028288581058276325, "maximum": 0.041324031313467686, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12973953705408128, "standard_deviation": 0.05242926065720134, "minimum": 0.05800439294965281, "maximum": 0.21303299324089, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.620985431595743, "standard_deviation": 0.01741939567678407, "minimum": 5.600291427628715, "maximum": 5.646486063456042, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46883930225473447, "standard_deviation": 0.0064393831957065915, "minimum": 0.4625911649726622, "maximum": 0.477311873167415, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05694797024904973, "standard_deviation": 0.02897482347162499, "minimum": -0.10685199226947299, "maximum": -0.009915474518774543, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9974726570071937, "standard_deviation": 0.00030693660271136926, "minimum": 0.9968044500744009, "maximum": 0.9978349416964757, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7656891527140448, "standard_deviation": 0.029549557548330888, "minimum": 0.729202591968515, "maximum": 0.8156570343995516, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.47940562347872784, "standard_deviation": 0.20465413357712994, "minimum": 0.17582004294270623, "maximum": 0.723151227206214, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05245826889452514, "standard_deviation": 0.0029731039007826867, "minimum": 0.04855112763700382, "maximum": 0.058768797975542936, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9993874898421211, "standard_deviation": 0.00013909473206583903, "minimum": 0.9991616842306829, "maximum": 0.9996424301816743, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41601538023875395, "standard_deviation": 0.031187971004682135, "minimum": 0.36591184556004647, "maximum": 0.4644877190148534, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46749361830752634, "standard_deviation": 0.22074794002283554, "minimum": 0.14567642942812667, "maximum": 0.7669055167316561, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02578965618201986, "standard_deviation": 0.0029668359655972993, "minimum": 0.019974750776647357, "maximum": 0.030589231056509746, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4446-2273-0010", "language": "en", "style": "normal", "speaker": "librispeech_4446", "source_utterance_id": "4446-2273-0010", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4446-2273-0010.wav", "source_file_sha256": "9b258d8ee36847c23d5e8891adbc03b35105e5fe35a38867fd79bc3ec3766bdd", "decoded_pcm_sha256": "263c284aa66f0a837f0cdfcf012d15e0d4edfeb7a8a0717217e17395942a83fd", "original_source_sha256": "ee6acd6b5c33e2d7e114405b48aec2700f720d68dd8d59676172e81ca63317dc", "corpus_entry_id": "librispeech_4446-2273-0010"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.523749828338623, "predicted_mos": 3.523749828338623, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ff17b39b1cb5e6e39d69b8e58d4e54f6176da9af1035a4bd5e02a7efc5cc19be", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.158949307999137, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4446-2273-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ff17b39b1cb5e6e39d69b8e58d4e54f6176da9af1035a4bd5e02a7efc5cc19be", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "66fab68f8236f2b9617cfd6928803d223cdcd7756c14f365c225f0ed88f58739"}, "rows": [{"case": "librispeech_4446-2273-0010", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.2935831546783447, "predicted_mos": 2.2935831546783447, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cd6b5f654297a8803c5a629aa3610cdea05d1f06ef05b14e4b9bda0749a32488", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.766411002998211, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4446-2273-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "cd6b5f654297a8803c5a629aa3610cdea05d1f06ef05b14e4b9bda0749a32488", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1148710286342531, "全帯域_10ms_rms_envelope_nrmse": 0.09034002966321346, "全帯域_10ms_envelope_correlation": 0.991821860206185, "全帯域_10ms_modulation_mae_db": 0.3880680544882938, "全帯域_40ms_log_rms_mae_db": 0.6089687293841078, "全帯域_40ms_rms_envelope_nrmse": 0.049409351718864944, "全帯域_40ms_envelope_correlation": 0.9972292477808535, "全帯域_40ms_modulation_mae_db": 0.42315474488889876, "300–1000 Hz_10ms_log_rms_mae_db": 1.5516001148829266, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09305226481321047, "300–1000 Hz_10ms_envelope_correlation": 0.9938332218150701, "300–1000 Hz_10ms_modulation_mae_db": 0.2545046959544026, "300–1000 Hz_40ms_log_rms_mae_db": 0.8488434351941943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056971651139885074, "300–1000 Hz_40ms_envelope_correlation": 0.9975052686273779, "300–1000 Hz_40ms_modulation_mae_db": 0.21799131366257232, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6077535946929284, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13807501765262173, "1000–3000 Hz_10ms_envelope_correlation": 0.9858620720663217, "1000–3000 Hz_10ms_modulation_mae_db": 0.5513705709937794, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9657003907530571, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07887572274448121, "1000–3000 Hz_40ms_envelope_correlation": 0.99520011264673, "1000–3000 Hz_40ms_modulation_mae_db": 0.7187380049987225, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6859987488654724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1810457047455501, "3000–6000 Hz_10ms_envelope_correlation": 0.9758690419009166, "3000–6000 Hz_10ms_modulation_mae_db": 0.4337331038336007, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9832453145407271, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10078277741846596, "3000–6000 Hz_40ms_envelope_correlation": 0.992188734128542, "3000–6000 Hz_40ms_modulation_mae_db": 0.9183564075954966, "mr_stft_db": 5.823145255653948, "band_mae_db": 0.14618462248597697, "rms_gain_error_db": -0.03487393914914078, "output_peak": 0.45819522637545207}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1080248829134673, "全帯域_10ms_rms_envelope_nrmse": 0.09145359098912074, "全帯域_10ms_envelope_correlation": 0.9916313429486027, "全帯域_10ms_modulation_mae_db": 0.3769188774198351, "全帯域_40ms_log_rms_mae_db": 0.5873235942018096, "全帯域_40ms_rms_envelope_nrmse": 0.04504616453093, "全帯域_40ms_envelope_correlation": 0.9977310277799322, "全帯域_40ms_modulation_mae_db": 0.4496783194620468, "300–1000 Hz_10ms_log_rms_mae_db": 1.6391287615809418, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09526731519567329, "300–1000 Hz_10ms_envelope_correlation": 0.9935418928512897, "300–1000 Hz_10ms_modulation_mae_db": 0.13997235312589948, "300–1000 Hz_40ms_log_rms_mae_db": 0.8350653742365742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049683185973211585, "300–1000 Hz_40ms_envelope_correlation": 0.9981380670052398, "300–1000 Hz_40ms_modulation_mae_db": 0.34737720744583367, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6142661126731843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15343384550329067, "1000–3000 Hz_10ms_envelope_correlation": 0.982779005703107, "1000–3000 Hz_10ms_modulation_mae_db": 0.508090185267085, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9781480854701616, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09040315108540285, "1000–3000 Hz_40ms_envelope_correlation": 0.994148005895638, "1000–3000 Hz_40ms_modulation_mae_db": 0.5648035537751586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6109234616253167, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15394665098857385, "3000–6000 Hz_10ms_envelope_correlation": 0.9825651390259432, "3000–6000 Hz_10ms_modulation_mae_db": 0.8978587531819242, "3000–6000 Hz_40ms_log_rms_mae_db": 0.947129562246195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08489431352655194, "3000–6000 Hz_40ms_envelope_correlation": 0.9943981349534577, "3000–6000 Hz_40ms_modulation_mae_db": 1.0710440524335934, "mr_stft_db": 5.787926021354333, "band_mae_db": 0.11569527210895018, "rms_gain_error_db": -0.05877697795903935, "output_peak": 0.44192149765450683}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0864263564948122, "全帯域_10ms_rms_envelope_nrmse": 0.08883010445867537, "全帯域_10ms_envelope_correlation": 0.9921026073949578, "全帯域_10ms_modulation_mae_db": 0.21655787517474634, "全帯域_40ms_log_rms_mae_db": 0.5172554658593013, "全帯域_40ms_rms_envelope_nrmse": 0.04555283840703297, "全帯域_40ms_envelope_correlation": 0.9976400788014221, "全帯域_40ms_modulation_mae_db": 0.31740876334186224, "300–1000 Hz_10ms_log_rms_mae_db": 1.5542316946255583, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08838913783826932, "300–1000 Hz_10ms_envelope_correlation": 0.9944412937762378, "300–1000 Hz_10ms_modulation_mae_db": 0.18186364771962626, "300–1000 Hz_40ms_log_rms_mae_db": 0.8587108036489361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05158609269005212, "300–1000 Hz_40ms_envelope_correlation": 0.9979891801460307, "300–1000 Hz_40ms_modulation_mae_db": 0.28221465696909914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6223898017186178, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.140627309879979, "1000–3000 Hz_10ms_envelope_correlation": 0.9853752818581996, "1000–3000 Hz_10ms_modulation_mae_db": 0.42529231831166897, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0212994821345778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07641754442804877, "1000–3000 Hz_40ms_envelope_correlation": 0.9956103479719435, "1000–3000 Hz_40ms_modulation_mae_db": 0.3429282364876798, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6302824269587046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15871674252578935, "3000–6000 Hz_10ms_envelope_correlation": 0.981749681079075, "3000–6000 Hz_10ms_modulation_mae_db": 0.6077386641143824, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9330897447980725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0856756201108581, "3000–6000 Hz_40ms_envelope_correlation": 0.9943333115719791, "3000–6000 Hz_40ms_modulation_mae_db": 1.0147514203535397, "mr_stft_db": 5.812012060799325, "band_mae_db": 0.16338094665892233, "rms_gain_error_db": -0.008949297261973342, "output_peak": 0.39985621032096236}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1027283053502879, "全帯域_10ms_rms_envelope_nrmse": 0.08781525691352693, "全帯域_10ms_envelope_correlation": 0.9922933884810441, "全帯域_10ms_modulation_mae_db": 0.27222832914896333, "全帯域_40ms_log_rms_mae_db": 0.566176655868709, "全帯域_40ms_rms_envelope_nrmse": 0.04608808954851727, "全帯域_40ms_envelope_correlation": 0.9975962392092366, "全帯域_40ms_modulation_mae_db": 0.350424283633505, "300–1000 Hz_10ms_log_rms_mae_db": 1.6257931008653026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09298954900346738, "300–1000 Hz_10ms_envelope_correlation": 0.9938315481333344, "300–1000 Hz_10ms_modulation_mae_db": 0.2967696853837006, "300–1000 Hz_40ms_log_rms_mae_db": 0.8893405949766054, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05418752862886352, "300–1000 Hz_40ms_envelope_correlation": 0.9977469479071733, "300–1000 Hz_40ms_modulation_mae_db": 0.5172277024693813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6456310482796825, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14640035569900417, "1000–3000 Hz_10ms_envelope_correlation": 0.9840700796219279, "1000–3000 Hz_10ms_modulation_mae_db": 0.3541198143868789, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0175624064219357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08764946436123984, "1000–3000 Hz_40ms_envelope_correlation": 0.9937257082979185, "1000–3000 Hz_40ms_modulation_mae_db": 0.1876438836891757, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6244237026696235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15058803729729892, "3000–6000 Hz_10ms_envelope_correlation": 0.983389860164744, "3000–6000 Hz_10ms_modulation_mae_db": 0.14780438481631705, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9198304473261221, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0718648276205171, "3000–6000 Hz_40ms_envelope_correlation": 0.995976018956821, "3000–6000 Hz_40ms_modulation_mae_db": 0.6180765658929616, "mr_stft_db": 5.799082223596386, "band_mae_db": 0.12824883132469525, "rms_gain_error_db": -0.005475185526420713, "output_peak": 0.4320577202743447}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0589077355925491, "全帯域_10ms_rms_envelope_nrmse": 0.08895629882358175, "全帯域_10ms_envelope_correlation": 0.9920990563011769, "全帯域_10ms_modulation_mae_db": 0.2169987735306476, "全帯域_40ms_log_rms_mae_db": 0.5477120610640828, "全帯域_40ms_rms_envelope_nrmse": 0.050363479050818845, "全帯域_40ms_envelope_correlation": 0.9971813918217566, "全帯域_40ms_modulation_mae_db": 0.2930901873360243, "300–1000 Hz_10ms_log_rms_mae_db": 1.6034395924996125, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09453326806511883, "300–1000 Hz_10ms_envelope_correlation": 0.9936320665573729, "300–1000 Hz_10ms_modulation_mae_db": 0.4897424956190097, "300–1000 Hz_40ms_log_rms_mae_db": 0.8987326790414882, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05627080765103989, "300–1000 Hz_40ms_envelope_correlation": 0.9975901772626474, "300–1000 Hz_40ms_modulation_mae_db": 0.15761672195018483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6472713214596304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1531964523188694, "1000–3000 Hz_10ms_envelope_correlation": 0.9832710936872102, "1000–3000 Hz_10ms_modulation_mae_db": 0.371031764166132, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0159308093190813, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09304867579205262, "1000–3000 Hz_40ms_envelope_correlation": 0.9943651031805362, "1000–3000 Hz_40ms_modulation_mae_db": 0.5188918529389418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6217333423047893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16114602883395943, "3000–6000 Hz_10ms_envelope_correlation": 0.9808848887985859, "3000–6000 Hz_10ms_modulation_mae_db": 0.3440326030592307, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9681165873613424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09892951748570128, "3000–6000 Hz_40ms_envelope_correlation": 0.9923406229560275, "3000–6000 Hz_40ms_modulation_mae_db": 0.4497918460226236, "mr_stft_db": 5.7882156503752755, "band_mae_db": 0.17128611995573295, "rms_gain_error_db": -0.07897978061582367, "output_peak": 0.4123540068490089}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0862660871195815, "全帯域_10ms_rms_envelope_nrmse": 0.09300747105719659, "全帯域_10ms_envelope_correlation": 0.9913656098989627, "全帯域_10ms_modulation_mae_db": 0.37971099313618056, "全帯域_40ms_log_rms_mae_db": 0.5546927304073093, "全帯域_40ms_rms_envelope_nrmse": 0.04889597694725294, "全帯域_40ms_envelope_correlation": 0.997359458238171, "全帯域_40ms_modulation_mae_db": 0.19290097967789574, "300–1000 Hz_10ms_log_rms_mae_db": 1.6086023006966053, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09594857692710862, "300–1000 Hz_10ms_envelope_correlation": 0.993506641928562, "300–1000 Hz_10ms_modulation_mae_db": 0.3700121227403092, "300–1000 Hz_40ms_log_rms_mae_db": 0.9560757469477652, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058626674651976664, "300–1000 Hz_40ms_envelope_correlation": 0.9974832334416973, "300–1000 Hz_40ms_modulation_mae_db": 0.32526849693576254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6254916414388, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15427166730367928, "1000–3000 Hz_10ms_envelope_correlation": 0.9823335549624007, "1000–3000 Hz_10ms_modulation_mae_db": 0.5627732496802321, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0007919061539763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09433972231271148, "1000–3000 Hz_40ms_envelope_correlation": 0.9931178590317451, "1000–3000 Hz_40ms_modulation_mae_db": 0.5065628881920635, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6616865622823123, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17312442237252393, "3000–6000 Hz_10ms_envelope_correlation": 0.9780509901275205, "3000–6000 Hz_10ms_modulation_mae_db": 0.4716255539089366, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9396097750889026, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08864380200623555, "3000–6000 Hz_40ms_envelope_correlation": 0.9938475852582196, "3000–6000 Hz_40ms_modulation_mae_db": 0.5921188045226901, "mr_stft_db": 5.804728194528182, "band_mae_db": 0.12643673039623973, "rms_gain_error_db": -0.07983803815772708, "output_peak": 0.41311670985273136}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.118642283992312, "全帯域_10ms_rms_envelope_nrmse": 0.0921707803466636, "全帯域_10ms_envelope_correlation": 0.9915357789815036, "全帯域_10ms_modulation_mae_db": 0.4841982975220449, "全帯域_40ms_log_rms_mae_db": 0.5451243186519932, "全帯域_40ms_rms_envelope_nrmse": 0.048652758475543505, "全帯域_40ms_envelope_correlation": 0.9974274884924308, "全帯域_40ms_modulation_mae_db": 0.4224546000466824, "300–1000 Hz_10ms_log_rms_mae_db": 1.66319093615978, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10600660543715583, "300–1000 Hz_10ms_envelope_correlation": 0.9919797611516767, "300–1000 Hz_10ms_modulation_mae_db": 0.3708969794023688, "300–1000 Hz_40ms_log_rms_mae_db": 0.925549018387473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06126866797128996, "300–1000 Hz_40ms_envelope_correlation": 0.9971264712704555, "300–1000 Hz_40ms_modulation_mae_db": 0.4960993270238401, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6284644975588858, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14159630476321228, "1000–3000 Hz_10ms_envelope_correlation": 0.9851772258308168, "1000–3000 Hz_10ms_modulation_mae_db": 0.4122703733045319, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9550317970903621, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07917025052499646, "1000–3000 Hz_40ms_envelope_correlation": 0.9952808177764566, "1000–3000 Hz_40ms_modulation_mae_db": 0.49433835550443556, "3000–6000 Hz_10ms_log_rms_mae_db": 1.617665118174514, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15215305325843012, "3000–6000 Hz_10ms_envelope_correlation": 0.9829839313936588, "3000–6000 Hz_10ms_modulation_mae_db": 0.5121693945478362, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9274519361802595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0830765300212863, "3000–6000 Hz_40ms_envelope_correlation": 0.9946800288899613, "3000–6000 Hz_40ms_modulation_mae_db": 0.745000240780236, "mr_stft_db": 5.809511720472699, "band_mae_db": 0.1519432629683853, "rms_gain_error_db": -0.09880999300104566, "output_peak": 0.4083588232801261}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1294256085633108, "全帯域_10ms_rms_envelope_nrmse": 0.09148629223325219, "全帯域_10ms_envelope_correlation": 0.991629455542463, "全帯域_10ms_modulation_mae_db": 0.3281970735245736, "全帯域_40ms_log_rms_mae_db": 0.5893117366504116, "全帯域_40ms_rms_envelope_nrmse": 0.048849526080238403, "全帯域_40ms_envelope_correlation": 0.997311124543508, "全帯域_40ms_modulation_mae_db": 0.5021887194270579, "300–1000 Hz_10ms_log_rms_mae_db": 1.6135060659628087, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09464760700935651, "300–1000 Hz_10ms_envelope_correlation": 0.9936256035565649, "300–1000 Hz_10ms_modulation_mae_db": 0.20345867550981614, "300–1000 Hz_40ms_log_rms_mae_db": 0.9092784390093469, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05772424186433933, "300–1000 Hz_40ms_envelope_correlation": 0.9974497843667569, "300–1000 Hz_40ms_modulation_mae_db": 0.4371070924524718, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6391437377719016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15152403737787448, "1000–3000 Hz_10ms_envelope_correlation": 0.9829260928157503, "1000–3000 Hz_10ms_modulation_mae_db": 0.1774430840642773, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0318541206049487, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09406755324284119, "1000–3000 Hz_40ms_envelope_correlation": 0.9930108831178326, "1000–3000 Hz_40ms_modulation_mae_db": 0.26357794968588394, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6384687153467659, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15289774687924254, "3000–6000 Hz_10ms_envelope_correlation": 0.982817351251624, "3000–6000 Hz_10ms_modulation_mae_db": 0.897737947519842, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9766479595727138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08536971368396513, "3000–6000 Hz_40ms_envelope_correlation": 0.9943891976139916, "3000–6000 Hz_40ms_modulation_mae_db": 1.1558518437539063, "mr_stft_db": 5.803996922551559, "band_mae_db": 0.1680434696144323, "rms_gain_error_db": -0.04178148525874072, "output_peak": 0.408935129196913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.966568930198528, "spectral_convergence": 0.4153151089704822, "spectral_convergence_status": "ok", "amplitude_floor": 9.383482434461407e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.675266223053317, "spectral_convergence": 0.3991315176891308, "spectral_convergence_status": "ok", "amplitude_floor": 6.715789372234631e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.827600613709999, "spectral_convergence": 0.4151848093048046, "spectral_convergence_status": "ok", "amplitude_floor": 5.217365851457555e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.0167397027219553e-05, "output_power": 3.094765412750902e-05, "delta_db": 0.11089895520848736, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003363761740772986, "output_power": 0.00034044353561762074, "delta_db": 0.0521986553445736, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005972956425120644, "output_power": 0.0005987556547049646, "delta_db": 0.010602815369330531, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016539661163734624, "output_power": 0.00015332596397672939, "delta_db": -0.329109044017824, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011856598981285755, "output_power": 0.00012518852407971662, "delta_db": 0.23604388081034422, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00046450684077542097, "output_power": 0.00044995250687827947, "delta_db": -0.1382543841653021, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14618462248597697, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03487393914914078, "pitch": null, "output_rms_dbfs": -27.69905341840812, "output_peak": 0.45819522637545207, "reference_rms_dbfs": -27.66417947925898, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.823145255653948}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014203698766261565, "log_rms_mae_db": 1.1148710286342531, "rms_envelope_nrmse": 0.09034002966321346, "envelope_correlation": 0.991821860206185, "rms_gain_error_db": -0.03487031701113736, "reference_mean_power": 0.0017137270951261038, "output_mean_power": 0.0017000223520461527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.499341088840717e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046355332033754323, "output_power": 0.0004457727340661002, "reference_active": true, "delta_db": -0.16986195089770512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011665881324257648, "output_power": 0.00011232400114502664, "reference_active": true, "delta_db": -0.1644498882069858}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5955876441113554e-05, "output_power": 2.1441070541749058e-05, "reference_active": true, "delta_db": -0.8298923243601906}], "active_band_count": 3, "active_band_mae_db": 0.3880680544882938}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274207517780892, "log_rms_mae_db": 0.6089687293841078, "rms_envelope_nrmse": 0.049409351718864944, "envelope_correlation": 0.9972292477808535, "rms_gain_error_db": -0.035833758199560695, "reference_mean_power": 0.0017200719561396216, "output_mean_power": 0.0017059379847859401}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.335969034308142e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004202066244366408, "output_power": 0.00040389960804416335, "reference_active": true, "delta_db": -0.17189462933271882}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.868302162815529e-05, "output_power": 5.66985865990423e-05, "reference_active": true, "delta_db": -0.1494023506863717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5858058061398606e-06, "output_power": 1.274774237825177e-06, "reference_active": true, "delta_db": -0.9481672546476057}], "active_band_count": 3, "active_band_mae_db": 0.42315474488889876}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000125392198184175, "log_rms_mae_db": 1.5516001148829266, "rms_envelope_nrmse": 0.09305226481321047, "envelope_correlation": 0.9938332218150701, "rms_gain_error_db": -0.007851557144429408, "reference_mean_power": 0.0008371717774342565, "output_mean_power": 0.0008356596320611822}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.199403135704403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003464696140419235, "output_power": 0.0003331272145912311, "reference_active": true, "delta_db": -0.1705503875393852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.952130707503092e-05, "output_power": 8.306954611734769e-05, "reference_active": true, "delta_db": -0.3248457704070951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1976615680578699e-05, "output_power": 1.2739310020856974e-05, "reference_active": true, "delta_db": 0.26811792991672756}], "active_band_count": 3, "active_band_mae_db": 0.2545046959544026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011133215930497508, "log_rms_mae_db": 0.8488434351941943, "rms_envelope_nrmse": 0.056971651139885074, "envelope_correlation": 0.9975052686273779, "rms_gain_error_db": -0.007922587538555476, "reference_mean_power": 0.0008413676993576814, "output_mean_power": 0.0008398342391838458}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.575111583882504e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003227285828337414, "output_power": 0.0003093568978009387, "reference_active": true, "delta_db": -0.18377626878575282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.553081813418887e-05, "output_power": 4.2633974339911734e-05, "reference_active": true, "delta_db": -0.28549634654970507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.055134444571575e-06, "output_power": 1.0112014137606185e-06, "reference_active": true, "delta_db": -0.18470132565225908}], "active_band_count": 3, "active_band_mae_db": 0.21799131366257232}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.277196005950351e-05, "log_rms_mae_db": 1.6077535946929284, "rms_envelope_nrmse": 0.13807501765262173, "envelope_correlation": 0.9858620720663217, "rms_gain_error_db": -0.15673140941510671, "reference_mean_power": 0.0001860033831933617, "output_mean_power": 0.0001794104364397182}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.45795784914854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.813551789415923e-05, "output_power": 7.96530420176945e-05, "reference_active": true, "delta_db": -0.43948594286314696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.039396685146323e-05, "output_power": 1.8516941945101146e-05, "reference_active": true, "delta_db": -0.41932444103376826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.503164081364961e-06, "output_power": 3.749623295526486e-06, "reference_active": true, "delta_db": -0.795301329084423}], "active_band_count": 3, "active_band_mae_db": 0.5513705709937794}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5906277437135206e-05, "log_rms_mae_db": 0.9657003907530571, "rms_envelope_nrmse": 0.07887572274448121, "envelope_correlation": 0.99520011264673, "rms_gain_error_db": -0.1569885923857682, "reference_mean_power": 0.00018692252747526107, "output_mean_power": 0.0001802863247576809}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315230721785866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.510469005068111e-05, "output_power": 7.654960925022811e-05, "reference_active": true, "delta_db": -0.46010516191209105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648548828731175e-06, "output_power": 9.053708509587838e-06, "reference_active": true, "delta_db": -0.27635491063501516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.651744211074769e-07, "output_power": 2.633449261368573e-07, "reference_active": true, "delta_db": -1.4197539424490615}], "active_band_count": 3, "active_band_mae_db": 0.7187380049987225}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.827493119721784e-05, "log_rms_mae_db": 1.6859987488654724, "rms_envelope_nrmse": 0.1810457047455501, "envelope_correlation": 0.9758690419009166, "rms_gain_error_db": -0.1634484661616389, "reference_mean_power": 0.0003389197953959056, "output_mean_power": 0.0003264014571769207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5434076582307222e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.258901619859077e-05, "output_power": 6.117415640897909e-05, "reference_active": true, "delta_db": -0.09930136028732738}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4223152137238325e-05, "output_power": 1.2854198446154855e-05, "reference_active": true, "delta_db": -0.43950855105723224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.429050160653041e-06, "output_power": 7.0719584156164645e-06, "reference_active": true, "delta_db": -0.7623894001562426}], "active_band_count": 3, "active_band_mae_db": 0.4337331038336007}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.761964716205697e-05, "log_rms_mae_db": 0.9832453145407271, "rms_envelope_nrmse": 0.10078277741846596, "envelope_correlation": 0.992188734128542, "rms_gain_error_db": -0.16720476629255632, "reference_mean_power": 0.00033859490771126537, "output_mean_power": 0.0003258066509287573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4129997856527214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.165966182107609e-05, "output_power": 5.973119040022797e-05, "reference_active": true, "delta_db": -0.13799968449007688}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.083964993529701e-06, "output_power": 5.989095229241363e-06, "reference_active": true, "delta_db": -0.729151877804567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.531889318612958e-07, "output_power": 4.22909120276325e-07, "reference_active": true, "delta_db": -1.887917660491846}], "active_band_count": 3, "active_band_mae_db": 0.9183564075954966}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8809866799543987, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "17fea53f0d872d16b5bf4194036a2f3b705996210ec5030d30adb7e56d2df459", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9839743008182167, "standard_deviation": 0.001346776468135158, "minimum": 0.9823335549624007, "maximum": 0.9858620720663217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.628801469449204, "standard_deviation": 0.014331784326752398, "minimum": 1.6077535946929284, "maximum": 1.6472713214596304, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.42029892002182323, "standard_deviation": 0.12595991078216257, "minimum": 0.1774430840642773, "maximum": 0.5627732496802321, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14739062381231638, "standard_deviation": 0.00656549136379728, "minimum": 0.13807501765262173, "maximum": 0.15427166730367928, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99430735473985, "standard_deviation": 0.0009932066791875232, "minimum": 0.9930108831178326, "maximum": 0.9956103479719435, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9982898747435125, "standard_deviation": 0.028495416342819532, "minimum": 0.9550317970903621, "maximum": 1.0318541206049487, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.44968559065900765, "standard_deviation": 0.1733529813933375, "minimum": 0.1876438836891757, "maximum": 0.7187380049987225, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0867465105614718, "standard_deviation": 0.007475624399212316, "minimum": 0.07641754442804877, "maximum": 0.09433972231271148, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9810388604677585, "standard_deviation": 0.0026988469121881956, "minimum": 0.9758690419009166, "maximum": 0.983389860164744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6363977597784374, "standard_deviation": 0.025353902811136655, "minimum": 1.6109234616253167, "maximum": 1.6859987488654724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5390875506227588, "standard_deviation": 0.25900470882693727, "minimum": 0.14780438481631705, "maximum": 0.8978587531819242, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16045229836267103, "standard_deviation": 0.011043681716187475, "minimum": 0.15058803729729892, "maximum": 0.1810457047455501, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994019204291125, "standard_deviation": 0.0012445692934343012, "minimum": 0.992188734128542, "maximum": 0.995976018956821, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9493901658892918, "standard_deviation": 0.02378886369993415, "minimum": 0.9198304473261221, "maximum": 0.9832453145407271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8206238976693809, "standard_deviation": 0.2560135363390224, "minimum": 0.4497918460226236, "maximum": 1.1558518437539063, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08740463773419768, "standard_deviation": 0.00915722480569963, "minimum": 0.0718648276205171, "maximum": 0.10078277741846596, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935490037212635, "standard_deviation": 0.000700400610825408, "minimum": 0.9919797611516767, "maximum": 0.9944412937762378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.607436570909192, "standard_deviation": 0.0386422208038597, "minimum": 1.5516001148829266, "maximum": 1.66319093615978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2884025819318916, "standard_deviation": 0.11699806058658592, "minimum": 0.13997235312589948, "maximum": 0.4897424956190097, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09510429053617003, "standard_deviation": 0.0049844144279922896, "minimum": 0.08838913783826932, "maximum": 0.10600660543715583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976286412534223, "standard_deviation": 0.0003219377816718208, "minimum": 0.9971264712704555, "maximum": 0.9981380670052398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.890199511430298, "standard_deviation": 0.04100254531484711, "minimum": 0.8350653742365742, "maximum": 0.9560757469477652, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3476128148636432, "standard_deviation": 0.12900895435072368, "minimum": 0.15761672195018483, "maximum": 0.5172277024693813, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.055789856321332265, "standard_deviation": 0.0037978418943777404, "minimum": 0.049683185973211585, "maximum": 0.06126866797128996, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14640240693916687, "standard_deviation": 0.020968216664799113, "minimum": 0.11569527210895018, "maximum": 0.17128611995573295, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.803577256166463, "standard_deviation": 0.011903880400706857, "minimum": 5.787926021354333, "maximum": 5.823145255653948, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4218494154755057, "standard_deviation": 0.02009624987460075, "minimum": 0.39985621032096236, "maximum": 0.45819522637545207, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05093558711623891, "standard_deviation": 0.03411439867582295, "minimum": -0.09880999300104566, "maximum": -0.005475185526420713, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918098874693619, "standard_deviation": 0.0003251751158271163, "minimum": 0.9913656098989627, "maximum": 0.9922933884810441, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1006615360825718, "standard_deviation": 0.022565233674544553, "minimum": 1.0589077355925491, "maximum": 1.1294256085633108, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33285978424316065, "standard_deviation": 0.09314762495674626, "minimum": 0.21655787517474634, "maximum": 0.4841982975220449, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09050747806065383, "standard_deviation": 0.0018270029803822815, "minimum": 0.08781525691352693, "maximum": 0.09300747105719659, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974345070834139, "standard_deviation": 0.00020122662886421397, "minimum": 0.9971813918217566, "maximum": 0.9977310277799322, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5645706615109656, "standard_deviation": 0.029525432900891296, "minimum": 0.5172554658593013, "maximum": 0.6089687293841078, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36891257472674666, "standard_deviation": 0.099870039325454, "minimum": 0.19290097967789574, "maximum": 0.5021887194270579, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04785727309489986, "standard_deviation": 0.0019902995477686634, "minimum": 0.04504616453093, "maximum": 0.050363479050818845, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2273-0010", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.3783435821533203, "predicted_mos": 2.3783435821533203, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "72283a49b5ca2956614328048b3f94935fae99ea7c5f5f1e4ab0c3a7d2b1ee90", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.990302599966526, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1235043090129402, "全帯域_10ms_rms_envelope_nrmse": 0.09078101317260613, "全帯域_10ms_envelope_correlation": 0.991741379669964, "全帯域_10ms_modulation_mae_db": 0.3061522643869776, "全帯域_40ms_log_rms_mae_db": 0.6066964188562818, "全帯域_40ms_rms_envelope_nrmse": 0.04928766035223408, "全帯域_40ms_envelope_correlation": 0.997244434504886, "全帯域_40ms_modulation_mae_db": 0.3406621888634418, "300–1000 Hz_10ms_log_rms_mae_db": 1.5800189087061758, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09338679727596381, "300–1000 Hz_10ms_envelope_correlation": 0.9937947575800259, "300–1000 Hz_10ms_modulation_mae_db": 0.24906410082222363, "300–1000 Hz_40ms_log_rms_mae_db": 0.8528131322612597, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05653486622799227, "300–1000 Hz_40ms_envelope_correlation": 0.997546567444657, "300–1000 Hz_40ms_modulation_mae_db": 0.16309171321352087, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6514585534447808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13373337993223217, "1000–3000 Hz_10ms_envelope_correlation": 0.9867961986751632, "1000–3000 Hz_10ms_modulation_mae_db": 0.5866697644151199, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9644957419651403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07525871392565724, "1000–3000 Hz_40ms_envelope_correlation": 0.995734298703279, "1000–3000 Hz_40ms_modulation_mae_db": 0.7799369710044846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7207300152214704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18318629313543736, "3000–6000 Hz_10ms_envelope_correlation": 0.9753126776948429, "3000–6000 Hz_10ms_modulation_mae_db": 0.33599289953160577, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0283902536795384, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10048451881652876, "3000–6000 Hz_40ms_envelope_correlation": 0.9922128873896855, "3000–6000 Hz_40ms_modulation_mae_db": 0.8008242165208456, "mr_stft_db": 5.799423142202208, "band_mae_db": 0.13272567397229912, "rms_gain_error_db": -0.03796481661679607, "output_peak": 0.4701823922341644}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.088581940865795, "全帯域_10ms_rms_envelope_nrmse": 0.09109890730745544, "全帯域_10ms_envelope_correlation": 0.9916961250886659, "全帯域_10ms_modulation_mae_db": 0.3869024721099321, "全帯域_40ms_log_rms_mae_db": 0.5778362394802834, "全帯域_40ms_rms_envelope_nrmse": 0.04553914250869623, "全帯域_40ms_envelope_correlation": 0.9976795333099271, "全帯域_40ms_modulation_mae_db": 0.3823022216996781, "300–1000 Hz_10ms_log_rms_mae_db": 1.6370910389858484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09528278127457608, "300–1000 Hz_10ms_envelope_correlation": 0.9935314676889785, "300–1000 Hz_10ms_modulation_mae_db": 0.1850539231393226, "300–1000 Hz_40ms_log_rms_mae_db": 0.8308504847555899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05061073748684386, "300–1000 Hz_40ms_envelope_correlation": 0.9980552614844288, "300–1000 Hz_40ms_modulation_mae_db": 0.34181612748127926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6716184641319936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15335532781841005, "1000–3000 Hz_10ms_envelope_correlation": 0.9830012040172779, "1000–3000 Hz_10ms_modulation_mae_db": 0.7579177382993295, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9846607207996446, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09354226768570659, "1000–3000 Hz_40ms_envelope_correlation": 0.9939349602854414, "1000–3000 Hz_40ms_modulation_mae_db": 0.8634311955810324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6466393779550794, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1542625573158405, "3000–6000 Hz_10ms_envelope_correlation": 0.9825254266819441, "3000–6000 Hz_10ms_modulation_mae_db": 0.6989456270023012, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9749320141336814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08395418221883567, "3000–6000 Hz_40ms_envelope_correlation": 0.994471257567382, "3000–6000 Hz_40ms_modulation_mae_db": 0.8320809605676655, "mr_stft_db": 5.76671310674993, "band_mae_db": 0.14712288245002267, "rms_gain_error_db": -0.05817830079497954, "output_peak": 0.4274009674707028}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0856719219709956, "全帯域_10ms_rms_envelope_nrmse": 0.08911839155882106, "全帯域_10ms_envelope_correlation": 0.992056721880117, "全帯域_10ms_modulation_mae_db": 0.17429015281853688, "全帯域_40ms_log_rms_mae_db": 0.5272517309797643, "全帯域_40ms_rms_envelope_nrmse": 0.04612853511629831, "全帯域_40ms_envelope_correlation": 0.9975811844050991, "全帯域_40ms_modulation_mae_db": 0.28406060473312533, "300–1000 Hz_10ms_log_rms_mae_db": 1.567387301977286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08877035509789057, "300–1000 Hz_10ms_envelope_correlation": 0.9943836486568738, "300–1000 Hz_10ms_modulation_mae_db": 0.16131158305167823, "300–1000 Hz_40ms_log_rms_mae_db": 0.8355159023595695, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05191001184133935, "300–1000 Hz_40ms_envelope_correlation": 0.9979475934335963, "300–1000 Hz_40ms_modulation_mae_db": 0.2692343124886102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6775402712390333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1431103546485466, "1000–3000 Hz_10ms_envelope_correlation": 0.984835933132637, "1000–3000 Hz_10ms_modulation_mae_db": 0.3343948783932387, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0310658691176386, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07741780619315092, "1000–3000 Hz_40ms_envelope_correlation": 0.9955004914969995, "1000–3000 Hz_40ms_modulation_mae_db": 0.38917096667004264, "3000–6000 Hz_10ms_log_rms_mae_db": 1.657391416182744, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16095827771793123, "3000–6000 Hz_10ms_envelope_correlation": 0.9814671318249928, "3000–6000 Hz_10ms_modulation_mae_db": 0.4446319511220791, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9763891461777191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08919978991566348, "3000–6000 Hz_40ms_envelope_correlation": 0.9940116626768692, "3000–6000 Hz_40ms_modulation_mae_db": 0.8148336593926704, "mr_stft_db": 5.795139317548315, "band_mae_db": 0.16255013134186072, "rms_gain_error_db": -0.0022710079794005297, "output_peak": 0.4171704849993194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.103099258502503, "全帯域_10ms_rms_envelope_nrmse": 0.08912619208590843, "全帯域_10ms_envelope_correlation": 0.9920557583315348, "全帯域_10ms_modulation_mae_db": 0.24198330269219145, "全帯域_40ms_log_rms_mae_db": 0.5684396508395217, "全帯域_40ms_rms_envelope_nrmse": 0.04654391897106737, "全帯域_40ms_envelope_correlation": 0.9975561987384337, "全帯域_40ms_modulation_mae_db": 0.32414967620776114, "300–1000 Hz_10ms_log_rms_mae_db": 1.6086809859499125, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09336539257732111, "300–1000 Hz_10ms_envelope_correlation": 0.9937830747102303, "300–1000 Hz_10ms_modulation_mae_db": 0.35758593070861533, "300–1000 Hz_40ms_log_rms_mae_db": 0.8266295541173133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052449591926880285, "300–1000 Hz_40ms_envelope_correlation": 0.9979010739192385, "300–1000 Hz_40ms_modulation_mae_db": 0.49527639960349695, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6836869357408921, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1414867183329191, "1000–3000 Hz_10ms_envelope_correlation": 0.9850463448570773, "1000–3000 Hz_10ms_modulation_mae_db": 0.3527628614086447, "1000–3000 Hz_40ms_log_rms_mae_db": 1.001011221436321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08115513925707898, "1000–3000 Hz_40ms_envelope_correlation": 0.9946050316655191, "1000–3000 Hz_40ms_modulation_mae_db": 0.20802336779725283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.662459044891694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.152547722689675, "3000–6000 Hz_10ms_envelope_correlation": 0.9830087181854582, "3000–6000 Hz_10ms_modulation_mae_db": 0.04290831364374598, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9387187894573739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07440309045149392, "3000–6000 Hz_40ms_envelope_correlation": 0.9956940893778765, "3000–6000 Hz_40ms_modulation_mae_db": 0.39908339938651644, "mr_stft_db": 5.773757912493903, "band_mae_db": 0.11793652708691742, "rms_gain_error_db": -0.019697309569324917, "output_peak": 0.4317643560670583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.057564774069697, "全帯域_10ms_rms_envelope_nrmse": 0.09113915974275022, "全帯域_10ms_envelope_correlation": 0.9916946770634478, "全帯域_10ms_modulation_mae_db": 0.10871344678323519, "全帯域_40ms_log_rms_mae_db": 0.5583830831612963, "全帯域_40ms_rms_envelope_nrmse": 0.05183539415160709, "全帯域_40ms_envelope_correlation": 0.9969976568179397, "全帯域_40ms_modulation_mae_db": 0.21314920522366612, "300–1000 Hz_10ms_log_rms_mae_db": 1.5832130370196429, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0968045230052823, "300–1000 Hz_10ms_envelope_correlation": 0.9933140360806678, "300–1000 Hz_10ms_modulation_mae_db": 0.44243733041772537, "300–1000 Hz_40ms_log_rms_mae_db": 0.8584465152670421, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056948878894392534, "300–1000 Hz_40ms_envelope_correlation": 0.9975175700930463, "300–1000 Hz_40ms_modulation_mae_db": 0.12254662054568129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6990109574832473, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1547728872934886, "1000–3000 Hz_10ms_envelope_correlation": 0.9829862568895039, "1000–3000 Hz_10ms_modulation_mae_db": 0.35535967017724285, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0425436914647748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0947007785699623, "1000–3000 Hz_40ms_envelope_correlation": 0.9942329668261514, "1000–3000 Hz_40ms_modulation_mae_db": 0.41761077024926013, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6362896342594913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1651746559807774, "3000–6000 Hz_10ms_envelope_correlation": 0.9799594401928399, "3000–6000 Hz_10ms_modulation_mae_db": 0.3164728159770545, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9799223254380517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10111433450018088, "3000–6000 Hz_40ms_envelope_correlation": 0.9919821206898197, "3000–6000 Hz_40ms_modulation_mae_db": 0.31181723911243675, "mr_stft_db": 5.770347016509452, "band_mae_db": 0.15741665367878171, "rms_gain_error_db": -0.07276657014335297, "output_peak": 0.4094519458662779}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.074658747161768, "全帯域_10ms_rms_envelope_nrmse": 0.09456168714782956, "全帯域_10ms_envelope_correlation": 0.9910648423596239, "全帯域_10ms_modulation_mae_db": 0.31567991926866323, "全帯域_40ms_log_rms_mae_db": 0.5626783196232037, "全帯域_40ms_rms_envelope_nrmse": 0.05102080543146742, "全帯域_40ms_envelope_correlation": 0.9971079944547854, "全帯域_40ms_modulation_mae_db": 0.1526741439266052, "300–1000 Hz_10ms_log_rms_mae_db": 1.5745376916857114, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09635370503746704, "300–1000 Hz_10ms_envelope_correlation": 0.993443266804067, "300–1000 Hz_10ms_modulation_mae_db": 0.2823506259393162, "300–1000 Hz_40ms_log_rms_mae_db": 0.8647266041835185, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059285659481843396, "300–1000 Hz_40ms_envelope_correlation": 0.9974174273135152, "300–1000 Hz_40ms_modulation_mae_db": 0.2439333000692945, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6763430481465804, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1555379416524942, "1000–3000 Hz_10ms_envelope_correlation": 0.9820802087676188, "1000–3000 Hz_10ms_modulation_mae_db": 0.5647070715776569, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0054588708510706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09337450017430488, "1000–3000 Hz_40ms_envelope_correlation": 0.9933665691754574, "1000–3000 Hz_40ms_modulation_mae_db": 0.5378440756898571, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7118197191131956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1760450357714862, "3000–6000 Hz_10ms_envelope_correlation": 0.9775068848541201, "3000–6000 Hz_10ms_modulation_mae_db": 0.4552927281501771, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9872228361378098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08966753620981076, "3000–6000 Hz_40ms_envelope_correlation": 0.9937739302142503, "3000–6000 Hz_40ms_modulation_mae_db": 0.5200086100363864, "mr_stft_db": 5.782182181895801, "band_mae_db": 0.114757841518642, "rms_gain_error_db": -0.07591587218888711, "output_peak": 0.418998519405993}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1246191075435017, "全帯域_10ms_rms_envelope_nrmse": 0.09410248618556864, "全帯域_10ms_envelope_correlation": 0.9911807251875362, "全帯域_10ms_modulation_mae_db": 0.4196325408996171, "全帯域_40ms_log_rms_mae_db": 0.5538747184148404, "全帯域_40ms_rms_envelope_nrmse": 0.05088411131355741, "全帯域_40ms_envelope_correlation": 0.9971914686330676, "全帯域_40ms_modulation_mae_db": 0.3608521682503747, "300–1000 Hz_10ms_log_rms_mae_db": 1.660969588992021, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10793561048576451, "300–1000 Hz_10ms_envelope_correlation": 0.991680838771397, "300–1000 Hz_10ms_modulation_mae_db": 0.3320430336062728, "300–1000 Hz_40ms_log_rms_mae_db": 0.9274441973389382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06377916209213001, "300–1000 Hz_40ms_envelope_correlation": 0.9968934170352193, "300–1000 Hz_40ms_modulation_mae_db": 0.44265510023422444, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6780100852270048, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14344020893045673, "1000–3000 Hz_10ms_envelope_correlation": 0.9848992502438326, "1000–3000 Hz_10ms_modulation_mae_db": 0.51010438971873, "1000–3000 Hz_40ms_log_rms_mae_db": 1.009037930626327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08262876382950098, "1000–3000 Hz_40ms_envelope_correlation": 0.9950062545193133, "1000–3000 Hz_40ms_modulation_mae_db": 0.583005583339657, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6441159809069235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15336831005898593, "3000–6000 Hz_10ms_envelope_correlation": 0.9827594938375214, "3000–6000 Hz_10ms_modulation_mae_db": 0.39070765313661665, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9551328637914231, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08420208573247047, "3000–6000 Hz_40ms_envelope_correlation": 0.9944861463254762, "3000–6000 Hz_40ms_modulation_mae_db": 0.5526938486178126, "mr_stft_db": 5.777816919690487, "band_mae_db": 0.1397539928727229, "rms_gain_error_db": -0.10498590560488427, "output_peak": 0.4170008981874211}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1390723685356376, "全帯域_10ms_rms_envelope_nrmse": 0.09178898000059106, "全帯域_10ms_envelope_correlation": 0.9915759585558126, "全帯域_10ms_modulation_mae_db": 0.32194983339049793, "全帯域_40ms_log_rms_mae_db": 0.5854041092395457, "全帯域_40ms_rms_envelope_nrmse": 0.049133709581445535, "全帯域_40ms_envelope_correlation": 0.997270042094244, "全帯域_40ms_modulation_mae_db": 0.48312911249869517, "300–1000 Hz_10ms_log_rms_mae_db": 1.5978495154867283, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09438406513568576, "300–1000 Hz_10ms_envelope_correlation": 0.9936649248615869, "300–1000 Hz_10ms_modulation_mae_db": 0.1805853492120448, "300–1000 Hz_40ms_log_rms_mae_db": 0.8490419044414272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05742675966660994, "300–1000 Hz_40ms_envelope_correlation": 0.9974710702990143, "300–1000 Hz_40ms_modulation_mae_db": 0.34511874005393683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.686877597421986, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14999600613239147, "1000–3000 Hz_10ms_envelope_correlation": 0.9832384874898107, "1000–3000 Hz_10ms_modulation_mae_db": 0.12736017243747197, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0188941399575955, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08977703348577397, "1000–3000 Hz_40ms_envelope_correlation": 0.9936397953657762, "1000–3000 Hz_40ms_modulation_mae_db": 0.2538308532002622, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6991615021479596, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15328881887217782, "3000–6000 Hz_10ms_envelope_correlation": 0.9827281913189329, "3000–6000 Hz_10ms_modulation_mae_db": 0.8649706195470696, "3000–6000 Hz_40ms_log_rms_mae_db": 1.032575619822872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0829420408445245, "3000–6000 Hz_40ms_envelope_correlation": 0.9946665212851222, "3000–6000 Hz_40ms_modulation_mae_db": 1.1116626765151756, "mr_stft_db": 5.780557876529286, "band_mae_db": 0.14223785144746962, "rms_gain_error_db": -0.028292994175558736, "output_peak": 0.4079432306866758}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.963784990047327, "spectral_convergence": 0.414928568806752, "spectral_convergence_status": "ok", "amplitude_floor": 9.383482434461407e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.658714088185671, "spectral_convergence": 0.398708248197182, "spectral_convergence_status": "ok", "amplitude_floor": 6.715789372234631e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.775770348373626, "spectral_convergence": 0.4135466699753702, "spectral_convergence_status": "ok", "amplitude_floor": 5.217365851457555e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.0167397027219553e-05, "output_power": 3.0267567827661552e-05, "delta_db": 0.01439685315395141, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003363761740772986, "output_power": 0.00034008241472139727, "delta_db": 0.04758949011957409, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005972956425120644, "output_power": 0.0006008810702371666, "delta_db": 0.025991760772793415, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016539661163734624, "output_power": 0.0001525582405827473, "delta_db": -0.35090936772021564, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011856598981285755, "output_power": 0.00012447843582642136, "delta_db": 0.21133991183340217, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00046450684077542097, "output_power": 0.00044913763538355285, "delta_db": -0.146126660233858, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13272567397229912, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03796481661679607, "pitch": null, "output_rms_dbfs": -27.702144295875776, "output_peak": 0.4701823922341644, "reference_rms_dbfs": -27.66417947925898, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.799423142202208}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014203698766261565, "log_rms_mae_db": 1.1235043090129402, "rms_envelope_nrmse": 0.09078101317260613, "envelope_correlation": 0.991741379669964, "rms_gain_error_db": -0.03796165392289162, "reference_mean_power": 0.0017137270951261038, "output_mean_power": 0.0016988126954383162}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.499341088840717e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046355332033754323, "output_power": 0.00044578109503897367, "reference_active": true, "delta_db": -0.1697804948091582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011665881324257648, "output_power": 0.00011291152854012951, "reference_active": true, "delta_db": -0.1417926738015991}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5955876441113554e-05, "output_power": 2.2570819912804472e-05, "reference_active": true, "delta_db": -0.6068836245501756}], "active_band_count": 3, "active_band_mae_db": 0.3061522643869776}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274207517780892, "log_rms_mae_db": 0.6066964188562818, "rms_envelope_nrmse": 0.04928766035223408, "envelope_correlation": 0.997244434504886, "rms_gain_error_db": -0.037775840065827224, "reference_mean_power": 0.0017200719561396216, "output_mean_power": 0.0017051752924878308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.335969034308142e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004202066244366408, "output_power": 0.0004041901160868458, "reference_active": true, "delta_db": -0.16877205416431418}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.868302162815529e-05, "output_power": 5.716659309676388e-05, "reference_active": true, "delta_db": -0.11370157112138302}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5858058061398606e-06, "output_power": 1.337515148504781e-06, "reference_active": true, "delta_db": -0.7395129413046282}], "active_band_count": 3, "active_band_mae_db": 0.3406621888634418}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000125392198184175, "log_rms_mae_db": 1.5800189087061758, "rms_envelope_nrmse": 0.09338679727596381, "envelope_correlation": 0.9937947575800259, "rms_gain_error_db": 0.003919463311807191, "reference_mean_power": 0.0008371717774342565, "output_mean_power": 0.0008379276574427049}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.199403135704403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003464696140419235, "output_power": 0.00033519324428136484, "reference_active": true, "delta_db": -0.14369895301430197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.952130707503092e-05, "output_power": 8.369130518577442e-05, "reference_active": true, "delta_db": -0.2924607373932716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1976615680578699e-05, "output_power": 1.2865817148900408e-05, "reference_active": true, "delta_db": 0.3110326120590974}], "active_band_count": 3, "active_band_mae_db": 0.24906410082222363}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011133215930497508, "log_rms_mae_db": 0.8528131322612597, "rms_envelope_nrmse": 0.05653486622799227, "envelope_correlation": 0.997546567444657, "rms_gain_error_db": 0.003949406175675271, "reference_mean_power": 0.0008413676993576814, "output_mean_power": 0.0008421331740027497}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.575111583882504e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003227285828337414, "output_power": 0.0003112955608571117, "reference_active": true, "delta_db": -0.1566450331292596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.553081813418887e-05, "output_power": 4.33197009261326e-05, "reference_active": true, "delta_db": -0.2162000448107476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.055134444571575e-06, "output_power": 1.0272231410461958e-06, "reference_active": true, "delta_db": -0.11643006170055535}], "active_band_count": 3, "active_band_mae_db": 0.16309171321352087}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.277196005950351e-05, "log_rms_mae_db": 1.6514585534447808, "rms_envelope_nrmse": 0.13373337993223217, "envelope_correlation": 0.9867961986751632, "rms_gain_error_db": -0.19421554582882544, "reference_mean_power": 0.0001860033831933617, "output_mean_power": 0.00017786860093463506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.45795784914854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.813551789415923e-05, "output_power": 8.002991523026564e-05, "reference_active": true, "delta_db": -0.41898603938949386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.039396685146323e-05, "output_power": 1.8144234082085446e-05, "reference_active": true, "delta_db": -0.5076306889955841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.503164081364961e-06, "output_power": 3.7168797864620616e-06, "reference_active": true, "delta_db": -0.8333925648602819}], "active_band_count": 3, "active_band_mae_db": 0.5866697644151199}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5906277437135206e-05, "log_rms_mae_db": 0.9644957419651403, "rms_envelope_nrmse": 0.07525871392565724, "envelope_correlation": 0.995734298703279, "rms_gain_error_db": -0.19420034930104757, "reference_mean_power": 0.00018692252747526107, "output_mean_power": 0.00017874817227743432}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315230721785866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.510469005068111e-05, "output_power": 7.67996823214704e-05, "reference_active": true, "delta_db": -0.44594070766881017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648548828731175e-06, "output_power": 8.858973957475353e-06, "reference_active": true, "delta_db": -0.3707857403537955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.651744211074769e-07, "output_power": 2.5715318220515556e-07, "reference_active": true, "delta_db": -1.523084464990848}], "active_band_count": 3, "active_band_mae_db": 0.7799369710044846}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.827493119721784e-05, "log_rms_mae_db": 1.7207300152214704, "rms_envelope_nrmse": 0.18318629313543736, "envelope_correlation": 0.9753126776948429, "rms_gain_error_db": -0.15114170775134422, "reference_mean_power": 0.0003389197953959056, "output_mean_power": 0.0003273277044364772}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5434076582307222e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.258901619859077e-05, "output_power": 6.21792665571654e-05, "reference_active": true, "delta_db": -0.02852530211312474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4223152137238325e-05, "output_power": 1.3492906520693371e-05, "reference_active": true, "delta_db": -0.22890343908059602}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.429050160653041e-06, "output_power": 7.0912638134452286e-06, "reference_active": true, "delta_db": -0.7505499574010965}], "active_band_count": 3, "active_band_mae_db": 0.33599289953160577}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.761964716205697e-05, "log_rms_mae_db": 1.0283902536795384, "rms_envelope_nrmse": 0.10048451881652876, "envelope_correlation": 0.9922128873896855, "rms_gain_error_db": -0.148906523445281, "reference_mean_power": 0.00033859490771126537, "output_mean_power": 0.0003271822765461011}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4129997856527214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.165966182107609e-05, "output_power": 6.076298677187708e-05, "reference_active": true, "delta_db": -0.0636202523639355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.083964993529701e-06, "output_power": 6.227403298778771e-06, "reference_active": true, "delta_db": -0.5596941400279789}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.531889318612958e-07, "output_power": 4.3363366524631163e-07, "reference_active": true, "delta_db": -1.7791582571706221}], "active_band_count": 3, "active_band_mae_db": 0.8008242165208456}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8807455856028628, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ac9ad9bf459d8e8ee84349684ee1e23dd0129c729b94e682f36e0e971675b2fe", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9841104855091152, "standard_deviation": 0.0015405987519086343, "minimum": 0.9820802087676188, "maximum": 0.9867961986751632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.67806823910444, "standard_deviation": 0.013631399586647327, "minimum": 1.6514585534447808, "maximum": 1.6990109574832473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4486595683034293, "standard_deviation": 0.1949801050520146, "minimum": 0.12736017243747197, "maximum": 0.7579177382993295, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14692910309261736, "standard_deviation": 0.007716783139676776, "minimum": 0.13373337993223217, "maximum": 0.1555379416524942, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945025460047421, "standard_deviation": 0.0008624668936236542, "minimum": 0.9933665691754574, "maximum": 0.995734298703279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0071460232773142, "standard_deviation": 0.02487646362540502, "minimum": 0.9644957419651403, "maximum": 1.0425436914647748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5041067229414811, "standard_deviation": 0.23408020102910945, "minimum": 0.20802336779725283, "maximum": 0.8634311955810324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08598187539014199, "standard_deviation": 0.007792882671595336, "minimum": 0.07525871392565724, "maximum": 0.0947007785699623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9806584955738316, "standard_deviation": 0.0028625613456747467, "minimum": 0.9753126776948429, "maximum": 0.9830087181854582, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6723258363348197, "standard_deviation": 0.03316080452295883, "minimum": 1.6362896342594913, "maximum": 1.7207300152214704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4437403260138312, "standard_deviation": 0.248993474495912, "minimum": 0.04290831364374598, "maximum": 0.8649706195470696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16235395894278892, "standard_deviation": 0.01168569143750642, "minimum": 0.152547722689675, "maximum": 0.18318629313543736, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9939123269408102, "standard_deviation": 0.0012551591148047216, "minimum": 0.9919821206898197, "maximum": 0.9956940893778765, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9841604810798086, "standard_deviation": 0.03247140847128776, "minimum": 0.9387187894573739, "maximum": 1.032575619822872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6678755762686887, "standard_deviation": 0.26654955626547466, "minimum": 0.31181723911243675, "maximum": 1.1116626765151756, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08824594733618855, "standard_deviation": 0.009046206197460672, "minimum": 0.07440309045149392, "maximum": 0.10111433450018088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9934495018942284, "standard_deviation": 0.0007840177692722974, "minimum": 0.991680838771397, "maximum": 0.9943836486568738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6012185086004158, "standard_deviation": 0.03285053881346425, "minimum": 1.567387301977286, "maximum": 1.660969588992021, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2738039846121499, "standard_deviation": 0.09910624989019089, "minimum": 0.16131158305167823, "maximum": 0.44243733041772537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09578540373624389, "standard_deviation": 0.005503585170219915, "minimum": 0.08877035509789057, "maximum": 0.10793561048576451, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975937476278395, "standard_deviation": 0.00037364648096522874, "minimum": 0.9968934170352193, "maximum": 0.9980552614844288, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8556835368405823, "standard_deviation": 0.03197968525662249, "minimum": 0.8266295541173133, "maximum": 0.9274441973389382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.30295903921125555, "standard_deviation": 0.1290189216587087, "minimum": 0.12254662054568129, "maximum": 0.49527639960349695, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05611820845225396, "standard_deviation": 0.004355600375616415, "minimum": 0.05061073748684386, "maximum": 0.06377916209213001, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1393126942960895, "standard_deviation": 0.017074426615776794, "minimum": 0.114757841518642, "maximum": 0.16255013134186072, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.780742184202422, "standard_deviation": 0.011467811799870211, "minimum": 5.76671310674993, "maximum": 5.799423142202208, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4249890993647016, "standard_deviation": 0.019953334294353774, "minimum": 0.4079432306866758, "maximum": 0.4701823922341644, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.050009097134148014, "standard_deviation": 0.03400132848663417, "minimum": -0.10498590560488427, "maximum": -0.0022710079794005297, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9916332735170879, "standard_deviation": 0.0003602397151455844, "minimum": 0.9910648423596239, "maximum": 0.992056721880117, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0995965534578547, "standard_deviation": 0.027959408386674385, "minimum": 1.057564774069697, "maximum": 1.1390723685356376, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2844129915437065, "standard_deviation": 0.10450242617665878, "minimum": 0.10871344678323519, "maximum": 0.4196325408996171, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09146460215019131, "standard_deviation": 0.0020119426927781054, "minimum": 0.08911839155882106, "maximum": 0.09456168714782956, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973285641197978, "standard_deviation": 0.0002467683448603611, "minimum": 0.9969976568179397, "maximum": 0.9976795333099271, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5675705338243422, "standard_deviation": 0.02353276014607726, "minimum": 0.5272517309797643, "maximum": 0.6066964188562818, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31762241517541845, "standard_deviation": 0.10230708046697673, "minimum": 0.1526741439266052, "maximum": 0.48312911249869517, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04879665967829668, "standard_deviation": 0.00244066787234834, "minimum": 0.04553914250869623, "maximum": 0.05183539415160709, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2273-0010", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.1733555793762207, "predicted_mos": 2.1733555793762207, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c042a7ba3d98b85e039ea20d71f3c0aede3cd4f79e0bb1b37d0a1067398bdeaa", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.867664999794215, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0364349292438881, "全帯域_10ms_rms_envelope_nrmse": 0.08945763037384565, "全帯域_10ms_envelope_correlation": 0.9920452972568216, "全帯域_10ms_modulation_mae_db": 0.49147130916683573, "全帯域_40ms_log_rms_mae_db": 0.5495850893895352, "全帯域_40ms_rms_envelope_nrmse": 0.04927283475948222, "全帯域_40ms_envelope_correlation": 0.9973601251444124, "全帯域_40ms_modulation_mae_db": 0.6120423113550415, "300–1000 Hz_10ms_log_rms_mae_db": 1.9391231183366642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09584527400897677, "300–1000 Hz_10ms_envelope_correlation": 0.9935571321854677, "300–1000 Hz_10ms_modulation_mae_db": 0.14208649933370565, "300–1000 Hz_40ms_log_rms_mae_db": 1.1564912352079675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05347777579481125, "300–1000 Hz_40ms_envelope_correlation": 0.997951173416276, "300–1000 Hz_40ms_modulation_mae_db": 0.3972914560183598, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9467221092082105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1594332726121034, "1000–3000 Hz_10ms_envelope_correlation": 0.9811237318270019, "1000–3000 Hz_10ms_modulation_mae_db": 0.38685909876181973, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1818347759658678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09042334163719318, "1000–3000 Hz_40ms_envelope_correlation": 0.9937055409579525, "1000–3000 Hz_40ms_modulation_mae_db": 0.35011396098871767, "3000–6000 Hz_10ms_log_rms_mae_db": 2.003145138183495, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16195903200436693, "3000–6000 Hz_10ms_envelope_correlation": 0.9806877077408879, "3000–6000 Hz_10ms_modulation_mae_db": 0.7747000204629327, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1635716182727776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09318333948209996, "3000–6000 Hz_40ms_envelope_correlation": 0.9932102285283563, "3000–6000 Hz_40ms_modulation_mae_db": 1.2620441504918212, "mr_stft_db": 6.104267494586069, "band_mae_db": 0.22494520143472616, "rms_gain_error_db": -0.09803392249114758, "output_peak": 0.45305673321044904}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0762754803778631, "全帯域_10ms_rms_envelope_nrmse": 0.09179772943501528, "全帯域_10ms_envelope_correlation": 0.9917106237351355, "全帯域_10ms_modulation_mae_db": 0.6079782118746123, "全帯域_40ms_log_rms_mae_db": 0.5443192016728651, "全帯域_40ms_rms_envelope_nrmse": 0.05084267446603165, "全帯域_40ms_envelope_correlation": 0.997306144614012, "全帯域_40ms_modulation_mae_db": 0.8234982975552237, "300–1000 Hz_10ms_log_rms_mae_db": 1.9407908649995385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10172653380195121, "300–1000 Hz_10ms_envelope_correlation": 0.9927669761240867, "300–1000 Hz_10ms_modulation_mae_db": 0.2751592335455923, "300–1000 Hz_40ms_log_rms_mae_db": 1.1347545864744324, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06246326674355813, "300–1000 Hz_40ms_envelope_correlation": 0.9972049529158323, "300–1000 Hz_40ms_modulation_mae_db": 0.4437471820588685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9170526622530861, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15981216090526956, "1000–3000 Hz_10ms_envelope_correlation": 0.9812095280952807, "1000–3000 Hz_10ms_modulation_mae_db": 0.9620992979605321, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1636058516472516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0894487590368661, "1000–3000 Hz_40ms_envelope_correlation": 0.9941594751805753, "1000–3000 Hz_40ms_modulation_mae_db": 1.1697306660104796, "3000–6000 Hz_10ms_log_rms_mae_db": 1.982979305043725, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17461927646931796, "3000–6000 Hz_10ms_envelope_correlation": 0.9777550008546186, "3000–6000 Hz_10ms_modulation_mae_db": 1.1360608077529328, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1851736178525913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10105359496378674, "3000–6000 Hz_40ms_envelope_correlation": 0.9926072869494409, "3000–6000 Hz_40ms_modulation_mae_db": 1.839138426190533, "mr_stft_db": 6.119473492242647, "band_mae_db": 0.24198330760349418, "rms_gain_error_db": -0.1353686383627127, "output_peak": 0.38991434006504905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0413068176539384, "全帯域_10ms_rms_envelope_nrmse": 0.08878282897212258, "全帯域_10ms_envelope_correlation": 0.9921036528801405, "全帯域_10ms_modulation_mae_db": 0.6137420457767692, "全帯域_40ms_log_rms_mae_db": 0.5310630775508631, "全帯域_40ms_rms_envelope_nrmse": 0.0459561197613194, "全帯域_40ms_envelope_correlation": 0.9976048415347931, "全帯域_40ms_modulation_mae_db": 0.6568442918182766, "300–1000 Hz_10ms_log_rms_mae_db": 1.984905721726971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09718589970670634, "300–1000 Hz_10ms_envelope_correlation": 0.9932800135768954, "300–1000 Hz_10ms_modulation_mae_db": 0.2582243626363725, "300–1000 Hz_40ms_log_rms_mae_db": 1.16070210623701, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05430220833750191, "300–1000 Hz_40ms_envelope_correlation": 0.9977554299910868, "300–1000 Hz_40ms_modulation_mae_db": 0.45900701599068094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8830976752576345, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.153335849563363, "1000–3000 Hz_10ms_envelope_correlation": 0.9825462738064775, "1000–3000 Hz_10ms_modulation_mae_db": 0.8823230677011015, "1000–3000 Hz_40ms_log_rms_mae_db": 1.115462445575521, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08686754698239307, "1000–3000 Hz_40ms_envelope_correlation": 0.9941309636935934, "1000–3000 Hz_40ms_modulation_mae_db": 1.0722512968503035, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0115507421706615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17532098184869882, "3000–6000 Hz_10ms_envelope_correlation": 0.9774939241306623, "3000–6000 Hz_10ms_modulation_mae_db": 1.0104954341047911, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2035439614471357, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09680498954091903, "3000–6000 Hz_40ms_envelope_correlation": 0.9926333924401397, "3000–6000 Hz_40ms_modulation_mae_db": 1.2965465777509877, "mr_stft_db": 6.099070581540551, "band_mae_db": 0.08875690209268873, "rms_gain_error_db": -0.034466754319612686, "output_peak": 0.4411555856480645}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1222259044880702, "全帯域_10ms_rms_envelope_nrmse": 0.0894581718591661, "全帯域_10ms_envelope_correlation": 0.9919907040276907, "全帯域_10ms_modulation_mae_db": 0.46091580055707926, "全帯域_40ms_log_rms_mae_db": 0.5860419680540867, "全帯域_40ms_rms_envelope_nrmse": 0.047081803823694555, "全帯域_40ms_envelope_correlation": 0.9974762720917382, "全帯域_40ms_modulation_mae_db": 0.7321604800331686, "300–1000 Hz_10ms_log_rms_mae_db": 2.0345128091868783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10167083583141388, "300–1000 Hz_10ms_envelope_correlation": 0.9926726246047674, "300–1000 Hz_10ms_modulation_mae_db": 0.3489205076039776, "300–1000 Hz_40ms_log_rms_mae_db": 1.1741012666544055, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05885157437375767, "300–1000 Hz_40ms_envelope_correlation": 0.9974062258310257, "300–1000 Hz_40ms_modulation_mae_db": 0.6728405920434494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9427671012871708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16338514569602017, "1000–3000 Hz_10ms_envelope_correlation": 0.9800794898220518, "1000–3000 Hz_10ms_modulation_mae_db": 0.9871037612490836, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2119822073775632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10770275392528529, "1000–3000 Hz_40ms_envelope_correlation": 0.9905302567041415, "1000–3000 Hz_40ms_modulation_mae_db": 0.9910293264291065, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9204609080747852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16259021051149894, "3000–6000 Hz_10ms_envelope_correlation": 0.9805886309105201, "3000–6000 Hz_10ms_modulation_mae_db": 1.0699051185646824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1012370589628708, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0827161202319279, "3000–6000 Hz_40ms_envelope_correlation": 0.9946222697081849, "3000–6000 Hz_40ms_modulation_mae_db": 1.6668869370256265, "mr_stft_db": 6.109780674812704, "band_mae_db": 0.17897556081230173, "rms_gain_error_db": -0.007304845721376865, "output_peak": 0.37820731654418516}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0748362761755472, "全帯域_10ms_rms_envelope_nrmse": 0.08929611910502305, "全帯域_10ms_envelope_correlation": 0.9920252018008454, "全帯域_10ms_modulation_mae_db": 0.36494440788372984, "全帯域_40ms_log_rms_mae_db": 0.5722258459496357, "全帯域_40ms_rms_envelope_nrmse": 0.05129143961294954, "全帯域_40ms_envelope_correlation": 0.9970348983145596, "全帯域_40ms_modulation_mae_db": 0.39338738975533943, "300–1000 Hz_10ms_log_rms_mae_db": 2.01938444174246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11016810418505228, "300–1000 Hz_10ms_envelope_correlation": 0.9913431953366576, "300–1000 Hz_10ms_modulation_mae_db": 0.40119927758791474, "300–1000 Hz_40ms_log_rms_mae_db": 1.23255597749451, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0670900470765622, "300–1000 Hz_40ms_envelope_correlation": 0.996576401391549, "300–1000 Hz_40ms_modulation_mae_db": 0.14307163881693968, "1000–3000 Hz_10ms_log_rms_mae_db": 1.903785355237673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15182272826990947, "1000–3000 Hz_10ms_envelope_correlation": 0.9833502501302485, "1000–3000 Hz_10ms_modulation_mae_db": 0.6316866067663137, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1783261046306772, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09182053624812153, "1000–3000 Hz_40ms_envelope_correlation": 0.9940999715942792, "1000–3000 Hz_40ms_modulation_mae_db": 0.6778147056089482, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9742971158185667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1705000794022945, "3000–6000 Hz_10ms_envelope_correlation": 0.9792569307881143, "3000–6000 Hz_10ms_modulation_mae_db": 0.8136815765557337, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1675344242947603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0982134639137334, "3000–6000 Hz_40ms_envelope_correlation": 0.9927619939962647, "3000–6000 Hz_40ms_modulation_mae_db": 1.3508130534656229, "mr_stft_db": 6.1014749365310905, "band_mae_db": 0.21461476330291993, "rms_gain_error_db": -0.013600598773503376, "output_peak": 0.4200364604778065}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0357116344679416, "全帯域_10ms_rms_envelope_nrmse": 0.0865563378117362, "全帯域_10ms_envelope_correlation": 0.9925759966568336, "全帯域_10ms_modulation_mae_db": 0.35988927750159144, "全帯域_40ms_log_rms_mae_db": 0.5619362152217819, "全帯域_40ms_rms_envelope_nrmse": 0.04808652475311197, "全帯域_40ms_envelope_correlation": 0.9975271049596821, "全帯域_40ms_modulation_mae_db": 0.46911829730864074, "300–1000 Hz_10ms_log_rms_mae_db": 1.9466156578310363, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09669678039161836, "300–1000 Hz_10ms_envelope_correlation": 0.9933739206971287, "300–1000 Hz_10ms_modulation_mae_db": 0.3958207735225146, "300–1000 Hz_40ms_log_rms_mae_db": 1.2257821876152553, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059290691306128525, "300–1000 Hz_40ms_envelope_correlation": 0.9973723949921126, "300–1000 Hz_40ms_modulation_mae_db": 0.6396808839542639, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9570315574323176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15929331228139543, "1000–3000 Hz_10ms_envelope_correlation": 0.9814152873518667, "1000–3000 Hz_10ms_modulation_mae_db": 0.4929385779861921, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1472324149360036, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0952331622988376, "1000–3000 Hz_40ms_envelope_correlation": 0.993453809208353, "1000–3000 Hz_40ms_modulation_mae_db": 0.3401510759890051, "3000–6000 Hz_10ms_log_rms_mae_db": 1.984868350576681, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1665734441235495, "3000–6000 Hz_10ms_envelope_correlation": 0.9795944426774335, "3000–6000 Hz_10ms_modulation_mae_db": 0.5748409178542252, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1519206500044832, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0911261802794952, "3000–6000 Hz_40ms_envelope_correlation": 0.9937729932141415, "3000–6000 Hz_40ms_modulation_mae_db": 0.7937506579879625, "mr_stft_db": 6.0948956364504925, "band_mae_db": 0.21901133079384172, "rms_gain_error_db": -0.10993345706300775, "output_peak": 0.44538428814468844}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0627652427211907, "全帯域_10ms_rms_envelope_nrmse": 0.08676394725330618, "全帯域_10ms_envelope_correlation": 0.9924679480672615, "全帯域_10ms_modulation_mae_db": 0.5114213295443056, "全帯域_40ms_log_rms_mae_db": 0.5797274073679392, "全帯域_40ms_rms_envelope_nrmse": 0.04434563368355743, "全帯域_40ms_envelope_correlation": 0.9977908021373671, "全帯域_40ms_modulation_mae_db": 0.5016653252235221, "300–1000 Hz_10ms_log_rms_mae_db": 2.0370420725875906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10891618599504337, "300–1000 Hz_10ms_envelope_correlation": 0.99158274700485, "300–1000 Hz_10ms_modulation_mae_db": 0.1987737267931703, "300–1000 Hz_40ms_log_rms_mae_db": 1.2606810712455319, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06609331301534915, "300–1000 Hz_40ms_envelope_correlation": 0.9967220242910002, "300–1000 Hz_40ms_modulation_mae_db": 0.33630212515156627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9112529296186083, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1569143545182945, "1000–3000 Hz_10ms_envelope_correlation": 0.981619447745981, "1000–3000 Hz_10ms_modulation_mae_db": 0.8068145388659135, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1742976851598586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09647270775527365, "1000–3000 Hz_40ms_envelope_correlation": 0.9925238858648638, "1000–3000 Hz_40ms_modulation_mae_db": 0.8533295049086799, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9394784871640534, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15536229796178297, "3000–6000 Hz_10ms_envelope_correlation": 0.9822381002261711, "3000–6000 Hz_10ms_modulation_mae_db": 1.228503208380459, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0946976082114799, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07007368667959948, "3000–6000 Hz_40ms_envelope_correlation": 0.9962311928055776, "3000–6000 Hz_40ms_modulation_mae_db": 1.688402749211803, "mr_stft_db": 6.1076971571738525, "band_mae_db": 0.22038657941157425, "rms_gain_error_db": -0.05114004960039136, "output_peak": 0.42624508579705}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0777596224278052, "全帯域_10ms_rms_envelope_nrmse": 0.09209756343077752, "全帯域_10ms_envelope_correlation": 0.9915240703054885, "全帯域_10ms_modulation_mae_db": 0.19667680964249434, "全帯域_40ms_log_rms_mae_db": 0.5683039516950028, "全帯域_40ms_rms_envelope_nrmse": 0.051638739591766306, "全帯域_40ms_envelope_correlation": 0.9969808631298764, "全帯域_40ms_modulation_mae_db": 0.17425729730045666, "300–1000 Hz_10ms_log_rms_mae_db": 1.964835359013992, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11751289038371672, "300–1000 Hz_10ms_envelope_correlation": 0.9902094148114877, "300–1000 Hz_10ms_modulation_mae_db": 0.329926332076138, "300–1000 Hz_40ms_log_rms_mae_db": 1.1999851603850307, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07105628449002027, "300–1000 Hz_40ms_envelope_correlation": 0.9961469452255596, "300–1000 Hz_40ms_modulation_mae_db": 0.19387004984244008, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9807337761871096, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16226948179600728, "1000–3000 Hz_10ms_envelope_correlation": 0.980307024556332, "1000–3000 Hz_10ms_modulation_mae_db": 0.22432846482520916, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2082651120635013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09358643982883277, "1000–3000 Hz_40ms_envelope_correlation": 0.9929147430251221, "1000–3000 Hz_40ms_modulation_mae_db": 0.3599719590571265, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9862377373860898, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1544367543532294, "3000–6000 Hz_10ms_envelope_correlation": 0.9824709212070442, "3000–6000 Hz_10ms_modulation_mae_db": 0.6974767819995075, "3000–6000 Hz_40ms_log_rms_mae_db": 1.152748999265676, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08488823986766703, "3000–6000 Hz_40ms_envelope_correlation": 0.9943346304836619, "3000–6000 Hz_40ms_modulation_mae_db": 1.1717485494569924, "mr_stft_db": 6.107605733198734, "band_mae_db": 0.11449877700392076, "rms_gain_error_db": -0.001616765725702773, "output_peak": 0.41639134075669143}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.319189612912897, "spectral_convergence": 0.42514612457985806, "spectral_convergence_status": "ok", "amplitude_floor": 9.383482434461407e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.053092352383682, "spectral_convergence": 0.4088750014870504, "spectral_convergence_status": "ok", "amplitude_floor": 6.715789372234631e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.940520518461629, "spectral_convergence": 0.4102827534229872, "spectral_convergence_status": "ok", "amplitude_floor": 5.217365851457555e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.0167397027219553e-05, "output_power": 3.473616319491067e-05, "delta_db": 0.612440074902874, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003363761740772986, "output_power": 0.00032567490299788147, "delta_db": -0.14040934160205581, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005972956425120644, "output_power": 0.0005921146056126753, "delta_db": -0.03783572524486642, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016539661163734624, "output_power": 0.0001586454112889505, "delta_db": -0.18099093281325646, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011856598981285755, "output_power": 0.00011194401789866361, "delta_db": -0.24959240666999272, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00046450684077542097, "output_power": 0.0004509743502072585, "delta_db": -0.12840272737531164, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22494520143472616, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09803392249114758, "pitch": null, "output_rms_dbfs": -27.762213401750127, "output_peak": 0.45305673321044904, "reference_rms_dbfs": -27.66417947925898, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.104267494586069}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014203698766261565, "log_rms_mae_db": 1.0364349292438881, "rms_envelope_nrmse": 0.08945763037384565, "envelope_correlation": 0.9920452972568216, "rms_gain_error_db": -0.09803224324070517, "reference_mean_power": 0.0017137270951261038, "output_mean_power": 0.0016754768783681922}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.499341088840717e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046355332033754323, "output_power": 0.0004557267435192725, "reference_active": true, "delta_db": -0.07395181050930351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011665881324257648, "output_power": 0.00010795298941132286, "reference_active": true, "delta_db": -0.3368288085832599}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5955876441113554e-05, "output_power": 2.0317598189561655e-05, "reference_active": true, "delta_db": -1.063633308407944}], "active_band_count": 3, "active_band_mae_db": 0.49147130916683573}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274207517780892, "log_rms_mae_db": 0.5495850893895352, "rms_envelope_nrmse": 0.04927283475948222, "envelope_correlation": 0.9973601251444124, "rms_gain_error_db": -0.09988522079616477, "reference_mean_power": 0.0017200719561396216, "output_mean_power": 0.0016809627636520904}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.335969034308142e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004202066244366408, "output_power": 0.0004134312773463135, "reference_active": true, "delta_db": -0.0705956540557909}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.868302162815529e-05, "output_power": 5.482937227550585e-05, "reference_active": true, "delta_db": -0.2949919405224856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5858058061398606e-06, "output_power": 1.1303060954872777e-06, "reference_active": true, "delta_db": -1.470539339486848}], "active_band_count": 3, "active_band_mae_db": 0.6120423113550415}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000125392198184175, "log_rms_mae_db": 1.9391231183366642, "rms_envelope_nrmse": 0.09584527400897677, "envelope_correlation": 0.9935571321854677, "rms_gain_error_db": -0.09339991428487032, "reference_mean_power": 0.0008371717774342565, "output_mean_power": 0.0008193596776605929}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.199403135704403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003464696140419235, "output_power": 0.0003424998409863482, "reference_active": true, "delta_db": -0.05004778076711212}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.952130707503092e-05, "output_power": 8.241871278937494e-05, "reference_active": true, "delta_db": -0.35900587120475513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1976615680578699e-05, "output_power": 1.1929260682181134e-05, "reference_active": true, "delta_db": -0.017205846029249704}], "active_band_count": 3, "active_band_mae_db": 0.14208649933370565}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011133215930497508, "log_rms_mae_db": 1.1564912352079675, "rms_envelope_nrmse": 0.05347777579481125, "envelope_correlation": 0.997951173416276, "rms_gain_error_db": -0.09363521806632318, "reference_mean_power": 0.0008413676993576814, "output_mean_power": 0.0008234217102072001}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.575111583882504e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003227285828337414, "output_power": 0.000318750049335961, "reference_active": true, "delta_db": -0.05387170243269834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.553081813418887e-05, "output_power": 4.299117632180616e-05, "reference_active": true, "delta_db": -0.24926125711906294}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.055134444571575e-06, "output_power": 8.598717081322214e-07, "reference_active": true, "delta_db": -0.8887414085033181}], "active_band_count": 3, "active_band_mae_db": 0.3972914560183598}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.277196005950351e-05, "log_rms_mae_db": 1.9467221092082105, "rms_envelope_nrmse": 0.1594332726121034, "envelope_correlation": 0.9811237318270019, "rms_gain_error_db": -0.1810552442721093, "reference_mean_power": 0.0001860033831933617, "output_mean_power": 0.00017840840854098779}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.45795784914854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.813551789415923e-05, "output_power": 8.404138721754101e-05, "reference_active": true, "delta_db": -0.2065774840240942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.039396685146323e-05, "output_power": 1.8044739750257468e-05, "reference_active": true, "delta_db": -0.5315108622016809}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.503164081364961e-06, "output_power": 4.963255889545496e-06, "reference_active": true, "delta_db": 0.42248895005968423}], "active_band_count": 3, "active_band_mae_db": 0.38685909876181973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5906277437135206e-05, "log_rms_mae_db": 1.1818347759658678, "rms_envelope_nrmse": 0.09042334163719318, "envelope_correlation": 0.9937055409579525, "rms_gain_error_db": -0.18150339253575581, "reference_mean_power": 0.00018692252747526107, "output_mean_power": 0.00017927152193188259}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315230721785866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.510469005068111e-05, "output_power": 8.197375998472688e-05, "reference_active": true, "delta_db": -0.16278638533994833}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648548828731175e-06, "output_power": 8.780829789679523e-06, "reference_active": true, "delta_db": -0.40926440299880784}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.651744211074769e-07, "output_power": 4.0768943668025257e-07, "reference_active": true, "delta_db": 0.4782910946273969}], "active_band_count": 3, "active_band_mae_db": 0.35011396098871767}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.827493119721784e-05, "log_rms_mae_db": 2.003145138183495, "rms_envelope_nrmse": 0.16195903200436693, "envelope_correlation": 0.9806877077408879, "rms_gain_error_db": -0.09894272409125064, "reference_mean_power": 0.0003389197953959056, "output_mean_power": 0.0003312856795935501}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5434076582307222e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.258901619859077e-05, "output_power": 5.785018508812958e-05, "reference_active": true, "delta_db": -0.3419337238854987}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4223152137238325e-05, "output_power": 1.416182667983639e-05, "reference_active": true, "delta_db": -0.018765805224315046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.429050160653041e-06, "output_power": 5.3633802085489514e-06, "reference_active": true, "delta_db": -1.9634005322789845}], "active_band_count": 3, "active_band_mae_db": 0.7747000204629327}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.761964716205697e-05, "log_rms_mae_db": 1.1635716182727776, "rms_envelope_nrmse": 0.09318333948209996, "envelope_correlation": 0.9932102285283563, "rms_gain_error_db": -0.10481775125179055, "reference_mean_power": 0.00033859490771126537, "output_mean_power": 0.0003305206872692919}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4129997856527214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.165966182107609e-05, "output_power": 5.6049067424641694e-05, "reference_active": true, "delta_db": -0.414327475035805}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.083964993529701e-06, "output_power": 6.527635870730763e-06, "reference_active": true, "delta_db": -0.3552048609735652}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.531889318612958e-07, "output_power": 3.261210301924393e-07, "reference_active": true, "delta_db": -3.016600115466093}], "active_band_count": 3, "active_band_mae_db": 1.2620441504918212}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8614454204729546, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0378aded09ad8b85532b89bffcc966ac6bbee9c9fd360f0ed7019b0ec34ad15f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.981456379166905, "standard_deviation": 0.0010817530377673118, "minimum": 0.9800794898220518, "maximum": 0.9833502501302485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9303053958102263, "standard_deviation": 0.031965661690392104, "minimum": 1.8830976752576345, "maximum": 1.9807337761871096, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6717691767645206, "standard_deviation": 0.28316113761418404, "minimum": 0.22432846482520916, "maximum": 0.9871037612490836, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15828328820529536, "standard_deviation": 0.004047728056489258, "minimum": 0.15182272826990947, "maximum": 0.16338514569602017, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9931898307786101, "standard_deviation": 0.0012295428588690025, "minimum": 0.9905302567041415, "maximum": 0.9941594751805753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1726258246695305, "standard_deviation": 0.03147110656637953, "minimum": 1.115462445575521, "maximum": 1.2119822073775632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7267990619802959, "standard_deviation": 0.3442804419680588, "minimum": 0.3401510759890051, "maximum": 1.1697306660104796, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09394440596410039, "standard_deviation": 0.006377135184375591, "minimum": 0.08686754698239307, "maximum": 0.10770275392528529, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9800107073169315, "standard_deviation": 0.001850487247325384, "minimum": 0.9774939241306623, "maximum": 0.9824709212070442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9753772230522573, "standard_deviation": 0.03082460065954427, "minimum": 1.9204609080747852, "maximum": 2.0115507421706615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.913207983209408, "standard_deviation": 0.23090365467969481, "minimum": 0.5748409178542252, "maximum": 1.228503208380459, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16517025958434237, "standard_deviation": 0.008027362760744951, "minimum": 0.1544367543532294, "maximum": 0.17532098184869882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937717485157209, "standard_deviation": 0.0012578794146598106, "minimum": 0.9926072869494409, "maximum": 0.9962311928055776, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.152553492288972, "standard_deviation": 0.037779522655890053, "minimum": 1.0946976082114799, "maximum": 1.2035439614471357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3836663876976687, "standard_deviation": 0.33740687968283484, "minimum": 0.7937506579879625, "maximum": 1.839138426190533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08975745186990358, "standard_deviation": 0.01016383707017605, "minimum": 0.07007368667959948, "maximum": 0.10105359496378674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9923482530426677, "standard_deviation": 0.0011851989590157092, "minimum": 0.9902094148114877, "maximum": 0.9935571321854677, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9834012556781415, "standard_deviation": 0.0418515280594008, "minimum": 1.9391231183366642, "maximum": 2.0370420725875906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2937638391374232, "standard_deviation": 0.09254056859816881, "minimum": 0.14208649933370565, "maximum": 0.40119927758791474, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10371531303805986, "standard_deviation": 0.0077554259060475545, "minimum": 0.09584527400897677, "maximum": 0.11751289038371672, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971419435068054, "standard_deviation": 0.0006146234089717691, "minimum": 0.9961469452255596, "maximum": 0.997951173416276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.193131698914268, "standard_deviation": 0.04374350965914959, "minimum": 1.1347545864744324, "maximum": 1.2606810712455319, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4107263679845711, "standard_deviation": 0.18857761324326633, "minimum": 0.14307163881693968, "maximum": 0.6728405920434494, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.061578145142211146, "standard_deviation": 0.006240515578061427, "minimum": 0.05347777579481125, "maximum": 0.07105628449002027, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18789655280693346, "standard_deviation": 0.05648142723100814, "minimum": 0.08875690209268873, "maximum": 0.24198330760349418, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.105533213317018, "standard_deviation": 0.0075056702641956895, "minimum": 6.0948956364504925, "maximum": 6.119473492242647, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.421298893830498, "standard_deviation": 0.026389856464400015, "minimum": 0.37820731654418516, "maximum": 0.45305673321044904, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05643312900718189, "standard_deviation": 0.051529896471509344, "minimum": -0.1353686383627127, "maximum": -0.001616765725702773, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920554368412771, "standard_deviation": 0.0003484509801678684, "minimum": 0.9915240703054885, "maximum": 0.9925759966568336, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0659144884445306, "standard_deviation": 0.029019558024176252, "minimum": 1.0357116344679416, "maximum": 1.1222259044880702, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45087989899342723, "standard_deviation": 0.1398473749411404, "minimum": 0.19667680964249434, "maximum": 0.6137420457767692, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08927629103012408, "standard_deviation": 0.00201200358208567, "minimum": 0.0865563378117362, "maximum": 0.09209756343077752, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973851314908051, "standard_deviation": 0.0002762671278132024, "minimum": 0.9969808631298764, "maximum": 0.9977908021373671, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5616503446127137, "standard_deviation": 0.018744312414378704, "minimum": 0.5310630775508631, "maximum": 0.5860419680540867, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5453717112937086, "standard_deviation": 0.2062126228742555, "minimum": 0.17425729730045666, "maximum": 0.8234982975552237, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04856447130648913, "standard_deviation": 0.002662123065102665, "minimum": 0.04434563368355743, "maximum": 0.051638739591766306, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2273-0010", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.5409226417541504, "predicted_mos": 2.5409226417541504, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "aae1ecfc7eb8844dd8a454870c85b4a42c7e5d76d2627ec80d799f0748e7af4a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.093623499851674, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0833029339573215, "全帯域_10ms_rms_envelope_nrmse": 0.09190652820602725, "全帯域_10ms_envelope_correlation": 0.9915736186699283, "全帯域_10ms_modulation_mae_db": 0.32562487145955804, "全帯域_40ms_log_rms_mae_db": 0.5679215168920871, "全帯域_40ms_rms_envelope_nrmse": 0.04745777543372253, "全帯域_40ms_envelope_correlation": 0.9974466109895902, "全帯域_40ms_modulation_mae_db": 0.308720409065345, "300–1000 Hz_10ms_log_rms_mae_db": 1.5102957509902444, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09515018828487473, "300–1000 Hz_10ms_envelope_correlation": 0.9935824028548804, "300–1000 Hz_10ms_modulation_mae_db": 0.14802910296738897, "300–1000 Hz_40ms_log_rms_mae_db": 0.9037620452076863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055175853352155364, "300–1000 Hz_40ms_envelope_correlation": 0.9976766385306699, "300–1000 Hz_40ms_modulation_mae_db": 0.2221269155338975, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5685749196706598, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15164748031478, "1000–3000 Hz_10ms_envelope_correlation": 0.9828833281301493, "1000–3000 Hz_10ms_modulation_mae_db": 0.15051068060352923, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9713519890192317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09294654364573067, "1000–3000 Hz_40ms_envelope_correlation": 0.9929370139273328, "1000–3000 Hz_40ms_modulation_mae_db": 0.17296740786558285, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4612269521491976, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16306597405974121, "3000–6000 Hz_10ms_envelope_correlation": 0.980454737257895, "3000–6000 Hz_10ms_modulation_mae_db": 0.4166143851723047, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9199490210198266, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08877206033360276, "3000–6000 Hz_40ms_envelope_correlation": 0.9938425427516795, "3000–6000 Hz_40ms_modulation_mae_db": 0.48815326994142266, "mr_stft_db": 5.694661265376766, "band_mae_db": 0.10070697471675927, "rms_gain_error_db": 0.017913922551487525, "output_peak": 0.39387814248094294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1322484649239144, "全帯域_10ms_rms_envelope_nrmse": 0.0907497826676003, "全帯域_10ms_envelope_correlation": 0.9917475546529432, "全帯域_10ms_modulation_mae_db": 0.22285863636328776, "全帯域_40ms_log_rms_mae_db": 0.5701747325741714, "全帯域_40ms_rms_envelope_nrmse": 0.04914164260905686, "全帯域_40ms_envelope_correlation": 0.9972630968178298, "全帯域_40ms_modulation_mae_db": 0.1828609526243612, "300–1000 Hz_10ms_log_rms_mae_db": 1.4833934730091016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09880850451134812, "300–1000 Hz_10ms_envelope_correlation": 0.9930876749352598, "300–1000 Hz_10ms_modulation_mae_db": 0.15819327534669972, "300–1000 Hz_40ms_log_rms_mae_db": 0.8139496546367477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05771566457744328, "300–1000 Hz_40ms_envelope_correlation": 0.9975540198168366, "300–1000 Hz_40ms_modulation_mae_db": 0.21916405855934765, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5342812751154187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15122648538399902, "1000–3000 Hz_10ms_envelope_correlation": 0.9828787753876329, "1000–3000 Hz_10ms_modulation_mae_db": 0.138297022115898, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9274891531014579, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08265281360967114, "1000–3000 Hz_40ms_envelope_correlation": 0.9945416175148399, "1000–3000 Hz_40ms_modulation_mae_db": 0.2700440639422838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4700235176102092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14912033795476012, "3000–6000 Hz_10ms_envelope_correlation": 0.9840708765683599, "3000–6000 Hz_10ms_modulation_mae_db": 0.2955779517524006, "3000–6000 Hz_40ms_log_rms_mae_db": 0.857676500117524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0873214093910432, "3000–6000 Hz_40ms_envelope_correlation": 0.9942248819679728, "3000–6000 Hz_40ms_modulation_mae_db": 0.40946788776340776, "mr_stft_db": 5.6870588639933475, "band_mae_db": 0.11872429926045636, "rms_gain_error_db": -0.03869834556635824, "output_peak": 0.4021152544756948}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1315874209964762, "全帯域_10ms_rms_envelope_nrmse": 0.08840142574944414, "全帯域_10ms_envelope_correlation": 0.9923240134203181, "全帯域_10ms_modulation_mae_db": 0.3864272045493145, "全帯域_40ms_log_rms_mae_db": 0.5686941361728053, "全帯域_40ms_rms_envelope_nrmse": 0.04732440862295115, "全帯域_40ms_envelope_correlation": 0.9977151034019439, "全帯域_40ms_modulation_mae_db": 0.4052606551810747, "300–1000 Hz_10ms_log_rms_mae_db": 1.474146717679734, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09321641366599981, "300–1000 Hz_10ms_envelope_correlation": 0.9938578486904197, "300–1000 Hz_10ms_modulation_mae_db": 0.18173278208401392, "300–1000 Hz_40ms_log_rms_mae_db": 0.8396856449165738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05621210131224352, "300–1000 Hz_40ms_envelope_correlation": 0.9976767859933863, "300–1000 Hz_40ms_modulation_mae_db": 0.17477451411053171, "1000–3000 Hz_10ms_log_rms_mae_db": 1.539513305992058, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1536378761577102, "1000–3000 Hz_10ms_envelope_correlation": 0.9832819986012032, "1000–3000 Hz_10ms_modulation_mae_db": 0.6941110288847722, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9299706471526642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09763831101700736, "1000–3000 Hz_40ms_envelope_correlation": 0.9937509796103483, "1000–3000 Hz_40ms_modulation_mae_db": 0.9051325166911841, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4929183061591809, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14350659256681805, "3000–6000 Hz_10ms_envelope_correlation": 0.9850291490588473, "3000–6000 Hz_10ms_modulation_mae_db": 0.6344468151514161, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8946779901530806, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08359102978928115, "3000–6000 Hz_40ms_envelope_correlation": 0.9948790732713765, "3000–6000 Hz_40ms_modulation_mae_db": 1.022022403223412, "mr_stft_db": 5.704148950251941, "band_mae_db": 0.16954851035303162, "rms_gain_error_db": -0.14200692226903855, "output_peak": 0.45722706625644494}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.126795858581949, "全帯域_10ms_rms_envelope_nrmse": 0.08575392434979139, "全帯域_10ms_envelope_correlation": 0.9927488541337645, "全帯域_10ms_modulation_mae_db": 0.4898533745149507, "全帯域_40ms_log_rms_mae_db": 0.5697092996107856, "全帯域_40ms_rms_envelope_nrmse": 0.048226555393937325, "全帯域_40ms_envelope_correlation": 0.9975418811652529, "全帯域_40ms_modulation_mae_db": 0.5277376375994777, "300–1000 Hz_10ms_log_rms_mae_db": 1.457084729897191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0850597288151791, "300–1000 Hz_10ms_envelope_correlation": 0.9949191238432, "300–1000 Hz_10ms_modulation_mae_db": 0.2171122001559802, "300–1000 Hz_40ms_log_rms_mae_db": 0.8321968231217682, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05156018383722788, "300–1000 Hz_40ms_envelope_correlation": 0.9980792956425024, "300–1000 Hz_40ms_modulation_mae_db": 0.29930345201904757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5284863518299, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15232797465830872, "1000–3000 Hz_10ms_envelope_correlation": 0.9828499947370986, "1000–3000 Hz_10ms_modulation_mae_db": 0.46028852788254154, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9768564455862464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08878284865805766, "1000–3000 Hz_40ms_envelope_correlation": 0.9940591585962097, "1000–3000 Hz_40ms_modulation_mae_db": 0.6998344165394627, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4959468082666236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15853476465762756, "3000–6000 Hz_10ms_envelope_correlation": 0.9816013887806416, "3000–6000 Hz_10ms_modulation_mae_db": 0.7557095633319589, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9061875869847461, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0864376155942067, "3000–6000 Hz_40ms_envelope_correlation": 0.9941836016237531, "3000–6000 Hz_40ms_modulation_mae_db": 1.022711936217903, "mr_stft_db": 5.692062420087418, "band_mae_db": 0.16829842096901063, "rms_gain_error_db": -0.11900050895672404, "output_peak": 0.40026419932232493}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1825656112407499, "全帯域_10ms_rms_envelope_nrmse": 0.09299356707621478, "全帯域_10ms_envelope_correlation": 0.9913546228455544, "全帯域_10ms_modulation_mae_db": 0.3418612103356547, "全帯域_40ms_log_rms_mae_db": 0.630873814566207, "全帯域_40ms_rms_envelope_nrmse": 0.047920439750788024, "全帯域_40ms_envelope_correlation": 0.9974511539062522, "全帯域_40ms_modulation_mae_db": 0.40451294445705416, "300–1000 Hz_10ms_log_rms_mae_db": 1.4839801373448371, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09321800740704138, "300–1000 Hz_10ms_envelope_correlation": 0.9938383048173672, "300–1000 Hz_10ms_modulation_mae_db": 0.26420363829290133, "300–1000 Hz_40ms_log_rms_mae_db": 0.8765246948943916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051760850918907225, "300–1000 Hz_40ms_envelope_correlation": 0.9980066901257029, "300–1000 Hz_40ms_modulation_mae_db": 0.4155587721900904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6219331750806778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15042866419429687, "1000–3000 Hz_10ms_envelope_correlation": 0.9830904252786543, "1000–3000 Hz_10ms_modulation_mae_db": 0.4241525602201457, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0106091060908475, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0903143917097385, "1000–3000 Hz_40ms_envelope_correlation": 0.9934944792600531, "1000–3000 Hz_40ms_modulation_mae_db": 0.3329651824627128, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4863532810981772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16234588133274605, "3000–6000 Hz_10ms_envelope_correlation": 0.9806250186082773, "3000–6000 Hz_10ms_modulation_mae_db": 0.499096624035886, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8726977498076193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08895677238838812, "3000–6000 Hz_40ms_envelope_correlation": 0.9938467436686008, "3000–6000 Hz_40ms_modulation_mae_db": 0.7783381609720449, "mr_stft_db": 5.693363016643608, "band_mae_db": 0.14115510323309718, "rms_gain_error_db": -0.07742868915102946, "output_peak": 0.4127916841886823}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.093243039011282, "全帯域_10ms_rms_envelope_nrmse": 0.0877501120673779, "全帯域_10ms_envelope_correlation": 0.9923445731976845, "全帯域_10ms_modulation_mae_db": 0.340871680323191, "全帯域_40ms_log_rms_mae_db": 0.5795000779045213, "全帯域_40ms_rms_envelope_nrmse": 0.04663311363423639, "全帯域_40ms_envelope_correlation": 0.997555131284779, "全帯域_40ms_modulation_mae_db": 0.5520727885845198, "300–1000 Hz_10ms_log_rms_mae_db": 1.5321120159614607, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08992019266421944, "300–1000 Hz_10ms_envelope_correlation": 0.9942367336547004, "300–1000 Hz_10ms_modulation_mae_db": 0.10062566424658015, "300–1000 Hz_40ms_log_rms_mae_db": 0.8737027493846755, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0541073063959847, "300–1000 Hz_40ms_envelope_correlation": 0.9977540408188674, "300–1000 Hz_40ms_modulation_mae_db": 0.07554159250350066, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5526276533947743, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14333944840104365, "1000–3000 Hz_10ms_envelope_correlation": 0.9846606384445138, "1000–3000 Hz_10ms_modulation_mae_db": 0.1654187738125209, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9500256770010451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09114336443310689, "1000–3000 Hz_40ms_envelope_correlation": 0.9931939041729722, "1000–3000 Hz_40ms_modulation_mae_db": 0.5761653596372782, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4757417045394512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.153599574014702, "3000–6000 Hz_10ms_envelope_correlation": 0.9832700421931234, "3000–6000 Hz_10ms_modulation_mae_db": 0.4284020781309365, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8956687550217397, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08301266233488068, "3000–6000 Hz_40ms_envelope_correlation": 0.9949140637199791, "3000–6000 Hz_40ms_modulation_mae_db": 0.6416975690632408, "mr_stft_db": 5.70341522064975, "band_mae_db": 0.14210011232616362, "rms_gain_error_db": 0.03243018200638419, "output_peak": 0.4192669965554489}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0501059622912108, "全帯域_10ms_rms_envelope_nrmse": 0.09315220548754771, "全帯域_10ms_envelope_correlation": 0.9914313376296656, "全帯域_10ms_modulation_mae_db": 0.579171803369077, "全帯域_40ms_log_rms_mae_db": 0.5547137642168081, "全帯域_40ms_rms_envelope_nrmse": 0.05207404849809411, "全帯域_40ms_envelope_correlation": 0.9971274232463482, "全帯域_40ms_modulation_mae_db": 0.7426387620238173, "300–1000 Hz_10ms_log_rms_mae_db": 1.5050565922485204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10197428907827127, "300–1000 Hz_10ms_envelope_correlation": 0.9927674818789258, "300–1000 Hz_10ms_modulation_mae_db": 0.3860222037155287, "300–1000 Hz_40ms_log_rms_mae_db": 0.8423022357901038, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05724788162862841, "300–1000 Hz_40ms_envelope_correlation": 0.9977837270674341, "300–1000 Hz_40ms_modulation_mae_db": 0.5080441763471012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5584296778772344, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15329944462119888, "1000–3000 Hz_10ms_envelope_correlation": 0.9827752942947051, "1000–3000 Hz_10ms_modulation_mae_db": 0.5453739586129455, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9731792891163452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.089124678748766, "1000–3000 Hz_40ms_envelope_correlation": 0.9942728934138929, "1000–3000 Hz_40ms_modulation_mae_db": 0.7319962757095574, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4751061843124629, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1453376176000116, "3000–6000 Hz_10ms_envelope_correlation": 0.9845506660543936, "3000–6000 Hz_10ms_modulation_mae_db": 0.5133572287866901, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9050020642628018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08034688746509809, "3000–6000 Hz_40ms_envelope_correlation": 0.9949240960830656, "3000–6000 Hz_40ms_modulation_mae_db": 0.6617737686445028, "mr_stft_db": 5.70400722231988, "band_mae_db": 0.12889335467842042, "rms_gain_error_db": -0.12109107110052843, "output_peak": 0.38756501346959576}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1143122322117982, "全帯域_10ms_rms_envelope_nrmse": 0.09489714637740576, "全帯域_10ms_envelope_correlation": 0.99102193029481, "全帯域_10ms_modulation_mae_db": 0.30639082516318045, "全帯域_40ms_log_rms_mae_db": 0.5796486591216241, "全帯域_40ms_rms_envelope_nrmse": 0.050880867800729646, "全帯域_40ms_envelope_correlation": 0.9971751574612238, "全帯域_40ms_modulation_mae_db": 0.32725027274562063, "300–1000 Hz_10ms_log_rms_mae_db": 1.5301255996298297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09631318044047, "300–1000 Hz_10ms_envelope_correlation": 0.9935154125876017, "300–1000 Hz_10ms_modulation_mae_db": 0.243188745612443, "300–1000 Hz_40ms_log_rms_mae_db": 0.8627792068475583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05921229695837952, "300–1000 Hz_40ms_envelope_correlation": 0.9975129879176916, "300–1000 Hz_40ms_modulation_mae_db": 0.3850783625325576, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5852168502118396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1425546468319646, "1000–3000 Hz_10ms_envelope_correlation": 0.9849617261728595, "1000–3000 Hz_10ms_modulation_mae_db": 0.4631523830473265, "1000–3000 Hz_40ms_log_rms_mae_db": 0.990151828241795, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08318584043774098, "1000–3000 Hz_40ms_envelope_correlation": 0.9947275180444061, "1000–3000 Hz_40ms_modulation_mae_db": 0.4554021750962165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5081379435381816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16229512023418302, "3000–6000 Hz_10ms_envelope_correlation": 0.9806134280143347, "3000–6000 Hz_10ms_modulation_mae_db": 0.28223745622111535, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8741411258678927, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07880367376447768, "3000–6000 Hz_40ms_envelope_correlation": 0.9952087230975022, "3000–6000 Hz_40ms_modulation_mae_db": 0.46671645293402536, "mr_stft_db": 5.699503311032982, "band_mae_db": 0.2002489409698142, "rms_gain_error_db": -0.1011658278633964, "output_peak": 0.4101805454787015}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.788353501564455, "spectral_convergence": 0.411482603196218, "spectral_convergence_status": "ok", "amplitude_floor": 9.383482434461407e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.522881311233252, "spectral_convergence": 0.3922580238938341, "spectral_convergence_status": "ok", "amplitude_floor": 6.715789372234631e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.772748983332589, "spectral_convergence": 0.41263994482193206, "spectral_convergence_status": "ok", "amplitude_floor": 5.217365851457555e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.0167397027219553e-05, "output_power": 3.241684416845642e-05, "delta_db": 0.3123289431349929, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003363761740772986, "output_power": 0.000337143649963789, "delta_db": 0.009897576132514335, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005972956425120644, "output_power": 0.0006075634894405618, "delta_db": 0.0740232131709414, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016539661163734624, "output_power": 0.00016018643560192008, "delta_db": -0.13900870464090417, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011856598981285755, "output_power": 0.0001198680831555692, "delta_db": 0.04743429227460912, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00046450684077542097, "output_power": 0.0004622077278084803, "delta_db": -0.021549118946593615, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10070697471675927, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.017913922551487525, "pitch": null, "output_rms_dbfs": -27.64626555670749, "output_peak": 0.39387814248094294, "reference_rms_dbfs": -27.66417947925898, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.694661265376766}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014203698766261565, "log_rms_mae_db": 1.0833029339573215, "rms_envelope_nrmse": 0.09190652820602725, "envelope_correlation": 0.9915736186699283, "rms_gain_error_db": 0.017921209802039578, "reference_mean_power": 0.0017137270951261038, "output_mean_power": 0.001720813419770381}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.499341088840717e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046355332033754323, "output_power": 0.0004612614895599369, "reference_active": true, "delta_db": -0.021524990453058935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011665881324257648, "output_power": 0.0001227238957705498, "reference_active": true, "delta_db": 0.22011579063948114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5955876441113554e-05, "output_power": 2.1913528591204328e-05, "reference_active": true, "delta_db": -0.735233833286134}], "active_band_count": 3, "active_band_mae_db": 0.32562487145955804}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274207517780892, "log_rms_mae_db": 0.5679215168920871, "rms_envelope_nrmse": 0.04745777543372253, "envelope_correlation": 0.9974466109895902, "rms_gain_error_db": 0.017562538299665455, "reference_mean_power": 0.0017200719561396216, "output_mean_power": 0.0017270418796344262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.335969034308142e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004202066244366408, "output_power": 0.00042147134564809026, "reference_active": true, "delta_db": 0.013051591610733647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.868302162815529e-05, "output_power": 6.251019646016236e-05, "reference_active": true, "delta_db": 0.27438395966065554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5858058061398606e-06, "output_power": 1.3689180060434033e-06, "reference_active": true, "delta_db": -0.6387256759246457}], "active_band_count": 3, "active_band_mae_db": 0.308720409065345}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000125392198184175, "log_rms_mae_db": 1.5102957509902444, "rms_envelope_nrmse": 0.09515018828487473, "envelope_correlation": 0.9935824028548804, "rms_gain_error_db": 0.027597385092735533, "reference_mean_power": 0.0008371717774342565, "output_mean_power": 0.0008425085513069365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.199403135704403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003464696140419235, "output_power": 0.00033879725434261473, "reference_active": true, "delta_db": -0.0972527018022929}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.952130707503092e-05, "output_power": 9.231627781784088e-05, "reference_active": true, "delta_db": 0.1335187088270947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1976615680578699e-05, "output_power": 1.1402564139014725e-05, "reference_active": true, "delta_db": -0.2133158982727793}], "active_band_count": 3, "active_band_mae_db": 0.14802910296738897}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011133215930497508, "log_rms_mae_db": 0.9037620452076863, "rms_envelope_nrmse": 0.055175853352155364, "envelope_correlation": 0.9976766385306699, "rms_gain_error_db": 0.02759634596280754, "reference_mean_power": 0.0008413676993576814, "output_mean_power": 0.0008467310186531306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.575111583882504e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003227285828337414, "output_power": 0.0003182843152924953, "reference_active": true, "delta_db": -0.06022193320139009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.553081813418887e-05, "output_power": 4.670015874958283e-05, "reference_active": true, "delta_db": 0.11012902752753974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.055134444571575e-06, "output_power": 9.412496378320482e-07, "reference_active": true, "delta_db": -0.4960297858727626}], "active_band_count": 3, "active_band_mae_db": 0.2221269155338975}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.277196005950351e-05, "log_rms_mae_db": 1.5685749196706598, "rms_envelope_nrmse": 0.15164748031478, "envelope_correlation": 0.9828833281301493, "rms_gain_error_db": 0.007811209090146046, "reference_mean_power": 0.0001860033831933617, "output_mean_power": 0.00018633822942400177}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.45795784914854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.813551789415923e-05, "output_power": 8.73292246802701e-05, "reference_active": true, "delta_db": -0.039913563622162314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.039396685146323e-05, "output_power": 1.912928919012039e-05, "reference_active": true, "delta_db": -0.27802876333555543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.503164081364961e-06, "output_power": 4.366754737764096e-06, "reference_active": true, "delta_db": -0.13358971485287}], "active_band_count": 3, "active_band_mae_db": 0.15051068060352923}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5906277437135206e-05, "log_rms_mae_db": 0.9713519890192317, "rms_envelope_nrmse": 0.09294654364573067, "envelope_correlation": 0.9929370139273328, "rms_gain_error_db": 0.007538537654967966, "reference_mean_power": 0.00018692252747526107, "output_mean_power": 0.00018724727169026472}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315230721785866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.510469005068111e-05, "output_power": 8.414377338410967e-05, "reference_active": true, "delta_db": -0.04931510532285055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648548828731175e-06, "output_power": 9.308747945180479e-06, "reference_active": true, "delta_db": -0.15570728051397642}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.651744211074769e-07, "output_power": 3.3971315574206664e-07, "reference_active": true, "delta_db": -0.31387983775992156}], "active_band_count": 3, "active_band_mae_db": 0.17296740786558285}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.827493119721784e-05, "log_rms_mae_db": 1.4612269521491976, "rms_envelope_nrmse": 0.16306597405974121, "envelope_correlation": 0.980454737257895, "rms_gain_error_db": -0.08901472902497139, "reference_mean_power": 0.0003389197953959056, "output_mean_power": 0.0003320438667073067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5434076582307222e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.258901619859077e-05, "output_power": 6.361179644348634e-05, "reference_active": true, "delta_db": 0.07039535365185998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4223152137238325e-05, "output_power": 1.3780914346453139e-05, "reference_active": true, "delta_db": -0.13717822048793438}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.429050160653041e-06, "output_power": 6.630582455603491e-06, "reference_active": true, "delta_db": -1.0422695813771197}], "active_band_count": 3, "active_band_mae_db": 0.4166143851723047}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.761964716205697e-05, "log_rms_mae_db": 0.9199490210198266, "rms_envelope_nrmse": 0.08877206033360276, "envelope_correlation": 0.9938425427516795, "rms_gain_error_db": -0.09049412443441246, "reference_mean_power": 0.00033859490771126537, "output_mean_power": 0.00033161258939519564}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4129997856527214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.165966182107609e-05, "output_power": 6.238897241329388e-05, "reference_active": true, "delta_db": 0.05106694115317973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.083964993529701e-06, "output_power": 6.707036980651366e-06, "reference_active": true, "delta_db": -0.23745705471614528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.531889318612958e-07, "output_power": 4.982476232635396e-07, "reference_active": true, "delta_db": -1.175935813954943}], "active_band_count": 3, "active_band_mae_db": 0.48815326994142266}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.886032757299161, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "68c21703f367350e3893fc5e985ec7533a4b442d50cfa6807d449e9d9f1330e8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9834227726308521, "standard_deviation": 0.0008754752136775947, "minimum": 0.9827752942947051, "maximum": 0.9849617261728595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5611329011465704, "standard_deviation": 0.030859020689870106, "minimum": 1.5284863518299, "maximum": 1.6219331750806778, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.38016311689745996, "standard_deviation": 0.20648585795732122, "minimum": 0.138297022115898, "maximum": 0.6941110288847722, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14980775257041273, "standard_deviation": 0.004366147588285139, "minimum": 0.1425546468319646, "maximum": 0.1536378761577102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938721955675068, "standard_deviation": 0.000640002573037824, "minimum": 0.9929370139273328, "maximum": 0.9947275180444061, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9662042669137041, "standard_deviation": 0.02880321493472134, "minimum": 0.9274891531014579, "maximum": 1.0106091060908475, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5180634247430348, "standard_deviation": 0.253771249823988, "minimum": 0.17296740786558285, "maximum": 0.9051325166911841, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0894735990324774, "standard_deviation": 0.004911747118172364, "minimum": 0.08265281360967114, "maximum": 0.09763831101700736, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.982526913316984, "standard_deviation": 0.0019169157024418542, "minimum": 0.980454737257895, "maximum": 0.9850291490588473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4831818372091856, "standard_deviation": 0.015428816775641964, "minimum": 1.4612269521491976, "maximum": 1.5081379435381816, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.47818026282283854, "standard_deviation": 0.16085577992541075, "minimum": 0.28223745622111535, "maximum": 0.7557095633319589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15472573280257368, "standard_deviation": 0.007983242603942262, "minimum": 0.14350659256681805, "maximum": 0.16306597405974121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945029657729912, "standard_deviation": 0.0005387833637431649, "minimum": 0.9938425427516795, "maximum": 0.9952087230975022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.890750099154404, "standard_deviation": 0.020802614345068242, "minimum": 0.857676500117524, "maximum": 0.9199490210198266, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.686360181094995, "standard_deviation": 0.2393621013585039, "minimum": 0.40946788776340776, "maximum": 1.022711936217903, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08465526388262229, "standard_deviation": 0.0038241721093919896, "minimum": 0.07880367376447768, "maximum": 0.08895677238838812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937256229077944, "standard_deviation": 0.000666072900116507, "minimum": 0.9927674818789258, "maximum": 0.9949191238432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4970243770951148, "standard_deviation": 0.026838127556914133, "minimum": 1.457084729897191, "maximum": 1.5321120159614607, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.21238845155269198, "standard_deviation": 0.08808632045039642, "minimum": 0.10062566424658015, "maximum": 0.3860222037155287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09420756310842548, "standard_deviation": 0.005221369223023997, "minimum": 0.0850597288151791, "maximum": 0.10197428907827127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977555232391364, "standard_deviation": 0.00020020003925875137, "minimum": 0.9975129879176916, "maximum": 0.9980792956425024, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8556128818499381, "standard_deviation": 0.0289068253465062, "minimum": 0.8139496546367477, "maximum": 0.9037620452076863, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.2874489804745093, "standard_deviation": 0.1420646045633974, "minimum": 0.07554159250350066, "maximum": 0.5080441763471012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05537401737262124, "standard_deviation": 0.0027690528934061747, "minimum": 0.05156018383722788, "maximum": 0.05921229695837952, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14620946456334416, "standard_deviation": 0.031895118219224826, "minimum": 0.10070697471675927, "maximum": 0.2002489409698142, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.697277533794461, "standard_deviation": 0.006424476459711468, "minimum": 5.6870588639933475, "maximum": 5.704148950251941, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4104111127784795, "standard_deviation": 0.021523890172809618, "minimum": 0.38756501346959576, "maximum": 0.45722706625644494, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06863090754365042, "standard_deviation": 0.0659095015360249, "minimum": -0.14200692226903855, "maximum": 0.03243018200638419, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918183131055835, "standard_deviation": 0.0005931099007956476, "minimum": 0.99102193029481, "maximum": 0.9927488541337645, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1142701904018377, "standard_deviation": 0.0395734826530334, "minimum": 1.0501059622912108, "maximum": 1.1825656112407499, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37413245075977675, "standard_deviation": 0.11183982238808175, "minimum": 0.22285863636328776, "maximum": 0.579171803369077, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09070058649767615, "standard_deviation": 0.0031349853499693446, "minimum": 0.08575392434979139, "maximum": 0.09489714637740576, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974094447841526, "standard_deviation": 0.00020394556155444964, "minimum": 0.9971274232463482, "maximum": 0.9977151034019439, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5776545001323762, "standard_deviation": 0.022866219125410422, "minimum": 0.5547137642168081, "maximum": 0.630873814566207, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43138180278515875, "standard_deviation": 0.17311751234052383, "minimum": 0.1828609526243612, "maximum": 0.7426387620238173, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0487073564679395, "standard_deviation": 0.0018850638344907913, "minimum": 0.04663311363423639, "maximum": 0.05207404849809411, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2273-0010", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.308213472366333, "predicted_mos": 2.308213472366333, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "92be33756e4a12b6a69c5f338f5a0f9b906bab45fb168ece4c39e968981fd80d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.085384499747306, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0902794596115246, "全帯域_10ms_rms_envelope_nrmse": 0.09035964970073625, "全帯域_10ms_envelope_correlation": 0.9918183917741347, "全帯域_10ms_modulation_mae_db": 0.394315611565629, "全帯域_40ms_log_rms_mae_db": 0.6033594101086622, "全帯域_40ms_rms_envelope_nrmse": 0.04955287601237036, "全帯域_40ms_envelope_correlation": 0.9972126366963167, "全帯域_40ms_modulation_mae_db": 0.4281195288712593, "300–1000 Hz_10ms_log_rms_mae_db": 1.5587071222401887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09311038296733562, "300–1000 Hz_10ms_envelope_correlation": 0.9938261349500163, "300–1000 Hz_10ms_modulation_mae_db": 0.2553831068044177, "300–1000 Hz_40ms_log_rms_mae_db": 0.8479043043084236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057002702355681556, "300–1000 Hz_40ms_envelope_correlation": 0.9975027322889456, "300–1000 Hz_40ms_modulation_mae_db": 0.21887627161163128, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6470804176138145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13838277630468518, "1000–3000 Hz_10ms_envelope_correlation": 0.9857976267506949, "1000–3000 Hz_10ms_modulation_mae_db": 0.5504853478630422, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9999927016687438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07913605544270376, "1000–3000 Hz_40ms_envelope_correlation": 0.995162647711932, "1000–3000 Hz_40ms_modulation_mae_db": 0.7188822428399252, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7293961974183079, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18125721271315498, "3000–6000 Hz_10ms_envelope_correlation": 0.9758083564308918, "3000–6000 Hz_10ms_modulation_mae_db": 0.43161596296543364, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0253317205243873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10089737296490923, "3000–6000 Hz_40ms_envelope_correlation": 0.992147238158374, "3000–6000 Hz_40ms_modulation_mae_db": 0.9173331522394995, "mr_stft_db": 5.836760640774211, "band_mae_db": 0.14031638335712224, "rms_gain_error_db": -0.03438707506692868, "output_peak": 0.4581293407352761}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0787523018950285, "全帯域_10ms_rms_envelope_nrmse": 0.09147386511176707, "全帯域_10ms_envelope_correlation": 0.9916274218471518, "全帯域_10ms_modulation_mae_db": 0.3777224822288961, "全帯域_40ms_log_rms_mae_db": 0.5563431658870884, "全帯域_40ms_rms_envelope_nrmse": 0.04515164667094036, "全帯域_40ms_envelope_correlation": 0.9977200355331932, "全帯域_40ms_modulation_mae_db": 0.45012921934736977, "300–1000 Hz_10ms_log_rms_mae_db": 1.6596602249204981, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09532874873670157, "300–1000 Hz_10ms_envelope_correlation": 0.9935343680547009, "300–1000 Hz_10ms_modulation_mae_db": 0.13985981130336236, "300–1000 Hz_40ms_log_rms_mae_db": 0.8520335698395026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04972190486948176, "300–1000 Hz_40ms_envelope_correlation": 0.9981355050292429, "300–1000 Hz_40ms_modulation_mae_db": 0.3473256974549095, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6404751981325763, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1536720017393464, "1000–3000 Hz_10ms_envelope_correlation": 0.9827144741997853, "1000–3000 Hz_10ms_modulation_mae_db": 0.5061539446001043, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0041450699970527, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09055915657371451, "1000–3000 Hz_40ms_envelope_correlation": 0.994111670807954, "1000–3000 Hz_40ms_modulation_mae_db": 0.5637021646967599, "3000–6000 Hz_10ms_log_rms_mae_db": 1.645135991784917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15396419145492457, "3000–6000 Hz_10ms_envelope_correlation": 0.9825588854283593, "3000–6000 Hz_10ms_modulation_mae_db": 0.8981460668552442, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9822337881167186, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08469726935189158, "3000–6000 Hz_40ms_envelope_correlation": 0.9944099971456185, "3000–6000 Hz_40ms_modulation_mae_db": 1.0730517432801456, "mr_stft_db": 5.801617005766915, "band_mae_db": 0.1100617358241938, "rms_gain_error_db": -0.05900794656742924, "output_peak": 0.4419147876761002}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0668678758295222, "全帯域_10ms_rms_envelope_nrmse": 0.08897533267164985, "全帯域_10ms_envelope_correlation": 0.9920773494089143, "全帯域_10ms_modulation_mae_db": 0.21620467900900045, "全帯域_40ms_log_rms_mae_db": 0.5178395561872985, "全帯域_40ms_rms_envelope_nrmse": 0.04568791103799448, "全帯域_40ms_envelope_correlation": 0.9976261624336729, "全帯域_40ms_modulation_mae_db": 0.3162565769843804, "300–1000 Hz_10ms_log_rms_mae_db": 1.5671337456969654, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08843689448491862, "300–1000 Hz_10ms_envelope_correlation": 0.9944354226644762, "300–1000 Hz_10ms_modulation_mae_db": 0.1816031567862171, "300–1000 Hz_40ms_log_rms_mae_db": 0.8622653420966918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051602521634157114, "300–1000 Hz_40ms_envelope_correlation": 0.9979878151398621, "300–1000 Hz_40ms_modulation_mae_db": 0.2825689035930132, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6525566825978488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14068400294044983, "1000–3000 Hz_10ms_envelope_correlation": 0.9853617042533777, "1000–3000 Hz_10ms_modulation_mae_db": 0.42570630975841967, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0391343127776962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07644421840400956, "1000–3000 Hz_40ms_envelope_correlation": 0.995601469917732, "1000–3000 Hz_40ms_modulation_mae_db": 0.3431808740044639, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6680573384987363, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15897452299087636, "3000–6000 Hz_10ms_envelope_correlation": 0.9817185016393309, "3000–6000 Hz_10ms_modulation_mae_db": 0.6115449419537848, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9800503603921221, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08593294973819955, "3000–6000 Hz_40ms_envelope_correlation": 0.9943146572489224, "3000–6000 Hz_40ms_modulation_mae_db": 1.0183972166040978, "mr_stft_db": 5.825217120877881, "band_mae_db": 0.15222455191486242, "rms_gain_error_db": -0.008109925645046806, "output_peak": 0.40000210473451775}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0825567988959857, "全帯域_10ms_rms_envelope_nrmse": 0.08788897379629201, "全帯域_10ms_envelope_correlation": 0.9922806380644934, "全帯域_10ms_modulation_mae_db": 0.2721561627739132, "全帯域_40ms_log_rms_mae_db": 0.5579122588366314, "全帯域_40ms_rms_envelope_nrmse": 0.04617707605937144, "全帯域_40ms_envelope_correlation": 0.9975864111168329, "全帯域_40ms_modulation_mae_db": 0.3496162224394748, "300–1000 Hz_10ms_log_rms_mae_db": 1.6391802719870578, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09310384997517766, "300–1000 Hz_10ms_envelope_correlation": 0.9938165505170292, "300–1000 Hz_10ms_modulation_mae_db": 0.29298937713641493, "300–1000 Hz_40ms_log_rms_mae_db": 0.9029882247904938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05427337781579182, "300–1000 Hz_40ms_envelope_correlation": 0.9977394792747803, "300–1000 Hz_40ms_modulation_mae_db": 0.5169178347665189, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6875115683206945, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1464559477787127, "1000–3000 Hz_10ms_envelope_correlation": 0.9840644765687536, "1000–3000 Hz_10ms_modulation_mae_db": 0.3550749055000109, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0483081442934874, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08760713449665074, "1000–3000 Hz_40ms_envelope_correlation": 0.9937360185497469, "1000–3000 Hz_40ms_modulation_mae_db": 0.18789567232066715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.661141547775573, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15067648773414197, "3000–6000 Hz_10ms_envelope_correlation": 0.9833769119726495, "3000–6000 Hz_10ms_modulation_mae_db": 0.15035142677219449, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9552134131030084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0719179000329486, "3000–6000 Hz_40ms_envelope_correlation": 0.9959616739014053, "3000–6000 Hz_40ms_modulation_mae_db": 0.6147617682879429, "mr_stft_db": 5.813379535426125, "band_mae_db": 0.129305420044633, "rms_gain_error_db": -0.004601190127547735, "output_peak": 0.4319331819620943}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0445387578968706, "全帯域_10ms_rms_envelope_nrmse": 0.08901586565785556, "全帯域_10ms_envelope_correlation": 0.9920885932092047, "全帯域_10ms_modulation_mae_db": 0.21607188678799474, "全帯域_40ms_log_rms_mae_db": 0.538832574501891, "全帯域_40ms_rms_envelope_nrmse": 0.05042287086544763, "全帯域_40ms_envelope_correlation": 0.9971749630718825, "全帯域_40ms_modulation_mae_db": 0.2894471776619067, "300–1000 Hz_10ms_log_rms_mae_db": 1.6159348414624999, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09461374038720499, "300–1000 Hz_10ms_envelope_correlation": 0.9936216805744872, "300–1000 Hz_10ms_modulation_mae_db": 0.49336124414436755, "300–1000 Hz_40ms_log_rms_mae_db": 0.9084800754995604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05631310518950799, "300–1000 Hz_40ms_envelope_correlation": 0.9975867799995491, "300–1000 Hz_40ms_modulation_mae_db": 0.15933710726209552, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6825023372780024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15327850899897164, "1000–3000 Hz_10ms_envelope_correlation": 0.9832460068953808, "1000–3000 Hz_10ms_modulation_mae_db": 0.37306125856962197, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0529629246681391, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09308746688448544, "1000–3000 Hz_40ms_envelope_correlation": 0.9943501515444648, "1000–3000 Hz_40ms_modulation_mae_db": 0.5180928990779208, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6598362033549228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16109846163685862, "3000–6000 Hz_10ms_envelope_correlation": 0.9808964875325096, "3000–6000 Hz_10ms_modulation_mae_db": 0.345470781980128, "3000–6000 Hz_40ms_log_rms_mae_db": 1.000924044000444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09873024958343993, "3000–6000 Hz_40ms_envelope_correlation": 0.9923600062428131, "3000–6000 Hz_40ms_modulation_mae_db": 0.45300147514967093, "mr_stft_db": 5.802096435427803, "band_mae_db": 0.15881112071206058, "rms_gain_error_db": -0.07920383657774535, "output_peak": 0.41240659610380265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0698375094908341, "全帯域_10ms_rms_envelope_nrmse": 0.09305310996806322, "全帯域_10ms_envelope_correlation": 0.9913586440217722, "全帯域_10ms_modulation_mae_db": 0.38228570653973754, "全帯域_40ms_log_rms_mae_db": 0.5424594086829353, "全帯域_40ms_rms_envelope_nrmse": 0.04892428047223574, "全帯域_40ms_envelope_correlation": 0.9973593046648745, "全帯域_40ms_modulation_mae_db": 0.1950441330212579, "300–1000 Hz_10ms_log_rms_mae_db": 1.6232653881966852, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09601283653703886, "300–1000 Hz_10ms_envelope_correlation": 0.9934984633903845, "300–1000 Hz_10ms_modulation_mae_db": 0.3720240781367932, "300–1000 Hz_40ms_log_rms_mae_db": 0.973407902608413, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05866605306934031, "300–1000 Hz_40ms_envelope_correlation": 0.9974799952793574, "300–1000 Hz_40ms_modulation_mae_db": 0.3268095720255142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.657144094698374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15430508912686036, "1000–3000 Hz_10ms_envelope_correlation": 0.9823246001220174, "1000–3000 Hz_10ms_modulation_mae_db": 0.5624504626430072, "1000–3000 Hz_40ms_log_rms_mae_db": 1.034772105542004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09425999844517036, "1000–3000 Hz_40ms_envelope_correlation": 0.9931258151056691, "1000–3000 Hz_40ms_modulation_mae_db": 0.5058770561100778, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6991315747912275, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1732637747818411, "3000–6000 Hz_10ms_envelope_correlation": 0.9780311653346916, "3000–6000 Hz_10ms_modulation_mae_db": 0.46891512438549676, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9799026086842356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08869755319547455, "3000–6000 Hz_40ms_envelope_correlation": 0.99383902031008, "3000–6000 Hz_40ms_modulation_mae_db": 0.5903242834702658, "mr_stft_db": 5.81904569327071, "band_mae_db": 0.11737993759505035, "rms_gain_error_db": -0.0813101525464821, "output_peak": 0.41293229669396436}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.096658925751519, "全帯域_10ms_rms_envelope_nrmse": 0.09214360606413148, "全帯域_10ms_envelope_correlation": 0.9915396082596308, "全帯域_10ms_modulation_mae_db": 0.4805991847325155, "全帯域_40ms_log_rms_mae_db": 0.5371519742815754, "全帯域_40ms_rms_envelope_nrmse": 0.048651567105570045, "全帯域_40ms_envelope_correlation": 0.9974254737317879, "全帯域_40ms_modulation_mae_db": 0.42173769089922003, "300–1000 Hz_10ms_log_rms_mae_db": 1.6735149686638648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1060558105460679, "300–1000 Hz_10ms_envelope_correlation": 0.9919728644786805, "300–1000 Hz_10ms_modulation_mae_db": 0.3703098792496216, "300–1000 Hz_40ms_log_rms_mae_db": 0.931776536780525, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06128983941650563, "300–1000 Hz_40ms_envelope_correlation": 0.9971243621589126, "300–1000 Hz_40ms_modulation_mae_db": 0.495741232456359, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6681419900181431, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1415578010629765, "1000–3000 Hz_10ms_envelope_correlation": 0.9851834438082787, "1000–3000 Hz_10ms_modulation_mae_db": 0.41129134672417517, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9934883893312341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0791019009407585, "1000–3000 Hz_40ms_envelope_correlation": 0.9952841962575296, "1000–3000 Hz_40ms_modulation_mae_db": 0.49387168909745816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.653831394571583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1521488553429198, "3000–6000 Hz_10ms_envelope_correlation": 0.9829809940615873, "3000–6000 Hz_10ms_modulation_mae_db": 0.514362437936834, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9655172267483902, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0829402975241358, "3000–6000 Hz_40ms_envelope_correlation": 0.9946793413045361, "3000–6000 Hz_40ms_modulation_mae_db": 0.746537736125095, "mr_stft_db": 5.823113368116215, "band_mae_db": 0.14047022485229302, "rms_gain_error_db": -0.09802187686261633, "output_peak": 0.4083491619091603}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1025085376405739, "全帯域_10ms_rms_envelope_nrmse": 0.0914720247309671, "全帯域_10ms_envelope_correlation": 0.9916324471152304, "全帯域_10ms_modulation_mae_db": 0.322843655318223, "全帯域_40ms_log_rms_mae_db": 0.5862939504471458, "全帯域_40ms_rms_envelope_nrmse": 0.048814453695598316, "全帯域_40ms_envelope_correlation": 0.997314562963243, "全帯域_40ms_modulation_mae_db": 0.49275602073993685, "300–1000 Hz_10ms_log_rms_mae_db": 1.6218955893820777, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09474742293805385, "300–1000 Hz_10ms_envelope_correlation": 0.9936132802038404, "300–1000 Hz_10ms_modulation_mae_db": 0.2016373826889201, "300–1000 Hz_40ms_log_rms_mae_db": 0.9200616218318841, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05780927690566567, "300–1000 Hz_40ms_envelope_correlation": 0.9974429023198651, "300–1000 Hz_40ms_modulation_mae_db": 0.43640545630442057, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6733055380168942, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15172213531059714, "1000–3000 Hz_10ms_envelope_correlation": 0.9828852763150185, "1000–3000 Hz_10ms_modulation_mae_db": 0.17758024998821842, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0664360481600297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09423125274284898, "1000–3000 Hz_40ms_envelope_correlation": 0.9929867242265614, "1000–3000 Hz_40ms_modulation_mae_db": 0.2639306678388474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6745321543706209, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1530530268214451, "3000–6000 Hz_10ms_envelope_correlation": 0.9827761810536325, "3000–6000 Hz_10ms_modulation_mae_db": 0.8937891195447262, "3000–6000 Hz_40ms_log_rms_mae_db": 1.019176812432578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08533301037891626, "3000–6000 Hz_40ms_envelope_correlation": 0.9943740036585721, "3000–6000 Hz_40ms_modulation_mae_db": 1.1518560942615494, "mr_stft_db": 5.815925856464624, "band_mae_db": 0.1511477442615827, "rms_gain_error_db": -0.0409792873842826, "output_peak": 0.40894564117870796}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.978500441442182, "spectral_convergence": 0.41527531043233806, "spectral_convergence_status": "ok", "amplitude_floor": 9.383482434461407e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.691695009445188, "spectral_convergence": 0.3990187363655018, "spectral_convergence_status": "ok", "amplitude_floor": 6.715789372234631e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.840086471435262, "spectral_convergence": 0.4151790101523915, "spectral_convergence_status": "ok", "amplitude_floor": 5.217365851457555e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.0167397027219553e-05, "output_power": 3.0644146204755164e-05, "delta_db": 0.06809686717750667, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003363761740772986, "output_power": 0.0003405153598580365, "delta_db": 0.053114800844760855, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005972956425120644, "output_power": 0.0005987839496253154, "delta_db": 0.010808041613741626, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016539661163734624, "output_power": 0.0001533783626138226, "delta_db": -0.3276251106528666, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011856598981285755, "output_power": 0.00012544244039418902, "delta_db": 0.2448436308492287, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00046450684077542097, "output_power": 0.0004500400137839988, "delta_db": -0.13740984900462905, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14031638335712224, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03438707506692868, "pitch": null, "output_rms_dbfs": -27.698566554325907, "output_peak": 0.4581293407352761, "reference_rms_dbfs": -27.66417947925898, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.836760640774211}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014203698766261565, "log_rms_mae_db": 1.0902794596115246, "rms_envelope_nrmse": 0.09035964970073625, "envelope_correlation": 0.9918183917741347, "rms_gain_error_db": -0.034384297881887564, "reference_mean_power": 0.0017137270951261038, "output_mean_power": 0.001700212612261713}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.499341088840717e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046355332033754323, "output_power": 0.00044577758461190734, "reference_active": true, "delta_db": -0.16981469466921997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011665881324257648, "output_power": 0.00011230488031403852, "reference_active": true, "delta_db": -0.16518924739834726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5955876441113554e-05, "output_power": 2.135214002467002e-05, "reference_active": true, "delta_db": -0.8479428926293197}], "active_band_count": 3, "active_band_mae_db": 0.394315611565629}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274207517780892, "log_rms_mae_db": 0.6033594101086622, "rms_envelope_nrmse": 0.04955287601237036, "envelope_correlation": 0.9972126366963167, "rms_gain_error_db": -0.03534785186191865, "reference_mean_power": 0.0017200719561396216, "output_mean_power": 0.0017061288627469816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.335969034308142e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004202066244366408, "output_power": 0.00040392228786735634, "reference_active": true, "delta_db": -0.17165077057831546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.868302162815529e-05, "output_power": 5.669889273419271e-05, "reference_active": true, "delta_db": -0.1493789016983838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5858058061398606e-06, "output_power": 1.2703316313130469e-06, "reference_active": true, "delta_db": -0.9633289143370787}], "active_band_count": 3, "active_band_mae_db": 0.4281195288712593}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000125392198184175, "log_rms_mae_db": 1.5587071222401887, "rms_envelope_nrmse": 0.09311038296733562, "envelope_correlation": 0.9938261349500163, "rms_gain_error_db": -0.007408104873591133, "reference_mean_power": 0.0008371717774342565, "output_mean_power": 0.0008357449645019657}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.199403135704403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003464696140419235, "output_power": 0.0003330563527603887, "reference_active": true, "delta_db": -0.17147430412144157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.952130707503092e-05, "output_power": 8.30509444797799e-05, "reference_active": true, "delta_db": -0.32581838836106103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1976615680578699e-05, "output_power": 1.2741477053544073e-05, "reference_active": true, "delta_db": 0.2688566279307505}], "active_band_count": 3, "active_band_mae_db": 0.2553831068044177}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011133215930497508, "log_rms_mae_db": 0.8479043043084236, "rms_envelope_nrmse": 0.057002702355681556, "envelope_correlation": 0.9975027322889456, "rms_gain_error_db": -0.007479283323083246, "reference_mean_power": 0.0008413676993576814, "output_mean_power": 0.0008399199692762082}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.575111583882504e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003227285828337414, "output_power": 0.0003092932396474992, "reference_active": true, "delta_db": -0.18467003355491662}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.553081813418887e-05, "output_power": 4.262696582016376e-05, "reference_active": true, "delta_db": -0.2862103338583122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.055134444571575e-06, "output_power": 1.0109576336945843e-06, "reference_active": true, "delta_db": -0.18574844742166505}], "active_band_count": 3, "active_band_mae_db": 0.21887627161163128}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.277196005950351e-05, "log_rms_mae_db": 1.6470804176138145, "rms_envelope_nrmse": 0.13838277630468518, "envelope_correlation": 0.9857976267506949, "rms_gain_error_db": -0.15322348573505273, "reference_mean_power": 0.0001860033831933617, "output_mean_power": 0.00017955541004371483}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.45795784914854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.813551789415923e-05, "output_power": 7.96403448977971e-05, "reference_active": true, "delta_db": -0.4401782866227922}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.039396685146323e-05, "output_power": 1.851915536863611e-05, "reference_active": true, "delta_db": -0.41880533794342334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.503164081364961e-06, "output_power": 3.7520665230990417e-06, "reference_active": true, "delta_db": -0.7924724190229109}], "active_band_count": 3, "active_band_mae_db": 0.5504853478630422}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5906277437135206e-05, "log_rms_mae_db": 0.9999927016687438, "rms_envelope_nrmse": 0.07913605544270376, "envelope_correlation": 0.995162647711932, "rms_gain_error_db": -0.15348102421737636, "reference_mean_power": 0.00018692252747526107, "output_mean_power": 0.00018043199135803036}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315230721785866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.510469005068111e-05, "output_power": 7.653436401539352e-05, "reference_active": true, "delta_db": -0.4609701670781614}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648548828731175e-06, "output_power": 9.054066823113664e-06, "reference_active": true, "delta_db": -0.2761830357534034}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.651744211074769e-07, "output_power": 2.6336071760823673e-07, "reference_active": true, "delta_db": -1.4194935256882109}], "active_band_count": 3, "active_band_mae_db": 0.7188822428399252}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.827493119721784e-05, "log_rms_mae_db": 1.7293961974183079, "rms_envelope_nrmse": 0.18125721271315498, "envelope_correlation": 0.9758083564308918, "rms_gain_error_db": -0.15416474380235454, "reference_mean_power": 0.0003389197953959056, "output_mean_power": 0.0003270999375228993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5434076582307222e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.258901619859077e-05, "output_power": 6.127871892273046e-05, "reference_active": true, "delta_db": -0.09188447707405886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4223152137238325e-05, "output_power": 1.2852863989405935e-05, "reference_active": true, "delta_db": -0.4399594366489774}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.429050160653041e-06, "output_power": 7.070957725615104e-06, "reference_active": true, "delta_db": -0.7630039751732647}], "active_band_count": 3, "active_band_mae_db": 0.43161596296543364}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.761964716205697e-05, "log_rms_mae_db": 1.0253317205243873, "rms_envelope_nrmse": 0.10089737296490923, "envelope_correlation": 0.992147238158374, "rms_gain_error_db": -0.15787402932208355, "reference_mean_power": 0.00033859490771126537, "output_mean_power": 0.00032650739301489013}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4129997856527214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.165966182107609e-05, "output_power": 5.982637414517351e-05, "reference_active": true, "delta_db": -0.13108455802566657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.083964993529701e-06, "output_power": 5.989029283235772e-06, "reference_active": true, "delta_db": -0.729199698289823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.531889318612958e-07, "output_power": 4.2253948345790675e-07, "reference_active": true, "delta_db": -1.8917152004030091}], "active_band_count": 3, "active_band_mae_db": 0.9173331522394995}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8808471160401535, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b9054164e8bc3e5b8f5fc5f1428c9780ada1d5757b9369631175e7216edb9790", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9839472011141633, "standard_deviation": 0.0013490076212228504, "minimum": 0.9823246001220174, "maximum": 0.9857976267506949, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6635897283345435, "standard_deviation": 0.016967087416195722, "minimum": 1.6404751981325763, "maximum": 1.6875115683206945, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.420225478205825, "standard_deviation": 0.12537652258727708, "minimum": 0.17758024998821842, "maximum": 0.5624504626430072, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14750728290782497, "standard_deviation": 0.006563985821732849, "minimum": 0.13838277630468518, "maximum": 0.15430508912686036, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942948367651987, "standard_deviation": 0.000990373348392504, "minimum": 0.9929867242265614, "maximum": 0.995601469917732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0299049620547984, "standard_deviation": 0.02725073822871131, "minimum": 0.9934883893312341, "maximum": 1.0664360481600297, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.449429158248265, "standard_deviation": 0.173055585593236, "minimum": 0.18789567232066715, "maximum": 0.7188822428399252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08680339799129272, "standard_deviation": 0.00746830401086802, "minimum": 0.07644421840400956, "maximum": 0.09425999844517036, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9810184354317066, "standard_deviation": 0.002711098186710958, "minimum": 0.9758083564308918, "maximum": 0.9833769119726495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6738828003207362, "standard_deviation": 0.027624212495428013, "minimum": 1.645135991784917, "maximum": 1.7293961974183079, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5392744827992302, "standard_deviation": 0.25792437566218207, "minimum": 0.15035142677219449, "maximum": 0.8981460668552442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1605545666845203, "standard_deviation": 0.011089599089322682, "minimum": 0.15067648773414197, "maximum": 0.18125721271315498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940107422462903, "standard_deviation": 0.0012457901647987405, "minimum": 0.992147238158374, "maximum": 0.9959616739014053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9885437467502356, "standard_deviation": 0.024706175366845655, "minimum": 0.9552134131030084, "maximum": 1.0253317205243873, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8206579336772835, "standard_deviation": 0.2557771561112957, "minimum": 0.45300147514967093, "maximum": 1.1518560942615494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08739332534623943, "standard_deviation": 0.009146043028447647, "minimum": 0.0719179000329486, "maximum": 0.10089737296490923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935398456042019, "standard_deviation": 0.0006999844295098923, "minimum": 0.9919728644786805, "maximum": 0.9944354226644762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6199115190687297, "standard_deviation": 0.040332625495572386, "minimum": 1.5587071222401887, "maximum": 1.6735149686638648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.28839600453126435, "standard_deviation": 0.1182128884660048, "minimum": 0.13985981130336236, "maximum": 0.49336124414436755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09517621082156238, "standard_deviation": 0.004979527250849407, "minimum": 0.08843689448491862, "maximum": 0.1060558105460679, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976249464363144, "standard_deviation": 0.00032217786472821567, "minimum": 0.9971243621589126, "maximum": 0.9981355050292429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8998646972194368, "standard_deviation": 0.04362220302585778, "minimum": 0.8479043043084236, "maximum": 0.973407902608413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3479977594343078, "standard_deviation": 0.1282706808510795, "minimum": 0.15933710726209552, "maximum": 0.5169178347665189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05583484765701648, "standard_deviation": 0.0037981665888739014, "minimum": 0.04972190486948176, "maximum": 0.06128983941650563, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13746463982022475, "standard_deviation": 0.017315591073122943, "minimum": 0.1100617358241938, "maximum": 0.15881112071206058, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.81714445701556, "standard_deviation": 0.011786247942742493, "minimum": 5.801617005766915, "maximum": 5.836760640774211, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42182663887420296, "standard_deviation": 0.020054502180282766, "minimum": 0.40000210473451775, "maximum": 0.4581293407352761, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.050702661347259856, "standard_deviation": 0.03455316931240437, "minimum": -0.09802187686261633, "maximum": -0.004601190127547735, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918028867125666, "standard_deviation": 0.0003189154439098447, "minimum": 0.9913586440217722, "maximum": 0.9922806380644934, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0790000208764825, "standard_deviation": 0.018605468700922712, "minimum": 1.0445387578968706, "maximum": 1.1025085376405739, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3327749211194887, "standard_deviation": 0.09342097813322958, "minimum": 0.21607188678799474, "maximum": 0.4805991847325155, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09054780346268282, "standard_deviation": 0.0017912165306960132, "minimum": 0.08788897379629201, "maximum": 0.09305310996806322, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974274437764754, "standard_deviation": 0.0001991579206838572, "minimum": 0.9971749630718825, "maximum": 0.9977200355331932, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5550240373666535, "standard_deviation": 0.027893489482670758, "minimum": 0.5178395561872985, "maximum": 0.6033594101086622, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3678883212456007, "standard_deviation": 0.09849647832835544, "minimum": 0.1950441330212579, "maximum": 0.49275602073993685, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04792283523994105, "standard_deviation": 0.001962439611237255, "minimum": 0.04515164667094036, "maximum": 0.05042287086544763, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4446-2273-0010", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.645928144454956, "predicted_mos": 2.645928144454956, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ea368f46d5587d16afcda44a266e2b58af7869e1e31beea058161ccb709bf5f8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.058146200142801, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.12169919451327, "全帯域_10ms_rms_envelope_nrmse": 0.08877978840878939, "全帯域_10ms_envelope_correlation": 0.992209311341307, "全帯域_10ms_modulation_mae_db": 0.41310534329554366, "全帯域_40ms_log_rms_mae_db": 0.5547641090596276, "全帯域_40ms_rms_envelope_nrmse": 0.04967270664586565, "全帯域_40ms_envelope_correlation": 0.9973747386803797, "全帯域_40ms_modulation_mae_db": 0.51343183933008, "300–1000 Hz_10ms_log_rms_mae_db": 1.5791273743592038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08953262227807007, "300–1000 Hz_10ms_envelope_correlation": 0.9942909840808891, "300–1000 Hz_10ms_modulation_mae_db": 0.29871558547298255, "300–1000 Hz_40ms_log_rms_mae_db": 0.8675859693299965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05017217419681565, "300–1000 Hz_40ms_envelope_correlation": 0.9980906203142462, "300–1000 Hz_40ms_modulation_mae_db": 0.3811956780020513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.636143889326999, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15328743312940052, "1000–3000 Hz_10ms_envelope_correlation": 0.9833184631618553, "1000–3000 Hz_10ms_modulation_mae_db": 0.7822541440908397, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0006677970195172, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10141377082010344, "1000–3000 Hz_40ms_envelope_correlation": 0.9930357422001046, "1000–3000 Hz_40ms_modulation_mae_db": 0.8168625484932123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6241269341228077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1495657650258971, "3000–6000 Hz_10ms_envelope_correlation": 0.9838483981256907, "3000–6000 Hz_10ms_modulation_mae_db": 0.33421920876190886, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9348650158129799, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08543795037709573, "3000–6000 Hz_40ms_envelope_correlation": 0.9948761051787769, "3000–6000 Hz_40ms_modulation_mae_db": 0.41952596647896484, "mr_stft_db": 5.79798337462752, "band_mae_db": 0.22093204400442942, "rms_gain_error_db": -0.12084273742387254, "output_peak": 0.41313157378978616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.072928280771221, "全帯域_10ms_rms_envelope_nrmse": 0.08069692782160877, "全帯域_10ms_envelope_correlation": 0.9935514231849121, "全帯域_10ms_modulation_mae_db": 0.5255770030927774, "全帯域_40ms_log_rms_mae_db": 0.5623529804738566, "全帯域_40ms_rms_envelope_nrmse": 0.04175287100691623, "全帯域_40ms_envelope_correlation": 0.9981382119098704, "全帯域_40ms_modulation_mae_db": 0.6041858522601914, "300–1000 Hz_10ms_log_rms_mae_db": 1.5713720637027642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09616303356877565, "300–1000 Hz_10ms_envelope_correlation": 0.9934333133207447, "300–1000 Hz_10ms_modulation_mae_db": 0.3528617821774951, "300–1000 Hz_40ms_log_rms_mae_db": 0.8683863905097775, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05483943566184116, "300–1000 Hz_40ms_envelope_correlation": 0.9977580461667948, "300–1000 Hz_40ms_modulation_mae_db": 0.52983944416727, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6256163701057516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14299206722227553, "1000–3000 Hz_10ms_envelope_correlation": 0.9847263324549679, "1000–3000 Hz_10ms_modulation_mae_db": 0.48275769790680007, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9079685613257537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07436488212473868, "1000–3000 Hz_40ms_envelope_correlation": 0.9955843167324306, "1000–3000 Hz_40ms_modulation_mae_db": 0.6145369066835494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7026270924126259, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13822405558178574, "3000–6000 Hz_10ms_envelope_correlation": 0.9862044044891938, "3000–6000 Hz_10ms_modulation_mae_db": 0.8401208373467431, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9949929839840099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07493769348907502, "3000–6000 Hz_40ms_envelope_correlation": 0.9960348262654858, "3000–6000 Hz_40ms_modulation_mae_db": 1.1452679961943515, "mr_stft_db": 5.79655882008399, "band_mae_db": 0.1495589421677326, "rms_gain_error_db": -0.09916748654491277, "output_peak": 0.41840229761772757}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.102633610096708, "全帯域_10ms_rms_envelope_nrmse": 0.08839240837665861, "全帯域_10ms_envelope_correlation": 0.9923070242746074, "全帯域_10ms_modulation_mae_db": 0.6340785589991385, "全帯域_40ms_log_rms_mae_db": 0.5586216047119844, "全帯域_40ms_rms_envelope_nrmse": 0.052070061298312056, "全帯域_40ms_envelope_correlation": 0.9970950028160894, "全帯域_40ms_modulation_mae_db": 0.7296605753158311, "300–1000 Hz_10ms_log_rms_mae_db": 1.553454118296258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0850030360129612, "300–1000 Hz_10ms_envelope_correlation": 0.9949880895318973, "300–1000 Hz_10ms_modulation_mae_db": 0.5978730681367277, "300–1000 Hz_40ms_log_rms_mae_db": 0.8683273965286145, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04718552327957964, "300–1000 Hz_40ms_envelope_correlation": 0.998458466212242, "300–1000 Hz_40ms_modulation_mae_db": 0.792466878625627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6402209053419985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15855792137863386, "1000–3000 Hz_10ms_envelope_correlation": 0.9811607452163948, "1000–3000 Hz_10ms_modulation_mae_db": 0.4951644848875129, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9890224592279391, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10465204151183843, "1000–3000 Hz_40ms_envelope_correlation": 0.9910798370597973, "1000–3000 Hz_40ms_modulation_mae_db": 0.32376881297094207, "3000–6000 Hz_10ms_log_rms_mae_db": 1.58582448264825, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15276043400048386, "3000–6000 Hz_10ms_envelope_correlation": 0.9828952885556604, "3000–6000 Hz_10ms_modulation_mae_db": 0.5061080003917752, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9001679380607802, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08921941205063223, "3000–6000 Hz_40ms_envelope_correlation": 0.993949613992093, "3000–6000 Hz_40ms_modulation_mae_db": 0.7655866823828886, "mr_stft_db": 5.779430134230085, "band_mae_db": 0.1688297546344338, "rms_gain_error_db": -0.11594417528939138, "output_peak": 0.40758357390917577}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0385543270325106, "全帯域_10ms_rms_envelope_nrmse": 0.08555860782957471, "全帯域_10ms_envelope_correlation": 0.9926870926392498, "全帯域_10ms_modulation_mae_db": 0.35110936282197347, "全帯域_40ms_log_rms_mae_db": 0.5256898027861823, "全帯域_40ms_rms_envelope_nrmse": 0.047775110667737784, "全帯域_40ms_envelope_correlation": 0.997438087311288, "全帯域_40ms_modulation_mae_db": 0.531356481966211, "300–1000 Hz_10ms_log_rms_mae_db": 1.595165923478933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09293988892205052, "300–1000 Hz_10ms_envelope_correlation": 0.993849904175088, "300–1000 Hz_10ms_modulation_mae_db": 0.24399498908943937, "300–1000 Hz_40ms_log_rms_mae_db": 0.8692941873971034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05689715564094644, "300–1000 Hz_40ms_envelope_correlation": 0.997512121277567, "300–1000 Hz_40ms_modulation_mae_db": 0.3611326894453301, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6360471733017348, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15974790724030338, "1000–3000 Hz_10ms_envelope_correlation": 0.9813637396879801, "1000–3000 Hz_10ms_modulation_mae_db": 0.5611415851794016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9867222382198828, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09331879500122045, "1000–3000 Hz_40ms_envelope_correlation": 0.9938818088818469, "1000–3000 Hz_40ms_modulation_mae_db": 0.5461197086660233, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6081726455976884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14710285325422506, "3000–6000 Hz_10ms_envelope_correlation": 0.9841382487909883, "3000–6000 Hz_10ms_modulation_mae_db": 0.437354136676132, "3000–6000 Hz_40ms_log_rms_mae_db": 0.937487392481848, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08602558914610682, "3000–6000 Hz_40ms_envelope_correlation": 0.9943623467253516, "3000–6000 Hz_40ms_modulation_mae_db": 0.775415132087575, "mr_stft_db": 5.797110621011311, "band_mae_db": 0.12824995624539817, "rms_gain_error_db": -0.05852311569057675, "output_peak": 0.4873100117914193}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0829326142846278, "全帯域_10ms_rms_envelope_nrmse": 0.087423390538148, "全帯域_10ms_envelope_correlation": 0.9925895368243075, "全帯域_10ms_modulation_mae_db": 0.5514873086809732, "全帯域_40ms_log_rms_mae_db": 0.5757016451318814, "全帯域_40ms_rms_envelope_nrmse": 0.04755506459030125, "全帯域_40ms_envelope_correlation": 0.9977937060867615, "全帯域_40ms_modulation_mae_db": 0.6545207042704928, "300–1000 Hz_10ms_log_rms_mae_db": 1.6072747469785027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09598739629010287, "300–1000 Hz_10ms_envelope_correlation": 0.9936607200280332, "300–1000 Hz_10ms_modulation_mae_db": 0.46795793520194423, "300–1000 Hz_40ms_log_rms_mae_db": 0.8570135964803618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05727151339015362, "300–1000 Hz_40ms_envelope_correlation": 0.9978095788021242, "300–1000 Hz_40ms_modulation_mae_db": 0.6158785935766079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6082344916291478, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1462846361034335, "1000–3000 Hz_10ms_envelope_correlation": 0.9840587353268578, "1000–3000 Hz_10ms_modulation_mae_db": 0.2110270302404971, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9320183222857522, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08105973165128594, "1000–3000 Hz_40ms_envelope_correlation": 0.994871203967999, "1000–3000 Hz_40ms_modulation_mae_db": 0.16655882854696616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6366095357155885, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14503444887631645, "3000–6000 Hz_10ms_envelope_correlation": 0.9850050585678461, "3000–6000 Hz_10ms_modulation_mae_db": 0.8857645166382295, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9510234765204211, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0760969747558608, "3000–6000 Hz_40ms_envelope_correlation": 0.9962923869361927, "3000–6000 Hz_40ms_modulation_mae_db": 1.2152642995065015, "mr_stft_db": 5.776761806065202, "band_mae_db": 0.1690399026187794, "rms_gain_error_db": -0.16543559843196914, "output_peak": 0.4197218602894628}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0797063423951083, "全帯域_10ms_rms_envelope_nrmse": 0.08713020061239707, "全帯域_10ms_envelope_correlation": 0.9924209904939146, "全帯域_10ms_modulation_mae_db": 0.2775544540074167, "全帯域_40ms_log_rms_mae_db": 0.5643939638799989, "全帯域_40ms_rms_envelope_nrmse": 0.04446662430920326, "全帯域_40ms_envelope_correlation": 0.9978256056072118, "全帯域_40ms_modulation_mae_db": 0.24563612887853656, "300–1000 Hz_10ms_log_rms_mae_db": 1.5899173301506422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09232163057953366, "300–1000 Hz_10ms_envelope_correlation": 0.9939224234349191, "300–1000 Hz_10ms_modulation_mae_db": 0.20503329926813643, "300–1000 Hz_40ms_log_rms_mae_db": 0.8741659121027757, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051280271895989805, "300–1000 Hz_40ms_envelope_correlation": 0.9979933279782703, "300–1000 Hz_40ms_modulation_mae_db": 0.3219416695988588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6575787313378731, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15953947056584777, "1000–3000 Hz_10ms_envelope_correlation": 0.98109057301118, "1000–3000 Hz_10ms_modulation_mae_db": 0.14472999336850836, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0094347784470805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10653165251581155, "1000–3000 Hz_40ms_envelope_correlation": 0.9910885975679494, "1000–3000 Hz_40ms_modulation_mae_db": 0.30803839104621744, "3000–6000 Hz_10ms_log_rms_mae_db": 1.608826151364791, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1448517650969755, "3000–6000 Hz_10ms_envelope_correlation": 0.984617904002972, "3000–6000 Hz_10ms_modulation_mae_db": 0.44902842623294736, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9128301473990514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07342428706573424, "3000–6000 Hz_40ms_envelope_correlation": 0.9959517219098325, "3000–6000 Hz_40ms_modulation_mae_db": 0.6675729595391271, "mr_stft_db": 5.786929820445387, "band_mae_db": 0.13616635802102575, "rms_gain_error_db": -0.07775895622326807, "output_peak": 0.4185099599584656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1133629282594666, "全帯域_10ms_rms_envelope_nrmse": 0.09638894403058472, "全帯域_10ms_envelope_correlation": 0.9907392846424231, "全帯域_10ms_modulation_mae_db": 0.5999736864809925, "全帯域_40ms_log_rms_mae_db": 0.5971076479772371, "全帯域_40ms_rms_envelope_nrmse": 0.05348958784390413, "全帯域_40ms_envelope_correlation": 0.9968298885175781, "全帯域_40ms_modulation_mae_db": 0.6714522965156317, "300–1000 Hz_10ms_log_rms_mae_db": 1.5877457087186084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10432525564691977, "300–1000 Hz_10ms_envelope_correlation": 0.9923865144744537, "300–1000 Hz_10ms_modulation_mae_db": 0.514026348041889, "300–1000 Hz_40ms_log_rms_mae_db": 0.8765554176677302, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06337448047290838, "300–1000 Hz_40ms_envelope_correlation": 0.9971376122908224, "300–1000 Hz_40ms_modulation_mae_db": 0.7606491147851037, "1000–3000 Hz_10ms_log_rms_mae_db": 1.659259964132762, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15090693401534022, "1000–3000 Hz_10ms_envelope_correlation": 0.9834049888949101, "1000–3000 Hz_10ms_modulation_mae_db": 0.7154612601895853, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0406864356202012, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09380179306182933, "1000–3000 Hz_40ms_envelope_correlation": 0.9936181470800205, "1000–3000 Hz_40ms_modulation_mae_db": 0.9341823195213372, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6179546446821151, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1587834909925898, "3000–6000 Hz_10ms_envelope_correlation": 0.9816517736198513, "3000–6000 Hz_10ms_modulation_mae_db": 0.7095246167203176, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9391526046393845, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07897918719859702, "3000–6000 Hz_40ms_envelope_correlation": 0.9951368171465573, "3000–6000 Hz_40ms_modulation_mae_db": 0.6807554708769873, "mr_stft_db": 5.773114027516868, "band_mae_db": 0.1778563763151851, "rms_gain_error_db": -0.075763419386384, "output_peak": 0.4261759568769823}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.10072793164958, "全帯域_10ms_rms_envelope_nrmse": 0.08759581870649152, "全帯域_10ms_envelope_correlation": 0.9926291388599494, "全帯域_10ms_modulation_mae_db": 0.5974097708599667, "全帯域_40ms_log_rms_mae_db": 0.5467139805238223, "全帯域_40ms_rms_envelope_nrmse": 0.047644060427220566, "全帯域_40ms_envelope_correlation": 0.9978934485842728, "全帯域_40ms_modulation_mae_db": 0.638365865693674, "300–1000 Hz_10ms_log_rms_mae_db": 1.5770838499463418, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09001113669660159, "300–1000 Hz_10ms_envelope_correlation": 0.9943825225864037, "300–1000 Hz_10ms_modulation_mae_db": 0.26334449660774023, "300–1000 Hz_40ms_log_rms_mae_db": 0.9044553960863659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05837172471365874, "300–1000 Hz_40ms_envelope_correlation": 0.9976088180609214, "300–1000 Hz_40ms_modulation_mae_db": 0.43446087534531985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6191594335050032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13856219604320433, "1000–3000 Hz_10ms_envelope_correlation": 0.9865312161568757, "1000–3000 Hz_10ms_modulation_mae_db": 0.5850062266229531, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9575443538564186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0811396293899833, "1000–3000 Hz_40ms_envelope_correlation": 0.9959676466657791, "1000–3000 Hz_40ms_modulation_mae_db": 0.6577572997394533, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6395331478580852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15928733307977344, "3000–6000 Hz_10ms_envelope_correlation": 0.9816339738706195, "3000–6000 Hz_10ms_modulation_mae_db": 0.9062354853172031, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9626000945449329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07870933636904912, "3000–6000 Hz_40ms_envelope_correlation": 0.9958843377242251, "3000–6000 Hz_40ms_modulation_mae_db": 1.1186729804608926, "mr_stft_db": 5.785871932334377, "band_mae_db": 0.2253076661014426, "rms_gain_error_db": -0.18725213873085955, "output_peak": 0.39995314986826863}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.963002857864651, "spectral_convergence": 0.4102459313998252, "spectral_convergence_status": "ok", "amplitude_floor": 9.383482434461407e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.657552562738281, "spectral_convergence": 0.39075667935584, "spectral_convergence_status": "ok", "amplitude_floor": 6.715789372234631e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.773394703279626, "spectral_convergence": 0.4081743180870821, "spectral_convergence_status": "ok", "amplitude_floor": 5.217365851457555e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.0167397027219553e-05, "output_power": 3.3166943585882864e-05, "delta_db": 0.4116761308033824, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003363761740772986, "output_power": 0.00033476612307651676, "delta_db": -0.020837238528256186, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005972956425120644, "output_power": 0.0005905220669419366, "delta_db": -0.04953215223757631, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016539661163734624, "output_power": 0.00014535212306478017, "delta_db": -0.5610522861337676, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011856598981285755, "output_power": 0.00011577798881419937, "delta_db": -0.10334130087394873, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00046450684077542097, "output_power": 0.00044573506320019486, "delta_db": -0.17915315544964533, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22093204400442942, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12084273742387254, "pitch": null, "output_rms_dbfs": -27.78502221668285, "output_peak": 0.41313157378978616, "reference_rms_dbfs": -27.66417947925898, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.79798337462752}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014203698766261565, "log_rms_mae_db": 1.12169919451327, "rms_envelope_nrmse": 0.08877978840878939, "envelope_correlation": 0.992209311341307, "rms_gain_error_db": -0.12083059010217528, "reference_mean_power": 0.0017137270951261038, "output_mean_power": 0.001666704485647864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.499341088840717e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046355332033754323, "output_power": 0.0004443172586226107, "reference_active": true, "delta_db": -0.18406513309809414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011665881324257648, "output_power": 0.0001033359786973722, "reference_active": true, "delta_db": -0.5266599704583514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5955876441113554e-05, "output_power": 2.2981406540382666e-05, "reference_active": true, "delta_db": -0.5285909263301855}], "active_band_count": 3, "active_band_mae_db": 0.41310534329554366}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274207517780892, "log_rms_mae_db": 0.5547641090596276, "rms_envelope_nrmse": 0.04967270664586565, "envelope_correlation": 0.9973747386803797, "rms_gain_error_db": -0.12116504355891121, "reference_mean_power": 0.0017200719561396216, "output_mean_power": 0.0016727464268849264}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.335969034308142e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004202066244366408, "output_power": 0.00040336749242386743, "reference_active": true, "delta_db": -0.17761999361337305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.868302162815529e-05, "output_power": 5.200266803323417e-05, "reference_active": true, "delta_db": -0.5248684187285233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5858058061398606e-06, "output_power": 1.307583082493401e-06, "reference_active": true, "delta_db": -0.8378071056483436}], "active_band_count": 3, "active_band_mae_db": 0.51343183933008}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000125392198184175, "log_rms_mae_db": 1.5791273743592038, "rms_envelope_nrmse": 0.08953262227807007, "envelope_correlation": 0.9942909840808891, "rms_gain_error_db": -0.0607431430163411, "reference_mean_power": 0.0008371717774342565, "output_mean_power": 0.0008255440749690482}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.199403135704403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003464696140419235, "output_power": 0.0003375884924994181, "reference_active": true, "delta_db": -0.11277518020279198}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.952130707503092e-05, "output_power": 8.044445982411373e-05, "reference_active": true, "delta_db": -0.4643027478839553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1976615680578699e-05, "output_power": 1.1128259151136405e-05, "reference_active": true, "delta_db": -0.3190688283322004}], "active_band_count": 3, "active_band_mae_db": 0.29871558547298255}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011133215930497508, "log_rms_mae_db": 0.8675859693299965, "rms_envelope_nrmse": 0.05017217419681565, "envelope_correlation": 0.9980906203142462, "rms_gain_error_db": -0.06082435418837305, "reference_mean_power": 0.0008413676993576814, "output_mean_power": 0.0008296662040710986}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.575111583882504e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003227285828337414, "output_power": 0.0003144482590943027, "reference_active": true, "delta_db": -0.11288236254104368}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.553081813418887e-05, "output_power": 4.118176355452531e-05, "reference_active": true, "delta_db": -0.4360051336252157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.055134444571575e-06, "output_power": 9.201059664551172e-07, "reference_active": true, "delta_db": -0.5946995378398944}], "active_band_count": 3, "active_band_mae_db": 0.3811956780020513}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.277196005950351e-05, "log_rms_mae_db": 1.636143889326999, "rms_envelope_nrmse": 0.15328743312940052, "envelope_correlation": 0.9833184631618553, "rms_gain_error_db": -0.39188201599975836, "reference_mean_power": 0.0001860033831933617, "output_mean_power": 0.00016995448976304253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.45795784914854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.813551789415923e-05, "output_power": 7.407055699394238e-05, "reference_active": true, "delta_db": -0.7550535035654493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.039396685146323e-05, "output_power": 1.672956664047287e-05, "reference_active": true, "delta_db": -0.8601701783507023}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.503164081364961e-06, "output_power": 3.805080915796822e-06, "reference_active": true, "delta_db": -0.7315387503563675}], "active_band_count": 3, "active_band_mae_db": 0.7822541440908397}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5906277437135206e-05, "log_rms_mae_db": 1.0006677970195172, "rms_envelope_nrmse": 0.10141377082010344, "envelope_correlation": 0.9930357422001046, "rms_gain_error_db": -0.39201204996379196, "reference_mean_power": 0.00018692252747526107, "output_mean_power": 0.00017078921394985346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315230721785866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.510469005068111e-05, "output_power": 7.07510565222091e-05, "reference_active": true, "delta_db": -0.8022056481808147}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648548828731175e-06, "output_power": 7.949882644350523e-06, "reference_active": true, "delta_db": -0.8410128135941488}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.651744211074769e-07, "output_power": 3.0322390129643326e-07, "reference_active": true, "delta_db": -0.8073691837046731}], "active_band_count": 3, "active_band_mae_db": 0.8168625484932123}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.827493119721784e-05, "log_rms_mae_db": 1.6241269341228077, "rms_envelope_nrmse": 0.1495657650258971, "envelope_correlation": 0.9838483981256907, "rms_gain_error_db": -0.27171120795893433, "reference_mean_power": 0.0003389197953959056, "output_mean_power": 0.0003183653650458571}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5434076582307222e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.258901619859077e-05, "output_power": 6.025373223272799e-05, "reference_active": true, "delta_db": -0.16514172056787832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4223152137238325e-05, "output_power": 1.4090906652875481e-05, "reference_active": true, "delta_db": -0.040569176030826964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.429050160653041e-06, "output_power": 7.015909290163336e-06, "reference_active": true, "delta_db": -0.7969467296870213}], "active_band_count": 3, "active_band_mae_db": 0.33421920876190886}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.761964716205697e-05, "log_rms_mae_db": 0.9348650158129799, "rms_envelope_nrmse": 0.08543795037709573, "envelope_correlation": 0.9948761051787769, "rms_gain_error_db": -0.2716747885890169, "reference_mean_power": 0.00033859490771126537, "output_mean_power": 0.0003180628480101958}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4129997856527214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.165966182107609e-05, "output_power": 5.9519111654401766e-05, "reference_active": true, "delta_db": -0.1534469788881935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.083964993529701e-06, "output_power": 6.846045549519538e-06, "reference_active": true, "delta_db": -0.14836622043419628}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.531889318612958e-07, "output_power": 5.240374770584309e-07, "reference_active": true, "delta_db": -0.9567647001145047}], "active_band_count": 3, "active_band_mae_db": 0.41952596647896484}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.882681435766504, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1f6a997bfc05f796774e6f94f85ccb5dba65b9c7e7cda5afe06d36b05194f70e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9832068492388778, "standard_deviation": 0.00193464249583362, "minimum": 0.98109057301118, "maximum": 0.9865312161568757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6352826198351589, "standard_deviation": 0.017658415150177638, "minimum": 1.6082344916291478, "maximum": 1.659259964132762, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4971928028107623, "standard_deviation": 0.22265768079043555, "minimum": 0.14472999336850836, "maximum": 0.7822541440908397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1512348207123049, "standard_deviation": 0.008039466142336133, "minimum": 0.13856219604320433, "maximum": 0.15974790724030338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9936409125194909, "standard_deviation": 0.0018594288036480131, "minimum": 0.9910798370597973, "maximum": 0.9959676466657791, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9780081182503182, "standard_deviation": 0.04320430797898658, "minimum": 0.9079685613257537, "maximum": 1.0406864356202012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5459781019584626, "standard_deviation": 0.2649062033194406, "minimum": 0.16655882854696616, "maximum": 0.9341823195213372, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09203528700960138, "standard_deviation": 0.012032529607114945, "minimum": 0.07436488212473868, "maximum": 0.10653165251581155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9837493812528528, "standard_deviation": 0.0016090635782249445, "minimum": 0.9816339738706195, "maximum": 0.9862044044891938, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6279593293002441, "standard_deviation": 0.034706466697948544, "minimum": 1.58582448264825, "maximum": 1.7026270924126259, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.633544403510657, "standard_deviation": 0.22828384684151903, "minimum": 0.33421920876190886, "maximum": 0.9062354853172031, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14945126823850585, "standard_deviation": 0.007241066614928279, "minimum": 0.13822405558178574, "maximum": 0.15928733307977344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9953110194848144, "standard_deviation": 0.0008618148259710592, "minimum": 0.993949613992093, "maximum": 0.9962923869361927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.941639956680426, "standard_deviation": 0.029260376227729364, "minimum": 0.9001679380607802, "maximum": 0.9949929839840099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8485076859409111, "standard_deviation": 0.28099016059749143, "minimum": 0.41952596647896484, "maximum": 1.2152642995065015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08035380380651887, "standard_deviation": 0.005814803713642074, "minimum": 0.07342428706573424, "maximum": 0.08921941205063223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938643089540536, "standard_deviation": 0.0007679422244787052, "minimum": 0.9923865144744537, "maximum": 0.9949880895318973, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5826426394539068, "standard_deviation": 0.016321715067166762, "minimum": 1.553454118296258, "maximum": 1.6072747469785027, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.36797593799954437, "standard_deviation": 0.14246531625972197, "minimum": 0.20503329926813643, "maximum": 0.5978730681367277, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09328549999937691, "standard_deviation": 0.005755865268917383, "minimum": 0.0850030360129612, "maximum": 0.10432525564691977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977960738878735, "standard_deviation": 0.0003991641909839844, "minimum": 0.9971376122908224, "maximum": 0.998458466212242, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8732230332628407, "standard_deviation": 0.013860389741212734, "minimum": 0.8570135964803618, "maximum": 0.9044553960863659, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5246956179432711, "standard_deviation": 0.1821744969589803, "minimum": 0.3219416695988588, "maximum": 0.792466878625627, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05492403490648668, "standard_deviation": 0.005188577623069449, "minimum": 0.04718552327957964, "maximum": 0.06337448047290838, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17199262501355336, "standard_deviation": 0.03583868388622675, "minimum": 0.12824995624539817, "maximum": 0.2253076661014426, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.786720067039342, "standard_deviation": 0.009782086350588994, "minimum": 5.773114027516868, "maximum": 5.79798337462752, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42384854801266103, "standard_deviation": 0.02688164245052671, "minimum": 0.39995314986826863, "maximum": 0.4873100117914193, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11258595346515426, "standard_deviation": 0.04491140606955492, "minimum": -0.18725213873085955, "maximum": -0.05852311569057675, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9923917252825838, "standard_deviation": 0.000783718366085528, "minimum": 0.9907392846424231, "maximum": 0.9935514231849121, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0890681536253117, "standard_deviation": 0.026491032073173548, "minimum": 1.0385543270325106, "maximum": 1.12169919451327, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4937869360298478, "standard_deviation": 0.1307439669720498, "minimum": 0.2775544540074167, "maximum": 0.6340785589991385, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0877457607905316, "standard_deviation": 0.004328757435975429, "minimum": 0.08069692782160877, "maximum": 0.09638894403058472, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975485861891815, "standard_deviation": 0.00044209069238391677, "minimum": 0.9968298885175781, "maximum": 0.9981382119098704, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5606682168180739, "standard_deviation": 0.020797709345342596, "minimum": 0.5256898027861823, "maximum": 0.5971076479772371, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5735762180288311, "standard_deviation": 0.15051021123833053, "minimum": 0.24563612887853656, "maximum": 0.7296605753158311, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04805326084868262, "standard_deviation": 0.0038053125943614736, "minimum": 0.04175287100691623, "maximum": 0.05348958784390413, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4507-16021-0016", "language": "en", "style": "normal", "speaker": "librispeech_4507", "source_utterance_id": "4507-16021-0016", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4507-16021-0016.wav", "source_file_sha256": "e88b190d01b5dd67996e066cab9e656ef03f616d3067aeb9f9ac68783f130755", "decoded_pcm_sha256": "0f759c362b7bb3f7ab6547020834b6f41d471a7cfe5c2018d0a8cc506e042103", "original_source_sha256": "e3cf855f3ed817d96563b7914780df19ad9175cd7deb4a04880dfb6093dcccf7", "corpus_entry_id": "librispeech_4507-16021-0016"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.2654337882995605, "predicted_mos": 3.2654337882995605, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ced7aaa55f66f55c9b8baa0204ce5bc103752bffba5ba59c179c646a6faf5360", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7256014730010065, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4507-16021-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ced7aaa55f66f55c9b8baa0204ce5bc103752bffba5ba59c179c646a6faf5360", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "bf03704b4a0939f52cf691a192ebddece3d996e02ed4c4bccd90a88cb5b78db0"}, "rows": [{"case": "librispeech_4507-16021-0016", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.860515832901001, "predicted_mos": 2.860515832901001, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bd221e591fd9fea5045d9127b8e395fb609c7902a0e54f4f3082f4830ee8b7f4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.13849158100129, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4507-16021-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "bd221e591fd9fea5045d9127b8e395fb609c7902a0e54f4f3082f4830ee8b7f4", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.838831521557922, "全帯域_10ms_rms_envelope_nrmse": 0.06191369594871921, "全帯域_10ms_envelope_correlation": 0.9964195043711009, "全帯域_10ms_modulation_mae_db": 0.4971920114962572, "全帯域_40ms_log_rms_mae_db": 0.44897197463956806, "全帯域_40ms_rms_envelope_nrmse": 0.03277499197009555, "全帯域_40ms_envelope_correlation": 0.9991453981719729, "全帯域_40ms_modulation_mae_db": 0.5343224440808723, "300–1000 Hz_10ms_log_rms_mae_db": 1.5776671285646693, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08175455926581905, "300–1000 Hz_10ms_envelope_correlation": 0.9945516868836927, "300–1000 Hz_10ms_modulation_mae_db": 0.6035350307686415, "300–1000 Hz_40ms_log_rms_mae_db": 0.8184153224885933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04646628374083927, "300–1000 Hz_40ms_envelope_correlation": 0.9984377775505101, "300–1000 Hz_40ms_modulation_mae_db": 0.6550584259025912, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4204548402879567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13886946454269147, "1000–3000 Hz_10ms_envelope_correlation": 0.984794211568571, "1000–3000 Hz_10ms_modulation_mae_db": 0.4735901476800773, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7842038568249698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07831064405500593, "1000–3000 Hz_40ms_envelope_correlation": 0.9950385764077252, "1000–3000 Hz_40ms_modulation_mae_db": 0.529750437981905, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3631975559416836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18830701526668678, "3000–6000 Hz_10ms_envelope_correlation": 0.970652464105377, "3000–6000 Hz_10ms_modulation_mae_db": 1.2501411635795208, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8048311521470541, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10413434402253083, "3000–6000 Hz_40ms_envelope_correlation": 0.9906164254519132, "3000–6000 Hz_40ms_modulation_mae_db": 1.7209779147794089, "mr_stft_db": 5.61899114669166, "band_mae_db": 0.1769121165977714, "rms_gain_error_db": -0.15069694780250972, "output_peak": 0.3548189989605716}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8021448683992501, "全帯域_10ms_rms_envelope_nrmse": 0.06736265630973282, "全帯域_10ms_envelope_correlation": 0.995594529796468, "全帯域_10ms_modulation_mae_db": 0.8173005117847912, "全帯域_40ms_log_rms_mae_db": 0.43698709445320394, "全帯域_40ms_rms_envelope_nrmse": 0.035389890799116426, "全帯域_40ms_envelope_correlation": 0.9988098859012117, "全帯域_40ms_modulation_mae_db": 1.2420867824019217, "300–1000 Hz_10ms_log_rms_mae_db": 1.4800754729448942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08076485677182314, "300–1000 Hz_10ms_envelope_correlation": 0.9945514478092037, "300–1000 Hz_10ms_modulation_mae_db": 0.8728020774886881, "300–1000 Hz_40ms_log_rms_mae_db": 0.7339731376600385, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04272004305783118, "300–1000 Hz_40ms_envelope_correlation": 0.9985787865997905, "300–1000 Hz_40ms_modulation_mae_db": 1.234768284323102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3806341445159793, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14975790424289367, "1000–3000 Hz_10ms_envelope_correlation": 0.9822989898191486, "1000–3000 Hz_10ms_modulation_mae_db": 1.0099276096025875, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7812822623770429, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08299916536689869, "1000–3000 Hz_40ms_envelope_correlation": 0.9944667998630168, "1000–3000 Hz_40ms_modulation_mae_db": 1.2509934333723505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3876607907189276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2095725879629103, "3000–6000 Hz_10ms_envelope_correlation": 0.9634184200357316, "3000–6000 Hz_10ms_modulation_mae_db": 0.9122486798810953, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8032509569044248, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10100810937841882, "3000–6000 Hz_40ms_envelope_correlation": 0.9907499620045721, "3000–6000 Hz_40ms_modulation_mae_db": 1.47469550810375, "mr_stft_db": 5.612081836315955, "band_mae_db": 0.14266240887519452, "rms_gain_error_db": -0.11192125937714706, "output_peak": 0.3584587200657448}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8606807234896213, "全帯域_10ms_rms_envelope_nrmse": 0.06301120521962064, "全帯域_10ms_envelope_correlation": 0.996132909858306, "全帯域_10ms_modulation_mae_db": 0.5205988816051844, "全帯域_40ms_log_rms_mae_db": 0.46991162611454723, "全帯域_40ms_rms_envelope_nrmse": 0.03375121701120196, "全帯域_40ms_envelope_correlation": 0.9988954666371108, "全帯域_40ms_modulation_mae_db": 0.6290905169956927, "300–1000 Hz_10ms_log_rms_mae_db": 1.5686589620577134, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07821301466110639, "300–1000 Hz_10ms_envelope_correlation": 0.9948048484554625, "300–1000 Hz_10ms_modulation_mae_db": 0.7804828964632038, "300–1000 Hz_40ms_log_rms_mae_db": 0.7966174151784605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04407726319670908, "300–1000 Hz_40ms_envelope_correlation": 0.9983356952442572, "300–1000 Hz_40ms_modulation_mae_db": 0.8507767078179286, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3976975511402003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1553735532408447, "1000–3000 Hz_10ms_envelope_correlation": 0.9809050900300594, "1000–3000 Hz_10ms_modulation_mae_db": 0.6104267460572247, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8224651809523988, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09703418988258106, "1000–3000 Hz_40ms_envelope_correlation": 0.99228798362604, "1000–3000 Hz_40ms_modulation_mae_db": 0.7720122103831454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3661791130627061, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20511996971003696, "3000–6000 Hz_10ms_envelope_correlation": 0.9649829708733857, "3000–6000 Hz_10ms_modulation_mae_db": 1.2902559581697985, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7857373923129035, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10159685596766298, "3000–6000 Hz_40ms_envelope_correlation": 0.9906237307182499, "3000–6000 Hz_40ms_modulation_mae_db": 1.8891093223514523, "mr_stft_db": 5.620546216638663, "band_mae_db": 0.20987469525447947, "rms_gain_error_db": -0.09779709713415476, "output_peak": 0.3726315434443948}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7940633110977657, "全帯域_10ms_rms_envelope_nrmse": 0.0602130949016484, "全帯域_10ms_envelope_correlation": 0.9964172355044081, "全帯域_10ms_modulation_mae_db": 0.3425360417194787, "全帯域_40ms_log_rms_mae_db": 0.41227072176343177, "全帯域_40ms_rms_envelope_nrmse": 0.03307383295293588, "全帯域_40ms_envelope_correlation": 0.9988668012427788, "全帯域_40ms_modulation_mae_db": 0.39569534562003, "300–1000 Hz_10ms_log_rms_mae_db": 1.47463811506823, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0739655156099535, "300–1000 Hz_10ms_envelope_correlation": 0.9952772721927251, "300–1000 Hz_10ms_modulation_mae_db": 0.4468992180879974, "300–1000 Hz_40ms_log_rms_mae_db": 0.756738439860903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04024023249889169, "300–1000 Hz_40ms_envelope_correlation": 0.9985211960665554, "300–1000 Hz_40ms_modulation_mae_db": 0.6299699413623682, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3863975108371138, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13330669845617618, "1000–3000 Hz_10ms_envelope_correlation": 0.9861177665413235, "1000–3000 Hz_10ms_modulation_mae_db": 0.4017141262468542, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7942261528742427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08106920305408861, "1000–3000 Hz_40ms_envelope_correlation": 0.994828842771264, "1000–3000 Hz_40ms_modulation_mae_db": 0.3198376761827697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3849535062434892, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18559537530109646, "3000–6000 Hz_10ms_envelope_correlation": 0.9713873194778824, "3000–6000 Hz_10ms_modulation_mae_db": 0.9083749808756414, "3000–6000 Hz_40ms_log_rms_mae_db": 0.809868248283927, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08959507131373007, "3000–6000 Hz_40ms_envelope_correlation": 0.9929571285946338, "3000–6000 Hz_40ms_modulation_mae_db": 1.2106756191358912, "mr_stft_db": 5.605233239985316, "band_mae_db": 0.18742731900931128, "rms_gain_error_db": -0.057211995753404514, "output_peak": 0.37734043352438157}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8093243443061457, "全帯域_10ms_rms_envelope_nrmse": 0.060074267370122204, "全帯域_10ms_envelope_correlation": 0.9964580501551006, "全帯域_10ms_modulation_mae_db": 0.26046454940445374, "全帯域_40ms_log_rms_mae_db": 0.4252618096903768, "全帯域_40ms_rms_envelope_nrmse": 0.0326597239458917, "全帯域_40ms_envelope_correlation": 0.99892387656534, "全帯域_40ms_modulation_mae_db": 0.4669982533923534, "300–1000 Hz_10ms_log_rms_mae_db": 1.5479609479542362, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07646476299023529, "300–1000 Hz_10ms_envelope_correlation": 0.9950517937244086, "300–1000 Hz_10ms_modulation_mae_db": 0.3641592965522604, "300–1000 Hz_40ms_log_rms_mae_db": 0.805613349750095, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04470158657546813, "300–1000 Hz_40ms_envelope_correlation": 0.9982920753825894, "300–1000 Hz_40ms_modulation_mae_db": 0.590009625069605, "1000–3000 Hz_10ms_log_rms_mae_db": 1.384306746700492, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13557018718165423, "1000–3000 Hz_10ms_envelope_correlation": 0.9855025663693405, "1000–3000 Hz_10ms_modulation_mae_db": 0.9235447397876048, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7666718070299039, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07286270212588439, "1000–3000 Hz_40ms_envelope_correlation": 0.9956991210408785, "1000–3000 Hz_40ms_modulation_mae_db": 1.0768324519689028, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3897026469917122, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1876786433792559, "3000–6000 Hz_10ms_envelope_correlation": 0.9707157453832068, "3000–6000 Hz_10ms_modulation_mae_db": 1.044799279656153, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7819225254508839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09639125533307698, "3000–6000 Hz_40ms_envelope_correlation": 0.9917426277835417, "3000–6000 Hz_40ms_modulation_mae_db": 1.3063114107350886, "mr_stft_db": 5.62076466174137, "band_mae_db": 0.17941335651309773, "rms_gain_error_db": -0.07744180207422441, "output_peak": 0.3554831761686961}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8073964073211923, "全帯域_10ms_rms_envelope_nrmse": 0.05914223183217175, "全帯域_10ms_envelope_correlation": 0.9965986389929757, "全帯域_10ms_modulation_mae_db": 0.29059645170323517, "全帯域_40ms_log_rms_mae_db": 0.41404672295189476, "全帯域_40ms_rms_envelope_nrmse": 0.031352660464623115, "全帯域_40ms_envelope_correlation": 0.9990498723598291, "全帯域_40ms_modulation_mae_db": 0.5217753328223624, "300–1000 Hz_10ms_log_rms_mae_db": 1.5305237362573123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0752240394710331, "300–1000 Hz_10ms_envelope_correlation": 0.9952799432392845, "300–1000 Hz_10ms_modulation_mae_db": 0.46479817272453694, "300–1000 Hz_40ms_log_rms_mae_db": 0.7664087078711884, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04201397801713104, "300–1000 Hz_40ms_envelope_correlation": 0.9985900137406674, "300–1000 Hz_40ms_modulation_mae_db": 0.7476466598661197, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3936561878343259, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13694220370953009, "1000–3000 Hz_10ms_envelope_correlation": 0.9852067465032691, "1000–3000 Hz_10ms_modulation_mae_db": 0.5293387466074808, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8031972628201383, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08297962193747888, "1000–3000 Hz_40ms_envelope_correlation": 0.9942303939599484, "1000–3000 Hz_40ms_modulation_mae_db": 0.750143731767777, "3000–6000 Hz_10ms_log_rms_mae_db": 1.388218591239993, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17028534628708406, "3000–6000 Hz_10ms_envelope_correlation": 0.9759373830597445, "3000–6000 Hz_10ms_modulation_mae_db": 1.0649085118203647, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8357655933074133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09081512364661166, "3000–6000 Hz_40ms_envelope_correlation": 0.9924754359621717, "3000–6000 Hz_40ms_modulation_mae_db": 1.2996607432331124, "mr_stft_db": 5.626883853128715, "band_mae_db": 0.20968889437036, "rms_gain_error_db": -0.09210471575958501, "output_peak": 0.3691562597682433}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.807713739962107, "全帯域_10ms_rms_envelope_nrmse": 0.062197688573239485, "全帯域_10ms_envelope_correlation": 0.9962528403347932, "全帯域_10ms_modulation_mae_db": 0.39197799662865257, "全帯域_40ms_log_rms_mae_db": 0.4394731154987593, "全帯域_40ms_rms_envelope_nrmse": 0.03583817496147133, "全帯域_40ms_envelope_correlation": 0.9987534340114846, "全帯域_40ms_modulation_mae_db": 0.334858013565911, "300–1000 Hz_10ms_log_rms_mae_db": 1.536823664529505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0800566770448248, "300–1000 Hz_10ms_envelope_correlation": 0.9946440656987561, "300–1000 Hz_10ms_modulation_mae_db": 0.6146113153945502, "300–1000 Hz_40ms_log_rms_mae_db": 0.8016542228648937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04666800453208887, "300–1000 Hz_40ms_envelope_correlation": 0.9982142935869786, "300–1000 Hz_40ms_modulation_mae_db": 0.7880899629318959, "1000–3000 Hz_10ms_log_rms_mae_db": 1.411497833746002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12941195522791912, "1000–3000 Hz_10ms_envelope_correlation": 0.9867788164467101, "1000–3000 Hz_10ms_modulation_mae_db": 1.1611091230417963, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8043969516811332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0816156590098348, "1000–3000 Hz_40ms_envelope_correlation": 0.9944880482637029, "1000–3000 Hz_40ms_modulation_mae_db": 1.488363186528099, "3000–6000 Hz_10ms_log_rms_mae_db": 1.36118207360726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17658749079095834, "3000–6000 Hz_10ms_envelope_correlation": 0.9742277339797542, "3000–6000 Hz_10ms_modulation_mae_db": 0.27706600781391505, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8068478229015633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09152804055932762, "3000–6000 Hz_40ms_envelope_correlation": 0.9923297500888018, "3000–6000 Hz_40ms_modulation_mae_db": 0.36467545314846966, "mr_stft_db": 5.617079234851111, "band_mae_db": 0.13717458248713174, "rms_gain_error_db": -0.10431991965568957, "output_peak": 0.36339204763714905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8240446113403327, "全帯域_10ms_rms_envelope_nrmse": 0.05982914287847298, "全帯域_10ms_envelope_correlation": 0.9964672102866923, "全帯域_10ms_modulation_mae_db": 0.18298038220783774, "全帯域_40ms_log_rms_mae_db": 0.45452876242842905, "全帯域_40ms_rms_envelope_nrmse": 0.0321907338849586, "全帯域_40ms_envelope_correlation": 0.9989319075941506, "全帯域_40ms_modulation_mae_db": 0.3973142001504917, "300–1000 Hz_10ms_log_rms_mae_db": 1.4749676358125214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07208380245644963, "300–1000 Hz_10ms_envelope_correlation": 0.9955587537685678, "300–1000 Hz_10ms_modulation_mae_db": 0.2884827699543366, "300–1000 Hz_40ms_log_rms_mae_db": 0.741902119593059, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04016909131054008, "300–1000 Hz_40ms_envelope_correlation": 0.9985806629883824, "300–1000 Hz_40ms_modulation_mae_db": 0.5401756346682857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4156677153204054, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12326328349080028, "1000–3000 Hz_10ms_envelope_correlation": 0.9880924846482774, "1000–3000 Hz_10ms_modulation_mae_db": 0.9419471524217152, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8051517437246167, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06947904673593715, "1000–3000 Hz_40ms_envelope_correlation": 0.9961712553458539, "1000–3000 Hz_40ms_modulation_mae_db": 1.2774223708482608, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3844877922771426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21232083718364683, "3000–6000 Hz_10ms_envelope_correlation": 0.9624923285656048, "3000–6000 Hz_10ms_modulation_mae_db": 1.1524084341683807, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7980759427879281, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1066901949484457, "3000–6000 Hz_40ms_envelope_correlation": 0.9896232512793902, "3000–6000 Hz_40ms_modulation_mae_db": 1.3959273800552774, "mr_stft_db": 5.6284213742760585, "band_mae_db": 0.17958911708704853, "rms_gain_error_db": -0.06422788599297267, "output_peak": 0.3720503370276248}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.498337322799208, "spectral_convergence": 0.18922675572998393, "spectral_convergence_status": "ok", "amplitude_floor": 9.54303147153e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.477203198153017, "spectral_convergence": 0.17808337261286708, "spectral_convergence_status": "ok", "amplitude_floor": 7.553612828866807e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.881432919122754, "spectral_convergence": 0.18751512392878458, "spectral_convergence_status": "ok", "amplitude_floor": 6.230060381303045e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.459798737171228e-05, "output_power": 5.959228959250923e-05, "delta_db": 0.3801343801277389, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011571142287648336, "output_power": 0.001127297533845635, "delta_db": -0.11337677127136922, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009701755098987012, "output_power": 0.0009195372612552401, "delta_db": -0.23280975249002472, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.8727504132367385e-05, "output_power": 4.7714833970186926e-05, "delta_db": -0.09120750004720317, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.5070297613490014e-05, "output_power": 2.422123188717889e-05, "delta_db": -0.14963262065401728, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.1228013015272054e-05, "output_power": 3.0557173790547165e-05, "delta_db": -0.09431167499627505, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1769121165977714, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15069694780250972, "pitch": null, "output_rms_dbfs": -26.558199488288675, "output_peak": 0.3548189989605716, "reference_rms_dbfs": -26.407502540486163, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.61899114669166}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014140982422473965, "log_rms_mae_db": 0.838831521557922, "rms_envelope_nrmse": 0.06191369594871921, "envelope_correlation": 0.9964195043711009, "rms_gain_error_db": -0.15069687174886864, "reference_mean_power": 0.0022888204000268467, "output_mean_power": 0.0022107621904950653}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0460573675773498e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006367944533168533, "output_power": 0.0006079684914423596, "reference_active": true, "delta_db": -0.20118199779825602}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.21105688014902e-05, "output_power": 8.716187514345182e-05, "reference_active": true, "delta_db": -0.2398289936739623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2041191998800966e-05, "output_power": 9.453942946272463e-06, "reference_active": true, "delta_db": -1.0505650430165532}], "active_band_count": 3, "active_band_mae_db": 0.4971920114962572}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274322977924003, "log_rms_mae_db": 0.44897197463956806, "rms_envelope_nrmse": 0.03277499197009555, "envelope_correlation": 0.9991453981719729, "rms_gain_error_db": -0.15069702436089993, "reference_mean_power": 0.0023003276499839287, "output_mean_power": 0.0022218769177216355}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.201135437818204e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005685810358795958, "output_power": 0.0005459986708581022, "reference_active": true, "delta_db": -0.1760078490183588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.696081479723381e-05, "output_power": 4.475202271352951e-05, "reference_active": true, "delta_db": -0.2092295411460232}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.36988853805451e-07, "output_power": 7.078829147401154e-07, "reference_active": true, "delta_db": -1.2177299420782348}], "active_band_count": 3, "active_band_mae_db": 0.5343224440808723}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012682596688393775, "log_rms_mae_db": 1.5776671285646693, "rms_envelope_nrmse": 0.08175455926581905, "envelope_correlation": 0.9945516868836927, "rms_gain_error_db": -0.2047302004589283, "reference_mean_power": 0.0016135328798776547, "output_mean_power": 0.0015392345378667514}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.436393406849114e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006358276871928121, "output_power": 0.0006000555811349893, "reference_active": true, "delta_db": -0.25147955908363584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.784078725837931e-05, "output_power": 8.186501966277686e-05, "reference_active": true, "delta_db": -0.30597848944585476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0842784949487312e-05, "output_power": 8.125051827308656e-06, "reference_active": true, "delta_db": -1.2531470437764336}], "active_band_count": 3, "active_band_mae_db": 0.6035350307686415}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011190972053745482, "log_rms_mae_db": 0.8184153224885933, "rms_envelope_nrmse": 0.04646628374083927, "envelope_correlation": 0.9984377775505101, "rms_gain_error_db": -0.2047303496638951, "reference_mean_power": 0.0016216465744125685, "output_mean_power": 0.0015469745679947302}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.451026691855673e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005862792692297138, "output_power": 0.0005548887770969547, "reference_active": true, "delta_db": -0.23898596771439032}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.32096116317955e-05, "output_power": 4.095030591435046e-05, "reference_active": true, "delta_db": -0.23323212434394885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.999199919742325e-07, "output_power": 7.090380354726421e-07, "reference_active": true, "delta_db": -1.4929571856494346}], "active_band_count": 3, "active_band_mae_db": 0.6550584259025912}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.381727808324483e-05, "log_rms_mae_db": 1.4204548402879567, "rms_envelope_nrmse": 0.13886946454269147, "envelope_correlation": 0.984794211568571, "rms_gain_error_db": -0.1482385541166029, "reference_mean_power": 6.869686019984948e-05, "output_mean_power": 6.639158434337604e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.573688907938692e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3586087172622807e-05, "output_power": 1.315743773338637e-05, "reference_active": true, "delta_db": -0.13923073406854242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.445561080668119e-06, "output_power": 2.1629262685010484e-06, "reference_active": true, "delta_db": -0.533367991650149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901755177746097e-07, "output_power": 4.967799025925023e-07, "reference_active": true, "delta_db": -0.7481717173215405}], "active_band_count": 3, "active_band_mae_db": 0.4735901476800773}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.1112752780143365e-05, "log_rms_mae_db": 0.7842038568249698, "rms_envelope_nrmse": 0.07831064405500593, "envelope_correlation": 0.9950385764077252, "rms_gain_error_db": -0.14824724087244073, "reference_mean_power": 6.904136883729842e-05, "output_mean_power": 6.672439876541677e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.335283050026158e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2478970323107004e-05, "output_power": 1.2099811650860382e-05, "reference_active": true, "delta_db": -0.13400141881009042}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3575517045580512e-06, "output_power": 1.161776456860246e-06, "reference_active": true, "delta_db": -0.6763380803651842}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9851801848052634e-08, "output_power": 3.330863080140834e-08, "reference_active": true, "delta_db": -0.7789118147704405}], "active_band_count": 3, "active_band_mae_db": 0.529750437981905}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.653785524000496e-05, "log_rms_mae_db": 1.3631975559416836, "rms_envelope_nrmse": 0.18830701526668678, "envelope_correlation": 0.970652464105377, "rms_gain_error_db": -0.16735365839325414, "reference_mean_power": 1.4641647362050651e-05, "output_mean_power": 1.4088169787227141e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3222330564047269e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.030474467255676e-06, "output_power": 5.371968701983927e-06, "reference_active": true, "delta_db": -0.5021800929035274}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.175765551110617e-06, "output_power": 2.093882959232601e-06, "reference_active": true, "delta_db": -0.16659693818852958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1480409466014962e-06, "output_power": 5.646674219003052e-07, "reference_active": true, "delta_db": -3.0816464596465054}], "active_band_count": 3, "active_band_mae_db": 1.2501411635795208}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.7197117785089643e-05, "log_rms_mae_db": 0.8048311521470541, "rms_envelope_nrmse": 0.10413434402253083, "envelope_correlation": 0.9906164254519132, "rms_gain_error_db": -0.16732425100833728, "reference_mean_power": 1.4714471928104246e-05, "output_mean_power": 1.4158337338805617e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.175274300154406e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.965217773921107e-06, "output_power": 5.1667018832011874e-06, "reference_active": true, "delta_db": -0.6241289957268946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4021233434065406e-06, "output_power": 1.1869700497176846e-06, "reference_active": true, "delta_db": -0.7234645903286248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593565327256392e-08, "output_power": 3.985162375191458e-08, "reference_active": true, "delta_db": -3.815340158282707}], "active_band_count": 3, "active_band_mae_db": 1.7209779147794089}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9004192990041546, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "accf2bf8b60470990c04aedb7e6ee889f4eae7ee7ed0c078296ff920d5ee157b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9849620839908375, "standard_deviation": 0.0023411523387154797, "minimum": 0.9809050900300594, "maximum": 0.9880924846482774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3987890662978095, "standard_deviation": 0.015288812832036246, "minimum": 1.3806341445159793, "maximum": 1.4204548402879567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7564497989306675, "standard_deviation": 0.28512636254904344, "minimum": 0.4017141262468542, "maximum": 1.1611091230417963, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1378119062615637, "standard_deviation": 0.010423441494592728, "minimum": 0.12326328349080028, "maximum": 0.1553735532408447, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946513776598037, "standard_deviation": 0.0011596067805441852, "minimum": 0.99228798362604, "maximum": 0.9961712553458539, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7951994022855557, "standard_deviation": 0.017404400852081267, "minimum": 0.7666718070299039, "maximum": 0.8224651809523988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9331694373791513, "standard_deviation": 0.4047563878576076, "minimum": 0.3198376761827697, "maximum": 1.488363186528099, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08079377902096369, "standard_deviation": 0.008204833374432674, "minimum": 0.06947904673593715, "maximum": 0.09703418988258106, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9692267956850859, "standard_deviation": 0.005015106544191997, "minimum": 0.9624923285656048, "maximum": 0.9759373830597445, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3781977587603642, "standard_deviation": 0.012343422925268398, "minimum": 1.36118207360726, "maximum": 1.3897026469917122, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9875253769956087, "standard_deviation": 0.3193188307626839, "minimum": 0.27706600781391505, "maximum": 1.2902559581697985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19193340823520944, "standard_deviation": 0.015474557477422069, "minimum": 0.17028534628708406, "maximum": 0.21232083718364683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9913897889854093, "standard_deviation": 0.0011562362189433465, "minimum": 0.9896232512793902, "maximum": 0.9929571285946338, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8032874542620123, "standard_deviation": 0.016495980615396724, "minimum": 0.7819225254508839, "maximum": 0.8357655933074133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3327541689428064, "standard_deviation": 0.45328340212665913, "minimum": 0.36467545314846966, "maximum": 1.8891093223514523, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09771987439622559, "standard_deviation": 0.006560134904123529, "minimum": 0.08959507131373007, "maximum": 0.1066901949484457, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949649764715126, "standard_deviation": 0.0003830852832564362, "minimum": 0.9945514478092037, "maximum": 0.9955587537685678, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5239144578986352, "standard_deviation": 0.042132088336685025, "minimum": 1.47463811506823, "maximum": 1.5776671285646693, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5544713471792768, "standard_deviation": 0.20178537965421317, "minimum": 0.2884827699543366, "maximum": 0.8728020774886881, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0773159035339056, "standard_deviation": 0.00345613135011617, "minimum": 0.07208380245644963, "maximum": 0.08175455926581905, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984438126449664, "standard_deviation": 0.00014723059113874035, "minimum": 0.9982142935869786, "maximum": 0.9985900137406674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7776653394084039, "standard_deviation": 0.03191274571164853, "minimum": 0.7339731376600385, "maximum": 0.8184153224885933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.754561905242737, "standard_deviation": 0.2202240520795795, "minimum": 0.5401756346682857, "maximum": 1.234768284323102, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.043382060366187414, "standard_deviation": 0.002536546809270457, "minimum": 0.04016909131054008, "maximum": 0.04666800453208887, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17784281127429935, "standard_deviation": 0.026806250131989326, "minimum": 0.13717458248713174, "maximum": 0.20987469525447947, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.618750195453606, "standard_deviation": 0.00753086144823938, "minimum": 5.605233239985316, "maximum": 5.6284213742760585, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3654164395746008, "standard_deviation": 0.008583353482207243, "minimum": 0.3548189989605716, "maximum": 0.37734043352438157, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09446520294371097, "standard_deviation": 0.029694648040381605, "minimum": -0.15069694780250972, "maximum": -0.057211995753404514, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962926149124806, "standard_deviation": 0.00031567272365955217, "minimum": 0.995594529796468, "maximum": 0.9965986389929757, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8180249409342921, "standard_deviation": 0.022071278675278735, "minimum": 0.7940633110977657, "maximum": 0.8606807234896213, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41295585331873635, "standard_deviation": 0.1995317780977907, "minimum": 0.18298038220783774, "maximum": 0.8173005117847912, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06171799787921594, "standard_deviation": 0.002644992082628617, "minimum": 0.05914223183217175, "maximum": 0.06736265630973282, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989220803104848, "standard_deviation": 0.00012597885295697583, "minimum": 0.9987534340114846, "maximum": 0.9991453981719729, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.43768147844252636, "standard_deviation": 0.02004401309804588, "minimum": 0.41227072176343177, "maximum": 0.46991162611454723, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5652676111287044, "standard_deviation": 0.28901687700599515, "minimum": 0.334858013565911, "maximum": 1.2420867824019217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033378903248786826, "standard_deviation": 0.0015456440020376082, "minimum": 0.031352660464623115, "maximum": 0.03583817496147133, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0016", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.988079071044922, "predicted_mos": 2.988079071044922, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b6a8bb4e79148851198df2c0cf0a16587427ba34320b892cd783f859413ea75b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.855291800107807, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8425112278109603, "全帯域_10ms_rms_envelope_nrmse": 0.06169291753110598, "全帯域_10ms_envelope_correlation": 0.9964220896339301, "全帯域_10ms_modulation_mae_db": 0.4594950238394023, "全帯域_40ms_log_rms_mae_db": 0.45264901110823563, "全帯域_40ms_rms_envelope_nrmse": 0.03171884519076587, "全帯域_40ms_envelope_correlation": 0.9991897589830053, "全帯域_40ms_modulation_mae_db": 0.4432429310841555, "300–1000 Hz_10ms_log_rms_mae_db": 1.5742206204878368, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07991387331111614, "300–1000 Hz_10ms_envelope_correlation": 0.9947513235297116, "300–1000 Hz_10ms_modulation_mae_db": 0.5897996069378401, "300–1000 Hz_40ms_log_rms_mae_db": 0.7953351732073839, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04338287846604039, "300–1000 Hz_40ms_envelope_correlation": 0.9986245317419865, "300–1000 Hz_40ms_modulation_mae_db": 0.654682165642751, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4077573937298322, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13456949595813653, "1000–3000 Hz_10ms_envelope_correlation": 0.9858460543951271, "1000–3000 Hz_10ms_modulation_mae_db": 0.44915216612727366, "1000–3000 Hz_40ms_log_rms_mae_db": 0.775141817893556, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07588981143883124, "1000–3000 Hz_40ms_envelope_correlation": 0.9955281861904237, "1000–3000 Hz_40ms_modulation_mae_db": 0.528914201848294, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2848232834092732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16549970562715813, "3000–6000 Hz_10ms_envelope_correlation": 0.9782302140030545, "3000–6000 Hz_10ms_modulation_mae_db": 1.3821282368672136, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7446247561005144, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09766979908757155, "3000–6000 Hz_40ms_envelope_correlation": 0.993044741858416, "3000–6000 Hz_40ms_modulation_mae_db": 1.7633534658661436, "mr_stft_db": 5.546670585798608, "band_mae_db": 0.20812078417584912, "rms_gain_error_db": -0.14331670695464327, "output_peak": 0.3556914298221382}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8023727201375077, "全帯域_10ms_rms_envelope_nrmse": 0.06649050213193956, "全帯域_10ms_envelope_correlation": 0.9957140984628471, "全帯域_10ms_modulation_mae_db": 0.8360220904680147, "全帯域_40ms_log_rms_mae_db": 0.4247703886783349, "全帯域_40ms_rms_envelope_nrmse": 0.034912036418152624, "全帯域_40ms_envelope_correlation": 0.9988468834069976, "全帯域_40ms_modulation_mae_db": 1.2469683977642665, "300–1000 Hz_10ms_log_rms_mae_db": 1.4753724675985531, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08168486630052438, "300–1000 Hz_10ms_envelope_correlation": 0.9944056724411742, "300–1000 Hz_10ms_modulation_mae_db": 0.9154403077852127, "300–1000 Hz_40ms_log_rms_mae_db": 0.7162122995399738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042625806908153166, "300–1000 Hz_40ms_envelope_correlation": 0.9985657498798508, "300–1000 Hz_40ms_modulation_mae_db": 1.2717421173793186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3835388592238782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14552755763261682, "1000–3000 Hz_10ms_envelope_correlation": 0.9834487921796058, "1000–3000 Hz_10ms_modulation_mae_db": 1.021532773102732, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8026866149783501, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08091179083849097, "1000–3000 Hz_40ms_envelope_correlation": 0.9949694759545503, "1000–3000 Hz_40ms_modulation_mae_db": 1.3095991937182672, "3000–6000 Hz_10ms_log_rms_mae_db": 1.310347962804205, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17691827300271545, "3000–6000 Hz_10ms_envelope_correlation": 0.9743566025490423, "3000–6000 Hz_10ms_modulation_mae_db": 1.0925837099195965, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7882090617637847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09479908895028112, "3000–6000 Hz_40ms_envelope_correlation": 0.9927699691297344, "3000–6000 Hz_40ms_modulation_mae_db": 1.5640825873921766, "mr_stft_db": 5.543166553042609, "band_mae_db": 0.1722805672444524, "rms_gain_error_db": -0.11162119951387, "output_peak": 0.36491740522580435}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8738841503397371, "全帯域_10ms_rms_envelope_nrmse": 0.06482029940272954, "全帯域_10ms_envelope_correlation": 0.995890118348082, "全帯域_10ms_modulation_mae_db": 0.5371580709708822, "全帯域_40ms_log_rms_mae_db": 0.4782221524786947, "全帯域_40ms_rms_envelope_nrmse": 0.03446239221892382, "全帯域_40ms_envelope_correlation": 0.9988297611605025, "全帯域_40ms_modulation_mae_db": 0.6582514357627899, "300–1000 Hz_10ms_log_rms_mae_db": 1.5669438629087322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08037571408259955, "300–1000 Hz_10ms_envelope_correlation": 0.9944788314547593, "300–1000 Hz_10ms_modulation_mae_db": 0.8574172175401836, "300–1000 Hz_40ms_log_rms_mae_db": 0.805485873708108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044253322736018466, "300–1000 Hz_40ms_envelope_correlation": 0.998285751242893, "300–1000 Hz_40ms_modulation_mae_db": 0.9552169036717123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3565716704719524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14207568086635255, "1000–3000 Hz_10ms_envelope_correlation": 0.9843651948619996, "1000–3000 Hz_10ms_modulation_mae_db": 0.744450780203215, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8224934693761643, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08953541497812419, "1000–3000 Hz_40ms_envelope_correlation": 0.9938812975790656, "1000–3000 Hz_40ms_modulation_mae_db": 1.015224763786269, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3009745888914985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18659297722180954, "3000–6000 Hz_10ms_envelope_correlation": 0.971235806043668, "3000–6000 Hz_10ms_modulation_mae_db": 1.4968836719578669, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7825327156986956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10672918741436056, "3000–6000 Hz_40ms_envelope_correlation": 0.9902774326917557, "3000–6000 Hz_40ms_modulation_mae_db": 2.041373440220572, "mr_stft_db": 5.548033650186505, "band_mae_db": 0.2541308834035976, "rms_gain_error_db": -0.09321057074180616, "output_peak": 0.37199027280974156}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8166948695469614, "全帯域_10ms_rms_envelope_nrmse": 0.062163835142342125, "全帯域_10ms_envelope_correlation": 0.9961756255434779, "全帯域_10ms_modulation_mae_db": 0.3802079847686633, "全帯域_40ms_log_rms_mae_db": 0.4281872736817024, "全帯域_40ms_rms_envelope_nrmse": 0.03480749379547974, "全帯域_40ms_envelope_correlation": 0.9987360688562466, "全帯域_40ms_modulation_mae_db": 0.4928917463501857, "300–1000 Hz_10ms_log_rms_mae_db": 1.4962813453761346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0771838703566454, "300–1000 Hz_10ms_envelope_correlation": 0.9948456164418297, "300–1000 Hz_10ms_modulation_mae_db": 0.4870693506379187, "300–1000 Hz_40ms_log_rms_mae_db": 0.7660916001459069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042831243785314985, "300–1000 Hz_40ms_envelope_correlation": 0.9983058925073132, "300–1000 Hz_40ms_modulation_mae_db": 0.7470368642010329, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3773270142656993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13105425812077978, "1000–3000 Hz_10ms_envelope_correlation": 0.9867549690080867, "1000–3000 Hz_10ms_modulation_mae_db": 0.2739680314352159, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8086525496607612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08282747949197578, "1000–3000 Hz_40ms_envelope_correlation": 0.9948044935337247, "1000–3000 Hz_40ms_modulation_mae_db": 0.5431218741663991, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3139939479978264, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1685579746570286, "3000–6000 Hz_10ms_envelope_correlation": 0.9770124836758093, "3000–6000 Hz_10ms_modulation_mae_db": 1.0815402486858596, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8018179319722379, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08846021586230195, "3000–6000 Hz_40ms_envelope_correlation": 0.9941582089591733, "3000–6000 Hz_40ms_modulation_mae_db": 1.4582193558582024, "mr_stft_db": 5.536620050078948, "band_mae_db": 0.22081202522858093, "rms_gain_error_db": -0.05495447019399058, "output_peak": 0.37774263446230444}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8319467291772834, "全帯域_10ms_rms_envelope_nrmse": 0.06303484575948129, "全帯域_10ms_envelope_correlation": 0.996114069204931, "全帯域_10ms_modulation_mae_db": 0.280109397779777, "全帯域_40ms_log_rms_mae_db": 0.44137091964336744, "全帯域_40ms_rms_envelope_nrmse": 0.03508200932670461, "全帯域_40ms_envelope_correlation": 0.99877331526913, "全帯域_40ms_modulation_mae_db": 0.4679391435446592, "300–1000 Hz_10ms_log_rms_mae_db": 1.5818515559734745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08018204371127287, "300–1000 Hz_10ms_envelope_correlation": 0.9945640983173796, "300–1000 Hz_10ms_modulation_mae_db": 0.3653596645396974, "300–1000 Hz_40ms_log_rms_mae_db": 0.7875615749533252, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047051925845849364, "300–1000 Hz_40ms_envelope_correlation": 0.9981157507195213, "300–1000 Hz_40ms_modulation_mae_db": 0.6055808560053184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3694277577354819, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13216783655832354, "1000–3000 Hz_10ms_envelope_correlation": 0.9864354073147925, "1000–3000 Hz_10ms_modulation_mae_db": 1.0178147067497407, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7753717330433934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07301963307043785, "1000–3000 Hz_40ms_envelope_correlation": 0.9959747938367459, "1000–3000 Hz_40ms_modulation_mae_db": 1.2183714725066483, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2891160875961292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16807775373749856, "3000–6000 Hz_10ms_envelope_correlation": 0.9770631126458473, "3000–6000 Hz_10ms_modulation_mae_db": 1.036095257506742, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7404107102724689, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0938059212766848, "3000–6000 Hz_40ms_envelope_correlation": 0.9930849564893242, "3000–6000 Hz_40ms_modulation_mae_db": 1.3902951961943817, "mr_stft_db": 5.546563932464621, "band_mae_db": 0.22639734710048298, "rms_gain_error_db": -0.0899495241794166, "output_peak": 0.3561959221498338}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8069691253818213, "全帯域_10ms_rms_envelope_nrmse": 0.059268597766794565, "全帯域_10ms_envelope_correlation": 0.9965680540171308, "全帯域_10ms_modulation_mae_db": 0.21171906432480184, "全帯域_40ms_log_rms_mae_db": 0.41441720696485734, "全帯域_40ms_rms_envelope_nrmse": 0.03234165205771742, "全帯域_40ms_envelope_correlation": 0.9989644324452561, "全帯域_40ms_modulation_mae_db": 0.4339370770440489, "300–1000 Hz_10ms_log_rms_mae_db": 1.5146633544262396, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07620032850100987, "300–1000 Hz_10ms_envelope_correlation": 0.9951113222435698, "300–1000 Hz_10ms_modulation_mae_db": 0.3685199319690681, "300–1000 Hz_40ms_log_rms_mae_db": 0.7356928647890847, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04314569168040519, "300–1000 Hz_40ms_envelope_correlation": 0.998453260662974, "300–1000 Hz_40ms_modulation_mae_db": 0.6608626815100561, "1000–3000 Hz_10ms_log_rms_mae_db": 1.355801925691841, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12625695872382905, "1000–3000 Hz_10ms_envelope_correlation": 0.9874400659617866, "1000–3000 Hz_10ms_modulation_mae_db": 0.704729145102594, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7844125875858959, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07691810564108231, "1000–3000 Hz_40ms_envelope_correlation": 0.9951343885255236, "1000–3000 Hz_40ms_modulation_mae_db": 0.9611534714330666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3455185269091363, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15449441515300136, "3000–6000 Hz_10ms_envelope_correlation": 0.9805613835867781, "3000–6000 Hz_10ms_modulation_mae_db": 0.9017267777762467, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8377492893898119, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09142601996636485, "3000–6000 Hz_40ms_envelope_correlation": 0.9930671269362874, "3000–6000 Hz_40ms_modulation_mae_db": 1.0501900566933813, "mr_stft_db": 5.5659400505620455, "band_mae_db": 0.22634538585848518, "rms_gain_error_db": -0.08563735883992758, "output_peak": 0.3684649090481876}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8146643947866385, "全帯域_10ms_rms_envelope_nrmse": 0.06244961975546093, "全帯域_10ms_envelope_correlation": 0.9962335621617318, "全帯域_10ms_modulation_mae_db": 0.4339914262585059, "全帯域_40ms_log_rms_mae_db": 0.43824775201084737, "全帯域_40ms_rms_envelope_nrmse": 0.03588349977651211, "全帯域_40ms_envelope_correlation": 0.9987650962008302, "全帯域_40ms_modulation_mae_db": 0.3752924463954825, "300–1000 Hz_10ms_log_rms_mae_db": 1.5397848727719732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08074637813779341, "300–1000 Hz_10ms_envelope_correlation": 0.9945475145895071, "300–1000 Hz_10ms_modulation_mae_db": 0.6636042728778816, "300–1000 Hz_40ms_log_rms_mae_db": 0.7971419181413504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04679046639641709, "300–1000 Hz_40ms_envelope_correlation": 0.9982036206478138, "300–1000 Hz_40ms_modulation_mae_db": 0.7889772359612645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3501094442733892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12483508992221885, "1000–3000 Hz_10ms_envelope_correlation": 0.9878367770909298, "1000–3000 Hz_10ms_modulation_mae_db": 1.0994019814839717, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8011955079183302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08013185900926824, "1000–3000 Hz_40ms_envelope_correlation": 0.9948902951247709, "1000–3000 Hz_40ms_modulation_mae_db": 1.3808772716465316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.320661165239754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16568751263037718, "3000–6000 Hz_10ms_envelope_correlation": 0.977513428286219, "3000–6000 Hz_10ms_modulation_mae_db": 0.24047610072627926, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8003062216466561, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08968118133909557, "3000–6000 Hz_40ms_envelope_correlation": 0.9932920743706737, "3000–6000 Hz_40ms_modulation_mae_db": 0.47291385650774226, "mr_stft_db": 5.5433276422245585, "band_mae_db": 0.17823249163466012, "rms_gain_error_db": -0.10962514334465728, "output_peak": 0.36278735901635456}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8397289776711888, "全帯域_10ms_rms_envelope_nrmse": 0.0627128616621752, "全帯域_10ms_envelope_correlation": 0.9961095533228831, "全帯域_10ms_modulation_mae_db": 0.17146616960425598, "全帯域_40ms_log_rms_mae_db": 0.4686806041898176, "全帯域_40ms_rms_envelope_nrmse": 0.03328935318174772, "全帯域_40ms_envelope_correlation": 0.9988480491586562, "全帯域_40ms_modulation_mae_db": 0.376091325897858, "300–1000 Hz_10ms_log_rms_mae_db": 1.497334321428417, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0756124310863686, "300–1000 Hz_10ms_envelope_correlation": 0.995073737366378, "300–1000 Hz_10ms_modulation_mae_db": 0.2567115062552002, "300–1000 Hz_40ms_log_rms_mae_db": 0.7476808977575723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041285956919133934, "300–1000 Hz_40ms_envelope_correlation": 0.998456178121554, "300–1000 Hz_40ms_modulation_mae_db": 0.47485217551456355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3699047007969556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12121992431864798, "1000–3000 Hz_10ms_envelope_correlation": 0.9886569797707312, "1000–3000 Hz_10ms_modulation_mae_db": 1.1594909482010367, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7997255589451308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07029965305812297, "1000–3000 Hz_40ms_envelope_correlation": 0.9963088785729209, "1000–3000 Hz_40ms_modulation_mae_db": 1.5103914844061403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3357373791512919, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19062109579335507, "3000–6000 Hz_10ms_envelope_correlation": 0.9700144374013702, "3000–6000 Hz_10ms_modulation_mae_db": 1.0738958063800659, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7767156702836926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10252235791463166, "3000–6000 Hz_40ms_envelope_correlation": 0.9912408817686499, "3000–6000 Hz_40ms_modulation_mae_db": 1.4507871816616806, "mr_stft_db": 5.549197301571964, "band_mae_db": 0.21211997099474833, "rms_gain_error_db": -0.06030647144202921, "output_peak": 0.3693065431888697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.425716508825052, "spectral_convergence": 0.18808005699373592, "spectral_convergence_status": "ok", "amplitude_floor": 9.54303147153e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.399987392722147, "spectral_convergence": 0.1767607321539435, "spectral_convergence_status": "ok", "amplitude_floor": 7.553612828866807e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8143078558486225, "spectral_convergence": 0.18676463329987458, "spectral_convergence_status": "ok", "amplitude_floor": 6.230060381303045e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.459798737171228e-05, "output_power": 5.929431212362253e-05, "delta_db": 0.3583640141506854, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011571142287648336, "output_power": 0.001130674657073453, "delta_db": -0.10038576029196862, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009701755098987012, "output_power": 0.0009218697971831243, "delta_db": -0.22180720941307336, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.8727504132367385e-05, "output_power": 4.7201411941144944e-05, "delta_db": -0.13819177059610224, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.5070297613490014e-05, "output_power": 2.3634757021831897e-05, "delta_db": -0.2560834768948628, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.1228013015272054e-05, "output_power": 3.000233833922897e-05, "delta_db": -0.17389247370840216, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20812078417584912, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14331670695464327, "pitch": null, "output_rms_dbfs": -26.55081924744081, "output_peak": 0.3556914298221382, "reference_rms_dbfs": -26.407502540486163, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.546670585798608}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014140982422473965, "log_rms_mae_db": 0.8425112278109603, "rms_envelope_nrmse": 0.06169291753110598, "envelope_correlation": 0.9964220896339301, "rms_gain_error_db": -0.14331663692607968, "reference_mean_power": 0.0022888204000268467, "output_mean_power": 0.002214522269424585}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0460573675773498e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006367944533168533, "output_power": 0.0006089558435993, "reference_active": true, "delta_db": -0.1941346954278635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.21105688014902e-05, "output_power": 8.801419533638136e-05, "reference_active": true, "delta_db": -0.19756741347454915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2041191998800966e-05, "output_power": 9.593811586713776e-06, "reference_active": true, "delta_db": -0.9867829626157941}], "active_band_count": 3, "active_band_mae_db": 0.4594950238394023}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274322977924003, "log_rms_mae_db": 0.45264901110823563, "rms_envelope_nrmse": 0.03171884519076587, "envelope_correlation": 0.9991897589830053, "rms_gain_error_db": -0.14331684913059511, "reference_mean_power": 0.0023003276499839287, "output_mean_power": 0.0022256558701128737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.201135437818204e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005685810358795958, "output_power": 0.0005465838883783623, "reference_active": true, "delta_db": -0.17135544519003273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.696081479723381e-05, "output_power": 4.531394907909935e-05, "reference_active": true, "delta_db": -0.155037112746743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.36988853805451e-07, "output_power": 7.437051709417889e-07, "reference_active": true, "delta_db": -1.003336235315691}], "active_band_count": 3, "active_band_mae_db": 0.4432429310841555}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012682596688393775, "log_rms_mae_db": 1.5742206204878368, "rms_envelope_nrmse": 0.07991387331111614, "envelope_correlation": 0.9947513235297116, "rms_gain_error_db": -0.18725978286267028, "reference_mean_power": 0.0016135328798776547, "output_mean_power": 0.0015454389064223245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.436393406849114e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006358276871928121, "output_power": 0.0006015997315920402, "reference_active": true, "delta_db": -0.24031801575674114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.784078725837931e-05, "output_power": 8.158926941922183e-05, "reference_active": true, "delta_db": -0.3206317510326406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0842784949487312e-05, "output_power": 8.209107424202684e-06, "reference_active": true, "delta_db": -1.2084490540241384}], "active_band_count": 3, "active_band_mae_db": 0.5897996069378401}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011190972053745482, "log_rms_mae_db": 0.7953351732073839, "rms_envelope_nrmse": 0.04338287846604039, "envelope_correlation": 0.9986245317419865, "rms_gain_error_db": -0.1872600515241193, "reference_mean_power": 0.0016216465744125685, "output_mean_power": 0.0015532100924518734}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.451026691855673e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005862792692297138, "output_power": 0.0005562200929612382, "reference_active": true, "delta_db": -0.2285786453394314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.32096116317955e-05, "output_power": 4.081972944012497e-05, "reference_active": true, "delta_db": -0.2471024107341268}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.999199919742325e-07, "output_power": 7.097880895417995e-07, "reference_active": true, "delta_db": -1.4883654408546945}], "active_band_count": 3, "active_band_mae_db": 0.654682165642751}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.381727808324483e-05, "log_rms_mae_db": 1.4077573937298322, "rms_envelope_nrmse": 0.13456949595813653, "envelope_correlation": 0.9858460543951271, "rms_gain_error_db": -0.20416950601168618, "reference_mean_power": 6.869686019984948e-05, "output_mean_power": 6.55420373920896e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.573688907938692e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3586087172622807e-05, "output_power": 1.3182248220758855e-05, "reference_active": true, "delta_db": -0.1310491171998311}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.445561080668119e-06, "output_power": 2.229798036648123e-06, "reference_active": true, "delta_db": -0.401129855664806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901755177746097e-07, "output_power": 4.891628167873328e-07, "reference_active": true, "delta_db": -0.8152775255171839}], "active_band_count": 3, "active_band_mae_db": 0.44915216612727366}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.1112752780143365e-05, "log_rms_mae_db": 0.775141817893556, "rms_envelope_nrmse": 0.07588981143883124, "envelope_correlation": 0.9955281861904237, "rms_gain_error_db": -0.20417878940508438, "reference_mean_power": 6.904136883729842e-05, "output_mean_power": 6.587058407023358e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.335283050026158e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2478970323107004e-05, "output_power": 1.215479705390647e-05, "reference_active": true, "delta_db": -0.11431039981428492}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3575517045580512e-06, "output_power": 1.1792606056165117e-06, "reference_active": true, "delta_db": -0.6114658864622222}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9851801848052634e-08, "output_power": 3.26852138154453e-08, "reference_active": true, "delta_db": -0.8609663192683747}], "active_band_count": 3, "active_band_mae_db": 0.528914201848294}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.653785524000496e-05, "log_rms_mae_db": 1.2848232834092732, "rms_envelope_nrmse": 0.16549970562715813, "envelope_correlation": 0.9782302140030545, "rms_gain_error_db": -0.4050605097450102, "reference_mean_power": 1.4641647362050651e-05, "output_mean_power": 1.3337790509320668e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3222330564047269e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.030474467255676e-06, "output_power": 5.150916554851656e-06, "reference_active": true, "delta_db": -0.6846696867820148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.175765551110617e-06, "output_power": 1.94930241856705e-06, "reference_active": true, "delta_db": -0.477328746556417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1480409466014962e-06, "output_power": 5.774558034463845e-07, "reference_active": true, "delta_db": -2.9843862772632095}], "active_band_count": 3, "active_band_mae_db": 1.3821282368672136}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.7197117785089643e-05, "log_rms_mae_db": 0.7446247561005144, "rms_envelope_nrmse": 0.09766979908757155, "envelope_correlation": 0.993044741858416, "rms_gain_error_db": -0.4050324776905153, "reference_mean_power": 1.4714471928104246e-05, "output_mean_power": 1.3404216476299437e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.175274300154406e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.965217773921107e-06, "output_power": 4.964650431946116e-06, "reference_active": true, "delta_db": -0.7973762858973084}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4021233434065406e-06, "output_power": 1.1055814908710525e-06, "reference_active": true, "delta_db": -1.0319546043243146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593565327256392e-08, "output_power": 4.324213083719883e-08, "reference_active": true, "delta_db": -3.460729507376808}], "active_band_count": 3, "active_band_mae_db": 1.7633534658661436}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9060162695392743, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e59275b46e7035194df3443f7db89716360f33296d3f9cd39489919ff25e12e4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9863480300728824, "standard_deviation": 0.001752654650578875, "minimum": 0.9834487921796058, "maximum": 0.9886569797707312, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3713048457736288, "standard_deviation": 0.01861973510128558, "minimum": 1.3501094442733892, "maximum": 1.4077573937298322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8088175665507225, "standard_deviation": 0.32221097672383625, "minimum": 0.2739680314352159, "maximum": 1.1594909482010367, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13221335026261313, "standard_deviation": 0.008385390452254844, "minimum": 0.12121992431864798, "maximum": 0.14552755763261682, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951864761647157, "standard_deviation": 0.0007545031432537952, "minimum": 0.9938812975790656, "maximum": 0.9963088785729209, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7962099799251978, "standard_deviation": 0.01665652766368108, "minimum": 0.775141817893556, "maximum": 0.8224934693761643, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0584567166889522, "standard_deviation": 0.3692013619495882, "minimum": 0.528914201848294, "maximum": 1.5103914844061403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07869171844079169, "standard_deviation": 0.006039024275738861, "minimum": 0.07029965305812297, "maximum": 0.08953541497812419, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9757484335239737, "standard_deviation": 0.0036044584478673535, "minimum": 0.9700144374013702, "maximum": 0.9805613835867781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3126466177498894, "standard_deviation": 0.021210300056851556, "minimum": 1.2848232834092732, "maximum": 1.3455185269091363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0381662262274838, "standard_deviation": 0.376028167516641, "minimum": 0.24047610072627926, "maximum": 1.4968836719578669, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17205621347786798, "standard_deviation": 0.01195015063381525, "minimum": 0.15449441515300136, "maximum": 0.19062109579335507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926169240255018, "standard_deviation": 0.0012436287444883634, "minimum": 0.9902774326917557, "maximum": 0.9941582089591733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7840457946409828, "standard_deviation": 0.031613079326525816, "minimum": 0.7404107102724689, "maximum": 0.8377492893898119, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.398901892549285, "standard_deviation": 0.47180774051497903, "minimum": 0.47291385650774226, "maximum": 2.041373440220572, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0956367214764115, "standard_deviation": 0.0063621913374502905, "minimum": 0.08846021586230195, "maximum": 0.10672918741436056, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9947222645480387, "standard_deviation": 0.00026886095295344873, "minimum": 0.9944056724411742, "maximum": 0.9951113222435698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.53080655012142, "standard_deviation": 0.040572351440900226, "minimum": 1.4753724675985531, "maximum": 1.5818515559734745, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5629902323178753, "standard_deviation": 0.2385064870040438, "minimum": 0.2567115062552002, "maximum": 0.9154403077852127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07898743818591628, "standard_deviation": 0.002298887084407623, "minimum": 0.0756124310863686, "maximum": 0.08168486630052438, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983763419404883, "standard_deviation": 0.00017755243782813468, "minimum": 0.9981157507195213, "maximum": 0.9986245317419865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7689002752803382, "standard_deviation": 0.032776406748325236, "minimum": 0.7162122995399738, "maximum": 0.805485873708108, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7698688749857522, "standard_deviation": 0.24677717181896416, "minimum": 0.47485217551456355, "maximum": 1.2717421173793186, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04392091159216657, "standard_deviation": 0.002029996339092341, "minimum": 0.041285956919133934, "maximum": 0.047051925845849364, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21230493195510708, "standard_deviation": 0.026720600998520704, "minimum": 0.1722805672444524, "maximum": 0.2541308834035976, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.547439970741232, "standard_deviation": 0.008445599398122266, "minimum": 5.536620050078948, "maximum": 5.5659400505620455, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36588705946540423, "standard_deviation": 0.007602249925893402, "minimum": 0.3556914298221382, "maximum": 0.37774263446230444, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09357768065129259, "standard_deviation": 0.028616012293964377, "minimum": -0.14331670695464327, "maximum": -0.05495447019399058, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9961533963368767, "standard_deviation": 0.00027172058428512254, "minimum": 0.9957140984628471, "maximum": 0.9965680540171308, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8285965243565123, "standard_deviation": 0.023515737418196174, "minimum": 0.8023727201375077, "maximum": 0.8738841503397371, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4137711535017879, "standard_deviation": 0.21197992496624204, "minimum": 0.17146616960425598, "maximum": 0.8360220904680147, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06282918489400366, "standard_deviation": 0.0021374635364102513, "minimum": 0.059268597766794565, "maximum": 0.06649050213193956, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988691706850781, "standard_deviation": 0.00014739813328258763, "minimum": 0.9987360688562466, "maximum": 0.9991897589830053, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4433181635944822, "standard_deviation": 0.021994598229496714, "minimum": 0.41441720696485734, "maximum": 0.4782221524786947, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5618268129804308, "standard_deviation": 0.2908102718869516, "minimum": 0.3752924463954825, "maximum": 1.2469683977642665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0340621602457505, "standard_deviation": 0.0014562119010607154, "minimum": 0.03171884519076587, "maximum": 0.03588349977651211, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0016", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5381920337677, "predicted_mos": 2.5381920337677, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c7415114a61f0188e3e5b96e73556930f566ea12f3ab3473a374817bcc89cb44", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.124621499795467, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8325440794701829, "全帯域_10ms_rms_envelope_nrmse": 0.06344824514730546, "全帯域_10ms_envelope_correlation": 0.9960654048184108, "全帯域_10ms_modulation_mae_db": 0.4724616420062066, "全帯域_40ms_log_rms_mae_db": 0.4506005272346971, "全帯域_40ms_rms_envelope_nrmse": 0.03266694348763602, "全帯域_40ms_envelope_correlation": 0.9989578614703776, "全帯域_40ms_modulation_mae_db": 0.48825410263659363, "300–1000 Hz_10ms_log_rms_mae_db": 1.7033176711828184, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07569546970772166, "300–1000 Hz_10ms_envelope_correlation": 0.9951525099410232, "300–1000 Hz_10ms_modulation_mae_db": 0.6310597207965869, "300–1000 Hz_40ms_log_rms_mae_db": 0.8568304748588527, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04154266719019155, "300–1000 Hz_40ms_envelope_correlation": 0.9985420665605879, "300–1000 Hz_40ms_modulation_mae_db": 0.783120191325247, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6163879939050314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15072808638782897, "1000–3000 Hz_10ms_envelope_correlation": 0.98205635773652, "1000–3000 Hz_10ms_modulation_mae_db": 0.478540962881739, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9452749883223934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08480435997069201, "1000–3000 Hz_40ms_envelope_correlation": 0.9941465182824277, "1000–3000 Hz_40ms_modulation_mae_db": 0.9176748917645359, "3000–6000 Hz_10ms_log_rms_mae_db": 1.75570816030348, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23252408788976972, "3000–6000 Hz_10ms_envelope_correlation": 0.9551116065840732, "3000–6000 Hz_10ms_modulation_mae_db": 1.31560772157465, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0562168253188842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12344610233113734, "3000–6000 Hz_40ms_envelope_correlation": 0.9866633246404057, "3000–6000 Hz_40ms_modulation_mae_db": 1.9912899236859332, "mr_stft_db": 5.883550300161821, "band_mae_db": 0.13373844111189528, "rms_gain_error_db": -0.09247377039026337, "output_peak": 0.37253523461825827}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8010609208018044, "全帯域_10ms_rms_envelope_nrmse": 0.055284972570254456, "全帯域_10ms_envelope_correlation": 0.9969859979314479, "全帯域_10ms_modulation_mae_db": 0.3477864266699777, "全帯域_40ms_log_rms_mae_db": 0.4241488108734945, "全帯域_40ms_rms_envelope_nrmse": 0.03072972692335424, "全帯域_40ms_envelope_correlation": 0.9990229300679808, "全帯域_40ms_modulation_mae_db": 0.3581312809544932, "300–1000 Hz_10ms_log_rms_mae_db": 1.625118135956798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0729196399031015, "300–1000 Hz_10ms_envelope_correlation": 0.9954924467264029, "300–1000 Hz_10ms_modulation_mae_db": 0.4363660472901201, "300–1000 Hz_40ms_log_rms_mae_db": 0.8524903435634285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04020243331220649, "300–1000 Hz_40ms_envelope_correlation": 0.9986204317981352, "300–1000 Hz_40ms_modulation_mae_db": 0.5253007895835758, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6205735047940737, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15284031158491163, "1000–3000 Hz_10ms_envelope_correlation": 0.9815907792693807, "1000–3000 Hz_10ms_modulation_mae_db": 0.45441582469682124, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9794273471681738, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08607783363154349, "1000–3000 Hz_40ms_envelope_correlation": 0.9938778804333259, "1000–3000 Hz_40ms_modulation_mae_db": 0.2932391931539009, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7686044558704612, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24783567363998407, "3000–6000 Hz_10ms_envelope_correlation": 0.9497211867914241, "3000–6000 Hz_10ms_modulation_mae_db": 1.247929648020884, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0795481554181219, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1336273656228841, "3000–6000 Hz_40ms_envelope_correlation": 0.9842515156945607, "3000–6000 Hz_40ms_modulation_mae_db": 1.7541951431011462, "mr_stft_db": 5.872546737706409, "band_mae_db": 0.1462605618007403, "rms_gain_error_db": -0.05854101679944085, "output_peak": 0.36494923771046267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8187787999216106, "全帯域_10ms_rms_envelope_nrmse": 0.060663005461488596, "全帯域_10ms_envelope_correlation": 0.9963426574492651, "全帯域_10ms_modulation_mae_db": 0.23626855309961706, "全帯域_40ms_log_rms_mae_db": 0.42444638818318253, "全帯域_40ms_rms_envelope_nrmse": 0.029163704731472454, "全帯域_40ms_envelope_correlation": 0.9990993181075243, "全帯域_40ms_modulation_mae_db": 0.34645110471416335, "300–1000 Hz_10ms_log_rms_mae_db": 1.6687919311257902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07413796169863765, "300–1000 Hz_10ms_envelope_correlation": 0.995269431214404, "300–1000 Hz_10ms_modulation_mae_db": 0.34999316415874676, "300–1000 Hz_40ms_log_rms_mae_db": 0.9048572550805235, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03834673577733335, "300–1000 Hz_40ms_envelope_correlation": 0.998658420016954, "300–1000 Hz_40ms_modulation_mae_db": 0.5852823326305775, "1000–3000 Hz_10ms_log_rms_mae_db": 1.593393091894923, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14267620554121752, "1000–3000 Hz_10ms_envelope_correlation": 0.983912658497592, "1000–3000 Hz_10ms_modulation_mae_db": 0.459199640215616, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8898713639612481, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08104979232093826, "1000–3000 Hz_40ms_envelope_correlation": 0.9945669025855265, "1000–3000 Hz_40ms_modulation_mae_db": 0.8686560660376802, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7371313776766806, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2436465972130518, "3000–6000 Hz_10ms_envelope_correlation": 0.9505754416683015, "3000–6000 Hz_10ms_modulation_mae_db": 1.5478991867410272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.106790334427766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14013344740538625, "3000–6000 Hz_40ms_envelope_correlation": 0.9831714340238726, "3000–6000 Hz_40ms_modulation_mae_db": 2.346857576931904, "mr_stft_db": 5.873743758808362, "band_mae_db": 0.12432449095631015, "rms_gain_error_db": -0.0387234659981904, "output_peak": 0.3602378447242811}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8197370181259388, "全帯域_10ms_rms_envelope_nrmse": 0.06232714220078426, "全帯域_10ms_envelope_correlation": 0.9962738470588421, "全帯域_10ms_modulation_mae_db": 0.6165126307214249, "全帯域_40ms_log_rms_mae_db": 0.4459561155685796, "全帯域_40ms_rms_envelope_nrmse": 0.03416790881716614, "全帯域_40ms_envelope_correlation": 0.9989302418073644, "全帯域_40ms_modulation_mae_db": 0.651397728308702, "300–1000 Hz_10ms_log_rms_mae_db": 1.7475271840596582, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08264435744427676, "300–1000 Hz_10ms_envelope_correlation": 0.9942983584532618, "300–1000 Hz_10ms_modulation_mae_db": 0.6641587102285711, "300–1000 Hz_40ms_log_rms_mae_db": 0.9610273272313423, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047103727564884484, "300–1000 Hz_40ms_envelope_correlation": 0.9982217691647215, "300–1000 Hz_40ms_modulation_mae_db": 0.775765579722033, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6347998841187452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1611593847419972, "1000–3000 Hz_10ms_envelope_correlation": 0.979459456169676, "1000–3000 Hz_10ms_modulation_mae_db": 0.4626501324219537, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0125543438181208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08829635124343449, "1000–3000 Hz_40ms_envelope_correlation": 0.9936624708255326, "1000–3000 Hz_40ms_modulation_mae_db": 0.7704849652408764, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7451128006830847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2382544403653818, "3000–6000 Hz_10ms_envelope_correlation": 0.9532334052168505, "3000–6000 Hz_10ms_modulation_mae_db": 1.174255767739084, "3000–6000 Hz_40ms_log_rms_mae_db": 1.05856152711333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12234133293759286, "3000–6000 Hz_40ms_envelope_correlation": 0.9867392801595389, "3000–6000 Hz_40ms_modulation_mae_db": 1.433245402123504, "mr_stft_db": 5.876565564247229, "band_mae_db": 0.1825576771304375, "rms_gain_error_db": -0.12064536140397375, "output_peak": 0.3626037996375217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8100941286662069, "全帯域_10ms_rms_envelope_nrmse": 0.0647997656185505, "全帯域_10ms_envelope_correlation": 0.9958468140395828, "全帯域_10ms_modulation_mae_db": 0.3950070380154604, "全帯域_40ms_log_rms_mae_db": 0.42444413548695936, "全帯域_40ms_rms_envelope_nrmse": 0.03294053023087083, "全帯域_40ms_envelope_correlation": 0.9988727305573953, "全帯域_40ms_modulation_mae_db": 0.4670236438535351, "300–1000 Hz_10ms_log_rms_mae_db": 1.7093032203746796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07655729496137911, "300–1000 Hz_10ms_envelope_correlation": 0.9950045923482495, "300–1000 Hz_10ms_modulation_mae_db": 0.4425014171053765, "300–1000 Hz_40ms_log_rms_mae_db": 0.883672938243121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04155679606999866, "300–1000 Hz_40ms_envelope_correlation": 0.9985023298017899, "300–1000 Hz_40ms_modulation_mae_db": 0.4742448384789619, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6055143454244762, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1565595000225907, "1000–3000 Hz_10ms_envelope_correlation": 0.9805911501735077, "1000–3000 Hz_10ms_modulation_mae_db": 0.8604367651593469, "1000–3000 Hz_40ms_log_rms_mae_db": 0.912595394394495, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08953615574126633, "1000–3000 Hz_40ms_envelope_correlation": 0.9933503754518423, "1000–3000 Hz_40ms_modulation_mae_db": 1.169887494813809, "3000–6000 Hz_10ms_log_rms_mae_db": 1.713734112520304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23990620418102723, "3000–6000 Hz_10ms_envelope_correlation": 0.9536850983001579, "3000–6000 Hz_10ms_modulation_mae_db": 1.143497223562563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0762797603193108, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12893656196342812, "3000–6000 Hz_40ms_envelope_correlation": 0.9856268280893675, "3000–6000 Hz_40ms_modulation_mae_db": 1.3931520454146928, "mr_stft_db": 5.869197833012257, "band_mae_db": 0.11171632855515719, "rms_gain_error_db": -0.05966249972437138, "output_peak": 0.3754882529914099}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7935775806097788, "全帯域_10ms_rms_envelope_nrmse": 0.06314383785917532, "全帯域_10ms_envelope_correlation": 0.9960581970202362, "全帯域_10ms_modulation_mae_db": 0.3745956633796344, "全帯域_40ms_log_rms_mae_db": 0.4370171045774804, "全帯域_40ms_rms_envelope_nrmse": 0.03505680181265348, "全帯域_40ms_envelope_correlation": 0.9987122994283077, "全帯域_40ms_modulation_mae_db": 0.534235070264776, "300–1000 Hz_10ms_log_rms_mae_db": 1.6155772780446964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07835907543709617, "300–1000 Hz_10ms_envelope_correlation": 0.9947731661130783, "300–1000 Hz_10ms_modulation_mae_db": 0.5527557109216336, "300–1000 Hz_40ms_log_rms_mae_db": 0.8186025480882069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04563565705748041, "300–1000 Hz_40ms_envelope_correlation": 0.9981753454648421, "300–1000 Hz_40ms_modulation_mae_db": 0.7277247814074372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6304806476609544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1568795769563129, "1000–3000 Hz_10ms_envelope_correlation": 0.9806370166547385, "1000–3000 Hz_10ms_modulation_mae_db": 0.35939651927410016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9958976110406527, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08779249095195271, "1000–3000 Hz_40ms_envelope_correlation": 0.993831016649768, "1000–3000 Hz_40ms_modulation_mae_db": 0.565614501220519, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7223410769193865, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24486666677880728, "3000–6000 Hz_10ms_envelope_correlation": 0.9512599719567291, "3000–6000 Hz_10ms_modulation_mae_db": 1.2507686563759848, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0700023067989883, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13231415346855357, "3000–6000 Hz_40ms_envelope_correlation": 0.9845018003842018, "3000–6000 Hz_40ms_modulation_mae_db": 1.3047286749724198, "mr_stft_db": 5.871828971156564, "band_mae_db": 0.13442847379042916, "rms_gain_error_db": -0.05464182040892599, "output_peak": 0.38009924612061924}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7921435602344793, "全帯域_10ms_rms_envelope_nrmse": 0.0631934703997266, "全帯域_10ms_envelope_correlation": 0.9960617895835986, "全帯域_10ms_modulation_mae_db": 0.5809495135213603, "全帯域_40ms_log_rms_mae_db": 0.4309574701655063, "全帯域_40ms_rms_envelope_nrmse": 0.03227978007229215, "全帯域_40ms_envelope_correlation": 0.9989402760976817, "全帯域_40ms_modulation_mae_db": 0.4890981079432531, "300–1000 Hz_10ms_log_rms_mae_db": 1.6403765977185083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08002701090962198, "300–1000 Hz_10ms_envelope_correlation": 0.994494307838487, "300–1000 Hz_10ms_modulation_mae_db": 0.5857096715875683, "300–1000 Hz_40ms_log_rms_mae_db": 0.8524571510726034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041994930014573814, "300–1000 Hz_40ms_envelope_correlation": 0.9984253248370158, "300–1000 Hz_40ms_modulation_mae_db": 0.5175393513479344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6420348638977509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16150803878740583, "1000–3000 Hz_10ms_envelope_correlation": 0.9793831471800376, "1000–3000 Hz_10ms_modulation_mae_db": 0.6476102050280326, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9775239317192838, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09234250762114309, "1000–3000 Hz_40ms_envelope_correlation": 0.9929717826612054, "1000–3000 Hz_40ms_modulation_mae_db": 0.8948022151951793, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7934800074441806, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26266775192817093, "3000–6000 Hz_10ms_envelope_correlation": 0.9424481654571457, "3000–6000 Hz_10ms_modulation_mae_db": 0.9341806577394732, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1036779072917435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1349978081209576, "3000–6000 Hz_40ms_envelope_correlation": 0.9836182609690194, "3000–6000 Hz_40ms_modulation_mae_db": 1.6045616410602068, "mr_stft_db": 5.86566570065171, "band_mae_db": 0.14008078779272282, "rms_gain_error_db": -0.07152844178155036, "output_peak": 0.36973525354068276}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8352987923059089, "全帯域_10ms_rms_envelope_nrmse": 0.061845800540457095, "全帯域_10ms_envelope_correlation": 0.9963441942355681, "全帯域_10ms_modulation_mae_db": 0.43504712034280096, "全帯域_40ms_log_rms_mae_db": 0.4440931876606256, "全帯域_40ms_rms_envelope_nrmse": 0.033421899871350386, "全帯域_40ms_envelope_correlation": 0.9989992125234556, "全帯域_40ms_modulation_mae_db": 0.5032105386347355, "300–1000 Hz_10ms_log_rms_mae_db": 1.6657037138249233, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07723345474544184, "300–1000 Hz_10ms_envelope_correlation": 0.9950940765950829, "300–1000 Hz_10ms_modulation_mae_db": 0.5867657817722473, "300–1000 Hz_40ms_log_rms_mae_db": 0.8702306118819094, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04529770495208554, "300–1000 Hz_40ms_envelope_correlation": 0.9984202925189427, "300–1000 Hz_40ms_modulation_mae_db": 0.759152196539837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6373324382880654, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16369304143432412, "1000–3000 Hz_10ms_envelope_correlation": 0.9787978415926356, "1000–3000 Hz_10ms_modulation_mae_db": 0.7427052770302544, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9945590217615278, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10225756399039725, "1000–3000 Hz_40ms_envelope_correlation": 0.9913173514489233, "1000–3000 Hz_40ms_modulation_mae_db": 0.7945169305510152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7650470910366682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2413794277338985, "3000–6000 Hz_10ms_envelope_correlation": 0.9518851751835631, "3000–6000 Hz_10ms_modulation_mae_db": 1.1642210825732824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0900931381027894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.125030406003738, "3000–6000 Hz_40ms_envelope_correlation": 0.9861836022195339, "3000–6000 Hz_40ms_modulation_mae_db": 1.6629808829280837, "mr_stft_db": 5.892396972738127, "band_mae_db": 0.17595016918597686, "rms_gain_error_db": -0.12439478702195989, "output_peak": 0.3698985170003836}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.782725690375875, "spectral_convergence": 0.19732717082325194, "spectral_convergence_status": "ok", "amplitude_floor": 9.54303147153e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.799659505478818, "spectral_convergence": 0.18434074791671984, "spectral_convergence_status": "ok", "amplitude_floor": 7.553612828866807e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.068265704630769, "spectral_convergence": 0.18807325569126937, "spectral_convergence_status": "ok", "amplitude_floor": 6.230060381303045e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.459798737171228e-05, "output_power": 5.9868650727786166e-05, "delta_db": 0.40022836730961087, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011571142287648336, "output_power": 0.0011358794262082722, "delta_db": -0.0804400064741275, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009701755098987012, "output_power": 0.0009392349298881014, "delta_db": -0.1407607205437786, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.8727504132367385e-05, "output_power": 4.733240044485014e-05, "delta_db": -0.12615636880209152, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.5070297613490014e-05, "output_power": 2.5299679070586156e-05, "delta_db": 0.03955522566338006, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.1228013015272054e-05, "output_power": 3.1118263640249384e-05, "delta_db": -0.015289957878382943, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13373844111189528, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09247377039026337, "pitch": null, "output_rms_dbfs": -26.49997631087643, "output_peak": 0.37253523461825827, "reference_rms_dbfs": -26.407502540486163, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.883550300161821}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014140982422473965, "log_rms_mae_db": 0.8325440794701829, "rms_envelope_nrmse": 0.06344824514730546, "envelope_correlation": 0.9960654048184108, "rms_gain_error_db": -0.0924736159794457, "reference_mean_power": 0.0022888204000268467, "output_mean_power": 0.002240600115583209}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0460573675773498e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006367944533168533, "output_power": 0.0006159223839255901, "reference_active": true, "delta_db": -0.1447328434287218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.21105688014902e-05, "output_power": 8.649350571092028e-05, "reference_active": true, "delta_db": -0.2732596410058513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2041191998800966e-05, "output_power": 9.56599693065862e-06, "reference_active": true, "delta_db": -0.9993924415840467}], "active_band_count": 3, "active_band_mae_db": 0.4724616420062066}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274322977924003, "log_rms_mae_db": 0.4506005272346971, "rms_envelope_nrmse": 0.03266694348763602, "envelope_correlation": 0.9989578614703776, "rms_gain_error_db": -0.09247329381284536, "reference_mean_power": 0.0023003276499839287, "output_mean_power": 0.002251865100779974}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.201135437818204e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005685810358795958, "output_power": 0.0005482866821074754, "reference_active": true, "delta_db": -0.15784673399525773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.696081479723381e-05, "output_power": 4.544414191642303e-05, "reference_active": true, "delta_db": -0.14257716146373728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.36988853805451e-07, "output_power": 7.166392381541877e-07, "reference_active": true, "delta_db": -1.1643384124507858}], "active_band_count": 3, "active_band_mae_db": 0.48825410263659363}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012682596688393775, "log_rms_mae_db": 1.7033176711828184, "rms_envelope_nrmse": 0.07569546970772166, "envelope_correlation": 0.9951525099410232, "rms_gain_error_db": -0.12269614964459091, "reference_mean_power": 0.0016135328798776547, "output_mean_power": 0.001568585531262956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.436393406849114e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006358276871928121, "output_power": 0.000600990645402756, "reference_active": true, "delta_db": -0.24471723262539968}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.784078725837931e-05, "output_power": 7.903934590279996e-05, "reference_active": true, "delta_db": -0.45852881863706363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0842784949487312e-05, "output_power": 8.244181286000608e-06, "reference_active": true, "delta_db": -1.1899331111272973}], "active_band_count": 3, "active_band_mae_db": 0.6310597207965869}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011190972053745482, "log_rms_mae_db": 0.8568304748588527, "rms_envelope_nrmse": 0.04154266719019155, "envelope_correlation": 0.9985420665605879, "rms_gain_error_db": -0.12269644282242989, "reference_mean_power": 0.0016216465744125685, "output_mean_power": 0.0015764731003946957}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.451026691855673e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005862792692297138, "output_power": 0.000553309936948867, "reference_active": true, "delta_db": -0.25136068028806224}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.32096116317955e-05, "output_power": 4.004074895998362e-05, "reference_active": true, "delta_db": -0.33078170550406316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.999199919742325e-07, "output_power": 6.656461627116643e-07, "reference_active": true, "delta_db": -1.7672181881836155}], "active_band_count": 3, "active_band_mae_db": 0.783120191325247}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.381727808324483e-05, "log_rms_mae_db": 1.6163879939050314, "rms_envelope_nrmse": 0.15072808638782897, "envelope_correlation": 0.98205635773652, "rms_gain_error_db": -0.12259247057347143, "reference_mean_power": 6.869686019984948e-05, "output_mean_power": 6.678480172700148e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.573688907938692e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3586087172622807e-05, "output_power": 1.276605516785993e-05, "reference_active": true, "delta_db": -0.27037679954680516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.445561080668119e-06, "output_power": 2.0591813606221595e-06, "reference_active": true, "delta_db": -0.746839158767161}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901755177746097e-07, "output_power": 5.359700187583384e-07, "reference_active": true, "delta_db": -0.41840693033125065}], "active_band_count": 3, "active_band_mae_db": 0.478540962881739}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.1112752780143365e-05, "log_rms_mae_db": 0.9452749883223934, "rms_envelope_nrmse": 0.08480435997069201, "envelope_correlation": 0.9941465182824277, "rms_gain_error_db": -0.12258401447264747, "reference_mean_power": 6.904136883729842e-05, "output_mean_power": 6.71198522496602e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.335283050026158e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2478970323107004e-05, "output_power": 1.1713022449674021e-05, "reference_active": true, "delta_db": -0.27509776223707844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3575517045580512e-06, "output_power": 1.1146670756533686e-06, "reference_active": true, "delta_db": -0.8561120598949994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9851801848052634e-08, "output_power": 2.743256879100329e-08, "reference_active": true, "delta_db": -1.62181485316153}], "active_band_count": 3, "active_band_mae_db": 0.9176748917645359}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.653785524000496e-05, "log_rms_mae_db": 1.75570816030348, "rms_envelope_nrmse": 0.23252408788976972, "envelope_correlation": 0.9551116065840732, "rms_gain_error_db": -0.0883280076242002, "reference_mean_power": 1.4641647362050651e-05, "output_mean_power": 1.4346869305906984e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3222330564047269e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.030474467255676e-06, "output_power": 5.602667737439336e-06, "reference_active": true, "delta_db": -0.3195661533996338}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.175765551110617e-06, "output_power": 2.1371552395685315e-06, "reference_active": true, "delta_db": -0.07776026518065347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1480409466014962e-06, "output_power": 5.069997013230723e-07, "reference_active": true, "delta_db": -3.549496746143663}], "active_band_count": 3, "active_band_mae_db": 1.31560772157465}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.7197117785089643e-05, "log_rms_mae_db": 1.0562168253188842, "rms_envelope_nrmse": 0.12344610233113734, "envelope_correlation": 0.9866633246404057, "rms_gain_error_db": -0.08835176965198174, "reference_mean_power": 1.4714471928104246e-05, "output_mean_power": 1.4418148818258375e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.175274300154406e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.965217773921107e-06, "output_power": 5.428499565456381e-06, "reference_active": true, "delta_db": -0.40946495810118505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4021233434065406e-06, "output_power": 1.1591641118937873e-06, "reference_active": true, "delta_db": -0.8264129299830264}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593565327256392e-08, "output_power": 3.222410388189144e-08, "reference_active": true, "delta_db": -4.737991882973588}], "active_band_count": 3, "active_band_mae_db": 1.9912899236859332}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8882017019337083, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ccec73780c6ac099ec3a43470b7fe55631e47f3f8bb2587f95db95a9a05ed3ea", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.980803550909261, "standard_deviation": 0.0016808016135790073, "minimum": 0.9787978415926356, "maximum": 0.983912658497592, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6225645962480026, "standard_deviation": 0.016874956936129278, "minimum": 1.593393091894923, "maximum": 1.6420348638977509, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.558119415838483, "standard_deviation": 0.17277308896643165, "minimum": 0.35939651927410016, "maximum": 0.8604367651593469, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15575551818207362, "standard_deviation": 0.006887352347675846, "minimum": 0.14267620554121752, "maximum": 0.16369304143432412, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934655372923189, "standard_deviation": 0.0009921753707417735, "minimum": 0.9913173514489233, "maximum": 0.9945669025855265, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.963463000273237, "standard_deviation": 0.04345037028548977, "minimum": 0.8898713639612481, "maximum": 1.0125543438181208, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7843595322471895, "standard_deviation": 0.2605084422130063, "minimum": 0.2932391931539009, "maximum": 1.169887494813809, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08901963193392096, "standard_deviation": 0.006307831789445209, "minimum": 0.08104979232093826, "maximum": 0.10225756399039725, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9509900063947807, "standard_deviation": 0.00387119341216777, "minimum": 0.9424481654571457, "maximum": 0.9551116065840732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7501448853067807, "standard_deviation": 0.026091004778915464, "minimum": 1.713734112520304, "maximum": 1.7934800074441806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2222949930408686, "standard_deviation": 0.17365505472743928, "minimum": 0.9341806577394732, "maximum": 1.5478991867410272, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24388510621626142, "standard_deviation": 0.00888258594902857, "minimum": 0.23252408788976972, "maximum": 0.26266775192817093, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9850945057725626, "standard_deviation": 0.0013926033772105482, "minimum": 0.9831714340238726, "maximum": 0.9867392801595389, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0801462443488667, "standard_deviation": 0.018971073407870895, "minimum": 1.0562168253188842, "maximum": 1.106790334427766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6863764112772364, "standard_deviation": 0.34574119948511245, "minimum": 1.3047286749724198, "maximum": 2.346857576931904, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13010339723170972, "standard_deviation": 0.00625104474515571, "minimum": 0.12234133293759286, "maximum": 0.14013344740538625, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949473611537487, "standard_deviation": 0.00040080989164074704, "minimum": 0.9942983584532618, "maximum": 0.9954924467264029, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.671964466535984, "standard_deviation": 0.04552562768998512, "minimum": 1.6155772780446964, "maximum": 1.7475271840596582, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5311637779826063, "standard_deviation": 0.1094599180329864, "minimum": 0.34999316415874676, "maximum": 0.6641587102285711, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07719678310090958, "standard_deviation": 0.0031476012011578043, "minimum": 0.0729196399031015, "maximum": 0.08264435744427676, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984457475203736, "standard_deviation": 0.00017426241226813847, "minimum": 0.9981753454648421, "maximum": 0.998658420016954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8750210812524984, "standard_deviation": 0.042954775177092494, "minimum": 0.8186025480882069, "maximum": 0.9610273272313423, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6435162576294504, "standard_deviation": 0.130565237391343, "minimum": 0.4742448384789619, "maximum": 0.783120191325247, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04271008149234429, "standard_deviation": 0.0030021507444294576, "minimum": 0.03834673577733335, "maximum": 0.047103727564884484, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14363211629045863, "standard_deviation": 0.02435795627182127, "minimum": 0.11171632855515719, "maximum": 0.1825576771304375, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.875686979810309, "standard_deviation": 0.008559740503895156, "minimum": 5.86566570065171, "maximum": 5.892396972738127, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36944342329295243, "standard_deviation": 0.0066725335865671755, "minimum": 0.3602378447242811, "maximum": 0.38009924612061924, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0775763954410845, "standard_deviation": 0.031677403620961705, "minimum": -0.12439478702195989, "maximum": -0.0387234659981904, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962473627671189, "standard_deviation": 0.00034403276378093973, "minimum": 0.9958468140395828, "maximum": 0.9969859979314479, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8129043600169888, "standard_deviation": 0.016566416550717045, "minimum": 0.7921435602344793, "maximum": 0.8352987923059089, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.43232857346956033, "standard_deviation": 0.12417895873292581, "minimum": 0.23626855309961706, "maximum": 0.6165126307214249, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06183827997471779, "standard_deviation": 0.002914730041581921, "minimum": 0.055284972570254456, "maximum": 0.0647997656185505, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998941858757511, "standard_deviation": 0.00011496906266677347, "minimum": 0.9987122994283077, "maximum": 0.9990993181075243, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4352079674688157, "standard_deviation": 0.01073523596277673, "minimum": 0.4241488108734945, "maximum": 0.4506005272346971, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4797251971637815, "standard_deviation": 0.09700024506457663, "minimum": 0.34645110471416335, "maximum": 0.651397728308702, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03255341199334946, "standard_deviation": 0.0018777417550164947, "minimum": 0.029163704731472454, "maximum": 0.03505680181265348, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0016", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.964154005050659, "predicted_mos": 2.964154005050659, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "84292d7fe09850d3b6e28ed00f2fd89fc4c3d45a366de37e8c5ee6c68cc4dd47", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.286793799605221, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8275104306188819, "全帯域_10ms_rms_envelope_nrmse": 0.05829732598662352, "全帯域_10ms_envelope_correlation": 0.9966303028660067, "全帯域_10ms_modulation_mae_db": 0.3763188253707126, "全帯域_40ms_log_rms_mae_db": 0.44365033161888057, "全帯域_40ms_rms_envelope_nrmse": 0.032552380277652636, "全帯域_40ms_envelope_correlation": 0.9988816099098636, "全帯域_40ms_modulation_mae_db": 0.48452558156286835, "300–1000 Hz_10ms_log_rms_mae_db": 1.4177060293915005, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07062226978732421, "300–1000 Hz_10ms_envelope_correlation": 0.995709406128958, "300–1000 Hz_10ms_modulation_mae_db": 0.40994545280996436, "300–1000 Hz_40ms_log_rms_mae_db": 0.7354255106320629, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03978428023594896, "300–1000 Hz_40ms_envelope_correlation": 0.9985643984149156, "300–1000 Hz_40ms_modulation_mae_db": 0.6510654357494393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2921536614811069, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13682672606543475, "1000–3000 Hz_10ms_envelope_correlation": 0.9853835943913724, "1000–3000 Hz_10ms_modulation_mae_db": 0.1584751558651837, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7521675251138599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09531501507404498, "1000–3000 Hz_40ms_envelope_correlation": 0.9924663811641211, "1000–3000 Hz_40ms_modulation_mae_db": 0.17891439701349784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.200470816461395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18401509542425729, "3000–6000 Hz_10ms_envelope_correlation": 0.9720386760856348, "3000–6000 Hz_10ms_modulation_mae_db": 1.2416382791760434, "3000–6000 Hz_40ms_log_rms_mae_db": 0.737401907050279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10491578758352813, "3000–6000 Hz_40ms_envelope_correlation": 0.9906557446408748, "3000–6000 Hz_40ms_modulation_mae_db": 1.7291622177986437, "mr_stft_db": 5.540032887536785, "band_mae_db": 0.15017879894769723, "rms_gain_error_db": -0.04696899882065369, "output_peak": 0.383463387018667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8775053226758943, "全帯域_10ms_rms_envelope_nrmse": 0.06625642795482835, "全帯域_10ms_envelope_correlation": 0.9957238721895717, "全帯域_10ms_modulation_mae_db": 0.5102520237915695, "全帯域_40ms_log_rms_mae_db": 0.5073910519712876, "全帯域_40ms_rms_envelope_nrmse": 0.037049495515128955, "全帯域_40ms_envelope_correlation": 0.9986648100215483, "全帯域_40ms_modulation_mae_db": 0.5439969607221813, "300–1000 Hz_10ms_log_rms_mae_db": 1.440440878542043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07885259443343148, "300–1000 Hz_10ms_envelope_correlation": 0.9947536767457803, "300–1000 Hz_10ms_modulation_mae_db": 0.7032007206218137, "300–1000 Hz_40ms_log_rms_mae_db": 0.7919255222771628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04536881413257523, "300–1000 Hz_40ms_envelope_correlation": 0.9982800332069174, "300–1000 Hz_40ms_modulation_mae_db": 0.7401970976863087, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3179136962373565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12718931057058794, "1000–3000 Hz_10ms_envelope_correlation": 0.9872730754424758, "1000–3000 Hz_10ms_modulation_mae_db": 0.5824882690503285, "1000–3000 Hz_40ms_log_rms_mae_db": 0.739066131774978, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07114047322850167, "1000–3000 Hz_40ms_envelope_correlation": 0.9959207358050635, "1000–3000 Hz_40ms_modulation_mae_db": 0.5134911055274008, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2116109816371077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17639710621634339, "3000–6000 Hz_10ms_envelope_correlation": 0.974180165266942, "3000–6000 Hz_10ms_modulation_mae_db": 0.8304041346673685, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7286052002004689, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.091493124476973, "3000–6000 Hz_40ms_envelope_correlation": 0.9926855284704537, "3000–6000 Hz_40ms_modulation_mae_db": 1.482504198429968, "mr_stft_db": 5.529720975145428, "band_mae_db": 0.18806580298384767, "rms_gain_error_db": -0.10271219405648416, "output_peak": 0.368311814968469}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8548584126951685, "全帯域_10ms_rms_envelope_nrmse": 0.06679549293111652, "全帯域_10ms_envelope_correlation": 0.9955939059624512, "全帯域_10ms_modulation_mae_db": 0.18523744686814955, "全帯域_40ms_log_rms_mae_db": 0.47801230917390347, "全帯域_40ms_rms_envelope_nrmse": 0.03719638513160165, "全帯域_40ms_envelope_correlation": 0.9985701966458675, "全帯域_40ms_modulation_mae_db": 0.28008301651964257, "300–1000 Hz_10ms_log_rms_mae_db": 1.4765859472773322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08076550405631615, "300–1000 Hz_10ms_envelope_correlation": 0.9944084269023324, "300–1000 Hz_10ms_modulation_mae_db": 0.3603937622275742, "300–1000 Hz_40ms_log_rms_mae_db": 0.8172883888412622, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04692256272319284, "300–1000 Hz_40ms_envelope_correlation": 0.9980364450286793, "300–1000 Hz_40ms_modulation_mae_db": 0.44433136481227375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.309557891101728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14356876760505882, "1000–3000 Hz_10ms_envelope_correlation": 0.983896168715248, "1000–3000 Hz_10ms_modulation_mae_db": 0.4888940885699032, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7399357094604633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09007864457152272, "1000–3000 Hz_40ms_envelope_correlation": 0.9936216238088421, "1000–3000 Hz_40ms_modulation_mae_db": 0.5964380922820475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2050010327540464, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17760213663574778, "3000–6000 Hz_10ms_envelope_correlation": 0.9738046360543441, "3000–6000 Hz_10ms_modulation_mae_db": 1.0578936428279506, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7431045005055471, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11343661950624666, "3000–6000 Hz_40ms_envelope_correlation": 0.9884173694095353, "3000–6000 Hz_40ms_modulation_mae_db": 1.5589365473333527, "mr_stft_db": 5.511302305510893, "band_mae_db": 0.21733548957447915, "rms_gain_error_db": -0.07316123300402307, "output_peak": 0.3581415238657365}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7885828339989933, "全帯域_10ms_rms_envelope_nrmse": 0.05970423694070254, "全帯域_10ms_envelope_correlation": 0.9965055844383383, "全帯域_10ms_modulation_mae_db": 0.3573788789597509, "全帯域_40ms_log_rms_mae_db": 0.4099124592622721, "全帯域_40ms_rms_envelope_nrmse": 0.03329718146665182, "全帯域_40ms_envelope_correlation": 0.9988837239277862, "全帯域_40ms_modulation_mae_db": 0.32706062196015684, "300–1000 Hz_10ms_log_rms_mae_db": 1.4225290721776596, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07510899480318836, "300–1000 Hz_10ms_envelope_correlation": 0.9951935686516366, "300–1000 Hz_10ms_modulation_mae_db": 0.5235764042421981, "300–1000 Hz_40ms_log_rms_mae_db": 0.7400607246509154, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043057693714248196, "300–1000 Hz_40ms_envelope_correlation": 0.9983828533695569, "300–1000 Hz_40ms_modulation_mae_db": 0.5660837016668943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.253458545596862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12096880743785579, "1000–3000 Hz_10ms_envelope_correlation": 0.9886786137073367, "1000–3000 Hz_10ms_modulation_mae_db": 0.3182093931134394, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6907939584708614, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06636701394286323, "1000–3000 Hz_40ms_envelope_correlation": 0.9967400502102484, "1000–3000 Hz_40ms_modulation_mae_db": 0.6136201032979861, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2356197090599104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17409968099236678, "3000–6000 Hz_10ms_envelope_correlation": 0.9755787277175547, "3000–6000 Hz_10ms_modulation_mae_db": 1.1008849736516717, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7790068768380264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10542978026793656, "3000–6000 Hz_40ms_envelope_correlation": 0.9913347994439691, "3000–6000 Hz_40ms_modulation_mae_db": 1.4715805737775163, "mr_stft_db": 5.505102207540763, "band_mae_db": 0.19626796956278905, "rms_gain_error_db": -0.07980856962890268, "output_peak": 0.37405982190397047}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8470176612077476, "全帯域_10ms_rms_envelope_nrmse": 0.06849331275277318, "全帯域_10ms_envelope_correlation": 0.9953968893831285, "全帯域_10ms_modulation_mae_db": 0.33060635702150704, "全帯域_40ms_log_rms_mae_db": 0.47429883452321114, "全帯域_40ms_rms_envelope_nrmse": 0.03691545066642117, "全帯域_40ms_envelope_correlation": 0.9986317337268431, "全帯域_40ms_modulation_mae_db": 0.36702620372030975, "300–1000 Hz_10ms_log_rms_mae_db": 1.4726036260260045, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08545294908840409, "300–1000 Hz_10ms_envelope_correlation": 0.9938137005687616, "300–1000 Hz_10ms_modulation_mae_db": 0.3730033216837924, "300–1000 Hz_40ms_log_rms_mae_db": 0.7982789068441583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04741909358086562, "300–1000 Hz_40ms_envelope_correlation": 0.9980996225367007, "300–1000 Hz_40ms_modulation_mae_db": 0.4565169015219017, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3048320761031809, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12365599835050002, "1000–3000 Hz_10ms_envelope_correlation": 0.9883669178565813, "1000–3000 Hz_10ms_modulation_mae_db": 0.25939104482216074, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7614709379959883, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06880298161430218, "1000–3000 Hz_40ms_envelope_correlation": 0.996759140903387, "1000–3000 Hz_40ms_modulation_mae_db": 0.3445965986402832, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2286539845521949, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17384699345945664, "3000–6000 Hz_10ms_envelope_correlation": 0.975519868868939, "3000–6000 Hz_10ms_modulation_mae_db": 0.7629718841014786, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7583844468323716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09573663346527368, "3000–6000 Hz_40ms_envelope_correlation": 0.9930142522036484, "3000–6000 Hz_40ms_modulation_mae_db": 1.242550455515522, "mr_stft_db": 5.5198060179023, "band_mae_db": 0.24922986738259242, "rms_gain_error_db": -0.08813628304374557, "output_peak": 0.3757311528153396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8119644977327557, "全帯域_10ms_rms_envelope_nrmse": 0.0567620575450317, "全帯域_10ms_envelope_correlation": 0.9969276447949009, "全帯域_10ms_modulation_mae_db": 0.5676805865988422, "全帯域_40ms_log_rms_mae_db": 0.46222412235776605, "全帯域_40ms_rms_envelope_nrmse": 0.03321221652057169, "全帯域_40ms_envelope_correlation": 0.9989820344551944, "全帯域_40ms_modulation_mae_db": 0.5940003827083504, "300–1000 Hz_10ms_log_rms_mae_db": 1.4246241512122595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07214073901683074, "300–1000 Hz_10ms_envelope_correlation": 0.9957237121406042, "300–1000 Hz_10ms_modulation_mae_db": 0.8079590555278141, "300–1000 Hz_40ms_log_rms_mae_db": 0.7813008321537818, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044288955275887554, "300–1000 Hz_40ms_envelope_correlation": 0.9984563260639604, "300–1000 Hz_40ms_modulation_mae_db": 0.8208039938204577, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2503859619203117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12016787819543491, "1000–3000 Hz_10ms_envelope_correlation": 0.9886270665356383, "1000–3000 Hz_10ms_modulation_mae_db": 0.6096710682220788, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7254846341170341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07110760209635357, "1000–3000 Hz_40ms_envelope_correlation": 0.9958524048336779, "1000–3000 Hz_40ms_modulation_mae_db": 0.6224799224533996, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2224949297874135, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1810398017430821, "3000–6000 Hz_10ms_envelope_correlation": 0.9728547162498312, "3000–6000 Hz_10ms_modulation_mae_db": 1.096226071159789, "3000–6000 Hz_40ms_log_rms_mae_db": 0.727908946503907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10633667906964608, "3000–6000 Hz_40ms_envelope_correlation": 0.9896606950391954, "3000–6000 Hz_40ms_modulation_mae_db": 1.6158352178310904, "mr_stft_db": 5.535717367491217, "band_mae_db": 0.1867039984084703, "rms_gain_error_db": -0.114878434587471, "output_peak": 0.3687677716681277}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8297360332022465, "全帯域_10ms_rms_envelope_nrmse": 0.0618798095765558, "全帯域_10ms_envelope_correlation": 0.9963136060590823, "全帯域_10ms_modulation_mae_db": 0.21217301987264583, "全帯域_40ms_log_rms_mae_db": 0.4497683188577787, "全帯域_40ms_rms_envelope_nrmse": 0.03588684136460005, "全帯域_40ms_envelope_correlation": 0.9987779091814867, "全帯域_40ms_modulation_mae_db": 0.2665074619923398, "300–1000 Hz_10ms_log_rms_mae_db": 1.4472078893770297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07902102237506467, "300–1000 Hz_10ms_envelope_correlation": 0.9947969114013602, "300–1000 Hz_10ms_modulation_mae_db": 0.2650260005407134, "300–1000 Hz_40ms_log_rms_mae_db": 0.8511080952809321, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04887968027441914, "300–1000 Hz_40ms_envelope_correlation": 0.9980398637703952, "300–1000 Hz_40ms_modulation_mae_db": 0.4140703281909755, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3327213244908227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12753462866238552, "1000–3000 Hz_10ms_envelope_correlation": 0.98720271432726, "1000–3000 Hz_10ms_modulation_mae_db": 0.4757229028936935, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7693622314842274, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07839696393242584, "1000–3000 Hz_40ms_envelope_correlation": 0.9950032292860856, "1000–3000 Hz_40ms_modulation_mae_db": 0.6148367851728835, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1831720081947634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15702394626768268, "3000–6000 Hz_10ms_envelope_correlation": 0.9797284958590147, "3000–6000 Hz_10ms_modulation_mae_db": 0.6274173142766459, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7281871677036312, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09869798166924616, "3000–6000 Hz_40ms_envelope_correlation": 0.9911851677685309, "3000–6000 Hz_40ms_modulation_mae_db": 0.7867923250707497, "mr_stft_db": 5.525536434540043, "band_mae_db": 0.14743789189612058, "rms_gain_error_db": -0.11400921287820207, "output_peak": 0.36317181448329705}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.886783329929151, "全帯域_10ms_rms_envelope_nrmse": 0.06875609331481142, "全帯域_10ms_envelope_correlation": 0.9953416860699533, "全帯域_10ms_modulation_mae_db": 0.4600590126192358, "全帯域_40ms_log_rms_mae_db": 0.44286436101154186, "全帯域_40ms_rms_envelope_nrmse": 0.038215801378456336, "全帯域_40ms_envelope_correlation": 0.9985063178342274, "全帯域_40ms_modulation_mae_db": 0.5950401793607876, "300–1000 Hz_10ms_log_rms_mae_db": 1.5747102177738077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08736694090033048, "300–1000 Hz_10ms_envelope_correlation": 0.9934239101200082, "300–1000 Hz_10ms_modulation_mae_db": 0.3748851934688447, "300–1000 Hz_40ms_log_rms_mae_db": 0.7899849932086939, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05057940444173718, "300–1000 Hz_40ms_envelope_correlation": 0.9976845986863192, "300–1000 Hz_40ms_modulation_mae_db": 0.6826676629220229, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2884436608406413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1299224585076689, "1000–3000 Hz_10ms_envelope_correlation": 0.9869705034404788, "1000–3000 Hz_10ms_modulation_mae_db": 0.4841615131343128, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7390512101629344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07717711752570511, "1000–3000 Hz_40ms_envelope_correlation": 0.9955551725182226, "1000–3000 Hz_40ms_modulation_mae_db": 0.44603767712387293, "3000–6000 Hz_10ms_log_rms_mae_db": 1.208175324164908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18481045798178458, "3000–6000 Hz_10ms_envelope_correlation": 0.9717493976385533, "3000–6000 Hz_10ms_modulation_mae_db": 0.8579536438246301, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7230647051275906, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09390440238313913, "3000–6000 Hz_40ms_envelope_correlation": 0.9919386461893809, "3000–6000 Hz_40ms_modulation_mae_db": 0.9780624663063135, "mr_stft_db": 5.534769887361722, "band_mae_db": 0.16885952934040435, "rms_gain_error_db": -0.08295377600588989, "output_peak": 0.37985347220372984}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.352319095063715, "spectral_convergence": 0.18566739360999313, "spectral_convergence_status": "ok", "amplitude_floor": 9.54303147153e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3813619928371335, "spectral_convergence": 0.18085990841048932, "spectral_convergence_status": "ok", "amplitude_floor": 7.553612828866807e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.886417574709506, "spectral_convergence": 0.19083419860531603, "spectral_convergence_status": "ok", "amplitude_floor": 6.230060381303045e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.459798737171228e-05, "output_power": 5.979175821218588e-05, "delta_db": 0.3946469055517826, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011571142287648336, "output_power": 0.0011290303930316563, "delta_db": -0.10670600926553378, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009701755098987012, "output_power": 0.0009675123912537176, "delta_db": -0.01193771669534906, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.8727504132367385e-05, "output_power": 5.0985073832289126e-05, "delta_db": 0.19668885560187924, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.5070297613490014e-05, "output_power": 2.4871095158671456e-05, "delta_db": -0.034645804177262624, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.1228013015272054e-05, "output_power": 3.0123095694588886e-05, "delta_db": -0.15644750239437616, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15017879894769723, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04696899882065369, "pitch": null, "output_rms_dbfs": -26.45447153930682, "output_peak": 0.383463387018667, "reference_rms_dbfs": -26.407502540486163, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.540032887536785}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014140982422473965, "log_rms_mae_db": 0.8275104306188819, "rms_envelope_nrmse": 0.05829732598662352, "envelope_correlation": 0.9966303028660067, "rms_gain_error_db": -0.04696908629702355, "reference_mean_power": 0.0022888204000268467, "output_mean_power": 0.0022642001092193203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0460573675773498e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006367944533168533, "output_power": 0.0006241290032770282, "reference_active": true, "delta_db": -0.08724907186435987}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.21105688014902e-05, "output_power": 8.863950282579745e-05, "reference_active": true, "delta_db": -0.16682152712141105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2041191998800966e-05, "output_power": 9.844210543752977e-06, "reference_active": true, "delta_db": -0.8748858771263668}], "active_band_count": 3, "active_band_mae_db": 0.3763188253707126}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274322977924003, "log_rms_mae_db": 0.44365033161888057, "rms_envelope_nrmse": 0.032552380277652636, "envelope_correlation": 0.9988816099098636, "rms_gain_error_db": -0.046968803593756284, "reference_mean_power": 0.0023003276499839287, "output_mean_power": 0.00227558372658371}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.201135437818204e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005685810358795958, "output_power": 0.0005572497877601991, "reference_active": true, "delta_db": -0.08742458648505752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.696081479723381e-05, "output_power": 4.670259135551578e-05, "reference_active": true, "delta_db": -0.023946449220537163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.36988853805451e-07, "output_power": 6.878818550150845e-07, "reference_active": true, "delta_db": -1.3422057089830104}], "active_band_count": 3, "active_band_mae_db": 0.48452558156286835}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012682596688393775, "log_rms_mae_db": 1.4177060293915005, "rms_envelope_nrmse": 0.07062226978732421, "envelope_correlation": 0.995709406128958, "rms_gain_error_db": -0.0755400000715625, "reference_mean_power": 0.0016135328798776547, "output_mean_power": 0.0015857101996741504}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.436393406849114e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006358276871928121, "output_power": 0.0006191871921628539, "reference_active": true, "delta_db": -0.11517470974356611}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.784078725837931e-05, "output_power": 8.054571844491587e-05, "reference_active": true, "delta_db": -0.37653759764454264}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0842784949487312e-05, "output_power": 9.148048094296542e-06, "reference_active": true, "delta_db": -0.7381240510417844}], "active_band_count": 3, "active_band_mae_db": 0.40994545280996436}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011190972053745482, "log_rms_mae_db": 0.7354255106320629, "rms_envelope_nrmse": 0.03978428023594896, "envelope_correlation": 0.9985643984149156, "rms_gain_error_db": -0.07553977772138053, "reference_mean_power": 0.0016216465744125685, "output_mean_power": 0.0015936840686886907}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.451026691855673e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005862792692297138, "output_power": 0.0005707779676639845, "reference_active": true, "delta_db": -0.11637337085356075}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.32096116317955e-05, "output_power": 4.1011236047950945e-05, "reference_active": true, "delta_db": -0.22677504092521686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.999199919742325e-07, "output_power": 6.90176967568402e-07, "reference_active": true, "delta_db": -1.6100478954695405}], "active_band_count": 3, "active_band_mae_db": 0.6510654357494393}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.381727808324483e-05, "log_rms_mae_db": 1.2921536614811069, "rms_envelope_nrmse": 0.13682672606543475, "envelope_correlation": 0.9853835943913724, "rms_gain_error_db": 0.052231466260776016, "reference_mean_power": 6.869686019984948e-05, "output_mean_power": 6.952804766490177e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.573688907938692e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3586087172622807e-05, "output_power": 1.3049074529301291e-05, "reference_active": true, "delta_db": -0.17514685241796615}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.445561080668119e-06, "output_power": 2.394673513859637e-06, "reference_active": true, "delta_db": -0.09132203552904215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901755177746097e-07, "output_power": 5.624520859259737e-07, "reference_active": true, "delta_db": -0.2089565796485428}], "active_band_count": 3, "active_band_mae_db": 0.1584751558651837}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.1112752780143365e-05, "log_rms_mae_db": 0.7521675251138599, "rms_envelope_nrmse": 0.09531501507404498, "envelope_correlation": 0.9924663811641211, "rms_gain_error_db": 0.052227009701060076, "reference_mean_power": 6.904136883729842e-05, "output_mean_power": 6.987665292876419e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.335283050026158e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2478970323107004e-05, "output_power": 1.2009282053953721e-05, "reference_active": true, "delta_db": -0.16661706969778461}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3575517045580512e-06, "output_power": 1.330783855956906e-06, "reference_active": true, "delta_db": -0.08648855682344453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9851801848052634e-08, "output_power": 3.7332253465855486e-08, "reference_active": true, "delta_db": -0.28363756451926436}], "active_band_count": 3, "active_band_mae_db": 0.17891439701349784}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.653785524000496e-05, "log_rms_mae_db": 1.200470816461395, "rms_envelope_nrmse": 0.18401509542425729, "envelope_correlation": 0.9720386760856348, "rms_gain_error_db": -0.2908720331397518, "reference_mean_power": 1.4641647362050651e-05, "output_mean_power": 1.3693130259872317e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3222330564047269e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.030474467255676e-06, "output_power": 5.8948448500783825e-06, "reference_active": true, "delta_db": -0.09879103892653888}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.175765551110617e-06, "output_power": 1.9351073549017882e-06, "reference_active": true, "delta_db": -0.5090703269890485}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1480409466014962e-06, "output_power": 5.600825358047881e-07, "reference_active": true, "delta_db": -3.117053471612543}], "active_band_count": 3, "active_band_mae_db": 1.2416382791760434}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.7197117785089643e-05, "log_rms_mae_db": 0.737401907050279, "rms_envelope_nrmse": 0.10491578758352813, "envelope_correlation": 0.9906557446408748, "rms_gain_error_db": -0.2908786347343033, "reference_mean_power": 1.4714471928104246e-05, "output_mean_power": 1.3761216177350371e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.175274300154406e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.965217773921107e-06, "output_power": 5.667128983797648e-06, "reference_active": true, "delta_db": -0.22263205917403645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4021233434065406e-06, "output_power": 1.1037977208622996e-06, "reference_active": true, "delta_db": -1.038967267754777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593565327256392e-08, "output_power": 3.885002517623163e-08, "reference_active": true, "delta_db": -3.9258873264671172}], "active_band_count": 3, "active_band_mae_db": 1.7291622177986437}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9038230087252312, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bfdc7c1a17b0e8a6740af842348712d2b8ebefb34f61971950dd416b3f30c3b5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9870498318020489, "standard_deviation": 0.0016765933712280686, "minimum": 0.983896168715248, "maximum": 0.9886786137073367, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2936833522215014, "standard_deviation": 0.02930105135434035, "minimum": 1.2503859619203117, "maximum": 1.3327213244908227, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.42212667945888754, "standard_deviation": 0.159848729796626, "minimum": 0.1584751558651837, "maximum": 0.6096710682220788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12872932192436584, "standard_deviation": 0.008020172065348415, "minimum": 0.12016787819543491, "maximum": 0.14356876760505882, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995239842316206, "standard_deviation": 0.0015054129696988607, "minimum": 0.9924663811641211, "maximum": 0.996759140903387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7396665423225434, "standard_deviation": 0.024213360380317427, "minimum": 0.6907939584708614, "maximum": 0.7693622314842274, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4913018351889214, "standard_deviation": 0.16041620482747274, "minimum": 0.17891439701349784, "maximum": 0.6224799224533996, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07729822649821491, "standard_deviation": 0.01039964283336391, "minimum": 0.06636701394286323, "maximum": 0.09531501507404498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9744318354676018, "standard_deviation": 0.0025751697966916146, "minimum": 0.9717493976385533, "maximum": 0.9797284958590147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2118998483264676, "standard_deviation": 0.01678881163815596, "minimum": 1.1831720081947634, "maximum": 1.2356197090599104, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9469237429606971, "standard_deviation": 0.20790841537975135, "minimum": 0.6274173142766459, "maximum": 1.2416382791760434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17610440234009014, "standard_deviation": 0.008768776810139135, "minimum": 0.15702394626768268, "maximum": 0.18481045798178458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9911115253956986, "standard_deviation": 0.0015310304461690735, "minimum": 0.9884173694095353, "maximum": 0.9930142522036484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7407079688452277, "standard_deviation": 0.019176938656556997, "minimum": 0.7230647051275906, "maximum": 0.7790068768380264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3581780002578945, "standard_deviation": 0.328845776043604, "minimum": 0.7867923250707497, "maximum": 1.7291622177986437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10124387605274868, "standard_deviation": 0.00747721007031144, "minimum": 0.091493124476973, "maximum": 0.11343661950624666, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9947279140824301, "standard_deviation": 0.0008294131895984299, "minimum": 0.9934239101200082, "maximum": 0.9957237121406042, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4595509764722046, "standard_deviation": 0.05152151354576659, "minimum": 1.4177060293915005, "maximum": 1.5747102177738077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47724873889033936, "standard_deviation": 0.18789760705398814, "minimum": 0.2650260005407134, "maximum": 0.8079590555278141, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07866637680761127, "standard_deviation": 0.0059314023588567825, "minimum": 0.07062226978732421, "maximum": 0.08736694090033048, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981930176346805, "standard_deviation": 0.00028436593310648733, "minimum": 0.9976845986863192, "maximum": 0.9985643984149156, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7881716217361212, "standard_deviation": 0.037932913584315706, "minimum": 0.7354255106320629, "maximum": 0.8511080952809321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5969670607962843, "standard_deviation": 0.1503889922966475, "minimum": 0.4140703281909755, "maximum": 0.8208039938204577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04578756054735934, "standard_deviation": 0.003430656247854315, "minimum": 0.03978428023594896, "maximum": 0.05057940444173718, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18800991851205012, "standard_deviation": 0.0340625382244371, "minimum": 0.14743789189612058, "maximum": 0.24922986738259242, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.525248510378644, "standard_deviation": 0.01235723102865163, "minimum": 5.505102207540763, "maximum": 5.540032887536785, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3714375948659171, "standard_deviation": 0.008474874360810764, "minimum": 0.3581415238657365, "maximum": 0.383463387018667, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08782858775317151, "standard_deviation": 0.022713927577706367, "minimum": -0.114878434587471, "maximum": -0.04696899882065369, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9960541864704291, "standard_deviation": 0.0006125543877971317, "minimum": 0.9953416860699533, "maximum": 0.9969276447949009, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8404948152576048, "standard_deviation": 0.03285127340980219, "minimum": 0.7885828339989933, "maximum": 0.886783329929151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3749632688878017, "standard_deviation": 0.13493884011371296, "minimum": 0.18523744686814955, "maximum": 0.5676805865988422, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06336809462530538, "standard_deviation": 0.0047865762490188375, "minimum": 0.0567620575450317, "maximum": 0.06875609331481142, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987372919628522, "standard_deviation": 0.0001696570920787161, "minimum": 0.9985063178342274, "maximum": 0.9989820344551944, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4585152235970802, "standard_deviation": 0.029121697374583356, "minimum": 0.4099124592622721, "maximum": 0.5073910519712876, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43228005106832956, "standard_deviation": 0.1382690285828619, "minimum": 0.2665074619923398, "maximum": 0.5950401793607876, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.035540719040135546, "standard_deviation": 0.002189877467841816, "minimum": 0.032552380277652636, "maximum": 0.038215801378456336, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0016", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.8517119884490967, "predicted_mos": 2.8517119884490967, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "01eb47320bda9777935ec1e60538d2e8ed189a4ec0804b374ffcda3d3f3ea211", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.052449299953878, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8394704624659316, "全帯域_10ms_rms_envelope_nrmse": 0.061919763346642634, "全帯域_10ms_envelope_correlation": 0.9964187600111154, "全帯域_10ms_modulation_mae_db": 0.4967273850093286, "全帯域_40ms_log_rms_mae_db": 0.4493675648628927, "全帯域_40ms_rms_envelope_nrmse": 0.032776001283676234, "全帯域_40ms_envelope_correlation": 0.9991453682512628, "全帯域_40ms_modulation_mae_db": 0.5342629433729703, "300–1000 Hz_10ms_log_rms_mae_db": 1.5780206554835923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08175660180174363, "300–1000 Hz_10ms_envelope_correlation": 0.9945512805183558, "300–1000 Hz_10ms_modulation_mae_db": 0.6035879174170322, "300–1000 Hz_40ms_log_rms_mae_db": 0.8187943170164776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046464064421714885, "300–1000 Hz_40ms_envelope_correlation": 0.9984378280003992, "300–1000 Hz_40ms_modulation_mae_db": 0.6551499359269494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4210880240430352, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1388842587198108, "1000–3000 Hz_10ms_envelope_correlation": 0.9847908188457887, "1000–3000 Hz_10ms_modulation_mae_db": 0.4734982454029182, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7852006120250756, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07832275288256367, "1000–3000 Hz_40ms_envelope_correlation": 0.9950365834987838, "1000–3000 Hz_40ms_modulation_mae_db": 0.5299289826123227, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3634430996052513, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18816471469016322, "3000–6000 Hz_10ms_envelope_correlation": 0.9706952698219125, "3000–6000 Hz_10ms_modulation_mae_db": 1.2448321606443482, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8052020529203628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10383766955653471, "3000–6000 Hz_40ms_envelope_correlation": 0.9906578576610477, "3000–6000 Hz_40ms_modulation_mae_db": 1.7154809299957208, "mr_stft_db": 5.618271215476082, "band_mae_db": 0.17699781600214412, "rms_gain_error_db": -0.15070299181381916, "output_peak": 0.3548987973978774}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8028553739045999, "全帯域_10ms_rms_envelope_nrmse": 0.06737928982479212, "全帯域_10ms_envelope_correlation": 0.9955921899674777, "全帯域_10ms_modulation_mae_db": 0.817037863422955, "全帯域_40ms_log_rms_mae_db": 0.4374550814668301, "全帯域_40ms_rms_envelope_nrmse": 0.035396543622048846, "全帯域_40ms_envelope_correlation": 0.9988092331838323, "全帯域_40ms_modulation_mae_db": 1.2427604677229045, "300–1000 Hz_10ms_log_rms_mae_db": 1.4803842855146798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08076945904003505, "300–1000 Hz_10ms_envelope_correlation": 0.9945507441205859, "300–1000 Hz_10ms_modulation_mae_db": 0.8728706578903492, "300–1000 Hz_40ms_log_rms_mae_db": 0.7346123577044176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042720311525892646, "300–1000 Hz_40ms_envelope_correlation": 0.9985786664778625, "300–1000 Hz_40ms_modulation_mae_db": 1.2348898081156092, "1000–3000 Hz_10ms_log_rms_mae_db": 1.381121587444003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1497699305025027, "1000–3000 Hz_10ms_envelope_correlation": 0.9822956374111759, "1000–3000 Hz_10ms_modulation_mae_db": 1.0100198430805236, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7820860209162411, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08300949142236866, "1000–3000 Hz_40ms_envelope_correlation": 0.9944646343909062, "1000–3000 Hz_40ms_modulation_mae_db": 1.2512334322769358, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3879368503189133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20961216875398841, "3000–6000 Hz_10ms_envelope_correlation": 0.9634173392027091, "3000–6000 Hz_10ms_modulation_mae_db": 0.904154887101433, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8039528929413736, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10097439985945399, "3000–6000 Hz_40ms_envelope_correlation": 0.9907484002073013, "3000–6000 Hz_40ms_modulation_mae_db": 1.466079722753434, "mr_stft_db": 5.611397285560251, "band_mae_db": 0.14166357049732012, "rms_gain_error_db": -0.11186665634786076, "output_peak": 0.35881049473782406}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8614130786376587, "全帯域_10ms_rms_envelope_nrmse": 0.06301351906891957, "全帯域_10ms_envelope_correlation": 0.9961327269462081, "全帯域_10ms_modulation_mae_db": 0.5206607411422196, "全帯域_40ms_log_rms_mae_db": 0.47035307195092013, "全帯域_40ms_rms_envelope_nrmse": 0.03375337217993537, "全帯域_40ms_envelope_correlation": 0.998895469801012, "全帯域_40ms_modulation_mae_db": 0.6297044113613514, "300–1000 Hz_10ms_log_rms_mae_db": 1.569065206160646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07821708683682571, "300–1000 Hz_10ms_envelope_correlation": 0.9948042424034734, "300–1000 Hz_10ms_modulation_mae_db": 0.7806013319947652, "300–1000 Hz_40ms_log_rms_mae_db": 0.7970478413535768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04407658557850841, "300–1000 Hz_40ms_envelope_correlation": 0.9983356566366929, "300–1000 Hz_40ms_modulation_mae_db": 0.8509174355201236, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3979746068821617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15538937588222504, "1000–3000 Hz_10ms_envelope_correlation": 0.9809009009236961, "1000–3000 Hz_10ms_modulation_mae_db": 0.6103652870295424, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8228772673015619, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09705135584576562, "1000–3000 Hz_40ms_envelope_correlation": 0.9922846534012951, "1000–3000 Hz_40ms_modulation_mae_db": 0.7722321590188951, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3662009895154394, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20521130948578786, "3000–6000 Hz_10ms_envelope_correlation": 0.9649630836664158, "3000–6000 Hz_10ms_modulation_mae_db": 1.289890082926212, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7859279526930267, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10166699992687851, "3000–6000 Hz_40ms_envelope_correlation": 0.9906033039871903, "3000–6000 Hz_40ms_modulation_mae_db": 1.8812858314140313, "mr_stft_db": 5.619202276396076, "band_mae_db": 0.21045692835165797, "rms_gain_error_db": -0.0978324588326456, "output_peak": 0.3725727690846558}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.794225224607329, "全帯域_10ms_rms_envelope_nrmse": 0.06021315521179314, "全帯域_10ms_envelope_correlation": 0.9964170949872772, "全帯域_10ms_modulation_mae_db": 0.3425395498954358, "全帯域_40ms_log_rms_mae_db": 0.41196089598660374, "全帯域_40ms_rms_envelope_nrmse": 0.033064130169703554, "全帯域_40ms_envelope_correlation": 0.9988672925466165, "全帯域_40ms_modulation_mae_db": 0.3956859742652747, "300–1000 Hz_10ms_log_rms_mae_db": 1.4756401323789128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07396969360695242, "300–1000 Hz_10ms_envelope_correlation": 0.9952766804241175, "300–1000 Hz_10ms_modulation_mae_db": 0.44689735301808325, "300–1000 Hz_40ms_log_rms_mae_db": 0.7577184435487592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040240518712552, "300–1000 Hz_40ms_envelope_correlation": 0.9985210786387482, "300–1000 Hz_40ms_modulation_mae_db": 0.6300345450946542, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3869713184835, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13332522494260052, "1000–3000 Hz_10ms_envelope_correlation": 0.9861138008178506, "1000–3000 Hz_10ms_modulation_mae_db": 0.4020117942234009, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7951495554103526, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08108648982817462, "1000–3000 Hz_40ms_envelope_correlation": 0.9948262681474687, "1000–3000 Hz_40ms_modulation_mae_db": 0.3197588564597961, "3000–6000 Hz_10ms_log_rms_mae_db": 1.384823678525392, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18553081413410707, "3000–6000 Hz_10ms_envelope_correlation": 0.9714080021077311, "3000–6000 Hz_10ms_modulation_mae_db": 0.9045938291869113, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8103683614228858, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08929924708228612, "3000–6000 Hz_40ms_envelope_correlation": 0.9929901715758507, "3000–6000 Hz_40ms_modulation_mae_db": 1.2028527946582723, "mr_stft_db": 5.605141021389464, "band_mae_db": 0.18626659384542185, "rms_gain_error_db": -0.057158673949849456, "output_peak": 0.37742356571523245}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8091688840573, "全帯域_10ms_rms_envelope_nrmse": 0.06008246393411387, "全帯域_10ms_envelope_correlation": 0.9964569620299055, "全帯域_10ms_modulation_mae_db": 0.26034771078359603, "全帯域_40ms_log_rms_mae_db": 0.425204187392461, "全帯域_40ms_rms_envelope_nrmse": 0.032658922403392764, "全帯域_40ms_envelope_correlation": 0.9989237518150436, "全帯域_40ms_modulation_mae_db": 0.4671591306136417, "300–1000 Hz_10ms_log_rms_mae_db": 1.5485407534885878, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0764699902475628, "300–1000 Hz_10ms_envelope_correlation": 0.9950510546163164, "300–1000 Hz_10ms_modulation_mae_db": 0.36423958315875815, "300–1000 Hz_40ms_log_rms_mae_db": 0.806599294344013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044704225216257144, "300–1000 Hz_40ms_envelope_correlation": 0.9982917548834189, "300–1000 Hz_40ms_modulation_mae_db": 0.5901518078207354, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3846925778892234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13557821883096827, "1000–3000 Hz_10ms_envelope_correlation": 0.9855003865174559, "1000–3000 Hz_10ms_modulation_mae_db": 0.9236177067945287, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7670078867327559, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07286753201639709, "1000–3000 Hz_40ms_envelope_correlation": 0.9956978454829484, "1000–3000 Hz_40ms_modulation_mae_db": 1.0769964644142311, "3000–6000 Hz_10ms_log_rms_mae_db": 1.390463725763893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18772534045982228, "3000–6000 Hz_10ms_envelope_correlation": 0.9707008447093449, "3000–6000 Hz_10ms_modulation_mae_db": 1.0400571229079156, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7828914083696455, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09636059949149303, "3000–6000 Hz_40ms_envelope_correlation": 0.991730204310291, "3000–6000 Hz_40ms_modulation_mae_db": 1.2968945778502168, "mr_stft_db": 5.619190724533752, "band_mae_db": 0.17807043579482185, "rms_gain_error_db": -0.07734590499064009, "output_peak": 0.35577623905643285}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.807920958285826, "全帯域_10ms_rms_envelope_nrmse": 0.05914527564410608, "全帯域_10ms_envelope_correlation": 0.996598182867475, "全帯域_10ms_modulation_mae_db": 0.2889238321784548, "全帯域_40ms_log_rms_mae_db": 0.41405732876887796, "全帯域_40ms_rms_envelope_nrmse": 0.03135081392449163, "全帯域_40ms_envelope_correlation": 0.9990498756117915, "全帯域_40ms_modulation_mae_db": 0.5212111725176688, "300–1000 Hz_10ms_log_rms_mae_db": 1.5310850764666721, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07522737390431847, "300–1000 Hz_10ms_envelope_correlation": 0.9952794631023432, "300–1000 Hz_10ms_modulation_mae_db": 0.4648839621271715, "300–1000 Hz_40ms_log_rms_mae_db": 0.7673545877105593, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04201189565445303, "300–1000 Hz_40ms_envelope_correlation": 0.9985900827480653, "300–1000 Hz_40ms_modulation_mae_db": 0.7477883043809976, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3941338349793437, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1369584754081679, "1000–3000 Hz_10ms_envelope_correlation": 0.9852036038008114, "1000–3000 Hz_10ms_modulation_mae_db": 0.529229540942118, "1000–3000 Hz_40ms_log_rms_mae_db": 0.803616103014757, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08299627476139658, "1000–3000 Hz_40ms_envelope_correlation": 0.9942283107417164, "1000–3000 Hz_40ms_modulation_mae_db": 0.7502057110052697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3885101408506981, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17038903245279313, "3000–6000 Hz_10ms_envelope_correlation": 0.9759131059737504, "3000–6000 Hz_10ms_modulation_mae_db": 1.0650986186158118, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8362332983025934, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09093200180132581, "3000–6000 Hz_40ms_envelope_correlation": 0.9924496386086926, "3000–6000 Hz_40ms_modulation_mae_db": 1.3001331386548225, "mr_stft_db": 5.627149539909226, "band_mae_db": 0.20909769494955013, "rms_gain_error_db": -0.09203973912058974, "output_peak": 0.3691496438313175}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8078917082969458, "全帯域_10ms_rms_envelope_nrmse": 0.062215292021572714, "全帯域_10ms_envelope_correlation": 0.9962506238081481, "全帯域_10ms_modulation_mae_db": 0.3929495956606253, "全帯域_40ms_log_rms_mae_db": 0.4407161573679413, "全帯域_40ms_rms_envelope_nrmse": 0.03586867180778716, "全帯域_40ms_envelope_correlation": 0.9987511095765984, "全帯域_40ms_modulation_mae_db": 0.335697818913183, "300–1000 Hz_10ms_log_rms_mae_db": 1.5375097443417733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08006251274515858, "300–1000 Hz_10ms_envelope_correlation": 0.9946432598732191, "300–1000 Hz_10ms_modulation_mae_db": 0.6147186429058252, "300–1000 Hz_40ms_log_rms_mae_db": 0.8025342725594679, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04667022357416939, "300–1000 Hz_40ms_envelope_correlation": 0.9982140398997349, "300–1000 Hz_40ms_modulation_mae_db": 0.7882994411272825, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4119889973176731, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12942671242325468, "1000–3000 Hz_10ms_envelope_correlation": 0.9867754452626223, "1000–3000 Hz_10ms_modulation_mae_db": 1.161186177649322, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8045154657216751, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08162855547232212, "1000–3000 Hz_40ms_envelope_correlation": 0.994485645207235, "1000–3000 Hz_40ms_modulation_mae_db": 1.4887485879074498, "3000–6000 Hz_10ms_log_rms_mae_db": 1.361295655818379, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17669943138375052, "3000–6000 Hz_10ms_envelope_correlation": 0.974209478300974, "3000–6000 Hz_10ms_modulation_mae_db": 0.280751546786328, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8072629894803287, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09169198815355978, "3000–6000 Hz_40ms_envelope_correlation": 0.9923040158242621, "3000–6000 Hz_40ms_modulation_mae_db": 0.36396354625495864, "mr_stft_db": 5.615904028985974, "band_mae_db": 0.13747876712676432, "rms_gain_error_db": -0.1043668102461147, "output_peak": 0.3634310914722346}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8243438719809475, "全帯域_10ms_rms_envelope_nrmse": 0.05983372221233082, "全帯域_10ms_envelope_correlation": 0.9964666089317473, "全帯域_10ms_modulation_mae_db": 0.1834773610872419, "全帯域_40ms_log_rms_mae_db": 0.4549445157771344, "全帯域_40ms_rms_envelope_nrmse": 0.032189021974103844, "全帯域_40ms_envelope_correlation": 0.9989319537168415, "全帯域_40ms_modulation_mae_db": 0.3981010500529112, "300–1000 Hz_10ms_log_rms_mae_db": 1.4753535953771113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07208854505154295, "300–1000 Hz_10ms_envelope_correlation": 0.9955581137242135, "300–1000 Hz_10ms_modulation_mae_db": 0.28854293143993504, "300–1000 Hz_40ms_log_rms_mae_db": 0.7424499845977854, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04017005663431575, "300–1000 Hz_40ms_envelope_correlation": 0.9985805197523866, "300–1000 Hz_40ms_modulation_mae_db": 0.5403333100580497, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4161473431430172, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12327804209392272, "1000–3000 Hz_10ms_envelope_correlation": 0.9880892608632104, "1000–3000 Hz_10ms_modulation_mae_db": 0.9419000317330789, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8059115266660991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06948994370443373, "1000–3000 Hz_40ms_envelope_correlation": 0.9961693885378877, "1000–3000 Hz_40ms_modulation_mae_db": 1.27762705528676, "3000–6000 Hz_10ms_log_rms_mae_db": 1.384599383843009, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2122899769058653, "3000–6000 Hz_10ms_envelope_correlation": 0.9625178421957173, "3000–6000 Hz_10ms_modulation_mae_db": 1.1504849349680633, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7981397314318367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10660660763010663, "3000–6000 Hz_40ms_envelope_correlation": 0.989636092076904, "3000–6000 Hz_40ms_modulation_mae_db": 1.3902708393182586, "mr_stft_db": 5.627352698374832, "band_mae_db": 0.17861814051050062, "rms_gain_error_db": -0.06418822694120199, "output_peak": 0.3719408426464498}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.498195616095585, "spectral_convergence": 0.18919505876828632, "spectral_convergence_status": "ok", "amplitude_floor": 9.54303147153e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.477251132791514, "spectral_convergence": 0.1780493045849487, "spectral_convergence_status": "ok", "amplitude_floor": 7.553612828866807e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.879366897541151, "spectral_convergence": 0.1874836011182091, "spectral_convergence_status": "ok", "amplitude_floor": 6.230060381303045e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.459798737171228e-05, "output_power": 5.958438108780488e-05, "delta_db": 0.37955798880238323, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011571142287648336, "output_power": 0.0011273036685778338, "delta_db": -0.11335313711079988, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009701755098987012, "output_power": 0.0009195438413152188, "delta_db": -0.23277867519022188, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.8727504132367385e-05, "output_power": 4.771712206128387e-05, "delta_db": -0.09099924583571949, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.5070297613490014e-05, "output_power": 2.4223707573319416e-05, "delta_db": -0.14918874484669264, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.1228013015272054e-05, "output_power": 3.054452960661843e-05, "delta_db": -0.0961091042270476, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17699781600214412, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15070299181381916, "pitch": null, "output_rms_dbfs": -26.558205532299986, "output_peak": 0.3548987973978774, "reference_rms_dbfs": -26.407502540486163, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.618271215476082}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014140982422473965, "log_rms_mae_db": 0.8394704624659316, "rms_envelope_nrmse": 0.061919763346642634, "envelope_correlation": 0.9964187600111154, "rms_gain_error_db": -0.15070292574674357, "reference_mean_power": 0.0022888204000268467, "output_mean_power": 0.002210759108728929}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0460573675773498e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006367944533168533, "output_power": 0.0006079659027225615, "reference_active": true, "delta_db": -0.20120049002495696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.21105688014902e-05, "output_power": 8.71477662108783e-05, "reference_active": true, "delta_db": -0.2405320450402551}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2041191998800966e-05, "output_power": 9.458549028366945e-06, "reference_active": true, "delta_db": -1.0484496199627737}], "active_band_count": 3, "active_band_mae_db": 0.4967273850093286}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274322977924003, "log_rms_mae_db": 0.4493675648628927, "rms_envelope_nrmse": 0.032776001283676234, "envelope_correlation": 0.9991453682512628, "rms_gain_error_db": -0.150703100135095, "reference_mean_power": 0.0023003276499839287, "output_mean_power": 0.0022218738093208704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.201135437818204e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005685810358795958, "output_power": 0.0005459991254918365, "reference_active": true, "delta_db": -0.17600423280369676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.696081479723381e-05, "output_power": 4.474351308910128e-05, "reference_active": true, "delta_db": -0.21005543327801485}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.36988853805451e-07, "output_power": 7.080460564187203e-07, "reference_active": true, "delta_db": -1.216729164037199}], "active_band_count": 3, "active_band_mae_db": 0.5342629433729703}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012682596688393775, "log_rms_mae_db": 1.5780206554835923, "rms_envelope_nrmse": 0.08175660180174363, "envelope_correlation": 0.9945512805183558, "rms_gain_error_db": -0.2046986120880369, "reference_mean_power": 0.0016135328798776547, "output_mean_power": 0.0015392457335163215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.436393406849114e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006358276871928121, "output_power": 0.0006000373286377534, "reference_active": true, "delta_db": -0.2516116648358911}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.784078725837931e-05, "output_power": 8.186198856968889e-05, "reference_active": true, "delta_db": -0.3061392921156899}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0842784949487312e-05, "output_power": 8.12530299160783e-06, "reference_active": true, "delta_db": -1.2530127952995156}], "active_band_count": 3, "active_band_mae_db": 0.6035879174170322}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011190972053745482, "log_rms_mae_db": 0.8187943170164776, "rms_envelope_nrmse": 0.046464064421714885, "envelope_correlation": 0.9984378280003992, "rms_gain_error_db": -0.20469876545174154, "reference_mean_power": 0.0016216465744125685, "output_mean_power": 0.0015469858184601762}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.451026691855673e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005862792692297138, "output_power": 0.0005548736423120875, "reference_active": true, "delta_db": -0.23910442466405932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.32096116317955e-05, "output_power": 4.094899794469447e-05, "reference_active": true, "delta_db": -0.2333708420075607}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.999199919742325e-07, "output_power": 7.090352019905139e-07, "reference_active": true, "delta_db": -1.492974541109228}], "active_band_count": 3, "active_band_mae_db": 0.6551499359269494}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.381727808324483e-05, "log_rms_mae_db": 1.4210880240430352, "rms_envelope_nrmse": 0.1388842587198108, "envelope_correlation": 0.9847908188457887, "rms_gain_error_db": -0.14798247702708867, "reference_mean_power": 6.869686019984948e-05, "output_mean_power": 6.63954991674502e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.573688907938692e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3586087172622807e-05, "output_power": 1.3156455727129344e-05, "reference_active": true, "delta_db": -0.1395548821460279}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.445561080668119e-06, "output_power": 2.16293557239388e-06, "reference_active": true, "delta_db": -0.5333493103816965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901755177746097e-07, "output_power": 4.9684638621488e-07, "reference_active": true, "delta_db": -0.7475905436810303}], "active_band_count": 3, "active_band_mae_db": 0.4734982454029182}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.1112752780143365e-05, "log_rms_mae_db": 0.7852006120250756, "rms_envelope_nrmse": 0.07832275288256367, "envelope_correlation": 0.9950365834987838, "rms_gain_error_db": -0.14799125230290794, "reference_mean_power": 6.904136883729842e-05, "output_mean_power": 6.672833185402659e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.335283050026158e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2478970323107004e-05, "output_power": 1.2098875736348039e-05, "reference_active": true, "delta_db": -0.13433735645988326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3575517045580512e-06, "output_power": 1.1617984106779904e-06, "reference_active": true, "delta_db": -0.6762560135289779}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9851801848052634e-08, "output_power": 3.330646986262628e-08, "reference_active": true, "delta_db": -0.7791935778481069}], "active_band_count": 3, "active_band_mae_db": 0.5299289826123227}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.653785524000496e-05, "log_rms_mae_db": 1.3634430996052513, "rms_envelope_nrmse": 0.18816471469016322, "envelope_correlation": 0.9706952698219125, "rms_gain_error_db": -0.16285338055276763, "reference_mean_power": 1.4641647362050651e-05, "output_mean_power": 1.4102775899221354e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3222330564047269e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.030474467255676e-06, "output_power": 5.383669447052688e-06, "reference_active": true, "delta_db": -0.49273096315088305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.175765551110617e-06, "output_power": 2.0962000967043683e-06, "reference_active": true, "delta_db": -0.16179359607359708}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1480409466014962e-06, "output_power": 5.64885186299942e-07, "reference_active": true, "delta_db": -3.0799719227085647}], "active_band_count": 3, "active_band_mae_db": 1.2448321606443482}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.7197117785089643e-05, "log_rms_mae_db": 0.8052020529203628, "rms_envelope_nrmse": 0.10383766955653471, "envelope_correlation": 0.9906578576610477, "rms_gain_error_db": -0.16282410009243836, "reference_mean_power": 1.4714471928104246e-05, "output_mean_power": 1.4173015783801845e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.175274300154406e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.965217773921107e-06, "output_power": 5.17773980841084e-06, "reference_active": true, "delta_db": -0.6148608071238684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4021233434065406e-06, "output_power": 1.1881853200728324e-06, "reference_active": true, "delta_db": -0.719020373651001}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593565327256392e-08, "output_power": 3.987712836361821e-08, "reference_active": true, "delta_db": -3.8125616092122927}], "active_band_count": 3, "active_band_mae_db": 1.7154809299957208}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9004019686958665, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f181cdcb93a77cf0d3a20cc0f166e63cb4916f301838c666c7067b9535a56131", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9849587318053263, "standard_deviation": 0.0023413809872217255, "minimum": 0.9809009009236961, "maximum": 0.9880892608632104, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3992647862727448, "standard_deviation": 0.015324236057420637, "minimum": 1.381121587444003, "maximum": 1.4210880240430352, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7564785783569291, "standard_deviation": 0.2851325020985517, "minimum": 0.4020117942234009, "maximum": 1.161186177649322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13782627985043158, "standard_deviation": 0.01042320738525969, "minimum": 0.12327804209392272, "maximum": 0.15538937588222504, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946491661760302, "standard_deviation": 0.0011601165368765212, "minimum": 0.9922846534012951, "maximum": 0.9961693885378877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7957955547235648, "standard_deviation": 0.017329995216178364, "minimum": 0.7670078867327559, "maximum": 0.8228772673015619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9333414061227076, "standard_deviation": 0.40486714979562477, "minimum": 0.3197588564597961, "maximum": 1.4887485879074498, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08080654949167776, "standard_deviation": 0.008207648475007157, "minimum": 0.06948994370443373, "maximum": 0.09705135584576562, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9692281207473195, "standard_deviation": 0.005007991289309276, "minimum": 0.9625178421957173, "maximum": 0.9759131059737504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.378409190530122, "standard_deviation": 0.012441167765244555, "minimum": 1.361295655818379, "maximum": 1.390463725763893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.984982897892128, "standard_deviation": 0.3176408618142518, "minimum": 0.280751546786328, "maximum": 1.289890082926212, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1919528485332847, "standard_deviation": 0.015456696999631916, "minimum": 0.17038903245279313, "maximum": 0.2122899769058653, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9913899605314425, "standard_deviation": 0.0011512299793959323, "minimum": 0.989636092076904, "maximum": 0.9929901715758507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8037473359452567, "standard_deviation": 0.016464851595793865, "minimum": 0.7828914083696455, "maximum": 0.8362332983025934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3271201726124642, "standard_deviation": 0.4513434506720041, "minimum": 0.36396354625495864, "maximum": 1.8812858314140313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09767118918770483, "standard_deviation": 0.0065216062733149055, "minimum": 0.08929924708228612, "maximum": 0.10660660763010663, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949643548478281, "standard_deviation": 0.0003830981119626962, "minimum": 0.9945507441205859, "maximum": 0.9955581137242135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.524449931151497, "standard_deviation": 0.04207123997711493, "minimum": 1.4753535953771113, "maximum": 1.5780206554835923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5545427974939899, "standard_deviation": 0.2017987450362022, "minimum": 0.28854293143993504, "maximum": 0.8728706578903492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07732015790426745, "standard_deviation": 0.0034558980294026593, "minimum": 0.07208854505154295, "maximum": 0.08175660180174363, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984437033796636, "standard_deviation": 0.00014730429447947454, "minimum": 0.9982140398997349, "maximum": 0.9985900827480653, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.778388887354382, "standard_deviation": 0.031885048657053355, "minimum": 0.7346123577044176, "maximum": 0.8187943170164776, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7546955735055502, "standard_deviation": 0.2202263824696426, "minimum": 0.5403333100580497, "maximum": 1.2348898081156092, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.043382235164732906, "standard_deviation": 0.002536667273566547, "minimum": 0.04017005663431575, "maximum": 0.04667022357416939, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17733124338477263, "standard_deviation": 0.02685669741942411, "minimum": 0.13747876712676432, "maximum": 0.21045692835165797, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6179510988282075, "standard_deviation": 0.007439729560151172, "minimum": 5.605141021389464, "maximum": 5.627352698374832, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36550043049275305, "standard_deviation": 0.0084765788343729, "minimum": 0.3548987973978774, "maximum": 0.37742356571523245, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09443768278034018, "standard_deviation": 0.029718438905074994, "minimum": -0.15070299181381916, "maximum": -0.057158673949849456, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962916436936693, "standard_deviation": 0.0003162229132055017, "minimum": 0.9955921899674777, "maximum": 0.996598182867475, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8184111952795674, "standard_deviation": 0.02223549643158302, "minimum": 0.794225224607329, "maximum": 0.8614130786376587, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41283300489748215, "standard_deviation": 0.19949666467141755, "minimum": 0.1834773610872419, "maximum": 0.817037863422955, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06172531015803387, "standard_deviation": 0.002649126591529322, "minimum": 0.05914527564410608, "maximum": 0.06737928982479212, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989217568128748, "standard_deviation": 0.00012647070159376873, "minimum": 0.9987511095765984, "maximum": 0.9991453682512628, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4380073504467077, "standard_deviation": 0.020303949964431783, "minimum": 0.41196089598660374, "maximum": 0.47035307195092013, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5655728711024882, "standard_deviation": 0.28910708343522357, "minimum": 0.335697818913183, "maximum": 1.2427604677229045, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033382184670642426, "standard_deviation": 0.001554710355914262, "minimum": 0.03135081392449163, "maximum": 0.03586867180778716, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0016", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.314694881439209, "predicted_mos": 2.314694881439209, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "409245a7524a30587c656e6789dc0e891179a41a8b0fc6e3d5e8de9361c72569", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7819556002505124, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8471597905970877, "全帯域_10ms_rms_envelope_nrmse": 0.06197856194452814, "全帯域_10ms_envelope_correlation": 0.9964075507406333, "全帯域_10ms_modulation_mae_db": 0.20906314285065597, "全帯域_40ms_log_rms_mae_db": 0.46075362195210706, "全帯域_40ms_rms_envelope_nrmse": 0.034622699291589794, "全帯域_40ms_envelope_correlation": 0.9990038224664813, "全帯域_40ms_modulation_mae_db": 0.2613976743099032, "300–1000 Hz_10ms_log_rms_mae_db": 1.5337022510593965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07733604036530543, "300–1000 Hz_10ms_envelope_correlation": 0.9951454273413276, "300–1000 Hz_10ms_modulation_mae_db": 0.28444689840843057, "300–1000 Hz_40ms_log_rms_mae_db": 0.8126855860634641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04580603510246661, "300–1000 Hz_40ms_envelope_correlation": 0.9984516610167261, "300–1000 Hz_40ms_modulation_mae_db": 0.4201868601417364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.419916780226886, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14747238105674698, "1000–3000 Hz_10ms_envelope_correlation": 0.9828254684320763, "1000–3000 Hz_10ms_modulation_mae_db": 0.9719521375753467, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8017498476087267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08592683115222066, "1000–3000 Hz_40ms_envelope_correlation": 0.9940031318761968, "1000–3000 Hz_40ms_modulation_mae_db": 1.1361958553574274, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3675511367805724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17935972760452928, "3000–6000 Hz_10ms_envelope_correlation": 0.9733062731004697, "3000–6000 Hz_10ms_modulation_mae_db": 1.1446250484982035, "3000–6000 Hz_40ms_log_rms_mae_db": 0.806150309235457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09784368397219054, "3000–6000 Hz_40ms_envelope_correlation": 0.9914990127130762, "3000–6000 Hz_40ms_modulation_mae_db": 1.5890904923038187, "mr_stft_db": 5.596321502635351, "band_mae_db": 0.22015141098615829, "rms_gain_error_db": -0.14956283015592375, "output_peak": 0.37086702318697057}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8347512462474953, "全帯域_10ms_rms_envelope_nrmse": 0.06078821926993862, "全帯域_10ms_envelope_correlation": 0.996427698599292, "全帯域_10ms_modulation_mae_db": 0.4640136411151232, "全帯域_40ms_log_rms_mae_db": 0.42416302374596937, "全帯域_40ms_rms_envelope_nrmse": 0.03229177727870819, "全帯域_40ms_envelope_correlation": 0.9990219220581863, "全帯域_40ms_modulation_mae_db": 0.6695041509975521, "300–1000 Hz_10ms_log_rms_mae_db": 1.4740420985679306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0766162576266344, "300–1000 Hz_10ms_envelope_correlation": 0.995098084895562, "300–1000 Hz_10ms_modulation_mae_db": 0.5359914220259111, "300–1000 Hz_40ms_log_rms_mae_db": 0.7880711125087515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044638886746924886, "300–1000 Hz_40ms_envelope_correlation": 0.998383959428438, "300–1000 Hz_40ms_modulation_mae_db": 0.8113921804870953, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3923440713575113, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12928514195258567, "1000–3000 Hz_10ms_envelope_correlation": 0.9869836170666098, "1000–3000 Hz_10ms_modulation_mae_db": 0.5176312090510394, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7845312178592571, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06776657905712351, "1000–3000 Hz_40ms_envelope_correlation": 0.9965617708722992, "1000–3000 Hz_40ms_modulation_mae_db": 0.5678679281692417, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3628525989492333, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19889454269502715, "3000–6000 Hz_10ms_envelope_correlation": 0.967654876559042, "3000–6000 Hz_10ms_modulation_mae_db": 1.085289358033067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7906790128830784, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10776633121364265, "3000–6000 Hz_40ms_envelope_correlation": 0.9895087135261718, "3000–6000 Hz_40ms_modulation_mae_db": 1.112153321384681, "mr_stft_db": 5.59557689572024, "band_mae_db": 0.15791881531926028, "rms_gain_error_db": -0.10620463008710458, "output_peak": 0.3651055616896942}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8410328313200884, "全帯域_10ms_rms_envelope_nrmse": 0.06343457047927145, "全帯域_10ms_envelope_correlation": 0.9960446314146784, "全帯域_10ms_modulation_mae_db": 0.4655486281237233, "全帯域_40ms_log_rms_mae_db": 0.45652778606848926, "全帯域_40ms_rms_envelope_nrmse": 0.03452359607347218, "全帯域_40ms_envelope_correlation": 0.9987921922586832, "全帯域_40ms_modulation_mae_db": 0.6737877576757318, "300–1000 Hz_10ms_log_rms_mae_db": 1.490938578649627, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0781523055193902, "300–1000 Hz_10ms_envelope_correlation": 0.9948217140744413, "300–1000 Hz_10ms_modulation_mae_db": 0.525844884283971, "300–1000 Hz_40ms_log_rms_mae_db": 0.8018488029830366, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0439905917394797, "300–1000 Hz_40ms_envelope_correlation": 0.9983534589671856, "300–1000 Hz_40ms_modulation_mae_db": 0.6229291778819074, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3708218514389559, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14367153874434654, "1000–3000 Hz_10ms_envelope_correlation": 0.9836844268348676, "1000–3000 Hz_10ms_modulation_mae_db": 0.6373725651881329, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7813107256929451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08766022223107253, "1000–3000 Hz_40ms_envelope_correlation": 0.9935717884485823, "1000–3000 Hz_40ms_modulation_mae_db": 0.8531738561049363, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3059565807032316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20215424082739358, "3000–6000 Hz_10ms_envelope_correlation": 0.9661124633201152, "3000–6000 Hz_10ms_modulation_mae_db": 1.387432119864411, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7472118118037387, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10387797617465504, "3000–6000 Hz_40ms_envelope_correlation": 0.9901116600384791, "3000–6000 Hz_40ms_modulation_mae_db": 1.8281180773200125, "mr_stft_db": 5.60727850826008, "band_mae_db": 0.2117044041867818, "rms_gain_error_db": -0.07917214912184209, "output_peak": 0.37345936728761836}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8456790881484229, "全帯域_10ms_rms_envelope_nrmse": 0.06767394586662365, "全帯域_10ms_envelope_correlation": 0.9954802127182815, "全帯域_10ms_modulation_mae_db": 0.3795540053453603, "全帯域_40ms_log_rms_mae_db": 0.467097139350151, "全帯域_40ms_rms_envelope_nrmse": 0.03735602371534815, "全帯域_40ms_envelope_correlation": 0.9985679905955185, "全帯域_40ms_modulation_mae_db": 0.403279848917083, "300–1000 Hz_10ms_log_rms_mae_db": 1.555736375959632, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08012258851368113, "300–1000 Hz_10ms_envelope_correlation": 0.9944583763045736, "300–1000 Hz_10ms_modulation_mae_db": 0.45282107487902207, "300–1000 Hz_40ms_log_rms_mae_db": 0.8494691794147199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045428371375105414, "300–1000 Hz_40ms_envelope_correlation": 0.9981134211723505, "300–1000 Hz_40ms_modulation_mae_db": 0.5939584100298995, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3972108368353813, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15056462781102056, "1000–3000 Hz_10ms_envelope_correlation": 0.9821657450934669, "1000–3000 Hz_10ms_modulation_mae_db": 0.5569719759234492, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8068397652013367, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09412437192785943, "1000–3000 Hz_40ms_envelope_correlation": 0.992897058419166, "1000–3000 Hz_40ms_modulation_mae_db": 0.8129618805055184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3494970560941963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21050702392739185, "3000–6000 Hz_10ms_envelope_correlation": 0.9631271559403831, "3000–6000 Hz_10ms_modulation_mae_db": 1.4321353230461067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8036519768157047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10667312877012143, "3000–6000 Hz_40ms_envelope_correlation": 0.9895762415124222, "3000–6000 Hz_40ms_modulation_mae_db": 1.9736837325448489, "mr_stft_db": 5.5991309378150484, "band_mae_db": 0.15762539403384665, "rms_gain_error_db": -0.0731741889061799, "output_peak": 0.3628600103622855}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8472893410988864, "全帯域_10ms_rms_envelope_nrmse": 0.06542746081765165, "全帯域_10ms_envelope_correlation": 0.9957727998248365, "全帯域_10ms_modulation_mae_db": 0.8320429715153912, "全帯域_40ms_log_rms_mae_db": 0.4439121877739859, "全帯域_40ms_rms_envelope_nrmse": 0.03359016969874943, "全帯域_40ms_envelope_correlation": 0.9988466002950839, "全帯域_40ms_modulation_mae_db": 0.9630538571557632, "300–1000 Hz_10ms_log_rms_mae_db": 1.458941373464485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07823805198178005, "300–1000 Hz_10ms_envelope_correlation": 0.994744714055477, "300–1000 Hz_10ms_modulation_mae_db": 0.8772844389226303, "300–1000 Hz_40ms_log_rms_mae_db": 0.7849991601856998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04333671902661801, "300–1000 Hz_40ms_envelope_correlation": 0.9983273110316897, "300–1000 Hz_40ms_modulation_mae_db": 1.0730213530563646, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4140446083713212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14723572731473117, "1000–3000 Hz_10ms_envelope_correlation": 0.9829875006645337, "1000–3000 Hz_10ms_modulation_mae_db": 0.8733337678712098, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8354678108263178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08492831818937199, "1000–3000 Hz_40ms_envelope_correlation": 0.9943338568647864, "1000–3000 Hz_40ms_modulation_mae_db": 1.0396644734355112, "3000–6000 Hz_10ms_log_rms_mae_db": 1.359053280159306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20951516843320242, "3000–6000 Hz_10ms_envelope_correlation": 0.9633756106915347, "3000–6000 Hz_10ms_modulation_mae_db": 1.3170957978349023, "3000–6000 Hz_40ms_log_rms_mae_db": 0.786544403935328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1081270409344192, "3000–6000 Hz_40ms_envelope_correlation": 0.9898444607421731, "3000–6000 Hz_40ms_modulation_mae_db": 1.9054721942092574, "mr_stft_db": 5.599227332533541, "band_mae_db": 0.20418566487958864, "rms_gain_error_db": -0.07113730939621549, "output_peak": 0.37133928467968347}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7770668687253557, "全帯域_10ms_rms_envelope_nrmse": 0.057380645937815174, "全帯域_10ms_envelope_correlation": 0.9967558437994942, "全帯域_10ms_modulation_mae_db": 0.3962684800526343, "全帯域_40ms_log_rms_mae_db": 0.4119450659958314, "全帯域_40ms_rms_envelope_nrmse": 0.03209067676225249, "全帯域_40ms_envelope_correlation": 0.998940476813919, "全帯域_40ms_modulation_mae_db": 0.38100329576366415, "300–1000 Hz_10ms_log_rms_mae_db": 1.4592246287586974, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0762825520523751, "300–1000 Hz_10ms_envelope_correlation": 0.9950403040535851, "300–1000 Hz_10ms_modulation_mae_db": 0.5994268903815588, "300–1000 Hz_40ms_log_rms_mae_db": 0.7920451477720801, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04487709357603328, "300–1000 Hz_40ms_envelope_correlation": 0.9982320223244773, "300–1000 Hz_40ms_modulation_mae_db": 0.6760368943479341, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3621489019179642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1367919815907963, "1000–3000 Hz_10ms_envelope_correlation": 0.9852628558796849, "1000–3000 Hz_10ms_modulation_mae_db": 0.7862849057130896, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7715906732376749, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07555758874680729, "1000–3000 Hz_40ms_envelope_correlation": 0.9954215509997261, "1000–3000 Hz_40ms_modulation_mae_db": 1.0455269983138276, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3767601992677447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17125819960068045, "3000–6000 Hz_10ms_envelope_correlation": 0.9763236987174762, "3000–6000 Hz_10ms_modulation_mae_db": 0.5495432632038058, "3000–6000 Hz_40ms_log_rms_mae_db": 0.806636680901633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09159301080028737, "3000–6000 Hz_40ms_envelope_correlation": 0.9926227251594583, "3000–6000 Hz_40ms_modulation_mae_db": 0.3826047299327146, "mr_stft_db": 5.604331455735244, "band_mae_db": 0.19230480603333452, "rms_gain_error_db": -0.06478497538305103, "output_peak": 0.3744809547525383}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.815367042130028, "全帯域_10ms_rms_envelope_nrmse": 0.06657899999649228, "全帯域_10ms_envelope_correlation": 0.9956460295094045, "全帯域_10ms_modulation_mae_db": 0.4394551667528133, "全帯域_40ms_log_rms_mae_db": 0.4315127423613699, "全帯域_40ms_rms_envelope_nrmse": 0.03720891275422263, "全帯域_40ms_envelope_correlation": 0.9985979987146655, "全帯域_40ms_modulation_mae_db": 0.44984129033270864, "300–1000 Hz_10ms_log_rms_mae_db": 1.5648033950407425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08259250339787808, "300–1000 Hz_10ms_envelope_correlation": 0.9942098941241464, "300–1000 Hz_10ms_modulation_mae_db": 0.48815891197031763, "300–1000 Hz_40ms_log_rms_mae_db": 0.826895667923558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048733294119492954, "300–1000 Hz_40ms_envelope_correlation": 0.9979502897533088, "300–1000 Hz_40ms_modulation_mae_db": 0.5732748169499685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3415464613281483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12888120035978062, "1000–3000 Hz_10ms_envelope_correlation": 0.9869166340877187, "1000–3000 Hz_10ms_modulation_mae_db": 0.692355851042367, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7738211009516026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07568779363792182, "1000–3000 Hz_40ms_envelope_correlation": 0.9952268190170322, "1000–3000 Hz_40ms_modulation_mae_db": 0.7952955401823125, "3000–6000 Hz_10ms_log_rms_mae_db": 1.362691334159301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20142896135964464, "3000–6000 Hz_10ms_envelope_correlation": 0.9662732216686746, "3000–6000 Hz_10ms_modulation_mae_db": 0.8184059261287397, "3000–6000 Hz_40ms_log_rms_mae_db": 0.800135263869007, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09496658636842625, "3000–6000 Hz_40ms_envelope_correlation": 0.9917517288095502, "3000–6000 Hz_40ms_modulation_mae_db": 1.329522375528892, "mr_stft_db": 5.600453307882225, "band_mae_db": 0.11984601775373067, "rms_gain_error_db": -0.08677636338763009, "output_peak": 0.370659620945856}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8549748728045056, "全帯域_10ms_rms_envelope_nrmse": 0.06561203812361018, "全帯域_10ms_envelope_correlation": 0.9958873448729311, "全帯域_10ms_modulation_mae_db": 0.44495223346515256, "全帯域_40ms_log_rms_mae_db": 0.4623958653761824, "全帯域_40ms_rms_envelope_nrmse": 0.036150403671083346, "全帯域_40ms_envelope_correlation": 0.9988253668982972, "全帯域_40ms_modulation_mae_db": 0.5220829633258077, "300–1000 Hz_10ms_log_rms_mae_db": 1.557025355809286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07684090313608583, "300–1000 Hz_10ms_envelope_correlation": 0.9951842216281841, "300–1000 Hz_10ms_modulation_mae_db": 0.5041213574988408, "300–1000 Hz_40ms_log_rms_mae_db": 0.807221476020437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04491749793055429, "300–1000 Hz_40ms_envelope_correlation": 0.9985015929664421, "300–1000 Hz_40ms_modulation_mae_db": 0.73244547197238, "1000–3000 Hz_10ms_log_rms_mae_db": 1.433153953379374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13979935299554508, "1000–3000 Hz_10ms_envelope_correlation": 0.9845863841370963, "1000–3000 Hz_10ms_modulation_mae_db": 0.4733877219988625, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8142639302886389, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08156359418840935, "1000–3000 Hz_40ms_envelope_correlation": 0.994605885420075, "1000–3000 Hz_40ms_modulation_mae_db": 0.5901783903629733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.356404962171918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1915440517798588, "3000–6000 Hz_10ms_envelope_correlation": 0.9694910505170533, "3000–6000 Hz_10ms_modulation_mae_db": 1.1594194681561245, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8065233211066175, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10393464320157525, "3000–6000 Hz_40ms_envelope_correlation": 0.9902514764315377, "3000–6000 Hz_40ms_modulation_mae_db": 1.5820544033999002, "mr_stft_db": 5.602792528516065, "band_mae_db": 0.19033863807059523, "rms_gain_error_db": -0.13393409894687403, "output_peak": 0.3772356325500895}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.483162667303823, "spectral_convergence": 0.18771767818814494, "spectral_convergence_status": "ok", "amplitude_floor": 9.54303147153e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.438599922950725, "spectral_convergence": 0.17830563505469277, "spectral_convergence_status": "ok", "amplitude_floor": 7.553612828866807e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.867201917651502, "spectral_convergence": 0.18342964265420275, "spectral_convergence_status": "ok", "amplitude_floor": 6.230060381303045e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.459798737171228e-05, "output_power": 5.8509340943923584e-05, "delta_db": 0.30048572454213973, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011571142287648336, "output_power": 0.0011256403749654123, "delta_db": -0.11976571864801734, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009701755098987012, "output_power": 0.0009249555237336057, "delta_db": -0.20729457256095424, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.8727504132367385e-05, "output_power": 4.815970573498444e-05, "delta_db": -0.0509034235218858, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.5070297613490014e-05, "output_power": 2.350589039941651e-05, "delta_db": -0.2798278272542566, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.1228013015272054e-05, "output_power": 2.87264018690235e-05, "delta_db": -0.362631199389696, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22015141098615829, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14956283015592375, "pitch": null, "output_rms_dbfs": -26.55706537064209, "output_peak": 0.37086702318697057, "reference_rms_dbfs": -26.407502540486163, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.596321502635351}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014140982422473965, "log_rms_mae_db": 0.8471597905970877, "rms_envelope_nrmse": 0.06197856194452814, "envelope_correlation": 0.9964075507406333, "rms_gain_error_db": -0.1495628025974394, "reference_mean_power": 0.0022888204000268467, "output_mean_power": 0.002211339560155557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0460573675773498e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006367944533168533, "output_power": 0.0006115610270001505, "reference_active": true, "delta_db": -0.17559470600436125}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.21105688014902e-05, "output_power": 8.421312308294031e-05, "reference_active": true, "delta_db": -0.38929690502800474}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2041191998800966e-05, "output_power": 1.2215162901398917e-05, "reference_active": true, "delta_db": 0.06229781751960198}], "active_band_count": 3, "active_band_mae_db": 0.20906314285065597}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274322977924003, "log_rms_mae_db": 0.46075362195210706, "rms_envelope_nrmse": 0.034622699291589794, "envelope_correlation": 0.9990038224664813, "rms_gain_error_db": -0.14956242482486, "reference_mean_power": 0.0023003276499839287, "output_mean_power": 0.002222457461558384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.201135437818204e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005685810358795958, "output_power": 0.0005476969112515304, "reference_active": true, "delta_db": -0.1625207866836283}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.696081479723381e-05, "output_power": 4.267678335297113e-05, "reference_active": true, "delta_db": -0.41543945349841604}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.36988853805451e-07, "output_power": 8.935341313024904e-07, "reference_active": true, "delta_db": -0.20623278274766538}], "active_band_count": 3, "active_band_mae_db": 0.2613976743099032}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012682596688393775, "log_rms_mae_db": 1.5337022510593965, "rms_envelope_nrmse": 0.07733604036530543, "envelope_correlation": 0.9951454273413276, "rms_gain_error_db": -0.1941662362552652, "reference_mean_power": 0.0016135328798776547, "output_mean_power": 0.001542983194953104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.436393406849114e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006358276871928121, "output_power": 0.0006017395609775035, "reference_active": true, "delta_db": -0.23930870556196948}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.784078725837931e-05, "output_power": 7.917537544602868e-05, "reference_active": true, "delta_db": -0.45106087961681146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0842784949487312e-05, "output_power": 1.0443443782818395e-05, "reference_active": true, "delta_db": -0.16297111004651071}], "active_band_count": 3, "active_band_mae_db": 0.28444689840843057}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011190972053745482, "log_rms_mae_db": 0.8126855860634641, "rms_envelope_nrmse": 0.04580603510246661, "envelope_correlation": 0.9984516610167261, "rms_gain_error_db": -0.19416621822240832, "reference_mean_power": 0.0016216465744125685, "output_mean_power": 0.001550742134892943}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.451026691855673e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005862792692297138, "output_power": 0.0005548899518516241, "reference_active": true, "delta_db": -0.23897677327846895}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.32096116317955e-05, "output_power": 3.924057109667306e-05, "reference_active": true, "delta_db": -0.418450435293119}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.999199919742325e-07, "output_power": 8.702657951393032e-07, "reference_active": true, "delta_db": -0.603133371853621}], "active_band_count": 3, "active_band_mae_db": 0.4201868601417364}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.381727808324483e-05, "log_rms_mae_db": 1.419916780226886, "rms_envelope_nrmse": 0.14747238105674698, "envelope_correlation": 0.9828254684320763, "rms_gain_error_db": -0.16519080140407316, "reference_mean_power": 6.869686019984948e-05, "output_mean_power": 6.613293661791663e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.573688907938692e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3586087172622807e-05, "output_power": 1.256643164497213e-05, "reference_active": true, "delta_db": -0.3388242359567557}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.445561080668119e-06, "output_power": 1.9581031885096268e-06, "reference_active": true, "delta_db": -0.9654293968435735}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901755177746097e-07, "output_power": 4.0721232266430243e-07, "reference_active": true, "delta_db": -1.611602779925711}], "active_band_count": 3, "active_band_mae_db": 0.9719521375753467}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.1112752780143365e-05, "log_rms_mae_db": 0.8017498476087267, "rms_envelope_nrmse": 0.08592683115222066, "envelope_correlation": 0.9940031318761968, "rms_gain_error_db": -0.16518415862658126, "reference_mean_power": 6.904136883729842e-05, "output_mean_power": 6.646468906827123e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.335283050026158e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2478970323107004e-05, "output_power": 1.1405156208532586e-05, "reference_active": true, "delta_db": -0.39077514013167164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3575517045580512e-06, "output_power": 1.073088172643849e-06, "reference_active": true, "delta_db": -1.0212097119452062}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9851801848052634e-08, "output_power": 2.5164464372661256e-08, "reference_active": true, "delta_db": -1.9966027139954046}], "active_band_count": 3, "active_band_mae_db": 1.1361958553574274}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.653785524000496e-05, "log_rms_mae_db": 1.3675511367805724, "rms_envelope_nrmse": 0.17935972760452928, "envelope_correlation": 0.9733062731004697, "rms_gain_error_db": -0.1692258742777964, "reference_mean_power": 1.4641647362050651e-05, "output_mean_power": 1.4082097775748283e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3222330564047269e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.030474467255676e-06, "output_power": 5.874365364694128e-06, "reference_active": true, "delta_db": -0.11390528416337732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.175765551110617e-06, "output_power": 1.9901442665669607e-06, "reference_active": true, "delta_db": -0.38727536512563604}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1480409466014962e-06, "output_power": 5.843700217568472e-07, "reference_active": true, "delta_db": -2.9326944962055967}], "active_band_count": 3, "active_band_mae_db": 1.1446250484982035}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.7197117785089643e-05, "log_rms_mae_db": 0.806150309235457, "rms_envelope_nrmse": 0.09784368397219054, "envelope_correlation": 0.9914990127130762, "rms_gain_error_db": -0.16923153581296838, "reference_mean_power": 1.4714471928104246e-05, "output_mean_power": 1.4152120807394729e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.175274300154406e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.965217773921107e-06, "output_power": 5.661950536089976e-06, "reference_active": true, "delta_db": -0.22660232245438064}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4021233434065406e-06, "output_power": 1.1424643173641131e-06, "reference_active": true, "delta_db": -0.8894357515207776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593565327256392e-08, "output_power": 4.138630909109218e-08, "reference_active": true, "delta_db": -3.651233402936298}], "active_band_count": 3, "active_band_mae_db": 1.5890904923038187}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8968315497001479, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3b7af3e82e52954de830ea46cbf1493ed3db6151df421e10dcd0c0543b843979", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9844265790245067, "standard_deviation": 0.0018423733145347743, "minimum": 0.9821657450934669, "maximum": 0.9869836170666098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.391398433106943, "standard_deviation": 0.03132433329978315, "minimum": 1.3415464613281483, "maximum": 1.433153953379374, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6886612667954372, "standard_deviation": 0.17706667627654823, "minimum": 0.4733877219988625, "maximum": 0.9719521375753467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14046274397819414, "standard_deviation": 0.008283703966238551, "minimum": 0.12888120035978062, "maximum": 0.15056462781102056, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945777327397329, "standard_deviation": 0.001152436785972163, "minimum": 0.992897058419166, "maximum": 0.9965617708722992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7961968839583125, "standard_deviation": 0.02228566781991871, "minimum": 0.7715906732376749, "maximum": 0.8354678108263178, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8551081153039686, "standard_deviation": 0.2095515595651767, "minimum": 0.5678679281692417, "maximum": 1.1361958553574274, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08165191239134832, "standard_deviation": 0.008334040132255405, "minimum": 0.06776657905712351, "maximum": 0.09412437192785943, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9682080438143437, "standard_deviation": 0.00464620770689895, "minimum": 0.9631271559403831, "maximum": 0.9763236987174762, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.355095893535688, "standard_deviation": 0.02140002016089181, "minimum": 1.3059565807032316, "maximum": 1.3767601992677447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1117432880956701, "standard_deviation": 0.29928361033835227, "minimum": 0.5495432632038058, "maximum": 1.4321353230461067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.195582739528466, "standard_deviation": 0.014022217964736908, "minimum": 0.17125819960068045, "maximum": 0.21050702392739185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906457523666086, "standard_deviation": 0.0011575309759558644, "minimum": 0.9895087135261718, "maximum": 0.9926227251594583, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7934415975688206, "standard_deviation": 0.02017143073775921, "minimum": 0.7472118118037387, "maximum": 0.806636680901633, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4628374158280155, "standard_deviation": 0.5248301612607905, "minimum": 0.3826047299327146, "maximum": 1.9736837325448489, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10184780017941472, "standard_deviation": 0.0062651297303324294, "minimum": 0.09159301080028737, "maximum": 0.1081270409344192, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9948378420596621, "standard_deviation": 0.0003526533120869626, "minimum": 0.9942098941241464, "maximum": 0.9951842216281841, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5118017571637246, "standard_deviation": 0.045800605867560025, "minimum": 1.458941373464485, "maximum": 1.5648033950407425, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5335119847963352, "standard_deviation": 0.16649940258235477, "minimum": 0.28444689840843057, "maximum": 0.8772844389226303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07827265032414127, "standard_deviation": 0.0021290624337034904, "minimum": 0.0762825520523751, "maximum": 0.08259250339787808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982892145825772, "standard_deviation": 0.00018326132998184615, "minimum": 0.9979502897533088, "maximum": 0.9985015929664421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8079045166089684, "standard_deviation": 0.021770952831982937, "minimum": 0.7849991601856998, "maximum": 0.8494691794147199, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6879056456084107, "standard_deviation": 0.194035819671902, "minimum": 0.4201868601417364, "maximum": 1.0730213530563646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04521606120208439, "standard_deviation": 0.0016184182778508627, "minimum": 0.04333671902661801, "maximum": 0.048733294119492954, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18175939390791201, "standard_deviation": 0.03389582080666581, "minimum": 0.11984601775373067, "maximum": 0.22015141098615829, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.600639058637224, "standard_deviation": 0.003980707146788908, "minimum": 5.59557689572024, "maximum": 5.60727850826008, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.37075093193184194, "standard_deviation": 0.004748004944139708, "minimum": 0.3628600103622855, "maximum": 0.3772356325500895, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09559331817310263, "standard_deviation": 0.031379212470981833, "minimum": -0.14956283015592375, "maximum": -0.06478497538305103, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9960527639349439, "standard_deviation": 0.0004407507023718608, "minimum": 0.9954802127182815, "maximum": 0.9967558437994942, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8329151351339837, "standard_deviation": 0.02550997354914534, "minimum": 0.7770668687253557, "maximum": 0.8549748728045056, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4538622836526067, "standard_deviation": 0.17415616685073124, "minimum": 0.20906314285065597, "maximum": 0.8320429715153912, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0636093053044914, "standard_deviation": 0.0034247335196329097, "minimum": 0.057380645937815174, "maximum": 0.06767394586662365, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988245462626044, "standard_deviation": 0.00017038422623107013, "minimum": 0.9985679905955185, "maximum": 0.9990219220581863, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44478842907801075, "standard_deviation": 0.020289003894923673, "minimum": 0.4119450659958314, "maximum": 0.467097139350151, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5404938548097767, "standard_deviation": 0.2215801450702333, "minimum": 0.2613976743099032, "maximum": 0.9630538571557632, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.034729282405678276, "standard_deviation": 0.0020466990076849793, "minimum": 0.03209067676225249, "maximum": 0.03735602371534815, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4507-16021-0032", "language": "en", "style": "normal", "speaker": "librispeech_4507", "source_utterance_id": "4507-16021-0032", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4507-16021-0032.wav", "source_file_sha256": "66d9c3987b9a67d196faf75935bc8c492dcfb4a3e22e6b21bf4daf50a2c9df82", "decoded_pcm_sha256": "7576336f86f07936b03cf91b64bf80c306a8a013d42a9e0eb9b1dae02be902c2", "original_source_sha256": "cf872613e0d590501ff569c17e5f9d525920c4babeda14ac94577f60cfc54401", "corpus_entry_id": "librispeech_4507-16021-0032"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 2.9888720512390137, "predicted_mos": 2.9888720512390137, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "939abfd897c6303be34824829576e19f920f3abb7f09fa57d03d4b550dc06020", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.645950804999302, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4507-16021-0032", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "939abfd897c6303be34824829576e19f920f3abb7f09fa57d03d4b550dc06020", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "ef0e7a805e8d327859d4b390159adeb5aea5e9adfcd0a91a422cd585becc0a79"}, "rows": [{"case": "librispeech_4507-16021-0032", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.7723941802978516, "predicted_mos": 2.7723941802978516, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "14aad94641ee9bfe902bf84db8aaf14f1cd49383dfda85b1ffc171da7a12cfd9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4962389430002077, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4507-16021-0032", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "14aad94641ee9bfe902bf84db8aaf14f1cd49383dfda85b1ffc171da7a12cfd9", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7656536738664086, "全帯域_10ms_rms_envelope_nrmse": 0.07812715187406447, "全帯域_10ms_envelope_correlation": 0.9941755301469443, "全帯域_10ms_modulation_mae_db": 0.4634483891366384, "全帯域_40ms_log_rms_mae_db": 0.42056371165590545, "全帯域_40ms_rms_envelope_nrmse": 0.03534377751118292, "全帯域_40ms_envelope_correlation": 0.998724042375039, "全帯域_40ms_modulation_mae_db": 0.46261710015757423, "300–1000 Hz_10ms_log_rms_mae_db": 1.4154265931196546, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0974443023992325, "300–1000 Hz_10ms_envelope_correlation": 0.9925657315012331, "300–1000 Hz_10ms_modulation_mae_db": 0.6762987238974597, "300–1000 Hz_40ms_log_rms_mae_db": 0.8260206803990873, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044176123972054795, "300–1000 Hz_40ms_envelope_correlation": 0.9984790847646888, "300–1000 Hz_40ms_modulation_mae_db": 0.7224004895016224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4007952915035924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1844328006651495, "1000–3000 Hz_10ms_envelope_correlation": 0.9771263171404044, "1000–3000 Hz_10ms_modulation_mae_db": 0.670031421914247, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8767922723151352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1155913398946023, "1000–3000 Hz_40ms_envelope_correlation": 0.9914346382705621, "1000–3000 Hz_40ms_modulation_mae_db": 0.8805060760394591, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4094058708427686, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3711364976223417, "3000–6000 Hz_10ms_envelope_correlation": 0.905461179558946, "3000–6000 Hz_10ms_modulation_mae_db": 0.7550797444280558, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9795720849977851, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2854453954415478, "3000–6000 Hz_40ms_envelope_correlation": 0.9415161304599206, "3000–6000 Hz_40ms_modulation_mae_db": 0.7733578423164028, "mr_stft_db": 5.644350626331914, "band_mae_db": 0.3901195430345914, "rms_gain_error_db": -0.029053093588536978, "output_peak": 0.48079240199429246}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7802004826370984, "全帯域_10ms_rms_envelope_nrmse": 0.09155412340712664, "全帯域_10ms_envelope_correlation": 0.9921148972091606, "全帯域_10ms_modulation_mae_db": 0.7467308004912975, "全帯域_40ms_log_rms_mae_db": 0.41452701298083555, "全帯域_40ms_rms_envelope_nrmse": 0.05255427092533497, "全帯域_40ms_envelope_correlation": 0.9972505416659685, "全帯域_40ms_modulation_mae_db": 0.8324379835463797, "300–1000 Hz_10ms_log_rms_mae_db": 1.3714341895703965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11784317626223503, "300–1000 Hz_10ms_envelope_correlation": 0.9891702439986243, "300–1000 Hz_10ms_modulation_mae_db": 0.7102114969662928, "300–1000 Hz_40ms_log_rms_mae_db": 0.835762035204348, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06813335955269523, "300–1000 Hz_40ms_envelope_correlation": 0.9961785208861162, "300–1000 Hz_40ms_modulation_mae_db": 0.773421217169116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.429867123230722, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18759519505380273, "1000–3000 Hz_10ms_envelope_correlation": 0.9734606519855632, "1000–3000 Hz_10ms_modulation_mae_db": 0.42464547474921316, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9021612166528049, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09541600584829524, "1000–3000 Hz_40ms_envelope_correlation": 0.9929843979316837, "1000–3000 Hz_40ms_modulation_mae_db": 0.2799706696396147, "3000–6000 Hz_10ms_log_rms_mae_db": 1.451986215117893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3762770907030337, "3000–6000 Hz_10ms_envelope_correlation": 0.9011551905934579, "3000–6000 Hz_10ms_modulation_mae_db": 0.492037239520618, "3000–6000 Hz_40ms_log_rms_mae_db": 0.991241499854766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2741668228236028, "3000–6000 Hz_40ms_envelope_correlation": 0.9452185385900327, "3000–6000 Hz_40ms_modulation_mae_db": 0.5119824294667755, "mr_stft_db": 5.651576612314123, "band_mae_db": 0.3219133984883112, "rms_gain_error_db": 0.04403461250688516, "output_peak": 0.49307106434173165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7703948110071657, "全帯域_10ms_rms_envelope_nrmse": 0.0805063087322838, "全帯域_10ms_envelope_correlation": 0.9938163371206543, "全帯域_10ms_modulation_mae_db": 0.5414013165168164, "全帯域_40ms_log_rms_mae_db": 0.423124577123263, "全帯域_40ms_rms_envelope_nrmse": 0.0417022487924543, "全帯域_40ms_envelope_correlation": 0.9982313557431504, "全帯域_40ms_modulation_mae_db": 0.47787708316628436, "300–1000 Hz_10ms_log_rms_mae_db": 1.3807325582684495, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09117886277780936, "300–1000 Hz_10ms_envelope_correlation": 0.9934754932621908, "300–1000 Hz_10ms_modulation_mae_db": 1.0034517874846702, "300–1000 Hz_40ms_log_rms_mae_db": 0.7622251716306047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04460631244570817, "300–1000 Hz_40ms_envelope_correlation": 0.9983951088366372, "300–1000 Hz_40ms_modulation_mae_db": 1.0409135023284466, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3887692208376867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19070113606246736, "1000–3000 Hz_10ms_envelope_correlation": 0.9739148665169427, "1000–3000 Hz_10ms_modulation_mae_db": 0.5195231183906212, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8749735829102325, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08910635138504677, "1000–3000 Hz_40ms_envelope_correlation": 0.9943285422790772, "1000–3000 Hz_40ms_modulation_mae_db": 0.36936196919805986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4283929903091621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.375153712516128, "3000–6000 Hz_10ms_envelope_correlation": 0.90774106733994, "3000–6000 Hz_10ms_modulation_mae_db": 0.6025883071347059, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9676991962691103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.28814538149294366, "3000–6000 Hz_40ms_envelope_correlation": 0.9443239870093348, "3000–6000 Hz_40ms_modulation_mae_db": 0.5252810857929014, "mr_stft_db": 5.664042047704413, "band_mae_db": 0.2976203038382677, "rms_gain_error_db": -0.042336031294354814, "output_peak": 0.44943901051172463}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.738599894632151, "全帯域_10ms_rms_envelope_nrmse": 0.08361600542698719, "全帯域_10ms_envelope_correlation": 0.9933579351156214, "全帯域_10ms_modulation_mae_db": 0.6010607209294305, "全帯域_40ms_log_rms_mae_db": 0.3950846004713734, "全帯域_40ms_rms_envelope_nrmse": 0.04351015443898291, "全帯域_40ms_envelope_correlation": 0.9981262818796678, "全帯域_40ms_modulation_mae_db": 0.5951593343732577, "300–1000 Hz_10ms_log_rms_mae_db": 1.4104983086894876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10200472549762664, "300–1000 Hz_10ms_envelope_correlation": 0.9919447753861389, "300–1000 Hz_10ms_modulation_mae_db": 0.7202642519859497, "300–1000 Hz_40ms_log_rms_mae_db": 0.795965250718838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054381920687466764, "300–1000 Hz_40ms_envelope_correlation": 0.9977844411286652, "300–1000 Hz_40ms_modulation_mae_db": 0.7515469479217831, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3770890474905688, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21448990954225236, "1000–3000 Hz_10ms_envelope_correlation": 0.9743254712764292, "1000–3000 Hz_10ms_modulation_mae_db": 0.34080559033771846, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8795053626299192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16883821074463495, "1000–3000 Hz_40ms_envelope_correlation": 0.9856381229190148, "1000–3000 Hz_40ms_modulation_mae_db": 0.5805351441171501, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4554225819839985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.361619831535582, "3000–6000 Hz_10ms_envelope_correlation": 0.9051250101578336, "3000–6000 Hz_10ms_modulation_mae_db": 0.4877300199497068, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9784859467577198, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2810610125273745, "3000–6000 Hz_40ms_envelope_correlation": 0.9387853957703107, "3000–6000 Hz_40ms_modulation_mae_db": 0.5660875777735731, "mr_stft_db": 5.643225689264198, "band_mae_db": 0.42239584074183023, "rms_gain_error_db": -0.08046477349966752, "output_peak": 0.4296999234350646}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7390421640826765, "全帯域_10ms_rms_envelope_nrmse": 0.08731524518710634, "全帯域_10ms_envelope_correlation": 0.9927633329178059, "全帯域_10ms_modulation_mae_db": 0.6029259528512133, "全帯域_40ms_log_rms_mae_db": 0.4133517131084064, "全帯域_40ms_rms_envelope_nrmse": 0.0468730420731689, "全帯域_40ms_envelope_correlation": 0.9978242446780297, "全帯域_40ms_modulation_mae_db": 0.6977647888700585, "300–1000 Hz_10ms_log_rms_mae_db": 1.375506137282295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09969676991080072, "300–1000 Hz_10ms_envelope_correlation": 0.9922315275222157, "300–1000 Hz_10ms_modulation_mae_db": 0.8022012214396986, "300–1000 Hz_40ms_log_rms_mae_db": 0.7683833668921348, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05581269128362451, "300–1000 Hz_40ms_envelope_correlation": 0.9975141540743688, "300–1000 Hz_40ms_modulation_mae_db": 0.8961498651758842, "1000–3000 Hz_10ms_log_rms_mae_db": 1.392573563099888, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18190440559992976, "1000–3000 Hz_10ms_envelope_correlation": 0.9760902098102733, "1000–3000 Hz_10ms_modulation_mae_db": 0.5977418096535316, "1000–3000 Hz_40ms_log_rms_mae_db": 0.895128816779515, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09624508903711206, "1000–3000 Hz_40ms_envelope_correlation": 0.9931752305654188, "1000–3000 Hz_40ms_modulation_mae_db": 0.6101001538103147, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4736461040858089, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3559563407808805, "3000–6000 Hz_10ms_envelope_correlation": 0.9143505764292851, "3000–6000 Hz_10ms_modulation_mae_db": 0.3781828961366089, "3000–6000 Hz_40ms_log_rms_mae_db": 1.001649215796441, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27388249547472754, "3000–6000 Hz_40ms_envelope_correlation": 0.9472586937543065, "3000–6000 Hz_40ms_modulation_mae_db": 0.31770629090956587, "mr_stft_db": 5.643386650455739, "band_mae_db": 0.30540187808865354, "rms_gain_error_db": -0.08578546176717979, "output_peak": 0.43828139846145675}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7244602993073266, "全帯域_10ms_rms_envelope_nrmse": 0.08268374527350145, "全帯域_10ms_envelope_correlation": 0.9935620271936262, "全帯域_10ms_modulation_mae_db": 0.57797195585493, "全帯域_40ms_log_rms_mae_db": 0.3810047893637573, "全帯域_40ms_rms_envelope_nrmse": 0.04436340545268918, "全帯域_40ms_envelope_correlation": 0.9981267761956634, "全帯域_40ms_modulation_mae_db": 0.628393570039291, "300–1000 Hz_10ms_log_rms_mae_db": 1.4192835484509525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09773819905464946, "300–1000 Hz_10ms_envelope_correlation": 0.9928112014378077, "300–1000 Hz_10ms_modulation_mae_db": 0.8016582185956566, "300–1000 Hz_40ms_log_rms_mae_db": 0.8051183368949697, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054052003665087694, "300–1000 Hz_40ms_envelope_correlation": 0.9980395651700421, "300–1000 Hz_40ms_modulation_mae_db": 0.949195433004582, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3738581073037601, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17018660289760693, "1000–3000 Hz_10ms_envelope_correlation": 0.9810919245685569, "1000–3000 Hz_10ms_modulation_mae_db": 0.43023506293593144, "1000–3000 Hz_40ms_log_rms_mae_db": 0.871685889797717, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11520184027959875, "1000–3000 Hz_40ms_envelope_correlation": 0.9918810460206172, "1000–3000 Hz_40ms_modulation_mae_db": 0.5969642561156768, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4401285631027059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3501086859753304, "3000–6000 Hz_10ms_envelope_correlation": 0.9125897892395801, "3000–6000 Hz_10ms_modulation_mae_db": 0.3950585828415097, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0097735853627179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2623899797284124, "3000–6000 Hz_40ms_envelope_correlation": 0.9480649822625117, "3000–6000 Hz_40ms_modulation_mae_db": 0.4519353938684176, "mr_stft_db": 5.6505261983930595, "band_mae_db": 0.36859892244433673, "rms_gain_error_db": -0.11271501470324656, "output_peak": 0.42362074576059827}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7813788004107801, "全帯域_10ms_rms_envelope_nrmse": 0.08359230221380771, "全帯域_10ms_envelope_correlation": 0.9935888838076684, "全帯域_10ms_modulation_mae_db": 0.35141509446400687, "全帯域_40ms_log_rms_mae_db": 0.4362417770366978, "全帯域_40ms_rms_envelope_nrmse": 0.04725869837095853, "全帯域_40ms_envelope_correlation": 0.9980755910279884, "全帯域_40ms_modulation_mae_db": 0.2974912255884244, "300–1000 Hz_10ms_log_rms_mae_db": 1.4421012321898279, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1198490684012302, "300–1000 Hz_10ms_envelope_correlation": 0.9893859111379538, "300–1000 Hz_10ms_modulation_mae_db": 0.4454687293359525, "300–1000 Hz_40ms_log_rms_mae_db": 0.8673332214704211, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07896191019599313, "300–1000 Hz_40ms_envelope_correlation": 0.9957859552025609, "300–1000 Hz_40ms_modulation_mae_db": 0.4203250223710278, "1000–3000 Hz_10ms_log_rms_mae_db": 1.445483111771706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24066654319163883, "1000–3000 Hz_10ms_envelope_correlation": 0.9666071806220894, "1000–3000 Hz_10ms_modulation_mae_db": 0.5400021898097246, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9406035360218552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1774588273100808, "1000–3000 Hz_40ms_envelope_correlation": 0.983843104165481, "1000–3000 Hz_40ms_modulation_mae_db": 0.6887899928957942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4315930812890312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3607846301797391, "3000–6000 Hz_10ms_envelope_correlation": 0.9182895384890621, "3000–6000 Hz_10ms_modulation_mae_db": 0.6473612813822399, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9697599421090688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27084114706138285, "3000–6000 Hz_40ms_envelope_correlation": 0.9545273698268868, "3000–6000 Hz_40ms_modulation_mae_db": 0.6113053000733121, "mr_stft_db": 5.6561325280747825, "band_mae_db": 0.5383933550318308, "rms_gain_error_db": -0.1724737679874921, "output_peak": 0.45665734738091557}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7610720081542393, "全帯域_10ms_rms_envelope_nrmse": 0.08119516347172567, "全帯域_10ms_envelope_correlation": 0.9937662334125524, "全帯域_10ms_modulation_mae_db": 0.7280508346044416, "全帯域_40ms_log_rms_mae_db": 0.39587177559052084, "全帯域_40ms_rms_envelope_nrmse": 0.04269958914821842, "全帯域_40ms_envelope_correlation": 0.9982343770413777, "全帯域_40ms_modulation_mae_db": 0.7625570187197592, "300–1000 Hz_10ms_log_rms_mae_db": 1.3721349609463507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09893613624051766, "300–1000 Hz_10ms_envelope_correlation": 0.9924599194260045, "300–1000 Hz_10ms_modulation_mae_db": 0.5543563477704465, "300–1000 Hz_40ms_log_rms_mae_db": 0.7616596177484928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054359933865845124, "300–1000 Hz_40ms_envelope_correlation": 0.9978083168400066, "300–1000 Hz_40ms_modulation_mae_db": 0.5809801828242654, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4643300044470393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20216370608222664, "1000–3000 Hz_10ms_envelope_correlation": 0.969190372844297, "1000–3000 Hz_10ms_modulation_mae_db": 0.2451597946144892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8816406563001956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09008920798654786, "1000–3000 Hz_40ms_envelope_correlation": 0.9944760304322493, "1000–3000 Hz_40ms_modulation_mae_db": 0.4400248554085297, "3000–6000 Hz_10ms_log_rms_mae_db": 1.42638968148415, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35218253191726234, "3000–6000 Hz_10ms_envelope_correlation": 0.9166470222382974, "3000–6000 Hz_10ms_modulation_mae_db": 0.3375398483065524, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9760522477819692, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27954976193509096, "3000–6000 Hz_40ms_envelope_correlation": 0.945453902455988, "3000–6000 Hz_40ms_modulation_mae_db": 0.249595554275212, "mr_stft_db": 5.648184617670656, "band_mae_db": 0.3482727260006047, "rms_gain_error_db": -0.09731264639455309, "output_peak": 0.419062579688579}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.399930818466698, "spectral_convergence": 0.26219529114101253, "spectral_convergence_status": "ok", "amplitude_floor": 1.3235749190080935e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.504184495208703, "spectral_convergence": 0.2942684830655435, "spectral_convergence_status": "ok", "amplitude_floor": 1.1987597823804542e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.028936565320341, "spectral_convergence": 0.295489415400422, "spectral_convergence_status": "ok", "amplitude_floor": 9.377214849682438e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.350500895926799e-05, "output_power": 5.1490037594175796e-05, "delta_db": 0.7318394648770872, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012608807022653607, "output_power": 0.0012435293103024408, "delta_db": -0.06017971756146703, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00036191686574782775, "output_power": 0.00034968438054261544, "delta_db": -0.14932588228645885, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.9809147109830135e-05, "output_power": 3.536378578693199e-05, "delta_db": 0.7420920089438465, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.6488056661262093e-05, "output_power": 1.8519163459542136e-05, "delta_db": 0.5045189377918795, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.109371525887873e-05, "output_power": 6.328090202879659e-05, "delta_db": 0.15276124674680971, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3901195430345914, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.029053093588536978, "pitch": null, "output_rms_dbfs": -27.54026735856096, "output_peak": 0.48079240199429246, "reference_rms_dbfs": -27.511214264972427, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.644350626331914}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018136510927426692, "log_rms_mae_db": 0.7656536738664086, "rms_envelope_nrmse": 0.07812715187406447, "envelope_correlation": 0.9941755301469443, "rms_gain_error_db": -0.028913442613555343, "reference_mean_power": 0.0017749427149271913, "output_mean_power": 0.001763165164721956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.578419076835135e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003061888162786739, "output_power": 0.00030386110735548013, "reference_active": true, "delta_db": -0.03314207600376219}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.561280729204581e-05, "output_power": 5.277733040452605e-05, "reference_active": true, "delta_db": -0.22727399631196868}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1447257663776473e-05, "output_power": 8.82487472204928e-06, "reference_active": true, "delta_db": -1.1299290950941843}], "active_band_count": 3, "active_band_mae_db": 0.4634483891366384}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015531555176710683, "log_rms_mae_db": 0.42056371165590545, "rms_envelope_nrmse": 0.03534377751118292, "envelope_correlation": 0.998724042375039, "rms_gain_error_db": -0.02891760338077872, "reference_mean_power": 0.0017815693662331493, "output_mean_power": 0.0017697461496802328}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.067921014076438e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002861407755621019, "output_power": 0.00028378162832990946, "reference_active": true, "delta_db": -0.03595473593947798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5605898871779113e-05, "output_power": 2.433777276926876e-05, "reference_active": true, "delta_db": -0.2205919397415423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.159270454376389e-07, "output_power": 7.058792580785061e-07, "reference_active": true, "delta_db": -1.1313046247917022}], "active_band_count": 3, "active_band_mae_db": 0.46261710015757423}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017559797263300874, "log_rms_mae_db": 1.4154265931196546, "rms_envelope_nrmse": 0.0974443023992325, "envelope_correlation": 0.9925657315012331, "rms_gain_error_db": -0.1061498311814307, "reference_mean_power": 0.0012870696988326355, "output_mean_power": 0.0012559926062744272}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.703155295091611e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025376682215949657, "output_power": 0.000248486890709417, "reference_active": true, "delta_db": -0.09131359351931329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.494529209431561e-05, "output_power": 4.0797715903637474e-05, "reference_active": true, "delta_db": -0.4204835779477566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.800066525463509e-06, "output_power": 6.205489645067107e-06, "reference_active": true, "delta_db": -1.517099000225309}], "active_band_count": 3, "active_band_mae_db": 0.6762987238974597}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014992083957937532, "log_rms_mae_db": 0.8260206803990873, "rms_envelope_nrmse": 0.044176123972054795, "envelope_correlation": 0.9984790847646888, "rms_gain_error_db": -0.10636133526520554, "reference_mean_power": 0.0012919551058812144, "output_mean_power": 0.0012606986538585252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4067631484646013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002445691472192921, "output_power": 0.00023811843891017649, "reference_active": true, "delta_db": -0.11608642545911681}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8830927140807284e-05, "output_power": 1.7153801115807027e-05, "reference_active": true, "delta_db": -0.4051133255033981}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.050384653726169e-07, "output_power": 4.82628139855598e-07, "reference_active": true, "delta_db": -1.6460017175423522}], "active_band_count": 3, "active_band_mae_db": 0.7224004895016224}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.2493337902299445e-05, "log_rms_mae_db": 1.4007952915035924, "rms_envelope_nrmse": 0.1844328006651495, "envelope_correlation": 0.9771263171404044, "rms_gain_error_db": 0.3341124919348789, "reference_mean_power": 4.199153922743893e-05, "output_mean_power": 4.5349556424502295e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0299126624198213e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.222152478948813e-06, "output_power": 3.8108442379259864e-06, "reference_active": true, "delta_db": -0.4451271532682211}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.960031981806193e-07, "output_power": 7.926962250580271e-07, "reference_active": true, "delta_db": -0.5320276980835918}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.633269352164066e-07, "output_power": 2.0758756903616274e-07, "reference_active": true, "delta_db": -1.032939414390928}], "active_band_count": 3, "active_band_mae_db": 0.670031421914247}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.820326553727612e-05, "log_rms_mae_db": 0.8767922723151352, "rms_envelope_nrmse": 0.1155913398946023, "envelope_correlation": 0.9914346382705621, "rms_gain_error_db": 0.3337564064395105, "reference_mean_power": 4.2190507757870256e-05, "output_mean_power": 4.5560700496292236e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.548028980657037e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.958931660389438e-06, "output_power": 3.5465058982899684e-06, "reference_active": true, "delta_db": -0.4777731849511624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7410748918107156e-07, "output_power": 3.2039503273336156e-07, "reference_active": true, "delta_db": -0.6731062798762367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8002833142836267e-08, "output_power": 1.2772531427703687e-08, "reference_active": true, "delta_db": -1.4906387632909779}], "active_band_count": 3, "active_band_mae_db": 0.8805060760394591}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.219318962627022e-05, "log_rms_mae_db": 1.4094058708427686, "rms_envelope_nrmse": 0.3711364976223417, "envelope_correlation": 0.905461179558946, "rms_gain_error_db": 0.7212080263560268, "reference_mean_power": 2.2095607497389627e-05, "output_mean_power": 2.6087156837763262e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.718240736767725e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.504526854650138e-06, "output_power": 9.813814208897086e-06, "reference_active": true, "delta_db": 0.13907329734643886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5665611692910136e-06, "output_power": 2.85730007836662e-06, "reference_active": true, "delta_db": 0.46604233765951697}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.91578688150447e-07, "output_power": 8.670115435862694e-07, "reference_active": true, "delta_db": 1.6601235982782114}], "active_band_count": 3, "active_band_mae_db": 0.7550797444280558}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.1857789632782756e-05, "log_rms_mae_db": 0.9795720849977851, "rms_envelope_nrmse": 0.2854453954415478, "envelope_correlation": 0.9415161304599206, "rms_gain_error_db": 0.7211219838277303, "reference_mean_power": 2.220381322589703e-05, "output_mean_power": 2.621439045734807e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6122764663239433e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.855212413255228e-06, "output_power": 1.0406863001486469e-05, "reference_active": true, "delta_db": 0.23653848204085942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3142139667203495e-06, "output_power": 1.5548825786838368e-06, "reference_active": true, "delta_db": 0.7303151932171594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2621117296192714e-08, "output_power": 5.820319367857249e-08, "reference_active": true, "delta_db": 1.3532198516911895}], "active_band_count": 3, "active_band_mae_db": 0.7733578423164028}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8714040042824527, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9b2b3cca75cb26fb0b8d367e39f376ad501998611e9b83f943bc11e29181b269", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9739758743455695, "standard_deviation": 0.004513850761750951, "minimum": 0.9666071806220894, "maximum": 0.9810919245685569, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4090956837106203, "standard_deviation": 0.03343967155126289, "minimum": 1.3738581073037601, "maximum": 1.4643300044470393, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47101805780068456, "standard_deviation": 0.13864532891653344, "minimum": 0.2451597946144892, "maximum": 0.670031421914247, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19651753738688427, "standard_deviation": 0.022265801470243565, "minimum": 0.17018660289760693, "maximum": 0.24066654319163883, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.990970139073013, "standard_deviation": 0.004013554665456944, "minimum": 0.983843104165481, "maximum": 0.9944760304322493, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8903114166759218, "standard_deviation": 0.022834190909550762, "minimum": 0.871685889797717, "maximum": 0.9406035360218552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5557816396530749, "standard_deviation": 0.18996993908509552, "minimum": 0.2799706696396147, "maximum": 0.8805060760394591, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11849335906073985, "standard_deviation": 0.03531233721434563, "minimum": 0.08910635138504677, "maximum": 0.1774588273100808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9101699217558004, "standard_deviation": 0.006164445711426828, "minimum": 0.9011551905934579, "maximum": 0.9182895384890621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4396206360269397, "standard_deviation": 0.020139128103931234, "minimum": 1.4094058708427686, "maximum": 1.4736461040858089, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5119472399624997, "standard_deviation": 0.14571148516803992, "minimum": 0.3375398483065524, "maximum": 0.7550797444280558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.36290241515378724, "standard_deviation": 0.010210287062086204, "minimum": 0.3501086859753304, "maximum": 0.3762770907030337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9456436250161615, "standard_deviation": 0.0046816052969421125, "minimum": 0.9387853957703107, "maximum": 0.9545273698268868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9842792148661973, "standard_deviation": 0.015162272653097091, "minimum": 0.9676991962691103, "maximum": 1.0097735853627179, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.50090643430952, "standard_deviation": 0.16496782189309883, "minimum": 0.249595554275212, "maximum": 0.7733578423164028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2769352495606353, "standard_deviation": 0.008346373804053914, "minimum": 0.2623899797284124, "maximum": 0.28814538149294366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9917556004590211, "standard_deviation": 0.0015940286528095481, "minimum": 0.9891702439986243, "maximum": 0.9934754932621908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.398389691064677, "standard_deviation": 0.026820749012438746, "minimum": 1.3714341895703965, "maximum": 1.4421012321898279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7142388471845158, "standard_deviation": 0.16831747999142543, "minimum": 0.4454687293359525, "maximum": 1.0034517874846702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1030864050680127, "standard_deviation": 0.010218049577049653, "minimum": 0.09117886277780936, "maximum": 0.1198490684012302, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974981433628858, "standard_deviation": 0.0009936478813781999, "minimum": 0.9957859552025609, "maximum": 0.9984790847646888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8028084601198621, "standard_deviation": 0.038481299226197384, "minimum": 0.7616596177484928, "maximum": 0.8673332214704211, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.766866582537091, "standard_deviation": 0.20069856057568342, "minimum": 0.4203250223710278, "maximum": 1.0409135023284466, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056810531958559424, "standard_deviation": 0.011637193879406165, "minimum": 0.044176123972054795, "maximum": 0.07896191019599313, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3740894959585533, "standard_deviation": 0.07889481859677265, "minimum": 0.2976203038382677, "maximum": 0.5383933550318308, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.65017812127611, "standard_deviation": 0.0071926002841217335, "minimum": 5.643225689264198, "maximum": 5.664042047704413, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44882805894679534, "standard_deviation": 0.02683106882786725, "minimum": 0.419062579688579, "maximum": 0.49307106434173165, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07201327209101821, "standard_deviation": 0.06419115380758246, "minimum": -0.1724737679874921, "maximum": 0.04403461250688516, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933931471155042, "standard_deviation": 0.0006578176406509279, "minimum": 0.9921148972091606, "maximum": 0.9941755301469443, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7576002667622308, "standard_deviation": 0.021113549799116862, "minimum": 0.7244602993073266, "maximum": 0.7813788004107801, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5766256331060968, "standard_deviation": 0.1297941118451501, "minimum": 0.35141509446400687, "maximum": 0.7467308004912975, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0835737556983254, "standard_deviation": 0.004195434338485655, "minimum": 0.07812715187406447, "maximum": 0.09155412340712664, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980741513258606, "standard_deviation": 0.00041730754902837307, "minimum": 0.9972505416659685, "maximum": 0.998724042375039, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40997124466634494, "standard_deviation": 0.017996018253636784, "minimum": 0.3810047893637573, "maximum": 0.4362417770366978, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5942872630576287, "standard_deviation": 0.1756125300648141, "minimum": 0.2974912255884244, "maximum": 0.8324379835463797, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04428814833912376, "standard_deviation": 0.004980787805389299, "minimum": 0.03534377751118292, "maximum": 0.05255427092533497, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0032", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.6427524089813232, "predicted_mos": 2.6427524089813232, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a66bfebbc5ec292cd90057f18ccf7ba60be9e9c3176b7c4189a451615d69a5c2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.896971399895847, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7679447827904554, "全帯域_10ms_rms_envelope_nrmse": 0.07532184431701482, "全帯域_10ms_envelope_correlation": 0.9946041930112467, "全帯域_10ms_modulation_mae_db": 0.5387644334841953, "全帯域_40ms_log_rms_mae_db": 0.4227975620286701, "全帯域_40ms_rms_envelope_nrmse": 0.03440031108809978, "全帯域_40ms_envelope_correlation": 0.9988279691901968, "全帯域_40ms_modulation_mae_db": 0.5450235066841679, "300–1000 Hz_10ms_log_rms_mae_db": 1.3718732510074594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09632311317966159, "300–1000 Hz_10ms_envelope_correlation": 0.9927903888828113, "300–1000 Hz_10ms_modulation_mae_db": 0.8536451286346239, "300–1000 Hz_40ms_log_rms_mae_db": 0.7458490872747943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045554182653945396, "300–1000 Hz_40ms_envelope_correlation": 0.9984491763275136, "300–1000 Hz_40ms_modulation_mae_db": 0.946319267853089, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3346365268596412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14889736113340563, "1000–3000 Hz_10ms_envelope_correlation": 0.9836838505343404, "1000–3000 Hz_10ms_modulation_mae_db": 0.8186751915849175, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8318901440177097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08410690813271364, "1000–3000 Hz_40ms_envelope_correlation": 0.9946234993432342, "1000–3000 Hz_40ms_modulation_mae_db": 0.9589303027659616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3935305552416228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.354679381759838, "3000–6000 Hz_10ms_envelope_correlation": 0.9107151150467991, "3000–6000 Hz_10ms_modulation_mae_db": 0.5886604293261548, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0003074604878535, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27033206324411524, "3000–6000 Hz_40ms_envelope_correlation": 0.9450078774762435, "3000–6000 Hz_40ms_modulation_mae_db": 0.54147371388665, "mr_stft_db": 5.5862624119605515, "band_mae_db": 0.3234216223692177, "rms_gain_error_db": -0.06145767431614394, "output_peak": 0.4968366049727698}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7767652028171526, "全帯域_10ms_rms_envelope_nrmse": 0.09483136501824703, "全帯域_10ms_envelope_correlation": 0.991519555771715, "全帯域_10ms_modulation_mae_db": 0.6280231812213449, "全帯域_40ms_log_rms_mae_db": 0.4200104594855955, "全帯域_40ms_rms_envelope_nrmse": 0.05623846071144614, "全帯域_40ms_envelope_correlation": 0.9968218676518537, "全帯域_40ms_modulation_mae_db": 0.6812263642350387, "300–1000 Hz_10ms_log_rms_mae_db": 1.387034316541329, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12371725292664594, "300–1000 Hz_10ms_envelope_correlation": 0.9880558038387067, "300–1000 Hz_10ms_modulation_mae_db": 0.7299855171378624, "300–1000 Hz_40ms_log_rms_mae_db": 0.8084965014902704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07430574044090182, "300–1000 Hz_40ms_envelope_correlation": 0.9954471523999078, "300–1000 Hz_40ms_modulation_mae_db": 0.7973690902275142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3467951490252357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19237972109358903, "1000–3000 Hz_10ms_envelope_correlation": 0.9721925260403094, "1000–3000 Hz_10ms_modulation_mae_db": 0.35270717655538214, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8405696305714415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08733596091315803, "1000–3000 Hz_40ms_envelope_correlation": 0.994355478219505, "1000–3000 Hz_40ms_modulation_mae_db": 0.2687335420228087, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4623587302431957, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3655167971234862, "3000–6000 Hz_10ms_envelope_correlation": 0.905989877623891, "3000–6000 Hz_10ms_modulation_mae_db": 0.5468080574775118, "3000–6000 Hz_40ms_log_rms_mae_db": 1.02410215767392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2639154606074256, "3000–6000 Hz_40ms_envelope_correlation": 0.9485552777818315, "3000–6000 Hz_40ms_modulation_mae_db": 0.6437665040689361, "mr_stft_db": 5.592415781640245, "band_mae_db": 0.2884758352357643, "rms_gain_error_db": 0.03743247845343332, "output_peak": 0.5107609988167245}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7629051654694031, "全帯域_10ms_rms_envelope_nrmse": 0.08272695588568425, "全帯域_10ms_envelope_correlation": 0.9934700410701562, "全帯域_10ms_modulation_mae_db": 0.5590510617822809, "全帯域_40ms_log_rms_mae_db": 0.43266641988437743, "全帯域_40ms_rms_envelope_nrmse": 0.0423642341101765, "全帯域_40ms_envelope_correlation": 0.9981744701897933, "全帯域_40ms_modulation_mae_db": 0.5046679215587383, "300–1000 Hz_10ms_log_rms_mae_db": 1.394897816794653, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09376808203345749, "300–1000 Hz_10ms_envelope_correlation": 0.9930925453693658, "300–1000 Hz_10ms_modulation_mae_db": 1.0888431477898444, "300–1000 Hz_40ms_log_rms_mae_db": 0.8202886681690957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04620392312473183, "300–1000 Hz_40ms_envelope_correlation": 0.9982693001444038, "300–1000 Hz_40ms_modulation_mae_db": 1.1292053242671447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.365423100808997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18381462687122177, "1000–3000 Hz_10ms_envelope_correlation": 0.9752927969391142, "1000–3000 Hz_10ms_modulation_mae_db": 0.6308499700640642, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8572093183467729, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08706057151488847, "1000–3000 Hz_40ms_envelope_correlation": 0.9942698309744469, "1000–3000 Hz_40ms_modulation_mae_db": 0.279476657701141, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4205217482468362, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36118432007292844, "3000–6000 Hz_10ms_envelope_correlation": 0.911003225679813, "3000–6000 Hz_10ms_modulation_mae_db": 0.6002340251356952, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0121914850745062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27494967251472985, "3000–6000 Hz_40ms_envelope_correlation": 0.9462656400055861, "3000–6000 Hz_40ms_modulation_mae_db": 0.3822618155696161, "mr_stft_db": 5.605704189660801, "band_mae_db": 0.2634428617413234, "rms_gain_error_db": -0.042274558777794034, "output_peak": 0.451373266684083}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7404883733103591, "全帯域_10ms_rms_envelope_nrmse": 0.08242032890910057, "全帯域_10ms_envelope_correlation": 0.9935299443501993, "全帯域_10ms_modulation_mae_db": 0.586220420333183, "全帯域_40ms_log_rms_mae_db": 0.39079593224222736, "全帯域_40ms_rms_envelope_nrmse": 0.04114597912031314, "全帯域_40ms_envelope_correlation": 0.9983034834473832, "全帯域_40ms_modulation_mae_db": 0.6215696223738946, "300–1000 Hz_10ms_log_rms_mae_db": 1.4183305521911638, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10043478215606683, "300–1000 Hz_10ms_envelope_correlation": 0.9921254036915474, "300–1000 Hz_10ms_modulation_mae_db": 0.8052047608887682, "300–1000 Hz_40ms_log_rms_mae_db": 0.7593478043353581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050010398133720005, "300–1000 Hz_40ms_envelope_correlation": 0.9980634540097796, "300–1000 Hz_40ms_modulation_mae_db": 0.936829189945113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3659180471473544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2007046089403087, "1000–3000 Hz_10ms_envelope_correlation": 0.9760268986218783, "1000–3000 Hz_10ms_modulation_mae_db": 0.37482216098762766, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8783214196947788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.153425326336336, "1000–3000 Hz_40ms_envelope_correlation": 0.9870616845570196, "1000–3000 Hz_40ms_modulation_mae_db": 0.5266724597383426, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4357658819097736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3533748213293486, "3000–6000 Hz_10ms_envelope_correlation": 0.908354552994577, "3000–6000 Hz_10ms_modulation_mae_db": 0.6578474662588859, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0351435362850128, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27141151586614404, "3000–6000 Hz_40ms_envelope_correlation": 0.941842473458833, "3000–6000 Hz_40ms_modulation_mae_db": 0.7689056669672368, "mr_stft_db": 5.590310446571839, "band_mae_db": 0.3369125949001121, "rms_gain_error_db": -0.06439016266353964, "output_peak": 0.44691150751250425}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7540007990165081, "全帯域_10ms_rms_envelope_nrmse": 0.09093569805148982, "全帯域_10ms_envelope_correlation": 0.9921469917692426, "全帯域_10ms_modulation_mae_db": 0.6565050245278753, "全帯域_40ms_log_rms_mae_db": 0.43329644867973305, "全帯域_40ms_rms_envelope_nrmse": 0.0487970650059775, "全帯域_40ms_envelope_correlation": 0.9976376808412859, "全帯域_40ms_modulation_mae_db": 0.7569109392076321, "300–1000 Hz_10ms_log_rms_mae_db": 1.3548335099839128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1059532941328708, "300–1000 Hz_10ms_envelope_correlation": 0.9912008083944857, "300–1000 Hz_10ms_modulation_mae_db": 0.8235231583005623, "300–1000 Hz_40ms_log_rms_mae_db": 0.7600911024951718, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059649625913755736, "300–1000 Hz_40ms_envelope_correlation": 0.9971244381018912, "300–1000 Hz_40ms_modulation_mae_db": 0.9168043285518955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3805066508398698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18312472731923649, "1000–3000 Hz_10ms_envelope_correlation": 0.97627057588129, "1000–3000 Hz_10ms_modulation_mae_db": 0.6586306697976559, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8945680801377011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1014548545798067, "1000–3000 Hz_40ms_envelope_correlation": 0.9927347589591831, "1000–3000 Hz_40ms_modulation_mae_db": 0.6323287716890644, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4626472754636375, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34402412329240467, "3000–6000 Hz_10ms_envelope_correlation": 0.9182039922915661, "3000–6000 Hz_10ms_modulation_mae_db": 0.29522868146080977, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0249945146344883, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26062252003983555, "3000–6000 Hz_40ms_envelope_correlation": 0.9507097021121288, "3000–6000 Hz_40ms_modulation_mae_db": 0.1628024586403599, "mr_stft_db": 5.586552494870425, "band_mae_db": 0.28085881853900047, "rms_gain_error_db": -0.08547876728979731, "output_peak": 0.4827059533111638}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7368117214328581, "全帯域_10ms_rms_envelope_nrmse": 0.07979023744153343, "全帯域_10ms_envelope_correlation": 0.9940232704701192, "全帯域_10ms_modulation_mae_db": 0.6165027431555034, "全帯域_40ms_log_rms_mae_db": 0.38362165084585803, "全帯域_40ms_rms_envelope_nrmse": 0.044183866873017966, "全帯域_40ms_envelope_correlation": 0.9981501006828347, "全帯域_40ms_modulation_mae_db": 0.6941464912765705, "300–1000 Hz_10ms_log_rms_mae_db": 1.4273454382711894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09530271207100392, "300–1000 Hz_10ms_envelope_correlation": 0.9931366018790934, "300–1000 Hz_10ms_modulation_mae_db": 0.8147675801737599, "300–1000 Hz_40ms_log_rms_mae_db": 0.795697133320861, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05347900656315808, "300–1000 Hz_40ms_envelope_correlation": 0.9980213656184515, "300–1000 Hz_40ms_modulation_mae_db": 0.9371268004673423, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3323198464251498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15672928864738014, "1000–3000 Hz_10ms_envelope_correlation": 0.9831485768667467, "1000–3000 Hz_10ms_modulation_mae_db": 0.4365873373340081, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8146201131586774, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10075280692236192, "1000–3000 Hz_40ms_envelope_correlation": 0.9932410733999709, "1000–3000 Hz_40ms_modulation_mae_db": 0.5420874758682832, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4125754101292933, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3358216352601981, "3000–6000 Hz_10ms_envelope_correlation": 0.9159258900370214, "3000–6000 Hz_10ms_modulation_mae_db": 0.2841584218817979, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9958379501381799, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2494227398306346, "3000–6000 Hz_40ms_envelope_correlation": 0.9499412622628394, "3000–6000 Hz_40ms_modulation_mae_db": 0.32322994734222493, "mr_stft_db": 5.593791224094113, "band_mae_db": 0.27062381621801807, "rms_gain_error_db": -0.11505650147208565, "output_peak": 0.40558665666876026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7815864832327521, "全帯域_10ms_rms_envelope_nrmse": 0.0815389886169868, "全帯域_10ms_envelope_correlation": 0.9939693741134157, "全帯域_10ms_modulation_mae_db": 0.3860645326982522, "全帯域_40ms_log_rms_mae_db": 0.4464810863604783, "全帯域_40ms_rms_envelope_nrmse": 0.04745851586199889, "全帯域_40ms_envelope_correlation": 0.9981138800071888, "全帯域_40ms_modulation_mae_db": 0.3347576801626426, "300–1000 Hz_10ms_log_rms_mae_db": 1.4440286584820305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.116126256967094, "300–1000 Hz_10ms_envelope_correlation": 0.9900867646213768, "300–1000 Hz_10ms_modulation_mae_db": 0.5217943839186413, "300–1000 Hz_40ms_log_rms_mae_db": 0.8588788502048108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07384050038265293, "300–1000 Hz_40ms_envelope_correlation": 0.99643425287245, "300–1000 Hz_40ms_modulation_mae_db": 0.5259737913429731, "1000–3000 Hz_10ms_log_rms_mae_db": 1.345790997962739, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22503522624771619, "1000–3000 Hz_10ms_envelope_correlation": 0.9692634478899472, "1000–3000 Hz_10ms_modulation_mae_db": 0.4255114868390599, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8752199529293565, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16029964092636584, "1000–3000 Hz_40ms_envelope_correlation": 0.9858340859581449, "1000–3000 Hz_40ms_modulation_mae_db": 0.63193096196012, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4336080953940986, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3464180239276107, "3000–6000 Hz_10ms_envelope_correlation": 0.9198882762714734, "3000–6000 Hz_10ms_modulation_mae_db": 0.5077153185261966, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0116337193639737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.25367994593511056, "3000–6000 Hz_40ms_envelope_correlation": 0.9562294799893184, "3000–6000 Hz_40ms_modulation_mae_db": 0.3528453892365728, "mr_stft_db": 5.593628932113394, "band_mae_db": 0.4644122020126016, "rms_gain_error_db": -0.18678671696018762, "output_peak": 0.430380764636663}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7560804878616646, "全帯域_10ms_rms_envelope_nrmse": 0.08245339290895703, "全帯域_10ms_envelope_correlation": 0.9936041706867876, "全帯域_10ms_modulation_mae_db": 0.7488383028985647, "全帯域_40ms_log_rms_mae_db": 0.39759571990811504, "全帯域_40ms_rms_envelope_nrmse": 0.04191139803945209, "全帯域_40ms_envelope_correlation": 0.9983525595668058, "全帯域_40ms_modulation_mae_db": 0.7871122803066758, "300–1000 Hz_10ms_log_rms_mae_db": 1.4155690433954213, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1063156991600854, "300–1000 Hz_10ms_envelope_correlation": 0.9912768546781391, "300–1000 Hz_10ms_modulation_mae_db": 0.6448738320034454, "300–1000 Hz_40ms_log_rms_mae_db": 0.805002939217928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05698812526468902, "300–1000 Hz_40ms_envelope_correlation": 0.9976021591302818, "300–1000 Hz_40ms_modulation_mae_db": 0.7282558685602215, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3953460060558733, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18923675385545627, "1000–3000 Hz_10ms_envelope_correlation": 0.9741617319423124, "1000–3000 Hz_10ms_modulation_mae_db": 0.4101794656108179, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8185530301336714, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10277811503648936, "1000–3000 Hz_40ms_envelope_correlation": 0.9941303834977415, "1000–3000 Hz_40ms_modulation_mae_db": 0.5722330503488433, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4123796651544236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3374892112374631, "3000–6000 Hz_10ms_envelope_correlation": 0.921236201081176, "3000–6000 Hz_10ms_modulation_mae_db": 0.3347715765141524, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9950631080144179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2601928787341116, "3000–6000 Hz_40ms_envelope_correlation": 0.9509023941632537, "3000–6000 Hz_40ms_modulation_mae_db": 0.2001053589337942, "mr_stft_db": 5.5961703256753, "band_mae_db": 0.35962211465466093, "rms_gain_error_db": -0.11509639780408032, "output_peak": 0.38770352126555907}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.348385261079312, "spectral_convergence": 0.25712972442159526, "spectral_convergence_status": "ok", "amplitude_floor": 1.3235749190080935e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4479968279981765, "spectral_convergence": 0.2940148671136831, "spectral_convergence_status": "ok", "amplitude_floor": 1.1987597823804542e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.962405146804166, "spectral_convergence": 0.29400888255691043, "spectral_convergence_status": "ok", "amplitude_floor": 9.377214849682438e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.350500895926799e-05, "output_power": 5.184909387826043e-05, "delta_db": 0.7620190857358788, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012608807022653607, "output_power": 0.0012373688667371334, "delta_db": -0.08174813314770824, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00036191686574782775, "output_power": 0.0003461666770320875, "delta_db": -0.19323563053170478, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.9809147109830135e-05, "output_power": 3.289829710678993e-05, "delta_db": 0.42823868118309544, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.6488056661262093e-05, "output_power": 1.797842552281568e-05, "delta_db": 0.37582184094861726, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.109371525887873e-05, "output_power": 6.250908944950784e-05, "delta_db": 0.09946636266830157, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3234216223692177, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06145767431614394, "pitch": null, "output_rms_dbfs": -27.572671939288572, "output_peak": 0.4968366049727698, "reference_rms_dbfs": -27.511214264972427, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5862624119605515}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018136510927426692, "log_rms_mae_db": 0.7679447827904554, "rms_envelope_nrmse": 0.07532184431701482, "envelope_correlation": 0.9946041930112467, "rms_gain_error_db": -0.061420172553721106, "reference_mean_power": 0.0017749427149271913, "output_mean_power": 0.0017500172267655451}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.578419076835135e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003061888162786739, "output_power": 0.00030280716226587114, "reference_active": true, "delta_db": -0.04823180526936654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.561280729204581e-05, "output_power": 5.258682705262521e-05, "reference_active": true, "delta_db": -0.242978510383076}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1447257663776473e-05, "output_power": 8.437099287247903e-06, "reference_active": true, "delta_db": -1.3250829848001433}], "active_band_count": 3, "active_band_mae_db": 0.5387644334841953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015531555176710683, "log_rms_mae_db": 0.4227975620286701, "rms_envelope_nrmse": 0.03440031108809978, "envelope_correlation": 0.9988279691901968, "rms_gain_error_db": -0.06172571219791349, "reference_mean_power": 0.0017815693662331493, "output_mean_power": 0.0017564272456743123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.067921014076438e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002861407755621019, "output_power": 0.00028225660869015583, "reference_active": true, "delta_db": -0.059356307548210585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5605898871779113e-05, "output_power": 2.4221499462408333e-05, "reference_active": true, "delta_db": -0.24139000764854532}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.159270454376389e-07, "output_power": 6.736409234691448e-07, "reference_active": true, "delta_db": -1.3343242048557478}], "active_band_count": 3, "active_band_mae_db": 0.5450235066841679}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017559797263300874, "log_rms_mae_db": 1.3718732510074594, "rms_envelope_nrmse": 0.09632311317966159, "envelope_correlation": 0.9927903888828113, "rms_gain_error_db": -0.1356518703837378, "reference_mean_power": 0.0012870696988326355, "output_mean_power": 0.0012474894425681228}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.703155295091611e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025376682215949657, "output_power": 0.00024848189400694583, "reference_active": true, "delta_db": -0.0914009245701864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.494529209431561e-05, "output_power": 4.0233783324878234e-05, "reference_active": true, "delta_db": -0.4809333446604839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.800066525463509e-06, "output_power": 5.567059251286255e-06, "reference_active": true, "delta_db": -1.9886011166732014}], "active_band_count": 3, "active_band_mae_db": 0.8536451286346239}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014992083957937532, "log_rms_mae_db": 0.7458490872747943, "rms_envelope_nrmse": 0.045554182653945396, "envelope_correlation": 0.9984491763275136, "rms_gain_error_db": -0.13612408660983444, "reference_mean_power": 0.0012919551058812144, "output_mean_power": 0.0012520884632431233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4067631484646013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002445691472192921, "output_power": 0.00023783347995440297, "reference_active": true, "delta_db": -0.12128678759417062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8830927140807284e-05, "output_power": 1.6886098992500968e-05, "reference_active": true, "delta_db": -0.47342372096323554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.050384653726169e-07, "output_power": 4.205213813188477e-07, "reference_active": true, "delta_db": -2.244247295001861}], "active_band_count": 3, "active_band_mae_db": 0.946319267853089}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.2493337902299445e-05, "log_rms_mae_db": 1.3346365268596412, "rms_envelope_nrmse": 0.14889736113340563, "envelope_correlation": 0.9836838505343404, "rms_gain_error_db": 0.015094354914415358, "reference_mean_power": 4.199153922743893e-05, "output_mean_power": 4.213773909415057e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0299126624198213e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.222152478948813e-06, "output_power": 3.729381613871815e-06, "reference_active": true, "delta_db": -0.5389708799117089}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.960031981806193e-07, "output_power": 8.090104686851821e-07, "reference_active": true, "delta_db": -0.4435541836182806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.633269352164066e-07, "output_power": 1.8756216341641755e-07, "reference_active": true, "delta_db": -1.4735005112247632}], "active_band_count": 3, "active_band_mae_db": 0.8186751915849175}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.820326553727612e-05, "log_rms_mae_db": 0.8318901440177097, "rms_envelope_nrmse": 0.08410690813271364, "envelope_correlation": 0.9946234993432342, "rms_gain_error_db": 0.014882830355575239, "reference_mean_power": 4.2190507757870256e-05, "output_mean_power": 4.2335338356657065e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.548028980657037e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.958931660389438e-06, "output_power": 3.474872792816282e-06, "reference_active": true, "delta_db": -0.5663909428252247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7410748918107156e-07, "output_power": 3.2426628981093554e-07, "reference_active": true, "delta_db": -0.6209459985634153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8002833142836267e-08, "output_power": 1.2201000912526684e-08, "reference_active": true, "delta_db": -1.6894539669092443}], "active_band_count": 3, "active_band_mae_db": 0.9589303027659616}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.219318962627022e-05, "log_rms_mae_db": 1.3935305552416228, "rms_envelope_nrmse": 0.354679381759838, "envelope_correlation": 0.9107151150467991, "rms_gain_error_db": 0.548256939624815, "reference_mean_power": 2.2095607497389627e-05, "output_mean_power": 2.5068690037872217e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.718240736767725e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.504526854650138e-06, "output_power": 1.0028190939841979e-05, "reference_active": true, "delta_db": 0.23292092362319983}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5665611692910136e-06, "output_power": 2.8039449786805214e-06, "reference_active": true, "delta_db": 0.38417867959206925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.91578688150447e-07, "output_power": 7.707272041957836e-07, "reference_active": true, "delta_db": 1.1488816847631953}], "active_band_count": 3, "active_band_mae_db": 0.5886604293261548}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.1857789632782756e-05, "log_rms_mae_db": 1.0003074604878535, "rms_envelope_nrmse": 0.27033206324411524, "envelope_correlation": 0.9450078774762435, "rms_gain_error_db": 0.548331513618084, "reference_mean_power": 2.220381322589703e-05, "output_mean_power": 2.5191888001381893e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6122764663239433e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.855212413255228e-06, "output_power": 1.0533164062700443e-05, "reference_active": true, "delta_db": 0.2889285949623222}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3142139667203495e-06, "output_power": 1.4971146364177261e-06, "reference_active": true, "delta_db": 0.5658897791213048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2621117296192714e-08, "output_power": 5.088445263301353e-08, "reference_active": true, "delta_db": 0.7696027675763227}], "active_band_count": 3, "active_band_mae_db": 0.54147371388665}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8720138207242304, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ec5537a696a1ded6820a5e60ab86b1678b28a46b79789d27f5203c6e0323ea28", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9762550505894924, "standard_deviation": 0.004977873851614056, "minimum": 0.9692634478899472, "maximum": 0.9836838505343404, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3583420406406075, "standard_deviation": 0.022354501122731615, "minimum": 1.3323198464251498, "maximum": 1.3953460060558733, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5134954323466917, "standard_deviation": 0.16792190712159782, "minimum": 0.35270717655538214, "maximum": 0.8186751915849175, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18499028926353928, "standard_deviation": 0.023992317536882034, "minimum": 0.14889736113340563, "maximum": 0.22503522624771619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9920313493636557, "standard_deviation": 0.0035172923836351996, "minimum": 0.9858340859581449, "maximum": 0.9946234993432342, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8513689611237637, "standard_deviation": 0.029565886469139612, "minimum": 0.8146201131586774, "maximum": 0.8945680801377011, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5515491527618206, "standard_deviation": 0.2184655227829375, "minimum": 0.2687335420228087, "maximum": 0.9589303027659616, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.109651773045265, "standard_deviation": 0.03008525199813098, "minimum": 0.08410690813271364, "maximum": 0.16029964092636584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9139146413782896, "standard_deviation": 0.005658911761327453, "minimum": 0.905989877623891, "maximum": 0.921236201081176, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.42917342022286, "standard_deviation": 0.024458945431098965, "minimum": 1.3935305552416228, "maximum": 1.4626472754636375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.47692799707265054, "standard_deviation": 0.1495873598690488, "minimum": 0.2841584218817979, "maximum": 0.6578474662588859, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3498135392504097, "standard_deviation": 0.010723508337150783, "minimum": 0.3358216352601981, "maximum": 0.3655167971234862, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9486817634062543, "standard_deviation": 0.0043707644812482245, "minimum": 0.941842473458833, "maximum": 0.9562294799893184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0124092414590442, "standard_deviation": 0.01480617413832139, "minimum": 0.9950631080144179, "maximum": 1.0351435362850128, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.42192385683067385, "standard_deviation": 0.2125812281737497, "minimum": 0.1628024586403599, "maximum": 0.7689056669672368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.26306584959651336, "standard_deviation": 0.00888697718817286, "minimum": 0.2494227398306346, "maximum": 0.27494967251472985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9914706464194408, "standard_deviation": 0.0017428645358895146, "minimum": 0.9880558038387067, "maximum": 0.9931366018790934, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.401739073333395, "standard_deviation": 0.02991839391137849, "minimum": 1.3548335099839128, "maximum": 1.4440286584820305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7853296886059385, "standard_deviation": 0.1655076390656803, "minimum": 0.5217943839186413, "maximum": 1.0888431477898444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10474264907836076, "standard_deviation": 0.010640637081269631, "minimum": 0.09376808203345749, "maximum": 0.12371725292664594, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974264123255849, "standard_deviation": 0.0010378702927369976, "minimum": 0.9954471523999078, "maximum": 0.9984491763275136, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7942065108135363, "standard_deviation": 0.03761308659670349, "minimum": 0.7458490872747943, "maximum": 0.8588788502048108, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8647354576519116, "standard_deviation": 0.1803464050736048, "minimum": 0.5259737913429731, "maximum": 1.1292053242671447, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05750393780969435, "standard_deviation": 0.011319351804745711, "minimum": 0.045554182653945396, "maximum": 0.07430574044090182, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3234712332088373, "standard_deviation": 0.06632029016063526, "minimum": 0.2634428617413234, "maximum": 0.4644122020126016, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.593104475823333, "standard_deviation": 0.006175089277945367, "minimum": 5.5862624119605515, "maximum": 5.605704189660801, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45153240923352844, "standard_deviation": 0.04340209848801497, "minimum": 0.38770352126555907, "maximum": 0.5107609988167245, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0791385376037744, "standard_deviation": 0.06512546032666248, "minimum": -0.18678671696018762, "maximum": 0.03743247845343332, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933584426553603, "standard_deviation": 0.0010224965764647261, "minimum": 0.991519555771715, "maximum": 0.9946041930112467, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7595728769913941, "standard_deviation": 0.015977221277094115, "minimum": 0.7368117214328581, "maximum": 0.7815864832327521, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.58999621251265, "standard_deviation": 0.10487241394351964, "minimum": 0.3860645326982522, "maximum": 0.7488383028985647, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08375235139362672, "standard_deviation": 0.006216668795753116, "minimum": 0.07532184431701482, "maximum": 0.09483136501824703, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980477514471677, "standard_deviation": 0.0005938856967843781, "minimum": 0.9968218676518537, "maximum": 0.9988279691901968, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41590815992938185, "standard_deviation": 0.02264816777081559, "minimum": 0.38362165084585803, "maximum": 0.4464810863604783, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.61567685072567, "standard_deviation": 0.14931107508487593, "minimum": 0.3347576801626426, "maximum": 0.7871122803066758, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04456247885131025, "standard_deviation": 0.0064319879527402375, "minimum": 0.03440031108809978, "maximum": 0.05623846071144614, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0032", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.51175856590271, "predicted_mos": 2.51175856590271, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3160e4f3d11bf6562193677c03101e70df92cbb70b1eec8565500e03011a0325", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.936247100122273, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7586985152539231, "全帯域_10ms_rms_envelope_nrmse": 0.09924625928408566, "全帯域_10ms_envelope_correlation": 0.9908485281866493, "全帯域_10ms_modulation_mae_db": 0.5134865692254015, "全帯域_40ms_log_rms_mae_db": 0.43019102947694654, "全帯域_40ms_rms_envelope_nrmse": 0.05572587524306625, "全帯域_40ms_envelope_correlation": 0.9972454559071846, "全帯域_40ms_modulation_mae_db": 0.5268940937568324, "300–1000 Hz_10ms_log_rms_mae_db": 1.5836005488868454, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11925888700093591, "300–1000 Hz_10ms_envelope_correlation": 0.9892376161181324, "300–1000 Hz_10ms_modulation_mae_db": 0.4645598739735757, "300–1000 Hz_40ms_log_rms_mae_db": 0.9911559198370041, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07195242127888407, "300–1000 Hz_40ms_envelope_correlation": 0.996407741496344, "300–1000 Hz_40ms_modulation_mae_db": 0.455431777118594, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5959224286332752, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20110339741681427, "1000–3000 Hz_10ms_envelope_correlation": 0.9696759481438857, "1000–3000 Hz_10ms_modulation_mae_db": 0.13378756839270103, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0058186471469517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08867648325175562, "1000–3000 Hz_40ms_envelope_correlation": 0.9940751804331581, "1000–3000 Hz_40ms_modulation_mae_db": 0.20078527030882543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7245977420731369, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39329052511073326, "3000–6000 Hz_10ms_envelope_correlation": 0.8970400944943506, "3000–6000 Hz_10ms_modulation_mae_db": 0.7436941373088679, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1832696041078437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3004468934135877, "3000–6000 Hz_40ms_envelope_correlation": 0.9381991067105436, "3000–6000 Hz_40ms_modulation_mae_db": 0.8284375239490541, "mr_stft_db": 5.897462280285285, "band_mae_db": 0.30726817539170215, "rms_gain_error_db": -0.18119635745537926, "output_peak": 0.35427319937601226}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7415088895091587, "全帯域_10ms_rms_envelope_nrmse": 0.07546397743717204, "全帯域_10ms_envelope_correlation": 0.9949792251046491, "全帯域_10ms_modulation_mae_db": 0.48427218878595646, "全帯域_40ms_log_rms_mae_db": 0.41151236937402347, "全帯域_40ms_rms_envelope_nrmse": 0.04652823953664527, "全帯域_40ms_envelope_correlation": 0.9982886281185078, "全帯域_40ms_modulation_mae_db": 0.5098462208493905, "300–1000 Hz_10ms_log_rms_mae_db": 1.5330018330225232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09250130270396674, "300–1000 Hz_10ms_envelope_correlation": 0.99401294045184, "300–1000 Hz_10ms_modulation_mae_db": 0.8663899959120743, "300–1000 Hz_40ms_log_rms_mae_db": 0.911591697207409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06158791742172439, "300–1000 Hz_40ms_envelope_correlation": 0.9977444182759582, "300–1000 Hz_40ms_modulation_mae_db": 0.9755592102500904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.647247661746413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21762554348728855, "1000–3000 Hz_10ms_envelope_correlation": 0.9668285163971994, "1000–3000 Hz_10ms_modulation_mae_db": 0.6548864293578781, "1000–3000 Hz_40ms_log_rms_mae_db": 1.11089132906687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12726457547730657, "1000–3000 Hz_40ms_envelope_correlation": 0.9887037564293117, "1000–3000 Hz_40ms_modulation_mae_db": 0.9371995893513887, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7224689313661619, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3983485954135244, "3000–6000 Hz_10ms_envelope_correlation": 0.9037626017203807, "3000–6000 Hz_10ms_modulation_mae_db": 1.2519932336730644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1799121332298523, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30642099000126694, "3000–6000 Hz_40ms_envelope_correlation": 0.9439355255768807, "3000–6000 Hz_40ms_modulation_mae_db": 1.2849891243817886, "mr_stft_db": 5.879171088466246, "band_mae_db": 0.45205672665075114, "rms_gain_error_db": -0.19464346034018504, "output_peak": 0.3889682221092622}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7596969142271024, "全帯域_10ms_rms_envelope_nrmse": 0.09218200511061378, "全帯域_10ms_envelope_correlation": 0.992395795277786, "全帯域_10ms_modulation_mae_db": 0.6320855006670925, "全帯域_40ms_log_rms_mae_db": 0.41695510880834324, "全帯域_40ms_rms_envelope_nrmse": 0.05928535136731324, "全帯域_40ms_envelope_correlation": 0.9968688857021042, "全帯域_40ms_modulation_mae_db": 0.6407854943170862, "300–1000 Hz_10ms_log_rms_mae_db": 1.5355245884520048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11332205948268735, "300–1000 Hz_10ms_envelope_correlation": 0.9900486203315925, "300–1000 Hz_10ms_modulation_mae_db": 0.7261271491694471, "300–1000 Hz_40ms_log_rms_mae_db": 0.9369073569349667, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0665948982936056, "300–1000 Hz_40ms_envelope_correlation": 0.9963892786277919, "300–1000 Hz_40ms_modulation_mae_db": 0.8139662491908436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6312870034910905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22719107021668736, "1000–3000 Hz_10ms_envelope_correlation": 0.9680204986874773, "1000–3000 Hz_10ms_modulation_mae_db": 0.30030430210916487, "1000–3000 Hz_40ms_log_rms_mae_db": 1.106103274099166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15819921203516474, "1000–3000 Hz_40ms_envelope_correlation": 0.9857178529123908, "1000–3000 Hz_40ms_modulation_mae_db": 0.2580335697459067, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6955995835981672, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3835301169324425, "3000–6000 Hz_10ms_envelope_correlation": 0.9002701880762979, "3000–6000 Hz_10ms_modulation_mae_db": 1.3009198520746608, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1815445130855087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2983175400119793, "3000–6000 Hz_40ms_envelope_correlation": 0.9373854346701675, "3000–6000 Hz_40ms_modulation_mae_db": 1.4557777698804815, "mr_stft_db": 5.896372354237502, "band_mae_db": 0.4431189815609668, "rms_gain_error_db": 0.13756000735556706, "output_peak": 0.5199280976944596}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7267854027464348, "全帯域_10ms_rms_envelope_nrmse": 0.07310363338488933, "全帯域_10ms_envelope_correlation": 0.9949431306188943, "全帯域_10ms_modulation_mae_db": 0.3999303196476349, "全帯域_40ms_log_rms_mae_db": 0.3842847151027591, "全帯域_40ms_rms_envelope_nrmse": 0.038209632579272, "全帯域_40ms_envelope_correlation": 0.99857114458172, "全帯域_40ms_modulation_mae_db": 0.40172839945348643, "300–1000 Hz_10ms_log_rms_mae_db": 1.5943545715359415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.092420327847317, "300–1000 Hz_10ms_envelope_correlation": 0.9935055847255578, "300–1000 Hz_10ms_modulation_mae_db": 0.8018459434335795, "300–1000 Hz_40ms_log_rms_mae_db": 0.9343462659748754, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04796641898195895, "300–1000 Hz_40ms_envelope_correlation": 0.998418903817574, "300–1000 Hz_40ms_modulation_mae_db": 1.0083612106787778, "1000–3000 Hz_10ms_log_rms_mae_db": 1.679501930425583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2753539440818146, "1000–3000 Hz_10ms_envelope_correlation": 0.9552421146376044, "1000–3000 Hz_10ms_modulation_mae_db": 0.4322121667852666, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0867557214751937, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19015497818198193, "1000–3000 Hz_40ms_envelope_correlation": 0.9810335460965568, "1000–3000 Hz_40ms_modulation_mae_db": 0.5768537191629152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7254634396036443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.38419248510279147, "3000–6000 Hz_10ms_envelope_correlation": 0.903309502232667, "3000–6000 Hz_10ms_modulation_mae_db": 0.9665294539811956, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1994849912488907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29478254995092185, "3000–6000 Hz_40ms_envelope_correlation": 0.941760497389703, "3000–6000 Hz_40ms_modulation_mae_db": 1.0320393541751394, "mr_stft_db": 5.907259468808571, "band_mae_db": 0.5305696520090498, "rms_gain_error_db": -0.07735100519100756, "output_peak": 0.48632192142238395}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.709422913932188, "全帯域_10ms_rms_envelope_nrmse": 0.07724025189337162, "全帯域_10ms_envelope_correlation": 0.9943167980065084, "全帯域_10ms_modulation_mae_db": 0.735451592431365, "全帯域_40ms_log_rms_mae_db": 0.3867239687996714, "全帯域_40ms_rms_envelope_nrmse": 0.04579501159377247, "全帯域_40ms_envelope_correlation": 0.9978678464845493, "全帯域_40ms_modulation_mae_db": 0.7841465116951071, "300–1000 Hz_10ms_log_rms_mae_db": 1.5068593952607556, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09017382492025341, "300–1000 Hz_10ms_envelope_correlation": 0.9936708875092441, "300–1000 Hz_10ms_modulation_mae_db": 1.111608252972843, "300–1000 Hz_40ms_log_rms_mae_db": 0.936821167313524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052512817412689504, "300–1000 Hz_40ms_envelope_correlation": 0.9978035399342856, "300–1000 Hz_40ms_modulation_mae_db": 1.2852826550148864, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6522322110665677, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21569924808683508, "1000–3000 Hz_10ms_envelope_correlation": 0.9651247720489742, "1000–3000 Hz_10ms_modulation_mae_db": 0.4173861748031336, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0452837139052913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11101917734307457, "1000–3000 Hz_40ms_envelope_correlation": 0.9906463050227425, "1000–3000 Hz_40ms_modulation_mae_db": 0.1494892880883409, "3000–6000 Hz_10ms_log_rms_mae_db": 1.756948614068809, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40304430714160683, "3000–6000 Hz_10ms_envelope_correlation": 0.8890266454629777, "3000–6000 Hz_10ms_modulation_mae_db": 0.2690587808618502, "3000–6000 Hz_40ms_log_rms_mae_db": 1.24226911284314, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31235398817886423, "3000–6000 Hz_40ms_envelope_correlation": 0.930448943248311, "3000–6000 Hz_40ms_modulation_mae_db": 0.46084079369680286, "mr_stft_db": 5.881499576729884, "band_mae_db": 0.33159143113233913, "rms_gain_error_db": -0.05133051166794641, "output_peak": 0.35183505327901077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7347741456064182, "全帯域_10ms_rms_envelope_nrmse": 0.08836330817511301, "全帯域_10ms_envelope_correlation": 0.9925441474036585, "全帯域_10ms_modulation_mae_db": 0.5803365263750918, "全帯域_40ms_log_rms_mae_db": 0.396129194454149, "全帯域_40ms_rms_envelope_nrmse": 0.03845214856023321, "全帯域_40ms_envelope_correlation": 0.9984902362666608, "全帯域_40ms_modulation_mae_db": 0.5136724528633155, "300–1000 Hz_10ms_log_rms_mae_db": 1.507439089383208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10742017542206796, "300–1000 Hz_10ms_envelope_correlation": 0.9909572489524792, "300–1000 Hz_10ms_modulation_mae_db": 0.5770228570888082, "300–1000 Hz_40ms_log_rms_mae_db": 0.9213832976418443, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0494665847645489, "300–1000 Hz_40ms_envelope_correlation": 0.9980960768834712, "300–1000 Hz_40ms_modulation_mae_db": 0.5473352005507613, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6324741118832014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17376804358982867, "1000–3000 Hz_10ms_envelope_correlation": 0.9781922530467124, "1000–3000 Hz_10ms_modulation_mae_db": 0.5096064182239295, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0248674420635202, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09537568023356834, "1000–3000 Hz_40ms_envelope_correlation": 0.9933321583800389, "1000–3000 Hz_40ms_modulation_mae_db": 0.5779684171214344, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7313512049633066, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.38678489790057813, "3000–6000 Hz_10ms_envelope_correlation": 0.8952088181170954, "3000–6000 Hz_10ms_modulation_mae_db": 0.30096546344535163, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2256642062348848, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29871309021518755, "3000–6000 Hz_40ms_envelope_correlation": 0.9342551824510547, "3000–6000 Hz_40ms_modulation_mae_db": 0.43281304782241503, "mr_stft_db": 5.883483608251308, "band_mae_db": 0.3545931489689404, "rms_gain_error_db": -0.03241810955566078, "output_peak": 0.5139717382050439}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7412908754322575, "全帯域_10ms_rms_envelope_nrmse": 0.0767185445295619, "全帯域_10ms_envelope_correlation": 0.9944062752963129, "全帯域_10ms_modulation_mae_db": 0.5457070353089367, "全帯域_40ms_log_rms_mae_db": 0.38919433499682815, "全帯域_40ms_rms_envelope_nrmse": 0.043405267137373915, "全帯域_40ms_envelope_correlation": 0.9980900245893588, "全帯域_40ms_modulation_mae_db": 0.6080661088369747, "300–1000 Hz_10ms_log_rms_mae_db": 1.491432067672651, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08952902285419576, "300–1000 Hz_10ms_envelope_correlation": 0.9937142419971766, "300–1000 Hz_10ms_modulation_mae_db": 0.9227976844563462, "300–1000 Hz_40ms_log_rms_mae_db": 0.9093069292750531, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049410874380821965, "300–1000 Hz_40ms_envelope_correlation": 0.9980125404587613, "300–1000 Hz_40ms_modulation_mae_db": 1.064473286603767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6837665276635752, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26496597256653404, "1000–3000 Hz_10ms_envelope_correlation": 0.9500950755017643, "1000–3000 Hz_10ms_modulation_mae_db": 0.35556474294319856, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0723277815399788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11932933652455946, "1000–3000 Hz_40ms_envelope_correlation": 0.9899867118038319, "1000–3000 Hz_40ms_modulation_mae_db": 0.18627568670243722, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7526648344486386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.38931499739014835, "3000–6000 Hz_10ms_envelope_correlation": 0.9026452202165091, "3000–6000 Hz_10ms_modulation_mae_db": 0.4161158973933467, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2070053011222972, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2921195282665036, "3000–6000 Hz_40ms_envelope_correlation": 0.9449729009407941, "3000–6000 Hz_40ms_modulation_mae_db": 0.6156958274454587, "mr_stft_db": 5.890947861784674, "band_mae_db": 0.3671501535889022, "rms_gain_error_db": -0.01648837123130193, "output_peak": 0.4553889084788515}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7634533889403365, "全帯域_10ms_rms_envelope_nrmse": 0.1091318207496345, "全帯域_10ms_envelope_correlation": 0.9888720346167416, "全帯域_10ms_modulation_mae_db": 0.28363767681390767, "全帯域_40ms_log_rms_mae_db": 0.4187558520990276, "全帯域_40ms_rms_envelope_nrmse": 0.05780854969346555, "全帯域_40ms_envelope_correlation": 0.9966922841134443, "全帯域_40ms_modulation_mae_db": 0.23182985743355022, "300–1000 Hz_10ms_log_rms_mae_db": 1.577522507470882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12202129195489105, "300–1000 Hz_10ms_envelope_correlation": 0.9883517441987186, "300–1000 Hz_10ms_modulation_mae_db": 0.33041638559993497, "300–1000 Hz_40ms_log_rms_mae_db": 0.9573480921341511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06407056874268068, "300–1000 Hz_40ms_envelope_correlation": 0.9965845670165676, "300–1000 Hz_40ms_modulation_mae_db": 0.3522059605487284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6124729611272612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2343208091003359, "1000–3000 Hz_10ms_envelope_correlation": 0.96349528907485, "1000–3000 Hz_10ms_modulation_mae_db": 0.34940116065242677, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0522198667980507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1298731303890852, "1000–3000 Hz_40ms_envelope_correlation": 0.9897856609270664, "1000–3000 Hz_40ms_modulation_mae_db": 0.4214848941602079, "3000–6000 Hz_10ms_log_rms_mae_db": 1.721379248436798, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3795351822856448, "3000–6000 Hz_10ms_envelope_correlation": 0.8965390373128549, "3000–6000 Hz_10ms_modulation_mae_db": 0.9650108025706302, "3000–6000 Hz_40ms_log_rms_mae_db": 1.148871624800344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27452346498637265, "3000–6000 Hz_40ms_envelope_correlation": 0.9429452436667443, "3000–6000 Hz_40ms_modulation_mae_db": 1.0367262027440203, "mr_stft_db": 5.899090011564844, "band_mae_db": 0.4587626476818188, "rms_gain_error_db": 0.06237008060423429, "output_peak": 0.5219797425480598}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.677549037570401, "spectral_convergence": 0.2709928830266073, "spectral_convergence_status": "ok", "amplitude_floor": 1.3235749190080935e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.817783276880349, "spectral_convergence": 0.2752861946413383, "spectral_convergence_status": "ok", "amplitude_floor": 1.1987597823804542e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.197054526405106, "spectral_convergence": 0.2736139355047201, "spectral_convergence_status": "ok", "amplitude_floor": 9.377214849682438e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.350500895926799e-05, "output_power": 4.7576940563483055e-05, "delta_db": 0.38857248851121595, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012608807022653607, "output_power": 0.001201977113858089, "delta_db": -0.20777799315554657, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00036191686574782775, "output_power": 0.00033737378389973405, "delta_db": -0.3049749010289867, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.9809147109830135e-05, "output_power": 2.991071301500025e-05, "delta_db": 0.014772156063397095, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.6488056661262093e-05, "output_power": 1.9108345797827913e-05, "delta_db": 0.6405362074727179, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.109371525887873e-05, "output_power": 6.526706255792397e-05, "delta_db": 0.28697530611834854, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.30726817539170215, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.18119635745537926, "pitch": null, "output_rms_dbfs": -27.692410622427804, "output_peak": 0.35427319937601226, "reference_rms_dbfs": -27.511214264972427, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.897462280285285}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018136510927426692, "log_rms_mae_db": 0.7586985152539231, "rms_envelope_nrmse": 0.09924625928408566, "envelope_correlation": 0.9908485281866493, "rms_gain_error_db": -0.18127514976287668, "reference_mean_power": 0.0017749427149271913, "output_mean_power": 0.0017023812451089743}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.578419076835135e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003061888162786739, "output_power": 0.0002999951071597579, "reference_active": true, "delta_db": -0.08875152247858661}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.561280729204581e-05, "output_power": 5.1874371295188156e-05, "reference_active": true, "delta_db": -0.3022197227591329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1447257663776473e-05, "output_power": 8.785219403812014e-06, "reference_active": true, "delta_db": -1.1494884624384851}], "active_band_count": 3, "active_band_mae_db": 0.5134865692254015}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015531555176710683, "log_rms_mae_db": 0.43019102947694654, "rms_envelope_nrmse": 0.05572587524306625, "envelope_correlation": 0.9972454559071846, "rms_gain_error_db": -0.18207668109822966, "reference_mean_power": 0.0017815693662331493, "output_mean_power": 0.0017084216577539168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.067921014076438e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002861407755621019, "output_power": 0.00028040366892592705, "reference_active": true, "delta_db": -0.08796058106842287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5605898871779113e-05, "output_power": 2.3816672006552713e-05, "reference_active": true, "delta_db": -0.3145895017100837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.159270454376389e-07, "output_power": 6.983090382595787e-07, "reference_active": true, "delta_db": -1.1781321984919904}], "active_band_count": 3, "active_band_mae_db": 0.5268940937568324}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017559797263300874, "log_rms_mae_db": 1.5836005488868454, "rms_envelope_nrmse": 0.11925888700093591, "envelope_correlation": 0.9892376161181324, "rms_gain_error_db": -0.2661717304517538, "reference_mean_power": 0.0012870696988326355, "output_mean_power": 0.0012105560231483333}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.703155295091611e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025376682215949657, "output_power": 0.00024190153330578423, "reference_active": true, "delta_db": -0.2079622001353398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.494529209431561e-05, "output_power": 4.1233830766395296e-05, "reference_active": true, "delta_db": -0.3743052305829467}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.800066525463509e-06, "output_power": 7.300367720768605e-06, "reference_active": true, "delta_db": -0.8114121912024406}], "active_band_count": 3, "active_band_mae_db": 0.4645598739735757}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014992083957937532, "log_rms_mae_db": 0.9911559198370041, "rms_envelope_nrmse": 0.07195242127888407, "envelope_correlation": 0.996407741496344, "rms_gain_error_db": -0.26702124416780804, "reference_mean_power": 0.0012919551058812144, "output_mean_power": 0.0012149133329626238}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4067631484646013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002445691472192921, "output_power": 0.0002326695033107912, "reference_active": true, "delta_db": -0.21662206489565558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8830927140807284e-05, "output_power": 1.737990441357132e-05, "reference_active": true, "delta_db": -0.348243194587379}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.050384653726169e-07, "output_power": 5.862323827009223e-07, "reference_active": true, "delta_db": -0.8014300718727474}], "active_band_count": 3, "active_band_mae_db": 0.455431777118594}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.2493337902299445e-05, "log_rms_mae_db": 1.5959224286332752, "rms_envelope_nrmse": 0.20110339741681427, "envelope_correlation": 0.9696759481438857, "rms_gain_error_db": -0.024560296905635673, "reference_mean_power": 4.199153922743893e-05, "output_mean_power": 4.1754738157964644e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0299126624198213e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.222152478948813e-06, "output_power": 4.170269065258118e-06, "reference_active": true, "delta_db": -0.053698368741382715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.960031981806193e-07, "output_power": 8.588515931687291e-07, "reference_active": true, "delta_db": -0.18391434223996417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.633269352164066e-07, "output_power": 2.734451872184375e-07, "reference_active": true, "delta_db": 0.16374999419675626}], "active_band_count": 3, "active_band_mae_db": 0.13378756839270103}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.820326553727612e-05, "log_rms_mae_db": 1.0058186471469517, "rms_envelope_nrmse": 0.08867648325175562, "envelope_correlation": 0.9940751804331581, "rms_gain_error_db": -0.024729041385711976, "reference_mean_power": 4.2190507757870256e-05, "output_mean_power": 4.195095462395658e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.548028980657037e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.958931660389438e-06, "output_power": 3.8996452110992304e-06, "reference_active": true, "delta_db": -0.06552908163970886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7410748918107156e-07, "output_power": 3.573144343418185e-07, "reference_active": true, "delta_db": -0.19945841695004857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8002833142836267e-08, "output_power": 1.665727766045136e-08, "reference_active": true, "delta_db": -0.3373683123367188}], "active_band_count": 3, "active_band_mae_db": 0.20078527030882543}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.219318962627022e-05, "log_rms_mae_db": 1.7245977420731369, "rms_envelope_nrmse": 0.39329052511073326, "envelope_correlation": 0.8970400944943506, "rms_gain_error_db": 0.8972244169875122, "reference_mean_power": 2.2095607497389627e-05, "output_mean_power": 2.7166168385093782e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.718240736767725e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.504526854650138e-06, "output_power": 9.689301708123181e-06, "reference_active": true, "delta_db": 0.08361977193557156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5665611692910136e-06, "output_power": 2.76687698400621e-06, "reference_active": true, "delta_db": 0.32638231401829754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.91578688150447e-07, "output_power": 8.997473010375699e-07, "reference_active": true, "delta_db": 1.8210803259727348}], "active_band_count": 3, "active_band_mae_db": 0.7436941373088679}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.1857789632782756e-05, "log_rms_mae_db": 1.1832696041078437, "rms_envelope_nrmse": 0.3004468934135877, "envelope_correlation": 0.9381991067105436, "rms_gain_error_db": 0.8968015632982974, "reference_mean_power": 2.220381322589703e-05, "output_mean_power": 2.7296547586548412e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6122764663239433e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.855212413255228e-06, "output_power": 1.0268263957331082e-05, "reference_active": true, "delta_db": 0.17831034937934026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3142139667203495e-06, "output_power": 1.4858523336688767e-06, "reference_active": true, "delta_db": 0.5330957238922682}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2621117296192714e-08, "output_power": 6.41232432643463e-08, "reference_active": true, "delta_db": 1.773906498575554}], "active_band_count": 3, "active_band_mae_db": 0.8284375239490541}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8499982151838101, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b4757344713027dd15348393b490fbf53c5310ed431a1a856c7f82dc06b08030", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9645843084423085, "standard_deviation": 0.00867374811334114, "minimum": 0.9500950755017643, "maximum": 0.9781922530467124, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6418631045046208, "standard_deviation": 0.030433588678773247, "minimum": 1.5959224286332752, "maximum": 1.6837665276635752, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.39414362040846235, "standard_deviation": 0.15277864923189183, "minimum": 0.13378756839270103, "maximum": 0.6548864293578781, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2262535035682673, "standard_deviation": 0.03283846680960363, "minimum": 0.17376804358982867, "maximum": 0.2753539440818146, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9891601465006372, "standard_deviation": 0.004190903388856974, "minimum": 0.9810335460965568, "maximum": 0.9940751804331581, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.063033472011878, "standard_deviation": 0.03771449444717912, "minimum": 1.0058186471469517, "maximum": 1.11089132906687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.413511304330182, "standard_deviation": 0.27224181134615205, "minimum": 0.1494892880883409, "maximum": 0.9371995893513887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12748657167956207, "standard_deviation": 0.033276736341497734, "minimum": 0.08867648325175562, "maximum": 0.19015497818198193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8984752634541416, "standard_deviation": 0.005038293829758508, "minimum": 0.8890266454629777, "maximum": 0.9037626017203807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7288091998198327, "standard_deviation": 0.019252428444052908, "minimum": 1.6955995835981672, "maximum": 1.756948614068809, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7767859526636209, "standard_deviation": 0.4118493952906385, "minimum": 0.2690587808618502, "maximum": 1.3009198520746608, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3897551384096837, "standard_deviation": 0.007983071626612897, "minimum": 0.3795351822856448, "maximum": 0.40304430714160683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9392378543317749, "standard_deviation": 0.005094111036155269, "minimum": 0.930448943248311, "maximum": 0.9449729009407941, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1960026858340953, "standard_deviation": 0.02928916177312076, "minimum": 1.148871624800344, "maximum": 1.24226911284314, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8934149555118951, "standard_deviation": 0.3760626098598822, "minimum": 0.43281304782241503, "maximum": 1.4557777698804815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2972097556280855, "standard_deviation": 0.011177978385239486, "minimum": 0.27452346498637265, "maximum": 0.31235398817886423, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916873605355927, "standard_deviation": 0.002302266129088737, "minimum": 0.9883517441987186, "maximum": 0.99401294045184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5412168252106015, "standard_deviation": 0.03934804560593508, "minimum": 1.491432067672651, "maximum": 1.5943545715359415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7250960178258262, "standard_deviation": 0.2563229031461586, "minimum": 0.33041638559993497, "maximum": 1.111608252972843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10333086152328941, "standard_deviation": 0.013729520657009388, "minimum": 0.08952902285419576, "maximum": 0.12202129195489105, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974321333138443, "standard_deviation": 0.0008317080050872097, "minimum": 0.9963892786277919, "maximum": 0.998418903817574, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9373575907898535, "standard_deviation": 0.02675106891101068, "minimum": 0.9093069292750531, "maximum": 0.9911559198370041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8128269437445561, "standard_deviation": 0.329858385220682, "minimum": 0.3522059605487284, "maximum": 1.2852826550148864, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05794531265961426, "standard_deviation": 0.009224327634333847, "minimum": 0.04796641898195895, "maximum": 0.07195242127888407, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.40563886462305876, "standard_deviation": 0.07675199012253284, "minimum": 0.30726817539170215, "maximum": 0.5305696520090498, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.89191078126604, "standard_deviation": 0.009859627225469198, "minimum": 5.879171088466246, "maximum": 5.907259468808571, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44908336038913554, "standard_deviation": 0.07370778306777849, "minimum": 0.35183505327901077, "maximum": 0.5219797425480598, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04418721593520995, "standard_deviation": 0.11202786542060393, "minimum": -0.19464346034018504, "maximum": 0.13756000735556706, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9929132418139, "standard_deviation": 0.002190441379462755, "minimum": 0.9888720346167416, "maximum": 0.9949792251046491, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7419538807059775, "standard_deviation": 0.018508529273592002, "minimum": 0.709422913932188, "maximum": 0.7634533889403365, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5218634261569233, "standard_deviation": 0.13880041920505867, "minimum": 0.28363767681390767, "maximum": 0.735451592431365, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08643122507055523, "standard_deviation": 0.01305516829402889, "minimum": 0.07310363338488933, "maximum": 0.1091318207496345, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977643132204412, "standard_deviation": 0.0007358089831600242, "minimum": 0.9966922841134443, "maximum": 0.99857114458172, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40421832163896854, "standard_deviation": 0.017303614106864464, "minimum": 0.3842847151027591, "maximum": 0.43019102947694654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5271211424007178, "standard_deviation": 0.16428778653248696, "minimum": 0.23182985743355022, "maximum": 0.7841465116951071, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04815125946389274, "standard_deviation": 0.008436530258361007, "minimum": 0.038209632579272, "maximum": 0.05928535136731324, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0032", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.3008475303649902, "predicted_mos": 2.3008475303649902, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a1ae3efcc1f3443ac20fb9b847532e1ee704556d2187ca18da1a003fd8d21fb6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8733300999738276, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7412347295592525, "全帯域_10ms_rms_envelope_nrmse": 0.09334236111015223, "全帯域_10ms_envelope_correlation": 0.9919110513791259, "全帯域_10ms_modulation_mae_db": 0.6243164947927756, "全帯域_40ms_log_rms_mae_db": 0.4070040300246179, "全帯域_40ms_rms_envelope_nrmse": 0.052220716651716584, "全帯域_40ms_envelope_correlation": 0.997387958068182, "全帯域_40ms_modulation_mae_db": 0.6169903382851514, "300–1000 Hz_10ms_log_rms_mae_db": 1.328789182905543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10541397817456691, "300–1000 Hz_10ms_envelope_correlation": 0.9913667662066186, "300–1000 Hz_10ms_modulation_mae_db": 0.8139452522363326, "300–1000 Hz_40ms_log_rms_mae_db": 0.736706015380783, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05900106294283724, "300–1000 Hz_40ms_envelope_correlation": 0.9971694770260985, "300–1000 Hz_40ms_modulation_mae_db": 0.744862412966785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2969954805070787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.187094930247277, "1000–3000 Hz_10ms_envelope_correlation": 0.9779750490813253, "1000–3000 Hz_10ms_modulation_mae_db": 0.260593031059321, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8404293946373884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12550620192685058, "1000–3000 Hz_40ms_envelope_correlation": 0.9910282450803921, "1000–3000 Hz_40ms_modulation_mae_db": 0.42145148851493897, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3604527208392057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3558107909563143, "3000–6000 Hz_10ms_envelope_correlation": 0.908874765122161, "3000–6000 Hz_10ms_modulation_mae_db": 0.5986814539351241, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9626977225492471, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2745526359681328, "3000–6000 Hz_40ms_envelope_correlation": 0.941989534768019, "3000–6000 Hz_40ms_modulation_mae_db": 0.9446051654625464, "mr_stft_db": 5.582503418427476, "band_mae_db": 0.3957299963572805, "rms_gain_error_db": 0.06899311309921365, "output_peak": 0.44727831188726974}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7822935960322657, "全帯域_10ms_rms_envelope_nrmse": 0.07617424445482597, "全帯域_10ms_envelope_correlation": 0.9945615489429213, "全帯域_10ms_modulation_mae_db": 0.31910263642652575, "全帯域_40ms_log_rms_mae_db": 0.4172884872699721, "全帯域_40ms_rms_envelope_nrmse": 0.03958075232674931, "全帯域_40ms_envelope_correlation": 0.9985482719002798, "全帯域_40ms_modulation_mae_db": 0.43242049480061673, "300–1000 Hz_10ms_log_rms_mae_db": 1.3694252812995917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10533311297399121, "300–1000 Hz_10ms_envelope_correlation": 0.9915384318614182, "300–1000 Hz_10ms_modulation_mae_db": 0.38233876075795487, "300–1000 Hz_40ms_log_rms_mae_db": 0.8153989168105378, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05915024818213085, "300–1000 Hz_40ms_envelope_correlation": 0.9975208148388379, "300–1000 Hz_40ms_modulation_mae_db": 0.5603334652180395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3614032675267502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2305454724717315, "1000–3000 Hz_10ms_envelope_correlation": 0.9650414118820002, "1000–3000 Hz_10ms_modulation_mae_db": 0.3068063414474665, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8557271757388333, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13301803398859888, "1000–3000 Hz_40ms_envelope_correlation": 0.9892082693445483, "1000–3000 Hz_40ms_modulation_mae_db": 0.3212718578537646, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3305308635870026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36535989795811125, "3000–6000 Hz_10ms_envelope_correlation": 0.9123847349184325, "3000–6000 Hz_10ms_modulation_mae_db": 0.2155537397184086, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9524045092232267, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2697190514240582, "3000–6000 Hz_40ms_envelope_correlation": 0.9516164503595019, "3000–6000 Hz_40ms_modulation_mae_db": 0.16584344518068136, "mr_stft_db": 5.57231879374138, "band_mae_db": 0.4780292760366487, "rms_gain_error_db": -0.11160998557527124, "output_peak": 0.3915641807634752}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7917169923783363, "全帯域_10ms_rms_envelope_nrmse": 0.09885529307045957, "全帯域_10ms_envelope_correlation": 0.9909210915000162, "全帯域_10ms_modulation_mae_db": 0.8896045142178292, "全帯域_40ms_log_rms_mae_db": 0.4362903028272204, "全帯域_40ms_rms_envelope_nrmse": 0.06234498243248149, "全帯域_40ms_envelope_correlation": 0.9963929184704541, "全帯域_40ms_modulation_mae_db": 0.9694065043539807, "300–1000 Hz_10ms_log_rms_mae_db": 1.3601856675961264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12171756629622266, "300–1000 Hz_10ms_envelope_correlation": 0.9888495989532217, "300–1000 Hz_10ms_modulation_mae_db": 1.0036229432255754, "300–1000 Hz_40ms_log_rms_mae_db": 0.8301914836784735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08548066978289737, "300–1000 Hz_40ms_envelope_correlation": 0.9946403993625953, "300–1000 Hz_40ms_modulation_mae_db": 1.08636429980656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2870911794439839, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17961797068219407, "1000–3000 Hz_10ms_envelope_correlation": 0.9760515837181111, "1000–3000 Hz_10ms_modulation_mae_db": 0.4263967502636738, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7808373116791059, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08285399303038915, "1000–3000 Hz_40ms_envelope_correlation": 0.9946748817935078, "1000–3000 Hz_40ms_modulation_mae_db": 0.6435543787661995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3206016087333483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3752923957704718, "3000–6000 Hz_10ms_envelope_correlation": 0.9047230197024488, "3000–6000 Hz_10ms_modulation_mae_db": 0.2076539953302996, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9594475330016514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29241809663387225, "3000–6000 Hz_40ms_envelope_correlation": 0.9393376869923749, "3000–6000 Hz_40ms_modulation_mae_db": 0.11146162300944717, "mr_stft_db": 5.564633928174032, "band_mae_db": 0.37096376353990296, "rms_gain_error_db": -0.18118883800027757, "output_peak": 0.4151190102899542}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7579826043747331, "全帯域_10ms_rms_envelope_nrmse": 0.1140667589156214, "全帯域_10ms_envelope_correlation": 0.9877866690625833, "全帯域_10ms_modulation_mae_db": 0.6289800077083572, "全帯域_40ms_log_rms_mae_db": 0.43358024686559377, "全帯域_40ms_rms_envelope_nrmse": 0.0636301297713275, "全帯域_40ms_envelope_correlation": 0.9959773929253373, "全帯域_40ms_modulation_mae_db": 0.6193265865823069, "300–1000 Hz_10ms_log_rms_mae_db": 1.3097298900714798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13671093782058652, "300–1000 Hz_10ms_envelope_correlation": 0.9854661967800323, "300–1000 Hz_10ms_modulation_mae_db": 0.8724241223894419, "300–1000 Hz_40ms_log_rms_mae_db": 0.7622175715140781, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07536350719187504, "300–1000 Hz_40ms_envelope_correlation": 0.9953601173453018, "300–1000 Hz_40ms_modulation_mae_db": 0.7999024628911613, "1000–3000 Hz_10ms_log_rms_mae_db": 1.246872811463755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18543591618519076, "1000–3000 Hz_10ms_envelope_correlation": 0.9779650530164847, "1000–3000 Hz_10ms_modulation_mae_db": 0.25879906836485406, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8207162759079784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13396001168517782, "1000–3000 Hz_40ms_envelope_correlation": 0.9890684883552296, "1000–3000 Hz_40ms_modulation_mae_db": 0.34614612135939815, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3332854157276075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34998531744015055, "3000–6000 Hz_10ms_envelope_correlation": 0.9139724175649109, "3000–6000 Hz_10ms_modulation_mae_db": 0.4510994343648644, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9597597583759965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27076603196671567, "3000–6000 Hz_40ms_envelope_correlation": 0.9454016496093537, "3000–6000 Hz_40ms_modulation_mae_db": 0.25851519877427004, "mr_stft_db": 5.553752119116683, "band_mae_db": 0.43269262863315916, "rms_gain_error_db": 0.04840676314766888, "output_peak": 0.4331349853841247}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7975634491476631, "全帯域_10ms_rms_envelope_nrmse": 0.09381614649358, "全帯域_10ms_envelope_correlation": 0.9916668782237482, "全帯域_10ms_modulation_mae_db": 0.450744552754948, "全帯域_40ms_log_rms_mae_db": 0.43362501807007836, "全帯域_40ms_rms_envelope_nrmse": 0.05334278717473959, "全帯域_40ms_envelope_correlation": 0.9972121436615001, "全帯域_40ms_modulation_mae_db": 0.517764487978211, "300–1000 Hz_10ms_log_rms_mae_db": 1.34940267095347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11175796979302417, "300–1000 Hz_10ms_envelope_correlation": 0.9904441426781146, "300–1000 Hz_10ms_modulation_mae_db": 0.6167404272792189, "300–1000 Hz_40ms_log_rms_mae_db": 0.8256857057338929, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06754184184839836, "300–1000 Hz_40ms_envelope_correlation": 0.9966310931893836, "300–1000 Hz_40ms_modulation_mae_db": 0.766964305435009, "1000–3000 Hz_10ms_log_rms_mae_db": 1.284959038873991, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16318218317664024, "1000–3000 Hz_10ms_envelope_correlation": 0.9812568495128844, "1000–3000 Hz_10ms_modulation_mae_db": 0.1765057800270252, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8267129471509788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09962848294878456, "1000–3000 Hz_40ms_envelope_correlation": 0.9929967916916639, "1000–3000 Hz_40ms_modulation_mae_db": 0.3095072884169683, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3067448315928707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3593588839850034, "3000–6000 Hz_10ms_envelope_correlation": 0.9211906992298364, "3000–6000 Hz_10ms_modulation_mae_db": 1.513863503619581, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9469179945818915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27945806729779227, "3000–6000 Hz_40ms_envelope_correlation": 0.9526510295622357, "3000–6000 Hz_40ms_modulation_mae_db": 1.7171964544681977, "mr_stft_db": 5.565067181973402, "band_mae_db": 0.4270229074041465, "rms_gain_error_db": -0.10880270086255672, "output_peak": 0.4091748220390768}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7858623530354696, "全帯域_10ms_rms_envelope_nrmse": 0.09879096507246396, "全帯域_10ms_envelope_correlation": 0.9912266134067621, "全帯域_10ms_modulation_mae_db": 0.42805608514297205, "全帯域_40ms_log_rms_mae_db": 0.42635080874489656, "全帯域_40ms_rms_envelope_nrmse": 0.07158820268474482, "全帯域_40ms_envelope_correlation": 0.9954493077116184, "全帯域_40ms_modulation_mae_db": 0.4786935792209382, "300–1000 Hz_10ms_log_rms_mae_db": 1.3583465204131902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11433477262299958, "300–1000 Hz_10ms_envelope_correlation": 0.990702087039399, "300–1000 Hz_10ms_modulation_mae_db": 0.4072083034412923, "300–1000 Hz_40ms_log_rms_mae_db": 0.8095972456119161, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08770592224129978, "300–1000 Hz_40ms_envelope_correlation": 0.9948170978394157, "300–1000 Hz_40ms_modulation_mae_db": 0.3770475759888963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2840826929288416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18872268209953963, "1000–3000 Hz_10ms_envelope_correlation": 0.9731565718647531, "1000–3000 Hz_10ms_modulation_mae_db": 0.22450350568391864, "1000–3000 Hz_40ms_log_rms_mae_db": 0.78975016037458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09157809741351343, "1000–3000 Hz_40ms_envelope_correlation": 0.9935157586637644, "1000–3000 Hz_40ms_modulation_mae_db": 0.28146639679405977, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2911280918901602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3669011215735093, "3000–6000 Hz_10ms_envelope_correlation": 0.9022319350303384, "3000–6000 Hz_10ms_modulation_mae_db": 0.33223308065675516, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9434439425973594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2775825723924702, "3000–6000 Hz_40ms_envelope_correlation": 0.9401747271467783, "3000–6000 Hz_40ms_modulation_mae_db": 0.39141682710105125, "mr_stft_db": 5.564832407535088, "band_mae_db": 0.3687603939804836, "rms_gain_error_db": -0.24232622448850813, "output_peak": 0.3975996466537862}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7682449058204059, "全帯域_10ms_rms_envelope_nrmse": 0.07935447486765931, "全帯域_10ms_envelope_correlation": 0.9941146379142732, "全帯域_10ms_modulation_mae_db": 0.6156579929297422, "全帯域_40ms_log_rms_mae_db": 0.4094214233001654, "全帯域_40ms_rms_envelope_nrmse": 0.03908939786495843, "全帯域_40ms_envelope_correlation": 0.9986253870554475, "全帯域_40ms_modulation_mae_db": 0.658480359764139, "300–1000 Hz_10ms_log_rms_mae_db": 1.301728190666513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09352720842799553, "300–1000 Hz_10ms_envelope_correlation": 0.9933158201040657, "300–1000 Hz_10ms_modulation_mae_db": 0.7150869724498218, "300–1000 Hz_40ms_log_rms_mae_db": 0.7406447229076282, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05335235960136915, "300–1000 Hz_40ms_envelope_correlation": 0.9979234179951217, "300–1000 Hz_40ms_modulation_mae_db": 0.80416707608784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2824409665789007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1400075522342196, "1000–3000 Hz_10ms_envelope_correlation": 0.9853446860064575, "1000–3000 Hz_10ms_modulation_mae_db": 0.3531248717219219, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8030698186916809, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06785286758313004, "1000–3000 Hz_40ms_envelope_correlation": 0.9966244818366041, "1000–3000 Hz_40ms_modulation_mae_db": 0.52680558139612, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3372642055973185, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.33893435301246794, "3000–6000 Hz_10ms_envelope_correlation": 0.919449073042528, "3000–6000 Hz_10ms_modulation_mae_db": 0.3978465795307738, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9502572701372807, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2625992832544113, "3000–6000 Hz_40ms_envelope_correlation": 0.9489788161523216, "3000–6000 Hz_40ms_modulation_mae_db": 0.5738858554447551, "mr_stft_db": 5.573783746274818, "band_mae_db": 0.29598058515362063, "rms_gain_error_db": -0.127180097539597, "output_peak": 0.4407019119410912}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.823195775848015, "全帯域_10ms_rms_envelope_nrmse": 0.09893675710311137, "全帯域_10ms_envelope_correlation": 0.9907026512828747, "全帯域_10ms_modulation_mae_db": 0.6520101180276665, "全帯域_40ms_log_rms_mae_db": 0.44030755297401025, "全帯域_40ms_rms_envelope_nrmse": 0.048180179770258166, "全帯域_40ms_envelope_correlation": 0.997756304033789, "全帯域_40ms_modulation_mae_db": 0.7233862613384607, "300–1000 Hz_10ms_log_rms_mae_db": 1.362102873679043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12911293544321809, "300–1000 Hz_10ms_envelope_correlation": 0.9871345254369185, "300–1000 Hz_10ms_modulation_mae_db": 0.8285807291334363, "300–1000 Hz_40ms_log_rms_mae_db": 0.8151601943835529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06093322404129855, "300–1000 Hz_40ms_envelope_correlation": 0.9974780906515215, "300–1000 Hz_40ms_modulation_mae_db": 0.9027185002438395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.291237633622342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1660440644867348, "1000–3000 Hz_10ms_envelope_correlation": 0.9794928511642689, "1000–3000 Hz_10ms_modulation_mae_db": 0.44144129091345524, "1000–3000 Hz_40ms_log_rms_mae_db": 0.786492591198041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08800566952820581, "1000–3000 Hz_40ms_envelope_correlation": 0.9939746235731027, "1000–3000 Hz_40ms_modulation_mae_db": 0.498129572936264, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2889230225758208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3412455372343769, "3000–6000 Hz_10ms_envelope_correlation": 0.9157001152628095, "3000–6000 Hz_10ms_modulation_mae_db": 0.4448305101970018, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9338475203476483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2697477495900066, "3000–6000 Hz_40ms_envelope_correlation": 0.9437181477931504, "3000–6000 Hz_40ms_modulation_mae_db": 0.36432955650889753, "mr_stft_db": 5.572987801202813, "band_mae_db": 0.324338138419926, "rms_gain_error_db": -0.11161358942772676, "output_peak": 0.38471529357397594}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.2904544622581495, "spectral_convergence": 0.2671379598840179, "spectral_convergence_status": "ok", "amplitude_floor": 1.3235749190080935e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.431944307790549, "spectral_convergence": 0.28648042149506037, "spectral_convergence_status": "ok", "amplitude_floor": 1.1987597823804542e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0251114852337295, "spectral_convergence": 0.2950349130634741, "spectral_convergence_status": "ok", "amplitude_floor": 9.377214849682438e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.350500895926799e-05, "output_power": 5.2248352707602916e-05, "delta_db": 0.7953334008086855, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012608807022653607, "output_power": 0.0012590063555660295, "delta_db": -0.006460754471415576, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00036191686574782775, "output_power": 0.00037338945797926704, "delta_db": 0.13553230006030442, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.9809147109830135e-05, "output_power": 3.736448460661133e-05, "delta_db": 0.981094458085824, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.6488056661262093e-05, "output_power": 1.8028482586737323e-05, "delta_db": 0.3878970352865827, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.109371525887873e-05, "output_power": 6.205870937539193e-05, "delta_db": 0.06806202943087111, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3957299963572805, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.06899311309921365, "pitch": null, "output_rms_dbfs": -27.442221151873213, "output_peak": 0.44727831188726974, "reference_rms_dbfs": -27.511214264972427, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.582503418427476}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018136510927426692, "log_rms_mae_db": 0.7412347295592525, "rms_envelope_nrmse": 0.09334236111015223, "envelope_correlation": 0.9919110513791259, "rms_gain_error_db": 0.06935188142223096, "reference_mean_power": 0.0017749427149271913, "output_mean_power": 0.001803514046791313}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.578419076835135e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003061888162786739, "output_power": 0.0003036390379503315, "reference_active": true, "delta_db": -0.036317170604036685}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.561280729204581e-05, "output_power": 4.997658163451247e-05, "reference_active": true, "delta_db": -0.46408271136190943}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1447257663776473e-05, "output_power": 8.345387329332858e-06, "reference_active": true, "delta_db": -1.3725496024123807}], "active_band_count": 3, "active_band_mae_db": 0.6243164947927756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015531555176710683, "log_rms_mae_db": 0.4070040300246179, "rms_envelope_nrmse": 0.052220716651716584, "envelope_correlation": 0.997387958068182, "rms_gain_error_db": 0.07018764107206034, "reference_mean_power": 0.0017815693662331493, "output_mean_power": 0.0018105957665195279}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.067921014076438e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002861407755621019, "output_power": 0.0002835256031971896, "reference_active": true, "delta_db": -0.039874668693312784}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5605898871779113e-05, "output_power": 2.2590083134834024e-05, "reference_active": true, "delta_db": -0.5442219675962036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.159270454376389e-07, "output_power": 6.841848371752152e-07, "reference_active": true, "delta_db": -1.2668743785659378}], "active_band_count": 3, "active_band_mae_db": 0.6169903382851514}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017559797263300874, "log_rms_mae_db": 1.328789182905543, "rms_envelope_nrmse": 0.10541397817456691, "envelope_correlation": 0.9913667662066186, "rms_gain_error_db": 0.03710785414537966, "reference_mean_power": 0.0012870696988326355, "output_mean_power": 0.001298114052601432}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.703155295091611e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025376682215949657, "output_power": 0.0002469471928082279, "reference_active": true, "delta_db": -0.1183074753419058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.494529209431561e-05, "output_power": 3.891705346237378e-05, "reference_active": true, "delta_db": -0.6254425622004668}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.800066525463509e-06, "output_power": 5.95219818870204e-06, "reference_active": true, "delta_db": -1.6980857191666254}], "active_band_count": 3, "active_band_mae_db": 0.8139452522363326}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014992083957937532, "log_rms_mae_db": 0.736706015380783, "rms_envelope_nrmse": 0.05900106294283724, "envelope_correlation": 0.9971694770260985, "rms_gain_error_db": 0.03783686497861466, "reference_mean_power": 0.0012919551058812144, "output_mean_power": 0.001303260129481172}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4067631484646013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002445691472192921, "output_power": 0.00023580928013773494, "reference_active": true, "delta_db": -0.15840776723719618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8830927140807284e-05, "output_power": 1.6139549033115133e-05, "reference_active": true, "delta_db": -0.6698030741699426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.050384653726169e-07, "output_power": 5.100058668784549e-07, "reference_active": true, "delta_db": -1.406376397493216}], "active_band_count": 3, "active_band_mae_db": 0.744862412966785}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.2493337902299445e-05, "log_rms_mae_db": 1.2969954805070787, "rms_envelope_nrmse": 0.187094930247277, "envelope_correlation": 0.9779750490813253, "rms_gain_error_db": 0.43643355210057133, "reference_mean_power": 4.199153922743893e-05, "output_mean_power": 4.6430691259979515e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0299126624198213e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.222152478948813e-06, "output_power": 3.843831007452925e-06, "reference_active": true, "delta_db": -0.40769627552679927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.960031981806193e-07, "output_power": 8.247753866155042e-07, "reference_active": true, "delta_db": -0.35973867806627835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.633269352164066e-07, "output_power": 2.641981051540926e-07, "reference_active": true, "delta_db": 0.014344139584885477}], "active_band_count": 3, "active_band_mae_db": 0.260593031059321}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.820326553727612e-05, "log_rms_mae_db": 0.8404293946373884, "rms_envelope_nrmse": 0.12550620192685058, "envelope_correlation": 0.9910282450803921, "rms_gain_error_db": 0.43651737708169985, "reference_mean_power": 4.2190507757870256e-05, "output_mean_power": 4.665159425971311e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.548028980657037e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.958931660389438e-06, "output_power": 3.592468896285465e-06, "reference_active": true, "delta_db": -0.42184988280625324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7410748918107156e-07, "output_power": 3.22717429897934e-07, "reference_active": true, "delta_db": -0.6417398014984681}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8002833142836267e-08, "output_power": 1.7189544878288722e-08, "reference_active": true, "delta_db": -0.20076478124009575}], "active_band_count": 3, "active_band_mae_db": 0.42145148851493897}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.219318962627022e-05, "log_rms_mae_db": 1.3604527208392057, "rms_envelope_nrmse": 0.3558107909563143, "envelope_correlation": 0.908874765122161, "rms_gain_error_db": 0.5185035796253602, "reference_mean_power": 2.2095607497389627e-05, "output_mean_power": 2.4897532304084497e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.718240736767725e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.504526854650138e-06, "output_power": 8.92827621897157e-06, "reference_active": true, "delta_db": -0.2716288429558175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5665611692910136e-06, "output_power": 1.8389591538101345e-06, "reference_active": true, "delta_db": -1.447795363565196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.91578688150447e-07, "output_power": 5.812333221201175e-07, "reference_active": true, "delta_db": -0.07662015528435874}], "active_band_count": 3, "active_band_mae_db": 0.5986814539351241}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.1857789632782756e-05, "log_rms_mae_db": 0.9626977225492471, "rms_envelope_nrmse": 0.2745526359681328, "envelope_correlation": 0.941989534768019, "rms_gain_error_db": 0.5184301850276752, "reference_mean_power": 2.220381322589703e-05, "output_mean_power": 2.5019036688542153e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6122764663239433e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.855212413255228e-06, "output_power": 9.031542774385503e-06, "reference_active": true, "delta_db": -0.37904046101098865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3142139667203495e-06, "output_power": 9.089666162050039e-07, "reference_active": true, "delta_db": -1.601181452180111}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2621117296192714e-08, "output_power": 3.501591441141999e-08, "reference_active": true, "delta_db": -0.8535935831965389}], "active_band_count": 3, "active_band_mae_db": 0.9446051654625464}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8759955718586094, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "746b9c826a215073027b7a1221c390f79b582c97754828a2c316d841a903e9a9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9770355070307857, "standard_deviation": 0.0060299263717510515, "minimum": 0.9650414118820002, "maximum": 0.9853446860064575, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2918853838682054, "standard_deviation": 0.03186228882850685, "minimum": 1.246872811463755, "maximum": 1.3614032675267502, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3060213299352046, "standard_deviation": 0.09468841625321203, "minimum": 0.1765057800270252, "maximum": 0.44144129091345524, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18008134644794094, "standard_deviation": 0.026153156531375586, "minimum": 0.1400075522342196, "maximum": 0.2305454724717315, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9926364425423516, "standard_deviation": 0.002667862612847224, "minimum": 0.9890684883552296, "maximum": 0.9966244818366041, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8129669594223233, "standard_deviation": 0.027260136048426684, "minimum": 0.7808373116791059, "maximum": 0.8557271757388333, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4185415857547141, "standard_deviation": 0.12765755176824198, "minimum": 0.28146639679405977, "maximum": 0.6435543787661995, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1028004197630813, "standard_deviation": 0.02499615570802624, "minimum": 0.06785286758313004, "maximum": 0.13396001168517782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9123158449841833, "standard_deviation": 0.006708475562450926, "minimum": 0.9022319350303384, "maximum": 0.9211906992298364, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3211163450679169, "standard_deviation": 0.024457934712933095, "minimum": 1.2889230225758208, "maximum": 1.3604527208392057, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5202202871691011, "standard_deviation": 0.42169356055849017, "minimum": 0.2076539953302996, "maximum": 1.513863503619581, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.35661103724130067, "standard_deviation": 0.012717123048512498, "minimum": 0.33893435301246794, "maximum": 0.3752923957704718, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9454835052979669, "standard_deviation": 0.005105862776076431, "minimum": 0.9393376869923749, "maximum": 0.9526510295622357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9510970313517877, "standard_deviation": 0.00967577384524335, "minimum": 0.9338475203476483, "maximum": 0.9626977225492471, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5659067657437308, "standard_deviation": 0.5346674299859434, "minimum": 0.11146162300944717, "maximum": 1.7171964544681977, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2746054360659324, "standard_deviation": 0.008912812902746845, "minimum": 0.2625992832544113, "maximum": 0.29241809663387225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9898521961324735, "standard_deviation": 0.002558868197330369, "minimum": 0.9854661967800323, "maximum": 0.9933158201040657, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3424637846981196, "standard_deviation": 0.025759830016695483, "minimum": 1.301728190666513, "maximum": 1.3694252812995917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7049934388641342, "standard_deviation": 0.2220133304462228, "minimum": 0.38233876075795487, "maximum": 1.0036229432255754, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11473856019407558, "standard_deviation": 0.014002516320153138, "minimum": 0.09352720842799553, "maximum": 0.13671093782058652, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964425635310346, "standard_deviation": 0.0013122196601221344, "minimum": 0.9946403993625953, "maximum": 0.9979234179951217, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7919502320026077, "standard_deviation": 0.038856228795145935, "minimum": 0.736706015380783, "maximum": 0.8301914836784735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7552950123297664, "standard_deviation": 0.21281950791436466, "minimum": 0.3770475759888963, "maximum": 1.08636429980656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06856610447901329, "standard_deviation": 0.01292592194745005, "minimum": 0.05335235960136915, "maximum": 0.08770592224129978, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.386689711190646, "standard_deviation": 0.05958464328845922, "minimum": 0.29598058515362063, "maximum": 0.4780292760366487, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.568734924555711, "standard_deviation": 0.008585516934734179, "minimum": 5.553752119116683, "maximum": 5.582503418427476, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4149110203165942, "standard_deviation": 0.023404554735061943, "minimum": 0.38471529357397594, "maximum": 0.44727831188726974, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09566519495588185, "standard_deviation": 0.10580791855794146, "minimum": -0.24232622448850813, "maximum": 0.06899311309921365, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9916113927140382, "standard_deviation": 0.0021092119430443715, "minimum": 0.9877866690625833, "maximum": 0.9945615489429213, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7810118007745176, "standard_deviation": 0.025266493025113086, "minimum": 0.7412347295592525, "maximum": 0.823195775848015, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5760590502501021, "standard_deviation": 0.17501510432480671, "minimum": 0.31910263642652575, "maximum": 0.8896045142178292, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09416712513598423, "standard_deviation": 0.011988479070039226, "minimum": 0.07617424445482597, "maximum": 0.1140667589156214, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997168710478326, "standard_deviation": 0.0011586787504034917, "minimum": 0.9954493077116184, "maximum": 0.9986253870554475, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42548348375956935, "standard_deviation": 0.01273786670184723, "minimum": 0.4070040300246179, "maximum": 0.44030755297401025, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6270585765404756, "standard_deviation": 0.16880714320347492, "minimum": 0.43242049480061673, "maximum": 0.9694065043539807, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05374714358462199, "standard_deviation": 0.011581514574565086, "minimum": 0.03908939786495843, "maximum": 0.07158820268474482, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0032", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.774578809738159, "predicted_mos": 2.774578809738159, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b971df5ab58959aa35b262fec7fa9629ac7ff56f75e85c2bda44134d85cb6b1e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.2069740002043545, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7658343517362223, "全帯域_10ms_rms_envelope_nrmse": 0.07811514613296058, "全帯域_10ms_envelope_correlation": 0.9941773191656993, "全帯域_10ms_modulation_mae_db": 0.4624876712349642, "全帯域_40ms_log_rms_mae_db": 0.4203790910690272, "全帯域_40ms_rms_envelope_nrmse": 0.03533583422094867, "全帯域_40ms_envelope_correlation": 0.9987245867692061, "全帯域_40ms_modulation_mae_db": 0.4618905182578324, "300–1000 Hz_10ms_log_rms_mae_db": 1.4160035814085463, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09744593557973465, "300–1000 Hz_10ms_envelope_correlation": 0.9925654356062091, "300–1000 Hz_10ms_modulation_mae_db": 0.676338279460995, "300–1000 Hz_40ms_log_rms_mae_db": 0.8263704310701557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04417447948295582, "300–1000 Hz_40ms_envelope_correlation": 0.998479148915275, "300–1000 Hz_40ms_modulation_mae_db": 0.7224475115972954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.401274285078899, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18444774146282117, "1000–3000 Hz_10ms_envelope_correlation": 0.9771225759951963, "1000–3000 Hz_10ms_modulation_mae_db": 0.6699544788933754, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8774579310817295, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11560715825160295, "1000–3000 Hz_40ms_envelope_correlation": 0.9914321486169919, "1000–3000 Hz_40ms_modulation_mae_db": 0.880645463526451, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4100385275228922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37119469648601544, "3000–6000 Hz_10ms_envelope_correlation": 0.9054949760539067, "3000–6000 Hz_10ms_modulation_mae_db": 0.7590258035296819, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9800544480945683, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.28548310363535, "3000–6000 Hz_40ms_envelope_correlation": 0.9415588465499863, "3000–6000 Hz_40ms_modulation_mae_db": 0.7768909793083827, "mr_stft_db": 5.644193781449182, "band_mae_db": 0.3900528659164319, "rms_gain_error_db": -0.029013423575384854, "output_peak": 0.48123037128047985}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7793087041764197, "全帯域_10ms_rms_envelope_nrmse": 0.09156544408247037, "全帯域_10ms_envelope_correlation": 0.9921131693930203, "全帯域_10ms_modulation_mae_db": 0.746870980705992, "全帯域_40ms_log_rms_mae_db": 0.41319995883026894, "全帯域_40ms_rms_envelope_nrmse": 0.05255525213293205, "全帯域_40ms_envelope_correlation": 0.9972507988616607, "全帯域_40ms_modulation_mae_db": 0.8322488918748352, "300–1000 Hz_10ms_log_rms_mae_db": 1.3718309589600917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11784853478888507, "300–1000 Hz_10ms_envelope_correlation": 0.9891692079230536, "300–1000 Hz_10ms_modulation_mae_db": 0.7102697986126904, "300–1000 Hz_40ms_log_rms_mae_db": 0.8359995952199604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06813485452205521, "300–1000 Hz_40ms_envelope_correlation": 0.996178290399345, "300–1000 Hz_40ms_modulation_mae_db": 0.7735345876511713, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4301555212054242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18760333480553992, "1000–3000 Hz_10ms_envelope_correlation": 0.973458325916785, "1000–3000 Hz_10ms_modulation_mae_db": 0.42505119175640743, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9025276575439131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09542355520277875, "1000–3000 Hz_40ms_envelope_correlation": 0.992983481372986, "1000–3000 Hz_40ms_modulation_mae_db": 0.28008007762130915, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4522366561023323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37631964569001874, "3000–6000 Hz_10ms_envelope_correlation": 0.9011924286927226, "3000–6000 Hz_10ms_modulation_mae_db": 0.49251652517123773, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9915507837568315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27419190338461163, "3000–6000 Hz_40ms_envelope_correlation": 0.945261020596478, "3000–6000 Hz_40ms_modulation_mae_db": 0.5098619790578779, "mr_stft_db": 5.65029021380681, "band_mae_db": 0.3217166009131013, "rms_gain_error_db": 0.04403278023635524, "output_peak": 0.49273878208177924}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7700375209566664, "全帯域_10ms_rms_envelope_nrmse": 0.08050618447065763, "全帯域_10ms_envelope_correlation": 0.993816335431304, "全帯域_10ms_modulation_mae_db": 0.5412042011357431, "全帯域_40ms_log_rms_mae_db": 0.4232417654116366, "全帯域_40ms_rms_envelope_nrmse": 0.04169305239164172, "全帯域_40ms_envelope_correlation": 0.9982321177064826, "全帯域_40ms_modulation_mae_db": 0.477585332285043, "300–1000 Hz_10ms_log_rms_mae_db": 1.3812910944844388, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09118691765547768, "300–1000 Hz_10ms_envelope_correlation": 0.9934742913206698, "300–1000 Hz_10ms_modulation_mae_db": 1.0036234415512562, "300–1000 Hz_40ms_log_rms_mae_db": 0.7628605157088394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04460921059715092, "300–1000 Hz_40ms_envelope_correlation": 0.998394802571281, "300–1000 Hz_40ms_modulation_mae_db": 1.0411901093960436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.389288062648354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1907136560990331, "1000–3000 Hz_10ms_envelope_correlation": 0.9739119594585082, "1000–3000 Hz_10ms_modulation_mae_db": 0.5195152990366303, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8759537710662711, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08912415544040393, "1000–3000 Hz_40ms_envelope_correlation": 0.9943268703829089, "1000–3000 Hz_40ms_modulation_mae_db": 0.3695671673406038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.42883578741075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37521601384596337, "3000–6000 Hz_10ms_envelope_correlation": 0.9077825487771978, "3000–6000 Hz_10ms_modulation_mae_db": 0.6040741884277382, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9678968245534866, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.28820529272699236, "3000–6000 Hz_40ms_envelope_correlation": 0.9443666974896768, "3000–6000 Hz_40ms_modulation_mae_db": 0.5261982907780705, "mr_stft_db": 5.661662404689648, "band_mae_db": 0.29762376904057825, "rms_gain_error_db": -0.042315008002587716, "output_peak": 0.44924559761615834}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7386996564454288, "全帯域_10ms_rms_envelope_nrmse": 0.0836162066184024, "全帯域_10ms_envelope_correlation": 0.9933578743638016, "全帯域_10ms_modulation_mae_db": 0.6017054952607049, "全帯域_40ms_log_rms_mae_db": 0.3951238000351985, "全帯域_40ms_rms_envelope_nrmse": 0.043514814543297975, "全帯域_40ms_envelope_correlation": 0.9981258126714536, "全帯域_40ms_modulation_mae_db": 0.5951239776461376, "300–1000 Hz_10ms_log_rms_mae_db": 1.4111391663892934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10200838850460633, "300–1000 Hz_10ms_envelope_correlation": 0.9919441201802319, "300–1000 Hz_10ms_modulation_mae_db": 0.7202934167628884, "300–1000 Hz_40ms_log_rms_mae_db": 0.7965773137197056, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05438213298076117, "300–1000 Hz_40ms_envelope_correlation": 0.9977843433533808, "300–1000 Hz_40ms_modulation_mae_db": 0.7516009330432243, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3775507626576389, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21450594121480632, "1000–3000 Hz_10ms_envelope_correlation": 0.9743212191472593, "1000–3000 Hz_10ms_modulation_mae_db": 0.3406162795421872, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8800625203742743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16885177357285822, "1000–3000 Hz_40ms_envelope_correlation": 0.9856353394764891, "1000–3000 Hz_40ms_modulation_mae_db": 0.5806088040964069, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4556935789031307, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36160274250430585, "3000–6000 Hz_10ms_envelope_correlation": 0.9051854298295228, "3000–6000 Hz_10ms_modulation_mae_db": 0.4845354159974478, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9785724200038044, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2810181059716329, "3000–6000 Hz_40ms_envelope_correlation": 0.9388513912622442, "3000–6000 Hz_40ms_modulation_mae_db": 0.5629911495690181, "mr_stft_db": 5.642400165088099, "band_mae_db": 0.42235672792682494, "rms_gain_error_db": -0.08044597217203675, "output_peak": 0.42974765208523485}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7390922883589565, "全帯域_10ms_rms_envelope_nrmse": 0.0873302867049242, "全帯域_10ms_envelope_correlation": 0.9927606327931591, "全帯域_10ms_modulation_mae_db": 0.6027396034979421, "全帯域_40ms_log_rms_mae_db": 0.41335332875081143, "全帯域_40ms_rms_envelope_nrmse": 0.046877256669264386, "全帯域_40ms_envelope_correlation": 0.9978236148649865, "全帯域_40ms_modulation_mae_db": 0.6972232417999255, "300–1000 Hz_10ms_log_rms_mae_db": 1.3757842172027754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09970223764763131, "300–1000 Hz_10ms_envelope_correlation": 0.9922305338189303, "300–1000 Hz_10ms_modulation_mae_db": 0.802390452823006, "300–1000 Hz_40ms_log_rms_mae_db": 0.7685177601335265, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05581424408529641, "300–1000 Hz_40ms_envelope_correlation": 0.9975138644025804, "300–1000 Hz_40ms_modulation_mae_db": 0.8964591238942227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3926612828157685, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18191424442097115, "1000–3000 Hz_10ms_envelope_correlation": 0.9760880559232749, "1000–3000 Hz_10ms_modulation_mae_db": 0.5976343864203758, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8952367468553757, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09625398944325793, "1000–3000 Hz_40ms_envelope_correlation": 0.993174439936346, "1000–3000 Hz_40ms_modulation_mae_db": 0.6102292345973455, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4742242201030746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3559818204607564, "3000–6000 Hz_10ms_envelope_correlation": 0.9144013548241342, "3000–6000 Hz_10ms_modulation_mae_db": 0.37692084430637074, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0021369437015726, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2738937483398146, "3000–6000 Hz_40ms_envelope_correlation": 0.9473134827713486, "3000–6000 Hz_40ms_modulation_mae_db": 0.31854029333275435, "mr_stft_db": 5.642490233631235, "band_mae_db": 0.305754448768161, "rms_gain_error_db": -0.08569507610080151, "output_peak": 0.43844044413308236}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7249790021490575, "全帯域_10ms_rms_envelope_nrmse": 0.08270867434440628, "全帯域_10ms_envelope_correlation": 0.9935579622335988, "全帯域_10ms_modulation_mae_db": 0.5774187228847326, "全帯域_40ms_log_rms_mae_db": 0.38115087686320065, "全帯域_40ms_rms_envelope_nrmse": 0.04436473939535938, "全帯域_40ms_envelope_correlation": 0.9981265587450665, "全帯域_40ms_modulation_mae_db": 0.6277942269222668, "300–1000 Hz_10ms_log_rms_mae_db": 1.4198883244506573, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09774317189397466, "300–1000 Hz_10ms_envelope_correlation": 0.9928103142172006, "300–1000 Hz_10ms_modulation_mae_db": 0.8017360244883757, "300–1000 Hz_40ms_log_rms_mae_db": 0.8058050435630922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054052075258394686, "300–1000 Hz_40ms_envelope_correlation": 0.9980394141241855, "300–1000 Hz_40ms_modulation_mae_db": 0.9493342135620969, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3741903747896238, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17019894097712845, "1000–3000 Hz_10ms_envelope_correlation": 0.9810891863065554, "1000–3000 Hz_10ms_modulation_mae_db": 0.4302772227577214, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8724908895098741, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11521408339165307, "1000–3000 Hz_40ms_envelope_correlation": 0.9918791650227462, "1000–3000 Hz_40ms_modulation_mae_db": 0.5972601162337251, "3000–6000 Hz_10ms_log_rms_mae_db": 1.440406389289659, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.350132700786853, "3000–6000 Hz_10ms_envelope_correlation": 0.9126313408512526, "3000–6000 Hz_10ms_modulation_mae_db": 0.39535863887242834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0100456424943607, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2623977623745907, "3000–6000 Hz_40ms_envelope_correlation": 0.9481102723293139, "3000–6000 Hz_40ms_modulation_mae_db": 0.4523471442085159, "mr_stft_db": 5.648348989228025, "band_mae_db": 0.36867025720184055, "rms_gain_error_db": -0.11266794027108479, "output_peak": 0.42405509397704233}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.781705613792275, "全帯域_10ms_rms_envelope_nrmse": 0.083595066030451, "全帯域_10ms_envelope_correlation": 0.9935885317962372, "全帯域_10ms_modulation_mae_db": 0.3521047031639765, "全帯域_40ms_log_rms_mae_db": 0.43619531773347325, "全帯域_40ms_rms_envelope_nrmse": 0.04726796765968978, "全帯域_40ms_envelope_correlation": 0.9980747676918156, "全帯域_40ms_modulation_mae_db": 0.29822307770681894, "300–1000 Hz_10ms_log_rms_mae_db": 1.4428161803751416, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11985106099068117, "300–1000 Hz_10ms_envelope_correlation": 0.9893853345861898, "300–1000 Hz_10ms_modulation_mae_db": 0.44555326377025845, "300–1000 Hz_40ms_log_rms_mae_db": 0.867923296981796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07895924770459542, "300–1000 Hz_40ms_envelope_correlation": 0.9957861392344195, "300–1000 Hz_40ms_modulation_mae_db": 0.42040728758326235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4457234967156636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2406813576327876, "1000–3000 Hz_10ms_envelope_correlation": 0.966602864631153, "1000–3000 Hz_10ms_modulation_mae_db": 0.540505397929937, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9409550726738168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17747143516442376, "1000–3000 Hz_40ms_envelope_correlation": 0.9838404994977703, "1000–3000 Hz_40ms_modulation_mae_db": 0.689032317390765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.432033035291137, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3608570049423787, "3000–6000 Hz_10ms_envelope_correlation": 0.9183266342598986, "3000–6000 Hz_10ms_modulation_mae_db": 0.6510537259126533, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9701056928026077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2709020133722009, "3000–6000 Hz_40ms_envelope_correlation": 0.9545715419845859, "3000–6000 Hz_40ms_modulation_mae_db": 0.6149161364364356, "mr_stft_db": 5.655144094425509, "band_mae_db": 0.5382107541532518, "rms_gain_error_db": -0.17249514258929097, "output_peak": 0.45673692138151284}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7614860298718742, "全帯域_10ms_rms_envelope_nrmse": 0.08120429920013139, "全帯域_10ms_envelope_correlation": 0.993764896381067, "全帯域_10ms_modulation_mae_db": 0.7284362188667676, "全帯域_40ms_log_rms_mae_db": 0.39580524830374525, "全帯域_40ms_rms_envelope_nrmse": 0.04270607638900658, "全帯域_40ms_envelope_correlation": 0.9982339149931377, "全帯域_40ms_modulation_mae_db": 0.7631685125503815, "300–1000 Hz_10ms_log_rms_mae_db": 1.3726494114689813, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09894106748346988, "300–1000 Hz_10ms_envelope_correlation": 0.9924590403568351, "300–1000 Hz_10ms_modulation_mae_db": 0.5544562188400032, "300–1000 Hz_40ms_log_rms_mae_db": 0.7619416525564056, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05436138131449005, "300–1000 Hz_40ms_envelope_correlation": 0.9978080657809929, "300–1000 Hz_40ms_modulation_mae_db": 0.5811408582644714, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4647271272120497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2021696486482451, "1000–3000 Hz_10ms_envelope_correlation": 0.969188670246087, "1000–3000 Hz_10ms_modulation_mae_db": 0.24503159320433973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8819599912167909, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09009362246707996, "1000–3000 Hz_40ms_envelope_correlation": 0.994475848569786, "1000–3000 Hz_40ms_modulation_mae_db": 0.4403085264673521, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4267981632337774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3521985252489963, "3000–6000 Hz_10ms_envelope_correlation": 0.9167026048823694, "3000–6000 Hz_10ms_modulation_mae_db": 0.3357701325330355, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9765049362339688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27955689240737935, "3000–6000 Hz_40ms_envelope_correlation": 0.9455114302838237, "3000–6000 Hz_40ms_modulation_mae_db": 0.24741807325359053, "mr_stft_db": 5.647323027047039, "band_mae_db": 0.3478108104076869, "rms_gain_error_db": -0.09736636383542757, "output_peak": 0.4191043964533226}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.403932672249502, "spectral_convergence": 0.26215348402798544, "spectral_convergence_status": "ok", "amplitude_floor": 1.3235749190080935e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.501946007295421, "spectral_convergence": 0.2942510372697704, "spectral_convergence_status": "ok", "amplitude_floor": 1.1987597823804542e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.026702664802624, "spectral_convergence": 0.2954814181783468, "spectral_convergence_status": "ok", "amplitude_floor": 9.377214849682438e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.350500895926799e-05, "output_power": 5.147227041812338e-05, "delta_db": 0.7303406277309281, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012608807022653607, "output_power": 0.0012435475554170568, "delta_db": -0.060115998158912574, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00036191686574782775, "output_power": 0.0003496907288978534, "delta_db": -0.14924703888434532, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.9809147109830135e-05, "output_power": 3.536524813701356e-05, "delta_db": 0.742271593065357, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.6488056661262093e-05, "output_power": 1.852233547024457e-05, "delta_db": 0.5052627450285059, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.109371525887873e-05, "output_power": 6.328553497709821e-05, "delta_db": 0.1530791926305427, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3900528659164319, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.029013423575384854, "pitch": null, "output_rms_dbfs": -27.540227688547812, "output_peak": 0.48123037128047985, "reference_rms_dbfs": -27.511214264972427, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.644193781449182}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018136510927426692, "log_rms_mae_db": 0.7658343517362223, "rms_envelope_nrmse": 0.07811514613296058, "envelope_correlation": 0.9941773191656993, "rms_gain_error_db": -0.02887449093091861, "reference_mean_power": 0.0017749427149271913, "output_mean_power": 0.0017631809785443233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.578419076835135e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003061888162786739, "output_power": 0.00030385244310343477, "reference_active": true, "delta_db": -0.03326591187513715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.561280729204581e-05, "output_power": 5.27828400092953e-05, "reference_active": true, "delta_db": -0.22682064521532952}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1447257663776473e-05, "output_power": 8.830063214479531e-06, "reference_active": true, "delta_db": -1.127376456614426}], "active_band_count": 3, "active_band_mae_db": 0.4624876712349642}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015531555176710683, "log_rms_mae_db": 0.4203790910690272, "rms_envelope_nrmse": 0.03533583422094867, "envelope_correlation": 0.9987245867692061, "rms_gain_error_db": -0.028881014629345537, "reference_mean_power": 0.0017815693662331493, "output_mean_power": 0.0017697610596266945}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.067921014076438e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002861407755621019, "output_power": 0.00028377220010734785, "reference_active": true, "delta_db": -0.036099026219196625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5605898871779113e-05, "output_power": 2.4339436835272228e-05, "reference_active": true, "delta_db": -0.22029500625296236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.159270454376389e-07, "output_power": 7.062088094895519e-07, "reference_active": true, "delta_db": -1.1292775223013383}], "active_band_count": 3, "active_band_mae_db": 0.4618905182578324}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017559797263300874, "log_rms_mae_db": 1.4160035814085463, "rms_envelope_nrmse": 0.09744593557973465, "envelope_correlation": 0.9925654356062091, "rms_gain_error_db": -0.10610478882890809, "reference_mean_power": 0.0012870696988326355, "output_mean_power": 0.0012560056327247887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.703155295091611e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025376682215949657, "output_power": 0.0002484826175312222, "reference_active": true, "delta_db": -0.0913882788945513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.494529209431561e-05, "output_power": 4.079626330367414e-05, "reference_active": true, "delta_db": -0.42063821096227216}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.800066525463509e-06, "output_power": 6.205647753620021e-06, "reference_active": true, "delta_db": -1.5169883485261615}], "active_band_count": 3, "active_band_mae_db": 0.676338279460995}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014992083957937532, "log_rms_mae_db": 0.8263704310701557, "rms_envelope_nrmse": 0.04417447948295582, "envelope_correlation": 0.998479148915275, "rms_gain_error_db": -0.10631768408984452, "reference_mean_power": 0.0012919551058812144, "output_mean_power": 0.0012607113252731692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4067631484646013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002445691472192921, "output_power": 0.00023811395566107374, "reference_active": true, "delta_db": -0.11616819437566143}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8830927140807284e-05, "output_power": 1.7153326801967666e-05, "reference_active": true, "delta_db": -0.40523341243322347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.050384653726169e-07, "output_power": 4.826348954001668e-07, "reference_active": true, "delta_db": -1.6459409279830008}], "active_band_count": 3, "active_band_mae_db": 0.7224475115972954}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.2493337902299445e-05, "log_rms_mae_db": 1.401274285078899, "rms_envelope_nrmse": 0.18444774146282117, "envelope_correlation": 0.9771225759951963, "rms_gain_error_db": 0.334334107645825, "reference_mean_power": 4.199153922743893e-05, "output_mean_power": 4.535187062167402e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0299126624198213e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.222152478948813e-06, "output_power": 3.8105216270740586e-06, "reference_active": true, "delta_db": -0.44549482519307004}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.960031981806193e-07, "output_power": 7.926388627936397e-07, "reference_active": true, "delta_db": -0.5323419800936815}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.633269352164066e-07, "output_power": 2.076312035572407e-07, "reference_active": true, "delta_db": -1.0320266313933748}], "active_band_count": 3, "active_band_mae_db": 0.6699544788933754}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.820326553727612e-05, "log_rms_mae_db": 0.8774579310817295, "rms_envelope_nrmse": 0.11560715825160295, "envelope_correlation": 0.9914321486169919, "rms_gain_error_db": 0.33397737223540647, "reference_mean_power": 4.2190507757870256e-05, "output_mean_power": 4.5563018649752706e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.548028980657037e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.958931660389438e-06, "output_power": 3.546165394723606e-06, "reference_active": true, "delta_db": -0.4781901754303317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7410748918107156e-07, "output_power": 3.203789911200608e-07, "reference_active": true, "delta_db": -0.6733237288950632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8002833142836267e-08, "output_power": 1.2773167510857412e-08, "reference_active": true, "delta_db": -1.4904224862539583}], "active_band_count": 3, "active_band_mae_db": 0.880645463526451}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.219318962627022e-05, "log_rms_mae_db": 1.4100385275228922, "rms_envelope_nrmse": 0.37119469648601544, "envelope_correlation": 0.9054949760539067, "rms_gain_error_db": 0.7237037548012675, "reference_mean_power": 2.2095607497389627e-05, "output_mean_power": 2.610215246234528e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.718240736767725e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.504526854650138e-06, "output_power": 9.824491330414111e-06, "reference_active": true, "delta_db": 0.1437957164014528}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5665611692910136e-06, "output_power": 2.85916770400348e-06, "reference_active": true, "delta_db": 0.46888010240546507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.91578688150447e-07, "output_power": 8.678660092322984e-07, "reference_active": true, "delta_db": 1.6644015917821278}], "active_band_count": 3, "active_band_mae_db": 0.7590258035296819}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.1857789632782756e-05, "log_rms_mae_db": 0.9800544480945683, "rms_envelope_nrmse": 0.28548310363535, "envelope_correlation": 0.9415588465499863, "rms_gain_error_db": 0.7236176127848599, "reference_mean_power": 2.220381322589703e-05, "output_mean_power": 2.6229458618495542e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6122764663239433e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.855212413255228e-06, "output_power": 1.0417692208906067e-05, "reference_active": true, "delta_db": 0.2410553280451445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3142139667203495e-06, "output_power": 1.5560094890148127e-06, "reference_active": true, "delta_db": 0.7334616280602807}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2621117296192714e-08, "output_power": 5.824255635484824e-08, "reference_active": true, "delta_db": 1.3561559818197226}], "active_band_count": 3, "active_band_mae_db": 0.7768909793083827}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8713903990893059, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a20661bed246831698575712622c2770fce5d80fc1d628dd293961f018bb593f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9739728572031023, "standard_deviation": 0.0045139779159222935, "minimum": 0.966602864631153, "maximum": 0.9810891863065554, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4094463641404276, "standard_deviation": 0.033415256202105094, "minimum": 1.3741903747896238, "maximum": 1.4647271272120497, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4710732311926218, "standard_deviation": 0.13868521250395616, "minimum": 0.24503159320433973, "maximum": 0.6699544788933754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1965293581576666, "standard_deviation": 0.02226696347441279, "minimum": 0.17019894097712845, "maximum": 0.2406813576327876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9909684741095031, "standard_deviation": 0.0040142912504251204, "minimum": 0.9838404994977703, "maximum": 0.994475848569786, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8908305725402557, "standard_deviation": 0.022677334748637673, "minimum": 0.8724908895098741, "maximum": 0.9409550726738168, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5559664634092448, "standard_deviation": 0.18996788731773756, "minimum": 0.28008007762130915, "maximum": 0.880645463526451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11850497161675733, "standard_deviation": 0.0353136288324027, "minimum": 0.08912415544040393, "maximum": 0.17747143516442376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9102146647713756, "standard_deviation": 0.006166160075047021, "minimum": 0.9011924286927226, "maximum": 0.9183266342598986, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.440033294732094, "standard_deviation": 0.020098216670601878, "minimum": 1.4100385275228922, "maximum": 1.4742242201030746, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5124069093438242, "standard_deviation": 0.14778192362611617, "minimum": 0.3357701325330355, "maximum": 0.7590258035296819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.362937893745661, "standard_deviation": 0.010224655039966857, "minimum": 0.350132700786853, "maximum": 0.37631964569001874, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9456930854084322, "standard_deviation": 0.0046778377884382945, "minimum": 0.9388513912622442, "maximum": 0.9545715419845859, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.98460846145515, "standard_deviation": 0.015189012803660766, "minimum": 0.9678968245534866, "maximum": 1.0100456424943607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5011455057430807, "standard_deviation": 0.16630925534377444, "minimum": 0.24741807325359053, "maximum": 0.7768909793083827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.27695610277657157, "standard_deviation": 0.008350655077790177, "minimum": 0.2623977623745907, "maximum": 0.28820529272699236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991754784751165, "standard_deviation": 0.0015939915200387368, "minimum": 0.9891692079230536, "maximum": 0.9934742913206698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3989253668424908, "standard_deviation": 0.026932997695093924, "minimum": 1.3718309589600917, "maximum": 1.4428161803751416, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7143326120386841, "standard_deviation": 0.1683453635422717, "minimum": 0.44555326377025845, "maximum": 1.0036234415512562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10309091431805759, "standard_deviation": 0.01021717999566016, "minimum": 0.09118691765547768, "maximum": 0.11985106099068117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974980085976826, "standard_deviation": 0.0009935882298668696, "minimum": 0.9957861392344195, "maximum": 0.998479148915275, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8032494511191852, "standard_deviation": 0.038516781399800096, "minimum": 0.7619416525564056, "maximum": 0.867923296981796, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7670143281239734, "standard_deviation": 0.20075590360715004, "minimum": 0.42040728758326235, "maximum": 1.0411901093960436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05681095324321246, "standard_deviation": 0.011636427334620609, "minimum": 0.04417447948295582, "maximum": 0.07895924770459542, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.37402452929098456, "standard_deviation": 0.07883060695874754, "minimum": 0.29762376904057825, "maximum": 0.5382107541532518, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.648981613670693, "standard_deviation": 0.0066763870164407635, "minimum": 5.642400165088099, "maximum": 5.661662404689648, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44891240737607657, "standard_deviation": 0.026752541534904722, "minimum": 0.4191043964533226, "maximum": 0.49273878208177924, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07199576828878236, "standard_deviation": 0.06419629213615988, "minimum": -0.17249514258929097, "maximum": 0.04403278023635524, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993392090194736, "standard_deviation": 0.0006587003910719288, "minimum": 0.9921131693930203, "maximum": 0.9941773191656993, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7576428959358625, "standard_deviation": 0.020886835616584602, "minimum": 0.7249790021490575, "maximum": 0.781705613792275, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5766209495938528, "standard_deviation": 0.12985337898933944, "minimum": 0.3521047031639765, "maximum": 0.746870980705992, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08358016344805047, "standard_deviation": 0.004201184146664128, "minimum": 0.07811514613296058, "maximum": 0.09156544408247037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980740215379762, "standard_deviation": 0.0004174133868488137, "minimum": 0.9972507988616607, "maximum": 0.9987245867692061, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40980617337467024, "standard_deviation": 0.017910542628805586, "minimum": 0.38115087686320065, "maximum": 0.43619531773347325, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.594157222380405, "standard_deviation": 0.17552694172543237, "minimum": 0.29822307770681894, "maximum": 0.8322488918748352, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04428937417526757, "standard_deviation": 0.004984448848980933, "minimum": 0.03533583422094867, "maximum": 0.05255525213293205, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0032", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6625335216522217, "predicted_mos": 2.6625335216522217, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "835a7930f7a80e6974738ebfbfe57496bff330da1b333584f8cccb4a1bcc88c3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9421239998191595, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7293748758387533, "全帯域_10ms_rms_envelope_nrmse": 0.07816974023468004, "全帯域_10ms_envelope_correlation": 0.9942972618721555, "全帯域_10ms_modulation_mae_db": 0.6779878170089036, "全帯域_40ms_log_rms_mae_db": 0.36628396572252864, "全帯域_40ms_rms_envelope_nrmse": 0.04354728217983628, "全帯域_40ms_envelope_correlation": 0.9982177745124796, "全帯域_40ms_modulation_mae_db": 0.7758065999855219, "300–1000 Hz_10ms_log_rms_mae_db": 1.3106343917247385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09245221864305184, "300–1000 Hz_10ms_envelope_correlation": 0.9936280161156958, "300–1000 Hz_10ms_modulation_mae_db": 0.701119041920426, "300–1000 Hz_40ms_log_rms_mae_db": 0.7388846060961988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05724334533419248, "300–1000 Hz_40ms_envelope_correlation": 0.997726244977143, "300–1000 Hz_40ms_modulation_mae_db": 0.7564853866349162, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3768154775853165, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16154261506353557, "1000–3000 Hz_10ms_envelope_correlation": 0.9809969944440078, "1000–3000 Hz_10ms_modulation_mae_db": 0.3395803833393323, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8656901198825516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08130388454005479, "1000–3000 Hz_40ms_envelope_correlation": 0.9950807595331992, "1000–3000 Hz_40ms_modulation_mae_db": 0.34823183724792234, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4553802080097153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3629341119725442, "3000–6000 Hz_10ms_envelope_correlation": 0.9108599743137182, "3000–6000 Hz_10ms_modulation_mae_db": 0.4268753984693067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9781518925795701, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2677253541916235, "3000–6000 Hz_40ms_envelope_correlation": 0.9502756433695825, "3000–6000 Hz_40ms_modulation_mae_db": 0.35602350182439935, "mr_stft_db": 5.634630094501332, "band_mae_db": 0.37575592570019606, "rms_gain_error_db": -0.10851008532271715, "output_peak": 0.3568691453840639}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7794592372866813, "全帯域_10ms_rms_envelope_nrmse": 0.08806040378372364, "全帯域_10ms_envelope_correlation": 0.9926924758639841, "全帯域_10ms_modulation_mae_db": 0.573211076021968, "全帯域_40ms_log_rms_mae_db": 0.4206727338579678, "全帯域_40ms_rms_envelope_nrmse": 0.04254182377114618, "全帯域_40ms_envelope_correlation": 0.9983198488459665, "全帯域_40ms_modulation_mae_db": 0.7506378731557611, "300–1000 Hz_10ms_log_rms_mae_db": 1.4400720243411655, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10567535780798984, "300–1000 Hz_10ms_envelope_correlation": 0.9913849578996041, "300–1000 Hz_10ms_modulation_mae_db": 0.7171779215678239, "300–1000 Hz_40ms_log_rms_mae_db": 0.8522167124499503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053572084894741155, "300–1000 Hz_40ms_envelope_correlation": 0.9979214182940975, "300–1000 Hz_40ms_modulation_mae_db": 0.923733347319963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4207313458462252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1722078694217775, "1000–3000 Hz_10ms_envelope_correlation": 0.9776985822292413, "1000–3000 Hz_10ms_modulation_mae_db": 0.6555386392118098, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9312137258311539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0912252765892619, "1000–3000 Hz_40ms_envelope_correlation": 0.9935216881057951, "1000–3000 Hz_40ms_modulation_mae_db": 0.7607501517295233, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4192301643199756, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3497396751025631, "3000–6000 Hz_10ms_envelope_correlation": 0.9207206614338871, "3000–6000 Hz_10ms_modulation_mae_db": 0.8224341967323469, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9608017844208312, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2692507481689999, "3000–6000 Hz_40ms_envelope_correlation": 0.9522929581250351, "3000–6000 Hz_40ms_modulation_mae_db": 0.6838296776451708, "mr_stft_db": 5.628020722779255, "band_mae_db": 0.32131381072517723, "rms_gain_error_db": -0.12240652846583959, "output_peak": 0.47655654386178714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7675715333502899, "全帯域_10ms_rms_envelope_nrmse": 0.0820657061743243, "全帯域_10ms_envelope_correlation": 0.9936040893932616, "全帯域_10ms_modulation_mae_db": 0.1927035490728076, "全帯域_40ms_log_rms_mae_db": 0.3996427407223809, "全帯域_40ms_rms_envelope_nrmse": 0.04530884301863831, "全帯域_40ms_envelope_correlation": 0.9979074611600725, "全帯域_40ms_modulation_mae_db": 0.3138629655614903, "300–1000 Hz_10ms_log_rms_mae_db": 1.3900772659759386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10172944304758702, "300–1000 Hz_10ms_envelope_correlation": 0.9918559960765021, "300–1000 Hz_10ms_modulation_mae_db": 0.22452341553336597, "300–1000 Hz_40ms_log_rms_mae_db": 0.7797500960096513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05918403095567815, "300–1000 Hz_40ms_envelope_correlation": 0.9970896222547669, "300–1000 Hz_40ms_modulation_mae_db": 0.3458081762813303, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3292396913046205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22166862225844347, "1000–3000 Hz_10ms_envelope_correlation": 0.9682252092489916, "1000–3000 Hz_10ms_modulation_mae_db": 0.5162548590120646, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8396513457127537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1393023586980334, "1000–3000 Hz_40ms_envelope_correlation": 0.9884412510233641, "1000–3000 Hz_40ms_modulation_mae_db": 0.5534836286046879, "3000–6000 Hz_10ms_log_rms_mae_db": 1.389370194105255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35028211487322236, "3000–6000 Hz_10ms_envelope_correlation": 0.9146730206625797, "3000–6000 Hz_10ms_modulation_mae_db": 0.8565907505618661, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9243613907038644, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2622550215176987, "3000–6000 Hz_40ms_envelope_correlation": 0.9501571087261873, "3000–6000 Hz_40ms_modulation_mae_db": 0.6485641569892612, "mr_stft_db": 5.630140454685432, "band_mae_db": 0.3206520249987366, "rms_gain_error_db": 0.00839479276768649, "output_peak": 0.35369893228382365}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7457177216794737, "全帯域_10ms_rms_envelope_nrmse": 0.09339244018946222, "全帯域_10ms_envelope_correlation": 0.9917112451216292, "全帯域_10ms_modulation_mae_db": 0.3722745459395261, "全帯域_40ms_log_rms_mae_db": 0.39186132058476564, "全帯域_40ms_rms_envelope_nrmse": 0.04885290772721837, "全帯域_40ms_envelope_correlation": 0.9976490063459693, "全帯域_40ms_modulation_mae_db": 0.4707794992680574, "300–1000 Hz_10ms_log_rms_mae_db": 1.4138872091272636, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10736174199608245, "300–1000 Hz_10ms_envelope_correlation": 0.991090822894649, "300–1000 Hz_10ms_modulation_mae_db": 0.4442252973643714, "300–1000 Hz_40ms_log_rms_mae_db": 0.8030212614258605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06257659349086837, "300–1000 Hz_40ms_envelope_correlation": 0.9970240597690696, "300–1000 Hz_40ms_modulation_mae_db": 0.5452115735172846, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3741308723994463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18466592081093752, "1000–3000 Hz_10ms_envelope_correlation": 0.9742980297470181, "1000–3000 Hz_10ms_modulation_mae_db": 0.5253745378544725, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8492848211735907, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07825010509661307, "1000–3000 Hz_40ms_envelope_correlation": 0.9953269421290264, "1000–3000 Hz_40ms_modulation_mae_db": 0.512056556912261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4264699861281682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3522652694011786, "3000–6000 Hz_10ms_envelope_correlation": 0.9087890587232846, "3000–6000 Hz_10ms_modulation_mae_db": 0.5667034539508043, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9697006093691973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2696260477756395, "3000–6000 Hz_40ms_envelope_correlation": 0.9428198094160283, "3000–6000 Hz_40ms_modulation_mae_db": 0.44382305554311013, "mr_stft_db": 5.639266599914062, "band_mae_db": 0.3178908912786122, "rms_gain_error_db": -0.0978671332243395, "output_peak": 0.38573926639330447}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7632354853005281, "全帯域_10ms_rms_envelope_nrmse": 0.08754779439516522, "全帯域_10ms_envelope_correlation": 0.9926991551154468, "全帯域_10ms_modulation_mae_db": 0.5518140975647754, "全帯域_40ms_log_rms_mae_db": 0.4233704776624381, "全帯域_40ms_rms_envelope_nrmse": 0.047289984804894714, "全帯域_40ms_envelope_correlation": 0.9977563693772616, "全帯域_40ms_modulation_mae_db": 0.5178680272558638, "300–1000 Hz_10ms_log_rms_mae_db": 1.4132426734506762, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10868157246978045, "300–1000 Hz_10ms_envelope_correlation": 0.9908872382086753, "300–1000 Hz_10ms_modulation_mae_db": 0.9654308191698115, "300–1000 Hz_40ms_log_rms_mae_db": 0.8058483410359585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06613400194577904, "300–1000 Hz_40ms_envelope_correlation": 0.9966695243304001, "300–1000 Hz_40ms_modulation_mae_db": 0.943624616347555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4398800503106586, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16532662059744252, "1000–3000 Hz_10ms_envelope_correlation": 0.9798506493577609, "1000–3000 Hz_10ms_modulation_mae_db": 0.5143956403514457, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8983093236386225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0849383589855668, "1000–3000 Hz_40ms_envelope_correlation": 0.9944583065285764, "1000–3000 Hz_40ms_modulation_mae_db": 0.6100505026382841, "3000–6000 Hz_10ms_log_rms_mae_db": 1.379754268664331, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35135729675204064, "3000–6000 Hz_10ms_envelope_correlation": 0.9197132253976866, "3000–6000 Hz_10ms_modulation_mae_db": 0.8761650888346096, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9311673285941837, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26215497620043526, "3000–6000 Hz_40ms_envelope_correlation": 0.9549710255069565, "3000–6000 Hz_40ms_modulation_mae_db": 0.9075391126500435, "mr_stft_db": 5.62766649096314, "band_mae_db": 0.4162608055967232, "rms_gain_error_db": -0.0722181582421737, "output_peak": 0.3855497502524769}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7483204327419057, "全帯域_10ms_rms_envelope_nrmse": 0.08217069095219491, "全帯域_10ms_envelope_correlation": 0.9935599588736941, "全帯域_10ms_modulation_mae_db": 0.48378133049641964, "全帯域_40ms_log_rms_mae_db": 0.39303115419603546, "全帯域_40ms_rms_envelope_nrmse": 0.038265603616091624, "全帯域_40ms_envelope_correlation": 0.9985220024506711, "全帯域_40ms_modulation_mae_db": 0.46444420623347976, "300–1000 Hz_10ms_log_rms_mae_db": 1.318868038798228, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08858952939041524, "300–1000 Hz_10ms_envelope_correlation": 0.9939012316579812, "300–1000 Hz_10ms_modulation_mae_db": 0.6028094871210931, "300–1000 Hz_40ms_log_rms_mae_db": 0.7515326960214985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04633180160315972, "300–1000 Hz_40ms_envelope_correlation": 0.9983451391068904, "300–1000 Hz_40ms_modulation_mae_db": 0.6232735849368201, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4319294953016528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1976769465399707, "1000–3000 Hz_10ms_envelope_correlation": 0.9729740050044692, "1000–3000 Hz_10ms_modulation_mae_db": 0.5602913378939839, "1000–3000 Hz_40ms_log_rms_mae_db": 0.874616334010437, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12050672431646943, "1000–3000 Hz_40ms_envelope_correlation": 0.9901340981214375, "1000–3000 Hz_40ms_modulation_mae_db": 0.6108005849662006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3891131001121106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3642684374176955, "3000–6000 Hz_10ms_envelope_correlation": 0.909672676984263, "3000–6000 Hz_10ms_modulation_mae_db": 0.2581301300935673, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9443487357470249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27418448789751654, "3000–6000 Hz_40ms_envelope_correlation": 0.9469307502646258, "3000–6000 Hz_40ms_modulation_mae_db": 0.1633988693641556, "mr_stft_db": 5.62456604147291, "band_mae_db": 0.38875351972014655, "rms_gain_error_db": -0.05092428995989161, "output_peak": 0.41463357680500534}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7562238093734056, "全帯域_10ms_rms_envelope_nrmse": 0.08493290672857444, "全帯域_10ms_envelope_correlation": 0.9931301804599955, "全帯域_10ms_modulation_mae_db": 0.6579439388783404, "全帯域_40ms_log_rms_mae_db": 0.4075628318202111, "全帯域_40ms_rms_envelope_nrmse": 0.04134686014948754, "全帯域_40ms_envelope_correlation": 0.998294449204599, "全帯域_40ms_modulation_mae_db": 0.6428908630023963, "300–1000 Hz_10ms_log_rms_mae_db": 1.4017483190728524, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10564365647845067, "300–1000 Hz_10ms_envelope_correlation": 0.9914062276810739, "300–1000 Hz_10ms_modulation_mae_db": 0.94859547722002, "300–1000 Hz_40ms_log_rms_mae_db": 0.788342342980278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0569753514253321, "300–1000 Hz_40ms_envelope_correlation": 0.9976227194773236, "300–1000 Hz_40ms_modulation_mae_db": 1.0682596288075603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3785130022597272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18400864680506468, "1000–3000 Hz_10ms_envelope_correlation": 0.9764688545099861, "1000–3000 Hz_10ms_modulation_mae_db": 0.44706917503594656, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8301811169836881, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11172039091490166, "1000–3000 Hz_40ms_envelope_correlation": 0.9914526844449467, "1000–3000 Hz_40ms_modulation_mae_db": 0.5360174413105466, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4333550942812112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3855595254390967, "3000–6000 Hz_10ms_envelope_correlation": 0.9185581766187666, "3000–6000 Hz_10ms_modulation_mae_db": 1.3300359953792682, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9613900973935383, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30085638734852926, "3000–6000 Hz_40ms_envelope_correlation": 0.9534559648214311, "3000–6000 Hz_40ms_modulation_mae_db": 1.683041339501494, "mr_stft_db": 5.639026408084369, "band_mae_db": 0.4232944199664881, "rms_gain_error_db": -0.06940780978677812, "output_peak": 0.4478794354261001}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7267708774535585, "全帯域_10ms_rms_envelope_nrmse": 0.11316767738692088, "全帯域_10ms_envelope_correlation": 0.9878353735860282, "全帯域_10ms_modulation_mae_db": 0.5854399865302174, "全帯域_40ms_log_rms_mae_db": 0.4197335449258332, "全帯域_40ms_rms_envelope_nrmse": 0.05779837530265805, "全帯域_40ms_envelope_correlation": 0.9966052758356986, "全帯域_40ms_modulation_mae_db": 0.6682176091027322, "300–1000 Hz_10ms_log_rms_mae_db": 1.3899459847467182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13663145665123877, "300–1000 Hz_10ms_envelope_correlation": 0.9852853388473987, "300–1000 Hz_10ms_modulation_mae_db": 0.7151806469038865, "300–1000 Hz_40ms_log_rms_mae_db": 0.7865495554267861, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06540369165937465, "300–1000 Hz_40ms_envelope_correlation": 0.9964502819830069, "300–1000 Hz_40ms_modulation_mae_db": 0.8438359922878043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3706146162958848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23298256384242266, "1000–3000 Hz_10ms_envelope_correlation": 0.9661387707460632, "1000–3000 Hz_10ms_modulation_mae_db": 0.18951010232870577, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8387343654567406, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14267010018018889, "1000–3000 Hz_40ms_envelope_correlation": 0.9889973495546336, "1000–3000 Hz_40ms_modulation_mae_db": 0.15932426195587643, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4365745841175768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3554710743073817, "3000–6000 Hz_10ms_envelope_correlation": 0.9096900095120884, "3000–6000 Hz_10ms_modulation_mae_db": 0.6021243720597297, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9324728777128789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26974856463864355, "3000–6000 Hz_40ms_envelope_correlation": 0.9450026364216837, "3000–6000 Hz_40ms_modulation_mae_db": 0.7410829146625305, "mr_stft_db": 5.6387826880635785, "band_mae_db": 0.41108661624722426, "rms_gain_error_db": 0.005082486154130211, "output_peak": 0.46834921216685055}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.41819805607502, "spectral_convergence": 0.2575839471696723, "spectral_convergence_status": "ok", "amplitude_floor": 1.3235749190080935e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.472802417426859, "spectral_convergence": 0.289109639734174, "spectral_convergence_status": "ok", "amplitude_floor": 1.1987597823804542e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.012889810002115, "spectral_convergence": 0.3023564389725072, "spectral_convergence_status": "ok", "amplitude_floor": 9.377214849682438e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.350500895926799e-05, "output_power": 5.080952884133208e-05, "delta_db": 0.6740590527093498, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012608807022653607, "output_power": 0.0012288797193738697, "delta_db": -0.11164620942008784, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00036191686574782775, "output_power": 0.00033423032656670524, "delta_db": -0.3456296892251847, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.9809147109830135e-05, "output_power": 3.1241545249336954e-05, "delta_db": 0.2038295622257238, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.6488056661262093e-05, "output_power": 1.8899168686542768e-05, "delta_db": 0.5927323015208971, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.109371525887873e-05, "output_power": 6.586586661981177e-05, "delta_db": 0.3266387390999329, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.37575592570019606, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10851008532271715, "pitch": null, "output_rms_dbfs": -27.619724350295144, "output_peak": 0.3568691453840639, "reference_rms_dbfs": -27.511214264972427, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.634630094501332}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018136510927426692, "log_rms_mae_db": 0.7293748758387533, "rms_envelope_nrmse": 0.07816974023468004, "envelope_correlation": 0.9942972618721555, "rms_gain_error_db": -0.10833750496898634, "reference_mean_power": 0.0017749427149271913, "output_mean_power": 0.0017312133425149499}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.578419076835135e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003061888162786739, "output_power": 0.0003040322504991489, "reference_active": true, "delta_db": -0.030696695612142732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.561280729204581e-05, "output_power": 5.0907518650810226e-05, "reference_active": true, "delta_db": -0.38392889439721023}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1447257663776473e-05, "output_power": 7.884382263010758e-06, "reference_active": true, "delta_db": -1.6193378610173583}], "active_band_count": 3, "active_band_mae_db": 0.6779878170089036}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015531555176710683, "log_rms_mae_db": 0.36628396572252864, "rms_envelope_nrmse": 0.04354728217983628, "envelope_correlation": 0.9982177745124796, "rms_gain_error_db": -0.10859174448954136, "reference_mean_power": 0.0017815693662331493, "output_mean_power": 0.0017375750105754933}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.067921014076438e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002861407755621019, "output_power": 0.00028317967692459813, "reference_active": true, "delta_db": -0.04517667999481367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5605898871779113e-05, "output_power": 2.3351134150119467e-05, "reference_active": true, "delta_db": -0.4003204712226835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.159270454376389e-07, "output_power": 5.938388659203984e-07, "reference_active": true, "delta_db": -1.8819226487390686}], "active_band_count": 3, "active_band_mae_db": 0.7758065999855219}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017559797263300874, "log_rms_mae_db": 1.3106343917247385, "rms_envelope_nrmse": 0.09245221864305184, "envelope_correlation": 0.9936280161156958, "rms_gain_error_db": -0.21278777264277818, "reference_mean_power": 0.0012870696988326355, "output_mean_power": 0.0012255281425615269}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.703155295091611e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025376682215949657, "output_power": 0.00024925898009022917, "reference_active": true, "delta_db": -0.07784027525666443}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.494529209431561e-05, "output_power": 3.911979252514472e-05, "reference_active": true, "delta_db": -0.6028766439575253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.800066525463509e-06, "output_power": 6.341937162648098e-06, "reference_active": true, "delta_db": -1.422640206547088}], "active_band_count": 3, "active_band_mae_db": 0.701119041920426}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014992083957937532, "log_rms_mae_db": 0.7388846060961988, "rms_envelope_nrmse": 0.05724334533419248, "envelope_correlation": 0.997726244977143, "rms_gain_error_db": -0.21354839185599012, "reference_mean_power": 0.0012919551058812144, "output_mean_power": 0.0012299645189851052}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4067631484646013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002445691472192921, "output_power": 0.00023839351932599835, "reference_active": true, "delta_db": -0.11107224195849638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8830927140807284e-05, "output_power": 1.5960609734373427e-05, "reference_active": true, "delta_db": -0.718222245901447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.050384653726169e-07, "output_power": 5.060537466795428e-07, "reference_active": true, "delta_db": -1.440161672044805}], "active_band_count": 3, "active_band_mae_db": 0.7564853866349162}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.2493337902299445e-05, "log_rms_mae_db": 1.3768154775853165, "rms_envelope_nrmse": 0.16154261506353557, "envelope_correlation": 0.9809969944440078, "rms_gain_error_db": 0.12884775987783853, "reference_mean_power": 4.199153922743893e-05, "output_mean_power": 4.3256021284587124e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0299126624198213e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.222152478948813e-06, "output_power": 4.079896606082554e-06, "reference_active": true, "delta_db": -0.14884756166357388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.960031981806193e-07, "output_power": 8.392297974560707e-07, "reference_active": true, "delta_db": -0.28428664432795353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.633269352164066e-07, "output_power": 2.30109521928985e-07, "reference_active": true, "delta_db": -0.5856069440264697}], "active_band_count": 3, "active_band_mae_db": 0.3395803833393323}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.820326553727612e-05, "log_rms_mae_db": 0.8656901198825516, "rms_envelope_nrmse": 0.08130388454005479, "envelope_correlation": 0.9950807595331992, "rms_gain_error_db": 0.12812116493505468, "reference_mean_power": 4.2190507757870256e-05, "output_mean_power": 4.3453710694110304e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.548028980657037e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.958931660389438e-06, "output_power": 3.775770179858134e-06, "reference_active": true, "delta_db": -0.2057245277783506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7410748918107156e-07, "output_power": 3.530152937676673e-07, "reference_active": true, "delta_db": -0.25202881496202467}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8002833142836267e-08, "output_power": 1.572702725407405e-08, "reference_active": true, "delta_db": -0.5869421690033918}], "active_band_count": 3, "active_band_mae_db": 0.34823183724792234}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.219318962627022e-05, "log_rms_mae_db": 1.4553802080097153, "rms_envelope_nrmse": 0.3629341119725442, "envelope_correlation": 0.9108599743137182, "rms_gain_error_db": 0.7618170928281469, "reference_mean_power": 2.2095607497389627e-05, "output_mean_power": 2.6332230973275914e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.718240736767725e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.504526854650138e-06, "output_power": 9.634325009011579e-06, "reference_active": true, "delta_db": 0.058907908238971116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5665611692910136e-06, "output_power": 2.612336085887972e-06, "reference_active": true, "delta_db": 0.07677430319231492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.91578688150447e-07, "output_power": 4.5448350081413326e-07, "reference_active": true, "delta_db": -1.144943983976634}], "active_band_count": 3, "active_band_mae_db": 0.4268753984693067}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.1857789632782756e-05, "log_rms_mae_db": 0.9781518925795701, "rms_envelope_nrmse": 0.2677253541916235, "envelope_correlation": 0.9502756433695825, "rms_gain_error_db": 0.7615885579480459, "reference_mean_power": 2.220381322589703e-05, "output_mean_power": 2.645979171897363e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6122764663239433e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.855212413255228e-06, "output_power": 9.963568584189396e-06, "reference_active": true, "delta_db": 0.04748925377443033}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3142139667203495e-06, "output_power": 1.31980897375542e-06, "reference_active": true, "delta_db": 0.018449986131057226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2621117296192714e-08, "output_power": 3.383854679133362e-08, "reference_active": true, "delta_db": -1.0021312655677104}], "active_band_count": 3, "active_band_mae_db": 0.35602350182439935}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.875812750809964, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ca66ee3f25a995f32adfeed71e70eacf79c7c047d8bb44d4f138d5c3d92f1bab", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9745813869109423, "standard_deviation": 0.005298885562010828, "minimum": 0.9661387707460632, "maximum": 0.9809969944440078, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3902318189129415, "standard_deviation": 0.03744578531691392, "minimum": 1.3292396913046205, "maximum": 1.4398800503106586, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46850183437847015, "standard_deviation": 0.14439314511012602, "minimum": 0.18951010232870577, "maximum": 0.6555386392118098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19000997566744932, "standard_deviation": 0.02593630708171956, "minimum": 0.16154261506353557, "maximum": 0.23298256384242266, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9921766349301224, "standard_deviation": 0.0027808163712870274, "minimum": 0.9884412510233641, "maximum": 0.9953269421290264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8659601440861923, "standard_deviation": 0.034581380113358935, "minimum": 0.8301811169836881, "maximum": 0.9312137258311539, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5113393706706628, "standard_deviation": 0.18312623384587473, "minimum": 0.15932426195587643, "maximum": 0.7607501517295233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10623964991513624, "standard_deviation": 0.026017279335310266, "minimum": 0.07825010509661307, "maximum": 0.14267010018018889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9140846004557843, "standard_deviation": 0.0049756021939228266, "minimum": 0.9087890587232846, "maximum": 0.9207206614338871, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.416155949967293, "standard_deviation": 0.02710028620713335, "minimum": 1.379754268664331, "maximum": 1.4553802080097153, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7173824232601873, "standard_deviation": 0.3299928003992294, "minimum": 0.2581301300935673, "maximum": 1.3300359953792682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3589846881582153, "standard_deviation": 0.01211030346308227, "minimum": 0.3497396751025631, "maximum": 0.3855595254390967, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9494882370814413, "standard_deviation": 0.004240400419697985, "minimum": 0.9428198094160283, "maximum": 0.9549710255069565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.950299339565136, "standard_deviation": 0.019919422321943936, "minimum": 0.9243613907038644, "maximum": 0.9781518925795701, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7034128285225206, "standard_deviation": 0.4607186532204417, "minimum": 0.1633988693641556, "maximum": 1.683041339501494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.27197519846738577, "standard_deviation": 0.012339731701507635, "minimum": 0.26215497620043526, "maximum": 0.30085638734852926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9911799786726976, "standard_deviation": 0.0026400415083796605, "minimum": 0.9852853388473987, "maximum": 0.9939012316579812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3848094884046975, "standard_deviation": 0.04613518108420663, "minimum": 1.3106343917247385, "maximum": 1.4400720243411655, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6648827633500998, "standard_deviation": 0.2460716883182918, "minimum": 0.22452341553336597, "maximum": 0.9654308191698115, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10584562206057453, "standard_deviation": 0.014403757172799058, "minimum": 0.08858952939041524, "maximum": 0.13663145665123877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973561262740873, "standard_deviation": 0.0006523616335533109, "minimum": 0.9964502819830069, "maximum": 0.9983451391068904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7882682014307728, "standard_deviation": 0.03480533693926808, "minimum": 0.7388846060961988, "maximum": 0.8522167124499503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7562790382666542, "standard_deviation": 0.23870595082908633, "minimum": 0.3458081762813303, "maximum": 1.0682596288075603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.058427612663640716, "standard_deviation": 0.006538202804584205, "minimum": 0.04633180160315972, "maximum": 0.06613400194577904, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.371876001779163, "standard_deviation": 0.04558518613445791, "minimum": 0.3178908912786122, "maximum": 0.4232944199664881, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.63276243755801, "standard_deviation": 0.005901346662602537, "minimum": 5.62456604147291, "maximum": 5.639266599914062, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41115948282167647, "standard_deviation": 0.04849015569645999, "minimum": 0.35369893228382365, "maximum": 0.47655654386178714, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06348209075999037, "standard_deviation": 0.04901670236271188, "minimum": -0.12240652846583959, "maximum": 0.00839479276768649, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924412175357744, "standard_deviation": 0.0020142259904772325, "minimum": 0.9878353735860282, "maximum": 0.9942972618721555, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7520842466280746, "standard_deviation": 0.01829095554299433, "minimum": 0.7267708774535585, "maximum": 0.7794592372866813, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5118945426891197, "standard_deviation": 0.16117719128366265, "minimum": 0.1927035490728076, "maximum": 0.6779878170089036, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0886884199806307, "standard_deviation": 0.010916206258788566, "minimum": 0.07816974023468004, "maximum": 0.11316767738692088, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979090234665898, "standard_deviation": 0.0006069918138991246, "minimum": 0.9966052758356986, "maximum": 0.9985220024506711, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4027698461865201, "standard_deviation": 0.019311268008571664, "minimum": 0.36628396572252864, "maximum": 0.4233704776624381, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5755634554456628, "standard_deviation": 0.15991480523225438, "minimum": 0.3138629655614903, "maximum": 0.7758065999855219, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045618960071246384, "standard_deviation": 0.005949410235605614, "minimum": 0.038265603616091624, "maximum": 0.05779837530265805, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4507-16021-0045", "language": "en", "style": "normal", "speaker": "librispeech_4507", "source_utterance_id": "4507-16021-0045", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 78240, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4507-16021-0045.wav", "source_file_sha256": "3ca0ecbf7e912204bcc67653acfdce7867114fbf05b01f94110211f0e292917f", "decoded_pcm_sha256": "ef91d0d7b9ebfc4c08837a0dcebf79c72e3e4ff97d381f356a678c2a085d6e6c", "original_source_sha256": "86e41fe1ab9a5fdf5d806ab2728428f4b8d2720df04ce6431b65f617878b924e", "corpus_entry_id": "librispeech_4507-16021-0045"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.727339029312134, "predicted_mos": 3.727339029312134, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "6c835bb3f8878645ff8533ebdcae1eabf5cefa5ea8d153ae2496b238459c99da", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.542686881999543, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4507-16021-0045", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6c835bb3f8878645ff8533ebdcae1eabf5cefa5ea8d153ae2496b238459c99da", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "364c398f7e47ebcead99b59ac008405f9502702cf89d4940bfeb13f7b50e73fa"}, "rows": [{"case": "librispeech_4507-16021-0045", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.0837388038635254, "predicted_mos": 2.0837388038635254, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "779548e307693921a9ea5c3c668b5ff70e7fc405c8b9173e15244364c1013ee2", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.893606515997817, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4507-16021-0045", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "779548e307693921a9ea5c3c668b5ff70e7fc405c8b9173e15244364c1013ee2", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9308492758577004, "全帯域_10ms_rms_envelope_nrmse": 0.09077278796500471, "全帯域_10ms_envelope_correlation": 0.9936211750084467, "全帯域_10ms_modulation_mae_db": 0.3462277678086789, "全帯域_40ms_log_rms_mae_db": 0.5418101485495196, "全帯域_40ms_rms_envelope_nrmse": 0.05525058959218527, "全帯域_40ms_envelope_correlation": 0.9976300452212132, "全帯域_40ms_modulation_mae_db": 0.4093340076916241, "300–1000 Hz_10ms_log_rms_mae_db": 1.578445697881438, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10205723810800815, "300–1000 Hz_10ms_envelope_correlation": 0.9930456210035079, "300–1000 Hz_10ms_modulation_mae_db": 0.5378254698052204, "300–1000 Hz_40ms_log_rms_mae_db": 0.8814451963351899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0669619397723135, "300–1000 Hz_40ms_envelope_correlation": 0.9971285452738139, "300–1000 Hz_40ms_modulation_mae_db": 0.6603134430169469, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5361140830683933, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.165847500790001, "1000–3000 Hz_10ms_envelope_correlation": 0.9843104858022177, "1000–3000 Hz_10ms_modulation_mae_db": 1.7597521496119775, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9928854387980178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08583647621740097, "1000–3000 Hz_40ms_envelope_correlation": 0.9960983467795493, "1000–3000 Hz_40ms_modulation_mae_db": 1.0161630939192958, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4694088347676644, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2163662732571935, "3000–6000 Hz_10ms_envelope_correlation": 0.9674550409224605, "3000–6000 Hz_10ms_modulation_mae_db": 0.9354017779943148, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8967094111862333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12192772111371103, "3000–6000 Hz_40ms_envelope_correlation": 0.9892198118160631, "3000–6000 Hz_40ms_modulation_mae_db": 0.779774995643964, "mr_stft_db": 5.822639680759889, "band_mae_db": 0.33706438210888884, "rms_gain_error_db": -0.12740408848425344, "output_peak": 0.4975156273204168}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.02984362024703, "全帯域_10ms_rms_envelope_nrmse": 0.12211836653335198, "全帯域_10ms_envelope_correlation": 0.9882475324943717, "全帯域_10ms_modulation_mae_db": 0.47676904815841575, "全帯域_40ms_log_rms_mae_db": 0.602598613273607, "全帯域_40ms_rms_envelope_nrmse": 0.0706128381085547, "全帯域_40ms_envelope_correlation": 0.9959825668890726, "全帯域_40ms_modulation_mae_db": 0.554822703563509, "300–1000 Hz_10ms_log_rms_mae_db": 1.6382715999601594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13788944358558416, "300–1000 Hz_10ms_envelope_correlation": 0.9869292077057251, "300–1000 Hz_10ms_modulation_mae_db": 0.5473738758316901, "300–1000 Hz_40ms_log_rms_mae_db": 0.9763354370077364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08494551836402524, "300–1000 Hz_40ms_envelope_correlation": 0.9951336684259326, "300–1000 Hz_40ms_modulation_mae_db": 0.6496736723195925, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6273148450481425, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16770580395147572, "1000–3000 Hz_10ms_envelope_correlation": 0.9841605083430145, "1000–3000 Hz_10ms_modulation_mae_db": 1.4267425432338694, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1111360179274248, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0964944231833248, "1000–3000 Hz_40ms_envelope_correlation": 0.995103685168285, "1000–3000 Hz_40ms_modulation_mae_db": 1.0354895766546823, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4117215207409242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20735143496202854, "3000–6000 Hz_10ms_envelope_correlation": 0.9700434826195222, "3000–6000 Hz_10ms_modulation_mae_db": 0.9959231973956761, "3000–6000 Hz_40ms_log_rms_mae_db": 0.821151413124307, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1250596083221352, "3000–6000 Hz_40ms_envelope_correlation": 0.988563879757655, "3000–6000 Hz_40ms_modulation_mae_db": 1.5427742773856152, "mr_stft_db": 5.810801346835731, "band_mae_db": 0.3528235446256896, "rms_gain_error_db": -0.14291638762749678, "output_peak": 0.48018709788355624}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9490873721603426, "全帯域_10ms_rms_envelope_nrmse": 0.09410872792363997, "全帯域_10ms_envelope_correlation": 0.9930788823266951, "全帯域_10ms_modulation_mae_db": 0.23304829827558182, "全帯域_40ms_log_rms_mae_db": 0.5538881568252636, "全帯域_40ms_rms_envelope_nrmse": 0.052609021612687365, "全帯域_40ms_envelope_correlation": 0.9977343935989247, "全帯域_40ms_modulation_mae_db": 0.550176903441053, "300–1000 Hz_10ms_log_rms_mae_db": 1.57441679923676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10588064291040981, "300–1000 Hz_10ms_envelope_correlation": 0.9922149363128879, "300–1000 Hz_10ms_modulation_mae_db": 0.16687883527066724, "300–1000 Hz_40ms_log_rms_mae_db": 0.9168013385035011, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05936011320494499, "300–1000 Hz_40ms_envelope_correlation": 0.9974346725212079, "300–1000 Hz_40ms_modulation_mae_db": 0.6254070476366917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6457167857919415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17809552358303377, "1000–3000 Hz_10ms_envelope_correlation": 0.9818310715054639, "1000–3000 Hz_10ms_modulation_mae_db": 1.250976813634173, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1182025911473177, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10807810013123341, "1000–3000 Hz_40ms_envelope_correlation": 0.9935072903865447, "1000–3000 Hz_40ms_modulation_mae_db": 0.7542539339524681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4691109544022356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22182282644464368, "3000–6000 Hz_10ms_envelope_correlation": 0.9656378833198639, "3000–6000 Hz_10ms_modulation_mae_db": 0.1375674782021172, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8749435693333565, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1331015419393671, "3000–6000 Hz_40ms_envelope_correlation": 0.9872294958027551, "3000–6000 Hz_40ms_modulation_mae_db": 0.8196181224645471, "mr_stft_db": 5.837201098757665, "band_mae_db": 0.34396368408642736, "rms_gain_error_db": 0.028413739399341872, "output_peak": 0.5100138455130592}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9927426595215999, "全帯域_10ms_rms_envelope_nrmse": 0.09379691837631719, "全帯域_10ms_envelope_correlation": 0.9930944897074083, "全帯域_10ms_modulation_mae_db": 0.2103128927039286, "全帯域_40ms_log_rms_mae_db": 0.5999599101653555, "全帯域_40ms_rms_envelope_nrmse": 0.05099854899489238, "全帯域_40ms_envelope_correlation": 0.9979222471122393, "全帯域_40ms_modulation_mae_db": 0.08491660898354071, "300–1000 Hz_10ms_log_rms_mae_db": 1.681555350323537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11114570446223937, "300–1000 Hz_10ms_envelope_correlation": 0.9915221227170955, "300–1000 Hz_10ms_modulation_mae_db": 0.29552306383654364, "300–1000 Hz_40ms_log_rms_mae_db": 1.026092941842474, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06411898461216683, "300–1000 Hz_40ms_envelope_correlation": 0.9972406639691802, "300–1000 Hz_40ms_modulation_mae_db": 0.20450716014928014, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5950068721093296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18130839182229003, "1000–3000 Hz_10ms_envelope_correlation": 0.9808167762589056, "1000–3000 Hz_10ms_modulation_mae_db": 1.892201264451466, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0842248751435952, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09216884477703512, "1000–3000 Hz_40ms_envelope_correlation": 0.995182422910445, "1000–3000 Hz_40ms_modulation_mae_db": 1.240140996022193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4654148528630884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21023122553186668, "3000–6000 Hz_10ms_envelope_correlation": 0.9698210736947334, "3000–6000 Hz_10ms_modulation_mae_db": 0.24026622006327084, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8759936402971767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1210743126800298, "3000–6000 Hz_40ms_envelope_correlation": 0.9893793696471733, "3000–6000 Hz_40ms_modulation_mae_db": 0.5206025086814242, "mr_stft_db": 5.820652052544886, "band_mae_db": 0.3215976996551752, "rms_gain_error_db": -0.0972862587539358, "output_peak": 0.490513755227274}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0070290305013139, "全帯域_10ms_rms_envelope_nrmse": 0.1097110600347834, "全帯域_10ms_envelope_correlation": 0.9905212741454968, "全帯域_10ms_modulation_mae_db": 0.6387881583977952, "全帯域_40ms_log_rms_mae_db": 0.5609660242337875, "全帯域_40ms_rms_envelope_nrmse": 0.05765835571687296, "全帯域_40ms_envelope_correlation": 0.9972503967057899, "全帯域_40ms_modulation_mae_db": 0.5672997801765457, "300–1000 Hz_10ms_log_rms_mae_db": 1.652408300268741, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12701883038374434, "300–1000 Hz_10ms_envelope_correlation": 0.9887756849138853, "300–1000 Hz_10ms_modulation_mae_db": 0.9165908687702456, "300–1000 Hz_40ms_log_rms_mae_db": 0.9587780237219784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0672726832429987, "300–1000 Hz_40ms_envelope_correlation": 0.996742471171396, "300–1000 Hz_40ms_modulation_mae_db": 0.7299003185701102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5869007935004549, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16475113743159026, "1000–3000 Hz_10ms_envelope_correlation": 0.984504204197832, "1000–3000 Hz_10ms_modulation_mae_db": 1.2204679942044037, "1000–3000 Hz_40ms_log_rms_mae_db": 1.066945722170251, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0935550846555328, "1000–3000 Hz_40ms_envelope_correlation": 0.9952491357374857, "1000–3000 Hz_40ms_modulation_mae_db": 0.4408403595730654, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4298739501385407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20273680749727657, "3000–6000 Hz_10ms_envelope_correlation": 0.9715287787654192, "3000–6000 Hz_10ms_modulation_mae_db": 0.5628155389427413, "3000–6000 Hz_40ms_log_rms_mae_db": 0.858621737530274, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12386708552687929, "3000–6000 Hz_40ms_envelope_correlation": 0.9887731368721484, "3000–6000 Hz_40ms_modulation_mae_db": 0.22680984388429973, "mr_stft_db": 5.812519480370618, "band_mae_db": 0.32576856189481224, "rms_gain_error_db": -0.04362124709785389, "output_peak": 0.6167552628603468}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9704095963523873, "全帯域_10ms_rms_envelope_nrmse": 0.11468641055433068, "全帯域_10ms_envelope_correlation": 0.9897443094010998, "全帯域_10ms_modulation_mae_db": 0.4630245141052598, "全帯域_40ms_log_rms_mae_db": 0.5518342970684044, "全帯域_40ms_rms_envelope_nrmse": 0.05437193779138555, "全帯域_40ms_envelope_correlation": 0.9975820944249012, "全帯域_40ms_modulation_mae_db": 0.36536789542508924, "300–1000 Hz_10ms_log_rms_mae_db": 1.6738285064273726, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1300041695176368, "300–1000 Hz_10ms_envelope_correlation": 0.9882963217169067, "300–1000 Hz_10ms_modulation_mae_db": 0.7719815976107535, "300–1000 Hz_40ms_log_rms_mae_db": 0.9852723199682677, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06157005124159142, "300–1000 Hz_40ms_envelope_correlation": 0.9972314216263768, "300–1000 Hz_40ms_modulation_mae_db": 0.47749599331153725, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5821238853643165, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16443556424194422, "1000–3000 Hz_10ms_envelope_correlation": 0.9838194727041913, "1000–3000 Hz_10ms_modulation_mae_db": 1.30788223765369, "1000–3000 Hz_40ms_log_rms_mae_db": 1.081193277138507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08664508498157958, "1000–3000 Hz_40ms_envelope_correlation": 0.9954520738464456, "1000–3000 Hz_40ms_modulation_mae_db": 0.39859271554040365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4530388883553773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2214357322622422, "3000–6000 Hz_10ms_envelope_correlation": 0.965997044814621, "3000–6000 Hz_10ms_modulation_mae_db": 1.2578114569022387, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9227925198985337, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12642227247770635, "3000–6000 Hz_40ms_envelope_correlation": 0.9882367266616218, "3000–6000 Hz_40ms_modulation_mae_db": 0.7490759631201053, "mr_stft_db": 5.8297530140271405, "band_mae_db": 0.2307069435619221, "rms_gain_error_db": 0.0424375122060058, "output_peak": 0.5643383693567033}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9624123527730007, "全帯域_10ms_rms_envelope_nrmse": 0.08451162215384726, "全帯域_10ms_envelope_correlation": 0.9943629885971761, "全帯域_10ms_modulation_mae_db": 0.1785041316971702, "全帯域_40ms_log_rms_mae_db": 0.5450648682597525, "全帯域_40ms_rms_envelope_nrmse": 0.04637997635092919, "全帯域_40ms_envelope_correlation": 0.9982068633836696, "全帯域_40ms_modulation_mae_db": 0.3163357669288129, "300–1000 Hz_10ms_log_rms_mae_db": 1.5876122313941479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09703908784007786, "300–1000 Hz_10ms_envelope_correlation": 0.9934403092813849, "300–1000 Hz_10ms_modulation_mae_db": 0.36854390875544113, "300–1000 Hz_40ms_log_rms_mae_db": 0.9347639068976934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053846815228799715, "300–1000 Hz_40ms_envelope_correlation": 0.9979017004566233, "300–1000 Hz_40ms_modulation_mae_db": 0.2944424765647849, "1000–3000 Hz_10ms_log_rms_mae_db": 1.648110096963897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18003068924183568, "1000–3000 Hz_10ms_envelope_correlation": 0.9810461944602834, "1000–3000 Hz_10ms_modulation_mae_db": 1.8662945228690297, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1655349138514781, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10266398339595458, "1000–3000 Hz_40ms_envelope_correlation": 0.9938932308451222, "1000–3000 Hz_40ms_modulation_mae_db": 1.1506809378119474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.483380084138349, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2114060769696383, "3000–6000 Hz_10ms_envelope_correlation": 0.9689983513975866, "3000–6000 Hz_10ms_modulation_mae_db": 0.5687029690841855, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8910274370898833, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12062907477575784, "3000–6000 Hz_40ms_envelope_correlation": 0.9899481202034468, "3000–6000 Hz_40ms_modulation_mae_db": 1.2474266710717785, "mr_stft_db": 5.835055027576978, "band_mae_db": 0.2681904686507518, "rms_gain_error_db": -0.019756542887286205, "output_peak": 0.5303200284950733}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9519391780441815, "全帯域_10ms_rms_envelope_nrmse": 0.09567573766663796, "全帯域_10ms_envelope_correlation": 0.9929138052660001, "全帯域_10ms_modulation_mae_db": 0.25320871399603934, "全帯域_40ms_log_rms_mae_db": 0.4828607643199806, "全帯域_40ms_rms_envelope_nrmse": 0.04653767650546809, "全帯域_40ms_envelope_correlation": 0.9984362461559295, "全帯域_40ms_modulation_mae_db": 0.3570269616141681, "300–1000 Hz_10ms_log_rms_mae_db": 1.6213215768598717, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1086057560511956, "300–1000 Hz_10ms_envelope_correlation": 0.9920835255874195, "300–1000 Hz_10ms_modulation_mae_db": 0.3838054253775558, "300–1000 Hz_40ms_log_rms_mae_db": 0.8773297714715662, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05632688997972709, "300–1000 Hz_40ms_envelope_correlation": 0.9981596987510852, "300–1000 Hz_40ms_modulation_mae_db": 0.40196473222094237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5832077374990274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18008267732037514, "1000–3000 Hz_10ms_envelope_correlation": 0.9807606331943556, "1000–3000 Hz_10ms_modulation_mae_db": 1.8238413465954715, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0702205488847887, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09982863810959816, "1000–3000 Hz_40ms_envelope_correlation": 0.9940694642153244, "1000–3000 Hz_40ms_modulation_mae_db": 0.9632424265936406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4342747710617432, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22001452780389966, "3000–6000 Hz_10ms_envelope_correlation": 0.9667443898907088, "3000–6000 Hz_10ms_modulation_mae_db": 0.29753933676767463, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8477002067350822, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13748665074134817, "3000–6000 Hz_40ms_envelope_correlation": 0.9873892691551021, "3000–6000 Hz_40ms_modulation_mae_db": 0.7778458340707091, "mr_stft_db": 5.823654710759968, "band_mae_db": 0.4140413610545644, "rms_gain_error_db": -0.15674386900755433, "output_peak": 0.4930083401586251}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 78240, "output_samples": 78240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.723012617053997, "spectral_convergence": 0.27311512375434177, "spectral_convergence_status": "ok", "amplitude_floor": 1.658911626731477e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.697477935797429, "spectral_convergence": 0.29837602747033115, "spectral_convergence_status": "ok", "amplitude_floor": 1.4097603217110054e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.047428489428243, "spectral_convergence": 0.33663841785308407, "spectral_convergence_status": "ok", "amplitude_floor": 8.986287240444543e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.8753665955686144e-05, "output_power": 5.903163385320927e-05, "delta_db": 0.8307752625411735, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00084480798498807, "output_power": 0.0008550571266042472, "delta_db": 0.052371209423536166, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015560766761226633, "output_power": 0.0014510140309022916, "delta_db": -0.3035938120062719, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.758941174067266e-05, "output_power": 6.47916552391933e-05, "delta_db": 0.5117643269464223, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.7471770650556e-05, "output_power": 8.045956808780466e-05, "delta_db": 0.164342148536038, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.273577609828158e-05, "output_power": 5.083366101116711e-05, "delta_db": -0.15953953319989134, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.33706438210888884, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12740408848425344, "pitch": null, "output_rms_dbfs": -25.91558596645335, "output_peak": 0.4975156273204168, "reference_rms_dbfs": -25.788181877969095, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.822639680759889}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 78240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.0002230061513489049, "log_rms_mae_db": 0.9308492758577004, "rms_envelope_nrmse": 0.09077278796500471, "envelope_correlation": 0.9936211750084467, "rms_gain_error_db": -0.12740425980948686, "reference_mean_power": 0.002640134243154231, "output_mean_power": 0.002563808505951401}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.2388906472962606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011616255671056531, "output_power": 0.0010993098237895625, "reference_active": true, "delta_db": -0.23946053197319805}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014996209998372379, "output_power": 0.00013880302794745563, "reference_active": true, "delta_db": -0.3358257317507052}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.935590189308993e-05, "output_power": 3.537285919293359e-05, "reference_active": true, "delta_db": -0.4633970397021336}], "active_band_count": 3, "active_band_mae_db": 0.3462277678086789}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.0001943907377270206, "log_rms_mae_db": 0.5418101485495196, "rms_envelope_nrmse": 0.05525058959218527, "envelope_correlation": 0.9976300452212132, "rms_gain_error_db": -0.1274044964678422, "reference_mean_power": 0.0026564436783668454, "output_mean_power": 0.002579646298179501}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 2.080120853269114e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001087632600289277, "output_power": 0.0010239975584485442, "reference_active": true, "delta_db": -0.2618329535126802}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.148356435560039e-05, "output_power": 6.693076135871683e-05, "reference_active": true, "delta_db": -0.2858043415125248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.554313443594102e-06, "output_power": 2.183924982375879e-06, "reference_active": true, "delta_db": -0.6803647280496672}], "active_band_count": 3, "active_band_mae_db": 0.4093340076916241}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.00021647011759154595, "log_rms_mae_db": 1.578445697881438, "rms_envelope_nrmse": 0.10205723810800815, "envelope_correlation": 0.9930456210035079, "rms_gain_error_db": -0.21384015044402488, "reference_mean_power": 0.0021208071215793162, "output_mean_power": 0.0020189109244019682}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.0688695374548394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066233690351074, "output_power": 0.0009919500709106833, "reference_active": true, "delta_db": -0.3136258823299314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015588791655839424, "output_power": 0.0001385006068991615, "reference_active": true, "delta_db": -0.5136077629230493}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.732583628298963e-05, "output_power": 3.1144780213096826e-05, "reference_active": true, "delta_db": -0.7862427641626804}], "active_band_count": 3, "active_band_mae_db": 0.5378254698052204}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.00018533448652616975, "log_rms_mae_db": 0.8814451963351899, "rms_envelope_nrmse": 0.0669619397723135, "envelope_correlation": 0.9971285452738139, "rms_gain_error_db": -0.21383982007851796, "reference_mean_power": 0.0021339103768736657, "output_mean_power": 0.002031384775847679}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.9317825081344165e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001010025588563974, "output_power": 0.0009317544186414192, "reference_active": true, "delta_db": -0.350309156086733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.797692317548564e-05, "output_power": 7.023865649210725e-05, "reference_active": true, "delta_db": -0.4538989883811709}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604806628732506e-06, "output_power": 1.9527573533417024e-06, "reference_active": true, "delta_db": -1.176732184582937}], "active_band_count": 3, "active_band_mae_db": 0.6603134430169469}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 6.255230128136302e-05, "log_rms_mae_db": 1.5361140830683933, "rms_envelope_nrmse": 0.165847500790001, "envelope_correlation": 0.9843104858022177, "rms_gain_error_db": 0.2619445104539581, "reference_mean_power": 0.0001286629346191704, "output_mean_power": 0.00013666204143812123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.8253387298901982e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.359313427297316e-05, "output_power": 7.38427148289085e-05, "reference_active": true, "delta_db": 0.014703555808030474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.331803440637771e-06, "output_power": 9.140837996538259e-06, "reference_active": true, "delta_db": 0.9577519860726059}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8895709731947531e-06, "output_power": 5.093816543512346e-06, "reference_active": true, "delta_db": 4.306800906955296}], "active_band_count": 3, "active_band_mae_db": 1.7597521496119775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 6.0793288183779695e-05, "log_rms_mae_db": 0.9928854387980178, "rms_envelope_nrmse": 0.08583647621740097, "envelope_correlation": 0.9960983467795493, "rms_gain_error_db": 0.2619451475325514, "reference_mean_power": 0.00012945711895499159, "output_mean_power": 0.00013750562120103376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.7711142240683522e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.969624557508824e-05, "output_power": 7.027835373404441e-05, "reference_active": true, "delta_db": 0.036121956406086675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.240093489969041e-06, "output_power": 3.933954055538192e-06, "reference_active": true, "delta_db": 0.8427174194625847}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.585555587494176e-07, "output_power": 2.613042437526744e-07, "reference_active": true, "delta_db": 2.169649905889216}], "active_band_count": 3, "active_band_mae_db": 1.0161630939192958}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 3.765201856958641e-05, "log_rms_mae_db": 1.4694088347676644, "rms_envelope_nrmse": 0.2163662732571935, "envelope_correlation": 0.9674550409224605, "rms_gain_error_db": -0.12148217052136173, "reference_mean_power": 2.5945814971953512e-05, "output_mean_power": 2.5230107422133076e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.30554638893879e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0948707154331005e-06, "output_power": 5.0533582482998005e-06, "reference_active": true, "delta_db": -0.03553080252718133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.046923144145412e-06, "output_power": 1.1971348194218698e-06, "reference_active": true, "delta_db": 0.5822826193310847}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.3665503073484234e-07, "output_power": 7.227326242524544e-07, "reference_active": true, "delta_db": 2.188391912124678}], "active_band_count": 3, "active_band_mae_db": 0.9354017779943148}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 2.0830708932641756e-05, "log_rms_mae_db": 0.8967094111862333, "rms_envelope_nrmse": 0.12192772111371103, "envelope_correlation": 0.9892198118160631, "rms_gain_error_db": -0.12150691805333119, "reference_mean_power": 2.6105327616819126e-05, "output_mean_power": 2.5385075305430665e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.2170827427715922e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.759181059225212e-06, "output_power": 4.80544960254626e-06, "reference_active": true, "delta_db": 0.04201799422134414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.967295837314181e-07, "output_power": 7.078229486274972e-07, "reference_active": true, "delta_db": 0.7414706982068213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240630896276443e-08, "output_power": 4.636744844032407e-08, "reference_active": true, "delta_db": 1.5558362945037265}], "active_band_count": 3, "active_band_mae_db": 0.779774995643964}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8603668534717299, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "58270cf471d308c57472bce2463154fbd5cb19a488bf9ec94b7a76c941e12f2b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826561683082831, "standard_deviation": 0.0016910493818144288, "minimum": 0.9807606331943556, "maximum": 0.984504204197832, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6005618874181877, "standard_deviation": 0.037848659962827665, "minimum": 1.5361140830683933, "maximum": 1.648110096963897, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.5685198590317602, "standard_deviation": 0.29403710087888707, "minimum": 1.2204679942044037, "maximum": 1.892201264451466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17278216104781824, "standard_deviation": 0.0076977646504933096, "minimum": 0.16443556424194422, "maximum": 0.18130839182229003, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9948194562361503, "standard_deviation": 0.0008924350438458491, "minimum": 0.9935072903865447, "maximum": 0.9960983467795493, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0862929231326723, "standard_deviation": 0.049717165913637694, "minimum": 0.9928854387980178, "maximum": 1.1655349138514781, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8749255050084621, "standard_deviation": 0.31460556659580935, "minimum": 0.39859271554040365, "maximum": 1.240140996022193, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09565882943145743, "standard_deviation": 0.007706471628728715, "minimum": 0.08583647621740097, "maximum": 0.10807810013123341, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9682782556781144, "standard_deviation": 0.0021309510872918875, "minimum": 0.9656378833198639, "maximum": 0.9715287787654192, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4520279820584903, "standard_deviation": 0.024458988958928105, "minimum": 1.4117215207409242, "maximum": 1.483380084138349, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6245034969190274, "standard_deviation": 0.40253818693645904, "minimum": 0.1375674782021172, "maximum": 1.2578114569022387, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21392061309109864, "standard_deviation": 0.007073139568986662, "minimum": 0.20273680749727657, "maximum": 0.22182282644464368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9885924762394956, "standard_deviation": 0.0009498840877906407, "minimum": 0.9872294958027551, "maximum": 0.9899481202034468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8736174918993559, "standard_deviation": 0.0314215545004925, "minimum": 0.821151413124307, "maximum": 0.9227925198985337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8329910270403054, "standard_deviation": 0.40597059878837827, "minimum": 0.22680984388429973, "maximum": 1.5427742773856152, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12619603344711683, "standard_deviation": 0.006063618363889672, "minimum": 0.12062907477575784, "maximum": 0.13748665074134817, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907884661548516, "standard_deviation": 0.0024358400128160237, "minimum": 0.9869292077057251, "maximum": 0.9934403092813849, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6259825077940033, "standard_deviation": 0.042493188849291534, "minimum": 1.57441679923676, "maximum": 1.681555350323537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.49856538065726463, "standard_deviation": 0.24921102288969296, "minimum": 0.16687883527066724, "maximum": 0.9165908687702456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11495510910736201, "standard_deviation": 0.014747890732153724, "minimum": 0.09703908784007786, "maximum": 0.13788944358558416, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997121605274452, "standard_deviation": 0.0009185482341950064, "minimum": 0.9951336684259326, "maximum": 0.9981596987510852, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9446023669685508, "standard_deviation": 0.05194340139087826, "minimum": 0.8773297714715662, "maximum": 1.026092941842474, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5054631054737357, "standard_deviation": 0.19128594377126357, "minimum": 0.20450716014928014, "maximum": 0.7299003185701102, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06430037445582093, "standard_deviation": 0.009613142937893141, "minimum": 0.053846815228799715, "maximum": 0.08494551836402524, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.32426958070477896, "standard_deviation": 0.055206218366864894, "minimum": 0.2307069435619221, "maximum": 0.4140413610545644, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.824034551454109, "standard_deviation": 0.00962701173542272, "minimum": 5.810801346835731, "maximum": 5.837201098757665, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5228315408518818, "standard_deviation": 0.04653441075577903, "minimum": 0.48018709788355624, "maximum": 0.6167552628603468, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0646096427816291, "standard_deviation": 0.07764049193922273, "minimum": -0.15674386900755433, "maximum": 0.0424375122060058, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9919480571183368, "standard_deviation": 0.002162771254166384, "minimum": 0.9882475324943717, "maximum": 0.9943629885971761, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9742891356821946, "standard_deviation": 0.03311562799044478, "minimum": 0.9308492758577004, "maximum": 1.02984362024703, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3499854406428587, "standard_deviation": 0.16222882217175333, "minimum": 0.1785041316971702, "maximum": 0.6387881583977952, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10067270390098915, "standard_deviation": 0.013161160820738751, "minimum": 0.08451162215384726, "maximum": 0.12211836653335198, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975931066864674, "standard_deviation": 0.0007491567788365106, "minimum": 0.9959825668890726, "maximum": 0.9984362461559295, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5548728478369589, "standard_deviation": 0.03743659750679496, "minimum": 0.4828607643199806, "maximum": 0.602598613273607, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40066007847804286, "standard_deviation": 0.16209315587428932, "minimum": 0.08491660898354071, "maximum": 0.5672997801765457, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054302368084121935, "standard_deviation": 0.007700050262489724, "minimum": 0.04637997635092919, "maximum": 0.0706128381085547, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0045", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1623151302337646, "predicted_mos": 2.1623151302337646, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "24569927d1a20947e878f3db46d7dadd5f763b7efafd4f512dcf90acf987ccbe", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6575190001167357, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9324681027842134, "全帯域_10ms_rms_envelope_nrmse": 0.10223761838103024, "全帯域_10ms_envelope_correlation": 0.991893591102772, "全帯域_10ms_modulation_mae_db": 0.2500330496541038, "全帯域_40ms_log_rms_mae_db": 0.5521703957003551, "全帯域_40ms_rms_envelope_nrmse": 0.0662989322718726, "全帯域_40ms_envelope_correlation": 0.9965590691480033, "全帯域_40ms_modulation_mae_db": 0.2951507969079004, "300–1000 Hz_10ms_log_rms_mae_db": 1.569510422102213, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11551786071532927, "300–1000 Hz_10ms_envelope_correlation": 0.9909896225065272, "300–1000 Hz_10ms_modulation_mae_db": 0.3504962169133661, "300–1000 Hz_40ms_log_rms_mae_db": 0.862215161329656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07871873843829251, "300–1000 Hz_40ms_envelope_correlation": 0.9958906033219057, "300–1000 Hz_40ms_modulation_mae_db": 0.33683329858517785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.525215410539605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15539749052636964, "1000–3000 Hz_10ms_envelope_correlation": 0.9854408780519329, "1000–3000 Hz_10ms_modulation_mae_db": 1.5848307441269995, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9430002652434204, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07323934965367215, "1000–3000 Hz_40ms_envelope_correlation": 0.9966986334816268, "1000–3000 Hz_40ms_modulation_mae_db": 0.8268319030071152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3257730565389159, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20328455264616835, "3000–6000 Hz_10ms_envelope_correlation": 0.9712882241671364, "3000–6000 Hz_10ms_modulation_mae_db": 0.5121911296970616, "3000–6000 Hz_40ms_log_rms_mae_db": 0.779306057852905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11849609028128054, "3000–6000 Hz_40ms_envelope_correlation": 0.9901126179785461, "3000–6000 Hz_40ms_modulation_mae_db": 0.4544009999637017, "mr_stft_db": 5.734830566821469, "band_mae_db": 0.30296308685864537, "rms_gain_error_db": -0.1514473123872489, "output_peak": 0.49409213714798544}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.031179105844674, "全帯域_10ms_rms_envelope_nrmse": 0.12750035926029563, "全帯域_10ms_envelope_correlation": 0.9872752856432835, "全帯域_10ms_modulation_mae_db": 0.5447124745348909, "全帯域_40ms_log_rms_mae_db": 0.6024043856582565, "全帯域_40ms_rms_envelope_nrmse": 0.07796994315908046, "全帯域_40ms_envelope_correlation": 0.9952207721887917, "全帯域_40ms_modulation_mae_db": 0.62427336032745, "300–1000 Hz_10ms_log_rms_mae_db": 1.6234290369757052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14665249727565927, "300–1000 Hz_10ms_envelope_correlation": 0.985279671906947, "300–1000 Hz_10ms_modulation_mae_db": 0.6391083845240013, "300–1000 Hz_40ms_log_rms_mae_db": 0.9159192521522097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09291811949335868, "300–1000 Hz_40ms_envelope_correlation": 0.9942595794883669, "300–1000 Hz_40ms_modulation_mae_db": 0.6768579878442802, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5599288722211724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13031050558735244, "1000–3000 Hz_10ms_envelope_correlation": 0.9896501556422059, "1000–3000 Hz_10ms_modulation_mae_db": 0.4000604913034495, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0056846748606438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06706327722424757, "1000–3000 Hz_40ms_envelope_correlation": 0.9971865715887864, "1000–3000 Hz_40ms_modulation_mae_db": 0.25563809195933923, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3383678833229298, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2045018947149041, "3000–6000 Hz_10ms_envelope_correlation": 0.9709595068917969, "3000–6000 Hz_10ms_modulation_mae_db": 1.2075153708644413, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7835174221166513, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1238959828652268, "3000–6000 Hz_40ms_envelope_correlation": 0.9890761650609056, "3000–6000 Hz_40ms_modulation_mae_db": 1.707524747967198, "mr_stft_db": 5.726274609002644, "band_mae_db": 0.3523226397403569, "rms_gain_error_db": -0.1945819464444601, "output_peak": 0.4816239955161608}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9569432448304034, "全帯域_10ms_rms_envelope_nrmse": 0.09638011492369718, "全帯域_10ms_envelope_correlation": 0.9926952453389414, "全帯域_10ms_modulation_mae_db": 0.1271581415328843, "全帯域_40ms_log_rms_mae_db": 0.550418459459903, "全帯域_40ms_rms_envelope_nrmse": 0.05243520985428343, "全帯域_40ms_envelope_correlation": 0.9977100746399763, "全帯域_40ms_modulation_mae_db": 0.34257421960573003, "300–1000 Hz_10ms_log_rms_mae_db": 1.6159808602335408, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10914378410361676, "300–1000 Hz_10ms_envelope_correlation": 0.9917025015049061, "300–1000 Hz_10ms_modulation_mae_db": 0.14048518201382584, "300–1000 Hz_40ms_log_rms_mae_db": 0.9131232250210453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0596143093152576, "300–1000 Hz_40ms_envelope_correlation": 0.9973924521361738, "300–1000 Hz_40ms_modulation_mae_db": 0.4644281638854122, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5623223099185017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16165343810546215, "1000–3000 Hz_10ms_envelope_correlation": 0.9842075474238877, "1000–3000 Hz_10ms_modulation_mae_db": 1.0398905441267414, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9881889856038286, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08762024108280994, "1000–3000 Hz_40ms_envelope_correlation": 0.9952485104466334, "1000–3000 Hz_40ms_modulation_mae_db": 0.6664179181966202, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3666623158973148, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19821817659692129, "3000–6000 Hz_10ms_envelope_correlation": 0.9729297445719192, "3000–6000 Hz_10ms_modulation_mae_db": 0.6624999510331468, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8202607804991059, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11173156944881274, "3000–6000 Hz_40ms_envelope_correlation": 0.9916030441372353, "3000–6000 Hz_40ms_modulation_mae_db": 1.084876200888474, "mr_stft_db": 5.7522718663595755, "band_mae_db": 0.3759330161982333, "rms_gain_error_db": 0.006945520652775582, "output_peak": 0.508886306463613}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0033408779804132, "全帯域_10ms_rms_envelope_nrmse": 0.09755578419316653, "全帯域_10ms_envelope_correlation": 0.992548029410853, "全帯域_10ms_modulation_mae_db": 0.18785124719174986, "全帯域_40ms_log_rms_mae_db": 0.614870334680942, "全帯域_40ms_rms_envelope_nrmse": 0.05504608053546957, "全帯域_40ms_envelope_correlation": 0.9976056450396379, "全帯域_40ms_modulation_mae_db": 0.14209619352353373, "300–1000 Hz_10ms_log_rms_mae_db": 1.6872907206917414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11420562063583255, "300–1000 Hz_10ms_envelope_correlation": 0.9910247019618698, "300–1000 Hz_10ms_modulation_mae_db": 0.18527112871372675, "300–1000 Hz_40ms_log_rms_mae_db": 1.001701404742997, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06769956559670565, "300–1000 Hz_40ms_envelope_correlation": 0.9968775557006765, "300–1000 Hz_40ms_modulation_mae_db": 0.2960614854856371, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4960416944942583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15054674090328418, "1000–3000 Hz_10ms_envelope_correlation": 0.9861187993684695, "1000–3000 Hz_10ms_modulation_mae_db": 1.3248239579929277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9384711220433779, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06447101785082987, "1000–3000 Hz_40ms_envelope_correlation": 0.9975177512916537, "1000–3000 Hz_40ms_modulation_mae_db": 0.932167254121318, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3326966959146738, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18742443057942781, "3000–6000 Hz_10ms_envelope_correlation": 0.9757226347461379, "3000–6000 Hz_10ms_modulation_mae_db": 0.31205396527176205, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7704933796265235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1007417699710394, "3000–6000 Hz_40ms_envelope_correlation": 0.9925802837722193, "3000–6000 Hz_40ms_modulation_mae_db": 0.7199804605439671, "mr_stft_db": 5.735094733191566, "band_mae_db": 0.3336797074933067, "rms_gain_error_db": -0.11715099779328894, "output_peak": 0.4905962311770639}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0204116608218896, "全帯域_10ms_rms_envelope_nrmse": 0.11677265914015278, "全帯域_10ms_envelope_correlation": 0.9892662847924395, "全帯域_10ms_modulation_mae_db": 0.7794596492170482, "全帯域_40ms_log_rms_mae_db": 0.5789033745706597, "全帯域_40ms_rms_envelope_nrmse": 0.061782764632794654, "全帯域_40ms_envelope_correlation": 0.9968549786110995, "全帯域_40ms_modulation_mae_db": 0.717747641679837, "300–1000 Hz_10ms_log_rms_mae_db": 1.6322603238370035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13415553925718782, "300–1000 Hz_10ms_envelope_correlation": 0.9874789073295651, "300–1000 Hz_10ms_modulation_mae_db": 1.0135183783526651, "300–1000 Hz_40ms_log_rms_mae_db": 0.9314194088930935, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07268039918935285, "300–1000 Hz_40ms_envelope_correlation": 0.996183018810252, "300–1000 Hz_40ms_modulation_mae_db": 0.8483391667146618, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5551245143739179, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14750525109508206, "1000–3000 Hz_10ms_envelope_correlation": 0.9868808120035785, "1000–3000 Hz_10ms_modulation_mae_db": 0.6978906282182741, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9931921036327165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07824488789739227, "1000–3000 Hz_40ms_envelope_correlation": 0.996230668435937, "1000–3000 Hz_40ms_modulation_mae_db": 0.06177811883013257, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2969419508977131, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1771977597240888, "3000–6000 Hz_10ms_envelope_correlation": 0.9785866975198473, "3000–6000 Hz_10ms_modulation_mae_db": 0.43145564321669333, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7218856660617138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10800577554770738, "3000–6000 Hz_40ms_envelope_correlation": 0.9922459651729975, "3000–6000 Hz_40ms_modulation_mae_db": 1.0348656669171683, "mr_stft_db": 5.726614132904436, "band_mae_db": 0.35797332978865465, "rms_gain_error_db": -0.05777285967657347, "output_peak": 0.5951817119637042}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9848228323071032, "全帯域_10ms_rms_envelope_nrmse": 0.11579682794462792, "全帯域_10ms_envelope_correlation": 0.9895470672765465, "全帯域_10ms_modulation_mae_db": 0.5029722355248677, "全帯域_40ms_log_rms_mae_db": 0.5581959577196872, "全帯域_40ms_rms_envelope_nrmse": 0.053687642244127906, "全帯域_40ms_envelope_correlation": 0.99764651534444, "全帯域_40ms_modulation_mae_db": 0.3597856689355581, "300–1000 Hz_10ms_log_rms_mae_db": 1.6797694183931875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13264736309222805, "300–1000 Hz_10ms_envelope_correlation": 0.9878545908133799, "300–1000 Hz_10ms_modulation_mae_db": 0.7838126892083519, "300–1000 Hz_40ms_log_rms_mae_db": 0.9302995372188084, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0630683693703458, "300–1000 Hz_40ms_envelope_correlation": 0.9971148885255858, "300–1000 Hz_40ms_modulation_mae_db": 0.45758929951799665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.541704185857032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15429866903525485, "1000–3000 Hz_10ms_envelope_correlation": 0.985421079017586, "1000–3000 Hz_10ms_modulation_mae_db": 0.8635790040704977, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0043544016263708, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08756661790474962, "1000–3000 Hz_40ms_envelope_correlation": 0.9952440134931588, "1000–3000 Hz_40ms_modulation_mae_db": 0.33659623851125353, "3000–6000 Hz_10ms_log_rms_mae_db": 1.317708769697576, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2025224331369464, "3000–6000 Hz_10ms_envelope_correlation": 0.971560703221788, "3000–6000 Hz_10ms_modulation_mae_db": 0.38955009137324853, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8059847723358777, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10692043968582145, "3000–6000 Hz_40ms_envelope_correlation": 0.9921637798619318, "3000–6000 Hz_40ms_modulation_mae_db": 0.48936118577931476, "mr_stft_db": 5.743078774858578, "band_mae_db": 0.2752039009755281, "rms_gain_error_db": 0.04250403415773076, "output_peak": 0.5732985543386362}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9752869539120961, "全帯域_10ms_rms_envelope_nrmse": 0.08884516788845508, "全帯域_10ms_envelope_correlation": 0.9937979681146347, "全帯域_10ms_modulation_mae_db": 0.30275254832094617, "全帯域_40ms_log_rms_mae_db": 0.5443026848928136, "全帯域_40ms_rms_envelope_nrmse": 0.04896199132072561, "全帯域_40ms_envelope_correlation": 0.998013076944522, "全帯域_40ms_modulation_mae_db": 0.14224851141968187, "300–1000 Hz_10ms_log_rms_mae_db": 1.6561280036307593, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10219554210744695, "300–1000 Hz_10ms_envelope_correlation": 0.9927278324563448, "300–1000 Hz_10ms_modulation_mae_db": 0.3873786904956808, "300–1000 Hz_40ms_log_rms_mae_db": 0.9399042144960954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05680066490000959, "300–1000 Hz_40ms_envelope_correlation": 0.9976352413895638, "300–1000 Hz_40ms_modulation_mae_db": 0.18854942281206344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6239499053803224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1785297482584983, "1000–3000 Hz_10ms_envelope_correlation": 0.9807189251373544, "1000–3000 Hz_10ms_modulation_mae_db": 1.8499511481270503, "1000–3000 Hz_40ms_log_rms_mae_db": 1.088904814674789, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0959761035312627, "1000–3000 Hz_40ms_envelope_correlation": 0.9942874641385636, "1000–3000 Hz_40ms_modulation_mae_db": 1.3045995699455848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3817514180217498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20305193397879132, "3000–6000 Hz_10ms_envelope_correlation": 0.972173069080589, "3000–6000 Hz_10ms_modulation_mae_db": 0.9079616777466596, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8389883927737762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12383333770235579, "3000–6000 Hz_40ms_envelope_correlation": 0.9904981529654513, "3000–6000 Hz_40ms_modulation_mae_db": 1.2698754356818027, "mr_stft_db": 5.762790483809394, "band_mae_db": 0.2838979201397922, "rms_gain_error_db": 0.014043106569149604, "output_peak": 0.5066753695210411}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9706366008365956, "全帯域_10ms_rms_envelope_nrmse": 0.10048549143998578, "全帯域_10ms_envelope_correlation": 0.9922283418446082, "全帯域_10ms_modulation_mae_db": 0.3081595589130487, "全帯域_40ms_log_rms_mae_db": 0.4919945689669679, "全帯域_40ms_rms_envelope_nrmse": 0.05066580951908806, "全帯域_40ms_envelope_correlation": 0.9981950876203479, "全帯域_40ms_modulation_mae_db": 0.37487540462134344, "300–1000 Hz_10ms_log_rms_mae_db": 1.6722770575301218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11428586362666487, "300–1000 Hz_10ms_envelope_correlation": 0.9912267120613811, "300–1000 Hz_10ms_modulation_mae_db": 0.390255038528067, "300–1000 Hz_40ms_log_rms_mae_db": 0.9216370549076154, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060824974430637044, "300–1000 Hz_40ms_envelope_correlation": 0.9978279412654909, "300–1000 Hz_40ms_modulation_mae_db": 0.47853399206232955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5121783887330433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16687574140627984, "1000–3000 Hz_10ms_envelope_correlation": 0.9829267212077962, "1000–3000 Hz_10ms_modulation_mae_db": 1.4710346826051899, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9363236062565591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09565152412398263, "1000–3000 Hz_40ms_envelope_correlation": 0.9942820378846098, "1000–3000 Hz_40ms_modulation_mae_db": 0.9548874899551899, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2962925308418114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21613678190893373, "3000–6000 Hz_10ms_envelope_correlation": 0.9684447666638787, "3000–6000 Hz_10ms_modulation_mae_db": 0.4540224237023778, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7480374099570772, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13647098668070876, "3000–6000 Hz_40ms_envelope_correlation": 0.9883963415618819, "3000–6000 Hz_40ms_modulation_mae_db": 0.5444051593398014, "mr_stft_db": 5.729709189204602, "band_mae_db": 0.39139977564777223, "rms_gain_error_db": -0.18392019291585116, "output_peak": 0.49752910743328976}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 78240, "output_samples": 78240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.647704802342586, "spectral_convergence": 0.2729619249666087, "spectral_convergence_status": "ok", "amplitude_floor": 1.658911626731477e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.606011115237456, "spectral_convergence": 0.30218262283366903, "spectral_convergence_status": "ok", "amplitude_floor": 1.4097603217110054e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.950775782884367, "spectral_convergence": 0.33251691997861343, "spectral_convergence_status": "ok", "amplitude_floor": 8.986287240444543e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.8753665955686144e-05, "output_power": 5.812643375974801e-05, "delta_db": 0.7636640107900841, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00084480798498807, "output_power": 0.000851565806757932, "delta_db": 0.034602046365843894, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015560766761226633, "output_power": 0.001449016338772345, "delta_db": -0.30957710662882465, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.758941174067266e-05, "output_power": 6.275878202905039e-05, "delta_db": 0.37331864206743615, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.7471770650556e-05, "output_power": 7.550930926719983e-05, "delta_db": -0.1114298455497498, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.273577609828158e-05, "output_power": 5.0071044935981316e-05, "delta_db": -0.22518686974993368, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.30296308685864537, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1514473123872489, "pitch": null, "output_rms_dbfs": -25.939629190356346, "output_peak": 0.49409213714798544, "reference_rms_dbfs": -25.788181877969095, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.734830566821469}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 78240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.0002230061513489049, "log_rms_mae_db": 0.9324681027842134, "rms_envelope_nrmse": 0.10223761838103024, "envelope_correlation": 0.991893591102772, "rms_gain_error_db": -0.15144749417714343, "reference_mean_power": 0.002640134243154231, "output_mean_power": 0.0025496540703936216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.2388906472962606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011616255671056531, "output_power": 0.0010854964752808895, "reference_active": true, "delta_db": -0.2943774458695022}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014996209998372379, "output_power": 0.0001377202579209164, "reference_active": true, "delta_db": -0.3698368602922625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.935590189308993e-05, "output_power": 4.014193963975775e-05, "reference_active": true, "delta_db": 0.08588484280054662}], "active_band_count": 3, "active_band_mae_db": 0.2500330496541038}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.0001943907377270206, "log_rms_mae_db": 0.5521703957003551, "rms_envelope_nrmse": 0.0662989322718726, "envelope_correlation": 0.9965590691480033, "rms_gain_error_db": -0.15144777502220225, "reference_mean_power": 0.0026564436783668454, "output_mean_power": 0.002565404398239013}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 2.080120853269114e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001087632600289277, "output_power": 0.0010101800131972199, "reference_active": true, "delta_db": -0.32083444912710635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.148356435560039e-05, "output_power": 6.685115350585279e-05, "reference_active": true, "delta_db": -0.29097294077061325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.554313443594102e-06, "output_power": 2.7204371018093036e-06, "reference_active": true, "delta_db": 0.27364500082598164}], "active_band_count": 3, "active_band_mae_db": 0.2951507969079004}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.00021647011759154595, "log_rms_mae_db": 1.569510422102213, "rms_envelope_nrmse": 0.11551786071532927, "envelope_correlation": 0.9909896225065272, "rms_gain_error_db": -0.2276445836821464, "reference_mean_power": 0.0021208071215793162, "output_mean_power": 0.002012503826080275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.0688695374548394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066233690351074, "output_power": 0.000981218720379081, "reference_active": true, "delta_db": -0.3608657565051677}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015588791655839424, "output_power": 0.0001387361716475154, "reference_active": true, "delta_db": -0.5062274654388859}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.732583628298963e-05, "output_power": 3.577420604042534e-05, "reference_active": true, "delta_db": -0.18439542879604476}], "active_band_count": 3, "active_band_mae_db": 0.3504962169133661}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.00018533448652616975, "log_rms_mae_db": 0.862215161329656, "rms_envelope_nrmse": 0.07871873843829251, "envelope_correlation": 0.9958906033219057, "rms_gain_error_db": -0.22764447278266497, "reference_mean_power": 0.0021339103768736657, "output_mean_power": 0.002024937988908263}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.9317825081344165e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001010025588563974, "output_power": 0.0009208006472642897, "reference_active": true, "delta_db": -0.40166760682729397}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.797692317548564e-05, "output_power": 7.07658730695394e-05, "reference_active": true, "delta_db": -0.4214222561702716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604806628732506e-06, "output_power": 2.452338955925096e-06, "reference_active": true, "delta_db": -0.18741003275796803}], "active_band_count": 3, "active_band_mae_db": 0.33683329858517785}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 6.255230128136302e-05, "log_rms_mae_db": 1.525215410539605, "rms_envelope_nrmse": 0.15539749052636964, "envelope_correlation": 0.9854408780519329, "rms_gain_error_db": 0.05949856145734165, "reference_mean_power": 0.0001286629346191704, "output_mean_power": 0.0001304377530561395}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.8253387298901982e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.359313427297316e-05, "output_power": 7.091579891490326e-05, "reference_active": true, "delta_db": -0.16094299693330846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.331803440637771e-06, "output_power": 9.186251392008603e-06, "reference_active": true, "delta_db": 0.9792751281345148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8895709731947531e-06, "output_power": 4.343007902489008e-06, "reference_active": true, "delta_db": 3.6142741073131757}], "active_band_count": 3, "active_band_mae_db": 1.5848307441269995}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 6.0793288183779695e-05, "log_rms_mae_db": 0.9430002652434204, "rms_envelope_nrmse": 0.07323934965367215, "envelope_correlation": 0.9966986334816268, "rms_gain_error_db": 0.059498731789344006, "reference_mean_power": 0.00012945711895499159, "output_mean_power": 0.00013124289777682264}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.7711142240683522e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.969624557508824e-05, "output_power": 6.731858603624767e-05, "reference_active": true, "delta_db": -0.150743985298756}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.240093489969041e-06, "output_power": 3.900225170740595e-06, "reference_active": true, "delta_db": 0.8053213920467678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.585555587494176e-07, "output_power": 2.2522910216142472e-07, "reference_active": true, "delta_db": 1.5244303316758216}], "active_band_count": 3, "active_band_mae_db": 0.8268319030071152}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 3.765201856958641e-05, "log_rms_mae_db": 1.3257730565389159, "rms_envelope_nrmse": 0.20328455264616835, "envelope_correlation": 0.9712882241671364, "rms_gain_error_db": -0.2726786559655472, "reference_mean_power": 2.5945814971953512e-05, "output_mean_power": 2.4366853435956095e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.30554638893879e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0948707154331005e-06, "output_power": 4.66290255304306e-06, "reference_active": true, "delta_db": -0.3847682855792136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.046923144145412e-06, "output_power": 1.14389950262069e-06, "reference_active": true, "delta_db": 0.38473070375352364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.3665503073484234e-07, "output_power": 5.210098188880522e-07, "reference_active": true, "delta_db": 0.7670743997584474}], "active_band_count": 3, "active_band_mae_db": 0.5121911296970616}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 2.0830708932641756e-05, "log_rms_mae_db": 0.779306057852905, "rms_envelope_nrmse": 0.11849609028128054, "envelope_correlation": 0.9901126179785461, "rms_gain_error_db": -0.272705775144675, "reference_mean_power": 2.6105327616819126e-05, "output_mean_power": 2.4516505668859223e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.2170827427715922e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.759181059225212e-06, "output_power": 4.389815903682803e-06, "reference_active": true, "delta_db": -0.3508591994198553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.967295837314181e-07, "output_power": 6.575419141692949e-07, "reference_active": true, "delta_db": 0.4214587243674428}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240630896276443e-08, "output_power": 3.712941236735529e-08, "reference_active": true, "delta_db": 0.5908850761038069}], "active_band_count": 3, "active_band_mae_db": 0.4544009999637017}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8649664278682989, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6d51d6f12ce88f413011e43b80272fcbff0b80c15c65604614c1e1831c37f8a9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9851706147316014, "standard_deviation": 0.002667422170084536, "minimum": 0.9807189251373544, "maximum": 0.9896501556422059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5470581601897315, "standard_deviation": 0.03909896788148102, "minimum": 1.4960416944942583, "maximum": 1.6239499053803224, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1540076500713914, "standard_deviation": 0.48894650735641093, "minimum": 0.4000604913034495, "maximum": 1.8499511481270503, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15563969811469794, "standard_deviation": 0.014252321075028116, "minimum": 0.13031050558735244, "maximum": 0.1785297482584983, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9958369563451211, "standard_deviation": 0.0012570132954414412, "minimum": 0.9942820378846098, "maximum": 0.9975177512916537, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9872649967427133, "standard_deviation": 0.0506150784557245, "minimum": 0.9363236062565591, "maximum": 1.088904814674789, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6673645730658191, "standard_deviation": 0.41917442572875685, "minimum": 0.06177811883013257, "maximum": 1.3045995699455848, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08122912740861835, "standard_deviation": 0.012309907351356472, "minimum": 0.06447101785082987, "maximum": 0.0959761035312627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9727081683578866, "standard_deviation": 0.0031292826343605057, "minimum": 0.9684447666638787, "maximum": 0.9785866975198473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3320243276415855, "standard_deviation": 0.030389882188566718, "minimum": 1.2962925308418114, "maximum": 1.3817514180217498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6096562816131739, "standard_deviation": 0.3051240375709824, "minimum": 0.31205396527176205, "maximum": 1.2075153708644413, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19904224541077273, "standard_deviation": 0.011838506747616799, "minimum": 0.1771977597240888, "maximum": 0.21613678190893373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9908345438138961, "standard_deviation": 0.0015615121448108555, "minimum": 0.9883963415618819, "maximum": 0.9925802837722193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7835592351529539, "standard_deviation": 0.038128610740687165, "minimum": 0.7218856660617138, "maximum": 0.8389883927737762, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9131612321351785, "standard_deviation": 0.4417374300719123, "minimum": 0.4544009999637017, "maximum": 1.707524747967198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1162619940228691, "standard_deviation": 0.011624048246608853, "minimum": 0.1007417699710394, "maximum": 0.13647098668070876, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9897855675676152, "standard_deviation": 0.0025841856018668213, "minimum": 0.985279671906947, "maximum": 0.9927278324563448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.642080730424284, "standard_deviation": 0.03957904099753959, "minimum": 1.569510422102213, "maximum": 1.6872907206917414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4862907135937106, "standard_deviation": 0.30177626946589564, "minimum": 0.14048518201382584, "maximum": 1.0135183783526651, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1211005088517457, "standard_deviation": 0.015034121389730313, "minimum": 0.10219554210744695, "maximum": 0.14665249727565927, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966476600797519, "standard_deviation": 0.00117632633349996, "minimum": 0.9942595794883669, "maximum": 0.9978279412654909, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9270274073451901, "standard_deviation": 0.03839347990060976, "minimum": 0.862215161329656, "maximum": 1.001701404742997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4683991021134448, "standard_deviation": 0.21158265986007035, "minimum": 0.18854942281206344, "maximum": 0.8483391667146618, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06904064259174496, "standard_deviation": 0.01207526291755521, "minimum": 0.05680066490000959, "maximum": 0.09291811949335868, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3341716721052862, "standard_deviation": 0.042918374721358316, "minimum": 0.2752039009755281, "maximum": 0.39139977564777223, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.738833044519033, "standard_deviation": 0.01305396871613694, "minimum": 5.726274609002644, "maximum": 5.762790483809394, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5184854266951868, "standard_deviation": 0.04189986689866003, "minimum": 0.4816239955161608, "maximum": 0.5951817119637042, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08017258097972083, "standard_deviation": 0.09435345353975039, "minimum": -0.1945819464444601, "maximum": 0.04250403415773076, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9911564766905099, "standard_deviation": 0.002210384758178179, "minimum": 0.9872752856432835, "maximum": 0.9937979681146347, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9843861724146735, "standard_deviation": 0.03287928721289348, "minimum": 0.9324681027842134, "maximum": 1.031179105844674, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3753873631111925, "standard_deviation": 0.21729504120468265, "minimum": 0.1271581415328843, "maximum": 0.7794596492170482, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10569675289642638, "standard_deviation": 0.01296290506988476, "minimum": 0.08884516788845508, "maximum": 0.12750035926029563, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972256524421024, "standard_deviation": 0.0009795993307358693, "minimum": 0.9952207721887917, "maximum": 0.9981950876203479, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5616575202061982, "standard_deviation": 0.03811964562506215, "minimum": 0.4919945689669679, "maximum": 0.614870334680942, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3748439746276293, "standard_deviation": 0.20552692408421921, "minimum": 0.14209619352353373, "maximum": 0.717747641679837, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.058356046692180294, "standard_deviation": 0.009817847966123587, "minimum": 0.04896199132072561, "maximum": 0.07796994315908046, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0045", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.9110057353973389, "predicted_mos": 1.9110057353973389, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "aa792600ce9066638c42cded1e968e9e9238f9ade499295b34a56f568005ee2f", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9663578998297453, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9923473082198312, "全帯域_10ms_rms_envelope_nrmse": 0.089006913426919, "全帯域_10ms_envelope_correlation": 0.9938927901723441, "全帯域_10ms_modulation_mae_db": 0.4280285900165859, "全帯域_40ms_log_rms_mae_db": 0.5681568852427196, "全帯域_40ms_rms_envelope_nrmse": 0.04353365920625151, "全帯域_40ms_envelope_correlation": 0.9986471414713922, "全帯域_40ms_modulation_mae_db": 0.8489233246209039, "300–1000 Hz_10ms_log_rms_mae_db": 1.7641496990875662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10153928763288037, "300–1000 Hz_10ms_envelope_correlation": 0.9931544529438244, "300–1000 Hz_10ms_modulation_mae_db": 0.4551403823595548, "300–1000 Hz_40ms_log_rms_mae_db": 1.0320261681455112, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05419129406754239, "300–1000 Hz_40ms_envelope_correlation": 0.9983366021856949, "300–1000 Hz_40ms_modulation_mae_db": 0.9703219464557504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8855506209760113, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1772790093177071, "1000–3000 Hz_10ms_envelope_correlation": 0.9815848717397992, "1000–3000 Hz_10ms_modulation_mae_db": 0.958670859929832, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3491915954656872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09695181965174403, "1000–3000 Hz_40ms_envelope_correlation": 0.994610646826842, "1000–3000 Hz_40ms_modulation_mae_db": 0.3796328203288779, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8725723493658848, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27918505313128394, "3000–6000 Hz_10ms_envelope_correlation": 0.9472371075525978, "3000–6000 Hz_10ms_modulation_mae_db": 1.2577414243198584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2698101632049255, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18259154504125655, "3000–6000 Hz_40ms_envelope_correlation": 0.9762172623528327, "3000–6000 Hz_40ms_modulation_mae_db": 0.8849972772061735, "mr_stft_db": 6.111062864454798, "band_mae_db": 0.27849170392230677, "rms_gain_error_db": -0.15247645929488848, "output_peak": 0.5328131989240519}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9855034009354945, "全帯域_10ms_rms_envelope_nrmse": 0.09338258360495884, "全帯域_10ms_envelope_correlation": 0.9931451058844369, "全帯域_10ms_modulation_mae_db": 0.20252071785125758, "全帯域_40ms_log_rms_mae_db": 0.5936677055975953, "全帯域_40ms_rms_envelope_nrmse": 0.0525173468642921, "全帯域_40ms_envelope_correlation": 0.9977555913152102, "全帯域_40ms_modulation_mae_db": 0.23277205850245805, "300–1000 Hz_10ms_log_rms_mae_db": 1.8191922535250473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10897333191639753, "300–1000 Hz_10ms_envelope_correlation": 0.9918472907842142, "300–1000 Hz_10ms_modulation_mae_db": 0.41714746097998984, "300–1000 Hz_40ms_log_rms_mae_db": 1.1383006934142996, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06368527103539592, "300–1000 Hz_40ms_envelope_correlation": 0.9972470278082826, "300–1000 Hz_40ms_modulation_mae_db": 0.329309223015282, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8742533954534268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20513214796115695, "1000–3000 Hz_10ms_envelope_correlation": 0.976385122235302, "1000–3000 Hz_10ms_modulation_mae_db": 2.015333384399914, "1000–3000 Hz_40ms_log_rms_mae_db": 1.373179362951651, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10872881733623616, "1000–3000 Hz_40ms_envelope_correlation": 0.9939419201789762, "1000–3000 Hz_40ms_modulation_mae_db": 1.290038117150046, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8626760508809252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2667265212841546, "3000–6000 Hz_10ms_envelope_correlation": 0.9534514715480737, "3000–6000 Hz_10ms_modulation_mae_db": 2.1827976185513767, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2605306537388297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15840926058651705, "3000–6000 Hz_40ms_envelope_correlation": 0.9832619565376991, "3000–6000 Hz_40ms_modulation_mae_db": 1.9851061862252604, "mr_stft_db": 6.138556731121149, "band_mae_db": 0.3572083423497734, "rms_gain_error_db": -0.08619184993689796, "output_peak": 0.5087647936515334}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.994794360865387, "全帯域_10ms_rms_envelope_nrmse": 0.11979775808035978, "全帯域_10ms_envelope_correlation": 0.9888040504500022, "全帯域_10ms_modulation_mae_db": 0.9686710290263184, "全帯域_40ms_log_rms_mae_db": 0.6058604335959459, "全帯域_40ms_rms_envelope_nrmse": 0.06915694918240825, "全帯域_40ms_envelope_correlation": 0.9960911012809165, "全帯域_40ms_modulation_mae_db": 1.009872884256902, "300–1000 Hz_10ms_log_rms_mae_db": 1.75967428674685, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1357540615227231, "300–1000 Hz_10ms_envelope_correlation": 0.9872000483627, "300–1000 Hz_10ms_modulation_mae_db": 1.2461670547392956, "300–1000 Hz_40ms_log_rms_mae_db": 1.0985260794370926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07495324583121504, "300–1000 Hz_40ms_envelope_correlation": 0.9959465328658926, "300–1000 Hz_40ms_modulation_mae_db": 1.1718413792111948, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9910452717353477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23925819916968605, "1000–3000 Hz_10ms_envelope_correlation": 0.9708082976659805, "1000–3000 Hz_10ms_modulation_mae_db": 2.1940747378024796, "1000–3000 Hz_40ms_log_rms_mae_db": 1.43234188445257, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1660431156342472, "1000–3000 Hz_40ms_envelope_correlation": 0.9872326881043497, "1000–3000 Hz_40ms_modulation_mae_db": 1.6519778735094874, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8989849234504983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2885429567782749, "3000–6000 Hz_10ms_envelope_correlation": 0.9422981272766525, "3000–6000 Hz_10ms_modulation_mae_db": 1.703343462779029, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3156242983311954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18358011613881678, "3000–6000 Hz_40ms_envelope_correlation": 0.9753217441097846, "3000–6000 Hz_40ms_modulation_mae_db": 1.3677955435390914, "mr_stft_db": 6.137226342096974, "band_mae_db": 0.38735517507432143, "rms_gain_error_db": -0.006866339351448, "output_peak": 0.5589640717805759}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9766996308274821, "全帯域_10ms_rms_envelope_nrmse": 0.09564073856476608, "全帯域_10ms_envelope_correlation": 0.992777053527163, "全帯域_10ms_modulation_mae_db": 0.1654481988312908, "全帯域_40ms_log_rms_mae_db": 0.5375274787174795, "全帯域_40ms_rms_envelope_nrmse": 0.04725515277003231, "全帯域_40ms_envelope_correlation": 0.9981266916983237, "全帯域_40ms_modulation_mae_db": 0.19903789098716282, "300–1000 Hz_10ms_log_rms_mae_db": 1.7516600204141284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10788639818611201, "300–1000 Hz_10ms_envelope_correlation": 0.991875961485892, "300–1000 Hz_10ms_modulation_mae_db": 0.19099151135430326, "300–1000 Hz_40ms_log_rms_mae_db": 1.0614102894895876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05563617192736029, "300–1000 Hz_40ms_envelope_correlation": 0.9977356213813785, "300–1000 Hz_40ms_modulation_mae_db": 0.24492833958791405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.899543695038504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1575041762319921, "1000–3000 Hz_10ms_envelope_correlation": 0.9848853670144321, "1000–3000 Hz_10ms_modulation_mae_db": 1.2156697376451582, "1000–3000 Hz_40ms_log_rms_mae_db": 1.399528075686321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08546771002685677, "1000–3000 Hz_40ms_envelope_correlation": 0.9955961227797919, "1000–3000 Hz_40ms_modulation_mae_db": 0.6232268162771, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8648530644854873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26806219561817507, "3000–6000 Hz_10ms_envelope_correlation": 0.9526431160578303, "3000–6000 Hz_10ms_modulation_mae_db": 1.2846772740559917, "3000–6000 Hz_40ms_log_rms_mae_db": 1.299899762255036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1662582752639727, "3000–6000 Hz_40ms_envelope_correlation": 0.9812521513210305, "3000–6000 Hz_40ms_modulation_mae_db": 1.1602098308092301, "mr_stft_db": 6.119115424974428, "band_mae_db": 0.2410723518616674, "rms_gain_error_db": -0.012202842475087919, "output_peak": 0.6254067280163368}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9873046513011657, "全帯域_10ms_rms_envelope_nrmse": 0.10366600056347658, "全帯域_10ms_envelope_correlation": 0.9916808252514497, "全帯域_10ms_modulation_mae_db": 0.44933308415909307, "全帯域_40ms_log_rms_mae_db": 0.5492783716092803, "全帯域_40ms_rms_envelope_nrmse": 0.06537482917606356, "全帯域_40ms_envelope_correlation": 0.9966975345526048, "全帯域_40ms_modulation_mae_db": 0.6405267064184067, "300–1000 Hz_10ms_log_rms_mae_db": 1.8106225026485832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13187103523447802, "300–1000 Hz_10ms_envelope_correlation": 0.9882487027797383, "300–1000 Hz_10ms_modulation_mae_db": 0.16717329335993691, "300–1000 Hz_40ms_log_rms_mae_db": 1.0890006432212553, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08915241197160967, "300–1000 Hz_40ms_envelope_correlation": 0.9947715144568886, "300–1000 Hz_40ms_modulation_mae_db": 0.6790421830678275, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8364515559322174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23618811427128972, "1000–3000 Hz_10ms_envelope_correlation": 0.9689603026619285, "1000–3000 Hz_10ms_modulation_mae_db": 2.228743294644923, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2963771867351044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14069710712130593, "1000–3000 Hz_40ms_envelope_correlation": 0.9895548800365653, "1000–3000 Hz_40ms_modulation_mae_db": 1.414576822068846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8992917901689077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2924145960110361, "3000–6000 Hz_10ms_envelope_correlation": 0.9401067507356978, "3000–6000 Hz_10ms_modulation_mae_db": 1.4230286873219846, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2824380069725978, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19045779943242358, "3000–6000 Hz_40ms_envelope_correlation": 0.9732766100571949, "3000–6000 Hz_40ms_modulation_mae_db": 1.0674454260282584, "mr_stft_db": 6.116156462020583, "band_mae_db": 0.38370283681540407, "rms_gain_error_db": -0.17883550569683354, "output_peak": 0.5205915477521827}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9623282795156545, "全帯域_10ms_rms_envelope_nrmse": 0.08975725509424562, "全帯域_10ms_envelope_correlation": 0.993658864854325, "全帯域_10ms_modulation_mae_db": 0.03404384025283734, "全帯域_40ms_log_rms_mae_db": 0.5395098373496074, "全帯域_40ms_rms_envelope_nrmse": 0.04648585015497369, "全帯域_40ms_envelope_correlation": 0.9981932928518461, "全帯域_40ms_modulation_mae_db": 0.3239589589170267, "300–1000 Hz_10ms_log_rms_mae_db": 1.759903092674404, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10377927326741805, "300–1000 Hz_10ms_envelope_correlation": 0.9924858115845419, "300–1000 Hz_10ms_modulation_mae_db": 0.28051750677656256, "300–1000 Hz_40ms_log_rms_mae_db": 1.0312539852919533, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054593770022134974, "300–1000 Hz_40ms_envelope_correlation": 0.9978138489186589, "300–1000 Hz_40ms_modulation_mae_db": 0.20925946531866865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.943266488870146, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1769019743990128, "1000–3000 Hz_10ms_envelope_correlation": 0.9809966180105116, "1000–3000 Hz_10ms_modulation_mae_db": 1.4407987553767105, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3855697622751522, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09037398875955059, "1000–3000 Hz_40ms_envelope_correlation": 0.9950638956451202, "1000–3000 Hz_40ms_modulation_mae_db": 0.3584378572371978, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8077724701554285, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25125955364662506, "3000–6000 Hz_10ms_envelope_correlation": 0.9574897386842379, "3000–6000 Hz_10ms_modulation_mae_db": 1.2673093111490512, "3000–6000 Hz_40ms_log_rms_mae_db": 1.227301370120947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16164807163149633, "3000–6000 Hz_40ms_envelope_correlation": 0.9815227707173344, "3000–6000 Hz_40ms_modulation_mae_db": 1.0771273900696101, "mr_stft_db": 6.12310816767913, "band_mae_db": 0.23909943062141678, "rms_gain_error_db": 0.013413014037742314, "output_peak": 0.6158924480498611}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9703848054876658, "全帯域_10ms_rms_envelope_nrmse": 0.10127288660609439, "全帯域_10ms_envelope_correlation": 0.9918942794802765, "全帯域_10ms_modulation_mae_db": 0.14077404083953995, "全帯域_40ms_log_rms_mae_db": 0.550579464363375, "全帯域_40ms_rms_envelope_nrmse": 0.05508878566947341, "全帯域_40ms_envelope_correlation": 0.997485423818225, "全帯域_40ms_modulation_mae_db": 0.20582226861395467, "300–1000 Hz_10ms_log_rms_mae_db": 1.8360445722194187, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12011798471611229, "300–1000 Hz_10ms_envelope_correlation": 0.9899454527425922, "300–1000 Hz_10ms_modulation_mae_db": 0.11350473595872856, "300–1000 Hz_40ms_log_rms_mae_db": 1.0687356154759573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06759794933977491, "300–1000 Hz_40ms_envelope_correlation": 0.9967293308610548, "300–1000 Hz_40ms_modulation_mae_db": 0.15169064942474705, "1000–3000 Hz_10ms_log_rms_mae_db": 1.918864820991914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16664554417691563, "1000–3000 Hz_10ms_envelope_correlation": 0.9831643392667939, "1000–3000 Hz_10ms_modulation_mae_db": 1.7991045036496314, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3792774389296518, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09191092603696174, "1000–3000 Hz_40ms_envelope_correlation": 0.9948106090527441, "1000–3000 Hz_40ms_modulation_mae_db": 0.829450797168342, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8762020993094422, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24729133152233546, "3000–6000 Hz_10ms_envelope_correlation": 0.9619550271394491, "3000–6000 Hz_10ms_modulation_mae_db": 1.8572048260321277, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2831679459613032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1538559250139826, "3000–6000 Hz_40ms_envelope_correlation": 0.9856549389287091, "3000–6000 Hz_40ms_modulation_mae_db": 1.6021010673238958, "mr_stft_db": 6.145307624439976, "band_mae_db": 0.39436688117555735, "rms_gain_error_db": -0.05765825099089823, "output_peak": 0.501763232138755}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0085199335881811, "全帯域_10ms_rms_envelope_nrmse": 0.10779607726072458, "全帯域_10ms_envelope_correlation": 0.9908545867722892, "全帯域_10ms_modulation_mae_db": 0.2594243852038772, "全帯域_40ms_log_rms_mae_db": 0.5934797325520359, "全帯域_40ms_rms_envelope_nrmse": 0.059721668718220704, "全帯域_40ms_envelope_correlation": 0.9970191117135976, "全帯域_40ms_modulation_mae_db": 0.14754409479653005, "300–1000 Hz_10ms_log_rms_mae_db": 1.8434979806252805, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11865113299610315, "300–1000 Hz_10ms_envelope_correlation": 0.9901680884258771, "300–1000 Hz_10ms_modulation_mae_db": 0.21991724716308483, "300–1000 Hz_40ms_log_rms_mae_db": 1.1621420319866853, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.066254880442883, "300–1000 Hz_40ms_envelope_correlation": 0.9967864755892796, "300–1000 Hz_40ms_modulation_mae_db": 0.09873883714487884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9542355842842793, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2180192454919117, "1000–3000 Hz_10ms_envelope_correlation": 0.9746970678376432, "1000–3000 Hz_10ms_modulation_mae_db": 1.7645889002603743, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4116633198644997, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1547823950593496, "1000–3000 Hz_40ms_envelope_correlation": 0.9879210356322838, "1000–3000 Hz_40ms_modulation_mae_db": 0.8031396066922237, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9025144740486701, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28366757324591807, "3000–6000 Hz_10ms_envelope_correlation": 0.9472968650541205, "3000–6000 Hz_10ms_modulation_mae_db": 1.822606199211324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2746344466270547, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1712580928272176, "3000–6000 Hz_40ms_envelope_correlation": 0.9802984154907844, "3000–6000 Hz_40ms_modulation_mae_db": 1.6051682070217288, "mr_stft_db": 6.135965867235275, "band_mae_db": 0.38497397114978055, "rms_gain_error_db": 0.013778344117044228, "output_peak": 0.5237391563601956}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 78240, "output_samples": 78240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0479730709144635, "spectral_convergence": 0.2834027208081071, "spectral_convergence_status": "ok", "amplitude_floor": 1.658911626731477e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.050854729143466, "spectral_convergence": 0.3058433000248013, "spectral_convergence_status": "ok", "amplitude_floor": 1.4097603217110054e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.234360793306465, "spectral_convergence": 0.3290905641928898, "spectral_convergence_status": "ok", "amplitude_floor": 8.986287240444543e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.8753665955686144e-05, "output_power": 5.6540894688975355e-05, "delta_db": 0.6435539901819248, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00084480798498807, "output_power": 0.0008492135534512221, "delta_db": 0.022589068005436946, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015560766761226633, "output_power": 0.0014429479849849343, "delta_db": -0.32780317124863795, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.758941174067266e-05, "output_power": 6.447055308132142e-05, "delta_db": 0.49018753602656034, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.7471770650556e-05, "output_power": 7.987165232592735e-05, "delta_db": 0.13249186822972486, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.273577609828158e-05, "output_power": 5.339957502729087e-05, "delta_db": 0.05432458984155556, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27849170392230677, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15247645929488848, "pitch": null, "output_rms_dbfs": -25.940658337263983, "output_peak": 0.5328131989240519, "reference_rms_dbfs": -25.788181877969095, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.111062864454798}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 78240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.0002230061513489049, "log_rms_mae_db": 0.9923473082198312, "rms_envelope_nrmse": 0.089006913426919, "envelope_correlation": 0.9938927901723441, "rms_gain_error_db": -0.1524764046912063, "reference_mean_power": 0.002640134243154231, "output_mean_power": 0.002549050089625686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.2388906472962606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011616255671056531, "output_power": 0.001124004751393326, "reference_active": true, "delta_db": -0.1429801509393707}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014996209998372379, "output_power": 0.00013067190503298476, "reference_active": true, "delta_db": -0.5979929078732733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.935590189308993e-05, "output_power": 3.472950538464363e-05, "reference_active": true, "delta_db": -0.5431127112371137}], "active_band_count": 3, "active_band_mae_db": 0.4280285900165859}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.0001943907377270206, "log_rms_mae_db": 0.5681568852427196, "rms_envelope_nrmse": 0.04353365920625151, "envelope_correlation": 0.9986471414713922, "rms_gain_error_db": -0.15247539483971484, "reference_mean_power": 0.0026564436783668454, "output_mean_power": 0.0025647974486597766}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 2.080120853269114e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001087632600289277, "output_power": 0.0010517514099044802, "reference_active": true, "delta_db": -0.14569113615255364}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.148356435560039e-05, "output_power": 6.628946795568661e-05, "reference_active": true, "delta_db": -0.327616660698702}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.554313443594102e-06, "output_power": 1.5846303566440852e-06, "reference_active": true, "delta_db": -2.073462177011456}], "active_band_count": 3, "active_band_mae_db": 0.8489233246209039}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.00021647011759154595, "log_rms_mae_db": 1.7641496990875662, "rms_envelope_nrmse": 0.10153928763288037, "envelope_correlation": 0.9931544529438244, "rms_gain_error_db": -0.23454520739836404, "reference_mean_power": 0.0021208071215793162, "output_mean_power": 0.0020093086428782147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.0688695374548394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066233690351074, "output_power": 0.0010174312776002163, "reference_active": true, "delta_db": -0.20347316573330687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015588791655839424, "output_power": 0.0001297242686944192, "reference_active": true, "delta_db": -0.7979122165355548}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.732583628298963e-05, "output_power": 3.432463836818616e-05, "reference_active": true, "delta_db": -0.3640357648098026}], "active_band_count": 3, "active_band_mae_db": 0.4551403823595548}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.00018533448652616975, "log_rms_mae_db": 1.0320261681455112, "rms_envelope_nrmse": 0.05419129406754239, "envelope_correlation": 0.9983366021856949, "rms_gain_error_db": -0.2345448257102117, "reference_mean_power": 0.0021339103768736657, "output_mean_power": 0.0020217231904702104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.9317825081344165e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001010025588563974, "output_power": 0.0009621958516255354, "reference_active": true, "delta_db": -0.2106889641603983}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.797692317548564e-05, "output_power": 6.877189847010921e-05, "reference_active": true, "delta_db": -0.5455508139199743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604806628732506e-06, "output_power": 1.5590099759259011e-06, "reference_active": true, "delta_db": -2.154726061286879}], "active_band_count": 3, "active_band_mae_db": 0.9703219464557504}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 6.255230128136302e-05, "log_rms_mae_db": 1.8855506209760113, "rms_envelope_nrmse": 0.1772790093177071, "envelope_correlation": 0.9815848717397992, "rms_gain_error_db": 0.1968513762200404, "reference_mean_power": 0.0001286629346191704, "output_mean_power": 0.00013462899085451148}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.8253387298901982e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.359313427297316e-05, "output_power": 7.377097615442726e-05, "reference_active": true, "delta_db": 0.010482306042978704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.331803440637771e-06, "output_power": 6.11656018823805e-06, "reference_active": true, "delta_db": -0.7870355974351662}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8895709731947531e-06, "output_power": 3.049387926816279e-06, "reference_active": true, "delta_db": 2.0784946763113514}], "active_band_count": 3, "active_band_mae_db": 0.958670859929832}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 6.0793288183779695e-05, "log_rms_mae_db": 1.3491915954656872, "rms_envelope_nrmse": 0.09695181965174403, "envelope_correlation": 0.994610646826842, "rms_gain_error_db": 0.19685460944606847, "reference_mean_power": 0.00012945711895499159, "output_mean_power": 0.00013546010209312736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.7711142240683522e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.969624557508824e-05, "output_power": 6.938005358532969e-05, "reference_active": true, "delta_db": -0.019747530499582604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.240093489969041e-06, "output_power": 2.6294801662955495e-06, "reference_active": true, "delta_db": -0.9068764220687178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.585555587494176e-07, "output_power": 1.6649796250772373e-07, "reference_active": true, "delta_db": 0.21227450841833348}], "active_band_count": 3, "active_band_mae_db": 0.3796328203288779}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 3.765201856958641e-05, "log_rms_mae_db": 1.8725723493658848, "rms_envelope_nrmse": 0.27918505313128394, "envelope_correlation": 0.9472371075525978, "rms_gain_error_db": 0.1478443924545213, "reference_mean_power": 2.5945814971953512e-05, "output_mean_power": 2.6844279741917733e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.30554638893879e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0948707154331005e-06, "output_power": 5.410902515214162e-06, "reference_active": true, "delta_db": 0.2613654152553234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.046923144145412e-06, "output_power": 1.3793656191331436e-06, "reference_active": true, "delta_db": 1.197645961592705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.3665503073484234e-07, "output_power": 7.439774250217426e-07, "reference_active": true, "delta_db": 2.3142128961115462}], "active_band_count": 3, "active_band_mae_db": 1.2577414243198584}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 2.0830708932641756e-05, "log_rms_mae_db": 1.2698101632049255, "rms_envelope_nrmse": 0.18259154504125655, "envelope_correlation": 0.9762172623528327, "rms_gain_error_db": 0.14787215643346283, "reference_mean_power": 2.6105327616819126e-05, "output_mean_power": 2.7009488739323967e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.2170827427715922e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.759181059225212e-06, "output_power": 5.130912508490107e-06, "reference_active": true, "delta_db": 0.3266238169940029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.967295837314181e-07, "output_power": 7.748942580698678e-07, "reference_active": true, "delta_db": 1.134648736488042}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240630896276443e-08, "output_power": 4.265809883027161e-08, "reference_active": true, "delta_db": 1.193719278136476}], "active_band_count": 3, "active_band_mae_db": 0.8849972772061735}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8364699336876389, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7a1081e33e79bdc8fb7c15b8ec550d7e3fe8d6d73472bc591a12209162f03aa3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9776852483040489, "standard_deviation": 0.005880610467897331, "minimum": 0.9689603026619285, "maximum": 0.9848853670144321, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9129014291602306, "standard_deviation": 0.04933460217263734, "minimum": 1.8364515559322174, "maximum": 1.9910452717353477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.7021230217136278, "standard_deviation": 0.4611169381526184, "minimum": 0.958670859929832, "maximum": 2.228743294644923, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19711605137745902, "standard_deviation": 0.03185560649313078, "minimum": 0.1575041762319921, "maximum": 0.23925819916968605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9923414747820842, "standard_deviation": 0.00348909718366893, "minimum": 0.9872326881043497, "maximum": 0.9955961227797919, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3783910782950797, "standard_deviation": 0.041615370657642564, "minimum": 1.2963771867351044, "maximum": 1.43234188445257, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9188100888040152, "standard_deviation": 0.4832321658159922, "minimum": 0.3584378572371978, "maximum": 1.6519778735094874, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1168694849532815, "standard_deviation": 0.032065082468133624, "minimum": 0.08546771002685677, "maximum": 0.1660431156342472, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9503097755060825, "standard_deviation": 0.007457360970936372, "minimum": 0.9401067507356978, "maximum": 0.9619550271394491, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8731084027331555, "standard_deviation": 0.030881127164224503, "minimum": 1.8077724701554285, "maximum": 1.9025144740486701, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5998386004275928, "standard_deviation": 0.34322251500275763, "minimum": 1.2577414243198584, "maximum": 2.1827976185513767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2721437226547254, "standard_deviation": 0.016741810141040047, "minimum": 0.24729133152233546, "maximum": 0.2924145960110361, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9796007311894213, "standard_deviation": 0.004254716543766037, "minimum": 0.9732766100571949, "maximum": 0.9856549389287091, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.276675830901486, "standard_deviation": 0.026407501908467412, "minimum": 1.227301370120947, "maximum": 1.3156242983311954, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3437438660279062, "standard_deviation": 0.366172010033433, "minimum": 0.8849972772061735, "maximum": 1.9851061862252604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1710073857419604, "standard_deviation": 0.013275562493126261, "minimum": 0.1538559250139826, "maximum": 0.19045779943242358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9906157261386725, "standard_deviation": 0.0021032830857291712, "minimum": 0.9872000483627, "maximum": 0.9931544529438244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7930930509926597, "standard_deviation": 0.03807275166320074, "minimum": 1.7516600204141284, "maximum": 1.8434979806252805, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38631989908643205, "standard_deviation": 0.3673290164821204, "minimum": 0.11350473595872856, "maximum": 1.2461670547392956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11607156318402806, "standard_deviation": 0.01275998770049072, "minimum": 0.10153928763288037, "maximum": 0.1357540615227231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969208692583913, "standard_deviation": 0.001146156240977643, "minimum": 0.9947715144568886, "maximum": 0.9983366021856949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0851744383077928, "standard_deviation": 0.04706597066591388, "minimum": 1.0312539852919533, "maximum": 1.1621420319866853, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48189150290328286, "standard_deviation": 0.4076173293724354, "minimum": 0.09873883714487884, "maximum": 1.1718413792111948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06575812432973953, "standard_deviation": 0.011956227827472904, "minimum": 0.05419129406754239, "maximum": 0.08915241197160967, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3332838366212785, "standard_deviation": 0.06847002847695656, "minimum": 0.23909943062141678, "maximum": 0.39436688117555735, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.128312435502789, "standard_deviation": 0.012471679842516665, "minimum": 6.111062864454798, "maximum": 6.145307624439976, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5484918970841866, "standard_deviation": 0.04776004314366643, "minimum": 0.501763232138755, "maximum": 0.6254067280163368, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.058379986198908454, "standard_deviation": 0.07491399165890138, "minimum": -0.17883550569683354, "maximum": 0.013778344117044228, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920884445490359, "standard_deviation": 0.0016829439232997588, "minimum": 0.9888040504500022, "maximum": 0.9938927901723441, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9847352963426077, "standard_deviation": 0.014660982979599495, "minimum": 0.9623282795156545, "maximum": 1.0085199335881811, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3310304857726001, "standard_deviation": 0.29361001094917855, "minimum": 0.03404384025283734, "maximum": 0.9686710290263184, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10004002665019311, "standard_deviation": 0.010394587634636085, "minimum": 0.089006913426919, "maximum": 0.11979775808035978, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975019860877645, "standard_deviation": 0.0008548215174261465, "minimum": 0.9960911012809165, "maximum": 0.9986471414713922, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5672574886285049, "standard_deviation": 0.02707479770179727, "minimum": 0.5375274787174795, "maximum": 0.6058604335959459, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4510572733891681, "standard_deviation": 0.3351007504270116, "minimum": 0.14754409479653005, "maximum": 1.009872884256902, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054891780217714445, "standard_deviation": 0.009261121351303712, "minimum": 0.04353365920625151, "maximum": 0.06915694918240825, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0045", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.0089714527130127, "predicted_mos": 2.0089714527130127, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "9c03ab0cbd9b808032496579a26823bac00a450430caaf8d668d23c3899d9b6f", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.123349600005895, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9907370205616333, "全帯域_10ms_rms_envelope_nrmse": 0.0947968092424582, "全帯域_10ms_envelope_correlation": 0.9930298045705465, "全帯域_10ms_modulation_mae_db": 0.24378818034447125, "全帯域_40ms_log_rms_mae_db": 0.5570980689225629, "全帯域_40ms_rms_envelope_nrmse": 0.057570489691644154, "全帯域_40ms_envelope_correlation": 0.9974253445239568, "全帯域_40ms_modulation_mae_db": 0.4821944846257947, "300–1000 Hz_10ms_log_rms_mae_db": 1.6336762913669207, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10868135586321892, "300–1000 Hz_10ms_envelope_correlation": 0.9920221993846253, "300–1000 Hz_10ms_modulation_mae_db": 0.382373164474944, "300–1000 Hz_40ms_log_rms_mae_db": 0.9173559154201519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07050174202848498, "300–1000 Hz_40ms_envelope_correlation": 0.9967369961185277, "300–1000 Hz_40ms_modulation_mae_db": 0.6083703532170495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.529682093028693, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1894225125433346, "1000–3000 Hz_10ms_envelope_correlation": 0.978745149788268, "1000–3000 Hz_10ms_modulation_mae_db": 1.8513732623719177, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0248742090622436, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10074095901650536, "1000–3000 Hz_40ms_envelope_correlation": 0.9939667982451026, "1000–3000 Hz_40ms_modulation_mae_db": 1.172968336060408, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2567926626970378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17765708393753576, "3000–6000 Hz_10ms_envelope_correlation": 0.9787427953001584, "3000–6000 Hz_10ms_modulation_mae_db": 0.7569347046721533, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7448110135574559, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10650548474575522, "3000–6000 Hz_40ms_envelope_correlation": 0.9920194133547104, "3000–6000 Hz_40ms_modulation_mae_db": 1.3442538710766858, "mr_stft_db": 5.713780995874327, "band_mae_db": 0.2476692942705725, "rms_gain_error_db": -0.14847718457320247, "output_peak": 0.5075121158851391}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0130847533437664, "全帯域_10ms_rms_envelope_nrmse": 0.09216506328846634, "全帯域_10ms_envelope_correlation": 0.9933857103381705, "全帯域_10ms_modulation_mae_db": 0.4369512728346577, "全帯域_40ms_log_rms_mae_db": 0.5793741313921649, "全帯域_40ms_rms_envelope_nrmse": 0.053574653847934924, "全帯域_40ms_envelope_correlation": 0.9977540930678334, "全帯域_40ms_modulation_mae_db": 0.34400866354343634, "300–1000 Hz_10ms_log_rms_mae_db": 1.5638940445252885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10612556617180119, "300–1000 Hz_10ms_envelope_correlation": 0.9923128559591665, "300–1000 Hz_10ms_modulation_mae_db": 0.6640535284120074, "300–1000 Hz_40ms_log_rms_mae_db": 0.9113378525468893, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06358133171479576, "300–1000 Hz_40ms_envelope_correlation": 0.9973009906229792, "300–1000 Hz_40ms_modulation_mae_db": 0.46689980691726385, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4763839036745445, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15650254534482336, "1000–3000 Hz_10ms_envelope_correlation": 0.985592891560465, "1000–3000 Hz_10ms_modulation_mae_db": 1.3121639532400475, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0092632001964508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08866166226295998, "1000–3000 Hz_40ms_envelope_correlation": 0.9954071560376833, "1000–3000 Hz_40ms_modulation_mae_db": 0.8045279337483328, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3184710581393033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2047274799110919, "3000–6000 Hz_10ms_envelope_correlation": 0.974287855911775, "3000–6000 Hz_10ms_modulation_mae_db": 0.8926956083278758, "3000–6000 Hz_40ms_log_rms_mae_db": 0.788418339736771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13772733381968405, "3000–6000 Hz_40ms_envelope_correlation": 0.9884975367783727, "3000–6000 Hz_40ms_modulation_mae_db": 0.8460295586856706, "mr_stft_db": 5.727795683063405, "band_mae_db": 0.2580469417491637, "rms_gain_error_db": -0.13096454692362405, "output_peak": 0.5586942048820697}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0177949791254093, "全帯域_10ms_rms_envelope_nrmse": 0.11206717933052171, "全帯域_10ms_envelope_correlation": 0.9901748004991624, "全帯域_10ms_modulation_mae_db": 0.17935565014958596, "全帯域_40ms_log_rms_mae_db": 0.5740256255453918, "全帯域_40ms_rms_envelope_nrmse": 0.062388562054174396, "全帯域_40ms_envelope_correlation": 0.9969573326079703, "全帯域_40ms_modulation_mae_db": 0.5467600788361576, "300–1000 Hz_10ms_log_rms_mae_db": 1.5795279785937222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1285857850477151, "300–1000 Hz_10ms_envelope_correlation": 0.9886874691135377, "300–1000 Hz_10ms_modulation_mae_db": 0.41117951568287364, "300–1000 Hz_40ms_log_rms_mae_db": 0.9800117030986828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07805950338344536, "300–1000 Hz_40ms_envelope_correlation": 0.9959264233879717, "300–1000 Hz_40ms_modulation_mae_db": 0.5278441733011731, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4523538321195646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14597382861363586, "1000–3000 Hz_10ms_envelope_correlation": 0.9871441615909338, "1000–3000 Hz_10ms_modulation_mae_db": 1.1028128057017932, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9354730246670839, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0697332151587432, "1000–3000 Hz_40ms_envelope_correlation": 0.9970098814268776, "1000–3000 Hz_40ms_modulation_mae_db": 0.46964656790669984, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2455030613024531, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20100364675032906, "3000–6000 Hz_10ms_envelope_correlation": 0.9718679981790431, "3000–6000 Hz_10ms_modulation_mae_db": 0.42403206355413975, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7143147885754896, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11411471220276652, "3000–6000 Hz_40ms_envelope_correlation": 0.9905885841657264, "3000–6000 Hz_40ms_modulation_mae_db": 1.0667979681193998, "mr_stft_db": 5.738119889811834, "band_mae_db": 0.5213265718966033, "rms_gain_error_db": -0.14352368905825205, "output_peak": 0.7197518371195368}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9954235561386149, "全帯域_10ms_rms_envelope_nrmse": 0.0924108299234702, "全帯域_10ms_envelope_correlation": 0.9933021567963611, "全帯域_10ms_modulation_mae_db": 0.5409534686575083, "全帯域_40ms_log_rms_mae_db": 0.5563093008028395, "全帯域_40ms_rms_envelope_nrmse": 0.04795679953566429, "全帯域_40ms_envelope_correlation": 0.9981759163194821, "全帯域_40ms_modulation_mae_db": 0.3254130599271738, "300–1000 Hz_10ms_log_rms_mae_db": 1.5565183407959833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10564045973191627, "300–1000 Hz_10ms_envelope_correlation": 0.9922947414986475, "300–1000 Hz_10ms_modulation_mae_db": 0.7355749561753555, "300–1000 Hz_40ms_log_rms_mae_db": 0.9416350848404441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05597904985098408, "300–1000 Hz_40ms_envelope_correlation": 0.9978821810036453, "300–1000 Hz_40ms_modulation_mae_db": 0.31867559068436496, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4563477078109333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15675802262425056, "1000–3000 Hz_10ms_envelope_correlation": 0.9854637043275537, "1000–3000 Hz_10ms_modulation_mae_db": 1.30051541526863, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9673795810491318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09320885481348816, "1000–3000 Hz_40ms_envelope_correlation": 0.9948406452858759, "1000–3000 Hz_40ms_modulation_mae_db": 0.7940335553159045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2970343633409624, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18363765577124547, "3000–6000 Hz_10ms_envelope_correlation": 0.9765948518804458, "3000–6000 Hz_10ms_modulation_mae_db": 0.5506841538949754, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7596349678419223, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0950206638075051, "3000–6000 Hz_40ms_envelope_correlation": 0.9936096272114271, "3000–6000 Hz_40ms_modulation_mae_db": 1.190799129001151, "mr_stft_db": 5.722115466305884, "band_mae_db": 0.24901340945680617, "rms_gain_error_db": -0.06674355788713895, "output_peak": 0.49544712871393176}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9772987439598847, "全帯域_10ms_rms_envelope_nrmse": 0.10060900762463001, "全帯域_10ms_envelope_correlation": 0.9921410387073152, "全帯域_10ms_modulation_mae_db": 0.6459992244101619, "全帯域_40ms_log_rms_mae_db": 0.556232406253967, "全帯域_40ms_rms_envelope_nrmse": 0.04629912052276442, "全帯域_40ms_envelope_correlation": 0.9984753935578898, "全帯域_40ms_modulation_mae_db": 0.3885846043559815, "300–1000 Hz_10ms_log_rms_mae_db": 1.5591384267060162, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11477520693723213, "300–1000 Hz_10ms_envelope_correlation": 0.9911023978451651, "300–1000 Hz_10ms_modulation_mae_db": 0.798274792399606, "300–1000 Hz_40ms_log_rms_mae_db": 0.8878672930513722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05442145383303582, "300–1000 Hz_40ms_envelope_correlation": 0.9983449382435945, "300–1000 Hz_40ms_modulation_mae_db": 0.4145050972402739, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5529906172773722, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1655392235764793, "1000–3000 Hz_10ms_envelope_correlation": 0.9831811565158459, "1000–3000 Hz_10ms_modulation_mae_db": 1.1587291475217558, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0829410558344843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0907102102863657, "1000–3000 Hz_40ms_envelope_correlation": 0.9949342300014723, "1000–3000 Hz_40ms_modulation_mae_db": 0.586489803879551, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2499430466049029, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21812383796699672, "3000–6000 Hz_10ms_envelope_correlation": 0.9670718738127999, "3000–6000 Hz_10ms_modulation_mae_db": 0.14744886881952193, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7066677614033106, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14242518762665912, "3000–6000 Hz_40ms_envelope_correlation": 0.9859516866912937, "3000–6000 Hz_40ms_modulation_mae_db": 0.6731002220495647, "mr_stft_db": 5.710326891170586, "band_mae_db": 0.3663572702748413, "rms_gain_error_db": -0.159349376393875, "output_peak": 0.4778966896973261}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0030922415219554, "全帯域_10ms_rms_envelope_nrmse": 0.10490272792325193, "全帯域_10ms_envelope_correlation": 0.9915170521513345, "全帯域_10ms_modulation_mae_db": 0.36934414812384087, "全帯域_40ms_log_rms_mae_db": 0.5917585944006424, "全帯域_40ms_rms_envelope_nrmse": 0.0615672444705253, "全帯域_40ms_envelope_correlation": 0.9971794468068855, "全帯域_40ms_modulation_mae_db": 0.36481265686920983, "300–1000 Hz_10ms_log_rms_mae_db": 1.6341735553578756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1266484665267382, "300–1000 Hz_10ms_envelope_correlation": 0.9891833988990348, "300–1000 Hz_10ms_modulation_mae_db": 0.58507334982006, "300–1000 Hz_40ms_log_rms_mae_db": 0.9381650907593312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07628081899331647, "300–1000 Hz_40ms_envelope_correlation": 0.9963473112159421, "300–1000 Hz_40ms_modulation_mae_db": 0.47519454616190565, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5286941216326755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1664928236721978, "1000–3000 Hz_10ms_envelope_correlation": 0.9833034324150894, "1000–3000 Hz_10ms_modulation_mae_db": 1.4549194428891044, "1000–3000 Hz_40ms_log_rms_mae_db": 1.032909834566313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09863298178416696, "1000–3000 Hz_40ms_envelope_correlation": 0.9940117278420851, "1000–3000 Hz_40ms_modulation_mae_db": 0.5289249301113007, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2234557296722486, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18112782937088734, "3000–6000 Hz_10ms_envelope_correlation": 0.9781814003722182, "3000–6000 Hz_10ms_modulation_mae_db": 0.27483200472928865, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7250293048267401, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1080021786275304, "3000–6000 Hz_40ms_envelope_correlation": 0.9919170511768859, "3000–6000 Hz_40ms_modulation_mae_db": 0.4641332193981292, "mr_stft_db": 5.719358179694109, "band_mae_db": 0.2966799000798485, "rms_gain_error_db": -0.1937738055103353, "output_peak": 0.5146112399458422}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9920393274414728, "全帯域_10ms_rms_envelope_nrmse": 0.09471321280460197, "全帯域_10ms_envelope_correlation": 0.9931561929551874, "全帯域_10ms_modulation_mae_db": 0.23023014041768364, "全帯域_40ms_log_rms_mae_db": 0.5796777363905931, "全帯域_40ms_rms_envelope_nrmse": 0.05474518247171035, "全帯域_40ms_envelope_correlation": 0.9978612315714666, "全帯域_40ms_modulation_mae_db": 0.3435873989707523, "300–1000 Hz_10ms_log_rms_mae_db": 1.5736760201128595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11192926245532266, "300–1000 Hz_10ms_envelope_correlation": 0.9917301078451912, "300–1000 Hz_10ms_modulation_mae_db": 0.3008938539843087, "300–1000 Hz_40ms_log_rms_mae_db": 0.9300031641520947, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06680788712185794, "300–1000 Hz_40ms_envelope_correlation": 0.9973987583479585, "300–1000 Hz_40ms_modulation_mae_db": 0.3839772669648949, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5207315833154158, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17190470628235122, "1000–3000 Hz_10ms_envelope_correlation": 0.9818542674902709, "1000–3000 Hz_10ms_modulation_mae_db": 1.7244234494017832, "1000–3000 Hz_40ms_log_rms_mae_db": 1.01946578494129, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09225210015888541, "1000–3000 Hz_40ms_envelope_correlation": 0.9947449069393974, "1000–3000 Hz_40ms_modulation_mae_db": 0.9011776833461221, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2583600921799398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20932613198127384, "3000–6000 Hz_10ms_envelope_correlation": 0.9700405492235711, "3000–6000 Hz_10ms_modulation_mae_db": 0.34006822761116656, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7645238547277264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13745627045910028, "3000–6000 Hz_40ms_envelope_correlation": 0.9862562845083569, "3000–6000 Hz_40ms_modulation_mae_db": 0.5289046054235215, "mr_stft_db": 5.7388193214685, "band_mae_db": 0.21914015348311955, "rms_gain_error_db": -0.18574581656516306, "output_peak": 0.5332738476521486}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9476461903898885, "全帯域_10ms_rms_envelope_nrmse": 0.10195776153391885, "全帯域_10ms_envelope_correlation": 0.9918959729459401, "全帯域_10ms_modulation_mae_db": 0.5119119175598451, "全帯域_40ms_log_rms_mae_db": 0.5382099437553246, "全帯域_40ms_rms_envelope_nrmse": 0.061804501203969235, "全帯域_40ms_envelope_correlation": 0.9968622848532125, "全帯域_40ms_modulation_mae_db": 0.3763909050077436, "300–1000 Hz_10ms_log_rms_mae_db": 1.5643265008474594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11499421476033973, "300–1000 Hz_10ms_envelope_correlation": 0.9908256948045273, "300–1000 Hz_10ms_modulation_mae_db": 0.6647227790998042, "300–1000 Hz_40ms_log_rms_mae_db": 0.915394137658989, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06871960977620994, "300–1000 Hz_40ms_envelope_correlation": 0.9965420601161494, "300–1000 Hz_40ms_modulation_mae_db": 0.5213667647481185, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5455121088825379, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18992592027897384, "1000–3000 Hz_10ms_envelope_correlation": 0.979054749126523, "1000–3000 Hz_10ms_modulation_mae_db": 1.9328321725758866, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9772359092779833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08681130385897093, "1000–3000 Hz_40ms_envelope_correlation": 0.9958304397049397, "1000–3000 Hz_40ms_modulation_mae_db": 1.1472209212365605, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2645495861659606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21230935607053114, "3000–6000 Hz_10ms_envelope_correlation": 0.9688303508754267, "3000–6000 Hz_10ms_modulation_mae_db": 0.41835800603252354, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7185645965017411, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11672762902181967, "3000–6000 Hz_40ms_envelope_correlation": 0.990956348776073, "3000–6000 Hz_40ms_modulation_mae_db": 0.286398000172567, "mr_stft_db": 5.736329476220533, "band_mae_db": 0.2872758252240914, "rms_gain_error_db": 0.05175865652351441, "output_peak": 0.5178459725122114}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 78240, "output_samples": 78240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.558107523678787, "spectral_convergence": 0.2711031631398995, "spectral_convergence_status": "ok", "amplitude_floor": 1.658911626731477e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.58864263905125, "spectral_convergence": 0.280975383244128, "spectral_convergence_status": "ok", "amplitude_floor": 1.4097603217110054e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.994592824892944, "spectral_convergence": 0.3122260694979353, "spectral_convergence_status": "ok", "amplitude_floor": 8.986287240444543e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.8753665955686144e-05, "output_power": 5.6092454397672475e-05, "delta_db": 0.6089716618341801, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00084480798498807, "output_power": 0.0008096747825798688, "delta_db": -0.18447396758093032, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015560766761226633, "output_power": 0.0014898545362499569, "delta_db": -0.18887125550546335, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.758941174067266e-05, "output_power": 6.149334835333538e-05, "delta_db": 0.2848549895021584, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.7471770650556e-05, "output_power": 7.993593191314882e-05, "delta_db": 0.13598560373267346, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.273577609828158e-05, "output_power": 5.1739176619061004e-05, "delta_db": -0.08285828746802915, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2476692942705725, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14847718457320247, "pitch": null, "output_rms_dbfs": -25.9366590625423, "output_peak": 0.5075121158851391, "reference_rms_dbfs": -25.788181877969095, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.713780995874327}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 78240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.0002230061513489049, "log_rms_mae_db": 0.9907370205616333, "rms_envelope_nrmse": 0.0947968092424582, "envelope_correlation": 0.9930298045705465, "rms_gain_error_db": -0.14847739638316115, "reference_mean_power": 0.002640134243154231, "output_mean_power": 0.0025513983504366466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.2388906472962606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011616255671056531, "output_power": 0.001107999149614948, "reference_active": true, "delta_db": -0.205267351074563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014996209998372379, "output_power": 0.00013409342502814224, "reference_active": true, "delta_db": -0.48574029691451553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.935590189308993e-05, "output_power": 3.97233218052744e-05, "reference_active": true, "delta_db": 0.040356893044335165}], "active_band_count": 3, "active_band_mae_db": 0.24378818034447125}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.0001943907377270206, "log_rms_mae_db": 0.5570980689225629, "rms_envelope_nrmse": 0.057570489691644154, "envelope_correlation": 0.9974253445239568, "rms_gain_error_db": -0.14847653858514057, "reference_mean_power": 0.0026564436783668454, "output_mean_power": 0.0025671601265261747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 2.080120853269114e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001087632600289277, "output_power": 0.0010354912246107518, "reference_active": true, "delta_db": -0.2133579372290194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.148356435560039e-05, "output_power": 6.49456034415954e-05, "reference_active": true, "delta_db": -0.4165644302152459}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.554313443594102e-06, "output_power": 2.1164503290382115e-06, "reference_active": true, "delta_db": -0.8166610864331186}], "active_band_count": 3, "active_band_mae_db": 0.4821944846257947}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.00021647011759154595, "log_rms_mae_db": 1.6336762913669207, "rms_envelope_nrmse": 0.10868135586321892, "envelope_correlation": 0.9920221993846253, "rms_gain_error_db": -0.22073958020437423, "reference_mean_power": 0.0021208071215793162, "output_mean_power": 0.0020157061230247886}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.0688695374548394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066233690351074, "output_power": 0.0010012823432788377, "reference_active": true, "delta_db": -0.272958432016716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015588791655839424, "output_power": 0.00013555554699156935, "reference_active": true, "delta_db": -0.6069515894442044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.732583628298963e-05, "output_power": 3.969451445852101e-05, "reference_active": true, "delta_db": 0.26720947196391187}], "active_band_count": 3, "active_band_mae_db": 0.382373164474944}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.00018533448652616975, "log_rms_mae_db": 0.9173559154201519, "rms_envelope_nrmse": 0.07050174202848498, "envelope_correlation": 0.9967369961185277, "rms_gain_error_db": -0.22073928061031953, "reference_mean_power": 0.0021339103768736657, "output_mean_power": 0.0020281601592185383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.9317825081344165e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001010025588563974, "output_power": 0.0009461202297537238, "reference_active": true, "delta_db": -0.2838604800574014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.797692317548564e-05, "output_power": 6.809818517834678e-05, "reference_active": true, "delta_db": -0.5883055664960889}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604806628732506e-06, "output_power": 2.0560184096066396e-06, "reference_active": true, "delta_db": -0.9529450130976581}], "active_band_count": 3, "active_band_mae_db": 0.6083703532170495}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 6.255230128136302e-05, "log_rms_mae_db": 1.529682093028693, "rms_envelope_nrmse": 0.1894225125433346, "envelope_correlation": 0.978745149788268, "rms_gain_error_db": 0.15052608396392994, "reference_mean_power": 0.0001286629346191704, "output_mean_power": 0.00013320056322719908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.8253387298901982e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.359313427297316e-05, "output_power": 7.350950063150049e-05, "reference_active": true, "delta_db": -0.004938270878715883}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.331803440637771e-06, "output_power": 5.954885647329483e-06, "reference_active": true, "delta_db": -0.9033738731475338}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8895709731947531e-06, "output_power": 5.507366432017329e-06, "reference_active": true, "delta_db": 4.645807643089503}], "active_band_count": 3, "active_band_mae_db": 1.8513732623719177}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 6.0793288183779695e-05, "log_rms_mae_db": 1.0248742090622436, "rms_envelope_nrmse": 0.10074095901650536, "envelope_correlation": 0.9939667982451026, "rms_gain_error_db": 0.15052965744708627, "reference_mean_power": 0.00012945711895499159, "output_mean_power": 0.0001340228667892368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.7711142240683522e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.969624557508824e-05, "output_power": 7.013704070925171e-05, "reference_active": true, "delta_db": 0.02738053772455156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.240093489969041e-06, "output_power": 2.740019119877356e-06, "reference_active": true, "delta_db": -0.7280394822048426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.585555587494176e-07, "output_power": 2.995918337664502e-07, "reference_active": true, "delta_db": 2.7634849882518298}], "active_band_count": 3, "active_band_mae_db": 1.172968336060408}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 3.765201856958641e-05, "log_rms_mae_db": 1.2567926626970378, "rms_envelope_nrmse": 0.17765708393753576, "envelope_correlation": 0.9787427953001584, "rms_gain_error_db": 0.11854983665944312, "reference_mean_power": 2.5945814971953512e-05, "output_mean_power": 2.666381579266989e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.30554638893879e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0948707154331005e-06, "output_power": 5.392827079715377e-06, "reference_active": true, "delta_db": 0.24683326846152667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.046923144145412e-06, "output_power": 9.597293528217897e-07, "reference_active": true, "delta_db": -0.37766023105144064}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.3665503073484234e-07, "output_power": 2.9888726198158637e-07, "reference_active": true, "delta_db": -1.6463106145034925}], "active_band_count": 3, "active_band_mae_db": 0.7569347046721533}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 2.0830708932641756e-05, "log_rms_mae_db": 0.7448110135574559, "rms_envelope_nrmse": 0.10650548474575522, "envelope_correlation": 0.9920194133547104, "rms_gain_error_db": 0.11857074525930557, "reference_mean_power": 2.6105327616819126e-05, "output_mean_power": 2.6827871804332534e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.2170827427715922e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.759181059225212e-06, "output_power": 5.021086029922081e-06, "reference_active": true, "delta_db": 0.23265435023405168}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.967295837314181e-07, "output_power": 5.287004365827153e-07, "reference_active": true, "delta_db": -0.52567900123573}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240630896276443e-08, "output_power": 1.5247082214562032e-08, "reference_active": true, "delta_db": -3.274428261760276}], "active_band_count": 3, "active_band_mae_db": 1.3442538710766858}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8538753087112633, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0aba1794013c1a45f29e9dbd8936efea4fd79c6483a50516c41b6051e982f4ac", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9830424391018688, "standard_deviation": 0.0030486282397012197, "minimum": 0.978745149788268, "maximum": 0.9871441615909338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5078369959677171, "standard_deviation": 0.040099140123112505, "minimum": 1.4523538321195646, "maximum": 1.5529906172773722, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.479721206121365, "standard_deviation": 0.31830244077464076, "minimum": 1.1028128057017932, "maximum": 1.9328321725758866, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1678149478670058, "standard_deviation": 0.015614533278149675, "minimum": 0.14597382861363586, "maximum": 0.18992592027897384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950932231854293, "standard_deviation": 0.0009974599745551965, "minimum": 0.9939667982451026, "maximum": 0.9970098814268776, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0061928249493726, "standard_deviation": 0.04548931326550378, "minimum": 0.9354730246670839, "maximum": 1.0829410558344843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8006237164506098, "standard_deviation": 0.2667688532317855, "minimum": 0.46964656790669984, "maximum": 1.172968336060408, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0900939109175107, "standard_deviation": 0.009474783244917137, "minimum": 0.0697332151587432, "maximum": 0.10074095901650536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9732022094444297, "standard_deviation": 0.004418297380980087, "minimum": 0.9670718738127999, "maximum": 0.9787427953001584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2642637000128512, "standard_deviation": 0.030060505382829027, "minimum": 1.2234557296722486, "maximum": 1.3184710581393033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4756317047052056, "standard_deviation": 0.2481506733399659, "minimum": 0.14744886881952193, "maximum": 0.8926956083278758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1984891277199864, "standard_deviation": 0.015561656224209049, "minimum": 0.17765708393753576, "maximum": 0.21812383796699672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9899745665828558, "standard_deviation": 0.002795405046285674, "minimum": 0.9859516866912937, "maximum": 0.9936096272114271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7402455783963946, "standard_deviation": 0.02880554725861328, "minimum": 0.7066677614033106, "maximum": 0.788418339736771, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8000520717408361, "standard_deviation": 0.3757708011401886, "minimum": 0.286398000172567, "maximum": 1.3442538710766858, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11974743253885255, "standard_deviation": 0.017389325053576953, "minimum": 0.0950206638075051, "maximum": 0.14242518762665912, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991019858168737, "standard_deviation": 0.0013971116568112848, "minimum": 0.9886874691135377, "maximum": 0.9923128559591665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5831163947882656, "standard_deviation": 0.0322270793094911, "minimum": 1.5565183407959833, "maximum": 1.6341735553578756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5677682425061199, "standard_deviation": 0.18141356519553808, "minimum": 0.3008938539843087, "maximum": 0.798274792399606, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11467253968678552, "standard_deviation": 0.008743685997172153, "minimum": 0.10564045973191627, "maximum": 0.1285857850477151, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970599573820961, "standard_deviation": 0.0008163854171845395, "minimum": 0.9959264233879717, "maximum": 0.9983449382435945, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9277212801909944, "standard_deviation": 0.02714595124617765, "minimum": 0.8878672930513722, "maximum": 0.9800117030986828, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4646041999043805, "standard_deviation": 0.09127133525801383, "minimum": 0.31867559068436496, "maximum": 0.6083703532170495, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06679392458776628, "standard_deviation": 0.008575139322076618, "minimum": 0.05442145383303582, "maximum": 0.07805950338344536, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3056886708043808, "standard_deviation": 0.09778934186853182, "minimum": 0.21914015348311955, "maximum": 0.5213265718966033, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7258307379511475, "standard_deviation": 0.011180811832825798, "minimum": 5.710326891170586, "maximum": 5.7388193214685, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5406291295510257, "standard_deviation": 0.07628415217663223, "minimum": 0.4778966896973261, "maximum": 0.7197518371195368, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12210241504850956, "standard_deviation": 0.08026733813615164, "minimum": -0.1937738055103353, "maximum": 0.05175865652351441, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9923253411205022, "standard_deviation": 0.001118775842260152, "minimum": 0.9901748004991624, "maximum": 0.9933857103381705, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9921396015603281, "standard_deviation": 0.0221344412988791, "minimum": 0.9476461903898885, "maximum": 1.0177949791254093, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.39481675031221936, "standard_deviation": 0.16771825531964354, "minimum": 0.17935565014958596, "maximum": 0.6459992244101619, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0992028239589149, "standard_deviation": 0.0069977385133049505, "minimum": 0.09216506328846634, "maximum": 0.11206717933052171, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975863804135872, "standard_deviation": 0.0005797919194211259, "minimum": 0.9968622848532125, "maximum": 0.9984753935578898, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5665857259329358, "standard_deviation": 0.017451836811702415, "minimum": 0.5382099437553246, "maximum": 0.5917585944006424, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3964689815170312, "standard_deviation": 0.077460756878979, "minimum": 0.3254130599271738, "maximum": 0.5467600788361576, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05573831922479838, "standard_deviation": 0.006252722845930766, "minimum": 0.04629912052276442, "maximum": 0.062388562054174396, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0045", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.0859761238098145, "predicted_mos": 2.0859761238098145, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "4aba4be5489bbb375f00a1cd0462d73f01ffa81493749a0b34a542f14da3934c", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.159411699976772, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9303341774932202, "全帯域_10ms_rms_envelope_nrmse": 0.09077767195416755, "全帯域_10ms_envelope_correlation": 0.9936204087445094, "全帯域_10ms_modulation_mae_db": 0.3461138233160425, "全帯域_40ms_log_rms_mae_db": 0.5417391932483494, "全帯域_40ms_rms_envelope_nrmse": 0.055256424672805345, "全帯域_40ms_envelope_correlation": 0.9976294400613716, "全帯域_40ms_modulation_mae_db": 0.40868069338733176, "300–1000 Hz_10ms_log_rms_mae_db": 1.5790742942053513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10206082767397348, "300–1000 Hz_10ms_envelope_correlation": 0.9930451006613001, "300–1000 Hz_10ms_modulation_mae_db": 0.5378101156090119, "300–1000 Hz_40ms_log_rms_mae_db": 0.8821719471320489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06696339473055045, "300–1000 Hz_40ms_envelope_correlation": 0.9971283780930403, "300–1000 Hz_40ms_modulation_mae_db": 0.6603217619344423, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5362629098130223, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16585097854886485, "1000–3000 Hz_10ms_envelope_correlation": 0.984310220524849, "1000–3000 Hz_10ms_modulation_mae_db": 1.7597499926079265, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9937245724728766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08583858568569004, "1000–3000 Hz_40ms_envelope_correlation": 0.9960986337858669, "1000–3000 Hz_40ms_modulation_mae_db": 1.016191698147881, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4702862342251986, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21637429238952421, "3000–6000 Hz_10ms_envelope_correlation": 0.9674570069794305, "3000–6000 Hz_10ms_modulation_mae_db": 0.935222264556503, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8979472950310868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12191636934699913, "3000–6000 Hz_40ms_envelope_correlation": 0.9892197545008102, "3000–6000 Hz_40ms_modulation_mae_db": 0.7807515638727237, "mr_stft_db": 5.820755533914213, "band_mae_db": 0.3369834700032884, "rms_gain_error_db": -0.12738627541031974, "output_peak": 0.4975397271403047}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0302316545344765, "全帯域_10ms_rms_envelope_nrmse": 0.1221162645440631, "全帯域_10ms_envelope_correlation": 0.9882479165876339, "全帯域_10ms_modulation_mae_db": 0.47647864379595145, "全帯域_40ms_log_rms_mae_db": 0.6032130535203757, "全帯域_40ms_rms_envelope_nrmse": 0.07062005389093592, "全帯域_40ms_envelope_correlation": 0.9959816466491493, "全帯域_40ms_modulation_mae_db": 0.555983820185802, "300–1000 Hz_10ms_log_rms_mae_db": 1.6393123541398107, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1378914173814942, "300–1000 Hz_10ms_envelope_correlation": 0.9869287393429365, "300–1000 Hz_10ms_modulation_mae_db": 0.5473622680895576, "300–1000 Hz_40ms_log_rms_mae_db": 0.9776704692900892, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08494518993283726, "300–1000 Hz_40ms_envelope_correlation": 0.9951336219276512, "300–1000 Hz_40ms_modulation_mae_db": 0.6496803182834912, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6275673188995707, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1677138839634276, "1000–3000 Hz_10ms_envelope_correlation": 0.9841594810055715, "1000–3000 Hz_10ms_modulation_mae_db": 1.4267135569827578, "1000–3000 Hz_40ms_log_rms_mae_db": 1.111460000095002, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09650361520654208, "1000–3000 Hz_40ms_envelope_correlation": 0.9951032440248088, "1000–3000 Hz_40ms_modulation_mae_db": 1.0354402281936654, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4119143036182629, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20741059842966272, "3000–6000 Hz_10ms_envelope_correlation": 0.9700276229447641, "3000–6000 Hz_10ms_modulation_mae_db": 0.9942965789643637, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8219055799190645, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12512498703631095, "3000–6000 Hz_40ms_envelope_correlation": 0.9885468700640646, "3000–6000 Hz_40ms_modulation_mae_db": 1.5412389277185021, "mr_stft_db": 5.809434814176627, "band_mae_db": 0.35272696854672797, "rms_gain_error_db": -0.14289358505014094, "output_peak": 0.4802123760284301}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.950274102394489, "全帯域_10ms_rms_envelope_nrmse": 0.09411348168986651, "全帯域_10ms_envelope_correlation": 0.9930782633103957, "全帯域_10ms_modulation_mae_db": 0.2329065431078461, "全帯域_40ms_log_rms_mae_db": 0.5544150824720773, "全帯域_40ms_rms_envelope_nrmse": 0.05261144444665843, "全帯域_40ms_envelope_correlation": 0.9977341827934992, "全帯域_40ms_modulation_mae_db": 0.5502456442626484, "300–1000 Hz_10ms_log_rms_mae_db": 1.5748872534146718, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10588348100864448, "300–1000 Hz_10ms_envelope_correlation": 0.9922144635114677, "300–1000 Hz_10ms_modulation_mae_db": 0.16688863588341704, "300–1000 Hz_40ms_log_rms_mae_db": 0.917901757730554, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059361199104378926, "300–1000 Hz_40ms_envelope_correlation": 0.9974344968239506, "300–1000 Hz_40ms_modulation_mae_db": 0.6254224117146837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6463050995661253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1781022388416096, "1000–3000 Hz_10ms_envelope_correlation": 0.9818302733276799, "1000–3000 Hz_10ms_modulation_mae_db": 1.2509774584260878, "1000–3000 Hz_40ms_log_rms_mae_db": 1.119019213107491, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10808448625800908, "1000–3000 Hz_40ms_envelope_correlation": 0.9935071412245751, "1000–3000 Hz_40ms_modulation_mae_db": 0.754244107102652, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4692169027362572, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22184735315338525, "3000–6000 Hz_10ms_envelope_correlation": 0.9656292106548051, "3000–6000 Hz_10ms_modulation_mae_db": 0.13860161412023608, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8752756144941611, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13307257247155035, "3000–6000 Hz_40ms_envelope_correlation": 0.9872272628685693, "3000–6000 Hz_40ms_modulation_mae_db": 0.8205404405497906, "mr_stft_db": 5.837333684691052, "band_mae_db": 0.34357793801079173, "rms_gain_error_db": 0.02848490330353524, "output_peak": 0.5101632825461776}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9930711170486551, "全帯域_10ms_rms_envelope_nrmse": 0.09380411373076962, "全帯域_10ms_envelope_correlation": 0.9930934449891282, "全帯域_10ms_modulation_mae_db": 0.2105360069250772, "全帯域_40ms_log_rms_mae_db": 0.599858003012455, "全帯域_40ms_rms_envelope_nrmse": 0.05100705949346375, "全帯域_40ms_envelope_correlation": 0.9979215192758738, "全帯域_40ms_modulation_mae_db": 0.08489213395871109, "300–1000 Hz_10ms_log_rms_mae_db": 1.6820962087652542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11114689537202421, "300–1000 Hz_10ms_envelope_correlation": 0.9915218865196198, "300–1000 Hz_10ms_modulation_mae_db": 0.29556772759692396, "300–1000 Hz_40ms_log_rms_mae_db": 1.0267566168467628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06411763124964744, "300–1000 Hz_40ms_envelope_correlation": 0.9972407398226706, "300–1000 Hz_40ms_modulation_mae_db": 0.204508171996978, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5955374164810874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18131655475340216, "1000–3000 Hz_10ms_envelope_correlation": 0.9808154746887117, "1000–3000 Hz_10ms_modulation_mae_db": 1.8922435324587383, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0854478891285742, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09217962259987791, "1000–3000 Hz_40ms_envelope_correlation": 0.995181779660492, "1000–3000 Hz_40ms_modulation_mae_db": 1.2400858926591123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4656251088262457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2102833824824062, "3000–6000 Hz_10ms_envelope_correlation": 0.9698182210669211, "3000–6000 Hz_10ms_modulation_mae_db": 0.24087898994335855, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8769311568207824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1211599965441978, "3000–6000 Hz_40ms_envelope_correlation": 0.9893709261943672, "3000–6000 Hz_40ms_modulation_mae_db": 0.5216112775033985, "mr_stft_db": 5.820007643529226, "band_mae_db": 0.3217832668636701, "rms_gain_error_db": -0.09722294604241097, "output_peak": 0.4905631925249568}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0078117019309214, "全帯域_10ms_rms_envelope_nrmse": 0.10972815314877574, "全帯域_10ms_envelope_correlation": 0.9905181703017575, "全帯域_10ms_modulation_mae_db": 0.6389512384278125, "全帯域_40ms_log_rms_mae_db": 0.5615841374724467, "全帯域_40ms_rms_envelope_nrmse": 0.05765980907156646, "全帯域_40ms_envelope_correlation": 0.9972500869749972, "全帯域_40ms_modulation_mae_db": 0.5675989125523461, "300–1000 Hz_10ms_log_rms_mae_db": 1.653407803121404, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12702082516758484, "300–1000 Hz_10ms_envelope_correlation": 0.9887751695326249, "300–1000 Hz_10ms_modulation_mae_db": 0.9166123030013584, "300–1000 Hz_40ms_log_rms_mae_db": 0.9600845280952479, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06727259796833064, "300–1000 Hz_40ms_envelope_correlation": 0.9967423275876873, "300–1000 Hz_40ms_modulation_mae_db": 0.7299197157775357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.586869341489561, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1647596714833527, "1000–3000 Hz_10ms_envelope_correlation": 0.9845031101449485, "1000–3000 Hz_10ms_modulation_mae_db": 1.2204482666578451, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0673067954297029, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09356423634111001, "1000–3000 Hz_40ms_envelope_correlation": 0.9952486850461417, "1000–3000 Hz_40ms_modulation_mae_db": 0.44080758132359027, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4306072199993467, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20282998790572535, "3000–6000 Hz_10ms_envelope_correlation": 0.9715082901385099, "3000–6000 Hz_10ms_modulation_mae_db": 0.5629722571381333, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8601062182728895, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12399951768814005, "3000–6000 Hz_40ms_envelope_correlation": 0.9887485013653627, "3000–6000 Hz_40ms_modulation_mae_db": 0.22741848866869627, "mr_stft_db": 5.812439291375905, "band_mae_db": 0.325073890742042, "rms_gain_error_db": -0.0435803502651285, "output_peak": 0.6162199397703035}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9699804330264677, "全帯域_10ms_rms_envelope_nrmse": 0.11467920525806448, "全帯域_10ms_envelope_correlation": 0.9897456587499622, "全帯域_10ms_modulation_mae_db": 0.46389478394624256, "全帯域_40ms_log_rms_mae_db": 0.5517331089013543, "全帯域_40ms_rms_envelope_nrmse": 0.05437136443527271, "全帯域_40ms_envelope_correlation": 0.9975821538059146, "全帯域_40ms_modulation_mae_db": 0.36565735429597424, "300–1000 Hz_10ms_log_rms_mae_db": 1.6753431285577987, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13000768624623618, "300–1000 Hz_10ms_envelope_correlation": 0.9882956572348579, "300–1000 Hz_10ms_modulation_mae_db": 0.7720437760721458, "300–1000 Hz_40ms_log_rms_mae_db": 0.987190485705595, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061573083087206514, "300–1000 Hz_40ms_envelope_correlation": 0.997231141408258, "300–1000 Hz_40ms_modulation_mae_db": 0.4775543740160957, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5823888576720853, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16443635021456193, "1000–3000 Hz_10ms_envelope_correlation": 0.9838194605068621, "1000–3000 Hz_10ms_modulation_mae_db": 1.3079855957478497, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0817005736959011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08664130174181471, "1000–3000 Hz_40ms_envelope_correlation": 0.9954526791515045, "1000–3000 Hz_40ms_modulation_mae_db": 0.39868855401163356, "3000–6000 Hz_10ms_log_rms_mae_db": 1.453532260210408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22143951062587472, "3000–6000 Hz_10ms_envelope_correlation": 0.9660031459875089, "3000–6000 Hz_10ms_modulation_mae_db": 1.2579573157155333, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9234458363772191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12641816233164854, "3000–6000 Hz_40ms_envelope_correlation": 0.9882385101224719, "3000–6000 Hz_40ms_modulation_mae_db": 0.7490194815521699, "mr_stft_db": 5.829219857958613, "band_mae_db": 0.2306143558978638, "rms_gain_error_db": 0.042458705184830685, "output_peak": 0.5643358990750955}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9624388637059819, "全帯域_10ms_rms_envelope_nrmse": 0.08452231242353093, "全帯域_10ms_envelope_correlation": 0.9943616039341183, "全帯域_10ms_modulation_mae_db": 0.1793922664254601, "全帯域_40ms_log_rms_mae_db": 0.5458564725187537, "全帯域_40ms_rms_envelope_nrmse": 0.046390700590998965, "全帯域_40ms_envelope_correlation": 0.9982060901284654, "全帯域_40ms_modulation_mae_db": 0.3169141330273639, "300–1000 Hz_10ms_log_rms_mae_db": 1.5886192703904898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09704102679165101, "300–1000 Hz_10ms_envelope_correlation": 0.9934400001112422, "300–1000 Hz_10ms_modulation_mae_db": 0.36857336813061153, "300–1000 Hz_40ms_log_rms_mae_db": 0.9362314465435259, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053844780349231326, "300–1000 Hz_40ms_envelope_correlation": 0.9979017759522133, "300–1000 Hz_40ms_modulation_mae_db": 0.2944765949977977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.648304369372225, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18003634118628417, "1000–3000 Hz_10ms_envelope_correlation": 0.9810452719396363, "1000–3000 Hz_10ms_modulation_mae_db": 1.8662622388748, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1662013143504217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10266963372925782, "1000–3000 Hz_40ms_envelope_correlation": 0.9938928428897902, "1000–3000 Hz_40ms_modulation_mae_db": 1.1506354485710755, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4836136391304988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21139057504740907, "3000–6000 Hz_10ms_envelope_correlation": 0.9689985020438588, "3000–6000 Hz_10ms_modulation_mae_db": 0.5683589619177125, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8910505231231322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12057740354484398, "3000–6000 Hz_40ms_envelope_correlation": 0.98994631674055, "3000–6000 Hz_40ms_modulation_mae_db": 1.247684274296365, "mr_stft_db": 5.833060066741171, "band_mae_db": 0.26829022960876014, "rms_gain_error_db": -0.019757818309379893, "output_peak": 0.5303438850137405}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9520658497578062, "全帯域_10ms_rms_envelope_nrmse": 0.09569358130416099, "全帯域_10ms_envelope_correlation": 0.9929109634174371, "全帯域_10ms_modulation_mae_db": 0.25358720745942326, "全帯域_40ms_log_rms_mae_db": 0.4829729004587066, "全帯域_40ms_rms_envelope_nrmse": 0.04654534130411696, "全帯域_40ms_envelope_correlation": 0.9984355361449169, "全帯域_40ms_modulation_mae_db": 0.3565449316192113, "300–1000 Hz_10ms_log_rms_mae_db": 1.6223258986991, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10860745648903375, "300–1000 Hz_10ms_envelope_correlation": 0.9920831717749501, "300–1000 Hz_10ms_modulation_mae_db": 0.383899494540953, "300–1000 Hz_40ms_log_rms_mae_db": 0.8783330507410136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056325215902846504, "300–1000 Hz_40ms_envelope_correlation": 0.9981597461851588, "300–1000 Hz_40ms_modulation_mae_db": 0.4019103481963442, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5836197585877114, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18008750918834862, "1000–3000 Hz_10ms_envelope_correlation": 0.9807599528233725, "1000–3000 Hz_10ms_modulation_mae_db": 1.8239004016465004, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0711004525189183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09983408955570547, "1000–3000 Hz_40ms_envelope_correlation": 0.9940692827853256, "1000–3000 Hz_40ms_modulation_mae_db": 0.9633245032566835, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4347667004295428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22002402469779955, "3000–6000 Hz_10ms_envelope_correlation": 0.9667318210874728, "3000–6000 Hz_10ms_modulation_mae_db": 0.29564330138484735, "3000–6000 Hz_40ms_log_rms_mae_db": 0.848401598095871, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13744086111063658, "3000–6000 Hz_40ms_envelope_correlation": 0.9873819268169044, "3000–6000 Hz_40ms_modulation_mae_db": 0.7763140620196879, "mr_stft_db": 5.821284283807567, "band_mae_db": 0.4134011568096301, "rms_gain_error_db": -0.15667957739363259, "output_peak": 0.4930451198866532}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 78240, "output_samples": 78240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.721697797778399, "spectral_convergence": 0.2730878540419999, "spectral_convergence_status": "ok", "amplitude_floor": 1.658911626731477e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.696264197931017, "spectral_convergence": 0.2983666839760422, "spectral_convergence_status": "ok", "amplitude_floor": 1.4097603217110054e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.044304606033225, "spectral_convergence": 0.33662372173530875, "spectral_convergence_status": "ok", "amplitude_floor": 8.986287240444543e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.8753665955686144e-05, "output_power": 5.9015237695647334e-05, "delta_db": 0.8295688331447941, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00084480798498807, "output_power": 0.0008550744225374165, "delta_db": 0.05245905679807085, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015560766761226633, "output_power": 0.001451024519995784, "delta_db": -0.3035624178304603, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.758941174067266e-05, "output_power": 6.47930303905178e-05, "delta_db": 0.5118565015161454, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.7471770650556e-05, "output_power": 8.046228891839689e-05, "delta_db": 0.16448900760596097, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.273577609828158e-05, "output_power": 5.08286811794153e-05, "delta_db": -0.15996500312429882, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3369834700032884, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12738627541031974, "pitch": null, "output_rms_dbfs": -25.915568153379418, "output_peak": 0.4975397271403047, "reference_rms_dbfs": -25.788181877969095, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.820755533914213}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 78240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.0002230061513489049, "log_rms_mae_db": 0.9303341774932202, "rms_envelope_nrmse": 0.09077767195416755, "envelope_correlation": 0.9936204087445094, "rms_gain_error_db": -0.1273864494058406, "reference_mean_power": 0.002640134243154231, "output_mean_power": 0.0025638190201439353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.2388906472962606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011616255671056531, "output_power": 0.0010993132751451185, "reference_active": true, "delta_db": -0.23944689703345995}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014996209998372379, "output_power": 0.0001388194882972329, "reference_active": true, "delta_db": -0.3353107418872359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.935590189308993e-05, "output_power": 3.537133782864422e-05, "reference_active": true, "delta_db": -0.46358383102743156}], "active_band_count": 3, "active_band_mae_db": 0.3461138233160425}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.0001943907377270206, "log_rms_mae_db": 0.5417391932483494, "rms_envelope_nrmse": 0.055256424672805345, "envelope_correlation": 0.9976294400613716, "rms_gain_error_db": -0.12738669386415097, "reference_mean_power": 0.0026564436783668454, "output_mean_power": 0.002579656872689828}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 2.080120853269114e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001087632600289277, "output_power": 0.001024005700622811, "reference_active": true, "delta_db": -0.2617984213278398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.148356435560039e-05, "output_power": 6.694515351579463e-05, "reference_active": true, "delta_db": -0.28487057617743666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.554313443594102e-06, "output_power": 2.184423705363699e-06, "reference_active": true, "delta_db": -0.6793730826567188}], "active_band_count": 3, "active_band_mae_db": 0.40868069338733176}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.00021647011759154595, "log_rms_mae_db": 1.5790742942053513, "rms_envelope_nrmse": 0.10206082767397348, "envelope_correlation": 0.9930451006613001, "rms_gain_error_db": -0.21379359750451876, "reference_mean_power": 0.0021208071215793162, "output_mean_power": 0.0020189325656490434}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.0688695374548394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066233690351074, "output_power": 0.000991946061077562, "reference_active": true, "delta_db": -0.31364343817239637}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015588791655839424, "output_power": 0.00013849812048793056, "reference_active": true, "delta_db": -0.5136857296839478}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.732583628298963e-05, "output_power": 3.114579558670856e-05, "reference_active": true, "delta_db": -0.7861011789706915}], "active_band_count": 3, "active_band_mae_db": 0.5378101156090119}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.00018533448652616975, "log_rms_mae_db": 0.8821719471320489, "rms_envelope_nrmse": 0.06696339473055045, "envelope_correlation": 0.9971283780930403, "rms_gain_error_db": -0.21379327121882025, "reference_mean_power": 0.0021339103768736657, "output_mean_power": 0.0020314065488969846}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.9317825081344165e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001010025588563974, "output_power": 0.0009317504979043033, "reference_active": true, "delta_db": -0.3503274308385654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.797692317548564e-05, "output_power": 7.023784819342055e-05, "reference_active": true, "delta_db": -0.4539489667971555}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604806628732506e-06, "output_power": 1.9527768211955154e-06, "reference_active": true, "delta_db": -1.1766888881676056}], "active_band_count": 3, "active_band_mae_db": 0.6603217619344423}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 6.255230128136302e-05, "log_rms_mae_db": 1.5362629098130223, "rms_envelope_nrmse": 0.16585097854886485, "envelope_correlation": 0.984310220524849, "rms_gain_error_db": 0.2620535590159607, "reference_mean_power": 0.0001286629346191704, "output_mean_power": 0.0001366654729775079}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.8253387298901982e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.359313427297316e-05, "output_power": 7.384097319160198e-05, "reference_active": true, "delta_db": 0.01460112290134115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.331803440637771e-06, "output_power": 9.140917697415256e-06, "reference_active": true, "delta_db": 0.9577898529511074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8895709731947531e-06, "output_power": 5.093884683298918e-06, "reference_active": true, "delta_db": 4.306859001971331}], "active_band_count": 3, "active_band_mae_db": 1.7597499926079265}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 6.0793288183779695e-05, "log_rms_mae_db": 0.9937245724728766, "rms_envelope_nrmse": 0.08583858568569004, "envelope_correlation": 0.9960986337858669, "rms_gain_error_db": 0.2620541625439303, "reference_mean_power": 0.00012945711895499159, "output_mean_power": 0.0001375090728601326}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.7711142240683522e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.969624557508824e-05, "output_power": 7.027683735899049e-05, "reference_active": true, "delta_db": 0.03602824897154488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.240093489969041e-06, "output_power": 3.934042408061507e-06, "reference_active": true, "delta_db": 0.8428149563984931}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.585555587494176e-07, "output_power": 2.613091765245818e-07, "reference_active": true, "delta_db": 2.1697318890736046}], "active_band_count": 3, "active_band_mae_db": 1.016191698147881}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 3.765201856958641e-05, "log_rms_mae_db": 1.4702862342251986, "rms_envelope_nrmse": 0.21637429238952421, "envelope_correlation": 0.9674570069794305, "rms_gain_error_db": -0.11898680552886799, "reference_mean_power": 2.5945814971953512e-05, "output_mean_power": 2.5244608278159085e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.30554638893879e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0948707154331005e-06, "output_power": 5.055723408437684e-06, "reference_active": true, "delta_db": -0.0334986179028576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.046923144145412e-06, "output_power": 1.1974371876590855e-06, "reference_active": true, "delta_db": 0.5833794070429204}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.3665503073484234e-07, "output_power": 7.227986702945852e-07, "reference_active": true, "delta_db": 2.188788768723731}], "active_band_count": 3, "active_band_mae_db": 0.935222264556503}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 2.0830708932641756e-05, "log_rms_mae_db": 0.8979472950310868, "rms_envelope_nrmse": 0.12191636934699913, "envelope_correlation": 0.9892197545008102, "rms_gain_error_db": -0.11901163148482435, "reference_mean_power": 2.6105327616819126e-05, "output_mean_power": 2.5399664769675415e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.2170827427715922e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.759181059225212e-06, "output_power": 4.807626707601926e-06, "reference_active": true, "delta_db": 0.04398511629103266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.967295837314181e-07, "output_power": 7.079460194366368e-07, "reference_active": true, "delta_db": 0.7422257504029494}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240630896276443e-08, "output_power": 4.636966419170652e-08, "reference_active": true, "delta_db": 1.5560438249241895}], "active_band_count": 3, "active_band_mae_db": 0.7807515638727237}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8603543860235295, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9afac594e709321d7bf4abe6f7c67fa9a6481414b10ff1711fafd2cb958318c6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826554056202039, "standard_deviation": 0.0016912021012353738, "minimum": 0.9807599528233725, "maximum": 0.9845031101449485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6008568839851733, "standard_deviation": 0.03791864895343766, "minimum": 1.5362629098130223, "maximum": 1.648304369372225, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.5685351304253132, "standard_deviation": 0.29403835133861533, "minimum": 1.2204482666578451, "maximum": 1.8922435324587383, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17278794102248146, "standard_deviation": 0.007698531381644202, "minimum": 0.16443635021456193, "maximum": 0.18131655475340216, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9948192860710631, "standard_deviation": 0.000892577372901097, "minimum": 0.9935071412245751, "maximum": 0.9960986337858669, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0869951013498609, "standard_deviation": 0.04966715770227519, "minimum": 0.9937245724728766, "maximum": 1.1662013143504217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8749272516582867, "standard_deviation": 0.3145785350951791, "minimum": 0.39868855401163356, "maximum": 1.2400858926591123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09566444638975088, "standard_deviation": 0.007708434392026519, "minimum": 0.08583858568569004, "maximum": 0.10808448625800908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9682717276129089, "standard_deviation": 0.002126122854870051, "minimum": 0.9656292106548051, "maximum": 0.9715082901385099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.45244529614697, "standard_deviation": 0.024431057775777483, "minimum": 1.4119143036182629, "maximum": 1.4836136391304988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.624241410467586, "standard_deviation": 0.4022991068613634, "minimum": 0.13860161412023608, "maximum": 1.2579573157155333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2139499655914734, "standard_deviation": 0.00704723966059966, "minimum": 0.20282998790572535, "maximum": 0.22184735315338525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9885850085841376, "standard_deviation": 0.0009496505787824021, "minimum": 0.9872272628685693, "maximum": 0.98994631674055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8743829777667759, "standard_deviation": 0.0313511755630652, "minimum": 0.8219055799190645, "maximum": 0.9234458363772191, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8330723145226668, "standard_deviation": 0.405393715216053, "minimum": 0.22741848866869627, "maximum": 1.5412389277185021, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1262137337592909, "standard_deviation": 0.006035582459043531, "minimum": 0.12057740354484398, "maximum": 0.13744086111063658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907880235861249, "standard_deviation": 0.0024359104267247305, "minimum": 0.9869287393429365, "maximum": 0.9934400001112422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6268832764117351, "standard_deviation": 0.04264311482687458, "minimum": 1.5748872534146718, "maximum": 1.6820962087652542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.49859471111549736, "standard_deviation": 0.24920979100627086, "minimum": 0.16688863588341704, "maximum": 0.9166123030013584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11495745201633026, "standard_deviation": 0.014747892037149128, "minimum": 0.09704102679165101, "maximum": 0.1378914173814942, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971215284750787, "standard_deviation": 0.0009185757872597451, "minimum": 0.9951336219276512, "maximum": 0.9981597461851588, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9457925377606047, "standard_deviation": 0.0520399677713039, "minimum": 0.8783330507410136, "maximum": 1.0267566168467628, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.505474212114671, "standard_deviation": 0.1912896317276833, "minimum": 0.204508171996978, "maximum": 0.7299197157775357, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06430038654062863, "standard_deviation": 0.009613411496287484, "minimum": 0.053844780349231326, "maximum": 0.08494518993283726, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.32405640956034676, "standard_deviation": 0.055032679178695434, "minimum": 0.2306143558978638, "maximum": 0.4134011568096301, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.822941897024297, "standard_deviation": 0.009693867173302671, "minimum": 5.809434814176627, "maximum": 5.837333684691052, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5228029277482078, "standard_deviation": 0.046361105259891226, "minimum": 0.4802123760284301, "maximum": 0.6162199397703035, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06457211799783084, "standard_deviation": 0.0776382760119592, "minimum": -0.15667957739363259, "maximum": 0.042458705184830685, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9919470537543678, "standard_deviation": 0.0021621617132525184, "minimum": 0.9882479165876339, "maximum": 0.9943616039341183, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9745259874865022, "standard_deviation": 0.03331080985471218, "minimum": 0.9303341774932202, "maximum": 1.0302316545344765, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.350232564175482, "standard_deviation": 0.16214625908565852, "minimum": 0.1793922664254601, "maximum": 0.6389512384278125, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10067934800667486, "standard_deviation": 0.013157011015991217, "minimum": 0.08452231242353093, "maximum": 0.1221162645440631, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975925819792735, "standard_deviation": 0.0007491993098346007, "minimum": 0.9959816466491493, "maximum": 0.9984355361449169, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5551714939505648, "standard_deviation": 0.03748947550157179, "minimum": 0.4829729004587066, "maximum": 0.6032130535203757, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4008147029111736, "standard_deviation": 0.1622731975901934, "minimum": 0.08489213395871109, "maximum": 0.5675989125523461, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054307774738227316, "standard_deviation": 0.00769914892451048, "minimum": 0.046390700590998965, "maximum": 0.07062005389093592, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4507-16021-0045", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.2285919189453125, "predicted_mos": 2.2285919189453125, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "33edeeb3677c7e6bc5ae4c032bb7ec92b72a36d7db582ed2607091169cfb4ae9", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8256363002583385, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9773769320742913, "全帯域_10ms_rms_envelope_nrmse": 0.09981673388751186, "全帯域_10ms_envelope_correlation": 0.9921541182766834, "全帯域_10ms_modulation_mae_db": 0.2860403251164387, "全帯域_40ms_log_rms_mae_db": 0.5016671818925461, "全帯域_40ms_rms_envelope_nrmse": 0.04325047161101707, "全帯域_40ms_envelope_correlation": 0.9985247325690787, "全帯域_40ms_modulation_mae_db": 0.39820750405472655, "300–1000 Hz_10ms_log_rms_mae_db": 1.6102685102175784, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11563978543581055, "300–1000 Hz_10ms_envelope_correlation": 0.9907435747304321, "300–1000 Hz_10ms_modulation_mae_db": 0.480447739579148, "300–1000 Hz_40ms_log_rms_mae_db": 0.8569170036970323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05302398312997936, "300–1000 Hz_40ms_envelope_correlation": 0.9981308980354444, "300–1000 Hz_40ms_modulation_mae_db": 0.47236640572670785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6354098854570278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14406760255361703, "1000–3000 Hz_10ms_envelope_correlation": 0.9873966557377422, "1000–3000 Hz_10ms_modulation_mae_db": 0.2006057972158277, "1000–3000 Hz_40ms_log_rms_mae_db": 1.155101734159026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07755163460040024, "1000–3000 Hz_40ms_envelope_correlation": 0.9962703627574993, "1000–3000 Hz_40ms_modulation_mae_db": 0.20990241123237685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4093942768694727, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21758613770755295, "3000–6000 Hz_10ms_envelope_correlation": 0.9675780050684848, "3000–6000 Hz_10ms_modulation_mae_db": 0.5907461547371289, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9050492815748468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1389257034480663, "3000–6000 Hz_40ms_envelope_correlation": 0.9872332528118117, "3000–6000 Hz_40ms_modulation_mae_db": 0.4348464319292204, "mr_stft_db": 5.825898603945208, "band_mae_db": 0.31895672051701707, "rms_gain_error_db": -0.09984140945449597, "output_peak": 0.5001892979400855}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9972318338562461, "全帯域_10ms_rms_envelope_nrmse": 0.10338067201995, "全帯域_10ms_envelope_correlation": 0.9921808099193019, "全帯域_10ms_modulation_mae_db": 0.3939965138138492, "全帯域_40ms_log_rms_mae_db": 0.5587610995900343, "全帯域_40ms_rms_envelope_nrmse": 0.06770700680984683, "全帯域_40ms_envelope_correlation": 0.9969998338581032, "全帯域_40ms_modulation_mae_db": 0.41695127873145815, "300–1000 Hz_10ms_log_rms_mae_db": 1.644071473492255, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12160888303213983, "300–1000 Hz_10ms_envelope_correlation": 0.9905160920167327, "300–1000 Hz_10ms_modulation_mae_db": 0.5035340267068692, "300–1000 Hz_40ms_log_rms_mae_db": 0.9090356604891996, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08102870196708095, "300–1000 Hz_40ms_envelope_correlation": 0.9963214272584698, "300–1000 Hz_40ms_modulation_mae_db": 0.45695062350599785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6524541964165842, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15939699977057994, "1000–3000 Hz_10ms_envelope_correlation": 0.9845992206661949, "1000–3000 Hz_10ms_modulation_mae_db": 1.2930200624926735, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0919106452394214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07708317522376822, "1000–3000 Hz_40ms_envelope_correlation": 0.996784263116472, "1000–3000 Hz_40ms_modulation_mae_db": 0.6148356312153723, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4100428421556312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20267098810564335, "3000–6000 Hz_10ms_envelope_correlation": 0.9714475417553019, "3000–6000 Hz_10ms_modulation_mae_db": 0.9797751092531725, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8800375152373634, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1205332349631492, "3000–6000 Hz_40ms_envelope_correlation": 0.9893604705085661, "3000–6000 Hz_40ms_modulation_mae_db": 0.8539118767923891, "mr_stft_db": 5.806398072508607, "band_mae_db": 0.3147054713828232, "rms_gain_error_db": -0.28300700773101933, "output_peak": 0.4947815229072634}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9811985109297937, "全帯域_10ms_rms_envelope_nrmse": 0.09183852491679725, "全帯域_10ms_envelope_correlation": 0.993425592925001, "全帯域_10ms_modulation_mae_db": 0.4173506925562725, "全帯域_40ms_log_rms_mae_db": 0.5355091058029388, "全帯域_40ms_rms_envelope_nrmse": 0.04517256499458065, "全帯域_40ms_envelope_correlation": 0.9984524438183324, "全帯域_40ms_modulation_mae_db": 0.25949194473417125, "300–1000 Hz_10ms_log_rms_mae_db": 1.5829469406660883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10741978237355129, "300–1000 Hz_10ms_envelope_correlation": 0.992167014035391, "300–1000 Hz_10ms_modulation_mae_db": 0.5962255327588283, "300–1000 Hz_40ms_log_rms_mae_db": 0.9013152440190373, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0565928141763187, "300–1000 Hz_40ms_envelope_correlation": 0.9980071821824277, "300–1000 Hz_40ms_modulation_mae_db": 0.3445381905935218, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6598063540035843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17156374404311284, "1000–3000 Hz_10ms_envelope_correlation": 0.9827856121445033, "1000–3000 Hz_10ms_modulation_mae_db": 1.2769339592043352, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1595313807037868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08739147123930968, "1000–3000 Hz_40ms_envelope_correlation": 0.9956537841462673, "1000–3000 Hz_40ms_modulation_mae_db": 0.8275039191495286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3868051804722117, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1847067328839357, "3000–6000 Hz_10ms_envelope_correlation": 0.9769772906279717, "3000–6000 Hz_10ms_modulation_mae_db": 0.4930655457887984, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7708247982664435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0954101325329604, "3000–6000 Hz_40ms_envelope_correlation": 0.9935723796903508, "3000–6000 Hz_40ms_modulation_mae_db": 0.6580443607835139, "mr_stft_db": 5.793034984973322, "band_mae_db": 0.28597532153807154, "rms_gain_error_db": -0.1267895539308772, "output_peak": 0.5068449331729348}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9737798609320988, "全帯域_10ms_rms_envelope_nrmse": 0.09064834583515981, "全帯域_10ms_envelope_correlation": 0.9936613964041575, "全帯域_10ms_modulation_mae_db": 0.26483124400878316, "全帯域_40ms_log_rms_mae_db": 0.5260290329785732, "全帯域_40ms_rms_envelope_nrmse": 0.05070212945674173, "全帯域_40ms_envelope_correlation": 0.9980807288246145, "全帯域_40ms_modulation_mae_db": 0.18548576959581123, "300–1000 Hz_10ms_log_rms_mae_db": 1.5565770955158145, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10406710790150425, "300–1000 Hz_10ms_envelope_correlation": 0.9927693465391791, "300–1000 Hz_10ms_modulation_mae_db": 0.24603181530262197, "300–1000 Hz_40ms_log_rms_mae_db": 0.83697457527265, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06164053296889649, "300–1000 Hz_40ms_envelope_correlation": 0.9976795183455716, "300–1000 Hz_40ms_modulation_mae_db": 0.3670647750801403, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6391789109757147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2019103128137739, "1000–3000 Hz_10ms_envelope_correlation": 0.9769324323385573, "1000–3000 Hz_10ms_modulation_mae_db": 2.2585527651647537, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0375006792248684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12093201377117471, "1000–3000 Hz_40ms_envelope_correlation": 0.99203137633941, "1000–3000 Hz_40ms_modulation_mae_db": 2.2729880142613474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4410445817258748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21219381928446496, "3000–6000 Hz_10ms_envelope_correlation": 0.9687823176087677, "3000–6000 Hz_10ms_modulation_mae_db": 0.8294747988510455, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9062035257297342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1254617484913624, "3000–6000 Hz_40ms_envelope_correlation": 0.988405906344907, "3000–6000 Hz_40ms_modulation_mae_db": 0.47924410924598176, "mr_stft_db": 5.808125274030288, "band_mae_db": 0.31936339395058155, "rms_gain_error_db": -0.15096020725544843, "output_peak": 0.4965181238868701}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9362054784293956, "全帯域_10ms_rms_envelope_nrmse": 0.0994607762175086, "全帯域_10ms_envelope_correlation": 0.9923477411949593, "全帯域_10ms_modulation_mae_db": 0.8552121693313963, "全帯域_40ms_log_rms_mae_db": 0.5118705438182928, "全帯域_40ms_rms_envelope_nrmse": 0.05031518363287174, "全帯域_40ms_envelope_correlation": 0.9981477451196636, "全帯域_40ms_modulation_mae_db": 0.4717442608048075, "300–1000 Hz_10ms_log_rms_mae_db": 1.6539228687996281, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12061238174575487, "300–1000 Hz_10ms_envelope_correlation": 0.9900922692244213, "300–1000 Hz_10ms_modulation_mae_db": 1.1913554274137939, "300–1000 Hz_40ms_log_rms_mae_db": 0.9586349475823442, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06202265327661841, "300–1000 Hz_40ms_envelope_correlation": 0.9975878344518051, "300–1000 Hz_40ms_modulation_mae_db": 0.6271616351012163, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5595668416049753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14741945735237547, "1000–3000 Hz_10ms_envelope_correlation": 0.986709808777909, "1000–3000 Hz_10ms_modulation_mae_db": 1.4227219559881135, "1000–3000 Hz_40ms_log_rms_mae_db": 1.104428278901848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0783445907486943, "1000–3000 Hz_40ms_envelope_correlation": 0.9962970355108595, "1000–3000 Hz_40ms_modulation_mae_db": 0.5309289277650356, "3000–6000 Hz_10ms_log_rms_mae_db": 1.384123350419568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19669165447122022, "3000–6000 Hz_10ms_envelope_correlation": 0.9734222464556687, "3000–6000 Hz_10ms_modulation_mae_db": 0.20000053780163426, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8170500383444107, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09909400086519284, "3000–6000 Hz_40ms_envelope_correlation": 0.9928237769488353, "3000–6000 Hz_40ms_modulation_mae_db": 0.5027554815292091, "mr_stft_db": 5.8058367376874385, "band_mae_db": 0.2526438902989892, "rms_gain_error_db": -0.16717659952168906, "output_peak": 0.5512088806375007}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.959454227243556, "全帯域_10ms_rms_envelope_nrmse": 0.10058397382771621, "全帯域_10ms_envelope_correlation": 0.9920941738875663, "全帯域_10ms_modulation_mae_db": 0.2681515540827047, "全帯域_40ms_log_rms_mae_db": 0.5181299633850468, "全帯域_40ms_rms_envelope_nrmse": 0.04863308967999988, "全帯域_40ms_envelope_correlation": 0.9982018358358202, "全帯域_40ms_modulation_mae_db": 0.24392975049151444, "300–1000 Hz_10ms_log_rms_mae_db": 1.5820880948233969, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11402426199438934, "300–1000 Hz_10ms_envelope_correlation": 0.9911588424019274, "300–1000 Hz_10ms_modulation_mae_db": 0.39505494913297273, "300–1000 Hz_40ms_log_rms_mae_db": 0.8868810208520734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059191305928221895, "300–1000 Hz_40ms_envelope_correlation": 0.9978266714092542, "300–1000 Hz_40ms_modulation_mae_db": 0.3363382166307261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6214389041850628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17792738215709158, "1000–3000 Hz_10ms_envelope_correlation": 0.9817785193572182, "1000–3000 Hz_10ms_modulation_mae_db": 1.7897408331936158, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1220507485246412, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09903279222507157, "1000–3000 Hz_40ms_envelope_correlation": 0.9945823749276925, "1000–3000 Hz_40ms_modulation_mae_db": 1.3748449414195718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3768517757693872, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21379026079586422, "3000–6000 Hz_10ms_envelope_correlation": 0.9681366822341191, "3000–6000 Hz_10ms_modulation_mae_db": 0.5389451686331124, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8140263127982743, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10776524069696955, "3000–6000 Hz_40ms_envelope_correlation": 0.9916117907886938, "3000–6000 Hz_40ms_modulation_mae_db": 0.7731858421314614, "mr_stft_db": 5.801014437437474, "band_mae_db": 0.32321052111191634, "rms_gain_error_db": -0.12750258560796635, "output_peak": 0.5665860786616508}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9886688990830902, "全帯域_10ms_rms_envelope_nrmse": 0.09958390523766975, "全帯域_10ms_envelope_correlation": 0.9923467431622658, "全帯域_10ms_modulation_mae_db": 0.546722374499289, "全帯域_40ms_log_rms_mae_db": 0.5330637331263944, "全帯域_40ms_rms_envelope_nrmse": 0.05328398092351304, "全帯域_40ms_envelope_correlation": 0.9979282747256063, "全帯域_40ms_modulation_mae_db": 0.3942140229609097, "300–1000 Hz_10ms_log_rms_mae_db": 1.613284209692654, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11678363384281346, "300–1000 Hz_10ms_envelope_correlation": 0.9907210641568296, "300–1000 Hz_10ms_modulation_mae_db": 0.7606568304076357, "300–1000 Hz_40ms_log_rms_mae_db": 0.9501480514662287, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06381202881483314, "300–1000 Hz_40ms_envelope_correlation": 0.9974325359568338, "300–1000 Hz_40ms_modulation_mae_db": 0.4718837852030339, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6047932387535029, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1860437975049006, "1000–3000 Hz_10ms_envelope_correlation": 0.9787698172742347, "1000–3000 Hz_10ms_modulation_mae_db": 1.2998884549329277, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0437132217916467, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09891781270697819, "1000–3000 Hz_40ms_envelope_correlation": 0.9938870528157966, "1000–3000 Hz_40ms_modulation_mae_db": 0.8404308083448747, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4318755116996345, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20883909902893658, "3000–6000 Hz_10ms_envelope_correlation": 0.9698683880272362, "3000–6000 Hz_10ms_modulation_mae_db": 0.21775331177904403, "3000–6000 Hz_40ms_log_rms_mae_db": 0.850305622551884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11353772255965029, "3000–6000 Hz_40ms_envelope_correlation": 0.9913140592830814, "3000–6000 Hz_40ms_modulation_mae_db": 1.031113143511807, "mr_stft_db": 5.800091289783455, "band_mae_db": 0.30945974554274386, "rms_gain_error_db": -0.17411483843988768, "output_peak": 0.4916150299773693}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.003017525496107, "全帯域_10ms_rms_envelope_nrmse": 0.1323602734679973, "全帯域_10ms_envelope_correlation": 0.9868583118449767, "全帯域_10ms_modulation_mae_db": 0.42910488372076855, "全帯域_40ms_log_rms_mae_db": 0.5986808502699478, "全帯域_40ms_rms_envelope_nrmse": 0.08715891426039356, "全帯域_40ms_envelope_correlation": 0.9947171663131038, "全帯域_40ms_modulation_mae_db": 0.5186353602870187, "300–1000 Hz_10ms_log_rms_mae_db": 1.745282781278519, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15763202764920628, "300–1000 Hz_10ms_envelope_correlation": 0.9837791121700458, "300–1000 Hz_10ms_modulation_mae_db": 0.45903672220641, "300–1000 Hz_40ms_log_rms_mae_db": 1.0469029446713047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10770082002452537, "300–1000 Hz_40ms_envelope_correlation": 0.9931918240802903, "300–1000 Hz_40ms_modulation_mae_db": 0.7066787283157895, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5722853975288869, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16734231320719956, "1000–3000 Hz_10ms_envelope_correlation": 0.9842359359415109, "1000–3000 Hz_10ms_modulation_mae_db": 0.4681731471325905, "1000–3000 Hz_40ms_log_rms_mae_db": 1.08924896130918, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1032908336035188, "1000–3000 Hz_40ms_envelope_correlation": 0.9942883324283668, "1000–3000 Hz_40ms_modulation_mae_db": 0.24379217055609692, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3785674901349052, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.199664581570887, "3000–6000 Hz_10ms_envelope_correlation": 0.9725814885006352, "3000–6000 Hz_10ms_modulation_mae_db": 0.8275372808095586, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8048845674167004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11050828900788744, "3000–6000 Hz_40ms_envelope_correlation": 0.9910577732156931, "3000–6000 Hz_40ms_modulation_mae_db": 0.564201176953446, "mr_stft_db": 5.809185393868215, "band_mae_db": 0.39232896756743346, "rms_gain_error_db": -0.338720557655609, "output_peak": 0.5323758081821013}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 78240, "output_samples": 78240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.753736093045354, "spectral_convergence": 0.27803844061869787, "spectral_convergence_status": "ok", "amplitude_floor": 1.658911626731477e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.699590715691127, "spectral_convergence": 0.29805043968680833, "spectral_convergence_status": "ok", "amplitude_floor": 1.4097603217110054e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.024369003099141, "spectral_convergence": 0.3053962973898062, "spectral_convergence_status": "ok", "amplitude_floor": 8.986287240444543e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.8753665955686144e-05, "output_power": 5.9257372366217225e-05, "delta_db": 0.8473511241022964, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00084480798498807, "output_power": 0.0008305977963860907, "delta_db": -0.07367235517867558, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015560766761226633, "output_power": 0.001499813320262425, "delta_db": -0.15993786793415174, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.758941174067266e-05, "output_power": 6.338242073147422e-05, "delta_db": 0.4162617959290585, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.7471770650556e-05, "output_power": 7.48555147260776e-05, "delta_db": -0.14919681036690094, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.273577609828158e-05, "output_power": 4.9587624609626056e-05, "delta_db": -0.2673203695910193, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.31895672051701707, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09984140945449597, "pitch": null, "output_rms_dbfs": -25.888023287423593, "output_peak": 0.5001892979400855, "reference_rms_dbfs": -25.788181877969095, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.825898603945208}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 78240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.0002230061513489049, "log_rms_mae_db": 0.9773769320742913, "rms_envelope_nrmse": 0.09981673388751186, "envelope_correlation": 0.9921541182766834, "rms_gain_error_db": -0.09984150758244198, "reference_mean_power": 0.002640134243154231, "output_mean_power": 0.002580131608922883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.2388906472962606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011616255671056531, "output_power": 0.0011463136050232313, "reference_active": true, "delta_db": -0.0576271531786669}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014996209998372379, "output_power": 0.0001263210932719343, "reference_active": true, "delta_db": -0.7450563769674146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.935590189308993e-05, "output_power": 3.9861497843403245e-05, "reference_active": true, "delta_db": 0.0554374452032345}], "active_band_count": 3, "active_band_mae_db": 0.2860403251164387}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.0001943907377270206, "log_rms_mae_db": 0.5016671818925461, "rms_envelope_nrmse": 0.04325047161101707, "envelope_correlation": 0.9985247325690787, "rms_gain_error_db": -0.09984091732772539, "reference_mean_power": 0.0026564436783668454, "output_mean_power": 0.0025960707305628526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 2.080120853269114e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001087632600289277, "output_power": 0.0010742279968223608, "reference_active": true, "delta_db": -0.05385749601586453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.148356435560039e-05, "output_power": 6.146513183644342e-05, "reference_active": true, "delta_db": -0.655773819448852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.554313443594102e-06, "output_power": 2.2844153485706227e-06, "reference_active": true, "delta_db": -0.484991196699463}], "active_band_count": 3, "active_band_mae_db": 0.39820750405472655}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.00021647011759154595, "log_rms_mae_db": 1.6102685102175784, "rms_envelope_nrmse": 0.11563978543581055, "envelope_correlation": 0.9907435747304321, "rms_gain_error_db": -0.15397246891863153, "reference_mean_power": 0.0021208071215793162, "output_mean_power": 0.002046934407207903}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.0688695374548394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066233690351074, "output_power": 0.001045237962780919, "reference_active": true, "delta_db": -0.08637226230186607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015588791655839424, "output_power": 0.00012724499640611307, "reference_active": true, "delta_db": -0.8817173895124052}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.732583628298963e-05, "output_power": 4.162314611175564e-05, "reference_active": true, "delta_db": 0.4732535669231727}], "active_band_count": 3, "active_band_mae_db": 0.480447739579148}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.00018533448652616975, "log_rms_mae_db": 0.8569170036970323, "rms_envelope_nrmse": 0.05302398312997936, "envelope_correlation": 0.9981308980354444, "rms_gain_error_db": -0.1539724501309814, "reference_mean_power": 0.0021339103768736657, "output_mean_power": 0.0020595812541236136}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.9317825081344165e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001010025588563974, "output_power": 0.0009910988910023226, "reference_active": true, "delta_db": -0.0821538640553983}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.797692317548564e-05, "output_power": 6.419616335088037e-05, "reference_active": true, "delta_db": -0.8445702126799454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604806628732506e-06, "output_power": 2.2870938510858357e-06, "reference_active": true, "delta_db": -0.4903751404447798}], "active_band_count": 3, "active_band_mae_db": 0.47236640572670785}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 6.255230128136302e-05, "log_rms_mae_db": 1.6354098854570278, "rms_envelope_nrmse": 0.14406760255361703, "envelope_correlation": 0.9873966557377422, "rms_gain_error_db": 0.025328085066508466, "reference_mean_power": 0.0001286629346191704, "output_mean_power": 0.00012941549009018674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.8253387298901982e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.359313427297316e-05, "output_power": 7.339357589073907e-05, "reference_active": true, "delta_db": -0.011792516122715438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.331803440637771e-06, "output_power": 7.45676319965256e-06, "reference_active": true, "delta_db": 0.07339538273832022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8895709731947531e-06, "output_power": 2.1282672305540662e-06, "reference_active": true, "delta_db": 0.5166294927864474}], "active_band_count": 3, "active_band_mae_db": 0.2006057972158277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 6.0793288183779695e-05, "log_rms_mae_db": 1.155101734159026, "rms_envelope_nrmse": 0.07755163460040024, "envelope_correlation": 0.9962703627574993, "rms_gain_error_db": 0.025330335293289284, "reference_mean_power": 0.00012945711895499159, "output_mean_power": 0.0001302143871154826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.7711142240683522e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.969624557508824e-05, "output_power": 6.93060794553112e-05, "reference_active": true, "delta_db": -0.024380518501051128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.240093489969041e-06, "output_power": 3.471161209540301e-06, "reference_active": true, "delta_db": 0.2991724231612868}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.585555587494176e-07, "output_power": 1.4776311680748272e-07, "reference_active": true, "delta_db": -0.3061542920347927}], "active_band_count": 3, "active_band_mae_db": 0.20990241123237685}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 3.765201856958641e-05, "log_rms_mae_db": 1.4093942768694727, "rms_envelope_nrmse": 0.21758613770755295, "envelope_correlation": 0.9675780050684848, "rms_gain_error_db": -0.43881950433135075, "reference_mean_power": 2.5945814971953512e-05, "output_mean_power": 2.3452295962077242e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.30554638893879e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0948707154331005e-06, "output_power": 4.876945220377241e-06, "reference_active": true, "delta_db": -0.18985290604977031}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.046923144145412e-06, "output_power": 1.2137351190553116e-06, "reference_active": true, "delta_db": 0.642091175184948}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.3665503073484234e-07, "output_power": 5.422104785837245e-07, "reference_active": true, "delta_db": 0.9402943829766683}], "active_band_count": 3, "active_band_mae_db": 0.5907461547371289}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 2.0830708932641756e-05, "log_rms_mae_db": 0.9050492815748468, "rms_envelope_nrmse": 0.1389257034480663, "envelope_correlation": 0.9872332528118117, "rms_gain_error_db": -0.43880168011492077, "reference_mean_power": 2.6105327616819126e-05, "output_mean_power": 2.359657551029505e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.2170827427715922e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.759181059225212e-06, "output_power": 4.603587550074334e-06, "reference_active": true, "delta_db": -0.1443582060966602}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.967295837314181e-07, "output_power": 6.983240445096732e-07, "reference_active": true, "delta_db": 0.6827942665739366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240630896276443e-08, "output_power": 2.9032931267278077e-08, "reference_active": true, "delta_db": -0.4773868231170644}], "active_band_count": 3, "active_band_mae_db": 0.4348464319292204}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8556317710968941, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b840f56bda052b5a1c953bbefbc48f424ea4661b60167d06cd58299c0789d641", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9829010002797338, "standard_deviation": 0.003651764739948074, "minimum": 0.9769324323385573, "maximum": 0.9873966557377422, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6181167161156673, "standard_deviation": 0.036599976132003095, "minimum": 1.5595668416049753, "maximum": 1.6598063540035843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2512046219156048, "standard_deviation": 0.6607210947008869, "minimum": 0.2006057972158277, "maximum": 2.2585527651647537, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16945895117533138, "standard_deviation": 0.01942191461591323, "minimum": 0.14406760255361703, "maximum": 0.2019103128137739, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949743227552954, "standard_deviation": 0.0015871869081668926, "minimum": 0.99203137633941, "maximum": 0.996784263116472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1004357062318024, "standard_deviation": 0.04526099208684857, "minimum": 1.0375006792248684, "maximum": 1.1595313807037868, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8644033529930255, "standard_deviation": 0.6796655245503852, "minimum": 0.20990241123237685, "maximum": 2.2729880142613474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09281804051486446, "standard_deviation": 0.015563156053686905, "minimum": 0.07708317522376822, "maximum": 0.12093201377117471, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9710992450347732, "standard_deviation": 0.0031696142805712844, "minimum": 0.9675780050684848, "maximum": 0.9769772906279717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4023381261558359, "standard_deviation": 0.024690791830972152, "minimum": 1.3768517757693872, "maximum": 1.4410445817258748, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5846622384566869, "standard_deviation": 0.28489082110235664, "minimum": 0.20000053780163426, "maximum": 0.9797751092531725, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2045179092310631, "standard_deviation": 0.010790513388525705, "minimum": 0.1847067328839357, "maximum": 0.21758613770755295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906724261989923, "standard_deviation": 0.0021767237845917143, "minimum": 0.9872332528118117, "maximum": 0.9935723796903508, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8435477077399571, "standard_deviation": 0.04991774634419918, "minimum": 0.7708247982664435, "maximum": 0.9062035257297342, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6621628028596286, "standard_deviation": 0.20886574404419134, "minimum": 0.4348464319292204, "maximum": 1.031113143511807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1139045090706548, "standard_deviation": 0.014214291269720663, "minimum": 0.0954101325329604, "maximum": 0.1389257034480663, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9902434144093699, "standard_deviation": 0.0027589151094749337, "minimum": 0.9837791121700458, "maximum": 0.9927693465391791, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6235552468107417, "standard_deviation": 0.058915634858734224, "minimum": 1.5565770955158145, "maximum": 1.745282781278519, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5790428804385349, "standard_deviation": 0.28853757386495227, "minimum": 0.24603181530262197, "maximum": 1.1913554274137939, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11972348299689624, "standard_deviation": 0.01645646282487698, "minimum": 0.10406710790150425, "maximum": 0.15763202764920628, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970222364650121, "standard_deviation": 0.0016444635656258605, "minimum": 0.9931918240802903, "maximum": 0.9981308980354444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9183511810062338, "standard_deviation": 0.06646488000230925, "minimum": 0.83697457527265, "maximum": 1.0469029446713047, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4728727950196417, "standard_deviation": 0.1336476228247835, "minimum": 0.3363382166307261, "maximum": 0.7066787283157895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06812660503580929, "standard_deviation": 0.018013814592704572, "minimum": 0.05302398312997936, "maximum": 0.10770082002452537, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.314580503988697, "standard_deviation": 0.039355461509144846, "minimum": 0.2526438902989892, "maximum": 0.39232896756743346, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.806198099279251, "standard_deviation": 0.009549490918420485, "minimum": 5.793034984973322, "maximum": 5.825898603945208, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.517514959420722, "standard_deviation": 0.028801315658481246, "minimum": 0.4916150299773693, "maximum": 0.5665860786616508, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.18351409494962412, "standard_deviation": 0.08345998371650414, "minimum": -0.338720557655609, "maximum": -0.09984140945449597, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918836109518641, "standard_deviation": 0.0021188813473600964, "minimum": 0.9868583118449767, "maximum": 0.9936613964041575, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9771166585055724, "standard_deviation": 0.02145586521713623, "minimum": 0.9362054784293956, "maximum": 1.003017525496107, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.43267621964118774, "standard_deviation": 0.1964067525269752, "minimum": 0.26483124400878316, "maximum": 0.8552121693313963, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10220915067628886, "standard_deviation": 0.01295307513811992, "minimum": 0.09064834583515981, "maximum": 0.1323602734679973, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976315951330403, "standard_deviation": 0.00126749639729756, "minimum": 0.9947171663131038, "maximum": 0.9985247325690787, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5354639388579718, "standard_deviation": 0.030783232541154292, "minimum": 0.5016671818925461, "maximum": 0.5986808502699478, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3610824864575522, "standard_deviation": 0.11804710478147266, "minimum": 0.18548576959581123, "maximum": 0.5186353602870187, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05577791767112056, "standard_deviation": 0.014684072525270722, "minimum": 0.04325047161101707, "maximum": 0.08715891426039356, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4970-29093-0004", "language": "en", "style": "normal", "speaker": "librispeech_4970", "source_utterance_id": "4970-29093-0004", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 60000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4970-29093-0004.wav", "source_file_sha256": "7fc375eb16ad38d7cc34b102ddd253d8f13917f0d769c790f719377b1735a2f6", "decoded_pcm_sha256": "222e257a563c92e3e220ca509960e218a09ffd8f995b3f1148bba432a24fc5cf", "original_source_sha256": "48c89f99c44d1be73c92460e6d61d4c1090efe126701f7492184112d37575229", "corpus_entry_id": "librispeech_4970-29093-0004"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.0058484077453613, "predicted_mos": 3.0058484077453613, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "6053e5c91a19f7740d1b7237aee5109b050af7e2b5c2b9624935c889dea23518", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.993600306002918, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4970-29093-0004", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6053e5c91a19f7740d1b7237aee5109b050af7e2b5c2b9624935c889dea23518", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "bfe001d3c4144b13a19f4dadaababd475cde8489e0343573d7c93e2ed3b27180"}, "rows": [{"case": "librispeech_4970-29093-0004", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1945033073425293, "predicted_mos": 2.1945033073425293, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "abbca9de21f86e9f1609c3bc622246451afb7dcd03f8643e2588ea16f834d47c", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.262926059000165, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4970-29093-0004", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "abbca9de21f86e9f1609c3bc622246451afb7dcd03f8643e2588ea16f834d47c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.774875613775231, "全帯域_10ms_rms_envelope_nrmse": 0.0914479429215703, "全帯域_10ms_envelope_correlation": 0.9938758349940534, "全帯域_10ms_modulation_mae_db": 0.4177925605290504, "全帯域_40ms_log_rms_mae_db": 0.4472044636935861, "全帯域_40ms_rms_envelope_nrmse": 0.05855232495314401, "全帯域_40ms_envelope_correlation": 0.9975349145732044, "全帯域_40ms_modulation_mae_db": 0.7370938296151809, "300–1000 Hz_10ms_log_rms_mae_db": 1.045997912674788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16463724153727313, "300–1000 Hz_10ms_envelope_correlation": 0.9810942221356553, "300–1000 Hz_10ms_modulation_mae_db": 0.5489925591213065, "300–1000 Hz_40ms_log_rms_mae_db": 0.7480804646483895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10795987128819078, "300–1000 Hz_40ms_envelope_correlation": 0.9919884526910484, "300–1000 Hz_40ms_modulation_mae_db": 0.4393727026875011, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2181706384098294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09988930740005628, "1000–3000 Hz_10ms_envelope_correlation": 0.9939679005509549, "1000–3000 Hz_10ms_modulation_mae_db": 1.1286623452670657, "1000–3000 Hz_40ms_log_rms_mae_db": 0.742151316296979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056039322705772464, "1000–3000 Hz_40ms_envelope_correlation": 0.9980916332369745, "1000–3000 Hz_40ms_modulation_mae_db": 1.118640169401953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2587555670830055, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21048636782248, "3000–6000 Hz_10ms_envelope_correlation": 0.9660502654739874, "3000–6000 Hz_10ms_modulation_mae_db": 0.7288805110527505, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7321516305066699, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11775878385142625, "3000–6000 Hz_40ms_envelope_correlation": 0.9885473828916433, "3000–6000 Hz_40ms_modulation_mae_db": 0.8643091473717357, "mr_stft_db": 5.110395927276678, "band_mae_db": 0.2457335491151613, "rms_gain_error_db": -0.16529317716588746, "output_peak": 0.6449172522586692}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7904427369107707, "全帯域_10ms_rms_envelope_nrmse": 0.09358338706184861, "全帯域_10ms_envelope_correlation": 0.9936802550284741, "全帯域_10ms_modulation_mae_db": 0.5128251057703711, "全帯域_40ms_log_rms_mae_db": 0.41980195475115295, "全帯域_40ms_rms_envelope_nrmse": 0.04763858672250346, "全帯域_40ms_envelope_correlation": 0.9986209242289481, "全帯域_40ms_modulation_mae_db": 0.5656792045359426, "300–1000 Hz_10ms_log_rms_mae_db": 1.098533098232865, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.138841747691016, "300–1000 Hz_10ms_envelope_correlation": 0.9874481099060817, "300–1000 Hz_10ms_modulation_mae_db": 0.9114287479444839, "300–1000 Hz_40ms_log_rms_mae_db": 0.717303283794726, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08342715397853008, "300–1000 Hz_40ms_envelope_correlation": 0.9963964215369634, "300–1000 Hz_40ms_modulation_mae_db": 1.157388095210506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1728601948062383, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13770946560025793, "1000–3000 Hz_10ms_envelope_correlation": 0.9884787318572597, "1000–3000 Hz_10ms_modulation_mae_db": 0.7719881475304063, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7074225831858106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07213848043541328, "1000–3000 Hz_40ms_envelope_correlation": 0.9967605766255085, "1000–3000 Hz_40ms_modulation_mae_db": 0.8119002772756646, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3378996780223888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2124483620921111, "3000–6000 Hz_10ms_envelope_correlation": 0.9670342022146765, "3000–6000 Hz_10ms_modulation_mae_db": 0.37041069292581197, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7879096285093686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12072673765049048, "3000–6000 Hz_40ms_envelope_correlation": 0.9892212670185903, "3000–6000 Hz_40ms_modulation_mae_db": 0.7190264354012567, "mr_stft_db": 5.122779583432425, "band_mae_db": 0.2746528940233948, "rms_gain_error_db": -0.20261767028042726, "output_peak": 0.7131243654013883}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7764448967584969, "全帯域_10ms_rms_envelope_nrmse": 0.09169335654396986, "全帯域_10ms_envelope_correlation": 0.9936594650344659, "全帯域_10ms_modulation_mae_db": 0.2426974868837132, "全帯域_40ms_log_rms_mae_db": 0.40995388083044776, "全帯域_40ms_rms_envelope_nrmse": 0.05003579590210803, "全帯域_40ms_envelope_correlation": 0.9980156598170135, "全帯域_40ms_modulation_mae_db": 0.3584317297820523, "300–1000 Hz_10ms_log_rms_mae_db": 1.0959011241840595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17245394262043862, "300–1000 Hz_10ms_envelope_correlation": 0.9790727922303533, "300–1000 Hz_10ms_modulation_mae_db": 0.8771912083309087, "300–1000 Hz_40ms_log_rms_mae_db": 0.6848059033528057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08142701752112601, "300–1000 Hz_40ms_envelope_correlation": 0.9950483539012275, "300–1000 Hz_40ms_modulation_mae_db": 0.6960534431629434, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1872168543660206, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11207598309519645, "1000–3000 Hz_10ms_envelope_correlation": 0.992415195962953, "1000–3000 Hz_10ms_modulation_mae_db": 1.2326847194256747, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7138392210856374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061930722243568034, "1000–3000 Hz_40ms_envelope_correlation": 0.9976190038933902, "1000–3000 Hz_40ms_modulation_mae_db": 0.9301318395801018, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3224828983502037, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21525304788842287, "3000–6000 Hz_10ms_envelope_correlation": 0.964389998718503, "3000–6000 Hz_10ms_modulation_mae_db": 0.4161347854983979, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8360636759076717, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13791041073120588, "3000–6000 Hz_40ms_envelope_correlation": 0.9840907215253462, "3000–6000 Hz_40ms_modulation_mae_db": 0.5305611727145253, "mr_stft_db": 5.114866406008306, "band_mae_db": 0.12248660487145657, "rms_gain_error_db": -0.028879643235063417, "output_peak": 0.7502322171085269}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.78123493138137, "全帯域_10ms_rms_envelope_nrmse": 0.07926765781392854, "全帯域_10ms_envelope_correlation": 0.9952616378364799, "全帯域_10ms_modulation_mae_db": 0.26583384924698844, "全帯域_40ms_log_rms_mae_db": 0.405508687094172, "全帯域_40ms_rms_envelope_nrmse": 0.03665784213125228, "全帯域_40ms_envelope_correlation": 0.9989429116095203, "全帯域_40ms_modulation_mae_db": 0.1418024945765361, "300–1000 Hz_10ms_log_rms_mae_db": 1.012916453917737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12169541890582264, "300–1000 Hz_10ms_envelope_correlation": 0.9895174939229908, "300–1000 Hz_10ms_modulation_mae_db": 0.2611451723691109, "300–1000 Hz_40ms_log_rms_mae_db": 0.6874137893384367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05719831253738758, "300–1000 Hz_40ms_envelope_correlation": 0.9975963323068399, "300–1000 Hz_40ms_modulation_mae_db": 0.2803171917529294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1865665054514243, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11475073828973975, "1000–3000 Hz_10ms_envelope_correlation": 0.9920092371789648, "1000–3000 Hz_10ms_modulation_mae_db": 1.1973442188880223, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6802288268426859, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057610587514396024, "1000–3000 Hz_40ms_envelope_correlation": 0.9979273610403647, "1000–3000 Hz_40ms_modulation_mae_db": 0.9050083978781142, "3000–6000 Hz_10ms_log_rms_mae_db": 1.334044422433618, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2023524977065468, "3000–6000 Hz_10ms_envelope_correlation": 0.968335819581589, "3000–6000 Hz_10ms_modulation_mae_db": 0.6800129640922425, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7772011449420215, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11111811505368747, "3000–6000 Hz_40ms_envelope_correlation": 0.989668455576772, "3000–6000 Hz_40ms_modulation_mae_db": 0.6619766539582207, "mr_stft_db": 5.118226891204205, "band_mae_db": 0.14120297873386956, "rms_gain_error_db": -0.03893858578545969, "output_peak": 0.6634661724136754}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7827477136833647, "全帯域_10ms_rms_envelope_nrmse": 0.08829970875047227, "全帯域_10ms_envelope_correlation": 0.9941168850326588, "全帯域_10ms_modulation_mae_db": 0.21305841665099656, "全帯域_40ms_log_rms_mae_db": 0.3999748062690177, "全帯域_40ms_rms_envelope_nrmse": 0.05622350848587346, "全帯域_40ms_envelope_correlation": 0.9974962666911396, "全帯域_40ms_modulation_mae_db": 0.41185521715691165, "300–1000 Hz_10ms_log_rms_mae_db": 1.0283780427905245, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15120673082731148, "300–1000 Hz_10ms_envelope_correlation": 0.9838076772498521, "300–1000 Hz_10ms_modulation_mae_db": 0.1700151671696356, "300–1000 Hz_40ms_log_rms_mae_db": 0.7009723302618405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09808436412272743, "300–1000 Hz_40ms_envelope_correlation": 0.9929302350433812, "300–1000 Hz_40ms_modulation_mae_db": 0.6832323530296486, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1612990654865147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10651817552823092, "1000–3000 Hz_10ms_envelope_correlation": 0.9931483507871265, "1000–3000 Hz_10ms_modulation_mae_db": 0.6673061613357721, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6737208965945078, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0705320515331581, "1000–3000 Hz_40ms_envelope_correlation": 0.9969252361997789, "1000–3000 Hz_40ms_modulation_mae_db": 0.7374015834505373, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3048302493066948, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2042122690561078, "3000–6000 Hz_10ms_envelope_correlation": 0.9680260522375438, "3000–6000 Hz_10ms_modulation_mae_db": 0.6357445991026673, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7754229451587495, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11690524069865733, "3000–6000 Hz_40ms_envelope_correlation": 0.9886910282968457, "3000–6000 Hz_40ms_modulation_mae_db": 0.900019843060599, "mr_stft_db": 5.105261945242871, "band_mae_db": 0.2993724096015386, "rms_gain_error_db": -0.047770048308768166, "output_peak": 0.7541485433036902}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7794315805189708, "全帯域_10ms_rms_envelope_nrmse": 0.08866785804041341, "全帯域_10ms_envelope_correlation": 0.9940733176189712, "全帯域_10ms_modulation_mae_db": 0.5969289016239226, "全帯域_40ms_log_rms_mae_db": 0.4169537911834684, "全帯域_40ms_rms_envelope_nrmse": 0.045868409503843584, "全帯域_40ms_envelope_correlation": 0.9983336136390295, "全帯域_40ms_modulation_mae_db": 0.3584205667622135, "300–1000 Hz_10ms_log_rms_mae_db": 1.1154564627557892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14191986426060604, "300–1000 Hz_10ms_envelope_correlation": 0.9861045587758601, "300–1000 Hz_10ms_modulation_mae_db": 0.4139987413556912, "300–1000 Hz_40ms_log_rms_mae_db": 0.7546019133496977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07833472605518296, "300–1000 Hz_40ms_envelope_correlation": 0.9961266030816294, "300–1000 Hz_40ms_modulation_mae_db": 0.5178774590525633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1723823652307777, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12129981084863319, "1000–3000 Hz_10ms_envelope_correlation": 0.9916234122620815, "1000–3000 Hz_10ms_modulation_mae_db": 0.3247786588031885, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6699676494650953, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06128754862335631, "1000–3000 Hz_40ms_envelope_correlation": 0.9980442826755925, "1000–3000 Hz_40ms_modulation_mae_db": 0.2775035746048324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3550896059849533, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21275487691845202, "3000–6000 Hz_10ms_envelope_correlation": 0.9649760810939196, "3000–6000 Hz_10ms_modulation_mae_db": 0.8986834654103965, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8353878681565181, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12628697664844774, "3000–6000 Hz_40ms_envelope_correlation": 0.9865374347337541, "3000–6000 Hz_40ms_modulation_mae_db": 0.9603887602792086, "mr_stft_db": 5.12527582951342, "band_mae_db": 0.26558160407190523, "rms_gain_error_db": -0.017736983050221207, "output_peak": 0.7495778302868192}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7828192457054656, "全帯域_10ms_rms_envelope_nrmse": 0.0881779808702637, "全帯域_10ms_envelope_correlation": 0.9941732403480427, "全帯域_10ms_modulation_mae_db": 0.2584360584214942, "全帯域_40ms_log_rms_mae_db": 0.40635747930857086, "全帯域_40ms_rms_envelope_nrmse": 0.048965078484243814, "全帯域_40ms_envelope_correlation": 0.9981727297997214, "全帯域_40ms_modulation_mae_db": 0.5833734317949362, "300–1000 Hz_10ms_log_rms_mae_db": 1.0715681621049358, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14259159759056875, "300–1000 Hz_10ms_envelope_correlation": 0.9857853148034404, "300–1000 Hz_10ms_modulation_mae_db": 0.5363358963283485, "300–1000 Hz_40ms_log_rms_mae_db": 0.691299020171515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08178368095745771, "300–1000 Hz_40ms_envelope_correlation": 0.9954402893675206, "300–1000 Hz_40ms_modulation_mae_db": 0.7849807885306447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1864491296312578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10365235480380443, "1000–3000 Hz_10ms_envelope_correlation": 0.9934950338655778, "1000–3000 Hz_10ms_modulation_mae_db": 0.34999618403230853, "1000–3000 Hz_40ms_log_rms_mae_db": 0.711492020111743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05975587248359329, "1000–3000 Hz_40ms_envelope_correlation": 0.9978120932497885, "1000–3000 Hz_40ms_modulation_mae_db": 0.35262005846319416, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3065665623606217, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20882306520072524, "3000–6000 Hz_10ms_envelope_correlation": 0.9666914414791827, "3000–6000 Hz_10ms_modulation_mae_db": 0.2787656804880945, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7709714743230562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11138985597366774, "3000–6000 Hz_40ms_envelope_correlation": 0.9898710159138636, "3000–6000 Hz_40ms_modulation_mae_db": 0.5173391947672622, "mr_stft_db": 5.119304682778071, "band_mae_db": 0.21083165701373427, "rms_gain_error_db": -0.09635060065906442, "output_peak": 0.7888368465535991}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7704859478447407, "全帯域_10ms_rms_envelope_nrmse": 0.09317918813234853, "全帯域_10ms_envelope_correlation": 0.9935849024470149, "全帯域_10ms_modulation_mae_db": 0.5618115449062009, "全帯域_40ms_log_rms_mae_db": 0.40368929381688884, "全帯域_40ms_rms_envelope_nrmse": 0.048017865027075654, "全帯域_40ms_envelope_correlation": 0.998393258777682, "全帯域_40ms_modulation_mae_db": 0.2600982869226603, "300–1000 Hz_10ms_log_rms_mae_db": 1.0663724103707335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14346091388429913, "300–1000 Hz_10ms_envelope_correlation": 0.9856588540845961, "300–1000 Hz_10ms_modulation_mae_db": 0.5164531620696292, "300–1000 Hz_40ms_log_rms_mae_db": 0.666327104728759, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08066783286099817, "300–1000 Hz_40ms_envelope_correlation": 0.9956750972804518, "300–1000 Hz_40ms_modulation_mae_db": 0.4950430031189475, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2377870833478268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11423145429645108, "1000–3000 Hz_10ms_envelope_correlation": 0.9920895574073472, "1000–3000 Hz_10ms_modulation_mae_db": 0.28802107119867565, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7141435547840262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05976780849025912, "1000–3000 Hz_40ms_envelope_correlation": 0.9977707177869094, "1000–3000 Hz_40ms_modulation_mae_db": 0.1663059829319701, "3000–6000 Hz_10ms_log_rms_mae_db": 1.322499030245107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20236164214204402, "3000–6000 Hz_10ms_envelope_correlation": 0.9672296683098138, "3000–6000 Hz_10ms_modulation_mae_db": 0.45491263975240726, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8148886762488831, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1158676631708034, "3000–6000 Hz_40ms_envelope_correlation": 0.988249028017072, "3000–6000 Hz_40ms_modulation_mae_db": 0.7102730929307662, "mr_stft_db": 5.107072047144676, "band_mae_db": 0.17530635511858617, "rms_gain_error_db": -0.15551038242860515, "output_peak": 0.6741035477451541}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.395401484272716, "spectral_convergence": 0.3663779759223918, "spectral_convergence_status": "ok", "amplitude_floor": 1.4952467880734047e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.957983340248814, "spectral_convergence": 0.3771698567493733, "spectral_convergence_status": "ok", "amplitude_floor": 1.087083783284318e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.977802957308505, "spectral_convergence": 0.3874080872620942, "spectral_convergence_status": "ok", "amplitude_floor": 6.922347653241366e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.706318352359072e-05, "output_power": 6.245291208459506e-05, "delta_db": 0.39196696628502276, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006570402135178566, "output_power": 0.0006486195598177092, "delta_db": -0.05601909801824952, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010679696452407233, "output_power": 0.0009569216823336384, "delta_db": -0.47682513853169617, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007107211898605409, "output_power": 0.0006860460351227732, "delta_db": -0.153460051952247, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005187917577577873, "output_power": 0.0005371645682625635, "delta_db": 0.15114290919248677, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.297230867177819e-05, "output_power": 8.778733949817671e-05, "delta_db": 0.24498713071126568, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2457335491151613, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16529317716588746, "pitch": null, "output_rms_dbfs": -25.259306489104894, "output_peak": 0.6449172522586692, "reference_rms_dbfs": -25.09401331193901, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.110395927276678}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.0002227499093660225, "log_rms_mae_db": 0.774875613775231, "rms_envelope_nrmse": 0.0914479429215703, "envelope_correlation": 0.9938758349940534, "rms_gain_error_db": -0.16529317430030982, "reference_mean_power": 0.003098689875324867, "output_mean_power": 0.0029829693822108704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.065525005901086e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018487969313489734, "output_power": 0.0017818647308087117, "reference_active": true, "delta_db": -0.16014479878121524}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000254937700664643, "output_power": 0.00022129449440631903, "reference_active": true, "delta_db": -0.6146345526627007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5927620235575996e-05, "output_power": 4.011329011145496e-05, "reference_active": true, "delta_db": 0.4785983301432353}], "active_band_count": 3, "active_band_mae_db": 0.4177925605290504}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00020550660315895724, "log_rms_mae_db": 0.4472044636935861, "rms_envelope_nrmse": 0.05855232495314401, "envelope_correlation": 0.9975349145732044, "rms_gain_error_db": -0.1652931801754341, "reference_mean_power": 0.003123712899469445, "output_mean_power": 0.003007057918099119}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 2.850551559013711e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017535539962877898, "output_power": 0.0016927312446947717, "reference_active": true, "delta_db": -0.15331132876867204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011841366649455362, "output_power": 0.00010347829176506653, "reference_active": true, "delta_db": -0.5855257786866277}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7582705881956915e-06, "output_power": 2.4679147267939112e-06, "reference_active": true, "delta_db": 1.472444381390243}], "active_band_count": 3, "active_band_mae_db": 0.7370938296151809}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00016827060195170467, "log_rms_mae_db": 1.045997912674788, "rms_envelope_nrmse": 0.16463724153727313, "envelope_correlation": 0.9810942221356553, "rms_gain_error_db": -0.31630115740937187, "reference_mean_power": 0.001201501254742638, "output_mean_power": 0.0011171053016774562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.2448894715706734e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007295274990436897, "output_power": 0.0006884444366973898, "reference_active": true, "delta_db": -0.25172772212866984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014476368795830424, "output_power": 0.0001223997633316787, "reference_active": true, "delta_db": -0.72879060466841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.785748904828138e-05, "output_power": 2.389439470442814e-05, "reference_active": true, "delta_db": -0.6664593505668397}], "active_band_count": 3, "active_band_mae_db": 0.5489925591213065}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012583494842815252, "log_rms_mae_db": 0.7480804646483895, "rms_envelope_nrmse": 0.10795987128819078, "envelope_correlation": 0.9919884526910484, "rms_gain_error_db": -0.31630115737791165, "reference_mean_power": 0.0012112038213410922, "output_mean_power": 0.0011261263397779407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1451125517965788e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007097609595329509, "output_power": 0.0006619872443932516, "reference_active": true, "delta_db": -0.3026248601263997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.709378020571589e-05, "output_power": 5.5208481093229826e-05, "reference_active": true, "delta_db": -0.846764627285041}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7774397828761626e-06, "output_power": 1.7097083561821023e-06, "reference_active": true, "delta_db": -0.16872862065106253}], "active_band_count": 3, "active_band_mae_db": 0.4393727026875011}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015339527329010622, "log_rms_mae_db": 1.2181706384098294, "rms_envelope_nrmse": 0.09988930740005628, "envelope_correlation": 0.9939679005509549, "rms_gain_error_db": -0.09250933190808375, "reference_mean_power": 0.0009604249251386078, "output_mean_power": 0.0009401832057628199}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.606398277844274e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000996843421347699, "output_power": 0.000964228869140457, "reference_active": true, "delta_db": -0.1444681691683408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073134134931949, "output_power": 9.954124651636117e-05, "reference_active": true, "delta_db": -0.4626748585070529}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0381145318442363e-05, "output_power": 1.9684739472823727e-05, "reference_active": true, "delta_db": 2.7788440081258035}], "active_band_count": 3, "active_band_mae_db": 1.1286623452670657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.0001445669043392124, "log_rms_mae_db": 0.742151316296979, "rms_envelope_nrmse": 0.056039322705772464, "envelope_correlation": 0.9980916332369745, "rms_gain_error_db": -0.09250934634434242, "reference_mean_power": 0.0009681806844812572, "output_mean_power": 0.0009477755031566547}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.5548349296058167e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009718305477517134, "output_power": 0.0009450817052821119, "reference_active": true, "delta_db": -0.12121189955798091}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.313569881232493e-05, "output_power": 4.869315107559919e-05, "reference_active": true, "delta_db": -0.3791851664549931}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.152693877304456e-07, "output_power": 1.1874563434217947e-06, "reference_active": true, "delta_db": 2.855523442192885}], "active_band_count": 3, "active_band_mae_db": 1.118640169401953}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.2745097516618104e-05, "log_rms_mae_db": 1.2587555670830055, "rms_envelope_nrmse": 0.21048636782248, "envelope_correlation": 0.9660502654739874, "rms_gain_error_db": 0.17179037792963422, "reference_mean_power": 0.0001383637841793036, "output_mean_power": 0.00014394661911650345}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 9.868150581916701e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.732185415785845e-05, "output_power": 5.892530789175033e-05, "reference_active": true, "delta_db": 0.1198163061019715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6922402540887435e-05, "output_power": 1.5805621004139603e-05, "reference_active": true, "delta_db": -0.29650457621210646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.249240297758961e-06, "output_power": 6.387683017304142e-06, "reference_active": true, "delta_db": 1.7703206508441738}], "active_band_count": 3, "active_band_mae_db": 0.7288805110527505}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.734940150914352e-05, "log_rms_mae_db": 0.7321516305066699, "rms_envelope_nrmse": 0.11775878385142625, "envelope_correlation": 0.9885473828916433, "rms_gain_error_db": 0.17179040339489188, "reference_mean_power": 0.00013948111618043138, "output_mean_power": 0.00014510903515387372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 8.397728752995274e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.502560023531223e-05, "output_power": 5.7783485089269926e-05, "reference_active": true, "delta_db": 0.2123894319338199}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.667910609926144e-06, "output_power": 7.622892487485068e-06, "reference_active": true, "delta_db": -0.025572468007911128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2898812052596812e-07, "output_power": 3.938306450399894e-07, "reference_active": true, "delta_db": 2.354965542173476}], "active_band_count": 3, "active_band_mae_db": 0.8643091473717357}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9085266868278109, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d5cb662c22f576adce414ae4911a844e0fa69b1d23146fc99e6debbbb8e4cf45", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9921534274840332, "standard_deviation": 0.0016886084560489472, "minimum": 0.9884787318572597, "maximum": 0.9939679005509549, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1903414795912362, "standard_deviation": 0.02545668147074132, "minimum": 1.1612990654865147, "maximum": 1.2377870833478268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7450976883101392, "standard_deviation": 0.4031761084585899, "minimum": 0.28802107119867565, "maximum": 1.2326847194256747, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11376591123279625, "standard_deviation": 0.011841028800452411, "minimum": 0.09988930740005628, "maximum": 0.13770946560025793, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9976188630885384, "standard_deviation": 0.0005039608973597018, "minimum": 0.9967605766255085, "maximum": 0.9980916332369745, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7016207585458106, "standard_deviation": 0.024836254294037678, "minimum": 0.6699676494650953, "maximum": 0.742151316296979, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6624389854482959, "standard_deviation": 0.3498750730760379, "minimum": 0.1663059829319701, "maximum": 1.118640169401953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06238279925368957, "standard_deviation": 0.005852230297039235, "minimum": 0.056039322705772464, "maximum": 0.07213848043541328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.966591691138652, "standard_deviation": 0.001386822880913639, "minimum": 0.964389998718503, "maximum": 0.968335819581589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3177710017233242, "standard_deviation": 0.028990260283300876, "minimum": 1.2587555670830055, "maximum": 1.3550896059849533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5579431672903461, "standard_deviation": 0.21047939470527535, "minimum": 0.2787656804880945, "maximum": 0.8986834654103965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20858651610336124, "standard_deviation": 0.005030524011805447, "minimum": 0.2023524977065468, "maximum": 0.21525304788842287, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9881095417467358, "standard_deviation": 0.0019250477166211483, "minimum": 0.9840907215253462, "maximum": 0.9898710159138636, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7912496304691172, "standard_deviation": 0.03561312744109797, "minimum": 0.7321516305066699, "maximum": 0.8360636759076717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7329867875604468, "standard_deviation": 0.16479013374856802, "minimum": 0.5173391947672622, "maximum": 0.9603887602792086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11974547297229829, "standard_deviation": 0.008823815310770738, "minimum": 0.11111811505368747, "maximum": 0.13791041073120588, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9848111278886037, "standard_deviation": 0.0033847345624164415, "minimum": 0.9790727922303533, "maximum": 0.9895174939229908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0668904583789292, "standard_deviation": 0.03596939022564589, "minimum": 1.012916453917737, "maximum": 1.1154564627557892, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5294450818361393, "standard_deviation": 0.26239006659796743, "minimum": 0.1700151671696356, "maximum": 0.9114287479444839, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14710093216466696, "standard_deviation": 0.01577174868629061, "minimum": 0.12169541890582264, "maximum": 0.17245394262043862, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9951502231511328, "standard_deviation": 0.001843324885463076, "minimum": 0.9919884526910484, "maximum": 0.9975963323068399, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7063504762057713, "standard_deviation": 0.03132699632634656, "minimum": 0.666327104728759, "maximum": 0.7546019133496977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6317831295682105, "standard_deviation": 0.2669216686053716, "minimum": 0.2803171917529294, "maximum": 1.157388095210506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08361036991520009, "standard_deviation": 0.014854628096124768, "minimum": 0.05719831253738758, "maximum": 0.10795987128819078, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21689600656870583, "standard_deviation": 0.06519109427012607, "minimum": 0.12248660487145657, "maximum": 0.2993724096015386, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.115397914075082, "standard_deviation": 0.007296657808992009, "minimum": 5.105261945242871, "maximum": 5.12527582951342, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7173008468839404, "standard_deviation": 0.05157856334798238, "minimum": 0.6449172522586692, "maximum": 0.7888368465535991, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09413713636418711, "standard_deviation": 0.07160915126536402, "minimum": -0.20261767028042726, "maximum": -0.017736983050221207, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940531922925202, "standard_deviation": 0.0005377291285228869, "minimum": 0.9935849024470149, "maximum": 0.9952616378364799, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7798103333223013, "standard_deviation": 0.006041812420913397, "minimum": 0.7704859478447407, "maximum": 0.7904427369107707, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3836729905040922, "standard_deviation": 0.1574845207496507, "minimum": 0.21305841665099656, "maximum": 0.5969289016239226, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0892896350168519, "standard_deviation": 0.00458325310034139, "minimum": 0.07926765781392854, "maximum": 0.09358338706184861, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981887848920323, "standard_deviation": 0.000500798556366293, "minimum": 0.9974962666911396, "maximum": 0.9989429116095203, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4136805446184131, "standard_deviation": 0.015090232770361728, "minimum": 0.3999748062690177, "maximum": 0.4472044636935861, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4270943451433042, "standard_deviation": 0.19223148011060104, "minimum": 0.1418024945765361, "maximum": 0.7370938296151809, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04899492640125554, "standard_deviation": 0.006648884210341393, "minimum": 0.03665784213125228, "maximum": 0.05855232495314401, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0004", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.356605291366577, "predicted_mos": 2.356605291366577, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "796aa6454c96dcc4cc1609bb3bde48eeff1a52c9918fa188d85bc06b5c686f6a", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7403007997199893, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7642017020772452, "全帯域_10ms_rms_envelope_nrmse": 0.08853187326346293, "全帯域_10ms_envelope_correlation": 0.9943625676896017, "全帯域_10ms_modulation_mae_db": 0.42438796196977185, "全帯域_40ms_log_rms_mae_db": 0.4332025545115897, "全帯域_40ms_rms_envelope_nrmse": 0.05440659332236477, "全帯域_40ms_envelope_correlation": 0.9980164283691977, "全帯域_40ms_modulation_mae_db": 0.7327334979841623, "300–1000 Hz_10ms_log_rms_mae_db": 1.00128705879046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1529440045574033, "300–1000 Hz_10ms_envelope_correlation": 0.9837860299672281, "300–1000 Hz_10ms_modulation_mae_db": 0.5965070331469953, "300–1000 Hz_40ms_log_rms_mae_db": 0.704725444051268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0983336003423571, "300–1000 Hz_40ms_envelope_correlation": 0.993481281235117, "300–1000 Hz_40ms_modulation_mae_db": 0.5151991494904161, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1900419023317845, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10687266515359972, "1000–3000 Hz_10ms_envelope_correlation": 0.9931618982429994, "1000–3000 Hz_10ms_modulation_mae_db": 1.2228113098588713, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6739239676744372, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059341949488304006, "1000–3000 Hz_40ms_envelope_correlation": 0.9979690620972932, "1000–3000 Hz_40ms_modulation_mae_db": 1.1840416514250862, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2479701794944005, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19811357002423557, "3000–6000 Hz_10ms_envelope_correlation": 0.9693315585181727, "3000–6000 Hz_10ms_modulation_mae_db": 0.6287129901120893, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7219183011058595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10642833953551485, "3000–6000 Hz_40ms_envelope_correlation": 0.9903062089361592, "3000–6000 Hz_40ms_modulation_mae_db": 0.8327502793617109, "mr_stft_db": 5.073339740825463, "band_mae_db": 0.19801966040929384, "rms_gain_error_db": -0.19236551156555814, "output_peak": 0.6449128279545484}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7793646203279915, "全帯域_10ms_rms_envelope_nrmse": 0.09558778035701275, "全帯域_10ms_envelope_correlation": 0.9933224407152773, "全帯域_10ms_modulation_mae_db": 0.4291409868580323, "全帯域_40ms_log_rms_mae_db": 0.41018366580461385, "全帯域_40ms_rms_envelope_nrmse": 0.05001086304028599, "全帯域_40ms_envelope_correlation": 0.9983515251152079, "全帯域_40ms_modulation_mae_db": 0.47967621072600125, "300–1000 Hz_10ms_log_rms_mae_db": 1.0959538076509465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1358517459573579, "300–1000 Hz_10ms_envelope_correlation": 0.9875550698890073, "300–1000 Hz_10ms_modulation_mae_db": 0.8765045301814599, "300–1000 Hz_40ms_log_rms_mae_db": 0.7220496113635159, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08478033266783537, "300–1000 Hz_40ms_envelope_correlation": 0.9955869833218024, "300–1000 Hz_40ms_modulation_mae_db": 1.1349604659283175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1578761041063412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13171961979875965, "1000–3000 Hz_10ms_envelope_correlation": 0.9894680212309037, "1000–3000 Hz_10ms_modulation_mae_db": 0.8233628526767879, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6323709191009232, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06723821350512495, "1000–3000 Hz_40ms_envelope_correlation": 0.9972132073908885, "1000–3000 Hz_40ms_modulation_mae_db": 0.7835789668363974, "3000–6000 Hz_10ms_log_rms_mae_db": 1.366874472584873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2024202283891365, "3000–6000 Hz_10ms_envelope_correlation": 0.9692443490645014, "3000–6000 Hz_10ms_modulation_mae_db": 0.2177280176972792, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7871720258318032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1110108741908349, "3000–6000 Hz_40ms_envelope_correlation": 0.9903389061938862, "3000–6000 Hz_40ms_modulation_mae_db": 0.5721588385107327, "mr_stft_db": 5.091568947020058, "band_mae_db": 0.23213724679198733, "rms_gain_error_db": -0.1848663870403591, "output_peak": 0.7104311479943319}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7506146006671502, "全帯域_10ms_rms_envelope_nrmse": 0.08694474819047156, "全帯域_10ms_envelope_correlation": 0.9943010551597189, "全帯域_10ms_modulation_mae_db": 0.22937373127601493, "全帯域_40ms_log_rms_mae_db": 0.39072364970069284, "全帯域_40ms_rms_envelope_nrmse": 0.04642518908025645, "全帯域_40ms_envelope_correlation": 0.9983028291214763, "全帯域_40ms_modulation_mae_db": 0.2936513233211561, "300–1000 Hz_10ms_log_rms_mae_db": 1.062236426598191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15947675717214826, "300–1000 Hz_10ms_envelope_correlation": 0.9821000925438289, "300–1000 Hz_10ms_modulation_mae_db": 0.8320312215219015, "300–1000 Hz_40ms_log_rms_mae_db": 0.6501364793253437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07781360037579371, "300–1000 Hz_40ms_envelope_correlation": 0.9954800562356979, "300–1000 Hz_40ms_modulation_mae_db": 0.6525186285651704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1957383048024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10615252302425716, "1000–3000 Hz_10ms_envelope_correlation": 0.9931781360353276, "1000–3000 Hz_10ms_modulation_mae_db": 1.261756666538535, "1000–3000 Hz_40ms_log_rms_mae_db": 0.707445470555274, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05943652524936877, "1000–3000 Hz_40ms_envelope_correlation": 0.9978131469589036, "1000–3000 Hz_40ms_modulation_mae_db": 0.9448803324722498, "3000–6000 Hz_10ms_log_rms_mae_db": 1.261715851337443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19773220263544347, "3000–6000 Hz_10ms_envelope_correlation": 0.9694000992522404, "3000–6000 Hz_10ms_modulation_mae_db": 0.2296850752073929, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7809333593000964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12200547136814825, "3000–6000 Hz_40ms_envelope_correlation": 0.987218923586805, "3000–6000 Hz_40ms_modulation_mae_db": 0.37396611263858426, "mr_stft_db": 5.086300496517377, "band_mae_db": 0.09611157231047911, "rms_gain_error_db": -0.04518228021065552, "output_peak": 0.7539570088052469}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7665173589118768, "全帯域_10ms_rms_envelope_nrmse": 0.07597038978435643, "全帯域_10ms_envelope_correlation": 0.995719808129573, "全帯域_10ms_modulation_mae_db": 0.31053199105578744, "全帯域_40ms_log_rms_mae_db": 0.41880707558768665, "全帯域_40ms_rms_envelope_nrmse": 0.039357773567611706, "全帯域_40ms_envelope_correlation": 0.9988807198390949, "全帯域_40ms_modulation_mae_db": 0.44201304491316923, "300–1000 Hz_10ms_log_rms_mae_db": 1.037754736156866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12435980670678443, "300–1000 Hz_10ms_envelope_correlation": 0.9891229742521583, "300–1000 Hz_10ms_modulation_mae_db": 0.4533547803075772, "300–1000 Hz_40ms_log_rms_mae_db": 0.6906230059249426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06133013753713466, "300–1000 Hz_40ms_envelope_correlation": 0.9973858489720835, "300–1000 Hz_40ms_modulation_mae_db": 0.5582346917233049, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1652699148776893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1080034627868118, "1000–3000 Hz_10ms_envelope_correlation": 0.99294817833163, "1000–3000 Hz_10ms_modulation_mae_db": 0.7092501296122711, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6505116558426884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05920854208163496, "1000–3000 Hz_40ms_envelope_correlation": 0.9978832018548772, "1000–3000 Hz_40ms_modulation_mae_db": 0.4530636774890661, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3164057772935245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19386065087060128, "3000–6000 Hz_10ms_envelope_correlation": 0.9706658004766415, "3000–6000 Hz_10ms_modulation_mae_db": 0.5557630989999524, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7529087750624359, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10121508063706024, "3000–6000 Hz_40ms_envelope_correlation": 0.9912950671157015, "3000–6000 Hz_40ms_modulation_mae_db": 0.564394997219886, "mr_stft_db": 5.083261205201718, "band_mae_db": 0.13206880457811715, "rms_gain_error_db": -0.1100095981294272, "output_peak": 0.6788705802894215}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7711536320241968, "全帯域_10ms_rms_envelope_nrmse": 0.09548511246812684, "全帯域_10ms_envelope_correlation": 0.9931232090035886, "全帯域_10ms_modulation_mae_db": 0.2367870406664614, "全帯域_40ms_log_rms_mae_db": 0.39493277261304244, "全帯域_40ms_rms_envelope_nrmse": 0.0623907813799574, "全帯域_40ms_envelope_correlation": 0.9969299402791058, "全帯域_40ms_modulation_mae_db": 0.3897472604514782, "300–1000 Hz_10ms_log_rms_mae_db": 1.0505333074045327, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15722222615312473, "300–1000 Hz_10ms_envelope_correlation": 0.982454630560311, "300–1000 Hz_10ms_modulation_mae_db": 0.15932614167785344, "300–1000 Hz_40ms_log_rms_mae_db": 0.7314957856125714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.104212505584956, "300–1000 Hz_40ms_envelope_correlation": 0.9919431463937639, "300–1000 Hz_40ms_modulation_mae_db": 0.5514664577592648, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1707116302191516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11759737826136547, "1000–3000 Hz_10ms_envelope_correlation": 0.9916410826774184, "1000–3000 Hz_10ms_modulation_mae_db": 0.6677955232624592, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6452679801885819, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07191976414135237, "1000–3000 Hz_40ms_envelope_correlation": 0.9968500105700456, "1000–3000 Hz_40ms_modulation_mae_db": 0.6267531849139788, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2714141072719851, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19266497503469635, "3000–6000 Hz_10ms_envelope_correlation": 0.9710986718330351, "3000–6000 Hz_10ms_modulation_mae_db": 0.4016591815897577, "3000–6000 Hz_40ms_log_rms_mae_db": 0.740176113460842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10360683989364226, "3000–6000 Hz_40ms_envelope_correlation": 0.9908738361323478, "3000–6000 Hz_40ms_modulation_mae_db": 0.6572864569954923, "mr_stft_db": 5.076497064561269, "band_mae_db": 0.2848417449076041, "rms_gain_error_db": -0.06771714291282749, "output_peak": 0.7560356662639514}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7912950113211835, "全帯域_10ms_rms_envelope_nrmse": 0.09455305818452474, "全帯域_10ms_envelope_correlation": 0.9932537955043456, "全帯域_10ms_modulation_mae_db": 0.3936590320033479, "全帯域_40ms_log_rms_mae_db": 0.4318258476236154, "全帯域_40ms_rms_envelope_nrmse": 0.046683593939179924, "全帯域_40ms_envelope_correlation": 0.9982931447515817, "全帯域_40ms_modulation_mae_db": 0.23770387111890937, "300–1000 Hz_10ms_log_rms_mae_db": 1.0857500507310296, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14136971199581833, "300–1000 Hz_10ms_envelope_correlation": 0.9862910385375051, "300–1000 Hz_10ms_modulation_mae_db": 0.470492586617079, "300–1000 Hz_40ms_log_rms_mae_db": 0.7497779297582071, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07874780475866831, "300–1000 Hz_40ms_envelope_correlation": 0.9961793575307378, "300–1000 Hz_40ms_modulation_mae_db": 0.4717965782335045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1914040256803158, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13138285350259266, "1000–3000 Hz_10ms_envelope_correlation": 0.9898162323021871, "1000–3000 Hz_10ms_modulation_mae_db": 0.5088677726070692, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6409587160408599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060485692702592714, "1000–3000 Hz_40ms_envelope_correlation": 0.9978691468318913, "1000–3000 Hz_40ms_modulation_mae_db": 0.4473842775372492, "3000–6000 Hz_10ms_log_rms_mae_db": 1.350891353553126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1953790064033915, "3000–6000 Hz_10ms_envelope_correlation": 0.9699985483603093, "3000–6000 Hz_10ms_modulation_mae_db": 0.5506894089093195, "3000–6000 Hz_40ms_log_rms_mae_db": 0.788488946941886, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10909307999856972, "3000–6000 Hz_40ms_envelope_correlation": 0.9897153883994044, "3000–6000 Hz_40ms_modulation_mae_db": 0.619653392924299, "mr_stft_db": 5.085719026928177, "band_mae_db": 0.2196508730992955, "rms_gain_error_db": -0.059611889382089736, "output_peak": 0.7495268296421056}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7676996131127212, "全帯域_10ms_rms_envelope_nrmse": 0.08684053544700354, "全帯域_10ms_envelope_correlation": 0.9943647437834732, "全帯域_10ms_modulation_mae_db": 0.1982699036086721, "全帯域_40ms_log_rms_mae_db": 0.4084492167346121, "全帯域_40ms_rms_envelope_nrmse": 0.046339739743423446, "全帯域_40ms_envelope_correlation": 0.998392783941362, "全帯域_40ms_modulation_mae_db": 0.5342455282685555, "300–1000 Hz_10ms_log_rms_mae_db": 1.0484987131414794, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14600846839467513, "300–1000 Hz_10ms_envelope_correlation": 0.985064916595543, "300–1000 Hz_10ms_modulation_mae_db": 0.5868518243977662, "300–1000 Hz_40ms_log_rms_mae_db": 0.6668813150997981, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0839353811643039, "300–1000 Hz_40ms_envelope_correlation": 0.9951605302235967, "300–1000 Hz_40ms_modulation_mae_db": 0.8034711212227167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1753210527650375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09857042539699866, "1000–3000 Hz_10ms_envelope_correlation": 0.9941274901007974, "1000–3000 Hz_10ms_modulation_mae_db": 0.2131335724686471, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6799682152005667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05537837821228081, "1000–3000 Hz_40ms_envelope_correlation": 0.9981361892936799, "1000–3000 Hz_40ms_modulation_mae_db": 0.21438798777233936, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2990183958945634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20233077057602927, "3000–6000 Hz_10ms_envelope_correlation": 0.9682027378262348, "3000–6000 Hz_10ms_modulation_mae_db": 0.21547928418729176, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7601968192333204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10391345410202017, "3000–6000 Hz_40ms_envelope_correlation": 0.9908803307610065, "3000–6000 Hz_40ms_modulation_mae_db": 0.49948230837013163, "mr_stft_db": 5.0863360989414925, "band_mae_db": 0.16759643513505576, "rms_gain_error_db": -0.10541775535887533, "output_peak": 0.7735313780031768}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7873330728178938, "全帯域_10ms_rms_envelope_nrmse": 0.09142219090251051, "全帯域_10ms_envelope_correlation": 0.9938390353523725, "全帯域_10ms_modulation_mae_db": 0.6816756478564425, "全帯域_40ms_log_rms_mae_db": 0.41393769267407404, "全帯域_40ms_rms_envelope_nrmse": 0.045415099900698606, "全帯域_40ms_envelope_correlation": 0.998596260674707, "全帯域_40ms_modulation_mae_db": 0.37526349094375594, "300–1000 Hz_10ms_log_rms_mae_db": 1.024675641940734, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14344833517906716, "300–1000 Hz_10ms_envelope_correlation": 0.9856314844198695, "300–1000 Hz_10ms_modulation_mae_db": 0.5761518105480261, "300–1000 Hz_40ms_log_rms_mae_db": 0.654965040386147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07893175450843319, "300–1000 Hz_40ms_envelope_correlation": 0.9958452025561719, "300–1000 Hz_40ms_modulation_mae_db": 0.5944883050567848, "1000–3000 Hz_10ms_log_rms_mae_db": 1.233433806814008, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11482192070316487, "1000–3000 Hz_10ms_envelope_correlation": 0.9920141332141927, "1000–3000 Hz_10ms_modulation_mae_db": 0.42854035830391024, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6798564428691146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058452250679155486, "1000–3000 Hz_40ms_envelope_correlation": 0.9978699700141714, "1000–3000 Hz_40ms_modulation_mae_db": 0.2849082303785057, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3309732838942891, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19649581733804983, "3000–6000 Hz_10ms_envelope_correlation": 0.9691084145881772, "3000–6000 Hz_10ms_modulation_mae_db": 0.32417614874509276, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8001059846757969, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10929575317805837, "3000–6000 Hz_40ms_envelope_correlation": 0.9897582067502241, "3000–6000 Hz_40ms_modulation_mae_db": 0.571743344199867, "mr_stft_db": 5.072758737515087, "band_mae_db": 0.18448821601606247, "rms_gain_error_db": -0.15760451823628066, "output_peak": 0.6752861811063371}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.349270527722449, "spectral_convergence": 0.3611299259052257, "spectral_convergence_status": "ok", "amplitude_floor": 1.4952467880734047e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.929804244288094, "spectral_convergence": 0.37261986996126095, "spectral_convergence_status": "ok", "amplitude_floor": 1.087083783284318e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.940944450465846, "spectral_convergence": 0.38136711360803727, "spectral_convergence_status": "ok", "amplitude_floor": 6.922347653241366e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.706318352359072e-05, "output_power": 6.113550750757408e-05, "delta_db": 0.2993752474324187, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006570402135178566, "output_power": 0.0006517876637065847, "delta_db": -0.03485814575275613, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010679696452407233, "output_power": 0.0009580123779765696, "delta_db": -0.47187788567112177, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007107211898605409, "output_power": 0.0006823666635071806, "delta_db": -0.17681462388701202, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005187917577577873, "output_power": 0.0005204718889419702, "delta_db": 0.014042102601512856, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.297230867177819e-05, "output_power": 8.670580407083204e-05, "delta_db": 0.19114995711094152, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19801966040929384, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.19236551156555814, "pitch": null, "output_rms_dbfs": -25.28637882350457, "output_peak": 0.6449128279545484, "reference_rms_dbfs": -25.09401331193901, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.073339740825463}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.0002227499093660225, "log_rms_mae_db": 0.7642017020772452, "rms_envelope_nrmse": 0.08853187326346293, "envelope_correlation": 0.9943625676896017, "rms_gain_error_db": -0.19236550908440955, "reference_mean_power": 0.003098689875324867, "output_mean_power": 0.002964432474709499}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.065525005901086e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018487969313489734, "output_power": 0.0017571500324832302, "reference_active": true, "delta_db": -0.2208036675068899}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000254937700664643, "output_power": 0.0002236047854383269, "reference_active": true, "delta_db": -0.5695297070940416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5927620235575996e-05, "output_power": 4.0152399394050667e-05, "reference_active": true, "delta_db": 0.4828305113083839}], "active_band_count": 3, "active_band_mae_db": 0.42438796196977185}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00020550660315895724, "log_rms_mae_db": 0.4332025545115897, "rms_envelope_nrmse": 0.05440659332236477, "envelope_correlation": 0.9980164283691977, "rms_gain_error_db": -0.19236551833039411, "reference_mean_power": 0.003123712899469445, "output_mean_power": 0.0029883713161747306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 2.850551559013711e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017535539962877898, "output_power": 0.0016692501731989356, "reference_active": true, "delta_db": -0.2139771346312126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011841366649455362, "output_power": 0.00010472335692538246, "reference_active": true, "delta_db": -0.5335827342402756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7582705881956915e-06, "output_power": 2.4555556102939805e-06, "reference_active": true, "delta_db": 1.450640625080999}], "active_band_count": 3, "active_band_mae_db": 0.7327334979841623}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00016827060195170467, "log_rms_mae_db": 1.00128705879046, "rms_envelope_nrmse": 0.1529440045574033, "envelope_correlation": 0.9837860299672281, "rms_gain_error_db": -0.30557963533744764, "reference_mean_power": 0.001201501254742638, "output_mean_power": 0.0011198665307160182}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.2448894715706734e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007295274990436897, "output_power": 0.0006815124304664117, "reference_active": true, "delta_db": -0.2956788540344194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014476368795830424, "output_power": 0.00012077570994459847, "reference_active": true, "delta_db": -0.7868003950025104}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.785748904828138e-05, "output_power": 2.3672154323506776e-05, "reference_active": true, "delta_db": -0.707041850404056}], "active_band_count": 3, "active_band_mae_db": 0.5965070331469953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012583494842815252, "log_rms_mae_db": 0.704725444051268, "rms_envelope_nrmse": 0.0983336003423571, "envelope_correlation": 0.993481281235117, "rms_gain_error_db": -0.3055796352901467, "reference_mean_power": 0.0012112038213410922, "output_mean_power": 0.0011289098667654279}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1451125517965788e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007097609595329509, "output_power": 0.0006541473746209802, "reference_active": true, "delta_db": -0.3543650454127111}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.709378020571589e-05, "output_power": 5.442963190333781e-05, "reference_active": true, "delta_db": -0.9084686435568279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7774397828761626e-06, "output_power": 1.6653998469989744e-06, "reference_active": true, "delta_db": -0.28276375950170934}], "active_band_count": 3, "active_band_mae_db": 0.5151991494904161}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015339527329010622, "log_rms_mae_db": 1.1900419023317845, "rms_envelope_nrmse": 0.10687266515359972, "envelope_correlation": 0.9931618982429994, "rms_gain_error_db": -0.15790947182455878, "reference_mean_power": 0.0009604249251386078, "output_mean_power": 0.0009261311148154009}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.606398277844274e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000996843421347699, "output_power": 0.000951136830059354, "reference_active": true, "delta_db": -0.2038394827843969}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073134134931949, "output_power": 9.820212454826194e-05, "reference_active": true, "delta_db": -0.521496771860652}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0381145318442363e-05, "output_power": 2.0443489991134927e-05, "reference_active": true, "delta_db": 2.9430976749315647}], "active_band_count": 3, "active_band_mae_db": 1.2228113098588713}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.0001445669043392124, "log_rms_mae_db": 0.6739239676744372, "rms_envelope_nrmse": 0.059341949488304006, "envelope_correlation": 0.9979690620972932, "rms_gain_error_db": -0.15790949586655198, "reference_mean_power": 0.0009681806844812572, "output_mean_power": 0.0009336099347568994}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.5548349296058167e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009718305477517134, "output_power": 0.0009321839882119293, "reference_active": true, "delta_db": -0.1808890726931783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.313569881232493e-05, "output_power": 4.7934892395894443e-05, "reference_active": true, "delta_db": -0.4473463975594891}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.152693877304456e-07, "output_power": 1.2062970229828968e-06, "reference_active": true, "delta_db": 2.9238894840225917}], "active_band_count": 3, "active_band_mae_db": 1.1840416514250862}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.2745097516618104e-05, "log_rms_mae_db": 1.2479701794944005, "rms_envelope_nrmse": 0.19811357002423557, "envelope_correlation": 0.9693315585181727, "rms_gain_error_db": 0.06388673191693191, "reference_mean_power": 0.0001383637841793036, "output_mean_power": 0.00014041422408276125}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 9.868150581916701e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.732185415785845e-05, "output_power": 5.8371414603577585e-05, "reference_active": true, "delta_db": 0.07879988526228324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6922402540887435e-05, "output_power": 1.6020562127150667e-05, "reference_active": true, "delta_db": -0.2378427112058219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.249240297758961e-06, "output_power": 6.099032236794464e-06, "reference_active": true, "delta_db": 1.5694963738681624}], "active_band_count": 3, "active_band_mae_db": 0.6287129901120893}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.734940150914352e-05, "log_rms_mae_db": 0.7219183011058595, "rms_envelope_nrmse": 0.10642833953551485, "envelope_correlation": 0.9903062089361592, "rms_gain_error_db": 0.06388675256268977, "reference_mean_power": 0.00013948111618043138, "output_mean_power": 0.00014154811471763212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 8.397728752995274e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.502560023531223e-05, "output_power": 5.730439181550116e-05, "reference_active": true, "delta_db": 0.17623118962168804}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.667910609926144e-06, "output_power": 7.720742345262309e-06, "reference_active": true, "delta_db": 0.029820181583565652}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2898812052596812e-07, "output_power": 3.881797707659115e-07, "reference_active": true, "delta_db": 2.292199466879879}], "active_band_count": 3, "active_band_mae_db": 0.8327502793617109}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9108961626833514, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d07212d5feeb4abd899b111f088a44c1a2caead39f0bac61adbbca2bace6cf2c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.992044396516932, "standard_deviation": 0.001667173791352914, "minimum": 0.9894680212309037, "maximum": 0.9941274901007974, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.184974592699591, "standard_deviation": 0.0237558820092778, "minimum": 1.1578761041063412, "maximum": 1.233433806814008, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7294397731660689, "standard_deviation": 0.3675343495083556, "minimum": 0.2131335724686471, "maximum": 1.261756666538535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11439010607844376, "standard_deviation": 0.01204135594856077, "minimum": 0.09857042539699866, "maximum": 0.13171961979875965, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9977004918764689, "standard_deviation": 0.0004352287185733017, "minimum": 0.9968500105700456, "maximum": 0.9981361892936799, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6637879209340558, "standard_deviation": 0.02551237353132054, "minimum": 0.6323709191009232, "maximum": 0.707445470555274, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.617374788603109, "standard_deviation": 0.33485824014148885, "minimum": 0.21438798777233936, "maximum": 1.1840416514250862, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06143266450747675, "standard_deviation": 0.00539073337736297, "minimum": 0.05537837821228081, "maximum": 0.07191976414135237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9696312724899141, "standard_deviation": 0.0009231523238368474, "minimum": 0.9682027378262348, "maximum": 0.9710986718330351, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3056579276655256, "standard_deviation": 0.04313668756798829, "minimum": 1.2479701794944005, "maximum": 1.366874472584873, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.39048665068102195, "standard_deviation": 0.16927796046645782, "minimum": 0.21547928418729176, "maximum": 0.6287129901120893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.197374652658948, "standard_deviation": 0.003585267623143997, "minimum": 0.19266497503469635, "maximum": 0.2024202283891365, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900483584844418, "standard_deviation": 0.0012693945351286151, "minimum": 0.987218923586805, "maximum": 0.9912950671157015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.766487540701505, "standard_deviation": 0.02713315216889166, "minimum": 0.7219183011058595, "maximum": 0.8001059846757969, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.586429466277588, "standard_deviation": 0.13115085159941356, "minimum": 0.37396611263858426, "maximum": 0.8327502793617109, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1083211116129811, "standard_deviation": 0.006456142711164491, "minimum": 0.10121508063706024, "maximum": 0.12200547136814825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9852507795956814, "standard_deviation": 0.0024335757024932067, "minimum": 0.9821000925438289, "maximum": 0.9891229742521583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0508362178017798, "standard_deviation": 0.030939371904912576, "minimum": 1.00128705879046, "maximum": 1.0959538076509465, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5689024910498324, "standard_deviation": 0.22544592444312644, "minimum": 0.15932614167785344, "maximum": 0.8765045301814599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14508513201454742, "standard_deviation": 0.011639344197966854, "minimum": 0.12435980670678443, "maximum": 0.15947675717214826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9951328008086214, "standard_deviation": 0.0016864193871147886, "minimum": 0.9919431463937639, "maximum": 0.9973858489720835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6963318264402243, "standard_deviation": 0.03697487872269975, "minimum": 0.6501364793253437, "maximum": 0.7497779297582071, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.660266924747435, "standard_deviation": 0.21680870829734702, "minimum": 0.4717965782335045, "maximum": 1.1349604659283175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08351063961743528, "standard_deviation": 0.013191781307216509, "minimum": 0.06133013753713466, "maximum": 0.104212505584956, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1893643191559869, "standard_deviation": 0.05905701892510006, "minimum": 0.09611157231047911, "maximum": 0.2848417449076041, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.08197266468883, "standard_deviation": 0.006917471805162909, "minimum": 5.072758737515087, "maximum": 5.091568947020058, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7178189525073899, "standard_deviation": 0.047164810042889714, "minimum": 0.6449128279545484, "maximum": 0.7735313780031768, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11534688535450915, "standard_deviation": 0.05729081907046015, "minimum": -0.19236551156555814, "maximum": -0.04518228021065552, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940358319172438, "standard_deviation": 0.0008546658462597091, "minimum": 0.9931232090035886, "maximum": 0.995719808129573, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7722724514075324, "standard_deviation": 0.013247437234060357, "minimum": 0.7506146006671502, "maximum": 0.7912950113211835, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36297828691181633, "standard_deviation": 0.15780269798629956, "minimum": 0.1982699036086721, "maximum": 0.6816756478564425, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08941696107468367, "standard_deviation": 0.006542666315286481, "minimum": 0.07597038978435643, "maximum": 0.09558778035701275, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982204540114666, "standard_deviation": 0.0005785672709134887, "minimum": 0.9969299402791058, "maximum": 0.9988807198390949, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41275780940624085, "standard_deviation": 0.015343634075230034, "minimum": 0.39072364970069284, "maximum": 0.4332025545115897, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4356292784658985, "standard_deviation": 0.15355704440140908, "minimum": 0.23770387111890937, "maximum": 0.7327334979841623, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04887870424672229, "standard_deviation": 0.006911117672340267, "minimum": 0.039357773567611706, "maximum": 0.0623907813799574, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0004", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.8962328433990479, "predicted_mos": 1.8962328433990479, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "8e0cc08943abcf9b222fef1908c7393f6a3eac8789942e4e08ad31e4d4a68be3", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8556238003075123, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7573664281459859, "全帯域_10ms_rms_envelope_nrmse": 0.09974160200793106, "全帯域_10ms_envelope_correlation": 0.9924920932856933, "全帯域_10ms_modulation_mae_db": 0.384457376535575, "全帯域_40ms_log_rms_mae_db": 0.3814576550316515, "全帯域_40ms_rms_envelope_nrmse": 0.05430168667169141, "全帯域_40ms_envelope_correlation": 0.9976775069343691, "全帯域_40ms_modulation_mae_db": 0.28498656911518516, "300–1000 Hz_10ms_log_rms_mae_db": 1.215689263013969, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16940290339086495, "300–1000 Hz_10ms_envelope_correlation": 0.9801922297591957, "300–1000 Hz_10ms_modulation_mae_db": 0.7589704948441346, "300–1000 Hz_40ms_log_rms_mae_db": 0.7872086350174321, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0982114770878151, "300–1000 Hz_40ms_envelope_correlation": 0.9939852238135563, "300–1000 Hz_40ms_modulation_mae_db": 0.4938052036786991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3968415763064925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12493745090219621, "1000–3000 Hz_10ms_envelope_correlation": 0.9915474146874311, "1000–3000 Hz_10ms_modulation_mae_db": 0.5813165908448432, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8481124639123745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07036321344733527, "1000–3000 Hz_40ms_envelope_correlation": 0.997709630023625, "1000–3000 Hz_40ms_modulation_mae_db": 0.5794416197206899, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5236038394976068, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22745835483429266, "3000–6000 Hz_10ms_envelope_correlation": 0.9608935155449362, "3000–6000 Hz_10ms_modulation_mae_db": 0.3750356798689378, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9620776244561442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13937220828847277, "3000–6000 Hz_40ms_envelope_correlation": 0.9842747268112552, "3000–6000 Hz_40ms_modulation_mae_db": 0.46303863566381587, "mr_stft_db": 5.349868673454154, "band_mae_db": 0.3039677410605605, "rms_gain_error_db": -0.05941623123737644, "output_peak": 0.6971527316853043}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7897268765051625, "全帯域_10ms_rms_envelope_nrmse": 0.09437393319933976, "全帯域_10ms_envelope_correlation": 0.9934180397984151, "全帯域_10ms_modulation_mae_db": 0.5750600937357896, "全帯域_40ms_log_rms_mae_db": 0.42397744216830796, "全帯域_40ms_rms_envelope_nrmse": 0.04879458596712546, "全帯域_40ms_envelope_correlation": 0.9983397020946003, "全帯域_40ms_modulation_mae_db": 0.8207587834340014, "300–1000 Hz_10ms_log_rms_mae_db": 1.2585352803369678, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1638064627601039, "300–1000 Hz_10ms_envelope_correlation": 0.9810237453928862, "300–1000 Hz_10ms_modulation_mae_db": 0.20832132201238018, "300–1000 Hz_40ms_log_rms_mae_db": 0.8586796870005869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07935401063714033, "300–1000 Hz_40ms_envelope_correlation": 0.9956618811496931, "300–1000 Hz_40ms_modulation_mae_db": 0.5188069116396214, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3300984678728252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09727459803126161, "1000–3000 Hz_10ms_envelope_correlation": 0.9943176566019895, "1000–3000 Hz_10ms_modulation_mae_db": 0.563808287639286, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8603219993011921, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05008644900549994, "1000–3000 Hz_40ms_envelope_correlation": 0.9985179436473727, "1000–3000 Hz_40ms_modulation_mae_db": 0.465427787917192, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5984226067855576, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22507753872321853, "3000–6000 Hz_10ms_envelope_correlation": 0.9605602976201243, "3000–6000 Hz_10ms_modulation_mae_db": 0.43585514906884093, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9425002584488595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1323387363807981, "3000–6000 Hz_40ms_envelope_correlation": 0.9851000004014595, "3000–6000 Hz_40ms_modulation_mae_db": 0.29594920521199447, "mr_stft_db": 5.344131961875715, "band_mae_db": 0.2533478032160189, "rms_gain_error_db": -0.1574768425064093, "output_peak": 0.6728162930246727}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7622328469817157, "全帯域_10ms_rms_envelope_nrmse": 0.10054568725699414, "全帯域_10ms_envelope_correlation": 0.9927292185799397, "全帯域_10ms_modulation_mae_db": 0.6902390674716772, "全帯域_40ms_log_rms_mae_db": 0.38894761039258824, "全帯域_40ms_rms_envelope_nrmse": 0.05842705038417637, "全帯域_40ms_envelope_correlation": 0.9977949089117969, "全帯域_40ms_modulation_mae_db": 0.5627485637909807, "300–1000 Hz_10ms_log_rms_mae_db": 1.2650201558812193, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1693973409732807, "300–1000 Hz_10ms_envelope_correlation": 0.9799758338509005, "300–1000 Hz_10ms_modulation_mae_db": 0.6208929743294324, "300–1000 Hz_40ms_log_rms_mae_db": 0.8429799862468115, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10092535384575697, "300–1000 Hz_40ms_envelope_correlation": 0.9931677682336052, "300–1000 Hz_40ms_modulation_mae_db": 0.32104659659750934, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3272829553843577, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11484765193243764, "1000–3000 Hz_10ms_envelope_correlation": 0.9925470220890215, "1000–3000 Hz_10ms_modulation_mae_db": 0.8015454056216038, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8211649293382324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07591330631981122, "1000–3000 Hz_40ms_envelope_correlation": 0.997113605642663, "1000–3000 Hz_40ms_modulation_mae_db": 0.6858385295661892, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6014301760737104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23494275128923686, "3000–6000 Hz_10ms_envelope_correlation": 0.9580372028866169, "3000–6000 Hz_10ms_modulation_mae_db": 0.583625750455479, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9758884821020865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14215202048013137, "3000–6000 Hz_40ms_envelope_correlation": 0.9835246055332209, "3000–6000 Hz_40ms_modulation_mae_db": 0.5934818756883944, "mr_stft_db": 5.330730158907233, "band_mae_db": 0.28074520536561604, "rms_gain_error_db": -0.23079706433644037, "output_peak": 0.7006791984345222}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.799835728739662, "全帯域_10ms_rms_envelope_nrmse": 0.08372649781722703, "全帯域_10ms_envelope_correlation": 0.9948640112568093, "全帯域_10ms_modulation_mae_db": 0.3178696787255586, "全帯域_40ms_log_rms_mae_db": 0.4397858314603572, "全帯域_40ms_rms_envelope_nrmse": 0.05112480766496146, "全帯域_40ms_envelope_correlation": 0.9981399958974286, "全帯域_40ms_modulation_mae_db": 0.4138916436416518, "300–1000 Hz_10ms_log_rms_mae_db": 1.2861452869465415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13915373660401362, "300–1000 Hz_10ms_envelope_correlation": 0.9865541088231364, "300–1000 Hz_10ms_modulation_mae_db": 0.43691825723051386, "300–1000 Hz_40ms_log_rms_mae_db": 0.8804344161668011, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0763027316403441, "300–1000 Hz_40ms_envelope_correlation": 0.9961990311351647, "300–1000 Hz_40ms_modulation_mae_db": 0.5190175983155344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3392157766177375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10916187287728585, "1000–3000 Hz_10ms_envelope_correlation": 0.992875970202781, "1000–3000 Hz_10ms_modulation_mae_db": 0.6714406202188853, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8321220658493316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06109525876883904, "1000–3000 Hz_40ms_envelope_correlation": 0.9978621131635488, "1000–3000 Hz_40ms_modulation_mae_db": 1.0680885001496057, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5608644693906388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23001242841536024, "3000–6000 Hz_10ms_envelope_correlation": 0.9603151563640697, "3000–6000 Hz_10ms_modulation_mae_db": 0.3406587631073535, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9767954877825432, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13791458550830515, "3000–6000 Hz_40ms_envelope_correlation": 0.9848999068157858, "3000–6000 Hz_40ms_modulation_mae_db": 0.6919540272412776, "mr_stft_db": 5.348645731691959, "band_mae_db": 0.21459210772331824, "rms_gain_error_db": -0.15418160644232093, "output_peak": 0.6772557654236903}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7753475804545157, "全帯域_10ms_rms_envelope_nrmse": 0.1018009783578192, "全帯域_10ms_envelope_correlation": 0.9921846164260966, "全帯域_10ms_modulation_mae_db": 0.3818192440838657, "全帯域_40ms_log_rms_mae_db": 0.4053599706586818, "全帯域_40ms_rms_envelope_nrmse": 0.05377582419017755, "全帯域_40ms_envelope_correlation": 0.9976989115340397, "全帯域_40ms_modulation_mae_db": 0.47846299510395784, "300–1000 Hz_10ms_log_rms_mae_db": 1.2272638333797012, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1751553472977765, "300–1000 Hz_10ms_envelope_correlation": 0.978161288191675, "300–1000 Hz_10ms_modulation_mae_db": 0.5035085872766997, "300–1000 Hz_40ms_log_rms_mae_db": 0.8028764277572686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08543141823275449, "300–1000 Hz_40ms_envelope_correlation": 0.994702348026053, "300–1000 Hz_40ms_modulation_mae_db": 0.43419231920137197, "1000–3000 Hz_10ms_log_rms_mae_db": 1.349314550554204, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12588697952819614, "1000–3000 Hz_10ms_envelope_correlation": 0.9906362756835074, "1000–3000 Hz_10ms_modulation_mae_db": 1.0651393427572928, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8791649311487074, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06836611081848723, "1000–3000 Hz_40ms_envelope_correlation": 0.9972266684001706, "1000–3000 Hz_40ms_modulation_mae_db": 1.1472979925497466, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5672134304565748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24415954883050017, "3000–6000 Hz_10ms_envelope_correlation": 0.9555251339377664, "3000–6000 Hz_10ms_modulation_mae_db": 0.8082244450519934, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9872968119610462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15319834321735679, "3000–6000 Hz_40ms_envelope_correlation": 0.9814361721825562, "3000–6000 Hz_40ms_modulation_mae_db": 0.9693992545227469, "mr_stft_db": 5.336787079704884, "band_mae_db": 0.27297402475145044, "rms_gain_error_db": -0.0074884678328896405, "output_peak": 0.683634580492961}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7575882923695157, "全帯域_10ms_rms_envelope_nrmse": 0.08969166199581177, "全帯域_10ms_envelope_correlation": 0.9939615407663408, "全帯域_10ms_modulation_mae_db": 0.43772236623963634, "全帯域_40ms_log_rms_mae_db": 0.4060596723538994, "全帯域_40ms_rms_envelope_nrmse": 0.03851386338877106, "全帯域_40ms_envelope_correlation": 0.9989001425386455, "全帯域_40ms_modulation_mae_db": 0.3933351475644679, "300–1000 Hz_10ms_log_rms_mae_db": 1.2693438655978706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14681452942109496, "300–1000 Hz_10ms_envelope_correlation": 0.9848022827735794, "300–1000 Hz_10ms_modulation_mae_db": 0.496918522103587, "300–1000 Hz_40ms_log_rms_mae_db": 0.8788129601144028, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07829261696304105, "300–1000 Hz_40ms_envelope_correlation": 0.9957054175386477, "300–1000 Hz_40ms_modulation_mae_db": 0.8320239306162261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3369489520669324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11817884455410271, "1000–3000 Hz_10ms_envelope_correlation": 0.9915311349958169, "1000–3000 Hz_10ms_modulation_mae_db": 0.6446592739386365, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8388941853817474, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0542597997991358, "1000–3000 Hz_40ms_envelope_correlation": 0.9981824570505708, "1000–3000 Hz_40ms_modulation_mae_db": 0.7502019911351946, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5813289366652497, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2268610895232365, "3000–6000 Hz_10ms_envelope_correlation": 0.9616520330418222, "3000–6000 Hz_10ms_modulation_mae_db": 0.603638033056679, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9921044831169923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14307838277778417, "3000–6000 Hz_40ms_envelope_correlation": 0.9838548941804107, "3000–6000 Hz_40ms_modulation_mae_db": 0.5919733007823486, "mr_stft_db": 5.365445561765522, "band_mae_db": 0.2598738170372417, "rms_gain_error_db": -0.09344286617241346, "output_peak": 0.7844839179974022}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.759675400731677, "全帯域_10ms_rms_envelope_nrmse": 0.08711741669347567, "全帯域_10ms_envelope_correlation": 0.994345124058069, "全帯域_10ms_modulation_mae_db": 0.5411619981695918, "全帯域_40ms_log_rms_mae_db": 0.38921077918757024, "全帯域_40ms_rms_envelope_nrmse": 0.04287949913461632, "全帯域_40ms_envelope_correlation": 0.9986672054470231, "全帯域_40ms_modulation_mae_db": 0.4704656760547697, "300–1000 Hz_10ms_log_rms_mae_db": 1.2726133581608394, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13033680363300484, "300–1000 Hz_10ms_envelope_correlation": 0.9881028304562434, "300–1000 Hz_10ms_modulation_mae_db": 0.5004108362360976, "300–1000 Hz_40ms_log_rms_mae_db": 0.825143068417632, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060047109482965676, "300–1000 Hz_40ms_envelope_correlation": 0.997637183498495, "300–1000 Hz_40ms_modulation_mae_db": 0.4751061285063187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3512753654083767, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12447769839993406, "1000–3000 Hz_10ms_envelope_correlation": 0.9909616905300768, "1000–3000 Hz_10ms_modulation_mae_db": 1.032049387150024, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8521265700856323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07904678780895424, "1000–3000 Hz_40ms_envelope_correlation": 0.9966430202004322, "1000–3000 Hz_40ms_modulation_mae_db": 0.8073716472548934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5636021626641559, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21890252227161472, "3000–6000 Hz_10ms_envelope_correlation": 0.9632306959756298, "3000–6000 Hz_10ms_modulation_mae_db": 0.35118468879813697, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9394907777157756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12436557478289385, "3000–6000 Hz_40ms_envelope_correlation": 0.9872055435544266, "3000–6000 Hz_40ms_modulation_mae_db": 0.41375980320902706, "mr_stft_db": 5.353571887209973, "band_mae_db": 0.263674160211462, "rms_gain_error_db": -0.11490669729426146, "output_peak": 0.6672079500221885}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7837401011052283, "全帯域_10ms_rms_envelope_nrmse": 0.10605072678884626, "全帯域_10ms_envelope_correlation": 0.9915804995320577, "全帯域_10ms_modulation_mae_db": 0.13570539395635486, "全帯域_40ms_log_rms_mae_db": 0.4267518959703558, "全帯域_40ms_rms_envelope_nrmse": 0.057483102823006074, "全帯域_40ms_envelope_correlation": 0.9975349139599762, "全帯域_40ms_modulation_mae_db": 0.8287925191385805, "300–1000 Hz_10ms_log_rms_mae_db": 1.313185861590317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1645505664797711, "300–1000 Hz_10ms_envelope_correlation": 0.980782794046147, "300–1000 Hz_10ms_modulation_mae_db": 0.6105360292397329, "300–1000 Hz_40ms_log_rms_mae_db": 0.9030932774803124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08635747704484452, "300–1000 Hz_40ms_envelope_correlation": 0.9946442826279197, "300–1000 Hz_40ms_modulation_mae_db": 0.16943215303976955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3590324395696307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1505173121145772, "1000–3000 Hz_10ms_envelope_correlation": 0.9863770554246674, "1000–3000 Hz_10ms_modulation_mae_db": 0.8809780948450486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8419913716715864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09472645124687382, "1000–3000 Hz_40ms_envelope_correlation": 0.994746731136751, "1000–3000 Hz_40ms_modulation_mae_db": 1.2172199500862917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5352601259734167, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23155789335479443, "3000–6000 Hz_10ms_envelope_correlation": 0.9583068201465572, "3000–6000 Hz_10ms_modulation_mae_db": 0.48391882488273263, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9286037846069084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13531036766492696, "3000–6000 Hz_40ms_envelope_correlation": 0.9844558281087221, "3000–6000 Hz_40ms_modulation_mae_db": 0.40787424409879763, "mr_stft_db": 5.340638822932022, "band_mae_db": 0.2168646031719452, "rms_gain_error_db": -0.13908746476935638, "output_peak": 0.6984341299928375}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.64387125751203, "spectral_convergence": 0.37421208170073234, "spectral_convergence_status": "ok", "amplitude_floor": 1.4952467880734047e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.252304666237198, "spectral_convergence": 0.39742898908020435, "spectral_convergence_status": "ok", "amplitude_floor": 1.087083783284318e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.153430096613231, "spectral_convergence": 0.38820629071031304, "spectral_convergence_status": "ok", "amplitude_floor": 6.922347653241366e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.706318352359072e-05, "output_power": 6.189208779129085e-05, "delta_db": 0.35279135846901655, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006570402135178566, "output_power": 0.0006449080039480977, "delta_db": -0.08094183955482913, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010679696452407233, "output_power": 0.0009439558645787293, "delta_db": -0.5360721999124337, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007107211898605409, "output_power": 0.0007921940005519638, "delta_db": 0.4713228513216294, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005187917577577873, "output_power": 0.0005189768185727241, "delta_db": 0.0015489174258810523, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.297230867177819e-05, "output_power": 9.058287535650199e-05, "delta_db": 0.38112927967957316, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3039677410605605, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05941623123737644, "pitch": null, "output_rms_dbfs": -25.153429543176387, "output_peak": 0.6971527316853043, "reference_rms_dbfs": -25.09401331193901, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.349868673454154}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.0002227499093660225, "log_rms_mae_db": 0.7573664281459859, "rms_envelope_nrmse": 0.09974160200793106, "envelope_correlation": 0.9924920932856933, "rms_gain_error_db": -0.05941622683126675, "reference_mean_power": 0.003098689875324867, "output_mean_power": 0.0030565850910689155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.065525005901086e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018487969313489734, "output_power": 0.0018022028697943179, "reference_active": true, "delta_db": -0.11085534708768588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000254937700664643, "output_power": 0.00024307323325043541, "reference_active": true, "delta_db": -0.206969266835259}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5927620235575996e-05, "output_power": 2.963969292346759e-05, "reference_active": true, "delta_db": -0.8355475156837802}], "active_band_count": 3, "active_band_mae_db": 0.384457376535575}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00020550660315895724, "log_rms_mae_db": 0.3814576550316515, "rms_envelope_nrmse": 0.05430168667169141, "envelope_correlation": 0.9976775069343691, "rms_gain_error_db": -0.0594162175358371, "reference_mean_power": 0.003123712899469445, "output_mean_power": 0.0030812681106826834}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 2.850551559013711e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017535539962877898, "output_power": 0.0017120361948234164, "reference_active": true, "delta_db": -0.10406201403563808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011841366649455362, "output_power": 0.00011392062123278157, "reference_active": true, "delta_db": -0.16799483977706037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7582705881956915e-06, "output_power": 1.5374302648856854e-06, "reference_active": true, "delta_db": -0.5829028535328571}], "active_band_count": 3, "active_band_mae_db": 0.28498656911518516}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00016827060195170467, "log_rms_mae_db": 1.215689263013969, "rms_envelope_nrmse": 0.16940290339086495, "envelope_correlation": 0.9801922297591957, "rms_gain_error_db": -0.38208307330552643, "reference_mean_power": 0.001201501254742638, "output_mean_power": 0.0011003121822872478}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.2448894715706734e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007295274990436897, "output_power": 0.000615025893448134, "reference_active": true, "delta_db": -0.7414826640970404}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014476368795830424, "output_power": 0.00012641999496752654, "reference_active": true, "delta_db": -0.5884386983555081}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.785748904828138e-05, "output_power": 3.464503310530338e-05, "reference_active": true, "delta_db": 0.9469901220798549}], "active_band_count": 3, "active_band_mae_db": 0.7589704948441346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012583494842815252, "log_rms_mae_db": 0.7872086350174321, "rms_envelope_nrmse": 0.0982114770878151, "envelope_correlation": 0.9939852238135563, "rms_gain_error_db": -0.38208307365422994, "reference_mean_power": 0.0012112038213410922, "output_mean_power": 0.001109197609646218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1451125517965788e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007097609595329509, "output_power": 0.0005954669845088998, "reference_active": true, "delta_db": -0.7625442010348862}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.709378020571589e-05, "output_power": 5.7538073753240694e-05, "reference_active": true, "delta_db": -0.6672694305220637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7774397828761626e-06, "output_power": 1.7564455793798903e-06, "reference_active": true, "delta_db": -0.05160197947914756}], "active_band_count": 3, "active_band_mae_db": 0.4938052036786991}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015339527329010622, "log_rms_mae_db": 1.3968415763064925, "rms_envelope_nrmse": 0.12493745090219621, "envelope_correlation": 0.9915474146874311, "rms_gain_error_db": 0.2887966126444942, "reference_mean_power": 0.0009604249251386078, "output_mean_power": 0.0010264624902627387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.606398277844274e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000996843421347699, "output_power": 0.0010656136481564712, "reference_active": true, "delta_db": 0.2897282712166058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073134134931949, "output_power": 0.00010612969475383675, "reference_active": true, "delta_db": -0.18433645663506512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0381145318442363e-05, "output_power": 1.3907010256253218e-05, "reference_active": true, "delta_db": 1.2698850446828587}], "active_band_count": 3, "active_band_mae_db": 0.5813165908448432}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.0001445669043392124, "log_rms_mae_db": 0.8481124639123745, "rms_envelope_nrmse": 0.07036321344733527, "envelope_correlation": 0.997709630023625, "rms_gain_error_db": 0.28879664247799425, "reference_mean_power": 0.0009681806844812572, "output_mean_power": 0.0010347515326097187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.5548349296058167e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009718305477517134, "output_power": 0.0010406608389379557, "reference_active": true, "delta_db": 0.2971866574244596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.313569881232493e-05, "output_power": 5.128248646212966e-05, "reference_active": true, "delta_db": -0.15417322511545964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.152693877304456e-07, "output_power": 8.274881920864268e-07, "reference_active": true, "delta_db": 1.2869649766221505}], "active_band_count": 3, "active_band_mae_db": 0.5794416197206899}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.2745097516618104e-05, "log_rms_mae_db": 1.5236038394976068, "rms_envelope_nrmse": 0.22745835483429266, "envelope_correlation": 0.9608935155449362, "rms_gain_error_db": 0.24024376227068486, "reference_mean_power": 0.0001383637841793036, "output_mean_power": 0.000146233477373315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 9.868150581916701e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.732185415785845e-05, "output_power": 5.775401204876626e-05, "reference_active": true, "delta_db": 0.032619292965231084}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6922402540887435e-05, "output_power": 1.6373130746463224e-05, "reference_active": true, "delta_db": -0.14330291732934053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.249240297758961e-06, "output_power": 5.287249129749729e-06, "reference_active": true, "delta_db": 0.9491848293122418}], "active_band_count": 3, "active_band_mae_db": 0.3750356798689378}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.734940150914352e-05, "log_rms_mae_db": 0.9620776244561442, "rms_envelope_nrmse": 0.13937220828847277, "envelope_correlation": 0.9842747268112552, "rms_gain_error_db": 0.24024380064656942, "reference_mean_power": 0.00013948111618043138, "output_mean_power": 0.00014741436097741978}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 8.397728752995274e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.502560023531223e-05, "output_power": 5.653998913477644e-05, "reference_active": true, "delta_db": 0.11790932033420069}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.667910609926144e-06, "output_power": 7.624702783515907e-06, "reference_active": true, "delta_db": -0.024541221353568836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2898812052596812e-07, "output_power": 3.051261920259604e-07, "reference_active": true, "delta_db": 1.246665365303678}], "active_band_count": 3, "active_band_mae_db": 0.46303863566381587}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8847044611049345, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "569b3c62b95c7f608dc76e9c684367364291ed5f996f5d5436963f37afc8fa91", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9913492775269115, "standard_deviation": 0.0023308666479050118, "minimum": 0.9863770554246674, "maximum": 0.9943176566019895, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3487512604725695, "standard_deviation": 0.02224571950238807, "minimum": 1.3272829553843577, "maximum": 1.3968415763064925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7801171253769525, "standard_deviation": 0.19705336963157896, "minimum": 0.563808287639286, "maximum": 1.0651393427572928, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12066030104249892, "standard_deviation": 0.015448880140823778, "minimum": 0.09727459803126161, "maximum": 0.1505173121145772, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972502711581417, "standard_deviation": 0.0011781963335372532, "minimum": 0.994746731136751, "maximum": 0.9985179436473727, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8467373145861006, "standard_deviation": 0.017802969728253206, "minimum": 0.8211649293382324, "maximum": 0.8791649311487074, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8401110022974754, "standard_deviation": 0.2751208240184841, "minimum": 0.465427787917192, "maximum": 1.2172199500862917, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06923217215186707, "standard_deviation": 0.014391921471807964, "minimum": 0.05008644900549994, "maximum": 0.09472645124687382, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9598151069396903, "standard_deviation": 0.002415681208081031, "minimum": 0.9555251339377664, "maximum": 0.9632306959756298, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5664657184383637, "standard_deviation": 0.02755704200868311, "minimum": 1.5236038394976068, "maximum": 1.6014301760737104, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.49776766678626916, "standard_deviation": 0.1604553449640068, "minimum": 0.3406587631073535, "maximum": 0.8082244450519934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22987151590528176, "standard_deviation": 0.007472961646048157, "minimum": 0.21890252227161472, "maximum": 0.24415954883050017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9843439596984795, "standard_deviation": 0.001623593862267737, "minimum": 0.9814361721825562, "maximum": 0.9872055435544266, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9630947137737945, "standard_deviation": 0.023756611240797615, "minimum": 0.9286037846069084, "maximum": 0.9921044831169923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5534287933023003, "standard_deviation": 0.2106012705089473, "minimum": 0.29594920521199447, "maximum": 0.9693992545227469, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13846627738758366, "standard_deviation": 0.008446059060105234, "minimum": 0.12436557478289385, "maximum": 0.15319834321735679, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9824493891617205, "standard_deviation": 0.003560844284444426, "minimum": 0.978161288191675, "maximum": 0.9881028304562434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2634746131134285, "standard_deviation": 0.030994419835839927, "minimum": 1.215689263013969, "maximum": 1.313185861590317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5170596279090722, "standard_deviation": 0.16064536225910356, "minimum": 0.20832132201238018, "maximum": 0.7589704948441346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1573272113199888, "standard_deviation": 0.016356668065078427, "minimum": 0.13033680363300484, "maximum": 0.1751553472977765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9952128920028918, "standard_deviation": 0.0013921978764103715, "minimum": 0.9931677682336052, "maximum": 0.997637183498495, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8474035572751559, "standard_deviation": 0.040403810883709805, "minimum": 0.7872086350174321, "maximum": 0.9030932774803124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4704288551993814, "standard_deviation": 0.18907586982824956, "minimum": 0.16943215303976955, "maximum": 0.8320239306162261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08311527436683278, "standard_deviation": 0.01296658765024115, "minimum": 0.060047109482965676, "maximum": 0.10092535384575697, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2582549328172016, "standard_deviation": 0.030443848264535382, "minimum": 0.21459210772331824, "maximum": 0.3039677410605605, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.346227484692683, "standard_deviation": 0.010739071640145691, "minimum": 5.330730158907233, "maximum": 5.365445561765522, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6977080708841974, "standard_deviation": 0.03723655790567125, "minimum": 0.6672079500221885, "maximum": 0.7844839179974022, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1195996550739335, "standard_deviation": 0.06789950084952909, "minimum": -0.23079706433644037, "maximum": -0.0074884678328896405, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931968929629278, "standard_deviation": 0.0011391854021728521, "minimum": 0.9915804995320577, "maximum": 0.9948640112568093, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7731891568791829, "standard_deviation": 0.016457413564069722, "minimum": 0.7573664281459859, "maximum": 0.799835728739662, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4330044023647561, "standard_deviation": 0.1710009424494376, "minimum": 0.13570539395635486, "maximum": 0.6902390674716772, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0953810630146806, "standard_deviation": 0.00791242257307975, "minimum": 0.08372649781722703, "maximum": 0.10605072678884626, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980941609147349, "standard_deviation": 0.0005032053344991974, "minimum": 0.9975349139599762, "maximum": 0.9989001425386455, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4076938571529265, "standard_deviation": 0.02086610589317143, "minimum": 0.3814576550316515, "maximum": 0.4397858314603572, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5316802372304494, "standard_deviation": 0.19757744764053842, "minimum": 0.28498656911518516, "maximum": 0.8287925191385805, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.050662552528065716, "standard_deviation": 0.006985828899111671, "minimum": 0.03851386338877106, "maximum": 0.05842705038417637, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0004", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.1553800106048584, "predicted_mos": 2.1553800106048584, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "2226620a4d1b3b74b36675db995a30df8f5620d7d1d7519e7b8c9e6cf0f677d5", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.299541000276804, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8163166442244378, "全帯域_10ms_rms_envelope_nrmse": 0.10002034056410451, "全帯域_10ms_envelope_correlation": 0.9929863254440223, "全帯域_10ms_modulation_mae_db": 0.6526047455903633, "全帯域_40ms_log_rms_mae_db": 0.4316062344393989, "全帯域_40ms_rms_envelope_nrmse": 0.05800809752222816, "全帯域_40ms_envelope_correlation": 0.9977508412499857, "全帯域_40ms_modulation_mae_db": 1.0548424282275077, "300–1000 Hz_10ms_log_rms_mae_db": 1.0105829319578736, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1554385018236558, "300–1000 Hz_10ms_envelope_correlation": 0.9847313895105051, "300–1000 Hz_10ms_modulation_mae_db": 1.166569287742745, "300–1000 Hz_40ms_log_rms_mae_db": 0.6609176873139524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10702636832274788, "300–1000 Hz_40ms_envelope_correlation": 0.9928971563575989, "300–1000 Hz_40ms_modulation_mae_db": 1.2815453609465939, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1503499392156962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11263233458805766, "1000–3000 Hz_10ms_envelope_correlation": 0.9924508388408657, "1000–3000 Hz_10ms_modulation_mae_db": 0.7881850523157085, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6670670562870924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05560465824901353, "1000–3000 Hz_40ms_envelope_correlation": 0.9981446553346187, "1000–3000 Hz_40ms_modulation_mae_db": 1.018061052675323, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2221956470681172, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20969559145551023, "3000–6000 Hz_10ms_envelope_correlation": 0.9650879410856262, "3000–6000 Hz_10ms_modulation_mae_db": 0.5473924989930034, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7394515173878377, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12300423054110926, "3000–6000 Hz_40ms_envelope_correlation": 0.9867588673908604, "3000–6000 Hz_40ms_modulation_mae_db": 0.6959301871129394, "mr_stft_db": 5.037562828890092, "band_mae_db": 0.1288354001966919, "rms_gain_error_db": 0.1726546772499718, "output_peak": 0.7651817834014648}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8038131518502191, "全帯域_10ms_rms_envelope_nrmse": 0.08113304014719272, "全帯域_10ms_envelope_correlation": 0.9950378934554652, "全帯域_10ms_modulation_mae_db": 0.27411997967194485, "全帯域_40ms_log_rms_mae_db": 0.42039469635727633, "全帯域_40ms_rms_envelope_nrmse": 0.040084152079567684, "全帯域_40ms_envelope_correlation": 0.9987233998540461, "全帯域_40ms_modulation_mae_db": 0.3150008036285849, "300–1000 Hz_10ms_log_rms_mae_db": 1.033201116343835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1294790551117668, "300–1000 Hz_10ms_envelope_correlation": 0.9881876671855799, "300–1000 Hz_10ms_modulation_mae_db": 0.31972922627083056, "300–1000 Hz_40ms_log_rms_mae_db": 0.6505255847735327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07171386206911627, "300–1000 Hz_40ms_envelope_correlation": 0.9963404040764647, "300–1000 Hz_40ms_modulation_mae_db": 0.4419703687250067, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1064930735652814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1101464903685553, "1000–3000 Hz_10ms_envelope_correlation": 0.9927716985440754, "1000–3000 Hz_10ms_modulation_mae_db": 0.44890646393200656, "1000–3000 Hz_40ms_log_rms_mae_db": 0.628571219362635, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053325756413037335, "1000–3000 Hz_40ms_envelope_correlation": 0.9982929411914035, "1000–3000 Hz_40ms_modulation_mae_db": 0.7280034323014642, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1928713449337294, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19269947506116558, "3000–6000 Hz_10ms_envelope_correlation": 0.9713384515394821, "3000–6000 Hz_10ms_modulation_mae_db": 0.5658087584899324, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6874601804552558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11111897683137567, "3000–6000 Hz_40ms_envelope_correlation": 0.9896543893048423, "3000–6000 Hz_40ms_modulation_mae_db": 0.9000920785234241, "mr_stft_db": 5.040016789578326, "band_mae_db": 0.22925109405601216, "rms_gain_error_db": -0.012012959224986273, "output_peak": 0.7328316071939913}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7446543291016718, "全帯域_10ms_rms_envelope_nrmse": 0.11370567178675026, "全帯域_10ms_envelope_correlation": 0.9902991113981575, "全帯域_10ms_modulation_mae_db": 0.6747738800508231, "全帯域_40ms_log_rms_mae_db": 0.3860366484702094, "全帯域_40ms_rms_envelope_nrmse": 0.0657399593713771, "全帯域_40ms_envelope_correlation": 0.9966726977143984, "全帯域_40ms_modulation_mae_db": 0.9321030266003009, "300–1000 Hz_10ms_log_rms_mae_db": 1.0074447103366073, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1675474239751269, "300–1000 Hz_10ms_envelope_correlation": 0.9800769851424103, "300–1000 Hz_10ms_modulation_mae_db": 0.5731322205393646, "300–1000 Hz_40ms_log_rms_mae_db": 0.6200725984957385, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.104185776359102, "300–1000 Hz_40ms_envelope_correlation": 0.9919182690655733, "300–1000 Hz_40ms_modulation_mae_db": 0.46621985305824615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1266275498250766, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13624335015010938, "1000–3000 Hz_10ms_envelope_correlation": 0.9888129147627325, "1000–3000 Hz_10ms_modulation_mae_db": 1.0296840781245444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6360327203955076, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08663426684965518, "1000–3000 Hz_40ms_envelope_correlation": 0.9954996770057262, "1000–3000 Hz_40ms_modulation_mae_db": 1.1636949312268492, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1902495285649257, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20142916986775541, "3000–6000 Hz_10ms_envelope_correlation": 0.9682166579265737, "3000–6000 Hz_10ms_modulation_mae_db": 0.48811590197876287, "3000–6000 Hz_40ms_log_rms_mae_db": 0.693819430332098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11537352752190556, "3000–6000 Hz_40ms_envelope_correlation": 0.9885325322928813, "3000–6000 Hz_40ms_modulation_mae_db": 0.6975034826776687, "mr_stft_db": 5.04926410488255, "band_mae_db": 0.11637675794937619, "rms_gain_error_db": -0.11457417292351642, "output_peak": 0.6683810098844499}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7778202028500223, "全帯域_10ms_rms_envelope_nrmse": 0.10230720493474456, "全帯域_10ms_envelope_correlation": 0.992131043807748, "全帯域_10ms_modulation_mae_db": 0.5488635266312812, "全帯域_40ms_log_rms_mae_db": 0.3981430293008032, "全帯域_40ms_rms_envelope_nrmse": 0.05816219093020227, "全帯域_40ms_envelope_correlation": 0.9973175350197515, "全帯域_40ms_modulation_mae_db": 0.8313233504303542, "300–1000 Hz_10ms_log_rms_mae_db": 0.9998483967122135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1864613444922985, "300–1000 Hz_10ms_envelope_correlation": 0.975288882251557, "300–1000 Hz_10ms_modulation_mae_db": 0.8296070956248087, "300–1000 Hz_40ms_log_rms_mae_db": 0.6346130320493543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10332577603353864, "300–1000 Hz_40ms_envelope_correlation": 0.9920062971338043, "300–1000 Hz_40ms_modulation_mae_db": 1.079603174769098, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1041643660856497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10425611388315453, "1000–3000 Hz_10ms_envelope_correlation": 0.9935788830599132, "1000–3000 Hz_10ms_modulation_mae_db": 0.9487748347796784, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6321968909034696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06635236878699137, "1000–3000 Hz_40ms_envelope_correlation": 0.9973856810909018, "1000–3000 Hz_40ms_modulation_mae_db": 0.6961315567166476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2714825395373865, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18859995621911288, "3000–6000 Hz_10ms_envelope_correlation": 0.9724371881112158, "3000–6000 Hz_10ms_modulation_mae_db": 0.4551014321834906, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7630635831292345, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10670204784917993, "3000–6000 Hz_40ms_envelope_correlation": 0.9903975669629272, "3000–6000 Hz_40ms_modulation_mae_db": 0.820465277210657, "mr_stft_db": 5.033378456396786, "band_mae_db": 0.17222109683547573, "rms_gain_error_db": 0.012022549483384118, "output_peak": 0.68647879434041}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7437518001189627, "全帯域_10ms_rms_envelope_nrmse": 0.08219693480874239, "全帯域_10ms_envelope_correlation": 0.9951632020256788, "全帯域_10ms_modulation_mae_db": 0.6246180388577135, "全帯域_40ms_log_rms_mae_db": 0.3926597238227866, "全帯域_40ms_rms_envelope_nrmse": 0.044804689362754155, "全帯域_40ms_envelope_correlation": 0.9987559426878877, "全帯域_40ms_modulation_mae_db": 0.5442463770774125, "300–1000 Hz_10ms_log_rms_mae_db": 0.9878520110892757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1156857631048075, "300–1000 Hz_10ms_envelope_correlation": 0.9907486608838256, "300–1000 Hz_10ms_modulation_mae_db": 0.642500037615886, "300–1000 Hz_40ms_log_rms_mae_db": 0.6247955667967476, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0632326975299186, "300–1000 Hz_40ms_envelope_correlation": 0.9973825479032072, "300–1000 Hz_40ms_modulation_mae_db": 0.8660659869585547, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1466054908664338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1266474118333309, "1000–3000 Hz_10ms_envelope_correlation": 0.9907412287719038, "1000–3000 Hz_10ms_modulation_mae_db": 1.0577502101494163, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7147757915022008, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07775709558021315, "1000–3000 Hz_40ms_envelope_correlation": 0.9969159136626483, "1000–3000 Hz_40ms_modulation_mae_db": 1.2531220537534151, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2176229664725602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21906698602518887, "3000–6000 Hz_10ms_envelope_correlation": 0.9636972327325796, "3000–6000 Hz_10ms_modulation_mae_db": 0.7633635175835695, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6989943377759091, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1332791895899357, "3000–6000 Hz_40ms_envelope_correlation": 0.9855442142514326, "3000–6000 Hz_40ms_modulation_mae_db": 1.2016652812615065, "mr_stft_db": 5.030206382931305, "band_mae_db": 0.237549187469328, "rms_gain_error_db": -0.18850495556317237, "output_peak": 0.65185217558263}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7785546361082364, "全帯域_10ms_rms_envelope_nrmse": 0.08330777791544393, "全帯域_10ms_envelope_correlation": 0.9948281170821257, "全帯域_10ms_modulation_mae_db": 0.2748849956796732, "全帯域_40ms_log_rms_mae_db": 0.421081259863005, "全帯域_40ms_rms_envelope_nrmse": 0.0391633606008449, "全帯域_40ms_envelope_correlation": 0.9988969644073511, "全帯域_40ms_modulation_mae_db": 0.2441408344814127, "300–1000 Hz_10ms_log_rms_mae_db": 0.9293373041665886, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12087687357302104, "300–1000 Hz_10ms_envelope_correlation": 0.9899608968982384, "300–1000 Hz_10ms_modulation_mae_db": 0.42850416845560185, "300–1000 Hz_40ms_log_rms_mae_db": 0.6307846249916984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07611678879551605, "300–1000 Hz_40ms_envelope_correlation": 0.9961413771856111, "300–1000 Hz_40ms_modulation_mae_db": 0.6973393803338045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1016561925271586, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11485501441083766, "1000–3000 Hz_10ms_envelope_correlation": 0.9921200830443917, "1000–3000 Hz_10ms_modulation_mae_db": 0.7408679581156159, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6142560764264682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05734813911618482, "1000–3000 Hz_40ms_envelope_correlation": 0.9981845730966116, "1000–3000 Hz_40ms_modulation_mae_db": 0.8718457425093917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.173647105106588, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20876231456568503, "3000–6000 Hz_10ms_envelope_correlation": 0.9675800624907438, "3000–6000 Hz_10ms_modulation_mae_db": 0.6615574381294997, "3000–6000 Hz_40ms_log_rms_mae_db": 0.700699973612327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11253874964687995, "3000–6000 Hz_40ms_envelope_correlation": 0.990259831981868, "3000–6000 Hz_40ms_modulation_mae_db": 1.0100191164442969, "mr_stft_db": 5.050762044110706, "band_mae_db": 0.25892979711468644, "rms_gain_error_db": -0.10958861728320522, "output_peak": 0.6864748511789522}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8087512771696638, "全帯域_10ms_rms_envelope_nrmse": 0.11192740911121968, "全帯域_10ms_envelope_correlation": 0.9908207682795301, "全帯域_10ms_modulation_mae_db": 0.7955535180709025, "全帯域_40ms_log_rms_mae_db": 0.43849003586568813, "全帯域_40ms_rms_envelope_nrmse": 0.05997069484014985, "全帯域_40ms_envelope_correlation": 0.9976149176235407, "全帯域_40ms_modulation_mae_db": 0.6670713976236144, "300–1000 Hz_10ms_log_rms_mae_db": 1.0541265255452426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1881578263281867, "300–1000 Hz_10ms_envelope_correlation": 0.9759305181207006, "300–1000 Hz_10ms_modulation_mae_db": 1.2038218091154256, "300–1000 Hz_40ms_log_rms_mae_db": 0.7134569809588374, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11536893294302765, "300–1000 Hz_40ms_envelope_correlation": 0.9921281538243041, "300–1000 Hz_40ms_modulation_mae_db": 1.3908033617807252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1091279242033585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12368541710369722, "1000–3000 Hz_10ms_envelope_correlation": 0.9907044456344273, "1000–3000 Hz_10ms_modulation_mae_db": 0.31164522165465697, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6072137657866297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060745933819893856, "1000–3000 Hz_40ms_envelope_correlation": 0.9977161875367878, "1000–3000 Hz_40ms_modulation_mae_db": 0.23202475481290133, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2326306030892924, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20537942482970656, "3000–6000 Hz_10ms_envelope_correlation": 0.9669874920477434, "3000–6000 Hz_10ms_modulation_mae_db": 0.4692345180266729, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7137110754874177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11185528708683984, "3000–6000 Hz_40ms_envelope_correlation": 0.9892379212930238, "3000–6000 Hz_40ms_modulation_mae_db": 0.8115156382730945, "mr_stft_db": 5.039769472065905, "band_mae_db": 0.23558322514033347, "rms_gain_error_db": -0.2241124692353721, "output_peak": 0.6466739284858664}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7579178253462029, "全帯域_10ms_rms_envelope_nrmse": 0.09156775292674818, "全帯域_10ms_envelope_correlation": 0.9936797246704735, "全帯域_10ms_modulation_mae_db": 0.4709114927162165, "全帯域_40ms_log_rms_mae_db": 0.4084748916339874, "全帯域_40ms_rms_envelope_nrmse": 0.04232880714814075, "全帯域_40ms_envelope_correlation": 0.9986033879781072, "全帯域_40ms_modulation_mae_db": 0.17920679913691107, "300–1000 Hz_10ms_log_rms_mae_db": 0.994488545130899, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1543123617892008, "300–1000 Hz_10ms_envelope_correlation": 0.983128320645263, "300–1000 Hz_10ms_modulation_mae_db": 0.34260433280573044, "300–1000 Hz_40ms_log_rms_mae_db": 0.6561572813434337, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07554685912232229, "300–1000 Hz_40ms_envelope_correlation": 0.9957486778974576, "300–1000 Hz_40ms_modulation_mae_db": 0.6293417082217203, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1031132842988354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10429612081951481, "1000–3000 Hz_10ms_envelope_correlation": 0.9934483523872063, "1000–3000 Hz_10ms_modulation_mae_db": 0.7627236014547836, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6432547516776268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05058445425275551, "1000–3000 Hz_40ms_envelope_correlation": 0.9985105201033322, "1000–3000 Hz_40ms_modulation_mae_db": 0.8545068594842515, "3000–6000 Hz_10ms_log_rms_mae_db": 1.242475062846635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20460220233817783, "3000–6000 Hz_10ms_envelope_correlation": 0.9691630537396371, "3000–6000 Hz_10ms_modulation_mae_db": 0.5547559502064211, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7212410640245185, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11692105314991708, "3000–6000 Hz_40ms_envelope_correlation": 0.9896476264078263, "3000–6000 Hz_40ms_modulation_mae_db": 0.7275235446245626, "mr_stft_db": 5.038976561001247, "band_mae_db": 0.15675287092055532, "rms_gain_error_db": -0.050818064601067234, "output_peak": 0.7108001025964653}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.286246455133729, "spectral_convergence": 0.3687550458344201, "spectral_convergence_status": "ok", "amplitude_floor": 1.4952467880734047e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.874063913474237, "spectral_convergence": 0.38314062208070704, "spectral_convergence_status": "ok", "amplitude_floor": 1.087083783284318e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.9523781180623105, "spectral_convergence": 0.38885893865106674, "spectral_convergence_status": "ok", "amplitude_floor": 6.922347653241366e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.706318352359072e-05, "output_power": 5.84271577644518e-05, "delta_db": 0.1025876317673132, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006570402135178566, "output_power": 0.0006641946454130457, "delta_db": 0.0470341914100034, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010679696452407233, "output_power": 0.0011432013954382097, "delta_db": 0.2956383692715566, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007107211898605409, "output_power": 0.0007493637020559185, "delta_db": 0.22993388810341395, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005187917577577873, "output_power": 0.0005230795219637251, "delta_db": 0.0357465050379061, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.297230867177819e-05, "output_power": 8.179485611418906e-05, "delta_db": -0.06207181558995825, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1288354001966919, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.1726546772499718, "pitch": null, "output_rms_dbfs": -24.921358634689035, "output_peak": 0.7651817834014648, "reference_rms_dbfs": -25.09401331193901, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.037562828890092}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.0002227499093660225, "log_rms_mae_db": 0.8163166442244378, "rms_envelope_nrmse": 0.10002034056410451, "envelope_correlation": 0.9929863254440223, "rms_gain_error_db": 0.17265468143735263, "reference_mean_power": 0.003098689875324867, "output_mean_power": 0.00322436041893633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.065525005901086e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018487969313489734, "output_power": 0.0019971204740408164, "reference_active": true, "delta_db": 0.3351505236040453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000254937700664643, "output_power": 0.00026729589115911245, "reference_active": true, "delta_db": 0.2055821838442914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5927620235575996e-05, "output_power": 4.978936694513495e-05, "reference_active": true, "delta_db": 1.4170815293227532}], "active_band_count": 3, "active_band_mae_db": 0.6526047455903633}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00020550660315895724, "log_rms_mae_db": 0.4316062344393989, "rms_envelope_nrmse": 0.05800809752222816, "envelope_correlation": 0.9977508412499857, "rms_gain_error_db": 0.17265468954820926, "reference_mean_power": 0.003123712899469445, "output_mean_power": 0.0032503982835405663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 2.850551559013711e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017535539962877898, "output_power": 0.0019079907881886754, "reference_active": true, "delta_db": 0.366571301617426}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011841366649455362, "output_power": 0.0001242755532281689, "reference_active": true, "delta_db": 0.20983876556692094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7582705881956915e-06, "output_power": 3.1907842022519395e-06, "reference_active": true, "delta_db": 2.5881172174981764}], "active_band_count": 3, "active_band_mae_db": 1.0548424282275077}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00016827060195170467, "log_rms_mae_db": 1.0105829319578736, "rms_envelope_nrmse": 0.1554385018236558, "envelope_correlation": 0.9847313895105051, "rms_gain_error_db": 0.31762542917755776, "reference_mean_power": 0.001201501254742638, "output_mean_power": 0.0012926673261762836}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.2448894715706734e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007295274990436897, "output_power": 0.0008613064273558456, "reference_active": true, "delta_db": 0.721160211049401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014476368795830424, "output_power": 0.0001802974234421476, "reference_active": true, "delta_db": 0.9532988190016121}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.785748904828138e-05, "output_power": 4.2409861266377986e-05, "reference_active": true, "delta_db": 1.8252488331772216}], "active_band_count": 3, "active_band_mae_db": 1.166569287742745}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012583494842815252, "log_rms_mae_db": 0.6609176873139524, "rms_envelope_nrmse": 0.10702636832274788, "envelope_correlation": 0.9928971563575989, "rms_gain_error_db": 0.3176254295094383, "reference_mean_power": 0.0012112038213410922, "output_mean_power": 0.0013031060925878962}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1451125517965788e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007097609595329509, "output_power": 0.0008550669204312864, "reference_active": true, "delta_db": 0.8088799815388266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.709378020571589e-05, "output_power": 8.507541126728907e-05, "reference_active": true, "delta_db": 1.0312179558115453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7774397828761626e-06, "output_power": 2.819997419588555e-06, "reference_active": true, "delta_db": 2.0045381454894096}], "active_band_count": 3, "active_band_mae_db": 1.2815453609465939}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015339527329010622, "log_rms_mae_db": 1.1503499392156962, "rms_envelope_nrmse": 0.11263233458805766, "envelope_correlation": 0.9924508388408657, "rms_gain_error_db": 0.08333378919475848, "reference_mean_power": 0.0009604249251386078, "output_mean_power": 0.0009790318066861437}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.606398277844274e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000996843421347699, "output_power": 0.0010055012849825098, "reference_active": true, "delta_db": 0.037556828477026086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073134134931949, "output_power": 9.254716985004167e-05, "reference_active": true, "delta_db": -0.779074184969646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0381145318442363e-05, "output_power": 1.4826467325418956e-05, "reference_active": true, "delta_db": 1.5479241435004532}], "active_band_count": 3, "active_band_mae_db": 0.7881850523157085}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.0001445669043392124, "log_rms_mae_db": 0.6670670562870924, "rms_envelope_nrmse": 0.05560465824901353, "envelope_correlation": 0.9981446553346187, "rms_gain_error_db": 0.0833338114200807, "reference_mean_power": 0.0009681806844812572, "output_mean_power": 0.0009869378280039486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.5548349296058167e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009718305477517134, "output_power": 0.0009791381799278088, "reference_active": true, "delta_db": 0.032534393062859455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.313569881232493e-05, "output_power": 4.451149245034181e-05, "reference_active": true, "delta_db": -0.7691424033587966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.152693877304456e-07, "output_power": 1.0335130000839813e-06, "reference_active": true, "delta_db": 2.2525063616043135}], "active_band_count": 3, "active_band_mae_db": 1.018061052675323}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.2745097516618104e-05, "log_rms_mae_db": 1.2221956470681172, "rms_envelope_nrmse": 0.20969559145551023, "envelope_correlation": 0.9650879410856262, "rms_gain_error_db": -0.02248761511358745, "reference_mean_power": 0.0001383637841793036, "output_mean_power": 0.00013764919304483467}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 9.868150581916701e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.732185415785845e-05, "output_power": 5.478143350841241e-05, "reference_active": true, "delta_db": -0.19686837244075311}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6922402540887435e-05, "output_power": 1.4248632605852816e-05, "reference_active": true, "delta_db": -0.7468883305853946}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.249240297758961e-06, "output_power": 4.990606993595939e-06, "reference_active": true, "delta_db": 0.6984207939528622}], "active_band_count": 3, "active_band_mae_db": 0.5473924989930034}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.734940150914352e-05, "log_rms_mae_db": 0.7394515173878377, "rms_envelope_nrmse": 0.12300423054110926, "envelope_correlation": 0.9867588673908604, "rms_gain_error_db": -0.022487583188491547, "reference_mean_power": 0.00013948111618043138, "output_mean_power": 0.00013876075551305132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 8.397728752995274e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.502560023531223e-05, "output_power": 5.2834319997109495e-05, "reference_active": true, "delta_db": -0.17648666443909256}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.667910609926144e-06, "output_power": 6.612834157892592e-06, "reference_active": true, "delta_db": -0.6428941012253111}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2898812052596812e-07, "output_power": 3.0665774099765094e-07, "reference_active": true, "delta_db": 1.2684097956744143}], "active_band_count": 3, "active_band_mae_db": 0.6959301871129394}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9038730302041081, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9e68104d841e9327a51e697ac300331607d32ecf00754352e7fd3a8ccdd3a435", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9918285556306895, "standard_deviation": 0.0016295978316144739, "minimum": 0.9888129147627325, "maximum": 0.9935788830599132, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1185172275734363, "standard_deviation": 0.020101063560543857, "minimum": 1.1016561925271586, "maximum": 1.1503499392156962, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7610671775658013, "standard_deviation": 0.26618483443822194, "minimum": 0.31164522165465697, "maximum": 1.0577502101494163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11659528164465718, "standard_deviation": 0.011346533419904025, "minimum": 0.10425611388315453, "maximum": 0.13624335015010938, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975812686277536, "standard_deviation": 0.0009916469355431754, "minimum": 0.9954996770057262, "maximum": 0.9985105201033322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6429210340427038, "standard_deviation": 0.03424973414083799, "minimum": 0.6072137657866297, "maximum": 0.7147757915022008, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8521737979350303, "standard_deviation": 0.31849382739642035, "minimum": 0.23202475481290133, "maximum": 1.2531220537534151, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06354408413346808, "standard_deviation": 0.012675470045969597, "minimum": 0.05058445425275551, "maximum": 0.08663426684965518, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9680635099592002, "standard_deviation": 0.0029373621961186033, "minimum": 0.9636972327325796, "maximum": 0.9724371881112158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2178968497024043, "standard_deviation": 0.03175766055601086, "minimum": 1.173647105106588, "maximum": 1.2714825395373865, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5631662519489191, "standard_deviation": 0.10434833760285991, "minimum": 0.4551014321834906, "maximum": 0.7633635175835695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20377939004528778, "standard_deviation": 0.009676790792916699, "minimum": 0.18859995621911288, "maximum": 0.21906698602518887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9887541187357076, "standard_deviation": 0.001738004508723107, "minimum": 0.9855442142514326, "maximum": 0.9903975669629272, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7148051452755748, "standard_deviation": 0.02567940629979747, "minimum": 0.6874601804552558, "maximum": 0.7630635831292345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8580893257660187, "standard_deviation": 0.17550206737893054, "minimum": 0.6959301871129394, "maximum": 1.2016652812615065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11634913277714287, "standard_deviation": 0.008343778657275178, "minimum": 0.10670204784917993, "maximum": 0.1332791895899357, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9835066650797599, "standard_deviation": 0.0060279248807916, "minimum": 0.975288882251557, "maximum": 0.9907486608838256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.002110192660317, "standard_deviation": 0.03651806419087037, "minimum": 0.9293373041665886, "maximum": 1.0541265255452426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6883085222712991, "standard_deviation": 0.34901432217784306, "minimum": 0.31972922627083056, "maximum": 1.2038218091154256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15224489377475803, "standard_deviation": 0.02815252170368546, "minimum": 0.1156857631048075, "maximum": 0.1881578263281867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9943203604305026, "standard_deviation": 0.0022919103934852508, "minimum": 0.9919182690655733, "maximum": 0.9973825479032072, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6489154195904119, "standard_deviation": 0.030027759532023735, "minimum": 0.6200725984957385, "maximum": 0.7134569809588374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8566111493492187, "standard_deviation": 0.3619582607775054, "minimum": 0.4419703687250067, "maximum": 1.3908033617807252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08956463264691117, "standard_deviation": 0.019868439917940028, "minimum": 0.0632326975299186, "maximum": 0.11536893294302765, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1919374287103074, "standard_deviation": 0.05501122346121167, "minimum": 0.11637675794937619, "maximum": 0.25892979711468644, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.039992079982115, "standard_deviation": 0.007055344086994008, "minimum": 5.030206382931305, "maximum": 5.050762044110706, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6935842815830288, "standard_deviation": 0.04078046815447551, "minimum": 0.6466739284858664, "maximum": 0.7651817834014648, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06436675151224547, "standard_deviation": 0.12560354853233385, "minimum": -0.2241124692353721, "maximum": 0.1726546772499718, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931182732704001, "standard_deviation": 0.0019020178717054556, "minimum": 0.9902991113981575, "maximum": 0.9951632020256788, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7789474833461771, "standard_deviation": 0.02868909681619791, "minimum": 0.7437518001189627, "maximum": 0.8163166442244378, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5395412721586147, "standard_deviation": 0.1886568660635962, "minimum": 0.27411997967194485, "maximum": 0.7955535180709025, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09577076652436828, "standard_deviation": 0.013174570648286633, "minimum": 0.08113304014719272, "maximum": 0.11370567178675026, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980419608168836, "standard_deviation": 0.0008182866718378863, "minimum": 0.9966726977143984, "maximum": 0.9988969644073511, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41211081496914437, "standard_deviation": 0.018866267434044497, "minimum": 0.3860366484702094, "maximum": 0.43849003586568813, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5959918771507623, "standard_deviation": 0.33017101006552524, "minimum": 0.17920679913691107, "maximum": 1.0548424282275077, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05103274398190811, "standard_deviation": 0.010494725396894826, "minimum": 0.0391633606008449, "maximum": 0.0657399593713771, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0004", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.1992058753967285, "predicted_mos": 2.1992058753967285, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "7336b31c400435c83ad5ef521231bd7e3a11795e97937d845e763505dcbd17c5", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.309517499990761, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7742494258545011, "全帯域_10ms_rms_envelope_nrmse": 0.09143735146451072, "全帯域_10ms_envelope_correlation": 0.9938774949926595, "全帯域_10ms_modulation_mae_db": 0.41970480830116474, "全帯域_40ms_log_rms_mae_db": 0.44676136014550416, "全帯域_40ms_rms_envelope_nrmse": 0.058544091717245915, "全帯域_40ms_envelope_correlation": 0.9975360207059404, "全帯域_40ms_modulation_mae_db": 0.7404980258598505, "300–1000 Hz_10ms_log_rms_mae_db": 1.0487026903552192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1646574711354349, "300–1000 Hz_10ms_envelope_correlation": 0.981089257962414, "300–1000 Hz_10ms_modulation_mae_db": 0.5494026636620782, "300–1000 Hz_40ms_log_rms_mae_db": 0.7502589279587776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1079686355324158, "300–1000 Hz_40ms_envelope_correlation": 0.9919870590459502, "300–1000 Hz_40ms_modulation_mae_db": 0.43966662372031284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2203127488518597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09989224945069401, "1000–3000 Hz_10ms_envelope_correlation": 0.9939672329665835, "1000–3000 Hz_10ms_modulation_mae_db": 1.1287873876195855, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7427471903923738, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056030131445962215, "1000–3000 Hz_40ms_envelope_correlation": 0.9980917266467766, "1000–3000 Hz_40ms_modulation_mae_db": 1.1186352162997086, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2611356548813664, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21050554748992548, "3000–6000 Hz_10ms_envelope_correlation": 0.96604898432202, "3000–6000 Hz_10ms_modulation_mae_db": 0.7291258797300734, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7326800232637372, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11775427040018181, "3000–6000 Hz_40ms_envelope_correlation": 0.9885519844683911, "3000–6000 Hz_40ms_modulation_mae_db": 0.8640219948389612, "mr_stft_db": 5.12389246265002, "band_mae_db": 0.24226109643186242, "rms_gain_error_db": -0.1654568870436282, "output_peak": 0.6435907585036993}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7896644874070083, "全帯域_10ms_rms_envelope_nrmse": 0.0935954840273609, "全帯域_10ms_envelope_correlation": 0.9936786971240985, "全帯域_10ms_modulation_mae_db": 0.5113189439988329, "全帯域_40ms_log_rms_mae_db": 0.41996964747701226, "全帯域_40ms_rms_envelope_nrmse": 0.04769103954846765, "全帯域_40ms_envelope_correlation": 0.9986171602633287, "全帯域_40ms_modulation_mae_db": 0.5660863136531472, "300–1000 Hz_10ms_log_rms_mae_db": 1.101377101442199, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13885192861470813, "300–1000 Hz_10ms_envelope_correlation": 0.9874449878359205, "300–1000 Hz_10ms_modulation_mae_db": 0.9113389816221624, "300–1000 Hz_40ms_log_rms_mae_db": 0.7198198086192509, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08342018129022652, "300–1000 Hz_40ms_envelope_correlation": 0.9963961172422703, "300–1000 Hz_40ms_modulation_mae_db": 1.1572920607197261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1744224021277825, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13770244904793147, "1000–3000 Hz_10ms_envelope_correlation": 0.9884798425611685, "1000–3000 Hz_10ms_modulation_mae_db": 0.771713256083606, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7084661936189899, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07212468186264663, "1000–3000 Hz_40ms_envelope_correlation": 0.9967613838403474, "1000–3000 Hz_40ms_modulation_mae_db": 0.8117619688793017, "3000–6000 Hz_10ms_log_rms_mae_db": 1.339071498751291, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2124580356144756, "3000–6000 Hz_10ms_envelope_correlation": 0.9670369528480097, "3000–6000 Hz_10ms_modulation_mae_db": 0.37064650376024133, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7877186260001313, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12071234862685468, "3000–6000 Hz_40ms_envelope_correlation": 0.9892291539287154, "3000–6000 Hz_40ms_modulation_mae_db": 0.7190934465407035, "mr_stft_db": 5.135259163735471, "band_mae_db": 0.27194383800409155, "rms_gain_error_db": -0.20265514673174836, "output_peak": 0.7145055593030621}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7755614531656283, "全帯域_10ms_rms_envelope_nrmse": 0.09168095065172993, "全帯域_10ms_envelope_correlation": 0.9936612975779083, "全帯域_10ms_modulation_mae_db": 0.24335153074312585, "全帯域_40ms_log_rms_mae_db": 0.41168225965621014, "全帯域_40ms_rms_envelope_nrmse": 0.05004189045234289, "全帯域_40ms_envelope_correlation": 0.9980151711942742, "全帯域_40ms_modulation_mae_db": 0.3594427977755415, "300–1000 Hz_10ms_log_rms_mae_db": 1.0983364671738038, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.172470538200879, "300–1000 Hz_10ms_envelope_correlation": 0.9790688920730333, "300–1000 Hz_10ms_modulation_mae_db": 0.8774558133411717, "300–1000 Hz_40ms_log_rms_mae_db": 0.6863757469246694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08143613021942912, "300–1000 Hz_40ms_envelope_correlation": 0.9950473893005193, "300–1000 Hz_40ms_modulation_mae_db": 0.6964465694781254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1886572485475815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11210376810820796, "1000–3000 Hz_10ms_envelope_correlation": 0.9924118348853165, "1000–3000 Hz_10ms_modulation_mae_db": 1.2324309979187384, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7149129744584628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06194816208639147, "1000–3000 Hz_40ms_envelope_correlation": 0.9976176578000857, "1000–3000 Hz_40ms_modulation_mae_db": 0.9297898652869371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3253947248294353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21527533519696262, "3000–6000 Hz_10ms_envelope_correlation": 0.9643880977292055, "3000–6000 Hz_10ms_modulation_mae_db": 0.4168589381806244, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8383709948388394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13791531317984304, "3000–6000 Hz_40ms_envelope_correlation": 0.9840937266627203, "3000–6000 Hz_40ms_modulation_mae_db": 0.5309564733258464, "mr_stft_db": 5.128505067166741, "band_mae_db": 0.1217025601470036, "rms_gain_error_db": -0.028547748367566468, "output_peak": 0.7512628030304722}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7798029315211423, "全帯域_10ms_rms_envelope_nrmse": 0.07925802240690871, "全帯域_10ms_envelope_correlation": 0.995262778110261, "全帯域_10ms_modulation_mae_db": 0.2658107648507263, "全帯域_40ms_log_rms_mae_db": 0.40502897319870446, "全帯域_40ms_rms_envelope_nrmse": 0.03665768170356265, "全帯域_40ms_envelope_correlation": 0.9989428712740004, "全帯域_40ms_modulation_mae_db": 0.14015406920492887, "300–1000 Hz_10ms_log_rms_mae_db": 1.015979496858151, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12171100456376052, "300–1000 Hz_10ms_envelope_correlation": 0.9895147062715863, "300–1000 Hz_10ms_modulation_mae_db": 0.26126846886161814, "300–1000 Hz_40ms_log_rms_mae_db": 0.6912002196105959, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057206816883919445, "300–1000 Hz_40ms_envelope_correlation": 0.9975953481059492, "300–1000 Hz_40ms_modulation_mae_db": 0.2804346302503928, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1886290978012224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11477210197754006, "1000–3000 Hz_10ms_envelope_correlation": 0.9920064951696019, "1000–3000 Hz_10ms_modulation_mae_db": 1.1975524401041733, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6809054457027242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057616663599279846, "1000–3000 Hz_40ms_envelope_correlation": 0.9979268189718773, "1000–3000 Hz_40ms_modulation_mae_db": 0.9049756213245295, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3375673758169837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20236750415521892, "3000–6000 Hz_10ms_envelope_correlation": 0.9683344058537405, "3000–6000 Hz_10ms_modulation_mae_db": 0.6802927340179322, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7781971101899459, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11110227898764201, "3000–6000 Hz_40ms_envelope_correlation": 0.9896730942796945, "3000–6000 Hz_40ms_modulation_mae_db": 0.6617315824609713, "mr_stft_db": 5.130935437780892, "band_mae_db": 0.13904086410228506, "rms_gain_error_db": -0.03886389555046952, "output_peak": 0.6634313613412882}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7804041768054443, "全帯域_10ms_rms_envelope_nrmse": 0.088297974408527, "全帯域_10ms_envelope_correlation": 0.9941170616214702, "全帯域_10ms_modulation_mae_db": 0.2094265603587452, "全帯域_40ms_log_rms_mae_db": 0.3997652269671664, "全帯域_40ms_rms_envelope_nrmse": 0.056219596232581104, "全帯域_40ms_envelope_correlation": 0.9974964756887218, "全帯域_40ms_modulation_mae_db": 0.40708047371039835, "300–1000 Hz_10ms_log_rms_mae_db": 1.0308418930892373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15121540212713452, "300–1000 Hz_10ms_envelope_correlation": 0.983805310090877, "300–1000 Hz_10ms_modulation_mae_db": 0.16998516000960995, "300–1000 Hz_40ms_log_rms_mae_db": 0.7033571892147872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09809088740420463, "300–1000 Hz_40ms_envelope_correlation": 0.9929286140217657, "300–1000 Hz_40ms_modulation_mae_db": 0.6831292990495185, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1628757973912114, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10652925203506868, "1000–3000 Hz_10ms_envelope_correlation": 0.9931468302851707, "1000–3000 Hz_10ms_modulation_mae_db": 0.6675619049448906, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6747651351607941, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07054422655714208, "1000–3000 Hz_40ms_envelope_correlation": 0.9969238044021855, "1000–3000 Hz_40ms_modulation_mae_db": 0.7377662562624634, "3000–6000 Hz_10ms_log_rms_mae_db": 1.306624731597167, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20423045928335476, "3000–6000 Hz_10ms_envelope_correlation": 0.9680249646286109, "3000–6000 Hz_10ms_modulation_mae_db": 0.6361209325022948, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7758864782829626, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11690579588044779, "3000–6000 Hz_40ms_envelope_correlation": 0.9886944373210093, "3000–6000 Hz_40ms_modulation_mae_db": 0.8999763923575159, "mr_stft_db": 5.120828607048634, "band_mae_db": 0.2976568492321619, "rms_gain_error_db": -0.04755997859689566, "output_peak": 0.7543838347098987}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7799137854439241, "全帯域_10ms_rms_envelope_nrmse": 0.08868839346775877, "全帯域_10ms_envelope_correlation": 0.9940704459271731, "全帯域_10ms_modulation_mae_db": 0.596478217430766, "全帯域_40ms_log_rms_mae_db": 0.41743117215612774, "全帯域_40ms_rms_envelope_nrmse": 0.045844932503731034, "全帯域_40ms_envelope_correlation": 0.9983352885066067, "全帯域_40ms_modulation_mae_db": 0.3576441264896187, "300–1000 Hz_10ms_log_rms_mae_db": 1.1172302353680583, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1419415858966281, "300–1000 Hz_10ms_envelope_correlation": 0.9860989554002786, "300–1000 Hz_10ms_modulation_mae_db": 0.4140091031111179, "300–1000 Hz_40ms_log_rms_mae_db": 0.75561702108372, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07834233086389861, "300–1000 Hz_40ms_envelope_correlation": 0.9961242893471385, "300–1000 Hz_40ms_modulation_mae_db": 0.5177906069004403, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1744320841633957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12131948697076368, "1000–3000 Hz_10ms_envelope_correlation": 0.9916234877785198, "1000–3000 Hz_10ms_modulation_mae_db": 0.32542174352845904, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6723837968688618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06132654139285711, "1000–3000 Hz_40ms_envelope_correlation": 0.9980438094872678, "1000–3000 Hz_40ms_modulation_mae_db": 0.27800127952986214, "3000–6000 Hz_10ms_log_rms_mae_db": 1.356312955645033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2127816270727048, "3000–6000 Hz_10ms_envelope_correlation": 0.9649714812174437, "3000–6000 Hz_10ms_modulation_mae_db": 0.8993875238816798, "3000–6000 Hz_40ms_log_rms_mae_db": 0.835669859796904, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1263061116704084, "3000–6000 Hz_40ms_envelope_correlation": 0.9865360389836334, "3000–6000 Hz_40ms_modulation_mae_db": 0.9606039697579543, "mr_stft_db": 5.139736019536377, "band_mae_db": 0.2601377348875332, "rms_gain_error_db": -0.017972156462649297, "output_peak": 0.7487665443280129}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7852983786869249, "全帯域_10ms_rms_envelope_nrmse": 0.08821670368351599, "全帯域_10ms_envelope_correlation": 0.9941680690641743, "全帯域_10ms_modulation_mae_db": 0.2598005333429047, "全帯域_40ms_log_rms_mae_db": 0.4129194939234026, "全帯域_40ms_rms_envelope_nrmse": 0.049018105193781504, "全帯域_40ms_envelope_correlation": 0.9981685825087525, "全帯域_40ms_modulation_mae_db": 0.5841991966314263, "300–1000 Hz_10ms_log_rms_mae_db": 1.0737862049408442, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14260462587613568, "300–1000 Hz_10ms_envelope_correlation": 0.9857820355549318, "300–1000 Hz_10ms_modulation_mae_db": 0.5369078162859727, "300–1000 Hz_40ms_log_rms_mae_db": 0.6940860143483398, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0817856934993976, "300–1000 Hz_40ms_envelope_correlation": 0.9954392930021573, "300–1000 Hz_40ms_modulation_mae_db": 0.7854611256562413, "1000–3000 Hz_10ms_log_rms_mae_db": 1.188299300029397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10366248134773881, "1000–3000 Hz_10ms_envelope_correlation": 0.993493330687763, "1000–3000 Hz_10ms_modulation_mae_db": 0.34927951619568204, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7140694987901399, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05976071782211809, "1000–3000 Hz_40ms_envelope_correlation": 0.9978109262204632, "1000–3000 Hz_40ms_modulation_mae_db": 0.3516735243278392, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3107505335196328, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2088385401036413, "3000–6000 Hz_10ms_envelope_correlation": 0.9666905899003483, "3000–6000 Hz_10ms_modulation_mae_db": 0.2789382913648343, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7743935305534156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11138612097413085, "3000–6000 Hz_40ms_envelope_correlation": 0.9898746237265336, "3000–6000 Hz_40ms_modulation_mae_db": 0.5170586113057899, "mr_stft_db": 5.1299347242201625, "band_mae_db": 0.20633160288961716, "rms_gain_error_db": -0.09652064713814207, "output_peak": 0.7907149249175432}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7709193337300496, "全帯域_10ms_rms_envelope_nrmse": 0.09315499052648114, "全帯域_10ms_envelope_correlation": 0.9935886546680626, "全帯域_10ms_modulation_mae_db": 0.5631769022182259, "全帯域_40ms_log_rms_mae_db": 0.40722225697209924, "全帯域_40ms_rms_envelope_nrmse": 0.0479958249874537, "全帯域_40ms_envelope_correlation": 0.998395182591271, "全帯域_40ms_modulation_mae_db": 0.25892805786471307, "300–1000 Hz_10ms_log_rms_mae_db": 1.0688276353549695, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14349163489947198, "300–1000 Hz_10ms_envelope_correlation": 0.9856522659109295, "300–1000 Hz_10ms_modulation_mae_db": 0.5166113006825158, "300–1000 Hz_40ms_log_rms_mae_db": 0.6692034565753348, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08068250518761512, "300–1000 Hz_40ms_envelope_correlation": 0.9956731722036323, "300–1000 Hz_40ms_modulation_mae_db": 0.4948023393551268, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2392498752648091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11423714173105996, "1000–3000 Hz_10ms_envelope_correlation": 0.9920889598590514, "1000–3000 Hz_10ms_modulation_mae_db": 0.28892445875543155, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7139389030692604, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05977728743119475, "1000–3000 Hz_40ms_envelope_correlation": 0.9977698395044128, "1000–3000 Hz_40ms_modulation_mae_db": 0.1671656105532666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3244248546839723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20236841250062904, "3000–6000 Hz_10ms_envelope_correlation": 0.9672267207507956, "3000–6000 Hz_10ms_modulation_mae_db": 0.45558561739462294, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8145297989401489, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11584404420474737, "3000–6000 Hz_40ms_envelope_correlation": 0.9882511948543321, "3000–6000 Hz_40ms_modulation_mae_db": 0.7103953480745888, "mr_stft_db": 5.121110869013072, "band_mae_db": 0.17141244858674867, "rms_gain_error_db": -0.15570066880317707, "output_peak": 0.6741037207544798}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.405821451335607, "spectral_convergence": 0.3660765759554479, "spectral_convergence_status": "ok", "amplitude_floor": 1.4952467880734047e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.9752217148840225, "spectral_convergence": 0.3771500729794749, "spectral_convergence_status": "ok", "amplitude_floor": 1.087083783284318e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.9906342217304305, "spectral_convergence": 0.3874372454306109, "spectral_convergence_status": "ok", "amplitude_floor": 6.922347653241366e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.706318352359072e-05, "output_power": 6.21470534940406e-05, "delta_db": 0.37064545719684433, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006570402135178566, "output_power": 0.000648646264914776, "delta_db": -0.055840293087520115, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010679696452407233, "output_power": 0.0009569378119443578, "delta_db": -0.4767519356560223, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007107211898605409, "output_power": 0.0006861038344332188, "delta_db": -0.15309417469414136, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005187917577577873, "output_power": 0.0005372489609216152, "delta_db": 0.15182516542286878, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.297230867177819e-05, "output_power": 8.779587865605775e-05, "delta_db": 0.2454095525337774, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24226109643186242, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1654568870436282, "pitch": null, "output_rms_dbfs": -25.25947019898264, "output_peak": 0.6435907585036993, "reference_rms_dbfs": -25.09401331193901, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.12389246265002}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.0002227499093660225, "log_rms_mae_db": 0.7742494258545011, "rms_envelope_nrmse": 0.09143735146451072, "envelope_correlation": 0.9938774949926595, "rms_gain_error_db": -0.16545688457974111, "reference_mean_power": 0.003098689875324867, "output_mean_power": 0.002982856939256296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.065525005901086e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018487969313489734, "output_power": 0.0017817539453983922, "reference_active": true, "delta_db": -0.16041482482364156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000254937700664643, "output_power": 0.00022129385317867857, "reference_active": true, "delta_db": -0.6146471368903434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5927620235575996e-05, "output_power": 4.01636984561266e-05, "reference_active": true, "delta_db": 0.48405246318950923}], "active_band_count": 3, "active_band_mae_db": 0.41970480830116474}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00020550660315895724, "log_rms_mae_db": 0.44676136014550416, "rms_envelope_nrmse": 0.058544091717245915, "envelope_correlation": 0.9975360207059404, "rms_gain_error_db": -0.16545689199642152, "reference_mean_power": 0.003123712899469445, "output_mean_power": 0.003006944566060466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 2.850551559013711e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017535539962877898, "output_power": 0.0016926158950328323, "reference_active": true, "delta_db": -0.1536072849564507}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011841366649455362, "output_power": 0.00010346168592468727, "reference_active": true, "delta_db": -0.5862227754622217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7582705881956915e-06, "output_power": 2.4731594269689204e-06, "reference_active": true, "delta_db": 1.4816640171608793}], "active_band_count": 3, "active_band_mae_db": 0.7404980258598505}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00016827060195170467, "log_rms_mae_db": 1.0487026903552192, "rms_envelope_nrmse": 0.1646574711354349, "envelope_correlation": 0.981089257962414, "rms_gain_error_db": -0.31622067980947177, "reference_mean_power": 0.001201501254742638, "output_mean_power": 0.0011171260025590554}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.2448894715706734e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007295274990436897, "output_power": 0.0006883643257256588, "reference_active": true, "delta_db": -0.2522331191591292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014476368795830424, "output_power": 0.00012237552378215863, "reference_active": true, "delta_db": -0.72965074890592}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.785748904828138e-05, "output_power": 2.389513872333249e-05, "reference_active": true, "delta_db": -0.6663241229211856}], "active_band_count": 3, "active_band_mae_db": 0.5494026636620782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012583494842815252, "log_rms_mae_db": 0.7502589279587776, "rms_envelope_nrmse": 0.1079686355324158, "envelope_correlation": 0.9919870590459502, "rms_gain_error_db": -0.31622067984221025, "reference_mean_power": 0.0012112038213410922, "output_mean_power": 0.001126147207810161}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1451125517965788e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007097609595329509, "output_power": 0.0006618985332825005, "reference_active": true, "delta_db": -0.30320688535229207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.709378020571589e-05, "output_power": 5.520007290565288e-05, "reference_active": true, "delta_db": -0.8474261031113746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7774397828761626e-06, "output_power": 1.7098507692648991e-06, "reference_active": true, "delta_db": -0.16836688269727187}], "active_band_count": 3, "active_band_mae_db": 0.43966662372031284}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015339527329010622, "log_rms_mae_db": 1.2203127488518597, "rms_envelope_nrmse": 0.09989224945069401, "envelope_correlation": 0.9939672329665835, "rms_gain_error_db": -0.0919895271751105, "reference_mean_power": 0.0009604249251386078, "output_mean_power": 0.0009402957425203817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.606398277844274e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000996843421347699, "output_power": 0.0009642952589992473, "reference_active": true, "delta_db": -0.14416915554525303}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073134134931949, "output_power": 9.954149727542251e-05, "reference_active": true, "delta_db": -0.46266391800316475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0381145318442363e-05, "output_power": 1.968784490218213e-05, "reference_active": true, "delta_db": 2.779529089310339}], "active_band_count": 3, "active_band_mae_db": 1.1287873876195855}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.0001445669043392124, "log_rms_mae_db": 0.7427471903923738, "rms_envelope_nrmse": 0.056030131445962215, "envelope_correlation": 0.9980917266467766, "rms_gain_error_db": -0.09198954406664592, "reference_mean_power": 0.0009681806844812572, "output_mean_power": 0.0009478889481507106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.5548349296058167e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009718305477517134, "output_power": 0.0009451473369161136, "reference_active": true, "delta_db": -0.12091031222797595}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.313569881232493e-05, "output_power": 4.869470590540552e-05, "reference_active": true, "delta_db": -0.37904649331315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.152693877304456e-07, "output_power": 1.1875726631251134e-06, "reference_active": true, "delta_db": 2.855948843358}], "active_band_count": 3, "active_band_mae_db": 1.1186352162997086}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.2745097516618104e-05, "log_rms_mae_db": 1.2611356548813664, "rms_envelope_nrmse": 0.21050554748992548, "envelope_correlation": 0.96604898432202, "rms_gain_error_db": 0.17299026278879898, "reference_mean_power": 0.0001383637841793036, "output_mean_power": 0.00014398639471531358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 9.868150581916701e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.732185415785845e-05, "output_power": 5.891802497835376e-05, "reference_active": true, "delta_db": 0.119279503717431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6922402540887435e-05, "output_power": 1.580620124312565e-05, "reference_active": true, "delta_db": -0.29634514561181874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.249240297758961e-06, "output_power": 6.389790075284764e-06, "reference_active": true, "delta_db": 1.7717529898609703}], "active_band_count": 3, "active_band_mae_db": 0.7291258797300734}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.734940150914352e-05, "log_rms_mae_db": 0.7326800232637372, "rms_envelope_nrmse": 0.11775427040018181, "envelope_correlation": 0.9885519844683911, "rms_gain_error_db": 0.17299028381239034, "reference_mean_power": 0.00013948111618043138, "output_mean_power": 0.00014514913180520712}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 8.397728752995274e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.502560023531223e-05, "output_power": 5.777622489027125e-05, "reference_active": true, "delta_db": 0.21184372887392927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.667910609926144e-06, "output_power": 7.6239304369215336e-06, "reference_active": true, "delta_db": -0.024981163556245203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2898812052596812e-07, "output_power": 3.938556334861561e-07, "reference_active": true, "delta_db": 2.3552410920867093}], "active_band_count": 3, "active_band_mae_db": 0.8640219948389612}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9084327652184436, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d442e297c543d5afa8e109a0badaf881504d0a707ecfaf8c5686a83ab7ea8e3b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.992152251774147, "standard_deviation": 0.0016877986860559064, "minimum": 0.9884798425611685, "maximum": 0.9939672329665835, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1921098192721573, "standard_deviation": 0.02545618341521694, "minimum": 1.1628757973912114, "maximum": 1.2392498752648091, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7452089631438208, "standard_deviation": 0.40303152462507746, "minimum": 0.28892445875543155, "maximum": 1.2324309979187384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11377736633362559, "standard_deviation": 0.011837817849371792, "minimum": 0.09989224945069401, "maximum": 0.13770244904793147, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997618245859177, "standard_deviation": 0.0005038528669699939, "minimum": 0.9967613838403474, "maximum": 0.9980917266467766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7027736422577009, "standard_deviation": 0.02454194571012218, "minimum": 0.6723837968688618, "maximum": 0.7427471903923738, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6624711678079885, "standard_deviation": 0.3497040164801116, "minimum": 0.1671656105532666, "maximum": 1.1186352162997086, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06239105152469902, "standard_deviation": 0.005849953088636282, "minimum": 0.056030131445962215, "maximum": 0.07212468186264663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9665902746562718, "standard_deviation": 0.001387600589855298, "minimum": 0.9643880977292055, "maximum": 0.9683344058537405, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3201602912156103, "standard_deviation": 0.028707649342560322, "minimum": 1.2611356548813664, "maximum": 1.356312955645033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.558369552604038, "standard_deviation": 0.21053438246065598, "minimum": 0.2789382913648343, "maximum": 0.8993875238816798, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20860318267711406, "standard_deviation": 0.005033999384803808, "minimum": 0.20236750415521892, "maximum": 0.21527533519696262, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9881130317781287, "standard_deviation": 0.001926293652378343, "minimum": 0.9840937266627203, "maximum": 0.9898746237265336, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7921808027332606, "standard_deviation": 0.03558089186067346, "minimum": 0.7326800232637372, "maximum": 0.8383709948388394, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7329797273327914, "standard_deviation": 0.16478872290492771, "minimum": 0.5170586113057899, "maximum": 0.9606039697579543, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11974078549053199, "standard_deviation": 0.008831381660464033, "minimum": 0.11110227898764201, "maximum": 0.13791531317984304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9848070513874965, "standard_deviation": 0.0033849808356565233, "minimum": 0.9790688920730333, "maximum": 0.9895147062715863, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0693852155728103, "standard_deviation": 0.035729152966543935, "minimum": 1.015979496858151, "maximum": 1.1172302353680583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5296224134470309, "standard_deviation": 0.262414194986033, "minimum": 0.16998516000960995, "maximum": 0.9113389816221624, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1471180239142691, "standard_deviation": 0.01577218451742711, "minimum": 0.12171100456376052, "maximum": 0.172470538200879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9951489102836728, "standard_deviation": 0.001843461242467037, "minimum": 0.9919870590459502, "maximum": 0.9975953481059492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7087397980419344, "standard_deviation": 0.03084221265077062, "minimum": 0.6692034565753348, "maximum": 0.75561702108372, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6318779068912356, "standard_deviation": 0.26691540468972313, "minimum": 0.2804346302503928, "maximum": 1.1572920607197261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08361664761013836, "standard_deviation": 0.014854414203612597, "minimum": 0.057206816883919445, "maximum": 0.1079686355324158, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21381087428516293, "standard_deviation": 0.06468963716686085, "minimum": 0.1217025601470036, "maximum": 0.2976568492321619, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.128775293893922, "standard_deviation": 0.006697514240766721, "minimum": 5.120828607048634, "maximum": 5.139736019536377, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7175949383610571, "standard_deviation": 0.052258275410758726, "minimum": 0.6435907585036993, "maximum": 0.7907149249175432, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0941596410867846, "standard_deviation": 0.07169981646349378, "minimum": -0.20265514673174836, "maximum": -0.017972156462649297, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994053062385726, "standard_deviation": 0.0005373432278289364, "minimum": 0.9935886546680626, "maximum": 0.995262778110261, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7794767465768279, "standard_deviation": 0.0060323261223712244, "minimum": 0.7709193337300496, "maximum": 0.7896644874070083, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38363353265556144, "standard_deviation": 0.15783685436773817, "minimum": 0.2094265603587452, "maximum": 0.596478217430766, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08929123382959914, "standard_deviation": 0.004581666575091045, "minimum": 0.07925802240690871, "maximum": 0.0935954840273609, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998188344091612, "standard_deviation": 0.0005003079713064539, "minimum": 0.9974964756887218, "maximum": 0.9989428712740004, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.41509754881202837, "standard_deviation": 0.01437589621648899, "minimum": 0.3997652269671664, "maximum": 0.44676136014550416, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42675413264870304, "standard_deviation": 0.19369965141807266, "minimum": 0.14015406920492887, "maximum": 0.7404980258598505, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049001645292395805, "standard_deviation": 0.006647310036845336, "minimum": 0.03665768170356265, "maximum": 0.058544091717245915, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0004", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.821998119354248, "predicted_mos": 1.821998119354248, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "d78b28c08b3f3c12872b92a7a75e630a3c873c59b6aa9667cb1e0064564af050", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.982836700044572, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8103731798835127, "全帯域_10ms_rms_envelope_nrmse": 0.08260989038880717, "全帯域_10ms_envelope_correlation": 0.9950665241211057, "全帯域_10ms_modulation_mae_db": 0.7407899075700376, "全帯域_40ms_log_rms_mae_db": 0.4412953971623942, "全帯域_40ms_rms_envelope_nrmse": 0.041957547956940254, "全帯域_40ms_envelope_correlation": 0.9988957309620276, "全帯域_40ms_modulation_mae_db": 0.6098344746115772, "300–1000 Hz_10ms_log_rms_mae_db": 1.057217620009633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12885063744387917, "300–1000 Hz_10ms_envelope_correlation": 0.9887144827243969, "300–1000 Hz_10ms_modulation_mae_db": 0.6881270762169421, "300–1000 Hz_40ms_log_rms_mae_db": 0.7492860952282754, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08269132322986932, "300–1000 Hz_40ms_envelope_correlation": 0.9956032624821316, "300–1000 Hz_40ms_modulation_mae_db": 0.7125148777696783, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2203326567473471, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11377626251136709, "1000–3000 Hz_10ms_envelope_correlation": 0.992140152098564, "1000–3000 Hz_10ms_modulation_mae_db": 0.44979056156595226, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7027186881831163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06329317521452295, "1000–3000 Hz_40ms_envelope_correlation": 0.9975105088191758, "1000–3000 Hz_40ms_modulation_mae_db": 0.5334396909959003, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3246380987849744, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21801205247246896, "3000–6000 Hz_10ms_envelope_correlation": 0.9627119796589384, "3000–6000 Hz_10ms_modulation_mae_db": 0.5220570158283754, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7991824647221533, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13010272256757294, "3000–6000 Hz_40ms_envelope_correlation": 0.9854207518682124, "3000–6000 Hz_40ms_modulation_mae_db": 0.5160929948961486, "mr_stft_db": 5.1018853869158445, "band_mae_db": 0.2484766652104866, "rms_gain_error_db": -0.17147548576406302, "output_peak": 0.6472035922842435}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.79516132181072, "全帯域_10ms_rms_envelope_nrmse": 0.09664686330380425, "全帯域_10ms_envelope_correlation": 0.9932484099094301, "全帯域_10ms_modulation_mae_db": 0.6783241628205765, "全帯域_40ms_log_rms_mae_db": 0.41906895056862475, "全帯域_40ms_rms_envelope_nrmse": 0.05881867530131796, "全帯域_40ms_envelope_correlation": 0.997669724973105, "全帯域_40ms_modulation_mae_db": 0.5204359746008101, "300–1000 Hz_10ms_log_rms_mae_db": 1.005819500877354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13347054240512524, "300–1000 Hz_10ms_envelope_correlation": 0.9877586734114312, "300–1000 Hz_10ms_modulation_mae_db": 0.9252240589122683, "300–1000 Hz_40ms_log_rms_mae_db": 0.6254668035816845, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08205601764090902, "300–1000 Hz_40ms_envelope_correlation": 0.9956000912962374, "300–1000 Hz_40ms_modulation_mae_db": 0.7419084356107989, "1000–3000 Hz_10ms_log_rms_mae_db": 1.204769137425657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13228943741275495, "1000–3000 Hz_10ms_envelope_correlation": 0.9894941578001818, "1000–3000 Hz_10ms_modulation_mae_db": 0.9218770357833314, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7343827280473788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07186180671851083, "1000–3000 Hz_40ms_envelope_correlation": 0.99706099109203, "1000–3000 Hz_40ms_modulation_mae_db": 0.7780788431856781, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3475306522513195, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20213903981036146, "3000–6000 Hz_10ms_envelope_correlation": 0.9689513595551132, "3000–6000 Hz_10ms_modulation_mae_db": 0.12632120945786032, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7870691587429528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11320156385357728, "3000–6000 Hz_40ms_envelope_correlation": 0.9896290507648062, "3000–6000 Hz_40ms_modulation_mae_db": 0.4614594857940941, "mr_stft_db": 5.111095272519308, "band_mae_db": 0.18824692688783276, "rms_gain_error_db": -0.21116882033998124, "output_peak": 0.6313510434653523}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7821366082425204, "全帯域_10ms_rms_envelope_nrmse": 0.08817243836795, "全帯域_10ms_envelope_correlation": 0.9944428344328592, "全帯域_10ms_modulation_mae_db": 0.5779204831821175, "全帯域_40ms_log_rms_mae_db": 0.402916785864968, "全帯域_40ms_rms_envelope_nrmse": 0.042915860473949394, "全帯域_40ms_envelope_correlation": 0.9989675217401807, "全帯域_40ms_modulation_mae_db": 0.4034314057981223, "300–1000 Hz_10ms_log_rms_mae_db": 0.9960569081242049, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13401863819508894, "300–1000 Hz_10ms_envelope_correlation": 0.9884360864445189, "300–1000 Hz_10ms_modulation_mae_db": 1.0518028807388606, "300–1000 Hz_40ms_log_rms_mae_db": 0.6731023320218505, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08110688212686423, "300–1000 Hz_40ms_envelope_correlation": 0.9967000385702202, "300–1000 Hz_40ms_modulation_mae_db": 1.0283177607294358, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2279540145661374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11579829370028186, "1000–3000 Hz_10ms_envelope_correlation": 0.9918607013733357, "1000–3000 Hz_10ms_modulation_mae_db": 0.610888105046523, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7081673930059239, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059569813410684756, "1000–3000 Hz_40ms_envelope_correlation": 0.997789927581306, "1000–3000 Hz_40ms_modulation_mae_db": 0.7700936190004741, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3807149442681754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24848964825705463, "3000–6000 Hz_10ms_envelope_correlation": 0.9534336937554796, "3000–6000 Hz_10ms_modulation_mae_db": 0.8959076368678147, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8490260533691295, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15255953978635628, "3000–6000 Hz_40ms_envelope_correlation": 0.9811888068097853, "3000–6000 Hz_40ms_modulation_mae_db": 1.238984504265034, "mr_stft_db": 5.1065339414952104, "band_mae_db": 0.21272653758552273, "rms_gain_error_db": -0.20888148245913873, "output_peak": 0.6939612886612225}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7715360627790971, "全帯域_10ms_rms_envelope_nrmse": 0.08614481412190433, "全帯域_10ms_envelope_correlation": 0.9945169783256675, "全帯域_10ms_modulation_mae_db": 0.4866902122791619, "全帯域_40ms_log_rms_mae_db": 0.39569487559687244, "全帯域_40ms_rms_envelope_nrmse": 0.04219080413548349, "全帯域_40ms_envelope_correlation": 0.9987677430561401, "全帯域_40ms_modulation_mae_db": 0.4515638131869813, "300–1000 Hz_10ms_log_rms_mae_db": 1.0784619842643848, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13394389728546474, "300–1000 Hz_10ms_envelope_correlation": 0.9873040946237067, "300–1000 Hz_10ms_modulation_mae_db": 0.5287916966886724, "300–1000 Hz_40ms_log_rms_mae_db": 0.6894422562854544, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06673800169817222, "300–1000 Hz_40ms_envelope_correlation": 0.9966792273835164, "300–1000 Hz_40ms_modulation_mae_db": 0.4465121060095467, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2545432095445133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11369057901029055, "1000–3000 Hz_10ms_envelope_correlation": 0.9927973540430911, "1000–3000 Hz_10ms_modulation_mae_db": 1.1476566926352259, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7212962394441447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0637232089161514, "1000–3000 Hz_40ms_envelope_correlation": 0.9983296603397681, "1000–3000 Hz_40ms_modulation_mae_db": 1.0953292012270202, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3610614858967371, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22811173769787083, "3000–6000 Hz_10ms_envelope_correlation": 0.95861347157445, "3000–6000 Hz_10ms_modulation_mae_db": 0.6907634233572232, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7914301410230221, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12866304885341998, "3000–6000 Hz_40ms_envelope_correlation": 0.9855005007461638, "3000–6000 Hz_40ms_modulation_mae_db": 1.0078461819679843, "mr_stft_db": 5.110267251699327, "band_mae_db": 0.19129722051452688, "rms_gain_error_db": -0.1336218417578487, "output_peak": 0.7086435317734467}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7188560062912482, "全帯域_10ms_rms_envelope_nrmse": 0.0843798922514813, "全帯域_10ms_envelope_correlation": 0.9950575959518017, "全帯域_10ms_modulation_mae_db": 0.9305499780007773, "全帯域_40ms_log_rms_mae_db": 0.3683361708284072, "全帯域_40ms_rms_envelope_nrmse": 0.04746385784088768, "全帯域_40ms_envelope_correlation": 0.9987597105783574, "全帯域_40ms_modulation_mae_db": 0.823980438943015, "300–1000 Hz_10ms_log_rms_mae_db": 1.0063939838747011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14214002905457568, "300–1000 Hz_10ms_envelope_correlation": 0.9862858475137273, "300–1000 Hz_10ms_modulation_mae_db": 1.047435674873489, "300–1000 Hz_40ms_log_rms_mae_db": 0.6401351538363416, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08069578036050329, "300–1000 Hz_40ms_envelope_correlation": 0.9961601558437825, "300–1000 Hz_40ms_modulation_mae_db": 1.1529386065056535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2124721553226472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11309380992238591, "1000–3000 Hz_10ms_envelope_correlation": 0.9922673624082051, "1000–3000 Hz_10ms_modulation_mae_db": 0.6259264430822712, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6864026805575792, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0561513696520895, "1000–3000 Hz_40ms_envelope_correlation": 0.9981230609983971, "1000–3000 Hz_40ms_modulation_mae_db": 0.44192307462420405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.291471476800052, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2143613472898674, "3000–6000 Hz_10ms_envelope_correlation": 0.964163178677599, "3000–6000 Hz_10ms_modulation_mae_db": 0.5446555990907492, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7540826208994924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12276900885080756, "3000–6000 Hz_40ms_envelope_correlation": 0.9871192521389778, "3000–6000 Hz_40ms_modulation_mae_db": 0.6600508976668961, "mr_stft_db": 5.10574296717176, "band_mae_db": 0.17654259763352034, "rms_gain_error_db": -0.23542019867087563, "output_peak": 0.6780021604713288}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7770058978063078, "全帯域_10ms_rms_envelope_nrmse": 0.09906776157323581, "全帯域_10ms_envelope_correlation": 0.992706245259849, "全帯域_10ms_modulation_mae_db": 0.8465144758196095, "全帯域_40ms_log_rms_mae_db": 0.4082304423760048, "全帯域_40ms_rms_envelope_nrmse": 0.051219232855905586, "全帯域_40ms_envelope_correlation": 0.997981539060804, "全帯域_40ms_modulation_mae_db": 0.6982239738511469, "300–1000 Hz_10ms_log_rms_mae_db": 1.0519086641849598, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14406584368559744, "300–1000 Hz_10ms_envelope_correlation": 0.98527673533801, "300–1000 Hz_10ms_modulation_mae_db": 0.5767130790849101, "300–1000 Hz_40ms_log_rms_mae_db": 0.6900927881777366, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07924845880632893, "300–1000 Hz_40ms_envelope_correlation": 0.9953303219859109, "300–1000 Hz_40ms_modulation_mae_db": 0.31695171825223917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2118217321338565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1537358872199725, "1000–3000 Hz_10ms_envelope_correlation": 0.9865729988310981, "1000–3000 Hz_10ms_modulation_mae_db": 1.7937984062943346, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7191445935652686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0860890325044475, "1000–3000 Hz_40ms_envelope_correlation": 0.9959850436139149, "1000–3000 Hz_40ms_modulation_mae_db": 1.4403374207600932, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3792320912711407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22774442885369733, "3000–6000 Hz_10ms_envelope_correlation": 0.959714486497824, "3000–6000 Hz_10ms_modulation_mae_db": 0.46695876055992386, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8401321850102842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1394996299418388, "3000–6000 Hz_40ms_envelope_correlation": 0.9834766232249791, "3000–6000 Hz_40ms_modulation_mae_db": 0.6060968330877475, "mr_stft_db": 5.122361642043637, "band_mae_db": 0.19108006633088284, "rms_gain_error_db": 0.05590475260471338, "output_peak": 0.7019680088239584}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7545710550949852, "全帯域_10ms_rms_envelope_nrmse": 0.08670226910422814, "全帯域_10ms_envelope_correlation": 0.9945063293286834, "全帯域_10ms_modulation_mae_db": 1.0000118561623936, "全帯域_40ms_log_rms_mae_db": 0.39223804546103846, "全帯域_40ms_rms_envelope_nrmse": 0.048453058189832456, "全帯域_40ms_envelope_correlation": 0.9983884007003977, "全帯域_40ms_modulation_mae_db": 0.8009735166144306, "300–1000 Hz_10ms_log_rms_mae_db": 1.0865404870142576, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14479711979136115, "300–1000 Hz_10ms_envelope_correlation": 0.9857390566046543, "300–1000 Hz_10ms_modulation_mae_db": 1.1933076146681094, "300–1000 Hz_40ms_log_rms_mae_db": 0.7688880288222361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09262272297304548, "300–1000 Hz_40ms_envelope_correlation": 0.9945388665816918, "300–1000 Hz_40ms_modulation_mae_db": 1.2011106205656719, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1874915106021589, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11842195021042892, "1000–3000 Hz_10ms_envelope_correlation": 0.9914826643146185, "1000–3000 Hz_10ms_modulation_mae_db": 0.26370192816574656, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6836273651670645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06003443111927639, "1000–3000 Hz_40ms_envelope_correlation": 0.9977738595470718, "1000–3000 Hz_40ms_modulation_mae_db": 0.41521404647491894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.343562994745113, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.226396909700853, "3000–6000 Hz_10ms_envelope_correlation": 0.9614625161712992, "3000–6000 Hz_10ms_modulation_mae_db": 0.5496133130888773, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8186175050512241, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1307042508722074, "3000–6000 Hz_40ms_envelope_correlation": 0.986409309602361, "3000–6000 Hz_40ms_modulation_mae_db": 0.9659090210826328, "mr_stft_db": 5.112873155226772, "band_mae_db": 0.25367007193941976, "rms_gain_error_db": -0.16025460796226376, "output_peak": 0.6701075404985763}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7916757330789428, "全帯域_10ms_rms_envelope_nrmse": 0.10179008438693884, "全帯域_10ms_envelope_correlation": 0.9922335501456243, "全帯域_10ms_modulation_mae_db": 0.4298514549465399, "全帯域_40ms_log_rms_mae_db": 0.41961551038662004, "全帯域_40ms_rms_envelope_nrmse": 0.05102505087124881, "全帯域_40ms_envelope_correlation": 0.9979556224880255, "全帯域_40ms_modulation_mae_db": 0.5852165649563181, "300–1000 Hz_10ms_log_rms_mae_db": 1.0105506150702628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14677002652345375, "300–1000 Hz_10ms_envelope_correlation": 0.9847703751197725, "300–1000 Hz_10ms_modulation_mae_db": 0.805828970688019, "300–1000 Hz_40ms_log_rms_mae_db": 0.615835807893391, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06142090052468737, "300–1000 Hz_40ms_envelope_correlation": 0.9971786292221653, "300–1000 Hz_40ms_modulation_mae_db": 0.8633280201607509, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1413937257703395, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15167612160929922, "1000–3000 Hz_10ms_envelope_correlation": 0.9864710241655678, "1000–3000 Hz_10ms_modulation_mae_db": 1.8502288735050445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6309240882689946, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08306028276303498, "1000–3000 Hz_40ms_envelope_correlation": 0.9959289475588836, "1000–3000 Hz_40ms_modulation_mae_db": 1.9244966980743738, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3215929495030194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21298782847258382, "3000–6000 Hz_10ms_envelope_correlation": 0.9640876328282685, "3000–6000 Hz_10ms_modulation_mae_db": 0.3352660030131894, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7584253172452913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.127974970615262, "3000–6000 Hz_40ms_envelope_correlation": 0.9857003036661425, "3000–6000 Hz_40ms_modulation_mae_db": 0.5211420878106181, "mr_stft_db": 5.076833186441661, "band_mae_db": 0.1689948547663623, "rms_gain_error_db": 0.016269201138021237, "output_peak": 0.7494366545849878}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.3907915776880895, "spectral_convergence": 0.3615268740122204, "spectral_convergence_status": "ok", "amplitude_floor": 1.4952467880734047e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.9403251177627885, "spectral_convergence": 0.37878985178162305, "spectral_convergence_status": "ok", "amplitude_floor": 1.087083783284318e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.974539465296656, "spectral_convergence": 0.37647675884826426, "spectral_convergence_status": "ok", "amplitude_floor": 6.922347653241366e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.706318352359072e-05, "output_power": 5.972099832353714e-05, "delta_db": 0.19771061931991507, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006570402135178566, "output_power": 0.0006768148950499704, "delta_db": 0.12877956949948494, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010679696452407233, "output_power": 0.0009440590322454694, "delta_db": -0.5355975728478268, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007107211898605409, "output_power": 0.0007305195120918338, "delta_db": 0.11932556433544331, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005187917577577873, "output_power": 0.00047772604340589914, "delta_db": -0.3581415001750479, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.297230867177819e-05, "output_power": 8.591395081519112e-05, "delta_db": 0.15130516508520142, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2484766652104866, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.17147548576406302, "pitch": null, "output_rms_dbfs": -25.26548879770307, "output_peak": 0.6472035922842435, "reference_rms_dbfs": -25.09401331193901, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.1018853869158445}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.0002227499093660225, "log_rms_mae_db": 0.8103731798835127, "rms_envelope_nrmse": 0.08260989038880717, "envelope_correlation": 0.9950665241211057, "rms_gain_error_db": -0.17147547933774854, "reference_mean_power": 0.003098689875324867, "output_mean_power": 0.0029787260617121737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.065525005901086e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018487969313489734, "output_power": 0.0017352232115821038, "reference_active": true, "delta_db": -0.27533863148032656}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000254937700664643, "output_power": 0.00023241250729405066, "reference_active": true, "delta_db": -0.40174568567200586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5927620235575996e-05, "output_power": 2.5170967616434573e-05, "reference_active": true, "delta_db": -1.54528540555778}], "active_band_count": 3, "active_band_mae_db": 0.7407899075700376}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00020550660315895724, "log_rms_mae_db": 0.4412953971623942, "rms_envelope_nrmse": 0.041957547956940254, "envelope_correlation": 0.9988957309620276, "rms_gain_error_db": -0.17147547069630736, "reference_mean_power": 0.003123712899469445, "output_mean_power": 0.0030027803413191748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 2.850551559013711e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017535539962877898, "output_power": 0.0016434974610689127, "reference_active": true, "delta_db": -0.2815010592520625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011841366649455362, "output_power": 0.00010844091520938732, "reference_active": true, "delta_db": -0.3820865433341539}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7582705881956915e-06, "output_power": 1.3442935946563108e-06, "reference_active": true, "delta_db": -1.1659158212485152}], "active_band_count": 3, "active_band_mae_db": 0.6098344746115772}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00016827060195170467, "log_rms_mae_db": 1.057217620009633, "rms_envelope_nrmse": 0.12885063744387917, "envelope_correlation": 0.9887144827243969, "rms_gain_error_db": -0.2968988146529725, "reference_mean_power": 0.001201501254742638, "output_mean_power": 0.001122107195314134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.2448894715706734e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007295274990436897, "output_power": 0.0006714463135089424, "reference_active": true, "delta_db": -0.36030373243501507}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014476368795830424, "output_power": 0.0001426333474420926, "reference_active": true, "delta_db": -0.0643856381602034}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.785748904828138e-05, "output_power": 1.9097335713045572e-05, "reference_active": true, "delta_db": -1.639691858055608}], "active_band_count": 3, "active_band_mae_db": 0.6881270762169421}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012583494842815252, "log_rms_mae_db": 0.7492860952282754, "rms_envelope_nrmse": 0.08269132322986932, "envelope_correlation": 0.9956032624821316, "rms_gain_error_db": -0.29689881435827425, "reference_mean_power": 0.0012112038213410922, "output_mean_power": 0.001131168625622583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1451125517965788e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007097609595329509, "output_power": 0.0006505380239598127, "reference_active": true, "delta_db": -0.3783942110032085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.709378020571589e-05, "output_power": 6.95672466393319e-05, "reference_active": true, "delta_db": 0.15722553378476367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7774397828761626e-06, "output_power": 1.2291429868238942e-06, "reference_active": true, "delta_db": -1.6019248885210626}], "active_band_count": 3, "active_band_mae_db": 0.7125148777696783}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015339527329010622, "log_rms_mae_db": 1.2203326567473471, "rms_envelope_nrmse": 0.11377626251136709, "envelope_correlation": 0.992140152098564, "rms_gain_error_db": -0.05844658559862673, "reference_mean_power": 0.0009604249251386078, "output_mean_power": 0.0009475862797273574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.606398277844274e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000996843421347699, "output_power": 0.0009806785131046504, "reference_active": true, "delta_db": -0.07100287182265955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073134134931949, "output_power": 8.4533557417942e-05, "reference_active": true, "delta_db": -1.172414150587832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0381145318442363e-05, "output_power": 1.013094142223412e-05, "reference_active": true, "delta_db": -0.10595466228736498}], "active_band_count": 3, "active_band_mae_db": 0.44979056156595226}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.0001445669043392124, "log_rms_mae_db": 0.7027186881831163, "rms_envelope_nrmse": 0.06329317521452295, "envelope_correlation": 0.9975105088191758, "rms_gain_error_db": -0.05844656387133187, "reference_mean_power": 0.0009681806844812572, "output_mean_power": 0.0009552383674017067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.5548349296058167e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009718305477517134, "output_power": 0.0009519990683961926, "reference_active": true, "delta_db": -0.08954022815952939}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.313569881232493e-05, "output_power": 3.965945927492464e-05, "reference_active": true, "delta_db": -1.2703960791037583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.152693877304456e-07, "output_power": 5.821394497619812e-07, "reference_active": true, "delta_db": -0.240382765724413}], "active_band_count": 3, "active_band_mae_db": 0.5334396909959003}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.2745097516618104e-05, "log_rms_mae_db": 1.3246380987849744, "rms_envelope_nrmse": 0.21801205247246896, "envelope_correlation": 0.9627119796589384, "rms_gain_error_db": 0.06086469807172001, "reference_mean_power": 0.0001383637841793036, "output_mean_power": 0.00014031655097099908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 9.868150581916701e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.732185415785845e-05, "output_power": 5.524256859819308e-05, "reference_active": true, "delta_db": -0.16046366187885475}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6922402540887435e-05, "output_power": 1.4703713902781258e-05, "reference_active": true, "delta_db": -0.6103497774961668}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.249240297758961e-06, "output_power": 5.1032524445663235e-06, "reference_active": true, "delta_db": 0.7953576081101048}], "active_band_count": 3, "active_band_mae_db": 0.5220570158283754}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.734940150914352e-05, "log_rms_mae_db": 0.7991824647221533, "rms_envelope_nrmse": 0.13010272256757294, "envelope_correlation": 0.9854207518682124, "rms_gain_error_db": 0.060864679141461805, "reference_mean_power": 0.00013948111618043138, "output_mean_power": 0.00014144965157466047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 8.397728752995274e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.502560023531223e-05, "output_power": 5.346347760520128e-05, "reference_active": true, "delta_db": -0.12507584042901862}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.667910609926144e-06, "output_power": 6.913096518137544e-06, "reference_active": true, "delta_db": -0.45004420981728965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2898812052596812e-07, "output_power": 2.8650278369836534e-07, "reference_active": true, "delta_db": 0.9731589344421375}], "active_band_count": 3, "active_band_mae_db": 0.5160929948961486}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8969602161897281, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8c44c308b130531b372105a352d7a8ad3da615583e8ee4addf929617456bb387", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9903858018793328, "standard_deviation": 0.002576542951600895, "minimum": 0.9864710241655678, "maximum": 0.9927973540430911, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.207597267764082, "standard_deviation": 0.03298759099001672, "minimum": 1.1413937257703395, "maximum": 1.2545432095445133, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9579835057598036, "standard_deviation": 0.5979718133344318, "minimum": 0.26370192816574656, "maximum": 1.8502288735050445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12656029269959762, "standard_deviation": 0.017298229703482375, "minimum": 0.11309380992238591, "maximum": 0.1537358872199725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973127499438184, "standard_deviation": 0.000918946256297876, "minimum": 0.9959289475588836, "maximum": 0.9983296603397681, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6983329720299338, "standard_deviation": 0.03224574131352095, "minimum": 0.6309240882689946, "maximum": 0.7343827280473788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9248640742928328, "standard_deviation": 0.5326114261293683, "minimum": 0.41521404647491894, "maximum": 1.9244966980743738, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06797289003733978, "standard_deviation": 0.011239511946456542, "minimum": 0.0561513696520895, "maximum": 0.0860890325044475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9616422898398715, "standard_deviation": 0.004585767630608045, "minimum": 0.9534336937554796, "maximum": 0.9689513595551132, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3437255866900664, "standard_deviation": 0.03053953088505618, "minimum": 1.291471476800052, "maximum": 1.3807149442681754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5164428701580016, "standard_deviation": 0.2279653995895137, "minimum": 0.12632120945786032, "maximum": 0.8959076368678147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22228037406934467, "standard_deviation": 0.013839317279274324, "minimum": 0.20213903981036146, "maximum": 0.24848964825705463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9855555748526784, "standard_deviation": 0.0024842364598671117, "minimum": 0.9811888068097853, "maximum": 0.9896290507648062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7997456807579437, "standard_deviation": 0.03472014175171627, "minimum": 0.7540826208994924, "maximum": 0.8490260533691295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7471977508213944, "standard_deviation": 0.2856789920153708, "minimum": 0.4614594857940941, "maximum": 1.238984504265034, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1306843419176303, "standard_deviation": 0.011565333711951994, "minimum": 0.11320156385357728, "maximum": 0.15255953978635628, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9867856689725272, "standard_deviation": 0.0014805024282690718, "minimum": 0.9847703751197725, "maximum": 0.9887144827243969, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0366187204274697, "standard_deviation": 0.036031874600523386, "minimum": 0.9960569081242049, "maximum": 1.0865404870142576, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8521538814839089, "standard_deviation": 0.2417795189884402, "minimum": 0.5287916966886724, "maximum": 1.1933076146681094, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13850709179806825, "standard_deviation": 0.0066710611543684215, "minimum": 0.12885063744387917, "maximum": 0.14677002652345375, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.995973824170707, "standard_deviation": 0.0008669689664672934, "minimum": 0.9945388665816918, "maximum": 0.9971786292221653, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6815311582308712, "standard_deviation": 0.05555127170474755, "minimum": 0.615835807893391, "maximum": 0.7688880288222361, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.807947768200472, "standard_deviation": 0.31856981784708555, "minimum": 0.31695171825223917, "maximum": 1.2011106205656719, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07832251092004748, "standard_deviation": 0.009795257280983318, "minimum": 0.06142090052468737, "maximum": 0.09262272297304548, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20387936760856928, "standard_deviation": 0.03179631510151665, "minimum": 0.1689948547663623, "maximum": 0.25367007193941976, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.10594910043919, "standard_deviation": 0.013239973670122827, "minimum": 5.076833186441661, "maximum": 5.122361642043637, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6850842275703896, "standard_deviation": 0.037142187015422955, "minimum": 0.6313510434653523, "maximum": 0.7494366545849878, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13108106040142956, "standard_deviation": 0.10857431005538419, "minimum": -0.23542019867087563, "maximum": 0.05590475260471338, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939723084343777, "standard_deviation": 0.001090780893859696, "minimum": 0.9922335501456243, "maximum": 0.9950665241211057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7751644831234168, "standard_deviation": 0.02821359315788654, "minimum": 0.7188560062912482, "maximum": 0.8103731798835127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7113315663476517, "standard_deviation": 0.20671513576736045, "minimum": 0.4298514549465399, "maximum": 1.0000118561623936, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09068925168729372, "standard_deviation": 0.007337646940627442, "minimum": 0.08260989038880717, "maximum": 0.10179008438693884, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984232491948798, "standard_deviation": 0.000497730514686512, "minimum": 0.997669724973105, "maximum": 0.9989675217401807, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40592452228061626, "standard_deviation": 0.02181261486248567, "minimum": 0.3683361708284072, "maximum": 0.4412953971623942, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6117075203203002, "standard_deviation": 0.15419247107549736, "minimum": 0.4034314057981223, "maximum": 0.823980438943015, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048005510953195704, "standard_deviation": 0.005771321381354855, "minimum": 0.041957547956940254, "maximum": 0.05881867530131796, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4970-29093-0005", "language": "en", "style": "normal", "speaker": "librispeech_4970", "source_utterance_id": "4970-29093-0005", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4970-29093-0005.wav", "source_file_sha256": "d974953fd37f602dadfd3f24c476fd2dcbd3d7b256873d7563a35b4ad558b763", "decoded_pcm_sha256": "e5b971d7dd1c4efc72382481eb6d7cbe047355a62af1c39714d28b6cb739b74e", "original_source_sha256": "f466147cca3169748874913da56b587fd3d7fed00a285a1c9124dac56a8323ba", "corpus_entry_id": "librispeech_4970-29093-0005"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 2.672722578048706, "predicted_mos": 2.672722578048706, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cbbe34022bfc8eb1a67cc4d76e1352743a0d14f0a9697cad60fed51a41dab133", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.4351513759975205, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4970-29093-0005", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "cbbe34022bfc8eb1a67cc4d76e1352743a0d14f0a9697cad60fed51a41dab133", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "05fbfba7cad8726c9fa6d847225778b1f0f993516f8e315e0732959dacfa0d1d"}, "rows": [{"case": "librispeech_4970-29093-0005", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.8581464290618896, "predicted_mos": 1.8581464290618896, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "419e7e13a9628a89b3ba8e35f6e41217415565befaf88d3ab67c2569d21f4b6e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4413986600011413, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4970-29093-0005", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "419e7e13a9628a89b3ba8e35f6e41217415565befaf88d3ab67c2569d21f4b6e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1788971890273028, "全帯域_10ms_rms_envelope_nrmse": 0.07858144308069683, "全帯域_10ms_envelope_correlation": 0.995722004395199, "全帯域_10ms_modulation_mae_db": 0.8214307565697978, "全帯域_40ms_log_rms_mae_db": 0.7254679911637909, "全帯域_40ms_rms_envelope_nrmse": 0.04222155067267396, "全帯域_40ms_envelope_correlation": 0.9988586346783744, "全帯域_40ms_modulation_mae_db": 0.7735536649126171, "300–1000 Hz_10ms_log_rms_mae_db": 1.2477713549665617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09695199127134971, "300–1000 Hz_10ms_envelope_correlation": 0.9938799720362791, "300–1000 Hz_10ms_modulation_mae_db": 0.7215178812070887, "300–1000 Hz_40ms_log_rms_mae_db": 0.827370488429993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060814998851206954, "300–1000 Hz_40ms_envelope_correlation": 0.9975539619382655, "300–1000 Hz_40ms_modulation_mae_db": 0.8027053022049202, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3062244444404956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10363962925181275, "1000–3000 Hz_10ms_envelope_correlation": 0.9936939669599567, "1000–3000 Hz_10ms_modulation_mae_db": 0.45924289732873347, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7894071018001106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050916392867333016, "1000–3000 Hz_40ms_envelope_correlation": 0.9987583149555733, "1000–3000 Hz_40ms_modulation_mae_db": 0.6086735567315175, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5661941022871284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1444441652939203, "3000–6000 Hz_10ms_envelope_correlation": 0.9859657947157494, "3000–6000 Hz_10ms_modulation_mae_db": 1.1107008885392309, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9459877875443379, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08871078475679312, "3000–6000 Hz_40ms_envelope_correlation": 0.9950291686155027, "3000–6000 Hz_40ms_modulation_mae_db": 1.2408424701437124, "mr_stft_db": 5.268214063095396, "band_mae_db": 0.1742422705302248, "rms_gain_error_db": -0.15563753246242443, "output_peak": 0.7151583906653085}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1412611371582888, "全帯域_10ms_rms_envelope_nrmse": 0.07439775643588105, "全帯域_10ms_envelope_correlation": 0.9960956445294876, "全帯域_10ms_modulation_mae_db": 0.20712782379023528, "全帯域_40ms_log_rms_mae_db": 0.6966716033225877, "全帯域_40ms_rms_envelope_nrmse": 0.04255167979369427, "全帯域_40ms_envelope_correlation": 0.9987154306042494, "全帯域_40ms_modulation_mae_db": 0.2475109067187232, "300–1000 Hz_10ms_log_rms_mae_db": 1.1883740381669674, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09242651544192519, "300–1000 Hz_10ms_envelope_correlation": 0.9944323999827757, "300–1000 Hz_10ms_modulation_mae_db": 0.12761927452357943, "300–1000 Hz_40ms_log_rms_mae_db": 0.7758724571763931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04897959480078466, "300–1000 Hz_40ms_envelope_correlation": 0.9984483037000995, "300–1000 Hz_40ms_modulation_mae_db": 0.05108791860560855, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2806259185708924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09413738802882315, "1000–3000 Hz_10ms_envelope_correlation": 0.9946299675104924, "1000–3000 Hz_10ms_modulation_mae_db": 0.7061906587867531, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7857568256221662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05586799223972507, "1000–3000 Hz_40ms_envelope_correlation": 0.9981152842867828, "1000–3000 Hz_40ms_modulation_mae_db": 0.7107869518119193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5960651628279008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1326259399557953, "3000–6000 Hz_10ms_envelope_correlation": 0.9882724641249702, "3000–6000 Hz_10ms_modulation_mae_db": 0.8529078209529865, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9449924839744074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07409312026973319, "3000–6000 Hz_40ms_envelope_correlation": 0.9967919822132318, "3000–6000 Hz_40ms_modulation_mae_db": 1.2471992493257806, "mr_stft_db": 5.272047542135623, "band_mae_db": 0.13782054309525732, "rms_gain_error_db": -0.10179801101169483, "output_peak": 0.6995319137368297}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0878726751990724, "全帯域_10ms_rms_envelope_nrmse": 0.06890991246647282, "全帯域_10ms_envelope_correlation": 0.9966090098032861, "全帯域_10ms_modulation_mae_db": 0.5496985236500844, "全帯域_40ms_log_rms_mae_db": 0.6513248644000125, "全帯域_40ms_rms_envelope_nrmse": 0.034213852381993315, "全帯域_40ms_envelope_correlation": 0.9991387708103316, "全帯域_40ms_modulation_mae_db": 0.4788996539014711, "300–1000 Hz_10ms_log_rms_mae_db": 1.2271967156517316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09048051207022903, "300–1000 Hz_10ms_envelope_correlation": 0.9946087406832004, "300–1000 Hz_10ms_modulation_mae_db": 0.5853401191089428, "300–1000 Hz_40ms_log_rms_mae_db": 0.8277194646382939, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049394014825858305, "300–1000 Hz_40ms_envelope_correlation": 0.998321469564362, "300–1000 Hz_40ms_modulation_mae_db": 0.5889033749080231, "1000–3000 Hz_10ms_log_rms_mae_db": 1.32158231161761, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09390088906741387, "1000–3000 Hz_10ms_envelope_correlation": 0.9946582401089863, "1000–3000 Hz_10ms_modulation_mae_db": 0.4294806874785541, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8261223918423799, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04646380635875864, "1000–3000 Hz_40ms_envelope_correlation": 0.9987352342309529, "1000–3000 Hz_40ms_modulation_mae_db": 0.42151803499479956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5479736458537412, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13637878719672183, "3000–6000 Hz_10ms_envelope_correlation": 0.9869852097619796, "3000–6000 Hz_10ms_modulation_mae_db": 0.12515967805578448, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9306711942786728, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08177282576658047, "3000–6000 Hz_40ms_envelope_correlation": 0.9951169556634865, "3000–6000 Hz_40ms_modulation_mae_db": 0.4266145832844716, "mr_stft_db": 5.275980218510315, "band_mae_db": 0.173147432048673, "rms_gain_error_db": -0.0590287065694029, "output_peak": 0.7144432069363774}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1224909370232488, "全帯域_10ms_rms_envelope_nrmse": 0.06914398740840169, "全帯域_10ms_envelope_correlation": 0.9966065541005081, "全帯域_10ms_modulation_mae_db": 0.7200603414472463, "全帯域_40ms_log_rms_mae_db": 0.6926482697709337, "全帯域_40ms_rms_envelope_nrmse": 0.0335241107078934, "全帯域_40ms_envelope_correlation": 0.9992082021272377, "全帯域_40ms_modulation_mae_db": 0.896591567007185, "300–1000 Hz_10ms_log_rms_mae_db": 1.2891608247632198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08707312415878145, "300–1000 Hz_10ms_envelope_correlation": 0.9950403655153309, "300–1000 Hz_10ms_modulation_mae_db": 0.7999815564007592, "300–1000 Hz_40ms_log_rms_mae_db": 0.8564386879588514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04790719926440365, "300–1000 Hz_40ms_envelope_correlation": 0.9984581656324207, "300–1000 Hz_40ms_modulation_mae_db": 0.8784707454134919, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3305410290628097, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09337967952358278, "1000–3000 Hz_10ms_envelope_correlation": 0.9947252849334954, "1000–3000 Hz_10ms_modulation_mae_db": 0.37630189697647437, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7958102359671235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04971074007417572, "1000–3000 Hz_40ms_envelope_correlation": 0.9985428303064451, "1000–3000 Hz_40ms_modulation_mae_db": 0.7265943901098709, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5607164623342247, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13994658842121896, "3000–6000 Hz_10ms_envelope_correlation": 0.987732620815571, "3000–6000 Hz_10ms_modulation_mae_db": 1.417439621621215, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9218153679747098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08716049490127555, "3000–6000 Hz_40ms_envelope_correlation": 0.9962636727814357, "3000–6000 Hz_40ms_modulation_mae_db": 1.6364397172006833, "mr_stft_db": 5.273407825349431, "band_mae_db": 0.2594398749525072, "rms_gain_error_db": -0.08619186376285085, "output_peak": 0.7691862611915767}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1919498751211506, "全帯域_10ms_rms_envelope_nrmse": 0.07469437301059176, "全帯域_10ms_envelope_correlation": 0.9960993764490457, "全帯域_10ms_modulation_mae_db": 0.7875714656891182, "全帯域_40ms_log_rms_mae_db": 0.7416933932973925, "全帯域_40ms_rms_envelope_nrmse": 0.04248039139856949, "全帯域_40ms_envelope_correlation": 0.9987671558739478, "全帯域_40ms_modulation_mae_db": 0.936757740449384, "300–1000 Hz_10ms_log_rms_mae_db": 1.3213159821431253, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10076946121444336, "300–1000 Hz_10ms_envelope_correlation": 0.9933421660167052, "300–1000 Hz_10ms_modulation_mae_db": 0.503639797606447, "300–1000 Hz_40ms_log_rms_mae_db": 0.8697025953586817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05839877548147067, "300–1000 Hz_40ms_envelope_correlation": 0.997714379372925, "300–1000 Hz_40ms_modulation_mae_db": 0.757080310431164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3199114163043182, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08763358346456893, "1000–3000 Hz_10ms_envelope_correlation": 0.995475934539571, "1000–3000 Hz_10ms_modulation_mae_db": 0.899199343936416, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8261643828872994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048476408602889785, "1000–3000 Hz_40ms_envelope_correlation": 0.9987436896933963, "1000–3000 Hz_40ms_modulation_mae_db": 1.1547081854014907, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5298828259275563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13499056929182493, "3000–6000 Hz_10ms_envelope_correlation": 0.988088435618343, "3000–6000 Hz_10ms_modulation_mae_db": 0.9347194277081994, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9247295493696721, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08155777027071026, "3000–6000 Hz_40ms_envelope_correlation": 0.9962444782577232, "3000–6000 Hz_40ms_modulation_mae_db": 1.0739766466704894, "mr_stft_db": 5.265683101747594, "band_mae_db": 0.2266625975118746, "rms_gain_error_db": -0.1256784216581608, "output_peak": 0.7137519922737587}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1390102920943963, "全帯域_10ms_rms_envelope_nrmse": 0.07093205358396032, "全帯域_10ms_envelope_correlation": 0.9963905300450587, "全帯域_10ms_modulation_mae_db": 0.38001072949013226, "全帯域_40ms_log_rms_mae_db": 0.6626328843817341, "全帯域_40ms_rms_envelope_nrmse": 0.03742221133643198, "全帯域_40ms_envelope_correlation": 0.9989333050599987, "全帯域_40ms_modulation_mae_db": 0.44000952081214534, "300–1000 Hz_10ms_log_rms_mae_db": 1.2399335245002785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08532418342622347, "300–1000 Hz_10ms_envelope_correlation": 0.9952417698774724, "300–1000 Hz_10ms_modulation_mae_db": 0.11171603234341548, "300–1000 Hz_40ms_log_rms_mae_db": 0.8949910429610172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0475743320732677, "300–1000 Hz_40ms_envelope_correlation": 0.9984504761293038, "300–1000 Hz_40ms_modulation_mae_db": 0.4929029760030312, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2997071186101903, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10310580310626437, "1000–3000 Hz_10ms_envelope_correlation": 0.9934921410714547, "1000–3000 Hz_10ms_modulation_mae_db": 0.28501941027719807, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8023300578572079, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055425072280509566, "1000–3000 Hz_40ms_envelope_correlation": 0.9980748415932177, "1000–3000 Hz_40ms_modulation_mae_db": 0.26001067053993115, "3000–6000 Hz_10ms_log_rms_mae_db": 1.564626787874887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13867858046438755, "3000–6000 Hz_10ms_envelope_correlation": 0.9866369784250838, "3000–6000 Hz_10ms_modulation_mae_db": 0.6539054669118788, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9256855179877476, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07784071994470389, "3000–6000 Hz_40ms_envelope_correlation": 0.9957796831519425, "3000–6000 Hz_40ms_modulation_mae_db": 0.5389268678823164, "mr_stft_db": 5.264062708025029, "band_mae_db": 0.13063323209760896, "rms_gain_error_db": -0.013742894815530376, "output_peak": 0.7339123409831727}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1279719748854735, "全帯域_10ms_rms_envelope_nrmse": 0.07144888582895023, "全帯域_10ms_envelope_correlation": 0.9963891868429184, "全帯域_10ms_modulation_mae_db": 0.43436702362678486, "全帯域_40ms_log_rms_mae_db": 0.6947283103183423, "全帯域_40ms_rms_envelope_nrmse": 0.036702238795520904, "全帯域_40ms_envelope_correlation": 0.9990617073902469, "全帯域_40ms_modulation_mae_db": 0.5048950315778339, "300–1000 Hz_10ms_log_rms_mae_db": 1.276607322831851, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09366189019275324, "300–1000 Hz_10ms_envelope_correlation": 0.9942224522556798, "300–1000 Hz_10ms_modulation_mae_db": 0.12903578875781774, "300–1000 Hz_40ms_log_rms_mae_db": 0.8608986778524027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05024585900038612, "300–1000 Hz_40ms_envelope_correlation": 0.9982746297946449, "300–1000 Hz_40ms_modulation_mae_db": 0.09639644867736268, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2821214368092346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08792199072917714, "1000–3000 Hz_10ms_envelope_correlation": 0.9954855309488707, "1000–3000 Hz_10ms_modulation_mae_db": 0.2771017256048412, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7980642676496387, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04600652482578346, "1000–3000 Hz_40ms_envelope_correlation": 0.9989498277794563, "1000–3000 Hz_40ms_modulation_mae_db": 0.2568082048124069, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5609198750348472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13752923265278008, "3000–6000 Hz_10ms_envelope_correlation": 0.9871229749567844, "3000–6000 Hz_10ms_modulation_mae_db": 1.1194068293249935, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9479415139619731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07614327099736794, "3000–6000 Hz_40ms_envelope_correlation": 0.9963386404366862, "3000–6000 Hz_40ms_modulation_mae_db": 1.301246565587881, "mr_stft_db": 5.257025316916972, "band_mae_db": 0.18022284184153625, "rms_gain_error_db": -0.10040230374582174, "output_peak": 0.7238265651685942}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1479388904269834, "全帯域_10ms_rms_envelope_nrmse": 0.07204159292667515, "全帯域_10ms_envelope_correlation": 0.9963337005576268, "全帯域_10ms_modulation_mae_db": 0.29875358943528313, "全帯域_40ms_log_rms_mae_db": 0.6784550391242479, "全帯域_40ms_rms_envelope_nrmse": 0.040102371896908795, "全帯域_40ms_envelope_correlation": 0.9988711284412986, "全帯域_40ms_modulation_mae_db": 0.49526453627593864, "300–1000 Hz_10ms_log_rms_mae_db": 1.2795411629920816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09341641560800058, "300–1000 Hz_10ms_envelope_correlation": 0.994284820959057, "300–1000 Hz_10ms_modulation_mae_db": 0.22148747078546682, "300–1000 Hz_40ms_log_rms_mae_db": 0.8703121144574344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05713133770818857, "300–1000 Hz_40ms_envelope_correlation": 0.9978083822454038, "300–1000 Hz_40ms_modulation_mae_db": 0.3388499421045628, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2744066937945595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08236890394400312, "1000–3000 Hz_10ms_envelope_correlation": 0.9959022798777775, "1000–3000 Hz_10ms_modulation_mae_db": 0.2724796456839141, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7759348351645156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.043585053721509316, "1000–3000 Hz_40ms_envelope_correlation": 0.9988862134492231, "1000–3000 Hz_40ms_modulation_mae_db": 0.4825954742178144, "3000–6000 Hz_10ms_log_rms_mae_db": 1.576164496424955, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15076427176591933, "3000–6000 Hz_10ms_envelope_correlation": 0.9846373931668884, "3000–6000 Hz_10ms_modulation_mae_db": 0.6000576340712894, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9494782953731079, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08708832845254269, "3000–6000 Hz_40ms_envelope_correlation": 0.9953343175729537, "3000–6000 Hz_40ms_modulation_mae_db": 1.00250418415374, "mr_stft_db": 5.268976964446587, "band_mae_db": 0.1488608418353702, "rms_gain_error_db": -0.10746015874148174, "output_peak": 0.7939032638595833}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.637359206122923, "spectral_convergence": 0.25629940336410884, "spectral_convergence_status": "ok", "amplitude_floor": 1.6417184973675735e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.139869446703831, "spectral_convergence": 0.26343221649622506, "spectral_convergence_status": "ok", "amplitude_floor": 1.2466848673527415e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.0274135364594335, "spectral_convergence": 0.2691992178490113, "spectral_convergence_status": "ok", "amplitude_floor": 1.127090317770565e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.270965260159295e-05, "output_power": 4.4021582096702956e-05, "delta_db": 0.13139607747882903, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004287949616170093, "output_power": 0.00042874205718938265, "delta_db": -0.0005358626044535924, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015047786428193433, "output_power": 0.001453599047155258, "delta_db": -0.15027989068762712, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0009641149894002944, "output_power": 0.0009329904712533477, "delta_db": -0.14251626761009153, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00041492564533297956, "output_power": 0.00037927361576491075, "delta_db": -0.39017646139930584, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.594979221160373e-05, "output_power": 8.1506064670816e-05, "delta_db": -0.2305490634010418, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1742422705302248, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15563753246242443, "pitch": null, "output_rms_dbfs": -24.78844539876413, "output_peak": 0.7151583906653085, "reference_rms_dbfs": -24.632807866301704, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.268214063095396}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027067935709258154, "log_rms_mae_db": 1.1788971890273028, "rms_envelope_nrmse": 0.07858144308069683, "envelope_correlation": 0.995722004395199, "rms_gain_error_db": -0.15522723583738382, "reference_mean_power": 0.0034432187972516922, "output_mean_power": 0.003322323338193487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.845747823685581e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011374442611865624, "output_power": 0.001095896174035499, "reference_active": true, "delta_db": -0.16160713140143057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011129046974770975, "output_power": 9.942775856078609e-05, "reference_active": true, "delta_db": -0.4895032648381121}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.356136872122755e-05, "output_power": 1.551966449530047e-05, "reference_active": true, "delta_db": -1.8131818734698508}], "active_band_count": 3, "active_band_mae_db": 0.8214307565697978}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026426858408099, "log_rms_mae_db": 0.7254679911637909, "rms_envelope_nrmse": 0.04222155067267396, "envelope_correlation": 0.9988586346783744, "rms_gain_error_db": -0.1546535510806586, "reference_mean_power": 0.0034555022621815964, "output_mean_power": 0.0033346159758945635}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7242545404192642e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010676968681649734, "output_power": 0.001026893730964023, "reference_active": true, "delta_db": -0.16922466264358907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.454467282708388e-05, "output_power": 4.006593250482129e-05, "reference_active": true, "delta_db": -0.46020518389016046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7651139324996653e-06, "output_power": 1.1957754370403346e-06, "reference_active": true, "delta_db": -1.691231148204102}], "active_band_count": 3, "active_band_mae_db": 0.7735536649126171}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001806334260766054, "log_rms_mae_db": 1.2477713549665617, "rms_envelope_nrmse": 0.09695199127134971, "envelope_correlation": 0.9938799720362791, "rms_gain_error_db": -0.12591264235816743, "reference_mean_power": 0.0015370357837470275, "output_mean_power": 0.001493113128821323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.498278279762008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005426794452872498, "output_power": 0.0005455522200847854, "reference_active": true, "delta_db": 0.022929550392824023}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.715116909567221e-05, "output_power": 5.269029124535606e-05, "reference_active": true, "delta_db": -0.3529451969610832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1787274842814488e-05, "output_power": 7.808102762112664e-06, "reference_active": true, "delta_db": -1.7886788962673592}], "active_band_count": 3, "active_band_mae_db": 0.7215178812070887}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017379577463928091, "log_rms_mae_db": 0.827370488429993, "rms_envelope_nrmse": 0.060814998851206954, "envelope_correlation": 0.9975539619382655, "rms_gain_error_db": -0.12470864381878706, "reference_mean_power": 0.0015404251144981946, "output_mean_power": 0.0014968205127464346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.036061947317895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005182860743704318, "output_power": 0.0005192692864746524, "reference_active": true, "delta_db": 0.008230957794307974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.185473992393753e-05, "output_power": 1.9694064745367597e-05, "reference_active": true, "delta_db": -0.45210281305711925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.514034995645172e-07, "output_power": 5.436973514140746e-07, "reference_active": true, "delta_db": -1.9477821357633336}], "active_band_count": 3, "active_band_mae_db": 0.8027053022049202}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018667130520360042, "log_rms_mae_db": 1.3062244444404956, "rms_envelope_nrmse": 0.10363962925181275, "envelope_correlation": 0.9936939669599567, "rms_gain_error_db": -0.22872068802590045, "reference_mean_power": 0.001062265092647352, "output_mean_power": 0.0010077686439642674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.90294470054441e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004039276312698453, "output_power": 0.0003769505633440844, "reference_active": true, "delta_db": -0.3001916605353621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.530941238269661e-05, "output_power": 3.595601239145557e-05, "reference_active": true, "delta_db": -1.004169084293715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.869428114247789e-06, "output_power": 4.7878569272234995e-06, "reference_active": true, "delta_db": -0.0733679471571233}], "active_band_count": 3, "active_band_mae_db": 0.45924289732873347}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017791979118134677, "log_rms_mae_db": 0.7894071018001106, "rms_envelope_nrmse": 0.050916392867333016, "envelope_correlation": 0.9987583149555733, "rms_gain_error_db": -0.22873104412768247, "reference_mean_power": 0.0010675566768630375, "output_mean_power": 0.0010127863436308361}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.672011383574165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003902673215159807, "output_power": 0.00036243003599160484, "reference_active": true, "delta_db": -0.32138005592461216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8376752084991478e-05, "output_power": 1.4550600014479834e-05, "reference_active": true, "delta_db": -1.0138785400456898}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2996790275380256e-07, "output_power": 3.840252929636375e-07, "reference_active": true, "delta_db": -0.4907620742242503}], "active_band_count": 3, "active_band_mae_db": 0.6086735567315175}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.103910936684106e-05, "log_rms_mae_db": 1.5661941022871284, "rms_envelope_nrmse": 0.1444441652939203, "envelope_correlation": 0.9859657947157494, "rms_gain_error_db": -0.3320536276963187, "reference_mean_power": 0.00018921068158653413, "output_mean_power": 0.00017528319766301135}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9505568897447032e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.310547772849957e-05, "output_power": 8.132463873386243e-05, "reference_active": true, "delta_db": -0.5875309012769935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9017487007684174e-05, "output_power": 1.501074737309127e-05, "reference_active": true, "delta_db": -1.0275081227095315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.040878232439125e-06, "output_power": 3.3946878232795923e-06, "reference_active": true, "delta_db": -1.7170636416311678}], "active_band_count": 3, "active_band_mae_db": 1.1107008885392309}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.917454226005249e-05, "log_rms_mae_db": 0.9459877875443379, "rms_envelope_nrmse": 0.08871078475679312, "envelope_correlation": 0.9950291686155027, "rms_gain_error_db": -0.33282421386628824, "reference_mean_power": 0.00019006200677954335, "output_mean_power": 0.00017604061985627823}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8235095603824036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.910190725222932e-05, "output_power": 7.683038423992712e-05, "reference_active": true, "delta_db": -0.6435399517261908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361035171170285e-06, "output_power": 7.572783011451981e-06, "reference_active": true, "delta_db": -0.9206836412979752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6146687278338854e-07, "output_power": 2.1990652717031783e-07, "reference_active": true, "delta_db": -2.1583038174069715}], "active_band_count": 3, "active_band_mae_db": 1.2408424701437124}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8983265475715918, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "79d013df67841c93906d7e53775d48fd6ed961b63e61a9b9694f515142669f8e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9947579182438255, "standard_deviation": 0.0008549091377759878, "minimum": 0.9934921410714547, "maximum": 0.9959022798777775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.301890046151264, "standard_deviation": 0.021230980726295295, "minimum": 1.2744066937945595, "maximum": 1.3305410290628097, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4631270332591106, "standard_deviation": 0.2269672476127259, "minimum": 0.2724796456839141, "maximum": 0.899199343936416, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09326098338945576, "standard_deviation": 0.0074127138419716335, "minimum": 0.08236890394400312, "maximum": 0.10363962925181275, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.998600779536881, "standard_deviation": 0.00033430107291321026, "minimum": 0.9980748415932177, "maximum": 0.9989498277794563, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7999487623488053, "standard_deviation": 0.018078690313878373, "minimum": 0.7759348351645156, "maximum": 0.8261643828872994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5777119335774688, "standard_deviation": 0.29546941188920345, "minimum": 0.2568082048124069, "maximum": 1.1547081854014907, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04955649887133557, "standard_deviation": 0.004393827152038089, "minimum": 0.043585053721509316, "maximum": 0.05586799223972507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9869302339481711, "standard_deviation": 0.0012012059274477138, "minimum": 0.9846373931668884, "maximum": 0.9882724641249702, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.562817919820655, "standard_deviation": 0.019314215576135604, "minimum": 1.5298828259275563, "maximum": 1.5960651628279008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8517871708981972, "standard_deviation": 0.3956930171379877, "minimum": 0.12515967805578448, "maximum": 1.417439621621215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13941976688032104, "standard_deviation": 0.005773892242759375, "minimum": 0.1326259399557953, "maximum": 0.15076427176591933, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9958623623366203, "standard_deviation": 0.0006473690269245689, "minimum": 0.9950291686155027, "maximum": 0.9967919822132318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9364127138080784, "standard_deviation": 0.01175058515947666, "minimum": 0.9218153679747098, "maximum": 0.9494782953731079, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0584687855311343, "standard_deviation": 0.4025644339912707, "minimum": 0.4266145832844716, "maximum": 1.6364397172006833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0817959144199634, "standard_deviation": 0.005497922026295536, "minimum": 0.07409312026973319, "maximum": 0.08871078475679312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943815859158125, "standard_deviation": 0.0006088379512094869, "minimum": 0.9933421660167052, "maximum": 0.9952417698774724, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.258737615751977, "standard_deviation": 0.041427133531969806, "minimum": 1.1883740381669674, "maximum": 1.3213159821431253, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.40004224009168965, "standard_deviation": 0.28561672094950713, "minimum": 0.11171603234341548, "maximum": 0.7999815564007592, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09251301167296325, "standard_deviation": 0.005005992457476017, "minimum": 0.08532418342622347, "maximum": 0.10076946121444336, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981287210471781, "standard_deviation": 0.000373645840212111, "minimum": 0.9975539619382655, "maximum": 0.9984581656324207, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8479131911041333, "standard_deviation": 0.036749689153021445, "minimum": 0.7758724571763931, "maximum": 0.8949910429610172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5007996272935205, "standard_deviation": 0.31606897491781166, "minimum": 0.05108791860560855, "maximum": 0.8784707454134919, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05255576400069583, "standard_deviation": 0.005316238486126355, "minimum": 0.0475743320732677, "maximum": 0.060814998851206954, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17887870423913155, "standard_deviation": 0.04431063018110872, "minimum": 0.13063323209760896, "maximum": 0.2594398749525072, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.268174717528368, "standard_deviation": 0.005992627458850677, "minimum": 5.257025316916972, "maximum": 5.275980218510315, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7329642418519001, "standard_deviation": 0.03219783329732372, "minimum": 0.6995319137368297, "maximum": 0.7939032638595833, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09374248659592095, "standard_deviation": 0.04277902178174051, "minimum": -0.15563753246242443, "maximum": -0.013742894815530376, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962807508403912, "standard_deviation": 0.00029757924039243834, "minimum": 0.995722004395199, "maximum": 0.9966090098032861, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1421741213669896, "standard_deviation": 0.03253143764080846, "minimum": 1.0878726751990724, "maximum": 1.1919498751211506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5248775317123352, "standard_deviation": 0.23205288362595103, "minimum": 0.20712782379023528, "maximum": 0.8214307565697978, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07251875059270373, "standard_deviation": 0.0032364635172154026, "minimum": 0.06890991246647282, "maximum": 0.07858144308069683, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989442918732107, "standard_deviation": 0.00017640152183160873, "minimum": 0.9987154306042494, "maximum": 0.9992082021272377, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6929527944723801, "standard_deviation": 0.030029331234950107, "minimum": 0.6513248644000125, "maximum": 0.7416933932973925, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5966853277069123, "standard_deviation": 0.24375942240424076, "minimum": 0.2475109067187232, "maximum": 0.936757740449384, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03865230087296076, "standard_deviation": 0.003703583301475439, "minimum": 0.0335241107078934, "maximum": 0.04255167979369427, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0005", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.940087080001831, "predicted_mos": 1.940087080001831, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a1911def1e6330b900c6e5a4a54dd9003b62bed21a3c5ae051f6b6a63c231bf5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7006449000909925, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1614750608777422, "全帯域_10ms_rms_envelope_nrmse": 0.07815008902886718, "全帯域_10ms_envelope_correlation": 0.9957829332248608, "全帯域_10ms_modulation_mae_db": 0.8124817694771659, "全帯域_40ms_log_rms_mae_db": 0.7166179371501995, "全帯域_40ms_rms_envelope_nrmse": 0.04311759399648901, "全帯域_40ms_envelope_correlation": 0.9988156926840825, "全帯域_40ms_modulation_mae_db": 0.7954479419145507, "300–1000 Hz_10ms_log_rms_mae_db": 1.2124128613110476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09451056634915303, "300–1000 Hz_10ms_envelope_correlation": 0.9941777780444788, "300–1000 Hz_10ms_modulation_mae_db": 0.7172811919154198, "300–1000 Hz_40ms_log_rms_mae_db": 0.7532586409632905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059867392601775996, "300–1000 Hz_40ms_envelope_correlation": 0.9976175752499175, "300–1000 Hz_40ms_modulation_mae_db": 0.8165909194363198, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2911171816383855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10652802631512498, "1000–3000 Hz_10ms_envelope_correlation": 0.993326683944388, "1000–3000 Hz_10ms_modulation_mae_db": 0.47825510988741904, "1000–3000 Hz_40ms_log_rms_mae_db": 0.766854271724191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0531571260233703, "1000–3000 Hz_40ms_envelope_correlation": 0.998624531857247, "1000–3000 Hz_40ms_modulation_mae_db": 0.6218530038046913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5648737404024404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14895624186120981, "3000–6000 Hz_10ms_envelope_correlation": 0.9852611158959655, "3000–6000 Hz_10ms_modulation_mae_db": 1.1534987025065364, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9607219044913037, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09245013101487018, "3000–6000 Hz_40ms_envelope_correlation": 0.9948643534607867, "3000–6000 Hz_40ms_modulation_mae_db": 1.3872652332173676, "mr_stft_db": 5.1905248292860255, "band_mae_db": 0.18555035802340344, "rms_gain_error_db": -0.16145727257442383, "output_peak": 0.7207001927002573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1408301350554149, "全帯域_10ms_rms_envelope_nrmse": 0.073257494586581, "全帯域_10ms_envelope_correlation": 0.9962466431178165, "全帯域_10ms_modulation_mae_db": 0.22769413220169374, "全帯域_40ms_log_rms_mae_db": 0.7067611909191569, "全帯域_40ms_rms_envelope_nrmse": 0.04338927979100021, "全帯域_40ms_envelope_correlation": 0.9986963555434885, "全帯域_40ms_modulation_mae_db": 0.2478982247338742, "300–1000 Hz_10ms_log_rms_mae_db": 1.1602400394813912, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09111814177774083, "300–1000 Hz_10ms_envelope_correlation": 0.9946159466953396, "300–1000 Hz_10ms_modulation_mae_db": 0.10682104443052282, "300–1000 Hz_40ms_log_rms_mae_db": 0.739561805507472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050338970921577554, "300–1000 Hz_40ms_envelope_correlation": 0.998385804660998, "300–1000 Hz_40ms_modulation_mae_db": 0.033872275309630315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2639249907519228, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09295035935771352, "1000–3000 Hz_10ms_envelope_correlation": 0.9947744646470589, "1000–3000 Hz_10ms_modulation_mae_db": 0.6596300370747884, "1000–3000 Hz_40ms_log_rms_mae_db": 0.766803488068009, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054301004327320254, "1000–3000 Hz_40ms_envelope_correlation": 0.9982352829170442, "1000–3000 Hz_40ms_modulation_mae_db": 0.662078215808307, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5643629163929822, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14075996792983514, "3000–6000 Hz_10ms_envelope_correlation": 0.9870673722458951, "3000–6000 Hz_10ms_modulation_mae_db": 0.8896148633951572, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9422025133911849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08521961479048024, "3000–6000 Hz_40ms_envelope_correlation": 0.9959643088072574, "3000–6000 Hz_40ms_modulation_mae_db": 1.3409314329290094, "mr_stft_db": 5.200819519527099, "band_mae_db": 0.1438217087092854, "rms_gain_error_db": -0.12025626386764819, "output_peak": 0.7079474439373106}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0853671223126655, "全帯域_10ms_rms_envelope_nrmse": 0.06781619721445233, "全帯域_10ms_envelope_correlation": 0.99671229368684, "全帯域_10ms_modulation_mae_db": 0.555197790854288, "全帯域_40ms_log_rms_mae_db": 0.6530480693502017, "全帯域_40ms_rms_envelope_nrmse": 0.033519204860067146, "全帯域_40ms_envelope_correlation": 0.9991670689123965, "全帯域_40ms_modulation_mae_db": 0.4692267748174707, "300–1000 Hz_10ms_log_rms_mae_db": 1.2090967360971128, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09043291007246415, "300–1000 Hz_10ms_envelope_correlation": 0.9946131677995679, "300–1000 Hz_10ms_modulation_mae_db": 0.5915811194962765, "300–1000 Hz_40ms_log_rms_mae_db": 0.8175848361154834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04819210678853972, "300–1000 Hz_40ms_envelope_correlation": 0.9983970379652934, "300–1000 Hz_40ms_modulation_mae_db": 0.6150101181502056, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2936921811545135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09297890457311868, "1000–3000 Hz_10ms_envelope_correlation": 0.9947397326535501, "1000–3000 Hz_10ms_modulation_mae_db": 0.40546719036149703, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7874476945435724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04639353017814076, "1000–3000 Hz_40ms_envelope_correlation": 0.9987026858545561, "1000–3000 Hz_40ms_modulation_mae_db": 0.45553690056778406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5099749546869121, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1344618382101235, "3000–6000 Hz_10ms_envelope_correlation": 0.9875047091052916, "3000–6000 Hz_10ms_modulation_mae_db": 0.36014600717033635, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9079226387654377, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0822623858468848, "3000–6000 Hz_40ms_envelope_correlation": 0.9952955155476942, "3000–6000 Hz_40ms_modulation_mae_db": 0.5859718637224004, "mr_stft_db": 5.200702352989282, "band_mae_db": 0.1312987002867735, "rms_gain_error_db": -0.0522784179884123, "output_peak": 0.710431808304512}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.130473985594762, "全帯域_10ms_rms_envelope_nrmse": 0.06909092871153444, "全帯域_10ms_envelope_correlation": 0.9966139093859164, "全帯域_10ms_modulation_mae_db": 0.7077186840490581, "全帯域_40ms_log_rms_mae_db": 0.7006839307326147, "全帯域_40ms_rms_envelope_nrmse": 0.03235422424885786, "全帯域_40ms_envelope_correlation": 0.9992668766693954, "全帯域_40ms_modulation_mae_db": 0.8826018803910597, "300–1000 Hz_10ms_log_rms_mae_db": 1.1863728118696868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08553373704843437, "300–1000 Hz_10ms_envelope_correlation": 0.9952258354597769, "300–1000 Hz_10ms_modulation_mae_db": 0.8000674729628626, "300–1000 Hz_40ms_log_rms_mae_db": 0.7660053238673893, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0468617774886291, "300–1000 Hz_40ms_envelope_correlation": 0.9985368618073521, "300–1000 Hz_40ms_modulation_mae_db": 0.914603799630962, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3044719499413415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09661923024547038, "1000–3000 Hz_10ms_envelope_correlation": 0.9943344147449602, "1000–3000 Hz_10ms_modulation_mae_db": 0.2832115348012484, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7708561340732699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049343669970387374, "1000–3000 Hz_40ms_envelope_correlation": 0.9985525016222585, "1000–3000 Hz_40ms_modulation_mae_db": 0.6793907931366038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.540480919898074, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14524457540278443, "3000–6000 Hz_10ms_envelope_correlation": 0.9864691727246918, "3000–6000 Hz_10ms_modulation_mae_db": 1.5248145256531327, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9245376203097319, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08794077890966526, "3000–6000 Hz_40ms_envelope_correlation": 0.9960082505784171, "3000–6000 Hz_40ms_modulation_mae_db": 1.7466019537798576, "mr_stft_db": 5.208679285729986, "band_mae_db": 0.22940406156681029, "rms_gain_error_db": -0.08484823116510583, "output_peak": 0.7630855843549831}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.170459267523478, "全帯域_10ms_rms_envelope_nrmse": 0.07461795239551235, "全帯域_10ms_envelope_correlation": 0.9961119751727524, "全帯域_10ms_modulation_mae_db": 0.7807710357419738, "全帯域_40ms_log_rms_mae_db": 0.7282943057293892, "全帯域_40ms_rms_envelope_nrmse": 0.04254978030100412, "全帯域_40ms_envelope_correlation": 0.9987681941270508, "全帯域_40ms_modulation_mae_db": 0.8835777359113924, "300–1000 Hz_10ms_log_rms_mae_db": 1.2657399780010896, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10102905371256043, "300–1000 Hz_10ms_envelope_correlation": 0.9933070575791465, "300–1000 Hz_10ms_modulation_mae_db": 0.5771240301319532, "300–1000 Hz_40ms_log_rms_mae_db": 0.7932261110838257, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05858555463706953, "300–1000 Hz_40ms_envelope_correlation": 0.9976994311177777, "300–1000 Hz_40ms_modulation_mae_db": 0.8453290406811499, "1000–3000 Hz_10ms_log_rms_mae_db": 1.300783210749777, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08953743890691408, "1000–3000 Hz_10ms_envelope_correlation": 0.9952511941492814, "1000–3000 Hz_10ms_modulation_mae_db": 0.8851128095284903, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7663904585364862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04828321300101186, "1000–3000 Hz_40ms_envelope_correlation": 0.9987394163559076, "1000–3000 Hz_40ms_modulation_mae_db": 1.1222832176591186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5166917390790562, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14040854575825898, "3000–6000 Hz_10ms_envelope_correlation": 0.987177756310651, "3000–6000 Hz_10ms_modulation_mae_db": 0.8782331217269904, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9326631911963926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08862200692077846, "3000–6000 Hz_40ms_envelope_correlation": 0.9955488032667635, "3000–6000 Hz_40ms_modulation_mae_db": 1.0142269680525597, "mr_stft_db": 5.19884522936181, "band_mae_db": 0.21923535136350844, "rms_gain_error_db": -0.12837508660975339, "output_peak": 0.712943881497001}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.16257707341886, "全帯域_10ms_rms_envelope_nrmse": 0.07193252210999446, "全帯域_10ms_envelope_correlation": 0.9962920537252726, "全帯域_10ms_modulation_mae_db": 0.4128592872877508, "全帯域_40ms_log_rms_mae_db": 0.6837848733713188, "全帯域_40ms_rms_envelope_nrmse": 0.038368453374988544, "全帯域_40ms_envelope_correlation": 0.9988788894180946, "全帯域_40ms_modulation_mae_db": 0.47618361707824847, "300–1000 Hz_10ms_log_rms_mae_db": 1.2362387659741294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0901215008382014, "300–1000 Hz_10ms_envelope_correlation": 0.9947163517386706, "300–1000 Hz_10ms_modulation_mae_db": 0.15255762089081729, "300–1000 Hz_40ms_log_rms_mae_db": 0.8579887869963434, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05186796715951764, "300–1000 Hz_40ms_envelope_correlation": 0.9981757929921481, "300–1000 Hz_40ms_modulation_mae_db": 0.5416294843968256, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2862692061734184, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10158989263999392, "1000–3000 Hz_10ms_envelope_correlation": 0.9936820275359249, "1000–3000 Hz_10ms_modulation_mae_db": 0.2023281867810304, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7799659224429588, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05313927895376132, "1000–3000 Hz_40ms_envelope_correlation": 0.9982300112762736, "1000–3000 Hz_40ms_modulation_mae_db": 0.2446442869268631, "3000–6000 Hz_10ms_log_rms_mae_db": 1.580333300768086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14207337836879086, "3000–6000 Hz_10ms_envelope_correlation": 0.9860621967478416, "3000–6000 Hz_10ms_modulation_mae_db": 0.8097414736569513, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9960540671622097, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07886207428963041, "3000–6000 Hz_40ms_envelope_correlation": 0.9958344632137821, "3000–6000 Hz_40ms_modulation_mae_db": 0.8154476675885073, "mr_stft_db": 5.197359327542653, "band_mae_db": 0.14418440169511607, "rms_gain_error_db": -0.0034629415449024072, "output_peak": 0.7355269635568726}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1185356352566316, "全帯域_10ms_rms_envelope_nrmse": 0.07282346218858808, "全帯域_10ms_envelope_correlation": 0.9962532800467693, "全帯域_10ms_modulation_mae_db": 0.4184506115436277, "全帯域_40ms_log_rms_mae_db": 0.6922262692170223, "全帯域_40ms_rms_envelope_nrmse": 0.03778936691229066, "全帯域_40ms_envelope_correlation": 0.9990114124958932, "全帯域_40ms_modulation_mae_db": 0.456433971346331, "300–1000 Hz_10ms_log_rms_mae_db": 1.2456066335878497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09701071726691503, "300–1000 Hz_10ms_envelope_correlation": 0.9937969627270596, "300–1000 Hz_10ms_modulation_mae_db": 0.11892012175499084, "300–1000 Hz_40ms_log_rms_mae_db": 0.7971576304979551, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05232873160650988, "300–1000 Hz_40ms_envelope_correlation": 0.9981229722799351, "300–1000 Hz_40ms_modulation_mae_db": 0.13502266082189238, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2596864001451267, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08893428065497758, "1000–3000 Hz_10ms_envelope_correlation": 0.9954152954219032, "1000–3000 Hz_10ms_modulation_mae_db": 0.24994048884856102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7653547635907006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04667620083851806, "1000–3000 Hz_40ms_envelope_correlation": 0.9989590354626059, "1000–3000 Hz_40ms_modulation_mae_db": 0.20685060882833362, "3000–6000 Hz_10ms_log_rms_mae_db": 1.529356602441766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13888913265089792, "3000–6000 Hz_10ms_envelope_correlation": 0.9870760598470273, "3000–6000 Hz_10ms_modulation_mae_db": 1.2101847896337679, "3000–6000 Hz_40ms_log_rms_mae_db": 0.943255052002615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0779089415661528, "3000–6000 Hz_40ms_envelope_correlation": 0.9964374159675052, "3000–6000 Hz_40ms_modulation_mae_db": 1.426195245869554, "mr_stft_db": 5.1805439887540325, "band_mae_db": 0.16693239032048313, "rms_gain_error_db": -0.10706934352702861, "output_peak": 0.7149116508645214}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.150378605760533, "全帯域_10ms_rms_envelope_nrmse": 0.07241532995210481, "全帯域_10ms_envelope_correlation": 0.9962986397306782, "全帯域_10ms_modulation_mae_db": 0.2953976957273318, "全帯域_40ms_log_rms_mae_db": 0.7001438175640811, "全帯域_40ms_rms_envelope_nrmse": 0.039755349408836634, "全帯域_40ms_envelope_correlation": 0.9988974749888118, "全帯域_40ms_modulation_mae_db": 0.538192433238533, "300–1000 Hz_10ms_log_rms_mae_db": 1.2834944911315758, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09475910796093738, "300–1000 Hz_10ms_envelope_correlation": 0.9941174944357253, "300–1000 Hz_10ms_modulation_mae_db": 0.22138404007126603, "300–1000 Hz_40ms_log_rms_mae_db": 0.8556700283612998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05660937032635202, "300–1000 Hz_40ms_envelope_correlation": 0.9978501991826081, "300–1000 Hz_40ms_modulation_mae_db": 0.32710033062633176, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2666022289879617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08621390832410075, "1000–3000 Hz_10ms_envelope_correlation": 0.9955002811680147, "1000–3000 Hz_10ms_modulation_mae_db": 0.28667736564865226, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7329912401983528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.044529733670572356, "1000–3000 Hz_40ms_envelope_correlation": 0.9988310657941009, "1000–3000 Hz_40ms_modulation_mae_db": 0.4504157413227401, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5554306682571943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1477188149392217, "3000–6000 Hz_10ms_envelope_correlation": 0.9854555451661622, "3000–6000 Hz_10ms_modulation_mae_db": 0.8469025033359591, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9314386510577339, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08795087320714147, "3000–6000 Hz_40ms_envelope_correlation": 0.9954244724610057, "3000–6000 Hz_40ms_modulation_mae_db": 1.2650384777783237, "mr_stft_db": 5.195644514872101, "band_mae_db": 0.1395410302888932, "rms_gain_error_db": -0.11001873820640992, "output_peak": 0.8027778818180268}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.574190215369231, "spectral_convergence": 0.25438047435052413, "spectral_convergence_status": "ok", "amplitude_floor": 1.6417184973675735e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.05564029990094, "spectral_convergence": 0.26212427550932854, "spectral_convergence_status": "ok", "amplitude_floor": 1.2466848673527415e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.941743972587903, "spectral_convergence": 0.2688555326423583, "spectral_convergence_status": "ok", "amplitude_floor": 1.127090317770565e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.270965260159295e-05, "output_power": 4.34166667056538e-05, "delta_db": 0.07130438732617309, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004287949616170093, "output_power": 0.0004268789724393533, "delta_db": -0.01944911220975851, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015047786428193433, "output_power": 0.0014563943006315875, "delta_db": -0.14193647963078068, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0009641149894002944, "output_power": 0.0009323142806050264, "delta_db": -0.14566498504998243, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00041492564533297956, "output_power": 0.0003767631066201395, "delta_db": -0.41901908884017447, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.594979221160373e-05, "output_power": 7.991936445600604e-05, "delta_db": -0.3159280950835514, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18555035802340344, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16145727257442383, "pitch": null, "output_rms_dbfs": -24.794265138876128, "output_peak": 0.7207001927002573, "reference_rms_dbfs": -24.632807866301704, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.1905248292860255}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027067935709258154, "log_rms_mae_db": 1.1614750608777422, "rms_envelope_nrmse": 0.07815008902886718, "envelope_correlation": 0.9957829332248608, "rms_gain_error_db": -0.16103945762388477, "reference_mean_power": 0.0034432187972516922, "output_mean_power": 0.0033178800018896933}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.845747823685581e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011374442611865624, "output_power": 0.001094913519647048, "reference_active": true, "delta_db": -0.1655030554369325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011129046974770975, "output_power": 9.897937843690932e-05, "reference_active": true, "delta_db": -0.5091325330232689}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.356136872122755e-05, "output_power": 1.5700719101133798e-05, "reference_active": true, "delta_db": -1.7628097199712964}], "active_band_count": 3, "active_band_mae_db": 0.8124817694771659}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026426858408099, "log_rms_mae_db": 0.7166179371501995, "rms_envelope_nrmse": 0.04311759399648901, "envelope_correlation": 0.9988156926840825, "rms_gain_error_db": -0.16052831597079703, "reference_mean_power": 0.0034555022621815964, "output_mean_power": 0.0033301082416845986}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7242545404192642e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010676968681649734, "output_power": 0.0010258833346486072, "reference_active": true, "delta_db": -0.1734999401406739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.454467282708388e-05, "output_power": 3.9901181474295345e-05, "reference_active": true, "delta_db": -0.47810018125065645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7651139324996653e-06, "output_power": 1.183854617812074e-06, "reference_active": true, "delta_db": -1.7347437043523217}], "active_band_count": 3, "active_band_mae_db": 0.7954479419145507}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001806334260766054, "log_rms_mae_db": 1.2124128613110476, "rms_envelope_nrmse": 0.09451056634915303, "envelope_correlation": 0.9941777780444788, "rms_gain_error_db": -0.11902116522344341, "reference_mean_power": 0.0015370357837470275, "output_mean_power": 0.0014954843132910468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.498278279762008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005426794452872498, "output_power": 0.000547389004496396, "reference_active": true, "delta_db": 0.03752697004313269}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.715116909567221e-05, "output_power": 5.182464291173026e-05, "reference_active": true, "delta_db": -0.4248880048727085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1787274842814488e-05, "output_power": 7.988597681936256e-06, "reference_active": true, "delta_db": -1.6894286008304185}], "active_band_count": 3, "active_band_mae_db": 0.7172811919154198}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017379577463928091, "log_rms_mae_db": 0.7532586409632905, "rms_envelope_nrmse": 0.059867392601775996, "envelope_correlation": 0.9976175752499175, "rms_gain_error_db": -0.11803413295340451, "reference_mean_power": 0.0015404251144981946, "output_mean_power": 0.0014991226893053733}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.036061947317895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005182860743704318, "output_power": 0.0005208003985616694, "reference_active": true, "delta_db": 0.02101767815517901}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.185473992393753e-05, "output_power": 1.9410910505639813e-05, "reference_active": true, "delta_db": -0.514997353855057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.514034995645172e-07, "output_power": 5.479736284548909e-07, "reference_active": true, "delta_db": -1.9137577262987235}], "active_band_count": 3, "active_band_mae_db": 0.8165909194363198}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018667130520360042, "log_rms_mae_db": 1.2911171816383855, "rms_envelope_nrmse": 0.10652802631512498, "envelope_correlation": 0.993326683944388, "rms_gain_error_db": -0.23378664280526432, "reference_mean_power": 0.001062265092647352, "output_mean_power": 0.0010065937881626715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.90294470054441e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004039276312698453, "output_power": 0.00037752003069574434, "reference_active": true, "delta_db": -0.2936356309864336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.530941238269661e-05, "output_power": 3.505345327323567e-05, "reference_active": true, "delta_db": -1.1145762114336175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.869428114247789e-06, "output_power": 4.899291810729761e-06, "reference_active": true, "delta_db": 0.02655348724220591}], "active_band_count": 3, "active_band_mae_db": 0.47825510988741904}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017791979118134677, "log_rms_mae_db": 0.766854271724191, "rms_envelope_nrmse": 0.0531571260233703, "envelope_correlation": 0.998624531857247, "rms_gain_error_db": -0.23379432438817027, "reference_mean_power": 0.0010675566768630375, "output_mean_power": 0.001011606261177398}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.672011383574165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003902673215159807, "output_power": 0.000362726973151724, "reference_active": true, "delta_db": -0.3178233592177365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8376752084991478e-05, "output_power": 1.410804555515651e-05, "reference_active": true, "delta_db": -1.1480190309709102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2996790275380256e-07, "output_power": 3.9216097570219055e-07, "reference_active": true, "delta_db": -0.3997166212254275}], "active_band_count": 3, "active_band_mae_db": 0.6218530038046913}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.103910936684106e-05, "log_rms_mae_db": 1.5648737404024404, "rms_envelope_nrmse": 0.14895624186120981, "envelope_correlation": 0.9852611158959655, "rms_gain_error_db": -0.3954087452959148, "reference_mean_power": 0.00018921068158653413, "output_mean_power": 0.0001727447175082557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9505568897447032e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.310547772849957e-05, "output_power": 8.056925116694886e-05, "reference_active": true, "delta_db": -0.62805905622826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9017487007684174e-05, "output_power": 1.4757271318917606e-05, "reference_active": true, "delta_db": -1.1014706617974115}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.040878232439125e-06, "output_power": 3.383838035940687e-06, "reference_active": true, "delta_db": -1.730966389493938}], "active_band_count": 3, "active_band_mae_db": 1.1534987025065364}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.917454226005249e-05, "log_rms_mae_db": 0.9607219044913037, "rms_envelope_nrmse": 0.09245013101487018, "envelope_correlation": 0.9948643534607867, "rms_gain_error_db": -0.3955095081327365, "reference_mean_power": 0.00019006200677954335, "output_mean_power": 0.0001735179306328967}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8235095603824036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.910190725222932e-05, "output_power": 7.613383792478414e-05, "reference_active": true, "delta_db": -0.6830927710328167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361035171170285e-06, "output_power": 7.339235468309674e-06, "reference_active": true, "delta_db": -1.0567305536483482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6146687278338854e-07, "output_power": 2.0695277984550313e-07, "reference_active": true, "delta_db": -2.421972374970938}], "active_band_count": 3, "active_band_mae_db": 1.3872652332173676}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9025355838058617, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1791e1accb9f2fbaea54bcde4e217de193cc1750b44ca13fd303a0c664229c1a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9946280117831352, "standard_deviation": 0.0007997763280243579, "minimum": 0.993326683944388, "maximum": 0.9955002811680147, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.283318418692806, "standard_deviation": 0.01749955913693441, "minimum": 1.2596864001451267, "maximum": 1.3044719499413415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4313278403664609, "standard_deviation": 0.2357716951223013, "minimum": 0.2023281867810304, "maximum": 0.8851128095284903, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09441900512717674, "standard_deviation": 0.00684895536634587, "minimum": 0.08621390832410075, "maximum": 0.10652802631512498, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9986093163924992, "standard_deviation": 0.0002630461454550871, "minimum": 0.9982300112762736, "maximum": 0.9989590354626059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7670829966471926, "standard_deviation": 0.015859976400368327, "minimum": 0.7329912401983528, "maximum": 0.7874476945435724, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5553815960068051, "standard_deviation": 0.2907076745930793, "minimum": 0.20685060882833362, "maximum": 1.1222832176591186, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04947796962038528, "standard_deviation": 0.003654688484419862, "minimum": 0.044529733670572356, "maximum": 0.054301004327320254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9865092410054408, "standard_deviation": 0.0008400857760507683, "minimum": 0.9852611158959655, "maximum": 0.9875047091052916, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5451881052408138, "standard_deviation": 0.025128513716824858, "minimum": 1.5099749546869121, "maximum": 1.580333300768086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9591419983848539, "standard_deviation": 0.34348572582775744, "minimum": 0.36014600717033635, "maximum": 1.5248145256531327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1423140618901403, "standard_deviation": 0.00480192677215573, "minimum": 0.1344618382101235, "maximum": 0.14895624186120981, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9956721979129015, "standard_deviation": 0.00049007947880875, "minimum": 0.9948643534607867, "maximum": 0.9964374159675052, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9423494547970761, "standard_deviation": 0.026559664492805675, "minimum": 0.9079226387654377, "maximum": 0.9960540671622097, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1977098553671974, "standard_deviation": 0.371959797152507, "minimum": 0.5859718637224004, "maximum": 1.7466019537798576, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08515210081820046, "standard_deviation": 0.005089122095602722, "minimum": 0.0779089415661528, "maximum": 0.09245013101487018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943213243099707, "standard_deviation": 0.0005990476108900002, "minimum": 0.9933070575791465, "maximum": 0.9952258354597769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2249002896817354, "standard_deviation": 0.04093457669414528, "minimum": 1.1602400394813912, "maximum": 1.2834944911315758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4107170802067637, "standard_deviation": 0.2893267872300505, "minimum": 0.10682104443052282, "maximum": 0.8000674729628626, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09306446687830083, "standard_deviation": 0.004781762394044571, "minimum": 0.08553373704843437, "maximum": 0.10102905371256043, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980982094070038, "standard_deviation": 0.00034280362278913225, "minimum": 0.9976175752499175, "maximum": 0.9985368618073521, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7975566454241325, "standard_deviation": 0.044353210098847344, "minimum": 0.739561805507472, "maximum": 0.8579887869963434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5286448286316647, "standard_deviation": 0.33384175215494266, "minimum": 0.033872275309630315, "maximum": 0.914603799630962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.053081483941246434, "standard_deviation": 0.004794948635775303, "minimum": 0.0468617774886291, "maximum": 0.059867392601775996, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1699960002817842, "standard_deviation": 0.03776141942027449, "minimum": 0.1312987002867735, "maximum": 0.22940406156681029, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.196639881007874, "standard_deviation": 0.00829598737983737, "minimum": 5.1805439887540325, "maximum": 5.208679285729986, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7335406758791856, "standard_deviation": 0.03337196737147814, "minimum": 0.7079474439373106, "maximum": 0.8027778818180268, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09597078693546056, "standard_deviation": 0.04904700592543499, "minimum": -0.16145727257442383, "maximum": -0.0034629415449024072, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962889660113633, "standard_deviation": 0.0002866947818718728, "minimum": 0.9957829332248608, "maximum": 0.99671229368684, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.140012110725011, "standard_deviation": 0.028140687622322778, "minimum": 1.0853671223126655, "maximum": 1.170459267523478, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5263213758603612, "standard_deviation": 0.22276988610565623, "minimum": 0.22769413220169374, "maximum": 0.8124817694771659, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07251299702345432, "standard_deviation": 0.0031842307569686992, "minimum": 0.06781619721445233, "maximum": 0.07815008902886718, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989377456049016, "standard_deviation": 0.00019760812661145157, "minimum": 0.9986963555434885, "maximum": 0.9992668766693954, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.697695049254248, "standard_deviation": 0.022703896296549397, "minimum": 0.6530480693502017, "maximum": 0.7282943057293892, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5936953224289325, "standard_deviation": 0.23268655635916555, "minimum": 0.2478982247338742, "maximum": 0.8835777359113924, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038855406611691776, "standard_deviation": 0.004230746753536085, "minimum": 0.03235422424885786, "maximum": 0.04338927979100021, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0005", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.0655157566070557, "predicted_mos": 2.0655157566070557, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "300d5e3a2940f144ba4b352c50d244f578460a68044ff45fb94472a6c2354818", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.311887599993497, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1108866275636835, "全帯域_10ms_rms_envelope_nrmse": 0.07119596499893027, "全帯域_10ms_envelope_correlation": 0.9964265547912078, "全帯域_10ms_modulation_mae_db": 0.5450118750370877, "全帯域_40ms_log_rms_mae_db": 0.6551422847985579, "全帯域_40ms_rms_envelope_nrmse": 0.040402717450766144, "全帯域_40ms_envelope_correlation": 0.9988466424107784, "全帯域_40ms_modulation_mae_db": 0.6747219300509818, "300–1000 Hz_10ms_log_rms_mae_db": 1.469411937738505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0879470834899093, "300–1000 Hz_10ms_envelope_correlation": 0.9949282144566864, "300–1000 Hz_10ms_modulation_mae_db": 0.20598373050099275, "300–1000 Hz_40ms_log_rms_mae_db": 1.085782780139886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05395519462670488, "300–1000 Hz_40ms_envelope_correlation": 0.9980152665110058, "300–1000 Hz_40ms_modulation_mae_db": 0.30107555209138664, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5025960150601656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1037594107321697, "1000–3000 Hz_10ms_envelope_correlation": 0.9935853769446982, "1000–3000 Hz_10ms_modulation_mae_db": 0.6575506743267806, "1000–3000 Hz_40ms_log_rms_mae_db": 0.968332460488526, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05750934286944169, "1000–3000 Hz_40ms_envelope_correlation": 0.9981748738868628, "1000–3000 Hz_40ms_modulation_mae_db": 0.9122096436790672, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8720090708918418, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15515323447439638, "3000–6000 Hz_10ms_envelope_correlation": 0.9834724329263107, "3000–6000 Hz_10ms_modulation_mae_db": 0.9192413621870426, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1588932784692993, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0856210479153254, "3000–6000 Hz_40ms_envelope_correlation": 0.995224575704315, "3000–6000 Hz_40ms_modulation_mae_db": 0.9954217610464887, "mr_stft_db": 5.549928073254397, "band_mae_db": 0.18311102617102012, "rms_gain_error_db": -0.09843732012596974, "output_peak": 0.7323943686082934}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1315537260439685, "全帯域_10ms_rms_envelope_nrmse": 0.08070506897913754, "全帯域_10ms_envelope_correlation": 0.9953379236599836, "全帯域_10ms_modulation_mae_db": 0.43679285329829703, "全帯域_40ms_log_rms_mae_db": 0.717663866191667, "全帯域_40ms_rms_envelope_nrmse": 0.04492222237189355, "全帯域_40ms_envelope_correlation": 0.9985009523739328, "全帯域_40ms_modulation_mae_db": 0.4444050708720453, "300–1000 Hz_10ms_log_rms_mae_db": 1.5731978820319434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10982568327149773, "300–1000 Hz_10ms_envelope_correlation": 0.9920462853356219, "300–1000 Hz_10ms_modulation_mae_db": 0.3020299989815837, "300–1000 Hz_40ms_log_rms_mae_db": 1.1275437250502365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06500292418925474, "300–1000 Hz_40ms_envelope_correlation": 0.9970992589088781, "300–1000 Hz_40ms_modulation_mae_db": 0.29177077789200334, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4643021788360946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0896496712944638, "1000–3000 Hz_10ms_envelope_correlation": 0.995115752681568, "1000–3000 Hz_10ms_modulation_mae_db": 0.43809802223729405, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9007530611001396, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04546260059275151, "1000–3000 Hz_40ms_envelope_correlation": 0.9987519155736244, "1000–3000 Hz_40ms_modulation_mae_db": 0.21632091778589402, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7972511032522696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14948753091355252, "3000–6000 Hz_10ms_envelope_correlation": 0.9847184007503315, "3000–6000 Hz_10ms_modulation_mae_db": 1.1888332300303814, "3000–6000 Hz_40ms_log_rms_mae_db": 1.091089565894307, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08415502860155893, "3000–6000 Hz_40ms_envelope_correlation": 0.9953492224690014, "3000–6000 Hz_40ms_modulation_mae_db": 1.1572566396190573, "mr_stft_db": 5.5605191562070955, "band_mae_db": 0.1803404984465524, "rms_gain_error_db": -0.07324280726108333, "output_peak": 0.7496963476109715}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0853947684779597, "全帯域_10ms_rms_envelope_nrmse": 0.06970649069930206, "全帯域_10ms_envelope_correlation": 0.996530352144873, "全帯域_10ms_modulation_mae_db": 0.2550693196835566, "全帯域_40ms_log_rms_mae_db": 0.662273893328799, "全帯域_40ms_rms_envelope_nrmse": 0.03755243366569362, "全帯域_40ms_envelope_correlation": 0.9989581679530858, "全帯域_40ms_modulation_mae_db": 0.39754924897304456, "300–1000 Hz_10ms_log_rms_mae_db": 1.4515375851852246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09699121189802448, "300–1000 Hz_10ms_envelope_correlation": 0.9938015439419868, "300–1000 Hz_10ms_modulation_mae_db": 0.35065969574393135, "300–1000 Hz_40ms_log_rms_mae_db": 1.0820504708344802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05824846867219757, "300–1000 Hz_40ms_envelope_correlation": 0.9976496719118715, "300–1000 Hz_40ms_modulation_mae_db": 0.05802025218349708, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4518055664515934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09744781671153291, "1000–3000 Hz_10ms_envelope_correlation": 0.9944772065704373, "1000–3000 Hz_10ms_modulation_mae_db": 0.5277112919986525, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9279900448886341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05276575718959976, "1000–3000 Hz_40ms_envelope_correlation": 0.99862451828914, "1000–3000 Hz_40ms_modulation_mae_db": 0.5760752305867202, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7527271055634193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1433563000369535, "3000–6000 Hz_10ms_envelope_correlation": 0.9857373938754214, "3000–6000 Hz_10ms_modulation_mae_db": 0.9743447203727369, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0507267521214867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07439206853407275, "3000–6000 Hz_40ms_envelope_correlation": 0.9960959585091794, "3000–6000 Hz_40ms_modulation_mae_db": 1.1676809557117591, "mr_stft_db": 5.544352707627415, "band_mae_db": 0.19502850376539857, "rms_gain_error_db": -0.06511032860910612, "output_peak": 0.7525117471896174}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1055065180809749, "全帯域_10ms_rms_envelope_nrmse": 0.07033485926983607, "全帯域_10ms_envelope_correlation": 0.9964758183799326, "全帯域_10ms_modulation_mae_db": 0.44580863081155403, "全帯域_40ms_log_rms_mae_db": 0.6706230010148864, "全帯域_40ms_rms_envelope_nrmse": 0.037439087244061954, "全帯域_40ms_envelope_correlation": 0.9989732037526287, "全帯域_40ms_modulation_mae_db": 0.7102239945342808, "300–1000 Hz_10ms_log_rms_mae_db": 1.4829000329392688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0935396967739816, "300–1000 Hz_10ms_envelope_correlation": 0.9942486060485505, "300–1000 Hz_10ms_modulation_mae_db": 0.1021414727512494, "300–1000 Hz_40ms_log_rms_mae_db": 1.1073718471576919, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05271220268395427, "300–1000 Hz_40ms_envelope_correlation": 0.9980995469160322, "300–1000 Hz_40ms_modulation_mae_db": 0.33569380581924285, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4443703463924538, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08489300380293636, "1000–3000 Hz_10ms_envelope_correlation": 0.9957134522020337, "1000–3000 Hz_10ms_modulation_mae_db": 0.5233603061802956, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9030999533576695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049830011136161166, "1000–3000 Hz_40ms_envelope_correlation": 0.9985883278658532, "1000–3000 Hz_40ms_modulation_mae_db": 0.5606574447111703, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8286299413085025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14778199172508522, "3000–6000 Hz_10ms_envelope_correlation": 0.9847664179250495, "3000–6000 Hz_10ms_modulation_mae_db": 1.0088298330047847, "3000–6000 Hz_40ms_log_rms_mae_db": 1.090253856403994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07492021495282054, "3000–6000 Hz_40ms_envelope_correlation": 0.9960681109987808, "3000–6000 Hz_40ms_modulation_mae_db": 1.1827386827903894, "mr_stft_db": 5.555729291924211, "band_mae_db": 0.15498504028525592, "rms_gain_error_db": -0.06704760367117746, "output_peak": 0.6931900203786314}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1327323431958491, "全帯域_10ms_rms_envelope_nrmse": 0.07729041467966628, "全帯域_10ms_envelope_correlation": 0.9957030405647236, "全帯域_10ms_modulation_mae_db": 0.2162607569843039, "全帯域_40ms_log_rms_mae_db": 0.6738127083837344, "全帯域_40ms_rms_envelope_nrmse": 0.04101434385597979, "全帯域_40ms_envelope_correlation": 0.998718424617945, "全帯域_40ms_modulation_mae_db": 0.2297263289331701, "300–1000 Hz_10ms_log_rms_mae_db": 1.4455613199047468, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10216155936091557, "300–1000 Hz_10ms_envelope_correlation": 0.9931335478647956, "300–1000 Hz_10ms_modulation_mae_db": 0.3959065692340012, "300–1000 Hz_40ms_log_rms_mae_db": 1.0081704848378101, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05576833821801297, "300–1000 Hz_40ms_envelope_correlation": 0.9978523263755629, "300–1000 Hz_40ms_modulation_mae_db": 0.1468916511988352, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4313116659421152, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09540277915440715, "1000–3000 Hz_10ms_envelope_correlation": 0.9944777635591758, "1000–3000 Hz_10ms_modulation_mae_db": 0.34763108936165593, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8763014937201592, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05154165836966261, "1000–3000 Hz_40ms_envelope_correlation": 0.9984113981457122, "1000–3000 Hz_40ms_modulation_mae_db": 0.238405331200368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.829285466298061, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1402962627979729, "3000–6000 Hz_10ms_envelope_correlation": 0.9862294824394202, "3000–6000 Hz_10ms_modulation_mae_db": 0.2828561603842589, "3000–6000 Hz_40ms_log_rms_mae_db": 1.090956168976582, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07666152768845262, "3000–6000 Hz_40ms_envelope_correlation": 0.9955917179380808, "3000–6000 Hz_40ms_modulation_mae_db": 0.21888453104523112, "mr_stft_db": 5.560036775233328, "band_mae_db": 0.07852205250904506, "rms_gain_error_db": -0.0255210909231069, "output_peak": 0.7116534301917617}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1119155088870414, "全帯域_10ms_rms_envelope_nrmse": 0.06916675746334186, "全帯域_10ms_envelope_correlation": 0.9966732516892938, "全帯域_10ms_modulation_mae_db": 0.5117045131643093, "全帯域_40ms_log_rms_mae_db": 0.6981926584829214, "全帯域_40ms_rms_envelope_nrmse": 0.0387059729601643, "全帯域_40ms_envelope_correlation": 0.9989977079161918, "全帯域_40ms_modulation_mae_db": 0.5460617452974826, "300–1000 Hz_10ms_log_rms_mae_db": 1.5081250821495873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09792473982313735, "300–1000 Hz_10ms_envelope_correlation": 0.9940700982077816, "300–1000 Hz_10ms_modulation_mae_db": 0.7269392863089855, "300–1000 Hz_40ms_log_rms_mae_db": 1.0842444487735887, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06308107547981462, "300–1000 Hz_40ms_envelope_correlation": 0.9977496371990965, "300–1000 Hz_40ms_modulation_mae_db": 0.7013094110321032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4959333914331152, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08887940505414856, "1000–3000 Hz_10ms_envelope_correlation": 0.9951690498055611, "1000–3000 Hz_10ms_modulation_mae_db": 0.41955312704833947, "1000–3000 Hz_40ms_log_rms_mae_db": 0.938826357443397, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04592327129976325, "1000–3000 Hz_40ms_envelope_correlation": 0.9986692694811741, "1000–3000 Hz_40ms_modulation_mae_db": 0.4546739102725763, "3000–6000 Hz_10ms_log_rms_mae_db": 1.789351805313752, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13693867018295477, "3000–6000 Hz_10ms_envelope_correlation": 0.9869828068643554, "3000–6000 Hz_10ms_modulation_mae_db": 1.255230288610291, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0652092619609388, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07143214320254342, "3000–6000 Hz_40ms_envelope_correlation": 0.9964240065978136, "3000–6000 Hz_40ms_modulation_mae_db": 1.5716205304689679, "mr_stft_db": 5.549104547979071, "band_mae_db": 0.1652898539761066, "rms_gain_error_db": -0.12386809864104961, "output_peak": 0.706416607660085}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.119609944320315, "全帯域_10ms_rms_envelope_nrmse": 0.0705218258598949, "全帯域_10ms_envelope_correlation": 0.996487169984336, "全帯域_10ms_modulation_mae_db": 0.5766074779660616, "全帯域_40ms_log_rms_mae_db": 0.6945322160282964, "全帯域_40ms_rms_envelope_nrmse": 0.033211830629824975, "全帯域_40ms_envelope_correlation": 0.9992452421575301, "全帯域_40ms_modulation_mae_db": 0.5959442289538268, "300–1000 Hz_10ms_log_rms_mae_db": 1.5348428337094964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09763653028579881, "300–1000 Hz_10ms_envelope_correlation": 0.9938143096753481, "300–1000 Hz_10ms_modulation_mae_db": 0.7313604917447676, "300–1000 Hz_40ms_log_rms_mae_db": 1.0700181621296285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05124597747290046, "300–1000 Hz_40ms_envelope_correlation": 0.9983410742528828, "300–1000 Hz_40ms_modulation_mae_db": 0.7301370034746699, "1000–3000 Hz_10ms_log_rms_mae_db": 1.473970546957956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09248320199620838, "1000–3000 Hz_10ms_envelope_correlation": 0.9947723641349409, "1000–3000 Hz_10ms_modulation_mae_db": 0.4123454667917901, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9136535228079122, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04895646875532101, "1000–3000 Hz_40ms_envelope_correlation": 0.9985024252238993, "1000–3000 Hz_40ms_modulation_mae_db": 0.5116750363423862, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7824438740618531, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15844029587499248, "3000–6000 Hz_10ms_envelope_correlation": 0.9823664386466135, "3000–6000 Hz_10ms_modulation_mae_db": 0.42279879920615143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0744039524978934, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08100369161869331, "3000–6000 Hz_40ms_envelope_correlation": 0.9952249971224472, "3000–6000 Hz_40ms_modulation_mae_db": 0.6267008818963327, "mr_stft_db": 5.550165401499882, "band_mae_db": 0.2179847152280323, "rms_gain_error_db": -0.09409543093527459, "output_peak": 0.7358874063458762}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1184272618725748, "全帯域_10ms_rms_envelope_nrmse": 0.0689948092384057, "全帯域_10ms_envelope_correlation": 0.9965785364911129, "全帯域_10ms_modulation_mae_db": 0.35634000489149903, "全帯域_40ms_log_rms_mae_db": 0.6581032745098981, "全帯域_40ms_rms_envelope_nrmse": 0.03451211888540714, "全帯域_40ms_envelope_correlation": 0.9990942511164407, "全帯域_40ms_modulation_mae_db": 0.3521800410223593, "300–1000 Hz_10ms_log_rms_mae_db": 1.4657538907225385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08666852801000872, "300–1000 Hz_10ms_envelope_correlation": 0.9950799550074174, "300–1000 Hz_10ms_modulation_mae_db": 0.15634887819279483, "300–1000 Hz_40ms_log_rms_mae_db": 1.0948092522482973, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04589509231456031, "300–1000 Hz_40ms_envelope_correlation": 0.9985537640042131, "300–1000 Hz_40ms_modulation_mae_db": 0.18956794470097546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4625748210753162, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09316657870879193, "1000–3000 Hz_10ms_envelope_correlation": 0.9947323565393301, "1000–3000 Hz_10ms_modulation_mae_db": 0.787458759417348, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9104888233933827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04787310193676458, "1000–3000 Hz_40ms_envelope_correlation": 0.9986238311303377, "1000–3000 Hz_40ms_modulation_mae_db": 0.9224204553678564, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7526061757265934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13856882976905502, "3000–6000 Hz_10ms_envelope_correlation": 0.9867371261611708, "3000–6000 Hz_10ms_modulation_mae_db": 0.882485122492433, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0528358713231325, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07945279437969335, "3000–6000 Hz_40ms_envelope_correlation": 0.9956405825796864, "3000–6000 Hz_40ms_modulation_mae_db": 0.9198928595661672, "mr_stft_db": 5.54762231974033, "band_mae_db": 0.1421384240381123, "rms_gain_error_db": -0.02605691352955535, "output_peak": 0.7443122129975247}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.951100306449472, "spectral_convergence": 0.26223591623334713, "spectral_convergence_status": "ok", "amplitude_floor": 1.6417184973675735e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.495539804421572, "spectral_convergence": 0.260480558958015, "spectral_convergence_status": "ok", "amplitude_floor": 1.2466848673527415e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.203144108892148, "spectral_convergence": 0.25910265914956004, "spectral_convergence_status": "ok", "amplitude_floor": 1.127090317770565e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.270965260159295e-05, "output_power": 4.8004600017163243e-05, "delta_db": 0.507568165152837, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004287949616170093, "output_power": 0.00042712053767656296, "delta_db": -0.01699219154188684, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015047786428193433, "output_power": 0.0014855217965468316, "delta_db": -0.05593590246690494, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0009641149894002944, "output_power": 0.0009369136012314726, "delta_db": -0.12429291352187277, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00041492564533297956, "output_power": 0.00038113278812438055, "delta_db": -0.36893966219808777, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.594979221160373e-05, "output_power": 8.54576800862529e-05, "delta_db": -0.024937322144531524, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18311102617102012, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09843732012596974, "pitch": null, "output_rms_dbfs": -24.73124518642767, "output_peak": 0.7323943686082934, "reference_rms_dbfs": -24.632807866301704, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.549928073254397}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027067935709258154, "log_rms_mae_db": 1.1108866275636835, "rms_envelope_nrmse": 0.07119596499893027, "envelope_correlation": 0.9964265547912078, "rms_gain_error_db": -0.0984273275118125, "reference_mean_power": 0.0034432187972516922, "output_mean_power": 0.003366060269757675}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.845747823685581e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011374442611865624, "output_power": 0.0010989441404335833, "reference_active": true, "delta_db": -0.14954506173825574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011129046974770975, "output_power": 0.00010071706449463391, "reference_active": true, "delta_db": -0.43354916224661116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.356136872122755e-05, "output_power": 1.849295767378789e-05, "reference_active": true, "delta_db": -1.0519414011263963}], "active_band_count": 3, "active_band_mae_db": 0.5450118750370877}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026426858408099, "log_rms_mae_db": 0.6551422847985579, "rms_envelope_nrmse": 0.040402717450766144, "envelope_correlation": 0.9988466424107784, "rms_gain_error_db": -0.09886102066293138, "reference_mean_power": 0.0034555022621815964, "output_mean_power": 0.0033777311542566857}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7242545404192642e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010676968681649734, "output_power": 0.0010301036882195917, "reference_active": true, "delta_db": -0.1556702668836305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.454467282708388e-05, "output_power": 4.0290454790513025e-05, "reference_active": true, "delta_db": -0.43593603820797255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7651139324996653e-06, "output_power": 1.2691612227925516e-06, "reference_active": true, "delta_db": -1.4325594850613423}], "active_band_count": 3, "active_band_mae_db": 0.6747219300509818}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001806334260766054, "log_rms_mae_db": 1.469411937738505, "rms_envelope_nrmse": 0.0879470834899093, "envelope_correlation": 0.9949282144566864, "rms_gain_error_db": -0.05865439286450383, "reference_mean_power": 0.0015370357837470275, "output_mean_power": 0.0015164166321386834}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.498278279762008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005426794452872498, "output_power": 0.0005393758393675012, "reference_active": true, "delta_db": -0.026518835027869603}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.715116909567221e-05, "output_power": 5.239107722479407e-05, "reference_active": true, "delta_db": -0.3776779066952829}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1787274842814488e-05, "output_power": 1.1221165429954916e-05, "reference_active": true, "delta_db": -0.2137544497798258}], "active_band_count": 3, "active_band_mae_db": 0.20598373050099275}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017379577463928091, "log_rms_mae_db": 1.085782780139886, "rms_envelope_nrmse": 0.05395519462670488, "envelope_correlation": 0.9980152665110058, "rms_gain_error_db": -0.05963712156320199, "reference_mean_power": 0.0015404251144981946, "output_mean_power": 0.0015194166404225865}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.036061947317895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005182860743704318, "output_power": 0.0005138744079556934, "reference_active": true, "delta_db": -0.03712550674670742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.185473992393753e-05, "output_power": 2.0424036008937087e-05, "reference_active": true, "delta_db": -0.29404075172543653}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.514034995645172e-07, "output_power": 7.463274085094954e-07, "reference_active": true, "delta_db": -0.572060397802016}], "active_band_count": 3, "active_band_mae_db": 0.30107555209138664}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018667130520360042, "log_rms_mae_db": 1.5025960150601656, "rms_envelope_nrmse": 0.1037594107321697, "envelope_correlation": 0.9935853769446982, "rms_gain_error_db": -0.19361989594134185, "reference_mean_power": 0.001062265092647352, "output_mean_power": 0.001015946692216931}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.90294470054441e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004039276312698453, "output_power": 0.0003792128836025925, "reference_active": true, "delta_db": -0.2742047885150754}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.530941238269661e-05, "output_power": 3.720367805723335e-05, "reference_active": true, "delta_db": -0.8560255231189583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.869428114247789e-06, "output_power": 4.010844158092497e-06, "reference_active": true, "delta_db": -0.8424217113463082}], "active_band_count": 3, "active_band_mae_db": 0.6575506743267806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017791979118134677, "log_rms_mae_db": 0.968332460488526, "rms_envelope_nrmse": 0.05750934286944169, "envelope_correlation": 0.9981748738868628, "rms_gain_error_db": -0.19363785673630973, "reference_mean_power": 0.0010675566768630375, "output_mean_power": 0.0010210033227200568}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.672011383574165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003902673215159807, "output_power": 0.0003645582731522536, "reference_active": true, "delta_db": -0.2959522936392899}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8376752084991478e-05, "output_power": 1.5138125910619473e-05, "reference_active": true, "delta_db": -0.8419664330094478}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2996790275380256e-07, "output_power": 2.975534678428245e-07, "reference_active": true, "delta_db": -1.598710204388464}], "active_band_count": 3, "active_band_mae_db": 0.9122096436790672}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.103910936684106e-05, "log_rms_mae_db": 1.8720090708918418, "rms_envelope_nrmse": 0.15515323447439638, "envelope_correlation": 0.9834724329263107, "rms_gain_error_db": -0.28640979265755007, "reference_mean_power": 0.00018921068158653413, "output_mean_power": 0.00017713511796932377}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9505568897447032e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.310547772849957e-05, "output_power": 8.341384344542598e-05, "reference_active": true, "delta_db": -0.47737100263114085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9017487007684174e-05, "output_power": 1.6668368532156323e-05, "reference_active": true, "delta_db": -0.572600342272019}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.040878232439125e-06, "output_power": 3.4019735490471394e-06, "reference_active": true, "delta_db": -1.707752741657968}], "active_band_count": 3, "active_band_mae_db": 0.9192413621870426}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.917454226005249e-05, "log_rms_mae_db": 1.1588932784692993, "rms_envelope_nrmse": 0.0856210479153254, "envelope_correlation": 0.995224575704315, "rms_gain_error_db": -0.286873800013371, "reference_mean_power": 0.00019006200677954335, "output_mean_power": 0.0001779131014245854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8235095603824036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.910190725222932e-05, "output_power": 7.97966164855531e-05, "reference_active": true, "delta_db": -0.4790252343148774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361035171170285e-06, "output_power": 8.432582936004358e-06, "reference_active": true, "delta_db": -0.4536325572359067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6146687278338854e-07, "output_power": 2.2527227900061673e-07, "reference_active": true, "delta_db": -2.053607491588682}], "active_band_count": 3, "active_band_mae_db": 0.9954217610464887}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8892889759219408, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4340192a5d5cc6707f69b834946246b027e093bff22ebf137e74475aeda02426", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9947554153047182, "standard_deviation": 0.0006259929879934578, "minimum": 0.9935853769446982, "maximum": 0.9957134522020337, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4658580665186012, "standard_deviation": 0.024453639883608592, "minimum": 1.4313116659421152, "maximum": 1.5025960150601656, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5142135921702695, "standard_deviation": 0.1455802771552981, "minimum": 0.34763108936165593, "maximum": 0.787458759417348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09321023343183235, "standard_deviation": 0.005796427185602718, "minimum": 0.08489300380293636, "maximum": 0.1037594107321697, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9985433199495755, "standard_deviation": 0.0001809911576517544, "minimum": 0.9981748738868628, "maximum": 0.9987519155736244, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9174307146499776, "standard_deviation": 0.027769061429980688, "minimum": 0.8763014937201592, "maximum": 0.968332460488526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5490547462432549, "standard_deviation": 0.2644854859762906, "minimum": 0.21632091778589402, "maximum": 0.9224204553678564, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0499827765186832, "standard_deviation": 0.003950827035029085, "minimum": 0.04546260059275151, "maximum": 0.05750934286944169, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9851263124485841, "standard_deviation": 0.0016159841658521601, "minimum": 0.9823664386466135, "maximum": 0.9869828068643554, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8005380678020364, "standard_deviation": 0.041005855678854675, "minimum": 1.7526061757265934, "maximum": 1.8720090708918418, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.86682743953601, "standard_deviation": 0.34381170476644324, "minimum": 0.2828561603842589, "maximum": 1.255230288610291, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14625288947187035, "standard_deviation": 0.00783965342820881, "minimum": 0.13693867018295477, "maximum": 0.15844029587499248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995702396489913, "standard_deviation": 0.0004482322042385981, "minimum": 0.995224575704315, "maximum": 0.9964240065978136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0842960884559543, "standard_deviation": 0.0343709816737087, "minimum": 1.0507267521214867, "maximum": 1.1588932784692993, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9800246052680492, "standard_deviation": 0.40808430521409694, "minimum": 0.21888453104523112, "maximum": 1.5716205304689679, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07845481461164504, "standard_deviation": 0.004969523821054235, "minimum": 0.07143214320254342, "maximum": 0.0856210479153254, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938903200672735, "standard_deviation": 0.0009739442721409029, "minimum": 0.9920462853356219, "maximum": 0.9950799550074174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.491416320547664, "standard_deviation": 0.044338863150757975, "minimum": 1.4455613199047468, "maximum": 1.5731978820319434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.37142126543228826, "standard_deviation": 0.24140432799046938, "minimum": 0.1021414727512494, "maximum": 0.7313604917447676, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0965868791141592, "standard_deviation": 0.007480481234491924, "minimum": 0.08666852801000872, "maximum": 0.10982568327149773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997920068259943, "standard_deviation": 0.00044730454991792894, "minimum": 0.9970992589088781, "maximum": 0.9985537640042131, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0824988963964524, "standard_deviation": 0.03483380655323049, "minimum": 1.0081704848378101, "maximum": 1.1275437250502365, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3443082997990892, "standard_deviation": 0.2466622358143691, "minimum": 0.05802025218349708, "maximum": 0.7301370034746699, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05573865920717498, "standard_deviation": 0.006273047676994404, "minimum": 0.04589509231456031, "maximum": 0.06500292418925474, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16467501430244041, "standard_deviation": 0.04204259203174846, "minimum": 0.07852205250904506, "maximum": 0.2179847152280323, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.552182284183216, "standard_deviation": 0.005909979650548702, "minimum": 5.544352707627415, "maximum": 5.5605191562070955, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7282577676228452, "standard_deviation": 0.02192006091821003, "minimum": 0.6931900203786314, "maximum": 0.7525117471896174, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07167244921204038, "standard_deviation": 0.034270303385033965, "minimum": -0.12386809864104961, "maximum": -0.0255210909231069, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962765809631828, "standard_deviation": 0.0004824766719937451, "minimum": 0.9953379236599836, "maximum": 0.9966732516892938, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.114503337305296, "standard_deviation": 0.015177596326359724, "minimum": 1.0853947684779597, "maximum": 1.1327323431958491, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41794942897958365, "standard_deviation": 0.13214842241939123, "minimum": 0.2162607569843039, "maximum": 0.5766074779660616, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07223952389856433, "standard_deviation": 0.00432980574045705, "minimum": 0.0689948092384057, "maximum": 0.08070506897913754, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989168240373167, "standard_deviation": 0.00022928648777672942, "minimum": 0.9985009523739328, "maximum": 0.9992452421575301, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6787929878423451, "standard_deviation": 0.022318691886073824, "minimum": 0.6551422847985579, "maximum": 0.717663866191667, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49385157357964893, "standard_deviation": 0.16657558776008913, "minimum": 0.2297263289331701, "maximum": 0.7102239945342808, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038470090882973935, "standard_deviation": 0.0037247016417632805, "minimum": 0.033211830629824975, "maximum": 0.04492222237189355, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0005", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.929351568222046, "predicted_mos": 1.929351568222046, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9cf6835a786b6d65c50f2130b10900d00f87553f3fce7431e84b51c4d4999be6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.447533499915153, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1442791376121613, "全帯域_10ms_rms_envelope_nrmse": 0.0695918158371838, "全帯域_10ms_envelope_correlation": 0.9965363583974323, "全帯域_10ms_modulation_mae_db": 0.6191985012504576, "全帯域_40ms_log_rms_mae_db": 0.7151751464802271, "全帯域_40ms_rms_envelope_nrmse": 0.036835705836613775, "全帯域_40ms_envelope_correlation": 0.998986739662948, "全帯域_40ms_modulation_mae_db": 0.8602770242410887, "300–1000 Hz_10ms_log_rms_mae_db": 1.1999790368572472, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08521084299997207, "300–1000 Hz_10ms_envelope_correlation": 0.9952227295109252, "300–1000 Hz_10ms_modulation_mae_db": 0.7109774706597517, "300–1000 Hz_40ms_log_rms_mae_db": 0.780568968177253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045788071361024904, "300–1000 Hz_40ms_envelope_correlation": 0.9985482991174887, "300–1000 Hz_40ms_modulation_mae_db": 0.8876679684701635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2716394861494273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09205389521934682, "1000–3000 Hz_10ms_envelope_correlation": 0.9948387011153763, "1000–3000 Hz_10ms_modulation_mae_db": 0.24237862922492448, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8277431044757775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0469780776562018, "1000–3000 Hz_40ms_envelope_correlation": 0.9986522427909763, "1000–3000 Hz_40ms_modulation_mae_db": 0.3469057911513673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4561847581777856, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13591396287634994, "3000–6000 Hz_10ms_envelope_correlation": 0.9876890205169859, "3000–6000 Hz_10ms_modulation_mae_db": 0.4363114547064713, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9132466753876665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08157751888345835, "3000–6000 Hz_40ms_envelope_correlation": 0.9960149064758083, "3000–6000 Hz_40ms_modulation_mae_db": 0.6604036172226686, "mr_stft_db": 5.163633362298478, "band_mae_db": 0.16230708506982675, "rms_gain_error_db": -0.04945198224028224, "output_peak": 0.7770326879607757}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.14614807922999, "全帯域_10ms_rms_envelope_nrmse": 0.07215789087932689, "全帯域_10ms_envelope_correlation": 0.9962954856398264, "全帯域_10ms_modulation_mae_db": 0.5194227159239726, "全帯域_40ms_log_rms_mae_db": 0.6882174713842588, "全帯域_40ms_rms_envelope_nrmse": 0.034447525664530874, "全帯域_40ms_envelope_correlation": 0.9991512984706442, "全帯域_40ms_modulation_mae_db": 0.5690494048502671, "300–1000 Hz_10ms_log_rms_mae_db": 1.2686097313520148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09130328423207082, "300–1000 Hz_10ms_envelope_correlation": 0.9945850812811292, "300–1000 Hz_10ms_modulation_mae_db": 0.1657783145336545, "300–1000 Hz_40ms_log_rms_mae_db": 0.8569665685322765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05059981368283448, "300–1000 Hz_40ms_envelope_correlation": 0.9983570493283406, "300–1000 Hz_40ms_modulation_mae_db": 0.2721576802698555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.227616853741943, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09669054080185807, "1000–3000 Hz_10ms_envelope_correlation": 0.9942794288464026, "1000–3000 Hz_10ms_modulation_mae_db": 0.20526312692802554, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7422873995996844, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04663504928377724, "1000–3000 Hz_40ms_envelope_correlation": 0.9986239317483182, "1000–3000 Hz_40ms_modulation_mae_db": 0.28308437046097484, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4167094874112622, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1321183931179702, "3000–6000 Hz_10ms_envelope_correlation": 0.9880539110338974, "3000–6000 Hz_10ms_modulation_mae_db": 0.9522795834017715, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8782133828243877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07467301984268668, "3000–6000 Hz_40ms_envelope_correlation": 0.9963575259919691, "3000–6000 Hz_40ms_modulation_mae_db": 1.025605402965802, "mr_stft_db": 5.169334552321415, "band_mae_db": 0.13417092699318314, "rms_gain_error_db": -0.0774136875973886, "output_peak": 0.7714204860301348}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1237784665800339, "全帯域_10ms_rms_envelope_nrmse": 0.07653751849206111, "全帯域_10ms_envelope_correlation": 0.9957929433586618, "全帯域_10ms_modulation_mae_db": 0.31058246453446525, "全帯域_40ms_log_rms_mae_db": 0.684957250872295, "全帯域_40ms_rms_envelope_nrmse": 0.044823768207536585, "全帯域_40ms_envelope_correlation": 0.9984815956746246, "全帯域_40ms_modulation_mae_db": 0.17183723441743357, "300–1000 Hz_10ms_log_rms_mae_db": 1.1921940668360835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10758266797105402, "300–1000 Hz_10ms_envelope_correlation": 0.9924133487168944, "300–1000 Hz_10ms_modulation_mae_db": 0.2596266132472212, "300–1000 Hz_40ms_log_rms_mae_db": 0.8036176709786268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0677647035001054, "300–1000 Hz_40ms_envelope_correlation": 0.9968353935439369, "300–1000 Hz_40ms_modulation_mae_db": 0.12481035617988634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2116875787434354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08764821016216685, "1000–3000 Hz_10ms_envelope_correlation": 0.9954752739870815, "1000–3000 Hz_10ms_modulation_mae_db": 0.3853702338838098, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7195948457555352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05200723628499642, "1000–3000 Hz_40ms_envelope_correlation": 0.9985258135763009, "1000–3000 Hz_40ms_modulation_mae_db": 0.18943935050074637, "3000–6000 Hz_10ms_log_rms_mae_db": 1.461991857690604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12303476934134362, "3000–6000 Hz_10ms_envelope_correlation": 0.9893910498717824, "3000–6000 Hz_10ms_modulation_mae_db": 0.5022199719865291, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8784435719260587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06075280720620387, "3000–6000 Hz_40ms_envelope_correlation": 0.9973128337657389, "3000–6000 Hz_40ms_modulation_mae_db": 0.4713527791541506, "mr_stft_db": 5.146710903242002, "band_mae_db": 0.13909166059516573, "rms_gain_error_db": -0.04841357684729081, "output_peak": 0.7816606624317518}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.098223970931001, "全帯域_10ms_rms_envelope_nrmse": 0.06129509827606578, "全帯域_10ms_envelope_correlation": 0.9973530099122948, "全帯域_10ms_modulation_mae_db": 0.506668593319454, "全帯域_40ms_log_rms_mae_db": 0.6177148135853984, "全帯域_40ms_rms_envelope_nrmse": 0.03089277451687953, "全帯域_40ms_envelope_correlation": 0.9993363310096828, "全帯域_40ms_modulation_mae_db": 0.5076314568229738, "300–1000 Hz_10ms_log_rms_mae_db": 1.202642876787769, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08595794609037571, "300–1000 Hz_10ms_envelope_correlation": 0.995145637059703, "300–1000 Hz_10ms_modulation_mae_db": 0.14342885547668735, "300–1000 Hz_40ms_log_rms_mae_db": 0.786396218663912, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04818212952177314, "300–1000 Hz_40ms_envelope_correlation": 0.9984235353569583, "300–1000 Hz_40ms_modulation_mae_db": 0.127991200309336, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1958557241857524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08403208041629442, "1000–3000 Hz_10ms_envelope_correlation": 0.9957635382683707, "1000–3000 Hz_10ms_modulation_mae_db": 0.39165199952943386, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7340846592429388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04390583097341562, "1000–3000 Hz_40ms_envelope_correlation": 0.9989030993062382, "1000–3000 Hz_40ms_modulation_mae_db": 0.6518933916804458, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4054389268141867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13602950917498502, "3000–6000 Hz_10ms_envelope_correlation": 0.9877307497137052, "3000–6000 Hz_10ms_modulation_mae_db": 0.8398163339243672, "3000–6000 Hz_40ms_log_rms_mae_db": 0.851352885946065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08175919805248748, "3000–6000 Hz_40ms_envelope_correlation": 0.9960390895448045, "3000–6000 Hz_40ms_modulation_mae_db": 1.1809786321039129, "mr_stft_db": 5.152638052859135, "band_mae_db": 0.17715864085975466, "rms_gain_error_db": -0.08371506629321121, "output_peak": 0.7184940572284017}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1613725795827778, "全帯域_10ms_rms_envelope_nrmse": 0.07309721628164102, "全帯域_10ms_envelope_correlation": 0.9961748884449823, "全帯域_10ms_modulation_mae_db": 0.661057893130813, "全帯域_40ms_log_rms_mae_db": 0.6717881770208606, "全帯域_40ms_rms_envelope_nrmse": 0.0337043657530664, "全帯域_40ms_envelope_correlation": 0.9991369470001326, "全帯域_40ms_modulation_mae_db": 0.6252862646113987, "300–1000 Hz_10ms_log_rms_mae_db": 1.2279112676899167, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08547643362478037, "300–1000 Hz_10ms_envelope_correlation": 0.9953453152904709, "300–1000 Hz_10ms_modulation_mae_db": 0.38875850104488824, "300–1000 Hz_40ms_log_rms_mae_db": 0.8366635677605507, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043876284731867124, "300–1000 Hz_40ms_envelope_correlation": 0.9987798251342642, "300–1000 Hz_40ms_modulation_mae_db": 0.47816376527417453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.253863183890393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09507483868528671, "1000–3000 Hz_10ms_envelope_correlation": 0.9944762851273393, "1000–3000 Hz_10ms_modulation_mae_db": 0.3733209058794149, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7756825150264587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047470147690572256, "1000–3000 Hz_40ms_envelope_correlation": 0.9986025138737568, "1000–3000 Hz_40ms_modulation_mae_db": 0.5875728088042718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4595725075572845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13127650225106136, "3000–6000 Hz_10ms_envelope_correlation": 0.9887476598058219, "3000–6000 Hz_10ms_modulation_mae_db": 0.9227826223520047, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8836801595123306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07595026877743741, "3000–6000 Hz_40ms_envelope_correlation": 0.9968832349185094, "3000–6000 Hz_40ms_modulation_mae_db": 1.1734472467948425, "mr_stft_db": 5.150160134260199, "band_mae_db": 0.17351099516065868, "rms_gain_error_db": 0.01608410000124091, "output_peak": 0.8057931745955694}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1271865649045716, "全帯域_10ms_rms_envelope_nrmse": 0.07685805821964944, "全帯域_10ms_envelope_correlation": 0.9957980937926135, "全帯域_10ms_modulation_mae_db": 0.63836355973953, "全帯域_40ms_log_rms_mae_db": 0.6683931416619846, "全帯域_40ms_rms_envelope_nrmse": 0.03980169343272491, "全帯域_40ms_envelope_correlation": 0.998857167931149, "全帯域_40ms_modulation_mae_db": 0.7644199182360878, "300–1000 Hz_10ms_log_rms_mae_db": 1.238066659992135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09434746003193319, "300–1000 Hz_10ms_envelope_correlation": 0.9941707726760212, "300–1000 Hz_10ms_modulation_mae_db": 0.6408264548303261, "300–1000 Hz_40ms_log_rms_mae_db": 0.8910836406088056, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050549295858714004, "300–1000 Hz_40ms_envelope_correlation": 0.9982847844632485, "300–1000 Hz_40ms_modulation_mae_db": 0.6730548738279131, "1000–3000 Hz_10ms_log_rms_mae_db": 1.239123118974447, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10787568529640822, "1000–3000 Hz_10ms_envelope_correlation": 0.9929096838320624, "1000–3000 Hz_10ms_modulation_mae_db": 0.2645480869741555, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7538273489977645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05962635018542886, "1000–3000 Hz_40ms_envelope_correlation": 0.9978377562762372, "1000–3000 Hz_40ms_modulation_mae_db": 0.6116151137809398, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5035343042856497, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13827161140453811, "3000–6000 Hz_10ms_envelope_correlation": 0.9869268120242314, "3000–6000 Hz_10ms_modulation_mae_db": 0.7198734887970927, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8980797031691867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08107387364091631, "3000–6000 Hz_40ms_envelope_correlation": 0.9956606172918999, "3000–6000 Hz_40ms_modulation_mae_db": 0.9656676773394114, "mr_stft_db": 5.166699565456445, "band_mae_db": 0.1689122882215155, "rms_gain_error_db": -0.08340613081222624, "output_peak": 0.6883276831878773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.131423871818683, "全帯域_10ms_rms_envelope_nrmse": 0.07580470523324567, "全帯域_10ms_envelope_correlation": 0.9959688822702629, "全帯域_10ms_modulation_mae_db": 0.7192491861516018, "全帯域_40ms_log_rms_mae_db": 0.6617331425643873, "全帯域_40ms_rms_envelope_nrmse": 0.0428363681105723, "全帯域_40ms_envelope_correlation": 0.9987349615198784, "全帯域_40ms_modulation_mae_db": 0.7284430724996288, "300–1000 Hz_10ms_log_rms_mae_db": 1.2547149032445526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10612078070450358, "300–1000 Hz_10ms_envelope_correlation": 0.992778725955785, "300–1000 Hz_10ms_modulation_mae_db": 0.7056728429991631, "300–1000 Hz_40ms_log_rms_mae_db": 0.8342432705371287, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06607605777053237, "300–1000 Hz_40ms_envelope_correlation": 0.997282288725969, "300–1000 Hz_40ms_modulation_mae_db": 0.6030229289488664, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2080191466730192, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09466343895310728, "1000–3000 Hz_10ms_envelope_correlation": 0.9945174437206253, "1000–3000 Hz_10ms_modulation_mae_db": 0.4754351672671319, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7108787614772667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049747680923279136, "1000–3000 Hz_40ms_envelope_correlation": 0.9984342408056067, "1000–3000 Hz_40ms_modulation_mae_db": 0.7820635089173753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.415957750962748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13251052419132608, "3000–6000 Hz_10ms_envelope_correlation": 0.9885798046911466, "3000–6000 Hz_10ms_modulation_mae_db": 0.7776618075218614, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8321546883020814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07910270371279149, "3000–6000 Hz_40ms_envelope_correlation": 0.9965906907957398, "3000–6000 Hz_40ms_modulation_mae_db": 1.445560994822486, "mr_stft_db": 5.159979980526902, "band_mae_db": 0.16253942625525822, "rms_gain_error_db": -0.12227663033719761, "output_peak": 0.7749847079543433}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1402157396807815, "全帯域_10ms_rms_envelope_nrmse": 0.06911826866924031, "全帯域_10ms_envelope_correlation": 0.9966665185425924, "全帯域_10ms_modulation_mae_db": 0.4901340480310707, "全帯域_40ms_log_rms_mae_db": 0.7056577365653665, "全帯域_40ms_rms_envelope_nrmse": 0.04136505409422638, "全帯域_40ms_envelope_correlation": 0.9988207303932176, "全帯域_40ms_modulation_mae_db": 0.7513640096885853, "300–1000 Hz_10ms_log_rms_mae_db": 1.285702083027321, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08872621453276118, "300–1000 Hz_10ms_envelope_correlation": 0.9948435790497392, "300–1000 Hz_10ms_modulation_mae_db": 0.48492125015986004, "300–1000 Hz_40ms_log_rms_mae_db": 0.8586872968192126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05536912833703988, "300–1000 Hz_40ms_envelope_correlation": 0.9979284583940059, "300–1000 Hz_40ms_modulation_mae_db": 0.6352905620176442, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2081064061899593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0882722146951048, "1000–3000 Hz_10ms_envelope_correlation": 0.9954947459998196, "1000–3000 Hz_10ms_modulation_mae_db": 0.7454872662187532, "1000–3000 Hz_40ms_log_rms_mae_db": 0.727449531805961, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05143664109175863, "1000–3000 Hz_40ms_envelope_correlation": 0.9986660543470858, "1000–3000 Hz_40ms_modulation_mae_db": 0.9700622350522657, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4770334884839569, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13327737256049565, "3000–6000 Hz_10ms_envelope_correlation": 0.9876984677219774, "3000–6000 Hz_10ms_modulation_mae_db": 0.709323111520502, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8985979041529593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07615020013267208, "3000–6000 Hz_40ms_envelope_correlation": 0.9959886796245683, "3000–6000 Hz_40ms_modulation_mae_db": 0.8645387592687445, "mr_stft_db": 5.154488802302787, "band_mae_db": 0.18058051400369032, "rms_gain_error_db": -0.11928037504119071, "output_peak": 0.743823964947856}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.489276583014818, "spectral_convergence": 0.2623900556917221, "spectral_convergence_status": "ok", "amplitude_floor": 1.6417184973675735e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.005714473153407, "spectral_convergence": 0.2617852910853482, "spectral_convergence_status": "ok", "amplitude_floor": 1.2466848673527415e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.99590903072721, "spectral_convergence": 0.26378788929852004, "spectral_convergence_status": "ok", "amplitude_floor": 1.127090317770565e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.270965260159295e-05, "output_power": 4.637586130520487e-05, "delta_db": 0.35765949624865895, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004287949616170093, "output_power": 0.00042667363525879763, "delta_db": -0.02153865667042678, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015047786428193433, "output_power": 0.0014972247994867842, "delta_db": -0.021856066879121786, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0009641149894002944, "output_power": 0.0009600643181402206, "delta_db": -0.01828506073262654, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00041492564533297956, "output_power": 0.0003918753159796016, "delta_db": -0.24822369764049795, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.594979221160373e-05, "output_power": 8.009711578530952e-05, "delta_db": -0.30627953224762866, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16230708506982675, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04945198224028224, "pitch": null, "output_rms_dbfs": -24.682259848541985, "output_peak": 0.7770326879607757, "reference_rms_dbfs": -24.632807866301704, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.163633362298478}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027067935709258154, "log_rms_mae_db": 1.1442791376121613, "rms_envelope_nrmse": 0.0695918158371838, "envelope_correlation": 0.9965363583974323, "rms_gain_error_db": -0.04904344607974294, "reference_mean_power": 0.0034432187972516922, "output_mean_power": 0.003404554384064613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.845747823685581e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011374442611865624, "output_power": 0.0011129393198547287, "reference_active": true, "delta_db": -0.0945863768269997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011129046974770975, "output_power": 9.542146399033561e-05, "reference_active": true, "delta_db": -0.6681190016115267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.356136872122755e-05, "output_power": 1.831097641478217e-05, "reference_active": true, "delta_db": -1.0948901253128465}], "active_band_count": 3, "active_band_mae_db": 0.6191985012504576}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026426858408099, "log_rms_mae_db": 0.7151751464802271, "rms_envelope_nrmse": 0.036835705836613775, "envelope_correlation": 0.998986739662948, "rms_gain_error_db": -0.04867052713892167, "reference_mean_power": 0.0034555022621815964, "output_mean_power": 0.0034169933131127203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7242545404192642e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010676968681649734, "output_power": 0.001042894367836409, "reference_active": true, "delta_db": -0.10207646774059366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.454467282708388e-05, "output_power": 3.7806291243920165e-05, "reference_active": true, "delta_db": -0.712316978297582}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7651139324996653e-06, "output_power": 1.1752465373428223e-06, "reference_active": true, "delta_db": -1.7664376266850903}], "active_band_count": 3, "active_band_mae_db": 0.8602770242410887}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001806334260766054, "log_rms_mae_db": 1.1999790368572472, "rms_envelope_nrmse": 0.08521084299997207, "envelope_correlation": 0.9952227295109252, "rms_gain_error_db": -0.00836537237925517, "reference_mean_power": 0.0015370357837470275, "output_mean_power": 0.001534077997782806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.498278279762008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005426794452872498, "output_power": 0.0005345085198082035, "reference_active": true, "delta_db": -0.06588740359574528}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.715116909567221e-05, "output_power": 4.963497315258512e-05, "reference_active": true, "delta_db": -0.6123732752943816}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1787274842814488e-05, "output_power": 8.43230364048132e-06, "reference_active": true, "delta_db": -1.4546717330891281}], "active_band_count": 3, "active_band_mae_db": 0.7109774706597517}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017379577463928091, "log_rms_mae_db": 0.780568968177253, "rms_envelope_nrmse": 0.045788071361024904, "envelope_correlation": 0.9985482991174887, "rms_gain_error_db": -0.007417177857784814, "reference_mean_power": 0.0015404251144981946, "output_mean_power": 0.0015377965165393427}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.036061947317895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005182860743704318, "output_power": 0.0005082384391709524, "reference_active": true, "delta_db": -0.08502031518289398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.185473992393753e-05, "output_power": 1.888529911936717e-05, "reference_active": true, "delta_db": -0.6342177476356997}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.514034995645172e-07, "output_power": 5.442003875025701e-07, "reference_active": true, "delta_db": -1.9437658425918967}], "active_band_count": 3, "active_band_mae_db": 0.8876679684701635}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018667130520360042, "log_rms_mae_db": 1.2716394861494273, "rms_envelope_nrmse": 0.09205389521934682, "envelope_correlation": 0.9948387011153763, "rms_gain_error_db": -0.08150975511669858, "reference_mean_power": 0.001062265092647352, "output_mean_power": 0.0010425140933422719}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.90294470054441e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004039276312698453, "output_power": 0.00039517294546718187, "reference_active": true, "delta_db": -0.09516358703718032}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.530941238269661e-05, "output_power": 3.979473510326366e-05, "reference_active": true, "delta_db": -0.5636281172226041}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.869428114247789e-06, "output_power": 4.9466636026085384e-06, "reference_active": true, "delta_db": 0.06834418341498909}], "active_band_count": 3, "active_band_mae_db": 0.24237862922492448}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017791979118134677, "log_rms_mae_db": 0.8277431044757775, "rms_envelope_nrmse": 0.0469780776562018, "envelope_correlation": 0.9986522427909763, "rms_gain_error_db": -0.08149668945755406, "reference_mean_power": 0.0010675566768630375, "output_mean_power": 0.0010477104416264328}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.672011383574165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003902673215159807, "output_power": 0.00038047938794115644, "reference_active": true, "delta_db": -0.11031053456851225}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8376752084991478e-05, "output_power": 1.629531702644112e-05, "reference_active": true, "delta_db": -0.5220594232978868}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2996790275380256e-07, "output_power": 3.9138240265686773e-07, "reference_active": true, "delta_db": -0.4083474155877029}], "active_band_count": 3, "active_band_mae_db": 0.3469057911513673}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.103910936684106e-05, "log_rms_mae_db": 1.4561847581777856, "rms_envelope_nrmse": 0.13591396287634994, "envelope_correlation": 0.9876890205169859, "rms_gain_error_db": -0.3487416147784361, "reference_mean_power": 0.00018921068158653413, "output_mean_power": 0.00017461095542121904}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9505568897447032e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.310547772849957e-05, "output_power": 8.555194826524838e-05, "reference_active": true, "delta_db": -0.3674532870089297}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9017487007684174e-05, "output_power": 1.6334535907928805e-05, "reference_active": true, "delta_db": -0.6604632828846093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.040878232439125e-06, "output_power": 4.7250284447764945e-06, "reference_active": true, "delta_db": -0.28101779422587486}], "active_band_count": 3, "active_band_mae_db": 0.4363114547064713}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.917454226005249e-05, "log_rms_mae_db": 0.9132466753876665, "rms_envelope_nrmse": 0.08157751888345835, "envelope_correlation": 0.9960149064758083, "rms_gain_error_db": -0.3499501981133593, "reference_mean_power": 0.00019006200677954335, "output_mean_power": 0.00017534778759809133}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8235095603824036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.910190725222932e-05, "output_power": 8.202958495559931e-05, "reference_active": true, "delta_db": -0.35916486175622814}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361035171170285e-06, "output_power": 7.96819833242875e-06, "reference_active": true, "delta_db": -0.6996374161562156}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6146687278338854e-07, "output_power": 2.9229921193232137e-07, "reference_active": true, "delta_db": -0.9224085737555622}], "active_band_count": 3, "active_band_mae_db": 0.6604036172226686}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9093741843073312, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "762c0747929a9f37de938390a1539303f5c55f49c2818975dcd0623f39e04cd5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9947193876121347, "standard_deviation": 0.000913822645088954, "minimum": 0.9929096838320624, "maximum": 0.9957635382683707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2269889373185472, "standard_deviation": 0.026140665518825407, "minimum": 1.1958557241857524, "maximum": 1.2716394861494273, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.38543192698820616, "standard_deviation": 0.17132414799458917, "minimum": 0.20526312692802554, "maximum": 0.7454872662187532, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09328886302869666, "standard_deviation": 0.007305798662189521, "minimum": 0.08403208041629442, "maximum": 0.10787568529640822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9985307065905651, "standard_deviation": 0.00031055787158707106, "minimum": 0.9978377562762372, "maximum": 0.9989030993062382, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7489435207976733, "standard_deviation": 0.03776913306208052, "minimum": 0.7108787614772667, "maximum": 0.8277431044757775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5528295712935484, "standard_deviation": 0.26412315744789605, "minimum": 0.18943935050074637, "maximum": 0.9700622350522657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04972587676117875, "standard_deviation": 0.004813908811906784, "minimum": 0.04390583097341562, "maximum": 0.05962635018542886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9881021844224436, "standard_deviation": 0.000771030818589849, "minimum": 0.9869268120242314, "maximum": 0.9893910498717824, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4495528851729347, "standard_deviation": 0.03406221811111117, "minimum": 1.4054389268141867, "maximum": 1.5035343042856497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.732533546776325, "standard_deviation": 0.18497878629015818, "minimum": 0.4363114547064713, "maximum": 0.9522795834017715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13280408061475873, "standard_deviation": 0.004606237020330938, "minimum": 0.12303476934134362, "maximum": 0.13827161140453811, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9963559473011298, "standard_deviation": 0.0005456121055923499, "minimum": 0.9956606172918999, "maximum": 0.9973128337657389, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.879221121402592, "standard_deviation": 0.0264635462296812, "minimum": 0.8321546883020814, "maximum": 0.9132466753876665, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9734443887090023, "standard_deviation": 0.3095383403351469, "minimum": 0.4713527791541506, "maximum": 1.445560994822486, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0763799487810817, "standard_deviation": 0.006894676541153697, "minimum": 0.06075280720620387, "maximum": 0.08175919805248748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943131486925835, "standard_deviation": 0.001129159595306506, "minimum": 0.9924133487168944, "maximum": 0.9953453152904709, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2337275782233799, "standard_deviation": 0.03431262562861769, "minimum": 1.1921940668360835, "maximum": 1.285702083027321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.43749878786894403, "standard_deviation": 0.23440569553622964, "minimum": 0.14342885547668735, "maximum": 0.7109774706597517, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09309070377343137, "standard_deviation": 0.009059868733263931, "minimum": 0.08521084299997207, "maximum": 0.10758266797105402, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980549542580265, "standard_deviation": 0.0006707249683525055, "minimum": 0.9968353935439369, "maximum": 0.9987798251342642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8310284002597207, "standard_deviation": 0.03850057426117409, "minimum": 0.780568968177253, "maximum": 0.8910836406088056, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.47526991691223, "standard_deviation": 0.27666530754726903, "minimum": 0.12481035617988634, "maximum": 0.8876679684701635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.053525685595486414, "standard_deviation": 0.008965127508113228, "minimum": 0.043876284731867124, "maximum": 0.0677647035001054, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16228394214488162, "standard_deviation": 0.017130975482608116, "minimum": 0.13417092699318314, "maximum": 0.18058051400369032, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.1579556691584205, "standard_deviation": 0.008192157174329867, "minimum": 5.146710903242002, "maximum": 5.169334552321415, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7576921780420887, "standard_deviation": 0.038340580188624775, "minimum": 0.6883276831878773, "maximum": 0.8057931745955694, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07098416864594331, "standard_deviation": 0.04452143923710347, "minimum": -0.12227663033719761, "maximum": 0.01608410000124091, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963232725448332, "standard_deviation": 0.0005249526004288425, "minimum": 0.9957929433586618, "maximum": 0.9973530099122948, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1340785512925002, "standard_deviation": 0.018816952861660657, "minimum": 1.098223970931001, "maximum": 1.1613725795827778, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5580846202601706, "standard_deviation": 0.12915664621286935, "minimum": 0.31058246453446525, "maximum": 0.7192491861516018, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07180757148605175, "standard_deviation": 0.005186477144895007, "minimum": 0.06129509827606578, "maximum": 0.07685805821964944, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989382214577847, "standard_deviation": 0.00027148649500029237, "minimum": 0.9984815956746246, "maximum": 0.9993363310096828, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6767046100168472, "standard_deviation": 0.030038333868077877, "minimum": 0.6177148135853984, "maximum": 0.7151751464802271, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.622288548170933, "standard_deviation": 0.2150299401891491, "minimum": 0.17183723441743357, "maximum": 0.8602770242410887, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038088406952018844, "standard_deviation": 0.004890255813858343, "minimum": 0.03089277451687953, "maximum": 0.044823768207536585, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0005", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.8778631687164307, "predicted_mos": 1.8778631687164307, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "37f6048dfe5522144459e19160e772f28b197a1f10b49dd51c71cb9526659cf4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.693915799725801, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.176101658507673, "全帯域_10ms_rms_envelope_nrmse": 0.07855316573945094, "全帯域_10ms_envelope_correlation": 0.9957252592909405, "全帯域_10ms_modulation_mae_db": 0.8208437286212479, "全帯域_40ms_log_rms_mae_db": 0.7190789252366353, "全帯域_40ms_rms_envelope_nrmse": 0.042208163365580004, "全帯域_40ms_envelope_correlation": 0.9988594725147196, "全帯域_40ms_modulation_mae_db": 0.7738570711226708, "300–1000 Hz_10ms_log_rms_mae_db": 1.249867627819122, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09696702737661263, "300–1000 Hz_10ms_envelope_correlation": 0.9938780596849286, "300–1000 Hz_10ms_modulation_mae_db": 0.7211781325827767, "300–1000 Hz_40ms_log_rms_mae_db": 0.8302989341605974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06082352747220963, "300–1000 Hz_40ms_envelope_correlation": 0.9975531626285824, "300–1000 Hz_40ms_modulation_mae_db": 0.8021481156187346, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3083538055470216, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10364667409446154, "1000–3000 Hz_10ms_envelope_correlation": 0.9936926553332646, "1000–3000 Hz_10ms_modulation_mae_db": 0.45880462355810225, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7908702607930741, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050903969748188574, "1000–3000 Hz_40ms_envelope_correlation": 0.9987585774442606, "1000–3000 Hz_40ms_modulation_mae_db": 0.6082069402506478, "3000–6000 Hz_10ms_log_rms_mae_db": 1.571155524400724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14446511031075465, "3000–6000 Hz_10ms_envelope_correlation": 0.9859597023095643, "3000–6000 Hz_10ms_modulation_mae_db": 1.1098560753330033, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9518281744202466, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08869551242438387, "3000–6000 Hz_40ms_envelope_correlation": 0.9950279460695379, "3000–6000 Hz_40ms_modulation_mae_db": 1.2398050456371348, "mr_stft_db": 5.282439067482076, "band_mae_db": 0.17232411326106242, "rms_gain_error_db": -0.15561232303191067, "output_peak": 0.7136517416988136}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1421297704574747, "全帯域_10ms_rms_envelope_nrmse": 0.0744077071740046, "全帯域_10ms_envelope_correlation": 0.996094587316809, "全帯域_10ms_modulation_mae_db": 0.2072152749448509, "全帯域_40ms_log_rms_mae_db": 0.6976161158482204, "全帯域_40ms_rms_envelope_nrmse": 0.04257381010308572, "全帯域_40ms_envelope_correlation": 0.9987139339388179, "全帯域_40ms_modulation_mae_db": 0.24781115674704443, "300–1000 Hz_10ms_log_rms_mae_db": 1.191164616981724, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09243943640124556, "300–1000 Hz_10ms_envelope_correlation": 0.9944309497661699, "300–1000 Hz_10ms_modulation_mae_db": 0.12798153045173857, "300–1000 Hz_40ms_log_rms_mae_db": 0.7791212811732618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04898746417261758, "300–1000 Hz_40ms_envelope_correlation": 0.99844790900948, "300–1000 Hz_40ms_modulation_mae_db": 0.05147181768407801, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2820665041208488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09416090726317373, "1000–3000 Hz_10ms_envelope_correlation": 0.9946271193194042, "1000–3000 Hz_10ms_modulation_mae_db": 0.706605818303943, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7859829569126723, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05588556033588606, "1000–3000 Hz_40ms_envelope_correlation": 0.9981138015131162, "1000–3000 Hz_40ms_modulation_mae_db": 0.7110478681287136, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6025383896077376, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13263670644622388, "3000–6000 Hz_10ms_envelope_correlation": 0.9882691479576013, "3000–6000 Hz_10ms_modulation_mae_db": 0.8522919479837494, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9498174789646194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07405929085123336, "3000–6000 Hz_40ms_envelope_correlation": 0.9967932730939303, "3000–6000 Hz_40ms_modulation_mae_db": 1.2466602029791387, "mr_stft_db": 5.287265484295001, "band_mae_db": 0.13368591134827265, "rms_gain_error_db": -0.1019729043564703, "output_peak": 0.6996961832163611}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0950228636322434, "全帯域_10ms_rms_envelope_nrmse": 0.06894795490351643, "全帯域_10ms_envelope_correlation": 0.9966052276920583, "全帯域_10ms_modulation_mae_db": 0.5497568122195217, "全帯域_40ms_log_rms_mae_db": 0.661098719212367, "全帯域_40ms_rms_envelope_nrmse": 0.03424176977537754, "全帯域_40ms_envelope_correlation": 0.9991372652971668, "全帯域_40ms_modulation_mae_db": 0.47825964181799435, "300–1000 Hz_10ms_log_rms_mae_db": 1.229216476054962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0905028817281953, "300–1000 Hz_10ms_envelope_correlation": 0.9946061782846116, "300–1000 Hz_10ms_modulation_mae_db": 0.5853753159922084, "300–1000 Hz_40ms_log_rms_mae_db": 0.8291112314645593, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0494074295379538, "300–1000 Hz_40ms_envelope_correlation": 0.9983206103455091, "300–1000 Hz_40ms_modulation_mae_db": 0.5889319336317425, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3238629963811133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.093913678829963, "1000–3000 Hz_10ms_envelope_correlation": 0.9946566199416413, "1000–3000 Hz_10ms_modulation_mae_db": 0.42979742202951554, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8280025518741765, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04646567234054391, "1000–3000 Hz_40ms_envelope_correlation": 0.9987348303572143, "1000–3000 Hz_40ms_modulation_mae_db": 0.4213660709196727, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5554088346843384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13644696176824828, "3000–6000 Hz_10ms_envelope_correlation": 0.986971637190051, "3000–6000 Hz_10ms_modulation_mae_db": 0.12438459035187695, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9386071752771575, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08183140791658763, "3000–6000 Hz_40ms_envelope_correlation": 0.9951080087187429, "3000–6000 Hz_40ms_modulation_mae_db": 0.4262217815647847, "mr_stft_db": 5.289373601450731, "band_mae_db": 0.17140750267322646, "rms_gain_error_db": -0.05901073009601538, "output_peak": 0.7144337287973183}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1222946197809474, "全帯域_10ms_rms_envelope_nrmse": 0.06915902392129689, "全帯域_10ms_envelope_correlation": 0.9966050212988631, "全帯域_10ms_modulation_mae_db": 0.7210020015692095, "全帯域_40ms_log_rms_mae_db": 0.6918346618520717, "全帯域_40ms_rms_envelope_nrmse": 0.03353488824409592, "全帯域_40ms_envelope_correlation": 0.9992076897687218, "全帯域_40ms_modulation_mae_db": 0.8976090583055875, "300–1000 Hz_10ms_log_rms_mae_db": 1.2922376057028155, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08708843332606808, "300–1000 Hz_10ms_envelope_correlation": 0.9950387589073286, "300–1000 Hz_10ms_modulation_mae_db": 0.7997066867991877, "300–1000 Hz_40ms_log_rms_mae_db": 0.8605018180340673, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04791202476579885, "300–1000 Hz_40ms_envelope_correlation": 0.9984578938866593, "300–1000 Hz_40ms_modulation_mae_db": 0.8783054062580872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3325341703839435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09338967424834121, "1000–3000 Hz_10ms_envelope_correlation": 0.9947238402214441, "1000–3000 Hz_10ms_modulation_mae_db": 0.3758244513471574, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7966896983734184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04970246384667569, "1000–3000 Hz_40ms_envelope_correlation": 0.9985428487319646, "1000–3000 Hz_40ms_modulation_mae_db": 0.7262340672505733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5673986535590854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13994070840391584, "3000–6000 Hz_10ms_envelope_correlation": 0.9877307887089919, "3000–6000 Hz_10ms_modulation_mae_db": 1.416897251500731, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9268043559044369, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08711162384112585, "3000–6000 Hz_40ms_envelope_correlation": 0.9962656106551302, "3000–6000 Hz_40ms_modulation_mae_db": 1.6359308330608247, "mr_stft_db": 5.28822488230766, "band_mae_db": 0.25675753785665395, "rms_gain_error_db": -0.08623351974809206, "output_peak": 0.7704320556690583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1924235401191572, "全帯域_10ms_rms_envelope_nrmse": 0.07470108313326963, "全帯域_10ms_envelope_correlation": 0.9960987413802291, "全帯域_10ms_modulation_mae_db": 0.7866543622477161, "全帯域_40ms_log_rms_mae_db": 0.7412018729004884, "全帯域_40ms_rms_envelope_nrmse": 0.042482530950735266, "全帯域_40ms_envelope_correlation": 0.9987670644734347, "全帯域_40ms_modulation_mae_db": 0.9360310327523855, "300–1000 Hz_10ms_log_rms_mae_db": 1.3231007515047737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1007825887508157, "300–1000 Hz_10ms_envelope_correlation": 0.993340537835026, "300–1000 Hz_10ms_modulation_mae_db": 0.5032661765912027, "300–1000 Hz_40ms_log_rms_mae_db": 0.8706927613242932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058405671860564355, "300–1000 Hz_40ms_envelope_correlation": 0.9977139008717275, "300–1000 Hz_40ms_modulation_mae_db": 0.7565736631291893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3217079966470433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08764099893980538, "1000–3000 Hz_10ms_envelope_correlation": 0.9954749596755602, "1000–3000 Hz_10ms_modulation_mae_db": 0.8988346499128611, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8260284532805607, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04846328098680708, "1000–3000 Hz_40ms_envelope_correlation": 0.9987441587569562, "1000–3000 Hz_40ms_modulation_mae_db": 1.1546067915122442, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5359659928887142, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13499893184585096, "3000–6000 Hz_10ms_envelope_correlation": 0.9880835974244243, "3000–6000 Hz_10ms_modulation_mae_db": 0.9341358960339402, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9288019353580077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08153242539772321, "3000–6000 Hz_40ms_envelope_correlation": 0.9962427965103732, "3000–6000 Hz_40ms_modulation_mae_db": 1.073413010760453, "mr_stft_db": 5.281955511247805, "band_mae_db": 0.22493575888184222, "rms_gain_error_db": -0.12564810247844332, "output_peak": 0.7131284654202776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.141399461595611, "全帯域_10ms_rms_envelope_nrmse": 0.07092290479471546, "全帯域_10ms_envelope_correlation": 0.9963914238208142, "全帯域_10ms_modulation_mae_db": 0.3788510623736992, "全帯域_40ms_log_rms_mae_db": 0.6690273254507527, "全帯域_40ms_rms_envelope_nrmse": 0.03742082710605573, "全帯域_40ms_envelope_correlation": 0.9989333188174724, "全帯域_40ms_modulation_mae_db": 0.43760134702157755, "300–1000 Hz_10ms_log_rms_mae_db": 1.2428389820246588, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08535139396544582, "300–1000 Hz_10ms_envelope_correlation": 0.9952390077902737, "300–1000 Hz_10ms_modulation_mae_db": 0.11095640540544993, "300–1000 Hz_40ms_log_rms_mae_db": 0.898114034485174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047597532142645435, "300–1000 Hz_40ms_envelope_correlation": 0.9984491471538022, "300–1000 Hz_40ms_modulation_mae_db": 0.49223841721116934, "1000–3000 Hz_10ms_log_rms_mae_db": 1.300993473353254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10313022654574072, "1000–3000 Hz_10ms_envelope_correlation": 0.9934889952165363, "1000–3000 Hz_10ms_modulation_mae_db": 0.2855139259984771, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8037230232313872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05544104232703255, "1000–3000 Hz_40ms_envelope_correlation": 0.9980734468643465, "1000–3000 Hz_40ms_modulation_mae_db": 0.25925577793073906, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5695317554915107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13872453543798607, "3000–6000 Hz_10ms_envelope_correlation": 0.9866261580534883, "3000–6000 Hz_10ms_modulation_mae_db": 0.6535512827313106, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9310238904792585, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07786961724254467, "3000–6000 Hz_40ms_envelope_correlation": 0.9957725357060725, "3000–6000 Hz_40ms_modulation_mae_db": 0.5384923567675156, "mr_stft_db": 5.278856025411394, "band_mae_db": 0.12812196992499333, "rms_gain_error_db": -0.013711536919779437, "output_peak": 0.7346534011876332}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1295874392287728, "全帯域_10ms_rms_envelope_nrmse": 0.07145102915884288, "全帯域_10ms_envelope_correlation": 0.9963888878626945, "全帯域_10ms_modulation_mae_db": 0.4335952285685884, "全帯域_40ms_log_rms_mae_db": 0.6960037545222612, "全帯域_40ms_rms_envelope_nrmse": 0.03671151303872988, "全帯域_40ms_envelope_correlation": 0.9990611912797212, "全帯域_40ms_modulation_mae_db": 0.5037998050184195, "300–1000 Hz_10ms_log_rms_mae_db": 1.280672760854449, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09367453657203595, "300–1000 Hz_10ms_envelope_correlation": 0.9942209266163603, "300–1000 Hz_10ms_modulation_mae_db": 0.1294045117148873, "300–1000 Hz_40ms_log_rms_mae_db": 0.8655171242011239, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05025048798520576, "300–1000 Hz_40ms_envelope_correlation": 0.998274281075744, "300–1000 Hz_40ms_modulation_mae_db": 0.09642141308253627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.28380156191625, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08793111740605478, "1000–3000 Hz_10ms_envelope_correlation": 0.9954842208469805, "1000–3000 Hz_10ms_modulation_mae_db": 0.27755385641160674, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7973973698291698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04599854273990162, "1000–3000 Hz_40ms_envelope_correlation": 0.9989498445398639, "1000–3000 Hz_40ms_modulation_mae_db": 0.25714314008063593, "3000–6000 Hz_10ms_log_rms_mae_db": 1.567275209310119, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13753826978745978, "3000–6000 Hz_10ms_envelope_correlation": 0.9871190953563393, "3000–6000 Hz_10ms_modulation_mae_db": 1.1190887544129262, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9531194983440073, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07611132264680191, "3000–6000 Hz_40ms_envelope_correlation": 0.9963384742298251, "3000–6000 Hz_40ms_modulation_mae_db": 1.3007720298182295, "mr_stft_db": 5.272538265499972, "band_mae_db": 0.17735212196112024, "rms_gain_error_db": -0.10045696667282894, "output_peak": 0.7237577193205262}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1463900803586644, "全帯域_10ms_rms_envelope_nrmse": 0.07207113186014083, "全帯域_10ms_envelope_correlation": 0.9963305332477834, "全帯域_10ms_modulation_mae_db": 0.3002676427206817, "全帯域_40ms_log_rms_mae_db": 0.6782929357333696, "全帯域_40ms_rms_envelope_nrmse": 0.04013423243426682, "全帯域_40ms_envelope_correlation": 0.9988691362130417, "全帯域_40ms_modulation_mae_db": 0.4969487821106555, "300–1000 Hz_10ms_log_rms_mae_db": 1.282650291759807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09343643689668646, "300–1000 Hz_10ms_envelope_correlation": 0.9942821934257638, "300–1000 Hz_10ms_modulation_mae_db": 0.22180566250112643, "300–1000 Hz_40ms_log_rms_mae_db": 0.8743467736026134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05713989244833748, "300–1000 Hz_40ms_envelope_correlation": 0.9978074530561414, "300–1000 Hz_40ms_modulation_mae_db": 0.3388076144733025, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2762958425402418, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08237829267880988, "1000–3000 Hz_10ms_envelope_correlation": 0.9959009038471234, "1000–3000 Hz_10ms_modulation_mae_db": 0.2721016622150498, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7762288425264944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04358152436247641, "1000–3000 Hz_40ms_envelope_correlation": 0.9988857642367955, "1000–3000 Hz_40ms_modulation_mae_db": 0.4823258127515988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5819025431346871, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1507909963269302, "3000–6000 Hz_10ms_envelope_correlation": 0.9846290742200161, "3000–6000 Hz_10ms_modulation_mae_db": 0.5996531165399911, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9551437976282467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0870828353805302, "3000–6000 Hz_40ms_envelope_correlation": 0.9953309948241684, "3000–6000 Hz_40ms_modulation_mae_db": 1.0019980801222201, "mr_stft_db": 5.28384610651184, "band_mae_db": 0.14583731580420936, "rms_gain_error_db": -0.10744789011143564, "output_peak": 0.794604647039014}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.650962171483877, "spectral_convergence": 0.256168390093675, "spectral_convergence_status": "ok", "amplitude_floor": 1.6417184973675735e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1581345562838745, "spectral_convergence": 0.26340561727634026, "spectral_convergence_status": "ok", "amplitude_floor": 1.2466848673527415e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.038220474678476, "spectral_convergence": 0.2692148537589734, "spectral_convergence_status": "ok", "amplitude_floor": 1.127090317770565e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.270965260159295e-05, "output_power": 4.3913164750681255e-05, "delta_db": 0.12068698223123654, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004287949616170093, "output_power": 0.00042876405294818836, "delta_db": -0.0003130621704100723, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015047786428193433, "output_power": 0.0014536364334059033, "delta_db": -0.15016819252995345, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0009641149894002944, "output_power": 0.000933018002368163, "delta_db": -0.1423881158742894, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00041492564533297956, "output_power": 0.0003793174925554484, "delta_db": -0.3896740709065522, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.594979221160373e-05, "output_power": 8.150296448620755e-05, "delta_db": -0.2307142558539327, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17232411326106242, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15561232303191067, "pitch": null, "output_rms_dbfs": -24.788420189333614, "output_peak": 0.7136517416988136, "reference_rms_dbfs": -24.632807866301704, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.282439067482076}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027067935709258154, "log_rms_mae_db": 1.176101658507673, "rms_envelope_nrmse": 0.07855316573945094, "envelope_correlation": 0.9957252592909405, "rms_gain_error_db": -0.15520208709338693, "reference_mean_power": 0.0034432187972516922, "output_mean_power": 0.0033223425768678203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.845747823685581e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011374442611865624, "output_power": 0.0010958787352681005, "reference_active": true, "delta_db": -0.1616762403269903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011129046974770975, "output_power": 9.943171729014524e-05, "reference_active": true, "delta_db": -0.48933035335789243}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.356136872122755e-05, "output_power": 1.5525587977824118e-05, "reference_active": true, "delta_db": -1.811524592178861}], "active_band_count": 3, "active_band_mae_db": 0.8208437286212479}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026426858408099, "log_rms_mae_db": 0.7190789252366353, "rms_envelope_nrmse": 0.042208163365580004, "envelope_correlation": 0.9988594725147196, "rms_gain_error_db": -0.15462818157754288, "reference_mean_power": 0.0034555022621815964, "output_mean_power": 0.0033346354552573013}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7242545404192642e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010676968681649734, "output_power": 0.0010268764537107586, "reference_active": true, "delta_db": -0.169297732316246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.454467282708388e-05, "output_power": 4.006943250791032e-05, "reference_active": true, "delta_db": -0.45982581779407194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7651139324996653e-06, "output_power": 1.1954405317716006e-06, "reference_active": true, "delta_db": -1.6924476632576944}], "active_band_count": 3, "active_band_mae_db": 0.7738570711226708}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001806334260766054, "log_rms_mae_db": 1.249867627819122, "rms_envelope_nrmse": 0.09696702737661263, "envelope_correlation": 0.9938780596849286, "rms_gain_error_db": -0.12578119826734435, "reference_mean_power": 0.0015370357837470275, "output_mean_power": 0.0014931583202469456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.498278279762008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005426794452872498, "output_power": 0.0005455399019049426, "reference_active": true, "delta_db": 0.022831488690696514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.715116909567221e-05, "output_power": 5.268876525865005e-05, "reference_active": true, "delta_db": -0.35307097671806953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1787274842814488e-05, "output_power": 7.809985306725982e-06, "reference_active": true, "delta_db": -1.7876319323395642}], "active_band_count": 3, "active_band_mae_db": 0.7211781325827767}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017379577463928091, "log_rms_mae_db": 0.8302989341605974, "rms_envelope_nrmse": 0.06082352747220963, "envelope_correlation": 0.9975531626285824, "rms_gain_error_db": -0.12457731861363489, "reference_mean_power": 0.0015404251144981946, "output_mean_power": 0.0014968657754060243}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.036061947317895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005182860743704318, "output_power": 0.0005192509259240439, "reference_active": true, "delta_db": 0.008077395336502018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.185473992393753e-05, "output_power": 1.9694889229781045e-05, "reference_active": true, "delta_db": -0.4519210011555122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.514034995645172e-07, "output_power": 5.438646554581434e-07, "reference_active": true, "delta_db": -1.9464459503641898}], "active_band_count": 3, "active_band_mae_db": 0.8021481156187346}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018667130520360042, "log_rms_mae_db": 1.3083538055470216, "rms_envelope_nrmse": 0.10364667409446154, "envelope_correlation": 0.9936926553332646, "rms_gain_error_db": -0.22851624490014522, "reference_mean_power": 0.001062265092647352, "output_mean_power": 0.0010078160855574055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.90294470054441e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004039276312698453, "output_power": 0.0003769511506979103, "reference_active": true, "delta_db": -0.30018489348547317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.530941238269661e-05, "output_power": 3.595733102416925e-05, "reference_active": true, "delta_db": -1.0040098162402287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.869428114247789e-06, "output_power": 4.789123568121162e-06, "reference_active": true, "delta_db": -0.07221916094860481}], "active_band_count": 3, "active_band_mae_db": 0.45880462355810225}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017791979118134677, "log_rms_mae_db": 0.7908702607930741, "rms_envelope_nrmse": 0.050903969748188574, "envelope_correlation": 0.9987585774442606, "rms_gain_error_db": -0.22852670695837635, "reference_mean_power": 0.0010675566768630375, "output_mean_power": 0.0010128339967260985}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.672011383574165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003902673215159807, "output_power": 0.00036243202540169343, "reference_active": true, "delta_db": -0.321356217187017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8376752084991478e-05, "output_power": 1.4551709263718925e-05, "reference_active": true, "delta_db": -1.0135474729683522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2996790275380256e-07, "output_power": 3.8411770331636835e-07, "reference_active": true, "delta_db": -0.4897171305965742}], "active_band_count": 3, "active_band_mae_db": 0.6082069402506478}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.103910936684106e-05, "log_rms_mae_db": 1.571155524400724, "rms_envelope_nrmse": 0.14446511031075465, "envelope_correlation": 0.9859597023095643, "rms_gain_error_db": -0.3310335299512811, "reference_mean_power": 0.00018921068158653413, "output_mean_power": 0.0001753243741004988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9505568897447032e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.310547772849957e-05, "output_power": 8.132092109132428e-05, "reference_active": true, "delta_db": -0.5877294374856696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9017487007684174e-05, "output_power": 1.5012938323569153e-05, "reference_active": true, "delta_db": -1.026874278008404}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.040878232439125e-06, "output_power": 3.3963290178084626e-06, "reference_active": true, "delta_db": -1.7149645105049367}], "active_band_count": 3, "active_band_mae_db": 1.1098560753330033}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.917454226005249e-05, "log_rms_mae_db": 0.9518281744202466, "rms_envelope_nrmse": 0.08869551242438387, "envelope_correlation": 0.9950279460695379, "rms_gain_error_db": -0.33180483073444095, "reference_mean_power": 0.00019006200677954335, "output_mean_power": 0.00017608194524913038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8235095603824036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.910190725222932e-05, "output_power": 7.682560815705985e-05, "reference_active": true, "delta_db": -0.64380993487526}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361035171170285e-06, "output_power": 7.574346680220256e-06, "reference_active": true, "delta_db": -0.9197869793774374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6146687278338854e-07, "output_power": 2.2003242215145478e-07, "reference_active": true, "delta_db": -2.155818222658707}], "active_band_count": 3, "active_band_mae_db": 1.2398050456371348}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8982290992225472, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a59705038c0d0a7c143d46d9f6e4046cf7a108d026187769dc382e0f133e32c9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9947561643002443, "standard_deviation": 0.000855364341673143, "minimum": 0.9934889952165363, "maximum": 0.9959009038471234, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3037020438612146, "standard_deviation": 0.02140150440433166, "minimum": 1.2762958425402418, "maximum": 1.3325341703839435, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4631295512220891, "standard_deviation": 0.226888533099306, "minimum": 0.2721016622150498, "maximum": 0.8988346499128611, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09327394625079377, "standard_deviation": 0.007415623246944261, "minimum": 0.08237829267880988, "maximum": 0.10364667409446154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9986004090555647, "standard_deviation": 0.0003348927934017147, "minimum": 0.9980734468643465, "maximum": 0.9989498445398639, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8006153946026191, "standard_deviation": 0.01826426586332688, "minimum": 0.7762288425264944, "maximum": 0.8280025518741765, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5775233086031032, "standard_deviation": 0.2955130765276185, "minimum": 0.25714314008063593, "maximum": 1.1546067915122442, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04955525708593899, "standard_deviation": 0.00440177801911606, "minimum": 0.04358152436247641, "maximum": 0.05588556033588606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9869236501525596, "standard_deviation": 0.0012030076565982917, "minimum": 0.9846290742200161, "maximum": 0.9882691479576013, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5688971128846148, "standard_deviation": 0.019189462040823686, "minimum": 1.5359659928887142, "maximum": 1.6025383896077376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8512323643609412, "standard_deviation": 0.3957202776390648, "minimum": 0.12438459035187695, "maximum": 1.416897251500731, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13944277754092121, "standard_deviation": 0.0057748485838316965, "minimum": 0.13263670644622388, "maximum": 0.1507909963269302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9958599549759726, "standard_deviation": 0.000649867076651123, "minimum": 0.9950279460695379, "maximum": 0.9967932730939303, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9418932882969976, "standard_deviation": 0.011898981347732221, "minimum": 0.9268043559044369, "maximum": 0.9551437976282467, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0579116675887876, "standard_deviation": 0.40249111352170913, "minimum": 0.4262217815647847, "maximum": 1.6359308330608247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08178675446261634, "standard_deviation": 0.005496343004058559, "minimum": 0.07405929085123336, "maximum": 0.08869551242438387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943795765388078, "standard_deviation": 0.0006086172714717758, "minimum": 0.993340537835026, "maximum": 0.9952390077902737, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2614686390877892, "standard_deviation": 0.041450852859767844, "minimum": 1.191164616981724, "maximum": 1.3231007515047737, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3999593027548223, "standard_deviation": 0.2854730694395413, "minimum": 0.11095640540544993, "maximum": 0.7997066867991877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09253034187713818, "standard_deviation": 0.0050026345434573665, "minimum": 0.08535139396544582, "maximum": 0.1007825887508157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981280447534557, "standard_deviation": 0.0003736825547981491, "minimum": 0.9975531626285824, "maximum": 0.9984578938866593, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8509629948057114, "standard_deviation": 0.036893344255926705, "minimum": 0.7791212811732618, "maximum": 0.898114034485174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5006122976361049, "standard_deviation": 0.315830160463544, "minimum": 0.05147181768407801, "maximum": 0.8783054062580872, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.052565503798166616, "standard_deviation": 0.005314377504266036, "minimum": 0.047597532142645435, "maximum": 0.06082352747220963, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1763027789639226, "standard_deviation": 0.04463304638333557, "minimum": 0.12812196992499333, "maximum": 0.25675753785665395, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.28306236802581, "standard_deviation": 0.005539821310939909, "minimum": 5.272538265499972, "maximum": 5.289373601450731, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7330447427936253, "standard_deviation": 0.032749072888273195, "minimum": 0.6996961832163611, "maximum": 0.794604647039014, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09376174667687197, "standard_deviation": 0.04278540559140716, "minimum": -0.15561232303191067, "maximum": -0.013711536919779437, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996279960238774, "standard_deviation": 0.00029597469292982437, "minimum": 0.9957252592909405, "maximum": 0.9966052276920583, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.143168679210068, "standard_deviation": 0.030386569708267454, "minimum": 1.0950228636322434, "maximum": 1.1924235401191572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5247732641581894, "standard_deviation": 0.2318318042527124, "minimum": 0.2072152749448509, "maximum": 0.8208437286212479, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0725267500856547, "standard_deviation": 0.0032220190447550577, "minimum": 0.06894795490351643, "maximum": 0.07855316573945094, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998943634037887, "standard_deviation": 0.00017635865928296665, "minimum": 0.9987139339388179, "maximum": 0.9992076897687218, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6942692888445208, "standard_deviation": 0.02627059681544532, "minimum": 0.661098719212367, "maximum": 0.7412018729004884, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5964897368620419, "standard_deviation": 0.24399085515458518, "minimum": 0.24781115674704443, "maximum": 0.9360310327523855, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038663466877240864, "standard_deviation": 0.003699653661495709, "minimum": 0.03353488824409592, "maximum": 0.04257381010308572, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29093-0005", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.9636211395263672, "predicted_mos": 1.9636211395263672, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2c17258e743fb181b949053fe6978987a69c1c592433c204a1c7f357db424d89", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.483983899932355, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0880189610196018, "全帯域_10ms_rms_envelope_nrmse": 0.06603378011552259, "全帯域_10ms_envelope_correlation": 0.9968815471164214, "全帯域_10ms_modulation_mae_db": 0.3567302165093335, "全帯域_40ms_log_rms_mae_db": 0.6600081157623431, "全帯域_40ms_rms_envelope_nrmse": 0.032949383697547276, "全帯域_40ms_envelope_correlation": 0.9991798585034555, "全帯域_40ms_modulation_mae_db": 0.3377096548777198, "300–1000 Hz_10ms_log_rms_mae_db": 1.2567161815886962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0818204759998464, "300–1000 Hz_10ms_envelope_correlation": 0.9956025987022465, "300–1000 Hz_10ms_modulation_mae_db": 0.28414266045939257, "300–1000 Hz_40ms_log_rms_mae_db": 0.8442286435122744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042951050175496476, "300–1000 Hz_40ms_envelope_correlation": 0.9987302323406775, "300–1000 Hz_40ms_modulation_mae_db": 0.2824075431391854, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2753098887661778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08425890729074385, "1000–3000 Hz_10ms_envelope_correlation": 0.9956579609959422, "1000–3000 Hz_10ms_modulation_mae_db": 0.3301836586817215, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7619442791070455, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047546127625757835, "1000–3000 Hz_40ms_envelope_correlation": 0.9985738966167679, "1000–3000 Hz_40ms_modulation_mae_db": 0.4255180413843475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5434899669099258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1296630444544643, "3000–6000 Hz_10ms_envelope_correlation": 0.9884733367790336, "3000–6000 Hz_10ms_modulation_mae_db": 0.4969627592185322, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9495959432929733, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07986416482834288, "3000–6000 Hz_40ms_envelope_correlation": 0.9956301748988337, "3000–6000 Hz_40ms_modulation_mae_db": 0.6470565254603683, "mr_stft_db": 5.2507927310271425, "band_mae_db": 0.13478800362529678, "rms_gain_error_db": -0.014956590076490698, "output_peak": 0.6975262409724131}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0884447402127815, "全帯域_10ms_rms_envelope_nrmse": 0.0767101342834722, "全帯域_10ms_envelope_correlation": 0.9957675918686487, "全帯域_10ms_modulation_mae_db": 0.41758890611342997, "全帯域_40ms_log_rms_mae_db": 0.641670740549178, "全帯域_40ms_rms_envelope_nrmse": 0.04125631714406978, "全帯域_40ms_envelope_correlation": 0.9987064437892993, "全帯域_40ms_modulation_mae_db": 0.4170762764851876, "300–1000 Hz_10ms_log_rms_mae_db": 1.2863823182612104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09478682936359471, "300–1000 Hz_10ms_envelope_correlation": 0.9941584172846288, "300–1000 Hz_10ms_modulation_mae_db": 0.39922479009448186, "300–1000 Hz_40ms_log_rms_mae_db": 0.8777919987936847, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05327298891691849, "300–1000 Hz_40ms_envelope_correlation": 0.9980794910205203, "300–1000 Hz_40ms_modulation_mae_db": 0.37548751892749194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3353767316802707, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11187272883865873, "1000–3000 Hz_10ms_envelope_correlation": 0.9923626303230517, "1000–3000 Hz_10ms_modulation_mae_db": 0.21103349038344363, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8394570614582127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06570230403606869, "1000–3000 Hz_40ms_envelope_correlation": 0.9973450662038028, "1000–3000 Hz_40ms_modulation_mae_db": 0.5577645856746732, "3000–6000 Hz_10ms_log_rms_mae_db": 1.49632840810533, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1358757678945432, "3000–6000 Hz_10ms_envelope_correlation": 0.987306389984763, "3000–6000 Hz_10ms_modulation_mae_db": 0.42074575940177555, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8619803039504311, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07438282909284383, "3000–6000 Hz_40ms_envelope_correlation": 0.9963341696080094, "3000–6000 Hz_40ms_modulation_mae_db": 0.5944025448576579, "mr_stft_db": 5.2472797378676175, "band_mae_db": 0.1823937913568543, "rms_gain_error_db": -0.0343677882647459, "output_peak": 0.7670687337189618}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0910675306459128, "全帯域_10ms_rms_envelope_nrmse": 0.07445633353067603, "全帯域_10ms_envelope_correlation": 0.9960790275005976, "全帯域_10ms_modulation_mae_db": 0.5575264083216296, "全帯域_40ms_log_rms_mae_db": 0.6578550302882566, "全帯域_40ms_rms_envelope_nrmse": 0.04780259560503942, "全帯域_40ms_envelope_correlation": 0.9983447763508887, "全帯域_40ms_modulation_mae_db": 0.7181917308054157, "300–1000 Hz_10ms_log_rms_mae_db": 1.2667522412830017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10293273301606626, "300–1000 Hz_10ms_envelope_correlation": 0.9930235770751259, "300–1000 Hz_10ms_modulation_mae_db": 0.43898741391128165, "300–1000 Hz_40ms_log_rms_mae_db": 0.9323412072026609, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06919213481506843, "300–1000 Hz_40ms_envelope_correlation": 0.9966832341861005, "300–1000 Hz_40ms_modulation_mae_db": 0.5687437670225927, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2591700649938917, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09148400588491731, "1000–3000 Hz_10ms_envelope_correlation": 0.9952704418106543, "1000–3000 Hz_10ms_modulation_mae_db": 0.4544847646254815, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8139655374851478, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055443463533412306, "1000–3000 Hz_40ms_envelope_correlation": 0.9985443969047407, "1000–3000 Hz_40ms_modulation_mae_db": 0.749154492510173, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4959377185288238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12371759955670224, "3000–6000 Hz_10ms_envelope_correlation": 0.9898378355770099, "3000–6000 Hz_10ms_modulation_mae_db": 1.07031971835883, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8535991213938969, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06953716351299342, "3000–6000 Hz_40ms_envelope_correlation": 0.9971667863337613, "3000–6000 Hz_40ms_modulation_mae_db": 1.1555074540248833, "mr_stft_db": 5.251008682539575, "band_mae_db": 0.1768941881555041, "rms_gain_error_db": -0.10232664013314655, "output_peak": 0.7804436672260017}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0971329308397701, "全帯域_10ms_rms_envelope_nrmse": 0.0711126660339406, "全帯域_10ms_envelope_correlation": 0.9964331047001622, "全帯域_10ms_modulation_mae_db": 0.6844753583533985, "全帯域_40ms_log_rms_mae_db": 0.6672853805530815, "全帯域_40ms_rms_envelope_nrmse": 0.038896955215820685, "全帯域_40ms_envelope_correlation": 0.9989419689159269, "全帯域_40ms_modulation_mae_db": 0.7433050858389331, "300–1000 Hz_10ms_log_rms_mae_db": 1.2337810441747032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09510929499833545, "300–1000 Hz_10ms_envelope_correlation": 0.9940732374683567, "300–1000 Hz_10ms_modulation_mae_db": 0.6406827192415531, "300–1000 Hz_40ms_log_rms_mae_db": 0.8210217268805594, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05602665119163706, "300–1000 Hz_40ms_envelope_correlation": 0.9978983746691841, "300–1000 Hz_40ms_modulation_mae_db": 0.6142346667069559, "1000–3000 Hz_10ms_log_rms_mae_db": 1.270844594530868, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09153406634102478, "1000–3000 Hz_10ms_envelope_correlation": 0.9949347099319785, "1000–3000 Hz_10ms_modulation_mae_db": 0.4431544650676085, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7718055560521673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04594615658237154, "1000–3000 Hz_40ms_envelope_correlation": 0.998772816374459, "1000–3000 Hz_40ms_modulation_mae_db": 0.5268153609935827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5455201732618635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12883950215650192, "3000–6000 Hz_10ms_envelope_correlation": 0.9892722119681896, "3000–6000 Hz_10ms_modulation_mae_db": 0.8192744655831236, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9165556845505363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07603477911098849, "3000–6000 Hz_40ms_envelope_correlation": 0.9969153465181079, "3000–6000 Hz_40ms_modulation_mae_db": 0.9635534158186102, "mr_stft_db": 5.25332896391859, "band_mae_db": 0.17438542023593107, "rms_gain_error_db": -0.10581536501260437, "output_peak": 0.7857096933842725}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0899182090815012, "全帯域_10ms_rms_envelope_nrmse": 0.07872477642464974, "全帯域_10ms_envelope_correlation": 0.9956503532454148, "全帯域_10ms_modulation_mae_db": 0.7918016733770109, "全帯域_40ms_log_rms_mae_db": 0.6295205480843307, "全帯域_40ms_rms_envelope_nrmse": 0.04063050907032244, "全帯域_40ms_envelope_correlation": 0.9988928715514153, "全帯域_40ms_modulation_mae_db": 0.7400634164718946, "300–1000 Hz_10ms_log_rms_mae_db": 1.2751528375174515, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10765132148857076, "300–1000 Hz_10ms_envelope_correlation": 0.9925327464533837, "300–1000 Hz_10ms_modulation_mae_db": 0.8170342295400603, "300–1000 Hz_40ms_log_rms_mae_db": 0.8923432365856123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06959336336529467, "300–1000 Hz_40ms_envelope_correlation": 0.9969047422991063, "300–1000 Hz_40ms_modulation_mae_db": 0.7627926023113408, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2748041195210922, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10112078593632243, "1000–3000 Hz_10ms_envelope_correlation": 0.9937502150602416, "1000–3000 Hz_10ms_modulation_mae_db": 0.4614212498128607, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7855296819577351, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049651711175899706, "1000–3000 Hz_40ms_envelope_correlation": 0.998481585957668, "1000–3000 Hz_40ms_modulation_mae_db": 0.6463531139667437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5259666699076764, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12365953420011763, "3000–6000 Hz_10ms_envelope_correlation": 0.9894940715307804, "3000–6000 Hz_10ms_modulation_mae_db": 0.47333442259732816, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8902028709840222, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06427336600810803, "3000–6000 Hz_40ms_envelope_correlation": 0.9972961505135132, "3000–6000 Hz_40ms_modulation_mae_db": 0.44020443462220665, "mr_stft_db": 5.248154441234718, "band_mae_db": 0.15120249000882072, "rms_gain_error_db": -0.12966535855788688, "output_peak": 0.731960862965024}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1114033158723882, "全帯域_10ms_rms_envelope_nrmse": 0.07676560054887355, "全帯域_10ms_envelope_correlation": 0.99579228390853, "全帯域_10ms_modulation_mae_db": 0.6646000675013187, "全帯域_40ms_log_rms_mae_db": 0.6676513857451187, "全帯域_40ms_rms_envelope_nrmse": 0.0395023779659058, "全帯域_40ms_envelope_correlation": 0.9988553535885097, "全帯域_40ms_modulation_mae_db": 0.813342933669821, "300–1000 Hz_10ms_log_rms_mae_db": 1.2610757931643792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09727170543121003, "300–1000 Hz_10ms_envelope_correlation": 0.9937737289068108, "300–1000 Hz_10ms_modulation_mae_db": 0.4679637030683696, "300–1000 Hz_40ms_log_rms_mae_db": 0.8601180602576661, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0563940576139194, "300–1000 Hz_40ms_envelope_correlation": 0.9977984154868367, "300–1000 Hz_40ms_modulation_mae_db": 0.5721874591706069, "1000–3000 Hz_10ms_log_rms_mae_db": 1.284245757478581, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10322328188530301, "1000–3000 Hz_10ms_envelope_correlation": 0.9935898746063809, "1000–3000 Hz_10ms_modulation_mae_db": 0.7149296790995766, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7649224547669948, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05581004505201461, "1000–3000 Hz_40ms_envelope_correlation": 0.9982148276273317, "1000–3000 Hz_40ms_modulation_mae_db": 0.8354197099493582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5046287188085818, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13062700751338918, "3000–6000 Hz_10ms_envelope_correlation": 0.9887270964937794, "3000–6000 Hz_10ms_modulation_mae_db": 0.944354709704989, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8851406641554643, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06897410762421836, "3000–6000 Hz_40ms_envelope_correlation": 0.9974532314771295, "3000–6000 Hz_40ms_modulation_mae_db": 1.1412136924004548, "mr_stft_db": 5.252055689024469, "band_mae_db": 0.15640485560361053, "rms_gain_error_db": -0.07363047089442713, "output_peak": 0.6971379025982077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.130270193802926, "全帯域_10ms_rms_envelope_nrmse": 0.07532702604729338, "全帯域_10ms_envelope_correlation": 0.9959926912975334, "全帯域_10ms_modulation_mae_db": 0.45496922685188457, "全帯域_40ms_log_rms_mae_db": 0.6822265130855363, "全帯域_40ms_rms_envelope_nrmse": 0.039462438081379225, "全帯域_40ms_envelope_correlation": 0.9989163721967399, "全帯域_40ms_modulation_mae_db": 0.4547701817697898, "300–1000 Hz_10ms_log_rms_mae_db": 1.2580913817031243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1033180856502291, "300–1000 Hz_10ms_envelope_correlation": 0.9930278392638447, "300–1000 Hz_10ms_modulation_mae_db": 0.2805226428423091, "300–1000 Hz_40ms_log_rms_mae_db": 0.8407635187301548, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06115789453109832, "300–1000 Hz_40ms_envelope_correlation": 0.9975366212052302, "300–1000 Hz_40ms_modulation_mae_db": 0.3148385868923999, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2637055632283312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09579460740804488, "1000–3000 Hz_10ms_envelope_correlation": 0.9944133561511754, "1000–3000 Hz_10ms_modulation_mae_db": 0.3899703197941385, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7769337434634369, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05072419958314123, "1000–3000 Hz_40ms_envelope_correlation": 0.9984400060712993, "1000–3000 Hz_40ms_modulation_mae_db": 0.610663595184041, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5344957453060095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1328086992974004, "3000–6000 Hz_10ms_envelope_correlation": 0.9877600902256284, "3000–6000 Hz_10ms_modulation_mae_db": 0.9314775298302581, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9121305129043018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06953968939400175, "3000–6000 Hz_40ms_envelope_correlation": 0.9966670306750399, "3000–6000 Hz_40ms_modulation_mae_db": 1.2084507752542124, "mr_stft_db": 5.241473598312249, "band_mae_db": 0.15753063031829825, "rms_gain_error_db": -0.10843644522906941, "output_peak": 0.6894936168857018}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.114339435166918, "全帯域_10ms_rms_envelope_nrmse": 0.07219543751903627, "全帯域_10ms_envelope_correlation": 0.996303819040213, "全帯域_10ms_modulation_mae_db": 0.36049248111831705, "全帯域_40ms_log_rms_mae_db": 0.6708974607293068, "全帯域_40ms_rms_envelope_nrmse": 0.03679931589724864, "全帯域_40ms_envelope_correlation": 0.9990266965556204, "全帯域_40ms_modulation_mae_db": 0.4737400387137957, "300–1000 Hz_10ms_log_rms_mae_db": 1.240961566957909, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09362812967904258, "300–1000 Hz_10ms_envelope_correlation": 0.994251547510841, "300–1000 Hz_10ms_modulation_mae_db": 0.09067544914817689, "300–1000 Hz_40ms_log_rms_mae_db": 0.8365268260532006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049275114895412885, "300–1000 Hz_40ms_envelope_correlation": 0.9983543646853936, "300–1000 Hz_40ms_modulation_mae_db": 0.2775418272175954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2893080231756335, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08391998042385829, "1000–3000 Hz_10ms_envelope_correlation": 0.9957682586715735, "1000–3000 Hz_10ms_modulation_mae_db": 0.2530527259971401, "1000–3000 Hz_40ms_log_rms_mae_db": 0.769787476192288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.043938393922991856, "1000–3000 Hz_40ms_envelope_correlation": 0.9988908783389822, "1000–3000 Hz_40ms_modulation_mae_db": 0.5524026813813355, "3000–6000 Hz_10ms_log_rms_mae_db": 1.53687572251786, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13099920746631905, "3000–6000 Hz_10ms_envelope_correlation": 0.9886779819768963, "3000–6000 Hz_10ms_modulation_mae_db": 0.8184252084087581, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9171007695895194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07363969305469222, "3000–6000 Hz_40ms_envelope_correlation": 0.996965612011838, "3000–6000 Hz_40ms_modulation_mae_db": 1.079426342008274, "mr_stft_db": 5.245111325547782, "band_mae_db": 0.17538453739717105, "rms_gain_error_db": -0.07452307937704537, "output_peak": 0.7107270425615534}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.640566533054363, "spectral_convergence": 0.2566830514812511, "spectral_convergence_status": "ok", "amplitude_floor": 1.6417184973675735e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1166612362236465, "spectral_convergence": 0.2544541761282634, "spectral_convergence_status": "ok", "amplitude_floor": 1.2466848673527415e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.995150423803416, "spectral_convergence": 0.2592534422624662, "spectral_convergence_status": "ok", "amplitude_floor": 1.127090317770565e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.270965260159295e-05, "output_power": 4.613229610007804e-05, "delta_db": 0.33479031974251444, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004287949616170093, "output_power": 0.00043694458476625797, "delta_db": 0.08176687614974273, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015047786428193433, "output_power": 0.0014874163653440724, "delta_db": -0.05040063169566018, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0009641149894002944, "output_power": 0.0009761603129287001, "delta_db": 0.053923118434577276, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00041492564533297956, "output_power": 0.000399178118222834, "delta_db": -0.16803551608015888, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.594979221160373e-05, "output_power": 8.361104956230135e-05, "delta_db": -0.11981155964912721, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13478800362529678, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.014956590076490698, "pitch": null, "output_rms_dbfs": -24.64776445637819, "output_peak": 0.6975262409724131, "reference_rms_dbfs": -24.632807866301704, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.2507927310271425}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027067935709258154, "log_rms_mae_db": 1.0880189610196018, "rms_envelope_nrmse": 0.06603378011552259, "envelope_correlation": 0.9968815471164214, "rms_gain_error_db": -0.014224761647293326, "reference_mean_power": 0.0034432187972516922, "output_mean_power": 0.0034319594227827347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.845747823685581e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011374442611865624, "output_power": 0.001113303445968339, "reference_active": true, "delta_db": -0.09316570549954423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011129046974770975, "output_power": 9.771132240379329e-05, "reference_active": true, "delta_db": -0.5651308456552645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.356136872122755e-05, "output_power": 2.142945400609321e-05, "reference_active": true, "delta_db": -0.41189409837319174}], "active_band_count": 3, "active_band_mae_db": 0.3567302165093335}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026426858408099, "log_rms_mae_db": 0.6600081157623431, "rms_envelope_nrmse": 0.032949383697547276, "envelope_correlation": 0.9991798585034555, "rms_gain_error_db": -0.013579052002189772, "reference_mean_power": 0.0034555022621815964, "output_mean_power": 0.003444714843221591}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7242545404192642e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010676968681649734, "output_power": 0.0010433050033396584, "reference_active": true, "delta_db": -0.10036678708492981}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.454467282708388e-05, "output_power": 3.9828481646967954e-05, "reference_active": true, "delta_db": -0.48602023041627146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7651139324996653e-06, "output_power": 1.5999210346612748e-06, "reference_active": true, "delta_db": -0.42674194713195807}], "active_band_count": 3, "active_band_mae_db": 0.3377096548777198}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001806334260766054, "log_rms_mae_db": 1.2567161815886962, "rms_envelope_nrmse": 0.0818204759998464, "envelope_correlation": 0.9956025987022465, "rms_gain_error_db": -0.010207334726975487, "reference_mean_power": 0.0015370357837470275, "output_mean_power": 0.0015334274910687408}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.498278279762008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005426794452872498, "output_power": 0.0005341519612384135, "reference_active": true, "delta_db": -0.06878545078337181}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.715116909567221e-05, "output_power": 4.8836154047625294e-05, "reference_active": true, "delta_db": -0.6828366384326217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1787274842814488e-05, "output_power": 1.1516826408799058e-05, "reference_active": true, "delta_db": -0.10080589216218412}], "active_band_count": 3, "active_band_mae_db": 0.28414266045939257}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017379577463928091, "log_rms_mae_db": 0.8442286435122744, "rms_envelope_nrmse": 0.042951050175496476, "envelope_correlation": 0.9987302323406775, "rms_gain_error_db": -0.008813351512001532, "reference_mean_power": 0.0015404251144981946, "output_mean_power": 0.0015373022238431005}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.036061947317895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005182860743704318, "output_power": 0.000507633907926526, "reference_active": true, "delta_db": -0.09018916568530232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.185473992393753e-05, "output_power": 1.900937934798645e-05, "reference_active": true, "delta_db": -0.6057770518099153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.514034995645172e-07, "output_power": 8.815786666691097e-07, "reference_active": true, "delta_db": 0.1512564119223386}], "active_band_count": 3, "active_band_mae_db": 0.2824075431391854}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018667130520360042, "log_rms_mae_db": 1.2753098887661778, "rms_envelope_nrmse": 0.08425890729074385, "envelope_correlation": 0.9956579609959422, "rms_gain_error_db": -0.036990255958792045, "reference_mean_power": 0.001062265092647352, "output_mean_power": 0.0010532558613808437}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.90294470054441e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004039276312698453, "output_power": 0.00040015813885531116, "reference_active": true, "delta_db": -0.04071908268030348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.530941238269661e-05, "output_power": 3.9468496530439746e-05, "reference_active": true, "delta_db": -0.5993784668146088}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.869428114247789e-06, "output_power": 4.491926239302863e-06, "reference_active": true, "delta_db": -0.3504534265502523}], "active_band_count": 3, "active_band_mae_db": 0.3301836586817215}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017791979118134677, "log_rms_mae_db": 0.7619442791070455, "rms_envelope_nrmse": 0.047546127625757835, "envelope_correlation": 0.9985738966167679, "rms_gain_error_db": -0.03702285822658078, "reference_mean_power": 0.0010675566768630375, "output_mean_power": 0.0010584946207731344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.672011383574165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003902673215159807, "output_power": 0.00038522968548483405, "reference_active": true, "delta_db": -0.05642441647451097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8376752084991478e-05, "output_power": 1.5974245969503198e-05, "reference_active": true, "delta_db": -0.6084838907849982}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2996790275380256e-07, "output_power": 3.734835299666754e-07, "reference_active": true, "delta_db": -0.6116458168935334}], "active_band_count": 3, "active_band_mae_db": 0.4255180413843475}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.103910936684106e-05, "log_rms_mae_db": 1.5434899669099258, "rms_envelope_nrmse": 0.1296630444544643, "envelope_correlation": 0.9884733367790336, "rms_gain_error_db": -0.2078648380659075, "reference_mean_power": 0.00018921068158653413, "output_mean_power": 0.0001803678657271879}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9505568897447032e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.310547772849957e-05, "output_power": 8.447856291566813e-05, "reference_active": true, "delta_db": -0.42228715462902777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9017487007684174e-05, "output_power": 1.6730371037858465e-05, "reference_active": true, "delta_db": -0.5564755559052513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.040878232439125e-06, "output_power": 4.480161297619505e-06, "reference_active": true, "delta_db": -0.5121255671213175}], "active_band_count": 3, "active_band_mae_db": 0.4969627592185322}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.917454226005249e-05, "log_rms_mae_db": 0.9495959432929733, "rms_envelope_nrmse": 0.07986416482834288, "envelope_correlation": 0.9956301748988337, "rms_gain_error_db": -0.2081378556484882, "reference_mean_power": 0.00019006200677954335, "output_mean_power": 0.0001811680145747179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8235095603824036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.910190725222932e-05, "output_power": 8.049538917725708e-05, "reference_active": true, "delta_db": -0.44115995896398535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361035171170285e-06, "output_power": 8.699074576957593e-06, "reference_active": true, "delta_db": -0.31850822916292054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6146687278338854e-07, "output_power": 2.7537110227688473e-07, "reference_active": true, "delta_db": -1.1815013882541991}], "active_band_count": 3, "active_band_mae_db": 0.6470565254603683}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9099722184288396, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dd03e5e7349c16124bac5087ff4655fd2150378765ad146dfedf8970a880f6ba", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9944684309438747, "standard_deviation": 0.0011765746999992614, "minimum": 0.9923626303230517, "maximum": 0.9957682586715735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2815955929218559, "standard_deviation": 0.023853994748173096, "minimum": 1.2591700649938917, "maximum": 1.3353767316802707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4072787941827464, "standard_deviation": 0.15567231264674494, "minimum": 0.21103349038344363, "maximum": 0.7149296790995766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09540104550110916, "standard_deviation": 0.009648337251492865, "minimum": 0.08391998042385829, "maximum": 0.11187272883865873, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9984079342618815, "standard_deviation": 0.00047599089851200696, "minimum": 0.9973450662038028, "maximum": 0.9988908783389822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7855432238103786, "standard_deviation": 0.02727569966435757, "minimum": 0.7619442791070455, "maximum": 0.8394570614582127, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6130114476305318, "standard_deviation": 0.1300696099245687, "minimum": 0.4255180413843475, "maximum": 0.8354197099493582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05184530018895722, "standard_deviation": 0.006995439156499442, "minimum": 0.043938393922991856, "maximum": 0.06570230403606869, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9886936268170101, "standard_deviation": 0.0008561120988629585, "minimum": 0.987306389984763, "maximum": 0.9898378355770099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5229053904182588, "standard_deviation": 0.020845209011185584, "minimum": 1.4959377185288238, "maximum": 1.5455201732618635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7468618216379493, "standard_deviation": 0.2483358285816095, "minimum": 0.42074575940177555, "maximum": 1.07031971835883, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12952379531742975, "standard_deviation": 0.004193893231364563, "minimum": 0.12365953420011763, "maximum": 0.1358757678945432, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9968035627545291, "standard_deviation": 0.0005917539406777526, "minimum": 0.9956301748988337, "maximum": 0.9974532314771295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8982882338526432, "standard_deviation": 0.0317699660978334, "minimum": 0.8535991213938969, "maximum": 0.9495959432929733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9037268980558334, "standard_deviation": 0.2984516911799343, "minimum": 0.44020443462220665, "maximum": 1.2084507752542124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07203072407827363, "standard_deviation": 0.004891085715337201, "minimum": 0.06427336600810803, "maximum": 0.07986416482834288, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938054615831549, "standard_deviation": 0.0009607903555538163, "minimum": 0.9925327464533837, "maximum": 0.9956025987022465, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2598641705813094, "standard_deviation": 0.017056877801400035, "minimum": 1.2337810441747032, "maximum": 1.2863823182612104, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.42740420103820315, "standard_deviation": 0.22531150568566521, "minimum": 0.09067544914817689, "maximum": 0.8170342295400603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09706482195336191, "standard_deviation": 0.007918033066853057, "minimum": 0.0818204759998464, "maximum": 0.10765132148857076, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977481844866312, "standard_deviation": 0.0006925098836899039, "minimum": 0.9966832341861005, "maximum": 0.9987302323406775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8631419022519766, "standard_deviation": 0.03626490933041147, "minimum": 0.8210217268805594, "maximum": 0.9323412072026609, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4710292464235212, "standard_deviation": 0.18203376136550467, "minimum": 0.2775418272175954, "maximum": 0.7627926023113408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05723290693810572, "standard_deviation": 0.009229975241711395, "minimum": 0.042951050175496476, "maximum": 0.06959336336529467, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16362298958768584, "standard_deviation": 0.016290116002989984, "minimum": 0.13478800362529678, "maximum": 0.1823937913568543, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.248650646184018, "standard_deviation": 0.003960379710937334, "minimum": 5.241473598312249, "maximum": 5.25332896391859, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.732508470039017, "standard_deviation": 0.039861199466276774, "minimum": 0.6894936168857018, "maximum": 0.7857096933842725, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08046521719317704, "standard_deviation": 0.03931915428332128, "minimum": -0.12966535855788688, "maximum": -0.014956590076490698, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9961125523346901, "standard_deviation": 0.00041081415936857537, "minimum": 0.9956503532454148, "maximum": 0.9968815471164214, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1013244145802248, "standard_deviation": 0.015605072348765374, "minimum": 1.0880189610196018, "maximum": 1.130270193802926, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5360230422682903, "standard_deviation": 0.1638851973263355, "minimum": 0.3567302165093335, "maximum": 0.7918016733770109, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07391571931293306, "standard_deviation": 0.0040405639894189995, "minimum": 0.06603378011552259, "maximum": 0.07872477642464974, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988580426814819, "standard_deviation": 0.0002478767923633057, "minimum": 0.9983447763508887, "maximum": 0.9991798585034555, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6596393968496439, "standard_deviation": 0.016873108075450328, "minimum": 0.6295205480843307, "maximum": 0.6822265130855363, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5872749148290697, "standard_deviation": 0.18424896595713386, "minimum": 0.3377096548777198, "maximum": 0.813342933669821, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03966248658466666, "standard_deviation": 0.0042029569090764295, "minimum": 0.032949383697547276, "maximum": 0.04780259560503942, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4970-29095-0008", "language": "en", "style": "normal", "speaker": "librispeech_4970", "source_utterance_id": "4970-29095-0008", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 48640, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4970-29095-0008.wav", "source_file_sha256": "97cc13834ebedb08201f7c89ee10174703a028aab5a4ce12c5e4897544885afc", "decoded_pcm_sha256": "44ec7bab382a51da3a9f07cbc8ffdb491029ea5b6b84d63bfceb71c60641803b", "original_source_sha256": "cd2a3b352b93beb33caced744406acf8dace3f1bbb5fc63c8feb4c8a0ed55bb4", "corpus_entry_id": "librispeech_4970-29095-0008"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.036919116973877, "predicted_mos": 4.036919116973877, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "04cec3b427dbbf48dfd02f834c89e555fcc8e973100feb0211cffc0c22b254cb", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.59282143299788, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4970-29095-0008", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "04cec3b427dbbf48dfd02f834c89e555fcc8e973100feb0211cffc0c22b254cb", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "98cb7d4b93856ed5577d455b6af75b342eb49cf42be0c4bd39ba08000ad1229e"}, "rows": [{"case": "librispeech_4970-29095-0008", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.32261061668396, "predicted_mos": 2.32261061668396, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "d12f1554199459cfc0c64d0d9bb243f5d45bd60385c7284ca9172dcfb82b2d45", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.661859073999949, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4970-29095-0008", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d12f1554199459cfc0c64d0d9bb243f5d45bd60385c7284ca9172dcfb82b2d45", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9209865131843162, "全帯域_10ms_rms_envelope_nrmse": 0.07848525122983815, "全帯域_10ms_envelope_correlation": 0.9949641509399305, "全帯域_10ms_modulation_mae_db": 0.2735062455654245, "全帯域_40ms_log_rms_mae_db": 0.5759442647435644, "全帯域_40ms_rms_envelope_nrmse": 0.03630879484021844, "全帯域_40ms_envelope_correlation": 0.9988556333336248, "全帯域_40ms_modulation_mae_db": 0.32244874394550904, "300–1000 Hz_10ms_log_rms_mae_db": 0.9646614152913712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09566610933565008, "300–1000 Hz_10ms_envelope_correlation": 0.9932503558404332, "300–1000 Hz_10ms_modulation_mae_db": 0.4648832855870899, "300–1000 Hz_40ms_log_rms_mae_db": 0.5709658029404485, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04578490260378495, "300–1000 Hz_40ms_envelope_correlation": 0.9983677755585287, "300–1000 Hz_40ms_modulation_mae_db": 0.443203038110858, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1735544202441273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12044916385780738, "1000–3000 Hz_10ms_envelope_correlation": 0.9906499326732031, "1000–3000 Hz_10ms_modulation_mae_db": 0.323096671083224, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6322990686832121, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0535991102241229, "1000–3000 Hz_40ms_envelope_correlation": 0.998114916720228, "1000–3000 Hz_40ms_modulation_mae_db": 0.17217082743095616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.393204155832888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1489503612665709, "3000–6000 Hz_10ms_envelope_correlation": 0.9845236642932139, "3000–6000 Hz_10ms_modulation_mae_db": 1.1501409840775159, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7875198360859189, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08073441416534424, "3000–6000 Hz_40ms_envelope_correlation": 0.9960784451654116, "3000–6000 Hz_40ms_modulation_mae_db": 1.1703283806988096, "mr_stft_db": 5.300523410420143, "band_mae_db": 0.17688574864324288, "rms_gain_error_db": -0.043470362196770126, "output_peak": 0.36357568484389957}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0380479883811733, "全帯域_10ms_rms_envelope_nrmse": 0.0860905591355317, "全帯域_10ms_envelope_correlation": 0.9940907341596766, "全帯域_10ms_modulation_mae_db": 0.520234335842999, "全帯域_40ms_log_rms_mae_db": 0.7029027984436614, "全帯域_40ms_rms_envelope_nrmse": 0.04552116038594956, "全帯域_40ms_envelope_correlation": 0.9984170987981736, "全帯域_40ms_modulation_mae_db": 0.6166297761765371, "300–1000 Hz_10ms_log_rms_mae_db": 1.0011845528741, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11449319273167753, "300–1000 Hz_10ms_envelope_correlation": 0.990374080843365, "300–1000 Hz_10ms_modulation_mae_db": 0.47353443944632706, "300–1000 Hz_40ms_log_rms_mae_db": 0.5809754849213435, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059976618697867905, "300–1000 Hz_40ms_envelope_correlation": 0.9973011380179776, "300–1000 Hz_40ms_modulation_mae_db": 0.46900127672561315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2228825727665906, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13805533700470377, "1000–3000 Hz_10ms_envelope_correlation": 0.9880975024977554, "1000–3000 Hz_10ms_modulation_mae_db": 1.1681004883610615, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7628122203702045, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06653132273615854, "1000–3000 Hz_40ms_envelope_correlation": 0.997727122078673, "1000–3000 Hz_40ms_modulation_mae_db": 1.2338467930744992, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3542312842519555, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16350652940936844, "3000–6000 Hz_10ms_envelope_correlation": 0.9838205970745723, "3000–6000 Hz_10ms_modulation_mae_db": 1.5290796643805133, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7764268302766494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10090429766549293, "3000–6000 Hz_40ms_envelope_correlation": 0.9965148572389897, "3000–6000 Hz_40ms_modulation_mae_db": 1.5755455087408918, "mr_stft_db": 5.291039517879445, "band_mae_db": 0.2916466033806696, "rms_gain_error_db": -0.1556900359720202, "output_peak": 0.33458594406990866}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0246660459121195, "全帯域_10ms_rms_envelope_nrmse": 0.07582577564314162, "全帯域_10ms_envelope_correlation": 0.9953095619782713, "全帯域_10ms_modulation_mae_db": 0.30914502322775056, "全帯域_40ms_log_rms_mae_db": 0.6490564589410361, "全帯域_40ms_rms_envelope_nrmse": 0.03790647240322788, "全帯域_40ms_envelope_correlation": 0.9987315117943876, "全帯域_40ms_modulation_mae_db": 0.3625647027007839, "300–1000 Hz_10ms_log_rms_mae_db": 0.9681633033877464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10977308625911135, "300–1000 Hz_10ms_envelope_correlation": 0.9911819060843305, "300–1000 Hz_10ms_modulation_mae_db": 0.44896163690394114, "300–1000 Hz_40ms_log_rms_mae_db": 0.5610517398414225, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05429800163293071, "300–1000 Hz_40ms_envelope_correlation": 0.9976885201007175, "300–1000 Hz_40ms_modulation_mae_db": 0.43324360105270293, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1769724410629474, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13837544196797164, "1000–3000 Hz_10ms_envelope_correlation": 0.9876360434904881, "1000–3000 Hz_10ms_modulation_mae_db": 0.6341371889980486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7120310894270766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0696297301961388, "1000–3000 Hz_40ms_envelope_correlation": 0.9967326022468571, "1000–3000 Hz_40ms_modulation_mae_db": 0.6406263554753803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.29981483381564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15118455321367227, "3000–6000 Hz_10ms_envelope_correlation": 0.9835236622307633, "3000–6000 Hz_10ms_modulation_mae_db": 0.9040026061986873, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7678525594319303, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0864040861555195, "3000–6000 Hz_40ms_envelope_correlation": 0.9946635868607399, "3000–6000 Hz_40ms_modulation_mae_db": 0.977013778941736, "mr_stft_db": 5.28750503690999, "band_mae_db": 0.1370335776028219, "rms_gain_error_db": 0.014022702223642654, "output_peak": 0.3308432751814231}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9581676671709439, "全帯域_10ms_rms_envelope_nrmse": 0.07879631829691557, "全帯域_10ms_envelope_correlation": 0.9950357120099795, "全帯域_10ms_modulation_mae_db": 0.43545602788574883, "全帯域_40ms_log_rms_mae_db": 0.5982079840972447, "全帯域_40ms_rms_envelope_nrmse": 0.03939626379269815, "全帯域_40ms_envelope_correlation": 0.9988113316179835, "全帯域_40ms_modulation_mae_db": 0.35895446235635786, "300–1000 Hz_10ms_log_rms_mae_db": 0.9641165920358112, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10515998327896925, "300–1000 Hz_10ms_envelope_correlation": 0.9920321495494747, "300–1000 Hz_10ms_modulation_mae_db": 0.5076645502251189, "300–1000 Hz_40ms_log_rms_mae_db": 0.5578490342628227, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0552096446854763, "300–1000 Hz_40ms_envelope_correlation": 0.9979145215510453, "300–1000 Hz_40ms_modulation_mae_db": 0.2758212671146609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.188419238657852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11964233205849956, "1000–3000 Hz_10ms_envelope_correlation": 0.9916604926278576, "1000–3000 Hz_10ms_modulation_mae_db": 1.0595857107809161, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6992729179526241, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0710436914083768, "1000–3000 Hz_40ms_envelope_correlation": 0.9977461878130005, "1000–3000 Hz_40ms_modulation_mae_db": 1.0420654498977933, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3080049568097443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14793761109932183, "3000–6000 Hz_10ms_envelope_correlation": 0.9841196712302026, "3000–6000 Hz_10ms_modulation_mae_db": 0.12772330666695117, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7432521168048484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08366702575586053, "3000–6000 Hz_40ms_envelope_correlation": 0.9945110534459343, "3000–6000 Hz_40ms_modulation_mae_db": 0.23055937906895138, "mr_stft_db": 5.277379933631534, "band_mae_db": 0.2790447301306666, "rms_gain_error_db": -0.1326798595123733, "output_peak": 0.3321689689857362}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9904912950321922, "全帯域_10ms_rms_envelope_nrmse": 0.08160302217897629, "全帯域_10ms_envelope_correlation": 0.9947281619268191, "全帯域_10ms_modulation_mae_db": 0.45859072216388935, "全帯域_40ms_log_rms_mae_db": 0.6036408861494709, "全帯域_40ms_rms_envelope_nrmse": 0.043712556392551054, "全帯域_40ms_envelope_correlation": 0.9985644233648009, "全帯域_40ms_modulation_mae_db": 0.5735786346034323, "300–1000 Hz_10ms_log_rms_mae_db": 0.9781527104405705, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10431998053948226, "300–1000 Hz_10ms_envelope_correlation": 0.9923915347891703, "300–1000 Hz_10ms_modulation_mae_db": 0.697088416716368, "300–1000 Hz_40ms_log_rms_mae_db": 0.5946634899981355, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0621595851788639, "300–1000 Hz_40ms_envelope_correlation": 0.9975320729714342, "300–1000 Hz_40ms_modulation_mae_db": 0.5949888354404872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1928744078519586, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14643867671598615, "1000–3000 Hz_10ms_envelope_correlation": 0.9861789434550101, "1000–3000 Hz_10ms_modulation_mae_db": 0.15628423153498675, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6811560197992232, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06440865024530257, "1000–3000 Hz_40ms_envelope_correlation": 0.9972242812733998, "1000–3000 Hz_40ms_modulation_mae_db": 0.1255067928453973, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2931222491766985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15682648171229008, "3000–6000 Hz_10ms_envelope_correlation": 0.98280360688261, "3000–6000 Hz_10ms_modulation_mae_db": 1.3748424928539995, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7816149757012257, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08497070806268593, "3000–6000 Hz_40ms_envelope_correlation": 0.9956367661633408, "3000–6000 Hz_40ms_modulation_mae_db": 1.5881913887976988, "mr_stft_db": 5.279916847053621, "band_mae_db": 0.2590690278499275, "rms_gain_error_db": -0.15439718418612802, "output_peak": 0.3437289246052531}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9761396811227003, "全帯域_10ms_rms_envelope_nrmse": 0.07942808804615198, "全帯域_10ms_envelope_correlation": 0.9949162312816164, "全帯域_10ms_modulation_mae_db": 0.35634778966666847, "全帯域_40ms_log_rms_mae_db": 0.6362399786143724, "全帯域_40ms_rms_envelope_nrmse": 0.04837422753280475, "全帯域_40ms_envelope_correlation": 0.9980501918583443, "全帯域_40ms_modulation_mae_db": 0.36616849510448035, "300–1000 Hz_10ms_log_rms_mae_db": 1.0355165529670016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11252810237871438, "300–1000 Hz_10ms_envelope_correlation": 0.99074954456006, "300–1000 Hz_10ms_modulation_mae_db": 0.22823975730087878, "300–1000 Hz_40ms_log_rms_mae_db": 0.6249628941537524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0708928926803241, "300–1000 Hz_40ms_envelope_correlation": 0.996208992304405, "300–1000 Hz_40ms_modulation_mae_db": 0.27893444479406554, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1337803202656298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12305027251507815, "1000–3000 Hz_10ms_envelope_correlation": 0.9906814644178011, "1000–3000 Hz_10ms_modulation_mae_db": 0.781552776433398, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6558893170015035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0686190343540278, "1000–3000 Hz_40ms_envelope_correlation": 0.9974845464329041, "1000–3000 Hz_40ms_modulation_mae_db": 0.83254684499878, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2979345813235184, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1447665166349805, "3000–6000 Hz_10ms_envelope_correlation": 0.9859066433457057, "3000–6000 Hz_10ms_modulation_mae_db": 0.5272740081477695, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7167907642981766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08255455886648184, "3000–6000 Hz_40ms_envelope_correlation": 0.9956068643213204, "3000–6000 Hz_40ms_modulation_mae_db": 0.6793257655279152, "mr_stft_db": 5.282374218022262, "band_mae_db": 0.24999943196232968, "rms_gain_error_db": -0.10662369099109857, "output_peak": 0.32736564966120574}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0619361698247878, "全帯域_10ms_rms_envelope_nrmse": 0.07161426604263142, "全帯域_10ms_envelope_correlation": 0.9958175379376909, "全帯域_10ms_modulation_mae_db": 0.49426479916217597, "全帯域_40ms_log_rms_mae_db": 0.6604751354908418, "全帯域_40ms_rms_envelope_nrmse": 0.033002606076596594, "全帯域_40ms_envelope_correlation": 0.9990716111095725, "全帯域_40ms_modulation_mae_db": 0.3688540624663746, "300–1000 Hz_10ms_log_rms_mae_db": 0.9785615203648609, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09548419114376187, "300–1000 Hz_10ms_envelope_correlation": 0.9933038423024266, "300–1000 Hz_10ms_modulation_mae_db": 0.878760750883604, "300–1000 Hz_40ms_log_rms_mae_db": 0.5303433875997594, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04583764068522174, "300–1000 Hz_40ms_envelope_correlation": 0.9984138128319836, "300–1000 Hz_40ms_modulation_mae_db": 0.714230764720674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1439162663667741, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09738275481771337, "1000–3000 Hz_10ms_envelope_correlation": 0.9939697373862522, "1000–3000 Hz_10ms_modulation_mae_db": 0.21670140392961965, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6357623209489078, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04933290825278083, "1000–3000 Hz_40ms_envelope_correlation": 0.998468215848783, "1000–3000 Hz_40ms_modulation_mae_db": 0.3244552580724582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2964892146758176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13382268985157983, "3000–6000 Hz_10ms_envelope_correlation": 0.9875749321439046, "3000–6000 Hz_10ms_modulation_mae_db": 1.2519799480888862, "3000–6000 Hz_40ms_log_rms_mae_db": 0.75537875830664, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07255183601060426, "3000–6000 Hz_40ms_envelope_correlation": 0.996787707663878, "3000–6000 Hz_40ms_modulation_mae_db": 1.215570246524345, "mr_stft_db": 5.282971080778356, "band_mae_db": 0.17323757519363583, "rms_gain_error_db": -0.060363095757890346, "output_peak": 0.3465159370605886}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9786518137983021, "全帯域_10ms_rms_envelope_nrmse": 0.0742556614080755, "全帯域_10ms_envelope_correlation": 0.9954924000956429, "全帯域_10ms_modulation_mae_db": 0.20392950327713102, "全帯域_40ms_log_rms_mae_db": 0.6398052539266995, "全帯域_40ms_rms_envelope_nrmse": 0.04008959530796168, "全帯域_40ms_envelope_correlation": 0.9985858734978682, "全帯域_40ms_modulation_mae_db": 0.15687780942324767, "300–1000 Hz_10ms_log_rms_mae_db": 1.0052794938586656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09821192040000501, "300–1000 Hz_10ms_envelope_correlation": 0.9929487881686788, "300–1000 Hz_10ms_modulation_mae_db": 0.4393637594548376, "300–1000 Hz_40ms_log_rms_mae_db": 0.5559524510736706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053240518797431566, "300–1000 Hz_40ms_envelope_correlation": 0.9978836926055832, "300–1000 Hz_40ms_modulation_mae_db": 0.3304620212327172, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2192076776253002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13537112419808187, "1000–3000 Hz_10ms_envelope_correlation": 0.9881717383732141, "1000–3000 Hz_10ms_modulation_mae_db": 0.14530714349826587, "1000–3000 Hz_40ms_log_rms_mae_db": 0.753639188547841, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06942164214454474, "1000–3000 Hz_40ms_envelope_correlation": 0.9967893049486378, "1000–3000 Hz_40ms_modulation_mae_db": 0.15201996223245737, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3198879422305612, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15708862586899516, "3000–6000 Hz_10ms_envelope_correlation": 0.9819528459594369, "3000–6000 Hz_10ms_modulation_mae_db": 0.6233168201582392, "3000–6000 Hz_40ms_log_rms_mae_db": 0.77962895329678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08407702139188267, "3000–6000 Hz_40ms_envelope_correlation": 0.9945958778386808, "3000–6000 Hz_40ms_modulation_mae_db": 0.7010337233996488, "mr_stft_db": 5.279894358944187, "band_mae_db": 0.22336997493212205, "rms_gain_error_db": -0.03214308126451359, "output_peak": 0.3307762550044561}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48640, "output_samples": 48640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.584326558040189, "spectral_convergence": 0.2654819778691751, "spectral_convergence_status": "ok", "amplitude_floor": 9.047519636617716e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1225197340404565, "spectral_convergence": 0.26835056106031135, "spectral_convergence_status": "ok", "amplitude_floor": 7.308511681145121e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.194723939179785, "spectral_convergence": 0.2644319221173708, "spectral_convergence_status": "ok", "amplitude_floor": 4.4367419091404295e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011152155483091157, "output_power": 0.00011869101907603846, "delta_db": 0.270590429714202, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042085142829602937, "output_power": 0.0004333606770797921, "delta_db": 0.12720695684621824, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006661514585812611, "output_power": 0.000639670345675265, "delta_db": -0.1761676531587171, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015794681682280444, "output_power": 0.00015434252548033486, "delta_db": -0.10025275528088481, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.332154632901203e-05, "output_power": 4.912538137681732e-05, "delta_db": -0.35596800528033273, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2768891588196896e-05, "output_power": 3.300461117494157e-05, "delta_db": 0.03112869157910231, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17688574864324288, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.043470362196770126, "pitch": null, "output_rms_dbfs": -28.45212571539087, "output_peak": 0.36357568484389957, "reference_rms_dbfs": -28.408655353194106, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.300523410420143}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00012692463938083332, "log_rms_mae_db": 0.9209865131843162, "rms_envelope_nrmse": 0.07848525122983815, "envelope_correlation": 0.9949641509399305, "rms_gain_error_db": -0.043470328885806334, "reference_mean_power": 0.0014449383611437946, "output_mean_power": 0.0014305475182379305}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.053245386897797e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005746543771154043, "output_power": 0.0005546555032648631, "reference_active": true, "delta_db": -0.15383392656997302}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014747895972213286, "output_power": 0.0001311642265267843, "reference_active": true, "delta_db": -0.5091466297916529}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.536231970329893e-05, "output_power": 2.4458800499160518e-05, "reference_active": true, "delta_db": -0.15753818033464764}], "active_band_count": 3, "active_band_mae_db": 0.2735062455654245}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 0.00010829935323771927, "log_rms_mae_db": 0.5759442647435644, "rms_envelope_nrmse": 0.03630879484021844, "envelope_correlation": 0.9988556333336248, "rms_gain_error_db": -0.0434703958741713, "reference_mean_power": 0.0014593630016451321, "output_mean_power": 0.0014448284744528993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 9.289432894914324e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005498374692309951, "output_power": 0.0005298033168141792, "reference_active": true, "delta_db": -0.16119659098695183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.817099868184177e-05, "output_power": 5.0986859662160816e-05, "reference_active": true, "delta_db": -0.5724825573875888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4585911877580322e-06, "output_power": 1.5392185680284364e-06, "reference_active": true, "delta_db": 0.23366708346198645}], "active_band_count": 3, "active_band_mae_db": 0.32244874394550904}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00011599347403670498, "log_rms_mae_db": 0.9646614152913712, "rms_envelope_nrmse": 0.09566610933565008, "envelope_correlation": 0.9932503558404332, "rms_gain_error_db": -0.06033581735288834, "reference_mean_power": 0.0008943859828976575, "output_mean_power": 0.0008820463404714874}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 8.128232732706927e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005209748261573112, "output_power": 0.0005083871676910336, "reference_active": true, "delta_db": -0.10622158513845142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010680702813776803, "output_power": 9.273400899173429e-05, "reference_active": true, "delta_db": -0.6136079595650714}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2431239684221878e-05, "output_power": 1.9203095633138628e-05, "reference_active": true, "delta_db": -0.674820312057747}], "active_band_count": 3, "active_band_mae_db": 0.4648832855870899}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 9.903058317661472e-05, "log_rms_mae_db": 0.5709658029404485, "rms_envelope_nrmse": 0.04578490260378495, "envelope_correlation": 0.9983677755585287, "rms_gain_error_db": -0.06033582322839358, "reference_mean_power": 0.0009033149500834516, "output_mean_power": 0.0008908521154982506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 7.323650960940424e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005139241705701183, "output_power": 0.0004992259004809168, "reference_active": true, "delta_db": -0.12601934625969186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0770880999134286e-05, "output_power": 3.4175437565715897e-05, "reference_active": true, "delta_db": -0.7663601225117396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2000817341803485e-06, "output_power": 1.0851452184035497e-06, "reference_active": true, "delta_db": -0.4372296455611424}], "active_band_count": 3, "active_band_mae_db": 0.443203038110858}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 7.992748424465529e-05, "log_rms_mae_db": 1.1735544202441273, "rms_envelope_nrmse": 0.12044916385780738, "envelope_correlation": 0.9906499326732031, "rms_gain_error_db": -0.0959604585333748, "reference_mean_power": 0.00017342262158164605, "output_mean_power": 0.00016963274940984336}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.6075439307790634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.767193199363673e-05, "output_power": 9.412168931388612e-05, "reference_active": true, "delta_db": -0.16080064965541027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.270979881634546e-05, "output_power": 2.0815600286473408e-05, "reference_active": true, "delta_db": -0.3782434715202012}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.175922545645147e-06, "output_power": 4.610807851765709e-06, "reference_active": true, "delta_db": 0.4302458920740605}], "active_band_count": 3, "active_band_mae_db": 0.323096671083224}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 6.970659359658012e-05, "log_rms_mae_db": 0.6322990686832121, "rms_envelope_nrmse": 0.0535991102241229, "envelope_correlation": 0.998114916720228, "rms_gain_error_db": -0.09596062862261057, "reference_mean_power": 0.00017515363198831316, "output_mean_power": 0.00017132592466123045}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 1.4903701525512362e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.997559182587721e-05, "output_power": 9.666499761968033e-05, "reference_active": true, "delta_db": -0.14624739123207026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.524639480988942e-06, "output_power": 8.909851224355975e-06, "reference_active": true, "delta_db": -0.2897809379308363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.337301598444833e-07, "output_power": 3.276023746787202e-07, "reference_active": true, "delta_db": -0.08048415312996196}], "active_band_count": 3, "active_band_mae_db": 0.17217082743095616}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 2.3755576133519224e-05, "log_rms_mae_db": 1.393204155832888, "rms_envelope_nrmse": 0.1489503612665709, "envelope_correlation": 0.9845236642932139, "rms_gain_error_db": -0.3681159796868733, "reference_mean_power": 3.0329423269641833e-05, "output_mean_power": 2.786458338201633e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 3.170053749158532e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1550798308210337e-05, "output_power": 1.0361751683046353e-05, "reference_active": true, "delta_db": -0.4717882027029967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.1283799417647094e-06, "output_power": 5.507761332725452e-06, "reference_active": true, "delta_db": -0.46370569703278475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7104244593685833e-06, "output_power": 9.585416606129378e-07, "reference_active": true, "delta_db": -2.514929052496766}], "active_band_count": 3, "active_band_mae_db": 1.1501409840775159}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 2.1008424841575618e-05, "log_rms_mae_db": 0.7875198360859189, "rms_envelope_nrmse": 0.08073441416534424, "envelope_correlation": 0.9960784451654116, "rms_gain_error_db": -0.3681197615004056, "reference_mean_power": 3.0632047924156045e-05, "output_mean_power": 2.814258954712936e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 2.7930999787597993e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1345569218097283e-05, "output_power": 1.0218447005137925e-05, "reference_active": true, "delta_db": -0.4544139279977263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6424993904119263e-06, "output_power": 2.2798046824113306e-06, "reference_active": true, "delta_db": -0.6411725448824235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1775222693841551e-07, "output_power": 6.751950696540163e-08, "reference_active": true, "delta_db": -2.415398669216279}], "active_band_count": 3, "active_band_mae_db": 1.1703283806988096}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9196920720156401, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "86ab83fdb65603b4cd74b4d5013d66cfb1e62b6c397e68874b1396a0ca5e95a3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9896307318651978, "standard_deviation": 0.0025484669367304845, "minimum": 0.9861789434550101, "maximum": 0.9939697373862522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1814509181051474, "standard_deviation": 0.03181409373521895, "minimum": 1.1337803202656298, "maximum": 1.2228825727665906, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5605957018274401, "standard_deviation": 0.41107609710378873, "minimum": 0.14530714349826587, "maximum": 1.1681004883610615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12734563789198025, "standard_deviation": 0.015530769439754491, "minimum": 0.09738275481771337, "maximum": 0.14643867671598615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975358971703103, "standard_deviation": 0.0006080055124684709, "minimum": 0.9967326022468571, "maximum": 0.998468215848783, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6916077678413242, "standard_deviation": 0.049878595907038775, "minimum": 0.6322990686832121, "maximum": 0.7628122203702045, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5654047855034652, "standard_deviation": 0.43558337689862575, "minimum": 0.1255067928453973, "maximum": 1.2338467930744992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06407326119518163, "standard_deviation": 0.008122903011081548, "minimum": 0.04933290825278083, "maximum": 0.0710436914083768, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9842782028950512, "standard_deviation": 0.0017714432617516773, "minimum": 0.9819528459594369, "maximum": 0.9875749321439046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3203361522646029, "standard_deviation": 0.03554295798286446, "minimum": 1.2931222491766985, "maximum": 1.393204155832888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9360449788215703, "standard_deviation": 0.4794622952365092, "minimum": 0.12772330666695117, "maximum": 1.5290796643805133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1505104211320974, "standard_deviation": 0.009047846118139394, "minimum": 0.13382268985157983, "maximum": 0.16350652940936844, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955493948372869, "standard_deviation": 0.0008886884660991409, "minimum": 0.9945110534459343, "maximum": 0.996787707663878, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7635580992752711, "standard_deviation": 0.023931308708404136, "minimum": 0.7167907642981766, "maximum": 0.7875198360859189, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0171960214624995, "standard_deviation": 0.468297591648782, "minimum": 0.23055937906895138, "maximum": 1.5881913887976988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.084482993509234, "standard_deviation": 0.007878918851631432, "minimum": 0.07255183601060426, "maximum": 0.10090429766549293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920290252672425, "standard_deviation": 0.0011456574150070955, "minimum": 0.990374080843365, "maximum": 0.9933038423024266, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.986954517652516, "standard_deviation": 0.025117288115157312, "minimum": 0.9641165920358112, "maximum": 1.0355165529670016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5173120745647706, "standard_deviation": 0.1934770342781132, "minimum": 0.22823975730087878, "maximum": 0.878760750883604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10445457075842146, "standard_deviation": 0.007477906702281115, "minimum": 0.09548419114376187, "maximum": 0.11449319273167753, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976638157427093, "standard_deviation": 0.0007005372555722813, "minimum": 0.996208992304405, "maximum": 0.9984138128319836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5720955355989195, "standard_deviation": 0.028568637218771918, "minimum": 0.5303433875997594, "maximum": 0.6249628941537524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4424856561489724, "standard_deviation": 0.15351451498064564, "minimum": 0.2758212671146609, "maximum": 0.714230764720674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05592497562023765, "standard_deviation": 0.008402934581781915, "minimum": 0.04578490260378495, "maximum": 0.0708928926803241, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.223785833711927, "standard_deviation": 0.05589730061523644, "minimum": 0.1370335776028219, "maximum": 0.2916466033806696, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.285200550454942, "standard_deviation": 0.007600640607009211, "minimum": 5.277379933631534, "maximum": 5.300523410420143, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3386950799265589, "standard_deviation": 0.012057834662002992, "minimum": 0.32736564966120574, "maximum": 0.36357568484389957, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08391807595714394, "standard_deviation": 0.06265059135212342, "minimum": -0.1556900359720202, "maximum": 0.014022702223642654, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9950443112912035, "standard_deviation": 0.0005211504903676681, "minimum": 0.9940907341596766, "maximum": 0.9958175379376909, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.993635896803317, "standard_deviation": 0.045789406771279906, "minimum": 0.9209865131843162, "maximum": 1.0619361698247878, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3814343058489735, "standard_deviation": 0.11331317057903634, "minimum": 0.20392950327713102, "maximum": 0.520234335842999, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07826236774765778, "standard_deviation": 0.004477278662320417, "minimum": 0.07161426604263142, "maximum": 0.0860905591355317, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986359594218445, "standard_deviation": 0.0003108937121761892, "minimum": 0.9980501918583443, "maximum": 0.9990716111095725, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6332840950508615, "standard_deviation": 0.04014804086838019, "minimum": 0.5759442647435644, "maximum": 0.7029027984436614, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.39075958584709036, "standard_deviation": 0.14476497946005865, "minimum": 0.15687780942324767, "maximum": 0.6166297761765371, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.040538959591501014, "standard_deviation": 0.005062665221188361, "minimum": 0.033002606076596594, "maximum": 0.04837422753280475, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29095-0008", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1524548530578613, "predicted_mos": 2.1524548530578613, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "0177612b789929fd8ec090b534c0d43fe4b216c16be6227f468140d710b15434", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.352536899968982, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9066811695836346, "全帯域_10ms_rms_envelope_nrmse": 0.08045666366383387, "全帯域_10ms_envelope_correlation": 0.9947050495088405, "全帯域_10ms_modulation_mae_db": 0.285191124097161, "全帯域_40ms_log_rms_mae_db": 0.5769772750349937, "全帯域_40ms_rms_envelope_nrmse": 0.03734438594616325, "全帯域_40ms_envelope_correlation": 0.9987870959789379, "全帯域_40ms_modulation_mae_db": 0.3029193028522708, "300–1000 Hz_10ms_log_rms_mae_db": 1.0075665891047547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09941609262949207, "300–1000 Hz_10ms_envelope_correlation": 0.9927133149511868, "300–1000 Hz_10ms_modulation_mae_db": 0.5676695811246758, "300–1000 Hz_40ms_log_rms_mae_db": 0.5981283642692121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047202451239138764, "300–1000 Hz_40ms_envelope_correlation": 0.9982729890900816, "300–1000 Hz_40ms_modulation_mae_db": 0.5599366512599223, "1000–3000 Hz_10ms_log_rms_mae_db": 1.16804135460224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12052123302353075, "1000–3000 Hz_10ms_envelope_correlation": 0.9906465092506744, "1000–3000 Hz_10ms_modulation_mae_db": 0.3254517349627199, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6388871123199394, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05394748187923339, "1000–3000 Hz_40ms_envelope_correlation": 0.9980988669716628, "1000–3000 Hz_40ms_modulation_mae_db": 0.2526333130576739, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3327183292629199, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14626712584153082, "3000–6000 Hz_10ms_envelope_correlation": 0.9857374337178791, "3000–6000 Hz_10ms_modulation_mae_db": 1.145134404733058, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7665389698003044, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08433445357686399, "3000–6000 Hz_40ms_envelope_correlation": 0.9964226333085241, "3000–6000 Hz_40ms_modulation_mae_db": 1.228439413263857, "mr_stft_db": 5.248621630339872, "band_mae_db": 0.18632263807151883, "rms_gain_error_db": -0.044346543855953706, "output_peak": 0.36268893075756936}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0230879440232044, "全帯域_10ms_rms_envelope_nrmse": 0.08728311579796393, "全帯域_10ms_envelope_correlation": 0.9939763483732517, "全帯域_10ms_modulation_mae_db": 0.5477040471473942, "全帯域_40ms_log_rms_mae_db": 0.7067653127371994, "全帯域_40ms_rms_envelope_nrmse": 0.04688851579618013, "全帯域_40ms_envelope_correlation": 0.9983759340608008, "全帯域_40ms_modulation_mae_db": 0.6300442320287835, "300–1000 Hz_10ms_log_rms_mae_db": 0.9845051886588473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1147540718843195, "300–1000 Hz_10ms_envelope_correlation": 0.9903367193463061, "300–1000 Hz_10ms_modulation_mae_db": 0.5298569737438267, "300–1000 Hz_40ms_log_rms_mae_db": 0.5739753660915017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061112315451316436, "300–1000 Hz_40ms_envelope_correlation": 0.9972043684334444, "300–1000 Hz_40ms_modulation_mae_db": 0.5111065867835038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2043431500234714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13051961345732532, "1000–3000 Hz_10ms_envelope_correlation": 0.9897314777976606, "1000–3000 Hz_10ms_modulation_mae_db": 1.180147826603552, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7721419464092178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07661950926655604, "1000–3000 Hz_40ms_envelope_correlation": 0.997078437001661, "1000–3000 Hz_40ms_modulation_mae_db": 1.2153245394632939, "3000–6000 Hz_10ms_log_rms_mae_db": 1.316105411054842, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15852649709420932, "3000–6000 Hz_10ms_envelope_correlation": 0.9852917428948166, "3000–6000 Hz_10ms_modulation_mae_db": 1.4538699447243333, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7627587527738974, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09851424265474218, "3000–6000 Hz_40ms_envelope_correlation": 0.9972421032863461, "3000–6000 Hz_40ms_modulation_mae_db": 1.5164058157580225, "mr_stft_db": 5.2345020684192525, "band_mae_db": 0.3240570062650762, "rms_gain_error_db": -0.17586495030392124, "output_peak": 0.339277847781994}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0237959938151382, "全帯域_10ms_rms_envelope_nrmse": 0.07581496672991095, "全帯域_10ms_envelope_correlation": 0.9953116718323113, "全帯域_10ms_modulation_mae_db": 0.22096299050837173, "全帯域_40ms_log_rms_mae_db": 0.6578837290068764, "全帯域_40ms_rms_envelope_nrmse": 0.038180527745055926, "全帯域_40ms_envelope_correlation": 0.9987137762025978, "全帯域_40ms_modulation_mae_db": 0.2661237464329263, "300–1000 Hz_10ms_log_rms_mae_db": 0.9665579371515346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10950226636163264, "300–1000 Hz_10ms_envelope_correlation": 0.9912525410727749, "300–1000 Hz_10ms_modulation_mae_db": 0.37895016077751964, "300–1000 Hz_40ms_log_rms_mae_db": 0.5450529768435987, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05442295756347446, "300–1000 Hz_40ms_envelope_correlation": 0.9976958389553288, "300–1000 Hz_40ms_modulation_mae_db": 0.36674329697213043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1796581646514892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12943859957460396, "1000–3000 Hz_10ms_envelope_correlation": 0.9892001526449393, "1000–3000 Hz_10ms_modulation_mae_db": 0.5758495581464059, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7122299976401358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06652346872183161, "1000–3000 Hz_40ms_envelope_correlation": 0.997076033031173, "1000–3000 Hz_40ms_modulation_mae_db": 0.5830952241061664, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2973517373724102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15194195270575292, "3000–6000 Hz_10ms_envelope_correlation": 0.9836565310225377, "3000–6000 Hz_10ms_modulation_mae_db": 0.868550163606087, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7872818444272681, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08675405739186809, "3000–6000 Hz_40ms_envelope_correlation": 0.995060925970848, "3000–6000 Hz_40ms_modulation_mae_db": 0.9343585031593197, "mr_stft_db": 5.238100999921517, "band_mae_db": 0.15879210957035533, "rms_gain_error_db": 0.015123223907548356, "output_peak": 0.33185560383086604}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9643030631309961, "全帯域_10ms_rms_envelope_nrmse": 0.0788378968890361, "全帯域_10ms_envelope_correlation": 0.9950993552534233, "全帯域_10ms_modulation_mae_db": 0.44924452526292624, "全帯域_40ms_log_rms_mae_db": 0.6168193938856926, "全帯域_40ms_rms_envelope_nrmse": 0.04097434367497762, "全帯域_40ms_envelope_correlation": 0.9987845078314451, "全帯域_40ms_modulation_mae_db": 0.4010024101420427, "300–1000 Hz_10ms_log_rms_mae_db": 1.0073719433207082, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10572710476163469, "300–1000 Hz_10ms_envelope_correlation": 0.9919985082160675, "300–1000 Hz_10ms_modulation_mae_db": 0.5716270076748451, "300–1000 Hz_40ms_log_rms_mae_db": 0.5921064884124365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057492819192267904, "300–1000 Hz_40ms_envelope_correlation": 0.997781591356775, "300–1000 Hz_40ms_modulation_mae_db": 0.33476709977929175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.193225291434216, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1149485680278379, "1000–3000 Hz_10ms_envelope_correlation": 0.9925255969358755, "1000–3000 Hz_10ms_modulation_mae_db": 0.9156190011620636, "1000–3000 Hz_40ms_log_rms_mae_db": 0.736666851985064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07205157301911065, "1000–3000 Hz_40ms_envelope_correlation": 0.9978039739188993, "1000–3000 Hz_40ms_modulation_mae_db": 0.9482359645639061, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2743615160177277, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13793361645924193, "3000–6000 Hz_10ms_envelope_correlation": 0.9861405825341609, "3000–6000 Hz_10ms_modulation_mae_db": 0.2778811482374768, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7195860657037337, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07656183417700284, "3000–6000 Hz_40ms_envelope_correlation": 0.9955106787723249, "3000–6000 Hz_40ms_modulation_mae_db": 0.40046226750911185, "mr_stft_db": 5.233834837627901, "band_mae_db": 0.28506580724829994, "rms_gain_error_db": -0.15962479529395082, "output_peak": 0.3278856237923871}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9867415468825456, "全帯域_10ms_rms_envelope_nrmse": 0.08226706031934716, "全帯域_10ms_envelope_correlation": 0.9946160737975203, "全帯域_10ms_modulation_mae_db": 0.5001984570869072, "全帯域_40ms_log_rms_mae_db": 0.6233796713010363, "全帯域_40ms_rms_envelope_nrmse": 0.045539396608703334, "全帯域_40ms_envelope_correlation": 0.9983874892996654, "全帯域_40ms_modulation_mae_db": 0.583303716455735, "300–1000 Hz_10ms_log_rms_mae_db": 1.012762956555055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10786133552512199, "300–1000 Hz_10ms_envelope_correlation": 0.9916920931197625, "300–1000 Hz_10ms_modulation_mae_db": 0.7138804813631051, "300–1000 Hz_40ms_log_rms_mae_db": 0.6024467343444436, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06392465219182099, "300–1000 Hz_40ms_envelope_correlation": 0.9971846374428853, "300–1000 Hz_40ms_modulation_mae_db": 0.6288803103272633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.203572840372745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12153118573948987, "1000–3000 Hz_10ms_envelope_correlation": 0.9904995875111836, "1000–3000 Hz_10ms_modulation_mae_db": 0.2053608707213177, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7000718131727844, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05175374979576479, "1000–3000 Hz_40ms_envelope_correlation": 0.9982962069111091, "1000–3000 Hz_40ms_modulation_mae_db": 0.14007085241501152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2131960179444718, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15244969174791165, "3000–6000 Hz_10ms_envelope_correlation": 0.9846636560642247, "3000–6000 Hz_10ms_modulation_mae_db": 1.479333027547051, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7467840632486544, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08825519678454972, "3000–6000 Hz_40ms_envelope_correlation": 0.9962893044453608, "3000–6000 Hz_40ms_modulation_mae_db": 1.7870143355925283, "mr_stft_db": 5.231690243754917, "band_mae_db": 0.28771357920468293, "rms_gain_error_db": -0.14518471928026497, "output_peak": 0.34543805577399056}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9866190186289695, "全帯域_10ms_rms_envelope_nrmse": 0.08075279090285052, "全帯域_10ms_envelope_correlation": 0.9947498464058625, "全帯域_10ms_modulation_mae_db": 0.3455887424521123, "全帯域_40ms_log_rms_mae_db": 0.6451042959545509, "全帯域_40ms_rms_envelope_nrmse": 0.04936122618157025, "全帯域_40ms_envelope_correlation": 0.9979770715466749, "全帯域_40ms_modulation_mae_db": 0.33299031467722834, "300–1000 Hz_10ms_log_rms_mae_db": 1.0454468679029858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11211478930838453, "300–1000 Hz_10ms_envelope_correlation": 0.9907961508638623, "300–1000 Hz_10ms_modulation_mae_db": 0.2378391875703925, "300–1000 Hz_40ms_log_rms_mae_db": 0.6181187618416847, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07070072084095327, "300–1000 Hz_40ms_envelope_correlation": 0.9961975492774695, "300–1000 Hz_40ms_modulation_mae_db": 0.23032522327342342, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1564562197207202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11156251100800735, "1000–3000 Hz_10ms_envelope_correlation": 0.9927834992588298, "1000–3000 Hz_10ms_modulation_mae_db": 0.742321333605147, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6859876968122453, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06661185257471147, "1000–3000 Hz_40ms_envelope_correlation": 0.9980432715603386, "1000–3000 Hz_40ms_modulation_mae_db": 0.8283196894785023, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2505111960227908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14205140167893893, "3000–6000 Hz_10ms_envelope_correlation": 0.9861853722895111, "3000–6000 Hz_10ms_modulation_mae_db": 0.5598574289243182, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7004504161655165, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08180320256921422, "3000–6000 Hz_40ms_envelope_correlation": 0.9954972638232727, "3000–6000 Hz_40ms_modulation_mae_db": 0.8023052566044182, "mr_stft_db": 5.237870458578176, "band_mae_db": 0.24053941914150687, "rms_gain_error_db": -0.11246557842028825, "output_peak": 0.31753777140350814}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0631442852152981, "全帯域_10ms_rms_envelope_nrmse": 0.07214665972065773, "全帯域_10ms_envelope_correlation": 0.9957497850324307, "全帯域_10ms_modulation_mae_db": 0.47695766060829453, "全帯域_40ms_log_rms_mae_db": 0.6452095861467806, "全帯域_40ms_rms_envelope_nrmse": 0.03379310969513135, "全帯域_40ms_envelope_correlation": 0.9990177890283433, "全帯域_40ms_modulation_mae_db": 0.32432767137949453, "300–1000 Hz_10ms_log_rms_mae_db": 1.0128422263967736, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09483986923126196, "300–1000 Hz_10ms_envelope_correlation": 0.9933753800573998, "300–1000 Hz_10ms_modulation_mae_db": 0.8153327678464968, "300–1000 Hz_40ms_log_rms_mae_db": 0.5325641736189426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04430103900944511, "300–1000 Hz_40ms_envelope_correlation": 0.9984879327062987, "300–1000 Hz_40ms_modulation_mae_db": 0.6547247589893687, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1416397094257802, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09234707937180336, "1000–3000 Hz_10ms_envelope_correlation": 0.9946102536278928, "1000–3000 Hz_10ms_modulation_mae_db": 0.22003185285267965, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6538045218155244, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05183216262435579, "1000–3000 Hz_40ms_envelope_correlation": 0.9983254174535705, "1000–3000 Hz_40ms_modulation_mae_db": 0.32702238479376083, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2557521508789615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1335968039670761, "3000–6000 Hz_10ms_envelope_correlation": 0.9882178982586927, "3000–6000 Hz_10ms_modulation_mae_db": 1.3887709118717135, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6973052889802328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07497360131697255, "3000–6000 Hz_40ms_envelope_correlation": 0.9973262272705234, "3000–6000 Hz_40ms_modulation_mae_db": 1.4371998510637602, "mr_stft_db": 5.2374467924226344, "band_mae_db": 0.18347519797508158, "rms_gain_error_db": -0.05471569229686602, "output_peak": 0.34591545173413324}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9701958100286717, "全帯域_10ms_rms_envelope_nrmse": 0.07653371454823886, "全帯域_10ms_envelope_correlation": 0.9952120338166658, "全帯域_10ms_modulation_mae_db": 0.25404046445365375, "全帯域_40ms_log_rms_mae_db": 0.6287684044982983, "全帯域_40ms_rms_envelope_nrmse": 0.040920975515533434, "全帯域_40ms_envelope_correlation": 0.9985336182237384, "全帯域_40ms_modulation_mae_db": 0.146830318345463, "300–1000 Hz_10ms_log_rms_mae_db": 1.0301629678362807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10009088264270304, "300–1000 Hz_10ms_envelope_correlation": 0.9926684075153617, "300–1000 Hz_10ms_modulation_mae_db": 0.574373326585345, "300–1000 Hz_40ms_log_rms_mae_db": 0.5666925114508561, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05311502691422143, "300–1000 Hz_40ms_envelope_correlation": 0.9978898818572363, "300–1000 Hz_40ms_modulation_mae_db": 0.3524211533817973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.218175141742531, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12802196765109994, "1000–3000 Hz_10ms_envelope_correlation": 0.989478426323988, "1000–3000 Hz_10ms_modulation_mae_db": 0.07053626090421998, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7490945688499155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07295107139810679, "1000–3000 Hz_40ms_envelope_correlation": 0.9965288581151923, "1000–3000 Hz_40ms_modulation_mae_db": 0.08173604913679335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.275037976918469, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15318296486111285, "3000–6000 Hz_10ms_envelope_correlation": 0.9829220362769309, "3000–6000 Hz_10ms_modulation_mae_db": 0.5905911042402415, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7736982374540139, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08478151042819752, "3000–6000 Hz_40ms_envelope_correlation": 0.994636295806033, "3000–6000 Hz_40ms_modulation_mae_db": 0.7515916391968175, "mr_stft_db": 5.230743667187877, "band_mae_db": 0.23180184219592126, "rms_gain_error_db": -0.04261247011814677, "output_peak": 0.331377075897412}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48640, "output_samples": 48640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.528116966990282, "spectral_convergence": 0.264675635850955, "spectral_convergence_status": "ok", "amplitude_floor": 9.047519636617716e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.067422078270768, "spectral_convergence": 0.2660809500502988, "spectral_convergence_status": "ok", "amplitude_floor": 7.308511681145121e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.150325845758564, "spectral_convergence": 0.2618516456546799, "spectral_convergence_status": "ok", "amplitude_floor": 4.4367419091404295e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011152155483091157, "output_power": 0.00011858090447263289, "delta_db": 0.26655942871298577, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042085142829602937, "output_power": 0.00043512000457765986, "delta_db": 0.14480244942272733, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006661514585812611, "output_power": 0.0006383095495132772, "delta_db": -0.18541641348312915, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015794681682280444, "output_power": 0.0001543698033984433, "delta_db": -0.09948526734892153, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.332154632901203e-05, "output_power": 4.860811643688984e-05, "delta_db": -0.4019394299642984, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2768891588196896e-05, "output_power": 3.2918120866700585e-05, "delta_db": 0.019732839497050736, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18632263807151883, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.044346543855953706, "pitch": null, "output_rms_dbfs": -28.453001897050058, "output_peak": 0.36268893075756936, "reference_rms_dbfs": -28.408655353194106, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.248621630339872}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00012692463938083332, "log_rms_mae_db": 0.9066811695836346, "rms_envelope_nrmse": 0.08045666366383387, "envelope_correlation": 0.9947050495088405, "rms_gain_error_db": -0.044346522195162996, "reference_mean_power": 0.0014449383611437946, "output_mean_power": 0.001430258933007414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.053245386897797e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005746543771154043, "output_power": 0.0005555135698622258, "reference_active": true, "delta_db": -0.1471204687450684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014747895972213286, "output_power": 0.00013164604165624656, "reference_active": true, "delta_db": -0.49322260308572363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.536231970329893e-05, "output_power": 2.4136035874309217e-05, "reference_active": true, "delta_db": -0.21523030046069097}], "active_band_count": 3, "active_band_mae_db": 0.285191124097161}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 0.00010829935323771927, "log_rms_mae_db": 0.5769772750349937, "rms_envelope_nrmse": 0.03734438594616325, "envelope_correlation": 0.9987870959789379, "rms_gain_error_db": -0.044346615258457386, "reference_mean_power": 0.0014593630016451321, "output_mean_power": 0.0014445369996433407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 9.289432894914324e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005498374692309951, "output_power": 0.000530238727112337, "reference_active": true, "delta_db": -0.15762887777082768}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.817099868184177e-05, "output_power": 5.1074928112915554e-05, "reference_active": true, "delta_db": -0.5649875580309792}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4585911877580322e-06, "output_power": 1.522466455747807e-06, "reference_active": true, "delta_db": 0.18614147275500564}], "active_band_count": 3, "active_band_mae_db": 0.3029193028522708}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00011599347403670498, "log_rms_mae_db": 1.0075665891047547, "rms_envelope_nrmse": 0.09941609262949207, "envelope_correlation": 0.9927133149511868, "rms_gain_error_db": -0.06705156736118337, "reference_mean_power": 0.0008943859828976575, "output_mean_power": 0.0008806834345820773}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 8.128232732706927e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005209748261573112, "output_power": 0.000507447634135144, "reference_active": true, "delta_db": -0.11425506370207116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010680702813776803, "output_power": 9.242903025467116e-05, "reference_active": true, "delta_db": -0.6279143462951399}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2431239684221878e-05, "output_power": 1.7979157735955254e-05, "reference_active": true, "delta_db": -0.9608393333768162}], "active_band_count": 3, "active_band_mae_db": 0.5676695811246758}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 9.903058317661472e-05, "log_rms_mae_db": 0.5981283642692121, "rms_envelope_nrmse": 0.047202451239138764, "envelope_correlation": 0.9982729890900816, "rms_gain_error_db": -0.06705157345512218, "reference_mean_power": 0.0009033149500834516, "output_mean_power": 0.000889475603201156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 7.323650960940424e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005139241705701183, "output_power": 0.0004978116588162269, "reference_active": true, "delta_db": -0.13833980013314684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0770880999134286e-05, "output_power": 3.412586062055865e-05, "reference_active": true, "delta_db": -0.772664833568335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2000817341803485e-06, "output_power": 1.0053801204951215e-06, "reference_active": true, "delta_db": -0.768805320078285}], "active_band_count": 3, "active_band_mae_db": 0.5599366512599223}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 7.992748424465529e-05, "log_rms_mae_db": 1.16804135460224, "rms_envelope_nrmse": 0.12052123302353075, "envelope_correlation": 0.9906465092506744, "rms_gain_error_db": -0.1008917433919483, "reference_mean_power": 0.00017342262158164605, "output_mean_power": 0.00016944024577266627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.6075439307790634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.767193199363673e-05, "output_power": 9.4709021191965e-05, "reference_active": true, "delta_db": -0.1337843009292136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.270979881634546e-05, "output_power": 2.07151190153358e-05, "reference_active": true, "delta_db": -0.39925854059321786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.175922545645147e-06, "output_power": 4.624701123138335e-06, "reference_active": true, "delta_db": 0.44331236336572827}], "active_band_count": 3, "active_band_mae_db": 0.3254517349627199}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 6.970659359658012e-05, "log_rms_mae_db": 0.6388871123199394, "rms_envelope_nrmse": 0.05394748187923339, "envelope_correlation": 0.9980988669716628, "rms_gain_error_db": -0.10089215715569097, "reference_mean_power": 0.00017515363198831316, "output_mean_power": 0.00017113148996306034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 1.4903701525512362e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.997559182587721e-05, "output_power": 9.717464707412294e-05, "reference_active": true, "delta_db": -0.12341011703592339}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.524639480988942e-06, "output_power": 8.818501990453872e-06, "reference_active": true, "delta_db": -0.3345372882165938}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.337301598444833e-07, "output_power": 3.114585110379789e-07, "reference_active": true, "delta_db": -0.2999525339205045}], "active_band_count": 3, "active_band_mae_db": 0.2526333130576739}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 2.3755576133519224e-05, "log_rms_mae_db": 1.3327183292629199, "rms_envelope_nrmse": 0.14626712584153082, "envelope_correlation": 0.9857374337178791, "rms_gain_error_db": -0.47487458291066464, "reference_mean_power": 3.0329423269641833e-05, "output_mean_power": 2.718796447324095e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 3.170053749158532e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1550798308210337e-05, "output_power": 1.0452912342338705e-05, "reference_active": true, "delta_db": -0.4337469214373386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.1283799417647094e-06, "output_power": 5.318701350178542e-06, "reference_active": true, "delta_db": -0.6154007747331814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7104244593685833e-06, "output_power": 9.873664056039066e-07, "reference_active": true, "delta_db": -2.3862555180286544}], "active_band_count": 3, "active_band_mae_db": 1.145134404733058}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 2.1008424841575618e-05, "log_rms_mae_db": 0.7665389698003044, "rms_envelope_nrmse": 0.08433445357686399, "envelope_correlation": 0.9964226333085241, "rms_gain_error_db": -0.4748786785870708, "reference_mean_power": 3.0632047924156045e-05, "output_mean_power": 2.745921799748977e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 2.7930999787597993e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1345569218097283e-05, "output_power": 1.0263836765220047e-05, "reference_active": true, "delta_db": -0.4351655328204369}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6424993904119263e-06, "output_power": 2.196889589829207e-06, "reference_active": true, "delta_db": -0.8020666487816586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1775222693841551e-07, "output_power": 6.701322572228698e-08, "reference_active": true, "delta_db": -2.4480860581894754}], "active_band_count": 3, "active_band_mae_db": 1.228439413263857}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9224501942239992, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "79782708fb3314c6c2ec4b8f9e698b492e5d82f99e963ea2d1a058cfc387d24e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9911844379188806, "standard_deviation": 0.0019201659901107652, "minimum": 0.9892001526449393, "maximum": 0.9946102536278928, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.183138983996649, "standard_deviation": 0.026378127591323852, "minimum": 1.1416397094257802, "maximum": 1.218175141742531, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5294148048697633, "standard_deviation": 0.3915156881243376, "minimum": 0.07053626090421998, "maximum": 1.180147826603552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11861134473171231, "standard_deviation": 0.012607723851137334, "minimum": 0.09234707937180336, "maximum": 0.13051961345732532, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9976563831204508, "standard_deviation": 0.0006725622296334146, "minimum": 0.9965288581151923, "maximum": 0.9983254174535705, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7061105636256033, "standard_deviation": 0.04609805254564435, "minimum": 0.6388871123199394, "maximum": 0.7721419464092178, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5470547521268886, "standard_deviation": 0.41493465422157666, "minimum": 0.08173604913679335, "maximum": 1.2153245394632939, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06403635865995881, "standard_deviation": 0.010116892839368246, "minimum": 0.05175374979576479, "maximum": 0.07661950926655604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9853519066323442, "standard_deviation": 0.0016446923610744725, "minimum": 0.9829220362769309, "maximum": 0.9882178982586927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2768792919340743, "standard_deviation": 0.03832703098159722, "minimum": 1.2131960179444718, "maximum": 1.3327183292629199, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9704985167355349, "standard_deviation": 0.4630763617400781, "minimum": 0.2778811482374768, "maximum": 1.479333027547051, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14699375679447182, "standard_deviation": 0.008553041387961537, "minimum": 0.1335968039670761, "maximum": 0.15852649709420932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9959981790854041, "standard_deviation": 0.0009854452862166584, "minimum": 0.994636295806033, "maximum": 0.9973262272705234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7443004548192027, "standard_deviation": 0.03442430750579887, "minimum": 0.6973052889802328, "maximum": 0.7872818444272681, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1072221352684795, "standard_deviation": 0.46321336336787916, "minimum": 0.40046226750911185, "maximum": 1.7870143355925283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08449726236242638, "standard_deviation": 0.007336088228445186, "minimum": 0.07497360131697255, "maximum": 0.09851424265474218, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918541393928402, "standard_deviation": 0.001038289361707216, "minimum": 0.9903367193463061, "maximum": 0.9933753800573998, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0084020846158674, "standard_deviation": 0.024519198389545197, "minimum": 0.9665579371515346, "maximum": 1.0454468679029858, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5486911858357759, "standard_deviation": 0.17957360143930848, "minimum": 0.2378391875703925, "maximum": 0.8153327678464968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1055383015430688, "standard_deviation": 0.006875817182787423, "minimum": 0.09483986923126196, "maximum": 0.1147540718843195, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99758934863994, "standard_deviation": 0.0007233396387903661, "minimum": 0.9961975492774695, "maximum": 0.9984879327062987, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5786356721090845, "standard_deviation": 0.029524921512522426, "minimum": 0.5325641736189426, "maximum": 0.6181187618416847, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45486313509583765, "standard_deviation": 0.15470257762673068, "minimum": 0.23032522327342342, "maximum": 0.6547247589893687, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0565339978003298, "standard_deviation": 0.008701945922148777, "minimum": 0.04430103900944511, "maximum": 0.07070072084095327, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23722094995905538, "standard_deviation": 0.058614774627770504, "minimum": 0.15879210957035533, "maximum": 0.3240570062650762, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.2366013372815186, "standard_deviation": 0.00560042709017153, "minimum": 5.230743667187877, "maximum": 5.248621630339872, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.33774704512148257, "standard_deviation": 0.013790594572026185, "minimum": 0.31753777140350814, "maximum": 0.36268893075756936, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08996144070773043, "standard_deviation": 0.06802954819433024, "minimum": -0.17586495030392124, "maximum": 0.015123223907548356, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949275205025383, "standard_deviation": 0.0005367834234823175, "minimum": 0.9939763483732517, "maximum": 0.9957497850324307, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9905711039135572, "standard_deviation": 0.04720937101045993, "minimum": 0.9066811695836346, "maximum": 1.0631442852152981, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38498600145210266, "standard_deviation": 0.12415454634126963, "minimum": 0.22096299050837173, "maximum": 0.5477040471473942, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07926160857147989, "standard_deviation": 0.004584942347950206, "minimum": 0.07214665972065773, "maximum": 0.08728311579796393, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985721602715255, "standard_deviation": 0.00032473133485921775, "minimum": 0.9979770715466749, "maximum": 0.9990177890283433, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6376134585706785, "standard_deviation": 0.03721875901547843, "minimum": 0.5769772750349937, "maximum": 0.7067653127371994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.373442714039243, "standard_deviation": 0.16148343146513716, "minimum": 0.146830318345463, "maximum": 0.6300442320287835, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041625310145414415, "standard_deviation": 0.0052823405226181605, "minimum": 0.03379310969513135, "maximum": 0.04936122618157025, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29095-0008", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.159803628921509, "predicted_mos": 2.159803628921509, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "4e181f59dd62d6472e10631afb6d495161df02046c89bd0bef880ef973a48650", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7879186999052763, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9960902395700382, "全帯域_10ms_rms_envelope_nrmse": 0.09798075308570478, "全帯域_10ms_envelope_correlation": 0.9923897567445545, "全帯域_10ms_modulation_mae_db": 0.567567089424122, "全帯域_40ms_log_rms_mae_db": 0.6960467044300289, "全帯域_40ms_rms_envelope_nrmse": 0.05881346333315142, "全帯域_40ms_envelope_correlation": 0.9973174507641216, "全帯域_40ms_modulation_mae_db": 0.6665650205513233, "300–1000 Hz_10ms_log_rms_mae_db": 1.1350533824138782, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12708934914945322, "300–1000 Hz_10ms_envelope_correlation": 0.9883682200752563, "300–1000 Hz_10ms_modulation_mae_db": 0.773945598504218, "300–1000 Hz_40ms_log_rms_mae_db": 0.7240744927607788, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07923191879831983, "300–1000 Hz_40ms_envelope_correlation": 0.9955310951767552, "300–1000 Hz_40ms_modulation_mae_db": 1.0209545223136471, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2681832847556163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11933478253949135, "1000–3000 Hz_10ms_envelope_correlation": 0.9910291841697326, "1000–3000 Hz_10ms_modulation_mae_db": 0.532802765855883, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7395775508761001, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07453767367577778, "1000–3000 Hz_40ms_envelope_correlation": 0.9965615493594054, "1000–3000 Hz_40ms_modulation_mae_db": 0.48010846168028715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5983523364373282, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17871055914977363, "3000–6000 Hz_10ms_envelope_correlation": 0.9766736255312807, "3000–6000 Hz_10ms_modulation_mae_db": 0.49191004626385887, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8571910840903176, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09117939592110937, "3000–6000 Hz_40ms_envelope_correlation": 0.9936145730195698, "3000–6000 Hz_40ms_modulation_mae_db": 0.6222941411354957, "mr_stft_db": 5.513695204786011, "band_mae_db": 0.20003216199350604, "rms_gain_error_db": -0.19454156129403688, "output_peak": 0.33734834364549426}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0082883700658272, "全帯域_10ms_rms_envelope_nrmse": 0.0851344136772144, "全帯域_10ms_envelope_correlation": 0.9940938401851678, "全帯域_10ms_modulation_mae_db": 0.2968523969203218, "全帯域_40ms_log_rms_mae_db": 0.6680499874019062, "全帯域_40ms_rms_envelope_nrmse": 0.05292896285301051, "全帯域_40ms_envelope_correlation": 0.997551308062789, "全帯域_40ms_modulation_mae_db": 0.29261697831310257, "300–1000 Hz_10ms_log_rms_mae_db": 1.109729431592883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10508069203878807, "300–1000 Hz_10ms_envelope_correlation": 0.9918843237400004, "300–1000 Hz_10ms_modulation_mae_db": 0.09382035988609679, "300–1000 Hz_40ms_log_rms_mae_db": 0.6926226184309776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06258323432338073, "300–1000 Hz_40ms_envelope_correlation": 0.9969113596302698, "300–1000 Hz_40ms_modulation_mae_db": 0.1424217486533079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2658006308370169, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11738285332254772, "1000–3000 Hz_10ms_envelope_correlation": 0.9914617971985924, "1000–3000 Hz_10ms_modulation_mae_db": 0.8215197604283883, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7621304720631836, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06305944625365108, "1000–3000 Hz_40ms_envelope_correlation": 0.99779082864699, "1000–3000 Hz_40ms_modulation_mae_db": 0.7668904902564382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6565866635430109, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17194141422938275, "3000–6000 Hz_10ms_envelope_correlation": 0.9786879063869565, "3000–6000 Hz_10ms_modulation_mae_db": 1.2254417317884498, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9496032466853461, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09725345755031481, "3000–6000 Hz_40ms_envelope_correlation": 0.9931261178346837, "3000–6000 Hz_40ms_modulation_mae_db": 1.4155348593577468, "mr_stft_db": 5.511023026404871, "band_mae_db": 0.1919310054724316, "rms_gain_error_db": 0.006798534467298325, "output_peak": 0.3610327759368158}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9345344321351305, "全帯域_10ms_rms_envelope_nrmse": 0.08315384928866822, "全帯域_10ms_envelope_correlation": 0.9944585226560234, "全帯域_10ms_modulation_mae_db": 0.5156094747295229, "全帯域_40ms_log_rms_mae_db": 0.5874549351578089, "全帯域_40ms_rms_envelope_nrmse": 0.04307558752543118, "全帯域_40ms_envelope_correlation": 0.9985444466293618, "全帯域_40ms_modulation_mae_db": 0.5270470473831523, "300–1000 Hz_10ms_log_rms_mae_db": 1.0619677870509043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10400188593021907, "300–1000 Hz_10ms_envelope_correlation": 0.9926344213981193, "300–1000 Hz_10ms_modulation_mae_db": 0.8849746330163496, "300–1000 Hz_40ms_log_rms_mae_db": 0.6094894665365898, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0625060060139519, "300–1000 Hz_40ms_envelope_correlation": 0.9977369483053751, "300–1000 Hz_40ms_modulation_mae_db": 0.9561625356676894, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4753814103347687, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14623042784273066, "1000–3000 Hz_10ms_envelope_correlation": 0.9861900215104191, "1000–3000 Hz_10ms_modulation_mae_db": 0.48519432938660917, "1000–3000 Hz_40ms_log_rms_mae_db": 0.84934189754067, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08900872326246745, "1000–3000 Hz_40ms_envelope_correlation": 0.9946213224145923, "1000–3000 Hz_40ms_modulation_mae_db": 0.6321294755716477, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6444278305113798, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.205162534017137, "3000–6000 Hz_10ms_envelope_correlation": 0.9690114873455172, "3000–6000 Hz_10ms_modulation_mae_db": 0.8515338321065936, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9119540788277323, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10203032891141071, "3000–6000 Hz_40ms_envelope_correlation": 0.992127808399539, "3000–6000 Hz_40ms_modulation_mae_db": 0.8471804102868195, "mr_stft_db": 5.517411648291801, "band_mae_db": 0.2670302959296175, "rms_gain_error_db": -0.13210137670703873, "output_peak": 0.3456157716916367}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0203471808151725, "全帯域_10ms_rms_envelope_nrmse": 0.07805782315209944, "全帯域_10ms_envelope_correlation": 0.9950831784189609, "全帯域_10ms_modulation_mae_db": 0.49623475745411166, "全帯域_40ms_log_rms_mae_db": 0.6493522512903753, "全帯域_40ms_rms_envelope_nrmse": 0.03895812839974181, "全帯域_40ms_envelope_correlation": 0.9987694915993028, "全帯域_40ms_modulation_mae_db": 0.7276303576887079, "300–1000 Hz_10ms_log_rms_mae_db": 1.0476922421379535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10732456821989213, "300–1000 Hz_10ms_envelope_correlation": 0.9916277883196326, "300–1000 Hz_10ms_modulation_mae_db": 0.6103858242414252, "300–1000 Hz_40ms_log_rms_mae_db": 0.6002333607811976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05452976476835388, "300–1000 Hz_40ms_envelope_correlation": 0.9978947143234554, "300–1000 Hz_40ms_modulation_mae_db": 0.9255599889488497, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3178987818246144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12347219375657278, "1000–3000 Hz_10ms_envelope_correlation": 0.9902022108396024, "1000–3000 Hz_10ms_modulation_mae_db": 0.43774928378355654, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7737455174952456, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06427499490559314, "1000–3000 Hz_40ms_envelope_correlation": 0.997302103733863, "1000–3000 Hz_40ms_modulation_mae_db": 0.5840110423259193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6651522051080592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.179857430826428, "3000–6000 Hz_10ms_envelope_correlation": 0.9767656810482512, "3000–6000 Hz_10ms_modulation_mae_db": 1.210751246665924, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9068968269099956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09811009496411618, "3000–6000 Hz_40ms_envelope_correlation": 0.9932987239221719, "3000–6000 Hz_40ms_modulation_mae_db": 1.3680358397636063, "mr_stft_db": 5.527341596289851, "band_mae_db": 0.16554924692788534, "rms_gain_error_db": -0.0995728701956338, "output_peak": 0.3396837087978167}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9474832630079925, "全帯域_10ms_rms_envelope_nrmse": 0.08251418090956014, "全帯域_10ms_envelope_correlation": 0.9944543972230543, "全帯域_10ms_modulation_mae_db": 0.545158530866926, "全帯域_40ms_log_rms_mae_db": 0.6034328951017481, "全帯域_40ms_rms_envelope_nrmse": 0.04121908921898318, "全帯域_40ms_envelope_correlation": 0.9985639496217161, "全帯域_40ms_modulation_mae_db": 0.4710564855179915, "300–1000 Hz_10ms_log_rms_mae_db": 1.0094319219875032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09717930097109206, "300–1000 Hz_10ms_envelope_correlation": 0.9931997317195831, "300–1000 Hz_10ms_modulation_mae_db": 0.6924574736800296, "300–1000 Hz_40ms_log_rms_mae_db": 0.5836966173120346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05376834336755672, "300–1000 Hz_40ms_envelope_correlation": 0.9979800346792811, "300–1000 Hz_40ms_modulation_mae_db": 0.7107322914248838, "1000–3000 Hz_10ms_log_rms_mae_db": 1.333762971329824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10748739608918753, "1000–3000 Hz_10ms_envelope_correlation": 0.9926135711801303, "1000–3000 Hz_10ms_modulation_mae_db": 0.5770286711670495, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7839807255479546, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.051267913281461026, "1000–3000 Hz_40ms_envelope_correlation": 0.9982590007674603, "1000–3000 Hz_40ms_modulation_mae_db": 0.5594295055280405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6375403476070665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17394205248413677, "3000–6000 Hz_10ms_envelope_correlation": 0.9783089433733515, "3000–6000 Hz_10ms_modulation_mae_db": 0.7249160648443659, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9384556532308972, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0893286618384298, "3000–6000 Hz_40ms_envelope_correlation": 0.9938821256089061, "3000–6000 Hz_40ms_modulation_mae_db": 0.9251514197785692, "mr_stft_db": 5.521720857610238, "band_mae_db": 0.151783433348704, "rms_gain_error_db": -0.08442820952938587, "output_peak": 0.31976690493866927}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0401573074298565, "全帯域_10ms_rms_envelope_nrmse": 0.07911900806428604, "全帯域_10ms_envelope_correlation": 0.9948958693523519, "全帯域_10ms_modulation_mae_db": 0.11963217832471155, "全帯域_40ms_log_rms_mae_db": 0.7503306901567622, "全帯域_40ms_rms_envelope_nrmse": 0.04222359806037593, "全帯域_40ms_envelope_correlation": 0.9984296167144993, "全帯域_40ms_modulation_mae_db": 0.20400948067638788, "300–1000 Hz_10ms_log_rms_mae_db": 1.0959173798394455, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0844666670684486, "300–1000 Hz_10ms_envelope_correlation": 0.9947710827663515, "300–1000 Hz_10ms_modulation_mae_db": 0.22955879599320006, "300–1000 Hz_40ms_log_rms_mae_db": 0.6429934636333546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04320426854433152, "300–1000 Hz_40ms_envelope_correlation": 0.9985859344949801, "300–1000 Hz_40ms_modulation_mae_db": 0.41364281860369284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2949931227155878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15391543903450663, "1000–3000 Hz_10ms_envelope_correlation": 0.9855065058023752, "1000–3000 Hz_10ms_modulation_mae_db": 0.23914872482762176, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7477394317873989, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08574398812331589, "1000–3000 Hz_40ms_envelope_correlation": 0.9955474315652751, "1000–3000 Hz_40ms_modulation_mae_db": 0.27109279977895745, "3000–6000 Hz_10ms_log_rms_mae_db": 1.68616043246352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1981487560215938, "3000–6000 Hz_10ms_envelope_correlation": 0.9711708204361952, "3000–6000 Hz_10ms_modulation_mae_db": 1.2775226253990868, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9739489732965463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10111730710478531, "3000–6000 Hz_40ms_envelope_correlation": 0.9921613067683462, "3000–6000 Hz_40ms_modulation_mae_db": 1.4749124858482354, "mr_stft_db": 5.521206592246415, "band_mae_db": 0.2787357878772539, "rms_gain_error_db": 0.013773848481496109, "output_peak": 0.35531642968860044}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1008366669602987, "全帯域_10ms_rms_envelope_nrmse": 0.07877702616879835, "全帯域_10ms_envelope_correlation": 0.9949522298488186, "全帯域_10ms_modulation_mae_db": 0.35115406826233375, "全帯域_40ms_log_rms_mae_db": 0.7369804274144944, "全帯域_40ms_rms_envelope_nrmse": 0.04810420638360696, "全帯域_40ms_envelope_correlation": 0.9979663074274207, "全帯域_40ms_modulation_mae_db": 0.42154055756256903, "300–1000 Hz_10ms_log_rms_mae_db": 1.0777076788030733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.099114838657216, "300–1000 Hz_10ms_envelope_correlation": 0.9927529248026815, "300–1000 Hz_10ms_modulation_mae_db": 0.5098574911082873, "300–1000 Hz_40ms_log_rms_mae_db": 0.6323082453341449, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06423732747842573, "300–1000 Hz_40ms_envelope_correlation": 0.9967176798028596, "300–1000 Hz_40ms_modulation_mae_db": 0.4653750838465611, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2726439967019585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11440762623886376, "1000–3000 Hz_10ms_envelope_correlation": 0.9917069402552946, "1000–3000 Hz_10ms_modulation_mae_db": 0.5415079538011115, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7375282956192909, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05986444261843961, "1000–3000 Hz_40ms_envelope_correlation": 0.9976408616336124, "1000–3000 Hz_40ms_modulation_mae_db": 0.5917551850113596, "3000–6000 Hz_10ms_log_rms_mae_db": 1.639738726961671, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16826148920051204, "3000–6000 Hz_10ms_envelope_correlation": 0.9797307716439628, "3000–6000 Hz_10ms_modulation_mae_db": 0.8142377577173202, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9854934366497933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09698330415642689, "3000–6000 Hz_40ms_envelope_correlation": 0.9927345563153542, "3000–6000 Hz_40ms_modulation_mae_db": 0.8553733340641134, "mr_stft_db": 5.516123996570074, "band_mae_db": 0.16869128751196324, "rms_gain_error_db": 0.02707037743995129, "output_peak": 0.3412890590276304}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9403067881964045, "全帯域_10ms_rms_envelope_nrmse": 0.07241211526852448, "全帯域_10ms_envelope_correlation": 0.9959056836272727, "全帯域_10ms_modulation_mae_db": 0.31715540610936976, "全帯域_40ms_log_rms_mae_db": 0.60919056042201, "全帯域_40ms_rms_envelope_nrmse": 0.04109531773986153, "全帯域_40ms_envelope_correlation": 0.9987669720349145, "全帯域_40ms_modulation_mae_db": 0.22505310690365474, "300–1000 Hz_10ms_log_rms_mae_db": 0.9474457523620078, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09926122559669444, "300–1000 Hz_10ms_envelope_correlation": 0.9931594094864762, "300–1000 Hz_10ms_modulation_mae_db": 0.677765604314445, "300–1000 Hz_40ms_log_rms_mae_db": 0.5695558966488001, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05780388742628248, "300–1000 Hz_40ms_envelope_correlation": 0.9979445247063267, "300–1000 Hz_40ms_modulation_mae_db": 0.6660658146381038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3074931260417735, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1363627914163777, "1000–3000 Hz_10ms_envelope_correlation": 0.9882006074775213, "1000–3000 Hz_10ms_modulation_mae_db": 0.663109007674578, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7802610488075173, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06698339564204826, "1000–3000 Hz_40ms_envelope_correlation": 0.9973847534738364, "1000–3000 Hz_40ms_modulation_mae_db": 0.8502659233558353, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6005056218381224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17621748450642882, "3000–6000 Hz_10ms_envelope_correlation": 0.9772880055480044, "3000–6000 Hz_10ms_modulation_mae_db": 0.9287655085183953, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8327602619740178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08281898503269541, "3000–6000 Hz_40ms_envelope_correlation": 0.9947758687404319, "3000–6000 Hz_40ms_modulation_mae_db": 1.1228561627636113, "mr_stft_db": 5.519340037247485, "band_mae_db": 0.24785640517661167, "rms_gain_error_db": -0.14954108124957394, "output_peak": 0.33708578313825466}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48640, "output_samples": 48640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.834876978487676, "spectral_convergence": 0.2692091325875288, "spectral_convergence_status": "ok", "amplitude_floor": 9.047519636617716e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.386064696316521, "spectral_convergence": 0.28243121611579625, "spectral_convergence_status": "ok", "amplitude_floor": 7.308511681145121e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.320143939553834, "spectral_convergence": 0.27345552476444474, "spectral_convergence_status": "ok", "amplitude_floor": 4.4367419091404295e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011152155483091157, "output_power": 0.00011884537906065517, "delta_db": 0.27623484459192005, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042085142829602937, "output_power": 0.00042162360863679015, "delta_db": 0.00796115588554628, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006661514585812611, "output_power": 0.0006075182350700915, "delta_db": -0.40013665037816776, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015794681682280444, "output_power": 0.00014804839239422352, "delta_db": -0.2810718188760742, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.332154632901203e-05, "output_power": 5.065169173508794e-05, "delta_db": -0.2230878039350696, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2768891588196896e-05, "output_power": 3.26807249426157e-05, "delta_db": -0.011700698294258224, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20003216199350604, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.19454156129403688, "pitch": null, "output_rms_dbfs": -28.603196914488144, "output_peak": 0.33734834364549426, "reference_rms_dbfs": -28.408655353194106, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.513695204786011}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00012692463938083332, "log_rms_mae_db": 0.9960902395700382, "rms_envelope_nrmse": 0.09798075308570478, "envelope_correlation": 0.9923897567445545, "rms_gain_error_db": -0.1945416093875631, "reference_mean_power": 0.0014449383611437946, "output_mean_power": 0.0013816408340429636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.053245386897797e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005746543771154043, "output_power": 0.0005502757809077405, "reference_active": true, "delta_db": -0.1882632026883938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014747895972213286, "output_power": 0.00012924434156413186, "reference_active": true, "delta_db": -0.5731852717435131}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.536231970329893e-05, "output_power": 2.042037406548009e-05, "reference_active": true, "delta_db": -0.9412527938404592}], "active_band_count": 3, "active_band_mae_db": 0.567567089424122}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 0.00010829935323771927, "log_rms_mae_db": 0.6960467044300289, "rms_envelope_nrmse": 0.05881346333315142, "envelope_correlation": 0.9973174507641216, "rms_gain_error_db": -0.19454146377241952, "reference_mean_power": 0.0014593630016451321, "output_mean_power": 0.0013954336299671854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 9.289432894914324e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005498374692309951, "output_power": 0.0005237470529065383, "reference_active": true, "delta_db": -0.21112739720996285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.817099868184177e-05, "output_power": 5.344326758982751e-05, "reference_active": true, "delta_db": -0.3681351624586879}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4585911877580322e-06, "output_power": 1.051696273794457e-06, "reference_active": true, "delta_db": -1.4204325019853192}], "active_band_count": 3, "active_band_mae_db": 0.6665650205513233}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00011599347403670498, "log_rms_mae_db": 1.1350533824138782, "rms_envelope_nrmse": 0.12708934914945322, "envelope_correlation": 0.9883682200752563, "rms_gain_error_db": -0.24960157776974268, "reference_mean_power": 0.0008943859828976575, "output_mean_power": 0.000844432279508048}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 8.128232732706927e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005209748261573112, "output_power": 0.0005042129214645838, "reference_active": true, "delta_db": -0.14202767266719202}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010680702813776803, "output_power": 8.746494551960577e-05, "reference_active": true, "delta_db": -0.8676580120422912}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2431239684221878e-05, "output_power": 1.658204788202553e-05, "reference_active": true, "delta_db": -1.3121511108031705}], "active_band_count": 3, "active_band_mae_db": 0.773945598504218}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 9.903058317661472e-05, "log_rms_mae_db": 0.7240744927607788, "rms_envelope_nrmse": 0.07923191879831983, "envelope_correlation": 0.9955310951767552, "rms_gain_error_db": -0.24960156272126965, "reference_mean_power": 0.0009033149500834516, "output_mean_power": 0.0008528625444066724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 7.323650960940424e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005139241705701183, "output_power": 0.0004931386675257653, "reference_active": true, "delta_db": -0.17929986283294533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0770880999134286e-05, "output_power": 3.3828564871892515e-05, "reference_active": true, "delta_db": -0.8106652216105991}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2000817341803485e-06, "output_power": 7.445964775485305e-07, "reference_active": true, "delta_db": -2.072898482497397}], "active_band_count": 3, "active_band_mae_db": 1.0209545223136471}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 7.992748424465529e-05, "log_rms_mae_db": 1.2681832847556163, "rms_envelope_nrmse": 0.11933478253949135, "envelope_correlation": 0.9910291841697326, "rms_gain_error_db": -0.21612685854091146, "reference_mean_power": 0.00017342262158164605, "output_mean_power": 0.00016500346381832738}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.6075439307790634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.767193199363673e-05, "output_power": 9.332647412482869e-05, "reference_active": true, "delta_db": -0.19764919850782423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.270979881634546e-05, "output_power": 1.9490965790731705e-05, "reference_active": true, "delta_db": -0.6637992762844362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.175922545645147e-06, "output_power": 3.5241667759417443e-06, "reference_active": true, "delta_db": -0.7369598227753886}], "active_band_count": 3, "active_band_mae_db": 0.532802765855883}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 6.970659359658012e-05, "log_rms_mae_db": 0.7395775508761001, "rms_envelope_nrmse": 0.07453767367577778, "envelope_correlation": 0.9965615493594054, "rms_gain_error_db": -0.21612553333651907, "reference_mean_power": 0.00017515363198831316, "output_mean_power": 0.0001666504896170829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 1.4903701525512362e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.997559182587721e-05, "output_power": 9.477021590368412e-05, "reference_active": true, "delta_db": -0.23222113680525677}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.524639480988942e-06, "output_power": 8.178264725251899e-06, "reference_active": true, "delta_db": -0.6618738180921545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.337301598444833e-07, "output_power": 2.942879050984241e-07, "reference_active": true, "delta_db": -0.5462304301434502}], "active_band_count": 3, "active_band_mae_db": 0.48010846168028715}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 2.3755576133519224e-05, "log_rms_mae_db": 1.5983523364373282, "rms_envelope_nrmse": 0.17871055914977363, "envelope_correlation": 0.9766736255312807, "rms_gain_error_db": -0.11414473605679976, "reference_mean_power": 3.0329423269641833e-05, "output_mean_power": 2.9542665587403812e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 3.170053749158532e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1550798308210337e-05, "output_power": 1.0175879439753353e-05, "reference_active": true, "delta_db": -0.5504004761422246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.1283799417647094e-06, "output_power": 5.129171019331834e-06, "reference_active": true, "delta_db": -0.7729850268981686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7104244593685833e-06, "output_power": 1.6514652313184242e-06, "reference_active": true, "delta_db": -0.15234463575118326}], "active_band_count": 3, "active_band_mae_db": 0.49191004626385887}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 2.1008424841575618e-05, "log_rms_mae_db": 0.8571910840903176, "rms_envelope_nrmse": 0.09117939592110937, "envelope_correlation": 0.9936145730195698, "rms_gain_error_db": -0.11414230760413674, "reference_mean_power": 3.0632047924156045e-05, "output_mean_power": 2.9837456718490788e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 2.7930999787597993e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1345569218097283e-05, "output_power": 9.964646688701402e-06, "reference_active": true, "delta_db": -0.5636438498279255}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6424993904119263e-06, "output_power": 2.414203335286461e-06, "reference_active": true, "delta_db": -0.3924105021718696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1775222693841551e-07, "output_power": 9.547426466241218e-08, "reference_active": true, "delta_db": -0.9108280714066923}], "active_band_count": 3, "active_band_mae_db": 0.6222941411354957}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9150039880214543, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ce74928c003b63c17a7bf7012024cc0efd9cbc97d42a943a6f0b4d421ee536a2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9896138548042085, "standard_deviation": 0.0026654866729225054, "minimum": 0.9855065058023752, "maximum": 0.9926135711801303, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.317019665567645, "standard_deviation": 0.06857015019162081, "minimum": 1.2658006308370169, "maximum": 1.4753814103347687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5372575621155997, "standard_deviation": 0.16892288706527447, "minimum": 0.23914872482762176, "maximum": 0.8215197604283883, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12732418878003476, "standard_deviation": 0.01640588467490768, "minimum": 0.10748739608918753, "maximum": 0.15391543903450663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968884814493795, "standard_deviation": 0.0012375845365932222, "minimum": 0.9946213224145923, "maximum": 0.9982590007674603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7717881174671701, "standard_deviation": 0.03609854739204158, "minimum": 0.7375282956192909, "maximum": 0.84934189754067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5919603604385606, "standard_deviation": 0.1754684346781418, "minimum": 0.27109279977895745, "maximum": 0.8502659233558353, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06934257222034428, "standard_deviation": 0.012928401635576549, "minimum": 0.051267913281461026, "maximum": 0.08900872326246745, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9759546551641899, "standard_deviation": 0.003804462597112117, "minimum": 0.9690114873455172, "maximum": 0.9797307716439628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6410580205587697, "standard_deviation": 0.030117955531160902, "minimum": 1.5983523364373282, "maximum": 1.68616043246352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9406348516629993, "standard_deviation": 0.27762149006634756, "minimum": 0.49191004626385887, "maximum": 1.2775226253990868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1815302150544241, "standard_deviation": 0.013088820606283273, "minimum": 0.16826148920051204, "maximum": 0.205162534017137, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932151350761254, "standard_deviation": 0.0008927499350751305, "minimum": 0.992127808399539, "maximum": 0.9947758687404319, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9195379452080807, "standard_deviation": 0.05372854941878422, "minimum": 0.8327602619740178, "maximum": 0.9854934366497933, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0789173316247747, "standard_deviation": 0.3142942815876524, "minimum": 0.6222941411354957, "maximum": 1.4749124858482354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09485269193491105, "standard_deviation": 0.006551944620672995, "minimum": 0.08281898503269541, "maximum": 0.10203032891141071, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922997377885125, "standard_deviation": 0.0018537559764728391, "minimum": 0.9883682200752563, "maximum": 0.9947710827663515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.060618197023456, "standard_deviation": 0.059865959282456666, "minimum": 0.9474457523620078, "maximum": 1.1350533824138782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5590957225930064, "standard_deviation": 0.2710709601404574, "minimum": 0.09382035988609679, "maximum": 0.8849746330163496, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10293981595397544, "standard_deviation": 0.012016301962573157, "minimum": 0.0844666670684486, "maximum": 0.12708934914945322, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974127863899129, "standard_deviation": 0.0009705294578768501, "minimum": 0.9955310951767552, "maximum": 0.9985859344949801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6318717701797347, "standard_deviation": 0.05348158575466129, "minimum": 0.5695558966488001, "maximum": 0.7240744927607788, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6626143505120919, "standard_deviation": 0.30648616104723697, "minimum": 0.1424217486533079, "maximum": 1.0209545223136471, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059733093840075344, "standard_deviation": 0.010391489806512604, "minimum": 0.04320426854433152, "maximum": 0.07923191879831983, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20895120302974668, "standard_deviation": 0.04913344070366953, "minimum": 0.151783433348704, "maximum": 0.2787357878772539, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.518482869930843, "standard_deviation": 0.0051045889477703025, "minimum": 5.511023026404871, "maximum": 5.527341596289851, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.34214234710811475, "standard_deviation": 0.012526852044738346, "minimum": 0.31976690493866927, "maximum": 0.3610327759368158, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07656779232336544, "standard_deviation": 0.08346970071373726, "minimum": -0.19454156129403688, "maximum": 0.02707037743995129, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945291847570255, "standard_deviation": 0.001020668288352802, "minimum": 0.9923897567445545, "maximum": 0.9959056836272727, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9985055310225901, "standard_deviation": 0.057156169634685026, "minimum": 0.9345344321351305, "maximum": 1.1008366669602987, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4011704877614274, "standard_deviation": 0.1559579350953602, "minimum": 0.11963217832471155, "maximum": 0.567567089424122, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08214364620185699, "standard_deviation": 0.00749090554700325, "minimum": 0.07241211526852448, "maximum": 0.09798075308570478, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982386928567657, "standard_deviation": 0.0005595832783124482, "minimum": 0.9973174507641216, "maximum": 0.9987694915993028, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6626048064218918, "standard_deviation": 0.06158977691946702, "minimum": 0.5874549351578089, "maximum": 0.7503306901567622, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44193987932461115, "standard_deviation": 0.1951302561503536, "minimum": 0.20400948067638788, "maximum": 0.7276303576887079, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045802294189270315, "standard_deviation": 0.006924333120514209, "minimum": 0.03895812839974181, "maximum": 0.05881346333315142, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29095-0008", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.457566261291504, "predicted_mos": 2.457566261291504, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "678eccb469d35dd0181ecde1853ef067d9ff39dec45480489db93bc79199829d", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.20693670026958, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0088033936679228, "全帯域_10ms_rms_envelope_nrmse": 0.08990597527613407, "全帯域_10ms_envelope_correlation": 0.9934471434410211, "全帯域_10ms_modulation_mae_db": 0.35996580665625183, "全帯域_40ms_log_rms_mae_db": 0.6530728647254048, "全帯域_40ms_rms_envelope_nrmse": 0.04759417356956496, "全帯域_40ms_envelope_correlation": 0.9981197882542983, "全帯域_40ms_modulation_mae_db": 0.3852264125811437, "300–1000 Hz_10ms_log_rms_mae_db": 0.9603237816001258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11155095653075563, "300–1000 Hz_10ms_envelope_correlation": 0.9908687623125428, "300–1000 Hz_10ms_modulation_mae_db": 0.4609225834282091, "300–1000 Hz_40ms_log_rms_mae_db": 0.5588120809064492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06574955815754435, "300–1000 Hz_40ms_envelope_correlation": 0.9967068211861072, "300–1000 Hz_40ms_modulation_mae_db": 0.5289840024609762, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1545908554795565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10230224194643663, "1000–3000 Hz_10ms_envelope_correlation": 0.9932746271316292, "1000–3000 Hz_10ms_modulation_mae_db": 0.23364143926149847, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6910744276062264, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04602376646153058, "1000–3000 Hz_40ms_envelope_correlation": 0.9986250780393281, "1000–3000 Hz_40ms_modulation_mae_db": 0.26887725832123494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1838452581620669, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.132524894802935, "3000–6000 Hz_10ms_envelope_correlation": 0.9871719356340829, "3000–6000 Hz_10ms_modulation_mae_db": 0.31146216574343005, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6941427808319227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0745655973187779, "3000–6000 Hz_40ms_envelope_correlation": 0.9956598222435976, "3000–6000 Hz_40ms_modulation_mae_db": 0.4037998773008691, "mr_stft_db": 5.184399359973718, "band_mae_db": 0.17149915966874305, "rms_gain_error_db": -0.11186470408179724, "output_peak": 0.33108847278066933}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0427952144057868, "全帯域_10ms_rms_envelope_nrmse": 0.07684961719088852, "全帯域_10ms_envelope_correlation": 0.9951990966452279, "全帯域_10ms_modulation_mae_db": 0.5183800519914333, "全帯域_40ms_log_rms_mae_db": 0.6055325999229536, "全帯域_40ms_rms_envelope_nrmse": 0.0339341796080156, "全帯域_40ms_envelope_correlation": 0.9989944669373223, "全帯域_40ms_modulation_mae_db": 0.4247335670759014, "300–1000 Hz_10ms_log_rms_mae_db": 0.9682456596765855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08780277227367835, "300–1000 Hz_10ms_envelope_correlation": 0.9943125751123482, "300–1000 Hz_10ms_modulation_mae_db": 0.6567959876202524, "300–1000 Hz_40ms_log_rms_mae_db": 0.5529239931553712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040305697066658985, "300–1000 Hz_40ms_envelope_correlation": 0.9987171328094232, "300–1000 Hz_40ms_modulation_mae_db": 0.3686530639663914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.072575265298634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14014237696657855, "1000–3000 Hz_10ms_envelope_correlation": 0.9884967028530847, "1000–3000 Hz_10ms_modulation_mae_db": 0.39767829221036904, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6386223018588603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08447716577476114, "1000–3000 Hz_40ms_envelope_correlation": 0.9960550244986243, "1000–3000 Hz_40ms_modulation_mae_db": 0.3930509633353146, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1501485445107444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12801966624369895, "3000–6000 Hz_10ms_envelope_correlation": 0.9888826182573425, "3000–6000 Hz_10ms_modulation_mae_db": 0.9279171704637346, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6644583296334468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07466901499912634, "3000–6000 Hz_40ms_envelope_correlation": 0.9968060474923072, "3000–6000 Hz_40ms_modulation_mae_db": 0.9763850516289546, "mr_stft_db": 5.206987559145688, "band_mae_db": 0.18651151511372596, "rms_gain_error_db": 0.030378504122457673, "output_peak": 0.34600866205890785}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.974552731036911, "全帯域_10ms_rms_envelope_nrmse": 0.07991983188516295, "全帯域_10ms_envelope_correlation": 0.9952942826519551, "全帯域_10ms_modulation_mae_db": 0.6280180584864633, "全帯域_40ms_log_rms_mae_db": 0.658846676447613, "全帯域_40ms_rms_envelope_nrmse": 0.05071687163743223, "全帯域_40ms_envelope_correlation": 0.9983751670562787, "全帯域_40ms_modulation_mae_db": 0.7303467689539378, "300–1000 Hz_10ms_log_rms_mae_db": 0.9350477896516008, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09615976230738518, "300–1000 Hz_10ms_envelope_correlation": 0.9939443712718831, "300–1000 Hz_10ms_modulation_mae_db": 0.755287796149276, "300–1000 Hz_40ms_log_rms_mae_db": 0.5652206602290074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06608721954902798, "300–1000 Hz_40ms_envelope_correlation": 0.997490428043309, "300–1000 Hz_40ms_modulation_mae_db": 0.8748604065241175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1181400990857373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12275830551633257, "1000–3000 Hz_10ms_envelope_correlation": 0.990597298123483, "1000–3000 Hz_10ms_modulation_mae_db": 0.36708284726626855, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7033881563117483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07518372348173702, "1000–3000 Hz_40ms_envelope_correlation": 0.9966644532515406, "1000–3000 Hz_40ms_modulation_mae_db": 0.3729414347225733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1693434316780602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1494840312868228, "3000–6000 Hz_10ms_envelope_correlation": 0.9836521857361419, "3000–6000 Hz_10ms_modulation_mae_db": 0.3110766165841176, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7213611137588483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08790306237685584, "3000–6000 Hz_40ms_envelope_correlation": 0.9939660901120564, "3000–6000 Hz_40ms_modulation_mae_db": 0.2600045871789229, "mr_stft_db": 5.169119349584688, "band_mae_db": 0.22646528792287976, "rms_gain_error_db": -0.24582720440124686, "output_peak": 0.309815421045364}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.050096258608808, "全帯域_10ms_rms_envelope_nrmse": 0.07584470456375368, "全帯域_10ms_envelope_correlation": 0.9953018513092757, "全帯域_10ms_modulation_mae_db": 0.7038830716086434, "全帯域_40ms_log_rms_mae_db": 0.6590959932837788, "全帯域_40ms_rms_envelope_nrmse": 0.03473536457293714, "全帯域_40ms_envelope_correlation": 0.998948290168159, "全帯域_40ms_modulation_mae_db": 0.4849071082720231, "300–1000 Hz_10ms_log_rms_mae_db": 0.94231700472243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09916216477641294, "300–1000 Hz_10ms_envelope_correlation": 0.992740682698681, "300–1000 Hz_10ms_modulation_mae_db": 0.8344231125487913, "300–1000 Hz_40ms_log_rms_mae_db": 0.5308406361587665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05109883751697321, "300–1000 Hz_40ms_envelope_correlation": 0.9979429046489116, "300–1000 Hz_40ms_modulation_mae_db": 0.8067972902415796, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0861623052889375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10635672732845691, "1000–3000 Hz_10ms_envelope_correlation": 0.9927410427869914, "1000–3000 Hz_10ms_modulation_mae_db": 0.21787558453531738, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6700588978483287, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052183113755890326, "1000–3000 Hz_40ms_envelope_correlation": 0.9982369527015839, "1000–3000 Hz_40ms_modulation_mae_db": 0.17714172665662534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1810016115425015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13778509484083135, "3000–6000 Hz_10ms_envelope_correlation": 0.9863793415628371, "3000–6000 Hz_10ms_modulation_mae_db": 0.47215343629529877, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7375018485690992, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07628660052131228, "3000–6000 Hz_40ms_envelope_correlation": 0.9958988822654987, "3000–6000 Hz_40ms_modulation_mae_db": 0.5374393848753289, "mr_stft_db": 5.194303599679231, "band_mae_db": 0.19705273227312203, "rms_gain_error_db": -0.03728969429549866, "output_peak": 0.32760799856337053}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0096472805242713, "全帯域_10ms_rms_envelope_nrmse": 0.08379362219186681, "全帯域_10ms_envelope_correlation": 0.9943508547549977, "全帯域_10ms_modulation_mae_db": 0.9180373405068448, "全帯域_40ms_log_rms_mae_db": 0.6400207481783273, "全帯域_40ms_rms_envelope_nrmse": 0.046284794197325195, "全帯域_40ms_envelope_correlation": 0.9982708865350558, "全帯域_40ms_modulation_mae_db": 0.8244069678678381, "300–1000 Hz_10ms_log_rms_mae_db": 0.9905703453125636, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.101971375050624, "300–1000 Hz_10ms_envelope_correlation": 0.9926286680244463, "300–1000 Hz_10ms_modulation_mae_db": 1.0587126670443125, "300–1000 Hz_40ms_log_rms_mae_db": 0.5881511008291159, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05950532088824203, "300–1000 Hz_40ms_envelope_correlation": 0.9976375626385348, "300–1000 Hz_40ms_modulation_mae_db": 1.0787407876199335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0901470053276736, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11983794805304945, "1000–3000 Hz_10ms_envelope_correlation": 0.9907678803636136, "1000–3000 Hz_10ms_modulation_mae_db": 0.4057558970866015, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7411215684281988, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06086369062361745, "1000–3000 Hz_40ms_envelope_correlation": 0.9975881315671528, "1000–3000 Hz_40ms_modulation_mae_db": 0.40963137266317723, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2186343458694202, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14317089070146508, "3000–6000 Hz_10ms_envelope_correlation": 0.9861222973032094, "3000–6000 Hz_10ms_modulation_mae_db": 0.4761863309360543, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6517160045585946, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07916912557885872, "3000–6000 Hz_40ms_envelope_correlation": 0.9967315324850257, "3000–6000 Hz_40ms_modulation_mae_db": 0.5939507203272564, "mr_stft_db": 5.170953840296261, "band_mae_db": 0.21924679302295383, "rms_gain_error_db": -0.1286856453066376, "output_peak": 0.33687132553857024}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9817514506323204, "全帯域_10ms_rms_envelope_nrmse": 0.07597969004738374, "全帯域_10ms_envelope_correlation": 0.9955734495701731, "全帯域_10ms_modulation_mae_db": 0.4415418841023519, "全帯域_40ms_log_rms_mae_db": 0.696351220501086, "全帯域_40ms_rms_envelope_nrmse": 0.04714344969150054, "全帯域_40ms_envelope_correlation": 0.998416965102234, "全帯域_40ms_modulation_mae_db": 0.5308584083042112, "300–1000 Hz_10ms_log_rms_mae_db": 0.9561007272342267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09954292632066493, "300–1000 Hz_10ms_envelope_correlation": 0.993143583164323, "300–1000 Hz_10ms_modulation_mae_db": 0.4011844423999036, "300–1000 Hz_40ms_log_rms_mae_db": 0.594770438822568, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061252711405644596, "300–1000 Hz_40ms_envelope_correlation": 0.9976462921005809, "300–1000 Hz_40ms_modulation_mae_db": 0.33239709497681624, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1251701246271542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12363109316354683, "1000–3000 Hz_10ms_envelope_correlation": 0.9901832754087629, "1000–3000 Hz_10ms_modulation_mae_db": 0.08817229718170427, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6944200276625955, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06284843892554702, "1000–3000 Hz_40ms_envelope_correlation": 0.997462196098334, "1000–3000 Hz_40ms_modulation_mae_db": 0.21390691410431117, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1463925937856922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14056324468619083, "3000–6000 Hz_10ms_envelope_correlation": 0.985932347207589, "3000–6000 Hz_10ms_modulation_mae_db": 0.7053259662879775, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6226601593833224, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0755318044152769, "3000–6000 Hz_40ms_envelope_correlation": 0.9962087494961891, "3000–6000 Hz_40ms_modulation_mae_db": 0.6788247764170654, "mr_stft_db": 5.193190517877153, "band_mae_db": 0.21462538646162352, "rms_gain_error_db": -0.19113117442703195, "output_peak": 0.3316656741689244}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9459570012383637, "全帯域_10ms_rms_envelope_nrmse": 0.07773310250314051, "全帯域_10ms_envelope_correlation": 0.9950518286825424, "全帯域_10ms_modulation_mae_db": 0.23656865758912293, "全帯域_40ms_log_rms_mae_db": 0.6469726845009881, "全帯域_40ms_rms_envelope_nrmse": 0.04191172327961523, "全帯域_40ms_envelope_correlation": 0.998452178458276, "全帯域_40ms_modulation_mae_db": 0.09771418163771557, "300–1000 Hz_10ms_log_rms_mae_db": 0.885978958841657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1005892245469653, "300–1000 Hz_10ms_envelope_correlation": 0.9925548990663144, "300–1000 Hz_10ms_modulation_mae_db": 0.2747443941240703, "300–1000 Hz_40ms_log_rms_mae_db": 0.5112339873258693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060687421194478536, "300–1000 Hz_40ms_envelope_correlation": 0.9970778606207196, "300–1000 Hz_40ms_modulation_mae_db": 0.10390835602151317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.146787351887504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12195820711528094, "1000–3000 Hz_10ms_envelope_correlation": 0.9915900174805503, "1000–3000 Hz_10ms_modulation_mae_db": 0.9421855868143219, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7294377098622006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07728255803009333, "1000–3000 Hz_40ms_envelope_correlation": 0.9974739920958934, "1000–3000 Hz_40ms_modulation_mae_db": 0.9114333908358025, "3000–6000 Hz_10ms_log_rms_mae_db": 1.208209435967956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15139214284827646, "3000–6000 Hz_10ms_envelope_correlation": 0.9837268448639365, "3000–6000 Hz_10ms_modulation_mae_db": 0.8796535865518155, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6871043858438027, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09544360894047338, "3000–6000 Hz_40ms_envelope_correlation": 0.9936541275636063, "3000–6000 Hz_40ms_modulation_mae_db": 0.7867973862469917, "mr_stft_db": 5.172004787908852, "band_mae_db": 0.17861901456673066, "rms_gain_error_db": -0.03525490808516006, "output_peak": 0.3182578512963936}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0778051562063076, "全帯域_10ms_rms_envelope_nrmse": 0.08225963550073177, "全帯域_10ms_envelope_correlation": 0.9944603285325403, "全帯域_10ms_modulation_mae_db": 0.1174846935699731, "全帯域_40ms_log_rms_mae_db": 0.7203441611274727, "全帯域_40ms_rms_envelope_nrmse": 0.04338462796329511, "全帯域_40ms_envelope_correlation": 0.9983390249695133, "全帯域_40ms_modulation_mae_db": 0.444288052532077, "300–1000 Hz_10ms_log_rms_mae_db": 0.9920886096584474, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10752060832773078, "300–1000 Hz_10ms_envelope_correlation": 0.9915182139085578, "300–1000 Hz_10ms_modulation_mae_db": 0.28851972401514375, "300–1000 Hz_40ms_log_rms_mae_db": 0.6134202707803634, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059975580651396077, "300–1000 Hz_40ms_envelope_correlation": 0.9971637567953444, "300–1000 Hz_40ms_modulation_mae_db": 0.5959786080945909, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0424201976609224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09982143003824402, "1000–3000 Hz_10ms_envelope_correlation": 0.9939284639636831, "1000–3000 Hz_10ms_modulation_mae_db": 0.2858706743494596, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6703040379059871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05395516152872773, "1000–3000 Hz_40ms_envelope_correlation": 0.9985127162344216, "1000–3000 Hz_40ms_modulation_mae_db": 0.32097930196269253, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1947772704641146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15089278507875378, "3000–6000 Hz_10ms_envelope_correlation": 0.9851667800787145, "3000–6000 Hz_10ms_modulation_mae_db": 1.0690622350539243, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7139040691886758, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09055108723709242, "3000–6000 Hz_40ms_envelope_correlation": 0.9960853117535128, "3000–6000 Hz_40ms_modulation_mae_db": 1.2986333925773899, "mr_stft_db": 5.1736887062660495, "band_mae_db": 0.19924015040279439, "rms_gain_error_db": -0.027352526427141276, "output_peak": 0.33184917092090205}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48640, "output_samples": 48640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.403833375556202, "spectral_convergence": 0.27052516539194366, "spectral_convergence_status": "ok", "amplitude_floor": 9.047519636617716e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.946358595901677, "spectral_convergence": 0.28398563220031914, "spectral_convergence_status": "ok", "amplitude_floor": 7.308511681145121e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.203006108463279, "spectral_convergence": 0.28985333709909067, "spectral_convergence_status": "ok", "amplitude_floor": 4.4367419091404295e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011152155483091157, "output_power": 0.00012047651927229091, "delta_db": 0.33543595970016016, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042085142829602937, "output_power": 0.00043091117581425277, "delta_db": 0.10258952361954968, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006661514585812611, "output_power": 0.000614989721347473, "delta_db": -0.3470512590612415, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015794681682280444, "output_power": 0.000154604801787565, "delta_db": -0.09287899475446759, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.332154632901203e-05, "output_power": 5.188343659280898e-05, "delta_db": -0.11874000856890084, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2768891588196896e-05, "output_power": 3.301350785444915e-05, "delta_db": 0.03229921230813843, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17149915966874305, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11186470408179724, "pitch": null, "output_rms_dbfs": -28.520520057275903, "output_peak": 0.33108847278066933, "reference_rms_dbfs": -28.408655353194106, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.184399359973718}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00012692463938083332, "log_rms_mae_db": 1.0088033936679228, "rms_envelope_nrmse": 0.08990597527613407, "envelope_correlation": 0.9934471434410211, "rms_gain_error_db": -0.11186475771508683, "reference_mean_power": 0.0014449383611437946, "output_mean_power": 0.0014081951537572818}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.053245386897797e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005746543771154043, "output_power": 0.0005513005213923308, "reference_active": true, "delta_db": -0.1801831583866999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014747895972213286, "output_power": 0.0001279669536320403, "reference_active": true, "delta_db": -0.6163223424316867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.536231970329893e-05, "output_power": 2.3760182976240743e-05, "reference_active": true, "delta_db": -0.2833919191503688}], "active_band_count": 3, "active_band_mae_db": 0.35996580665625183}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 0.00010829935323771927, "log_rms_mae_db": 0.6530728647254048, "rms_envelope_nrmse": 0.04759417356956496, "envelope_correlation": 0.9981197882542983, "rms_gain_error_db": -0.11186470537176754, "reference_mean_power": 0.0014593630016451321, "output_mean_power": 0.001422253008516768}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 9.289432894914324e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005498374692309951, "output_power": 0.000526866597686471, "reference_active": true, "delta_db": -0.18533665929660598}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.817099868184177e-05, "output_power": 5.1364779981749e-05, "reference_active": true, "delta_db": -0.5404108766871445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4585911877580322e-06, "output_power": 1.3211143661480393e-06, "reference_active": true, "delta_db": -0.4299317017596807}], "active_band_count": 3, "active_band_mae_db": 0.3852264125811437}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00011599347403670498, "log_rms_mae_db": 0.9603237816001258, "rms_envelope_nrmse": 0.11155095653075563, "envelope_correlation": 0.9908687623125428, "rms_gain_error_db": -0.13573960042408334, "reference_mean_power": 0.0008943859828976575, "output_mean_power": 0.0008668641130624806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 8.128232732706927e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005209748261573112, "output_power": 0.0004999298907558582, "reference_active": true, "delta_db": -0.17907634463968936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010680702813776803, "output_power": 9.297720833018631e-05, "reference_active": true, "delta_db": -0.6022332895162986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2431239684221878e-05, "output_power": 1.953023478941221e-05, "reference_active": true, "delta_db": -0.6014581161286393}], "active_band_count": 3, "active_band_mae_db": 0.4609225834282091}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 9.903058317661472e-05, "log_rms_mae_db": 0.5588120809064492, "rms_envelope_nrmse": 0.06574955815754435, "envelope_correlation": 0.9967068211861072, "rms_gain_error_db": -0.1357396023725512, "reference_mean_power": 0.0009033149500834516, "output_mean_power": 0.000875518319430612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 7.323650960940424e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005139241705701183, "output_power": 0.0004917458741808811, "reference_active": true, "delta_db": -0.19158318827323995}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0770880999134286e-05, "output_power": 3.6130231832104215e-05, "reference_active": true, "delta_db": -0.5247934776518117}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2000817341803485e-06, "output_power": 9.820945284934695e-07, "reference_active": true, "delta_db": -0.870575341457877}], "active_band_count": 3, "active_band_mae_db": 0.5289840024609762}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 7.992748424465529e-05, "log_rms_mae_db": 1.1545908554795565, "rms_envelope_nrmse": 0.10230224194643663, "envelope_correlation": 0.9932746271316292, "rms_gain_error_db": -0.09129670501742826, "reference_mean_power": 0.00017342262158164605, "output_mean_power": 0.00016981501059413538}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.6075439307790634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.767193199363673e-05, "output_power": 9.503675708287614e-05, "reference_active": true, "delta_db": -0.11878169778489114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.270979881634546e-05, "output_power": 2.0384345792384716e-05, "reference_active": true, "delta_db": -0.46916508942681073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.175922545645147e-06, "output_power": 4.068690751948142e-06, "reference_active": true, "delta_db": -0.1129775305727935}], "active_band_count": 3, "active_band_mae_db": 0.23364143926149847}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 6.970659359658012e-05, "log_rms_mae_db": 0.6910744276062264, "rms_envelope_nrmse": 0.04602376646153058, "envelope_correlation": 0.9986250780393281, "rms_gain_error_db": -0.09129502081304336, "reference_mean_power": 0.00017515363198831316, "output_mean_power": 0.00017151007829955989}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 1.4903701525512362e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.997559182587721e-05, "output_power": 9.707782528703196e-05, "reference_active": true, "delta_db": -0.1277394489443081}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.524639480988942e-06, "output_power": 8.633158607846558e-06, "reference_active": true, "delta_db": -0.426788262771698}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.337301598444833e-07, "output_power": 3.1490898356066924e-07, "reference_active": true, "delta_db": -0.2521040632476987}], "active_band_count": 3, "active_band_mae_db": 0.26887725832123494}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 2.3755576133519224e-05, "log_rms_mae_db": 1.1838452581620669, "rms_envelope_nrmse": 0.132524894802935, "envelope_correlation": 0.9871719356340829, "rms_gain_error_db": -0.09962946045345414, "reference_mean_power": 3.0329423269641833e-05, "output_mean_power": 2.9641570216654025e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 3.170053749158532e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1550798308210337e-05, "output_power": 1.1426113311052126e-05, "reference_active": true, "delta_db": -0.047134740013778806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.1283799417647094e-06, "output_power": 6.102811667007346e-06, "reference_active": true, "delta_db": -0.01815714636576824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7104244593685833e-06, "output_power": 1.4002140699383785e-06, "reference_active": true, "delta_db": -0.869094610850743}], "active_band_count": 3, "active_band_mae_db": 0.31146216574343005}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 2.1008424841575618e-05, "log_rms_mae_db": 0.6941427808319227, "rms_envelope_nrmse": 0.0745655973187779, "envelope_correlation": 0.9956598222435976, "rms_gain_error_db": -0.0996306621769906, "reference_mean_power": 3.0632047924156045e-05, "output_mean_power": 2.993732324239811e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 2.7930999787597993e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1345569218097283e-05, "output_power": 1.1452476095359417e-05, "reference_active": true, "delta_db": 0.04073104211913689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6424993904119263e-06, "output_power": 2.6542849013630975e-06, "reference_active": true, "delta_db": 0.019326408391265155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1775222693841551e-07, "output_power": 9.033060337057225e-08, "reference_active": true, "delta_db": -1.1513421813922051}], "active_band_count": 3, "active_band_mae_db": 0.4037998773008691}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9269056678913318, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "01c3d92cbdadb5ff83d170e367106306615e36242f3d50a677ae43bda5886029", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9914474135139748, "standard_deviation": 0.001799894185729413, "minimum": 0.9884967028530847, "maximum": 0.9939284639636831, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.104499150582015, "standard_deviation": 0.038407570550835116, "minimum": 1.0424201976609224, "maximum": 1.1545908554795565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3672828273381926, "standard_deviation": 0.2556868776267115, "minimum": 0.08817229718170427, "maximum": 0.9421855868143219, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11710104126599075, "standard_deviation": 0.013467748744789756, "minimum": 0.09982143003824402, "maximum": 0.14014237696657855, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975773180608598, "standard_deviation": 0.0008930393270618277, "minimum": 0.9960550244986243, "maximum": 0.9986250780393281, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6923033909355182, "standard_deviation": 0.0332769395744275, "minimum": 0.6386223018588603, "maximum": 0.7411215684281988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3834952953252164, "standard_deviation": 0.22928210327480836, "minimum": 0.17714172665662534, "maximum": 0.9114333908358025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06410220232273808, "standard_deviation": 0.013604504802506572, "minimum": 0.04602376646153058, "maximum": 0.08447716577476114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9858792938304817, "standard_deviation": 0.0017363145037909188, "minimum": 0.9836521857361419, "maximum": 0.9888826182573425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1815440614975694, "standard_deviation": 0.025739348713003773, "minimum": 1.1463925937856922, "maximum": 1.2186343458694202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.644104688489544, "standard_deviation": 0.29269494653516615, "minimum": 0.3110766165841176, "maximum": 1.0690622350539243, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14172909381112175, "standard_deviation": 0.008694379165681373, "minimum": 0.12801966624369895, "maximum": 0.15139214284827646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9956263204264743, "standard_deviation": 0.0011883041948732825, "minimum": 0.9936541275636063, "maximum": 0.9968060474923072, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.686606086470964, "standard_deviation": 0.03854022571256304, "minimum": 0.6226601593833224, "maximum": 0.7375018485690992, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6919793970690973, "standard_deviation": 0.32976637846341633, "minimum": 0.2600045871789229, "maximum": 1.2986333925773899, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08176498767347173, "standard_deviation": 0.008278751073682542, "minimum": 0.0745655973187779, "maximum": 0.09544360894047338, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992713969444887, "standard_deviation": 0.0011415472372026075, "minimum": 0.9908687623125428, "maximum": 0.9943125751123482, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9538341095872046, "standard_deviation": 0.03413794489008997, "minimum": 0.885978958841657, "maximum": 0.9920886096584474, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5913238384162449, "standard_deviation": 0.2812702588090196, "minimum": 0.2747443941240703, "maximum": 1.0587126670443125, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10053747376677713, "standard_deviation": 0.0071347970349094954, "minimum": 0.08780277227367835, "maximum": 0.11155095653075563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975478448553663, "standard_deviation": 0.0006117975881326272, "minimum": 0.9967068211861072, "maximum": 0.9987171328094232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5644216460259388, "standard_deviation": 0.03381866434180653, "minimum": 0.5112339873258693, "maximum": 0.6134202707803634, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5862899512382398, "standard_deviation": 0.3212256578601942, "minimum": 0.10390835602151317, "maximum": 1.0787407876199335, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05808279330374572, "standard_deviation": 0.008536389398902371, "minimum": 0.040305697066658985, "maximum": 0.06608721954902798, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19915750492907164, "standard_deviation": 0.019783548261775034, "minimum": 0.17149915966874305, "maximum": 0.22646528792287976, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.183080965091455, "standard_deviation": 0.013909421674009626, "minimum": 5.169119349584688, "maximum": 5.206987559145688, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.32914557204663775, "standard_deviation": 0.01105859478095654, "minimum": 0.309815421045364, "maximum": 0.34600866205890785, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09337841911275699, "standard_deviation": 0.09303152952498844, "minimum": -0.24582720440124686, "maximum": 0.030378504122457673, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9948348544484666, "standard_deviation": 0.0007025159375205713, "minimum": 0.9934471434410211, "maximum": 0.9955734495701731, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0114260607900865, "standard_deviation": 0.04378854618983765, "minimum": 0.9459570012383637, "maximum": 1.0778051562063076, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.49048494556388555, "standard_deviation": 0.2594132508198552, "minimum": 0.1174846935699731, "maximum": 0.9180373405068448, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08028577239488277, "standard_deviation": 0.004861981908031793, "minimum": 0.07584470456375368, "maximum": 0.08990597527613407, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984895959351421, "standard_deviation": 0.0003144899590072966, "minimum": 0.9981197882542983, "maximum": 0.9989944669373223, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.660029618585953, "standard_deviation": 0.034945945413905284, "minimum": 0.6055325999229536, "maximum": 0.7203441611274727, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.490310183403106, "standard_deviation": 0.22119936843982652, "minimum": 0.09771418163771557, "maximum": 0.8244069678678381, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04321314806496075, "standard_deviation": 0.006095393027901869, "minimum": 0.0339341796080156, "maximum": 0.05071687163743223, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29095-0008", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2970845699310303, "predicted_mos": 2.2970845699310303, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "9c99e577370d9d7a4eb700d06fd66fa4cb7f8736b25df747e0b6a96fc6b772c9", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.908848499879241, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9280350763854831, "全帯域_10ms_rms_envelope_nrmse": 0.0784887623420381, "全帯域_10ms_envelope_correlation": 0.9949639291310449, "全帯域_10ms_modulation_mae_db": 0.2731357833768024, "全帯域_40ms_log_rms_mae_db": 0.5802318204007478, "全帯域_40ms_rms_envelope_nrmse": 0.03631721984119529, "全帯域_40ms_envelope_correlation": 0.9988554044965622, "全帯域_40ms_modulation_mae_db": 0.32255061931061424, "300–1000 Hz_10ms_log_rms_mae_db": 0.9688191416212932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09569787402317512, "300–1000 Hz_10ms_envelope_correlation": 0.9932459518585514, "300–1000 Hz_10ms_modulation_mae_db": 0.4651238773445323, "300–1000 Hz_40ms_log_rms_mae_db": 0.5761098550195216, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04580570302258149, "300–1000 Hz_40ms_envelope_correlation": 0.998366313709408, "300–1000 Hz_40ms_modulation_mae_db": 0.44375972096786453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1787953323160638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12046418404356876, "1000–3000 Hz_10ms_envelope_correlation": 0.9906472018430259, "1000–3000 Hz_10ms_modulation_mae_db": 0.3234262644656076, "1000–3000 Hz_40ms_log_rms_mae_db": 0.639319739355476, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0536042250038937, "1000–3000 Hz_40ms_envelope_correlation": 0.9981138810872582, "1000–3000 Hz_40ms_modulation_mae_db": 0.17239049501218726, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4012440743523251, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1489308199246621, "3000–6000 Hz_10ms_envelope_correlation": 0.9845183288186717, "3000–6000 Hz_10ms_modulation_mae_db": 1.1491752187140432, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7907561482275259, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08059460520391688, "3000–6000 Hz_40ms_envelope_correlation": 0.9960814963290858, "3000–6000 Hz_40ms_modulation_mae_db": 1.1694432747732277, "mr_stft_db": 5.336998347501614, "band_mae_db": 0.1764839285530525, "rms_gain_error_db": -0.04349508597023424, "output_peak": 0.3636148484627496}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0245362885018439, "全帯域_10ms_rms_envelope_nrmse": 0.08610650963133851, "全帯域_10ms_envelope_correlation": 0.9940884170307346, "全帯域_10ms_modulation_mae_db": 0.5200691641472003, "全帯域_40ms_log_rms_mae_db": 0.6778834332370179, "全帯域_40ms_rms_envelope_nrmse": 0.04551603844846007, "全帯域_40ms_envelope_correlation": 0.9984174012636425, "全帯域_40ms_modulation_mae_db": 0.6172569438290554, "300–1000 Hz_10ms_log_rms_mae_db": 1.0040966489415963, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11452116969019936, "300–1000 Hz_10ms_envelope_correlation": 0.9903689245256408, "300–1000 Hz_10ms_modulation_mae_db": 0.4737633115380023, "300–1000 Hz_40ms_log_rms_mae_db": 0.5837677389120097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059998257392964736, "300–1000 Hz_40ms_envelope_correlation": 0.9972986301137684, "300–1000 Hz_40ms_modulation_mae_db": 0.46945053644463025, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2271470525816055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13806682161678546, "1000–3000 Hz_10ms_envelope_correlation": 0.9880938659186352, "1000–3000 Hz_10ms_modulation_mae_db": 1.1684120596142042, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7678557703848198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06651750437187143, "1000–3000 Hz_40ms_envelope_correlation": 0.9977260852380637, "1000–3000 Hz_40ms_modulation_mae_db": 1.2342574982006864, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3628138616321825, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16348146945245154, "3000–6000 Hz_10ms_envelope_correlation": 0.9838040299616766, "3000–6000 Hz_10ms_modulation_mae_db": 1.5280254263978534, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7756710830706759, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1007583609615103, "3000–6000 Hz_40ms_envelope_correlation": 0.9965091681068221, "3000–6000 Hz_40ms_modulation_mae_db": 1.5745096091512103, "mr_stft_db": 5.327664219146572, "band_mae_db": 0.29069603949428685, "rms_gain_error_db": -0.155682759411275, "output_peak": 0.33452591678587185}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0304176595611565, "全帯域_10ms_rms_envelope_nrmse": 0.07585413323897985, "全帯域_10ms_envelope_correlation": 0.9953063552464759, "全帯域_10ms_modulation_mae_db": 0.3101577790528614, "全帯域_40ms_log_rms_mae_db": 0.6500333940519629, "全帯域_40ms_rms_envelope_nrmse": 0.037892060277701284, "全帯域_40ms_envelope_correlation": 0.9987325679107224, "全帯域_40ms_modulation_mae_db": 0.36519178384393397, "300–1000 Hz_10ms_log_rms_mae_db": 0.9710385160816617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10979178712268055, "300–1000 Hz_10ms_envelope_correlation": 0.9911793652414648, "300–1000 Hz_10ms_modulation_mae_db": 0.4491370929448864, "300–1000 Hz_40ms_log_rms_mae_db": 0.5629756748784817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05430730354192936, "300–1000 Hz_40ms_envelope_correlation": 0.9976880158024475, "300–1000 Hz_40ms_modulation_mae_db": 0.43354418458705024, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1829238635323671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1383941552572182, "1000–3000 Hz_10ms_envelope_correlation": 0.9876322980692568, "1000–3000 Hz_10ms_modulation_mae_db": 0.6342969862545776, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7160748429161068, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06964349850854422, "1000–3000 Hz_40ms_envelope_correlation": 0.9967305871158404, "1000–3000 Hz_40ms_modulation_mae_db": 0.6408479311191074, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3126858602246028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15117063014294016, "3000–6000 Hz_10ms_envelope_correlation": 0.9835179409871749, "3000–6000 Hz_10ms_modulation_mae_db": 0.9027196936878936, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7777775729905594, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08630388171843419, "3000–6000 Hz_40ms_envelope_correlation": 0.9946628986718009, "3000–6000 Hz_40ms_modulation_mae_db": 0.9757604634106413, "mr_stft_db": 5.324273936745802, "band_mae_db": 0.13641160345099854, "rms_gain_error_db": 0.014154797312377426, "output_peak": 0.33092432721862997}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9470512607266901, "全帯域_10ms_rms_envelope_nrmse": 0.07882707145909133, "全帯域_10ms_envelope_correlation": 0.9950314269628954, "全帯域_10ms_modulation_mae_db": 0.43498512681712914, "全帯域_40ms_log_rms_mae_db": 0.5923992381607653, "全帯域_40ms_rms_envelope_nrmse": 0.03939751796736337, "全帯域_40ms_envelope_correlation": 0.9988109493647623, "全帯域_40ms_modulation_mae_db": 0.3583229190908988, "300–1000 Hz_10ms_log_rms_mae_db": 0.9678960270334298, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10517505126027506, "300–1000 Hz_10ms_envelope_correlation": 0.992029159491962, "300–1000 Hz_10ms_modulation_mae_db": 0.5078630467142663, "300–1000 Hz_40ms_log_rms_mae_db": 0.5628325576184847, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05521491553745603, "300–1000 Hz_40ms_envelope_correlation": 0.9979133033943315, "300–1000 Hz_40ms_modulation_mae_db": 0.27582360657663885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1918834062513848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11965096454880507, "1000–3000 Hz_10ms_envelope_correlation": 0.9916586190023005, "1000–3000 Hz_10ms_modulation_mae_db": 1.0597983274347549, "1000–3000 Hz_40ms_log_rms_mae_db": 0.704077688733968, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07103441254708953, "1000–3000 Hz_40ms_envelope_correlation": 0.9977458115752253, "1000–3000 Hz_40ms_modulation_mae_db": 1.0423264945054163, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3159584385462046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14803000582797565, "3000–6000 Hz_10ms_envelope_correlation": 0.9841042897264124, "3000–6000 Hz_10ms_modulation_mae_db": 0.12720228635087483, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7462006187938574, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08372159438578998, "3000–6000 Hz_40ms_envelope_correlation": 0.9945040281824993, "3000–6000 Hz_40ms_modulation_mae_db": 0.2308265469317016, "mr_stft_db": 5.314570278576955, "band_mae_db": 0.2784125417916939, "rms_gain_error_db": -0.13256797798763476, "output_peak": 0.33219801458422343}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9828906889010586, "全帯域_10ms_rms_envelope_nrmse": 0.0815989271378413, "全帯域_10ms_envelope_correlation": 0.9947290510813439, "全帯域_10ms_modulation_mae_db": 0.4591298454926407, "全帯域_40ms_log_rms_mae_db": 0.5957282255217968, "全帯域_40ms_rms_envelope_nrmse": 0.04370982369484286, "全帯域_40ms_envelope_correlation": 0.9985650527364454, "全帯域_40ms_modulation_mae_db": 0.5742469718668941, "300–1000 Hz_10ms_log_rms_mae_db": 0.9817439708372964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10434229678057802, "300–1000 Hz_10ms_envelope_correlation": 0.9923879353716006, "300–1000 Hz_10ms_modulation_mae_db": 0.6973092306177966, "300–1000 Hz_40ms_log_rms_mae_db": 0.5995594802208032, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06216995508053619, "300–1000 Hz_40ms_envelope_correlation": 0.9975308294872334, "300–1000 Hz_40ms_modulation_mae_db": 0.5954369731146419, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1943129760257487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14647936827803706, "1000–3000 Hz_10ms_envelope_correlation": 0.9861724640969052, "1000–3000 Hz_10ms_modulation_mae_db": 0.15664259971970834, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6843085055689172, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06444562102012186, "1000–3000 Hz_40ms_envelope_correlation": 0.9972223394930076, "1000–3000 Hz_40ms_modulation_mae_db": 0.1259016018691491, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2989292564873478, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1568284040693993, "3000–6000 Hz_10ms_envelope_correlation": 0.9827925326429179, "3000–6000 Hz_10ms_modulation_mae_db": 1.374301229338614, "3000–6000 Hz_40ms_log_rms_mae_db": 0.78432589479812, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0848308520857302, "3000–6000 Hz_40ms_envelope_correlation": 0.9956385295935453, "3000–6000 Hz_40ms_modulation_mae_db": 1.587663491556263, "mr_stft_db": 5.313960461073168, "band_mae_db": 0.25829580253556655, "rms_gain_error_db": -0.15447085515833459, "output_peak": 0.34398844125194794}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9698339857702728, "全帯域_10ms_rms_envelope_nrmse": 0.0794831295456602, "全帯域_10ms_envelope_correlation": 0.9949091535477997, "全帯域_10ms_modulation_mae_db": 0.35645814352234706, "全帯域_40ms_log_rms_mae_db": 0.610460877789147, "全帯域_40ms_rms_envelope_nrmse": 0.04839957777072103, "全帯域_40ms_envelope_correlation": 0.9980482598959209, "全帯域_40ms_modulation_mae_db": 0.366313298156185, "300–1000 Hz_10ms_log_rms_mae_db": 1.0385481409461976, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1125470100698333, "300–1000 Hz_10ms_envelope_correlation": 0.9907460439492263, "300–1000 Hz_10ms_modulation_mae_db": 0.22827802919375836, "300–1000 Hz_40ms_log_rms_mae_db": 0.6274489387335012, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07089787150767146, "300–1000 Hz_40ms_envelope_correlation": 0.9962080497696124, "300–1000 Hz_40ms_modulation_mae_db": 0.27917681425415375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1396163297472857, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12306202040698636, "1000–3000 Hz_10ms_envelope_correlation": 0.990678806670512, "1000–3000 Hz_10ms_modulation_mae_db": 0.7816854271491787, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6626037157723629, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06861823958663503, "1000–3000 Hz_40ms_envelope_correlation": 0.9974833413222584, "1000–3000 Hz_40ms_modulation_mae_db": 0.8327097006545873, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3107116271382613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1448645376689887, "3000–6000 Hz_10ms_envelope_correlation": 0.9859003990197555, "3000–6000 Hz_10ms_modulation_mae_db": 0.5249523961357365, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7235815431771259, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08258932848021824, "3000–6000 Hz_40ms_envelope_correlation": 0.9956123467746798, "3000–6000 Hz_40ms_modulation_mae_db": 0.6792377560647495, "mr_stft_db": 5.316974586491999, "band_mae_db": 0.24902533419178663, "rms_gain_error_db": -0.10663230166651293, "output_peak": 0.32726207626249376}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0457734915727852, "全帯域_10ms_rms_envelope_nrmse": 0.07163299772855958, "全帯域_10ms_envelope_correlation": 0.9958153014062613, "全帯域_10ms_modulation_mae_db": 0.4938363101830803, "全帯域_40ms_log_rms_mae_db": 0.6411274244451642, "全帯域_40ms_rms_envelope_nrmse": 0.03301824916507569, "全帯域_40ms_envelope_correlation": 0.9990707336405564, "全帯域_40ms_modulation_mae_db": 0.3690614661766954, "300–1000 Hz_10ms_log_rms_mae_db": 0.9810557479572171, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09550456864319586, "300–1000 Hz_10ms_envelope_correlation": 0.9933008953043879, "300–1000 Hz_10ms_modulation_mae_db": 0.8785806458751351, "300–1000 Hz_40ms_log_rms_mae_db": 0.5334886406620456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04584458616251974, "300–1000 Hz_40ms_envelope_correlation": 0.9984130459468584, "300–1000 Hz_40ms_modulation_mae_db": 0.714113554416894, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1465093781963362, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09740002970999713, "1000–3000 Hz_10ms_envelope_correlation": 0.9939681592349163, "1000–3000 Hz_10ms_modulation_mae_db": 0.2171249027304731, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6403003786276933, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04933971499018084, "1000–3000 Hz_40ms_envelope_correlation": 0.998467705965909, "1000–3000 Hz_40ms_modulation_mae_db": 0.32448551792140107, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3070426058001747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13385529455520326, "3000–6000 Hz_10ms_envelope_correlation": 0.9875608847786049, "3000–6000 Hz_10ms_modulation_mae_db": 1.250464279235721, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7631904229454302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07246804649971533, "3000–6000 Hz_40ms_envelope_correlation": 0.9967841948926426, "3000–6000 Hz_40ms_modulation_mae_db": 1.2139724407000483, "mr_stft_db": 5.316700193660952, "band_mae_db": 0.17280873259208787, "rms_gain_error_db": -0.060378425934981376, "output_peak": 0.34653475716050963}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9644556009967319, "全帯域_10ms_rms_envelope_nrmse": 0.07429308763026343, "全帯域_10ms_envelope_correlation": 0.9954880429520239, "全帯域_10ms_modulation_mae_db": 0.20348230987824256, "全帯域_40ms_log_rms_mae_db": 0.6340674005996976, "全帯域_40ms_rms_envelope_nrmse": 0.04010253800100898, "全帯域_40ms_envelope_correlation": 0.9985850799796662, "全帯域_40ms_modulation_mae_db": 0.15762554140203278, "300–1000 Hz_10ms_log_rms_mae_db": 1.0077057818193889, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09824041294202387, "300–1000 Hz_10ms_envelope_correlation": 0.9929445620716549, "300–1000 Hz_10ms_modulation_mae_db": 0.43956531497102375, "300–1000 Hz_40ms_log_rms_mae_db": 0.559437631206766, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05325345881324278, "300–1000 Hz_40ms_envelope_correlation": 0.9978824091865797, "300–1000 Hz_40ms_modulation_mae_db": 0.3309214910987483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2232769250828368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13538357453224445, "1000–3000 Hz_10ms_envelope_correlation": 0.9881688757233921, "1000–3000 Hz_10ms_modulation_mae_db": 0.14525918359108655, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7573997033606857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06941848811138056, "1000–3000 Hz_40ms_envelope_correlation": 0.9967884545374252, "1000–3000 Hz_40ms_modulation_mae_db": 0.1521202738343982, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3376321744873914, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1570966476319489, "3000–6000 Hz_10ms_envelope_correlation": 0.9819484584596977, "3000–6000 Hz_10ms_modulation_mae_db": 0.6224661816607249, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7904659491802511, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08401722980714904, "3000–6000 Hz_40ms_envelope_correlation": 0.9945956373186025, "3000–6000 Hz_40ms_modulation_mae_db": 0.7000304512349159, "mr_stft_db": 5.312712271994447, "band_mae_db": 0.2229515610289309, "rms_gain_error_db": -0.032086174952220055, "output_peak": 0.33077139282447}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48640, "output_samples": 48640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.614300472915338, "spectral_convergence": 0.26535735257218857, "spectral_convergence_status": "ok", "amplitude_floor": 9.047519636617716e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.161108543305769, "spectral_convergence": 0.2682713509396771, "spectral_convergence_status": "ok", "amplitude_floor": 7.308511681145121e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.235586026283733, "spectral_convergence": 0.2643858504128024, "spectral_convergence_status": "ok", "amplitude_floor": 4.4367419091404295e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011152155483091157, "output_power": 0.00011859566111897014, "delta_db": 0.26709984720399116, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042085142829602937, "output_power": 0.00043338596170391, "delta_db": 0.12746034052821112, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006661514585812611, "output_power": 0.000639693373358866, "delta_db": -0.1760113130066368, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015794681682280444, "output_power": 0.00015435082699032118, "delta_db": -0.10001917072517365, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.332154632901203e-05, "output_power": 4.9138181762030695e-05, "delta_db": -0.3548365305976991, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2768891588196896e-05, "output_power": 3.302245739192909e-05, "delta_db": 0.03347636925660328, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1764839285530525, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04349508597023424, "pitch": null, "output_rms_dbfs": -28.452150439164335, "output_peak": 0.3636148484627496, "reference_rms_dbfs": -28.408655353194106, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.336998347501614}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00012692463938083332, "log_rms_mae_db": 0.9280350763854831, "rms_envelope_nrmse": 0.0784887623420381, "envelope_correlation": 0.9949639291310449, "rms_gain_error_db": -0.043495053822956306, "reference_mean_power": 0.0014449383611437946, "output_mean_power": 0.0014305393739721688}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.053245386897797e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005746543771154043, "output_power": 0.0005546914149457595, "reference_active": true, "delta_db": -0.15355274771422348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014747895972213286, "output_power": 0.0001311521421048183, "reference_active": true, "delta_db": -0.5095467723986822}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.536231970329893e-05, "output_power": 2.446573062543325e-05, "reference_active": true, "delta_db": -0.15630783001750148}], "active_band_count": 3, "active_band_mae_db": 0.2731357833768024}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 0.00010829935323771927, "log_rms_mae_db": 0.5802318204007478, "rms_envelope_nrmse": 0.03631721984119529, "envelope_correlation": 0.9988554044965622, "rms_gain_error_db": -0.043495154068554406, "reference_mean_power": 0.0014593630016451321, "output_mean_power": 0.001444820237819981}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 9.289432894914324e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005498374692309951, "output_power": 0.0005298166717085207, "reference_active": true, "delta_db": -0.16108711859152539}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.817099868184177e-05, "output_power": 5.0986767262081536e-05, "reference_active": true, "delta_db": -0.5724904278234593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4585911877580322e-06, "output_power": 1.5393629038185356e-06, "reference_active": true, "delta_db": 0.2340743115168581}], "active_band_count": 3, "active_band_mae_db": 0.32255061931061424}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00011599347403670498, "log_rms_mae_db": 0.9688191416212932, "rms_envelope_nrmse": 0.09569787402317512, "envelope_correlation": 0.9932459518585514, "rms_gain_error_db": -0.06015210556958142, "reference_mean_power": 0.0008943859828976575, "output_mean_power": 0.0008820836528805232}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 8.128232732706927e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005209748261573112, "output_power": 0.0005083823740526408, "reference_active": true, "delta_db": -0.10626253543479239}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010680702813776803, "output_power": 9.271620890995453e-05, "reference_active": true, "delta_db": -0.6144416579457561}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2431239684221878e-05, "output_power": 1.9203771601727e-05, "reference_active": true, "delta_db": -0.6746674386530485}], "active_band_count": 3, "active_band_mae_db": 0.4651238773445323}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 9.903058317661472e-05, "log_rms_mae_db": 0.5761098550195216, "rms_envelope_nrmse": 0.04580570302258149, "envelope_correlation": 0.998366313709408, "rms_gain_error_db": -0.06015211287486462, "reference_mean_power": 0.0009033149500834516, "output_mean_power": 0.0008908898001167302}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 7.323650960940424e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005139241705701183, "output_power": 0.0004992180016183982, "reference_active": true, "delta_db": -0.12608806183596025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0770880999134286e-05, "output_power": 3.416797712046113e-05, "reference_active": true, "delta_db": -0.7673082841552382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2000817341803485e-06, "output_power": 1.0849820267212371e-06, "reference_active": true, "delta_db": -0.43788281691239506}], "active_band_count": 3, "active_band_mae_db": 0.44375972096786453}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 7.992748424465529e-05, "log_rms_mae_db": 1.1787953323160638, "rms_envelope_nrmse": 0.12046418404356876, "envelope_correlation": 0.9906472018430259, "rms_gain_error_db": -0.09559731588018434, "reference_mean_power": 0.00017342262158164605, "output_mean_power": 0.00016964693413121525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.6075439307790634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.767193199363673e-05, "output_power": 9.41161196854068e-05, "reference_active": true, "delta_db": -0.1610576499815812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.270979881634546e-05, "output_power": 2.0813368083465192e-05, "reference_active": true, "delta_db": -0.37870922096720616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.175922545645147e-06, "output_power": 4.611090298943286e-06, "reference_active": true, "delta_db": 0.4305119224480355}], "active_band_count": 3, "active_band_mae_db": 0.3234262644656076}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 6.970659359658012e-05, "log_rms_mae_db": 0.639319739355476, "rms_envelope_nrmse": 0.0536042250038937, "envelope_correlation": 0.9981138810872582, "rms_gain_error_db": -0.09559764400521015, "reference_mean_power": 0.00017515363198831316, "output_mean_power": 0.00017134024473131003}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 1.4903701525512362e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.997559182587721e-05, "output_power": 9.665820205136255e-05, "reference_active": true, "delta_db": -0.14655271183798277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.524639480988942e-06, "output_power": 8.908829125293659e-06, "reference_active": true, "delta_db": -0.29027917009449716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.337301598444833e-07, "output_power": 3.2761327873617384e-07, "reference_active": true, "delta_db": -0.0803396031040818}], "active_band_count": 3, "active_band_mae_db": 0.17239049501218726}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 2.3755576133519224e-05, "log_rms_mae_db": 1.4012440743523251, "rms_envelope_nrmse": 0.1489308199246621, "envelope_correlation": 0.9845183288186717, "rms_gain_error_db": -0.3654852844011724, "reference_mean_power": 3.0329423269641833e-05, "output_mean_power": 2.7881467187136524e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 3.170053749158532e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1550798308210337e-05, "output_power": 1.0358529389338486e-05, "reference_active": true, "delta_db": -0.47313898012034056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.1283799417647094e-06, "output_power": 5.510136229463805e-06, "reference_active": true, "delta_db": -0.4618334622280802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7104244593685833e-06, "output_power": 9.59066181071265e-07, "reference_active": true, "delta_db": -2.512553213793709}], "active_band_count": 3, "active_band_mae_db": 1.1491752187140432}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 2.1008424841575618e-05, "log_rms_mae_db": 0.7907561482275259, "rms_envelope_nrmse": 0.08059460520391688, "envelope_correlation": 0.9960814963290858, "rms_gain_error_db": -0.3654895632692637, "reference_mean_power": 3.0632047924156045e-05, "output_mean_power": 2.8159638579810317e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 2.7930999787597993e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1345569218097283e-05, "output_power": 1.0214960390536153e-05, "reference_active": true, "delta_db": -0.45589602784279604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6424993904119263e-06, "output_power": 2.281253844854668e-06, "reference_active": true, "delta_db": -0.6384128202705228}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1775222693841551e-07, "output_power": 6.754092927504609e-08, "reference_active": true, "delta_db": -2.414020976206364}], "active_band_count": 3, "active_band_mae_db": 1.1694432747732277}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9196663870354668, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f5dec17e826e866bb8754c5626ae6b837caf2499b671209658e876e27fd9ce81", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.989627536319868, "standard_deviation": 0.00254977664738082, "minimum": 0.9861724640969052, "maximum": 0.9939681592349163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1855581579667036, "standard_deviation": 0.03152579726601519, "minimum": 1.1396163297472857, "maximum": 1.2271470525816055, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5608307188699488, "standard_deviation": 0.41107179669134747, "minimum": 0.14525918359108655, "maximum": 1.1684120596142042, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1273626397992053, "standard_deviation": 0.01553507782363146, "minimum": 0.09740002970999713, "maximum": 0.14647936827803706, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975347757918736, "standard_deviation": 0.0006083741618726502, "minimum": 0.9967305871158404, "maximum": 0.998467705965909, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6964925430900037, "standard_deviation": 0.049232082316014746, "minimum": 0.639319739355476, "maximum": 0.7678557703848198, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5656299391396167, "standard_deviation": 0.43563269426100326, "minimum": 0.1259016018691491, "maximum": 1.2342574982006864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06407771301746465, "standard_deviation": 0.008119679548404275, "minimum": 0.04933971499018084, "maximum": 0.07103441254708953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.984268358049364, "standard_deviation": 0.001770085343729117, "minimum": 0.9819484584596977, "maximum": 0.9875608847786049, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3308772373335613, "standard_deviation": 0.03498158828579449, "minimum": 1.2989292564873478, "maximum": 1.4012440743523251, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9349133389401827, "standard_deviation": 0.47950107873400016, "minimum": 0.12720228635087483, "maximum": 1.5280254263978534, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1505322261591962, "standard_deviation": 0.00902292553224849, "minimum": 0.13385529455520326, "maximum": 0.16348146945245154, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955485374837098, "standard_deviation": 0.0008887591890430474, "minimum": 0.9945040281824993, "maximum": 0.9967841948926426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7689961541479433, "standard_deviation": 0.023624890684534258, "minimum": 0.7235815431771259, "maximum": 0.7907561482275259, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0164305042278445, "standard_deviation": 0.467948518332415, "minimum": 0.2308265469317016, "maximum": 1.587663491556263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08441048739280801, "standard_deviation": 0.00785714701965082, "minimum": 0.07246804649971533, "maximum": 0.1007583609615103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920253547268111, "standard_deviation": 0.0011457609918051606, "minimum": 0.9903689245256408, "maximum": 0.9933008953043879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9901129969047602, "standard_deviation": 0.024824941321287975, "minimum": 0.9678960270334298, "maximum": 1.0385481409461976, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5174525686499252, "standard_deviation": 0.19341157146739732, "minimum": 0.22827802919375836, "maximum": 0.8785806458751351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10447752131649514, "standard_deviation": 0.0074760129108130844, "minimum": 0.09550456864319586, "maximum": 0.11452116969019936, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99766257467628, "standard_deviation": 0.0007005865146046914, "minimum": 0.9962080497696124, "maximum": 0.9984130459468584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5757025646564518, "standard_deviation": 0.02850119500072388, "minimum": 0.5334886406620456, "maximum": 0.6274489387335012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4427783601825777, "standard_deviation": 0.15347237654193985, "minimum": 0.27582360657663885, "maximum": 0.714113554416894, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05593650638236272, "standard_deviation": 0.00840110469178915, "minimum": 0.04580570302258149, "maximum": 0.07089787150767146, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22313569295480046, "standard_deviation": 0.05575037339651945, "minimum": 0.13641160345099854, "maximum": 0.29069603949428685, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.320481786898938, "standard_deviation": 0.008478581356441953, "minimum": 5.312712271994447, "maximum": 5.336998347501614, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.33872747181886204, "standard_deviation": 0.012094529166716167, "minimum": 0.32726207626249376, "maximum": 0.3636148484627496, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08389484797110194, "standard_deviation": 0.06268239308597516, "minimum": -0.155682759411275, "maximum": 0.014154797312377426, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9950414596698225, "standard_deviation": 0.0005207060145579642, "minimum": 0.9940884170307346, "maximum": 0.9958153014062613, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9866242565520027, "standard_deviation": 0.04249341521739075, "minimum": 0.9280350763854831, "maximum": 1.0457734915727852, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.381406807808788, "standard_deviation": 0.11329969774751365, "minimum": 0.20348230987824256, "maximum": 0.5200691641472003, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07828557733922153, "standard_deviation": 0.004472497066759279, "minimum": 0.07163299772855958, "maximum": 0.08610650963133851, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986356811610348, "standard_deviation": 0.00031119899463791267, "minimum": 0.9980482598959209, "maximum": 0.9990707336405564, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6227414767757875, "standard_deviation": 0.03350917767082561, "minimum": 0.5802318204007478, "maximum": 0.6778834332370179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3913211929595387, "standard_deviation": 0.1447877144574612, "minimum": 0.15762554140203278, "maximum": 0.6172569438290554, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.040544128145796075, "standard_deviation": 0.005063855083670599, "minimum": 0.03301824916507569, "maximum": 0.04839957777072103, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4970-29095-0008", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.9814953804016113, "predicted_mos": 1.9814953804016113, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "019cd5c246f24edda952f4fac19437ad432ce8c3d1903d99f46fc74d28f27bbf", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8366453000344336, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0412217388200704, "全帯域_10ms_rms_envelope_nrmse": 0.0781025026026463, "全帯域_10ms_envelope_correlation": 0.9950585150948217, "全帯域_10ms_modulation_mae_db": 0.40420876813231094, "全帯域_40ms_log_rms_mae_db": 0.6434581276419181, "全帯域_40ms_rms_envelope_nrmse": 0.03890886038456197, "全帯域_40ms_envelope_correlation": 0.9987561717625837, "全帯域_40ms_modulation_mae_db": 0.3555227954600079, "300–1000 Hz_10ms_log_rms_mae_db": 0.9175433270973595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09866288046842016, "300–1000 Hz_10ms_envelope_correlation": 0.9928886387076461, "300–1000 Hz_10ms_modulation_mae_db": 0.6633187235664918, "300–1000 Hz_40ms_log_rms_mae_db": 0.5455486276603617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052741223335263536, "300–1000 Hz_40ms_envelope_correlation": 0.9979399657108335, "300–1000 Hz_40ms_modulation_mae_db": 0.6721851757526105, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1894313212834018, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11436809723233772, "1000–3000 Hz_10ms_envelope_correlation": 0.9916108196334195, "1000–3000 Hz_10ms_modulation_mae_db": 0.18959258007477917, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7160318044116539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06286862482353159, "1000–3000 Hz_40ms_envelope_correlation": 0.9974324448349731, "1000–3000 Hz_40ms_modulation_mae_db": 0.2993866454235111, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2908877031409434, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13770966232470222, "3000–6000 Hz_10ms_envelope_correlation": 0.9861512141964185, "3000–6000 Hz_10ms_modulation_mae_db": 0.45921884737973806, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6907949048705284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07165046218537931, "3000–6000 Hz_40ms_envelope_correlation": 0.9960374067669879, "3000–6000 Hz_40ms_modulation_mae_db": 0.5829773054783024, "mr_stft_db": 5.280189634469683, "band_mae_db": 0.16444231583108526, "rms_gain_error_db": -0.09778082818027632, "output_peak": 0.32764780298551305}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9875528877447564, "全帯域_10ms_rms_envelope_nrmse": 0.08446756330198277, "全帯域_10ms_envelope_correlation": 0.9941547981530816, "全帯域_10ms_modulation_mae_db": 0.09762294791581046, "全帯域_40ms_log_rms_mae_db": 0.6632155563340509, "全帯域_40ms_rms_envelope_nrmse": 0.03886598004004691, "全帯域_40ms_envelope_correlation": 0.9986737059409834, "全帯域_40ms_modulation_mae_db": 0.1083782749707175, "300–1000 Hz_10ms_log_rms_mae_db": 1.002805183532302, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1128692293835473, "300–1000 Hz_10ms_envelope_correlation": 0.9905863821421971, "300–1000 Hz_10ms_modulation_mae_db": 0.17599293498391344, "300–1000 Hz_40ms_log_rms_mae_db": 0.6145901326974355, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04993491333714424, "300–1000 Hz_40ms_envelope_correlation": 0.9980311390858451, "300–1000 Hz_40ms_modulation_mae_db": 0.16851556630041942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1338379014685986, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10760748077774668, "1000–3000 Hz_10ms_envelope_correlation": 0.9927360331698182, "1000–3000 Hz_10ms_modulation_mae_db": 0.5575920987843981, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6474356949576358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06250860713914884, "1000–3000 Hz_40ms_envelope_correlation": 0.9976532145359414, "1000–3000 Hz_40ms_modulation_mae_db": 0.5875300413044319, "3000–6000 Hz_10ms_log_rms_mae_db": 1.403064987636045, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1806142457995765, "3000–6000 Hz_10ms_envelope_correlation": 0.9774488086458741, "3000–6000 Hz_10ms_modulation_mae_db": 1.7792134928848122, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8068836935263894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10761250593402603, "3000–6000 Hz_40ms_envelope_correlation": 0.9931506555159809, "3000–6000 Hz_40ms_modulation_mae_db": 1.8195576603221333, "mr_stft_db": 5.277093291478234, "band_mae_db": 0.18411079382281217, "rms_gain_error_db": -0.03556913081548194, "output_peak": 0.3354446384356939}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1108820018778145, "全帯域_10ms_rms_envelope_nrmse": 0.08668789007158574, "全帯域_10ms_envelope_correlation": 0.9938507212914292, "全帯域_10ms_modulation_mae_db": 0.3556351369579068, "全帯域_40ms_log_rms_mae_db": 0.6794656372593053, "全帯域_40ms_rms_envelope_nrmse": 0.04737741934028225, "全帯域_40ms_envelope_correlation": 0.9980409254917813, "全帯域_40ms_modulation_mae_db": 0.3671306412663367, "300–1000 Hz_10ms_log_rms_mae_db": 1.0170544960955374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10955457178932505, "300–1000 Hz_10ms_envelope_correlation": 0.9911694771636186, "300–1000 Hz_10ms_modulation_mae_db": 0.8018858587526728, "300–1000 Hz_40ms_log_rms_mae_db": 0.604165604642934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06551616549196017, "300–1000 Hz_40ms_envelope_correlation": 0.9966861495112626, "300–1000 Hz_40ms_modulation_mae_db": 0.7933088406606782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1681524590203, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12879857383805418, "1000–3000 Hz_10ms_envelope_correlation": 0.9893492367484041, "1000–3000 Hz_10ms_modulation_mae_db": 0.4180505353898161, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7513277094033912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07537266046167707, "1000–3000 Hz_40ms_envelope_correlation": 0.99615914956292, "1000–3000 Hz_40ms_modulation_mae_db": 0.5219020109624007, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2598501476498103, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13667953865462856, "3000–6000 Hz_10ms_envelope_correlation": 0.9863447310234809, "3000–6000 Hz_10ms_modulation_mae_db": 0.6057080889889191, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7080579708196086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07707984832300767, "3000–6000 Hz_40ms_envelope_correlation": 0.9953421689446075, "3000–6000 Hz_40ms_modulation_mae_db": 0.5485778899758211, "mr_stft_db": 5.273237140629938, "band_mae_db": 0.17580383360998367, "rms_gain_error_db": -0.06092961748940902, "output_peak": 0.34429171654764457}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0149096914116207, "全帯域_10ms_rms_envelope_nrmse": 0.08364726135106047, "全帯域_10ms_envelope_correlation": 0.994325576503522, "全帯域_10ms_modulation_mae_db": 0.16430986586637555, "全帯域_40ms_log_rms_mae_db": 0.6997333544248163, "全帯域_40ms_rms_envelope_nrmse": 0.04666043016318681, "全帯域_40ms_envelope_correlation": 0.9981752189945526, "全帯域_40ms_modulation_mae_db": 0.14413796194353026, "300–1000 Hz_10ms_log_rms_mae_db": 0.961170774207094, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10077994251370746, "300–1000 Hz_10ms_envelope_correlation": 0.9926153401640301, "300–1000 Hz_10ms_modulation_mae_db": 0.14745260976601418, "300–1000 Hz_40ms_log_rms_mae_db": 0.5380102150752774, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05924348791646097, "300–1000 Hz_40ms_envelope_correlation": 0.9974117204162972, "300–1000 Hz_40ms_modulation_mae_db": 0.21929659372617583, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1837231677059734, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10811176952852282, "1000–3000 Hz_10ms_envelope_correlation": 0.9926703320511369, "1000–3000 Hz_10ms_modulation_mae_db": 0.3533200446973637, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6878940689599217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06260489160444627, "1000–3000 Hz_40ms_envelope_correlation": 0.9976541513785934, "1000–3000 Hz_40ms_modulation_mae_db": 0.3490533863098224, "3000–6000 Hz_10ms_log_rms_mae_db": 1.329717393903587, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17790635916625605, "3000–6000 Hz_10ms_envelope_correlation": 0.9771037719301322, "3000–6000 Hz_10ms_modulation_mae_db": 0.9015162492378052, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7552822126648023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10003476959803485, "3000–6000 Hz_40ms_envelope_correlation": 0.9921720376855382, "3000–6000 Hz_40ms_modulation_mae_db": 0.9827781767539058, "mr_stft_db": 5.275117134635671, "band_mae_db": 0.20877486807219583, "rms_gain_error_db": -0.10412471287194468, "output_peak": 0.32581685354668644}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.014919775220278, "全帯域_10ms_rms_envelope_nrmse": 0.08731840692096432, "全帯域_10ms_envelope_correlation": 0.9937871681103375, "全帯域_10ms_modulation_mae_db": 0.4729744552918216, "全帯域_40ms_log_rms_mae_db": 0.6868031326656298, "全帯域_40ms_rms_envelope_nrmse": 0.048937225160510106, "全帯域_40ms_envelope_correlation": 0.9978920122684926, "全帯域_40ms_modulation_mae_db": 0.24206115549326399, "300–1000 Hz_10ms_log_rms_mae_db": 0.8997605558731148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10611695115523348, "300–1000 Hz_10ms_envelope_correlation": 0.9917855758857629, "300–1000 Hz_10ms_modulation_mae_db": 0.3735791197204588, "300–1000 Hz_40ms_log_rms_mae_db": 0.5057664003521974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05999225248141557, "300–1000 Hz_40ms_envelope_correlation": 0.9971919497875756, "300–1000 Hz_40ms_modulation_mae_db": 0.19959047358168366, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1696108329804116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12808734664739935, "1000–3000 Hz_10ms_envelope_correlation": 0.9898465053566328, "1000–3000 Hz_10ms_modulation_mae_db": 0.43497397143415867, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7081833743088648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07587146832960007, "1000–3000 Hz_40ms_envelope_correlation": 0.9967598970090172, "1000–3000 Hz_40ms_modulation_mae_db": 0.4953561875891346, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3366486885065354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15079364008851912, "3000–6000 Hz_10ms_envelope_correlation": 0.983365156658421, "3000–6000 Hz_10ms_modulation_mae_db": 0.522500137612131, "3000–6000 Hz_40ms_log_rms_mae_db": 0.774501480750596, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08082295113161136, "3000–6000 Hz_40ms_envelope_correlation": 0.9949437926768532, "3000–6000 Hz_40ms_modulation_mae_db": 0.5680816521892614, "mr_stft_db": 5.274460817735744, "band_mae_db": 0.22406838596646425, "rms_gain_error_db": 0.023961763559820257, "output_peak": 0.30953953601943196}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9099234915774774, "全帯域_10ms_rms_envelope_nrmse": 0.07747100429651133, "全帯域_10ms_envelope_correlation": 0.9950854186122751, "全帯域_10ms_modulation_mae_db": 0.09919636893605488, "全帯域_40ms_log_rms_mae_db": 0.6099758003378328, "全帯域_40ms_rms_envelope_nrmse": 0.040475943627833244, "全帯域_40ms_envelope_correlation": 0.9985575558917036, "全帯域_40ms_modulation_mae_db": 0.13533429238832673, "300–1000 Hz_10ms_log_rms_mae_db": 1.0036761608549754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10308796226264197, "300–1000 Hz_10ms_envelope_correlation": 0.9921504875358645, "300–1000 Hz_10ms_modulation_mae_db": 0.07927916340950522, "300–1000 Hz_40ms_log_rms_mae_db": 0.6062414772771048, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05703311458795817, "300–1000 Hz_40ms_envelope_correlation": 0.9974401908384626, "300–1000 Hz_40ms_modulation_mae_db": 0.21187549150428875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1976463019683263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.127431086532405, "1000–3000 Hz_10ms_envelope_correlation": 0.989549553959505, "1000–3000 Hz_10ms_modulation_mae_db": 0.5059585830198987, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7352104686585171, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0720288245142395, "1000–3000 Hz_40ms_envelope_correlation": 0.9964932995208041, "1000–3000 Hz_40ms_modulation_mae_db": 0.6032581392412475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2561708151489612, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1467555039523478, "3000–6000 Hz_10ms_envelope_correlation": 0.9848696370780704, "3000–6000 Hz_10ms_modulation_mae_db": 1.1317061847893797, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6965752101052879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08925975000227164, "3000–6000 Hz_40ms_envelope_correlation": 0.9946431986057986, "3000–6000 Hz_40ms_modulation_mae_db": 1.2003210601749903, "mr_stft_db": 5.258280397464741, "band_mae_db": 0.17165156264222584, "rms_gain_error_db": -0.03210724031199584, "output_peak": 0.3325506857908958}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0025105985960363, "全帯域_10ms_rms_envelope_nrmse": 0.08048382024998882, "全帯域_10ms_envelope_correlation": 0.9947204868069708, "全帯域_10ms_modulation_mae_db": 0.17903511187064458, "全帯域_40ms_log_rms_mae_db": 0.7067802345363026, "全帯域_40ms_rms_envelope_nrmse": 0.04138577854277853, "全帯域_40ms_envelope_correlation": 0.9985398281752075, "全帯域_40ms_modulation_mae_db": 0.22645467171128053, "300–1000 Hz_10ms_log_rms_mae_db": 0.998641584524473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10126555846966222, "300–1000 Hz_10ms_envelope_correlation": 0.9925264425211909, "300–1000 Hz_10ms_modulation_mae_db": 0.4277815714876083, "300–1000 Hz_40ms_log_rms_mae_db": 0.5983587947707314, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05476648036633755, "300–1000 Hz_40ms_envelope_correlation": 0.9978008606971377, "300–1000 Hz_40ms_modulation_mae_db": 0.3106109225929268, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2018622632988751, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12068730118781097, "1000–3000 Hz_10ms_envelope_correlation": 0.9908668569318811, "1000–3000 Hz_10ms_modulation_mae_db": 0.5076415422766278, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7522924009213484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06431159886783112, "1000–3000 Hz_40ms_envelope_correlation": 0.9976344230805295, "1000–3000 Hz_40ms_modulation_mae_db": 0.36652404608670164, "3000–6000 Hz_10ms_log_rms_mae_db": 1.286377909060346, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14310956938197, "3000–6000 Hz_10ms_envelope_correlation": 0.9850244200396218, "3000–6000 Hz_10ms_modulation_mae_db": 0.43963216337846106, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6912700669088171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06970745680873919, "3000–6000 Hz_40ms_envelope_correlation": 0.9961945440515338, "3000–6000 Hz_40ms_modulation_mae_db": 0.5989636628674562, "mr_stft_db": 5.284623239915247, "band_mae_db": 0.21770795840893317, "rms_gain_error_db": -0.07673060565770334, "output_peak": 0.36389221931910415}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9444794800581393, "全帯域_10ms_rms_envelope_nrmse": 0.08145286734271376, "全帯域_10ms_envelope_correlation": 0.9945922961403668, "全帯域_10ms_modulation_mae_db": 0.35204572069813267, "全帯域_40ms_log_rms_mae_db": 0.6654301430594083, "全帯域_40ms_rms_envelope_nrmse": 0.04231154536702221, "全帯域_40ms_envelope_correlation": 0.9984747793570734, "全帯域_40ms_modulation_mae_db": 0.13058106527880903, "300–1000 Hz_10ms_log_rms_mae_db": 1.017523554024678, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10781767627609579, "300–1000 Hz_10ms_envelope_correlation": 0.9915132037471541, "300–1000 Hz_10ms_modulation_mae_db": 0.4902593213499493, "300–1000 Hz_40ms_log_rms_mae_db": 0.5972158854309922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060995994352128304, "300–1000 Hz_40ms_envelope_correlation": 0.9972430762362535, "300–1000 Hz_40ms_modulation_mae_db": 0.2404047011619276, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2193350994172372, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11796031856998333, "1000–3000 Hz_10ms_envelope_correlation": 0.9911183982471489, "1000–3000 Hz_10ms_modulation_mae_db": 0.20147545029466418, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7429462828937468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07500276047606237, "1000–3000 Hz_40ms_envelope_correlation": 0.9963101511058471, "1000–3000 Hz_40ms_modulation_mae_db": 0.1407977938246712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2926824685061529, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13847265608517279, "3000–6000 Hz_10ms_envelope_correlation": 0.9861638153745463, "3000–6000 Hz_10ms_modulation_mae_db": 0.9006014205783138, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7427713942573528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06506155164418147, "3000–6000 Hz_40ms_envelope_correlation": 0.997042086407599, "3000–6000 Hz_40ms_modulation_mae_db": 1.104761703311934, "mr_stft_db": 5.271614534256666, "band_mae_db": 0.2431500016464158, "rms_gain_error_db": -0.0793170517452109, "output_peak": 0.3283972624720987}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48640, "output_samples": 48640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5717430217168635, "spectral_convergence": 0.27070042186615245, "spectral_convergence_status": "ok", "amplitude_floor": 9.047519636617716e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.062284121533152, "spectral_convergence": 0.27189457759216445, "spectral_convergence_status": "ok", "amplitude_floor": 7.308511681145121e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.206541760159034, "spectral_convergence": 0.28568458134322094, "spectral_convergence_status": "ok", "amplitude_floor": 4.4367419091404295e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011152155483091157, "output_power": 0.0001187142813967485, "delta_db": 0.2714415225527226, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042085142829602937, "output_power": 0.00040992382898632286, "delta_db": -0.11425640724355243, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006661514585812611, "output_power": 0.0006434093739346213, "delta_db": -0.15085599227787622, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015794681682280444, "output_power": 0.00015261877477193078, "delta_db": -0.14902915083066332, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.332154632901203e-05, "output_power": 5.171518746343267e-05, "delta_db": -0.1328463219258774, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2768891588196896e-05, "output_power": 3.406310251541106e-05, "delta_db": 0.1682245001558194, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16444231583108526, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09778082818027632, "pitch": null, "output_rms_dbfs": -28.50643618137438, "output_peak": 0.32764780298551305, "reference_rms_dbfs": -28.408655353194106, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.280189634469683}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00012692463938083332, "log_rms_mae_db": 1.0412217388200704, "rms_envelope_nrmse": 0.0781025026026463, "envelope_correlation": 0.9950585150948217, "rms_gain_error_db": -0.09778072004818533, "reference_mean_power": 0.0014449383611437946, "output_mean_power": 0.0014127693006170667}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.053245386897797e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005746543771154043, "output_power": 0.0005672306398313158, "reference_active": true, "delta_db": -0.056470371621600166}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014747895972213286, "output_power": 0.00013417852215435856, "reference_active": true, "delta_db": -0.410470613098457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.536231970329893e-05, "output_power": 2.1360943964729874e-05, "reference_active": true, "delta_db": -0.7456853196768756}], "active_band_count": 3, "active_band_mae_db": 0.40420876813231094}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 0.00010829935323771927, "log_rms_mae_db": 0.6434581276419181, "rms_envelope_nrmse": 0.03890886038456197, "envelope_correlation": 0.9987561717625837, "rms_gain_error_db": -0.09778051785823752, "reference_mean_power": 0.0014593630016451321, "output_mean_power": 0.0014268728678052379}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 9.289432894914324e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005498374692309951, "output_power": 0.0005428582507167556, "reference_active": true, "delta_db": -0.05547889045817048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.817099868184177e-05, "output_power": 5.3438168629479215e-05, "reference_active": true, "delta_db": -0.3685495375686637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4585911877580322e-06, "output_power": 1.2579969540338606e-06, "reference_active": true, "delta_db": -0.6425399583531894}], "active_band_count": 3, "active_band_mae_db": 0.3555227954600079}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00011599347403670498, "log_rms_mae_db": 0.9175433270973595, "rms_envelope_nrmse": 0.09866288046842016, "envelope_correlation": 0.9928886387076461, "rms_gain_error_db": -0.12753446106089356, "reference_mean_power": 0.0008943859828976575, "output_mean_power": 0.0008685034302651205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 8.128232732706927e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005209748261573112, "output_power": 0.0005187650130787296, "reference_active": true, "delta_db": -0.0184606001404377}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010680702813776803, "output_power": 9.615653525800179e-05, "reference_active": true, "delta_db": -0.45621024858972037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2431239684221878e-05, "output_power": 1.5824308598687894e-05, "reference_active": true, "delta_db": -1.5152853219693172}], "active_band_count": 3, "active_band_mae_db": 0.6633187235664918}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 9.903058317661472e-05, "log_rms_mae_db": 0.5455486276603617, "rms_envelope_nrmse": 0.052741223335263536, "envelope_correlation": 0.9979399657108335, "rms_gain_error_db": -0.12753448002148562, "reference_mean_power": 0.0009033149500834516, "output_mean_power": 0.0008771739990718034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 7.323650960940424e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005139241705701183, "output_power": 0.0005103209744720478, "reference_active": true, "delta_db": -0.030556251998204838}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0770880999134286e-05, "output_power": 3.637534206227147e-05, "reference_active": true, "delta_db": -0.49543010040571384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2000817341803485e-06, "output_power": 8.514397235640842e-07, "reference_active": true, "delta_db": -1.490569174853913}], "active_band_count": 3, "active_band_mae_db": 0.6721851757526105}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 7.992748424465529e-05, "log_rms_mae_db": 1.1894313212834018, "rms_envelope_nrmse": 0.11436809723233772, "envelope_correlation": 0.9916108196334195, "rms_gain_error_db": -0.12904520697599695, "reference_mean_power": 0.00017342262158164605, "output_mean_power": 0.00016834538932501613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.6075439307790634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.767193199363673e-05, "output_power": 9.378400842693335e-05, "reference_active": true, "delta_db": -0.1764098733758593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.270979881634546e-05, "output_power": 2.266242864367404e-05, "reference_active": true, "delta_db": -0.009068372096522535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.175922545645147e-06, "output_power": 3.8231601026347356e-06, "reference_active": true, "delta_db": -0.38329949475195557}], "active_band_count": 3, "active_band_mae_db": 0.18959258007477917}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 6.970659359658012e-05, "log_rms_mae_db": 0.7160318044116539, "rms_envelope_nrmse": 0.06286862482353159, "envelope_correlation": 0.9974324448349731, "rms_gain_error_db": -0.12904409709247394, "reference_mean_power": 0.00017515363198831316, "output_mean_power": 0.00017002576500901507}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 1.4903701525512362e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.997559182587721e-05, "output_power": 9.693036822242619e-05, "reference_active": true, "delta_db": -0.13434121154506393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.524639480988942e-06, "output_power": 9.650519871014953e-06, "reference_active": true, "delta_db": 0.057021632131197965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.337301598444833e-07, "output_power": 2.8360621720767805e-07, "reference_active": true, "delta_db": -0.7067970925942715}], "active_band_count": 3, "active_band_mae_db": 0.2993866454235111}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 2.3755576133519224e-05, "log_rms_mae_db": 1.2908877031409434, "rms_envelope_nrmse": 0.13770966232470222, "envelope_correlation": 0.9861512141964185, "rms_gain_error_db": -0.11906069392785923, "reference_mean_power": 3.0329423269641833e-05, "output_mean_power": 2.95092439483336e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 3.170053749158532e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1550798308210337e-05, "output_power": 1.1403314114605843e-05, "reference_active": true, "delta_db": -0.05580913058487896}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.1283799417647094e-06, "output_power": 5.540716401850594e-06, "reference_active": true, "delta_db": -0.43779760822510994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7104244593685833e-06, "output_power": 1.395400640742118e-06, "reference_active": true, "delta_db": -0.8840498033292252}], "active_band_count": 3, "active_band_mae_db": 0.45921884737973806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 2.1008424841575618e-05, "log_rms_mae_db": 0.6907949048705284, "rms_envelope_nrmse": 0.07165046218537931, "envelope_correlation": 0.9960374067669879, "rms_gain_error_db": -0.11906503863007739, "reference_mean_power": 3.0632047924156045e-05, "output_mean_power": 2.9803655100881798e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 2.7930999787597993e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1345569218097283e-05, "output_power": 1.1169986168864486e-05, "reference_active": true, "delta_db": -0.06773654383875508}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6424993904119263e-06, "output_power": 2.524959341147639e-06, "reference_active": true, "delta_db": -0.19760506629118718}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1775222693841551e-07, "output_power": 8.367777016593967e-08, "reference_active": true, "delta_db": -1.4835903063049651}], "active_band_count": 3, "active_band_mae_db": 0.5829773054783024}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9268752605210397, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1ce87dc0b5c5f5535d14f7748903c5bc0bd09ac46272cc8be7b4ab73eb861703", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9909684670122434, "standard_deviation": 0.001328805825909642, "minimum": 0.9893492367484041, "maximum": 0.9927360331698182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1829499183928904, "standard_deviation": 0.026046756271969982, "minimum": 1.1338379014685986, "maximum": 1.2193350994172372, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3960756007464633, "standard_deviation": 0.1388407531725389, "minimum": 0.18959258007477917, "maximum": 0.5575920987843981, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1191314967892825, "standard_deviation": 0.008645909935378785, "minimum": 0.10760748077774668, "maximum": 0.12879857383805418, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970120913785783, "standard_deviation": 0.0006481016896835451, "minimum": 0.99615914956292, "maximum": 0.9976541513785934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.717665225564385, "standard_deviation": 0.036233870912333876, "minimum": 0.6474356949576358, "maximum": 0.7522924009213484, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.42047603134274014, "standard_deviation": 0.15951049012233387, "minimum": 0.1407977938246712, "maximum": 0.6032581392412475, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0688211795270671, "standard_deviation": 0.006272268158201974, "minimum": 0.06250860713914884, "maximum": 0.07587146832960007, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9833089443683206, "standard_deviation": 0.0038484119333517643, "minimum": 0.9771037719301322, "maximum": 0.9863447310234809, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3069250141940476, "standard_deviation": 0.04828508294437145, "minimum": 1.2561708151489612, "maximum": 1.403064987636045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.842512073106195, "standard_deviation": 0.4528652170097005, "minimum": 0.43963216337846106, "maximum": 1.7792134928848122, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15150514693164663, "standard_deviation": 0.017799978295165386, "minimum": 0.13667953865462856, "maximum": 0.1806142457995765, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9949407363318624, "standard_deviation": 0.0016173240547177128, "minimum": 0.9921720376855382, "maximum": 0.997042086407599, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7332671167379228, "standard_deviation": 0.04349730191787594, "minimum": 0.6907949048705284, "maximum": 0.8068836935263894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9257523888842255, "standard_deviation": 0.4478594615994316, "minimum": 0.5485778899758211, "maximum": 1.8195576603221333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08265366195340645, "standard_deviation": 0.015124545204622794, "minimum": 0.06506155164418147, "maximum": 0.10761250593402603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991904443483433, "standard_deviation": 0.000789764344246178, "minimum": 0.9905863821421971, "maximum": 0.9928886387076461, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9772719545261918, "standard_deviation": 0.04604873927366017, "minimum": 0.8997605558731148, "maximum": 1.017523554024678, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3949436628795767, "standard_deviation": 0.25553369627381334, "minimum": 0.07927916340950522, "maximum": 0.8018858587526728, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10501934653982918, "standard_deviation": 0.004890554912189131, "minimum": 0.09866288046842016, "maximum": 0.1128692293835473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974681315354584, "standard_deviation": 0.00044604591604115884, "minimum": 0.9966861495112626, "maximum": 0.9980311390858451, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5762371422383793, "standard_deviation": 0.04044377850191204, "minimum": 0.5057664003521974, "maximum": 0.6145901326974355, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3519734706600888, "standard_deviation": 0.24070695684316282, "minimum": 0.16851556630041942, "maximum": 0.7933088406606782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05752795398358356, "standard_deviation": 0.004975544621818398, "minimum": 0.04993491333714424, "maximum": 0.06551616549196017, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1987137150000145, "standard_deviation": 0.028596822645687407, "minimum": 0.16444231583108526, "maximum": 0.2431500016464158, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.27432702382324, "standard_deviation": 0.0076970898449087236, "minimum": 5.258280397464741, "maximum": 5.284623239915247, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3334475893896336, "standard_deviation": 0.015758125777261583, "minimum": 0.30953953601943196, "maximum": 0.36389221931910415, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.057824677939025224, "standard_deviation": 0.04211362529793433, "minimum": -0.10412471287194468, "maximum": 0.023961763559820257, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944468725891005, "standard_deviation": 0.0005026340494361669, "minimum": 0.9937871681103375, "maximum": 0.9950854186122751, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0032999581632742, "standard_deviation": 0.0606033592944114, "minimum": 0.9099234915774774, "maximum": 1.1108820018778145, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2656285469586322, "standard_deviation": 0.1470934430378817, "minimum": 0.09762294791581046, "maximum": 0.4729744552918216, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08245391451718169, "standard_deviation": 0.003697803213116975, "minimum": 0.07747100429651133, "maximum": 0.08731840692096432, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983887747352973, "standard_deviation": 0.00031356145037105747, "minimum": 0.9978920122684926, "maximum": 0.9987561717625837, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.669357748282408, "standard_deviation": 0.0315770926860438, "minimum": 0.6099758003378328, "maximum": 0.7067802345363026, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.21370010731403408, "standard_deviation": 0.10254141192927961, "minimum": 0.1083782749707175, "maximum": 0.3671306412663367, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04311539782827775, "standard_deviation": 0.00398179327947785, "minimum": 0.03886598004004691, "maximum": 0.048937225160510106, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4992-23283-0016", "language": "en", "style": "normal", "speaker": "librispeech_4992", "source_utterance_id": "4992-23283-0016", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 71920, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4992-23283-0016.wav", "source_file_sha256": "6fa2445be2e2c330cd3ea90114781f5b092d29e49c240ef3f318312286086034", "decoded_pcm_sha256": "4af74b1b8682fb324d2e81cb2f0cc0dc50b77d4a3b44fd510cfb4df09ae57d6d", "original_source_sha256": "e7442a8f1bbd834ad37ee3c9cc5fa26b1cfb61446f827b54298716f59b010002", "corpus_entry_id": "librispeech_4992-23283-0016"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.6870718002319336, "predicted_mos": 3.6870718002319336, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "3cdb94e1da84305f01e63847deb8e69762ad8006b40fa32d05ff8533b0693535", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.842964009003481, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4992-23283-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "3cdb94e1da84305f01e63847deb8e69762ad8006b40fa32d05ff8533b0693535", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "421415266b6d6a08914ccb879fcef3847317f4a6ef5cf11d1a65a92b0b747247"}, "rows": [{"case": "librispeech_4992-23283-0016", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.5118801593780518, "predicted_mos": 2.5118801593780518, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "08b12ef8c4491f501aac75aeb5541c4c3b4fe6ce8015a68486b3c3bd7a0b8497", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3267180850016302, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4992-23283-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "08b12ef8c4491f501aac75aeb5541c4c3b4fe6ce8015a68486b3c3bd7a0b8497", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1895908237143207, "全帯域_10ms_rms_envelope_nrmse": 0.0737965840217634, "全帯域_10ms_envelope_correlation": 0.9943900721815288, "全帯域_10ms_modulation_mae_db": 0.2704819479725986, "全帯域_40ms_log_rms_mae_db": 0.6676847596833934, "全帯域_40ms_rms_envelope_nrmse": 0.036823071221479015, "全帯域_40ms_envelope_correlation": 0.9985555041968565, "全帯域_40ms_modulation_mae_db": 0.34297411960554774, "300–1000 Hz_10ms_log_rms_mae_db": 1.2961407446001387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0900095086869385, "300–1000 Hz_10ms_envelope_correlation": 0.9934642558040438, "300–1000 Hz_10ms_modulation_mae_db": 0.5079820497100723, "300–1000 Hz_40ms_log_rms_mae_db": 0.7823181666406602, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04772687432233516, "300–1000 Hz_40ms_envelope_correlation": 0.9981881874038763, "300–1000 Hz_40ms_modulation_mae_db": 0.9360133157409626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4038465335194434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13618914760483944, "1000–3000 Hz_10ms_envelope_correlation": 0.9873269580211301, "1000–3000 Hz_10ms_modulation_mae_db": 0.3793403167949497, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8972573975838656, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07523467542693593, "1000–3000 Hz_40ms_envelope_correlation": 0.9967544835661238, "1000–3000 Hz_40ms_modulation_mae_db": 0.42680474071695723, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7066040398564446, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21302429989240404, "3000–6000 Hz_10ms_envelope_correlation": 0.9698751657322965, "3000–6000 Hz_10ms_modulation_mae_db": 0.6908901793527065, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5042641609892315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1390003283870487, "3000–6000 Hz_40ms_envelope_correlation": 0.9879058386659588, "3000–6000 Hz_40ms_modulation_mae_db": 0.8808069473514313, "mr_stft_db": 5.702166029533075, "band_mae_db": 0.25893526009829165, "rms_gain_error_db": -0.0763547464854987, "output_peak": 0.32798290730587415}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1306726318087974, "全帯域_10ms_rms_envelope_nrmse": 0.08250866028757221, "全帯域_10ms_envelope_correlation": 0.9929938891353022, "全帯域_10ms_modulation_mae_db": 0.3200158020698338, "全帯域_40ms_log_rms_mae_db": 0.6646354844838193, "全帯域_40ms_rms_envelope_nrmse": 0.0515617610456809, "全帯域_40ms_envelope_correlation": 0.997119631647813, "全帯域_40ms_modulation_mae_db": 0.3485653619825098, "300–1000 Hz_10ms_log_rms_mae_db": 1.3120631647073246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1052495333634083, "300–1000 Hz_10ms_envelope_correlation": 0.9910109596018, "300–1000 Hz_10ms_modulation_mae_db": 0.459122437993047, "300–1000 Hz_40ms_log_rms_mae_db": 0.8684239668498376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07119735953358358, "300–1000 Hz_40ms_envelope_correlation": 0.9957248508902288, "300–1000 Hz_40ms_modulation_mae_db": 0.7034936132782126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3996399780106572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.124663652441339, "1000–3000 Hz_10ms_envelope_correlation": 0.9893175612767398, "1000–3000 Hz_10ms_modulation_mae_db": 1.0780872100186067, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9189571941948772, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06700022265467762, "1000–3000 Hz_40ms_envelope_correlation": 0.9973457065942533, "1000–3000 Hz_40ms_modulation_mae_db": 0.8006633594175687, "3000–6000 Hz_10ms_log_rms_mae_db": 1.682329074266544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19956767320302116, "3000–6000 Hz_10ms_envelope_correlation": 0.9730020211743864, "3000–6000 Hz_10ms_modulation_mae_db": 0.6543370697149228, "3000–6000 Hz_40ms_log_rms_mae_db": 1.454799982988652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1251747673379354, "3000–6000 Hz_40ms_envelope_correlation": 0.9892037034916954, "3000–6000 Hz_40ms_modulation_mae_db": 0.9222800992081702, "mr_stft_db": 5.691322796361244, "band_mae_db": 0.1703071449887185, "rms_gain_error_db": -0.08503008430374007, "output_peak": 0.3218736332334208}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1495406415580294, "全帯域_10ms_rms_envelope_nrmse": 0.07095559995280724, "全帯域_10ms_envelope_correlation": 0.994950636186341, "全帯域_10ms_modulation_mae_db": 0.5316912155825947, "全帯域_40ms_log_rms_mae_db": 0.618704494010991, "全帯域_40ms_rms_envelope_nrmse": 0.03535849406604452, "全帯域_40ms_envelope_correlation": 0.9988400214933713, "全帯域_40ms_modulation_mae_db": 0.734229457095012, "300–1000 Hz_10ms_log_rms_mae_db": 1.2542208905402652, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09363356816266334, "300–1000 Hz_10ms_envelope_correlation": 0.9929680064856038, "300–1000 Hz_10ms_modulation_mae_db": 0.6321211586705514, "300–1000 Hz_40ms_log_rms_mae_db": 0.8160513170039412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050663513756467306, "300–1000 Hz_40ms_envelope_correlation": 0.9980094250818804, "300–1000 Hz_40ms_modulation_mae_db": 1.2388756078401642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4157019556989603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13706272464686253, "1000–3000 Hz_10ms_envelope_correlation": 0.98772793078185, "1000–3000 Hz_10ms_modulation_mae_db": 0.6816988555737319, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9425223776089493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0841125812096877, "1000–3000 Hz_40ms_envelope_correlation": 0.9963479384295432, "1000–3000 Hz_40ms_modulation_mae_db": 0.7731584893325909, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7247931920209485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20143797619538123, "3000–6000 Hz_10ms_envelope_correlation": 0.9725555364341377, "3000–6000 Hz_10ms_modulation_mae_db": 0.5206575143660962, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4593889014601258, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1200555771233834, "3000–6000 Hz_40ms_envelope_correlation": 0.9903064622875525, "3000–6000 Hz_40ms_modulation_mae_db": 0.756150262579734, "mr_stft_db": 5.684974042503413, "band_mae_db": 0.2276066333461391, "rms_gain_error_db": -0.1302853994472758, "output_peak": 0.29845553226265437}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1622573128288722, "全帯域_10ms_rms_envelope_nrmse": 0.06795415195935307, "全帯域_10ms_envelope_correlation": 0.9952864347460493, "全帯域_10ms_modulation_mae_db": 0.19452907276412887, "全帯域_40ms_log_rms_mae_db": 0.6115109682033278, "全帯域_40ms_rms_envelope_nrmse": 0.031863824277912844, "全帯域_40ms_envelope_correlation": 0.9989710380353508, "全帯域_40ms_modulation_mae_db": 0.44773396612385025, "300–1000 Hz_10ms_log_rms_mae_db": 1.3569268909640906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09371017072174821, "300–1000 Hz_10ms_envelope_correlation": 0.9928427778232688, "300–1000 Hz_10ms_modulation_mae_db": 0.4042699019472848, "300–1000 Hz_40ms_log_rms_mae_db": 0.9084215407278305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04822923888901806, "300–1000 Hz_40ms_envelope_correlation": 0.9980737075645978, "300–1000 Hz_40ms_modulation_mae_db": 0.7811256320153689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4754040575280452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12298276158380114, "1000–3000 Hz_10ms_envelope_correlation": 0.9907201446889057, "1000–3000 Hz_10ms_modulation_mae_db": 0.22655016681476214, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9705285935899853, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0807311345376691, "1000–3000 Hz_40ms_envelope_correlation": 0.9970761503153822, "1000–3000 Hz_40ms_modulation_mae_db": 0.2787811500264871, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6928515992384465, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1989127140485344, "3000–6000 Hz_10ms_envelope_correlation": 0.9732919038639436, "3000–6000 Hz_10ms_modulation_mae_db": 0.5995200480607888, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3778749095243663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1191467599467963, "3000–6000 Hz_40ms_envelope_correlation": 0.9905313900026289, "3000–6000 Hz_40ms_modulation_mae_db": 0.7095892928749583, "mr_stft_db": 5.69563813780929, "band_mae_db": 0.22955824986112092, "rms_gain_error_db": -0.08702941500242493, "output_peak": 0.29325653778322686}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1907362453556751, "全帯域_10ms_rms_envelope_nrmse": 0.07359321972418673, "全帯域_10ms_envelope_correlation": 0.9943938164445613, "全帯域_10ms_modulation_mae_db": 0.24324514378827353, "全帯域_40ms_log_rms_mae_db": 0.6571882306699914, "全帯域_40ms_rms_envelope_nrmse": 0.03767098039098375, "全帯域_40ms_envelope_correlation": 0.9984509806524204, "全帯域_40ms_modulation_mae_db": 0.3459293068678558, "300–1000 Hz_10ms_log_rms_mae_db": 1.277145303483407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09111081035654951, "300–1000 Hz_10ms_envelope_correlation": 0.9933190133282407, "300–1000 Hz_10ms_modulation_mae_db": 0.502629771850576, "300–1000 Hz_40ms_log_rms_mae_db": 0.8304993929431773, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055463718381661545, "300–1000 Hz_40ms_envelope_correlation": 0.9974978288075598, "300–1000 Hz_40ms_modulation_mae_db": 0.8869140402357502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4295266136291038, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12694938310663326, "1000–3000 Hz_10ms_envelope_correlation": 0.9886621814253503, "1000–3000 Hz_10ms_modulation_mae_db": 0.7035182067155761, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8967447638655964, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07004469230434063, "1000–3000 Hz_40ms_envelope_correlation": 0.9967162017759554, "1000–3000 Hz_40ms_modulation_mae_db": 0.8704666753457758, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7285047101677538, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19880550318641443, "3000–6000 Hz_10ms_envelope_correlation": 0.9760155070247399, "3000–6000 Hz_10ms_modulation_mae_db": 0.7588979619066701, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4639344525440605, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12244061338381684, "3000–6000 Hz_40ms_envelope_correlation": 0.9912092393012849, "3000–6000 Hz_40ms_modulation_mae_db": 0.9286327558691595, "mr_stft_db": 5.677733638280896, "band_mae_db": 0.1504041961604584, "rms_gain_error_db": -0.054967081078444065, "output_peak": 0.3276528329091352}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.168649262495065, "全帯域_10ms_rms_envelope_nrmse": 0.0778957067048499, "全帯域_10ms_envelope_correlation": 0.9937026083196105, "全帯域_10ms_modulation_mae_db": 0.34093647146048717, "全帯域_40ms_log_rms_mae_db": 0.6446879804522724, "全帯域_40ms_rms_envelope_nrmse": 0.0454410594593084, "全帯域_40ms_envelope_correlation": 0.9977021512444827, "全帯域_40ms_modulation_mae_db": 0.2973537440447442, "300–1000 Hz_10ms_log_rms_mae_db": 1.3469392366551936, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09165652506472251, "300–1000 Hz_10ms_envelope_correlation": 0.9931397962025983, "300–1000 Hz_10ms_modulation_mae_db": 0.6423809245040961, "300–1000 Hz_40ms_log_rms_mae_db": 0.8528616796733068, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05437846265937707, "300–1000 Hz_40ms_envelope_correlation": 0.9974618064115203, "300–1000 Hz_40ms_modulation_mae_db": 0.9580696748966296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4365090873514443, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12386732464778863, "1000–3000 Hz_10ms_envelope_correlation": 0.9896568211563322, "1000–3000 Hz_10ms_modulation_mae_db": 0.421803681151534, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0005674468876673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07565112493440762, "1000–3000 Hz_40ms_envelope_correlation": 0.9966342100335092, "1000–3000 Hz_40ms_modulation_mae_db": 0.3583761365515137, "3000–6000 Hz_10ms_log_rms_mae_db": 1.772082019999498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22382900479600273, "3000–6000 Hz_10ms_envelope_correlation": 0.9660250017888439, "3000–6000 Hz_10ms_modulation_mae_db": 0.2623860926667827, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5081761476675462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1468634838271066, "3000–6000 Hz_40ms_envelope_correlation": 0.9845773667544424, "3000–6000 Hz_40ms_modulation_mae_db": 0.6086786800066982, "mr_stft_db": 5.683257319995799, "band_mae_db": 0.16725504791690446, "rms_gain_error_db": -0.0382061716815364, "output_peak": 0.31531172553598796}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.137174294379966, "全帯域_10ms_rms_envelope_nrmse": 0.07290192083258415, "全帯域_10ms_envelope_correlation": 0.994516088032536, "全帯域_10ms_modulation_mae_db": 0.33947404408563236, "全帯域_40ms_log_rms_mae_db": 0.6168861964912632, "全帯域_40ms_rms_envelope_nrmse": 0.03460072599851033, "全帯域_40ms_envelope_correlation": 0.9987067897904337, "全帯域_40ms_modulation_mae_db": 0.5447448332065177, "300–1000 Hz_10ms_log_rms_mae_db": 1.3430107078478386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08950083006876865, "300–1000 Hz_10ms_envelope_correlation": 0.9934714197544992, "300–1000 Hz_10ms_modulation_mae_db": 0.6486047764359174, "300–1000 Hz_40ms_log_rms_mae_db": 0.834673893553006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045535184034136, "300–1000 Hz_40ms_envelope_correlation": 0.9982672452275143, "300–1000 Hz_40ms_modulation_mae_db": 1.1302630616108502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4963735141414745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15068349164848877, "1000–3000 Hz_10ms_envelope_correlation": 0.9842743750387357, "1000–3000 Hz_10ms_modulation_mae_db": 1.0439416651343603, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0525332132563308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08575916624819253, "1000–3000 Hz_40ms_envelope_correlation": 0.9954907189603672, "1000–3000 Hz_40ms_modulation_mae_db": 1.4339848413324754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7275020077441592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2127362647550419, "3000–6000 Hz_10ms_envelope_correlation": 0.9698423455612549, "3000–6000 Hz_10ms_modulation_mae_db": 0.5461181645202419, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4556326134146813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12634931254375054, "3000–6000 Hz_40ms_envelope_correlation": 0.9902483066125713, "3000–6000 Hz_40ms_modulation_mae_db": 0.6252112323520929, "mr_stft_db": 5.685203472689518, "band_mae_db": 0.198333747117262, "rms_gain_error_db": -0.05491559435512599, "output_peak": 0.32065375817416397}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1709952116678224, "全帯域_10ms_rms_envelope_nrmse": 0.07771129350909133, "全帯域_10ms_envelope_correlation": 0.9939016516550162, "全帯域_10ms_modulation_mae_db": 0.1610509327687112, "全帯域_40ms_log_rms_mae_db": 0.6520550030302086, "全帯域_40ms_rms_envelope_nrmse": 0.038761329243823556, "全帯域_40ms_envelope_correlation": 0.9985608546892819, "全帯域_40ms_modulation_mae_db": 0.501563135035581, "300–1000 Hz_10ms_log_rms_mae_db": 1.3159834558072854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09618046389264198, "300–1000 Hz_10ms_envelope_correlation": 0.992691763531239, "300–1000 Hz_10ms_modulation_mae_db": 0.44021986328289925, "300–1000 Hz_40ms_log_rms_mae_db": 0.8661676289917449, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052220467445398766, "300–1000 Hz_40ms_envelope_correlation": 0.9980303618240365, "300–1000 Hz_40ms_modulation_mae_db": 0.9700800168556437, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4366187718382761, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12679540471688672, "1000–3000 Hz_10ms_envelope_correlation": 0.9889108507073261, "1000–3000 Hz_10ms_modulation_mae_db": 0.708730794372885, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9280994838750958, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06830813210242949, "1000–3000 Hz_40ms_envelope_correlation": 0.9972054201753157, "1000–3000 Hz_40ms_modulation_mae_db": 0.5464906621014832, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6979488377314804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19046200412474337, "3000–6000 Hz_10ms_envelope_correlation": 0.9759405246370779, "3000–6000 Hz_10ms_modulation_mae_db": 0.4194833582446669, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4322064919392763, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10617852832065401, "3000–6000 Hz_40ms_envelope_correlation": 0.9921434804020278, "3000–6000 Hz_40ms_modulation_mae_db": 0.5882008471268111, "mr_stft_db": 5.69786255457722, "band_mae_db": 0.16438622050950655, "rms_gain_error_db": -0.1332324520679374, "output_peak": 0.3327268250662582}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 71920, "output_samples": 71920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6465499563175205, "spectral_convergence": 0.2793955915094199, "spectral_convergence_status": "ok", "amplitude_floor": 7.580332405627278e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.56308875385439, "spectral_convergence": 0.22704156655592767, "spectral_convergence_status": "ok", "amplitude_floor": 6.818562526348602e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.896859378427313, "spectral_convergence": 0.23145793829633715, "spectral_convergence_status": "ok", "amplitude_floor": 5.715517707749406e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020741335082478346, "output_power": 0.00021477201721649365, "delta_db": 0.15140988499192937, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006538184008046484, "output_power": 0.0006511241020735132, "delta_db": -0.017933674957607323, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004887436095925812, "output_power": 0.000467738607435432, "delta_db": -0.19077873640891407, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 3.3579219734638266e-05, "output_power": 3.021502530478572e-05, "delta_db": -0.458476381100026, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.100095928785279e-05, "output_power": 5.6947794539412117e-05, "delta_db": -0.2985975519746843, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.8128919162253983e-05, "output_power": 1.639571570858841e-05, "delta_db": -0.4364153311565885, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25893526009829165, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0763547464854987, "pitch": null, "output_rms_dbfs": -28.42484827483424, "output_peak": 0.32798290730587415, "reference_rms_dbfs": -28.34849352834874, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.702166029533075}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 71920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 0.00010701965100918563, "log_rms_mae_db": 1.1895908237143207, "rms_envelope_nrmse": 0.0737965840217634, "envelope_correlation": 0.9943900721815288, "rms_gain_error_db": -0.07635475091794815, "reference_mean_power": 0.0014643132072841782, "output_mean_power": 0.0014387936233899135}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 5.509210445243936e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038836857080924806, "output_power": 0.0003745790874974863, "reference_active": true, "delta_db": -0.15700549515436532}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3200931448411458e-05, "output_power": 2.0231455198685985e-05, "reference_active": true, "delta_db": -0.5947829924751966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.17358638897755e-06, "output_power": 9.300469883142477e-06, "reference_active": true, "delta_db": 0.05965735628823389}], "active_band_count": 3, "active_band_mae_db": 0.2704819479725986}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 0.00010161161302678278, "log_rms_mae_db": 0.6676847596833934, "rms_envelope_nrmse": 0.036823071221479015, "envelope_correlation": 0.9985555041968565, "rms_gain_error_db": -0.07635453336761802, "reference_mean_power": 0.0014741623423244587, "output_mean_power": 0.0014484711834016663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 4.939959887293225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035657053350397007, "output_power": 0.0003419642509083308, "reference_active": true, "delta_db": -0.18164743731596672}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.1722039793266e-06, "output_power": 7.895402122103394e-06, "reference_active": true, "delta_db": -0.14964944109574002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.984441222185337e-07, "output_power": 5.947974786005115e-07, "reference_active": true, "delta_db": -0.6976254804049364}], "active_band_count": 3, "active_band_mae_db": 0.34297411960554774}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 9.15514590150253e-05, "log_rms_mae_db": 1.2961407446001387, "rms_envelope_nrmse": 0.0900095086869385, "envelope_correlation": 0.9934642558040438, "rms_gain_error_db": -0.1250080992608429, "reference_mean_power": 0.0006933001790048939, "output_mean_power": 0.0006736285764874784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 3.8791214865659925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002719560122465178, "output_power": 0.000254297227640779, "reference_active": true, "delta_db": -0.2915703888445193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.187110727790427e-05, "output_power": 1.784931717088741e-05, "reference_active": true, "delta_db": -0.8824916406094343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.698791354460839e-06, "output_power": 6.180278402746489e-06, "reference_active": true, "delta_db": -0.3498841196762632}], "active_band_count": 3, "active_band_mae_db": 0.5079820497100723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 8.797925167476037e-05, "log_rms_mae_db": 0.7823181666406602, "rms_envelope_nrmse": 0.04772687432233516, "envelope_correlation": 0.9981881874038763, "rms_gain_error_db": -0.12500812545425227, "reference_mean_power": 0.000697963583829786, "output_mean_power": 0.0006781596584245713}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.5511880459540643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002583969788516632, "output_power": 0.00023981139936071153, "reference_active": true, "delta_db": -0.3241760832322428}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.105856462043575e-06, "output_power": 7.0388460886611615e-06, "reference_active": true, "delta_db": -0.6129743998830549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388944870715119e-07, "output_power": 4.152792143745539e-07, "reference_active": true, "delta_db": -1.8708894641075897}], "active_band_count": 3, "active_band_mae_db": 0.9360133157409626}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 4.592612274041097e-05, "log_rms_mae_db": 1.4038465335194434, "rms_envelope_nrmse": 0.13618914760483944, "envelope_correlation": 0.9873269580211301, "rms_gain_error_db": -0.34468084128078924, "reference_mean_power": 8.288074215773485e-05, "output_mean_power": 7.655711291825955e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 4.287875385442721e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2411523150580774e-05, "output_power": 1.9259035153696306e-05, "reference_active": true, "delta_db": -0.6583684753389453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.113877152498063e-06, "output_power": 2.876904048715343e-06, "reference_active": true, "delta_db": -0.34376097524233706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.881539635700238e-07, "output_power": 9.577131861651961e-07, "reference_active": true, "delta_db": -0.13589149980356668}], "active_band_count": 3, "active_band_mae_db": 0.3793403167949497}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 4.2997336393594785e-05, "log_rms_mae_db": 0.8972573975838656, "rms_envelope_nrmse": 0.07523467542693593, "envelope_correlation": 0.9967544835661238, "rms_gain_error_db": -0.34468082890250207, "reference_mean_power": 8.343814618382176e-05, "output_mean_power": 7.707198838946989e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.9683446107037247e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.063395086340567e-05, "output_power": 1.7778354748159235e-05, "reference_active": true, "delta_db": -0.646908241664742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4408129681959194e-06, "output_power": 1.3892660420230652e-06, "reference_active": true, "delta_db": -0.15822188407123022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9896369199765786e-08, "output_power": 4.472398930037258e-08, "reference_active": true, "delta_db": -0.47528409641489944}], "active_band_count": 3, "active_band_mae_db": 0.42680474071695723}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 2.431018241105218e-05, "log_rms_mae_db": 1.7066040398564446, "rms_envelope_nrmse": 0.21302429989240404, "envelope_correlation": 0.9698751657322965, "rms_gain_error_db": -0.45876954056564545, "reference_mean_power": 1.5917473583289094e-05, "output_mean_power": 1.4321786119692614e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 8.500689890351627e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.331755201814068e-06, "output_power": 3.902842502030963e-06, "reference_active": true, "delta_db": -0.45282879773823165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3672340908387822e-06, "output_power": 1.2969268882535205e-06, "reference_active": true, "delta_db": -0.22927384355270589}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.790743563061608e-07, "output_power": 3.843941072949849e-07, "reference_active": true, "delta_db": 1.390567896767182}], "active_band_count": 3, "active_band_mae_db": 0.6908901793527065}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 1.765504974955162e-05, "log_rms_mae_db": 1.5042641609892315, "rms_envelope_nrmse": 0.1390003283870487, "envelope_correlation": 0.9879058386659588, "rms_gain_error_db": -0.4587699852345512, "reference_mean_power": 1.6024474415231737e-05, "output_mean_power": 1.4418058905920757e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 7.931595726775402e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.29977249297526e-06, "output_power": 3.879674542499666e-06, "reference_active": true, "delta_db": -0.44650181956047297}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.410418312822663e-07, "output_power": 7.396964302694088e-07, "reference_active": true, "delta_db": -0.007892014568595207}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9106902570925117e-08, "output_power": 3.1622266169128904e-08, "reference_active": true, "delta_db": 2.1880270079252258}], "active_band_count": 3, "active_band_mae_db": 0.8808069473514313}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8905708107189653, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "93452c7d1adeb74da91663ea7df0ccf09c12a36fd8316a873fbe1184a2b3742b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9883246028870463, "standard_deviation": 0.0019534839457559174, "minimum": 0.9842743750387357, "maximum": 0.9907201446889057, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4367025639646755, "standard_deviation": 0.03381492945499392, "minimum": 1.3996399780106572, "maximum": 1.4963735141414745, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6554588620720507, "standard_deviation": 0.3047642200795873, "minimum": 0.22655016681476214, "maximum": 1.0780872100186067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13114923629957992, "standard_deviation": 0.009564309722666777, "minimum": 0.12298276158380114, "maximum": 0.15068349164848877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966963537313063, "standard_deviation": 0.0005864708634167714, "minimum": 0.9954907189603672, "maximum": 0.9973457065942533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.950901308857796, "standard_deviation": 0.05425094642980015, "minimum": 0.8967447638655964, "maximum": 1.0525332132563308, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6860907568531065, "standard_deviation": 0.3730306622800889, "minimum": 0.2787811500264871, "maximum": 1.4339848413324754, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07585521617729257, "standard_deviation": 0.007167438117697713, "minimum": 0.06700022265467762, "maximum": 0.08575916624819253, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9720685007770851, "standard_deviation": 0.0033693401633018306, "minimum": 0.9660250017888439, "maximum": 0.9760155070247399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7165769351281592, "standard_deviation": 0.028264623376631732, "minimum": 1.682329074266544, "maximum": 1.772082019999498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5565362986041095, "standard_deviation": 0.15906758598799597, "minimum": 0.2623860926667827, "maximum": 0.7588979619066701, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2048469300251929, "standard_deviation": 0.010740853334714875, "minimum": 0.19046200412474337, "maximum": 0.22382900479600273, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9895157234397702, "standard_deviation": 0.0023616906597888117, "minimum": 0.9845773667544424, "maximum": 0.9921434804020278, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4570347075659924, "standard_deviation": 0.04103811930930929, "minimum": 1.3778749095243663, "maximum": 1.5081761476675462, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7524437646711319, "standard_deviation": 0.14245126710974557, "minimum": 0.5882008471268111, "maximum": 0.9286327558691595, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1256511713588115, "standard_deviation": 0.012489916547753643, "minimum": 0.10617852832065401, "maximum": 0.1468634838271066, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9928634990664117, "standard_deviation": 0.0008006297792165345, "minimum": 0.9910109596018, "maximum": 0.9934714197544992, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.312803799325693, "standard_deviation": 0.03588938222438043, "minimum": 1.2542208905402652, "maximum": 1.3569268909640906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5296663605493055, "standard_deviation": 0.09801638677648182, "minimum": 0.4042699019472848, "maximum": 0.6486047764359174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09388142628968012, "standard_deviation": 0.00509086102376569, "minimum": 0.08950083006876865, "maximum": 0.1052495333634083, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976566766514018, "standard_deviation": 0.0008356727257179204, "minimum": 0.9957248508902288, "maximum": 0.9982672452275143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.844927198297938, "standard_deviation": 0.03810650625897066, "minimum": 0.7823181666406602, "maximum": 0.9084215407278305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9506043703091978, "standard_deviation": 0.17315092892833414, "minimum": 0.7034936132782126, "maximum": 1.2388756078401642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05317685237774719, "standard_deviation": 0.00802869923478486, "minimum": 0.045535184034136, "maximum": 0.07119735953358358, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1958483124998002, "standard_deviation": 0.03901656813278468, "minimum": 0.1504041961604584, "maximum": 0.25893526009829165, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.689769748968807, "standard_deviation": 0.008342152766103153, "minimum": 5.677733638280896, "maximum": 5.702166029533075, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3172392190338402, "standard_deviation": 0.014285777765869354, "minimum": 0.29325653778322686, "maximum": 0.3327268250662582, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08250261805274792, "standard_deviation": 0.034646812659342115, "minimum": -0.1332324520679374, "maximum": -0.0382061716815364, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9942668995876182, "standard_deviation": 0.0007243428778045504, "minimum": 0.9929938891353022, "maximum": 0.9952864347460493, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1624520529760685, "standard_deviation": 0.022243902984558875, "minimum": 1.1306726318087974, "maximum": 1.1907362453556751, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3001780788115326, "standard_deviation": 0.11454168407913444, "minimum": 0.1610509327687112, "maximum": 0.5316912155825947, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.074664642124026, "standard_deviation": 0.004552029581794929, "minimum": 0.06795415195935307, "maximum": 0.08250866028757221, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983633714687513, "standard_deviation": 0.0006304285333500228, "minimum": 0.997119631647813, "maximum": 0.9989710380353508, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6416691396281584, "standard_deviation": 0.022719317616505814, "minimum": 0.6115109682033278, "maximum": 0.6676847596833934, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4453867404952023, "standard_deviation": 0.14547052508921424, "minimum": 0.2973537440447442, "maximum": 0.734229457095012, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03901015571296792, "standard_deviation": 0.00642968249961029, "minimum": 0.031863824277912844, "maximum": 0.0515617610456809, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-23283-0016", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.4443726539611816, "predicted_mos": 2.4443726539611816, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "d4e78ff31900ee25e344014c5a9b83d5e5c19224b03b07349e364a0c70eee950", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9157011001370847, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2503875693673656, "全帯域_10ms_rms_envelope_nrmse": 0.0747150627733529, "全帯域_10ms_envelope_correlation": 0.9942647534862588, "全帯域_10ms_modulation_mae_db": 0.33170890705781847, "全帯域_40ms_log_rms_mae_db": 0.7038498434973199, "全帯域_40ms_rms_envelope_nrmse": 0.038246753323050955, "全帯域_40ms_envelope_correlation": 0.9984548887815313, "全帯域_40ms_modulation_mae_db": 0.27407694788309594, "300–1000 Hz_10ms_log_rms_mae_db": 1.2723420883202616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09636586797179215, "300–1000 Hz_10ms_envelope_correlation": 0.9924831531191755, "300–1000 Hz_10ms_modulation_mae_db": 0.6163910016821116, "300–1000 Hz_40ms_log_rms_mae_db": 0.7370991570874621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05181642142708618, "300–1000 Hz_40ms_envelope_correlation": 0.9978340284001379, "300–1000 Hz_40ms_modulation_mae_db": 1.1577063129886584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3225759137879365, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13853031182523293, "1000–3000 Hz_10ms_envelope_correlation": 0.9870164265551205, "1000–3000 Hz_10ms_modulation_mae_db": 0.384490308317378, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7770060902939815, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07920190592668101, "1000–3000 Hz_40ms_envelope_correlation": 0.9964882706161476, "1000–3000 Hz_40ms_modulation_mae_db": 0.4229070591081217, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5532970781790565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20194937659481207, "3000–6000 Hz_10ms_envelope_correlation": 0.9734551523881811, "3000–6000 Hz_10ms_modulation_mae_db": 0.7406786886325852, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3092646560239392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13568992518343573, "3000–6000 Hz_40ms_envelope_correlation": 0.9890176478992451, "3000–6000 Hz_40ms_modulation_mae_db": 0.9837594067881558, "mr_stft_db": 5.607414662172384, "band_mae_db": 0.27828055529032314, "rms_gain_error_db": -0.08542608316332613, "output_peak": 0.3272565609098552}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1430748273202378, "全帯域_10ms_rms_envelope_nrmse": 0.0836991941456895, "全帯域_10ms_envelope_correlation": 0.9927891034554742, "全帯域_10ms_modulation_mae_db": 0.2052468958630729, "全帯域_40ms_log_rms_mae_db": 0.6737814578316287, "全帯域_40ms_rms_envelope_nrmse": 0.0526623253591842, "全帯域_40ms_envelope_correlation": 0.9969956759546963, "全帯域_40ms_modulation_mae_db": 0.2070796086882202, "300–1000 Hz_10ms_log_rms_mae_db": 1.304473564172558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11040693621278716, "300–1000 Hz_10ms_envelope_correlation": 0.9900982568669315, "300–1000 Hz_10ms_modulation_mae_db": 0.5414941711228244, "300–1000 Hz_40ms_log_rms_mae_db": 0.815348130607872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07351863201343284, "300–1000 Hz_40ms_envelope_correlation": 0.9954471533536563, "300–1000 Hz_40ms_modulation_mae_db": 0.8608641791246211, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3066261787035252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12190639691303304, "1000–3000 Hz_10ms_envelope_correlation": 0.9897800950694249, "1000–3000 Hz_10ms_modulation_mae_db": 1.0902183379356571, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7492251988156976, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06173969013157905, "1000–3000 Hz_40ms_envelope_correlation": 0.9978285619570268, "1000–3000 Hz_40ms_modulation_mae_db": 0.7942589486401541, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5335646188101126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18123856155587337, "3000–6000 Hz_10ms_envelope_correlation": 0.9779822608449118, "3000–6000 Hz_10ms_modulation_mae_db": 0.5799568214448066, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3086202307492207, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11325582473107074, "3000–6000 Hz_40ms_envelope_correlation": 0.9915161952459075, "3000–6000 Hz_40ms_modulation_mae_db": 0.8507762089411767, "mr_stft_db": 5.614454496564481, "band_mae_db": 0.1852375608795901, "rms_gain_error_db": -0.09086170320497611, "output_peak": 0.31992365095911224}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1774875015825084, "全帯域_10ms_rms_envelope_nrmse": 0.07080342034822944, "全帯域_10ms_envelope_correlation": 0.9949584798157566, "全帯域_10ms_modulation_mae_db": 0.47368431345921663, "全帯域_40ms_log_rms_mae_db": 0.6564709699885102, "全帯域_40ms_rms_envelope_nrmse": 0.035435229663349985, "全帯域_40ms_envelope_correlation": 0.9988152614360744, "全帯域_40ms_modulation_mae_db": 0.6949702438549994, "300–1000 Hz_10ms_log_rms_mae_db": 1.2030466578889716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09629767153310932, "300–1000 Hz_10ms_envelope_correlation": 0.992485353438797, "300–1000 Hz_10ms_modulation_mae_db": 0.6499976462862174, "300–1000 Hz_40ms_log_rms_mae_db": 0.7462389453273752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052403004963120045, "300–1000 Hz_40ms_envelope_correlation": 0.9977688769015666, "300–1000 Hz_40ms_modulation_mae_db": 1.2464293496393457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3048503124443893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13802290772898548, "1000–3000 Hz_10ms_envelope_correlation": 0.9875789194217897, "1000–3000 Hz_10ms_modulation_mae_db": 0.7263334268824391, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7925215727732887, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08387939263712252, "1000–3000 Hz_40ms_envelope_correlation": 0.9964471008670968, "1000–3000 Hz_40ms_modulation_mae_db": 0.7789461681863951, "3000–6000 Hz_10ms_log_rms_mae_db": 1.543013469664684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18819820728610298, "3000–6000 Hz_10ms_envelope_correlation": 0.9761448790445374, "3000–6000 Hz_10ms_modulation_mae_db": 0.5230670266920484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2666053523092646, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11322607312645533, "3000–6000 Hz_40ms_envelope_correlation": 0.9914872654400735, "3000–6000 Hz_40ms_modulation_mae_db": 0.8761432553454348, "mr_stft_db": 5.605007962273277, "band_mae_db": 0.23338729493887078, "rms_gain_error_db": -0.12570249615324966, "output_peak": 0.2964876539111115}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1901483668177848, "全帯域_10ms_rms_envelope_nrmse": 0.06855382917060239, "全帯域_10ms_envelope_correlation": 0.9952044157850125, "全帯域_10ms_modulation_mae_db": 0.21988819220400582, "全帯域_40ms_log_rms_mae_db": 0.6371604580487009, "全帯域_40ms_rms_envelope_nrmse": 0.033113592437654465, "全帯域_40ms_envelope_correlation": 0.9988821386480787, "全帯域_40ms_modulation_mae_db": 0.46613315556374885, "300–1000 Hz_10ms_log_rms_mae_db": 1.2914515346761373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09661694878003736, "300–1000 Hz_10ms_envelope_correlation": 0.9923765275527119, "300–1000 Hz_10ms_modulation_mae_db": 0.37850631201526336, "300–1000 Hz_40ms_log_rms_mae_db": 0.8471085439604107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050118888699433324, "300–1000 Hz_40ms_envelope_correlation": 0.9978949913806966, "300–1000 Hz_40ms_modulation_mae_db": 1.0158472288692042, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3662177933841242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12204043202413495, "1000–3000 Hz_10ms_envelope_correlation": 0.9905628163918335, "1000–3000 Hz_10ms_modulation_mae_db": 0.27122479364780333, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8157389600961852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07783187940706181, "1000–3000 Hz_40ms_envelope_correlation": 0.9970578493135118, "1000–3000 Hz_40ms_modulation_mae_db": 0.27851551573819006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5387552679835286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18924099219680873, "3000–6000 Hz_10ms_envelope_correlation": 0.9761586857886403, "3000–6000 Hz_10ms_modulation_mae_db": 0.7268973312765487, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2567129578214637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12371584378627078, "3000–6000 Hz_40ms_envelope_correlation": 0.9901500280876186, "3000–6000 Hz_40ms_modulation_mae_db": 0.902791047344995, "mr_stft_db": 5.609261594514014, "band_mae_db": 0.23194813537136097, "rms_gain_error_db": -0.08737169362812044, "output_peak": 0.294695982387826}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2295846073194283, "全帯域_10ms_rms_envelope_nrmse": 0.07551368138525588, "全帯域_10ms_envelope_correlation": 0.9940946189515584, "全帯域_10ms_modulation_mae_db": 0.2137016746217638, "全帯域_40ms_log_rms_mae_db": 0.6713415627333915, "全帯域_40ms_rms_envelope_nrmse": 0.03924076009362596, "全帯域_40ms_envelope_correlation": 0.9983161751738875, "全帯域_40ms_modulation_mae_db": 0.3032518315222637, "300–1000 Hz_10ms_log_rms_mae_db": 1.2247455656013222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09832914257416238, "300–1000 Hz_10ms_envelope_correlation": 0.9921668978809746, "300–1000 Hz_10ms_modulation_mae_db": 0.6059948722152518, "300–1000 Hz_40ms_log_rms_mae_db": 0.7558730491762422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05958047982534548, "300–1000 Hz_40ms_envelope_correlation": 0.997065215298194, "300–1000 Hz_40ms_modulation_mae_db": 0.996853914153491, "1000–3000 Hz_10ms_log_rms_mae_db": 1.329127812234594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13270658475782451, "1000–3000 Hz_10ms_envelope_correlation": 0.9874322986538977, "1000–3000 Hz_10ms_modulation_mae_db": 0.6539635809626488, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7553357390096865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0699518912861819, "1000–3000 Hz_40ms_envelope_correlation": 0.9965354788789164, "1000–3000 Hz_40ms_modulation_mae_db": 0.7946506863389838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5427918021614042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19575126768981876, "3000–6000 Hz_10ms_envelope_correlation": 0.9767324424938265, "3000–6000 Hz_10ms_modulation_mae_db": 0.9111274383201243, "3000–6000 Hz_40ms_log_rms_mae_db": 1.265345483901997, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12010592684234865, "3000–6000 Hz_40ms_envelope_correlation": 0.9914637306197269, "3000–6000 Hz_40ms_modulation_mae_db": 1.1192463299948339, "mr_stft_db": 5.605093553245143, "band_mae_db": 0.13296788165044676, "rms_gain_error_db": -0.05578651909051105, "output_peak": 0.3298909707193925}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.20383018253888, "全帯域_10ms_rms_envelope_nrmse": 0.0766327803841628, "全帯域_10ms_envelope_correlation": 0.9939039364272854, "全帯域_10ms_modulation_mae_db": 0.2747093588917782, "全帯域_40ms_log_rms_mae_db": 0.6613184975250505, "全帯域_40ms_rms_envelope_nrmse": 0.044446929647838994, "全帯域_40ms_envelope_correlation": 0.9977934588621834, "全帯域_40ms_modulation_mae_db": 0.23201075369765944, "300–1000 Hz_10ms_log_rms_mae_db": 1.2831090653966553, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09326929794365638, "300–1000 Hz_10ms_envelope_correlation": 0.9928833780074504, "300–1000 Hz_10ms_modulation_mae_db": 0.7567523221249406, "300–1000 Hz_40ms_log_rms_mae_db": 0.81302282060484, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05517360437143669, "300–1000 Hz_40ms_envelope_correlation": 0.9973575556388096, "300–1000 Hz_40ms_modulation_mae_db": 1.1602017207032753, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3120994071950451, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12317970193966073, "1000–3000 Hz_10ms_envelope_correlation": 0.9897541737413089, "1000–3000 Hz_10ms_modulation_mae_db": 0.4748982541984743, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8034997165495485, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07465624674171441, "1000–3000 Hz_40ms_envelope_correlation": 0.9967177995304903, "1000–3000 Hz_40ms_modulation_mae_db": 0.4383570594896214, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6203833905247789, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20876715264309306, "3000–6000 Hz_10ms_envelope_correlation": 0.9704264471952192, "3000–6000 Hz_10ms_modulation_mae_db": 0.33378736669311687, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3343915803308088, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14478577081889066, "3000–6000 Hz_40ms_envelope_correlation": 0.9849846797670297, "3000–6000 Hz_40ms_modulation_mae_db": 0.6412250320678948, "mr_stft_db": 5.595187802795845, "band_mae_db": 0.17323134259836537, "rms_gain_error_db": -0.025393640031477985, "output_peak": 0.3159013670270146}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1480460771604044, "全帯域_10ms_rms_envelope_nrmse": 0.07269724157049429, "全帯域_10ms_envelope_correlation": 0.9945413421270031, "全帯域_10ms_modulation_mae_db": 0.30294933902230375, "全帯域_40ms_log_rms_mae_db": 0.6116969150424976, "全帯域_40ms_rms_envelope_nrmse": 0.03380139621247389, "全帯域_40ms_envelope_correlation": 0.9987631387244564, "全帯域_40ms_modulation_mae_db": 0.5020180952574266, "300–1000 Hz_10ms_log_rms_mae_db": 1.2884606774410017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09234849475176954, "300–1000 Hz_10ms_envelope_correlation": 0.9930409859896986, "300–1000 Hz_10ms_modulation_mae_db": 0.6624432555713696, "300–1000 Hz_40ms_log_rms_mae_db": 0.7465765331109036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04556574229825469, "300–1000 Hz_40ms_envelope_correlation": 0.9982604350847502, "300–1000 Hz_40ms_modulation_mae_db": 1.2086252006877307, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3534403141459244, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1498553736846159, "1000–3000 Hz_10ms_envelope_correlation": 0.9844948624382472, "1000–3000 Hz_10ms_modulation_mae_db": 1.0335478666189044, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8826856236560057, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08644352979433835, "1000–3000 Hz_40ms_envelope_correlation": 0.9954478223234332, "1000–3000 Hz_40ms_modulation_mae_db": 1.3611803184507778, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5540444930571267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19973675303395622, "3000–6000 Hz_10ms_envelope_correlation": 0.973683169014507, "3000–6000 Hz_10ms_modulation_mae_db": 0.5506355278949974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3098186894081068, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12120963042888619, "3000–6000 Hz_40ms_envelope_correlation": 0.9912434516942894, "3000–6000 Hz_40ms_modulation_mae_db": 0.6044204136961514, "mr_stft_db": 5.592078115935364, "band_mae_db": 0.20440781247701365, "rms_gain_error_db": -0.05381565976661152, "output_peak": 0.31811215242320806}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1808617992594277, "全帯域_10ms_rms_envelope_nrmse": 0.07706567055301689, "全帯域_10ms_envelope_correlation": 0.9940172419113643, "全帯域_10ms_modulation_mae_db": 0.10234488590775799, "全帯域_40ms_log_rms_mae_db": 0.660690125362523, "全帯域_40ms_rms_envelope_nrmse": 0.038290890989634535, "全帯域_40ms_envelope_correlation": 0.9986129116585044, "全帯域_40ms_modulation_mae_db": 0.4343919327779058, "300–1000 Hz_10ms_log_rms_mae_db": 1.275162289547616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09729378168383718, "300–1000 Hz_10ms_envelope_correlation": 0.9924635249646274, "300–1000 Hz_10ms_modulation_mae_db": 0.5252018921116414, "300–1000 Hz_40ms_log_rms_mae_db": 0.8010198891543016, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05151857580522791, "300–1000 Hz_40ms_envelope_correlation": 0.9980338602366202, "300–1000 Hz_40ms_modulation_mae_db": 1.1160703121675561, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3421937478621846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1260252325226833, "1000–3000 Hz_10ms_envelope_correlation": 0.9892486141648348, "1000–3000 Hz_10ms_modulation_mae_db": 0.8090093992131114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7528525130881953, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06763632292873434, "1000–3000 Hz_40ms_envelope_correlation": 0.9975250223805272, "1000–3000 Hz_40ms_modulation_mae_db": 0.6645760784546838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5233077315319696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1761128034410467, "3000–6000 Hz_10ms_envelope_correlation": 0.9791856165190684, "3000–6000 Hz_10ms_modulation_mae_db": 0.4288145643417778, "3000–6000 Hz_40ms_log_rms_mae_db": 1.247444906663631, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10446072477556177, "3000–6000 Hz_40ms_envelope_correlation": 0.9922352359512135, "3000–6000 Hz_40ms_modulation_mae_db": 0.6240682518433313, "mr_stft_db": 5.618144188507311, "band_mae_db": 0.1632014580906306, "rms_gain_error_db": -0.13582441580997637, "output_peak": 0.3324195904876777}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 71920, "output_samples": 71920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.550415048080089, "spectral_convergence": 0.2805717768421111, "spectral_convergence_status": "ok", "amplitude_floor": 7.580332405627278e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.466769580782194, "spectral_convergence": 0.22617011780475071, "spectral_convergence_status": "ok", "amplitude_floor": 6.818562526348602e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.805059357654866, "spectral_convergence": 0.23102202934930044, "spectral_convergence_status": "ok", "amplitude_floor": 5.715517707749406e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020741335082478346, "output_power": 0.0002132813695855491, "delta_db": 0.12116213190826305, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006538184008046484, "output_power": 0.0006534534776250361, "delta_db": -0.0024246544266084574, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004887436095925812, "output_power": 0.00046485096861573547, "delta_db": -0.21767351724699108, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 3.3579219734638266e-05, "output_power": 2.9982922251569067e-05, "delta_db": -0.4919664181690635, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.100095928785279e-05, "output_power": 5.6474224431311285e-05, "delta_db": -0.33486389280831685, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.8128919162253983e-05, "output_power": 1.615149177988243e-05, "delta_db": -0.5015927171826958, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27828055529032314, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08542608316332613, "pitch": null, "output_rms_dbfs": -28.43391961151206, "output_peak": 0.3272565609098552, "reference_rms_dbfs": -28.34849352834874, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.607414662172384}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 71920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 0.00010701965100918563, "log_rms_mae_db": 1.2503875693673656, "rms_envelope_nrmse": 0.0747150627733529, "envelope_correlation": 0.9942647534862588, "rms_gain_error_db": -0.08542609157804046, "reference_mean_power": 0.0014643132072841782, "output_mean_power": 0.0014357914748146794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 5.509210445243936e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038836857080924806, "output_power": 0.0003764306422127019, "reference_active": true, "delta_db": -0.13559108217989557}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3200931448411458e-05, "output_power": 1.9999194995668043e-05, "reference_active": true, "delta_db": -0.6449290600779839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.17358638897755e-06, "output_power": 9.638286086620063e-06, "reference_active": true, "delta_db": 0.21460657891557597}], "active_band_count": 3, "active_band_mae_db": 0.33170890705781847}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 0.00010161161302678278, "log_rms_mae_db": 0.7038498434973199, "rms_envelope_nrmse": 0.038246753323050955, "envelope_correlation": 0.9984548887815313, "rms_gain_error_db": -0.08542586306025467, "reference_mean_power": 0.0014741623423244587, "output_mean_power": 0.0014454488455369109}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 4.939959887293225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035657053350397007, "output_power": 0.00034364855119125085, "reference_active": true, "delta_db": -0.16030933545971623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.1722039793266e-06, "output_power": 7.83946137114312e-06, "reference_active": true, "delta_db": -0.18052973672057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.984441222185337e-07, "output_power": 6.251618817418058e-07, "reference_active": true, "delta_db": -0.4813917714690017}], "active_band_count": 3, "active_band_mae_db": 0.27407694788309594}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 9.15514590150253e-05, "log_rms_mae_db": 1.2723420883202616, "rms_envelope_nrmse": 0.09636586797179215, "envelope_correlation": 0.9924831531191755, "rms_gain_error_db": -0.13046171827436182, "reference_mean_power": 0.0006933001790048939, "output_mean_power": 0.0006727832035637155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 3.8791214865659925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002719560122465178, "output_power": 0.00025686787068217575, "reference_active": true, "delta_db": -0.24788878760522773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.187110727790427e-05, "output_power": 1.6785679501399473e-05, "reference_active": true, "delta_db": -1.1493184418399078}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.698791354460839e-06, "output_power": 6.036703890653302e-06, "reference_active": true, "delta_db": -0.4519657756011991}], "active_band_count": 3, "active_band_mae_db": 0.6163910016821116}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 8.797925167476037e-05, "log_rms_mae_db": 0.7370991570874621, "rms_envelope_nrmse": 0.05181642142708618, "envelope_correlation": 0.9978340284001379, "rms_gain_error_db": -0.1304617418658655, "reference_mean_power": 0.000697963583829786, "output_mean_power": 0.0006773085996068727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.5511880459540643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002583969788516632, "output_power": 0.0002419462357954959, "reference_active": true, "delta_db": -0.28568561895887223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.105856462043575e-06, "output_power": 6.666946136406437e-06, "reference_active": true, "delta_db": -0.8487196262892898}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388944870715119e-07, "output_power": 3.7287028210254556e-07, "reference_active": true, "delta_db": -2.3387136937178137}], "active_band_count": 3, "active_band_mae_db": 1.1577063129886584}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 4.592612274041097e-05, "log_rms_mae_db": 1.3225759137879365, "rms_envelope_nrmse": 0.13853031182523293, "envelope_correlation": 0.9870164265551205, "rms_gain_error_db": -0.37599444269378335, "reference_mean_power": 8.288074215773485e-05, "output_mean_power": 7.600710427326643e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 4.287875385442721e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2411523150580774e-05, "output_power": 1.9296221964479016e-05, "reference_active": true, "delta_db": -0.6499908724190664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.113877152498063e-06, "output_power": 2.780233541220683e-06, "reference_active": true, "delta_db": -0.4922019651138079}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.881539635700238e-07, "output_power": 9.907234113080915e-07, "reference_active": true, "delta_db": 0.01127808741925965}], "active_band_count": 3, "active_band_mae_db": 0.384490308317378}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 4.2997336393594785e-05, "log_rms_mae_db": 0.7770060902939815, "rms_envelope_nrmse": 0.07920190592668101, "envelope_correlation": 0.9964882706161476, "rms_gain_error_db": -0.3759942547764732, "reference_mean_power": 8.343814618382176e-05, "output_mean_power": 7.651828382162727e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.9683446107037247e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.063395086340567e-05, "output_power": 1.7800084761348692e-05, "reference_active": true, "delta_db": -0.6416032164171861}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4408129681959194e-06, "output_power": 1.3543185834673268e-06, "reference_active": true, "delta_db": -0.2688677100744382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9896369199765786e-08, "output_power": 4.5945599168827015e-08, "reference_active": true, "delta_db": -0.35825025083274087}], "active_band_count": 3, "active_band_mae_db": 0.4229070591081217}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 2.431018241105218e-05, "log_rms_mae_db": 1.5532970781790565, "rms_envelope_nrmse": 0.20194937659481207, "envelope_correlation": 0.9734551523881811, "rms_gain_error_db": -0.5339290566444478, "reference_mean_power": 1.5917473583289094e-05, "output_mean_power": 1.4076063978267798e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 8.500689890351627e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.331755201814068e-06, "output_power": 3.892663267450867e-06, "reference_active": true, "delta_db": -0.4641706863890568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3672340908387822e-06, "output_power": 1.3126179877915037e-06, "reference_active": true, "delta_db": -0.1770452719183686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.790743563061608e-07, "output_power": 4.0160761819877676e-07, "reference_active": true, "delta_db": 1.5808201075903303}], "active_band_count": 3, "active_band_mae_db": 0.7406786886325852}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 1.765504974955162e-05, "log_rms_mae_db": 1.3092646560239392, "rms_envelope_nrmse": 0.13568992518343573, "envelope_correlation": 0.9890176478992451, "rms_gain_error_db": -0.5339291542294851, "reference_mean_power": 1.6024474415231737e-05, "output_mean_power": 1.41706861229161e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 7.931595726775402e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.29977249297526e-06, "output_power": 3.867755724890125e-06, "reference_active": true, "delta_db": -0.459864394307525}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.410418312822663e-07, "output_power": 7.477491745114171e-07, "reference_active": true, "delta_db": 0.03913217973194542}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9106902570925117e-08, "output_power": 3.360612618148414e-08, "reference_active": true, "delta_db": 2.452281646324997}], "active_band_count": 3, "active_band_mae_db": 0.9837594067881558}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8911743602495455, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "01406d44fcddd3f522b7d522fec4108a8bee0c4c22bae84befce0ef3feaa9de1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9882335258045571, "standard_deviation": 0.0019888853472164922, "minimum": 0.9844948624382472, "maximum": 0.9905628163918335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3296414349697154, "standard_deviation": 0.022564153357701733, "minimum": 1.3048503124443893, "maximum": 1.3662177933841242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.680460745972052, "standard_deviation": 0.29501367049691235, "minimum": 0.27122479364780333, "maximum": 1.0902183379356571, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13153336767452137, "standard_deviation": 0.010078700153511812, "minimum": 0.12190639691303304, "maximum": 0.1498553736846159, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967559882333938, "standard_deviation": 0.0007330379986098659, "minimum": 0.9954478223234332, "maximum": 0.9978285619570268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7911081767853236, "standard_deviation": 0.044472893837770845, "minimum": 0.7492251988156976, "maximum": 0.8826856236560057, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6916739793008659, "standard_deviation": 0.3348394887827035, "minimum": 0.27851551573819006, "maximum": 1.3611803184507778, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07516760735667667, "standard_deviation": 0.008374365050969959, "minimum": 0.06173969013157905, "maximum": 0.08644352979433835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9754710816611115, "standard_deviation": 0.0028114704582771796, "minimum": 0.9704264471952192, "maximum": 0.9791856165190684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5511447314890827, "standard_deviation": 0.02971382482489039, "minimum": 1.5233077315319696, "maximum": 1.6203833905247789, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5993705956620007, "standard_deviation": 0.1857884441301275, "minimum": 0.33378736669311687, "maximum": 0.9111274383201243, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.192624389305189, "standard_deviation": 0.01096237257246588, "minimum": 0.1761128034410467, "maximum": 0.20876715264309306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.990262279338138, "standard_deviation": 0.0023539147702790245, "minimum": 0.9849846797670297, "maximum": 0.9922352359512135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.287275482151054, "standard_deviation": 0.031838805364760726, "minimum": 1.247444906663631, "maximum": 1.3343915803308088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8253037432527467, "standard_deviation": 0.18666944297757393, "minimum": 0.6044204136961514, "maximum": 1.1192463299948339, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12205621496161498, "standard_deviation": 0.012954894324776143, "minimum": 0.10446072477556177, "maximum": 0.14478577081889066, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922497597275458, "standard_deviation": 0.0009130079249787088, "minimum": 0.9900982568669315, "maximum": 0.9930409859896986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2678489303805653, "standard_deviation": 0.03522797353051818, "minimum": 1.2030466578889716, "maximum": 1.304473564172558, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5920976841412025, "standard_deviation": 0.1126640953878897, "minimum": 0.37850631201526336, "maximum": 0.7567523221249406, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09761601768139394, "standard_deviation": 0.005544106263835165, "minimum": 0.09234849475176954, "maximum": 0.11040693621278716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974577645368039, "standard_deviation": 0.0008954163463388278, "minimum": 0.9954471533536563, "maximum": 0.9982604350847502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7827858836286758, "standard_deviation": 0.04123595740219363, "minimum": 0.7370991570874621, "maximum": 0.8471085439604107, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0953247772917354, "standard_deviation": 0.12827726243738502, "minimum": 0.8608641791246211, "maximum": 1.2464293496393457, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054961918675417144, "standard_deviation": 0.00850016302487162, "minimum": 0.04556574229825469, "maximum": 0.07351863201343284, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20033275516207516, "standard_deviation": 0.04638920958493739, "minimum": 0.13296788165044676, "maximum": 0.27828055529032314, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.605830297000978, "standard_deviation": 0.008814735637345845, "minimum": 5.592078115935364, "maximum": 5.618144188507311, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31683599110314975, "standard_deviation": 0.014333476815356426, "minimum": 0.294695982387826, "maximum": 0.3324195904876777, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08252277635603116, "standard_deviation": 0.03705150097788901, "minimum": -0.13582441580997637, "maximum": -0.025393640031477985, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9942217364949641, "standard_deviation": 0.0007390977842527457, "minimum": 0.9927891034554742, "maximum": 0.9952044157850125, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1904276164207546, "standard_deviation": 0.037043614425923986, "minimum": 1.1430748273202378, "maximum": 1.2503875693673656, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2655291958784647, "standard_deviation": 0.10969120132996925, "minimum": 0.10234488590775799, "maximum": 0.47368431345921663, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0749601100413505, "standard_deviation": 0.004588936525403506, "minimum": 0.06855382917060239, "maximum": 0.0836991941456895, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983292061549265, "standard_deviation": 0.0006427690377998719, "minimum": 0.9969956759546963, "maximum": 0.9988821386480787, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6595387287537027, "standard_deviation": 0.027016311700148295, "minimum": 0.6116969150424976, "maximum": 0.7038498434973199, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.389241571155665, "standard_deviation": 0.16591782514701706, "minimum": 0.2070796086882202, "maximum": 0.6949702438549994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.039404734715851625, "standard_deviation": 0.006442368017799595, "minimum": 0.033113592437654465, "maximum": 0.0526623253591842, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-23283-0016", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.4139208793640137, "predicted_mos": 2.4139208793640137, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "96a5b5d89501efb07196020d413f0833c62f07bbbd2928e3a1c280900646a70d", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.108491099905223, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0922556569421031, "全帯域_10ms_rms_envelope_nrmse": 0.07552482921502539, "全帯域_10ms_envelope_correlation": 0.994093117770037, "全帯域_10ms_modulation_mae_db": 0.163432228780397, "全帯域_40ms_log_rms_mae_db": 0.5990804517271816, "全帯域_40ms_rms_envelope_nrmse": 0.039648256467815716, "全帯域_40ms_envelope_correlation": 0.9982613182835842, "全帯域_40ms_modulation_mae_db": 0.5593375981866445, "300–1000 Hz_10ms_log_rms_mae_db": 1.5856540023190586, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09287982567067526, "300–1000 Hz_10ms_envelope_correlation": 0.9929716476193721, "300–1000 Hz_10ms_modulation_mae_db": 0.3519545445170044, "300–1000 Hz_40ms_log_rms_mae_db": 1.1043515282643483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047545007643450314, "300–1000 Hz_40ms_envelope_correlation": 0.9980808941403423, "300–1000 Hz_40ms_modulation_mae_db": 0.7668285065171353, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8235194963405579, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14840381758856086, "1000–3000 Hz_10ms_envelope_correlation": 0.9854729778171413, "1000–3000 Hz_10ms_modulation_mae_db": 1.0186534987692974, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3059437546460266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08969005596848806, "1000–3000 Hz_40ms_envelope_correlation": 0.9958037982127335, "1000–3000 Hz_40ms_modulation_mae_db": 1.0573892648389518, "3000–6000 Hz_10ms_log_rms_mae_db": 2.460372596894543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24659501410190807, "3000–6000 Hz_10ms_envelope_correlation": 0.9593766709999021, "3000–6000 Hz_10ms_modulation_mae_db": 0.9667432113784985, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0520101237191732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15507938451474615, "3000–6000 Hz_40ms_envelope_correlation": 0.9831690670871652, "3000–6000 Hz_40ms_modulation_mae_db": 0.9061202708942551, "mr_stft_db": 6.053483793032643, "band_mae_db": 0.24343159768014347, "rms_gain_error_db": -0.04050818356730812, "output_peak": 0.30556490588560353}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.10721951038491, "全帯域_10ms_rms_envelope_nrmse": 0.07088542519358257, "全帯域_10ms_envelope_correlation": 0.9948040277074486, "全帯域_10ms_modulation_mae_db": 0.4504587173825992, "全帯域_40ms_log_rms_mae_db": 0.6338779239423425, "全帯域_40ms_rms_envelope_nrmse": 0.03324218599202344, "全帯域_40ms_envelope_correlation": 0.9987903174353984, "全帯域_40ms_modulation_mae_db": 0.2373930170828433, "300–1000 Hz_10ms_log_rms_mae_db": 1.5115163660137763, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09781004721033758, "300–1000 Hz_10ms_envelope_correlation": 0.992158051290777, "300–1000 Hz_10ms_modulation_mae_db": 0.45972329520058824, "300–1000 Hz_40ms_log_rms_mae_db": 0.9692250080592343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04804660424145783, "300–1000 Hz_40ms_envelope_correlation": 0.9980124855837936, "300–1000 Hz_40ms_modulation_mae_db": 0.5347213920109296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8108181086035828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14666511426903994, "1000–3000 Hz_10ms_envelope_correlation": 0.9847435420470094, "1000–3000 Hz_10ms_modulation_mae_db": 0.7184471868400258, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2257497481394526, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08215082244561417, "1000–3000 Hz_40ms_envelope_correlation": 0.995318270904218, "1000–3000 Hz_40ms_modulation_mae_db": 0.7609593180920827, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4623708903381316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2463773389548064, "3000–6000 Hz_10ms_envelope_correlation": 0.9598744635099592, "3000–6000 Hz_10ms_modulation_mae_db": 0.3247730699085671, "3000–6000 Hz_40ms_log_rms_mae_db": 2.061145955315519, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1420819941526735, "3000–6000 Hz_40ms_envelope_correlation": 0.9861267593406452, "3000–6000 Hz_40ms_modulation_mae_db": 0.4262934465321068, "mr_stft_db": 6.037865545052225, "band_mae_db": 0.10663148128827939, "rms_gain_error_db": -0.03958878974583732, "output_peak": 0.34331657658193326}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1132962514557232, "全帯域_10ms_rms_envelope_nrmse": 0.07007897713674216, "全帯域_10ms_envelope_correlation": 0.9949267706694324, "全帯域_10ms_modulation_mae_db": 0.128191004141478, "全帯域_40ms_log_rms_mae_db": 0.589412139256356, "全帯域_40ms_rms_envelope_nrmse": 0.03499556941302672, "全帯域_40ms_envelope_correlation": 0.9986329328119994, "全帯域_40ms_modulation_mae_db": 0.19286936085312656, "300–1000 Hz_10ms_log_rms_mae_db": 1.627887884417442, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08524655839556342, "300–1000 Hz_10ms_envelope_correlation": 0.9940710561854997, "300–1000 Hz_10ms_modulation_mae_db": 0.15803299986562266, "300–1000 Hz_40ms_log_rms_mae_db": 1.1193702834830754, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04569719961550469, "300–1000 Hz_40ms_envelope_correlation": 0.998185369438, "300–1000 Hz_40ms_modulation_mae_db": 0.35372163592741085, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7159272154251515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13298381956843974, "1000–3000 Hz_10ms_envelope_correlation": 0.9879334323305919, "1000–3000 Hz_10ms_modulation_mae_db": 0.6015385700163252, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1239866209585878, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07491937602546646, "1000–3000 Hz_40ms_envelope_correlation": 0.9967041347791703, "1000–3000 Hz_40ms_modulation_mae_db": 0.5560574134141548, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4546713044744344, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24251554524324587, "3000–6000 Hz_10ms_envelope_correlation": 0.9603675059846174, "3000–6000 Hz_10ms_modulation_mae_db": 0.42756265246246644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.99099835420347, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13450363452384323, "3000–6000 Hz_40ms_envelope_correlation": 0.9875543136754403, "3000–6000 Hz_40ms_modulation_mae_db": 0.7593527267751137, "mr_stft_db": 6.028283267478398, "band_mae_db": 0.1385216082718716, "rms_gain_error_db": 0.0017107108761174594, "output_peak": 0.3207183397440306}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1672234351184523, "全帯域_10ms_rms_envelope_nrmse": 0.07201976106351278, "全帯域_10ms_envelope_correlation": 0.9946639061556938, "全帯域_10ms_modulation_mae_db": 0.27238721843167923, "全帯域_40ms_log_rms_mae_db": 0.6159981792393339, "全帯域_40ms_rms_envelope_nrmse": 0.03528054680709513, "全帯域_40ms_envelope_correlation": 0.9986819807336451, "全帯域_40ms_modulation_mae_db": 0.3596908107517089, "300–1000 Hz_10ms_log_rms_mae_db": 1.566793916253549, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08832738977909893, "300–1000 Hz_10ms_envelope_correlation": 0.9936836510025429, "300–1000 Hz_10ms_modulation_mae_db": 0.40074604118461865, "300–1000 Hz_40ms_log_rms_mae_db": 1.0264639911392837, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04685268659462796, "300–1000 Hz_40ms_envelope_correlation": 0.9982154057717036, "300–1000 Hz_40ms_modulation_mae_db": 0.559483782717756, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7802556009834327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13293915120605126, "1000–3000 Hz_10ms_envelope_correlation": 0.9881671417188428, "1000–3000 Hz_10ms_modulation_mae_db": 0.8657242321899655, "1000–3000 Hz_40ms_log_rms_mae_db": 1.276701352975241, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07597163683508738, "1000–3000 Hz_40ms_envelope_correlation": 0.9968854802491097, "1000–3000 Hz_40ms_modulation_mae_db": 1.0474827893965568, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4261778555565856, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22094989590706965, "3000–6000 Hz_10ms_envelope_correlation": 0.9679020522724912, "3000–6000 Hz_10ms_modulation_mae_db": 0.3235817149078878, "3000–6000 Hz_40ms_log_rms_mae_db": 1.974671811296664, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11708188901935494, "3000–6000 Hz_40ms_envelope_correlation": 0.9907598755797155, "3000–6000 Hz_40ms_modulation_mae_db": 0.428664098054587, "mr_stft_db": 6.035039256377165, "band_mae_db": 0.21320182499620358, "rms_gain_error_db": -0.07313861839430086, "output_peak": 0.3139019133842182}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1723117364503797, "全帯域_10ms_rms_envelope_nrmse": 0.06999837455012628, "全帯域_10ms_envelope_correlation": 0.9949402316456621, "全帯域_10ms_modulation_mae_db": 0.10135394356849887, "全帯域_40ms_log_rms_mae_db": 0.6325533256115643, "全帯域_40ms_rms_envelope_nrmse": 0.03324873065314988, "全帯域_40ms_envelope_correlation": 0.9988144165445341, "全帯域_40ms_modulation_mae_db": 0.42409428903634055, "300–1000 Hz_10ms_log_rms_mae_db": 1.588861869413143, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08806583585095375, "300–1000 Hz_10ms_envelope_correlation": 0.9936735791419581, "300–1000 Hz_10ms_modulation_mae_db": 0.1548335143625612, "300–1000 Hz_40ms_log_rms_mae_db": 1.0603024577051432, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044274262940812055, "300–1000 Hz_40ms_envelope_correlation": 0.9983247930482764, "300–1000 Hz_40ms_modulation_mae_db": 0.4917479595524686, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7905694745931728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13573439806386212, "1000–3000 Hz_10ms_envelope_correlation": 0.9876020819973165, "1000–3000 Hz_10ms_modulation_mae_db": 0.7230733837424773, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2481082202618596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07921869153206051, "1000–3000 Hz_40ms_envelope_correlation": 0.9964625577470119, "1000–3000 Hz_40ms_modulation_mae_db": 0.6133934833555745, "3000–6000 Hz_10ms_log_rms_mae_db": 2.418347333992719, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23527895085800118, "3000–6000 Hz_10ms_envelope_correlation": 0.9627657751850751, "3000–6000 Hz_10ms_modulation_mae_db": 0.6056311262117017, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9948195333791445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13477079948080337, "3000–6000 Hz_40ms_envelope_correlation": 0.9874528327108488, "3000–6000 Hz_40ms_modulation_mae_db": 0.8872393924915264, "mr_stft_db": 6.049915329386646, "band_mae_db": 0.148039961327436, "rms_gain_error_db": -0.06306443851167448, "output_peak": 0.3132137503951868}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1213022841618345, "全帯域_10ms_rms_envelope_nrmse": 0.07410880075419449, "全帯域_10ms_envelope_correlation": 0.9943737453727578, "全帯域_10ms_modulation_mae_db": 0.42230419577911665, "全帯域_40ms_log_rms_mae_db": 0.6236909302445263, "全帯域_40ms_rms_envelope_nrmse": 0.03811389720520798, "全帯域_40ms_envelope_correlation": 0.9985021224964368, "全帯域_40ms_modulation_mae_db": 0.586884978420996, "300–1000 Hz_10ms_log_rms_mae_db": 1.6302600973272887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1013892719267039, "300–1000 Hz_10ms_envelope_correlation": 0.9916687806741995, "300–1000 Hz_10ms_modulation_mae_db": 0.6042336358602332, "300–1000 Hz_40ms_log_rms_mae_db": 1.0913536917971483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05901298939888554, "300–1000 Hz_40ms_envelope_correlation": 0.9971284008866734, "300–1000 Hz_40ms_modulation_mae_db": 1.0442967900117281, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8149231028131838, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1284184310538868, "1000–3000 Hz_10ms_envelope_correlation": 0.9889150917169358, "1000–3000 Hz_10ms_modulation_mae_db": 1.0610244638017852, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2897804929913472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07182806534038419, "1000–3000 Hz_40ms_envelope_correlation": 0.9971274663544953, "1000–3000 Hz_40ms_modulation_mae_db": 0.8294696117808241, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4507497248725927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23573974160645614, "3000–6000 Hz_10ms_envelope_correlation": 0.9626335760041127, "3000–6000 Hz_10ms_modulation_mae_db": 0.33980724679931534, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0134964625214185, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14432290020415944, "3000–6000 Hz_40ms_envelope_correlation": 0.9853870889216961, "3000–6000 Hz_40ms_modulation_mae_db": 0.48476777293843254, "mr_stft_db": 6.0498999198132095, "band_mae_db": 0.1804213199239971, "rms_gain_error_db": -0.09654879351381229, "output_peak": 0.32273388817618215}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.150748425312144, "全帯域_10ms_rms_envelope_nrmse": 0.06934352733743224, "全帯域_10ms_envelope_correlation": 0.9950885533959415, "全帯域_10ms_modulation_mae_db": 0.40825616701246775, "全帯域_40ms_log_rms_mae_db": 0.6365377048625857, "全帯域_40ms_rms_envelope_nrmse": 0.03496384199063431, "全帯域_40ms_envelope_correlation": 0.9987536690629857, "全帯域_40ms_modulation_mae_db": 0.6318062068373962, "300–1000 Hz_10ms_log_rms_mae_db": 1.6712118045199156, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.090275030582912, "300–1000 Hz_10ms_envelope_correlation": 0.9935897127908652, "300–1000 Hz_10ms_modulation_mae_db": 1.0643099894912038, "300–1000 Hz_40ms_log_rms_mae_db": 1.103753132388791, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053071086532968485, "300–1000 Hz_40ms_envelope_correlation": 0.9978763615283255, "300–1000 Hz_40ms_modulation_mae_db": 1.431104994793569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8237788005666526, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14787031545274149, "1000–3000 Hz_10ms_envelope_correlation": 0.9852323133064569, "1000–3000 Hz_10ms_modulation_mae_db": 0.6100378396094791, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2576714542734548, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08641419305604008, "1000–3000 Hz_40ms_envelope_correlation": 0.9958489678657627, "1000–3000 Hz_40ms_modulation_mae_db": 0.7020755768814698, "3000–6000 Hz_10ms_log_rms_mae_db": 2.411508676929591, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24232496945079154, "3000–6000 Hz_10ms_envelope_correlation": 0.9605021340994749, "3000–6000 Hz_10ms_modulation_mae_db": 0.8030655263124337, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0276789953361, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1337548319164224, "3000–6000 Hz_40ms_envelope_correlation": 0.9875170194910784, "3000–6000 Hz_40ms_modulation_mae_db": 1.0191089237093305, "mr_stft_db": 6.058343416758753, "band_mae_db": 0.18948196968221712, "rms_gain_error_db": -0.09598150420355372, "output_peak": 0.3216898922016787}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1704747619369735, "全帯域_10ms_rms_envelope_nrmse": 0.08517322686568744, "全帯域_10ms_envelope_correlation": 0.9924727956590828, "全帯域_10ms_modulation_mae_db": 0.5651236117863901, "全帯域_40ms_log_rms_mae_db": 0.664058341659721, "全帯域_40ms_rms_envelope_nrmse": 0.049748120460023464, "全帯域_40ms_envelope_correlation": 0.9972310144100008, "全帯域_40ms_modulation_mae_db": 0.7998842457186764, "300–1000 Hz_10ms_log_rms_mae_db": 1.6003873314138348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09223050993157542, "300–1000 Hz_10ms_envelope_correlation": 0.9930322514798849, "300–1000 Hz_10ms_modulation_mae_db": 0.5231337988581349, "300–1000 Hz_40ms_log_rms_mae_db": 1.1329358205006352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05469115192064036, "300–1000 Hz_40ms_envelope_correlation": 0.9973741597139444, "300–1000 Hz_40ms_modulation_mae_db": 0.8956510564791315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7942421667666568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13784655718178465, "1000–3000 Hz_10ms_envelope_correlation": 0.986819685930619, "1000–3000 Hz_10ms_modulation_mae_db": 0.5468532421007284, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2427791384664482, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07651362961776535, "1000–3000 Hz_40ms_envelope_correlation": 0.9963540610648821, "1000–3000 Hz_40ms_modulation_mae_db": 0.5770017962662572, "3000–6000 Hz_10ms_log_rms_mae_db": 2.462052376548395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2223195848956856, "3000–6000 Hz_10ms_envelope_correlation": 0.9677605800324036, "3000–6000 Hz_10ms_modulation_mae_db": 0.0995051233870043, "3000–6000 Hz_40ms_log_rms_mae_db": 2.035226291214679, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13289527134415796, "3000–6000 Hz_40ms_envelope_correlation": 0.9882979441484491, "3000–6000 Hz_40ms_modulation_mae_db": 0.37612221909576754, "mr_stft_db": 6.05211759106274, "band_mae_db": 0.129987552989539, "rms_gain_error_db": -0.02473203541386082, "output_peak": 0.3186766881902493}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 71920, "output_samples": 71920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.012161926838132, "spectral_convergence": 0.2835806637345464, "spectral_convergence_status": "ok", "amplitude_floor": 7.580332405627278e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.9776885706094705, "spectral_convergence": 0.24052057204316588, "spectral_convergence_status": "ok", "amplitude_floor": 6.818562526348602e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.170600881650327, "spectral_convergence": 0.23581847028246722, "spectral_convergence_status": "ok", "amplitude_floor": 5.715517707749406e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020741335082478346, "output_power": 0.00021432886125379093, "delta_db": 0.14243948817823202, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006538184008046484, "output_power": 0.0006529775120589383, "delta_db": -0.005589141945814195, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004887436095925812, "output_power": 0.00047729603039098464, "delta_db": -0.10293269554875528, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 3.3579219734638266e-05, "output_power": 2.9620530683633537e-05, "delta_db": -0.5447776530480496, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.100095928785279e-05, "output_power": 5.561833955024553e-05, "delta_db": -0.40118645546028575, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.8128919162253983e-05, "output_power": 1.926363885065515e-05, "delta_db": 0.263664151899724, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24343159768014347, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04050818356730812, "pitch": null, "output_rms_dbfs": -28.389001711916045, "output_peak": 0.30556490588560353, "reference_rms_dbfs": -28.34849352834874, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.053483793032643}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 71920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 0.00010701965100918563, "log_rms_mae_db": 1.0922556569421031, "rms_envelope_nrmse": 0.07552482921502539, "envelope_correlation": 0.994093117770037, "rms_gain_error_db": -0.04050834901466299, "reference_mean_power": 0.0014643132072841782, "output_mean_power": 0.001450718484122001}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 5.509210445243936e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038836857080924806, "output_power": 0.0003836434062029236, "reference_active": true, "delta_db": -0.053163386097803036}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3200931448411458e-05, "output_power": 2.1435234686788826e-05, "reference_active": true, "delta_db": -0.34377178063942304}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.17358638897755e-06, "output_power": 8.978483795568637e-06, "reference_active": true, "delta_db": -0.09336151960396497}], "active_band_count": 3, "active_band_mae_db": 0.163432228780397}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 0.00010161161302678278, "log_rms_mae_db": 0.5990804517271816, "rms_envelope_nrmse": 0.039648256467815716, "envelope_correlation": 0.9982613182835842, "rms_gain_error_db": -0.040508296822372315, "reference_mean_power": 0.0014741623423244587, "output_mean_power": 0.0014604761970795162}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 4.939959887293225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035657053350397007, "output_power": 0.00035304593170746164, "reference_active": true, "delta_db": -0.04314239519639835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.1722039793266e-06, "output_power": 7.589675532154878e-06, "reference_active": true, "delta_db": -0.3211598846935625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.984441222185337e-07, "output_power": 5.161317972603643e-07, "reference_active": true, "delta_db": -1.3137105146699728}], "active_band_count": 3, "active_band_mae_db": 0.5593375981866445}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 9.15514590150253e-05, "log_rms_mae_db": 1.5856540023190586, "rms_envelope_nrmse": 0.09287982567067526, "envelope_correlation": 0.9929716476193721, "rms_gain_error_db": -0.027302194137690913, "reference_mean_power": 0.0006933001790048939, "output_mean_power": 0.0006889553753317095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 3.8791214865659925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002719560122465178, "output_power": 0.00025875495236237796, "reference_active": true, "delta_db": -0.21609993796221835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.187110727790427e-05, "output_power": 1.862208579397192e-05, "reference_active": true, "delta_db": -0.6984244764800117}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.698791354460839e-06, "output_power": 6.920386863276802e-06, "reference_active": true, "delta_db": 0.14133921910878317}], "active_band_count": 3, "active_band_mae_db": 0.3519545445170044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 8.797925167476037e-05, "log_rms_mae_db": 1.1043515282643483, "rms_envelope_nrmse": 0.047545007643450314, "envelope_correlation": 0.9980808941403423, "rms_gain_error_db": -0.02730224607276611, "reference_mean_power": 0.000697963583829786, "output_mean_power": 0.0006935895470344183}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.5511880459540643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002583969788516632, "output_power": 0.0002456780227440391, "reference_active": true, "delta_db": -0.21921123452891986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.105856462043575e-06, "output_power": 7.144171390351945e-06, "reference_active": true, "delta_db": -0.5484704418430348}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388944870715119e-07, "output_power": 4.488991224115942e-07, "reference_active": true, "delta_db": -1.5328038431794513}], "active_band_count": 3, "active_band_mae_db": 0.7668285065171353}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 4.592612274041097e-05, "log_rms_mae_db": 1.8235194963405579, "rms_envelope_nrmse": 0.14840381758856086, "envelope_correlation": 0.9854729778171413, "rms_gain_error_db": -0.45845996974075676, "reference_mean_power": 8.288074215773485e-05, "output_mean_power": 7.4577467986376e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 4.287875385442721e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2411523150580774e-05, "output_power": 1.8345292712431914e-05, "reference_active": true, "delta_db": -0.8694672779072312}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.113877152498063e-06, "output_power": 2.653757775073381e-06, "reference_active": true, "delta_db": -0.6944019536141905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.881539635700238e-07, "output_power": 7.008345286985243e-07, "reference_active": true, "delta_db": -1.4920912647864704}], "active_band_count": 3, "active_band_mae_db": 1.0186534987692974}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 4.2997336393594785e-05, "log_rms_mae_db": 1.3059437546460266, "rms_envelope_nrmse": 0.08969005596848806, "envelope_correlation": 0.9958037982127335, "rms_gain_error_db": -0.4584611738207456, "reference_mean_power": 8.343814618382176e-05, "output_mean_power": 7.507900856882848e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.9683446107037247e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.063395086340567e-05, "output_power": 1.7039061173808458e-05, "reference_active": true, "delta_db": -0.8313672987244464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4408129681959194e-06, "output_power": 1.232099009603617e-06, "reference_active": true, "delta_db": -0.6796200026927495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9896369199765786e-08, "output_power": 3.4036970263677227e-08, "reference_active": true, "delta_db": -1.6611804930996596}], "active_band_count": 3, "active_band_mae_db": 1.0573892648389518}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 2.431018241105218e-05, "log_rms_mae_db": 2.460372596894543, "rms_envelope_nrmse": 0.24659501410190807, "envelope_correlation": 0.9593766709999021, "rms_gain_error_db": 0.05836111391078121, "reference_mean_power": 1.5917473583289094e-05, "output_mean_power": 1.613281854852401e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 8.500689890351627e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.331755201814068e-06, "output_power": 3.966575176890019e-06, "reference_active": true, "delta_db": -0.38248215885555503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3672340908387822e-06, "output_power": 1.101210681435565e-06, "reference_active": true, "delta_db": -0.9397246334072672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.790743563061608e-07, "output_power": 4.0134902833451936e-07, "reference_active": true, "delta_db": 1.5780228418726732}], "active_band_count": 3, "active_band_mae_db": 0.9667432113784985}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 1.765504974955162e-05, "log_rms_mae_db": 2.0520101237191732, "rms_envelope_nrmse": 0.15507938451474615, "envelope_correlation": 0.9831690670871652, "rms_gain_error_db": 0.05836308416617763, "reference_mean_power": 1.6024474415231737e-05, "output_mean_power": 1.6241274345827942e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 7.931595726775402e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.29977249297526e-06, "output_power": 3.924417574184261e-06, "reference_active": true, "delta_db": -0.3967026516697284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.410418312822663e-07, "output_power": 6.174110347755842e-07, "reference_active": true, "delta_db": -0.7926833707300734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9106902570925117e-08, "output_power": 2.7169882991101314e-08, "reference_active": true, "delta_db": 1.5289747902829631}], "active_band_count": 3, "active_band_mae_db": 0.9061202708942551}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8657395656137046, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6ddee88b50f2cd100f5e574dea511c3b04dfb7a4e84228fa0ac6de3071f60724", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9868607833581142, "standard_deviation": 0.0015440749975820414, "minimum": 0.9847435420470094, "maximum": 0.9889150917169358, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.794254245761549, "standard_deviation": 0.03540456155713475, "minimum": 1.7159272154251515, "maximum": 1.8237788005666526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7681690521337605, "standard_deviation": 0.19430518418823528, "minimum": 0.5468532421007284, "maximum": 1.0610244638017852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13885770054804586, "standard_deviation": 0.007769893986636999, "minimum": 0.1284184310538868, "maximum": 0.14840381758856086, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996313092147173, "standard_deviation": 0.0006132148554101138, "minimum": 0.995318270904218, "maximum": 0.9971274663544953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2463400978390524, "standard_deviation": 0.05593364918989138, "minimum": 1.1239866209585878, "maximum": 1.3059437546460266, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.767978656753234, "standard_deviation": 0.19832658322412777, "minimum": 0.5560574134141548, "maximum": 1.0573892648389518, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07958830885261328, "standard_deviation": 0.006093267535599148, "minimum": 0.07182806534038419, "maximum": 0.08969005596848806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9626478447610045, "standard_deviation": 0.0034217186623221717, "minimum": 0.9593766709999021, "maximum": 0.9679020522724912, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.443281344950874, "standard_deviation": 0.021105577546655877, "minimum": 2.411508676929591, "maximum": 2.4623708903381316, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4863337089209844, "standard_deviation": 0.2859806825070794, "minimum": 0.0995051233870043, "maximum": 0.9667432113784985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23651263012724555, "standard_deviation": 0.010103364190243798, "minimum": 0.22094989590706965, "maximum": 0.24659501410190807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9870331126193799, "standard_deviation": 0.0022242899499516876, "minimum": 0.9831690670871652, "maximum": 0.9907598755797155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.018755940873271, "standard_deviation": 0.030631811562013885, "minimum": 1.974671811296664, "maximum": 2.061145955315519, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6609586063113899, "standard_deviation": 0.25924719086413833, "minimum": 0.37612221909576754, "maximum": 1.0191089237093305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13681133814452012, "standard_deviation": 0.010963205760605483, "minimum": 0.11708188901935494, "maximum": 0.15507938451474615, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931060912731373, "standard_deviation": 0.000828840304613937, "minimum": 0.9916687806741995, "maximum": 0.9940710561854997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.597821658959751, "standard_deviation": 0.04779208010039668, "minimum": 1.5115163660137763, "maximum": 1.6712118045199156, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4646209774174959, "standard_deviation": 0.29014886228330294, "minimum": 0.1548335143625612, "maximum": 1.0643099894912038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09202805866847753, "standard_deviation": 0.005349048139742693, "minimum": 0.08524655839556342, "maximum": 0.1013892719267039, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978997337638824, "standard_deviation": 0.0004274797832845487, "minimum": 0.9971284008866734, "maximum": 0.9983247930482764, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0759694891672074, "standard_deviation": 0.05481601821966385, "minimum": 0.9692250080592343, "maximum": 1.1329358205006352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7596945147512661, "standard_deviation": 0.35408476403086375, "minimum": 0.35372163592741085, "maximum": 1.431104994793569, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04989887361104341, "standard_deviation": 0.005121901495666815, "minimum": 0.044274262940812055, "maximum": 0.05901298939888554, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16871466451996092, "standard_deviation": 0.046017433184020975, "minimum": 0.10663148128827939, "maximum": 0.24343159768014347, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.045618514870222, "standard_deviation": 0.010524507566851358, "minimum": 6.028283267478398, "maximum": 6.058343416758753, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31997699431988536, "standard_deviation": 0.011005630649297849, "minimum": 0.30556490588560353, "maximum": 0.34331657658193326, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05398145655927877, "standard_deviation": 0.03458179471992261, "minimum": -0.09654879351381229, "maximum": 0.0017107108761174594, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944203935470071, "standard_deviation": 0.0008520549614820817, "minimum": 0.9924727956590828, "maximum": 0.9950885533959415, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.136854007720315, "standard_deviation": 0.031997774279334174, "minimum": 1.0922556569421031, "maximum": 1.1723117364503797, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31393838586032835, "standard_deviation": 0.1717322039010091, "minimum": 0.10135394356849887, "maximum": 0.5651236117863901, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07339161526453791, "standard_deviation": 0.005222328040023921, "minimum": 0.06934352733743224, "maximum": 0.08517322686568744, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998458471472323, "standard_deviation": 0.0005280393702438769, "minimum": 0.9972310144100008, "maximum": 0.9988144165445341, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6244011245679514, "standard_deviation": 0.023338316102088072, "minimum": 0.589412139256356, "maximum": 0.664058341659721, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4739950633609666, "standard_deviation": 0.20760654259994782, "minimum": 0.19286936085312656, "maximum": 0.7998842457186764, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037405143623622086, "standard_deviation": 0.005461146690827129, "minimum": 0.03324218599202344, "maximum": 0.049748120460023464, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-23283-0016", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.494011402130127, "predicted_mos": 2.494011402130127, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "8f3cea11d7afc1582f5e3ffad005b322dff3e7ed02e15907f0e1f1c473e43b42", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9651454999111593, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1314421689164778, "全帯域_10ms_rms_envelope_nrmse": 0.07551978998612752, "全帯域_10ms_envelope_correlation": 0.9941063368328028, "全帯域_10ms_modulation_mae_db": 0.6146399450383749, "全帯域_40ms_log_rms_mae_db": 0.6192207977427003, "全帯域_40ms_rms_envelope_nrmse": 0.03747630588017523, "全帯域_40ms_envelope_correlation": 0.9984509991538073, "全帯域_40ms_modulation_mae_db": 0.6933450538792919, "300–1000 Hz_10ms_log_rms_mae_db": 1.1756163298835307, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08984377229504348, "300–1000 Hz_10ms_envelope_correlation": 0.9933827045908403, "300–1000 Hz_10ms_modulation_mae_db": 0.6855770266857396, "300–1000 Hz_40ms_log_rms_mae_db": 0.7165646561412707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05274224345577519, "300–1000 Hz_40ms_envelope_correlation": 0.9975804004759444, "300–1000 Hz_40ms_modulation_mae_db": 1.079210350802961, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2581107807043816, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1331915347782187, "1000–3000 Hz_10ms_envelope_correlation": 0.9876784093975545, "1000–3000 Hz_10ms_modulation_mae_db": 0.8121625411893394, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8156786284561113, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08065713703136261, "1000–3000 Hz_40ms_envelope_correlation": 0.9957737595833374, "1000–3000 Hz_40ms_modulation_mae_db": 0.6057583501195211, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5050331997172923, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18484027953974433, "3000–6000 Hz_10ms_envelope_correlation": 0.9770312741835073, "3000–6000 Hz_10ms_modulation_mae_db": 0.19015489107800723, "3000–6000 Hz_40ms_log_rms_mae_db": 1.285983378442175, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11229817582911662, "3000–6000 Hz_40ms_envelope_correlation": 0.9915938985125611, "3000–6000 Hz_40ms_modulation_mae_db": 0.507048185977231, "mr_stft_db": 5.552945903816354, "band_mae_db": 0.2324861887946373, "rms_gain_error_db": -0.023922572171655748, "output_peak": 0.3259018841732439}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2017229717098792, "全帯域_10ms_rms_envelope_nrmse": 0.07155590381162048, "全帯域_10ms_envelope_correlation": 0.99472818591086, "全帯域_10ms_modulation_mae_db": 0.12023456072507686, "全帯域_40ms_log_rms_mae_db": 0.6593899535306476, "全帯域_40ms_rms_envelope_nrmse": 0.039244228489901944, "全帯域_40ms_envelope_correlation": 0.9982900680563481, "全帯域_40ms_modulation_mae_db": 0.10769892262983526, "300–1000 Hz_10ms_log_rms_mae_db": 1.0619348611853885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08886630177640907, "300–1000 Hz_10ms_envelope_correlation": 0.9935279239798059, "300–1000 Hz_10ms_modulation_mae_db": 0.3594500694851724, "300–1000 Hz_40ms_log_rms_mae_db": 0.6239423146443744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05642749099139365, "300–1000 Hz_40ms_envelope_correlation": 0.9972063626440418, "300–1000 Hz_40ms_modulation_mae_db": 0.3931189335115468, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2570424986565185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11495383099487064, "1000–3000 Hz_10ms_envelope_correlation": 0.9905735094134486, "1000–3000 Hz_10ms_modulation_mae_db": 0.12216476969198659, "1000–3000 Hz_40ms_log_rms_mae_db": 0.766693396167871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05934724335847336, "1000–3000 Hz_40ms_envelope_correlation": 0.9973896893449218, "1000–3000 Hz_40ms_modulation_mae_db": 0.2734675435909622, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3991766242454393, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16599924107705968, "3000–6000 Hz_10ms_envelope_correlation": 0.9817604202892442, "3000–6000 Hz_10ms_modulation_mae_db": 0.22119400561285243, "3000–6000 Hz_40ms_log_rms_mae_db": 1.196169273839135, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10515612316672746, "3000–6000 Hz_40ms_envelope_correlation": 0.9922858535174306, "3000–6000 Hz_40ms_modulation_mae_db": 0.6708390936742331, "mr_stft_db": 5.563556943009799, "band_mae_db": 0.09846242406562937, "rms_gain_error_db": -0.0058563686343364755, "output_peak": 0.35134429874539763}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.194882806766256, "全帯域_10ms_rms_envelope_nrmse": 0.07399666633712304, "全帯域_10ms_envelope_correlation": 0.9943800145685514, "全帯域_10ms_modulation_mae_db": 0.2901083209711733, "全帯域_40ms_log_rms_mae_db": 0.650808104711699, "全帯域_40ms_rms_envelope_nrmse": 0.04254618842711308, "全帯域_40ms_envelope_correlation": 0.9980244665243964, "全帯域_40ms_modulation_mae_db": 0.41318585987744266, "300–1000 Hz_10ms_log_rms_mae_db": 1.1322606226759264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09062655675015836, "300–1000 Hz_10ms_envelope_correlation": 0.993266994937063, "300–1000 Hz_10ms_modulation_mae_db": 0.5895144437025576, "300–1000 Hz_40ms_log_rms_mae_db": 0.7411372323660009, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055979195407492514, "300–1000 Hz_40ms_envelope_correlation": 0.9972657698859465, "300–1000 Hz_40ms_modulation_mae_db": 0.8725250744430305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.257428761913375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11651394715036034, "1000–3000 Hz_10ms_envelope_correlation": 0.9911511265813951, "1000–3000 Hz_10ms_modulation_mae_db": 0.7114324788038777, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7574569854400762, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06635483500383643, "1000–3000 Hz_40ms_envelope_correlation": 0.9978846249511277, "1000–3000 Hz_40ms_modulation_mae_db": 0.5152392398218971, "3000–6000 Hz_10ms_log_rms_mae_db": 1.403825073353109, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1897078398138373, "3000–6000 Hz_10ms_envelope_correlation": 0.9758075888829021, "3000–6000 Hz_10ms_modulation_mae_db": 0.6549571328145763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2192213689196931, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12497576663653209, "3000–6000 Hz_40ms_envelope_correlation": 0.9894832783029389, "3000–6000 Hz_40ms_modulation_mae_db": 0.6229025973922494, "mr_stft_db": 5.545611689678864, "band_mae_db": 0.24832389808839894, "rms_gain_error_db": -0.04343350170449756, "output_peak": 0.314744564104991}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1758465670232667, "全帯域_10ms_rms_envelope_nrmse": 0.07035095391695322, "全帯域_10ms_envelope_correlation": 0.9949228516920368, "全帯域_10ms_modulation_mae_db": 0.14660835071964398, "全帯域_40ms_log_rms_mae_db": 0.6533020624733534, "全帯域_40ms_rms_envelope_nrmse": 0.03839020267924741, "全帯域_40ms_envelope_correlation": 0.9984351994191888, "全帯域_40ms_modulation_mae_db": 0.35462657398020464, "300–1000 Hz_10ms_log_rms_mae_db": 1.1822166313750657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07943478194806108, "300–1000 Hz_10ms_envelope_correlation": 0.9949564004871717, "300–1000 Hz_10ms_modulation_mae_db": 0.1131891106512729, "300–1000 Hz_40ms_log_rms_mae_db": 0.7435970806283467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045182373769549505, "300–1000 Hz_40ms_envelope_correlation": 0.9983996337569876, "300–1000 Hz_40ms_modulation_mae_db": 0.4651326293045271, "1000–3000 Hz_10ms_log_rms_mae_db": 1.296670828191912, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14770228370336466, "1000–3000 Hz_10ms_envelope_correlation": 0.9853019994381459, "1000–3000 Hz_10ms_modulation_mae_db": 0.8195396547686684, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8335001624831934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09233294193247375, "1000–3000 Hz_40ms_envelope_correlation": 0.9950390144645487, "1000–3000 Hz_40ms_modulation_mae_db": 0.9728483829366508, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3828619612244157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17531474230613942, "3000–6000 Hz_10ms_envelope_correlation": 0.979839445203952, "3000–6000 Hz_10ms_modulation_mae_db": 0.48483737524653653, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1747374717381331, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1079154906194503, "3000–6000 Hz_40ms_envelope_correlation": 0.9929936989785095, "3000–6000 Hz_40ms_modulation_mae_db": 0.6653412497137, "mr_stft_db": 5.542273542576107, "band_mae_db": 0.25584896993626255, "rms_gain_error_db": -0.07984514945817422, "output_peak": 0.3198821739415237}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2099423858876452, "全帯域_10ms_rms_envelope_nrmse": 0.06718946976459983, "全帯域_10ms_envelope_correlation": 0.9953882866620125, "全帯域_10ms_modulation_mae_db": 0.6659652675371582, "全帯域_40ms_log_rms_mae_db": 0.660381374801302, "全帯域_40ms_rms_envelope_nrmse": 0.033355947920078365, "全帯域_40ms_envelope_correlation": 0.9988660528066735, "全帯域_40ms_modulation_mae_db": 0.750421275212731, "300–1000 Hz_10ms_log_rms_mae_db": 1.091987705446357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08090125800839096, "300–1000 Hz_10ms_envelope_correlation": 0.9947076765432228, "300–1000 Hz_10ms_modulation_mae_db": 0.5771528152916612, "300–1000 Hz_40ms_log_rms_mae_db": 0.6591206671878017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04662210550854738, "300–1000 Hz_40ms_envelope_correlation": 0.9982110689152657, "300–1000 Hz_40ms_modulation_mae_db": 0.9714534383194144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.206207816768565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10945846981927915, "1000–3000 Hz_10ms_envelope_correlation": 0.991969737816989, "1000–3000 Hz_10ms_modulation_mae_db": 0.7803094822365813, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7460764558987257, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06451838232257948, "1000–3000 Hz_40ms_envelope_correlation": 0.9976302320594538, "1000–3000 Hz_40ms_modulation_mae_db": 0.5721705858647268, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4145754754846689, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1792280165604565, "3000–6000 Hz_10ms_envelope_correlation": 0.9785480596559226, "3000–6000 Hz_10ms_modulation_mae_db": 0.5861356508301899, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1954731316981102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10383174938406557, "3000–6000 Hz_40ms_envelope_correlation": 0.9923663630864751, "3000–6000 Hz_40ms_modulation_mae_db": 0.8296284124571155, "mr_stft_db": 5.557705209781251, "band_mae_db": 0.15708754983818568, "rms_gain_error_db": -0.09158305864682123, "output_peak": 0.3168044326540288}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1693306154037286, "全帯域_10ms_rms_envelope_nrmse": 0.06782074086224622, "全帯域_10ms_envelope_correlation": 0.9952544219589778, "全帯域_10ms_modulation_mae_db": 0.22037528310623702, "全帯域_40ms_log_rms_mae_db": 0.6044975016593447, "全帯域_40ms_rms_envelope_nrmse": 0.033848079660100325, "全帯域_40ms_envelope_correlation": 0.9987666165274941, "全帯域_40ms_modulation_mae_db": 0.4726714953499355, "300–1000 Hz_10ms_log_rms_mae_db": 1.158141281578346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08418048927101389, "300–1000 Hz_10ms_envelope_correlation": 0.9942003961646975, "300–1000 Hz_10ms_modulation_mae_db": 0.42988351828573396, "300–1000 Hz_40ms_log_rms_mae_db": 0.6922854734076821, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047460532226257575, "300–1000 Hz_40ms_envelope_correlation": 0.998057014870747, "300–1000 Hz_40ms_modulation_mae_db": 0.9709724085129577, "1000–3000 Hz_10ms_log_rms_mae_db": 1.250053941321108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12586904677015212, "1000–3000 Hz_10ms_envelope_correlation": 0.9894824359928097, "1000–3000 Hz_10ms_modulation_mae_db": 0.8358903449170644, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7532852261642905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07512321229854162, "1000–3000 Hz_40ms_envelope_correlation": 0.9969340595495927, "1000–3000 Hz_40ms_modulation_mae_db": 0.6406189235259775, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3868553028800914, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18757041077487513, "3000–6000 Hz_10ms_envelope_correlation": 0.9761775527403338, "3000–6000 Hz_10ms_modulation_mae_db": 0.4501693228301779, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2280013618848526, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12257513317797099, "3000–6000 Hz_40ms_envelope_correlation": 0.9892826256518081, "3000–6000 Hz_40ms_modulation_mae_db": 0.6562852845498844, "mr_stft_db": 5.5500415230392806, "band_mae_db": 0.1944834613975803, "rms_gain_error_db": -0.062372303484472744, "output_peak": 0.311539682800537}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.144810233783925, "全帯域_10ms_rms_envelope_nrmse": 0.07368487221540414, "全帯域_10ms_envelope_correlation": 0.994503332836587, "全帯域_10ms_modulation_mae_db": 0.31014749936389246, "全帯域_40ms_log_rms_mae_db": 0.6382087515611762, "全帯域_40ms_rms_envelope_nrmse": 0.04057460653931289, "全帯域_40ms_envelope_correlation": 0.9983536200307028, "全帯域_40ms_modulation_mae_db": 0.6107437797048773, "300–1000 Hz_10ms_log_rms_mae_db": 1.0824188479984465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08232217023686797, "300–1000 Hz_10ms_envelope_correlation": 0.9946486706185884, "300–1000 Hz_10ms_modulation_mae_db": 0.33919511375692196, "300–1000 Hz_40ms_log_rms_mae_db": 0.6574556548660468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046654281517443094, "300–1000 Hz_40ms_envelope_correlation": 0.9983790828014599, "300–1000 Hz_40ms_modulation_mae_db": 0.7213342835145305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.251378370009486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.131487506935809, "1000–3000 Hz_10ms_envelope_correlation": 0.9888534475966959, "1000–3000 Hz_10ms_modulation_mae_db": 0.2667232902833593, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7816659199115684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07995795110196366, "1000–3000 Hz_40ms_envelope_correlation": 0.996908430931826, "1000–3000 Hz_40ms_modulation_mae_db": 0.16911361487731288, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4557526635410056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19337419243159384, "3000–6000 Hz_10ms_envelope_correlation": 0.9747846752007245, "3000–6000 Hz_10ms_modulation_mae_db": 0.3769572522464422, "3000–6000 Hz_40ms_log_rms_mae_db": 1.244661634845439, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12075009891740648, "3000–6000 Hz_40ms_envelope_correlation": 0.9896034094584429, "3000–6000 Hz_40ms_modulation_mae_db": 0.5729841508353521, "mr_stft_db": 5.551591954665903, "band_mae_db": 0.20719666579400572, "rms_gain_error_db": -0.12234201722339709, "output_peak": 0.29746369059460226}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2725180674739323, "全帯域_10ms_rms_envelope_nrmse": 0.07247213025688089, "全帯域_10ms_envelope_correlation": 0.9945541605511458, "全帯域_10ms_modulation_mae_db": 0.4331716263207315, "全帯域_40ms_log_rms_mae_db": 0.6699009512621679, "全帯域_40ms_rms_envelope_nrmse": 0.0364143337011714, "全帯域_40ms_envelope_correlation": 0.9985178726515117, "全帯域_40ms_modulation_mae_db": 0.7088424031710844, "300–1000 Hz_10ms_log_rms_mae_db": 1.2329170594379688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.099598723819477, "300–1000 Hz_10ms_envelope_correlation": 0.9919030622771374, "300–1000 Hz_10ms_modulation_mae_db": 0.3726212978619448, "300–1000 Hz_40ms_log_rms_mae_db": 0.7851934505936905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05721579218264054, "300–1000 Hz_40ms_envelope_correlation": 0.9971517309955925, "300–1000 Hz_40ms_modulation_mae_db": 0.527821911295057, "1000–3000 Hz_10ms_log_rms_mae_db": 1.26459319266879, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10907594456208297, "1000–3000 Hz_10ms_envelope_correlation": 0.9916581638169573, "1000–3000 Hz_10ms_modulation_mae_db": 0.7656221121404907, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7775040020440914, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06190110293341896, "1000–3000 Hz_40ms_envelope_correlation": 0.9974217736818116, "1000–3000 Hz_40ms_modulation_mae_db": 0.9371002794530382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4252601619544991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18393759929861495, "3000–6000 Hz_10ms_envelope_correlation": 0.9773477624991241, "3000–6000 Hz_10ms_modulation_mae_db": 0.2932330025701387, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2520281939078848, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11898607355372767, "3000–6000 Hz_40ms_envelope_correlation": 0.9899416697194727, "3000–6000 Hz_40ms_modulation_mae_db": 0.6993801613532469, "mr_stft_db": 5.560117761998586, "band_mae_db": 0.11328754909616466, "rms_gain_error_db": -0.004906014510293626, "output_peak": 0.32405391135553674}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 71920, "output_samples": 71920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.463791587889921, "spectral_convergence": 0.27335958242691016, "spectral_convergence_status": "ok", "amplitude_floor": 7.580332405627278e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.388947062668453, "spectral_convergence": 0.22836596736070583, "spectral_convergence_status": "ok", "amplitude_floor": 6.818562526348602e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.806099060890685, "spectral_convergence": 0.2297658767999113, "spectral_convergence_status": "ok", "amplitude_floor": 5.715517707749406e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020741335082478346, "output_power": 0.0002196677594770833, "delta_db": 0.24929612713869131, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006538184008046484, "output_power": 0.0006437815727187977, "delta_db": -0.0671859752742844, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004887436095925812, "output_power": 0.00048608756568807967, "delta_db": -0.02366580173159808, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 3.3579219734638266e-05, "output_power": 3.182057765962413e-05, "delta_db": -0.2336254099990393, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.100095928785279e-05, "output_power": 5.7323797459969756e-05, "delta_db": -0.27001711828044095, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.8128919162253983e-05, "output_power": 1.5968320560541585e-05, "delta_db": -0.5511267003437699, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2324861887946373, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.023922572171655748, "pitch": null, "output_rms_dbfs": -28.372416100520393, "output_peak": 0.3259018841732439, "reference_rms_dbfs": -28.34849352834874, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.552945903816354}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 71920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 0.00010701965100918563, "log_rms_mae_db": 1.1314421689164778, "rms_envelope_nrmse": 0.07551978998612752, "envelope_correlation": 0.9941063368328028, "rms_gain_error_db": -0.023922560366475693, "reference_mean_power": 0.0014643132072841782, "output_mean_power": 0.0014562693983091995}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 5.509210445243936e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038836857080924806, "output_power": 0.0003907030622896429, "reference_active": true, "delta_db": 0.026027383347415743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3200931448411458e-05, "output_power": 2.1279842098109645e-05, "reference_active": true, "delta_db": -0.3753701981457163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.17358638897755e-06, "output_power": 6.580924824350659e-06, "reference_active": true, "delta_db": -1.4425222536219928}], "active_band_count": 3, "active_band_mae_db": 0.6146399450383749}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 0.00010161161302678278, "log_rms_mae_db": 0.6192207977427003, "rms_envelope_nrmse": 0.03747630588017523, "envelope_correlation": 0.9984509991538073, "rms_gain_error_db": -0.02392233385460448, "reference_mean_power": 0.0014741623423244587, "output_mean_power": 0.0014660645062516604}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 4.939959887293225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035657053350397007, "output_power": 0.00035818424433436144, "reference_active": true, "delta_db": 0.019610275342675454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.1722039793266e-06, "output_power": 7.459878983463295e-06, "reference_active": true, "delta_db": -0.396074158790538}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.984441222185337e-07, "output_power": 4.7609825986061466e-07, "reference_active": true, "delta_db": -1.664350727504662}], "active_band_count": 3, "active_band_mae_db": 0.6933450538792919}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 9.15514590150253e-05, "log_rms_mae_db": 1.1756163298835307, "rms_envelope_nrmse": 0.08984377229504348, "envelope_correlation": 0.9933827045908403, "rms_gain_error_db": -0.04672926043082038, "reference_mean_power": 0.0006933001790048939, "output_mean_power": 0.0006858803902540648}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 3.8791214865659925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002719560122465178, "output_power": 0.00026961680758463645, "reference_active": true, "delta_db": -0.037517022907911986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.187110727790427e-05, "output_power": 1.8925900980785455e-05, "reference_active": true, "delta_db": -0.6281420724920396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.698791354460839e-06, "output_power": 4.8628318516989875e-06, "reference_active": true, "delta_db": -1.3910719846572672}], "active_band_count": 3, "active_band_mae_db": 0.6855770266857396}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 8.797925167476037e-05, "log_rms_mae_db": 0.7165646561412707, "rms_envelope_nrmse": 0.05274224345577519, "envelope_correlation": 0.9975804004759444, "rms_gain_error_db": -0.046729238972262235, "reference_mean_power": 0.000697963583829786, "output_mean_power": 0.0006904938901253056}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.5511880459540643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002583969788516632, "output_power": 0.00025569266894192666, "reference_active": true, "delta_db": -0.045691552417928556}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.105856462043575e-06, "output_power": 6.958212538108152e-06, "reference_active": true, "delta_db": -0.6630121890823741}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388944870715119e-07, "output_power": 3.5689166846558127e-07, "reference_active": true, "delta_db": -2.52892731090858}], "active_band_count": 3, "active_band_mae_db": 1.079210350802961}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 4.592612274041097e-05, "log_rms_mae_db": 1.2581107807043816, "rms_envelope_nrmse": 0.1331915347782187, "envelope_correlation": 0.9876784093975545, "rms_gain_error_db": -0.28900216692615377, "reference_mean_power": 8.288074215773485e-05, "output_mean_power": 7.754493113424112e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 4.287875385442721e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2411523150580774e-05, "output_power": 2.0602752082992594e-05, "reference_active": true, "delta_db": -0.3654613680781661}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.113877152498063e-06, "output_power": 2.9059033418976382e-06, "reference_active": true, "delta_db": -0.30020310540222706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.881539635700238e-07, "output_power": 6.572677175942456e-07, "reference_active": true, "delta_db": -1.770823150087625}], "active_band_count": 3, "active_band_mae_db": 0.8121625411893394}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 4.2997336393594785e-05, "log_rms_mae_db": 0.8156786284561113, "rms_envelope_nrmse": 0.08065713703136261, "envelope_correlation": 0.9957737595833374, "rms_gain_error_db": -0.2890019656277531, "reference_mean_power": 8.343814618382176e-05, "output_mean_power": 7.80664534516083e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.9683446107037247e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.063395086340567e-05, "output_power": 1.887975079476685e-05, "reference_active": true, "delta_db": -0.3858613451635798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4408129681959194e-06, "output_power": 1.3390325297881046e-06, "reference_active": true, "delta_db": -0.31816481070821295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9896369199765786e-08, "output_power": 3.861393445839452e-08, "reference_active": true, "delta_db": -1.1132488944867702}], "active_band_count": 3, "active_band_mae_db": 0.6057583501195211}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 2.431018241105218e-05, "log_rms_mae_db": 1.5050331997172923, "rms_envelope_nrmse": 0.18484027953974433, "envelope_correlation": 0.9770312741835073, "rms_gain_error_db": -0.3003673961470123, "reference_mean_power": 1.5917473583289094e-05, "output_mean_power": 1.4853794053389254e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 8.500689890351627e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.331755201814068e-06, "output_power": 3.999559353687879e-06, "reference_active": true, "delta_db": -0.34651759465760734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3672340908387822e-06, "output_power": 1.4296754826807416e-06, "reference_active": true, "delta_db": 0.19394590962915187}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.790743563061608e-07, "output_power": 2.771531473518239e-07, "reference_active": true, "delta_db": -0.030001168947262603}], "active_band_count": 3, "active_band_mae_db": 0.19015489107800723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 1.765504974955162e-05, "log_rms_mae_db": 1.285983378442175, "rms_envelope_nrmse": 0.11229817582911662, "envelope_correlation": 0.9915938985125611, "rms_gain_error_db": -0.3003666228830582, "reference_mean_power": 1.6024474415231737e-05, "output_mean_power": 1.4953647255171871e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 7.931595726775402e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.29977249297526e-06, "output_power": 3.98596669897147e-06, "reference_active": true, "delta_db": -0.32911811017170123}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.410418312822663e-07, "output_power": 7.994466245385276e-07, "reference_active": true, "delta_db": 0.329467488729584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9106902570925117e-08, "output_power": 2.3304843185352496e-08, "reference_active": true, "delta_db": 0.8625589590304079}], "active_band_count": 3, "active_band_mae_db": 0.507048185977231}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8897072408149636, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9c2c0fd6d755afaae8bde170d9549bd13f9e6052ff9ee7f80e6425deea5eed76", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895836037567496, "standard_deviation": 0.0022658880388969167, "minimum": 0.9853019994381459, "maximum": 0.991969737816989, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.255185773779267, "standard_deviation": 0.024687231109917968, "minimum": 1.206207816768565, "maximum": 1.296670828191912, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.639230584253921, "standard_deviation": 0.27987288239675223, "minimum": 0.12216476969198659, "maximum": 0.8358903449170644, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1235315705892672, "standard_deviation": 0.013502068323154237, "minimum": 0.10907594456208297, "maximum": 0.14770228370336466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968726980708275, "standard_deviation": 0.000981378126828334, "minimum": 0.9950390144645487, "maximum": 0.9978846249511277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.778982597070741, "standard_deviation": 0.03090468188270795, "minimum": 0.7460764558987257, "maximum": 0.8335001624831934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5857896150237608, "standard_deviation": 0.28106061242845715, "minimum": 0.16911361487731288, "maximum": 0.9728483829366508, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07252410074783124, "standard_deviation": 0.01139663484641237, "minimum": 0.05934724335847336, "maximum": 0.09233294193247375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9776620973319639, "standard_deviation": 0.002291019755560735, "minimum": 0.9747846752007245, "maximum": 0.9817604202892442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.421667557800065, "standard_deviation": 0.04091243765378691, "minimum": 1.3828619612244157, "maximum": 1.5050331997172923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.40720482915361517, "standard_deviation": 0.1677095269895404, "minimum": 0.19015489107800723, "maximum": 0.6549571328145763, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18249654022529016, "standard_deviation": 0.00876732924265065, "minimum": 0.16599924107705968, "maximum": 0.19337419243159384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9909438496534548, "standard_deviation": 0.0015186010010521576, "minimum": 0.9892826256518081, "maximum": 0.9929936989785095, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2245344769094277, "standard_deviation": 0.03606761729869789, "minimum": 1.1747374717381331, "maximum": 1.285983378442175, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6530511419941265, "standard_deviation": 0.0944460156538742, "minimum": 0.507048185977231, "maximum": 0.8296284124571155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11456107641062466, "standard_deviation": 0.008310334454247528, "minimum": 0.10383174938406557, "maximum": 0.12497576663653209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9938242286998158, "standard_deviation": 0.001011902691754161, "minimum": 0.9919030622771374, "maximum": 0.9949564004871717, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1396866674476285, "standard_deviation": 0.058294205649653034, "minimum": 1.0619348611853885, "maximum": 1.2329170594379688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.43332292446512555, "standard_deviation": 0.1809555507575383, "minimum": 0.1131891106512729, "maximum": 0.6855770266857396, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08697175676317773, "standard_deviation": 0.006630044253053786, "minimum": 0.07943478194806108, "maximum": 0.099598723819477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977813830432481, "standard_deviation": 0.0005389729492165747, "minimum": 0.9971517309955925, "maximum": 0.9983996337569876, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7024120662294018, "standard_deviation": 0.05398964351941318, "minimum": 0.6239423146443744, "maximum": 0.7851934505936905, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7501961287130031, "standard_deviation": 0.26185064248425094, "minimum": 0.3931189335115468, "maximum": 1.079210350802961, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05103550188238744, "standard_deviation": 0.005075646101052566, "minimum": 0.045182373769549505, "maximum": 0.05721579218264054, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18839708837635805, "standard_deviation": 0.05998246125875095, "minimum": 0.09846242406562937, "maximum": 0.25584896993626255, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.552980566070769, "standard_deviation": 0.007217337078179501, "minimum": 5.542273542576107, "maximum": 5.563556943009799, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3202168297962326, "standard_deviation": 0.015356072439871131, "minimum": 0.29746369059460226, "maximum": 0.35134429874539763, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05428262322920609, "standard_deviation": 0.0423672330054759, "minimum": -0.12234201722339709, "maximum": -0.004906014510293626, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947296988766217, "standard_deviation": 0.00043741074586186435, "minimum": 0.9941063368328028, "maximum": 0.9953882866620125, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1875619771206387, "standard_deviation": 0.043784090577929605, "minimum": 1.1314421689164778, "maximum": 1.2725180674739323, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.350156356722786, "standard_deviation": 0.2045635762093788, "minimum": 0.12023456072507686, "maximum": 0.6659652675371582, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07157381589386941, "standard_deviation": 0.002964268609530697, "minimum": 0.06718946976459983, "maximum": 0.07551978998612752, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984631118962652, "standard_deviation": 0.00026531474707253925, "minimum": 0.9980244665243964, "maximum": 0.9988660528066735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6444636872177989, "standard_deviation": 0.022409222571519415, "minimum": 0.6044975016593447, "maximum": 0.6699009512621679, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5139419204756753, "standard_deviation": 0.2196538030180691, "minimum": 0.10769892262983526, "maximum": 0.750421275212731, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03773123666213758, "standard_deviation": 0.003162575266887741, "minimum": 0.033355947920078365, "maximum": 0.04254618842711308, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-23283-0016", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.5217108726501465, "predicted_mos": 2.5217108726501465, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "2005be3837f554acfdab07e75acdf5bb325dd6a37ea79d0b6f8350dc8bb5310f", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9432676997967064, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1514060584275403, "全帯域_10ms_rms_envelope_nrmse": 0.07402440335624258, "全帯域_10ms_envelope_correlation": 0.994356204922796, "全帯域_10ms_modulation_mae_db": 0.2777036855802846, "全帯域_40ms_log_rms_mae_db": 0.625751848377752, "全帯域_40ms_rms_envelope_nrmse": 0.036937082857968144, "全帯域_40ms_envelope_correlation": 0.9985486312477239, "全帯域_40ms_modulation_mae_db": 0.36664977252567926, "300–1000 Hz_10ms_log_rms_mae_db": 1.3384644388660762, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0909020541596782, "300–1000 Hz_10ms_envelope_correlation": 0.993326083232789, "300–1000 Hz_10ms_modulation_mae_db": 0.4973980130347006, "300–1000 Hz_40ms_log_rms_mae_db": 0.8371680750748337, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04803666064079027, "300–1000 Hz_40ms_envelope_correlation": 0.9981552584449143, "300–1000 Hz_40ms_modulation_mae_db": 0.9293879916500388, "1000–3000 Hz_10ms_log_rms_mae_db": 1.470569665889269, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1363498506768761, "1000–3000 Hz_10ms_envelope_correlation": 0.9871823443419636, "1000–3000 Hz_10ms_modulation_mae_db": 0.3490064805074278, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9644428784891288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07409909456591034, "1000–3000 Hz_40ms_envelope_correlation": 0.996740727058357, "1000–3000 Hz_40ms_modulation_mae_db": 0.39556762325409794, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7917014925754708, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2127689860803638, "3000–6000 Hz_10ms_envelope_correlation": 0.9698488736231436, "3000–6000 Hz_10ms_modulation_mae_db": 0.6887603621565043, "3000–6000 Hz_40ms_log_rms_mae_db": 1.58878663059684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13794042600513542, "3000–6000 Hz_40ms_envelope_correlation": 0.9879522385852132, "3000–6000 Hz_40ms_modulation_mae_db": 0.8777770274773312, "mr_stft_db": 5.742021871830697, "band_mae_db": 0.2347118878362462, "rms_gain_error_db": -0.07760540009767623, "output_peak": 0.32807196276476047}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1221574485361239, "全帯域_10ms_rms_envelope_nrmse": 0.08328604145675955, "全帯域_10ms_envelope_correlation": 0.9928519286117576, "全帯域_10ms_modulation_mae_db": 0.31083992670903987, "全帯域_40ms_log_rms_mae_db": 0.6534725372763811, "全帯域_40ms_rms_envelope_nrmse": 0.05210673782170844, "全帯域_40ms_envelope_correlation": 0.9970469311506587, "全帯域_40ms_modulation_mae_db": 0.3289218927337476, "300–1000 Hz_10ms_log_rms_mae_db": 1.3600852417395797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1059870245551311, "300–1000 Hz_10ms_envelope_correlation": 0.9908738227312827, "300–1000 Hz_10ms_modulation_mae_db": 0.44985204198864054, "300–1000 Hz_40ms_log_rms_mae_db": 0.9233253069236749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07151689611795264, "300–1000 Hz_40ms_envelope_correlation": 0.9956710137482037, "300–1000 Hz_40ms_modulation_mae_db": 0.6964049087692507, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4540693660519497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12508791258569435, "1000–3000 Hz_10ms_envelope_correlation": 0.9891698849300081, "1000–3000 Hz_10ms_modulation_mae_db": 1.0604262432436211, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9869136123272203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06615900425684469, "1000–3000 Hz_40ms_envelope_correlation": 0.9973360913217774, "1000–3000 Hz_40ms_modulation_mae_db": 0.7832650916273333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7589423439444387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20039170365205325, "3000–6000 Hz_10ms_envelope_correlation": 0.9727545599304689, "3000–6000 Hz_10ms_modulation_mae_db": 0.6593690393592518, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5296387997406984, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12555662398510017, "3000–6000 Hz_40ms_envelope_correlation": 0.9890501900757172, "3000–6000 Hz_40ms_modulation_mae_db": 0.9179548065603816, "mr_stft_db": 5.732648770419725, "band_mae_db": 0.15031470000672922, "rms_gain_error_db": -0.08228143412330613, "output_peak": 0.3213835333899559}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.111699176392844, "全帯域_10ms_rms_envelope_nrmse": 0.07086015083039103, "全帯域_10ms_envelope_correlation": 0.9949582745855897, "全帯域_10ms_modulation_mae_db": 0.5315721878618492, "全帯域_40ms_log_rms_mae_db": 0.5915607330070809, "全帯域_40ms_rms_envelope_nrmse": 0.03504744591592866, "全帯域_40ms_envelope_correlation": 0.9988574155060915, "全帯域_40ms_modulation_mae_db": 0.7407615701404696, "300–1000 Hz_10ms_log_rms_mae_db": 1.2978907782638687, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0942858693706378, "300–1000 Hz_10ms_envelope_correlation": 0.9928583307354144, "300–1000 Hz_10ms_modulation_mae_db": 0.6247246633880136, "300–1000 Hz_40ms_log_rms_mae_db": 0.8591593994183356, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0508184576244869, "300–1000 Hz_40ms_envelope_correlation": 0.997984400529944, "300–1000 Hz_40ms_modulation_mae_db": 1.2369678316023998, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4810119114944238, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13728678577736025, "1000–3000 Hz_10ms_envelope_correlation": 0.9875416521993308, "1000–3000 Hz_10ms_modulation_mae_db": 0.658513698881059, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0139286352996844, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08307890906215713, "1000–3000 Hz_40ms_envelope_correlation": 0.9963096038317771, "1000–3000 Hz_40ms_modulation_mae_db": 0.7540552699600114, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8079767283239472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20206584781823028, "3000–6000 Hz_10ms_envelope_correlation": 0.9723695878334213, "3000–6000 Hz_10ms_modulation_mae_db": 0.5294616442677932, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5429266030622542, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11998905276200691, "3000–6000 Hz_40ms_envelope_correlation": 0.9902399471045962, "3000–6000 Hz_40ms_modulation_mae_db": 0.7554073886731979, "mr_stft_db": 5.724888204582915, "band_mae_db": 0.20586831610964518, "rms_gain_error_db": -0.1288548485526829, "output_peak": 0.2991742392094184}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1305407935746847, "全帯域_10ms_rms_envelope_nrmse": 0.0678898612617723, "全帯域_10ms_envelope_correlation": 0.9952911728556889, "全帯域_10ms_modulation_mae_db": 0.18562884936954002, "全帯域_40ms_log_rms_mae_db": 0.5618126374665273, "全帯域_40ms_rms_envelope_nrmse": 0.03185186310548623, "全帯域_40ms_envelope_correlation": 0.9989692889507715, "全帯域_40ms_modulation_mae_db": 0.42724308237536895, "300–1000 Hz_10ms_log_rms_mae_db": 1.412989335327339, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09462587866013814, "300–1000 Hz_10ms_envelope_correlation": 0.992697639333745, "300–1000 Hz_10ms_modulation_mae_db": 0.4178325103642983, "300–1000 Hz_40ms_log_rms_mae_db": 0.9639463102187964, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04879135617433539, "300–1000 Hz_40ms_envelope_correlation": 0.9980211057392031, "300–1000 Hz_40ms_modulation_mae_db": 0.7751633992586201, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5429331237964343, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12307143874624611, "1000–3000 Hz_10ms_envelope_correlation": 0.9905716898843329, "1000–3000 Hz_10ms_modulation_mae_db": 0.245198009468069, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0475546145108454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07983128638962983, "1000–3000 Hz_40ms_envelope_correlation": 0.9970357149734467, "1000–3000 Hz_40ms_modulation_mae_db": 0.27585974092216065, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7655767288847806, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19937697351141456, "3000–6000 Hz_10ms_envelope_correlation": 0.9731538578408598, "3000–6000 Hz_10ms_modulation_mae_db": 0.5954903902578941, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4687542827712645, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1190534998144739, "3000–6000 Hz_40ms_envelope_correlation": 0.9904745354140824, "3000–6000 Hz_40ms_modulation_mae_db": 0.7100143587806156, "mr_stft_db": 5.73644026588367, "band_mae_db": 0.20894340239843845, "rms_gain_error_db": -0.08700300407049048, "output_peak": 0.2933878227377906}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1711455788544414, "全帯域_10ms_rms_envelope_nrmse": 0.07355606574613838, "全帯域_10ms_envelope_correlation": 0.9944016424692571, "全帯域_10ms_modulation_mae_db": 0.24918733143112445, "全帯域_40ms_log_rms_mae_db": 0.642076729845829, "全帯域_40ms_rms_envelope_nrmse": 0.03772478614975542, "全帯域_40ms_envelope_correlation": 0.99845033473369, "全帯域_40ms_modulation_mae_db": 0.33862868318968165, "300–1000 Hz_10ms_log_rms_mae_db": 1.330646768198198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09184240583145314, "300–1000 Hz_10ms_envelope_correlation": 0.9932047618437182, "300–1000 Hz_10ms_modulation_mae_db": 0.4974999090372989, "300–1000 Hz_40ms_log_rms_mae_db": 0.8928383575828068, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055850527095489384, "300–1000 Hz_40ms_envelope_correlation": 0.9974529536607536, "300–1000 Hz_40ms_modulation_mae_db": 0.8856008571835879, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4900282343600062, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12754014985620601, "1000–3000 Hz_10ms_envelope_correlation": 0.988515838356205, "1000–3000 Hz_10ms_modulation_mae_db": 0.6786411016614334, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9787831461438391, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06969983302896554, "1000–3000 Hz_40ms_envelope_correlation": 0.996693033694815, "1000–3000 Hz_40ms_modulation_mae_db": 0.8464205732943971, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8092470000564878, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20084225986358484, "3000–6000 Hz_10ms_envelope_correlation": 0.9758590981662463, "3000–6000 Hz_10ms_modulation_mae_db": 0.7790446006548869, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5431685569344098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12468488708897012, "3000–6000 Hz_40ms_envelope_correlation": 0.991163509790348, "3000–6000 Hz_40ms_modulation_mae_db": 0.9442612693185973, "mr_stft_db": 5.71700127611538, "band_mae_db": 0.1403485004470598, "rms_gain_error_db": -0.05659682227500545, "output_peak": 0.3287214714687884}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1363888197597756, "全帯域_10ms_rms_envelope_nrmse": 0.07767942394998685, "全帯域_10ms_envelope_correlation": 0.9937382960950358, "全帯域_10ms_modulation_mae_db": 0.32630343577328896, "全帯域_40ms_log_rms_mae_db": 0.6486024535363137, "全帯域_40ms_rms_envelope_nrmse": 0.04549258355238751, "全帯域_40ms_envelope_correlation": 0.9976990420424571, "全帯域_40ms_modulation_mae_db": 0.2814537863844651, "300–1000 Hz_10ms_log_rms_mae_db": 1.3914366724030856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09257519357791402, "300–1000 Hz_10ms_envelope_correlation": 0.9930046512715134, "300–1000 Hz_10ms_modulation_mae_db": 0.6382878527204013, "300–1000 Hz_40ms_log_rms_mae_db": 0.8987357040313231, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0548513680568429, "300–1000 Hz_40ms_envelope_correlation": 0.9974170990569962, "300–1000 Hz_40ms_modulation_mae_db": 0.9579392155464704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4966397983007147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12414633287962343, "1000–3000 Hz_10ms_envelope_correlation": 0.9895121264034562, "1000–3000 Hz_10ms_modulation_mae_db": 0.4312154316643309, "1000–3000 Hz_40ms_log_rms_mae_db": 1.073165078107226, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07501475087661803, "1000–3000 Hz_40ms_envelope_correlation": 0.9965908534031932, "1000–3000 Hz_40ms_modulation_mae_db": 0.35511163396993456, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8524257779437234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22398867614699672, "3000–6000 Hz_10ms_envelope_correlation": 0.9660849438175942, "3000–6000 Hz_10ms_modulation_mae_db": 0.2502229802847476, "3000–6000 Hz_40ms_log_rms_mae_db": 1.596481461028217, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1461214863345922, "3000–6000 Hz_40ms_envelope_correlation": 0.9847834448338868, "3000–6000 Hz_40ms_modulation_mae_db": 0.6129785931253132, "mr_stft_db": 5.722900142890481, "band_mae_db": 0.14251487295544965, "rms_gain_error_db": -0.03923334243254313, "output_peak": 0.3165016003670673}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.117147160562977, "全帯域_10ms_rms_envelope_nrmse": 0.07324893454654578, "全帯域_10ms_envelope_correlation": 0.9944556504350851, "全帯域_10ms_modulation_mae_db": 0.3549164254315344, "全帯域_40ms_log_rms_mae_db": 0.6288312281047274, "全帯域_40ms_rms_envelope_nrmse": 0.03464911034303614, "全帯域_40ms_envelope_correlation": 0.9986926719829367, "全帯域_40ms_modulation_mae_db": 0.5628145923415647, "300–1000 Hz_10ms_log_rms_mae_db": 1.3915070563922318, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0903568242447482, "300–1000 Hz_10ms_envelope_correlation": 0.9933453278138983, "300–1000 Hz_10ms_modulation_mae_db": 0.6453021098483962, "300–1000 Hz_40ms_log_rms_mae_db": 0.8857812968978557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04603784723023569, "300–1000 Hz_40ms_envelope_correlation": 0.9982237927911489, "300–1000 Hz_40ms_modulation_mae_db": 1.132015250415612, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5553466705609698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15122096414937578, "1000–3000 Hz_10ms_envelope_correlation": 0.984044550682843, "1000–3000 Hz_10ms_modulation_mae_db": 1.021063262203102, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1204741403913976, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08499537325145437, "1000–3000 Hz_40ms_envelope_correlation": 0.9954407140379068, "1000–3000 Hz_40ms_modulation_mae_db": 1.4166650881199694, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8088134893962955, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21306708097597798, "3000–6000 Hz_10ms_envelope_correlation": 0.9696292776523743, "3000–6000 Hz_10ms_modulation_mae_db": 0.5474442530196958, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5276862567076457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12567795430186862, "3000–6000 Hz_40ms_envelope_correlation": 0.9901711363515707, "3000–6000 Hz_40ms_modulation_mae_db": 0.6174479896730166, "mr_stft_db": 5.726637556714949, "band_mae_db": 0.17942439193995266, "rms_gain_error_db": -0.048687984821115415, "output_peak": 0.32160249420773523}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1430586351966043, "全帯域_10ms_rms_envelope_nrmse": 0.07844787820160608, "全帯域_10ms_envelope_correlation": 0.9937736352478557, "全帯域_10ms_modulation_mae_db": 0.15638135402577166, "全帯域_40ms_log_rms_mae_db": 0.6432011103479683, "全帯域_40ms_rms_envelope_nrmse": 0.039477726866472114, "全帯域_40ms_envelope_correlation": 0.9984871759932732, "全帯域_40ms_modulation_mae_db": 0.49311226970150424, "300–1000 Hz_10ms_log_rms_mae_db": 1.372928454679291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09673566671249909, "300–1000 Hz_10ms_envelope_correlation": 0.9925866563536501, "300–1000 Hz_10ms_modulation_mae_db": 0.43435781641595445, "300–1000 Hz_40ms_log_rms_mae_db": 0.9229137761902401, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0522890807259745, "300–1000 Hz_40ms_envelope_correlation": 0.9980037733935463, "300–1000 Hz_40ms_modulation_mae_db": 0.9688532044686711, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5015184671244455, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12694605027940334, "1000–3000 Hz_10ms_envelope_correlation": 0.9887879713698887, "1000–3000 Hz_10ms_modulation_mae_db": 0.6811442481033252, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9958864511473025, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06745378172694895, "1000–3000 Hz_40ms_envelope_correlation": 0.9971683149676153, "1000–3000 Hz_40ms_modulation_mae_db": 0.5243888203597905, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7714374130621104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19098025372819627, "3000–6000 Hz_10ms_envelope_correlation": 0.9759783053629332, "3000–6000 Hz_10ms_modulation_mae_db": 0.4295482054382655, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5143991857002492, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10684503430249427, "3000–6000 Hz_40ms_envelope_correlation": 0.9921625046278608, "3000–6000 Hz_40ms_modulation_mae_db": 0.5928174099475213, "mr_stft_db": 5.736663117376323, "band_mae_db": 0.16059547653679587, "rms_gain_error_db": -0.13053002130946237, "output_peak": 0.33100314183001517}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 71920, "output_samples": 71920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.683078178234254, "spectral_convergence": 0.2784395992477099, "spectral_convergence_status": "ok", "amplitude_floor": 7.580332405627278e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.604328021201619, "spectral_convergence": 0.22882135892808425, "spectral_convergence_status": "ok", "amplitude_floor": 6.818562526348602e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9386594160562165, "spectral_convergence": 0.2343022193287057, "spectral_convergence_status": "ok", "amplitude_floor": 5.715517707749406e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020741335082478346, "output_power": 0.00021241150829496563, "delta_db": 0.10341335085260964, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006538184008046484, "output_power": 0.0006517968684534214, "delta_db": -0.013448694533155019, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004887436095925812, "output_power": 0.0004683624745699021, "delta_db": -0.18499000057122983, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 3.3579219734638266e-05, "output_power": 3.036132387887783e-05, "delta_db": -0.4374989574597703, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.100095928785279e-05, "output_power": 5.730345521329412e-05, "delta_db": -0.2715585537362259, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.8128919162253983e-05, "output_power": 1.654381768447827e-05, "delta_db": -0.39736176986448646, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2347118878362462, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07760540009767623, "pitch": null, "output_rms_dbfs": -28.426098928446415, "output_peak": 0.32807196276476047, "reference_rms_dbfs": -28.34849352834874, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.742021871830697}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 71920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 0.00010701965100918563, "log_rms_mae_db": 1.1514060584275403, "rms_envelope_nrmse": 0.07402440335624258, "envelope_correlation": 0.994356204922796, "rms_gain_error_db": -0.07760540681568939, "reference_mean_power": 0.0014643132072841782, "output_mean_power": 0.0014383793476541937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 5.509210445243936e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038836857080924806, "output_power": 0.00037365709981419565, "reference_active": true, "delta_db": -0.16770838209522396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3200931448411458e-05, "output_power": 2.0171152299691722e-05, "reference_active": true, "delta_db": -0.6077471239235614}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.17358638897755e-06, "output_power": 9.29618397967067e-06, "reference_active": true, "delta_db": 0.05765555072206824}], "active_band_count": 3, "active_band_mae_db": 0.2777036855802846}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 0.00010161161302678278, "log_rms_mae_db": 0.625751848377752, "rms_envelope_nrmse": 0.036937082857968144, "envelope_correlation": 0.9985486312477239, "rms_gain_error_db": -0.07760516107941709, "reference_mean_power": 0.0014741623423244587, "output_mean_power": 0.0014480541305778908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 4.939959887293225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035657053350397007, "output_power": 0.00034122580365394365, "reference_active": true, "delta_db": -0.19103585538524367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.1722039793266e-06, "output_power": 7.82315513876347e-06, "reference_active": true, "delta_db": -0.18957255443521553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.984441222185337e-07, "output_power": 5.918308179642125e-07, "reference_active": true, "delta_db": -0.7193409077565787}], "active_band_count": 3, "active_band_mae_db": 0.36664977252567926}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 9.15514590150253e-05, "log_rms_mae_db": 1.3384644388660762, "rms_envelope_nrmse": 0.0909020541596782, "envelope_correlation": 0.993326083232789, "rms_gain_error_db": -0.11890294543461719, "reference_mean_power": 0.0006933001790048939, "output_mean_power": 0.0006745762049478213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 3.8791214865659925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002719560122465178, "output_power": 0.00025402478155564517, "reference_active": true, "delta_db": -0.2962257781677846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.187110727790427e-05, "output_power": 1.782258807831141e-05, "reference_active": true, "delta_db": -0.8890000115548395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.698791354460839e-06, "output_power": 6.241653078743019e-06, "reference_active": true, "delta_db": -0.3069682493814775}], "active_band_count": 3, "active_band_mae_db": 0.4973980130347006}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 8.797925167476037e-05, "log_rms_mae_db": 0.8371680750748337, "rms_envelope_nrmse": 0.04803666064079027, "envelope_correlation": 0.9981552584449143, "rms_gain_error_db": -0.11890298010661496, "reference_mean_power": 0.000697963583829786, "output_mean_power": 0.0006791136596684408}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.5511880459540643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002583969788516632, "output_power": 0.00023956937709221642, "reference_active": true, "delta_db": -0.3285612797106336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.105856462043575e-06, "output_power": 7.0349069546133465e-06, "reference_active": true, "delta_db": -0.6154055129200054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388944870715119e-07, "output_power": 4.178394319914061e-07, "reference_active": true, "delta_db": -1.8441971823194776}], "active_band_count": 3, "active_band_mae_db": 0.9293879916500388}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 4.592612274041097e-05, "log_rms_mae_db": 1.470569665889269, "rms_envelope_nrmse": 0.1363498506768761, "envelope_correlation": 0.9871823443419636, "rms_gain_error_db": -0.3203581825831451, "reference_mean_power": 8.288074215773485e-05, "output_mean_power": 7.698707383689881e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 4.287875385442721e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2411523150580774e-05, "output_power": 1.9268526594422286e-05, "reference_active": true, "delta_db": -0.6562286667297503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.113877152498063e-06, "output_power": 2.8905027503044284e-06, "reference_active": true, "delta_db": -0.3232808800748313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.881539635700238e-07, "output_power": 9.729121510889649e-07, "reference_active": true, "delta_db": -0.06750989471770172}], "active_band_count": 3, "active_band_mae_db": 0.3490064805074278}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 4.2997336393594785e-05, "log_rms_mae_db": 0.9644428784891288, "rms_envelope_nrmse": 0.07409909456591034, "envelope_correlation": 0.996740727058357, "rms_gain_error_db": -0.3203583129964733, "reference_mean_power": 8.343814618382176e-05, "output_mean_power": 7.750483840924114e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.9683446107037247e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.063395086340567e-05, "output_power": 1.7787172450197766e-05, "reference_active": true, "delta_db": -0.6447547634410076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4408129681959194e-06, "output_power": 1.4002600117767134e-06, "reference_active": true, "delta_db": -0.12398922212541044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9896369199765786e-08, "output_power": 4.531824219815435e-08, "reference_active": true, "delta_db": -0.41795888419587585}], "active_band_count": 3, "active_band_mae_db": 0.39556762325409794}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 2.431018241105218e-05, "log_rms_mae_db": 1.7917014925754708, "rms_envelope_nrmse": 0.2127689860803638, "envelope_correlation": 0.9698488736231436, "rms_gain_error_db": -0.4208507484819863, "reference_mean_power": 1.5917473583289094e-05, "output_mean_power": 1.4447378903196757e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 8.500689890351627e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.331755201814068e-06, "output_power": 3.912379894188388e-06, "reference_active": true, "delta_db": -0.4422288720241755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3672340908387822e-06, "output_power": 1.3008572546938511e-06, "reference_active": true, "delta_db": -0.2161323532255421}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.790743563061608e-07, "output_power": 3.859330021457475e-07, "reference_active": true, "delta_db": 1.4079198612197952}], "active_band_count": 3, "active_band_mae_db": 0.6887603621565043}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 1.765504974955162e-05, "log_rms_mae_db": 1.58878663059684, "rms_envelope_nrmse": 0.13794042600513542, "envelope_correlation": 0.9879522385852132, "rms_gain_error_db": -0.4208516970658088, "reference_mean_power": 1.6024474415231737e-05, "output_mean_power": 1.4544494251765774e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 7.931595726775402e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.29977249297526e-06, "output_power": 3.888836771953724e-06, "reference_active": true, "delta_db": -0.4362576237866849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.410418312822663e-07, "output_power": 7.425488666471424e-07, "reference_active": true, "delta_db": 0.008823152760767335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9106902570925117e-08, "output_power": 3.162389210943709e-08, "reference_active": true, "delta_db": 2.1882503058845413}], "active_band_count": 3, "active_band_mae_db": 0.8777770274773312}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.889459514252344, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "203a358ea820766ff66bfb3249bafa9625922d72a49e0d9ad61ccf43fa140aad", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9881657572710035, "standard_deviation": 0.0019806259096968246, "minimum": 0.984044550682843, "maximum": 0.9905716898843329, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4990146546972767, "standard_deviation": 0.03453380917275372, "minimum": 1.4540693660519497, "maximum": 1.5553466705609698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.640651059466546, "standard_deviation": 0.2951100573894906, "minimum": 0.245198009468069, "maximum": 1.0604262432436211, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13145618561884817, "standard_deviation": 0.009625559811218156, "minimum": 0.12307143874624611, "maximum": 0.15122096414937578, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966643816611112, "standard_deviation": 0.0005950170223472078, "minimum": 0.9954407140379068, "maximum": 0.9973360913217774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0226435695520806, "standard_deviation": 0.053616873766924845, "minimum": 0.9644428784891288, "maximum": 1.1204741403913976, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6689167301884619, "standard_deviation": 0.37023830666012475, "minimum": 0.27585974092216065, "maximum": 1.4166650881199694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07504150414481611, "standard_deviation": 0.007096438136359264, "minimum": 0.06615900425684469, "maximum": 0.08499537325145437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9719598130283802, "standard_deviation": 0.003337653898600807, "minimum": 0.9660849438175942, "maximum": 0.9759783053629332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7957651217734067, "standard_deviation": 0.030676144192183667, "minimum": 1.7589423439444387, "maximum": 1.8524257779437234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5599176844298799, "standard_deviation": 0.1646481955637741, "minimum": 0.2502229802847476, "maximum": 0.7790446006548869, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20543522272210218, "standard_deviation": 0.010418346257430763, "minimum": 0.19098025372819627, "maximum": 0.22398867614699672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9894996883479095, "standard_deviation": 0.002288094869796502, "minimum": 0.9847834448338868, "maximum": 0.9921625046278608, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5389802220676974, "standard_deviation": 0.0406569491654362, "minimum": 1.4687542827712645, "maximum": 1.596481461028217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7535823554444968, "standard_deviation": 0.1438500805630351, "minimum": 0.5928174099475213, "maximum": 0.9442612693185973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12573362057433018, "standard_deviation": 0.011953530271125746, "minimum": 0.10684503430249427, "maximum": 0.1461214863345922, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927371591645013, "standard_deviation": 0.000803369963101608, "minimum": 0.9908738227312827, "maximum": 0.9933453278138983, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3619935932337088, "standard_deviation": 0.037998371323449666, "minimum": 1.2978907782638687, "maximum": 1.412989335327339, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5256568645997131, "standard_deviation": 0.09570379040663302, "minimum": 0.4178325103642983, "maximum": 0.6453021098483962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09466386463902496, "standard_deviation": 0.0050380782804512365, "minimum": 0.0903568242447482, "maximum": 0.1059870245551311, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976161746705887, "standard_deviation": 0.0008417266719574862, "minimum": 0.9956710137482037, "maximum": 0.9982237927911489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8979835282922333, "standard_deviation": 0.0396350799487032, "minimum": 0.8371680750748337, "maximum": 0.9639463102187964, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9477915823618314, "standard_deviation": 0.17538119995834206, "minimum": 0.6964049087692507, "maximum": 1.2369678316023998, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05352402420826346, "standard_deviation": 0.008002814007461558, "minimum": 0.04603784723023569, "maximum": 0.07151689611795264, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17784019352878963, "standard_deviation": 0.035250983647702246, "minimum": 0.1403485004470598, "maximum": 0.2347118878362462, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.729900150726767, "standard_deviation": 0.008401930532321534, "minimum": 5.71700127611538, "maximum": 5.742021871830697, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3174807832469414, "standard_deviation": 0.013990606014941918, "minimum": 0.2933878227377906, "maximum": 0.33100314183001517, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08134910721028527, "standard_deviation": 0.034175118489285075, "minimum": -0.13053002130946237, "maximum": -0.03923334243254313, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9942283506528833, "standard_deviation": 0.0007654950915017816, "minimum": 0.9928519286117576, "maximum": 0.9952911728556889, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1354429589131239, "standard_deviation": 0.019593841056383243, "minimum": 1.111699176392844, "maximum": 1.1711455788544414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29906664952280415, "standard_deviation": 0.11602341270175406, "minimum": 0.15638135402577166, "maximum": 0.5315721878618492, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07487409491868031, "standard_deviation": 0.004801699584388537, "minimum": 0.0678898612617723, "maximum": 0.08328604145675955, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983439364509503, "standard_deviation": 0.000648976405948254, "minimum": 0.9970469311506587, "maximum": 0.9989692889507715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6244136597453225, "standard_deviation": 0.03187918465213599, "minimum": 0.5618126374665273, "maximum": 0.6534725372763811, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4424482061740601, "standard_deviation": 0.15194373775739448, "minimum": 0.2814537863844651, "maximum": 0.7407615701404696, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03916091707659283, "standard_deviation": 0.006606186161279966, "minimum": 0.03185186310548623, "maximum": 0.05210673782170844, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-23283-0016", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.8675806522369385, "predicted_mos": 2.8675806522369385, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "3e3dac08970723ce100dda2cc8a23d980a157dfec81358ad53b3b92374a2766c", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.096128299832344, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0975665262843155, "全帯域_10ms_rms_envelope_nrmse": 0.07372561439768179, "全帯域_10ms_envelope_correlation": 0.9944381260756961, "全帯域_10ms_modulation_mae_db": 0.17641041527959145, "全帯域_40ms_log_rms_mae_db": 0.6216689611779687, "全帯域_40ms_rms_envelope_nrmse": 0.04101947416839685, "全帯域_40ms_envelope_correlation": 0.9982469665009791, "全帯域_40ms_modulation_mae_db": 0.37537606185901556, "300–1000 Hz_10ms_log_rms_mae_db": 1.2397990419140577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09552312052529986, "300–1000 Hz_10ms_envelope_correlation": 0.9926075759926738, "300–1000 Hz_10ms_modulation_mae_db": 0.19728517362534925, "300–1000 Hz_40ms_log_rms_mae_db": 0.8049868269160341, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05940148975786664, "300–1000 Hz_40ms_envelope_correlation": 0.9970647997129889, "300–1000 Hz_40ms_modulation_mae_db": 0.6144973195309427, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4261202290807298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13186128995894086, "1000–3000 Hz_10ms_envelope_correlation": 0.9879071404770365, "1000–3000 Hz_10ms_modulation_mae_db": 0.6577826159765169, "1000–3000 Hz_40ms_log_rms_mae_db": 0.937879820844202, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06931068391173188, "1000–3000 Hz_40ms_envelope_correlation": 0.9970624999205192, "1000–3000 Hz_40ms_modulation_mae_db": 0.77244791982433, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7489359761182457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18193736504805022, "3000–6000 Hz_10ms_envelope_correlation": 0.9777062753914059, "3000–6000 Hz_10ms_modulation_mae_db": 0.42571173551521424, "3000–6000 Hz_40ms_log_rms_mae_db": 1.444371149842914, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10736453327443772, "3000–6000 Hz_40ms_envelope_correlation": 0.9918522518777343, "3000–6000 Hz_40ms_modulation_mae_db": 0.4789228906186476, "mr_stft_db": 5.672434237126144, "band_mae_db": 0.13311969301252466, "rms_gain_error_db": -0.09837616596990614, "output_peak": 0.3311809934957389}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1368542475510073, "全帯域_10ms_rms_envelope_nrmse": 0.07675763802293883, "全帯域_10ms_envelope_correlation": 0.9938984612627217, "全帯域_10ms_modulation_mae_db": 0.2676068604221781, "全帯域_40ms_log_rms_mae_db": 0.6053282858579404, "全帯域_40ms_rms_envelope_nrmse": 0.042564107578110605, "全帯域_40ms_envelope_correlation": 0.9980131703020144, "全帯域_40ms_modulation_mae_db": 0.2750015860322664, "300–1000 Hz_10ms_log_rms_mae_db": 1.3106089511308168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09183775852127214, "300–1000 Hz_10ms_envelope_correlation": 0.993151499139408, "300–1000 Hz_10ms_modulation_mae_db": 0.7387130525280506, "300–1000 Hz_40ms_log_rms_mae_db": 0.8579759045852904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049689329782748724, "300–1000 Hz_40ms_envelope_correlation": 0.9979560019782502, "300–1000 Hz_40ms_modulation_mae_db": 0.9208128165551971, "1000–3000 Hz_10ms_log_rms_mae_db": 1.439428163301042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13850607538423584, "1000–3000 Hz_10ms_envelope_correlation": 0.9864900850377639, "1000–3000 Hz_10ms_modulation_mae_db": 0.8773877712746062, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9461143853386164, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07961798968053206, "1000–3000 Hz_40ms_envelope_correlation": 0.9957373728002727, "1000–3000 Hz_40ms_modulation_mae_db": 0.7680329056835257, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6835307217379154, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20259026596855603, "3000–6000 Hz_10ms_envelope_correlation": 0.9725242121192776, "3000–6000 Hz_10ms_modulation_mae_db": 0.8184936144421355, "3000–6000 Hz_40ms_log_rms_mae_db": 1.416034249393058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10564114250012181, "3000–6000 Hz_40ms_envelope_correlation": 0.9921919061212036, "3000–6000 Hz_40ms_modulation_mae_db": 0.7970265792965741, "mr_stft_db": 5.66776316725856, "band_mae_db": 0.12814425927706447, "rms_gain_error_db": -0.05761997375631447, "output_peak": 0.32395272629875105}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1870167363870079, "全帯域_10ms_rms_envelope_nrmse": 0.07061182349059907, "全帯域_10ms_envelope_correlation": 0.994899926763067, "全帯域_10ms_modulation_mae_db": 0.22088575674336428, "全帯域_40ms_log_rms_mae_db": 0.6376277347226648, "全帯域_40ms_rms_envelope_nrmse": 0.03846551962175278, "全帯域_40ms_envelope_correlation": 0.998457131787003, "全帯域_40ms_modulation_mae_db": 0.5101384697185404, "300–1000 Hz_10ms_log_rms_mae_db": 1.2283266687799985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08837981887472426, "300–1000 Hz_10ms_envelope_correlation": 0.9936845610966587, "300–1000 Hz_10ms_modulation_mae_db": 0.512670208940012, "300–1000 Hz_40ms_log_rms_mae_db": 0.7889411831481642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050352148591431604, "300–1000 Hz_40ms_envelope_correlation": 0.997923714460061, "300–1000 Hz_40ms_modulation_mae_db": 1.004830220861261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3688927178460037, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12241242435648894, "1000–3000 Hz_10ms_envelope_correlation": 0.9892621934371649, "1000–3000 Hz_10ms_modulation_mae_db": 0.741268181873988, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8385234750047826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06268047436608334, "1000–3000 Hz_40ms_envelope_correlation": 0.9970932366129004, "1000–3000 Hz_40ms_modulation_mae_db": 0.771698219645117, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7552549584031254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18400855350075954, "3000–6000 Hz_10ms_envelope_correlation": 0.9773997189968114, "3000–6000 Hz_10ms_modulation_mae_db": 0.9042774644498919, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4519143872616136, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11286151245150346, "3000–6000 Hz_40ms_envelope_correlation": 0.9917071842459719, "3000–6000 Hz_40ms_modulation_mae_db": 0.7906526145558946, "mr_stft_db": 5.671192917383876, "band_mae_db": 0.10050792878970453, "rms_gain_error_db": -0.09373692567280043, "output_peak": 0.34276076917195697}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1199003773810843, "全帯域_10ms_rms_envelope_nrmse": 0.07048331540847812, "全帯域_10ms_envelope_correlation": 0.9948573309685594, "全帯域_10ms_modulation_mae_db": 0.1597963230945055, "全帯域_40ms_log_rms_mae_db": 0.5781455537996545, "全帯域_40ms_rms_envelope_nrmse": 0.031292806183364955, "全帯域_40ms_envelope_correlation": 0.9989194143177356, "全帯域_40ms_modulation_mae_db": 0.21855949397723387, "300–1000 Hz_10ms_log_rms_mae_db": 1.328714823476846, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0860743412672747, "300–1000 Hz_10ms_envelope_correlation": 0.993968353487766, "300–1000 Hz_10ms_modulation_mae_db": 0.5798765442494237, "300–1000 Hz_40ms_log_rms_mae_db": 0.8177239301121406, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044108789423816096, "300–1000 Hz_40ms_envelope_correlation": 0.9983541841138738, "300–1000 Hz_40ms_modulation_mae_db": 0.7981886142624849, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4123942712423354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12484899575596206, "1000–3000 Hz_10ms_envelope_correlation": 0.9890987715147583, "1000–3000 Hz_10ms_modulation_mae_db": 0.36872953718054774, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8801042169198052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06160118476981623, "1000–3000 Hz_40ms_envelope_correlation": 0.9976460578001471, "1000–3000 Hz_40ms_modulation_mae_db": 0.33509394160301104, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6914982464341621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19469732538729917, "3000–6000 Hz_10ms_envelope_correlation": 0.9743057753870048, "3000–6000 Hz_10ms_modulation_mae_db": 0.4567505780352683, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3881488875714652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10671364655632842, "3000–6000 Hz_40ms_envelope_correlation": 0.9921123878391623, "3000–6000 Hz_40ms_modulation_mae_db": 0.5521139931054341, "mr_stft_db": 5.669838090390343, "band_mae_db": 0.11780493691837379, "rms_gain_error_db": -0.035432764244371785, "output_peak": 0.3362136562469435}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1177976645083143, "全帯域_10ms_rms_envelope_nrmse": 0.0710175599826462, "全帯域_10ms_envelope_correlation": 0.995028624751011, "全帯域_10ms_modulation_mae_db": 0.25515013290575383, "全帯域_40ms_log_rms_mae_db": 0.6037913548384125, "全帯域_40ms_rms_envelope_nrmse": 0.038631491024475675, "全帯域_40ms_envelope_correlation": 0.9986786908268075, "全帯域_40ms_modulation_mae_db": 0.3172301465213878, "300–1000 Hz_10ms_log_rms_mae_db": 1.2654927472354642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08925958143754631, "300–1000 Hz_10ms_envelope_correlation": 0.9938757493492208, "300–1000 Hz_10ms_modulation_mae_db": 0.22450248412826204, "300–1000 Hz_40ms_log_rms_mae_db": 0.7911738018076012, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05762352759288484, "300–1000 Hz_40ms_envelope_correlation": 0.997632087946037, "300–1000 Hz_40ms_modulation_mae_db": 0.4410351216941594, "1000–3000 Hz_10ms_log_rms_mae_db": 1.439913722292158, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12555627104211195, "1000–3000 Hz_10ms_envelope_correlation": 0.9889635422509468, "1000–3000 Hz_10ms_modulation_mae_db": 0.9352873437007992, "1000–3000 Hz_40ms_log_rms_mae_db": 0.927236074536183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06758940942849795, "1000–3000 Hz_40ms_envelope_correlation": 0.9970429649848294, "1000–3000 Hz_40ms_modulation_mae_db": 0.9648614626750005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7378607311521872, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2135024916684514, "3000–6000 Hz_10ms_envelope_correlation": 0.9692360223877933, "3000–6000 Hz_10ms_modulation_mae_db": 0.32447635840436323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4588152678291424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1332774525875335, "3000–6000 Hz_40ms_envelope_correlation": 0.9873730037502485, "3000–6000 Hz_40ms_modulation_mae_db": 0.4469602036273903, "mr_stft_db": 5.67034788692374, "band_mae_db": 0.18442853192901923, "rms_gain_error_db": -0.16002179116260257, "output_peak": 0.3211805893675085}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1569943923733654, "全帯域_10ms_rms_envelope_nrmse": 0.0682454849300145, "全帯域_10ms_envelope_correlation": 0.9953028024164517, "全帯域_10ms_modulation_mae_db": 0.2530897255753981, "全帯域_40ms_log_rms_mae_db": 0.6311447688345885, "全帯域_40ms_rms_envelope_nrmse": 0.03570718291004211, "全帯域_40ms_envelope_correlation": 0.9987642452929119, "全帯域_40ms_modulation_mae_db": 0.06392118107305443, "300–1000 Hz_10ms_log_rms_mae_db": 1.3090582274173357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09011747101011536, "300–1000 Hz_10ms_envelope_correlation": 0.9935303607701849, "300–1000 Hz_10ms_modulation_mae_db": 0.37502248906869234, "300–1000 Hz_40ms_log_rms_mae_db": 0.8377338061401134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054548733444231105, "300–1000 Hz_40ms_envelope_correlation": 0.9976624151616301, "300–1000 Hz_40ms_modulation_mae_db": 0.5081745635929474, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4226911167811938, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11935977583406443, "1000–3000 Hz_10ms_envelope_correlation": 0.9901088179646653, "1000–3000 Hz_10ms_modulation_mae_db": 1.0289903754002556, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9616162384801362, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06295055169709445, "1000–3000 Hz_40ms_envelope_correlation": 0.9975436669734246, "1000–3000 Hz_40ms_modulation_mae_db": 1.0437665631377169, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7131564162594028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19444973080869987, "3000–6000 Hz_10ms_envelope_correlation": 0.9749417142691981, "3000–6000 Hz_10ms_modulation_mae_db": 0.38072449587071283, "3000–6000 Hz_40ms_log_rms_mae_db": 1.428475103933104, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1033161809085638, "3000–6000 Hz_40ms_envelope_correlation": 0.9925975563373237, "3000–6000 Hz_40ms_modulation_mae_db": 0.4347655172540435, "mr_stft_db": 5.675606608455699, "band_mae_db": 0.16721577648373528, "rms_gain_error_db": -0.11918859878848137, "output_peak": 0.3285309203345391}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1688350782532768, "全帯域_10ms_rms_envelope_nrmse": 0.07457531506774305, "全帯域_10ms_envelope_correlation": 0.9943222991076277, "全帯域_10ms_modulation_mae_db": 0.4880030634550832, "全帯域_40ms_log_rms_mae_db": 0.6237364415980078, "全帯域_40ms_rms_envelope_nrmse": 0.039341870072184655, "全帯域_40ms_envelope_correlation": 0.9984015619204845, "全帯域_40ms_modulation_mae_db": 0.806470289953556, "300–1000 Hz_10ms_log_rms_mae_db": 1.365748701623651, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09173467187449297, "300–1000 Hz_10ms_envelope_correlation": 0.9931252721900119, "300–1000 Hz_10ms_modulation_mae_db": 0.527389738400209, "300–1000 Hz_40ms_log_rms_mae_db": 0.8641026655849954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049685689155723294, "300–1000 Hz_40ms_envelope_correlation": 0.9979088763159706, "300–1000 Hz_40ms_modulation_mae_db": 0.9985624039379846, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4519396954524002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14104539447981113, "1000–3000 Hz_10ms_envelope_correlation": 0.9863840470783709, "1000–3000 Hz_10ms_modulation_mae_db": 0.8702962079406836, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9349246278487701, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07613317129066467, "1000–3000 Hz_40ms_envelope_correlation": 0.9967322653671046, "1000–3000 Hz_40ms_modulation_mae_db": 0.9267339530628345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7486623835994763, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19893803575251645, "3000–6000 Hz_10ms_envelope_correlation": 0.9732423813441977, "3000–6000 Hz_10ms_modulation_mae_db": 0.769233211296083, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4926867850513128, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11300805951425479, "3000–6000 Hz_40ms_envelope_correlation": 0.9914798900326186, "3000–6000 Hz_40ms_modulation_mae_db": 0.9135646603755112, "mr_stft_db": 5.684613893753845, "band_mae_db": 0.16360754947002742, "rms_gain_error_db": -0.08566811175746925, "output_peak": 0.3275699157928468}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1795936496882335, "全帯域_10ms_rms_envelope_nrmse": 0.07236797536029359, "全帯域_10ms_envelope_correlation": 0.9945830393891874, "全帯域_10ms_modulation_mae_db": 0.4576658944908339, "全帯域_40ms_log_rms_mae_db": 0.6244218108424299, "全帯域_40ms_rms_envelope_nrmse": 0.038382153765180387, "全帯域_40ms_envelope_correlation": 0.9983785647860021, "全帯域_40ms_modulation_mae_db": 0.5759406086205396, "300–1000 Hz_10ms_log_rms_mae_db": 1.2241941217585497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09770595146213339, "300–1000 Hz_10ms_envelope_correlation": 0.9921975117139821, "300–1000 Hz_10ms_modulation_mae_db": 0.7240434732612894, "300–1000 Hz_40ms_log_rms_mae_db": 0.7927945780497928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05684548643639902, "300–1000 Hz_40ms_envelope_correlation": 0.9972381252581081, "300–1000 Hz_40ms_modulation_mae_db": 1.3221088884672376, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4189190184613993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12678385168843456, "1000–3000 Hz_10ms_envelope_correlation": 0.9887454395195854, "1000–3000 Hz_10ms_modulation_mae_db": 0.6183019784984428, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9215191385204541, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06324522028520216, "1000–3000 Hz_40ms_envelope_correlation": 0.9974992476089668, "1000–3000 Hz_40ms_modulation_mae_db": 0.5565100122405376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7737301905313516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22931370289839575, "3000–6000 Hz_10ms_envelope_correlation": 0.9652636050643206, "3000–6000 Hz_10ms_modulation_mae_db": 0.900930611604307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5173392993246775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15158557521902583, "3000–6000 Hz_40ms_envelope_correlation": 0.9859978768164205, "3000–6000 Hz_40ms_modulation_mae_db": 1.0363252581510012, "mr_stft_db": 5.6818879042546016, "band_mae_db": 0.19214175917025778, "rms_gain_error_db": -0.037201785940805095, "output_peak": 0.32138530661067277}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 71920, "output_samples": 71920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.62522157424955, "spectral_convergence": 0.2809352052450187, "spectral_convergence_status": "ok", "amplitude_floor": 7.580332405627278e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.506298945852237, "spectral_convergence": 0.23002203424967127, "spectral_convergence_status": "ok", "amplitude_floor": 6.818562526348602e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.885782191276642, "spectral_convergence": 0.2366959369856768, "spectral_convergence_status": "ok", "amplitude_floor": 5.715517707749406e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020741335082478346, "output_power": 0.0002095846068460177, "delta_db": 0.04522674575998093, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006538184008046484, "output_power": 0.0006360905760984599, "delta_db": -0.1193817762578713, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004887436095925812, "output_power": 0.00047645350051280666, "delta_db": -0.1106056989999995, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 3.3579219734638266e-05, "output_power": 3.158711850912389e-05, "delta_db": -0.2656059060890779, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.100095928785279e-05, "output_power": 5.793673925647091e-05, "delta_db": -0.2238261570174915, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.8128919162253983e-05, "output_power": 1.827170605045737e-05, "delta_db": 0.03407187395072666, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13311969301252466, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09837616596990614, "pitch": null, "output_rms_dbfs": -28.446869694318643, "output_peak": 0.3311809934957389, "reference_rms_dbfs": -28.34849352834874, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.672434237126144}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 71920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 0.00010701965100918563, "log_rms_mae_db": 1.0975665262843155, "rms_envelope_nrmse": 0.07372561439768179, "envelope_correlation": 0.9944381260756961, "rms_gain_error_db": -0.09837619247221928, "reference_mean_power": 0.0014643132072841782, "output_mean_power": 0.0014315165068274202}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 5.509210445243936e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038836857080924806, "output_power": 0.00037433643670245346, "reference_active": true, "delta_db": -0.15981974860738007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3200931448411458e-05, "output_power": 2.152207909428816e-05, "reference_active": true, "delta_db": -0.3262119776518071}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.17358638897755e-06, "output_power": 9.265291919434902e-06, "reference_active": true, "delta_db": 0.04319951957958721}], "active_band_count": 3, "active_band_mae_db": 0.17641041527959145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 0.00010161161302678278, "log_rms_mae_db": 0.6216689611779687, "rms_envelope_nrmse": 0.04101947416839685, "envelope_correlation": 0.9982469665009791, "rms_gain_error_db": -0.09837610992184657, "reference_mean_power": 0.0014741623423244587, "output_mean_power": 0.0014411450749713054}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 4.939959887293225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035657053350397007, "output_power": 0.00034350850955292955, "reference_active": true, "delta_db": -0.16207950752302233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.1722039793266e-06, "output_power": 7.814067080086284e-06, "reference_active": true, "delta_db": -0.19462063026845394}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.984441222185337e-07, "output_power": 5.850444477333615e-07, "reference_active": true, "delta_db": -0.7694280477855703}], "active_band_count": 3, "active_band_mae_db": 0.37537606185901556}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 9.15514590150253e-05, "log_rms_mae_db": 1.2397990419140577, "rms_envelope_nrmse": 0.09552312052529986, "envelope_correlation": 0.9926075759926738, "rms_gain_error_db": -0.11980436325257232, "reference_mean_power": 0.0006933001790048939, "output_mean_power": 0.0006744362050319458}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 3.8791214865659925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002719560122465178, "output_power": 0.00026324370194635843, "reference_active": true, "delta_db": -0.14140674817000207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.187110727790427e-05, "output_power": 1.9868170178512707e-05, "reference_active": true, "delta_db": -0.4171289956850413}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.698791354460839e-06, "output_power": 6.750383225724581e-06, "reference_active": true, "delta_db": 0.03331977702100439}], "active_band_count": 3, "active_band_mae_db": 0.19728517362534925}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 8.797925167476037e-05, "log_rms_mae_db": 0.8049868269160341, "rms_envelope_nrmse": 0.05940148975786664, "envelope_correlation": 0.9970647997129889, "rms_gain_error_db": -0.11980440513655241, "reference_mean_power": 0.000697963583829786, "output_mean_power": 0.0006789727169326588}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.5511880459540643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002583969788516632, "output_power": 0.00024903178006430805, "reference_active": true, "delta_db": -0.16032658728203422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.105856462043575e-06, "output_power": 7.699509628278145e-06, "reference_active": true, "delta_db": -0.2233584261217052}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388944870715119e-07, "output_power": 4.565080562417582e-07, "reference_active": true, "delta_db": -1.4598069451890885}], "active_band_count": 3, "active_band_mae_db": 0.6144973195309427}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 4.592612274041097e-05, "log_rms_mae_db": 1.4261202290807298, "rms_envelope_nrmse": 0.13186128995894086, "envelope_correlation": 0.9879071404770365, "rms_gain_error_db": -0.2701786285664257, "reference_mean_power": 8.288074215773485e-05, "output_mean_power": 7.788176200156028e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 4.287875385442721e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2411523150580774e-05, "output_power": 1.951862932248998e-05, "reference_active": true, "delta_db": -0.6002205694378219}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.113877152498063e-06, "output_power": 2.7313237952262857e-06, "reference_active": true, "delta_db": -0.5692828665646616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.881539635700238e-07, "output_power": 8.211834267385808e-07, "reference_active": true, "delta_db": -0.803844411927067}], "active_band_count": 3, "active_band_mae_db": 0.6577826159765169}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 4.2997336393594785e-05, "log_rms_mae_db": 0.937879820844202, "rms_envelope_nrmse": 0.06931068391173188, "envelope_correlation": 0.9970624999205192, "rms_gain_error_db": -0.2701789634728499, "reference_mean_power": 8.343814618382176e-05, "output_mean_power": 7.840553996795805e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.9683446107037247e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.063395086340567e-05, "output_power": 1.8056882262403818e-05, "reference_active": true, "delta_db": -0.5793962570913844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4408129681959194e-06, "output_power": 1.2231739120652402e-06, "reference_active": true, "delta_db": -0.7111939890304831}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9896369199765786e-08, "output_power": 3.9390690441751186e-08, "reference_active": true, "delta_db": -1.0267535133511223}], "active_band_count": 3, "active_band_mae_db": 0.77244791982433}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 2.431018241105218e-05, "log_rms_mae_db": 1.7489359761182457, "rms_envelope_nrmse": 0.18193736504805022, "envelope_correlation": 0.9777062753914059, "rms_gain_error_db": -0.029555133918752356, "reference_mean_power": 1.5917473583289094e-05, "output_mean_power": 1.5809517818261977e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 8.500689890351627e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.331755201814068e-06, "output_power": 4.221494259945391e-06, "reference_active": true, "delta_db": -0.11197702502575431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3672340908387822e-06, "output_power": 1.3913483149839214e-06, "reference_active": true, "delta_db": 0.07592987779362097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.790743563061608e-07, "output_power": 3.586268173629695e-07, "reference_active": true, "delta_db": 1.0892283037262676}], "active_band_count": 3, "active_band_mae_db": 0.42571173551521424}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 1.765504974955162e-05, "log_rms_mae_db": 1.444371149842914, "rms_envelope_nrmse": 0.10736453327443772, "envelope_correlation": 0.9918522518777343, "rms_gain_error_db": -0.029552939855783902, "reference_mean_power": 1.6024474415231737e-05, "output_mean_power": 1.5915800987996148e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 7.931595726775402e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.29977249297526e-06, "output_power": 4.241293200007018e-06, "reference_active": true, "delta_db": -0.059471808397402307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.410418312822663e-07, "output_power": 7.745021201399105e-07, "reference_active": true, "delta_db": 0.1917988665983926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9106902570925117e-08, "output_power": 2.510383598250284e-08, "reference_active": true, "delta_db": 1.185497996860148}], "active_band_count": 3, "active_band_mae_db": 0.4789228906186476}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8847842614012157, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7a95fa79f3a3ddd8a1c5a2dc0522bb30272ff37598f93782c372db24bfafe17e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9883700046600365, "standard_deviation": 0.0013382387364849677, "minimum": 0.9863840470783709, "maximum": 0.9901088179646653, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4225373668071577, "standard_deviation": 0.025236927030449056, "minimum": 1.3688927178460037, "maximum": 1.4519396954524002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.76225550148073, "standard_deviation": 0.21144269670677887, "minimum": 0.36872953718054774, "maximum": 1.0289903754002556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1287967598125062, "standard_deviation": 0.0076871415694411875, "minimum": 0.11935977583406443, "maximum": 0.14104539447981113, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970446640085205, "standard_deviation": 0.0006125599935679532, "minimum": 0.9957373728002727, "maximum": 0.9976460578001471, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9184897471866187, "standard_deviation": 0.040040736787904076, "minimum": 0.8385234750047826, "maximum": 0.9616162384801362, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7673931222340091, "standard_deviation": 0.23057824408049377, "minimum": 0.33509394160301104, "maximum": 1.0437665631377169, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06789108567870285, "standard_deviation": 0.006764030765526771, "minimum": 0.06160118476981623, "maximum": 0.07961798968053206, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9730774631200012, "standard_deviation": 0.0041617014306469914, "minimum": 0.9652636050643206, "maximum": 0.9777062753914059, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7315787030294834, "standard_deviation": 0.032125504047832945, "minimum": 1.6835307217379154, "maximum": 1.7737301905313516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.622574758702247, "standard_deviation": 0.24794533087127885, "minimum": 0.32447635840436323, "maximum": 0.9042774644498919, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19992968387909105, "standard_deviation": 0.015539429749407866, "minimum": 0.18193736504805022, "maximum": 0.22931370289839575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906640071275854, "standard_deviation": 0.002505427790063836, "minimum": 0.9859978768164205, "maximum": 0.9925975563373237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4497231412759108, "standard_deviation": 0.041251391143190444, "minimum": 1.3881488875714652, "maximum": 1.5173392993246775, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6812914646230621, "standard_deviation": 0.23258803291697006, "minimum": 0.4347655172540435, "maximum": 1.0363252581510012, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11672101287647116, "standard_deviation": 0.01693887881387319, "minimum": 0.1033161809085638, "maximum": 0.15158557521902583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932676104674882, "standard_deviation": 0.0006231553469032008, "minimum": 0.9921975117139821, "maximum": 0.993968353487766, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.28399291041709, "standard_deviation": 0.05208608571691764, "minimum": 1.2241941217585497, "maximum": 1.365748701623651, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48493789552516103, "standard_deviation": 0.20556889123708744, "minimum": 0.19728517362534925, "maximum": 0.7387130525280506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09132908937160737, "standard_deviation": 0.0037935841544266325, "minimum": 0.0860743412672747, "maximum": 0.09770595146213339, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997717525618365, "standard_deviation": 0.00041533919241636857, "minimum": 0.9970647997129889, "maximum": 0.9983541841138738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8194290870430165, "standard_deviation": 0.03039127092319253, "minimum": 0.7889411831481642, "maximum": 0.8641026655849954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8260262436127768, "standard_deviation": 0.2955840097198026, "minimum": 0.4410351216941594, "maximum": 1.3221088884672376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.052781899273137665, "standard_deviation": 0.0051709602945041845, "minimum": 0.044108789423816096, "maximum": 0.05940148975786664, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1483713043813384, "standard_deviation": 0.03310197277240898, "minimum": 0.10050792878970453, "maximum": 0.19214175917025778, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.674210588193351, "standard_deviation": 0.006059101962413955, "minimum": 5.66776316725856, "maximum": 5.684613893753845, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3290968596648697, "standard_deviation": 0.007492321491885413, "minimum": 0.3211805893675085, "maximum": 0.34276076917195697, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08590576466159389, "standard_deviation": 0.04232107577585884, "minimum": -0.16002179116260257, "maximum": -0.035432764244371785, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946663263417903, "standard_deviation": 0.0004463178640875965, "minimum": 0.9938984612627217, "maximum": 0.9953028024164517, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1455698340533256, "standard_deviation": 0.032420785387151124, "minimum": 1.0975665262843155, "maximum": 1.1870167363870079, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2848260214958386, "standard_deviation": 0.12236573343690246, "minimum": 0.1597963230945055, "maximum": 0.4880030634550832, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07222309083254938, "standard_deviation": 0.0027063760688958677, "minimum": 0.0682454849300145, "maximum": 0.07675763802293883, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984824682167422, "standard_deviation": 0.0002933814693217469, "minimum": 0.9980131703020144, "maximum": 0.9989194143177356, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6157331139589584, "standard_deviation": 0.01910633870460074, "minimum": 0.5781455537996545, "maximum": 0.6376277347226648, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3928297297194493, "standard_deviation": 0.2320233346970224, "minimum": 0.06392118107305443, "maximum": 0.806470289953556, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0381755756654385, "standard_deviation": 0.0034331845493409947, "minimum": 0.031292806183364955, "maximum": 0.042564107578110605, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4992-41797-0000", "language": "en", "style": "normal", "speaker": "librispeech_4992", "source_utterance_id": "4992-41797-0000", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 87760, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4992-41797-0000.wav", "source_file_sha256": "1ce7a56ab790d2da03a45673e203d7684b9289bacdb91c2f0c3d38a47409492d", "decoded_pcm_sha256": "70049242cb761e6a91df355dd7a9dbdfb974c89071ee6b5dd83506a6ce8390e3", "original_source_sha256": "f7dbbd42eb06171dccf937d53a6c9144e51b88ea5b1b2ee69978900b87216fb5", "corpus_entry_id": "librispeech_4992-41797-0000"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.5616118907928467, "predicted_mos": 3.5616118907928467, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "9a82559561ea973f3dc9b0d5fd2463947b7f738edf3be97771c3e1a07abb4e55", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7662989090022165, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4992-41797-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9a82559561ea973f3dc9b0d5fd2463947b7f738edf3be97771c3e1a07abb4e55", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "b07633293b756654df0f8a661b93eaeafa136a1adfd7b75d4b0518561159c264"}, "rows": [{"case": "librispeech_4992-41797-0000", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.3656582832336426, "predicted_mos": 2.3656582832336426, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "306b62fa6567b25cfa8291738e3d0f9ef604c0789ba562217846c333fa798506", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4393541290010035, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4992-41797-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "306b62fa6567b25cfa8291738e3d0f9ef604c0789ba562217846c333fa798506", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8359216633920362, "全帯域_10ms_rms_envelope_nrmse": 0.04979669415040021, "全帯域_10ms_envelope_correlation": 0.9977058414399969, "全帯域_10ms_modulation_mae_db": 0.18570919992192425, "全帯域_40ms_log_rms_mae_db": 0.45091352315381816, "全帯域_40ms_rms_envelope_nrmse": 0.025260560670114642, "全帯域_40ms_envelope_correlation": 0.9993963186623355, "全帯域_40ms_modulation_mae_db": 0.24625709769024506, "300–1000 Hz_10ms_log_rms_mae_db": 0.9073094100238165, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061446358299589415, "300–1000 Hz_10ms_envelope_correlation": 0.9969288304046637, "300–1000 Hz_10ms_modulation_mae_db": 0.3379853446485975, "300–1000 Hz_40ms_log_rms_mae_db": 0.6700605641596509, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03327486924112801, "300–1000 Hz_40ms_envelope_correlation": 0.9990821107390186, "300–1000 Hz_40ms_modulation_mae_db": 0.43011660991836065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1996975890036519, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09368040487042885, "1000–3000 Hz_10ms_envelope_correlation": 0.9943369194800347, "1000–3000 Hz_10ms_modulation_mae_db": 0.3105870664899408, "1000–3000 Hz_40ms_log_rms_mae_db": 0.874622768771467, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05755367892534826, "1000–3000 Hz_40ms_envelope_correlation": 0.9982302458014773, "1000–3000 Hz_40ms_modulation_mae_db": 0.26667150131508893, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6094153851912598, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23330254164341038, "3000–6000 Hz_10ms_envelope_correlation": 0.9625419532493344, "3000–6000 Hz_10ms_modulation_mae_db": 1.2265665795595575, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4813969215181342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20139953761229595, "3000–6000 Hz_40ms_envelope_correlation": 0.9704788544092385, "3000–6000 Hz_40ms_modulation_mae_db": 1.4999361096285455, "mr_stft_db": 5.497830877877239, "band_mae_db": 0.15565472566667296, "rms_gain_error_db": -0.03902388534155727, "output_peak": 0.5020820260342704}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8937569371550189, "全帯域_10ms_rms_envelope_nrmse": 0.067137418397127, "全帯域_10ms_envelope_correlation": 0.9959380195684325, "全帯域_10ms_modulation_mae_db": 0.26367603090311814, "全帯域_40ms_log_rms_mae_db": 0.4919847861865074, "全帯域_40ms_rms_envelope_nrmse": 0.04077252373856719, "全帯域_40ms_envelope_correlation": 0.9985574891102188, "全帯域_40ms_modulation_mae_db": 0.2885085586600232, "300–1000 Hz_10ms_log_rms_mae_db": 0.9024877900197138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08498594354605096, "300–1000 Hz_10ms_envelope_correlation": 0.9943138337690163, "300–1000 Hz_10ms_modulation_mae_db": 0.15323631959270909, "300–1000 Hz_40ms_log_rms_mae_db": 0.6938727790880158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05579439066882097, "300–1000 Hz_40ms_envelope_correlation": 0.9976450615845248, "300–1000 Hz_40ms_modulation_mae_db": 0.1611438546476429, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2380565019613992, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08916843993890763, "1000–3000 Hz_10ms_envelope_correlation": 0.994598851897751, "1000–3000 Hz_10ms_modulation_mae_db": 0.2963789926151849, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9017602923955131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047715886369029674, "1000–3000 Hz_40ms_envelope_correlation": 0.9985858671856647, "1000–3000 Hz_40ms_modulation_mae_db": 0.22233152826940158, "3000–6000 Hz_10ms_log_rms_mae_db": 1.598005963209814, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22076255006188264, "3000–6000 Hz_10ms_envelope_correlation": 0.9640238760411557, "3000–6000 Hz_10ms_modulation_mae_db": 0.84315372952769, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4153245388972109, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18348955952473844, "3000–6000 Hz_40ms_envelope_correlation": 0.973819993440403, "3000–6000 Hz_40ms_modulation_mae_db": 1.0266708455063622, "mr_stft_db": 5.507189402883601, "band_mae_db": 0.19399214571147075, "rms_gain_error_db": -0.12141367396083978, "output_peak": 0.46452734567531906}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8863500933053141, "全帯域_10ms_rms_envelope_nrmse": 0.05409270480274795, "全帯域_10ms_envelope_correlation": 0.997320725043444, "全帯域_10ms_modulation_mae_db": 0.5437233608362545, "全帯域_40ms_log_rms_mae_db": 0.5066782487674553, "全帯域_40ms_rms_envelope_nrmse": 0.028955075512519694, "全帯域_40ms_envelope_correlation": 0.9992398221757637, "全帯域_40ms_modulation_mae_db": 0.8196915103782009, "300–1000 Hz_10ms_log_rms_mae_db": 0.904125013897934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0677464569601315, "300–1000 Hz_10ms_envelope_correlation": 0.9962934794966254, "300–1000 Hz_10ms_modulation_mae_db": 0.7598738895735412, "300–1000 Hz_40ms_log_rms_mae_db": 0.6599731117714925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03719312385583142, "300–1000 Hz_40ms_envelope_correlation": 0.9988900665967466, "300–1000 Hz_40ms_modulation_mae_db": 1.1423845984345082, "1000–3000 Hz_10ms_log_rms_mae_db": 1.188143153858721, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08403161837742727, "1000–3000 Hz_10ms_envelope_correlation": 0.9951890051942768, "1000–3000 Hz_10ms_modulation_mae_db": 0.3445323513996285, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8185740418659146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.040639760167459765, "1000–3000 Hz_40ms_envelope_correlation": 0.9990086520816985, "1000–3000 Hz_40ms_modulation_mae_db": 0.26339572919391846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6695936100334396, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23237205239862457, "3000–6000 Hz_10ms_envelope_correlation": 0.9597947673039933, "3000–6000 Hz_10ms_modulation_mae_db": 0.3400520648403676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5012663731253193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18707721243687175, "3000–6000 Hz_40ms_envelope_correlation": 0.9726194825498689, "3000–6000 Hz_40ms_modulation_mae_db": 0.45213524497699237, "mr_stft_db": 5.505252925467943, "band_mae_db": 0.16078317979843074, "rms_gain_error_db": -0.06873221349171238, "output_peak": 0.5118940214127017}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8335025505049722, "全帯域_10ms_rms_envelope_nrmse": 0.05327594798181663, "全帯域_10ms_envelope_correlation": 0.9975769015116741, "全帯域_10ms_modulation_mae_db": 0.3872556057279312, "全帯域_40ms_log_rms_mae_db": 0.4528247181779269, "全帯域_40ms_rms_envelope_nrmse": 0.0304221285769029, "全帯域_40ms_envelope_correlation": 0.9993536316267349, "全帯域_40ms_modulation_mae_db": 0.4523286570653749, "300–1000 Hz_10ms_log_rms_mae_db": 0.902109244043123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0687839501436078, "300–1000 Hz_10ms_envelope_correlation": 0.9964547245431361, "300–1000 Hz_10ms_modulation_mae_db": 0.5643294948485136, "300–1000 Hz_40ms_log_rms_mae_db": 0.6912767562887124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043161679643907956, "300–1000 Hz_40ms_envelope_correlation": 0.998796977854258, "300–1000 Hz_40ms_modulation_mae_db": 0.5981159157110516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2128105644059053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08991391152861536, "1000–3000 Hz_10ms_envelope_correlation": 0.994575092359826, "1000–3000 Hz_10ms_modulation_mae_db": 0.5200644640325379, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8834037977772692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05125870664790054, "1000–3000 Hz_40ms_envelope_correlation": 0.9984336471172062, "1000–3000 Hz_40ms_modulation_mae_db": 0.7733111818488196, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5845337180999652, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23576593276891317, "3000–6000 Hz_10ms_envelope_correlation": 0.9574315195079998, "3000–6000 Hz_10ms_modulation_mae_db": 0.46147112863469086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4520085504123614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19587088158703891, "3000–6000 Hz_40ms_envelope_correlation": 0.9698345180698302, "3000–6000 Hz_40ms_modulation_mae_db": 0.6064673414227281, "mr_stft_db": 5.519958501449323, "band_mae_db": 0.18071549120755734, "rms_gain_error_db": -0.14298112288874867, "output_peak": 0.48853599445321816}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8204720060081304, "全帯域_10ms_rms_envelope_nrmse": 0.059505137390352576, "全帯域_10ms_envelope_correlation": 0.9967652366357302, "全帯域_10ms_modulation_mae_db": 0.4434375165605701, "全帯域_40ms_log_rms_mae_db": 0.45440426608869106, "全帯域_40ms_rms_envelope_nrmse": 0.03188945569826045, "全帯域_40ms_envelope_correlation": 0.9990912847141749, "全帯域_40ms_modulation_mae_db": 0.5754115287535537, "300–1000 Hz_10ms_log_rms_mae_db": 0.9195128453724634, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07695924130103984, "300–1000 Hz_10ms_envelope_correlation": 0.9952330756531563, "300–1000 Hz_10ms_modulation_mae_db": 0.5071941683146287, "300–1000 Hz_40ms_log_rms_mae_db": 0.7130383994932864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04361287446872797, "300–1000 Hz_40ms_envelope_correlation": 0.9984978249441818, "300–1000 Hz_40ms_modulation_mae_db": 0.7490211912454874, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2036804921733397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08656005552797048, "1000–3000 Hz_10ms_envelope_correlation": 0.9949632488093542, "1000–3000 Hz_10ms_modulation_mae_db": 0.3404093193198294, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8262716843746675, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04925773165846369, "1000–3000 Hz_40ms_envelope_correlation": 0.9985302472924036, "1000–3000 Hz_40ms_modulation_mae_db": 0.2931882542525998, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6119999116770043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22159781377253635, "3000–6000 Hz_10ms_envelope_correlation": 0.9633998003346885, "3000–6000 Hz_10ms_modulation_mae_db": 0.6755595793150976, "3000–6000 Hz_40ms_log_rms_mae_db": 1.484024788516433, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.188105922049725, "3000–6000 Hz_40ms_envelope_correlation": 0.9721551658146232, "3000–6000 Hz_40ms_modulation_mae_db": 0.9102966790915575, "mr_stft_db": 5.504155739678247, "band_mae_db": 0.23140393873248488, "rms_gain_error_db": -0.08342632429521107, "output_peak": 0.5218041278121403}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8230250691022836, "全帯域_10ms_rms_envelope_nrmse": 0.05983436585219612, "全帯域_10ms_envelope_correlation": 0.9968651125302561, "全帯域_10ms_modulation_mae_db": 0.7906389345891505, "全帯域_40ms_log_rms_mae_db": 0.45408471675039785, "全帯域_40ms_rms_envelope_nrmse": 0.03526842555246606, "全帯域_40ms_envelope_correlation": 0.9990270218829612, "全帯域_40ms_modulation_mae_db": 0.7473259239585496, "300–1000 Hz_10ms_log_rms_mae_db": 0.9430231921062988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07587642665559546, "300–1000 Hz_10ms_envelope_correlation": 0.9956689352023447, "300–1000 Hz_10ms_modulation_mae_db": 0.8473760881640486, "300–1000 Hz_40ms_log_rms_mae_db": 0.6839829398500977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04985592428797992, "300–1000 Hz_40ms_envelope_correlation": 0.9983353580203775, "300–1000 Hz_40ms_modulation_mae_db": 0.970018076674874, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1672316205970716, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08919567301860044, "1000–3000 Hz_10ms_envelope_correlation": 0.9944990580451958, "1000–3000 Hz_10ms_modulation_mae_db": 0.1979668907203128, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8078332592728127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04440293705959739, "1000–3000 Hz_40ms_envelope_correlation": 0.9986913954908406, "1000–3000 Hz_40ms_modulation_mae_db": 0.3386372020387149, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6020429853700455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2287760435034224, "3000–6000 Hz_10ms_envelope_correlation": 0.9616331064642131, "3000–6000 Hz_10ms_modulation_mae_db": 0.7839021933474092, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4942732660590703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19039249040771536, "3000–6000 Hz_40ms_envelope_correlation": 0.9719893312732261, "3000–6000 Hz_40ms_modulation_mae_db": 1.1672101153156502, "mr_stft_db": 5.524676551763972, "band_mae_db": 0.21291839302026316, "rms_gain_error_db": -0.14050172632595448, "output_peak": 0.47988304999178}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8646742745254475, "全帯域_10ms_rms_envelope_nrmse": 0.06540866558104542, "全帯域_10ms_envelope_correlation": 0.9961878081937205, "全帯域_10ms_modulation_mae_db": 0.6893891772175268, "全帯域_40ms_log_rms_mae_db": 0.4979098907330335, "全帯域_40ms_rms_envelope_nrmse": 0.038648838926926485, "全帯域_40ms_envelope_correlation": 0.998760887303353, "全帯域_40ms_modulation_mae_db": 0.7633768308012904, "300–1000 Hz_10ms_log_rms_mae_db": 0.8906571371220854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07849823097147773, "300–1000 Hz_10ms_envelope_correlation": 0.9952027315490076, "300–1000 Hz_10ms_modulation_mae_db": 0.9851974056794326, "300–1000 Hz_40ms_log_rms_mae_db": 0.6490384042375477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04875547260517192, "300–1000 Hz_40ms_envelope_correlation": 0.9982824426355135, "300–1000 Hz_40ms_modulation_mae_db": 1.040199303905095, "1000–3000 Hz_10ms_log_rms_mae_db": 1.198987861523814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09291616720552914, "1000–3000 Hz_10ms_envelope_correlation": 0.9941512710285827, "1000–3000 Hz_10ms_modulation_mae_db": 0.34193880098204615, "1000–3000 Hz_40ms_log_rms_mae_db": 0.876751204072625, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04638766218269725, "1000–3000 Hz_40ms_envelope_correlation": 0.9987421019720637, "1000–3000 Hz_40ms_modulation_mae_db": 0.43439654976331044, "3000–6000 Hz_10ms_log_rms_mae_db": 1.63043363322359, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2253035366523308, "3000–6000 Hz_10ms_envelope_correlation": 0.9623213632956505, "3000–6000 Hz_10ms_modulation_mae_db": 0.7418697751941691, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4848984676272303, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1873512388809644, "3000–6000 Hz_40ms_envelope_correlation": 0.972475577469727, "3000–6000 Hz_40ms_modulation_mae_db": 0.9200726880526015, "mr_stft_db": 5.508960170968451, "band_mae_db": 0.13034799200521432, "rms_gain_error_db": -0.13382230147654084, "output_peak": 0.46030313232757364}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8863828609968208, "全帯域_10ms_rms_envelope_nrmse": 0.056554585628027225, "全帯域_10ms_envelope_correlation": 0.9970466873929859, "全帯域_10ms_modulation_mae_db": 0.33032462375953814, "全帯域_40ms_log_rms_mae_db": 0.4676941230769086, "全帯域_40ms_rms_envelope_nrmse": 0.033385363321264634, "全帯域_40ms_envelope_correlation": 0.9989232713154323, "全帯域_40ms_modulation_mae_db": 0.33842851052633033, "300–1000 Hz_10ms_log_rms_mae_db": 0.875448128540126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07218245836730501, "300–1000 Hz_10ms_envelope_correlation": 0.9957590622965681, "300–1000 Hz_10ms_modulation_mae_db": 0.3769507340641555, "300–1000 Hz_40ms_log_rms_mae_db": 0.6371531160778764, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044940896687160094, "300–1000 Hz_40ms_envelope_correlation": 0.9982767833880377, "300–1000 Hz_40ms_modulation_mae_db": 0.2773104720253071, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1448283353864914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0832797156823121, "1000–3000 Hz_10ms_envelope_correlation": 0.9953411135443141, "1000–3000 Hz_10ms_modulation_mae_db": 0.5229245229465341, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8198097213279494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04457484980576928, "1000–3000 Hz_40ms_envelope_correlation": 0.998834017362511, "1000–3000 Hz_40ms_modulation_mae_db": 0.6635787015184841, "3000–6000 Hz_10ms_log_rms_mae_db": 1.675020175615174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2502128487659414, "3000–6000 Hz_10ms_envelope_correlation": 0.9519798096079146, "3000–6000 Hz_10ms_modulation_mae_db": 0.9231008712250016, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5259777505581984, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20990477051604264, "3000–6000 Hz_40ms_envelope_correlation": 0.9646149053468281, "3000–6000 Hz_40ms_modulation_mae_db": 1.0569252324853344, "mr_stft_db": 5.511125250887191, "band_mae_db": 0.1385973630974743, "rms_gain_error_db": 0.0023513414332034773, "output_peak": 0.47791151194797693}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87760, "output_samples": 87760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.175039740326897, "spectral_convergence": 0.16327439822397252, "spectral_convergence_status": "ok", "amplitude_floor": 1.7869460513029852e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.281551236256423, "spectral_convergence": 0.16252305682088022, "spectral_convergence_status": "ok", "amplitude_floor": 1.6383670687778497e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0369016570484, "spectral_convergence": 0.16890481241982924, "spectral_convergence_status": "ok", "amplitude_floor": 9.631824221799184e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.013402842456697e-05, "output_power": 9.792142293654634e-05, "delta_db": 0.3598893402541722, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00191544195180521, "output_power": 0.0019024036767776161, "delta_db": -0.02966318422705618, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015517446606579416, "output_power": 0.0015553730880964645, "delta_db": 0.010143205750617801, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955960266218775, "output_power": 0.0004680120516160857, "delta_db": -0.2487077923244705, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000158155871967105, "output_power": 0.00014971058692841294, "delta_db": -0.23832807933711866, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.514760940569477e-06, "output_power": 9.41191954000947e-06, "delta_db": -0.04719675210660261, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15565472566667296, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03902388534155727, "pitch": null, "output_rms_dbfs": -23.785295007043416, "output_peak": 0.5020820260342704, "reference_rms_dbfs": -23.74627112170186, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.497830877877239}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002415922668587691, "log_rms_mae_db": 0.8359216633920362, "rms_envelope_nrmse": 0.04979669415040021, "envelope_correlation": 0.9977058414399969, "rms_gain_error_db": -0.03902393777604282, "reference_mean_power": 0.004224437839878649, "output_mean_power": 0.004186648790455308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.932071872304039e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008459874532766979, "output_power": 0.0008395211660196441, "reference_active": true, "delta_db": -0.0333227207440359}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.267978041368795e-05, "output_power": 9.008233702346599e-05, "reference_active": true, "delta_db": -0.12345351505694646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0604587197812338e-05, "output_power": 1.8790086599996842e-05, "reference_active": true, "delta_db": -0.4003513639647903}], "active_band_count": 3, "active_band_mae_db": 0.18570919992192425}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00018058591873130477, "log_rms_mae_db": 0.45091352315381816, "rms_envelope_nrmse": 0.025260560670114642, "envelope_correlation": 0.9993963186623355, "rms_gain_error_db": -0.03902430552506363, "reference_mean_power": 0.004247690453399677, "output_mean_power": 0.004209693044885298}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8509413564750568e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008019972220007684, "output_power": 0.0007959752936537295, "reference_active": true, "delta_db": -0.03273276112475453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.130657727931199e-05, "output_power": 4.0090128473110816e-05, "reference_active": true, "delta_db": -0.12981762224134222}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0523532353898244e-06, "output_power": 9.215935989442178e-07, "reference_active": true, "delta_db": -0.5762209097046385}], "active_band_count": 3, "active_band_mae_db": 0.24625709769024506}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002374784601360974, "log_rms_mae_db": 0.9073094100238165, "rms_envelope_nrmse": 0.061446358299589415, "envelope_correlation": 0.9969288304046637, "rms_gain_error_db": -0.05741002138071196, "reference_mean_power": 0.0028785349502786395, "output_mean_power": 0.0028407335780675884}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.6369199856429154e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008090138264371756, "output_power": 0.0008043488203760437, "reference_active": true, "delta_db": -0.02511514742971984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000105081165545653, "output_power": 0.00010329222873257405, "reference_active": true, "delta_db": -0.0745723299335311}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7476383653056024e-05, "output_power": 1.4158743508446207e-05, "reference_active": true, "delta_db": -0.9142685565825415}], "active_band_count": 3, "active_band_mae_db": 0.3379853446485975}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.0001750973468816212, "log_rms_mae_db": 0.6700605641596509, "rms_envelope_nrmse": 0.03327486924112801, "envelope_correlation": 0.9990821107390186, "rms_gain_error_db": -0.057410029025251524, "reference_mean_power": 0.002894380018498304, "output_mean_power": 0.002856370561339651}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.567165950582092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007701013013376472, "output_power": 0.0007661727425437102, "reference_active": true, "delta_db": -0.022211598664221142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.715572645231493e-05, "output_power": 4.570116369746094e-05, "reference_active": true, "delta_db": -0.13607180982621767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1197000044443696e-06, "output_power": 8.627700045305811e-07, "reference_active": true, "delta_db": -1.1320664212646432}], "active_band_count": 3, "active_band_mae_db": 0.43011660991836065}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.00012170766433419248, "log_rms_mae_db": 1.1996975890036519, "rms_envelope_nrmse": 0.09368040487042885, "envelope_correlation": 0.9943369194800347, "rms_gain_error_db": -0.2671534336635239, "reference_mean_power": 0.0006352125631157451, "output_mean_power": 0.0005973154325285966}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 2.616269223692633e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010398636594123316, "output_power": 9.654643416619499e-05, "reference_active": true, "delta_db": -0.32240162750787693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0847306945521593e-05, "output_power": 9.738637996599453e-06, "reference_active": true, "delta_db": -0.4682370693767257}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2384759945240008e-06, "output_power": 3.3454375900027736e-06, "reference_active": true, "delta_db": 0.14112250258521958}], "active_band_count": 3, "active_band_mae_db": 0.3105870664899408}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00010363700534457545, "log_rms_mae_db": 0.874622768771467, "rms_envelope_nrmse": 0.05755367892534826, "envelope_correlation": 0.9982302458014773, "rms_gain_error_db": -0.2671535023077274, "reference_mean_power": 0.0006387091058518984, "output_mean_power": 0.0006006033601105537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.5308592276752824e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.945508810428335e-05, "output_power": 9.192587369349285e-05, "reference_active": true, "delta_db": -0.3418924021885411}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.423063898822258e-06, "output_power": 5.219774742999275e-06, "reference_active": true, "delta_db": -0.16592960040055643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8432082747665845e-07, "output_power": 1.7232772763048725e-07, "reference_active": true, "delta_db": -0.2921925013561693}], "active_band_count": 3, "active_band_mae_db": 0.26667150131508893}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 2.9435709173680706e-05, "log_rms_mae_db": 1.6094153851912598, "rms_envelope_nrmse": 0.23330254164341038, "envelope_correlation": 0.9625419532493344, "rms_gain_error_db": 0.3524027015945341, "reference_mean_power": 1.4688872454843858e-05, "output_mean_power": 1.5930475168610007e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 9.038086104701173e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4018145980274633e-06, "output_power": 3.755435024988079e-06, "reference_active": true, "delta_db": 0.42949611914999464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2285201909756879e-06, "output_power": 1.5684397834294348e-06, "reference_active": true, "delta_db": 1.0608555104311488}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.73608399821043e-07, "output_power": 4.5296457353851593e-07, "reference_active": true, "delta_db": 2.189348109097529}], "active_band_count": 3, "active_band_mae_db": 1.2265665795595575}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 2.1616481165467548e-05, "log_rms_mae_db": 1.4813969215181342, "rms_envelope_nrmse": 0.20139953761229595, "envelope_correlation": 0.9704788544092385, "rms_gain_error_db": 0.3524028148401025, "reference_mean_power": 1.4769720024935061e-05, "output_mean_power": 1.601815693908418e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.51604517212682e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5641156211998326e-06, "output_power": 4.130796405958163e-06, "reference_active": true, "delta_db": 0.6408200622323569}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790775026435671e-07, "output_power": 7.644454650763155e-07, "reference_active": true, "delta_db": 1.206098158610061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8657206259627155e-08, "output_power": 3.436652525964017e-08, "reference_active": true, "delta_db": 2.6528901080432195}], "active_band_count": 3, "active_band_mae_db": 1.4999361096285455}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9080659362371144, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d3d62520a4fa274610ee7acd50cf971c6326d9211bc0ca0e146749b9d4b6b1ea", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.994706820044917, "standard_deviation": 0.0004171710787591704, "minimum": 0.9941512710285827, "maximum": 0.9953411135443141, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1941795148637993, "standard_deviation": 0.028320385108487684, "minimum": 1.1448283353864914, "maximum": 1.2380565019613992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.35935030106325183, "standard_deviation": 0.11074739911037348, "minimum": 0.1979668907203128, "maximum": 0.5229245229465341, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08859324826872392, "standard_deviation": 0.003783057558572028, "minimum": 0.0832797156823121, "maximum": 0.09368040487042885, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9986320217879832, "standard_deviation": 0.0002424753956282172, "minimum": 0.9982302458014773, "maximum": 0.9990086520816985, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8511283462322773, "standard_deviation": 0.03654112100726584, "minimum": 0.8078332592728127, "maximum": 0.9017602923955131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4069388310250422, "standard_deviation": 0.2045239620015096, "minimum": 0.22233152826940158, "maximum": 0.7733111818488196, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04772390160203323, "standard_deviation": 0.005132362254538968, "minimum": 0.040639760167459765, "maximum": 0.05755367892534826, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9603907744756188, "standard_deviation": 0.004000152500703994, "minimum": 0.9519798096079146, "maximum": 0.9640238760411557, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6226306728025366, "standard_deviation": 0.033338797073920054, "minimum": 1.5845337180999652, "maximum": 1.675020175615174, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.749459490205498, "standard_deviation": 0.27355280272320737, "minimum": 0.3400520648403676, "maximum": 1.2265665795595575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2310116649458827, "standard_deviation": 0.009484069156512802, "minimum": 0.22076255006188264, "maximum": 0.2502128487659414, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9709984785467181, "standard_deviation": 0.002865184015471523, "minimum": 0.9646149053468281, "maximum": 0.973819993440403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4798963320892446, "standard_deviation": 0.03333473978622007, "minimum": 1.4153245388972109, "maximum": 1.5259777505581984, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9549642820599715, "standard_deviation": 0.32405471969261096, "minimum": 0.45213524497699237, "maximum": 1.4999361096285455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19294895162692405, "standard_deviation": 0.008886041299899854, "minimum": 0.18348955952473844, "maximum": 0.20990477051604264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9957318341143149, "standard_deviation": 0.0008293670891672025, "minimum": 0.9943138337690163, "maximum": 0.9969288304046637, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.905584095140695, "standard_deviation": 0.019833067439182853, "minimum": 0.875448128540126, "maximum": 0.9430231921062988, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5665179306107033, "standard_deviation": 0.28133584080679014, "minimum": 0.15323631959270909, "maximum": 0.9851974056794326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07330988328059972, "standard_deviation": 0.007334891069742782, "minimum": 0.061446358299589415, "maximum": 0.08498594354605096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984758282203323, "standard_deviation": 0.0004519346947315804, "minimum": 0.9976450615845248, "maximum": 0.9990821107390186, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.674799508870835, "standard_deviation": 0.025368716759049587, "minimum": 0.6371531160778764, "maximum": 0.7130383994932864, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6710387528202908, "standard_deviation": 0.3649446730754615, "minimum": 0.1611438546476429, "maximum": 1.1423845984345082, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04457365393234103, "standard_deviation": 0.007136238664634846, "minimum": 0.03327486924112801, "maximum": 0.05579439066882097, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17555165365494604, "standard_deviation": 0.035668021101952475, "minimum": 0.13034799200521432, "maximum": 0.23140393873248488, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.509893677621996, "standard_deviation": 0.008697347755797107, "minimum": 5.497830877877239, "maximum": 5.524676551763972, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.48836765120687253, "standard_deviation": 0.022028867985104738, "minimum": 0.46030313232757364, "maximum": 0.5218041278121403, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09094373829342012, "standard_deviation": 0.053292157172324266, "minimum": -0.14298112288874867, "maximum": 0.0023513414332034773, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.99692579153953, "standard_deviation": 0.0006271002457820094, "minimum": 0.9959380195684325, "maximum": 0.9977058414399969, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.855510681873753, "standard_deviation": 0.03071067648125558, "minimum": 0.8204720060081304, "maximum": 0.8937569371550189, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4542693061895017, "standard_deviation": 0.20868591705387718, "minimum": 0.18570919992192425, "maximum": 0.7906389345891505, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05820068997296414, "standard_deviation": 0.0059839778739949855, "minimum": 0.04979669415040021, "maximum": 0.067137418397127, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990437158488717, "standard_deviation": 0.00029088314381721055, "minimum": 0.9985574891102188, "maximum": 0.9993963186623355, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47206178411684235, "standard_deviation": 0.023098247480482382, "minimum": 0.45091352315381816, "maximum": 0.5066782487674553, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5289160772291961, "standard_deviation": 0.22966743202904003, "minimum": 0.24625709769024506, "maximum": 0.8196915103782009, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03307529649962776, "standard_deviation": 0.005094948118803552, "minimum": 0.025260560670114642, "maximum": 0.04077252373856719, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41797-0000", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.0868520736694336, "predicted_mos": 2.0868520736694336, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "663944d6fb8b8e37e0d97a348c8dbbd71c1d34430698ba49b97d60e0285fdfb0", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.17776960041374, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8503538201867293, "全帯域_10ms_rms_envelope_nrmse": 0.05087240624945553, "全帯域_10ms_envelope_correlation": 0.9976027889248533, "全帯域_10ms_modulation_mae_db": 0.12390601858648496, "全帯域_40ms_log_rms_mae_db": 0.4450996302008955, "全帯域_40ms_rms_envelope_nrmse": 0.026180531864028227, "全帯域_40ms_envelope_correlation": 0.9993475243347913, "全帯域_40ms_modulation_mae_db": 0.15898739739074785, "300–1000 Hz_10ms_log_rms_mae_db": 0.8787768362290846, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06115970571380142, "300–1000 Hz_10ms_envelope_correlation": 0.9969485520027105, "300–1000 Hz_10ms_modulation_mae_db": 0.2839754042336983, "300–1000 Hz_40ms_log_rms_mae_db": 0.6232140840626511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0319870155147784, "300–1000 Hz_40ms_envelope_correlation": 0.9991428305992791, "300–1000 Hz_40ms_modulation_mae_db": 0.3392919351190113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2071235271794167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09539069228480658, "1000–3000 Hz_10ms_envelope_correlation": 0.9941737266649306, "1000–3000 Hz_10ms_modulation_mae_db": 0.38894789202104874, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8693283153419263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058615581120197144, "1000–3000 Hz_40ms_envelope_correlation": 0.998217966771417, "1000–3000 Hz_40ms_modulation_mae_db": 0.31981360911541157, "3000–6000 Hz_10ms_log_rms_mae_db": 1.579949002894304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23327253277520965, "3000–6000 Hz_10ms_envelope_correlation": 0.9627189708263825, "3000–6000 Hz_10ms_modulation_mae_db": 1.2584575533365712, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4767717447753146, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20389299570743089, "3000–6000 Hz_40ms_envelope_correlation": 0.9697485943473653, "3000–6000 Hz_40ms_modulation_mae_db": 1.5510186955222611, "mr_stft_db": 5.44339048079357, "band_mae_db": 0.15659201500677924, "rms_gain_error_db": -0.03599264244857402, "output_peak": 0.49788403591468944}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9026535059168875, "全帯域_10ms_rms_envelope_nrmse": 0.06519147371573056, "全帯域_10ms_envelope_correlation": 0.9961830726208978, "全帯域_10ms_modulation_mae_db": 0.2707101169390544, "全帯域_40ms_log_rms_mae_db": 0.5057001254646323, "全帯域_40ms_rms_envelope_nrmse": 0.03875681951572807, "全帯域_40ms_envelope_correlation": 0.9987167383639572, "全帯域_40ms_modulation_mae_db": 0.2974472795613419, "300–1000 Hz_10ms_log_rms_mae_db": 0.9009295821969626, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08427142224581242, "300–1000 Hz_10ms_envelope_correlation": 0.9944230686657818, "300–1000 Hz_10ms_modulation_mae_db": 0.18223722731565176, "300–1000 Hz_40ms_log_rms_mae_db": 0.6435410968428924, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05465002918454301, "300–1000 Hz_40ms_envelope_correlation": 0.9977631159441114, "300–1000 Hz_40ms_modulation_mae_db": 0.20040804736375586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2204331205317256, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08888583937352613, "1000–3000 Hz_10ms_envelope_correlation": 0.9946276961867088, "1000–3000 Hz_10ms_modulation_mae_db": 0.33615765489293487, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8828604991972194, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04886693528521996, "1000–3000 Hz_40ms_envelope_correlation": 0.9984874661881579, "1000–3000 Hz_40ms_modulation_mae_db": 0.26868757386562486, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5351908195851094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2184540394107831, "3000–6000 Hz_10ms_envelope_correlation": 0.9644057215699028, "3000–6000 Hz_10ms_modulation_mae_db": 0.765579844061904, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3930334966730509, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1829135494283235, "3000–6000 Hz_40ms_envelope_correlation": 0.9735853494975792, "3000–6000 Hz_40ms_modulation_mae_db": 0.962117595998214, "mr_stft_db": 5.447431173756219, "band_mae_db": 0.1843215292792528, "rms_gain_error_db": -0.12309886596521585, "output_peak": 0.4616555694652295}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9004082367535649, "全帯域_10ms_rms_envelope_nrmse": 0.05393593329623099, "全帯域_10ms_envelope_correlation": 0.997337552138411, "全帯域_10ms_modulation_mae_db": 0.4962557639784751, "全帯域_40ms_log_rms_mae_db": 0.4913924519361104, "全帯域_40ms_rms_envelope_nrmse": 0.02851855644887662, "全帯域_40ms_envelope_correlation": 0.9992659266442178, "全帯域_40ms_modulation_mae_db": 0.7918457111730959, "300–1000 Hz_10ms_log_rms_mae_db": 0.9158652117227326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06779934356256423, "300–1000 Hz_10ms_envelope_correlation": 0.996286467550029, "300–1000 Hz_10ms_modulation_mae_db": 0.704909338591023, "300–1000 Hz_40ms_log_rms_mae_db": 0.6302711483885399, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03751828239862174, "300–1000 Hz_40ms_envelope_correlation": 0.9988679388021919, "300–1000 Hz_40ms_modulation_mae_db": 1.1028910606767812, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1851552410101533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08345423282724217, "1000–3000 Hz_10ms_envelope_correlation": 0.9952425523648186, "1000–3000 Hz_10ms_modulation_mae_db": 0.38999074875442835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7905676303260044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.040244817890030815, "1000–3000 Hz_40ms_envelope_correlation": 0.99900714934695, "1000–3000 Hz_40ms_modulation_mae_db": 0.2982134353371457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5956980637752853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2295490601184851, "3000–6000 Hz_10ms_envelope_correlation": 0.960358038521016, "3000–6000 Hz_10ms_modulation_mae_db": 0.29732971153349735, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4656509707759158, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18761388624911224, "3000–6000 Hz_40ms_envelope_correlation": 0.9720710335260027, "3000–6000 Hz_40ms_modulation_mae_db": 0.4435654347378925, "mr_stft_db": 5.451762712212065, "band_mae_db": 0.13890343937956554, "rms_gain_error_db": -0.06940799168316847, "output_peak": 0.5110161378984202}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8523048022456876, "全帯域_10ms_rms_envelope_nrmse": 0.05392972414415139, "全帯域_10ms_envelope_correlation": 0.9975153322590087, "全帯域_10ms_modulation_mae_db": 0.39443159887910334, "全帯域_40ms_log_rms_mae_db": 0.46955235472089385, "全帯域_40ms_rms_envelope_nrmse": 0.030888873368330412, "全帯域_40ms_envelope_correlation": 0.9993303768051714, "全帯域_40ms_modulation_mae_db": 0.440305521274363, "300–1000 Hz_10ms_log_rms_mae_db": 0.8850815033139809, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06874753493690967, "300–1000 Hz_10ms_envelope_correlation": 0.9964787562577989, "300–1000 Hz_10ms_modulation_mae_db": 0.5870168952626165, "300–1000 Hz_40ms_log_rms_mae_db": 0.6441933249977381, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04291234516469686, "300–1000 Hz_40ms_envelope_correlation": 0.9988376067160067, "300–1000 Hz_40ms_modulation_mae_db": 0.591337284097104, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1972742460610069, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08909037050742306, "1000–3000 Hz_10ms_envelope_correlation": 0.9946671685694446, "1000–3000 Hz_10ms_modulation_mae_db": 0.5917262314237554, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8550057836425496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.051907818981869365, "1000–3000 Hz_40ms_envelope_correlation": 0.9983658455878233, "1000–3000 Hz_40ms_modulation_mae_db": 0.8280920136698255, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5487370769091773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23363065705132133, "3000–6000 Hz_10ms_envelope_correlation": 0.9581964176018472, "3000–6000 Hz_10ms_modulation_mae_db": 0.417792321295738, "3000–6000 Hz_40ms_log_rms_mae_db": 1.435302237783333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19727129986983127, "3000–6000 Hz_40ms_envelope_correlation": 0.9693214329691757, "3000–6000 Hz_40ms_modulation_mae_db": 0.6289695281550073, "mr_stft_db": 5.458210081634313, "band_mae_db": 0.16444141964668021, "rms_gain_error_db": -0.1436662130944173, "output_peak": 0.49436054164958565}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8412786844549898, "全帯域_10ms_rms_envelope_nrmse": 0.05975348916416084, "全帯域_10ms_envelope_correlation": 0.9967646035919951, "全帯域_10ms_modulation_mae_db": 0.4991200838733072, "全帯域_40ms_log_rms_mae_db": 0.46550114789740493, "全帯域_40ms_rms_envelope_nrmse": 0.033695351751113295, "全帯域_40ms_envelope_correlation": 0.9990074946298279, "全帯域_40ms_modulation_mae_db": 0.6016912479041431, "300–1000 Hz_10ms_log_rms_mae_db": 0.89130693651212, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07690766967127227, "300–1000 Hz_10ms_envelope_correlation": 0.9952808213890713, "300–1000 Hz_10ms_modulation_mae_db": 0.6162421141242714, "300–1000 Hz_40ms_log_rms_mae_db": 0.6509159263680337, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04522675800937467, "300–1000 Hz_40ms_envelope_correlation": 0.9984224235050854, "300–1000 Hz_40ms_modulation_mae_db": 0.8527110803770022, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1903388420702992, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08681732597499774, "1000–3000 Hz_10ms_envelope_correlation": 0.9949517354120281, "1000–3000 Hz_10ms_modulation_mae_db": 0.19716529906859878, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8168109282369092, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049988223887165326, "1000–3000 Hz_40ms_envelope_correlation": 0.998501405728377, "1000–3000 Hz_40ms_modulation_mae_db": 0.2511720671580837, "3000–6000 Hz_10ms_log_rms_mae_db": 1.535383997242272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21894341834579115, "3000–6000 Hz_10ms_envelope_correlation": 0.9639844307287327, "3000–6000 Hz_10ms_modulation_mae_db": 0.6640180524501379, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4391214140929067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18763554544277852, "3000–6000 Hz_40ms_envelope_correlation": 0.971959368527767, "3000–6000 Hz_40ms_modulation_mae_db": 0.7887843600409997, "mr_stft_db": 5.443856758560055, "band_mae_db": 0.2154298319310112, "rms_gain_error_db": -0.09807784478136905, "output_peak": 0.5161436937730849}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8307580192611815, "全帯域_10ms_rms_envelope_nrmse": 0.059336192873386304, "全帯域_10ms_envelope_correlation": 0.9969273315808229, "全帯域_10ms_modulation_mae_db": 0.7271194453878559, "全帯域_40ms_log_rms_mae_db": 0.4451994997959518, "全帯域_40ms_rms_envelope_nrmse": 0.03578592283178194, "全帯域_40ms_envelope_correlation": 0.9989977790234179, "全帯域_40ms_modulation_mae_db": 0.6752672197825466, "300–1000 Hz_10ms_log_rms_mae_db": 0.9299408102433908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07421153416667327, "300–1000 Hz_10ms_envelope_correlation": 0.9958808340283593, "300–1000 Hz_10ms_modulation_mae_db": 0.7755126663749694, "300–1000 Hz_40ms_log_rms_mae_db": 0.654916261082544, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04971046329736961, "300–1000 Hz_40ms_envelope_correlation": 0.9983494844036195, "300–1000 Hz_40ms_modulation_mae_db": 0.8528289802862172, "1000–3000 Hz_10ms_log_rms_mae_db": 1.139640809157084, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08913165476015766, "1000–3000 Hz_10ms_envelope_correlation": 0.9945111479075769, "1000–3000 Hz_10ms_modulation_mae_db": 0.23390980203113554, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7949669541959326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04430358930189068, "1000–3000 Hz_40ms_envelope_correlation": 0.9986944773172721, "1000–3000 Hz_40ms_modulation_mae_db": 0.3334941769562268, "3000–6000 Hz_10ms_log_rms_mae_db": 1.554614255922668, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23212553865327673, "3000–6000 Hz_10ms_envelope_correlation": 0.9605784718896311, "3000–6000 Hz_10ms_modulation_mae_db": 0.8952679297188885, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4473158411777607, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19259186069002865, "3000–6000 Hz_40ms_envelope_correlation": 0.9712315580646312, "3000–6000 Hz_40ms_modulation_mae_db": 1.2836995417480874, "mr_stft_db": 5.464537723266555, "band_mae_db": 0.19855842049388964, "rms_gain_error_db": -0.14245474145736267, "output_peak": 0.4807683584742521}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9019300581948302, "全帯域_10ms_rms_envelope_nrmse": 0.06589914912132809, "全帯域_10ms_envelope_correlation": 0.9961287558333851, "全帯域_10ms_modulation_mae_db": 0.7005472240102165, "全帯域_40ms_log_rms_mae_db": 0.5294255846576333, "全帯域_40ms_rms_envelope_nrmse": 0.038184643156080625, "全帯域_40ms_envelope_correlation": 0.9987984355658409, "全帯域_40ms_modulation_mae_db": 0.7743558697538998, "300–1000 Hz_10ms_log_rms_mae_db": 0.899805949515379, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07874794594153292, "300–1000 Hz_10ms_envelope_correlation": 0.995195124843151, "300–1000 Hz_10ms_modulation_mae_db": 1.0133659602296294, "300–1000 Hz_40ms_log_rms_mae_db": 0.6179218051523607, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04821000389850375, "300–1000 Hz_40ms_envelope_correlation": 0.9983572752673244, "300–1000 Hz_40ms_modulation_mae_db": 1.0855009128404778, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1828779253932677, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09466622261224944, "1000–3000 Hz_10ms_envelope_correlation": 0.9938653429513488, "1000–3000 Hz_10ms_modulation_mae_db": 0.3289617955250694, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8499537028176175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.046779798020996625, "1000–3000 Hz_40ms_envelope_correlation": 0.998642898935356, "1000–3000 Hz_40ms_modulation_mae_db": 0.43259060235147784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5863983476211962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22528520043406344, "3000–6000 Hz_10ms_envelope_correlation": 0.9622311613639504, "3000–6000 Hz_10ms_modulation_mae_db": 0.8432152067273155, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4369823377053141, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.187429425520936, "3000–6000 Hz_40ms_envelope_correlation": 0.9722459902314803, "3000–6000 Hz_40ms_modulation_mae_db": 1.0087024406559897, "mr_stft_db": 5.450977244718501, "band_mae_db": 0.12335927400785714, "rms_gain_error_db": -0.1341160234877157, "output_peak": 0.45674011617997523}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8876315722189908, "全帯域_10ms_rms_envelope_nrmse": 0.05700624525242955, "全帯域_10ms_envelope_correlation": 0.9970038449133144, "全帯域_10ms_modulation_mae_db": 0.3337559281053725, "全帯域_40ms_log_rms_mae_db": 0.4651005920299955, "全帯域_40ms_rms_envelope_nrmse": 0.03322041308566124, "全帯域_40ms_envelope_correlation": 0.998936244672542, "全帯域_40ms_modulation_mae_db": 0.3316477223206823, "300–1000 Hz_10ms_log_rms_mae_db": 0.8499650022740813, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07190219759949813, "300–1000 Hz_10ms_envelope_correlation": 0.9958006254705692, "300–1000 Hz_10ms_modulation_mae_db": 0.3513446694200581, "300–1000 Hz_40ms_log_rms_mae_db": 0.5755023546128829, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04359047663782514, "300–1000 Hz_40ms_envelope_correlation": 0.9983851613465414, "300–1000 Hz_40ms_modulation_mae_db": 0.2496694897033294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1299628730199989, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08290345933087943, "1000–3000 Hz_10ms_envelope_correlation": 0.9954272821538026, "1000–3000 Hz_10ms_modulation_mae_db": 0.5916679195268032, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7930162001444148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04609036397645545, "1000–3000 Hz_40ms_envelope_correlation": 0.9987757187166071, "1000–3000 Hz_40ms_modulation_mae_db": 0.7460672375202662, "3000–6000 Hz_10ms_log_rms_mae_db": 1.604811363526088, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24936377783587477, "3000–6000 Hz_10ms_envelope_correlation": 0.9522664136587677, "3000–6000 Hz_10ms_modulation_mae_db": 0.9223796227025542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4978728344862533, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2097569656779101, "3000–6000 Hz_40ms_envelope_correlation": 0.9645731247308053, "3000–6000 Hz_40ms_modulation_mae_db": 1.0372067242042187, "mr_stft_db": 5.4532979790533815, "band_mae_db": 0.1332919729672682, "rms_gain_error_db": 0.005573900037677566, "output_peak": 0.47382869096690666}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87760, "output_samples": 87760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.13113569549646, "spectral_convergence": 0.16320256423747564, "spectral_convergence_status": "ok", "amplitude_floor": 1.7869460513029852e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.220571711728019, "spectral_convergence": 0.16206882381232282, "spectral_convergence_status": "ok", "amplitude_floor": 1.6383670687778497e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9784640351562315, "spectral_convergence": 0.16872096984594753, "spectral_convergence_status": "ok", "amplitude_floor": 9.631824221799184e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.013402842456697e-05, "output_power": 9.742232348661091e-05, "delta_db": 0.33769701389055823, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00191544195180521, "output_power": 0.0019091066033052934, "delta_db": -0.014388152595818066, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015517446606579416, "output_power": 0.0015543742863794098, "delta_db": 0.007353435155690933, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955960266218775, "output_power": 0.0004655273939977284, "delta_db": -0.2718257365343314, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000158155871967105, "output_power": 0.00014997653241468983, "delta_db": -0.23062012794110248, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.514760940569477e-06, "output_power": 9.346114959951279e-06, "delta_db": -0.0776676239231744, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15659201500677924, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03599264244857402, "pitch": null, "output_rms_dbfs": -23.78226376415043, "output_peak": 0.49788403591468944, "reference_rms_dbfs": -23.74627112170186, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.44339048079357}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002415922668587691, "log_rms_mae_db": 0.8503538201867293, "rms_envelope_nrmse": 0.05087240624945553, "envelope_correlation": 0.9976027889248533, "rms_gain_error_db": -0.03599271343455946, "reference_mean_power": 0.004224437839878649, "output_mean_power": 0.004189571945619842}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.932071872304039e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008459874532766979, "output_power": 0.000838680250965559, "reference_active": true, "delta_db": -0.03767505646021138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.267978041368795e-05, "output_power": 9.04074119447227e-05, "reference_active": true, "delta_db": -0.10780959090409402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0604587197812338e-05, "output_power": 1.9558726656776205e-05, "reference_active": true, "delta_db": -0.2262334083951495}], "active_band_count": 3, "active_band_mae_db": 0.12390601858648496}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00018058591873130477, "log_rms_mae_db": 0.4450996302008955, "rms_envelope_nrmse": 0.026180531864028227, "envelope_correlation": 0.9993475243347913, "rms_gain_error_db": -0.03599309342850631, "reference_mean_power": 0.004247690453399677, "output_mean_power": 0.004212632277874306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8509413564750568e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008019972220007684, "output_power": 0.0007947154608877836, "reference_active": true, "delta_db": -0.03961201821743864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.130657727931199e-05, "output_power": 4.029998340423847e-05, "reference_active": true, "delta_db": -0.1071434292385353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0523532353898244e-06, "output_power": 9.753059330081182e-07, "reference_active": true, "delta_db": -0.33020674471626965}], "active_band_count": 3, "active_band_mae_db": 0.15898739739074785}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002374784601360974, "log_rms_mae_db": 0.8787768362290846, "rms_envelope_nrmse": 0.06115970571380142, "envelope_correlation": 0.9969485520027105, "rms_gain_error_db": -0.047624916768673656, "reference_mean_power": 0.0028785349502786395, "output_mean_power": 0.0028471412609987258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.6369199856429154e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008090138264371756, "output_power": 0.0008030412478391437, "reference_active": true, "delta_db": -0.03218090793122702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000105081165545653, "output_power": 0.00010281134699402543, "reference_active": true, "delta_db": -0.09483832131543189}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7476383653056024e-05, "output_power": 1.4789751321159768e-05, "reference_active": true, "delta_db": -0.7249069834544359}], "active_band_count": 3, "active_band_mae_db": 0.2839754042336983}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.0001750973468816212, "log_rms_mae_db": 0.6232140840626511, "rms_envelope_nrmse": 0.0319870155147784, "envelope_correlation": 0.9991428305992791, "rms_gain_error_db": -0.04762492112736334, "reference_mean_power": 0.002894380018498304, "output_mean_power": 0.0028628135179003117}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.567165950582092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007701013013376472, "output_power": 0.0007642693459975145, "reference_active": true, "delta_db": -0.03301416413572341}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.715572645231493e-05, "output_power": 4.571292769818379e-05, "reference_active": true, "delta_db": -0.13495403016025148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1197000044443696e-06, "output_power": 9.206846846294935e-07, "reference_active": true, "delta_db": -0.8499076110610588}], "active_band_count": 3, "active_band_mae_db": 0.3392919351190113}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.00012170766433419248, "log_rms_mae_db": 1.2071235271794167, "rms_envelope_nrmse": 0.09539069228480658, "envelope_correlation": 0.9941737266649306, "rms_gain_error_db": -0.28310951323977807, "reference_mean_power": 0.0006352125631157451, "output_mean_power": 0.0005951249083153317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 2.616269223692633e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010398636594123316, "output_power": 9.649804202195209e-05, "reference_active": true, "delta_db": -0.3245789953422095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0847306945521593e-05, "output_power": 9.390309509493972e-06, "reference_active": true, "delta_db": -0.6264202245826022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2384759945240008e-06, "output_power": 3.403495084972404e-06, "reference_active": true, "delta_db": 0.2158444561383346}], "active_band_count": 3, "active_band_mae_db": 0.38894789202104874}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00010363700534457545, "log_rms_mae_db": 0.8693283153419263, "rms_envelope_nrmse": 0.058615581120197144, "envelope_correlation": 0.998217966771417, "rms_gain_error_db": -0.2831095099874944, "reference_mean_power": 0.0006387091058518984, "output_mean_power": 0.0005984007880456062}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.5308592276752824e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.945508810428335e-05, "output_power": 9.189218890024636e-05, "reference_active": true, "delta_db": -0.3434840977801854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.423063898822258e-06, "output_power": 5.040515018498505e-06, "reference_active": true, "delta_db": -0.317698085808534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8432082747665845e-07, "output_power": 1.7208719157805674e-07, "reference_active": true, "delta_db": -0.29825864375751526}], "active_band_count": 3, "active_band_mae_db": 0.31981360911541157}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 2.9435709173680706e-05, "log_rms_mae_db": 1.579949002894304, "rms_envelope_nrmse": 0.23327253277520965, "envelope_correlation": 0.9627189708263825, "rms_gain_error_db": 0.3345835263014724, "reference_mean_power": 1.4688872454843858e-05, "output_mean_power": 1.5865246071865756e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 9.038086104701173e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4018145980274633e-06, "output_power": 3.7295695602444803e-06, "reference_active": true, "delta_db": 0.3994807111538396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2285201909756879e-06, "output_power": 1.5957122038280346e-06, "reference_active": true, "delta_db": 1.1357226789679888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.73608399821043e-07, "output_power": 4.582963022328578e-07, "reference_active": true, "delta_db": 2.2401692698878852}], "active_band_count": 3, "active_band_mae_db": 1.2584575533365712}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 2.1616481165467548e-05, "log_rms_mae_db": 1.4767717447753146, "rms_envelope_nrmse": 0.20389299570743089, "envelope_correlation": 0.9697485943473653, "rms_gain_error_db": 0.33458394268963343, "reference_mean_power": 1.4769720024935061e-05, "output_mean_power": 1.5952569933117565e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.51604517212682e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5641156211998326e-06, "output_power": 4.1186055331336985e-06, "reference_active": true, "delta_db": 0.6279841434822685}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790775026435671e-07, "output_power": 7.741542367701608e-07, "reference_active": true, "delta_db": 1.2609080215320703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8657206259627155e-08, "output_power": 3.525843283864535e-08, "reference_active": true, "delta_db": 2.764163921552445}], "active_band_count": 3, "active_band_mae_db": 1.5510186955222611}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.909327098994202, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "411dccac747785067a97912294bde9b47136e776cb34000418285f3d726dcd08", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9946833315263324, "standard_deviation": 0.0005209931837356812, "minimum": 0.9938653429513488, "maximum": 0.9954272821538026, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.181600823052869, "standard_deviation": 0.031447803570118456, "minimum": 1.1299628730199989, "maximum": 1.2204331205317256, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3823159179054718, "standard_deviation": 0.14592865213805928, "minimum": 0.19716529906859878, "maximum": 0.5917262314237554, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08879247470891027, "standard_deviation": 0.004556446494745317, "minimum": 0.08290345933087943, "maximum": 0.09539069228480658, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.998586616073995, "standard_deviation": 0.00024760085105639533, "minimum": 0.998217966771417, "maximum": 0.99900714934695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8315637517378217, "standard_deviation": 0.03716527722963848, "minimum": 0.7905676303260044, "maximum": 0.8828604991972194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4347663394967578, "standard_deviation": 0.22520592446450324, "minimum": 0.2511720671580837, "maximum": 0.8280920136698255, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04834964105797817, "standard_deviation": 0.005485437031596505, "minimum": 0.040244817890030815, "maximum": 0.058615581120197144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9605924532700288, "standard_deviation": 0.003934699763946117, "minimum": 0.9522664136587677, "maximum": 0.9644057215699028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5675978659345127, "standard_deviation": 0.02749734454283275, "minimum": 1.5351908195851094, "maximum": 1.604811363526088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7580050302283259, "standard_deviation": 0.30247337041284555, "minimum": 0.29732971153349735, "maximum": 1.2584575533365712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23007802807810068, "standard_deviation": 0.009865568744105331, "minimum": 0.2184540394107831, "maximum": 0.24936377783587477, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9705920564868509, "standard_deviation": 0.0027957781803386625, "minimum": 0.9645731247308053, "maximum": 0.9735853494975792, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.449006359683731, "standard_deviation": 0.03162685004116174, "minimum": 1.3930334966730509, "maximum": 1.4978728344862533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9630080401328338, "standard_deviation": 0.35212619986025906, "minimum": 0.4435654347378925, "maximum": 1.5510186955222611, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19363819107329389, "standard_deviation": 0.009297933033856369, "minimum": 0.1829135494283235, "maximum": 0.2097569656779101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9957867812759339, "standard_deviation": 0.000806494121514349, "minimum": 0.9944230686657818, "maximum": 0.9969485520027105, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.8939589790009665, "standard_deviation": 0.024228560289926145, "minimum": 0.8499650022740813, "maximum": 0.9299408102433908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5643255344439897, "standard_deviation": 0.27752988580187227, "minimum": 0.18223722731565176, "maximum": 1.0133659602296294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07296841922975805, "standard_deviation": 0.007204577770060659, "minimum": 0.06115970571380142, "maximum": 0.08427142224581242, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99851572957302, "standard_deviation": 0.0004252458978162908, "minimum": 0.9977631159441114, "maximum": 0.9991428305992791, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6300595001884554, "standard_deviation": 0.025648942663425944, "minimum": 0.5755023546128829, "maximum": 0.654916261082544, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6593298488079599, "standard_deviation": 0.3661936692547905, "minimum": 0.20040804736375586, "maximum": 1.1028910606767812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.044225671763214146, "standard_deviation": 0.00709657181335391, "minimum": 0.0319870155147784, "maximum": 0.05465002918454301, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.164362237839038, "standard_deviation": 0.03279980148691378, "minimum": 0.12335927400785714, "maximum": 0.2154298319310112, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.451683019249333, "standard_deviation": 0.007161935722846876, "minimum": 5.44339048079357, "maximum": 5.464537723266555, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.48654964304026793, "standard_deviation": 0.021933224868799443, "minimum": 0.45674011617997523, "maximum": 0.5161436937730849, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09265505286001818, "standard_deviation": 0.054977104015115956, "minimum": -0.1436662130944173, "maximum": 0.005573900037677566, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.996932910232836, "standard_deviation": 0.000559846747955536, "minimum": 0.9961287558333851, "maximum": 0.9976027889248533, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8709148374041077, "standard_deviation": 0.030186461653996927, "minimum": 0.8307580192611815, "maximum": 0.9026535059168875, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4432307724699837, "standard_deviation": 0.20661820401660372, "minimum": 0.12390601858648496, "maximum": 0.7271194453878559, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05824057672710915, "standard_deviation": 0.00539026760907676, "minimum": 0.05087240624945553, "maximum": 0.06589914912132809, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990500650049707, "standard_deviation": 0.00024070674468399305, "minimum": 0.9987167383639572, "maximum": 0.9993475243347913, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4771214233379397, "standard_deviation": 0.029610220698174552, "minimum": 0.4450996302008955, "maximum": 0.5294255846576333, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5089434961451025, "standard_deviation": 0.23606536671268197, "minimum": 0.15898739739074785, "maximum": 0.7918457111730959, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.033153889002700054, "standard_deviation": 0.004458873025720421, "minimum": 0.026180531864028227, "maximum": 0.03875681951572807, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41797-0000", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.065336227416992, "predicted_mos": 2.065336227416992, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "baea4b74407ed7b3fb704aab5b43a2734b8a5e3cff424742cec41c1a33179466", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.1331541002728045, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8275420661134133, "全帯域_10ms_rms_envelope_nrmse": 0.056607472630907596, "全帯域_10ms_envelope_correlation": 0.9970439760544739, "全帯域_10ms_modulation_mae_db": 0.1493228137973512, "全帯域_40ms_log_rms_mae_db": 0.4215301551953602, "全帯域_40ms_rms_envelope_nrmse": 0.030177689252995656, "全帯域_40ms_envelope_correlation": 0.9991531576765941, "全帯域_40ms_modulation_mae_db": 0.028520065256957255, "300–1000 Hz_10ms_log_rms_mae_db": 0.9491065074758759, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0707896827750863, "300–1000 Hz_10ms_envelope_correlation": 0.9959668304724308, "300–1000 Hz_10ms_modulation_mae_db": 0.34333155310960556, "300–1000 Hz_40ms_log_rms_mae_db": 0.7150646433416529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03975031924835561, "300–1000 Hz_40ms_envelope_correlation": 0.9987578383749357, "300–1000 Hz_40ms_modulation_mae_db": 0.17518117439913958, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3629832669339723, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0964721936510582, "1000–3000 Hz_10ms_envelope_correlation": 0.9935540883020151, "1000–3000 Hz_10ms_modulation_mae_db": 0.6654858140907641, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9540469685089648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048909835524925976, "1000–3000 Hz_40ms_envelope_correlation": 0.9984265003458918, "1000–3000 Hz_40ms_modulation_mae_db": 0.4950509676050674, "3000–6000 Hz_10ms_log_rms_mae_db": 2.091617792047925, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24788093861751617, "3000–6000 Hz_10ms_envelope_correlation": 0.9532680308119481, "3000–6000 Hz_10ms_modulation_mae_db": 0.4127435344948358, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8304584245145914, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19187246963963028, "3000–6000 Hz_40ms_envelope_correlation": 0.971140984800925, "3000–6000 Hz_40ms_modulation_mae_db": 0.5197198832966515, "mr_stft_db": 5.7524631334615215, "band_mae_db": 0.1083544293595697, "rms_gain_error_db": -0.05188552611639122, "output_peak": 0.45646229537896515}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8664947652682231, "全帯域_10ms_rms_envelope_nrmse": 0.06106836548161476, "全帯域_10ms_envelope_correlation": 0.9966839005548226, "全帯域_10ms_modulation_mae_db": 0.19812838493095974, "全帯域_40ms_log_rms_mae_db": 0.48544612896665895, "全帯域_40ms_rms_envelope_nrmse": 0.03509295097745678, "全帯域_40ms_envelope_correlation": 0.9989967189002675, "全帯域_40ms_modulation_mae_db": 0.32214410827502343, "300–1000 Hz_10ms_log_rms_mae_db": 0.9658331306033531, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07554277360794581, "300–1000 Hz_10ms_envelope_correlation": 0.9956212206478106, "300–1000 Hz_10ms_modulation_mae_db": 0.4583293482063313, "300–1000 Hz_40ms_log_rms_mae_db": 0.7501825076893546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04564022363927089, "300–1000 Hz_40ms_envelope_correlation": 0.9985937275707638, "300–1000 Hz_40ms_modulation_mae_db": 0.47546561634665513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3583830031394166, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09471328714392753, "1000–3000 Hz_10ms_envelope_correlation": 0.9937352425328854, "1000–3000 Hz_10ms_modulation_mae_db": 0.2882638537558754, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9921112474878253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050376457942870066, "1000–3000 Hz_40ms_envelope_correlation": 0.9982344416633812, "1000–3000 Hz_40ms_modulation_mae_db": 0.3797301286707712, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1075159389111566, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2500376427211096, "3000–6000 Hz_10ms_envelope_correlation": 0.9536539383516341, "3000–6000 Hz_10ms_modulation_mae_db": 0.564363758782779, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8053560633212065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19504892350322509, "3000–6000 Hz_40ms_envelope_correlation": 0.9706448770988978, "3000–6000 Hz_40ms_modulation_mae_db": 0.6217825998520637, "mr_stft_db": 5.768628224412098, "band_mae_db": 0.14045513854977026, "rms_gain_error_db": -0.1249479952019996, "output_peak": 0.4798090847803578}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8053511521768386, "全帯域_10ms_rms_envelope_nrmse": 0.06600949309417375, "全帯域_10ms_envelope_correlation": 0.9961731070812988, "全帯域_10ms_modulation_mae_db": 1.0003421890236972, "全帯域_40ms_log_rms_mae_db": 0.46935689204897996, "全帯域_40ms_rms_envelope_nrmse": 0.04318369875443548, "全帯域_40ms_envelope_correlation": 0.9984546913405569, "全帯域_40ms_modulation_mae_db": 1.275303415075724, "300–1000 Hz_10ms_log_rms_mae_db": 1.0274619797923856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08593958732506009, "300–1000 Hz_10ms_envelope_correlation": 0.9943135072125019, "300–1000 Hz_10ms_modulation_mae_db": 1.0649773745857962, "300–1000 Hz_40ms_log_rms_mae_db": 0.8543684001438351, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05980540950219633, "300–1000 Hz_40ms_envelope_correlation": 0.9973762832290595, "300–1000 Hz_40ms_modulation_mae_db": 1.3695454053562681, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3497490145097089, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09162309665879396, "1000–3000 Hz_10ms_envelope_correlation": 0.9943393185723196, "1000–3000 Hz_10ms_modulation_mae_db": 0.6199784021690752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9513004200156998, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05380653143030523, "1000–3000 Hz_40ms_envelope_correlation": 0.9982219913682451, "1000–3000 Hz_40ms_modulation_mae_db": 0.6785421298825431, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1252768905745802, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2469359761580451, "3000–6000 Hz_10ms_envelope_correlation": 0.9547591132121652, "3000–6000 Hz_10ms_modulation_mae_db": 0.9643212193174876, "3000–6000 Hz_40ms_log_rms_mae_db": 1.895677329768622, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19761175238007112, "3000–6000 Hz_40ms_envelope_correlation": 0.9696997433344405, "3000–6000 Hz_40ms_modulation_mae_db": 1.2701815585608598, "mr_stft_db": 5.757762195628047, "band_mae_db": 0.1877863190851089, "rms_gain_error_db": -0.1503157365137477, "output_peak": 0.5378566657567535}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8541248476010014, "全帯域_10ms_rms_envelope_nrmse": 0.05790023938236324, "全帯域_10ms_envelope_correlation": 0.9969611774444034, "全帯域_10ms_modulation_mae_db": 0.5572581173348913, "全帯域_40ms_log_rms_mae_db": 0.4351373559505791, "全帯域_40ms_rms_envelope_nrmse": 0.029810583352930776, "全帯域_40ms_envelope_correlation": 0.9992405171572176, "全帯域_40ms_modulation_mae_db": 0.5679548564471378, "300–1000 Hz_10ms_log_rms_mae_db": 1.1023617748910801, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07460260911067729, "300–1000 Hz_10ms_envelope_correlation": 0.9955126433844154, "300–1000 Hz_10ms_modulation_mae_db": 0.5695384593682998, "300–1000 Hz_40ms_log_rms_mae_db": 0.8497528456649571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03983684301395045, "300–1000 Hz_40ms_envelope_correlation": 0.9987516533309359, "300–1000 Hz_40ms_modulation_mae_db": 0.6431347992767641, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3748318670632358, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09208277718136361, "1000–3000 Hz_10ms_envelope_correlation": 0.9944899367202864, "1000–3000 Hz_10ms_modulation_mae_db": 0.6688218340316316, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0045919821664882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05707466382910838, "1000–3000 Hz_40ms_envelope_correlation": 0.9982046006837767, "1000–3000 Hz_40ms_modulation_mae_db": 0.9309926079589731, "3000–6000 Hz_10ms_log_rms_mae_db": 2.150723206484414, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28512346636896185, "3000–6000 Hz_10ms_envelope_correlation": 0.9383237944431764, "3000–6000 Hz_10ms_modulation_mae_db": 0.8989876532860338, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8725364085723093, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2159021095712923, "3000–6000 Hz_40ms_envelope_correlation": 0.9629112203318476, "3000–6000 Hz_40ms_modulation_mae_db": 1.0119211943793316, "mr_stft_db": 5.765583762904105, "band_mae_db": 0.1543196822187535, "rms_gain_error_db": -0.09386653081046024, "output_peak": 0.5521622975025217}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8495319486503639, "全帯域_10ms_rms_envelope_nrmse": 0.064800093944379, "全帯域_10ms_envelope_correlation": 0.9961412498772222, "全帯域_10ms_modulation_mae_db": 0.5283363636101347, "全帯域_40ms_log_rms_mae_db": 0.4690962541609164, "全帯域_40ms_rms_envelope_nrmse": 0.0339984465368716, "全帯域_40ms_envelope_correlation": 0.9989471868813776, "全帯域_40ms_modulation_mae_db": 0.4708315569985216, "300–1000 Hz_10ms_log_rms_mae_db": 0.9834218423156482, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08286259039306074, "300–1000 Hz_10ms_envelope_correlation": 0.9944193879797465, "300–1000 Hz_10ms_modulation_mae_db": 0.7289504887546693, "300–1000 Hz_40ms_log_rms_mae_db": 0.7484953661062299, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04535238920973325, "300–1000 Hz_40ms_envelope_correlation": 0.9983230761962437, "300–1000 Hz_40ms_modulation_mae_db": 0.5713332315949767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3603923734657235, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09046843645195601, "1000–3000 Hz_10ms_envelope_correlation": 0.9945313717920994, "1000–3000 Hz_10ms_modulation_mae_db": 0.1807553848841769, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9469402504230884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047988355766296086, "1000–3000 Hz_40ms_envelope_correlation": 0.9987156461059137, "1000–3000 Hz_40ms_modulation_mae_db": 0.2205171105187925, "3000–6000 Hz_10ms_log_rms_mae_db": 2.068475754172867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2490104903090512, "3000–6000 Hz_10ms_envelope_correlation": 0.9528842016266299, "3000–6000 Hz_10ms_modulation_mae_db": 0.7094829758156364, "3000–6000 Hz_40ms_log_rms_mae_db": 1.82746966413031, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19222135205491211, "3000–6000 Hz_40ms_envelope_correlation": 0.9709102193660405, "3000–6000 Hz_40ms_modulation_mae_db": 0.7053769569013074, "mr_stft_db": 5.757591061794119, "band_mae_db": 0.18272507534075585, "rms_gain_error_db": -0.07891615145144412, "output_peak": 0.4627995060999195}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8093799321560577, "全帯域_10ms_rms_envelope_nrmse": 0.05144186684781015, "全帯域_10ms_envelope_correlation": 0.9975982941718695, "全帯域_10ms_modulation_mae_db": 0.09333782572108788, "全帯域_40ms_log_rms_mae_db": 0.42481842202689446, "全帯域_40ms_rms_envelope_nrmse": 0.026738083839602602, "全帯域_40ms_envelope_correlation": 0.9993732919010335, "全帯域_40ms_modulation_mae_db": 0.0902562984121547, "300–1000 Hz_10ms_log_rms_mae_db": 0.9670971425686222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06564350949489604, "300–1000 Hz_10ms_envelope_correlation": 0.9965302884534586, "300–1000 Hz_10ms_modulation_mae_db": 0.13816320912219737, "300–1000 Hz_40ms_log_rms_mae_db": 0.7701303614199473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03482467109895333, "300–1000 Hz_40ms_envelope_correlation": 0.9990470374525674, "300–1000 Hz_40ms_modulation_mae_db": 0.18537225438325908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3292734119419443, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0894787921281093, "1000–3000 Hz_10ms_envelope_correlation": 0.9944270481653278, "1000–3000 Hz_10ms_modulation_mae_db": 0.3634610295088467, "1000–3000 Hz_40ms_log_rms_mae_db": 0.96588451878327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04678377673286979, "1000–3000 Hz_40ms_envelope_correlation": 0.9984978276632518, "1000–3000 Hz_40ms_modulation_mae_db": 0.393704599195195, "3000–6000 Hz_10ms_log_rms_mae_db": 2.056747513946732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23707369642443482, "3000–6000 Hz_10ms_envelope_correlation": 0.9576267815469838, "3000–6000 Hz_10ms_modulation_mae_db": 0.6956814188790436, "3000–6000 Hz_40ms_log_rms_mae_db": 1.845094456913663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19478125291666992, "3000–6000 Hz_40ms_envelope_correlation": 0.9702066224725631, "3000–6000 Hz_40ms_modulation_mae_db": 0.5931560183695148, "mr_stft_db": 5.748497106362255, "band_mae_db": 0.11099104659396634, "rms_gain_error_db": -0.07468395235680322, "output_peak": 0.5135552701390396}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8562616618923583, "全帯域_10ms_rms_envelope_nrmse": 0.05652832159207387, "全帯域_10ms_envelope_correlation": 0.9971985579063182, "全帯域_10ms_modulation_mae_db": 0.4205232659114108, "全帯域_40ms_log_rms_mae_db": 0.4385620034645408, "全帯域_40ms_rms_envelope_nrmse": 0.03595147673397792, "全帯域_40ms_envelope_correlation": 0.9989452567632457, "全帯域_40ms_modulation_mae_db": 0.47682760978526195, "300–1000 Hz_10ms_log_rms_mae_db": 1.022118682796808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07252656440598973, "300–1000 Hz_10ms_envelope_correlation": 0.9959596161835079, "300–1000 Hz_10ms_modulation_mae_db": 0.5805274823133109, "300–1000 Hz_40ms_log_rms_mae_db": 0.795438395340402, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04924331635489136, "300–1000 Hz_40ms_envelope_correlation": 0.9982442994094237, "300–1000 Hz_40ms_modulation_mae_db": 0.8199466916541098, "1000–3000 Hz_10ms_log_rms_mae_db": 1.375610805098619, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0925552274424635, "1000–3000 Hz_10ms_envelope_correlation": 0.9941182497300465, "1000–3000 Hz_10ms_modulation_mae_db": 0.6173951825940867, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9657509910187273, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049515319271407676, "1000–3000 Hz_40ms_envelope_correlation": 0.9984281343550119, "1000–3000 Hz_40ms_modulation_mae_db": 0.7905602538835336, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0364556541263505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25229927956306736, "3000–6000 Hz_10ms_envelope_correlation": 0.9542087243350861, "3000–6000 Hz_10ms_modulation_mae_db": 0.9555126914581585, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8144576469769644, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19945187321936841, "3000–6000 Hz_40ms_envelope_correlation": 0.9701236850782364, "3000–6000 Hz_40ms_modulation_mae_db": 0.8775286001878463, "mr_stft_db": 5.766768624609742, "band_mae_db": 0.20017362455660556, "rms_gain_error_db": -0.1306246138145634, "output_peak": 0.48855366561608743}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8613926266971135, "全帯域_10ms_rms_envelope_nrmse": 0.05059729581140704, "全帯域_10ms_envelope_correlation": 0.9977297614994296, "全帯域_10ms_modulation_mae_db": 0.12513399903704883, "全帯域_40ms_log_rms_mae_db": 0.46661174486414475, "全帯域_40ms_rms_envelope_nrmse": 0.029021707547657483, "全帯域_40ms_envelope_correlation": 0.9993130005482292, "全帯域_40ms_modulation_mae_db": 0.14212027089460702, "300–1000 Hz_10ms_log_rms_mae_db": 1.0140765904635827, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06666425570318972, "300–1000 Hz_10ms_envelope_correlation": 0.9964897786940151, "300–1000 Hz_10ms_modulation_mae_db": 0.30678641106168403, "300–1000 Hz_40ms_log_rms_mae_db": 0.7708328641477106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04068706464702641, "300–1000 Hz_40ms_envelope_correlation": 0.9987466406757463, "300–1000 Hz_40ms_modulation_mae_db": 0.28282497469939505, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3623690375431914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09324571377889541, "1000–3000 Hz_10ms_envelope_correlation": 0.9942509095070026, "1000–3000 Hz_10ms_modulation_mae_db": 0.4728360590181084, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9705876438417566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0554917881610085, "1000–3000 Hz_40ms_envelope_correlation": 0.9982505010043485, "1000–3000 Hz_40ms_modulation_mae_db": 0.5530545720893437, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0807144272980613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2571229442323587, "3000–6000 Hz_10ms_envelope_correlation": 0.952906534195435, "3000–6000 Hz_10ms_modulation_mae_db": 1.1675035995753413, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8437000682508435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20606523464329154, "3000–6000 Hz_40ms_envelope_correlation": 0.9684123659041988, "3000–6000 Hz_40ms_modulation_mae_db": 1.215876195503086, "mr_stft_db": 5.759566065085199, "band_mae_db": 0.15737518224772437, "rms_gain_error_db": -0.1049792243172891, "output_peak": 0.4925025973417593}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87760, "output_samples": 87760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.440500522797263, "spectral_convergence": 0.16947232468054216, "spectral_convergence_status": "ok", "amplitude_floor": 1.7869460513029852e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.600930399362355, "spectral_convergence": 0.1652538587376477, "spectral_convergence_status": "ok", "amplitude_floor": 1.6383670687778497e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.215958478224946, "spectral_convergence": 0.16537672605160372, "spectral_convergence_status": "ok", "amplitude_floor": 9.631824221799184e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.013402842456697e-05, "output_power": 9.580373513040282e-05, "delta_db": 0.26493659650406953, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00191544195180521, "output_power": 0.0019038330482279528, "delta_db": -0.026401336856530965, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015517446606579416, "output_power": 0.001526501405126939, "delta_db": -0.07123051313909541, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955960266218775, "output_power": 0.000477068235277929, "delta_db": -0.16547315005015417, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000158155871967105, "output_power": 0.00015751105081884836, "delta_db": -0.01774292051542775, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.514760940569477e-06, "output_power": 9.746127464036914e-06, "delta_db": 0.10434205909214045, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1083544293595697, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05188552611639122, "pitch": null, "output_rms_dbfs": -23.79815664781825, "output_peak": 0.45646229537896515, "reference_rms_dbfs": -23.74627112170186, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7524631334615215}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002415922668587691, "log_rms_mae_db": 0.8275420661134133, "rms_envelope_nrmse": 0.056607472630907596, "envelope_correlation": 0.9970439760544739, "rms_gain_error_db": -0.05188543692433433, "reference_mean_power": 0.004224437839878649, "output_mean_power": 0.00417426849822548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.932071872304039e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008459874532766979, "output_power": 0.0008355102032618239, "reference_active": true, "delta_db": -0.05412164253350737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.267978041368795e-05, "output_power": 9.009725929354473e-05, "reference_active": true, "delta_db": -0.12273415951392681}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0604587197812338e-05, "output_power": 2.193184571237709e-05, "reference_active": true, "delta_db": 0.27111263934461943}], "active_band_count": 3, "active_band_mae_db": 0.1493228137973512}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00018058591873130477, "log_rms_mae_db": 0.4215301551953602, "rms_envelope_nrmse": 0.030177689252995656, "envelope_correlation": 0.9991531576765941, "rms_gain_error_db": -0.05188559308012844, "reference_mean_power": 0.004247690453399677, "output_mean_power": 0.004197244813243441}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8509413564750568e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008019972220007684, "output_power": 0.000790788873499184, "reference_active": true, "delta_db": -0.06112313849609806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.130657727931199e-05, "output_power": 4.1479800963948175e-05, "reference_active": true, "delta_db": 0.018174535898349428}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0523532353898244e-06, "output_power": 1.0508368368339477e-06, "reference_active": true, "delta_db": -0.006262521376424274}], "active_band_count": 3, "active_band_mae_db": 0.028520065256957255}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002374784601360974, "log_rms_mae_db": 0.9491065074758759, "rms_envelope_nrmse": 0.0707896827750863, "envelope_correlation": 0.9959668304724308, "rms_gain_error_db": -0.09722762566114904, "reference_mean_power": 0.0028785349502786395, "output_mean_power": 0.002814807790362705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.6369199856429154e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008090138264371756, "output_power": 0.0007869827515232208, "reference_active": true, "delta_db": -0.11990730044036643}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000105081165545653, "output_power": 0.00010076202767582909, "reference_active": true, "delta_db": -0.18227982873716794}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7476383653056024e-05, "output_power": 2.0664853133808472e-05, "reference_active": true, "delta_db": 0.7278075301512823}], "active_band_count": 3, "active_band_mae_db": 0.34333155310960556}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.0001750973468816212, "log_rms_mae_db": 0.7150646433416529, "rms_envelope_nrmse": 0.03975031924835561, "envelope_correlation": 0.9987578383749357, "rms_gain_error_db": -0.09722770100791979, "reference_mean_power": 0.002894380018498304, "output_mean_power": 0.0028303020195062016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.567165950582092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007701013013376472, "output_power": 0.0007482342726034883, "reference_active": true, "delta_db": -0.12510260252739025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.715572645231493e-05, "output_power": 4.651743363938693e-05, "reference_active": true, "delta_db": -0.0591869296797161}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1197000044443696e-06, "output_power": 1.2112312830476007e-06, "reference_active": true, "delta_db": 0.34125399099031234}], "active_band_count": 3, "active_band_mae_db": 0.17518117439913958}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.00012170766433419248, "log_rms_mae_db": 1.3629832669339723, "rms_envelope_nrmse": 0.0964721936510582, "envelope_correlation": 0.9935540883020151, "rms_gain_error_db": -0.13177262739061837, "reference_mean_power": 0.0006352125631157451, "output_mean_power": 0.0006162285510746522}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 2.616269223692633e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010398636594123316, "output_power": 9.722178273404306e-05, "reference_active": true, "delta_db": -0.29212820612738016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0847306945521593e-05, "output_power": 9.825513995753716e-06, "reference_active": true, "delta_db": -0.42966650924839633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2384759945240008e-06, "output_power": 4.343171260401067e-06, "reference_active": true, "delta_db": 1.2746627268965158}], "active_band_count": 3, "active_band_mae_db": 0.6654858140907641}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00010363700534457545, "log_rms_mae_db": 0.9540469685089648, "rms_envelope_nrmse": 0.048909835524925976, "envelope_correlation": 0.9984265003458918, "rms_gain_error_db": -0.1317726887689262, "reference_mean_power": 0.0006387091058518984, "output_mean_power": 0.0006196205871057584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.5308592276752824e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.945508810428335e-05, "output_power": 9.320042375815814e-05, "reference_active": true, "delta_db": -0.2820911947294831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.423063898822258e-06, "output_power": 4.859246366613266e-06, "reference_active": true, "delta_db": -0.4767580306251174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8432082747665845e-07, "output_power": 2.1787368514494645e-07, "reference_active": true, "delta_db": 0.7263036774606018}], "active_band_count": 3, "active_band_mae_db": 0.4950509676050674}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 2.9435709173680706e-05, "log_rms_mae_db": 2.091617792047925, "rms_envelope_nrmse": 0.24788093861751617, "envelope_correlation": 0.9532680308119481, "rms_gain_error_db": -0.05648947691783926, "reference_mean_power": 1.4688872454843858e-05, "output_mean_power": 1.4499048818150988e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 9.038086104701173e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4018145980274633e-06, "output_power": 3.334176263101515e-06, "reference_active": true, "delta_db": -0.08722085158760479}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2285201909756879e-06, "output_power": 1.0167007085723111e-06, "reference_active": true, "delta_db": -0.8218917236033454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.73608399821043e-07, "output_power": 2.9514894078799466e-07, "reference_active": true, "delta_db": 0.3291180282935573}], "active_band_count": 3, "active_band_mae_db": 0.4127435344948358}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 2.1616481165467548e-05, "log_rms_mae_db": 1.8304584245145914, "rms_envelope_nrmse": 0.19187246963963028, "envelope_correlation": 0.971140984800925, "rms_gain_error_db": -0.056489829787514136, "reference_mean_power": 1.4769720024935061e-05, "output_mean_power": 1.4578850414187978e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.51604517212682e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5641156211998326e-06, "output_power": 3.4373132380421395e-06, "reference_active": true, "delta_db": -0.15732673611639822}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790775026435671e-07, "output_power": 5.123676308004898e-07, "reference_active": true, "delta_db": -0.5315500770589658}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8657206259627155e-08, "output_power": 2.279685287711101e-08, "reference_active": true, "delta_db": 0.8702828367145904}], "active_band_count": 3, "active_band_mae_db": 0.5197198832966515}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8993350360192544, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1530fdc0acb9f13c3779e3ca79a76edfe915fcce916cf7db1be603ffc1d937b4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9941807706652479, "standard_deviation": 0.00035943393416288906, "minimum": 0.9935540883020151, "maximum": 0.9945313717920994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3591990974619765, "standard_deviation": 0.014760162266244168, "minimum": 1.3292734119419443, "maximum": 1.375610805098619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4846246950065706, "standard_deviation": 0.18833924643143185, "minimum": 0.1807553848841769, "maximum": 0.6688218340316316, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09257994055457094, "standard_deviation": 0.0022482678844079453, "minimum": 0.0894787921281093, "maximum": 0.0964721936510582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9983724553987277, "standard_deviation": 0.0001789869763620585, "minimum": 0.9982046006837767, "maximum": 0.9987156461059137, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9689017527807275, "standard_deviation": 0.020167602960417922, "minimum": 0.9469402504230884, "maximum": 1.0045919821664882, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5552690462255275, "standard_deviation": 0.23431766041164143, "minimum": 0.2205171105187925, "maximum": 0.9309926079589731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05124334108234896, "standard_deviation": 0.0037471705488444714, "minimum": 0.04678377673286979, "maximum": 0.05707466382910838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9522038898153824, "standard_deviation": 0.005816293621450343, "minimum": 0.9383237944431764, "maximum": 0.9576267815469838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0896908971952612, "standard_deviation": 0.03734523807404146, "minimum": 2.0364556541263505, "maximum": 2.150723206484414, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7960746064511646, "standard_deviation": 0.24509110495462827, "minimum": 0.4127435344948358, "maximum": 1.1675035995753413, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2531855542993181, "standard_deviation": 0.014093805098310791, "minimum": 0.23707369642443482, "maximum": 0.28512346636896185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9692562147983936, "standard_deviation": 0.0027002646640027304, "minimum": 0.9629112203318476, "maximum": 0.971140984800925, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8418437578060638, "standard_deviation": 0.029932534885695014, "minimum": 1.8053560633212065, "maximum": 1.895677329768622, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8519428758813326, "standard_deviation": 0.2891902053738639, "minimum": 0.5197198832966515, "maximum": 1.2701815585608598, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1991193709910576, "standard_deviation": 0.00817467223768652, "minimum": 0.19187246963963028, "maximum": 0.2159021095712923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9956016591284859, "standard_deviation": 0.0008432927836827003, "minimum": 0.9943135072125019, "maximum": 0.9965302884534586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0039347063634194, "standard_deviation": 0.04915075990171397, "minimum": 0.9491065074758759, "maximum": 1.1023617748910801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5238255408152368, "standard_deviation": 0.28609079011688354, "minimum": 0.13816320912219737, "maximum": 1.0649773745857962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07432144660198821, "standard_deviation": 0.007162772267121024, "minimum": 0.06564350949489604, "maximum": 0.08593958732506009, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984800695299595, "standard_deviation": 0.0005147708086527798, "minimum": 0.9973762832290595, "maximum": 0.9990470374525674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7817831729817611, "standard_deviation": 0.04911894646990566, "minimum": 0.7150646433416529, "maximum": 0.8543684001438351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.565350518463821, "standard_deviation": 0.3967997680433204, "minimum": 0.17518117439913958, "maximum": 1.3695454053562681, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0443925295892972, "standard_deviation": 0.007663865622330539, "minimum": 0.03482467109895333, "maximum": 0.05980540950219633, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1552725622440318, "standard_deviation": 0.03425521673270662, "minimum": 0.1083544293595697, "maximum": 0.20017362455660556, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.759607521782136, "standard_deviation": 0.007074366211164624, "minimum": 5.748497106362255, "maximum": 5.768628224412098, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.49796267282692547, "standard_deviation": 0.03418204060688387, "minimum": 0.45646229537896515, "maximum": 0.5521622975025217, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10127746632283732, "standard_deviation": 0.03282362622356819, "minimum": -0.1503157365137477, "maximum": -0.05188552611639122, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9969412530737298, "standard_deviation": 0.0005888107881066391, "minimum": 0.9961412498772222, "maximum": 0.9977297614994296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8412598750694212, "standard_deviation": 0.023889809905913422, "minimum": 0.8053511521768386, "maximum": 0.8664947652682231, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38404786992082274, "standard_deviation": 0.3101107476594251, "minimum": 0.09333782572108788, "maximum": 1.0003421890236972, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05811914359809117, "standard_deviation": 0.0056267044930854385, "minimum": 0.05059729581140704, "maximum": 0.06600949309417375, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990529776460653, "standard_deviation": 0.0002926058447497254, "minimum": 0.9984546913405569, "maximum": 0.9993732919010335, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45131986958475934, "standard_deviation": 0.02406588503748256, "minimum": 0.4215301551953602, "maximum": 0.48544612896665895, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4217447726431735, "standard_deviation": 0.3978520560263803, "minimum": 0.028520065256957255, "maximum": 1.275303415075724, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.032996829624491034, "standard_deviation": 0.005223195730045448, "minimum": 0.026738083839602602, "maximum": 0.04318369875443548, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41797-0000", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.9795260429382324, "predicted_mos": 1.9795260429382324, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "3851cc9b370000fe8b200702b5a9ac2d852a1715fc18d1c3f340642dcd403350", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9227784001268446, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8766377433095273, "全帯域_10ms_rms_envelope_nrmse": 0.055964468078092154, "全帯域_10ms_envelope_correlation": 0.9972340547723003, "全帯域_10ms_modulation_mae_db": 0.37641106935575697, "全帯域_40ms_log_rms_mae_db": 0.47435630950850916, "全帯域_40ms_rms_envelope_nrmse": 0.030911572027056942, "全帯域_40ms_envelope_correlation": 0.9992507546294856, "全帯域_40ms_modulation_mae_db": 0.680350170929629, "300–1000 Hz_10ms_log_rms_mae_db": 0.8754695073085043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07007953742097403, "300–1000 Hz_10ms_envelope_correlation": 0.9961663598556619, "300–1000 Hz_10ms_modulation_mae_db": 0.5106564096401498, "300–1000 Hz_40ms_log_rms_mae_db": 0.66152710028914, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040890673773473725, "300–1000 Hz_40ms_envelope_correlation": 0.9988082550523165, "300–1000 Hz_40ms_modulation_mae_db": 0.9242735871009856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1312099256870032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0998306789174672, "1000–3000 Hz_10ms_envelope_correlation": 0.9932573841839364, "1000–3000 Hz_10ms_modulation_mae_db": 0.7965653871595961, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8110583340575681, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05376157896901107, "1000–3000 Hz_40ms_envelope_correlation": 0.9982664849371191, "1000–3000 Hz_40ms_modulation_mae_db": 0.6306088778101943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.482938056600749, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23211500514472827, "3000–6000 Hz_10ms_envelope_correlation": 0.9586925399200039, "3000–6000 Hz_10ms_modulation_mae_db": 0.4814345556223632, "3000–6000 Hz_40ms_log_rms_mae_db": 1.403452593253958, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18979607637329918, "3000–6000 Hz_40ms_envelope_correlation": 0.9712868928366408, "3000–6000 Hz_40ms_modulation_mae_db": 0.6486987982063502, "mr_stft_db": 5.435992866322722, "band_mae_db": 0.11590291778615565, "rms_gain_error_db": -0.1215623968381594, "output_peak": 0.5129057025607064}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.888571261073876, "全帯域_10ms_rms_envelope_nrmse": 0.05728616642417684, "全帯域_10ms_envelope_correlation": 0.9970381485269928, "全帯域_10ms_modulation_mae_db": 0.3689124568633219, "全帯域_40ms_log_rms_mae_db": 0.49836117579262323, "全帯域_40ms_rms_envelope_nrmse": 0.02900094465221021, "全帯域_40ms_envelope_correlation": 0.9992985192639653, "全帯域_40ms_modulation_mae_db": 0.44173781161358766, "300–1000 Hz_10ms_log_rms_mae_db": 0.8475361589435647, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06828821132028376, "300–1000 Hz_10ms_envelope_correlation": 0.9962934081659577, "300–1000 Hz_10ms_modulation_mae_db": 0.43303017044280295, "300–1000 Hz_40ms_log_rms_mae_db": 0.6476294265877874, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036695527156644536, "300–1000 Hz_40ms_envelope_correlation": 0.9989993750723825, "300–1000 Hz_40ms_modulation_mae_db": 0.626331622711808, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1491355877876637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09917476391612776, "1000–3000 Hz_10ms_envelope_correlation": 0.9934149438458546, "1000–3000 Hz_10ms_modulation_mae_db": 0.4574289947339087, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8290674512432802, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06195019636744051, "1000–3000 Hz_40ms_envelope_correlation": 0.9976433245744228, "1000–3000 Hz_40ms_modulation_mae_db": 0.6312596728959169, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4419310978689635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23331244027233258, "3000–6000 Hz_10ms_envelope_correlation": 0.9583923835590208, "3000–6000 Hz_10ms_modulation_mae_db": 0.7407175576166969, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3898863455503783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19226668971661337, "3000–6000 Hz_40ms_envelope_correlation": 0.970467477397518, "3000–6000 Hz_40ms_modulation_mae_db": 0.6716677634999827, "mr_stft_db": 5.433487331408933, "band_mae_db": 0.18457510562959886, "rms_gain_error_db": -0.09879595542819251, "output_peak": 0.4713745934950585}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7854715873605351, "全帯域_10ms_rms_envelope_nrmse": 0.05185835378677209, "全帯域_10ms_envelope_correlation": 0.9976463200376539, "全帯域_10ms_modulation_mae_db": 0.2845551545745672, "全帯域_40ms_log_rms_mae_db": 0.414190002949919, "全帯域_40ms_rms_envelope_nrmse": 0.03073658392891684, "全帯域_40ms_envelope_correlation": 0.9992587413623515, "全帯域_40ms_modulation_mae_db": 0.22573343950222116, "300–1000 Hz_10ms_log_rms_mae_db": 0.8556539256579477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06681965621583245, "300–1000 Hz_10ms_envelope_correlation": 0.9965326427252955, "300–1000 Hz_10ms_modulation_mae_db": 0.47675021126864214, "300–1000 Hz_40ms_log_rms_mae_db": 0.6395508502978408, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03933478254492918, "300–1000 Hz_40ms_envelope_correlation": 0.9989143702850968, "300–1000 Hz_40ms_modulation_mae_db": 0.5586736846319901, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1239497687231237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08576924536242339, "1000–3000 Hz_10ms_envelope_correlation": 0.9951835520136836, "1000–3000 Hz_10ms_modulation_mae_db": 0.5210329892553661, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8044515742980479, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04788708473420307, "1000–3000 Hz_40ms_envelope_correlation": 0.9987766163028434, "1000–3000 Hz_40ms_modulation_mae_db": 0.2485949148878749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4901100306827977, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23504463194604683, "3000–6000 Hz_10ms_envelope_correlation": 0.9578393739164276, "3000–6000 Hz_10ms_modulation_mae_db": 0.46345319554326597, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4018707579173875, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19742001282329888, "3000–6000 Hz_40ms_envelope_correlation": 0.9688740623091414, "3000–6000 Hz_40ms_modulation_mae_db": 0.6721790154603425, "mr_stft_db": 5.433060853466553, "band_mae_db": 0.16751357628225563, "rms_gain_error_db": -0.11928354077417651, "output_peak": 0.472911484646256}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.865159472204134, "全帯域_10ms_rms_envelope_nrmse": 0.055541676133544916, "全帯域_10ms_envelope_correlation": 0.9971716048258282, "全帯域_10ms_modulation_mae_db": 0.12307792717388906, "全帯域_40ms_log_rms_mae_db": 0.473247121936323, "全帯域_40ms_rms_envelope_nrmse": 0.02921841543459409, "全帯域_40ms_envelope_correlation": 0.9992245922767985, "全帯域_40ms_modulation_mae_db": 0.1569976688115617, "300–1000 Hz_10ms_log_rms_mae_db": 0.8544995100368074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06919089843593731, "300–1000 Hz_10ms_envelope_correlation": 0.9961099157816791, "300–1000 Hz_10ms_modulation_mae_db": 0.1660864778953174, "300–1000 Hz_40ms_log_rms_mae_db": 0.650140970178406, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04054163530647732, "300–1000 Hz_40ms_envelope_correlation": 0.9986416709502721, "300–1000 Hz_40ms_modulation_mae_db": 0.12926584898964408, "1000–3000 Hz_10ms_log_rms_mae_db": 1.097293596443173, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08851116259504287, "1000–3000 Hz_10ms_envelope_correlation": 0.9947629048806865, "1000–3000 Hz_10ms_modulation_mae_db": 0.5299747698149047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7886722598231679, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05026237746756263, "1000–3000 Hz_40ms_envelope_correlation": 0.9985107937938598, "1000–3000 Hz_40ms_modulation_mae_db": 0.5024921899916651, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4864199245416814, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22740536465078567, "3000–6000 Hz_10ms_envelope_correlation": 0.961332070816416, "3000–6000 Hz_10ms_modulation_mae_db": 0.3653407869163552, "3000–6000 Hz_40ms_log_rms_mae_db": 1.470800338414845, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19255840853791398, "3000–6000 Hz_40ms_envelope_correlation": 0.9705909812267444, "3000–6000 Hz_40ms_modulation_mae_db": 0.484805175180192, "mr_stft_db": 5.438079968964371, "band_mae_db": 0.12591330439848744, "rms_gain_error_db": -0.0690557493708654, "output_peak": 0.5224443151690787}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9125930104024954, "全帯域_10ms_rms_envelope_nrmse": 0.05870528459091849, "全帯域_10ms_envelope_correlation": 0.9968624562409114, "全帯域_10ms_modulation_mae_db": 0.314917847640315, "全帯域_40ms_log_rms_mae_db": 0.5213787373792649, "全帯域_40ms_rms_envelope_nrmse": 0.029772238412306317, "全帯域_40ms_envelope_correlation": 0.9992277437059731, "全帯域_40ms_modulation_mae_db": 0.13552083358184347, "300–1000 Hz_10ms_log_rms_mae_db": 0.9663147814782506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07326655374585626, "300–1000 Hz_10ms_envelope_correlation": 0.9956810541872321, "300–1000 Hz_10ms_modulation_mae_db": 0.40382374096838286, "300–1000 Hz_40ms_log_rms_mae_db": 0.6967671298036409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04031776581189696, "300–1000 Hz_40ms_envelope_correlation": 0.9987230051769054, "300–1000 Hz_40ms_modulation_mae_db": 0.3953144759290934, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1132713342065466, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09408726216760906, "1000–3000 Hz_10ms_envelope_correlation": 0.9941092203990253, "1000–3000 Hz_10ms_modulation_mae_db": 0.614110652222371, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7989223759322855, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0519084716400361, "1000–3000 Hz_40ms_envelope_correlation": 0.9984586599467962, "1000–3000 Hz_40ms_modulation_mae_db": 0.5168319858741689, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4989796889182097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.249760696596666, "3000–6000 Hz_10ms_envelope_correlation": 0.9521179225361617, "3000–6000 Hz_10ms_modulation_mae_db": 0.5973155291890494, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4424939702817907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2000299515865367, "3000–6000 Hz_40ms_envelope_correlation": 0.9680915465302261, "3000–6000 Hz_40ms_modulation_mae_db": 0.49893509975750333, "mr_stft_db": 5.438276941752551, "band_mae_db": 0.12330027702780764, "rms_gain_error_db": -0.08725123230886102, "output_peak": 0.48628047565193366}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9208017792115629, "全帯域_10ms_rms_envelope_nrmse": 0.060995720635075384, "全帯域_10ms_envelope_correlation": 0.9968210781997118, "全帯域_10ms_modulation_mae_db": 0.7182750020519341, "全帯域_40ms_log_rms_mae_db": 0.524990819019326, "全帯域_40ms_rms_envelope_nrmse": 0.039388415277692684, "全帯域_40ms_envelope_correlation": 0.9988212164727338, "全帯域_40ms_modulation_mae_db": 0.887925218696569, "300–1000 Hz_10ms_log_rms_mae_db": 0.8979944109873617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07842250938563959, "300–1000 Hz_10ms_envelope_correlation": 0.9954144081440621, "300–1000 Hz_10ms_modulation_mae_db": 0.9537114189471231, "300–1000 Hz_40ms_log_rms_mae_db": 0.6786504215226808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05174971874260093, "300–1000 Hz_40ms_envelope_correlation": 0.9982481502531448, "300–1000 Hz_40ms_modulation_mae_db": 1.074088829510869, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1280728986058184, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0971173329239049, "1000–3000 Hz_10ms_envelope_correlation": 0.9935605895712252, "1000–3000 Hz_10ms_modulation_mae_db": 0.4904679785106487, "1000–3000 Hz_40ms_log_rms_mae_db": 0.815261712369166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06040916785004219, "1000–3000 Hz_40ms_envelope_correlation": 0.9975920273292941, "1000–3000 Hz_40ms_modulation_mae_db": 0.38007883186096897, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4559141401548936, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22415226819774225, "3000–6000 Hz_10ms_envelope_correlation": 0.9616611089381902, "3000–6000 Hz_10ms_modulation_mae_db": 0.5146453660530225, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3933182256424457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18374721825731027, "3000–6000 Hz_40ms_envelope_correlation": 0.9729945349787478, "3000–6000 Hz_40ms_modulation_mae_db": 0.6692528612283283, "mr_stft_db": 5.431799936708717, "band_mae_db": 0.18392578231874834, "rms_gain_error_db": -0.16602860066960104, "output_peak": 0.5039802140443006}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8336889544014279, "全帯域_10ms_rms_envelope_nrmse": 0.053734309594126, "全帯域_10ms_envelope_correlation": 0.9973467768073004, "全帯域_10ms_modulation_mae_db": 0.46469441678086926, "全帯域_40ms_log_rms_mae_db": 0.4501111635542779, "全帯域_40ms_rms_envelope_nrmse": 0.02880498372309278, "全帯域_40ms_envelope_correlation": 0.9992369990792357, "全帯域_40ms_modulation_mae_db": 0.48787120693198643, "300–1000 Hz_10ms_log_rms_mae_db": 0.8894696237682544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06627542021095113, "300–1000 Hz_10ms_envelope_correlation": 0.9964283191400548, "300–1000 Hz_10ms_modulation_mae_db": 0.6144920816776912, "300–1000 Hz_40ms_log_rms_mae_db": 0.67653911086001, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03588199784999022, "300–1000 Hz_40ms_envelope_correlation": 0.9989393606164053, "300–1000 Hz_40ms_modulation_mae_db": 0.7305380080117092, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1348178626564802, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0901894885794593, "1000–3000 Hz_10ms_envelope_correlation": 0.9945872191649766, "1000–3000 Hz_10ms_modulation_mae_db": 0.4320904210951408, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8161111315068805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0536419516225458, "1000–3000 Hz_40ms_envelope_correlation": 0.9982956025349777, "1000–3000 Hz_40ms_modulation_mae_db": 0.6645236309222321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.483615346140995, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22382959004781566, "3000–6000 Hz_10ms_envelope_correlation": 0.963710567736408, "3000–6000 Hz_10ms_modulation_mae_db": 1.1701659650015606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.38220848949297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18908347524864186, "3000–6000 Hz_40ms_envelope_correlation": 0.9725611882662726, "3000–6000 Hz_40ms_modulation_mae_db": 1.30705067887071, "mr_stft_db": 5.435064001460934, "band_mae_db": 0.13479827931709606, "rms_gain_error_db": -0.06200256908746243, "output_peak": 0.4947660130961175}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9131555232982427, "全帯域_10ms_rms_envelope_nrmse": 0.06422859841439588, "全帯域_10ms_envelope_correlation": 0.9961773906951895, "全帯域_10ms_modulation_mae_db": 0.022236129216278572, "全帯域_40ms_log_rms_mae_db": 0.5195884995360984, "全帯域_40ms_rms_envelope_nrmse": 0.03525773550509326, "全帯域_40ms_envelope_correlation": 0.9987937685624934, "全帯域_40ms_modulation_mae_db": 0.09576409771533308, "300–1000 Hz_10ms_log_rms_mae_db": 0.9187214486490076, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07803987886861, "300–1000 Hz_10ms_envelope_correlation": 0.9950636418130453, "300–1000 Hz_10ms_modulation_mae_db": 0.10981535129615454, "300–1000 Hz_40ms_log_rms_mae_db": 0.7028855778848762, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04407695240238844, "300–1000 Hz_40ms_envelope_correlation": 0.9983607638741214, "300–1000 Hz_40ms_modulation_mae_db": 0.121891656605759, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1809058130978067, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0865943857792535, "1000–3000 Hz_10ms_envelope_correlation": 0.9951477980353289, "1000–3000 Hz_10ms_modulation_mae_db": 0.7344022327340015, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8177581184000338, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05081172752312717, "1000–3000 Hz_40ms_envelope_correlation": 0.9986380237930458, "1000–3000 Hz_40ms_modulation_mae_db": 0.5972073412021156, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4637535643169783, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23463044467200098, "3000–6000 Hz_10ms_envelope_correlation": 0.9580074538553325, "3000–6000 Hz_10ms_modulation_mae_db": 0.6067369691246872, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3997495756037968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18993328836652068, "3000–6000 Hz_40ms_envelope_correlation": 0.9714242424886571, "3000–6000 Hz_40ms_modulation_mae_db": 0.5769610789158194, "mr_stft_db": 5.4511938419148045, "band_mae_db": 0.13046258333692554, "rms_gain_error_db": 0.002392514281325519, "output_peak": 0.4685883050702253}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87760, "output_samples": 87760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.059495838804595, "spectral_convergence": 0.170550829788229, "spectral_convergence_status": "ok", "amplitude_floor": 1.7869460513029852e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.192125151636012, "spectral_convergence": 0.16719765226334082, "spectral_convergence_status": "ok", "amplitude_floor": 1.6383670687778497e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.05635760852756, "spectral_convergence": 0.17171875267802492, "spectral_convergence_status": "ok", "amplitude_floor": 9.631824221799184e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.013402842456697e-05, "output_power": 9.132050221184274e-05, "delta_db": 0.056795094682620915, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00191544195180521, "output_power": 0.001843430223238487, "delta_db": -0.16642291542144, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015517446606579416, "output_power": 0.0015353753975375472, "delta_db": -0.04605682408890761, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955960266218775, "output_power": 0.0004714481413070381, "delta_db": -0.21693887954541075, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000158155871967105, "output_power": 0.00015314991315718436, "delta_db": -0.13968566037114918, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.514760940569477e-06, "output_power": 9.363669313744668e-06, "delta_db": -0.06951813260740544, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11590291778615565, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1215623968381594, "pitch": null, "output_rms_dbfs": -23.867833518540014, "output_peak": 0.5129057025607064, "reference_rms_dbfs": -23.74627112170186, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.435992866322722}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002415922668587691, "log_rms_mae_db": 0.8766377433095273, "rms_envelope_nrmse": 0.055964468078092154, "envelope_correlation": 0.9972340547723003, "rms_gain_error_db": -0.12156243423302227, "reference_mean_power": 0.004224437839878649, "output_mean_power": 0.004107832065709353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.932071872304039e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008459874532766979, "output_power": 0.0008173891073016494, "reference_active": true, "delta_db": -0.14935076210232062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.267978041368795e-05, "output_power": 8.46599134666402e-05, "reference_active": true, "delta_db": -0.3930717591153604}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0604587197812338e-05, "output_power": 1.800042874027456e-05, "reference_active": true, "delta_db": -0.5868106868495899}], "active_band_count": 3, "active_band_mae_db": 0.37641106935575697}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00018058591873130477, "log_rms_mae_db": 0.47435630950850916, "rms_envelope_nrmse": 0.030911572027056942, "envelope_correlation": 0.9992507546294856, "rms_gain_error_db": -0.1215623965800023, "reference_mean_power": 0.004247690453399677, "output_mean_power": 0.004130442880765353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8509413564750568e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008019972220007684, "output_power": 0.0007752260395939494, "reference_active": true, "delta_db": -0.14744511855178627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.130657727931199e-05, "output_power": 3.712760351935165e-05, "reference_active": true, "delta_db": -0.46322292556273564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0523532353898244e-06, "output_power": 7.570477860891384e-07, "reference_active": true, "delta_db": -1.4303824686743654}], "active_band_count": 3, "active_band_mae_db": 0.680350170929629}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002374784601360974, "log_rms_mae_db": 0.8754695073085043, "rms_envelope_nrmse": 0.07007953742097403, "envelope_correlation": 0.9961663598556619, "rms_gain_error_db": -0.15087084338340617, "reference_mean_power": 0.0028785349502786395, "output_mean_power": 0.0027802536690602134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.6369199856429154e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008090138264371756, "output_power": 0.0007757180453501102, "reference_active": true, "delta_db": -0.18252049524088157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000105081165545653, "output_power": 9.593015822983593e-05, "reference_active": true, "delta_db": -0.39569720466852415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7476383653056024e-05, "output_power": 1.4030605611856693e-05, "reference_active": true, "delta_db": -0.9537515290110437}], "active_band_count": 3, "active_band_mae_db": 0.5106564096401498}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.0001750973468816212, "log_rms_mae_db": 0.66152710028914, "rms_envelope_nrmse": 0.040890673773473725, "envelope_correlation": 0.9988082550523165, "rms_gain_error_db": -0.15087086311577108, "reference_mean_power": 0.002894380018498304, "output_mean_power": 0.002795557729373102}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.567165950582092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007701013013376472, "output_power": 0.0007392135552659662, "reference_active": true, "delta_db": -0.1777793515073326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.715572645231493e-05, "output_power": 4.195736048132008e-05, "reference_active": true, "delta_db": -0.5072628060630595}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1197000044443696e-06, "output_power": 6.923470077334804e-07, "reference_active": true, "delta_db": -2.087778603732565}], "active_band_count": 3, "active_band_mae_db": 0.9242735871009856}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.00012170766433419248, "log_rms_mae_db": 1.1312099256870032, "rms_envelope_nrmse": 0.0998306789174672, "envelope_correlation": 0.9932573841839364, "rms_gain_error_db": -0.2013582585723192, "reference_mean_power": 0.0006352125631157451, "output_mean_power": 0.0006064335961940981}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 2.616269223692633e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010398636594123316, "output_power": 9.764692173505358e-05, "reference_active": true, "delta_db": -0.2731784403740865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0847306945521593e-05, "output_power": 8.968955936052231e-06, "reference_active": true, "delta_db": -0.825800391755373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2384759945240008e-06, "output_power": 4.359256412686071e-06, "reference_active": true, "delta_db": 1.290717329349329}], "active_band_count": 3, "active_band_mae_db": 0.7965653871595961}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00010363700534457545, "log_rms_mae_db": 0.8110583340575681, "rms_envelope_nrmse": 0.05376157896901107, "envelope_correlation": 0.9982664849371191, "rms_gain_error_db": -0.20135833731305125, "reference_mean_power": 0.0006387091058518984, "output_mean_power": 0.0006097717133633705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.5308592276752824e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.945508810428335e-05, "output_power": 9.311032650674229e-05, "reference_active": true, "delta_db": -0.2862915687947554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.423063898822258e-06, "output_power": 4.693700109762134e-06, "reference_active": true, "delta_db": -0.6272938349270594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8432082747665845e-07, "output_power": 2.3088649684867076e-07, "reference_active": true, "delta_db": 0.9782412297087681}], "active_band_count": 3, "active_band_mae_db": 0.6306088778101943}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 2.9435709173680706e-05, "log_rms_mae_db": 1.482938056600749, "rms_envelope_nrmse": 0.23211500514472827, "envelope_correlation": 0.9586925399200039, "rms_gain_error_db": -0.22883572006587802, "reference_mean_power": 1.4688872454843858e-05, "output_mean_power": 1.3934933092123454e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 9.038086104701173e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4018145980274633e-06, "output_power": 3.152616000588239e-06, "reference_active": true, "delta_db": -0.33039565063503673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2285201909756879e-06, "output_power": 9.839878368229066e-07, "reference_active": true, "delta_db": -0.9639256848687794}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.73608399821043e-07, "output_power": 2.832224395609469e-07, "reference_active": true, "delta_db": 0.1499823313632736}], "active_band_count": 3, "active_band_mae_db": 0.4814345556223632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 2.1616481165467548e-05, "log_rms_mae_db": 1.403452593253958, "rms_envelope_nrmse": 0.18979607637329918, "envelope_correlation": 0.9712868928366408, "rms_gain_error_db": -0.22883578256761195, "reference_mean_power": 1.4769720024935061e-05, "output_mean_power": 1.4011630777486718e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.51604517212682e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5641156211998326e-06, "output_power": 3.273738891283503e-06, "reference_active": true, "delta_db": -0.36907746510926576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790775026435671e-07, "output_power": 4.699455613991825e-07, "reference_active": true, "delta_db": -0.9068914069617801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8657206259627155e-08, "output_power": 2.1770046919030533e-08, "reference_active": true, "delta_db": 0.6701275225480047}], "active_band_count": 3, "active_band_mae_db": 0.6486987982063502}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9056185916194324, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "aa9f89ec09b002b23391cf7f0623840ed4ceeed0f0a384a3323b6861170b96e6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9942529515118397, "standard_deviation": 0.0007775409717459505, "minimum": 0.9932573841839364, "maximum": 0.9951835520136836, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1323320984009517, "standard_deviation": 0.024867117977066623, "minimum": 1.097293596443173, "maximum": 1.1809058130978067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5720091781907422, "standard_deviation": 0.13217403941412126, "minimum": 0.4320904210951408, "maximum": 0.7965653871595961, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09265929003016099, "standard_deviation": 0.005648718298611077, "minimum": 0.08576924536242339, "maximum": 0.0998306789174672, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9982726916515449, "standard_deviation": 0.00043730583185937026, "minimum": 0.9975920273292941, "maximum": 0.9987766163028434, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8101628697038037, "standard_deviation": 0.012524643056415107, "minimum": 0.7886722598231679, "maximum": 0.8290674512432802, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5214496806806421, "standard_deviation": 0.1439243316786766, "minimum": 0.2485949148878749, "maximum": 0.6645236309222321, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05382906952174607, "standard_deviation": 0.004929669471989258, "minimum": 0.04788708473420307, "maximum": 0.06195019636744051, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9589691776597451, "standard_deviation": 0.003481806258524164, "minimum": 0.9521179225361617, "maximum": 0.963710567736408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4754577311531585, "standard_deviation": 0.0194701919551607, "minimum": 1.4419310978689635, "maximum": 1.4989796889182097, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6174762406333751, "standard_deviation": 0.2500151578661799, "minimum": 0.3653407869163552, "maximum": 1.1701659650015606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23253130519101478, "standard_deviation": 0.008285632812121241, "minimum": 0.22382959004781566, "maximum": 0.249760696596666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9707863657542435, "standard_deviation": 0.001679380854264727, "minimum": 0.9680915465302261, "maximum": 0.9729945349787478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4104725370196967, "standard_deviation": 0.030284333492723147, "minimum": 1.38220848949297, "maximum": 1.470800338414845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6911938088899034, "standard_deviation": 0.26047169854398705, "minimum": 0.484805175180192, "maximum": 1.30705067887071, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19185439011376687, "standard_deviation": 0.005068291999160858, "minimum": 0.18374721825731027, "maximum": 0.2000299515865367, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9959612187266236, "standard_deviation": 0.0005213156435156054, "minimum": 0.9950636418130453, "maximum": 0.9965326427252955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.8882074208537123, "standard_deviation": 0.03989558388745633, "minimum": 0.8475361589435647, "maximum": 0.9663147814782506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.45854573276703303, "standard_deviation": 0.2627726016427052, "minimum": 0.10981535129615454, "maximum": 0.9537114189471231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07129783320051056, "standard_deviation": 0.004786170479825764, "minimum": 0.06627542021095113, "maximum": 0.07842250938563959, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9987043689100805, "standard_deviation": 0.00027454400484463066, "minimum": 0.9982481502531448, "maximum": 0.9989993750723825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6692113234280477, "standard_deviation": 0.023312530400030892, "minimum": 0.6395508502978408, "maximum": 0.7028855778848762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5700472141864823, "standard_deviation": 0.3452229104881591, "minimum": 0.121891656605759, "maximum": 1.074088829510869, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04118613169855016, "standard_deviation": 0.004970337795373118, "minimum": 0.03588199784999022, "maximum": 0.05174971874260093, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1457989782621344, "standard_deviation": 0.028234708653338256, "minimum": 0.11590291778615565, "maximum": 0.18457510562959886, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.437119467749948, "standard_deviation": 0.006138946675853601, "minimum": 5.431799936708717, "maximum": 5.4511938419148045, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.49165638796670963, "standard_deviation": 0.020300372125329084, "minimum": 0.4685883050702253, "maximum": 0.5224443151690787, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09019844127449911, "standard_deviation": 0.0499679365275694, "minimum": -0.16602860066960104, "maximum": 0.002392514281325519, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9970372287632361, "standard_deviation": 0.00043755790246988, "minimum": 0.9961773906951895, "maximum": 0.9976463200376539, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8745099164077252, "standard_deviation": 0.04624561197498217, "minimum": 0.7854715873605351, "maximum": 0.9208017792115629, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3341350004571165, "standard_deviation": 0.2112072420416854, "minimum": 0.022236129216278572, "maximum": 0.7182750020519341, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05728932220713771, "standard_deviation": 0.00397921278901746, "minimum": 0.05185835378677209, "maximum": 0.06422859841439588, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991390419191296, "standard_deviation": 0.00020606860541392088, "minimum": 0.9987937685624934, "maximum": 0.9992985192639653, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48452797870954273, "standard_deviation": 0.03924007773747953, "minimum": 0.414190002949919, "maximum": 0.524990819019326, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38898755597284146, "standard_deviation": 0.2871256869380878, "minimum": 0.09576409771533308, "maximum": 0.887925218696569, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03163636112012039, "standard_deviation": 0.0037574145186166762, "minimum": 0.02880498372309278, "maximum": 0.039388415277692684, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41797-0000", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.3610382080078125, "predicted_mos": 2.3610382080078125, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "7c9796819fcf3e7e17114e9cea9f829febac762197d7d4338de7b6b6f07ec710", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7760886000469327, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8403889192508148, "全帯域_10ms_rms_envelope_nrmse": 0.04981563686733227, "全帯域_10ms_envelope_correlation": 0.9977039442091213, "全帯域_10ms_modulation_mae_db": 0.18462195592055283, "全帯域_40ms_log_rms_mae_db": 0.4544693218540625, "全帯域_40ms_rms_envelope_nrmse": 0.02527323250247233, "全帯域_40ms_envelope_correlation": 0.9993955193359264, "全帯域_40ms_modulation_mae_db": 0.2445856875655145, "300–1000 Hz_10ms_log_rms_mae_db": 0.9132516042255122, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061473860409789825, "300–1000 Hz_10ms_envelope_correlation": 0.9969257289906771, "300–1000 Hz_10ms_modulation_mae_db": 0.3378047495280796, "300–1000 Hz_40ms_log_rms_mae_db": 0.680290300596278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03328585533858586, "300–1000 Hz_40ms_envelope_correlation": 0.9990811249522961, "300–1000 Hz_40ms_modulation_mae_db": 0.42997442035503314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2024988187723917, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09372008878145975, "1000–3000 Hz_10ms_envelope_correlation": 0.9943280012932711, "1000–3000 Hz_10ms_modulation_mae_db": 0.3109343796098578, "1000–3000 Hz_40ms_log_rms_mae_db": 0.877479237870214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057534636113211274, "1000–3000 Hz_40ms_envelope_correlation": 0.9982280648980514, "1000–3000 Hz_40ms_modulation_mae_db": 0.26578769698435056, "3000–6000 Hz_10ms_log_rms_mae_db": 1.619818158902463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23333002676002576, "3000–6000 Hz_10ms_envelope_correlation": 0.9625488258388155, "3000–6000 Hz_10ms_modulation_mae_db": 1.2279275506545668, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4952402990798628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20139858274417838, "3000–6000 Hz_40ms_envelope_correlation": 0.9704992559659535, "3000–6000 Hz_40ms_modulation_mae_db": 1.5011471217311465, "mr_stft_db": 5.499392057149868, "band_mae_db": 0.15395067581562585, "rms_gain_error_db": -0.038832876930999134, "output_peak": 0.5017533371669768}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8933748770090059, "全帯域_10ms_rms_envelope_nrmse": 0.06718526124979267, "全帯域_10ms_envelope_correlation": 0.9959319729925418, "全帯域_10ms_modulation_mae_db": 0.26331527722661613, "全帯域_40ms_log_rms_mae_db": 0.4944357243687097, "全帯域_40ms_rms_envelope_nrmse": 0.04079646217363213, "全帯域_40ms_envelope_correlation": 0.998555613838639, "全帯域_40ms_modulation_mae_db": 0.28855444268283764, "300–1000 Hz_10ms_log_rms_mae_db": 0.9066524493691023, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08501248498527289, "300–1000 Hz_10ms_envelope_correlation": 0.9943100663281358, "300–1000 Hz_10ms_modulation_mae_db": 0.1530317693395378, "300–1000 Hz_40ms_log_rms_mae_db": 0.7031424490725713, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05580924547045784, "300–1000 Hz_40ms_envelope_correlation": 0.9976436761485091, "300–1000 Hz_40ms_modulation_mae_db": 0.1611305439739341, "1000–3000 Hz_10ms_log_rms_mae_db": 1.241005519320016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08922763034232825, "1000–3000 Hz_10ms_envelope_correlation": 0.994589093380947, "1000–3000 Hz_10ms_modulation_mae_db": 0.29564608882066845, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9055493483945486, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04769347938167332, "1000–3000 Hz_40ms_envelope_correlation": 0.9985850219838158, "1000–3000 Hz_40ms_modulation_mae_db": 0.22164585366376488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.608357164564705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2208208482622061, "3000–6000 Hz_10ms_envelope_correlation": 0.9640208694064442, "3000–6000 Hz_10ms_modulation_mae_db": 0.8457291334911354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4290149056502086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1835093089096941, "3000–6000 Hz_40ms_envelope_correlation": 0.9738345331608863, "3000–6000 Hz_40ms_modulation_mae_db": 1.0291710465247326, "mr_stft_db": 5.5091999155135225, "band_mae_db": 0.1925896389013848, "rms_gain_error_db": -0.12144309522714661, "output_peak": 0.4641447110226578}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8791322178297635, "全帯域_10ms_rms_envelope_nrmse": 0.0540790347214518, "全帯域_10ms_envelope_correlation": 0.9973219756115743, "全帯域_10ms_modulation_mae_db": 0.5417331730070472, "全帯域_40ms_log_rms_mae_db": 0.5058339058961696, "全帯域_40ms_rms_envelope_nrmse": 0.02896242526202544, "全帯域_40ms_envelope_correlation": 0.9992392418350614, "全帯域_40ms_modulation_mae_db": 0.8177735021924035, "300–1000 Hz_10ms_log_rms_mae_db": 0.9078685257939939, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06776922994153928, "300–1000 Hz_10ms_envelope_correlation": 0.9962907992062343, "300–1000 Hz_10ms_modulation_mae_db": 0.7597472603502412, "300–1000 Hz_40ms_log_rms_mae_db": 0.6677539290644221, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0371999445888086, "300–1000 Hz_40ms_envelope_correlation": 0.9988893841973805, "300–1000 Hz_40ms_modulation_mae_db": 1.1425579462563025, "1000–3000 Hz_10ms_log_rms_mae_db": 1.190799250581376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08405811603155625, "1000–3000 Hz_10ms_envelope_correlation": 0.9951839183840888, "1000–3000 Hz_10ms_modulation_mae_db": 0.3445881166949039, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8231589278957627, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.040614845979913786, "1000–3000 Hz_40ms_envelope_correlation": 0.999008016057468, "1000–3000 Hz_40ms_modulation_mae_db": 0.26292653978222763, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6806706609194848, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23241594145220015, "3000–6000 Hz_10ms_envelope_correlation": 0.9597938957310562, "3000–6000 Hz_10ms_modulation_mae_db": 0.34145205221664426, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5150845702566968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18708941328628156, "3000–6000 Hz_40ms_envelope_correlation": 0.9726345429141835, "3000–6000 Hz_40ms_modulation_mae_db": 0.45392839981740746, "mr_stft_db": 5.506953245476885, "band_mae_db": 0.16013358660405905, "rms_gain_error_db": -0.06863645930512063, "output_peak": 0.5119897719181961}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8345155545969353, "全帯域_10ms_rms_envelope_nrmse": 0.05329930357708096, "全帯域_10ms_envelope_correlation": 0.9975747546795362, "全帯域_10ms_modulation_mae_db": 0.3856956843660293, "全帯域_40ms_log_rms_mae_db": 0.45182617810435327, "全帯域_40ms_rms_envelope_nrmse": 0.030435117475633437, "全帯域_40ms_envelope_correlation": 0.9993531019791727, "全帯域_40ms_modulation_mae_db": 0.4515329898779891, "300–1000 Hz_10ms_log_rms_mae_db": 0.9065585254976041, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0688133870596484, "300–1000 Hz_10ms_envelope_correlation": 0.9964510860558962, "300–1000 Hz_10ms_modulation_mae_db": 0.5642218008502837, "300–1000 Hz_40ms_log_rms_mae_db": 0.7005851051476619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04317802770796006, "300–1000 Hz_40ms_envelope_correlation": 0.9987954204845626, "300–1000 Hz_40ms_modulation_mae_db": 0.598214315320244, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2165387780471215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08995508409953615, "1000–3000 Hz_10ms_envelope_correlation": 0.994567818811719, "1000–3000 Hz_10ms_modulation_mae_db": 0.519464716700127, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8865000843265011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0512563806917823, "1000–3000 Hz_40ms_envelope_correlation": 0.9984317229542293, "1000–3000 Hz_40ms_modulation_mae_db": 0.7726868090722405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5906162311907548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23577655421383484, "3000–6000 Hz_10ms_envelope_correlation": 0.9574324398857188, "3000–6000 Hz_10ms_modulation_mae_db": 0.46074826880284725, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4598350208318656, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19585420905582226, "3000–6000 Hz_40ms_envelope_correlation": 0.9698486376075118, "3000–6000 Hz_40ms_modulation_mae_db": 0.6056460673384642, "mr_stft_db": 5.521787257999651, "band_mae_db": 0.17855344939891382, "rms_gain_error_db": -0.14305281295760394, "output_peak": 0.48852215682312583}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8191109842393893, "全帯域_10ms_rms_envelope_nrmse": 0.05953528289973011, "全帯域_10ms_envelope_correlation": 0.9967620139700639, "全帯域_10ms_modulation_mae_db": 0.44265018317752663, "全帯域_40ms_log_rms_mae_db": 0.4582041686356765, "全帯域_40ms_rms_envelope_nrmse": 0.031912022475934665, "全帯域_40ms_envelope_correlation": 0.9990900295772568, "全帯域_40ms_modulation_mae_db": 0.5742688772885924, "300–1000 Hz_10ms_log_rms_mae_db": 0.9253020378947152, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07699512027413716, "300–1000 Hz_10ms_envelope_correlation": 0.9952282550520751, "300–1000 Hz_10ms_modulation_mae_db": 0.5069286666409849, "300–1000 Hz_40ms_log_rms_mae_db": 0.7226635986717445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043634305825636564, "300–1000 Hz_40ms_envelope_correlation": 0.9984958949988694, "300–1000 Hz_40ms_modulation_mae_db": 0.7489404120893587, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2071134586632413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08660574537865193, "1000–3000 Hz_10ms_envelope_correlation": 0.9949549931931713, "1000–3000 Hz_10ms_modulation_mae_db": 0.3412668566837445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8312026346029061, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04926454486552358, "1000–3000 Hz_40ms_envelope_correlation": 0.9985269783566147, "1000–3000 Hz_40ms_modulation_mae_db": 0.2931308090154798, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6196645312037252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2216069552410025, "3000–6000 Hz_10ms_envelope_correlation": 0.9634084174159847, "3000–6000 Hz_10ms_modulation_mae_db": 0.6767213283439689, "3000–6000 Hz_40ms_log_rms_mae_db": 1.492955380483996, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18809260363457334, "3000–6000 Hz_40ms_envelope_correlation": 0.9721752963927843, "3000–6000 Hz_40ms_modulation_mae_db": 0.911566214986402, "mr_stft_db": 5.505195346563902, "band_mae_db": 0.2281391150377369, "rms_gain_error_db": -0.08351115775150499, "output_peak": 0.52164738505368}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8198251568821316, "全帯域_10ms_rms_envelope_nrmse": 0.05986637029745743, "全帯域_10ms_envelope_correlation": 0.996861401606195, "全帯域_10ms_modulation_mae_db": 0.7892444687634913, "全帯域_40ms_log_rms_mae_db": 0.45434975468842376, "全帯域_40ms_rms_envelope_nrmse": 0.0352754091502322, "全帯域_40ms_envelope_correlation": 0.9990264352558931, "全帯域_40ms_modulation_mae_db": 0.7453822659436046, "300–1000 Hz_10ms_log_rms_mae_db": 0.9488310652737385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07589206389927271, "300–1000 Hz_10ms_envelope_correlation": 0.9956665824351847, "300–1000 Hz_10ms_modulation_mae_db": 0.8472797658397567, "300–1000 Hz_40ms_log_rms_mae_db": 0.695390800278264, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04985704801006236, "300–1000 Hz_40ms_envelope_correlation": 0.9983348399310615, "300–1000 Hz_40ms_modulation_mae_db": 0.9701364605270032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1703029946866976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08924269900536978, "1000–3000 Hz_10ms_envelope_correlation": 0.9944914193647008, "1000–3000 Hz_10ms_modulation_mae_db": 0.19772281955328164, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8104502086625985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.044400217614325756, "1000–3000 Hz_40ms_envelope_correlation": 0.9986896528933906, "1000–3000 Hz_40ms_modulation_mae_db": 0.33844536788378604, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6108665807379814, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22883847916108102, "3000–6000 Hz_10ms_envelope_correlation": 0.9616291647903135, "3000–6000 Hz_10ms_modulation_mae_db": 0.7860599454588081, "3000–6000 Hz_40ms_log_rms_mae_db": 1.508420519470714, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19043740794963315, "3000–6000 Hz_40ms_envelope_correlation": 0.9719978469197678, "3000–6000 Hz_40ms_modulation_mae_db": 1.1694024107638519, "mr_stft_db": 5.526326246411036, "band_mae_db": 0.21174927014196074, "rms_gain_error_db": -0.14046913507800765, "output_peak": 0.4797205703275153}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8614089755333791, "全帯域_10ms_rms_envelope_nrmse": 0.0654245092635465, "全帯域_10ms_envelope_correlation": 0.996185723703329, "全帯域_10ms_modulation_mae_db": 0.6878315617085442, "全帯域_40ms_log_rms_mae_db": 0.5017873892496459, "全帯域_40ms_rms_envelope_nrmse": 0.0386726410786637, "全帯域_40ms_envelope_correlation": 0.9987589842774027, "全帯域_40ms_modulation_mae_db": 0.7612182485203182, "300–1000 Hz_10ms_log_rms_mae_db": 0.8947098224121129, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07852009043707947, "300–1000 Hz_10ms_envelope_correlation": 0.9951997414521383, "300–1000 Hz_10ms_modulation_mae_db": 0.9849350121882178, "300–1000 Hz_40ms_log_rms_mae_db": 0.6575357442891051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048764624840599236, "300–1000 Hz_40ms_envelope_correlation": 0.9982814539685108, "300–1000 Hz_40ms_modulation_mae_db": 1.0400622434339741, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2023924266691384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0929384295746873, "1000–3000 Hz_10ms_envelope_correlation": 0.9941461645121098, "1000–3000 Hz_10ms_modulation_mae_db": 0.3423808741647889, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8800774011237934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.046372149119878484, "1000–3000 Hz_40ms_envelope_correlation": 0.9987408755781416, "1000–3000 Hz_40ms_modulation_mae_db": 0.43476857269737734, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6415422131083697, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22531760891608807, "3000–6000 Hz_10ms_envelope_correlation": 0.9623283256987214, "3000–6000 Hz_10ms_modulation_mae_db": 0.7436060018857119, "3000–6000 Hz_40ms_log_rms_mae_db": 1.497738386668867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.187331236435198, "3000–6000 Hz_40ms_envelope_correlation": 0.9724971948579477, "3000–6000 Hz_40ms_modulation_mae_db": 0.9216405010556775, "mr_stft_db": 5.5105866886041275, "band_mae_db": 0.12929344013690525, "rms_gain_error_db": -0.1337921045605644, "output_peak": 0.46019250060375844}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8807188408972398, "全帯域_10ms_rms_envelope_nrmse": 0.0565809041365377, "全帯域_10ms_envelope_correlation": 0.9970439323312046, "全帯域_10ms_modulation_mae_db": 0.3307651198294476, "全帯域_40ms_log_rms_mae_db": 0.4656456902376322, "全帯域_40ms_rms_envelope_nrmse": 0.03339751736177972, "全帯域_40ms_envelope_correlation": 0.9989224710960176, "全帯域_40ms_modulation_mae_db": 0.33905751503323717, "300–1000 Hz_10ms_log_rms_mae_db": 0.879290628305538, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07220555723425792, "300–1000 Hz_10ms_envelope_correlation": 0.995756231405182, "300–1000 Hz_10ms_modulation_mae_db": 0.3773608885996629, "300–1000 Hz_40ms_log_rms_mae_db": 0.6451893779079922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04495007715597294, "300–1000 Hz_40ms_envelope_correlation": 0.9982760205425969, "300–1000 Hz_40ms_modulation_mae_db": 0.27748915198408314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1484589538044978, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08334144779714166, "1000–3000 Hz_10ms_envelope_correlation": 0.9953316626336549, "1000–3000 Hz_10ms_modulation_mae_db": 0.5222205345085947, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8224236366643437, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.044580105816491876, "1000–3000 Hz_40ms_envelope_correlation": 0.9988314235571915, "1000–3000 Hz_40ms_modulation_mae_db": 0.6628033293656692, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6846884367401271, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25020485676312854, "3000–6000 Hz_10ms_envelope_correlation": 0.951987295621959, "3000–6000 Hz_10ms_modulation_mae_db": 0.9226653095396206, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5361984635102328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20985732133340448, "3000–6000 Hz_40ms_envelope_correlation": 0.9646396685424585, "3000–6000 Hz_40ms_modulation_mae_db": 1.0564185436556766, "mr_stft_db": 5.5128623142267585, "band_mae_db": 0.1377797642462155, "rms_gain_error_db": 0.002420742446910849, "output_peak": 0.4783591256001628}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87760, "output_samples": 87760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.178773667873688, "spectral_convergence": 0.16319704375264268, "spectral_convergence_status": "ok", "amplitude_floor": 1.7869460513029852e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.282205346551909, "spectral_convergence": 0.16243354465530374, "spectral_convergence_status": "ok", "amplitude_floor": 1.6383670687778497e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.037197157024008, "spectral_convergence": 0.16880866425741284, "spectral_convergence_status": "ok", "amplitude_floor": 9.631824221799184e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.013402842456697e-05, "output_power": 9.777232679443595e-05, "delta_db": 0.3532716895208818, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00191544195180521, "output_power": 0.0019025336232114604, "delta_db": -0.02936654323282086, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015517446606579416, "output_power": 0.0015554555404146272, "delta_db": 0.010373424712076893, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955960266218775, "output_power": 0.0004680933894015802, "delta_db": -0.24795307923443716, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000158155871967105, "output_power": 0.00014974611273946858, "delta_db": -0.23729763561713546, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.514760940569477e-06, "output_power": 9.41572385069815e-06, "delta_db": -0.04544168257640303, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15395067581562585, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.038832876930999134, "pitch": null, "output_rms_dbfs": -23.785103998632856, "output_peak": 0.5017533371669768, "reference_rms_dbfs": -23.74627112170186, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.499392057149868}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002415922668587691, "log_rms_mae_db": 0.8403889192508148, "rms_envelope_nrmse": 0.04981563686733227, "envelope_correlation": 0.9977039442091213, "rms_gain_error_db": -0.03883297065970351, "reference_mean_power": 0.004224437839878649, "output_mean_power": 0.004186832889000929}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.932071872304039e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008459874532766979, "output_power": 0.0008395128294794051, "reference_active": true, "delta_db": -0.033365846891971696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.267978041368795e-05, "output_power": 9.007370025663704e-05, "reference_active": true, "delta_db": -0.12386992077221429}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0604587197812338e-05, "output_power": 1.880619383566304e-05, "reference_active": true, "delta_db": -0.39663010009747246}], "active_band_count": 3, "active_band_mae_db": 0.18462195592055283}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00018058591873130477, "log_rms_mae_db": 0.4544693218540625, "rms_envelope_nrmse": 0.02527323250247233, "envelope_correlation": 0.9993955193359264, "rms_gain_error_db": -0.03883342206954433, "reference_mean_power": 0.004247690453399677, "output_mean_power": 0.0042098780756531515}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8509413564750568e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008019972220007684, "output_power": 0.0007959640393924201, "reference_active": true, "delta_db": -0.03279416627359613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.130657727931199e-05, "output_power": 4.008224865269472e-05, "reference_active": true, "delta_db": -0.13067132339348533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0523532353898244e-06, "output_power": 9.228526917710751e-07, "reference_active": true, "delta_db": -0.570291573029462}], "active_band_count": 3, "active_band_mae_db": 0.2445856875655145}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002374784601360974, "log_rms_mae_db": 0.9132516042255122, "rms_envelope_nrmse": 0.061473860409789825, "envelope_correlation": 0.9969257289906771, "rms_gain_error_db": -0.057195441871745176, "reference_mean_power": 0.0028785349502786395, "output_mean_power": 0.002840873938652587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.6369199856429154e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008090138264371756, "output_power": 0.0008043297595378309, "reference_active": true, "delta_db": -0.025218064407169605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000105081165545653, "output_power": 0.00010328721610067098, "reference_active": true, "delta_db": -0.07478309227510796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7476383653056024e-05, "output_power": 1.416153274443812e-05, "reference_active": true, "delta_db": -0.9134130919019613}], "active_band_count": 3, "active_band_mae_db": 0.3378047495280796}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.0001750973468816212, "log_rms_mae_db": 0.680290300596278, "rms_envelope_nrmse": 0.03328585533858586, "envelope_correlation": 0.9990811249522961, "rms_gain_error_db": -0.057195452551304586, "reference_mean_power": 0.002894380018498304, "output_mean_power": 0.0028565116925514146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.567165950582092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007701013013376472, "output_power": 0.0007661584603303985, "reference_active": true, "delta_db": -0.022292556180705296}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.715572645231493e-05, "output_power": 4.56990665394979e-05, "reference_active": true, "delta_db": -0.13627110563298475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1197000044443696e-06, "output_power": 8.629104333144374e-07, "reference_active": true, "delta_db": -1.1313595992514094}], "active_band_count": 3, "active_band_mae_db": 0.42997442035503314}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.00012170766433419248, "log_rms_mae_db": 1.2024988187723917, "rms_envelope_nrmse": 0.09372008878145975, "envelope_correlation": 0.9943280012932711, "rms_gain_error_db": -0.2663669134266241, "reference_mean_power": 0.0006352125631157451, "output_mean_power": 0.0005974236179278575}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 2.616269223692633e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010398636594123316, "output_power": 9.654725459139313e-05, "reference_active": true, "delta_db": -0.32236472250713244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0847306945521593e-05, "output_power": 9.740245422708306e-06, "reference_active": true, "delta_db": -0.4675202969871601}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2384759945240008e-06, "output_power": 3.346821067351719e-06, "reference_active": true, "delta_db": 0.14291811933528087}], "active_band_count": 3, "active_band_mae_db": 0.3109343796098578}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00010363700534457545, "log_rms_mae_db": 0.877479237870214, "rms_envelope_nrmse": 0.057534636113211274, "envelope_correlation": 0.9982280648980514, "rms_gain_error_db": -0.266366994123706, "reference_mean_power": 0.0006387091058518984, "output_mean_power": 0.0006007121393500682}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.5308592276752824e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.945508810428335e-05, "output_power": 9.192689096690586e-05, "reference_active": true, "delta_db": -0.3418443424021486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.423063898822258e-06, "output_power": 5.220895139529522e-06, "reference_active": true, "delta_db": -0.16499751072054075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8432082747665845e-07, "output_power": 1.7239405600448657e-07, "reference_active": true, "delta_db": -0.2905212378303623}], "active_band_count": 3, "active_band_mae_db": 0.26578769698435056}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 2.9435709173680706e-05, "log_rms_mae_db": 1.619818158902463, "rms_envelope_nrmse": 0.23333002676002576, "envelope_correlation": 0.9625488258388155, "rms_gain_error_db": 0.35484894497295183, "reference_mean_power": 1.4688872454843858e-05, "output_mean_power": 1.59394508287472e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 9.038086104701173e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4018145980274633e-06, "output_power": 3.7570064827532387e-06, "reference_active": true, "delta_db": 0.4313130396416989}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2285201909756879e-06, "output_power": 1.5687696006444129e-06, "reference_active": true, "delta_db": 1.0617686646360067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.73608399821043e-07, "output_power": 4.531056951581219e-07, "reference_active": true, "delta_db": 2.190700947685995}], "active_band_count": 3, "active_band_mae_db": 1.2279275506545668}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 2.1616481165467548e-05, "log_rms_mae_db": 1.4952402990798628, "rms_envelope_nrmse": 0.20139858274417838, "envelope_correlation": 0.9704992559659535, "rms_gain_error_db": 0.3548489844792858, "reference_mean_power": 1.4769720024935061e-05, "output_mean_power": 1.602718172937304e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.51604517212682e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5641156211998326e-06, "output_power": 4.13220633305998e-06, "reference_active": true, "delta_db": 0.6423021471002611}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790775026435671e-07, "output_power": 7.646338452753802e-07, "reference_active": true, "delta_db": 1.2071682467449758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8657206259627155e-08, "output_power": 3.437507939522209e-08, "reference_active": true, "delta_db": 2.6539709713482025}], "active_band_count": 3, "active_band_mae_db": 1.5011471217311465}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9080200234607657, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d742b6e2101da4291cd0a2988aafbc85c61c78df499d12616ff0dd75f708b5f6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9946991339467078, "standard_deviation": 0.0004168907059502239, "minimum": 0.9941461645121098, "maximum": 0.9953316626336549, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.19738877506806, "standard_deviation": 0.028249020921042024, "minimum": 1.1484589538044978, "maximum": 1.241005519320016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.35927804834199584, "standard_deviation": 0.11053245331946104, "minimum": 0.19772281955328164, "maximum": 0.5222205345085947, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08863615512634138, "standard_deviation": 0.0037782874102647596, "minimum": 0.08334144779714166, "maximum": 0.09372008878145975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9986302195348629, "standard_deviation": 0.0002428465591546141, "minimum": 0.9982280648980514, "maximum": 0.999008016057468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8546051849425835, "standard_deviation": 0.03646232677235048, "minimum": 0.8104502086625985, "maximum": 0.9055493483945486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.406524372308112, "standard_deviation": 0.20446852212023675, "minimum": 0.22164585366376488, "maximum": 0.7726868090722405, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.047714544947850046, "standard_deviation": 0.005132514849635109, "minimum": 0.040614845979913786, "maximum": 0.057534636113211274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9603936542986267, "standard_deviation": 0.003999197373055655, "minimum": 0.951987295621959, "maximum": 0.9640208694064442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6320279971709515, "standard_deviation": 0.03434017763840818, "minimum": 1.5906162311907548, "maximum": 1.6846884367401271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.750613698799163, "standard_deviation": 0.2737772981872344, "minimum": 0.34145205221664426, "maximum": 1.2279275506545668, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23103890884619588, "standard_deviation": 0.009470781819252226, "minimum": 0.2208208482622061, "maximum": 0.25020485676312854, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9710158720451867, "standard_deviation": 0.002862394155972836, "minimum": 0.9646396685424585, "maximum": 0.9738345331608863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4918109432440556, "standard_deviation": 0.033384766516075956, "minimum": 1.4290149056502086, "maximum": 1.5361984635102328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9561150382341699, "standard_deviation": 0.32428877333995276, "minimum": 0.45392839981740746, "maximum": 1.5011471217311465, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19294626041859816, "standard_deviation": 0.008869058820831296, "minimum": 0.1835093089096941, "maximum": 0.20985732133340448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9957285613656904, "standard_deviation": 0.0008296346647345479, "minimum": 0.9943100663281358, "maximum": 0.9969257289906771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9103080823465396, "standard_deviation": 0.020550515012704322, "minimum": 0.879290628305538, "maximum": 0.9488310652737385, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5664137391670956, "standard_deviation": 0.2812864866999885, "minimum": 0.1530317693395378, "maximum": 0.9849350121882178, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0733352242801247, "standard_deviation": 0.007334545730321281, "minimum": 0.061473860409789825, "maximum": 0.08501248498527289, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984747269029733, "standard_deviation": 0.00045198041895638005, "minimum": 0.9976436761485091, "maximum": 0.9990811249522961, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6840689131285049, "standard_deviation": 0.02605050912126518, "minimum": 0.6451893779079922, "maximum": 0.7226635986717445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6710631867424915, "standard_deviation": 0.36495398335186674, "minimum": 0.1611305439739341, "maximum": 1.1425579462563025, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04458489111726043, "standard_deviation": 0.0071361636912108005, "minimum": 0.03328585533858586, "maximum": 0.05580924547045784, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17402361753535023, "standard_deviation": 0.035106348795986106, "minimum": 0.12929344013690525, "maximum": 0.2281391150377369, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.511537883993219, "standard_deviation": 0.00878828860574125, "minimum": 5.499392057149868, "maximum": 5.526326246411036, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.48829119481450917, "standard_deviation": 0.022039676893579642, "minimum": 0.46019250060375844, "maximum": 0.52164738505368, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09091461242050457, "standard_deviation": 0.05334474556110029, "minimum": -0.14305281295760394, "maximum": 0.002420742446910849, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9969232148879458, "standard_deviation": 0.0006283635832294749, "minimum": 0.9959319729925418, "maximum": 0.9977039442091213, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8535594407798324, "standard_deviation": 0.029030315202147744, "minimum": 0.8191109842393893, "maximum": 0.8933748770090059, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4532321779999069, "standard_deviation": 0.20828029672482776, "minimum": 0.18462195592055283, "maximum": 0.7892444687634913, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05822328787661618, "standard_deviation": 0.00599287770640171, "minimum": 0.04981563686733227, "maximum": 0.06718526124979267, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990426746494212, "standard_deviation": 0.00029134337135837277, "minimum": 0.998555613838639, "maximum": 0.9993955193359264, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4733190166293342, "standard_deviation": 0.023231770662433485, "minimum": 0.45182617810435327, "maximum": 0.5058339058961696, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5277966911380622, "standard_deviation": 0.2289601573468829, "minimum": 0.2445856875655145, "maximum": 0.8177735021924035, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0330906034350467, "standard_deviation": 0.005099030403753398, "minimum": 0.02527323250247233, "maximum": 0.04079646217363213, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41797-0000", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.1801841259002686, "predicted_mos": 2.1801841259002686, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "172409dfefefcdc4870b6f9048c1e91cd74d53349dcfc9e358f3fc5c2a3da8fd", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.888434300199151, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8290869400416463, "全帯域_10ms_rms_envelope_nrmse": 0.06325490935063223, "全帯域_10ms_envelope_correlation": 0.9962879740503745, "全帯域_10ms_modulation_mae_db": 0.302013888085881, "全帯域_40ms_log_rms_mae_db": 0.4630859286625164, "全帯域_40ms_rms_envelope_nrmse": 0.035980625958901134, "全帯域_40ms_envelope_correlation": 0.9987587401548519, "全帯域_40ms_modulation_mae_db": 0.7559746625971456, "300–1000 Hz_10ms_log_rms_mae_db": 0.9466454519954833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08008545225994845, "300–1000 Hz_10ms_envelope_correlation": 0.9947465805953202, "300–1000 Hz_10ms_modulation_mae_db": 0.3031629894591504, "300–1000 Hz_40ms_log_rms_mae_db": 0.7343955315768869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045815077708361615, "300–1000 Hz_40ms_envelope_correlation": 0.9982164167402161, "300–1000 Hz_40ms_modulation_mae_db": 0.9588289379696938, "1000–3000 Hz_10ms_log_rms_mae_db": 1.158927027672861, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09273961464455763, "1000–3000 Hz_10ms_envelope_correlation": 0.9942218284156971, "1000–3000 Hz_10ms_modulation_mae_db": 0.5194154650046618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8005386732992708, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05417306907059657, "1000–3000 Hz_40ms_envelope_correlation": 0.9982167427184622, "1000–3000 Hz_40ms_modulation_mae_db": 0.5604722013721183, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5722346271028902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24297353090690296, "3000–6000 Hz_10ms_envelope_correlation": 0.9549895803575069, "3000–6000 Hz_10ms_modulation_mae_db": 0.32941148197380615, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4336420890679211, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19234168039511812, "3000–6000 Hz_40ms_envelope_correlation": 0.9706554622602701, "3000–6000 Hz_40ms_modulation_mae_db": 0.4359477840720234, "mr_stft_db": 5.4994818216204235, "band_mae_db": 0.20438078318557631, "rms_gain_error_db": -0.04328839113226012, "output_peak": 0.5110863185689147}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.842411076053551, "全帯域_10ms_rms_envelope_nrmse": 0.05935186944651391, "全帯域_10ms_envelope_correlation": 0.9967675860180341, "全帯域_10ms_modulation_mae_db": 0.2761766696048561, "全帯域_40ms_log_rms_mae_db": 0.4588144615895259, "全帯域_40ms_rms_envelope_nrmse": 0.033095534213763725, "全帯域_40ms_envelope_correlation": 0.9989945874956297, "全帯域_40ms_modulation_mae_db": 0.5187398161379141, "300–1000 Hz_10ms_log_rms_mae_db": 0.9145289918191061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07424227603057326, "300–1000 Hz_10ms_envelope_correlation": 0.9955522994953542, "300–1000 Hz_10ms_modulation_mae_db": 0.29007940631677326, "300–1000 Hz_40ms_log_rms_mae_db": 0.6961910713255501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0410654086318872, "300–1000 Hz_40ms_envelope_correlation": 0.9986584534277944, "300–1000 Hz_40ms_modulation_mae_db": 0.6831801527932857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1819049251723033, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09192800798111019, "1000–3000 Hz_10ms_envelope_correlation": 0.994160789081288, "1000–3000 Hz_10ms_modulation_mae_db": 0.3400855407639097, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8048085842206881, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05200578412770427, "1000–3000 Hz_40ms_envelope_correlation": 0.9981601841792775, "1000–3000 Hz_40ms_modulation_mae_db": 0.3544927203715837, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5662163362203227, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22452328837692367, "3000–6000 Hz_10ms_envelope_correlation": 0.962056249183698, "3000–6000 Hz_10ms_modulation_mae_db": 0.8278047940780606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4407768157548075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18482140367742173, "3000–6000 Hz_40ms_envelope_correlation": 0.9729274552696852, "3000–6000 Hz_40ms_modulation_mae_db": 1.175025025371451, "mr_stft_db": 5.487035266009767, "band_mae_db": 0.13970478609226114, "rms_gain_error_db": -0.0726565394914743, "output_peak": 0.5063240725650877}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8510186924418319, "全帯域_10ms_rms_envelope_nrmse": 0.0636258436826461, "全帯域_10ms_envelope_correlation": 0.9964166970016234, "全帯域_10ms_modulation_mae_db": 0.31010969907188196, "全帯域_40ms_log_rms_mae_db": 0.4178636370572113, "全帯域_40ms_rms_envelope_nrmse": 0.04214605889767774, "全帯域_40ms_envelope_correlation": 0.9984952911083598, "全帯域_40ms_modulation_mae_db": 0.38600107655964494, "300–1000 Hz_10ms_log_rms_mae_db": 0.8979660528562776, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0797736063001839, "300–1000 Hz_10ms_envelope_correlation": 0.9951093230350994, "300–1000 Hz_10ms_modulation_mae_db": 0.4187158463628931, "300–1000 Hz_40ms_log_rms_mae_db": 0.6711188172222805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05362611987966139, "300–1000 Hz_40ms_envelope_correlation": 0.997939142916455, "300–1000 Hz_40ms_modulation_mae_db": 0.5910656909749116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.145954199782628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08195423671478047, "1000–3000 Hz_10ms_envelope_correlation": 0.9954461176983399, "1000–3000 Hz_10ms_modulation_mae_db": 0.367287135975028, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7827274092544642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04463295081180958, "1000–3000 Hz_40ms_envelope_correlation": 0.9987774079896229, "1000–3000 Hz_40ms_modulation_mae_db": 0.3362203878897759, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6696956445329392, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22725073459600592, "3000–6000 Hz_10ms_envelope_correlation": 0.960545125306513, "3000–6000 Hz_10ms_modulation_mae_db": 0.6105943088108398, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5313266589660688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19106466983616005, "3000–6000 Hz_40ms_envelope_correlation": 0.9714132989709141, "3000–6000 Hz_40ms_modulation_mae_db": 0.6101970420677835, "mr_stft_db": 5.50866702038192, "band_mae_db": 0.17779763741812885, "rms_gain_error_db": -0.13842457183818974, "output_peak": 0.4996963556769951}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8635780487786547, "全帯域_10ms_rms_envelope_nrmse": 0.05961046442503667, "全帯域_10ms_envelope_correlation": 0.9967755347375903, "全帯域_10ms_modulation_mae_db": 0.2778136240438681, "全帯域_40ms_log_rms_mae_db": 0.4757705876255025, "全帯域_40ms_rms_envelope_nrmse": 0.03605558614892249, "全帯域_40ms_envelope_correlation": 0.9988392942600121, "全帯域_40ms_modulation_mae_db": 0.3261765955515684, "300–1000 Hz_10ms_log_rms_mae_db": 0.9419660610561001, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07692825986818505, "300–1000 Hz_10ms_envelope_correlation": 0.9952668012765694, "300–1000 Hz_10ms_modulation_mae_db": 0.5327847290982491, "300–1000 Hz_40ms_log_rms_mae_db": 0.6675717352507782, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04767276271525224, "300–1000 Hz_40ms_envelope_correlation": 0.998211206513589, "300–1000 Hz_40ms_modulation_mae_db": 0.5806457376250959, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1745226033852962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09026837614925166, "1000–3000 Hz_10ms_envelope_correlation": 0.9945160638296064, "1000–3000 Hz_10ms_modulation_mae_db": 0.40557344883328766, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8074294146810694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05258756905764057, "1000–3000 Hz_40ms_envelope_correlation": 0.9983143883754573, "1000–3000 Hz_40ms_modulation_mae_db": 0.40634502134736533, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6247077604696945, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22633504571871302, "3000–6000 Hz_10ms_envelope_correlation": 0.9612080816961369, "3000–6000 Hz_10ms_modulation_mae_db": 1.1238828697189307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4484411020498544, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18714415028550566, "3000–6000 Hz_40ms_envelope_correlation": 0.9721663934916388, "3000–6000 Hz_40ms_modulation_mae_db": 1.5084963254916015, "mr_stft_db": 5.495458327801287, "band_mae_db": 0.19987402365693005, "rms_gain_error_db": -0.09554785611112881, "output_peak": 0.5454687330407906}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8254887528545167, "全帯域_10ms_rms_envelope_nrmse": 0.0566928590850433, "全帯域_10ms_envelope_correlation": 0.9970573284833524, "全帯域_10ms_modulation_mae_db": 0.16923557815963353, "全帯域_40ms_log_rms_mae_db": 0.4609266607464274, "全帯域_40ms_rms_envelope_nrmse": 0.03242198160299517, "全帯域_40ms_envelope_correlation": 0.9990377496758717, "全帯域_40ms_modulation_mae_db": 0.2324452110916756, "300–1000 Hz_10ms_log_rms_mae_db": 0.91182235631691, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07459420508501746, "300–1000 Hz_10ms_envelope_correlation": 0.9954698302499538, "300–1000 Hz_10ms_modulation_mae_db": 0.18398249335468284, "300–1000 Hz_40ms_log_rms_mae_db": 0.6963790957738134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04507987307407867, "300–1000 Hz_40ms_envelope_correlation": 0.9983032203240348, "300–1000 Hz_40ms_modulation_mae_db": 0.30041877935795785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2223349486678656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09535818862255106, "1000–3000 Hz_10ms_envelope_correlation": 0.9941425669119018, "1000–3000 Hz_10ms_modulation_mae_db": 0.6634377238493817, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8666541898876833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05762557193574788, "1000–3000 Hz_40ms_envelope_correlation": 0.9982646177524949, "1000–3000 Hz_40ms_modulation_mae_db": 0.9113345330695789, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6174298298092629, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22575752950460984, "3000–6000 Hz_10ms_envelope_correlation": 0.9611294199354916, "3000–6000 Hz_10ms_modulation_mae_db": 0.7893842298040538, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4604994065128754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18696626124195592, "3000–6000 Hz_40ms_envelope_correlation": 0.9721103423076253, "3000–6000 Hz_40ms_modulation_mae_db": 1.0872648329628658, "mr_stft_db": 5.504707466245442, "band_mae_db": 0.17373593162944156, "rms_gain_error_db": -0.0728726140934326, "output_peak": 0.49032485345352567}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8214790353249547, "全帯域_10ms_rms_envelope_nrmse": 0.057844774039034576, "全帯域_10ms_envelope_correlation": 0.9969642243541077, "全帯域_10ms_modulation_mae_db": 0.656762602382091, "全帯域_40ms_log_rms_mae_db": 0.4508268689927406, "全帯域_40ms_rms_envelope_nrmse": 0.030948416233071076, "全帯域_40ms_envelope_correlation": 0.9991685308861571, "全帯域_40ms_modulation_mae_db": 0.698798799710492, "300–1000 Hz_10ms_log_rms_mae_db": 0.9059897301108839, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07387488712511477, "300–1000 Hz_10ms_envelope_correlation": 0.9956581782315558, "300–1000 Hz_10ms_modulation_mae_db": 0.7771451082381665, "300–1000 Hz_40ms_log_rms_mae_db": 0.6748852270805059, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04311801950997606, "300–1000 Hz_40ms_envelope_correlation": 0.9985796357284371, "300–1000 Hz_40ms_modulation_mae_db": 0.7267042907778257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2286782858855678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08448210601334175, "1000–3000 Hz_10ms_envelope_correlation": 0.9950921521011148, "1000–3000 Hz_10ms_modulation_mae_db": 0.35406772157445027, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8351255338537087, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04632551210276492, "1000–3000 Hz_40ms_envelope_correlation": 0.9985994867514071, "1000–3000 Hz_40ms_modulation_mae_db": 0.5106461586439878, "3000–6000 Hz_10ms_log_rms_mae_db": 1.608309049871408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2351283429816132, "3000–6000 Hz_10ms_envelope_correlation": 0.9577868406562156, "3000–6000 Hz_10ms_modulation_mae_db": 0.31803241830909085, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4719313008971067, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1901036006811651, "3000–6000 Hz_40ms_envelope_correlation": 0.9713195803949449, "3000–6000 Hz_40ms_modulation_mae_db": 0.6116103001236803, "mr_stft_db": 5.497055982584224, "band_mae_db": 0.14387337014453574, "rms_gain_error_db": -0.09142939418175813, "output_peak": 0.49196371833745184}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8619989395028195, "全帯域_10ms_rms_envelope_nrmse": 0.05474824335029005, "全帯域_10ms_envelope_correlation": 0.9972514458140644, "全帯域_10ms_modulation_mae_db": 0.40166843884887443, "全帯域_40ms_log_rms_mae_db": 0.4574832801364111, "全帯域_40ms_rms_envelope_nrmse": 0.027162458342920755, "全帯域_40ms_envelope_correlation": 0.9993365434848366, "全帯域_40ms_modulation_mae_db": 0.521179233727298, "300–1000 Hz_10ms_log_rms_mae_db": 0.9488956842952022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0669875565939998, "300–1000 Hz_10ms_envelope_correlation": 0.9963806161009909, "300–1000 Hz_10ms_modulation_mae_db": 0.49463880773684127, "300–1000 Hz_40ms_log_rms_mae_db": 0.7138978665696585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03512987942043235, "300–1000 Hz_40ms_envelope_correlation": 0.9990226577953525, "300–1000 Hz_40ms_modulation_mae_db": 0.6788713652477102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2200340808003072, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08863517570672735, "1000–3000 Hz_10ms_envelope_correlation": 0.9947233315039142, "1000–3000 Hz_10ms_modulation_mae_db": 0.32896254948042936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8598414942536782, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04887580175399618, "1000–3000 Hz_40ms_envelope_correlation": 0.9985913323920085, "1000–3000 Hz_40ms_modulation_mae_db": 0.38089538615353985, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6157568559521391, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2380784586583334, "3000–6000 Hz_10ms_envelope_correlation": 0.9567350961342926, "3000–6000 Hz_10ms_modulation_mae_db": 0.26563444048476953, "3000–6000 Hz_40ms_log_rms_mae_db": 1.47571458211844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1923402643053025, "3000–6000 Hz_40ms_envelope_correlation": 0.9706080606805295, "3000–6000 Hz_40ms_modulation_mae_db": 0.6174331623525962, "mr_stft_db": 5.495929709416778, "band_mae_db": 0.13716936926575393, "rms_gain_error_db": -0.06782482624307089, "output_peak": 0.5058214699063048}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8342963749026873, "全帯域_10ms_rms_envelope_nrmse": 0.06523298234660567, "全帯域_10ms_envelope_correlation": 0.9960615536333479, "全帯域_10ms_modulation_mae_db": 0.3043973588390328, "全帯域_40ms_log_rms_mae_db": 0.4422666641058226, "全帯域_40ms_rms_envelope_nrmse": 0.043695298175714395, "全帯域_40ms_envelope_correlation": 0.9981746619026761, "全帯域_40ms_modulation_mae_db": 0.15337557779074384, "300–1000 Hz_10ms_log_rms_mae_db": 0.9715370299082134, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08033956239914357, "300–1000 Hz_10ms_envelope_correlation": 0.9947419639456111, "300–1000 Hz_10ms_modulation_mae_db": 0.2599243799227949, "300–1000 Hz_40ms_log_rms_mae_db": 0.7565707395826792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0536302798761568, "300–1000 Hz_40ms_envelope_correlation": 0.9975922543724319, "300–1000 Hz_40ms_modulation_mae_db": 0.2474840528242487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1284383638987268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07976919732413482, "1000–3000 Hz_10ms_envelope_correlation": 0.9957152230807869, "1000–3000 Hz_10ms_modulation_mae_db": 0.6512858868934185, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7873226608536695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04753954596337827, "1000–3000 Hz_40ms_envelope_correlation": 0.9986053304159325, "1000–3000 Hz_40ms_modulation_mae_db": 0.6986927387227534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5833859653461926, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22555563862442016, "3000–6000 Hz_10ms_envelope_correlation": 0.9615707323857476, "3000–6000 Hz_10ms_modulation_mae_db": 0.24220013509644045, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4468901599684614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1818757575716272, "3000–6000 Hz_40ms_envelope_correlation": 0.9737986902149806, "3000–6000 Hz_40ms_modulation_mae_db": 0.2823147965202882, "mr_stft_db": 5.497447769664167, "band_mae_db": 0.15053668195876888, "rms_gain_error_db": -0.057496030721571736, "output_peak": 0.4702475165802076}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87760, "output_samples": 87760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.180044976928791, "spectral_convergence": 0.17121562834104118, "spectral_convergence_status": "ok", "amplitude_floor": 1.7869460513029852e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2758141935710166, "spectral_convergence": 0.16599734416601217, "spectral_convergence_status": "ok", "amplitude_floor": 1.6383670687778497e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.042586294361464, "spectral_convergence": 0.17002856425047974, "spectral_convergence_status": "ok", "amplitude_floor": 9.631824221799184e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.013402842456697e-05, "output_power": 9.771605182314649e-05, "delta_db": 0.35077129432110576, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00191544195180521, "output_power": 0.0018775239285288097, "delta_db": -0.08683514383708486, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015517446606579416, "output_power": 0.0015688418536760568, "delta_db": 0.04758907241934869, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955960266218775, "output_power": 0.00048322719257220626, "delta_db": -0.10976450527444384, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000158155871967105, "output_power": 0.00014156897394788958, "delta_db": -0.4811723648573169, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.514760940569477e-06, "output_power": 9.84947569620609e-06, "delta_db": 0.15015231840415788, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20438078318557631, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04328839113226012, "pitch": null, "output_rms_dbfs": -23.789559512834117, "output_peak": 0.5110863185689147, "reference_rms_dbfs": -23.74627112170186, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.4994818216204235}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002415922668587691, "log_rms_mae_db": 0.8290869400416463, "rms_envelope_nrmse": 0.06325490935063223, "envelope_correlation": 0.9962879740503745, "rms_gain_error_db": -0.04328833481841882, "reference_mean_power": 0.004224437839878649, "output_mean_power": 0.004182539880253349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.932071872304039e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008459874532766979, "output_power": 0.0008323858331849605, "reference_active": true, "delta_db": -0.0703924201034266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.267978041368795e-05, "output_power": 7.904625955246179e-05, "reference_active": true, "delta_db": -0.6910367208725179}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0604587197812338e-05, "output_power": 1.992978739287682e-05, "reference_active": true, "delta_db": -0.14461252328169844}], "active_band_count": 3, "active_band_mae_db": 0.302013888085881}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00018058591873130477, "log_rms_mae_db": 0.4630859286625164, "rms_envelope_nrmse": 0.035980625958901134, "envelope_correlation": 0.9987587401548519, "rms_gain_error_db": -0.043288077714541254, "reference_mean_power": 0.004247690453399677, "output_mean_power": 0.004205562123406094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8509413564750568e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008019972220007684, "output_power": 0.0007880208835310463, "reference_active": true, "delta_db": -0.07635136970574886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.130657727931199e-05, "output_power": 3.413732007865874e-05, "reference_active": true, "delta_db": -0.8278978602289586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0523532353898244e-06, "output_power": 7.687658179543988e-07, "reference_active": true, "delta_db": -1.3636747578567294}], "active_band_count": 3, "active_band_mae_db": 0.7559746625971456}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002374784601360974, "log_rms_mae_db": 0.9466454519954833, "rms_envelope_nrmse": 0.08008545225994845, "envelope_correlation": 0.9947465805953202, "rms_gain_error_db": -0.04971853531455919, "reference_mean_power": 0.0028785349502786395, "output_mean_power": 0.0028457690604648824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.6369199856429154e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008090138264371756, "output_power": 0.0007948871816783434, "reference_active": true, "delta_db": -0.07650450354984227}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000105081165545653, "output_power": 8.811087567446371e-05, "reference_active": true, "delta_db": -0.7649536385130764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7476383653056024e-05, "output_power": 1.720475479753464e-05, "reference_active": true, "delta_db": -0.06803082631453236}], "active_band_count": 3, "active_band_mae_db": 0.3031629894591504}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.0001750973468816212, "log_rms_mae_db": 0.7343955315768869, "rms_envelope_nrmse": 0.045815077708361615, "envelope_correlation": 0.9982164167402161, "rms_gain_error_db": -0.04971857628123062, "reference_mean_power": 0.002894380018498304, "output_mean_power": 0.002861433739887861}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.567165950582092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007701013013376472, "output_power": 0.0007554866196985915, "reference_active": true, "delta_db": -0.08321080250299429}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.715572645231493e-05, "output_power": 3.804502639275083e-05, "reference_active": true, "delta_db": -0.9323654999940411}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1197000044443696e-06, "output_power": 7.294753629944665e-07, "reference_active": true, "delta_db": -1.8609105114120459}], "active_band_count": 3, "active_band_mae_db": 0.9588289379696938}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.00012170766433419248, "log_rms_mae_db": 1.158927027672861, "rms_envelope_nrmse": 0.09273961464455763, "envelope_correlation": 0.9942218284156971, "rms_gain_error_db": -0.19079068668216564, "reference_mean_power": 0.0006352125631157451, "output_mean_power": 0.0006079110116513393}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 2.616269223692633e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010398636594123316, "output_power": 9.680759662207613e-05, "reference_active": true, "delta_db": -0.3106696266820654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0847306945521593e-05, "output_power": 8.760831295969845e-06, "reference_active": true, "delta_db": -0.9277661216329602}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2384759945240008e-06, "output_power": 3.4859547616113874e-06, "reference_active": true, "delta_db": 0.31981064669895976}], "active_band_count": 3, "active_band_mae_db": 0.5194154650046618}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00010363700534457545, "log_rms_mae_db": 0.8005386732992708, "rms_envelope_nrmse": 0.05417306907059657, "envelope_correlation": 0.9982167427184622, "rms_gain_error_db": -0.19079076680310877, "reference_mean_power": 0.0006387091058518984, "output_mean_power": 0.0006112572610681136}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.5308592276752824e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.945508810428335e-05, "output_power": 9.258240100041143e-05, "reference_active": true, "delta_db": -0.31098567017995726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.423063898822258e-06, "output_power": 4.453652459677435e-06, "reference_active": true, "delta_db": -0.8552839777225294}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8432082747665845e-07, "output_power": 1.637041600399212e-07, "reference_active": true, "delta_db": -0.5151469562138682}], "active_band_count": 3, "active_band_mae_db": 0.5604722013721183}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 2.9435709173680706e-05, "log_rms_mae_db": 1.5722346271028902, "rms_envelope_nrmse": 0.24297353090690296, "envelope_correlation": 0.9549895803575069, "rms_gain_error_db": -0.07622672340708855, "reference_mean_power": 1.4688872454843858e-05, "output_mean_power": 1.443330494634319e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 9.038086104701173e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4018145980274633e-06, "output_power": 3.2036433827221063e-06, "reference_active": true, "delta_db": -0.2606647437054607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2285201909756879e-06, "output_power": 1.057756295439494e-06, "reference_active": true, "delta_db": -0.6499667979371148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.73608399821043e-07, "output_power": 2.785413759378069e-07, "reference_active": true, "delta_db": 0.07760290427884307}], "active_band_count": 3, "active_band_mae_db": 0.32941148197380615}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 2.1616481165467548e-05, "log_rms_mae_db": 1.4336420890679211, "rms_envelope_nrmse": 0.19234168039511812, "envelope_correlation": 0.9706554622602701, "rms_gain_error_db": -0.07622729036177063, "reference_mean_power": 1.4769720024935061e-05, "output_mean_power": 1.451274397800466e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.51604517212682e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5641156211998326e-06, "output_power": 3.3222537264610523e-06, "reference_active": true, "delta_db": -0.3051898703485413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790775026435671e-07, "output_power": 5.215796524062561e-07, "reference_active": true, "delta_db": -0.454160521897103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8657206259627155e-08, "output_power": 2.1168783866148757e-08, "reference_active": true, "delta_db": 0.5484929599704259}], "active_band_count": 3, "active_band_mae_db": 0.4359477840720234}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9020582565548585, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3a80ac1be2f72022df6341385040abd5275f2f3fd7ac71221a43e6be07b646f1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9947522590778312, "standard_deviation": 0.0006075489051067238, "minimum": 0.9941425669119018, "maximum": 0.9957152230807869, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1825993044081944, "standard_deviation": 0.03780763256258165, "minimum": 1.1284383638987268, "maximum": 1.2286782858855678, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4537644340468209, "standard_deviation": 0.13914340004977618, "minimum": 0.32896254948042936, "maximum": 0.6634377238493817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08814186289455686, "standard_deviation": 0.005530397245626194, "minimum": 0.07976919732413482, "maximum": 0.09535818862255106, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9984411863218328, "standard_deviation": 0.00022809223573235432, "minimum": 0.9981601841792775, "maximum": 0.9987774079896229, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8180559950380291, "standard_deviation": 0.03204261249966472, "minimum": 0.7827274092544642, "maximum": 0.8666541898876833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5198873934463379, "standard_deviation": 0.20008368136444515, "minimum": 0.3362203878897759, "maximum": 0.9113345330695789, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05047072560295478, "standard_deviation": 0.004377452124598382, "minimum": 0.04463295081180958, "maximum": 0.05762557193574788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9595026407069502, "standard_deviation": 0.0026297914014024463, "minimum": 0.9549895803575069, "maximum": 0.962056249183698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6072170086631061, "standard_deviation": 0.033510489586644876, "minimum": 1.5662163362203227, "maximum": 1.6696956445329392, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.563368084784499, "standard_deviation": 0.32604484080214413, "minimum": 0.24220013509644045, "maximum": 1.1238828697189307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23070032117094028, "standard_deviation": 0.007017182244553684, "minimum": 0.22452328837692367, "maximum": 0.24297353090690296, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9718749104488236, "standard_deviation": 0.0011059971944072443, "minimum": 0.9706080606805295, "maximum": 0.9737986902149806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.463652764416942, "standard_deviation": 0.03103005576165169, "minimum": 1.4336420890679211, "maximum": 1.5313266589660688, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7910361586202862, "standard_deviation": 0.4191603122264933, "minimum": 0.2823147965202882, "maximum": 1.5084963254916015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18833222349928203, "standard_deviation": 0.0037802750231199977, "minimum": 0.1818757575716272, "maximum": 0.19234168039511812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9953656991163069, "standard_deviation": 0.0005357406484783306, "minimum": 0.9947419639456111, "maximum": 0.9963806161009909, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.929918919794772, "standard_deviation": 0.025842994324719117, "minimum": 0.8979660528562776, "maximum": 0.9715370299082134, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4075542200611939, "standard_deviation": 0.19141024002394708, "minimum": 0.18398249335468284, "maximum": 0.7771451082381665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07585322570777078, "standard_deviation": 0.004492187585229361, "minimum": 0.0669875565939998, "maximum": 0.08033956239914357, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983153734772889, "standard_deviation": 0.0004428851935057712, "minimum": 0.9975922543724319, "maximum": 0.9990226577953525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.701376260547769, "standard_deviation": 0.03185992757806958, "minimum": 0.6675717352507782, "maximum": 0.7565707395826792, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5958998759463412, "standard_deviation": 0.23061660703647283, "minimum": 0.2474840528242487, "maximum": 0.9588289379696938, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.045642177601975795, "standard_deviation": 0.0062088539350095865, "minimum": 0.03512987942043235, "maximum": 0.0536302798761568, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16588407291892454, "standard_deviation": 0.02691468231112547, "minimum": 0.13716936926575393, "maximum": 0.20438078318557631, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.498222920465501, "standard_deviation": 0.0064668538792793445, "minimum": 5.487035266009767, "maximum": 5.50866702038192, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5026166297661598, "standard_deviation": 0.02155694880322784, "minimum": 0.4702475165802076, "maximum": 0.5454687330407906, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07994252797661079, "standard_deviation": 0.029014535293975807, "minimum": -0.13842457183818974, "maximum": -0.04328839113226012, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9966977930115619, "standard_deviation": 0.00040873756430987005, "minimum": 0.9960615536333479, "maximum": 0.9972514458140644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8411697324875828, "standard_deviation": 0.016306812970122393, "minimum": 0.8214790353249547, "maximum": 0.8635780487786547, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33727223237951487, "standard_deviation": 0.14381882966214465, "minimum": 0.16923557815963353, "maximum": 0.656762602382091, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06004524321572531, "standard_deviation": 0.003680913027335315, "minimum": 0.05474824335029005, "maximum": 0.06523298234660567, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988506748710494, "standard_deviation": 0.00037505832781785977, "minimum": 0.9981746619026761, "maximum": 0.9993365434848366, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4533797611145197, "standard_deviation": 0.01728047032821046, "minimum": 0.4178636370572113, "maximum": 0.4757705876255025, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44908637164581033, "standard_deviation": 0.2138258759264471, "minimum": 0.15337557779074384, "maximum": 0.7559746625971456, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.035188244946745814, "standard_deviation": 0.005562403479433029, "minimum": 0.027162458342920755, "maximum": 0.043695298175714395, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_4992-41806-0000", "language": "en", "style": "normal", "speaker": "librispeech_4992", "source_utterance_id": "4992-41806-0000", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_4992-41806-0000.wav", "source_file_sha256": "00ac6bb3f09d3c9acfa680aad38256f9bad29db9664cc522f7cefe07cc276b5a", "decoded_pcm_sha256": "9e17e42f2de437c240d5850615912f710137bb8bb0f2e42082d93bc758515234", "original_source_sha256": "ce4dd04d130b3b575f1826c6505b5519c09a7301eee9b76e58eeb1f591131035", "corpus_entry_id": "librispeech_4992-41806-0000"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.29105281829834, "predicted_mos": 4.29105281829834, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d3117e8022d7a4b21392ff93373a9fc410b97f793c7b132c978c4013e925a9bd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2760890340032347, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_4992-41806-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d3117e8022d7a4b21392ff93373a9fc410b97f793c7b132c978c4013e925a9bd", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "31ad5018e5e57c365e8a2ba88f046ad939b81e19a335769d1abb4bd9cc8ca7c3"}, "rows": [{"case": "librispeech_4992-41806-0000", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.7210562229156494, "predicted_mos": 2.7210562229156494, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6d776dce6cfc569371d73a9e4512f23bd52c5bee92520fb633e1020312010815", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2512716350029223, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_4992-41806-0000", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6d776dce6cfc569371d73a9e4512f23bd52c5bee92520fb633e1020312010815", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.860302182522901, "全帯域_10ms_rms_envelope_nrmse": 0.07892030244190043, "全帯域_10ms_envelope_correlation": 0.993978247443152, "全帯域_10ms_modulation_mae_db": 0.162804398105875, "全帯域_40ms_log_rms_mae_db": 0.4819617396236542, "全帯域_40ms_rms_envelope_nrmse": 0.04186108224851523, "全帯域_40ms_envelope_correlation": 0.9981761525973084, "全帯域_40ms_modulation_mae_db": 0.2075130651101058, "300–1000 Hz_10ms_log_rms_mae_db": 1.2737748600070369, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09956881958732217, "300–1000 Hz_10ms_envelope_correlation": 0.9918102129068713, "300–1000 Hz_10ms_modulation_mae_db": 0.2786023974040238, "300–1000 Hz_40ms_log_rms_mae_db": 0.8530887251055521, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05160536929161956, "300–1000 Hz_40ms_envelope_correlation": 0.997637285853372, "300–1000 Hz_40ms_modulation_mae_db": 0.3048392488726788, "1000–3000 Hz_10ms_log_rms_mae_db": 1.419860467422227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13250030259360868, "1000–3000 Hz_10ms_envelope_correlation": 0.9873143418133714, "1000–3000 Hz_10ms_modulation_mae_db": 0.5069920780868279, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8984281199974924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07293396695593635, "1000–3000 Hz_40ms_envelope_correlation": 0.9960009146967592, "1000–3000 Hz_40ms_modulation_mae_db": 0.55392340623846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6859481308953888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4247886860442077, "3000–6000 Hz_10ms_envelope_correlation": 0.8878902178776725, "3000–6000 Hz_10ms_modulation_mae_db": 1.366707017483871, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2813301468433789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3500903760810466, "3000–6000 Hz_40ms_envelope_correlation": 0.9182705420852922, "3000–6000 Hz_40ms_modulation_mae_db": 1.5668682950300639, "mr_stft_db": 5.674962661711402, "band_mae_db": 0.13332320893191682, "rms_gain_error_db": -0.028088229228717468, "output_peak": 0.597174802474431}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8541322139729006, "全帯域_10ms_rms_envelope_nrmse": 0.0980326936340883, "全帯域_10ms_envelope_correlation": 0.990712905275784, "全帯域_10ms_modulation_mae_db": 0.8934827831787343, "全帯域_40ms_log_rms_mae_db": 0.4825430899680464, "全帯域_40ms_rms_envelope_nrmse": 0.048391506641037474, "全帯域_40ms_envelope_correlation": 0.9976068097382571, "全帯域_40ms_modulation_mae_db": 1.1116924655775484, "300–1000 Hz_10ms_log_rms_mae_db": 1.311994465146582, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12298868092249547, "300–1000 Hz_10ms_envelope_correlation": 0.9875929046199087, "300–1000 Hz_10ms_modulation_mae_db": 0.9838683265782979, "300–1000 Hz_40ms_log_rms_mae_db": 0.839484951675349, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06120099650923726, "300–1000 Hz_40ms_envelope_correlation": 0.9969188919961463, "300–1000 Hz_40ms_modulation_mae_db": 1.153937732131762, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4346444337032418, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17878919861565876, "1000–3000 Hz_10ms_envelope_correlation": 0.9779400851218424, "1000–3000 Hz_10ms_modulation_mae_db": 1.3058168807111064, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9092282597993856, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12624245535509224, "1000–3000 Hz_40ms_envelope_correlation": 0.9885788129078004, "1000–3000 Hz_40ms_modulation_mae_db": 1.535519909459076, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7205356926628979, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4154257625347618, "3000–6000 Hz_10ms_envelope_correlation": 0.8920889361086944, "3000–6000 Hz_10ms_modulation_mae_db": 1.143462007318518, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3032073486753544, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3393850302833094, "3000–6000 Hz_40ms_envelope_correlation": 0.9226822811616935, "3000–6000 Hz_40ms_modulation_mae_db": 1.2351541247049584, "mr_stft_db": 5.685309896704871, "band_mae_db": 0.1869828925601694, "rms_gain_error_db": -0.07781499855869463, "output_peak": 0.6118127046185288}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8021445199609983, "全帯域_10ms_rms_envelope_nrmse": 0.07800061383540953, "全帯域_10ms_envelope_correlation": 0.9941518150767014, "全帯域_10ms_modulation_mae_db": 0.12006303149976001, "全帯域_40ms_log_rms_mae_db": 0.4378018738930019, "全帯域_40ms_rms_envelope_nrmse": 0.04328795497790983, "全帯域_40ms_envelope_correlation": 0.99809170511701, "全帯域_40ms_modulation_mae_db": 0.0783457564981368, "300–1000 Hz_10ms_log_rms_mae_db": 1.2505243114647826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10132604287272955, "300–1000 Hz_10ms_envelope_correlation": 0.9915700262550168, "300–1000 Hz_10ms_modulation_mae_db": 0.2300334203471616, "300–1000 Hz_40ms_log_rms_mae_db": 0.8286859128331453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059979295339465166, "300–1000 Hz_40ms_envelope_correlation": 0.9968687708362719, "300–1000 Hz_40ms_modulation_mae_db": 0.054943361841328815, "1000–3000 Hz_10ms_log_rms_mae_db": 1.446626927360296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15645975379435248, "1000–3000 Hz_10ms_envelope_correlation": 0.9823582409454858, "1000–3000 Hz_10ms_modulation_mae_db": 0.6160692602514386, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9498833427629878, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11075013639578742, "1000–3000 Hz_40ms_envelope_correlation": 0.9908468592839448, "1000–3000 Hz_40ms_modulation_mae_db": 0.6387137569434481, "3000–6000 Hz_10ms_log_rms_mae_db": 1.728685435498653, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.42076933987134146, "3000–6000 Hz_10ms_envelope_correlation": 0.8970597927549294, "3000–6000 Hz_10ms_modulation_mae_db": 0.305569190602393, "3000–6000 Hz_40ms_log_rms_mae_db": 1.308807320979435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34326690960616363, "3000–6000 Hz_40ms_envelope_correlation": 0.9275232377102152, "3000–6000 Hz_40ms_modulation_mae_db": 0.25500163764284794, "mr_stft_db": 5.687632778390097, "band_mae_db": 0.2677911942595987, "rms_gain_error_db": -0.06293759031072324, "output_peak": 0.5923567069577741}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8404941832315705, "全帯域_10ms_rms_envelope_nrmse": 0.08096500191702781, "全帯域_10ms_envelope_correlation": 0.9936559986689737, "全帯域_10ms_modulation_mae_db": 0.19585821102645165, "全帯域_40ms_log_rms_mae_db": 0.47812351261652963, "全帯域_40ms_rms_envelope_nrmse": 0.04241940955554231, "全帯域_40ms_envelope_correlation": 0.9981462717950551, "全帯域_40ms_modulation_mae_db": 0.20426618563276114, "300–1000 Hz_10ms_log_rms_mae_db": 1.2281537330664614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10754143607502058, "300–1000 Hz_10ms_envelope_correlation": 0.9904476889598227, "300–1000 Hz_10ms_modulation_mae_db": 0.22940008181046942, "300–1000 Hz_40ms_log_rms_mae_db": 0.8027043985775446, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05490193342014471, "300–1000 Hz_40ms_envelope_correlation": 0.9973815900972411, "300–1000 Hz_40ms_modulation_mae_db": 0.2857061507328525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3944864747156542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12785499074131806, "1000–3000 Hz_10ms_envelope_correlation": 0.988319841051229, "1000–3000 Hz_10ms_modulation_mae_db": 0.4046037278055213, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8637737466853962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0653458976918908, "1000–3000 Hz_40ms_envelope_correlation": 0.9970537589831859, "1000–3000 Hz_40ms_modulation_mae_db": 0.22019176285711797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7602156450477995, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4247058760224514, "3000–6000 Hz_10ms_envelope_correlation": 0.8946328631172101, "3000–6000 Hz_10ms_modulation_mae_db": 0.6343272832295417, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3470533743591073, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.345232071889254, "3000–6000 Hz_40ms_envelope_correlation": 0.9263897136003106, "3000–6000 Hz_40ms_modulation_mae_db": 0.7876008136139578, "mr_stft_db": 5.674083699836722, "band_mae_db": 0.25195324512712647, "rms_gain_error_db": -0.05497029151668719, "output_peak": 0.59509692009914}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8752437321434775, "全帯域_10ms_rms_envelope_nrmse": 0.08439172992958824, "全帯域_10ms_envelope_correlation": 0.9930939386651616, "全帯域_10ms_modulation_mae_db": 0.1805186463939513, "全帯域_40ms_log_rms_mae_db": 0.47854064876741764, "全帯域_40ms_rms_envelope_nrmse": 0.044314517783521176, "全帯域_40ms_envelope_correlation": 0.9979399004268412, "全帯域_40ms_modulation_mae_db": 0.14102411597831022, "300–1000 Hz_10ms_log_rms_mae_db": 1.2045651473751149, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10443807648269313, "300–1000 Hz_10ms_envelope_correlation": 0.9909866283121721, "300–1000 Hz_10ms_modulation_mae_db": 0.1545794762054592, "300–1000 Hz_40ms_log_rms_mae_db": 0.7935640434683615, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05830192181929606, "300–1000 Hz_40ms_envelope_correlation": 0.9969776668637008, "300–1000 Hz_40ms_modulation_mae_db": 0.13663557064433932, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4412040725711863, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13924241664894427, "1000–3000 Hz_10ms_envelope_correlation": 0.9860009248661831, "1000–3000 Hz_10ms_modulation_mae_db": 0.9249494680987763, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9274269925172774, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06982228610015094, "1000–3000 Hz_40ms_envelope_correlation": 0.9964196762855916, "1000–3000 Hz_40ms_modulation_mae_db": 0.8130730998737113, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7429700501528953, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4101220090370069, "3000–6000 Hz_10ms_envelope_correlation": 0.903372944495179, "3000–6000 Hz_10ms_modulation_mae_db": 0.35153153763903605, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3309372584141506, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32974048467225536, "3000–6000 Hz_40ms_envelope_correlation": 0.9343623641586846, "3000–6000 Hz_40ms_modulation_mae_db": 0.3102976616897777, "mr_stft_db": 5.691396851768512, "band_mae_db": 0.2486269215867746, "rms_gain_error_db": -0.029649605186242614, "output_peak": 0.6863979229542898}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8321749390327963, "全帯域_10ms_rms_envelope_nrmse": 0.08301022704331723, "全帯域_10ms_envelope_correlation": 0.9933406212601836, "全帯域_10ms_modulation_mae_db": 0.20584680581304884, "全帯域_40ms_log_rms_mae_db": 0.47557567027719233, "全帯域_40ms_rms_envelope_nrmse": 0.04415147780641495, "全帯域_40ms_envelope_correlation": 0.9980005767049261, "全帯域_40ms_modulation_mae_db": 0.41951481042407535, "300–1000 Hz_10ms_log_rms_mae_db": 1.2615841542088682, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10257707025101438, "300–1000 Hz_10ms_envelope_correlation": 0.9913008705747289, "300–1000 Hz_10ms_modulation_mae_db": 0.3331131397387529, "300–1000 Hz_40ms_log_rms_mae_db": 0.8193843331031088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05582653158945413, "300–1000 Hz_40ms_envelope_correlation": 0.9972396861489531, "300–1000 Hz_40ms_modulation_mae_db": 0.5960829136506004, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4028391697792968, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12715011769166676, "1000–3000 Hz_10ms_envelope_correlation": 0.9889994381999784, "1000–3000 Hz_10ms_modulation_mae_db": 0.34507563765481947, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9090871388242225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07620606588078326, "1000–3000 Hz_40ms_envelope_correlation": 0.996606221759937, "1000–3000 Hz_40ms_modulation_mae_db": 0.2387384815593595, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7714381183860604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4158044507143842, "3000–6000 Hz_10ms_envelope_correlation": 0.9019848515281726, "3000–6000 Hz_10ms_modulation_mae_db": 0.2565630904052955, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3106742278211667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33572671469198934, "3000–6000 Hz_40ms_envelope_correlation": 0.9331063767272404, "3000–6000 Hz_40ms_modulation_mae_db": 0.3972787011408028, "mr_stft_db": 5.6878564760563615, "band_mae_db": 0.23185120039274754, "rms_gain_error_db": -0.06930031763310986, "output_peak": 0.7237760879939561}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8377248857260164, "全帯域_10ms_rms_envelope_nrmse": 0.0947425582696822, "全帯域_10ms_envelope_correlation": 0.9913717898390806, "全帯域_10ms_modulation_mae_db": 0.6055178211498625, "全帯域_40ms_log_rms_mae_db": 0.43329099776109065, "全帯域_40ms_rms_envelope_nrmse": 0.05496473803674961, "全帯域_40ms_envelope_correlation": 0.9969684604378907, "全帯域_40ms_modulation_mae_db": 0.6449710489333156, "300–1000 Hz_10ms_log_rms_mae_db": 1.2987841317872897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12687072402822586, "300–1000 Hz_10ms_envelope_correlation": 0.9868359814966958, "300–1000 Hz_10ms_modulation_mae_db": 0.5546591278574855, "300–1000 Hz_40ms_log_rms_mae_db": 0.86275008519228, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08165643846006339, "300–1000 Hz_40ms_envelope_correlation": 0.9943390030209156, "300–1000 Hz_40ms_modulation_mae_db": 0.696110647308852, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4013309969049457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13646354915077502, "1000–3000 Hz_10ms_envelope_correlation": 0.9867131902134131, "1000–3000 Hz_10ms_modulation_mae_db": 0.5126571115303773, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9063166243754148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.079511544559449, "1000–3000 Hz_40ms_envelope_correlation": 0.9952372877643577, "1000–3000 Hz_40ms_modulation_mae_db": 0.6712542148937505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6935192276902018, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.406984710488575, "3000–6000 Hz_10ms_envelope_correlation": 0.9101161677405168, "3000–6000 Hz_10ms_modulation_mae_db": 0.34772296132290137, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2445543233431449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32860904070153235, "3000–6000 Hz_40ms_envelope_correlation": 0.9396465586204658, "3000–6000 Hz_40ms_modulation_mae_db": 0.5205162404209931, "mr_stft_db": 5.672336981098726, "band_mae_db": 0.26377116771024034, "rms_gain_error_db": -0.11622837439506818, "output_peak": 0.5904023605540463}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8362206403581669, "全帯域_10ms_rms_envelope_nrmse": 0.07991947552251531, "全帯域_10ms_envelope_correlation": 0.9938190012205632, "全帯域_10ms_modulation_mae_db": 0.4777776522167467, "全帯域_40ms_log_rms_mae_db": 0.46120474096188546, "全帯域_40ms_rms_envelope_nrmse": 0.04225834146772481, "全帯域_40ms_envelope_correlation": 0.9981470415498446, "全帯域_40ms_modulation_mae_db": 0.5756455838560088, "300–1000 Hz_10ms_log_rms_mae_db": 1.2688003373322108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10498974684165449, "300–1000 Hz_10ms_envelope_correlation": 0.9909299056776861, "300–1000 Hz_10ms_modulation_mae_db": 0.4631280463357557, "300–1000 Hz_40ms_log_rms_mae_db": 0.8589386512537934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05573308133560242, "300–1000 Hz_40ms_envelope_correlation": 0.9973438524184641, "300–1000 Hz_40ms_modulation_mae_db": 0.6062952909112261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.398390177163452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14496668728416165, "1000–3000 Hz_10ms_envelope_correlation": 0.985310010009555, "1000–3000 Hz_10ms_modulation_mae_db": 0.7996682771423144, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9040802768529765, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09709784572799204, "1000–3000 Hz_40ms_envelope_correlation": 0.9931253933769821, "1000–3000 Hz_40ms_modulation_mae_db": 0.8780212032650839, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7795001201516603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.429073396220601, "3000–6000 Hz_10ms_envelope_correlation": 0.8852026412838296, "3000–6000 Hz_10ms_modulation_mae_db": 1.460503939667351, "3000–6000 Hz_40ms_log_rms_mae_db": 1.374676349206395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34440332623880004, "3000–6000 Hz_40ms_envelope_correlation": 0.9206236174737985, "3000–6000 Hz_40ms_modulation_mae_db": 1.415758979400345, "mr_stft_db": 5.679600888519221, "band_mae_db": 0.16946673576747118, "rms_gain_error_db": -0.04840121625380201, "output_peak": 0.5901450169091975}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.449170473682822, "spectral_convergence": 0.2593046587938499, "spectral_convergence_status": "ok", "amplitude_floor": 1.726332693189446e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.568858819408275, "spectral_convergence": 0.28079045271598124, "spectral_convergence_status": "ok", "amplitude_floor": 1.4419419879485463e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.00685869204311, "spectral_convergence": 0.2920663300777653, "spectral_convergence_status": "ok", "amplitude_floor": 1.208114881083552e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024021305054769385, "output_power": 0.00024292238134817625, "delta_db": 0.04870931308542892, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0023981297439394636, "output_power": 0.0024373853537258874, "delta_db": 0.07051521466123345, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0027873624928200164, "output_power": 0.002715915140631054, "delta_db": -0.11277255618014978, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955109173134605, "output_power": 0.00048814677908708056, "delta_db": -0.06502799418555129, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00027155220884703173, "output_power": 0.00026479672168366815, "delta_db": -0.1094073574322358, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6324739911259635e-05, "output_power": 3.9769780608836785e-05, "delta_db": 0.3935068180469016, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13332320893191682, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.028088229228717468, "pitch": null, "output_rms_dbfs": -22.08383997315076, "output_peak": 0.597174802474431, "reference_rms_dbfs": -22.055751743922045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.674962661711402}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024330354027503592, "log_rms_mae_db": 0.860302182522901, "rms_envelope_nrmse": 0.07892030244190043, "envelope_correlation": 0.993978247443152, "rms_gain_error_db": -0.028001559410919606, "reference_mean_power": 0.006233969602179056, "output_mean_power": 0.006193904776157892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8072421002875186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017414213578018356, "output_power": 0.001751322014380448, "reference_active": true, "delta_db": 0.024621402592259947}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002371798239966289, "output_power": 0.00021574648138322178, "reference_active": true, "delta_db": -0.41134020929661674}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.501630810521967e-05, "output_power": 6.580629849336616e-05, "reference_active": true, "delta_db": 0.0524515824287483}], "active_band_count": 3, "active_band_mae_db": 0.162804398105875}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022924128639936837, "log_rms_mae_db": 0.4819617396236542, "rms_envelope_nrmse": 0.04186108224851523, "envelope_correlation": 0.9981761525973084, "rms_gain_error_db": -0.027963840490184428, "reference_mean_power": 0.006262494274967633, "output_mean_power": 0.006222300166056878}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.530420990426951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00159929971914277, "output_power": 0.0016113714404740757, "reference_active": true, "delta_db": 0.03265801077405954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010914154654466241, "output_power": 9.880187968835527e-05, "reference_active": true, "delta_db": -0.43224896436909344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.773901654262189e-06, "output_power": 4.603734248799194e-06, "reference_active": true, "delta_db": -0.15763222018716438}], "active_band_count": 3, "active_band_mae_db": 0.2075130651101058}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00021113569197140015, "log_rms_mae_db": 1.2737748600070369, "rms_envelope_nrmse": 0.09956881958732217, "envelope_correlation": 0.9918102129068713, "rms_gain_error_db": -0.0485949398395093, "reference_mean_power": 0.0043153114305736935, "output_mean_power": 0.0042672948311402795}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.429798497423617e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014993372451038876, "output_power": 0.0014662302902132172, "reference_active": true, "delta_db": -0.09697142292273994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023063830674655453, "output_power": 0.00020749600997189695, "reference_active": true, "delta_db": -0.45921691079859916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9359539965921096e-05, "output_power": 6.330709894275473e-05, "reference_active": true, "delta_db": 0.2796188584907323}], "active_band_count": 3, "active_band_mae_db": 0.2786023974040238}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001994421712728477, "log_rms_mae_db": 0.8530887251055521, "rms_envelope_nrmse": 0.05160536929161956, "envelope_correlation": 0.997637285853372, "rms_gain_error_db": -0.0485763905213652, "reference_mean_power": 0.004336654014858995, "output_mean_power": 0.004288418252240031}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.221204429766169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014035050384262429, "output_power": 0.001370567228219568, "reference_active": true, "delta_db": -0.10313632898462868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010608827780695941, "output_power": 9.273880721446151e-05, "reference_active": true, "delta_db": -0.5840589351290048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7233786501496316e-06, "output_power": 4.482502226836652e-06, "reference_active": true, "delta_db": -0.22732248250440318}], "active_band_count": 3, "active_band_mae_db": 0.3048392488726788}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011279772332784735, "log_rms_mae_db": 1.419860467422227, "rms_envelope_nrmse": 0.13250030259360868, "envelope_correlation": 0.9873143418133714, "rms_gain_error_db": -0.09459042648518742, "reference_mean_power": 0.0007258838363645759, "output_mean_power": 0.0007102448334361271}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4338411984908935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00026856860022608645, "output_power": 0.00027159771987888827, "reference_active": true, "delta_db": 0.04870883990031356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.89737119770249e-05, "output_power": 1.97121036210047e-05, "reference_active": true, "delta_db": 0.16580669796655564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310520152307534e-06, "output_power": 7.174359432689656e-06, "reference_active": true, "delta_db": 1.3064606963936145}], "active_band_count": 3, "active_band_mae_db": 0.5069920780868279}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010728134925657107, "log_rms_mae_db": 0.8984281199974924, "rms_envelope_nrmse": 0.07293396695593635, "envelope_correlation": 0.9960009146967592, "rms_gain_error_db": -0.09459330643880291, "reference_mean_power": 0.0007295243282152702, "output_mean_power": 0.0007138064183646966}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.220316674170232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000246564866107528, "output_power": 0.0002516431156819069, "reference_active": true, "delta_db": 0.08853861174413671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.89658406854447e-06, "output_power": 9.393535601922075e-06, "reference_active": true, "delta_db": 0.23605798718404603}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.477241032022853e-07, "output_power": 4.7309913904228105e-07, "reference_active": true, "delta_db": 1.3371736197871975}], "active_band_count": 3, "active_band_mae_db": 0.55392340623846}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.800534128114117e-05, "log_rms_mae_db": 1.6859481308953888, "rms_envelope_nrmse": 0.4247886860442077, "envelope_correlation": 0.8878902178776725, "rms_gain_error_db": 0.46022538367424454, "reference_mean_power": 3.947976781678993e-05, "output_mean_power": 4.389318831176522e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2531812232892484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6595680637210926e-05, "output_power": 1.4653755961541104e-05, "reference_active": true, "delta_db": -0.54046114115004}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.516181974676621e-06, "output_power": 6.532496567114406e-06, "reference_active": true, "delta_db": -1.1516574274205653}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1029203509908326e-06, "output_power": 1.207876621385315e-06, "reference_active": true, "delta_db": -2.4080024838810075}], "active_band_count": 3, "active_band_mae_db": 1.366707017483871}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.0235226592584236e-05, "log_rms_mae_db": 1.2813301468433789, "rms_envelope_nrmse": 0.3500903760810466, "envelope_correlation": 0.9182705420852922, "rms_gain_error_db": 0.4602001738922207, "reference_mean_power": 3.967785924701116e-05, "output_mean_power": 4.411316820142679e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9337568551130464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8381327852746327e-05, "output_power": 1.547718007627267e-05, "reference_active": true, "delta_db": -0.7468504565847143}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.503290667812924e-06, "output_power": 3.4092402292466344e-06, "reference_active": true, "delta_db": -1.2087237518358316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6513948696073343e-07, "output_power": 8.777035670646899e-08, "reference_active": true, "delta_db": -2.745030676669646}], "active_band_count": 3, "active_band_mae_db": 1.5668682950300639}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8888578305977107, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c0636aed9345a08cad0a0b4780f6146758e44405ac04b2b69c118cf7a0254342", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9853695090276323, "standard_deviation": 0.003627966759731528, "minimum": 0.9779400851218424, "maximum": 0.9889994381999784, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4174228399525375, "standard_deviation": 0.02097924490228556, "minimum": 1.3944864747156542, "maximum": 1.446626927360296, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6769790551601477, "standard_deviation": 0.31980673982870816, "minimum": 0.34507563765481947, "maximum": 1.3058168807111064, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14292837706506073, "standard_deviation": 0.017376243521686596, "minimum": 0.12715011769166676, "maximum": 0.17878919861565876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942336156323198, "standard_deviation": 0.0030976281767908127, "minimum": 0.9885788129078004, "maximum": 0.9970537589831859, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9085280627268941, "standard_deviation": 0.024488663506524767, "minimum": 0.8637737466853962, "maximum": 0.9498833427629878, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6936794793862509, "standard_deviation": 0.4159584165002093, "minimum": 0.22019176285711797, "maximum": 1.535519909459076, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08723877483338525, "standard_deviation": 0.02184347868665253, "minimum": 0.0653458976918908, "maximum": 0.12624245535509224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8965435518632756, "standard_deviation": 0.008347125484825373, "minimum": 0.8852026412838296, "maximum": 0.9101161677405168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7353515525606946, "standard_deviation": 0.03460427131290224, "minimum": 1.6859481308953888, "maximum": 1.7795001201516603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7332983784586135, "standard_deviation": 0.5088793048655718, "minimum": 0.2565630904052955, "maximum": 1.460503939667351, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.4184592788666662, "standard_deviation": 0.007695113122943828, "minimum": 0.406984710488575, "maximum": 0.429073396220601, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9278255864422126, "standard_deviation": 0.007388461628543586, "minimum": 0.9182705420852922, "maximum": 0.9396465586204658, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3126550437052664, "standard_deviation": 0.03980500988365409, "minimum": 1.2445543233431449, "maximum": 1.374676349206395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8110595567054684, "standard_deviation": 0.525480413647605, "minimum": 0.25500163764284794, "maximum": 1.5668682950300639, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.33955674427054383, "standard_deviation": 0.007665611563465412, "minimum": 0.32860904070153235, "maximum": 0.3500903760810466, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9901842773503627, "standard_deviation": 0.0018900551384857624, "minimum": 0.9868359814966958, "maximum": 0.9918102129068713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2622726425485433, "standard_deviation": 0.03507749090521566, "minimum": 1.2045651473751149, "maximum": 1.311994465146582, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.40342300203467574, "standard_deviation": 0.2688994812400347, "minimum": 0.1545794762054592, "maximum": 0.9838683265782979, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10878757463264445, "standard_deviation": 0.010301154365090248, "minimum": 0.09956881958732217, "maximum": 0.12687072402822586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996838343404383, "standard_deviation": 0.0010435905499098339, "minimum": 0.9943390030209156, "maximum": 0.997637285853372, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8323251376511418, "standard_deviation": 0.025848086486614495, "minimum": 0.7935640434683615, "maximum": 0.86275008519228, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.479318864511705, "standard_deviation": 0.35803463041657074, "minimum": 0.054943361841328815, "maximum": 1.153937732131762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05990069597061033, "standard_deviation": 0.009299257023344616, "minimum": 0.05160536929161956, "maximum": 0.08165643846006339, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21922082079200564, "standard_deviation": 0.0497603171808305, "minimum": 0.13332320893191682, "maximum": 0.2677911942595987, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.681647529260739, "standard_deviation": 0.00732520567451434, "minimum": 5.672336981098726, "maximum": 5.691396851768512, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6233953153201705, "standard_deviation": 0.051859166493250224, "minimum": 0.5901450169091975, "maximum": 0.7237760879939561, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06092382788538065, "standard_deviation": 0.028452063126052647, "minimum": -0.11622837439506818, "maximum": -0.028088229228717468, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930155396811999, "standard_deviation": 0.0012755687443956702, "minimum": 0.990712905275784, "maximum": 0.9941518150767014, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8423046621186034, "standard_deviation": 0.021796132920362907, "minimum": 0.8021445199609983, "maximum": 0.8752437321434775, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3552336686730538, "standard_deviation": 0.2772177733981136, "minimum": 0.12006303149976001, "maximum": 0.8934827831787343, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08474782532419113, "standard_deviation": 0.007527436149881898, "minimum": 0.07800061383540953, "maximum": 0.0980326936340883, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978846147958917, "standard_deviation": 0.00041373404159560393, "minimum": 0.9969684604378907, "maximum": 0.9981761525973084, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46613028423360225, "standard_deviation": 0.02004375110756525, "minimum": 0.43329099776109065, "maximum": 0.4825430899680464, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4228716290012827, "standard_deviation": 0.34629821893886137, "minimum": 0.0783457564981368, "maximum": 1.1116924655775484, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04520612856467693, "standard_deviation": 0.004449852753516174, "minimum": 0.04186108224851523, "maximum": 0.05496473803674961, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41806-0000", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.65397310256958, "predicted_mos": 2.65397310256958, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7de3cb82a2dcf93874f087a667a88cc56d59ed966b650ad1228b87baf2248a93", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.26513580000028, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8803550792354257, "全帯域_10ms_rms_envelope_nrmse": 0.08100191871777947, "全帯域_10ms_envelope_correlation": 0.9936539313971334, "全帯域_10ms_modulation_mae_db": 0.21845182029380203, "全帯域_40ms_log_rms_mae_db": 0.49375402668528756, "全帯域_40ms_rms_envelope_nrmse": 0.04343245266635059, "全帯域_40ms_envelope_correlation": 0.9980305847433425, "全帯域_40ms_modulation_mae_db": 0.15908835146463893, "300–1000 Hz_10ms_log_rms_mae_db": 1.2179638244821251, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10026237090721811, "300–1000 Hz_10ms_envelope_correlation": 0.9916961123861219, "300–1000 Hz_10ms_modulation_mae_db": 0.3137291017975455, "300–1000 Hz_40ms_log_rms_mae_db": 0.786285159730951, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052115039542294773, "300–1000 Hz_40ms_envelope_correlation": 0.9975780302194011, "300–1000 Hz_40ms_modulation_mae_db": 0.2915866556386734, "1000–3000 Hz_10ms_log_rms_mae_db": 1.41532939298209, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13125687488014043, "1000–3000 Hz_10ms_envelope_correlation": 0.987585801603927, "1000–3000 Hz_10ms_modulation_mae_db": 0.37411889093443174, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8748838888478182, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07072209315647336, "1000–3000 Hz_40ms_envelope_correlation": 0.9963206814259098, "1000–3000 Hz_40ms_modulation_mae_db": 0.3659630954390143, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6707610795722676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4237878135268213, "3000–6000 Hz_10ms_envelope_correlation": 0.8875427889284901, "3000–6000 Hz_10ms_modulation_mae_db": 1.327790817223056, "3000–6000 Hz_40ms_log_rms_mae_db": 1.255119810747908, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3449716064212015, "3000–6000 Hz_40ms_envelope_correlation": 0.919895249391327, "3000–6000 Hz_40ms_modulation_mae_db": 1.4673641715083923, "mr_stft_db": 5.603598624550823, "band_mae_db": 0.1342430842731498, "rms_gain_error_db": -0.02344208637528024, "output_peak": 0.6072137400683572}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8616305429810289, "全帯域_10ms_rms_envelope_nrmse": 0.09908843277038552, "全帯域_10ms_envelope_correlation": 0.9905264191936112, "全帯域_10ms_modulation_mae_db": 0.9357654503915046, "全帯域_40ms_log_rms_mae_db": 0.49673979512940963, "全帯域_40ms_rms_envelope_nrmse": 0.04965198658018291, "全帯域_40ms_envelope_correlation": 0.997509810066431, "全帯域_40ms_modulation_mae_db": 1.1338779585737098, "300–1000 Hz_10ms_log_rms_mae_db": 1.3085626829403199, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1258049859745044, "300–1000 Hz_10ms_envelope_correlation": 0.987015972846861, "300–1000 Hz_10ms_modulation_mae_db": 1.0866893337366037, "300–1000 Hz_40ms_log_rms_mae_db": 0.8175180122268498, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06394911007576409, "300–1000 Hz_40ms_envelope_correlation": 0.9966200704062719, "300–1000 Hz_40ms_modulation_mae_db": 1.2344897908261012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4540950994607895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1619361294030459, "1000–3000 Hz_10ms_envelope_correlation": 0.9814481498037103, "1000–3000 Hz_10ms_modulation_mae_db": 0.8726565722426134, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9045694682838472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10402896955245297, "1000–3000 Hz_40ms_envelope_correlation": 0.9919628962198869, "1000–3000 Hz_40ms_modulation_mae_db": 0.9740678143186102, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6794931853691457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4139141065384137, "3000–6000 Hz_10ms_envelope_correlation": 0.8926056567152596, "3000–6000 Hz_10ms_modulation_mae_db": 1.1942346825406334, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2971373633387673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33477041496341314, "3000–6000 Hz_40ms_envelope_correlation": 0.92456571123822, "3000–6000 Hz_40ms_modulation_mae_db": 1.3239941205813743, "mr_stft_db": 5.6087566565671585, "band_mae_db": 0.15689930535537813, "rms_gain_error_db": -0.0945910305589373, "output_peak": 0.6442197440542881}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8048300244225425, "全帯域_10ms_rms_envelope_nrmse": 0.07754876436089711, "全帯域_10ms_envelope_correlation": 0.9942297941430925, "全帯域_10ms_modulation_mae_db": 0.09048591547527464, "全帯域_40ms_log_rms_mae_db": 0.4463655623104455, "全帯域_40ms_rms_envelope_nrmse": 0.042789269793707244, "全帯域_40ms_envelope_correlation": 0.9981497406254006, "全帯域_40ms_modulation_mae_db": 0.09076105630798864, "300–1000 Hz_10ms_log_rms_mae_db": 1.2406398025484233, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10079580383946087, "300–1000 Hz_10ms_envelope_correlation": 0.9916426711629122, "300–1000 Hz_10ms_modulation_mae_db": 0.19782487234845103, "300–1000 Hz_40ms_log_rms_mae_db": 0.8131950520112563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05936551415836655, "300–1000 Hz_40ms_envelope_correlation": 0.9969083469411049, "300–1000 Hz_40ms_modulation_mae_db": 0.20738867611923884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4627686731916414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15341199961657473, "1000–3000 Hz_10ms_envelope_correlation": 0.9832431687867487, "1000–3000 Hz_10ms_modulation_mae_db": 0.673394945414409, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9292932907113338, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10517379887061197, "1000–3000 Hz_40ms_envelope_correlation": 0.9920688353000725, "1000–3000 Hz_40ms_modulation_mae_db": 0.5228542724493773, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6870557512049384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41771685305281475, "3000–6000 Hz_10ms_envelope_correlation": 0.8994209715159895, "3000–6000 Hz_10ms_modulation_mae_db": 0.27921808014295646, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2856356498579515, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34095819722912096, "3000–6000 Hz_40ms_envelope_correlation": 0.929199762828133, "3000–6000 Hz_40ms_modulation_mae_db": 0.3444761949691944, "mr_stft_db": 5.612921862561675, "band_mae_db": 0.28239198241989133, "rms_gain_error_db": -0.06942646660142612, "output_peak": 0.591819900844012}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8456615244560904, "全帯域_10ms_rms_envelope_nrmse": 0.08001233512973857, "全帯域_10ms_envelope_correlation": 0.9938037296757349, "全帯域_10ms_modulation_mae_db": 0.19386086161904403, "全帯域_40ms_log_rms_mae_db": 0.4879305122494879, "全帯域_40ms_rms_envelope_nrmse": 0.041643530111738734, "全帯域_40ms_envelope_correlation": 0.9982082267650844, "全帯域_40ms_modulation_mae_db": 0.31280833724551615, "300–1000 Hz_10ms_log_rms_mae_db": 1.217882869243386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10628395516500946, "300–1000 Hz_10ms_envelope_correlation": 0.9906587445304149, "300–1000 Hz_10ms_modulation_mae_db": 0.17883472276042375, "300–1000 Hz_40ms_log_rms_mae_db": 0.7732517766762198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05228998083740718, "300–1000 Hz_40ms_envelope_correlation": 0.9975943099198915, "300–1000 Hz_40ms_modulation_mae_db": 0.5039552567418654, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3996219590229926, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12495450614610928, "1000–3000 Hz_10ms_envelope_correlation": 0.9891055650859146, "1000–3000 Hz_10ms_modulation_mae_db": 0.19985383831708958, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8645746378862151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06621375799161311, "1000–3000 Hz_40ms_envelope_correlation": 0.997300347645699, "1000–3000 Hz_40ms_modulation_mae_db": 0.3499089517162062, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7695600564427565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41916656035598665, "3000–6000 Hz_10ms_envelope_correlation": 0.8974473343959561, "3000–6000 Hz_10ms_modulation_mae_db": 0.5109421490307735, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3371609075454074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33781471007112557, "3000–6000 Hz_40ms_envelope_correlation": 0.929650790605626, "3000–6000 Hz_40ms_modulation_mae_db": 0.6520985092442295, "mr_stft_db": 5.604998861566116, "band_mae_db": 0.23401700839280556, "rms_gain_error_db": -0.05265788146221831, "output_peak": 0.607734275283023}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8880994869416734, "全帯域_10ms_rms_envelope_nrmse": 0.08415718082877582, "全帯域_10ms_envelope_correlation": 0.9931333507868222, "全帯域_10ms_modulation_mae_db": 0.14714679862705493, "全帯域_40ms_log_rms_mae_db": 0.4848175651277781, "全帯域_40ms_rms_envelope_nrmse": 0.04411578469620984, "全帯域_40ms_envelope_correlation": 0.9979687770365168, "全帯域_40ms_modulation_mae_db": 0.15970375839756615, "300–1000 Hz_10ms_log_rms_mae_db": 1.1949161320607495, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10634330055456634, "300–1000 Hz_10ms_envelope_correlation": 0.9906519719008539, "300–1000 Hz_10ms_modulation_mae_db": 0.18702418386658773, "300–1000 Hz_40ms_log_rms_mae_db": 0.7806434820971856, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059357904563067006, "300–1000 Hz_40ms_envelope_correlation": 0.9968604223945005, "300–1000 Hz_40ms_modulation_mae_db": 0.17492362215661408, "1000–3000 Hz_10ms_log_rms_mae_db": 1.42451552975855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13066653878142717, "1000–3000 Hz_10ms_envelope_correlation": 0.9879726084261712, "1000–3000 Hz_10ms_modulation_mae_db": 0.7546477951315614, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8999755190724952, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06430860342052254, "1000–3000 Hz_40ms_envelope_correlation": 0.9974299179809399, "1000–3000 Hz_40ms_modulation_mae_db": 0.6929358094359509, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7178074523466944, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4099132335388353, "3000–6000 Hz_10ms_envelope_correlation": 0.9043526666637102, "3000–6000 Hz_10ms_modulation_mae_db": 0.4847170343455636, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2913170394341997, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32978694466083464, "3000–6000 Hz_40ms_envelope_correlation": 0.9350493691337768, "3000–6000 Hz_40ms_modulation_mae_db": 0.4410718115226664, "mr_stft_db": 5.618280485025099, "band_mae_db": 0.2561654861896233, "rms_gain_error_db": -0.04214019545506729, "output_peak": 0.6970175214838973}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8312045790989452, "全帯域_10ms_rms_envelope_nrmse": 0.08184547149938255, "全帯域_10ms_envelope_correlation": 0.9935387429507425, "全帯域_10ms_modulation_mae_db": 0.2746077410182997, "全帯域_40ms_log_rms_mae_db": 0.4744175580836517, "全帯域_40ms_rms_envelope_nrmse": 0.04271977092155581, "全帯域_40ms_envelope_correlation": 0.9981484434328075, "全帯域_40ms_modulation_mae_db": 0.48730333386956165, "300–1000 Hz_10ms_log_rms_mae_db": 1.2043346418496426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10186008755644757, "300–1000 Hz_10ms_envelope_correlation": 0.9914272802139915, "300–1000 Hz_10ms_modulation_mae_db": 0.4293300160253488, "300–1000 Hz_40ms_log_rms_mae_db": 0.7816778668382556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054819585676257714, "300–1000 Hz_40ms_envelope_correlation": 0.9973491828211597, "300–1000 Hz_40ms_modulation_mae_db": 0.7321519115421048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3720013775927593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12483782564478978, "1000–3000 Hz_10ms_envelope_correlation": 0.9895028773857328, "1000–3000 Hz_10ms_modulation_mae_db": 0.3686868909522605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8631059524545646, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07277530469469047, "1000–3000 Hz_40ms_envelope_correlation": 0.9970871866846669, "1000–3000 Hz_40ms_modulation_mae_db": 0.3420088144579803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.732265918862404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.415427851394437, "3000–6000 Hz_10ms_envelope_correlation": 0.9033718385073408, "3000–6000 Hz_10ms_modulation_mae_db": 0.254240946696252, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2689486914451344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3336899223339245, "3000–6000 Hz_40ms_envelope_correlation": 0.9349741785803114, "3000–6000 Hz_40ms_modulation_mae_db": 0.38449166253162387, "mr_stft_db": 5.6154923184832795, "band_mae_db": 0.24978071927918633, "rms_gain_error_db": -0.078106419797284, "output_peak": 0.669304315485259}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8582232700286311, "全帯域_10ms_rms_envelope_nrmse": 0.09442382926385237, "全帯域_10ms_envelope_correlation": 0.9914243671748157, "全帯域_10ms_modulation_mae_db": 0.5853745330897292, "全帯域_40ms_log_rms_mae_db": 0.4498340572899458, "全帯域_40ms_rms_envelope_nrmse": 0.053942302649806886, "全帯域_40ms_envelope_correlation": 0.9970818124529118, "全帯域_40ms_modulation_mae_db": 0.6670968025766641, "300–1000 Hz_10ms_log_rms_mae_db": 1.2615713194588285, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12489927307118506, "300–1000 Hz_10ms_envelope_correlation": 0.9872014107406744, "300–1000 Hz_10ms_modulation_mae_db": 0.6373418204464419, "300–1000 Hz_40ms_log_rms_mae_db": 0.8095295710794459, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07833002282762404, "300–1000 Hz_40ms_envelope_correlation": 0.99474944598664, "300–1000 Hz_40ms_modulation_mae_db": 0.8865905608484147, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3531544193622016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12353764059824668, "1000–3000 Hz_10ms_envelope_correlation": 0.9889945814456788, "1000–3000 Hz_10ms_modulation_mae_db": 0.2001811931813071, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8269971967831207, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06895722477410657, "1000–3000 Hz_40ms_envelope_correlation": 0.9963889429836906, "1000–3000 Hz_40ms_modulation_mae_db": 0.23966315956526238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6574113119940947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4041720114313461, "3000–6000 Hz_10ms_envelope_correlation": 0.9082553822805233, "3000–6000 Hz_10ms_modulation_mae_db": 0.37883984039037005, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1833976665455768, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32181531686732584, "3000–6000 Hz_40ms_envelope_correlation": 0.939503516995596, "3000–6000 Hz_40ms_modulation_mae_db": 0.5624327803217085, "mr_stft_db": 5.601397495881333, "band_mae_db": 0.22736442213128805, "rms_gain_error_db": -0.11454067165716139, "output_peak": 0.602508083858133}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8322252787702362, "全帯域_10ms_rms_envelope_nrmse": 0.08040129599189359, "全帯域_10ms_envelope_correlation": 0.9937518887759277, "全帯域_10ms_modulation_mae_db": 0.4864630234340253, "全帯域_40ms_log_rms_mae_db": 0.4545714203243086, "全帯域_40ms_rms_envelope_nrmse": 0.04304672731188587, "全帯域_40ms_envelope_correlation": 0.998090638994243, "全帯域_40ms_modulation_mae_db": 0.5782266559366371, "300–1000 Hz_10ms_log_rms_mae_db": 1.2278960912239962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10559062958779372, "300–1000 Hz_10ms_envelope_correlation": 0.9908208751482824, "300–1000 Hz_10ms_modulation_mae_db": 0.526304382967126, "300–1000 Hz_40ms_log_rms_mae_db": 0.816772878720122, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05580185868068895, "300–1000 Hz_40ms_envelope_correlation": 0.9973313907731021, "300–1000 Hz_40ms_modulation_mae_db": 0.6760011524217554, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3622583797297756, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12968986918263337, "1000–3000 Hz_10ms_envelope_correlation": 0.9880168051463423, "1000–3000 Hz_10ms_modulation_mae_db": 0.2956105084969249, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8575142653252242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08652364983158099, "1000–3000 Hz_40ms_envelope_correlation": 0.9943918888719, "1000–3000 Hz_40ms_modulation_mae_db": 0.3730052312127894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7609687166000512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.42754767171284797, "3000–6000 Hz_10ms_envelope_correlation": 0.8864246470124383, "3000–6000 Hz_10ms_modulation_mae_db": 1.2756798441201134, "3000–6000 Hz_40ms_log_rms_mae_db": 1.344091864224117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34020491654111035, "3000–6000 Hz_40ms_envelope_correlation": 0.9228374156665771, "3000–6000 Hz_40ms_modulation_mae_db": 1.2426931065825104, "mr_stft_db": 5.611358796379051, "band_mae_db": 0.16668853556460148, "rms_gain_error_db": -0.06015697826869201, "output_peak": 0.604268418235936}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.384452430853743, "spectral_convergence": 0.2592854653856663, "spectral_convergence_status": "ok", "amplitude_floor": 1.726332693189446e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.503016889835927, "spectral_convergence": 0.28163796723203927, "spectral_convergence_status": "ok", "amplitude_floor": 1.4419419879485463e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.923326552962799, "spectral_convergence": 0.29311820051519144, "spectral_convergence_status": "ok", "amplitude_floor": 1.208114881083552e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024021305054769385, "output_power": 0.00024101494039518206, "delta_db": 0.01447366590251997, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0023981297439394636, "output_power": 0.0024363196904006166, "delta_db": 0.06861599546385759, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0027873624928200164, "output_power": 0.00273229858702379, "delta_db": -0.08665294279522859, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955109173134605, "output_power": 0.00048441952669420516, "delta_db": -0.09831586488770293, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00027155220884703173, "output_power": 0.00026186496050908053, "delta_db": -0.157759492986526, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6324739911259635e-05, "output_power": 3.964300504632322e-05, "delta_db": 0.3796405436030637, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1342430842731498, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.02344208637528024, "pitch": null, "output_rms_dbfs": -22.079193830297324, "output_peak": 0.6072137400683572, "reference_rms_dbfs": -22.055751743922045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.603598624550823}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024330354027503592, "log_rms_mae_db": 0.8803550792354257, "rms_envelope_nrmse": 0.08100191871777947, "envelope_correlation": 0.9936539313971334, "rms_gain_error_db": -0.02335333909880817, "reference_mean_power": 0.006233969602179056, "output_mean_power": 0.006200537613526775}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8072421002875186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017414213578018356, "output_power": 0.0017517961607612811, "reference_active": true, "delta_db": 0.02579703611515247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002371798239966289, "output_power": 0.00021615530805007267, "reference_active": true, "delta_db": -0.40311837633177094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.501630810521967e-05, "output_power": 6.849617336010857e-05, "reference_active": true, "delta_db": 0.22644004843448268}], "active_band_count": 3, "active_band_mae_db": 0.21845182029380203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022924128639936837, "log_rms_mae_db": 0.49375402668528756, "rms_envelope_nrmse": 0.04343245266635059, "envelope_correlation": 0.9980305847433425, "rms_gain_error_db": -0.023313557714039448, "reference_mean_power": 0.006262494274967633, "output_mean_power": 0.006228966369196748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.530420990426951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00159929971914277, "output_power": 0.0016126613625398094, "reference_active": true, "delta_db": 0.03613319914436266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010914154654466241, "output_power": 9.944583038296364e-05, "reference_active": true, "delta_db": -0.4040352496421843}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.773901654262189e-06, "output_power": 4.814854065731267e-06, "reference_active": true, "delta_db": 0.037096605607369854}], "active_band_count": 3, "active_band_mae_db": 0.15908835146463893}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00021113569197140015, "log_rms_mae_db": 1.2179638244821251, "rms_envelope_nrmse": 0.10026237090721811, "envelope_correlation": 0.9916961123861219, "rms_gain_error_db": -0.03055164321868923, "reference_mean_power": 0.0043153114305736935, "output_mean_power": 0.0042850607103198735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.429798497423617e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014993372451038876, "output_power": 0.001478090181828199, "reference_active": true, "delta_db": -0.06198397240175016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023063830674655453, "output_power": 0.00020547623507880702, "reference_active": true, "delta_db": -0.5016984138750381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9359539965921096e-05, "output_power": 6.475018572126709e-05, "reference_active": true, "delta_db": 0.3775049191158483}], "active_band_count": 3, "active_band_mae_db": 0.3137291017975455}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001994421712728477, "log_rms_mae_db": 0.786285159730951, "rms_envelope_nrmse": 0.052115039542294773, "envelope_correlation": 0.9975780302194011, "rms_gain_error_db": -0.030531726449136383, "reference_mean_power": 0.004336654014858995, "output_mean_power": 0.004306273429723767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.221204429766169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014035050384262429, "output_power": 0.0013799986946338287, "reference_active": true, "delta_db": -0.07335300447628051}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010608827780695941, "output_power": 9.233150525209553e-05, "reference_active": true, "delta_db": -0.6031748352379551}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7233786501496316e-06, "output_power": 4.512628163451253e-06, "reference_active": true, "delta_db": -0.19823212720178462}], "active_band_count": 3, "active_band_mae_db": 0.2915866556386734}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011279772332784735, "log_rms_mae_db": 1.41532939298209, "rms_envelope_nrmse": 0.13125687488014043, "envelope_correlation": 0.987585801603927, "rms_gain_error_db": -0.1334556953511799, "reference_mean_power": 0.0007258838363645759, "output_mean_power": 0.0007039171661707448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4338411984908935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00026856860022608645, "output_power": 0.0002669881292676217, "reference_active": true, "delta_db": -0.025632832932190684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.89737119770249e-05, "output_power": 1.9354609167508545e-05, "reference_active": true, "delta_db": 0.08632102267738148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310520152307534e-06, "output_power": 6.701582108078182e-06, "reference_active": true, "delta_db": 1.0104028171937232}], "active_band_count": 3, "active_band_mae_db": 0.37411889093443174}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010728134925657107, "log_rms_mae_db": 0.8748838888478182, "rms_envelope_nrmse": 0.07072209315647336, "envelope_correlation": 0.9963206814259098, "rms_gain_error_db": -0.13345739861108702, "reference_mean_power": 0.0007295243282152702, "output_mean_power": 0.0007074472121359301}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.220316674170232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000246564866107528, "output_power": 0.00024689055393127313, "reference_active": true, "delta_db": 0.005732815548083924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.89658406854447e-06, "output_power": 9.248389036233753e-06, "reference_active": true, "delta_db": 0.16842803199718337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.477241032022853e-07, "output_power": 4.3013782980427057e-07, "reference_active": true, "delta_db": 0.9237284387717755}], "active_band_count": 3, "active_band_mae_db": 0.3659630954390143}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.800534128114117e-05, "log_rms_mae_db": 1.6707610795722676, "rms_envelope_nrmse": 0.4237878135268213, "envelope_correlation": 0.8875427889284901, "rms_gain_error_db": 0.4043929174546316, "reference_mean_power": 3.947976781678993e-05, "output_mean_power": 4.333251356417137e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2531812232892484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6595680637210926e-05, "output_power": 1.4488320237257387e-05, "reference_active": true, "delta_db": -0.589770321113823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.516181974676621e-06, "output_power": 6.696502884801469e-06, "reference_active": true, "delta_db": -1.043968726538123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1029203509908326e-06, "output_power": 1.22422003733767e-06, "reference_active": true, "delta_db": -2.349633404017222}], "active_band_count": 3, "active_band_mae_db": 1.327790817223056}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.0235226592584236e-05, "log_rms_mae_db": 1.255119810747908, "rms_envelope_nrmse": 0.3449716064212015, "envelope_correlation": 0.919895249391327, "rms_gain_error_db": 0.4043709413025188, "reference_mean_power": 3.967785924701116e-05, "output_mean_power": 4.354971594095398e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9337568551130464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8381327852746327e-05, "output_power": 1.5388595123377396e-05, "reference_active": true, "delta_db": -0.7717790784022835}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.503290667812924e-06, "output_power": 3.408147174217152e-06, "reference_active": true, "delta_db": -1.2101163902285015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6513948696073343e-07, "output_power": 9.45869510415129e-08, "reference_active": true, "delta_db": -2.4201970458943918}], "active_band_count": 3, "active_band_mae_db": 1.4673641715083923}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8895110875482266, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "18c77da9a2fcd3039827f50acbd96311ef69846405e4b04a03b91c0b846df458", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9869836947105282, "standard_deviation": 0.0029745077029680878, "minimum": 0.9814481498037103, "maximum": 0.9895028773857328, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4054681038876, "standard_deviation": 0.04117209600314726, "minimum": 1.3531544193622016, "maximum": 1.4627686731916414, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46739382933382473, "standard_deviation": 0.2618948028534595, "minimum": 0.19985383831708958, "maximum": 0.8726565722426134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13503642303162092, "standard_deviation": 0.014447446450315204, "minimum": 0.12353764059824668, "maximum": 0.1619361294030459, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953688371390956, "standard_deviation": 0.002277937955836767, "minimum": 0.9919628962198869, "maximum": 0.9974299179809399, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8776142774205774, "standard_deviation": 0.03218563546089838, "minimum": 0.8269971967831207, "maximum": 0.9292932907113338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.48255089357439884, "standard_deviation": 0.24184069060890817, "minimum": 0.23966315956526238, "maximum": 0.9740678143186102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07983792528650649, "standard_deviation": 0.01669028137408699, "minimum": 0.06430860342052254, "maximum": 0.10517379887061197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8974276607524635, "standard_deviation": 0.007986872351089653, "minimum": 0.8864246470124383, "maximum": 0.9082553822805233, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7094154340490442, "standard_deviation": 0.042192706522155066, "minimum": 1.6574113119940947, "maximum": 1.7695600564427565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7132079243112148, "standard_deviation": 0.467423477700635, "minimum": 0.254240946696252, "maximum": 1.327790817223056, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.41645576269393786, "standard_deviation": 0.007434860344895855, "minimum": 0.4041720114313461, "maximum": 0.42754767171284797, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9294594993049459, "standard_deviation": 0.006781494870479243, "minimum": 0.919895249391327, "maximum": 0.939503516995596, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.282851124142383, "standard_deviation": 0.05047884473441887, "minimum": 1.1833976665455768, "maximum": 1.344091864224117, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8023277946577124, "standard_deviation": 0.4633989654377451, "minimum": 0.3444761949691944, "maximum": 1.4673641715083923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.33550150363600706, "standard_deviation": 0.007277318763585572, "minimum": 0.32181531686732584, "maximum": 0.3449716064212015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990139379866264, "standard_deviation": 0.0019170754596564293, "minimum": 0.987015972846861, "maximum": 0.9916961123861219, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2342209204759338, "standard_deviation": 0.036498979660261735, "minimum": 1.1949161320607495, "maximum": 1.3085626829403199, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.44463480424356605, "standard_deviation": 0.30968613835624975, "minimum": 0.17883472276042375, "maximum": 1.0866893337366037, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1089800508320232, "standard_deviation": 0.010391503113071885, "minimum": 0.10026237090721811, "maximum": 0.1258049859745044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996873899932759, "standard_deviation": 0.0009280612619288039, "minimum": 0.99474944598664, "maximum": 0.9975943099198915, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7973592249225359, "standard_deviation": 0.018559873902369208, "minimum": 0.7732517766762198, "maximum": 0.8175180122268498, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5883859532868461, "standard_deviation": 0.36762745521649637, "minimum": 0.17492362215661408, "maximum": 1.2344897908261012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05950362704518379, "standard_deviation": 0.00858991981679327, "minimum": 0.052115039542294773, "maximum": 0.07833002282762404, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2134438179507405, "standard_deviation": 0.05368225918091928, "minimum": 0.1342430842731498, "maximum": 0.28239198241989133, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.609600637626817, "standard_deviation": 0.0059686999522586, "minimum": 5.601397495881333, "maximum": 5.618280485025099, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6280107499141132, "standard_deviation": 0.03797995234139619, "minimum": 0.591819900844012, "maximum": 0.6970175214838973, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06688271627200834, "standard_deviation": 0.029077079136899234, "minimum": -0.11454067165716139, "maximum": -0.02344208637528024, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993007778012235, "standard_deviation": 0.0013125820385040503, "minimum": 0.9905264191936112, "maximum": 0.9942297941430925, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8502787232418216, "standard_deviation": 0.02752564926650268, "minimum": 0.8048300244225425, "maximum": 0.8880994869416734, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36651951799359184, "standard_deviation": 0.2855637689286342, "minimum": 0.09048591547527464, "maximum": 0.9357654503915046, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08480990357033813, "standard_deviation": 0.007701532712174297, "minimum": 0.07754876436089711, "maximum": 0.09908843277038552, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978985042645921, "standard_deviation": 0.0003962115756476431, "minimum": 0.9970818124529118, "maximum": 0.9982082267650844, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47355381215003933, "standard_deviation": 0.02050123713099569, "minimum": 0.4463655623104455, "maximum": 0.49673979512940963, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44860828179653534, "standard_deviation": 0.34834586036260035, "minimum": 0.09076105630798864, "maximum": 1.1338779585737098, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045167728091429736, "standard_deviation": 0.004305937304110799, "minimum": 0.041643530111738734, "maximum": 0.053942302649806886, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41806-0000", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5079665184020996, "predicted_mos": 2.5079665184020996, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3354f0b08fb12954186ce1217250a5da6d54d0a0e39e21491ca685d70a5af612", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9068124000914395, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8282408171222271, "全帯域_10ms_rms_envelope_nrmse": 0.08252761565967646, "全帯域_10ms_envelope_correlation": 0.993400757408658, "全帯域_10ms_modulation_mae_db": 0.2522552479923285, "全帯域_40ms_log_rms_mae_db": 0.46782327953606057, "全帯域_40ms_rms_envelope_nrmse": 0.04726200300707512, "全帯域_40ms_envelope_correlation": 0.9976771646825406, "全帯域_40ms_modulation_mae_db": 0.3280197884672051, "300–1000 Hz_10ms_log_rms_mae_db": 1.5290828334957869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10297382757865543, "300–1000 Hz_10ms_envelope_correlation": 0.9913276282631434, "300–1000 Hz_10ms_modulation_mae_db": 0.30762754619753513, "300–1000 Hz_40ms_log_rms_mae_db": 1.1308399151659385, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06894805568670014, "300–1000 Hz_40ms_envelope_correlation": 0.9959268304708354, "300–1000 Hz_40ms_modulation_mae_db": 0.4566009124585437, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6971869439654794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15680399731598466, "1000–3000 Hz_10ms_envelope_correlation": 0.9828534269576452, "1000–3000 Hz_10ms_modulation_mae_db": 0.4517798390562349, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1450189253164982, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0992782456885892, "1000–3000 Hz_40ms_envelope_correlation": 0.9928901824696384, "1000–3000 Hz_40ms_modulation_mae_db": 0.4988334571143825, "3000–6000 Hz_10ms_log_rms_mae_db": 2.261412980223361, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4399508188547662, "3000–6000 Hz_10ms_envelope_correlation": 0.893404263557606, "3000–6000 Hz_10ms_modulation_mae_db": 0.4286887021422478, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6466706960294182, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3444153921162822, "3000–6000 Hz_40ms_envelope_correlation": 0.9328702698077802, "3000–6000 Hz_40ms_modulation_mae_db": 0.49919085499742755, "mr_stft_db": 5.9935966007676145, "band_mae_db": 0.3104455471621677, "rms_gain_error_db": -0.04802775376999885, "output_peak": 0.5957269730501199}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8347376979649743, "全帯域_10ms_rms_envelope_nrmse": 0.0866684490533559, "全帯域_10ms_envelope_correlation": 0.9927576659065416, "全帯域_10ms_modulation_mae_db": 0.28844031040099605, "全帯域_40ms_log_rms_mae_db": 0.45549909045297604, "全帯域_40ms_rms_envelope_nrmse": 0.05617258344994061, "全帯域_40ms_envelope_correlation": 0.9967309908113335, "全帯域_40ms_modulation_mae_db": 0.4117992413257317, "300–1000 Hz_10ms_log_rms_mae_db": 1.4531593727786638, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.108110145957393, "300–1000 Hz_10ms_envelope_correlation": 0.9903697617899692, "300–1000 Hz_10ms_modulation_mae_db": 0.4038430029328681, "300–1000 Hz_40ms_log_rms_mae_db": 1.0683759251531213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07028512207292976, "300–1000 Hz_40ms_envelope_correlation": 0.9956431758522173, "300–1000 Hz_40ms_modulation_mae_db": 0.4272185667533697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8136435148743346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18196792169320625, "1000–3000 Hz_10ms_envelope_correlation": 0.9759609618701166, "1000–3000 Hz_10ms_modulation_mae_db": 0.8446192308928234, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2229965028707654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11537085768752653, "1000–3000 Hz_40ms_envelope_correlation": 0.9898931429382463, "1000–3000 Hz_40ms_modulation_mae_db": 0.9186726152738848, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2755784616588715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4372503274873756, "3000–6000 Hz_10ms_envelope_correlation": 0.8830078436395595, "3000–6000 Hz_10ms_modulation_mae_db": 0.7101988371258376, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7173348090321476, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3466371171412798, "3000–6000 Hz_40ms_envelope_correlation": 0.9216636100328236, "3000–6000 Hz_40ms_modulation_mae_db": 0.80331855505901, "mr_stft_db": 6.0007102729728174, "band_mae_db": 0.21673102678911496, "rms_gain_error_db": -0.05526556979765499, "output_peak": 0.6563310312204772}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8551838786503653, "全帯域_10ms_rms_envelope_nrmse": 0.09422049913745358, "全帯域_10ms_envelope_correlation": 0.9914554029797162, "全帯域_10ms_modulation_mae_db": 0.6476090008658903, "全帯域_40ms_log_rms_mae_db": 0.4785684297247244, "全帯域_40ms_rms_envelope_nrmse": 0.053619359724545165, "全帯域_40ms_envelope_correlation": 0.9970974112603382, "全帯域_40ms_modulation_mae_db": 0.8265518820699636, "300–1000 Hz_10ms_log_rms_mae_db": 1.4536906820504767, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1093682428844186, "300–1000 Hz_10ms_envelope_correlation": 0.9903049686046309, "300–1000 Hz_10ms_modulation_mae_db": 1.0695760780907426, "300–1000 Hz_40ms_log_rms_mae_db": 1.008097567877905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06371410026847302, "300–1000 Hz_40ms_envelope_correlation": 0.9967124772858331, "300–1000 Hz_40ms_modulation_mae_db": 1.2849821289776464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7810570303452922, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18607873867150893, "1000–3000 Hz_10ms_envelope_correlation": 0.9753716550974311, "1000–3000 Hz_10ms_modulation_mae_db": 1.321571565639412, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1538033988092036, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11140800434931966, "1000–3000 Hz_40ms_envelope_correlation": 0.9907046193895851, "1000–3000 Hz_40ms_modulation_mae_db": 1.2015284355335385, "3000–6000 Hz_10ms_log_rms_mae_db": 2.329875488662005, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4310741424572931, "3000–6000 Hz_10ms_envelope_correlation": 0.8854232862960186, "3000–6000 Hz_10ms_modulation_mae_db": 0.8309785282394458, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7834187246374307, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34298005843247836, "3000–6000 Hz_40ms_envelope_correlation": 0.9226129889250798, "3000–6000 Hz_40ms_modulation_mae_db": 0.9205339479709513, "mr_stft_db": 5.9995041415116335, "band_mae_db": 0.24756287764185889, "rms_gain_error_db": -0.10269075100828706, "output_peak": 0.7480495247874359}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8561213932082764, "全帯域_10ms_rms_envelope_nrmse": 0.08777517302648946, "全帯域_10ms_envelope_correlation": 0.9925754000228487, "全帯域_10ms_modulation_mae_db": 0.13188324551721986, "全帯域_40ms_log_rms_mae_db": 0.48447244541497364, "全帯域_40ms_rms_envelope_nrmse": 0.0494443024979457, "全帯域_40ms_envelope_correlation": 0.9975317881365289, "全帯域_40ms_modulation_mae_db": 0.09139710738475652, "300–1000 Hz_10ms_log_rms_mae_db": 1.4659431552965878, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11190291519672839, "300–1000 Hz_10ms_envelope_correlation": 0.989746897635067, "300–1000 Hz_10ms_modulation_mae_db": 0.19308700322938968, "300–1000 Hz_40ms_log_rms_mae_db": 1.0158145600974688, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0637562964485818, "300–1000 Hz_40ms_envelope_correlation": 0.9965900889501206, "300–1000 Hz_40ms_modulation_mae_db": 0.28411569428459993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.748189277727289, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14609900350456806, "1000–3000 Hz_10ms_envelope_correlation": 0.9846929275571912, "1000–3000 Hz_10ms_modulation_mae_db": 0.8256961567043041, "1000–3000 Hz_40ms_log_rms_mae_db": 1.139091140631303, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08146583291739158, "1000–3000 Hz_40ms_envelope_correlation": 0.995330129079532, "1000–3000 Hz_40ms_modulation_mae_db": 0.5833851135080721, "3000–6000 Hz_10ms_log_rms_mae_db": 2.282457366738097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.443583045346786, "3000–6000 Hz_10ms_envelope_correlation": 0.8812724081590702, "3000–6000 Hz_10ms_modulation_mae_db": 0.5933473391612338, "3000–6000 Hz_40ms_log_rms_mae_db": 1.712420394443126, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3517363300073568, "3000–6000 Hz_40ms_envelope_correlation": 0.9206069673370684, "3000–6000 Hz_40ms_modulation_mae_db": 0.8062529185032, "mr_stft_db": 5.998373423731517, "band_mae_db": 0.23719139730658986, "rms_gain_error_db": -0.09582463119624311, "output_peak": 0.5952051580186282}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8241671650102814, "全帯域_10ms_rms_envelope_nrmse": 0.07744528341179087, "全帯域_10ms_envelope_correlation": 0.9941907012954322, "全帯域_10ms_modulation_mae_db": 0.3068573062864948, "全帯域_40ms_log_rms_mae_db": 0.4447992380657228, "全帯域_40ms_rms_envelope_nrmse": 0.04264757053022625, "全帯域_40ms_envelope_correlation": 0.998106485687998, "全帯域_40ms_modulation_mae_db": 0.37551679328869086, "300–1000 Hz_10ms_log_rms_mae_db": 1.430170506065066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09940570665025666, "300–1000 Hz_10ms_envelope_correlation": 0.9918717331899587, "300–1000 Hz_10ms_modulation_mae_db": 0.3119733607502506, "300–1000 Hz_40ms_log_rms_mae_db": 0.9812922367433322, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057684072235139446, "300–1000 Hz_40ms_envelope_correlation": 0.9971278132001133, "300–1000 Hz_40ms_modulation_mae_db": 0.5124749502599831, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7417784555358753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16642875157492737, "1000–3000 Hz_10ms_envelope_correlation": 0.9800989016405479, "1000–3000 Hz_10ms_modulation_mae_db": 0.7953614072359837, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1568332794713898, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10709066537310967, "1000–3000 Hz_40ms_envelope_correlation": 0.9912966468839983, "1000–3000 Hz_40ms_modulation_mae_db": 0.8545989420362566, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2590915249803136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.43485167333759356, "3000–6000 Hz_10ms_envelope_correlation": 0.8853935246941774, "3000–6000 Hz_10ms_modulation_mae_db": 0.43403381695332266, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6885271559090338, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34244310479356277, "3000–6000 Hz_40ms_envelope_correlation": 0.9244721494728083, "3000–6000 Hz_40ms_modulation_mae_db": 0.402184765585243, "mr_stft_db": 5.996812567538343, "band_mae_db": 0.20045963051521987, "rms_gain_error_db": -0.04424484529936541, "output_peak": 0.5730017812869915}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8175447883147478, "全帯域_10ms_rms_envelope_nrmse": 0.08245192071244135, "全帯域_10ms_envelope_correlation": 0.9934560775793319, "全帯域_10ms_modulation_mae_db": 0.4609028096287467, "全帯域_40ms_log_rms_mae_db": 0.47940374589224605, "全帯域_40ms_rms_envelope_nrmse": 0.04529796040598112, "全帯域_40ms_envelope_correlation": 0.9979190916059612, "全帯域_40ms_modulation_mae_db": 0.5605248143665705, "300–1000 Hz_10ms_log_rms_mae_db": 1.5070312346746857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10586698283943782, "300–1000 Hz_10ms_envelope_correlation": 0.9908701462797328, "300–1000 Hz_10ms_modulation_mae_db": 0.523983716935764, "300–1000 Hz_40ms_log_rms_mae_db": 1.105858924503085, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06049260656233407, "300–1000 Hz_40ms_envelope_correlation": 0.9969686974915749, "300–1000 Hz_40ms_modulation_mae_db": 0.6151556068035479, "1000–3000 Hz_10ms_log_rms_mae_db": 1.717838213460895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1402103617955362, "1000–3000 Hz_10ms_envelope_correlation": 0.9860579930378202, "1000–3000 Hz_10ms_modulation_mae_db": 0.29724288823233386, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1481396217663309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07653061334628522, "1000–3000 Hz_40ms_envelope_correlation": 0.9956621154677042, "1000–3000 Hz_40ms_modulation_mae_db": 0.23514115768627877, "3000–6000 Hz_10ms_log_rms_mae_db": 2.303248580042319, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4253819232021422, "3000–6000 Hz_10ms_envelope_correlation": 0.8937898669790856, "3000–6000 Hz_10ms_modulation_mae_db": 0.2470042039875333, "3000–6000 Hz_40ms_log_rms_mae_db": 1.736851203318128, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33591112246230115, "3000–6000 Hz_40ms_envelope_correlation": 0.93037685390104, "3000–6000 Hz_40ms_modulation_mae_db": 0.3658713461415298, "mr_stft_db": 6.005440814061186, "band_mae_db": 0.24328433723041196, "rms_gain_error_db": -0.07110485227683129, "output_peak": 0.8034345199229584}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8576030051630124, "全帯域_10ms_rms_envelope_nrmse": 0.0831777060778522, "全帯域_10ms_envelope_correlation": 0.9933690827340887, "全帯域_10ms_modulation_mae_db": 0.2275630423687969, "全帯域_40ms_log_rms_mae_db": 0.4779477508186066, "全帯域_40ms_rms_envelope_nrmse": 0.05280522565140405, "全帯域_40ms_envelope_correlation": 0.997198427428592, "全帯域_40ms_modulation_mae_db": 0.1435413214112257, "300–1000 Hz_10ms_log_rms_mae_db": 1.5346161302244372, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10321910404120506, "300–1000 Hz_10ms_envelope_correlation": 0.9913308704185217, "300–1000 Hz_10ms_modulation_mae_db": 0.2535882016386534, "300–1000 Hz_40ms_log_rms_mae_db": 1.080486915702914, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06914465547987596, "300–1000 Hz_40ms_envelope_correlation": 0.9959622850754771, "300–1000 Hz_40ms_modulation_mae_db": 0.1704394597200283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7456964810030535, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1550420200312234, "1000–3000 Hz_10ms_envelope_correlation": 0.982585343722075, "1000–3000 Hz_10ms_modulation_mae_db": 1.1462577554133404, "1000–3000 Hz_40ms_log_rms_mae_db": 1.173086040590947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08488240112991999, "1000–3000 Hz_40ms_envelope_correlation": 0.9946271938909356, "1000–3000 Hz_40ms_modulation_mae_db": 0.9289126079931052, "3000–6000 Hz_10ms_log_rms_mae_db": 2.249339604372035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4375307934828097, "3000–6000 Hz_10ms_envelope_correlation": 0.8813746245067232, "3000–6000 Hz_10ms_modulation_mae_db": 1.1968165064121659, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7019361115882827, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3473131951050935, "3000–6000 Hz_40ms_envelope_correlation": 0.9201138709977291, "3000–6000 Hz_40ms_modulation_mae_db": 1.5675167617035168, "mr_stft_db": 5.997898358854766, "band_mae_db": 0.21196945950897353, "rms_gain_error_db": -0.10652443661602047, "output_peak": 0.7302902243422973}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8644369425575948, "全帯域_10ms_rms_envelope_nrmse": 0.08087775359458682, "全帯域_10ms_envelope_correlation": 0.993658591199008, "全帯域_10ms_modulation_mae_db": 0.49234315341384766, "全帯域_40ms_log_rms_mae_db": 0.4648953014035883, "全帯域_40ms_rms_envelope_nrmse": 0.037558118562651684, "全帯域_40ms_envelope_correlation": 0.9985212300991955, "全帯域_40ms_modulation_mae_db": 0.6213423332055168, "300–1000 Hz_10ms_log_rms_mae_db": 1.4818977405469793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10129438478945156, "300–1000 Hz_10ms_envelope_correlation": 0.9915169437505804, "300–1000 Hz_10ms_modulation_mae_db": 0.5764142224528224, "300–1000 Hz_40ms_log_rms_mae_db": 1.028312934338978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051480489685141836, "300–1000 Hz_40ms_envelope_correlation": 0.997648386868499, "300–1000 Hz_40ms_modulation_mae_db": 0.7448055765116203, "1000–3000 Hz_10ms_log_rms_mae_db": 1.739604258559229, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15149532603322927, "1000–3000 Hz_10ms_envelope_correlation": 0.9833784804448664, "1000–3000 Hz_10ms_modulation_mae_db": 0.5468689893870597, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1560621142220557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09010390321966037, "1000–3000 Hz_40ms_envelope_correlation": 0.9938806007194781, "1000–3000 Hz_40ms_modulation_mae_db": 0.598059692415751, "3000–6000 Hz_10ms_log_rms_mae_db": 2.267462661739965, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.43402612500556953, "3000–6000 Hz_10ms_envelope_correlation": 0.8832721849715338, "3000–6000 Hz_10ms_modulation_mae_db": 0.8035335356276868, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7038484653178088, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.35560984978571747, "3000–6000 Hz_40ms_envelope_correlation": 0.9161315326333258, "3000–6000 Hz_40ms_modulation_mae_db": 0.9555784920288793, "mr_stft_db": 5.990609062376685, "band_mae_db": 0.22333741703867724, "rms_gain_error_db": -0.024983136723843082, "output_peak": 0.5850864997982428}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.783969771923012, "spectral_convergence": 0.2679848654605698, "spectral_convergence_status": "ok", "amplitude_floor": 1.726332693189446e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.965662451428738, "spectral_convergence": 0.29522038687145896, "spectral_convergence_status": "ok", "amplitude_floor": 1.4419419879485463e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.2311575789510965, "spectral_convergence": 0.3002040708357989, "spectral_convergence_status": "ok", "amplitude_floor": 1.208114881083552e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024021305054769385, "output_power": 0.00025974553849940807, "delta_db": 0.33951498149264614, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0023981297439394636, "output_power": 0.00241412861645941, "delta_db": 0.028877284349664575, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0027873624928200164, "output_power": 0.002647098121766552, "delta_db": -0.2242341192598449, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955109173134605, "output_power": 0.0005056240554868129, "delta_db": 0.08774500195752138, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00027155220884703173, "output_power": 0.0002892114224079671, "delta_db": 0.273621016655618, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6324739911259635e-05, "output_power": 4.477860928461073e-05, "delta_db": 0.9086808792577112, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3104455471621677, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04802775376999885, "pitch": null, "output_rms_dbfs": -22.103779497692038, "output_peak": 0.5957269730501199, "reference_rms_dbfs": -22.055751743922045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9935966007676145}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024330354027503592, "log_rms_mae_db": 0.8282408171222271, "rms_envelope_nrmse": 0.08252761565967646, "envelope_correlation": 0.993400757408658, "rms_gain_error_db": -0.04812003449234729, "reference_mean_power": 0.006233969602179056, "output_mean_power": 0.006165278178307677}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8072421002875186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017414213578018356, "output_power": 0.0017047616607253674, "reference_active": true, "delta_db": -0.0924019685491243}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002371798239966289, "output_power": 0.00021020187786117243, "reference_active": true, "delta_db": -0.5244115092483682}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.501630810521967e-05, "output_power": 6.714559574691729e-05, "reference_active": true, "delta_db": 0.1399522661794929}], "active_band_count": 3, "active_band_mae_db": 0.2522552479923285}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022924128639936837, "log_rms_mae_db": 0.46782327953606057, "rms_envelope_nrmse": 0.04726200300707512, "envelope_correlation": 0.9976771646825406, "rms_gain_error_db": -0.04850916757669317, "reference_mean_power": 0.006262494274967633, "output_mean_power": 0.006192933621702446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.530420990426951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00159929971914277, "output_power": 0.0015642219046347553, "reference_active": true, "delta_db": -0.09631497724902142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010914154654466241, "output_power": 9.779856528556643e-05, "reference_active": true, "delta_db": -0.47657619786960126}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.773901654262189e-06, "output_power": 4.342668204459263e-06, "reference_active": true, "delta_db": -0.4111681902829927}], "active_band_count": 3, "active_band_mae_db": 0.3280197884672051}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00021113569197140015, "log_rms_mae_db": 1.5290828334957869, "rms_envelope_nrmse": 0.10297382757865543, "envelope_correlation": 0.9913276282631434, "rms_gain_error_db": -0.13241370841193575, "reference_mean_power": 0.0043153114305736935, "output_mean_power": 0.0041857257788591885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.429798497423617e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014993372451038876, "output_power": 0.0014186296208932082, "reference_active": true, "delta_db": -0.24030305808229746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023063830674655453, "output_power": 0.00020452046924329982, "reference_active": true, "delta_db": -0.5219466049396736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9359539965921096e-05, "output_power": 5.720410945532133e-05, "reference_active": true, "delta_db": -0.16063297557063433}], "active_band_count": 3, "active_band_mae_db": 0.30762754619753513}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001994421712728477, "log_rms_mae_db": 1.1308399151659385, "rms_envelope_nrmse": 0.06894805568670014, "envelope_correlation": 0.9959268304708354, "rms_gain_error_db": -0.13248693553301502, "reference_mean_power": 0.004336654014858995, "output_mean_power": 0.004206356536239244}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.221204429766169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014035050384262429, "output_power": 0.0013237384416128954, "reference_active": true, "delta_db": -0.25411795006186016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010608827780695941, "output_power": 9.41768506991804e-05, "reference_active": true, "delta_db": -0.517232358934719}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7233786501496316e-06, "output_power": 4.115357037246265e-06, "reference_active": true, "delta_db": -0.5984524283790518}], "active_band_count": 3, "active_band_mae_db": 0.4566009124585437}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011279772332784735, "log_rms_mae_db": 1.6971869439654794, "rms_envelope_nrmse": 0.15680399731598466, "envelope_correlation": 0.9828534269576452, "rms_gain_error_db": 0.14692458243747497, "reference_mean_power": 0.0007258838363645759, "output_mean_power": 0.0007508610632708605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4338411984908935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00026856860022608645, "output_power": 0.0002823987075915867, "reference_active": true, "delta_db": 0.21807469180707226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.89737119770249e-05, "output_power": 2.277441679910373e-05, "reference_active": true, "delta_db": 0.7929496083772876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310520152307534e-06, "output_power": 5.7486859379905566e-06, "reference_active": true, "delta_db": 0.34431521698434503}], "active_band_count": 3, "active_band_mae_db": 0.4517798390562349}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010728134925657107, "log_rms_mae_db": 1.1450189253164982, "rms_envelope_nrmse": 0.0992782456885892, "envelope_correlation": 0.9928901824696384, "rms_gain_error_db": 0.1469321993668052, "reference_mean_power": 0.0007295243282152702, "output_mean_power": 0.0007546281457730397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.220316674170232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000246564866107528, "output_power": 0.00026277212488828774, "reference_active": true, "delta_db": 0.27648100443544943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.89658406854447e-06, "output_power": 1.1061307797322328e-05, "reference_active": true, "delta_db": 0.9458319039562159}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.477241032022853e-07, "output_power": 3.2644953132471173e-07, "reference_active": true, "delta_db": -0.2741874629514823}], "active_band_count": 3, "active_band_mae_db": 0.4988334571143825}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.800534128114117e-05, "log_rms_mae_db": 2.261412980223361, "rms_envelope_nrmse": 0.4399508188547662, "envelope_correlation": 0.893404263557606, "rms_gain_error_db": 0.9767622061753534, "reference_mean_power": 3.947976781678993e-05, "output_mean_power": 4.9436852250406354e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2531812232892484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6595680637210926e-05, "output_power": 1.856675658735396e-05, "reference_active": true, "delta_db": 0.48740974999616055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.516181974676621e-06, "output_power": 7.5116396526752475e-06, "reference_active": true, "delta_db": -0.5451018707209586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1029203509908326e-06, "output_power": 1.9836606365456757e-06, "reference_active": true, "delta_db": -0.25355448570962413}], "active_band_count": 3, "active_band_mae_db": 0.4286887021422478}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.0235226592584236e-05, "log_rms_mae_db": 1.6466706960294182, "rms_envelope_nrmse": 0.3444153921162822, "envelope_correlation": 0.9328702698077802, "rms_gain_error_db": 0.9767302800502019, "reference_mean_power": 3.967785924701116e-05, "output_mean_power": 4.9684538533913265e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9337568551130464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8381327852746327e-05, "output_power": 1.9884587126802222e-05, "reference_active": true, "delta_db": 0.341396966744652}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.503290667812924e-06, "output_power": 3.953855554732131e-06, "reference_active": true, "delta_db": -0.565091803961774}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6513948696073343e-07, "output_power": 1.441260723457345e-07, "reference_active": true, "delta_db": -0.5910837942858568}], "active_band_count": 3, "active_band_mae_db": 0.49919085499742755}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8822134375466543, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c0de5fac2cdb7a65cbeef09274d23441d3fc3a27e0f5f4ecf275b9f0bd2d82b5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9813749612909617, "standard_deviation": 0.00392035102394638, "minimum": 0.9753716550974311, "maximum": 0.9860579930378202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7481242719339312, "standard_deviation": 0.035872697335125464, "minimum": 1.6971869439654794, "maximum": 1.8136435148743346, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7786747290701865, "standard_deviation": 0.34414763880343524, "minimum": 0.29724288823233386, "maximum": 1.321571565639412, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16051576507752302, "standard_deviation": 0.016448551363674055, "minimum": 0.1402103617955362, "maximum": 0.18607873867150893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930355788548897, "standard_deviation": 0.0021962388535792326, "minimum": 0.9898931429382463, "maximum": 0.9956621154677042, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1618788779598117, "standard_deviation": 0.026666664244316834, "minimum": 1.139091140631303, "maximum": 1.2229965028707654, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7273915026951587, "standard_deviation": 0.3047716998609389, "minimum": 0.23514115768627877, "maximum": 1.2015284355335385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09576631546397528, "standard_deviation": 0.014612194527806713, "minimum": 0.07653061334628522, "maximum": 0.11537085768752653, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8858672503504719, "standard_deviation": 0.005017153080080898, "minimum": 0.8812724081590702, "maximum": 0.8937898669790856, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.278558333552121, "standard_deviation": 0.026503122784213934, "minimum": 2.249339604372035, "maximum": 2.329875488662005, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6555751837061843, "standard_deviation": 0.2975962084287964, "minimum": 0.2470042039875333, "maximum": 1.1968165064121659, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.435456106146792, "standard_deviation": 0.005566148882977599, "minimum": 0.4253819232021422, "maximum": 0.443583045346786, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9236060303884569, "standard_deviation": 0.005532206310057057, "minimum": 0.9161315326333258, "maximum": 0.9328702698077802, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.711375945034422, "standard_deviation": 0.03915237564755505, "minimum": 1.6466706960294182, "maximum": 1.7834187246374307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7900559552487196, "standard_deviation": 0.3899239686750803, "minimum": 0.3658713461415298, "maximum": 1.5675167617035168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.345880771230509, "standard_deviation": 0.006015923550924332, "minimum": 0.33591112246230115, "maximum": 0.35560984978571747, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9909173687414505, "standard_deviation": 0.0007227663203559941, "minimum": 0.989746897635067, "maximum": 0.9918717331899587, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4819489568915853, "standard_deviation": 0.03815217772776851, "minimum": 1.430170506065066, "maximum": 1.5346161302244372, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.45501164152850326, "standard_deviation": 0.2806017385430236, "minimum": 0.19308700322938968, "maximum": 1.0695760780907426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10526766374219332, "standard_deviation": 0.0042909866908653616, "minimum": 0.09940570665025666, "maximum": 0.11190291519672839, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965724693993339, "standard_deviation": 0.0006861565343311588, "minimum": 0.9956431758522173, "maximum": 0.997648386868499, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0523848724478428, "standard_deviation": 0.05208849711246028, "minimum": 0.9812922367433322, "maximum": 1.1308399151659385, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5619741119711674, "standard_deviation": 0.34251457274931923, "minimum": 0.1704394597200283, "maximum": 1.2849821289776464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.063188174804897, "standard_deviation": 0.0064835289579112864, "minimum": 0.051480489685141836, "maximum": 0.07028512207292976, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23637271164912677, "standard_deviation": 0.034019801778914074, "minimum": 0.20045963051521987, "maximum": 0.3104455471621677, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.99786815522682, "standard_deviation": 0.00447868486152208, "minimum": 5.990609062376685, "maximum": 6.005440814061186, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.660890714053394, "standard_deviation": 0.08846639913470396, "minimum": 0.5730017812869915, "maximum": 0.8034345199229584, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06858324708603053, "standard_deviation": 0.030342152307786942, "minimum": -0.10652443661602047, "maximum": -0.024983136723843082, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931079598907031, "standard_deviation": 0.0008356259818187738, "minimum": 0.9914554029797162, "maximum": 0.9941907012954322, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8422544609989349, "standard_deviation": 0.018038467179547348, "minimum": 0.8175447883147478, "maximum": 0.8644369425575948, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3509817645592901, "standard_deviation": 0.16856328363981343, "minimum": 0.13188324551721986, "maximum": 0.6476090008658903, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08439305008420582, "standard_deviation": 0.005106698849660735, "minimum": 0.07744528341179087, "maximum": 0.09422049913745358, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997597823714061, "standard_deviation": 0.0005839048844901286, "minimum": 0.9967309908113335, "maximum": 0.9985212300991955, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4691761601636123, "standard_deviation": 0.013654868242325043, "minimum": 0.4447992380657228, "maximum": 0.48447244541497364, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4198366601899576, "standard_deviation": 0.24509059765343053, "minimum": 0.09139710738475652, "maximum": 0.8265518820699636, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04810089047872121, "standard_deviation": 0.0061989646926695775, "minimum": 0.037558118562651684, "maximum": 0.05617258344994061, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41806-0000", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.072784900665283, "predicted_mos": 3.072784900665283, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0a6f9639da3d815291030479d19dd89661f45ff2a58200142995d0486422eafd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7425283999182284, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7714043051159599, "全帯域_10ms_rms_envelope_nrmse": 0.08705367398490636, "全帯域_10ms_envelope_correlation": 0.9926676227618948, "全帯域_10ms_modulation_mae_db": 0.6764448668022921, "全帯域_40ms_log_rms_mae_db": 0.4165526482361037, "全帯域_40ms_rms_envelope_nrmse": 0.043278022129206, "全帯域_40ms_envelope_correlation": 0.9980722125582717, "全帯域_40ms_modulation_mae_db": 0.6112526609825181, "300–1000 Hz_10ms_log_rms_mae_db": 1.1346076465133765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11703397564747706, "300–1000 Hz_10ms_envelope_correlation": 0.9886990137554468, "300–1000 Hz_10ms_modulation_mae_db": 0.7202184039087699, "300–1000 Hz_40ms_log_rms_mae_db": 0.7557255998803712, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056579268372262743, "300–1000 Hz_40ms_envelope_correlation": 0.9972142622218161, "300–1000 Hz_40ms_modulation_mae_db": 0.8423291761929136, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3259398803135638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13358409867216117, "1000–3000 Hz_10ms_envelope_correlation": 0.9871061761556961, "1000–3000 Hz_10ms_modulation_mae_db": 0.5471015014738473, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8447601555642407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08107108270359725, "1000–3000 Hz_40ms_envelope_correlation": 0.9950313660386532, "1000–3000 Hz_40ms_modulation_mae_db": 0.8664341760094493, "3000–6000 Hz_10ms_log_rms_mae_db": 1.605203348691883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4181365146062772, "3000–6000 Hz_10ms_envelope_correlation": 0.8952032241325638, "3000–6000 Hz_10ms_modulation_mae_db": 0.07984354282401468, "3000–6000 Hz_40ms_log_rms_mae_db": 1.231906934859912, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3438621424566289, "3000–6000 Hz_40ms_envelope_correlation": 0.9241954600168487, "3000–6000 Hz_40ms_modulation_mae_db": 0.18151258111990817, "mr_stft_db": 5.603488292291374, "band_mae_db": 0.21861914390022882, "rms_gain_error_db": -0.0607577450574322, "output_peak": 0.6066764760135345}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8508978318141917, "全帯域_10ms_rms_envelope_nrmse": 0.08038471254485247, "全帯域_10ms_envelope_correlation": 0.9937802912888739, "全帯域_10ms_modulation_mae_db": 0.5056887747617292, "全帯域_40ms_log_rms_mae_db": 0.46402951067472104, "全帯域_40ms_rms_envelope_nrmse": 0.04283548899132819, "全帯域_40ms_envelope_correlation": 0.9981585890053111, "全帯域_40ms_modulation_mae_db": 0.6547163950412499, "300–1000 Hz_10ms_log_rms_mae_db": 1.1935835022042713, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10052809720064809, "300–1000 Hz_10ms_envelope_correlation": 0.9917178911484886, "300–1000 Hz_10ms_modulation_mae_db": 0.5518824305933226, "300–1000 Hz_40ms_log_rms_mae_db": 0.7851744021301623, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05629781201366582, "300–1000 Hz_40ms_envelope_correlation": 0.9973143889846681, "300–1000 Hz_40ms_modulation_mae_db": 0.7068055949893969, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3033302358107721, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13925528640815246, "1000–3000 Hz_10ms_envelope_correlation": 0.9862292696854293, "1000–3000 Hz_10ms_modulation_mae_db": 0.5024044392902739, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8219903018151017, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0861403187485705, "1000–3000 Hz_40ms_envelope_correlation": 0.9948477918708348, "1000–3000 Hz_40ms_modulation_mae_db": 0.3458826635234429, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5815186525804512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3994006463180553, "3000–6000 Hz_10ms_envelope_correlation": 0.9041422418612304, "3000–6000 Hz_10ms_modulation_mae_db": 0.5015198102598452, "3000–6000 Hz_40ms_log_rms_mae_db": 1.187644800410083, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3202593235071585, "3000–6000 Hz_40ms_envelope_correlation": 0.9344208492252375, "3000–6000 Hz_40ms_modulation_mae_db": 0.37561365572361777, "mr_stft_db": 5.586295086057614, "band_mae_db": 0.26675327399698306, "rms_gain_error_db": -0.08730868749715323, "output_peak": 0.5587767620181787}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.831368222898553, "全帯域_10ms_rms_envelope_nrmse": 0.1077117764585619, "全帯域_10ms_envelope_correlation": 0.9888442514896442, "全帯域_10ms_modulation_mae_db": 0.5405257515529226, "全帯域_40ms_log_rms_mae_db": 0.46485463500476787, "全帯域_40ms_rms_envelope_nrmse": 0.05889596352984077, "全帯域_40ms_envelope_correlation": 0.9963785041222569, "全帯域_40ms_modulation_mae_db": 0.6748337826093832, "300–1000 Hz_10ms_log_rms_mae_db": 1.2336659880466776, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1318740452569293, "300–1000 Hz_10ms_envelope_correlation": 0.9857413866321805, "300–1000 Hz_10ms_modulation_mae_db": 0.708893146301465, "300–1000 Hz_40ms_log_rms_mae_db": 0.8559088205170414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07272693393789581, "300–1000 Hz_40ms_envelope_correlation": 0.9953017651768608, "300–1000 Hz_40ms_modulation_mae_db": 0.83490961557913, "1000–3000 Hz_10ms_log_rms_mae_db": 1.328206607557359, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1464497544253987, "1000–3000 Hz_10ms_envelope_correlation": 0.9845616875475192, "1000–3000 Hz_10ms_modulation_mae_db": 0.23830890490526413, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8776997375171766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08966730742675069, "1000–3000 Hz_40ms_envelope_correlation": 0.994113227900577, "1000–3000 Hz_40ms_modulation_mae_db": 0.09125968568937946, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5757742002494832, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4017012073600944, "3000–6000 Hz_10ms_envelope_correlation": 0.9028591165231945, "3000–6000 Hz_10ms_modulation_mae_db": 0.12729204650935433, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2236006041356686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3250251693574411, "3000–6000 Hz_40ms_envelope_correlation": 0.9320961777690394, "3000–6000 Hz_40ms_modulation_mae_db": 0.20729278989769892, "mr_stft_db": 5.604572403106967, "band_mae_db": 0.1979690186000339, "rms_gain_error_db": 0.010197073494851772, "output_peak": 0.8963970760039629}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8509631897292258, "全帯域_10ms_rms_envelope_nrmse": 0.0857494974209118, "全帯域_10ms_envelope_correlation": 0.9930691438034447, "全帯域_10ms_modulation_mae_db": 0.7339974342626601, "全帯域_40ms_log_rms_mae_db": 0.45241358931340664, "全帯域_40ms_rms_envelope_nrmse": 0.04626657870396372, "全帯域_40ms_envelope_correlation": 0.9980510422686155, "全帯域_40ms_modulation_mae_db": 0.9306906572862775, "300–1000 Hz_10ms_log_rms_mae_db": 1.2298815040101307, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11292920785675432, "300–1000 Hz_10ms_envelope_correlation": 0.9898529296090264, "300–1000 Hz_10ms_modulation_mae_db": 0.8032870654801876, "300–1000 Hz_40ms_log_rms_mae_db": 0.8504321055431233, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06644358183365681, "300–1000 Hz_40ms_envelope_correlation": 0.9966695039362735, "300–1000 Hz_40ms_modulation_mae_db": 1.3241407957219775, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3035102626674604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13189582047774684, "1000–3000 Hz_10ms_envelope_correlation": 0.9874724230411468, "1000–3000 Hz_10ms_modulation_mae_db": 0.825621782213688, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8312207158477087, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06810546079117931, "1000–3000 Hz_40ms_envelope_correlation": 0.9964682929972131, "1000–3000 Hz_40ms_modulation_mae_db": 0.7459039927047354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5364550551224707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4042407644855507, "3000–6000 Hz_10ms_envelope_correlation": 0.9091366263592763, "3000–6000 Hz_10ms_modulation_mae_db": 0.5401924879652504, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1881972185393244, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33533919055101813, "3000–6000 Hz_40ms_envelope_correlation": 0.9344939405133378, "3000–6000 Hz_40ms_modulation_mae_db": 0.602682816022119, "mr_stft_db": 5.590846178637713, "band_mae_db": 0.2629847379027637, "rms_gain_error_db": -0.15704064948593385, "output_peak": 0.5634595394121261}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8290291477789805, "全帯域_10ms_rms_envelope_nrmse": 0.08326013208113643, "全帯域_10ms_envelope_correlation": 0.9933110892682572, "全帯域_10ms_modulation_mae_db": 0.24654585457533196, "全帯域_40ms_log_rms_mae_db": 0.47669498089648393, "全帯域_40ms_rms_envelope_nrmse": 0.04496704634586357, "全帯域_40ms_envelope_correlation": 0.9979305992169567, "全帯域_40ms_modulation_mae_db": 0.4386352979611788, "300–1000 Hz_10ms_log_rms_mae_db": 1.1679621960530195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10210052707506724, "300–1000 Hz_10ms_envelope_correlation": 0.991404095541239, "300–1000 Hz_10ms_modulation_mae_db": 0.167772615331913, "300–1000 Hz_40ms_log_rms_mae_db": 0.8027903294512171, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0575950577149401, "300–1000 Hz_40ms_envelope_correlation": 0.997094156419184, "300–1000 Hz_40ms_modulation_mae_db": 0.33453573356072935, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2772175239515031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11248506026703838, "1000–3000 Hz_10ms_envelope_correlation": 0.9908852863701608, "1000–3000 Hz_10ms_modulation_mae_db": 0.1825859220909117, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8469541219945613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06197627325011469, "1000–3000 Hz_40ms_envelope_correlation": 0.9971048877670176, "1000–3000 Hz_40ms_modulation_mae_db": 0.0286588398551471, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5962222053932256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41907835874131666, "3000–6000 Hz_10ms_envelope_correlation": 0.8877520797180828, "3000–6000 Hz_10ms_modulation_mae_db": 1.8944442106512323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2642052077412929, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3409610844152558, "3000–6000 Hz_40ms_envelope_correlation": 0.919767044312073, "3000–6000 Hz_40ms_modulation_mae_db": 1.9398361817094425, "mr_stft_db": 5.587150279863497, "band_mae_db": 0.14465793961469275, "rms_gain_error_db": -0.05343642490841714, "output_peak": 0.68152576934016}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.867780221304594, "全帯域_10ms_rms_envelope_nrmse": 0.08826811189464664, "全帯域_10ms_envelope_correlation": 0.992494831726996, "全帯域_10ms_modulation_mae_db": 0.28902433494352747, "全帯域_40ms_log_rms_mae_db": 0.4898975757556602, "全帯域_40ms_rms_envelope_nrmse": 0.05342972606908585, "全帯域_40ms_envelope_correlation": 0.9970123073776389, "全帯域_40ms_modulation_mae_db": 0.35618033224279166, "300–1000 Hz_10ms_log_rms_mae_db": 1.2215543949587413, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11003384494970433, "300–1000 Hz_10ms_envelope_correlation": 0.9900488079103804, "300–1000 Hz_10ms_modulation_mae_db": 0.20835379474291962, "300–1000 Hz_40ms_log_rms_mae_db": 0.761454639927969, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06654969168869608, "300–1000 Hz_40ms_envelope_correlation": 0.9960544550192787, "300–1000 Hz_40ms_modulation_mae_db": 0.46980821705754466, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2707982587423878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13860312144768258, "1000–3000 Hz_10ms_envelope_correlation": 0.9862147475715904, "1000–3000 Hz_10ms_modulation_mae_db": 1.2566240934370054, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8251437622948543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0802724259690449, "1000–3000 Hz_40ms_envelope_correlation": 0.9951064709821248, "1000–3000 Hz_40ms_modulation_mae_db": 1.2121300603625074, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5970414213817568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4186426397868371, "3000–6000 Hz_10ms_envelope_correlation": 0.8977377075359959, "3000–6000 Hz_10ms_modulation_mae_db": 0.24035833857173508, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2460402488511302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3384831111648672, "3000–6000 Hz_40ms_envelope_correlation": 0.9292960037625866, "3000–6000 Hz_40ms_modulation_mae_db": 0.20614890211298775, "mr_stft_db": 5.589309125836652, "band_mae_db": 0.23546246984084207, "rms_gain_error_db": 0.01085377276945826, "output_peak": 0.5733339013309204}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8527005529532325, "全帯域_10ms_rms_envelope_nrmse": 0.08392021675582406, "全帯域_10ms_envelope_correlation": 0.9933842559914047, "全帯域_10ms_modulation_mae_db": 0.4942123968481679, "全帯域_40ms_log_rms_mae_db": 0.4806738332299082, "全帯域_40ms_rms_envelope_nrmse": 0.04970623579645949, "全帯域_40ms_envelope_correlation": 0.9977076813606583, "全帯域_40ms_modulation_mae_db": 0.4305255037874404, "300–1000 Hz_10ms_log_rms_mae_db": 1.1614762264558534, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10154096929873746, "300–1000 Hz_10ms_envelope_correlation": 0.9917223136414236, "300–1000 Hz_10ms_modulation_mae_db": 0.7146881405828577, "300–1000 Hz_40ms_log_rms_mae_db": 0.7878955562975103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06371159822183008, "300–1000 Hz_40ms_envelope_correlation": 0.9967430799961735, "300–1000 Hz_40ms_modulation_mae_db": 0.7337279739064707, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2799985596274015, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14158904902010128, "1000–3000 Hz_10ms_envelope_correlation": 0.9857395990978451, "1000–3000 Hz_10ms_modulation_mae_db": 1.0714445413446363, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7950610793366897, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06837511556951846, "1000–3000 Hz_40ms_envelope_correlation": 0.9969623032228683, "1000–3000 Hz_40ms_modulation_mae_db": 0.7902033853197808, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5705995548216438, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4022929854396215, "3000–6000 Hz_10ms_envelope_correlation": 0.9103640761067712, "3000–6000 Hz_10ms_modulation_mae_db": 0.4718213614493066, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2000714393115326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.330621915339481, "3000–6000 Hz_40ms_envelope_correlation": 0.9368092757796159, "3000–6000 Hz_40ms_modulation_mae_db": 0.5765958088719443, "mr_stft_db": 5.580320534075327, "band_mae_db": 0.2938824575475915, "rms_gain_error_db": -0.15937347354736497, "output_peak": 0.5889008818153156}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8621565755298192, "全帯域_10ms_rms_envelope_nrmse": 0.08497570727316871, "全帯域_10ms_envelope_correlation": 0.9932147291912042, "全帯域_10ms_modulation_mae_db": 0.4401939131435148, "全帯域_40ms_log_rms_mae_db": 0.5009321593245584, "全帯域_40ms_rms_envelope_nrmse": 0.05286694031438317, "全帯域_40ms_envelope_correlation": 0.9973722067447376, "全帯域_40ms_modulation_mae_db": 0.628962182244544, "300–1000 Hz_10ms_log_rms_mae_db": 1.2406401708477885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12168523996180518, "300–1000 Hz_10ms_envelope_correlation": 0.9879449928993403, "300–1000 Hz_10ms_modulation_mae_db": 0.5075412932854366, "300–1000 Hz_40ms_log_rms_mae_db": 0.8286626013769894, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07251576702981188, "300–1000 Hz_40ms_envelope_correlation": 0.9957052783253145, "300–1000 Hz_40ms_modulation_mae_db": 0.7805615460365943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.263597573049611, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1175668780842124, "1000–3000 Hz_10ms_envelope_correlation": 0.9900650115481743, "1000–3000 Hz_10ms_modulation_mae_db": 0.610765461607999, "1000–3000 Hz_40ms_log_rms_mae_db": 0.819350005404089, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05350944575793574, "1000–3000 Hz_40ms_envelope_correlation": 0.9979403161942234, "1000–3000 Hz_40ms_modulation_mae_db": 0.4185782237903461, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6173961340948402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3985720979328623, "3000–6000 Hz_10ms_envelope_correlation": 0.9065984913541318, "3000–6000 Hz_10ms_modulation_mae_db": 0.7503423380117452, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2638067960546855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3226837901040066, "3000–6000 Hz_40ms_envelope_correlation": 0.9351808892217247, "3000–6000 Hz_40ms_modulation_mae_db": 0.9846921993968586, "mr_stft_db": 5.590444770221272, "band_mae_db": 0.19639263464451573, "rms_gain_error_db": -0.15704916824242554, "output_peak": 0.5691111824565284}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.320347802513801, "spectral_convergence": 0.26418903310664266, "spectral_convergence_status": "ok", "amplitude_floor": 1.726332693189446e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.480952247022753, "spectral_convergence": 0.28306914362922103, "spectral_convergence_status": "ok", "amplitude_floor": 1.4419419879485463e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0091648273375675, "spectral_convergence": 0.28771159047288947, "spectral_convergence_status": "ok", "amplitude_floor": 1.208114881083552e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024021305054769385, "output_power": 0.0002557486256132016, "delta_db": 0.27216709986618765, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0023981297439394636, "output_power": 0.002321040000008927, "delta_db": -0.1419005065433335, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0027873624928200164, "output_power": 0.002763468383954926, "delta_db": -0.03738951503120329, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955109173134605, "output_power": 0.0004769861611934668, "delta_db": -0.1654744846217747, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00027155220884703173, "output_power": 0.00028464601416668036, "delta_db": 0.2045176710292201, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6324739911259635e-05, "output_power": 4.066577674261009e-05, "delta_db": 0.4902655863096538, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21861914390022882, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0607577450574322, "pitch": null, "output_rms_dbfs": -22.116509488979474, "output_peak": 0.6066764760135345, "reference_rms_dbfs": -22.055751743922045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.603488292291374}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024330354027503592, "log_rms_mae_db": 0.7714043051159599, "rms_envelope_nrmse": 0.08705367398490636, "envelope_correlation": 0.9926676227618948, "rms_gain_error_db": -0.0606573053734162, "reference_mean_power": 0.006233969602179056, "output_mean_power": 0.006147505836300277}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8072421002875186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017414213578018356, "output_power": 0.001669815946581622, "reference_active": true, "delta_db": -0.182352623876354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002371798239966289, "output_power": 0.00021099126703314402, "reference_active": true, "delta_db": -0.5081326230881884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.501630810521967e-05, "output_power": 4.776803424052935e-05, "reference_active": true, "delta_db": -1.3388493534423338}], "active_band_count": 3, "active_band_mae_db": 0.6764448668022921}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022924128639936837, "log_rms_mae_db": 0.4165526482361037, "rms_envelope_nrmse": 0.043278022129206, "envelope_correlation": 0.9980722125582717, "rms_gain_error_db": -0.060757256233650386, "reference_mean_power": 0.006262494274967633, "output_mean_power": 0.006175492750512712}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.530420990426951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00159929971914277, "output_power": 0.0015365925194369817, "reference_active": true, "delta_db": -0.17371146387067138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010914154654466241, "output_power": 9.430601780760324e-05, "reference_active": true, "delta_db": -0.6345069687916554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.773901654262189e-06, "output_power": 3.769810441203673e-06, "reference_active": true, "delta_db": -1.0255395502852274}], "active_band_count": 3, "active_band_mae_db": 0.6112526609825181}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00021113569197140015, "log_rms_mae_db": 1.1346076465133765, "rms_envelope_nrmse": 0.11703397564747706, "envelope_correlation": 0.9886990137554468, "rms_gain_error_db": -0.0894472927687001, "reference_mean_power": 0.0043153114305736935, "output_mean_power": 0.00422734229094673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.429798497423617e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014993372451038876, "output_power": 0.0014111607132717072, "reference_active": true, "delta_db": -0.2632285226229097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023063830674655453, "output_power": 0.00019347050960877215, "reference_active": true, "delta_db": -0.7631666536999498}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9359539965921096e-05, "output_power": 4.5715610409158466e-05, "reference_active": true, "delta_db": -1.1342600354034502}], "active_band_count": 3, "active_band_mae_db": 0.7202184039087699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001994421712728477, "log_rms_mae_db": 0.7557255998803712, "rms_envelope_nrmse": 0.056579268372262743, "envelope_correlation": 0.9972142622218161, "rms_gain_error_db": -0.08945539206669197, "reference_mean_power": 0.004336654014858995, "output_mean_power": 0.0042482418764621995}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.221204429766169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014035050384262429, "output_power": 0.0013224987292699052, "reference_active": true, "delta_db": -0.2581871262870344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010608827780695941, "output_power": 8.466451443627662e-05, "reference_active": true, "delta_db": -0.9796597733619504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7233786501496316e-06, "output_power": 3.510255264417081e-06, "reference_active": true, "delta_db": -1.2891406289297558}], "active_band_count": 3, "active_band_mae_db": 0.8423291761929136}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011279772332784735, "log_rms_mae_db": 1.3259398803135638, "rms_envelope_nrmse": 0.13358409867216117, "envelope_correlation": 0.9871061761556961, "rms_gain_error_db": -0.09189049945518585, "reference_mean_power": 0.0007258838363645759, "output_mean_power": 0.0007106865165562128}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4338411984908935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00026856860022608645, "output_power": 0.00026088176080073147, "reference_active": true, "delta_db": -0.12611518594495402}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.89737119770249e-05, "output_power": 2.3781651143725847e-05, "reference_active": true, "delta_db": 0.980897004393749}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310520152307534e-06, "output_power": 6.005736761504439e-06, "reference_active": true, "delta_db": 0.5342923140828387}], "active_band_count": 3, "active_band_mae_db": 0.5471015014738473}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010728134925657107, "log_rms_mae_db": 0.8447601555642407, "rms_envelope_nrmse": 0.08107108270359725, "envelope_correlation": 0.9950313660386532, "rms_gain_error_db": -0.09189406925197546, "reference_mean_power": 0.0007295243282152702, "output_mean_power": 0.0007142502028903495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.220316674170232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000246564866107528, "output_power": 0.0002423373955971211, "reference_active": true, "delta_db": -0.07510756313807407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.89658406854447e-06, "output_power": 1.1522977777340515e-05, "reference_active": true, "delta_db": 1.123414373387885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.477241032022853e-07, "output_power": 4.800791611187944e-07, "reference_active": true, "delta_db": 1.4007805915023888}], "active_band_count": 3, "active_band_mae_db": 0.8664341760094493}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.800534128114117e-05, "log_rms_mae_db": 1.605203348691883, "rms_envelope_nrmse": 0.4181365146062772, "envelope_correlation": 0.8952032241325638, "rms_gain_error_db": 0.5959739236617818, "reference_mean_power": 3.947976781678993e-05, "output_mean_power": 4.528683629557294e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2531812232892484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6595680637210926e-05, "output_power": 1.70474520237341e-05, "reference_active": true, "delta_db": 0.11664408101253679}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.516181974676621e-06, "output_power": 8.509867075403933e-06, "reference_active": true, "delta_db": -0.0032215648217554333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1029203509908326e-06, "output_power": 2.1616696355592866e-06, "reference_active": true, "delta_db": 0.11966498263775181}], "active_band_count": 3, "active_band_mae_db": 0.07984354282401468}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.0235226592584236e-05, "log_rms_mae_db": 1.231906934859912, "rms_envelope_nrmse": 0.3438621424566289, "envelope_correlation": 0.9241954600168487, "rms_gain_error_db": 0.5959483664839219, "reference_mean_power": 3.967785924701116e-05, "output_mean_power": 4.551379710277811e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9337568551130464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8381327852746327e-05, "output_power": 1.8783348749688816e-05, "reference_active": true, "delta_db": 0.09396140831000356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.503290667812924e-06, "output_power": 4.460246779505418e-06, "reference_active": true, "delta_db": -0.04171091374912919}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6513948696073343e-07, "output_power": 1.8144185326754612e-07, "reference_active": true, "delta_db": 0.40886542130059167}], "active_band_count": 3, "active_band_mae_db": 0.18151258111990817}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8950956629568667, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "84ffb822fcbd78657aa4b6c5586b33dfa668a811204837aca9383e8fad44a088", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9872842751271953, "standard_deviation": 0.0021653877493454043, "minimum": 0.9845616875475192, "maximum": 0.9908852863701608, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2940748627150072, "standard_deviation": 0.024807895380856367, "minimum": 1.263597573049611, "maximum": 1.328206607557359, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6543570807954533, "standard_deviation": 0.3776197647070399, "minimum": 0.1825859220909117, "maximum": 1.2566240934370054, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13267863360031174, "standard_deviation": 0.011862121059831815, "minimum": 0.11248506026703838, "maximum": 0.1464497544253987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995946832121689, "standard_deviation": 0.0013486741390774598, "minimum": 0.994113227900577, "maximum": 0.9979403161942234, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8327724849718028, "standard_deviation": 0.02432651416399262, "minimum": 0.7950610793366897, "maximum": 0.8776997375171766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5623813784068485, "standard_deviation": 0.40962897995913344, "minimum": 0.0286588398551471, "maximum": 1.2121300603625074, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07363967877708895, "standard_deviation": 0.012603709679713426, "minimum": 0.05350944575793574, "maximum": 0.08966730742675069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9017241954489058, "standard_deviation": 0.007691559338797851, "minimum": 0.8877520797180828, "maximum": 0.9103640761067712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5850263215419693, "standard_deviation": 0.02504234310761991, "minimum": 1.5364550551224707, "maximum": 1.6173961340948402, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5757267670303104, "standard_deviation": 0.5789815982635552, "minimum": 0.07984354282401468, "maximum": 1.8944442106512323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.4077581518338269, "standard_deviation": 0.009161157871406317, "minimum": 0.3985720979328623, "maximum": 0.41907835874131666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9307824550750579, "standard_deviation": 0.005992197425381776, "minimum": 0.919767044312073, "maximum": 0.9368092757796159, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2256841562379537, "standard_deviation": 0.031402720772273136, "minimum": 1.187644800410083, "maximum": 1.2642052077412929, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6342968668568221, "standard_deviation": 0.5939110595488326, "minimum": 0.18151258111990817, "maximum": 1.9398361817094425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.33215446586198216, "standard_deviation": 0.00885747573691521, "minimum": 0.3202593235071585, "maximum": 0.3438621424566289, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9896414288921906, "standard_deviation": 0.00210223798341666, "minimum": 0.9857413866321805, "maximum": 0.9917223136414236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1979214536362324, "standard_deviation": 0.03952537877302679, "minimum": 1.1346076465133765, "maximum": 1.2406401708477885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5478296112783589, "standard_deviation": 0.2420408353796608, "minimum": 0.167772615331913, "maximum": 0.8032870654801876, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11221573840589039, "standard_deviation": 0.011067808845575981, "minimum": 0.10052809720064809, "maximum": 0.1318740452569293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965121112599462, "standard_deviation": 0.0007444007522618728, "minimum": 0.9953017651768608, "maximum": 0.9973143889846681, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.803505506890548, "standard_deviation": 0.03819592107498419, "minimum": 0.7557255998803712, "maximum": 0.8559088205170414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7533523316305946, "standard_deviation": 0.29263936932623114, "minimum": 0.33453573356072935, "maximum": 1.3241407957219775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06405246385159491, "standard_deviation": 0.0067254854655380975, "minimum": 0.05629781201366582, "maximum": 0.07272693393789581, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22709020950595646, "standard_deviation": 0.04788393397326881, "minimum": 0.14465793961469275, "maximum": 0.2938824575475915, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.591553333761302, "standard_deviation": 0.008382947774819701, "minimum": 5.580320534075327, "maximum": 5.604572403106967, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6297726985488408, "standard_deviation": 0.11481282614557763, "minimum": 0.5587767620181787, "maximum": 0.8963970760039629, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08173941280930211, "standard_deviation": 0.07138127480649356, "minimum": -0.15937347354736497, "maximum": 0.01085377276945826, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.992595776940215, "standard_deviation": 0.0015690269151677984, "minimum": 0.9888442514896442, "maximum": 0.9937802912888739, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8395375058905696, "standard_deviation": 0.03061614460226323, "minimum": 0.7714043051159599, "maximum": 0.867780221304594, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4908291658612683, "standard_deviation": 0.16871338960476415, "minimum": 0.24654585457533196, "maximum": 0.7339974342626601, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08766547855175105, "standard_deviation": 0.008451043907291709, "minimum": 0.08038471254485247, "maximum": 0.1077117764585619, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975853928318059, "standard_deviation": 0.0006264359769784024, "minimum": 0.9963785041222569, "maximum": 0.9981585890053111, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4682561165544512, "standard_deviation": 0.025953692899391922, "minimum": 0.4165526482361037, "maximum": 0.5009321593245584, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.590724601519423, "standard_deviation": 0.18222264963576384, "minimum": 0.35618033224279166, "maximum": 0.9306906572862775, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04903075023501635, "standard_deviation": 0.00570057055914557, "minimum": 0.04283548899132819, "maximum": 0.05889596352984077, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41806-0000", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.7264857292175293, "predicted_mos": 2.7264857292175293, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e9a05f6cd70e65c70c7dc59ff0729abe5465e01690fa9c63e4dd7b1bb7aadb97", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8071917998604476, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.861654648035521, "全帯域_10ms_rms_envelope_nrmse": 0.0789298975807124, "全帯域_10ms_envelope_correlation": 0.993976790121826, "全帯域_10ms_modulation_mae_db": 0.16298922291059004, "全帯域_40ms_log_rms_mae_db": 0.4826955381380171, "全帯域_40ms_rms_envelope_nrmse": 0.04186800461737251, "全帯域_40ms_envelope_correlation": 0.9981756056642476, "全帯域_40ms_modulation_mae_db": 0.21058992367944715, "300–1000 Hz_10ms_log_rms_mae_db": 1.2782442584491425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0995999641127924, "300–1000 Hz_10ms_envelope_correlation": 0.991804745666292, "300–1000 Hz_10ms_modulation_mae_db": 0.27878264195229796, "300–1000 Hz_40ms_log_rms_mae_db": 0.8576242623001881, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051617811673567174, "300–1000 Hz_40ms_envelope_correlation": 0.9976357097704183, "300–1000 Hz_40ms_modulation_mae_db": 0.30475531570883974, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4242894974233744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13253627850362207, "1000–3000 Hz_10ms_envelope_correlation": 0.9873065932331118, "1000–3000 Hz_10ms_modulation_mae_db": 0.5079367703417833, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9011977798563691, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07294113433767939, "1000–3000 Hz_40ms_envelope_correlation": 0.9959987538377175, "1000–3000 Hz_40ms_modulation_mae_db": 0.554891688166396, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6905068641955359, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.42482102780975156, "3000–6000 Hz_10ms_envelope_correlation": 0.887952664591455, "3000–6000 Hz_10ms_modulation_mae_db": 1.361923885082491, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2859207938704478, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.35007930771050216, "3000–6000 Hz_40ms_envelope_correlation": 0.9183469995102173, "3000–6000 Hz_40ms_modulation_mae_db": 1.5621005625605477, "mr_stft_db": 5.676701485273878, "band_mae_db": 0.1311740780348266, "rms_gain_error_db": -0.028113636400025174, "output_peak": 0.5976233114832721}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8553716395959062, "全帯域_10ms_rms_envelope_nrmse": 0.09806761578550427, "全帯域_10ms_envelope_correlation": 0.9907059405370474, "全帯域_10ms_modulation_mae_db": 0.8924306603685936, "全帯域_40ms_log_rms_mae_db": 0.48442181766877274, "全帯域_40ms_rms_envelope_nrmse": 0.04841774918999081, "全帯域_40ms_envelope_correlation": 0.9976039961061502, "全帯域_40ms_modulation_mae_db": 1.1111562861706439, "300–1000 Hz_10ms_log_rms_mae_db": 1.315456323349987, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12300493851856936, "300–1000 Hz_10ms_envelope_correlation": 0.9875885359465224, "300–1000 Hz_10ms_modulation_mae_db": 0.9838065242148603, "300–1000 Hz_40ms_log_rms_mae_db": 0.8427206865697795, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06120375441292303, "300–1000 Hz_40ms_envelope_correlation": 0.9969177027931998, "300–1000 Hz_40ms_modulation_mae_db": 1.154033121804907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4383455162350745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1788128510858812, "1000–3000 Hz_10ms_envelope_correlation": 0.9779338368584959, "1000–3000 Hz_10ms_modulation_mae_db": 1.3060803699693895, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9123018664589458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12624909404340995, "1000–3000 Hz_40ms_envelope_correlation": 0.9885773455536804, "1000–3000 Hz_40ms_modulation_mae_db": 1.5355263700673063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7227360823674904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.415390144977059, "3000–6000 Hz_10ms_envelope_correlation": 0.8921671205496271, "3000–6000 Hz_10ms_modulation_mae_db": 1.1455646992489996, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3045711419678665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3393115073634277, "3000–6000 Hz_40ms_envelope_correlation": 0.9227689105489548, "3000–6000 Hz_40ms_modulation_mae_db": 1.2332122697237937, "mr_stft_db": 5.687107645873735, "band_mae_db": 0.18582242918907962, "rms_gain_error_db": -0.07782921795943364, "output_peak": 0.6117404703578844}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8050978408195781, "全帯域_10ms_rms_envelope_nrmse": 0.07803642974979896, "全帯域_10ms_envelope_correlation": 0.9941466367140802, "全帯域_10ms_modulation_mae_db": 0.11854175372508019, "全帯域_40ms_log_rms_mae_db": 0.4395880671812267, "全帯域_40ms_rms_envelope_nrmse": 0.04332467301649882, "全帯域_40ms_envelope_correlation": 0.9980887233378902, "全帯域_40ms_modulation_mae_db": 0.07866849883991328, "300–1000 Hz_10ms_log_rms_mae_db": 1.2540943143750307, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10134748967767081, "300–1000 Hz_10ms_envelope_correlation": 0.9915653556504601, "300–1000 Hz_10ms_modulation_mae_db": 0.23012768987250962, "300–1000 Hz_40ms_log_rms_mae_db": 0.832246228271868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05997448759995722, "300–1000 Hz_40ms_envelope_correlation": 0.9968682294089536, "300–1000 Hz_40ms_modulation_mae_db": 0.05498610797369991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.450809508345295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15648483103332173, "1000–3000 Hz_10ms_envelope_correlation": 0.982349808620097, "1000–3000 Hz_10ms_modulation_mae_db": 0.614863949335165, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9533301924728969, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11074889832005512, "1000–3000 Hz_40ms_envelope_correlation": 0.9908439223338997, "1000–3000 Hz_40ms_modulation_mae_db": 0.6393516021483326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7338798551476182, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4208088860752229, "3000–6000 Hz_10ms_envelope_correlation": 0.8971305576775942, "3000–6000 Hz_10ms_modulation_mae_db": 0.30552926863572394, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3125204397647252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3432590697864232, "3000–6000 Hz_40ms_envelope_correlation": 0.9276123971845036, "3000–6000 Hz_40ms_modulation_mae_db": 0.25707096773893057, "mr_stft_db": 5.688851087873805, "band_mae_db": 0.2655348600654233, "rms_gain_error_db": -0.06297684416829286, "output_peak": 0.5922572845336203}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8421663622934773, "全帯域_10ms_rms_envelope_nrmse": 0.080926654168284, "全帯域_10ms_envelope_correlation": 0.9936621107839423, "全帯域_10ms_modulation_mae_db": 0.19880318699434088, "全帯域_40ms_log_rms_mae_db": 0.48019820938612845, "全帯域_40ms_rms_envelope_nrmse": 0.04235968658443182, "全帯域_40ms_envelope_correlation": 0.9981517061520392, "全帯域_40ms_modulation_mae_db": 0.20525940155397873, "300–1000 Hz_10ms_log_rms_mae_db": 1.2322722878123724, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10757612938597953, "300–1000 Hz_10ms_envelope_correlation": 0.9904413098787767, "300–1000 Hz_10ms_modulation_mae_db": 0.22998262323366062, "300–1000 Hz_40ms_log_rms_mae_db": 0.8074385876614494, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05492398993854168, "300–1000 Hz_40ms_envelope_correlation": 0.9973792317663368, "300–1000 Hz_40ms_modulation_mae_db": 0.285711680112818, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3991705049097136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12789154231080874, "1000–3000 Hz_10ms_envelope_correlation": 0.9883110181413741, "1000–3000 Hz_10ms_modulation_mae_db": 0.40535487509300144, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8665808876717632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06536227098323348, "1000–3000 Hz_40ms_envelope_correlation": 0.9970500446112059, "1000–3000 Hz_40ms_modulation_mae_db": 0.2206749566380953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7631440131360907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4247539790374393, "3000–6000 Hz_10ms_envelope_correlation": 0.89469730333348, "3000–6000 Hz_10ms_modulation_mae_db": 0.635361898739956, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3488762124354519, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34525463940565065, "3000–6000 Hz_40ms_envelope_correlation": 0.9264623726263836, "3000–6000 Hz_40ms_modulation_mae_db": 0.7880776146693274, "mr_stft_db": 5.676011564006166, "band_mae_db": 0.2502267679959778, "rms_gain_error_db": -0.05503692730161125, "output_peak": 0.5954305506302695}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8801539536739917, "全帯域_10ms_rms_envelope_nrmse": 0.0844026832741727, "全帯域_10ms_envelope_correlation": 0.993092174941068, "全帯域_10ms_modulation_mae_db": 0.17815402275404976, "全帯域_40ms_log_rms_mae_db": 0.4807937810586163, "全帯域_40ms_rms_envelope_nrmse": 0.0443213377988349, "全帯域_40ms_envelope_correlation": 0.9979393247768253, "全帯域_40ms_modulation_mae_db": 0.14116435332186372, "300–1000 Hz_10ms_log_rms_mae_db": 1.2084801026392002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10447318953347982, "300–1000 Hz_10ms_envelope_correlation": 0.990980180037695, "300–1000 Hz_10ms_modulation_mae_db": 0.1544697518611079, "300–1000 Hz_40ms_log_rms_mae_db": 0.7984309375995562, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05832191603328884, "300–1000 Hz_40ms_envelope_correlation": 0.9969751571627589, "300–1000 Hz_40ms_modulation_mae_db": 0.13686646693186758, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4462208841285682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13928700690846552, "1000–3000 Hz_10ms_envelope_correlation": 0.9859905913520333, "1000–3000 Hz_10ms_modulation_mae_db": 0.9257201933524261, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9317922416485699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06984519029764913, "1000–3000 Hz_40ms_envelope_correlation": 0.9964157952086965, "1000–3000 Hz_40ms_modulation_mae_db": 0.8133830730140806, "3000–6000 Hz_10ms_log_rms_mae_db": 1.747330193535002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41016348081870335, "3000–6000 Hz_10ms_envelope_correlation": 0.9034400663007174, "3000–6000 Hz_10ms_modulation_mae_db": 0.353597883887935, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3352480833734441, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3297324726557871, "3000–6000 Hz_40ms_envelope_correlation": 0.9344479561966599, "3000–6000 Hz_40ms_modulation_mae_db": 0.3120846320018344, "mr_stft_db": 5.691627645105547, "band_mae_db": 0.24670996852212124, "rms_gain_error_db": -0.02977451150305291, "output_peak": 0.6871422166980583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8326756080109443, "全帯域_10ms_rms_envelope_nrmse": 0.08303792336832293, "全帯域_10ms_envelope_correlation": 0.9933361757310396, "全帯域_10ms_modulation_mae_db": 0.2044173411010245, "全帯域_40ms_log_rms_mae_db": 0.4780905068442054, "全帯域_40ms_rms_envelope_nrmse": 0.044162483767683836, "全帯域_40ms_envelope_correlation": 0.9979996658120839, "全帯域_40ms_modulation_mae_db": 0.41887797759242257, "300–1000 Hz_10ms_log_rms_mae_db": 1.2658887233238603, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10260360956506086, "300–1000 Hz_10ms_envelope_correlation": 0.9912962072973052, "300–1000 Hz_10ms_modulation_mae_db": 0.33308979765806496, "300–1000 Hz_40ms_log_rms_mae_db": 0.8243284734363638, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05583615863356471, "300–1000 Hz_40ms_envelope_correlation": 0.9972386590455189, "300–1000 Hz_40ms_modulation_mae_db": 0.5961452362847343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4081667561027302, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1271941049300011, "1000–3000 Hz_10ms_envelope_correlation": 0.9889902527604394, "1000–3000 Hz_10ms_modulation_mae_db": 0.3452723642109541, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9132220121667756, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07622270491906354, "1000–3000 Hz_40ms_envelope_correlation": 0.9966032776526277, "1000–3000 Hz_40ms_modulation_mae_db": 0.23926022341935316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7754273538966163, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4159113493142384, "3000–6000 Hz_10ms_envelope_correlation": 0.9020356438795796, "3000–6000 Hz_10ms_modulation_mae_db": 0.25926811319236504, "3000–6000 Hz_40ms_log_rms_mae_db": 1.314947431163556, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33581451733731366, "3000–6000 Hz_40ms_envelope_correlation": 0.9331660577696935, "3000–6000 Hz_40ms_modulation_mae_db": 0.39972854766343763, "mr_stft_db": 5.689420288443748, "band_mae_db": 0.23058709252252604, "rms_gain_error_db": -0.06936204940400653, "output_peak": 0.7242156537308434}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8384006161402275, "全帯域_10ms_rms_envelope_nrmse": 0.09477403332206681, "全帯域_10ms_envelope_correlation": 0.9913660021302529, "全帯域_10ms_modulation_mae_db": 0.6057406514972273, "全帯域_40ms_log_rms_mae_db": 0.4331590018882514, "全帯域_40ms_rms_envelope_nrmse": 0.05497645273886347, "全帯域_40ms_envelope_correlation": 0.9969670887254186, "全帯域_40ms_modulation_mae_db": 0.6456269668391488, "300–1000 Hz_10ms_log_rms_mae_db": 1.3031377854362336, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1268963218793609, "300–1000 Hz_10ms_envelope_correlation": 0.9868291684437606, "300–1000 Hz_10ms_modulation_mae_db": 0.5546446291302035, "300–1000 Hz_40ms_log_rms_mae_db": 0.8679425089222171, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08166149142900313, "300–1000 Hz_40ms_envelope_correlation": 0.9943369728763836, "300–1000 Hz_40ms_modulation_mae_db": 0.696180831759742, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4059945984093127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.136505990415706, "1000–3000 Hz_10ms_envelope_correlation": 0.9867053780517765, "1000–3000 Hz_10ms_modulation_mae_db": 0.5132796867368056, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9094490367528694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07953623165937308, "1000–3000 Hz_40ms_envelope_correlation": 0.9952351057669964, "1000–3000 Hz_40ms_modulation_mae_db": 0.6717300481137753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6963676751073906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4070656069548721, "3000–6000 Hz_10ms_envelope_correlation": 0.9101826754926258, "3000–6000 Hz_10ms_modulation_mae_db": 0.34822581039691064, "3000–6000 Hz_40ms_log_rms_mae_db": 1.245928018231867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3286470894214315, "3000–6000 Hz_40ms_envelope_correlation": 0.9397312758504961, "3000–6000 Hz_40ms_modulation_mae_db": 0.522444533312182, "mr_stft_db": 5.673367567942442, "band_mae_db": 0.2623384503280963, "rms_gain_error_db": -0.11622160091949824, "output_peak": 0.5907876047193101}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8370254235698789, "全帯域_10ms_rms_envelope_nrmse": 0.07995866946152491, "全帯域_10ms_envelope_correlation": 0.9938128412856649, "全帯域_10ms_modulation_mae_db": 0.47602599868329626, "全帯域_40ms_log_rms_mae_db": 0.46424933829853626, "全帯域_40ms_rms_envelope_nrmse": 0.042274832530355154, "全帯域_40ms_envelope_correlation": 0.9981453978988506, "全帯域_40ms_modulation_mae_db": 0.573691970492737, "300–1000 Hz_10ms_log_rms_mae_db": 1.273408174663723, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10501512001124139, "300–1000 Hz_10ms_envelope_correlation": 0.9909249968798124, "300–1000 Hz_10ms_modulation_mae_db": 0.46306006101605995, "300–1000 Hz_40ms_log_rms_mae_db": 0.86392683461727, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055742374137254753, "300–1000 Hz_40ms_envelope_correlation": 0.99734242982779, "300–1000 Hz_40ms_modulation_mae_db": 0.6063326087302335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4026666631965394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14500532883548786, "1000–3000 Hz_10ms_envelope_correlation": 0.9853023033435979, "1000–3000 Hz_10ms_modulation_mae_db": 0.8004880549938278, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9066317365166954, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0971211251093225, "1000–3000 Hz_40ms_envelope_correlation": 0.993122313627141, "1000–3000 Hz_40ms_modulation_mae_db": 0.8786903182381517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7817523679354625, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.429156309275477, "3000–6000 Hz_10ms_envelope_correlation": 0.8852432518024084, "3000–6000 Hz_10ms_modulation_mae_db": 1.4545885728050572, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3761490682619781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34441305743833334, "3000–6000 Hz_40ms_envelope_correlation": 0.9206937842383537, "3000–6000 Hz_40ms_modulation_mae_db": 1.4103547765614823, "mr_stft_db": 5.680762786458362, "band_mae_db": 0.16860873512516242, "rms_gain_error_db": -0.048305066790888314, "output_peak": 0.5906924004356756}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.45259885753818, "spectral_convergence": 0.25915900056688895, "spectral_convergence_status": "ok", "amplitude_floor": 1.726332693189446e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5701021566746896, "spectral_convergence": 0.2806307383958693, "spectral_convergence_status": "ok", "amplitude_floor": 1.4419419879485463e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.007403441608767, "spectral_convergence": 0.29192313209981857, "spectral_convergence_status": "ok", "amplitude_floor": 1.208114881083552e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024021305054769385, "output_power": 0.0002423614061474929, "delta_db": 0.03866865032532096, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0023981297439394636, "output_power": 0.00243760207148155, "delta_db": 0.07090134622329637, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0027873624928200164, "output_power": 0.002716084883066144, "delta_db": -0.11250113422651574, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955109173134605, "output_power": 0.00048821624039368627, "delta_db": -0.0644100547180856, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00027155220884703173, "output_power": 0.0002648752962334261, "delta_db": -0.10811884331504981, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6324739911259635e-05, "output_power": 3.9760053246376084e-05, "delta_db": 0.3924444394006912, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1311740780348266, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.028113636400025174, "pitch": null, "output_rms_dbfs": -22.083865380322067, "output_peak": 0.5976233114832721, "reference_rms_dbfs": -22.055751743922045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.676701485273878}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024330354027503592, "log_rms_mae_db": 0.861654648035521, "rms_envelope_nrmse": 0.0789298975807124, "envelope_correlation": 0.993976790121826, "rms_gain_error_db": -0.028026869184043544, "reference_mean_power": 0.006233969602179056, "output_mean_power": 0.006193868679482855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8072421002875186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017414213578018356, "output_power": 0.0017514736039885787, "reference_active": true, "delta_db": 0.02499729966156572}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002371798239966289, "output_power": 0.00021567868467488899, "reference_active": true, "delta_db": -0.412705161500566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.501630810521967e-05, "output_power": 6.578832445066134e-05, "reference_active": true, "delta_db": 0.05126520756963835}], "active_band_count": 3, "active_band_mae_db": 0.16298922291059004}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022924128639936837, "log_rms_mae_db": 0.4826955381380171, "rms_envelope_nrmse": 0.04186800461737251, "envelope_correlation": 0.9981756056642476, "rms_gain_error_db": -0.027989356889601174, "reference_mean_power": 0.006262494274967633, "output_mean_power": 0.006222263607860418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.530420990426951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00159929971914277, "output_power": 0.0016116516537026356, "reference_active": true, "delta_db": 0.03341317172494748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010914154654466241, "output_power": 9.870239814008807e-05, "reference_active": true, "delta_db": -0.43662398769440147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.773901654262189e-06, "output_power": 4.599389684305451e-06, "reference_active": true, "delta_db": -0.16173261161899244}], "active_band_count": 3, "active_band_mae_db": 0.21058992367944715}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00021113569197140015, "log_rms_mae_db": 1.2782442584491425, "rms_envelope_nrmse": 0.0995999641127924, "envelope_correlation": 0.991804745666292, "rms_gain_error_db": -0.0482911313677012, "reference_mean_power": 0.0043153114305736935, "output_mean_power": 0.0042675933579976065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.429798497423617e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014993372451038876, "output_power": 0.0014661978129370205, "reference_active": true, "delta_db": -0.09706762103122832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023063830674655453, "output_power": 0.00020749312223086798, "reference_active": true, "delta_db": -0.4592773523811247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9359539965921096e-05, "output_power": 6.33126981272346e-05, "reference_active": true, "delta_db": 0.2800029524445409}], "active_band_count": 3, "active_band_mae_db": 0.27878264195229796}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001994421712728477, "log_rms_mae_db": 0.8576242623001881, "rms_envelope_nrmse": 0.051617811673567174, "envelope_correlation": 0.9976357097704183, "rms_gain_error_db": -0.04827328380835261, "reference_mean_power": 0.004336654014858995, "output_mean_power": 0.004288717563830617}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.221204429766169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014035050384262429, "output_power": 0.0013705517801760245, "reference_active": true, "delta_db": -0.10318527979666328}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010608827780695941, "output_power": 9.274170230805135e-05, "reference_active": true, "delta_db": -0.5839233604344993}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7233786501496316e-06, "output_power": 4.482672713487128e-06, "reference_active": true, "delta_db": -0.22715730689535663}], "active_band_count": 3, "active_band_mae_db": 0.30475531570883974}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011279772332784735, "log_rms_mae_db": 1.4242894974233744, "rms_envelope_nrmse": 0.13253627850362207, "envelope_correlation": 0.9873065932331118, "rms_gain_error_db": -0.0938416668012855, "reference_mean_power": 0.0007258838363645759, "output_mean_power": 0.0007103672960888592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4338411984908935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00026856860022608645, "output_power": 0.00027158839565346324, "reference_active": true, "delta_db": 0.04855973963671292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.89737119770249e-05, "output_power": 1.9713648758460774e-05, "reference_active": true, "delta_db": 0.16614710728320686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310520152307534e-06, "output_power": 7.178726500713751e-06, "reference_active": true, "delta_db": 1.3091034641054302}], "active_band_count": 3, "active_band_mae_db": 0.5079367703417833}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010728134925657107, "log_rms_mae_db": 0.9011977798563691, "rms_envelope_nrmse": 0.07294113433767939, "envelope_correlation": 0.9959987538377175, "rms_gain_error_db": -0.0938448819449035, "reference_mean_power": 0.0007295243282152702, "output_mean_power": 0.0007139294400158011}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.220316674170232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000246564866107528, "output_power": 0.00025164108771972605, "reference_active": true, "delta_db": 0.0885036123231853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.89658406854447e-06, "output_power": 9.395329525085352e-06, "reference_active": true, "delta_db": 0.23688729851059026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.477241032022853e-07, "output_power": 4.7332910614049206e-07, "reference_active": true, "delta_db": 1.3392841536654125}], "active_band_count": 3, "active_band_mae_db": 0.554891688166396}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.800534128114117e-05, "log_rms_mae_db": 1.6905068641955359, "rms_envelope_nrmse": 0.42482102780975156, "envelope_correlation": 0.887952664591455, "rms_gain_error_db": 0.4646563184626459, "reference_mean_power": 3.947976781678993e-05, "output_mean_power": 4.393799364801861e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2531812232892484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6595680637210926e-05, "output_power": 1.4678807376516755e-05, "reference_active": true, "delta_db": -0.5330429732881902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.516181974676621e-06, "output_power": 6.538466222965204e-06, "reference_active": true, "delta_db": -1.1476904835668358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1029203509908326e-06, "output_power": 1.2087013413630416e-06, "reference_active": true, "delta_db": -2.405038198392448}], "active_band_count": 3, "active_band_mae_db": 1.361923885082491}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.0235226592584236e-05, "log_rms_mae_db": 1.2859207938704478, "rms_envelope_nrmse": 0.35007930771050216, "envelope_correlation": 0.9183469995102173, "rms_gain_error_db": 0.4646308184437891, "reference_mean_power": 3.967785924701116e-05, "output_mean_power": 4.415819513791973e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9337568551130464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8381327852746327e-05, "output_power": 1.550196874167751e-05, "reference_active": true, "delta_db": -0.7399002443887752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.503290667812924e-06, "output_power": 3.4127428566554133e-06, "reference_active": true, "delta_db": -1.20426413394963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6513948696073343e-07, "output_power": 8.782885077119205e-08, "reference_active": true, "delta_db": -2.7421373093432386}], "active_band_count": 3, "active_band_mae_db": 1.5621005625605477}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.888797883889831, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "440aeadc9e438036e1b23d929ab9d686eb85163ee36081c4a8b2732af458c203", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9853612227951157, "standard_deviation": 0.003627211129312669, "minimum": 0.9779338368584959, "maximum": 0.9889902527604394, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.421957991093826, "standard_deviation": 0.02081049822757139, "minimum": 1.3991705049097136, "maximum": 1.450809508345295, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.677374533004169, "standard_deviation": 0.31980662184024766, "minimum": 0.3452723642109541, "maximum": 1.3060803699693895, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14296474175291177, "standard_deviation": 0.017369727886369358, "minimum": 0.1271941049300011, "maximum": 0.1788128510858812, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942308198239956, "standard_deviation": 0.0030971534127644785, "minimum": 0.9885773455536804, "maximum": 0.9970500446112059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9118132191931106, "standard_deviation": 0.024831237156018798, "minimum": 0.8665808876717632, "maximum": 0.9533301924728969, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6941885349756863, "standard_deviation": 0.41579314234922243, "minimum": 0.2206749566380953, "maximum": 1.5355263700673063, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08725333120872328, "standard_deviation": 0.021838412286136682, "minimum": 0.06536227098323348, "maximum": 0.12624909404340995, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8966061604534359, "standard_deviation": 0.008350584298455319, "minimum": 0.8852432518024084, "maximum": 0.9101826754926258, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.738893050665151, "standard_deviation": 0.0343652260106235, "minimum": 1.6905068641955359, "maximum": 1.7817523679354625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7330075164986798, "standard_deviation": 0.5064058818926988, "minimum": 0.25926811319236504, "maximum": 1.4545885728050572, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.41850884803284544, "standard_deviation": 0.007695732670597178, "minimum": 0.4070656069548721, "maximum": 0.429156309275477, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9279037192406578, "standard_deviation": 0.00738969362535531, "minimum": 0.9183469995102173, "maximum": 0.9397312758504961, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3155201486336672, "standard_deviation": 0.039688209329667636, "minimum": 1.245928018231867, "maximum": 1.3761490682619781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.810634238028942, "standard_deviation": 0.5224021646660912, "minimum": 0.25707096773893057, "maximum": 1.5621005625605477, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3395639576398587, "standard_deviation": 0.007653964529331973, "minimum": 0.3286470894214315, "maximum": 0.35007930771050216, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9901788124750781, "standard_deviation": 0.0018902859028800195, "minimum": 0.9868291684437606, "maximum": 0.991804745666292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2663727462561938, "standard_deviation": 0.03508497708535541, "minimum": 1.2084801026392002, "maximum": 1.315456323349987, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4034954648673456, "standard_deviation": 0.26881806914823586, "minimum": 0.1544697518611079, "maximum": 0.9838065242148603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10881459533551938, "standard_deviation": 0.010298236528386186, "minimum": 0.0995999641127924, "maximum": 0.1268963218793609, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99683676158142, "standard_deviation": 0.0010437194935120416, "minimum": 0.9943369728763836, "maximum": 0.9976357097704183, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8368323149223366, "standard_deviation": 0.02587047514577466, "minimum": 0.7984309375995562, "maximum": 0.8679425089222171, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4793764211633552, "standard_deviation": 0.3580377903027549, "minimum": 0.05498610797369991, "maximum": 1.154033121804907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05991024798226256, "standard_deviation": 0.009296030805023036, "minimum": 0.051617811673567174, "maximum": 0.08166149142900313, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21762529772290168, "standard_deviation": 0.049654856596127166, "minimum": 0.1311740780348266, "maximum": 0.2655348600654233, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.682981258872211, "standard_deviation": 0.007101910558129187, "minimum": 5.673367567942442, "maximum": 5.691627645105547, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6237361865736167, "standard_deviation": 0.05197782298757676, "minimum": 0.5906924004356756, "maximum": 0.7242156537308434, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06095248180585111, "standard_deviation": 0.028434699908994016, "minimum": -0.11622160091949824, "maximum": -0.028113636400025174, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930123340306152, "standard_deviation": 0.00127732686919537, "minimum": 0.9907059405370474, "maximum": 0.9941466367140802, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8440682615174406, "standard_deviation": 0.022274430438729124, "minimum": 0.8050978408195781, "maximum": 0.8801539536739917, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35463785475427534, "standard_deviation": 0.2770963097536174, "minimum": 0.11854175372508019, "maximum": 0.8924306603685936, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08476673833879837, "standard_deviation": 0.00753477810394964, "minimum": 0.07803642974979896, "maximum": 0.09806761578550427, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978839385591882, "standard_deviation": 0.0004144719188666134, "minimum": 0.9969670887254186, "maximum": 0.9981756056642476, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4678995325579693, "standard_deviation": 0.020476642444341466, "minimum": 0.4331590018882514, "maximum": 0.48442181766877274, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4231294223112694, "standard_deviation": 0.3456611244645303, "minimum": 0.07866849883991328, "maximum": 1.1111562861706439, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04521315253050391, "standard_deviation": 0.004456505210484809, "minimum": 0.04186800461737251, "maximum": 0.05497645273886347, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_4992-41806-0000", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.610459566116333, "predicted_mos": 2.610459566116333, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "51109adb211bd7ade8fbc5495b1bfe41a9cb4692b0afeef4d8cb684b23bafbb2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.030819300096482, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8182107519723105, "全帯域_10ms_rms_envelope_nrmse": 0.07642911102420166, "全帯域_10ms_envelope_correlation": 0.9945273186217276, "全帯域_10ms_modulation_mae_db": 0.44208458354476626, "全帯域_40ms_log_rms_mae_db": 0.44163113460666337, "全帯域_40ms_rms_envelope_nrmse": 0.04262938748628197, "全帯域_40ms_envelope_correlation": 0.9983583467261192, "全帯域_40ms_modulation_mae_db": 0.5410266286255286, "300–1000 Hz_10ms_log_rms_mae_db": 1.249244558374566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10253892129780576, "300–1000 Hz_10ms_envelope_correlation": 0.9917421849417862, "300–1000 Hz_10ms_modulation_mae_db": 0.5577789908895301, "300–1000 Hz_40ms_log_rms_mae_db": 0.8306221455908258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06381316179563329, "300–1000 Hz_40ms_envelope_correlation": 0.9969761609773286, "300–1000 Hz_40ms_modulation_mae_db": 0.7784940127686159, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3811984773824666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12710456010040602, "1000–3000 Hz_10ms_envelope_correlation": 0.9883726265152479, "1000–3000 Hz_10ms_modulation_mae_db": 0.6322852935827529, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8718604137288596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06392413539178095, "1000–3000 Hz_40ms_envelope_correlation": 0.9969854508947206, "1000–3000 Hz_40ms_modulation_mae_db": 0.4253781330317057, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7137701710177853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40513469076730685, "3000–6000 Hz_10ms_envelope_correlation": 0.9053584020253477, "3000–6000 Hz_10ms_modulation_mae_db": 0.48709768561896444, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2630071292158942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3286254711206083, "3000–6000 Hz_40ms_envelope_correlation": 0.9344158776343144, "3000–6000 Hz_40ms_modulation_mae_db": 0.5662424418734474, "mr_stft_db": 5.667015915585534, "band_mae_db": 0.27326081616667075, "rms_gain_error_db": -0.1468613548322389, "output_peak": 0.6434296147731933}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8430928981659782, "全帯域_10ms_rms_envelope_nrmse": 0.0907457864398811, "全帯域_10ms_envelope_correlation": 0.99205522069382, "全帯域_10ms_modulation_mae_db": 0.3499246424477575, "全帯域_40ms_log_rms_mae_db": 0.4760603015235544, "全帯域_40ms_rms_envelope_nrmse": 0.04343828515802796, "全帯域_40ms_envelope_correlation": 0.9981090999107552, "全帯域_40ms_modulation_mae_db": 0.5288864815986302, "300–1000 Hz_10ms_log_rms_mae_db": 1.238015421632297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11540726999514268, "300–1000 Hz_10ms_envelope_correlation": 0.9890498514227057, "300–1000 Hz_10ms_modulation_mae_db": 0.38792334040085824, "300–1000 Hz_40ms_log_rms_mae_db": 0.8203968156883984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05618349941125024, "300–1000 Hz_40ms_envelope_correlation": 0.9973732352477716, "300–1000 Hz_40ms_modulation_mae_db": 0.7133913424880075, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3803894017216214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.142823795631884, "1000–3000 Hz_10ms_envelope_correlation": 0.9853065486743144, "1000–3000 Hz_10ms_modulation_mae_db": 0.3882535721061455, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8902713164024513, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07694279286250118, "1000–3000 Hz_40ms_envelope_correlation": 0.9955078758997423, "1000–3000 Hz_40ms_modulation_mae_db": 0.3603541057920007, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7169627264172223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39884602468556, "3000–6000 Hz_10ms_envelope_correlation": 0.9072873643875576, "3000–6000 Hz_10ms_modulation_mae_db": 0.23786575665549856, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2630553250982952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3248411608692964, "3000–6000 Hz_40ms_envelope_correlation": 0.9350641420450008, "3000–6000 Hz_40ms_modulation_mae_db": 0.2288042344275871, "mr_stft_db": 5.67156855220781, "band_mae_db": 0.20227585794603012, "rms_gain_error_db": -0.09073882825887301, "output_peak": 0.7719771529233725}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8253528298154343, "全帯域_10ms_rms_envelope_nrmse": 0.08260076989574285, "全帯域_10ms_envelope_correlation": 0.9933844868947386, "全帯域_10ms_modulation_mae_db": 0.2669211101255397, "全帯域_40ms_log_rms_mae_db": 0.41265768300238953, "全帯域_40ms_rms_envelope_nrmse": 0.04139658734741841, "全帯域_40ms_envelope_correlation": 0.9982037814325199, "全帯域_40ms_modulation_mae_db": 0.4658166542303048, "300–1000 Hz_10ms_log_rms_mae_db": 1.2516926476499326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10390765685943969, "300–1000 Hz_10ms_envelope_correlation": 0.9910923097569143, "300–1000 Hz_10ms_modulation_mae_db": 0.23263280613082027, "300–1000 Hz_40ms_log_rms_mae_db": 0.7878857833727706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05467571736523113, "300–1000 Hz_40ms_envelope_correlation": 0.9974127365354226, "300–1000 Hz_40ms_modulation_mae_db": 0.43421087399517827, "1000–3000 Hz_10ms_log_rms_mae_db": 1.445048694339238, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14124806900099793, "1000–3000 Hz_10ms_envelope_correlation": 0.9861253406576153, "1000–3000 Hz_10ms_modulation_mae_db": 0.33274806719807665, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9418845317634063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09086603060312916, "1000–3000 Hz_40ms_envelope_correlation": 0.994068038381185, "1000–3000 Hz_40ms_modulation_mae_db": 0.23364083223083917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6920263023389408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.41866404160479503, "3000–6000 Hz_10ms_envelope_correlation": 0.8985259824196256, "3000–6000 Hz_10ms_modulation_mae_db": 0.6208665208982753, "3000–6000 Hz_40ms_log_rms_mae_db": 1.255454326911206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34490820459204885, "3000–6000 Hz_40ms_envelope_correlation": 0.9272804662140964, "3000–6000 Hz_40ms_modulation_mae_db": 0.546685383860335, "mr_stft_db": 5.6757540838438345, "band_mae_db": 0.243643328274737, "rms_gain_error_db": -0.024098801012271444, "output_peak": 0.6051717689038161}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8125801442252217, "全帯域_10ms_rms_envelope_nrmse": 0.0779482864044646, "全帯域_10ms_envelope_correlation": 0.9942353180368484, "全帯域_10ms_modulation_mae_db": 0.6563838614240612, "全帯域_40ms_log_rms_mae_db": 0.436149361442924, "全帯域_40ms_rms_envelope_nrmse": 0.048035307563982665, "全帯域_40ms_envelope_correlation": 0.9977557070365104, "全帯域_40ms_modulation_mae_db": 0.9094180698384237, "300–1000 Hz_10ms_log_rms_mae_db": 1.2474552842847078, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11230792214546155, "300–1000 Hz_10ms_envelope_correlation": 0.989807537772216, "300–1000 Hz_10ms_modulation_mae_db": 0.9284489488435765, "300–1000 Hz_40ms_log_rms_mae_db": 0.8675190622396756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07153396559573665, "300–1000 Hz_40ms_envelope_correlation": 0.9958220668284576, "300–1000 Hz_40ms_modulation_mae_db": 1.1730738118910886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3777137294522377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14341295563255596, "1000–3000 Hz_10ms_envelope_correlation": 0.985457582237394, "1000–3000 Hz_10ms_modulation_mae_db": 0.6634919214158519, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9051262583328865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10001244507593954, "1000–3000 Hz_40ms_envelope_correlation": 0.9925649953619399, "1000–3000 Hz_40ms_modulation_mae_db": 0.6263883182172986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7272017847407406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4099132429455349, "3000–6000 Hz_10ms_envelope_correlation": 0.8952579041742478, "3000–6000 Hz_10ms_modulation_mae_db": 1.0399914837395448, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3123483378853755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3346641242596395, "3000–6000 Hz_40ms_envelope_correlation": 0.9250602596707597, "3000–6000 Hz_40ms_modulation_mae_db": 1.1329955645888052, "mr_stft_db": 5.656615881169856, "band_mae_db": 0.25359163435850407, "rms_gain_error_db": -0.12180613908603699, "output_peak": 0.6512062858529271}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8252303767554027, "全帯域_10ms_rms_envelope_nrmse": 0.07759763793159884, "全帯域_10ms_envelope_correlation": 0.9941891031143723, "全帯域_10ms_modulation_mae_db": 0.7164978220618378, "全帯域_40ms_log_rms_mae_db": 0.4587810469268207, "全帯域_40ms_rms_envelope_nrmse": 0.03919427232915817, "全帯域_40ms_envelope_correlation": 0.9983950301882736, "全帯域_40ms_modulation_mae_db": 0.9353153557200852, "300–1000 Hz_10ms_log_rms_mae_db": 1.2729710893626414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1023198120538177, "300–1000 Hz_10ms_envelope_correlation": 0.9913873061243459, "300–1000 Hz_10ms_modulation_mae_db": 0.8696351336742497, "300–1000 Hz_40ms_log_rms_mae_db": 0.8308366485261269, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055469595444738444, "300–1000 Hz_40ms_envelope_correlation": 0.9972593293857397, "300–1000 Hz_40ms_modulation_mae_db": 0.9888552328340917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3965594314404628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12716636883733445, "1000–3000 Hz_10ms_envelope_correlation": 0.9883194940358083, "1000–3000 Hz_10ms_modulation_mae_db": 0.22775552228810736, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9303287037465484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07565260352998956, "1000–3000 Hz_40ms_envelope_correlation": 0.9956822883284162, "1000–3000 Hz_40ms_modulation_mae_db": 0.2505037346336039, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7192219948804066, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40553043992373566, "3000–6000 Hz_10ms_envelope_correlation": 0.9001631428146144, "3000–6000 Hz_10ms_modulation_mae_db": 0.9548898657661175, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2636788210241363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32751593327965817, "3000–6000 Hz_40ms_envelope_correlation": 0.9306315420503524, "3000–6000 Hz_40ms_modulation_mae_db": 1.2521849985026283, "mr_stft_db": 5.6716977843975505, "band_mae_db": 0.1493266984781277, "rms_gain_error_db": 0.005564056422603766, "output_peak": 0.6293230753685704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8555630539326383, "全帯域_10ms_rms_envelope_nrmse": 0.08080639834794683, "全帯域_10ms_envelope_correlation": 0.9937889814956915, "全帯域_10ms_modulation_mae_db": 0.8617334637751264, "全帯域_40ms_log_rms_mae_db": 0.4597817854259181, "全帯域_40ms_rms_envelope_nrmse": 0.04231432426777509, "全帯域_40ms_envelope_correlation": 0.9983084411958998, "全帯域_40ms_modulation_mae_db": 0.8849333327723629, "300–1000 Hz_10ms_log_rms_mae_db": 1.2214615830231037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0981751280875523, "300–1000 Hz_10ms_envelope_correlation": 0.9921925294581023, "300–1000 Hz_10ms_modulation_mae_db": 1.0066084688148242, "300–1000 Hz_40ms_log_rms_mae_db": 0.7755526566907073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055615549784078976, "300–1000 Hz_40ms_envelope_correlation": 0.9975009996382266, "300–1000 Hz_40ms_modulation_mae_db": 1.1901389882060096, "1000–3000 Hz_10ms_log_rms_mae_db": 1.426948379288817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13323512401347545, "1000–3000 Hz_10ms_envelope_correlation": 0.9871778946220471, "1000–3000 Hz_10ms_modulation_mae_db": 0.4028798005931184, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9274044940806748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06728593991748995, "1000–3000 Hz_40ms_envelope_correlation": 0.9965736369763372, "1000–3000 Hz_40ms_modulation_mae_db": 0.12334294144166152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7262796410980592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40845567171648184, "3000–6000 Hz_10ms_envelope_correlation": 0.9069125641928879, "3000–6000 Hz_10ms_modulation_mae_db": 0.21104071144571565, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2904861056843497, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3281382170771718, "3000–6000 Hz_40ms_envelope_correlation": 0.9375333681412116, "3000–6000 Hz_40ms_modulation_mae_db": 0.26526340592713865, "mr_stft_db": 5.6787598628851645, "band_mae_db": 0.2996552526603331, "rms_gain_error_db": -0.12502257599048916, "output_peak": 0.5468441847972483}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8534166189505349, "全帯域_10ms_rms_envelope_nrmse": 0.07906070784406342, "全帯域_10ms_envelope_correlation": 0.9940069962055514, "全帯域_10ms_modulation_mae_db": 0.21624225929782334, "全帯域_40ms_log_rms_mae_db": 0.48385710887659544, "全帯域_40ms_rms_envelope_nrmse": 0.04547231671019887, "全帯域_40ms_envelope_correlation": 0.9979508575240773, "全帯域_40ms_modulation_mae_db": 0.33404169849238174, "300–1000 Hz_10ms_log_rms_mae_db": 1.2412386563342277, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10041189494054024, "300–1000 Hz_10ms_envelope_correlation": 0.9917274278225728, "300–1000 Hz_10ms_modulation_mae_db": 0.1732202517615777, "300–1000 Hz_40ms_log_rms_mae_db": 0.8086017100333207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057661235316085295, "300–1000 Hz_40ms_envelope_correlation": 0.9971710675672273, "300–1000 Hz_40ms_modulation_mae_db": 0.2958805595814163, "1000–3000 Hz_10ms_log_rms_mae_db": 1.421353413876095, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13048248112045865, "1000–3000 Hz_10ms_envelope_correlation": 0.9880220866722792, "1000–3000 Hz_10ms_modulation_mae_db": 0.20024748500620623, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8854886867421122, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07804327069992176, "1000–3000 Hz_40ms_envelope_correlation": 0.9959210796426279, "1000–3000 Hz_40ms_modulation_mae_db": 0.2661761858597545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7495779398693743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.411029573704586, "3000–6000 Hz_10ms_envelope_correlation": 0.8948426625212886, "3000–6000 Hz_10ms_modulation_mae_db": 1.0701341873259305, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3381373480146417, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3266519268008584, "3000–6000 Hz_40ms_envelope_correlation": 0.9288252985218108, "3000–6000 Hz_40ms_modulation_mae_db": 1.2305290399743807, "mr_stft_db": 5.6678092668072635, "band_mae_db": 0.2586224647953264, "rms_gain_error_db": -0.09625322951464044, "output_peak": 0.5692342545555042}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7645901377176016, "全帯域_10ms_rms_envelope_nrmse": 0.07326369961350104, "全帯域_10ms_envelope_correlation": 0.9948157242292762, "全帯域_10ms_modulation_mae_db": 0.45619129974952055, "全帯域_40ms_log_rms_mae_db": 0.4297589911204588, "全帯域_40ms_rms_envelope_nrmse": 0.040333596028222875, "全帯域_40ms_envelope_correlation": 0.9983141344924703, "全帯域_40ms_modulation_mae_db": 0.4017436007070185, "300–1000 Hz_10ms_log_rms_mae_db": 1.174278686235512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08774838883645562, "300–1000 Hz_10ms_envelope_correlation": 0.9936907992618833, "300–1000 Hz_10ms_modulation_mae_db": 0.42826693000201627, "300–1000 Hz_40ms_log_rms_mae_db": 0.7529698612800846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04850780111655176, "300–1000 Hz_40ms_envelope_correlation": 0.9979789576416433, "300–1000 Hz_40ms_modulation_mae_db": 0.306102121068476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.418351362262189, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15065955994578004, "1000–3000 Hz_10ms_envelope_correlation": 0.9836003039543177, "1000–3000 Hz_10ms_modulation_mae_db": 0.7349441076514425, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9549128553585013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07843314836372835, "1000–3000 Hz_40ms_envelope_correlation": 0.9953476610568488, "1000–3000 Hz_40ms_modulation_mae_db": 0.4604508588963127, "3000–6000 Hz_10ms_log_rms_mae_db": 1.668823893116084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39291657884637543, "3000–6000 Hz_10ms_envelope_correlation": 0.9076130723269128, "3000–6000 Hz_10ms_modulation_mae_db": 0.5263299500192825, "3000–6000 Hz_40ms_log_rms_mae_db": 1.262232943154772, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32317184392446563, "3000–6000 Hz_40ms_envelope_correlation": 0.9335708106712901, "3000–6000 Hz_40ms_modulation_mae_db": 0.570238577288995, "mr_stft_db": 5.657776300844103, "band_mae_db": 0.18489916891835653, "rms_gain_error_db": -0.03896341830567095, "output_peak": 0.5747862189130214}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.449872222932832, "spectral_convergence": 0.2621974758214525, "spectral_convergence_status": "ok", "amplitude_floor": 1.726332693189446e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.56139133103651, "spectral_convergence": 0.28402124910774607, "spectral_convergence_status": "ok", "amplitude_floor": 1.4419419879485463e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9897841927872575, "spectral_convergence": 0.28541944898873844, "spectral_convergence_status": "ok", "amplitude_floor": 1.208114881083552e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024021305054769385, "output_power": 0.0002608002531862, "delta_db": 0.35711410159329593, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0023981297439394636, "output_power": 0.0023895564740930605, "delta_db": -0.01555376687741997, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0027873624928200164, "output_power": 0.002579298354797043, "delta_db": -0.33691870686378195, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955109173134605, "output_power": 0.0004791514624416581, "delta_db": -0.145804094751025, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00027155220884703173, "output_power": 0.0002699142495234332, "delta_db": -0.026275269859609356, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6324739911259635e-05, "output_power": 4.325062713196642e-05, "delta_db": 0.7578989570548922, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27326081616667075, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1468613548322389, "pitch": null, "output_rms_dbfs": -22.202613098754284, "output_peak": 0.6434296147731933, "reference_rms_dbfs": -22.055751743922045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.667015915585534}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024330354027503592, "log_rms_mae_db": 0.8182107519723105, "rms_envelope_nrmse": 0.07642911102420166, "envelope_correlation": 0.9945273186217276, "rms_gain_error_db": -0.14678377582694693, "reference_mean_power": 0.006233969602179056, "output_mean_power": 0.006026793381629551}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8072421002875186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017414213578018356, "output_power": 0.0016955519775788308, "reference_active": true, "delta_db": -0.11592758869229114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002371798239966289, "output_power": 0.00022675749315274874, "reference_active": true, "delta_db": -0.19516095313734055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.501630810521967e-05, "output_power": 5.1464266099046275e-05, "reference_active": true, "delta_db": -1.0151652088046672}], "active_band_count": 3, "active_band_mae_db": 0.44208458354476626}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022924128639936837, "log_rms_mae_db": 0.44163113460666337, "rms_envelope_nrmse": 0.04262938748628197, "envelope_correlation": 0.9983583467261192, "rms_gain_error_db": -0.14709189627055888, "reference_mean_power": 0.006262494274967633, "output_mean_power": 0.006053940555272362}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.530420990426951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00159929971914277, "output_power": 0.0015634918443477581, "reference_active": true, "delta_db": -0.0983424080518617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010914154654466241, "output_power": 0.00010518558944154549, "reference_active": true, "delta_db": -0.16033858478005025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.773901654262189e-06, "output_power": 3.4868525188097346e-06, "reference_active": true, "delta_db": -1.364398893044674}], "active_band_count": 3, "active_band_mae_db": 0.5410266286255286}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00021113569197140015, "log_rms_mae_db": 1.249244558374566, "rms_envelope_nrmse": 0.10253892129780576, "envelope_correlation": 0.9917421849417862, "rms_gain_error_db": -0.2461421668460249, "reference_mean_power": 0.0043153114305736935, "output_mean_power": 0.0040775371541433205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.429798497423617e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014993372451038876, "output_power": 0.0014375986172670756, "reference_active": true, "delta_db": -0.1826168307423293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023063830674655453, "output_power": 0.00021989534998165594, "reference_active": true, "delta_db": -0.20715395329882286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9359539965921096e-05, "output_power": 4.4170661942760945e-05, "reference_active": true, "delta_db": -1.283566188627438}], "active_band_count": 3, "active_band_mae_db": 0.5577789908895301}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001994421712728477, "log_rms_mae_db": 0.8306221455908258, "rms_envelope_nrmse": 0.06381316179563329, "envelope_correlation": 0.9969761609773286, "rms_gain_error_db": -0.2460980136593914, "reference_mean_power": 0.004336654014858995, "output_mean_power": 0.004097745419104039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.221204429766169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014035050384262429, "output_power": 0.001346963460626326, "reference_active": true, "delta_db": -0.17858161360943509}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010608827780695941, "output_power": 9.915173308731648e-05, "reference_active": true, "delta_db": -0.29367089608326225}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7233786501496316e-06, "output_power": 3.075599696488352e-06, "reference_active": true, "delta_db": -1.8632295286131502}], "active_band_count": 3, "active_band_mae_db": 0.7784940127686159}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011279772332784735, "log_rms_mae_db": 1.3811984773824666, "rms_envelope_nrmse": 0.12710456010040602, "envelope_correlation": 0.9883726265152479, "rms_gain_error_db": -0.10843596773392443, "reference_mean_power": 0.0007258838363645759, "output_mean_power": 0.0007079841402955402}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4338411984908935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00026856860022608645, "output_power": 0.0002607639535610398, "reference_active": true, "delta_db": -0.12807678683949023}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.89737119770249e-05, "output_power": 1.867328599289738e-05, "reference_active": true, "delta_db": -0.06931554910320542}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310520152307534e-06, "output_power": 7.853864707032026e-06, "reference_active": true, "delta_db": 1.699463544805563}], "active_band_count": 3, "active_band_mae_db": 0.6322852935827529}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010728134925657107, "log_rms_mae_db": 0.8718604137288596, "rms_envelope_nrmse": 0.06392413539178095, "envelope_correlation": 0.9969854508947206, "rms_gain_error_db": -0.108420246755361, "reference_mean_power": 0.0007295243282152702, "output_mean_power": 0.0007115374363049574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.220316674170232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000246564866107528, "output_power": 0.00024193077877385656, "reference_active": true, "delta_db": -0.08240069093124926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.89658406854447e-06, "output_power": 8.99732566048455e-06, "reference_active": true, "delta_db": 0.04890153175161521}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.477241032022853e-07, "output_power": 4.5260360976468433e-07, "reference_active": true, "delta_db": 1.1448321764122527}], "active_band_count": 3, "active_band_mae_db": 0.4253781330317057}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.800534128114117e-05, "log_rms_mae_db": 1.7137701710177853, "rms_envelope_nrmse": 0.40513469076730685, "envelope_correlation": 0.9053584020253477, "rms_gain_error_db": 0.7564528739162891, "reference_mean_power": 3.947976781678993e-05, "output_mean_power": 4.699156157426262e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2531812232892484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6595680637210926e-05, "output_power": 1.744397756672678e-05, "reference_active": true, "delta_db": 0.2165045069931047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.516181974676621e-06, "output_power": 8.251555983958763e-06, "reference_active": true, "delta_db": -0.13709082362254255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1029203509908326e-06, "output_power": 1.6294950716096319e-06, "reference_active": true, "delta_db": -1.1076977262412462}], "active_band_count": 3, "active_band_mae_db": 0.48709768561896444}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.0235226592584236e-05, "log_rms_mae_db": 1.2630071292158942, "rms_envelope_nrmse": 0.3286254711206083, "envelope_correlation": 0.9344158776343144, "rms_gain_error_db": 0.7564422708278856, "reference_mean_power": 3.967785924701116e-05, "output_mean_power": 4.722722844908639e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9337568551130464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8381327852746327e-05, "output_power": 1.8966580082406674e-05, "reference_active": true, "delta_db": 0.13612147806904948}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.503290667812924e-06, "output_power": 4.266148763061345e-06, "reference_active": true, "delta_db": -0.23493984210406632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6513948696073343e-07, "output_power": 1.2164221803081733e-07, "reference_active": true, "delta_db": -1.3276660054472265}], "active_band_count": 3, "active_band_mae_db": 0.5662424418734474}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8888668712050005, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4b36f08087bc9d31b8667ecce2d104ec1080ca33f0cadcefbe241402faba0cd1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.986547734671128, "standard_deviation": 0.0017169290033004112, "minimum": 0.9836003039543177, "maximum": 0.9883726265152479, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.405945361220391, "standard_deviation": 0.025388834478377097, "minimum": 1.3777137294522377, "maximum": 1.445048694339238, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4478257212302127, "standard_deviation": 0.20406999158588726, "minimum": 0.20024748500620623, "maximum": 0.7349441076514425, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13701661428536158, "standard_deviation": 0.008709128475093295, "minimum": 0.12710456010040602, "maximum": 0.15065955994578004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953313783177272, "standard_deviation": 0.0014157465032588671, "minimum": 0.9925649953619399, "maximum": 0.9969854508947206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.91340965751943, "standard_deviation": 0.029581271271185056, "minimum": 0.8718604137288596, "maximum": 0.9549128553585013, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3432793887628971, "standard_deviation": 0.15850031498749398, "minimum": 0.12334294144166152, "maximum": 0.6263883182172986, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07889504580556006, "standard_deviation": 0.011731594220294449, "minimum": 0.06392413539178095, "maximum": 0.10001244507593954, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9019951368578103, "standard_deviation": 0.005437974194783258, "minimum": 0.8948426625212886, "maximum": 0.9076130723269128, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7142330566848267, "standard_deviation": 0.024349408351985187, "minimum": 1.668823893116084, "maximum": 1.7495779398693743, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6435270201836661, "standard_deviation": 0.3436045949504378, "minimum": 0.21104071144571565, "maximum": 1.0701341873259305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.406311283024297, "standard_deviation": 0.007833697329888896, "minimum": 0.39291657884637543, "maximum": 0.41866404160479503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9315477206186046, "standard_deviation": 0.0042938492369167325, "minimum": 0.9250602596707597, "maximum": 0.9375333681412116, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2810500421235838, "standard_deviation": 0.029971231032800187, "minimum": 1.255454326911206, "maximum": 1.3381373480146417, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7241179558054147, "standard_deviation": 0.420621466381195, "minimum": 0.2288042344275871, "maximum": 1.2521849985026283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3298146102404684, "standard_deviation": 0.006966314334429064, "minimum": 0.32317184392446563, "maximum": 0.34490820459204885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913362433200659, "standard_deviation": 0.0014253627659622868, "minimum": 0.9890498514227057, "maximum": 0.9936907992618833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2370447408621235, "standard_deviation": 0.029187625120702755, "minimum": 1.174278686235512, "maximum": 1.2729710893626414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5730643588146815, "standard_deviation": 0.3236971717995383, "minimum": 0.1732202517615777, "maximum": 1.0066084688148242, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10285212427702695, "standard_deviation": 0.0084894155579025, "minimum": 0.08774838883645562, "maximum": 0.11540726999514268, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971868192277271, "standard_deviation": 0.0006238605621663054, "minimum": 0.9958220668284576, "maximum": 0.9979789576416433, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8092980854277387, "standard_deviation": 0.03625718374291945, "minimum": 0.7529698612800846, "maximum": 0.8675190622396756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7350183678541105, "standard_deviation": 0.3651055962404433, "minimum": 0.2958805595814163, "maximum": 1.1901389882060096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.057932565728663225, "standard_deviation": 0.006903212258335557, "minimum": 0.04850780111655176, "maximum": 0.07153396559573665, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2331594026997607, "standard_deviation": 0.0500233424650463, "minimum": 0.1493266984781277, "maximum": 0.2996552526603331, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.66837470596764, "standard_deviation": 0.007894956714191133, "minimum": 5.656615881169856, "maximum": 5.6787598628851645, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6239965695109566, "standard_deviation": 0.07048005794292589, "minimum": 0.5468441847972483, "maximum": 0.7719771529233725, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07977253632220213, "standard_deviation": 0.054451098906982444, "minimum": -0.1468613548322389, "maximum": 0.005564056422603766, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9938753936615032, "standard_deviation": 0.0008549011877052641, "minimum": 0.99205522069382, "maximum": 0.9948157242292762, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8247546014418903, "standard_deviation": 0.029077117729654555, "minimum": 0.7645901377176016, "maximum": 0.8555630539326383, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.49574738030330406, "standard_deviation": 0.22824606854090593, "minimum": 0.21624225929782334, "maximum": 0.8617334637751264, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07980654968767505, "standard_deviation": 0.005231483773211447, "minimum": 0.07326369961350104, "maximum": 0.0907457864398811, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981744248133282, "standard_deviation": 0.0002237374897175682, "minimum": 0.9977557070365104, "maximum": 0.9983950301882736, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44983467661566556, "standard_deviation": 0.024095167770681204, "minimum": 0.41265768300238953, "maximum": 0.48385710887659544, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6251477277480919, "standard_deviation": 0.24520639920452142, "minimum": 0.33404169849238174, "maximum": 0.9353153557200852, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04285175961138325, "standard_deviation": 0.0028348513335044716, "minimum": 0.03919427232915817, "maximum": 0.048035307563982665, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5105-28240-0011", "language": "en", "style": "normal", "speaker": "librispeech_5105", "source_utterance_id": "5105-28240-0011", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5105-28240-0011.wav", "source_file_sha256": "fc1351af6c67d17a820120299468d6cea16d4a39aeb70eb25aad817fe7531fa1", "decoded_pcm_sha256": "b9bc3828aa0284c0660c9ec7acd1980026fcc21a26b62a3c5c68458711b73ba2", "original_source_sha256": "2e162ff81e45f909e947ebfae4f39e339d73d5fc19afafb1c903db96f7f8532d", "corpus_entry_id": "librispeech_5105-28240-0011"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.9131500720977783, "predicted_mos": 3.9131500720977783, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c1978e86ebbcccd9c43b5680cc6f89db524854c938b34fb6986e5c7e6bd3fc76", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9846279980010877, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5105-28240-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c1978e86ebbcccd9c43b5680cc6f89db524854c938b34fb6986e5c7e6bd3fc76", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "3ac3db42bbe992b387196f6c806baaddd2efa79e26f56a4565441675620504a3"}, "rows": [{"case": "librispeech_5105-28240-0011", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.8853964805603027, "predicted_mos": 1.8853964805603027, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "47ce5a94716d6e3e2771d3c85211fb4d6e992deeb0d6a7c621b8619f9b65178d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.770301047999965, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5105-28240-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "47ce5a94716d6e3e2771d3c85211fb4d6e992deeb0d6a7c621b8619f9b65178d", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2213290113154007, "全帯域_10ms_rms_envelope_nrmse": 0.07342340761909585, "全帯域_10ms_envelope_correlation": 0.9949114822204569, "全帯域_10ms_modulation_mae_db": 0.2892789957026915, "全帯域_40ms_log_rms_mae_db": 0.6896151267442013, "全帯域_40ms_rms_envelope_nrmse": 0.043905838638546274, "全帯域_40ms_envelope_correlation": 0.9980986151369761, "全帯域_40ms_modulation_mae_db": 0.12749372500489206, "300–1000 Hz_10ms_log_rms_mae_db": 1.6835380555147352, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09969069869653954, "300–1000 Hz_10ms_envelope_correlation": 0.9925632010974296, "300–1000 Hz_10ms_modulation_mae_db": 0.41981217281515404, "300–1000 Hz_40ms_log_rms_mae_db": 1.1213911021759004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06304859275362172, "300–1000 Hz_40ms_envelope_correlation": 0.9968962699596899, "300–1000 Hz_40ms_modulation_mae_db": 0.18402542078734777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.558839083132504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12470484970182143, "1000–3000 Hz_10ms_envelope_correlation": 0.9884466305151257, "1000–3000 Hz_10ms_modulation_mae_db": 1.2307300921395978, "1000–3000 Hz_40ms_log_rms_mae_db": 0.957049481628198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07634153114386863, "1000–3000 Hz_40ms_envelope_correlation": 0.9957606353935848, "1000–3000 Hz_40ms_modulation_mae_db": 1.2095314253618026, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5958467499314242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20393826084884292, "3000–6000 Hz_10ms_envelope_correlation": 0.9656702633406995, "3000–6000 Hz_10ms_modulation_mae_db": 0.8473575835103878, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1509979698244668, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1287276398469526, "3000–6000 Hz_40ms_envelope_correlation": 0.9857349929576618, "3000–6000 Hz_40ms_modulation_mae_db": 0.8143320805051658, "mr_stft_db": 5.6218478998830586, "band_mae_db": 0.1675252556231341, "rms_gain_error_db": -0.029298818565698596, "output_peak": 0.6074618846159984}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3074461374497346, "全帯域_10ms_rms_envelope_nrmse": 0.08036946403272403, "全帯域_10ms_envelope_correlation": 0.9939469104816469, "全帯域_10ms_modulation_mae_db": 0.765430403329478, "全帯域_40ms_log_rms_mae_db": 0.6561438680248486, "全帯域_40ms_rms_envelope_nrmse": 0.037382152330089416, "全帯域_40ms_envelope_correlation": 0.9987242721550104, "全帯域_40ms_modulation_mae_db": 0.6344825136700405, "300–1000 Hz_10ms_log_rms_mae_db": 1.6820864314473059, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11378283418508375, "300–1000 Hz_10ms_envelope_correlation": 0.9902879909827171, "300–1000 Hz_10ms_modulation_mae_db": 1.0607027849281563, "300–1000 Hz_40ms_log_rms_mae_db": 1.000120866730683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05131402610156062, "300–1000 Hz_40ms_envelope_correlation": 0.9980775268377112, "300–1000 Hz_40ms_modulation_mae_db": 0.8414005158707903, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6291854407288957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11911064290263837, "1000–3000 Hz_10ms_envelope_correlation": 0.9895927323148975, "1000–3000 Hz_10ms_modulation_mae_db": 0.4952220767187178, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0116726649173864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06608784184782951, "1000–3000 Hz_40ms_envelope_correlation": 0.9970674992663249, "1000–3000 Hz_40ms_modulation_mae_db": 0.37809991494608436, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6040467612695906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18969125269519113, "3000–6000 Hz_10ms_envelope_correlation": 0.9703799681555882, "3000–6000 Hz_10ms_modulation_mae_db": 0.6823161846272899, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1620715991145527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1220017793717015, "3000–6000 Hz_40ms_envelope_correlation": 0.9871586324090038, "3000–6000 Hz_40ms_modulation_mae_db": 0.7523563830033334, "mr_stft_db": 5.640295833341878, "band_mae_db": 0.16886377738086744, "rms_gain_error_db": -0.08825739853693144, "output_peak": 0.624667839616073}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2884021122239517, "全帯域_10ms_rms_envelope_nrmse": 0.08024896153273697, "全帯域_10ms_envelope_correlation": 0.9939874294883316, "全帯域_10ms_modulation_mae_db": 0.4144289880462118, "全帯域_40ms_log_rms_mae_db": 0.7129689563899667, "全帯域_40ms_rms_envelope_nrmse": 0.044853022264296036, "全帯域_40ms_envelope_correlation": 0.9981163384324643, "全帯域_40ms_modulation_mae_db": 0.2630661993487701, "300–1000 Hz_10ms_log_rms_mae_db": 1.6642019791332874, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11558912448944189, "300–1000 Hz_10ms_envelope_correlation": 0.9900761857683149, "300–1000 Hz_10ms_modulation_mae_db": 0.3572445889632427, "300–1000 Hz_40ms_log_rms_mae_db": 1.0110646825468246, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06711325039722935, "300–1000 Hz_40ms_envelope_correlation": 0.9967254393533509, "300–1000 Hz_40ms_modulation_mae_db": 0.3157548360408013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5604364505870474, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11623709833518511, "1000–3000 Hz_10ms_envelope_correlation": 0.990030091045696, "1000–3000 Hz_10ms_modulation_mae_db": 0.3740702222052022, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9444639205914962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06853568443662608, "1000–3000 Hz_40ms_envelope_correlation": 0.9966872487048506, "1000–3000 Hz_40ms_modulation_mae_db": 0.2311405249791015, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5628969450575168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2018614380389807, "3000–6000 Hz_10ms_envelope_correlation": 0.9663509959853331, "3000–6000 Hz_10ms_modulation_mae_db": 0.35068683088521313, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1484878199129789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12543652200799654, "3000–6000 Hz_40ms_envelope_correlation": 0.9863034564105825, "3000–6000 Hz_40ms_modulation_mae_db": 0.4745203083457849, "mr_stft_db": 5.623495282140301, "band_mae_db": 0.2114432079230858, "rms_gain_error_db": -0.10402616877630674, "output_peak": 0.6244100091677796}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.274793807664178, "全帯域_10ms_rms_envelope_nrmse": 0.07969453490491218, "全帯域_10ms_envelope_correlation": 0.9940491096178747, "全帯域_10ms_modulation_mae_db": 0.5118508524847108, "全帯域_40ms_log_rms_mae_db": 0.68327495995749, "全帯域_40ms_rms_envelope_nrmse": 0.04035600755352026, "全帯域_40ms_envelope_correlation": 0.9983974606257882, "全帯域_40ms_modulation_mae_db": 0.46589388850818597, "300–1000 Hz_10ms_log_rms_mae_db": 1.7050029073909028, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09868640947765668, "300–1000 Hz_10ms_envelope_correlation": 0.9926731796276194, "300–1000 Hz_10ms_modulation_mae_db": 0.09706488247458132, "300–1000 Hz_40ms_log_rms_mae_db": 1.0614941673442233, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05345361698562176, "300–1000 Hz_40ms_envelope_correlation": 0.9977439498771687, "300–1000 Hz_40ms_modulation_mae_db": 0.082691382146117, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5762151097504347, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13596041057090494, "1000–3000 Hz_10ms_envelope_correlation": 0.9859902071892351, "1000–3000 Hz_10ms_modulation_mae_db": 0.4980166252466895, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9908290464507344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07613355880424512, "1000–3000 Hz_40ms_envelope_correlation": 0.9953482736429659, "1000–3000 Hz_40ms_modulation_mae_db": 0.5924559501019596, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6240663563461044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21472422450957152, "3000–6000 Hz_10ms_envelope_correlation": 0.9618006889168983, "3000–6000 Hz_10ms_modulation_mae_db": 0.35763645654974185, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2152978663053855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13593118030063686, "3000–6000 Hz_40ms_envelope_correlation": 0.983853199084924, "3000–6000 Hz_40ms_modulation_mae_db": 0.6304672840567407, "mr_stft_db": 5.6127147533032895, "band_mae_db": 0.0777504996879523, "rms_gain_error_db": 0.01519278714780244, "output_peak": 0.6469382503477547}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3193269185022376, "全帯域_10ms_rms_envelope_nrmse": 0.07972015973609464, "全帯域_10ms_envelope_correlation": 0.9940030994901486, "全帯域_10ms_modulation_mae_db": 0.3511488713883857, "全帯域_40ms_log_rms_mae_db": 0.7558874495441295, "全帯域_40ms_rms_envelope_nrmse": 0.04226453511722156, "全帯域_40ms_envelope_correlation": 0.998247229326836, "全帯域_40ms_modulation_mae_db": 0.3967581580667707, "300–1000 Hz_10ms_log_rms_mae_db": 1.6788814680947737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10799515912272614, "300–1000 Hz_10ms_envelope_correlation": 0.9912187815324057, "300–1000 Hz_10ms_modulation_mae_db": 0.540094375222716, "300–1000 Hz_40ms_log_rms_mae_db": 0.9943592494337508, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05768131445399668, "300–1000 Hz_40ms_envelope_correlation": 0.9973925978378533, "300–1000 Hz_40ms_modulation_mae_db": 0.6165279291881628, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5904529817836524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1155568185132739, "1000–3000 Hz_10ms_envelope_correlation": 0.9900802515282869, "1000–3000 Hz_10ms_modulation_mae_db": 0.3875145451315323, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9482920942718266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0612877036526013, "1000–3000 Hz_40ms_envelope_correlation": 0.9973617726387957, "1000–3000 Hz_40ms_modulation_mae_db": 0.15420464917429813, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5689179196228504, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2095715287603039, "3000–6000 Hz_10ms_envelope_correlation": 0.963760628945501, "3000–6000 Hz_10ms_modulation_mae_db": 0.7427027517912336, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1115947753333704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13034196136131532, "3000–6000 Hz_40ms_envelope_correlation": 0.985624457651234, "3000–6000 Hz_40ms_modulation_mae_db": 0.8641425958462571, "mr_stft_db": 5.624115422208761, "band_mae_db": 0.16404389137746436, "rms_gain_error_db": -0.052596769737408504, "output_peak": 0.6082228552397279}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3130784814582876, "全帯域_10ms_rms_envelope_nrmse": 0.07591859552417601, "全帯域_10ms_envelope_correlation": 0.9945834115715707, "全帯域_10ms_modulation_mae_db": 0.23998816817734736, "全帯域_40ms_log_rms_mae_db": 0.6698814036394274, "全帯域_40ms_rms_envelope_nrmse": 0.040527420334028695, "全帯域_40ms_envelope_correlation": 0.9984161391931158, "全帯域_40ms_modulation_mae_db": 0.16864533879859414, "300–1000 Hz_10ms_log_rms_mae_db": 1.7587617839090153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11221092989983514, "300–1000 Hz_10ms_envelope_correlation": 0.9905549417734258, "300–1000 Hz_10ms_modulation_mae_db": 0.6628404455374131, "300–1000 Hz_40ms_log_rms_mae_db": 1.1352775232008683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059184581051715264, "300–1000 Hz_40ms_envelope_correlation": 0.9972501511190027, "300–1000 Hz_40ms_modulation_mae_db": 0.38772502533116454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6361755480633864, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12638160108315527, "1000–3000 Hz_10ms_envelope_correlation": 0.9881346007209161, "1000–3000 Hz_10ms_modulation_mae_db": 0.5190607125810885, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9165172701406639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06806439256712093, "1000–3000 Hz_40ms_envelope_correlation": 0.9967626266229909, "1000–3000 Hz_40ms_modulation_mae_db": 0.49198345403584404, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6114524139025193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20088826564007337, "3000–6000 Hz_10ms_envelope_correlation": 0.966686273682954, "3000–6000 Hz_10ms_modulation_mae_db": 0.6975073298502631, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1721253118087027, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1256196727524313, "3000–6000 Hz_40ms_envelope_correlation": 0.9864691764202622, "3000–6000 Hz_40ms_modulation_mae_db": 0.513136355658395, "mr_stft_db": 5.631958889618656, "band_mae_db": 0.1663118954281827, "rms_gain_error_db": -0.06722987126977066, "output_peak": 0.6090258958814124}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3018084211913052, "全帯域_10ms_rms_envelope_nrmse": 0.07184456514188968, "全帯域_10ms_envelope_correlation": 0.995146012303159, "全帯域_10ms_modulation_mae_db": 0.18909074675436635, "全帯域_40ms_log_rms_mae_db": 0.6749319664348241, "全帯域_40ms_rms_envelope_nrmse": 0.03568082703484363, "全帯域_40ms_envelope_correlation": 0.9987709288235445, "全帯域_40ms_modulation_mae_db": 0.17970611491978117, "300–1000 Hz_10ms_log_rms_mae_db": 1.738202346780004, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10616367620084335, "300–1000 Hz_10ms_envelope_correlation": 0.991508984739214, "300–1000 Hz_10ms_modulation_mae_db": 0.43208921501466935, "300–1000 Hz_40ms_log_rms_mae_db": 1.097730036343872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05718257224462234, "300–1000 Hz_40ms_envelope_correlation": 0.9974397151747331, "300–1000 Hz_40ms_modulation_mae_db": 0.3282514213855697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.579128650799151, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12293457472499264, "1000–3000 Hz_10ms_envelope_correlation": 0.9886972878225432, "1000–3000 Hz_10ms_modulation_mae_db": 0.45329880207204454, "1000–3000 Hz_40ms_log_rms_mae_db": 0.968575132926219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07492484077878996, "1000–3000 Hz_40ms_envelope_correlation": 0.9958071354136607, "1000–3000 Hz_40ms_modulation_mae_db": 0.5834095847181289, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6460416763010748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21237802247719684, "3000–6000 Hz_10ms_envelope_correlation": 0.9626718461049278, "3000–6000 Hz_10ms_modulation_mae_db": 0.8978164494177009, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1970161868982625, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1328265380139358, "3000–6000 Hz_40ms_envelope_correlation": 0.9847579291084311, "3000–6000 Hz_40ms_modulation_mae_db": 1.2295052114190943, "mr_stft_db": 5.623359892191083, "band_mae_db": 0.15272975703610955, "rms_gain_error_db": -0.058129667108233835, "output_peak": 0.5909363986984115}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2921571942439984, "全帯域_10ms_rms_envelope_nrmse": 0.07375232289141428, "全帯域_10ms_envelope_correlation": 0.9951550095219351, "全帯域_10ms_modulation_mae_db": 0.20955586641756127, "全帯域_40ms_log_rms_mae_db": 0.6855585140324937, "全帯域_40ms_rms_envelope_nrmse": 0.04159372091701381, "全帯域_40ms_envelope_correlation": 0.9986634022285821, "全帯域_40ms_modulation_mae_db": 0.2984413145600152, "300–1000 Hz_10ms_log_rms_mae_db": 1.667442694541612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11391036181951732, "300–1000 Hz_10ms_envelope_correlation": 0.9904286048084288, "300–1000 Hz_10ms_modulation_mae_db": 0.19544500987171118, "300–1000 Hz_40ms_log_rms_mae_db": 1.0096617582976202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06470119042748573, "300–1000 Hz_40ms_envelope_correlation": 0.9970665176362491, "300–1000 Hz_40ms_modulation_mae_db": 0.019102042795985593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5697489067635984, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1269654943259836, "1000–3000 Hz_10ms_envelope_correlation": 0.9884249430662602, "1000–3000 Hz_10ms_modulation_mae_db": 0.40798266921957005, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9372642898517963, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07123697453989625, "1000–3000 Hz_40ms_envelope_correlation": 0.9969307879935512, "1000–3000 Hz_40ms_modulation_mae_db": 0.49160480633171905, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6285184726958766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20537655292202026, "3000–6000 Hz_10ms_envelope_correlation": 0.9652987267091588, "3000–6000 Hz_10ms_modulation_mae_db": 0.5701862781460824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1690651058198427, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13238443283416465, "3000–6000 Hz_40ms_envelope_correlation": 0.9852819374346866, "3000–6000 Hz_40ms_modulation_mae_db": 0.8155512167023661, "mr_stft_db": 5.62323587183338, "band_mae_db": 0.1757458050513485, "rms_gain_error_db": -0.1763316175243601, "output_peak": 0.6091818204653618}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.699744827085052, "spectral_convergence": 0.2825737742510139, "spectral_convergence_status": "ok", "amplitude_floor": 1.4461531006911672e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4068304619881165, "spectral_convergence": 0.25978876599186745, "spectral_convergence_status": "ok", "amplitude_floor": 7.398993822544586e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.758968410576007, "spectral_convergence": 0.28194759743776665, "spectral_convergence_status": "ok", "amplitude_floor": 6.518159095373565e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005975557583734567, "output_power": 0.0006216200557103202, "delta_db": 0.17146582076013067, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005550914808547427, "output_power": 0.000571644966814421, "delta_db": 0.12761821851171748, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010492685783192576, "output_power": 0.0010473386860047457, "delta_db": -0.007995220028455496, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007612304608209731, "output_power": 0.0007160814459588618, "delta_db": -0.26553737284167345, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00020798712521433106, "output_power": 0.00019285614693462523, "delta_db": -0.3280296640771516, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.709473434650137e-06, "output_power": 8.921593320170155e-06, "delta_db": 0.10450523751967591, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1675252556231341, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.029298818565698596, "pitch": null, "output_rms_dbfs": -25.005242207809975, "output_peak": 0.6074618846159984, "reference_rms_dbfs": -24.975943389244275, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6218478998830586}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002040524003006053, "log_rms_mae_db": 1.2213290113154007, "rms_envelope_nrmse": 0.07342340761909585, "envelope_correlation": 0.9949114822204569, "rms_gain_error_db": -0.029299047889771303, "reference_mean_power": 0.0031824932955688964, "output_mean_power": 0.0031610953260302935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.3108768872499092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006941576008860869, "output_power": 0.0006885528494804886, "reference_active": true, "delta_db": -0.035208035617128514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.971136227750714e-05, "output_power": 6.344057588595258e-05, "reference_active": true, "delta_db": -0.40936453001709006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2294065042747786e-05, "output_power": 2.4576255081184012e-05, "reference_active": true, "delta_db": 0.423264421473856}], "active_band_count": 3, "active_band_mae_db": 0.2892789957026915}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019294226027584624, "log_rms_mae_db": 0.6896151267442013, "rms_envelope_nrmse": 0.043905838638546274, "envelope_correlation": 0.9980986151369761, "rms_gain_error_db": -0.029287327058329225, "reference_mean_power": 0.0031984676882902833, "output_mean_power": 0.0031769708865983623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.195649150719305e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000632939746621434, "output_power": 0.0006270917585104559, "reference_active": true, "delta_db": -0.040312757138371146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313010835721604e-05, "output_power": 3.100137146505599e-05, "reference_active": true, "delta_db": -0.28841949364332353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3632381122717001e-06, "output_power": 1.380214582737282e-06, "reference_active": true, "delta_db": 0.053748924232981485}], "active_band_count": 3, "active_band_mae_db": 0.12749372500489206}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015457760673902765, "log_rms_mae_db": 1.6835380555147352, "rms_envelope_nrmse": 0.09969069869653954, "envelope_correlation": 0.9925632010974296, "rms_gain_error_db": 0.01088040352291741, "reference_mean_power": 0.0011955559481795221, "output_mean_power": 0.0011985549361889498}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2205605502187216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134156506493551, "output_power": 0.0003954874446344329, "reference_active": true, "delta_db": -0.1925421296332228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.135405678849972e-05, "output_power": 3.9117519019132114e-05, "reference_active": true, "delta_db": -0.24146817408114943}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1249926033986765e-05, "output_power": 1.36048039721032e-05, "reference_active": true, "delta_db": 0.8254262147310899}], "active_band_count": 3, "active_band_mae_db": 0.41981217281515404}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014615313182534672, "log_rms_mae_db": 1.1213911021759004, "rms_envelope_nrmse": 0.06304859275362172, "envelope_correlation": 0.9968962699596899, "rms_gain_error_db": 0.010878496920471596, "reference_mean_power": 0.0012015673286161978, "output_mean_power": 0.0012045808670263234}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.16593985496473e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003798908177653278, "output_power": 0.0003617491735664687, "reference_active": true, "delta_db": -0.21251248845761753}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8446225039409073e-05, "output_power": 1.7555505195269123e-05, "reference_active": true, "delta_db": -0.21494170934174675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.60382048073901e-07, "output_power": 7.825175048799891e-07, "reference_active": true, "delta_db": 0.12462206456267909}], "active_band_count": 3, "active_band_mae_db": 0.18402542078734777}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011311095186173674, "log_rms_mae_db": 1.558839083132504, "rms_envelope_nrmse": 0.12470484970182143, "envelope_correlation": 0.9884466305151257, "rms_gain_error_db": -0.2324240551138879, "reference_mean_power": 0.0008539373095994977, "output_mean_power": 0.000809437989459965}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.101259078601872e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000229953654454443, "output_power": 0.00023669427345955754, "reference_active": true, "delta_db": 0.12547434979079788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8372953922112745e-05, "output_power": 4.487498832345795e-05, "reference_active": true, "delta_db": -1.1420732244953693}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3078079511035874e-05, "output_power": 7.483068197384358e-06, "reference_active": true, "delta_db": -2.4246427021326262}], "active_band_count": 3, "active_band_mae_db": 1.2307300921395978}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.65333606126804e-05, "log_rms_mae_db": 0.957049481628198, "rms_envelope_nrmse": 0.07634153114386863, "envelope_correlation": 0.9957606353935848, "rms_gain_error_db": -0.2324243268430033, "reference_mean_power": 0.0008582264011938477, "output_mean_power": 0.0008135035223427088}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.585136403392408e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214290189767478, "output_power": 0.00021832639333894187, "reference_active": true, "delta_db": 0.12477735206650514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7723754282185457e-05, "output_power": 2.1178910515957498e-05, "reference_active": true, "delta_db": -1.1694842560643635}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.120050210892624e-07, "output_power": 5.327996888881615e-07, "reference_active": true, "delta_db": -2.334332667954539}], "active_band_count": 3, "active_band_mae_db": 1.2095314253618026}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.9261609894090753e-05, "log_rms_mae_db": 1.5958467499314242, "rms_envelope_nrmse": 0.20393826084884292, "envelope_correlation": 0.9656702633406995, "rms_gain_error_db": -0.22356648310818103, "reference_mean_power": 3.945170478466558e-05, "output_mean_power": 3.7472194390984554e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2228241150412126e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1970652723962295e-05, "output_power": 1.0523311351495025e-05, "reference_active": true, "delta_db": -0.5596541184542485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3243801130743035e-06, "output_power": 2.738608358971884e-06, "reference_active": true, "delta_db": -0.8418074622661386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0798067724676823e-06, "output_power": 8.303953910161836e-07, "reference_active": true, "delta_db": -1.1406111698107764}], "active_band_count": 3, "active_band_mae_db": 0.8473575835103878}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5144247671385576e-05, "log_rms_mae_db": 1.1509979698244668, "rms_envelope_nrmse": 0.1287276398469526, "envelope_correlation": 0.9857349929576618, "rms_gain_error_db": -0.22356298170686625, "reference_mean_power": 3.9649231901027684e-05, "output_mean_power": 3.765984084099405e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.976126261669901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1283346332236661e-05, "output_power": 9.562630756036544e-06, "reference_active": true, "delta_db": -0.7186053203615155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7283331039583737e-06, "output_power": 1.344734812706259e-06, "reference_active": true, "delta_db": -1.089908002628877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.035315524247478e-08, "output_power": 6.079044014780615e-08, "reference_active": true, "delta_db": -0.634482918525105}], "active_band_count": 3, "active_band_mae_db": 0.8143320805051658}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8019038889876061, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "45ea888df8bb989b95bb601e41537c2063fa07a73c9d019d42c3f81b2c5d39d4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9886745930253701, "standard_deviation": 0.0013238772349749507, "minimum": 0.9859902071892351, "maximum": 0.9900802515282869, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5875227714510838, "standard_deviation": 0.02971799433273569, "minimum": 1.558839083132504, "maximum": 1.6361755480633864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5457369681643054, "standard_deviation": 0.2820286348526918, "minimum": 0.3740702222052022, "maximum": 1.2307300921395978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12348143626974441, "standard_deviation": 0.006681039776030718, "minimum": 0.1155568185132739, "maximum": 0.13596041057090494, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9964657474595906, "standard_deviation": 0.0007268454030681842, "minimum": 0.9953482736429659, "maximum": 0.9973617726387957, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9593329875972901, "standard_deviation": 0.030437943321777124, "minimum": 0.9165172701406639, "maximum": 1.0116726649173864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5165537887061173, "standard_deviation": 0.3215147405479041, "minimum": 0.15420464917429813, "maximum": 1.2095314253618026, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07032656597137223, "standard_deviation": 0.005343439851806978, "minimum": 0.0612877036526013, "maximum": 0.07634153114386863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9653274239801326, "standard_deviation": 0.0026855576145694525, "minimum": 0.9618006889168983, "maximum": 0.9703799681555882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6052234118908697, "standard_deviation": 0.028788818386854348, "minimum": 1.5628969450575168, "maximum": 1.6460416763010748, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.643276233097239, "standard_deviation": 0.2047058737899298, "minimum": 0.35068683088521313, "maximum": 0.8978164494177009, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2048036932365226, "standard_deviation": 0.007855572173670248, "minimum": 0.18969125269519113, "maximum": 0.21472422450957152, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9856479726845983, "standard_deviation": 0.001037966722828819, "minimum": 0.983853199084924, "maximum": 0.9871586324090038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1658320793771952, "standard_deviation": 0.03150072647021442, "minimum": 1.1115947753333704, "maximum": 1.2152978663053855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7617514294421421, "standard_deviation": 0.2380191045911846, "minimum": 0.4745203083457849, "maximum": 1.2295052114190943, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1291587158111418, "standard_deviation": 0.004612276064747132, "minimum": 0.1220017793717015, "maximum": 0.13593118030063686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9911639837911944, "standard_deviation": 0.0010155851203062466, "minimum": 0.9900761857683149, "maximum": 0.9926731796276194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6972647083514545, "standard_deviation": 0.03434468444597476, "minimum": 1.6642019791332874, "maximum": 1.7587617839090153, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47066168435345546, "standard_deviation": 0.29812696461339905, "minimum": 0.09706488247458132, "maximum": 1.0607027849281563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10850364923645547, "standard_deviation": 0.006555592529167214, "minimum": 0.09868640947765668, "maximum": 0.11558912448944189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973240209744698, "standard_deviation": 0.0004432404887312933, "minimum": 0.9967254393533509, "maximum": 0.9980775268377112, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.053887423259218, "standard_deviation": 0.0577992522897898, "minimum": 0.9943592494337508, "maximum": 1.1352775232008683, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3469348216932424, "standard_deviation": 0.27360040391551455, "minimum": 0.019102042795985593, "maximum": 0.8414005158707903, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05920989305198168, "standard_deviation": 0.005467823692511051, "minimum": 0.05131402610156062, "maximum": 0.06711325039722935, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1605517611885181, "standard_deviation": 0.03760912566350911, "minimum": 0.0777504996879523, "maximum": 0.2114432079230858, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.625127980565051, "standard_deviation": 0.008035362977065386, "minimum": 5.6127147533032895, "maximum": 5.640295833341878, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6151056192540649, "standard_deviation": 0.01670469952098188, "minimum": 0.5909363986984115, "maximum": 0.6469382503477547, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07008469054636343, "standard_deviation": 0.05626889049179405, "minimum": -0.1763316175243601, "maximum": 0.01519278714780244, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944728080868905, "standard_deviation": 0.0005392558463424773, "minimum": 0.9939469104816469, "maximum": 0.9951550095219351, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2897927605061368, "standard_deviation": 0.031145481436011903, "minimum": 1.2213290113154007, "maximum": 1.3193269185022376, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3713466115375941, "standard_deviation": 0.19298078964912155, "minimum": 0.18909074675436635, "maximum": 0.765430403329478, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07687150142288046, "standard_deviation": 0.0035360206660562154, "minimum": 0.07184456514188968, "maximum": 0.08036946403272403, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984292982402897, "standard_deviation": 0.00026730314057055567, "minimum": 0.9980986151369761, "maximum": 0.9987709288235445, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6910327805959227, "standard_deviation": 0.030945431684442793, "minimum": 0.6561438680248486, "maximum": 0.7558874495441295, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31681090660963124, "standard_deviation": 0.17293691605823452, "minimum": 0.12749372500489206, "maximum": 0.6344825136700405, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04082044052369496, "standard_deviation": 0.003093859820351744, "minimum": 0.03568082703484363, "maximum": 0.044853022264296036, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28240-0011", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.772688388824463, "predicted_mos": 1.772688388824463, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "958b81d021e67894d62aa816103769ee74fb791e8086f2c95e9041d2e45013a7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.928327299654484, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2691787165639539, "全帯域_10ms_rms_envelope_nrmse": 0.07387046525537495, "全帯域_10ms_envelope_correlation": 0.9948519620909806, "全帯域_10ms_modulation_mae_db": 0.3476991134153755, "全帯域_40ms_log_rms_mae_db": 0.7249214522510508, "全帯域_40ms_rms_envelope_nrmse": 0.04182728538981818, "全帯域_40ms_envelope_correlation": 0.9982865112256701, "全帯域_40ms_modulation_mae_db": 0.19267163743171092, "300–1000 Hz_10ms_log_rms_mae_db": 1.6210282352886412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1003126811867327, "300–1000 Hz_10ms_envelope_correlation": 0.9924639844683949, "300–1000 Hz_10ms_modulation_mae_db": 0.4813630383997604, "300–1000 Hz_40ms_log_rms_mae_db": 0.955355636261156, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058486374243163164, "300–1000 Hz_40ms_envelope_correlation": 0.9973284379637449, "300–1000 Hz_40ms_modulation_mae_db": 0.20187509320326436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5082497184822632, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12626493543465392, "1000–3000 Hz_10ms_envelope_correlation": 0.9882230224475793, "1000–3000 Hz_10ms_modulation_mae_db": 1.2752459629770596, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8378127612574093, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07790953372362676, "1000–3000 Hz_40ms_envelope_correlation": 0.9956673044140786, "1000–3000 Hz_40ms_modulation_mae_db": 1.2802898978054194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4861799154968443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20205383930668458, "3000–6000 Hz_10ms_envelope_correlation": 0.9667313644537909, "3000–6000 Hz_10ms_modulation_mae_db": 1.0546930738728006, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9827679104633825, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1298579759761925, "3000–6000 Hz_40ms_envelope_correlation": 0.9862749784083642, "3000–6000 Hz_40ms_modulation_mae_db": 1.0038727459241177, "mr_stft_db": 5.566974465029991, "band_mae_db": 0.19253643020955094, "rms_gain_error_db": -0.03613389056091363, "output_peak": 0.6078446821364158}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.301655008062343, "全帯域_10ms_rms_envelope_nrmse": 0.07743761995766145, "全帯域_10ms_envelope_correlation": 0.9943900525682843, "全帯域_10ms_modulation_mae_db": 0.6329675543170253, "全帯域_40ms_log_rms_mae_db": 0.6738279005783496, "全帯域_40ms_rms_envelope_nrmse": 0.03851297110508872, "全帯域_40ms_envelope_correlation": 0.9986381128592237, "全帯域_40ms_modulation_mae_db": 0.5115416341437937, "300–1000 Hz_10ms_log_rms_mae_db": 1.6065183883883487, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11558462989538656, "300–1000 Hz_10ms_envelope_correlation": 0.9899685673740373, "300–1000 Hz_10ms_modulation_mae_db": 0.9997849498643747, "300–1000 Hz_40ms_log_rms_mae_db": 0.834519377537504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05695061140607234, "300–1000 Hz_40ms_envelope_correlation": 0.9975755290031393, "300–1000 Hz_40ms_modulation_mae_db": 0.749583945154369, "1000–3000 Hz_10ms_log_rms_mae_db": 1.539252981468402, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11671642061231223, "1000–3000 Hz_10ms_envelope_correlation": 0.9899477739161333, "1000–3000 Hz_10ms_modulation_mae_db": 0.3917416141087695, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8539187304698934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06571226932176308, "1000–3000 Hz_40ms_envelope_correlation": 0.9969995300813791, "1000–3000 Hz_40ms_modulation_mae_db": 0.32749633084730717, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5073822652807283, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18861653990597638, "3000–6000 Hz_10ms_envelope_correlation": 0.9708036958314546, "3000–6000 Hz_10ms_modulation_mae_db": 0.8107345704857339, "3000–6000 Hz_40ms_log_rms_mae_db": 1.003243182559379, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12078594166126615, "3000–6000 Hz_40ms_envelope_correlation": 0.9876593015908198, "3000–6000 Hz_40ms_modulation_mae_db": 0.8537976599543257, "mr_stft_db": 5.583803954847208, "band_mae_db": 0.19039199177198451, "rms_gain_error_db": -0.08865011908755552, "output_peak": 0.6232794234604704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3264810280526662, "全帯域_10ms_rms_envelope_nrmse": 0.08197637721303613, "全帯域_10ms_envelope_correlation": 0.9937369189745388, "全帯域_10ms_modulation_mae_db": 0.4772698909187257, "全帯域_40ms_log_rms_mae_db": 0.7369611231724216, "全帯域_40ms_rms_envelope_nrmse": 0.04615657570637996, "全帯域_40ms_envelope_correlation": 0.9980162451975585, "全帯域_40ms_modulation_mae_db": 0.3401623874833762, "300–1000 Hz_10ms_log_rms_mae_db": 1.61084926671281, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1218423384909827, "300–1000 Hz_10ms_envelope_correlation": 0.9889150580237313, "300–1000 Hz_10ms_modulation_mae_db": 0.46540341152660286, "300–1000 Hz_40ms_log_rms_mae_db": 0.8713628392525495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07106847856121962, "300–1000 Hz_40ms_envelope_correlation": 0.9962567846773593, "300–1000 Hz_40ms_modulation_mae_db": 0.3917699261117236, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5440207815797231, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1145692811935412, "1000–3000 Hz_10ms_envelope_correlation": 0.9903234521400276, "1000–3000 Hz_10ms_modulation_mae_db": 0.36646458824312944, "1000–3000 Hz_40ms_log_rms_mae_db": 0.85233240377974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06740915592360038, "1000–3000 Hz_40ms_envelope_correlation": 0.9968073070426896, "1000–3000 Hz_40ms_modulation_mae_db": 0.2688944088455775, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4744239914890522, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1971045398786346, "3000–6000 Hz_10ms_envelope_correlation": 0.9679885344872077, "3000–6000 Hz_10ms_modulation_mae_db": 0.4450450910897133, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9790555882217653, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12337963821211999, "3000–6000 Hz_40ms_envelope_correlation": 0.9869918641184023, "3000–6000 Hz_40ms_modulation_mae_db": 0.5570448373783111, "mr_stft_db": 5.574963227812084, "band_mae_db": 0.18324496980052943, "rms_gain_error_db": -0.11061303073303055, "output_peak": 0.620963511956263}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3192593958234338, "全帯域_10ms_rms_envelope_nrmse": 0.07949049358229049, "全帯域_10ms_envelope_correlation": 0.994069756399623, "全帯域_10ms_modulation_mae_db": 0.5181420920168969, "全帯域_40ms_log_rms_mae_db": 0.7063688081744143, "全帯域_40ms_rms_envelope_nrmse": 0.04082657998369674, "全帯域_40ms_envelope_correlation": 0.9983537699670705, "全帯域_40ms_modulation_mae_db": 0.5268349968917211, "300–1000 Hz_10ms_log_rms_mae_db": 1.641831302398889, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09954422335618393, "300–1000 Hz_10ms_envelope_correlation": 0.9925533321447875, "300–1000 Hz_10ms_modulation_mae_db": 0.06448247850759192, "300–1000 Hz_40ms_log_rms_mae_db": 0.9032043036476161, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053472188652770224, "300–1000 Hz_40ms_envelope_correlation": 0.9977445416477128, "300–1000 Hz_40ms_modulation_mae_db": 0.06100887043137746, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5323469894056225, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13742185866579254, "1000–3000 Hz_10ms_envelope_correlation": 0.9856822228717935, "1000–3000 Hz_10ms_modulation_mae_db": 0.46474167593740007, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8578398397428829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07862202663160395, "1000–3000 Hz_40ms_envelope_correlation": 0.9950520925527486, "1000–3000 Hz_40ms_modulation_mae_db": 0.5429114816004549, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5052256292239665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2053176075949675, "3000–6000 Hz_10ms_envelope_correlation": 0.9653528662537403, "3000–6000 Hz_10ms_modulation_mae_db": 0.5750276394252569, "3000–6000 Hz_40ms_log_rms_mae_db": 1.013159046251542, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1280766011071306, "3000–6000 Hz_40ms_envelope_correlation": 0.98633245374523, "3000–6000 Hz_40ms_modulation_mae_db": 0.8138324507578071, "mr_stft_db": 5.562082230845615, "band_mae_db": 0.13051777691741237, "rms_gain_error_db": 0.005881082339285574, "output_peak": 0.6450318935985452}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3299099995858923, "全帯域_10ms_rms_envelope_nrmse": 0.08104192621550144, "全帯域_10ms_envelope_correlation": 0.9938010325602709, "全帯域_10ms_modulation_mae_db": 0.3596762720729246, "全帯域_40ms_log_rms_mae_db": 0.7588206010604202, "全帯域_40ms_rms_envelope_nrmse": 0.04151235332360211, "全帯域_40ms_envelope_correlation": 0.9983080712211466, "全帯域_40ms_modulation_mae_db": 0.4181308077050004, "300–1000 Hz_10ms_log_rms_mae_db": 1.6544085026042796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1117344140741231, "300–1000 Hz_10ms_envelope_correlation": 0.9906035895520641, "300–1000 Hz_10ms_modulation_mae_db": 0.5767613136557236, "300–1000 Hz_40ms_log_rms_mae_db": 0.8998087552142936, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058305160740113965, "300–1000 Hz_40ms_envelope_correlation": 0.99732931801089, "300–1000 Hz_40ms_modulation_mae_db": 0.6299975932087235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.540957873570952, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11928861877835885, "1000–3000 Hz_10ms_envelope_correlation": 0.9894467339407081, "1000–3000 Hz_10ms_modulation_mae_db": 0.24095859609557746, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8489970054748384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06537484384004699, "1000–3000 Hz_40ms_envelope_correlation": 0.9969958563199757, "1000–3000 Hz_40ms_modulation_mae_db": 0.13431725330491187, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4514226668797634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19998732013051726, "3000–6000 Hz_10ms_envelope_correlation": 0.9672616159229237, "3000–6000 Hz_10ms_modulation_mae_db": 0.9084124159028355, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9443817227056305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12408152072667512, "3000–6000 Hz_40ms_envelope_correlation": 0.9873784349297241, "3000–6000 Hz_40ms_modulation_mae_db": 0.9462283977038637, "mr_stft_db": 5.56832237352468, "band_mae_db": 0.13403037463201012, "rms_gain_error_db": -0.050500348512378224, "output_peak": 0.608059270098663}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3248771671359456, "全帯域_10ms_rms_envelope_nrmse": 0.07563471598164266, "全帯域_10ms_envelope_correlation": 0.9946300574250279, "全帯域_10ms_modulation_mae_db": 0.20044477032773791, "全帯域_40ms_log_rms_mae_db": 0.6606140928429236, "全帯域_40ms_rms_envelope_nrmse": 0.041388187269399856, "全帯域_40ms_envelope_correlation": 0.9983521586786197, "全帯域_40ms_modulation_mae_db": 0.11633888664980402, "300–1000 Hz_10ms_log_rms_mae_db": 1.6634588435670705, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11210650150727836, "300–1000 Hz_10ms_envelope_correlation": 0.9906042956608672, "300–1000 Hz_10ms_modulation_mae_db": 0.5765741555488474, "300–1000 Hz_40ms_log_rms_mae_db": 0.9499849484035007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059121877018588155, "300–1000 Hz_40ms_envelope_correlation": 0.9972698764279351, "300–1000 Hz_40ms_modulation_mae_db": 0.3184194206453107, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5854161000630698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12982343791005013, "1000–3000 Hz_10ms_envelope_correlation": 0.9874980339631723, "1000–3000 Hz_10ms_modulation_mae_db": 0.668672519447771, "1000–3000 Hz_40ms_log_rms_mae_db": 0.839034662707779, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07195410637608117, "1000–3000 Hz_40ms_envelope_correlation": 0.9963756073840135, "1000–3000 Hz_40ms_modulation_mae_db": 0.6068774450489541, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4919883605925937, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1943720398287926, "3000–6000 Hz_10ms_envelope_correlation": 0.969289700706387, "3000–6000 Hz_10ms_modulation_mae_db": 0.7481980085667876, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9949735965141869, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1205828776329883, "3000–6000 Hz_40ms_envelope_correlation": 0.9883477427356733, "3000–6000 Hz_40ms_modulation_mae_db": 0.5244396312537606, "mr_stft_db": 5.580698456710522, "band_mae_db": 0.1379097548935463, "rms_gain_error_db": -0.0715315682360338, "output_peak": 0.6074591986941162}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3097870165330854, "全帯域_10ms_rms_envelope_nrmse": 0.07391841689322025, "全帯域_10ms_envelope_correlation": 0.9948628251673467, "全帯域_10ms_modulation_mae_db": 0.21025892376086622, "全帯域_40ms_log_rms_mae_db": 0.6751341044484448, "全帯域_40ms_rms_envelope_nrmse": 0.03722823287239851, "全帯域_40ms_envelope_correlation": 0.9986602794102, "全帯域_40ms_modulation_mae_db": 0.20991281667997785, "300–1000 Hz_10ms_log_rms_mae_db": 1.7012093051470907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11252447295755219, "300–1000 Hz_10ms_envelope_correlation": 0.9904544575813957, "300–1000 Hz_10ms_modulation_mae_db": 0.2877405072614949, "300–1000 Hz_40ms_log_rms_mae_db": 0.9711520074167103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060751006052216554, "300–1000 Hz_40ms_envelope_correlation": 0.9970996087900822, "300–1000 Hz_40ms_modulation_mae_db": 0.19785508928408543, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5273454041395556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12371598302777398, "1000–3000 Hz_10ms_envelope_correlation": 0.9885530616686484, "1000–3000 Hz_10ms_modulation_mae_db": 0.4251335643855914, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8679295228254187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07496128137607456, "1000–3000 Hz_40ms_envelope_correlation": 0.9958110875917106, "1000–3000 Hz_40ms_modulation_mae_db": 0.5876476726685843, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5127360780438714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20015955648592557, "3000–6000 Hz_10ms_envelope_correlation": 0.9671760967791087, "3000–6000 Hz_10ms_modulation_mae_db": 0.9467373162449544, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9896930232272351, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12359348779169145, "3000–6000 Hz_40ms_envelope_correlation": 0.9874016856719252, "3000–6000 Hz_40ms_modulation_mae_db": 1.2421704445451056, "mr_stft_db": 5.575769647019403, "band_mae_db": 0.17553213806780518, "rms_gain_error_db": -0.058063038998776496, "output_peak": 0.5749781198971805}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3328349918882572, "全帯域_10ms_rms_envelope_nrmse": 0.07519897161483018, "全帯域_10ms_envelope_correlation": 0.9949781160156388, "全帯域_10ms_modulation_mae_db": 0.235191279798767, "全帯域_40ms_log_rms_mae_db": 0.764272787240943, "全帯域_40ms_rms_envelope_nrmse": 0.04342592599317783, "全帯域_40ms_envelope_correlation": 0.9985397507784611, "全帯域_40ms_modulation_mae_db": 0.3492118851232368, "300–1000 Hz_10ms_log_rms_mae_db": 1.6313574805432758, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11484972966604577, "300–1000 Hz_10ms_envelope_correlation": 0.9902664711355763, "300–1000 Hz_10ms_modulation_mae_db": 0.22933717622202432, "300–1000 Hz_40ms_log_rms_mae_db": 0.8739490973785243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06655354756993241, "300–1000 Hz_40ms_envelope_correlation": 0.9968655181013184, "300–1000 Hz_40ms_modulation_mae_db": 0.14229710423665456, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5235885214569749, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12905869629072295, "1000–3000 Hz_10ms_envelope_correlation": 0.988064840727287, "1000–3000 Hz_10ms_modulation_mae_db": 0.513877828910254, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8171388072742183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0738336399273952, "1000–3000 Hz_40ms_envelope_correlation": 0.9966937783554688, "1000–3000 Hz_40ms_modulation_mae_db": 0.6108284534388924, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5507341441670413, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20494286955814442, "3000–6000 Hz_10ms_envelope_correlation": 0.96558761148923, "3000–6000 Hz_10ms_modulation_mae_db": 0.49280194099924085, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0335678319384531, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1332702867361309, "3000–6000 Hz_40ms_envelope_correlation": 0.9852794807704919, "3000–6000 Hz_40ms_modulation_mae_db": 0.7175575082579506, "mr_stft_db": 5.574877025603045, "band_mae_db": 0.22091287177970406, "rms_gain_error_db": -0.1845663503844738, "output_peak": 0.6102885333676957}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.652913790853278, "spectral_convergence": 0.2824760639578791, "spectral_convergence_status": "ok", "amplitude_floor": 1.4461531006911672e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.356345981827601, "spectral_convergence": 0.25872121439666956, "spectral_convergence_status": "ok", "amplitude_floor": 7.398993822544586e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.691663622409093, "spectral_convergence": 0.2800996920865963, "spectral_convergence_status": "ok", "amplitude_floor": 6.518159095373565e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005975557583734567, "output_power": 0.0006214208780270659, "delta_db": 0.1700740440752579, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005550914808547427, "output_power": 0.0005730070391721262, "delta_db": 0.1379539499831438, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010492685783192576, "output_power": 0.0010473877597812855, "delta_db": -0.007791733120614203, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007612304608209731, "output_power": 0.0007141617489706559, "delta_db": -0.27719573014088666, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00020798712521433106, "output_power": 0.00018910238074182318, "delta_db": -0.4133945565169871, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.709473434650137e-06, "output_power": 8.416103002979686e-06, "delta_db": -0.14880856742041598, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19253643020955094, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03613389056091363, "pitch": null, "output_rms_dbfs": -25.01207727980519, "output_peak": 0.6078446821364158, "reference_rms_dbfs": -24.975943389244275, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.566974465029991}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002040524003006053, "log_rms_mae_db": 1.2691787165639539, "rms_envelope_nrmse": 0.07387046525537495, "envelope_correlation": 0.9948519620909806, "rms_gain_error_db": -0.03613454158422617, "reference_mean_power": 0.0031824932955688964, "output_mean_power": 0.0031561238947739267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.3108768872499092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006941576008860869, "output_power": 0.0006865832737499162, "reference_active": true, "delta_db": -0.047648643378640505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.971136227750714e-05, "output_power": 6.197236930779357e-05, "reference_active": true, "delta_db": -0.5110546939079766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2294065042747786e-05, "output_power": 2.492462679205598e-05, "reference_active": true, "delta_db": 0.48439400295950946}], "active_band_count": 3, "active_band_mae_db": 0.3476991134153755}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019294226027584624, "log_rms_mae_db": 0.7249214522510508, "rms_envelope_nrmse": 0.04182728538981818, "envelope_correlation": 0.9982865112256701, "rms_gain_error_db": -0.03612176462313427, "reference_mean_power": 0.0031984676882902833, "output_mean_power": 0.003171975259335077}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.195649150719305e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000632939746621434, "output_power": 0.0006247243210096975, "reference_active": true, "delta_db": -0.0567395515444824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313010835721604e-05, "output_power": 3.042135525856535e-05, "reference_active": true, "delta_db": -0.37044298528277403}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3632381122717001e-06, "output_power": 1.4114157414215602e-06, "reference_active": true, "delta_db": 0.1508323754678764}], "active_band_count": 3, "active_band_mae_db": 0.19267163743171092}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015457760673902765, "log_rms_mae_db": 1.6210282352886412, "rms_envelope_nrmse": 0.1003126811867327, "envelope_correlation": 0.9924639844683949, "rms_gain_error_db": 0.011767692638827112, "reference_mean_power": 0.0011955559481795221, "output_mean_power": 0.0011987998330129856}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2205605502187216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134156506493551, "output_power": 0.00039678965058831584, "reference_active": true, "delta_db": -0.17826577675735197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.135405678849972e-05, "output_power": 3.8288149618304945e-05, "reference_active": true, "delta_db": -0.3345374143818832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1249926033986765e-05, "output_power": 1.3940497069467332e-05, "reference_active": true, "delta_db": 0.931285924060046}], "active_band_count": 3, "active_band_mae_db": 0.4813630383997604}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014615313182534672, "log_rms_mae_db": 0.955355636261156, "rms_envelope_nrmse": 0.058486374243163164, "envelope_correlation": 0.9973284379637449, "rms_gain_error_db": 0.01176662055250048, "reference_mean_power": 0.0012015673286161978, "output_mean_power": 0.0012048272266220446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.16593985496473e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003798908177653278, "output_power": 0.00036278691644639737, "reference_active": true, "delta_db": -0.20007180204362926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8446225039409073e-05, "output_power": 1.693716887304166e-05, "reference_active": true, "delta_db": -0.3706668491167109}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.60382048073901e-07, "output_power": 7.665147526264688e-07, "reference_active": true, "delta_db": 0.03488662844945287}], "active_band_count": 3, "active_band_mae_db": 0.20187509320326436}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011311095186173674, "log_rms_mae_db": 1.5082497184822632, "rms_envelope_nrmse": 0.12626493543465392, "envelope_correlation": 0.9882230224475793, "rms_gain_error_db": -0.2557299425589885, "reference_mean_power": 0.0008539373095994977, "output_mean_power": 0.0008051058727977398}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.101259078601872e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000229953654454443, "output_power": 0.00023401846236236666, "reference_active": true, "delta_db": 0.07609805620756133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8372953922112745e-05, "output_power": 4.425718937799593e-05, "reference_active": true, "delta_db": -1.20227841114908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3078079511035874e-05, "output_power": 7.2745784422564995e-06, "reference_active": true, "delta_db": -2.547361421574537}], "active_band_count": 3, "active_band_mae_db": 1.2752459629770596}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.65333606126804e-05, "log_rms_mae_db": 0.8378127612574093, "rms_envelope_nrmse": 0.07790953372362676, "envelope_correlation": 0.9956673044140786, "rms_gain_error_db": -0.2557275607021431, "reference_mean_power": 0.0008582264011938477, "output_mean_power": 0.0008091501413249608}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.585136403392408e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214290189767478, "output_power": 0.00021536110823915927, "reference_active": true, "delta_db": 0.06538772150574229}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7723754282185457e-05, "output_power": 2.107388733690779e-05, "reference_active": true, "delta_db": -1.1910738706612993}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.120050210892624e-07, "output_power": 5.029866046469523e-07, "reference_active": true, "delta_db": -2.5844081012492164}], "active_band_count": 3, "active_band_mae_db": 1.2802898978054194}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.9261609894090753e-05, "log_rms_mae_db": 1.4861799154968443, "rms_envelope_nrmse": 0.20205383930668458, "envelope_correlation": 0.9667313644537909, "rms_gain_error_db": -0.39673796914285464, "reference_mean_power": 3.945170478466558e-05, "output_mean_power": 3.600741778475593e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2228241150412126e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1970652723962295e-05, "output_power": 1.0434729821866097e-05, "reference_active": true, "delta_db": -0.5963662312731435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3243801130743035e-06, "output_power": 2.599293574645289e-06, "reference_active": true, "delta_db": -1.0685534240079184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0798067724676823e-06, "output_power": 7.645925998702419e-07, "reference_active": true, "delta_db": -1.4991595663373398}], "active_band_count": 3, "active_band_mae_db": 1.0546930738728006}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5144247671385576e-05, "log_rms_mae_db": 0.9827679104633825, "rms_envelope_nrmse": 0.1298579759761925, "envelope_correlation": 0.9862749784083642, "rms_gain_error_db": -0.39671789439533506, "reference_mean_power": 3.9649231901027684e-05, "output_mean_power": 3.6187867290695225e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.976126261669901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1283346332236661e-05, "output_power": 9.45637945125912e-06, "reference_active": true, "delta_db": -0.7671302803040638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7283331039583737e-06, "output_power": 1.2618797579022734e-06, "reference_active": true, "delta_db": -1.3660947457534067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.035315524247478e-08, "output_power": 5.7470405117003815e-08, "reference_active": true, "delta_db": -0.8783932117148828}], "active_band_count": 3, "active_band_mae_db": 1.0038727459241177}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8029533427232153, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "19583adf1374f0862955ae72c9440ec264fe696d7cb1eb98c5873a66183b28a0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9884673927094187, "standard_deviation": 0.0014890805346903658, "minimum": 0.9856822228717935, "maximum": 0.9903234521400276, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5376472962708203, "standard_deviation": 0.022463007816628095, "minimum": 1.5082497184822632, "maximum": 1.5854161000630698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.543354543763194, "standard_deviation": 0.32029329197470763, "minimum": 0.24095859609557746, "maximum": 1.2752459629770596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12460740398915073, "standard_deviation": 0.007616657567770653, "minimum": 0.1145692811935412, "maximum": 0.13742185866579254, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963003204677581, "standard_deviation": 0.0007159716555495443, "minimum": 0.9950520925527486, "maximum": 0.9969995300813791, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8468754666915225, "standard_deviation": 0.015470050143708452, "minimum": 0.8171388072742183, "maximum": 0.8679295228254187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5449078679450128, "standard_deviation": 0.3469950632180549, "minimum": 0.13431725330491187, "maximum": 1.2802898978054194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07197210714002401, "standard_deviation": 0.00528346789528041, "minimum": 0.06537484384004699, "maximum": 0.07862202663160395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9675239357404803, "standard_deviation": 0.0018272333247321211, "minimum": 0.9653528662537403, "maximum": 0.9708036958314546, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4975116313967325, "standard_deviation": 0.02936835247437159, "minimum": 1.4514226668797634, "maximum": 1.5507341441670413, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7477062570734154, "standard_deviation": 0.2236232305020271, "minimum": 0.4450450910897133, "maximum": 1.0546930738728006, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19906928908620536, "standard_deviation": 0.00560183377152711, "minimum": 0.18861653990597638, "maximum": 0.2053176075949675, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9869582427463289, "standard_deviation": 0.0009616378157439659, "minimum": 0.9852794807704919, "maximum": 0.9883477427356733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9926052377351968, "standard_deviation": 0.0262774280289388, "minimum": 0.9443817227056305, "maximum": 1.0335678319384531, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8323679594719053, "standard_deviation": 0.2375238579521035, "minimum": 0.5244396312537606, "maximum": 1.2421704445451056, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12545354123052438, "standard_deviation": 0.004512429837236509, "minimum": 0.1205828776329883, "maximum": 0.1332702867361309, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907287194926068, "standard_deviation": 0.0012263527204792355, "minimum": 0.9889150580237313, "maximum": 0.9925533321447875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6413326655813008, "standard_deviation": 0.03139392930300004, "minimum": 1.6065183883883487, "maximum": 1.7012093051470907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4601808788733025, "standard_deviation": 0.28243548584622136, "minimum": 0.06448247850759192, "maximum": 0.9997849498643747, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11106237389178567, "standard_deviation": 0.007586696618413655, "minimum": 0.09954422335618393, "maximum": 0.1218423384909827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971837018277727, "standard_deviation": 0.00046061084427403753, "minimum": 0.9962567846773593, "maximum": 0.9977445416477128, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9074171206389818, "standard_deviation": 0.04778223343042742, "minimum": 0.834519377537504, "maximum": 0.9711520074167103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3366008802844386, "standard_deviation": 0.24234484804641251, "minimum": 0.06100887043137746, "maximum": 0.749583945154369, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06058865553050956, "standard_deviation": 0.005620089293064248, "minimum": 0.053472188652770224, "maximum": 0.07106847856121962, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17063453850906785, "standard_deviation": 0.032952565283160186, "minimum": 0.13051777691741237, "maximum": 0.22091287177970406, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.573436422674068, "standard_deviation": 0.007239784286993047, "minimum": 5.562082230845615, "maximum": 5.583803954847208, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6122380791511688, "standard_deviation": 0.01973099845790802, "minimum": 0.5749781198971805, "maximum": 0.6450318935985452, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07427215802173456, "standard_deviation": 0.056576886178122274, "minimum": -0.1845663503844738, "maximum": 0.005881082339285574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944150901502138, "standard_deviation": 0.0004944546692589524, "minimum": 0.9937369189745388, "maximum": 0.9949781160156388, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3142479154556972, "standard_deviation": 0.021004213470969518, "minimum": 1.2691787165639539, "maximum": 1.3328349918882572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3727062370785399, "standard_deviation": 0.15838007679794053, "minimum": 0.20044477032773791, "maximum": 0.6329675543170253, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07732112333919469, "standard_deviation": 0.003186320122777717, "minimum": 0.07387046525537495, "maximum": 0.08197637721303613, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983943624172438, "standard_deviation": 0.0002127182862449438, "minimum": 0.9980162451975585, "maximum": 0.9986602794102, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.712615108721121, "standard_deviation": 0.04000005875495197, "minimum": 0.6606140928429236, "maximum": 0.764272787240943, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3331006315135776, "standard_deviation": 0.15061538606675084, "minimum": 0.11633888664980402, "maximum": 0.5268349968917211, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04135976395544524, "standard_deviation": 0.0027510818651804527, "minimum": 0.03722823287239851, "maximum": 0.04615657570637996, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28240-0011", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.6886096000671387, "predicted_mos": 2.6886096000671387, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3f59bc7659ab69011cee9f446ff48bc993d70f210f8a885f6cc353f33667ee76", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9321377999149263, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2023771211933298, "全帯域_10ms_rms_envelope_nrmse": 0.07558418566007809, "全帯域_10ms_envelope_correlation": 0.9947103633557373, "全帯域_10ms_modulation_mae_db": 0.49363988362155037, "全帯域_40ms_log_rms_mae_db": 0.6587604048328334, "全帯域_40ms_rms_envelope_nrmse": 0.03865651753608751, "全帯域_40ms_envelope_correlation": 0.9986812555710721, "全帯域_40ms_modulation_mae_db": 0.5018895899107614, "300–1000 Hz_10ms_log_rms_mae_db": 1.9828276250271148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11639252454703658, "300–1000 Hz_10ms_envelope_correlation": 0.9898212756276336, "300–1000 Hz_10ms_modulation_mae_db": 0.3791814328120207, "300–1000 Hz_40ms_log_rms_mae_db": 1.3483451100642658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06342619351335063, "300–1000 Hz_40ms_envelope_correlation": 0.9969211042198463, "300–1000 Hz_40ms_modulation_mae_db": 0.35852343284436333, "1000–3000 Hz_10ms_log_rms_mae_db": 1.962795576042538, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12520123276101838, "1000–3000 Hz_10ms_envelope_correlation": 0.9885661312816579, "1000–3000 Hz_10ms_modulation_mae_db": 0.6681767611087767, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1783745000176027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07108150370551922, "1000–3000 Hz_40ms_envelope_correlation": 0.996680140430797, "1000–3000 Hz_40ms_modulation_mae_db": 0.58972255795333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9835951111770418, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22351216983342323, "3000–6000 Hz_10ms_envelope_correlation": 0.9589818498380783, "3000–6000 Hz_10ms_modulation_mae_db": 0.7939175000784519, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4887437982320988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14278444956811917, "3000–6000 Hz_40ms_envelope_correlation": 0.9832961605798443, "3000–6000 Hz_40ms_modulation_mae_db": 0.9996203245855116, "mr_stft_db": 5.8593312993097975, "band_mae_db": 0.1752907396448471, "rms_gain_error_db": -0.1183055411849886, "output_peak": 0.5984231055455785}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2316952332223077, "全帯域_10ms_rms_envelope_nrmse": 0.07383522545200204, "全帯域_10ms_envelope_correlation": 0.9949413951172005, "全帯域_10ms_modulation_mae_db": 0.1487694741099545, "全帯域_40ms_log_rms_mae_db": 0.6351796560426447, "全帯域_40ms_rms_envelope_nrmse": 0.038586850989630994, "全帯域_40ms_envelope_correlation": 0.9986584662755561, "全帯域_40ms_modulation_mae_db": 0.1309555992888173, "300–1000 Hz_10ms_log_rms_mae_db": 1.8263849383611979, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10372492846808516, "300–1000 Hz_10ms_envelope_correlation": 0.9919478065274326, "300–1000 Hz_10ms_modulation_mae_db": 0.7050704783366636, "300–1000 Hz_40ms_log_rms_mae_db": 1.165575304993692, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05433482125653275, "300–1000 Hz_40ms_envelope_correlation": 0.9977875058073843, "300–1000 Hz_40ms_modulation_mae_db": 0.4155705055840946, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9847383974928947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1311277840311757, "1000–3000 Hz_10ms_envelope_correlation": 0.9874187022447718, "1000–3000 Hz_10ms_modulation_mae_db": 0.42795530411487537, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2475917317233374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07335422959845313, "1000–3000 Hz_40ms_envelope_correlation": 0.9964532882509207, "1000–3000 Hz_40ms_modulation_mae_db": 0.3311072982361784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.98964194419907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2410181698524039, "3000–6000 Hz_10ms_envelope_correlation": 0.9520922595530654, "3000–6000 Hz_10ms_modulation_mae_db": 0.5216448389029934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.504913901844929, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15766426171424525, "3000–6000 Hz_40ms_envelope_correlation": 0.9781017334897372, "3000–6000 Hz_40ms_modulation_mae_db": 0.6261774800880161, "mr_stft_db": 5.873241623158837, "band_mae_db": 0.2482204164236699, "rms_gain_error_db": -0.1098468298017693, "output_peak": 0.5992563140925878}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2823590422867848, "全帯域_10ms_rms_envelope_nrmse": 0.07376621739917662, "全帯域_10ms_envelope_correlation": 0.9948783665549424, "全帯域_10ms_modulation_mae_db": 0.32059564708312854, "全帯域_40ms_log_rms_mae_db": 0.6977154813101075, "全帯域_40ms_rms_envelope_nrmse": 0.03868926838631979, "全帯域_40ms_envelope_correlation": 0.9985469179255667, "全帯域_40ms_modulation_mae_db": 0.25408711246634974, "300–1000 Hz_10ms_log_rms_mae_db": 1.9414899001621715, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10350328011377993, "300–1000 Hz_10ms_envelope_correlation": 0.9919524953259695, "300–1000 Hz_10ms_modulation_mae_db": 0.4182173675259134, "300–1000 Hz_40ms_log_rms_mae_db": 1.269956349478422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05232076917384203, "300–1000 Hz_40ms_envelope_correlation": 0.997843218638021, "300–1000 Hz_40ms_modulation_mae_db": 0.31678349206300976, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8851190972602976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13021097253606861, "1000–3000 Hz_10ms_envelope_correlation": 0.9873819222673054, "1000–3000 Hz_10ms_modulation_mae_db": 0.3484933840635764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1701311107460002, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06984394921434471, "1000–3000 Hz_40ms_envelope_correlation": 0.9965847243284982, "1000–3000 Hz_40ms_modulation_mae_db": 0.30926100672711976, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0235553279044045, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22744915765891113, "3000–6000 Hz_10ms_envelope_correlation": 0.957242037193361, "3000–6000 Hz_10ms_modulation_mae_db": 0.525243816780339, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5418257077521518, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14156639516875894, "3000–6000 Hz_40ms_envelope_correlation": 0.9825695086109439, "3000–6000 Hz_40ms_modulation_mae_db": 0.763827956651736, "mr_stft_db": 5.863931540199313, "band_mae_db": 0.23335036272652207, "rms_gain_error_db": -0.05336685336227699, "output_peak": 0.6364411444291814}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.298001458324749, "全帯域_10ms_rms_envelope_nrmse": 0.07278844587671124, "全帯域_10ms_envelope_correlation": 0.9950269746403829, "全帯域_10ms_modulation_mae_db": 0.04293387384655805, "全帯域_40ms_log_rms_mae_db": 0.6981842294232345, "全帯域_40ms_rms_envelope_nrmse": 0.034917472432440554, "全帯域_40ms_envelope_correlation": 0.9988491679886863, "全帯域_40ms_modulation_mae_db": 0.03418386090893632, "300–1000 Hz_10ms_log_rms_mae_db": 1.9746718690958809, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10141943321778911, "300–1000 Hz_10ms_envelope_correlation": 0.9922582231926615, "300–1000 Hz_10ms_modulation_mae_db": 0.31705991130018446, "300–1000 Hz_40ms_log_rms_mae_db": 1.2919650171724286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04986757888733773, "300–1000 Hz_40ms_envelope_correlation": 0.9980732308467705, "300–1000 Hz_40ms_modulation_mae_db": 0.3029342331771591, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9869475450163645, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13068238042432076, "1000–3000 Hz_10ms_envelope_correlation": 0.9874466496765383, "1000–3000 Hz_10ms_modulation_mae_db": 0.687431179781421, "1000–3000 Hz_40ms_log_rms_mae_db": 1.203756995242599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07171865990367389, "1000–3000 Hz_40ms_envelope_correlation": 0.9965804182145785, "1000–3000 Hz_40ms_modulation_mae_db": 0.6307770564190892, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9689593595058166, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22368724369415302, "3000–6000 Hz_10ms_envelope_correlation": 0.9587571347204664, "3000–6000 Hz_10ms_modulation_mae_db": 0.739370394182088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4804079028466008, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13796497619266326, "3000–6000 Hz_40ms_envelope_correlation": 0.983570840241517, "3000–6000 Hz_40ms_modulation_mae_db": 1.0135473926295935, "mr_stft_db": 5.8719953989604114, "band_mae_db": 0.22446117759008422, "rms_gain_error_db": -0.07364115992221604, "output_peak": 0.6229069140740572}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2549523333322854, "全帯域_10ms_rms_envelope_nrmse": 0.075298456304638, "全帯域_10ms_envelope_correlation": 0.9946714150656527, "全帯域_10ms_modulation_mae_db": 0.1266871001419049, "全帯域_40ms_log_rms_mae_db": 0.712522516419257, "全帯域_40ms_rms_envelope_nrmse": 0.03868544246617765, "全帯域_40ms_envelope_correlation": 0.9985669511585867, "全帯域_40ms_modulation_mae_db": 0.18941536641838544, "300–1000 Hz_10ms_log_rms_mae_db": 2.0066015177721135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10960815437108976, "300–1000 Hz_10ms_envelope_correlation": 0.9909591397754672, "300–1000 Hz_10ms_modulation_mae_db": 0.1663980982494272, "300–1000 Hz_40ms_log_rms_mae_db": 1.381862507935701, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05703122428444189, "300–1000 Hz_40ms_envelope_correlation": 0.9974330321746696, "300–1000 Hz_40ms_modulation_mae_db": 0.17683511690017162, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9118554566700467, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12300964949622444, "1000–3000 Hz_10ms_envelope_correlation": 0.9890337454821013, "1000–3000 Hz_10ms_modulation_mae_db": 0.6039208610872229, "1000–3000 Hz_40ms_log_rms_mae_db": 1.199474840998654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07304741734854892, "1000–3000 Hz_40ms_envelope_correlation": 0.9964961102372452, "1000–3000 Hz_40ms_modulation_mae_db": 0.8354640628994697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9759326104066133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2275871327465843, "3000–6000 Hz_10ms_envelope_correlation": 0.9575604426242524, "3000–6000 Hz_10ms_modulation_mae_db": 0.41487255169513154, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5308149874571673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14464082591088903, "3000–6000 Hz_40ms_envelope_correlation": 0.9820270599907043, "3000–6000 Hz_40ms_modulation_mae_db": 0.5236327643752817, "mr_stft_db": 5.864975189705557, "band_mae_db": 0.20821383876722988, "rms_gain_error_db": -0.07251077854918352, "output_peak": 0.6056202843295001}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2158925387025492, "全帯域_10ms_rms_envelope_nrmse": 0.07208463453791199, "全帯域_10ms_envelope_correlation": 0.9951447857162686, "全帯域_10ms_modulation_mae_db": 0.35401104831869074, "全帯域_40ms_log_rms_mae_db": 0.6296746046900026, "全帯域_40ms_rms_envelope_nrmse": 0.03729332991658017, "全帯域_40ms_envelope_correlation": 0.9986973044397117, "全帯域_40ms_modulation_mae_db": 0.2692531000741542, "300–1000 Hz_10ms_log_rms_mae_db": 2.041148793086503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09984653943558905, "300–1000 Hz_10ms_envelope_correlation": 0.992583219591872, "300–1000 Hz_10ms_modulation_mae_db": 0.17824745530556818, "300–1000 Hz_40ms_log_rms_mae_db": 1.3534874664596581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056169264148857465, "300–1000 Hz_40ms_envelope_correlation": 0.9976758236073009, "300–1000 Hz_40ms_modulation_mae_db": 0.13936522619330893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.874345885884355, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11087331895104019, "1000–3000 Hz_10ms_envelope_correlation": 0.990751464556061, "1000–3000 Hz_10ms_modulation_mae_db": 0.2873491231222381, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1579569787606185, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05752055916594104, "1000–3000 Hz_40ms_envelope_correlation": 0.9975121915701933, "1000–3000 Hz_40ms_modulation_mae_db": 0.29198235785504106, "3000–6000 Hz_10ms_log_rms_mae_db": 2.023327450317768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23269380634351255, "3000–6000 Hz_10ms_envelope_correlation": 0.9553777268632504, "3000–6000 Hz_10ms_modulation_mae_db": 0.6932437977248563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5386994227023214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14394963600229793, "3000–6000 Hz_40ms_envelope_correlation": 0.9819530731142415, "3000–6000 Hz_40ms_modulation_mae_db": 0.8527311286063165, "mr_stft_db": 5.8619732985095085, "band_mae_db": 0.1514679027315936, "rms_gain_error_db": -0.08295674985082674, "output_peak": 0.5984476303380818}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2977139080211364, "全帯域_10ms_rms_envelope_nrmse": 0.07437639017399054, "全帯域_10ms_envelope_correlation": 0.9948468767343541, "全帯域_10ms_modulation_mae_db": 0.22006921923792122, "全帯域_40ms_log_rms_mae_db": 0.7218939486739696, "全帯域_40ms_rms_envelope_nrmse": 0.03760027347016437, "全帯域_40ms_envelope_correlation": 0.998709844091625, "全帯域_40ms_modulation_mae_db": 0.22843835757105305, "300–1000 Hz_10ms_log_rms_mae_db": 2.0810978063415573, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10613493565895511, "300–1000 Hz_10ms_envelope_correlation": 0.9916036418137528, "300–1000 Hz_10ms_modulation_mae_db": 0.30332954026731906, "300–1000 Hz_40ms_log_rms_mae_db": 1.3800407382219257, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06148129130913971, "300–1000 Hz_40ms_envelope_correlation": 0.9971859375226413, "300–1000 Hz_40ms_modulation_mae_db": 0.1644903201427198, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8963509291991043, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12056084749299106, "1000–3000 Hz_10ms_envelope_correlation": 0.9894792212415082, "1000–3000 Hz_10ms_modulation_mae_db": 0.7075819216154547, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1880096283176163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06526689787769537, "1000–3000 Hz_40ms_envelope_correlation": 0.9973603563500336, "1000–3000 Hz_40ms_modulation_mae_db": 0.8148581612100531, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0702312247355796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23639860542198574, "3000–6000 Hz_10ms_envelope_correlation": 0.953852389801716, "3000–6000 Hz_10ms_modulation_mae_db": 0.6398955209344938, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5614492765567864, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14999593656960145, "3000–6000 Hz_40ms_envelope_correlation": 0.9808021631722987, "3000–6000 Hz_40ms_modulation_mae_db": 0.9360312476546581, "mr_stft_db": 5.880315557105914, "band_mae_db": 0.22221221713956152, "rms_gain_error_db": -0.09988662922666663, "output_peak": 0.6054112892933075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1811666620210115, "全帯域_10ms_rms_envelope_nrmse": 0.06598197284485219, "全帯域_10ms_envelope_correlation": 0.995923663092067, "全帯域_10ms_modulation_mae_db": 0.23685770234088518, "全帯域_40ms_log_rms_mae_db": 0.6378496187153373, "全帯域_40ms_rms_envelope_nrmse": 0.03176609012222211, "全帯域_40ms_envelope_correlation": 0.9990526177910791, "全帯域_40ms_modulation_mae_db": 0.34941687789127945, "300–1000 Hz_10ms_log_rms_mae_db": 1.9725108767054278, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09603325283437666, "300–1000 Hz_10ms_envelope_correlation": 0.9930860671843058, "300–1000 Hz_10ms_modulation_mae_db": 0.3662379483907583, "300–1000 Hz_40ms_log_rms_mae_db": 1.2615597256724236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04796391357071836, "300–1000 Hz_40ms_envelope_correlation": 0.9982294446301024, "300–1000 Hz_40ms_modulation_mae_db": 0.4293064945815237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.927763301710589, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11504797836725526, "1000–3000 Hz_10ms_envelope_correlation": 0.9902077829687025, "1000–3000 Hz_10ms_modulation_mae_db": 0.43459052697278966, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2291590067535307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06305936606166937, "1000–3000 Hz_40ms_envelope_correlation": 0.9972127543113389, "1000–3000 Hz_40ms_modulation_mae_db": 0.4701402874588371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.933336041104786, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22527551419122832, "3000–6000 Hz_10ms_envelope_correlation": 0.958316032132935, "3000–6000 Hz_10ms_modulation_mae_db": 0.4155399123520569, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4484342276484923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14004497268100669, "3000–6000 Hz_40ms_envelope_correlation": 0.9829662558599918, "3000–6000 Hz_40ms_modulation_mae_db": 0.5935711377606627, "mr_stft_db": 5.86698852084496, "band_mae_db": 0.20563106346063242, "rms_gain_error_db": -0.06674797843384643, "output_peak": 0.590547481811643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.99479935263719, "spectral_convergence": 0.2812946135353807, "spectral_convergence_status": "ok", "amplitude_floor": 1.4461531006911672e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.699122685224705, "spectral_convergence": 0.26495440127760345, "spectral_convergence_status": "ok", "amplitude_floor": 7.398993822544586e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.884071860067497, "spectral_convergence": 0.2704080415317841, "spectral_convergence_status": "ok", "amplitude_floor": 6.518159095373565e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005975557583734567, "output_power": 0.0006001280129956727, "delta_db": 0.0186546354352046, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005550914808547427, "output_power": 0.0005622013894765337, "delta_db": 0.05527352485009661, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010492685783192576, "output_power": 0.001016266434083812, "delta_db": -0.13879085905431407, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007612304608209731, "output_power": 0.0007149540853760839, "delta_db": -0.27238006221983724, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00020798712521433106, "output_power": 0.00019169452340410783, "delta_db": -0.3542674664122498, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.709473434650137e-06, "output_power": 9.145968216543044e-06, "delta_db": 0.2123778898973802, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1752907396448471, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1183055411849886, "pitch": null, "output_rms_dbfs": -25.094248930429263, "output_peak": 0.5984231055455785, "reference_rms_dbfs": -24.975943389244275, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8593312993097975}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002040524003006053, "log_rms_mae_db": 1.2023771211933298, "rms_envelope_nrmse": 0.07558418566007809, "envelope_correlation": 0.9947103633557373, "rms_gain_error_db": -0.11830552084058618, "reference_mean_power": 0.0031824932955688964, "output_mean_power": 0.003096969621459217}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.3108768872499092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006941576008860869, "output_power": 0.0006805163039870899, "reference_active": true, "delta_db": -0.08619548854309625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.971136227750714e-05, "output_power": 5.990568713761111e-05, "reference_active": true, "delta_db": -0.6583551563618322}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2294065042747786e-05, "output_power": 2.6413512889343845e-05, "reference_active": true, "delta_db": 0.7363690059597229}], "active_band_count": 3, "active_band_mae_db": 0.49363988362155037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019294226027584624, "log_rms_mae_db": 0.6587604048328334, "rms_envelope_nrmse": 0.03865651753608751, "envelope_correlation": 0.9986812555710721, "rms_gain_error_db": -0.11829677426837709, "reference_mean_power": 0.0031984676882902833, "output_mean_power": 0.0031125210001817084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.195649150719305e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000632939746621434, "output_power": 0.0006217393993360164, "reference_active": true, "delta_db": -0.07753979576578154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313010835721604e-05, "output_power": 2.8560334492674916e-05, "reference_active": true, "delta_db": -0.6445956681228284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3632381122717001e-06, "output_power": 1.632770141462512e-06, "reference_active": true, "delta_db": 0.7835333058436742}], "active_band_count": 3, "active_band_mae_db": 0.5018895899107614}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015457760673902765, "log_rms_mae_db": 1.9828276250271148, "rms_envelope_nrmse": 0.11639252454703658, "envelope_correlation": 0.9898212756276336, "rms_gain_error_db": -0.0906641180618461, "reference_mean_power": 0.0011955559481795221, "output_mean_power": 0.0011708560186089732}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2205605502187216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134156506493551, "output_power": 0.0004099134631482374, "reference_active": true, "delta_db": -0.03694731194172631}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.135405678849972e-05, "output_power": 3.692611342186041e-05, "reference_active": true, "delta_db": -0.49184520469277265}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1249926033986765e-05, "output_power": 1.2942698822777444e-05, "reference_active": true, "delta_db": 0.6087517818015632}], "active_band_count": 3, "active_band_mae_db": 0.3791814328120207}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014615313182534672, "log_rms_mae_db": 1.3483451100642658, "rms_envelope_nrmse": 0.06342619351335063, "envelope_correlation": 0.9969211042198463, "rms_gain_error_db": -0.09066639450468984, "reference_mean_power": 0.0012015673286161978, "output_mean_power": 0.0011767425884009058}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.16593985496473e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003798908177653278, "output_power": 0.00037493219130315514, "reference_active": true, "delta_db": -0.05706066358675918}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8446225039409073e-05, "output_power": 1.6554453186845077e-05, "reference_active": true, "delta_db": -0.46992662554989767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.60382048073901e-07, "output_power": 8.627603062989797e-07, "reference_active": true, "delta_db": 0.5485830093964331}], "active_band_count": 3, "active_band_mae_db": 0.35852343284436333}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011311095186173674, "log_rms_mae_db": 1.962795576042538, "rms_envelope_nrmse": 0.12520123276101838, "envelope_correlation": 0.9885661312816579, "rms_gain_error_db": -0.2867270840498037, "reference_mean_power": 0.0008539373095994977, "output_mean_power": 0.0007993800039822773}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.101259078601872e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000229953654454443, "output_power": 0.00021968159648210714, "reference_active": true, "delta_db": -0.19846639798541466}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8372953922112745e-05, "output_power": 4.476056390376709e-05, "reference_active": true, "delta_db": -1.153161218003623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3078079511035874e-05, "output_power": 1.5199672837536268e-05, "reference_active": true, "delta_db": 0.6529026673372923}], "active_band_count": 3, "active_band_mae_db": 0.6681767611087767}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.65333606126804e-05, "log_rms_mae_db": 1.1783745000176027, "rms_envelope_nrmse": 0.07108150370551922, "envelope_correlation": 0.996680140430797, "rms_gain_error_db": -0.2867304595940904, "reference_mean_power": 0.0008582264011938477, "output_mean_power": 0.0008033944448403485}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.585136403392408e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214290189767478, "output_power": 0.0002025253747543715, "reference_active": true, "delta_db": -0.20149060793813572}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7723754282185457e-05, "output_power": 2.1385570510903723e-05, "reference_active": true, "delta_db": -1.127312004214406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.120050210892624e-07, "output_power": 1.0093312868177777e-06, "reference_active": true, "delta_db": 0.44036506170744816}], "active_band_count": 3, "active_band_mae_db": 0.58972255795333}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.9261609894090753e-05, "log_rms_mae_db": 1.9835951111770418, "rms_envelope_nrmse": 0.22351216983342323, "envelope_correlation": 0.9589818498380783, "rms_gain_error_db": -0.2952258005598317, "reference_mean_power": 3.945170478466558e-05, "output_mean_power": 3.685896996830348e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2228241150412126e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1970652723962295e-05, "output_power": 8.786548517379397e-06, "reference_active": true, "delta_db": -1.342995203806034}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3243801130743035e-06, "output_power": 2.9992667904894917e-06, "reference_active": true, "delta_db": -0.4469557683474765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0798067724676823e-06, "output_power": 9.422494534244165e-07, "reference_active": true, "delta_db": -0.5918015280818449}], "active_band_count": 3, "active_band_mae_db": 0.7939175000784519}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5144247671385576e-05, "log_rms_mae_db": 1.4887437982320988, "rms_envelope_nrmse": 0.14278444956811917, "envelope_correlation": 0.9832961605798443, "rms_gain_error_db": -0.2952746930502849, "reference_mean_power": 3.9649231901027684e-05, "output_mean_power": 3.7043098728609324e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.976126261669901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1283346332236661e-05, "output_power": 8.041661045961774e-06, "reference_active": true, "delta_db": -1.4709215488092175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7283331039583737e-06, "output_power": 1.4260234149554947e-06, "reference_active": true, "delta_db": -0.8350079178469837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.035315524247478e-08, "output_power": 5.997778581928334e-08, "reference_active": true, "delta_db": -0.692931507100334}], "active_band_count": 3, "active_band_mae_db": 0.9996203245855116}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7775239850495226, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "33680393674ad305f7d1938f6ae649505ee03322f0a2e73556d3367c67ea3a2f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9887857024648308, "standard_deviation": 0.001315272351467847, "minimum": 0.9873819222673054, "maximum": 0.990751464556061, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.928739523659524, "standard_deviation": 0.044529793450141186, "minimum": 1.874345885884355, "maximum": 1.9869475450163645, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5206873827332943, "standard_deviation": 0.16541005960227778, "minimum": 0.2873491231222381, "maximum": 0.7075819216154547, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12333927050751181, "standard_deviation": 0.007533194696839792, "minimum": 0.11087331895104019, "maximum": 0.1311277840311757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968599979617007, "standard_deviation": 0.0004283303575439628, "minimum": 0.9964532882509207, "maximum": 0.9975121915701933, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1968068490699948, "standard_deviation": 0.030045772577345566, "minimum": 1.1579569787606185, "maximum": 1.2475917317233374, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5341640985948897, "standard_deviation": 0.21927026381253756, "minimum": 0.29198235785504106, "maximum": 0.8354640628994697, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0681115728594807, "standard_deviation": 0.005637467071162348, "minimum": 0.05752055916594104, "maximum": 0.07335422959845313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9565224840908906, "standard_deviation": 0.0025048762318139175, "minimum": 0.9520922595530654, "maximum": 0.9589818498380783, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.996072383668885, "standard_deviation": 0.04186934216873197, "minimum": 1.933336041104786, "maximum": 2.0702312247355796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5929660415813014, "standard_deviation": 0.1448861936135273, "minimum": 0.41487255169513154, "maximum": 0.7939175000784519, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22970272496777527, "standard_deviation": 0.006386192191384463, "minimum": 0.22351216983342323, "maximum": 0.2410181698524039, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9819108493824098, "standard_deviation": 0.001771851667591338, "minimum": 0.9781017334897372, "maximum": 0.983570840241517, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5119111531300684, "standard_deviation": 0.03785561912498722, "minimum": 1.4484342276484923, "maximum": 1.5614492765567864, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.788642429043972, "standard_deviation": 0.19131809303048475, "minimum": 0.5236327643752817, "maximum": 1.0135473926295935, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14482643172594772, "standard_deviation": 0.006291449202998469, "minimum": 0.13796497619266326, "maximum": 0.15766426171424525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9917764836298869, "standard_deviation": 0.0010127050239634582, "minimum": 0.9898212756276336, "maximum": 0.9930860671843058, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9783416658189958, "standard_deviation": 0.07542135722014699, "minimum": 1.8263849383611979, "maximum": 2.0810978063415573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3542177790234819, "standard_deviation": 0.1682571754864813, "minimum": 0.1663980982494272, "maximum": 0.7050704783366636, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10458288108083767, "standard_deviation": 0.0062629069300712175, "minimum": 0.09603325283437666, "maximum": 0.11639252454703658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997643662180842, "standard_deviation": 0.0004416180171795382, "minimum": 0.9969211042198463, "maximum": 0.9982294446301024, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3065990274998147, "standard_deviation": 0.07413559643181483, "minimum": 1.165575304993692, "maximum": 1.381862507935701, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.2879761026857939, "standard_deviation": 0.11462447899301677, "minimum": 0.13936522619330893, "maximum": 0.4293064945815237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05532438201802757, "standard_deviation": 0.005360278972637971, "minimum": 0.04796391357071836, "maximum": 0.06342619351335063, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20860596481051757, "standard_deviation": 0.03162678351021255, "minimum": 0.1514679027315936, "maximum": 0.2482204164236699, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.867844053474287, "standard_deviation": 0.006898329544484385, "minimum": 5.8593312993097975, "maximum": 5.880315557105914, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6071317704892422, "standard_deviation": 0.015125455277648046, "minimum": 0.590547481811643, "maximum": 0.6364411444291814, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08465781504147178, "standard_deviation": 0.02258403373806968, "minimum": -0.1183055411849886, "maximum": -0.05336685336227699, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9950179800345756, "standard_deviation": 0.00039745461481526727, "minimum": 0.9946714150656527, "maximum": 0.995923663092067, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2455197871380193, "standard_deviation": 0.04471522941412063, "minimum": 1.1811666620210115, "maximum": 1.298001458324749, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24294549358757417, "standard_deviation": 0.14350856163267098, "minimum": 0.04293387384655805, "maximum": 0.49363988362155037, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07296444103117009, "standard_deviation": 0.003053459307178295, "minimum": 0.06598197284485219, "maximum": 0.07558418566007809, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987203156552356, "standard_deviation": 0.00016327246309443742, "minimum": 0.9985469179255667, "maximum": 0.9990526177910791, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6739725575134233, "standard_deviation": 0.0376752375572043, "minimum": 0.6296746046900026, "maximum": 0.7218939486739696, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24470498306621713, "standard_deviation": 0.14060011548182463, "minimum": 0.03418386090893632, "maximum": 0.5018895899107614, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037024405664952895, "standard_deviation": 0.0024821888285459595, "minimum": 0.03176609012222211, "maximum": 0.03868926838631979, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28240-0011", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.142855644226074, "predicted_mos": 2.142855644226074, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3bf5b63af191c6290d7dc6519aaae7c34910426c989fa8f70194c0d9c756dfbc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.957165400031954, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2908828768110623, "全帯域_10ms_rms_envelope_nrmse": 0.0748422427193381, "全帯域_10ms_envelope_correlation": 0.9947713823555405, "全帯域_10ms_modulation_mae_db": 0.28544685756813226, "全帯域_40ms_log_rms_mae_db": 0.6957418265327335, "全帯域_40ms_rms_envelope_nrmse": 0.038423798426152206, "全帯域_40ms_envelope_correlation": 0.9986332858979915, "全帯域_40ms_modulation_mae_db": 0.13044333402731964, "300–1000 Hz_10ms_log_rms_mae_db": 1.5731497255131548, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1044357201326327, "300–1000 Hz_10ms_envelope_correlation": 0.9917805065955132, "300–1000 Hz_10ms_modulation_mae_db": 0.5992996843275146, "300–1000 Hz_40ms_log_rms_mae_db": 0.9620851390117648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05725498687864976, "300–1000 Hz_40ms_envelope_correlation": 0.9974191248581415, "300–1000 Hz_40ms_modulation_mae_db": 0.4960117663132775, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4469773496209948, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11802663693710737, "1000–3000 Hz_10ms_envelope_correlation": 0.9901660528011386, "1000–3000 Hz_10ms_modulation_mae_db": 0.32431474833129376, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8918824078048804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07121396096401805, "1000–3000 Hz_40ms_envelope_correlation": 0.9969516917369154, "1000–3000 Hz_40ms_modulation_mae_db": 0.5835055534710053, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3934137682273935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20116949775113963, "3000–6000 Hz_10ms_envelope_correlation": 0.9666381200968488, "3000–6000 Hz_10ms_modulation_mae_db": 0.26354738666939304, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0207956160895713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1371304541972128, "3000–6000 Hz_40ms_envelope_correlation": 0.9837114840835275, "3000–6000 Hz_40ms_modulation_mae_db": 0.4301848679280178, "mr_stft_db": 5.530944753196972, "band_mae_db": 0.17026242295853575, "rms_gain_error_db": -0.09492103625258475, "output_peak": 0.6204120722930955}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3229117833188995, "全帯域_10ms_rms_envelope_nrmse": 0.07730628679410076, "全帯域_10ms_envelope_correlation": 0.9943919261597252, "全帯域_10ms_modulation_mae_db": 0.380480850480553, "全帯域_40ms_log_rms_mae_db": 0.7479474464066934, "全帯域_40ms_rms_envelope_nrmse": 0.038928205977865796, "全帯域_40ms_envelope_correlation": 0.9985030468307654, "全帯域_40ms_modulation_mae_db": 0.39308357683078166, "300–1000 Hz_10ms_log_rms_mae_db": 1.5700532879992022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10688203633478077, "300–1000 Hz_10ms_envelope_correlation": 0.9914219166652021, "300–1000 Hz_10ms_modulation_mae_db": 0.08100123891797889, "300–1000 Hz_40ms_log_rms_mae_db": 0.9233430224076292, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056647226767008896, "300–1000 Hz_40ms_envelope_correlation": 0.9974767390967599, "300–1000 Hz_40ms_modulation_mae_db": 0.11593107524131467, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4079711109843516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12284779889548174, "1000–3000 Hz_10ms_envelope_correlation": 0.9885405408908466, "1000–3000 Hz_10ms_modulation_mae_db": 0.1866420099352973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.815019927899377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0656484271877645, "1000–3000 Hz_40ms_envelope_correlation": 0.9965362727541033, "1000–3000 Hz_40ms_modulation_mae_db": 0.27250251313959684, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3817574578638387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18414203157503214, "3000–6000 Hz_10ms_envelope_correlation": 0.9721893548870069, "3000–6000 Hz_10ms_modulation_mae_db": 0.6582580070236826, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9519975504502202, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11181907082090868, "3000–6000 Hz_40ms_envelope_correlation": 0.9895072939255828, "3000–6000 Hz_40ms_modulation_mae_db": 0.7481120824478991, "mr_stft_db": 5.538321989182205, "band_mae_db": 0.13811063643519067, "rms_gain_error_db": 0.023700222697129836, "output_peak": 0.6070632717060588}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2931300052937105, "全帯域_10ms_rms_envelope_nrmse": 0.07339558347420894, "全帯域_10ms_envelope_correlation": 0.9949385191445373, "全帯域_10ms_modulation_mae_db": 0.3451471778955523, "全帯域_40ms_log_rms_mae_db": 0.6714800137443527, "全帯域_40ms_rms_envelope_nrmse": 0.037426458192229714, "全帯域_40ms_envelope_correlation": 0.9986601946413567, "全帯域_40ms_modulation_mae_db": 0.6583505634877648, "300–1000 Hz_10ms_log_rms_mae_db": 1.5957963041447596, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10447737090987771, "300–1000 Hz_10ms_envelope_correlation": 0.991775766111011, "300–1000 Hz_10ms_modulation_mae_db": 0.4359518767672545, "300–1000 Hz_40ms_log_rms_mae_db": 0.9319455697020025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05401566108189075, "300–1000 Hz_40ms_envelope_correlation": 0.9977163815791069, "300–1000 Hz_40ms_modulation_mae_db": 0.8111317659002278, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4660469536635143, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12137321810747131, "1000–3000 Hz_10ms_envelope_correlation": 0.9891671567145576, "1000–3000 Hz_10ms_modulation_mae_db": 0.4473668231617605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8720220592626684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0733135916460669, "1000–3000 Hz_40ms_envelope_correlation": 0.9962277018084364, "1000–3000 Hz_40ms_modulation_mae_db": 0.5601654468817339, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3627302909834615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19832642335427897, "3000–6000 Hz_10ms_envelope_correlation": 0.9675705486612404, "3000–6000 Hz_10ms_modulation_mae_db": 0.6215419433238222, "3000–6000 Hz_40ms_log_rms_mae_db": 0.973844237000768, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13050264149211235, "3000–6000 Hz_40ms_envelope_correlation": 0.9853367028194215, "3000–6000 Hz_40ms_modulation_mae_db": 0.8845915160525465, "mr_stft_db": 5.534711259149017, "band_mae_db": 0.15929273639653302, "rms_gain_error_db": -0.07076182130887099, "output_peak": 0.6482361014828981}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2819093300243354, "全帯域_10ms_rms_envelope_nrmse": 0.08588207127913414, "全帯域_10ms_envelope_correlation": 0.9930672006690195, "全帯域_10ms_modulation_mae_db": 0.4844813823625924, "全帯域_40ms_log_rms_mae_db": 0.6219510072232094, "全帯域_40ms_rms_envelope_nrmse": 0.044904286778431286, "全帯域_40ms_envelope_correlation": 0.9980098816977193, "全帯域_40ms_modulation_mae_db": 0.42686053418177955, "300–1000 Hz_10ms_log_rms_mae_db": 1.5532929515449614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10647602710813518, "300–1000 Hz_10ms_envelope_correlation": 0.9916161226308843, "300–1000 Hz_10ms_modulation_mae_db": 0.33218211420844485, "300–1000 Hz_40ms_log_rms_mae_db": 0.8585541402966324, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056770619732244995, "300–1000 Hz_40ms_envelope_correlation": 0.9975471545284614, "300–1000 Hz_40ms_modulation_mae_db": 0.321512821329464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4356739378211354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13519404814242025, "1000–3000 Hz_10ms_envelope_correlation": 0.9865455894335866, "1000–3000 Hz_10ms_modulation_mae_db": 0.5034183222528329, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8519203039447439, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0781685151434729, "1000–3000 Hz_40ms_envelope_correlation": 0.9958207404415302, "1000–3000 Hz_40ms_modulation_mae_db": 0.4289102521155248, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3917762385477455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19588776634130844, "3000–6000 Hz_10ms_envelope_correlation": 0.9685068780583396, "3000–6000 Hz_10ms_modulation_mae_db": 0.7264551946309575, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9934410300044989, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12523709057588628, "3000–6000 Hz_40ms_envelope_correlation": 0.9868460955151822, "3000–6000 Hz_40ms_modulation_mae_db": 0.8733247876392629, "mr_stft_db": 5.54034250416802, "band_mae_db": 0.19222946182937825, "rms_gain_error_db": 0.0011161405456254674, "output_peak": 0.6215077612926784}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3024534440247397, "全帯域_10ms_rms_envelope_nrmse": 0.07242335150873111, "全帯域_10ms_envelope_correlation": 0.9950539172105789, "全帯域_10ms_modulation_mae_db": 0.31737387306459897, "全帯域_40ms_log_rms_mae_db": 0.7221545220171415, "全帯域_40ms_rms_envelope_nrmse": 0.03291656190683716, "全帯域_40ms_envelope_correlation": 0.9989286283260113, "全帯域_40ms_modulation_mae_db": 0.28427255286473985, "300–1000 Hz_10ms_log_rms_mae_db": 1.5690615661087735, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1053963591364231, "300–1000 Hz_10ms_envelope_correlation": 0.9916604791787251, "300–1000 Hz_10ms_modulation_mae_db": 0.5448684212267348, "300–1000 Hz_40ms_log_rms_mae_db": 0.9320198835997473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04984717568970138, "300–1000 Hz_40ms_envelope_correlation": 0.9980424349758198, "300–1000 Hz_40ms_modulation_mae_db": 0.5662223590794567, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4397569101781722, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12342979756218025, "1000–3000 Hz_10ms_envelope_correlation": 0.9885071515017326, "1000–3000 Hz_10ms_modulation_mae_db": 0.07616164837880464, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7942109933176282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06808856120981122, "1000–3000 Hz_40ms_envelope_correlation": 0.9964401747257192, "1000–3000 Hz_40ms_modulation_mae_db": 0.14676404574919608, "3000–6000 Hz_10ms_log_rms_mae_db": 1.334790241104065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1958035728516654, "3000–6000 Hz_10ms_envelope_correlation": 0.9683425274623695, "3000–6000 Hz_10ms_modulation_mae_db": 0.4777705473105524, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9421275628206017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1244369218875338, "3000–6000 Hz_40ms_envelope_correlation": 0.986562131281853, "3000–6000 Hz_40ms_modulation_mae_db": 0.5437588101956674, "mr_stft_db": 5.524521228339641, "band_mae_db": 0.1406080456511293, "rms_gain_error_db": -0.023891547910107235, "output_peak": 0.5942236265022645}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.299549596095545, "全帯域_10ms_rms_envelope_nrmse": 0.07233029153400176, "全帯域_10ms_envelope_correlation": 0.9951131706520361, "全帯域_10ms_modulation_mae_db": 0.2795262754251047, "全帯域_40ms_log_rms_mae_db": 0.7053483643119696, "全帯域_40ms_rms_envelope_nrmse": 0.03626086833098568, "全帯域_40ms_envelope_correlation": 0.9987841410017586, "全帯域_40ms_modulation_mae_db": 0.30334903032729654, "300–1000 Hz_10ms_log_rms_mae_db": 1.5252705810188723, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09449841834307038, "300–1000 Hz_10ms_envelope_correlation": 0.9934141232885726, "300–1000 Hz_10ms_modulation_mae_db": 0.6506725722469975, "300–1000 Hz_40ms_log_rms_mae_db": 0.9682212736814337, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050225119314793794, "300–1000 Hz_40ms_envelope_correlation": 0.9982318527500549, "300–1000 Hz_40ms_modulation_mae_db": 0.6017470976911558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4396174118601885, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12658952700198117, "1000–3000 Hz_10ms_envelope_correlation": 0.9880605172054568, "1000–3000 Hz_10ms_modulation_mae_db": 0.33209367625943986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.797176094505683, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06782150447850838, "1000–3000 Hz_40ms_envelope_correlation": 0.9967482507408333, "1000–3000 Hz_40ms_modulation_mae_db": 0.46370735691582016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.394916442765509, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19791551689058526, "3000–6000 Hz_10ms_envelope_correlation": 0.9677786006970804, "3000–6000 Hz_10ms_modulation_mae_db": 0.37670814694040516, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9744886930464769, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12715808267062123, "3000–6000 Hz_40ms_envelope_correlation": 0.9855830481517, "3000–6000 Hz_40ms_modulation_mae_db": 0.4502436807411902, "mr_stft_db": 5.515516536604569, "band_mae_db": 0.21214915634567502, "rms_gain_error_db": -0.08993842620155532, "output_peak": 0.6010688912443325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.333343964782453, "全帯域_10ms_rms_envelope_nrmse": 0.07707094372512172, "全帯域_10ms_envelope_correlation": 0.9943998831009624, "全帯域_10ms_modulation_mae_db": 0.3114350360284005, "全帯域_40ms_log_rms_mae_db": 0.644870766186536, "全帯域_40ms_rms_envelope_nrmse": 0.03634765489933137, "全帯域_40ms_envelope_correlation": 0.9987191946295976, "全帯域_40ms_modulation_mae_db": 0.3051085461267595, "300–1000 Hz_10ms_log_rms_mae_db": 1.5542252469864068, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09364795909426875, "300–1000 Hz_10ms_envelope_correlation": 0.9934040434667551, "300–1000 Hz_10ms_modulation_mae_db": 0.39997520510816736, "300–1000 Hz_40ms_log_rms_mae_db": 0.8439982584682849, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04679254982085814, "300–1000 Hz_40ms_envelope_correlation": 0.9982721660824427, "300–1000 Hz_40ms_modulation_mae_db": 0.24337475084663415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4493253598636309, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12578122070164846, "1000–3000 Hz_10ms_envelope_correlation": 0.9881140613267031, "1000–3000 Hz_10ms_modulation_mae_db": 0.3804899515592661, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8274576736628995, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06315429201142404, "1000–3000 Hz_40ms_envelope_correlation": 0.9970994352601744, "1000–3000 Hz_40ms_modulation_mae_db": 0.27883624209420993, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4046637651636378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1965594725695514, "3000–6000 Hz_10ms_envelope_correlation": 0.9683683029940224, "3000–6000 Hz_10ms_modulation_mae_db": 0.7496372235839752, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9685007012464988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12564130942335205, "3000–6000 Hz_40ms_envelope_correlation": 0.9868821150803584, "3000–6000 Hz_40ms_modulation_mae_db": 1.0235926296727635, "mr_stft_db": 5.530787697205468, "band_mae_db": 0.14474706284986918, "rms_gain_error_db": -0.05208158047432989, "output_peak": 0.6073874306735393}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3593378779055028, "全帯域_10ms_rms_envelope_nrmse": 0.07363089219046758, "全帯域_10ms_envelope_correlation": 0.994881416670051, "全帯域_10ms_modulation_mae_db": 0.44372972486726225, "全帯域_40ms_log_rms_mae_db": 0.7855200850661176, "全帯域_40ms_rms_envelope_nrmse": 0.03559671179354179, "全帯域_40ms_envelope_correlation": 0.9987649971999539, "全帯域_40ms_modulation_mae_db": 0.5709792095991745, "300–1000 Hz_10ms_log_rms_mae_db": 1.6115643647310376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09961521461667293, "300–1000 Hz_10ms_envelope_correlation": 0.9925265250495541, "300–1000 Hz_10ms_modulation_mae_db": 0.544314315317529, "300–1000 Hz_40ms_log_rms_mae_db": 0.9269645170841693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05126754999178503, "300–1000 Hz_40ms_envelope_correlation": 0.9979392305246438, "300–1000 Hz_40ms_modulation_mae_db": 0.6347027738134857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.456219151787815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11577239826138201, "1000–3000 Hz_10ms_envelope_correlation": 0.990144489147693, "1000–3000 Hz_10ms_modulation_mae_db": 0.11191483038693761, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8179406961996946, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06215021971932776, "1000–3000 Hz_40ms_envelope_correlation": 0.997408092914502, "1000–3000 Hz_40ms_modulation_mae_db": 0.1835053834440119, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3334049131952277, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18424604608680895, "3000–6000 Hz_10ms_envelope_correlation": 0.9720117998970214, "3000–6000 Hz_10ms_modulation_mae_db": 0.5426186456589194, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9014678467193603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1189372251357219, "3000–6000 Hz_40ms_envelope_correlation": 0.9876330839865352, "3000–6000 Hz_40ms_modulation_mae_db": 0.5453081759899826, "mr_stft_db": 5.531409107970741, "band_mae_db": 0.14400563142039569, "rms_gain_error_db": -0.042393460994501676, "output_peak": 0.6034533185149011}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.55856978937317, "spectral_convergence": 0.2753408239820216, "spectral_convergence_status": "ok", "amplitude_floor": 1.4461531006911672e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.311066294729619, "spectral_convergence": 0.2595703438250202, "spectral_convergence_status": "ok", "amplitude_floor": 7.398993822544586e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.723198175488125, "spectral_convergence": 0.2759352681614436, "spectral_convergence_status": "ok", "amplitude_floor": 6.518159095373565e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005975557583734567, "output_power": 0.0006097306461751982, "delta_db": 0.08759588150906687, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005550914808547427, "output_power": 0.0005664589669527101, "delta_db": 0.08803893854107611, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010492685783192576, "output_power": 0.0010269329060228183, "delta_db": -0.09344597373667982, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007612304608209731, "output_power": 0.0007106369949514333, "delta_db": -0.29868345744193675, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00020798712521433106, "output_power": 0.00018857130998762114, "delta_db": -0.4256083408354009, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.709473434650137e-06, "output_power": 8.766214504448507e-06, "delta_db": 0.02820194568705392, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17026242295853575, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09492103625258475, "pitch": null, "output_rms_dbfs": -25.070864425496865, "output_peak": 0.6204120722930955, "reference_rms_dbfs": -24.975943389244275, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.530944753196972}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002040524003006053, "log_rms_mae_db": 1.2908828768110623, "rms_envelope_nrmse": 0.0748422427193381, "envelope_correlation": 0.9947713823555405, "rms_gain_error_db": -0.09492059914973347, "reference_mean_power": 0.0031824932955688964, "output_mean_power": 0.0031136904704500237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.3108768872499092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006941576008860869, "output_power": 0.0006979701528541253, "reference_active": true, "delta_db": 0.023787679209749946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.971136227750714e-05, "output_power": 6.357280883535946e-05, "reference_active": true, "delta_db": -0.4003216939211073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2294065042747786e-05, "output_power": 2.4627049527792664e-05, "reference_active": true, "delta_db": 0.43223119957353956}], "active_band_count": 3, "active_band_mae_db": 0.28544685756813226}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019294226027584624, "log_rms_mae_db": 0.6957418265327335, "rms_envelope_nrmse": 0.038423798426152206, "envelope_correlation": 0.9986332858979915, "rms_gain_error_db": -0.09492471359658475, "reference_mean_power": 0.0031984676882902833, "output_mean_power": 0.0031293165455792415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.195649150719305e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000632939746621434, "output_power": 0.0006374618537950119, "reference_active": true, "delta_db": 0.030918325138033873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313010835721604e-05, "output_power": 3.122551492304466e-05, "reference_active": true, "delta_db": -0.2571324749233854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3632381122717001e-06, "output_power": 1.3312015022123893e-06, "reference_active": true, "delta_db": -0.10327920202053964}], "active_band_count": 3, "active_band_mae_db": 0.13044333402731964}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015457760673902765, "log_rms_mae_db": 1.5731497255131548, "rms_envelope_nrmse": 0.1044357201326327, "envelope_correlation": 0.9917805065955132, "rms_gain_error_db": -0.04215051961613857, "reference_mean_power": 0.0011955559481795221, "output_mean_power": 0.0011840085882053272}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2205605502187216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134156506493551, "output_power": 0.00040915591964549666, "reference_active": true, "delta_db": -0.04498074753979042}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.135405678849972e-05, "output_power": 3.812044961666e-05, "reference_active": true, "delta_db": -0.3536010546011982}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1249926033986765e-05, "output_power": 1.552677973691862e-05, "reference_active": true, "delta_db": 1.399317250841555}], "active_band_count": 3, "active_band_mae_db": 0.5992996843275146}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014615313182534672, "log_rms_mae_db": 0.9620851390117648, "rms_envelope_nrmse": 0.05725498687864976, "envelope_correlation": 0.9974191248581415, "rms_gain_error_db": -0.04215161925875941, "reference_mean_power": 0.0012015673286161978, "output_mean_power": 0.0011899616060067025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.16593985496473e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003798908177653278, "output_power": 0.00037433338459963054, "reference_active": true, "delta_db": -0.06400235451180009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8446225039409073e-05, "output_power": 1.748814242580054e-05, "reference_active": true, "delta_db": -0.23163821046837857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.60382048073901e-07, "output_power": 1.0006249519627645e-06, "reference_active": true, "delta_db": 1.192394733959654}], "active_band_count": 3, "active_band_mae_db": 0.4960117663132775}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011311095186173674, "log_rms_mae_db": 1.4469773496209948, "rms_envelope_nrmse": 0.11802663693710737, "envelope_correlation": 0.9901660528011386, "rms_gain_error_db": -0.3283959412778464, "reference_mean_power": 0.0008539373095994977, "output_mean_power": 0.000791746942119819}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.101259078601872e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000229953654454443, "output_power": 0.00022419800619232882, "reference_active": true, "delta_db": -0.11008569764662898}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8372953922112745e-05, "output_power": 5.2154949492860207e-05, "reference_active": true, "delta_db": -0.48916141865845025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3078079511035874e-05, "output_power": 1.1999806985227292e-05, "reference_active": true, "delta_db": -0.37369712868880195}], "active_band_count": 3, "active_band_mae_db": 0.32431474833129376}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.65333606126804e-05, "log_rms_mae_db": 0.8918824078048804, "rms_envelope_nrmse": 0.07121396096401805, "envelope_correlation": 0.9969516917369154, "rms_gain_error_db": -0.3283963386424485, "reference_mean_power": 0.0008582264011938477, "output_mean_power": 0.0007957235958441286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.585136403392408e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214290189767478, "output_power": 0.000206872697879928, "reference_active": true, "delta_db": -0.10925327027836161}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7723754282185457e-05, "output_power": 2.5390766513877303e-05, "reference_active": true, "delta_db": -0.3817622915459992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.120050210892624e-07, "output_power": 6.824127296356606e-07, "reference_active": true, "delta_db": -1.2595010985886552}], "active_band_count": 3, "active_band_mae_db": 0.5835055534710053}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.9261609894090753e-05, "log_rms_mae_db": 1.3934137682273935, "rms_envelope_nrmse": 0.20116949775113963, "envelope_correlation": 0.9666381200968488, "rms_gain_error_db": -0.2824619409611092, "reference_mean_power": 3.945170478466558e-05, "output_mean_power": 3.696745735578218e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2228241150412126e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1970652723962295e-05, "output_power": 1.0368121295774267e-05, "reference_active": true, "delta_db": -0.6241776250318843}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3243801130743035e-06, "output_power": 3.214514998025148e-06, "reference_active": true, "delta_db": -0.14595219234847415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0798067724676823e-06, "output_power": 1.07471871761813e-06, "reference_active": true, "delta_db": -0.020512342627820652}], "active_band_count": 3, "active_band_mae_db": 0.26354738666939304}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5144247671385576e-05, "log_rms_mae_db": 1.0207956160895713, "rms_envelope_nrmse": 0.1371304541972128, "envelope_correlation": 0.9837114840835275, "rms_gain_error_db": -0.282453441988216, "reference_mean_power": 3.9649231901027684e-05, "output_mean_power": 3.715261902806287e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.976126261669901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1283346332236661e-05, "output_power": 9.517136935194528e-06, "reference_active": true, "delta_db": -0.7393160052247195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7283331039583737e-06, "output_power": 1.577241932802477e-06, "reference_active": true, "delta_db": -0.3972913359202173}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.035315524247478e-08, "output_power": 6.790298300916424e-08, "reference_active": true, "delta_db": -0.15394726263911643}], "active_band_count": 3, "active_band_mae_db": 0.4301848679280178}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7940457541170142, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "650af07f40c8ea89eecf9b64598b351ef9e760ced811496a63408a6d750b42a1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9886556948777143, "standard_deviation": 0.001188922475631416, "minimum": 0.9865455894335866, "maximum": 0.9901660528011386, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4426985232224752, "standard_deviation": 0.01717000796832195, "minimum": 1.4079711109843516, "maximum": 1.4660469536635143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.2953002512832041, "standard_deviation": 0.1555399853200774, "minimum": 0.07616164837880464, "maximum": 0.5034183222528329, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12362683070120907, "standard_deviation": 0.005928119991785123, "minimum": 0.11577239826138201, "maximum": 0.13519404814242025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966540450477768, "standard_deviation": 0.0005071214060204413, "minimum": 0.9958207404415302, "maximum": 0.997408092914502, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8334537695746969, "standard_deviation": 0.035284379291564326, "minimum": 0.7942109933176282, "maximum": 0.8918824078048804, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.36473709922638736, "standard_deviation": 0.1674210022978694, "minimum": 0.14676404574919608, "maximum": 0.5835055534710053, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06869488404504921, "standard_deviation": 0.00536350104831485, "minimum": 0.06215021971932776, "maximum": 0.0781685151434729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9689257665942411, "standard_deviation": 0.00204893501901904, "minimum": 0.9666381200968488, "maximum": 0.9721893548870069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3746816397313597, "standard_deviation": 0.027875967178790213, "minimum": 1.3334049131952277, "maximum": 1.4046637651636378, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5520671368927135, "standard_deviation": 0.17119415139412342, "minimum": 0.26354738666939304, "maximum": 0.7496372235839752, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19425629092754626, "standard_deviation": 0.006443291353708363, "minimum": 0.18414203157503214, "maximum": 0.20116949775113963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9865077443555201, "standard_deviation": 0.0017140908731688232, "minimum": 0.9837114840835275, "maximum": 0.9895072939255828, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9658329046722495, "standard_deviation": 0.03550764591978793, "minimum": 0.9014678467193603, "maximum": 1.0207956160895713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6873895688334163, "standard_deviation": 0.22468695136086345, "minimum": 0.4301848679280178, "maximum": 1.0235926296727635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12510784952541865, "standard_deviation": 0.0074954265182281205, "minimum": 0.11181907082090868, "maximum": 0.1371304541972128, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921999353732772, "standard_deviation": 0.0008127372522077094, "minimum": 0.9914219166652021, "maximum": 0.9934141232885726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5690517535058959, "standard_deviation": 0.026525852196921416, "minimum": 1.5252705810188723, "maximum": 1.6115643647310376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4485331785150777, "standard_deviation": 0.18251373398539125, "minimum": 0.08100123891797889, "maximum": 0.6506725722469975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1019286382094827, "standard_deviation": 0.005333870585365898, "minimum": 0.09364795909426875, "maximum": 0.10688203633478077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978306355494289, "standard_deviation": 0.0003381832822094387, "minimum": 0.9974191248581415, "maximum": 0.9982721660824427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9183914755314581, "standard_deviation": 0.04468055549949178, "minimum": 0.8439982584682849, "maximum": 0.9682212736814337, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.47382930127687706, "standard_deviation": 0.22973446734643116, "minimum": 0.11593107524131467, "maximum": 0.8111317659002278, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05285261115961659, "standard_deviation": 0.0038850742587131152, "minimum": 0.04679254982085814, "maximum": 0.05725498687864976, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16267564423583836, "standard_deviation": 0.027125322213371664, "minimum": 0.13811063643519067, "maximum": 0.21214915634567502, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.530819384477079, "standard_deviation": 0.007895152160030745, "minimum": 5.515516536604569, "maximum": 5.54034250416802, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.612919059213721, "standard_deviation": 0.016994283143128154, "minimum": 0.5942236265022645, "maximum": 0.6482361014828981, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04364643873739932, "standard_deviation": 0.04227379299132504, "minimum": -0.09492103625258475, "maximum": 0.023700222697129836, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945771769953065, "standard_deviation": 0.0006677462719966224, "minimum": 0.9930672006690195, "maximum": 0.9951131706520361, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.310439859782031, "standard_deviation": 0.026048420660600174, "minimum": 1.2819093300243354, "maximum": 1.3593378779055028, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3559526472115246, "standard_deviation": 0.07485704303333847, "minimum": 0.2795262754251047, "maximum": 0.4844813823625924, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07586020790313801, "standard_deviation": 0.004477683355157612, "minimum": 0.07233029153400176, "maximum": 0.08588207127913414, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986254212781442, "standard_deviation": 0.0002778840620031733, "minimum": 0.9980098816977193, "maximum": 0.9989286283260113, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6993767539360942, "standard_deviation": 0.053574197253690836, "minimum": 0.6219510072232094, "maximum": 0.7855200850661176, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38405591843070197, "standard_deviation": 0.16865601307584857, "minimum": 0.13044333402731964, "maximum": 0.6583505634877648, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037600568288171876, "standard_deviation": 0.0034874241901093122, "minimum": 0.03291656190683716, "maximum": 0.044904286778431286, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28240-0011", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.8479759693145752, "predicted_mos": 1.8479759693145752, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2f70b189e89c566c067807793cd5ce6a4a2030a86bbc96a33f41e400f74e3272", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9427208001725376, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1973313748156198, "全帯域_10ms_rms_envelope_nrmse": 0.07269462136998299, "全帯域_10ms_envelope_correlation": 0.9950130784264862, "全帯域_10ms_modulation_mae_db": 0.2806283775646296, "全帯域_40ms_log_rms_mae_db": 0.6802693768241425, "全帯域_40ms_rms_envelope_nrmse": 0.043457491261058474, "全帯域_40ms_envelope_correlation": 0.9981385574747237, "全帯域_40ms_modulation_mae_db": 0.12831793771863745, "300–1000 Hz_10ms_log_rms_mae_db": 1.7590703270979702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10004561839924679, "300–1000 Hz_10ms_envelope_correlation": 0.9925201324163961, "300–1000 Hz_10ms_modulation_mae_db": 0.42177936228597507, "300–1000 Hz_40ms_log_rms_mae_db": 1.2230027385597488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06335122059051093, "300–1000 Hz_40ms_envelope_correlation": 0.9968761280525622, "300–1000 Hz_40ms_modulation_mae_db": 0.1844944716071293, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6245437400359621, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12482489368594925, "1000–3000 Hz_10ms_envelope_correlation": 0.988397470725695, "1000–3000 Hz_10ms_modulation_mae_db": 1.2213818334979576, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0412902499693109, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0761575988853795, "1000–3000 Hz_40ms_envelope_correlation": 0.9957452554289289, "1000–3000 Hz_40ms_modulation_mae_db": 1.2012926756643854, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6851687734112906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20444351476401726, "3000–6000 Hz_10ms_envelope_correlation": 0.9655211489887378, "3000–6000 Hz_10ms_modulation_mae_db": 0.8312979729915186, "3000–6000 Hz_40ms_log_rms_mae_db": 1.261872563588652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12895524106159462, "3000–6000 Hz_40ms_envelope_correlation": 0.9856579329379647, "3000–6000 Hz_40ms_modulation_mae_db": 0.7993247606686906, "mr_stft_db": 5.6473473903124765, "band_mae_db": 0.17204650165442978, "rms_gain_error_db": -0.02975086253090796, "output_peak": 0.6080046910379034}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2727242319512055, "全帯域_10ms_rms_envelope_nrmse": 0.07986371160903465, "全帯域_10ms_envelope_correlation": 0.9940276572311317, "全帯域_10ms_modulation_mae_db": 0.769095412704999, "全帯域_40ms_log_rms_mae_db": 0.6273497717609646, "全帯域_40ms_rms_envelope_nrmse": 0.03707464605247392, "全帯域_40ms_envelope_correlation": 0.9987512283578637, "全帯域_40ms_modulation_mae_db": 0.6496851417494484, "300–1000 Hz_10ms_log_rms_mae_db": 1.7500851930846582, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11400644035872998, "300–1000 Hz_10ms_envelope_correlation": 0.9902518780555885, "300–1000 Hz_10ms_modulation_mae_db": 1.0628970886584823, "300–1000 Hz_40ms_log_rms_mae_db": 1.0992800006706858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051424772594925804, "300–1000 Hz_40ms_envelope_correlation": 0.9980714649687338, "300–1000 Hz_40ms_modulation_mae_db": 0.8407060297996537, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6992987255271625, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11947028970465749, "1000–3000 Hz_10ms_envelope_correlation": 0.9895110852388365, "1000–3000 Hz_10ms_modulation_mae_db": 0.5011018375350261, "1000–3000 Hz_40ms_log_rms_mae_db": 1.100617523706583, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0662596323514175, "1000–3000 Hz_40ms_envelope_correlation": 0.9970262104864275, "1000–3000 Hz_40ms_modulation_mae_db": 0.3837123409528842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6821565822447422, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1902304783178006, "3000–6000 Hz_10ms_envelope_correlation": 0.9702474820767046, "3000–6000 Hz_10ms_modulation_mae_db": 0.666188607549362, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2677879801094945, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12206213639661076, "3000–6000 Hz_40ms_envelope_correlation": 0.9871526068214833, "3000–6000 Hz_40ms_modulation_mae_db": 0.7375179750013975, "mr_stft_db": 5.666765120084919, "band_mae_db": 0.17297878941584766, "rms_gain_error_db": -0.08960861655077962, "output_peak": 0.6239490551524085}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2320180542811061, "全帯域_10ms_rms_envelope_nrmse": 0.08037159037464396, "全帯域_10ms_envelope_correlation": 0.9939678405047184, "全帯域_10ms_modulation_mae_db": 0.3995390322850469, "全帯域_40ms_log_rms_mae_db": 0.6754685319598525, "全帯域_40ms_rms_envelope_nrmse": 0.04493635927784847, "全帯域_40ms_envelope_correlation": 0.9981082697801162, "全帯域_40ms_modulation_mae_db": 0.2516918255867592, "300–1000 Hz_10ms_log_rms_mae_db": 1.731567122426124, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11571959140790954, "300–1000 Hz_10ms_envelope_correlation": 0.9900498305536933, "300–1000 Hz_10ms_modulation_mae_db": 0.35882979374056556, "300–1000 Hz_40ms_log_rms_mae_db": 1.0979592311858573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06711242180742483, "300–1000 Hz_40ms_envelope_correlation": 0.9967223252239966, "300–1000 Hz_40ms_modulation_mae_db": 0.31629574478382644, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6221006735349384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11642700468017998, "1000–3000 Hz_10ms_envelope_correlation": 0.9899674768202816, "1000–3000 Hz_10ms_modulation_mae_db": 0.3781202222463295, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0159786233112627, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06847880036766871, "1000–3000 Hz_40ms_envelope_correlation": 0.9966542005697403, "1000–3000 Hz_40ms_modulation_mae_db": 0.2291407335014786, "3000–6000 Hz_10ms_log_rms_mae_db": 1.64833272249891, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20240305111283988, "3000–6000 Hz_10ms_envelope_correlation": 0.9662314709804989, "3000–6000 Hz_10ms_modulation_mae_db": 0.34528857726768, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2599977979736092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12566015739789285, "3000–6000 Hz_40ms_envelope_correlation": 0.9862829666570141, "3000–6000 Hz_40ms_modulation_mae_db": 0.46844739061599155, "mr_stft_db": 5.650688868261416, "band_mae_db": 0.21691284385135137, "rms_gain_error_db": -0.10385906873520699, "output_peak": 0.6236641205471136}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2547083650562354, "全帯域_10ms_rms_envelope_nrmse": 0.07964934329528549, "全帯域_10ms_envelope_correlation": 0.9940536121358787, "全帯域_10ms_modulation_mae_db": 0.5028274506216125, "全帯域_40ms_log_rms_mae_db": 0.6554682024959525, "全帯域_40ms_rms_envelope_nrmse": 0.04039506570516809, "全帯域_40ms_envelope_correlation": 0.9983929759503706, "全帯域_40ms_modulation_mae_db": 0.4608137033273705, "300–1000 Hz_10ms_log_rms_mae_db": 1.7596610472632668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09893681804052411, "300–1000 Hz_10ms_envelope_correlation": 0.9926379178376686, "300–1000 Hz_10ms_modulation_mae_db": 0.09221373691956451, "300–1000 Hz_40ms_log_rms_mae_db": 1.1441820760456978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05359011346305966, "300–1000 Hz_40ms_envelope_correlation": 0.9977322386847736, "300–1000 Hz_40ms_modulation_mae_db": 0.0841583150778446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6500947815258884, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13643775454097098, "1000–3000 Hz_10ms_envelope_correlation": 0.9858863501760504, "1000–3000 Hz_10ms_modulation_mae_db": 0.5031431625601764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.080145267310828, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07647867720902665, "1000–3000 Hz_40ms_envelope_correlation": 0.9952929175949972, "1000–3000 Hz_40ms_modulation_mae_db": 0.5964893578161935, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7156291932650407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21526202586494134, "3000–6000 Hz_10ms_envelope_correlation": 0.9616602100021482, "3000–6000 Hz_10ms_modulation_mae_db": 0.3405208854504989, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3218298070777272, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13640560360404755, "3000–6000 Hz_40ms_envelope_correlation": 0.983749663843214, "3000–6000 Hz_40ms_modulation_mae_db": 0.6182147067567051, "mr_stft_db": 5.6388002941957085, "band_mae_db": 0.07677048260833501, "rms_gain_error_db": 0.016024312988903368, "output_peak": 0.6465789484610381}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2824507107763572, "全帯域_10ms_rms_envelope_nrmse": 0.07944815098777351, "全帯域_10ms_envelope_correlation": 0.9940434276902518, "全帯域_10ms_modulation_mae_db": 0.34748740662939176, "全帯域_40ms_log_rms_mae_db": 0.7368045803591735, "全帯域_40ms_rms_envelope_nrmse": 0.0422509280040832, "全帯域_40ms_envelope_correlation": 0.9982483740184268, "全帯域_40ms_modulation_mae_db": 0.39502702208168955, "300–1000 Hz_10ms_log_rms_mae_db": 1.7512503188832516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10833691469902829, "300–1000 Hz_10ms_envelope_correlation": 0.9911683886715676, "300–1000 Hz_10ms_modulation_mae_db": 0.5446213255351461, "300–1000 Hz_40ms_log_rms_mae_db": 1.0942931077632125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05788476447782571, "300–1000 Hz_40ms_envelope_correlation": 0.9973791344231789, "300–1000 Hz_40ms_modulation_mae_db": 0.6224818154848842, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6506992827485742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1158071796738533, "1000–3000 Hz_10ms_envelope_correlation": 0.9900129632201321, "1000–3000 Hz_10ms_modulation_mae_db": 0.3937731997033212, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0326813666880441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0613206543266654, "1000–3000 Hz_40ms_envelope_correlation": 0.9973295882572492, "1000–3000 Hz_40ms_modulation_mae_db": 0.16244129857536518, "3000–6000 Hz_10ms_log_rms_mae_db": 1.647870188684256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21022203979889165, "3000–6000 Hz_10ms_envelope_correlation": 0.9635559771892415, "3000–6000 Hz_10ms_modulation_mae_db": 0.7252610716539595, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2275605842904962, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13061714280499478, "3000–6000 Hz_40ms_envelope_correlation": 0.9855454700481515, "3000–6000 Hz_40ms_modulation_mae_db": 0.8486551979817635, "mr_stft_db": 5.650192526652945, "band_mae_db": 0.16904003313345917, "rms_gain_error_db": -0.05261111635689998, "output_peak": 0.6065247241158122}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2872798689384783, "全帯域_10ms_rms_envelope_nrmse": 0.07567677165286817, "全帯域_10ms_envelope_correlation": 0.9946176738938723, "全帯域_10ms_modulation_mae_db": 0.23419853094369722, "全帯域_40ms_log_rms_mae_db": 0.6542130798194855, "全帯域_40ms_rms_envelope_nrmse": 0.04051863979760242, "全帯域_40ms_envelope_correlation": 0.9984168607572192, "全帯域_40ms_modulation_mae_db": 0.16577175640206024, "300–1000 Hz_10ms_log_rms_mae_db": 1.8379957495573294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11248411659144598, "300–1000 Hz_10ms_envelope_correlation": 0.9905151073415106, "300–1000 Hz_10ms_modulation_mae_db": 0.6651498575091908, "300–1000 Hz_40ms_log_rms_mae_db": 1.2346337020847364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05937516235130043, "300–1000 Hz_40ms_envelope_correlation": 0.9972387271935826, "300–1000 Hz_40ms_modulation_mae_db": 0.38959811230521413, "1000–3000 Hz_10ms_log_rms_mae_db": 1.697758838117364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12671925383190488, "1000–3000 Hz_10ms_envelope_correlation": 0.9880418173917896, "1000–3000 Hz_10ms_modulation_mae_db": 0.5172718910421298, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0030259399062988, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06822287693348712, "1000–3000 Hz_40ms_envelope_correlation": 0.9967109266769054, "1000–3000 Hz_40ms_modulation_mae_db": 0.49035432440033394, "3000–6000 Hz_10ms_log_rms_mae_db": 1.699775692523941, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20156384182056486, "3000–6000 Hz_10ms_envelope_correlation": 0.9664520306802485, "3000–6000 Hz_10ms_modulation_mae_db": 0.6993314553720547, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2862855818268069, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12613383124804176, "3000–6000 Hz_40ms_envelope_correlation": 0.9863032970848439, "3000–6000 Hz_40ms_modulation_mae_db": 0.502386901448542, "mr_stft_db": 5.658860466801772, "band_mae_db": 0.17872337017622386, "rms_gain_error_db": -0.06769312330523865, "output_peak": 0.6100188972717162}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2639405530537708, "全帯域_10ms_rms_envelope_nrmse": 0.07269752620140128, "全帯域_10ms_envelope_correlation": 0.9950251030901056, "全帯域_10ms_modulation_mae_db": 0.24313014255646603, "全帯域_40ms_log_rms_mae_db": 0.6409874402709522, "全帯域_40ms_rms_envelope_nrmse": 0.03617169376428394, "全帯域_40ms_envelope_correlation": 0.9987311191027135, "全帯域_40ms_modulation_mae_db": 0.24353752485026878, "300–1000 Hz_10ms_log_rms_mae_db": 1.8081397706530669, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10642176474862838, "300–1000 Hz_10ms_envelope_correlation": 0.9914707368231463, "300–1000 Hz_10ms_modulation_mae_db": 0.4352193126070754, "300–1000 Hz_40ms_log_rms_mae_db": 1.198098286229234, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05732052647087746, "300–1000 Hz_40ms_envelope_correlation": 0.9974299747181391, "300–1000 Hz_40ms_modulation_mae_db": 0.3297354541318944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.65093523053108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12306160554593337, "1000–3000 Hz_10ms_envelope_correlation": 0.9886512857480225, "1000–3000 Hz_10ms_modulation_mae_db": 0.4529256688741549, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0574192298598686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07472986439533028, "1000–3000 Hz_40ms_envelope_correlation": 0.9958000045790433, "1000–3000 Hz_40ms_modulation_mae_db": 0.5822306331554326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7330265730877628, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21352058707692764, "3000–6000 Hz_10ms_envelope_correlation": 0.9622880795616392, "3000–6000 Hz_10ms_modulation_mae_db": 0.8771921314416679, "3000–6000 Hz_40ms_log_rms_mae_db": 1.311636487005471, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13378478880672634, "3000–6000 Hz_40ms_envelope_correlation": 0.9844964167792297, "3000–6000 Hz_40ms_modulation_mae_db": 1.2097113354745137, "mr_stft_db": 5.650136769236357, "band_mae_db": 0.160661464939074, "rms_gain_error_db": -0.05551905576583451, "output_peak": 0.5898319674106884}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2596574504531295, "全帯域_10ms_rms_envelope_nrmse": 0.07362823695569164, "全帯域_10ms_envelope_correlation": 0.9951704446091254, "全帯域_10ms_modulation_mae_db": 0.19932394963859032, "全帯域_40ms_log_rms_mae_db": 0.6582644880787764, "全帯域_40ms_rms_envelope_nrmse": 0.04141833756748639, "全帯域_40ms_envelope_correlation": 0.998675623824326, "全帯域_40ms_modulation_mae_db": 0.28310331271762884, "300–1000 Hz_10ms_log_rms_mae_db": 1.7433810935848464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11412020360818259, "300–1000 Hz_10ms_envelope_correlation": 0.9903917280649038, "300–1000 Hz_10ms_modulation_mae_db": 0.19614063767813142, "300–1000 Hz_40ms_log_rms_mae_db": 1.110725728828191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06482481535126927, "300–1000 Hz_40ms_envelope_correlation": 0.9970540186036692, "300–1000 Hz_40ms_modulation_mae_db": 0.01949840362037174, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6483238636820552, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12712408529495264, "1000–3000 Hz_10ms_envelope_correlation": 0.9883620145777088, "1000–3000 Hz_10ms_modulation_mae_db": 0.40503498189553494, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0373166551969333, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07108637818405834, "1000–3000 Hz_40ms_envelope_correlation": 0.9969089110546907, "1000–3000 Hz_40ms_modulation_mae_db": 0.48825293491854477, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7095619320560091, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20579158018239813, "3000–6000 Hz_10ms_envelope_correlation": 0.9651426822964682, "3000–6000 Hz_10ms_modulation_mae_db": 0.5535318870079974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2743666788442647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13252053878665324, "3000–6000 Hz_40ms_envelope_correlation": 0.9851978443025979, "3000–6000 Hz_40ms_modulation_mae_db": 0.7985244505763576, "mr_stft_db": 5.649737562127375, "band_mae_db": 0.1662630138151405, "rms_gain_error_db": -0.17548286746053793, "output_peak": 0.6092692922112611}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.724667393326483, "spectral_convergence": 0.2818451906158469, "spectral_convergence_status": "ok", "amplitude_floor": 1.4461531006911672e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.431968482554777, "spectral_convergence": 0.2601301939079789, "spectral_convergence_status": "ok", "amplitude_floor": 7.398993822544586e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.785406295056169, "spectral_convergence": 0.2826228631766563, "spectral_convergence_status": "ok", "amplitude_floor": 6.518159095373565e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005975557583734567, "output_power": 0.0006183777693202412, "delta_db": 0.14875432452051013, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005550914808547427, "output_power": 0.0005720776633917864, "delta_db": 0.13090429049963961, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010492685783192576, "output_power": 0.0010480352554090924, "delta_db": -0.00510775203302823, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007612304608209731, "output_power": 0.0007170652823280852, "delta_db": -0.2595746229719518, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00020798712521433106, "output_power": 0.00019351135192113885, "delta_db": -0.313300051272267, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.709473434650137e-06, "output_power": 9.066834650550932e-06, "delta_db": 0.1746379686291819, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17204650165442978, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.02975086253090796, "pitch": null, "output_rms_dbfs": -25.005694251775186, "output_peak": 0.6080046910379034, "reference_rms_dbfs": -24.975943389244275, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6473473903124765}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002040524003006053, "log_rms_mae_db": 1.1973313748156198, "rms_envelope_nrmse": 0.07269462136998299, "envelope_correlation": 0.9950130784264862, "rms_gain_error_db": -0.02975112858941644, "reference_mean_power": 0.0031824932955688964, "output_mean_power": 0.00316076628758543}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.3108768872499092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006941576008860869, "output_power": 0.0006892024774209438, "reference_active": true, "delta_db": -0.03111253556382973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.971136227750714e-05, "output_power": 6.33072536388824e-05, "reference_active": true, "delta_db": -0.4185009606430223}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2294065042747786e-05, "output_power": 2.440149458047106e-05, "reference_active": true, "delta_db": 0.3922716364870369}], "active_band_count": 3, "active_band_mae_db": 0.2806283775646296}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019294226027584624, "log_rms_mae_db": 0.6802693768241425, "rms_envelope_nrmse": 0.043457491261058474, "envelope_correlation": 0.9981385574747237, "rms_gain_error_db": -0.029740999508628622, "reference_mean_power": 0.0031984676882902833, "output_mean_power": 0.003176639031382924}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.195649150719305e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000632939746621434, "output_power": 0.0006277731916062727, "reference_active": true, "delta_db": -0.0355960318467509}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313010835721604e-05, "output_power": 3.0812923836865266e-05, "reference_active": true, "delta_db": -0.3148994585766571}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3632381122717001e-06, "output_power": 1.3740975064472319e-06, "reference_active": true, "delta_db": 0.03445832273250433}], "active_band_count": 3, "active_band_mae_db": 0.12831793771863745}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015457760673902765, "log_rms_mae_db": 1.7590703270979702, "rms_envelope_nrmse": 0.10004561839924679, "envelope_correlation": 0.9925201324163961, "rms_gain_error_db": 0.013863523136711391, "reference_mean_power": 0.0011955559481795221, "output_mean_power": 0.001199378492814863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2205605502187216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134156506493551, "output_power": 0.00039508823261401543, "reference_active": true, "delta_db": -0.19692818898977754}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.135405678849972e-05, "output_power": 3.916134370663786e-05, "reference_active": true, "delta_db": -0.23660534872046998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1249926033986765e-05, "output_power": 1.3624799561891861e-05, "reference_active": true, "delta_db": 0.8318045491476777}], "active_band_count": 3, "active_band_mae_db": 0.42177936228597507}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014615313182534672, "log_rms_mae_db": 1.2230027385597488, "rms_envelope_nrmse": 0.06335122059051093, "envelope_correlation": 0.9968761280525622, "rms_gain_error_db": 0.013861104316335452, "reference_mean_power": 0.0012015673286161978, "output_mean_power": 0.0012054084220488423}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.16593985496473e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003798908177653278, "output_power": 0.000361372622580625, "reference_active": true, "delta_db": -0.21703548945025058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8446225039409073e-05, "output_power": 1.7588031793673503e-05, "reference_active": true, "delta_db": -0.20690260602661495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.60382048073901e-07, "output_power": 7.834050863922014e-07, "reference_active": true, "delta_db": 0.1295453193445224}], "active_band_count": 3, "active_band_mae_db": 0.1844944716071293}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011311095186173674, "log_rms_mae_db": 1.6245437400359621, "rms_envelope_nrmse": 0.12482489368594925, "envelope_correlation": 0.988397470725695, "rms_gain_error_db": -0.22521431478643333, "reference_mean_power": 0.0008539373095994977, "output_mean_power": 0.0008107828567537387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.101259078601872e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000229953654454443, "output_power": 0.00023638948734534934, "reference_active": true, "delta_db": 0.11987842970892079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8372953922112745e-05, "output_power": 4.491236103496801e-05, "reference_active": true, "delta_db": -1.138457845839936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3078079511035874e-05, "output_power": 7.5155894987884576e-06, "reference_active": true, "delta_db": -2.4058092249450156}], "active_band_count": 3, "active_band_mae_db": 1.2213818334979576}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.65333606126804e-05, "log_rms_mae_db": 1.0412902499693109, "rms_envelope_nrmse": 0.0761575988853795, "envelope_correlation": 0.9957452554289289, "rms_gain_error_db": -0.22521612378447634, "reference_mean_power": 0.0008582264011938477, "output_mean_power": 0.0008148548560158084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.585136403392408e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214290189767478, "output_power": 0.00021802948347321845, "reference_active": true, "delta_db": 0.11886720656103118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7723754282185457e-05, "output_power": 2.1206966210808522e-05, "reference_active": true, "delta_db": -1.1637349658855858}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.120050210892624e-07, "output_power": 5.344039307810893e-07, "reference_active": true, "delta_db": -2.3212758545465393}], "active_band_count": 3, "active_band_mae_db": 1.2012926756643854}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.9261609894090753e-05, "log_rms_mae_db": 1.6851687734112906, "rms_envelope_nrmse": 0.20444351476401726, "envelope_correlation": 0.9655211489887378, "rms_gain_error_db": -0.1926201976654, "reference_mean_power": 3.945170478466558e-05, "output_mean_power": 3.7740161552034365e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2228241150412126e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1970652723962295e-05, "output_power": 1.0548573656506398e-05, "reference_active": true, "delta_db": -0.549240921041037}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3243801130743035e-06, "output_power": 2.755555600964559e-06, "reference_active": true, "delta_db": -0.8150149708963337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0798067724676823e-06, "output_power": 8.324961702767918e-07, "reference_active": true, "delta_db": -1.1296380270371853}], "active_band_count": 3, "active_band_mae_db": 0.8312979729915186}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5144247671385576e-05, "log_rms_mae_db": 1.261872563588652, "rms_envelope_nrmse": 0.12895524106159462, "envelope_correlation": 0.9856579329379647, "rms_gain_error_db": -0.1926220761058767, "reference_mean_power": 3.9649231901027684e-05, "output_mean_power": 3.792910289445899e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.976126261669901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1283346332236661e-05, "output_power": 9.58774370029598e-06, "reference_active": true, "delta_db": -0.7072150267744732}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7283331039583737e-06, "output_power": 1.3530021289470005e-06, "reference_active": true, "delta_db": -1.0632896840826778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.035315524247478e-08, "output_power": 6.088868889957988e-08, "reference_active": true, "delta_db": -0.6274695711489209}], "active_band_count": 3, "active_band_mae_db": 0.7993247606686906}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8007084530632556, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "292b29977640558f092703aba4f224a5425804cb56de7db95cd72fef9217e308", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9886038079873145, "standard_deviation": 0.0013348176338251626, "minimum": 0.9858863501760504, "maximum": 0.9900129632201321, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6554693919628782, "standard_deviation": 0.029038232499678747, "minimum": 1.6221006735349384, "maximum": 1.6992987255271625, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5465940996693288, "standard_deviation": 0.27792759619496954, "minimum": 0.3781202222463295, "maximum": 1.2213818334979576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12373400836980025, "standard_deviation": 0.006738353424242821, "minimum": 0.1158071796738533, "maximum": 0.13643775454097098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9964335018309978, "standard_deviation": 0.000725248245137264, "minimum": 0.9952929175949972, "maximum": 0.9973295882572492, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0460593569936414, "standard_deviation": 0.032298529190265386, "minimum": 1.0030259399062988, "maximum": 1.100617523706583, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5167392873730772, "standard_deviation": 0.317738961054833, "minimum": 0.16244129857536518, "maximum": 1.2012926756643854, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07034181033162919, "standard_deviation": 0.005309022628981495, "minimum": 0.0613206543266654, "maximum": 0.07647867720902665, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9651373852219609, "standard_deviation": 0.0027228097489214896, "minimum": 0.9616602100021482, "maximum": 0.9702474820767046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.690190207221494, "standard_deviation": 0.030664144783609293, "minimum": 1.647870188684256, "maximum": 1.7330265730877628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6298265735918422, "standard_deviation": 0.20266857013911338, "minimum": 0.3405208854504989, "maximum": 0.8771921314416679, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20542963986729768, "standard_deviation": 0.007940213451660197, "minimum": 0.1902304783178006, "maximum": 0.21526202586494134, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9855482748093125, "standard_deviation": 0.0010796793828079523, "minimum": 0.983749663843214, "maximum": 0.9871526068214833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2764171850895654, "standard_deviation": 0.030111384181794012, "minimum": 1.2275605842904962, "maximum": 1.3218298070777272, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7478478398154952, "standard_deviation": 0.23419318732074587, "minimum": 0.46844739061599155, "maximum": 1.2097113354745137, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12951743001332022, "standard_deviation": 0.00475259431377383, "minimum": 0.12206213639661076, "maximum": 0.13640560360404755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9911257149705593, "standard_deviation": 0.0010131080421029095, "minimum": 0.9900498305536933, "maximum": 0.9926379178376686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.767643827818814, "standard_deviation": 0.036240872174526456, "minimum": 1.731567122426124, "maximum": 1.8379957495573294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4721063893667664, "standard_deviation": 0.29970424078731966, "minimum": 0.09221373691956451, "maximum": 1.0628970886584823, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10875893348171195, "standard_deviation": 0.00650981196531271, "minimum": 0.09893681804052411, "maximum": 0.11571959140790954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973130014835794, "standard_deviation": 0.00044423749006996654, "minimum": 0.9967223252239966, "maximum": 0.9980714649687338, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1502718589209204, "standard_deviation": 0.05948043094092348, "minimum": 1.0942931077632125, "maximum": 1.2346337020847364, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3483710433513524, "standard_deviation": 0.2739727163608403, "minimum": 0.01949840362037174, "maximum": 0.8407060297996537, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05936047463839926, "standard_deviation": 0.005457491801469519, "minimum": 0.051424772594925804, "maximum": 0.06711242180742483, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16417456244923267, "standard_deviation": 0.03930162743248451, "minimum": 0.07677048260833501, "maximum": 0.21691284385135137, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.651566124709121, "standard_deviation": 0.008220393493211178, "minimum": 5.6388002941957085, "maximum": 5.666765120084919, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6147302120259928, "standard_deviation": 0.016748761932188466, "minimum": 0.5898319674106884, "maximum": 0.6465789484610381, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06981254971456279, "standard_deviation": 0.05630891428798578, "minimum": -0.17548286746053793, "maximum": 0.016024312988903368, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944898546976964, "standard_deviation": 0.0005231121496638898, "minimum": 0.9939678405047184, "maximum": 0.9951704446091254, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2562638261657377, "standard_deviation": 0.029404694354834972, "minimum": 1.1973313748156198, "maximum": 1.2872798689384783, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37202878786805416, "standard_deviation": 0.1889528755752165, "minimum": 0.19932394963859032, "maximum": 0.769095412704999, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07675374405583521, "standard_deviation": 0.0034280998128926577, "minimum": 0.07269462136998299, "maximum": 0.08037159037464396, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.99843287615822, "standard_deviation": 0.0002610708042336846, "minimum": 0.9981082697801162, "maximum": 0.9987512283578637, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6661031839461624, "standard_deviation": 0.03324289035626157, "minimum": 0.6273497717609646, "maximum": 0.7368045803591735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.32224352805423284, "standard_deviation": 0.17176055795636616, "minimum": 0.12831793771863745, "maximum": 0.6496851417494484, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04077789517875062, "standard_deviation": 0.002980842841163359, "minimum": 0.03617169376428394, "maximum": 0.04493635927784847, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28240-0011", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.4572489261627197, "predicted_mos": 2.4572489261627197, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4894c5e34775939bb1eb712f89249bdbdbee01a9395be291498b855150c34cf2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7883788999170065, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.198595038013074, "全帯域_10ms_rms_envelope_nrmse": 0.07572798650596906, "全帯域_10ms_envelope_correlation": 0.9946928031092455, "全帯域_10ms_modulation_mae_db": 0.21317747578098045, "全帯域_40ms_log_rms_mae_db": 0.5893088348351507, "全帯域_40ms_rms_envelope_nrmse": 0.041709569288942705, "全帯域_40ms_envelope_correlation": 0.9984367909173051, "全帯域_40ms_modulation_mae_db": 0.20933387538417658, "300–1000 Hz_10ms_log_rms_mae_db": 1.683769855041734, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11672198555610504, "300–1000 Hz_10ms_envelope_correlation": 0.989766171913109, "300–1000 Hz_10ms_modulation_mae_db": 0.3913639052525455, "300–1000 Hz_40ms_log_rms_mae_db": 1.031746844774823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06464062418014042, "300–1000 Hz_40ms_envelope_correlation": 0.9968275096759669, "300–1000 Hz_40ms_modulation_mae_db": 0.36105394316113065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5908943486888203, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12787752795865248, "1000–3000 Hz_10ms_envelope_correlation": 0.9884520906870443, "1000–3000 Hz_10ms_modulation_mae_db": 0.6169732901543085, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9687279556532157, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07956565389673995, "1000–3000 Hz_40ms_envelope_correlation": 0.9961280552306571, "1000–3000 Hz_40ms_modulation_mae_db": 0.7982725224709123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5981972646618374, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20061666981516119, "3000–6000 Hz_10ms_envelope_correlation": 0.9667650627852369, "3000–6000 Hz_10ms_modulation_mae_db": 0.3714706045127211, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1435725341666818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12998406294196152, "3000–6000 Hz_40ms_envelope_correlation": 0.9850355111051189, "3000–6000 Hz_40ms_modulation_mae_db": 0.36266463484514216, "mr_stft_db": 5.624306487252556, "band_mae_db": 0.20542173130086758, "rms_gain_error_db": -0.11881398613374194, "output_peak": 0.6092621847260133}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2852452309072346, "全帯域_10ms_rms_envelope_nrmse": 0.08037633930707598, "全帯域_10ms_envelope_correlation": 0.9940224242184491, "全帯域_10ms_modulation_mae_db": 0.2749737096473485, "全帯域_40ms_log_rms_mae_db": 0.740102478358841, "全帯域_40ms_rms_envelope_nrmse": 0.04551510885066645, "全帯域_40ms_envelope_correlation": 0.9981412208208378, "全帯域_40ms_modulation_mae_db": 0.2470335472946998, "300–1000 Hz_10ms_log_rms_mae_db": 1.663892973739177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11020798864483532, "300–1000 Hz_10ms_envelope_correlation": 0.9911474463622774, "300–1000 Hz_10ms_modulation_mae_db": 0.4455253590822738, "300–1000 Hz_40ms_log_rms_mae_db": 1.0118618749246004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06879055564038698, "300–1000 Hz_40ms_envelope_correlation": 0.9967061504242788, "300–1000 Hz_40ms_modulation_mae_db": 0.53806909055603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6117662182743149, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12402378934774427, "1000–3000 Hz_10ms_envelope_correlation": 0.9885375455423236, "1000–3000 Hz_10ms_modulation_mae_db": 0.3396841658711842, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9590683139263103, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07328632321959828, "1000–3000 Hz_40ms_envelope_correlation": 0.9960658552697711, "1000–3000 Hz_40ms_modulation_mae_db": 0.5369669271691503, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5874792590261406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21343072465251528, "3000–6000 Hz_10ms_envelope_correlation": 0.9623065012042, "3000–6000 Hz_10ms_modulation_mae_db": 0.360713173718967, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1411153463936101, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13277513923527687, "3000–6000 Hz_40ms_envelope_correlation": 0.9844965692811333, "3000–6000 Hz_40ms_modulation_mae_db": 0.3560371417799457, "mr_stft_db": 5.6137378285257755, "band_mae_db": 0.17800440516445817, "rms_gain_error_db": -0.12966283513508203, "output_peak": 0.6067005119420465}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2667477716511046, "全帯域_10ms_rms_envelope_nrmse": 0.07325284985413459, "全帯域_10ms_envelope_correlation": 0.9949928840130632, "全帯域_10ms_modulation_mae_db": 0.26138859768323297, "全帯域_40ms_log_rms_mae_db": 0.663961827124292, "全帯域_40ms_rms_envelope_nrmse": 0.03798089843370327, "全帯域_40ms_envelope_correlation": 0.9986666893085262, "全帯域_40ms_modulation_mae_db": 0.30627624318114777, "300–1000 Hz_10ms_log_rms_mae_db": 1.6587696563312866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10237932197876543, "300–1000 Hz_10ms_envelope_correlation": 0.9921553232839663, "300–1000 Hz_10ms_modulation_mae_db": 0.6181301571861252, "300–1000 Hz_40ms_log_rms_mae_db": 1.0361133545972512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05139115073887833, "300–1000 Hz_40ms_envelope_correlation": 0.9980320571601785, "300–1000 Hz_40ms_modulation_mae_db": 0.5454868344921255, "1000–3000 Hz_10ms_log_rms_mae_db": 1.561950680591725, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1192910031964767, "1000–3000 Hz_10ms_envelope_correlation": 0.9894269127684712, "1000–3000 Hz_10ms_modulation_mae_db": 0.6749041507821282, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9403435022726735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06807928540879907, "1000–3000 Hz_40ms_envelope_correlation": 0.9966763469422331, "1000–3000 Hz_40ms_modulation_mae_db": 0.6915815809340975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5642893636033957, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20276032227131924, "3000–6000 Hz_10ms_envelope_correlation": 0.9660784085473688, "3000–6000 Hz_10ms_modulation_mae_db": 0.31311893256636736, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1205248147687572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12110127753941428, "3000–6000 Hz_40ms_envelope_correlation": 0.9870297200250413, "3000–6000 Hz_40ms_modulation_mae_db": 0.45850929682964675, "mr_stft_db": 5.5991114959369055, "band_mae_db": 0.16368998310917113, "rms_gain_error_db": -0.0943436084257439, "output_peak": 0.6270271710279606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2417716184953036, "全帯域_10ms_rms_envelope_nrmse": 0.07504571555167794, "全帯域_10ms_envelope_correlation": 0.9947025934926947, "全帯域_10ms_modulation_mae_db": 0.21179484219249178, "全帯域_40ms_log_rms_mae_db": 0.6533705013707297, "全帯域_40ms_rms_envelope_nrmse": 0.038747323012435664, "全帯域_40ms_envelope_correlation": 0.9985621296358543, "全帯域_40ms_modulation_mae_db": 0.21717932011088367, "300–1000 Hz_10ms_log_rms_mae_db": 1.7368067993577585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10479250760868887, "300–1000 Hz_10ms_envelope_correlation": 0.9917471780372308, "300–1000 Hz_10ms_modulation_mae_db": 0.3508049268349411, "300–1000 Hz_40ms_log_rms_mae_db": 1.0926514823658686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05600337125962462, "300–1000 Hz_40ms_envelope_correlation": 0.9975959228415642, "300–1000 Hz_40ms_modulation_mae_db": 0.3470881177819288, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5233415193757511, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11887491353797565, "1000–3000 Hz_10ms_envelope_correlation": 0.9897700538896491, "1000–3000 Hz_10ms_modulation_mae_db": 0.6386234505175066, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8976328849303252, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06640151243919105, "1000–3000 Hz_40ms_envelope_correlation": 0.997209393255765, "1000–3000 Hz_40ms_modulation_mae_db": 0.45965681288781485, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6239231016012055, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2195794313346728, "3000–6000 Hz_10ms_envelope_correlation": 0.960079279602206, "3000–6000 Hz_10ms_modulation_mae_db": 0.49669309247035726, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1868747622358142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14018659355047888, "3000–6000 Hz_40ms_envelope_correlation": 0.9831564932629283, "3000–6000 Hz_40ms_modulation_mae_db": 0.7004939563649603, "mr_stft_db": 5.6076395735690125, "band_mae_db": 0.19290315238000402, "rms_gain_error_db": -0.06788841179815953, "output_peak": 0.6081705613306129}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1795531707108344, "全帯域_10ms_rms_envelope_nrmse": 0.07547889744186492, "全帯域_10ms_envelope_correlation": 0.9946907586180997, "全帯域_10ms_modulation_mae_db": 0.3807603364125494, "全帯域_40ms_log_rms_mae_db": 0.6072621804791406, "全帯域_40ms_rms_envelope_nrmse": 0.03558900946344834, "全帯域_40ms_envelope_correlation": 0.9988635133036929, "全帯域_40ms_modulation_mae_db": 0.3729763527663019, "300–1000 Hz_10ms_log_rms_mae_db": 1.6803145706760427, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11264849788164628, "300–1000 Hz_10ms_envelope_correlation": 0.9905180261784676, "300–1000 Hz_10ms_modulation_mae_db": 0.616811044073656, "300–1000 Hz_40ms_log_rms_mae_db": 1.028624634504186, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05878833551102349, "300–1000 Hz_40ms_envelope_correlation": 0.9974639465545759, "300–1000 Hz_40ms_modulation_mae_db": 0.44043426297034166, "1000–3000 Hz_10ms_log_rms_mae_db": 1.570062391315616, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11568855262588758, "1000–3000 Hz_10ms_envelope_correlation": 0.990285619000641, "1000–3000 Hz_10ms_modulation_mae_db": 0.6070266789626011, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8431720440857011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05735272801672511, "1000–3000 Hz_40ms_envelope_correlation": 0.9980500596230258, "1000–3000 Hz_40ms_modulation_mae_db": 0.6731741939809698, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5882975511625392, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20363224201483915, "3000–6000 Hz_10ms_envelope_correlation": 0.965727340641747, "3000–6000 Hz_10ms_modulation_mae_db": 0.495537611535852, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1489635591615035, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12628582660705243, "3000–6000 Hz_40ms_envelope_correlation": 0.9862506956015875, "3000–6000 Hz_40ms_modulation_mae_db": 0.7043134670195479, "mr_stft_db": 5.596552194708871, "band_mae_db": 0.1683433154421848, "rms_gain_error_db": -0.10001992463078291, "output_peak": 0.6322214611595272}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3245005581359968, "全帯域_10ms_rms_envelope_nrmse": 0.07889481944712574, "全帯域_10ms_envelope_correlation": 0.9941769098184756, "全帯域_10ms_modulation_mae_db": 0.18823414933419758, "全帯域_40ms_log_rms_mae_db": 0.6788903698341006, "全帯域_40ms_rms_envelope_nrmse": 0.04155987856483944, "全帯域_40ms_envelope_correlation": 0.9983893729659077, "全帯域_40ms_modulation_mae_db": 0.2818612380803168, "300–1000 Hz_10ms_log_rms_mae_db": 1.6957424785615371, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10247032034798086, "300–1000 Hz_10ms_envelope_correlation": 0.9921551417791101, "300–1000 Hz_10ms_modulation_mae_db": 0.43588139953313704, "300–1000 Hz_40ms_log_rms_mae_db": 1.0322935957200108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05527294739233041, "300–1000 Hz_40ms_envelope_correlation": 0.9977019904277244, "300–1000 Hz_40ms_modulation_mae_db": 0.4910806069071358, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5534947846988971, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12788105092607135, "1000–3000 Hz_10ms_envelope_correlation": 0.988002552164608, "1000–3000 Hz_10ms_modulation_mae_db": 0.30739029069568763, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8587011173467093, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07263667658278039, "1000–3000 Hz_40ms_envelope_correlation": 0.9964537168739492, "1000–3000 Hz_40ms_modulation_mae_db": 0.5588666150220571, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5897292529663978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2063616231625531, "3000–6000 Hz_10ms_envelope_correlation": 0.9648290452716178, "3000–6000 Hz_10ms_modulation_mae_db": 0.6078517622363395, "3000–6000 Hz_40ms_log_rms_mae_db": 1.148202293349651, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12823514955329923, "3000–6000 Hz_40ms_envelope_correlation": 0.9859987125532861, "3000–6000 Hz_40ms_modulation_mae_db": 0.8993174823649784, "mr_stft_db": 5.6057400936585635, "band_mae_db": 0.16606407790893427, "rms_gain_error_db": -0.09578083323111537, "output_peak": 0.6513489563238647}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.304099795791562, "全帯域_10ms_rms_envelope_nrmse": 0.07458133395592856, "全帯域_10ms_envelope_correlation": 0.9947974668750217, "全帯域_10ms_modulation_mae_db": 0.3895058511509714, "全帯域_40ms_log_rms_mae_db": 0.6830116132334672, "全帯域_40ms_rms_envelope_nrmse": 0.03908510007547313, "全帯域_40ms_envelope_correlation": 0.9985696767004906, "全帯域_40ms_modulation_mae_db": 0.387828606453924, "300–1000 Hz_10ms_log_rms_mae_db": 1.732137347465006, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10880276423049612, "300–1000 Hz_10ms_envelope_correlation": 0.9911230013431243, "300–1000 Hz_10ms_modulation_mae_db": 0.5487101593468502, "300–1000 Hz_40ms_log_rms_mae_db": 1.0801247923320882, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06194375204753963, "300–1000 Hz_40ms_envelope_correlation": 0.9970837537385915, "300–1000 Hz_40ms_modulation_mae_db": 0.34345448967692355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5773029309499524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12929527251100661, "1000–3000 Hz_10ms_envelope_correlation": 0.9875273949516233, "1000–3000 Hz_10ms_modulation_mae_db": 0.394939215788548, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9617821826987727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07166428784674315, "1000–3000 Hz_40ms_envelope_correlation": 0.9963156130057111, "1000–3000 Hz_40ms_modulation_mae_db": 0.6236022278329104, "3000–6000 Hz_10ms_log_rms_mae_db": 1.62691356685963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1991919737684687, "3000–6000 Hz_10ms_envelope_correlation": 0.9673822945748832, "3000–6000 Hz_10ms_modulation_mae_db": 0.7423544044311893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.189931138580133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12710511631596236, "3000–6000 Hz_40ms_envelope_correlation": 0.9863875952415434, "3000–6000 Hz_40ms_modulation_mae_db": 0.8829692786637106, "mr_stft_db": 5.615867883645836, "band_mae_db": 0.16507456374091314, "rms_gain_error_db": -0.08968956509479183, "output_peak": 0.6126357660984186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2556035111307091, "全帯域_10ms_rms_envelope_nrmse": 0.07776097352131128, "全帯域_10ms_envelope_correlation": 0.9943959839335564, "全帯域_10ms_modulation_mae_db": 0.2958968379015618, "全帯域_40ms_log_rms_mae_db": 0.6522950407030461, "全帯域_40ms_rms_envelope_nrmse": 0.039663167512266014, "全帯域_40ms_envelope_correlation": 0.9986063923779014, "全帯域_40ms_modulation_mae_db": 0.3435932818525866, "300–1000 Hz_10ms_log_rms_mae_db": 1.6520285763804576, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10773076246072266, "300–1000 Hz_10ms_envelope_correlation": 0.9914015001964022, "300–1000 Hz_10ms_modulation_mae_db": 0.18579674788258202, "300–1000 Hz_40ms_log_rms_mae_db": 0.9976481425132073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06590149132673266, "300–1000 Hz_40ms_envelope_correlation": 0.9968294120698062, "300–1000 Hz_40ms_modulation_mae_db": 0.31347216919257215, "1000–3000 Hz_10ms_log_rms_mae_db": 1.595676966933476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1280509902979037, "1000–3000 Hz_10ms_envelope_correlation": 0.9879666771105752, "1000–3000 Hz_10ms_modulation_mae_db": 0.17933040731224925, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9940250635718387, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06646453903421526, "1000–3000 Hz_40ms_envelope_correlation": 0.9971669098991321, "1000–3000 Hz_40ms_modulation_mae_db": 0.35083592405012576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5531952444296382, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19109740555006385, "3000–6000 Hz_10ms_envelope_correlation": 0.9699122596674167, "3000–6000 Hz_10ms_modulation_mae_db": 0.5385792926256049, "3000–6000 Hz_40ms_log_rms_mae_db": 1.127451667041643, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11672201721398363, "3000–6000 Hz_40ms_envelope_correlation": 0.9881495330741408, "3000–6000 Hz_40ms_modulation_mae_db": 0.7030758233116527, "mr_stft_db": 5.6040080648541375, "band_mae_db": 0.1672635485542009, "rms_gain_error_db": -0.12082473346298618, "output_peak": 0.60152802726209}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.733395071855854, "spectral_convergence": 0.28493947993126795, "spectral_convergence_status": "ok", "amplitude_floor": 1.4461531006911672e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.392700646969951, "spectral_convergence": 0.26549559607597367, "spectral_convergence_status": "ok", "amplitude_floor": 7.398993822544586e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.746823742931861, "spectral_convergence": 0.2758825200727522, "spectral_convergence_status": "ok", "amplitude_floor": 6.518159095373565e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005975557583734567, "output_power": 0.0006226112308268815, "delta_db": 0.17838514483495188, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005550914808547427, "output_power": 0.0005589612339276585, "delta_db": 0.03017126825217801, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010492685783192576, "output_power": 0.001015098602947994, "delta_db": -0.14378437488933996, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007612304608209731, "output_power": 0.0006967576357701408, "delta_db": -0.3843442151475882, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00020798712521433106, "output_power": 0.00019160555859660963, "delta_db": -0.35628348088697603, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.709473434650137e-06, "output_power": 8.993900718902765e-06, "delta_db": 0.1395619037941714, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20542173130086758, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11881398613374194, "pitch": null, "output_rms_dbfs": -25.09475737537802, "output_peak": 0.6092621847260133, "reference_rms_dbfs": -24.975943389244275, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.624306487252556}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002040524003006053, "log_rms_mae_db": 1.198595038013074, "rms_envelope_nrmse": 0.07572798650596906, "envelope_correlation": 0.9946928031092455, "rms_gain_error_db": -0.1188130696180989, "reference_mean_power": 0.0031824932955688964, "output_mean_power": 0.0030966077077429167}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.3108768872499092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006941576008860869, "output_power": 0.0006896223609706666, "reference_active": true, "delta_db": -0.02846748436884118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.971136227750714e-05, "output_power": 6.400071587793186e-05, "reference_active": true, "delta_db": -0.371187378676492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2294065042747786e-05, "output_power": 2.356009407643297e-05, "reference_active": true, "delta_db": 0.2398775642976082}], "active_band_count": 3, "active_band_mae_db": 0.21317747578098045}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019294226027584624, "log_rms_mae_db": 0.5893088348351507, "rms_envelope_nrmse": 0.041709569288942705, "envelope_correlation": 0.9984367909173051, "rms_gain_error_db": -0.11880161975752357, "reference_mean_power": 0.0031984676882902833, "output_mean_power": 0.0031121592063009396}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.195649150719305e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000632939746621434, "output_power": 0.0006286859006520796, "reference_active": true, "delta_db": -0.029286481868714534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313010835721604e-05, "output_power": 3.025835890243966e-05, "reference_active": true, "delta_db": -0.3937748645686524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3632381122717001e-06, "output_power": 1.3004021871415537e-06, "reference_active": true, "delta_db": -0.20494027971516285}], "active_band_count": 3, "active_band_mae_db": 0.20933387538417658}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015457760673902765, "log_rms_mae_db": 1.683769855041734, "rms_envelope_nrmse": 0.11672198555610504, "envelope_correlation": 0.989766171913109, "rms_gain_error_db": -0.12468030416138959, "reference_mean_power": 0.0011955559481795221, "output_mean_power": 0.0011617210911893194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2205605502187216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134156506493551, "output_power": 0.0004155653590580408, "reference_active": true, "delta_db": 0.02252424558333569}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.135405678849972e-05, "output_power": 3.9495715343539226e-05, "reference_active": true, "delta_db": -0.19968135656817318}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1249926033986765e-05, "output_power": 1.4006779065926667e-05, "reference_active": true, "delta_db": 0.9518861136061275}], "active_band_count": 3, "active_band_mae_db": 0.3913639052525455}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014615313182534672, "log_rms_mae_db": 1.031746844774823, "rms_envelope_nrmse": 0.06464062418014042, "envelope_correlation": 0.9968275096759669, "rms_gain_error_db": -0.1246815646255088, "reference_mean_power": 0.0012015673286161978, "output_mean_power": 0.0011675620075592323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.16593985496473e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003798908177653278, "output_power": 0.0003815787251520029, "reference_active": true, "delta_db": 0.019253562310765235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8446225039409073e-05, "output_power": 1.705859795261193e-05, "reference_active": true, "delta_db": -0.3396416901623913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.60382048073901e-07, "output_power": 8.983767683405319e-07, "reference_active": true, "delta_db": 0.7242665770102354}], "active_band_count": 3, "active_band_mae_db": 0.36105394316113065}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011311095186173674, "log_rms_mae_db": 1.5908943486888203, "rms_envelope_nrmse": 0.12787752795865248, "envelope_correlation": 0.9884520906870443, "rms_gain_error_db": -0.3673977456577553, "reference_mean_power": 0.0008539373095994977, "output_mean_power": 0.0007846684923870352}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.101259078601872e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000229953654454443, "output_power": 0.0002206404687040555, "reference_active": true, "delta_db": -0.17955144431672898}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8372953922112745e-05, "output_power": 4.63198459113602e-05, "reference_active": true, "delta_db": -1.0044456504382007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3078079511035874e-05, "output_power": 1.1216353695755119e-05, "reference_active": true, "delta_db": -0.666922775707996}], "active_band_count": 3, "active_band_mae_db": 0.6169732901543085}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.65333606126804e-05, "log_rms_mae_db": 0.9687279556532157, "rms_envelope_nrmse": 0.07956565389673995, "envelope_correlation": 0.9961280552306571, "rms_gain_error_db": -0.3673980286720592, "reference_mean_power": 0.0008582264011938477, "output_mean_power": 0.0007886096144256389}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.585136403392408e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214290189767478, "output_power": 0.00020323050209079444, "reference_active": true, "delta_db": -0.1863961515422054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7723754282185457e-05, "output_power": 2.199708203077564e-05, "reference_active": true, "delta_db": -1.0048696663745444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.120050210892624e-07, "output_power": 6.912609986349094e-07, "reference_active": true, "delta_db": -1.2035517494959869}], "active_band_count": 3, "active_band_mae_db": 0.7982725224709123}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.9261609894090753e-05, "log_rms_mae_db": 1.5981972646618374, "rms_envelope_nrmse": 0.20061666981516119, "envelope_correlation": 0.9667650627852369, "rms_gain_error_db": -0.10911270351464628, "reference_mean_power": 3.945170478466558e-05, "output_mean_power": 3.8472862849396743e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2228241150412126e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1970652723962295e-05, "output_power": 1.0731818208094357e-05, "reference_active": true, "delta_db": -0.47444524513981284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3243801130743035e-06, "output_power": 3.6672677009567535e-06, "reference_active": true, "delta_db": 0.42631937846488716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0798067724676823e-06, "output_power": 1.1342551661296696e-06, "reference_active": true, "delta_db": 0.21364718993346338}], "active_band_count": 3, "active_band_mae_db": 0.3714706045127211}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5144247671385576e-05, "log_rms_mae_db": 1.1435725341666818, "rms_envelope_nrmse": 0.12998406294196152, "envelope_correlation": 0.9850355111051189, "rms_gain_error_db": -0.1091382366135412, "reference_mean_power": 3.9649231901027684e-05, "output_mean_power": 3.866526177000771e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.976126261669901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1283346332236661e-05, "output_power": 9.945472426790471e-06, "reference_active": true, "delta_db": -0.5481250095368796}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7283331039583737e-06, "output_power": 1.7840554706465106e-06, "reference_active": true, "delta_db": 0.13780905159102644}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.035315524247478e-08, "output_power": 7.717729699354418e-08, "reference_active": true, "delta_db": 0.4020598434075206}], "active_band_count": 3, "active_band_mae_db": 0.36266463484514216}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7980101361445809, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5bb0c60d5a8917eed1d27ca29fe8ca4aba58fe9d997e339e02cca6711f5d3f1e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.988746105764367, "standard_deviation": 0.0009752397656727104, "minimum": 0.9875273949516233, "maximum": 0.990285619000641, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5730612301035691, "standard_deviation": 0.027601481770177797, "minimum": 1.5233415193757511, "maximum": 1.6117662182743149, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46985895626052665, "standard_deviation": 0.18683221896978236, "minimum": 0.17933040731224925, "maximum": 0.6749041507821282, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1238728875502148, "standard_deviation": 0.005236422410738958, "minimum": 0.11568855262588758, "maximum": 0.12929527251100661, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967582437625306, "standard_deviation": 0.0006773095100062862, "minimum": 0.9960658552697711, "maximum": 0.9980500596230258, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9279316330606934, "standard_deviation": 0.05502876515938694, "minimum": 0.8431720440857011, "maximum": 0.9940250635718387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5866196005435047, "standard_deviation": 0.14123055961435574, "minimum": 0.35083592405012576, "maximum": 0.7982725224709123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06943137580559902, "standard_deviation": 0.006530457592202507, "minimum": 0.05735272801672511, "maximum": 0.07956565389673995, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9653850240368346, "standard_deviation": 0.0030454828602732133, "minimum": 0.960079279602206, "maximum": 0.9699122596674167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.591503075538848, "standard_deviation": 0.02559983852473785, "minimum": 1.5531952444296382, "maximum": 1.62691356685963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4907898592621748, "standard_deviation": 0.14245471242167984, "minimum": 0.31311893256636736, "maximum": 0.7423544044311893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20458379907119917, "standard_deviation": 0.008760471049222686, "minimum": 0.19109740555006385, "maximum": 0.2195794313346728, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9858131037680975, "standard_deviation": 0.0015533616765305129, "minimum": 0.9831564932629283, "maximum": 0.9881495330741408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1508295144622243, "standard_deviation": 0.025214815297708058, "minimum": 1.1205248147687572, "maximum": 1.189931138580133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6334226351474481, "standard_deviation": 0.2165299446271971, "minimum": 0.3560371417799457, "maximum": 0.8993174823649784, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12779939786967864, "standard_deviation": 0.007106429312072048, "minimum": 0.11672201721398363, "maximum": 0.14018659355047888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9912517236367109, "standard_deviation": 0.0008170665129353448, "minimum": 0.989766171913109, "maximum": 0.9921553232839663, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.687932782194125, "standard_deviation": 0.03206373988259896, "minimum": 1.6520285763804576, "maximum": 1.7368067993577585, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.44912796239901387, "standard_deviation": 0.14596078534484816, "minimum": 0.18579674788258202, "maximum": 0.6181301571861252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10821926858865508, "standard_deviation": 0.005004946894473695, "minimum": 0.10237932197876543, "maximum": 0.11672198555610504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972800928615858, "standard_deviation": 0.00048593096333519377, "minimum": 0.9967061504242788, "maximum": 0.9980320571601785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0388830902165045, "standard_deviation": 0.032137294226465364, "minimum": 0.9976481425132073, "maximum": 1.0926514823658686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.42251743934227354, "standard_deviation": 0.09343044901565094, "minimum": 0.31347216919257215, "maximum": 0.5454868344921255, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06034152851208206, "standard_deviation": 0.005980687770195606, "minimum": 0.05139115073887833, "maximum": 0.06879055564038698, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17584559720009174, "standard_deviation": 0.015398313334233898, "minimum": 0.16368998310917113, "maximum": 0.20542173130086758, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.608370452768957, "standard_deviation": 0.009187416038857659, "minimum": 5.596552194708871, "maximum": 5.624306487252556, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6186118299838168, "standard_deviation": 0.016876207159603487, "minimum": 0.60152802726209, "maximum": 0.6513489563238647, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10212798723905045, "standard_deviation": 0.02006613509805804, "minimum": -0.12966283513508203, "maximum": -0.06788841179815953, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945589780098257, "standard_deviation": 0.00032982358793765557, "minimum": 0.9940224242184491, "maximum": 0.9949928840130632, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2570145868544773, "standard_deviation": 0.04972859376841237, "minimum": 1.1795531707108344, "maximum": 1.3245005581359968, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27696647501291677, "standard_deviation": 0.07581281518943495, "minimum": 0.18823414933419758, "maximum": 0.3895058511509714, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.076389864448136, "standard_deviation": 0.002397410629806989, "minimum": 0.07325284985413459, "maximum": 0.08037633930707598, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985294732538146, "standard_deviation": 0.00021312919171950095, "minimum": 0.9981412208208378, "maximum": 0.9988635133036929, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.658525355742346, "standard_deviation": 0.04657240448100453, "minimum": 0.5893088348351507, "maximum": 0.740102478358841, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.29576030814050464, "standard_deviation": 0.0687120790957866, "minimum": 0.20933387538417658, "maximum": 0.387828606453924, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.039981256900221876, "standard_deviation": 0.0029718385308848444, "minimum": 0.03558900946344834, "maximum": 0.04551510885066645, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5105-28241-0004", "language": "en", "style": "normal", "speaker": "librispeech_5105", "source_utterance_id": "5105-28241-0004", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5105-28241-0004.wav", "source_file_sha256": "7b65775c5ac812f91f307cc5a6a14676f363b992860e544dddd97abf36367019", "decoded_pcm_sha256": "6704bc85da6844b255174407f722499ab1188299d4775a37e25cc238c7fb903e", "original_source_sha256": "70c67685f8e24f7a27f6de3b63d5b3ecbde28f017d1b7aac9304f6c6034fa1bc", "corpus_entry_id": "librispeech_5105-28241-0004"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.691318988800049, "predicted_mos": 3.691318988800049, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ad077fa222d7666f411b813a32a7331436a48171bff1e7e354b8f3bca4c35557", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.264204802999302, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5105-28241-0004", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ad077fa222d7666f411b813a32a7331436a48171bff1e7e354b8f3bca4c35557", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "eda03c395cdc9d3f1e9518a18019741a17681b4698ba9987763b4c461fbd7fcb"}, "rows": [{"case": "librispeech_5105-28241-0004", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.7222840785980225, "predicted_mos": 2.7222840785980225, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "93131142d605d47b07afbff55c0fa9a2d3a715a63bed1a5ae7e6bd898b8ce8a5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7389266998507082, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2614892651937146, "全帯域_10ms_rms_envelope_nrmse": 0.11585885170986612, "全帯域_10ms_envelope_correlation": 0.9878664906019984, "全帯域_10ms_modulation_mae_db": 0.48366792006047254, "全帯域_40ms_log_rms_mae_db": 0.6612474420938518, "全帯域_40ms_rms_envelope_nrmse": 0.059035446762091555, "全帯域_40ms_envelope_correlation": 0.9968313591793979, "全帯域_40ms_modulation_mae_db": 0.8902790381197123, "300–1000 Hz_10ms_log_rms_mae_db": 1.5414372791422868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15984110997916512, "300–1000 Hz_10ms_envelope_correlation": 0.9805339644700632, "300–1000 Hz_10ms_modulation_mae_db": 0.494020531861375, "300–1000 Hz_40ms_log_rms_mae_db": 0.8980423131375651, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0852160523648199, "300–1000 Hz_40ms_envelope_correlation": 0.9943891262494678, "300–1000 Hz_40ms_modulation_mae_db": 0.6286966684985661, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5585238136947472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13016988782192665, "1000–3000 Hz_10ms_envelope_correlation": 0.9874610182248521, "1000–3000 Hz_10ms_modulation_mae_db": 0.6954287681461727, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8600233668609636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07111865730594907, "1000–3000 Hz_40ms_envelope_correlation": 0.9961786967723468, "1000–3000 Hz_40ms_modulation_mae_db": 0.6065342002618365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6602938549754978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16972872814939008, "3000–6000 Hz_10ms_envelope_correlation": 0.9825384223357306, "3000–6000 Hz_10ms_modulation_mae_db": 0.30226169164105826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0583374383244164, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09983021021829139, "3000–6000 Hz_40ms_envelope_correlation": 0.993714259462713, "3000–6000 Hz_40ms_modulation_mae_db": 0.5605447705438781, "mr_stft_db": 5.865667766593063, "band_mae_db": 0.20031948671095098, "rms_gain_error_db": -0.15421319718135051, "output_peak": 0.6223492583482954}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1752394134255526, "全帯域_10ms_rms_envelope_nrmse": 0.10178982778533797, "全帯域_10ms_envelope_correlation": 0.990733668530303, "全帯域_10ms_modulation_mae_db": 0.28190869223096365, "全帯域_40ms_log_rms_mae_db": 0.6546966259539158, "全帯域_40ms_rms_envelope_nrmse": 0.05746767977078474, "全帯域_40ms_envelope_correlation": 0.9970570409171265, "全帯域_40ms_modulation_mae_db": 0.14253065395512296, "300–1000 Hz_10ms_log_rms_mae_db": 1.4915325180570103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13606097250174182, "300–1000 Hz_10ms_envelope_correlation": 0.9860114735786966, "300–1000 Hz_10ms_modulation_mae_db": 0.4724904023140218, "300–1000 Hz_40ms_log_rms_mae_db": 0.9258821151712597, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07584485813330204, "300–1000 Hz_40ms_envelope_correlation": 0.9956559411295887, "300–1000 Hz_40ms_modulation_mae_db": 0.5957037235687104, "1000–3000 Hz_10ms_log_rms_mae_db": 1.563253918081072, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12974401020821244, "1000–3000 Hz_10ms_envelope_correlation": 0.9876330257860141, "1000–3000 Hz_10ms_modulation_mae_db": 0.1707280811884855, "1000–3000 Hz_40ms_log_rms_mae_db": 0.879116223062892, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07392575244710305, "1000–3000 Hz_40ms_envelope_correlation": 0.9959866355602883, "1000–3000 Hz_40ms_modulation_mae_db": 0.2045368743484294, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6321860848578154, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19580897912362247, "3000–6000 Hz_10ms_envelope_correlation": 0.978204060325719, "3000–6000 Hz_10ms_modulation_mae_db": 1.3484972579907264, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0954419047384298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12831730123854526, "3000–6000 Hz_40ms_envelope_correlation": 0.9921180284112909, "3000–6000 Hz_40ms_modulation_mae_db": 1.6003492427297108, "mr_stft_db": 5.868989694901614, "band_mae_db": 0.35260305318887647, "rms_gain_error_db": -0.16363796104512923, "output_peak": 0.5944295078663901}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2284544064581788, "全帯域_10ms_rms_envelope_nrmse": 0.09922743058423379, "全帯域_10ms_envelope_correlation": 0.9911537772366291, "全帯域_10ms_modulation_mae_db": 0.513614855088527, "全帯域_40ms_log_rms_mae_db": 0.64322757925353, "全帯域_40ms_rms_envelope_nrmse": 0.04573631981223475, "全帯域_40ms_envelope_correlation": 0.9981883502367422, "全帯域_40ms_modulation_mae_db": 0.5376325354544577, "300–1000 Hz_10ms_log_rms_mae_db": 1.399569409374576, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12626111597501166, "300–1000 Hz_10ms_envelope_correlation": 0.9879397656186008, "300–1000 Hz_10ms_modulation_mae_db": 0.597665480657868, "300–1000 Hz_40ms_log_rms_mae_db": 0.7940198644721898, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059893627618588186, "300–1000 Hz_40ms_envelope_correlation": 0.9973219310292745, "300–1000 Hz_40ms_modulation_mae_db": 0.7034003586999344, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5067492531610391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11230847401413646, "1000–3000 Hz_10ms_envelope_correlation": 0.9908354741754052, "1000–3000 Hz_10ms_modulation_mae_db": 0.451003574558605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8785714810722867, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06700771455091897, "1000–3000 Hz_40ms_envelope_correlation": 0.9967858567004446, "1000–3000 Hz_40ms_modulation_mae_db": 0.45581086977516816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6239637360816526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1852737187163981, "3000–6000 Hz_10ms_envelope_correlation": 0.9791545215260522, "3000–6000 Hz_10ms_modulation_mae_db": 0.9866919224531722, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0770257135525052, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10081474676620751, "3000–6000 Hz_40ms_envelope_correlation": 0.9939182240085709, "3000–6000 Hz_40ms_modulation_mae_db": 1.0865274494281367, "mr_stft_db": 5.864754896831694, "band_mae_db": 0.21816211572426336, "rms_gain_error_db": -0.14077502577237513, "output_peak": 0.5819696827424531}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2218740491785636, "全帯域_10ms_rms_envelope_nrmse": 0.08620454139534968, "全帯域_10ms_envelope_correlation": 0.9934678737860531, "全帯域_10ms_modulation_mae_db": 0.199696876370441, "全帯域_40ms_log_rms_mae_db": 0.7195022258919828, "全帯域_40ms_rms_envelope_nrmse": 0.05044169565128392, "全帯域_40ms_envelope_correlation": 0.997832516585648, "全帯域_40ms_modulation_mae_db": 0.23988525044980094, "300–1000 Hz_10ms_log_rms_mae_db": 1.4694265003559577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12710522987228756, "300–1000 Hz_10ms_envelope_correlation": 0.988140957910336, "300–1000 Hz_10ms_modulation_mae_db": 0.6854684101607552, "300–1000 Hz_40ms_log_rms_mae_db": 0.8527668724592141, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07602594660023278, "300–1000 Hz_40ms_envelope_correlation": 0.9960226053951341, "300–1000 Hz_40ms_modulation_mae_db": 0.9996896170152284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4995103977614612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12620174593382563, "1000–3000 Hz_10ms_envelope_correlation": 0.9881869514837764, "1000–3000 Hz_10ms_modulation_mae_db": 0.2828404084198089, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8370417317067275, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07755891079540807, "1000–3000 Hz_40ms_envelope_correlation": 0.9952685462013442, "1000–3000 Hz_40ms_modulation_mae_db": 0.24690828714298577, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6144987396018398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1825015344828653, "3000–6000 Hz_10ms_envelope_correlation": 0.9797998630075118, "3000–6000 Hz_10ms_modulation_mae_db": 0.5646759064329833, "3000–6000 Hz_40ms_log_rms_mae_db": 1.100019982592352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11586298286446142, "3000–6000 Hz_40ms_envelope_correlation": 0.9918724250895904, "3000–6000 Hz_40ms_modulation_mae_db": 0.6417016796517773, "mr_stft_db": 5.839820325364404, "band_mae_db": 0.22463683748124727, "rms_gain_error_db": -0.1689733594572776, "output_peak": 0.5497129760613996}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1686030129770626, "全帯域_10ms_rms_envelope_nrmse": 0.10086338872522548, "全帯域_10ms_envelope_correlation": 0.9907626676806787, "全帯域_10ms_modulation_mae_db": 0.09846103206810108, "全帯域_40ms_log_rms_mae_db": 0.590440834796712, "全帯域_40ms_rms_envelope_nrmse": 0.05321167311258062, "全帯域_40ms_envelope_correlation": 0.9973002592082778, "全帯域_40ms_modulation_mae_db": 0.09473990624908175, "300–1000 Hz_10ms_log_rms_mae_db": 1.4717944401804648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12506908747370046, "300–1000 Hz_10ms_envelope_correlation": 0.9880967107903106, "300–1000 Hz_10ms_modulation_mae_db": 0.11628787007828001, "300–1000 Hz_40ms_log_rms_mae_db": 0.8709096801837548, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06385615868177641, "300–1000 Hz_40ms_envelope_correlation": 0.9967317536977489, "300–1000 Hz_40ms_modulation_mae_db": 0.04544119119159989, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5689348991154866, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12996962858731798, "1000–3000 Hz_10ms_envelope_correlation": 0.9877111152144362, "1000–3000 Hz_10ms_modulation_mae_db": 0.4619904691246322, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9024455675351142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07538808236692045, "1000–3000 Hz_40ms_envelope_correlation": 0.9959912384608691, "1000–3000 Hz_40ms_modulation_mae_db": 0.45962494270826443, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6619191933807, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18267295043182297, "3000–6000 Hz_10ms_envelope_correlation": 0.9797671215853951, "3000–6000 Hz_10ms_modulation_mae_db": 0.38198777378963195, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0609729845701072, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11053945709878672, "3000–6000 Hz_40ms_envelope_correlation": 0.9923044702154712, "3000–6000 Hz_40ms_modulation_mae_db": 0.3336363304125607, "mr_stft_db": 5.849890746907062, "band_mae_db": 0.21217735811378488, "rms_gain_error_db": -0.09204243762771486, "output_peak": 0.5469108498120919}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2195822828740526, "全帯域_10ms_rms_envelope_nrmse": 0.09753722718859896, "全帯域_10ms_envelope_correlation": 0.9913466960366172, "全帯域_10ms_modulation_mae_db": 0.3582830213261406, "全帯域_40ms_log_rms_mae_db": 0.6174420348810004, "全帯域_40ms_rms_envelope_nrmse": 0.046632264296312245, "全帯域_40ms_envelope_correlation": 0.997892145359258, "全帯域_40ms_modulation_mae_db": 0.3103940903081712, "300–1000 Hz_10ms_log_rms_mae_db": 1.4849811833157565, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11533807231418378, "300–1000 Hz_10ms_envelope_correlation": 0.9898928407151913, "300–1000 Hz_10ms_modulation_mae_db": 0.38269142602293016, "300–1000 Hz_40ms_log_rms_mae_db": 0.8726916079143221, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055719793981241196, "300–1000 Hz_40ms_envelope_correlation": 0.9975236005953421, "300–1000 Hz_40ms_modulation_mae_db": 0.6369896034862913, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5454191406128195, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13349799150460953, "1000–3000 Hz_10ms_envelope_correlation": 0.9869677388686103, "1000–3000 Hz_10ms_modulation_mae_db": 0.641350889067185, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8726235887177454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07688477593625624, "1000–3000 Hz_40ms_envelope_correlation": 0.9957579574416316, "1000–3000 Hz_40ms_modulation_mae_db": 0.5605957907700664, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5841180625571039, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19816811387212438, "3000–6000 Hz_10ms_envelope_correlation": 0.979122987768728, "3000–6000 Hz_10ms_modulation_mae_db": 0.602582917321414, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0151620744216783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12630069422124143, "3000–6000 Hz_40ms_envelope_correlation": 0.9921846718422024, "3000–6000 Hz_40ms_modulation_mae_db": 0.6866995334836622, "mr_stft_db": 5.850568215283228, "band_mae_db": 0.26437532558489796, "rms_gain_error_db": -0.054072574001494955, "output_peak": 0.6802094470741336}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1889295577419199, "全帯域_10ms_rms_envelope_nrmse": 0.09441233651594495, "全帯域_10ms_envelope_correlation": 0.9918948296964607, "全帯域_10ms_modulation_mae_db": 0.073578744353872, "全帯域_40ms_log_rms_mae_db": 0.5765568457902192, "全帯域_40ms_rms_envelope_nrmse": 0.049646404542862656, "全帯域_40ms_envelope_correlation": 0.9975840176550966, "全帯域_40ms_modulation_mae_db": 0.11499516909359991, "300–1000 Hz_10ms_log_rms_mae_db": 1.483465998579434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13466937904457574, "300–1000 Hz_10ms_envelope_correlation": 0.9863092639292786, "300–1000 Hz_10ms_modulation_mae_db": 0.3838359035093169, "300–1000 Hz_40ms_log_rms_mae_db": 0.9032492791584427, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07362855533826675, "300–1000 Hz_40ms_envelope_correlation": 0.9956325274778859, "300–1000 Hz_40ms_modulation_mae_db": 0.22779087814483254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5720421986980535, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13477637321638639, "1000–3000 Hz_10ms_envelope_correlation": 0.986619442903143, "1000–3000 Hz_10ms_modulation_mae_db": 0.2645192712347964, "1000–3000 Hz_40ms_log_rms_mae_db": 0.940070832613285, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0783689754997263, "1000–3000 Hz_40ms_envelope_correlation": 0.9954495063228516, "1000–3000 Hz_40ms_modulation_mae_db": 0.27418576492320584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6093022678723128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18643552464881394, "3000–6000 Hz_10ms_envelope_correlation": 0.9791573301533799, "3000–6000 Hz_10ms_modulation_mae_db": 0.1689109470521756, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0562710032839522, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10024815210371021, "3000–6000 Hz_40ms_envelope_correlation": 0.9937459569924985, "3000–6000 Hz_40ms_modulation_mae_db": 0.09026583334963438, "mr_stft_db": 5.853924195024238, "band_mae_db": 0.10715434529711358, "rms_gain_error_db": -0.028267124963916434, "output_peak": 0.6277302082852353}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1362756830942844, "全帯域_10ms_rms_envelope_nrmse": 0.08108558434596105, "全帯域_10ms_envelope_correlation": 0.9940807855700596, "全帯域_10ms_modulation_mae_db": 0.7893803043447374, "全帯域_40ms_log_rms_mae_db": 0.5407047495772624, "全帯域_40ms_rms_envelope_nrmse": 0.03781049383446409, "全帯域_40ms_envelope_correlation": 0.9987069080771266, "全帯域_40ms_modulation_mae_db": 0.9478666838116553, "300–1000 Hz_10ms_log_rms_mae_db": 1.4611917487814674, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10052584126710562, "300–1000 Hz_10ms_envelope_correlation": 0.9924802169896633, "300–1000 Hz_10ms_modulation_mae_db": 0.7775398040948582, "300–1000 Hz_40ms_log_rms_mae_db": 0.8087658229319953, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05356210686853766, "300–1000 Hz_40ms_envelope_correlation": 0.9979381199075936, "300–1000 Hz_40ms_modulation_mae_db": 0.9841167267358498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5040768338276185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14203515297815617, "1000–3000 Hz_10ms_envelope_correlation": 0.9851373695546038, "1000–3000 Hz_10ms_modulation_mae_db": 0.3980626632445378, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8432299062423022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08974116221677837, "1000–3000 Hz_40ms_envelope_correlation": 0.9937070834324381, "1000–3000 Hz_40ms_modulation_mae_db": 0.3976488469097326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6182955766499854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18151301461910144, "3000–6000 Hz_10ms_envelope_correlation": 0.9799923395220936, "3000–6000 Hz_10ms_modulation_mae_db": 1.3599278115966846, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0609304151031418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10532516860933219, "3000–6000 Hz_40ms_envelope_correlation": 0.993263484460282, "3000–6000 Hz_40ms_modulation_mae_db": 1.250932943945397, "mr_stft_db": 5.8648344884509624, "band_mae_db": 0.25670962201361686, "rms_gain_error_db": -0.0984599234036364, "output_peak": 0.6160553122625827}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.959656957671598, "spectral_convergence": 0.3381110919727833, "spectral_convergence_status": "ok", "amplitude_floor": 1.7451878151186278e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.558549318990131, "spectral_convergence": 0.32793321465086805, "spectral_convergence_status": "ok", "amplitude_floor": 9.668633122862388e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.07879702311746, "spectral_convergence": 0.3465318753818924, "spectral_convergence_status": "ok", "amplitude_floor": 6.131486475721294e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.299449699297122e-05, "output_power": 9.541412691613578e-05, "delta_db": 0.6056338351555963, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007475936848096536, "output_power": 0.0007161726907006438, "delta_db": -0.18647867730236065, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001248382978500818, "output_power": 0.0011797881758047785, "delta_db": -0.24543799275117317, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000559570006980701, "output_power": 0.0005448335528833119, "delta_db": -0.11590583004319951, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012219591005283435, "output_power": 0.00012344440766146807, "delta_db": 0.044147499583889756, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001408377861622848, "output_power": 0.00014069798612013538, "delta_db": -0.004313085429486586, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20031948671095098, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15421319718135051, "pitch": null, "output_rms_dbfs": -25.527875394461603, "output_peak": 0.6223492583482954, "reference_rms_dbfs": -25.37366219728025, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.865667766593063}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023508705421562093, "log_rms_mae_db": 1.2614892651937146, "rms_envelope_nrmse": 0.11585885170986612, "envelope_correlation": 0.9878664906019984, "rms_gain_error_db": -0.15425108233976825, "reference_mean_power": 0.0029034514839332195, "output_mean_power": 0.002802137648791421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1197992634528758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005958091728481294, "output_power": 0.000608230601154909, "reference_active": true, "delta_db": 0.08961081318826009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.320217055945433e-05, "output_power": 7.990211269899077e-05, "reference_active": true, "delta_db": -0.6686776394978233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9969137080619498e-05, "output_power": 1.7025025360468724e-05, "reference_active": true, "delta_db": -0.6927153074953343}], "active_band_count": 3, "active_band_mae_db": 0.48366792006047254}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019484731874036573, "log_rms_mae_db": 0.6612474420938518, "rms_envelope_nrmse": 0.059035446762091555, "envelope_correlation": 0.9968313591793979, "rms_gain_error_db": -0.15408641448686733, "reference_mean_power": 0.002910052669271509, "output_mean_power": 0.0028086149804681894}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0516863068209195e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005686142820757538, "output_power": 0.0005806428243248428, "reference_active": true, "delta_db": 0.09091299516754264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3606311869244326e-05, "output_power": 3.528760144962285e-05, "reference_active": true, "delta_db": -0.9192721678078826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3652354435919545e-06, "output_power": 9.314131380056076e-07, "reference_active": true, "delta_db": -1.6606519513837115}], "active_band_count": 3, "active_band_mae_db": 0.8902790381197123}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020607260287182458, "log_rms_mae_db": 1.5414372791422868, "rms_envelope_nrmse": 0.15984110997916512, "envelope_correlation": 0.9805339644700632, "rms_gain_error_db": -0.19340473823558163, "reference_mean_power": 0.001692101437139437, "output_mean_power": 0.0016184001890713895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.505279098297708e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003919860661962002, "output_power": 0.00040442672702957635, "reference_active": true, "delta_db": 0.1356921917640764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.427449102286661e-05, "output_power": 3.863761822896332e-05, "reference_active": true, "delta_db": -0.591432302747787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.188127359577471e-06, "output_power": 7.722062215582951e-06, "reference_active": true, "delta_db": -0.7549371010722614}], "active_band_count": 3, "active_band_mae_db": 0.494020531861375}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016462000863024626, "log_rms_mae_db": 0.8980423131375651, "rms_envelope_nrmse": 0.0852160523648199, "envelope_correlation": 0.9943891262494678, "rms_gain_error_db": -0.19328529909777592, "reference_mean_power": 0.001695759255771571, "output_mean_power": 0.0016219432934597637}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.197777273221207e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003745482187229297, "output_power": 0.0003858129641020604, "reference_active": true, "delta_db": 0.12869080739450356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.969988957093567e-05, "output_power": 1.6869863755236244e-05, "reference_active": true, "delta_db": -0.6735221655663748}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.294406999255848e-07, "output_power": 6.462456588551894e-07, "reference_active": true, "delta_db": -1.08387703253482}], "active_band_count": 3, "active_band_mae_db": 0.6286966684985661}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010781730421004015, "log_rms_mae_db": 1.5585238136947472, "rms_envelope_nrmse": 0.13016988782192665, "envelope_correlation": 0.9874610182248521, "rms_gain_error_db": -0.14279886575859077, "reference_mean_power": 0.0005546344869318713, "output_mean_power": 0.0005366943011023663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0245820082367704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001805463155146596, "output_power": 0.0001709592723662944, "reference_active": true, "delta_db": -0.23695969297563496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.244891302260328e-05, "output_power": 1.9806451463348988e-05, "reference_active": true, "delta_db": -0.5438864326209174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.880443381781143e-06, "output_power": 3.613393516636821e-06, "reference_active": true, "delta_db": -1.3054401788419656}], "active_band_count": 3, "active_band_mae_db": 0.6954287681461727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010092628732279422, "log_rms_mae_db": 0.8600233668609636, "rms_envelope_nrmse": 0.07111865730594907, "envelope_correlation": 0.9961786967723468, "rms_gain_error_db": -0.14322167916293554, "reference_mean_power": 0.0005569522649053947, "output_mean_power": 0.0005388846419043424}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.880183972268188e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017516791642605334, "output_power": 0.00016536062778227208, "reference_active": true, "delta_db": -0.25022451792053824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.918962387005773e-06, "output_power": 8.679785310905856e-06, "reference_active": true, "delta_db": -0.5795726010222298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2471992219002067e-07, "output_power": 2.5854038194337844e-07, "reference_active": true, "delta_db": -0.9898054818427414}], "active_band_count": 3, "active_band_mae_db": 0.6065342002618365}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001308702624055787, "log_rms_mae_db": 1.6602938549754978, "rms_envelope_nrmse": 0.16972872814939008, "envelope_correlation": 0.9825384223357306, "rms_gain_error_db": -0.03376097558149848, "reference_mean_power": 0.00016841175600957635, "output_mean_power": 0.00016710764030306634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6472123519121696e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.520693965195547e-05, "output_power": 9.605212057960563e-05, "reference_active": true, "delta_db": 0.03838352044160854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1413816053513955e-05, "output_power": 1.8993244278681134e-05, "reference_active": true, "delta_db": -0.5209491384056517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.37748433230818e-06, "output_power": 4.964024452082697e-06, "reference_active": true, "delta_db": -0.3474524160759146}], "active_band_count": 3, "active_band_mae_db": 0.30226169164105826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.575610534879353e-05, "log_rms_mae_db": 1.0583374383244164, "rms_envelope_nrmse": 0.09983021021829139, "envelope_correlation": 0.993714259462713, "rms_gain_error_db": -0.03369682970044424, "reference_mean_power": 0.00016921252708073642, "output_mean_power": 0.00016790469046841494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5742070644083465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.764878819980257e-05, "output_power": 9.8333310561249e-05, "reference_active": true, "delta_db": 0.030338025269812424}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.033680531591513e-05, "output_power": 9.048956192738534e-06, "reference_active": true, "delta_db": -0.5778785108116085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.467040891819506e-07, "output_power": 2.707803756489823e-07, "reference_active": true, "delta_db": -1.0734177755502134}], "active_band_count": 3, "active_band_mae_db": 0.5605447705438781}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8010231936839901, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4e46135daec9f94c71983e58a27c8dd97ee8bf45bb9d2418288db0825abda977", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9875690170263551, "standard_deviation": 0.0016152697219736316, "minimum": 0.9851373695546038, "maximum": 0.9908354741754052, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5398138068690372, "standard_deviation": 0.0311961328388758, "minimum": 1.4995103977614612, "maximum": 1.5720421986980535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.42074051562302794, "standard_deviation": 0.18251497139008543, "minimum": 0.1707280811884855, "maximum": 0.6954287681461727, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12983790803307138, "standard_deviation": 0.008512147145262087, "minimum": 0.11230847401413646, "maximum": 0.14203515297815617, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9956406901115268, "standard_deviation": 0.000907984331934555, "minimum": 0.9937070834324381, "maximum": 0.9967858567004446, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8766403372264147, "standard_deviation": 0.033114005267287555, "minimum": 0.8370417317067275, "maximum": 0.940070832613285, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4007306971049611, "standard_deviation": 0.1476437581412963, "minimum": 0.2045368743484294, "maximum": 0.6065342002618365, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07624925388988256, "standard_deviation": 0.006617220451146037, "minimum": 0.06700771455091897, "maximum": 0.08974116221677837, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9797170807780763, "standard_deviation": 0.0012715043737864555, "minimum": 0.978204060325719, "maximum": 0.9825384223357306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6255721894971136, "standard_deviation": 0.02599932651953018, "minimum": 1.5841180625571039, "maximum": 1.6619191933807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7144420285347308, "standard_deviation": 0.4637984336584863, "minimum": 0.1689109470521756, "maximum": 1.3599278115966846, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18526282050551734, "standard_deviation": 0.008854737305671323, "minimum": 0.16972872814939008, "maximum": 0.19816811387212438, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9928901900603274, "standard_deviation": 0.0008519333282591368, "minimum": 0.9918724250895904, "maximum": 0.9939182240085709, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0655201895733228, "standard_deviation": 0.02654576658739454, "minimum": 1.0151620744216783, "maximum": 1.100019982592352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7813322229430946, "standard_deviation": 0.4985208611225857, "minimum": 0.09026583334963438, "maximum": 1.6003492427297108, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11090483914007201, "standard_deviation": 0.01155173428494465, "minimum": 0.09983021021829139, "maximum": 0.12831730123854526, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9874256492502675, "standard_deviation": 0.0034559167192201395, "minimum": 0.9805339644700632, "maximum": 0.9924802169896633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.475424884723369, "standard_deviation": 0.03924077774293536, "minimum": 1.399569409374576, "maximum": 1.5414372791422868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48874997858742564, "standard_deviation": 0.20536930003483767, "minimum": 0.11628787007828001, "maximum": 0.7775398040948582, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12810885105347147, "standard_deviation": 0.01711636552324103, "minimum": 0.10052584126710562, "maximum": 0.15984110997916512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964019506852544, "standard_deviation": 0.0011902554136770961, "minimum": 0.9943891262494678, "maximum": 0.9979381199075936, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.865790944428593, "standard_deviation": 0.04582673251078928, "minimum": 0.7940198644721898, "maximum": 0.9258821151712597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6027285959176266, "standard_deviation": 0.3308462768548939, "minimum": 0.04544119119159989, "maximum": 0.9996896170152284, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06796838744834562, "standard_deviation": 0.011314741518123407, "minimum": 0.05356210686853766, "maximum": 0.0852160523648199, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2295172680143439, "standard_deviation": 0.06905819973623435, "minimum": 0.10715434529711358, "maximum": 0.35260305318887647, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.857306291169533, "standard_deviation": 0.010255042052403966, "minimum": 5.839820325364404, "maximum": 5.868989694901614, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6024209053065728, "standard_deviation": 0.04410419701329949, "minimum": 0.5469108498120919, "maximum": 0.6802094470741336, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11255520043161191, "standard_deviation": 0.05273701853505141, "minimum": -0.1689733594572776, "maximum": -0.028267124963916434, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.99141334864235, "standard_deviation": 0.0018938053130594051, "minimum": 0.9878664906019984, "maximum": 0.9940807855700596, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.200055958867916, "standard_deviation": 0.040061913616542905, "minimum": 1.1362756830942844, "maximum": 1.2614892651937146, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3498239307304069, "standard_deviation": 0.24043661596351912, "minimum": 0.073578744353872, "maximum": 0.7893803043447374, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09712239853131475, "standard_deviation": 0.010523180356629507, "minimum": 0.08108558434596105, "maximum": 0.11585885170986612, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976740746523342, "standard_deviation": 0.0006144255374173393, "minimum": 0.9968313591793979, "maximum": 0.9987069080771266, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6254772922798093, "standard_deviation": 0.05626562856147724, "minimum": 0.5407047495772624, "maximum": 0.7195022258919828, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40979041593020027, "standard_deviation": 0.3449793430706307, "minimum": 0.09473990624908175, "maximum": 0.9478666838116553, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04999774722282682, "standard_deviation": 0.006825845220299055, "minimum": 0.03781049383446409, "maximum": 0.059035446762091555, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0004", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.4031307697296143, "predicted_mos": 2.4031307697296143, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2751e0fcfdcf606ee605cffc3e1f537b361535f53dd3b2f42d0fa612f37d64a9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.610440700314939, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2585275637939037, "全帯域_10ms_rms_envelope_nrmse": 0.11321056424140981, "全帯域_10ms_envelope_correlation": 0.9884495676921626, "全帯域_10ms_modulation_mae_db": 0.48903228462521087, "全帯域_40ms_log_rms_mae_db": 0.6756653975705004, "全帯域_40ms_rms_envelope_nrmse": 0.06205744599961545, "全帯域_40ms_envelope_correlation": 0.9964923044273631, "全帯域_40ms_modulation_mae_db": 0.9125464119722873, "300–1000 Hz_10ms_log_rms_mae_db": 1.5556212573819646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15438827825993584, "300–1000 Hz_10ms_envelope_correlation": 0.9818785890799269, "300–1000 Hz_10ms_modulation_mae_db": 0.59181537888206, "300–1000 Hz_40ms_log_rms_mae_db": 0.8873384980987579, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08608981252201313, "300–1000 Hz_40ms_envelope_correlation": 0.9942886349462075, "300–1000 Hz_40ms_modulation_mae_db": 0.7920320556417854, "1000–3000 Hz_10ms_log_rms_mae_db": 1.595688136081446, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12956471419728258, "1000–3000 Hz_10ms_envelope_correlation": 0.9876188843656669, "1000–3000 Hz_10ms_modulation_mae_db": 0.8340356438796143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8785122895827319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06857302697808691, "1000–3000 Hz_40ms_envelope_correlation": 0.9965308851478535, "1000–3000 Hz_40ms_modulation_mae_db": 0.7552174063339324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6868297469190734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.177743271218923, "3000–6000 Hz_10ms_envelope_correlation": 0.9811358046543316, "3000–6000 Hz_10ms_modulation_mae_db": 0.13144784780400062, "3000–6000 Hz_40ms_log_rms_mae_db": 1.099583397244189, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1036782108622776, "3000–6000 Hz_40ms_envelope_correlation": 0.9933576542094051, "3000–6000 Hz_40ms_modulation_mae_db": 0.3867060048296315, "mr_stft_db": 5.840933633966947, "band_mae_db": 0.21879563682421374, "rms_gain_error_db": -0.16217883379198522, "output_peak": 0.640236872036343}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.178957516397373, "全帯域_10ms_rms_envelope_nrmse": 0.10268780088689512, "全帯域_10ms_envelope_correlation": 0.9905906043593484, "全帯域_10ms_modulation_mae_db": 0.3485675126880023, "全帯域_40ms_log_rms_mae_db": 0.6537221844960277, "全帯域_40ms_rms_envelope_nrmse": 0.05827131507588454, "全帯域_40ms_envelope_correlation": 0.9969996113311972, "全帯域_40ms_modulation_mae_db": 0.180390994547542, "300–1000 Hz_10ms_log_rms_mae_db": 1.5206732895514348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1412216382337994, "300–1000 Hz_10ms_envelope_correlation": 0.9848666396359735, "300–1000 Hz_10ms_modulation_mae_db": 0.48098266800528716, "300–1000 Hz_40ms_log_rms_mae_db": 0.9155860842985742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07543243245738503, "300–1000 Hz_40ms_envelope_correlation": 0.9956539676891599, "300–1000 Hz_40ms_modulation_mae_db": 0.5985340025068525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.607818185638267, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1298481549010128, "1000–3000 Hz_10ms_envelope_correlation": 0.9877144914675331, "1000–3000 Hz_10ms_modulation_mae_db": 0.20044377878687594, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8929068147936172, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07272020677596348, "1000–3000 Hz_40ms_envelope_correlation": 0.9962613165990294, "1000–3000 Hz_40ms_modulation_mae_db": 0.2772933416194986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6451123931002758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19352954292270247, "3000–6000 Hz_10ms_envelope_correlation": 0.9793215698771665, "3000–6000 Hz_10ms_modulation_mae_db": 1.538545270657324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1239574340317813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12913678383668947, "3000–6000 Hz_40ms_envelope_correlation": 0.9926527360365189, "3000–6000 Hz_40ms_modulation_mae_db": 1.861600804980669, "mr_stft_db": 5.841186175502926, "band_mae_db": 0.3503284305536836, "rms_gain_error_db": -0.17377995081340913, "output_peak": 0.5701219203152583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2252045557665496, "全帯域_10ms_rms_envelope_nrmse": 0.1047568116653956, "全帯域_10ms_envelope_correlation": 0.990081747630739, "全帯域_10ms_modulation_mae_db": 0.484290260287137, "全帯域_40ms_log_rms_mae_db": 0.6520423415389225, "全帯域_40ms_rms_envelope_nrmse": 0.048342994527895186, "全帯域_40ms_envelope_correlation": 0.9978991185021402, "全帯域_40ms_modulation_mae_db": 0.49051833419429797, "300–1000 Hz_10ms_log_rms_mae_db": 1.448589237127045, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1340833089808423, "300–1000 Hz_10ms_envelope_correlation": 0.9863150000219997, "300–1000 Hz_10ms_modulation_mae_db": 0.6092695186011782, "300–1000 Hz_40ms_log_rms_mae_db": 0.7948089897261931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06297063736712619, "300–1000 Hz_40ms_envelope_correlation": 0.9969058903280138, "300–1000 Hz_40ms_modulation_mae_db": 0.7459685210544057, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5236787163993561, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10714077729007355, "1000–3000 Hz_10ms_envelope_correlation": 0.9917109142930355, "1000–3000 Hz_10ms_modulation_mae_db": 0.6471951964285539, "1000–3000 Hz_40ms_log_rms_mae_db": 0.863374950192704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05997406240004168, "1000–3000 Hz_40ms_envelope_correlation": 0.9975243248659726, "1000–3000 Hz_40ms_modulation_mae_db": 0.6799855759169455, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6225536656749082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18569769274972733, "3000–6000 Hz_10ms_envelope_correlation": 0.9789790094266323, "3000–6000 Hz_10ms_modulation_mae_db": 0.8894031907315384, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0857981167958102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10081655855013497, "3000–6000 Hz_40ms_envelope_correlation": 0.9937510858118384, "3000–6000 Hz_40ms_modulation_mae_db": 0.9821326835105072, "mr_stft_db": 5.839021349650198, "band_mae_db": 0.18935740334393159, "rms_gain_error_db": -0.1296211272258266, "output_peak": 0.60006236308875}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.211948246036442, "全帯域_10ms_rms_envelope_nrmse": 0.08711852401650437, "全帯域_10ms_envelope_correlation": 0.9933217366471769, "全帯域_10ms_modulation_mae_db": 0.19757918786112927, "全帯域_40ms_log_rms_mae_db": 0.7127311003154938, "全帯域_40ms_rms_envelope_nrmse": 0.04974860594125545, "全帯域_40ms_envelope_correlation": 0.9979040037958566, "全帯域_40ms_modulation_mae_db": 0.28015064172925414, "300–1000 Hz_10ms_log_rms_mae_db": 1.4775844209854716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12287607782507587, "300–1000 Hz_10ms_envelope_correlation": 0.9888795574764742, "300–1000 Hz_10ms_modulation_mae_db": 0.6520271939980816, "300–1000 Hz_40ms_log_rms_mae_db": 0.8367390606841446, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06975951271119071, "300–1000 Hz_40ms_envelope_correlation": 0.9966613033539977, "300–1000 Hz_40ms_modulation_mae_db": 0.9159872891935142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5079004480481806, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12148936800234504, "1000–3000 Hz_10ms_envelope_correlation": 0.9890498972174169, "1000–3000 Hz_10ms_modulation_mae_db": 0.264586557168808, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8317876218589877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07235860383528352, "1000–3000 Hz_40ms_envelope_correlation": 0.9959044063947653, "1000–3000 Hz_40ms_modulation_mae_db": 0.16621188275920298, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6028650834795117, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1789109618211475, "3000–6000 Hz_10ms_envelope_correlation": 0.9805138662842952, "3000–6000 Hz_10ms_modulation_mae_db": 0.49319414926964694, "3000–6000 Hz_40ms_log_rms_mae_db": 1.086622996243681, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10867699283508087, "3000–6000 Hz_40ms_envelope_correlation": 0.99271130004521, "3000–6000 Hz_40ms_modulation_mae_db": 0.7097411541398442, "mr_stft_db": 5.80949338050165, "band_mae_db": 0.17513811763147255, "rms_gain_error_db": -0.16893851962514878, "output_peak": 0.5504831298950705}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1823561564346847, "全帯域_10ms_rms_envelope_nrmse": 0.10632493103379385, "全帯域_10ms_envelope_correlation": 0.9897321819747029, "全帯域_10ms_modulation_mae_db": 0.07686446570490414, "全帯域_40ms_log_rms_mae_db": 0.6122987799386812, "全帯域_40ms_rms_envelope_nrmse": 0.05835256443679744, "全帯域_40ms_envelope_correlation": 0.9967489624238035, "全帯域_40ms_modulation_mae_db": 0.17636923124687906, "300–1000 Hz_10ms_log_rms_mae_db": 1.5276665920570869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1304584742497669, "300–1000 Hz_10ms_envelope_correlation": 0.9870366312228471, "300–1000 Hz_10ms_modulation_mae_db": 0.09545481317343114, "300–1000 Hz_40ms_log_rms_mae_db": 0.8965251988242791, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06616899013051722, "300–1000 Hz_40ms_envelope_correlation": 0.9964771582185882, "300–1000 Hz_40ms_modulation_mae_db": 0.06123182527425628, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5823358137369898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1355655393366981, "1000–3000 Hz_10ms_envelope_correlation": 0.9865694886350017, "1000–3000 Hz_10ms_modulation_mae_db": 0.40978196318193993, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8893418558464498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08126432980364759, "1000–3000 Hz_40ms_envelope_correlation": 0.9952325172730254, "1000–3000 Hz_40ms_modulation_mae_db": 0.4104013167721369, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6790389033161217, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1835911011671739, "3000–6000 Hz_10ms_envelope_correlation": 0.97946173960861, "3000–6000 Hz_10ms_modulation_mae_db": 0.30326617585718985, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0803149187757497, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11683482482075516, "3000–6000 Hz_40ms_envelope_correlation": 0.9914292133292766, "3000–6000 Hz_40ms_modulation_mae_db": 0.49643060098762987, "mr_stft_db": 5.828080732583863, "band_mae_db": 0.24066608334197911, "rms_gain_error_db": -0.10090213689468627, "output_peak": 0.5494096665386455}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2052709390070395, "全帯域_10ms_rms_envelope_nrmse": 0.10026944384700971, "全帯域_10ms_envelope_correlation": 0.9908841014231391, "全帯域_10ms_modulation_mae_db": 0.32872461332771435, "全帯域_40ms_log_rms_mae_db": 0.6029977531753172, "全帯域_40ms_rms_envelope_nrmse": 0.04750259588440862, "全帯域_40ms_envelope_correlation": 0.997789526920645, "全帯域_40ms_modulation_mae_db": 0.3934249702843588, "300–1000 Hz_10ms_log_rms_mae_db": 1.5058076022542388, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12091818440180968, "300–1000 Hz_10ms_envelope_correlation": 0.9889580125030083, "300–1000 Hz_10ms_modulation_mae_db": 0.20114701511040176, "300–1000 Hz_40ms_log_rms_mae_db": 0.8392585458907824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05910074285445972, "300–1000 Hz_40ms_envelope_correlation": 0.9971908604849216, "300–1000 Hz_40ms_modulation_mae_db": 0.45397615192313506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5860101328056389, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1411158772796594, "1000–3000 Hz_10ms_envelope_correlation": 0.9854120246868265, "1000–3000 Hz_10ms_modulation_mae_db": 0.650196078526334, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8897901060469795, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08100834845584617, "1000–3000 Hz_40ms_envelope_correlation": 0.9952869632840301, "1000–3000 Hz_40ms_modulation_mae_db": 0.6275186680679232, "3000–6000 Hz_10ms_log_rms_mae_db": 1.592243826561617, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19234429877504364, "3000–6000 Hz_10ms_envelope_correlation": 0.9820962832914526, "3000–6000 Hz_10ms_modulation_mae_db": 0.9305212077662549, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0426940749628024, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12096673543057487, "3000–6000 Hz_40ms_envelope_correlation": 0.9944813310298468, "3000–6000 Hz_40ms_modulation_mae_db": 0.9637790282996739, "mr_stft_db": 5.8292161831282, "band_mae_db": 0.2663444882996549, "rms_gain_error_db": -0.0025390076063477396, "output_peak": 0.7015880415012874}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1980553490517947, "全帯域_10ms_rms_envelope_nrmse": 0.09868429465374755, "全帯域_10ms_envelope_correlation": 0.9911466454185768, "全帯域_10ms_modulation_mae_db": 0.07345932297047268, "全帯域_40ms_log_rms_mae_db": 0.5923338948493705, "全帯域_40ms_rms_envelope_nrmse": 0.05354307247345415, "全帯域_40ms_envelope_correlation": 0.9971887838854712, "全帯域_40ms_modulation_mae_db": 0.15593636861237245, "300–1000 Hz_10ms_log_rms_mae_db": 1.5045489644229664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1402272854664545, "300–1000 Hz_10ms_envelope_correlation": 0.9852166129081248, "300–1000 Hz_10ms_modulation_mae_db": 0.3414026974775605, "300–1000 Hz_40ms_log_rms_mae_db": 0.8920951075884356, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07996320560302146, "300–1000 Hz_40ms_envelope_correlation": 0.9948842782322666, "300–1000 Hz_40ms_modulation_mae_db": 0.18424447121050322, "1000–3000 Hz_10ms_log_rms_mae_db": 1.599078994119733, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1309221982774752, "1000–3000 Hz_10ms_envelope_correlation": 0.9874929201458031, "1000–3000 Hz_10ms_modulation_mae_db": 0.19560891561715674, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9245559477409826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07707355540657411, "1000–3000 Hz_40ms_envelope_correlation": 0.9957454684292589, "1000–3000 Hz_40ms_modulation_mae_db": 0.20864465145199015, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6411283598606226, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1822085565003448, "3000–6000 Hz_10ms_envelope_correlation": 0.9804189337997289, "3000–6000 Hz_10ms_modulation_mae_db": 0.08164538255777458, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0866931344159805, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10157476702712416, "3000–6000 Hz_40ms_envelope_correlation": 0.9937663292209813, "3000–6000 Hz_40ms_modulation_mae_db": 0.16314967237462388, "mr_stft_db": 5.827738857114994, "band_mae_db": 0.1399344006662773, "rms_gain_error_db": -0.02815605433654644, "output_peak": 0.6286326913501146}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1478198412556953, "全帯域_10ms_rms_envelope_nrmse": 0.08229318764573172, "全帯域_10ms_envelope_correlation": 0.9939804282518991, "全帯域_10ms_modulation_mae_db": 0.7594931034490812, "全帯域_40ms_log_rms_mae_db": 0.5580103298014295, "全帯域_40ms_rms_envelope_nrmse": 0.04087956909117257, "全帯域_40ms_envelope_correlation": 0.9985742901264713, "全帯域_40ms_modulation_mae_db": 0.903168401910281, "300–1000 Hz_10ms_log_rms_mae_db": 1.49901498870882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10290591394617009, "300–1000 Hz_10ms_envelope_correlation": 0.9921140479188754, "300–1000 Hz_10ms_modulation_mae_db": 0.7511992010387215, "300–1000 Hz_40ms_log_rms_mae_db": 0.8161483440303899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054649766885234236, "300–1000 Hz_40ms_envelope_correlation": 0.9978490011803384, "300–1000 Hz_40ms_modulation_mae_db": 0.9653567883348669, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5414273480191782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12510664851306108, "1000–3000 Hz_10ms_envelope_correlation": 0.9884049152008114, "1000–3000 Hz_10ms_modulation_mae_db": 0.3969072963903273, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8555008722569291, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07515580812053385, "1000–3000 Hz_40ms_envelope_correlation": 0.9956183858553582, "1000–3000 Hz_40ms_modulation_mae_db": 0.39219153984688554, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6409203031672261, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18437753114873207, "3000–6000 Hz_10ms_envelope_correlation": 0.9793877519468378, "3000–6000 Hz_10ms_modulation_mae_db": 1.3684223896774776, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0843316594739068, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10786128921820817, "3000–6000 Hz_40ms_envelope_correlation": 0.9930668688918426, "3000–6000 Hz_40ms_modulation_mae_db": 1.28332149360395, "mr_stft_db": 5.839110329292595, "band_mae_db": 0.24515647427764967, "rms_gain_error_db": -0.13722749503529758, "output_peak": 0.6221047926465981}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9364146519908, "spectral_convergence": 0.33649335585157913, "spectral_convergence_status": "ok", "amplitude_floor": 1.7451878151186278e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.541491258975623, "spectral_convergence": 0.3259141473240945, "spectral_convergence_status": "ok", "amplitude_floor": 9.668633122862388e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.044894990934416, "spectral_convergence": 0.3407392684312954, "spectral_convergence_status": "ok", "amplitude_floor": 6.131486475721294e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.299449699297122e-05, "output_power": 9.187212631219508e-05, "delta_db": 0.4413447077780203, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007475936848096536, "output_power": 0.000712918674429992, "delta_db": -0.20625633165217877, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001248382978500818, "output_power": 0.0011844966425506478, "delta_db": -0.22814004443771144, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000559570006980701, "output_power": 0.0005409766026140443, "delta_db": -0.14675946049773808, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012219591005283435, "output_power": 0.00011869564179706416, "delta_db": -0.12621897080180028, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001408377861622848, "output_power": 0.00014625968048066556, "delta_db": 0.1640543057778337, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21879563682421374, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16217883379198522, "pitch": null, "output_rms_dbfs": -25.535841031072238, "output_peak": 0.640236872036343, "reference_rms_dbfs": -25.37366219728025, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.840933633966947}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023508705421562093, "log_rms_mae_db": 1.2585275637939037, "rms_envelope_nrmse": 0.11321056424140981, "envelope_correlation": 0.9884495676921626, "rms_gain_error_db": -0.1622189661008328, "reference_mean_power": 0.0029034514839332195, "output_mean_power": 0.0027970013555510586}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1197992634528758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005958091728481294, "output_power": 0.0006080163767676258, "reference_active": true, "delta_db": 0.08808091881750943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.320217055945433e-05, "output_power": 8.039929973432033e-05, "reference_active": true, "delta_db": -0.6417376050927691}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9969137080619498e-05, "output_power": 1.6851224529172786e-05, "reference_active": true, "delta_db": -0.737278329965354}], "active_band_count": 3, "active_band_mae_db": 0.48903228462521087}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019484731874036573, "log_rms_mae_db": 0.6756653975705004, "rms_envelope_nrmse": 0.06205744599961545, "envelope_correlation": 0.9964923044273631, "rms_gain_error_db": -0.16206356173824787, "reference_mean_power": 0.002910052669271509, "output_mean_power": 0.0028034608345555664}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0516863068209195e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005686142820757538, "output_power": 0.0005808235343061507, "reference_active": true, "delta_db": 0.09226441353204272}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3606311869244326e-05, "output_power": 3.5272191366098305e-05, "reference_active": true, "delta_db": -0.9211691445646769}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3652354435919545e-06, "output_power": 9.178822824682102e-07, "reference_active": true, "delta_db": -1.7242056778201422}], "active_band_count": 3, "active_band_mae_db": 0.9125464119722873}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020607260287182458, "log_rms_mae_db": 1.5556212573819646, "rms_envelope_nrmse": 0.15438827825993584, "envelope_correlation": 0.9818785890799269, "rms_gain_error_db": -0.2002834218336269, "reference_mean_power": 0.001692101437139437, "output_mean_power": 0.0016158388737117153}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.505279098297708e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003919860661962002, "output_power": 0.00040480139323810336, "reference_active": true, "delta_db": 0.13971369013938578}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.427449102286661e-05, "output_power": 3.796213856462326e-05, "reference_active": true, "delta_db": -0.6680290721236146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.188127359577471e-06, "output_power": 7.352866298673831e-06, "reference_active": true, "delta_db": -0.9677033743831797}], "active_band_count": 3, "active_band_mae_db": 0.59181537888206}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016462000863024626, "log_rms_mae_db": 0.8873384980987579, "rms_envelope_nrmse": 0.08608981252201313, "envelope_correlation": 0.9942886349462075, "rms_gain_error_db": -0.20012892423159967, "reference_mean_power": 0.001695759255771571, "output_mean_power": 0.001619389443190218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.197777273221207e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003745482187229297, "output_power": 0.0003864273180892983, "reference_active": true, "delta_db": 0.13560084845383183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.969988957093567e-05, "output_power": 1.639607132133927e-05, "reference_active": true, "delta_db": -0.7972399291334613}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.294406999255848e-07, "output_power": 5.949216901254523e-07, "reference_active": true, "delta_db": -1.443255389338063}], "active_band_count": 3, "active_band_mae_db": 0.7920320556417854}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010781730421004015, "log_rms_mae_db": 1.595688136081446, "rms_envelope_nrmse": 0.12956471419728258, "envelope_correlation": 0.9876188843656669, "rms_gain_error_db": -0.16773387023928088, "reference_mean_power": 0.0005546344869318713, "output_mean_power": 0.0005336217015048028}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0245820082367704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001805463155146596, "output_power": 0.00017148035501164847, "reference_active": true, "delta_db": -0.2237425607240626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.244891302260328e-05, "output_power": 1.9127533964315863e-05, "reference_active": true, "delta_db": -0.695363354864605}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.880443381781143e-06, "output_power": 3.3896837178993864e-06, "reference_active": true, "delta_db": -1.583001016050175}], "active_band_count": 3, "active_band_mae_db": 0.8340356438796143}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010092628732279422, "log_rms_mae_db": 0.8785122895827319, "rms_envelope_nrmse": 0.06857302697808691, "envelope_correlation": 0.9965308851478535, "rms_gain_error_db": -0.16813187796192025, "reference_mean_power": 0.0005569522649053947, "output_mean_power": 0.0005358025628498805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.880183972268188e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017516791642605334, "output_power": 0.00016551668140383976, "reference_active": true, "delta_db": -0.24612794013345848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.918962387005773e-06, "output_power": 8.374102021087118e-06, "reference_active": true, "delta_db": -0.7352799584140535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2471992219002067e-07, "output_power": 2.4159308648398764e-07, "reference_active": true, "delta_db": -1.2842443204542853}], "active_band_count": 3, "active_band_mae_db": 0.7552174063339324}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001308702624055787, "log_rms_mae_db": 1.6868297469190734, "rms_envelope_nrmse": 0.177743271218923, "envelope_correlation": 0.9811358046543316, "rms_gain_error_db": 0.05862862812514886, "reference_mean_power": 0.00016841175600957635, "output_mean_power": 0.00017070068623564218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6472123519121696e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.520693965195547e-05, "output_power": 9.846113093369092e-05, "reference_active": true, "delta_db": 0.14596214466433366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1413816053513955e-05, "output_power": 2.0615544770429158e-05, "reference_active": true, "delta_db": -0.16499252110260837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.37748433230818e-06, "output_power": 5.275216229690568e-06, "reference_active": true, "delta_db": -0.08338887764505984}], "active_band_count": 3, "active_band_mae_db": 0.13144784780400062}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.575610534879353e-05, "log_rms_mae_db": 1.099583397244189, "rms_envelope_nrmse": 0.1036782108622776, "envelope_correlation": 0.9933576542094051, "rms_gain_error_db": 0.05888426944422598, "reference_mean_power": 0.00016921252708073642, "output_mean_power": 0.00017152243693448123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5742070644083465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.764878819980257e-05, "output_power": 0.00010131487124381106, "reference_active": true, "delta_db": 0.16006338684709476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.033680531591513e-05, "output_power": 9.705222032591017e-06, "reference_active": true, "delta_db": -0.2738086134762175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.467040891819506e-07, "output_power": 2.933150603487331e-07, "reference_active": true, "delta_db": -0.7262460141655822}], "active_band_count": 3, "active_band_mae_db": 0.3867060048296315}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8010677496214437, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "de21a99c16594bb0456458ec899bf4166f1b6e6179c6be2b52416c0a9c40a6d6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.987996692001512, "standard_deviation": 0.001861831319993926, "minimum": 0.9854120246868265, "maximum": 0.9917109142930355, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5679922218560987, "standard_deviation": 0.038041974516835275, "minimum": 1.5079004480481806, "maximum": 1.607818185638267, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.44984442874745123, "standard_deviation": 0.23672020835540447, "minimum": 0.19560891561715674, "maximum": 0.8340356438796143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12759415972470098, "standard_deviation": 0.010199553249082878, "minimum": 0.10714077729007355, "maximum": 0.1411158772796594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960130334811617, "standard_deviation": 0.000754920775114328, "minimum": 0.9952325172730254, "maximum": 0.9975243248659726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8782213072899228, "standard_deviation": 0.02806337481433711, "minimum": 0.8317876218589877, "maximum": 0.9245559477409826, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4396830478460644, "standard_deviation": 0.22368279681763653, "minimum": 0.16621188275920298, "maximum": 0.7552174063339324, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07351599272199716, "standard_deviation": 0.006982702917787557, "minimum": 0.05997406240004168, "maximum": 0.08126432980364759, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9801643698611318, "standard_deviation": 0.0010740538319537016, "minimum": 0.9789790094266323, "maximum": 0.9820962832914526, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6388365352599195, "standard_deviation": 0.03316305472740179, "minimum": 1.592243826561617, "maximum": 1.6868297469190734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7170557017901509, "standard_deviation": 0.5532260519558871, "minimum": 0.08164538255777458, "maximum": 1.538545270657324, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18480036953797435, "standard_deviation": 0.005691609141267827, "minimum": 0.177743271218923, "maximum": 0.19352954292270247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993152064821865, "standard_deviation": 0.0009230725851836397, "minimum": 0.9914292133292766, "maximum": 0.9944813310298468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0862494664929878, "standard_deviation": 0.022475998560854003, "minimum": 1.0426940749628024, "maximum": 1.1239574340317813, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8558576803408162, "standard_deviation": 0.5441011567868345, "minimum": 0.16314967237462388, "maximum": 1.861600804980669, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11119327032260566, "standard_deviation": 0.010165902789107346, "minimum": 0.10081655855013497, "maximum": 0.12913678383668947, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9869081363459037, "standard_deviation": 0.003114106992591871, "minimum": 0.9818785890799269, "maximum": 0.9921140479188754, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5049382940611287, "standard_deviation": 0.03227642172159652, "minimum": 1.448589237127045, "maximum": 1.5556212573819646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4654123107858403, "standard_deviation": 0.2316084781004871, "minimum": 0.09545481317343114, "maximum": 0.7511992010387215, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13088489517048182, "standard_deviation": 0.015598759497938649, "minimum": 0.10290591394617009, "maximum": 0.15438827825993584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962388868041867, "standard_deviation": 0.0012047170364962304, "minimum": 0.9942886349462075, "maximum": 0.9978490011803384, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8598124786426946, "standard_deviation": 0.043663037294313235, "minimum": 0.7948089897261931, "maximum": 0.9155860842985742, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5896663881424149, "standard_deviation": 0.3327614589949292, "minimum": 0.06123182527425628, "maximum": 0.9653567883348669, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06926688756636847, "standard_deviation": 0.010703331239423654, "minimum": 0.054649766885234236, "maximum": 0.08608981252201313, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22821512936735783, "standard_deviation": 0.06438372432726519, "minimum": 0.1399344006662773, "maximum": 0.3503284305536836, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.831847580217672, "standard_deviation": 0.010768216915998038, "minimum": 5.80949338050165, "maximum": 5.841186175502926, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6078299346715084, "standard_deviation": 0.05163212137402971, "minimum": 0.5494096665386455, "maximum": 0.7015880415012874, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11291789066615596, "standard_deviation": 0.06508405750623313, "minimum": -0.17377995081340913, "maximum": -0.0025390076063477396, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9910233766747181, "standard_deviation": 0.0018293964241742494, "minimum": 0.9884495676921626, "maximum": 0.9939804282518991, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2010175209679352, "standard_deviation": 0.03317410330547374, "minimum": 1.1478198412556953, "maximum": 1.2585275637939037, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34475134386420647, "standard_deviation": 0.2328680279337395, "minimum": 0.07345932297047268, "maximum": 0.7594931034490812, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09941819474881097, "standard_deviation": 0.010162498427280174, "minimum": 0.08229318764573172, "maximum": 0.11321056424140981, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974495751766186, "standard_deviation": 0.0007037502148900258, "minimum": 0.9964923044273631, "maximum": 0.9985742901264713, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6324752227107178, "standard_deviation": 0.0501042615103497, "minimum": 0.5580103298014295, "maximum": 0.7127311003154938, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4365631693121591, "standard_deviation": 0.312956843898288, "minimum": 0.15593636861237245, "maximum": 0.9125464119722873, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05233727042881042, "standard_deviation": 0.007016723301907185, "minimum": 0.04087956909117257, "maximum": 0.06205744599961545, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0004", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.490093231201172, "predicted_mos": 2.490093231201172, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "211bceae686e81d782abb5a4e9bbdba8c69c259690315d76e49e7d4c6dd1a3ed", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.008099500089884, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2223306421972153, "全帯域_10ms_rms_envelope_nrmse": 0.10277757334543927, "全帯域_10ms_envelope_correlation": 0.9905428003236215, "全帯域_10ms_modulation_mae_db": 0.4210037268896893, "全帯域_40ms_log_rms_mae_db": 0.6312315109481456, "全帯域_40ms_rms_envelope_nrmse": 0.055513233715646926, "全帯域_40ms_envelope_correlation": 0.9972768267238361, "全帯域_40ms_modulation_mae_db": 0.5540241864159418, "300–1000 Hz_10ms_log_rms_mae_db": 1.8778978289719073, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13264738760032635, "300–1000 Hz_10ms_envelope_correlation": 0.9867799266356423, "300–1000 Hz_10ms_modulation_mae_db": 0.3927105110772773, "300–1000 Hz_40ms_log_rms_mae_db": 1.1182459584779547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07486052178027074, "300–1000 Hz_40ms_envelope_correlation": 0.9958378503392655, "300–1000 Hz_40ms_modulation_mae_db": 0.5322190037495719, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7942850289245758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14460957524005835, "1000–3000 Hz_10ms_envelope_correlation": 0.9845860707828177, "1000–3000 Hz_10ms_modulation_mae_db": 0.7380584993783045, "1000–3000 Hz_40ms_log_rms_mae_db": 1.032720523366981, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07261164416090037, "1000–3000 Hz_40ms_envelope_correlation": 0.996227008116526, "1000–3000 Hz_40ms_modulation_mae_db": 0.8720374823305924, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0343690473458516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21053219007522064, "3000–6000 Hz_10ms_envelope_correlation": 0.9729850748737193, "3000–6000 Hz_10ms_modulation_mae_db": 0.9183136762597798, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3034842376485545, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12543308908067805, "3000–6000 Hz_40ms_envelope_correlation": 0.9903645659882989, "3000–6000 Hz_40ms_modulation_mae_db": 0.9870692791845169, "mr_stft_db": 6.169919799454978, "band_mae_db": 0.2871980374276486, "rms_gain_error_db": -0.15869471688473494, "output_peak": 0.5344383265667301}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1666128694464093, "全帯域_10ms_rms_envelope_nrmse": 0.10512787690993539, "全帯域_10ms_envelope_correlation": 0.9899521250684141, "全帯域_10ms_modulation_mae_db": 0.03422215602239753, "全帯域_40ms_log_rms_mae_db": 0.604322582868054, "全帯域_40ms_rms_envelope_nrmse": 0.062117109352948355, "全帯域_40ms_envelope_correlation": 0.9962845364231346, "全帯域_40ms_modulation_mae_db": 0.05160247416716898, "300–1000 Hz_10ms_log_rms_mae_db": 1.7974980791628987, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14014626439677214, "300–1000 Hz_10ms_envelope_correlation": 0.9852509356528589, "300–1000 Hz_10ms_modulation_mae_db": 0.34407920162553013, "300–1000 Hz_40ms_log_rms_mae_db": 1.0822812397614916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08662661254071108, "300–1000 Hz_40ms_envelope_correlation": 0.9943702088087807, "300–1000 Hz_40ms_modulation_mae_db": 0.3936451916984722, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7752955759200877, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13147158607681655, "1000–3000 Hz_10ms_envelope_correlation": 0.9871963515039129, "1000–3000 Hz_10ms_modulation_mae_db": 0.3305244769292756, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0425633523023106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07513787597726566, "1000–3000 Hz_40ms_envelope_correlation": 0.9955772296589901, "1000–3000 Hz_40ms_modulation_mae_db": 0.21336340866687256, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0057465746015013, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21817975870302617, "3000–6000 Hz_10ms_envelope_correlation": 0.9751126770725748, "3000–6000 Hz_10ms_modulation_mae_db": 0.38031477967584665, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2823607056563826, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13860972246544148, "3000–6000 Hz_40ms_envelope_correlation": 0.9909318489869199, "3000–6000 Hz_40ms_modulation_mae_db": 0.6202951176064653, "mr_stft_db": 6.149759087085745, "band_mae_db": 0.306354026399104, "rms_gain_error_db": -0.09117327130080924, "output_peak": 0.5984877297219866}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1974652961561627, "全帯域_10ms_rms_envelope_nrmse": 0.08943895084931365, "全帯域_10ms_envelope_correlation": 0.9931260721934458, "全帯域_10ms_modulation_mae_db": 0.450517820234467, "全帯域_40ms_log_rms_mae_db": 0.5596980361493843, "全帯域_40ms_rms_envelope_nrmse": 0.046185696756988026, "全帯域_40ms_envelope_correlation": 0.9984831393605628, "全帯域_40ms_modulation_mae_db": 0.6135480078957891, "300–1000 Hz_10ms_log_rms_mae_db": 1.8836824485301837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1223661063549324, "300–1000 Hz_10ms_envelope_correlation": 0.9892555755855506, "300–1000 Hz_10ms_modulation_mae_db": 0.9750770339904605, "300–1000 Hz_40ms_log_rms_mae_db": 1.1644833419227236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07205379442936652, "300–1000 Hz_40ms_envelope_correlation": 0.9967383712529472, "300–1000 Hz_40ms_modulation_mae_db": 1.2051271804300108, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7964686963212542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13026099291698975, "1000–3000 Hz_10ms_envelope_correlation": 0.9876805234975128, "1000–3000 Hz_10ms_modulation_mae_db": 0.3449096524137601, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0168068099428857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07465383007603224, "1000–3000 Hz_40ms_envelope_correlation": 0.9961277791204767, "1000–3000 Hz_40ms_modulation_mae_db": 0.45590685010795456, "3000–6000 Hz_10ms_log_rms_mae_db": 1.975741477598263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18244262055570815, "3000–6000 Hz_10ms_envelope_correlation": 0.979907351928904, "3000–6000 Hz_10ms_modulation_mae_db": 0.3919946717159002, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2374125690051518, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1016988229842917, "3000–6000 Hz_40ms_envelope_correlation": 0.9935743046677048, "3000–6000 Hz_40ms_modulation_mae_db": 0.5137427498514192, "mr_stft_db": 6.18440697285678, "band_mae_db": 0.364903815612186, "rms_gain_error_db": -0.2094158341520634, "output_peak": 0.5957637699686652}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2207979692771653, "全帯域_10ms_rms_envelope_nrmse": 0.10265311229957641, "全帯域_10ms_envelope_correlation": 0.9904719941470672, "全帯域_10ms_modulation_mae_db": 0.7630665628998187, "全帯域_40ms_log_rms_mae_db": 0.5769774904573131, "全帯域_40ms_rms_envelope_nrmse": 0.06045478509153379, "全帯域_40ms_envelope_correlation": 0.9965611910385597, "全帯域_40ms_modulation_mae_db": 0.9207044827756571, "300–1000 Hz_10ms_log_rms_mae_db": 1.852637801718908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14191380872871606, "300–1000 Hz_10ms_envelope_correlation": 0.9846809076398139, "300–1000 Hz_10ms_modulation_mae_db": 1.0902549190349187, "300–1000 Hz_40ms_log_rms_mae_db": 1.1362571683630323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08678131140599334, "300–1000 Hz_40ms_envelope_correlation": 0.9940540005764855, "300–1000 Hz_40ms_modulation_mae_db": 1.441794808657774, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8669510540261174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14037558493735577, "1000–3000 Hz_10ms_envelope_correlation": 0.9855708749901942, "1000–3000 Hz_10ms_modulation_mae_db": 0.9171993991242972, "1000–3000 Hz_40ms_log_rms_mae_db": 1.143185256177678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08229990764007382, "1000–3000 Hz_40ms_envelope_correlation": 0.9950979358740906, "1000–3000 Hz_40ms_modulation_mae_db": 1.01953130625139, "3000–6000 Hz_10ms_log_rms_mae_db": 2.088279822243379, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17540659963995311, "3000–6000 Hz_10ms_envelope_correlation": 0.9816585519753399, "3000–6000 Hz_10ms_modulation_mae_db": 0.5003799203669721, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3177602254557281, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09796928874134918, "3000–6000 Hz_40ms_envelope_correlation": 0.9940861812579247, "3000–6000 Hz_40ms_modulation_mae_db": 0.5964195748097303, "mr_stft_db": 6.178853110170988, "band_mae_db": 0.22863570847734682, "rms_gain_error_db": -0.12198554329848572, "output_peak": 0.5500257236472982}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1869101814875125, "全帯域_10ms_rms_envelope_nrmse": 0.0899224462414678, "全帯域_10ms_envelope_correlation": 0.9926969336042522, "全帯域_10ms_modulation_mae_db": 0.5842531140724064, "全帯域_40ms_log_rms_mae_db": 0.6189689106363528, "全帯域_40ms_rms_envelope_nrmse": 0.047743522883412354, "全帯域_40ms_envelope_correlation": 0.9978770799419744, "全帯域_40ms_modulation_mae_db": 0.7573011670888828, "300–1000 Hz_10ms_log_rms_mae_db": 1.8870832343115915, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12014268195272193, "300–1000 Hz_10ms_envelope_correlation": 0.9890170435368167, "300–1000 Hz_10ms_modulation_mae_db": 0.8243540419586123, "300–1000 Hz_40ms_log_rms_mae_db": 1.1691933706451967, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06548595001621972, "300–1000 Hz_40ms_envelope_correlation": 0.9965408054558836, "300–1000 Hz_40ms_modulation_mae_db": 1.1258878469814746, "1000–3000 Hz_10ms_log_rms_mae_db": 1.821513558713551, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13568417781839084, "1000–3000 Hz_10ms_envelope_correlation": 0.9866383081981766, "1000–3000 Hz_10ms_modulation_mae_db": 0.6544576315050762, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0684495806650551, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08040586611415826, "1000–3000 Hz_40ms_envelope_correlation": 0.9954688133221041, "1000–3000 Hz_40ms_modulation_mae_db": 0.7313984821968474, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0532452307750657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18582481745630905, "3000–6000 Hz_10ms_envelope_correlation": 0.9800663235441097, "3000–6000 Hz_10ms_modulation_mae_db": 0.9179758376732864, "3000–6000 Hz_40ms_log_rms_mae_db": 1.301069067968164, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12032269699325933, "3000–6000 Hz_40ms_envelope_correlation": 0.9926150167856707, "3000–6000 Hz_40ms_modulation_mae_db": 1.0035187508896632, "mr_stft_db": 6.189759318121414, "band_mae_db": 0.2863486156048019, "rms_gain_error_db": -0.10175314744990094, "output_peak": 0.5641628029764587}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1335699876982548, "全帯域_10ms_rms_envelope_nrmse": 0.09842335738409119, "全帯域_10ms_envelope_correlation": 0.9914318377834009, "全帯域_10ms_modulation_mae_db": 0.6543561760714892, "全帯域_40ms_log_rms_mae_db": 0.6250715363892316, "全帯域_40ms_rms_envelope_nrmse": 0.053673280259395766, "全帯域_40ms_envelope_correlation": 0.9975626540625806, "全帯域_40ms_modulation_mae_db": 0.7525664438107778, "300–1000 Hz_10ms_log_rms_mae_db": 1.9281827187439533, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12031202171746229, "300–1000 Hz_10ms_envelope_correlation": 0.9892034164928806, "300–1000 Hz_10ms_modulation_mae_db": 0.8362524928347318, "300–1000 Hz_40ms_log_rms_mae_db": 1.1826425524867699, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0705771576108391, "300–1000 Hz_40ms_envelope_correlation": 0.9963354264526105, "300–1000 Hz_40ms_modulation_mae_db": 1.2524063896057938, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8894606756249095, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13851869952086962, "1000–3000 Hz_10ms_envelope_correlation": 0.9859505853519949, "1000–3000 Hz_10ms_modulation_mae_db": 0.8584969721666403, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0914527708768804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0777517739956101, "1000–3000 Hz_40ms_envelope_correlation": 0.9956676518053758, "1000–3000 Hz_40ms_modulation_mae_db": 1.0791694108320102, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1178075023594634, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2219274860137455, "3000–6000 Hz_10ms_envelope_correlation": 0.9705062459764843, "3000–6000 Hz_10ms_modulation_mae_db": 0.6747662983358227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3489097421756233, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13341368916559082, "3000–6000 Hz_40ms_envelope_correlation": 0.9903146877326351, "3000–6000 Hz_40ms_modulation_mae_db": 0.5674170436344851, "mr_stft_db": 6.176199994303872, "band_mae_db": 0.326084607841496, "rms_gain_error_db": -0.1858510414924787, "output_peak": 0.5998701575498625}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2161911446207212, "全帯域_10ms_rms_envelope_nrmse": 0.09518501936013926, "全帯域_10ms_envelope_correlation": 0.9918539352641582, "全帯域_10ms_modulation_mae_db": 0.45947717793636594, "全帯域_40ms_log_rms_mae_db": 0.6690068872498686, "全帯域_40ms_rms_envelope_nrmse": 0.05624311106056019, "全帯域_40ms_envelope_correlation": 0.9970806999640955, "全帯域_40ms_modulation_mae_db": 0.4951917872914082, "300–1000 Hz_10ms_log_rms_mae_db": 1.8788594885739798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12012512974683394, "300–1000 Hz_10ms_envelope_correlation": 0.9891077764089335, "300–1000 Hz_10ms_modulation_mae_db": 0.5758852498519944, "300–1000 Hz_40ms_log_rms_mae_db": 1.141269218600957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07029944961791651, "300–1000 Hz_40ms_envelope_correlation": 0.9961894312616814, "300–1000 Hz_40ms_modulation_mae_db": 0.6462603847461217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8292021443840198, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11855716959440975, "1000–3000 Hz_10ms_envelope_correlation": 0.9897507144469008, "1000–3000 Hz_10ms_modulation_mae_db": 0.4636158855064772, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1166042085652146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07453804156655383, "1000–3000 Hz_40ms_envelope_correlation": 0.9959336067710868, "1000–3000 Hz_40ms_modulation_mae_db": 0.40259576190328766, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0605316681432755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19452441358854997, "3000–6000 Hz_10ms_envelope_correlation": 0.9775336342717859, "3000–6000 Hz_10ms_modulation_mae_db": 1.452584781295833, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3126157894407593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12236936097059906, "3000–6000 Hz_40ms_envelope_correlation": 0.991605603456889, "3000–6000 Hz_40ms_modulation_mae_db": 1.53187675045723, "mr_stft_db": 6.1795581968444475, "band_mae_db": 0.34291749346096, "rms_gain_error_db": -0.12209113891977907, "output_peak": 0.5647882967297573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.159679303754977, "全帯域_10ms_rms_envelope_nrmse": 0.10464531116248246, "全帯域_10ms_envelope_correlation": 0.9901878111557498, "全帯域_10ms_modulation_mae_db": 0.2014924152627824, "全帯域_40ms_log_rms_mae_db": 0.5722002843443627, "全帯域_40ms_rms_envelope_nrmse": 0.053736624923665266, "全帯域_40ms_envelope_correlation": 0.9974766272396858, "全帯域_40ms_modulation_mae_db": 0.3684913656312781, "300–1000 Hz_10ms_log_rms_mae_db": 1.8696455382365802, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1456575533417521, "300–1000 Hz_10ms_envelope_correlation": 0.9841008980496747, "300–1000 Hz_10ms_modulation_mae_db": 0.7440311192993683, "300–1000 Hz_40ms_log_rms_mae_db": 1.1161624682459603, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07912295614800824, "300–1000 Hz_40ms_envelope_correlation": 0.9955400384115897, "300–1000 Hz_40ms_modulation_mae_db": 1.0566823059468105, "1000–3000 Hz_10ms_log_rms_mae_db": 1.811697748162283, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14412931756603387, "1000–3000 Hz_10ms_envelope_correlation": 0.9846843502331022, "1000–3000 Hz_10ms_modulation_mae_db": 0.7720837838647117, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0605787170661785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07566740609967049, "1000–3000 Hz_40ms_envelope_correlation": 0.9958443252122323, "1000–3000 Hz_40ms_modulation_mae_db": 0.6144635388609384, "3000–6000 Hz_10ms_log_rms_mae_db": 2.069338307613657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1898211835276871, "3000–6000 Hz_10ms_envelope_correlation": 0.9786174319653775, "3000–6000 Hz_10ms_modulation_mae_db": 0.5640152486251897, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3039638162992337, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10532360075355611, "3000–6000 Hz_40ms_envelope_correlation": 0.9932262599299053, "3000–6000 Hz_40ms_modulation_mae_db": 0.39122245744610895, "mr_stft_db": 6.175370905125836, "band_mae_db": 0.2639283436555268, "rms_gain_error_db": -0.1646889793906398, "output_peak": 0.6448439042337076}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.334998588943892, "spectral_convergence": 0.3604116223327677, "spectral_convergence_status": "ok", "amplitude_floor": 1.7451878151186278e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.941651804582128, "spectral_convergence": 0.34743654747120206, "spectral_convergence_status": "ok", "amplitude_floor": 9.668633122862388e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.233109004838916, "spectral_convergence": 0.36762790743310275, "spectral_convergence_status": "ok", "amplitude_floor": 6.131486475721294e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.299449699297122e-05, "output_power": 9.7305271196932e-05, "delta_db": 0.6908707026088013, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007475936848096536, "output_power": 0.0007288796326078215, "delta_db": -0.1100980935843556, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001248382978500818, "output_power": 0.001169769492551963, "delta_db": -0.2824754767192345, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000559570006980701, "output_power": 0.0005407202330860868, "delta_db": -0.1488180756240862, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012219591005283435, "output_power": 0.00012850716042052792, "delta_db": 0.21870657099707208, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001408377861622848, "output_power": 0.00013228093599764842, "delta_db": -0.2722193050323417, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2871980374276486, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15869471688473494, "pitch": null, "output_rms_dbfs": -25.532356914164986, "output_peak": 0.5344383265667301, "reference_rms_dbfs": -25.37366219728025, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.169919799454978}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023508705421562093, "log_rms_mae_db": 1.2223306421972153, "rms_envelope_nrmse": 0.10277757334543927, "envelope_correlation": 0.9905428003236215, "rms_gain_error_db": -0.1585572417545106, "reference_mean_power": 0.0029034514839332195, "output_mean_power": 0.0027993606226551027}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1197992634528758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005958091728481294, "output_power": 0.0005866063945091022, "reference_active": true, "delta_db": -0.0676039231084121}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.320217055945433e-05, "output_power": 7.7142416489629e-05, "reference_active": true, "delta_db": -0.8213278759820364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9969137080619498e-05, "output_power": 1.8321092192197676e-05, "reference_active": true, "delta_db": -0.3740793815786192}], "active_band_count": 3, "active_band_mae_db": 0.4210037268896893}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019484731874036573, "log_rms_mae_db": 0.6312315109481456, "rms_envelope_nrmse": 0.055513233715646926, "envelope_correlation": 0.9972768267238361, "rms_gain_error_db": -0.1584070119548291, "reference_mean_power": 0.002910052669271509, "output_mean_power": 0.0028058222071189074}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0516863068209195e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005686142820757538, "output_power": 0.0005553344498610926, "reference_active": true, "delta_db": -0.1026314828885565}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3606311869244326e-05, "output_power": 3.752754572856819e-05, "reference_active": true, "delta_db": -0.6519919367619513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3652354435919545e-06, "output_power": 1.1078032648523498e-06, "reference_active": true, "delta_db": -0.9074491395973172}], "active_band_count": 3, "active_band_mae_db": 0.5540241864159418}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020607260287182458, "log_rms_mae_db": 1.8778978289719073, "rms_envelope_nrmse": 0.13264738760032635, "envelope_correlation": 0.9867799266356423, "rms_gain_error_db": -0.19273481974695572, "reference_mean_power": 0.001692101437139437, "output_mean_power": 0.0016186498537296274}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.505279098297708e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003919860661962002, "output_power": 0.00038126731006560955, "reference_active": true, "delta_db": -0.12041059191912873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.427449102286661e-05, "output_power": 3.59808526071627e-05, "reference_active": true, "delta_db": -0.9008212703273726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.188127359577471e-06, "output_power": 8.862108020599035e-06, "reference_active": true, "delta_db": -0.15689967098533056}], "active_band_count": 3, "active_band_mae_db": 0.3927105110772773}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016462000863024626, "log_rms_mae_db": 1.1182459584779547, "rms_envelope_nrmse": 0.07486052178027074, "envelope_correlation": 0.9958378503392655, "rms_gain_error_db": -0.19190062643458985, "reference_mean_power": 0.001695759255771571, "output_mean_power": 0.001622460504407218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.197777273221207e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003745482187229297, "output_power": 0.0003632331374666582, "reference_active": true, "delta_db": -0.13322274001171508}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.969988957093567e-05, "output_power": 1.643908838006754e-05, "reference_active": true, "delta_db": -0.7858606137340208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.294406999255848e-07, "output_power": 7.096234787987824e-07, "reference_active": true, "delta_db": -0.6775736575029796}], "active_band_count": 3, "active_band_mae_db": 0.5322190037495719}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010781730421004015, "log_rms_mae_db": 1.7942850289245758, "rms_envelope_nrmse": 0.14460957524005835, "envelope_correlation": 0.9845860707828177, "rms_gain_error_db": -0.19664582018869037, "reference_mean_power": 0.0005546344869318713, "output_mean_power": 0.0005300810615991959}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0245820082367704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001805463155146596, "output_power": 0.00016712734945194233, "reference_active": true, "delta_db": -0.33541104459795607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.244891302260328e-05, "output_power": 1.9382675940384982e-05, "reference_active": true, "delta_db": -0.6378158245333476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.880443381781143e-06, "output_power": 3.6674518100768893e-06, "reference_active": true, "delta_db": -1.2409486290036096}], "active_band_count": 3, "active_band_mae_db": 0.7380584993783045}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010092628732279422, "log_rms_mae_db": 1.032720523366981, "rms_envelope_nrmse": 0.07261164416090037, "envelope_correlation": 0.996227008116526, "rms_gain_error_db": -0.19726359968067397, "reference_mean_power": 0.0005569522649053947, "output_mean_power": 0.0005322205193454425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.880183972268188e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017516791642605334, "output_power": 0.0001607538929382435, "reference_active": true, "delta_db": -0.3729306526055318}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.918962387005773e-06, "output_power": 9.254689744151124e-06, "reference_active": true, "delta_db": -0.30104379439251394}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2471992219002067e-07, "output_power": 2.0763240492051967e-07, "reference_active": true, "delta_db": -1.9421379999937316}], "active_band_count": 3, "active_band_mae_db": 0.8720374823305924}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001308702624055787, "log_rms_mae_db": 2.0343690473458516, "rms_envelope_nrmse": 0.21053219007522064, "envelope_correlation": 0.9729850748737193, "rms_gain_error_db": -0.20191891565486314, "reference_mean_power": 0.00016841175600957635, "output_mean_power": 0.0001607609310129989}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6472123519121696e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.520693965195547e-05, "output_power": 9.040935270922162e-05, "reference_active": true, "delta_db": -0.22455245446615918}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1413816053513955e-05, "output_power": 1.6416343770946337e-05, "reference_active": true, "delta_db": -1.1541762973951568}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.37748433230818e-06, "output_power": 3.917039274173123e-06, "reference_active": true, "delta_db": -1.3762122769180232}], "active_band_count": 3, "active_band_mae_db": 0.9183136762597798}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.575610534879353e-05, "log_rms_mae_db": 1.3034842376485545, "rms_envelope_nrmse": 0.12543308908067805, "envelope_correlation": 0.9903645659882989, "rms_gain_error_db": -0.20170199478965564, "reference_mean_power": 0.00016921252708073642, "output_mean_power": 0.00016153339167827995}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5742070644083465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.764878819980257e-05, "output_power": 9.110363525357575e-05, "reference_active": true, "delta_db": -0.30131151447052135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.033680531591513e-05, "output_power": 8.060059288606662e-06, "reference_active": true, "delta_db": -1.0804810039619175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.467040891819506e-07, "output_power": 2.410002161212734e-07, "reference_active": true, "delta_db": -1.5794153191211122}], "active_band_count": 3, "active_band_mae_db": 0.9870692791845169}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7895585998591725, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6f7ffed46688b3762cada244484600e56815291bdbda0a5e154fe9221b902b72", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9865072223755764, "standard_deviation": 0.0017150275416443841, "minimum": 0.9845860707828177, "maximum": 0.9897507144469008, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8231093102595997, "standard_deviation": 0.03839867869653323, "minimum": 1.7752955759200877, "maximum": 1.8894606756249095, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6349182876110678, "standard_deviation": 0.22853859634882626, "minimum": 0.3305244769292756, "maximum": 0.9171993991242972, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13545088795886556, "standard_deviation": 0.008627885021598359, "minimum": 0.11855716959440975, "maximum": 0.14460957524005835, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957430437351102, "standard_deviation": 0.0003690450553593602, "minimum": 0.9950979358740906, "maximum": 0.996227008116526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.071545152370398, "standard_deviation": 0.04314819023667298, "minimum": 1.0168068099428857, "maximum": 1.143185256177678, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6735582801437368, "standard_deviation": 0.30776186208432954, "minimum": 0.21336340866687256, "maximum": 1.0791694108320102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07663329320378309, "standard_deviation": 0.0032786114246337746, "minimum": 0.07261164416090037, "maximum": 0.08229990764007382, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9770484114510369, "standard_deviation": 0.0038598611576153834, "minimum": 0.9705062459764843, "maximum": 0.9816585519753399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.050632453835057, "standard_deviation": 0.0451544023673621, "minimum": 1.975741477598263, "maximum": 2.1178075023594634, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7250431517435788, "standard_deviation": 0.3610665936478938, "minimum": 0.38031477967584665, "maximum": 1.452584781295833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19733238369502495, "standard_deviation": 0.01737378546157124, "minimum": 0.17540659963995311, "maximum": 0.2219274860137455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920898086007436, "standard_deviation": 0.001486099346659988, "minimum": 0.9903146877326351, "maximum": 0.9940861812579247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3009470192061996, "standard_deviation": 0.03188572844564165, "minimum": 1.2374125690051518, "maximum": 1.3489097421756233, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7764452154849524, "standard_deviation": 0.37502299176702086, "minimum": 0.39122245744610895, "maximum": 1.53187675045723, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1181425338943457, "standard_deviation": 0.014974100783319188, "minimum": 0.09796928874134918, "maximum": 0.13860972246544148, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9871745600002714, "standard_deviation": 0.002239283106032444, "minimum": 0.9841008980496747, "maximum": 0.9892555755855506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8719358922812503, "standard_deviation": 0.036903851829531216, "minimum": 1.7974980791628987, "maximum": 1.9281827187439533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7228305712091116, "standard_deviation": 0.26626375794102936, "minimum": 0.34407920162553013, "maximum": 1.0902549190349187, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13041386922993964, "standard_deviation": 0.01097071356230544, "minimum": 0.12012512974683394, "maximum": 0.1456575533417521, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9957007665699056, "standard_deviation": 0.0009964739871452803, "minimum": 0.9940540005764855, "maximum": 0.9967383712529472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.138816914813011, "standard_deviation": 0.03305281036268259, "minimum": 1.0822812397614916, "maximum": 1.1826425524867699, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9567528889770037, "standard_deviation": 0.38110438408014385, "minimum": 0.3936451916984722, "maximum": 1.441794808657774, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07572596919366566, "standard_deviation": 0.007816040028409885, "minimum": 0.06548595001621972, "maximum": 0.08678131140599334, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3007963310598838, "standard_deviation": 0.04393892000108981, "minimum": 0.22863570847734682, "maximum": 0.364903815612186, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.175478422995507, "standard_deviation": 0.011979809186192585, "minimum": 6.149759087085745, "maximum": 6.189759318121414, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5815475889243082, "standard_deviation": 0.03510641673998022, "minimum": 0.5344383265667301, "maximum": 0.6448439042337076, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14445670911111147, "standard_deviation": 0.04177000846781589, "minimum": -0.2094158341520634, "maximum": -0.09117327130080924, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9912829386925137, "standard_deviation": 0.0011904702445156927, "minimum": 0.9899521250684141, "maximum": 0.9931260721934458, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1879446743298023, "standard_deviation": 0.032429908801177014, "minimum": 1.1335699876982548, "maximum": 1.2223306421972153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.446048643673677, "standard_deviation": 0.23701392861914708, "minimum": 0.03422215602239753, "maximum": 0.7630665628998187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09852170594405568, "standard_deviation": 0.006368536710785088, "minimum": 0.08943895084931365, "maximum": 0.10512787690993539, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973253443443038, "standard_deviation": 0.0007023959121612944, "minimum": 0.9962845364231346, "maximum": 0.9984831393605628, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6071846548803391, "standard_deviation": 0.03636835024090315, "minimum": 0.5596980361493843, "maximum": 0.6690068872498686, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5641787393846129, "standard_deviation": 0.2694503452933823, "minimum": 0.05160247416716898, "maximum": 0.9207044827756571, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05445842050551883, "standard_deviation": 0.005523649114212873, "minimum": 0.046185696756988026, "maximum": 0.062117109352948355, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0004", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8414130210876465, "predicted_mos": 2.8414130210876465, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "85ed10400c7a9179d8c377b85262b2e6fadbe18b58f828abdae426825ba7f5ec", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9250679002143443, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.218526181280846, "全帯域_10ms_rms_envelope_nrmse": 0.09781167988922364, "全帯域_10ms_envelope_correlation": 0.9913231951301356, "全帯域_10ms_modulation_mae_db": 0.3035954083605464, "全帯域_40ms_log_rms_mae_db": 0.6482508304139102, "全帯域_40ms_rms_envelope_nrmse": 0.058412472856589785, "全帯域_40ms_envelope_correlation": 0.9967309785122709, "全帯域_40ms_modulation_mae_db": 0.5704106667043233, "300–1000 Hz_10ms_log_rms_mae_db": 1.385928834807048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13284607191994713, "300–1000 Hz_10ms_envelope_correlation": 0.9865566095660284, "300–1000 Hz_10ms_modulation_mae_db": 0.33409271365995424, "300–1000 Hz_40ms_log_rms_mae_db": 0.8534981271722146, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08558843507866724, "300–1000 Hz_40ms_envelope_correlation": 0.9940911642379998, "300–1000 Hz_40ms_modulation_mae_db": 0.40230648214518433, "1000–3000 Hz_10ms_log_rms_mae_db": 1.384591292598296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12362837381876146, "1000–3000 Hz_10ms_envelope_correlation": 0.9887862326459798, "1000–3000 Hz_10ms_modulation_mae_db": 0.19946326561432748, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8550252765663676, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07819371870401595, "1000–3000 Hz_40ms_envelope_correlation": 0.9954477634077876, "1000–3000 Hz_40ms_modulation_mae_db": 0.18892677429360213, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5180133281970432, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16690094410124043, "3000–6000 Hz_10ms_envelope_correlation": 0.9831507579066814, "3000–6000 Hz_10ms_modulation_mae_db": 0.46061544190199616, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0642273633734156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09572522666110153, "3000–6000 Hz_40ms_envelope_correlation": 0.9945521242938951, "3000–6000 Hz_40ms_modulation_mae_db": 0.41837263634765115, "mr_stft_db": 5.74607520199327, "band_mae_db": 0.2236860504774073, "rms_gain_error_db": -0.09486437731941146, "output_peak": 0.5772032768348578}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3086744619063932, "全帯域_10ms_rms_envelope_nrmse": 0.09918986906411074, "全帯域_10ms_envelope_correlation": 0.9910521577645147, "全帯域_10ms_modulation_mae_db": 0.11792783635842403, "全帯域_40ms_log_rms_mae_db": 0.7148423500836759, "全帯域_40ms_rms_envelope_nrmse": 0.05243120298753345, "全帯域_40ms_envelope_correlation": 0.9973056899841571, "全帯域_40ms_modulation_mae_db": 0.3059733699577764, "300–1000 Hz_10ms_log_rms_mae_db": 1.3191355736946246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11666148168887364, "300–1000 Hz_10ms_envelope_correlation": 0.9896770399124043, "300–1000 Hz_10ms_modulation_mae_db": 0.22431779717018605, "300–1000 Hz_40ms_log_rms_mae_db": 0.8103983781880524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06409166859174471, "300–1000 Hz_40ms_envelope_correlation": 0.9967595592352428, "300–1000 Hz_40ms_modulation_mae_db": 0.33064826028406924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4247315038296056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12808903604554148, "1000–3000 Hz_10ms_envelope_correlation": 0.9878190538876604, "1000–3000 Hz_10ms_modulation_mae_db": 0.5355803746703517, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8683123678619568, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07642666452762166, "1000–3000 Hz_40ms_envelope_correlation": 0.9954416657320403, "1000–3000 Hz_40ms_modulation_mae_db": 0.3231923392488291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4977950948033536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20560651668312852, "3000–6000 Hz_10ms_envelope_correlation": 0.9765919917819833, "3000–6000 Hz_10ms_modulation_mae_db": 0.49205638038554095, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0043277480875787, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13015554660389383, "3000–6000 Hz_40ms_envelope_correlation": 0.9908789484541559, "3000–6000 Hz_40ms_modulation_mae_db": 0.6747799854336577, "mr_stft_db": 5.745498960958585, "band_mae_db": 0.20029782160699305, "rms_gain_error_db": -0.03173044417582151, "output_peak": 0.6379160948670346}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.239934592466212, "全帯域_10ms_rms_envelope_nrmse": 0.09970287446157287, "全帯域_10ms_envelope_correlation": 0.9909741009158407, "全帯域_10ms_modulation_mae_db": 0.3351523523889785, "全帯域_40ms_log_rms_mae_db": 0.6064820765440389, "全帯域_40ms_rms_envelope_nrmse": 0.05576954543743807, "全帯域_40ms_envelope_correlation": 0.9970163655951099, "全帯域_40ms_modulation_mae_db": 0.3117842267448319, "300–1000 Hz_10ms_log_rms_mae_db": 1.435573831210476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12237425448378454, "300–1000 Hz_10ms_envelope_correlation": 0.9886672641340786, "300–1000 Hz_10ms_modulation_mae_db": 0.36435771143059537, "300–1000 Hz_40ms_log_rms_mae_db": 0.8292213268603531, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0664788780691218, "300–1000 Hz_40ms_envelope_correlation": 0.9966002893440611, "300–1000 Hz_40ms_modulation_mae_db": 0.5373494403358637, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4106094552431003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12336588835540663, "1000–3000 Hz_10ms_envelope_correlation": 0.9888448369652882, "1000–3000 Hz_10ms_modulation_mae_db": 0.9488939862611168, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8643464320775471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07589101872422237, "1000–3000 Hz_40ms_envelope_correlation": 0.995743667962031, "1000–3000 Hz_40ms_modulation_mae_db": 0.8502049002265739, "3000–6000 Hz_10ms_log_rms_mae_db": 1.409849616068699, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20525014761780802, "3000–6000 Hz_10ms_envelope_correlation": 0.9780373395251765, "3000–6000 Hz_10ms_modulation_mae_db": 0.4186046497684463, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0177998985562189, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15201445135522257, "3000–6000 Hz_40ms_envelope_correlation": 0.988457711495099, "3000–6000 Hz_40ms_modulation_mae_db": 0.6220478961639833, "mr_stft_db": 5.752073184733195, "band_mae_db": 0.28777899339793434, "rms_gain_error_db": -0.08403895110462731, "output_peak": 0.5760537428575692}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2890240288241013, "全帯域_10ms_rms_envelope_nrmse": 0.10403867517530871, "全帯域_10ms_envelope_correlation": 0.9903276780702458, "全帯域_10ms_modulation_mae_db": 0.6554581051481306, "全帯域_40ms_log_rms_mae_db": 0.6757914684858888, "全帯域_40ms_rms_envelope_nrmse": 0.05365597484008463, "全帯域_40ms_envelope_correlation": 0.997497791974377, "全帯域_40ms_modulation_mae_db": 0.9824524035881774, "300–1000 Hz_10ms_log_rms_mae_db": 1.3788933327240942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14118146244668137, "300–1000 Hz_10ms_envelope_correlation": 0.9849857851828072, "300–1000 Hz_10ms_modulation_mae_db": 0.4116810726340138, "300–1000 Hz_40ms_log_rms_mae_db": 0.8548682880997777, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07312319434073095, "300–1000 Hz_40ms_envelope_correlation": 0.9960860814587635, "300–1000 Hz_40ms_modulation_mae_db": 0.36761784076090703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4405017416331878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12766818031142146, "1000–3000 Hz_10ms_envelope_correlation": 0.9879374215742455, "1000–3000 Hz_10ms_modulation_mae_db": 0.49261318470916676, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8620677904133119, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07687262014496192, "1000–3000 Hz_40ms_envelope_correlation": 0.9954784746856418, "1000–3000 Hz_40ms_modulation_mae_db": 0.2714049515619885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4359343861703517, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19832609803871692, "3000–6000 Hz_10ms_envelope_correlation": 0.9770815237077717, "3000–6000 Hz_10ms_modulation_mae_db": 1.4815935828623175, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0129335521125062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12344751627005465, "3000–6000 Hz_40ms_envelope_correlation": 0.9923485707578614, "3000–6000 Hz_40ms_modulation_mae_db": 1.4547375954815749, "mr_stft_db": 5.75244154437101, "band_mae_db": 0.28632280347503697, "rms_gain_error_db": -0.17198036052772042, "output_peak": 0.5627603079396489}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2565193869474927, "全帯域_10ms_rms_envelope_nrmse": 0.11240782803162089, "全帯域_10ms_envelope_correlation": 0.9885260046704403, "全帯域_10ms_modulation_mae_db": 0.49255741595716335, "全帯域_40ms_log_rms_mae_db": 0.7049324296667141, "全帯域_40ms_rms_envelope_nrmse": 0.06282494296166001, "全帯域_40ms_envelope_correlation": 0.9962544130309086, "全帯域_40ms_modulation_mae_db": 0.6725262670999711, "300–1000 Hz_10ms_log_rms_mae_db": 1.3797834901542319, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14852457905802488, "300–1000 Hz_10ms_envelope_correlation": 0.9831750086849304, "300–1000 Hz_10ms_modulation_mae_db": 0.5198986747754829, "300–1000 Hz_40ms_log_rms_mae_db": 0.837337682720903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08233769014239334, "300–1000 Hz_40ms_envelope_correlation": 0.9946225985834312, "300–1000 Hz_40ms_modulation_mae_db": 0.8878807149210545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3867597229082926, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1089122443032049, "1000–3000 Hz_10ms_envelope_correlation": 0.9913285679389691, "1000–3000 Hz_10ms_modulation_mae_db": 0.25186762440701577, "1000–3000 Hz_40ms_log_rms_mae_db": 0.839472778627354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06485957604752142, "1000–3000 Hz_40ms_envelope_correlation": 0.9969147280527099, "1000–3000 Hz_40ms_modulation_mae_db": 0.10503036099403407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4401844488992774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18828695301773032, "3000–6000 Hz_10ms_envelope_correlation": 0.979747769677775, "3000–6000 Hz_10ms_modulation_mae_db": 0.6199344086631695, "3000–6000 Hz_40ms_log_rms_mae_db": 1.013189754459458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13705656053327844, "3000–6000 Hz_40ms_envelope_correlation": 0.9901876683091168, "3000–6000 Hz_40ms_modulation_mae_db": 0.5049655325376717, "mr_stft_db": 5.74464520765443, "band_mae_db": 0.2509086917374302, "rms_gain_error_db": -0.11738476169013357, "output_peak": 0.566786410935363}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1505068091860098, "全帯域_10ms_rms_envelope_nrmse": 0.09465048430529335, "全帯域_10ms_envelope_correlation": 0.9918862826273929, "全帯域_10ms_modulation_mae_db": 0.3393634027244196, "全帯域_40ms_log_rms_mae_db": 0.5681476732710761, "全帯域_40ms_rms_envelope_nrmse": 0.0522142397345457, "全帯域_40ms_envelope_correlation": 0.9974123916083047, "全帯域_40ms_modulation_mae_db": 0.35892854912385297, "300–1000 Hz_10ms_log_rms_mae_db": 1.3418546185144462, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.115572732115846, "300–1000 Hz_10ms_envelope_correlation": 0.98986780918621, "300–1000 Hz_10ms_modulation_mae_db": 0.5729641462268319, "300–1000 Hz_40ms_log_rms_mae_db": 0.7710374203791848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059911632195031, "300–1000 Hz_40ms_envelope_correlation": 0.9971672856130949, "300–1000 Hz_40ms_modulation_mae_db": 0.7566979605692278, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4051205081886766, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12177085592458127, "1000–3000 Hz_10ms_envelope_correlation": 0.9891791997576869, "1000–3000 Hz_10ms_modulation_mae_db": 0.39583425968630154, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8214955482446055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06276626992783159, "1000–3000 Hz_40ms_envelope_correlation": 0.9972743709954128, "1000–3000 Hz_40ms_modulation_mae_db": 0.3244671980150944, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4604115752863907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1878679482646274, "3000–6000 Hz_10ms_envelope_correlation": 0.9785572613569179, "3000–6000 Hz_10ms_modulation_mae_db": 0.36899144760123187, "3000–6000 Hz_40ms_log_rms_mae_db": 1.031672409434023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11966985574638693, "3000–6000 Hz_40ms_envelope_correlation": 0.9909549002118992, "3000–6000 Hz_40ms_modulation_mae_db": 0.4450737495744512, "mr_stft_db": 5.752571222798683, "band_mae_db": 0.23190258751679893, "rms_gain_error_db": -0.09350893980239881, "output_peak": 0.5631827781048834}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.276631477519807, "全帯域_10ms_rms_envelope_nrmse": 0.10407983452131003, "全帯域_10ms_envelope_correlation": 0.9903225349553189, "全帯域_10ms_modulation_mae_db": 0.39906334956180345, "全帯域_40ms_log_rms_mae_db": 0.618693183410151, "全帯域_40ms_rms_envelope_nrmse": 0.059855037158751845, "全帯域_40ms_envelope_correlation": 0.9968161383679891, "全帯域_40ms_modulation_mae_db": 0.30469570685450614, "300–1000 Hz_10ms_log_rms_mae_db": 1.330886269800876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14103640541915247, "300–1000 Hz_10ms_envelope_correlation": 0.9852148260368598, "300–1000 Hz_10ms_modulation_mae_db": 0.35819563453329467, "300–1000 Hz_40ms_log_rms_mae_db": 0.8016287509504232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08382150117012041, "300–1000 Hz_40ms_envelope_correlation": 0.9949950304087958, "300–1000 Hz_40ms_modulation_mae_db": 0.6145492461274001, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3769470649119253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11510387543344155, "1000–3000 Hz_10ms_envelope_correlation": 0.9902682027782852, "1000–3000 Hz_10ms_modulation_mae_db": 0.2951836377828632, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8186468824881443, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06443954773475304, "1000–3000 Hz_40ms_envelope_correlation": 0.9969360225775068, "1000–3000 Hz_40ms_modulation_mae_db": 0.31693737736203925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4603905066290708, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17992877999275803, "3000–6000 Hz_10ms_envelope_correlation": 0.9823271847319727, "3000–6000 Hz_10ms_modulation_mae_db": 0.519984512261428, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9756757740111087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10495205558774541, "3000–6000 Hz_40ms_envelope_correlation": 0.9944785135779052, "3000–6000 Hz_40ms_modulation_mae_db": 0.5513719175805801, "mr_stft_db": 5.754897387193824, "band_mae_db": 0.2676122473805369, "rms_gain_error_db": -0.16829180314774556, "output_peak": 0.5334793354606341}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2631633104350142, "全帯域_10ms_rms_envelope_nrmse": 0.0952056434084044, "全帯域_10ms_envelope_correlation": 0.9917956290896093, "全帯域_10ms_modulation_mae_db": 0.18901214090929344, "全帯域_40ms_log_rms_mae_db": 0.6855952473854645, "全帯域_40ms_rms_envelope_nrmse": 0.05766906302154702, "全帯域_40ms_envelope_correlation": 0.9968319044395025, "全帯域_40ms_modulation_mae_db": 0.33116752481004724, "300–1000 Hz_10ms_log_rms_mae_db": 1.4098625894358543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12372077329603051, "300–1000 Hz_10ms_envelope_correlation": 0.9884715668391697, "300–1000 Hz_10ms_modulation_mae_db": 0.5926708955842265, "300–1000 Hz_40ms_log_rms_mae_db": 0.8541757390262628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08211154783712943, "300–1000 Hz_40ms_envelope_correlation": 0.9947703002716795, "300–1000 Hz_40ms_modulation_mae_db": 0.785447093638583, "1000–3000 Hz_10ms_log_rms_mae_db": 1.46099305823725, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12255501320194756, "1000–3000 Hz_10ms_envelope_correlation": 0.9889062312029616, "1000–3000 Hz_10ms_modulation_mae_db": 0.22935368130143927, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8404416600944201, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06603083225832367, "1000–3000 Hz_40ms_envelope_correlation": 0.9967145873460144, "1000–3000 Hz_40ms_modulation_mae_db": 0.12773839074458884, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4560475868848406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19347594105630805, "3000–6000 Hz_10ms_envelope_correlation": 0.9793885569167436, "3000–6000 Hz_10ms_modulation_mae_db": 0.3019379835505888, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0162534212990966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14021165914083178, "3000–6000 Hz_40ms_envelope_correlation": 0.9892465730063777, "3000–6000 Hz_40ms_modulation_mae_db": 0.4253464431609176, "mr_stft_db": 5.7514032929204575, "band_mae_db": 0.2359597902141235, "rms_gain_error_db": -0.10123180376454577, "output_peak": 0.5714797444552938}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.757109893652346, "spectral_convergence": 0.331043648520804, "spectral_convergence_status": "ok", "amplitude_floor": 1.7451878151186278e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4275154936920105, "spectral_convergence": 0.32885512967205954, "spectral_convergence_status": "ok", "amplitude_floor": 9.668633122862388e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.053600218635455, "spectral_convergence": 0.37070902582777954, "spectral_convergence_status": "ok", "amplitude_floor": 6.131486475721294e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.299449699297122e-05, "output_power": 9.683895154321762e-05, "delta_db": 0.670007816346585, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007475936848096536, "output_power": 0.000756247409564918, "delta_db": 0.04998275957260889, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001248382978500818, "output_power": 0.0011790156709566143, "delta_db": -0.24828260914652422, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000559570006980701, "output_power": 0.0005503422865869588, "delta_db": -0.07221544279073351, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012219591005283435, "output_power": 0.00012357097105757052, "delta_db": 0.04859789369564136, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001408377861622848, "output_power": 0.00013286671927998156, "delta_db": -0.253029781312351, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2236860504774073, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09486437731941146, "pitch": null, "output_rms_dbfs": -25.468526574599665, "output_peak": 0.5772032768348578, "reference_rms_dbfs": -25.37366219728025, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.74607520199327}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023508705421562093, "log_rms_mae_db": 1.218526181280846, "rms_envelope_nrmse": 0.09781167988922364, "envelope_correlation": 0.9913231951301356, "rms_gain_error_db": -0.09455290906921189, "reference_mean_power": 0.0029034514839332195, "output_mean_power": 0.002840921820774472}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1197992634528758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005958091728481294, "output_power": 0.0005800693307137149, "reference_active": true, "delta_db": -0.11627280999672321}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.320217055945433e-05, "output_power": 9.36940702932315e-05, "reference_active": true, "delta_db": 0.022860795095121567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9969137080619498e-05, "output_power": 1.6718373993468496e-05, "reference_active": true, "delta_db": -0.7716526199897943}], "active_band_count": 3, "active_band_mae_db": 0.3035954083605464}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019484731874036573, "log_rms_mae_db": 0.6482508304139102, "rms_envelope_nrmse": 0.058412472856589785, "envelope_correlation": 0.9967309785122709, "rms_gain_error_db": -0.09365633278534145, "reference_mean_power": 0.002910052669271509, "output_mean_power": 0.0028479687271446346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0516863068209195e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005686142820757538, "output_power": 0.0005519252358071339, "reference_active": true, "delta_db": -0.12937511943868632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3606311869244326e-05, "output_power": 4.759467937107899e-05, "reference_active": true, "delta_db": 0.38009048967247855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3652354435919545e-06, "output_power": 1.0352158567230625e-06, "reference_active": true, "delta_db": -1.201766391001805}], "active_band_count": 3, "active_band_mae_db": 0.5704106667043233}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020607260287182458, "log_rms_mae_db": 1.385928834807048, "rms_envelope_nrmse": 0.13284607191994713, "envelope_correlation": 0.9865566095660284, "rms_gain_error_db": -0.09169210629221236, "reference_mean_power": 0.001692101437139437, "output_mean_power": 0.0016567507798647316}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.505279098297708e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003919860661962002, "output_power": 0.0003854767202303044, "reference_active": true, "delta_db": -0.07272474367839553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.427449102286661e-05, "output_power": 4.690270026047247e-05, "reference_active": true, "delta_db": 0.2504426916987932}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.188127359577471e-06, "output_power": 7.858070772497871e-06, "reference_active": true, "delta_db": -0.679110705602674}], "active_band_count": 3, "active_band_mae_db": 0.33409271365995424}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016462000863024626, "log_rms_mae_db": 0.8534981271722146, "rms_envelope_nrmse": 0.08558843507866724, "envelope_correlation": 0.9940911642379998, "rms_gain_error_db": -0.09101786926023862, "reference_mean_power": 0.001695759255771571, "output_mean_power": 0.0016605899655309127}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.197777273221207e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003745482187229297, "output_power": 0.00037048046699395833, "reference_active": true, "delta_db": -0.047424205752080346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.969988957093567e-05, "output_power": 2.136739468834376e-05, "reference_active": true, "delta_db": 0.3528778046496649}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.294406999255848e-07, "output_power": 6.88848311802909e-07, "reference_active": true, "delta_db": -0.8066174360338079}], "active_band_count": 3, "active_band_mae_db": 0.40230648214518433}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010781730421004015, "log_rms_mae_db": 1.384591292598296, "rms_envelope_nrmse": 0.12362837381876146, "envelope_correlation": 0.9887862326459798, "rms_gain_error_db": -0.18135224372306985, "reference_mean_power": 0.0005546344869318713, "output_mean_power": 0.0005319510199731265}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0245820082367704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001805463155146596, "output_power": 0.00017104635029047233, "reference_active": true, "delta_db": -0.23474818180283744}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.244891302260328e-05, "output_power": 2.098676241624469e-05, "reference_active": true, "delta_db": -0.29249871237545555}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.880443381781143e-06, "output_power": 4.801146837933413e-06, "reference_active": true, "delta_db": -0.0711429026646895}], "active_band_count": 3, "active_band_mae_db": 0.19946326561432748}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010092628732279422, "log_rms_mae_db": 0.8550252765663676, "rms_envelope_nrmse": 0.07819371870401595, "envelope_correlation": 0.9954477634077876, "rms_gain_error_db": -0.18074416032186702, "reference_mean_power": 0.0005569522649053947, "output_mean_power": 0.0005342488037011934}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.880183972268188e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017516791642605334, "output_power": 0.00016529928623109142, "reference_active": true, "delta_db": -0.2518358591424851}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.918962387005773e-06, "output_power": 9.492230189879197e-06, "reference_active": true, "delta_db": -0.19097981942476927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2471992219002067e-07, "output_power": 3.1558217848650966e-07, "reference_active": true, "delta_db": -0.12396464431355204}], "active_band_count": 3, "active_band_mae_db": 0.18892677429360213}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001308702624055787, "log_rms_mae_db": 1.5180133281970432, "rms_envelope_nrmse": 0.16690094410124043, "envelope_correlation": 0.9831507579066814, "rms_gain_error_db": -0.2375609081881484, "reference_mean_power": 0.00016841175600957635, "output_mean_power": 0.00015944698568387568}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6472123519121696e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.520693965195547e-05, "output_power": 9.235212218771227e-05, "reference_active": true, "delta_db": -0.13221725668797554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1413816053513955e-05, "output_power": 1.8807488892112803e-05, "reference_active": true, "delta_db": -0.5636325373637194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.37748433230818e-06, "output_power": 4.591763384059086e-06, "reference_active": true, "delta_db": -0.6859965316542935}], "active_band_count": 3, "active_band_mae_db": 0.46061544190199616}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.575610534879353e-05, "log_rms_mae_db": 1.0642273633734156, "rms_envelope_nrmse": 0.09572522666110153, "envelope_correlation": 0.9945521242938951, "rms_gain_error_db": -0.23757225365886078, "reference_mean_power": 0.00016921252708073642, "output_mean_power": 0.00016020471218233803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5742070644083465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.764878819980257e-05, "output_power": 9.415134575901711e-05, "reference_active": true, "delta_db": -0.15840326140974734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.033680531591513e-05, "output_power": 9.290478766911901e-06, "reference_active": true, "delta_db": -0.46348241719982053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.467040891819506e-07, "output_power": 2.9966480633609694e-07, "reference_active": true, "delta_db": -0.6332322304333855}], "active_band_count": 3, "active_band_mae_db": 0.41837263634765115}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8130226763789753, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e5913c280fdf0bba08c3143a1ba2e41747adf7b7579a43c08adfc5037db7d96f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9891337183438846, "standard_deviation": 0.0011663071140437097, "minimum": 0.9878190538876604, "maximum": 0.9913285679389691, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4112817934437918, "standard_deviation": 0.029366919595641104, "minimum": 1.3769470649119253, "maximum": 1.46099305823725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4185987518040728, "standard_deviation": 0.2472011070782673, "minimum": 0.19946326561432748, "maximum": 0.9488939862611168, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12138668342428828, "standard_deviation": 0.006435604671352399, "minimum": 0.1089122443032049, "maximum": 0.12808903604554148, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996243910094893, "standard_deviation": 0.0007861281507386612, "minimum": 0.9954416657320403, "maximum": 0.9972743709954128, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8462260920467134, "standard_deviation": 0.019256964785847187, "minimum": 0.8186468824881443, "maximum": 0.8683123678619568, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.31348778655584375, "standard_deviation": 0.2340743571406775, "minimum": 0.10503036099403407, "maximum": 0.8502049002265739, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07068503100865646, "standard_deviation": 0.00667667200758948, "minimum": 0.06276626992783159, "maximum": 0.07819371870401595, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793602982006278, "standard_deviation": 0.002347410567894457, "minimum": 0.9765919917819833, "maximum": 0.9831507579066814, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4598283178673783, "standard_deviation": 0.03442713618978832, "minimum": 1.409849616068699, "maximum": 1.5180133281970432, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5829648008743399, "standard_deviation": 0.3756661873573667, "minimum": 0.3019379835505888, "maximum": 1.4815935828623175, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1907054160965397, "standard_deviation": 0.013071405955546952, "minimum": 0.16690094410124043, "maximum": 0.20560651668312852, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9913881262632888, "standard_deviation": 0.002253424636212915, "minimum": 0.988457711495099, "maximum": 0.9945521242938951, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0170099901666756, "standard_deviation": 0.02491766860353006, "minimum": 0.9756757740111087, "maximum": 1.0642273633734156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6370869695350609, "standard_deviation": 0.3432079157891627, "minimum": 0.41837263634765115, "maximum": 1.4547375954815749, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12540410898731438, "standard_deviation": 0.018599393648433965, "minimum": 0.09572522666110153, "maximum": 0.15201445135522257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.987076988692811, "standard_deviation": 0.002459959170719444, "minimum": 0.9831750086849304, "maximum": 0.98986780918621, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3727398175427066, "standard_deviation": 0.03994044254010266, "minimum": 1.3191355736946246, "maximum": 1.435573831210476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.42227233075182324, "standard_deviation": 0.1286236705106564, "minimum": 0.22431779717018605, "maximum": 0.5926708955842265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13023972005354256, "standard_deviation": 0.012428484495911133, "minimum": 0.115572732115846, "maximum": 0.14852457905802488, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956365386441335, "standard_deviation": 0.001153468305378204, "minimum": 0.9940911642379998, "maximum": 0.9971672856130949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8265207141746465, "standard_deviation": 0.030186093575989555, "minimum": 0.7710374203791848, "maximum": 0.8548682880997777, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5853121298477862, "standard_deviation": 0.21039610295945627, "minimum": 0.33064826028406924, "maximum": 0.8878807149210545, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07468306842811737, "standard_deviation": 0.010115960092695157, "minimum": 0.059911632195031, "maximum": 0.08558843507866724, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24805862322578265, "standard_deviation": 0.03097196457762787, "minimum": 0.20029782160699305, "maximum": 0.28777899339793434, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7499507503279315, "standard_deviation": 0.0039122854400214435, "minimum": 5.74464520765443, "maximum": 5.754897387193824, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5736077114319106, "standard_deviation": 0.02937970553131605, "minimum": 0.5334793354606341, "maximum": 0.6379160948670346, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10787893019155054, "standard_deviation": 0.0457259936200389, "minimum": -0.17198036052772042, "maximum": -0.03173044417582151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9907759479029372, "standard_deviation": 0.0010794444767064594, "minimum": 0.9885260046704403, "maximum": 0.9918862826273929, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2503725310707345, "standard_deviation": 0.04910670091761398, "minimum": 1.1505068091860098, "maximum": 1.3086744619063932, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3540162514260949, "standard_deviation": 0.16825448276322938, "minimum": 0.11792783635842403, "maximum": 0.6554581051481306, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10088586110710558, "standard_deviation": 0.005828117290563223, "minimum": 0.09465048430529335, "maximum": 0.11240782803162089, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969832091890775, "standard_deviation": 0.00041398001044629464, "minimum": 0.9962544130309086, "maximum": 0.997497791974377, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6528419074076149, "standard_deviation": 0.05163954960926556, "minimum": 0.5681476732710761, "maximum": 0.7148423500836759, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4797423393604358, "standard_deviation": 0.24606072141306223, "minimum": 0.30469570685450614, "maximum": 0.9824524035881774, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05660405987476881, "standard_deviation": 0.0037747980850429443, "minimum": 0.0522142397345457, "maximum": 0.06282494296166001, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0004", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.7528254985809326, "predicted_mos": 2.7528254985809326, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9e945b465f1b43ab9f1c018b7f4f9049dfbfb6bdfea44329dd22d8da69050095", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.878984599839896, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2246101138220316, "全帯域_10ms_rms_envelope_nrmse": 0.11584996964025866, "全帯域_10ms_envelope_correlation": 0.9878682831119829, "全帯域_10ms_modulation_mae_db": 0.47826841498035516, "全帯域_40ms_log_rms_mae_db": 0.6318649670272461, "全帯域_40ms_rms_envelope_nrmse": 0.059022470028070095, "全帯域_40ms_envelope_correlation": 0.9968337912611887, "全帯域_40ms_modulation_mae_db": 0.8842050827924187, "300–1000 Hz_10ms_log_rms_mae_db": 1.5825032485965778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1599229553857101, "300–1000 Hz_10ms_envelope_correlation": 0.9805099391444866, "300–1000 Hz_10ms_modulation_mae_db": 0.4917758735694678, "300–1000 Hz_40ms_log_rms_mae_db": 0.9474946295144397, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08521938536455342, "300–1000 Hz_40ms_envelope_correlation": 0.9943857471546801, "300–1000 Hz_40ms_modulation_mae_db": 0.6268788421685421, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5992170619569046, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13033333822757515, "1000–3000 Hz_10ms_envelope_correlation": 0.9874266844628329, "1000–3000 Hz_10ms_modulation_mae_db": 0.6976282999151474, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9235781210264223, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07114186154503603, "1000–3000 Hz_40ms_envelope_correlation": 0.996172934134259, "1000–3000 Hz_40ms_modulation_mae_db": 0.6101832875462263, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7032956007180233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16982484299370296, "3000–6000 Hz_10ms_envelope_correlation": 0.9825236090293855, "3000–6000 Hz_10ms_modulation_mae_db": 0.3045115954637107, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1156255877101071, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09976555719496756, "3000–6000 Hz_40ms_envelope_correlation": 0.9937204895376373, "3000–6000 Hz_40ms_modulation_mae_db": 0.563676025525662, "mr_stft_db": 5.900500097243789, "band_mae_db": 0.18906354329844055, "rms_gain_error_db": -0.15427748519946344, "output_peak": 0.6226843441125598}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1485739884266257, "全帯域_10ms_rms_envelope_nrmse": 0.10169824284778095, "全帯域_10ms_envelope_correlation": 0.9907526086060471, "全帯域_10ms_modulation_mae_db": 0.27661787872571875, "全帯域_40ms_log_rms_mae_db": 0.6241638125300769, "全帯域_40ms_rms_envelope_nrmse": 0.05741676151453472, "全帯域_40ms_envelope_correlation": 0.9970650554331685, "全帯域_40ms_modulation_mae_db": 0.13761888665205615, "300–1000 Hz_10ms_log_rms_mae_db": 1.5397875069417388, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1361474485138621, "300–1000 Hz_10ms_envelope_correlation": 0.9859923376398531, "300–1000 Hz_10ms_modulation_mae_db": 0.47216147471619063, "300–1000 Hz_40ms_log_rms_mae_db": 0.9901989777856981, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07585677797415841, "300–1000 Hz_40ms_envelope_correlation": 0.9956532913275848, "300–1000 Hz_40ms_modulation_mae_db": 0.5973356355476686, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6089398957204306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1299131485400353, "1000–3000 Hz_10ms_envelope_correlation": 0.9875884300920988, "1000–3000 Hz_10ms_modulation_mae_db": 0.17224267393242074, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9441384139825661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0738758988227396, "1000–3000 Hz_40ms_envelope_correlation": 0.9959813631215675, "1000–3000 Hz_40ms_modulation_mae_db": 0.20769455199467124, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6789703359196864, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19589236068215626, "3000–6000 Hz_10ms_envelope_correlation": 0.9781598364568864, "3000–6000 Hz_10ms_modulation_mae_db": 1.347420147698249, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1519515425770193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12819970363683214, "3000–6000 Hz_40ms_envelope_correlation": 0.9921041589815088, "3000–6000 Hz_40ms_modulation_mae_db": 1.6001855843351864, "mr_stft_db": 5.900891062938164, "band_mae_db": 0.3350052804660104, "rms_gain_error_db": -0.16419751768177546, "output_peak": 0.5933700358612805}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1685448010018693, "全帯域_10ms_rms_envelope_nrmse": 0.0991431728080198, "全帯域_10ms_envelope_correlation": 0.9911686098461947, "全帯域_10ms_modulation_mae_db": 0.5110692682378403, "全帯域_40ms_log_rms_mae_db": 0.6134326436557737, "全帯域_40ms_rms_envelope_nrmse": 0.04566333932310618, "全帯域_40ms_envelope_correlation": 0.9981929550093566, "全帯域_40ms_modulation_mae_db": 0.5305116879623447, "300–1000 Hz_10ms_log_rms_mae_db": 1.4357171802874187, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12633235516393576, "300–1000 Hz_10ms_envelope_correlation": 0.9879234794299729, "300–1000 Hz_10ms_modulation_mae_db": 0.5968797709492697, "300–1000 Hz_40ms_log_rms_mae_db": 0.8347346473999705, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059899984046961695, "300–1000 Hz_40ms_envelope_correlation": 0.9973183955243646, "300–1000 Hz_40ms_modulation_mae_db": 0.7038824021164904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5435053204899867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.112532081371834, "1000–3000 Hz_10ms_envelope_correlation": 0.9908016221371101, "1000–3000 Hz_10ms_modulation_mae_db": 0.45322132150391825, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9253933655384511, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06704238952201885, "1000–3000 Hz_40ms_envelope_correlation": 0.9967840060192803, "1000–3000 Hz_40ms_modulation_mae_db": 0.46022462795017754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6713612216899072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18540841417008852, "3000–6000 Hz_10ms_envelope_correlation": 0.9791308296497713, "3000–6000 Hz_10ms_modulation_mae_db": 0.986447137633959, "3000–6000 Hz_40ms_log_rms_mae_db": 1.137597030363471, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10083176556427902, "3000–6000 Hz_40ms_envelope_correlation": 0.9939123359632446, "3000–6000 Hz_40ms_modulation_mae_db": 1.0874125321235224, "mr_stft_db": 5.896169541011385, "band_mae_db": 0.20346004379222427, "rms_gain_error_db": -0.1407657337796706, "output_peak": 0.5828702134431357}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1787236120803277, "全帯域_10ms_rms_envelope_nrmse": 0.08612593466996957, "全帯域_10ms_envelope_correlation": 0.9934806808550044, "全帯域_10ms_modulation_mae_db": 0.2038067001214476, "全帯域_40ms_log_rms_mae_db": 0.6716940304587925, "全帯域_40ms_rms_envelope_nrmse": 0.05031237294544779, "全帯域_40ms_envelope_correlation": 0.9978454637043201, "全帯域_40ms_modulation_mae_db": 0.24345787081339365, "300–1000 Hz_10ms_log_rms_mae_db": 1.495940016441386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12718355525888872, "300–1000 Hz_10ms_envelope_correlation": 0.9881254298400396, "300–1000 Hz_10ms_modulation_mae_db": 0.685179897764162, "300–1000 Hz_40ms_log_rms_mae_db": 0.8846566117446669, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07602388478782754, "300–1000 Hz_40ms_envelope_correlation": 0.9960215247773611, "300–1000 Hz_40ms_modulation_mae_db": 1.001463530887726, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5520126526680027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12640484005566827, "1000–3000 Hz_10ms_envelope_correlation": 0.9881497934918985, "1000–3000 Hz_10ms_modulation_mae_db": 0.2860240700365556, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9035685770640876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07763841814152946, "1000–3000 Hz_40ms_envelope_correlation": 0.9952608834864793, "1000–3000 Hz_40ms_modulation_mae_db": 0.24563829653706773, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6572111318129272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18258902353945813, "3000–6000 Hz_10ms_envelope_correlation": 0.9797730864608039, "3000–6000 Hz_10ms_modulation_mae_db": 0.5649002893374665, "3000–6000 Hz_40ms_log_rms_mae_db": 1.14399538640833, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1157748136049969, "3000–6000 Hz_40ms_envelope_correlation": 0.9918714554021457, "3000–6000 Hz_40ms_modulation_mae_db": 0.6431839393527806, "mr_stft_db": 5.871198951826588, "band_mae_db": 0.2154700662709953, "rms_gain_error_db": -0.16868103865826212, "output_peak": 0.5499004372598965}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1520857739039019, "全帯域_10ms_rms_envelope_nrmse": 0.10092601013459004, "全帯域_10ms_envelope_correlation": 0.9907506644298621, "全帯域_10ms_modulation_mae_db": 0.09561075322837463, "全帯域_40ms_log_rms_mae_db": 0.5855949271462231, "全帯域_40ms_rms_envelope_nrmse": 0.05328091336726766, "全帯域_40ms_envelope_correlation": 0.9972931925067189, "全帯域_40ms_modulation_mae_db": 0.09133374601950058, "300–1000 Hz_10ms_log_rms_mae_db": 1.5136868848533849, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12516819620140363, "300–1000 Hz_10ms_envelope_correlation": 0.9880749997550843, "300–1000 Hz_10ms_modulation_mae_db": 0.12007386012983733, "300–1000 Hz_40ms_log_rms_mae_db": 0.9167620619664333, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06389808734375749, "300–1000 Hz_40ms_envelope_correlation": 0.9967253645071896, "300–1000 Hz_40ms_modulation_mae_db": 0.04500070765932651, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6177762590112739, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1301291260066445, "1000–3000 Hz_10ms_envelope_correlation": 0.9876751591676911, "1000–3000 Hz_10ms_modulation_mae_db": 0.4661338876620285, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9584811204028326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07535910805264068, "1000–3000 Hz_40ms_envelope_correlation": 0.9959861222597495, "1000–3000 Hz_40ms_modulation_mae_db": 0.46476641297986415, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7116927869397902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1828859952917899, "3000–6000 Hz_10ms_envelope_correlation": 0.9797388010987463, "3000–6000 Hz_10ms_modulation_mae_db": 0.38595911211747486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1256043667973707, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11067040921504916, "3000–6000 Hz_40ms_envelope_correlation": 0.9922977555228387, "3000–6000 Hz_40ms_modulation_mae_db": 0.3370153679465757, "mr_stft_db": 5.880968306429767, "band_mae_db": 0.19829609311176563, "rms_gain_error_db": -0.09200551063188027, "output_peak": 0.5476675703711866}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.18377386353184, "全帯域_10ms_rms_envelope_nrmse": 0.09747732345019448, "全帯域_10ms_envelope_correlation": 0.9913559675412538, "全帯域_10ms_modulation_mae_db": 0.35571227270905853, "全帯域_40ms_log_rms_mae_db": 0.5951075886295244, "全帯域_40ms_rms_envelope_nrmse": 0.046550426844264726, "全帯域_40ms_envelope_correlation": 0.9978993236054404, "全帯域_40ms_modulation_mae_db": 0.30558760871877827, "300–1000 Hz_10ms_log_rms_mae_db": 1.5309699228052907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1154261544525772, "300–1000 Hz_10ms_envelope_correlation": 0.9898738285242769, "300–1000 Hz_10ms_modulation_mae_db": 0.38290548934745366, "300–1000 Hz_40ms_log_rms_mae_db": 0.9357868535738703, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055745485281746, "300–1000 Hz_40ms_envelope_correlation": 0.9975178066085577, "300–1000 Hz_40ms_modulation_mae_db": 0.6372932152990302, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5912327723914068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13364472007349068, "1000–3000 Hz_10ms_envelope_correlation": 0.9869319192494015, "1000–3000 Hz_10ms_modulation_mae_db": 0.6440216790116859, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9211737683504734, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07683049557473014, "1000–3000 Hz_40ms_envelope_correlation": 0.9957541739854598, "1000–3000 Hz_40ms_modulation_mae_db": 0.564195932008659, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6305113512809228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1984695398921491, "3000–6000 Hz_10ms_envelope_correlation": 0.9791039511712415, "3000–6000 Hz_10ms_modulation_mae_db": 0.6016448767236566, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0776492531619937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1266241942634165, "3000–6000 Hz_40ms_envelope_correlation": 0.9921793588202874, "3000–6000 Hz_40ms_modulation_mae_db": 0.6871682387743615, "mr_stft_db": 5.88144370109207, "band_mae_db": 0.2521908054311633, "rms_gain_error_db": -0.05390045997086212, "output_peak": 0.6788260464534102}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1350875047369224, "全帯域_10ms_rms_envelope_nrmse": 0.09429051914171759, "全帯域_10ms_envelope_correlation": 0.9919156593052719, "全帯域_10ms_modulation_mae_db": 0.07809083521348063, "全帯域_40ms_log_rms_mae_db": 0.5540274217794018, "全帯域_40ms_rms_envelope_nrmse": 0.04959604295824536, "全帯域_40ms_envelope_correlation": 0.9975888185576474, "全帯域_40ms_modulation_mae_db": 0.1155447416835278, "300–1000 Hz_10ms_log_rms_mae_db": 1.5281492189088204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13476910595676556, "300–1000 Hz_10ms_envelope_correlation": 0.9862869298472663, "300–1000 Hz_10ms_modulation_mae_db": 0.3852983332346107, "300–1000 Hz_40ms_log_rms_mae_db": 0.9600152828992724, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07367481331121198, "300–1000 Hz_40ms_envelope_correlation": 0.9956263952267556, "300–1000 Hz_40ms_modulation_mae_db": 0.22849211390190372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6237701456604188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1350136764921696, "1000–3000 Hz_10ms_envelope_correlation": 0.9865696956530245, "1000–3000 Hz_10ms_modulation_mae_db": 0.26755864051122663, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0144628219701126, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07844902473844689, "1000–3000 Hz_40ms_envelope_correlation": 0.9954380418338044, "1000–3000 Hz_40ms_modulation_mae_db": 0.2759958054015232, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6503157793276266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18661120726843186, "3000–6000 Hz_10ms_envelope_correlation": 0.9791391382757894, "3000–6000 Hz_10ms_modulation_mae_db": 0.16852008062894008, "3000–6000 Hz_40ms_log_rms_mae_db": 1.112489906338399, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10033922408952571, "3000–6000 Hz_40ms_envelope_correlation": 0.9937494294243843, "3000–6000 Hz_40ms_modulation_mae_db": 0.09122812732006712, "mr_stft_db": 5.885310080085536, "band_mae_db": 0.09829733481460927, "rms_gain_error_db": -0.027922933489723523, "output_peak": 0.6287131192909142}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1021592820979227, "全帯域_10ms_rms_envelope_nrmse": 0.0810019934908333, "全帯域_10ms_envelope_correlation": 0.9940929167016143, "全帯域_10ms_modulation_mae_db": 0.7936440730006442, "全帯域_40ms_log_rms_mae_db": 0.5226932624054518, "全帯域_40ms_rms_envelope_nrmse": 0.03781812940711765, "全帯域_40ms_envelope_correlation": 0.9987058410083162, "全帯域_40ms_modulation_mae_db": 0.9528079555387792, "300–1000 Hz_10ms_log_rms_mae_db": 1.4968542558185134, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10060225353174633, "300–1000 Hz_10ms_envelope_correlation": 0.9924663180374246, "300–1000 Hz_10ms_modulation_mae_db": 0.7785950038416853, "300–1000 Hz_40ms_log_rms_mae_db": 0.8603113749736782, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05353334274368573, "300–1000 Hz_40ms_envelope_correlation": 0.9979365481158581, "300–1000 Hz_40ms_modulation_mae_db": 0.9865504143970361, "1000–3000 Hz_10ms_log_rms_mae_db": 1.56296277909103, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14219230284554246, "1000–3000 Hz_10ms_envelope_correlation": 0.9850998507256087, "1000–3000 Hz_10ms_modulation_mae_db": 0.4016246984945016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9160572888360959, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08974010155422138, "1000–3000 Hz_40ms_envelope_correlation": 0.9937018509469856, "1000–3000 Hz_40ms_modulation_mae_db": 0.40176392530043614, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6507979213515966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18162028980828684, "3000–6000 Hz_10ms_envelope_correlation": 0.9799806928525785, "3000–6000 Hz_10ms_modulation_mae_db": 1.3599616055946484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1088477675886264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10534639116886077, "3000–6000 Hz_40ms_envelope_correlation": 0.9932659303640352, "3000–6000 Hz_40ms_modulation_mae_db": 1.252636407034932, "mr_stft_db": 5.896226907080109, "band_mae_db": 0.24660221950936992, "rms_gain_error_db": -0.09849156752117004, "output_peak": 0.6172150161647306}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.996868545009875, "spectral_convergence": 0.3377876088033808, "spectral_convergence_status": "ok", "amplitude_floor": 1.7451878151186278e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.596022419479423, "spectral_convergence": 0.3282362786298202, "spectral_convergence_status": "ok", "amplitude_floor": 9.668633122862388e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.10860932724207, "spectral_convergence": 0.34715461486098537, "spectral_convergence_status": "ok", "amplitude_floor": 6.131486475721294e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.299449699297122e-05, "output_power": 9.394788934181724e-05, "delta_db": 0.5383773043389618, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007475936848096536, "output_power": 0.0007164899681575874, "delta_db": -0.18455510024556257, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001248382978500818, "output_power": 0.0011801210812673862, "delta_db": -0.24421269982152527, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000559570006980701, "output_power": 0.0005451833031284502, "delta_db": -0.113118816138178, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012219591005283435, "output_power": 0.00012371841900942008, "delta_db": 0.053776914067487434, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001408377861622848, "output_power": 0.00014084882627669884, "delta_db": 0.00034042517892805813, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18906354329844055, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15427748519946344, "pitch": null, "output_rms_dbfs": -25.527939682479715, "output_peak": 0.6226843441125598, "reference_rms_dbfs": -25.37366219728025, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.900500097243789}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023508705421562093, "log_rms_mae_db": 1.2246101138220316, "rms_envelope_nrmse": 0.11584996964025866, "envelope_correlation": 0.9878682831119829, "rms_gain_error_db": -0.15431954595728292, "reference_mean_power": 0.0029034514839332195, "output_mean_power": 0.0028020934753155737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1197992634528758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005958091728481294, "output_power": 0.0006080289813906676, "reference_active": true, "delta_db": 0.08817095029860442}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.320217055945433e-05, "output_power": 7.990593119318287e-05, "reference_active": true, "delta_db": -0.6684700966336805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9969137080619498e-05, "output_power": 1.7082163661567907e-05, "reference_active": true, "delta_db": -0.6781641980087805}], "active_band_count": 3, "active_band_mae_db": 0.47826841498035516}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019484731874036573, "log_rms_mae_db": 0.6318649670272461, "rms_envelope_nrmse": 0.059022470028070095, "envelope_correlation": 0.9968337912611887, "rms_gain_error_db": -0.15417021472012216, "reference_mean_power": 0.002910052669271509, "output_mean_power": 0.002808560786751814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0516863068209195e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005686142820757538, "output_power": 0.000580601128787696, "reference_active": true, "delta_db": 0.09060112027403586}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3606311869244326e-05, "output_power": 3.5291197199327e-05, "reference_active": true, "delta_db": -0.9188296512771217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3652354435919545e-06, "output_power": 9.351668576578042e-07, "reference_active": true, "delta_db": -1.6431844768260986}], "active_band_count": 3, "active_band_mae_db": 0.8842050827924187}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020607260287182458, "log_rms_mae_db": 1.5825032485965778, "rms_envelope_nrmse": 0.1599229553857101, "envelope_correlation": 0.9805099391444866, "rms_gain_error_db": -0.1920200147628266, "reference_mean_power": 0.001692101437139437, "output_mean_power": 0.0016189162891215752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.505279098297708e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003919860661962002, "output_power": 0.0004043271750321352, "reference_active": true, "delta_db": 0.13462301897222273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.427449102286661e-05, "output_power": 3.8647810764148104e-05, "reference_active": true, "delta_db": -0.5902867926896321}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.188127359577471e-06, "output_power": 7.730102017976144e-06, "reference_active": true, "delta_db": -0.7504178090465484}], "active_band_count": 3, "active_band_mae_db": 0.4917758735694678}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016462000863024626, "log_rms_mae_db": 0.9474946295144397, "rms_envelope_nrmse": 0.08521938536455342, "envelope_correlation": 0.9943857471546801, "rms_gain_error_db": -0.19190212974501059, "reference_mean_power": 0.001695759255771571, "output_mean_power": 0.0016224599427925848}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.197777273221207e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003745482187229297, "output_power": 0.00038573702875541714, "reference_active": true, "delta_db": 0.1278359489293267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.969988957093567e-05, "output_power": 1.687803477802124e-05, "reference_active": true, "delta_db": -0.6714191426638532}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.294406999255848e-07, "output_power": 6.466171193872822e-07, "reference_active": true, "delta_db": -1.0813814349124464}], "active_band_count": 3, "active_band_mae_db": 0.6268788421685421}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010781730421004015, "log_rms_mae_db": 1.5992170619569046, "rms_envelope_nrmse": 0.13033333822757515, "envelope_correlation": 0.9874266844628329, "rms_gain_error_db": -0.13905465420131052, "reference_mean_power": 0.0005546344869318713, "output_mean_power": 0.0005371572044014189}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0245820082367704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001805463155146596, "output_power": 0.00017076975358983808, "reference_active": true, "delta_db": -0.2417767826003793}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.244891302260328e-05, "output_power": 1.9790627356967074e-05, "reference_active": true, "delta_db": -0.5473575585995502}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.880443381781143e-06, "output_power": 3.6147995788959276e-06, "reference_active": true, "delta_db": -1.303750558545513}], "active_band_count": 3, "active_band_mae_db": 0.6976282999151474}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010092628732279422, "log_rms_mae_db": 0.9235781210264223, "rms_envelope_nrmse": 0.07114186154503603, "envelope_correlation": 0.996172934134259, "rms_gain_error_db": -0.1394929406809553, "reference_mean_power": 0.0005569522649053947, "output_mean_power": 0.0005393475127959029}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.880183972268188e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017516791642605334, "output_power": 0.0001651930798199307, "reference_active": true, "delta_db": -0.25462714055383945}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.918962387005773e-06, "output_power": 8.67329758132882e-06, "reference_active": true, "delta_db": -0.5828199609341962}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2471992219002067e-07, "output_power": 2.5834416569925533e-07, "reference_active": true, "delta_db": -0.9931027611506433}], "active_band_count": 3, "active_band_mae_db": 0.6101832875462263}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001308702624055787, "log_rms_mae_db": 1.7032956007180233, "rms_envelope_nrmse": 0.16982484299370296, "envelope_correlation": 0.9825236090293855, "rms_gain_error_db": -0.02730938358320256, "reference_mean_power": 0.00016841175600957635, "output_mean_power": 0.00016735606885590548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6472123519121696e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.520693965195547e-05, "output_power": 9.612766747804286e-05, "reference_active": true, "delta_db": 0.04179799010440025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1413816053513955e-05, "output_power": 1.8968448032090034e-05, "reference_active": true, "delta_db": -0.5266226866288524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.37748433230818e-06, "output_power": 4.9666978767016784e-06, "reference_active": true, "delta_db": -0.3451141096578796}], "active_band_count": 3, "active_band_mae_db": 0.3045115954637107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.575610534879353e-05, "log_rms_mae_db": 1.1156255877101071, "rms_envelope_nrmse": 0.09976555719496756, "envelope_correlation": 0.9937204895376373, "rms_gain_error_db": -0.02725252659895042, "reference_mean_power": 0.00016921252708073642, "output_mean_power": 0.00016815402172798245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5742070644083465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.764878819980257e-05, "output_power": 9.840007281767476e-05, "reference_active": true, "delta_db": 0.0332856165916454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.033680531591513e-05, "output_power": 9.039831128205829e-06, "reference_active": true, "delta_db": -0.5822601926119592}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.467040891819506e-07, "output_power": 2.706516862354519e-07, "reference_active": true, "delta_db": -1.0754822673733813}], "active_band_count": 3, "active_band_mae_db": 0.563676025525662}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8000137484906993, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a45e86723d58ee5dcb8eb75f2501ebd2d6914ecb95db8c63835bffb810006b90", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9875303943724583, "standard_deviation": 0.0016172429809100635, "minimum": 0.9850998507256087, "maximum": 0.9908016221371101, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5874271108736817, "standard_deviation": 0.030798849557050884, "minimum": 1.5435053204899867, "maximum": 1.6237701456604188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.42355690888343556, "standard_deviation": 0.18256240685849187, "minimum": 0.17224267393242074, "maximum": 0.6976282999151474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13002040420162, "standard_deviation": 0.008495869094610644, "minimum": 0.112532081371834, "maximum": 0.14219230284554246, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9956349219734482, "standard_deviation": 0.0009089134450864642, "minimum": 0.9937018509469856, "maximum": 0.9967840060192803, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9383566846463802, "standard_deviation": 0.03509120184203174, "minimum": 0.9035685770640876, "maximum": 1.0144628219701126, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4038078549648282, "standard_deviation": 0.148820163273603, "minimum": 0.20769455199467124, "maximum": 0.6101832875462263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07625966224392039, "standard_deviation": 0.00661584350491741, "minimum": 0.06704238952201885, "maximum": 0.08974010155422138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9796937431244004, "standard_deviation": 0.0012774969445565499, "minimum": 0.9781598364568864, "maximum": 0.9825236090293855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.66926951613006, "standard_deviation": 0.027784973849112563, "minimum": 1.6305113512809228, "maximum": 1.7116927869397902, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7149206056497632, "standard_deviation": 0.4629721465788672, "minimum": 0.16852008062894008, "maximum": 1.3599616055946484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18541270920575795, "standard_deviation": 0.008891905151555184, "minimum": 0.16982484299370296, "maximum": 0.1984695398921491, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9928876142520102, "standard_deviation": 0.0008556972382477867, "minimum": 0.9918714554021457, "maximum": 0.9939123359632446, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1217201051181647, "standard_deviation": 0.023626767474293795, "minimum": 1.0776492531619937, "maximum": 1.1519515425770193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.782813277801636, "standard_deviation": 0.49789603386407094, "minimum": 0.09122812732006712, "maximum": 1.6001855843351864, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11094400734224097, "standard_deviation": 0.011576137881189972, "minimum": 0.09976555719496756, "maximum": 0.12819970363683214, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9874066577773005, "standard_deviation": 0.0034586606063499413, "minimum": 0.9805099391444866, "maximum": 0.9924663180374246, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5154510293316412, "standard_deviation": 0.04238844657662152, "minimum": 1.4357171802874187, "maximum": 1.5825032485965778, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48910871294408464, "standard_deviation": 0.20437975435682226, "minimum": 0.12007386012983733, "maximum": 0.7785950038416853, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12819400305811118, "standard_deviation": 0.01711799609960638, "minimum": 0.10060225353174633, "maximum": 0.1599229553857101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9963981341552939, "standard_deviation": 0.0011902132151641602, "minimum": 0.9943857471546801, "maximum": 0.9979365481158581, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9162450549822536, "standard_deviation": 0.05279888188276491, "minimum": 0.8347346473999705, "maximum": 0.9901989777856981, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6033621077472154, "standard_deviation": 0.331545728726638, "minimum": 0.04500070765932651, "maximum": 1.001463530887726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06798147010673779, "standard_deviation": 0.011318208640748471, "minimum": 0.05353334274368573, "maximum": 0.08521938536455342, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21729817333682233, "standard_deviation": 0.06702397729308392, "minimum": 0.09829733481460927, "maximum": 0.3350052804660104, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.889088580963426, "standard_deviation": 0.010884562631919096, "minimum": 5.871198951826588, "maximum": 5.900891062938164, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6026558478696392, "standard_deviation": 0.04371748952825857, "minimum": 0.5476675703711866, "maximum": 0.6788260464534102, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11253028086660094, "standard_deviation": 0.052883553513505155, "minimum": -0.16868103865826212, "maximum": -0.027922933489723523, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.991423173799654, "standard_deviation": 0.0018978144662751256, "minimum": 0.9878682831119829, "maximum": 0.9940929167016143, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1616948674501801, "standard_deviation": 0.0364488679939225, "minimum": 1.1021592820979227, "maximum": 1.2246101138220316, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.349102524527115, "standard_deviation": 0.2404306388671868, "minimum": 0.07809083521348063, "maximum": 0.7936440730006442, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09706414577292055, "standard_deviation": 0.010550012202988766, "minimum": 0.0810019934908333, "maximum": 0.11584996964025866, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976780551357696, "standard_deviation": 0.0006144781269554888, "minimum": 0.9968337912611887, "maximum": 0.9987058410083162, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5998223317040613, "standard_deviation": 0.0466609882773919, "minimum": 0.5226932624054518, "maximum": 0.6716940304587925, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4076334475225999, "standard_deviation": 0.34539231922239677, "minimum": 0.09133374601950058, "maximum": 0.9528079555387792, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04995755704850677, "standard_deviation": 0.006829852624795896, "minimum": 0.03781812940711765, "maximum": 0.059022470028070095, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0004", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7673025131225586, "predicted_mos": 2.7673025131225586, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d2fe43b19f07e3a8661c88dedfcbea0a497ae115768597ad6c88cf7c07c1fb84", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.815145299769938, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.254188840482445, "全帯域_10ms_rms_envelope_nrmse": 0.10602276750092558, "全帯域_10ms_envelope_correlation": 0.9900754539921346, "全帯域_10ms_modulation_mae_db": 0.4809184175294065, "全帯域_40ms_log_rms_mae_db": 0.6000198914477681, "全帯域_40ms_rms_envelope_nrmse": 0.06815258148097358, "全帯域_40ms_envelope_correlation": 0.9959280798159349, "全帯域_40ms_modulation_mae_db": 0.5038345715454076, "300–1000 Hz_10ms_log_rms_mae_db": 1.4678884104417267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1266423597881821, "300–1000 Hz_10ms_envelope_correlation": 0.9884805472003105, "300–1000 Hz_10ms_modulation_mae_db": 0.8242886760007119, "300–1000 Hz_40ms_log_rms_mae_db": 0.8450149110843522, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08782182324543328, "300–1000 Hz_40ms_envelope_correlation": 0.9947193279104768, "300–1000 Hz_40ms_modulation_mae_db": 0.9355860162747441, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5518873716791293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13908295927899086, "1000–3000 Hz_10ms_envelope_correlation": 0.9856476727610555, "1000–3000 Hz_10ms_modulation_mae_db": 0.17673356714106533, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8654438634075784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08428125997815245, "1000–3000 Hz_40ms_envelope_correlation": 0.9944242172177746, "1000–3000 Hz_40ms_modulation_mae_db": 0.08651213049799693, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6493741543571292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17129997564413854, "3000–6000 Hz_10ms_envelope_correlation": 0.9822326338505484, "3000–6000 Hz_10ms_modulation_mae_db": 0.6917931270993852, "3000–6000 Hz_40ms_log_rms_mae_db": 1.096058327448306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10143650682833455, "3000–6000 Hz_40ms_envelope_correlation": 0.9935463942839717, "3000–6000 Hz_40ms_modulation_mae_db": 0.5914617879759696, "mr_stft_db": 5.843292869073319, "band_mae_db": 0.22165204058264218, "rms_gain_error_db": -0.20419605985940656, "output_peak": 0.6421875700389033}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2122146221160472, "全帯域_10ms_rms_envelope_nrmse": 0.10139728887381662, "全帯域_10ms_envelope_correlation": 0.9907627894772084, "全帯域_10ms_modulation_mae_db": 0.27048423039732206, "全帯域_40ms_log_rms_mae_db": 0.6659461127019196, "全帯域_40ms_rms_envelope_nrmse": 0.049751127289146525, "全帯域_40ms_envelope_correlation": 0.9978223554888173, "全帯域_40ms_modulation_mae_db": 0.2935465575284772, "300–1000 Hz_10ms_log_rms_mae_db": 1.4760570243600986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12968221698335872, "300–1000 Hz_10ms_envelope_correlation": 0.9873710568766816, "300–1000 Hz_10ms_modulation_mae_db": 0.45277180191526806, "300–1000 Hz_40ms_log_rms_mae_db": 0.8599616859164596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06532444270905205, "300–1000 Hz_40ms_envelope_correlation": 0.9969371948558368, "300–1000 Hz_40ms_modulation_mae_db": 0.7548975111561328, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5002448272440765, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11990182077266476, "1000–3000 Hz_10ms_envelope_correlation": 0.989392214479907, "1000–3000 Hz_10ms_modulation_mae_db": 0.4054079838091155, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8733294204195156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0656080966714387, "1000–3000 Hz_40ms_envelope_correlation": 0.9967667383768988, "1000–3000 Hz_40ms_modulation_mae_db": 0.3427681495832113, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5947776683800805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21102285729161926, "3000–6000 Hz_10ms_envelope_correlation": 0.9730106299218269, "3000–6000 Hz_10ms_modulation_mae_db": 0.21543526989092746, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0492441673104467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11753868331897492, "3000–6000 Hz_40ms_envelope_correlation": 0.9913121469417092, "3000–6000 Hz_40ms_modulation_mae_db": 0.17437749137086053, "mr_stft_db": 5.83856631460478, "band_mae_db": 0.18526815091250726, "rms_gain_error_db": -0.14533048626709932, "output_peak": 0.6115810147697249}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2549485683090402, "全帯域_10ms_rms_envelope_nrmse": 0.09148618352293467, "全帯域_10ms_envelope_correlation": 0.9924343943093185, "全帯域_10ms_modulation_mae_db": 0.37304949696003725, "全帯域_40ms_log_rms_mae_db": 0.5881500095695946, "全帯域_40ms_rms_envelope_nrmse": 0.04793482566565048, "全帯域_40ms_envelope_correlation": 0.9978555987829701, "全帯域_40ms_modulation_mae_db": 0.5431741285804944, "300–1000 Hz_10ms_log_rms_mae_db": 1.4375382616649286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11126504688114885, "300–1000 Hz_10ms_envelope_correlation": 0.9906683471494232, "300–1000 Hz_10ms_modulation_mae_db": 0.6066250843068252, "300–1000 Hz_40ms_log_rms_mae_db": 0.8266692770882905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06200301064293039, "300–1000 Hz_40ms_envelope_correlation": 0.9970619956219222, "300–1000 Hz_40ms_modulation_mae_db": 0.8033978171105597, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5404869615381265, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13629087614350688, "1000–3000 Hz_10ms_envelope_correlation": 0.986196081390687, "1000–3000 Hz_10ms_modulation_mae_db": 0.8243197372594525, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9104545774516672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08531345159356479, "1000–3000 Hz_40ms_envelope_correlation": 0.9943087018504272, "1000–3000 Hz_40ms_modulation_mae_db": 0.794560096281077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5708380706866252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19831299605691363, "3000–6000 Hz_10ms_envelope_correlation": 0.9760755888208396, "3000–6000 Hz_10ms_modulation_mae_db": 0.6072393147861049, "3000–6000 Hz_40ms_log_rms_mae_db": 1.012720897269623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1005545153190788, "3000–6000 Hz_40ms_envelope_correlation": 0.9936632402524048, "3000–6000 Hz_40ms_modulation_mae_db": 0.46381370493693536, "mr_stft_db": 5.8317844251311195, "band_mae_db": 0.1727497593358912, "rms_gain_error_db": -0.10084315191368183, "output_peak": 0.6173406754394325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.16417174706971, "全帯域_10ms_rms_envelope_nrmse": 0.09929259651670026, "全帯域_10ms_envelope_correlation": 0.9911191020545351, "全帯域_10ms_modulation_mae_db": 0.22063627793937224, "全帯域_40ms_log_rms_mae_db": 0.5445827940698154, "全帯域_40ms_rms_envelope_nrmse": 0.055178384080893754, "全帯域_40ms_envelope_correlation": 0.9972128088322336, "全帯域_40ms_modulation_mae_db": 0.22984886042496247, "300–1000 Hz_10ms_log_rms_mae_db": 1.4242348140687717, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11842982926312869, "300–1000 Hz_10ms_envelope_correlation": 0.9894380750095724, "300–1000 Hz_10ms_modulation_mae_db": 0.33483616713879377, "300–1000 Hz_40ms_log_rms_mae_db": 0.8004288080756642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06721061601915643, "300–1000 Hz_40ms_envelope_correlation": 0.9965760567560468, "300–1000 Hz_40ms_modulation_mae_db": 0.6526272021047853, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5452542960697178, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12304071200138653, "1000–3000 Hz_10ms_envelope_correlation": 0.9889445447145868, "1000–3000 Hz_10ms_modulation_mae_db": 0.5035896455990562, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8960608269536322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07216316076981395, "1000–3000 Hz_40ms_envelope_correlation": 0.9962400256802979, "1000–3000 Hz_40ms_modulation_mae_db": 0.5382542586142685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6246588609339907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24893527746792662, "3000–6000 Hz_10ms_envelope_correlation": 0.9623063599609942, "3000–6000 Hz_10ms_modulation_mae_db": 0.345660916209704, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0519985173131776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14718302061206562, "3000–6000 Hz_40ms_envelope_correlation": 0.9863252141940727, "3000–6000 Hz_40ms_modulation_mae_db": 0.3419566418608475, "mr_stft_db": 5.841403578576844, "band_mae_db": 0.2934817336371673, "rms_gain_error_db": -0.12759046284605408, "output_peak": 0.5656958044896545}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.194145500737899, "全帯域_10ms_rms_envelope_nrmse": 0.09499167763279488, "全帯域_10ms_envelope_correlation": 0.9918266951627405, "全帯域_10ms_modulation_mae_db": 0.18794041584599344, "全帯域_40ms_log_rms_mae_db": 0.6106969446044495, "全帯域_40ms_rms_envelope_nrmse": 0.05212589649169815, "全帯域_40ms_envelope_correlation": 0.9974215540162764, "全帯域_40ms_modulation_mae_db": 0.15161276019292919, "300–1000 Hz_10ms_log_rms_mae_db": 1.521162295126314, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12679363024940607, "300–1000 Hz_10ms_envelope_correlation": 0.9878069745837913, "300–1000 Hz_10ms_modulation_mae_db": 0.19557851673377744, "300–1000 Hz_40ms_log_rms_mae_db": 0.8991665593889286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07735380147083917, "300–1000 Hz_40ms_envelope_correlation": 0.9952944111638665, "300–1000 Hz_40ms_modulation_mae_db": 0.3504566854122276, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4914824934021325, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1315883364688581, "1000–3000 Hz_10ms_envelope_correlation": 0.9871752581038549, "1000–3000 Hz_10ms_modulation_mae_db": 0.6716496043572215, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8454278055616663, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06385033970660295, "1000–3000 Hz_40ms_envelope_correlation": 0.9969395092093638, "1000–3000 Hz_40ms_modulation_mae_db": 0.7239247595836643, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6096910478740427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18582127380381372, "3000–6000 Hz_10ms_envelope_correlation": 0.9791972011511231, "3000–6000 Hz_10ms_modulation_mae_db": 0.14428203669766246, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0378924221141508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1072427496917341, "3000–6000 Hz_40ms_envelope_correlation": 0.9927872344376626, "3000–6000 Hz_40ms_modulation_mae_db": 0.09363380721727638, "mr_stft_db": 5.843854088547636, "band_mae_db": 0.20520049938582122, "rms_gain_error_db": -0.09498154270406316, "output_peak": 0.6562547559829246}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2689947516998958, "全帯域_10ms_rms_envelope_nrmse": 0.10446891027469389, "全帯域_10ms_envelope_correlation": 0.9902203988449482, "全帯域_10ms_modulation_mae_db": 0.432764095752509, "全帯域_40ms_log_rms_mae_db": 0.626410578956378, "全帯域_40ms_rms_envelope_nrmse": 0.05794430754943918, "全帯域_40ms_envelope_correlation": 0.9970039607354109, "全帯域_40ms_modulation_mae_db": 0.7302401429422506, "300–1000 Hz_10ms_log_rms_mae_db": 1.5330542081481917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13152318364268523, "300–1000 Hz_10ms_envelope_correlation": 0.986913801995234, "300–1000 Hz_10ms_modulation_mae_db": 0.2583753682746059, "300–1000 Hz_40ms_log_rms_mae_db": 0.9100383532414594, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0795114985315129, "300–1000 Hz_40ms_envelope_correlation": 0.9951081164439556, "300–1000 Hz_40ms_modulation_mae_db": 0.14656851669204127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5770134226294372, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13163316772890726, "1000–3000 Hz_10ms_envelope_correlation": 0.9873941405718135, "1000–3000 Hz_10ms_modulation_mae_db": 0.16433771611998668, "1000–3000 Hz_40ms_log_rms_mae_db": 0.936714258055893, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07652621822696287, "1000–3000 Hz_40ms_envelope_correlation": 0.9958932146033119, "1000–3000 Hz_40ms_modulation_mae_db": 0.1198185469610001, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6288980048202721, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19064433629677896, "3000–6000 Hz_10ms_envelope_correlation": 0.9780901897559604, "3000–6000 Hz_10ms_modulation_mae_db": 1.433675431705324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.061504830494132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10651637201642551, "3000–6000 Hz_40ms_envelope_correlation": 0.993571176258984, "3000–6000 Hz_40ms_modulation_mae_db": 1.3641727249918139, "mr_stft_db": 5.849013987055556, "band_mae_db": 0.3017579532473874, "rms_gain_error_db": -0.16173674641209473, "output_peak": 0.5537407662191425}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2066673931632321, "全帯域_10ms_rms_envelope_nrmse": 0.10140547395974604, "全帯域_10ms_envelope_correlation": 0.9907569952147175, "全帯域_10ms_modulation_mae_db": 0.5255402695719633, "全帯域_40ms_log_rms_mae_db": 0.6231202281834506, "全帯域_40ms_rms_envelope_nrmse": 0.05351062796476459, "全帯域_40ms_envelope_correlation": 0.9974240309329484, "全帯域_40ms_modulation_mae_db": 0.5771166554690165, "300–1000 Hz_10ms_log_rms_mae_db": 1.469635862327221, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14763796037630367, "300–1000 Hz_10ms_envelope_correlation": 0.9834903723437806, "300–1000 Hz_10ms_modulation_mae_db": 0.763456778180757, "300–1000 Hz_40ms_log_rms_mae_db": 0.8594836336723086, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07347046752004284, "300–1000 Hz_40ms_envelope_correlation": 0.9959937184927873, "300–1000 Hz_40ms_modulation_mae_db": 1.0236859815714843, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5221401740755456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12588274011328665, "1000–3000 Hz_10ms_envelope_correlation": 0.9883083942856217, "1000–3000 Hz_10ms_modulation_mae_db": 0.4219838578206258, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8524440768980964, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06582678457382445, "1000–3000 Hz_40ms_envelope_correlation": 0.9966038942663225, "1000–3000 Hz_40ms_modulation_mae_db": 0.32655481848716694, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5735714603340545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17704721978234922, "3000–6000 Hz_10ms_envelope_correlation": 0.9818373906028008, "3000–6000 Hz_10ms_modulation_mae_db": 1.1814018347245445, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0129995547788564, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11583787247479499, "3000–6000 Hz_40ms_envelope_correlation": 0.9930057390338616, "3000–6000 Hz_40ms_modulation_mae_db": 1.1970452692691975, "mr_stft_db": 5.836459238782701, "band_mae_db": 0.3081646994481464, "rms_gain_error_db": -0.14445983985150704, "output_peak": 0.5512114793255986}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1895344860450703, "全帯域_10ms_rms_envelope_nrmse": 0.09739628695021206, "全帯域_10ms_envelope_correlation": 0.9916769507103066, "全帯域_10ms_modulation_mae_db": 0.6294262979905138, "全帯域_40ms_log_rms_mae_db": 0.6335081174858375, "全帯域_40ms_rms_envelope_nrmse": 0.05206233195529121, "全帯域_40ms_envelope_correlation": 0.997832364013532, "全帯域_40ms_modulation_mae_db": 0.5652052553406165, "300–1000 Hz_10ms_log_rms_mae_db": 1.4594649193333613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10971479586475534, "300–1000 Hz_10ms_envelope_correlation": 0.9912892022612979, "300–1000 Hz_10ms_modulation_mae_db": 0.7300111770532273, "300–1000 Hz_40ms_log_rms_mae_db": 0.8413023274324383, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06066445936206157, "300–1000 Hz_40ms_envelope_correlation": 0.9976705143008467, "300–1000 Hz_40ms_modulation_mae_db": 0.8609523922107387, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5367180910114133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14358998417033048, "1000–3000 Hz_10ms_envelope_correlation": 0.9847117028408074, "1000–3000 Hz_10ms_modulation_mae_db": 0.6443147637338683, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9130883611874911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07625921320103521, "1000–3000 Hz_40ms_envelope_correlation": 0.9956051059692537, "1000–3000 Hz_40ms_modulation_mae_db": 0.5410499629874761, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5843552201954436, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22777446665274306, "3000–6000 Hz_10ms_envelope_correlation": 0.968322790925694, "3000–6000 Hz_10ms_modulation_mae_db": 0.7180727868418222, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0411424062756702, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12268669179231755, "3000–6000 Hz_40ms_envelope_correlation": 0.9905009964228605, "3000–6000 Hz_40ms_modulation_mae_db": 0.5681834422228867, "mr_stft_db": 5.8414247622176205, "band_mae_db": 0.2371791021366334, "rms_gain_error_db": -0.19917538440953828, "output_peak": 0.6042753714674456}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9432384624379475, "spectral_convergence": 0.3362024714175492, "spectral_convergence_status": "ok", "amplitude_floor": 1.7451878151186278e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5301407830426275, "spectral_convergence": 0.3321584955778757, "spectral_convergence_status": "ok", "amplitude_floor": 9.668633122862388e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.056499361739379, "spectral_convergence": 0.3555591972502704, "spectral_convergence_status": "ok", "amplitude_floor": 6.131486475721294e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.299449699297122e-05, "output_power": 9.604557158802705e-05, "delta_db": 0.6342804833088366, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007475936848096536, "output_power": 0.000741505486922211, "delta_db": -0.03551255054443927, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001248382978500818, "output_power": 0.00110701208350238, "delta_db": -0.5219547708744817, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000559570006980701, "output_power": 0.000563699627809484, "delta_db": 0.031933198638593, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012219591005283435, "output_power": 0.0001223904014447757, "delta_db": 0.006906891525032892, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001408377861622848, "output_power": 0.00013765334043633652, "delta_db": -0.09932434860446947, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22165204058264218, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.20419605985940656, "pitch": null, "output_rms_dbfs": -25.57785825713966, "output_peak": 0.6421875700389033, "reference_rms_dbfs": -25.37366219728025, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.843292869073319}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023508705421562093, "log_rms_mae_db": 1.254188840482445, "rms_envelope_nrmse": 0.10602276750092558, "envelope_correlation": 0.9900754539921346, "rms_gain_error_db": -0.20385916659409906, "reference_mean_power": 0.0029034514839332195, "output_mean_power": 0.002770311831388891}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1197992634528758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005958091728481294, "output_power": 0.0005766793605706937, "reference_active": true, "delta_db": -0.14172776927704628}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.320217055945433e-05, "output_power": 8.124458258942032e-05, "reference_active": true, "delta_db": -0.5963161489156962}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9969137080619498e-05, "output_power": 1.6978063940097526e-05, "reference_active": true, "delta_db": -0.704711334395477}], "active_band_count": 3, "active_band_mae_db": 0.4809184175294065}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019484731874036573, "log_rms_mae_db": 0.6000198914477681, "rms_envelope_nrmse": 0.06815258148097358, "envelope_correlation": 0.9959280798159349, "rms_gain_error_db": -0.20403338367333668, "reference_mean_power": 0.002910052669271509, "output_mean_power": 0.0027764989336969436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0516863068209195e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005686142820757538, "output_power": 0.0005456431138823575, "reference_active": true, "delta_db": -0.17909085115156642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3606311869244326e-05, "output_power": 3.999061429876942e-05, "reference_active": true, "delta_db": -0.3759128109247287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3652354435919545e-06, "output_power": 1.0953616948561278e-06, "reference_active": true, "delta_db": -0.9565000525599275}], "active_band_count": 3, "active_band_mae_db": 0.5038345715454076}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020607260287182458, "log_rms_mae_db": 1.4678884104417267, "rms_envelope_nrmse": 0.1266423597881821, "envelope_correlation": 0.9884805472003105, "rms_gain_error_db": -0.3271487117799879, "reference_mean_power": 0.001692101437139437, "output_mean_power": 0.0015693200489695148}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.505279098297708e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003919860661962002, "output_power": 0.0003638534960519875, "reference_active": true, "delta_db": -0.3234407741218623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.427449102286661e-05, "output_power": 3.488157616523911e-05, "reference_active": true, "delta_db": -1.0355747645184936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.188127359577471e-06, "output_power": 7.109547456968123e-06, "reference_active": true, "delta_db": -1.1138504893617798}], "active_band_count": 3, "active_band_mae_db": 0.8242886760007119}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016462000863024626, "log_rms_mae_db": 0.8450149110843522, "rms_envelope_nrmse": 0.08782182324543328, "envelope_correlation": 0.9947193279104768, "rms_gain_error_db": -0.32822683282758697, "reference_mean_power": 0.001695759255771571, "output_mean_power": 0.0015723220788840203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.197777273221207e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003745482187229297, "output_power": 0.00034440716996074516, "reference_active": true, "delta_db": -0.36435551858867815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.969988957093567e-05, "output_power": 1.553801876585026e-05, "reference_active": true, "delta_db": -1.0306815007508499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.294406999255848e-07, "output_power": 5.992571700849447e-07, "reference_active": true, "delta_db": -1.4117210294847045}], "active_band_count": 3, "active_band_mae_db": 0.9355860162747441}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010781730421004015, "log_rms_mae_db": 1.5518873716791293, "rms_envelope_nrmse": 0.13908295927899086, "envelope_correlation": 0.9856476727610555, "rms_gain_error_db": -0.048660266555689446, "reference_mean_power": 0.0005546344869318713, "output_mean_power": 0.0005484548024602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0245820082367704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001805463155146596, "output_power": 0.0001787035763106189, "reference_active": true, "delta_db": -0.044553860975291945}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.244891302260328e-05, "output_power": 2.0358418390294636e-05, "reference_active": true, "delta_db": -0.4245128188867036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.880443381781143e-06, "output_power": 4.8122244651884805e-06, "reference_active": true, "delta_db": -0.061134021561200375}], "active_band_count": 3, "active_band_mae_db": 0.17673356714106533}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010092628732279422, "log_rms_mae_db": 0.8654438634075784, "rms_envelope_nrmse": 0.08428125997815245, "envelope_correlation": 0.9944242172177746, "rms_gain_error_db": -0.04838184747077681, "reference_mean_power": 0.0005569522649053947, "output_mean_power": 0.0005507820645779278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.880183972268188e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017516791642605334, "output_power": 0.0001727360850707171, "reference_active": true, "delta_db": -0.06071491762905028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.918962387005773e-06, "output_power": 9.502198460052716e-06, "reference_active": true, "delta_db": -0.18642146685970365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2471992219002067e-07, "output_power": 3.2564838967930977e-07, "reference_active": true, "delta_db": 0.012400007005236838}], "active_band_count": 3, "active_band_mae_db": 0.08651213049799693}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001308702624055787, "log_rms_mae_db": 1.6493741543571292, "rms_envelope_nrmse": 0.17129997564413854, "envelope_correlation": 0.9822326338505484, "rms_gain_error_db": -0.030204326697224632, "reference_mean_power": 0.00016841175600957635, "output_mean_power": 0.00016724454893579142}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6472123519121696e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.520693965195547e-05, "output_power": 9.782408014998209e-05, "reference_active": true, "delta_db": 0.11777167529964608}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1413816053513955e-05, "output_power": 1.863826919165513e-05, "reference_active": true, "delta_db": -0.6028848779128682}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.37748433230818e-06, "output_power": 3.9364693191959565e-06, "reference_active": true, "delta_db": -1.3547228280856416}], "active_band_count": 3, "active_band_mae_db": 0.6917931270993852}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.575610534879353e-05, "log_rms_mae_db": 1.096058327448306, "rms_envelope_nrmse": 0.10143650682833455, "envelope_correlation": 0.9935463942839717, "rms_gain_error_db": -0.029746364073253637, "reference_mean_power": 0.00016921252708073642, "output_mean_power": 0.0001680574908162194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5742070644083465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.764878819980257e-05, "output_power": 9.867069901813876e-05, "reference_active": true, "delta_db": 0.04521346751600041}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.033680531591513e-05, "output_power": 9.196990434143221e-06, "reference_active": true, "delta_db": -0.507406020469273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.467040891819506e-07, "output_power": 2.6168713733217646e-07, "reference_active": true, "delta_db": -1.2217658759426355}], "active_band_count": 3, "active_band_mae_db": 0.5914617879759696}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.809370828626246, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f1d1d7fe82b7236a554bbec47bb4df3865e1f645f24bac52d72e4a96f503f51a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9872212511435416, "standard_deviation": 0.0016353970027872874, "minimum": 0.9847117028408074, "maximum": 0.989392214479907, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5331534547061973, "standard_deviation": 0.027847320659058174, "minimum": 1.4914824934021325, "maximum": 1.5770134226294372, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.476542109480049, "standard_deviation": 0.23399904078510592, "minimum": 0.16433771611998668, "maximum": 0.8243197372594525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13137632458474144, "standard_deviation": 0.008142968732636691, "minimum": 0.11990182077266476, "maximum": 0.14358998417033048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9958476758967063, "standard_deviation": 0.0010154640086191262, "minimum": 0.9943087018504272, "maximum": 0.9969395092093638, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8866203987419425, "standard_deviation": 0.03241931522542142, "minimum": 0.8454278055616663, "maximum": 0.936714258055893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.43418034037448267, "standard_deviation": 0.26068970513100326, "minimum": 0.08651213049799693, "maximum": 0.794560096281077, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07372856559017442, "standard_deviation": 0.008351304241281022, "minimum": 0.06385033970660295, "maximum": 0.08531345159356479, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9751340981237234, "standard_deviation": 0.006937588275157307, "minimum": 0.9623063599609942, "maximum": 0.9822326338505484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6045205609477047, "standard_deviation": 0.028365752930280087, "minimum": 1.5708380706866252, "maximum": 1.6493741543571292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6671950897444343, "standard_deviation": 0.45349408726362156, "minimum": 0.14428203669766246, "maximum": 1.433675431705324, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20135730037453536, "standard_deviation": 0.026497384957360804, "minimum": 0.17129997564413854, "maximum": 0.24893527746792662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9918390177281908, "standard_deviation": 0.0025063796452289492, "minimum": 0.9863252141940727, "maximum": 0.9936632402524048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0454451403755454, "standard_deviation": 0.02692883980510722, "minimum": 1.012720897269623, "maximum": 1.096058327448306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5993306087307235, "standard_deviation": 0.4573524648298994, "minimum": 0.09363380721727638, "maximum": 1.3641727249918139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11487455150671576, "standard_deviation": 0.015243923107269281, "minimum": 0.1005545153190788, "maximum": 0.14718302061206562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9881822971775114, "standard_deviation": 0.002447198142774269, "minimum": 0.9834903723437806, "maximum": 0.9912892022612979, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4736294744338267, "standard_deviation": 0.03739463770759198, "minimum": 1.4242348140687717, "maximum": 1.5330542081481917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5207429462004958, "standard_deviation": 0.24377067897575055, "minimum": 0.19557851673377744, "maximum": 0.8242886760007119, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1252111278811211, "standard_deviation": 0.012235567195372044, "minimum": 0.10971479586475534, "maximum": 0.14763796037630367, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961701669432173, "standard_deviation": 0.0010576555169155464, "minimum": 0.9947193279104768, "maximum": 0.9976705143008467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8552581944874877, "standard_deviation": 0.0360397922180585, "minimum": 0.8004288080756642, "maximum": 0.9100383532414594, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6910215153165893, "standard_deviation": 0.29998812252083246, "minimum": 0.14656851669204127, "maximum": 1.0236859815714843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07167001493762858, "standard_deviation": 0.009506708808630235, "minimum": 0.06066445936206157, "maximum": 0.08782182324543328, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24068174233577455, "standard_deviation": 0.05397198901619491, "minimum": 0.1727497593358912, "maximum": 0.3081646994481464, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.840724907998697, "standard_deviation": 0.0051938577664512675, "minimum": 5.8317844251311195, "maximum": 5.849013987055556, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6002859297166033, "standard_deviation": 0.03980998906993888, "minimum": 0.5512114793255986, "maximum": 0.6562547559829246, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14728920928293063, "standard_deviation": 0.040413058192673726, "minimum": -0.20419605985940656, "maximum": -0.09498154270406316, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9911090974707386, "standard_deviation": 0.0008190608615726942, "minimum": 0.9900754539921346, "maximum": 0.9924343943093185, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2181082387029174, "standard_deviation": 0.03725184915797425, "minimum": 1.16417174706971, "maximum": 1.2689947516998958, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3900949377483897, "standard_deviation": 0.15578119082638373, "minimum": 0.18794041584599344, "maximum": 0.6294262979905138, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.099557648153978, "standard_deviation": 0.004832263070672111, "minimum": 0.09148618352293467, "maximum": 0.10602276750092558, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973125940772654, "standard_deviation": 0.0006404522296831868, "minimum": 0.9959280798159349, "maximum": 0.9978555987829701, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6115543346274017, "standard_deviation": 0.03578036206476282, "minimum": 0.5445827940698154, "maximum": 0.6659461127019196, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4493223665030193, "standard_deviation": 0.20054229596701068, "minimum": 0.15161276019292919, "maximum": 0.7302401429422506, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054582510309732185, "standard_deviation": 0.006289137267588362, "minimum": 0.04793482566565048, "maximum": 0.06815258148097358, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5105-28241-0019", "language": "en", "style": "normal", "speaker": "librispeech_5105", "source_utterance_id": "5105-28241-0019", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 84640, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5105-28241-0019.wav", "source_file_sha256": "12e1f4ec6493b355308fac9ac860430bf018c85b9c46665d98137123c4e9035e", "decoded_pcm_sha256": "3d0b0ee3168cffdc7dff07188417fdc9c6c2c31369b906ef6a97ab560d3e8f4b", "original_source_sha256": "96d523790a28d90ffd89ea67603421a3881a96d21cb523495d1a066c4c5ceecd", "corpus_entry_id": "librispeech_5105-28241-0019"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.247161626815796, "predicted_mos": 3.247161626815796, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "9a975ab933be439cf9354ae2e4bb0361afc3e57c4e37a574497cd3764b94d55f", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.891810835000797, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5105-28241-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9a975ab933be439cf9354ae2e4bb0361afc3e57c4e37a574497cd3764b94d55f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "9b33cdb5987a1a118cb734b090e58c46dfc7460fe25e94984a7d20e8f55b8af7"}, "rows": [{"case": "librispeech_5105-28241-0019", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.9378697872161865, "predicted_mos": 1.9378697872161865, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "5209d662d177347f3f3b28e4783d656a94257ad76315dc2d3552de68b8169ce2", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.302158552000037, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5105-28241-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5209d662d177347f3f3b28e4783d656a94257ad76315dc2d3552de68b8169ce2", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3566486037260803, "全帯域_10ms_rms_envelope_nrmse": 0.09286043207115922, "全帯域_10ms_envelope_correlation": 0.9924844461007319, "全帯域_10ms_modulation_mae_db": 0.4043757642150103, "全帯域_40ms_log_rms_mae_db": 0.691460838308341, "全帯域_40ms_rms_envelope_nrmse": 0.05476591218205565, "全帯域_40ms_envelope_correlation": 0.9973565002146216, "全帯域_40ms_modulation_mae_db": 0.35126043510617216, "300–1000 Hz_10ms_log_rms_mae_db": 1.7183167751978057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12224591903553741, "300–1000 Hz_10ms_envelope_correlation": 0.9890163586197261, "300–1000 Hz_10ms_modulation_mae_db": 0.475815155347503, "300–1000 Hz_40ms_log_rms_mae_db": 1.026659639654823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08182557315366902, "300–1000 Hz_40ms_envelope_correlation": 0.9946929007458434, "300–1000 Hz_40ms_modulation_mae_db": 0.49064516581793677, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9083675830375146, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15326135203064972, "1000–3000 Hz_10ms_envelope_correlation": 0.9852780581620497, "1000–3000 Hz_10ms_modulation_mae_db": 0.3081340347387656, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1644297891121653, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09819634215800081, "1000–3000 Hz_40ms_envelope_correlation": 0.9951067319841093, "1000–3000 Hz_40ms_modulation_mae_db": 0.54274103613265, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1174301158085056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18304669305099386, "3000–6000 Hz_10ms_envelope_correlation": 0.978825631889856, "3000–6000 Hz_10ms_modulation_mae_db": 1.323539144730556, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3857747148418158, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10540472544297991, "3000–6000 Hz_40ms_envelope_correlation": 0.9926458764251231, "3000–6000 Hz_40ms_modulation_mae_db": 1.6875068910145103, "mr_stft_db": 5.961647346753815, "band_mae_db": 0.2057632754502882, "rms_gain_error_db": -0.16219803482764855, "output_peak": 0.5891324490673722}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4239354823464025, "全帯域_10ms_rms_envelope_nrmse": 0.09986212844402123, "全帯域_10ms_envelope_correlation": 0.9911112655038015, "全帯域_10ms_modulation_mae_db": 0.1136266863623494, "全帯域_40ms_log_rms_mae_db": 0.6678469427950439, "全帯域_40ms_rms_envelope_nrmse": 0.05369409241811392, "全帯域_40ms_envelope_correlation": 0.9972241052689883, "全帯域_40ms_modulation_mae_db": 0.08657531743531512, "300–1000 Hz_10ms_log_rms_mae_db": 1.7313624152248202, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1333938070260488, "300–1000 Hz_10ms_envelope_correlation": 0.986916466819981, "300–1000 Hz_10ms_modulation_mae_db": 0.7468673573355432, "300–1000 Hz_40ms_log_rms_mae_db": 1.0241132343607102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07403637955812181, "300–1000 Hz_40ms_envelope_correlation": 0.9957302842090394, "300–1000 Hz_40ms_modulation_mae_db": 0.39204756077353187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9401443412097308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1324723907101745, "1000–3000 Hz_10ms_envelope_correlation": 0.9882185055981898, "1000–3000 Hz_10ms_modulation_mae_db": 0.2413011144307702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1382924152965377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07771609625149974, "1000–3000 Hz_40ms_envelope_correlation": 0.9961518419107493, "1000–3000 Hz_40ms_modulation_mae_db": 0.4084005203059469, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1222588842827004, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19833881882665635, "3000–6000 Hz_10ms_envelope_correlation": 0.9762590551004335, "3000–6000 Hz_10ms_modulation_mae_db": 0.5749836831085782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.433235954071112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12224382060222243, "3000–6000 Hz_40ms_envelope_correlation": 0.9907273135293758, "3000–6000 Hz_40ms_modulation_mae_db": 0.6323924940748095, "mr_stft_db": 5.9541969712891385, "band_mae_db": 0.24828945734926497, "rms_gain_error_db": -0.07443389360586258, "output_peak": 0.6008184843216063}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4449598900339973, "全帯域_10ms_rms_envelope_nrmse": 0.10381065247689071, "全帯域_10ms_envelope_correlation": 0.9903743677489174, "全帯域_10ms_modulation_mae_db": 0.28325486183390103, "全帯域_40ms_log_rms_mae_db": 0.7001496325785945, "全帯域_40ms_rms_envelope_nrmse": 0.04661939384091266, "全帯域_40ms_envelope_correlation": 0.997907959561005, "全帯域_40ms_modulation_mae_db": 0.12993858687057244, "300–1000 Hz_10ms_log_rms_mae_db": 1.7424140244689337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14434310262396233, "300–1000 Hz_10ms_envelope_correlation": 0.9846480264207538, "300–1000 Hz_10ms_modulation_mae_db": 0.8612461789092608, "300–1000 Hz_40ms_log_rms_mae_db": 0.9851789654626113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060638906215097155, "300–1000 Hz_40ms_envelope_correlation": 0.9970966001615371, "300–1000 Hz_40ms_modulation_mae_db": 0.2727210492983651, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8507435637353837, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12307231631017689, "1000–3000 Hz_10ms_envelope_correlation": 0.9896366874604252, "1000–3000 Hz_10ms_modulation_mae_db": 0.8431451889839364, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1257600529109937, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0711030120470918, "1000–3000 Hz_40ms_envelope_correlation": 0.996507208215395, "1000–3000 Hz_40ms_modulation_mae_db": 1.1171872691169484, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1871016336657347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19446906087224838, "3000–6000 Hz_10ms_envelope_correlation": 0.9760392121845795, "3000–6000 Hz_10ms_modulation_mae_db": 1.000161547123883, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4145496608274282, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1000447800709791, "3000–6000 Hz_40ms_envelope_correlation": 0.9934393776129766, "3000–6000 Hz_40ms_modulation_mae_db": 1.3902540143398585, "mr_stft_db": 5.948154193626482, "band_mae_db": 0.14894246603905953, "rms_gain_error_db": -0.06224736020245601, "output_peak": 0.5767292206138716}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4343949203745883, "全帯域_10ms_rms_envelope_nrmse": 0.10619093758689532, "全帯域_10ms_envelope_correlation": 0.990200030308972, "全帯域_10ms_modulation_mae_db": 0.48606391756048994, "全帯域_40ms_log_rms_mae_db": 0.6958737234938657, "全帯域_40ms_rms_envelope_nrmse": 0.05877876057329231, "全帯域_40ms_envelope_correlation": 0.9970778390685131, "全帯域_40ms_modulation_mae_db": 0.6126418824834268, "300–1000 Hz_10ms_log_rms_mae_db": 1.7493698349445568, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1476222800135807, "300–1000 Hz_10ms_envelope_correlation": 0.9843136934222773, "300–1000 Hz_10ms_modulation_mae_db": 0.9409051243567775, "300–1000 Hz_40ms_log_rms_mae_db": 1.0301218705666044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09288241164747393, "300–1000 Hz_40ms_envelope_correlation": 0.9938045313814504, "300–1000 Hz_40ms_modulation_mae_db": 0.6534448578790025, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9266367599670273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1364280877821399, "1000–3000 Hz_10ms_envelope_correlation": 0.9873015141547935, "1000–3000 Hz_10ms_modulation_mae_db": 0.5965335789783747, "1000–3000 Hz_40ms_log_rms_mae_db": 1.117827264546478, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07137935111144955, "1000–3000 Hz_40ms_envelope_correlation": 0.9966629171214836, "1000–3000 Hz_40ms_modulation_mae_db": 0.7971656169844602, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1103893069651103, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19600798997195026, "3000–6000 Hz_10ms_envelope_correlation": 0.97579855235666, "3000–6000 Hz_10ms_modulation_mae_db": 1.168480785007567, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4375466977591755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10837827362711279, "3000–6000 Hz_40ms_envelope_correlation": 0.992658877832479, "3000–6000 Hz_40ms_modulation_mae_db": 1.419415109921494, "mr_stft_db": 5.942200993023003, "band_mae_db": 0.25513441639179163, "rms_gain_error_db": -0.20324507476609333, "output_peak": 0.5909888826781371}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3084373950084969, "全帯域_10ms_rms_envelope_nrmse": 0.10246750894585707, "全帯域_10ms_envelope_correlation": 0.9906395841401846, "全帯域_10ms_modulation_mae_db": 0.11672088364425048, "全帯域_40ms_log_rms_mae_db": 0.6095359538994842, "全帯域_40ms_rms_envelope_nrmse": 0.045081037139302865, "全帯域_40ms_envelope_correlation": 0.9980106839658323, "全帯域_40ms_modulation_mae_db": 0.2834406147218047, "300–1000 Hz_10ms_log_rms_mae_db": 1.7601902478633111, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1344847434107876, "300–1000 Hz_10ms_envelope_correlation": 0.9866898248466847, "300–1000 Hz_10ms_modulation_mae_db": 0.35135645865083726, "300–1000 Hz_40ms_log_rms_mae_db": 1.023813208647888, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05975342421289852, "300–1000 Hz_40ms_envelope_correlation": 0.9971727780663046, "300–1000 Hz_40ms_modulation_mae_db": 0.622697159741576, "1000–3000 Hz_10ms_log_rms_mae_db": 1.969946800748236, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15032668392192433, "1000–3000 Hz_10ms_envelope_correlation": 0.9842684909076873, "1000–3000 Hz_10ms_modulation_mae_db": 0.1776175102389591, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1831802748315252, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08309456662632768, "1000–3000 Hz_40ms_envelope_correlation": 0.9948917551762078, "1000–3000 Hz_40ms_modulation_mae_db": 0.28253373241099955, "3000–6000 Hz_10ms_log_rms_mae_db": 2.166756095893791, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1851871709510014, "3000–6000 Hz_10ms_envelope_correlation": 0.9788186859068639, "3000–6000 Hz_10ms_modulation_mae_db": 0.6268708205849141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4244043164106457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10612742892209416, "3000–6000 Hz_40ms_envelope_correlation": 0.9926727610188635, "3000–6000 Hz_40ms_modulation_mae_db": 0.9384090497864311, "mr_stft_db": 5.935621933976759, "band_mae_db": 0.20050012499702055, "rms_gain_error_db": -0.003446101136000762, "output_peak": 0.6084639356786987}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4539543515739952, "全帯域_10ms_rms_envelope_nrmse": 0.0989445707659574, "全帯域_10ms_envelope_correlation": 0.9912587902421119, "全帯域_10ms_modulation_mae_db": 0.20933781406819138, "全帯域_40ms_log_rms_mae_db": 0.6855347706417056, "全帯域_40ms_rms_envelope_nrmse": 0.04905042254197577, "全帯域_40ms_envelope_correlation": 0.9976647702733519, "全帯域_40ms_modulation_mae_db": 0.39978557810149445, "300–1000 Hz_10ms_log_rms_mae_db": 1.7918455545785401, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13572481295070715, "300–1000 Hz_10ms_envelope_correlation": 0.9864311254924089, "300–1000 Hz_10ms_modulation_mae_db": 0.6658333123275654, "300–1000 Hz_40ms_log_rms_mae_db": 1.0655922708953047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0664062696569225, "300–1000 Hz_40ms_envelope_correlation": 0.9965131314394097, "300–1000 Hz_40ms_modulation_mae_db": 0.3510369254702146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9575677791998625, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15043897258252692, "1000–3000 Hz_10ms_envelope_correlation": 0.9842562472942562, "1000–3000 Hz_10ms_modulation_mae_db": 0.8350123763861764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.160140807319133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09219347490896958, "1000–3000 Hz_40ms_envelope_correlation": 0.993737770276179, "1000–3000 Hz_40ms_modulation_mae_db": 1.308112984195768, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1087317689173903, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19604393407419995, "3000–6000 Hz_10ms_envelope_correlation": 0.9758184895732169, "3000–6000 Hz_10ms_modulation_mae_db": 0.4522725911091198, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4101300090908182, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11304749093823861, "3000–6000 Hz_40ms_envelope_correlation": 0.9920681528467141, "3000–6000 Hz_40ms_modulation_mae_db": 0.5986594408819202, "mr_stft_db": 5.952742100341347, "band_mae_db": 0.16707492607456068, "rms_gain_error_db": -0.05218927099938483, "output_peak": 0.5870524661776184}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3837885056651136, "全帯域_10ms_rms_envelope_nrmse": 0.10898105675050797, "全帯域_10ms_envelope_correlation": 0.9894392835065235, "全帯域_10ms_modulation_mae_db": 0.2377241433565113, "全帯域_40ms_log_rms_mae_db": 0.6342560695719983, "全帯域_40ms_rms_envelope_nrmse": 0.055150122881726146, "全帯域_40ms_envelope_correlation": 0.9970379511326242, "全帯域_40ms_modulation_mae_db": 0.23653684572476827, "300–1000 Hz_10ms_log_rms_mae_db": 1.6924085471619275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14620741377485444, "300–1000 Hz_10ms_envelope_correlation": 0.9846143226427484, "300–1000 Hz_10ms_modulation_mae_db": 0.6007788637034929, "300–1000 Hz_40ms_log_rms_mae_db": 1.0115691167320966, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07348029409791118, "300–1000 Hz_40ms_envelope_correlation": 0.9958546213138241, "300–1000 Hz_40ms_modulation_mae_db": 0.3271767992671048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9916670057232562, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1460348540779463, "1000–3000 Hz_10ms_envelope_correlation": 0.9856788087034284, "1000–3000 Hz_10ms_modulation_mae_db": 0.8281681974700196, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2289278808516577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09348443861435293, "1000–3000 Hz_40ms_envelope_correlation": 0.9943160967535886, "1000–3000 Hz_40ms_modulation_mae_db": 1.036651345608911, "3000–6000 Hz_10ms_log_rms_mae_db": 2.055894255070026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18789073428053937, "3000–6000 Hz_10ms_envelope_correlation": 0.9778720184628479, "3000–6000 Hz_10ms_modulation_mae_db": 0.45452025496445464, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3658936475348749, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1150810436940546, "3000–6000 Hz_40ms_envelope_correlation": 0.9916823124880566, "3000–6000 Hz_40ms_modulation_mae_db": 0.8351665552505275, "mr_stft_db": 5.943485144379568, "band_mae_db": 0.2488682456252662, "rms_gain_error_db": -0.018314037664383685, "output_peak": 0.5806859823308284}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4687661294623156, "全帯域_10ms_rms_envelope_nrmse": 0.1017935895740735, "全帯域_10ms_envelope_correlation": 0.9908114446097093, "全帯域_10ms_modulation_mae_db": 0.23879126776337947, "全帯域_40ms_log_rms_mae_db": 0.7100619642943156, "全帯域_40ms_rms_envelope_nrmse": 0.05374748969785206, "全帯域_40ms_envelope_correlation": 0.9971984275838426, "全帯域_40ms_modulation_mae_db": 0.19186677620495796, "300–1000 Hz_10ms_log_rms_mae_db": 1.6967451701867433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15430249675705857, "300–1000 Hz_10ms_envelope_correlation": 0.982496583078328, "300–1000 Hz_10ms_modulation_mae_db": 0.8563825860748903, "300–1000 Hz_40ms_log_rms_mae_db": 1.0008000931014494, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08885063528513472, "300–1000 Hz_40ms_envelope_correlation": 0.9937071208035514, "300–1000 Hz_40ms_modulation_mae_db": 0.43605753180951196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8778773110520561, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1447505481575462, "1000–3000 Hz_10ms_envelope_correlation": 0.9854852314618164, "1000–3000 Hz_10ms_modulation_mae_db": 0.1699759526185458, "1000–3000 Hz_40ms_log_rms_mae_db": 1.076786895281194, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08755556599287934, "1000–3000 Hz_40ms_envelope_correlation": 0.9943136592567573, "1000–3000 Hz_40ms_modulation_mae_db": 0.13456078350534711, "3000–6000 Hz_10ms_log_rms_mae_db": 2.114645144076755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19684731393571156, "3000–6000 Hz_10ms_envelope_correlation": 0.9755537941171696, "3000–6000 Hz_10ms_modulation_mae_db": 0.19289737383380642, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3781589439874051, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11183568573174264, "3000–6000 Hz_40ms_envelope_correlation": 0.9917110624699502, "3000–6000 Hz_40ms_modulation_mae_db": 0.35687217292581735, "mr_stft_db": 5.934641956918417, "band_mae_db": 0.12956416442546653, "rms_gain_error_db": -0.004704701691813916, "output_peak": 0.5769083842833976}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84640, "output_samples": 84640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.242937564265599, "spectral_convergence": 0.3912590405947108, "spectral_convergence_status": "ok", "amplitude_floor": 1.2121984197282631e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.703665000081431, "spectral_convergence": 0.34501825998511887, "spectral_convergence_status": "ok", "amplitude_floor": 7.081048851832571e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9383394759144155, "spectral_convergence": 0.3859306625385233, "spectral_convergence_status": "ok", "amplitude_floor": 5.009180364479501e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.291800237989852e-05, "output_power": 8.03470681097805e-05, "delta_db": 0.4213527138251336, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025616304806087414, "output_power": 0.00025244428872013504, "delta_db": -0.06350932538397144, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038484147909669214, "output_power": 0.00037649477855186925, "delta_db": -0.09522917612727215, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002720810642199278, "output_power": 0.0002373828556801652, "delta_db": -0.5924896759665212, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.5951933766776914e-05, "output_power": 4.632753475784208e-05, "delta_db": 0.03535398312186294, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.6368160234296276e-05, "output_power": 2.6530431098291388e-05, "delta_db": 0.02664477827696774, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2057632754502882, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16219803482764855, "pitch": null, "output_rms_dbfs": -29.916012867215468, "output_peak": 0.5891324490673722, "reference_rms_dbfs": -29.753814832387818, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.961647346753815}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00018085971712132593, "log_rms_mae_db": 1.3566486037260803, "rms_envelope_nrmse": 0.09286043207115922, "envelope_correlation": 0.9924844461007319, "rms_gain_error_db": -0.16225722183133715, "reference_mean_power": 0.0010593154039372696, "output_mean_power": 0.0010204684107523565}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 5.491358535179395e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002450078451267542, "output_power": 0.0002270538613658602, "reference_active": true, "delta_db": -0.3305109857328356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.385267752666162e-05, "output_power": 8.256188690366765e-05, "reference_active": true, "delta_db": -0.06737324816403259}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4530628910939782e-05, "output_power": 1.2043704904151693e-05, "reference_active": true, "delta_db": -0.8152430587481628}], "active_band_count": 3, "active_band_mae_db": 0.4043757642150103}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00014864066644098504, "log_rms_mae_db": 0.691460838308341, "rms_envelope_nrmse": 0.05476591218205565, "envelope_correlation": 0.9973565002146216, "rms_gain_error_db": -0.16226164690636918, "reference_mean_power": 0.0010653029350414017, "output_mean_power": 0.001026235322723554}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 4.840611714703517e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022860940218938906, "output_power": 0.00021083139751871716, "reference_active": true, "delta_db": -0.35158800444173843}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8392485325428115e-05, "output_power": 3.840813536850003e-05, "reference_active": true, "delta_db": 0.0017699667998368998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1188041388515455e-06, "output_power": 9.521639419582892e-07, "reference_active": true, "delta_db": -0.7004233340769412}], "active_band_count": 3, "active_band_mae_db": 0.35126043510617216}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013930286763496605, "log_rms_mae_db": 1.7183167751978057, "rms_envelope_nrmse": 0.12224591903553741, "envelope_correlation": 0.9890163586197261, "rms_gain_error_db": -0.08939352322762753, "reference_mean_power": 0.0005230599489297363, "output_mean_power": 0.0005124035320769044}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 3.2510596296193775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001486768204336917, "output_power": 0.00013348477840574577, "reference_active": true, "delta_db": -0.46811519939917523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.417268122003869e-05, "output_power": 5.3931280434587264e-05, "reference_active": true, "delta_db": -0.019395996539588886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.810050335491182e-06, "output_power": 6.290141364622518e-06, "reference_active": true, "delta_db": -0.9399342701037449}], "active_band_count": 3, "active_band_mae_db": 0.475815155347503}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00011121771745577188, "log_rms_mae_db": 1.026659639654823, "rms_envelope_nrmse": 0.08182557315366902, "envelope_correlation": 0.9946929007458434, "rms_gain_error_db": -0.08939500564283707, "reference_mean_power": 0.0005260459982805144, "output_mean_power": 0.0005153285700774844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.9069199937213013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014215463322126283, "output_power": 0.0001285698431204109, "reference_active": true, "delta_db": -0.4362190515901775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2639933776267858e-05, "output_power": 2.3728741115401437e-05, "reference_active": true, "delta_db": 0.20399545965921403}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.874281484444505e-07, "output_power": 5.676167631801259e-07, "reference_active": true, "delta_db": -0.8317209862044189}], "active_band_count": 3, "active_band_mae_db": 0.49064516581793677}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010000130422823669, "log_rms_mae_db": 1.9083675830375146, "rms_envelope_nrmse": 0.15326135203064972, "envelope_correlation": 0.9852780581620497, "rms_gain_error_db": -0.5230717019949453, "reference_mean_power": 0.0002857956004291321, "output_mean_power": 0.0002533660196286846}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.149273337334192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04716085544297e-05, "output_power": 8.685842188271054e-05, "reference_active": true, "delta_db": -0.17700377525183456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.048324869712114e-05, "output_power": 1.829654179694062e-05, "reference_active": true, "delta_db": -0.4902982590258489}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.217073445558843e-06, "output_power": 7.74474569276605e-06, "reference_active": true, "delta_db": -0.25710006993861345}], "active_band_count": 3, "active_band_mae_db": 0.3081340347387656}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 8.407177238322724e-05, "log_rms_mae_db": 1.1644297891121653, "rms_envelope_nrmse": 0.09819634215800081, "envelope_correlation": 0.9951067319841093, "rms_gain_error_db": -0.5230726779244722, "reference_mean_power": 0.00028742710472676696, "output_mean_power": 0.00025481233785225977}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 1.945560884297879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.432816729085121e-05, "output_power": 8.042744892171317e-05, "reference_active": true, "delta_db": -0.20568368347331387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0296492874768435e-05, "output_power": 9.188237922617914e-06, "reference_active": true, "delta_db": -0.4945709089432449}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.97290149260778e-07, "output_power": 4.823791197242096e-07, "reference_active": true, "delta_db": -0.9279685159813913}], "active_band_count": 3, "active_band_mae_db": 0.54274103613265}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 5.528232384026309e-05, "log_rms_mae_db": 2.1174301158085056, "rms_envelope_nrmse": 0.18304669305099386, "envelope_correlation": 0.978825631889856, "rms_gain_error_db": -0.13743180247638803, "reference_mean_power": 3.470766577035859e-05, "output_mean_power": 3.3626543363614444e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.6534989625639428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5332454345272327e-05, "output_power": 1.498832008034846e-05, "reference_active": true, "delta_db": -0.09858721165460409}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.828360450157823e-06, "output_power": 3.1551069568075175e-06, "reference_active": true, "delta_db": -0.8399873476101964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1450631921352025e-06, "output_power": 5.69672384521247e-07, "reference_active": true, "delta_db": -3.0320428749268675}], "active_band_count": 3, "active_band_mae_db": 1.323539144730556}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 4.431598350735027e-05, "log_rms_mae_db": 1.3857747148418158, "rms_envelope_nrmse": 0.10540472544297991, "envelope_correlation": 0.9926458764251231, "rms_gain_error_db": -0.13743364977760494, "reference_mean_power": 3.49057787168493e-05, "output_mean_power": 3.381847082895101e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.464962908732827e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.55740476492126e-05, "output_power": 1.4895116892548057e-05, "reference_active": true, "delta_db": -0.19357583316368282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.044495719000328e-06, "output_power": 1.5353171092570417e-06, "reference_active": true, "delta_db": -1.243881158802958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.011419545771505e-07, "output_power": 4.389600271494565e-08, "reference_active": true, "delta_db": -3.625063681076891}], "active_band_count": 3, "active_band_mae_db": 1.6875068910145103}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8350966710067954, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "052ffeb6d045615f12ea0b03e5199831f256dc58c523015031809ccd3c0d6471", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9862654429678308, "standard_deviation": 0.0019348558405364092, "minimum": 0.9842562472942562, "maximum": 0.9896366874604252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9278688930841332, "standard_deviation": 0.047362966092685074, "minimum": 1.8507435637353837, "maximum": 1.9916670057232562, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4999859942306935, "standard_deviation": 0.307891217515326, "minimum": 0.1699759526185458, "maximum": 0.8431451889839364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1420981506966356, "standard_deviation": 0.010496361420381397, "minimum": 0.12307231631017689, "maximum": 0.15326135203064972, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9952109975868086, "standard_deviation": 0.001106001142599322, "minimum": 0.993737770276179, "maximum": 0.9966629171214836, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1494181725187105, "standard_deviation": 0.04591727623577337, "minimum": 1.076786895281194, "maximum": 1.2289278808516577, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7034191610326289, "standard_deviation": 0.4259927245835673, "minimum": 0.13456078350534711, "maximum": 1.308112984195768, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08434035596382143, "standard_deviation": 0.010255725802784707, "minimum": 0.0711030120470918, "maximum": 0.09819634215800081, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9768731799489534, "standard_deviation": 0.0013977209601002307, "minimum": 0.9755537941171696, "maximum": 0.978825631889856, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1229009005850017, "standard_deviation": 0.039624498699839826, "minimum": 2.055894255070026, "maximum": 2.1871016336657347, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7242157750578598, "standard_deviation": 0.3952287293247102, "minimum": 0.19289737383380642, "maximum": 1.323539144730556, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19222896449541266, "standard_deviation": 0.00591890115666959, "minimum": 0.18304669305099386, "maximum": 0.19833881882665635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922007167779424, "standard_deviation": 0.0008334249277069318, "minimum": 0.9907273135293758, "maximum": 0.9934393776129766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4062117430654093, "standard_deviation": 0.02661420474174089, "minimum": 1.3658936475348749, "maximum": 1.4375466977591755, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9823344660244211, "standard_deviation": 0.4689140678243771, "minimum": 0.35687217292581735, "maximum": 1.6875068910145103, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11027040612867803, "standard_deviation": 0.006820660740518757, "minimum": 0.1000447800709791, "maximum": 0.12224382060222243, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9856408001678635, "standard_deviation": 0.0020154175890400475, "minimum": 0.982496583078328, "maximum": 0.9890163586197261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7353315712033297, "standard_deviation": 0.03315671001667628, "minimum": 1.6924085471619275, "maximum": 1.7918455545785401, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6873981295882338, "standard_deviation": 0.20432487240844996, "minimum": 0.35135645865083726, "maximum": 0.9409051243567775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13979057194906713, "standard_deviation": 0.01019277522457206, "minimum": 0.12224591903553741, "maximum": 0.15430249675705857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955714960151201, "standard_deviation": 0.0013759197628070978, "minimum": 0.9937071208035514, "maximum": 0.9971727780663046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.020981049927686, "standard_deviation": 0.023608226407479335, "minimum": 0.9851789654626113, "maximum": 1.0655922708953047, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44322838125715547, "standard_deviation": 0.13750804801042885, "minimum": 0.2727210492983651, "maximum": 0.6534448578790025, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0747342367284036, "standard_deviation": 0.01238336159991857, "minimum": 0.05975342421289852, "maximum": 0.09288241164747393, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2005171345440898, "standard_deviation": 0.0484668461594805, "minimum": 0.12956416442546653, "maximum": 0.25513441639179163, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.946586330038566, "standard_deviation": 0.009387243262291046, "minimum": 5.934641956918417, "maximum": 5.961647346753815, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5888474756439412, "standard_deviation": 0.011288448119126833, "minimum": 0.5767292206138716, "maximum": 0.6084639356786987, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07259730936170546, "standard_deviation": 0.0736420691974014, "minimum": -0.20324507476609333, "maximum": -0.003446101136000762, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9907899015201189, "standard_deviation": 0.0008914240275394178, "minimum": 0.9894392835065235, "maximum": 0.9924844461007319, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4093606597738737, "standard_deviation": 0.0550870134017568, "minimum": 1.3084373950084969, "maximum": 1.4687661294623156, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2612369173505104, "standard_deviation": 0.1298045444520049, "minimum": 0.1136266863623494, "maximum": 0.48606391756048994, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1018638595769203, "standard_deviation": 0.0048862535207479536, "minimum": 0.09286043207115922, "maximum": 0.10898105675050797, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974347796335974, "standard_deviation": 0.0003779103844852034, "minimum": 0.9970379511326242, "maximum": 0.9980106839658323, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6743399869479185, "standard_deviation": 0.03519900317159471, "minimum": 0.6095359538994842, "maximum": 0.7100619642943156, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.286505754581064, "standard_deviation": 0.16875110236878843, "minimum": 0.08657531743531512, "maximum": 0.6126418824834268, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05211090390940393, "standard_deviation": 0.0047036974531423055, "minimum": 0.045081037139302865, "maximum": 0.05877876057329231, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0019", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.0090103149414062, "predicted_mos": 2.0090103149414062, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "604624b029c85caa888feca07d597ba50688c5181181f810b451273e7125cee7", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.800594999920577, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.325664819124269, "全帯域_10ms_rms_envelope_nrmse": 0.090014366151423, "全帯域_10ms_envelope_correlation": 0.9929015358633975, "全帯域_10ms_modulation_mae_db": 0.3419415998156117, "全帯域_40ms_log_rms_mae_db": 0.682630896239164, "全帯域_40ms_rms_envelope_nrmse": 0.05403659539262486, "全帯域_40ms_envelope_correlation": 0.9973615627839932, "全帯域_40ms_modulation_mae_db": 0.26073545172775614, "300–1000 Hz_10ms_log_rms_mae_db": 1.8208297298212104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12096911690256937, "300–1000 Hz_10ms_envelope_correlation": 0.989237181188382, "300–1000 Hz_10ms_modulation_mae_db": 0.3592875959531328, "300–1000 Hz_40ms_log_rms_mae_db": 1.054333566820424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08087538355440897, "300–1000 Hz_40ms_envelope_correlation": 0.9947783476258113, "300–1000 Hz_40ms_modulation_mae_db": 0.33873594519594724, "1000–3000 Hz_10ms_log_rms_mae_db": 2.008781634148317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1472666732020107, "1000–3000 Hz_10ms_envelope_correlation": 0.9863022702990475, "1000–3000 Hz_10ms_modulation_mae_db": 0.30086447976998315, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2395079945437428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09441056955590309, "1000–3000 Hz_40ms_envelope_correlation": 0.9953052429703525, "1000–3000 Hz_40ms_modulation_mae_db": 0.48860548030915085, "3000–6000 Hz_10ms_log_rms_mae_db": 2.172369284397862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17664548127960958, "3000–6000 Hz_10ms_envelope_correlation": 0.9802950556110666, "3000–6000 Hz_10ms_modulation_mae_db": 1.206682324700904, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4157657766530711, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09241002381444935, "3000–6000 Hz_40ms_envelope_correlation": 0.9942699482168638, "3000–6000 Hz_40ms_modulation_mae_db": 1.620411080311892, "mr_stft_db": 5.978856307330408, "band_mae_db": 0.17227044028067184, "rms_gain_error_db": -0.1418669383108531, "output_peak": 0.5961068602179965}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4164176820103886, "全帯域_10ms_rms_envelope_nrmse": 0.09859246698876648, "全帯域_10ms_envelope_correlation": 0.9913277712313282, "全帯域_10ms_modulation_mae_db": 0.1901729708305636, "全帯域_40ms_log_rms_mae_db": 0.6639528340707962, "全帯域_40ms_rms_envelope_nrmse": 0.05146779684739289, "全帯域_40ms_envelope_correlation": 0.9974242463206182, "全帯域_40ms_modulation_mae_db": 0.050914054532586794, "300–1000 Hz_10ms_log_rms_mae_db": 1.8694137528772505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13008869147320895, "300–1000 Hz_10ms_envelope_correlation": 0.9875541447991685, "300–1000 Hz_10ms_modulation_mae_db": 0.8673842357269094, "300–1000 Hz_40ms_log_rms_mae_db": 1.0998737274609898, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06904163677057386, "300–1000 Hz_40ms_envelope_correlation": 0.996276209736864, "300–1000 Hz_40ms_modulation_mae_db": 0.47133492404987165, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0745913555580575, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14212315773634773, "1000–3000 Hz_10ms_envelope_correlation": 0.9861000875662249, "1000–3000 Hz_10ms_modulation_mae_db": 0.20280116340033014, "1000–3000 Hz_40ms_log_rms_mae_db": 1.22648425592468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0882604214723004, "1000–3000 Hz_40ms_envelope_correlation": 0.9944954556849718, "1000–3000 Hz_40ms_modulation_mae_db": 0.41971860339704864, "3000–6000 Hz_10ms_log_rms_mae_db": 2.196440747889404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19705481941906147, "3000–6000 Hz_10ms_envelope_correlation": 0.9769414665136825, "3000–6000 Hz_10ms_modulation_mae_db": 0.7315307125472362, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4897232468496031, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11858038128936732, "3000–6000 Hz_40ms_envelope_correlation": 0.9916394233379231, "3000–6000 Hz_40ms_modulation_mae_db": 0.8388675363896904, "mr_stft_db": 5.972467744710261, "band_mae_db": 0.24247840227101838, "rms_gain_error_db": -0.05043238215980442, "output_peak": 0.5963544581138865}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.434382043409638, "全帯域_10ms_rms_envelope_nrmse": 0.10417279448808867, "全帯域_10ms_envelope_correlation": 0.9903078342513587, "全帯域_10ms_modulation_mae_db": 0.2701694117392999, "全帯域_40ms_log_rms_mae_db": 0.6901482774699975, "全帯域_40ms_rms_envelope_nrmse": 0.04660915205813644, "全帯域_40ms_envelope_correlation": 0.9978804507791564, "全帯域_40ms_modulation_mae_db": 0.11177208205754381, "300–1000 Hz_10ms_log_rms_mae_db": 1.876848532049677, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14613799755426082, "300–1000 Hz_10ms_envelope_correlation": 0.9843109078669631, "300–1000 Hz_10ms_modulation_mae_db": 0.8733094295748627, "300–1000 Hz_40ms_log_rms_mae_db": 1.058661775413532, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06263022340980201, "300–1000 Hz_40ms_envelope_correlation": 0.9968699337958963, "300–1000 Hz_40ms_modulation_mae_db": 0.3993088603016071, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9662420240359153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1258548194812575, "1000–3000 Hz_10ms_envelope_correlation": 0.9890863825504868, "1000–3000 Hz_10ms_modulation_mae_db": 0.7087181658501692, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1908093018708943, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06749184646073758, "1000–3000 Hz_40ms_envelope_correlation": 0.9968071050592141, "1000–3000 Hz_40ms_modulation_mae_db": 1.006351825233435, "3000–6000 Hz_10ms_log_rms_mae_db": 2.256362563024736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19933091087535168, "3000–6000 Hz_10ms_envelope_correlation": 0.9747795765698527, "3000–6000 Hz_10ms_modulation_mae_db": 1.0648065917493283, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4675548088467878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10040656475427226, "3000–6000 Hz_40ms_envelope_correlation": 0.9934652532016323, "3000–6000 Hz_40ms_modulation_mae_db": 1.4816363699269246, "mr_stft_db": 5.974958087277375, "band_mae_db": 0.13065353094728113, "rms_gain_error_db": -0.036192336993119856, "output_peak": 0.5768570820254507}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4225211479462518, "全帯域_10ms_rms_envelope_nrmse": 0.1090790982921882, "全帯域_10ms_envelope_correlation": 0.9895753375023201, "全帯域_10ms_modulation_mae_db": 0.5426419725717465, "全帯域_40ms_log_rms_mae_db": 0.6862030722978708, "全帯域_40ms_rms_envelope_nrmse": 0.05870679140352369, "全帯域_40ms_envelope_correlation": 0.9970083526766401, "全帯域_40ms_modulation_mae_db": 0.4864023602633007, "300–1000 Hz_10ms_log_rms_mae_db": 1.856136851895172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14798925450292966, "300–1000 Hz_10ms_envelope_correlation": 0.984130177305469, "300–1000 Hz_10ms_modulation_mae_db": 1.0051815974621843, "300–1000 Hz_40ms_log_rms_mae_db": 1.081665804593955, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09126811317927461, "300–1000 Hz_40ms_envelope_correlation": 0.9939039227537818, "300–1000 Hz_40ms_modulation_mae_db": 0.6398624374461276, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0412896792517823, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13575293825938634, "1000–3000 Hz_10ms_envelope_correlation": 0.9874209832736005, "1000–3000 Hz_10ms_modulation_mae_db": 0.5648324611159327, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1684837826713181, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06446819401998266, "1000–3000 Hz_40ms_envelope_correlation": 0.9973617961727088, "1000–3000 Hz_40ms_modulation_mae_db": 0.7838190806892124, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1811836152701325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2017560656140403, "3000–6000 Hz_10ms_envelope_correlation": 0.9741640785140333, "3000–6000 Hz_10ms_modulation_mae_db": 1.182423162716002, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4992792166361524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10473935967744247, "3000–6000 Hz_40ms_envelope_correlation": 0.9928929625766324, "3000–6000 Hz_40ms_modulation_mae_db": 1.482018001261302, "mr_stft_db": 5.9562098201107245, "band_mae_db": 0.21198222138765688, "rms_gain_error_db": -0.1870253124637056, "output_peak": 0.6034428172580211}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2962470747293413, "全帯域_10ms_rms_envelope_nrmse": 0.10259330402120502, "全帯域_10ms_envelope_correlation": 0.9906122288321064, "全帯域_10ms_modulation_mae_db": 0.06319886563740511, "全帯域_40ms_log_rms_mae_db": 0.6168919050106597, "全帯域_40ms_rms_envelope_nrmse": 0.04670275494006829, "全帯域_40ms_envelope_correlation": 0.9978657809163656, "全帯域_40ms_modulation_mae_db": 0.22047473057079384, "300–1000 Hz_10ms_log_rms_mae_db": 1.8924258266243554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13425253603712756, "300–1000 Hz_10ms_envelope_correlation": 0.9867313204586124, "300–1000 Hz_10ms_modulation_mae_db": 0.28199914155051603, "300–1000 Hz_40ms_log_rms_mae_db": 1.0701353421135598, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05985528578262709, "300–1000 Hz_40ms_envelope_correlation": 0.9971565380766295, "300–1000 Hz_40ms_modulation_mae_db": 0.6003777453373976, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0804681444333504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15176384949620814, "1000–3000 Hz_10ms_envelope_correlation": 0.9839637182262885, "1000–3000 Hz_10ms_modulation_mae_db": 0.1032362745383985, "1000–3000 Hz_40ms_log_rms_mae_db": 1.272713734571645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08355396347296361, "1000–3000 Hz_40ms_envelope_correlation": 0.9948600265779576, "1000–3000 Hz_40ms_modulation_mae_db": 0.2905462521233431, "3000–6000 Hz_10ms_log_rms_mae_db": 2.239313288771434, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18974693649800906, "3000–6000 Hz_10ms_envelope_correlation": 0.9782816012726843, "3000–6000 Hz_10ms_modulation_mae_db": 0.5374321840457937, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4693198625443207, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1075755707211507, "3000–6000 Hz_40ms_envelope_correlation": 0.9928774321876026, "3000–6000 Hz_40ms_modulation_mae_db": 0.7414238488805255, "mr_stft_db": 5.961581684886645, "band_mae_db": 0.18838144533993986, "rms_gain_error_db": -0.007503683757050038, "output_peak": 0.6070983383653284}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4603160903246202, "全帯域_10ms_rms_envelope_nrmse": 0.09856464700989989, "全帯域_10ms_envelope_correlation": 0.9913316349100795, "全帯域_10ms_modulation_mae_db": 0.26834950795340656, "全帯域_40ms_log_rms_mae_db": 0.7000461259576051, "全帯域_40ms_rms_envelope_nrmse": 0.050541231640536866, "全帯域_40ms_envelope_correlation": 0.9975445614105378, "全帯域_40ms_modulation_mae_db": 0.4456294281967906, "300–1000 Hz_10ms_log_rms_mae_db": 1.9282086059955732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13031957354680546, "300–1000 Hz_10ms_envelope_correlation": 0.987507771647644, "300–1000 Hz_10ms_modulation_mae_db": 0.5591525637916697, "300–1000 Hz_40ms_log_rms_mae_db": 1.1626642093593131, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06734718242110391, "300–1000 Hz_40ms_envelope_correlation": 0.9964293579886982, "300–1000 Hz_40ms_modulation_mae_db": 0.2693577233719089, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0555734499995104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1541991278129027, "1000–3000 Hz_10ms_envelope_correlation": 0.9834667674410761, "1000–3000 Hz_10ms_modulation_mae_db": 0.8321290744342825, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2293285017448514, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08904371519869207, "1000–3000 Hz_40ms_envelope_correlation": 0.9942192834663497, "1000–3000 Hz_40ms_modulation_mae_db": 1.3195440014115147, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2002147123844833, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2023287080508649, "3000–6000 Hz_10ms_envelope_correlation": 0.9740135190034549, "3000–6000 Hz_10ms_modulation_mae_db": 0.6080915359758997, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4933454397054144, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11391008570937926, "3000–6000 Hz_40ms_envelope_correlation": 0.9915611244565233, "3000–6000 Hz_40ms_modulation_mae_db": 0.6146715833061177, "mr_stft_db": 5.975800457464783, "band_mae_db": 0.13692324977278253, "rms_gain_error_db": -0.07111133392865736, "output_peak": 0.5826609926044459}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3793180428469005, "全帯域_10ms_rms_envelope_nrmse": 0.1056214616854675, "全帯域_10ms_envelope_correlation": 0.9900723678970583, "全帯域_10ms_modulation_mae_db": 0.31528041166473175, "全帯域_40ms_log_rms_mae_db": 0.6239366537412279, "全帯域_40ms_rms_envelope_nrmse": 0.052173079200720736, "全帯域_40ms_envelope_correlation": 0.9973464942137842, "全帯域_40ms_modulation_mae_db": 0.3396944374386985, "300–1000 Hz_10ms_log_rms_mae_db": 1.8219844276083428, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14045255298982792, "300–1000 Hz_10ms_envelope_correlation": 0.9858811221815429, "300–1000 Hz_10ms_modulation_mae_db": 0.622452138082006, "300–1000 Hz_40ms_log_rms_mae_db": 1.0419706793534578, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06855200205081442, "300–1000 Hz_40ms_envelope_correlation": 0.9964603318790576, "300–1000 Hz_40ms_modulation_mae_db": 0.3110716417351349, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0867926181236776, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14938671129986872, "1000–3000 Hz_10ms_envelope_correlation": 0.9852211847283177, "1000–3000 Hz_10ms_modulation_mae_db": 0.7573364715036698, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2870949100864584, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09798213758172716, "1000–3000 Hz_40ms_envelope_correlation": 0.993979029734169, "1000–3000 Hz_40ms_modulation_mae_db": 1.0007778854497846, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1196424087108228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18112651087835283, "3000–6000 Hz_10ms_envelope_correlation": 0.9794251438569915, "3000–6000 Hz_10ms_modulation_mae_db": 0.4975965284661297, "3000–6000 Hz_40ms_log_rms_mae_db": 1.409801999798452, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10467846215178535, "3000–6000 Hz_40ms_envelope_correlation": 0.9931900813379965, "3000–6000 Hz_40ms_modulation_mae_db": 0.9551153509038612, "mr_stft_db": 5.957673514695831, "band_mae_db": 0.2509361727300668, "rms_gain_error_db": -0.0203122705441905, "output_peak": 0.5811766384463259}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4518853113453016, "全帯域_10ms_rms_envelope_nrmse": 0.10034789548801394, "全帯域_10ms_envelope_correlation": 0.9910593606377844, "全帯域_10ms_modulation_mae_db": 0.33889153632830654, "全帯域_40ms_log_rms_mae_db": 0.694882979773074, "全帯域_40ms_rms_envelope_nrmse": 0.05029942686212072, "全帯域_40ms_envelope_correlation": 0.9975390580000542, "全帯域_40ms_modulation_mae_db": 0.3172667475219852, "300–1000 Hz_10ms_log_rms_mae_db": 1.8346053979636492, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15157198547454917, "300–1000 Hz_10ms_envelope_correlation": 0.983067478610368, "300–1000 Hz_10ms_modulation_mae_db": 0.9543272287320659, "300–1000 Hz_40ms_log_rms_mae_db": 1.0581826930917801, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08422192083042107, "300–1000 Hz_40ms_envelope_correlation": 0.9943377820515876, "300–1000 Hz_40ms_modulation_mae_db": 0.614352953352832, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9940248856597973, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14704399149212302, "1000–3000 Hz_10ms_envelope_correlation": 0.9850926041679738, "1000–3000 Hz_10ms_modulation_mae_db": 0.2358995699604478, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1670600090872068, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08770494643797339, "1000–3000 Hz_40ms_envelope_correlation": 0.9943246216748892, "1000–3000 Hz_40ms_modulation_mae_db": 0.28052161319496804, "3000–6000 Hz_10ms_log_rms_mae_db": 2.206501103988978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19404668465848704, "3000–6000 Hz_10ms_envelope_correlation": 0.9762057620672904, "3000–6000 Hz_10ms_modulation_mae_db": 0.30326684485593974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4367235335916448, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10500368613265693, "3000–6000 Hz_40ms_envelope_correlation": 0.992753695947431, "3000–6000 Hz_40ms_modulation_mae_db": 0.5112568498339162, "mr_stft_db": 5.959269438535176, "band_mae_db": 0.10854060611931687, "rms_gain_error_db": -0.0014450403338264616, "output_peak": 0.5847565203036693}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84640, "output_samples": 84640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.27266666307828, "spectral_convergence": 0.391847588001708, "spectral_convergence_status": "ok", "amplitude_floor": 1.2121984197282631e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.737937173438016, "spectral_convergence": 0.34411879306765425, "spectral_convergence_status": "ok", "amplitude_floor": 7.081048851832571e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.925965085474927, "spectral_convergence": 0.3846540834357745, "spectral_convergence_status": "ok", "amplitude_floor": 5.009180364479501e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.291800237989852e-05, "output_power": 7.712759970437685e-05, "delta_db": 0.2437505339010338, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025616304806087414, "output_power": 0.00025470561678880485, "delta_db": -0.02477960070460424, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038484147909669214, "output_power": 0.00037918809821535103, "delta_db": -0.06427177702205676, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002720810642199278, "output_power": 0.00024050205022403454, "delta_db": -0.5357953464563445, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.5951933766776914e-05, "output_power": 4.5605000168395755e-05, "delta_db": -0.03291330585989098, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.6368160234296276e-05, "output_power": 2.7182602661605785e-05, "delta_db": 0.1321120777401008, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17227044028067184, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1418669383108531, "pitch": null, "output_rms_dbfs": -29.895681770698673, "output_peak": 0.5961068602179965, "reference_rms_dbfs": -29.753814832387818, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.978856307330408}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00018085971712132593, "log_rms_mae_db": 1.325664819124269, "rms_envelope_nrmse": 0.090014366151423, "envelope_correlation": 0.9929015358633975, "rms_gain_error_db": -0.14193414468395113, "reference_mean_power": 0.0010593154039372696, "output_mean_power": 0.001025254946094942}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 5.491358535179395e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002450078451267542, "output_power": 0.00023144660545803303, "reference_active": true, "delta_db": -0.24729175124537303}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.385267752666162e-05, "output_power": 8.391056420852085e-05, "reference_active": true, "delta_db": 0.0029970650596452767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4530628910939782e-05, "output_power": 1.2154324117008429e-05, "reference_active": true, "delta_db": -0.7755359831418167}], "active_band_count": 3, "active_band_mae_db": 0.3419415998156117}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00014864066644098504, "log_rms_mae_db": 0.682630896239164, "rms_envelope_nrmse": 0.05403659539262486, "envelope_correlation": 0.9973615627839932, "rms_gain_error_db": -0.14194544652632085, "reference_mean_power": 0.0010653029350414017, "output_mean_power": 0.0010310472753296495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 4.840611714703517e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022860940218938906, "output_power": 0.00021523026495345448, "reference_active": true, "delta_db": -0.2619074766693237}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8392485325428115e-05, "output_power": 3.889503373062388e-05, "reference_active": true, "delta_db": 0.056479255111470804}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1188041388515455e-06, "output_power": 1.0054769103719108e-06, "reference_active": true, "delta_db": -0.46381962340247396}], "active_band_count": 3, "active_band_mae_db": 0.26073545172775614}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013930286763496605, "log_rms_mae_db": 1.8208297298212104, "rms_envelope_nrmse": 0.12096911690256937, "envelope_correlation": 0.989237181188382, "rms_gain_error_db": -0.05315032979274686, "reference_mean_power": 0.0005230599489297363, "output_mean_power": 0.0005166975878436798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 3.2510596296193775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001486768204336917, "output_power": 0.0001361400375310166, "reference_active": true, "delta_db": -0.3825739881219543}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.417268122003869e-05, "output_power": 5.481700423619334e-05, "reference_active": true, "delta_db": 0.051349664744527346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.810050335491182e-06, "output_power": 6.733795443003085e-06, "reference_active": true, "delta_db": -0.6439391349929169}], "active_band_count": 3, "active_band_mae_db": 0.3592875959531328}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00011121771745577188, "log_rms_mae_db": 1.054333566820424, "rms_envelope_nrmse": 0.08087538355440897, "envelope_correlation": 0.9947783476258113, "rms_gain_error_db": -0.05315143425003218, "reference_mean_power": 0.0005260459982805144, "output_mean_power": 0.0005196471835382964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.9069199937213013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014215463322126283, "output_power": 0.0001314579847157547, "reference_active": true, "delta_db": -0.3397404899090385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2639933776267858e-05, "output_power": 2.398027669222911e-05, "reference_active": true, "delta_db": 0.24979037652883326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.874281484444505e-07, "output_power": 6.23102673295697e-07, "reference_active": true, "delta_db": -0.4266769691499698}], "active_band_count": 3, "active_band_mae_db": 0.33873594519594724}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010000130422823669, "log_rms_mae_db": 2.008781634148317, "rms_envelope_nrmse": 0.1472666732020107, "envelope_correlation": 0.9863022702990475, "rms_gain_error_db": -0.48300462210254047, "reference_mean_power": 0.0002857956004291321, "output_mean_power": 0.00025571433620453067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.149273337334192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04716085544297e-05, "output_power": 8.740345293544965e-05, "reference_active": true, "delta_db": -0.1498372190842607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.048324869712114e-05, "output_power": 1.8736074968437255e-05, "reference_active": true, "delta_db": -0.38720222544647515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.217073445558843e-06, "output_power": 7.5537354254623236e-06, "reference_active": true, "delta_db": -0.3655539947792137}], "active_band_count": 3, "active_band_mae_db": 0.30086447976998315}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 8.407177238322724e-05, "log_rms_mae_db": 1.2395079945437428, "rms_envelope_nrmse": 0.09441056955590309, "envelope_correlation": 0.9953052429703525, "rms_gain_error_db": -0.483005286814954, "reference_mean_power": 0.00028742710472676696, "output_mean_power": 0.0002571740780213517}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 1.945560884297879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.432816729085121e-05, "output_power": 8.103043032546e-05, "reference_active": true, "delta_db": -0.17324516575794502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0296492874768435e-05, "output_power": 9.51005353921312e-06, "reference_active": true, "delta_db": -0.34506361383804185}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.97290149260778e-07, "output_power": 4.802137445070038e-07, "reference_active": true, "delta_db": -0.9475076613314658}], "active_band_count": 3, "active_band_mae_db": 0.48860548030915085}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 5.528232384026309e-05, "log_rms_mae_db": 2.172369284397862, "rms_envelope_nrmse": 0.17664548127960958, "envelope_correlation": 0.9802950556110666, "rms_gain_error_db": -0.05125826044355436, "reference_mean_power": 3.470766577035859e-05, "output_mean_power": 3.4300431267448226e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.6534989625639428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5332454345272327e-05, "output_power": 1.5393623243293162e-05, "reference_active": true, "delta_db": 0.017291728755616048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.828360450157823e-06, "output_power": 3.352198795156123e-06, "reference_active": true, "delta_db": -0.5768305534203687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1450631921352025e-06, "output_power": 5.704754838516476e-07, "reference_active": true, "delta_db": -3.0259246919267273}], "active_band_count": 3, "active_band_mae_db": 1.206682324700904}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 4.431598350735027e-05, "log_rms_mae_db": 1.4157657766530711, "rms_envelope_nrmse": 0.09241002381444935, "envelope_correlation": 0.9942699482168638, "rms_gain_error_db": -0.05125937924968727, "reference_mean_power": 3.49057787168493e-05, "output_mean_power": 3.449621081400741e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.464962908732827e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.55740476492126e-05, "output_power": 1.5327154092669635e-05, "reference_active": true, "delta_db": -0.06939975462316951}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.044495719000328e-06, "output_power": 1.6429200928092477e-06, "reference_active": true, "delta_db": -0.9496976447621418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.011419545771505e-07, "output_power": 4.1755891266644756e-08, "reference_active": true, "delta_db": -3.842135841550365}], "active_band_count": 3, "active_band_mae_db": 1.620411080311892}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8324094816236002, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "805a7a07f40152931455e3e56871d284be559dfed47d2b85f752733ba96e449d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9858317497816269, "standard_deviation": 0.0018288755937330172, "minimum": 0.9834667674410761, "maximum": 0.9890863825504868, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.038470473901301, "standard_deviation": 0.044365041327958786, "minimum": 1.9662420240359153, "maximum": 2.0867926181236776, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4632272075716517, "standard_deviation": 0.2850145337646674, "minimum": 0.1032362745383985, "maximum": 0.8321290744342825, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1441739085975131, "standard_deviation": 0.009364030347758899, "minimum": 0.1258548194812575, "maximum": 0.1541991278129027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951690701675766, "standard_deviation": 0.0012587066690657924, "minimum": 0.993979029734169, "maximum": 0.9973617961727088, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2226853113125995, "standard_deviation": 0.0447869019612742, "minimum": 1.1670600090872068, "maximum": 1.2870949100864584, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6987355927260571, "standard_deviation": 0.38582493962378106, "minimum": 0.28052161319496804, "maximum": 1.3195440014115147, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08411447427503499, "standard_deviation": 0.012043384098993621, "minimum": 0.06446819401998266, "maximum": 0.09798213758172716, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9767632754261321, "standard_deviation": 0.0024041317864846785, "minimum": 0.9740135190034549, "maximum": 0.9802950556110666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1965034655547315, "standard_deviation": 0.0418403447585245, "minimum": 2.1196424087108228, "maximum": 2.256362563024736, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7664787356321541, "standard_deviation": 0.34255793361950043, "minimum": 0.30326684485593974, "maximum": 1.206682324700904, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1927545146592221, "standard_deviation": 0.00956079319010698, "minimum": 0.17664548127960958, "maximum": 0.2023287080508649, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9928312401578256, "standard_deviation": 0.0008981089739403903, "minimum": 0.9915611244565233, "maximum": 0.9942699482168638, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4601892355781807, "standard_deviation": 0.0352545736313616, "minimum": 1.409801999798452, "maximum": 1.4992792166361524, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0306750776017788, "standard_deviation": 0.43488667750355176, "minimum": 0.5112568498339162, "maximum": 1.620411080311892, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10591301678131296, "standard_deviation": 0.007960255128463772, "minimum": 0.09241002381444935, "maximum": 0.11858038128936732, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9860525130072687, "standard_deviation": 0.0020922965820088764, "minimum": 0.983067478610368, "maximum": 0.989237181188382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8625566406044038, "standard_deviation": 0.03715479624136898, "minimum": 1.8208297298212104, "maximum": 1.9282086059955732, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6903867413591683, "standard_deviation": 0.2756275438834075, "minimum": 0.28199914155051603, "maximum": 1.0051815974621843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13772271356015986, "standard_deviation": 0.010564652193513536, "minimum": 0.12096911690256937, "maximum": 0.15157198547454917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9957765529885407, "standard_deviation": 0.0012430621746356272, "minimum": 0.9939039227537818, "maximum": 0.9971565380766295, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0784359747758763, "standard_deviation": 0.038443210924634236, "minimum": 1.0419706793534578, "maximum": 1.1626642093593131, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45555027884885335, "standard_deviation": 0.14772747172424702, "minimum": 0.2693577233719089, "maximum": 0.6398624374461276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07297396849987825, "standard_deviation": 0.011144474360811051, "minimum": 0.05985528578262709, "maximum": 0.09126811317927461, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1802707586060918, "standard_deviation": 0.0527594100833216, "minimum": 0.10854060611931687, "maximum": 0.2509361727300668, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.9671021318764, "standard_deviation": 0.009286665113198339, "minimum": 5.9562098201107245, "maximum": 5.978856307330408, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5910567134168905, "standard_deviation": 0.011172321380715956, "minimum": 0.5768570820254507, "maximum": 0.6070983383653284, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06448616231140092, "standard_deviation": 0.06678118708270325, "minimum": -0.1870253124637056, "maximum": -0.0014450403338264616, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9908985088906792, "standard_deviation": 0.0010193834908239269, "minimum": 0.9895753375023201, "maximum": 0.9929015358633975, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3983440264670888, "standard_deviation": 0.05974371164144822, "minimum": 1.2962470747293413, "maximum": 1.4603160903246202, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29133078456763395, "standard_deviation": 0.1372210917106745, "minimum": 0.06319886563740511, "maximum": 0.5426419725717465, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10112325426563158, "standard_deviation": 0.0057563629106887, "minimum": 0.090014366151423, "maximum": 0.1090790982921882, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974963133876438, "standard_deviation": 0.00028588848765465416, "minimum": 0.9970083526766401, "maximum": 0.9978804507791564, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6698365930700494, "standard_deviation": 0.03235654221169333, "minimum": 0.6168919050106597, "maximum": 0.7000461259576051, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.279111161538682, "standard_deviation": 0.15103279737256234, "minimum": 0.050914054532586794, "maximum": 0.4864023602633007, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05131710354314056, "standard_deviation": 0.003925251377642502, "minimum": 0.04660915205813644, "maximum": 0.05870679140352369, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0019", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.6715872287750244, "predicted_mos": 1.6715872287750244, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "5b622f5fad1c8ce3da2291098e0fc725a85e9e83368349f576f81a52b82f8c35", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8106643999926746, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4151558780168831, "全帯域_10ms_rms_envelope_nrmse": 0.09803711880646994, "全帯域_10ms_envelope_correlation": 0.9914226939181526, "全帯域_10ms_modulation_mae_db": 0.12269862353213674, "全帯域_40ms_log_rms_mae_db": 0.6558860025927993, "全帯域_40ms_rms_envelope_nrmse": 0.04671737472364879, "全帯域_40ms_envelope_correlation": 0.9978733639996272, "全帯域_40ms_modulation_mae_db": 0.04717910516430898, "300–1000 Hz_10ms_log_rms_mae_db": 2.12443090311934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1352015690138884, "300–1000 Hz_10ms_envelope_correlation": 0.9865487090381698, "300–1000 Hz_10ms_modulation_mae_db": 0.6311362794733014, "300–1000 Hz_40ms_log_rms_mae_db": 1.2817773488713824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059779514979258305, "300–1000 Hz_40ms_envelope_correlation": 0.9972521751305929, "300–1000 Hz_40ms_modulation_mae_db": 0.3330049869177876, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3811279174804607, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16674651984979313, "1000–3000 Hz_10ms_envelope_correlation": 0.9806449072261316, "1000–3000 Hz_10ms_modulation_mae_db": 0.43661514079373714, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4090244916017698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0828548670117455, "1000–3000 Hz_40ms_envelope_correlation": 0.994902914653883, "1000–3000 Hz_40ms_modulation_mae_db": 0.8731228048002523, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7831354635727905, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20598921791922797, "3000–6000 Hz_10ms_envelope_correlation": 0.9733002197698529, "3000–6000 Hz_10ms_modulation_mae_db": 0.6844330922825095, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7874653575560049, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12169779593777193, "3000–6000 Hz_40ms_envelope_correlation": 0.9901998372246528, "3000–6000 Hz_40ms_modulation_mae_db": 1.030962526137655, "mr_stft_db": 6.259432695312647, "band_mae_db": 0.1825303847044334, "rms_gain_error_db": -0.03910961498299732, "output_peak": 0.5611377248890362}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3880145316709362, "全帯域_10ms_rms_envelope_nrmse": 0.10195564972511113, "全帯域_10ms_envelope_correlation": 0.9907285487175235, "全帯域_10ms_modulation_mae_db": 0.23642743868320623, "全帯域_40ms_log_rms_mae_db": 0.6719365379056967, "全帯域_40ms_rms_envelope_nrmse": 0.052820070234938475, "全帯域_40ms_envelope_correlation": 0.9972739689553001, "全帯域_40ms_modulation_mae_db": 0.3623094827107167, "300–1000 Hz_10ms_log_rms_mae_db": 2.0943129386355714, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14399091197965358, "300–1000 Hz_10ms_envelope_correlation": 0.9847448772491164, "300–1000 Hz_10ms_modulation_mae_db": 0.12225571779232682, "300–1000 Hz_40ms_log_rms_mae_db": 1.208694292822657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08487836406857029, "300–1000 Hz_40ms_envelope_correlation": 0.9943162840810463, "300–1000 Hz_40ms_modulation_mae_db": 0.43393144764604213, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4392955523535464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14569552249242865, "1000–3000 Hz_10ms_envelope_correlation": 0.9852880180323856, "1000–3000 Hz_10ms_modulation_mae_db": 0.5915942188996076, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4624842615502556, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07848530838024931, "1000–3000 Hz_40ms_envelope_correlation": 0.9954557636131286, "1000–3000 Hz_40ms_modulation_mae_db": 0.8706057643411896, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8055190205712908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22737572831732594, "3000–6000 Hz_10ms_envelope_correlation": 0.9693803092583936, "3000–6000 Hz_10ms_modulation_mae_db": 0.9541047044936554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7972065123756382, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12931146877575547, "3000–6000 Hz_40ms_envelope_correlation": 0.990036162092186, "3000–6000 Hz_40ms_modulation_mae_db": 1.2121071150692286, "mr_stft_db": 6.2489992741914095, "band_mae_db": 0.1573353003517415, "rms_gain_error_db": -0.0294840811096279, "output_peak": 0.561723190477601}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3775072987817485, "全帯域_10ms_rms_envelope_nrmse": 0.09349204064980489, "全帯域_10ms_envelope_correlation": 0.992329731096494, "全帯域_10ms_modulation_mae_db": 0.2591378630931242, "全帯域_40ms_log_rms_mae_db": 0.6479431023824621, "全帯域_40ms_rms_envelope_nrmse": 0.05334560206236303, "全帯域_40ms_envelope_correlation": 0.997456316042976, "全帯域_40ms_modulation_mae_db": 0.3384311562644551, "300–1000 Hz_10ms_log_rms_mae_db": 2.083408169139897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1268095961093429, "300–1000 Hz_10ms_envelope_correlation": 0.988333470504127, "300–1000 Hz_10ms_modulation_mae_db": 0.39088332067937265, "300–1000 Hz_40ms_log_rms_mae_db": 1.1648052520347658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0796263814647086, "300–1000 Hz_40ms_envelope_correlation": 0.9952314307110264, "300–1000 Hz_40ms_modulation_mae_db": 0.4376922787382127, "1000–3000 Hz_10ms_log_rms_mae_db": 2.401531239502634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1472455337182943, "1000–3000 Hz_10ms_envelope_correlation": 0.9854692118942603, "1000–3000 Hz_10ms_modulation_mae_db": 0.5381303173076201, "1000–3000 Hz_40ms_log_rms_mae_db": 1.37594245569468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08134838967837019, "1000–3000 Hz_40ms_envelope_correlation": 0.9959832764953348, "1000–3000 Hz_40ms_modulation_mae_db": 0.6445330179865937, "3000–6000 Hz_10ms_log_rms_mae_db": 2.761922326725561, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22958220285966952, "3000–6000 Hz_10ms_envelope_correlation": 0.9668297630818068, "3000–6000 Hz_10ms_modulation_mae_db": 1.3623151710076176, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8215461376976312, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14450643976398517, "3000–6000 Hz_40ms_envelope_correlation": 0.9866832561603665, "3000–6000 Hz_40ms_modulation_mae_db": 1.7482156392063384, "mr_stft_db": 6.260628950388077, "band_mae_db": 0.31444122517575335, "rms_gain_error_db": -0.13628850418536387, "output_peak": 0.5818653100170328}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4346126631186455, "全帯域_10ms_rms_envelope_nrmse": 0.10370982758530102, "全帯域_10ms_envelope_correlation": 0.990413107290628, "全帯域_10ms_modulation_mae_db": 0.35080983010052086, "全帯域_40ms_log_rms_mae_db": 0.7192013335200999, "全帯域_40ms_rms_envelope_nrmse": 0.04658911305648317, "全帯域_40ms_envelope_correlation": 0.9979678118675064, "全帯域_40ms_modulation_mae_db": 0.16468864579766787, "300–1000 Hz_10ms_log_rms_mae_db": 2.140998477536958, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14774620359447477, "300–1000 Hz_10ms_envelope_correlation": 0.9839200941787953, "300–1000 Hz_10ms_modulation_mae_db": 0.5550647953595561, "300–1000 Hz_40ms_log_rms_mae_db": 1.336212133146321, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07541355473747222, "300–1000 Hz_40ms_envelope_correlation": 0.9955712577012211, "300–1000 Hz_40ms_modulation_mae_db": 0.2089948110946147, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3889395737485346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1612235142173485, "1000–3000 Hz_10ms_envelope_correlation": 0.9821902279645606, "1000–3000 Hz_10ms_modulation_mae_db": 0.6969997776513014, "1000–3000 Hz_40ms_log_rms_mae_db": 1.388951038818792, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09135586067638646, "1000–3000 Hz_40ms_envelope_correlation": 0.9943815835279041, "1000–3000 Hz_40ms_modulation_mae_db": 0.7567593005485427, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7819711155477007, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22887175217814648, "3000–6000 Hz_10ms_envelope_correlation": 0.9680377024766227, "3000–6000 Hz_10ms_modulation_mae_db": 0.3736852140703533, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8565879795672393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13517033418987368, "3000–6000 Hz_40ms_envelope_correlation": 0.9883457249011678, "3000–6000 Hz_40ms_modulation_mae_db": 0.7648376816500289, "mr_stft_db": 6.261704436461464, "band_mae_db": 0.21341842196150762, "rms_gain_error_db": -0.09351801025679798, "output_peak": 0.5656436381643252}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4135990325392174, "全帯域_10ms_rms_envelope_nrmse": 0.09279211188650757, "全帯域_10ms_envelope_correlation": 0.9923765370363093, "全帯域_10ms_modulation_mae_db": 0.5876118144038013, "全帯域_40ms_log_rms_mae_db": 0.671731584400957, "全帯域_40ms_rms_envelope_nrmse": 0.048418318582619434, "全帯域_40ms_envelope_correlation": 0.9978349174741623, "全帯域_40ms_modulation_mae_db": 0.6101631021182179, "300–1000 Hz_10ms_log_rms_mae_db": 2.136085826061213, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1304784120549054, "300–1000 Hz_10ms_envelope_correlation": 0.987840734474843, "300–1000 Hz_10ms_modulation_mae_db": 0.8819939366670155, "300–1000 Hz_40ms_log_rms_mae_db": 1.2421965052102595, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07661024554368934, "300–1000 Hz_40ms_envelope_correlation": 0.9959343281308024, "300–1000 Hz_40ms_modulation_mae_db": 0.6373614278187906, "1000–3000 Hz_10ms_log_rms_mae_db": 2.434726583462971, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15485880057344042, "1000–3000 Hz_10ms_envelope_correlation": 0.983305280687409, "1000–3000 Hz_10ms_modulation_mae_db": 1.1131373682012917, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4658306455600685, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08754118495962321, "1000–3000 Hz_40ms_envelope_correlation": 0.9943608682634095, "1000–3000 Hz_40ms_modulation_mae_db": 1.4426060205673326, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7605763581310883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22937162589924645, "3000–6000 Hz_10ms_envelope_correlation": 0.9686061760299731, "3000–6000 Hz_10ms_modulation_mae_db": 0.5226813267191629, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8041622399562647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13654857748808522, "3000–6000 Hz_40ms_envelope_correlation": 0.9886145330235095, "3000–6000 Hz_40ms_modulation_mae_db": 0.826217137180039, "mr_stft_db": 6.26028730221083, "band_mae_db": 0.241649710089358, "rms_gain_error_db": -0.11137610525355437, "output_peak": 0.5835263889047719}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3226535287142673, "全帯域_10ms_rms_envelope_nrmse": 0.10275545154342199, "全帯域_10ms_envelope_correlation": 0.990607175017963, "全帯域_10ms_modulation_mae_db": 0.3194314808498418, "全帯域_40ms_log_rms_mae_db": 0.6743360318724814, "全帯域_40ms_rms_envelope_nrmse": 0.046477238171276546, "全帯域_40ms_envelope_correlation": 0.9979175840083087, "全帯域_40ms_modulation_mae_db": 0.39386083586968784, "300–1000 Hz_10ms_log_rms_mae_db": 2.134840081872623, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14749961695724664, "300–1000 Hz_10ms_envelope_correlation": 0.9841774054091157, "300–1000 Hz_10ms_modulation_mae_db": 0.09794140789877413, "300–1000 Hz_40ms_log_rms_mae_db": 1.2993718616686734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08527469251347136, "300–1000 Hz_40ms_envelope_correlation": 0.9943420831262868, "300–1000 Hz_40ms_modulation_mae_db": 0.42803845408219443, "1000–3000 Hz_10ms_log_rms_mae_db": 2.425561347924383, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16724799360201037, "1000–3000 Hz_10ms_envelope_correlation": 0.980712637982845, "1000–3000 Hz_10ms_modulation_mae_db": 1.005343477094413, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4568533315718928, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09028880606794595, "1000–3000 Hz_40ms_envelope_correlation": 0.9944460752575083, "1000–3000 Hz_40ms_modulation_mae_db": 1.4622213456304187, "3000–6000 Hz_10ms_log_rms_mae_db": 2.746073653067085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20282700599724113, "3000–6000 Hz_10ms_envelope_correlation": 0.9749206929115363, "3000–6000 Hz_10ms_modulation_mae_db": 0.4817239282925678, "3000–6000 Hz_40ms_log_rms_mae_db": 1.810947588483274, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1263278876393574, "3000–6000 Hz_40ms_envelope_correlation": 0.9897908551612357, "3000–6000 Hz_40ms_modulation_mae_db": 0.8044094778768646, "mr_stft_db": 6.2515278768074865, "band_mae_db": 0.2155768367947478, "rms_gain_error_db": -0.020838974244791874, "output_peak": 0.6067891023251423}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4066173650669254, "全帯域_10ms_rms_envelope_nrmse": 0.10073119134916578, "全帯域_10ms_envelope_correlation": 0.9909665298826095, "全帯域_10ms_modulation_mae_db": 0.0782297327271438, "全帯域_40ms_log_rms_mae_db": 0.754574784564877, "全帯域_40ms_rms_envelope_nrmse": 0.051534726484380323, "全帯域_40ms_envelope_correlation": 0.9974053030688739, "全帯域_40ms_modulation_mae_db": 0.130345997952051, "300–1000 Hz_10ms_log_rms_mae_db": 2.1919294845880803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1415403507016725, "300–1000 Hz_10ms_envelope_correlation": 0.9852419632537281, "300–1000 Hz_10ms_modulation_mae_db": 0.29891795636070445, "300–1000 Hz_40ms_log_rms_mae_db": 1.23556975009326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07776105506150068, "300–1000 Hz_40ms_envelope_correlation": 0.9951879067534074, "300–1000 Hz_40ms_modulation_mae_db": 0.09346803378868751, "1000–3000 Hz_10ms_log_rms_mae_db": 2.365410193663836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13851442388054555, "1000–3000 Hz_10ms_envelope_correlation": 0.986673856503978, "1000–3000 Hz_10ms_modulation_mae_db": 0.48036267857605947, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3927447230204077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07537215987101396, "1000–3000 Hz_40ms_envelope_correlation": 0.9958630756174189, "1000–3000 Hz_40ms_modulation_mae_db": 0.7096428663545146, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7730162529648577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22318902636071994, "3000–6000 Hz_10ms_envelope_correlation": 0.969529066017768, "3000–6000 Hz_10ms_modulation_mae_db": 0.7879126616244366, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8094728137755238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1440650020985003, "3000–6000 Hz_40ms_envelope_correlation": 0.986626684349118, "3000–6000 Hz_40ms_modulation_mae_db": 1.1393392745755717, "mr_stft_db": 6.254041835139152, "band_mae_db": 0.21291558156943502, "rms_gain_error_db": -0.022875200925767093, "output_peak": 0.5889542419457207}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3709758647804826, "全帯域_10ms_rms_envelope_nrmse": 0.09133803209969395, "全帯域_10ms_envelope_correlation": 0.9928052119408051, "全帯域_10ms_modulation_mae_db": 0.31167909091965385, "全帯域_40ms_log_rms_mae_db": 0.6901920339190538, "全帯域_40ms_rms_envelope_nrmse": 0.049118947207019904, "全帯域_40ms_envelope_correlation": 0.9980323736402507, "全帯域_40ms_modulation_mae_db": 0.27020196602883867, "300–1000 Hz_10ms_log_rms_mae_db": 2.083235269763939, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12925364242804643, "300–1000 Hz_10ms_envelope_correlation": 0.9881400058614426, "300–1000 Hz_10ms_modulation_mae_db": 0.549265637646979, "300–1000 Hz_40ms_log_rms_mae_db": 1.2240987240835501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08132206669924119, "300–1000 Hz_40ms_envelope_correlation": 0.9954011799842661, "300–1000 Hz_40ms_modulation_mae_db": 0.6167483746222272, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4080729491296387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15393348726345135, "1000–3000 Hz_10ms_envelope_correlation": 0.9839320532940375, "1000–3000 Hz_10ms_modulation_mae_db": 0.7612206989788702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4582060793929905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08658363311970937, "1000–3000 Hz_40ms_envelope_correlation": 0.995209592533505, "1000–3000 Hz_40ms_modulation_mae_db": 0.9398979485855561, "3000–6000 Hz_10ms_log_rms_mae_db": 2.672771545633703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2259641307723081, "3000–6000 Hz_10ms_envelope_correlation": 0.9678729621786206, "3000–6000 Hz_10ms_modulation_mae_db": 0.5861646613517838, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7627166596978354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13466620252410685, "3000–6000 Hz_40ms_envelope_correlation": 0.9879377735145276, "3000–6000 Hz_40ms_modulation_mae_db": 1.0920394514654606, "mr_stft_db": 6.232926173892619, "band_mae_db": 0.2888383655252463, "rms_gain_error_db": -0.18237323183645823, "output_peak": 0.5726137952131329}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84640, "output_samples": 84640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.639287779058652, "spectral_convergence": 0.4055699306380008, "spectral_convergence_status": "ok", "amplitude_floor": 1.2121984197282631e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.0992110663039085, "spectral_convergence": 0.36666717452446423, "spectral_convergence_status": "ok", "amplitude_floor": 7.081048851832571e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0397992405753795, "spectral_convergence": 0.394087928389566, "spectral_convergence_status": "ok", "amplitude_floor": 5.009180364479501e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.291800237989852e-05, "output_power": 8.260707838298179e-05, "delta_db": 0.5418250000377579, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025616304806087414, "output_power": 0.0002637287249879547, "delta_db": 0.12640952920300869, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038484147909669214, "output_power": 0.0003619027722115355, "delta_db": -0.2668996528301041, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002720810642199278, "output_power": 0.0002694353988327023, "delta_db": -0.042436642253008985, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.5951933766776914e-05, "output_power": 4.48928409561658e-05, "delta_db": -0.1012670249917087, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.6368160234296276e-05, "output_power": 2.6269111461536214e-05, "delta_db": -0.016344458911011887, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1825303847044334, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03910961498299732, "pitch": null, "output_rms_dbfs": -29.79292444737082, "output_peak": 0.5611377248890362, "reference_rms_dbfs": -29.753814832387818, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.259432695312647}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00018085971712132593, "log_rms_mae_db": 1.4151558780168831, "rms_envelope_nrmse": 0.09803711880646994, "envelope_correlation": 0.9914226939181526, "rms_gain_error_db": -0.0391124914076604, "reference_mean_power": 0.0010593154039372696, "output_mean_power": 0.001049818057063047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 5.491358535179395e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002450078451267542, "output_power": 0.00024562764999292455, "reference_active": true, "delta_db": 0.010972625466413534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.385267752666162e-05, "output_power": 8.189810635147565e-05, "reference_active": true, "delta_db": -0.10243074642702397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4530628910939782e-05, "output_power": 1.5408262250810063e-05, "reference_active": true, "delta_db": 0.2546924987029727}], "active_band_count": 3, "active_band_mae_db": 0.12269862353213674}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00014864066644098504, "log_rms_mae_db": 0.6558860025927993, "rms_envelope_nrmse": 0.04671737472364879, "envelope_correlation": 0.9978733639996272, "rms_gain_error_db": -0.0391478343660263, "reference_mean_power": 0.0010653029350414017, "output_mean_power": 0.0010557433149564303}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 4.840611714703517e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022860940218938906, "output_power": 0.0002278427115864851, "reference_active": true, "delta_db": -0.01458947395600465}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8392485325428115e-05, "output_power": 3.833186604335808e-05, "reference_active": true, "delta_db": -0.006862651741818436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1188041388515455e-06, "output_power": 1.150171450319092e-06, "reference_active": true, "delta_db": 0.12008518979510385}], "active_band_count": 3, "active_band_mae_db": 0.04717910516430898}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013930286763496605, "log_rms_mae_db": 2.12443090311934, "rms_envelope_nrmse": 0.1352015690138884, "envelope_correlation": 0.9865487090381698, "rms_gain_error_db": -0.10670636162889705, "reference_mean_power": 0.0005230599489297363, "output_mean_power": 0.0005103649381578548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 3.2510596296193775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001486768204336917, "output_power": 0.0001443465711807022, "reference_active": true, "delta_db": -0.12836792731365693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.417268122003869e-05, "output_power": 5.031145214632427e-05, "reference_active": true, "delta_db": -0.3211347846135172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.810050335491182e-06, "output_power": 1.0890415397669368e-05, "reference_active": true, "delta_db": 1.4439061264927302}], "active_band_count": 3, "active_band_mae_db": 0.6311362794733014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00011121771745577188, "log_rms_mae_db": 1.2817773488713824, "rms_envelope_nrmse": 0.059779514979258305, "envelope_correlation": 0.9972521751305929, "rms_gain_error_db": -0.10670831709269814, "reference_mean_power": 0.0005260459982805144, "output_mean_power": 0.0005132782830068679}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.9069199937213013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014215463322126283, "output_power": 0.00013965173168986917, "reference_active": true, "delta_db": -0.07714693793635251}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2639933776267858e-05, "output_power": 2.227222026506514e-05, "reference_active": true, "delta_db": -0.07111639185921338}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.874281484444505e-07, "output_power": 8.361851961561387e-07, "reference_active": true, "delta_db": 0.8507516309577969}], "active_band_count": 3, "active_band_mae_db": 0.3330049869177876}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010000130422823669, "log_rms_mae_db": 2.3811279174804607, "rms_envelope_nrmse": 0.16674651984979313, "envelope_correlation": 0.9806449072261316, "rms_gain_error_db": -0.06549059930945363, "reference_mean_power": 0.0002857956004291321, "output_mean_power": 0.0002815182012939395}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.149273337334192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04716085544297e-05, "output_power": 9.089033104749377e-05, "reference_active": true, "delta_db": 0.020053732519193262}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.048324869712114e-05, "output_power": 1.879627213506738e-05, "reference_active": true, "delta_db": -0.3732711384090839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.217073445558843e-06, "output_power": 6.6537290788424845e-06, "reference_active": true, "delta_db": -0.9165205514529342}], "active_band_count": 3, "active_band_mae_db": 0.43661514079373714}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 8.407177238322724e-05, "log_rms_mae_db": 1.4090244916017698, "rms_envelope_nrmse": 0.0828548670117455, "envelope_correlation": 0.994902914653883, "rms_gain_error_db": -0.06549196399640138, "reference_mean_power": 0.00028742710472676696, "output_mean_power": 0.000283125198491594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 1.945560884297879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.432816729085121e-05, "output_power": 8.433878060364047e-05, "reference_active": true, "delta_db": 0.0005465567786357894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0296492874768435e-05, "output_power": 8.982148691187565e-06, "reference_active": true, "delta_db": -0.5930908316615522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.97290149260778e-07, "output_power": 3.746383651999355e-07, "reference_active": true, "delta_db": -2.025731025960569}], "active_band_count": 3, "active_band_mae_db": 0.8731228048002523}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 5.528232384026309e-05, "log_rms_mae_db": 2.7831354635727905, "rms_envelope_nrmse": 0.20598921791922797, "envelope_correlation": 0.9733002197698529, "rms_gain_error_db": -0.06598329432488736, "reference_mean_power": 3.470766577035859e-05, "output_mean_power": 3.418433038295631e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.6534989625639428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5332454345272327e-05, "output_power": 1.4827457699096143e-05, "reference_active": true, "delta_db": -0.14544986464547527}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.828360450157823e-06, "output_power": 3.741464168798587e-06, "reference_active": true, "delta_db": -0.09971230558507593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1450631921352025e-06, "output_power": 7.551196015292668e-07, "reference_active": true, "delta_db": -1.8081371066169774}], "active_band_count": 3, "active_band_mae_db": 0.6844330922825095}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 4.431598350735027e-05, "log_rms_mae_db": 1.7874653575560049, "rms_envelope_nrmse": 0.12169779593777193, "envelope_correlation": 0.9901998372246528, "rms_gain_error_db": -0.06598585685745711, "reference_mean_power": 3.49057787168493e-05, "output_mean_power": 3.437943582165921e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.464962908732827e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.55740476492126e-05, "output_power": 1.4914821742252899e-05, "reference_active": true, "delta_db": -0.18783431933851613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.044495719000328e-06, "output_power": 1.8737298527544938e-06, "reference_active": true, "delta_db": -0.37879229377624607}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.011419545771505e-07, "output_power": 5.653341826149134e-08, "reference_active": true, "delta_db": -2.526260965298203}], "active_band_count": 3, "active_band_mae_db": 1.030962526137655}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8081606073922151, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ff05f2e1c2850659d509b3250d37a231c8d632266ac2a5974f2253513ffe053a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9835270241982009, "standard_deviation": 0.0022361752047587854, "minimum": 0.9806449072261316, "maximum": 0.986673856503978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.4055831696582506, "standard_deviation": 0.02644818304846508, "minimum": 2.365410193663836, "maximum": 2.4392955523535464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7029254596878626, "standard_deviation": 0.24587202751002132, "minimum": 0.43661514079373714, "maximum": 1.1131373682012917, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15443322444966404, "standard_deviation": 0.01030806192510753, "minimum": 0.13851442388054555, "maximum": 0.16724799360201037, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950753937452614, "standard_deviation": 0.0006573859975767102, "minimum": 0.9943608682634095, "maximum": 0.9959832764953348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.426254628401357, "standard_deviation": 0.03813263918674117, "minimum": 1.37594245569468, "maximum": 1.4658306455600685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9624236336017999, "standard_deviation": 0.31725064921766144, "minimum": 0.6445330179865937, "maximum": 1.4622213456304187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08422877622063049, "standard_deviation": 0.005677625590282416, "minimum": 0.07537215987101396, "maximum": 0.09135586067638646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9698096114655718, "standard_deviation": 0.002823081679356262, "minimum": 0.9668297630818068, "maximum": 0.9749206929115363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.76062321702676, "standard_deviation": 0.03975469494504962, "minimum": 2.672771545633703, "maximum": 2.8055190205712908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7191275949802608, "standard_deviation": 0.3177601996520048, "minimum": 0.3736852140703533, "maximum": 1.3623151710076176, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2216463362879857, "standard_deviation": 0.010876097223517673, "minimum": 0.20282700599724113, "maximum": 0.22958220285966952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9885293533033455, "standard_deviation": 0.0014178223557363952, "minimum": 0.986626684349118, "maximum": 0.9901998372246528, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8062631611386766, "standard_deviation": 0.027083190400444313, "minimum": 1.7627166596978354, "maximum": 1.8565879795672393, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0772660378951482, "standard_deviation": 0.31840194048583004, "minimum": 0.7648376816500289, "maximum": 1.7482156392063384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1340367135521795, "standard_deviation": 0.008037193661910542, "minimum": 0.12169779593777193, "maximum": 0.14450643976398517, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9861184074961672, "standard_deviation": 0.001828338203755418, "minimum": 0.9839200941787953, "maximum": 0.988333470504127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1236551438397027, "standard_deviation": 0.036561012664040805, "minimum": 2.083235269763939, "maximum": 2.1919294845880803, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4409323814847538, "standard_deviation": 0.2665509198022134, "minimum": 0.09794140789877413, "maximum": 0.8819939366670155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13781503785490384, "standard_deviation": 0.008447417521937249, "minimum": 0.1268095961093429, "maximum": 0.14774620359447477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954045807023312, "standard_deviation": 0.000934440202993167, "minimum": 0.9943162840810463, "maximum": 0.9972521751305929, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2490907334913586, "standard_deviation": 0.0544814616787582, "minimum": 1.1648052520347658, "maximum": 1.336212133146321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3986549768385696, "standard_deviation": 0.18577913108170355, "minimum": 0.09346803378868751, "maximum": 0.6373614278187906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07758323438348899, "standard_deviation": 0.008046197187270326, "minimum": 0.059779514979258305, "maximum": 0.08527469251347136, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22833822827152786, "standard_deviation": 0.05211837392115014, "minimum": 0.1573353003517415, "maximum": 0.31444122517575335, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.25369356805046, "standard_deviation": 0.00961264803893993, "minimum": 6.232926173892619, "maximum": 6.261704436461464, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5777816739920953, "standard_deviation": 0.015697451226908122, "minimum": 0.5611377248890362, "maximum": 0.6067891023251423, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07948296534941983, "standard_deviation": 0.06071112344475506, "minimum": -0.18237323183645823, "maximum": -0.020838974244791874, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9914561918625606, "standard_deviation": 0.0009265274449403012, "minimum": 0.990413107290628, "maximum": 0.9928052119408051, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3911420203361384, "standard_deviation": 0.03485316129956006, "minimum": 1.3226535287142673, "maximum": 1.4346126631186455, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2832532342886786, "standard_deviation": 0.15579361772483558, "minimum": 0.0782297327271438, "maximum": 0.5876118144038013, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09810142795568454, "standard_deviation": 0.004929035120644713, "minimum": 0.09133803209969395, "maximum": 0.10370982758530102, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977202048821256, "standard_deviation": 0.00029332106739980365, "minimum": 0.9972739689553001, "maximum": 0.9980323736402507, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6857251763948033, "standard_deviation": 0.03528333408338085, "minimum": 0.6479431023824621, "maximum": 0.754574784564877, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28964753648824304, "standard_deviation": 0.17784865737401362, "minimum": 0.04717910516430898, "maximum": 0.6101631021182179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04937767381534121, "standard_deviation": 0.002841038922372434, "minimum": 0.046477238171276546, "maximum": 0.05334560206236303, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0019", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.086650848388672, "predicted_mos": 2.086650848388672, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "ab487c2a7be58eaa8a9a0366a4b1bc7903dd40c47906b3f1c2254be57c281a3f", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.904186500236392, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3702693295839947, "全帯域_10ms_rms_envelope_nrmse": 0.09960466597509848, "全帯域_10ms_envelope_correlation": 0.9911503160810883, "全帯域_10ms_modulation_mae_db": 0.3655929291442752, "全帯域_40ms_log_rms_mae_db": 0.7102736481150312, "全帯域_40ms_rms_envelope_nrmse": 0.04729672750810993, "全帯域_40ms_envelope_correlation": 0.9978560381326222, "全帯域_40ms_modulation_mae_db": 0.5955697665062213, "300–1000 Hz_10ms_log_rms_mae_db": 1.6394417900214644, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13290823959068276, "300–1000 Hz_10ms_envelope_correlation": 0.987098005563266, "300–1000 Hz_10ms_modulation_mae_db": 0.909605546917103, "300–1000 Hz_40ms_log_rms_mae_db": 1.0084966554214718, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0686774517843532, "300–1000 Hz_40ms_envelope_correlation": 0.9965091068630287, "300–1000 Hz_40ms_modulation_mae_db": 0.6672986919482579, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7265520899291296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13474409438693344, "1000–3000 Hz_10ms_envelope_correlation": 0.9874655393632075, "1000–3000 Hz_10ms_modulation_mae_db": 0.778232355183138, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1515265336892342, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07686584259755215, "1000–3000 Hz_40ms_envelope_correlation": 0.9958421573243901, "1000–3000 Hz_40ms_modulation_mae_db": 0.9543402167476024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.899774593761012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19361660516768409, "3000–6000 Hz_10ms_envelope_correlation": 0.9762266034556812, "3000–6000 Hz_10ms_modulation_mae_db": 0.1580987114851424, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2920218116756463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10178209356740744, "3000–6000 Hz_40ms_envelope_correlation": 0.993058000328378, "3000–6000 Hz_40ms_modulation_mae_db": 0.45681207425881426, "mr_stft_db": 5.824039072286159, "band_mae_db": 0.18733204798306924, "rms_gain_error_db": -0.06237380795664413, "output_peak": 0.5840386738600585}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5478857690307002, "全帯域_10ms_rms_envelope_nrmse": 0.09587678147451847, "全帯域_10ms_envelope_correlation": 0.9918416333367477, "全帯域_10ms_modulation_mae_db": 0.08643024233068325, "全帯域_40ms_log_rms_mae_db": 0.7396802303290231, "全帯域_40ms_rms_envelope_nrmse": 0.05192942001808591, "全帯域_40ms_envelope_correlation": 0.9973784943182298, "全帯域_40ms_modulation_mae_db": 0.15556141488506411, "300–1000 Hz_10ms_log_rms_mae_db": 1.5255331840110642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13461038185584515, "300–1000 Hz_10ms_envelope_correlation": 0.9867698279738351, "300–1000 Hz_10ms_modulation_mae_db": 0.07918684995825478, "300–1000 Hz_40ms_log_rms_mae_db": 0.9470866875571665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07963602253116485, "300–1000 Hz_40ms_envelope_correlation": 0.995023132999152, "300–1000 Hz_40ms_modulation_mae_db": 0.1687080827881348, "1000–3000 Hz_10ms_log_rms_mae_db": 1.728673872033933, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1267965411625323, "1000–3000 Hz_10ms_envelope_correlation": 0.988960458539174, "1000–3000 Hz_10ms_modulation_mae_db": 0.23213855615977366, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1140150368723958, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07414969379674734, "1000–3000 Hz_40ms_envelope_correlation": 0.9961648315635604, "1000–3000 Hz_40ms_modulation_mae_db": 0.3340513522529654, "3000–6000 Hz_10ms_log_rms_mae_db": 1.883541967982904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1913404589166605, "3000–6000 Hz_10ms_envelope_correlation": 0.9768608944341456, "3000–6000 Hz_10ms_modulation_mae_db": 0.5116147892755148, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2704266696827131, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11208167818743699, "3000–6000 Hz_40ms_envelope_correlation": 0.991939200353466, "3000–6000 Hz_40ms_modulation_mae_db": 0.7300123845308683, "mr_stft_db": 5.8442224051081295, "band_mae_db": 0.18813128516797725, "rms_gain_error_db": 0.005522338831459301, "output_peak": 0.6000238277545262}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.490644081265574, "全帯域_10ms_rms_envelope_nrmse": 0.101314687107881, "全帯域_10ms_envelope_correlation": 0.9909840909049092, "全帯域_10ms_modulation_mae_db": 0.6105524944449521, "全帯域_40ms_log_rms_mae_db": 0.6901857724108209, "全帯域_40ms_rms_envelope_nrmse": 0.05491363633552909, "全帯域_40ms_envelope_correlation": 0.9971187056716466, "全帯域_40ms_modulation_mae_db": 0.5873871864590438, "300–1000 Hz_10ms_log_rms_mae_db": 1.5592822690470782, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1526892216077111, "300–1000 Hz_10ms_envelope_correlation": 0.9829945345226472, "300–1000 Hz_10ms_modulation_mae_db": 1.01276298878222, "300–1000 Hz_40ms_log_rms_mae_db": 1.0145694444301951, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08492520161003356, "300–1000 Hz_40ms_envelope_correlation": 0.9943301098598429, "300–1000 Hz_40ms_modulation_mae_db": 0.9634868334973411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6995273594207685, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12809295799967407, "1000–3000 Hz_10ms_envelope_correlation": 0.9886184037993013, "1000–3000 Hz_10ms_modulation_mae_db": 0.4296122304062435, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0547041113221811, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06576027670796118, "1000–3000 Hz_40ms_envelope_correlation": 0.9968417907146784, "1000–3000 Hz_40ms_modulation_mae_db": 0.5522876302128312, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8279271424835908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19899697783284878, "3000–6000 Hz_10ms_envelope_correlation": 0.9789968315695229, "3000–6000 Hz_10ms_modulation_mae_db": 0.506039000892713, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2717920245845296, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12234296142052159, "3000–6000 Hz_40ms_envelope_correlation": 0.993186511065494, "3000–6000 Hz_40ms_modulation_mae_db": 0.3605701828800969, "mr_stft_db": 5.831636036680558, "band_mae_db": 0.2680600668664166, "rms_gain_error_db": 0.04309839373105753, "output_peak": 0.596131243142876}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4930664736220747, "全帯域_10ms_rms_envelope_nrmse": 0.1025815557420947, "全帯域_10ms_envelope_correlation": 0.9906832777824858, "全帯域_10ms_modulation_mae_db": 0.23873854670414651, "全帯域_40ms_log_rms_mae_db": 0.670977111298917, "全帯域_40ms_rms_envelope_nrmse": 0.055675631844000685, "全帯域_40ms_envelope_correlation": 0.9971550670598718, "全帯域_40ms_modulation_mae_db": 0.46399499334554745, "300–1000 Hz_10ms_log_rms_mae_db": 1.6874671561118242, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14349986045096738, "300–1000 Hz_10ms_envelope_correlation": 0.9849088083284304, "300–1000 Hz_10ms_modulation_mae_db": 0.7282083745354891, "300–1000 Hz_40ms_log_rms_mae_db": 1.0438477234491657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08641075875747695, "300–1000 Hz_40ms_envelope_correlation": 0.9942901230177438, "300–1000 Hz_40ms_modulation_mae_db": 0.32869338482431026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.716703644668475, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12183612670640605, "1000–3000 Hz_10ms_envelope_correlation": 0.9902972541906218, "1000–3000 Hz_10ms_modulation_mae_db": 0.812120869099855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0677501373782483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06959173794005755, "1000–3000 Hz_40ms_envelope_correlation": 0.9972517100709452, "1000–3000 Hz_40ms_modulation_mae_db": 1.0073294917419926, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9049671723672938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20451517698325922, "3000–6000 Hz_10ms_envelope_correlation": 0.9746041417026824, "3000–6000 Hz_10ms_modulation_mae_db": 1.404018323983329, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3238962848879088, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13713249323995763, "3000–6000 Hz_40ms_envelope_correlation": 0.9894040171933534, "3000–6000 Hz_40ms_modulation_mae_db": 1.6493483193452037, "mr_stft_db": 5.829490090512799, "band_mae_db": 0.32386643255357583, "rms_gain_error_db": -0.1317708320839846, "output_peak": 0.6034745404248347}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5107681676905091, "全帯域_10ms_rms_envelope_nrmse": 0.09808826712385366, "全帯域_10ms_envelope_correlation": 0.9914303345396837, "全帯域_10ms_modulation_mae_db": 0.619560877156777, "全帯域_40ms_log_rms_mae_db": 0.7445522694043617, "全帯域_40ms_rms_envelope_nrmse": 0.049919235226985215, "全帯域_40ms_envelope_correlation": 0.9976410381728926, "全帯域_40ms_modulation_mae_db": 0.6271082385688381, "300–1000 Hz_10ms_log_rms_mae_db": 1.5926451979053582, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11817574761471693, "300–1000 Hz_10ms_envelope_correlation": 0.9897333408861696, "300–1000 Hz_10ms_modulation_mae_db": 0.5171387341955157, "300–1000 Hz_40ms_log_rms_mae_db": 0.9674690004487638, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06363107134123432, "300–1000 Hz_40ms_envelope_correlation": 0.9967919011202125, "300–1000 Hz_40ms_modulation_mae_db": 0.38731418768339404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.70857481209413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1370817195375467, "1000–3000 Hz_10ms_envelope_correlation": 0.9874653206911604, "1000–3000 Hz_10ms_modulation_mae_db": 0.9692013210828828, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0823591361521596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07974754924445533, "1000–3000 Hz_40ms_envelope_correlation": 0.9961221257555265, "1000–3000 Hz_40ms_modulation_mae_db": 1.3187888272904604, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8166336398416079, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20462940731012377, "3000–6000 Hz_10ms_envelope_correlation": 0.973464384980452, "3000–6000 Hz_10ms_modulation_mae_db": 0.10270185139088357, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2444054118419987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1158796381238786, "3000–6000 Hz_40ms_envelope_correlation": 0.9910818861563098, "3000–6000 Hz_40ms_modulation_mae_db": 0.5145675498694022, "mr_stft_db": 5.83535156978857, "band_mae_db": 0.1488482460545863, "rms_gain_error_db": -0.08577804259449914, "output_peak": 0.5934443573838522}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3639073223511067, "全帯域_10ms_rms_envelope_nrmse": 0.11826251982100718, "全帯域_10ms_envelope_correlation": 0.987846206220775, "全帯域_10ms_modulation_mae_db": 0.2566179934748382, "全帯域_40ms_log_rms_mae_db": 0.7169027247556956, "全帯域_40ms_rms_envelope_nrmse": 0.07518594749592326, "全帯域_40ms_envelope_correlation": 0.9946956352040333, "全帯域_40ms_modulation_mae_db": 0.24694109578288756, "300–1000 Hz_10ms_log_rms_mae_db": 1.5962935859908356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1594439928236902, "300–1000 Hz_10ms_envelope_correlation": 0.9829139631110602, "300–1000 Hz_10ms_modulation_mae_db": 0.3954139694044003, "300–1000 Hz_40ms_log_rms_mae_db": 0.9427500522542218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10769863152273748, "300–1000 Hz_40ms_envelope_correlation": 0.9921244683663022, "300–1000 Hz_40ms_modulation_mae_db": 0.5427787434943975, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8055927938638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14210599842707003, "1000–3000 Hz_10ms_envelope_correlation": 0.9865639147974653, "1000–3000 Hz_10ms_modulation_mae_db": 0.49921328992911973, "1000–3000 Hz_40ms_log_rms_mae_db": 1.175468997958611, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08588644347396626, "1000–3000 Hz_40ms_envelope_correlation": 0.9954850751677546, "1000–3000 Hz_40ms_modulation_mae_db": 0.709728681626587, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8825321685913357, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19425162694400613, "3000–6000 Hz_10ms_envelope_correlation": 0.9765895909907916, "3000–6000 Hz_10ms_modulation_mae_db": 0.566404015225798, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3220026443158204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12607503454542565, "3000–6000 Hz_40ms_envelope_correlation": 0.9895509820323792, "3000–6000 Hz_40ms_modulation_mae_db": 0.7760024336952598, "mr_stft_db": 5.819253412713294, "band_mae_db": 0.25872628718910406, "rms_gain_error_db": 0.09628319478119, "output_peak": 0.5664355795788806}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5139836769155728, "全帯域_10ms_rms_envelope_nrmse": 0.10364892892053452, "全帯域_10ms_envelope_correlation": 0.9905170411189083, "全帯域_10ms_modulation_mae_db": 0.2486299924784361, "全帯域_40ms_log_rms_mae_db": 0.7040808218379826, "全帯域_40ms_rms_envelope_nrmse": 0.053931523607204976, "全帯域_40ms_envelope_correlation": 0.9973998680330151, "全帯域_40ms_modulation_mae_db": 0.32967902043174363, "300–1000 Hz_10ms_log_rms_mae_db": 1.6001851530461733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14084945708531293, "300–1000 Hz_10ms_envelope_correlation": 0.9858673569985075, "300–1000 Hz_10ms_modulation_mae_db": 0.626082080517502, "300–1000 Hz_40ms_log_rms_mae_db": 1.0368549342188158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0805625983586425, "300–1000 Hz_40ms_envelope_correlation": 0.9956695565423104, "300–1000 Hz_40ms_modulation_mae_db": 0.6758091750605743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7473608883927232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13939478062552102, "1000–3000 Hz_10ms_envelope_correlation": 0.9866553879136652, "1000–3000 Hz_10ms_modulation_mae_db": 0.42666121160356507, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1195328627573633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07697066781842855, "1000–3000 Hz_40ms_envelope_correlation": 0.995957132833772, "1000–3000 Hz_40ms_modulation_mae_db": 0.6047979842133578, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8349262076219757, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18642585006048049, "3000–6000 Hz_10ms_envelope_correlation": 0.9803528062180616, "3000–6000 Hz_10ms_modulation_mae_db": 0.5908495559961402, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2350216273112657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10192667673913111, "3000–6000 Hz_40ms_envelope_correlation": 0.9946722143394706, "3000–6000 Hz_40ms_modulation_mae_db": 0.5572461701036998, "mr_stft_db": 5.827068648617587, "band_mae_db": 0.3008997419494785, "rms_gain_error_db": -0.14607335972500585, "output_peak": 0.6022237039857379}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4897390735070835, "全帯域_10ms_rms_envelope_nrmse": 0.1164772373490483, "全帯域_10ms_envelope_correlation": 0.987936516213779, "全帯域_10ms_modulation_mae_db": 0.48461109246724093, "全帯域_40ms_log_rms_mae_db": 0.6680751678972493, "全帯域_40ms_rms_envelope_nrmse": 0.05417848966811275, "全帯域_40ms_envelope_correlation": 0.9973337276373935, "全帯域_40ms_modulation_mae_db": 0.3152508270049231, "300–1000 Hz_10ms_log_rms_mae_db": 1.70334178254592, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15556765153140906, "300–1000 Hz_10ms_envelope_correlation": 0.9821787764239868, "300–1000 Hz_10ms_modulation_mae_db": 0.78887004700218, "300–1000 Hz_40ms_log_rms_mae_db": 1.0371460048340388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07507185103973547, "300–1000 Hz_40ms_envelope_correlation": 0.9956837310154156, "300–1000 Hz_40ms_modulation_mae_db": 0.4233962369185546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7286080576608758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14109184974592803, "1000–3000 Hz_10ms_envelope_correlation": 0.9865805270501369, "1000–3000 Hz_10ms_modulation_mae_db": 0.35311475805798404, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0898085267084174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08202117536229055, "1000–3000 Hz_40ms_envelope_correlation": 0.9957237185917113, "1000–3000 Hz_40ms_modulation_mae_db": 0.4216813496443839, "3000–6000 Hz_10ms_log_rms_mae_db": 1.897293655997232, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19388635100656992, "3000–6000 Hz_10ms_envelope_correlation": 0.9762366682177658, "3000–6000 Hz_10ms_modulation_mae_db": 0.6104323820430407, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3075805970022354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10924541073760367, "3000–6000 Hz_40ms_envelope_correlation": 0.9919742555439098, "3000–6000 Hz_40ms_modulation_mae_db": 1.1654291126691836, "mr_stft_db": 5.842053496967668, "band_mae_db": 0.18512164798759453, "rms_gain_error_db": -0.13927499944136526, "output_peak": 0.5840598343529194}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84640, "output_samples": 84640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.011069768818513, "spectral_convergence": 0.39717191002032404, "spectral_convergence_status": "ok", "amplitude_floor": 1.2121984197282631e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.543925408949141, "spectral_convergence": 0.35431130624515317, "spectral_convergence_status": "ok", "amplitude_floor": 7.081048851832571e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.917122039090826, "spectral_convergence": 0.3963377545105649, "spectral_convergence_status": "ok", "amplitude_floor": 5.009180364479501e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.291800237989852e-05, "output_power": 8.118397836713199e-05, "delta_db": 0.46635567355011903, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025616304806087414, "output_power": 0.00026076696949187345, "delta_db": 0.0773609763043219, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038484147909669214, "output_power": 0.00036699651841994615, "delta_db": -0.20619930886189683, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002720810642199278, "output_power": 0.0002641922358779447, "delta_db": -0.12778267277674554, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.5951933766776914e-05, "output_power": 4.417810626552414e-05, "delta_db": -0.17096696743769835, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.6368160234296276e-05, "output_power": 2.591475815720627e-05, "delta_db": -0.07532668896763392, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18733204798306924, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06237380795664413, "pitch": null, "output_rms_dbfs": -29.816188640344464, "output_peak": 0.5840386738600585, "reference_rms_dbfs": -29.753814832387818, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.824039072286159}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00018085971712132593, "log_rms_mae_db": 1.3702693295839947, "rms_envelope_nrmse": 0.09960466597509848, "envelope_correlation": 0.9911503160810883, "rms_gain_error_db": -0.062397235895720105, "reference_mean_power": 0.0010593154039372696, "output_mean_power": 0.0010442045084662047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 5.491358535179395e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002450078451267542, "output_power": 0.00023444714553216077, "reference_active": true, "delta_db": -0.1913504121819631}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.385267752666162e-05, "output_power": 6.938867832916567e-05, "reference_active": true, "delta_db": -0.8222831935189272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4530628910939782e-05, "output_power": 1.4255087701246068e-05, "reference_active": true, "delta_db": -0.08314518173193537}], "active_band_count": 3, "active_band_mae_db": 0.3655929291442752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00014864066644098504, "log_rms_mae_db": 0.7102736481150312, "rms_envelope_nrmse": 0.04729672750810993, "envelope_correlation": 0.9978560381326222, "rms_gain_error_db": -0.062365294520396075, "reference_mean_power": 0.0010653029350414017, "output_mean_power": 0.0010501143521134781}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 4.840611714703517e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022860940218938906, "output_power": 0.00021755297155769246, "reference_active": true, "delta_db": -0.2152906827876656}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8392485325428115e-05, "output_power": 3.2873398305543626e-05, "reference_active": true, "delta_db": -0.674016251116075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1188041388515455e-06, "output_power": 9.099422443409034e-07, "reference_active": true, "delta_db": -0.8974023656149233}], "active_band_count": 3, "active_band_mae_db": 0.5955697665062213}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013930286763496605, "log_rms_mae_db": 1.6394417900214644, "rms_envelope_nrmse": 0.13290823959068276, "envelope_correlation": 0.987098005563266, "rms_gain_error_db": -0.182373702865352, "reference_mean_power": 0.0005230599489297363, "output_mean_power": 0.0005015498406645835}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 3.2510596296193775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001486768204336917, "output_power": 0.00013039886836576042, "reference_active": true, "delta_db": -0.5696944231157338}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.417268122003869e-05, "output_power": 4.295844889397175e-05, "reference_active": true, "delta_db": -1.0073173921902159}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.810050335491182e-06, "output_power": 1.018202866712396e-05, "reference_active": true, "delta_db": 1.1518048254453594}], "active_band_count": 3, "active_band_mae_db": 0.909605546917103}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00011121771745577188, "log_rms_mae_db": 1.0084966554214718, "rms_envelope_nrmse": 0.0686774517843532, "envelope_correlation": 0.9965091068630287, "rms_gain_error_db": -0.182375221045462, "reference_mean_power": 0.0005260459982805144, "output_mean_power": 0.0005044129165858117}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.9069199937213013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014215463322126283, "output_power": 0.00012541672416857747, "reference_active": true, "delta_db": -0.5440556614117047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2639933776267858e-05, "output_power": 1.8159825846118653e-05, "reference_active": true, "delta_db": -0.9576347287637152}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.874281484444505e-07, "output_power": 6.126419662535983e-07, "reference_active": true, "delta_db": -0.5002056856693534}], "active_band_count": 3, "active_band_mae_db": 0.6672986919482579}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010000130422823669, "log_rms_mae_db": 1.7265520899291296, "rms_envelope_nrmse": 0.13474409438693344, "envelope_correlation": 0.9874655393632075, "rms_gain_error_db": -0.17413976947578874, "reference_mean_power": 0.0002857956004291321, "output_mean_power": 0.00027456271581504277}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.149273337334192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04716085544297e-05, "output_power": 9.219863258698707e-05, "reference_active": true, "delta_db": 0.08212168015315649}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.048324869712114e-05, "output_power": 1.7924102392703728e-05, "reference_active": true, "delta_db": -0.579614218277716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.217073445558843e-06, "output_power": 5.590118987382128e-06, "reference_active": true, "delta_db": -1.6729611671185416}], "active_band_count": 3, "active_band_mae_db": 0.778232355183138}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 8.407177238322724e-05, "log_rms_mae_db": 1.1515265336892342, "rms_envelope_nrmse": 0.07686584259755215, "envelope_correlation": 0.9958421573243901, "rms_gain_error_db": -0.1741402166183033, "reference_mean_power": 0.00028742710472676696, "output_mean_power": 0.00027613006718437387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 1.945560884297879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.432816729085121e-05, "output_power": 8.542160901142967e-05, "reference_active": true, "delta_db": 0.05595085708082429}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0296492874768435e-05, "output_power": 8.96215298339875e-06, "reference_active": true, "delta_db": -0.6027697027664475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.97290149260778e-07, "output_power": 3.595467299354962e-07, "reference_active": true, "delta_db": -2.2043000903955354}], "active_band_count": 3, "active_band_mae_db": 0.9543402167476024}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 5.528232384026309e-05, "log_rms_mae_db": 1.899774593761012, "rms_envelope_nrmse": 0.19361660516768409, "envelope_correlation": 0.9762266034556812, "rms_gain_error_db": -0.0838807277163721, "reference_mean_power": 3.470766577035859e-05, "output_mean_power": 3.4043745392317876e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.6534989625639428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5332454345272327e-05, "output_power": 1.5074405891497303e-05, "reference_active": true, "delta_db": -0.07371475333136394}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.828360450157823e-06, "output_power": 3.932859633258932e-06, "reference_active": true, "delta_db": 0.1169562539494061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1450631921352025e-06, "output_power": 1.072671993795571e-06, "reference_active": true, "delta_db": -0.2836251271746571}], "active_band_count": 3, "active_band_mae_db": 0.1580987114851424}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 4.431598350735027e-05, "log_rms_mae_db": 1.2920218116756463, "rms_envelope_nrmse": 0.10178209356740744, "envelope_correlation": 0.993058000328378, "rms_gain_error_db": -0.08388158658692885, "reference_mean_power": 3.49057787168493e-05, "output_mean_power": 3.4238061879873546e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.464962908732827e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.55740476492126e-05, "output_power": 1.5517381319343215e-05, "reference_active": true, "delta_db": -0.015830666159684436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.044495719000328e-06, "output_power": 2.004875749720154e-06, "reference_active": true, "delta_db": -0.08498742715596516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.011419545771505e-07, "output_power": 7.550392510559431e-08, "reference_active": true, "delta_db": -1.2696181294607933}], "active_band_count": 3, "active_band_mae_db": 0.45681207425881426}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8343525031385889, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c0199f7289cc2c77d43b990423c7634603543d17080df18c4aa779bc47608870", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9878258507930916, "standard_deviation": 0.0013520573251418779, "minimum": 0.9865639147974653, "maximum": 0.9902972541906218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7326991897579793, "standard_deviation": 0.032814160043694925, "minimum": 1.6995273594207685, "maximum": 1.8055927938638, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5625368239403202, "standard_deviation": 0.25838907197621247, "minimum": 0.23213855615977366, "maximum": 0.9692013210828828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13389300857395145, "standard_deviation": 0.00746235383851189, "minimum": 0.12183612670640605, "maximum": 0.14210599842707003, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961735677527923, "standard_deviation": 0.0005909675648183152, "minimum": 0.9954850751677546, "maximum": 0.9972517100709452, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1068956678548263, "standard_deviation": 0.0414986106600671, "minimum": 1.0547041113221811, "maximum": 1.175468997958611, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7378756917162727, "standard_deviation": 0.33229077845273036, "minimum": 0.3340513522529654, "maximum": 1.3187888272904604, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07637417336768237, "standard_deviation": 0.006524171158786867, "minimum": 0.06576027670796118, "maximum": 0.08588644347396626, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9766664901961378, "standard_deviation": 0.0022022108898572946, "minimum": 0.973464384980452, "maximum": 0.9803528062180616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.868449568580869, "standard_deviation": 0.035903933983396884, "minimum": 1.8166336398416079, "maximum": 1.9049671723672938, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5562698287865702, "standard_deviation": 0.39491245140967995, "minimum": 0.10270185139088357, "maximum": 1.404018323983329, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19595780677770414, "standard_deviation": 0.006352986981115534, "minimum": 0.18642585006048049, "maximum": 0.20462940731012377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991858383376595, "standard_deviation": 0.0018162586614644063, "minimum": 0.9894040171933534, "maximum": 0.9946722143394706, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2833933839127647, "standard_deviation": 0.0337072417782041, "minimum": 1.2350216273112657, "maximum": 1.3238962848879088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.776248528419066, "standard_deviation": 0.43213253858306055, "minimum": 0.3605701828800969, "maximum": 1.6493483193452037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11580824832017034, "standard_deviation": 0.012242940247922713, "minimum": 0.10178209356740744, "maximum": 0.13713249323995763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9853080767259879, "standard_deviation": 0.0025710618269204908, "minimum": 0.9821787764239868, "maximum": 0.9897333408861696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6130237648349648, "standard_deviation": 0.06070989541857213, "minimum": 1.5255331840110642, "maximum": 1.70334178254592, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6321585739140831, "standard_deviation": 0.3001696701159241, "minimum": 0.07918684995825478, "maximum": 1.01276298878222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14221806907004192, "standard_deviation": 0.013683608416539934, "minimum": 0.11817574761471693, "maximum": 0.1594439928236902, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9950527662230011, "standard_deviation": 0.0014930184412312605, "minimum": 0.9921244683663022, "maximum": 0.9967919011202125, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9997775628267299, "standard_deviation": 0.04154247709691735, "minimum": 0.9427500522542218, "maximum": 1.0438477234491657, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5196856670268706, "standard_deviation": 0.2478807604536744, "minimum": 0.1687080827881348, "maximum": 0.9634868334973411, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08082669836817229, "standard_deviation": 0.013352259079378138, "minimum": 0.06363107134123432, "maximum": 0.10769863152273748, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2326232194689753, "standard_deviation": 0.06350434816394862, "minimum": 0.1488482460545863, "maximum": 0.32386643255357583, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.831639341584346, "standard_deviation": 0.008596133773163095, "minimum": 5.819253412713294, "maximum": 5.8442224051081295, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5912289700604607, "standard_deviation": 0.012506780958900012, "minimum": 0.5664355795788806, "maximum": 0.6034745404248347, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05254588930722402, "standard_deviation": 0.09135232117634912, "minimum": -0.14607335972500585, "maximum": 0.09628319478119, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9902986770247971, "standard_deviation": 0.00154209790049328, "minimum": 0.987846206220775, "maximum": 0.9918416333367477, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.472532986745827, "standard_deviation": 0.06778243173823104, "minimum": 1.3639073223511067, "maximum": 1.5478857690307002, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3638417710251687, "standard_deviation": 0.1920659868838199, "minimum": 0.08643024233068325, "maximum": 0.619560877156777, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10448183043925455, "standard_deviation": 0.008339612917343565, "minimum": 0.09587678147451847, "maximum": 0.11826251982100718, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970723217787132, "standard_deviation": 0.0009902848203095101, "minimum": 0.9946956352040333, "maximum": 0.9978560381326222, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7055909682561352, "standard_deviation": 0.028473293434029368, "minimum": 0.6680751678972493, "maximum": 0.7445522694043617, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.41518656787303365, "standard_deviation": 0.17828334599463405, "minimum": 0.15556141488506411, "maximum": 0.6271082385688381, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05537882646299398, "standard_deviation": 0.008478442391336905, "minimum": 0.04729672750810993, "maximum": 0.07518594749592326, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0019", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.9651846885681152, "predicted_mos": 1.9651846885681152, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "a4181bb8fd8dee9a554c9bd9c9730f5626f5a4f2ca3269e9b6f6a8d2393417d9", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.288721899967641, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2841603305593785, "全帯域_10ms_rms_envelope_nrmse": 0.09248533605115368, "全帯域_10ms_envelope_correlation": 0.9925543335158971, "全帯域_10ms_modulation_mae_db": 0.3968058495302999, "全帯域_40ms_log_rms_mae_db": 0.6433859419793871, "全帯域_40ms_rms_envelope_nrmse": 0.05454392917226289, "全帯域_40ms_envelope_correlation": 0.9973901382424853, "全帯域_40ms_modulation_mae_db": 0.36448195760356134, "300–1000 Hz_10ms_log_rms_mae_db": 1.7617659116019724, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12244180905406767, "300–1000 Hz_10ms_envelope_correlation": 0.9889761947425362, "300–1000 Hz_10ms_modulation_mae_db": 0.4757950957893988, "300–1000 Hz_40ms_log_rms_mae_db": 1.0663710840193041, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08185805622496677, "300–1000 Hz_40ms_envelope_correlation": 0.9946798762347286, "300–1000 Hz_40ms_modulation_mae_db": 0.4925024535769115, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9979265347204502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15340584420229753, "1000–3000 Hz_10ms_envelope_correlation": 0.9852107096377635, "1000–3000 Hz_10ms_modulation_mae_db": 0.3103448957740867, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2613452049080682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09803382268510147, "1000–3000 Hz_40ms_envelope_correlation": 0.9950818760195023, "1000–3000 Hz_40ms_modulation_mae_db": 0.5478981517189345, "3000–6000 Hz_10ms_log_rms_mae_db": 2.208410330442552, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18358959952312898, "3000–6000 Hz_10ms_envelope_correlation": 0.9787521077755452, "3000–6000 Hz_10ms_modulation_mae_db": 1.32170393335613, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4829754540804743, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1055501233144711, "3000–6000 Hz_40ms_envelope_correlation": 0.9926363877360761, "3000–6000 Hz_40ms_modulation_mae_db": 1.6882500831008482, "mr_stft_db": 5.999170021201951, "band_mae_db": 0.19295590686178532, "rms_gain_error_db": -0.16391658277535862, "output_peak": 0.5884663073557219}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.359729069055651, "全帯域_10ms_rms_envelope_nrmse": 0.09966556026156895, "全帯域_10ms_envelope_correlation": 0.9911428508339332, "全帯域_10ms_modulation_mae_db": 0.11404542042831384, "全帯域_40ms_log_rms_mae_db": 0.6192641015087134, "全帯域_40ms_rms_envelope_nrmse": 0.05358194105193418, "全帯域_40ms_envelope_correlation": 0.9972350071708094, "全帯域_40ms_modulation_mae_db": 0.08523625685093317, "300–1000 Hz_10ms_log_rms_mae_db": 1.8019104721293322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13362659798134208, "300–1000 Hz_10ms_envelope_correlation": 0.9868671578660717, "300–1000 Hz_10ms_modulation_mae_db": 0.7506506866433232, "300–1000 Hz_40ms_log_rms_mae_db": 1.084193373158611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07412024529110808, "300–1000 Hz_40ms_envelope_correlation": 0.9957147383326059, "300–1000 Hz_40ms_modulation_mae_db": 0.3943240109972747, "1000–3000 Hz_10ms_log_rms_mae_db": 2.022618119354189, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13291951210327996, "1000–3000 Hz_10ms_envelope_correlation": 0.9881395741533334, "1000–3000 Hz_10ms_modulation_mae_db": 0.245701023260991, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2202234319755099, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07788385662193906, "1000–3000 Hz_40ms_envelope_correlation": 0.996126377057249, "1000–3000 Hz_40ms_modulation_mae_db": 0.414037291148762, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2104593122081106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19867219342795403, "3000–6000 Hz_10ms_envelope_correlation": 0.9762510750285529, "3000–6000 Hz_10ms_modulation_mae_db": 0.572696668526164, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5093411181951206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12225050216787958, "3000–6000 Hz_40ms_envelope_correlation": 0.990764780142369, "3000–6000 Hz_40ms_modulation_mae_db": 0.6322117526544339, "mr_stft_db": 5.994087337565396, "band_mae_db": 0.23140811666760594, "rms_gain_error_db": -0.07424423054090497, "output_peak": 0.6005197522181027}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3586215777543584, "全帯域_10ms_rms_envelope_nrmse": 0.10345279987918084, "全帯域_10ms_envelope_correlation": 0.990439107780812, "全帯域_10ms_modulation_mae_db": 0.2805021192260307, "全帯域_40ms_log_rms_mae_db": 0.6706101502347706, "全帯域_40ms_rms_envelope_nrmse": 0.04637682595958823, "全帯域_40ms_envelope_correlation": 0.9979307654334021, "全帯域_40ms_modulation_mae_db": 0.13799173503199577, "300–1000 Hz_10ms_log_rms_mae_db": 1.7869660472507503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14455487024379676, "300–1000 Hz_10ms_envelope_correlation": 0.9846012098960943, "300–1000 Hz_10ms_modulation_mae_db": 0.863776219043875, "300–1000 Hz_40ms_log_rms_mae_db": 1.019532932831637, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06074689350053036, "300–1000 Hz_40ms_envelope_correlation": 0.9970826297259837, "300–1000 Hz_40ms_modulation_mae_db": 0.27247131805964725, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9410071445608368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12345834235564473, "1000–3000 Hz_10ms_envelope_correlation": 0.9895701402367232, "1000–3000 Hz_10ms_modulation_mae_db": 0.8473931869333372, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2027436282344146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07116240862744992, "1000–3000 Hz_40ms_envelope_correlation": 0.9964789757544776, "1000–3000 Hz_40ms_modulation_mae_db": 1.1233763551941893, "3000–6000 Hz_10ms_log_rms_mae_db": 2.284221943532331, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19489036497692477, "3000–6000 Hz_10ms_envelope_correlation": 0.9759869180352262, "3000–6000 Hz_10ms_modulation_mae_db": 0.9966770967302008, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5154853730829587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10023136389655331, "3000–6000 Hz_40ms_envelope_correlation": 0.9934321745824525, "3000–6000 Hz_40ms_modulation_mae_db": 1.3895903506691276, "mr_stft_db": 5.987469768698097, "band_mae_db": 0.12990381237199844, "rms_gain_error_db": -0.06195484960778054, "output_peak": 0.5771956997915925}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3789225772515665, "全帯域_10ms_rms_envelope_nrmse": 0.10623131601246598, "全帯域_10ms_envelope_correlation": 0.9901907749716143, "全帯域_10ms_modulation_mae_db": 0.4915119792467657, "全帯域_40ms_log_rms_mae_db": 0.6628543365385744, "全帯域_40ms_rms_envelope_nrmse": 0.05874484992933992, "全帯域_40ms_envelope_correlation": 0.9970794749423301, "全帯域_40ms_modulation_mae_db": 0.62120487792205, "300–1000 Hz_10ms_log_rms_mae_db": 1.800760127702186, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1477537211006772, "300–1000 Hz_10ms_envelope_correlation": 0.9842733055965589, "300–1000 Hz_10ms_modulation_mae_db": 0.9461976461802458, "300–1000 Hz_40ms_log_rms_mae_db": 1.0740385899539204, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09277274177986108, "300–1000 Hz_40ms_envelope_correlation": 0.9938054679753052, "300–1000 Hz_40ms_modulation_mae_db": 0.6553635704400752, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0046004643450215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13678098988091514, "1000–3000 Hz_10ms_envelope_correlation": 0.9872323417116576, "1000–3000 Hz_10ms_modulation_mae_db": 0.6004937097359228, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1904592424841234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07137902893701506, "1000–3000 Hz_40ms_envelope_correlation": 0.9966451678510726, "1000–3000 Hz_40ms_modulation_mae_db": 0.8046395420965916, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2101198420874453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1964452227247293, "3000–6000 Hz_10ms_envelope_correlation": 0.9757242778170899, "3000–6000 Hz_10ms_modulation_mae_db": 1.1638566938703192, "3000–6000 Hz_40ms_log_rms_mae_db": 1.520278879611836, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10849220297590416, "3000–6000 Hz_40ms_envelope_correlation": 0.9926419458467899, "3000–6000 Hz_40ms_modulation_mae_db": 1.4162731882892459, "mr_stft_db": 5.983774083940697, "band_mae_db": 0.23676603393540172, "rms_gain_error_db": -0.20240161337268445, "output_peak": 0.59065358497058}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2668562650048831, "全帯域_10ms_rms_envelope_nrmse": 0.10231223133752337, "全帯域_10ms_envelope_correlation": 0.990667273963174, "全帯域_10ms_modulation_mae_db": 0.1193128543084601, "全帯域_40ms_log_rms_mae_db": 0.5902568804589897, "全帯域_40ms_rms_envelope_nrmse": 0.045006440142659976, "全帯域_40ms_envelope_correlation": 0.9980168691957374, "全帯域_40ms_modulation_mae_db": 0.2875642768241314, "300–1000 Hz_10ms_log_rms_mae_db": 1.8190796196433656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13473045739349218, "300–1000 Hz_10ms_envelope_correlation": 0.986638908679271, "300–1000 Hz_10ms_modulation_mae_db": 0.3506164501205675, "300–1000 Hz_40ms_log_rms_mae_db": 1.0911498284147023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05980617948911474, "300–1000 Hz_40ms_envelope_correlation": 0.9971620837491761, "300–1000 Hz_40ms_modulation_mae_db": 0.6246195782041937, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0515981639449117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1507776633755317, "1000–3000 Hz_10ms_envelope_correlation": 0.9841839430898505, "1000–3000 Hz_10ms_modulation_mae_db": 0.1801865777912696, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2635845430521562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08334737319473749, "1000–3000 Hz_40ms_envelope_correlation": 0.9948588503238738, "1000–3000 Hz_40ms_modulation_mae_db": 0.28671170094712223, "3000–6000 Hz_10ms_log_rms_mae_db": 2.251903935319195, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18573867011341513, "3000–6000 Hz_10ms_envelope_correlation": 0.9787770684769468, "3000–6000 Hz_10ms_modulation_mae_db": 0.6303373032724636, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5140691237784454, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1064002795657585, "3000–6000 Hz_40ms_envelope_correlation": 0.9926819863163044, "3000–6000 Hz_40ms_modulation_mae_db": 0.9431001093441055, "mr_stft_db": 5.9791606648262645, "band_mae_db": 0.18179999849331938, "rms_gain_error_db": -0.0040496966051799315, "output_peak": 0.6086018691193836}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3953771602806313, "全帯域_10ms_rms_envelope_nrmse": 0.09890177374325959, "全帯域_10ms_envelope_correlation": 0.9912639833378645, "全帯域_10ms_modulation_mae_db": 0.2257206432603254, "全帯域_40ms_log_rms_mae_db": 0.6642596701089042, "全帯域_40ms_rms_envelope_nrmse": 0.04893878097508646, "全帯域_40ms_envelope_correlation": 0.9976776003752426, "全帯域_40ms_modulation_mae_db": 0.4062249731465329, "300–1000 Hz_10ms_log_rms_mae_db": 1.8427886866672596, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13603835219294194, "300–1000 Hz_10ms_envelope_correlation": 0.9863696660368569, "300–1000 Hz_10ms_modulation_mae_db": 0.6714831544363543, "300–1000 Hz_40ms_log_rms_mae_db": 1.1188765279339647, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06654393699130773, "300–1000 Hz_40ms_envelope_correlation": 0.9964974680565314, "300–1000 Hz_40ms_modulation_mae_db": 0.3526126918346333, "1000–3000 Hz_10ms_log_rms_mae_db": 2.025250625454509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1507747698030443, "1000–3000 Hz_10ms_envelope_correlation": 0.9841931492917169, "1000–3000 Hz_10ms_modulation_mae_db": 0.8343190283090297, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2364093200898196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0922614536388486, "1000–3000 Hz_40ms_envelope_correlation": 0.9937151597142311, "1000–3000 Hz_40ms_modulation_mae_db": 1.309691911368034, "3000–6000 Hz_10ms_log_rms_mae_db": 2.208829079570324, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19627119970084095, "3000–6000 Hz_10ms_envelope_correlation": 0.975787198245151, "3000–6000 Hz_10ms_modulation_mae_db": 0.44553014012599085, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4997063355689604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11289424823269026, "3000–6000 Hz_40ms_envelope_correlation": 0.9920901347427167, "3000–6000 Hz_40ms_modulation_mae_db": 0.5940748224021855, "mr_stft_db": 5.993728085060902, "band_mae_db": 0.14328608001357593, "rms_gain_error_db": -0.053433504890313135, "output_peak": 0.5870888998479732}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3290060652594822, "全帯域_10ms_rms_envelope_nrmse": 0.10912427554143495, "全帯域_10ms_envelope_correlation": 0.9894043937522852, "全帯域_10ms_modulation_mae_db": 0.23840670683721676, "全帯域_40ms_log_rms_mae_db": 0.6280332136196399, "全帯域_40ms_rms_envelope_nrmse": 0.05531387582953343, "全帯域_40ms_envelope_correlation": 0.9970156752489747, "全帯域_40ms_modulation_mae_db": 0.23357313904469526, "300–1000 Hz_10ms_log_rms_mae_db": 1.7529030963825756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14653881368924704, "300–1000 Hz_10ms_envelope_correlation": 0.9845487214159215, "300–1000 Hz_10ms_modulation_mae_db": 0.6033419786652813, "300–1000 Hz_40ms_log_rms_mae_db": 1.068843981874768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07369725802705589, "300–1000 Hz_40ms_envelope_correlation": 0.9958319513027631, "300–1000 Hz_40ms_modulation_mae_db": 0.3280864549826767, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0606070746623355, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14620686388160625, "1000–3000 Hz_10ms_envelope_correlation": 0.9856229975203138, "1000–3000 Hz_10ms_modulation_mae_db": 0.8295934711053073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.298890532769515, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09332056580813218, "1000–3000 Hz_40ms_envelope_correlation": 0.9943026013867086, "1000–3000 Hz_40ms_modulation_mae_db": 1.0405510439739791, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1575092674457848, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1882910832243379, "3000–6000 Hz_10ms_envelope_correlation": 0.9778258040450746, "3000–6000 Hz_10ms_modulation_mae_db": 0.4524552992824522, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4678464659770147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11512634833976759, "3000–6000 Hz_40ms_envelope_correlation": 0.9917027329102186, "3000–6000 Hz_40ms_modulation_mae_db": 0.8374453680739372, "mr_stft_db": 5.986317176290664, "band_mae_db": 0.23415220365911985, "rms_gain_error_db": -0.01706243345350572, "output_peak": 0.5810406228703109}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.422842413402213, "全帯域_10ms_rms_envelope_nrmse": 0.10189030772533099, "全帯域_10ms_envelope_correlation": 0.9907896131762473, "全帯域_10ms_modulation_mae_db": 0.22350899540469057, "全帯域_40ms_log_rms_mae_db": 0.696561266796047, "全帯域_40ms_rms_envelope_nrmse": 0.0539083965964495, "全帯域_40ms_envelope_correlation": 0.9971782378879467, "全帯域_40ms_modulation_mae_db": 0.18427182616862725, "300–1000 Hz_10ms_log_rms_mae_db": 1.7450595894306071, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15459300297670603, "300–1000 Hz_10ms_envelope_correlation": 0.9824260214069447, "300–1000 Hz_10ms_modulation_mae_db": 0.8592793842203169, "300–1000 Hz_40ms_log_rms_mae_db": 1.0467085940022178, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08898252654507605, "300–1000 Hz_40ms_envelope_correlation": 0.9936809677924601, "300–1000 Hz_40ms_modulation_mae_db": 0.43666911407801523, "1000–3000 Hz_10ms_log_rms_mae_db": 1.967555436161055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14504956961902873, "1000–3000 Hz_10ms_envelope_correlation": 0.9854252515062206, "1000–3000 Hz_10ms_modulation_mae_db": 0.16776757418674038, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1576325795800442, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08760837266491936, "1000–3000 Hz_40ms_envelope_correlation": 0.9942926731195657, "1000–3000 Hz_40ms_modulation_mae_db": 0.13659652375157397, "3000–6000 Hz_10ms_log_rms_mae_db": 2.206579862559824, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19741881442526293, "3000–6000 Hz_10ms_envelope_correlation": 0.9754928233724135, "3000–6000 Hz_10ms_modulation_mae_db": 0.19141891553315438, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4742069368080866, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11216247928833063, "3000–6000 Hz_40ms_envelope_correlation": 0.9917133454450594, "3000–6000 Hz_40ms_modulation_mae_db": 0.3585060365094342, "mr_stft_db": 5.974220918610215, "band_mae_db": 0.12318143072065156, "rms_gain_error_db": -0.0023902307539454505, "output_peak": 0.5768052583560093}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84640, "output_samples": 84640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.279969259371027, "spectral_convergence": 0.3902316564613535, "spectral_convergence_status": "ok", "amplitude_floor": 1.2121984197282631e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.74741425544167, "spectral_convergence": 0.34511540254321266, "spectral_convergence_status": "ok", "amplitude_floor": 7.081048851832571e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.970126548793155, "spectral_convergence": 0.3870557358629913, "spectral_convergence_status": "ok", "amplitude_floor": 5.009180364479501e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.291800237989852e-05, "output_power": 7.867172722102832e-05, "delta_db": 0.32983922655322045, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025616304806087414, "output_power": 0.00025271129274640847, "delta_db": -0.05891832845908673, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038484147909669214, "output_power": 0.0003768600572720146, "delta_db": -0.09101765355108235, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002720810642199278, "output_power": 0.00023773584762191427, "delta_db": -0.586036447348104, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.5951933766776914e-05, "output_power": 4.656974244077e-05, "delta_db": 0.058000437721250825, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.6368160234296276e-05, "output_power": 2.657493212120966e-05, "delta_db": 0.03392334753796738, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19295590686178532, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16391658277535862, "pitch": null, "output_rms_dbfs": -29.917731415163182, "output_peak": 0.5884663073557219, "reference_rms_dbfs": -29.753814832387818, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.999170021201951}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00018085971712132593, "log_rms_mae_db": 1.2841603305593785, "rms_envelope_nrmse": 0.09248533605115368, "envelope_correlation": 0.9925543335158971, "rms_gain_error_db": -0.16399871576777855, "reference_mean_power": 0.0010593154039372696, "output_mean_power": 0.001020059291281503}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 5.491358535179395e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002450078451267542, "output_power": 0.000227067746729846, "reference_active": true, "delta_db": -0.3302454032609773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.385267752666162e-05, "output_power": 8.259965571138119e-05, "reference_active": true, "delta_db": -0.0653869764443254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4530628910939782e-05, "output_power": 1.2100571852952806e-05, "reference_active": true, "delta_db": -0.7947851688855969}], "active_band_count": 3, "active_band_mae_db": 0.3968058495302999}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00014864066644098504, "log_rms_mae_db": 0.6433859419793871, "rms_envelope_nrmse": 0.05454392917226289, "envelope_correlation": 0.9973901382424853, "rms_gain_error_db": -0.1640568169089374, "reference_mean_power": 0.0010653029350414017, "output_mean_power": 0.0010258112127606803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 4.840611714703517e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022860940218938906, "output_power": 0.00021087149144688892, "reference_active": true, "delta_db": -0.3507621826477263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8392485325428115e-05, "output_power": 3.8484679922307515e-05, "reference_active": true, "delta_db": 0.010416519327126382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1188041388515455e-06, "output_power": 9.452079096646491e-07, "reference_active": true, "delta_db": -0.7322671708358313}], "active_band_count": 3, "active_band_mae_db": 0.36448195760356134}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013930286763496605, "log_rms_mae_db": 1.7617659116019724, "rms_envelope_nrmse": 0.12244180905406767, "envelope_correlation": 0.9889761947425362, "rms_gain_error_db": -0.08519719573240137, "reference_mean_power": 0.0005230599489297363, "output_mean_power": 0.000512898876196079}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 3.2510596296193775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001486768204336917, "output_power": 0.00013344734664805648, "reference_active": true, "delta_db": -0.4693332173440503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.417268122003869e-05, "output_power": 5.389508330627144e-05, "reference_active": true, "delta_db": -0.022311835140361776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.810050335491182e-06, "output_power": 6.2962187655306705e-06, "reference_active": true, "delta_db": -0.9357402348837844}], "active_band_count": 3, "active_band_mae_db": 0.4757950957893988}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00011121771745577188, "log_rms_mae_db": 1.0663710840193041, "rms_envelope_nrmse": 0.08185805622496677, "envelope_correlation": 0.9946798762347286, "rms_gain_error_db": -0.08519892330060733, "reference_mean_power": 0.0005260459982805144, "output_mean_power": 0.000515826712733908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.9069199937213013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014215463322126283, "output_power": 0.00012853294846693324, "reference_active": true, "delta_db": -0.43746549037274685}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2639933776267858e-05, "output_power": 2.3720938695156727e-05, "reference_active": true, "delta_db": 0.20256718949939945}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.874281484444505e-07, "output_power": 5.668652613217239e-07, "reference_active": true, "delta_db": -0.8374746808585882}], "active_band_count": 3, "active_band_mae_db": 0.4925024535769115}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010000130422823669, "log_rms_mae_db": 1.9979265347204502, "rms_envelope_nrmse": 0.15340584420229753, "envelope_correlation": 0.9852107096377635, "rms_gain_error_db": -0.5151169060353433, "reference_mean_power": 0.0002857956004291321, "output_mean_power": 0.0002538305251733582}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.149273337334192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04716085544297e-05, "output_power": 8.67725773795984e-05, "reference_active": true, "delta_db": -0.18129814632605937}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.048324869712114e-05, "output_power": 1.8268954867566752e-05, "reference_active": true, "delta_db": -0.49685135118418233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.217073445558843e-06, "output_power": 7.752265710055959e-06, "reference_active": true, "delta_db": -0.25288518981201835}], "active_band_count": 3, "active_band_mae_db": 0.3103448957740867}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 8.407177238322724e-05, "log_rms_mae_db": 1.2613452049080682, "rms_envelope_nrmse": 0.09803382268510147, "envelope_correlation": 0.9950818760195023, "rms_gain_error_db": -0.5151181799662747, "reference_mean_power": 0.00028742710472676696, "output_mean_power": 0.00025527947747044845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 1.945560884297879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.432816729085121e-05, "output_power": 8.033761673949e-05, "reference_active": true, "delta_db": -0.21053717884394701}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0296492874768435e-05, "output_power": 9.169228815378701e-06, "reference_active": true, "delta_db": -0.5035651274249228}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.97290149260778e-07, "output_power": 4.821988134435367e-07, "reference_active": true, "delta_db": -0.9295921488879337}], "active_band_count": 3, "active_band_mae_db": 0.5478981517189345}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 5.528232384026309e-05, "log_rms_mae_db": 2.208410330442552, "rms_envelope_nrmse": 0.18358959952312898, "envelope_correlation": 0.9787521077755452, "rms_gain_error_db": -0.12058170881794228, "reference_mean_power": 3.470766577035859e-05, "output_mean_power": 3.3757263656218266e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.6534989625639428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5332454345272327e-05, "output_power": 1.500704101896162e-05, "reference_active": true, "delta_db": -0.09316610579458795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.828360450157823e-06, "output_power": 3.1481318001043753e-06, "reference_active": true, "delta_db": -0.8495991469404176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1450631921352025e-06, "output_power": 5.709456912737605e-07, "reference_active": true, "delta_db": -3.0223465473333846}], "active_band_count": 3, "active_band_mae_db": 1.32170393335613}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 4.431598350735027e-05, "log_rms_mae_db": 1.4829754540804743, "rms_envelope_nrmse": 0.1055501233144711, "envelope_correlation": 0.9926363877360761, "rms_gain_error_db": -0.12058442275189998, "reference_mean_power": 3.49057787168493e-05, "output_mean_power": 3.3949930448359896e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.464962908732827e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.55740476492126e-05, "output_power": 1.4909554863893178e-05, "reference_active": true, "delta_db": -0.18936821643381346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.044495719000328e-06, "output_power": 1.531891478855547e-06, "reference_active": true, "delta_db": -1.2535820510393616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.011419545771505e-07, "output_power": 4.3929002718699055e-08, "reference_active": true, "delta_db": -3.6217999818293696}], "active_band_count": 3, "active_band_mae_db": 1.6882500831008482}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8331453451853934, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6c149c6e6f3c9b8c21f625b1c4b875d104a95d6f7e8f9f0c0b0ec5941454ca0e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9861972633934475, "standard_deviation": 0.0019342499598498577, "minimum": 0.9841839430898505, "maximum": 0.9895701402367232, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0088954454004133, "standard_deviation": 0.04037325862785717, "minimum": 1.9410071445608368, "maximum": 2.0606070746623355, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5019749333870855, "standard_deviation": 0.30809249908102315, "minimum": 0.16776757418674038, "maximum": 0.8473931869333372, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14242169440266855, "standard_deviation": 0.010441665205795553, "minimum": 0.12345834235564473, "maximum": 0.15340584420229753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951877101533351, "standard_deviation": 0.0011048534885089304, "minimum": 0.9937151597142311, "maximum": 0.9966451678510726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2289110603867064, "standard_deviation": 0.04555522253816223, "minimum": 1.1576325795800442, "maximum": 1.298890532769515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7079378150248984, "standard_deviation": 0.4260336135757708, "minimum": 0.13659652375157397, "maximum": 1.309691911368034, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08437461027226789, "standard_deviation": 0.010183293230980169, "minimum": 0.07116240862744992, "maximum": 0.09803382268510147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9768246590995, "standard_deviation": 0.001394940405799472, "minimum": 0.9754928233724135, "maximum": 0.9787770684769468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.217254196645696, "standard_deviation": 0.03706162343820897, "minimum": 2.1575092674457848, "maximum": 2.284221943532331, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7218345063371093, "standard_deviation": 0.3949012869461297, "minimum": 0.19141891553315438, "maximum": 1.32170393335613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19266464351457424, "standard_deviation": 0.005860280592249506, "minimum": 0.18358959952312898, "maximum": 0.19867219342795403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922079359652483, "standard_deviation": 0.000818696279895582, "minimum": 0.990764780142369, "maximum": 0.9934321745824525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4979887108878622, "standard_deviation": 0.020329139751565507, "minimum": 1.4678464659770147, "maximum": 1.520278879611836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9824314638804147, "standard_deviation": 0.4686610888821896, "minimum": 0.3585060365094342, "maximum": 1.6882500831008482, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11038844347266939, "standard_deviation": 0.006747054663519253, "minimum": 0.10023136389655331, "maximum": 0.12225050216787958, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9855876482050319, "standard_deviation": 0.002021735569710118, "minimum": 0.9824260214069447, "maximum": 0.9889761947425362, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.788904193851006, "standard_deviation": 0.033980020308866564, "minimum": 1.7450595894306071, "maximum": 1.8427886866672596, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6901425768874204, "standard_deviation": 0.2060120254632736, "minimum": 0.3506164501205675, "maximum": 0.9461976461802458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14003470307903385, "standard_deviation": 0.010204636726021126, "minimum": 0.12244180905406767, "maximum": 0.15459300297670603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955568978961942, "standard_deviation": 0.0013755781621846814, "minimum": 0.9936809677924601, "maximum": 0.9971620837491761, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0712143640236407, "standard_deviation": 0.029631341289880358, "minimum": 1.019532932831637, "maximum": 1.1188765279339647, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44458114902167845, "standard_deviation": 0.13803669366281054, "minimum": 0.27247131805964725, "maximum": 0.6553635704400752, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0748159797311276, "standard_deviation": 0.012341120345486218, "minimum": 0.05980617948911474, "maximum": 0.09277274177986108, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18418169784043226, "standard_deviation": 0.0476815788364311, "minimum": 0.12318143072065156, "maximum": 0.23676603393540172, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.987241007024274, "standard_deviation": 0.008272120005734046, "minimum": 5.974220918610215, "maximum": 5.999170021201951, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5887964993162093, "standard_deviation": 0.011178314786512662, "minimum": 0.5768052583560093, "maximum": 0.6086018691193836, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0724316427499591, "standard_deviation": 0.07404950536861742, "minimum": -0.20240161337268445, "maximum": -0.0023902307539454505, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9908065414164784, "standard_deviation": 0.0009162821002609024, "minimum": 0.9894043937522852, "maximum": 0.9925543335158971, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3494394323210206, "standard_deviation": 0.05352780302372356, "minimum": 1.2668562650048831, "maximum": 1.422842413402213, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26122682103026285, "standard_deviation": 0.12916793247538141, "minimum": 0.11404542042831384, "maximum": 0.4915119792467657, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1017579500689898, "standard_deviation": 0.0050140976259346135, "minimum": 0.09248533605115368, "maximum": 0.10912427554143495, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974404710621161, "standard_deviation": 0.00038787671319294513, "minimum": 0.9970156752489747, "maximum": 0.9980168691957374, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6469031951556283, "standard_deviation": 0.03364690958383574, "minimum": 0.5902568804589897, "maximum": 0.696561266796047, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2900686303240659, "standard_deviation": 0.17244274955687083, "minimum": 0.08523625685093317, "maximum": 0.62120487792205, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05205187995710682, "standard_deviation": 0.0047655064147529924, "minimum": 0.045006440142659976, "maximum": 0.05874484992933992, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5105-28241-0019", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.1059412956237793, "predicted_mos": 2.1059412956237793, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "26558fc4a5fd6d67937094ace1956c59c8140cc850b13e41aaf1fc54d3f44eb4", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.045146800111979, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4066082057331495, "全帯域_10ms_rms_envelope_nrmse": 0.09709391110523638, "全帯域_10ms_envelope_correlation": 0.9915844859647783, "全帯域_10ms_modulation_mae_db": 0.3280545881440784, "全帯域_40ms_log_rms_mae_db": 0.6739191990363109, "全帯域_40ms_rms_envelope_nrmse": 0.04995507408059172, "全帯域_40ms_envelope_correlation": 0.9975558837921411, "全帯域_40ms_modulation_mae_db": 0.4648078487582796, "300–1000 Hz_10ms_log_rms_mae_db": 1.7241132677159292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13982436738398282, "300–1000 Hz_10ms_envelope_correlation": 0.9857259420535128, "300–1000 Hz_10ms_modulation_mae_db": 0.3849420941662034, "300–1000 Hz_40ms_log_rms_mae_db": 1.0304046356142926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08332657958286424, "300–1000 Hz_40ms_envelope_correlation": 0.9947750657175435, "300–1000 Hz_40ms_modulation_mae_db": 0.7305469422219497, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9232505786645504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13820409797931055, "1000–3000 Hz_10ms_envelope_correlation": 0.986934184771975, "1000–3000 Hz_10ms_modulation_mae_db": 0.40107486318530833, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1129275399258023, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07521295415909728, "1000–3000 Hz_40ms_envelope_correlation": 0.9958723594821712, "1000–3000 Hz_40ms_modulation_mae_db": 0.5060378576460437, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1045608937855538, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22890733296783444, "3000–6000 Hz_10ms_envelope_correlation": 0.9732016996707192, "3000–6000 Hz_10ms_modulation_mae_db": 1.0786375296289157, "3000–6000 Hz_40ms_log_rms_mae_db": 1.374018102457265, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13824062037863316, "3000–6000 Hz_40ms_envelope_correlation": 0.9921510293796298, "3000–6000 Hz_40ms_modulation_mae_db": 0.8308512945627151, "mr_stft_db": 5.920078593037346, "band_mae_db": 0.2989498372517127, "rms_gain_error_db": -0.021909437539657697, "output_peak": 0.6130131182252418}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3396007856603926, "全帯域_10ms_rms_envelope_nrmse": 0.1100175854585707, "全帯域_10ms_envelope_correlation": 0.9893499021943978, "全帯域_10ms_modulation_mae_db": 0.7152511735405577, "全帯域_40ms_log_rms_mae_db": 0.6955455954143116, "全帯域_40ms_rms_envelope_nrmse": 0.06788253230100427, "全帯域_40ms_envelope_correlation": 0.9957851161433966, "全帯域_40ms_modulation_mae_db": 0.7052234155724233, "300–1000 Hz_10ms_log_rms_mae_db": 1.6989999611445772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14880148367017232, "300–1000 Hz_10ms_envelope_correlation": 0.9840228981153469, "300–1000 Hz_10ms_modulation_mae_db": 0.9504785964939665, "300–1000 Hz_40ms_log_rms_mae_db": 0.9929125106845199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10284186375543908, "300–1000 Hz_40ms_envelope_correlation": 0.9921507341330595, "300–1000 Hz_40ms_modulation_mae_db": 0.8134410153298045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9245906165944162, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15701650363784853, "1000–3000 Hz_10ms_envelope_correlation": 0.9831510838382681, "1000–3000 Hz_10ms_modulation_mae_db": 0.49555229677845175, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1167687601539338, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08701506180824252, "1000–3000 Hz_40ms_envelope_correlation": 0.9950282192140789, "1000–3000 Hz_40ms_modulation_mae_db": 0.7273415164201201, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0964969362372123, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20442218934936743, "3000–6000 Hz_10ms_envelope_correlation": 0.9782916426652127, "3000–6000 Hz_10ms_modulation_mae_db": 0.9667201981243659, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3438609158346226, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1352426401966147, "3000–6000 Hz_40ms_envelope_correlation": 0.9917570056169989, "3000–6000 Hz_40ms_modulation_mae_db": 1.1164395862893979, "mr_stft_db": 5.925882273104373, "band_mae_db": 0.3509827282873703, "rms_gain_error_db": -0.16832353544172382, "output_peak": 0.5560442713459524}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3210010112722077, "全帯域_10ms_rms_envelope_nrmse": 0.09382649196295628, "全帯域_10ms_envelope_correlation": 0.9922661162449227, "全帯域_10ms_modulation_mae_db": 0.0954164686258897, "全帯域_40ms_log_rms_mae_db": 0.5928471639582238, "全帯域_40ms_rms_envelope_nrmse": 0.04153324973136107, "全帯域_40ms_envelope_correlation": 0.9983820545973354, "全帯域_40ms_modulation_mae_db": 0.18434623614660509, "300–1000 Hz_10ms_log_rms_mae_db": 1.703601623732827, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12237984282162447, "300–1000 Hz_10ms_envelope_correlation": 0.9894802145438578, "300–1000 Hz_10ms_modulation_mae_db": 0.48267081441685794, "300–1000 Hz_40ms_log_rms_mae_db": 0.9417357511595018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06584863114373594, "300–1000 Hz_40ms_envelope_correlation": 0.9968843601131171, "300–1000 Hz_40ms_modulation_mae_db": 0.3862063923045775, "1000–3000 Hz_10ms_log_rms_mae_db": 1.894670425716922, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14371106269169123, "1000–3000 Hz_10ms_envelope_correlation": 0.9858756149396891, "1000–3000 Hz_10ms_modulation_mae_db": 1.3148462659989777, "1000–3000 Hz_40ms_log_rms_mae_db": 1.117997313553582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07786757541193591, "1000–3000 Hz_40ms_envelope_correlation": 0.9959726886399913, "1000–3000 Hz_40ms_modulation_mae_db": 1.6993078503018886, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0999084589874535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2024290091290414, "3000–6000 Hz_10ms_envelope_correlation": 0.9741464522262958, "3000–6000 Hz_10ms_modulation_mae_db": 0.6225110840635165, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3766238716637904, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09908145989375758, "3000–6000 Hz_40ms_envelope_correlation": 0.9934085648513235, "3000–6000 Hz_40ms_modulation_mae_db": 0.8988045344483323, "mr_stft_db": 5.926866045347261, "band_mae_db": 0.22595865897107548, "rms_gain_error_db": 0.05379001503717536, "output_peak": 0.579431106371067}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3727384770129643, "全帯域_10ms_rms_envelope_nrmse": 0.0971523340881169, "全帯域_10ms_envelope_correlation": 0.9916582563167177, "全帯域_10ms_modulation_mae_db": 0.1883377453411107, "全帯域_40ms_log_rms_mae_db": 0.6078208923477791, "全帯域_40ms_rms_envelope_nrmse": 0.04481939566413747, "全帯域_40ms_envelope_correlation": 0.998222971423264, "全帯域_40ms_modulation_mae_db": 0.18326710146048183, "300–1000 Hz_10ms_log_rms_mae_db": 1.6342910719415675, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12144382964489184, "300–1000 Hz_10ms_envelope_correlation": 0.9892525541457813, "300–1000 Hz_10ms_modulation_mae_db": 0.30480028762952827, "300–1000 Hz_40ms_log_rms_mae_db": 0.9468853424144071, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0586892546025214, "300–1000 Hz_40ms_envelope_correlation": 0.9974946952844858, "300–1000 Hz_40ms_modulation_mae_db": 0.21103014691364605, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9624738626327647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15618350501595726, "1000–3000 Hz_10ms_envelope_correlation": 0.9831195486821741, "1000–3000 Hz_10ms_modulation_mae_db": 0.19933307370646092, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1895869859806938, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08951900321956945, "1000–3000 Hz_40ms_envelope_correlation": 0.9943531735073439, "1000–3000 Hz_40ms_modulation_mae_db": 0.12842710039241287, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0658678098426435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19399600977706505, "3000–6000 Hz_10ms_envelope_correlation": 0.9761745884743082, "3000–6000 Hz_10ms_modulation_mae_db": 0.8012152818738438, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3277124146963437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10637311662441626, "3000–6000 Hz_40ms_envelope_correlation": 0.9925304807062101, "3000–6000 Hz_40ms_modulation_mae_db": 0.9894640409624489, "mr_stft_db": 5.919978094119379, "band_mae_db": 0.1450050567599305, "rms_gain_error_db": -0.12746786237084498, "output_peak": 0.5930203614876481}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.376293246655521, "全帯域_10ms_rms_envelope_nrmse": 0.10890402740546935, "全帯域_10ms_envelope_correlation": 0.9894055361749607, "全帯域_10ms_modulation_mae_db": 0.6227271685854306, "全帯域_40ms_log_rms_mae_db": 0.6564765041484084, "全帯域_40ms_rms_envelope_nrmse": 0.054182479485796166, "全帯域_40ms_envelope_correlation": 0.9971378831440716, "全帯域_40ms_modulation_mae_db": 0.5422383141866421, "300–1000 Hz_10ms_log_rms_mae_db": 1.7034512527728034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14620228842911756, "300–1000 Hz_10ms_envelope_correlation": 0.9842431838890892, "300–1000 Hz_10ms_modulation_mae_db": 1.2552472945218829, "300–1000 Hz_40ms_log_rms_mae_db": 0.9831564435737095, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07893794966077433, "300–1000 Hz_40ms_envelope_correlation": 0.9950839571688943, "300–1000 Hz_40ms_modulation_mae_db": 0.938210701312011, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9575604390178394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15245556517333667, "1000–3000 Hz_10ms_envelope_correlation": 0.9838442577876977, "1000–3000 Hz_10ms_modulation_mae_db": 0.9834247946108589, "1000–3000 Hz_40ms_log_rms_mae_db": 1.137162015941484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0757709082687853, "1000–3000 Hz_40ms_envelope_correlation": 0.9958784272200135, "1000–3000 Hz_40ms_modulation_mae_db": 1.280019056443513, "3000–6000 Hz_10ms_log_rms_mae_db": 2.079596336182687, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20975468644380688, "3000–6000 Hz_10ms_envelope_correlation": 0.976126591417196, "3000–6000 Hz_10ms_modulation_mae_db": 0.9768148996790806, "3000–6000 Hz_40ms_log_rms_mae_db": 1.332747691087797, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12815915418055515, "3000–6000 Hz_40ms_envelope_correlation": 0.9920748291573088, "3000–6000 Hz_40ms_modulation_mae_db": 1.0600877049194595, "mr_stft_db": 5.9219606091763985, "band_mae_db": 0.20351333260942775, "rms_gain_error_db": -0.044445464193229416, "output_peak": 0.6021820469881398}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3792341256135339, "全帯域_10ms_rms_envelope_nrmse": 0.09435449347709453, "全帯域_10ms_envelope_correlation": 0.9921723658301258, "全帯域_10ms_modulation_mae_db": 0.3586284221851952, "全帯域_40ms_log_rms_mae_db": 0.6655062981803525, "全帯域_40ms_rms_envelope_nrmse": 0.05226443243636007, "全帯域_40ms_envelope_correlation": 0.9975465448658978, "全帯域_40ms_modulation_mae_db": 0.628249716948239, "300–1000 Hz_10ms_log_rms_mae_db": 1.6928820640569424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13749729473916686, "300–1000 Hz_10ms_envelope_correlation": 0.9864398754616469, "300–1000 Hz_10ms_modulation_mae_db": 1.1079813185845915, "300–1000 Hz_40ms_log_rms_mae_db": 0.9467831589661991, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0770326677020545, "300–1000 Hz_40ms_envelope_correlation": 0.995932049592093, "300–1000 Hz_40ms_modulation_mae_db": 0.624096361227657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9445380351362083, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1319442682152771, "1000–3000 Hz_10ms_envelope_correlation": 0.9879171204889919, "1000–3000 Hz_10ms_modulation_mae_db": 0.44867066968762814, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1569140123284911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07022944326042976, "1000–3000 Hz_40ms_envelope_correlation": 0.9963855593449303, "1000–3000 Hz_40ms_modulation_mae_db": 0.7800373559737741, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1496831553868905, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23154518434668697, "3000–6000 Hz_10ms_envelope_correlation": 0.9664346055534793, "3000–6000 Hz_10ms_modulation_mae_db": 0.34849592642362576, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3901536098513252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10339979532384543, "3000–6000 Hz_40ms_envelope_correlation": 0.9929164041936582, "3000–6000 Hz_40ms_modulation_mae_db": 0.48664451372238376, "mr_stft_db": 5.933469257105057, "band_mae_db": 0.19203649017245525, "rms_gain_error_db": -0.13934833868726507, "output_peak": 0.5687876731292685}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.48089949285162, "全帯域_10ms_rms_envelope_nrmse": 0.10476980051422928, "全帯域_10ms_envelope_correlation": 0.9902650856127209, "全帯域_10ms_modulation_mae_db": 0.26710273100792664, "全帯域_40ms_log_rms_mae_db": 0.7769353537227615, "全帯域_40ms_rms_envelope_nrmse": 0.05245923741202997, "全帯域_40ms_envelope_correlation": 0.9974869317113065, "全帯域_40ms_modulation_mae_db": 0.23915453722277902, "300–1000 Hz_10ms_log_rms_mae_db": 1.7025328422667825, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13270556099890626, "300–1000 Hz_10ms_envelope_correlation": 0.9870853685341855, "300–1000 Hz_10ms_modulation_mae_db": 0.6579987776982169, "300–1000 Hz_40ms_log_rms_mae_db": 1.039019259172954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07544715496729251, "300–1000 Hz_40ms_envelope_correlation": 0.9956202916705789, "300–1000 Hz_40ms_modulation_mae_db": 0.3869581111843458, "1000–3000 Hz_10ms_log_rms_mae_db": 1.931110683769145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15304811931667148, "1000–3000 Hz_10ms_envelope_correlation": 0.9840291664536023, "1000–3000 Hz_10ms_modulation_mae_db": 0.8699461525005132, "1000–3000 Hz_40ms_log_rms_mae_db": 1.120847557454385, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07616807966509409, "1000–3000 Hz_40ms_envelope_correlation": 0.9963877221692967, "1000–3000 Hz_40ms_modulation_mae_db": 1.1293326219683197, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1339346597644213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20938731042845124, "3000–6000 Hz_10ms_envelope_correlation": 0.9722484177179178, "3000–6000 Hz_10ms_modulation_mae_db": 1.05816378768083, "3000–6000 Hz_40ms_log_rms_mae_db": 1.40743812920317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1116183183456112, "3000–6000 Hz_40ms_envelope_correlation": 0.9920556697912483, "3000–6000 Hz_40ms_modulation_mae_db": 1.2930763344015035, "mr_stft_db": 5.928565746563091, "band_mae_db": 0.27147363287674603, "rms_gain_error_db": -0.1265489165381783, "output_peak": 0.5942023675716682}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2988965503707237, "全帯域_10ms_rms_envelope_nrmse": 0.09768363869231421, "全帯域_10ms_envelope_correlation": 0.9920220082443514, "全帯域_10ms_modulation_mae_db": 0.47320365801367786, "全帯域_40ms_log_rms_mae_db": 0.625671282404369, "全帯域_40ms_rms_envelope_nrmse": 0.054258610912308866, "全帯域_40ms_envelope_correlation": 0.997917393528885, "全帯域_40ms_modulation_mae_db": 0.43795699667494814, "300–1000 Hz_10ms_log_rms_mae_db": 1.6625660009284533, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14033258900276477, "300–1000 Hz_10ms_envelope_correlation": 0.9862245277822134, "300–1000 Hz_10ms_modulation_mae_db": 0.4991288672395897, "300–1000 Hz_40ms_log_rms_mae_db": 0.9225004368256564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08428735910112845, "300–1000 Hz_40ms_envelope_correlation": 0.9955066899407519, "300–1000 Hz_40ms_modulation_mae_db": 0.4723427923385324, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9414931217933533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14384930015026245, "1000–3000 Hz_10ms_envelope_correlation": 0.9865845604419681, "1000–3000 Hz_10ms_modulation_mae_db": 0.9944225946720993, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1426179950535131, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08558607416336418, "1000–3000 Hz_40ms_envelope_correlation": 0.9959866848524784, "1000–3000 Hz_40ms_modulation_mae_db": 1.3642954405817342, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0848558449034296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18768285547811414, "3000–6000 Hz_10ms_envelope_correlation": 0.977869667539776, "3000–6000 Hz_10ms_modulation_mae_db": 0.26907280426585956, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3744260135803459, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09659467319168195, "3000–6000 Hz_40ms_envelope_correlation": 0.9937582193488059, "3000–6000 Hz_40ms_modulation_mae_db": 0.5198672170733224, "mr_stft_db": 5.919125843247656, "band_mae_db": 0.2854541276280315, "rms_gain_error_db": -0.2425001234266988, "output_peak": 0.6028297911826117}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84640, "output_samples": 84640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.219952780322276, "spectral_convergence": 0.3839597973017034, "spectral_convergence_status": "ok", "amplitude_floor": 1.2121984197282631e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.655443368334002, "spectral_convergence": 0.3367933788463125, "spectral_convergence_status": "ok", "amplitude_floor": 7.081048851832571e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.884839630455757, "spectral_convergence": 0.3760941595077057, "spectral_convergence_status": "ok", "amplitude_floor": 5.009180364479501e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.291800237989852e-05, "output_power": 8.186428527689639e-05, "delta_db": 0.5025971206967508, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025616304806087414, "output_power": 0.0002505358084530061, "delta_db": -0.0964667497893848, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038484147909669214, "output_power": 0.000365758173128256, "delta_db": -0.22087835616319287, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002720810642199278, "output_power": 0.0002756218955272628, "delta_db": 0.05615397524170765, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.5951933766776914e-05, "output_power": 4.81023307302798e-05, "delta_db": 0.1986232775495348, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.6368160234296276e-05, "output_power": 3.111557160548817e-05, "delta_db": 0.7189795440697053, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2989498372517127, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.021909437539657697, "pitch": null, "output_rms_dbfs": -29.775724269927476, "output_peak": 0.6130131182252418, "reference_rms_dbfs": -29.753814832387818, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.920078593037346}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00018085971712132593, "log_rms_mae_db": 1.4066082057331495, "rms_envelope_nrmse": 0.09709391110523638, "envelope_correlation": 0.9915844859647783, "rms_gain_error_db": -0.021929183279150388, "reference_mean_power": 0.0010593154039372696, "output_mean_power": 0.0010539799983924215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 5.491358535179395e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002450078451267542, "output_power": 0.00023923829104817914, "reference_active": true, "delta_db": -0.10349299199472295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.385267752666162e-05, "output_power": 7.727768825093366e-05, "reference_active": true, "delta_db": -0.35462813004235655}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4530628910939782e-05, "output_power": 1.2873011211333701e-05, "reference_active": true, "delta_db": -0.5260426423951557}], "active_band_count": 3, "active_band_mae_db": 0.3280545881440784}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00014864066644098504, "log_rms_mae_db": 0.6739191990363109, "rms_envelope_nrmse": 0.04995507408059172, "envelope_correlation": 0.9975558837921411, "rms_gain_error_db": -0.02184645721780749, "reference_mean_power": 0.0010653029350414017, "output_mean_power": 0.0010599575626536089}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 4.840611714703517e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022860940218938906, "output_power": 0.00022068363302239186, "reference_active": true, "delta_db": -0.15323963026596898}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8392485325428115e-05, "output_power": 3.725917083855293e-05, "reference_active": true, "delta_db": -0.13013041017095558}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1188041388515455e-06, "output_power": 8.662609028448775e-07, "reference_active": true, "delta_db": -1.111053505837914}], "active_band_count": 3, "active_band_mae_db": 0.4648078487582796}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013930286763496605, "log_rms_mae_db": 1.7241132677159292, "rms_envelope_nrmse": 0.13982436738398282, "envelope_correlation": 0.9857259420535128, "rms_gain_error_db": -0.2003472445575011, "reference_mean_power": 0.0005230599489297363, "output_mean_power": 0.0004994784353711752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 3.2510596296193775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001486768204336917, "output_power": 0.00013698097168400656, "reference_active": true, "delta_db": -0.35583022224092936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.417268122003869e-05, "output_power": 4.587929715613203e-05, "reference_active": true, "delta_db": -0.7216357492505897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.810050335491182e-06, "output_power": 7.672162658585291e-06, "reference_active": true, "delta_db": -0.07736031100709109}], "active_band_count": 3, "active_band_mae_db": 0.3849420941662034}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00011121771745577188, "log_rms_mae_db": 1.0304046356142926, "rms_envelope_nrmse": 0.08332657958286424, "envelope_correlation": 0.9947750657175435, "rms_gain_error_db": -0.200348786092084, "reference_mean_power": 0.0005260459982805144, "output_mean_power": 0.0005023296840621266}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.9069199937213013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014215463322126283, "output_power": 0.0001298850881140774, "reference_active": true, "delta_db": -0.39201725817087163}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2639933776267858e-05, "output_power": 2.0871316277158327e-05, "reference_active": true, "delta_db": -0.35325312884874527}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.874281484444505e-07, "output_power": 4.927087414860172e-07, "reference_active": true, "delta_db": -1.446370439646232}], "active_band_count": 3, "active_band_mae_db": 0.7305469422219497}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010000130422823669, "log_rms_mae_db": 1.9232505786645504, "rms_envelope_nrmse": 0.13820409797931055, "envelope_correlation": 0.986934184771975, "rms_gain_error_db": 0.07068948137399872, "reference_mean_power": 0.0002857956004291321, "output_mean_power": 0.00029048551888068315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.149273337334192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04716085544297e-05, "output_power": 9.480858133611215e-05, "reference_active": true, "delta_db": 0.20335336085620878}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.048324869712114e-05, "output_power": 1.9918673075659675e-05, "reference_active": true, "delta_db": -0.12138434403734427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.217073445558843e-06, "output_power": 6.712255513840964e-06, "reference_active": true, "delta_db": -0.8784868846623719}], "active_band_count": 3, "active_band_mae_db": 0.40107486318530833}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 8.407177238322724e-05, "log_rms_mae_db": 1.1129275399258023, "rms_envelope_nrmse": 0.07521295415909728, "envelope_correlation": 0.9958723594821712, "rms_gain_error_db": 0.07068866796387939, "reference_mean_power": 0.00028742710472676696, "output_mean_power": 0.0002921437415190972}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 1.945560884297879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.432816729085121e-05, "output_power": 8.750399191100259e-05, "reference_active": true, "delta_db": 0.16055204002264348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0296492874768435e-05, "output_power": 9.786120170350329e-06, "reference_active": true, "delta_db": -0.22078778846315433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.97290149260778e-07, "output_power": 4.597354336020207e-07, "reference_active": true, "delta_db": -1.1367737444523331}], "active_band_count": 3, "active_band_mae_db": 0.5060378576460437}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 5.528232384026309e-05, "log_rms_mae_db": 2.1045608937855538, "rms_envelope_nrmse": 0.22890733296783444, "envelope_correlation": 0.9732016996707192, "rms_gain_error_db": 0.6523421722114431, "reference_mean_power": 3.470766577035859e-05, "output_mean_power": 4.033291618092281e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.6534989625639428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5332454345272327e-05, "output_power": 1.783498188098532e-05, "reference_active": true, "delta_db": 0.6566099228049713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.828360450157823e-06, "output_power": 4.494564764390293e-06, "reference_active": true, "delta_db": 0.6967482177887534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1450631921352025e-06, "output_power": 1.7663830937946437e-06, "reference_active": true, "delta_db": 1.882554448293022}], "active_band_count": 3, "active_band_mae_db": 1.0786375296289157}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 4.431598350735027e-05, "log_rms_mae_db": 1.374018102457265, "rms_envelope_nrmse": 0.13824062037863316, "envelope_correlation": 0.9921510293796298, "rms_gain_error_db": 0.6523406528257224, "reference_mean_power": 3.49057787168493e-05, "output_mean_power": 4.056312412342892e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.464962908732827e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.55740476492126e-05, "output_power": 1.8818894888018802e-05, "reference_active": true, "delta_db": 0.8219261743634019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.044495719000328e-06, "output_power": 2.4198725022014726e-06, "reference_active": true, "delta_db": 0.7320627903951042}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.011419545771505e-07, "output_power": 1.255416505041133e-07, "reference_active": true, "delta_db": 0.9385649189296391}], "active_band_count": 3, "active_band_mae_db": 0.8308512945627151}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8262035700932118, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "95b5b1cde5f4fcb490bd606b44d8a38cb2c39189bd8360c9483e6c80d569ed58", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9851819421755458, "standard_deviation": 0.0018706843026890306, "minimum": 0.9831195486821741, "maximum": 0.9879171204889919, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9349609704156499, "standard_deviation": 0.021662721772855914, "minimum": 1.894670425716922, "maximum": 1.9624738626327647, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7134088388925373, "standard_deviation": 0.38125402713834433, "minimum": 0.19933307370646092, "maximum": 1.3148462659989777, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1470515527725444, "standard_deviation": 0.009069381430471847, "minimum": 0.1319442682152771, "maximum": 0.15701650363784853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995733104303788, "standard_deviation": 0.0006987242074707437, "minimum": 0.9943531735073439, "maximum": 0.9963877221692967, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1368527725489856, "standard_deviation": 0.026197501472643454, "minimum": 1.1129275399258023, "maximum": 1.1895869859806938, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9518498499659759, "standard_deviation": 0.5105594790951234, "minimum": 0.12842710039241287, "maximum": 1.6993078503018886, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07967113749456481, "standard_deviation": 0.00682073714573361, "minimum": 0.07022944326042976, "maximum": 0.08951900321956945, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9743117081581131, "standard_deviation": 0.003832724119037261, "minimum": 0.9664346055534793, "maximum": 0.9782916426652127, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1018630118862864, "standard_deviation": 0.027866911970805784, "minimum": 2.0658678098426435, "maximum": 2.1496831553868905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7652039389675047, "standard_deviation": 0.31868117258209006, "minimum": 0.26907280426585956, "maximum": 1.0786375296289157, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20851557224004597, "standard_deviation": 0.015339510546456865, "minimum": 0.18768285547811414, "maximum": 0.23154518434668697, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9925815253806479, "standard_deviation": 0.0007153198255861068, "minimum": 0.9917570056169989, "maximum": 0.9937582193488059, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3658725935468325, "standard_deviation": 0.028310356930013214, "minimum": 1.3277124146963437, "maximum": 1.40743812920317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8994044032974454, "standard_deviation": 0.28161037229295194, "minimum": 0.48664451372238376, "maximum": 1.2930763344015035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11483872226688943, "standard_deviation": 0.016627970336388256, "minimum": 0.09659467319168195, "maximum": 0.13824062037863316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9865593205657042, "standard_deviation": 0.002024163186899466, "minimum": 0.9840228981153469, "maximum": 0.9894802145438578, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6903047605699855, "standard_deviation": 0.02836345503621384, "minimum": 1.6342910719415675, "maximum": 1.7241132677159292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7054060063438548, "standard_deviation": 0.3549735661579743, "minimum": 0.30480028762952827, "maximum": 1.2552472945218829, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13614840708632836, "standard_deviation": 0.010083023825331461, "minimum": 0.12144382964489184, "maximum": 0.14880148367017232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954309804525655, "standard_deviation": 0.0016024328605194597, "minimum": 0.9921507341330595, "maximum": 0.9974946952844858, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9754246923014052, "standard_deviation": 0.04310024920129307, "minimum": 0.9225004368256564, "maximum": 1.039019259172954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5703540578540656, "standard_deviation": 0.2476241655437423, "minimum": 0.21103014691364605, "maximum": 0.938210701312011, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0783014325644763, "standard_deviation": 0.01316482211167621, "minimum": 0.0586892546025214, "maximum": 0.10284186375543908, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2466717330695937, "standard_deviation": 0.06691764028490949, "minimum": 0.1450050567599305, "maximum": 0.3509827282873703, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.92449080771257, "standard_deviation": 0.005068284868296753, "minimum": 5.919125843247656, "maximum": 5.933469257105057, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5886888420376997, "standard_deviation": 0.019162070602416814, "minimum": 0.5560442713459524, "maximum": 0.6130131182252418, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10209420789505283, "standard_deviation": 0.09318394736637454, "minimum": -0.2425001234266988, "maximum": 0.05379001503717536, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.991090469572872, "standard_deviation": 0.0012268820483220496, "minimum": 0.9893499021943978, "maximum": 0.9922661162449227, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3719089868962642, "standard_deviation": 0.05621742403719592, "minimum": 1.2988965503707237, "maximum": 1.48089949285162, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38109024443048334, "standard_deviation": 0.211663584229091, "minimum": 0.0954164686258897, "maximum": 0.7152511735405577, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10047528533799846, "standard_deviation": 0.006463983502938778, "minimum": 0.09382649196295628, "maximum": 0.1100175854585707, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975043474007873, "standard_deviation": 0.0008064197429081614, "minimum": 0.9957851161433966, "maximum": 0.9983820545973354, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6618402861515646, "standard_deviation": 0.05801154446529715, "minimum": 0.5928471639582238, "maximum": 0.7769353537227615, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42315552087129976, "standard_deviation": 0.20219980516999553, "minimum": 0.18326710146048183, "maximum": 0.7052234155724233, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0521693765029487, "standard_deviation": 0.007814513515592937, "minimum": 0.04153324973136107, "maximum": 0.06788253230100427, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5142-33396-0019", "language": "en", "style": "normal", "speaker": "librispeech_5142", "source_utterance_id": "5142-33396-0019", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 79760, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5142-33396-0019.wav", "source_file_sha256": "b67ef3cd0afe2691656adf0b768c15d77d779b6301a29b53372511547d154386", "decoded_pcm_sha256": "f3e12cfe88aca2fb306552ad3fe2c213b91a4bfc2f0aa979df8a1f01006ba85d", "original_source_sha256": "2025e96df477d9ebc46b91bae88c601221bbbb2ab12d9fa83ceced80b0f9b72f", "corpus_entry_id": "librispeech_5142-33396-0019"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 2.7795631885528564, "predicted_mos": 2.7795631885528564, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "f9fb1a92a39a8c50e49a864ca20c1761a0d76de0b668e405071c19a50398c553", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.495495570001367, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5142-33396-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f9fb1a92a39a8c50e49a864ca20c1761a0d76de0b668e405071c19a50398c553", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "4114fe2feba025e8426b7975a60af7eba2625ecf55c140d4d942ca0dc0361c57"}, "rows": [{"case": "librispeech_5142-33396-0019", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.204514980316162, "predicted_mos": 2.204514980316162, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "47a7f852e6661043bf59c0a3bf63854ba92bb9dcf0b0b8c265856ef3c0028f7f", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.362505881999823, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5142-33396-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "47a7f852e6661043bf59c0a3bf63854ba92bb9dcf0b0b8c265856ef3c0028f7f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6531825736006394, "全帯域_10ms_rms_envelope_nrmse": 0.04928888432550335, "全帯域_10ms_envelope_correlation": 0.9972609698567187, "全帯域_10ms_modulation_mae_db": 0.2371647883206948, "全帯域_40ms_log_rms_mae_db": 0.32839641740450726, "全帯域_40ms_rms_envelope_nrmse": 0.02732449220114666, "全帯域_40ms_envelope_correlation": 0.9991740381661354, "全帯域_40ms_modulation_mae_db": 0.2987812598165768, "300–1000 Hz_10ms_log_rms_mae_db": 1.3007927168722102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06507743078811146, "300–1000 Hz_10ms_envelope_correlation": 0.996187392575549, "300–1000 Hz_10ms_modulation_mae_db": 0.27378787543327293, "300–1000 Hz_40ms_log_rms_mae_db": 0.831379438847356, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03794394800913527, "300–1000 Hz_40ms_envelope_correlation": 0.9986491233962429, "300–1000 Hz_40ms_modulation_mae_db": 0.32108367837275936, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1564891749206798, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09728123655942439, "1000–3000 Hz_10ms_envelope_correlation": 0.9917298750905675, "1000–3000 Hz_10ms_modulation_mae_db": 0.2957707989654643, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7091937999291317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054100638790565925, "1000–3000 Hz_40ms_envelope_correlation": 0.9975206992186629, "1000–3000 Hz_40ms_modulation_mae_db": 0.2730525525829827, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2555932781568533, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11822938488510293, "3000–6000 Hz_10ms_envelope_correlation": 0.9877521205508717, "3000–6000 Hz_10ms_modulation_mae_db": 0.5026175797028688, "3000–6000 Hz_40ms_log_rms_mae_db": 0.762758587279392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06706495681287376, "3000–6000 Hz_40ms_envelope_correlation": 0.9967089223732511, "3000–6000 Hz_40ms_modulation_mae_db": 0.2586759824779791, "mr_stft_db": 5.48564027438776, "band_mae_db": 0.17794666566244852, "rms_gain_error_db": -0.08592253511444631, "output_peak": 0.36437919278005204}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6674745538572359, "全帯域_10ms_rms_envelope_nrmse": 0.049898478220595635, "全帯域_10ms_envelope_correlation": 0.997229563795054, "全帯域_10ms_modulation_mae_db": 0.49454883133266997, "全帯域_40ms_log_rms_mae_db": 0.3274136767048151, "全帯域_40ms_rms_envelope_nrmse": 0.026348936838792133, "全帯域_40ms_envelope_correlation": 0.9992877336664827, "全帯域_40ms_modulation_mae_db": 0.4845150742585023, "300–1000 Hz_10ms_log_rms_mae_db": 1.264181317732183, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06168652483536612, "300–1000 Hz_10ms_envelope_correlation": 0.9966040017976892, "300–1000 Hz_10ms_modulation_mae_db": 0.5639468840456715, "300–1000 Hz_40ms_log_rms_mae_db": 0.8409834061917922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034864810884306215, "300–1000 Hz_40ms_envelope_correlation": 0.9988963518291487, "300–1000 Hz_40ms_modulation_mae_db": 0.504341382170633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.168509795558298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10070033654603452, "1000–3000 Hz_10ms_envelope_correlation": 0.9912656067863099, "1000–3000 Hz_10ms_modulation_mae_db": 0.2113625617951449, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6926448597414343, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059668719894586664, "1000–3000 Hz_40ms_envelope_correlation": 0.9971187366353491, "1000–3000 Hz_40ms_modulation_mae_db": 0.13587248406921074, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3224904250810274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12814871422959215, "3000–6000 Hz_10ms_envelope_correlation": 0.9857386725014522, "3000–6000 Hz_10ms_modulation_mae_db": 0.2705559631313215, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8392168658015285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07835157825658434, "3000–6000 Hz_40ms_envelope_correlation": 0.9954002374906078, "3000–6000 Hz_40ms_modulation_mae_db": 0.49147097585964455, "mr_stft_db": 5.488007672793724, "band_mae_db": 0.18170818315173753, "rms_gain_error_db": -0.09976287866855854, "output_peak": 0.40014143149112236}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6772112108795886, "全帯域_10ms_rms_envelope_nrmse": 0.04691660706458713, "全帯域_10ms_envelope_correlation": 0.9975032192693565, "全帯域_10ms_modulation_mae_db": 0.3407253572639357, "全帯域_40ms_log_rms_mae_db": 0.33378340503658777, "全帯域_40ms_rms_envelope_nrmse": 0.022464952278019958, "全帯域_40ms_envelope_correlation": 0.9994506025683784, "全帯域_40ms_modulation_mae_db": 0.5742640725324671, "300–1000 Hz_10ms_log_rms_mae_db": 1.246197251834057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06024922678482586, "300–1000 Hz_10ms_envelope_correlation": 0.9967132276822321, "300–1000 Hz_10ms_modulation_mae_db": 0.2858508548982457, "300–1000 Hz_40ms_log_rms_mae_db": 0.7929995304598838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02968586566583459, "300–1000 Hz_40ms_envelope_correlation": 0.9991581950058271, "300–1000 Hz_40ms_modulation_mae_db": 0.41787126600319696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.181089194024154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10032312872399017, "1000–3000 Hz_10ms_envelope_correlation": 0.9913731206637058, "1000–3000 Hz_10ms_modulation_mae_db": 0.28794925606170335, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7065561671581211, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05746300318361915, "1000–3000 Hz_40ms_envelope_correlation": 0.9974240042974888, "1000–3000 Hz_40ms_modulation_mae_db": 0.5555966579825218, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2760268283411857, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12394413667723495, "3000–6000 Hz_10ms_envelope_correlation": 0.9867491675839184, "3000–6000 Hz_10ms_modulation_mae_db": 0.4699347633497055, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7688068589556628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07871043864247299, "3000–6000 Hz_40ms_envelope_correlation": 0.9953198567434121, "3000–6000 Hz_40ms_modulation_mae_db": 0.6019653396761034, "mr_stft_db": 5.502050556788191, "band_mae_db": 0.18807464707088475, "rms_gain_error_db": -0.07389506371412227, "output_peak": 0.37955129288834455}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6376791352840688, "全帯域_10ms_rms_envelope_nrmse": 0.044699858817777646, "全帯域_10ms_envelope_correlation": 0.9976972041257812, "全帯域_10ms_modulation_mae_db": 0.3901213989471251, "全帯域_40ms_log_rms_mae_db": 0.34091774154049126, "全帯域_40ms_rms_envelope_nrmse": 0.024993630903693068, "全帯域_40ms_envelope_correlation": 0.9992405230063165, "全帯域_40ms_modulation_mae_db": 0.4873040601067331, "300–1000 Hz_10ms_log_rms_mae_db": 1.2296013283175415, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06494633543689603, "300–1000 Hz_10ms_envelope_correlation": 0.9961753788984303, "300–1000 Hz_10ms_modulation_mae_db": 0.32931574613761033, "300–1000 Hz_40ms_log_rms_mae_db": 0.7795974422731944, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036957905699290944, "300–1000 Hz_40ms_envelope_correlation": 0.9986791544936435, "300–1000 Hz_40ms_modulation_mae_db": 0.24532430609445824, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1738650104764012, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10506406324441696, "1000–3000 Hz_10ms_envelope_correlation": 0.9901150297310337, "1000–3000 Hz_10ms_modulation_mae_db": 0.16346813239392557, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7378418746962798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0580301860719739, "1000–3000 Hz_40ms_envelope_correlation": 0.9968682388506309, "1000–3000 Hz_40ms_modulation_mae_db": 0.2656057382682854, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3417762067340966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14433608181616664, "3000–6000 Hz_10ms_envelope_correlation": 0.9817300294027651, "3000–6000 Hz_10ms_modulation_mae_db": 0.5415607681465797, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8720492499314114, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09462655559192408, "3000–6000 Hz_40ms_envelope_correlation": 0.9928435258050694, "3000–6000 Hz_40ms_modulation_mae_db": 0.5532706508214188, "mr_stft_db": 5.474990199746299, "band_mae_db": 0.14409654514285708, "rms_gain_error_db": -0.04387470772107756, "output_peak": 0.37580671827724144}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6677454727217006, "全帯域_10ms_rms_envelope_nrmse": 0.04768763339572523, "全帯域_10ms_envelope_correlation": 0.99742469659332, "全帯域_10ms_modulation_mae_db": 0.2585059789441263, "全帯域_40ms_log_rms_mae_db": 0.333853130160692, "全帯域_40ms_rms_envelope_nrmse": 0.02529710616304077, "全帯域_40ms_envelope_correlation": 0.9992866630939418, "全帯域_40ms_modulation_mae_db": 0.3630294357192077, "300–1000 Hz_10ms_log_rms_mae_db": 1.267705201926344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06194799714298096, "300–1000 Hz_10ms_envelope_correlation": 0.9965399239599166, "300–1000 Hz_10ms_modulation_mae_db": 0.11980576790560277, "300–1000 Hz_40ms_log_rms_mae_db": 0.8246112649728808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035191514179192876, "300–1000 Hz_40ms_envelope_correlation": 0.998839185241356, "300–1000 Hz_40ms_modulation_mae_db": 0.24484448331095043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.188098047527538, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09314709333182673, "1000–3000 Hz_10ms_envelope_correlation": 0.9924149910497978, "1000–3000 Hz_10ms_modulation_mae_db": 0.06053963638908725, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7334885389452049, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05105884454959241, "1000–3000 Hz_40ms_envelope_correlation": 0.9978196761806608, "1000–3000 Hz_40ms_modulation_mae_db": 0.15652965575619796, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3063437377773226, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13442036207610344, "3000–6000 Hz_10ms_envelope_correlation": 0.9847216966173301, "3000–6000 Hz_10ms_modulation_mae_db": 0.734427622997072, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8239122036178493, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08754261094468496, "3000–6000 Hz_40ms_envelope_correlation": 0.994555910781097, "3000–6000 Hz_40ms_modulation_mae_db": 0.5278935946899654, "mr_stft_db": 5.4842868072336195, "band_mae_db": 0.1975953944068164, "rms_gain_error_db": -0.07751844372447361, "output_peak": 0.3457581037128943}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.703186212131072, "全帯域_10ms_rms_envelope_nrmse": 0.04981945158166143, "全帯域_10ms_envelope_correlation": 0.9972825758683026, "全帯域_10ms_modulation_mae_db": 0.23819382847959755, "全帯域_40ms_log_rms_mae_db": 0.3787757962751256, "全帯域_40ms_rms_envelope_nrmse": 0.0287177084530661, "全帯域_40ms_envelope_correlation": 0.9991745656972578, "全帯域_40ms_modulation_mae_db": 0.3647804408065844, "300–1000 Hz_10ms_log_rms_mae_db": 1.3646902853608351, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06641495205679737, "300–1000 Hz_10ms_envelope_correlation": 0.9960727997804243, "300–1000 Hz_10ms_modulation_mae_db": 0.2626215846285018, "300–1000 Hz_40ms_log_rms_mae_db": 0.8845732607177428, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03562474931767331, "300–1000 Hz_40ms_envelope_correlation": 0.9988860862538906, "300–1000 Hz_40ms_modulation_mae_db": 0.25132585027258014, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1902206023971502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10112711536075548, "1000–3000 Hz_10ms_envelope_correlation": 0.9913800817851849, "1000–3000 Hz_10ms_modulation_mae_db": 0.3454368960574404, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7464333836152259, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06593440650300302, "1000–3000 Hz_40ms_envelope_correlation": 0.9965578026520326, "1000–3000 Hz_40ms_modulation_mae_db": 0.2715541144580419, "3000–6000 Hz_10ms_log_rms_mae_db": 1.33405634894057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13504658610063502, "3000–6000 Hz_10ms_envelope_correlation": 0.9838916273532748, "3000–6000 Hz_10ms_modulation_mae_db": 0.6261863730683385, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8768949305135328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0858146312993366, "3000–6000 Hz_40ms_envelope_correlation": 0.9940455959414131, "3000–6000 Hz_40ms_modulation_mae_db": 0.6129537703608477, "mr_stft_db": 5.481222431718707, "band_mae_db": 0.22679160985766222, "rms_gain_error_db": -0.11108535726786951, "output_peak": 0.35622373455641965}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6627776200877441, "全帯域_10ms_rms_envelope_nrmse": 0.04819668924890731, "全帯域_10ms_envelope_correlation": 0.9973387231507277, "全帯域_10ms_modulation_mae_db": 0.2525684150685687, "全帯域_40ms_log_rms_mae_db": 0.3119818453909679, "全帯域_40ms_rms_envelope_nrmse": 0.027111761991157567, "全帯域_40ms_envelope_correlation": 0.999129888401407, "全帯域_40ms_modulation_mae_db": 0.16028691997951194, "300–1000 Hz_10ms_log_rms_mae_db": 1.2640896749052646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06081380744424202, "300–1000 Hz_10ms_envelope_correlation": 0.9966629815515882, "300–1000 Hz_10ms_modulation_mae_db": 0.1507378643557791, "300–1000 Hz_40ms_log_rms_mae_db": 0.7885443110268487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03327011489796141, "300–1000 Hz_40ms_envelope_correlation": 0.9989424655372233, "300–1000 Hz_40ms_modulation_mae_db": 0.1331829088391708, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1716626307089943, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09646172164166587, "1000–3000 Hz_10ms_envelope_correlation": 0.9917395490779681, "1000–3000 Hz_10ms_modulation_mae_db": 0.614348080624973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.731806146541986, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060208509925740526, "1000–3000 Hz_40ms_envelope_correlation": 0.9966770022795863, "1000–3000 Hz_40ms_modulation_mae_db": 0.326091751936256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3147851164577131, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13878863893278748, "3000–6000 Hz_10ms_envelope_correlation": 0.9839218243009128, "3000–6000 Hz_10ms_modulation_mae_db": 0.5903591841972967, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8349760384470685, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09226919776637357, "3000–6000 Hz_40ms_envelope_correlation": 0.994108999503731, "3000–6000 Hz_40ms_modulation_mae_db": 0.2493099974801507, "mr_stft_db": 5.489270071092477, "band_mae_db": 0.18341075171232293, "rms_gain_error_db": -0.05667840308935969, "output_peak": 0.3520150634322808}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6705531202460714, "全帯域_10ms_rms_envelope_nrmse": 0.05215002672523808, "全帯域_10ms_envelope_correlation": 0.9968773269120458, "全帯域_10ms_modulation_mae_db": 0.10812607007935043, "全帯域_40ms_log_rms_mae_db": 0.3768088365264468, "全帯域_40ms_rms_envelope_nrmse": 0.03069512527463229, "全帯域_40ms_envelope_correlation": 0.9988691057531381, "全帯域_40ms_modulation_mae_db": 0.21606413290686907, "300–1000 Hz_10ms_log_rms_mae_db": 1.2574516169848267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06956000776750575, "300–1000 Hz_10ms_envelope_correlation": 0.9956035955602743, "300–1000 Hz_10ms_modulation_mae_db": 0.16687600972326555, "300–1000 Hz_40ms_log_rms_mae_db": 0.8410723835404976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04101827602253392, "300–1000 Hz_40ms_envelope_correlation": 0.9983600053047156, "300–1000 Hz_40ms_modulation_mae_db": 0.2981622000990704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1733461144057562, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09970355105674306, "1000–3000 Hz_10ms_envelope_correlation": 0.9917128779239763, "1000–3000 Hz_10ms_modulation_mae_db": 0.46727706043734774, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7535615922876261, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062472272779911485, "1000–3000 Hz_40ms_envelope_correlation": 0.9970996957038899, "1000–3000 Hz_40ms_modulation_mae_db": 0.41234188165596364, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2715894363388314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13017187313580256, "3000–6000 Hz_10ms_envelope_correlation": 0.9848122682304463, "3000–6000 Hz_10ms_modulation_mae_db": 0.40665516804585883, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7896347201476824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0845807818098722, "3000–6000 Hz_40ms_envelope_correlation": 0.993800918947271, "3000–6000 Hz_40ms_modulation_mae_db": 0.6645169601561619, "mr_stft_db": 5.504423392709718, "band_mae_db": 0.20091689349756361, "rms_gain_error_db": -0.06243977038933916, "output_peak": 0.36183071377248305}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79760, "output_samples": 79760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.013367299726588, "spectral_convergence": 0.21153769152737648, "spectral_convergence_status": "ok", "amplitude_floor": 8.669731759960068e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.306923620552491, "spectral_convergence": 0.2122930675930767, "spectral_convergence_status": "ok", "amplitude_floor": 7.258845275738271e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.136629902884201, "spectral_convergence": 0.21748652851115433, "spectral_convergence_status": "ok", "amplitude_floor": 5.316334054535358e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.354150869521306e-05, "output_power": 5.698595028086326e-05, "delta_db": 0.2707719024302063, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007507214796955326, "output_power": 0.00074408436069301, "delta_db": -0.038566655590369314, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006075331721980808, "output_power": 0.0006003441880475901, "delta_db": -0.05169685459923942, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029803976258038344, "output_power": 0.0002914836352826494, "delta_db": -0.09660031508585008, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002296848467085225, "output_power": 0.00021436723591145945, "delta_db": -0.2997393575747473, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011626570661661749, "output_power": 0.00010824831493240962, "delta_db": -0.3103049086942785, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17794666566244852, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08592253511444631, "pitch": null, "output_rms_dbfs": -26.956142486823143, "output_peak": 0.36437919278005204, "reference_rms_dbfs": -26.870219951708698, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.48564027438776}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.000134521265514355, "log_rms_mae_db": 0.6531825736006394, "rms_envelope_nrmse": 0.04928888432550335, "envelope_correlation": 0.9972609698567187, "rms_gain_error_db": -0.08592134498626242, "reference_mean_power": 0.0020578494395295825, "output_mean_power": 0.0020175367876832506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 5.569199266163003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039409678072813294, "output_power": 0.0003849301262332215, "reference_active": true, "delta_db": -0.10220985126058181}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.428667833313107e-05, "output_power": 5.0622979362980244e-05, "reference_active": true, "delta_db": -0.3034556754309981}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.373716545850661e-06, "output_power": 6.872322870068015e-06, "reference_active": true, "delta_db": -0.3058288382705045}], "active_band_count": 3, "active_band_mae_db": 0.2371647883206948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00012498981593367327, "log_rms_mae_db": 0.32839641740450726, "rms_envelope_nrmse": 0.02732449220114666, "envelope_correlation": 0.9991740381661354, "rms_gain_error_db": -0.0859218213940198, "reference_mean_power": 0.0020703160518770356, "output_mean_power": 0.0020297589601844337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 4.869724313408462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003604621899100367, "output_power": 0.0003523613711256294, "reference_active": true, "delta_db": -0.09871425572197474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.585585870556879e-05, "output_power": 2.3585397807692436e-05, "reference_active": true, "delta_db": -0.39915760057546834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932164327326652e-07, "output_power": 3.587438516531109e-07, "reference_active": true, "delta_db": -0.39847192315228736}], "active_band_count": 3, "active_band_mae_db": 0.2987812598165768}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.00011089795767449635, "log_rms_mae_db": 1.3007927168722102, "rms_envelope_nrmse": 0.06507743078811146, "envelope_correlation": 0.996187392575549, "rms_gain_error_db": -0.06288959041559737, "reference_mean_power": 0.001090370195617981, "output_mean_power": 0.0010746944667955351}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 3.4652335334077796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023090422993351253, "output_power": 0.00022626679221576007, "reference_active": true, "delta_db": -0.08811068924776398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.174073740387618e-05, "output_power": 3.693886148154337e-05, "reference_active": true, "delta_db": -0.5307661155161332}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.679949697911218e-06, "output_power": 4.9033158250156975e-06, "reference_active": true, "delta_db": 0.2024868215359216}], "active_band_count": 3, "active_band_mae_db": 0.27378787543327293}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00010160775009558359, "log_rms_mae_db": 0.831379438847356, "rms_envelope_nrmse": 0.03794394800913527, "envelope_correlation": 0.9986491233962429, "rms_gain_error_db": -0.0628894861833232, "reference_mean_power": 0.0010969783403958974, "output_mean_power": 0.0010812076353978895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 3.031226473561925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002111816957536322, "output_power": 0.00020754250072876516, "reference_active": true, "delta_db": -0.07549227522267216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.975380679045621e-05, "output_power": 1.69759571707018e-05, "reference_active": true, "delta_db": -0.6581653081236473}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.773024908316542e-07, "output_power": 2.630234230070089e-07, "reference_active": true, "delta_db": -0.22959345177195858}], "active_band_count": 3, "active_band_mae_db": 0.32108367837275936}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 7.327188751847531e-05, "log_rms_mae_db": 1.1564891749206798, "rms_envelope_nrmse": 0.09728123655942439, "envelope_correlation": 0.9917298750905675, "rms_gain_error_db": -0.17971413865850933, "reference_mean_power": 0.0003949163505892704, "output_mean_power": 0.0003789079358110755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 2.1879792460690968e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000156993183862756, "output_power": 0.00015202594711136364, "reference_active": true, "delta_db": -0.1396307943574399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3083527056032412e-05, "output_power": 1.2275986072727839e-05, "reference_active": true, "delta_db": -0.2766844979167994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5497810931869906e-06, "output_power": 2.8418595268109788e-06, "reference_active": true, "delta_db": 0.47099710462215366}], "active_band_count": 3, "active_band_mae_db": 0.2957707989654643}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 7.131903461227575e-05, "log_rms_mae_db": 0.7091937999291317, "rms_envelope_nrmse": 0.054100638790565925, "envelope_correlation": 0.9975206992186629, "rms_gain_error_db": -0.17971501631247322, "reference_mean_power": 0.0003973070610776639, "output_mean_power": 0.0003812016589049505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 1.9730997112139496e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014395632169995676, "output_power": 0.00013970892740395265, "reference_active": true, "delta_db": -0.1300658281779283}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.343232783051936e-06, "output_power": 5.750562218735606e-06, "reference_active": true, "delta_db": -0.42600342667107516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5197840492846542e-07, "output_power": 1.4304511849519055e-07, "reference_active": true, "delta_db": -0.2630884028999447}], "active_band_count": 3, "active_band_mae_db": 0.2730525525829827}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 4.2584167929876904e-05, "log_rms_mae_db": 1.2555932781568533, "rms_envelope_nrmse": 0.11822938488510293, "envelope_correlation": 0.9877521205508717, "rms_gain_error_db": -0.330561653067318, "reference_mean_power": 0.0001744638441038751, "output_mean_power": 0.00016167738347281762}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 1.0543054012178789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.162385199300134e-05, "output_power": 6.568520823913676e-05, "reference_active": true, "delta_db": -0.37590092962545424}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.778561151416546e-06, "output_power": 9.320145776245786e-06, "reference_active": true, "delta_db": -0.20852250817598658}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4459128360402158e-06, "output_power": 1.788483934337093e-06, "reference_active": true, "delta_db": 0.9234293013071655}], "active_band_count": 3, "active_band_mae_db": 0.5026175797028688}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 3.853974797463785e-05, "log_rms_mae_db": 0.762758587279392, "rms_envelope_nrmse": 0.06706495681287376, "envelope_correlation": 0.9967089223732511, "rms_gain_error_db": -0.3305697912634004, "reference_mean_power": 0.00017551968398739576, "output_mean_power": 0.00016265553601932083}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 9.383868144458744e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.640153736830121e-05, "output_power": 6.0835036695010416e-05, "reference_active": true, "delta_db": -0.3802436018085343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.23065423970599e-06, "output_power": 4.322795394055027e-06, "reference_active": true, "delta_db": 0.09357147023923443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2239857829027987e-07, "output_power": 1.1417081093749479e-07, "reference_active": true, "delta_db": -0.30221287538616853}], "active_band_count": 3, "active_band_mae_db": 0.2586759824779791}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9295573719562925, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ebe4034b5486d9e5fb7a9358b3d14cf5226b9abb13c52ee1e0f15d8390e7e089", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.991466391513568, "standard_deviation": 0.0006532640327759406, "minimum": 0.9901150297310337, "maximum": 0.9924149910497978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1754100712523714, "standard_deviation": 0.010947234315751373, "minimum": 1.1564891749206798, "maximum": 1.1902206023971502, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3057690528406358, "standard_deviation": 0.17427257186624326, "minimum": 0.06053963638908725, "maximum": 0.614348080624973, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09922603080810716, "standard_deviation": 0.0035792872936902765, "minimum": 0.09314709333182673, "maximum": 0.10506406324441696, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971357319772878, "standard_deviation": 0.00043369766441833576, "minimum": 0.9965578026520326, "maximum": 0.9978196761806608, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7264407953643761, "standard_deviation": 0.02130964254767278, "minimum": 0.6926448597414343, "maximum": 0.7535615922876261, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.2995806045886825, "standard_deviation": 0.1356832456519474, "minimum": 0.13587248406921074, "maximum": 0.5555966579825218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05861707271237413, "standard_deviation": 0.004648285033994025, "minimum": 0.05105884454959241, "maximum": 0.06593440650300302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9849146758176215, "standard_deviation": 0.0018638480378649442, "minimum": 0.9817300294027651, "maximum": 0.9877521205508717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.30283267222845, "standard_deviation": 0.031530905533607476, "minimum": 1.2555932781568533, "maximum": 1.3417762067340966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5177871778298802, "standard_deviation": 0.14167816401680108, "minimum": 0.2705559631313215, "maximum": 0.734427622997072, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13163572223167813, "standard_deviation": 0.008324181733082052, "minimum": 0.11822938488510293, "maximum": 0.14433608181616664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945979959482315, "standard_deviation": 0.0011880486132457833, "minimum": 0.9928435258050694, "maximum": 0.9967089223732511, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8210311818367659, "standard_deviation": 0.04369683395757955, "minimum": 0.762758587279392, "maximum": 0.8768949305135328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.49500715894028396, "standard_deviation": 0.1580204349945469, "minimum": 0.2493099974801507, "maximum": 0.6645169601561619, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0836200938905153, "standard_deviation": 0.008807535637950989, "minimum": 0.06706495681287376, "maximum": 0.09462655559192408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963199127257629, "standard_deviation": 0.00038062563628113086, "minimum": 0.9956035955602743, "maximum": 0.9967132276822321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2743386742416578, "standard_deviation": 0.04173725522359864, "minimum": 1.2296013283175415, "maximum": 1.3646902853608351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2691178233909937, "standard_deviation": 0.1402310658640664, "minimum": 0.11980576790560277, "maximum": 0.5639468840456715, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0638370352820907, "standard_deviation": 0.0032154233992699945, "minimum": 0.06024922678482586, "maximum": 0.06956000776750575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998801320882756, "standard_deviation": 0.00023838110325437317, "minimum": 0.9983600053047156, "maximum": 0.9991581950058271, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8229701297537746, "standard_deviation": 0.034816024851734524, "minimum": 0.7795974422731944, "maximum": 0.8845732607177428, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3020170093953524, "standard_deviation": 0.11482475186370601, "minimum": 0.1331829088391708, "maximum": 0.504341382170633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03556964808449106, "standard_deviation": 0.0033338452926115507, "minimum": 0.02968586566583459, "maximum": 0.04101827602253392, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18756758631278664, "standard_deviation": 0.023456487025379936, "minimum": 0.14409654514285708, "maximum": 0.22679160985766222, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.488736425808812, "standard_deviation": 0.009992618488282862, "minimum": 5.474990199746299, "maximum": 5.504423392709718, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36696328136385475, "standard_deviation": 0.017561815170200116, "minimum": 0.3457581037128943, "maximum": 0.40014143149112236, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07639714496115584, "standard_deviation": 0.022324958402636104, "minimum": -0.11108535726786951, "maximum": -0.04387470772107756, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973267849464134, "standard_deviation": 0.00023779318614120453, "minimum": 0.9968773269120458, "maximum": 0.9976972041257812, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.667476237351015, "standard_deviation": 0.018874642196974697, "minimum": 0.6376791352840688, "maximum": 0.703186212131072, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2899943335545086, "standard_deviation": 0.11673176996718465, "minimum": 0.10812607007935043, "maximum": 0.49454883133266997, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04858220367249948, "standard_deviation": 0.0022452429374706898, "minimum": 0.044699858817777646, "maximum": 0.05215002672523808, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992016400441321, "standard_deviation": 0.00016705751399866787, "minimum": 0.9988691057531381, "maximum": 0.9994506025683784, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.34149135612995424, "standard_deviation": 0.023892639752749127, "minimum": 0.3119818453909679, "maximum": 0.3787757962751256, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3686281745158066, "standard_deviation": 0.1421299553544997, "minimum": 0.16028691997951194, "maximum": 0.5742640725324671, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02661921426294357, "standard_deviation": 0.0024922216791723427, "minimum": 0.022464952278019958, "maximum": 0.03069512527463229, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0019", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.2213523387908936, "predicted_mos": 2.2213523387908936, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "c65ca004cbfc4b5da59e51fa89e59b05a4e2bc8f5e68c205e0515f9ee3207760", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7344855000264943, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6679376900724818, "全帯域_10ms_rms_envelope_nrmse": 0.05028452686201155, "全帯域_10ms_envelope_correlation": 0.9971315034416927, "全帯域_10ms_modulation_mae_db": 0.21339576466240806, "全帯域_40ms_log_rms_mae_db": 0.3310233902172323, "全帯域_40ms_rms_envelope_nrmse": 0.02711694720576546, "全帯域_40ms_envelope_correlation": 0.9991733634697845, "全帯域_40ms_modulation_mae_db": 0.2835286451996763, "300–1000 Hz_10ms_log_rms_mae_db": 1.296397875175759, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06601569584954717, "300–1000 Hz_10ms_envelope_correlation": 0.9960765897737436, "300–1000 Hz_10ms_modulation_mae_db": 0.28969924482662646, "300–1000 Hz_40ms_log_rms_mae_db": 0.8131758168263051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03784908238108857, "300–1000 Hz_40ms_envelope_correlation": 0.9986538699240891, "300–1000 Hz_40ms_modulation_mae_db": 0.354785389929592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1522965358199369, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09619748507639225, "1000–3000 Hz_10ms_envelope_correlation": 0.9918793437138081, "1000–3000 Hz_10ms_modulation_mae_db": 0.2586279581535418, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6947667699573683, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05163746913474219, "1000–3000 Hz_40ms_envelope_correlation": 0.9977273705545683, "1000–3000 Hz_40ms_modulation_mae_db": 0.288488098194495, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2700906563127596, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11728659886857482, "3000–6000 Hz_10ms_envelope_correlation": 0.9880149745030414, "3000–6000 Hz_10ms_modulation_mae_db": 0.4809362272373863, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7631910649329275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06705464496927777, "3000–6000 Hz_40ms_envelope_correlation": 0.9967551245618906, "3000–6000 Hz_40ms_modulation_mae_db": 0.23757506104026713, "mr_stft_db": 5.4382984825877285, "band_mae_db": 0.17298639637062088, "rms_gain_error_db": -0.08047696729258452, "output_peak": 0.36529811672523055}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6760377292752665, "全帯域_10ms_rms_envelope_nrmse": 0.04882221951686871, "全帯域_10ms_envelope_correlation": 0.9973529302005871, "全帯域_10ms_modulation_mae_db": 0.5509858861055225, "全帯域_40ms_log_rms_mae_db": 0.33856152304103393, "全帯域_40ms_rms_envelope_nrmse": 0.025445970053269318, "全帯域_40ms_envelope_correlation": 0.999344887033143, "全帯域_40ms_modulation_mae_db": 0.5565263955225654, "300–1000 Hz_10ms_log_rms_mae_db": 1.2151697920606217, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.060795840265862336, "300–1000 Hz_10ms_envelope_correlation": 0.9966973463090212, "300–1000 Hz_10ms_modulation_mae_db": 0.7208998398252904, "300–1000 Hz_40ms_log_rms_mae_db": 0.7841732751227416, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033595545425470046, "300–1000 Hz_40ms_envelope_correlation": 0.9989729596954982, "300–1000 Hz_40ms_modulation_mae_db": 0.6280347162048497, "1000–3000 Hz_10ms_log_rms_mae_db": 1.168886155138413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10267297021579255, "1000–3000 Hz_10ms_envelope_correlation": 0.9908786558768323, "1000–3000 Hz_10ms_modulation_mae_db": 0.14476248961713933, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6834103080187319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05968668869675342, "1000–3000 Hz_40ms_envelope_correlation": 0.9970980867453807, "1000–3000 Hz_40ms_modulation_mae_db": 0.08828398181992825, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3310417300879587, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12716251009612975, "3000–6000 Hz_10ms_envelope_correlation": 0.9860460698549908, "3000–6000 Hz_10ms_modulation_mae_db": 0.1926047273696757, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8570311561962598, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07965260403465901, "3000–6000 Hz_40ms_envelope_correlation": 0.9952381299601248, "3000–6000 Hz_40ms_modulation_mae_db": 0.41849438432708713, "mr_stft_db": 5.437946954772858, "band_mae_db": 0.1836409551860506, "rms_gain_error_db": -0.09884529464219532, "output_peak": 0.39816069379495994}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6718471561595524, "全帯域_10ms_rms_envelope_nrmse": 0.046242791164893306, "全帯域_10ms_envelope_correlation": 0.9975744891927406, "全帯域_10ms_modulation_mae_db": 0.3137784603901852, "全帯域_40ms_log_rms_mae_db": 0.335770804122475, "全帯域_40ms_rms_envelope_nrmse": 0.022864718081296514, "全帯域_40ms_envelope_correlation": 0.999423816590862, "全帯域_40ms_modulation_mae_db": 0.5199077083712594, "300–1000 Hz_10ms_log_rms_mae_db": 1.2578961422656025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061810836531718344, "300–1000 Hz_10ms_envelope_correlation": 0.99653785532414, "300–1000 Hz_10ms_modulation_mae_db": 0.18951106439111418, "300–1000 Hz_40ms_log_rms_mae_db": 0.7930140299551575, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031566549578767505, "300–1000 Hz_40ms_envelope_correlation": 0.9990417336709077, "300–1000 Hz_40ms_modulation_mae_db": 0.30251327632225095, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1939461294703813, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.100797453774763, "1000–3000 Hz_10ms_envelope_correlation": 0.9911862795352461, "1000–3000 Hz_10ms_modulation_mae_db": 0.29718395259212765, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7206521312830523, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05670920479762205, "1000–3000 Hz_40ms_envelope_correlation": 0.9973837225402323, "1000–3000 Hz_40ms_modulation_mae_db": 0.547285293632556, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2762108733916036, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12319303000600514, "3000–6000 Hz_10ms_envelope_correlation": 0.9871496055863489, "3000–6000 Hz_10ms_modulation_mae_db": 0.3410958098062103, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7627071455489715, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07639002272918115, "3000–6000 Hz_40ms_envelope_correlation": 0.9959822786699442, "3000–6000 Hz_40ms_modulation_mae_db": 0.6924683943518587, "mr_stft_db": 5.451596714654727, "band_mae_db": 0.18863001215111064, "rms_gain_error_db": -0.07283659111720331, "output_peak": 0.3860190029268821}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.660210645942436, "全帯域_10ms_rms_envelope_nrmse": 0.044198276412065564, "全帯域_10ms_envelope_correlation": 0.997748158885642, "全帯域_10ms_modulation_mae_db": 0.3902471875335, "全帯域_40ms_log_rms_mae_db": 0.35431761135862677, "全帯域_40ms_rms_envelope_nrmse": 0.024184273661771272, "全帯域_40ms_envelope_correlation": 0.9992897780741438, "全帯域_40ms_modulation_mae_db": 0.5843449007674697, "300–1000 Hz_10ms_log_rms_mae_db": 1.2295170089952199, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0635225856085863, "300–1000 Hz_10ms_envelope_correlation": 0.9963512889677926, "300–1000 Hz_10ms_modulation_mae_db": 0.28028492414804845, "300–1000 Hz_40ms_log_rms_mae_db": 0.7589488327590054, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03575274245161093, "300–1000 Hz_40ms_envelope_correlation": 0.9987686782576021, "300–1000 Hz_40ms_modulation_mae_db": 0.23162121577800224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1725480774902142, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10437349942371889, "1000–3000 Hz_10ms_envelope_correlation": 0.9902735625286544, "1000–3000 Hz_10ms_modulation_mae_db": 0.2087613517852285, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7243831723314778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05786517700192623, "1000–3000 Hz_40ms_envelope_correlation": 0.9969283426128122, "1000–3000 Hz_40ms_modulation_mae_db": 0.20967962059161824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3493231749795698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14411345505856743, "3000–6000 Hz_10ms_envelope_correlation": 0.9820217969690467, "3000–6000 Hz_10ms_modulation_mae_db": 0.4533985782747605, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8633760026820227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0952698831427578, "3000–6000 Hz_40ms_envelope_correlation": 0.9929821788269746, "3000–6000 Hz_40ms_modulation_mae_db": 0.8450962984949522, "mr_stft_db": 5.429262243630846, "band_mae_db": 0.15461506374888817, "rms_gain_error_db": -0.04324610822836506, "output_peak": 0.3671251099635034}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6575369027221571, "全帯域_10ms_rms_envelope_nrmse": 0.047212103883864896, "全帯域_10ms_envelope_correlation": 0.9974596758945093, "全帯域_10ms_modulation_mae_db": 0.21892034622311687, "全帯域_40ms_log_rms_mae_db": 0.3287609048356072, "全帯域_40ms_rms_envelope_nrmse": 0.025498583845971354, "全帯域_40ms_envelope_correlation": 0.9992516179618134, "全帯域_40ms_modulation_mae_db": 0.3143852213596974, "300–1000 Hz_10ms_log_rms_mae_db": 1.2561447561444035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061415966852278986, "300–1000 Hz_10ms_envelope_correlation": 0.9965943234512906, "300–1000 Hz_10ms_modulation_mae_db": 0.1126688744716431, "300–1000 Hz_40ms_log_rms_mae_db": 0.7794086329656017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034198294250642636, "300–1000 Hz_40ms_envelope_correlation": 0.9988953926305626, "300–1000 Hz_40ms_modulation_mae_db": 0.20458558216670977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.183471301333892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09435022025296277, "1000–3000 Hz_10ms_envelope_correlation": 0.9921836558291462, "1000–3000 Hz_10ms_modulation_mae_db": 0.06757613443578865, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7142902544520804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05329651742510835, "1000–3000 Hz_40ms_envelope_correlation": 0.9975574818764176, "1000–3000 Hz_40ms_modulation_mae_db": 0.18717467842953694, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2850094592207615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12760609178863955, "3000–6000 Hz_10ms_envelope_correlation": 0.9864493848064828, "3000–6000 Hz_10ms_modulation_mae_db": 0.5455830724144376, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7822267972558318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08033963284073241, "3000–6000 Hz_40ms_envelope_correlation": 0.9958001819231951, "3000–6000 Hz_40ms_modulation_mae_db": 0.5032704587211875, "mr_stft_db": 5.437669808891563, "band_mae_db": 0.19923273997453803, "rms_gain_error_db": -0.06832684754925396, "output_peak": 0.34685828188110196}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7014302245183622, "全帯域_10ms_rms_envelope_nrmse": 0.048909453232644, "全帯域_10ms_envelope_correlation": 0.9974250948739732, "全帯域_10ms_modulation_mae_db": 0.27612448350840796, "全帯域_40ms_log_rms_mae_db": 0.3879818810950122, "全帯域_40ms_rms_envelope_nrmse": 0.02910171232746201, "全帯域_40ms_envelope_correlation": 0.9991882658894146, "全帯域_40ms_modulation_mae_db": 0.39607245315141065, "300–1000 Hz_10ms_log_rms_mae_db": 1.3125681484888898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06551327628070829, "300–1000 Hz_10ms_envelope_correlation": 0.9962227753001212, "300–1000 Hz_10ms_modulation_mae_db": 0.29718059621160714, "300–1000 Hz_40ms_log_rms_mae_db": 0.8263353783968241, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03673926501762515, "300–1000 Hz_40ms_envelope_correlation": 0.998851810176782, "300–1000 Hz_40ms_modulation_mae_db": 0.3170277605793947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1917173386650561, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10086788883707956, "1000–3000 Hz_10ms_envelope_correlation": 0.9913611388930926, "1000–3000 Hz_10ms_modulation_mae_db": 0.33602567171205333, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7387747732291978, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06404046675115797, "1000–3000 Hz_40ms_envelope_correlation": 0.9967214011261651, "1000–3000 Hz_40ms_modulation_mae_db": 0.30044353719729305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.32975975279293, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13561217108586798, "3000–6000 Hz_10ms_envelope_correlation": 0.9839920813117792, "3000–6000 Hz_10ms_modulation_mae_db": 0.5883164530352719, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8280960246506562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08563160930586527, "3000–6000 Hz_40ms_envelope_correlation": 0.9943911849909319, "3000–6000 Hz_40ms_modulation_mae_db": 0.5133977933567214, "mr_stft_db": 5.4420237736156425, "band_mae_db": 0.23479315013533006, "rms_gain_error_db": -0.12219601735644543, "output_peak": 0.352211383789912}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6720339730356597, "全帯域_10ms_rms_envelope_nrmse": 0.04862433002575204, "全帯域_10ms_envelope_correlation": 0.9972834530541513, "全帯域_10ms_modulation_mae_db": 0.2539222066590266, "全帯域_40ms_log_rms_mae_db": 0.31439765545417014, "全帯域_40ms_rms_envelope_nrmse": 0.027721307148322116, "全帯域_40ms_envelope_correlation": 0.9990784308604886, "全帯域_40ms_modulation_mae_db": 0.10856604903219547, "300–1000 Hz_10ms_log_rms_mae_db": 1.19175956807215, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.060732433535654065, "300–1000 Hz_10ms_envelope_correlation": 0.9966795564710474, "300–1000 Hz_10ms_modulation_mae_db": 0.12900570124319385, "300–1000 Hz_40ms_log_rms_mae_db": 0.7322923252212578, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033736006512524404, "300–1000 Hz_40ms_envelope_correlation": 0.9989155774876716, "300–1000 Hz_40ms_modulation_mae_db": 0.09840603267899407, "1000–3000 Hz_10ms_log_rms_mae_db": 1.173473771171627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09564382233803649, "1000–3000 Hz_10ms_envelope_correlation": 0.9918915959665103, "1000–3000 Hz_10ms_modulation_mae_db": 0.5663369944326048, "1000–3000 Hz_40ms_log_rms_mae_db": 0.724598820024042, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05860841095421713, "1000–3000 Hz_40ms_envelope_correlation": 0.9968750992303572, "1000–3000 Hz_40ms_modulation_mae_db": 0.26233745169617545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.311522711455033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13448099923499207, "3000–6000 Hz_10ms_envelope_correlation": 0.9850137706234683, "3000–6000 Hz_10ms_modulation_mae_db": 0.42809134436058044, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7930281361259672, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08789811861633369, "3000–6000 Hz_40ms_envelope_correlation": 0.9948573666831623, "3000–6000 Hz_40ms_modulation_mae_db": 0.46232356036906924, "mr_stft_db": 5.43983980280624, "band_mae_db": 0.178653398281457, "rms_gain_error_db": -0.05088827150549494, "output_peak": 0.34692663187631284}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6838783101020903, "全帯域_10ms_rms_envelope_nrmse": 0.051537534682836214, "全帯域_10ms_envelope_correlation": 0.9969555678367448, "全帯域_10ms_modulation_mae_db": 0.15291099518100887, "全帯域_40ms_log_rms_mae_db": 0.38287763921612605, "全帯域_40ms_rms_envelope_nrmse": 0.030588676503913885, "全帯域_40ms_envelope_correlation": 0.998881449744102, "全帯域_40ms_modulation_mae_db": 0.23356435228215147, "300–1000 Hz_10ms_log_rms_mae_db": 1.238284093041843, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06971158203786396, "300–1000 Hz_10ms_envelope_correlation": 0.9955862871733968, "300–1000 Hz_10ms_modulation_mae_db": 0.1318194935786454, "300–1000 Hz_40ms_log_rms_mae_db": 0.8171071651900279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041161033527495776, "300–1000 Hz_40ms_envelope_correlation": 0.9983485130931652, "300–1000 Hz_40ms_modulation_mae_db": 0.2815586012873588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1751628443308795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10115029660545326, "1000–3000 Hz_10ms_envelope_correlation": 0.9913850218169071, "1000–3000 Hz_10ms_modulation_mae_db": 0.5389655965148252, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7365721441795934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06302623031617291, "1000–3000 Hz_40ms_envelope_correlation": 0.996957758694723, "1000–3000 Hz_40ms_modulation_mae_db": 0.39770186916292133, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2843746940017264, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12947642564606013, "3000–6000 Hz_10ms_envelope_correlation": 0.9852575936184746, "3000–6000 Hz_10ms_modulation_mae_db": 0.41122090151469926, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7761816332749959, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08093627110027443, "3000–6000 Hz_40ms_envelope_correlation": 0.9947963745425686, "3000–6000 Hz_40ms_modulation_mae_db": 0.8595741348834015, "mr_stft_db": 5.453143786391273, "band_mae_db": 0.21307317189465022, "rms_gain_error_db": -0.0630778923267322, "output_peak": 0.3626442109895158}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79760, "output_samples": 79760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.982629601268889, "spectral_convergence": 0.2119637982466495, "spectral_convergence_status": "ok", "amplitude_floor": 8.669731759960068e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.256244785175483, "spectral_convergence": 0.2126636893154138, "spectral_convergence_status": "ok", "amplitude_floor": 7.258845275738271e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.076021061318811, "spectral_convergence": 0.2169191657842892, "spectral_convergence_status": "ok", "amplitude_floor": 5.316334054535358e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.354150869521306e-05, "output_power": 5.6811539734304446e-05, "delta_db": 0.2574595530533311, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007507214796955326, "output_power": 0.0007437698904419637, "delta_db": -0.04040248993612789, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006075331721980808, "output_power": 0.0006023615202389799, "delta_db": -0.037127753372199646, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029803976258038344, "output_power": 0.00029222172146010076, "delta_db": -0.08561713937308958, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002296848467085225, "output_power": 0.0002152573841120744, "delta_db": -0.28174285612796884, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011626570661661749, "output_power": 0.00010762044334117519, "delta_db": -0.33556858636100834, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17298639637062088, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08047696729258452, "pitch": null, "output_rms_dbfs": -26.95069691900128, "output_peak": 0.36529811672523055, "reference_rms_dbfs": -26.870219951708698, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.4382984825877285}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.000134521265514355, "log_rms_mae_db": 0.6679376900724818, "rms_envelope_nrmse": 0.05028452686201155, "envelope_correlation": 0.9971315034416927, "rms_gain_error_db": -0.08047579100860548, "reference_mean_power": 0.0020578494395295825, "output_mean_power": 0.0020200681337601226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 5.569199266163003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039409678072813294, "output_power": 0.0003863123468345042, "reference_active": true, "delta_db": -0.08664298497452468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.428667833313107e-05, "output_power": 5.038542507740401e-05, "reference_active": true, "delta_db": -0.32388342224313504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.373716545850661e-06, "output_power": 6.993915004720938e-06, "reference_active": true, "delta_db": -0.22966088676956453}], "active_band_count": 3, "active_band_mae_db": 0.21339576466240806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00012498981593367327, "log_rms_mae_db": 0.3310233902172323, "rms_envelope_nrmse": 0.02711694720576546, "envelope_correlation": 0.9991733634697845, "rms_gain_error_db": -0.08047653086385519, "reference_mean_power": 0.0020703160518770356, "output_mean_power": 0.0020323055177920925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 4.869724313408462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003604621899100367, "output_power": 0.0003534879704095867, "reference_active": true, "delta_db": -0.0848507797923176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.585585870556879e-05, "output_power": 2.354657315914743e-05, "reference_active": true, "delta_db": -0.40631254618576856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932164327326652e-07, "output_power": 3.619840193468689e-07, "reference_active": true, "delta_db": -0.35942260962094263}], "active_band_count": 3, "active_band_mae_db": 0.2835286451996763}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.00011089795767449635, "log_rms_mae_db": 1.296397875175759, "rms_envelope_nrmse": 0.06601569584954717, "envelope_correlation": 0.9960765897737436, "rms_gain_error_db": -0.056604704507308326, "reference_mean_power": 0.001090370195617981, "output_mean_power": 0.0010762508351149064}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 3.4652335334077796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023090422993351253, "output_power": 0.000227575919442213, "reference_active": true, "delta_db": -0.06305582804415688}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.174073740387618e-05, "output_power": 3.66622681497903e-05, "reference_active": true, "delta_db": -0.5634078762610135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.679949697911218e-06, "output_power": 4.94885339144067e-06, "reference_active": true, "delta_db": 0.24263403017470903}], "active_band_count": 3, "active_band_mae_db": 0.28969924482662646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00010160775009558359, "log_rms_mae_db": 0.8131758168263051, "rms_envelope_nrmse": 0.03784908238108857, "envelope_correlation": 0.9986538699240891, "rms_gain_error_db": -0.05660456329306038, "reference_mean_power": 0.0010969783403958974, "output_mean_power": 0.0010827734452827959}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 3.031226473561925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002111816957536322, "output_power": 0.00020875706809247275, "reference_active": true, "delta_db": -0.05015084130793577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.975380679045621e-05, "output_power": 1.6920700570662253e-05, "reference_active": true, "delta_db": -0.6723246142865381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.773024908316542e-07, "output_power": 2.5631008766583915e-07, "reference_active": true, "delta_db": -0.3418807141943022}], "active_band_count": 3, "active_band_mae_db": 0.354785389929592}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 7.327188751847531e-05, "log_rms_mae_db": 1.1522965358199369, "rms_envelope_nrmse": 0.09619748507639225, "envelope_correlation": 0.9918793437138081, "rms_gain_error_db": -0.1675320693951056, "reference_mean_power": 0.0003949163505892704, "output_mean_power": 0.0003799722741349346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 2.1879792460690968e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000156993183862756, "output_power": 0.00015198125346584896, "reference_active": true, "delta_db": -0.1409077512009722}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3083527056032412e-05, "output_power": 1.2234403498686532e-05, "reference_active": true, "delta_db": -0.29142037122703174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5497810931869906e-06, "output_power": 2.759678382889645e-06, "reference_active": true, "delta_db": 0.34355575203262134}], "active_band_count": 3, "active_band_mae_db": 0.2586279581535418}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 7.131903461227575e-05, "log_rms_mae_db": 0.6947667699573683, "rms_envelope_nrmse": 0.05163746913474219, "envelope_correlation": 0.9977273705545683, "rms_gain_error_db": -0.16753378352406464, "reference_mean_power": 0.0003973070610776639, "output_mean_power": 0.00038227236658405875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 1.9730997112139496e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014395632169995676, "output_power": 0.00013955243245869747, "reference_active": true, "delta_db": -0.13493330417374602}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.343232783051936e-06, "output_power": 5.7177556416561185e-06, "reference_active": true, "delta_db": -0.45085058286832896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5197840492846542e-07, "output_power": 1.424996643394353e-07, "reference_active": true, "delta_db": -0.27968040754140994}], "active_band_count": 3, "active_band_mae_db": 0.288488098194495}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 4.2584167929876904e-05, "log_rms_mae_db": 1.2700906563127596, "rms_envelope_nrmse": 0.11728659886857482, "envelope_correlation": 0.9880149745030414, "rms_gain_error_db": -0.33812658189532974, "reference_mean_power": 0.0001744638441038751, "output_mean_power": 0.00016139600451565484}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 1.0543054012178789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.162385199300134e-05, "output_power": 6.593865568285678e-05, "reference_active": true, "delta_db": -0.35917585272560276}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.778561151416546e-06, "output_power": 9.247772365995469e-06, "reference_active": true, "delta_db": -0.24237824972276711}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4459128360402158e-06, "output_power": 1.7549614097307476e-06, "reference_active": true, "delta_db": 0.8412545792637891}], "active_band_count": 3, "active_band_mae_db": 0.4809362272373863}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 3.853974797463785e-05, "log_rms_mae_db": 0.7631910649329275, "rms_envelope_nrmse": 0.06705464496927777, "envelope_correlation": 0.9967551245618906, "rms_gain_error_db": -0.3381360576127443, "reference_mean_power": 0.00017551968398739576, "output_mean_power": 0.00016237240470509296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 9.383868144458744e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.640153736830121e-05, "output_power": 6.113446968163831e-05, "reference_active": true, "delta_db": -0.35891985312604535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.23065423970599e-06, "output_power": 4.254619133637132e-06, "reference_active": true, "delta_db": 0.02453155816906781}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2239857829027987e-07, "output_power": 1.1346162418980961e-07, "reference_active": true, "delta_db": -0.3292737718256882}], "active_band_count": 3, "active_band_mae_db": 0.23757506104026713}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9326692700661066, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f046658668ad93d390ae63c9205b2f0a9563d21b5aeae7f41fe00953e584195d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9913799067700246, "standard_deviation": 0.0006176447876468425, "minimum": 0.9902735625286544, "maximum": 0.9921836558291462, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.17643776917755, "standard_deviation": 0.013388934343866324, "minimum": 1.1522965358199369, "maximum": 1.1939461294703813, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.30228001865541365, "standard_deviation": 0.17625461164143788, "minimum": 0.06757613443578865, "maximum": 0.5663369944326048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09950670456552485, "standard_deviation": 0.003633048788398785, "minimum": 0.09435022025296277, "maximum": 0.10437349942371889, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997156157922582, "standard_deviation": 0.000358921446474836, "minimum": 0.9967214011261651, "maximum": 0.9977273705545683, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7171810466844429, "standard_deviation": 0.019318626698294664, "minimum": 0.6834103080187319, "maximum": 0.7387747732291978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.28517431634056556, "standard_deviation": 0.13948858076572374, "minimum": 0.08828398181992825, "maximum": 0.547285293632556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05810877063471253, "standard_deviation": 0.00429550497352724, "minimum": 0.05163746913474219, "maximum": 0.06404046675115797, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.985493159659204, "standard_deviation": 0.0018865062103528658, "minimum": 0.9820217969690467, "maximum": 0.9880149745030414, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3046666315302928, "standard_deviation": 0.029688667026711142, "minimum": 1.2700906563127596, "maximum": 1.3493231749795698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.43015588925162773, "standard_deviation": 0.12316197942821315, "minimum": 0.1926047273696757, "maximum": 0.5883164530352719, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1298664102231046, "standard_deviation": 0.00821709321835617, "minimum": 0.11728659886857482, "maximum": 0.14411345505856743, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995100352519849, "standard_deviation": 0.0011442921184818874, "minimum": 0.9929821788269746, "maximum": 0.9967551245618906, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.803229745083454, "standard_deviation": 0.04079998245230565, "minimum": 0.7627071455489715, "maximum": 0.8633760026820227, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.566525010693068, "standard_deviation": 0.21618253034987533, "minimum": 0.23757506104026713, "maximum": 0.8595741348834015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08164659834238519, "standard_deviation": 0.008343527296222087, "minimum": 0.06705464496927777, "maximum": 0.0952698831427578, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963432528463192, "standard_deviation": 0.0003776004169248764, "minimum": 0.9955862871733968, "maximum": 0.9966973463090212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2497171730305614, "standard_deviation": 0.040229537621620605, "minimum": 1.19175956807215, "maximum": 1.3125681484888898, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2688837173370211, "standard_deviation": 0.19811184360382633, "minimum": 0.1126688744716431, "maximum": 0.7208998398252904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06368977712027743, "standard_deviation": 0.003181341070666256, "minimum": 0.060732433535654065, "maximum": 0.06971158203786396, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9988060668670348, "standard_deviation": 0.0002201861065516093, "minimum": 0.9983485130931652, "maximum": 0.9990417336709077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7880569320546151, "standard_deviation": 0.0316488216425729, "minimum": 0.7322923252212578, "maximum": 0.8263353783968241, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.302316571868394, "standard_deviation": 0.15381868976197227, "minimum": 0.09840603267899407, "maximum": 0.6280347162048497, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.035574814893153124, "standard_deviation": 0.0030006060659696298, "minimum": 0.031566549578767505, "maximum": 0.041161033527495776, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1907031109678307, "standard_deviation": 0.024878855731791584, "minimum": 0.15461506374888817, "maximum": 0.23479315013533006, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.44122269591886, "standard_deviation": 0.007812478209752078, "minimum": 5.429262243630846, "maximum": 5.453143786391273, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36565542899342734, "standard_deviation": 0.018390468741363686, "minimum": 0.34685828188110196, "maximum": 0.39816069379495994, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07498674875228434, "standard_deviation": 0.02565454831152337, "minimum": -0.12219601735644543, "maximum": -0.04324610822836506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.997366359172505, "standard_deviation": 0.00024848326103026573, "minimum": 0.9969555678367448, "maximum": 0.997748158885642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6738640789785008, "standard_deviation": 0.013940866080879636, "minimum": 0.6575369027221571, "maximum": 0.7014302245183622, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29628566628289704, "standard_deviation": 0.12514022315184334, "minimum": 0.15291099518100887, "maximum": 0.5509858861055225, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04822890447261704, "standard_deviation": 0.0023104928376939614, "minimum": 0.044198276412065564, "maximum": 0.051537534682836214, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999203951202969, "standard_deviation": 0.00016851716935203028, "minimum": 0.998881449744102, "maximum": 0.999423816590862, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3467114261675355, "standard_deviation": 0.02637025669424955, "minimum": 0.31439765545417014, "maximum": 0.3879818810950122, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37461196571080324, "standard_deviation": 0.16955844345299168, "minimum": 0.10856604903219547, "maximum": 0.5843449007674697, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.026565273603471492, "standard_deviation": 0.002563925254968636, "minimum": 0.022864718081296514, "maximum": 0.030588676503913885, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0019", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.111036777496338, "predicted_mos": 2.111036777496338, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "1c3e71928deead8e7492892df7ffc9be83b8d3b6a737e41b3b2fe86324e90818", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.11500640027225, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6340093709881068, "全帯域_10ms_rms_envelope_nrmse": 0.04473266483028321, "全帯域_10ms_envelope_correlation": 0.9976887846977377, "全帯域_10ms_modulation_mae_db": 0.3069915390784097, "全帯域_40ms_log_rms_mae_db": 0.34415487089639224, "全帯域_40ms_rms_envelope_nrmse": 0.024867491474545787, "全帯域_40ms_envelope_correlation": 0.9992460730755071, "全帯域_40ms_modulation_mae_db": 0.3833353570083758, "300–1000 Hz_10ms_log_rms_mae_db": 1.462621290341833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06630947133978565, "300–1000 Hz_10ms_envelope_correlation": 0.9960350545515352, "300–1000 Hz_10ms_modulation_mae_db": 0.793881915510413, "300–1000 Hz_40ms_log_rms_mae_db": 0.9762837952901072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03705617099154183, "300–1000 Hz_40ms_envelope_correlation": 0.9986798705520754, "300–1000 Hz_40ms_modulation_mae_db": 0.7304323174101217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3037838910285835, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10321913432619065, "1000–3000 Hz_10ms_envelope_correlation": 0.9909766053898146, "1000–3000 Hz_10ms_modulation_mae_db": 0.26909415573447637, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8523560450489819, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062067374322827225, "1000–3000 Hz_40ms_envelope_correlation": 0.9970588226425362, "1000–3000 Hz_40ms_modulation_mae_db": 0.08967441494572657, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5914867906872865, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13076832019939075, "3000–6000 Hz_10ms_envelope_correlation": 0.9845399681665793, "3000–6000 Hz_10ms_modulation_mae_db": 0.7892386173216743, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9760088293456874, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07582066161801217, "3000–6000 Hz_40ms_envelope_correlation": 0.9951403959021592, "3000–6000 Hz_40ms_modulation_mae_db": 0.4891369840322448, "mr_stft_db": 5.746800737997006, "band_mae_db": 0.1816554602926842, "rms_gain_error_db": -0.04399813953613358, "output_peak": 0.34952655642734043}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.633539440009279, "全帯域_10ms_rms_envelope_nrmse": 0.05200858076933791, "全帯域_10ms_envelope_correlation": 0.9970520844646431, "全帯域_10ms_modulation_mae_db": 0.49278902650458006, "全帯域_40ms_log_rms_mae_db": 0.32432420329924316, "全帯域_40ms_rms_envelope_nrmse": 0.030132463447864594, "全帯域_40ms_envelope_correlation": 0.9990984248392757, "全帯域_40ms_modulation_mae_db": 0.5671091235925414, "300–1000 Hz_10ms_log_rms_mae_db": 1.3623928710310667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06561478916975347, "300–1000 Hz_10ms_envelope_correlation": 0.9961504280316048, "300–1000 Hz_10ms_modulation_mae_db": 0.5617193301390845, "300–1000 Hz_40ms_log_rms_mae_db": 0.8788451898123011, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03760769680162161, "300–1000 Hz_40ms_envelope_correlation": 0.9987117071958033, "300–1000 Hz_40ms_modulation_mae_db": 0.5495959168476586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.32889786179904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10099099971335548, "1000–3000 Hz_10ms_envelope_correlation": 0.9912591879306769, "1000–3000 Hz_10ms_modulation_mae_db": 0.43804470263357537, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8375551416325963, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05971521395464474, "1000–3000 Hz_40ms_envelope_correlation": 0.9971828751371581, "1000–3000 Hz_40ms_modulation_mae_db": 0.2080671431548371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.537043047389606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13200016384879817, "3000–6000 Hz_10ms_envelope_correlation": 0.9849561882692368, "3000–6000 Hz_10ms_modulation_mae_db": 0.6938417368222193, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9345714114360654, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07805795269022872, "3000–6000 Hz_40ms_envelope_correlation": 0.9957254409284084, "3000–6000 Hz_40ms_modulation_mae_db": 0.7820081691867627, "mr_stft_db": 5.736098183670676, "band_mae_db": 0.2214719325630392, "rms_gain_error_db": -0.12010907192324838, "output_peak": 0.3578567746284411}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6493168645555798, "全帯域_10ms_rms_envelope_nrmse": 0.04983085438390517, "全帯域_10ms_envelope_correlation": 0.9971707716841827, "全帯域_10ms_modulation_mae_db": 0.33570657066880333, "全帯域_40ms_log_rms_mae_db": 0.3037190293511938, "全帯域_40ms_rms_envelope_nrmse": 0.02695883360734572, "全帯域_40ms_envelope_correlation": 0.9991657718497318, "全帯域_40ms_modulation_mae_db": 0.498071952796183, "300–1000 Hz_10ms_log_rms_mae_db": 1.4155547331328093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06412908368794702, "300–1000 Hz_10ms_envelope_correlation": 0.9962762305996946, "300–1000 Hz_10ms_modulation_mae_db": 0.330895199707117, "300–1000 Hz_40ms_log_rms_mae_db": 0.9459472221264952, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03326113494166316, "300–1000 Hz_40ms_envelope_correlation": 0.9989370705208402, "300–1000 Hz_40ms_modulation_mae_db": 0.16332365203758947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3017867571729214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1077052060888592, "1000–3000 Hz_10ms_envelope_correlation": 0.9899888762396033, "1000–3000 Hz_10ms_modulation_mae_db": 0.5032031230971108, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8178867569624756, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0701421942083816, "1000–3000 Hz_40ms_envelope_correlation": 0.9958508450890227, "1000–3000 Hz_40ms_modulation_mae_db": 0.1869083159398346, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6089600480324355, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14492763001515033, "3000–6000 Hz_10ms_envelope_correlation": 0.982679892560905, "3000–6000 Hz_10ms_modulation_mae_db": 0.7540823973925956, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0477196420829162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09920433392990326, "3000–6000 Hz_40ms_envelope_correlation": 0.9932199028544436, "3000–6000 Hz_40ms_modulation_mae_db": 1.1635873757852309, "mr_stft_db": 5.731828988042305, "band_mae_db": 0.24156511694144392, "rms_gain_error_db": -0.07034889510636774, "output_peak": 0.33737610561026965}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6662401782131153, "全帯域_10ms_rms_envelope_nrmse": 0.047836434966275904, "全帯域_10ms_envelope_correlation": 0.9973803903061271, "全帯域_10ms_modulation_mae_db": 0.39816123741149195, "全帯域_40ms_log_rms_mae_db": 0.3434409252217673, "全帯域_40ms_rms_envelope_nrmse": 0.026003307794102745, "全帯域_40ms_envelope_correlation": 0.9992099455858029, "全帯域_40ms_modulation_mae_db": 0.3711242775854651, "300–1000 Hz_10ms_log_rms_mae_db": 1.3894552272360543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07645410267673484, "300–1000 Hz_10ms_envelope_correlation": 0.9946905172286472, "300–1000 Hz_10ms_modulation_mae_db": 0.2289146243200236, "300–1000 Hz_40ms_log_rms_mae_db": 0.9226545538685853, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047984770895400465, "300–1000 Hz_40ms_envelope_correlation": 0.9977787011085905, "300–1000 Hz_40ms_modulation_mae_db": 0.42143279193759, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3575708443122836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09777202586732378, "1000–3000 Hz_10ms_envelope_correlation": 0.9915276820891669, "1000–3000 Hz_10ms_modulation_mae_db": 0.15782176048112506, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8286620369652918, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055522728790762435, "1000–3000 Hz_40ms_envelope_correlation": 0.9972411828171174, "1000–3000 Hz_40ms_modulation_mae_db": 0.24737162177530622, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5572450340468862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15015520872484595, "3000–6000 Hz_10ms_envelope_correlation": 0.9793709629263997, "3000–6000 Hz_10ms_modulation_mae_db": 1.018028494565436, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9032489647405132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08461482326580806, "3000–6000 Hz_40ms_envelope_correlation": 0.9939028296943552, "3000–6000 Hz_40ms_modulation_mae_db": 0.4060735934569361, "mr_stft_db": 5.717148991761285, "band_mae_db": 0.18471306582128677, "rms_gain_error_db": -0.06457379896926158, "output_peak": 0.3598623744907108}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6678750858111635, "全帯域_10ms_rms_envelope_nrmse": 0.04508115156832392, "全帯域_10ms_envelope_correlation": 0.9976425567709343, "全帯域_10ms_modulation_mae_db": 0.08783753843344529, "全帯域_40ms_log_rms_mae_db": 0.32638437088700795, "全帯域_40ms_rms_envelope_nrmse": 0.021548104424612555, "全帯域_40ms_envelope_correlation": 0.9994299644574179, "全帯域_40ms_modulation_mae_db": 0.0989058467511885, "300–1000 Hz_10ms_log_rms_mae_db": 1.4284906992872752, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06673315302895168, "300–1000 Hz_10ms_envelope_correlation": 0.9959736836487127, "300–1000 Hz_10ms_modulation_mae_db": 0.18344562307780235, "300–1000 Hz_40ms_log_rms_mae_db": 0.9825513115549137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03773141416558045, "300–1000 Hz_40ms_envelope_correlation": 0.9986289852334297, "300–1000 Hz_40ms_modulation_mae_db": 0.12624173150246354, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3030727952003105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10065815775579652, "1000–3000 Hz_10ms_envelope_correlation": 0.9912304487244589, "1000–3000 Hz_10ms_modulation_mae_db": 0.5336003518282095, "1000–3000 Hz_40ms_log_rms_mae_db": 0.787178497103164, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059367603194506686, "1000–3000 Hz_40ms_envelope_correlation": 0.9970854833670798, "1000–3000 Hz_40ms_modulation_mae_db": 0.3367655708396476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6059403163004595, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14570587611038352, "3000–6000 Hz_10ms_envelope_correlation": 0.9805014306524852, "3000–6000 Hz_10ms_modulation_mae_db": 0.8895066842027376, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9729872287672239, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08761778358064395, "3000–6000 Hz_40ms_envelope_correlation": 0.9930891446519151, "3000–6000 Hz_40ms_modulation_mae_db": 0.287291320314784, "mr_stft_db": 5.729018968671589, "band_mae_db": 0.19068792909727672, "rms_gain_error_db": -0.0341483595878018, "output_peak": 0.3777175390238105}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6655873182349732, "全帯域_10ms_rms_envelope_nrmse": 0.048980863876390016, "全帯域_10ms_envelope_correlation": 0.9972417330276533, "全帯域_10ms_modulation_mae_db": 0.1108833661396119, "全帯域_40ms_log_rms_mae_db": 0.3409486129519896, "全帯域_40ms_rms_envelope_nrmse": 0.026275081017394612, "全帯域_40ms_envelope_correlation": 0.9991796930894307, "全帯域_40ms_modulation_mae_db": 0.1906339959260693, "300–1000 Hz_10ms_log_rms_mae_db": 1.4288096724047676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06381324305977354, "300–1000 Hz_10ms_envelope_correlation": 0.9963093377039579, "300–1000 Hz_10ms_modulation_mae_db": 0.47685415725610686, "300–1000 Hz_40ms_log_rms_mae_db": 0.9576069579838158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03322938816965809, "300–1000 Hz_40ms_envelope_correlation": 0.9989257203472004, "300–1000 Hz_40ms_modulation_mae_db": 0.585144182838595, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3654115948761465, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11641531073275468, "1000–3000 Hz_10ms_envelope_correlation": 0.9885086254878598, "1000–3000 Hz_10ms_modulation_mae_db": 0.10217975500939382, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8609014395821906, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0766497381877473, "1000–3000 Hz_40ms_envelope_correlation": 0.9953114049413863, "1000–3000 Hz_40ms_modulation_mae_db": 0.2659937889906621, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6239364950615165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13265806474332908, "3000–6000 Hz_10ms_envelope_correlation": 0.9836884897630805, "3000–6000 Hz_10ms_modulation_mae_db": 0.5927392099735054, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9915800999701259, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07298629856122944, "3000–6000 Hz_40ms_envelope_correlation": 0.9950901805610944, "3000–6000 Hz_40ms_modulation_mae_db": 0.17655732594762283, "mr_stft_db": 5.72410428201736, "band_mae_db": 0.17356334562123318, "rms_gain_error_db": -0.059494509388316685, "output_peak": 0.36922183297940364}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6758678878220105, "全帯域_10ms_rms_envelope_nrmse": 0.04592099610292391, "全帯域_10ms_envelope_correlation": 0.9975472213531961, "全帯域_10ms_modulation_mae_db": 0.14614355385238645, "全帯域_40ms_log_rms_mae_db": 0.3272135620778907, "全帯域_40ms_rms_envelope_nrmse": 0.023604132429572735, "全帯域_40ms_envelope_correlation": 0.9992983872935575, "全帯域_40ms_modulation_mae_db": 0.1578766141834101, "300–1000 Hz_10ms_log_rms_mae_db": 1.5080947645411333, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06528610062705952, "300–1000 Hz_10ms_envelope_correlation": 0.9963362869481062, "300–1000 Hz_10ms_modulation_mae_db": 0.35093221644371225, "300–1000 Hz_40ms_log_rms_mae_db": 1.000731710542681, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03970753050911323, "300–1000 Hz_40ms_envelope_correlation": 0.9986506971125072, "300–1000 Hz_40ms_modulation_mae_db": 0.48312282251465805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3554768305145966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10584413927200134, "1000–3000 Hz_10ms_envelope_correlation": 0.9905296395861958, "1000–3000 Hz_10ms_modulation_mae_db": 0.23558333102973247, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8372068208505153, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06118856017642201, "1000–3000 Hz_40ms_envelope_correlation": 0.9972555517594872, "1000–3000 Hz_40ms_modulation_mae_db": 0.2957335517716427, "3000–6000 Hz_10ms_log_rms_mae_db": 1.605876042097683, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14018017592747004, "3000–6000 Hz_10ms_envelope_correlation": 0.9819275145094164, "3000–6000 Hz_10ms_modulation_mae_db": 0.9331353941646606, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9613124814673177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0810441053592781, "3000–6000 Hz_40ms_envelope_correlation": 0.9940753911046148, "3000–6000 Hz_40ms_modulation_mae_db": 0.5853807878846581, "mr_stft_db": 5.72241745620263, "band_mae_db": 0.202437195888961, "rms_gain_error_db": 0.005467691618961447, "output_peak": 0.4109550880288222}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6789775264501505, "全帯域_10ms_rms_envelope_nrmse": 0.05143478341041859, "全帯域_10ms_envelope_correlation": 0.9969242336386444, "全帯域_10ms_modulation_mae_db": 0.22722037904917666, "全帯域_40ms_log_rms_mae_db": 0.3251033884330006, "全帯域_40ms_rms_envelope_nrmse": 0.02506769158502363, "全帯域_40ms_envelope_correlation": 0.9992156340949638, "全帯域_40ms_modulation_mae_db": 0.1898161134767011, "300–1000 Hz_10ms_log_rms_mae_db": 1.5263575441382902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06805576500437344, "300–1000 Hz_10ms_envelope_correlation": 0.9958103640404238, "300–1000 Hz_10ms_modulation_mae_db": 0.18291776095354353, "300–1000 Hz_40ms_log_rms_mae_db": 1.0229266026055441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03865311262254915, "300–1000 Hz_40ms_envelope_correlation": 0.9985599120186122, "300–1000 Hz_40ms_modulation_mae_db": 0.41045473950204964, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2947633234612006, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10592709795517252, "1000–3000 Hz_10ms_envelope_correlation": 0.9898960017308192, "1000–3000 Hz_10ms_modulation_mae_db": 0.5188859944663081, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7677946566992921, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060778015047483445, "1000–3000 Hz_40ms_envelope_correlation": 0.9964566838069345, "1000–3000 Hz_40ms_modulation_mae_db": 0.28858149927977267, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5910028003109997, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14410071129736, "3000–6000 Hz_10ms_envelope_correlation": 0.9813391571271695, "3000–6000 Hz_10ms_modulation_mae_db": 1.0300743119762463, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9890145763893802, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08990379183830012, "3000–6000 Hz_40ms_envelope_correlation": 0.9931770069005297, "3000–6000 Hz_40ms_modulation_mae_db": 0.460148874624632, "mr_stft_db": 5.730120802522181, "band_mae_db": 0.1861520646841778, "rms_gain_error_db": -0.015956071946755645, "output_peak": 0.3781068107564494}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79760, "output_samples": 79760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.273794169632038, "spectral_convergence": 0.21648122678700646, "spectral_convergence_status": "ok", "amplitude_floor": 8.669731759960068e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.619236992141558, "spectral_convergence": 0.22169790564214856, "spectral_convergence_status": "ok", "amplitude_floor": 7.258845275738271e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.347371052217424, "spectral_convergence": 0.22262384947290664, "spectral_convergence_status": "ok", "amplitude_floor": 5.316334054535358e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.354150869521306e-05, "output_power": 5.5821374799185414e-05, "delta_db": 0.1810992373090577, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007507214796955326, "output_power": 0.0007607602066331605, "delta_db": 0.05768945612332546, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006075331721980808, "output_power": 0.0006127355795540744, "delta_db": 0.03703103095967176, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029803976258038344, "output_power": 0.0002848106917617248, "delta_db": -0.19717920127599803, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002296848467085225, "output_power": 0.00021128156762535973, "delta_db": -0.3627073337263528, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011626570661661749, "output_power": 0.00010965513784779785, "delta_db": -0.2542265023616994, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1816554602926842, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04399813953613358, "pitch": null, "output_rms_dbfs": -26.91421809124483, "output_peak": 0.34952655642734043, "reference_rms_dbfs": -26.870219951708698, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.746800737997006}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.000134521265514355, "log_rms_mae_db": 0.6340093709881068, "rms_envelope_nrmse": 0.04473266483028321, "envelope_correlation": 0.9976887846977377, "rms_gain_error_db": -0.043997289473170484, "reference_mean_power": 0.0020578494395295825, "output_mean_power": 0.0020371071257315365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 5.569199266163003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039409678072813294, "output_power": 0.00039071460597763413, "reference_active": true, "delta_db": -0.0374324060338031}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.428667833313107e-05, "output_power": 4.93105643380991e-05, "reference_active": true, "delta_db": -0.41753296196865053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.373716545850661e-06, "output_power": 8.208945337124034e-06, "reference_active": true, "delta_db": 0.4660092492327754}], "active_band_count": 3, "active_band_mae_db": 0.3069915390784097}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00012498981593367327, "log_rms_mae_db": 0.34415487089639224, "rms_envelope_nrmse": 0.024867491474545787, "envelope_correlation": 0.9992460730755071, "rms_gain_error_db": -0.04399799611491838, "reference_mean_power": 0.0020703160518770356, "output_mean_power": 0.0020494477460585663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 4.869724313408462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003604621899100367, "output_power": 0.00035768279501940877, "reference_active": true, "delta_db": -0.03361666282585233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.585585870556879e-05, "output_power": 2.374239793426011e-05, "reference_active": true, "delta_db": -0.37034386113594114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932164327326652e-07, "output_power": 4.669132663395091e-07, "reference_active": true, "delta_db": 0.746045547063334}], "active_band_count": 3, "active_band_mae_db": 0.3833353570083758}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.00011089795767449635, "log_rms_mae_db": 1.462621290341833, "rms_envelope_nrmse": 0.06630947133978565, "envelope_correlation": 0.9960350545515352, "rms_gain_error_db": 0.039400416307465505, "reference_mean_power": 0.001090370195617981, "output_mean_power": 0.0011003073487285374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 3.4652335334077796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023090422993351253, "output_power": 0.00023503069832300902, "reference_active": true, "delta_db": 0.07692702130618664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.174073740387618e-05, "output_power": 3.64542784555128e-05, "reference_active": true, "delta_db": -0.5881161056173776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.679949697911218e-06, "output_power": 6.948666160550252e-06, "reference_active": true, "delta_db": 1.7166026196076747}], "active_band_count": 3, "active_band_mae_db": 0.793881915510413}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00010160775009558359, "log_rms_mae_db": 0.9762837952901072, "rms_envelope_nrmse": 0.03705617099154183, "envelope_correlation": 0.9986798705520754, "rms_gain_error_db": 0.039400237552824305, "reference_mean_power": 0.0010969783403958974, "output_mean_power": 0.0011069756716609006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 3.031226473561925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002111816957536322, "output_power": 0.00021573808050083946, "reference_active": true, "delta_db": 0.09270537471002771}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.975380679045621e-05, "output_power": 1.7576221916862527e-05, "reference_active": true, "delta_db": -0.5072527433622112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.773024908316542e-07, "output_power": 4.000254790757125e-07, "reference_active": true, "delta_db": 1.5913388341581265}], "active_band_count": 3, "active_band_mae_db": 0.7304323174101217}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 7.327188751847531e-05, "log_rms_mae_db": 1.3037838910285835, "rms_envelope_nrmse": 0.10321913432619065, "envelope_correlation": 0.9909766053898146, "rms_gain_error_db": -0.2685648392550127, "reference_mean_power": 0.0003949163505892704, "output_mean_power": 0.00037123475758998745}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 2.1879792460690968e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000156993183862756, "output_power": 0.00015074059409573974, "reference_active": true, "delta_db": -0.1765057458246207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3083527056032412e-05, "output_power": 1.3036524838939238e-05, "reference_active": true, "delta_db": -0.015630003826732738}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5497810931869906e-06, "output_power": 2.93776848916501e-06, "reference_active": true, "delta_db": 0.6151467175520756}], "active_band_count": 3, "active_band_mae_db": 0.26909415573447637}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 7.131903461227575e-05, "log_rms_mae_db": 0.8523560450489819, "rms_envelope_nrmse": 0.062067374322827225, "envelope_correlation": 0.9970588226425362, "rms_gain_error_db": -0.26856660259710824, "reference_mean_power": 0.0003973070610776639, "output_mean_power": 0.0003734819548535505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 1.9730997112139496e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014395632169995676, "output_power": 0.00013913773085488196, "reference_active": true, "delta_db": -0.14785824913509574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.343232783051936e-06, "output_power": 6.272187987233569e-06, "reference_active": true, "delta_db": -0.048915830140260984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5197840492846542e-07, "output_power": 1.4947100876983003e-07, "reference_active": true, "delta_db": -0.07224916556182298}], "active_band_count": 3, "active_band_mae_db": 0.08967441494572657}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 4.2584167929876904e-05, "log_rms_mae_db": 1.5914867906872865, "rms_envelope_nrmse": 0.13076832019939075, "envelope_correlation": 0.9845399681665793, "rms_gain_error_db": -0.3041120259291662, "reference_mean_power": 0.0001744638441038751, "output_mean_power": 0.0001626650439326262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 1.0543054012178789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.162385199300134e-05, "output_power": 6.480141235439314e-05, "reference_active": true, "delta_db": -0.4347320256563193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.778561151416546e-06, "output_power": 8.860572779631144e-06, "reference_active": true, "delta_db": -0.42813158849103916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4459128360402158e-06, "output_power": 2.0446893595826512e-06, "reference_active": true, "delta_db": 1.5048522378176645}], "active_band_count": 3, "active_band_mae_db": 0.7892386173216743}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 3.853974797463785e-05, "log_rms_mae_db": 0.9760088293456874, "rms_envelope_nrmse": 0.07582066161801217, "envelope_correlation": 0.9951403959021592, "rms_gain_error_db": -0.30412703703919475, "reference_mean_power": 0.00017551968398739576, "output_mean_power": 0.0001636489128693586}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 9.383868144458744e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.640153736830121e-05, "output_power": 6.005622236822643e-05, "reference_active": true, "delta_db": -0.43620123621883894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.23065423970599e-06, "output_power": 4.39832797552448e-06, "reference_active": true, "delta_db": 0.16880077771607813}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2239857829027987e-07, "output_power": 1.4928537628820684e-07, "reference_active": true, "delta_db": 0.8624089381618173}], "active_band_count": 3, "active_band_mae_db": 0.4891369840322448}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9231599563742179, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "11c00fbac8f267e44847e9ddab5f7c7840e751d06569ed3a2d564ce5fb2b0fc5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9904896333973243, "standard_deviation": 0.0009989178041276577, "minimum": 0.9885086254878598, "maximum": 0.9915276820891669, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3263454872956353, "standard_deviation": 0.029295395973824922, "minimum": 1.2947633234612006, "maximum": 1.3654115948761465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3448016467849914, "standard_deviation": 0.17370859475570924, "minimum": 0.10217975500939382, "maximum": 0.5336003518282095, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10481650896393177, "standard_deviation": 0.005722998204458927, "minimum": 0.09777202586732378, "maximum": 0.11641531073275468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966803561950903, "standard_deviation": 0.0007384745037740572, "minimum": 0.9953114049413863, "maximum": 0.9972555517594872, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8236926743555635, "standard_deviation": 0.031841477764376964, "minimum": 0.7677946566992921, "maximum": 0.8609014395821906, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.2398869883371787, "standard_deviation": 0.07745606870111713, "minimum": 0.08967441494572657, "maximum": 0.3367655708396476, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06317892848534692, "standard_deviation": 0.006824872580197545, "minimum": 0.055522728790762435, "maximum": 0.0766497381877473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823754504969091, "standard_deviation": 0.0019630714932170235, "minimum": 0.9793709629263997, "maximum": 0.9849561882692368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5901863217408592, "standard_deviation": 0.02902603896983945, "minimum": 1.537043047389606, "maximum": 1.6239364950615165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8375808558023845, "standard_deviation": 0.1564605714353507, "minimum": 0.5927392099735054, "maximum": 1.0300743119762463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14006201885834096, "standard_deviation": 0.007367008052289218, "minimum": 0.13076832019939075, "maximum": 0.15015520872484595, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9941775365746901, "standard_deviation": 0.0010243214851392488, "minimum": 0.9930891446519151, "maximum": 0.9957254409284084, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9720554042749037, "standard_deviation": 0.042536015397934585, "minimum": 0.9032489647405132, "maximum": 1.0477196420829162, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5437730539041089, "standard_deviation": 0.3098908754273579, "minimum": 0.17655732594762283, "maximum": 1.1635873757852309, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08365621885542547, "standard_deviation": 0.008542296824889889, "minimum": 0.07298629856122944, "maximum": 0.09920433392990326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9959477378440853, "standard_deviation": 0.0005396411899841307, "minimum": 0.9946905172286472, "maximum": 0.9963362869481062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4402221002641538, "standard_deviation": 0.05611054085256583, "minimum": 1.3623928710310667, "maximum": 1.5263575441382902, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38869510342597535, "standard_deviation": 0.2126882642855925, "minimum": 0.18291776095354353, "maximum": 0.793881915510413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06704946357429739, "standard_deviation": 0.004040456153764836, "minimum": 0.06381324305977354, "maximum": 0.07645410267673484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986090830111323, "standard_deviation": 0.00036202946423281786, "minimum": 0.9977787011085905, "maximum": 0.9989370705208402, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9609434179730554, "standard_deviation": 0.04562177688144082, "minimum": 0.8788451898123011, "maximum": 1.0229266026055441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.43371851932384076, "standard_deviation": 0.2053514992915599, "minimum": 0.12624173150246354, "maximum": 0.7304323174101217, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.038153902387141, "standard_deviation": 0.00461644283956536, "minimum": 0.03322938816965809, "maximum": 0.047984770895400465, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19778076386376287, "standard_deviation": 0.022999477793233983, "minimum": 0.17356334562123318, "maximum": 0.24156511694144392, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.729692301360629, "standard_deviation": 0.00910588742361558, "minimum": 5.717148991761285, "maximum": 5.746800737997006, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36757788524315593, "standard_deviation": 0.02232865615458909, "minimum": 0.33737610561026965, "maximum": 0.4109550880288222, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0503951443548655, "standard_deviation": 0.03808562232286059, "minimum": -0.12010907192324838, "maximum": 0.005467691618961447, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973309719928899, "standard_deviation": 0.0002806021675987641, "minimum": 0.9969242336386444, "maximum": 0.9976887846977377, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6589267090105473, "standard_deviation": 0.017827342906804208, "minimum": 0.633539440009279, "maximum": 0.6789775264501505, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26321665139223815, "standard_deviation": 0.14493823844635736, "minimum": 0.08783753843344529, "maximum": 0.49278902650458006, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04822829123848233, "standard_deviation": 0.0028108923205986346, "minimum": 0.04473266483028321, "maximum": 0.05200858076933791, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992304867857109, "standard_deviation": 9.963911770415035e-05, "minimum": 0.9990984248392757, "maximum": 0.9994299644574179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3294111203898107, "standard_deviation": 0.013447435413315608, "minimum": 0.3037190293511938, "maximum": 0.34415487089639224, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3071091601649918, "standard_deviation": 0.17192948416551088, "minimum": 0.0989058467511885, "maximum": 0.5671091235925414, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0255571382225578, "standard_deviation": 0.0025132396361450256, "minimum": 0.021548104424612555, "maximum": 0.030132463447864594, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0019", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.0683937072753906, "predicted_mos": 2.0683937072753906, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "c1a937c98cf3bca078e48567694b9915a7083acb3be4fa545cfc41dffad7c16a", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.28620360000059, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6486230686859211, "全帯域_10ms_rms_envelope_nrmse": 0.05029567185203369, "全帯域_10ms_envelope_correlation": 0.9970555591184497, "全帯域_10ms_modulation_mae_db": 0.22883660547142018, "全帯域_40ms_log_rms_mae_db": 0.3338456568740333, "全帯域_40ms_rms_envelope_nrmse": 0.026752769151968076, "全帯域_40ms_envelope_correlation": 0.9990969594024955, "全帯域_40ms_modulation_mae_db": 0.30308207863975084, "300–1000 Hz_10ms_log_rms_mae_db": 1.1284784135073815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06513867638931646, "300–1000 Hz_10ms_envelope_correlation": 0.9962838120666921, "300–1000 Hz_10ms_modulation_mae_db": 0.275189055885147, "300–1000 Hz_40ms_log_rms_mae_db": 0.7012340304844653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040302955876650254, "300–1000 Hz_40ms_envelope_correlation": 0.9985365710226719, "300–1000 Hz_40ms_modulation_mae_db": 0.1945088233932312, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0703997760212112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09095148719753746, "1000–3000 Hz_10ms_envelope_correlation": 0.9927868504133867, "1000–3000 Hz_10ms_modulation_mae_db": 0.2541947879782999, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6516439090453032, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0548060724069857, "1000–3000 Hz_40ms_envelope_correlation": 0.9974162835327156, "1000–3000 Hz_40ms_modulation_mae_db": 0.2498627366780716, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2407241785446588, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12899406912007014, "3000–6000 Hz_10ms_envelope_correlation": 0.9857299546263509, "3000–6000 Hz_10ms_modulation_mae_db": 0.4862873152362263, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8122541183177021, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08494343659522598, "3000–6000 Hz_40ms_envelope_correlation": 0.9944710771541393, "3000–6000 Hz_40ms_modulation_mae_db": 0.7082772487025714, "mr_stft_db": 5.423623861051404, "band_mae_db": 0.19768353710928685, "rms_gain_error_db": -0.002671982551508284, "output_peak": 0.36573620016378294}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6876017719093215, "全帯域_10ms_rms_envelope_nrmse": 0.04477648547814474, "全帯域_10ms_envelope_correlation": 0.9976725297648308, "全帯域_10ms_modulation_mae_db": 0.10207138879866591, "全帯域_40ms_log_rms_mae_db": 0.3602240603079694, "全帯域_40ms_rms_envelope_nrmse": 0.025486399062305366, "全帯域_40ms_envelope_correlation": 0.9991909554282516, "全帯域_40ms_modulation_mae_db": 0.16368013698262077, "300–1000 Hz_10ms_log_rms_mae_db": 1.1051213380422267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.058616420283856135, "300–1000 Hz_10ms_envelope_correlation": 0.9968850212963106, "300–1000 Hz_10ms_modulation_mae_db": 0.08183714942034113, "300–1000 Hz_40ms_log_rms_mae_db": 0.7424464484579081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035011106773901325, "300–1000 Hz_40ms_envelope_correlation": 0.9988055095355265, "300–1000 Hz_40ms_modulation_mae_db": 0.18119793383012928, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1739420256566127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09486690805329835, "1000–3000 Hz_10ms_envelope_correlation": 0.9920369670607936, "1000–3000 Hz_10ms_modulation_mae_db": 0.28988502384596104, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7217180350148591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05156250957102445, "1000–3000 Hz_40ms_envelope_correlation": 0.9975951901296244, "1000–3000 Hz_40ms_modulation_mae_db": 0.24105985754291526, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2529816780639147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13372788023775498, "3000–6000 Hz_10ms_envelope_correlation": 0.9836235099393941, "3000–6000 Hz_10ms_modulation_mae_db": 0.4963205112746249, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8267272574936572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08619263932037553, "3000–6000 Hz_40ms_envelope_correlation": 0.9930836558475414, "3000–6000 Hz_40ms_modulation_mae_db": 0.2541067041581557, "mr_stft_db": 5.420405054960991, "band_mae_db": 0.15222605727643243, "rms_gain_error_db": -0.02015811803779479, "output_peak": 0.3750272256956097}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.72494803718897, "全帯域_10ms_rms_envelope_nrmse": 0.04823795554611927, "全帯域_10ms_envelope_correlation": 0.9973287175047624, "全帯域_10ms_modulation_mae_db": 0.10574344036941641, "全帯域_40ms_log_rms_mae_db": 0.3385834942303408, "全帯域_40ms_rms_envelope_nrmse": 0.02420957210888115, "全帯域_40ms_envelope_correlation": 0.9993140098610175, "全帯域_40ms_modulation_mae_db": 0.16076425130773894, "300–1000 Hz_10ms_log_rms_mae_db": 1.2064026684391436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06472704398318928, "300–1000 Hz_10ms_envelope_correlation": 0.9961941263827793, "300–1000 Hz_10ms_modulation_mae_db": 0.3214707016963659, "300–1000 Hz_40ms_log_rms_mae_db": 0.7771222678845058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03355598685038145, "300–1000 Hz_40ms_envelope_correlation": 0.9989068468379516, "300–1000 Hz_40ms_modulation_mae_db": 0.3068131070075477, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1176037788363207, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09551410362262154, "1000–3000 Hz_10ms_envelope_correlation": 0.9920592216687253, "1000–3000 Hz_10ms_modulation_mae_db": 0.21607489283263293, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7105233949020169, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05442562316334916, "1000–3000 Hz_40ms_envelope_correlation": 0.9975220674248398, "1000–3000 Hz_40ms_modulation_mae_db": 0.18318859485061803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2316762661682816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.133289335261523, "3000–6000 Hz_10ms_envelope_correlation": 0.9845086956469347, "3000–6000 Hz_10ms_modulation_mae_db": 0.5473489669577735, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7915458524002494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08504818923488229, "3000–6000 Hz_40ms_envelope_correlation": 0.9943424660139908, "3000–6000 Hz_40ms_modulation_mae_db": 0.2975171058527338, "mr_stft_db": 5.4097590761190455, "band_mae_db": 0.21152760261170844, "rms_gain_error_db": -0.05990182787095709, "output_peak": 0.37936823603794984}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6905164472204376, "全帯域_10ms_rms_envelope_nrmse": 0.04941151542042923, "全帯域_10ms_envelope_correlation": 0.9973068957209087, "全帯域_10ms_modulation_mae_db": 0.30451399084710906, "全帯域_40ms_log_rms_mae_db": 0.3463358557238071, "全帯域_40ms_rms_envelope_nrmse": 0.028198481065950017, "全帯域_40ms_envelope_correlation": 0.9991769450198288, "全帯域_40ms_modulation_mae_db": 0.20986781786442968, "300–1000 Hz_10ms_log_rms_mae_db": 1.1716462775350078, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06528230124004396, "300–1000 Hz_10ms_envelope_correlation": 0.9962395476246322, "300–1000 Hz_10ms_modulation_mae_db": 0.3085866574082667, "300–1000 Hz_40ms_log_rms_mae_db": 0.7795857158344344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04018321691278947, "300–1000 Hz_40ms_envelope_correlation": 0.9985718074359854, "300–1000 Hz_40ms_modulation_mae_db": 0.37584184703408274, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1260171918758444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09771931959758957, "1000–3000 Hz_10ms_envelope_correlation": 0.991670779369391, "1000–3000 Hz_10ms_modulation_mae_db": 0.5439226933886904, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6691382043241619, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057619098062532, "1000–3000 Hz_40ms_envelope_correlation": 0.9971881916111436, "1000–3000 Hz_40ms_modulation_mae_db": 0.6252885380700735, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2360829052195415, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13143620335665912, "3000–6000 Hz_10ms_envelope_correlation": 0.984955541835132, "3000–6000 Hz_10ms_modulation_mae_db": 0.26063943425700326, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7869749933063835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08524431368299039, "3000–6000 Hz_40ms_envelope_correlation": 0.9942350848182208, "3000–6000 Hz_40ms_modulation_mae_db": 0.4113373554230086, "mr_stft_db": 5.404377597677633, "band_mae_db": 0.1706552504842668, "rms_gain_error_db": -0.10205718333325854, "output_peak": 0.35052703952791536}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6796699938682987, "全帯域_10ms_rms_envelope_nrmse": 0.04933658648960398, "全帯域_10ms_envelope_correlation": 0.9971842586500039, "全帯域_10ms_modulation_mae_db": 0.21620748933356412, "全帯域_40ms_log_rms_mae_db": 0.3325489894957899, "全帯域_40ms_rms_envelope_nrmse": 0.02554597521475369, "全帯域_40ms_envelope_correlation": 0.9992032203913996, "全帯域_40ms_modulation_mae_db": 0.1484698622014893, "300–1000 Hz_10ms_log_rms_mae_db": 1.1942907830045468, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06574653259961871, "300–1000 Hz_10ms_envelope_correlation": 0.9960765861630912, "300–1000 Hz_10ms_modulation_mae_db": 0.4657294454866656, "300–1000 Hz_40ms_log_rms_mae_db": 0.7653933849813808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03490884011861199, "300–1000 Hz_40ms_envelope_correlation": 0.9988142241023802, "300–1000 Hz_40ms_modulation_mae_db": 0.436996388311544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1328927537792328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09798472575374174, "1000–3000 Hz_10ms_envelope_correlation": 0.9916162837295451, "1000–3000 Hz_10ms_modulation_mae_db": 0.5887325560139489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7130024521978816, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05984235582911964, "1000–3000 Hz_40ms_envelope_correlation": 0.9969008812262756, "1000–3000 Hz_40ms_modulation_mae_db": 0.41963319148832584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.218499148701631, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12298779129608689, "3000–6000 Hz_10ms_envelope_correlation": 0.9871586327872606, "3000–6000 Hz_10ms_modulation_mae_db": 0.39372591426899906, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7609687845155501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08117159942526167, "3000–6000 Hz_40ms_envelope_correlation": 0.9951159043204145, "3000–6000 Hz_40ms_modulation_mae_db": 0.6140652854333061, "mr_stft_db": 5.411427246598355, "band_mae_db": 0.19881837924780169, "rms_gain_error_db": -0.041719268362381476, "output_peak": 0.3448122181657005}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6548452636076028, "全帯域_10ms_rms_envelope_nrmse": 0.048892301175317295, "全帯域_10ms_envelope_correlation": 0.9972917476349911, "全帯域_10ms_modulation_mae_db": 0.3191547490146697, "全帯域_40ms_log_rms_mae_db": 0.3330176237460941, "全帯域_40ms_rms_envelope_nrmse": 0.02582058504571158, "全帯域_40ms_envelope_correlation": 0.9992581509333645, "全帯域_40ms_modulation_mae_db": 0.27786491534075, "300–1000 Hz_10ms_log_rms_mae_db": 1.1892395822007178, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06672380662907551, "300–1000 Hz_10ms_envelope_correlation": 0.9959562200627246, "300–1000 Hz_10ms_modulation_mae_db": 0.34994683264014464, "300–1000 Hz_40ms_log_rms_mae_db": 0.7484270976524022, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03843129418832956, "300–1000 Hz_40ms_envelope_correlation": 0.9985780719339018, "300–1000 Hz_40ms_modulation_mae_db": 0.215870073428467, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0752408284508084, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08851147746833404, "1000–3000 Hz_10ms_envelope_correlation": 0.9932257064711437, "1000–3000 Hz_10ms_modulation_mae_db": 0.3521451589867697, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6596103251941571, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05121442667875599, "1000–3000 Hz_40ms_envelope_correlation": 0.9978459232785791, "1000–3000 Hz_40ms_modulation_mae_db": 0.47980702065480313, "3000–6000 Hz_10ms_log_rms_mae_db": 1.21055564446002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1356917787188633, "3000–6000 Hz_10ms_envelope_correlation": 0.9837508703874859, "3000–6000 Hz_10ms_modulation_mae_db": 0.8945948985465401, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7869992580445904, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.082041354392988, "3000–6000 Hz_40ms_envelope_correlation": 0.9947192577764187, "3000–6000 Hz_40ms_modulation_mae_db": 0.3361132328896546, "mr_stft_db": 5.420016226106067, "band_mae_db": 0.22808976318828436, "rms_gain_error_db": -0.07848690748560545, "output_peak": 0.35212330683750975}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6768407261593508, "全帯域_10ms_rms_envelope_nrmse": 0.055251421646691286, "全帯域_10ms_envelope_correlation": 0.9965157826136731, "全帯域_10ms_modulation_mae_db": 0.20580703696848965, "全帯域_40ms_log_rms_mae_db": 0.3448973650945085, "全帯域_40ms_rms_envelope_nrmse": 0.03272675429223278, "全帯域_40ms_envelope_correlation": 0.9987416055910359, "全帯域_40ms_modulation_mae_db": 0.25522389183005506, "300–1000 Hz_10ms_log_rms_mae_db": 1.2332543432800198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06899443214019879, "300–1000 Hz_10ms_envelope_correlation": 0.9956864633293402, "300–1000 Hz_10ms_modulation_mae_db": 0.4469292275110788, "300–1000 Hz_40ms_log_rms_mae_db": 0.8193809239783442, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03829007678610507, "300–1000 Hz_40ms_envelope_correlation": 0.9986050258658838, "300–1000 Hz_40ms_modulation_mae_db": 0.43863472819508126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0781680117982666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09333086899975694, "1000–3000 Hz_10ms_envelope_correlation": 0.9924427461470755, "1000–3000 Hz_10ms_modulation_mae_db": 0.3347293803721059, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7140382849001615, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05656493304409538, "1000–3000 Hz_40ms_envelope_correlation": 0.9972781964712913, "1000–3000 Hz_40ms_modulation_mae_db": 0.11152676905506032, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2257384581075215, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13511218826396562, "3000–6000 Hz_10ms_envelope_correlation": 0.9847549229476596, "3000–6000 Hz_10ms_modulation_mae_db": 0.5186445934526457, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7708268381232987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08917341510741322, "3000–6000 Hz_40ms_envelope_correlation": 0.9945200575972599, "3000–6000 Hz_40ms_modulation_mae_db": 0.4474880505609377, "mr_stft_db": 5.418185460489879, "band_mae_db": 0.3094685580495564, "rms_gain_error_db": -0.07779694729815684, "output_peak": 0.3676731968730821}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6566882309339588, "全帯域_10ms_rms_envelope_nrmse": 0.04555448037200302, "全帯域_10ms_envelope_correlation": 0.9976272262005159, "全帯域_10ms_modulation_mae_db": 0.14104345475252433, "全帯域_40ms_log_rms_mae_db": 0.34415796163435225, "全帯域_40ms_rms_envelope_nrmse": 0.0236424446477903, "全帯域_40ms_envelope_correlation": 0.9993490784300432, "全帯域_40ms_modulation_mae_db": 0.09491577040994294, "300–1000 Hz_10ms_log_rms_mae_db": 1.2397428956099161, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05749912032044551, "300–1000 Hz_10ms_envelope_correlation": 0.9970401287175769, "300–1000 Hz_10ms_modulation_mae_db": 0.0813353824450438, "300–1000 Hz_40ms_log_rms_mae_db": 0.7854041642281767, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030044797558397212, "300–1000 Hz_40ms_envelope_correlation": 0.9991683370785838, "300–1000 Hz_40ms_modulation_mae_db": 0.09193993567678815, "1000–3000 Hz_10ms_log_rms_mae_db": 1.123395782988803, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08895538515574274, "1000–3000 Hz_10ms_envelope_correlation": 0.9931987521494602, "1000–3000 Hz_10ms_modulation_mae_db": 0.17686778032246764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7106350802865572, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052894359503866444, "1000–3000 Hz_40ms_envelope_correlation": 0.9976903295153151, "1000–3000 Hz_40ms_modulation_mae_db": 0.39325849189697437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2087174976380337, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11971638816225338, "3000–6000 Hz_10ms_envelope_correlation": 0.9872730122485224, "3000–6000 Hz_10ms_modulation_mae_db": 0.43212370842122566, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7671024122917828, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06868123310123687, "3000–6000 Hz_40ms_envelope_correlation": 0.9962840996267807, "3000–6000 Hz_40ms_modulation_mae_db": 0.17260095411649426, "mr_stft_db": 5.417568106133255, "band_mae_db": 0.2134628500606389, "rms_gain_error_db": -0.04434296139604297, "output_peak": 0.3548808698634989}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79760, "output_samples": 79760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.888456224776768, "spectral_convergence": 0.21767335087264827, "spectral_convergence_status": "ok", "amplitude_floor": 8.669731759960068e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.250521828249389, "spectral_convergence": 0.2174358180420926, "spectral_convergence_status": "ok", "amplitude_floor": 7.258845275738271e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.131893530128053, "spectral_convergence": 0.21983778412088736, "spectral_convergence_status": "ok", "amplitude_floor": 5.316334054535358e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.354150869521306e-05, "output_power": 5.529749702298531e-05, "delta_db": 0.1401486931979567, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007507214796955326, "output_power": 0.0007673957141858107, "delta_db": 0.09540527277541458, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006075331721980808, "output_power": 0.000623032067773153, "delta_db": 0.10940404558907196, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029803976258038344, "output_power": 0.0002917541182155622, "delta_db": -0.09257213779691356, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002296848467085225, "output_power": 0.0002100247695366841, "delta_db": -0.3886182705295444, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011626570661661749, "output_power": 0.0001070178828822766, "delta_db": -0.35995280276681996, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19768353710928685, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.002671982551508284, "pitch": null, "output_rms_dbfs": -26.872891934260203, "output_peak": 0.36573620016378294, "reference_rms_dbfs": -26.870219951708698, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.423623861051404}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.000134521265514355, "log_rms_mae_db": 0.6486230686859211, "rms_envelope_nrmse": 0.05029567185203369, "envelope_correlation": 0.9970555591184497, "rms_gain_error_db": -0.0026717649717810147, "reference_mean_power": 0.0020578494395295825, "output_mean_power": 0.002056583846844823}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 5.569199266163003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039409678072813294, "output_power": 0.00039077833821842385, "reference_active": true, "delta_db": -0.03672405515843748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.428667833313107e-05, "output_power": 5.175167579305722e-05, "reference_active": true, "delta_db": -0.2076885163785237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.373716545850661e-06, "output_power": 6.660038110473358e-06, "reference_active": true, "delta_db": -0.4420972448772994}], "active_band_count": 3, "active_band_mae_db": 0.22883660547142018}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00012498981593367327, "log_rms_mae_db": 0.3338456568740333, "rms_envelope_nrmse": 0.026752769151968076, "envelope_correlation": 0.9990969594024955, "rms_gain_error_db": -0.0026729171080519536, "reference_mean_power": 0.0020703160518770356, "output_mean_power": 0.002069042243238525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 4.869724313408462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003604621899100367, "output_power": 0.0003574269134434599, "reference_active": true, "delta_db": -0.03672466037875846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.585585870556879e-05, "output_power": 2.4879088976589206e-05, "reference_active": true, "delta_db": -0.1672449251490431}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932164327326652e-07, "output_power": 3.34275074221417e-07, "reference_active": true, "delta_db": -0.705276650391451}], "active_band_count": 3, "active_band_mae_db": 0.30308207863975084}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.00011089795767449635, "log_rms_mae_db": 1.1284784135073815, "rms_envelope_nrmse": 0.06513867638931646, "envelope_correlation": 0.9962838120666921, "rms_gain_error_db": 0.09259920855182784, "reference_mean_power": 0.001090370195617981, "output_mean_power": 0.0011138684241026285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 3.4652335334077796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023090422993351253, "output_power": 0.00023884658086623158, "reference_active": true, "delta_db": 0.14687139827966156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.174073740387618e-05, "output_power": 3.959716718040493e-05, "reference_active": true, "delta_db": -0.22895999978441378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.679949697911218e-06, "output_power": 5.1905665311954306e-06, "reference_active": true, "delta_db": 0.44973576959136574}], "active_band_count": 3, "active_band_mae_db": 0.275189055885147}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00010160775009558359, "log_rms_mae_db": 0.7012340304844653, "rms_envelope_nrmse": 0.040302955876650254, "envelope_correlation": 0.9985365710226719, "rms_gain_error_db": 0.09259922698937328, "reference_mean_power": 0.0010969783403958974, "output_mean_power": 0.0011206189837079928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 3.031226473561925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002111816957536322, "output_power": 0.00021832381410638506, "reference_active": true, "delta_db": 0.1444483696262242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.975380679045621e-05, "output_power": 1.880815674317669e-05, "reference_active": true, "delta_db": -0.21304566222908997}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.773024908316542e-07, "output_power": 2.921171228001738e-07, "reference_active": true, "delta_db": 0.22603243832437944}], "active_band_count": 3, "active_band_mae_db": 0.1945088233932312}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 7.327188751847531e-05, "log_rms_mae_db": 1.0703997760212112, "rms_envelope_nrmse": 0.09095148719753746, "envelope_correlation": 0.9927868504133867, "rms_gain_error_db": -0.17518449875880615, "reference_mean_power": 0.0003949163505892704, "output_mean_power": 0.0003793033384551446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 2.1879792460690968e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000156993183862756, "output_power": 0.00014702211479062242, "reference_active": true, "delta_db": -0.2849813173133632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3083527056032412e-05, "output_power": 1.1985746295271823e-05, "reference_active": true, "delta_db": -0.38059756180603305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5497810931869906e-06, "output_power": 2.4934596688234936e-06, "reference_active": true, "delta_db": -0.09700548481550356}], "active_band_count": 3, "active_band_mae_db": 0.2541947879782999}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 7.131903461227575e-05, "log_rms_mae_db": 0.6516439090453032, "rms_envelope_nrmse": 0.0548060724069857, "envelope_correlation": 0.9974162835327156, "rms_gain_error_db": -0.175191665928402, "reference_mean_power": 0.0003973070610776639, "output_mean_power": 0.0003815989024855385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 1.9730997112139496e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014395632169995676, "output_power": 0.00013503701635251492, "reference_active": true, "delta_db": -0.2777790761073074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.343232783051936e-06, "output_power": 5.748159819642394e-06, "reference_active": true, "delta_db": -0.4278181477874046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5197840492846542e-07, "output_power": 1.5352566270584268e-07, "reference_active": true, "delta_db": 0.04399098613950281}], "active_band_count": 3, "active_band_mae_db": 0.2498627366780716}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 4.2584167929876904e-05, "log_rms_mae_db": 1.2407241785446588, "rms_envelope_nrmse": 0.12899406912007014, "envelope_correlation": 0.9857299546263509, "rms_gain_error_db": -0.4074891742568669, "reference_mean_power": 0.0001744638441038751, "output_mean_power": 0.00015883877169163785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 1.0543054012178789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.162385199300134e-05, "output_power": 6.492780143136333e-05, "reference_active": true, "delta_db": -0.4262697688575831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.778561151416546e-06, "output_power": 7.964924640297327e-06, "reference_active": true, "delta_db": -0.8909328491461168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4459128360402158e-06, "output_power": 1.493853627248731e-06, "reference_active": true, "delta_db": 0.14165932770497897}], "active_band_count": 3, "active_band_mae_db": 0.4862873152362263}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 3.853974797463785e-05, "log_rms_mae_db": 0.8122541183177021, "rms_envelope_nrmse": 0.08494343659522598, "envelope_correlation": 0.9944710771541393, "rms_gain_error_db": -0.4074903546490313, "reference_mean_power": 0.00017551968398739576, "output_mean_power": 0.00015980000657438175}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 9.383868144458744e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.640153736830121e-05, "output_power": 6.023380201535766e-05, "reference_active": true, "delta_db": -0.42337857411233504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.23065423970599e-06, "output_power": 3.5676981021903303e-06, "reference_active": true, "delta_db": -0.7401943567305426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2239857829027987e-07, "output_power": 9.809581772418087e-08, "reference_active": true, "delta_db": -0.9612588152648365}], "active_band_count": 3, "active_band_mae_db": 0.7082772487025714}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9330086560937276, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4eb85f37d97f7ca007d15301010854e6b9461f1dc876d45b3173ea6c1685c176", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9923796633761901, "standard_deviation": 0.0006394042683552051, "minimum": 0.9916162837295451, "maximum": 0.9932257064711437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1122075186758873, "standard_deviation": 0.0355711436418551, "minimum": 1.0703997760212112, "maximum": 1.1739420256566127, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.34456903421760954, "standard_deviation": 0.14895072222388725, "minimum": 0.17686778032246764, "maximum": 0.5887325560139489, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0934792844810778, "standard_deviation": 0.0037013403913998197, "minimum": 0.08851147746833404, "maximum": 0.09798472575374174, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974296328987231, "standard_deviation": 0.000301945399079152, "minimum": 0.9969008812262756, "maximum": 0.9978459232785791, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6937887107331373, "standard_deviation": 0.028473305555487122, "minimum": 0.6516439090453032, "maximum": 0.7217180350148591, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.33795315002960524, "standard_deviation": 0.17104037135709743, "minimum": 0.11152676905506032, "maximum": 0.6252885380700735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0548661722824661, "standard_deviation": 0.0030133504273532627, "minimum": 0.05121442667875599, "maximum": 0.05984235582911964, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9852193925523425, "standard_deviation": 0.0014006385352845383, "minimum": 0.9836235099393941, "maximum": 0.9872730122485224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2281219721129504, "standard_deviation": 0.015288741961477465, "minimum": 1.2087174976380337, "maximum": 1.2529816780639147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5037106678018798, "standard_deviation": 0.18188731052165824, "minimum": 0.26063943425700326, "maximum": 0.8945948985465401, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13011945430214705, "standard_deviation": 0.005869190709259625, "minimum": 0.11971638816225338, "maximum": 0.1356917787188633, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945964503943457, "standard_deviation": 0.0008974781144054861, "minimum": 0.9930836558475414, "maximum": 0.9962840996267807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7879249393116518, "standard_deviation": 0.022567220600286767, "minimum": 0.7609687845155501, "maximum": 0.8267272574936572, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.40518824214210775, "standard_deviation": 0.18150425365423667, "minimum": 0.17260095411649426, "maximum": 0.7082772487025714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08281202260754675, "standard_deviation": 0.006214262631004906, "minimum": 0.06868123310123687, "maximum": 0.08917341510741322, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9962952382053933, "standard_deviation": 0.0004550526553443245, "minimum": 0.9956864633293402, "maximum": 0.9970401287175769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.18352203770237, "standard_deviation": 0.047243700240028046, "minimum": 1.1051213380422267, "maximum": 1.2397428956099161, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.29137805656163174, "standard_deviation": 0.14516688860363278, "minimum": 0.0813353824450438, "maximum": 0.4657294454866656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06409104169821805, "standard_deviation": 0.003965230724003671, "minimum": 0.05749912032044551, "maximum": 0.06899443214019879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9987482992266106, "standard_deviation": 0.00021865531335929674, "minimum": 0.9985365710226719, "maximum": 0.9991683370785838, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7648742541877023, "standard_deviation": 0.035015829479507134, "minimum": 0.7012340304844653, "maximum": 0.8193809239783442, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.28022535460960896, "standard_deviation": 0.12887805079524076, "minimum": 0.09193993567678815, "maximum": 0.43863472819508126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03634103438314579, "standard_deviation": 0.003582619086322242, "minimum": 0.030044797558397212, "maximum": 0.040302955876650254, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21024149975349699, "standard_deviation": 0.0469005841708465, "minimum": 0.15222605727643243, "maximum": 0.3094685580495564, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.415670328642078, "standard_deviation": 0.006491345972498504, "minimum": 5.404377597677633, "maximum": 5.423623861051404, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3612685366456312, "standard_deviation": 0.012470009885507182, "minimum": 0.3448122181657005, "maximum": 0.37936823603794984, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.053391899541963175, "standard_deviation": 0.032800658978176, "minimum": -0.10205718333325854, "maximum": -0.002671982551508284, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9972478396510169, "standard_deviation": 0.00036118346291935503, "minimum": 0.9965157826136731, "maximum": 0.9976725297648308, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6774666924467327, "standard_deviation": 0.024808318138600575, "minimum": 0.6486230686859211, "maximum": 0.72494803718897, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.20292226944448244, "standard_deviation": 0.08292066051575885, "minimum": 0.10207138879866591, "maximum": 0.3191547490146697, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04896955224754281, "standard_deviation": 0.0031923177483236076, "minimum": 0.04477648547814474, "maximum": 0.055251421646691286, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991663656321795, "standard_deviation": 0.00018932983252526307, "minimum": 0.9987416055910359, "maximum": 0.9993490784300432, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3417013758883619, "standard_deviation": 0.00934592835104076, "minimum": 0.3325489894957899, "maximum": 0.3602240603079694, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.20173359057209717, "standard_deviation": 0.07207981308533379, "minimum": 0.09491577040994294, "maximum": 0.30308207863975084, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02654787257369912, "standard_deviation": 0.002866398992069119, "minimum": 0.0236424446477903, "maximum": 0.03272675429223278, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0019", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2309815883636475, "predicted_mos": 2.2309815883636475, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "d69b020b126967c4031b00d53fd58e15cd28f935cd145e16a3370a1d6adbfc64", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9718663999810815, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.653862575850154, "全帯域_10ms_rms_envelope_nrmse": 0.049342164330041155, "全帯域_10ms_envelope_correlation": 0.9972545164423788, "全帯域_10ms_modulation_mae_db": 0.23599816999890347, "全帯域_40ms_log_rms_mae_db": 0.3278750457104926, "全帯域_40ms_rms_envelope_nrmse": 0.027333586592990636, "全帯域_40ms_envelope_correlation": 0.9991731019143797, "全帯域_40ms_modulation_mae_db": 0.29728066334591546, "300–1000 Hz_10ms_log_rms_mae_db": 1.3105037650255897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06510860761960656, "300–1000 Hz_10ms_envelope_correlation": 0.9961839838904035, "300–1000 Hz_10ms_modulation_mae_db": 0.2744477595226087, "300–1000 Hz_40ms_log_rms_mae_db": 0.8414833742793675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03794925941984412, "300–1000 Hz_40ms_envelope_correlation": 0.9986488667617401, "300–1000 Hz_40ms_modulation_mae_db": 0.3202879603653062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1679983548583976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09730174914802205, "1000–3000 Hz_10ms_envelope_correlation": 0.9917232230914776, "1000–3000 Hz_10ms_modulation_mae_db": 0.2964468173597145, "1000–3000 Hz_40ms_log_rms_mae_db": 0.715757863641419, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05409081154185515, "1000–3000 Hz_40ms_envelope_correlation": 0.9975189595935922, "1000–3000 Hz_40ms_modulation_mae_db": 0.2739401310662875, "3000–6000 Hz_10ms_log_rms_mae_db": 1.268002273099682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11829267377190321, "3000–6000 Hz_10ms_envelope_correlation": 0.9877212568638194, "3000–6000 Hz_10ms_modulation_mae_db": 0.5022062529054346, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7726241161657965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06709433755193632, "3000–6000 Hz_40ms_envelope_correlation": 0.9966851368654909, "3000–6000 Hz_40ms_modulation_mae_db": 0.2579849130706617, "mr_stft_db": 5.490741751946281, "band_mae_db": 0.17548323961486156, "rms_gain_error_db": -0.0857631197165745, "output_peak": 0.3646007451718416}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6630184093502912, "全帯域_10ms_rms_envelope_nrmse": 0.049935887896453056, "全帯域_10ms_envelope_correlation": 0.9972250576153722, "全帯域_10ms_modulation_mae_db": 0.4886576111635841, "全帯域_40ms_log_rms_mae_db": 0.32948847632738565, "全帯域_40ms_rms_envelope_nrmse": 0.026386113189840896, "全帯域_40ms_envelope_correlation": 0.9992853347722624, "全帯域_40ms_modulation_mae_db": 0.47455727301010286, "300–1000 Hz_10ms_log_rms_mae_db": 1.2725121566698931, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06171264407675102, "300–1000 Hz_10ms_envelope_correlation": 0.9966013196811386, "300–1000 Hz_10ms_modulation_mae_db": 0.5637045000795574, "300–1000 Hz_40ms_log_rms_mae_db": 0.849091722353978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03486978626158622, "300–1000 Hz_40ms_envelope_correlation": 0.9988961038371184, "300–1000 Hz_40ms_modulation_mae_db": 0.5044450075911598, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1757238790675872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10073777918743702, "1000–3000 Hz_10ms_envelope_correlation": 0.9912557200205334, "1000–3000 Hz_10ms_modulation_mae_db": 0.21218409712024752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7022475681702169, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05966651154606739, "1000–3000 Hz_40ms_envelope_correlation": 0.9971159487630817, "1000–3000 Hz_40ms_modulation_mae_db": 0.1369657653608637, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3323783895108692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12819519956265485, "3000–6000 Hz_10ms_envelope_correlation": 0.9857063168186, "3000–6000 Hz_10ms_modulation_mae_db": 0.2737146614654455, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8448585485369883, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07827795766261698, "3000–6000 Hz_40ms_envelope_correlation": 0.9953879126252269, "3000–6000 Hz_40ms_modulation_mae_db": 0.48816434979167206, "mr_stft_db": 5.492366317082442, "band_mae_db": 0.177772749225548, "rms_gain_error_db": -0.09974195959551205, "output_peak": 0.3999056572338909}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.66795269270273, "全帯域_10ms_rms_envelope_nrmse": 0.0468812411617492, "全帯域_10ms_envelope_correlation": 0.9975067595946719, "全帯域_10ms_modulation_mae_db": 0.33941964492662, "全帯域_40ms_log_rms_mae_db": 0.3271729956737051, "全帯域_40ms_rms_envelope_nrmse": 0.02236746030147345, "全帯域_40ms_envelope_correlation": 0.9994556504934944, "全帯域_40ms_modulation_mae_db": 0.5705853458395386, "300–1000 Hz_10ms_log_rms_mae_db": 1.2511940588999702, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.060270060850676294, "300–1000 Hz_10ms_envelope_correlation": 0.9967115180465584, "300–1000 Hz_10ms_modulation_mae_db": 0.2863331024173061, "300–1000 Hz_40ms_log_rms_mae_db": 0.8012373677987563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02970357936006464, "300–1000 Hz_40ms_envelope_correlation": 0.9991575764492898, "300–1000 Hz_40ms_modulation_mae_db": 0.41875258140074, "1000–3000 Hz_10ms_log_rms_mae_db": 1.188903356027376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10034561043384436, "1000–3000 Hz_10ms_envelope_correlation": 0.9913663482695131, "1000–3000 Hz_10ms_modulation_mae_db": 0.2886392776453644, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7129543518785892, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05743561629441311, "1000–3000 Hz_40ms_envelope_correlation": 0.9974244818409856, "1000–3000 Hz_40ms_modulation_mae_db": 0.5566511265966442, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2876185870945613, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12404430973028942, "3000–6000 Hz_10ms_envelope_correlation": 0.9867071180619299, "3000–6000 Hz_10ms_modulation_mae_db": 0.4729864706882048, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7784787458044994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07872247706337868, "3000–6000 Hz_40ms_envelope_correlation": 0.9952959440553963, "3000–6000 Hz_40ms_modulation_mae_db": 0.600565066753692, "mr_stft_db": 5.507011338568361, "band_mae_db": 0.18688999843589713, "rms_gain_error_db": -0.07349161088174991, "output_peak": 0.3793485355523145}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6302883920910987, "全帯域_10ms_rms_envelope_nrmse": 0.04459569031366802, "全帯域_10ms_envelope_correlation": 0.9977075193580676, "全帯域_10ms_modulation_mae_db": 0.3867316631002451, "全帯域_40ms_log_rms_mae_db": 0.3307545659019261, "全帯域_40ms_rms_envelope_nrmse": 0.024872818459244588, "全帯域_40ms_envelope_correlation": 0.9992479416944346, "全帯域_40ms_modulation_mae_db": 0.48307824028582586, "300–1000 Hz_10ms_log_rms_mae_db": 1.2390578280768552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06499954343764948, "300–1000 Hz_10ms_envelope_correlation": 0.9961695944312866, "300–1000 Hz_10ms_modulation_mae_db": 0.3295593443966612, "300–1000 Hz_40ms_log_rms_mae_db": 0.7898907835501396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037000650523419124, "300–1000 Hz_40ms_envelope_correlation": 0.9986764642862518, "300–1000 Hz_40ms_modulation_mae_db": 0.24514680173050993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.184721688256731, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10510322077017233, "1000–3000 Hz_10ms_envelope_correlation": 0.9901056119938895, "1000–3000 Hz_10ms_modulation_mae_db": 0.16431586506962936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7473171908465999, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05803759382043157, "1000–3000 Hz_40ms_envelope_correlation": 0.9968653169549027, "1000–3000 Hz_40ms_modulation_mae_db": 0.2654941128414894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3529516892869056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14444863518366857, "3000–6000 Hz_10ms_envelope_correlation": 0.9816762081672517, "3000–6000 Hz_10ms_modulation_mae_db": 0.5438155460342621, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8832507180602343, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09466287541391077, "3000–6000 Hz_40ms_envelope_correlation": 0.9928102873044946, "3000–6000 Hz_40ms_modulation_mae_db": 0.5509029206576023, "mr_stft_db": 5.479541480756189, "band_mae_db": 0.13997767981790693, "rms_gain_error_db": -0.0438321179500748, "output_peak": 0.375960943282298}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6617466570583241, "全帯域_10ms_rms_envelope_nrmse": 0.04776903453275406, "全帯域_10ms_envelope_correlation": 0.9974151619142598, "全帯域_10ms_modulation_mae_db": 0.2573070291875526, "全帯域_40ms_log_rms_mae_db": 0.33336170959686096, "全帯域_40ms_rms_envelope_nrmse": 0.025392366480401798, "全帯域_40ms_envelope_correlation": 0.9992799236417176, "全帯域_40ms_modulation_mae_db": 0.3606261255669419, "300–1000 Hz_10ms_log_rms_mae_db": 1.2766941372238676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0619631063418293, "300–1000 Hz_10ms_envelope_correlation": 0.9965385854083458, "300–1000 Hz_10ms_modulation_mae_db": 0.12000985126005909, "300–1000 Hz_40ms_log_rms_mae_db": 0.834655162821487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03520413934702818, "300–1000 Hz_40ms_envelope_correlation": 0.9988385273302637, "300–1000 Hz_40ms_modulation_mae_db": 0.24525048162179397, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1983915170932955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09320091792874213, "1000–3000 Hz_10ms_envelope_correlation": 0.9924036297301104, "1000–3000 Hz_10ms_modulation_mae_db": 0.06107241039733407, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7434322067100695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05107323413900157, "1000–3000 Hz_40ms_envelope_correlation": 0.9978158517299721, "1000–3000 Hz_40ms_modulation_mae_db": 0.15633095833840038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.318357372983527, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13468666912877017, "3000–6000 Hz_10ms_envelope_correlation": 0.9846245924472389, "3000–6000 Hz_10ms_modulation_mae_db": 0.7363782596440402, "3000–6000 Hz_40ms_log_rms_mae_db": 0.832170032313072, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08775167160225084, "3000–6000 Hz_40ms_envelope_correlation": 0.9944845032364731, "3000–6000 Hz_40ms_modulation_mae_db": 0.5247806555384523, "mr_stft_db": 5.489806174923757, "band_mae_db": 0.195193226078182, "rms_gain_error_db": -0.07716638313178839, "output_peak": 0.3457411424458196}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.70182655104693, "全帯域_10ms_rms_envelope_nrmse": 0.049763719313064184, "全帯域_10ms_envelope_correlation": 0.997288846331277, "全帯域_10ms_modulation_mae_db": 0.23542879917790357, "全帯域_40ms_log_rms_mae_db": 0.37547310702109915, "全帯域_40ms_rms_envelope_nrmse": 0.028714668291478097, "全帯域_40ms_envelope_correlation": 0.9991747739407785, "全帯域_40ms_modulation_mae_db": 0.3598142977714632, "300–1000 Hz_10ms_log_rms_mae_db": 1.3759184955298744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06642746640549749, "300–1000 Hz_10ms_envelope_correlation": 0.9960717813151226, "300–1000 Hz_10ms_modulation_mae_db": 0.2630805702702325, "300–1000 Hz_40ms_log_rms_mae_db": 0.8980213298658688, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03562102654697398, "300–1000 Hz_40ms_envelope_correlation": 0.998886745023535, "300–1000 Hz_40ms_modulation_mae_db": 0.2516087223588886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2011201918177963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10112707171702436, "1000–3000 Hz_10ms_envelope_correlation": 0.9913770615702263, "1000–3000 Hz_10ms_modulation_mae_db": 0.34611937840019974, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7570012624095406, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06590753911091261, "1000–3000 Hz_40ms_envelope_correlation": 0.9965584565105087, "1000–3000 Hz_40ms_modulation_mae_db": 0.2726822148837509, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3448399019651969, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13523094185974938, "3000–6000 Hz_10ms_envelope_correlation": 0.9838231443254329, "3000–6000 Hz_10ms_modulation_mae_db": 0.6276764445090149, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8838839865298164, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08603354909843215, "3000–6000 Hz_40ms_envelope_correlation": 0.9939810372105783, "3000–6000 Hz_40ms_modulation_mae_db": 0.6129683610560633, "mr_stft_db": 5.486650925889191, "band_mae_db": 0.22347277872826385, "rms_gain_error_db": -0.11107422138115693, "output_peak": 0.3558118899070675}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6550012805136701, "全帯域_10ms_rms_envelope_nrmse": 0.04810686441982453, "全帯域_10ms_envelope_correlation": 0.9973488227965582, "全帯域_10ms_modulation_mae_db": 0.2561287823412512, "全帯域_40ms_log_rms_mae_db": 0.31201094883387537, "全帯域_40ms_rms_envelope_nrmse": 0.02695342548407986, "全帯域_40ms_envelope_correlation": 0.9991406160471076, "全帯域_40ms_modulation_mae_db": 0.17162119407483856, "300–1000 Hz_10ms_log_rms_mae_db": 1.274083335370836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06084252014097984, "300–1000 Hz_10ms_envelope_correlation": 0.996660649500645, "300–1000 Hz_10ms_modulation_mae_db": 0.1512416844029778, "300–1000 Hz_40ms_log_rms_mae_db": 0.7987514036564732, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03328239767938159, "300–1000 Hz_40ms_envelope_correlation": 0.9989423393533516, "300–1000 Hz_40ms_modulation_mae_db": 0.13346497560963724, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1784824208807823, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09647414336096989, "1000–3000 Hz_10ms_envelope_correlation": 0.9917356656002373, "1000–3000 Hz_10ms_modulation_mae_db": 0.6149455562385028, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7392829471409021, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06019648422666003, "1000–3000 Hz_40ms_envelope_correlation": 0.996676828866614, "1000–3000 Hz_40ms_modulation_mae_db": 0.3265179146255074, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3214706166073007, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13889083344640243, "3000–6000 Hz_10ms_envelope_correlation": 0.9838680752501447, "3000–6000 Hz_10ms_modulation_mae_db": 0.5925538031591312, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8379118451998394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09228287059927334, "3000–6000 Hz_40ms_envelope_correlation": 0.9940754850023288, "3000–6000 Hz_40ms_modulation_mae_db": 0.24792225422807135, "mr_stft_db": 5.49199950962447, "band_mae_db": 0.18194344218716177, "rms_gain_error_db": -0.05626033110113367, "output_peak": 0.3520075517760813}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6551631980563258, "全帯域_10ms_rms_envelope_nrmse": 0.05194185703067442, "全帯域_10ms_envelope_correlation": 0.9969025317900952, "全帯域_10ms_modulation_mae_db": 0.10997181701024629, "全帯域_40ms_log_rms_mae_db": 0.361174137493481, "全帯域_40ms_rms_envelope_nrmse": 0.03063359663647983, "全帯域_40ms_envelope_correlation": 0.9988738501393979, "全帯域_40ms_modulation_mae_db": 0.22026733925102704, "300–1000 Hz_10ms_log_rms_mae_db": 1.2636326894544483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06958088731973414, "300–1000 Hz_10ms_envelope_correlation": 0.9956013670016052, "300–1000 Hz_10ms_modulation_mae_db": 0.16695942155952895, "300–1000 Hz_40ms_log_rms_mae_db": 0.8484624035765718, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041028044734586544, "300–1000 Hz_40ms_envelope_correlation": 0.9983595125332279, "300–1000 Hz_40ms_modulation_mae_db": 0.2985576222736673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1803602385088234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09971864186511285, "1000–3000 Hz_10ms_envelope_correlation": 0.9917071483003324, "1000–3000 Hz_10ms_modulation_mae_db": 0.46791689443785217, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7569471947826569, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0624485060487916, "1000–3000 Hz_40ms_envelope_correlation": 0.9970995097077269, "1000–3000 Hz_40ms_modulation_mae_db": 0.41337926229168964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.27982349925654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1301951546685806, "3000–6000 Hz_10ms_envelope_correlation": 0.9847933701053455, "3000–6000 Hz_10ms_modulation_mae_db": 0.4064091626760605, "3000–6000 Hz_40ms_log_rms_mae_db": 0.798364337984993, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08457404686831578, "3000–6000 Hz_40ms_envelope_correlation": 0.9937855782935121, "3000–6000 Hz_40ms_modulation_mae_db": 0.6654601966252396, "mr_stft_db": 5.506786544365046, "band_mae_db": 0.19607956022545117, "rms_gain_error_db": -0.06269794137398538, "output_peak": 0.36194645637312245}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79760, "output_samples": 79760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.018001360939092, "spectral_convergence": 0.2116113025643865, "spectral_convergence_status": "ok", "amplitude_floor": 8.669731759960068e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3129139926700555, "spectral_convergence": 0.21261853108860093, "spectral_convergence_status": "ok", "amplitude_floor": 7.258845275738271e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.141309902229696, "spectral_convergence": 0.21764404840857568, "spectral_convergence_status": "ok", "amplitude_floor": 5.316334054535358e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.354150869521306e-05, "output_power": 5.6856475594105594e-05, "delta_db": 0.2608933064995349, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007507214796955326, "output_power": 0.000744113405410695, "delta_db": -0.03839713563935324, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006075331721980808, "output_power": 0.0006003638465128391, "delta_db": -0.05155464579018062, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029803976258038344, "output_power": 0.0002915106412601284, "delta_db": -0.09619795960903521, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002296848467085225, "output_power": 0.00021441437216180822, "delta_db": -0.29878451188542166, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011626570661661749, "output_power": 0.00010832892852699745, "delta_db": -0.30707187826564386, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17548323961486156, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0857631197165745, "pitch": null, "output_rms_dbfs": -26.95598307142527, "output_peak": 0.3646007451718416, "reference_rms_dbfs": -26.870219951708698, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.490741751946281}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.000134521265514355, "log_rms_mae_db": 0.653862575850154, "rms_envelope_nrmse": 0.049342164330041155, "envelope_correlation": 0.9972545164423788, "rms_gain_error_db": -0.08576196539393448, "reference_mean_power": 0.0020578494395295825, "output_mean_power": 0.002017610829630473}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 5.569199266163003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039409678072813294, "output_power": 0.00038490373102824225, "reference_active": true, "delta_db": -0.1025076633603628}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.428667833313107e-05, "output_power": 5.062375456880507e-05, "reference_active": true, "delta_db": -0.303389171041376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.373716545850661e-06, "output_power": 6.878229639238796e-06, "reference_active": true, "delta_db": -0.3020976755949716}], "active_band_count": 3, "active_band_mae_db": 0.23599816999890347}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00012498981593367327, "log_rms_mae_db": 0.3278750457104926, "rms_envelope_nrmse": 0.027333586592990636, "envelope_correlation": 0.9991731019143797, "rms_gain_error_db": -0.08576262831888634, "reference_mean_power": 0.0020703160518770356, "output_mean_power": 0.0020298333634997706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 4.869724313408462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003604621899100367, "output_power": 0.0003523316214091618, "reference_active": true, "delta_db": -0.09908094413365476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.585585870556879e-05, "output_power": 2.35904083229808e-05, "reference_active": true, "delta_db": -0.39823507721958373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932164327326652e-07, "output_power": 3.590699507061293e-07, "reference_active": true, "delta_db": -0.39452596868450784}], "active_band_count": 3, "active_band_mae_db": 0.29728066334591546}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.00011089795767449635, "log_rms_mae_db": 1.3105037650255897, "rms_envelope_nrmse": 0.06510860761960656, "envelope_correlation": 0.9961839838904035, "rms_gain_error_db": -0.06271042408788895, "reference_mean_power": 0.001090370195617981, "output_mean_power": 0.0010747388037698387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 3.4652335334077796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023090422993351253, "output_power": 0.00022626621919427108, "reference_active": true, "delta_db": -0.08812168778552387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.174073740387618e-05, "output_power": 3.693114655754547e-05, "reference_active": true, "delta_db": -0.5316732628203551}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.679949697911218e-06, "output_power": 4.904514444298032e-06, "reference_active": true, "delta_db": 0.20354832796194713}], "active_band_count": 3, "active_band_mae_db": 0.2744477595226087}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00010160775009558359, "log_rms_mae_db": 0.8414833742793675, "rms_envelope_nrmse": 0.03794925941984412, "envelope_correlation": 0.9986488667617401, "rms_gain_error_db": -0.06271032941655562, "reference_mean_power": 0.0010969783403958974, "output_mean_power": 0.0010812522386953519}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 3.031226473561925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002111816957536322, "output_power": 0.000207542855468915, "reference_active": true, "delta_db": -0.0754848520897049}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.975380679045621e-05, "output_power": 1.6974344432000042e-05, "reference_active": true, "delta_db": -0.6585779133061578}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.773024908316542e-07, "output_power": 2.6319259069935745e-07, "reference_active": true, "delta_db": -0.22680111570005596}], "active_band_count": 3, "active_band_mae_db": 0.3202879603653062}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 7.327188751847531e-05, "log_rms_mae_db": 1.1679983548583976, "rms_envelope_nrmse": 0.09730174914802205, "envelope_correlation": 0.9917232230914776, "rms_gain_error_db": -0.17924622982054894, "reference_mean_power": 0.0003949163505892704, "output_mean_power": 0.0003789487615481082}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 2.1879792460690968e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000156993183862756, "output_power": 0.0001520140412127369, "reference_active": true, "delta_db": -0.13997092501635985}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3083527056032412e-05, "output_power": 1.227116051553355e-05, "reference_active": true, "delta_db": -0.2783919981193581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5497810931869906e-06, "output_power": 2.8418467172529427e-06, "reference_active": true, "delta_db": 0.4709775289434256}], "active_band_count": 3, "active_band_mae_db": 0.2964468173597145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 7.131903461227575e-05, "log_rms_mae_db": 0.715757863641419, "rms_envelope_nrmse": 0.05409081154185515, "envelope_correlation": 0.9975189595935922, "rms_gain_error_db": -0.17924746908818034, "reference_mean_power": 0.0003973070610776639, "output_mean_power": 0.00038124270003697165}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 1.9730997112139496e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014395632169995676, "output_power": 0.0001397008378815842, "reference_active": true, "delta_db": -0.1303173036353623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.343232783051936e-06, "output_power": 5.7485885112155265e-06, "reference_active": true, "delta_db": -0.42749426771387744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5197840492846542e-07, "output_power": 1.4301480554109923e-07, "reference_active": true, "delta_db": -0.2640088218496229}], "active_band_count": 3, "active_band_mae_db": 0.2739401310662875}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 4.2584167929876904e-05, "log_rms_mae_db": 1.268002273099682, "rms_envelope_nrmse": 0.11829267377190321, "envelope_correlation": 0.9877212568638194, "rms_gain_error_db": -0.32792698883251953, "reference_mean_power": 0.0001744638441038751, "output_mean_power": 0.00016177549543844053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 1.0543054012178789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.162385199300134e-05, "output_power": 6.566534019520752e-05, "reference_active": true, "delta_db": -0.37721475465524407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.778561151416546e-06, "output_power": 9.320956554293442e-06, "reference_active": true, "delta_db": -0.20814472318653326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4459128360402158e-06, "output_power": 1.787590513538253e-06, "reference_active": true, "delta_db": 0.9212592808745267}], "active_band_count": 3, "active_band_mae_db": 0.5022062529054346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 3.853974797463785e-05, "log_rms_mae_db": 0.7726241161657965, "rms_envelope_nrmse": 0.06709433755193632, "envelope_correlation": 0.9966851368654909, "rms_gain_error_db": -0.3279357921358314, "reference_mean_power": 0.00017551968398739576, "output_mean_power": 0.00016275421663973489}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 9.383868144458744e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.640153736830121e-05, "output_power": 6.081526095144233e-05, "reference_active": true, "delta_db": -0.3816555994031093}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.23065423970599e-06, "output_power": 4.32278635322682e-06, "reference_active": true, "delta_db": 0.09356238726039702}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2239857829027987e-07, "output_power": 1.1426223061968451e-07, "reference_active": true, "delta_db": -0.29873675254847865}], "active_band_count": 3, "active_band_mae_db": 0.2579849130706617}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9291252895896104, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "de0d0544c011501cbb78c1f3d6cc7125e1ade60ded5b2f5017ea70a365c58e64", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9914593010720401, "standard_deviation": 0.000653402082707751, "minimum": 0.9901056119938895, "maximum": 0.9924036297301104, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1844627058138486, "standard_deviation": 0.011284034619190801, "minimum": 1.1679983548583976, "maximum": 1.2011201918177963, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.30645503708360555, "standard_deviation": 0.17424545303724004, "minimum": 0.06107241039733407, "maximum": 0.6149455562385028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09925114180141562, "standard_deviation": 0.0035759017021504075, "minimum": 0.09320091792874213, "maximum": 0.10510322077017233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997134419245923, "standard_deviation": 0.0004328404067164586, "minimum": 0.9965584565105087, "maximum": 0.9978158517299721, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7343675731974992, "standard_deviation": 0.021154824972168928, "minimum": 0.7022475681702169, "maximum": 0.7570012624095406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3002451857505791, "standard_deviation": 0.1358909049654112, "minimum": 0.1369657653608637, "maximum": 0.5566511265966442, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.058607037091016626, "standard_deviation": 0.0046376412456855046, "minimum": 0.05107323413900157, "maximum": 0.06590753911091261, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9848650102549703, "standard_deviation": 0.0018735016262129737, "minimum": 0.9816762081672517, "maximum": 0.9877212568638194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3131802912255728, "standard_deviation": 0.03129427906298273, "minimum": 1.268002273099682, "maximum": 1.3529516892869056, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5194675751351993, "standard_deviation": 0.1415874138875787, "minimum": 0.2737146614654455, "maximum": 0.7363782596440402, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13174805216900232, "standard_deviation": 0.00835394886252473, "minimum": 0.11829267377190321, "maximum": 0.14444863518366857, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945632355741876, "standard_deviation": 0.001194035885158741, "minimum": 0.9928102873044946, "maximum": 0.9966851368654909, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8289427913244048, "standard_deviation": 0.04303737752050811, "minimum": 0.7726241161657965, "maximum": 0.8838839865298164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4935935897151818, "standard_deviation": 0.1582871198745887, "minimum": 0.24792225422807135, "maximum": 0.6654601966252396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08367497323251435, "standard_deviation": 0.008834939158003485, "minimum": 0.06709433755193632, "maximum": 0.09466287541391077, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963173499093883, "standard_deviation": 0.00038085631749674005, "minimum": 0.9956013670016052, "maximum": 0.9967115180465584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.282949558281417, "standard_deviation": 0.04299032166614323, "minimum": 1.2390578280768552, "maximum": 1.3759184955298744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.26941702923861643, "standard_deviation": 0.14008125416935832, "minimum": 0.12000985126005909, "maximum": 0.5637045000795574, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06386310452409052, "standard_deviation": 0.0032155899126617496, "minimum": 0.060270060850676294, "maximum": 0.06958088731973414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9988007669468473, "standard_deviation": 0.00023859506682580992, "minimum": 0.9983595125332279, "maximum": 0.9991575764492898, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8326991934878303, "standard_deviation": 0.0355367451995617, "minimum": 0.7898907835501396, "maximum": 0.8980213298658688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.30218926911896293, "standard_deviation": 0.1148646958921992, "minimum": 0.13346497560963724, "maximum": 0.5044450075911598, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03558236048411055, "standard_deviation": 0.003333197497183137, "minimum": 0.02970357936006464, "maximum": 0.041028044734586544, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18460158428915907, "standard_deviation": 0.02356834793722262, "minimum": 0.13997767981790693, "maximum": 0.22347277872826385, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.493113005394467, "standard_deviation": 0.009444798057850867, "minimum": 5.479541480756189, "maximum": 5.507011338568361, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3669153652178045, "standard_deviation": 0.017520155213946283, "minimum": 0.3457411424458196, "maximum": 0.3999056572338909, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07625346064149695, "standard_deviation": 0.022353365397754094, "minimum": -0.11107422138115693, "maximum": -0.0438321179500748, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973311519803352, "standard_deviation": 0.000233756411480145, "minimum": 0.9969025317900952, "maximum": 0.9977075193580676, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6611074695836905, "standard_deviation": 0.019942362869863485, "minimum": 0.6302883920910987, "maximum": 0.70182655104693, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.28870543961328826, "standard_deviation": 0.1144996602182922, "minimum": 0.10997181701024629, "maximum": 0.4886576111635841, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04854205737477858, "standard_deviation": 0.002228167438934934, "minimum": 0.04459569031366802, "maximum": 0.05194185703067442, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992038990804466, "standard_deviation": 0.00016581565350491377, "minimum": 0.9988738501393979, "maximum": 0.9994556504934944, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3371638733198532, "standard_deviation": 0.02062352029914254, "minimum": 0.31201094883387537, "maximum": 0.37547310702109915, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3672288098932067, "standard_deviation": 0.13687621993434, "minimum": 0.17162119407483856, "maximum": 0.5705853458395386, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.026581754429498644, "standard_deviation": 0.0025015365027981234, "minimum": 0.02236746030147345, "maximum": 0.03063359663647983, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0019", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.0727920532226562, "predicted_mos": 2.0727920532226562, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "f3e6b8dec7dd4ae3c58c130ec1af121613409ad182e19db248bff69b5295aac4", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7803376000374556, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6733619861457124, "全帯域_10ms_rms_envelope_nrmse": 0.04859625315026933, "全帯域_10ms_envelope_correlation": 0.9972501088697775, "全帯域_10ms_modulation_mae_db": 0.17443092782872258, "全帯域_40ms_log_rms_mae_db": 0.3154262242722984, "全帯域_40ms_rms_envelope_nrmse": 0.026187565934685533, "全帯域_40ms_envelope_correlation": 0.9991362718621282, "全帯域_40ms_modulation_mae_db": 0.10703877484286807, "300–1000 Hz_10ms_log_rms_mae_db": 1.2416611908508055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.060585373365252854, "300–1000 Hz_10ms_envelope_correlation": 0.996678630215155, "300–1000 Hz_10ms_modulation_mae_db": 0.6085272738615929, "300–1000 Hz_40ms_log_rms_mae_db": 0.8088045493709651, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031000930834230715, "300–1000 Hz_40ms_envelope_correlation": 0.9990730465553305, "300–1000 Hz_40ms_modulation_mae_db": 0.6218678785608932, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1632183694458234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09109719895648027, "1000–3000 Hz_10ms_envelope_correlation": 0.9926775295576137, "1000–3000 Hz_10ms_modulation_mae_db": 0.43466585564517374, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7109917943037259, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05297679130762832, "1000–3000 Hz_40ms_envelope_correlation": 0.9974840561414406, "1000–3000 Hz_40ms_modulation_mae_db": 0.3601245950625857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3323184445074816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13730100278974025, "3000–6000 Hz_10ms_envelope_correlation": 0.9833635498929308, "3000–6000 Hz_10ms_modulation_mae_db": 0.6207208499108164, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8365145942704558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08785216933226897, "3000–6000 Hz_40ms_envelope_correlation": 0.993768902405299, "3000–6000 Hz_40ms_modulation_mae_db": 0.3929428896949018, "mr_stft_db": 5.475243279319771, "band_mae_db": 0.1984900069672422, "rms_gain_error_db": -0.014588615576409978, "output_peak": 0.3799137948650395}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6595146233403731, "全帯域_10ms_rms_envelope_nrmse": 0.04563270237350013, "全帯域_10ms_envelope_correlation": 0.9976270474599114, "全帯域_10ms_modulation_mae_db": 0.15493399875201655, "全帯域_40ms_log_rms_mae_db": 0.3004138775870769, "全帯域_40ms_rms_envelope_nrmse": 0.021743497229639232, "全帯域_40ms_envelope_correlation": 0.9994726748966013, "全帯域_40ms_modulation_mae_db": 0.2875360809347647, "300–1000 Hz_10ms_log_rms_mae_db": 1.2028106961961091, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.059247151410629625, "300–1000 Hz_10ms_envelope_correlation": 0.9968162176235106, "300–1000 Hz_10ms_modulation_mae_db": 0.20597616395199023, "300–1000 Hz_40ms_log_rms_mae_db": 0.7603694760710247, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0349407730948746, "300–1000 Hz_40ms_envelope_correlation": 0.9988141360982161, "300–1000 Hz_40ms_modulation_mae_db": 0.21894981449196993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1460224027421848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10126154141427257, "1000–3000 Hz_10ms_envelope_correlation": 0.9911697411232382, "1000–3000 Hz_10ms_modulation_mae_db": 0.5574492464489148, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7373086684314267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.062216758697977666, "1000–3000 Hz_40ms_envelope_correlation": 0.9968120845442314, "1000–3000 Hz_40ms_modulation_mae_db": 0.49112709185172543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3008577156947472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12445496526921786, "3000–6000 Hz_10ms_envelope_correlation": 0.9862789329350082, "3000–6000 Hz_10ms_modulation_mae_db": 0.4790569875454584, "3000–6000 Hz_40ms_log_rms_mae_db": 0.814670085404138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07712688124210633, "3000–6000 Hz_40ms_envelope_correlation": 0.9951384889783998, "3000–6000 Hz_40ms_modulation_mae_db": 0.4597257600016286, "mr_stft_db": 5.4997179906611535, "band_mae_db": 0.17508713672909418, "rms_gain_error_db": -0.06708740840921242, "output_peak": 0.40382747901072336}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6575869854068654, "全帯域_10ms_rms_envelope_nrmse": 0.049414218924261726, "全帯域_10ms_envelope_correlation": 0.9971761624315273, "全帯域_10ms_modulation_mae_db": 0.13622092505843267, "全帯域_40ms_log_rms_mae_db": 0.31238660914808003, "全帯域_40ms_rms_envelope_nrmse": 0.024303851634889056, "全帯域_40ms_envelope_correlation": 0.9992828663697744, "全帯域_40ms_modulation_mae_db": 0.19750754168901943, "300–1000 Hz_10ms_log_rms_mae_db": 1.2193428185377915, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0619097514771923, "300–1000 Hz_10ms_envelope_correlation": 0.9965458980704727, "300–1000 Hz_10ms_modulation_mae_db": 0.27428402771429977, "300–1000 Hz_40ms_log_rms_mae_db": 0.7715207966644274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03076375040581614, "300–1000 Hz_40ms_envelope_correlation": 0.9990974247943455, "300–1000 Hz_40ms_modulation_mae_db": 0.1771145532545493, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1531888129164374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10336331183116854, "1000–3000 Hz_10ms_envelope_correlation": 0.990597552551613, "1000–3000 Hz_10ms_modulation_mae_db": 0.2207228543313047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7119139755644504, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05841715270907362, "1000–3000 Hz_40ms_envelope_correlation": 0.9970648572206822, "1000–3000 Hz_40ms_modulation_mae_db": 0.18603934459295998, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2962827783263282, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13245330500664532, "3000–6000 Hz_10ms_envelope_correlation": 0.9852722884881708, "3000–6000 Hz_10ms_modulation_mae_db": 0.7726427935696255, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7785470420184368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0864679373259592, "3000–6000 Hz_40ms_envelope_correlation": 0.9947840070221742, "3000–6000 Hz_40ms_modulation_mae_db": 0.5873737979244259, "mr_stft_db": 5.476892403173937, "band_mae_db": 0.20830115221596604, "rms_gain_error_db": -0.040398407858954904, "output_peak": 0.3570550266098616}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6596068527097374, "全帯域_10ms_rms_envelope_nrmse": 0.04680081699676395, "全帯域_10ms_envelope_correlation": 0.9974747805581688, "全帯域_10ms_modulation_mae_db": 0.2354480918526443, "全帯域_40ms_log_rms_mae_db": 0.34466781800871943, "全帯域_40ms_rms_envelope_nrmse": 0.024776612597372988, "全帯域_40ms_envelope_correlation": 0.9992615345088287, "全帯域_40ms_modulation_mae_db": 0.2673115597185905, "300–1000 Hz_10ms_log_rms_mae_db": 1.2080429640592867, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06153744077826149, "300–1000 Hz_10ms_envelope_correlation": 0.9965699875589231, "300–1000 Hz_10ms_modulation_mae_db": 0.3611358474374919, "300–1000 Hz_40ms_log_rms_mae_db": 0.76865916887368, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03431667353821965, "300–1000 Hz_40ms_envelope_correlation": 0.9988552804533481, "300–1000 Hz_40ms_modulation_mae_db": 0.24865992359905906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.090693115667298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0904702446362515, "1000–3000 Hz_10ms_envelope_correlation": 0.9928043379652963, "1000–3000 Hz_10ms_modulation_mae_db": 0.2730096659317714, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6775178353148327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05134160441775641, "1000–3000 Hz_40ms_envelope_correlation": 0.9977135702449894, "1000–3000 Hz_40ms_modulation_mae_db": 0.16244832482258328, "3000–6000 Hz_10ms_log_rms_mae_db": 1.359872145521688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15578409845017221, "3000–6000 Hz_10ms_envelope_correlation": 0.9809600671668977, "3000–6000 Hz_10ms_modulation_mae_db": 0.8233484050291605, "3000–6000 Hz_40ms_log_rms_mae_db": 0.89578356798745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11065666777413914, "3000–6000 Hz_40ms_envelope_correlation": 0.9925335707224948, "3000–6000 Hz_40ms_modulation_mae_db": 0.7535090481480947, "mr_stft_db": 5.468303382537467, "band_mae_db": 0.2525947156403883, "rms_gain_error_db": -0.05050864693857769, "output_peak": 0.39583638080861083}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6628958110165103, "全帯域_10ms_rms_envelope_nrmse": 0.04921578450662277, "全帯域_10ms_envelope_correlation": 0.9972374846336411, "全帯域_10ms_modulation_mae_db": 0.17894345079382024, "全帯域_40ms_log_rms_mae_db": 0.3292202789391306, "全帯域_40ms_rms_envelope_nrmse": 0.02550237671326992, "全帯域_40ms_envelope_correlation": 0.9992579228969474, "全帯域_40ms_modulation_mae_db": 0.16427856574658314, "300–1000 Hz_10ms_log_rms_mae_db": 1.279623461720476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06571940170093236, "300–1000 Hz_10ms_envelope_correlation": 0.9960877932304167, "300–1000 Hz_10ms_modulation_mae_db": 0.34531456149905965, "300–1000 Hz_40ms_log_rms_mae_db": 0.8369967579496415, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03478177597143068, "300–1000 Hz_40ms_envelope_correlation": 0.9988408385899863, "300–1000 Hz_40ms_modulation_mae_db": 0.17272351365443192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.171543640419185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09821862696355327, "1000–3000 Hz_10ms_envelope_correlation": 0.991547963936028, "1000–3000 Hz_10ms_modulation_mae_db": 0.2200921721568041, "1000–3000 Hz_40ms_log_rms_mae_db": 0.746844829132335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060878080375398054, "1000–3000 Hz_40ms_envelope_correlation": 0.9967481335575081, "1000–3000 Hz_40ms_modulation_mae_db": 0.26915155922183376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2848982641420543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1300255788583672, "3000–6000 Hz_10ms_envelope_correlation": 0.985845609455695, "3000–6000 Hz_10ms_modulation_mae_db": 0.6667685415989691, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8429658835146305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08799380956632534, "3000–6000 Hz_40ms_envelope_correlation": 0.9944261539920333, "3000–6000 Hz_40ms_modulation_mae_db": 0.5474954429765128, "mr_stft_db": 5.480894988291651, "band_mae_db": 0.20932556735205307, "rms_gain_error_db": -0.0648182608980402, "output_peak": 0.3773926430382642}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6659187202592622, "全帯域_10ms_rms_envelope_nrmse": 0.052109104038010855, "全帯域_10ms_envelope_correlation": 0.9969301653442737, "全帯域_10ms_modulation_mae_db": 0.2422008377628555, "全帯域_40ms_log_rms_mae_db": 0.3297827286357988, "全帯域_40ms_rms_envelope_nrmse": 0.028114602155309233, "全帯域_40ms_envelope_correlation": 0.9991260924413631, "全帯域_40ms_modulation_mae_db": 0.27069270006854534, "300–1000 Hz_10ms_log_rms_mae_db": 1.1977011415945102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0654002970423587, "300–1000 Hz_10ms_envelope_correlation": 0.9961195067351514, "300–1000 Hz_10ms_modulation_mae_db": 0.6733209397481521, "300–1000 Hz_40ms_log_rms_mae_db": 0.7641882855241987, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03334025625590302, "300–1000 Hz_40ms_envelope_correlation": 0.9989405783747368, "300–1000 Hz_40ms_modulation_mae_db": 0.64835836418411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2086157576466134, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1006146785109212, "1000–3000 Hz_10ms_envelope_correlation": 0.9914354871445167, "1000–3000 Hz_10ms_modulation_mae_db": 0.4100809390288372, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7743409268026614, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05989124180063537, "1000–3000 Hz_40ms_envelope_correlation": 0.9972793387132928, "1000–3000 Hz_40ms_modulation_mae_db": 0.3262116015685253, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3072612073132879, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13663812144271686, "3000–6000 Hz_10ms_envelope_correlation": 0.9833288236737389, "3000–6000 Hz_10ms_modulation_mae_db": 0.6388809532047656, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8030877704154797, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0805297747106603, "3000–6000 Hz_40ms_envelope_correlation": 0.9947926723836286, "3000–6000 Hz_40ms_modulation_mae_db": 0.4487853259484349, "mr_stft_db": 5.487075884393396, "band_mae_db": 0.19232884968049838, "rms_gain_error_db": -0.09008841037658177, "output_peak": 0.3379444032888984}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6774698780127623, "全帯域_10ms_rms_envelope_nrmse": 0.04798941489137507, "全帯域_10ms_envelope_correlation": 0.9973938852154242, "全帯域_10ms_modulation_mae_db": 0.18357910588397433, "全帯域_40ms_log_rms_mae_db": 0.31438210284510926, "全帯域_40ms_rms_envelope_nrmse": 0.02507762032602143, "全帯域_40ms_envelope_correlation": 0.9993067228333965, "全帯域_40ms_modulation_mae_db": 0.22491997863250668, "300–1000 Hz_10ms_log_rms_mae_db": 1.2442338403536035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0641380903704373, "300–1000 Hz_10ms_envelope_correlation": 0.9962954831821245, "300–1000 Hz_10ms_modulation_mae_db": 0.2811884398693992, "300–1000 Hz_40ms_log_rms_mae_db": 0.7921869770840241, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03717910527500785, "300–1000 Hz_40ms_envelope_correlation": 0.9987029083617055, "300–1000 Hz_40ms_modulation_mae_db": 0.18831214643284935, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1798444803257406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09555271295210627, "1000–3000 Hz_10ms_envelope_correlation": 0.9919837475577947, "1000–3000 Hz_10ms_modulation_mae_db": 0.2558142233257545, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7063203844934749, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05280375679314074, "1000–3000 Hz_40ms_envelope_correlation": 0.9975630330796459, "1000–3000 Hz_40ms_modulation_mae_db": 0.17023377253204186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.291436338533939, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12106079612125785, "3000–6000 Hz_10ms_envelope_correlation": 0.9877083412768004, "3000–6000 Hz_10ms_modulation_mae_db": 0.37064317594559126, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8205502199740531, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07661275483020293, "3000–6000 Hz_40ms_envelope_correlation": 0.9959831622904601, "3000–6000 Hz_40ms_modulation_mae_db": 0.66901573894492, "mr_stft_db": 5.479100842183286, "band_mae_db": 0.2125011999225387, "rms_gain_error_db": -0.07432215597917317, "output_peak": 0.37815076736646336}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6959514508893461, "全帯域_10ms_rms_envelope_nrmse": 0.04919498891331571, "全帯域_10ms_envelope_correlation": 0.9972630308069039, "全帯域_10ms_modulation_mae_db": 0.16230062086016875, "全帯域_40ms_log_rms_mae_db": 0.3448166662471821, "全帯域_40ms_rms_envelope_nrmse": 0.02709350159640964, "全帯域_40ms_envelope_correlation": 0.9991802345395531, "全帯域_40ms_modulation_mae_db": 0.1580411276893755, "300–1000 Hz_10ms_log_rms_mae_db": 1.262898073671714, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06830615519157654, "300–1000 Hz_10ms_envelope_correlation": 0.9957520722363913, "300–1000 Hz_10ms_modulation_mae_db": 0.19281686658601616, "300–1000 Hz_40ms_log_rms_mae_db": 0.8089888946056693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039055730223573996, "300–1000 Hz_40ms_envelope_correlation": 0.9985164187614067, "300–1000 Hz_40ms_modulation_mae_db": 0.28211002508166516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1628754468770321, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10295706646820037, "1000–3000 Hz_10ms_envelope_correlation": 0.9906764926519117, "1000–3000 Hz_10ms_modulation_mae_db": 0.6307723578610136, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7181833002628498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06557591541018559, "1000–3000 Hz_40ms_envelope_correlation": 0.9961768478520064, "1000–3000 Hz_40ms_modulation_mae_db": 0.2700334688060754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2965301219494, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13603224063455008, "3000–6000 Hz_10ms_envelope_correlation": 0.9850735043836512, "3000–6000 Hz_10ms_modulation_mae_db": 0.6627994499053148, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8279067000037081, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0916326315894797, "3000–6000 Hz_40ms_envelope_correlation": 0.9947144390123238, "3000–6000 Hz_40ms_modulation_mae_db": 0.9516912754773621, "mr_stft_db": 5.470818563701703, "band_mae_db": 0.20956049543108093, "rms_gain_error_db": -0.08384842858382988, "output_peak": 0.3588842460110647}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79760, "output_samples": 79760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.008566325922181, "spectral_convergence": 0.21763283649994183, "spectral_convergence_status": "ok", "amplitude_floor": 8.669731759960068e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2893015993054044, "spectral_convergence": 0.21711226829750868, "spectral_convergence_status": "ok", "amplitude_floor": 7.258845275738271e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.127861912731727, "spectral_convergence": 0.21504977208706336, "spectral_convergence_status": "ok", "amplitude_floor": 5.316334054535358e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.354150869521306e-05, "output_power": 5.798138429935719e-05, "delta_db": 0.3459797528294614, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007507214796955326, "output_power": 0.0007690725372342879, "delta_db": 0.10488461327631432, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006075331721980808, "output_power": 0.0006060860930928402, "delta_db": -0.01035676980568615, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029803976258038344, "output_power": 0.0002904727686058659, "delta_db": -0.11168784598225283, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002296848467085225, "output_power": 0.00021999346445713555, "delta_db": -0.18722564837371658, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011626570661661749, "output_power": 0.00010528611342766659, "delta_db": -0.4308054115360219, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1984900069672422, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.014588615576409978, "pitch": null, "output_rms_dbfs": -26.884808567285106, "output_peak": 0.3799137948650395, "reference_rms_dbfs": -26.870219951708698, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.475243279319771}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.000134521265514355, "log_rms_mae_db": 0.6733619861457124, "rms_envelope_nrmse": 0.04859625315026933, "envelope_correlation": 0.9972501088697775, "rms_gain_error_db": -0.014587495939086556, "reference_mean_power": 0.0020578494395295825, "output_mean_power": 0.0020509489347172115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 5.569199266163003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039409678072813294, "output_power": 0.0003867092827280318, "reference_active": true, "delta_db": -0.08218290063052239}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.428667833313107e-05, "output_power": 5.3754785990851235e-05, "reference_active": true, "delta_db": -0.04276131888260541}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.373716545850661e-06, "output_power": 8.082045576674316e-06, "reference_active": true, "delta_db": 0.39834856397303986}], "active_band_count": 3, "active_band_mae_db": 0.17443092782872258}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00012498981593367327, "log_rms_mae_db": 0.3154262242722984, "rms_envelope_nrmse": 0.026187565934685533, "envelope_correlation": 0.9991362718621282, "rms_gain_error_db": -0.014588516894939563, "reference_mean_power": 0.0020703160518770356, "output_mean_power": 0.0020633732582028778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 4.869724313408462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003604621899100367, "output_power": 0.00035372619485189195, "reference_active": true, "delta_db": -0.08192494573121704}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.585585870556879e-05, "output_power": 2.634218821179314e-05, "reference_active": true, "delta_db": 0.08092882596656492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932164327326652e-07, "output_power": 4.07810040588978e-07, "reference_active": true, "delta_db": 0.15826255283082225}], "active_band_count": 3, "active_band_mae_db": 0.10703877484286807}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.00011089795767449635, "log_rms_mae_db": 1.2416611908508055, "rms_envelope_nrmse": 0.060585373365252854, "envelope_correlation": 0.996678630215155, "rms_gain_error_db": 0.016740096088564096, "reference_mean_power": 0.001090370195617981, "output_mean_power": 0.0010945811921212274}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 3.4652335334077796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023090422993351253, "output_power": 0.00022572381128881867, "reference_active": true, "delta_db": -0.09854514217483305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.174073740387618e-05, "output_power": 3.946741399843863e-05, "reference_active": true, "delta_db": -0.24321445878132553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.679949697911218e-06, "output_power": 6.586025489155759e-06, "reference_active": true, "delta_db": 1.4838222206286202}], "active_band_count": 3, "active_band_mae_db": 0.6085272738615929}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00010160775009558359, "log_rms_mae_db": 0.8088045493709651, "rms_envelope_nrmse": 0.031000930834230715, "envelope_correlation": 0.9990730465553305, "rms_gain_error_db": 0.01674008513402326, "reference_mean_power": 0.0010969783403958974, "output_mean_power": 0.001101214854696647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 3.031226473561925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002111816957536322, "output_power": 0.00020694476503428644, "reference_active": true, "delta_db": -0.08801828097234654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.975380679045621e-05, "output_power": 1.8698171052329296e-05, "reference_active": true, "delta_db": -0.23851673249920818}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.773024908316542e-07, "output_power": 3.9523976536908813e-07, "reference_active": true, "delta_db": 1.5390686222111247}], "active_band_count": 3, "active_band_mae_db": 0.6218678785608932}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 7.327188751847531e-05, "log_rms_mae_db": 1.1632183694458234, "rms_envelope_nrmse": 0.09109719895648027, "envelope_correlation": 0.9926775295576137, "rms_gain_error_db": -0.1410654532295943, "reference_mean_power": 0.0003949163505892704, "output_mean_power": 0.00038229495754894176}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 2.1879792460690968e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000156993183862756, "output_power": 0.00015293950726077968, "reference_active": true, "delta_db": -0.11361110505266464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3083527056032412e-05, "output_power": 1.355895941184184e-05, "reference_active": true, "delta_db": 0.15501523854794977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5497810931869906e-06, "output_power": 3.2362347639305267e-06, "reference_active": true, "delta_db": 1.0353712233349068}], "active_band_count": 3, "active_band_mae_db": 0.43466585564517374}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 7.131903461227575e-05, "log_rms_mae_db": 0.7109917943037259, "rms_envelope_nrmse": 0.05297679130762832, "envelope_correlation": 0.9974840561414406, "rms_gain_error_db": -0.1410712862734252, "reference_mean_power": 0.0003973070610776639, "output_mean_power": 0.0003846087451670247}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 1.9730997112139496e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014395632169995676, "output_power": 0.00014066398482624886, "reference_active": true, "delta_db": -0.10047825002046513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.343232783051936e-06, "output_power": 6.915672061846067e-06, "reference_active": true, "delta_db": 0.37523741668307464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5197840492846542e-07, "output_power": 1.746818142873503e-07, "reference_active": true, "delta_db": 0.6046581184842172}], "active_band_count": 3, "active_band_mae_db": 0.3601245950625857}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 4.2584167929876904e-05, "log_rms_mae_db": 1.3323184445074816, "rms_envelope_nrmse": 0.13730100278974025, "envelope_correlation": 0.9833635498929308, "rms_gain_error_db": -0.38312984929779803, "reference_mean_power": 0.0001744638441038751, "output_mean_power": 0.00015973219235600826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 1.0543054012178789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.162385199300134e-05, "output_power": 6.541799785181784e-05, "reference_active": true, "delta_db": -0.39360425830416435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.778561151416546e-06, "output_power": 8.560603182966935e-06, "reference_active": true, "delta_db": -0.5777058972761339}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4459128360402158e-06, "output_power": 1.7751185098998254e-06, "reference_active": true, "delta_db": 0.8908523941521512}], "active_band_count": 3, "active_band_mae_db": 0.6207208499108164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 3.853974797463785e-05, "log_rms_mae_db": 0.8365145942704558, "rms_envelope_nrmse": 0.08785216933226897, "envelope_correlation": 0.993768902405299, "rms_gain_error_db": -0.38313544830913715, "reference_mean_power": 0.00017551968398739576, "output_mean_power": 0.0001606986703990862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 9.383868144458744e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.640153736830121e-05, "output_power": 6.038982444254437e-05, "reference_active": true, "delta_db": -0.41214367445650263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.23065423970599e-06, "output_power": 4.029367632242573e-06, "reference_active": true, "delta_db": -0.2117063958859741}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2239857829027987e-07, "output_power": 1.0771557731790871e-07, "reference_active": true, "delta_db": -0.5549785987422287}], "active_band_count": 3, "active_band_mae_db": 0.3929428896949018}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9359863817075367, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6ec7c9d8a1016af082279ca5e517dda86882d9c2d09528ca378086c1e15c7919", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9916116065610014, "standard_deviation": 0.0008306921394311085, "minimum": 0.990597552551613, "maximum": 0.9928043379652963, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1595002532550394, "standard_deviation": 0.033705422958618454, "minimum": 1.090693115667298, "maximum": 1.2086157576466134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3753259143411968, "standard_deviation": 0.15852093770444767, "minimum": 0.2200921721568041, "maximum": 0.6307723578610136, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09794192271661925, "standard_deviation": 0.005087347384057477, "minimum": 0.0904702446362515, "maximum": 0.10336331183116854, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971052401692245, "standard_deviation": 0.0005113651342330223, "minimum": 0.9961768478520064, "maximum": 0.9977135702449894, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7229277142882196, "standard_deviation": 0.029389738850430835, "minimum": 0.6775178353148327, "maximum": 0.7743409268026614, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.27942121980729134, "standard_deviation": 0.11214791131141331, "minimum": 0.16244832482258328, "maximum": 0.49112709185172543, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.058012662688974474, "standard_deviation": 0.005123840981274567, "minimum": 0.05134160441775641, "maximum": 0.06557591541018559, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9847288896591117, "standard_deviation": 0.002103372503437022, "minimum": 0.9809600671668977, "maximum": 0.9877083412768004, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3086821269986157, "standard_deviation": 0.025090801823826885, "minimum": 1.2848982641420543, "maximum": 1.359872145521688, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6293576445887127, "standard_deviation": 0.14656530969158058, "minimum": 0.37064317594559126, "maximum": 0.8233484050291605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13421876357158347, "standard_deviation": 0.010507611970695416, "minimum": 0.12106079612125785, "maximum": 0.15578409845017221, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945176746008517, "standard_deviation": 0.0010158769368206044, "minimum": 0.9925335707224948, "maximum": 0.9959831622904601, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.827503232948544, "standard_deviation": 0.034202177542417424, "minimum": 0.7785470420184368, "maximum": 0.89578356798745, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6013174098895351, "standard_deviation": 0.18529228079381968, "minimum": 0.3929428896949018, "maximum": 0.9516912754773621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08735907829639274, "standard_deviation": 0.010884469933088536, "minimum": 0.07661275483020293, "maximum": 0.11065666777413914, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963581986065181, "standard_deviation": 0.0003573631132078209, "minimum": 0.9957520722363913, "maximum": 0.9968162176235106, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.232039273373037, "standard_deviation": 0.029828675417750024, "minimum": 1.1977011415945102, "maximum": 1.279623461720476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3678205150835002, "standard_deviation": 0.17924265987198984, "minimum": 0.19281686658601616, "maximum": 0.6733209397481521, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06335545766708015, "standard_deviation": 0.0030434547691920374, "minimum": 0.059247151410629625, "maximum": 0.06830615519157654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9988550789986343, "standard_deviation": 0.00019021690390723275, "minimum": 0.9985164187614067, "maximum": 0.9990974247943455, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7889643632679538, "standard_deviation": 0.027412748744159224, "minimum": 0.7603694760710247, "maximum": 0.8369967579496415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.319762027407441, "standard_deviation": 0.19825217713787743, "minimum": 0.17272351365443192, "maximum": 0.64835836418411, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03442237444988208, "standard_deviation": 0.0028218387104862273, "minimum": 0.03076375040581614, "maximum": 0.039055730223573996, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20727364049235775, "standard_deviation": 0.022130435198988205, "minimum": 0.17508713672909418, "maximum": 0.2525947156403883, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.479755916782795, "standard_deviation": 0.00995269779489691, "minimum": 5.468303382537467, "maximum": 5.4997179906611535, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3736255926248657, "standard_deviation": 0.02152978218373937, "minimum": 0.3379444032888984, "maximum": 0.40382747901072336, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0607075418275975, "standard_deviation": 0.024741891144964983, "minimum": -0.09008841037658177, "maximum": -0.014588615576409978, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9972940831649535, "standard_deviation": 0.00020905231054631028, "minimum": 0.9969301653442737, "maximum": 0.9976270474599114, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6690382884725712, "standard_deviation": 0.012941354699190062, "minimum": 0.6575869854068654, "maximum": 0.6959514508893461, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.18350724484907935, "standard_deviation": 0.03731745987430136, "minimum": 0.13622092505843267, "maximum": 0.2422008377628555, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04861916047426494, "standard_deviation": 0.0019317561173600368, "minimum": 0.04563270237350013, "maximum": 0.052109104038010855, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992530400435741, "standard_deviation": 0.00011158472875552009, "minimum": 0.9991260924413631, "maximum": 0.9994726748966013, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3238870382104244, "standard_deviation": 0.015938592284615066, "minimum": 0.3004138775870769, "maximum": 0.3448166662471821, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.20966579116528167, "standard_deviation": 0.06408590677055924, "minimum": 0.10703877484286807, "maximum": 0.2875360809347647, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02534995352344963, "standard_deviation": 0.0019239769329240809, "minimum": 0.021743497229639232, "maximum": 0.028114602155309233, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5142-33396-0030", "language": "en", "style": "normal", "speaker": "librispeech_5142", "source_utterance_id": "5142-33396-0030", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 44240, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5142-33396-0030.wav", "source_file_sha256": "bd387c102028ebcdbedb3ff002cca7538dbc777739626322fc12bb6eb919bc14", "decoded_pcm_sha256": "cdc8b64c805f94ff6dd7f431b45e924a4edf992efb5d08cda66456142b76481e", "original_source_sha256": "87b13e7bd9a84ee5f9ba237f84dc992d0fbbad061a6df1750b87a47756642b33", "corpus_entry_id": "librispeech_5142-33396-0030"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.437169075012207, "predicted_mos": 4.437169075012207, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "38f2b978cd8d9e0037d87bf71013ec19c8e3acb556672c6fc5de03fd60ac87ad", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.575701243997173, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5142-33396-0030", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "38f2b978cd8d9e0037d87bf71013ec19c8e3acb556672c6fc5de03fd60ac87ad", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "bf44bc8f7b5de1f328031c764965206870e3b2a80609be308d2d2082012fb5c3"}, "rows": [{"case": "librispeech_5142-33396-0030", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.2879910469055176, "predicted_mos": 3.2879910469055176, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "dbf02dcdcc02726d8587e2ef16506bdf64c2e0bcb8b87e71f319a11af53c4d20", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5166037249982764, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5142-33396-0030", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "dbf02dcdcc02726d8587e2ef16506bdf64c2e0bcb8b87e71f319a11af53c4d20", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8608693616696496, "全帯域_10ms_rms_envelope_nrmse": 0.058106879790551146, "全帯域_10ms_envelope_correlation": 0.9971500473816876, "全帯域_10ms_modulation_mae_db": 0.6956862034332228, "全帯域_40ms_log_rms_mae_db": 0.45805904093351657, "全帯域_40ms_rms_envelope_nrmse": 0.025853609114452045, "全帯域_40ms_envelope_correlation": 0.999493506312677, "全帯域_40ms_modulation_mae_db": 0.48243822381977636, "300–1000 Hz_10ms_log_rms_mae_db": 1.069439185997, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06296221118072791, "300–1000 Hz_10ms_envelope_correlation": 0.997088662692182, "300–1000 Hz_10ms_modulation_mae_db": 0.7623490828550911, "300–1000 Hz_40ms_log_rms_mae_db": 0.5697361571565741, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030018571393876016, "300–1000 Hz_40ms_envelope_correlation": 0.9993522244435352, "300–1000 Hz_40ms_modulation_mae_db": 0.7671849606155804, "1000–3000 Hz_10ms_log_rms_mae_db": 1.316132146256139, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09998905029719755, "1000–3000 Hz_10ms_envelope_correlation": 0.992926960214774, "1000–3000 Hz_10ms_modulation_mae_db": 1.1878373716503114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7966211490477387, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0581518646146437, "1000–3000 Hz_40ms_envelope_correlation": 0.9976694182186628, "1000–3000 Hz_40ms_modulation_mae_db": 0.6540144492555865, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4454142914504537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2273348875917786, "3000–6000 Hz_10ms_envelope_correlation": 0.9634673748913626, "3000–6000 Hz_10ms_modulation_mae_db": 1.893959189896302, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0674905870168054, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13212232646098965, "3000–6000 Hz_40ms_envelope_correlation": 0.9874396831970457, "3000–6000 Hz_40ms_modulation_mae_db": 1.7216431897922364, "mr_stft_db": 5.40337523081761, "band_mae_db": 0.19048980014288744, "rms_gain_error_db": -0.09483114515583305, "output_peak": 0.40028888858653905}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9109512902384506, "全帯域_10ms_rms_envelope_nrmse": 0.04770619905293834, "全帯域_10ms_envelope_correlation": 0.9981051122394816, "全帯域_10ms_modulation_mae_db": 0.3543692708156478, "全帯域_40ms_log_rms_mae_db": 0.5421987636476663, "全帯域_40ms_rms_envelope_nrmse": 0.02627414216067795, "全帯域_40ms_envelope_correlation": 0.9994643728314222, "全帯域_40ms_modulation_mae_db": 0.3641547783817682, "300–1000 Hz_10ms_log_rms_mae_db": 1.0326437028910707, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061824595727998664, "300–1000 Hz_10ms_envelope_correlation": 0.9971746358867396, "300–1000 Hz_10ms_modulation_mae_db": 0.1986265210094702, "300–1000 Hz_40ms_log_rms_mae_db": 0.5777611101567949, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03490402472468954, "300–1000 Hz_40ms_envelope_correlation": 0.999072805799247, "300–1000 Hz_40ms_modulation_mae_db": 0.6737239062462267, "1000–3000 Hz_10ms_log_rms_mae_db": 1.209827805176862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08468579773518495, "1000–3000 Hz_10ms_envelope_correlation": 0.9949318382574789, "1000–3000 Hz_10ms_modulation_mae_db": 0.6527215434019892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7851860752764331, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05030106862134669, "1000–3000 Hz_40ms_envelope_correlation": 0.9982248744086233, "1000–3000 Hz_40ms_modulation_mae_db": 0.2650512361947156, "3000–6000 Hz_10ms_log_rms_mae_db": 1.32480786125913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13900330311166562, "3000–6000 Hz_10ms_envelope_correlation": 0.9875552829765374, "3000–6000 Hz_10ms_modulation_mae_db": 0.6778246307466039, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9240250732285216, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07846769599165614, "3000–6000 Hz_40ms_envelope_correlation": 0.9971668609570276, "3000–6000 Hz_40ms_modulation_mae_db": 0.42081998248707997, "mr_stft_db": 5.401855592190523, "band_mae_db": 0.23137458799108054, "rms_gain_error_db": -0.09005058027210908, "output_peak": 0.4153885772462026}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9222118540882571, "全帯域_10ms_rms_envelope_nrmse": 0.05244533249251678, "全帯域_10ms_envelope_correlation": 0.9977936536686122, "全帯域_10ms_modulation_mae_db": 0.35106358277804856, "全帯域_40ms_log_rms_mae_db": 0.5115249526052555, "全帯域_40ms_rms_envelope_nrmse": 0.03317942330590424, "全帯域_40ms_envelope_correlation": 0.9991988587864329, "全帯域_40ms_modulation_mae_db": 0.529562902878824, "300–1000 Hz_10ms_log_rms_mae_db": 1.05144188666329, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05229355958158044, "300–1000 Hz_10ms_envelope_correlation": 0.9979797754132346, "300–1000 Hz_10ms_modulation_mae_db": 0.19343134575096013, "300–1000 Hz_40ms_log_rms_mae_db": 0.581319502053006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.027476122270646573, "300–1000 Hz_40ms_envelope_correlation": 0.9994190318408567, "300–1000 Hz_40ms_modulation_mae_db": 0.24994732333367056, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2116971700606134, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08678917489959434, "1000–3000 Hz_10ms_envelope_correlation": 0.9953055511042587, "1000–3000 Hz_10ms_modulation_mae_db": 0.9022479428641864, "1000–3000 Hz_40ms_log_rms_mae_db": 0.70930988311634, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05657557597953017, "1000–3000 Hz_40ms_envelope_correlation": 0.9984638675661063, "1000–3000 Hz_40ms_modulation_mae_db": 0.710599056175778, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3200440108773612, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1657211608299523, "3000–6000 Hz_10ms_envelope_correlation": 0.982076797447186, "3000–6000 Hz_10ms_modulation_mae_db": 0.8662177957489233, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9448444292348501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09764674851111156, "3000–6000 Hz_40ms_envelope_correlation": 0.9953211409746494, "3000–6000 Hz_40ms_modulation_mae_db": 0.9294216709701874, "mr_stft_db": 5.404688672800618, "band_mae_db": 0.24049428739084186, "rms_gain_error_db": -0.13148866126973618, "output_peak": 0.39706113135849935}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8042675213502922, "全帯域_10ms_rms_envelope_nrmse": 0.04207193855138991, "全帯域_10ms_envelope_correlation": 0.998535647499916, "全帯域_10ms_modulation_mae_db": 0.1266513608957462, "全帯域_40ms_log_rms_mae_db": 0.41951959501635233, "全帯域_40ms_rms_envelope_nrmse": 0.020891266841937206, "全帯域_40ms_envelope_correlation": 0.999683421044742, "全帯域_40ms_modulation_mae_db": 0.30154696244253665, "300–1000 Hz_10ms_log_rms_mae_db": 1.0813743371880562, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05261624934048992, "300–1000 Hz_10ms_envelope_correlation": 0.9979529336436296, "300–1000 Hz_10ms_modulation_mae_db": 0.2845123912532068, "300–1000 Hz_40ms_log_rms_mae_db": 0.583847628586382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.025914618979125482, "300–1000 Hz_40ms_envelope_correlation": 0.999480635661453, "300–1000 Hz_40ms_modulation_mae_db": 0.3219358153311806, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1486152233628508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08431224122601678, "1000–3000 Hz_10ms_envelope_correlation": 0.9950652399972176, "1000–3000 Hz_10ms_modulation_mae_db": 0.9273779333065114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6929746145197004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04493157858636651, "1000–3000 Hz_40ms_envelope_correlation": 0.9987385040450348, "1000–3000 Hz_40ms_modulation_mae_db": 0.5858550786296208, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3876075428547676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14342926478440163, "3000–6000 Hz_10ms_envelope_correlation": 0.987711494141462, "3000–6000 Hz_10ms_modulation_mae_db": 0.5569304747827709, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9680486834165395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09532590125886346, "3000–6000 Hz_40ms_envelope_correlation": 0.9962461502666317, "3000–6000 Hz_40ms_modulation_mae_db": 0.8242830981756879, "mr_stft_db": 5.403890924287981, "band_mae_db": 0.16634417254564324, "rms_gain_error_db": -0.08400921599558586, "output_peak": 0.4209698828233987}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9391689943141662, "全帯域_10ms_rms_envelope_nrmse": 0.044275445039863366, "全帯域_10ms_envelope_correlation": 0.9983223337885413, "全帯域_10ms_modulation_mae_db": 0.38573049697268663, "全帯域_40ms_log_rms_mae_db": 0.5157727869054887, "全帯域_40ms_rms_envelope_nrmse": 0.02058869267944883, "全帯域_40ms_envelope_correlation": 0.9996372429029807, "全帯域_40ms_modulation_mae_db": 0.5155967804353038, "300–1000 Hz_10ms_log_rms_mae_db": 1.0810124807710182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05571490448264815, "300–1000 Hz_10ms_envelope_correlation": 0.9977314601154208, "300–1000 Hz_10ms_modulation_mae_db": 0.3304609726126363, "300–1000 Hz_40ms_log_rms_mae_db": 0.6383489896522698, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.028537789093435934, "300–1000 Hz_40ms_envelope_correlation": 0.999400917891871, "300–1000 Hz_40ms_modulation_mae_db": 0.3334772087769286, "1000–3000 Hz_10ms_log_rms_mae_db": 1.255131329641565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08666935034261507, "1000–3000 Hz_10ms_envelope_correlation": 0.9947354134075548, "1000–3000 Hz_10ms_modulation_mae_db": 0.849862843489873, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7577407881253958, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04698166005021716, "1000–3000 Hz_40ms_envelope_correlation": 0.9985579908498972, "1000–3000 Hz_40ms_modulation_mae_db": 0.8018792961782354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3227852733009318, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12185860985232029, "3000–6000 Hz_10ms_envelope_correlation": 0.990604826206923, "3000–6000 Hz_10ms_modulation_mae_db": 0.2887927678836681, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9298612623795507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08048803586026834, "3000–6000 Hz_40ms_envelope_correlation": 0.996642745189248, "3000–6000 Hz_40ms_modulation_mae_db": 0.33409126195443023, "mr_stft_db": 5.386789653910135, "band_mae_db": 0.15549190691704182, "rms_gain_error_db": -0.045684112228224355, "output_peak": 0.3926908876143679}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8590103048754829, "全帯域_10ms_rms_envelope_nrmse": 0.0527349065056998, "全帯域_10ms_envelope_correlation": 0.9976258960979602, "全帯域_10ms_modulation_mae_db": 0.3801234384671935, "全帯域_40ms_log_rms_mae_db": 0.42474876755756824, "全帯域_40ms_rms_envelope_nrmse": 0.02491506705808776, "全帯域_40ms_envelope_correlation": 0.9994821948367093, "全帯域_40ms_modulation_mae_db": 0.18107051962383822, "300–1000 Hz_10ms_log_rms_mae_db": 1.0878334463644528, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06025958259619183, "300–1000 Hz_10ms_envelope_correlation": 0.9973514291607266, "300–1000 Hz_10ms_modulation_mae_db": 0.14036666154512964, "300–1000 Hz_40ms_log_rms_mae_db": 0.5941066940792681, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02788360278143795, "300–1000 Hz_40ms_envelope_correlation": 0.9994268276422945, "300–1000 Hz_40ms_modulation_mae_db": 0.31563809050377656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2245134094356465, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10251352524579578, "1000–3000 Hz_10ms_envelope_correlation": 0.9927700632200106, "1000–3000 Hz_10ms_modulation_mae_db": 1.2880435345248855, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7607004191984117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.060995942498979654, "1000–3000 Hz_40ms_envelope_correlation": 0.9976851961999531, "1000–3000 Hz_40ms_modulation_mae_db": 0.9169425919967219, "3000–6000 Hz_10ms_log_rms_mae_db": 1.429553313359431, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17976515774280954, "3000–6000 Hz_10ms_envelope_correlation": 0.9775997757144109, "3000–6000 Hz_10ms_modulation_mae_db": 1.2530086323161302, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9752231379851984, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08062104540519782, "3000–6000 Hz_40ms_envelope_correlation": 0.9964889679821098, "3000–6000 Hz_40ms_modulation_mae_db": 0.7367197769573397, "mr_stft_db": 5.403481308141424, "band_mae_db": 0.19787324663018926, "rms_gain_error_db": -0.06453773779591823, "output_peak": 0.4143260996812422}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.888504301844857, "全帯域_10ms_rms_envelope_nrmse": 0.043244132457089156, "全帯域_10ms_envelope_correlation": 0.998386481728847, "全帯域_10ms_modulation_mae_db": 0.4270416838148012, "全帯域_40ms_log_rms_mae_db": 0.4480242828516786, "全帯域_40ms_rms_envelope_nrmse": 0.021711737098113054, "全帯域_40ms_envelope_correlation": 0.9995765089276638, "全帯域_40ms_modulation_mae_db": 0.21657874322334184, "300–1000 Hz_10ms_log_rms_mae_db": 1.0809151413288973, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.057613260609844376, "300–1000 Hz_10ms_envelope_correlation": 0.9978463684375575, "300–1000 Hz_10ms_modulation_mae_db": 0.6273645244621232, "300–1000 Hz_40ms_log_rms_mae_db": 0.5943514393871777, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03347057753706233, "300–1000 Hz_40ms_envelope_correlation": 0.9994246267489556, "300–1000 Hz_40ms_modulation_mae_db": 0.49483206462418394, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2592638750839393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08822899004603293, "1000–3000 Hz_10ms_envelope_correlation": 0.9945702844061795, "1000–3000 Hz_10ms_modulation_mae_db": 0.7231090539568178, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7916309461191748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05352286072922747, "1000–3000 Hz_40ms_envelope_correlation": 0.9980922693973705, "1000–3000 Hz_40ms_modulation_mae_db": 0.3660521765173815, "3000–6000 Hz_10ms_log_rms_mae_db": 1.403171808482855, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14543735845273403, "3000–6000 Hz_10ms_envelope_correlation": 0.9852584963435485, "3000–6000 Hz_10ms_modulation_mae_db": 1.0404923273361881, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9846656410621276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07058156123152501, "3000–6000 Hz_40ms_envelope_correlation": 0.9967914501283741, "3000–6000 Hz_40ms_modulation_mae_db": 0.9099053154613115, "mr_stft_db": 5.402755308066941, "band_mae_db": 0.1585711551115356, "rms_gain_error_db": 0.00541854838630079, "output_peak": 0.4130714559952835}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.881324430076609, "全帯域_10ms_rms_envelope_nrmse": 0.044609616761716805, "全帯域_10ms_envelope_correlation": 0.9983080045704176, "全帯域_10ms_modulation_mae_db": 0.40488953030685293, "全帯域_40ms_log_rms_mae_db": 0.4677371087198264, "全帯域_40ms_rms_envelope_nrmse": 0.022196786867192207, "全帯域_40ms_envelope_correlation": 0.9995906720124569, "全帯域_40ms_modulation_mae_db": 0.04482105782014421, "300–1000 Hz_10ms_log_rms_mae_db": 1.1103871204277962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.054730732509183, "300–1000 Hz_10ms_envelope_correlation": 0.9978127246445349, "300–1000 Hz_10ms_modulation_mae_db": 0.5113737273448259, "300–1000 Hz_40ms_log_rms_mae_db": 0.6327171321291162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.028703563345829127, "300–1000 Hz_40ms_envelope_correlation": 0.9993841224544734, "300–1000 Hz_40ms_modulation_mae_db": 0.4274506455337422, "1000–3000 Hz_10ms_log_rms_mae_db": 1.253423506020843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09433750438273145, "1000–3000 Hz_10ms_envelope_correlation": 0.993820885482851, "1000–3000 Hz_10ms_modulation_mae_db": 0.8926327436255116, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7650699282248828, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05780346972526308, "1000–3000 Hz_40ms_envelope_correlation": 0.997809305718932, "1000–3000 Hz_40ms_modulation_mae_db": 0.29443207642973884, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3979526636201896, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12279252610187553, "3000–6000 Hz_10ms_envelope_correlation": 0.989774107749687, "3000–6000 Hz_10ms_modulation_mae_db": 0.9721110961909293, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9851246008619013, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06105620234326897, "3000–6000 Hz_40ms_envelope_correlation": 0.9979079398769942, "3000–6000 Hz_40ms_modulation_mae_db": 0.9351179981850081, "mr_stft_db": 5.416443597131978, "band_mae_db": 0.1691729741289789, "rms_gain_error_db": -0.05975913016894428, "output_peak": 0.41874831301625814}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44240, "output_samples": 44240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.1110963501388955, "spectral_convergence": 0.2184908077702738, "spectral_convergence_status": "ok", "amplitude_floor": 7.692579600611731e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.278248019994025, "spectral_convergence": 0.22396533048627848, "spectral_convergence_status": "ok", "amplitude_floor": 5.999960941709396e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.82078132231991, "spectral_convergence": 0.21907793745177984, "spectral_convergence_status": "ok", "amplitude_floor": 5.417405012540354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 6.282816067608343e-05, "output_power": 6.47702238599798e-05, "delta_db": 0.13221052345586376, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019406375287345015, "output_power": 0.00019689940777249969, "delta_db": 0.06299984217446966, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005567857807558596, "output_power": 0.0005424408842629976, "delta_db": -0.11335720430640696, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00030647737825017173, "output_power": 0.0002968384409453679, "delta_db": -0.13878281912490453, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00015415279416012865, "output_power": 0.00015132023779588667, "delta_db": -0.08054386013924467, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.966178555848869e-05, "output_power": 4.3104029882154245e-05, "delta_db": -0.6150445516564351, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19048980014288744, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09483114515583305, "pitch": null, "output_rms_dbfs": -28.87605084095371, "output_peak": 0.40028888858653905, "reference_rms_dbfs": -28.781219695797876, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.40337523081761}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 0.00013943874660832744, "log_rms_mae_db": 0.8608693616696496, "rms_envelope_nrmse": 0.058106879790551146, "envelope_correlation": 0.9971500473816876, "rms_gain_error_db": -0.09483131562101942, "reference_mean_power": 0.0013263679977567136, "output_mean_power": 0.0012977197171873572}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 5.047819293813421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030810693936871264, "output_power": 0.0003003008944306123, "reference_active": true, "delta_db": -0.11144854142631469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2023354078518551e-05, "output_power": 1.1440249048083321e-05, "reference_active": true, "delta_db": -0.21590158016542232}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2193973896229425e-06, "output_power": 4.827757380453419e-06, "reference_active": true, "delta_db": 1.7597084887079313}], "active_band_count": 3, "active_band_mae_db": 0.6956862034332228}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 0.0001327212418155818, "log_rms_mae_db": 0.45805904093351657, "rms_envelope_nrmse": 0.025853609114452045, "envelope_correlation": 0.999493506312677, "rms_gain_error_db": -0.09483130077941675, "reference_mean_power": 0.0013409420747202037, "output_mean_power": 0.0013119790125067335}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 4.5645799716966823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002745054731183271, "output_power": 0.00026906187265002696, "reference_active": true, "delta_db": -0.08698847355092554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.376005868866601e-06, "output_power": 4.946392458839987e-06, "reference_active": true, "delta_db": -0.3617116278009139}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7970463694996416e-07, "output_power": 2.2616257511503224e-07, "reference_active": true, "delta_db": 0.9986145701074896}], "active_band_count": 3, "active_band_mae_db": 0.48243822381977636}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 9.713079411954574e-05, "log_rms_mae_db": 1.069439185997, "rms_envelope_nrmse": 0.06296221118072791, "envelope_correlation": 0.997088662692182, "rms_gain_error_db": -0.07200636937752183, "reference_mean_power": 0.0005849439303269558, "output_mean_power": 0.0005753254713392417}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 2.471278629444775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014231281864882344, "output_power": 0.00013840096391691222, "reference_active": true, "delta_db": -0.12104905528841692}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.153370236673122e-06, "output_power": 8.587694714966619e-06, "reference_active": true, "delta_db": 0.22539433584717172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4861340139926787e-06, "output_power": 2.323369907209274e-06, "reference_active": true, "delta_db": 1.9406038574296844}], "active_band_count": 3, "active_band_mae_db": 0.7623490828550911}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 9.335285483371196e-05, "log_rms_mae_db": 0.5697361571565741, "rms_envelope_nrmse": 0.030018571393876016, "envelope_correlation": 0.9993522244435352, "rms_gain_error_db": -0.07200639945014556, "reference_mean_power": 0.0005913717973798096, "output_mean_power": 0.0005816476384637016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 2.253861652661314e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012623197394568146, "output_power": 0.00012347111162569726, "reference_active": true, "delta_db": -0.09604015354931295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1401147597866345e-06, "output_power": 3.465455775971244e-06, "reference_active": true, "delta_db": 0.42814840768907925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.001702937868675e-08, "output_power": 1.2048113669381503e-07, "reference_active": true, "delta_db": 1.777366320608349}], "active_band_count": 3, "active_band_mae_db": 0.7671849606155804}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 7.293407135281025e-05, "log_rms_mae_db": 1.316132146256139, "rms_envelope_nrmse": 0.09998905029719755, "envelope_correlation": 0.992926960214774, "rms_gain_error_db": -0.1642316106363849, "reference_mean_power": 0.0003354294563993909, "output_mean_power": 0.000322981790144858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 1.7752008532728783e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001256086055891369, "output_power": 0.00012381090985720623, "reference_active": true, "delta_db": -0.06260479295177698}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.726051831560619e-06, "output_power": 7.797243722098559e-06, "reference_active": true, "delta_db": -0.48876678929750883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2668523428769663e-06, "output_power": 2.5347787184060396e-06, "reference_active": true, "delta_db": 3.0121405327016486}], "active_band_count": 3, "active_band_mae_db": 1.1878373716503114}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 7.108743338455805e-05, "log_rms_mae_db": 0.7966211490477387, "rms_envelope_nrmse": 0.0581518646146437, "envelope_correlation": 0.9976694182186628, "rms_gain_error_db": -0.16423170869975517, "reference_mean_power": 0.00033911524225685186, "output_mean_power": 0.00032653079044428534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 1.626160013872791e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011427350869395253, "output_power": 0.00011394386994548581, "reference_active": true, "delta_db": -0.01254596734860118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179974955820808e-06, "output_power": 3.752519513937595e-06, "reference_active": true, "delta_db": -0.46850720372812227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.39572209062608e-08, "output_power": 1.3213844151172532e-07, "reference_active": true, "delta_db": 1.480990176690036}], "active_band_count": 3, "active_band_mae_db": 0.6540144492555865}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 4.804880216814636e-05, "log_rms_mae_db": 1.4454142914504537, "rms_envelope_nrmse": 0.2273348875917786, "envelope_correlation": 0.9634673748913626, "rms_gain_error_db": -0.25014046020101965, "reference_mean_power": 0.00012431296346075482, "output_mean_power": 0.00011735520963652537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 7.893026272799499e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.194673883780046e-05, "output_power": 3.5113439822461526e-05, "reference_active": true, "delta_db": -0.7722482572210494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.051190096168371e-06, "output_power": 6.182098779810125e-06, "reference_active": true, "delta_db": -0.5712648307584617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.501810525086e-07, "output_power": 2.580137281490514e-06, "reference_active": true, "delta_db": 4.3383644817093945}], "active_band_count": 3, "active_band_mae_db": 1.893959189896302}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 4.532286462232191e-05, "log_rms_mae_db": 1.0674905870168054, "rms_envelope_nrmse": 0.13212232646098965, "envelope_correlation": 0.9874396831970457, "rms_gain_error_db": -0.250141571720559, "reference_mean_power": 0.00012567880182247303, "output_mean_power": 0.00011864457212921419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 7.206643574889899e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.830114378682998e-05, "output_power": 3.2663512330864015e-05, "reference_active": true, "delta_db": -0.6914886058760601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6851867496780266e-06, "output_power": 3.3451897383843853e-06, "reference_active": true, "delta_db": -0.42038745075707606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.995617022655143e-08, "output_power": 1.5245414449745257e-07, "reference_active": true, "delta_db": 4.053053512743573}], "active_band_count": 3, "active_band_mae_db": 1.7216431897922364}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9576000783868702, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "46c2ff730354a17ff91def935b5c9e486af5d1ac36dd2b0c5103abf4030e4ae6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9942657795112906, "standard_deviation": 0.0009786806384307345, "minimum": 0.9927700632200106, "maximum": 0.9953055511042587, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2348255581298073, "standard_deviation": 0.048791938516576774, "minimum": 1.1486152233628508, "maximum": 1.316132146256139, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9279791208525108, "standard_deviation": 0.21478399277742857, "minimum": 0.6527215434019892, "maximum": 1.2880435345248855, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0909407042718961, "standard_deviation": 0.007103665399688178, "minimum": 0.08431224122601678, "maximum": 0.10251352524579578, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9981551783005725, "standard_deviation": 0.0004111143200569845, "minimum": 0.9976694182186628, "maximum": 0.9987385040450348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7574042254535097, "standard_deviation": 0.03782525305334391, "minimum": 0.6929746145197004, "maximum": 0.7966211490477387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5743532451722224, "standard_deviation": 0.2424840535962744, "minimum": 0.2650512361947156, "maximum": 0.9169425919967219, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05365800260069681, "standard_deviation": 0.005753863110359711, "minimum": 0.04493157858636651, "maximum": 0.060995942498979654, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9830060194338897, "standard_deviation": 0.00897066156400282, "minimum": 0.9634673748913626, "maximum": 0.990604826206923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.37891709565064, "standard_deviation": 0.05008093769896224, "minimum": 1.3200440108773612, "maximum": 1.4454142914504537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9436671143626896, "standard_deviation": 0.48756357045446275, "minimum": 0.2887927678836681, "maximum": 1.893959189896302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15566778355844219, "standard_deviation": 0.034997792308705714, "minimum": 0.12185860985232029, "maximum": 0.2273348875917786, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955006173215101, "standard_deviation": 0.0033399273358047806, "minimum": 0.9874396831970457, "maximum": 0.9979079398769942, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9724104268981868, "standard_deviation": 0.04518291644601362, "minimum": 0.9240250732285216, "maximum": 1.0674905870168054, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8515002867479102, "standard_deviation": 0.42100838930901785, "minimum": 0.33409126195443023, "maximum": 1.7216431897922364, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08703868963286013, "standard_deviation": 0.021769011312282108, "minimum": 0.06105620234326897, "maximum": 0.13212232646098965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9976172487492532, "standard_deviation": 0.000357337424215272, "minimum": 0.997088662692182, "maximum": 0.9979797754132346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0743809127039476, "standard_deviation": 0.023620855623822986, "minimum": 1.0326437028910707, "maximum": 1.1103871204277962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3810606533541804, "standard_deviation": 0.22719734111302303, "minimum": 0.14036666154512964, "maximum": 0.7623490828550911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.05725188700358304, "standard_deviation": 0.0040968101453464625, "minimum": 0.05229355958158044, "maximum": 0.06296221118072791, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9993701490603357, "standard_deviation": 0.00012572838013991228, "minimum": 0.999072805799247, "maximum": 0.999480635661453, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5965235816500736, "standard_deviation": 0.025442758353289758, "minimum": 0.5697361571565741, "maximum": 0.6383489896522698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44802375187066124, "standard_deviation": 0.18562559123633757, "minimum": 0.24994732333367056, "maximum": 0.7671849606155804, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.02961360876576287, "standard_deviation": 0.0030766068774677716, "minimum": 0.025914618979125482, "maximum": 0.03490402472468954, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18872651635727483, "standard_deviation": 0.03269496986418944, "minimum": 0.15549190691704182, "maximum": 0.24049428739084186, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.402910035918401, "standard_deviation": 0.00800739155759705, "minimum": 5.386789653910135, "maximum": 5.416443597131978, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40906815454022394, "standard_deviation": 0.01074666556531234, "minimum": 0.3926908876143679, "maximum": 0.4209698828233987, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07061775431250628, "standard_deviation": 0.04036724934105062, "minimum": -0.13148866126973618, "maximum": 0.00541854838630079, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9980283971219329, "standard_deviation": 0.0004697418226344293, "minimum": 0.9971500473816876, "maximum": 0.998535647499916, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8832885073072206, "standard_deviation": 0.04271218101655606, "minimum": 0.8042675213502922, "maximum": 0.9391689943141662, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.39069444593552494, "standard_deviation": 0.15459063715453594, "minimum": 0.1266513608957462, "maximum": 0.6956862034332228, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04814930633147066, "standard_deviation": 0.005699179344640914, "minimum": 0.04207193855138991, "maximum": 0.058106879790551146, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9995158472068856, "standard_deviation": 0.00014969462709312926, "minimum": 0.9991988587864329, "maximum": 0.999683421044742, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4734481622796691, "standard_deviation": 0.04498189675030949, "minimum": 0.41951959501635233, "maximum": 0.5421987636476663, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.32947124607819167, "standard_deviation": 0.17560272429480814, "minimum": 0.04482105782014421, "maximum": 0.529562902878824, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02445134064072666, "standard_deviation": 0.004172461528846409, "minimum": 0.02058869267944883, "maximum": 0.03317942330590424, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0030", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.2231812477111816, "predicted_mos": 3.2231812477111816, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "a4fa00cc781a3942ce5d7714835812c8a53136c674f98f7ec58e52e9d2bd0ade", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.924157800152898, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8786288254871258, "全帯域_10ms_rms_envelope_nrmse": 0.05897115940747921, "全帯域_10ms_envelope_correlation": 0.9970521592412, "全帯域_10ms_modulation_mae_db": 0.6729275479965234, "全帯域_40ms_log_rms_mae_db": 0.47716314996575676, "全帯域_40ms_rms_envelope_nrmse": 0.02596723869414944, "全帯域_40ms_envelope_correlation": 0.999477524063127, "全帯域_40ms_modulation_mae_db": 0.4468536708838399, "300–1000 Hz_10ms_log_rms_mae_db": 1.068823198837208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06365897664985413, "300–1000 Hz_10ms_envelope_correlation": 0.9970413937538707, "300–1000 Hz_10ms_modulation_mae_db": 0.7336749612481855, "300–1000 Hz_40ms_log_rms_mae_db": 0.5593431113289933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0303559025971317, "300–1000 Hz_40ms_envelope_correlation": 0.9993594470412793, "300–1000 Hz_40ms_modulation_mae_db": 0.7586975709302782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.309551959527833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10025185444170362, "1000–3000 Hz_10ms_envelope_correlation": 0.9928349182695848, "1000–3000 Hz_10ms_modulation_mae_db": 1.1488113398976134, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7711731177347557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05688933729877763, "1000–3000 Hz_40ms_envelope_correlation": 0.9977097481217154, "1000–3000 Hz_40ms_modulation_mae_db": 0.5630927470971605, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4021847479878606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23080847330306, "3000–6000 Hz_10ms_envelope_correlation": 0.9624638216548413, "3000–6000 Hz_10ms_modulation_mae_db": 1.9704762125035498, "3000–6000 Hz_40ms_log_rms_mae_db": 1.024247859607332, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13132694740183684, "3000–6000 Hz_40ms_envelope_correlation": 0.9875816442586107, "3000–6000 Hz_40ms_modulation_mae_db": 1.70620514122165, "mr_stft_db": 5.342013237344965, "band_mae_db": 0.18709795830750572, "rms_gain_error_db": -0.08887806974023525, "output_peak": 0.3954568079806314}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9234224224232046, "全帯域_10ms_rms_envelope_nrmse": 0.048304920641871425, "全帯域_10ms_envelope_correlation": 0.9981058265008491, "全帯域_10ms_modulation_mae_db": 0.3741456881015184, "全帯域_40ms_log_rms_mae_db": 0.5426825738359272, "全帯域_40ms_rms_envelope_nrmse": 0.028459259698840347, "全帯域_40ms_envelope_correlation": 0.9994118470672135, "全帯域_40ms_modulation_mae_db": 0.40031581201145167, "300–1000 Hz_10ms_log_rms_mae_db": 1.0790751433692707, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061937982477072856, "300–1000 Hz_10ms_envelope_correlation": 0.9971735197918731, "300–1000 Hz_10ms_modulation_mae_db": 0.22738736044256447, "300–1000 Hz_40ms_log_rms_mae_db": 0.5737707058454297, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035055750565590064, "300–1000 Hz_40ms_envelope_correlation": 0.9990757430595347, "300–1000 Hz_40ms_modulation_mae_db": 0.6194600230914972, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2397991512830004, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0874822317150538, "1000–3000 Hz_10ms_envelope_correlation": 0.9946524132504562, "1000–3000 Hz_10ms_modulation_mae_db": 0.5780367271374107, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7846465613836415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05462448846040794, "1000–3000 Hz_40ms_envelope_correlation": 0.9979696903288752, "1000–3000 Hz_40ms_modulation_mae_db": 0.33330472967988506, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3513822859672076, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14043837264188697, "3000–6000 Hz_10ms_envelope_correlation": 0.9874748548322789, "3000–6000 Hz_10ms_modulation_mae_db": 0.6861116663268071, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9028419972496443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08313165794306375, "3000–6000 Hz_40ms_envelope_correlation": 0.9968287875103836, "3000–6000 Hz_40ms_modulation_mae_db": 0.4768236738408049, "mr_stft_db": 5.350582359578348, "band_mae_db": 0.2609745184068584, "rms_gain_error_db": -0.11388064432798856, "output_peak": 0.4147593703038796}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9310571158539453, "全帯域_10ms_rms_envelope_nrmse": 0.05262397529165055, "全帯域_10ms_envelope_correlation": 0.9977679385422216, "全帯域_10ms_modulation_mae_db": 0.3188355734536383, "全帯域_40ms_log_rms_mae_db": 0.5048546909219491, "全帯域_40ms_rms_envelope_nrmse": 0.033501337984572256, "全帯域_40ms_envelope_correlation": 0.9991689879663801, "全帯域_40ms_modulation_mae_db": 0.5312897397581546, "300–1000 Hz_10ms_log_rms_mae_db": 1.0938394996205505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.052396633925161085, "300–1000 Hz_10ms_envelope_correlation": 0.9979724223981162, "300–1000 Hz_10ms_modulation_mae_db": 0.22468276622876793, "300–1000 Hz_40ms_log_rms_mae_db": 0.6138434255669365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02585418866016807, "300–1000 Hz_40ms_envelope_correlation": 0.9994849605353632, "300–1000 Hz_40ms_modulation_mae_db": 0.23015337916202802, "1000–3000 Hz_10ms_log_rms_mae_db": 1.199341609697869, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0880469176097863, "1000–3000 Hz_10ms_envelope_correlation": 0.9952048560480458, "1000–3000 Hz_10ms_modulation_mae_db": 0.8518672344403319, "1000–3000 Hz_40ms_log_rms_mae_db": 0.702745488892905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.058980968308231914, "1000–3000 Hz_40ms_envelope_correlation": 0.9983184901837907, "1000–3000 Hz_40ms_modulation_mae_db": 0.6524300855929456, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3098308691465015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1581481820794985, "3000–6000 Hz_10ms_envelope_correlation": 0.983690396005713, "3000–6000 Hz_10ms_modulation_mae_db": 1.013561194535019, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8977735658583084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09020575991093538, "3000–6000 Hz_40ms_envelope_correlation": 0.99609788741457, "3000–6000 Hz_40ms_modulation_mae_db": 1.0091123763185414, "mr_stft_db": 5.34642916313337, "band_mae_db": 0.24457127061560366, "rms_gain_error_db": -0.12945018709941403, "output_peak": 0.39126206266287644}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8224841537078695, "全帯域_10ms_rms_envelope_nrmse": 0.04243694393396016, "全帯域_10ms_envelope_correlation": 0.9985260590865148, "全帯域_10ms_modulation_mae_db": 0.11306719166390443, "全帯域_40ms_log_rms_mae_db": 0.42272651607042533, "全帯域_40ms_rms_envelope_nrmse": 0.021561889511903482, "全帯域_40ms_envelope_correlation": 0.9996764823237054, "全帯域_40ms_modulation_mae_db": 0.2726794580292924, "300–1000 Hz_10ms_log_rms_mae_db": 1.092871830175239, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05179029804698739, "300–1000 Hz_10ms_envelope_correlation": 0.9980138646106659, "300–1000 Hz_10ms_modulation_mae_db": 0.27106509395086714, "300–1000 Hz_40ms_log_rms_mae_db": 0.5846432146924198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.024796511464267937, "300–1000 Hz_40ms_envelope_correlation": 0.9995233021869222, "300–1000 Hz_40ms_modulation_mae_db": 0.3507255632109157, "1000–3000 Hz_10ms_log_rms_mae_db": 1.169550063574127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08410340143323772, "1000–3000 Hz_10ms_envelope_correlation": 0.9950343132836208, "1000–3000 Hz_10ms_modulation_mae_db": 0.9088389665541072, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7053666583907974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.044274269577344184, "1000–3000 Hz_40ms_envelope_correlation": 0.9987145139257089, "1000–3000 Hz_40ms_modulation_mae_db": 0.5377673707463674, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3337106867991089, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1435789200809175, "3000–6000 Hz_10ms_envelope_correlation": 0.9881068868517447, "3000–6000 Hz_10ms_modulation_mae_db": 0.4601000626604302, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9058656056562204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10084066292547196, "3000–6000 Hz_40ms_envelope_correlation": 0.9958710814563478, "3000–6000 Hz_40ms_modulation_mae_db": 0.687637781651841, "mr_stft_db": 5.3445626686423395, "band_mae_db": 0.16105437574215672, "rms_gain_error_db": -0.09303081508738083, "output_peak": 0.42279261232319987}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.930632258286001, "全帯域_10ms_rms_envelope_nrmse": 0.044264739065190524, "全帯域_10ms_envelope_correlation": 0.9983261277575811, "全帯域_10ms_modulation_mae_db": 0.3655711227652823, "全帯域_40ms_log_rms_mae_db": 0.4916932126616069, "全帯域_40ms_rms_envelope_nrmse": 0.020659074932375263, "全帯域_40ms_envelope_correlation": 0.9996386422886844, "全帯域_40ms_modulation_mae_db": 0.45802521193679285, "300–1000 Hz_10ms_log_rms_mae_db": 1.113779783701601, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05649811435335791, "300–1000 Hz_10ms_envelope_correlation": 0.9976760593626784, "300–1000 Hz_10ms_modulation_mae_db": 0.41905566712038267, "300–1000 Hz_40ms_log_rms_mae_db": 0.6533835720810025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.028769159062710662, "300–1000 Hz_40ms_envelope_correlation": 0.9993980741619322, "300–1000 Hz_40ms_modulation_mae_db": 0.3757026593194095, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2296145799516673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08826849400964938, "1000–3000 Hz_10ms_envelope_correlation": 0.9945666773639452, "1000–3000 Hz_10ms_modulation_mae_db": 0.8533616588716669, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7179074517873809, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049143889631161314, "1000–3000 Hz_40ms_envelope_correlation": 0.9984456865340907, "1000–3000 Hz_40ms_modulation_mae_db": 0.7913624685025619, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3395898799736885, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12213110878086694, "3000–6000 Hz_10ms_envelope_correlation": 0.9908087045697251, "3000–6000 Hz_10ms_modulation_mae_db": 0.31314308881847047, "3000–6000 Hz_40ms_log_rms_mae_db": 0.899711287941734, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08017852342447424, "3000–6000 Hz_40ms_envelope_correlation": 0.9969774339514984, "3000–6000 Hz_40ms_modulation_mae_db": 0.3610271536455973, "mr_stft_db": 5.32348491424273, "band_mae_db": 0.14749886122940034, "rms_gain_error_db": -0.049274252442776216, "output_peak": 0.3962060032077147}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8666243348583673, "全帯域_10ms_rms_envelope_nrmse": 0.0529208908775564, "全帯域_10ms_envelope_correlation": 0.9976315391688457, "全帯域_10ms_modulation_mae_db": 0.45627911148766803, "全帯域_40ms_log_rms_mae_db": 0.4278468922164444, "全帯域_40ms_rms_envelope_nrmse": 0.02583652185773708, "全帯域_40ms_envelope_correlation": 0.9994666511852609, "全帯域_40ms_modulation_mae_db": 0.24220706587528337, "300–1000 Hz_10ms_log_rms_mae_db": 1.093244847447447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06037435504304875, "300–1000 Hz_10ms_envelope_correlation": 0.9973392840230333, "300–1000 Hz_10ms_modulation_mae_db": 0.21585204478283296, "300–1000 Hz_40ms_log_rms_mae_db": 0.5959552684392981, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.027573376354696838, "300–1000 Hz_40ms_envelope_correlation": 0.9994373540249952, "300–1000 Hz_40ms_modulation_mae_db": 0.32993232490242497, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2284190815903042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1038762306979083, "1000–3000 Hz_10ms_envelope_correlation": 0.9925590845821731, "1000–3000 Hz_10ms_modulation_mae_db": 1.3420081767818626, "1000–3000 Hz_40ms_log_rms_mae_db": 0.740028749621514, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06137377673920628, "1000–3000 Hz_40ms_envelope_correlation": 0.9976473629311279, "1000–3000 Hz_40ms_modulation_mae_db": 0.9107246988374059, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3638291642090494, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17684033663867135, "3000–6000 Hz_10ms_envelope_correlation": 0.9790842845717083, "3000–6000 Hz_10ms_modulation_mae_db": 1.2613742910050212, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9101236597330942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08922531896201782, "3000–6000 Hz_40ms_envelope_correlation": 0.9963846579314211, "3000–6000 Hz_40ms_modulation_mae_db": 0.8226683146242625, "mr_stft_db": 5.350991281363027, "band_mae_db": 0.2299424278391354, "rms_gain_error_db": -0.07896927273053858, "output_peak": 0.40858447249277097}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.897565059017051, "全帯域_10ms_rms_envelope_nrmse": 0.04463194256154105, "全帯域_10ms_envelope_correlation": 0.9982925763489159, "全帯域_10ms_modulation_mae_db": 0.4300810319561184, "全帯域_40ms_log_rms_mae_db": 0.4675147239346803, "全帯域_40ms_rms_envelope_nrmse": 0.0237824500438361, "全帯域_40ms_envelope_correlation": 0.9995019877396153, "全帯域_40ms_modulation_mae_db": 0.26093784422864674, "300–1000 Hz_10ms_log_rms_mae_db": 1.1159009572515102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06041941238513206, "300–1000 Hz_10ms_envelope_correlation": 0.9977108146176803, "300–1000 Hz_10ms_modulation_mae_db": 0.712245517802935, "300–1000 Hz_40ms_log_rms_mae_db": 0.6234632543480456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036666165043603365, "300–1000 Hz_40ms_envelope_correlation": 0.9993605613484852, "300–1000 Hz_40ms_modulation_mae_db": 0.5579011872606562, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2458339085539138, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09182902908451468, "1000–3000 Hz_10ms_envelope_correlation": 0.9940386989894042, "1000–3000 Hz_10ms_modulation_mae_db": 0.7006550942561417, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7763915981265, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05492058409896444, "1000–3000 Hz_40ms_envelope_correlation": 0.9979039308471398, "1000–3000 Hz_40ms_modulation_mae_db": 0.36985905857186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3845595756368088, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1479221143669631, "3000–6000 Hz_10ms_envelope_correlation": 0.9849270210534271, "3000–6000 Hz_10ms_modulation_mae_db": 1.1297424447249158, "3000–6000 Hz_40ms_log_rms_mae_db": 0.926545401735154, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07275829041428425, "3000–6000 Hz_40ms_envelope_correlation": 0.9968241972492676, "3000–6000 Hz_40ms_modulation_mae_db": 1.01335089532707, "mr_stft_db": 5.347796634998133, "band_mae_db": 0.17576476363416402, "rms_gain_error_db": 0.01951587613949251, "output_peak": 0.42494935745889667}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8832868292072212, "全帯域_10ms_rms_envelope_nrmse": 0.04544496646387075, "全帯域_10ms_envelope_correlation": 0.9982336325669698, "全帯域_10ms_modulation_mae_db": 0.3786778097983012, "全帯域_40ms_log_rms_mae_db": 0.45280605147454706, "全帯域_40ms_rms_envelope_nrmse": 0.022053299988645057, "全帯域_40ms_envelope_correlation": 0.999585960079979, "全帯域_40ms_modulation_mae_db": 0.08704234857347072, "300–1000 Hz_10ms_log_rms_mae_db": 1.1180998452516393, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05528374279318362, "300–1000 Hz_10ms_envelope_correlation": 0.9977658371802827, "300–1000 Hz_10ms_modulation_mae_db": 0.464019817296026, "300–1000 Hz_40ms_log_rms_mae_db": 0.6351460964138855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.028947681521909747, "300–1000 Hz_40ms_envelope_correlation": 0.999369515087418, "300–1000 Hz_40ms_modulation_mae_db": 0.393283806443079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2465415542796656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0944365160763842, "1000–3000 Hz_10ms_envelope_correlation": 0.9937331661532642, "1000–3000 Hz_10ms_modulation_mae_db": 0.8108885164645064, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7454812745667672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055791355081386905, "1000–3000 Hz_40ms_envelope_correlation": 0.997891163304072, "1000–3000 Hz_40ms_modulation_mae_db": 0.27210724774707395, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3518892176523705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12897825927657508, "3000–6000 Hz_10ms_envelope_correlation": 0.9884925902781098, "3000–6000 Hz_10ms_modulation_mae_db": 1.0804191040677178, "3000–6000 Hz_40ms_log_rms_mae_db": 0.93104348922472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.060880260022825705, "3000–6000 Hz_40ms_envelope_correlation": 0.9976673362428887, "3000–6000 Hz_40ms_modulation_mae_db": 1.0868247874558505, "mr_stft_db": 5.349501543828775, "band_mae_db": 0.15247830334793097, "rms_gain_error_db": -0.05161107910074558, "output_peak": 0.4127535951245676}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44240, "output_samples": 44240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.06042832446136, "spectral_convergence": 0.2191964400314609, "spectral_convergence_status": "ok", "amplitude_floor": 7.692579600611731e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.228397029120849, "spectral_convergence": 0.22527382151842662, "spectral_convergence_status": "ok", "amplitude_floor": 5.999960941709396e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.737214358452684, "spectral_convergence": 0.2196223086065476, "spectral_convergence_status": "ok", "amplitude_floor": 5.417405012540354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 6.282816067608343e-05, "output_power": 6.444665253462204e-05, "delta_db": 0.11046018576812429, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019406375287345015, "output_power": 0.000197297940132921, "delta_db": 0.07178125406777253, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005567857807558596, "output_power": 0.000541092534211872, "delta_db": -0.1241659394899762, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00030647737825017173, "output_power": 0.0002992075678254075, "delta_db": -0.10425849938420072, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00015415279416012865, "output_power": 0.0001524935593786371, "delta_db": -0.04699899634294363, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.966178555848869e-05, "output_power": 4.2611815976397984e-05, "delta_db": -0.6649228747920168, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18709795830750572, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08887806974023525, "pitch": null, "output_rms_dbfs": -28.87009776553811, "output_peak": 0.3954568079806314, "reference_rms_dbfs": -28.781219695797876, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.342013237344965}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 0.00013943874660832744, "log_rms_mae_db": 0.8786288254871258, "rms_envelope_nrmse": 0.05897115940747921, "envelope_correlation": 0.9970521592412, "rms_gain_error_db": -0.08887838085184441, "reference_mean_power": 0.0013263679977567136, "output_mean_power": 0.0012994997392955308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 5.047819293813421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030810693936871264, "output_power": 0.00030119334193905465, "reference_active": true, "delta_db": -0.09856112574444124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2023354078518551e-05, "output_power": 1.1654050568979474e-05, "reference_active": true, "delta_db": -0.13548738695778134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2193973896229425e-06, "output_power": 4.855656997731785e-06, "reference_active": true, "delta_db": 1.784734131287348}], "active_band_count": 3, "active_band_mae_db": 0.6729275479965234}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 0.0001327212418155818, "log_rms_mae_db": 0.47716314996575676, "rms_envelope_nrmse": 0.02596723869414944, "envelope_correlation": 0.999477524063127, "rms_gain_error_db": -0.08887844446337016, "reference_mean_power": 0.0013409420747202037, "output_mean_power": 0.001313778569697959}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 4.5645799716966823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002745054731183271, "output_power": 0.00026990816312519246, "reference_active": true, "delta_db": -0.07334988286442357}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.376005868866601e-06, "output_power": 4.965774696135929e-06, "reference_active": true, "delta_db": -0.34472723013802764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7970463694996416e-07, "output_power": 2.2223255194588952e-07, "reference_active": true, "delta_db": 0.9224838996490685}], "active_band_count": 3, "active_band_mae_db": 0.4468536708838399}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 9.713079411954574e-05, "log_rms_mae_db": 1.068823198837208, "rms_envelope_nrmse": 0.06365897664985413, "envelope_correlation": 0.9970413937538707, "rms_gain_error_db": -0.0804344102417054, "reference_mean_power": 0.0005849439303269558, "output_mean_power": 0.0005742100611963323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 2.471278629444775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014231281864882344, "output_power": 0.0001378785854913804, "reference_active": true, "delta_db": -0.13747201084246116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.153370236673122e-06, "output_power": 8.577567625737933e-06, "reference_active": true, "delta_db": 0.22026987064209325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4861340139926787e-06, "output_power": 2.271884635472745e-06, "reference_active": true, "delta_db": 1.843283002260002}], "active_band_count": 3, "active_band_mae_db": 0.7336749612481855}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 9.335285483371196e-05, "log_rms_mae_db": 0.5593431113289933, "rms_envelope_nrmse": 0.0303559025971317, "envelope_correlation": 0.9993594470412793, "rms_gain_error_db": -0.08043446798991917, "reference_mean_power": 0.0005913717973798096, "output_mean_power": 0.0005805199675431304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 2.253861652661314e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012623197394568146, "output_power": 0.0001229543035500516, "reference_active": true, "delta_db": -0.11425639349034448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1401147597866345e-06, "output_power": 3.4913887670739487e-06, "reference_active": true, "delta_db": 0.4605269006479212}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.001702937868675e-08, "output_power": 1.183895489854978e-07, "reference_active": true, "delta_db": 1.7013094186525688}], "active_band_count": 3, "active_band_mae_db": 0.7586975709302782}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 7.293407135281025e-05, "log_rms_mae_db": 1.309551959527833, "rms_envelope_nrmse": 0.10025185444170362, "envelope_correlation": 0.9928349182695848, "rms_gain_error_db": -0.13879706201058273, "reference_mean_power": 0.0003354294563993909, "output_mean_power": 0.00032487888964202196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 1.7752008532728783e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001256086055891369, "output_power": 0.0001243689780325395, "reference_active": true, "delta_db": -0.0430732878298475}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.726051831560619e-06, "output_power": 8.115555040950671e-06, "reference_active": true, "delta_db": -0.31499560760158296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2668523428769663e-06, "output_power": 2.579660248122921e-06, "reference_active": true, "delta_db": 3.0883651242614096}], "active_band_count": 3, "active_band_mae_db": 1.1488113398976134}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 7.108743338455805e-05, "log_rms_mae_db": 0.7711731177347557, "rms_envelope_nrmse": 0.05688933729877763, "envelope_correlation": 0.9977097481217154, "rms_gain_error_db": -0.13879703127905751, "reference_mean_power": 0.00033911524225685186, "output_mean_power": 0.0003284487454599279}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 1.626160013872791e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011427350869395253, "output_power": 0.00011455140292950433, "reference_active": true, "delta_db": 0.01054850116881928}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179974955820808e-06, "output_power": 3.864147738595541e-06, "reference_active": true, "delta_db": -0.34119957208881513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.39572209062608e-08, "output_power": 1.2784483277018542e-07, "reference_active": true, "delta_db": 1.3375301680338472}], "active_band_count": 3, "active_band_mae_db": 0.5630927470971605}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 4.804880216814636e-05, "log_rms_mae_db": 1.4021847479878606, "rms_envelope_nrmse": 0.23080847330306, "envelope_correlation": 0.9624638216548413, "rms_gain_error_db": -0.21961294603300807, "reference_mean_power": 0.00012431296346075482, "output_mean_power": 0.00011818303127366782}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 7.893026272799499e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.194673883780046e-05, "output_power": 3.55204542189938e-05, "reference_active": true, "delta_db": -0.7221969142093511}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.051190096168371e-06, "output_power": 6.250551755235303e-06, "reference_active": true, "delta_db": -0.523440676403864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.501810525086e-07, "output_power": 2.7821815778902445e-06, "reference_active": true, "delta_db": 4.665791046897435}], "active_band_count": 3, "active_band_mae_db": 1.9704762125035498}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 4.532286462232191e-05, "log_rms_mae_db": 1.024247859607332, "rms_envelope_nrmse": 0.13132694740183684, "envelope_correlation": 0.9875816442586107, "rms_gain_error_db": -0.21961370817239037, "reference_mean_power": 0.00012567880182247303, "output_mean_power": 0.00011948149851929639}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 7.206643574889899e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.830114378682998e-05, "output_power": 3.315821701149918e-05, "reference_active": true, "delta_db": -0.626205739362145}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6851867496780266e-06, "output_power": 3.4096821600396155e-06, "reference_active": true, "delta_db": -0.33745603661097645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.995617022655143e-08, "output_power": 1.560735284121935e-07, "reference_active": true, "delta_db": 4.154953647691828}], "active_band_count": 3, "active_band_mae_db": 1.70620514122165}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.959538342474737, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "51fd5ca73c0c4cdcb8ac3f693e8aac919c724f22b158b34a57345f9475b4d358", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9940780159925617, "standard_deviation": 0.0009802264302573539, "minimum": 0.9925590845821731, "maximum": 0.9952048560480458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2335814885572978, "standard_deviation": 0.040431664373040216, "minimum": 1.169550063574127, "maximum": 1.309551959527833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8993084643004552, "standard_deviation": 0.24295328997283341, "minimum": 0.5780367271374107, "maximum": 1.3420081767818626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09228683438352975, "standard_deviation": 0.0068323087897945054, "minimum": 0.08410340143323772, "maximum": 0.1038762306979083, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9980750732720651, "standard_deviation": 0.0003772280674889662, "minimum": 0.9976473629311279, "maximum": 0.9987145139257089, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7429676125630327, "standard_deviation": 0.03234736943397639, "minimum": 0.702745488892905, "maximum": 0.7846465613836415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5538310508469075, "standard_deviation": 0.2254220468176686, "minimum": 0.27210724774707395, "maximum": 0.9107246988374059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05449983364943507, "standard_deviation": 0.0054530500431683995, "minimum": 0.044274269577344184, "maximum": 0.06137377673920628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9831310699771936, "standard_deviation": 0.009089783154101678, "minimum": 0.9624638216548413, "maximum": 0.9908087045697251, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3546220534215745, "standard_deviation": 0.029107441870461753, "minimum": 1.3098308691465015, "maximum": 1.4021847479878606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9893660080802414, "standard_deviation": 0.5202348960961388, "minimum": 0.31314308881847047, "maximum": 1.9704762125035498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15610572089605493, "standard_deviation": 0.034587628987747435, "minimum": 0.12213110878086694, "maximum": 0.23080847330306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955291282518735, "standard_deviation": 0.00325950681071333, "minimum": 0.9875816442586107, "maximum": 0.9976673362428887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9247691083757759, "standard_deviation": 0.041999708304759936, "minimum": 0.8977735658583084, "maximum": 1.024247859607332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8954562655107021, "standard_deviation": 0.4192010830762932, "minimum": 0.3610271536455973, "maximum": 1.70620514122165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08856842762561373, "standard_deviation": 0.02103926654369222, "minimum": 0.060880260022825705, "maximum": 0.13132694740183684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.997586649467275, "standard_deviation": 0.0003617462544142377, "minimum": 0.9970413937538707, "maximum": 0.9980138646106659, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0969543882068082, "standard_deviation": 0.017889975168460193, "minimum": 1.068823198837208, "maximum": 1.1180998452516393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.40849790360907023, "standard_deviation": 0.21507012772673959, "minimum": 0.21585204478283296, "maximum": 0.7336749612481855, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.057794939459224724, "standard_deviation": 0.0044445514910010885, "minimum": 0.05179029804698739, "maximum": 0.06365897664985413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9993761196807412, "standard_deviation": 0.00013550458070841568, "minimum": 0.9990757430595347, "maximum": 0.9995233021869222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6049435810895014, "standard_deviation": 0.032126659650330416, "minimum": 0.5593431113289933, "maximum": 0.6533835720810025, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45198206429003607, "standard_deviation": 0.17605321672992638, "minimum": 0.23015337916202802, "maximum": 0.7586975709302782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.029752341908759798, "standard_deviation": 0.004181750335899529, "minimum": 0.024796511464267937, "maximum": 0.036666165043603365, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1949228098903444, "standard_deviation": 0.04420741438159083, "minimum": 0.14749886122940034, "maximum": 0.2609745184068584, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.3444202253914606, "standard_deviation": 0.008996091043914575, "minimum": 5.32348491424273, "maximum": 5.350991281363027, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40834553519431716, "standard_deviation": 0.012819420954889377, "minimum": 0.39126206266287644, "maximum": 0.42494935745889667, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07319730554869831, "standard_deviation": 0.04648517814361866, "minimum": -0.12945018709941403, "maximum": 0.01951587613949251, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9979919824016372, "standard_deviation": 0.00048136496023600634, "minimum": 0.9970521592412, "maximum": 0.9985260590865148, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8917126248550982, "standard_deviation": 0.03735289706937711, "minimum": 0.8224841537078695, "maximum": 0.9310571158539453, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3886981346528693, "standard_deviation": 0.15527226111796869, "minimum": 0.11306719166390443, "maximum": 0.6729275479965234, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04869994228039001, "standard_deviation": 0.005668308523540223, "minimum": 0.04243694393396016, "maximum": 0.05897115940747921, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9994910103392457, "standard_deviation": 0.00015859719254853158, "minimum": 0.9991689879663801, "maximum": 0.9996764823237054, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47341097638516716, "standard_deviation": 0.04007120336558141, "minimum": 0.42272651607042533, "maximum": 0.5426825738359272, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33741889391211655, "standard_deviation": 0.14631572208594237, "minimum": 0.08704234857347072, "maximum": 0.5312897397581546, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.025227634089007377, "standard_deviation": 0.004247515399750511, "minimum": 0.020659074932375263, "maximum": 0.033501337984572256, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0030", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.9396677017211914, "predicted_mos": 2.9396677017211914, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "613c11a9b6dc38ba16f4a24b3462660e518cc956a5f1e276a893b9b348ca9c8a", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.874252500012517, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8686818760239738, "全帯域_10ms_rms_envelope_nrmse": 0.053625509343245836, "全帯域_10ms_envelope_correlation": 0.9975482269717438, "全帯域_10ms_modulation_mae_db": 0.581002051527919, "全帯域_40ms_log_rms_mae_db": 0.48071879802778567, "全帯域_40ms_rms_envelope_nrmse": 0.030643239159186764, "全帯域_40ms_envelope_correlation": 0.9991742377931452, "全帯域_40ms_modulation_mae_db": 0.49908284111836004, "300–1000 Hz_10ms_log_rms_mae_db": 1.1781998397899096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.063100333875971, "300–1000 Hz_10ms_envelope_correlation": 0.997177556592754, "300–1000 Hz_10ms_modulation_mae_db": 0.9393475154944504, "300–1000 Hz_40ms_log_rms_mae_db": 0.6597129756113318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03649719166850579, "300–1000 Hz_40ms_envelope_correlation": 0.9990836671646475, "300–1000 Hz_40ms_modulation_mae_db": 0.7978949827556123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3096466518429957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09559573545145361, "1000–3000 Hz_10ms_envelope_correlation": 0.9934303384407072, "1000–3000 Hz_10ms_modulation_mae_db": 1.1198632697154378, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8416016200821941, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06019839017575239, "1000–3000 Hz_40ms_envelope_correlation": 0.9972755863524505, "1000–3000 Hz_40ms_modulation_mae_db": 0.5888096088491502, "3000–6000 Hz_10ms_log_rms_mae_db": 1.772388001843872, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15192995797700537, "3000–6000 Hz_10ms_envelope_correlation": 0.9841274456435944, "3000–6000 Hz_10ms_modulation_mae_db": 0.7560662111770421, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2757392152148832, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.080086847334503, "3000–6000 Hz_40ms_envelope_correlation": 0.9961614434145329, "3000–6000 Hz_40ms_modulation_mae_db": 0.39771529643539444, "mr_stft_db": 5.6958237719418925, "band_mae_db": 0.13895565391209216, "rms_gain_error_db": 0.033685413921041504, "output_peak": 0.4195642748759858}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7384311756577949, "全帯域_10ms_rms_envelope_nrmse": 0.0392859727207127, "全帯域_10ms_envelope_correlation": 0.9987541372026113, "全帯域_10ms_modulation_mae_db": 0.17961692121569942, "全帯域_40ms_log_rms_mae_db": 0.40655140541647194, "全帯域_40ms_rms_envelope_nrmse": 0.020467190812392816, "全帯域_40ms_envelope_correlation": 0.9997219919253897, "全帯域_40ms_modulation_mae_db": 0.15994154734288588, "300–1000 Hz_10ms_log_rms_mae_db": 1.0906186511791196, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05146259705628811, "300–1000 Hz_10ms_envelope_correlation": 0.9980576702592676, "300–1000 Hz_10ms_modulation_mae_db": 0.5152743426006295, "300–1000 Hz_40ms_log_rms_mae_db": 0.5750748002419337, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0220132542560757, "300–1000 Hz_40ms_envelope_correlation": 0.9996379595045272, "300–1000 Hz_40ms_modulation_mae_db": 0.39124764106227267, "1000–3000 Hz_10ms_log_rms_mae_db": 1.407265950054873, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08187828299187276, "1000–3000 Hz_10ms_envelope_correlation": 0.9954048665113758, "1000–3000 Hz_10ms_modulation_mae_db": 0.45317800558623283, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9033297664753337, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0439550043391544, "1000–3000 Hz_40ms_envelope_correlation": 0.9988646472180445, "1000–3000 Hz_40ms_modulation_mae_db": 0.3200697717405425, "3000–6000 Hz_10ms_log_rms_mae_db": 1.816838379300829, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15748549341716578, "3000–6000 Hz_10ms_envelope_correlation": 0.9858453098563262, "3000–6000 Hz_10ms_modulation_mae_db": 0.980580670913081, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3208322815593003, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10787638498354045, "3000–6000 Hz_40ms_envelope_correlation": 0.9957555452319145, "3000–6000 Hz_40ms_modulation_mae_db": 1.132785264453153, "mr_stft_db": 5.70015400638529, "band_mae_db": 0.24802509866081554, "rms_gain_error_db": -0.09332214469836793, "output_peak": 0.4124104141745237}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8555829126852085, "全帯域_10ms_rms_envelope_nrmse": 0.04330124704124233, "全帯域_10ms_envelope_correlation": 0.9984614878740459, "全帯域_10ms_modulation_mae_db": 0.14091714453113577, "全帯域_40ms_log_rms_mae_db": 0.45396969776913904, "全帯域_40ms_rms_envelope_nrmse": 0.02334532721590351, "全帯域_40ms_envelope_correlation": 0.9996014068363954, "全帯域_40ms_modulation_mae_db": 0.22557115600659192, "300–1000 Hz_10ms_log_rms_mae_db": 1.1523730760715112, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.059772782490652356, "300–1000 Hz_10ms_envelope_correlation": 0.9973739916949567, "300–1000 Hz_10ms_modulation_mae_db": 0.35175128295692826, "300–1000 Hz_40ms_log_rms_mae_db": 0.6308249761345432, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030144589001689554, "300–1000 Hz_40ms_envelope_correlation": 0.9993117941105448, "300–1000 Hz_40ms_modulation_mae_db": 0.28463636243882856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3671659354812007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0963766127886999, "1000–3000 Hz_10ms_envelope_correlation": 0.9934389305916902, "1000–3000 Hz_10ms_modulation_mae_db": 0.6396666609717765, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8124083964489324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05724603651316722, "1000–3000 Hz_40ms_envelope_correlation": 0.9977415190896574, "1000–3000 Hz_40ms_modulation_mae_db": 0.07831516622230465, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8278374202609897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17057937459490513, "3000–6000 Hz_10ms_envelope_correlation": 0.9810627031277533, "3000–6000 Hz_10ms_modulation_mae_db": 0.26337421448079984, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2825114768611867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10277191158632482, "3000–6000 Hz_40ms_envelope_correlation": 0.9948138983841587, "3000–6000 Hz_40ms_modulation_mae_db": 0.3945856452722323, "mr_stft_db": 5.684170575609776, "band_mae_db": 0.16923635321495442, "rms_gain_error_db": -0.08827515390288013, "output_peak": 0.3796380002184969}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.831386925240896, "全帯域_10ms_rms_envelope_nrmse": 0.0425563068249484, "全帯域_10ms_envelope_correlation": 0.9984327387939934, "全帯域_10ms_modulation_mae_db": 0.4563389119215897, "全帯域_40ms_log_rms_mae_db": 0.45778943333015937, "全帯域_40ms_rms_envelope_nrmse": 0.024009941406595035, "全帯域_40ms_envelope_correlation": 0.9994743072025584, "全帯域_40ms_modulation_mae_db": 0.26598536832091374, "300–1000 Hz_10ms_log_rms_mae_db": 1.133884853454964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.04925272027167014, "300–1000 Hz_10ms_envelope_correlation": 0.9982116298126709, "300–1000 Hz_10ms_modulation_mae_db": 0.5492845158495704, "300–1000 Hz_40ms_log_rms_mae_db": 0.6162285880239937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.026934544454657174, "300–1000 Hz_40ms_envelope_correlation": 0.9994415528036793, "300–1000 Hz_40ms_modulation_mae_db": 0.43926342676891433, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3309910478983387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09040833156255211, "1000–3000 Hz_10ms_envelope_correlation": 0.9941439712508912, "1000–3000 Hz_10ms_modulation_mae_db": 0.5306864221369236, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7969086281062263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05640324178361764, "1000–3000 Hz_40ms_envelope_correlation": 0.997640583927421, "1000–3000 Hz_40ms_modulation_mae_db": 0.12598428519054491, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6849549767278609, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1388051518009172, "3000–6000 Hz_10ms_envelope_correlation": 0.9865617156917708, "3000–6000 Hz_10ms_modulation_mae_db": 0.9555237638304056, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1242490601578399, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06171605290673139, "3000–6000 Hz_40ms_envelope_correlation": 0.9976347735436976, "3000–6000 Hz_40ms_modulation_mae_db": 0.3936632853652075, "mr_stft_db": 5.673401901035991, "band_mae_db": 0.09366036510821314, "rms_gain_error_db": -0.006828541013982865, "output_peak": 0.4413890034522534}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8303784548567996, "全帯域_10ms_rms_envelope_nrmse": 0.044307421946623404, "全帯域_10ms_envelope_correlation": 0.9984731989653605, "全帯域_10ms_modulation_mae_db": 0.5611846221734951, "全帯域_40ms_log_rms_mae_db": 0.445261780691877, "全帯域_40ms_rms_envelope_nrmse": 0.025929054001281905, "全帯域_40ms_envelope_correlation": 0.9995812932920788, "全帯域_40ms_modulation_mae_db": 0.43878415993223946, "300–1000 Hz_10ms_log_rms_mae_db": 1.1365819219481164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.04799961237608585, "300–1000 Hz_10ms_envelope_correlation": 0.9983331614113569, "300–1000 Hz_10ms_modulation_mae_db": 0.6877266094601077, "300–1000 Hz_40ms_log_rms_mae_db": 0.5976876216652235, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.024833923582001385, "300–1000 Hz_40ms_envelope_correlation": 0.9995603553308888, "300–1000 Hz_40ms_modulation_mae_db": 0.16113034814017124, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3498614479213522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09736953428308114, "1000–3000 Hz_10ms_envelope_correlation": 0.9937143280585117, "1000–3000 Hz_10ms_modulation_mae_db": 0.38032532600225794, "1000–3000 Hz_40ms_log_rms_mae_db": 0.834484340082237, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06038683181043528, "1000–3000 Hz_40ms_envelope_correlation": 0.9979616615159754, "1000–3000 Hz_40ms_modulation_mae_db": 0.4430119121318546, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8210194347057298, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18361203099031154, "3000–6000 Hz_10ms_envelope_correlation": 0.9817798242004705, "3000–6000 Hz_10ms_modulation_mae_db": 0.6793208317927867, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3161575316637302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14160466486263004, "3000–6000 Hz_40ms_envelope_correlation": 0.9922466743010748, "3000–6000 Hz_40ms_modulation_mae_db": 0.7608380627538395, "mr_stft_db": 5.67939505351961, "band_mae_db": 0.28358574919048257, "rms_gain_error_db": -0.12900200070140513, "output_peak": 0.41311606865778194}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8613010923349786, "全帯域_10ms_rms_envelope_nrmse": 0.04729276028689783, "全帯域_10ms_envelope_correlation": 0.9982782279079389, "全帯域_10ms_modulation_mae_db": 0.44585837851159343, "全帯域_40ms_log_rms_mae_db": 0.45348497978275887, "全帯域_40ms_rms_envelope_nrmse": 0.027316156947995003, "全帯域_40ms_envelope_correlation": 0.9995609517999609, "全帯域_40ms_modulation_mae_db": 0.25128628733147756, "300–1000 Hz_10ms_log_rms_mae_db": 1.1682141912470807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05301427545015854, "300–1000 Hz_10ms_envelope_correlation": 0.9979391925503514, "300–1000 Hz_10ms_modulation_mae_db": 0.1876297190318791, "300–1000 Hz_40ms_log_rms_mae_db": 0.6639646128229664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02712278283080442, "300–1000 Hz_40ms_envelope_correlation": 0.9994578461713711, "300–1000 Hz_40ms_modulation_mae_db": 0.187802037285175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3933336505713598, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09937770655407373, "1000–3000 Hz_10ms_envelope_correlation": 0.9936848771567334, "1000–3000 Hz_10ms_modulation_mae_db": 0.9911387384652647, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8479451224249961, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05828342228674621, "1000–3000 Hz_40ms_envelope_correlation": 0.9984633194324052, "1000–3000 Hz_40ms_modulation_mae_db": 0.8510462178587667, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7143534443132038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14336185553288988, "3000–6000 Hz_10ms_envelope_correlation": 0.9865427219614282, "3000–6000 Hz_10ms_modulation_mae_db": 0.8595538474412566, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2168958666164682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1018393868914778, "3000–6000 Hz_40ms_envelope_correlation": 0.993785008822288, "3000–6000 Hz_40ms_modulation_mae_db": 0.9510948886889293, "mr_stft_db": 5.702960048582205, "band_mae_db": 0.2059349059546893, "rms_gain_error_db": -0.14118589476793642, "output_peak": 0.3925617699699156}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8640284648317984, "全帯域_10ms_rms_envelope_nrmse": 0.04534189613889026, "全帯域_10ms_envelope_correlation": 0.9982538035331944, "全帯域_10ms_modulation_mae_db": 0.4247637307578456, "全帯域_40ms_log_rms_mae_db": 0.475140042942339, "全帯域_40ms_rms_envelope_nrmse": 0.02414576623382963, "全帯域_40ms_envelope_correlation": 0.9995092291141714, "全帯域_40ms_modulation_mae_db": 0.11539175531013128, "300–1000 Hz_10ms_log_rms_mae_db": 1.1749980603755779, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0667056557570777, "300–1000 Hz_10ms_envelope_correlation": 0.9968630976927378, "300–1000 Hz_10ms_modulation_mae_db": 0.6834409400516673, "300–1000 Hz_40ms_log_rms_mae_db": 0.6193375544967581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.035242288350351036, "300–1000 Hz_40ms_envelope_correlation": 0.9991869279874528, "300–1000 Hz_40ms_modulation_mae_db": 0.5330446425664047, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4809406641567975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09801467160221919, "1000–3000 Hz_10ms_envelope_correlation": 0.9933446494387568, "1000–3000 Hz_10ms_modulation_mae_db": 1.0768372552385825, "1000–3000 Hz_40ms_log_rms_mae_db": 0.904843391833539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0635562808696463, "1000–3000 Hz_40ms_envelope_correlation": 0.9973296934924809, "1000–3000 Hz_40ms_modulation_mae_db": 0.6253344994134148, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7450219853154765, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1535747850986946, "3000–6000 Hz_10ms_envelope_correlation": 0.9854525864281589, "3000–6000 Hz_10ms_modulation_mae_db": 1.0625338220494063, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2317619021788937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09709205872636481, "3000–6000 Hz_40ms_envelope_correlation": 0.9959935035964002, "3000–6000 Hz_40ms_modulation_mae_db": 0.7842474528463423, "mr_stft_db": 5.668166628916179, "band_mae_db": 0.1895381239640803, "rms_gain_error_db": -0.059649231793522425, "output_peak": 0.3903258071347808}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8319087156304353, "全帯域_10ms_rms_envelope_nrmse": 0.04660053043572705, "全帯域_10ms_envelope_correlation": 0.9982284122773409, "全帯域_10ms_modulation_mae_db": 0.4464764772262398, "全帯域_40ms_log_rms_mae_db": 0.4262324086793152, "全帯域_40ms_rms_envelope_nrmse": 0.02629189775819491, "全帯域_40ms_envelope_correlation": 0.999496687582043, "全帯域_40ms_modulation_mae_db": 0.6489788128783702, "300–1000 Hz_10ms_log_rms_mae_db": 1.0508314704291668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05350135572625732, "300–1000 Hz_10ms_envelope_correlation": 0.9979326489924409, "300–1000 Hz_10ms_modulation_mae_db": 0.36619494827571913, "300–1000 Hz_40ms_log_rms_mae_db": 0.5826345775249366, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.026105510867747334, "300–1000 Hz_40ms_envelope_correlation": 0.9995171049338172, "300–1000 Hz_40ms_modulation_mae_db": 0.6877532645920655, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4463946465749327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10344544627881523, "1000–3000 Hz_10ms_envelope_correlation": 0.9927426130475808, "1000–3000 Hz_10ms_modulation_mae_db": 1.6698966575869238, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8674783454846654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05809766282841127, "1000–3000 Hz_40ms_envelope_correlation": 0.9980940382483522, "1000–3000 Hz_40ms_modulation_mae_db": 1.121141338912808, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7173727792500924, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15745056210151412, "3000–6000 Hz_10ms_envelope_correlation": 0.985286218757908, "3000–6000 Hz_10ms_modulation_mae_db": 0.5565973321782867, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2234680160115463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10385425429550765, "3000–6000 Hz_40ms_envelope_correlation": 0.9957633221095976, "3000–6000 Hz_40ms_modulation_mae_db": 0.6214930299557906, "mr_stft_db": 5.685512494442027, "band_mae_db": 0.2537411441593817, "rms_gain_error_db": -0.09870584143775894, "output_peak": 0.39169430272470623}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44240, "output_samples": 44240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.386350436463629, "spectral_convergence": 0.20951375244736953, "spectral_convergence_status": "ok", "amplitude_floor": 7.692579600611731e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.642390010311616, "spectral_convergence": 0.2234612659720484, "spectral_convergence_status": "ok", "amplitude_floor": 5.999960941709396e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.058730869050432, "spectral_convergence": 0.2266986488884139, "spectral_convergence_status": "ok", "amplitude_floor": 5.417405012540354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 6.282816067608343e-05, "output_power": 6.345652213181398e-05, "delta_db": 0.043219202668743587, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019406375287345015, "output_power": 0.00019816464945858592, "delta_db": 0.09081757710410783, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005567857807558596, "output_power": 0.0005754039260677211, "delta_db": 0.1428468512760348, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00030647737825017173, "output_power": 0.00030264013737622467, "delta_db": -0.05471898411931839, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00015415279416012865, "output_power": 0.00014877499503142483, "delta_db": -0.1542145668496255, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.966178555848869e-05, "output_power": 4.58385272966642e-05, "delta_db": -0.3479167414547227, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13895565391209216, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.033685413921041504, "pitch": null, "output_rms_dbfs": -28.74753428187683, "output_peak": 0.4195642748759858, "reference_rms_dbfs": -28.781219695797876, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6958237719418925}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 0.00013943874660832744, "log_rms_mae_db": 0.8686818760239738, "rms_envelope_nrmse": 0.053625509343245836, "envelope_correlation": 0.9975482269717438, "rms_gain_error_db": 0.03368547883319027, "reference_mean_power": 0.0013263679977567136, "output_mean_power": 0.0013366957975689691}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 5.047819293813421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030810693936871264, "output_power": 0.0003190912078398227, "reference_active": true, "delta_db": 0.15213358048595332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2023354078518551e-05, "output_power": 1.205128495362184e-05, "reference_active": true, "delta_db": 0.010077185671096106}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2193973896229425e-06, "output_power": 4.632912803935865e-06, "reference_active": true, "delta_db": 1.5807953884267074}], "active_band_count": 3, "active_band_mae_db": 0.581002051527919}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 0.0001327212418155818, "log_rms_mae_db": 0.48071879802778567, "rms_envelope_nrmse": 0.030643239159186764, "envelope_correlation": 0.9991742377931452, "rms_gain_error_db": 0.03368661743032756, "reference_mean_power": 0.0013409420747202037, "output_mean_power": 0.0013513837102661747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 4.5645799716966823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002745054731183271, "output_power": 0.0002858980880620824, "reference_active": true, "delta_db": 0.17660243151341815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.376005868866601e-06, "output_power": 6.0877840118008266e-06, "reference_active": true, "delta_db": 0.5399950156689483}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7970463694996416e-07, "output_power": 2.1509206599030304e-07, "reference_active": true, "delta_db": 0.7806510761727138}], "active_band_count": 3, "active_band_mae_db": 0.49908284111836004}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 9.713079411954574e-05, "log_rms_mae_db": 1.1781998397899096, "rms_envelope_nrmse": 0.063100333875971, "envelope_correlation": 0.997177556592754, "rms_gain_error_db": 0.10001029492059396, "reference_mean_power": 0.0005849439303269558, "output_mean_power": 0.0005985704436861421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 2.471278629444775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014231281864882344, "output_power": 0.00014576130037002353, "reference_active": true, "delta_db": 0.10398213709523214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.153370236673122e-06, "output_power": 7.048073469658554e-06, "reference_active": true, "delta_db": -0.6326674111847174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4861340139926787e-06, "output_power": 2.3999228131384972e-06, "reference_active": true, "delta_db": 2.0813929982034014}], "active_band_count": 3, "active_band_mae_db": 0.9393475154944504}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 9.335285483371196e-05, "log_rms_mae_db": 0.6597129756113318, "rms_envelope_nrmse": 0.03649719166850579, "envelope_correlation": 0.9990836671646475, "rms_gain_error_db": 0.10001027548020325, "reference_mean_power": 0.0005913717973798096, "output_mean_power": 0.0006051480478799592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 2.253861652661314e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012623197394568146, "output_power": 0.0001310141817177147, "reference_active": true, "delta_db": 0.16148935172445128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1401147597866345e-06, "output_power": 2.8088434487219886e-06, "reference_active": true, "delta_db": -0.4841798583570255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.001702937868675e-08, "output_power": 1.196696437160727e-07, "reference_active": true, "delta_db": 1.74801573818536}], "active_band_count": 3, "active_band_mae_db": 0.7978949827556123}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 7.293407135281025e-05, "log_rms_mae_db": 1.3096466518429957, "rms_envelope_nrmse": 0.09559573545145361, "envelope_correlation": 0.9934303384407072, "rms_gain_error_db": -0.006924923214806733, "reference_mean_power": 0.0003354294563993909, "output_mean_power": 0.0003348950327736667}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 1.7752008532728783e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001256086055891369, "output_power": 0.00012692049636122085, "reference_active": true, "delta_db": 0.04512367440606774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.726051831560619e-06, "output_power": 7.689193130278128e-06, "reference_active": true, "delta_db": -0.5493701904720211}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2668523428769663e-06, "output_power": 2.3946144196055695e-06, "reference_active": true, "delta_db": 2.765095944268224}], "active_band_count": 3, "active_band_mae_db": 1.1198632697154378}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 7.108743338455805e-05, "log_rms_mae_db": 0.8416016200821941, "rms_envelope_nrmse": 0.06019839017575239, "envelope_correlation": 0.9972755863524505, "rms_gain_error_db": -0.006925219153400947, "reference_mean_power": 0.00033911524225685186, "output_mean_power": 0.0003385749231744286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 1.626160013872791e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011427350869395253, "output_power": 0.0001158591326804216, "reference_active": true, "delta_db": 0.0598471073813624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179974955820808e-06, "output_power": 4.14796728517146e-06, "reference_active": true, "delta_db": -0.03338357252299956}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.39572209062608e-08, "output_power": 1.3811790341327864e-07, "reference_active": true, "delta_db": 1.673198146643089}], "active_band_count": 3, "active_band_mae_db": 0.5888096088491502}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 4.804880216814636e-05, "log_rms_mae_db": 1.772388001843872, "rms_envelope_nrmse": 0.15192995797700537, "envelope_correlation": 0.9841274456435944, "rms_gain_error_db": -0.31859212532328735, "reference_mean_power": 0.00012431296346075482, "output_mean_power": 0.00011552000713940717}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 7.893026272799499e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.194673883780046e-05, "output_power": 3.8680333413017615e-05, "reference_active": true, "delta_db": -0.35207993201286636}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.051190096168371e-06, "output_power": 5.9294092831253865e-06, "reference_active": true, "delta_db": -0.7525099420282924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.501810525086e-07, "output_power": 1.2421305459024919e-06, "reference_active": true, "delta_db": 1.1636087594899671}], "active_band_count": 3, "active_band_mae_db": 0.7560662111770421}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 4.532286462232191e-05, "log_rms_mae_db": 1.2757392152148832, "rms_envelope_nrmse": 0.080086847334503, "envelope_correlation": 0.9961614434145329, "rms_gain_error_db": -0.31859356150849893, "reference_mean_power": 0.00012567880182247303, "output_mean_power": 0.0001167891978316267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 7.206643574889899e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.830114378682998e-05, "output_power": 3.569904774498812e-05, "reference_active": true, "delta_db": -0.30555111826558323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6851867496780266e-06, "output_power": 3.1053793824699927e-06, "reference_active": true, "delta_db": -0.7434483569133771}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.995617022655143e-08, "output_power": 5.7998831315914223e-08, "reference_active": true, "delta_db": -0.14414641412722295}], "active_band_count": 3, "active_band_mae_db": 0.39771529643539444}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.956971276376921, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5ed9b67af75ed6fbfcc38c68fde0bbc76a7655d4bb489ddddf2a5b23a7c79369", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9937380718120309, "standard_deviation": 0.0007815969891794936, "minimum": 0.9927426130475808, "maximum": 0.9954048665113758, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3856999993127312, "standard_deviation": 0.05816611878715763, "minimum": 1.3096466518429957, "maximum": 1.4809406641567975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.857699041962925, "standard_deviation": 0.43756159803199096, "minimum": 0.38032532600225794, "maximum": 1.6698966575869238, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09530829018909595, "standard_deviation": 0.006545939986666509, "minimum": 0.08187828299187276, "maximum": 0.10344544627881523, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9979213811595984, "standard_deviation": 0.0005468381327643013, "minimum": 0.9972755863524505, "maximum": 0.9988646472180445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8511249513672655, "standard_deviation": 0.039086979003338004, "minimum": 0.7969086281062263, "maximum": 0.904843391833539, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5192141000399233, "standard_deviation": 0.3551701450183832, "minimum": 0.07831516622230465, "maximum": 1.121141338912808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.057265858825866336, "standard_deviation": 0.005825563167441929, "minimum": 0.0439550043391544, "maximum": 0.0635562808696463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9845823157084264, "standard_deviation": 0.0021063767117721374, "minimum": 0.9810627031277533, "maximum": 0.9865617156917708, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7624733027147568, "standard_deviation": 0.05531364798890815, "minimum": 1.6849549767278609, "maximum": 1.8278374202609897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7641938367328831, "standard_deviation": 0.2626014822859318, "minimum": 0.26337421448079984, "maximum": 1.0625338220494063, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15709990143917546, "standard_deviation": 0.01437978434286721, "minimum": 0.1388051518009172, "maximum": 0.18361203099031154, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995269271175458, "standard_deviation": 0.0016450412551708166, "minimum": 0.9922466743010748, "maximum": 0.9976347735436976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2489519187829812, "standard_deviation": 0.06442105195526904, "minimum": 1.1242490601578399, "maximum": 1.3208322815593003, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6795528657213612, "standard_deviation": 0.2783793522218608, "minimum": 0.3936632853652075, "maximum": 1.132785264453153, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09960519519838498, "standard_deviation": 0.022954336978362532, "minimum": 0.06171605290673139, "maximum": 0.14160466486263004, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9977361186258171, "standard_deviation": 0.0005306495536903639, "minimum": 0.9968630976927378, "maximum": 0.9983331614113569, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1357127580619308, "standard_deviation": 0.04460109638910451, "minimum": 1.0508314704291668, "maximum": 1.1781998397899096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.535081234215119, "standard_deviation": 0.23670098353754565, "minimum": 0.1876297190318791, "maximum": 0.9393475154944504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.05560116662552012, "standard_deviation": 0.006795638677115338, "minimum": 0.04799961237608585, "maximum": 0.0667056557570777, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.999399651000866, "standard_deviation": 0.000190626931197343, "minimum": 0.9990836671646475, "maximum": 0.9996379595045272, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.618183213315211, "standard_deviation": 0.0327931503940395, "minimum": 0.5750748002419337, "maximum": 0.6639646128229664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.43534658820118055, "standard_deviation": 0.2286112921911989, "minimum": 0.16113034814017124, "maximum": 0.7978949827556123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.02861176062647905, "standard_deviation": 0.005037250090426344, "minimum": 0.0220132542560757, "maximum": 0.03649719166850579, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19783467427058865, "standard_deviation": 0.06354885086201117, "minimum": 0.09366036510821314, "maximum": 0.28358574919048257, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.686198060054121, "standard_deviation": 0.012581862299977955, "minimum": 5.668166628916179, "maximum": 5.702960048582205, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40508745515105554, "standard_deviation": 0.020168275695845016, "minimum": 0.3796380002184969, "maximum": 0.4413890034522534, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07291042429935155, "standard_deviation": 0.05979879221532535, "minimum": -0.14118589476793642, "maximum": 0.033685413921041504, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9983037791907786, "standard_deviation": 0.00034902266987684024, "minimum": 0.9975482269717438, "maximum": 0.9987541372026113, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8352124521577357, "standard_deviation": 0.04219287358728006, "minimum": 0.7384311756577949, "maximum": 0.8686818760239738, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40451977973318975, "standard_deviation": 0.16138764087784438, "minimum": 0.14091714453113577, "maximum": 0.581002051527919, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.045288955592285976, "standard_deviation": 0.0042029574484047545, "minimum": 0.0392859727207127, "maximum": 0.053625509343245836, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9995150131932178, "standard_deviation": 0.00015809653416868995, "minimum": 0.9991742377931452, "maximum": 0.9997219919253897, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44989356832998073, "standard_deviation": 0.024317201544126744, "minimum": 0.40655140541647194, "maximum": 0.48071879802778567, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.32562774103012126, "standard_deviation": 0.18446572720310261, "minimum": 0.11539175531013128, "maximum": 0.6489788128783702, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.025268571691922447, "standard_deviation": 0.0030274944577078234, "minimum": 0.020467190812392816, "maximum": 0.030643239159186764, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0030", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.94852876663208, "predicted_mos": 2.94852876663208, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "4d7b78a26bd838da9795cddcbd0f9a93ac5dc02ea92566c30ef53d0e37694956", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.887445699889213, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8628694920606541, "全帯域_10ms_rms_envelope_nrmse": 0.0471450684735722, "全帯域_10ms_envelope_correlation": 0.9981631646001868, "全帯域_10ms_modulation_mae_db": 0.5948327962176356, "全帯域_40ms_log_rms_mae_db": 0.48138910110963473, "全帯域_40ms_rms_envelope_nrmse": 0.02358498581965127, "全帯域_40ms_envelope_correlation": 0.9995999378436499, "全帯域_40ms_modulation_mae_db": 0.30485962628821334, "300–1000 Hz_10ms_log_rms_mae_db": 1.1144561082137456, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06718253318937431, "300–1000 Hz_10ms_envelope_correlation": 0.9966833481082855, "300–1000 Hz_10ms_modulation_mae_db": 1.081353114284231, "300–1000 Hz_40ms_log_rms_mae_db": 0.6686769878816569, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03150525367060204, "300–1000 Hz_40ms_envelope_correlation": 0.9992501630488098, "300–1000 Hz_40ms_modulation_mae_db": 0.7473407543715392, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1762602677791656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09584127257440801, "1000–3000 Hz_10ms_envelope_correlation": 0.9937805573669095, "1000–3000 Hz_10ms_modulation_mae_db": 0.7964534888918618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7118657975191367, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05997663774587988, "1000–3000 Hz_40ms_envelope_correlation": 0.9978225375218964, "1000–3000 Hz_40ms_modulation_mae_db": 0.5042087530087002, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2243282970341671, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15003675096234304, "3000–6000 Hz_10ms_envelope_correlation": 0.9865191138011564, "3000–6000 Hz_10ms_modulation_mae_db": 1.4361870888059807, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8388937566401646, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09459376004433008, "3000–6000 Hz_40ms_envelope_correlation": 0.9966936472751536, "3000–6000 Hz_40ms_modulation_mae_db": 1.099118031882955, "mr_stft_db": 5.307233926636753, "band_mae_db": 0.19478036464388201, "rms_gain_error_db": -0.09191188990587448, "output_peak": 0.39836758418038987}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8741621061443159, "全帯域_10ms_rms_envelope_nrmse": 0.04520473094803741, "全帯域_10ms_envelope_correlation": 0.9982447363135037, "全帯域_10ms_modulation_mae_db": 0.4991981979878384, "全帯域_40ms_log_rms_mae_db": 0.46450797879715994, "全帯域_40ms_rms_envelope_nrmse": 0.02542498086569601, "全帯域_40ms_envelope_correlation": 0.9994314893825982, "全帯域_40ms_modulation_mae_db": 0.2568100595834149, "300–1000 Hz_10ms_log_rms_mae_db": 1.0201812570409343, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.057396353482413065, "300–1000 Hz_10ms_envelope_correlation": 0.99758158703941, "300–1000 Hz_10ms_modulation_mae_db": 0.55041869174797, "300–1000 Hz_40ms_log_rms_mae_db": 0.575695956787112, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033319662799815866, "300–1000 Hz_40ms_envelope_correlation": 0.9991566132230012, "300–1000 Hz_40ms_modulation_mae_db": 0.46702965738258007, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1859248232473745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08622662584117081, "1000–3000 Hz_10ms_envelope_correlation": 0.9948174981451229, "1000–3000 Hz_10ms_modulation_mae_db": 0.7314220958954487, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7301082906121317, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053954662977887195, "1000–3000 Hz_40ms_envelope_correlation": 0.9980419450417001, "1000–3000 Hz_40ms_modulation_mae_db": 0.198095821413483, "3000–6000 Hz_10ms_log_rms_mae_db": 1.175261548979149, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14517191986894684, "3000–6000 Hz_10ms_envelope_correlation": 0.9855157822743492, "3000–6000 Hz_10ms_modulation_mae_db": 1.0304889601859424, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8443261486225496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08506599339563152, "3000–6000 Hz_40ms_envelope_correlation": 0.995313915952993, "3000–6000 Hz_40ms_modulation_mae_db": 0.6318529151171498, "mr_stft_db": 5.317121336284267, "band_mae_db": 0.16794633418059757, "rms_gain_error_db": -0.03598095628811669, "output_peak": 0.41216290921767595}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8498865450797586, "全帯域_10ms_rms_envelope_nrmse": 0.048974222076126284, "全帯域_10ms_envelope_correlation": 0.9980426614098354, "全帯域_10ms_modulation_mae_db": 0.34332822965861104, "全帯域_40ms_log_rms_mae_db": 0.47072170282895537, "全帯域_40ms_rms_envelope_nrmse": 0.03131363312345381, "全帯域_40ms_envelope_correlation": 0.999245574537855, "全帯域_40ms_modulation_mae_db": 0.103568934700168, "300–1000 Hz_10ms_log_rms_mae_db": 1.0508993006184977, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05962792156681279, "300–1000 Hz_10ms_envelope_correlation": 0.9973734772697775, "300–1000 Hz_10ms_modulation_mae_db": 0.1915462475420903, "300–1000 Hz_40ms_log_rms_mae_db": 0.6275785952678165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03188077178864715, "300–1000 Hz_40ms_envelope_correlation": 0.9992152233483953, "300–1000 Hz_40ms_modulation_mae_db": 0.25667441081434034, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1476910048046514, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0946917722916267, "1000–3000 Hz_10ms_envelope_correlation": 0.9936364611897478, "1000–3000 Hz_10ms_modulation_mae_db": 1.028264132346299, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7060467220991455, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047796815453356185, "1000–3000 Hz_40ms_envelope_correlation": 0.9984442817577897, "1000–3000 Hz_40ms_modulation_mae_db": 0.3839066737456894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2288982172937175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1665385571773631, "3000–6000 Hz_10ms_envelope_correlation": 0.9856379293371773, "3000–6000 Hz_10ms_modulation_mae_db": 0.5574156883291673, "3000–6000 Hz_40ms_log_rms_mae_db": 0.863375814516593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12771137347142297, "3000–6000 Hz_40ms_envelope_correlation": 0.9943771085638307, "3000–6000 Hz_40ms_modulation_mae_db": 0.8297632830980296, "mr_stft_db": 5.315721197032865, "band_mae_db": 0.21283790078814713, "rms_gain_error_db": -0.10873395145888365, "output_peak": 0.3990315380718198}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9620334752942146, "全帯域_10ms_rms_envelope_nrmse": 0.04694883808455911, "全帯域_10ms_envelope_correlation": 0.9981048789809953, "全帯域_10ms_modulation_mae_db": 0.3092226776788641, "全帯域_40ms_log_rms_mae_db": 0.4657041664179341, "全帯域_40ms_rms_envelope_nrmse": 0.02537396989351043, "全帯域_40ms_envelope_correlation": 0.9994294038688223, "全帯域_40ms_modulation_mae_db": 0.14452952035831654, "300–1000 Hz_10ms_log_rms_mae_db": 1.0599816319763151, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05704604616198838, "300–1000 Hz_10ms_envelope_correlation": 0.9977775800105355, "300–1000 Hz_10ms_modulation_mae_db": 0.24003267664348601, "300–1000 Hz_40ms_log_rms_mae_db": 0.6319472868338467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03285577045710208, "300–1000 Hz_40ms_envelope_correlation": 0.9993415369513589, "300–1000 Hz_40ms_modulation_mae_db": 0.09778863006061743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.180294371944139, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08400806463140591, "1000–3000 Hz_10ms_envelope_correlation": 0.9951177867690219, "1000–3000 Hz_10ms_modulation_mae_db": 0.9957200896949358, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6667315696433637, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.042333914039035594, "1000–3000 Hz_40ms_envelope_correlation": 0.9989374311786122, "1000–3000 Hz_40ms_modulation_mae_db": 0.8013357881360977, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2160676515133602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.10661074514687484, "3000–6000 Hz_10ms_envelope_correlation": 0.99205083495027, "3000–6000 Hz_10ms_modulation_mae_db": 0.2570380317510596, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8561954850839671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.055154218555715465, "3000–6000 Hz_40ms_envelope_correlation": 0.9978038387912008, "3000–6000 Hz_40ms_modulation_mae_db": 0.21191571512238028, "mr_stft_db": 5.3212572546775565, "band_mae_db": 0.13029127028217555, "rms_gain_error_db": 0.012330526142064384, "output_peak": 0.4158232021040301}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9226537178329206, "全帯域_10ms_rms_envelope_nrmse": 0.04312403819798159, "全帯域_10ms_envelope_correlation": 0.9984083819570011, "全帯域_10ms_modulation_mae_db": 0.43357721064687826, "全帯域_40ms_log_rms_mae_db": 0.5208115733294616, "全帯域_40ms_rms_envelope_nrmse": 0.021640533973076692, "全帯域_40ms_envelope_correlation": 0.9995946111944886, "全帯域_40ms_modulation_mae_db": 0.26864962523947844, "300–1000 Hz_10ms_log_rms_mae_db": 1.0349406525122602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05772896519261343, "300–1000 Hz_10ms_envelope_correlation": 0.9978729513143553, "300–1000 Hz_10ms_modulation_mae_db": 0.5348627207626047, "300–1000 Hz_40ms_log_rms_mae_db": 0.5697600436523791, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038206392384435565, "300–1000 Hz_40ms_envelope_correlation": 0.999204361010767, "300–1000 Hz_40ms_modulation_mae_db": 0.40473272597183724, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1211805356525373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0844271001319155, "1000–3000 Hz_10ms_envelope_correlation": 0.9950456835119219, "1000–3000 Hz_10ms_modulation_mae_db": 0.5189285821792206, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7213158014970791, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04966968698995214, "1000–3000 Hz_40ms_envelope_correlation": 0.9983911619958374, "1000–3000 Hz_40ms_modulation_mae_db": 0.2653278070393375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2350178243711376, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13981159750024655, "3000–6000 Hz_10ms_envelope_correlation": 0.9870868662993009, "3000–6000 Hz_10ms_modulation_mae_db": 0.8422558963092167, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8514933778240951, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08101855715946392, "3000–6000 Hz_40ms_envelope_correlation": 0.9964841507793719, "3000–6000 Hz_40ms_modulation_mae_db": 1.0145932103519208, "mr_stft_db": 5.327688116237574, "band_mae_db": 0.15353625726518236, "rms_gain_error_db": 0.00595336618171414, "output_peak": 0.41115448097597357}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.903775409969632, "全帯域_10ms_rms_envelope_nrmse": 0.04777857951095219, "全帯域_10ms_envelope_correlation": 0.9982535704000424, "全帯域_10ms_modulation_mae_db": 0.4627288776466311, "全帯域_40ms_log_rms_mae_db": 0.4925602250660579, "全帯域_40ms_rms_envelope_nrmse": 0.02746191369412413, "全帯域_40ms_envelope_correlation": 0.9995686888144163, "全帯域_40ms_modulation_mae_db": 0.35416713274591594, "300–1000 Hz_10ms_log_rms_mae_db": 1.08708727229048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.051376510914823745, "300–1000 Hz_10ms_envelope_correlation": 0.998072704634263, "300–1000 Hz_10ms_modulation_mae_db": 0.5901233333839612, "300–1000 Hz_40ms_log_rms_mae_db": 0.6525414659580528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02476424804291432, "300–1000 Hz_40ms_envelope_correlation": 0.9995549635150383, "300–1000 Hz_40ms_modulation_mae_db": 0.4167381617303605, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2061657524419085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08566487923950789, "1000–3000 Hz_10ms_envelope_correlation": 0.9953647976508163, "1000–3000 Hz_10ms_modulation_mae_db": 0.5561732091948731, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7146579495026993, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053069938842797204, "1000–3000 Hz_40ms_envelope_correlation": 0.9986709648897741, "1000–3000 Hz_40ms_modulation_mae_db": 0.45799333122880315, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2390635837889639, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14296570924677413, "3000–6000 Hz_10ms_envelope_correlation": 0.9880057982884937, "3000–6000 Hz_10ms_modulation_mae_db": 0.6831698218456664, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8746953367535275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08831517904132735, "3000–6000 Hz_40ms_envelope_correlation": 0.997465146208678, "3000–6000 Hz_40ms_modulation_mae_db": 0.43552252583746576, "mr_stft_db": 5.315862923599723, "band_mae_db": 0.24618324388373472, "rms_gain_error_db": -0.14255688533236954, "output_peak": 0.39729912490736724}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9060520536624473, "全帯域_10ms_rms_envelope_nrmse": 0.04464403146338474, "全帯域_10ms_envelope_correlation": 0.9984161993181501, "全帯域_10ms_modulation_mae_db": 0.4650989522655891, "全帯域_40ms_log_rms_mae_db": 0.42864677550899627, "全帯域_40ms_rms_envelope_nrmse": 0.025975091255649167, "全帯域_40ms_envelope_correlation": 0.9995485052780129, "全帯域_40ms_modulation_mae_db": 0.3556942459612382, "300–1000 Hz_10ms_log_rms_mae_db": 0.8991269173597439, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.04726983473411702, "300–1000 Hz_10ms_envelope_correlation": 0.9983508915217366, "300–1000 Hz_10ms_modulation_mae_db": 0.6220954931214152, "300–1000 Hz_40ms_log_rms_mae_db": 0.4803493478069921, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.024383327256808803, "300–1000 Hz_40ms_envelope_correlation": 0.9995441291009324, "300–1000 Hz_40ms_modulation_mae_db": 0.29016199391136493, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0817558376308782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08704968194156433, "1000–3000 Hz_10ms_envelope_correlation": 0.9950107669216157, "1000–3000 Hz_10ms_modulation_mae_db": 0.7835739306227808, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6488442883798313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05607122583810454, "1000–3000 Hz_40ms_envelope_correlation": 0.998215768350117, "1000–3000 Hz_40ms_modulation_mae_db": 0.6372357746194605, "3000–6000 Hz_10ms_log_rms_mae_db": 1.200804935032702, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16386658858621223, "3000–6000 Hz_10ms_envelope_correlation": 0.9824004115823046, "3000–6000 Hz_10ms_modulation_mae_db": 0.4666446043533443, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8740574735694149, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.101545303214416, "3000–6000 Hz_40ms_envelope_correlation": 0.9945470560738768, "3000–6000 Hz_40ms_modulation_mae_db": 0.21671017186471245, "mr_stft_db": 5.307155866794548, "band_mae_db": 0.19566889348660466, "rms_gain_error_db": -0.1148093184543516, "output_peak": 0.38478383252079806}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8659353590211647, "全帯域_10ms_rms_envelope_nrmse": 0.04975292460098309, "全帯域_10ms_envelope_correlation": 0.997882976083483, "全帯域_10ms_modulation_mae_db": 0.1402283468837414, "全帯域_40ms_log_rms_mae_db": 0.5356478758896583, "全帯域_40ms_rms_envelope_nrmse": 0.03146007112665577, "全帯域_40ms_envelope_correlation": 0.9991216410782291, "全帯域_40ms_modulation_mae_db": 0.20620409035666806, "300–1000 Hz_10ms_log_rms_mae_db": 1.0776905660596356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06734214365276084, "300–1000 Hz_10ms_envelope_correlation": 0.9970012866215596, "300–1000 Hz_10ms_modulation_mae_db": 0.5681891915813186, "300–1000 Hz_40ms_log_rms_mae_db": 0.6432416293948177, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04464197533465035, "300–1000 Hz_40ms_envelope_correlation": 0.9987966686279519, "300–1000 Hz_40ms_modulation_mae_db": 0.7110573951120398, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1299716746317414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08944360639804606, "1000–3000 Hz_10ms_envelope_correlation": 0.9943906902921088, "1000–3000 Hz_10ms_modulation_mae_db": 0.6022875575403943, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7334518507020051, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.053956132695284986, "1000–3000 Hz_40ms_envelope_correlation": 0.9980290874222498, "1000–3000 Hz_40ms_modulation_mae_db": 0.4356107646754063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1622660814210082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12733425223833428, "3000–6000 Hz_10ms_envelope_correlation": 0.9887228677311728, "3000–6000 Hz_10ms_modulation_mae_db": 0.4455875602387553, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8262688338867469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07038546649157007, "3000–6000 Hz_40ms_envelope_correlation": 0.9964462832565815, "3000–6000 Hz_40ms_modulation_mae_db": 0.7375393696334999, "mr_stft_db": 5.3140104567660815, "band_mae_db": 0.1179782887138569, "rms_gain_error_db": 0.028137725882980504, "output_peak": 0.4267938989356276}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44240, "output_samples": 44240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.96557126607057, "spectral_convergence": 0.21064281847547076, "spectral_convergence_status": "ok", "amplitude_floor": 7.692579600611731e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.168986511482174, "spectral_convergence": 0.20912776989469106, "spectral_convergence_status": "ok", "amplitude_floor": 5.999960941709396e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7871440023575165, "spectral_convergence": 0.21287572652337922, "spectral_convergence_status": "ok", "amplitude_floor": 5.417405012540354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 6.282816067608343e-05, "output_power": 6.376994577275835e-05, "delta_db": 0.06461701954830376, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019406375287345015, "output_power": 0.00019695561297060257, "delta_db": 0.06423936465205571, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005567857807558596, "output_power": 0.0005637037131605776, "delta_db": 0.053627598679033515, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00030647737825017173, "output_power": 0.0002820382581222754, "delta_db": -0.36090400038190373, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00015415279416012865, "output_power": 0.00014362086271625302, "delta_db": -0.3073386995505679, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.966178555848869e-05, "output_power": 4.6155852815805484e-05, "delta_db": -0.31795550505142756, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19478036464388201, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09191188990587448, "pitch": null, "output_rms_dbfs": -28.87313158570375, "output_peak": 0.39836758418038987, "reference_rms_dbfs": -28.781219695797876, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.307233926636753}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 0.00013943874660832744, "log_rms_mae_db": 0.8628694920606541, "rms_envelope_nrmse": 0.0471450684735722, "envelope_correlation": 0.9981631646001868, "rms_gain_error_db": -0.09191201348275319, "reference_mean_power": 0.0013263679977567136, "output_mean_power": 0.0012985923300504654}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 5.047819293813421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030810693936871264, "output_power": 0.00030511642291802246, "reference_active": true, "delta_db": -0.04235895657262813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2023354078518551e-05, "output_power": 1.0077799101038507e-05, "reference_active": true, "delta_db": -0.7665994040413565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2193973896229425e-06, "output_power": 4.03021848278896e-06, "reference_active": true, "delta_db": 0.9755400280389224}], "active_band_count": 3, "active_band_mae_db": 0.5948327962176356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 0.0001327212418155818, "log_rms_mae_db": 0.48138910110963473, "rms_envelope_nrmse": 0.02358498581965127, "envelope_correlation": 0.9995999378436499, "rms_gain_error_db": -0.09191120287534399, "reference_mean_power": 0.0013409420747202037, "output_mean_power": 0.0013128614541650869}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 4.5645799716966823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002745054731183271, "output_power": 0.0002730387083835752, "reference_active": true, "delta_db": -0.023267870458802728}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.376005868866601e-06, "output_power": 4.586611417219104e-06, "reference_active": true, "delta_db": -0.6896778637297446}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7970463694996416e-07, "output_power": 1.7155207245200685e-07, "reference_active": true, "delta_db": -0.20163314467609267}], "active_band_count": 3, "active_band_mae_db": 0.30485962628821334}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 9.713079411954574e-05, "log_rms_mae_db": 1.1144561082137456, "rms_envelope_nrmse": 0.06718253318937431, "envelope_correlation": 0.9966833481082855, "rms_gain_error_db": 0.039369975197897956, "reference_mean_power": 0.0005849439303269558, "output_mean_power": 0.0005902707139347089}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 2.471278629444775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014231281864882344, "output_power": 0.0001403643186672257, "reference_active": true, "delta_db": -0.05987298448141788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.153370236673122e-06, "output_power": 6.992679634552931e-06, "reference_active": true, "delta_db": -0.6669353191172557}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4861340139926787e-06, "output_power": 2.6532799680525116e-06, "reference_active": true, "delta_db": 2.5172510392540195}], "active_band_count": 3, "active_band_mae_db": 1.081353114284231}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 9.335285483371196e-05, "log_rms_mae_db": 0.6686769878816569, "rms_envelope_nrmse": 0.03150525367060204, "envelope_correlation": 0.9992501630488098, "rms_gain_error_db": 0.03936992888441883, "reference_mean_power": 0.0005913717973798096, "output_mean_power": 0.0005967571099038794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 2.253861652661314e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012623197394568146, "output_power": 0.000125387542748013, "reference_active": true, "delta_db": -0.02914982042021556}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1401147597866345e-06, "output_power": 2.8896512346673023e-06, "reference_active": true, "delta_db": -0.3610009133298675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.001702937868675e-08, "output_power": 1.2256587701885296e-07, "reference_active": true, "delta_db": 1.8518715293645347}], "active_band_count": 3, "active_band_mae_db": 0.7473407543715392}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 7.293407135281025e-05, "log_rms_mae_db": 1.1762602677791656, "rms_envelope_nrmse": 0.09584127257440801, "envelope_correlation": 0.9937805573669095, "rms_gain_error_db": -0.2477632985844874, "reference_mean_power": 0.0003354294563993909, "output_mean_power": 0.0003168289576310817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 1.7752008532728783e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001256086055891369, "output_power": 0.0001228905529724689, "reference_active": true, "delta_db": -0.09500896001181774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.726051831560619e-06, "output_power": 7.641596998267714e-06, "reference_active": true, "delta_db": -0.5763365812861821}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2668523428769663e-06, "output_power": 1.88160081050445e-06, "reference_active": true, "delta_db": 1.7180149253775858}], "active_band_count": 3, "active_band_mae_db": 0.7964534888918618}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 7.108743338455805e-05, "log_rms_mae_db": 0.7118657975191367, "rms_envelope_nrmse": 0.05997663774587988, "envelope_correlation": 0.9978225375218964, "rms_gain_error_db": -0.24776362536240296, "reference_mean_power": 0.00033911524225685186, "output_mean_power": 0.0003203103322829702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 1.626160013872791e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011427350869395253, "output_power": 0.00011280824762625184, "reference_active": true, "delta_db": -0.05604709409231076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179974955820808e-06, "output_power": 3.4705222108448285e-06, "reference_active": true, "delta_db": -0.8077885155234372}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.39572209062608e-08, "output_power": 1.0909610049548858e-07, "reference_active": true, "delta_db": 0.6487906494103526}], "active_band_count": 3, "active_band_mae_db": 0.5042087530087002}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 4.804880216814636e-05, "log_rms_mae_db": 1.2243282970341671, "rms_envelope_nrmse": 0.15003675096234304, "envelope_correlation": 0.9865191138011564, "rms_gain_error_db": -0.6105375901214017, "reference_mean_power": 0.00012431296346075482, "output_mean_power": 0.00010800967532192501}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 7.893026272799499e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.194673883780046e-05, "output_power": 3.585566535562827e-05, "reference_active": true, "delta_db": -0.6814041622043808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.051190096168371e-06, "output_power": 5.902755500636304e-06, "reference_active": true, "delta_db": -0.7720762858857344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.501810525086e-07, "output_power": 1.8336415767303122e-06, "reference_active": true, "delta_db": 2.855080818327827}], "active_band_count": 3, "active_band_mae_db": 1.4361870888059807}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 4.532286462232191e-05, "log_rms_mae_db": 0.8388937566401646, "rms_envelope_nrmse": 0.09459376004433008, "envelope_correlation": 0.9966936472751536, "rms_gain_error_db": -0.6105384684882053, "reference_mean_power": 0.00012567880182247303, "output_mean_power": 0.00010919636581998268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 7.206643574889899e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.830114378682998e-05, "output_power": 3.3208794383733325e-05, "reference_active": true, "delta_db": -0.6195863429297879}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6851867496780266e-06, "output_power": 3.1035522579033976e-06, "reference_active": true, "delta_db": -0.7460043848611404}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.995617022655143e-08, "output_power": 9.354276883994606e-08, "reference_active": true, "delta_db": 1.931763367857937}], "active_band_count": 3, "active_band_mae_db": 1.099118031882955}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9618253742681245, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0a511f75d95a919d315ee8679c542d3d1c32fada47e85b3b4d03bc667d501031", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9946455302309081, "standard_deviation": 0.0006430737672511908, "minimum": 0.9936364611897478, "maximum": 0.9953647976508163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1536555335165495, "standard_deviation": 0.04112684977051564, "minimum": 1.0817558376308782, "maximum": 1.2061657524419085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7516028857957266, "standard_deviation": 0.19056337546512758, "minimum": 0.5189285821792206, "maximum": 1.028264132346299, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08841912538120565, "standard_deviation": 0.004553584984541983, "minimum": 0.08400806463140591, "maximum": 0.09584127257440801, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9983191472697471, "standard_deviation": 0.0003674583793036251, "minimum": 0.9978225375218964, "maximum": 0.9989374311786122, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.704127783744424, "standard_deviation": 0.03037739021895866, "minimum": 0.6488442883798313, "maximum": 0.7334518507020051, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4604643392333722, "standard_deviation": 0.19374336719174176, "minimum": 0.198095821413483, "maximum": 0.8013357881360977, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05210362682278721, "standard_deviation": 0.005417405676881386, "minimum": 0.042333914039035594, "maximum": 0.05997663774587988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9869924505330281, "standard_deviation": 0.0028016202150479724, "minimum": 0.9824004115823046, "maximum": 0.99205083495027, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2102135174292756, "standard_deviation": 0.02838338047680258, "minimum": 1.1622660814210082, "maximum": 1.2390635837889639, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7148484564773916, "standard_deviation": 0.37903572873714014, "minimum": 0.2570380317510596, "maximum": 1.4361870888059807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14279201509088688, "standard_deviation": 0.019358002551606638, "minimum": 0.10661074514687484, "maximum": 0.1665385571773631, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9961413933627108, "standard_deviation": 0.001274519650888239, "minimum": 0.9943771085638307, "maximum": 0.9978038387912008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8536632783621323, "standard_deviation": 0.01698529165355821, "minimum": 0.8262688338867469, "maximum": 0.8746953367535275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6471269028635143, "standard_deviation": 0.33817471746358624, "minimum": 0.21191571512238028, "maximum": 1.099118031882955, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08797373142173467, "standard_deviation": 0.021548967491000666, "minimum": 0.055154218555715465, "maximum": 0.12771137347142297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9975892283149903, "standard_deviation": 0.0005532703523812496, "minimum": 0.9966833481082855, "maximum": 0.9983508915217366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0430454632589516, "standard_deviation": 0.06534410959654227, "minimum": 0.8991269173597439, "maximum": 1.1144561082137456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5473276836333847, "standard_deviation": 0.2710339033319675, "minimum": 0.1915462475420903, "maximum": 1.081353114284231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.05812128861186295, "standard_deviation": 0.006912534538443407, "minimum": 0.04726983473411702, "maximum": 0.06734214365276084, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9992579573532818, "standard_deviation": 0.00024050836446250814, "minimum": 0.9987966686279519, "maximum": 0.9995549635150383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6062239141978343, "standard_deviation": 0.06169012893752984, "minimum": 0.4803493478069921, "maximum": 0.6686769878816569, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.42394046616933495, "standard_deviation": 0.22089137578640683, "minimum": 0.09778863006061743, "maximum": 0.7473407543715392, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03269467521687203, "standard_deviation": 0.006628735172119303, "minimum": 0.024383327256808803, "maximum": 0.04464197533465035, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1774028191555226, "standard_deviation": 0.04314730770408554, "minimum": 0.1179782887138569, "maximum": 0.24618324388373472, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.315756384753671, "standard_deviation": 0.006804944983124251, "minimum": 5.307155866794548, "maximum": 5.327688116237574, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4056770713642103, "standard_deviation": 0.013230061095956949, "minimum": 0.38478383252079806, "maximum": 0.4267938989356276, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05594642290410462, "standard_deviation": 0.06655646395198053, "minimum": -0.14255688533236954, "maximum": 0.028137725882980504, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9981895711328997, "standard_deviation": 0.00018119865620025837, "minimum": 0.997882976083483, "maximum": 0.9984161993181501, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8934210198831385, "standard_deviation": 0.03738599763752356, "minimum": 0.8498865450797586, "maximum": 0.9620334752942146, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40602691112322364, "standard_deviation": 0.1392547767943983, "minimum": 0.1402283468837414, "maximum": 0.5948327962176356, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04669655416944958, "standard_deviation": 0.002241364977979361, "minimum": 0.04312403819798159, "maximum": 0.04975292460098309, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999442481499759, "standard_deviation": 0.00017611199800230995, "minimum": 0.9991216410782291, "maximum": 0.9995999378436499, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4824986748684823, "standard_deviation": 0.03388979302353352, "minimum": 0.42864677550899627, "maximum": 0.5356478758896583, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24931040440417668, "standard_deviation": 0.09250960222662308, "minimum": 0.103568934700168, "maximum": 0.3556942459612382, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02652939746897716, "standard_deviation": 0.0034538299768388362, "minimum": 0.021640533973076692, "maximum": 0.03146007112665577, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0030", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.3131749629974365, "predicted_mos": 3.3131749629974365, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "b2a6a5f91a2c6fd5e2a759aa6ab0f5ffb5e0867fd527762a1d9f83c14dca3256", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9418778996914625, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8482315188265768, "全帯域_10ms_rms_envelope_nrmse": 0.058127506684149885, "全帯域_10ms_envelope_correlation": 0.9971479531107569, "全帯域_10ms_modulation_mae_db": 0.6959954548668473, "全帯域_40ms_log_rms_mae_db": 0.4558034829427855, "全帯域_40ms_rms_envelope_nrmse": 0.025862450094136673, "全帯域_40ms_envelope_correlation": 0.9994930891556656, "全帯域_40ms_modulation_mae_db": 0.4838200916504261, "300–1000 Hz_10ms_log_rms_mae_db": 1.0773496893624526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06298216858610617, "300–1000 Hz_10ms_envelope_correlation": 0.9970869271713757, "300–1000 Hz_10ms_modulation_mae_db": 0.762675007149606, "300–1000 Hz_40ms_log_rms_mae_db": 0.5745175123141262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030016065887931404, "300–1000 Hz_40ms_envelope_correlation": 0.9993524483563588, "300–1000 Hz_40ms_modulation_mae_db": 0.7674787473567659, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3379194111168993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0999758510466163, "1000–3000 Hz_10ms_envelope_correlation": 0.9929267253197317, "1000–3000 Hz_10ms_modulation_mae_db": 1.1885409343238702, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8200023120767234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05809170620981216, "1000–3000 Hz_40ms_envelope_correlation": 0.9976725338994851, "1000–3000 Hz_40ms_modulation_mae_db": 0.6538865655987502, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4771932224819204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2273786130831781, "3000–6000 Hz_10ms_envelope_correlation": 0.9634476320278303, "3000–6000 Hz_10ms_modulation_mae_db": 1.8939805608777156, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0986493784713018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1321724656946064, "3000–6000 Hz_40ms_envelope_correlation": 0.9874243493385941, "3000–6000 Hz_40ms_modulation_mae_db": 1.7215191407554127, "mr_stft_db": 5.411737452066066, "band_mae_db": 0.18911017739057193, "rms_gain_error_db": -0.0948146087173639, "output_peak": 0.40021942349072953}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8771192859575702, "全帯域_10ms_rms_envelope_nrmse": 0.047668640952257445, "全帯域_10ms_envelope_correlation": 0.998108321550594, "全帯域_10ms_modulation_mae_db": 0.3533080405866618, "全帯域_40ms_log_rms_mae_db": 0.5081273228511896, "全帯域_40ms_rms_envelope_nrmse": 0.026272770189788935, "全帯域_40ms_envelope_correlation": 0.9994645251644204, "全帯域_40ms_modulation_mae_db": 0.36311106618252226, "300–1000 Hz_10ms_log_rms_mae_db": 1.0431445772715902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061862248700127166, "300–1000 Hz_10ms_envelope_correlation": 0.9971713893709642, "300–1000 Hz_10ms_modulation_mae_db": 0.19919630089284976, "300–1000 Hz_40ms_log_rms_mae_db": 0.5815501054200578, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034917694211078605, "300–1000 Hz_40ms_envelope_correlation": 0.9990721794746927, "300–1000 Hz_40ms_modulation_mae_db": 0.6743304309422515, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2354285218530092, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08466213070616148, "1000–3000 Hz_10ms_envelope_correlation": 0.9949321836766403, "1000–3000 Hz_10ms_modulation_mae_db": 0.6528185035777989, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8115942070594522, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050228243313793976, "1000–3000 Hz_40ms_envelope_correlation": 0.9982279526412542, "1000–3000 Hz_40ms_modulation_mae_db": 0.26513146108026964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3502093471933216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13902807732375824, "3000–6000 Hz_10ms_envelope_correlation": 0.9875442617507343, "3000–6000 Hz_10ms_modulation_mae_db": 0.6787183833023599, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9578633681101518, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07844513066806857, "3000–6000 Hz_40ms_envelope_correlation": 0.9971632041163765, "3000–6000 Hz_40ms_modulation_mae_db": 0.4213737510701754, "mr_stft_db": 5.410918161676257, "band_mae_db": 0.23004757929188083, "rms_gain_error_db": -0.08999509430527393, "output_peak": 0.4153248898927057}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9079347170175097, "全帯域_10ms_rms_envelope_nrmse": 0.05239698397763868, "全帯域_10ms_envelope_correlation": 0.997797818482493, "全帯域_10ms_modulation_mae_db": 0.3512441493934941, "全帯域_40ms_log_rms_mae_db": 0.5010037612560698, "全帯域_40ms_rms_envelope_nrmse": 0.033139546021875535, "全帯域_40ms_envelope_correlation": 0.9992009319107139, "全帯域_40ms_modulation_mae_db": 0.5299525210370387, "300–1000 Hz_10ms_log_rms_mae_db": 1.0622440260035388, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05232183073038108, "300–1000 Hz_10ms_envelope_correlation": 0.9979780296026297, "300–1000 Hz_10ms_modulation_mae_db": 0.19368485522709844, "300–1000 Hz_40ms_log_rms_mae_db": 0.5763348812085396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.027461907066259773, "300–1000 Hz_40ms_envelope_correlation": 0.9994199321590558, "300–1000 Hz_40ms_modulation_mae_db": 0.24947045953202565, "1000–3000 Hz_10ms_log_rms_mae_db": 1.227235143274602, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08675647915517125, "1000–3000 Hz_10ms_envelope_correlation": 0.9953057221701465, "1000–3000 Hz_10ms_modulation_mae_db": 0.9034569979737118, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7300476323221622, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05649606226643881, "1000–3000 Hz_40ms_envelope_correlation": 0.9984666034045622, "1000–3000 Hz_40ms_modulation_mae_db": 0.7104247975867194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3414858072857905, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16571017722438208, "3000–6000 Hz_10ms_envelope_correlation": 0.9820745631947242, "3000–6000 Hz_10ms_modulation_mae_db": 0.8668184212512319, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9787940233926745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09761591055368479, "3000–6000 Hz_40ms_envelope_correlation": 0.995320756975933, "3000–6000 Hz_40ms_modulation_mae_db": 0.9297687679399217, "mr_stft_db": 5.413405723993989, "band_mae_db": 0.23945129707186666, "rms_gain_error_db": -0.13138368600639364, "output_peak": 0.3969668973448961}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8050833371455085, "全帯域_10ms_rms_envelope_nrmse": 0.04214012387263752, "全帯域_10ms_envelope_correlation": 0.9985306394647053, "全帯域_10ms_modulation_mae_db": 0.12387320714071326, "全帯域_40ms_log_rms_mae_db": 0.4257159110194161, "全帯域_40ms_rms_envelope_nrmse": 0.02091886920637349, "全帯域_40ms_envelope_correlation": 0.9996823399192802, "全帯域_40ms_modulation_mae_db": 0.30232407616300666, "300–1000 Hz_10ms_log_rms_mae_db": 1.0898881994664165, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.052674424541444304, "300–1000 Hz_10ms_envelope_correlation": 0.9979486272539335, "300–1000 Hz_10ms_modulation_mae_db": 0.2841917071843978, "300–1000 Hz_40ms_log_rms_mae_db": 0.5851957752057273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02592271479419143, "300–1000 Hz_40ms_envelope_correlation": 0.9994803905827033, "300–1000 Hz_40ms_modulation_mae_db": 0.3204890553993474, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1652709433621493, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08440276792063414, "1000–3000 Hz_10ms_envelope_correlation": 0.9950509369100553, "1000–3000 Hz_10ms_modulation_mae_db": 0.927906452762353, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7155483674522672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04496765377454143, "1000–3000 Hz_40ms_envelope_correlation": 0.9987326496243887, "1000–3000 Hz_40ms_modulation_mae_db": 0.5856289029285534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4151741556922026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14340933651327473, "3000–6000 Hz_10ms_envelope_correlation": 0.9877130400334703, "3000–6000 Hz_10ms_modulation_mae_db": 0.5570799385364736, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0094911701477385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09529358374017048, "3000–6000 Hz_40ms_envelope_correlation": 0.9962481138640329, "3000–6000 Hz_40ms_modulation_mae_db": 0.82417411493689, "mr_stft_db": 5.414456257661821, "band_mae_db": 0.16492160893939012, "rms_gain_error_db": -0.08399829114014404, "output_peak": 0.42075770101840937}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.899656572514425, "全帯域_10ms_rms_envelope_nrmse": 0.044180403629131525, "全帯域_10ms_envelope_correlation": 0.9983294403964592, "全帯域_10ms_modulation_mae_db": 0.38860958925971173, "全帯域_40ms_log_rms_mae_db": 0.48694042110458224, "全帯域_40ms_rms_envelope_nrmse": 0.020520121115018036, "全帯域_40ms_envelope_correlation": 0.9996397785605685, "全帯域_40ms_modulation_mae_db": 0.5174389623481833, "300–1000 Hz_10ms_log_rms_mae_db": 1.0958584943618086, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.055748571395029534, "300–1000 Hz_10ms_envelope_correlation": 0.9977286435505702, "300–1000 Hz_10ms_modulation_mae_db": 0.33015136664528355, "300–1000 Hz_40ms_log_rms_mae_db": 0.6489045694818785, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.028560425586953157, "300–1000 Hz_40ms_envelope_correlation": 0.9993997049426937, "300–1000 Hz_40ms_modulation_mae_db": 0.33286110982711953, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2753664666687634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08668462946626078, "1000–3000 Hz_10ms_envelope_correlation": 0.9947317345902305, "1000–3000 Hz_10ms_modulation_mae_db": 0.8502020683024495, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7780210265905084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04691272385857321, "1000–3000 Hz_40ms_envelope_correlation": 0.998561059542128, "1000–3000 Hz_40ms_modulation_mae_db": 0.801627593907424, "3000–6000 Hz_10ms_log_rms_mae_db": 1.349376146224732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12185207049165768, "3000–6000 Hz_10ms_envelope_correlation": 0.990604512461362, "3000–6000 Hz_10ms_modulation_mae_db": 0.28892171741883926, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9668867049434728, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08047680032050095, "3000–6000 Hz_40ms_envelope_correlation": 0.9966424006336734, "3000–6000 Hz_40ms_modulation_mae_db": 0.33405239956841953, "mr_stft_db": 5.393953526607067, "band_mae_db": 0.15384554201148412, "rms_gain_error_db": -0.04586115775494248, "output_peak": 0.39269708070752274}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8473332831661474, "全帯域_10ms_rms_envelope_nrmse": 0.05280510509791382, "全帯域_10ms_envelope_correlation": 0.9976195743785683, "全帯域_10ms_modulation_mae_db": 0.37986681643384373, "全帯域_40ms_log_rms_mae_db": 0.42255235245009815, "全帯域_40ms_rms_envelope_nrmse": 0.024962514790504774, "全帯域_40ms_envelope_correlation": 0.9994801014780328, "全帯域_40ms_modulation_mae_db": 0.17996095261883602, "300–1000 Hz_10ms_log_rms_mae_db": 1.0995905445974294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06029937681008503, "300–1000 Hz_10ms_envelope_correlation": 0.9973485400288289, "300–1000 Hz_10ms_modulation_mae_db": 0.14059886928496945, "300–1000 Hz_40ms_log_rms_mae_db": 0.6073620451362213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.027902283787894073, "300–1000 Hz_40ms_envelope_correlation": 0.9994264942589339, "300–1000 Hz_40ms_modulation_mae_db": 0.3151695035067621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2447784802397557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10256823590164335, "1000–3000 Hz_10ms_envelope_correlation": 0.9927587607982059, "1000–3000 Hz_10ms_modulation_mae_db": 1.2887875076438897, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7858131013579855, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.061022225086121734, "1000–3000 Hz_40ms_envelope_correlation": 0.9976794830083432, "1000–3000 Hz_40ms_modulation_mae_db": 0.9170299364103434, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4570938600452379, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17976448878860293, "3000–6000 Hz_10ms_envelope_correlation": 0.9775959571445043, "3000–6000 Hz_10ms_modulation_mae_db": 1.2530892005989431, "3000–6000 Hz_40ms_log_rms_mae_db": 1.014388928590764, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08059833543438558, "3000–6000 Hz_40ms_envelope_correlation": 0.996487564868983, "3000–6000 Hz_40ms_modulation_mae_db": 0.7363583316862007, "mr_stft_db": 5.4104496380377185, "band_mae_db": 0.19621981171068867, "rms_gain_error_db": -0.06456932469909463, "output_peak": 0.41437749680507824}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8686299463220707, "全帯域_10ms_rms_envelope_nrmse": 0.04326004405409009, "全帯域_10ms_envelope_correlation": 0.9983851347924014, "全帯域_10ms_modulation_mae_db": 0.42779970236288634, "全帯域_40ms_log_rms_mae_db": 0.4498708338515491, "全帯域_40ms_rms_envelope_nrmse": 0.02167020141714844, "全帯域_40ms_envelope_correlation": 0.9995778716761643, "全帯域_40ms_modulation_mae_db": 0.21572479965662372, "300–1000 Hz_10ms_log_rms_mae_db": 1.0954890767673024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05766434126591224, "300–1000 Hz_10ms_envelope_correlation": 0.9978424633160857, "300–1000 Hz_10ms_modulation_mae_db": 0.6274993338901036, "300–1000 Hz_40ms_log_rms_mae_db": 0.6056686569405163, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03349379642703669, "300–1000 Hz_40ms_envelope_correlation": 0.9994236902206022, "300–1000 Hz_40ms_modulation_mae_db": 0.49431117629199867, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2829541211023683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08828098313163499, "1000–3000 Hz_10ms_envelope_correlation": 0.9945609128106523, "1000–3000 Hz_10ms_modulation_mae_db": 0.7238121163882516, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8196051293277944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05346374726464718, "1000–3000 Hz_40ms_envelope_correlation": 0.9980942559731947, "1000–3000 Hz_40ms_modulation_mae_db": 0.3661301600603097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4285444786487365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14543212757703078, "3000–6000 Hz_10ms_envelope_correlation": 0.9852568693201075, "3000–6000 Hz_10ms_modulation_mae_db": 1.0405655745707476, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0187557611823939, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07057005663117095, "3000–6000 Hz_40ms_envelope_correlation": 0.9967898939673862, "3000–6000 Hz_40ms_modulation_mae_db": 0.9099551407028527, "mr_stft_db": 5.410599214294412, "band_mae_db": 0.1577824046334538, "rms_gain_error_db": 0.005601008720736204, "output_peak": 0.4132944014178722}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8615999977906272, "全帯域_10ms_rms_envelope_nrmse": 0.044699698663751876, "全帯域_10ms_envelope_correlation": 0.9983007495501195, "全帯域_10ms_modulation_mae_db": 0.40421192062845607, "全帯域_40ms_log_rms_mae_db": 0.4490121025974439, "全帯域_40ms_rms_envelope_nrmse": 0.022235607629485312, "全帯域_40ms_envelope_correlation": 0.9995887086428525, "全帯域_40ms_modulation_mae_db": 0.046909432507556, "300–1000 Hz_10ms_log_rms_mae_db": 1.1196386258268942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05477195950527133, "300–1000 Hz_10ms_envelope_correlation": 0.9978097115827891, "300–1000 Hz_10ms_modulation_mae_db": 0.5117135468441719, "300–1000 Hz_40ms_log_rms_mae_db": 0.6371762397830942, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02872721379619935, "300–1000 Hz_40ms_envelope_correlation": 0.9993832006701071, "300–1000 Hz_40ms_modulation_mae_db": 0.42657466501870694, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2761255465576107, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09436771572882081, "1000–3000 Hz_10ms_envelope_correlation": 0.993813828040242, "1000–3000 Hz_10ms_modulation_mae_db": 0.8935386451623675, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7899618841808501, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057746067916685566, "1000–3000 Hz_40ms_envelope_correlation": 0.9978113709537267, "1000–3000 Hz_40ms_modulation_mae_db": 0.2940682078236286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4277227621437822, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12281013664437147, "3000–6000 Hz_10ms_envelope_correlation": 0.9897676394636209, "3000–6000 Hz_10ms_modulation_mae_db": 0.9721647995290178, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0213574681922313, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06105415269451851, "3000–6000 Hz_40ms_envelope_correlation": 0.9979049291647468, "3000–6000 Hz_40ms_modulation_mae_db": 0.9349820475584615, "mr_stft_db": 5.4230008659754425, "band_mae_db": 0.168923266880197, "rms_gain_error_db": -0.059469295344991, "output_peak": 0.4189280876381904}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44240, "output_samples": 44240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.115532822675517, "spectral_convergence": 0.2184851273581556, "spectral_convergence_status": "ok", "amplitude_floor": 7.692579600611731e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.286128061124065, "spectral_convergence": 0.22397549799180544, "spectral_convergence_status": "ok", "amplitude_floor": 5.999960941709396e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8335514723986135, "spectral_convergence": 0.21911898040618827, "spectral_convergence_status": "ok", "amplitude_floor": 5.417405012540354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 6.282816067608343e-05, "output_power": 6.467960461989503e-05, "delta_db": 0.12613010715652917, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019406375287345015, "output_power": 0.00019690845263076285, "delta_db": 0.0631993370260933, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005567857807558596, "output_power": 0.0005424503674886143, "delta_db": -0.11328127941440667, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00030647737825017173, "output_power": 0.00029687739451246965, "delta_db": -0.13821293979015634, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00015415279416012865, "output_power": 0.00015134916789526624, "delta_db": -0.07971363532474567, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.966178555848869e-05, "output_power": 4.31131697164729e-05, "delta_db": -0.6141237656315005, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18911017739057193, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0948146087173639, "pitch": null, "output_rms_dbfs": -28.876034304515237, "output_peak": 0.40021942349072953, "reference_rms_dbfs": -28.781219695797876, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.411737452066066}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 0.00013943874660832744, "log_rms_mae_db": 0.8482315188265768, "rms_envelope_nrmse": 0.058127506684149885, "envelope_correlation": 0.9971479531107569, "rms_gain_error_db": -0.09481478230288892, "reference_mean_power": 0.0013263679977567136, "output_mean_power": 0.00129772465753421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 5.047819293813421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030810693936871264, "output_power": 0.0003002769465462493, "reference_active": true, "delta_db": -0.11179488900453088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2023354078518551e-05, "output_power": 1.1445296299376708e-05, "reference_active": true, "delta_db": -0.21398596628097588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2193973896229425e-06, "output_power": 4.830533947139731e-06, "reference_active": true, "delta_db": 1.7622055093150353}], "active_band_count": 3, "active_band_mae_db": 0.6959954548668473}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 0.0001327212418155818, "log_rms_mae_db": 0.4558034829427855, "rms_envelope_nrmse": 0.025862450094136673, "envelope_correlation": 0.9994930891556656, "rms_gain_error_db": -0.09481503682194929, "reference_mean_power": 0.0013409420747202037, "output_mean_power": 0.0013119839257652946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 4.5645799716966823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002745054731183271, "output_power": 0.00026903718325772927, "reference_active": true, "delta_db": -0.08738700488920767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.376005868866601e-06, "output_power": 4.9471999802971275e-06, "reference_active": true, "delta_db": -0.36100267983122347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7970463694996416e-07, "output_power": 2.2639474527073617e-07, "reference_active": true, "delta_db": 1.0030705902308472}], "active_band_count": 3, "active_band_mae_db": 0.4838200916504261}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 9.713079411954574e-05, "log_rms_mae_db": 1.0773496893624526, "rms_envelope_nrmse": 0.06298216858610617, "envelope_correlation": 0.9970869271713757, "rms_gain_error_db": -0.07191198855374743, "reference_mean_power": 0.0005849439303269558, "output_mean_power": 0.0005753379744412176}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 2.471278629444775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014231281864882344, "output_power": 0.0001383813971757579, "reference_active": true, "delta_db": -0.12166309208561467}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.153370236673122e-06, "output_power": 8.58779124756939e-06, "reference_active": true, "delta_db": 0.22544315376917853}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4861340139926787e-06, "output_power": 2.3235383868803463e-06, "reference_active": true, "delta_db": 1.9409187755940247}], "active_band_count": 3, "active_band_mae_db": 0.762675007149606}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 9.335285483371196e-05, "log_rms_mae_db": 0.5745175123141262, "rms_envelope_nrmse": 0.030016065887931404, "envelope_correlation": 0.9993524483563588, "rms_gain_error_db": -0.07191203902335513, "reference_mean_power": 0.0005913717973798096, "output_mean_power": 0.0005816602762286156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 2.253861652661314e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012623197394568146, "output_power": 0.00012345676785296164, "reference_active": true, "delta_db": -0.09654470745945991}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1401147597866345e-06, "output_power": 3.4656911934847305e-06, "reference_active": true, "delta_db": 0.4284434253175912}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.001702937868675e-08, "output_power": 1.2048340568108616e-07, "reference_active": true, "delta_db": 1.7774481092932466}], "active_band_count": 3, "active_band_mae_db": 0.7674787473567659}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 7.293407135281025e-05, "log_rms_mae_db": 1.3379194111168993, "rms_envelope_nrmse": 0.0999758510466163, "envelope_correlation": 0.9929267253197317, "rms_gain_error_db": -0.16358596179861656, "reference_mean_power": 0.0003354294563993909, "output_mean_power": 0.0003230298101699262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 1.7752008532728783e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001256086055891369, "output_power": 0.0001237866509144958, "reference_active": true, "delta_db": -0.06345581304719458}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.726051831560619e-06, "output_power": 7.794488829973113e-06, "reference_active": true, "delta_db": -0.4903014929704647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2668523428769663e-06, "output_power": 2.5346181976731245e-06, "reference_active": true, "delta_db": 3.0118654969539516}], "active_band_count": 3, "active_band_mae_db": 1.1885409343238702}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 7.108743338455805e-05, "log_rms_mae_db": 0.8200023120767234, "rms_envelope_nrmse": 0.05809170620981216, "envelope_correlation": 0.9976725338994851, "rms_gain_error_db": -0.16358617444840837, "reference_mean_power": 0.00033911524225685186, "output_mean_power": 0.0003265793295081144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 1.626160013872791e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011427350869395253, "output_power": 0.00011392355973658652, "reference_active": true, "delta_db": -0.013320155365460676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179974955820808e-06, "output_power": 3.752828161318184e-06, "reference_active": true, "delta_db": -0.4681500081387402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.39572209062608e-08, "output_power": 1.32114083380587e-07, "reference_active": true, "delta_db": 1.4801895332920498}], "active_band_count": 3, "active_band_mae_db": 0.6538865655987502}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 4.804880216814636e-05, "log_rms_mae_db": 1.4771932224819204, "rms_envelope_nrmse": 0.2273786130831781, "envelope_correlation": 0.9634476320278303, "rms_gain_error_db": -0.2493355561679909, "reference_mean_power": 0.00012431296346075482, "output_mean_power": 0.00011737696179765119}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 7.893026272799499e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.194673883780046e-05, "output_power": 3.510991108756534e-05, "reference_active": true, "delta_db": -0.7726847245997605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.051190096168371e-06, "output_power": 6.182719855255978e-06, "reference_active": true, "delta_db": -0.5708285451206897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.501810525086e-07, "output_power": 2.5801752632061628e-06, "reference_active": true, "delta_db": 4.338428412912696}], "active_band_count": 3, "active_band_mae_db": 1.8939805608777156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 4.532286462232191e-05, "log_rms_mae_db": 1.0986493784713018, "rms_envelope_nrmse": 0.1321724656946064, "envelope_correlation": 0.9874243493385941, "rms_gain_error_db": -0.24933699842786294, "reference_mean_power": 0.00012567880182247303, "output_mean_power": 0.00011866655424063245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 7.206643574889899e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.830114378682998e-05, "output_power": 3.266123918089164e-05, "reference_active": true, "delta_db": -0.6917908547073506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6851867496780266e-06, "output_power": 3.3456309316938566e-06, "reference_active": true, "delta_db": -0.41981470242176144}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.995617022655143e-08, "output_power": 1.5245057631545219e-07, "reference_active": true, "delta_db": 4.052951865137126}], "active_band_count": 3, "active_band_mae_db": 1.7215191407554127}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9575199959773745, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7c4be0c50743382423a49cbf3e9c655dbdaed57fc5ba8bcde4d088f42adf9ccc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.994260100539488, "standard_deviation": 0.0009793903965555411, "minimum": 0.9927587607982059, "maximum": 0.9953057221701465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2556348292718946, "standard_deviation": 0.05046162946811159, "minimum": 1.1652709433621493, "maximum": 1.3379194111168993, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9286329032668366, "standard_deviation": 0.2149106187093005, "minimum": 0.6528185035777989, "maximum": 1.2887875076438897, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0909623491321179, "standard_deviation": 0.0071056833937314014, "minimum": 0.08440276792063414, "maximum": 0.10256823590164335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9981557386308854, "standard_deviation": 0.00041085947389783717, "minimum": 0.9976725338994851, "maximum": 0.9987326496243887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7813242075459679, "standard_deviation": 0.039531391774762385, "minimum": 0.7155483674522672, "maximum": 0.8200023120767234, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5742409531744999, "standard_deviation": 0.24248230198140475, "minimum": 0.26513146108026964, "maximum": 0.9170299364103434, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05361605371132676, "standard_deviation": 0.005750331859476238, "minimum": 0.04496765377454143, "maximum": 0.061022225086121734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9830005594245442, "standard_deviation": 0.00897569159539065, "minimum": 0.9634476320278303, "maximum": 0.990604512461362, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4058499724644655, "standard_deviation": 0.05240517049119581, "minimum": 1.3414858072857905, "maximum": 1.4771932224819204, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9439173245106661, "standard_deviation": 0.487454515672051, "minimum": 0.28892171741883926, "maximum": 1.8939805608777156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.155673128455782, "standard_deviation": 0.03500814013079913, "minimum": 0.12185207049165768, "maximum": 0.2273786130831781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954976516162157, "standard_deviation": 0.0033445481041453613, "minimum": 0.9874243493385941, "maximum": 0.9979049291647468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.008273350378841, "standard_deviation": 0.04409541013224103, "minimum": 0.9578633681101518, "maximum": 1.0986493784713018, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8515229617772917, "standard_deviation": 0.42091914286716925, "minimum": 0.33405239956841953, "maximum": 1.7215191407554127, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08702830446713827, "standard_deviation": 0.021784252503601043, "minimum": 0.06105415269451851, "maximum": 0.1321724656946064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9976142914846471, "standard_deviation": 0.0003570339833735638, "minimum": 0.9970869271713757, "maximum": 0.9979780296026297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0854004042071792, "standard_deviation": 0.023879147941857706, "minimum": 1.0431445772715902, "maximum": 1.1196386258268942, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38121387338981005, "standard_deviation": 0.22722332603982376, "minimum": 0.14059886928496945, "maximum": 0.762675007149606, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.057290615191794604, "standard_deviation": 0.004091899511958182, "minimum": 0.05232183073038108, "maximum": 0.06298216858610617, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9993697550831434, "standard_deviation": 0.00012582006952375506, "minimum": 0.9990721794746927, "maximum": 0.9994803905827033, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6020887231862702, "standard_deviation": 0.028269417327284222, "minimum": 0.5745175123141262, "maximum": 0.6489045694818785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4475856434843722, "standard_deviation": 0.1861138110870584, "minimum": 0.24947045953202565, "maximum": 0.7674787473567659, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.02962526269469306, "standard_deviation": 0.003080494080586974, "minimum": 0.02592271479419143, "maximum": 0.034917694211078605, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18753771099119165, "standard_deviation": 0.032641968941745425, "minimum": 0.15384554201148412, "maximum": 0.23945129707186666, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.411065105039096, "standard_deviation": 0.008047327998961067, "minimum": 5.393953526607067, "maximum": 5.4230008659754425, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4090707472894255, "standard_deviation": 0.010769091116576935, "minimum": 0.39269708070752274, "maximum": 0.42075770101840937, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07056130615593342, "standard_deviation": 0.04038306106547011, "minimum": -0.13138368600639364, "maximum": 0.005601008720736204, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9980274539657622, "standard_deviation": 0.00046998131177186664, "minimum": 0.9971479531107569, "maximum": 0.9985306394647053, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8644485823425545, "standard_deviation": 0.03250703308950165, "minimum": 0.8050833371455085, "maximum": 0.9079347170175097, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3906136100840768, "standard_deviation": 0.15539691246140575, "minimum": 0.12387320714071326, "maximum": 0.6959954548668473, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.048159813366446355, "standard_deviation": 0.005696878932544217, "minimum": 0.04214012387263752, "maximum": 0.058127506684149885, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9995159183134623, "standard_deviation": 0.0001492054521838761, "minimum": 0.9992009319107139, "maximum": 0.9996823399192802, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4623782735091418, "standard_deviation": 0.03272559978286794, "minimum": 0.42255235245009815, "maximum": 0.5081273228511896, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3299052377705241, "standard_deviation": 0.1758034739016715, "minimum": 0.046909432507556, "maximum": 0.5299525210370387, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0244477600580414, "standard_deviation": 0.004168452851656986, "minimum": 0.020520121115018036, "maximum": 0.033139546021875535, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0030", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.322763442993164, "predicted_mos": 3.322763442993164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "c2b13b44a2fc7b5ee8ccae88e088abb566592fd7c3dd9d64ef8a81df91870b54", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9544581999070942, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8174393821138654, "全帯域_10ms_rms_envelope_nrmse": 0.05204952511672779, "全帯域_10ms_envelope_correlation": 0.997662037276792, "全帯域_10ms_modulation_mae_db": 0.5639694268326955, "全帯域_40ms_log_rms_mae_db": 0.4243094614136136, "全帯域_40ms_rms_envelope_nrmse": 0.02609559736718512, "全帯域_40ms_envelope_correlation": 0.999383611166008, "全帯域_40ms_modulation_mae_db": 0.2719202725392768, "300–1000 Hz_10ms_log_rms_mae_db": 0.9835510404944305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05899510760735549, "300–1000 Hz_10ms_envelope_correlation": 0.9974156575859113, "300–1000 Hz_10ms_modulation_mae_db": 0.5738372847968937, "300–1000 Hz_40ms_log_rms_mae_db": 0.48443023285731907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02811627742946719, "300–1000 Hz_40ms_envelope_correlation": 0.9993848132791158, "300–1000 Hz_40ms_modulation_mae_db": 0.19516698765516827, "1000–3000 Hz_10ms_log_rms_mae_db": 1.186176695338142, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09935580697670837, "1000–3000 Hz_10ms_envelope_correlation": 0.9929924242214603, "1000–3000 Hz_10ms_modulation_mae_db": 0.7235088341239907, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7502318002275877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06148136413130725, "1000–3000 Hz_40ms_envelope_correlation": 0.9972140184553768, "1000–3000 Hz_40ms_modulation_mae_db": 0.8628336120699975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4350969223471437, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17577946238659647, "3000–6000 Hz_10ms_envelope_correlation": 0.9783478001753034, "3000–6000 Hz_10ms_modulation_mae_db": 1.459933966257422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9950593305042595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1040715779525051, "3000–6000 Hz_40ms_envelope_correlation": 0.9926477464285631, "3000–6000 Hz_40ms_modulation_mae_db": 1.1218181620272647, "mr_stft_db": 5.376985379016631, "band_mae_db": 0.1260440390347458, "rms_gain_error_db": -0.00469801306216684, "output_peak": 0.39149578300497034}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8574727346892097, "全帯域_10ms_rms_envelope_nrmse": 0.04972265678929299, "全帯域_10ms_envelope_correlation": 0.9978777893110258, "全帯域_10ms_modulation_mae_db": 0.38445423302964854, "全帯域_40ms_log_rms_mae_db": 0.4799690565833399, "全帯域_40ms_rms_envelope_nrmse": 0.025732835808842832, "全帯域_40ms_envelope_correlation": 0.9994257228328413, "全帯域_40ms_modulation_mae_db": 0.14708202809248885, "300–1000 Hz_10ms_log_rms_mae_db": 1.0684822213397667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.056301998399708786, "300–1000 Hz_10ms_envelope_correlation": 0.9976894474349898, "300–1000 Hz_10ms_modulation_mae_db": 0.38731896461219834, "300–1000 Hz_40ms_log_rms_mae_db": 0.5876726820549714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.02877854694036663, "300–1000 Hz_40ms_envelope_correlation": 0.9993908888679384, "300–1000 Hz_40ms_modulation_mae_db": 0.26396514826777273, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2439308838650212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0863837867468797, "1000–3000 Hz_10ms_envelope_correlation": 0.9950071827002822, "1000–3000 Hz_10ms_modulation_mae_db": 1.0477757290665226, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7458855896004117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05442904037460761, "1000–3000 Hz_40ms_envelope_correlation": 0.9982557161239209, "1000–3000 Hz_40ms_modulation_mae_db": 0.6873357174375481, "3000–6000 Hz_10ms_log_rms_mae_db": 1.353589353977035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19127943106213546, "3000–6000 Hz_10ms_envelope_correlation": 0.9746809013316167, "3000–6000 Hz_10ms_modulation_mae_db": 1.024565374902117, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9590327644531184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10374400142231761, "3000–6000 Hz_40ms_envelope_correlation": 0.9922815301804345, "3000–6000 Hz_40ms_modulation_mae_db": 0.8627141833669546, "mr_stft_db": 5.388590122725216, "band_mae_db": 0.1282381049282401, "rms_gain_error_db": -0.03953052036170808, "output_peak": 0.47362621310700104}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8818831158083124, "全帯域_10ms_rms_envelope_nrmse": 0.043607407244635486, "全帯域_10ms_envelope_correlation": 0.9984252473162587, "全帯域_10ms_modulation_mae_db": 0.20874128219904722, "全帯域_40ms_log_rms_mae_db": 0.4730244336596377, "全帯域_40ms_rms_envelope_nrmse": 0.02215352118709882, "全帯域_40ms_envelope_correlation": 0.9996392837022707, "全帯域_40ms_modulation_mae_db": 0.2558402304801324, "300–1000 Hz_10ms_log_rms_mae_db": 1.0499570096138793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.055492879228509845, "300–1000 Hz_10ms_envelope_correlation": 0.9977246925726417, "300–1000 Hz_10ms_modulation_mae_db": 0.3630110390930463, "300–1000 Hz_40ms_log_rms_mae_db": 0.5775343731091471, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.031304301995902624, "300–1000 Hz_40ms_envelope_correlation": 0.9992435452712332, "300–1000 Hz_40ms_modulation_mae_db": 0.20590236490344152, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2005730426667185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08397116667994356, "1000–3000 Hz_10ms_envelope_correlation": 0.9951167302401102, "1000–3000 Hz_10ms_modulation_mae_db": 0.8578031216396026, "1000–3000 Hz_40ms_log_rms_mae_db": 0.738662054049452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052319261525232516, "1000–3000 Hz_40ms_envelope_correlation": 0.9982077024193314, "1000–3000 Hz_40ms_modulation_mae_db": 0.49411564409213504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3530986300426862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1643532963365134, "3000–6000 Hz_10ms_envelope_correlation": 0.9866849332440236, "3000–6000 Hz_10ms_modulation_mae_db": 0.6685403843176864, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9851856932307403, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12898538502922802, "3000–6000 Hz_40ms_envelope_correlation": 0.9946706830847195, "3000–6000 Hz_40ms_modulation_mae_db": 0.3634975332583745, "mr_stft_db": 5.405291361072105, "band_mae_db": 0.2794555801051058, "rms_gain_error_db": -0.08695136979308746, "output_peak": 0.4146681069307692}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8394722094184508, "全帯域_10ms_rms_envelope_nrmse": 0.04591370503338528, "全帯域_10ms_envelope_correlation": 0.9981909192297844, "全帯域_10ms_modulation_mae_db": 0.36057155819496606, "全帯域_40ms_log_rms_mae_db": 0.43404559971592094, "全帯域_40ms_rms_envelope_nrmse": 0.0229680364654027, "全帯域_40ms_envelope_correlation": 0.999541329330357, "全帯域_40ms_modulation_mae_db": 0.5243743008948415, "300–1000 Hz_10ms_log_rms_mae_db": 1.1373915100460978, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06337761035680838, "300–1000 Hz_10ms_envelope_correlation": 0.9970684079878533, "300–1000 Hz_10ms_modulation_mae_db": 0.39332460322356794, "300–1000 Hz_40ms_log_rms_mae_db": 0.6783396590839551, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.032727080930947366, "300–1000 Hz_40ms_envelope_correlation": 0.9992049955235198, "300–1000 Hz_40ms_modulation_mae_db": 0.6459390529708733, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2447032516743386, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0912247031227835, "1000–3000 Hz_10ms_envelope_correlation": 0.9942815663283399, "1000–3000 Hz_10ms_modulation_mae_db": 0.47323671200236256, "1000–3000 Hz_40ms_log_rms_mae_db": 0.753863491046612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05319855206166014, "1000–3000 Hz_40ms_envelope_correlation": 0.9982515382126625, "1000–3000 Hz_40ms_modulation_mae_db": 0.36590764825432626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3699412267474511, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12907598513012092, "3000–6000 Hz_10ms_envelope_correlation": 0.98852105538239, "3000–6000 Hz_10ms_modulation_mae_db": 1.1843074931275617, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9463551541159327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06596383357350884, "3000–6000 Hz_40ms_envelope_correlation": 0.9972860429535055, "3000–6000 Hz_40ms_modulation_mae_db": 0.8503964373159608, "mr_stft_db": 5.39431249179092, "band_mae_db": 0.14745632770252973, "rms_gain_error_db": -0.0425395990191406, "output_peak": 0.38424855270335423}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8187345559242964, "全帯域_10ms_rms_envelope_nrmse": 0.04476008856342749, "全帯域_10ms_envelope_correlation": 0.9984093728632004, "全帯域_10ms_modulation_mae_db": 0.5102618877879367, "全帯域_40ms_log_rms_mae_db": 0.43892007757503887, "全帯域_40ms_rms_envelope_nrmse": 0.02608623738902904, "全帯域_40ms_envelope_correlation": 0.9995445720062772, "全帯域_40ms_modulation_mae_db": 0.23792980434425096, "300–1000 Hz_10ms_log_rms_mae_db": 1.0068900687953426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.04900509731076695, "300–1000 Hz_10ms_envelope_correlation": 0.998220516135602, "300–1000 Hz_10ms_modulation_mae_db": 0.07650925915984218, "300–1000 Hz_40ms_log_rms_mae_db": 0.5413571825375395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.027112816048271295, "300–1000 Hz_40ms_envelope_correlation": 0.9994296567455361, "300–1000 Hz_40ms_modulation_mae_db": 0.09139910029643487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2672735980226677, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0888401514858261, "1000–3000 Hz_10ms_envelope_correlation": 0.9949469047381009, "1000–3000 Hz_10ms_modulation_mae_db": 0.9847388550421167, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7593165368711784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05398758097728539, "1000–3000 Hz_40ms_envelope_correlation": 0.9985910997541538, "1000–3000 Hz_40ms_modulation_mae_db": 0.49789520029851975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3641008941973214, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13737465437546892, "3000–6000 Hz_10ms_envelope_correlation": 0.9878296771315551, "3000–6000 Hz_10ms_modulation_mae_db": 0.6418486889867324, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9682979421444394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.080280088342838, "3000–6000 Hz_40ms_envelope_correlation": 0.9968588180144732, "3000–6000 Hz_40ms_modulation_mae_db": 0.6963976773437969, "mr_stft_db": 5.368017651455049, "band_mae_db": 0.21604902260810177, "rms_gain_error_db": -0.11942065247057758, "output_peak": 0.39957031335744575}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8795755795243938, "全帯域_10ms_rms_envelope_nrmse": 0.047038128360378455, "全帯域_10ms_envelope_correlation": 0.9980865957100512, "全帯域_10ms_modulation_mae_db": 0.10845819386770239, "全帯域_40ms_log_rms_mae_db": 0.44620018326135397, "全帯域_40ms_rms_envelope_nrmse": 0.02365179865897931, "全帯域_40ms_envelope_correlation": 0.9994944865625802, "全帯域_40ms_modulation_mae_db": 0.5044978203886324, "300–1000 Hz_10ms_log_rms_mae_db": 0.9975368546781909, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.04889920588826932, "300–1000 Hz_10ms_envelope_correlation": 0.9982356840887109, "300–1000 Hz_10ms_modulation_mae_db": 0.01871809658399778, "300–1000 Hz_40ms_log_rms_mae_db": 0.5300312143451765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.024740394436136695, "300–1000 Hz_40ms_envelope_correlation": 0.9995296688705002, "300–1000 Hz_40ms_modulation_mae_db": 0.11182642351540954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1666659790461025, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0938315199734354, "1000–3000 Hz_10ms_envelope_correlation": 0.9936861476227723, "1000–3000 Hz_10ms_modulation_mae_db": 0.7743121570218016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7019009047913641, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05020327316168782, "1000–3000 Hz_40ms_envelope_correlation": 0.9981746166453213, "1000–3000 Hz_40ms_modulation_mae_db": 0.37596992185243455, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3672185752341532, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.157189238411735, "3000–6000 Hz_10ms_envelope_correlation": 0.9825695628819631, "3000–6000 Hz_10ms_modulation_mae_db": 0.5308115083816264, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9643354326773587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07963443605802617, "3000–6000 Hz_40ms_envelope_correlation": 0.9955336563167194, "3000–6000 Hz_40ms_modulation_mae_db": 0.5500615839187192, "mr_stft_db": 5.397021940901859, "band_mae_db": 0.09383342298442461, "rms_gain_error_db": -0.016587560838015835, "output_peak": 0.4295190829863058}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8353590294524562, "全帯域_10ms_rms_envelope_nrmse": 0.04811720899780227, "全帯域_10ms_envelope_correlation": 0.998010268097522, "全帯域_10ms_modulation_mae_db": 0.0890610620334026, "全帯域_40ms_log_rms_mae_db": 0.4524079169074678, "全帯域_40ms_rms_envelope_nrmse": 0.028395855792149198, "全帯域_40ms_envelope_correlation": 0.9992841835708722, "全帯域_40ms_modulation_mae_db": 0.4351036343850909, "300–1000 Hz_10ms_log_rms_mae_db": 1.0280360218564462, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05933964274556899, "300–1000 Hz_10ms_envelope_correlation": 0.9974268109006388, "300–1000 Hz_10ms_modulation_mae_db": 0.2024574132246438, "300–1000 Hz_40ms_log_rms_mae_db": 0.5788474842140332, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0353510295608242, "300–1000 Hz_40ms_envelope_correlation": 0.999057801034115, "300–1000 Hz_40ms_modulation_mae_db": 0.8313177554004155, "1000–3000 Hz_10ms_log_rms_mae_db": 1.177203387226082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08291389279191357, "1000–3000 Hz_10ms_envelope_correlation": 0.9950841064992516, "1000–3000 Hz_10ms_modulation_mae_db": 0.4501661384660607, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6852113758249406, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04740415884372992, "1000–3000 Hz_40ms_envelope_correlation": 0.9983773782761449, "1000–3000 Hz_40ms_modulation_mae_db": 0.21086636606026368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3877773750238445, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19509153038588972, "3000–6000 Hz_10ms_envelope_correlation": 0.9755233768552278, "3000–6000 Hz_10ms_modulation_mae_db": 0.7145418445776515, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0110956324911662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13927366185500423, "3000–6000 Hz_40ms_envelope_correlation": 0.9893319149440025, "3000–6000 Hz_40ms_modulation_mae_db": 0.5841808356660376, "mr_stft_db": 5.387120289285874, "band_mae_db": 0.17316480210471621, "rms_gain_error_db": -0.04029489946460513, "output_peak": 0.4108223048901493}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9308941560047082, "全帯域_10ms_rms_envelope_nrmse": 0.04771764431373547, "全帯域_10ms_envelope_correlation": 0.9980663542340009, "全帯域_10ms_modulation_mae_db": 0.20075442969907065, "全帯域_40ms_log_rms_mae_db": 0.5288241502253279, "全帯域_40ms_rms_envelope_nrmse": 0.021443981891252042, "全帯域_40ms_envelope_correlation": 0.999629021361324, "全帯域_40ms_modulation_mae_db": 0.48519009731956625, "300–1000 Hz_10ms_log_rms_mae_db": 1.0204692530230601, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05895814944578647, "300–1000 Hz_10ms_envelope_correlation": 0.9974330450110903, "300–1000 Hz_10ms_modulation_mae_db": 0.3465545844137419, "300–1000 Hz_40ms_log_rms_mae_db": 0.5475624089836222, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.023739594629486258, "300–1000 Hz_40ms_envelope_correlation": 0.999587498704599, "300–1000 Hz_40ms_modulation_mae_db": 0.2783567330228253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1761623583899075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0746983872912606, "1000–3000 Hz_10ms_envelope_correlation": 0.9960435206355738, "1000–3000 Hz_10ms_modulation_mae_db": 0.2109100678205503, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7067761748228358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04323475223846408, "1000–3000 Hz_40ms_envelope_correlation": 0.9986882256824132, "1000–3000 Hz_40ms_modulation_mae_db": 0.5262454635990784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3351206283358261, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17771983838174216, "3000–6000 Hz_10ms_envelope_correlation": 0.9779629154730016, "3000–6000 Hz_10ms_modulation_mae_db": 0.7299188761141145, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9569368843239157, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09999489453275687, "3000–6000 Hz_40ms_envelope_correlation": 0.9934719179163143, "3000–6000 Hz_40ms_modulation_mae_db": 0.40906983321481966, "mr_stft_db": 5.383941450299943, "band_mae_db": 0.12735164255738665, "rms_gain_error_db": -0.06425427793484562, "output_peak": 0.4074177538544712}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44240, "output_samples": 44240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.0788203101899345, "spectral_convergence": 0.20941172017430035, "spectral_convergence_status": "ok", "amplitude_floor": 7.692579600611731e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.254615824743655, "spectral_convergence": 0.2139544641535785, "spectral_convergence_status": "ok", "amplitude_floor": 5.999960941709396e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.797520002116302, "spectral_convergence": 0.2168545139503631, "spectral_convergence_status": "ok", "amplitude_floor": 5.417405012540354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 6.282816067608343e-05, "output_power": 6.317637598966702e-05, "delta_db": 0.024003639580541764, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019406375287345015, "output_power": 0.0001963902212430245, "delta_db": 0.051754337220696736, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005567857807558596, "output_power": 0.0005604913143894079, "delta_db": 0.028807512785374023, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00030647737825017173, "output_power": 0.00031178565457542546, "delta_db": 0.07457705357100601, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00015415279416012865, "output_power": 0.00013940225931953932, "delta_db": -0.4368158860365113, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.966178555848869e-05, "output_power": 5.129238705159689e-05, "delta_db": 0.140305805014345, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1260440390347458, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.00469801306216684, "pitch": null, "output_rms_dbfs": -28.78591770886004, "output_peak": 0.39149578300497034, "reference_rms_dbfs": -28.781219695797876, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.376985379016631}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 0.00013943874660832744, "log_rms_mae_db": 0.8174393821138654, "rms_envelope_nrmse": 0.05204952511672779, "envelope_correlation": 0.997662037276792, "rms_gain_error_db": -0.0046979518769044324, "reference_mean_power": 0.0013263679977567136, "output_mean_power": 0.0013249339836920175}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 5.047819293813421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030810693936871264, "output_power": 0.00030164076096562346, "reference_active": true, "delta_db": -0.09211452131115008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2023354078518551e-05, "output_power": 1.2172329982917033e-05, "reference_active": true, "delta_db": 0.053480803155377295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2193973896229425e-06, "output_power": 4.596273709795224e-06, "reference_active": true, "delta_db": 1.546312956031559}], "active_band_count": 3, "active_band_mae_db": 0.5639694268326955}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 0.0001327212418155818, "log_rms_mae_db": 0.4243094614136136, "rms_envelope_nrmse": 0.02609559736718512, "envelope_correlation": 0.999383611166008, "rms_gain_error_db": -0.004698441216745, "reference_mean_power": 0.0013409420747202037, "output_mean_power": 0.0013394921528411571}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 4.5645799716966823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002745054731183271, "output_power": 0.0002690907246310172, "reference_active": true, "delta_db": -0.08652279685582706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.376005868866601e-06, "output_power": 5.799906819234516e-06, "reference_active": true, "delta_db": 0.3296128212838512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7970463694996416e-07, "output_power": 1.9702521195305532e-07, "reference_active": true, "delta_db": 0.3996251994781522}], "active_band_count": 3, "active_band_mae_db": 0.2719202725392768}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 9.713079411954574e-05, "log_rms_mae_db": 0.9835510404944305, "rms_envelope_nrmse": 0.05899510760735549, "envelope_correlation": 0.9974156575859113, "rms_gain_error_db": -0.006105333344378094, "reference_mean_power": 0.0005849439303269558, "output_mean_power": 0.0005841221909894571}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 2.471278629444775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014231281864882344, "output_power": 0.00013984444933351476, "reference_active": true, "delta_db": -0.07598787222044058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.153370236673122e-06, "output_power": 7.788316385134607e-06, "reference_active": true, "delta_db": -0.19893578094002456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4861340139926787e-06, "output_power": 2.0735608968790978e-06, "reference_active": true, "delta_db": 1.4465882012302158}], "active_band_count": 3, "active_band_mae_db": 0.5738372847968937}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 9.335285483371196e-05, "log_rms_mae_db": 0.48443023285731907, "rms_envelope_nrmse": 0.02811627742946719, "envelope_correlation": 0.9993848132791158, "rms_gain_error_db": -0.00610533340394197, "reference_mean_power": 0.0005913717973798096, "output_mean_power": 0.0005905410280555424}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 2.253861652661314e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012623197394568146, "output_power": 0.0001249802536939252, "reference_active": true, "delta_db": -0.04327971629949148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1401147597866345e-06, "output_power": 2.9949577490445654e-06, "reference_active": true, "delta_db": -0.20554820230058962}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.001702937868675e-08, "output_power": 8.646686807864447e-08, "reference_active": true, "delta_db": 0.3366730443654237}], "active_band_count": 3, "active_band_mae_db": 0.19516698765516827}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 7.293407135281025e-05, "log_rms_mae_db": 1.186176695338142, "rms_envelope_nrmse": 0.09935580697670837, "envelope_correlation": 0.9929924242214603, "rms_gain_error_db": 0.03686965584575762, "reference_mean_power": 0.0003354294563993909, "output_mean_power": 0.0003382892241070803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 1.7752008532728783e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001256086055891369, "output_power": 0.0001194631968339746, "reference_active": true, "delta_db": -0.2178526214918214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.726051831560619e-06, "output_power": 8.759054321199782e-06, "reference_active": true, "delta_db": 0.01639431414879796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2668523428769663e-06, "output_power": 1.9785815526460363e-06, "reference_active": true, "delta_db": 1.9362795667313528}], "active_band_count": 3, "active_band_mae_db": 0.7235088341239907}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 7.108743338455805e-05, "log_rms_mae_db": 0.7502318002275877, "rms_envelope_nrmse": 0.06148136413130725, "envelope_correlation": 0.9972140184553768, "rms_gain_error_db": 0.03686966692925627, "reference_mean_power": 0.00033911524225685186, "output_mean_power": 0.00034200643470659207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 1.626160013872791e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011427350869395253, "output_power": 0.0001085890182021986, "reference_active": true, "delta_db": -0.22159655804488657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179974955820808e-06, "output_power": 4.421519211870286e-06, "reference_active": true, "delta_db": 0.24397836678983564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.39572209062608e-08, "output_power": 1.5318729723475044e-07, "reference_active": true, "delta_db": 2.1229259113752703}], "active_band_count": 3, "active_band_mae_db": 0.8628336120699975}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 4.804880216814636e-05, "log_rms_mae_db": 1.4350969223471437, "rms_envelope_nrmse": 0.17577946238659647, "envelope_correlation": 0.9783478001753034, "rms_gain_error_db": -0.3100014648160374, "reference_mean_power": 0.00012431296346075482, "output_mean_power": 0.0001157487402605759}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 7.893026272799499e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.194673883780046e-05, "output_power": 3.7488921189999855e-05, "reference_active": true, "delta_db": -0.4879525910851537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.051190096168371e-06, "output_power": 6.794730491926833e-06, "reference_active": true, "delta_db": -0.16090187739419878}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.501810525086e-07, "output_power": 2.2433710387606646e-06, "reference_active": true, "delta_db": 3.7309474302929138}], "active_band_count": 3, "active_band_mae_db": 1.459933966257422}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 4.532286462232191e-05, "log_rms_mae_db": 0.9950593305042595, "rms_envelope_nrmse": 0.1040715779525051, "envelope_correlation": 0.9926477464285631, "rms_gain_error_db": -0.310002854795821, "reference_mean_power": 0.00012567880182247303, "output_mean_power": 0.00011702044523390004}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 7.206643574889899e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.830114378682998e-05, "output_power": 3.51033699482871e-05, "reference_active": true, "delta_db": -0.37862932451868003}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6851867496780266e-06, "output_power": 3.517226511032595e-06, "reference_active": true, "delta_db": -0.20259162974860295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.995617022655143e-08, "output_power": 1.138301313935307e-07, "reference_active": true, "delta_db": 2.784233531814511}], "active_band_count": 3, "active_band_mae_db": 1.1218181620272647}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.962619078617351, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "276743f7c914e7a561da4efd4804185f20a0a4d8576f4f67d90c8f50f582e998", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9946448228732364, "standard_deviation": 0.0009543894871556948, "minimum": 0.9929924242214603, "maximum": 0.9960435206355738, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2078361495286225, "standard_deviation": 0.03845830049171949, "minimum": 1.1666659790461025, "maximum": 1.2672735980226677, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6903064518978759, "standard_deviation": 0.2891605462057477, "minimum": 0.2109100678205503, "maximum": 1.0477757290665226, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08765242688359384, "standard_deviation": 0.007509461801695613, "minimum": 0.0746983872912606, "maximum": 0.09935580697670837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9982200369461657, "standard_deviation": 0.00044692573303579237, "minimum": 0.9972140184553768, "maximum": 0.9986882256824132, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7302309909042979, "standard_deviation": 0.028030936035855308, "minimum": 0.6852113758249406, "maximum": 0.7593165368711784, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5026461967080379, "standard_deviation": 0.20155723744400028, "minimum": 0.21086636606026368, "maximum": 0.8628336120699975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05203224791424685, "standard_deviation": 0.005377953251160643, "minimum": 0.04323475223846408, "maximum": 0.06148136413130725, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9815150278093852, "standard_deviation": 0.005632218335422128, "minimum": 0.9746809013316167, "maximum": 0.98852105538239, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3707429507381828, "standard_deviation": 0.03015034697673037, "minimum": 1.3351206283358261, "maximum": 1.4350969223471437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.869308517083114, "standard_deviation": 0.3211976926577685, "minimum": 0.5308115083816264, "maximum": 1.459933966257422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16598292955877525, "standard_deviation": 0.023849202978713704, "minimum": 0.12907598513012092, "maximum": 0.19509153038588972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940102887298415, "standard_deviation": 0.0026355700490939067, "minimum": 0.9893319149440025, "maximum": 0.9972860429535055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9732873542426164, "standard_deviation": 0.021863083883316467, "minimum": 0.9463551541159327, "maximum": 1.0110956324911662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6797670307639909, "standard_deviation": 0.2556597704345919, "minimum": 0.3634975332583745, "maximum": 1.1218181620272647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10024348484577311, "standard_deviation": 0.024999388098297187, "minimum": 0.06596383357350884, "maximum": 0.13927366185500423, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9976517827146798, "standard_deviation": 0.00040807433595111884, "minimum": 0.9970684079878533, "maximum": 0.9982356840887109, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0365392474809019, "standard_deviation": 0.049150103724242084, "minimum": 0.9835510404944305, "maximum": 1.1373915100460978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2952164056384915, "standard_deviation": 0.18355916778114875, "minimum": 0.01871809658399778, "maximum": 0.5738372847968937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.05629621137284678, "standard_deviation": 0.00510322099558232, "minimum": 0.04889920588826932, "maximum": 0.06337761035680838, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9993536085370697, "standard_deviation": 0.00017547916830722916, "minimum": 0.999057801034115, "maximum": 0.999587498704599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5657219046482205, "standard_deviation": 0.0564193213310295, "minimum": 0.48443023285731907, "maximum": 0.6783396590839551, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3279841957540426, "standard_deviation": 0.26625507669878207, "minimum": 0.09139910029643487, "maximum": 0.8313177554004155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.02898375524642528, "standard_deviation": 0.00395947290556355, "minimum": 0.023739594629486258, "maximum": 0.0353510295608242, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16144911775315635, "standard_deviation": 0.06015544289103437, "minimum": 0.09383342298442461, "maximum": 0.2794555801051058, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.38766008581845, "standard_deviation": 0.011695664001090292, "minimum": 5.368017651455049, "maximum": 5.405291361072105, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41392101385430835, "standard_deviation": 0.02790225707341581, "minimum": 0.38424855270335423, "maximum": 0.47362621310700104, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.051784611618018395, "standard_deviation": 0.037398346265606716, "minimum": -0.11942065247057758, "maximum": -0.00469801306216684, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9980910730048295, "standard_deviation": 0.0002562584146955267, "minimum": 0.997662037276792, "maximum": 0.9984252473162587, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8576038453669617, "standard_deviation": 0.03852887734425939, "minimum": 0.8174393821138654, "maximum": 0.9308941560047082, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3032840092055587, "standard_deviation": 0.17887202488752868, "minimum": 0.0890610620334026, "maximum": 0.5639694268326955, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04736579555242315, "standard_deviation": 0.0027074411083681557, "minimum": 0.043607407244635486, "maximum": 0.05204952511672779, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9994927763165663, "standard_deviation": 0.000122471205467981, "minimum": 0.9992841835708722, "maximum": 0.9996392837022707, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45971260991771257, "standard_deviation": 0.03368845762187338, "minimum": 0.4243094614136136, "maximum": 0.5288241502253279, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.357742273555535, "standard_deviation": 0.14541239877942652, "minimum": 0.14708202809248885, "maximum": 0.5243743008948415, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02456598306999238, "standard_deviation": 0.0023795675975671095, "minimum": 0.021443981891252042, "maximum": 0.028395855792149198, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5142-33396-0065", "language": "en", "style": "normal", "speaker": "librispeech_5142", "source_utterance_id": "5142-33396-0065", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 83120, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5142-33396-0065.wav", "source_file_sha256": "8220322c62eaf1418ba3c053857998b922c37a96cbce5e7d19b77c33319f6c7f", "decoded_pcm_sha256": "0b650128a600d9d1c16e5bf3b64c63018c80e6b3463b2d6fb23e3aead6248f75", "original_source_sha256": "67800641844bf8c9d654fc9a6099941ba97ff31b0ac0a579209378598c4ea09b", "corpus_entry_id": "librispeech_5142-33396-0065"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.326630592346191, "predicted_mos": 4.326630592346191, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "5aad7ecb1f54d37f9f8c6a166eccb362abb255b16dc9eaee2147a04cc1e03d0f", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.134002695998788, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5142-33396-0065", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5aad7ecb1f54d37f9f8c6a166eccb362abb255b16dc9eaee2147a04cc1e03d0f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "fef7806050fadfa47764bf831e8f4678f2d63580b079ce66cfa32ee7dd9c929f"}, "rows": [{"case": "librispeech_5142-33396-0065", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.2076332569122314, "predicted_mos": 2.2076332569122314, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "ed8493fc46e1d5bb3a7fccb08268e6fee01914c334c95da2935db983152e0405", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.925071608999133, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5142-33396-0065", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ed8493fc46e1d5bb3a7fccb08268e6fee01914c334c95da2935db983152e0405", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0329522295372924, "全帯域_10ms_rms_envelope_nrmse": 0.07961685000574302, "全帯域_10ms_envelope_correlation": 0.9948608186148465, "全帯域_10ms_modulation_mae_db": 0.21284225002070167, "全帯域_40ms_log_rms_mae_db": 0.5522950432502546, "全帯域_40ms_rms_envelope_nrmse": 0.041887274036175634, "全帯域_40ms_envelope_correlation": 0.9985476285529108, "全帯域_40ms_modulation_mae_db": 0.21394151397643013, "300–1000 Hz_10ms_log_rms_mae_db": 1.3907394809245477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09924424739141821, "300–1000 Hz_10ms_envelope_correlation": 0.9930254074764885, "300–1000 Hz_10ms_modulation_mae_db": 0.8890465023808888, "300–1000 Hz_40ms_log_rms_mae_db": 0.918320954757808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04945734342166605, "300–1000 Hz_40ms_envelope_correlation": 0.9982883139945044, "300–1000 Hz_40ms_modulation_mae_db": 0.6907792599375734, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2766091304349896, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10771260293468372, "1000–3000 Hz_10ms_envelope_correlation": 0.9918050172437375, "1000–3000 Hz_10ms_modulation_mae_db": 0.2876040887491151, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7572799729013562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05328716247340266, "1000–3000 Hz_40ms_envelope_correlation": 0.9980652430401693, "1000–3000 Hz_40ms_modulation_mae_db": 0.3841044450784659, "3000–6000 Hz_10ms_log_rms_mae_db": 1.514863757068072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2317057861168755, "3000–6000 Hz_10ms_envelope_correlation": 0.9656112092157441, "3000–6000 Hz_10ms_modulation_mae_db": 0.6025142726569958, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9550175883982785, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15891921620305974, "3000–6000 Hz_40ms_envelope_correlation": 0.9829210346434841, "3000–6000 Hz_40ms_modulation_mae_db": 0.4793243888106263, "mr_stft_db": 5.762814913701953, "band_mae_db": 0.17741129847677004, "rms_gain_error_db": -0.07136257451027513, "output_peak": 0.5755080039895948}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0685346260011235, "全帯域_10ms_rms_envelope_nrmse": 0.07548061562883471, "全帯域_10ms_envelope_correlation": 0.9953578795799743, "全帯域_10ms_modulation_mae_db": 0.3652030778576589, "全帯域_40ms_log_rms_mae_db": 0.6133290442685366, "全帯域_40ms_rms_envelope_nrmse": 0.03620385898238611, "全帯域_40ms_envelope_correlation": 0.9988738241647492, "全帯域_40ms_modulation_mae_db": 0.46025728303603913, "300–1000 Hz_10ms_log_rms_mae_db": 1.384536783292683, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08747379262835749, "300–1000 Hz_10ms_envelope_correlation": 0.9945709892659469, "300–1000 Hz_10ms_modulation_mae_db": 1.0761190145123394, "300–1000 Hz_40ms_log_rms_mae_db": 0.9291871776120746, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042012857859472126, "300–1000 Hz_40ms_envelope_correlation": 0.9987422792262117, "300–1000 Hz_40ms_modulation_mae_db": 0.43544511800851665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.33640406270277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10678910862854138, "1000–3000 Hz_10ms_envelope_correlation": 0.9918631552285848, "1000–3000 Hz_10ms_modulation_mae_db": 0.4284170612146967, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7612907720109034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049494734755090225, "1000–3000 Hz_40ms_envelope_correlation": 0.9982016307178526, "1000–3000 Hz_40ms_modulation_mae_db": 0.9684282757937707, "3000–6000 Hz_10ms_log_rms_mae_db": 1.484215270718477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20778334243915247, "3000–6000 Hz_10ms_envelope_correlation": 0.9723755632256538, "3000–6000 Hz_10ms_modulation_mae_db": 0.4337093437700294, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9282023869318105, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1334234361204262, "3000–6000 Hz_40ms_envelope_correlation": 0.9879897207175387, "3000–6000 Hz_40ms_modulation_mae_db": 0.7315430386312789, "mr_stft_db": 5.779352881235333, "band_mae_db": 0.13979468263513137, "rms_gain_error_db": -0.028210526682122768, "output_peak": 0.5296595665841133}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0336602783890305, "全帯域_10ms_rms_envelope_nrmse": 0.07505881868152463, "全帯域_10ms_envelope_correlation": 0.9954396410593529, "全帯域_10ms_modulation_mae_db": 0.0857983625932195, "全帯域_40ms_log_rms_mae_db": 0.5318164355582388, "全帯域_40ms_rms_envelope_nrmse": 0.03500640880258943, "全帯域_40ms_envelope_correlation": 0.9990010086002113, "全帯域_40ms_modulation_mae_db": 0.12278947623496482, "300–1000 Hz_10ms_log_rms_mae_db": 1.4037939874807397, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08662479907967174, "300–1000 Hz_10ms_envelope_correlation": 0.9948056384832283, "300–1000 Hz_10ms_modulation_mae_db": 0.8064881815239913, "300–1000 Hz_40ms_log_rms_mae_db": 0.9771390874684034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04646008193548467, "300–1000 Hz_40ms_envelope_correlation": 0.9986231473977558, "300–1000 Hz_40ms_modulation_mae_db": 0.6671891188139786, "1000–3000 Hz_10ms_log_rms_mae_db": 1.267235947419948, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11024198760016372, "1000–3000 Hz_10ms_envelope_correlation": 0.9913381010250933, "1000–3000 Hz_10ms_modulation_mae_db": 0.23979474079135987, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7536173399688807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05461932068722409, "1000–3000 Hz_40ms_envelope_correlation": 0.997801914192137, "1000–3000 Hz_40ms_modulation_mae_db": 0.7315972317418676, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4925927190623962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23225287400314135, "3000–6000 Hz_10ms_envelope_correlation": 0.9655632524946824, "3000–6000 Hz_10ms_modulation_mae_db": 0.1524042395282695, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9069227224892885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13761543751537686, "3000–6000 Hz_40ms_envelope_correlation": 0.9872312529072989, "3000–6000 Hz_40ms_modulation_mae_db": 0.14508614278193321, "mr_stft_db": 5.765997057732181, "band_mae_db": 0.1273389116416173, "rms_gain_error_db": -0.07680301378201157, "output_peak": 0.5705298032287132}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.038383356549997, "全帯域_10ms_rms_envelope_nrmse": 0.09409257718284987, "全帯域_10ms_envelope_correlation": 0.9928324840583826, "全帯域_10ms_modulation_mae_db": 0.5996126064613682, "全帯域_40ms_log_rms_mae_db": 0.5435784617298256, "全帯域_40ms_rms_envelope_nrmse": 0.0473367711588417, "全帯域_40ms_envelope_correlation": 0.9980866347758339, "全帯域_40ms_modulation_mae_db": 0.42821007014593143, "300–1000 Hz_10ms_log_rms_mae_db": 1.4069987763781489, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1379336993795061, "300–1000 Hz_10ms_envelope_correlation": 0.9867529180746702, "300–1000 Hz_10ms_modulation_mae_db": 2.1087961727048192, "300–1000 Hz_40ms_log_rms_mae_db": 0.9653831281512575, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061665030666117567, "300–1000 Hz_40ms_envelope_correlation": 0.9973031310417788, "300–1000 Hz_40ms_modulation_mae_db": 1.7570499890400277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2914103242746946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11853337538067517, "1000–3000 Hz_10ms_envelope_correlation": 0.9900230300670447, "1000–3000 Hz_10ms_modulation_mae_db": 0.4568906293707929, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7523143305860666, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06020798048251132, "1000–3000 Hz_40ms_envelope_correlation": 0.9974508513127615, "1000–3000 Hz_40ms_modulation_mae_db": 0.2215306092479261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.50466125738526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21079605109709615, "3000–6000 Hz_10ms_envelope_correlation": 0.9715009384626553, "3000–6000 Hz_10ms_modulation_mae_db": 0.3137627514394215, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9420950784672576, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1396629078200754, "3000–6000 Hz_40ms_envelope_correlation": 0.9868607700347651, "3000–6000 Hz_40ms_modulation_mae_db": 0.5403811280523975, "mr_stft_db": 5.756463753928091, "band_mae_db": 0.16590652423858657, "rms_gain_error_db": 0.015577926443488268, "output_peak": 0.4959778644494527}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0168766218167853, "全帯域_10ms_rms_envelope_nrmse": 0.07509894559220003, "全帯域_10ms_envelope_correlation": 0.9954187533487748, "全帯域_10ms_modulation_mae_db": 0.5379961014401532, "全帯域_40ms_log_rms_mae_db": 0.5512842335506949, "全帯域_40ms_rms_envelope_nrmse": 0.03902953747755202, "全帯域_40ms_envelope_correlation": 0.9987122508201396, "全帯域_40ms_modulation_mae_db": 0.4978261463048983, "300–1000 Hz_10ms_log_rms_mae_db": 1.3684824704019982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09916920075482345, "300–1000 Hz_10ms_envelope_correlation": 0.9930187880973195, "300–1000 Hz_10ms_modulation_mae_db": 0.8525000229319422, "300–1000 Hz_40ms_log_rms_mae_db": 0.9398343661518813, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057100801774012236, "300–1000 Hz_40ms_envelope_correlation": 0.9976453890071041, "300–1000 Hz_40ms_modulation_mae_db": 0.49423809357896237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2807603491741284, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13313811893205008, "1000–3000 Hz_10ms_envelope_correlation": 0.9873815397747362, "1000–3000 Hz_10ms_modulation_mae_db": 0.4997955958486475, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7476749181670139, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07337018206552527, "1000–3000 Hz_40ms_envelope_correlation": 0.9960117043678202, "1000–3000 Hz_40ms_modulation_mae_db": 0.7540449118260714, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5068044268364829, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2179215854470819, "3000–6000 Hz_10ms_envelope_correlation": 0.9695042934932583, "3000–6000 Hz_10ms_modulation_mae_db": 0.8733300085896522, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9780998839146713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14122960176940647, "3000–6000 Hz_40ms_envelope_correlation": 0.9865749074003584, "3000–6000 Hz_40ms_modulation_mae_db": 1.0956065437928855, "mr_stft_db": 5.772458647442023, "band_mae_db": 0.1443261843380745, "rms_gain_error_db": -0.04344694172916725, "output_peak": 0.49554664719373004}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0843769747294356, "全帯域_10ms_rms_envelope_nrmse": 0.0908871201426771, "全帯域_10ms_envelope_correlation": 0.9932709718054674, "全帯域_10ms_modulation_mae_db": 0.6040698638418411, "全帯域_40ms_log_rms_mae_db": 0.5974213321258485, "全帯域_40ms_rms_envelope_nrmse": 0.04304016341478716, "全帯域_40ms_envelope_correlation": 0.9984120955665325, "全帯域_40ms_modulation_mae_db": 0.49295204239035284, "300–1000 Hz_10ms_log_rms_mae_db": 1.401659320100619, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12266647837831175, "300–1000 Hz_10ms_envelope_correlation": 0.9893058563609333, "300–1000 Hz_10ms_modulation_mae_db": 2.2184924722690558, "300–1000 Hz_40ms_log_rms_mae_db": 0.9434701981511442, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04789574548505913, "300–1000 Hz_40ms_envelope_correlation": 0.9983029107611214, "300–1000 Hz_40ms_modulation_mae_db": 1.8478728880386013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.315473835702351, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12241320459522417, "1000–3000 Hz_10ms_envelope_correlation": 0.9893104226309958, "1000–3000 Hz_10ms_modulation_mae_db": 0.5889783912484096, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7793793879282483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0657458697006918, "1000–3000 Hz_40ms_envelope_correlation": 0.9968363378347616, "1000–3000 Hz_40ms_modulation_mae_db": 0.7223956594215624, "3000–6000 Hz_10ms_log_rms_mae_db": 1.495936606263939, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2303109490020265, "3000–6000 Hz_10ms_envelope_correlation": 0.9659736022866486, "3000–6000 Hz_10ms_modulation_mae_db": 0.4455235688798982, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9455032392192665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15364718036957545, "3000–6000 Hz_40ms_envelope_correlation": 0.9840466506719631, "3000–6000 Hz_40ms_modulation_mae_db": 0.7327630966684495, "mr_stft_db": 5.775480905559419, "band_mae_db": 0.07961527338040687, "rms_gain_error_db": -0.026006585450869483, "output_peak": 0.5033171152462784}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.988491074222662, "全帯域_10ms_rms_envelope_nrmse": 0.0810583044121375, "全帯域_10ms_envelope_correlation": 0.9946539022230777, "全帯域_10ms_modulation_mae_db": 0.16489570095403117, "全帯域_40ms_log_rms_mae_db": 0.5467146652555328, "全帯域_40ms_rms_envelope_nrmse": 0.04093352540197079, "全帯域_40ms_envelope_correlation": 0.9985550127864933, "全帯域_40ms_modulation_mae_db": 0.3605460188754992, "300–1000 Hz_10ms_log_rms_mae_db": 1.4286507414866276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08263350460981529, "300–1000 Hz_10ms_envelope_correlation": 0.9951720827840222, "300–1000 Hz_10ms_modulation_mae_db": 0.9732433469138221, "300–1000 Hz_40ms_log_rms_mae_db": 0.9598266780232856, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04002676296563095, "300–1000 Hz_40ms_envelope_correlation": 0.9988741431428472, "300–1000 Hz_40ms_modulation_mae_db": 0.9265386133862226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2881409993142194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15105240254288446, "1000–3000 Hz_10ms_envelope_correlation": 0.9839915039882359, "1000–3000 Hz_10ms_modulation_mae_db": 0.7033129334325187, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7410007008856561, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08560016286427846, "1000–3000 Hz_40ms_envelope_correlation": 0.9946615156546249, "1000–3000 Hz_40ms_modulation_mae_db": 0.5125855874293457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4809900289730544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2131416564468992, "3000–6000 Hz_10ms_envelope_correlation": 0.9712478335950027, "3000–6000 Hz_10ms_modulation_mae_db": 0.06285670283235295, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9414019917786973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13308359571364964, "3000–6000 Hz_40ms_envelope_correlation": 0.9881444975293273, "3000–6000 Hz_40ms_modulation_mae_db": 0.08326313594049804, "mr_stft_db": 5.780500773743836, "band_mae_db": 0.06409041557367613, "rms_gain_error_db": -0.0035568408030341794, "output_peak": 0.5258428839225249}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0281162046284025, "全帯域_10ms_rms_envelope_nrmse": 0.07449258389950258, "全帯域_10ms_envelope_correlation": 0.9954990846856951, "全帯域_10ms_modulation_mae_db": 0.2914266139452915, "全帯域_40ms_log_rms_mae_db": 0.5363334629763608, "全帯域_40ms_rms_envelope_nrmse": 0.04085878420168218, "全帯域_40ms_envelope_correlation": 0.9985992179767152, "全帯域_40ms_modulation_mae_db": 0.23334820837171, "300–1000 Hz_10ms_log_rms_mae_db": 1.3560936937212638, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10555677238351205, "300–1000 Hz_10ms_envelope_correlation": 0.9921385727056019, "300–1000 Hz_10ms_modulation_mae_db": 1.4356987758478432, "300–1000 Hz_40ms_log_rms_mae_db": 0.9075288513212629, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06397743283904757, "300–1000 Hz_40ms_envelope_correlation": 0.9971298803201456, "300–1000 Hz_40ms_modulation_mae_db": 1.1686749574362214, "1000–3000 Hz_10ms_log_rms_mae_db": 1.295910987140692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10216683949946964, "1000–3000 Hz_10ms_envelope_correlation": 0.9925746425894594, "1000–3000 Hz_10ms_modulation_mae_db": 0.2582511098101581, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7731377404063943, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05568567869605499, "1000–3000 Hz_40ms_envelope_correlation": 0.9977088749751527, "1000–3000 Hz_40ms_modulation_mae_db": 0.30681990472581494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.499250062743252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1971922794763904, "3000–6000 Hz_10ms_envelope_correlation": 0.9753439746477909, "3000–6000 Hz_10ms_modulation_mae_db": 0.25774755914715414, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9329745087804249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1253614223357883, "3000–6000 Hz_40ms_envelope_correlation": 0.9894541909363997, "3000–6000 Hz_40ms_modulation_mae_db": 0.4704567459063816, "mr_stft_db": 5.763990015945725, "band_mae_db": 0.15050881622837842, "rms_gain_error_db": -0.06543103679092627, "output_peak": 0.5382612436674523}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83120, "output_samples": 83120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5592132196126895, "spectral_convergence": 0.3067745924700744, "spectral_convergence_status": "ok", "amplitude_floor": 8.933262458231008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.542728299986408, "spectral_convergence": 0.32190050641691326, "spectral_convergence_status": "ok", "amplitude_floor": 6.540934930277787e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.186503221506761, "spectral_convergence": 0.33770604197192605, "spectral_convergence_status": "ok", "amplitude_floor": 6.044182898288457e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.147025669043969e-05, "output_power": 3.239535193743276e-05, "delta_db": 0.12582417110178573, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00012755476455897642, "output_power": 0.0001308505791150838, "delta_db": 0.11078963598844394, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005321662836872518, "output_power": 0.0005189906312203709, "delta_db": -0.10887837542962145, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022563854189948138, "output_power": 0.00021504413637056684, "delta_db": -0.20885679503212154, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.229072811272528e-05, "output_power": 8.779400168127657e-05, "delta_db": 0.2811393976863976, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.381343402576783e-05, "output_power": 6.053606323343698e-05, "delta_db": -0.22897941562225, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17741129847677004, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07136257451027513, "pitch": null, "output_rms_dbfs": -29.806299547600887, "output_peak": 0.5755080039895948, "reference_rms_dbfs": -29.734936973090612, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.762814913701953}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.00015201529585382603, "log_rms_mae_db": 1.0329522295372924, "rms_envelope_nrmse": 0.07961685000574302, "envelope_correlation": 0.9948608186148465, "rms_gain_error_db": -0.07136263807048014, "reference_mean_power": 0.0010639579210132953, "output_mean_power": 0.0010466179729972987}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.575807359242935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021400202307720183, "output_power": 0.00021354860825852282, "reference_active": true, "delta_db": -0.009211336068923064}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3329698566280647e-05, "output_power": 2.2705025700018174e-05, "reference_active": true, "delta_db": -0.11787129665416189}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.358098056371455e-06, "output_power": 6.0277477405130765e-06, "reference_active": true, "delta_db": 0.5114441173390201}], "active_band_count": 3, "active_band_mae_db": 0.21284225002070167}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.0001490763524181648, "log_rms_mae_db": 0.5522950432502546, "rms_envelope_nrmse": 0.041887274036175634, "envelope_correlation": 0.9985476285529108, "rms_gain_error_db": -0.07136317551592179, "reference_mean_power": 0.0010701430709341084, "output_mean_power": 0.0010527021896186326}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 5.309268989226658e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019126377610844124, "output_power": 0.0001918928000317835, "reference_active": true, "delta_db": 0.014259541990177833}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0962051121960504e-05, "output_power": 1.0942255067401726e-05, "reference_active": true, "delta_db": -0.007849889902632964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.152513485967136e-07, "output_power": 3.6360384248116396e-07, "reference_active": true, "delta_db": 0.6197151100364796}], "active_band_count": 3, "active_band_mae_db": 0.21394151397643013}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.0001087945365454221, "log_rms_mae_db": 1.3907394809245477, "rms_envelope_nrmse": 0.09924424739141821, "envelope_correlation": 0.9930254074764885, "rms_gain_error_db": -0.10279692575379748, "reference_mean_power": 0.0005739909463294839, "output_mean_power": 0.0005605641886898825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 3.2669273354505384e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013940368795916017, "output_power": 0.00013743980118556518, "reference_active": true, "delta_db": -0.0616174502614101}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.236537072685246e-06, "output_power": 8.228870626917353e-06, "reference_active": true, "delta_db": -0.5016894323693654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7105571121059603e-06, "output_power": 2.7766480920297285e-06, "reference_active": true, "delta_db": 2.103832624511891}], "active_band_count": 3, "active_band_mae_db": 0.8890465023808888}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.00010533749157464783, "log_rms_mae_db": 0.918320954757808, "rms_envelope_nrmse": 0.04945734342166605, "envelope_correlation": 0.9982883139945044, "rms_gain_error_db": -0.10279691285522188, "reference_mean_power": 0.0005773280667275119, "output_mean_power": 0.0005638232490664007}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 3.141212974645643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001274169753045263, "output_power": 0.000125986511698834, "reference_active": true, "delta_db": -0.04903239942270593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.860144772024442e-06, "output_power": 3.3937386107360787e-06, "reference_active": true, "delta_db": -0.559252034285729}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1444270061935373e-07, "output_power": 1.6032211456693507e-07, "reference_active": true, "delta_db": 1.4640533461042855}], "active_band_count": 3, "active_band_mae_db": 0.6907792599375734}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 8.389504112280413e-05, "log_rms_mae_db": 1.2766091304349896, "rms_envelope_nrmse": 0.10771260293468372, "envelope_correlation": 0.9918050172437375, "rms_gain_error_db": -0.13588595620360888, "reference_mean_power": 0.00021157950445783077, "output_mean_power": 0.0002050619108606558}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.2662436125548267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5392735637289177e-05, "output_power": 5.2405631744159085e-05, "reference_active": true, "delta_db": -0.2407485311098787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790041823950801e-06, "output_power": 5.472797250574101e-06, "reference_active": true, "delta_db": -0.2447234159820384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4111913003034675e-06, "output_power": 1.293754455788257e-06, "reference_active": true, "delta_db": -0.37734031915542815}], "active_band_count": 3, "active_band_mae_db": 0.2876040887491151}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 7.722064559288104e-05, "log_rms_mae_db": 0.7572799729013562, "rms_envelope_nrmse": 0.05328716247340266, "envelope_correlation": 0.9980652430401693, "rms_gain_error_db": -0.13588612315642012, "reference_mean_power": 0.0002128095577406816, "output_mean_power": 0.00020625406508040998}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 1.2078803728484267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0224253455435505e-05, "output_power": 4.754002500713965e-05, "reference_active": true, "delta_db": -0.2385408417666412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7667526770926532e-06, "output_power": 2.6458371334519226e-06, "reference_active": true, "delta_db": -0.19407231497328514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.983894982848384e-08, "output_power": 8.459213503304738e-08, "reference_active": true, "delta_db": -0.7197001784954714}], "active_band_count": 3, "active_band_mae_db": 0.3841044450784659}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.54681310079566e-05, "log_rms_mae_db": 1.514863757068072, "rms_envelope_nrmse": 0.2317057861168755, "envelope_correlation": 0.9656112092157441, "rms_gain_error_db": -0.09156145472754494, "reference_mean_power": 0.00010955764718773985, "output_mean_power": 0.0001072720429098081}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 6.937978158746457e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2971450195505076e-05, "output_power": 2.109108831428538e-05, "reference_active": true, "delta_db": -0.37089423015745737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1375769637538917e-05, "output_power": 1.156112836352052e-05, "reference_active": true, "delta_db": 0.07019434304462073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4470462261987434e-06, "output_power": 3.3518733305032147e-06, "reference_active": true, "delta_db": 1.366454244768909}], "active_band_count": 3, "active_band_mae_db": 0.6025142726569958}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 5.762149573106217e-05, "log_rms_mae_db": 0.9550175883982785, "rms_envelope_nrmse": 0.15891921620305974, "envelope_correlation": 0.9829210346434841, "rms_gain_error_db": -0.09156158116593523, "reference_mean_power": 0.00011019456862153096, "output_mean_power": 0.00010789567367392812}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 6.599778128222787e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3073624841125754e-05, "output_power": 2.231966545330521e-05, "reference_active": true, "delta_db": -0.1442814633882891}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.235055585786147e-06, "output_power": 6.345715033364117e-06, "reference_active": true, "delta_db": 0.0764023578099272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9515593840599056e-07, "output_power": 2.582918515481924e-07, "reference_active": true, "delta_db": 1.2172893452336626}], "active_band_count": 3, "active_band_mae_db": 0.4793243888106263}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9117448677924884, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2701d82314f2bd673f80d3d3f4b3f50c86a3677e6b0d1f0b9722df314512ff6e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897859265684861, "standard_deviation": 0.002883859507343534, "minimum": 0.9839915039882359, "maximum": 0.9925746425894594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.293993204520474, "standard_deviation": 0.022354166035566683, "minimum": 1.267235947419948, "maximum": 1.33640406270277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4328805688082123, "standard_deviation": 0.16538140392447453, "minimum": 0.23979474079135987, "maximum": 0.7033129334325187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11900595501421155, "standard_deviation": 0.016369307561856377, "minimum": 0.10216683949946964, "maximum": 0.15105240254288446, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99709225901191, "standard_deviation": 0.001214873592315281, "minimum": 0.9946615156546249, "maximum": 0.9982016307178526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7582118953568149, "standard_deviation": 0.012785794941374948, "minimum": 0.7410007008856561, "maximum": 0.7793793879282483, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5751883281581031, "standard_deviation": 0.25919893733129024, "minimum": 0.2215306092479261, "maximum": 0.9684282757937707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06225138646559735, "standard_deviation": 0.012128044932578616, "minimum": 0.049494734755090225, "maximum": 0.08560016286427846, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9696400834276795, "standard_deviation": 0.003632762969999096, "minimum": 0.9655632524946824, "maximum": 0.9753439746477909, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4974142661313667, "standard_deviation": 0.01145144235716531, "minimum": 1.4809900289730544, "maximum": 1.514863757068072, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.39273105585547163, "standard_deviation": 0.2592520080779012, "minimum": 0.06285670283235295, "maximum": 0.8733300085896522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21763806550358294, "standard_deviation": 0.012841630620457302, "minimum": 0.1971922794763904, "maximum": 0.23225287400314135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9866528781051419, "standard_deviation": 0.002170000763931587, "minimum": 0.9829210346434841, "maximum": 0.9894541909363997, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9412771749974619, "standard_deviation": 0.02064450038477229, "minimum": 0.9069227224892885, "maximum": 0.9780998839146713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5348030275730563, "standard_deviation": 0.3285057875695916, "minimum": 0.08326313594049804, "maximum": 1.0956065437928855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14036784973091976, "standard_deviation": 0.01105691285395202, "minimum": 0.1253614223357883, "maximum": 0.15891921620305974, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9923487816560264, "standard_deviation": 0.0029391297035029763, "minimum": 0.9867529180746702, "maximum": 0.9951720827840222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3926194067233286, "standard_deviation": 0.022981971111806675, "minimum": 1.3560936937212638, "maximum": 1.4286507414866276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.2950480611355877, "standard_deviation": 0.5714121832831797, "minimum": 0.8064881815239913, "maximum": 2.2184924722690558, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.102662811825677, "standard_deviation": 0.01914436283223067, "minimum": 0.08263350460981529, "maximum": 0.1379336993795061, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981136493614337, "standard_deviation": 0.0006701227383805937, "minimum": 0.9971298803201456, "maximum": 0.9988741431428472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9425863052046397, "standard_deviation": 0.023953954979674383, "minimum": 0.9075288513212629, "maximum": 0.9771390874684034, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9984735047800128, "standard_deviation": 0.5486008391160883, "minimum": 0.43544511800851665, "maximum": 1.8478728880386013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05107450711831128, "standard_deviation": 0.008889317106286283, "minimum": 0.04002676296563095, "maximum": 0.06397743283904757, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13112401331408013, "standard_deviation": 0.039890143348310904, "minimum": 0.06409041557367613, "maximum": 0.17741129847677004, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.76963236866107, "standard_deviation": 0.008616807207576564, "minimum": 5.756463753928091, "maximum": 5.780500773743836, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5293303910352325, "standard_deviation": 0.031287189503666546, "minimum": 0.49554664719373004, "maximum": 0.5755080039895948, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0374049491631148, "standard_deviation": 0.03315469095014849, "minimum": -0.07680301378201157, "maximum": 0.015577926443488268, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946666919219465, "standard_deviation": 0.0010477095242267984, "minimum": 0.9928324840583826, "maximum": 0.9954990846856951, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.036423920734341, "standard_deviation": 0.02952673206305501, "minimum": 0.988491074222662, "maximum": 1.0843769747294356, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35773057213928317, "standard_deviation": 0.2029295160719278, "minimum": 0.0857983625932195, "maximum": 0.6040698638418411, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08072322694318368, "standard_deviation": 0.0076849067792501774, "minimum": 0.07449258389950258, "maximum": 0.09409257718284987, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985984591554482, "standard_deviation": 0.00028081208321584954, "minimum": 0.9980866347758339, "maximum": 0.9990010086002113, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5590965848394116, "standard_deviation": 0.02969783553952658, "minimum": 0.5318164355582388, "maximum": 0.6133290442685366, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35123384491697823, "standard_deviation": 0.1436287641487287, "minimum": 0.12278947623496482, "maximum": 0.4978261463048983, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04053704043449813, "standard_deviation": 0.003895468533130764, "minimum": 0.03500640880258943, "maximum": 0.0473367711588417, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0065", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.27384352684021, "predicted_mos": 2.27384352684021, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "7d497b0de450d77c5adc60f44434a82896f589fe83d6e5220a35bf178fd0045d", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.893732599914074, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.042539310618311, "全帯域_10ms_rms_envelope_nrmse": 0.07720946180531159, "全帯域_10ms_envelope_correlation": 0.995171263180057, "全帯域_10ms_modulation_mae_db": 0.2985319957364271, "全帯域_40ms_log_rms_mae_db": 0.5454902132916837, "全帯域_40ms_rms_envelope_nrmse": 0.03824224162888772, "全帯域_40ms_envelope_correlation": 0.9988015613520472, "全帯域_40ms_modulation_mae_db": 0.2539696790141707, "300–1000 Hz_10ms_log_rms_mae_db": 1.2289452069635785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10064315871164049, "300–1000 Hz_10ms_envelope_correlation": 0.992800992641825, "300–1000 Hz_10ms_modulation_mae_db": 1.0388268093250763, "300–1000 Hz_40ms_log_rms_mae_db": 0.7619635549766582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04809733713984722, "300–1000 Hz_40ms_envelope_correlation": 0.9983426464190439, "300–1000 Hz_40ms_modulation_mae_db": 0.8224811711038061, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2452132333547643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10901783263033227, "1000–3000 Hz_10ms_envelope_correlation": 0.9917036145843784, "1000–3000 Hz_10ms_modulation_mae_db": 0.19213710249421853, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6998907874475598, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05568115073274424, "1000–3000 Hz_40ms_envelope_correlation": 0.9980132132230161, "1000–3000 Hz_40ms_modulation_mae_db": 0.3737490180689244, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4956245221510274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20899824880217016, "3000–6000 Hz_10ms_envelope_correlation": 0.9719981919732, "3000–6000 Hz_10ms_modulation_mae_db": 0.5933336336345555, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9368315616318283, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1389488805026005, "3000–6000 Hz_40ms_envelope_correlation": 0.9869881641780568, "3000–6000 Hz_40ms_modulation_mae_db": 0.4233900919982705, "mr_stft_db": 5.698208494296646, "band_mae_db": 0.14833754483623382, "rms_gain_error_db": -0.073466447755048, "output_peak": 0.5839702133070274}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0841848185104161, "全帯域_10ms_rms_envelope_nrmse": 0.08081504442779447, "全帯域_10ms_envelope_correlation": 0.9946810769321907, "全帯域_10ms_modulation_mae_db": 0.348140187044943, "全帯域_40ms_log_rms_mae_db": 0.6371483934658537, "全帯域_40ms_rms_envelope_nrmse": 0.0381505893546663, "全帯域_40ms_envelope_correlation": 0.9987447236496417, "全帯域_40ms_modulation_mae_db": 0.49147324012227084, "300–1000 Hz_10ms_log_rms_mae_db": 1.1908813137799372, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08560549200475161, "300–1000 Hz_10ms_envelope_correlation": 0.9947942857586037, "300–1000 Hz_10ms_modulation_mae_db": 0.9058710823047221, "300–1000 Hz_40ms_log_rms_mae_db": 0.7454463025654291, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03961000023503606, "300–1000 Hz_40ms_envelope_correlation": 0.998848209879298, "300–1000 Hz_40ms_modulation_mae_db": 0.3554291209871346, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3281762930812966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10803868328959856, "1000–3000 Hz_10ms_envelope_correlation": 0.9916808864534087, "1000–3000 Hz_10ms_modulation_mae_db": 0.604270794375637, "1000–3000 Hz_40ms_log_rms_mae_db": 0.73532654822645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049872792349525324, "1000–3000 Hz_40ms_envelope_correlation": 0.9982008131576939, "1000–3000 Hz_40ms_modulation_mae_db": 1.140816467253176, "3000–6000 Hz_10ms_log_rms_mae_db": 1.486796773830528, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20685604181837955, "3000–6000 Hz_10ms_envelope_correlation": 0.9725888077143438, "3000–6000 Hz_10ms_modulation_mae_db": 0.4751213372411797, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9420082818777201, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12964194738200618, "3000–6000 Hz_40ms_envelope_correlation": 0.9886804368471459, "3000–6000 Hz_40ms_modulation_mae_db": 0.8197555946468661, "mr_stft_db": 5.70251941698654, "band_mae_db": 0.12129045992125354, "rms_gain_error_db": -0.009398145196849466, "output_peak": 0.5218973691742274}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0484687598334526, "全帯域_10ms_rms_envelope_nrmse": 0.07685502270775858, "全帯域_10ms_envelope_correlation": 0.9952374152768154, "全帯域_10ms_modulation_mae_db": 0.013604631893340192, "全帯域_40ms_log_rms_mae_db": 0.5408883247119989, "全帯域_40ms_rms_envelope_nrmse": 0.03586378219425274, "全帯域_40ms_envelope_correlation": 0.9989792403263411, "全帯域_40ms_modulation_mae_db": 0.04276716414533436, "300–1000 Hz_10ms_log_rms_mae_db": 1.1873900673602817, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09487782237246722, "300–1000 Hz_10ms_envelope_correlation": 0.9936881192394343, "300–1000 Hz_10ms_modulation_mae_db": 1.0189380620014128, "300–1000 Hz_40ms_log_rms_mae_db": 0.7717081314751988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04720909957938858, "300–1000 Hz_40ms_envelope_correlation": 0.9985209995474857, "300–1000 Hz_40ms_modulation_mae_db": 0.6092520467225925, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2583804655335369, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10666499785685295, "1000–3000 Hz_10ms_envelope_correlation": 0.9919275519530734, "1000–3000 Hz_10ms_modulation_mae_db": 0.10580841644517534, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7200263707716513, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05498909654568583, "1000–3000 Hz_40ms_envelope_correlation": 0.9978531319701195, "1000–3000 Hz_40ms_modulation_mae_db": 0.7559562711243427, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5100047445052407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2152766630235689, "3000–6000 Hz_10ms_envelope_correlation": 0.9703008403737313, "3000–6000 Hz_10ms_modulation_mae_db": 0.2015218710404761, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9018463241464262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1239095570417557, "3000–6000 Hz_40ms_envelope_correlation": 0.989684344937523, "3000–6000 Hz_40ms_modulation_mae_db": 0.1810194809747833, "mr_stft_db": 5.686995799975001, "band_mae_db": 0.10751186687812504, "rms_gain_error_db": -0.09344322597177687, "output_peak": 0.55918345297485}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0747991305280555, "全帯域_10ms_rms_envelope_nrmse": 0.09872704044782392, "全帯域_10ms_envelope_correlation": 0.9920949700328234, "全帯域_10ms_modulation_mae_db": 0.7212926555997415, "全帯域_40ms_log_rms_mae_db": 0.5636787031131255, "全帯域_40ms_rms_envelope_nrmse": 0.04812096232556475, "全帯域_40ms_envelope_correlation": 0.9980129654835246, "全帯域_40ms_modulation_mae_db": 0.5342825623918388, "300–1000 Hz_10ms_log_rms_mae_db": 1.2400400076480376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13998949754847426, "300–1000 Hz_10ms_envelope_correlation": 0.9864354765978259, "300–1000 Hz_10ms_modulation_mae_db": 2.091836120426774, "300–1000 Hz_40ms_log_rms_mae_db": 0.7873571394984743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06401381237314649, "300–1000 Hz_40ms_envelope_correlation": 0.9971395256938319, "300–1000 Hz_40ms_modulation_mae_db": 1.724479184802723, "1000–3000 Hz_10ms_log_rms_mae_db": 1.276269671391018, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11798333735943425, "1000–3000 Hz_10ms_envelope_correlation": 0.9903597556555213, "1000–3000 Hz_10ms_modulation_mae_db": 0.4559620188632129, "1000–3000 Hz_40ms_log_rms_mae_db": 0.699445997136947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06914377338462271, "1000–3000 Hz_40ms_envelope_correlation": 0.9969090480364481, "1000–3000 Hz_40ms_modulation_mae_db": 0.35233157296640644, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5132776158378578, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20864914464459405, "3000–6000 Hz_10ms_envelope_correlation": 0.9721459606381642, "3000–6000 Hz_10ms_modulation_mae_db": 0.2510608861595138, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9111138599782682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1367439629264667, "3000–6000 Hz_40ms_envelope_correlation": 0.9873829256843709, "3000–6000 Hz_40ms_modulation_mae_db": 0.47264153909161416, "mr_stft_db": 5.680689223913261, "band_mae_db": 0.18405035461879957, "rms_gain_error_db": 0.005101032452408528, "output_peak": 0.5054843917517651}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0039180078816488, "全帯域_10ms_rms_envelope_nrmse": 0.07497569802719865, "全帯域_10ms_envelope_correlation": 0.9954270200270792, "全帯域_10ms_modulation_mae_db": 0.3987590687813434, "全帯域_40ms_log_rms_mae_db": 0.5580709329172268, "全帯域_40ms_rms_envelope_nrmse": 0.0374975583318558, "全帯域_40ms_envelope_correlation": 0.9988028227259539, "全帯域_40ms_modulation_mae_db": 0.4134985831227582, "300–1000 Hz_10ms_log_rms_mae_db": 1.1788979856579116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09752946116652954, "300–1000 Hz_10ms_envelope_correlation": 0.993241392224047, "300–1000 Hz_10ms_modulation_mae_db": 0.9422094462447506, "300–1000 Hz_40ms_log_rms_mae_db": 0.7544037079905217, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051080467306330576, "300–1000 Hz_40ms_envelope_correlation": 0.9980820922122684, "300–1000 Hz_40ms_modulation_mae_db": 0.5965609010029504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.28017867312632, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12062549099959338, "1000–3000 Hz_10ms_envelope_correlation": 0.9896422644961991, "1000–3000 Hz_10ms_modulation_mae_db": 0.483670850543408, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7417974484781646, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06594510418231919, "1000–3000 Hz_40ms_envelope_correlation": 0.9968381374126799, "1000–3000 Hz_40ms_modulation_mae_db": 0.5421236114242759, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4995772422797713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2115515452364322, "3000–6000 Hz_10ms_envelope_correlation": 0.9712714001588855, "3000–6000 Hz_10ms_modulation_mae_db": 0.885053804465894, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9542223426043884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1334733160450479, "3000–6000 Hz_40ms_envelope_correlation": 0.9880461713586833, "3000–6000 Hz_40ms_modulation_mae_db": 1.1523186349291217, "mr_stft_db": 5.702590758646132, "band_mae_db": 0.1319341127095134, "rms_gain_error_db": -0.032001772358685815, "output_peak": 0.4923084372318775}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0975494645093278, "全帯域_10ms_rms_envelope_nrmse": 0.09238701715236819, "全帯域_10ms_envelope_correlation": 0.99305355676143, "全帯域_10ms_modulation_mae_db": 0.6656436618475543, "全帯域_40ms_log_rms_mae_db": 0.6106007643924138, "全帯域_40ms_rms_envelope_nrmse": 0.04351356288316288, "全帯域_40ms_envelope_correlation": 0.9983758794485805, "全帯域_40ms_modulation_mae_db": 0.5849083385076249, "300–1000 Hz_10ms_log_rms_mae_db": 1.254181680571897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11985422995829947, "300–1000 Hz_10ms_envelope_correlation": 0.9898176966197517, "300–1000 Hz_10ms_modulation_mae_db": 2.162624355406018, "300–1000 Hz_40ms_log_rms_mae_db": 0.8403575159479478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052170337057321256, "300–1000 Hz_40ms_envelope_correlation": 0.9979900245812644, "300–1000 Hz_40ms_modulation_mae_db": 1.8248459542583888, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3122289023939324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12161352596167846, "1000–3000 Hz_10ms_envelope_correlation": 0.9894519672354564, "1000–3000 Hz_10ms_modulation_mae_db": 0.6172983860795306, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7580770368029716, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0597899492636706, "1000–3000 Hz_40ms_envelope_correlation": 0.99739210507152, "1000–3000 Hz_40ms_modulation_mae_db": 0.6705586909545115, "3000–6000 Hz_10ms_log_rms_mae_db": 1.504730709131591, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20664202834404086, "3000–6000 Hz_10ms_envelope_correlation": 0.9726359804715098, "3000–6000 Hz_10ms_modulation_mae_db": 0.5277911765421076, "3000–6000 Hz_40ms_log_rms_mae_db": 0.943665473869386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13069735681680625, "3000–6000 Hz_40ms_envelope_correlation": 0.988509319679205, "3000–6000 Hz_40ms_modulation_mae_db": 0.8735101422142065, "mr_stft_db": 5.705685323463133, "band_mae_db": 0.08956446698607727, "rms_gain_error_db": -0.015513992310158889, "output_peak": 0.5104393189989074}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0090029704411692, "全帯域_10ms_rms_envelope_nrmse": 0.08016284381526664, "全帯域_10ms_envelope_correlation": 0.9947653667713835, "全帯域_10ms_modulation_mae_db": 0.20454662256202316, "全帯域_40ms_log_rms_mae_db": 0.5663969282639005, "全帯域_40ms_rms_envelope_nrmse": 0.04012711219480143, "全帯域_40ms_envelope_correlation": 0.9986256236020871, "全帯域_40ms_modulation_mae_db": 0.3511620618238334, "300–1000 Hz_10ms_log_rms_mae_db": 1.246421776471971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09355213442874781, "300–1000 Hz_10ms_envelope_correlation": 0.9938006996825189, "300–1000 Hz_10ms_modulation_mae_db": 1.059976403087861, "300–1000 Hz_40ms_log_rms_mae_db": 0.7783999937647911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047170804908274605, "300–1000 Hz_40ms_envelope_correlation": 0.9984327237232549, "300–1000 Hz_40ms_modulation_mae_db": 0.7861896443846269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2882263223303014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1279804686708195, "1000–3000 Hz_10ms_envelope_correlation": 0.9883010309459317, "1000–3000 Hz_10ms_modulation_mae_db": 0.5337470906608944, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7141095661183077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06558239942556365, "1000–3000 Hz_40ms_envelope_correlation": 0.9968065838327143, "1000–3000 Hz_40ms_modulation_mae_db": 0.692069781859702, "3000–6000 Hz_10ms_log_rms_mae_db": 1.45883234599555, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20587339978501934, "3000–6000 Hz_10ms_envelope_correlation": 0.9731697202527464, "3000–6000 Hz_10ms_modulation_mae_db": 0.09335366717151487, "3000–6000 Hz_40ms_log_rms_mae_db": 0.920760124709867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12501242489614908, "3000–6000 Hz_40ms_envelope_correlation": 0.9895438889864758, "3000–6000 Hz_40ms_modulation_mae_db": 0.1516431551974341, "mr_stft_db": 5.705994184031574, "band_mae_db": 0.07816142970466998, "rms_gain_error_db": -0.04476273107794855, "output_peak": 0.5256001390308275}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0432826286340227, "全帯域_10ms_rms_envelope_nrmse": 0.07187920552478638, "全帯域_10ms_envelope_correlation": 0.9957983503959431, "全帯域_10ms_modulation_mae_db": 0.20145998433800885, "全帯域_40ms_log_rms_mae_db": 0.5478443200877281, "全帯域_40ms_rms_envelope_nrmse": 0.038973121493814926, "全帯域_40ms_envelope_correlation": 0.9987082914224793, "全帯域_40ms_modulation_mae_db": 0.14625138208818214, "300–1000 Hz_10ms_log_rms_mae_db": 1.1840024870185746, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10059457409573896, "300–1000 Hz_10ms_envelope_correlation": 0.9928335575718691, "300–1000 Hz_10ms_modulation_mae_db": 1.3452137372304396, "300–1000 Hz_40ms_log_rms_mae_db": 0.747313688904009, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06025456728660035, "300–1000 Hz_40ms_envelope_correlation": 0.9974092082788512, "300–1000 Hz_40ms_modulation_mae_db": 1.1439702814739392, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2779043849032465, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09813290503249049, "1000–3000 Hz_10ms_envelope_correlation": 0.9931717188365632, "1000–3000 Hz_10ms_modulation_mae_db": 0.18461487252654582, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7231593366416034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05050744431625798, "1000–3000 Hz_40ms_envelope_correlation": 0.9981330825964326, "1000–3000 Hz_40ms_modulation_mae_db": 0.2858079302793261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.485345261229025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18833562274546245, "3000–6000 Hz_10ms_envelope_correlation": 0.9775923204857412, "3000–6000 Hz_10ms_modulation_mae_db": 0.3586059565017686, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9222512307946247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11497496858271726, "3000–6000 Hz_40ms_envelope_correlation": 0.9911675690495677, "3000–6000 Hz_40ms_modulation_mae_db": 0.6320378860459271, "mr_stft_db": 5.699601186933191, "band_mae_db": 0.13526820619926083, "rms_gain_error_db": -0.045322702972413254, "output_peak": 0.537670865955579}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83120, "output_samples": 83120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5228446707623755, "spectral_convergence": 0.30362784676120136, "spectral_convergence_status": "ok", "amplitude_floor": 8.933262458231008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.481824425715364, "spectral_convergence": 0.3206956946094241, "spectral_convergence_status": "ok", "amplitude_floor": 6.540934930277787e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0899563864122, "spectral_convergence": 0.3339753606122469, "spectral_convergence_status": "ok", "amplitude_floor": 6.044182898288457e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.147025669043969e-05, "output_power": 3.1804301209602524e-05, "delta_db": 0.04585572590191163, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00012755476455897642, "output_power": 0.0001308035037129904, "delta_db": 0.10922691719864694, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005321662836872518, "output_power": 0.0005232374121450448, "delta_db": -0.0734856657299146, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022563854189948138, "output_power": 0.00021322140192646697, "delta_db": -0.24582490103557428, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.229072811272528e-05, "output_power": 8.566821618588879e-05, "delta_db": 0.1746881895880878, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.381343402576783e-05, "output_power": 6.036952089432338e-05, "delta_db": -0.24094386956326766, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14833754483623382, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.073466447755048, "pitch": null, "output_rms_dbfs": -29.808403420845657, "output_peak": 0.5839702133070274, "reference_rms_dbfs": -29.734936973090612, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.698208494296646}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.00015201529585382603, "log_rms_mae_db": 1.042539310618311, "rms_envelope_nrmse": 0.07720946180531159, "envelope_correlation": 0.995171263180057, "rms_gain_error_db": -0.07346654318321161, "reference_mean_power": 0.0010639579210132953, "output_mean_power": 0.0010461110700281681}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.575807359242935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021400202307720183, "output_power": 0.00021262021950013406, "reference_active": true, "delta_db": -0.02813316828075501}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3329698566280647e-05, "output_power": 2.2447116602247782e-05, "reference_active": true, "delta_db": -0.16748564958293805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.358098056371455e-06, "output_power": 6.295183211991265e-06, "reference_active": true, "delta_db": 0.6999771693455882}], "active_band_count": 3, "active_band_mae_db": 0.2985319957364271}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.0001490763524181648, "log_rms_mae_db": 0.5454902132916837, "rms_envelope_nrmse": 0.03824224162888772, "envelope_correlation": 0.9988015613520472, "rms_gain_error_db": -0.0734669040481946, "reference_mean_power": 0.0010701430709341084, "output_mean_power": 0.0010521923826941689}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 5.309268989226658e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019126377610844124, "output_power": 0.0001910295715544329, "reference_active": true, "delta_db": -0.005321241334132155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0962051121960504e-05, "output_power": 1.079369113381836e-05, "reference_active": true, "delta_db": -0.06721836709561764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.152513485967136e-07, "output_power": 3.694825191101865e-07, "reference_active": true, "delta_db": 0.6893694286127623}], "active_band_count": 3, "active_band_mae_db": 0.2539696790141707}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.0001087945365454221, "log_rms_mae_db": 1.2289452069635785, "rms_envelope_nrmse": 0.10064315871164049, "envelope_correlation": 0.992800992641825, "rms_gain_error_db": -0.08036818097762369, "reference_mean_power": 0.0005739909463294839, "output_mean_power": 0.0005634666599440995}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 3.2669273354505384e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013940368795916017, "output_power": 0.00013803514552620407, "reference_active": true, "delta_db": -0.042845858318543095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.236537072685246e-06, "output_power": 8.133041572911913e-06, "reference_active": true, "delta_db": -0.5525618545289783}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7105571121059603e-06, "output_power": 3.0566439301275835e-06, "reference_active": true, "delta_db": 2.5210727151277075}], "active_band_count": 3, "active_band_mae_db": 1.0388268093250763}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.00010533749157464783, "log_rms_mae_db": 0.7619635549766582, "rms_envelope_nrmse": 0.04809733713984722, "envelope_correlation": 0.9983426464190439, "rms_gain_error_db": -0.0803681600524126, "reference_mean_power": 0.0005773280667275119, "output_mean_power": 0.000566742596026228}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 3.141212974645643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001274169753045263, "output_power": 0.00012652410979058386, "reference_active": true, "delta_db": -0.030540009984944383}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.860144772024442e-06, "output_power": 3.3468788998425122e-06, "reference_active": true, "delta_db": -0.6196359432752224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1444270061935373e-07, "output_power": 1.739061152106053e-07, "reference_active": true, "delta_db": 1.8172675600512518}], "active_band_count": 3, "active_band_mae_db": 0.8224811711038061}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 8.389504112280413e-05, "log_rms_mae_db": 1.2452132333547643, "rms_envelope_nrmse": 0.10901783263033227, "envelope_correlation": 0.9917036145843784, "rms_gain_error_db": -0.18573968177177103, "reference_mean_power": 0.00021157950445783077, "output_mean_power": 0.0002027214143088805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.2662436125548267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5392735637289177e-05, "output_power": 5.1695067226205836e-05, "reference_active": true, "delta_db": -0.30003709399483447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790041823950801e-06, "output_power": 5.590639824949865e-06, "reference_active": true, "delta_db": -0.15220186946619008}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4111913003034675e-06, "output_power": 1.3714142445338052e-06, "reference_active": true, "delta_db": -0.12417234402163106}], "active_band_count": 3, "active_band_mae_db": 0.19213710249421853}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 7.722064559288104e-05, "log_rms_mae_db": 0.6998907874475598, "rms_envelope_nrmse": 0.05568115073274424, "envelope_correlation": 0.9980132132230161, "rms_gain_error_db": -0.18573985312884692, "reference_mean_power": 0.0002128095577406816, "output_mean_power": 0.0002038999615392563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 1.2078803728484267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0224253455435505e-05, "output_power": 4.6939501512155445e-05, "reference_active": true, "delta_db": -0.29375016848652674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7667526770926532e-06, "output_power": 2.678806884282681e-06, "reference_active": true, "delta_db": -0.1402893247764187}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.983894982848384e-08, "output_power": 8.522740162616819e-08, "reference_active": true, "delta_db": -0.6872075609438276}], "active_band_count": 3, "active_band_mae_db": 0.3737490180689244}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.54681310079566e-05, "log_rms_mae_db": 1.4956245221510274, "rms_envelope_nrmse": 0.20899824880217016, "envelope_correlation": 0.9719981919732, "rms_gain_error_db": -0.12667532795479103, "reference_mean_power": 0.00010955764718773985, "output_mean_power": 0.00010640821652247791}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 6.937978158746457e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2971450195505076e-05, "output_power": 2.08297467609449e-05, "reference_active": true, "delta_db": -0.4250442320431979}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1375769637538917e-05, "output_power": 1.1186697373762884e-05, "reference_active": true, "delta_db": -0.07278899330215549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4470462261987434e-06, "output_power": 3.287448371507191e-06, "reference_active": true, "delta_db": 1.2821676755583131}], "active_band_count": 3, "active_band_mae_db": 0.5933336336345555}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 5.762149573106217e-05, "log_rms_mae_db": 0.9368315616318283, "rms_envelope_nrmse": 0.1389488805026005, "envelope_correlation": 0.9869881641780568, "rms_gain_error_db": -0.12667586555939964, "reference_mean_power": 0.00011019456862153096, "output_mean_power": 0.00010702681526099119}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 6.599778128222787e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3073624841125754e-05, "output_power": 2.1889700094632877e-05, "reference_active": true, "delta_db": -0.22876015619670328}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.235055585786147e-06, "output_power": 6.1700243831725705e-06, "reference_active": true, "delta_db": -0.045534492709623155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9515593840599056e-07, "output_power": 2.4545355904024376e-07, "reference_active": true, "delta_db": 0.995875627088485}], "active_band_count": 3, "active_band_mae_db": 0.4233900919982705}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9122120564159654, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0870f22e43b9318952925a9d605d17ad4e7688663df46328735b969eb6b1d716", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9907798487700665, "standard_deviation": 0.0016072096750947278, "minimum": 0.9883010309459317, "maximum": 0.9931717188365632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.283322243264302, "standard_deviation": 0.02681921918439539, "minimum": 1.2452132333547643, "maximum": 1.3281762930812966, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.39718869149857783, "standard_deviation": 0.20460735448330894, "minimum": 0.10580841644517534, "maximum": 0.6172983860795306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11375715522509998, "standard_deviation": 0.009846230957080088, "minimum": 0.09813290503249049, "maximum": 0.1279804686708195, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975182644125781, "standard_deviation": 0.0006044014218017849, "minimum": 0.9968065838327143, "maximum": 0.9982008131576939, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7239791364529569, "standard_deviation": 0.020370676175822507, "minimum": 0.699445997136947, "maximum": 0.7580770368029716, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6016766679913331, "standard_deviation": 0.27898622168632947, "minimum": 0.2858079302793261, "maximum": 1.140816467253176, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05893896377504869, "standard_deviation": 0.007344427798071206, "minimum": 0.049872792349525324, "maximum": 0.06914377338462271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9727129027585402, "standard_deviation": 0.0021635594272049353, "minimum": 0.9703008403737313, "maximum": 0.9775923204857412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4942736518700739, "standard_deviation": 0.01748316299044987, "minimum": 1.45883234599555, "maximum": 1.5132776158378578, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4232302915946263, "standard_deviation": 0.25287873526163845, "minimum": 0.09335366717151487, "maximum": 0.885053804465894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20652283679995845, "standard_deviation": 0.00796320096651732, "minimum": 0.18833562274546245, "maximum": 0.2152766630235689, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9887503525901286, "standard_deviation": 0.001356413219451395, "minimum": 0.9869881641780568, "maximum": 0.9911675690495677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9290873999515636, "standard_deviation": 0.017938019810607355, "minimum": 0.9018463241464262, "maximum": 0.9542223426043884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.588289565637278, "standard_deviation": 0.3488891183854568, "minimum": 0.1516431551974341, "maximum": 1.1523186349291217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12917530177419367, "standard_deviation": 0.007750777688028624, "minimum": 0.11497496858271726, "maximum": 0.1389488805026005, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921765275419845, "standard_deviation": 0.0027322108749891654, "minimum": 0.9864354765978259, "maximum": 0.9947942857586037, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2138450656840236, "standard_deviation": 0.03149048060040494, "minimum": 1.1788979856579116, "maximum": 1.254181680571897, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3206870020033818, "standard_deviation": 0.5151859782759501, "minimum": 0.9058710823047221, "maximum": 2.162624355406018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10408079628583117, "standard_deviation": 0.01749290240932187, "minimum": 0.08560549200475161, "maximum": 0.13998949754847426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980956787919123, "standard_deviation": 0.0005755259702274126, "minimum": 0.9971395256938319, "maximum": 0.998848209879298, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7733687543903788, "standard_deviation": 0.030844127579745702, "minimum": 0.7454463025654291, "maximum": 0.8403575159479478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9829010380920201, "standard_deviation": 0.5387519802910097, "minimum": 0.3554291209871346, "maximum": 1.8248459542583888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05120080323574314, "standard_deviation": 0.0077758887762107, "minimum": 0.03961000023503606, "maximum": 0.06401381237314649, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12451480523174169, "standard_deviation": 0.033705328532695256, "minimum": 0.07816142970466998, "maximum": 0.18405035461879957, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.697785548530685, "standard_deviation": 0.00916098958183581, "minimum": 5.680689223913261, "maximum": 5.705994184031574, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5295692735531325, "standard_deviation": 0.03002645602519571, "minimum": 0.4923084372318775, "maximum": 0.5839702133070274, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03860099814880904, "standard_deviation": 0.033048992887364806, "minimum": -0.09344322597177687, "maximum": 0.005101032452408528, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945286274222153, "standard_deviation": 0.001282434891600412, "minimum": 0.9920949700328234, "maximum": 0.9957983503959431, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0504681363695505, "standard_deviation": 0.03370085447027603, "minimum": 1.0039180078816488, "maximum": 1.0975494645093278, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3564973509754227, "standard_deviation": 0.23861623669739176, "minimum": 0.013604631893340192, "maximum": 0.7212926555997415, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08162641673853854, "standard_deviation": 0.009200451083219808, "minimum": 0.07187920552478638, "maximum": 0.09872704044782392, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986313885013319, "standard_deviation": 0.000303755879151356, "minimum": 0.9980129654835246, "maximum": 0.9989792403263411, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5712648225304913, "standard_deviation": 0.03439033834979427, "minimum": 0.5408883247119989, "maximum": 0.6371483934658537, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35228912640200166, "standard_deviation": 0.19208461859240628, "minimum": 0.04276716414533436, "maximum": 0.5849083385076249, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04006111630087582, "standard_deviation": 0.003949981100337895, "minimum": 0.03586378219425274, "maximum": 0.04812096232556475, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0065", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.09330677986145, "predicted_mos": 2.09330677986145, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "8ea904671e0286480366beb93291bb41a40278439c9f8e5999f1fce72edba411", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.859691399615258, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9493932048674874, "全帯域_10ms_rms_envelope_nrmse": 0.0823118399581696, "全帯域_10ms_envelope_correlation": 0.994571419352762, "全帯域_10ms_modulation_mae_db": 0.1561659635433735, "全帯域_40ms_log_rms_mae_db": 0.5112438069749181, "全帯域_40ms_rms_envelope_nrmse": 0.0421494972680832, "全帯域_40ms_envelope_correlation": 0.9985287934511176, "全帯域_40ms_modulation_mae_db": 0.18531863666100104, "300–1000 Hz_10ms_log_rms_mae_db": 1.7536617182286651, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10435778153704502, "300–1000 Hz_10ms_envelope_correlation": 0.9925199142248062, "300–1000 Hz_10ms_modulation_mae_db": 1.0009957932364035, "300–1000 Hz_40ms_log_rms_mae_db": 1.330530458525382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05066820658803702, "300–1000 Hz_40ms_envelope_correlation": 0.9982804884833493, "300–1000 Hz_40ms_modulation_mae_db": 0.6624221882346891, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4333904879997579, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10909287823430928, "1000–3000 Hz_10ms_envelope_correlation": 0.9915815465101918, "1000–3000 Hz_10ms_modulation_mae_db": 0.1387905010989163, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8299643057521932, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052733612602417955, "1000–3000 Hz_40ms_envelope_correlation": 0.9979477929646083, "1000–3000 Hz_40ms_modulation_mae_db": 0.31997639575850223, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7259794368482309, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21131401245146839, "3000–6000 Hz_10ms_envelope_correlation": 0.9717281334762373, "3000–6000 Hz_10ms_modulation_mae_db": 1.2065429430033763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1264600829101739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1511220395877662, "3000–6000 Hz_40ms_envelope_correlation": 0.9853997671133969, "3000–6000 Hz_40ms_modulation_mae_db": 1.4700475290795871, "mr_stft_db": 6.0329668477110125, "band_mae_db": 0.17912358620581456, "rms_gain_error_db": 0.061237440327726704, "output_peak": 0.5268993789384584}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0112028054090092, "全帯域_10ms_rms_envelope_nrmse": 0.0884641893382881, "全帯域_10ms_envelope_correlation": 0.9939038977850859, "全帯域_10ms_modulation_mae_db": 0.3309409104976163, "全帯域_40ms_log_rms_mae_db": 0.5493057919783186, "全帯域_40ms_rms_envelope_nrmse": 0.052465999017717396, "全帯域_40ms_envelope_correlation": 0.9980191180793508, "全帯域_40ms_modulation_mae_db": 0.5410301520551819, "300–1000 Hz_10ms_log_rms_mae_db": 1.7339969856720527, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1285185792755294, "300–1000 Hz_10ms_envelope_correlation": 0.9885319718485416, "300–1000 Hz_10ms_modulation_mae_db": 1.520423921048789, "300–1000 Hz_40ms_log_rms_mae_db": 1.2377735016546219, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07710020300462818, "300–1000 Hz_40ms_envelope_correlation": 0.9961874972192698, "300–1000 Hz_40ms_modulation_mae_db": 0.9093648512210498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4816639483840033, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11582990546117379, "1000–3000 Hz_10ms_envelope_correlation": 0.9905445200901629, "1000–3000 Hz_10ms_modulation_mae_db": 0.8274200574562318, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9041585424646837, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06475851328382176, "1000–3000 Hz_40ms_envelope_correlation": 0.997108319406645, "1000–3000 Hz_40ms_modulation_mae_db": 0.507670663955098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8290069491257337, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23330993039883635, "3000–6000 Hz_10ms_envelope_correlation": 0.9650017504518961, "3000–6000 Hz_10ms_modulation_mae_db": 1.0806043663671385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1530576669920194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17037470166072552, "3000–6000 Hz_40ms_envelope_correlation": 0.9806728856728154, "3000–6000 Hz_40ms_modulation_mae_db": 1.2449767371896214, "mr_stft_db": 6.04317120093302, "band_mae_db": 0.2026989535963762, "rms_gain_error_db": -0.19133411362241343, "output_peak": 0.5392228093529653}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.044510574844885, "全帯域_10ms_rms_envelope_nrmse": 0.07913904661692099, "全帯域_10ms_envelope_correlation": 0.9953077838419492, "全帯域_10ms_modulation_mae_db": 0.7260418606089569, "全帯域_40ms_log_rms_mae_db": 0.598945933266448, "全帯域_40ms_rms_envelope_nrmse": 0.04699179134641089, "全帯域_40ms_envelope_correlation": 0.9986028641371087, "全帯域_40ms_modulation_mae_db": 0.6931341198379521, "300–1000 Hz_10ms_log_rms_mae_db": 1.8112274084617996, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10301819737819957, "300–1000 Hz_10ms_envelope_correlation": 0.9931551120685917, "300–1000 Hz_10ms_modulation_mae_db": 0.4221255255564913, "300–1000 Hz_40ms_log_rms_mae_db": 1.2610133171969904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06862293116899322, "300–1000 Hz_40ms_envelope_correlation": 0.9974260608914371, "300–1000 Hz_40ms_modulation_mae_db": 0.2788740079229508, "1000–3000 Hz_10ms_log_rms_mae_db": 1.456358403477253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1274337367453186, "1000–3000 Hz_10ms_envelope_correlation": 0.9884171862033774, "1000–3000 Hz_10ms_modulation_mae_db": 0.5411961992549732, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9057523958269027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07604595729116817, "1000–3000 Hz_40ms_envelope_correlation": 0.9957683066987351, "1000–3000 Hz_40ms_modulation_mae_db": 0.6663672945240803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8549744895194868, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2529610948460889, "3000–6000 Hz_10ms_envelope_correlation": 0.958798926846696, "3000–6000 Hz_10ms_modulation_mae_db": 1.143153816087595, "3000–6000 Hz_40ms_log_rms_mae_db": 1.188019189472903, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17358886958673356, "3000–6000 Hz_40ms_envelope_correlation": 0.980319375796018, "3000–6000 Hz_40ms_modulation_mae_db": 1.2637310968516708, "mr_stft_db": 6.052891552423624, "band_mae_db": 0.23271361271636035, "rms_gain_error_db": -0.21730779789714816, "output_peak": 0.5018723259523052}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0571845778277245, "全帯域_10ms_rms_envelope_nrmse": 0.07427818979944524, "全帯域_10ms_envelope_correlation": 0.995568192217556, "全帯域_10ms_modulation_mae_db": 0.4507210206849635, "全帯域_40ms_log_rms_mae_db": 0.5781486418626044, "全帯域_40ms_rms_envelope_nrmse": 0.04119937369666815, "全帯域_40ms_envelope_correlation": 0.9986288126752868, "全帯域_40ms_modulation_mae_db": 0.47727757991023606, "300–1000 Hz_10ms_log_rms_mae_db": 1.7252962589862237, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10436221992402729, "300–1000 Hz_10ms_envelope_correlation": 0.9922807631547381, "300–1000 Hz_10ms_modulation_mae_db": 0.8748568863263452, "300–1000 Hz_40ms_log_rms_mae_db": 1.2707672771819338, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06452745541938441, "300–1000 Hz_40ms_envelope_correlation": 0.9969821107781598, "300–1000 Hz_40ms_modulation_mae_db": 0.5749175720935519, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4340510001551867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11982129756291937, "1000–3000 Hz_10ms_envelope_correlation": 0.9905611183586703, "1000–3000 Hz_10ms_modulation_mae_db": 1.3367313589024157, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9100713493005373, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07835490719188216, "1000–3000 Hz_40ms_envelope_correlation": 0.9964810678629515, "1000–3000 Hz_40ms_modulation_mae_db": 1.4624283577091397, "3000–6000 Hz_10ms_log_rms_mae_db": 1.842258020783915, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23486628771703058, "3000–6000 Hz_10ms_envelope_correlation": 0.9644835865930421, "3000–6000 Hz_10ms_modulation_mae_db": 1.5277405322623687, "3000–6000 Hz_40ms_log_rms_mae_db": 1.144188251821956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1575449991701813, "3000–6000 Hz_40ms_envelope_correlation": 0.9835621241036407, "3000–6000 Hz_40ms_modulation_mae_db": 1.8125094001141255, "mr_stft_db": 6.048861052239807, "band_mae_db": 0.2141982878957169, "rms_gain_error_db": -0.09645170951109258, "output_peak": 0.5511540965521182}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9838460000146659, "全帯域_10ms_rms_envelope_nrmse": 0.0704453552208698, "全帯域_10ms_envelope_correlation": 0.9960149252740176, "全帯域_10ms_modulation_mae_db": 0.1623146820244243, "全帯域_40ms_log_rms_mae_db": 0.522849529743834, "全帯域_40ms_rms_envelope_nrmse": 0.038656043378709516, "全帯域_40ms_envelope_correlation": 0.9987948097798037, "全帯域_40ms_modulation_mae_db": 0.15639376757360235, "300–1000 Hz_10ms_log_rms_mae_db": 1.6794148507703426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10474259197089363, "300–1000 Hz_10ms_envelope_correlation": 0.9922190967385358, "300–1000 Hz_10ms_modulation_mae_db": 0.9577880876410635, "300–1000 Hz_40ms_log_rms_mae_db": 1.195318825447687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06219975916737131, "300–1000 Hz_40ms_envelope_correlation": 0.9972254786720101, "300–1000 Hz_40ms_modulation_mae_db": 0.8836771606559316, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5042882711213954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11979585463298137, "1000–3000 Hz_10ms_envelope_correlation": 0.9901015732734464, "1000–3000 Hz_10ms_modulation_mae_db": 0.4079317374522136, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9349222172705073, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07952990106130259, "1000–3000 Hz_40ms_envelope_correlation": 0.9957890516218545, "1000–3000 Hz_40ms_modulation_mae_db": 0.29979479956838767, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8125194692888762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.207348445870231, "3000–6000 Hz_10ms_envelope_correlation": 0.9725472208313395, "3000–6000 Hz_10ms_modulation_mae_db": 0.7476195117811414, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1076971490950631, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1391523498356822, "3000–6000 Hz_40ms_envelope_correlation": 0.9869362610487636, "3000–6000 Hz_40ms_modulation_mae_db": 0.921978522195544, "mr_stft_db": 6.06642252006802, "band_mae_db": 0.20759105842199851, "rms_gain_error_db": -0.08661257819198928, "output_peak": 0.5236246370045505}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9646100314692629, "全帯域_10ms_rms_envelope_nrmse": 0.0741151732679144, "全帯域_10ms_envelope_correlation": 0.9956872147622938, "全帯域_10ms_modulation_mae_db": 0.16822998342869044, "全帯域_40ms_log_rms_mae_db": 0.48245374675324526, "全帯域_40ms_rms_envelope_nrmse": 0.04367634688461185, "全帯域_40ms_envelope_correlation": 0.9985685938790755, "全帯域_40ms_modulation_mae_db": 0.1262993249036111, "300–1000 Hz_10ms_log_rms_mae_db": 1.7854528780264975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1432539006244942, "300–1000 Hz_10ms_envelope_correlation": 0.9858402208323568, "300–1000 Hz_10ms_modulation_mae_db": 1.7845193336628522, "300–1000 Hz_40ms_log_rms_mae_db": 1.2921092883985985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09916224307594171, "300–1000 Hz_40ms_envelope_correlation": 0.9935533158491362, "300–1000 Hz_40ms_modulation_mae_db": 1.5811035480714768, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4790286064516094, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13515719983153487, "1000–3000 Hz_10ms_envelope_correlation": 0.9873636968159312, "1000–3000 Hz_10ms_modulation_mae_db": 0.24935828145577274, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9187493317121377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08057867683721982, "1000–3000 Hz_40ms_envelope_correlation": 0.995422887941114, "1000–3000 Hz_40ms_modulation_mae_db": 0.44437518902190637, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8074688371475731, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23971497894010702, "3000–6000 Hz_10ms_envelope_correlation": 0.9629741050761997, "3000–6000 Hz_10ms_modulation_mae_db": 0.8023974753172554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1346180479986383, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16477939186724913, "3000–6000 Hz_40ms_envelope_correlation": 0.9816683380383724, "3000–6000 Hz_40ms_modulation_mae_db": 0.9606309285804429, "mr_stft_db": 6.0473285010182005, "band_mae_db": 0.1792874944691152, "rms_gain_error_db": -0.13853939520840783, "output_peak": 0.5082016754964639}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0153678685421323, "全帯域_10ms_rms_envelope_nrmse": 0.08022489239770786, "全帯域_10ms_envelope_correlation": 0.9950528392128755, "全帯域_10ms_modulation_mae_db": 0.3112767299636608, "全帯域_40ms_log_rms_mae_db": 0.5586682149108767, "全帯域_40ms_rms_envelope_nrmse": 0.04780899710661698, "全帯域_40ms_envelope_correlation": 0.9984198834497532, "全帯域_40ms_modulation_mae_db": 0.558286462059647, "300–1000 Hz_10ms_log_rms_mae_db": 1.7627854746514142, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11899376477884907, "300–1000 Hz_10ms_envelope_correlation": 0.990149379958003, "300–1000 Hz_10ms_modulation_mae_db": 1.244727223406007, "300–1000 Hz_40ms_log_rms_mae_db": 1.294696444492981, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07183344599709539, "300–1000 Hz_40ms_envelope_correlation": 0.9966267949322203, "300–1000 Hz_40ms_modulation_mae_db": 1.0057319504968858, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4966510761641894, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11925804908373543, "1000–3000 Hz_10ms_envelope_correlation": 0.9902532863366332, "1000–3000 Hz_10ms_modulation_mae_db": 0.5872954486440835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8563960388546006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06417675168042272, "1000–3000 Hz_40ms_envelope_correlation": 0.997582341459875, "1000–3000 Hz_40ms_modulation_mae_db": 0.9313259351195491, "3000–6000 Hz_10ms_log_rms_mae_db": 1.829046873044991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23135046461776268, "3000–6000 Hz_10ms_envelope_correlation": 0.9656746318969257, "3000–6000 Hz_10ms_modulation_mae_db": 0.6327612719915996, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1166682298128638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1651600936685221, "3000–6000 Hz_40ms_envelope_correlation": 0.9820973725947072, "3000–6000 Hz_40ms_modulation_mae_db": 0.8199593035544952, "mr_stft_db": 6.05622221454385, "band_mae_db": 0.32291967132883376, "rms_gain_error_db": -0.1835536534284945, "output_peak": 0.5511693514635719}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9755186323804484, "全帯域_10ms_rms_envelope_nrmse": 0.0678680978864329, "全帯域_10ms_envelope_correlation": 0.9964671852253133, "全帯域_10ms_modulation_mae_db": 0.444020614910883, "全帯域_40ms_log_rms_mae_db": 0.5560575543721658, "全帯域_40ms_rms_envelope_nrmse": 0.03413256108219947, "全帯域_40ms_envelope_correlation": 0.9992694950023542, "全帯域_40ms_modulation_mae_db": 0.6837355427781318, "300–1000 Hz_10ms_log_rms_mae_db": 1.6903967130393796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10129893731894585, "300–1000 Hz_10ms_envelope_correlation": 0.9930751472272432, "300–1000 Hz_10ms_modulation_mae_db": 0.7475159351029559, "300–1000 Hz_40ms_log_rms_mae_db": 1.183412324860427, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058048786597793424, "300–1000 Hz_40ms_envelope_correlation": 0.9980756979234788, "300–1000 Hz_40ms_modulation_mae_db": 0.3408679235207033, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4596857899029616, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10374918985567627, "1000–3000 Hz_10ms_envelope_correlation": 0.9923813208753056, "1000–3000 Hz_10ms_modulation_mae_db": 0.4011108511559316, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8580195406548496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05838268734192107, "1000–3000 Hz_40ms_envelope_correlation": 0.9975806414432028, "1000–3000 Hz_40ms_modulation_mae_db": 0.5560132707111306, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8023154205433285, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23198321239106934, "3000–6000 Hz_10ms_envelope_correlation": 0.9653544121614904, "3000–6000 Hz_10ms_modulation_mae_db": 0.9714022541214892, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1305152086984056, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1567348917958262, "3000–6000 Hz_40ms_envelope_correlation": 0.9836706139731112, "3000–6000 Hz_40ms_modulation_mae_db": 1.2148979972283092, "mr_stft_db": 6.0564416052098435, "band_mae_db": 0.22282231236279237, "rms_gain_error_db": -0.14697145675599424, "output_peak": 0.5245263165210273}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83120, "output_samples": 83120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.846523624303594, "spectral_convergence": 0.3081155495655643, "spectral_convergence_status": "ok", "amplitude_floor": 8.933262458231008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.863373950192937, "spectral_convergence": 0.33940459069384554, "spectral_convergence_status": "ok", "amplitude_floor": 6.540934930277787e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.3890029686365075, "spectral_convergence": 0.3556478052750495, "spectral_convergence_status": "ok", "amplitude_floor": 6.044182898288457e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.147025669043969e-05, "output_power": 3.225816755916429e-05, "delta_db": 0.10739408149156415, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00012755476455897642, "output_power": 0.00013444032174693887, "delta_db": 0.228328576958135, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005321662836872518, "output_power": 0.0005452170466658283, "delta_db": 0.10522070343178541, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022563854189948138, "output_power": 0.00022501228666279723, "delta_db": -0.012070514526865922, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.229072811272528e-05, "output_power": 8.438856915070289e-05, "delta_db": 0.10932718271003312, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.381343402576783e-05, "output_power": 5.6711622222741274e-05, "delta_db": -0.5124004581165037, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17912358620581456, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.061237440327726704, "pitch": null, "output_rms_dbfs": -29.673699532762882, "output_peak": 0.5268993789384584, "reference_rms_dbfs": -29.734936973090612, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0329668477110125}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.00015201529585382603, "log_rms_mae_db": 0.9493932048674874, "rms_envelope_nrmse": 0.0823118399581696, "envelope_correlation": 0.994571419352762, "rms_gain_error_db": 0.06123756991452205, "reference_mean_power": 0.0010639579210132953, "output_mean_power": 0.0010790664981250057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.575807359242935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021400202307720183, "output_power": 0.00021597970062962055, "reference_active": true, "delta_db": 0.039950558701913545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3329698566280647e-05, "output_power": 2.1548414003692142e-05, "reference_active": true, "delta_db": -0.34493816551113304}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.358098056371455e-06, "output_power": 5.255932033408615e-06, "reference_active": true, "delta_db": -0.08360916641707393}], "active_band_count": 3, "active_band_mae_db": 0.1561659635433735}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.0001490763524181648, "log_rms_mae_db": 0.5112438069749181, "rms_envelope_nrmse": 0.0421494972680832, "envelope_correlation": 0.9985287934511176, "rms_gain_error_db": 0.061237489850064095, "reference_mean_power": 0.0010701430709341084, "output_mean_power": 0.0010853394593437651}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 5.309268989226658e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019126377610844124, "output_power": 0.00019313669776905844, "reference_active": true, "delta_db": 0.04232075838478941}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0962051121960504e-05, "output_power": 1.0181401547188261e-05, "reference_active": true, "delta_db": -0.32084257043248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.152513485967136e-07, "output_power": 3.0156274994223413e-07, "reference_active": true, "delta_db": -0.19279258116573367}], "active_band_count": 3, "active_band_mae_db": 0.18531863666100104}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.0001087945365454221, "log_rms_mae_db": 1.7536617182286651, "rms_envelope_nrmse": 0.10435778153704502, "envelope_correlation": 0.9925199142248062, "rms_gain_error_db": 0.1238496242128744, "reference_mean_power": 0.0005739909463294839, "output_mean_power": 0.0005905953247041397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 3.2669273354505384e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013940368795916017, "output_power": 0.00013951933229534391, "reference_active": true, "delta_db": 0.003601258863465654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.236537072685246e-06, "output_power": 7.4125784951799255e-06, "reference_active": true, "delta_db": -0.9553987236208201}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7105571121059603e-06, "output_power": 2.7386486494794673e-06, "reference_active": true, "delta_db": 2.0439873972249245}], "active_band_count": 3, "active_band_mae_db": 1.0009957932364035}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.00010533749157464783, "log_rms_mae_db": 1.330530458525382, "rms_envelope_nrmse": 0.05066820658803702, "envelope_correlation": 0.9982804884833493, "rms_gain_error_db": 0.12384953886890132, "reference_mean_power": 0.0005773280667275119, "output_mean_power": 0.000594028969463253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 3.141212974645643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001274169753045263, "output_power": 0.00012735809176562407, "reference_active": true, "delta_db": -0.0020074803056353997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.860144772024442e-06, "output_power": 2.9662281250540745e-06, "reference_active": true, "delta_db": -1.1439904442754423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1444270061935373e-07, "output_power": 1.3890406700980262e-07, "reference_active": true, "delta_db": 0.8412686401229899}], "active_band_count": 3, "active_band_mae_db": 0.6624221882346891}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 8.389504112280413e-05, "log_rms_mae_db": 1.4333904879997579, "rms_envelope_nrmse": 0.10909287823430928, "envelope_correlation": 0.9915815465101918, "rms_gain_error_db": 0.03961538942528921, "reference_mean_power": 0.00021157950445783077, "output_mean_power": 0.00021351831551980375}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.2662436125548267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5392735637289177e-05, "output_power": 5.445380100304041e-05, "reference_active": true, "delta_db": -0.07424613911511115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790041823950801e-06, "output_power": 6.138753855525558e-06, "reference_active": true, "delta_db": 0.25398519070409836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4111913003034675e-06, "output_power": 1.382839807901407e-06, "reference_active": true, "delta_db": -0.08814017347753941}], "active_band_count": 3, "active_band_mae_db": 0.1387905010989163}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 7.722064559288104e-05, "log_rms_mae_db": 0.8299643057521932, "rms_envelope_nrmse": 0.052733612602417955, "envelope_correlation": 0.9979477929646083, "rms_gain_error_db": 0.03961525095388303, "reference_mean_power": 0.0002128095577406816, "output_mean_power": 0.0002147596335616823}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 1.2078803728484267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0224253455435505e-05, "output_power": 4.9687926994622055e-05, "reference_active": true, "delta_db": -0.04662611879072889}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7667526770926532e-06, "output_power": 2.9386913923098244e-06, "reference_active": true, "delta_db": 0.26183642043037497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.983894982848384e-08, "output_power": 8.593168729134628e-08, "reference_active": true, "delta_db": -0.6514666480544028}], "active_band_count": 3, "active_band_mae_db": 0.31997639575850223}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.54681310079566e-05, "log_rms_mae_db": 1.7259794368482309, "rms_envelope_nrmse": 0.21131401245146839, "envelope_correlation": 0.9717281334762373, "rms_gain_error_db": -0.3976681800875214, "reference_mean_power": 0.00010955764718773985, "output_mean_power": 9.997142415673775e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 6.937978158746457e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2971450195505076e-05, "output_power": 1.999437580958324e-05, "reference_active": true, "delta_db": -0.6028056247376259}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1375769637538917e-05, "output_power": 8.813261073423663e-06, "reference_active": true, "delta_db": -1.1084415356317785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4470462261987434e-06, "output_power": 1.5768994679937585e-06, "reference_active": true, "delta_db": -1.9083816686407244}], "active_band_count": 3, "active_band_mae_db": 1.2065429430033763}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 5.762149573106217e-05, "log_rms_mae_db": 1.1264600829101739, "rms_envelope_nrmse": 0.1511220395877662, "envelope_correlation": 0.9853997671133969, "rms_gain_error_db": -0.3976689147101352, "reference_mean_power": 0.00011019456862153096, "output_mean_power": 0.00010055259836961919}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 6.599778128222787e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3073624841125754e-05, "output_power": 2.000184621278062e-05, "reference_active": true, "delta_db": -0.6204574325219746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.235055585786147e-06, "output_power": 4.733369080748999e-06, "reference_active": true, "delta_db": -1.1966996002503307}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9515593840599056e-07, "output_power": 1.074195220074563e-07, "reference_active": true, "delta_db": -2.592985554466456}], "active_band_count": 3, "active_band_mae_db": 1.4700475290795871}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8981371764288133, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1439e59a1690f3dd3c86c7d6fd0afeedc8db54e4a3cd635b5a6adac1cb4709e2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9901505310579648, "standard_deviation": 0.0016094330284789323, "minimum": 0.9873636968159312, "maximum": 0.9923813208753056, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4681396979570445, "standard_deviation": 0.026728545677907226, "minimum": 1.4333904879997579, "maximum": 1.5042882711213954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5612293044275674, "standard_deviation": 0.3776237476104888, "minimum": 0.1387905010989163, "maximum": 1.3367313589024157, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11876726392595613, "standard_deviation": 0.009811887600406202, "minimum": 0.10374918985567627, "maximum": 0.13515719983153487, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967100511748733, "standard_deviation": 0.0009749568094163778, "minimum": 0.995422887941114, "maximum": 0.9979477929646083, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8897542152295516, "standard_deviation": 0.0367457836790766, "minimum": 0.8299643057521932, "maximum": 0.9349222172705073, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6484939882959742, "standard_deviation": 0.38588781341175693, "minimum": 0.29979479956838767, "maximum": 1.4624283577091397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06932012591126953, "standard_deviation": 0.01068952192269339, "minimum": 0.052733612602417955, "maximum": 0.08057867683721982, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9658203459167284, "standard_deviation": 0.004476120187151303, "minimum": 0.958798926846696, "maximum": 0.9725472208313395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8129461870377668, "standard_deviation": 0.03936850492544141, "minimum": 1.7259794368482309, "maximum": 1.8549744895194868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0140277713664956, "standard_deviation": 0.28908987522222507, "minimum": 0.6327612719915996, "maximum": 1.5277405322623687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23035605340407428, "standard_deviation": 0.01475142035168676, "minimum": 0.207348445870231, "maximum": 0.2529610948460889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9830408422926031, "standard_deviation": 0.0023051080587021084, "minimum": 0.980319375796018, "maximum": 0.9869362610487636, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.137652978350253, "standard_deviation": 0.024899570558744148, "minimum": 1.1076971490950631, "maximum": 1.188019189472903, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2135914393492246, "standard_deviation": 0.32336325233941343, "minimum": 0.8199593035544952, "maximum": 1.8125094001141255, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15980716714658577, "standard_deviation": 0.011158559478985877, "minimum": 0.1391523498356822, "maximum": 0.17358886958673356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9909714507566021, "standard_deviation": 0.0026136715392836934, "minimum": 0.9858402208323568, "maximum": 0.9931551120685917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.742779035979547, "standard_deviation": 0.044986706726415665, "minimum": 1.6794148507703426, "maximum": 1.8112274084617996, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0691190882476134, "standard_deviation": 0.434760067947921, "minimum": 0.4221255255564913, "maximum": 1.7845193336628522, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.113568246600998, "standard_deviation": 0.015322728271443549, "minimum": 0.10129893731894585, "maximum": 0.1432539006244942, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967946805936327, "standard_deviation": 0.0014818565226309248, "minimum": 0.9935533158491362, "maximum": 0.9982804884833493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2582026797198278, "standard_deviation": 0.0504888346685346, "minimum": 1.183412324860427, "maximum": 1.330530458525382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7796199002771549, "standard_deviation": 0.41748438043635855, "minimum": 0.2788740079229508, "maximum": 1.5811035480714768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06902037887740559, "standard_deviation": 0.014666639141300532, "minimum": 0.05066820658803702, "maximum": 0.09916224307594171, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.220169372124626, "standard_deviation": 0.045633928259638, "minimum": 0.17912358620581456, "maximum": 0.32291967132883376, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.0505381867684225, "standard_deviation": 0.010008198030599477, "minimum": 6.0329668477110125, "maximum": 6.06642252006802, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5283338239101827, "standard_deviation": 0.01814639525872567, "minimum": 0.5018723259523052, "maximum": 0.5511693514635719, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12494165803597668, "standard_deviation": 0.08782164607849646, "minimum": -0.21730779789714816, "maximum": 0.061237440327726704, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9953216822089817, "standard_deviation": 0.000813512761049197, "minimum": 0.9939038977850859, "maximum": 0.9964671852253133, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0002042119194519, "standard_deviation": 0.038333861363204484, "minimum": 0.9493932048674874, "maximum": 1.0571845778277245, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3437139707078211, "standard_deviation": 0.19558573545128988, "minimum": 0.1561659635433735, "maximum": 0.7260418606089569, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0771058480607186, "standard_deviation": 0.006725373938304391, "minimum": 0.0678680978864329, "maximum": 0.0884641893382881, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986040463067314, "standard_deviation": 0.0003510808404429282, "minimum": 0.9980191180793508, "maximum": 0.9992694950023542, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5447091524828014, "standard_deviation": 0.03756538410848117, "minimum": 0.48245374675324526, "maximum": 0.598945933266448, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4276844482224204, "standard_deviation": 0.23651098345648558, "minimum": 0.1262993249036111, "maximum": 0.6931341198379521, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04338507622262718, "standard_deviation": 0.005724685971779774, "minimum": 0.03413256108219947, "maximum": 0.052465999017717396, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0065", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.572547197341919, "predicted_mos": 2.572547197341919, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "b7e2e53886d45f5a782eac96aee183a5f857a708b36070a5d02d32026b7d0b5c", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9304201998747885, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9782912608194722, "全帯域_10ms_rms_envelope_nrmse": 0.07048391033738402, "全帯域_10ms_envelope_correlation": 0.9960597653384419, "全帯域_10ms_modulation_mae_db": 0.19746097838742693, "全帯域_40ms_log_rms_mae_db": 0.5250852334466238, "全帯域_40ms_rms_envelope_nrmse": 0.03781486802915607, "全帯域_40ms_envelope_correlation": 0.9989146570501292, "全帯域_40ms_modulation_mae_db": 0.2073594396006868, "300–1000 Hz_10ms_log_rms_mae_db": 1.1946283598223981, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0823570825673776, "300–1000 Hz_10ms_envelope_correlation": 0.9953372075646402, "300–1000 Hz_10ms_modulation_mae_db": 0.8455565657499594, "300–1000 Hz_40ms_log_rms_mae_db": 0.8081703666131165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04830657059920618, "300–1000 Hz_40ms_envelope_correlation": 0.9985001300319225, "300–1000 Hz_40ms_modulation_mae_db": 0.7180603167340438, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2397054578507638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10659308794937947, "1000–3000 Hz_10ms_envelope_correlation": 0.9921587117951783, "1000–3000 Hz_10ms_modulation_mae_db": 0.5825494194678501, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7333903445028866, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06214996794858249, "1000–3000 Hz_40ms_envelope_correlation": 0.9975155795994285, "1000–3000 Hz_40ms_modulation_mae_db": 0.77168576648137, "3000–6000 Hz_10ms_log_rms_mae_db": 1.359149862849004, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20600895377578662, "3000–6000 Hz_10ms_envelope_correlation": 0.9728198185166104, "3000–6000 Hz_10ms_modulation_mae_db": 0.43499047593475076, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8869424778281214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1291256059526933, "3000–6000 Hz_40ms_envelope_correlation": 0.9887982067077503, "3000–6000 Hz_40ms_modulation_mae_db": 0.4332300212323097, "mr_stft_db": 5.663277646501698, "band_mae_db": 0.23106820437550543, "rms_gain_error_db": -0.11558140000677586, "output_peak": 0.5200132684034662}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0511849059416642, "全帯域_10ms_rms_envelope_nrmse": 0.07152782063079718, "全帯域_10ms_envelope_correlation": 0.9959597083919842, "全帯域_10ms_modulation_mae_db": 0.30619096906779125, "全帯域_40ms_log_rms_mae_db": 0.5678389949338881, "全帯域_40ms_rms_envelope_nrmse": 0.04378021780520857, "全帯域_40ms_envelope_correlation": 0.9985188819502219, "全帯域_40ms_modulation_mae_db": 0.19399098444378163, "300–1000 Hz_10ms_log_rms_mae_db": 1.2112219364360657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11018215433077473, "300–1000 Hz_10ms_envelope_correlation": 0.9916436484881852, "300–1000 Hz_10ms_modulation_mae_db": 1.1468340560058612, "300–1000 Hz_40ms_log_rms_mae_db": 0.7938628152966656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06974031060575592, "300–1000 Hz_40ms_envelope_correlation": 0.996860243223508, "300–1000 Hz_40ms_modulation_mae_db": 0.6459133303021022, "1000–3000 Hz_10ms_log_rms_mae_db": 1.212632574191878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10697808098816249, "1000–3000 Hz_10ms_envelope_correlation": 0.9918403387231682, "1000–3000 Hz_10ms_modulation_mae_db": 0.3471557529331766, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7014617796479017, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054795182163391916, "1000–3000 Hz_40ms_envelope_correlation": 0.9977680018993095, "1000–3000 Hz_40ms_modulation_mae_db": 0.46485661498860925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.294921938733684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22197463012713073, "3000–6000 Hz_10ms_envelope_correlation": 0.9687419101894009, "3000–6000 Hz_10ms_modulation_mae_db": 0.09162067008359709, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8680663111292718, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14412176739820004, "3000–6000 Hz_40ms_envelope_correlation": 0.9860527925672675, "3000–6000 Hz_40ms_modulation_mae_db": 0.22877595545151233, "mr_stft_db": 5.6742170375138175, "band_mae_db": 0.1465204501713991, "rms_gain_error_db": -0.12559958671520421, "output_peak": 0.5770940775963627}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0291192407033145, "全帯域_10ms_rms_envelope_nrmse": 0.06706046463440563, "全帯域_10ms_envelope_correlation": 0.9964280566715908, "全帯域_10ms_modulation_mae_db": 0.14326275666460253, "全帯域_40ms_log_rms_mae_db": 0.5645749106715221, "全帯域_40ms_rms_envelope_nrmse": 0.039205017836277785, "全帯域_40ms_envelope_correlation": 0.9987982104353493, "全帯域_40ms_modulation_mae_db": 0.21079093781051836, "300–1000 Hz_10ms_log_rms_mae_db": 1.1598959569468754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08389951262981067, "300–1000 Hz_10ms_envelope_correlation": 0.9950823828757271, "300–1000 Hz_10ms_modulation_mae_db": 0.26602971347667964, "300–1000 Hz_40ms_log_rms_mae_db": 0.8190068762440668, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06109097861224268, "300–1000 Hz_40ms_envelope_correlation": 0.9973654521600679, "300–1000 Hz_40ms_modulation_mae_db": 0.19706733750461516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2036730988995061, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1030478639802678, "1000–3000 Hz_10ms_envelope_correlation": 0.9927221366492407, "1000–3000 Hz_10ms_modulation_mae_db": 0.4261589583104031, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6863197714635276, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055058483995775204, "1000–3000 Hz_40ms_envelope_correlation": 0.9981779388176764, "1000–3000 Hz_40ms_modulation_mae_db": 0.4793729444797088, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3526128778857336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18816671028173226, "3000–6000 Hz_10ms_envelope_correlation": 0.977382502332818, "3000–6000 Hz_10ms_modulation_mae_db": 0.38479278874208395, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8606610085757713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11488916871304701, "3000–6000 Hz_40ms_envelope_correlation": 0.9911129106467815, "3000–6000 Hz_40ms_modulation_mae_db": 0.19878239653540208, "mr_stft_db": 5.6470589442727315, "band_mae_db": 0.16673094915830053, "rms_gain_error_db": -0.10606952278514595, "output_peak": 0.506808890663156}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0430007549165712, "全帯域_10ms_rms_envelope_nrmse": 0.08540510626683293, "全帯域_10ms_envelope_correlation": 0.9944810826272088, "全帯域_10ms_modulation_mae_db": 0.25899867250115355, "全帯域_40ms_log_rms_mae_db": 0.6041030025725947, "全帯域_40ms_rms_envelope_nrmse": 0.048850654293216236, "全帯域_40ms_envelope_correlation": 0.9984844631759228, "全帯域_40ms_modulation_mae_db": 0.3830754484433831, "300–1000 Hz_10ms_log_rms_mae_db": 1.2640752220201716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09248070550994864, "300–1000 Hz_10ms_envelope_correlation": 0.9941254932407307, "300–1000 Hz_10ms_modulation_mae_db": 0.7872842242492073, "300–1000 Hz_40ms_log_rms_mae_db": 0.9025835303249907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052252880897463556, "300–1000 Hz_40ms_envelope_correlation": 0.9982948435554907, "300–1000 Hz_40ms_modulation_mae_db": 0.44858337161519407, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1903488548994006, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11012049089502822, "1000–3000 Hz_10ms_envelope_correlation": 0.992299122457918, "1000–3000 Hz_10ms_modulation_mae_db": 0.6405432403453202, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6993769092523324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06928121014083681, "1000–3000 Hz_40ms_envelope_correlation": 0.9976295524106205, "1000–3000 Hz_40ms_modulation_mae_db": 0.8447081397490518, "3000–6000 Hz_10ms_log_rms_mae_db": 1.384671953787811, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22516032882100628, "3000–6000 Hz_10ms_envelope_correlation": 0.9683007592854785, "3000–6000 Hz_10ms_modulation_mae_db": 0.45180107841156086, "3000–6000 Hz_40ms_log_rms_mae_db": 0.922240782277735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14754773425776957, "3000–6000 Hz_40ms_envelope_correlation": 0.9872254197661687, "3000–6000 Hz_40ms_modulation_mae_db": 0.8295956491118046, "mr_stft_db": 5.672420372256075, "band_mae_db": 0.295570182805676, "rms_gain_error_db": -0.22565700183330112, "output_peak": 0.4847234905653919}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0837218761588752, "全帯域_10ms_rms_envelope_nrmse": 0.07155030689242928, "全帯域_10ms_envelope_correlation": 0.9958387895340642, "全帯域_10ms_modulation_mae_db": 0.2682831755093423, "全帯域_40ms_log_rms_mae_db": 0.6045106977191627, "全帯域_40ms_rms_envelope_nrmse": 0.03298618290499031, "全帯域_40ms_envelope_correlation": 0.9990874817769855, "全帯域_40ms_modulation_mae_db": 0.26360607304880623, "300–1000 Hz_10ms_log_rms_mae_db": 1.2364039060306116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08948678480530357, "300–1000 Hz_10ms_envelope_correlation": 0.9943124249046975, "300–1000 Hz_10ms_modulation_mae_db": 0.7340813379283547, "300–1000 Hz_40ms_log_rms_mae_db": 0.8451865575657814, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04632607418274129, "300–1000 Hz_40ms_envelope_correlation": 0.9984123663784482, "300–1000 Hz_40ms_modulation_mae_db": 0.365939673418545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2626865734849064, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1322200850645185, "1000–3000 Hz_10ms_envelope_correlation": 0.9875727851265201, "1000–3000 Hz_10ms_modulation_mae_db": 0.6213821867794714, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7911217199216665, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08653239379047384, "1000–3000 Hz_40ms_envelope_correlation": 0.9944850806306561, "1000–3000 Hz_40ms_modulation_mae_db": 0.7474693361473975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3937898647446931, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22838042158892208, "3000–6000 Hz_10ms_envelope_correlation": 0.9664860798020394, "3000–6000 Hz_10ms_modulation_mae_db": 0.408971098530166, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9130678828599462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.141939979506679, "3000–6000 Hz_40ms_envelope_correlation": 0.9869178056331188, "3000–6000 Hz_40ms_modulation_mae_db": 0.6417107375185354, "mr_stft_db": 5.6700399463793, "band_mae_db": 0.17342304319150706, "rms_gain_error_db": -0.04259451928699559, "output_peak": 0.4987896729837246}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0134227304810173, "全帯域_10ms_rms_envelope_nrmse": 0.06871380326221246, "全帯域_10ms_envelope_correlation": 0.9962866195552048, "全帯域_10ms_modulation_mae_db": 0.4560273226561464, "全帯域_40ms_log_rms_mae_db": 0.5218930894179206, "全帯域_40ms_rms_envelope_nrmse": 0.03880494902348356, "全帯域_40ms_envelope_correlation": 0.998864011004971, "全帯域_40ms_modulation_mae_db": 0.6562843530118981, "300–1000 Hz_10ms_log_rms_mae_db": 1.2114947287973103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08958698218920444, "300–1000 Hz_10ms_envelope_correlation": 0.9944596293636531, "300–1000 Hz_10ms_modulation_mae_db": 1.016180224004456, "300–1000 Hz_40ms_log_rms_mae_db": 0.8231794638831934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05020462695213567, "300–1000 Hz_40ms_envelope_correlation": 0.9984037726951669, "300–1000 Hz_40ms_modulation_mae_db": 0.5117972847270156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.219907128322872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12647226393716018, "1000–3000 Hz_10ms_envelope_correlation": 0.9889099242629885, "1000–3000 Hz_10ms_modulation_mae_db": 0.20180009517485611, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7230776781700434, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06930428960171502, "1000–3000 Hz_40ms_envelope_correlation": 0.997007722355402, "1000–3000 Hz_40ms_modulation_mae_db": 0.4430085206576076, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3392048501962734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22095078096138546, "3000–6000 Hz_10ms_envelope_correlation": 0.9689521532170506, "3000–6000 Hz_10ms_modulation_mae_db": 0.5340203636453703, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9380960210461151, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14683431258009494, "3000–6000 Hz_40ms_envelope_correlation": 0.9855311780606361, "3000–6000 Hz_40ms_modulation_mae_db": 0.784733264974248, "mr_stft_db": 5.678954118669121, "band_mae_db": 0.2227512204209338, "rms_gain_error_db": -0.11881670779933948, "output_peak": 0.5577511713245664}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0703946330492637, "全帯域_10ms_rms_envelope_nrmse": 0.07185539135233332, "全帯域_10ms_envelope_correlation": 0.9958141027520305, "全帯域_10ms_modulation_mae_db": 0.2534094555008081, "全帯域_40ms_log_rms_mae_db": 0.5788686099346643, "全帯域_40ms_rms_envelope_nrmse": 0.03675959911111323, "全帯域_40ms_envelope_correlation": 0.9988722947270122, "全帯域_40ms_modulation_mae_db": 0.3086879129101219, "300–1000 Hz_10ms_log_rms_mae_db": 1.1961637654304975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0900303185058402, "300–1000 Hz_10ms_envelope_correlation": 0.9942346722509534, "300–1000 Hz_10ms_modulation_mae_db": 0.7224152194964294, "300–1000 Hz_40ms_log_rms_mae_db": 0.801726184278882, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0472040423408495, "300–1000 Hz_40ms_envelope_correlation": 0.9983635590712149, "300–1000 Hz_40ms_modulation_mae_db": 0.5790123654681493, "1000–3000 Hz_10ms_log_rms_mae_db": 1.213591952845791, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10747808083074942, "1000–3000 Hz_10ms_envelope_correlation": 0.9919204160386257, "1000–3000 Hz_10ms_modulation_mae_db": 0.2658678640533677, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6736440945704375, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05296137283919126, "1000–3000 Hz_40ms_envelope_correlation": 0.9981935774694564, "1000–3000 Hz_40ms_modulation_mae_db": 0.24323954431297992, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3940843533819725, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21371433963257933, "3000–6000 Hz_10ms_envelope_correlation": 0.9707041810381649, "3000–6000 Hz_10ms_modulation_mae_db": 0.36783540958947253, "3000–6000 Hz_40ms_log_rms_mae_db": 0.89523908283507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14463424211805995, "3000–6000 Hz_40ms_envelope_correlation": 0.985887415505498, "3000–6000 Hz_40ms_modulation_mae_db": 0.25725865869974224, "mr_stft_db": 5.663333709515011, "band_mae_db": 0.2374824258601602, "rms_gain_error_db": -0.05185108950783805, "output_peak": 0.5056175140507936}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0101000652137408, "全帯域_10ms_rms_envelope_nrmse": 0.07601230763708683, "全帯域_10ms_envelope_correlation": 0.9956050029648279, "全帯域_10ms_modulation_mae_db": 0.6531544310002931, "全帯域_40ms_log_rms_mae_db": 0.5469265479673634, "全帯域_40ms_rms_envelope_nrmse": 0.04664751161211971, "全帯域_40ms_envelope_correlation": 0.9985083846692013, "全帯域_40ms_modulation_mae_db": 0.7096049006731032, "300–1000 Hz_10ms_log_rms_mae_db": 1.195547630884531, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09807342337558818, "300–1000 Hz_10ms_envelope_correlation": 0.9934152741793632, "300–1000 Hz_10ms_modulation_mae_db": 0.36728937888692476, "300–1000 Hz_40ms_log_rms_mae_db": 0.7838505225425716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06939675350776794, "300–1000 Hz_40ms_envelope_correlation": 0.9967959019457089, "300–1000 Hz_40ms_modulation_mae_db": 0.35926507925576234, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2173533536626608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10213823819860963, "1000–3000 Hz_10ms_envelope_correlation": 0.9930434369507679, "1000–3000 Hz_10ms_modulation_mae_db": 0.7424537266466641, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7450240462445323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05721146398450663, "1000–3000 Hz_40ms_envelope_correlation": 0.9982166153487368, "1000–3000 Hz_40ms_modulation_mae_db": 0.795477280440184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3702438791641331, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21200942415711627, "3000–6000 Hz_10ms_envelope_correlation": 0.9715411979384732, "3000–6000 Hz_10ms_modulation_mae_db": 0.9011242545423491, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9174758632516633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14039240182740853, "3000–6000 Hz_40ms_envelope_correlation": 0.9877175687973249, "3000–6000 Hz_40ms_modulation_mae_db": 1.037020798915262, "mr_stft_db": 5.667554507984175, "band_mae_db": 0.2702527580572439, "rms_gain_error_db": -0.1901531222293461, "output_peak": 0.5223000054158105}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83120, "output_samples": 83120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.409166901822185, "spectral_convergence": 0.29901238304083366, "spectral_convergence_status": "ok", "amplitude_floor": 8.933262458231008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.417364816246433, "spectral_convergence": 0.3117896430571401, "spectral_convergence_status": "ok", "amplitude_floor": 6.540934930277787e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.163301221436476, "spectral_convergence": 0.33790622455408836, "spectral_convergence_status": "ok", "amplitude_floor": 6.044182898288457e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.147025669043969e-05, "output_power": 3.344892847545777e-05, "delta_db": 0.264819245611916, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00012755476455897642, "output_power": 0.00013392136670011679, "delta_db": 0.21153187282096775, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005321662836872518, "output_power": 0.0005101753875233119, "delta_db": -0.18327852573337094, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022563854189948138, "output_power": 0.00021358965257042034, "delta_db": -0.23833075295173742, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.229072811272528e-05, "output_power": 8.498852753894205e-05, "delta_db": 0.14009399973664421, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.381343402576783e-05, "output_power": 5.889475686527768e-05, "delta_db": -0.3483548293983963, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23106820437550543, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11558140000677586, "pitch": null, "output_rms_dbfs": -29.850518373097387, "output_peak": 0.5200132684034662, "reference_rms_dbfs": -29.734936973090612, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.663277646501698}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.00015201529585382603, "log_rms_mae_db": 0.9782912608194722, "rms_envelope_nrmse": 0.07048391033738402, "envelope_correlation": 0.9960597653384419, "rms_gain_error_db": -0.1155814163991328, "reference_mean_power": 0.0010639579210132953, "output_mean_power": 0.0010360156372300705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.575807359242935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021400202307720183, "output_power": 0.000206459777197486, "reference_active": true, "delta_db": -0.15582424657006125}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3329698566280647e-05, "output_power": 2.1372460725932904e-05, "reference_active": true, "delta_db": -0.38054599790620436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.358098056371455e-06, "output_power": 5.2894362607395855e-06, "reference_active": true, "delta_db": -0.05601269068601523}], "active_band_count": 3, "active_band_mae_db": 0.19746097838742693}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.0001490763524181648, "log_rms_mae_db": 0.5250852334466238, "rms_envelope_nrmse": 0.03781486802915607, "envelope_correlation": 0.9989146570501292, "rms_gain_error_db": -0.11558072403554145, "reference_mean_power": 0.0010701430709341084, "output_mean_power": 0.0010420385152588682}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 5.309268989226658e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019126377610844124, "output_power": 0.00018573972504754168, "reference_active": true, "delta_db": -0.1272792745106693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0962051121960504e-05, "output_power": 9.860671579769303e-06, "reference_active": true, "delta_db": -0.4598532869127567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.152513485967136e-07, "output_power": 3.127248388212665e-07, "reference_active": true, "delta_db": -0.03494575737863442}], "active_band_count": 3, "active_band_mae_db": 0.2073594396006868}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.0001087945365454221, "log_rms_mae_db": 1.1946283598223981, "rms_envelope_nrmse": 0.0823570825673776, "envelope_correlation": 0.9953372075646402, "rms_gain_error_db": -0.15325907948387105, "reference_mean_power": 0.0005739909463294839, "output_mean_power": 0.0005540884971132014}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 3.2669273354505384e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013940368795916017, "output_power": 0.0001333303982534312, "reference_active": true, "delta_db": -0.19345086946732903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.236537072685246e-06, "output_power": 8.54973298763288e-06, "reference_active": true, "delta_db": -0.3355662599176944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7105571121059603e-06, "output_power": 2.7158315908927096e-06, "reference_active": true, "delta_db": 2.0076525678648545}], "active_band_count": 3, "active_band_mae_db": 0.8455565657499594}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.00010533749157464783, "log_rms_mae_db": 0.8081703666131165, "rms_envelope_nrmse": 0.04830657059920618, "envelope_correlation": 0.9985001300319225, "rms_gain_error_db": -0.15325909446014152, "reference_mean_power": 0.0005773280667275119, "output_mean_power": 0.0005573099049328498}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 3.141212974645643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001274169753045263, "output_power": 0.00012222773217869797, "reference_active": true, "delta_db": -0.18057537435094168}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.860144772024442e-06, "output_power": 3.601648496240747e-06, "reference_active": true, "delta_db": -0.301022674110922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1444270061935373e-07, "output_power": 1.682078996245412e-07, "reference_active": true, "delta_db": 1.6725829017402676}], "active_band_count": 3, "active_band_mae_db": 0.7180603167340438}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 8.389504112280413e-05, "log_rms_mae_db": 1.2397054578507638, "rms_envelope_nrmse": 0.10659308794937947, "envelope_correlation": 0.9921587117951783, "rms_gain_error_db": -0.21897609892201958, "reference_mean_power": 0.00021157950445783077, "output_mean_power": 0.0002011759152096474}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.2662436125548267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5392735637289177e-05, "output_power": 4.919763914900409e-05, "reference_active": true, "delta_db": -0.5150855109114689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790041823950801e-06, "output_power": 5.898022350755056e-06, "reference_active": true, "delta_db": 0.08024713158786353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4111913003034675e-06, "output_power": 1.0823166234124945e-06, "reference_active": true, "delta_db": -1.1523156159042178}], "active_band_count": 3, "active_band_mae_db": 0.5825494194678501}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 7.722064559288104e-05, "log_rms_mae_db": 0.7333903445028866, "rms_envelope_nrmse": 0.06214996794858249, "envelope_correlation": 0.9975155795994285, "rms_gain_error_db": -0.2189760860132635, "reference_mean_power": 0.0002128095577406816, "output_mean_power": 0.0002023454860658653}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 1.2078803728484267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0224253455435505e-05, "output_power": 4.466079668824817e-05, "reference_active": true, "delta_db": -0.5098702398663576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7667526770926532e-06, "output_power": 2.797265998563683e-06, "reference_active": true, "delta_db": 0.04763427573520131}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.983894982848384e-08, "output_power": 6.661081168598103e-08, "reference_active": true, "delta_db": -1.757552783842551}], "active_band_count": 3, "active_band_mae_db": 0.77168576648137}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.54681310079566e-05, "log_rms_mae_db": 1.359149862849004, "rms_envelope_nrmse": 0.20600895377578662, "envelope_correlation": 0.9728198185166104, "rms_gain_error_db": -0.14044770728572165, "reference_mean_power": 0.00010955764718773985, "output_mean_power": 0.00010607130846955858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 6.937978158746457e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2971450195505076e-05, "output_power": 2.1085099796516464e-05, "reference_active": true, "delta_db": -0.3721275233286966}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1375769637538917e-05, "output_power": 1.0349800349581326e-05, "reference_active": true, "delta_db": -0.41048816675186406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4470462261987434e-06, "output_power": 2.759800870460522e-06, "reference_active": true, "delta_db": 0.5223557377236917}], "active_band_count": 3, "active_band_mae_db": 0.43499047593475076}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 5.762149573106217e-05, "log_rms_mae_db": 0.8869424778281214, "rms_envelope_nrmse": 0.1291256059526933, "envelope_correlation": 0.9887982067077503, "rms_gain_error_db": -0.14044758304491217, "reference_mean_power": 0.00011019456862153096, "output_mean_power": 0.00010668796486906174}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 6.599778128222787e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3073624841125754e-05, "output_power": 2.2110049551659657e-05, "reference_active": true, "delta_db": -0.18526111173653717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.235055585786147e-06, "output_power": 5.437188622622032e-06, "reference_active": true, "delta_db": -0.594659301239225}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9515593840599056e-07, "output_power": 2.1996760996338207e-07, "reference_active": true, "delta_db": 0.5197696507211671}], "active_band_count": 3, "active_band_mae_db": 0.4332300212323097}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9183326966623196, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e97be220accc0cf268c34c78fa7f6ee8f67df78ffc9e61dec301ab81ce2359a2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9913083590005509, "standard_deviation": 0.0019668173144473345, "minimum": 0.9875727851265201, "maximum": 0.9930434369507679, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2199873742697225, "standard_deviation": 0.02222791310939676, "minimum": 1.1903488548994006, "maximum": 1.2626865734849064, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47848890546388867, "standard_deviation": 0.19600923447933236, "minimum": 0.20180009517485611, "maximum": 0.7424537266466641, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11188102398048447, "standard_deviation": 0.01117404378915766, "minimum": 0.10213823819860963, "maximum": 0.1322200850645185, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973742585664108, "standard_deviation": 0.0012391603546521713, "minimum": 0.9944850806306561, "maximum": 0.9982166153487368, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.719177042971666, "standard_deviation": 0.03764507660087072, "minimum": 0.6736440945704375, "maximum": 0.7911217199216665, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5987272684071137, "standard_deviation": 0.21844994782789706, "minimum": 0.24323954431297992, "maximum": 0.8447081397490518, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06341179555805915, "standard_deviation": 0.011320052197094359, "minimum": 0.05296137283919126, "maximum": 0.08653239379047384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9706160752900045, "standard_deviation": 0.0033849101294823854, "minimum": 0.9664860798020394, "maximum": 0.977382502332818, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3610849475929132, "standard_deviation": 0.033232101946724664, "minimum": 1.294921938733684, "maximum": 1.3940843533819725, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4468945174349188, "standard_deviation": 0.2241957076333149, "minimum": 0.09162067008359709, "maximum": 0.9011242545423491, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21454569866820739, "standard_deviation": 0.012962224904209773, "minimum": 0.18816671028173226, "maximum": 0.22838042158892208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9874054122105682, "standard_deviation": 0.0018400181637221581, "minimum": 0.9855311780606361, "maximum": 0.9911129106467815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9002236787254618, "standard_deviation": 0.0272087597922182, "minimum": 0.8606610085757713, "maximum": 0.9380960210461151, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.551388435304852, "standard_deviation": 0.31729426208093486, "minimum": 0.19878239653540208, "maximum": 1.037020798915262, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13868565154424406, "standard_deviation": 0.011218893413442203, "minimum": 0.11488916871304701, "maximum": 0.14754773425776957, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9940763416084939, "standard_deviation": 0.0011456150721957318, "minimum": 0.9916436484881852, "maximum": 0.9953372075646402, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2086789382960577, "standard_deviation": 0.031071522131462196, "minimum": 1.1598959569468754, "maximum": 1.2640752220201716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.735708839974734, "standard_deviation": 0.29735150770036867, "minimum": 0.26602971347667964, "maximum": 1.1468340560058612, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09201212048923099, "standard_deviation": 0.008802024224574283, "minimum": 0.0823570825673776, "maximum": 0.11018215433077473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978745336326911, "standard_deviation": 0.0007394911869665683, "minimum": 0.9967959019457089, "maximum": 0.9985001300319225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8221957895936585, "standard_deviation": 0.03758938912693705, "minimum": 0.7838505225425716, "maximum": 0.9025835303249907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.47820484487817844, "standard_deviation": 0.170550896460578, "minimum": 0.19706733750461516, "maximum": 0.7180603167340438, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05556527971227035, "standard_deviation": 0.00978661744883755, "minimum": 0.04632607418274129, "maximum": 0.06974031060575592, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21797490425509075, "standard_deviation": 0.0521609358776455, "minimum": 0.1465204501713991, "maximum": 0.295570182805676, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.667107035386492, "standard_deviation": 0.00970101190990042, "minimum": 5.6470589442727315, "maximum": 5.678954118669121, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.521637261375409, "standard_deviation": 0.031038918411456364, "minimum": 0.4847234905653919, "maximum": 0.5770940775963627, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1220403687704933, "standard_deviation": 0.06202969101970196, "minimum": -0.22565700183330112, "maximum": -0.04259451928699559, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9958091409794192, "standard_deviation": 0.000597953714318338, "minimum": 0.9944810826272088, "maximum": 0.9964280566715908, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0349044334104898, "standard_deviation": 0.03441042450547629, "minimum": 0.9782912608194722, "maximum": 1.0837218761588752, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3170984701609455, "standard_deviation": 0.1633141596714923, "minimum": 0.14326275666460253, "maximum": 0.6531544310002931, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07282613887668521, "standard_deviation": 0.005710618661159909, "minimum": 0.06706046463440563, "maximum": 0.08540510626683293, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987560480987241, "standard_deviation": 0.00022465698917822742, "minimum": 0.9984844631759228, "maximum": 0.9990874817769855, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5642251358329675, "standard_deviation": 0.0317669182385841, "minimum": 0.5218930894179206, "maximum": 0.6045106977191627, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36667500624278737, "standard_deviation": 0.20536734775500626, "minimum": 0.19399098444378163, "maximum": 0.7096049006731032, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04060612507694568, "standard_deviation": 0.005350610261655769, "minimum": 0.03298618290499031, "maximum": 0.048850654293216236, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0065", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.17651629447937, "predicted_mos": 2.17651629447937, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "5554b08eb14408312986351b794cc61fb574c052f78f7e9a6c48b6faa23de822", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9217228000052273, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0126018135482013, "全帯域_10ms_rms_envelope_nrmse": 0.07962617556111845, "全帯域_10ms_envelope_correlation": 0.9948598013677236, "全帯域_10ms_modulation_mae_db": 0.20841432645907856, "全帯域_40ms_log_rms_mae_db": 0.547160018585393, "全帯域_40ms_rms_envelope_nrmse": 0.0418546822906201, "全帯域_40ms_envelope_correlation": 0.998550226743756, "全帯域_40ms_modulation_mae_db": 0.2128250864501963, "300–1000 Hz_10ms_log_rms_mae_db": 1.410380353900788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09935698772214432, "300–1000 Hz_10ms_envelope_correlation": 0.9930090931332237, "300–1000 Hz_10ms_modulation_mae_db": 0.8902518963359075, "300–1000 Hz_40ms_log_rms_mae_db": 0.9329757849304141, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04953361620005532, "300–1000 Hz_40ms_envelope_correlation": 0.9982825165872146, "300–1000 Hz_40ms_modulation_mae_db": 0.6916309634746111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3036272553422341, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10788020054700352, "1000–3000 Hz_10ms_envelope_correlation": 0.9917780031723367, "1000–3000 Hz_10ms_modulation_mae_db": 0.28811996591710737, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7880155902188192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05341347032337685, "1000–3000 Hz_40ms_envelope_correlation": 0.998052916204116, "1000–3000 Hz_40ms_modulation_mae_db": 0.384807167187442, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5426902319834124, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23167832070318792, "3000–6000 Hz_10ms_envelope_correlation": 0.9656283569797596, "3000–6000 Hz_10ms_modulation_mae_db": 0.600153580926936, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9889864773728022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15869279877016515, "3000–6000 Hz_40ms_envelope_correlation": 0.9829735169281411, "3000–6000 Hz_40ms_modulation_mae_db": 0.4770762487376084, "mr_stft_db": 5.777098673232402, "band_mae_db": 0.1695668193091817, "rms_gain_error_db": -0.07129544638742717, "output_peak": 0.5754588164586549}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.045791754380437, "全帯域_10ms_rms_envelope_nrmse": 0.0754602005002488, "全帯域_10ms_envelope_correlation": 0.9953604382206335, "全帯域_10ms_modulation_mae_db": 0.36854522487178976, "全帯域_40ms_log_rms_mae_db": 0.5920803987382874, "全帯域_40ms_rms_envelope_nrmse": 0.03617907285453352, "全帯域_40ms_envelope_correlation": 0.9988754818514627, "全帯域_40ms_modulation_mae_db": 0.46010514433009114, "300–1000 Hz_10ms_log_rms_mae_db": 1.4049187968469736, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08756432059372266, "300–1000 Hz_10ms_envelope_correlation": 0.9945603408526928, "300–1000 Hz_10ms_modulation_mae_db": 1.0770459875380742, "300–1000 Hz_40ms_log_rms_mae_db": 0.9532085162567216, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042077667674883926, "300–1000 Hz_40ms_envelope_correlation": 0.9987391665354131, "300–1000 Hz_40ms_modulation_mae_db": 0.43587728678883847, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3623414163817218, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10690014917175192, "1000–3000 Hz_10ms_envelope_correlation": 0.9918450402078424, "1000–3000 Hz_10ms_modulation_mae_db": 0.42890456895687606, "1000–3000 Hz_40ms_log_rms_mae_db": 0.781954109173111, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049506041301659025, "1000–3000 Hz_40ms_envelope_correlation": 0.9981984348808097, "1000–3000 Hz_40ms_modulation_mae_db": 0.9685681891715756, "3000–6000 Hz_10ms_log_rms_mae_db": 1.511433830021437, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20772948487165693, "3000–6000 Hz_10ms_envelope_correlation": 0.9724002139112173, "3000–6000 Hz_10ms_modulation_mae_db": 0.4323567903812231, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9522925654859871, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1331014346005298, "3000–6000 Hz_40ms_envelope_correlation": 0.9880481911444533, "3000–6000 Hz_40ms_modulation_mae_db": 0.7313140041815344, "mr_stft_db": 5.793232717460388, "band_mae_db": 0.13079704331292422, "rms_gain_error_db": -0.028456953934218255, "output_peak": 0.5296639188426081}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0198532899832478, "全帯域_10ms_rms_envelope_nrmse": 0.07518714440572308, "全帯域_10ms_envelope_correlation": 0.9954246097324536, "全帯域_10ms_modulation_mae_db": 0.07925979915962979, "全帯域_40ms_log_rms_mae_db": 0.5255548219206315, "全帯域_40ms_rms_envelope_nrmse": 0.035030429355904884, "全帯域_40ms_envelope_correlation": 0.9990007155402691, "全帯域_40ms_modulation_mae_db": 0.1220242303348981, "300–1000 Hz_10ms_log_rms_mae_db": 1.418569836823376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08669733091621003, "300–1000 Hz_10ms_envelope_correlation": 0.9947955698150923, "300–1000 Hz_10ms_modulation_mae_db": 0.8073167803389821, "300–1000 Hz_40ms_log_rms_mae_db": 0.9902950700473934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046482435877667004, "300–1000 Hz_40ms_envelope_correlation": 0.9986206948503513, "300–1000 Hz_40ms_modulation_mae_db": 0.6672351413037366, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2897298122934044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11034609456905854, "1000–3000 Hz_10ms_envelope_correlation": 0.991319320402306, "1000–3000 Hz_10ms_modulation_mae_db": 0.24053927318866708, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7824219537620485, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05459495726647743, "1000–3000 Hz_40ms_envelope_correlation": 0.997800092996072, "1000–3000 Hz_40ms_modulation_mae_db": 0.7333265347574246, "3000–6000 Hz_10ms_log_rms_mae_db": 1.527376069544642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23221740581055583, "3000–6000 Hz_10ms_envelope_correlation": 0.9655855060425389, "3000–6000 Hz_10ms_modulation_mae_db": 0.1494463310145487, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9477159687249408, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13737780292861926, "3000–6000 Hz_40ms_envelope_correlation": 0.9872752675049422, "3000–6000 Hz_40ms_modulation_mae_db": 0.145498698372914, "mr_stft_db": 5.780825047030748, "band_mae_db": 0.11418682420301866, "rms_gain_error_db": -0.07767498782856763, "output_peak": 0.5704741613824599}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0151173539802958, "全帯域_10ms_rms_envelope_nrmse": 0.0940598209864969, "全帯域_10ms_envelope_correlation": 0.9928370855572313, "全帯域_10ms_modulation_mae_db": 0.5919567359642559, "全帯域_40ms_log_rms_mae_db": 0.5328131649079756, "全帯域_40ms_rms_envelope_nrmse": 0.0473015230678783, "全帯域_40ms_envelope_correlation": 0.9980891250583557, "全帯域_40ms_modulation_mae_db": 0.42067956480636476, "300–1000 Hz_10ms_log_rms_mae_db": 1.4244595780604765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1379774681966081, "300–1000 Hz_10ms_envelope_correlation": 0.9867445373314069, "300–1000 Hz_10ms_modulation_mae_db": 2.10879533351649, "300–1000 Hz_40ms_log_rms_mae_db": 0.9846570977356912, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06169755697100049, "300–1000 Hz_40ms_envelope_correlation": 0.9973006889960251, "300–1000 Hz_40ms_modulation_mae_db": 1.7568463641918148, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3136803945149103, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11863934158568352, "1000–3000 Hz_10ms_envelope_correlation": 0.9900017715432113, "1000–3000 Hz_10ms_modulation_mae_db": 0.4567854420312876, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7799764711211554, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0602447437349053, "1000–3000 Hz_40ms_envelope_correlation": 0.9974430037535451, "1000–3000 Hz_40ms_modulation_mae_db": 0.2212446798041899, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5356108000407491, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2107553893087196, "3000–6000 Hz_10ms_envelope_correlation": 0.9715192492181636, "3000–6000 Hz_10ms_modulation_mae_db": 0.30981773766359993, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9834819655111356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13949743297746167, "3000–6000 Hz_40ms_envelope_correlation": 0.9868905060252425, "3000–6000 Hz_40ms_modulation_mae_db": 0.5376763245748813, "mr_stft_db": 5.7700844008814185, "band_mae_db": 0.15682698008085058, "rms_gain_error_db": 0.015559145732980457, "output_peak": 0.49597688058647316}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9942522737660394, "全帯域_10ms_rms_envelope_nrmse": 0.07507821120973056, "全帯域_10ms_envelope_correlation": 0.9954212150033159, "全帯域_10ms_modulation_mae_db": 0.5423387299270875, "全帯域_40ms_log_rms_mae_db": 0.5425068179781009, "全帯域_40ms_rms_envelope_nrmse": 0.03902120431049327, "全帯域_40ms_envelope_correlation": 0.9987125926913055, "全帯域_40ms_modulation_mae_db": 0.5017258312869837, "300–1000 Hz_10ms_log_rms_mae_db": 1.389338199264461, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09926645177848975, "300–1000 Hz_10ms_envelope_correlation": 0.9930058238293136, "300–1000 Hz_10ms_modulation_mae_db": 0.8530398766046087, "300–1000 Hz_40ms_log_rms_mae_db": 0.9580398389935242, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057164531039649844, "300–1000 Hz_40ms_envelope_correlation": 0.9976408557820524, "300–1000 Hz_40ms_modulation_mae_db": 0.49408402392767226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3000848277010344, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13325000972539972, "1000–3000 Hz_10ms_envelope_correlation": 0.9873575347052653, "1000–3000 Hz_10ms_modulation_mae_db": 0.5002998369177264, "1000–3000 Hz_40ms_log_rms_mae_db": 0.76704083842156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07341162135419237, "1000–3000 Hz_40ms_envelope_correlation": 0.9960031006359484, "1000–3000 Hz_40ms_modulation_mae_db": 0.7547074532100829, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5377662116376392, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21798580371379647, "3000–6000 Hz_10ms_envelope_correlation": 0.9694926928571559, "3000–6000 Hz_10ms_modulation_mae_db": 0.8710448829770906, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0174710169364214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14119125977068991, "3000–6000 Hz_40ms_envelope_correlation": 0.9865783074469867, "3000–6000 Hz_40ms_modulation_mae_db": 1.095532907974999, "mr_stft_db": 5.788382152012609, "band_mae_db": 0.13642961787056548, "rms_gain_error_db": -0.04323820960524367, "output_peak": 0.4955545435951636}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0576969375497607, "全帯域_10ms_rms_envelope_nrmse": 0.09088742156792134, "全帯域_10ms_envelope_correlation": 0.9932707027902915, "全帯域_10ms_modulation_mae_db": 0.6003706852867633, "全帯域_40ms_log_rms_mae_db": 0.5835088049440253, "全帯域_40ms_rms_envelope_nrmse": 0.043011172263856305, "全帯域_40ms_envelope_correlation": 0.9984141756927288, "全帯域_40ms_modulation_mae_db": 0.4917629913922223, "300–1000 Hz_10ms_log_rms_mae_db": 1.416754318854517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12274384880154655, "300–1000 Hz_10ms_envelope_correlation": 0.9892920031356706, "300–1000 Hz_10ms_modulation_mae_db": 2.218870192872583, "300–1000 Hz_40ms_log_rms_mae_db": 0.9586896800643462, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04794222551443175, "300–1000 Hz_40ms_envelope_correlation": 0.99829970292642, "300–1000 Hz_40ms_modulation_mae_db": 1.8480610129472506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3372572133550633, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12249325701482336, "1000–3000 Hz_10ms_envelope_correlation": 0.9892932194094374, "1000–3000 Hz_10ms_modulation_mae_db": 0.5896458517749131, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8036938653125189, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06565472226994298, "1000–3000 Hz_40ms_envelope_correlation": 0.9968409107448274, "1000–3000 Hz_40ms_modulation_mae_db": 0.7233009759379531, "3000–6000 Hz_10ms_log_rms_mae_db": 1.528518327693229, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2303501869575026, "3000–6000 Hz_10ms_envelope_correlation": 0.9659764403277208, "3000–6000 Hz_10ms_modulation_mae_db": 0.4427510779680908, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9853074582136716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1534882252738586, "3000–6000 Hz_40ms_envelope_correlation": 0.9840820093906093, "3000–6000 Hz_40ms_modulation_mae_db": 0.7307554855810712, "mr_stft_db": 5.789113384188049, "band_mae_db": 0.06893421491958011, "rms_gain_error_db": -0.026074451453186397, "output_peak": 0.503336295526323}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9728762531674527, "全帯域_10ms_rms_envelope_nrmse": 0.08126229041766025, "全帯域_10ms_envelope_correlation": 0.9946269776984419, "全帯域_10ms_modulation_mae_db": 0.17196369204427622, "全帯域_40ms_log_rms_mae_db": 0.535300265983757, "全帯域_40ms_rms_envelope_nrmse": 0.04116328199294211, "全帯域_40ms_envelope_correlation": 0.9985384350140688, "全帯域_40ms_modulation_mae_db": 0.3675113968913586, "300–1000 Hz_10ms_log_rms_mae_db": 1.4484242904518172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08275044339640446, "300–1000 Hz_10ms_envelope_correlation": 0.9951591834554274, "300–1000 Hz_10ms_modulation_mae_db": 0.9744726071634245, "300–1000 Hz_40ms_log_rms_mae_db": 0.9754166465893039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04010203610866533, "300–1000 Hz_40ms_envelope_correlation": 0.9988706201842562, "300–1000 Hz_40ms_modulation_mae_db": 0.9275996437957742, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3149058669364104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15114502368919022, "1000–3000 Hz_10ms_envelope_correlation": 0.9839720119850944, "1000–3000 Hz_10ms_modulation_mae_db": 0.704864909883815, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7688814363416385, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08563266652077631, "1000–3000 Hz_40ms_envelope_correlation": 0.9946568976002104, "1000–3000 Hz_40ms_modulation_mae_db": 0.5113725537804369, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5137370790196893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21367561819725703, "3000–6000 Hz_10ms_envelope_correlation": 0.9711290087321836, "3000–6000 Hz_10ms_modulation_mae_db": 0.05882274373113861, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9714667049937384, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13358227687023472, "3000–6000 Hz_40ms_envelope_correlation": 0.9880667156354151, "3000–6000 Hz_40ms_modulation_mae_db": 0.08112553529116473, "mr_stft_db": 5.79461967234602, "band_mae_db": 0.06189362593122907, "rms_gain_error_db": -0.0027372268392358906, "output_peak": 0.5258146081857139}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0180070257298302, "全帯域_10ms_rms_envelope_nrmse": 0.07439542847259367, "全帯域_10ms_envelope_correlation": 0.9955111423848819, "全帯域_10ms_modulation_mae_db": 0.2901464936498405, "全帯域_40ms_log_rms_mae_db": 0.5338413014521585, "全帯域_40ms_rms_envelope_nrmse": 0.0408481654208512, "全帯域_40ms_envelope_correlation": 0.9986001530170839, "全帯域_40ms_modulation_mae_db": 0.23514559486349018, "300–1000 Hz_10ms_log_rms_mae_db": 1.372401681704467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1056270621937886, "300–1000 Hz_10ms_envelope_correlation": 0.9921272070203238, "300–1000 Hz_10ms_modulation_mae_db": 1.4359981899137526, "300–1000 Hz_40ms_log_rms_mae_db": 0.9246291360020472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06401066351152612, "300–1000 Hz_40ms_envelope_correlation": 0.9971262896725638, "300–1000 Hz_40ms_modulation_mae_db": 1.1689529056792904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3206558899705405, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10232297477432067, "1000–3000 Hz_10ms_envelope_correlation": 0.9925498322768072, "1000–3000 Hz_10ms_modulation_mae_db": 0.2587908284979387, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7994198352850074, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05576647199098782, "1000–3000 Hz_40ms_envelope_correlation": 0.9976988181066825, "1000–3000 Hz_40ms_modulation_mae_db": 0.30741521720077686, "3000–6000 Hz_10ms_log_rms_mae_db": 1.536044089140021, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1972668641837451, "3000–6000 Hz_10ms_envelope_correlation": 0.9753414072810793, "3000–6000 Hz_10ms_modulation_mae_db": 0.2545595170832926, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9758935836192261, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1253648138582545, "3000–6000 Hz_40ms_envelope_correlation": 0.9894612894910936, "3000–6000 Hz_40ms_modulation_mae_db": 0.4687638770454612, "mr_stft_db": 5.777846846729129, "band_mae_db": 0.14459190346716197, "rms_gain_error_db": -0.06531963246541916, "output_peak": 0.5382148243044971}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83120, "output_samples": 83120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.568832983519937, "spectral_convergence": 0.3062696026948447, "spectral_convergence_status": "ok", "amplitude_floor": 8.933262458231008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.558976272321916, "spectral_convergence": 0.322119845585967, "spectral_convergence_status": "ok", "amplitude_floor": 6.540934930277787e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.203486763855351, "spectral_convergence": 0.33799103404207864, "spectral_convergence_status": "ok", "amplitude_floor": 6.044182898288457e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.147025669043969e-05, "output_power": 3.206161551313699e-05, "delta_db": 0.080851164015862, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00012755476455897642, "output_power": 0.00013085912099077996, "delta_db": 0.11107313250995396, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005321662836872518, "output_power": 0.0005190673227396079, "delta_db": -0.10823666359557671, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022563854189948138, "output_power": 0.00021513504049231862, "delta_db": -0.20702131990128822, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.229072811272528e-05, "output_power": 8.789767224673883e-05, "delta_db": 0.2862646900505634, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.381343402576783e-05, "output_power": 6.0606153519421156e-05, "delta_db": -0.22395394578184571, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1695668193091817, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07129544638742717, "pitch": null, "output_rms_dbfs": -29.80623241947804, "output_peak": 0.5754588164586549, "reference_rms_dbfs": -29.734936973090612, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.777098673232402}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.00015201529585382603, "log_rms_mae_db": 1.0126018135482013, "rms_envelope_nrmse": 0.07962617556111845, "envelope_correlation": 0.9948598013677236, "rms_gain_error_db": -0.07129557917271678, "reference_mean_power": 0.0010639579210132953, "output_mean_power": 0.0010466341338265137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.575807359242935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021400202307720183, "output_power": 0.0002135629688992401, "reference_active": true, "delta_db": -0.00891929308268716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3329698566280647e-05, "output_power": 2.2740746173549868e-05, "reference_active": true, "delta_db": -0.11104416748420182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.358098056371455e-06, "output_power": 6.019197715948964e-06, "reference_active": true, "delta_db": 0.5052795188103467}], "active_band_count": 3, "active_band_mae_db": 0.20841432645907856}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.0001490763524181648, "log_rms_mae_db": 0.547160018585393, "rms_envelope_nrmse": 0.0418546822906201, "envelope_correlation": 0.998550226743756, "rms_gain_error_db": -0.07129610121167715, "reference_mean_power": 0.0010701430709341084, "output_mean_power": 0.0010527184481287336}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 5.309268989226658e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019126377610844124, "output_power": 0.00019190058900119846, "reference_active": true, "delta_db": 0.014435819463097403}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0962051121960504e-05, "output_power": 1.0961808692103653e-05, "reference_active": true, "delta_db": -9.604690960083702e-05}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.152513485967136e-07, "output_power": 3.6395801888593367e-07, "reference_active": true, "delta_db": 0.6239433929778907}], "active_band_count": 3, "active_band_mae_db": 0.2128250864501963}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.0001087945365454221, "log_rms_mae_db": 1.410380353900788, "rms_envelope_nrmse": 0.09935698772214432, "envelope_correlation": 0.9930090931332237, "rms_gain_error_db": -0.10219435075338235, "reference_mean_power": 0.0005739909463294839, "output_mean_power": 0.0005606419712578593}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 3.2669273354505384e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013940368795916017, "output_power": 0.00013740590852655176, "reference_active": true, "delta_db": -0.06268855265384574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.236537072685246e-06, "output_power": 8.226554631264227e-06, "reference_active": true, "delta_db": -0.5029119156818588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7105571121059603e-06, "output_power": 2.777493818491033e-06, "reference_active": true, "delta_db": 2.1051552206720183}], "active_band_count": 3, "active_band_mae_db": 0.8902518963359075}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.00010533749157464783, "log_rms_mae_db": 0.9329757849304141, "rms_envelope_nrmse": 0.04953361620005532, "envelope_correlation": 0.9982825165872146, "rms_gain_error_db": -0.10219435399917197, "reference_mean_power": 0.0005773280667275119, "output_mean_power": 0.0005639014817576931}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 3.141212974645643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001274169753045263, "output_power": 0.00012595806370991483, "reference_active": true, "delta_db": -0.050013155181491326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.860144772024442e-06, "output_power": 3.393148910908712e-06, "reference_active": true, "delta_db": -0.560006734819169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1444270061935373e-07, "output_power": 1.6035237539695113e-07, "reference_active": true, "delta_db": 1.4648730004231727}], "active_band_count": 3, "active_band_mae_db": 0.6916309634746111}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 8.389504112280413e-05, "log_rms_mae_db": 1.3036272553422341, "rms_envelope_nrmse": 0.10788020054700352, "envelope_correlation": 0.9917780031723367, "rms_gain_error_db": -0.13353077784477208, "reference_mean_power": 0.00021157950445783077, "output_mean_power": 0.00020517314606449647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.2662436125548267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5392735637289177e-05, "output_power": 5.240434690730978e-05, "reference_active": true, "delta_db": -0.2408550090542513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790041823950801e-06, "output_power": 5.470808092235721e-06, "reference_active": true, "delta_db": -0.24630220189845708}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4111913003034675e-06, "output_power": 1.2937954568379663e-06, "reference_active": true, "delta_db": -0.3772026867986138}], "active_band_count": 3, "active_band_mae_db": 0.28811996591710737}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 7.722064559288104e-05, "log_rms_mae_db": 0.7880155902188192, "rms_envelope_nrmse": 0.05341347032337685, "envelope_correlation": 0.998052916204116, "rms_gain_error_db": -0.1335310262878866, "reference_mean_power": 0.0002128095577406816, "output_mean_power": 0.0002063659430924296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 1.2078803728484267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0224253455435505e-05, "output_power": 4.7539053307018474e-05, "reference_active": true, "delta_db": -0.2386296108225029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7667526770926532e-06, "output_power": 2.6443504411232286e-06, "reference_active": true, "delta_db": -0.19651329568535003}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.983894982848384e-08, "output_power": 8.460034705872438e-08, "reference_active": true, "delta_db": -0.7192785950544731}], "active_band_count": 3, "active_band_mae_db": 0.384807167187442}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.54681310079566e-05, "log_rms_mae_db": 1.5426902319834124, "rms_envelope_nrmse": 0.23167832070318792, "envelope_correlation": 0.9656283569797596, "rms_gain_error_db": -0.0867291609798207, "reference_mean_power": 0.00010955764718773985, "output_mean_power": 0.00010739146844708729}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 6.937978158746457e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2971450195505076e-05, "output_power": 2.111362589441277e-05, "reference_active": true, "delta_db": -0.36625591057696555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1375769637538917e-05, "output_power": 1.156224425480787e-05, "reference_active": true, "delta_db": 0.07061350804868353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4470462261987434e-06, "output_power": 3.349664464077252e-06, "reference_active": true, "delta_db": 1.3635913241551592}], "active_band_count": 3, "active_band_mae_db": 0.600153580926936}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 5.762149573106217e-05, "log_rms_mae_db": 0.9889864773728022, "rms_envelope_nrmse": 0.15869279877016515, "envelope_correlation": 0.9829735169281411, "rms_gain_error_db": -0.08672938806664443, "reference_mean_power": 0.00011019456862153096, "output_mean_power": 0.00010801579099356519}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 6.599778128222787e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3073624841125754e-05, "output_power": 2.2339062652052797e-05, "reference_active": true, "delta_db": -0.140508809187153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.235055585786147e-06, "output_power": 6.344795220615147e-06, "reference_active": true, "delta_db": 0.0757728014028687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9515593840599056e-07, "output_power": 2.58152588772279e-07, "reference_active": true, "delta_db": 1.2149471356228034}], "active_band_count": 3, "active_band_mae_db": 0.4770762487376084}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9113322554030279, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ab124cfcbb2079ee25d6bbac98c7ecd83a5608db9d8622a6edeb1f1f0f461268", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897645917127876, "standard_deviation": 0.0028830338039482965, "minimum": 0.9839720119850944, "maximum": 0.9925498322768072, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3177853345619148, "standard_deviation": 0.022986492698041885, "minimum": 1.2897298122934044, "maximum": 1.3623414163817218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4334938346460414, "standard_deviation": 0.16558919911999453, "minimum": 0.24053927318866708, "maximum": 0.704864909883815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11912213138465393, "standard_deviation": 0.016351708059061405, "minimum": 0.10232297477432067, "maximum": 0.15114502368919022, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970867718652764, "standard_deviation": 0.0012141215699842114, "minimum": 0.9946568976002104, "maximum": 0.9981984348808097, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7839255124544824, "standard_deviation": 0.012987626205593268, "minimum": 0.76704083842156, "maximum": 0.8036938653125189, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5755928463812352, "standard_deviation": 0.25945397492290645, "minimum": 0.2212446798041899, "maximum": 0.9685681891715756, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06227808684528976, "standard_deviation": 0.01211885375921031, "minimum": 0.049506041301659025, "maximum": 0.08563266652077631, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9696341094187273, "standard_deviation": 0.0036223284282931134, "minimum": 0.9655855060425389, "maximum": 0.9753414072810793, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5291470798851023, "standard_deviation": 0.011350355376362026, "minimum": 1.511433830021437, "maximum": 1.5426902319834124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.38986908271824006, "standard_deviation": 0.25979747776093415, "minimum": 0.05882274373113861, "maximum": 0.8710448829770906, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21770738421830266, "standard_deviation": 0.01280404219944631, "minimum": 0.1972668641837451, "maximum": 0.23221740581055583, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9866719754458604, "standard_deviation": 0.002152292892962586, "minimum": 0.9829735169281411, "maximum": 0.9894612894910936, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9778269676072404, "standard_deviation": 0.021996723476676854, "minimum": 0.9477159687249408, "maximum": 1.0174710169364214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5334678852199543, "standard_deviation": 0.32874232237242235, "minimum": 0.08112553529116473, "maximum": 1.095532907974999, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1402870056312267, "standard_deviation": 0.010968836428202242, "minimum": 0.1253648138582545, "maximum": 0.15869279877016515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9923367198216438, "standard_deviation": 0.0029384938100791504, "minimum": 0.9867445373314069, "maximum": 0.9951591834554274, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4106558819883595, "standard_deviation": 0.022889903620151152, "minimum": 1.372401681704467, "maximum": 1.4484242904518172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.295723858035478, "standard_deviation": 0.5710768947303607, "minimum": 0.8073167803389821, "maximum": 2.218870192872583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1027479891998643, "standard_deviation": 0.01912712699125109, "minimum": 0.08275044339640446, "maximum": 0.1379774681966081, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998110066941787, "standard_deviation": 0.000670151216214566, "minimum": 0.9971262896725638, "maximum": 0.9988706201842562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9597389713274302, "standard_deviation": 0.02328697147477646, "minimum": 0.9246291360020472, "maximum": 0.9902950700473934, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9987859177636236, "standard_deviation": 0.5484794994083613, "minimum": 0.43587728678883847, "maximum": 1.8480610129472506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05112634161223498, "standard_deviation": 0.008879117704095506, "minimum": 0.04010203610866533, "maximum": 0.06401066351152612, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12290337863681397, "standard_deviation": 0.03919691589214847, "minimum": 0.06189362593122907, "maximum": 0.1695668193091817, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.783900361735095, "standard_deviation": 0.008721933565113098, "minimum": 5.7700844008814185, "maximum": 5.79461967234602, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5293117561102367, "standard_deviation": 0.03126157236604457, "minimum": 0.4955545435951636, "maximum": 0.5754588164586549, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.037404720347539715, "standard_deviation": 0.03337876014842415, "minimum": -0.07767498782856763, "maximum": 0.015559145732980457, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946639965943717, "standard_deviation": 0.0010469807246058403, "minimum": 0.9928370855572313, "maximum": 0.9955111423848819, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.017024587763158, "standard_deviation": 0.026676926514378977, "minimum": 0.9728762531674527, "maximum": 1.0576969375497607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35662446092034017, "standard_deviation": 0.2024287229552678, "minimum": 0.07925979915962979, "maximum": 0.6003706852867633, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08074458664018663, "standard_deviation": 0.007680410588287802, "minimum": 0.07439542847259367, "maximum": 0.0940598209864969, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985976132011288, "standard_deviation": 0.0002805317047221167, "minimum": 0.9980891250583557, "maximum": 0.9990007155402691, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5490956993137911, "standard_deviation": 0.024846416559308596, "minimum": 0.5255548219206315, "maximum": 0.5920803987382874, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35147248004445064, "standard_deviation": 0.14367968520271268, "minimum": 0.1220242303348981, "maximum": 0.5017258312869837, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04055119144463496, "standard_deviation": 0.0038861570268077113, "minimum": 0.035030429355904884, "maximum": 0.0473015230678783, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5142-33396-0065", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7207581996917725, "predicted_mos": 2.7207581996917725, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "7a7a7257a4eec31820b96d405148b0e9f45448f5a0fc17e4b45806b480df11ef", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.006951100192964, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0224042973584908, "全帯域_10ms_rms_envelope_nrmse": 0.06654973702578257, "全帯域_10ms_envelope_correlation": 0.9964389664393608, "全帯域_10ms_modulation_mae_db": 0.16954377495372164, "全帯域_40ms_log_rms_mae_db": 0.5467144514812528, "全帯域_40ms_rms_envelope_nrmse": 0.03466807950692692, "全帯域_40ms_envelope_correlation": 0.9990306176707541, "全帯域_40ms_modulation_mae_db": 0.26388176709405814, "300–1000 Hz_10ms_log_rms_mae_db": 1.333422079230325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0830041548146728, "300–1000 Hz_10ms_envelope_correlation": 0.995130793407323, "300–1000 Hz_10ms_modulation_mae_db": 0.37893682939592127, "300–1000 Hz_40ms_log_rms_mae_db": 0.8998528342675867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050453751729886254, "300–1000 Hz_40ms_envelope_correlation": 0.9981769596622787, "300–1000 Hz_40ms_modulation_mae_db": 0.1470285362296024, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3410167569563152, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11956770395374972, "1000–3000 Hz_10ms_envelope_correlation": 0.9899140194079197, "1000–3000 Hz_10ms_modulation_mae_db": 0.6155485311098384, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8071900739555495, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056386297544601845, "1000–3000 Hz_40ms_envelope_correlation": 0.9978963959534491, "1000–3000 Hz_40ms_modulation_mae_db": 0.18289396356459955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4858083759414251, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20502756574314174, "3000–6000 Hz_10ms_envelope_correlation": 0.9730301716264365, "3000–6000 Hz_10ms_modulation_mae_db": 0.784767435041287, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9277726045007801, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13809536123467456, "3000–6000 Hz_40ms_envelope_correlation": 0.987251649155939, "3000–6000 Hz_40ms_modulation_mae_db": 0.9777758711517398, "mr_stft_db": 5.7502346806262326, "band_mae_db": 0.19969579432914256, "rms_gain_error_db": -0.08224311142474473, "output_peak": 0.5261171082880993}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.02980982503484, "全帯域_10ms_rms_envelope_nrmse": 0.07194501712559659, "全帯域_10ms_envelope_correlation": 0.9958031605988837, "全帯域_10ms_modulation_mae_db": 0.2824005326386985, "全帯域_40ms_log_rms_mae_db": 0.5873891836301437, "全帯域_40ms_rms_envelope_nrmse": 0.03610907905837792, "全帯域_40ms_envelope_correlation": 0.9989133135945071, "全帯域_40ms_modulation_mae_db": 0.4623291797120663, "300–1000 Hz_10ms_log_rms_mae_db": 1.3447330455196471, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.096042087205196, "300–1000 Hz_10ms_envelope_correlation": 0.9934365051041657, "300–1000 Hz_10ms_modulation_mae_db": 0.9494441738016245, "300–1000 Hz_40ms_log_rms_mae_db": 0.9068911248255443, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04965808214096063, "300–1000 Hz_40ms_envelope_correlation": 0.9981945025851898, "300–1000 Hz_40ms_modulation_mae_db": 0.7866439611284085, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3107259116067673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11673354430876093, "1000–3000 Hz_10ms_envelope_correlation": 0.9903089903953709, "1000–3000 Hz_10ms_modulation_mae_db": 0.16618508742479787, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7871732455648354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059764352220851864, "1000–3000 Hz_40ms_envelope_correlation": 0.9974532189299691, "1000–3000 Hz_40ms_modulation_mae_db": 0.7753155070260555, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5000729991219715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22793061525654756, "3000–6000 Hz_10ms_envelope_correlation": 0.9667495175837479, "3000–6000 Hz_10ms_modulation_mae_db": 0.8659508993666621, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9367289265893336, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15396913154972217, "3000–6000 Hz_40ms_envelope_correlation": 0.9846796909504899, "3000–6000 Hz_40ms_modulation_mae_db": 1.2225242039255233, "mr_stft_db": 5.760565769757459, "band_mae_db": 0.18961317757872084, "rms_gain_error_db": -0.06437978966248549, "output_peak": 0.5409023844097953}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.013776562868347, "全帯域_10ms_rms_envelope_nrmse": 0.07216708787628417, "全帯域_10ms_envelope_correlation": 0.9958281923936982, "全帯域_10ms_modulation_mae_db": 0.08531234012341493, "全帯域_40ms_log_rms_mae_db": 0.5493583244247692, "全帯域_40ms_rms_envelope_nrmse": 0.03773795452461372, "全帯域_40ms_envelope_correlation": 0.9988785787761935, "全帯域_40ms_modulation_mae_db": 0.09545830236343796, "300–1000 Hz_10ms_log_rms_mae_db": 1.3056230842314753, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0944754311821573, "300–1000 Hz_10ms_envelope_correlation": 0.9937582880314949, "300–1000 Hz_10ms_modulation_mae_db": 0.8170481710162404, "300–1000 Hz_40ms_log_rms_mae_db": 0.8623710425692653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06320846855194176, "300–1000 Hz_40ms_envelope_correlation": 0.9972257760580606, "300–1000 Hz_40ms_modulation_mae_db": 0.6966399678915262, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3035865317180308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11833903552294923, "1000–3000 Hz_10ms_envelope_correlation": 0.990012475244311, "1000–3000 Hz_10ms_modulation_mae_db": 0.5522441880399287, "1000–3000 Hz_40ms_log_rms_mae_db": 0.777208998730195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06081878960217926, "1000–3000 Hz_40ms_envelope_correlation": 0.997311390061066, "1000–3000 Hz_40ms_modulation_mae_db": 0.8108281396952653, "3000–6000 Hz_10ms_log_rms_mae_db": 1.514776395782698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21306172698999262, "3000–6000 Hz_10ms_envelope_correlation": 0.9708464405571967, "3000–6000 Hz_10ms_modulation_mae_db": 0.2522883961081515, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9254207277457861, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14100637397912255, "3000–6000 Hz_40ms_envelope_correlation": 0.986701596780784, "3000–6000 Hz_40ms_modulation_mae_db": 0.1809153563294298, "mr_stft_db": 5.75729441330017, "band_mae_db": 0.1576127353068581, "rms_gain_error_db": -0.10078887440799057, "output_peak": 0.5503309343059386}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0090372046695788, "全帯域_10ms_rms_envelope_nrmse": 0.07531321163120153, "全帯域_10ms_envelope_correlation": 0.9955356824920122, "全帯域_10ms_modulation_mae_db": 0.6937099861689894, "全帯域_40ms_log_rms_mae_db": 0.5301385942449127, "全帯域_40ms_rms_envelope_nrmse": 0.04259225621237664, "全帯域_40ms_envelope_correlation": 0.9986425864751702, "全帯域_40ms_modulation_mae_db": 0.6236693293829046, "300–1000 Hz_10ms_log_rms_mae_db": 1.3366216792659065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08203236606664037, "300–1000 Hz_10ms_envelope_correlation": 0.9953738949800714, "300–1000 Hz_10ms_modulation_mae_db": 0.6938700581312266, "300–1000 Hz_40ms_log_rms_mae_db": 0.8701205123383322, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051697838120232896, "300–1000 Hz_40ms_envelope_correlation": 0.9982427331706295, "300–1000 Hz_40ms_modulation_mae_db": 0.42542742340644174, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2786234952912194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10284788600210652, "1000–3000 Hz_10ms_envelope_correlation": 0.9927704391546567, "1000–3000 Hz_10ms_modulation_mae_db": 0.5339737444131957, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7496635679904516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052787168675696584, "1000–3000 Hz_40ms_envelope_correlation": 0.9983952060579793, "1000–3000 Hz_40ms_modulation_mae_db": 0.512692525396868, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4976309569276365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2431042647377164, "3000–6000 Hz_10ms_envelope_correlation": 0.9619332643221822, "3000–6000 Hz_10ms_modulation_mae_db": 0.7802282987927676, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9540889391151137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15149738420162448, "3000–6000 Hz_40ms_envelope_correlation": 0.9845419932244541, "3000–6000 Hz_40ms_modulation_mae_db": 0.8583419492420704, "mr_stft_db": 5.742567659632719, "band_mae_db": 0.1444363886570957, "rms_gain_error_db": -0.13827676020171156, "output_peak": 0.5255036629270692}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.019911920560804, "全帯域_10ms_rms_envelope_nrmse": 0.07413496257072012, "全帯域_10ms_envelope_correlation": 0.9955265448365446, "全帯域_10ms_modulation_mae_db": 0.21700251413721974, "全帯域_40ms_log_rms_mae_db": 0.5795221942590094, "全帯域_40ms_rms_envelope_nrmse": 0.034142150966994626, "全帯域_40ms_envelope_correlation": 0.9990085566757494, "全帯域_40ms_modulation_mae_db": 0.16487997090286519, "300–1000 Hz_10ms_log_rms_mae_db": 1.331965234441609, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08578954597504242, "300–1000 Hz_10ms_envelope_correlation": 0.9947665544969573, "300–1000 Hz_10ms_modulation_mae_db": 0.9376307776638014, "300–1000 Hz_40ms_log_rms_mae_db": 0.9116293105839056, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04434845907286464, "300–1000 Hz_40ms_envelope_correlation": 0.9985500662002226, "300–1000 Hz_40ms_modulation_mae_db": 0.8978396955449339, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3151652566211351, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1289654995618396, "1000–3000 Hz_10ms_envelope_correlation": 0.9881502713228961, "1000–3000 Hz_10ms_modulation_mae_db": 0.8863945746047467, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7443802986847011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.055863762655960415, "1000–3000 Hz_40ms_envelope_correlation": 0.9976965920594781, "1000–3000 Hz_40ms_modulation_mae_db": 0.6365268110686705, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5029693097358807, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20958565955661237, "3000–6000 Hz_10ms_envelope_correlation": 0.9719704679236784, "3000–6000 Hz_10ms_modulation_mae_db": 0.9480155953178762, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9650417478870607, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14054406128279875, "3000–6000 Hz_40ms_envelope_correlation": 0.9872257242039363, "3000–6000 Hz_40ms_modulation_mae_db": 1.0987555350564058, "mr_stft_db": 5.7596538033861515, "band_mae_db": 0.200938397809007, "rms_gain_error_db": -0.03630940150185158, "output_peak": 0.5244591447473906}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0139945585774186, "全帯域_10ms_rms_envelope_nrmse": 0.0672846610211628, "全帯域_10ms_envelope_correlation": 0.9964373273139118, "全帯域_10ms_modulation_mae_db": 0.423113202034263, "全帯域_40ms_log_rms_mae_db": 0.5667327540050311, "全帯域_40ms_rms_envelope_nrmse": 0.03738812851571181, "全帯域_40ms_envelope_correlation": 0.9989537045237409, "全帯域_40ms_modulation_mae_db": 0.3564592723415955, "300–1000 Hz_10ms_log_rms_mae_db": 1.391792915750177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1048308071605266, "300–1000 Hz_10ms_envelope_correlation": 0.9925184738327731, "300–1000 Hz_10ms_modulation_mae_db": 0.36884516354390917, "300–1000 Hz_40ms_log_rms_mae_db": 0.9204694725501733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07441457129333799, "300–1000 Hz_40ms_envelope_correlation": 0.9963832688944155, "300–1000 Hz_40ms_modulation_mae_db": 0.4828475954171075, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3647022795038555, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1220586527209253, "1000–3000 Hz_10ms_envelope_correlation": 0.9894044179853293, "1000–3000 Hz_10ms_modulation_mae_db": 0.43648292472393585, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7729747965915219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05815797466925936, "1000–3000 Hz_40ms_envelope_correlation": 0.9975194760485024, "1000–3000 Hz_40ms_modulation_mae_db": 0.7674230942018845, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5121580857155177, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23309721486835955, "3000–6000 Hz_10ms_envelope_correlation": 0.9652636155771491, "3000–6000 Hz_10ms_modulation_mae_db": 0.15931368581423008, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9669947917452119, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15498793235573896, "3000–6000 Hz_40ms_envelope_correlation": 0.9837691470511237, "3000–6000 Hz_40ms_modulation_mae_db": 0.17840934433399736, "mr_stft_db": 5.7682189907935735, "band_mae_db": 0.15008974356163432, "rms_gain_error_db": -0.11852883056045818, "output_peak": 0.5214706332959291}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0634068940549803, "全帯域_10ms_rms_envelope_nrmse": 0.08238128030029893, "全帯域_10ms_envelope_correlation": 0.9947851957081987, "全帯域_10ms_modulation_mae_db": 0.31322672512321886, "全帯域_40ms_log_rms_mae_db": 0.648302993417753, "全帯域_40ms_rms_envelope_nrmse": 0.046760844289075065, "全帯域_40ms_envelope_correlation": 0.9985306664476432, "全帯域_40ms_modulation_mae_db": 0.4799982029149241, "300–1000 Hz_10ms_log_rms_mae_db": 1.3759805141677333, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1162201700404317, "300–1000 Hz_10ms_envelope_correlation": 0.9911043191076436, "300–1000 Hz_10ms_modulation_mae_db": 0.9508334889991933, "300–1000 Hz_40ms_log_rms_mae_db": 0.9309394492481533, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0815336579165358, "300–1000 Hz_40ms_envelope_correlation": 0.9960534799586342, "300–1000 Hz_40ms_modulation_mae_db": 0.7220312426242229, "1000–3000 Hz_10ms_log_rms_mae_db": 1.321884289235642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13522134800503555, "1000–3000 Hz_10ms_envelope_correlation": 0.9873109140410304, "1000–3000 Hz_10ms_modulation_mae_db": 0.7782970303848007, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7910624116720929, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08019686217513639, "1000–3000 Hz_40ms_envelope_correlation": 0.9954437717736228, "1000–3000 Hz_40ms_modulation_mae_db": 0.6964290200648545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5137955586273966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2223406566382625, "3000–6000 Hz_10ms_envelope_correlation": 0.9685014318984583, "3000–6000 Hz_10ms_modulation_mae_db": 0.8071445849666868, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9604393271508546, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14946182292692853, "3000–6000 Hz_40ms_envelope_correlation": 0.9857672460040214, "3000–6000 Hz_40ms_modulation_mae_db": 0.983980948735122, "mr_stft_db": 5.777498160892022, "band_mae_db": 0.1871477589242684, "rms_gain_error_db": -0.19999488587203373, "output_peak": 0.5275020484049739}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0742213344090552, "全帯域_10ms_rms_envelope_nrmse": 0.07311826756480559, "全帯域_10ms_envelope_correlation": 0.9960482376275098, "全帯域_10ms_modulation_mae_db": 0.42204553042530263, "全帯域_40ms_log_rms_mae_db": 0.5919239683533802, "全帯域_40ms_rms_envelope_nrmse": 0.041723809847828336, "全帯域_40ms_envelope_correlation": 0.9989914164950995, "全帯域_40ms_modulation_mae_db": 0.6386065582993231, "300–1000 Hz_10ms_log_rms_mae_db": 1.357935614425341, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11270202701493093, "300–1000 Hz_10ms_envelope_correlation": 0.9914267659539682, "300–1000 Hz_10ms_modulation_mae_db": 1.1564401556650519, "300–1000 Hz_40ms_log_rms_mae_db": 0.9076173470130652, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07172819026524235, "300–1000 Hz_40ms_envelope_correlation": 0.996893655498781, "300–1000 Hz_40ms_modulation_mae_db": 0.8755244438819775, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2858249393811998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11615876735136613, "1000–3000 Hz_10ms_envelope_correlation": 0.9904007674799445, "1000–3000 Hz_10ms_modulation_mae_db": 0.6533116482961635, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7703154569151208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0664713781635581, "1000–3000 Hz_40ms_envelope_correlation": 0.9967971506600679, "1000–3000 Hz_40ms_modulation_mae_db": 0.9160311296927066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4972087557956761, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21397620039515083, "3000–6000 Hz_10ms_envelope_correlation": 0.9712195208523989, "3000–6000 Hz_10ms_modulation_mae_db": 1.207972442507043, "3000–6000 Hz_40ms_log_rms_mae_db": 0.972501567105879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15583661459770556, "3000–6000 Hz_40ms_envelope_correlation": 0.9847932046597502, "3000–6000 Hz_40ms_modulation_mae_db": 1.4363016613434745, "mr_stft_db": 5.746297048609443, "band_mae_db": 0.24009067956615507, "rms_gain_error_db": -0.21086687796189668, "output_peak": 0.5213385057486603}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83120, "output_samples": 83120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.549963229223995, "spectral_convergence": 0.3032040143463802, "spectral_convergence_status": "ok", "amplitude_floor": 8.933262458231008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5228325969808445, "spectral_convergence": 0.30738410932526034, "spectral_convergence_status": "ok", "amplitude_floor": 6.540934930277787e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.177908215673859, "spectral_convergence": 0.32952761301436784, "spectral_convergence_status": "ok", "amplitude_floor": 6.044182898288457e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.147025669043969e-05, "output_power": 3.388223557496564e-05, "delta_db": 0.32071772379187147, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00012755476455897642, "output_power": 0.00013232447743602958, "delta_db": 0.15943502119324093, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005321662836872518, "output_power": 0.0005188994309865331, "delta_db": -0.10964161153425764, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022563854189948138, "output_power": 0.00021453353422280777, "delta_db": -0.21918097223360636, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.229072811272528e-05, "output_power": 8.388264357691989e-05, "delta_db": 0.08321203877557698, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.381343402576783e-05, "output_power": 5.947211401190476e-05, "delta_db": -0.3059873984463019, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19969579432914256, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08224311142474473, "pitch": null, "output_rms_dbfs": -29.817180084515357, "output_peak": 0.5261171082880993, "reference_rms_dbfs": -29.734936973090612, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7502346806262326}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.00015201529585382603, "log_rms_mae_db": 1.0224042973584908, "rms_envelope_nrmse": 0.06654973702578257, "envelope_correlation": 0.9964389664393608, "rms_gain_error_db": -0.0822431183530383, "reference_mean_power": 0.0010639579210132953, "output_mean_power": 0.0010439991386220822}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.575807359242935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021400202307720183, "output_power": 0.00020520795266128588, "reference_active": true, "delta_db": -0.1822369150729916}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3329698566280647e-05, "output_power": 2.3585190204634817e-05, "reference_active": true, "delta_db": 0.04730255513675877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.358098056371455e-06, "output_power": 5.024599742170142e-06, "reference_active": true, "delta_db": -0.2790918546514145}], "active_band_count": 3, "active_band_mae_db": 0.16954377495372164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.0001490763524181648, "log_rms_mae_db": 0.5467144514812528, "rms_envelope_nrmse": 0.03466807950692692, "envelope_correlation": 0.9990306176707541, "rms_gain_error_db": -0.08224243847293615, "reference_mean_power": 0.0010701430709341084, "output_mean_power": 0.0010500684257265188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 5.309268989226658e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019126377610844124, "output_power": 0.00018408639354309628, "reference_active": true, "delta_db": -0.16611036277736926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0962051121960504e-05, "output_power": 1.1180718845655743e-05, "reference_active": true, "delta_db": 0.08577903596397576}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.152513485967136e-07, "output_power": 2.784077771389293e-07, "reference_active": true, "delta_db": -0.5397559025408295}], "active_band_count": 3, "active_band_mae_db": 0.26388176709405814}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.0001087945365454221, "log_rms_mae_db": 1.333422079230325, "rms_envelope_nrmse": 0.0830041548146728, "envelope_correlation": 0.995130793407323, "rms_gain_error_db": -0.0767198931353866, "reference_mean_power": 0.0005739909463294839, "output_mean_power": 0.0005639401985987207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 3.2669273354505384e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013940368795916017, "output_power": 0.00013436242120229284, "reference_active": true, "delta_db": -0.15996442083157592}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.236537072685246e-06, "output_power": 9.393497171933163e-06, "reference_active": true, "delta_db": 0.07318131254242737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7105571121059603e-06, "output_power": 2.1062215621076273e-06, "reference_active": true, "delta_db": 0.9036647548137605}], "active_band_count": 3, "active_band_mae_db": 0.37893682939592127}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.00010533749157464783, "log_rms_mae_db": 0.8998528342675867, "rms_envelope_nrmse": 0.050453751729886254, "envelope_correlation": 0.9981769596622787, "rms_gain_error_db": -0.07671990382542923, "reference_mean_power": 0.0005773280667275119, "output_mean_power": 0.0005672188836557645}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 3.141212974645643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001274169753045263, "output_power": 0.0001231982445522164, "reference_active": true, "delta_db": -0.14622771718372943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.860144772024442e-06, "output_power": 3.963884881902041e-06, "reference_active": true, "delta_db": 0.11517440443344665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1444270061935373e-07, "output_power": 1.098044014772692e-07, "reference_active": true, "delta_db": -0.17968348707163115}], "active_band_count": 3, "active_band_mae_db": 0.1470285362296024}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 8.389504112280413e-05, "log_rms_mae_db": 1.3410167569563152, "rms_envelope_nrmse": 0.11956770395374972, "envelope_correlation": 0.9899140194079197, "rms_gain_error_db": -0.18001502413691725, "reference_mean_power": 0.00021157950445783077, "output_mean_power": 0.00020298880796479438}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.2662436125548267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5392735637289177e-05, "output_power": 4.969921962374048e-05, "reference_active": true, "delta_db": -0.4710324433225569}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790041823950801e-06, "output_power": 6.018601831537505e-06, "reference_active": true, "delta_db": 0.16813912122899785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4111913003034675e-06, "output_power": 1.8635167183672736e-06, "reference_active": true, "delta_db": 1.2074740287779604}], "active_band_count": 3, "active_band_mae_db": 0.6155485311098384}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 7.722064559288104e-05, "log_rms_mae_db": 0.8071900739555495, "rms_envelope_nrmse": 0.056386297544601845, "envelope_correlation": 0.9978963959534491, "rms_gain_error_db": -0.18001513999825497, "reference_mean_power": 0.0002128095577406816, "output_mean_power": 0.00020416891233180867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 1.2078803728484267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0224253455435505e-05, "output_power": 4.5589260405392284e-05, "reference_active": true, "delta_db": -0.42050943318321315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7667526770926532e-06, "output_power": 2.745038063465894e-06, "reference_active": true, "delta_db": -0.034219679626703046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.983894982848384e-08, "output_power": 1.0202234037648522e-07, "reference_active": true, "delta_db": 0.09395277788388252}], "active_band_count": 3, "active_band_mae_db": 0.18289396356459955}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.54681310079566e-05, "log_rms_mae_db": 1.4858083759414251, "rms_envelope_nrmse": 0.20502756574314174, "envelope_correlation": 0.9730301716264365, "rms_gain_error_db": -0.20269699904560384, "reference_mean_power": 0.00010955764718773985, "output_mean_power": 0.00010456178704038544}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 6.937978158746457e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2971450195505076e-05, "output_power": 2.090123407760453e-05, "reference_active": true, "delta_db": -0.41016484176925183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1375769637538917e-05, "output_power": 9.681596134591682e-06, "reference_active": true, "delta_db": -0.7003382669721683}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4470462261987434e-06, "output_power": 1.8376476937454628e-06, "reference_active": true, "delta_db": -1.2437991963824406}], "active_band_count": 3, "active_band_mae_db": 0.784767435041287}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 5.762149573106217e-05, "log_rms_mae_db": 0.9277726045007801, "rms_envelope_nrmse": 0.13809536123467456, "envelope_correlation": 0.987251649155939, "rms_gain_error_db": -0.2026975878878961, "reference_mean_power": 0.00011019456862153096, "output_mean_power": 0.00010516965041451478}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 6.599778128222787e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3073624841125754e-05, "output_power": 2.1192401773563068e-05, "reference_active": true, "delta_db": -0.3693564818091387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.235055585786147e-06, "output_power": 5.103463616014704e-06, "reference_active": true, "delta_db": -0.8697530667154897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9515593840599056e-07, "output_power": 1.3211738359256165e-07, "reference_active": true, "delta_db": -1.694218064930591}], "active_band_count": 3, "active_band_mae_db": 0.9777758711517398}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9048819287123497, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "702fed22ff3153cd3b2785cc1aee186c19dc194252fcde3c1fbd51b4579e5851", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897840368789323, "standard_deviation": 0.0016289018164543998, "minimum": 0.9873109140410304, "maximum": 0.9927704391546567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3151911825392706, "standard_deviation": 0.028074609573885842, "minimum": 1.2786234952912194, "maximum": 1.3647022795038555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.577804716124676, "standard_deviation": 0.21860852922276974, "minimum": 0.16618508742479787, "maximum": 0.8863945746047467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11998655467834163, "standard_deviation": 0.009553396709337346, "minimum": 0.10284788600210652, "maximum": 0.13522134800503555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973141501930168, "standard_deviation": 0.0008848780446717549, "minimum": 0.9954437717736228, "maximum": 0.9983952060579793, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7749961062630585, "standard_deviation": 0.02088340653670895, "minimum": 0.7443802986847011, "maximum": 0.8071900739555495, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.662267523838863, "standard_deviation": 0.2281306164985014, "minimum": 0.18289396356459955, "maximum": 0.9160311296927066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06130582321340547, "standard_deviation": 0.008639294176383507, "minimum": 0.052787168675696584, "maximum": 0.08019686217513639, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.968689303792656, "standard_deviation": 0.0038142942514547843, "minimum": 0.9619332643221822, "maximum": 0.9730301716264365, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5030525547060254, "standard_deviation": 0.010039645472217652, "minimum": 1.4858083759414251, "maximum": 1.514776395782698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.725710167239338, "standard_deviation": 0.35044640265520255, "minimum": 0.15931368581423008, "maximum": 1.207972442507043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22101548802322296, "standard_deviation": 0.012997394020720995, "minimum": 0.20502756574314174, "maximum": 0.2431042647377164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9855912815038123, "standard_deviation": 0.001340633701044594, "minimum": 0.9837691470511237, "maximum": 0.987251649155939, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9511235789800025, "standard_deviation": 0.018560004270063563, "minimum": 0.9254207277457861, "maximum": 0.972501567105879, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8671256087647203, "standard_deviation": 0.4590537426029667, "minimum": 0.17840934433399736, "maximum": 1.4363016613434745, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14817483526603945, "standard_deviation": 0.007197241913467804, "minimum": 0.13809536123467456, "maximum": 0.15583661459770556, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9934394493642997, "standard_deviation": 0.0016395784765423573, "minimum": 0.9911043191076436, "maximum": 0.9953738949800714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.347259270879027, "standard_deviation": 0.027257049638038107, "minimum": 1.3056230842314753, "maximum": 1.391792915750177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7816311022771212, "standard_deviation": 0.2836294559869233, "minimum": 0.36884516354390917, "maximum": 1.1564401556650519, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09688707368244977, "standard_deviation": 0.013256088535458423, "minimum": 0.08203236606664037, "maximum": 0.1162201700404317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974650552535265, "standard_deviation": 0.0009537513712468213, "minimum": 0.9960534799586342, "maximum": 0.9985500662002226, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9012363866745032, "standard_deviation": 0.023644739462439363, "minimum": 0.8623710425692653, "maximum": 0.9309394492481533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6292478582655276, "standard_deviation": 0.258077784204232, "minimum": 0.1470285362296024, "maximum": 0.8978396955449339, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06088037738637529, "standard_deviation": 0.013758660845100565, "minimum": 0.04434845907286464, "maximum": 0.0815336579165358, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18370308446661027, "standard_deviation": 0.03191868029954036, "minimum": 0.1444363886570957, "maximum": 0.24009067956615507, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.757791315874721, "standard_deviation": 0.011523207245842344, "minimum": 5.742567659632719, "maximum": 5.777498160892022, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5297030527659821, "standard_deviation": 0.010361396538199837, "minimum": 0.5213385057486603, "maximum": 0.5503309343059386, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11892356644914658, "standard_deviation": 0.06196229263726362, "minimum": -0.21086687796189668, "maximum": -0.03630940150185158, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.995800413426265, "standard_deviation": 0.0005413420875913547, "minimum": 0.9947851957081987, "maximum": 0.9964389664393608, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0308203246916894, "standard_deviation": 0.02444958835174487, "minimum": 1.0090372046695788, "maximum": 1.0742213344090552, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3257943257006036, "standard_deviation": 0.18906523267280625, "minimum": 0.08531234012341493, "maximum": 0.6937099861689894, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07286177813948154, "standard_deviation": 0.004936545591898034, "minimum": 0.06654973702578257, "maximum": 0.08238128030029893, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988686800823572, "standard_deviation": 0.00018343714282325845, "minimum": 0.9985306664476432, "maximum": 0.9990306176707541, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5750103079770315, "standard_deviation": 0.03659069295152129, "minimum": 0.5301385942449127, "maximum": 0.648302993417753, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38566032287639684, "standard_deviation": 0.20144184873046253, "minimum": 0.09545830236343796, "maximum": 0.6386065582993231, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03889028786523813, "standard_deviation": 0.004397570751188066, "minimum": 0.034142150966994626, "maximum": 0.046760844289075065, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5639-40744-0013", "language": "en", "style": "normal", "speaker": "librispeech_5639", "source_utterance_id": "5639-40744-0013", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5639-40744-0013.wav", "source_file_sha256": "9a13d3ca62f03bd2a6ae4c168a38c53491e373a85d1cc18ed7a1e77d3ba397bb", "decoded_pcm_sha256": "ed7f1ce67f421ec6afa8566971a1c9c8ec8d332f4950f7baa3aacb009b96f5ef", "original_source_sha256": "a2688cb66fd448ee0d578c6d82312afea0f9ad95b24350f251eb79ca5a4f759e", "corpus_entry_id": "librispeech_5639-40744-0013"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.079737186431885, "predicted_mos": 4.079737186431885, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "af23c5afae81553f107efba31683fcf029d531d83ad635a7b328e5995e91ed40", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.90743279200251, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5639-40744-0013", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "af23c5afae81553f107efba31683fcf029d531d83ad635a7b328e5995e91ed40", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "7ddaea57f9e5ab7533b5a1b89a5446fc88a177a7e3314f61f6d0dd3f1f5f3321"}, "rows": [{"case": "librispeech_5639-40744-0013", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.798445224761963, "predicted_mos": 2.798445224761963, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6f9d69a0f2632bd021ed25ef8d6ed704d46ca1385c047d070b7793e4c8a1410f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3142361040008836, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5639-40744-0013", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6f9d69a0f2632bd021ed25ef8d6ed704d46ca1385c047d070b7793e4c8a1410f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9014545113645236, "全帯域_10ms_rms_envelope_nrmse": 0.08061944540538701, "全帯域_10ms_envelope_correlation": 0.9940347561419645, "全帯域_10ms_modulation_mae_db": 0.14931724032907134, "全帯域_40ms_log_rms_mae_db": 0.498885231044676, "全帯域_40ms_rms_envelope_nrmse": 0.046925673443768204, "全帯域_40ms_envelope_correlation": 0.9978631359446367, "全帯域_40ms_modulation_mae_db": 0.11141935237404409, "300–1000 Hz_10ms_log_rms_mae_db": 1.8175860748900303, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11285923284560619, "300–1000 Hz_10ms_envelope_correlation": 0.9911494182731452, "300–1000 Hz_10ms_modulation_mae_db": 0.33517322218644513, "300–1000 Hz_40ms_log_rms_mae_db": 1.3263044038247092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07172425827774326, "300–1000 Hz_40ms_envelope_correlation": 0.9964584820061426, "300–1000 Hz_40ms_modulation_mae_db": 0.30059660585631914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8518672325853556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1571137278826271, "1000–3000 Hz_10ms_envelope_correlation": 0.9826330607940433, "1000–3000 Hz_10ms_modulation_mae_db": 0.5874249147893577, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1510367207335064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09700217540608558, "1000–3000 Hz_40ms_envelope_correlation": 0.9930292857555245, "1000–3000 Hz_40ms_modulation_mae_db": 0.465791709441064, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8941202918804405, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2025318590338308, "3000–6000 Hz_10ms_envelope_correlation": 0.9732906489094019, "3000–6000 Hz_10ms_modulation_mae_db": 1.6669975875132501, "3000–6000 Hz_40ms_log_rms_mae_db": 1.25971999369282, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13038008224663633, "3000–6000 Hz_40ms_envelope_correlation": 0.9887791395897425, "3000–6000 Hz_40ms_modulation_mae_db": 2.019772692631667, "mr_stft_db": 5.863745220170164, "band_mae_db": 0.1785753609457441, "rms_gain_error_db": -0.0855157752248928, "output_peak": 0.2944582731796088}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8768445889642417, "全帯域_10ms_rms_envelope_nrmse": 0.07385661100824768, "全帯域_10ms_envelope_correlation": 0.9950053985321595, "全帯域_10ms_modulation_mae_db": 0.4533683045924519, "全帯域_40ms_log_rms_mae_db": 0.4311435575253929, "全帯域_40ms_rms_envelope_nrmse": 0.03588156045471693, "全帯域_40ms_envelope_correlation": 0.998784411484429, "全帯域_40ms_modulation_mae_db": 0.3944715561882277, "300–1000 Hz_10ms_log_rms_mae_db": 1.8268974249251944, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10706052428408774, "300–1000 Hz_10ms_envelope_correlation": 0.9920908022415051, "300–1000 Hz_10ms_modulation_mae_db": 0.8016850928090696, "300–1000 Hz_40ms_log_rms_mae_db": 1.2549909729546616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06496652441288736, "300–1000 Hz_40ms_envelope_correlation": 0.9971692355526685, "300–1000 Hz_40ms_modulation_mae_db": 0.9400184464304164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.888471443434016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1386725824397427, "1000–3000 Hz_10ms_envelope_correlation": 0.9866766431732884, "1000–3000 Hz_10ms_modulation_mae_db": 0.192831826627498, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1045413867797154, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0875388574902267, "1000–3000 Hz_40ms_envelope_correlation": 0.994627939647176, "1000–3000 Hz_40ms_modulation_mae_db": 0.27234075410714514, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8418360518949297, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1847204225634046, "3000–6000 Hz_10ms_envelope_correlation": 0.9766143753940251, "3000–6000 Hz_10ms_modulation_mae_db": 0.5757682941586837, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2130400092204183, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10736935268469255, "3000–6000 Hz_40ms_envelope_correlation": 0.9916886598940146, "3000–6000 Hz_40ms_modulation_mae_db": 0.49201650881882114, "mr_stft_db": 5.862931868827154, "band_mae_db": 0.15273734869332412, "rms_gain_error_db": -0.08559479388741979, "output_peak": 0.2858540308680077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8399278388419584, "全帯域_10ms_rms_envelope_nrmse": 0.0691660311965521, "全帯域_10ms_envelope_correlation": 0.9958417229817774, "全帯域_10ms_modulation_mae_db": 0.40884472873905775, "全帯域_40ms_log_rms_mae_db": 0.45111843403269003, "全帯域_40ms_rms_envelope_nrmse": 0.04091227677002316, "全帯域_40ms_envelope_correlation": 0.9986423611020905, "全帯域_40ms_modulation_mae_db": 0.4296035177563499, "300–1000 Hz_10ms_log_rms_mae_db": 1.7795806809800567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10158001606447498, "300–1000 Hz_10ms_envelope_correlation": 0.9930333972600089, "300–1000 Hz_10ms_modulation_mae_db": 0.5447575434796363, "300–1000 Hz_40ms_log_rms_mae_db": 1.2304274946407405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06607202666752407, "300–1000 Hz_40ms_envelope_correlation": 0.9971845814826464, "300–1000 Hz_40ms_modulation_mae_db": 0.6553869046215562, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8814638445805731, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17485679016186842, "1000–3000 Hz_10ms_envelope_correlation": 0.9793852625718602, "1000–3000 Hz_10ms_modulation_mae_db": 1.2149275078771538, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1644556099764374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10790052383075908, "1000–3000 Hz_40ms_envelope_correlation": 0.992935037803926, "1000–3000 Hz_40ms_modulation_mae_db": 1.55415723765548, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8536678460201406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16629055810075613, "3000–6000 Hz_10ms_envelope_correlation": 0.9810472825928539, "3000–6000 Hz_10ms_modulation_mae_db": 1.0625755613187733, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2591618796800514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09329598335556796, "3000–6000 Hz_40ms_envelope_correlation": 0.9944042631239065, "3000–6000 Hz_40ms_modulation_mae_db": 1.1267006684101395, "mr_stft_db": 5.860092748053252, "band_mae_db": 0.27396361385713813, "rms_gain_error_db": -0.16565894694742458, "output_peak": 0.27416814972927483}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8613827953333949, "全帯域_10ms_rms_envelope_nrmse": 0.08313916578793384, "全帯域_10ms_envelope_correlation": 0.9936868446398601, "全帯域_10ms_modulation_mae_db": 0.43398544776498843, "全帯域_40ms_log_rms_mae_db": 0.4635227905679879, "全帯域_40ms_rms_envelope_nrmse": 0.04346679891697831, "全帯域_40ms_envelope_correlation": 0.998228874530993, "全帯域_40ms_modulation_mae_db": 0.7110994965418801, "300–1000 Hz_10ms_log_rms_mae_db": 1.8383277022214226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11653608730313282, "300–1000 Hz_10ms_envelope_correlation": 0.9904968452080164, "300–1000 Hz_10ms_modulation_mae_db": 0.37518620071882114, "300–1000 Hz_40ms_log_rms_mae_db": 1.272696315156162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0649254154934318, "300–1000 Hz_40ms_envelope_correlation": 0.9971481343563389, "300–1000 Hz_40ms_modulation_mae_db": 0.6273729016367747, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8321937980891612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14241455483224139, "1000–3000 Hz_10ms_envelope_correlation": 0.9861217663426293, "1000–3000 Hz_10ms_modulation_mae_db": 1.3205850462206603, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1568965611443753, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09182586011043686, "1000–3000 Hz_40ms_envelope_correlation": 0.9943076924682809, "1000–3000 Hz_40ms_modulation_mae_db": 1.4939475830856945, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8383876392527296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18089587680247332, "3000–6000 Hz_10ms_envelope_correlation": 0.9771099271022939, "3000–6000 Hz_10ms_modulation_mae_db": 2.027215743100173, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1462992464255577, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09180001997578614, "3000–6000 Hz_40ms_envelope_correlation": 0.994004294395892, "3000–6000 Hz_40ms_modulation_mae_db": 2.186083690899373, "mr_stft_db": 5.850023618403692, "band_mae_db": 0.2182944887400822, "rms_gain_error_db": -0.10888293069873803, "output_peak": 0.27947683398072676}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.915323284246818, "全帯域_10ms_rms_envelope_nrmse": 0.08489135649341817, "全帯域_10ms_envelope_correlation": 0.9933254556840903, "全帯域_10ms_modulation_mae_db": 0.1171087790763801, "全帯域_40ms_log_rms_mae_db": 0.5198234305532147, "全帯域_40ms_rms_envelope_nrmse": 0.05070034087092733, "全帯域_40ms_envelope_correlation": 0.9973993619094427, "全帯域_40ms_modulation_mae_db": 0.1253664303161491, "300–1000 Hz_10ms_log_rms_mae_db": 1.9026263369309542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13345448311658717, "300–1000 Hz_10ms_envelope_correlation": 0.9872286535534023, "300–1000 Hz_10ms_modulation_mae_db": 0.3456734665107422, "300–1000 Hz_40ms_log_rms_mae_db": 1.3246116996638542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08685795496398388, "300–1000 Hz_40ms_envelope_correlation": 0.9942478928581708, "300–1000 Hz_40ms_modulation_mae_db": 0.48319431883303104, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8599563524643268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15763251194918104, "1000–3000 Hz_10ms_envelope_correlation": 0.9825501791333919, "1000–3000 Hz_10ms_modulation_mae_db": 0.5867715492194887, "1000–3000 Hz_40ms_log_rms_mae_db": 1.15541607598888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09463303611949687, "1000–3000 Hz_40ms_envelope_correlation": 0.9933656741361937, "1000–3000 Hz_40ms_modulation_mae_db": 0.5425704296897981, "3000–6000 Hz_10ms_log_rms_mae_db": 1.892421168088738, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20134133531301537, "3000–6000 Hz_10ms_envelope_correlation": 0.9721427411772253, "3000–6000 Hz_10ms_modulation_mae_db": 0.7209491036420034, "3000–6000 Hz_40ms_log_rms_mae_db": 1.216667965069829, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10461527486188803, "3000–6000 Hz_40ms_envelope_correlation": 0.9920989691487477, "3000–6000 Hz_40ms_modulation_mae_db": 0.5939056868336365, "mr_stft_db": 5.844151679977483, "band_mae_db": 0.1394111641767064, "rms_gain_error_db": -0.0500798482326022, "output_peak": 0.27094478228217095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8420595838748818, "全帯域_10ms_rms_envelope_nrmse": 0.07195552251946462, "全帯域_10ms_envelope_correlation": 0.9952050595137134, "全帯域_10ms_modulation_mae_db": 0.1497982036119782, "全帯域_40ms_log_rms_mae_db": 0.4467727848164033, "全帯域_40ms_rms_envelope_nrmse": 0.0371324057656668, "全帯域_40ms_envelope_correlation": 0.9986063338587664, "全帯域_40ms_modulation_mae_db": 0.07787665555154816, "300–1000 Hz_10ms_log_rms_mae_db": 1.8341847735703518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11097751993988983, "300–1000 Hz_10ms_envelope_correlation": 0.9911896846278127, "300–1000 Hz_10ms_modulation_mae_db": 0.3474811351577136, "300–1000 Hz_40ms_log_rms_mae_db": 1.2590858920973624, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07038120377189218, "300–1000 Hz_40ms_envelope_correlation": 0.9962239369360729, "300–1000 Hz_40ms_modulation_mae_db": 0.17949581415327187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7665096632594293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13243989607780546, "1000–3000 Hz_10ms_envelope_correlation": 0.9877326406684491, "1000–3000 Hz_10ms_modulation_mae_db": 0.33751595259430617, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0920954439004371, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08116078482464203, "1000–3000 Hz_40ms_envelope_correlation": 0.9950812496800407, "1000–3000 Hz_40ms_modulation_mae_db": 0.24975183173449586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.870004930028582, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1776606245339415, "3000–6000 Hz_10ms_envelope_correlation": 0.977938704112087, "3000–6000 Hz_10ms_modulation_mae_db": 0.42362654208083866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.210334204961108, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09922651325781076, "3000–6000 Hz_40ms_envelope_correlation": 0.9928465715047713, "3000–6000 Hz_40ms_modulation_mae_db": 0.49826095458859637, "mr_stft_db": 5.859114112202789, "band_mae_db": 0.09777101331509395, "rms_gain_error_db": -0.036372614927327775, "output_peak": 0.3074738660980778}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8304120484226238, "全帯域_10ms_rms_envelope_nrmse": 0.07102011510224397, "全帯域_10ms_envelope_correlation": 0.9954006541046077, "全帯域_10ms_modulation_mae_db": 0.3509579417900188, "全帯域_40ms_log_rms_mae_db": 0.4287581000823879, "全帯域_40ms_rms_envelope_nrmse": 0.037946441707576314, "全帯域_40ms_envelope_correlation": 0.9986388317651091, "全帯域_40ms_modulation_mae_db": 0.29354708170483285, "300–1000 Hz_10ms_log_rms_mae_db": 1.8311870329404927, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11425253372499872, "300–1000 Hz_10ms_envelope_correlation": 0.9907403949960485, "300–1000 Hz_10ms_modulation_mae_db": 0.549453188188545, "300–1000 Hz_40ms_log_rms_mae_db": 1.2244958480017798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07195241664418839, "300–1000 Hz_40ms_envelope_correlation": 0.9961738603042626, "300–1000 Hz_40ms_modulation_mae_db": 0.5784486241029111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8658365619854775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15854556584023916, "1000–3000 Hz_10ms_envelope_correlation": 0.9832729926520579, "1000–3000 Hz_10ms_modulation_mae_db": 0.40212692255760935, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1586847307070316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1006058743689581, "1000–3000 Hz_40ms_envelope_correlation": 0.9939471717531838, "1000–3000 Hz_40ms_modulation_mae_db": 0.5193250325424995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8822523036964176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17251409713242344, "3000–6000 Hz_10ms_envelope_correlation": 0.9800424126971461, "3000–6000 Hz_10ms_modulation_mae_db": 0.48988449128362593, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2300658187294318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11338209357634274, "3000–6000 Hz_40ms_envelope_correlation": 0.9910509852046954, "3000–6000 Hz_40ms_modulation_mae_db": 0.5967471375534243, "mr_stft_db": 5.856500025529894, "band_mae_db": 0.16398209380575746, "rms_gain_error_db": -0.09153030338588189, "output_peak": 0.2927911296571862}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8554365676236645, "全帯域_10ms_rms_envelope_nrmse": 0.08119724285912905, "全帯域_10ms_envelope_correlation": 0.9940351865305371, "全帯域_10ms_modulation_mae_db": 0.5854088836420779, "全帯域_40ms_log_rms_mae_db": 0.47560043455215095, "全帯域_40ms_rms_envelope_nrmse": 0.04901451575991056, "全帯域_40ms_envelope_correlation": 0.9977613508671079, "全帯域_40ms_modulation_mae_db": 0.9203201007898546, "300–1000 Hz_10ms_log_rms_mae_db": 1.873627935803931, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11829264790938968, "300–1000 Hz_10ms_envelope_correlation": 0.990173560793129, "300–1000 Hz_10ms_modulation_mae_db": 0.6081093048456265, "300–1000 Hz_40ms_log_rms_mae_db": 1.2506798909459966, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07332947062129723, "300–1000 Hz_40ms_envelope_correlation": 0.9961761381976714, "300–1000 Hz_40ms_modulation_mae_db": 1.200843663059482, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8616255518362193, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16430181044094994, "1000–3000 Hz_10ms_envelope_correlation": 0.9812109947503788, "1000–3000 Hz_10ms_modulation_mae_db": 0.34794378824845795, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1266074757147186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09289899823591977, "1000–3000 Hz_40ms_envelope_correlation": 0.9941175021934763, "1000–3000 Hz_40ms_modulation_mae_db": 0.4080481429341993, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8550221731501326, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17252090682760274, "3000–6000 Hz_10ms_envelope_correlation": 0.9792187059321766, "3000–6000 Hz_10ms_modulation_mae_db": 0.4982095913677634, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1631695201016614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09640425894278491, "3000–6000 Hz_40ms_envelope_correlation": 0.99339372091292, "3000–6000 Hz_40ms_modulation_mae_db": 0.7944419774838926, "mr_stft_db": 5.85914101078707, "band_mae_db": 0.20434005030255373, "rms_gain_error_db": -0.13041907110289586, "output_peak": 0.2752053804525345}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.669771433308448, "spectral_convergence": 0.2457576975778883, "spectral_convergence_status": "ok", "amplitude_floor": 7.181372005990482e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.544399691113733, "spectral_convergence": 0.24453113669013612, "spectral_convergence_status": "ok", "amplitude_floor": 5.417622086660404e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.377064536088314, "spectral_convergence": 0.24794020367439468, "spectral_convergence_status": "ok", "amplitude_floor": 3.166881446630363e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001077391554152542, "output_power": 0.00011393829255664332, "delta_db": 0.24296140250873416, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002614688899274717, "output_power": 0.0002502437302704098, "delta_db": -0.19056817822588573, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015854777413607497, "output_power": 0.00015162261586035384, "delta_db": -0.19396164305100128, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.449556222199797e-05, "output_power": 6.307864219309069e-05, "delta_db": -0.09647496649092316, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5950148027800597e-05, "output_power": 1.706428627422296e-05, "delta_db": 0.29323410373717307, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.2322545231279567e-05, "output_power": 1.2477443389081831e-05, "delta_db": 0.05425187166074734, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1785753609457441, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0855157752248928, "pitch": null, "output_rms_dbfs": -32.1579294169677, "output_peak": 0.2944582731796088, "reference_rms_dbfs": -32.072413641742806, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.863745220170164}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001113800968506657, "log_rms_mae_db": 0.9014545113645236, "rms_envelope_nrmse": 0.08061944540538701, "envelope_correlation": 0.9940347561419645, "rms_gain_error_db": -0.08551869772549706, "reference_mean_power": 0.0006210401213146882, "output_mean_power": 0.0006089305851870816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.9949527250696107e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014723003466329858, "output_power": 0.00014238876422047867, "reference_active": true, "delta_db": -0.14520693500798657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0749733198322775e-05, "output_power": 2.9858190775686797e-05, "reference_active": true, "delta_db": -0.12777863360289007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6784890109842875e-06, "output_power": 5.911931657254671e-06, "reference_active": true, "delta_db": 0.1749661523763374}], "active_band_count": 3, "active_band_mae_db": 0.14931724032907134}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.537149969198292e-05, "log_rms_mae_db": 0.498885231044676, "rms_envelope_nrmse": 0.046925673443768204, "envelope_correlation": 0.9978631359446367, "rms_gain_error_db": -0.0855073131581678, "reference_mean_power": 0.0006241534506116662, "output_mean_power": 0.0006119848125546344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.688739338888854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013414910111714758, "output_power": 0.0001295539098107302, "reference_active": true, "delta_db": -0.15137242869337317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.37620762285614e-05, "output_power": 1.3777144788689742e-05, "reference_active": true, "delta_db": 0.004752634915098151}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.149987894354722e-07, "output_power": 4.3237455616281457e-07, "reference_active": true, "delta_db": 0.17813299351366094}], "active_band_count": 3, "active_band_mae_db": 0.11141935237404409}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.440030149355929e-05, "log_rms_mae_db": 1.8175860748900303, "rms_envelope_nrmse": 0.11285923284560619, "envelope_correlation": 0.9911494182731452, "rms_gain_error_db": -0.22522713791089197, "reference_mean_power": 0.000315142987429675, "output_mean_power": 0.0002992160833448344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.94905977338963e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010605229421605894, "output_power": 0.00010217925416643975, "reference_active": true, "delta_db": -0.1615733966573567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.109806584151368e-05, "output_power": 1.9485537268747457e-05, "reference_active": true, "delta_db": -0.3453025835137008}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3612767263836796e-06, "output_power": 2.9966767599884795e-06, "reference_active": true, "delta_db": -0.4986436863882779}], "active_band_count": 3, "active_band_mae_db": 0.33517322218644513}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.828793840107377e-05, "log_rms_mae_db": 1.3263044038247092, "rms_envelope_nrmse": 0.07172425827774326, "envelope_correlation": 0.9964584820061426, "rms_gain_error_db": -0.22522705324607273, "reference_mean_power": 0.00031672629889435445, "output_mean_power": 0.00030071938223323526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7988521248624547e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.967318213226487e-05, "output_power": 9.535087229852515e-05, "reference_active": true, "delta_db": -0.19253653163811874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.491202888187961e-06, "output_power": 9.262441679918737e-06, "reference_active": true, "delta_db": -0.1059577066660368}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5805315595038687e-07, "output_power": 3.116147081606148e-07, "reference_active": true, "delta_db": -0.6032955792648018}], "active_band_count": 3, "active_band_mae_db": 0.30059660585631914}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.58934180861454e-05, "log_rms_mae_db": 1.8518672325853556, "rms_envelope_nrmse": 0.1571137278826271, "envelope_correlation": 0.9826330607940433, "rms_gain_error_db": -0.12977945794115747, "reference_mean_power": 6.36645119573909e-05, "output_mean_power": 6.179018114435589e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.512623526912829e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3991998554391248e-05, "output_power": 1.1663722123204422e-05, "reference_active": true, "delta_db": -0.7904258714570416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3957717792014314e-06, "output_power": 2.2756558300309194e-06, "reference_active": true, "delta_db": -0.2233886483789158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.685934249147775e-07, "output_power": 6.46920096162291e-07, "reference_active": true, "delta_db": -0.7484602245321155}], "active_band_count": 3, "active_band_mae_db": 0.5874249147893577}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6382034637347546e-05, "log_rms_mae_db": 1.1510367207335064, "rms_envelope_nrmse": 0.09700217540608558, "envelope_correlation": 0.9930292857555245, "rms_gain_error_db": -0.12973852861738214, "reference_mean_power": 6.398182095924481e-05, "output_mean_power": 6.20987335679421e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.275432891493173e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2825234350232933e-05, "output_power": 1.066213656872079e-05, "reference_active": true, "delta_db": -0.8022106854358757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.068139769619595e-06, "output_power": 1.0444561636911266e-06, "reference_active": true, "delta_db": -0.09737868121203407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.783047579212978e-08, "output_power": 5.156775077538919e-08, "reference_active": true, "delta_db": -0.4977857616752823}], "active_band_count": 3, "active_band_mae_db": 0.465791709441064}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.0860899025143115e-05, "log_rms_mae_db": 1.8941202918804405, "rms_envelope_nrmse": 0.2025318590338308, "envelope_correlation": 0.9732906489094019, "rms_gain_error_db": 0.2752829149582267, "reference_mean_power": 1.1507041647425964e-05, "output_mean_power": 1.2260042521098562e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.593308449162533e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.586447862335724e-06, "output_power": 5.2792154665699475e-06, "reference_active": true, "delta_db": 0.6109292679898277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.405946680666416e-07, "output_power": 1.4590652262975665e-06, "reference_active": true, "delta_db": 2.394880761741967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329768801508531e-07, "output_power": 5.271477450716925e-07, "reference_active": true, "delta_db": 1.9951827328079563}], "active_band_count": 3, "active_band_mae_db": 1.6669975875132501}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4734203651083813e-05, "log_rms_mae_db": 1.25971999369282, "rms_envelope_nrmse": 0.13038008224663633, "envelope_correlation": 0.9887791395897425, "rms_gain_error_db": 0.2756713756176905, "reference_mean_power": 1.156194575111937e-05, "output_mean_power": 1.2319641352183139e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.1073798497935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.414350374295127e-06, "output_power": 5.339414674529935e-06, "reference_active": true, "delta_db": 0.826268498794138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649548666103306e-07, "output_power": 8.141840544816831e-07, "reference_active": true, "delta_db": 2.4331179485230776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1590180837860948e-08, "output_power": 4.113859384504508e-08, "reference_active": true, "delta_db": 2.7999316305777855}], "active_band_count": 3, "active_band_mae_db": 2.019772692631667}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8933429328442459, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "28134492fcd820e5dfc8997c4e4381fa334230d6915b63649bc95923566ec6d0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9836979425107624, "standard_deviation": 0.002889448846857286, "minimum": 0.9793852625718602, "maximum": 0.9877326406684491, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8509905560293198, "standard_deviation": 0.038244399560479155, "minimum": 1.7665096632594293, "maximum": 1.888471443434016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6237659385168165, "standard_deviation": 0.419278819607345, "minimum": 0.192831826627498, "maximum": 1.3205850462206603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1532471799530819, "standard_deviation": 0.01421338430175957, "minimum": 0.13243989607780546, "maximum": 0.17485679016186842, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939264441797253, "standard_deviation": 0.000766270893042733, "minimum": 0.992935037803926, "maximum": 0.9950812496800407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1387167506181377, "standard_deviation": 0.02754347219627606, "minimum": 1.0920954439004371, "maximum": 1.1644556099764374, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.688241590148797, "standard_deviation": 0.526710009264741, "minimum": 0.24975183173449586, "maximum": 1.55415723765548, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09419576379831562, "standard_deviation": 0.008093209156202175, "minimum": 0.08116078482464203, "maximum": 0.10790052383075908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9771755997396512, "standard_deviation": 0.003132934493193021, "minimum": 0.9721427411772253, "maximum": 0.9810472825928539, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8659640505015138, "standard_deviation": 0.021999769472324567, "minimum": 1.8383876392527296, "maximum": 1.8941202918804405, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9331533643081389, "standard_deviation": 0.6059568155629492, "minimum": 0.42362654208083866, "maximum": 2.027215743100173, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.182309460038431, "standard_deviation": 0.013357450862007113, "minimum": 0.16629055810075613, "maximum": 0.2025318590338308, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922833254718362, "standard_deviation": 0.0018203582977112036, "minimum": 0.9887791395897425, "maximum": 0.9944042631239065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.21230732973511, "standard_deviation": 0.040626715461573176, "minimum": 1.1462992464255577, "maximum": 1.25971999369282, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0384911646524437, "standard_deviation": 0.6897722427067604, "minimum": 0.49201650881882114, "maximum": 2.186083690899373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10455919736268868, "standard_deviation": 0.012745076635773496, "minimum": 0.09180001997578614, "maximum": 0.13038008224663633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907628446191336, "standard_deviation": 0.0016970366573093867, "minimum": 0.9872286535534023, "maximum": 0.9930333972600089, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8380022452828042, "standard_deviation": 0.03676140328909088, "minimum": 1.7795806809800567, "maximum": 1.9026263369309542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48843989423707496, "standard_deviation": 0.16729201091963916, "minimum": 0.33517322218644513, "maximum": 0.8016850928090696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11437663064852088, "standard_deviation": 0.009374079047492283, "minimum": 0.10158001606447498, "maximum": 0.13345448311658717, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9963477827117468, "standard_deviation": 0.0009645443082438472, "minimum": 0.9942478928581708, "maximum": 0.9971845814826464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2679115646606582, "standard_deviation": 0.03869157814567816, "minimum": 1.2244958480017798, "maximum": 1.3263044038247092, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6206696598367203, "standard_deviation": 0.32898088292101885, "minimum": 0.17949581415327187, "maximum": 1.200843663059482, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07127615885661853, "standard_deviation": 0.007128353259491893, "minimum": 0.0649254154934318, "maximum": 0.08685795496398388, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17863439172955003, "standard_deviation": 0.05383050446191601, "minimum": 0.09777101331509395, "maximum": 0.27396361385713813, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.8569625354939365, "standard_deviation": 0.0066899874479822395, "minimum": 5.844151679977483, "maximum": 5.863745220170164, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.28504655578094845, "standard_deviation": 0.01252007176956922, "minimum": 0.27094478228217095, "maximum": 0.3074738660980778, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09425678555089786, "standard_deviation": 0.041543353836723425, "minimum": -0.16565894694742458, "maximum": -0.036372614927327775, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945668847660887, "standard_deviation": 0.0009103708658154606, "minimum": 0.9933254556840903, "maximum": 0.9958417229817774, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8653551523340133, "standard_deviation": 0.03037008706503672, "minimum": 0.8304120484226238, "maximum": 0.915323284246818, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3310986911932531, "standard_deviation": 0.17252498716484937, "minimum": 0.1171087790763801, "maximum": 0.5854088836420779, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07698068629654706, "standard_deviation": 0.0061311249305119105, "minimum": 0.0691660311965521, "maximum": 0.08489135649341817, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982405826828219, "standard_deviation": 0.0005112268942229514, "minimum": 0.9973993619094427, "maximum": 0.998784411484429, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46445309539686297, "standard_deviation": 0.0321839338746696, "minimum": 0.4287581000823879, "maximum": 0.5198234305532147, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3829630239028608, "standard_deviation": 0.3023476778683192, "minimum": 0.07787665555154816, "maximum": 0.9203201007898546, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04274750171119595, "standard_deviation": 0.0056768490108312815, "minimum": 0.03588156045471693, "maximum": 0.05070034087092733, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0013", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.7486329078674316, "predicted_mos": 2.7486329078674316, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ee4726390dea5bf2373e2111cc11db6fd51715742198e596fabc55a53d3703ca", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9423846998251975, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9053270590157035, "全帯域_10ms_rms_envelope_nrmse": 0.08258388460439403, "全帯域_10ms_envelope_correlation": 0.9937040242423295, "全帯域_10ms_modulation_mae_db": 0.15129324814690848, "全帯域_40ms_log_rms_mae_db": 0.4992789453323945, "全帯域_40ms_rms_envelope_nrmse": 0.046997434724099, "全帯域_40ms_envelope_correlation": 0.9978142078256388, "全帯域_40ms_modulation_mae_db": 0.11162176711405762, "300–1000 Hz_10ms_log_rms_mae_db": 1.821147181143219, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11995281483365167, "300–1000 Hz_10ms_envelope_correlation": 0.9897914962987464, "300–1000 Hz_10ms_modulation_mae_db": 0.3855509659674525, "300–1000 Hz_40ms_log_rms_mae_db": 1.2609900030769463, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07485546351225093, "300–1000 Hz_40ms_envelope_correlation": 0.9958980162232947, "300–1000 Hz_40ms_modulation_mae_db": 0.3877144287513526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9245127268493623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15563114290002336, "1000–3000 Hz_10ms_envelope_correlation": 0.9830013627569152, "1000–3000 Hz_10ms_modulation_mae_db": 0.5881657666511816, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2224420859517118, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0977219856073585, "1000–3000 Hz_40ms_envelope_correlation": 0.9930040434008404, "1000–3000 Hz_40ms_modulation_mae_db": 0.43346299857763815, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8611359400130685, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1915628843428381, "3000–6000 Hz_10ms_envelope_correlation": 0.975176094361993, "3000–6000 Hz_10ms_modulation_mae_db": 1.424897555517726, "3000–6000 Hz_40ms_log_rms_mae_db": 1.280720614570036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11827915275284112, "3000–6000 Hz_40ms_envelope_correlation": 0.9901296444785128, "3000–6000 Hz_40ms_modulation_mae_db": 1.6392357140060145, "mr_stft_db": 5.736659510000621, "band_mae_db": 0.13103127335324774, "rms_gain_error_db": -0.06443601591925649, "output_peak": 0.2934354706124147}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8675913000428285, "全帯域_10ms_rms_envelope_nrmse": 0.07480557484937976, "全帯域_10ms_envelope_correlation": 0.9948746693715371, "全帯域_10ms_modulation_mae_db": 0.41193915328866487, "全帯域_40ms_log_rms_mae_db": 0.4463941761845704, "全帯域_40ms_rms_envelope_nrmse": 0.036741778979456526, "全帯域_40ms_envelope_correlation": 0.9987238358536115, "全帯域_40ms_modulation_mae_db": 0.3856241291740756, "300–1000 Hz_10ms_log_rms_mae_db": 1.8255970090843985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10854453514251253, "300–1000 Hz_10ms_envelope_correlation": 0.9918018699436104, "300–1000 Hz_10ms_modulation_mae_db": 0.8083531981836645, "300–1000 Hz_40ms_log_rms_mae_db": 1.2347882804374901, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06415757535937304, "300–1000 Hz_40ms_envelope_correlation": 0.9971762058384209, "300–1000 Hz_40ms_modulation_mae_db": 1.0060155202950178, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9465616862115471, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13570544373945634, "1000–3000 Hz_10ms_envelope_correlation": 0.9873091698181805, "1000–3000 Hz_10ms_modulation_mae_db": 0.2607172967987146, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1536224061566975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08405114146164079, "1000–3000 Hz_40ms_envelope_correlation": 0.9951606078303356, "1000–3000 Hz_40ms_modulation_mae_db": 0.3516604880228374, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8044752677945912, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17699884503747504, "3000–6000 Hz_10ms_envelope_correlation": 0.9782293940266988, "3000–6000 Hz_10ms_modulation_mae_db": 0.611474237027051, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2089180832951467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09817774796327354, "3000–6000 Hz_40ms_envelope_correlation": 0.9929599821538961, "3000–6000 Hz_40ms_modulation_mae_db": 0.4385274956070933, "mr_stft_db": 5.747586445576431, "band_mae_db": 0.1828203859959271, "rms_gain_error_db": -0.08838589881299043, "output_peak": 0.29056417907071064}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8398598841452871, "全帯域_10ms_rms_envelope_nrmse": 0.06988839421875223, "全帯域_10ms_envelope_correlation": 0.9958162181039238, "全帯域_10ms_modulation_mae_db": 0.4468103214046124, "全帯域_40ms_log_rms_mae_db": 0.4407670133602104, "全帯域_40ms_rms_envelope_nrmse": 0.04163578211445227, "全帯域_40ms_envelope_correlation": 0.9986652694427868, "全帯域_40ms_modulation_mae_db": 0.4581181768380551, "300–1000 Hz_10ms_log_rms_mae_db": 1.796444339703919, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10090379384347306, "300–1000 Hz_10ms_envelope_correlation": 0.993110481041252, "300–1000 Hz_10ms_modulation_mae_db": 0.5606610438049219, "300–1000 Hz_40ms_log_rms_mae_db": 1.1982127457228495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06485141300556009, "300–1000 Hz_40ms_envelope_correlation": 0.9972794759486519, "300–1000 Hz_40ms_modulation_mae_db": 0.6839136413195991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9345618254909613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17192925896256445, "1000–3000 Hz_10ms_envelope_correlation": 0.9802776495564735, "1000–3000 Hz_10ms_modulation_mae_db": 1.1865675252336554, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2057576122524427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10390989850755787, "1000–3000 Hz_40ms_envelope_correlation": 0.9937922587655749, "1000–3000 Hz_40ms_modulation_mae_db": 1.481734989450204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.804076791481276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15832173460121265, "3000–6000 Hz_10ms_envelope_correlation": 0.9836993170891231, "3000–6000 Hz_10ms_modulation_mae_db": 1.2648264351484002, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2587318944870147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09572113913823037, "3000–6000 Hz_40ms_envelope_correlation": 0.9950955143429202, "3000–6000 Hz_40ms_modulation_mae_db": 1.4398248946519148, "mr_stft_db": 5.742774018070697, "band_mae_db": 0.3103473731809035, "rms_gain_error_db": -0.18528072407072008, "output_peak": 0.275726148344287}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.850362359675424, "全帯域_10ms_rms_envelope_nrmse": 0.08515758394408716, "全帯域_10ms_envelope_correlation": 0.9933671441774071, "全帯域_10ms_modulation_mae_db": 0.3758845475324792, "全帯域_40ms_log_rms_mae_db": 0.46141860479741986, "全帯域_40ms_rms_envelope_nrmse": 0.047592382681123375, "全帯域_40ms_envelope_correlation": 0.9978433463128379, "全帯域_40ms_modulation_mae_db": 0.6557601067625957, "300–1000 Hz_10ms_log_rms_mae_db": 1.8545061675048076, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12121030383829494, "300–1000 Hz_10ms_envelope_correlation": 0.9896513699560072, "300–1000 Hz_10ms_modulation_mae_db": 0.37502612685174636, "300–1000 Hz_40ms_log_rms_mae_db": 1.2547278069741772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06788653703967595, "300–1000 Hz_40ms_envelope_correlation": 0.9967904540838919, "300–1000 Hz_40ms_modulation_mae_db": 0.6550279798188132, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8950111441027935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1431744220021601, "1000–3000 Hz_10ms_envelope_correlation": 0.9860929345348262, "1000–3000 Hz_10ms_modulation_mae_db": 1.4200805542105404, "1000–3000 Hz_40ms_log_rms_mae_db": 1.200645782759867, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09165008843223671, "1000–3000 Hz_40ms_envelope_correlation": 0.9945169884969683, "1000–3000 Hz_40ms_modulation_mae_db": 1.6565446723219672, "3000–6000 Hz_10ms_log_rms_mae_db": 1.760800250411928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1701289405526388, "3000–6000 Hz_10ms_envelope_correlation": 0.9799884625797765, "3000–6000 Hz_10ms_modulation_mae_db": 1.793451615888601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1784564295880033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09283126961824847, "3000–6000 Hz_40ms_envelope_correlation": 0.994257652946683, "3000–6000 Hz_40ms_modulation_mae_db": 2.075531464379738, "mr_stft_db": 5.7289010156389395, "band_mae_db": 0.2547268958133281, "rms_gain_error_db": -0.11411315542691473, "output_peak": 0.27840896087406153}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9010804088826081, "全帯域_10ms_rms_envelope_nrmse": 0.08664012961803078, "全帯域_10ms_envelope_correlation": 0.9930462288134461, "全帯域_10ms_modulation_mae_db": 0.05157868536783614, "全帯域_40ms_log_rms_mae_db": 0.5093672622508739, "全帯域_40ms_rms_envelope_nrmse": 0.05445028219911139, "全帯域_40ms_envelope_correlation": 0.9970008393562411, "全帯域_40ms_modulation_mae_db": 0.08082974294411953, "300–1000 Hz_10ms_log_rms_mae_db": 1.9038171698760982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13097712366202827, "300–1000 Hz_10ms_envelope_correlation": 0.9877083387158945, "300–1000 Hz_10ms_modulation_mae_db": 0.40834621794206877, "300–1000 Hz_40ms_log_rms_mae_db": 1.3004277750851059, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08739512177345492, "300–1000 Hz_40ms_envelope_correlation": 0.994181733452283, "300–1000 Hz_40ms_modulation_mae_db": 0.5622489065765334, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9018319590023354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1571796532325081, "1000–3000 Hz_10ms_envelope_correlation": 0.9826167649740036, "1000–3000 Hz_10ms_modulation_mae_db": 0.6107461722559987, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1882000546401268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.093164872200718, "1000–3000 Hz_40ms_envelope_correlation": 0.9935704582071402, "1000–3000 Hz_40ms_modulation_mae_db": 0.5241175464474565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.850568640402479, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20059659042415268, "3000–6000 Hz_10ms_envelope_correlation": 0.9721958802833688, "3000–6000 Hz_10ms_modulation_mae_db": 0.6428599471988646, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2478739050372567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11147418060772779, "3000–6000 Hz_40ms_envelope_correlation": 0.9909582193172037, "3000–6000 Hz_40ms_modulation_mae_db": 0.5760764642269882, "mr_stft_db": 5.717950911955664, "band_mae_db": 0.155862729370952, "rms_gain_error_db": -0.05754323771568151, "output_peak": 0.26974623395944775}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8442159805646332, "全帯域_10ms_rms_envelope_nrmse": 0.07130605827589764, "全帯域_10ms_envelope_correlation": 0.995290344582681, "全帯域_10ms_modulation_mae_db": 0.11308357333810783, "全帯域_40ms_log_rms_mae_db": 0.4459492511357849, "全帯域_40ms_rms_envelope_nrmse": 0.03524965362437273, "全帯域_40ms_envelope_correlation": 0.998742003534808, "全帯域_40ms_modulation_mae_db": 0.13229925617541904, "300–1000 Hz_10ms_log_rms_mae_db": 1.8521838585064565, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11145807059445481, "300–1000 Hz_10ms_envelope_correlation": 0.9911078042636547, "300–1000 Hz_10ms_modulation_mae_db": 0.33547519857322267, "300–1000 Hz_40ms_log_rms_mae_db": 1.2032436884875006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06613374301378395, "300–1000 Hz_40ms_envelope_correlation": 0.9966566527029456, "300–1000 Hz_40ms_modulation_mae_db": 0.1960403490594029, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8196770470267878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13183460017950405, "1000–3000 Hz_10ms_envelope_correlation": 0.9878204536005004, "1000–3000 Hz_10ms_modulation_mae_db": 0.34488594595797323, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1623566203841447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08186248325261075, "1000–3000 Hz_40ms_envelope_correlation": 0.9950088177630665, "1000–3000 Hz_40ms_modulation_mae_db": 0.30229332068915366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8611062287186557, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17206322766796064, "3000–6000 Hz_10ms_envelope_correlation": 0.9793805652921616, "3000–6000 Hz_10ms_modulation_mae_db": 0.5131551875914969, "3000–6000 Hz_40ms_log_rms_mae_db": 1.248874429387954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09604404293196957, "3000–6000 Hz_40ms_envelope_correlation": 0.9935166518655026, "3000–6000 Hz_40ms_modulation_mae_db": 0.5526151668286989, "mr_stft_db": 5.737290897173786, "band_mae_db": 0.11220005017978041, "rms_gain_error_db": -0.03195483468766554, "output_peak": 0.3068530502978564}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.840922524514827, "全帯域_10ms_rms_envelope_nrmse": 0.07188692023468217, "全帯域_10ms_envelope_correlation": 0.9952868401295217, "全帯域_10ms_modulation_mae_db": 0.38494407636012323, "全帯域_40ms_log_rms_mae_db": 0.43882918869015314, "全帯域_40ms_rms_envelope_nrmse": 0.04043746704710063, "全帯域_40ms_envelope_correlation": 0.9984384153238376, "全帯域_40ms_modulation_mae_db": 0.3333405031973867, "300–1000 Hz_10ms_log_rms_mae_db": 1.8656143746495606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10982061115744075, "300–1000 Hz_10ms_envelope_correlation": 0.9914536862767975, "300–1000 Hz_10ms_modulation_mae_db": 0.6090960656954564, "300–1000 Hz_40ms_log_rms_mae_db": 1.1822387581784872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06418155001792268, "300–1000 Hz_40ms_envelope_correlation": 0.9969784836506148, "300–1000 Hz_40ms_modulation_mae_db": 0.6690447951637285, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9405565054553642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1541541076601526, "1000–3000 Hz_10ms_envelope_correlation": 0.9843549150908313, "1000–3000 Hz_10ms_modulation_mae_db": 0.531817459873223, "1000–3000 Hz_40ms_log_rms_mae_db": 1.230971422216715, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.099256388552227, "1000–3000 Hz_40ms_envelope_correlation": 0.9942394690118074, "1000–3000 Hz_40ms_modulation_mae_db": 0.6692460238647367, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8443104039135438, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16124131219220958, "3000–6000 Hz_10ms_envelope_correlation": 0.9821027032969001, "3000–6000 Hz_10ms_modulation_mae_db": 0.465339424856323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.264971683035855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10280320884707281, "3000–6000 Hz_40ms_envelope_correlation": 0.9923319410832976, "3000–6000 Hz_40ms_modulation_mae_db": 0.6635418231632806, "mr_stft_db": 5.744654667368884, "band_mae_db": 0.15087050231200738, "rms_gain_error_db": -0.09308827768478571, "output_peak": 0.28941182296720563}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8565656936737579, "全帯域_10ms_rms_envelope_nrmse": 0.0840713292426383, "全帯域_10ms_envelope_correlation": 0.9935588136727311, "全帯域_10ms_modulation_mae_db": 0.5654772522338439, "全帯域_40ms_log_rms_mae_db": 0.4778264174700561, "全帯域_40ms_rms_envelope_nrmse": 0.051790163824718204, "全帯域_40ms_envelope_correlation": 0.9974380025574856, "全帯域_40ms_modulation_mae_db": 0.8597639999088654, "300–1000 Hz_10ms_log_rms_mae_db": 1.8883948099393555, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12324834823407235, "300–1000 Hz_10ms_envelope_correlation": 0.9892120371021318, "300–1000 Hz_10ms_modulation_mae_db": 0.5706343232151591, "300–1000 Hz_40ms_log_rms_mae_db": 1.2406067735829094, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07638414843656467, "300–1000 Hz_40ms_envelope_correlation": 0.995692005036005, "300–1000 Hz_40ms_modulation_mae_db": 1.1542909058167272, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9250568576486877, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16553371436669465, "1000–3000 Hz_10ms_envelope_correlation": 0.9810018971365752, "1000–3000 Hz_10ms_modulation_mae_db": 0.46532932866484283, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1858761834544558, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09320659604719057, "1000–3000 Hz_40ms_envelope_correlation": 0.9942107005649435, "1000–3000 Hz_40ms_modulation_mae_db": 0.49663990547028963, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8304672669904505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1677042734585571, "3000–6000 Hz_10ms_envelope_correlation": 0.9805413290166411, "3000–6000 Hz_10ms_modulation_mae_db": 0.619708015039121, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2075743144449793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09949393785883966, "3000–6000 Hz_40ms_envelope_correlation": 0.9932289249881967, "3000–6000 Hz_40ms_modulation_mae_db": 1.0158427580113656, "mr_stft_db": 5.739210930965453, "band_mae_db": 0.22621450842728139, "rms_gain_error_db": -0.12127558610048039, "output_peak": 0.2746462163730615}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.562704821270231, "spectral_convergence": 0.24674388803084216, "spectral_convergence_status": "ok", "amplitude_floor": 7.181372005990482e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.416465354626969, "spectral_convergence": 0.24542951820202122, "spectral_convergence_status": "ok", "amplitude_floor": 5.417622086660404e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.230808354104665, "spectral_convergence": 0.2457971497066368, "spectral_convergence_status": "ok", "amplitude_floor": 3.166881446630363e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001077391554152542, "output_power": 0.00011377735763779463, "delta_db": 0.23682276787571593, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002614688899274717, "output_power": 0.0002528962180243252, "delta_db": -0.14477688413688344, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015854777413607497, "output_power": 0.00015324503932694279, "delta_db": -0.14773724379581735, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.449556222199797e-05, "output_power": 6.28159252818356e-05, "delta_db": -0.11460071611932689, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5950148027800597e-05, "output_power": 1.639371135840882e-05, "delta_db": 0.1191256631219591, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.2322545231279567e-05, "output_power": 1.2257107200526767e-05, "delta_db": -0.023124365069783763, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13103127335324774, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06443601591925649, "pitch": null, "output_rms_dbfs": -32.13684965766206, "output_peak": 0.2934354706124147, "reference_rms_dbfs": -32.072413641742806, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.736659510000621}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001113800968506657, "log_rms_mae_db": 0.9053270590157035, "rms_envelope_nrmse": 0.08258388460439403, "envelope_correlation": 0.9937040242423295, "rms_gain_error_db": -0.06444182334619428, "reference_mean_power": 0.0006210401213146882, "output_mean_power": 0.0006118929869257068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.9949527250696107e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014723003466329858, "output_power": 0.000142925120957383, "reference_active": true, "delta_db": -0.1288784590615048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0749733198322775e-05, "output_power": 2.9559767918053544e-05, "reference_active": true, "delta_db": -0.17140331979275325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6784890109842875e-06, "output_power": 5.882915172786319e-06, "reference_active": true, "delta_db": 0.1535979655864674}], "active_band_count": 3, "active_band_mae_db": 0.15129324814690848}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.537149969198292e-05, "log_rms_mae_db": 0.4992789453323945, "rms_envelope_nrmse": 0.046997434724099, "envelope_correlation": 0.9978142078256388, "rms_gain_error_db": -0.06442988256617405, "reference_mean_power": 0.0006241534506116662, "output_mean_power": 0.0006149621516407561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.688739338888854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013414910111714758, "output_power": 0.00013002293558775875, "reference_active": true, "delta_db": -0.13567799892257035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.37620762285614e-05, "output_power": 1.3650568971134297e-05, "reference_active": true, "delta_db": -0.035332054488418356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.149987894354722e-07, "output_power": 4.3095542752173145e-07, "reference_active": true, "delta_db": 0.1638552479311841}], "active_band_count": 3, "active_band_mae_db": 0.11162176711405762}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.440030149355929e-05, "log_rms_mae_db": 1.821147181143219, "rms_envelope_nrmse": 0.11995281483365167, "envelope_correlation": 0.9897914962987464, "rms_gain_error_db": -0.16922374624198003, "reference_mean_power": 0.000315142987429675, "output_mean_power": 0.000303099537035716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.94905977338963e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010605229421605894, "output_power": 0.00010330484269649298, "reference_active": true, "delta_db": -0.11399387256429701}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.109806584151368e-05, "output_power": 1.9349472135372076e-05, "reference_active": true, "delta_db": -0.37573521598276405}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3612767263836796e-06, "output_power": 2.882782569679661e-06, "reference_active": true, "delta_db": -0.6669238093552963}], "active_band_count": 3, "active_band_mae_db": 0.3855509659674525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.828793840107377e-05, "log_rms_mae_db": 1.2609900030769463, "rms_envelope_nrmse": 0.07485546351225093, "envelope_correlation": 0.9958980162232947, "rms_gain_error_db": -0.16921992198291866, "reference_mean_power": 0.00031672629889435445, "output_mean_power": 0.0003046226091816932}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7988521248624547e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.967318213226487e-05, "output_power": 9.632507428482614e-05, "reference_active": true, "delta_db": -0.14838971012613958}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.491202888187961e-06, "output_power": 9.227523644429784e-06, "reference_active": true, "delta_db": -0.12236090471803224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5805315595038687e-07, "output_power": 2.915467730812953e-07, "reference_active": true, "delta_db": -0.892392671409886}], "active_band_count": 3, "active_band_mae_db": 0.3877144287513526}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.58934180861454e-05, "log_rms_mae_db": 1.9245127268493623, "rms_envelope_nrmse": 0.15563114290002336, "envelope_correlation": 0.9830013627569152, "rms_gain_error_db": -0.16431965274627547, "reference_mean_power": 6.36645119573909e-05, "output_mean_power": 6.130070213253557e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.512623526912829e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3991998554391248e-05, "output_power": 1.1463301991983394e-05, "reference_active": true, "delta_db": -0.8657001788368744}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3957717792014314e-06, "output_power": 2.2052902969715356e-06, "reference_active": true, "delta_db": -0.359796782448764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.685934249147775e-07, "output_power": 6.788856534110022e-07, "reference_active": true, "delta_db": -0.5390003386679063}], "active_band_count": 3, "active_band_mae_db": 0.5881657666511816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6382034637347546e-05, "log_rms_mae_db": 1.2224420859517118, "rms_envelope_nrmse": 0.0977219856073585, "envelope_correlation": 0.9930040434008404, "rms_gain_error_db": -0.16428534430975555, "reference_mean_power": 6.398182095924481e-05, "output_mean_power": 6.160671639721746e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.275432891493173e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2825234350232933e-05, "output_power": 1.0470057313408236e-05, "reference_active": true, "delta_db": -0.8811625051443417}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.068139769619595e-06, "output_power": 1.0207059058003677e-06, "reference_active": true, "delta_db": -0.19727457675125631}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.783047579212978e-08, "output_power": 5.494922531474365e-08, "reference_active": true, "delta_db": -0.22195191383731658}], "active_band_count": 3, "active_band_mae_db": 0.43346299857763815}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.0860899025143115e-05, "log_rms_mae_db": 1.8611359400130685, "rms_envelope_nrmse": 0.1915628843428381, "envelope_correlation": 0.975176094361993, "rms_gain_error_db": 0.09256442671223758, "reference_mean_power": 1.1507041647425964e-05, "output_mean_power": 1.1754932174596754e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.593308449162533e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.586447862335724e-06, "output_power": 5.065482180555486e-06, "reference_active": true, "delta_db": 0.43144331011022785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.405946680666416e-07, "output_power": 1.3395854086706983e-06, "reference_active": true, "delta_db": 2.023837774480345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329768801508531e-07, "output_power": 5.06238578540455e-07, "reference_active": true, "delta_db": 1.8194115819626047}], "active_band_count": 3, "active_band_mae_db": 1.424897555517726}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4734203651083813e-05, "log_rms_mae_db": 1.280720614570036, "rms_envelope_nrmse": 0.11827915275284112, "envelope_correlation": 0.9901296444785128, "rms_gain_error_db": 0.09289508303210839, "reference_mean_power": 1.156194575111937e-05, "output_mean_power": 1.1811918333411742e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.1073798497935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.414350374295127e-06, "output_power": 5.105600295345236e-06, "reference_active": true, "delta_db": 0.6318001110669441}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649548666103306e-07, "output_power": 7.302756103084585e-07, "reference_active": true, "delta_db": 1.9607599858387932}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1590180837860948e-08, "output_power": 3.6878307774713635e-08, "reference_active": true, "delta_db": 2.325147045112306}], "active_band_count": 3, "active_band_mae_db": 1.6392357140060145}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8925354675113965, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3e611cf63876995530f7a3ad33231c450ae54d7e6e7f5ca5befaf1222d846a55", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9840593934335382, "standard_deviation": 0.002822858701307678, "minimum": 0.9802776495564735, "maximum": 0.9878204536005004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9109712189734798, "standard_deviation": 0.04097584793313024, "minimum": 1.8196770470267878, "maximum": 1.9465616862115471, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6760387562057663, "standard_deviation": 0.40945520399408836, "minimum": 0.2607172967987146, "maximum": 1.4200805542105404, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15189279288038296, "standard_deviation": 0.01401502136612023, "minimum": 0.13183460017950405, "maximum": 0.17192925896256445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941879180050845, "standard_deviation": 0.0007241574180895979, "minimum": 0.9930040434008404, "maximum": 0.9951606078303356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1937340209770202, "standard_deviation": 0.026962856856866155, "minimum": 1.1536224061566975, "maximum": 1.230971422216715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7394624931055354, "standard_deviation": 0.5260931152025291, "minimum": 0.30229332068915366, "maximum": 1.6565446723219672, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09310293175769252, "standard_deviation": 0.0074296620391492135, "minimum": 0.08186248325261075, "maximum": 0.10390989850755787, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9789142182433328, "standard_deviation": 0.0037160116318820384, "minimum": 0.9721958802833688, "maximum": 0.9836993170891231, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.827117598715749, "standard_deviation": 0.0350729511814155, "minimum": 1.760800250411928, "maximum": 1.8611359400130685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9169640522834479, "standard_deviation": 0.5030023818326348, "minimum": 0.465339424856323, "maximum": 1.793451615888601, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17482722603463058, "standard_deviation": 0.014571580257361636, "minimum": 0.15832173460121265, "maximum": 0.20059659042415268, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9928098163970267, "standard_deviation": 0.0016413653721337243, "minimum": 0.9901296444785128, "maximum": 0.9950955143429202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2370151692307807, "standard_deviation": 0.03486143822175502, "minimum": 1.1784564295880033, "maximum": 1.280720614570036, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.050149472609387, "standard_deviation": 0.6032397336913549, "minimum": 0.4385274956070933, "maximum": 2.075531464379738, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10185308496477541, "standard_deviation": 0.008743131818650978, "minimum": 0.09283126961824847, "maximum": 0.11827915275284112, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904796354497618, "standard_deviation": 0.0017096145778151231, "minimum": 0.9877083387158945, "maximum": 0.993110481041252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8509631138009768, "standard_deviation": 0.035713366542946226, "minimum": 1.796444339703919, "maximum": 1.9038171698760982, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5066428925292116, "standard_deviation": 0.16012739056913086, "minimum": 0.33547519857322267, "maximum": 0.8083531981836645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11576445016324105, "standard_deviation": 0.009720922152434712, "minimum": 0.10090379384347306, "maximum": 0.13097712366202827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9963316283670135, "standard_deviation": 0.0010396081266997792, "minimum": 0.994181733452283, "maximum": 0.9972794759486519, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2344044789431834, "standard_deviation": 0.0387659976005599, "minimum": 1.1822387581784872, "maximum": 1.3004277750851059, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6642870658501469, "standard_deviation": 0.3078809756595023, "minimum": 0.1960403490594029, "maximum": 1.1542909058167272, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07073069401982328, "standard_deviation": 0.00825108418304749, "minimum": 0.06415757535937304, "maximum": 0.08739512177345492, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19050921482917846, "standard_deviation": 0.0678871312261291, "minimum": 0.11220005017978041, "maximum": 0.3103473731809035, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7368785495938095, "standard_deviation": 0.0095426719744478, "minimum": 5.717950911955664, "maximum": 5.747586445576431, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.28484901031238063, "standard_deviation": 0.012353595164494193, "minimum": 0.26974623395944775, "maximum": 0.3068530502978564, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09450971630231186, "standard_deviation": 0.04715273859474557, "minimum": -0.18528072407072008, "maximum": -0.03195483468766554, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943680353866972, "standard_deviation": 0.0010619035467111819, "minimum": 0.9930462288134461, "maximum": 0.9958162181039238, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8632406513143837, "standard_deviation": 0.02629270327844179, "minimum": 0.8398598841452871, "maximum": 0.9053270590157035, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31262635720907206, "standard_deviation": 0.18323481251497087, "minimum": 0.05157868536783614, "maximum": 0.5654772522338439, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07829248437348275, "standard_deviation": 0.006982146703436803, "minimum": 0.06988839421875223, "maximum": 0.08664012961803078, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980832400259059, "standard_deviation": 0.0006576027924216004, "minimum": 0.9970008393562411, "maximum": 0.998742003534808, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46497885740268285, "standard_deviation": 0.0275216708681281, "minimum": 0.43882918869015314, "maximum": 0.5093672622508739, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37716971026432183, "standard_deviation": 0.2770635857598715, "minimum": 0.08082974294411953, "maximum": 0.8597639999088654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04436186814930427, "standard_deviation": 0.006951554724123488, "minimum": 0.03524965362437273, "maximum": 0.05445028219911139, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0013", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.8373324871063232, "predicted_mos": 2.8373324871063232, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "671e16adc994b779d48e0d1d9b4830208f59a2b8abdd5d41a17970cda0ecd800", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9943911996670067, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8633912221922009, "全帯域_10ms_rms_envelope_nrmse": 0.0780677452457456, "全帯域_10ms_envelope_correlation": 0.9944533928896357, "全帯域_10ms_modulation_mae_db": 0.309090455104751, "全帯域_40ms_log_rms_mae_db": 0.49151867907826086, "全帯域_40ms_rms_envelope_nrmse": 0.04260414148851901, "全帯域_40ms_envelope_correlation": 0.9982978950883751, "全帯域_40ms_modulation_mae_db": 0.2575250044679351, "300–1000 Hz_10ms_log_rms_mae_db": 2.124482060120214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11933959418606231, "300–1000 Hz_10ms_envelope_correlation": 0.9901767851981146, "300–1000 Hz_10ms_modulation_mae_db": 0.8033462298501268, "300–1000 Hz_40ms_log_rms_mae_db": 1.5143945710664464, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07651684021358697, "300–1000 Hz_40ms_envelope_correlation": 0.996011675409836, "300–1000 Hz_40ms_modulation_mae_db": 1.0499877038051209, "1000–3000 Hz_10ms_log_rms_mae_db": 2.173851682270572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16401052893128612, "1000–3000 Hz_10ms_envelope_correlation": 0.9814719300222412, "1000–3000 Hz_10ms_modulation_mae_db": 0.5377132184014649, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3152754205490849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10373618234695647, "1000–3000 Hz_40ms_envelope_correlation": 0.9927115782431692, "1000–3000 Hz_40ms_modulation_mae_db": 0.44879256795635847, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3894301327917047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21730478095435923, "3000–6000 Hz_10ms_envelope_correlation": 0.9675197397748442, "3000–6000 Hz_10ms_modulation_mae_db": 0.5611984000407256, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5435637686100119, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12203287717320353, "3000–6000 Hz_40ms_envelope_correlation": 0.9892990790501431, "3000–6000 Hz_40ms_modulation_mae_db": 0.6365854544242914, "mr_stft_db": 6.267408439080662, "band_mae_db": 0.21307556519870874, "rms_gain_error_db": -0.10602326155742173, "output_peak": 0.28202953062181446}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8561223344079804, "全帯域_10ms_rms_envelope_nrmse": 0.07169103506551863, "全帯域_10ms_envelope_correlation": 0.995313519344873, "全帯域_10ms_modulation_mae_db": 0.2621319439858601, "全帯域_40ms_log_rms_mae_db": 0.4485231599912975, "全帯域_40ms_rms_envelope_nrmse": 0.03916561502727853, "全帯域_40ms_envelope_correlation": 0.9985559003361786, "全帯域_40ms_modulation_mae_db": 0.23669348519734937, "300–1000 Hz_10ms_log_rms_mae_db": 2.0412518406608364, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11265338536641917, "300–1000 Hz_10ms_envelope_correlation": 0.991163351586459, "300–1000 Hz_10ms_modulation_mae_db": 0.5360708493658174, "300–1000 Hz_40ms_log_rms_mae_db": 1.4942768701810425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0671411661407201, "300–1000 Hz_40ms_envelope_correlation": 0.9969237709287468, "300–1000 Hz_40ms_modulation_mae_db": 0.7024364585947369, "1000–3000 Hz_10ms_log_rms_mae_db": 2.131009052826876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15528439420307835, "1000–3000 Hz_10ms_envelope_correlation": 0.9831117998801872, "1000–3000 Hz_10ms_modulation_mae_db": 0.30306545668483853, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3109004170061203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09784742029345747, "1000–3000 Hz_40ms_envelope_correlation": 0.9928607646350087, "1000–3000 Hz_40ms_modulation_mae_db": 0.5435821865337674, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3507580797054217, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21472998891904205, "3000–6000 Hz_10ms_envelope_correlation": 0.9687119466952205, "3000–6000 Hz_10ms_modulation_mae_db": 0.5939862027402021, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5250326281704576, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12517344708926287, "3000–6000 Hz_40ms_envelope_correlation": 0.9889946723489723, "3000–6000 Hz_40ms_modulation_mae_db": 0.7987590422180558, "mr_stft_db": 6.25864105206695, "band_mae_db": 0.12587734486758298, "rms_gain_error_db": -0.10084819581982116, "output_peak": 0.3008129130835902}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8243369960268745, "全帯域_10ms_rms_envelope_nrmse": 0.07796380165284132, "全帯域_10ms_envelope_correlation": 0.9943700198327062, "全帯域_10ms_modulation_mae_db": 0.4075113330894878, "全帯域_40ms_log_rms_mae_db": 0.4433109694301962, "全帯域_40ms_rms_envelope_nrmse": 0.0364438349389658, "全帯域_40ms_envelope_correlation": 0.9986626754187207, "全帯域_40ms_modulation_mae_db": 0.501213936597547, "300–1000 Hz_10ms_log_rms_mae_db": 2.0695371459559855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11821595007274517, "300–1000 Hz_10ms_envelope_correlation": 0.9901042100569875, "300–1000 Hz_10ms_modulation_mae_db": 0.7719601450258802, "300–1000 Hz_40ms_log_rms_mae_db": 1.5129449914597837, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06635434568524998, "300–1000 Hz_40ms_envelope_correlation": 0.9968247452076243, "300–1000 Hz_40ms_modulation_mae_db": 0.8663776358647565, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0972695471620826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16994938697351794, "1000–3000 Hz_10ms_envelope_correlation": 0.9800526815065725, "1000–3000 Hz_10ms_modulation_mae_db": 0.9786762080185724, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2526566840196602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10923179906326796, "1000–3000 Hz_40ms_envelope_correlation": 0.9912726937956811, "1000–3000 Hz_40ms_modulation_mae_db": 1.294034194605773, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3690632470352924, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21047139246620378, "3000–6000 Hz_10ms_envelope_correlation": 0.9692189363430669, "3000–6000 Hz_10ms_modulation_mae_db": 0.9703619211627381, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5691632929948924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11527520724913269, "3000–6000 Hz_40ms_envelope_correlation": 0.9904757511596849, "3000–6000 Hz_40ms_modulation_mae_db": 1.5204699391853425, "mr_stft_db": 6.255152932861446, "band_mae_db": 0.19983700261591916, "rms_gain_error_db": -0.04067553151237897, "output_peak": 0.29320899613488144}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8542267091189777, "全帯域_10ms_rms_envelope_nrmse": 0.07878772224595708, "全帯域_10ms_envelope_correlation": 0.9942619577480407, "全帯域_10ms_modulation_mae_db": 0.2047460189314878, "全帯域_40ms_log_rms_mae_db": 0.4817409208779267, "全帯域_40ms_rms_envelope_nrmse": 0.04658229292304344, "全帯域_40ms_envelope_correlation": 0.9977897952607546, "全帯域_40ms_modulation_mae_db": 0.16014844433976838, "300–1000 Hz_10ms_log_rms_mae_db": 2.089234454124318, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09830764903426421, "300–1000 Hz_10ms_envelope_correlation": 0.9931260048426742, "300–1000 Hz_10ms_modulation_mae_db": 0.3608191196729447, "300–1000 Hz_40ms_log_rms_mae_db": 1.4834525856635858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055308639279410536, "300–1000 Hz_40ms_envelope_correlation": 0.9976976024270687, "300–1000 Hz_40ms_modulation_mae_db": 0.5229433379261965, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1820892768446214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16153539794998104, "1000–3000 Hz_10ms_envelope_correlation": 0.9817147520769365, "1000–3000 Hz_10ms_modulation_mae_db": 0.6206340938129585, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2874674307738025, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09809445827438701, "1000–3000 Hz_40ms_envelope_correlation": 0.9928306049173538, "1000–3000 Hz_40ms_modulation_mae_db": 0.9873060492755265, "3000–6000 Hz_10ms_log_rms_mae_db": 2.405623604363381, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23898477441891025, "3000–6000 Hz_10ms_envelope_correlation": 0.96050300834064, "3000–6000 Hz_10ms_modulation_mae_db": 0.5140955816945674, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5700416808377855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13676511601582336, "3000–6000 Hz_40ms_envelope_correlation": 0.9865185597015071, "3000–6000 Hz_40ms_modulation_mae_db": 0.4665227068558158, "mr_stft_db": 6.255410633937818, "band_mae_db": 0.13267139852116608, "rms_gain_error_db": 0.005515550967584604, "output_peak": 0.28065789510570965}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8682516013273399, "全帯域_10ms_rms_envelope_nrmse": 0.07556725048051346, "全帯域_10ms_envelope_correlation": 0.9948061884211198, "全帯域_10ms_modulation_mae_db": 0.38801238475414374, "全帯域_40ms_log_rms_mae_db": 0.4361550318048866, "全帯域_40ms_rms_envelope_nrmse": 0.039942792741946995, "全帯域_40ms_envelope_correlation": 0.9985180147177007, "全帯域_40ms_modulation_mae_db": 0.3417093330720337, "300–1000 Hz_10ms_log_rms_mae_db": 2.1395278996788667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11271817845359694, "300–1000 Hz_10ms_envelope_correlation": 0.9910985349941022, "300–1000 Hz_10ms_modulation_mae_db": 0.8095800319298448, "300–1000 Hz_40ms_log_rms_mae_db": 1.46381047393495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0683718492475925, "300–1000 Hz_40ms_envelope_correlation": 0.9967069307701135, "300–1000 Hz_40ms_modulation_mae_db": 1.0659786968285676, "1000–3000 Hz_10ms_log_rms_mae_db": 2.132250445189975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18054790832631332, "1000–3000 Hz_10ms_envelope_correlation": 0.9773525385829568, "1000–3000 Hz_10ms_modulation_mae_db": 1.1597408569185743, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3130488391043393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12231951692425927, "1000–3000 Hz_40ms_envelope_correlation": 0.9894922393609052, "1000–3000 Hz_40ms_modulation_mae_db": 1.481227122789327, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2725893709759433, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22084347091264228, "3000–6000 Hz_10ms_envelope_correlation": 0.9678063020757193, "3000–6000 Hz_10ms_modulation_mae_db": 0.7467003865364612, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5174784467416924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12486601874481476, "3000–6000 Hz_40ms_envelope_correlation": 0.989670783963861, "3000–6000 Hz_40ms_modulation_mae_db": 0.4218779004824172, "mr_stft_db": 6.255445985668885, "band_mae_db": 0.19532421897737265, "rms_gain_error_db": -0.1073282260419329, "output_peak": 0.2910641928688889}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8821699333222669, "全帯域_10ms_rms_envelope_nrmse": 0.07643881010970296, "全帯域_10ms_envelope_correlation": 0.9946682076056155, "全帯域_10ms_modulation_mae_db": 0.4313352962862988, "全帯域_40ms_log_rms_mae_db": 0.47724017994306156, "全帯域_40ms_rms_envelope_nrmse": 0.03862180045753882, "全帯域_40ms_envelope_correlation": 0.9985908039900147, "全帯域_40ms_modulation_mae_db": 0.40284835097710353, "300–1000 Hz_10ms_log_rms_mae_db": 2.121056909555284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11429296475847223, "300–1000 Hz_10ms_envelope_correlation": 0.9909045972575401, "300–1000 Hz_10ms_modulation_mae_db": 0.6801811144813233, "300–1000 Hz_40ms_log_rms_mae_db": 1.5429374681388108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06674045482683133, "300–1000 Hz_40ms_envelope_correlation": 0.9969523904277036, "300–1000 Hz_40ms_modulation_mae_db": 0.7900705595008247, "1000–3000 Hz_10ms_log_rms_mae_db": 2.093699126973803, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15689308587770331, "1000–3000 Hz_10ms_envelope_correlation": 0.9828320774659207, "1000–3000 Hz_10ms_modulation_mae_db": 0.9101668182571622, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2466709431166252, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0996731302323689, "1000–3000 Hz_40ms_envelope_correlation": 0.9928157047945827, "1000–3000 Hz_40ms_modulation_mae_db": 1.1232917330683845, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3275814234508925, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22575150028103877, "3000–6000 Hz_10ms_envelope_correlation": 0.9649139891882613, "3000–6000 Hz_10ms_modulation_mae_db": 0.3021074024384577, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5553817921160709, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1287400221499884, "3000–6000 Hz_40ms_envelope_correlation": 0.9881478303054763, "3000–6000 Hz_40ms_modulation_mae_db": 0.25902659813628665, "mr_stft_db": 6.248900725849651, "band_mae_db": 0.18227690954202816, "rms_gain_error_db": -0.09249308004808462, "output_peak": 0.28180690383211354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8277582607957652, "全帯域_10ms_rms_envelope_nrmse": 0.07448233718730338, "全帯域_10ms_envelope_correlation": 0.9949629894887889, "全帯域_10ms_modulation_mae_db": 0.3804486597149304, "全帯域_40ms_log_rms_mae_db": 0.44526896781009545, "全帯域_40ms_rms_envelope_nrmse": 0.042287339830695526, "全帯域_40ms_envelope_correlation": 0.9983349255026046, "全帯域_40ms_modulation_mae_db": 0.4520746115155421, "300–1000 Hz_10ms_log_rms_mae_db": 2.1878633023560745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11322249020196103, "300–1000 Hz_10ms_envelope_correlation": 0.9912227979070718, "300–1000 Hz_10ms_modulation_mae_db": 0.40123357391767644, "300–1000 Hz_40ms_log_rms_mae_db": 1.5583959998592307, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07244845412544382, "300–1000 Hz_40ms_envelope_correlation": 0.9964946083557892, "300–1000 Hz_40ms_modulation_mae_db": 0.34807947693680924, "1000–3000 Hz_10ms_log_rms_mae_db": 2.113605714085291, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1577346170309741, "1000–3000 Hz_10ms_envelope_correlation": 0.9827031115138288, "1000–3000 Hz_10ms_modulation_mae_db": 0.49921452926058096, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3350913015729589, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10074259051102193, "1000–3000 Hz_40ms_envelope_correlation": 0.9928392910566022, "1000–3000 Hz_40ms_modulation_mae_db": 0.6969914695536809, "3000–6000 Hz_10ms_log_rms_mae_db": 2.384310627256069, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2378197851136412, "3000–6000 Hz_10ms_envelope_correlation": 0.9623376081989746, "3000–6000 Hz_10ms_modulation_mae_db": 1.2603588533327186, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5726242537262158, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13133264914111895, "3000–6000 Hz_40ms_envelope_correlation": 0.988411797389665, "3000–6000 Hz_40ms_modulation_mae_db": 1.3514884513793952, "mr_stft_db": 6.258749272647234, "band_mae_db": 0.1932275117295809, "rms_gain_error_db": -0.10948325638778887, "output_peak": 0.27469985670044345}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8732404984703391, "全帯域_10ms_rms_envelope_nrmse": 0.07664523246322173, "全帯域_10ms_envelope_correlation": 0.9946353561171675, "全帯域_10ms_modulation_mae_db": 0.2723136202555054, "全帯域_40ms_log_rms_mae_db": 0.48662317910300984, "全帯域_40ms_rms_envelope_nrmse": 0.046740768392353714, "全帯域_40ms_envelope_correlation": 0.9978867405274581, "全帯域_40ms_modulation_mae_db": 0.3814100447466591, "300–1000 Hz_10ms_log_rms_mae_db": 2.1465394101672817, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11719851537603274, "300–1000 Hz_10ms_envelope_correlation": 0.9902955117893759, "300–1000 Hz_10ms_modulation_mae_db": 0.5189697674749761, "300–1000 Hz_40ms_log_rms_mae_db": 1.5625293810711978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07048802502814779, "300–1000 Hz_40ms_envelope_correlation": 0.9964246673479702, "300–1000 Hz_40ms_modulation_mae_db": 0.5710256202948224, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1188840124979813, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1791359742131621, "1000–3000 Hz_10ms_envelope_correlation": 0.9781215719358204, "1000–3000 Hz_10ms_modulation_mae_db": 0.6249759478669361, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2761382639563192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12089290275031535, "1000–3000 Hz_40ms_envelope_correlation": 0.9903340151935041, "1000–3000 Hz_40ms_modulation_mae_db": 0.7617824538841754, "3000–6000 Hz_10ms_log_rms_mae_db": 2.444403213145029, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21828443911733297, "3000–6000 Hz_10ms_envelope_correlation": 0.9670287006676795, "3000–6000 Hz_10ms_modulation_mae_db": 0.5999621104602225, "3000–6000 Hz_40ms_log_rms_mae_db": 1.601143938772976, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12128322179248127, "3000–6000 Hz_40ms_envelope_correlation": 0.9894641411385827, "3000–6000 Hz_40ms_modulation_mae_db": 0.8037086509569832, "mr_stft_db": 6.25796193387728, "band_mae_db": 0.24151413461721716, "rms_gain_error_db": -0.09426068445171759, "output_peak": 0.2939746437078774}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.100345069107726, "spectral_convergence": 0.2499772551419698, "spectral_convergence_status": "ok", "amplitude_floor": 7.181372005990482e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.997527460348404, "spectral_convergence": 0.25630807306466635, "spectral_convergence_status": "ok", "amplitude_floor": 5.417622086660404e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.704352787785859, "spectral_convergence": 0.25938971351623447, "spectral_convergence_status": "ok", "amplitude_floor": 3.166881446630363e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001077391554152542, "output_power": 0.000116914824500544, "delta_db": 0.3549601536395177, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002614688899274717, "output_power": 0.0002503368435075027, "delta_db": -0.18895251163208288, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015854777413607497, "output_power": 0.00014916687722642277, "delta_db": -0.26487751343019367, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.449556222199797e-05, "output_power": 6.056059517178816e-05, "delta_db": -0.2733969831646678, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5950148027800597e-05, "output_power": 1.5804870435895646e-05, "delta_db": -0.03973777988517545, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.2322545231279567e-05, "output_power": 1.2774775181872456e-05, "delta_db": 0.15652844944061473, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21307556519870874, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10602326155742173, "pitch": null, "output_rms_dbfs": -32.17843690330022, "output_peak": 0.28202953062181446, "reference_rms_dbfs": -32.072413641742806, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.267408439080662}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001113800968506657, "log_rms_mae_db": 0.8633912221922009, "rms_envelope_nrmse": 0.0780677452457456, "envelope_correlation": 0.9944533928896357, "rms_gain_error_db": -0.10602928661895938, "reference_mean_power": 0.0006210401213146882, "output_mean_power": 0.0006060615459946561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.9949527250696107e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014723003466329858, "output_power": 0.00014266589619771054, "reference_active": true, "delta_db": -0.13676245448526492}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0749733198322775e-05, "output_power": 2.8882364405627383e-05, "reference_active": true, "delta_db": -0.2720860881701709}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6784890109842875e-06, "output_power": 5.03953548327348e-06, "reference_active": true, "delta_db": -0.5184228226588173}], "active_band_count": 3, "active_band_mae_db": 0.309090455104751}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.537149969198292e-05, "log_rms_mae_db": 0.49151867907826086, "rms_envelope_nrmse": 0.04260414148851901, "envelope_correlation": 0.9982978950883751, "rms_gain_error_db": -0.10600101302635404, "reference_mean_power": 0.0006241534506116662, "output_mean_power": 0.0006091037517587933}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.688739338888854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013414910111714758, "output_power": 0.0001299928309831425, "reference_active": true, "delta_db": -0.13668365053328277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.37620762285614e-05, "output_power": 1.2927887102003756e-05, "reference_active": true, "delta_db": -0.27156408292384226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.149987894354722e-07, "output_power": 3.816050399205397e-07, "reference_active": true, "delta_db": -0.3643272799466802}], "active_band_count": 3, "active_band_mae_db": 0.2575250044679351}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.440030149355929e-05, "log_rms_mae_db": 2.124482060120214, "rms_envelope_nrmse": 0.11933959418606231, "envelope_correlation": 0.9901767851981146, "rms_gain_error_db": -0.25029733312800817, "reference_mean_power": 0.000315142987429675, "output_mean_power": 0.0002974937967095529}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.94905977338963e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010605229421605894, "output_power": 0.00010086148683671422, "reference_active": true, "delta_db": -0.21794702025696933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.109806584151368e-05, "output_power": 1.7839327026787106e-05, "reference_active": true, "delta_db": -0.7286417634914102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3612767263836796e-06, "output_power": 2.399712854284317e-06, "reference_active": true, "delta_db": -1.4634499058020012}], "active_band_count": 3, "active_band_mae_db": 0.8033462298501268}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.828793840107377e-05, "log_rms_mae_db": 1.5143945710664464, "rms_envelope_nrmse": 0.07651684021358697, "envelope_correlation": 0.996011675409836, "rms_gain_error_db": -0.25029464214051467, "reference_mean_power": 0.00031672629889435445, "output_mean_power": 0.000298988622046664}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7988521248624547e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.967318213226487e-05, "output_power": 9.366430512128417e-05, "reference_active": true, "delta_db": -0.27004208105092314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.491202888187961e-06, "output_power": 8.048556890241967e-06, "reference_active": true, "delta_db": -0.7160323901063249}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5805315595038687e-07, "output_power": 2.175497767773885e-07, "reference_active": true, "delta_db": -2.163888640258115}], "active_band_count": 3, "active_band_mae_db": 1.0499877038051209}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.58934180861454e-05, "log_rms_mae_db": 2.173851682270572, "rms_envelope_nrmse": 0.16401052893128612, "envelope_correlation": 0.9814719300222412, "rms_gain_error_db": -0.30562536636078913, "reference_mean_power": 6.36645119573909e-05, "output_mean_power": 5.933826961621499e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.512623526912829e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3991998554391248e-05, "output_power": 1.2230741118946426e-05, "reference_active": true, "delta_db": -0.5842697779278341}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3957717792014314e-06, "output_power": 1.941522565646071e-06, "reference_active": true, "delta_db": -0.9130300230549525}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.685934249147775e-07, "output_power": 7.893700605127136e-07, "reference_active": true, "delta_db": 0.11583985422160818}], "active_band_count": 3, "active_band_mae_db": 0.5377132184014649}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6382034637347546e-05, "log_rms_mae_db": 1.3152754205490849, "rms_envelope_nrmse": 0.10373618234695647, "envelope_correlation": 0.9927115782431692, "rms_gain_error_db": -0.3055468825524604, "reference_mean_power": 6.398182095924481e-05, "output_mean_power": 5.963509397114748e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.275432891493173e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2825234350232933e-05, "output_power": 1.1166286016289724e-05, "reference_active": true, "delta_db": -0.6015656174685751}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.068139769619595e-06, "output_power": 9.049623795129241e-07, "reference_active": true, "delta_db": -0.7199755989699969}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.783047579212978e-08, "output_power": 5.816214491698292e-08, "reference_active": true, "delta_db": 0.024836487430503516}], "active_band_count": 3, "active_band_mae_db": 0.44879256795635847}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.0860899025143115e-05, "log_rms_mae_db": 2.3894301327917047, "rms_envelope_nrmse": 0.21730478095435923, "envelope_correlation": 0.9675197397748442, "rms_gain_error_db": 0.07738182195316544, "reference_mean_power": 1.1507041647425964e-05, "output_mean_power": 1.1713909573531315e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.593308449162533e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.586447862335724e-06, "output_power": 4.024272826912164e-06, "reference_active": true, "delta_db": -0.5678904472926599}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.405946680666416e-07, "output_power": 7.829494286989109e-07, "reference_active": true, "delta_db": -0.30852919375050075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329768801508531e-07, "output_power": 4.009882397501561e-07, "reference_active": true, "delta_db": 0.8071755590790164}], "active_band_count": 3, "active_band_mae_db": 0.5611984000407256}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4734203651083813e-05, "log_rms_mae_db": 1.5435637686100119, "rms_envelope_nrmse": 0.12203287717320353, "envelope_correlation": 0.9892990790501431, "rms_gain_error_db": 0.07804658745198328, "reference_mean_power": 1.156194575111937e-05, "output_mean_power": 1.1771602432872488e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.1073798497935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.414350374295127e-06, "output_power": 3.943907830859452e-06, "reference_active": true, "delta_db": -0.4894004394138351}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649548666103306e-07, "output_power": 4.23476241759959e-07, "reference_active": true, "delta_db": -0.405817475887365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1590180837860948e-08, "output_power": 2.7271569005278206e-08, "reference_active": true, "delta_db": 1.0145384479716741}], "active_band_count": 3, "active_band_mae_db": 0.6365854544242914}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8759989458845358, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bcd808d8e719313d6988ddfa45d2805521196721fcad588bee2e26e597bd1c1a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9809200578730581, "standard_deviation": 0.0022001945748109297, "minimum": 0.9773525385829568, "maximum": 0.9831117998801872, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1303323572314, "standard_deviation": 0.03256386146573854, "minimum": 2.093699126973803, "maximum": 2.1820892768446214, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.704273391152636, "standard_deviation": 0.28515220818725134, "minimum": 0.30306545668483853, "maximum": 1.1597408569185743, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16563641168825205, "standard_deviation": 0.009918977882477935, "minimum": 0.15528439420307835, "maximum": 0.18054790832631332, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9918946114996009, "standard_deviation": 0.0013528507940399374, "minimum": 0.9894922393609052, "maximum": 0.9928607646350087, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2921561625123639, "standard_deviation": 0.031788247006371645, "minimum": 1.2466709431166252, "maximum": 1.3350913015729589, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.917125972208374, "standard_deviation": 0.3662866608983478, "minimum": 0.44879256795635847, "maximum": 1.481227122789327, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10656725004950429, "standard_deviation": 0.00998393123983379, "minimum": 0.09784742029345747, "maximum": 0.12231951692425927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9660050289105507, "standard_deviation": 0.0031431561814084814, "minimum": 0.96050300834064, "maximum": 0.9692189363430669, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.367969962340467, "standard_deviation": 0.05211735423852932, "minimum": 2.2725893709759433, "maximum": 2.444403213145029, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6935963573007616, "standard_deviation": 0.29817105506530167, "minimum": 0.3021074024384577, "maximum": 1.2603588533327186, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22302376652289632, "standard_deviation": 0.010470277804946401, "minimum": 0.21047139246620378, "maximum": 0.23898477441891025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9888728268822365, "standard_deviation": 0.0011972011234010836, "minimum": 0.9865185597015071, "maximum": 0.9904757511596849, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5568037252462628, "standard_deviation": 0.027457875593025413, "minimum": 1.5174784467416924, "maximum": 1.601143938772976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7823048429548235, "standard_deviation": 0.44629694680559223, "minimum": 0.25902659813628665, "maximum": 1.5204699391853425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12568356991947824, "standard_deviation": 0.006611763435014864, "minimum": 0.11527520724913269, "maximum": 0.13676511601582336, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9910114742040406, "standard_deviation": 0.0009703020648736945, "minimum": 0.9901042100569875, "maximum": 0.9931260048426742, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1149366278273574, "standard_deviation": 0.04660089213405581, "minimum": 2.0412518406608364, "maximum": 2.1878633023560745, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6102701039648237, "standard_deviation": 0.18039536018210126, "minimum": 0.3608191196729447, "maximum": 0.8095800319298448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11324359093119424, "standard_deviation": 0.006571968941532065, "minimum": 0.09830764903426421, "maximum": 0.11933959418606231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967545488593565, "standard_deviation": 0.0004917278973454737, "minimum": 0.996011675409836, "maximum": 0.9976976024270687, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.516592792671881, "standard_deviation": 0.035761323343005455, "minimum": 1.46381047393495, "maximum": 1.5625293810711978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7396124362189793, "standard_deviation": 0.25397140025006165, "minimum": 0.34807947693680924, "maximum": 1.0659786968285676, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06792122181837287, "standard_deviation": 0.006149414239505672, "minimum": 0.055308639279410536, "maximum": 0.07651684021358697, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18547551075869698, "standard_deviation": 0.03894296211230756, "minimum": 0.12587734486758298, "maximum": 0.24151413461721716, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.25720887199874, "standard_deviation": 0.005190869061064391, "minimum": 6.248900725849651, "maximum": 6.267408439080662, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2872818665069149, "standard_deviation": 0.008759203836060452, "minimum": 0.27469985670044345, "maximum": 0.3008129130835902, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08069958560644516, "standard_deviation": 0.041305142962264, "minimum": -0.10948325638778887, "maximum": 0.005515550967584604, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946839539309934, "standard_deviation": 0.0003425864245499879, "minimum": 0.9942619577480407, "maximum": 0.995313519344873, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8561871944577181, "standard_deviation": 0.020666926456796637, "minimum": 0.8243369960268745, "maximum": 0.8821699333222669, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3319487140153081, "standard_deviation": 0.08125491938384986, "minimum": 0.2047460189314878, "maximum": 0.4313352962862988, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07620549180635053, "standard_deviation": 0.0023062739957408415, "minimum": 0.07169103506551863, "maximum": 0.07878772224595708, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983295938552259, "standard_deviation": 0.0003282912335075304, "minimum": 0.9977897952607546, "maximum": 0.9986626754187207, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4637976360048418, "standard_deviation": 0.022527961493054284, "minimum": 0.4361550318048866, "maximum": 0.49151867907826086, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3417029013642423, "standard_deviation": 0.11589159018132256, "minimum": 0.16014844433976838, "maximum": 0.501213936597547, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041548573225042726, "standard_deviation": 0.003719420032441672, "minimum": 0.0364438349389658, "maximum": 0.046740768392353714, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0013", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.599548578262329, "predicted_mos": 2.599548578262329, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "585c99d4cc4a014bf3a8c9d94d7ddef4b655795913913776748525dbd0e03228", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.0265011000446975, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8673257770961577, "全帯域_10ms_rms_envelope_nrmse": 0.07818337995988621, "全帯域_10ms_envelope_correlation": 0.9943815014843981, "全帯域_10ms_modulation_mae_db": 0.28306776052002897, "全帯域_40ms_log_rms_mae_db": 0.4663692245157994, "全帯域_40ms_rms_envelope_nrmse": 0.045235855552626755, "全帯域_40ms_envelope_correlation": 0.9979863158264959, "全帯域_40ms_modulation_mae_db": 0.4881419549861726, "300–1000 Hz_10ms_log_rms_mae_db": 1.8065742705687629, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11999218114573447, "300–1000 Hz_10ms_envelope_correlation": 0.989764120525079, "300–1000 Hz_10ms_modulation_mae_db": 0.11671152540574618, "300–1000 Hz_40ms_log_rms_mae_db": 1.2357539885275357, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07185321643270129, "300–1000 Hz_40ms_envelope_correlation": 0.996213309103605, "300–1000 Hz_40ms_modulation_mae_db": 0.40362195229420705, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7012594181210463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14475218359432976, "1000–3000 Hz_10ms_envelope_correlation": 0.986128964641859, "1000–3000 Hz_10ms_modulation_mae_db": 0.960388649250607, "1000–3000 Hz_40ms_log_rms_mae_db": 1.076621740263146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09549401170734685, "1000–3000 Hz_40ms_envelope_correlation": 0.9944353254113976, "1000–3000 Hz_40ms_modulation_mae_db": 1.211123080017858, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6744293183928707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17189912837585644, "3000–6000 Hz_10ms_envelope_correlation": 0.9799674477319322, "3000–6000 Hz_10ms_modulation_mae_db": 1.2639929605476539, "3000–6000 Hz_40ms_log_rms_mae_db": 1.119676345062546, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09651104858613245, "3000–6000 Hz_40ms_envelope_correlation": 0.9944143886210813, "3000–6000 Hz_40ms_modulation_mae_db": 1.7736604164325616, "mr_stft_db": 5.73707538401849, "band_mae_db": 0.2215144540401207, "rms_gain_error_db": -0.0735065998902512, "output_peak": 0.2644148823085071}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.896437559018693, "全帯域_10ms_rms_envelope_nrmse": 0.08445510335616033, "全帯域_10ms_envelope_correlation": 0.9933868893670013, "全帯域_10ms_modulation_mae_db": 0.18455414975244122, "全帯域_40ms_log_rms_mae_db": 0.5291155442803767, "全帯域_40ms_rms_envelope_nrmse": 0.042293861513966344, "全帯域_40ms_envelope_correlation": 0.9981916760061533, "全帯域_40ms_modulation_mae_db": 0.15681079393700606, "300–1000 Hz_10ms_log_rms_mae_db": 1.7227643556831809, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12129000927529973, "300–1000 Hz_10ms_envelope_correlation": 0.9894621850108954, "300–1000 Hz_10ms_modulation_mae_db": 0.22533895833443499, "300–1000 Hz_40ms_log_rms_mae_db": 1.233054615737273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06853173691375744, "300–1000 Hz_40ms_envelope_correlation": 0.9963803680125123, "300–1000 Hz_40ms_modulation_mae_db": 0.24988154335185617, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7499197998285425, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1412516627549502, "1000–3000 Hz_10ms_envelope_correlation": 0.9862913753162056, "1000–3000 Hz_10ms_modulation_mae_db": 0.6076366264432146, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1136045103050918, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08996369911561225, "1000–3000 Hz_40ms_envelope_correlation": 0.9944939535270454, "1000–3000 Hz_40ms_modulation_mae_db": 0.6590559614219008, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7155401764160367, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16698836938885292, "3000–6000 Hz_10ms_envelope_correlation": 0.9805583101308691, "3000–6000 Hz_10ms_modulation_mae_db": 0.7107981451540394, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1616171044766932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09720646969519318, "3000–6000 Hz_40ms_envelope_correlation": 0.9932950060028881, "3000–6000 Hz_40ms_modulation_mae_db": 0.9909727854849487, "mr_stft_db": 5.744008056673633, "band_mae_db": 0.19857719064445337, "rms_gain_error_db": -0.042541814747227444, "output_peak": 0.2945212955779474}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8485197567231402, "全帯域_10ms_rms_envelope_nrmse": 0.07869082855894467, "全帯域_10ms_envelope_correlation": 0.994345228999976, "全帯域_10ms_modulation_mae_db": 0.8516431363214498, "全帯域_40ms_log_rms_mae_db": 0.4557010552705565, "全帯域_40ms_rms_envelope_nrmse": 0.04235580155081323, "全帯域_40ms_envelope_correlation": 0.9983070461322574, "全帯域_40ms_modulation_mae_db": 1.015733587486679, "300–1000 Hz_10ms_log_rms_mae_db": 1.7266605039263956, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12243811835133415, "300–1000 Hz_10ms_envelope_correlation": 0.9894379808248308, "300–1000 Hz_10ms_modulation_mae_db": 0.8448678340412453, "300–1000 Hz_40ms_log_rms_mae_db": 1.2129728064652274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07430262677687939, "300–1000 Hz_40ms_envelope_correlation": 0.9960842621429824, "300–1000 Hz_40ms_modulation_mae_db": 1.0427056940423407, "1000–3000 Hz_10ms_log_rms_mae_db": 1.693257235058931, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14373512525604476, "1000–3000 Hz_10ms_envelope_correlation": 0.9858456232973847, "1000–3000 Hz_10ms_modulation_mae_db": 0.6372828834690845, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0866858141207674, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09192198267959154, "1000–3000 Hz_40ms_envelope_correlation": 0.9943027547608406, "1000–3000 Hz_40ms_modulation_mae_db": 0.6277794077835925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.622324741079368, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17794470821701483, "3000–6000 Hz_10ms_envelope_correlation": 0.9779151000182628, "3000–6000 Hz_10ms_modulation_mae_db": 2.0346921063128494, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1018735053415787, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1015557049859649, "3000–6000 Hz_40ms_envelope_correlation": 0.9927552239659199, "3000–6000 Hz_40ms_modulation_mae_db": 2.119739278991216, "mr_stft_db": 5.742759350252217, "band_mae_db": 0.23954046405656282, "rms_gain_error_db": -0.10871344209732875, "output_peak": 0.292322046135709}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8370982666030007, "全帯域_10ms_rms_envelope_nrmse": 0.07344766072363938, "全帯域_10ms_envelope_correlation": 0.9950619926342239, "全帯域_10ms_modulation_mae_db": 0.7298418112327992, "全帯域_40ms_log_rms_mae_db": 0.4295613284983723, "全帯域_40ms_rms_envelope_nrmse": 0.03859386034208683, "全帯域_40ms_envelope_correlation": 0.9985794325889766, "全帯域_40ms_modulation_mae_db": 0.8845219705327375, "300–1000 Hz_10ms_log_rms_mae_db": 1.8037813545155188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11336647265560758, "300–1000 Hz_10ms_envelope_correlation": 0.9909052675858191, "300–1000 Hz_10ms_modulation_mae_db": 0.8313877418348482, "300–1000 Hz_40ms_log_rms_mae_db": 1.1923509967439299, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06284804245518634, "300–1000 Hz_40ms_envelope_correlation": 0.9971757817674008, "300–1000 Hz_40ms_modulation_mae_db": 1.0046108536745522, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7299427413238453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13243799653260596, "1000–3000 Hz_10ms_envelope_correlation": 0.9878939251214385, "1000–3000 Hz_10ms_modulation_mae_db": 0.9312097483486573, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0939230602288657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08340259906969041, "1000–3000 Hz_40ms_envelope_correlation": 0.9951762157359647, "1000–3000 Hz_40ms_modulation_mae_db": 1.084679003501093, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6147367827728265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1568979161133582, "3000–6000 Hz_10ms_envelope_correlation": 0.9828202288420107, "3000–6000 Hz_10ms_modulation_mae_db": 0.4379215910466376, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1008982663013467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08027095161047061, "3000–6000 Hz_40ms_envelope_correlation": 0.9953231833587932, "3000–6000 Hz_40ms_modulation_mae_db": 0.31308960070786657, "mr_stft_db": 5.732430425271677, "band_mae_db": 0.1530529172953615, "rms_gain_error_db": -0.09170847571981422, "output_peak": 0.29438224524604223}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8675332005229036, "全帯域_10ms_rms_envelope_nrmse": 0.07929939489747637, "全帯域_10ms_envelope_correlation": 0.9941873434589055, "全帯域_10ms_modulation_mae_db": 0.5303250902460062, "全帯域_40ms_log_rms_mae_db": 0.4765662010043102, "全帯域_40ms_rms_envelope_nrmse": 0.04886422301053832, "全帯域_40ms_envelope_correlation": 0.9975976992687066, "全帯域_40ms_modulation_mae_db": 0.7640423674907314, "300–1000 Hz_10ms_log_rms_mae_db": 1.7546913877029835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11067927106946011, "300–1000 Hz_10ms_envelope_correlation": 0.9912209545364598, "300–1000 Hz_10ms_modulation_mae_db": 0.627752111670246, "300–1000 Hz_40ms_log_rms_mae_db": 1.2607655501032176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07350446459716148, "300–1000 Hz_40ms_envelope_correlation": 0.9958352862203553, "300–1000 Hz_40ms_modulation_mae_db": 0.9925455488577608, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7125149331647618, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15129500619622646, "1000–3000 Hz_10ms_envelope_correlation": 0.9850771844344486, "1000–3000 Hz_10ms_modulation_mae_db": 0.20725833790771633, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1274392023559299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.107727218107437, "1000–3000 Hz_40ms_envelope_correlation": 0.9929570382123745, "1000–3000 Hz_40ms_modulation_mae_db": 0.16404052765317478, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6827704619366823, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19210311318130238, "3000–6000 Hz_10ms_envelope_correlation": 0.9745231845774301, "3000–6000 Hz_10ms_modulation_mae_db": 1.1263887744617787, "3000–6000 Hz_40ms_log_rms_mae_db": 1.104142696758786, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10604127007982826, "3000–6000 Hz_40ms_envelope_correlation": 0.9918039858432137, "3000–6000 Hz_40ms_modulation_mae_db": 1.1648452009311607, "mr_stft_db": 5.732342263565222, "band_mae_db": 0.21348373028194403, "rms_gain_error_db": -0.054534487542060577, "output_peak": 0.2955807321606376}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8380324636072205, "全帯域_10ms_rms_envelope_nrmse": 0.07562705415332165, "全帯域_10ms_envelope_correlation": 0.9948324003741374, "全帯域_10ms_modulation_mae_db": 0.41847515446928685, "全帯域_40ms_log_rms_mae_db": 0.4635017487883275, "全帯域_40ms_rms_envelope_nrmse": 0.040292340069305335, "全帯域_40ms_envelope_correlation": 0.9985392245896277, "全帯域_40ms_modulation_mae_db": 0.38087154334947854, "300–1000 Hz_10ms_log_rms_mae_db": 1.751560721637009, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11100077104119668, "300–1000 Hz_10ms_envelope_correlation": 0.9912976155136046, "300–1000 Hz_10ms_modulation_mae_db": 0.27566191103567944, "300–1000 Hz_40ms_log_rms_mae_db": 1.2312892144528895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06242145132284464, "300–1000 Hz_40ms_envelope_correlation": 0.9972355383907752, "300–1000 Hz_40ms_modulation_mae_db": 0.2496814560985797, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6584509763904682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15164452038296128, "1000–3000 Hz_10ms_envelope_correlation": 0.9852456960016851, "1000–3000 Hz_10ms_modulation_mae_db": 0.5603928268058705, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0556857802054846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10788251043415652, "1000–3000 Hz_40ms_envelope_correlation": 0.9932134237209107, "1000–3000 Hz_40ms_modulation_mae_db": 0.6445972396799707, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6967089365524781, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17806873155991437, "3000–6000 Hz_10ms_envelope_correlation": 0.977818575076219, "3000–6000 Hz_10ms_modulation_mae_db": 0.6501660887518025, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1226091437761596, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09534005367322315, "3000–6000 Hz_40ms_envelope_correlation": 0.9934090927401475, "3000–6000 Hz_40ms_modulation_mae_db": 0.705402722142868, "mr_stft_db": 5.731005817233313, "band_mae_db": 0.22122302946730696, "rms_gain_error_db": -0.12455180736040325, "output_peak": 0.3027679731692564}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8477216705438604, "全帯域_10ms_rms_envelope_nrmse": 0.06970817853462484, "全帯域_10ms_envelope_correlation": 0.9956730121829289, "全帯域_10ms_modulation_mae_db": 0.49472391019136497, "全帯域_40ms_log_rms_mae_db": 0.4319340849934923, "全帯域_40ms_rms_envelope_nrmse": 0.03969444791898115, "全帯域_40ms_envelope_correlation": 0.9986185194082362, "全帯域_40ms_modulation_mae_db": 0.6578473336478226, "300–1000 Hz_10ms_log_rms_mae_db": 1.7495828646332847, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09598652719678649, "300–1000 Hz_10ms_envelope_correlation": 0.9936642732959321, "300–1000 Hz_10ms_modulation_mae_db": 0.7100503550546396, "300–1000 Hz_40ms_log_rms_mae_db": 1.2280869705622777, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05637811532950268, "300–1000 Hz_40ms_envelope_correlation": 0.9979055219685602, "300–1000 Hz_40ms_modulation_mae_db": 0.8374340455842875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7725624966860747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14692054746176286, "1000–3000 Hz_10ms_envelope_correlation": 0.9857966248469742, "1000–3000 Hz_10ms_modulation_mae_db": 0.7816819669807452, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0979491450779684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0973795612546227, "1000–3000 Hz_40ms_envelope_correlation": 0.9943138539580616, "1000–3000 Hz_40ms_modulation_mae_db": 1.0355790083982173, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7229152618567178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16357743332114638, "3000–6000 Hz_10ms_envelope_correlation": 0.9815245135576333, "3000–6000 Hz_10ms_modulation_mae_db": 0.6272392984027139, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2207254347570888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09693854691160837, "3000–6000 Hz_40ms_envelope_correlation": 0.9931424914665132, "3000–6000 Hz_40ms_modulation_mae_db": 0.8610322454555323, "mr_stft_db": 5.73094813218336, "band_mae_db": 0.18812735055194363, "rms_gain_error_db": -0.12201101599008798, "output_peak": 0.2710298648401066}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.867850502516721, "全帯域_10ms_rms_envelope_nrmse": 0.06960661026107307, "全帯域_10ms_envelope_correlation": 0.9956478888496249, "全帯域_10ms_modulation_mae_db": 0.5671035373621117, "全帯域_40ms_log_rms_mae_db": 0.45104415306956375, "全帯域_40ms_rms_envelope_nrmse": 0.037601743215039785, "全帯域_40ms_envelope_correlation": 0.9987419322695024, "全帯域_40ms_modulation_mae_db": 0.5044944033367402, "300–1000 Hz_10ms_log_rms_mae_db": 1.7907260181047049, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11173437474751916, "300–1000 Hz_10ms_envelope_correlation": 0.9911886109071474, "300–1000 Hz_10ms_modulation_mae_db": 0.5762977843895816, "300–1000 Hz_40ms_log_rms_mae_db": 1.2186137808029702, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06605747354607705, "300–1000 Hz_40ms_envelope_correlation": 0.9968772662630166, "300–1000 Hz_40ms_modulation_mae_db": 0.8053292483178297, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7310275823372754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15040876414708718, "1000–3000 Hz_10ms_envelope_correlation": 0.9856672659999453, "1000–3000 Hz_10ms_modulation_mae_db": 0.7143410945070209, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0804569934737334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10067560485328426, "1000–3000 Hz_40ms_envelope_correlation": 0.9945851394413185, "1000–3000 Hz_40ms_modulation_mae_db": 0.7462016848489711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.704378852086153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17166127560027714, "3000–6000 Hz_10ms_envelope_correlation": 0.9802772704140444, "3000–6000 Hz_10ms_modulation_mae_db": 1.0084534706003019, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1618864877005701, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10550403036907233, "3000–6000 Hz_40ms_envelope_correlation": 0.9922842383502626, "3000–6000 Hz_40ms_modulation_mae_db": 1.1100652564010838, "mr_stft_db": 5.740615421170671, "band_mae_db": 0.16588653016493696, "rms_gain_error_db": -0.11513125082980559, "output_peak": 0.2914195811517594}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.470826283793723, "spectral_convergence": 0.2360803439055029, "spectral_convergence_status": "ok", "amplitude_floor": 7.181372005990482e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.416579263848254, "spectral_convergence": 0.2448670767986142, "spectral_convergence_status": "ok", "amplitude_floor": 5.417622086660404e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.323820604413492, "spectral_convergence": 0.2502422054296613, "spectral_convergence_status": "ok", "amplitude_floor": 3.166881446630363e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001077391554152542, "output_power": 0.00011404216333618135, "delta_db": 0.24691880397361687, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002614688899274717, "output_power": 0.00026122603686354556, "delta_db": -0.0040356141894530885, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015854777413607497, "output_power": 0.00014917183084894357, "delta_db": -0.2647332927261295, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.449556222199797e-05, "output_power": 5.879813342640454e-05, "delta_db": -0.4016629353007495, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5950148027800597e-05, "output_power": 1.4759433014742181e-05, "delta_db": -0.33695043611874664, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.2322545231279567e-05, "output_power": 1.2112167184210696e-05, "delta_db": -0.07478564193202848, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2215144540401207, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0735065998902512, "pitch": null, "output_rms_dbfs": -32.14592024163306, "output_peak": 0.2644148823085071, "reference_rms_dbfs": -32.072413641742806, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.73707538401849}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001113800968506657, "log_rms_mae_db": 0.8673257770961577, "rms_envelope_nrmse": 0.07818337995988621, "envelope_correlation": 0.9943815014843981, "rms_gain_error_db": -0.07350635218845285, "reference_mean_power": 0.0006210401213146882, "output_mean_power": 0.0006106171850041673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.9949527250696107e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014723003466329858, "output_power": 0.00014550846049804348, "reference_active": true, "delta_db": -0.05108168499635992}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0749733198322775e-05, "output_power": 2.8390073884361504e-05, "reference_active": true, "delta_db": -0.3467482918248033}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6784890109842875e-06, "output_power": 5.1179431706906515e-06, "reference_active": true, "delta_db": -0.4513733047389237}], "active_band_count": 3, "active_band_mae_db": 0.28306776052002897}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.537149969198292e-05, "log_rms_mae_db": 0.4663692245157994, "rms_envelope_nrmse": 0.045235855552626755, "envelope_correlation": 0.9979863158264959, "rms_gain_error_db": -0.0734997330173079, "reference_mean_power": 0.0006241534506116662, "output_mean_power": 0.0006136791985158427}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.688739338888854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013414910111714758, "output_power": 0.0001331711218363441, "reference_active": true, "delta_db": -0.031777087068112345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.37620762285614e-05, "output_power": 1.2684635738484758e-05, "reference_active": true, "delta_db": -0.3540595887677345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.149987894354722e-07, "output_power": 3.2373370003681595e-07, "reference_active": true, "delta_db": -1.0785891891226709}], "active_band_count": 3, "active_band_mae_db": 0.4881419549861726}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.440030149355929e-05, "log_rms_mae_db": 1.8065742705687629, "rms_envelope_nrmse": 0.11999218114573447, "envelope_correlation": 0.989764120525079, "rms_gain_error_db": -0.15422533564010082, "reference_mean_power": 0.000315142987429675, "output_mean_power": 0.0003041481044007946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.94905977338963e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010605229421605894, "output_power": 0.00010378394592008429, "reference_active": true, "delta_db": -0.0938988921567234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.109806584151368e-05, "output_power": 2.0206440319999154e-05, "reference_active": true, "delta_db": -0.1875283081510779}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3612767263836796e-06, "output_power": 3.4148765323064113e-06, "reference_active": true, "delta_db": 0.06870737590943723}], "active_band_count": 3, "active_band_mae_db": 0.11671152540574618}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.828793840107377e-05, "log_rms_mae_db": 1.2357539885275357, "rms_envelope_nrmse": 0.07185321643270129, "envelope_correlation": 0.996213309103605, "rms_gain_error_db": -0.15421755091591238, "reference_mean_power": 0.00031672629889435445, "output_mean_power": 0.00030567672434319887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7988521248624547e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.967318213226487e-05, "output_power": 9.712883689222953e-05, "reference_active": true, "delta_db": -0.11230135485402902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.491202888187961e-06, "output_power": 9.470327002041272e-06, "reference_active": true, "delta_db": -0.009562821010704861}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5805315595038687e-07, "output_power": 2.786424871332002e-07, "reference_active": true, "delta_db": -1.0890016810178873}], "active_band_count": 3, "active_band_mae_db": 0.40362195229420705}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.58934180861454e-05, "log_rms_mae_db": 1.7012594181210463, "rms_envelope_nrmse": 0.14475218359432976, "envelope_correlation": 0.986128964641859, "rms_gain_error_db": -0.3810623618074282, "reference_mean_power": 6.36645119573909e-05, "output_mean_power": 5.831646335640996e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.512623526912829e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3991998554391248e-05, "output_power": 1.2130532446076408e-05, "reference_active": true, "delta_db": -0.6199988780601875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3957717792014314e-06, "output_power": 2.1087969794874516e-06, "reference_active": true, "delta_db": -0.5541067401427245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.685934249147775e-07, "output_power": 5.187888481300507e-07, "reference_active": true, "delta_db": -1.7070603295489089}], "active_band_count": 3, "active_band_mae_db": 0.960388649250607}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6382034637347546e-05, "log_rms_mae_db": 1.076621740263146, "rms_envelope_nrmse": 0.09549401170734685, "envelope_correlation": 0.9944353254113976, "rms_gain_error_db": -0.38105609487492764, "reference_mean_power": 6.398182095924481e-05, "output_mean_power": 5.8607201828763686e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.275432891493173e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2825234350232933e-05, "output_power": 1.1052624835978126e-05, "reference_active": true, "delta_db": -0.645998807133958}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.068139769619595e-06, "output_power": 9.109006913928534e-07, "reference_active": true, "delta_db": -0.6915705356559215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.783047579212978e-08, "output_power": 3.4086057767877884e-08, "reference_active": true, "delta_db": -2.2957998972636946}], "active_band_count": 3, "active_band_mae_db": 1.211123080017858}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.0860899025143115e-05, "log_rms_mae_db": 1.6744293183928707, "rms_envelope_nrmse": 0.17189912837585644, "envelope_correlation": 0.9799674477319322, "rms_gain_error_db": -0.37980626646372595, "reference_mean_power": 1.1507041647425964e-05, "output_mean_power": 1.0543457658512648e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.593308449162533e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.586447862335724e-06, "output_power": 4.040881042466652e-06, "reference_active": true, "delta_db": -0.5500039526717918}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.405946680666416e-07, "output_power": 5.895907788695587e-07, "reference_active": true, "delta_db": -1.5403594833738852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329768801508531e-07, "output_power": 2.2503627753343912e-07, "reference_active": true, "delta_db": -1.7016154455972847}], "active_band_count": 3, "active_band_mae_db": 1.2639929605476539}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4734203651083813e-05, "log_rms_mae_db": 1.119676345062546, "rms_envelope_nrmse": 0.09651104858613245, "envelope_correlation": 0.9944143886210813, "rms_gain_error_db": -0.37980696880622833, "reference_mean_power": 1.156194575111937e-05, "output_mean_power": 1.0593762454144051e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.1073798497935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.414350374295127e-06, "output_power": 3.852868322134923e-06, "reference_active": true, "delta_db": -0.5908263429927836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649548666103306e-07, "output_power": 2.9256041436811156e-07, "reference_active": true, "delta_db": -2.0119523532069863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1590180837860948e-08, "output_power": 1.1546118878270757e-08, "reference_active": true, "delta_db": -2.718202553097915}], "active_band_count": 3, "active_band_mae_db": 1.7736604164325616}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8944130874009951, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b67d9268196d4fd6806dcca2a3ab67a236a5ed02988a5856d2beba432b5c96bd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9859933324574925, "standard_deviation": 0.0008688742136682779, "minimum": 0.9850771844344486, "maximum": 0.9878939251214385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.718616897863868, "standard_deviation": 0.03540352243050298, "minimum": 1.6584509763904682, "maximum": 1.7725624966860747, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6750240167141145, "standard_deviation": 0.23841630667870434, "minimum": 0.20725833790771633, "maximum": 0.960388649250607, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14530572579074608, "standard_deviation": 0.00643081337351728, "minimum": 0.13243799653260596, "maximum": 0.15164452038296128, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941847130959892, "standard_deviation": 0.0007351437941793531, "minimum": 0.9929570382123745, "maximum": 0.9951762157359647, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0915457807538735, "standard_deviation": 0.022286533086163088, "minimum": 1.0556857802054846, "maximum": 1.1274392023559299, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7716319891630974, "standard_deviation": 0.33383865228896614, "minimum": 0.16404052765317478, "maximum": 1.211123080017858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0968058984027177, "standard_deviation": 0.008529847418298738, "minimum": 0.08340259906969041, "maximum": 0.10788251043415652, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9794255787935502, "standard_deviation": 0.002594740093482922, "minimum": 0.9745231845774301, "maximum": 0.9828202288420107, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6792255663866418, "standard_deviation": 0.040704464159543166, "minimum": 1.6147367827728265, "maximum": 1.7229152618567178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9824565544097221, "standard_deviation": 0.5091684535266362, "minimum": 0.4379215910466376, "maximum": 2.0346921063128494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17239258446971534, "standard_deviation": 0.010685939730386153, "minimum": 0.1568979161133582, "maximum": 0.19210311318130238, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933034512936025, "standard_deviation": 0.0011300052925568, "minimum": 0.9918039858432137, "maximum": 0.9953231833587932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1366786230218462, "standard_deviation": 0.04201781119855653, "minimum": 1.1008982663013467, "maximum": 1.2207254347570888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1298509383184046, "standard_deviation": 0.5775306602289274, "minimum": 0.31308960070786657, "maximum": 2.119739278991216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09742100948893666, "standard_deviation": 0.008073780444878125, "minimum": 0.08027095161047061, "maximum": 0.10604127007982826, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9908676260249709, "standard_deviation": 0.001387304187874083, "minimum": 0.9894379808248308, "maximum": 0.9936642732959321, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.76329268459648, "standard_deviation": 0.03305678749228028, "minimum": 1.7227643556831809, "maximum": 1.8065742705687629, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5260085277208026, "standard_deviation": 0.28341779775861003, "minimum": 0.11671152540574618, "maximum": 0.8448678340412453, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1133109656853673, "standard_deviation": 0.008506269933112168, "minimum": 0.09598652719678649, "maximum": 0.12243811835133415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967134167336509, "standard_deviation": 0.0007032914248899883, "minimum": 0.9958352862203553, "maximum": 0.9979055219685602, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2266109904244151, "standard_deviation": 0.0197824437179892, "minimum": 1.1923509967439299, "maximum": 1.2607655501032176, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6982262927776768, "standard_deviation": 0.3420543869367954, "minimum": 0.2496814560985797, "maximum": 1.0427056940423407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06698714092176379, "standard_deviation": 0.0062501391061010325, "minimum": 0.05637811532950268, "maximum": 0.07430262677687939, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20017570831282874, "standard_deviation": 0.02968655923692502, "minimum": 0.1530529172953615, "maximum": 0.23954046405656282, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.736398106296074, "standard_deviation": 0.005445739768947043, "minimum": 5.73094813218336, "maximum": 5.744008056673633, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.28830482757374576, "standard_deviation": 0.013270237403067207, "minimum": 0.2644148823085071, "maximum": 0.3027679731692564, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09158736177212239, "standard_deviation": 0.031544029407164005, "minimum": -0.12455180736040325, "maximum": -0.042541814747227444, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946895321688995, "standard_deviation": 0.0007752711905512933, "minimum": 0.9933868893670013, "maximum": 0.9956730121829289, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8588148995789622, "standard_deviation": 0.01992171077280767, "minimum": 0.8370982666030007, "maximum": 0.896437559018693, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5074668187619361, "standard_deviation": 0.21866361667605977, "minimum": 0.18455414975244122, "maximum": 0.8516431363214498, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07612727630564081, "standard_deviation": 0.005090170722144937, "minimum": 0.06960661026107307, "maximum": 0.08445510335616033, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983202307612445, "standard_deviation": 0.0003840020952725719, "minimum": 0.9975976992687066, "maximum": 0.9987419322695024, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4629741675525998, "standard_deviation": 0.03127380873445372, "minimum": 0.4295613284983723, "maximum": 0.5291155442803767, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.606557994345921, "standard_deviation": 0.2802465281081949, "minimum": 0.15681079393700606, "maximum": 1.015733587486679, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04186651664666972, "standard_deviation": 0.003719386442042334, "minimum": 0.037601743215039785, "maximum": 0.04886422301053832, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0013", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.786191463470459, "predicted_mos": 2.786191463470459, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0e609a2823383d54ee50d2bc1981c62381348487d90fb5f60dd61297af58d58a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.986823100131005, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9013270388345159, "全帯域_10ms_rms_envelope_nrmse": 0.0805476938802604, "全帯域_10ms_envelope_correlation": 0.9940456415128455, "全帯域_10ms_modulation_mae_db": 0.14975570675780073, "全帯域_40ms_log_rms_mae_db": 0.4990853920582688, "全帯域_40ms_rms_envelope_nrmse": 0.046945366651766254, "全帯域_40ms_envelope_correlation": 0.9978612696402591, "全帯域_40ms_modulation_mae_db": 0.11265987832769948, "300–1000 Hz_10ms_log_rms_mae_db": 1.8203508048555181, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11289238194486047, "300–1000 Hz_10ms_envelope_correlation": 0.9911431253116509, "300–1000 Hz_10ms_modulation_mae_db": 0.33561470251191866, "300–1000 Hz_40ms_log_rms_mae_db": 1.3282751791232663, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0717467544972506, "300–1000 Hz_40ms_envelope_correlation": 0.9964547736848842, "300–1000 Hz_40ms_modulation_mae_db": 0.3010483112174921, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8537957846795488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15712643942149282, "1000–3000 Hz_10ms_envelope_correlation": 0.9826296753964495, "1000–3000 Hz_10ms_modulation_mae_db": 0.5879022121568903, "1000–3000 Hz_40ms_log_rms_mae_db": 1.151557967068087, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0969992842179514, "1000–3000 Hz_40ms_envelope_correlation": 0.9930289365500822, "1000–3000 Hz_40ms_modulation_mae_db": 0.46627072983615797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8973828253828484, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20259986622328133, "3000–6000 Hz_10ms_envelope_correlation": 0.9732986194430238, "3000–6000 Hz_10ms_modulation_mae_db": 1.6675713705395545, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2614290814632414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13040304141569775, "3000–6000 Hz_40ms_envelope_correlation": 0.9887968562731905, "3000–6000 Hz_40ms_modulation_mae_db": 2.020291629768318, "mr_stft_db": 5.861036609048255, "band_mae_db": 0.17842329448710645, "rms_gain_error_db": -0.08556873475659393, "output_peak": 0.2944960879581819}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8784864736823965, "全帯域_10ms_rms_envelope_nrmse": 0.07409579272491416, "全帯域_10ms_envelope_correlation": 0.9949718096264797, "全帯域_10ms_modulation_mae_db": 0.4525518191287116, "全帯域_40ms_log_rms_mae_db": 0.4309029598334382, "全帯域_40ms_rms_envelope_nrmse": 0.035872859267667236, "全帯域_40ms_envelope_correlation": 0.9987847271727502, "全帯域_40ms_modulation_mae_db": 0.3908202827626029, "300–1000 Hz_10ms_log_rms_mae_db": 1.8299088957990677, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1070928175467302, "300–1000 Hz_10ms_envelope_correlation": 0.9920852344169333, "300–1000 Hz_10ms_modulation_mae_db": 0.8020422134702841, "300–1000 Hz_40ms_log_rms_mae_db": 1.2562200509188515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06494560384685233, "300–1000 Hz_40ms_envelope_correlation": 0.9971702764587884, "300–1000 Hz_40ms_modulation_mae_db": 0.9405364012539481, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8910973849183699, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1387028202754865, "1000–3000 Hz_10ms_envelope_correlation": 0.9866707967582836, "1000–3000 Hz_10ms_modulation_mae_db": 0.1931041727369495, "1000–3000 Hz_40ms_log_rms_mae_db": 1.106061683330293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08755122826602775, "1000–3000 Hz_40ms_envelope_correlation": 0.9946260881041509, "1000–3000 Hz_40ms_modulation_mae_db": 0.27250640532078857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8443513383848023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18481228256496693, "3000–6000 Hz_10ms_envelope_correlation": 0.976606510377002, "3000–6000 Hz_10ms_modulation_mae_db": 0.5748561763795115, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2145679449714224, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10740370987140997, "3000–6000 Hz_40ms_envelope_correlation": 0.991692959785604, "3000–6000 Hz_40ms_modulation_mae_db": 0.49127351672617037, "mr_stft_db": 5.862952163280911, "band_mae_db": 0.15439631733953513, "rms_gain_error_db": -0.0855121120143561, "output_peak": 0.285919750313891}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8402129149812874, "全帯域_10ms_rms_envelope_nrmse": 0.06907424598168621, "全帯域_10ms_envelope_correlation": 0.9958523670757893, "全帯域_10ms_modulation_mae_db": 0.408366812561322, "全帯域_40ms_log_rms_mae_db": 0.45077019210541025, "全帯域_40ms_rms_envelope_nrmse": 0.04082681823305296, "全帯域_40ms_envelope_correlation": 0.9986481440309825, "全帯域_40ms_modulation_mae_db": 0.42912098056533243, "300–1000 Hz_10ms_log_rms_mae_db": 1.7823170298553193, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10160450359438321, "300–1000 Hz_10ms_envelope_correlation": 0.9930287765744954, "300–1000 Hz_10ms_modulation_mae_db": 0.5449890640807595, "300–1000 Hz_40ms_log_rms_mae_db": 1.2322556555738529, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06604516365213296, "300–1000 Hz_40ms_envelope_correlation": 0.997185838458315, "300–1000 Hz_40ms_modulation_mae_db": 0.6558328966348577, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8825093094261764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17488341744278746, "1000–3000 Hz_10ms_envelope_correlation": 0.9793778198032422, "1000–3000 Hz_10ms_modulation_mae_db": 1.2154055641606059, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1646582616972672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10790521831013608, "1000–3000 Hz_40ms_envelope_correlation": 0.9929332799275004, "1000–3000 Hz_40ms_modulation_mae_db": 1.5545286377379857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8569072861363773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16628319299722524, "3000–6000 Hz_10ms_envelope_correlation": 0.9810372830977618, "3000–6000 Hz_10ms_modulation_mae_db": 1.0617678069250784, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2605646139428526, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09323899873696091, "3000–6000 Hz_40ms_envelope_correlation": 0.9943938191740924, "3000–6000 Hz_40ms_modulation_mae_db": 1.1260223433854695, "mr_stft_db": 5.859049352212377, "band_mae_db": 0.27328363206695305, "rms_gain_error_db": -0.16532032170298983, "output_peak": 0.2741371957634726}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8620371926599618, "全帯域_10ms_rms_envelope_nrmse": 0.08307959144955285, "全帯域_10ms_envelope_correlation": 0.9936957315448188, "全帯域_10ms_modulation_mae_db": 0.43551404242257724, "全帯域_40ms_log_rms_mae_db": 0.4637475455069267, "全帯域_40ms_rms_envelope_nrmse": 0.04329928074770395, "全帯域_40ms_envelope_correlation": 0.9982434661343065, "全帯域_40ms_modulation_mae_db": 0.7124531431302349, "300–1000 Hz_10ms_log_rms_mae_db": 1.840955894036656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11655492277204538, "300–1000 Hz_10ms_envelope_correlation": 0.9904932025943566, "300–1000 Hz_10ms_modulation_mae_db": 0.3752596653214457, "300–1000 Hz_40ms_log_rms_mae_db": 1.2742272093520033, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06490974607917388, "300–1000 Hz_40ms_envelope_correlation": 0.9971489362016747, "300–1000 Hz_40ms_modulation_mae_db": 0.627448435129783, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8344300205948607, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14243916906193285, "1000–3000 Hz_10ms_envelope_correlation": 0.9861165929673847, "1000–3000 Hz_10ms_modulation_mae_db": 1.3212651267774527, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1580655120036794, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09182985143092219, "1000–3000 Hz_40ms_envelope_correlation": 0.9943064575321278, "1000–3000 Hz_40ms_modulation_mae_db": 1.4946926960573486, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8410696735291694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18100311766158356, "3000–6000 Hz_10ms_envelope_correlation": 0.9770816564863711, "3000–6000 Hz_10ms_modulation_mae_db": 2.0262447676582314, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1475894413810959, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09190407447300428, "3000–6000 Hz_40ms_envelope_correlation": 0.9939808712450421, "3000–6000 Hz_40ms_modulation_mae_db": 2.1849446424915526, "mr_stft_db": 5.848623468930067, "band_mae_db": 0.21755493056140882, "rms_gain_error_db": -0.1086736575874163, "output_peak": 0.2794505819573225}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9153911396180525, "全帯域_10ms_rms_envelope_nrmse": 0.0850249770728744, "全帯域_10ms_envelope_correlation": 0.9933040928756874, "全帯域_10ms_modulation_mae_db": 0.11694819905047614, "全帯域_40ms_log_rms_mae_db": 0.5199895699391114, "全帯域_40ms_rms_envelope_nrmse": 0.05081509085136901, "全帯域_40ms_envelope_correlation": 0.9973873693487058, "全帯域_40ms_modulation_mae_db": 0.1253603548175834, "300–1000 Hz_10ms_log_rms_mae_db": 1.9053881266533135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13349001462113946, "300–1000 Hz_10ms_envelope_correlation": 0.9872216879836295, "300–1000 Hz_10ms_modulation_mae_db": 0.346006533609055, "300–1000 Hz_40ms_log_rms_mae_db": 1.3259071744352784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08683333043955167, "300–1000 Hz_40ms_envelope_correlation": 0.9942506663493951, "300–1000 Hz_40ms_modulation_mae_db": 0.4835253185310558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8622733814038583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.157663355308836, "1000–3000 Hz_10ms_envelope_correlation": 0.9825444564929198, "1000–3000 Hz_10ms_modulation_mae_db": 0.5872128407718087, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1564119849451322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09465024873294096, "1000–3000 Hz_40ms_envelope_correlation": 0.9933640048654354, "1000–3000 Hz_40ms_modulation_mae_db": 0.5425665302628002, "3000–6000 Hz_10ms_log_rms_mae_db": 1.895689028783174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2014473811041922, "3000–6000 Hz_10ms_envelope_correlation": 0.9721302316966874, "3000–6000 Hz_10ms_modulation_mae_db": 0.7220214755730906, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2182145824502344, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10470291857997749, "3000–6000 Hz_40ms_envelope_correlation": 0.992094691628688, "3000–6000 Hz_40ms_modulation_mae_db": 0.5947036878621577, "mr_stft_db": 5.842346040394468, "band_mae_db": 0.1389958518384731, "rms_gain_error_db": -0.05019440295299265, "output_peak": 0.2708601433110817}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8420306108340396, "全帯域_10ms_rms_envelope_nrmse": 0.07195365033849883, "全帯域_10ms_envelope_correlation": 0.9952051785379554, "全帯域_10ms_modulation_mae_db": 0.15011885690579085, "全帯域_40ms_log_rms_mae_db": 0.44707232764197946, "全帯域_40ms_rms_envelope_nrmse": 0.03709477179435118, "全帯域_40ms_envelope_correlation": 0.9986088336832356, "全帯域_40ms_modulation_mae_db": 0.07796195178416547, "300–1000 Hz_10ms_log_rms_mae_db": 1.8377242864104018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11099442215118496, "300–1000 Hz_10ms_envelope_correlation": 0.9911870115055041, "300–1000 Hz_10ms_modulation_mae_db": 0.3479211675098058, "300–1000 Hz_40ms_log_rms_mae_db": 1.2607171280816507, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07035709983691132, "300–1000 Hz_40ms_envelope_correlation": 0.9962260897787787, "300–1000 Hz_40ms_modulation_mae_db": 0.17976240531136167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7687725652625987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13246416988960913, "1000–3000 Hz_10ms_envelope_correlation": 0.9877282497710576, "1000–3000 Hz_10ms_modulation_mae_db": 0.33806307319073897, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0927522385556405, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08116477520597114, "1000–3000 Hz_40ms_envelope_correlation": 0.9950805229627401, "1000–3000 Hz_40ms_modulation_mae_db": 0.25021326028557717, "3000–6000 Hz_10ms_log_rms_mae_db": 1.871747997493188, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17775669243802628, "3000–6000 Hz_10ms_envelope_correlation": 0.9779179593030553, "3000–6000 Hz_10ms_modulation_mae_db": 0.42342373827385904, "3000–6000 Hz_40ms_log_rms_mae_db": 1.210750163689026, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09930231736826024, "3000–6000 Hz_40ms_envelope_correlation": 0.9928319591879027, "3000–6000 Hz_40ms_modulation_mae_db": 0.49788035321026053, "mr_stft_db": 5.857030242729383, "band_mae_db": 0.09739052611076156, "rms_gain_error_db": -0.035961962443355346, "output_peak": 0.30752215783372144}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8323927827698706, "全帯域_10ms_rms_envelope_nrmse": 0.07108430866441273, "全帯域_10ms_envelope_correlation": 0.9953916234292189, "全帯域_10ms_modulation_mae_db": 0.3511907109340799, "全帯域_40ms_log_rms_mae_db": 0.4298843063548515, "全帯域_40ms_rms_envelope_nrmse": 0.03796541581531107, "全帯域_40ms_envelope_correlation": 0.9986366281320833, "全帯域_40ms_modulation_mae_db": 0.29339009315192727, "300–1000 Hz_10ms_log_rms_mae_db": 1.8342091786097217, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11428918490187667, "300–1000 Hz_10ms_envelope_correlation": 0.9907342062339304, "300–1000 Hz_10ms_modulation_mae_db": 0.5498786633544243, "300–1000 Hz_40ms_log_rms_mae_db": 1.2260633646745793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0719539256217956, "300–1000 Hz_40ms_envelope_correlation": 0.9961730811205074, "300–1000 Hz_40ms_modulation_mae_db": 0.5788491745435298, "1000–3000 Hz_10ms_log_rms_mae_db": 1.867919850128963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1585656107785075, "1000–3000 Hz_10ms_envelope_correlation": 0.9832678211402534, "1000–3000 Hz_10ms_modulation_mae_db": 0.40255537563073823, "1000–3000 Hz_40ms_log_rms_mae_db": 1.158984521821188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10060595841256842, "1000–3000 Hz_40ms_envelope_correlation": 0.9939462736696942, "1000–3000 Hz_40ms_modulation_mae_db": 0.5199118236691894, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8844620778985524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1725065669424334, "3000–6000 Hz_10ms_envelope_correlation": 0.9800618113413438, "3000–6000 Hz_10ms_modulation_mae_db": 0.4905501464493185, "3000–6000 Hz_40ms_log_rms_mae_db": 1.231415322794435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11331128809365851, "3000–6000 Hz_40ms_envelope_correlation": 0.9910768245006584, "3000–6000 Hz_40ms_modulation_mae_db": 0.5970568376967741, "mr_stft_db": 5.85520233099536, "band_mae_db": 0.1634009704030667, "rms_gain_error_db": -0.09114856355326605, "output_peak": 0.2927838727689681}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8561364326450631, "全帯域_10ms_rms_envelope_nrmse": 0.08104882001386263, "全帯域_10ms_envelope_correlation": 0.994058537173271, "全帯域_10ms_modulation_mae_db": 0.5873766996234424, "全帯域_40ms_log_rms_mae_db": 0.475827333524942, "全帯域_40ms_rms_envelope_nrmse": 0.04893230540617979, "全帯域_40ms_envelope_correlation": 0.9977704528771936, "全帯域_40ms_modulation_mae_db": 0.9220364874166811, "300–1000 Hz_10ms_log_rms_mae_db": 1.8764531832214875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11837091565008424, "300–1000 Hz_10ms_envelope_correlation": 0.9901599247222316, "300–1000 Hz_10ms_modulation_mae_db": 0.6086541145883557, "300–1000 Hz_40ms_log_rms_mae_db": 1.2521024709318949, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07336215469118287, "300–1000 Hz_40ms_envelope_correlation": 0.9961717712648074, "300–1000 Hz_40ms_modulation_mae_db": 1.2017335128688218, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8639575238543977, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1643219038128842, "1000–3000 Hz_10ms_envelope_correlation": 0.9812060462343852, "1000–3000 Hz_10ms_modulation_mae_db": 0.34780936956115466, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1274434605641848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09290291172827132, "1000–3000 Hz_40ms_envelope_correlation": 0.994116456252067, "1000–3000 Hz_40ms_modulation_mae_db": 0.4079717352984264, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8574884959155233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17255638081323696, "3000–6000 Hz_10ms_envelope_correlation": 0.9792083841957028, "3000–6000 Hz_10ms_modulation_mae_db": 0.49786916712590973, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1652522786089656, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09640988919962307, "3000–6000 Hz_40ms_envelope_correlation": 0.993384862197573, "3000–6000 Hz_40ms_modulation_mae_db": 0.7938818168637654, "mr_stft_db": 5.858298125054645, "band_mae_db": 0.20327996302120488, "rms_gain_error_db": -0.13066532812323065, "output_peak": 0.2752100902072414}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6697137946582306, "spectral_convergence": 0.24548293302304833, "spectral_convergence_status": "ok", "amplitude_floor": 7.181372005990482e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.53956174083629, "spectral_convergence": 0.24418220760357506, "spectral_convergence_status": "ok", "amplitude_floor": 5.417622086660404e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.373834291650244, "spectral_convergence": 0.24768317269767923, "spectral_convergence_status": "ok", "amplitude_floor": 3.166881446630363e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001077391554152542, "output_power": 0.00011387345841080127, "delta_db": 0.2404894393708598, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002614688899274717, "output_power": 0.0002502800227946012, "delta_db": -0.18993837223079518, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015854777413607497, "output_power": 0.0001516315840692547, "delta_db": -0.1937047731616316, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.449556222199797e-05, "output_power": 6.308226196576366e-05, "delta_db": -0.0962257534231308, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5950148027800597e-05, "output_power": 1.7067220593124853e-05, "delta_db": 0.29398083813441395, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.2322545231279567e-05, "output_power": 1.2483043387996311e-05, "delta_db": 0.056200590601807286, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17842329448710645, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08556873475659393, "pitch": null, "output_rms_dbfs": -32.157982376499405, "output_peak": 0.2944960879581819, "reference_rms_dbfs": -32.072413641742806, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.861036609048255}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001113800968506657, "log_rms_mae_db": 0.9013270388345159, "rms_envelope_nrmse": 0.0805476938802604, "envelope_correlation": 0.9940456415128455, "rms_gain_error_db": -0.08557166415140123, "reference_mean_power": 0.0006210401213146882, "output_mean_power": 0.0006089231587330536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.9949527250696107e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014723003466329858, "output_power": 0.00014238359683206686, "reference_active": true, "delta_db": -0.14536454639028804}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0749733198322775e-05, "output_power": 2.9861841444910515e-05, "reference_active": true, "delta_db": -0.12724766754606243}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6784890109842875e-06, "output_power": 5.914230958712359e-06, "reference_active": true, "delta_db": 0.17665490633705166}], "active_band_count": 3, "active_band_mae_db": 0.14975570675780073}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.537149969198292e-05, "log_rms_mae_db": 0.4990853920582688, "rms_envelope_nrmse": 0.046945366651766254, "envelope_correlation": 0.9978612696402591, "rms_gain_error_db": -0.08556032387167323, "reference_mean_power": 0.0006241534506116662, "output_mean_power": 0.000611977342610869}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.688739338888854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013414910111714758, "output_power": 0.0001295478037797979, "reference_active": true, "delta_db": -0.1515771217078626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.37620762285614e-05, "output_power": 1.3778616001957279e-05, "reference_active": true, "delta_db": 0.005216378093343767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.149987894354722e-07, "output_power": 4.32678627469435e-07, "reference_active": true, "delta_db": 0.18118613518189208}], "active_band_count": 3, "active_band_mae_db": 0.11265987832769948}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.440030149355929e-05, "log_rms_mae_db": 1.8203508048555181, "rms_envelope_nrmse": 0.11289238194486047, "envelope_correlation": 0.9911431253116509, "rms_gain_error_db": -0.22485122305436536, "reference_mean_power": 0.000315142987429675, "output_mean_power": 0.00029924198389016894}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.94905977338963e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010605229421605894, "output_power": 0.00010216969923550323, "reference_active": true, "delta_db": -0.16197953074462648}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.109806584151368e-05, "output_power": 1.948209111555277e-05, "reference_active": true, "delta_db": -0.34607073152896495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3612767263836796e-06, "output_power": 2.9965731506064926e-06, "reference_active": true, "delta_db": -0.4987938452621644}], "active_band_count": 3, "active_band_mae_db": 0.33561470251191866}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.828793840107377e-05, "log_rms_mae_db": 1.3282751791232663, "rms_envelope_nrmse": 0.0717467544972506, "envelope_correlation": 0.9964547736848842, "rms_gain_error_db": -0.2248511664888175, "reference_mean_power": 0.00031672629889435445, "output_mean_power": 0.0003007454109602853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7988521248624547e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.967318213226487e-05, "output_power": 9.534234313836311e-05, "reference_active": true, "delta_db": -0.1929250265499725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.491202888187961e-06, "output_power": 9.261102846314752e-06, "reference_active": true, "delta_db": -0.10658550017130972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5805315595038687e-07, "output_power": 3.1159039756718693e-07, "reference_active": true, "delta_db": -0.6036344069311941}], "active_band_count": 3, "active_band_mae_db": 0.3010483112174921}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.58934180861454e-05, "log_rms_mae_db": 1.8537957846795488, "rms_envelope_nrmse": 0.15712643942149282, "envelope_correlation": 0.9826296753964495, "rms_gain_error_db": -0.12947045618553216, "reference_mean_power": 6.36645119573909e-05, "output_mean_power": 6.179457768967519e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.512623526912829e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3991998554391248e-05, "output_power": 1.1662431890780763e-05, "reference_active": true, "delta_db": -0.7909063114148702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3957717792014314e-06, "output_power": 2.275254581631099e-06, "reference_active": true, "delta_db": -0.2241544730188362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.685934249147775e-07, "output_power": 6.468924458922674e-07, "reference_active": true, "delta_db": -0.7486458520369645}], "active_band_count": 3, "active_band_mae_db": 0.5879022121568903}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6382034637347546e-05, "log_rms_mae_db": 1.151557967068087, "rms_envelope_nrmse": 0.0969992842179514, "envelope_correlation": 0.9930289365500822, "rms_gain_error_db": -0.12942961458841826, "reference_mean_power": 6.398182095924481e-05, "output_mean_power": 6.210315081316298e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.275432891493173e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2825234350232933e-05, "output_power": 1.0660972259214401e-05, "reference_active": true, "delta_db": -0.8026849626141568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.068139769619595e-06, "output_power": 1.044306575598446e-06, "reference_active": true, "delta_db": -0.09800072680963766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.783047579212978e-08, "output_power": 5.1563705035728875e-08, "reference_active": true, "delta_db": -0.4981265000846795}], "active_band_count": 3, "active_band_mae_db": 0.46627072983615797}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.0860899025143115e-05, "log_rms_mae_db": 1.8973828253828484, "rms_envelope_nrmse": 0.20259986622328133, "envelope_correlation": 0.9732986194430238, "rms_gain_error_db": 0.2787024001612432, "reference_mean_power": 1.1507041647425964e-05, "output_mean_power": 1.2269699457681005e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.593308449162533e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.586447862335724e-06, "output_power": 5.2808182758922355e-06, "reference_active": true, "delta_db": 0.6122476184166056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.405946680666416e-07, "output_power": 1.4592063307397791e-06, "reference_active": true, "delta_db": 2.39530074240158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329768801508531e-07, "output_power": 5.271456837953421e-07, "reference_active": true, "delta_db": 1.995165750800478}], "active_band_count": 3, "active_band_mae_db": 1.6675713705395545}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4734203651083813e-05, "log_rms_mae_db": 1.2614290814632414, "rms_envelope_nrmse": 0.13040304141569775, "envelope_correlation": 0.9887968562731905, "rms_gain_error_db": 0.27909121972555556, "reference_mean_power": 1.156194575111937e-05, "output_mean_power": 1.2329346252219923e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.1073798497935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.414350374295127e-06, "output_power": 5.341016340458465e-06, "reference_active": true, "delta_db": 0.8275710579864617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649548666103306e-07, "output_power": 8.141883557479868e-07, "reference_active": true, "delta_db": 2.433140891876934}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1590180837860948e-08, "output_power": 4.114078497921975e-08, "reference_active": true, "delta_db": 2.80016293944156}], "active_band_count": 3, "active_band_mae_db": 2.020291629768318}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8932494776187272, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b0c632d1ff6fe96a4b0134415f49432c2c844d604123a91f62985cd91952745c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.983692682320497, "standard_deviation": 0.002889899255248538, "minimum": 0.9793778198032422, "maximum": 0.9877282497710576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8530944775335967, "standard_deviation": 0.03815483811814314, "minimum": 1.7687725652625987, "maximum": 1.8910973849183699, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6241647168732924, "standard_deviation": 0.4194120804732407, "minimum": 0.1931041727369495, "maximum": 1.3212651267774527, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15327086074894206, "standard_deviation": 0.014212136490814596, "minimum": 0.13246416988960913, "maximum": 0.17488341744278746, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939252524829747, "standard_deviation": 0.0007663018056127508, "minimum": 0.9929332799275004, "maximum": 0.9950805229627401, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.139491953748184, "standard_deviation": 0.027353523172864685, "minimum": 1.0927522385556405, "maximum": 1.1646582616972672, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6885827273085342, "standard_deviation": 0.5268367398829972, "minimum": 0.25021326028557717, "maximum": 1.5545286377379857, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09420118453809866, "standard_deviation": 0.008091718077912211, "minimum": 0.08116477520597114, "maximum": 0.10790521831013608, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9771678069926184, "standard_deviation": 0.0031338028761476824, "minimum": 0.9721302316966874, "maximum": 0.9810372830977618, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8686373404404544, "standard_deviation": 0.02213553159411636, "minimum": 1.8410696735291694, "maximum": 1.8973828253828484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9330380811155692, "standard_deviation": 0.6057943986716652, "minimum": 0.42342373827385904, "maximum": 2.0262447676582314, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1823706850931182, "standard_deviation": 0.013388104871412808, "minimum": 0.16628319299722524, "maximum": 0.20259986622328133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922816054990938, "standard_deviation": 0.0018065527631499925, "minimum": 0.9887968562731905, "maximum": 0.9943938191740924, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2137229286626592, "standard_deviation": 0.04059523142224884, "minimum": 1.1475894413810959, "maximum": 1.2614290814632414, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0382568535005585, "standard_deviation": 0.6896481012013339, "minimum": 0.49127351672617037, "maximum": 2.1849446424915526, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10458452971732402, "standard_deviation": 0.012733281386451092, "minimum": 0.09190407447300428, "maximum": 0.13040304141569775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907566461678414, "standard_deviation": 0.0016980751997377017, "minimum": 0.9872216879836295, "maximum": 0.9930287765744954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8409134249301857, "standard_deviation": 0.03674747693971045, "minimum": 1.7823170298553193, "maximum": 1.9053881266533135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4887957655557561, "standard_deviation": 0.16731811849317996, "minimum": 0.33561470251191866, "maximum": 0.8020422134702841, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11441114539778807, "standard_deviation": 0.009379630849206991, "minimum": 0.10160450359438321, "maximum": 0.13349001462113946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996347679164644, "standard_deviation": 0.0009640935178314243, "minimum": 0.9942506663493951, "maximum": 0.997185838458315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2694710291364224, "standard_deviation": 0.03870895648858984, "minimum": 1.2260633646745793, "maximum": 1.3282751791232663, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6210920569363563, "standard_deviation": 0.3291429233969625, "minimum": 0.17976240531136167, "maximum": 1.2017335128688218, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07126922233310641, "standard_deviation": 0.007130142536145683, "minimum": 0.06490974607917388, "maximum": 0.08683333043955167, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1783406857285637, "standard_deviation": 0.05354753109001613, "minimum": 0.09739052611076156, "maximum": 0.27328363206695305, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.855567291580684, "standard_deviation": 0.006859024707011319, "minimum": 5.842346040394468, "maximum": 5.862952163280911, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2850474850142351, "standard_deviation": 0.012555103087329912, "minimum": 0.2708601433110817, "maximum": 0.30752215783372144, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0941306353917751, "standard_deviation": 0.041549831662531164, "minimum": -0.16532032170298983, "maximum": -0.035961962443355346, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945656227220082, "standard_deviation": 0.0009092926660955953, "minimum": 0.9933040928756874, "maximum": 0.9958523670757893, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8660018232531485, "standard_deviation": 0.030060274135469016, "minimum": 0.8323927827698706, "maximum": 0.9153911396180525, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3314778559230251, "standard_deviation": 0.17287656987414035, "minimum": 0.11694819905047614, "maximum": 0.5873766996234424, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07698863501575778, "standard_deviation": 0.006118505616137546, "minimum": 0.06907424598168621, "maximum": 0.0850249770728744, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982426113774396, "standard_deviation": 0.000513741051643975, "minimum": 0.9973873693487058, "maximum": 0.9987847271727502, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46465995337061605, "standard_deviation": 0.0321228678022463, "minimum": 0.4298843063548515, "maximum": 0.5199895699391114, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3829753964945284, "standard_deviation": 0.3028031231045776, "minimum": 0.07796195178416547, "maximum": 0.9220364874166811, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.042718988595925184, "standard_deviation": 0.005694975884065011, "minimum": 0.035872859267667236, "maximum": 0.05081509085136901, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0013", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.4262125492095947, "predicted_mos": 2.4262125492095947, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a9b314092bb43212797024a259ccd1edae72b265c6f7334a7f842841db1f6741", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.115829499904066, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8533356832148651, "全帯域_10ms_rms_envelope_nrmse": 0.07800570541767234, "全帯域_10ms_envelope_correlation": 0.99436941615412, "全帯域_10ms_modulation_mae_db": 0.24474216419599662, "全帯域_40ms_log_rms_mae_db": 0.4426708795874064, "全帯域_40ms_rms_envelope_nrmse": 0.039569432201331516, "全帯域_40ms_envelope_correlation": 0.9984033957680783, "全帯域_40ms_modulation_mae_db": 0.40523053053206487, "300–1000 Hz_10ms_log_rms_mae_db": 1.7794258725760732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10109559738987557, "300–1000 Hz_10ms_envelope_correlation": 0.9926855213611748, "300–1000 Hz_10ms_modulation_mae_db": 0.4000529078704063, "300–1000 Hz_40ms_log_rms_mae_db": 1.238851329125392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055656828841411696, "300–1000 Hz_40ms_envelope_correlation": 0.9976284441750767, "300–1000 Hz_40ms_modulation_mae_db": 0.6181575999940415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8408293940544838, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13769973839498353, "1000–3000 Hz_10ms_envelope_correlation": 0.9866943595914553, "1000–3000 Hz_10ms_modulation_mae_db": 0.46015002042372427, "1000–3000 Hz_40ms_log_rms_mae_db": 1.147661360258402, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08488227531208398, "1000–3000 Hz_40ms_envelope_correlation": 0.9946910736285581, "1000–3000 Hz_40ms_modulation_mae_db": 0.522846032871718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8083203169256772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1817680868879786, "3000–6000 Hz_10ms_envelope_correlation": 0.9768817296178972, "3000–6000 Hz_10ms_modulation_mae_db": 0.8494869629016356, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1337924521595106, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08836376316190778, "3000–6000 Hz_40ms_envelope_correlation": 0.9944008544675575, "3000–6000 Hz_40ms_modulation_mae_db": 1.496507306622983, "mr_stft_db": 5.841148925491354, "band_mae_db": 0.11434385457863022, "rms_gain_error_db": 0.0012208739391305313, "output_peak": 0.29028168147138667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8791707392045831, "全帯域_10ms_rms_envelope_nrmse": 0.07830570233768244, "全帯域_10ms_envelope_correlation": 0.9943641920948635, "全帯域_10ms_modulation_mae_db": 0.4391742330263558, "全帯域_40ms_log_rms_mae_db": 0.45958447010819314, "全帯域_40ms_rms_envelope_nrmse": 0.042943388867763437, "全帯域_40ms_envelope_correlation": 0.9982061776378157, "全帯域_40ms_modulation_mae_db": 0.4776901272007814, "300–1000 Hz_10ms_log_rms_mae_db": 1.8840646436892712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1077475122074497, "300–1000 Hz_10ms_envelope_correlation": 0.9919596458951111, "300–1000 Hz_10ms_modulation_mae_db": 0.8357437852908308, "300–1000 Hz_40ms_log_rms_mae_db": 1.3216654459434531, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06208070560778464, "300–1000 Hz_40ms_envelope_correlation": 0.9974486553273181, "300–1000 Hz_40ms_modulation_mae_db": 1.0417948089832028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8995395417473313, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1484422166092735, "1000–3000 Hz_10ms_envelope_correlation": 0.9845108754858751, "1000–3000 Hz_10ms_modulation_mae_db": 0.7376649096335943, "1000–3000 Hz_40ms_log_rms_mae_db": 1.164935522087993, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09321763269687001, "1000–3000 Hz_40ms_envelope_correlation": 0.9935481275750075, "1000–3000 Hz_40ms_modulation_mae_db": 0.5561070384540022, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8535461304868028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.174805450120539, "3000–6000 Hz_10ms_envelope_correlation": 0.9788718963879145, "3000–6000 Hz_10ms_modulation_mae_db": 0.21333018522596295, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1596191020049094, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09107202755212705, "3000–6000 Hz_40ms_envelope_correlation": 0.9939641488396043, "3000–6000 Hz_40ms_modulation_mae_db": 0.16169359845947318, "mr_stft_db": 5.845086608069003, "band_mae_db": 0.15104692004609635, "rms_gain_error_db": -0.08660129271356197, "output_peak": 0.285759236641749}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8534206781371693, "全帯域_10ms_rms_envelope_nrmse": 0.07250538977514133, "全帯域_10ms_envelope_correlation": 0.9951676394318817, "全帯域_10ms_modulation_mae_db": 0.2078791111517894, "全帯域_40ms_log_rms_mae_db": 0.43871543672812174, "全帯域_40ms_rms_envelope_nrmse": 0.04410234558329497, "全帯域_40ms_envelope_correlation": 0.9980802411909917, "全帯域_40ms_modulation_mae_db": 0.21145030405579787, "300–1000 Hz_10ms_log_rms_mae_db": 1.8498500880253257, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10948961415781117, "300–1000 Hz_10ms_envelope_correlation": 0.9915267788265493, "300–1000 Hz_10ms_modulation_mae_db": 0.42155179064178755, "300–1000 Hz_40ms_log_rms_mae_db": 1.3026633518176507, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07054364489815736, "300–1000 Hz_40ms_envelope_correlation": 0.9963608112083084, "300–1000 Hz_40ms_modulation_mae_db": 0.6323958198411744, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7549321051479954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13956596547604402, "1000–3000 Hz_10ms_envelope_correlation": 0.9864074006519253, "1000–3000 Hz_10ms_modulation_mae_db": 0.7817263827105227, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0714570670702441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08519540490832209, "1000–3000 Hz_40ms_envelope_correlation": 0.9948022558281414, "1000–3000 Hz_40ms_modulation_mae_db": 0.5226581573181753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.816964403322015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1663350115858417, "3000–6000 Hz_10ms_envelope_correlation": 0.98068161021812, "3000–6000 Hz_10ms_modulation_mae_db": 0.7811750681015068, "3000–6000 Hz_40ms_log_rms_mae_db": 1.184076959963622, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08799007835308745, "3000–6000 Hz_40ms_envelope_correlation": 0.9944571032669675, "3000–6000 Hz_40ms_modulation_mae_db": 1.4013740775471142, "mr_stft_db": 5.848756681417573, "band_mae_db": 0.16226974301642955, "rms_gain_error_db": -0.07597947881421167, "output_peak": 0.29029320053650814}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.857634273497273, "全帯域_10ms_rms_envelope_nrmse": 0.07635783977822681, "全帯域_10ms_envelope_correlation": 0.9946045237741524, "全帯域_10ms_modulation_mae_db": 0.5578910418883641, "全帯域_40ms_log_rms_mae_db": 0.46058966786066147, "全帯域_40ms_rms_envelope_nrmse": 0.0410150423715255, "全帯域_40ms_envelope_correlation": 0.9983097762945079, "全帯域_40ms_modulation_mae_db": 0.6543941867558791, "300–1000 Hz_10ms_log_rms_mae_db": 1.8466736351454098, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10821254840065458, "300–1000 Hz_10ms_envelope_correlation": 0.9916269113961081, "300–1000 Hz_10ms_modulation_mae_db": 0.31741360467764174, "300–1000 Hz_40ms_log_rms_mae_db": 1.263624554610445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06231407168408498, "300–1000 Hz_40ms_envelope_correlation": 0.9970494767472647, "300–1000 Hz_40ms_modulation_mae_db": 0.6524678187212393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8994368899681882, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15388700843911896, "1000–3000 Hz_10ms_envelope_correlation": 0.9833569877240145, "1000–3000 Hz_10ms_modulation_mae_db": 0.975162727674095, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1446378573487352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09568679306551077, "1000–3000 Hz_40ms_envelope_correlation": 0.993254426987289, "1000–3000 Hz_40ms_modulation_mae_db": 1.2292519742836931, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7730618533812694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16723474211535458, "3000–6000 Hz_10ms_envelope_correlation": 0.980536535545329, "3000–6000 Hz_10ms_modulation_mae_db": 1.0380274529453453, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1565284015039647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09260493908854074, "3000–6000 Hz_40ms_envelope_correlation": 0.9938244793914458, "3000–6000 Hz_40ms_modulation_mae_db": 1.1739343539975036, "mr_stft_db": 5.833111012689984, "band_mae_db": 0.12431810732536291, "rms_gain_error_db": -0.05087332788675555, "output_peak": 0.29697650201997855}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8509841855454335, "全帯域_10ms_rms_envelope_nrmse": 0.07701679144500057, "全帯域_10ms_envelope_correlation": 0.9945637085020204, "全帯域_10ms_modulation_mae_db": 0.3937156302960056, "全帯域_40ms_log_rms_mae_db": 0.4348649248882284, "全帯域_40ms_rms_envelope_nrmse": 0.03821236888614001, "全帯域_40ms_envelope_correlation": 0.9986101571844939, "全帯域_40ms_modulation_mae_db": 0.3593518160820304, "300–1000 Hz_10ms_log_rms_mae_db": 1.822275229093778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11173632095227524, "300–1000 Hz_10ms_envelope_correlation": 0.9911785408262735, "300–1000 Hz_10ms_modulation_mae_db": 0.5851541128516915, "300–1000 Hz_40ms_log_rms_mae_db": 1.268720849238929, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06230732816576226, "300–1000 Hz_40ms_envelope_correlation": 0.9972410573271344, "300–1000 Hz_40ms_modulation_mae_db": 0.6793356543794965, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8306216206860537, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15131769492979938, "1000–3000 Hz_10ms_envelope_correlation": 0.9840068657557053, "1000–3000 Hz_10ms_modulation_mae_db": 0.3225797180050707, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1000916757385355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0951170809830761, "1000–3000 Hz_40ms_envelope_correlation": 0.993520223832034, "1000–3000 Hz_40ms_modulation_mae_db": 0.267507976023324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.861348963564213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1731126771833753, "3000–6000 Hz_10ms_envelope_correlation": 0.9791622042005547, "3000–6000 Hz_10ms_modulation_mae_db": 0.6370187212060499, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2184873866178425, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09748377279803909, "3000–6000 Hz_40ms_envelope_correlation": 0.9930240151350991, "3000–6000 Hz_40ms_modulation_mae_db": 0.6875775088593007, "mr_stft_db": 5.834769353532619, "band_mae_db": 0.13338457952154797, "rms_gain_error_db": -0.087757694486092, "output_peak": 0.2795631511375579}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.837110618551252, "全帯域_10ms_rms_envelope_nrmse": 0.07017999158879858, "全帯域_10ms_envelope_correlation": 0.9954817410662564, "全帯域_10ms_modulation_mae_db": 0.4122855041095059, "全帯域_40ms_log_rms_mae_db": 0.4335148004051669, "全帯域_40ms_rms_envelope_nrmse": 0.04230266106339551, "全帯域_40ms_envelope_correlation": 0.9982414328975622, "全帯域_40ms_modulation_mae_db": 0.5150135590602556, "300–1000 Hz_10ms_log_rms_mae_db": 1.8044266345936955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10540882287852912, "300–1000 Hz_10ms_envelope_correlation": 0.9921119371129261, "300–1000 Hz_10ms_modulation_mae_db": 0.5149966469700787, "300–1000 Hz_40ms_log_rms_mae_db": 1.2472613841679387, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06305633413406984, "300–1000 Hz_40ms_envelope_correlation": 0.9970541159009867, "300–1000 Hz_40ms_modulation_mae_db": 0.7209152471906197, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8345794106833195, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14575162108566075, "1000–3000 Hz_10ms_envelope_correlation": 0.9852022826002449, "1000–3000 Hz_10ms_modulation_mae_db": 0.3379103348954895, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1232255239732423, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08679142266041659, "1000–3000 Hz_40ms_envelope_correlation": 0.994678187865699, "1000–3000 Hz_40ms_modulation_mae_db": 0.3146375455968733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8498285466887863, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16495565916775998, "3000–6000 Hz_10ms_envelope_correlation": 0.981313878189544, "3000–6000 Hz_10ms_modulation_mae_db": 0.16003224186982268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.188640086600498, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08565211370954384, "3000–6000 Hz_40ms_envelope_correlation": 0.9947350091812375, "3000–6000 Hz_40ms_modulation_mae_db": 0.04804725819449552, "mr_stft_db": 5.858077051864427, "band_mae_db": 0.10786166643449109, "rms_gain_error_db": -0.07836769473245667, "output_peak": 0.2923465159843636}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.851948671315241, "全帯域_10ms_rms_envelope_nrmse": 0.071572416051097, "全帯域_10ms_envelope_correlation": 0.9952622232696672, "全帯域_10ms_modulation_mae_db": 0.28808570147057705, "全帯域_40ms_log_rms_mae_db": 0.47640412812845173, "全帯域_40ms_rms_envelope_nrmse": 0.03783352527633977, "全帯域_40ms_envelope_correlation": 0.9985654739905004, "全帯域_40ms_modulation_mae_db": 0.34584843983666075, "300–1000 Hz_10ms_log_rms_mae_db": 1.8662836331091823, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11198054583924451, "300–1000 Hz_10ms_envelope_correlation": 0.9910647450187587, "300–1000 Hz_10ms_modulation_mae_db": 0.276805180914213, "300–1000 Hz_40ms_log_rms_mae_db": 1.2089297693962495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060572008905836305, "300–1000 Hz_40ms_envelope_correlation": 0.9973001398016569, "300–1000 Hz_40ms_modulation_mae_db": 0.6265097721171607, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8828045337089014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15803835637587313, "1000–3000 Hz_10ms_envelope_correlation": 0.9826478743727385, "1000–3000 Hz_10ms_modulation_mae_db": 0.5576798227246199, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1663681298763053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1050738420038068, "1000–3000 Hz_40ms_envelope_correlation": 0.9918643533997291, "1000–3000 Hz_40ms_modulation_mae_db": 0.6274645422899784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8520030135895253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18846555830733522, "3000–6000 Hz_10ms_envelope_correlation": 0.9751316030683986, "3000–6000 Hz_10ms_modulation_mae_db": 0.1509781312688158, "3000–6000 Hz_40ms_log_rms_mae_db": 1.203479520940519, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10844628963878324, "3000–6000 Hz_40ms_envelope_correlation": 0.991466441820678, "3000–6000 Hz_40ms_modulation_mae_db": 0.33175836060119673, "mr_stft_db": 5.856200910797251, "band_mae_db": 0.1565665205789675, "rms_gain_error_db": -0.04802607606691655, "output_peak": 0.2934429961755648}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8504341443400819, "全帯域_10ms_rms_envelope_nrmse": 0.08248837196527237, "全帯域_10ms_envelope_correlation": 0.9938242141203151, "全帯域_10ms_modulation_mae_db": 0.5557144803587377, "全帯域_40ms_log_rms_mae_db": 0.47606028313577914, "全帯域_40ms_rms_envelope_nrmse": 0.05185161049965327, "全帯域_40ms_envelope_correlation": 0.997456084245869, "全帯域_40ms_modulation_mae_db": 0.6170423165237221, "300–1000 Hz_10ms_log_rms_mae_db": 1.8849507928721407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1154958204138038, "300–1000 Hz_10ms_envelope_correlation": 0.9906441655739717, "300–1000 Hz_10ms_modulation_mae_db": 0.7743187190322095, "300–1000 Hz_40ms_log_rms_mae_db": 1.3501912602238126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07671300013163275, "300–1000 Hz_40ms_envelope_correlation": 0.995782783190512, "300–1000 Hz_40ms_modulation_mae_db": 1.050759223358106, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8947760925914534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14282956038731498, "1000–3000 Hz_10ms_envelope_correlation": 0.9858076790869751, "1000–3000 Hz_10ms_modulation_mae_db": 0.7241029269585025, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1453120029161488, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09012450424622945, "1000–3000 Hz_40ms_envelope_correlation": 0.9942377341206498, "1000–3000 Hz_40ms_modulation_mae_db": 0.8232056886406071, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8491982492103847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18211701299072275, "3000–6000 Hz_10ms_envelope_correlation": 0.9778452806368059, "3000–6000 Hz_10ms_modulation_mae_db": 0.5038609999868752, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1935600554892927, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09940741860591634, "3000–6000 Hz_40ms_envelope_correlation": 0.9932732696212987, "3000–6000 Hz_40ms_modulation_mae_db": 0.8623952695707015, "mr_stft_db": 5.844399268461512, "band_mae_db": 0.150669231481875, "rms_gain_error_db": -0.13358513252240553, "output_peak": 0.28540623058662795}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6589889880399005, "spectral_convergence": 0.24356624010201403, "spectral_convergence_status": "ok", "amplitude_floor": 7.181372005990482e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.523750524798566, "spectral_convergence": 0.23960908447793514, "spectral_convergence_status": "ok", "amplitude_floor": 5.417622086660404e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.340707263635596, "spectral_convergence": 0.24475803095942605, "spectral_convergence_status": "ok", "amplitude_floor": 3.166881446630363e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001077391554152542, "output_power": 0.0001142669531077703, "delta_db": 0.2554708049148051, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002614688899274717, "output_power": 0.00025949625536938266, "delta_db": -0.032889278889975684, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015854777413607497, "output_power": 0.00015616925921263836, "delta_db": -0.06564599082229147, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.449556222199797e-05, "output_power": 6.318661194734728e-05, "delta_db": -0.08904763737585851, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5950148027800597e-05, "output_power": 1.517833952765461e-05, "delta_db": -0.21540454519616803, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.2322545231279567e-05, "output_power": 1.2401119942725384e-05, "delta_db": 0.02760487027268259, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11434385457863022, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.0012208739391305313, "pitch": null, "output_rms_dbfs": -32.07119276780367, "output_peak": 0.29028168147138667, "reference_rms_dbfs": -32.072413641742806, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.841148925491354}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001113800968506657, "log_rms_mae_db": 0.8533356832148651, "rms_envelope_nrmse": 0.07800570541767234, "envelope_correlation": 0.99436941615412, "rms_gain_error_db": 0.001196829098037169, "reference_mean_power": 0.0006210401213146882, "output_mean_power": 0.0006212112911879084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.9949527250696107e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014723003466329858, "output_power": 0.00014564923976329356, "reference_active": true, "delta_db": -0.04688192227161089}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0749733198322775e-05, "output_power": 3.112877942051506e-05, "reference_active": true, "delta_db": 0.05320740101215647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6784890109842875e-06, "output_power": 4.9070340632740475e-06, "reference_active": true, "delta_db": -0.6341371693042225}], "active_band_count": 3, "active_band_mae_db": 0.24474216419599662}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.537149969198292e-05, "log_rms_mae_db": 0.4426708795874064, "rms_envelope_nrmse": 0.039569432201331516, "envelope_correlation": 0.9984033957680783, "rms_gain_error_db": 0.0012174955806271792, "reference_mean_power": 0.0006241534506116662, "output_mean_power": 0.0006243284495178558}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.688739338888854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013414910111714758, "output_power": 0.00013308572599899222, "reference_active": true, "delta_db": -0.03456288901180716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.37620762285614e-05, "output_power": 1.4212169026947755e-05, "reference_active": true, "delta_db": 0.13976404874366513}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.149987894354722e-07, "output_power": 3.2652043355023776e-07, "reference_active": true, "delta_db": -1.0413646538407222}], "active_band_count": 3, "active_band_mae_db": 0.40523053053206487}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.440030149355929e-05, "log_rms_mae_db": 1.7794258725760732, "rms_envelope_nrmse": 0.10109559738987557, "envelope_correlation": 0.9926855213611748, "rms_gain_error_db": -0.05551781283688696, "reference_mean_power": 0.000315142987429675, "output_mean_power": 0.0003111400135959289}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.94905977338963e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010605229421605894, "output_power": 0.00010447637176610349, "reference_active": true, "delta_db": -0.06501985888928294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.109806584151368e-05, "output_power": 2.1376682568928147e-05, "reference_active": true, "delta_db": 0.056976649585327103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3612767263836796e-06, "output_power": 2.622334247932048e-06, "reference_active": true, "delta_db": -1.078162215136609}], "active_band_count": 3, "active_band_mae_db": 0.4000529078704063}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.828793840107377e-05, "log_rms_mae_db": 1.238851329125392, "rms_envelope_nrmse": 0.055656828841411696, "envelope_correlation": 0.9976284441750767, "rms_gain_error_db": -0.05550811510026347, "reference_mean_power": 0.00031672629889435445, "output_mean_power": 0.00031270391196247776}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7988521248624547e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.967318213226487e-05, "output_power": 9.790635634636594e-05, "reference_active": true, "delta_db": -0.07767435325998434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.491202888187961e-06, "output_power": 9.759491346882314e-06, "reference_active": true, "delta_db": 0.12105926174264002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5805315595038687e-07, "output_power": 2.445533337722387e-07, "reference_active": true, "delta_db": -1.6557391849795}], "active_band_count": 3, "active_band_mae_db": 0.6181575999940415}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.58934180861454e-05, "log_rms_mae_db": 1.8408293940544838, "rms_envelope_nrmse": 0.13769973839498353, "envelope_correlation": 0.9866943595914553, "rms_gain_error_db": -0.12455401564593252, "reference_mean_power": 6.36645119573909e-05, "output_mean_power": 6.186457199273416e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.512623526912829e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3991998554391248e-05, "output_power": 1.2814974329203867e-05, "reference_active": true, "delta_db": -0.38162011052833616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3957717792014314e-06, "output_power": 2.2766378759844016e-06, "reference_active": true, "delta_db": -0.22151488024033889}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.685934249147775e-07, "output_power": 6.426361584465385e-07, "reference_active": true, "delta_db": -0.7773150705024976}], "active_band_count": 3, "active_band_mae_db": 0.46015002042372427}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6382034637347546e-05, "log_rms_mae_db": 1.147661360258402, "rms_envelope_nrmse": 0.08488227531208398, "envelope_correlation": 0.9946910736285581, "rms_gain_error_db": -0.1244869600748643, "reference_mean_power": 6.398182095924481e-05, "output_mean_power": 6.21738699146858e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.275432891493173e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2825234350232933e-05, "output_power": 1.1765777985365971e-05, "reference_active": true, "delta_db": -0.3744466034237415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.068139769619595e-06, "output_power": 1.0217288781234204e-06, "reference_active": true, "delta_db": -0.19292416834959397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.783047579212978e-08, "output_power": 4.5924034309571695e-08, "reference_active": true, "delta_db": -1.0011673268418186}], "active_band_count": 3, "active_band_mae_db": 0.522846032871718}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.0860899025143115e-05, "log_rms_mae_db": 1.8083203169256772, "rms_envelope_nrmse": 0.1817680868879786, "envelope_correlation": 0.9768817296178972, "rms_gain_error_db": -0.12715127680254956, "reference_mean_power": 1.1507041647425964e-05, "output_mean_power": 1.117502638638874e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.593308449162533e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.586447862335724e-06, "output_power": 4.332809070792088e-06, "reference_active": true, "delta_db": -0.24706909042677053}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.405946680666416e-07, "output_power": 7.604648310181038e-07, "reference_active": true, "delta_db": -0.43507496756065905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329768801508531e-07, "output_power": 2.1666182488176337e-07, "reference_active": true, "delta_db": -1.8663168307174773}], "active_band_count": 3, "active_band_mae_db": 0.8494869629016356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4734203651083813e-05, "log_rms_mae_db": 1.1337924521595106, "rms_envelope_nrmse": 0.08836376316190778, "envelope_correlation": 0.9944008544675575, "rms_gain_error_db": -0.1270092887044414, "reference_mean_power": 1.156194575111937e-05, "output_mean_power": 1.1228713434822673e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.1073798497935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.414350374295127e-06, "output_power": 4.163619045943377e-06, "reference_active": true, "delta_db": -0.2539581437216971}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649548666103306e-07, "output_power": 3.881090152549511e-07, "reference_active": true, "delta_db": -0.7845706682541752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1590180837860948e-08, "output_power": 9.753420966868743e-09, "reference_active": true, "delta_db": -3.450993107893077}], "active_band_count": 3, "active_band_mae_db": 1.496507306622983}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8998834914638822, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "59c8a907fb67a3ede9945f56917812cb6ffa8ae13c69791b4068656d90a6f1df", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9848292906586168, "standard_deviation": 0.0014525496828080115, "minimum": 0.9826478743727385, "maximum": 0.9866943595914553, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8546899485734658, "standard_deviation": 0.050055061182110526, "minimum": 1.7549321051479954, "maximum": 1.8995395417473313, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6121221053782024, "standard_deviation": 0.23113112244273024, "minimum": 0.3225797180050707, "maximum": 0.975162727674095, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14719152021225854, "standard_deviation": 0.007075268664706642, "minimum": 0.13769973839498353, "maximum": 0.15803835637587313, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938245479046385, "standard_deviation": 0.0009958933780704505, "minimum": 0.9918643533997291, "maximum": 0.9948022558281414, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1329611424087007, "standard_deviation": 0.032953016311475625, "minimum": 1.0714570670702441, "maximum": 1.1663681298763053, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6079598694347965, "standard_deviation": 0.3051627236247931, "minimum": 0.267507976023324, "maximum": 1.2292519742836931, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09201111948453947, "standard_deviation": 0.006796289988599254, "minimum": 0.08488227531208398, "maximum": 0.1050738420038068, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9788030922330705, "standard_deviation": 0.0021064617290726478, "minimum": 0.9751316030683986, "maximum": 0.981313878189544, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8330339346460842, "standard_deviation": 0.030685744304953046, "minimum": 1.7730618533812694, "maximum": 1.861348963564213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5417387204382518, "standard_deviation": 0.3412372637952868, "minimum": 0.1509781312688158, "maximum": 1.0380274529453453, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1748492747948634, "standard_deviation": 0.008597908914534151, "minimum": 0.16495565916775998, "maximum": 0.18846555830733522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993643165215486, "standard_deviation": 0.0010571464979844106, "minimum": 0.991466441820678, "maximum": 0.9947350091812375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1797729956600198, "standard_deviation": 0.0277970625500066, "minimum": 1.1337924521595106, "maximum": 1.2184873866178425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7704109667315959, "standard_deviation": 0.5591123360813864, "minimum": 0.04804725819449552, "maximum": 1.496507306622983, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09387755036349318, "standard_deviation": 0.00754079314628493, "minimum": 0.08565211370954384, "maximum": 0.10844628963878324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915997807513591, "standard_deviation": 0.0006495725805693929, "minimum": 0.9906441655739717, "maximum": 0.9926855213611748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8422438161381096, "standard_deviation": 0.03780256353582548, "minimum": 1.7794258725760732, "maximum": 1.8849507928721407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5157545935311074, "standard_deviation": 0.20450720399129826, "minimum": 0.276805180914213, "maximum": 0.8357437852908308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10889584777995545, "standard_deviation": 0.00441239599114347, "minimum": 0.10109559738987557, "maximum": 0.1154958204138038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969831854597823, "standard_deviation": 0.0006138543267519089, "minimum": 0.995782783190512, "maximum": 0.9976284441750767, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2752384930654839, "standard_deviation": 0.046608485646467065, "minimum": 1.2089297693962495, "maximum": 1.3501912602238126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7527919930731302, "standard_deviation": 0.1841400306318217, "minimum": 0.6181575999940415, "maximum": 1.050759223358106, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06415549029609247, "standard_deviation": 0.006501511466257517, "minimum": 0.055656828841411696, "maximum": 0.07671300013163275, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13755757787292508, "standard_deviation": 0.02049156768203236, "minimum": 0.10786166643449109, "maximum": 0.16226974301642955, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.845193726540465, "standard_deviation": 0.009035607373485863, "minimum": 5.833111012689984, "maximum": 5.858077051864427, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.28925868931921705, "standard_deviation": 0.005474819300714818, "minimum": 0.2795631511375579, "maximum": 0.29697650201997855, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06999622791040867, "standard_deviation": 0.0390254054305078, "minimum": -0.13358513252240553, "maximum": 0.0012208739391305313, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947047073016595, "standard_deviation": 0.0005558093308544895, "minimum": 0.9938242141203151, "maximum": 0.9954817410662564, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8542548742257374, "standard_deviation": 0.011700432947473666, "minimum": 0.837110618551252, "maximum": 0.8791707392045831, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38743598331216655, "standard_deviation": 0.13265370252332853, "minimum": 0.2078791111517894, "maximum": 0.5578910418883641, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07580402604486143, "standard_deviation": 0.004105415486676434, "minimum": 0.07017999158879858, "maximum": 0.08248837196527237, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982340924012274, "standard_deviation": 0.0003616292818124951, "minimum": 0.997456084245869, "maximum": 0.9986101571844939, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45280057385525113, "standard_deviation": 0.017725943211755593, "minimum": 0.4335148004051669, "maximum": 0.47640412812845173, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44825266000589903, "standard_deviation": 0.14765989061858184, "minimum": 0.21145030405579787, "maximum": 0.6543941867558791, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0422287968436805, "standard_deviation": 0.0044820331982190365, "minimum": 0.03783352527633977, "maximum": 0.05185161049965327, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5639-40744-0023", "language": "en", "style": "normal", "speaker": "librispeech_5639", "source_utterance_id": "5639-40744-0023", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5639-40744-0023.wav", "source_file_sha256": "543345a0df507df4346809e1b02820148ad6a4aadc077647dcb4019bf986b6c7", "decoded_pcm_sha256": "da6b45c38da036a67d46d62fc3e6e54310d46e3e14d8f0a3eff1b445bb1755f4", "original_source_sha256": "dd277d4b42ef738b14b59fa1d835988e71ec031e2f01d024279202edc26ad9c0", "corpus_entry_id": "librispeech_5639-40744-0023"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.042726039886475, "predicted_mos": 4.042726039886475, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "689f52cca314c6a0cfca585009a65fc6c0481fff0b5471b87373963fdac9d14f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.6476064449998375, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5639-40744-0023", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "689f52cca314c6a0cfca585009a65fc6c0481fff0b5471b87373963fdac9d14f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "7c6399830af60d82e6872b227aa2254101ab88790d9e2ca6eee36fdb0cf762a8"}, "rows": [{"case": "librispeech_5639-40744-0023", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.3031084537506104, "predicted_mos": 3.3031084537506104, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "238a4a10f21dbc78962dfc542c8a75a7f55916c22b5b6437fd471cc2c05ed861", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.189313454000512, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5639-40744-0023", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "238a4a10f21dbc78962dfc542c8a75a7f55916c22b5b6437fd471cc2c05ed861", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7558173264050408, "全帯域_10ms_rms_envelope_nrmse": 0.06280713538890483, "全帯域_10ms_envelope_correlation": 0.996161306139302, "全帯域_10ms_modulation_mae_db": 0.5071885225248, "全帯域_40ms_log_rms_mae_db": 0.4447718461755786, "全帯域_40ms_rms_envelope_nrmse": 0.03636160735744728, "全帯域_40ms_envelope_correlation": 0.9987201176071835, "全帯域_40ms_modulation_mae_db": 0.59363590419178, "300–1000 Hz_10ms_log_rms_mae_db": 1.2973172098207524, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08488629503007973, "300–1000 Hz_10ms_envelope_correlation": 0.9944936758807649, "300–1000 Hz_10ms_modulation_mae_db": 0.8225750408012309, "300–1000 Hz_40ms_log_rms_mae_db": 0.7869936240529904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05243241034565682, "300–1000 Hz_40ms_envelope_correlation": 0.9979002222924428, "300–1000 Hz_40ms_modulation_mae_db": 1.0679895239747217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6433393348645917, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1601830737760231, "1000–3000 Hz_10ms_envelope_correlation": 0.9823668983239033, "1000–3000 Hz_10ms_modulation_mae_db": 1.1056747440536956, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9467926406915637, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09836286261737234, "1000–3000 Hz_40ms_envelope_correlation": 0.9941770113492783, "1000–3000 Hz_40ms_modulation_mae_db": 1.0828997468117978, "3000–6000 Hz_10ms_log_rms_mae_db": 1.711732217952592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20456900498869002, "3000–6000 Hz_10ms_envelope_correlation": 0.967303767834235, "3000–6000 Hz_10ms_modulation_mae_db": 0.7362170191056436, "3000–6000 Hz_40ms_log_rms_mae_db": 1.13568306956851, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12175123837106262, "3000–6000 Hz_40ms_envelope_correlation": 0.9885358611248347, "3000–6000 Hz_40ms_modulation_mae_db": 0.6151975328529233, "mr_stft_db": 5.422270956627546, "band_mae_db": 0.22252352612249404, "rms_gain_error_db": -0.1009558130568691, "output_peak": 0.24339632315626616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7893683641553675, "全帯域_10ms_rms_envelope_nrmse": 0.06911992451450429, "全帯域_10ms_envelope_correlation": 0.9955096413472427, "全帯域_10ms_modulation_mae_db": 0.47783319403370655, "全帯域_40ms_log_rms_mae_db": 0.44679395498614566, "全帯域_40ms_rms_envelope_nrmse": 0.04007944529410721, "全帯域_40ms_envelope_correlation": 0.9986393854280675, "全帯域_40ms_modulation_mae_db": 0.511914576817008, "300–1000 Hz_10ms_log_rms_mae_db": 1.442015852890291, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08986514748571497, "300–1000 Hz_10ms_envelope_correlation": 0.9940688864529936, "300–1000 Hz_10ms_modulation_mae_db": 0.6405956201744475, "300–1000 Hz_40ms_log_rms_mae_db": 0.9025597528487651, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054154526972453744, "300–1000 Hz_40ms_envelope_correlation": 0.998073126001829, "300–1000 Hz_40ms_modulation_mae_db": 0.8122882445371261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.605378457233578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17021881093030283, "1000–3000 Hz_10ms_envelope_correlation": 0.9802865444840363, "1000–3000 Hz_10ms_modulation_mae_db": 0.6523579002320294, "1000–3000 Hz_40ms_log_rms_mae_db": 0.987395119576272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11277075571662981, "1000–3000 Hz_40ms_envelope_correlation": 0.9923042671802549, "1000–3000 Hz_40ms_modulation_mae_db": 0.6383125659155877, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6511238247074191, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18659747222909331, "3000–6000 Hz_10ms_envelope_correlation": 0.9726907551136192, "3000–6000 Hz_10ms_modulation_mae_db": 0.5028144293189671, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1165091947597685, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11228960997704635, "3000–6000 Hz_40ms_envelope_correlation": 0.9898126642058225, "3000–6000 Hz_40ms_modulation_mae_db": 0.6126500558199907, "mr_stft_db": 5.427387192231102, "band_mae_db": 0.24535647414294084, "rms_gain_error_db": -0.16401982462297732, "output_peak": 0.2534264087041444}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7676751514586934, "全帯域_10ms_rms_envelope_nrmse": 0.06424908021763673, "全帯域_10ms_envelope_correlation": 0.9959994295979145, "全帯域_10ms_modulation_mae_db": 0.34297644711654024, "全帯域_40ms_log_rms_mae_db": 0.41876486433268, "全帯域_40ms_rms_envelope_nrmse": 0.032574190177410686, "全帯域_40ms_envelope_correlation": 0.9990273462906268, "全帯域_40ms_modulation_mae_db": 0.5626270591756362, "300–1000 Hz_10ms_log_rms_mae_db": 1.4222272457765412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08731905656017555, "300–1000 Hz_10ms_envelope_correlation": 0.9942281777265679, "300–1000 Hz_10ms_modulation_mae_db": 0.3961427567783092, "300–1000 Hz_40ms_log_rms_mae_db": 0.7989378873943687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049159339161234154, "300–1000 Hz_40ms_envelope_correlation": 0.9982659843936598, "300–1000 Hz_40ms_modulation_mae_db": 0.8364591805149496, "1000–3000 Hz_10ms_log_rms_mae_db": 1.617641686568826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16002492039607108, "1000–3000 Hz_10ms_envelope_correlation": 0.9817129058825194, "1000–3000 Hz_10ms_modulation_mae_db": 0.48314363457371656, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9991192291821304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09423941240278018, "1000–3000 Hz_40ms_envelope_correlation": 0.9938682965851093, "1000–3000 Hz_40ms_modulation_mae_db": 0.5893710896821042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.648282009786356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1929952597680323, "3000–6000 Hz_10ms_envelope_correlation": 0.9709680056626881, "3000–6000 Hz_10ms_modulation_mae_db": 0.933423892087417, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0828900198748366, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11189615588440815, "3000–6000 Hz_40ms_envelope_correlation": 0.9904516563618625, "3000–6000 Hz_40ms_modulation_mae_db": 0.8098938820986761, "mr_stft_db": 5.432332441077441, "band_mae_db": 0.17461883619965982, "rms_gain_error_db": -0.10844519270722837, "output_peak": 0.2340298596620907}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7690947837728488, "全帯域_10ms_rms_envelope_nrmse": 0.07118847073526893, "全帯域_10ms_envelope_correlation": 0.9951015968576072, "全帯域_10ms_modulation_mae_db": 0.3426016457646985, "全帯域_40ms_log_rms_mae_db": 0.4557293119367755, "全帯域_40ms_rms_envelope_nrmse": 0.040992824885192555, "全帯域_40ms_envelope_correlation": 0.9984231008816814, "全帯域_40ms_modulation_mae_db": 0.23988890081095302, "300–1000 Hz_10ms_log_rms_mae_db": 1.4139622923734123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09701276821770391, "300–1000 Hz_10ms_envelope_correlation": 0.9928809732285198, "300–1000 Hz_10ms_modulation_mae_db": 0.5437691126038086, "300–1000 Hz_40ms_log_rms_mae_db": 0.8891518426609769, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058800050366239684, "300–1000 Hz_40ms_envelope_correlation": 0.9974839215875873, "300–1000 Hz_40ms_modulation_mae_db": 0.7103879286079909, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6420714676871522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17254667866105425, "1000–3000 Hz_10ms_envelope_correlation": 0.9787190547547266, "1000–3000 Hz_10ms_modulation_mae_db": 0.8469553138974065, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9857774897167194, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10257891731397463, "1000–3000 Hz_40ms_envelope_correlation": 0.992739927810924, "1000–3000 Hz_40ms_modulation_mae_db": 0.8685076001513746, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6641485044738544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18796486839000512, "3000–6000 Hz_10ms_envelope_correlation": 0.972691338046206, "3000–6000 Hz_10ms_modulation_mae_db": 0.2157544044165057, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1141524464874042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11379406843985643, "3000–6000 Hz_40ms_envelope_correlation": 0.9893460974985062, "3000–6000 Hz_40ms_modulation_mae_db": 0.11963156853667467, "mr_stft_db": 5.433634195594588, "band_mae_db": 0.1882421615148091, "rms_gain_error_db": -0.13004865438077867, "output_peak": 0.24814040567525714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7731915265934767, "全帯域_10ms_rms_envelope_nrmse": 0.06612259914993301, "全帯域_10ms_envelope_correlation": 0.9956789646453156, "全帯域_10ms_modulation_mae_db": 0.1689012727874157, "全帯域_40ms_log_rms_mae_db": 0.4385456634302921, "全帯域_40ms_rms_envelope_nrmse": 0.03697963063392366, "全帯域_40ms_envelope_correlation": 0.9986059467064926, "全帯域_40ms_modulation_mae_db": 0.33057420808664406, "300–1000 Hz_10ms_log_rms_mae_db": 1.4189310263903703, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09476088779175566, "300–1000 Hz_10ms_envelope_correlation": 0.9930896218598972, "300–1000 Hz_10ms_modulation_mae_db": 0.3043104964728305, "300–1000 Hz_40ms_log_rms_mae_db": 0.8748309037886025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05659554471472545, "300–1000 Hz_40ms_envelope_correlation": 0.9975263876859398, "300–1000 Hz_40ms_modulation_mae_db": 0.35963906666406187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6527219080370468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17861458894883, "1000–3000 Hz_10ms_envelope_correlation": 0.9768332252231013, "1000–3000 Hz_10ms_modulation_mae_db": 0.6005351450793297, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9996768379270112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10441634076588728, "1000–3000 Hz_40ms_envelope_correlation": 0.9917846853589565, "1000–3000 Hz_40ms_modulation_mae_db": 0.5492148149374727, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6426994480286308, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17723848408015505, "3000–6000 Hz_10ms_envelope_correlation": 0.9754443891378201, "3000–6000 Hz_10ms_modulation_mae_db": 1.115393264014276, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0652393960512476, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10763018704367223, "3000–6000 Hz_40ms_envelope_correlation": 0.9907768751647628, "3000–6000 Hz_40ms_modulation_mae_db": 0.8117400366525503, "mr_stft_db": 5.435337366353331, "band_mae_db": 0.16600963670784843, "rms_gain_error_db": -0.07200307332693655, "output_peak": 0.23107996461918925}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7647631296039783, "全帯域_10ms_rms_envelope_nrmse": 0.0705754617306768, "全帯域_10ms_envelope_correlation": 0.9952786104290958, "全帯域_10ms_modulation_mae_db": 0.4096916339716701, "全帯域_40ms_log_rms_mae_db": 0.4756486060229394, "全帯域_40ms_rms_envelope_nrmse": 0.04164872636217895, "全帯域_40ms_envelope_correlation": 0.9984693115594022, "全帯域_40ms_modulation_mae_db": 0.5843694807833589, "300–1000 Hz_10ms_log_rms_mae_db": 1.4280616682769751, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09839077906757247, "300–1000 Hz_10ms_envelope_correlation": 0.9927483248333769, "300–1000 Hz_10ms_modulation_mae_db": 0.46165816065839566, "300–1000 Hz_40ms_log_rms_mae_db": 0.8737015885203696, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05888345902761794, "300–1000 Hz_40ms_envelope_correlation": 0.9975764073392672, "300–1000 Hz_40ms_modulation_mae_db": 0.7661794547627031, "1000–3000 Hz_10ms_log_rms_mae_db": 1.640032747804643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16879620658831937, "1000–3000 Hz_10ms_envelope_correlation": 0.9794035004071514, "1000–3000 Hz_10ms_modulation_mae_db": 0.6558111767515986, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9761453713188504, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09964021520791316, "1000–3000 Hz_40ms_envelope_correlation": 0.9927827019913326, "1000–3000 Hz_40ms_modulation_mae_db": 0.33480912752820674, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6230684666197859, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18871443163572035, "3000–6000 Hz_10ms_envelope_correlation": 0.9719611906967558, "3000–6000 Hz_10ms_modulation_mae_db": 1.144209929345317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0744602654052025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1108714888678491, "3000–6000 Hz_40ms_envelope_correlation": 0.9897144602986593, "3000–6000 Hz_40ms_modulation_mae_db": 0.9109074870987185, "mr_stft_db": 5.434933328774673, "band_mae_db": 0.17710748982262856, "rms_gain_error_db": -0.15659456300616104, "output_peak": 0.24353699610738377}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7588089149995968, "全帯域_10ms_rms_envelope_nrmse": 0.06531502658694772, "全帯域_10ms_envelope_correlation": 0.9958134542190085, "全帯域_10ms_modulation_mae_db": 0.35021027360851703, "全帯域_40ms_log_rms_mae_db": 0.43000999861524897, "全帯域_40ms_rms_envelope_nrmse": 0.03640627761402024, "全帯域_40ms_envelope_correlation": 0.9986875746995072, "全帯域_40ms_modulation_mae_db": 0.5136278008101244, "300–1000 Hz_10ms_log_rms_mae_db": 1.3898811576788945, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08839354829620541, "300–1000 Hz_10ms_envelope_correlation": 0.9940372209574251, "300–1000 Hz_10ms_modulation_mae_db": 0.1650086648933422, "300–1000 Hz_40ms_log_rms_mae_db": 0.8227881951277766, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052565926782662094, "300–1000 Hz_40ms_envelope_correlation": 0.9979237979896111, "300–1000 Hz_40ms_modulation_mae_db": 0.37997846035223426, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6379302173169932, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16638753289921865, "1000–3000 Hz_10ms_envelope_correlation": 0.9804112121045985, "1000–3000 Hz_10ms_modulation_mae_db": 1.1230096136663883, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9907388765158017, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10344133604589698, "1000–3000 Hz_40ms_envelope_correlation": 0.9928128792974391, "1000–3000 Hz_40ms_modulation_mae_db": 0.9768554076814181, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6295722315543444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18710387835387654, "3000–6000 Hz_10ms_envelope_correlation": 0.9725397932038263, "3000–6000 Hz_10ms_modulation_mae_db": 1.029033060697505, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0702697596241435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11115703843582063, "3000–6000 Hz_40ms_envelope_correlation": 0.9901281955086827, "3000–6000 Hz_40ms_modulation_mae_db": 0.6077653444453818, "mr_stft_db": 5.422911635784524, "band_mae_db": 0.18398484789391087, "rms_gain_error_db": -0.08798486486642423, "output_peak": 0.23436922106323735}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7369735531054159, "全帯域_10ms_rms_envelope_nrmse": 0.06478544921936479, "全帯域_10ms_envelope_correlation": 0.9958906980165028, "全帯域_10ms_modulation_mae_db": 0.10237162128914039, "全帯域_40ms_log_rms_mae_db": 0.4061633743461146, "全帯域_40ms_rms_envelope_nrmse": 0.03495839242908276, "全帯域_40ms_envelope_correlation": 0.9988061924855967, "全帯域_40ms_modulation_mae_db": 0.19250838881295093, "300–1000 Hz_10ms_log_rms_mae_db": 1.3303010416739738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08673834446478801, "300–1000 Hz_10ms_envelope_correlation": 0.9943039151245046, "300–1000 Hz_10ms_modulation_mae_db": 0.39826469181159113, "300–1000 Hz_40ms_log_rms_mae_db": 0.7924131154988131, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04649936715411478, "300–1000 Hz_40ms_envelope_correlation": 0.9984699072041654, "300–1000 Hz_40ms_modulation_mae_db": 0.7347633374693018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6075130506620994, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16698741225624547, "1000–3000 Hz_10ms_envelope_correlation": 0.9801374885475227, "1000–3000 Hz_10ms_modulation_mae_db": 1.244793793692419, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9383692936839154, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10227657774374532, "1000–3000 Hz_40ms_envelope_correlation": 0.9928049439442764, "1000–3000 Hz_40ms_modulation_mae_db": 1.393965353341373, "3000–6000 Hz_10ms_log_rms_mae_db": 1.618343068911068, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19632209195752034, "3000–6000 Hz_10ms_envelope_correlation": 0.9696635272822987, "3000–6000 Hz_10ms_modulation_mae_db": 0.4076718642519546, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0445353580183587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11416916929462753, "3000–6000 Hz_40ms_envelope_correlation": 0.9891637235978308, "3000–6000 Hz_40ms_modulation_mae_db": 0.611122387313373, "mr_stft_db": 5.428807803826444, "band_mae_db": 0.1968679099161722, "rms_gain_error_db": -0.08615753414276403, "output_peak": 0.23516583778718023}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.255528286718019, "spectral_convergence": 0.18839926535100535, "spectral_convergence_status": "ok", "amplitude_floor": 7.733796592545008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.257046455174314, "spectral_convergence": 0.19592300489571238, "spectral_convergence_status": "ok", "amplitude_floor": 4.871370963739547e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7542381279903045, "spectral_convergence": 0.20035744151582288, "spectral_convergence_status": "ok", "amplitude_floor": 4.7483837933045734e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001233024392514864, "output_power": 0.00012714118602459503, "delta_db": 0.13314590203354526, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000393494821723947, "output_power": 0.00038919358283159497, "delta_db": -0.04773350698552247, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000358215617892152, "output_power": 0.0003435128848316742, "delta_db": -0.1820148512933486, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.947495410559829e-05, "output_power": 5.360089086707933e-05, "delta_db": -0.45162107690750997, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.0477761937191392e-05, "output_power": 9.4401631835544e-06, "delta_db": -0.45289025100943286, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.39686924640271e-06, "output_power": 9.54457820392586e-06, "delta_db": 0.06773556850560508, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22252352612249404, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1009558130568691, "pitch": null, "output_rms_dbfs": -30.303822314587393, "output_peak": 0.24339632315626616, "reference_rms_dbfs": -30.202866501530522, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.422270956627546}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.933633747331572e-05, "log_rms_mae_db": 0.7558173264050408, "rms_envelope_nrmse": 0.06280713538890483, "envelope_correlation": 0.996161306139302, "rms_gain_error_db": -0.10095572395266135, "reference_mean_power": 0.0009551583992700257, "output_mean_power": 0.0009332109513408438}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1194144575506277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002166069295810575, "output_power": 0.00020557974775534677, "reference_active": true, "delta_db": -0.22692017396444014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.38201712334803e-05, "output_power": 2.1287029152512115e-05, "reference_active": true, "delta_db": -0.4882982420002503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.302931519380675e-06, "output_power": 6.89599187017995e-06, "reference_active": true, "delta_db": -0.8063471516097096}], "active_band_count": 3, "active_band_mae_db": 0.5071885225248}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.652181544798992e-05, "log_rms_mae_db": 0.4447718461755786, "rms_envelope_nrmse": 0.03636160735744728, "envelope_correlation": 0.9987201176071835, "rms_gain_error_db": -0.10095552320307212, "reference_mean_power": 0.0009599620519447541, "output_mean_power": 0.000937904269952529}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.73215395386014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001986653950491428, "output_power": 0.00018847549507894357, "reference_active": true, "delta_db": -0.2286733242624911}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916927878865493e-05, "output_power": 9.720183672458492e-06, "reference_active": true, "delta_db": -0.5042596978408918}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.81959177520348e-07, "output_power": 4.571882417085727e-07, "reference_active": true, "delta_db": -1.0479746904719571}], "active_band_count": 3, "active_band_mae_db": 0.59363590419178}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.027389787958093e-05, "log_rms_mae_db": 1.2973172098207524, "rms_envelope_nrmse": 0.08488629503007973, "envelope_correlation": 0.9944936758807649, "rms_gain_error_db": -0.13563596280492823, "reference_mean_power": 0.0006127358569067375, "output_mean_power": 0.0005938950416527131}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.366962347376557e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001936423107817767, "output_power": 0.00018193884054374442, "reference_active": true, "delta_db": -0.2707483382340336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9545968551344243e-05, "output_power": 1.6675511037829836e-05, "reference_active": true, "delta_db": -0.6897804348865069}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.863910891753386e-06, "output_power": 4.851231525613167e-06, "reference_active": true, "delta_db": -1.5071963492831522}], "active_band_count": 3, "active_band_mae_db": 0.8225750408012309}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.943169310162246e-05, "log_rms_mae_db": 0.7869936240529904, "rms_envelope_nrmse": 0.05243241034565682, "envelope_correlation": 0.9979002222924428, "rms_gain_error_db": -0.13563596847845072, "reference_mean_power": 0.000615817475654304, "output_mean_power": 0.000596881903929552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.097820742933078e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018026482035346189, "output_power": 0.00016905339279092528, "reference_active": true, "delta_db": -0.27887088898841605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.739141870245847e-06, "output_power": 7.6072463949901274e-06, "reference_active": true, "delta_db": -0.6024130662362219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.763892611677605e-07, "output_power": 3.376349680381685e-07, "reference_active": true, "delta_db": -2.322684616699527}], "active_band_count": 3, "active_band_mae_db": 1.0679895239747217}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.05554473589972e-05, "log_rms_mae_db": 1.6433393348645917, "rms_envelope_nrmse": 0.1601830737760231, "envelope_correlation": 0.9823668983239033, "rms_gain_error_db": -0.43387738652033037, "reference_mean_power": 5.214627438247735e-05, "output_mean_power": 4.7188432020556295e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.666296423769008e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.72355480524535e-05, "output_power": 2.3427266925496854e-05, "reference_active": true, "delta_db": -0.6541449303029938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.163900077209868e-06, "output_power": 3.7335584865130746e-06, "reference_active": true, "delta_db": -0.4737734004587019}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1939677607224186e-06, "output_power": 7.212436625057523e-07, "reference_active": true, "delta_db": -2.189105901399391}], "active_band_count": 3, "active_band_mae_db": 1.1056747440536956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.291047760123656e-05, "log_rms_mae_db": 0.9467926406915637, "rms_envelope_nrmse": 0.09836286261737234, "envelope_correlation": 0.9941770113492783, "rms_gain_error_db": -0.43387814042985307, "reference_mean_power": 5.2408520364582374e-05, "output_mean_power": 4.7425736555433663e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.33304878770606e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.583453225376265e-05, "output_power": 2.2309950457360717e-05, "reference_active": true, "delta_db": -0.6370199698108376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0765400661096766e-06, "output_power": 1.8928792501321903e-06, "reference_active": true, "delta_db": -0.4021740459484547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.119613733354675e-08, "output_power": 3.6793702129298034e-08, "reference_active": true, "delta_db": -2.209505224676101}], "active_band_count": 3, "active_band_mae_db": 1.0828997468117978}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.3449414786854306e-05, "log_rms_mae_db": 1.711732217952592, "rms_envelope_nrmse": 0.20456900498869002, "envelope_correlation": 0.967303767834235, "rms_gain_error_db": -0.37083274942991823, "reference_mean_power": 6.886139457761324e-06, "output_mean_power": 6.32255387695584e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.997833564030985e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1829042042252175e-06, "output_power": 1.98557560741766e-06, "reference_active": true, "delta_db": -0.4114824838168184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.111376843686143e-07, "output_power": 6.347836346460427e-07, "reference_active": true, "delta_db": -1.0646885692549979}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6537364416157584e-07, "output_power": 1.9575564606544379e-07, "reference_active": true, "delta_db": 0.7324800042451147}], "active_band_count": 3, "active_band_mae_db": 0.7362170191056436}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.0291390653962812e-05, "log_rms_mae_db": 1.13568306956851, "rms_envelope_nrmse": 0.12175123837106262, "envelope_correlation": 0.9885358611248347, "rms_gain_error_db": -0.3708328456343888, "reference_mean_power": 6.9207717017349184e-06, "output_mean_power": 6.3543515569834515e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.443349031906855e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9318994555375575e-06, "output_power": 1.7889473963204314e-06, "reference_active": true, "delta_db": -0.33386949687893375}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.056544990756342e-07, "output_power": 3.4394746373486505e-07, "reference_active": true, "delta_db": -0.7166418574526408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.969546554060793e-09, "output_power": 9.570656655551752e-09, "reference_active": true, "delta_db": 0.7950812442271954}], "active_band_count": 3, "active_band_mae_db": 0.6151975328529233}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9103070675637412, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8dc7f7bef2e6f366795a6528370cfb79117823f94a0d56027b94ffd3a6efe7e1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9799838537159449, "standard_deviation": 0.001724999785855634, "minimum": 0.9768332252231013, "maximum": 0.9823668983239033, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6308286087718662, "standard_deviation": 0.01797858287921918, "minimum": 1.605378457233578, "maximum": 1.6527219080370468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.839035165243323, "standard_deviation": 0.28495649244664006, "minimum": 0.48314363457371656, "maximum": 1.244793793692419, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1679699030570081, "standard_deviation": 0.006178571495065021, "minimum": 0.16002492039607108, "maximum": 0.17861458894883, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929093391896964, "standard_deviation": 0.0007766678311268775, "minimum": 0.9917846853589565, "maximum": 0.9941770113492783, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.978001857326533, "standard_deviation": 0.023222143111563252, "minimum": 0.9383692936839154, "maximum": 0.9996768379270112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8042419632561668, "standard_deviation": 0.3419393014813905, "minimum": 0.33480912752820674, "maximum": 1.393965353341373, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10221580222677497, "standard_deviation": 0.0053850119930454726, "minimum": 0.09423941240278018, "maximum": 0.11277075571662981, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9716578458721812, "standard_deviation": 0.0024128805656569714, "minimum": 0.967303767834235, "maximum": 0.9754443891378201, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6486212215042564, "standard_deviation": 0.029756424762087796, "minimum": 1.618343068911068, "maximum": 1.711732217952592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7605647329046983, "standard_deviation": 0.35120652371945243, "minimum": 0.2157544044165057, "maximum": 1.144209929345317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19018818642538662, "standard_deviation": 0.00801363545281988, "minimum": 0.17723848408015505, "maximum": 0.20456900498869002, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9897411917201202, "standard_deviation": 0.0007246068630924429, "minimum": 0.9885358611248347, "maximum": 0.9907768751647628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.087967438723684, "standard_deviation": 0.03094097886755083, "minimum": 1.0445353580183587, "maximum": 1.13568306956851, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6373635368522861, "standard_deviation": 0.24074684577275748, "minimum": 0.11963156853667467, "maximum": 0.9109074870987185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11294486953929288, "standard_deviation": 0.004089052345351453, "minimum": 0.10763018704367223, "maximum": 0.12175123837106262, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937313495080062, "standard_deviation": 0.0007035208797135418, "minimum": 0.9927483248333769, "maximum": 0.9944936758807649, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3928371868601515, "standard_deviation": 0.05167731329806797, "minimum": 1.2973172098207524, "maximum": 1.442015852890291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4665405680242445, "standard_deviation": 0.2035054666890531, "minimum": 0.1650086648933422, "maximum": 0.8225750408012309, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09092085336424946, "standard_deviation": 0.0050995949700327, "minimum": 0.08488629503007973, "maximum": 0.09839077906757247, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979024693118128, "standard_deviation": 0.00035989058665926877, "minimum": 0.9974839215875873, "maximum": 0.9984699072041654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8426721137365829, "standard_deviation": 0.04732565672714262, "minimum": 0.7869936240529904, "maximum": 0.9025597528487651, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7084606496103862, "standard_deviation": 0.23604746996369816, "minimum": 0.35963906666406187, "maximum": 1.0679895239747217, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05363632806558808, "standard_deviation": 0.004420677422215606, "minimum": 0.04649936715411478, "maximum": 0.05888345902761794, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19433886029005798, "standard_deviation": 0.02682293288288292, "minimum": 0.16600963670784843, "maximum": 0.24535647414294084, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.4297018650337066, "standard_deviation": 0.005196054967556889, "minimum": 5.422270956627546, "maximum": 5.435337366353331, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.24039312709684363, "standard_deviation": 0.007924897646599349, "minimum": 0.23107996461918925, "maximum": 0.2534264087041444, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11327619001376742, "standard_deviation": 0.03377739409161729, "minimum": -0.16401982462297732, "maximum": -0.07200307332693655, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956792126564986, "standard_deviation": 0.0003626443587423837, "minimum": 0.9951015968576072, "maximum": 0.996161306139302, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7644615937618022, "standard_deviation": 0.015074632332631625, "minimum": 0.7369735531054159, "maximum": 0.7893683641553675, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33772182638706105, "standard_deviation": 0.14021783715496602, "minimum": 0.10237162128914039, "maximum": 0.5071885225248, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06677039344290464, "standard_deviation": 0.003118159172534347, "minimum": 0.06280713538890483, "maximum": 0.07118847073526893, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986723719573198, "standard_deviation": 0.00019097325951853407, "minimum": 0.9984231008816814, "maximum": 0.9990273462906268, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.43955345248072186, "standard_deviation": 0.02165749923942707, "minimum": 0.4061633743461146, "maximum": 0.4756486060229394, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4411432899360569, "standard_deviation": 0.16183624428896556, "minimum": 0.19250838881295093, "maximum": 0.59363590419178, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03750013684417042, "standard_deviation": 0.0031521334312440224, "minimum": 0.032574190177410686, "maximum": 0.04164872636217895, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0023", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.1958889961242676, "predicted_mos": 3.1958889961242676, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "98afaa4bf17f0b8a6f9bf56a9c9d790a38eda5289b3ca421febfc961feaf889e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.018841200042516, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7593998234983109, "全帯域_10ms_rms_envelope_nrmse": 0.06251203153732786, "全帯域_10ms_envelope_correlation": 0.9961989653312537, "全帯域_10ms_modulation_mae_db": 0.5160746799653081, "全帯域_40ms_log_rms_mae_db": 0.4453344584333102, "全帯域_40ms_rms_envelope_nrmse": 0.03567839401337641, "全帯域_40ms_envelope_correlation": 0.9987737562052346, "全帯域_40ms_modulation_mae_db": 0.5725947382176363, "300–1000 Hz_10ms_log_rms_mae_db": 1.3099623751575467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08546054313473896, "300–1000 Hz_10ms_envelope_correlation": 0.9944020332017672, "300–1000 Hz_10ms_modulation_mae_db": 0.9259394131631348, "300–1000 Hz_40ms_log_rms_mae_db": 0.8024523031053571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05225925683871, "300–1000 Hz_40ms_envelope_correlation": 0.9978956499584319, "300–1000 Hz_40ms_modulation_mae_db": 1.2292221608447116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.615190063653197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15903300892871808, "1000–3000 Hz_10ms_envelope_correlation": 0.9832143672938439, "1000–3000 Hz_10ms_modulation_mae_db": 0.8848997297921782, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9687321453325018, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1034564533677923, "1000–3000 Hz_40ms_envelope_correlation": 0.994060138075804, "1000–3000 Hz_40ms_modulation_mae_db": 0.8623577398914589, "3000–6000 Hz_10ms_log_rms_mae_db": 1.615556997285818, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18455931351774024, "3000–6000 Hz_10ms_envelope_correlation": 0.9751315301671574, "3000–6000 Hz_10ms_modulation_mae_db": 0.39031555450580235, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1552423132240996, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12126347053106717, "3000–6000 Hz_40ms_envelope_correlation": 0.9906138393848497, "3000–6000 Hz_40ms_modulation_mae_db": 0.4409181115594234, "mr_stft_db": 5.303764324752638, "band_mae_db": 0.24860291364900866, "rms_gain_error_db": -0.10107519198075383, "output_peak": 0.24370959236475714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7856219024719011, "全帯域_10ms_rms_envelope_nrmse": 0.06949519153653973, "全帯域_10ms_envelope_correlation": 0.9954690543235508, "全帯域_10ms_modulation_mae_db": 0.4888677776823966, "全帯域_40ms_log_rms_mae_db": 0.44543215762650457, "全帯域_40ms_rms_envelope_nrmse": 0.04116326718717523, "全帯域_40ms_envelope_correlation": 0.9985579118109439, "全帯域_40ms_modulation_mae_db": 0.5023209797280564, "300–1000 Hz_10ms_log_rms_mae_db": 1.4067401794727221, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09143182426238412, "300–1000 Hz_10ms_envelope_correlation": 0.9938869569480504, "300–1000 Hz_10ms_modulation_mae_db": 0.7182729463028004, "300–1000 Hz_40ms_log_rms_mae_db": 0.9042435931653927, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0558987441985609, "300–1000 Hz_40ms_envelope_correlation": 0.997965907029656, "300–1000 Hz_40ms_modulation_mae_db": 0.9600912587572089, "1000–3000 Hz_10ms_log_rms_mae_db": 1.617048618887849, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16947917233249615, "1000–3000 Hz_10ms_envelope_correlation": 0.9806168460236253, "1000–3000 Hz_10ms_modulation_mae_db": 0.5482644206581396, "1000–3000 Hz_40ms_log_rms_mae_db": 1.027760884855439, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11255106145913424, "1000–3000 Hz_40ms_envelope_correlation": 0.9924925915321445, "1000–3000 Hz_40ms_modulation_mae_db": 0.6445702813034694, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5982924780357761, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1770378391828827, "3000–6000 Hz_10ms_envelope_correlation": 0.976392566117279, "3000–6000 Hz_10ms_modulation_mae_db": 0.42708490303160973, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1500364692939298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11666981709270072, "3000–6000 Hz_40ms_envelope_correlation": 0.9902432473280782, "3000–6000 Hz_40ms_modulation_mae_db": 0.5409332442864229, "mr_stft_db": 5.311329256041677, "band_mae_db": 0.27722874562568073, "rms_gain_error_db": -0.16716249614103623, "output_peak": 0.2507838051596635}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7634562569765946, "全帯域_10ms_rms_envelope_nrmse": 0.06840827282203216, "全帯域_10ms_envelope_correlation": 0.9954541855448786, "全帯域_10ms_modulation_mae_db": 0.3606284126767032, "全帯域_40ms_log_rms_mae_db": 0.4049900088554088, "全帯域_40ms_rms_envelope_nrmse": 0.03596564978309697, "全帯域_40ms_envelope_correlation": 0.9987921008331175, "全帯域_40ms_modulation_mae_db": 0.580810381007817, "300–1000 Hz_10ms_log_rms_mae_db": 1.3991575524290085, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09378140901107487, "300–1000 Hz_10ms_envelope_correlation": 0.9933145154618752, "300–1000 Hz_10ms_modulation_mae_db": 0.41590282187575034, "300–1000 Hz_40ms_log_rms_mae_db": 0.795907825580399, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053230746755355135, "300–1000 Hz_40ms_envelope_correlation": 0.9979364050932001, "300–1000 Hz_40ms_modulation_mae_db": 0.9604641606630869, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5725769392281335, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15881469238663856, "1000–3000 Hz_10ms_envelope_correlation": 0.9821028357965014, "1000–3000 Hz_10ms_modulation_mae_db": 0.578608841772788, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9594427296265433, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09407241010894563, "1000–3000 Hz_40ms_envelope_correlation": 0.9940250821582087, "1000–3000 Hz_40ms_modulation_mae_db": 0.6927433384857421, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6105858327280724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18380131974642194, "3000–6000 Hz_10ms_envelope_correlation": 0.9746868589229948, "3000–6000 Hz_10ms_modulation_mae_db": 0.755785162011703, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1329591106439467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11677229914355022, "3000–6000 Hz_40ms_envelope_correlation": 0.9906946121519085, "3000–6000 Hz_40ms_modulation_mae_db": 0.7606637369989674, "mr_stft_db": 5.320331987670716, "band_mae_db": 0.2035000123997507, "rms_gain_error_db": -0.11388482133987535, "output_peak": 0.2389190758928155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7617274943970356, "全帯域_10ms_rms_envelope_nrmse": 0.07192048495142757, "全帯域_10ms_envelope_correlation": 0.9949910890420226, "全帯域_10ms_modulation_mae_db": 0.3625306025134412, "全帯域_40ms_log_rms_mae_db": 0.4395730253813644, "全帯域_40ms_rms_envelope_nrmse": 0.040078307703467535, "全帯域_40ms_envelope_correlation": 0.9984976298815099, "全帯域_40ms_modulation_mae_db": 0.32855296694597125, "300–1000 Hz_10ms_log_rms_mae_db": 1.4442342333260145, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09897877672984812, "300–1000 Hz_10ms_envelope_correlation": 0.9925378343963305, "300–1000 Hz_10ms_modulation_mae_db": 0.5248837286750443, "300–1000 Hz_40ms_log_rms_mae_db": 0.9040445445224378, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05737620670836638, "300–1000 Hz_40ms_envelope_correlation": 0.9975760100905965, "300–1000 Hz_40ms_modulation_mae_db": 0.8091094610976666, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6465277708604338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16989053318368944, "1000–3000 Hz_10ms_envelope_correlation": 0.9795936841711889, "1000–3000 Hz_10ms_modulation_mae_db": 0.9016973942360339, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9982226489951163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10439136308870804, "1000–3000 Hz_40ms_envelope_correlation": 0.9926990367665909, "1000–3000 Hz_40ms_modulation_mae_db": 0.9669173916975442, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5954338193522695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17973032652464535, "3000–6000 Hz_10ms_envelope_correlation": 0.9748659683086264, "3000–6000 Hz_10ms_modulation_mae_db": 0.0638084628923055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1084627305171044, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11060411581424044, "3000–6000 Hz_40ms_envelope_correlation": 0.9900236189993492, "3000–6000 Hz_40ms_modulation_mae_db": 0.13553025616783623, "mr_stft_db": 5.313500079729966, "band_mae_db": 0.20903651541122029, "rms_gain_error_db": -0.12825412502367212, "output_peak": 0.24960090374994753}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7756094104520942, "全帯域_10ms_rms_envelope_nrmse": 0.06590414776725077, "全帯域_10ms_envelope_correlation": 0.9957103624522101, "全帯域_10ms_modulation_mae_db": 0.18748447599873083, "全帯域_40ms_log_rms_mae_db": 0.4393860904574612, "全帯域_40ms_rms_envelope_nrmse": 0.03632335823174574, "全帯域_40ms_envelope_correlation": 0.9986611000871868, "全帯域_40ms_modulation_mae_db": 0.2552737332180191, "300–1000 Hz_10ms_log_rms_mae_db": 1.4315531773084924, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09596281623973676, "300–1000 Hz_10ms_envelope_correlation": 0.992905989732337, "300–1000 Hz_10ms_modulation_mae_db": 0.30070821322993474, "300–1000 Hz_40ms_log_rms_mae_db": 0.8734234655734749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05601793184017128, "300–1000 Hz_40ms_envelope_correlation": 0.9975786224834867, "300–1000 Hz_40ms_modulation_mae_db": 0.43490231107453936, "1000–3000 Hz_10ms_log_rms_mae_db": 1.609766342798911, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17393252723512426, "1000–3000 Hz_10ms_envelope_correlation": 0.97815052442139, "1000–3000 Hz_10ms_modulation_mae_db": 0.5337151368054952, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9816484141705177, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10219471287770003, "1000–3000 Hz_40ms_envelope_correlation": 0.9922257882824904, "1000–3000 Hz_40ms_modulation_mae_db": 0.469865599607543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5914558049305687, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16695948854222314, "3000–6000 Hz_10ms_envelope_correlation": 0.9792739599958232, "3000–6000 Hz_10ms_modulation_mae_db": 0.9121063411949115, "3000–6000 Hz_40ms_log_rms_mae_db": 1.07324682248328, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10828565617057709, "3000–6000 Hz_40ms_envelope_correlation": 0.9919938755934719, "3000–6000 Hz_40ms_modulation_mae_db": 0.5987186381334887, "mr_stft_db": 5.325237930520816, "band_mae_db": 0.19319356749587313, "rms_gain_error_db": -0.07345010450677252, "output_peak": 0.2315700062749369}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7665291456644167, "全帯域_10ms_rms_envelope_nrmse": 0.07205601037873742, "全帯域_10ms_envelope_correlation": 0.9950478168707882, "全帯域_10ms_modulation_mae_db": 0.36064259413902994, "全帯域_40ms_log_rms_mae_db": 0.4714941114215371, "全帯域_40ms_rms_envelope_nrmse": 0.04243329085061231, "全帯域_40ms_envelope_correlation": 0.998379667893988, "全帯域_40ms_modulation_mae_db": 0.5479893510331096, "300–1000 Hz_10ms_log_rms_mae_db": 1.4108996076865783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10032971371494756, "300–1000 Hz_10ms_envelope_correlation": 0.9923874250446484, "300–1000 Hz_10ms_modulation_mae_db": 0.4062855003622203, "300–1000 Hz_40ms_log_rms_mae_db": 0.8553884012802944, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059369057187373696, "300–1000 Hz_40ms_envelope_correlation": 0.9974619272286288, "300–1000 Hz_40ms_modulation_mae_db": 0.7588589489841816, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6542152854411336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1731878837676288, "1000–3000 Hz_10ms_envelope_correlation": 0.9784170077149358, "1000–3000 Hz_10ms_modulation_mae_db": 0.6293706541061468, "1000–3000 Hz_40ms_log_rms_mae_db": 1.018045885905204, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10251641596861087, "1000–3000 Hz_40ms_envelope_correlation": 0.9925363630741898, "1000–3000 Hz_40ms_modulation_mae_db": 0.30111697006288424, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5416440847969972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1706980660382361, "3000–6000 Hz_10ms_envelope_correlation": 0.9776153640858403, "3000–6000 Hz_10ms_modulation_mae_db": 0.9178036591297692, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0636458344775015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1083560842258035, "3000–6000 Hz_40ms_envelope_correlation": 0.9911172527849199, "3000–6000 Hz_40ms_modulation_mae_db": 0.8316634354061777, "mr_stft_db": 5.320835485335703, "band_mae_db": 0.21366190496214119, "rms_gain_error_db": -0.15253707527104057, "output_peak": 0.2439525493923463}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.762242051922267, "全帯域_10ms_rms_envelope_nrmse": 0.06560629374364921, "全帯域_10ms_envelope_correlation": 0.9957736535348753, "全帯域_10ms_modulation_mae_db": 0.28940247189069646, "全帯域_40ms_log_rms_mae_db": 0.43087948785983077, "全帯域_40ms_rms_envelope_nrmse": 0.035284563018842335, "全帯域_40ms_envelope_correlation": 0.9987713620947715, "全帯域_40ms_modulation_mae_db": 0.4503295840309763, "300–1000 Hz_10ms_log_rms_mae_db": 1.412901512889271, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09008222858871023, "300–1000 Hz_10ms_envelope_correlation": 0.9937795414772632, "300–1000 Hz_10ms_modulation_mae_db": 0.14632128855753826, "300–1000 Hz_40ms_log_rms_mae_db": 0.8208737934006365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05165297190295183, "300–1000 Hz_40ms_envelope_correlation": 0.9979779060634735, "300–1000 Hz_40ms_modulation_mae_db": 0.41178587848488113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.628696980411237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16219334724305035, "1000–3000 Hz_10ms_envelope_correlation": 0.9817835256861009, "1000–3000 Hz_10ms_modulation_mae_db": 1.1863368032057657, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0074781574276903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1021077045299836, "1000–3000 Hz_40ms_envelope_correlation": 0.993470494776802, "1000–3000 Hz_40ms_modulation_mae_db": 1.1327852642537417, "3000–6000 Hz_10ms_log_rms_mae_db": 1.577387509897097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18415483352747922, "3000–6000 Hz_10ms_envelope_correlation": 0.9740912086512046, "3000–6000 Hz_10ms_modulation_mae_db": 0.8099024834546756, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0956887559989879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11578182156279275, "3000–6000 Hz_40ms_envelope_correlation": 0.9902091801933716, "3000–6000 Hz_40ms_modulation_mae_db": 0.43610906000983585, "mr_stft_db": 5.3135265203215765, "band_mae_db": 0.22464249883394574, "rms_gain_error_db": -0.08668362214389622, "output_peak": 0.23788174014802743}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7259402632933107, "全帯域_10ms_rms_envelope_nrmse": 0.06526843250380314, "全帯域_10ms_envelope_correlation": 0.9958200972497909, "全帯域_10ms_modulation_mae_db": 0.1029572985319895, "全帯域_40ms_log_rms_mae_db": 0.3974581688928524, "全帯域_40ms_rms_envelope_nrmse": 0.03484534348444088, "全帯域_40ms_envelope_correlation": 0.9988062059834107, "全帯域_40ms_modulation_mae_db": 0.2016994594886777, "300–1000 Hz_10ms_log_rms_mae_db": 1.3407537420053048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09007688185040615, "300–1000 Hz_10ms_envelope_correlation": 0.9938137662242614, "300–1000 Hz_10ms_modulation_mae_db": 0.40392506521506427, "300–1000 Hz_40ms_log_rms_mae_db": 0.8025334884215274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04758165890966687, "300–1000 Hz_40ms_envelope_correlation": 0.9983561625507941, "300–1000 Hz_40ms_modulation_mae_db": 0.7957818821209103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.615230435236927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16093661226645678, "1000–3000 Hz_10ms_envelope_correlation": 0.9815049393079324, "1000–3000 Hz_10ms_modulation_mae_db": 1.0579526436729014, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9575470561075596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0990283922270946, "1000–3000 Hz_40ms_envelope_correlation": 0.9931143942900569, "1000–3000 Hz_40ms_modulation_mae_db": 1.255815648472077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5626582681894636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1860948486302677, "3000–6000 Hz_10ms_envelope_correlation": 0.9731568143612028, "3000–6000 Hz_10ms_modulation_mae_db": 0.2809836293163624, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0916269882680572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12241597150878662, "3000–6000 Hz_40ms_envelope_correlation": 0.9881874577947684, "3000–6000 Hz_40ms_modulation_mae_db": 0.4950384099958141, "mr_stft_db": 5.319873482532389, "band_mae_db": 0.20638472834817115, "rms_gain_error_db": -0.08285964733865647, "output_peak": 0.23610659631538905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.144184355083442, "spectral_convergence": 0.18769263475503412, "spectral_convergence_status": "ok", "amplitude_floor": 7.733796592545008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.145048819610348, "spectral_convergence": 0.19506291361196054, "spectral_convergence_status": "ok", "amplitude_floor": 4.871370963739547e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.622059799564125, "spectral_convergence": 0.1991230712932342, "spectral_convergence_status": "ok", "amplitude_floor": 4.7483837933045734e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001233024392514864, "output_power": 0.00012690368469588282, "delta_db": 0.12502563995058344, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000393494821723947, "output_power": 0.0003900006342906244, "delta_db": -0.03873708188629964, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000358215617892152, "output_power": 0.00034412705126596576, "delta_db": -0.17425703647468932, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.947495410559829e-05, "output_power": 5.275509028400215e-05, "delta_db": -0.520697449659402, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.0477761937191392e-05, "output_power": 9.139059164189693e-06, "delta_db": -0.5936703785999751, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.39686924640271e-06, "output_power": 9.482135855687399e-06, "delta_db": 0.03922989532310256, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24860291364900866, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10107519198075383, "pitch": null, "output_rms_dbfs": -30.303941693511277, "output_peak": 0.24370959236475714, "reference_rms_dbfs": -30.202866501530522, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.303764324752638}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.933633747331572e-05, "log_rms_mae_db": 0.7593998234983109, "rms_envelope_nrmse": 0.06251203153732786, "envelope_correlation": 0.9961989653312537, "rms_gain_error_db": -0.1010751015612475, "reference_mean_power": 0.0009551583992700257, "output_mean_power": 0.0009331852998612127}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1194144575506277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002166069295810575, "output_power": 0.0002069245790638941, "reference_active": true, "delta_db": -0.19860265835712695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.38201712334803e-05, "output_power": 2.1827714384690858e-05, "reference_active": true, "delta_db": -0.37936616711969523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.302931519380675e-06, "output_power": 6.640578799161689e-06, "reference_active": true, "delta_db": -0.9702552144191022}], "active_band_count": 3, "active_band_mae_db": 0.5160746799653081}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.652181544798992e-05, "log_rms_mae_db": 0.4453344584333102, "rms_envelope_nrmse": 0.03567839401337641, "envelope_correlation": 0.9987737562052346, "rms_gain_error_db": -0.10107490072779621, "reference_mean_power": 0.0009599620519447541, "output_mean_power": 0.0009378784894841994}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.73215395386014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001986653950491428, "output_power": 0.00018969793391046422, "reference_active": true, "delta_db": -0.20059624388523087}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916927878865493e-05, "output_power": 1.0067109310062919e-05, "reference_active": true, "delta_db": -0.35195656944083725}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.81959177520348e-07, "output_power": 4.4500960139205647e-07, "reference_active": true, "delta_db": -1.1652314013268408}], "active_band_count": 3, "active_band_mae_db": 0.5725947382176363}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.027389787958093e-05, "log_rms_mae_db": 1.3099623751575467, "rms_envelope_nrmse": 0.08546054313473896, "envelope_correlation": 0.9944020332017672, "rms_gain_error_db": -0.1277177761583443, "reference_mean_power": 0.0006127358569067375, "output_mean_power": 0.0005949788365259152}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.366962347376557e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001936423107817767, "output_power": 0.00018338856869499736, "reference_active": true, "delta_db": -0.2362799558812469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9545968551344243e-05, "output_power": 1.6598594326622966e-05, "reference_active": true, "delta_db": -0.7098588486421731}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.863910891753386e-06, "output_power": 4.501981368225904e-06, "reference_active": true, "delta_db": -1.8316794349659844}], "active_band_count": 3, "active_band_mae_db": 0.9259394131631348}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.943169310162246e-05, "log_rms_mae_db": 0.8024523031053571, "rms_envelope_nrmse": 0.05225925683871, "envelope_correlation": 0.9978956499584319, "rms_gain_error_db": -0.12771779530770616, "reference_mean_power": 0.000615817475654304, "output_mean_power": 0.0005979711476511983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.097820742933078e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018026482035346189, "output_power": 0.00017040934568551266, "reference_active": true, "delta_db": -0.24417571187134332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.739141870245847e-06, "output_power": 7.638863542800034e-06, "reference_active": true, "delta_db": -0.5844003763534568}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.763892611677605e-07, "output_power": 2.9840551227726587e-07, "reference_active": true, "delta_db": -2.8590903943093346}], "active_band_count": 3, "active_band_mae_db": 1.2292221608447116}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.05554473589972e-05, "log_rms_mae_db": 1.615190063653197, "rms_envelope_nrmse": 0.15903300892871808, "envelope_correlation": 0.9832143672938439, "rms_gain_error_db": -0.5178310698239974, "reference_mean_power": 5.214627438247735e-05, "output_mean_power": 4.628499045401239e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.666296423769008e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.72355480524535e-05, "output_power": 2.3244759051316518e-05, "reference_active": true, "delta_db": -0.6881107010264593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.163900077209868e-06, "output_power": 3.969113073967032e-06, "reference_active": true, "delta_db": -0.2080682755401359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1939677607224186e-06, "output_power": 7.964171049072774e-07, "reference_active": true, "delta_db": -1.7585202128099398}], "active_band_count": 3, "active_band_mae_db": 0.8848997297921782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.291047760123656e-05, "log_rms_mae_db": 0.9687321453325018, "rms_envelope_nrmse": 0.1034564533677923, "envelope_correlation": 0.994060138075804, "rms_gain_error_db": -0.5178317487048057, "reference_mean_power": 5.2408520364582374e-05, "output_mean_power": 4.651775250148309e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.33304878770606e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.583453225376265e-05, "output_power": 2.213978717269825e-05, "reference_active": true, "delta_db": -0.6702716111060774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0765400661096766e-06, "output_power": 2.0168956336165263e-06, "reference_active": true, "delta_db": -0.12656889315941128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.119613733354675e-08, "output_power": 4.052292063822318e-08, "reference_active": true, "delta_db": -1.7902327154088882}], "active_band_count": 3, "active_band_mae_db": 0.8623577398914589}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.3449414786854306e-05, "log_rms_mae_db": 1.615556997285818, "rms_envelope_nrmse": 0.18455931351774024, "envelope_correlation": 0.9751315301671574, "rms_gain_error_db": -0.5851248338107039, "reference_mean_power": 6.886139457761324e-06, "output_mean_power": 6.018154465251665e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.997833564030985e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1829042042252175e-06, "output_power": 1.8815165290673379e-06, "reference_active": true, "delta_db": -0.6452663937131932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.111376843686143e-07, "output_power": 7.199152095731151e-07, "reference_active": true, "delta_db": -0.5181322965367381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6537364416157584e-07, "output_power": 1.6566131099448376e-07, "reference_active": true, "delta_db": 0.0075479732674758005}], "active_band_count": 3, "active_band_mae_db": 0.39031555450580235}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.0291390653962812e-05, "log_rms_mae_db": 1.1552423132240996, "rms_envelope_nrmse": 0.12126347053106717, "envelope_correlation": 0.9906138393848497, "rms_gain_error_db": -0.585124923881974, "reference_mean_power": 6.9207717017349184e-06, "output_mean_power": 6.048421254237951e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.443349031906855e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9318994555375575e-06, "output_power": 1.694833394534365e-06, "reference_active": true, "delta_db": -0.5685750745799936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.056544990756342e-07, "output_power": 3.7755759085003673e-07, "reference_active": true, "delta_db": -0.31173090814287385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.969546554060793e-09, "output_power": 8.824263074719122e-09, "reference_active": true, "delta_db": 0.4424483519554029}], "active_band_count": 3, "active_band_mae_db": 0.4409181115594234}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9130941531833326, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "49e49bb9b91cac70eb52e602ce8ca9da1af4f3d8ebf2448d9c8dc00f91f083f4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9806729663019398, "standard_deviation": 0.001814577030290397, "minimum": 0.97815052442139, "maximum": 0.9832143672938439, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6199065545647278, "standard_deviation": 0.02495208355274031, "minimum": 1.5725769392281335, "maximum": 1.6542152854411336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7901057030311811, "standard_deviation": 0.2521251639270421, "minimum": 0.5337151368054952, "maximum": 1.1863368032057657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1659334721679753, "standard_deviation": 0.006348140931285025, "minimum": 0.15881469238663856, "maximum": 0.17393252723512426, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930779861195358, "standard_deviation": 0.0007094150592807367, "minimum": 0.9922257882824904, "maximum": 0.994060138075804, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9898597403025715, "standard_deviation": 0.026977259393165923, "minimum": 0.9575470561075596, "maximum": 1.027760884855439, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7907715292218076, "standard_deviation": 0.32582157962903, "minimum": 0.30111697006288424, "maximum": 1.255815648472077, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10253981420349617, "standard_deviation": 0.005193772504706876, "minimum": 0.09407241010894563, "maximum": 0.11255106145913424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9756517838262662, "standard_deviation": 0.002000145834573219, "minimum": 0.9731568143612028, "maximum": 0.9792739599958232, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.586626849402008, "standard_deviation": 0.02487472082416063, "minimum": 1.5416440847969972, "maximum": 1.615556997285818, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5697237744421424, "standard_deviation": 0.3212796759463685, "minimum": 0.0638084628923055, "maximum": 0.9178036591297692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17912950446373704, "standard_deviation": 0.0070610918642182, "minimum": 0.16695948854222314, "maximum": 0.1860948486302677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9903853855288396, "standard_deviation": 0.0010878350171461674, "minimum": 0.9881874577947684, "maximum": 0.9919938755934719, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1088636281133635, "standard_deviation": 0.03424871841964889, "minimum": 1.0636458344775015, "maximum": 1.1552423132240996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5299468615697458, "standard_deviation": 0.21468519655213486, "minimum": 0.13553025616783623, "maximum": 0.8316634354061777, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11501865450618981, "standard_deviation": 0.005468263220884696, "minimum": 0.10828565617057709, "maximum": 0.12241597150878662, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9933785078108166, "standard_deviation": 0.0007144176283224426, "minimum": 0.9923874250446484, "maximum": 0.9944020332017672, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3945252975343672, "standard_deviation": 0.04576457559063386, "minimum": 1.3099623751575467, "maximum": 1.4442342333260145, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.480279872172686, "standard_deviation": 0.24391643132737822, "minimum": 0.14632128855753826, "maximum": 0.9259394131631348, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09326302419148084, "standard_deviation": 0.004995233375518694, "minimum": 0.08546054313473896, "maximum": 0.10032971371494756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978435738122835, "standard_deviation": 0.00029143267366436425, "minimum": 0.9974619272286288, "maximum": 0.9983561625507941, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8448584268811901, "standard_deviation": 0.04554046874185411, "minimum": 0.795907825580399, "maximum": 0.9042435931653927, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7950270077533983, "standard_deviation": 0.27287141493332084, "minimum": 0.41178587848488113, "maximum": 1.2292221608447116, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054173321792644516, "standard_deviation": 0.003741429869138569, "minimum": 0.04758165890966687, "maximum": 0.059369057187373696, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22203136084072395, "standard_deviation": 0.0278218230254084, "minimum": 0.19319356749587313, "maximum": 0.27722874562568073, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.316049883363185, "standard_deviation": 0.006828162867408705, "minimum": 5.303764324752638, "maximum": 5.325237930520816, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2415655336622354, "standard_deviation": 0.00665607318544631, "minimum": 0.2315700062749369, "maximum": 0.2507838051596635, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11323838546821291, "standard_deviation": 0.03388459576065682, "minimum": -0.16716249614103623, "maximum": -0.07345010450677252, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955581530436712, "standard_deviation": 0.00040525401819533697, "minimum": 0.9949910890420226, "maximum": 0.9961989653312537, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7625657935844914, "standard_deviation": 0.017198345321850668, "minimum": 0.7259402632933107, "maximum": 0.7856219024719011, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33357353917478694, "standard_deviation": 0.13949046375518226, "minimum": 0.1029572985319895, "maximum": 0.5160746799653081, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06764635815509598, "standard_deviation": 0.003400102760967778, "minimum": 0.06251203153732786, "maximum": 0.07205601037873742, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986549668487703, "standard_deviation": 0.00016006226590320284, "minimum": 0.998379667893988, "maximum": 0.9988062059834107, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4343184386160337, "standard_deviation": 0.023636692200455924, "minimum": 0.3974581688928524, "maximum": 0.4714941114215371, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42994639920878297, "standard_deviation": 0.1490848349442254, "minimum": 0.2016994594886777, "maximum": 0.580810381007817, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03772152178409467, "standard_deviation": 0.0030005786953994917, "minimum": 0.03484534348444088, "maximum": 0.04243329085061231, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0023", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.8767971992492676, "predicted_mos": 2.8767971992492676, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5c8dffb21b0ddb0ad29d9fd8f99e6eb4fc08cfb7944b2f0968ef015da95320a9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9378084000200033, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7522479726997671, "全帯域_10ms_rms_envelope_nrmse": 0.06744991682947006, "全帯域_10ms_envelope_correlation": 0.9955433268654348, "全帯域_10ms_modulation_mae_db": 0.218913372378036, "全帯域_40ms_log_rms_mae_db": 0.40750865039897116, "全帯域_40ms_rms_envelope_nrmse": 0.03831133518004465, "全帯域_40ms_envelope_correlation": 0.9985508160662959, "全帯域_40ms_modulation_mae_db": 0.2672184207987761, "300–1000 Hz_10ms_log_rms_mae_db": 1.5455757145379676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08976235304628036, "300–1000 Hz_10ms_envelope_correlation": 0.9939214938730841, "300–1000 Hz_10ms_modulation_mae_db": 0.4879020722611354, "300–1000 Hz_40ms_log_rms_mae_db": 0.9986435697345272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05258839786619213, "300–1000 Hz_40ms_envelope_correlation": 0.9980100430858748, "300–1000 Hz_40ms_modulation_mae_db": 0.7689559575657938, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9466660786854795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18044103650608065, "1000–3000 Hz_10ms_envelope_correlation": 0.9764672181906014, "1000–3000 Hz_10ms_modulation_mae_db": 0.6800202238042656, "1000–3000 Hz_40ms_log_rms_mae_db": 1.144550436730099, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10295405306368793, "1000–3000 Hz_40ms_envelope_correlation": 0.9924354377198449, "1000–3000 Hz_40ms_modulation_mae_db": 0.483906669888039, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0129176368127646, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2463626757273926, "3000–6000 Hz_10ms_envelope_correlation": 0.9531727604440718, "3000–6000 Hz_10ms_modulation_mae_db": 1.603738791406928, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2933189895195447, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1349657106891097, "3000–6000 Hz_40ms_envelope_correlation": 0.9849927042754525, "3000–6000 Hz_40ms_modulation_mae_db": 1.6651491790813135, "mr_stft_db": 5.77214463929088, "band_mae_db": 0.17809339094633325, "rms_gain_error_db": -0.09119419839129803, "output_peak": 0.23719099326195378}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7494674257639516, "全帯域_10ms_rms_envelope_nrmse": 0.06094035359324863, "全帯域_10ms_envelope_correlation": 0.9963611977485164, "全帯域_10ms_modulation_mae_db": 0.37311561430372436, "全帯域_40ms_log_rms_mae_db": 0.396900090981319, "全帯域_40ms_rms_envelope_nrmse": 0.032190565169726466, "全帯域_40ms_envelope_correlation": 0.9989852270923824, "全帯域_40ms_modulation_mae_db": 0.3983006818005095, "300–1000 Hz_10ms_log_rms_mae_db": 1.5400493204654662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08616266209456713, "300–1000 Hz_10ms_envelope_correlation": 0.9943562027489782, "300–1000 Hz_10ms_modulation_mae_db": 0.541231342757729, "300–1000 Hz_40ms_log_rms_mae_db": 0.948151203212455, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04966729674273355, "300–1000 Hz_40ms_envelope_correlation": 0.9981864804350534, "300–1000 Hz_40ms_modulation_mae_db": 0.7024753839781228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8971976837272029, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1704978350811873, "1000–3000 Hz_10ms_envelope_correlation": 0.9791582203882785, "1000–3000 Hz_10ms_modulation_mae_db": 0.4703259180429231, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1272706260056358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10248171084252877, "1000–3000 Hz_40ms_envelope_correlation": 0.9926138568467496, "1000–3000 Hz_40ms_modulation_mae_db": 0.48380395893600764, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0706979108611496, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24749915514556356, "3000–6000 Hz_10ms_envelope_correlation": 0.9523099276322465, "3000–6000 Hz_10ms_modulation_mae_db": 0.9687650754011362, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3324135972838416, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12806097815162124, "3000–6000 Hz_40ms_envelope_correlation": 0.9863185872916698, "3000–6000 Hz_40ms_modulation_mae_db": 1.0097551949322974, "mr_stft_db": 5.763044213134324, "band_mae_db": 0.14960173100205104, "rms_gain_error_db": -0.08268786620843196, "output_peak": 0.25124857929168004}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7295634526715779, "全帯域_10ms_rms_envelope_nrmse": 0.06054185613175356, "全帯域_10ms_envelope_correlation": 0.9964322712348277, "全帯域_10ms_modulation_mae_db": 0.40785516568341135, "全帯域_40ms_log_rms_mae_db": 0.406241865226783, "全帯域_40ms_rms_envelope_nrmse": 0.031925350984969486, "全帯域_40ms_envelope_correlation": 0.9990326577604856, "全帯域_40ms_modulation_mae_db": 0.5418843023278122, "300–1000 Hz_10ms_log_rms_mae_db": 1.5640961284899955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08450219746041958, "300–1000 Hz_10ms_envelope_correlation": 0.9945928498734242, "300–1000 Hz_10ms_modulation_mae_db": 0.49439922034096656, "300–1000 Hz_40ms_log_rms_mae_db": 0.9756262056912686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04957851793279781, "300–1000 Hz_40ms_envelope_correlation": 0.9982069600734312, "300–1000 Hz_40ms_modulation_mae_db": 0.7539340116313277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8415212846362856, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1742818509221732, "1000–3000 Hz_10ms_envelope_correlation": 0.9779513339301857, "1000–3000 Hz_10ms_modulation_mae_db": 0.8331883930475334, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1089607613809345, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10220518427460362, "1000–3000 Hz_40ms_envelope_correlation": 0.992182045924379, "1000–3000 Hz_40ms_modulation_mae_db": 0.7039346909721226, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0708836408739306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26422463562191234, "3000–6000 Hz_10ms_envelope_correlation": 0.9460644284672681, "3000–6000 Hz_10ms_modulation_mae_db": 1.3078253574494576, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3414016039283563, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15280108271774256, "3000–6000 Hz_40ms_envelope_correlation": 0.9807085168508464, "3000–6000 Hz_40ms_modulation_mae_db": 1.192334588208803, "mr_stft_db": 5.764841299403319, "band_mae_db": 0.14595353188932136, "rms_gain_error_db": -0.09570077485068243, "output_peak": 0.24368833785652405}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7273459946437076, "全帯域_10ms_rms_envelope_nrmse": 0.060769030709687616, "全帯域_10ms_envelope_correlation": 0.9963669441608101, "全帯域_10ms_modulation_mae_db": 0.371328110955226, "全帯域_40ms_log_rms_mae_db": 0.3803667745047269, "全帯域_40ms_rms_envelope_nrmse": 0.033903750324710116, "全帯域_40ms_envelope_correlation": 0.9988469563133652, "全帯域_40ms_modulation_mae_db": 0.32701397181512376, "300–1000 Hz_10ms_log_rms_mae_db": 1.5138146145673337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08376229327754806, "300–1000 Hz_10ms_envelope_correlation": 0.9946188578524513, "300–1000 Hz_10ms_modulation_mae_db": 0.6878431265239127, "300–1000 Hz_40ms_log_rms_mae_db": 0.9544025495029281, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049155955012711125, "300–1000 Hz_40ms_envelope_correlation": 0.9981486277305875, "300–1000 Hz_40ms_modulation_mae_db": 0.9431304836093789, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9267260697315354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1799777419805849, "1000–3000 Hz_10ms_envelope_correlation": 0.9764993166368474, "1000–3000 Hz_10ms_modulation_mae_db": 0.6532081055989721, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0962304922642274, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10361676328728793, "1000–3000 Hz_40ms_envelope_correlation": 0.9921133779090585, "1000–3000 Hz_40ms_modulation_mae_db": 0.4664428287266298, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0772507667263875, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2573568733402641, "3000–6000 Hz_10ms_envelope_correlation": 0.947907884151165, "3000–6000 Hz_10ms_modulation_mae_db": 1.352545699248001, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3305021961234125, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.142252042657674, "3000–6000 Hz_40ms_envelope_correlation": 0.9829252918059509, "3000–6000 Hz_40ms_modulation_mae_db": 1.130221309852426, "mr_stft_db": 5.780509468352869, "band_mae_db": 0.16355321883438065, "rms_gain_error_db": -0.07489191480170514, "output_peak": 0.24071183638910396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7578700877096746, "全帯域_10ms_rms_envelope_nrmse": 0.06098265159136462, "全帯域_10ms_envelope_correlation": 0.9965907611727711, "全帯域_10ms_modulation_mae_db": 0.5549015330979912, "全帯域_40ms_log_rms_mae_db": 0.39461935328197717, "全帯域_40ms_rms_envelope_nrmse": 0.03426029156105052, "全帯域_40ms_envelope_correlation": 0.9991091490799735, "全帯域_40ms_modulation_mae_db": 0.8246454301437721, "300–1000 Hz_10ms_log_rms_mae_db": 1.4373865532644474, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08803677826609435, "300–1000 Hz_10ms_envelope_correlation": 0.9945034602846451, "300–1000 Hz_10ms_modulation_mae_db": 0.711249227001795, "300–1000 Hz_40ms_log_rms_mae_db": 0.8807735905151396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05422761599263694, "300–1000 Hz_40ms_envelope_correlation": 0.9982635647876905, "300–1000 Hz_40ms_modulation_mae_db": 1.1162010302378917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9175341055390909, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16948387403266682, "1000–3000 Hz_10ms_envelope_correlation": 0.9791875369741686, "1000–3000 Hz_10ms_modulation_mae_db": 0.8178491502866226, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1377024574013912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.099121816536698, "1000–3000 Hz_40ms_envelope_correlation": 0.9927438088105272, "1000–3000 Hz_40ms_modulation_mae_db": 0.6689253652840473, "3000–6000 Hz_10ms_log_rms_mae_db": 2.042739634974279, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23754194175789084, "3000–6000 Hz_10ms_envelope_correlation": 0.955850456987029, "3000–6000 Hz_10ms_modulation_mae_db": 1.3564528927042738, "3000–6000 Hz_40ms_log_rms_mae_db": 1.34532639553881, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13519589629558001, "3000–6000 Hz_40ms_envelope_correlation": 0.9846455245359143, "3000–6000 Hz_40ms_modulation_mae_db": 1.1841083124821654, "mr_stft_db": 5.757947506205434, "band_mae_db": 0.20283351802256644, "rms_gain_error_db": -0.1628813825020604, "output_peak": 0.2418366629513304}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7522326288796624, "全帯域_10ms_rms_envelope_nrmse": 0.05941806205861691, "全帯域_10ms_envelope_correlation": 0.9965055982876577, "全帯域_10ms_modulation_mae_db": 0.5404416654257003, "全帯域_40ms_log_rms_mae_db": 0.391156471173042, "全帯域_40ms_rms_envelope_nrmse": 0.02962033737293081, "全帯域_40ms_envelope_correlation": 0.9991069909730231, "全帯域_40ms_modulation_mae_db": 0.70291475558446, "300–1000 Hz_10ms_log_rms_mae_db": 1.5026877366428555, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07563092111765042, "300–1000 Hz_10ms_envelope_correlation": 0.9955940751763135, "300–1000 Hz_10ms_modulation_mae_db": 0.6289416276414199, "300–1000 Hz_40ms_log_rms_mae_db": 0.9238141758249637, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04043205106277702, "300–1000 Hz_40ms_envelope_correlation": 0.9987331623280872, "300–1000 Hz_40ms_modulation_mae_db": 0.9153222557205795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9140539340252865, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17685150958657722, "1000–3000 Hz_10ms_envelope_correlation": 0.9772949466735357, "1000–3000 Hz_10ms_modulation_mae_db": 0.546035118357114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1110055970467019, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10124531956400883, "1000–3000 Hz_40ms_envelope_correlation": 0.9924219700250555, "1000–3000 Hz_40ms_modulation_mae_db": 0.5607522038816328, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0726989800572424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2373396067538156, "3000–6000 Hz_10ms_envelope_correlation": 0.9556585938197122, "3000–6000 Hz_10ms_modulation_mae_db": 1.1608219838959881, "3000–6000 Hz_40ms_log_rms_mae_db": 1.343282847645571, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12943652968778704, "3000–6000 Hz_40ms_envelope_correlation": 0.9858575815194617, "3000–6000 Hz_40ms_modulation_mae_db": 0.9447559663918984, "mr_stft_db": 5.764706589796127, "band_mae_db": 0.14466594609046082, "rms_gain_error_db": -0.05752179516160827, "output_peak": 0.24606171036125823}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7542332458753028, "全帯域_10ms_rms_envelope_nrmse": 0.062290869936296664, "全帯域_10ms_envelope_correlation": 0.9962065674061907, "全帯域_10ms_modulation_mae_db": 0.5850744514342455, "全帯域_40ms_log_rms_mae_db": 0.4172049303246626, "全帯域_40ms_rms_envelope_nrmse": 0.03376535395188922, "全帯域_40ms_envelope_correlation": 0.9988920833960885, "全帯域_40ms_modulation_mae_db": 0.8250926645373059, "300–1000 Hz_10ms_log_rms_mae_db": 1.5449042500974033, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08850221661092643, "300–1000 Hz_10ms_envelope_correlation": 0.994044472770528, "300–1000 Hz_10ms_modulation_mae_db": 0.7374268096043721, "300–1000 Hz_40ms_log_rms_mae_db": 0.9627159161450707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05219526973313064, "300–1000 Hz_40ms_envelope_correlation": 0.9979846251137658, "300–1000 Hz_40ms_modulation_mae_db": 1.034256941548958, "1000–3000 Hz_10ms_log_rms_mae_db": 1.894678502297934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17404234050811543, "1000–3000 Hz_10ms_envelope_correlation": 0.9780163666854341, "1000–3000 Hz_10ms_modulation_mae_db": 0.3698899119796821, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1393425581294112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10405960597149819, "1000–3000 Hz_40ms_envelope_correlation": 0.9919612915412694, "1000–3000 Hz_40ms_modulation_mae_db": 0.35191003168746054, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0837138493381087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2526140313936873, "3000–6000 Hz_10ms_envelope_correlation": 0.9506608459053483, "3000–6000 Hz_10ms_modulation_mae_db": 0.949599100378074, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3772350502178947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13559712560989642, "3000–6000 Hz_40ms_envelope_correlation": 0.9847779951171183, "3000–6000 Hz_40ms_modulation_mae_db": 0.8065787806172727, "mr_stft_db": 5.77533761674776, "band_mae_db": 0.12997764884194823, "rms_gain_error_db": -0.09152741735831195, "output_peak": 0.23801080749985737}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7425984457542301, "全帯域_10ms_rms_envelope_nrmse": 0.058561212225430634, "全帯域_10ms_envelope_correlation": 0.9966851705452543, "全帯域_10ms_modulation_mae_db": 0.08709531441642802, "全帯域_40ms_log_rms_mae_db": 0.41497764594506614, "全帯域_40ms_rms_envelope_nrmse": 0.029178504731510912, "全帯域_40ms_envelope_correlation": 0.9992309250262794, "全帯域_40ms_modulation_mae_db": 0.08870305288913487, "300–1000 Hz_10ms_log_rms_mae_db": 1.4721767661853395, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07784618151528486, "300–1000 Hz_10ms_envelope_correlation": 0.9955011683837551, "300–1000 Hz_10ms_modulation_mae_db": 0.45281438255023304, "300–1000 Hz_40ms_log_rms_mae_db": 0.9223152507483097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04084736936013609, "300–1000 Hz_40ms_envelope_correlation": 0.9989250931363161, "300–1000 Hz_40ms_modulation_mae_db": 0.5779030557979999, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8608124660960506, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1734989998394841, "1000–3000 Hz_10ms_envelope_correlation": 0.97825851654439, "1000–3000 Hz_10ms_modulation_mae_db": 0.7714173081905287, "1000–3000 Hz_40ms_log_rms_mae_db": 1.104242309800464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09650691497007291, "1000–3000 Hz_40ms_envelope_correlation": 0.9933582922331756, "1000–3000 Hz_40ms_modulation_mae_db": 0.8527773298559019, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0214963993505792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2502796636510191, "3000–6000 Hz_10ms_envelope_correlation": 0.9508764152152329, "3000–6000 Hz_10ms_modulation_mae_db": 1.207553820249509, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3424519386610403, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1414863027442064, "3000–6000 Hz_40ms_envelope_correlation": 0.9831656872172018, "3000–6000 Hz_40ms_modulation_mae_db": 1.0829207111725394, "mr_stft_db": 5.771151198374599, "band_mae_db": 0.14342983605566859, "rms_gain_error_db": -0.09941237964926743, "output_peak": 0.24629384942010965}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.639571180190431, "spectral_convergence": 0.1978004796068417, "spectral_convergence_status": "ok", "amplitude_floor": 7.733796592545008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6539596924646265, "spectral_convergence": 0.20303883192903052, "spectral_convergence_status": "ok", "amplitude_floor": 4.871370963739547e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.022903045217583, "spectral_convergence": 0.20622122372607174, "spectral_convergence_status": "ok", "amplitude_floor": 4.7483837933045734e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001233024392514864, "output_power": 0.00012956493653906896, "delta_db": 0.21515818473360293, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000393494821723947, "output_power": 0.00038792294021188834, "delta_db": -0.0619355874616433, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000358215617892152, "output_power": 0.0003425335143939851, "delta_db": -0.1944144635131903, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.947495410559829e-05, "output_power": 5.478374555425215e-05, "delta_db": -0.3568239401799323, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.0477761937191392e-05, "output_power": 1.0127257843253459e-05, "delta_db": -0.14776659321572422, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.39686924640271e-06, "output_power": 9.599074040388347e-06, "delta_db": 0.09246157657390659, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17809339094633325, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09119419839129803, "pitch": null, "output_rms_dbfs": -30.294060699921822, "output_peak": 0.23719099326195378, "reference_rms_dbfs": -30.202866501530522, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.77214463929088}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.933633747331572e-05, "log_rms_mae_db": 0.7522479726997671, "rms_envelope_nrmse": 0.06744991682947006, "envelope_correlation": 0.9955433268654348, "rms_gain_error_db": -0.09119418773041911, "reference_mean_power": 0.0009551583992700257, "output_mean_power": 0.0009353108670078544}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1194144575506277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002166069295810575, "output_power": 0.0002101307955055199, "reference_active": true, "delta_db": -0.1318264158696902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.38201712334803e-05, "output_power": 2.121952272648306e-05, "reference_active": true, "delta_db": -0.5020926768201845}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.302931519380675e-06, "output_power": 8.259416244697935e-06, "reference_active": true, "delta_db": -0.022821024444233173}], "active_band_count": 3, "active_band_mae_db": 0.218913372378036}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.652181544798992e-05, "log_rms_mae_db": 0.40750865039897116, "rms_envelope_nrmse": 0.03831133518004465, "envelope_correlation": 0.9985508160662959, "rms_gain_error_db": -0.09119409899491351, "reference_mean_power": 0.0009599620519447541, "output_mean_power": 0.000940014722302085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.73215395386014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001986653950491428, "output_power": 0.0001923160859075811, "reference_active": true, "delta_db": -0.1410661371627497}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916927878865493e-05, "output_power": 9.87440912644796e-06, "reference_active": true, "delta_db": -0.43589323828553983}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.81959177520348e-07, "output_power": 5.526153354651753e-07, "reference_active": true, "delta_db": -0.22469588694803877}], "active_band_count": 3, "active_band_mae_db": 0.2672184207987761}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.027389787958093e-05, "log_rms_mae_db": 1.5455757145379676, "rms_envelope_nrmse": 0.08976235304628036, "envelope_correlation": 0.9939214938730841, "rms_gain_error_db": -0.17218463525080696, "reference_mean_power": 0.0006127358569067375, "output_mean_power": 0.0005889180049567683}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.366962347376557e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001936423107817767, "output_power": 0.00018490650355095677, "reference_active": true, "delta_db": -0.20048070064741605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9545968551344243e-05, "output_power": 1.554227834901678e-05, "reference_active": true, "delta_db": -0.9954251313351279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.863910891753386e-06, "output_power": 6.453444780499358e-06, "reference_active": true, "delta_db": -0.2678003848008623}], "active_band_count": 3, "active_band_mae_db": 0.4879020722611354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.943169310162246e-05, "log_rms_mae_db": 0.9986435697345272, "rms_envelope_nrmse": 0.05258839786619213, "envelope_correlation": 0.9980100430858748, "rms_gain_error_db": -0.1721846574392261, "reference_mean_power": 0.000615817475654304, "output_mean_power": 0.0005918798340900087}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.097820742933078e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018026482035346189, "output_power": 0.000171578556068832, "reference_active": true, "delta_db": -0.21447971423555215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.739141870245847e-06, "output_power": 7.1752027429459074e-06, "reference_active": true, "delta_db": -0.8563461266277492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.763892611677605e-07, "output_power": 4.33622368820268e-07, "reference_active": true, "delta_db": -1.23604203183408}], "active_band_count": 3, "active_band_mae_db": 0.7689559575657938}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.05554473589972e-05, "log_rms_mae_db": 1.9466660786854795, "rms_envelope_nrmse": 0.18044103650608065, "envelope_correlation": 0.9764672181906014, "rms_gain_error_db": -0.2672064481051803, "reference_mean_power": 5.214627438247735e-05, "output_mean_power": 4.903460042629467e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.666296423769008e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.72355480524535e-05, "output_power": 2.4229237009552715e-05, "reference_active": true, "delta_db": -0.5079638066115739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.163900077209868e-06, "output_power": 4.429926119278317e-06, "reference_active": true, "delta_db": 0.2689618425195465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1939677607224186e-06, "output_power": 8.926456603788186e-07, "reference_active": true, "delta_db": -1.2631350222816766}], "active_band_count": 3, "active_band_mae_db": 0.6800202238042656}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.291047760123656e-05, "log_rms_mae_db": 1.144550436730099, "rms_envelope_nrmse": 0.10295405306368793, "envelope_correlation": 0.9924354377198449, "rms_gain_error_db": -0.2672067197824821, "reference_mean_power": 5.2408520364582374e-05, "output_mean_power": 4.92811945759216e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.33304878770606e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.583453225376265e-05, "output_power": 2.327795998497146e-05, "reference_active": true, "delta_db": -0.4525568555203749}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0765400661096766e-06, "output_power": 2.1241382152813017e-06, "reference_active": true, "delta_db": 0.09842457295761498}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.119613733354675e-08, "output_power": 4.9733629246928525e-08, "reference_active": true, "delta_db": -0.9007385811861269}], "active_band_count": 3, "active_band_mae_db": 0.483906669888039}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.3449414786854306e-05, "log_rms_mae_db": 2.0129176368127646, "rms_envelope_nrmse": 0.2463626757273926, "envelope_correlation": 0.9531727604440718, "rms_gain_error_db": 0.10461612263705922, "reference_mean_power": 6.886139457761324e-06, "output_mean_power": 7.054032009050702e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.997833564030985e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1829042042252175e-06, "output_power": 2.2986071523816875e-06, "reference_active": true, "delta_db": 0.22430076418069492}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.111376843686143e-07, "output_power": 5.134418871633407e-07, "reference_active": true, "delta_db": -1.9860328259837137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6537364416157584e-07, "output_power": 3.009917382059337e-07, "reference_active": true, "delta_db": 2.6008827840563753}], "active_band_count": 3, "active_band_mae_db": 1.603738791406928}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.0291390653962812e-05, "log_rms_mae_db": 1.2933189895195447, "rms_envelope_nrmse": 0.1349657106891097, "envelope_correlation": 0.9849927042754525, "rms_gain_error_db": 0.10461612108144362, "reference_mean_power": 6.9207717017349184e-06, "output_mean_power": 7.089508627197443e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.443349031906855e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9318994555375575e-06, "output_power": 2.0910301316917774e-06, "reference_active": true, "delta_db": 0.343757709442319}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.056544990756342e-07, "output_power": 2.705329736076262e-07, "reference_active": true, "delta_db": -1.759360908957058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.969546554060793e-09, "output_power": 1.5511953842634254e-08, "reference_active": true, "delta_db": 2.8923289188445627}], "active_band_count": 3, "active_band_mae_db": 1.6651491790813135}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8957966062353697, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0a05bd64d88dc4a173a27d66d0efec2d6359ccb3e241661009447c5fd586dd81", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9778541820029302, "standard_deviation": 0.0010521850265028051, "minimum": 0.9764672181906014, "maximum": 0.9791875369741686, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8998987655923583, "standard_deviation": 0.0346172457287743, "minimum": 1.8415212846362856, "maximum": 1.9466660786854795, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6427417661634552, "standard_deviation": 0.16840286356226047, "minimum": 0.3698899119796821, "maximum": 0.8331883930475334, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1748843985571087, "standard_deviation": 0.00400155251321467, "minimum": 0.16948387403266682, "maximum": 0.18044103650608065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9924787601262575, "standard_deviation": 0.0004402106348432608, "minimum": 0.9919612915412694, "maximum": 0.9933582922331756, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.121163154844858, "standard_deviation": 0.01831600563849506, "minimum": 1.0962304922642274, "maximum": 1.144550436730099, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5715566349039802, "standard_deviation": 0.16074959473116895, "minimum": 0.35191003168746054, "maximum": 0.8527773298559019, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10152392106379826, "standard_deviation": 0.002542622942176672, "minimum": 0.09650691497007291, "maximum": 0.10405960597149819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9515626640777592, "standard_deviation": 0.0034472524519058175, "minimum": 0.9460644284672681, "maximum": 0.955850456987029, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0565498523743053, "standard_deviation": 0.027140025452698984, "minimum": 2.0129176368127646, "maximum": 2.0837138493381087, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2384128400916712, "standard_deviation": 0.21657622384709888, "minimum": 0.949599100378074, "maximum": 1.603738791406928, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24915232292394318, "standard_deviation": 0.009200930772523954, "minimum": 0.2373396067538156, "maximum": 0.26422463562191234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.984173986076702, "standard_deviation": 0.0018235234277838928, "minimum": 0.9807085168508464, "maximum": 0.9863185872916698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3382415773648089, "standard_deviation": 0.023101376466142593, "minimum": 1.2933189895195447, "maximum": 1.3772350502178947, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1269780053423397, "standard_deviation": 0.2531845327365778, "minimum": 0.8065787806172727, "maximum": 1.6651491790813135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13747445856920215, "standard_deviation": 0.00794676543788335, "minimum": 0.12806097815162124, "maximum": 0.15280108271774256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9946415726203974, "standard_deviation": 0.0006123267657073971, "minimum": 0.9939214938730841, "maximum": 0.9955940751763135, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.515086385531351, "standard_deviation": 0.04289130867390281, "minimum": 1.4373865532644474, "maximum": 1.5640961284899955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5927259760851955, "standard_deviation": 0.11226759545792551, "minimum": 0.45281438255023304, "maximum": 0.7374268096043721, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08427570042359639, "standard_deviation": 0.0050998450594805875, "minimum": 0.07563092111765042, "maximum": 0.08976235304628036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983073195863508, "standard_deviation": 0.0003395928149815106, "minimum": 0.9979846251137658, "maximum": 0.9989250931363161, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9458053076718328, "standard_deviation": 0.03647125924995389, "minimum": 0.8807735905151396, "maximum": 0.9986435697345272, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8515223900112565, "standard_deviation": 0.1810021396951103, "minimum": 0.5779030557979999, "maximum": 1.1162010302378917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.048586559212889416, "standard_deviation": 0.005206126553710992, "minimum": 0.04043205106277702, "maximum": 0.05422761599263694, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1572636027103413, "standard_deviation": 0.023394980662021832, "minimum": 0.12997764884194823, "maximum": 0.20283351802256644, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.768710316413164, "standard_deviation": 0.007366289082995217, "minimum": 5.757947506205434, "maximum": 5.780509468352869, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2431303471289772, "standard_deviation": 0.004687766869620344, "minimum": 0.23719099326195378, "maximum": 0.25124857929168004, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0944772161154207, "standard_deviation": 0.03073559678676874, "minimum": -0.1628813825020604, "maximum": -0.05752179516160827, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963364796776829, "standard_deviation": 0.0003528514815352439, "minimum": 0.9955433268654348, "maximum": 0.9966851705452543, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7456949067497343, "standard_deviation": 0.011511169153626457, "minimum": 0.7273459946437076, "maximum": 0.7578700877096746, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3923406534618453, "standard_deviation": 0.1731266399445515, "minimum": 0.08709531441642802, "maximum": 0.5850744514342455, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06136924413448359, "standard_deviation": 0.002696409178113089, "minimum": 0.058561212225430634, "maximum": 0.06744991682947006, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989693507134867, "standard_deviation": 0.00020952029068302833, "minimum": 0.9985508160662959, "maximum": 0.9992309250262794, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4011219727295685, "standard_deviation": 0.012580267859390697, "minimum": 0.3803667745047269, "maximum": 0.4172049303246626, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4969716599871118, "standard_deviation": 0.27191242852545455, "minimum": 0.08870305288913487, "maximum": 0.8250926645373059, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.032894436159604024, "standard_deviation": 0.0029019205106123047, "minimum": 0.029178504731510912, "maximum": 0.03831133518004465, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0023", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.962435483932495, "predicted_mos": 2.962435483932495, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fd4192f92a2132e61670a95bb896a83c07253aaff2f532d01a471aeeaa4b32dd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.019497099798173, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.793510817968802, "全帯域_10ms_rms_envelope_nrmse": 0.06857539303186494, "全帯域_10ms_envelope_correlation": 0.9953579201467803, "全帯域_10ms_modulation_mae_db": 0.14205338208929677, "全帯域_40ms_log_rms_mae_db": 0.44577879258951864, "全帯域_40ms_rms_envelope_nrmse": 0.040223262692485474, "全帯域_40ms_envelope_correlation": 0.9983509187381467, "全帯域_40ms_modulation_mae_db": 0.20309467873100986, "300–1000 Hz_10ms_log_rms_mae_db": 1.3854558399575392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09365758874451903, "300–1000 Hz_10ms_envelope_correlation": 0.9932439542878575, "300–1000 Hz_10ms_modulation_mae_db": 0.2843406375630083, "300–1000 Hz_40ms_log_rms_mae_db": 0.8602406689314588, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05711817383514745, "300–1000 Hz_40ms_envelope_correlation": 0.9974609498561092, "300–1000 Hz_40ms_modulation_mae_db": 0.42478049016757863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.457742179663484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14282907825027427, "1000–3000 Hz_10ms_envelope_correlation": 0.9852745988176701, "1000–3000 Hz_10ms_modulation_mae_db": 0.5940722971087222, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8766807208926427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08024650710609413, "1000–3000 Hz_40ms_envelope_correlation": 0.9951678719395395, "1000–3000 Hz_40ms_modulation_mae_db": 0.4454743973113105, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4560556858763285, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17507993920513618, "3000–6000 Hz_10ms_envelope_correlation": 0.9768006256080431, "3000–6000 Hz_10ms_modulation_mae_db": 0.9519392391865948, "3000–6000 Hz_40ms_log_rms_mae_db": 1.006550952876402, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10906938203113413, "3000–6000 Hz_40ms_envelope_correlation": 0.9915871463073115, "3000–6000 Hz_40ms_modulation_mae_db": 0.8901395918663351, "mr_stft_db": 5.323324840023443, "band_mae_db": 0.19583094506626866, "rms_gain_error_db": -0.07902833978565867, "output_peak": 0.24079551567786467}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7901653249764359, "全帯域_10ms_rms_envelope_nrmse": 0.06630296186644258, "全帯域_10ms_envelope_correlation": 0.9956834354115368, "全帯域_10ms_modulation_mae_db": 0.5486195459190187, "全帯域_40ms_log_rms_mae_db": 0.4426587196563143, "全帯域_40ms_rms_envelope_nrmse": 0.0322915802886629, "全帯域_40ms_envelope_correlation": 0.998992049220059, "全帯域_40ms_modulation_mae_db": 0.6460875349650824, "300–1000 Hz_10ms_log_rms_mae_db": 1.3260126449093022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08750892198354307, "300–1000 Hz_10ms_envelope_correlation": 0.9941304718592198, "300–1000 Hz_10ms_modulation_mae_db": 0.6315884357377906, "300–1000 Hz_40ms_log_rms_mae_db": 0.7940452513784313, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046246694060878966, "300–1000 Hz_40ms_envelope_correlation": 0.9984075923417098, "300–1000 Hz_40ms_modulation_mae_db": 0.8645097099963427, "1000–3000 Hz_10ms_log_rms_mae_db": 1.566213378548029, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1745598794183156, "1000–3000 Hz_10ms_envelope_correlation": 0.9787819644331619, "1000–3000 Hz_10ms_modulation_mae_db": 1.3761883698081618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9632511738283152, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11315373560441327, "1000–3000 Hz_40ms_envelope_correlation": 0.9917721989796084, "1000–3000 Hz_40ms_modulation_mae_db": 1.2716372433137475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4657610713329128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1738177086901002, "3000–6000 Hz_10ms_envelope_correlation": 0.9765103944380609, "3000–6000 Hz_10ms_modulation_mae_db": 0.4536139623010642, "3000–6000 Hz_40ms_log_rms_mae_db": 0.988266938803776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10756444624376205, "3000–6000 Hz_40ms_envelope_correlation": 0.9908863261672232, "3000–6000 Hz_40ms_modulation_mae_db": 0.7579660438686489, "mr_stft_db": 5.327561421602671, "band_mae_db": 0.2162411250991559, "rms_gain_error_db": -0.08923948460656386, "output_peak": 0.2445526610099186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.795713629019137, "全帯域_10ms_rms_envelope_nrmse": 0.07427228871897727, "全帯域_10ms_envelope_correlation": 0.9945789021396713, "全帯域_10ms_modulation_mae_db": 0.1493242677672151, "全帯域_40ms_log_rms_mae_db": 0.44285078685396106, "全帯域_40ms_rms_envelope_nrmse": 0.0462521432836756, "全帯域_40ms_envelope_correlation": 0.9978433012804212, "全帯域_40ms_modulation_mae_db": 0.18221456746306633, "300–1000 Hz_10ms_log_rms_mae_db": 1.3689277040010754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09334338775669572, "300–1000 Hz_10ms_envelope_correlation": 0.9934773660796262, "300–1000 Hz_10ms_modulation_mae_db": 0.389091704124447, "300–1000 Hz_40ms_log_rms_mae_db": 0.8076168565959665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05744052442282382, "300–1000 Hz_40ms_envelope_correlation": 0.9976640151900231, "300–1000 Hz_40ms_modulation_mae_db": 0.4499493147091143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5129227949461794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16299280089809226, "1000–3000 Hz_10ms_envelope_correlation": 0.9807764454646387, "1000–3000 Hz_10ms_modulation_mae_db": 0.2060891191532324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8809542813737352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09138270616815386, "1000–3000 Hz_40ms_envelope_correlation": 0.9937875330958089, "1000–3000 Hz_40ms_modulation_mae_db": 0.4509846550009582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4426435948935852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17411947888825538, "3000–6000 Hz_10ms_envelope_correlation": 0.9763659198138828, "3000–6000 Hz_10ms_modulation_mae_db": 0.6724067682699305, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9839434979038775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11316707661471717, "3000–6000 Hz_40ms_envelope_correlation": 0.989753904618617, "3000–6000 Hz_40ms_modulation_mae_db": 0.7940202791859118, "mr_stft_db": 5.325743040862753, "band_mae_db": 0.19171587171585538, "rms_gain_error_db": -0.10230033830820906, "output_peak": 0.2475138398591227}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7464589419678835, "全帯域_10ms_rms_envelope_nrmse": 0.06480753123082796, "全帯域_10ms_envelope_correlation": 0.9959630819001449, "全帯域_10ms_modulation_mae_db": 0.1433748567051021, "全帯域_40ms_log_rms_mae_db": 0.4144128537203642, "全帯域_40ms_rms_envelope_nrmse": 0.03867927034115055, "全帯域_40ms_envelope_correlation": 0.998603843023822, "全帯域_40ms_modulation_mae_db": 0.1079938606760781, "300–1000 Hz_10ms_log_rms_mae_db": 1.3292238233973255, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09039786576336103, "300–1000 Hz_10ms_envelope_correlation": 0.9938461169355528, "300–1000 Hz_10ms_modulation_mae_db": 0.36587569010367615, "300–1000 Hz_40ms_log_rms_mae_db": 0.8010741156595254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057214333735755324, "300–1000 Hz_40ms_envelope_correlation": 0.9976099335950418, "300–1000 Hz_40ms_modulation_mae_db": 0.16554137677060934, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5396441345473606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15539806856593352, "1000–3000 Hz_10ms_envelope_correlation": 0.9828225322728855, "1000–3000 Hz_10ms_modulation_mae_db": 0.3299605330214884, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9498784090548354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10113952311054544, "1000–3000 Hz_40ms_envelope_correlation": 0.9927707385211579, "1000–3000 Hz_40ms_modulation_mae_db": 0.19329509199432737, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4913962953925946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16239493864547036, "3000–6000 Hz_10ms_envelope_correlation": 0.9797891544129027, "3000–6000 Hz_10ms_modulation_mae_db": 0.6288511779486669, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0303270217523326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09990924817581355, "3000–6000 Hz_40ms_envelope_correlation": 0.9925711304041807, "3000–6000 Hz_40ms_modulation_mae_db": 0.26376433884242667, "mr_stft_db": 5.337288805209989, "band_mae_db": 0.1920889010424891, "rms_gain_error_db": -0.12339623431673045, "output_peak": 0.2491493828135066}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.746850735016722, "全帯域_10ms_rms_envelope_nrmse": 0.06709574849812003, "全帯域_10ms_envelope_correlation": 0.9955638548478167, "全帯域_10ms_modulation_mae_db": 0.17811202208592267, "全帯域_40ms_log_rms_mae_db": 0.42081898350491614, "全帯域_40ms_rms_envelope_nrmse": 0.0381443738265894, "全帯域_40ms_envelope_correlation": 0.9985322776479043, "全帯域_40ms_modulation_mae_db": 0.2509037801902259, "300–1000 Hz_10ms_log_rms_mae_db": 1.3383197951789292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08962603155057502, "300–1000 Hz_10ms_envelope_correlation": 0.9938890835163987, "300–1000 Hz_10ms_modulation_mae_db": 0.4014760313357863, "300–1000 Hz_40ms_log_rms_mae_db": 0.8620347407980807, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053918141639724054, "300–1000 Hz_40ms_envelope_correlation": 0.9978343425373124, "300–1000 Hz_40ms_modulation_mae_db": 0.6977083036190926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.504569737562817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15446575578190153, "1000–3000 Hz_10ms_envelope_correlation": 0.9830656536103387, "1000–3000 Hz_10ms_modulation_mae_db": 0.45492357788770593, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9524312109417604, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10024314090325709, "1000–3000 Hz_40ms_envelope_correlation": 0.9929360180977478, "1000–3000 Hz_40ms_modulation_mae_db": 0.5075469445921431, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4590068910466192, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16146120825302177, "3000–6000 Hz_10ms_envelope_correlation": 0.9797659475615141, "3000–6000 Hz_10ms_modulation_mae_db": 0.5271616949722618, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9869620756372053, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09751731624242099, "3000–6000 Hz_40ms_envelope_correlation": 0.9924713968602781, "3000–6000 Hz_40ms_modulation_mae_db": 0.8443222113907854, "mr_stft_db": 5.3160765733986395, "band_mae_db": 0.19276073449110717, "rms_gain_error_db": -0.08180547696193495, "output_peak": 0.25076257419756515}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7422868960141772, "全帯域_10ms_rms_envelope_nrmse": 0.06251773531534174, "全帯域_10ms_envelope_correlation": 0.9962663940596878, "全帯域_10ms_modulation_mae_db": 0.3037323094980305, "全帯域_40ms_log_rms_mae_db": 0.40004025337945726, "全帯域_40ms_rms_envelope_nrmse": 0.03621573495802703, "全帯域_40ms_envelope_correlation": 0.9988128759755592, "全帯域_40ms_modulation_mae_db": 0.3447238902388105, "300–1000 Hz_10ms_log_rms_mae_db": 1.3247155794857464, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08580637207550051, "300–1000 Hz_10ms_envelope_correlation": 0.9945219527979704, "300–1000 Hz_10ms_modulation_mae_db": 0.5461683505680343, "300–1000 Hz_40ms_log_rms_mae_db": 0.7836115599220234, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0529976493621656, "300–1000 Hz_40ms_envelope_correlation": 0.9980373094305093, "300–1000 Hz_40ms_modulation_mae_db": 0.7418541646448465, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4995833018979516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1484824329655634, "1000–3000 Hz_10ms_envelope_correlation": 0.9843014617654883, "1000–3000 Hz_10ms_modulation_mae_db": 0.610966385795784, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9523234908131293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08693584530386034, "1000–3000 Hz_40ms_envelope_correlation": 0.9947831654869635, "1000–3000 Hz_40ms_modulation_mae_db": 0.3235674962372797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4892289773882914, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17527079782174004, "3000–6000 Hz_10ms_envelope_correlation": 0.975929121933486, "3000–6000 Hz_10ms_modulation_mae_db": 0.37369327058666896, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0076353951871388, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10626612354951061, "3000–6000 Hz_40ms_envelope_correlation": 0.9908306867392793, "3000–6000 Hz_40ms_modulation_mae_db": 0.5595887582181628, "mr_stft_db": 5.32378810138566, "band_mae_db": 0.2207315547120743, "rms_gain_error_db": -0.12749459260307264, "output_peak": 0.23624431315817265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7972804071737061, "全帯域_10ms_rms_envelope_nrmse": 0.07419802758827677, "全帯域_10ms_envelope_correlation": 0.9946610152452815, "全帯域_10ms_modulation_mae_db": 0.2932123723658259, "全帯域_40ms_log_rms_mae_db": 0.446522684162291, "全帯域_40ms_rms_envelope_nrmse": 0.04343932933564387, "全帯域_40ms_envelope_correlation": 0.9982056015610097, "全帯域_40ms_modulation_mae_db": 0.30123329208538213, "300–1000 Hz_10ms_log_rms_mae_db": 1.37373782800587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10049067899603, "300–1000 Hz_10ms_envelope_correlation": 0.992361523666661, "300–1000 Hz_10ms_modulation_mae_db": 0.6192894759692126, "300–1000 Hz_40ms_log_rms_mae_db": 0.88958580574161, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06131334183263683, "300–1000 Hz_40ms_envelope_correlation": 0.9972637485967426, "300–1000 Hz_40ms_modulation_mae_db": 0.4646732244723782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.484531726852167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14942425638840331, "1000–3000 Hz_10ms_envelope_correlation": 0.9846753419456262, "1000–3000 Hz_10ms_modulation_mae_db": 0.659699631333789, "1000–3000 Hz_40ms_log_rms_mae_db": 0.865974587430166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0923449904625842, "1000–3000 Hz_40ms_envelope_correlation": 0.9948532563916359, "1000–3000 Hz_40ms_modulation_mae_db": 0.8401734766288733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5128478310295774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1658403768192955, "3000–6000 Hz_10ms_envelope_correlation": 0.9785656798424591, "3000–6000 Hz_10ms_modulation_mae_db": 0.36082127602465613, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9906392235203176, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09449305990510731, "3000–6000 Hz_40ms_envelope_correlation": 0.9929676540481998, "3000–6000 Hz_40ms_modulation_mae_db": 0.19528710537337388, "mr_stft_db": 5.313437107912168, "band_mae_db": 0.21211883898138342, "rms_gain_error_db": -0.13012784725879242, "output_peak": 0.2420457221664679}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7536104805078896, "全帯域_10ms_rms_envelope_nrmse": 0.0635685244473559, "全帯域_10ms_envelope_correlation": 0.9960522630104621, "全帯域_10ms_modulation_mae_db": 0.4800237518474919, "全帯域_40ms_log_rms_mae_db": 0.40291639828410303, "全帯域_40ms_rms_envelope_nrmse": 0.03264038100143097, "全帯域_40ms_envelope_correlation": 0.9989758525353515, "全帯域_40ms_modulation_mae_db": 0.5797202649086309, "300–1000 Hz_10ms_log_rms_mae_db": 1.3943571588673769, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0846482333339214, "300–1000 Hz_10ms_envelope_correlation": 0.9945901740585487, "300–1000 Hz_10ms_modulation_mae_db": 0.5387706473060133, "300–1000 Hz_40ms_log_rms_mae_db": 0.834661339236806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04499985450523699, "300–1000 Hz_40ms_envelope_correlation": 0.9985884319785512, "300–1000 Hz_40ms_modulation_mae_db": 0.8087695784020266, "1000–3000 Hz_10ms_log_rms_mae_db": 1.488541540672782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1608595170315316, "1000–3000 Hz_10ms_envelope_correlation": 0.9812483899506316, "1000–3000 Hz_10ms_modulation_mae_db": 0.42859679845642235, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8926496737109471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09255815534938587, "1000–3000 Hz_40ms_envelope_correlation": 0.9935494834796628, "1000–3000 Hz_40ms_modulation_mae_db": 0.3610542946676527, "3000–6000 Hz_10ms_log_rms_mae_db": 1.482402308706039, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1828588769574228, "3000–6000 Hz_10ms_envelope_correlation": 0.9738713883901852, "3000–6000 Hz_10ms_modulation_mae_db": 0.6384524703869974, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9971014301795743, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1052740949839678, "3000–6000 Hz_40ms_envelope_correlation": 0.9911171724760084, "3000–6000 Hz_40ms_modulation_mae_db": 1.0251821472122806, "mr_stft_db": 5.332661658767822, "band_mae_db": 0.18576589606785646, "rms_gain_error_db": -0.09014752061251036, "output_peak": 0.2561972920578169}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.093120236035607, "spectral_convergence": 0.1935824927241485, "spectral_convergence_status": "ok", "amplitude_floor": 7.733796592545008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.158808203957205, "spectral_convergence": 0.19357993222308856, "spectral_convergence_status": "ok", "amplitude_floor": 4.871370963739547e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.718046080077516, "spectral_convergence": 0.19607192993796238, "spectral_convergence_status": "ok", "amplitude_floor": 4.7483837933045734e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001233024392514864, "output_power": 0.0001280206239872295, "delta_db": 0.16308271480926145, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000393494821723947, "output_power": 0.0003932481519160196, "delta_db": -0.0027233122769352897, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000358215617892152, "output_power": 0.000340026839322465, "delta_db": -0.22631318249746252, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.947495410559829e-05, "output_power": 5.739730990105834e-05, "delta_db": -0.1544257723466646, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.0477761937191392e-05, "output_power": 9.243465777538139e-06, "delta_db": -0.5443368906549142, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.39686924640271e-06, "output_power": 9.216643813926385e-06, "delta_db": -0.08410379781237402, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19583094506626866, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07902833978565867, "pitch": null, "output_rms_dbfs": -30.28189484131618, "output_peak": 0.24079551567786467, "reference_rms_dbfs": -30.202866501530522, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.323324840023443}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.933633747331572e-05, "log_rms_mae_db": 0.793510817968802, "rms_envelope_nrmse": 0.06857539303186494, "envelope_correlation": 0.9953579201467803, "rms_gain_error_db": -0.07902826103088305, "reference_mean_power": 0.0009551583992700257, "output_mean_power": 0.0009379346342331759}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1194144575506277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002166069295810575, "output_power": 0.0002086228510801222, "reference_active": true, "delta_db": -0.16310469984792086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.38201712334803e-05, "output_power": 2.3160913296870018e-05, "reference_active": true, "delta_db": -0.12189198335444862}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.302931519380675e-06, "output_power": 8.037391248644354e-06, "reference_active": true, "delta_db": -0.14116346306552086}], "active_band_count": 3, "active_band_mae_db": 0.14205338208929677}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.652181544798992e-05, "log_rms_mae_db": 0.44577879258951864, "rms_envelope_nrmse": 0.040223262692485474, "envelope_correlation": 0.9983509187381467, "rms_gain_error_db": -0.07902834298055919, "reference_mean_power": 0.0009599620519447541, "output_mean_power": 0.0009426516479013248}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.73215395386014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001986653950491428, "output_power": 0.00019209552295288055, "reference_active": true, "delta_db": -0.14604982040448394}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916927878865493e-05, "output_power": 1.086527438030819e-05, "reference_active": true, "delta_db": -0.020597431955644292}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.81959177520348e-07, "output_power": 5.25567913297565e-07, "reference_active": true, "delta_db": -0.4426367838329014}], "active_band_count": 3, "active_band_mae_db": 0.20309467873100986}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.027389787958093e-05, "log_rms_mae_db": 1.3854558399575392, "rms_envelope_nrmse": 0.09365758874451903, "envelope_correlation": 0.9932439542878575, "rms_gain_error_db": -0.13194483850008293, "reference_mean_power": 0.0006127358569067375, "output_mean_power": 0.0005944000151999737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.366962347376557e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001936423107817767, "output_power": 0.00018433024115910383, "reference_active": true, "delta_db": -0.21403665268585187}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9545968551344243e-05, "output_power": 1.7346167616366885e-05, "reference_active": true, "delta_db": -0.5185365700206253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.863910891753386e-06, "output_power": 7.056941313628566e-06, "reference_active": true, "delta_db": 0.12044868998254778}], "active_band_count": 3, "active_band_mae_db": 0.2843406375630083}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.943169310162246e-05, "log_rms_mae_db": 0.8602406689314588, "rms_envelope_nrmse": 0.05711817383514745, "envelope_correlation": 0.9974609498561092, "rms_gain_error_db": -0.13194482199899615, "reference_mean_power": 0.000615817475654304, "output_mean_power": 0.0005973894201785285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.097820742933078e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018026482035346189, "output_power": 0.00017266674579104478, "reference_active": true, "delta_db": -0.18702276053376102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.739141870245847e-06, "output_power": 7.906445468079183e-06, "reference_active": true, "delta_db": -0.434875098044562}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.763892611677605e-07, "output_power": 4.959884001218327e-07, "reference_active": true, "delta_db": -0.6524436119244128}], "active_band_count": 3, "active_band_mae_db": 0.42478049016757863}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.05554473589972e-05, "log_rms_mae_db": 1.457742179663484, "rms_envelope_nrmse": 0.14282907825027427, "envelope_correlation": 0.9852745988176701, "rms_gain_error_db": -0.10262391533459822, "reference_mean_power": 5.214627438247735e-05, "output_mean_power": 5.092850104124157e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.666296423769008e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.72355480524535e-05, "output_power": 2.6703410241002825e-05, "reference_active": true, "delta_db": -0.08569391052914653}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.163900077209868e-06, "output_power": 4.4477716775205065e-06, "reference_active": true, "delta_db": 0.2864218599414908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1939677607224186e-06, "output_power": 8.629437509624216e-07, "reference_active": true, "delta_db": -1.4101011208555294}], "active_band_count": 3, "active_band_mae_db": 0.5940722971087222}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.291047760123656e-05, "log_rms_mae_db": 0.8766807208926427, "rms_envelope_nrmse": 0.08024650710609413, "envelope_correlation": 0.9951678719395395, "rms_gain_error_db": -0.10262410345385922, "reference_mean_power": 5.2408520364582374e-05, "output_mean_power": 5.118462056879188e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.33304878770606e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.583453225376265e-05, "output_power": 2.558994828364342e-05, "reference_active": true, "delta_db": -0.0413119465433148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0765400661096766e-06, "output_power": 2.2213198980534586e-06, "reference_active": true, "delta_db": 0.2927079184396392}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.119613733354675e-08, "output_power": 4.858292716468856e-08, "reference_active": true, "delta_db": -1.0024033269509776}], "active_band_count": 3, "active_band_mae_db": 0.4454743973113105}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.3449414786854306e-05, "log_rms_mae_db": 1.4560556858763285, "rms_envelope_nrmse": 0.17507993920513618, "envelope_correlation": 0.9768006256080431, "rms_gain_error_db": -0.43259985260740624, "reference_mean_power": 6.886139457761324e-06, "output_mean_power": 6.233268411608614e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.997833564030985e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1829042042252175e-06, "output_power": 1.905316870369876e-06, "reference_active": true, "delta_db": -0.5906746443084119}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.111376843686143e-07, "output_power": 6.195374931095495e-07, "reference_active": true, "delta_db": -1.170269845101293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6537364416157584e-07, "output_power": 2.1279117829290722e-07, "reference_active": true, "delta_db": 1.0948732281500797}], "active_band_count": 3, "active_band_mae_db": 0.9519392391865948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.0291390653962812e-05, "log_rms_mae_db": 1.006550952876402, "rms_envelope_nrmse": 0.10906938203113413, "envelope_correlation": 0.9915871463073115, "rms_gain_error_db": -0.43259989210837285, "reference_mean_power": 6.9207717017349184e-06, "output_mean_power": 6.26461713484312e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.443349031906855e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9318994555375575e-06, "output_power": 1.719477123337707e-06, "reference_active": true, "delta_db": -0.5058811793500777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.056544990756342e-07, "output_power": 3.34946767469815e-07, "reference_active": true, "delta_db": -0.8318050623500619}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.969546554060793e-09, "output_power": 1.0831955453153501e-08, "reference_active": true, "delta_db": 1.3327325338988658}], "active_band_count": 3, "active_band_mae_db": 0.8901395918663351}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9159953385971497, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1eda5a21c7fd22b5f1eed746c02d8f4cf976209b22d742d8d19d4ac3e408af27", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826182985325551, "standard_deviation": 0.002215167418108788, "minimum": 0.9787819644331619, "maximum": 0.9852745988176701, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5067185993363463, "standard_deviation": 0.03367723732786204, "minimum": 1.457742179663484, "maximum": 1.566213378548029, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5825620890706633, "standard_deviation": 0.35501881364895965, "minimum": 0.2060891191532324, "maximum": 1.3761883698081618, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15612647366250193, "standard_deviation": 0.009948430990263555, "minimum": 0.14282907825027427, "maximum": 0.1745598794183156, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937025332490156, "standard_deviation": 0.001186337325678648, "minimum": 0.9917721989796084, "maximum": 0.9951678719395395, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9167679435056915, "standard_deviation": 0.041134481521142345, "minimum": 0.865974587430166, "maximum": 0.9632511738283152, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5492166999682866, "standard_deviation": 0.3467602096138603, "minimum": 0.19329509199432737, "maximum": 1.2716372433137475, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09475057550103677, "standard_deviation": 0.010021671698638358, "minimum": 0.08024650710609413, "maximum": 0.11315373560441327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9771997790000668, "standard_deviation": 0.002041652702573768, "minimum": 0.9738713883901852, "maximum": 0.9797891544129027, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4749178319582437, "standard_deviation": 0.02300764500639169, "minimum": 1.4426435948935852, "maximum": 1.5128478310295774, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5758674824596051, "standard_deviation": 0.19343194166293953, "minimum": 0.36082127602465613, "maximum": 0.9519392391865948, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1713554156600553, "standard_deviation": 0.007404978917522029, "minimum": 0.16146120825302177, "maximum": 0.1828588769574228, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9915231772026372, "standard_deviation": 0.0010868183862765206, "minimum": 0.989753904618617, "maximum": 0.9929676540481998, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.998928316982578, "standard_deviation": 0.015460751022665517, "minimum": 0.9839434979038775, "maximum": 1.0303270217523326, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6662838094947406, "standard_deviation": 0.3001387755143054, "minimum": 0.19528710537337388, "maximum": 1.0251821472122806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1041575934683042, "standard_deviation": 0.00630351945206454, "minimum": 0.09449305990510731, "maximum": 0.11316707661471717, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937575804002293, "standard_deviation": 0.0007302605642862951, "minimum": 0.992361523666661, "maximum": 0.9945901740585487, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3550937967253955, "standard_deviation": 0.028589744140933418, "minimum": 1.3247155794857464, "maximum": 1.3943571588673769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47207512158849607, "standard_deviation": 0.12842323894980498, "minimum": 0.2843406375630083, "maximum": 0.6315884357377906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09068488502551822, "standard_deviation": 0.005121237905542459, "minimum": 0.0846482333339214, "maximum": 0.10049067899603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978582904407499, "standard_deviation": 0.00045954077772236956, "minimum": 0.9972637485967426, "maximum": 0.9985884319785512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8291087922829877, "standard_deviation": 0.038338544964664385, "minimum": 0.7836115599220234, "maximum": 0.88958580574161, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5772232703477486, "standard_deviation": 0.2388526506679567, "minimum": 0.16554137677060934, "maximum": 0.8645097099963427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05390608917429613, "standard_deviation": 0.00570005453677428, "minimum": 0.04499985450523699, "maximum": 0.06131334183263683, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2009067333970238, "standard_deviation": 0.013295919927020085, "minimum": 0.18576589606785646, "maximum": 0.2207315547120743, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.324985193645393, "standard_deviation": 0.007877906357403158, "minimum": 5.313437107912168, "maximum": 5.337288805209989, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.24590766261755442, "standard_deviation": 0.006310990166134918, "minimum": 0.23624431315817265, "maximum": 0.2561972920578169, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10294247930668404, "standard_deviation": 0.021148322817570706, "minimum": -0.13012784725879242, "maximum": -0.07902833978565867, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955158583451726, "standard_deviation": 0.0006227836818025574, "minimum": 0.9945789021396713, "maximum": 0.9962663940596878, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7707346540805942, "standard_deviation": 0.025318527538905275, "minimum": 0.7422868960141772, "maximum": 0.7972804071737061, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.279806563534738, "standard_deviation": 0.15948867309559214, "minimum": 0.14205338208929677, "maximum": 0.5486195459190187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06766727633715089, "standard_deviation": 0.004484660463460545, "minimum": 0.06251773531534174, "maximum": 0.07427228871897727, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985395899977842, "standard_deviation": 0.0003974696026130615, "minimum": 0.9978433012804212, "maximum": 0.998992049220059, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4269999340188657, "standard_deviation": 0.019761832690394162, "minimum": 0.40004025337945726, "maximum": 0.446522684162291, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.32699648365728573, "standard_deviation": 0.1915098358284219, "minimum": 0.1079938606760781, "maximum": 0.6460875349650824, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038485759465958225, "standard_deviation": 0.004865575820780453, "minimum": 0.0322915802886629, "maximum": 0.0462521432836756, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0023", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.2799899578094482, "predicted_mos": 3.2799899578094482, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4f8c1c41f21df802cc494aa84673e91d63fee2cf878983f6eb60a5d529980bf9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.176897500175983, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7575882295279549, "全帯域_10ms_rms_envelope_nrmse": 0.0628143741687676, "全帯域_10ms_envelope_correlation": 0.9961601538895931, "全帯域_10ms_modulation_mae_db": 0.5076632018790252, "全帯域_40ms_log_rms_mae_db": 0.4457909469791468, "全帯域_40ms_rms_envelope_nrmse": 0.03636263007566104, "全帯域_40ms_envelope_correlation": 0.9987197486770477, "全帯域_40ms_modulation_mae_db": 0.5926878869656648, "300–1000 Hz_10ms_log_rms_mae_db": 1.2992026161025856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08490046061840407, "300–1000 Hz_10ms_envelope_correlation": 0.9944915991523962, "300–1000 Hz_10ms_modulation_mae_db": 0.8226283209785669, "300–1000 Hz_40ms_log_rms_mae_db": 0.7886057799384689, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05243379275998184, "300–1000 Hz_40ms_envelope_correlation": 0.9978997923319162, "300–1000 Hz_40ms_modulation_mae_db": 1.0678053311702138, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6452500193984234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16020950295461092, "1000–3000 Hz_10ms_envelope_correlation": 0.9823581868097605, "1000–3000 Hz_10ms_modulation_mae_db": 1.1057687913307723, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9466121028798098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09836749084067528, "1000–3000 Hz_40ms_envelope_correlation": 0.9941733625276111, "1000–3000 Hz_40ms_modulation_mae_db": 1.0829199846435928, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7141199465354235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20447129238019202, "3000–6000 Hz_10ms_envelope_correlation": 0.9673183480083613, "3000–6000 Hz_10ms_modulation_mae_db": 0.7359665113573156, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1373838739212847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12156005101372765, "3000–6000 Hz_40ms_envelope_correlation": 0.9885434107919246, "3000–6000 Hz_40ms_modulation_mae_db": 0.6140629482710206, "mr_stft_db": 5.420009969151484, "band_mae_db": 0.22266718384524978, "rms_gain_error_db": -0.10084130567494962, "output_peak": 0.24340173916946145}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7895614040309127, "全帯域_10ms_rms_envelope_nrmse": 0.06916276916504825, "全帯域_10ms_envelope_correlation": 0.9955034746248452, "全帯域_10ms_modulation_mae_db": 0.4757683790365797, "全帯域_40ms_log_rms_mae_db": 0.44716339040818603, "全帯域_40ms_rms_envelope_nrmse": 0.040090062760851516, "全帯域_40ms_envelope_correlation": 0.9986383391455163, "全帯域_40ms_modulation_mae_db": 0.5096712249950877, "300–1000 Hz_10ms_log_rms_mae_db": 1.4443172772527377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08987935017375982, "300–1000 Hz_10ms_envelope_correlation": 0.9940663433647506, "300–1000 Hz_10ms_modulation_mae_db": 0.6404549788091777, "300–1000 Hz_40ms_log_rms_mae_db": 0.9038685787970328, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05415495490879822, "300–1000 Hz_40ms_envelope_correlation": 0.9980724410217509, "300–1000 Hz_40ms_modulation_mae_db": 0.8118684621028649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6061806428635543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17023620014458016, "1000–3000 Hz_10ms_envelope_correlation": 0.9802789543169118, "1000–3000 Hz_10ms_modulation_mae_db": 0.6522619779740264, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9870793438842372, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11276968096290876, "1000–3000 Hz_40ms_envelope_correlation": 0.9923008169163793, "1000–3000 Hz_40ms_modulation_mae_db": 0.6384651422166477, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6516748848269684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18659710442975802, "3000–6000 Hz_10ms_envelope_correlation": 0.9726817649785765, "3000–6000 Hz_10ms_modulation_mae_db": 0.495831939174782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.116567139108742, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.112219191965114, "3000–6000 Hz_40ms_envelope_correlation": 0.9898048025283386, "3000–6000 Hz_40ms_modulation_mae_db": 0.6149470148154478, "mr_stft_db": 5.425528357881633, "band_mae_db": 0.24541916394665073, "rms_gain_error_db": -0.16399716845361345, "output_peak": 0.2534014914718217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7680119734339882, "全帯域_10ms_rms_envelope_nrmse": 0.06422848600533157, "全帯域_10ms_envelope_correlation": 0.9960023213728182, "全帯域_10ms_modulation_mae_db": 0.3421233501050813, "全帯域_40ms_log_rms_mae_db": 0.41873669060672714, "全帯域_40ms_rms_envelope_nrmse": 0.03259810516205753, "全帯域_40ms_envelope_correlation": 0.999025952825018, "全帯域_40ms_modulation_mae_db": 0.5576418032735854, "300–1000 Hz_10ms_log_rms_mae_db": 1.4241379996852856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0873258326180682, "300–1000 Hz_10ms_envelope_correlation": 0.9942269098943187, "300–1000 Hz_10ms_modulation_mae_db": 0.3963406548274652, "300–1000 Hz_40ms_log_rms_mae_db": 0.8007438603486867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04914803397151455, "300–1000 Hz_40ms_envelope_correlation": 0.9982664545745175, "300–1000 Hz_40ms_modulation_mae_db": 0.8365135734874035, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6186390594486335, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16005707445719913, "1000–3000 Hz_10ms_envelope_correlation": 0.9817042392277676, "1000–3000 Hz_10ms_modulation_mae_db": 0.4830702229423863, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9998695772503448, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09425719922798922, "1000–3000 Hz_40ms_envelope_correlation": 0.9938642301217488, "1000–3000 Hz_40ms_modulation_mae_db": 0.5892757145757154, "3000–6000 Hz_10ms_log_rms_mae_db": 1.650006085879676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19300489624215622, "3000–6000 Hz_10ms_envelope_correlation": 0.9709446434847363, "3000–6000 Hz_10ms_modulation_mae_db": 0.9263561159774459, "3000–6000 Hz_40ms_log_rms_mae_db": 1.083177008048112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11185896949862598, "3000–6000 Hz_40ms_envelope_correlation": 0.9904232822122604, "3000–6000 Hz_40ms_modulation_mae_db": 0.8082264101203557, "mr_stft_db": 5.429846537231715, "band_mae_db": 0.17617450090887457, "rms_gain_error_db": -0.10856648749446685, "output_peak": 0.2339905374003855}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7697603742633856, "全帯域_10ms_rms_envelope_nrmse": 0.07120136426154455, "全帯域_10ms_envelope_correlation": 0.9950995920749718, "全帯域_10ms_modulation_mae_db": 0.34122079607585903, "全帯域_40ms_log_rms_mae_db": 0.45636240737465666, "全帯域_40ms_rms_envelope_nrmse": 0.04098774146489558, "全帯域_40ms_envelope_correlation": 0.9984233533655005, "全帯域_40ms_modulation_mae_db": 0.23848505680997475, "300–1000 Hz_10ms_log_rms_mae_db": 1.4157591000756697, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09702239223081932, "300–1000 Hz_10ms_envelope_correlation": 0.9928790860401545, "300–1000 Hz_10ms_modulation_mae_db": 0.5437038494886081, "300–1000 Hz_40ms_log_rms_mae_db": 0.8906134965944394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05879189756916586, "300–1000 Hz_40ms_envelope_correlation": 0.9974841546707222, "300–1000 Hz_40ms_modulation_mae_db": 0.7104106808507386, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6434074281224993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17257643460911679, "1000–3000 Hz_10ms_envelope_correlation": 0.9787091075582746, "1000–3000 Hz_10ms_modulation_mae_db": 0.8470855682910642, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9859335344793015, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10258833896741831, "1000–3000 Hz_40ms_envelope_correlation": 0.9927353539533404, "1000–3000 Hz_40ms_modulation_mae_db": 0.8690042796594328, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6668153547490667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18804744089413497, "3000–6000 Hz_10ms_envelope_correlation": 0.9726894102182744, "3000–6000 Hz_10ms_modulation_mae_db": 0.21121548330450465, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1152720481679093, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11387745405344867, "3000–6000 Hz_40ms_envelope_correlation": 0.9893428693604235, "3000–6000 Hz_40ms_modulation_mae_db": 0.12034164135587144, "mr_stft_db": 5.432132268581088, "band_mae_db": 0.1884626121036362, "rms_gain_error_db": -0.12998359119017666, "output_peak": 0.2481040219075438}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7737938824180062, "全帯域_10ms_rms_envelope_nrmse": 0.06611884139520338, "全帯域_10ms_envelope_correlation": 0.9956794330211493, "全帯域_10ms_modulation_mae_db": 0.16864542268118407, "全帯域_40ms_log_rms_mae_db": 0.4388234577460682, "全帯域_40ms_rms_envelope_nrmse": 0.03697219712658218, "全帯域_40ms_envelope_correlation": 0.9986065015439632, "全帯域_40ms_modulation_mae_db": 0.32982202343700423, "300–1000 Hz_10ms_log_rms_mae_db": 1.4211839063436902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09478062904620124, "300–1000 Hz_10ms_envelope_correlation": 0.9930864906061806, "300–1000 Hz_10ms_modulation_mae_db": 0.30494270827126696, "300–1000 Hz_40ms_log_rms_mae_db": 0.8762768112730962, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05660000897050664, "300–1000 Hz_40ms_envelope_correlation": 0.9975257538025832, "300–1000 Hz_40ms_modulation_mae_db": 0.35924814636473235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6543173540471612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17864463442606981, "1000–3000 Hz_10ms_envelope_correlation": 0.9768242532032547, "1000–3000 Hz_10ms_modulation_mae_db": 0.6002591734910506, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0000018123032695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10442425601607964, "1000–3000 Hz_40ms_envelope_correlation": 0.9917819747091985, "1000–3000 Hz_40ms_modulation_mae_db": 0.5489128377680715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6442055712378492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1771622565056952, "3000–6000 Hz_10ms_envelope_correlation": 0.9754531608890878, "3000–6000 Hz_10ms_modulation_mae_db": 1.1073154674343801, "3000–6000 Hz_40ms_log_rms_mae_db": 1.066124755506595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10750259490478149, "3000–6000 Hz_40ms_envelope_correlation": 0.9907766647575109, "3000–6000 Hz_40ms_modulation_mae_db": 0.8047250244653777, "mr_stft_db": 5.433414300437801, "band_mae_db": 0.16529042847888, "rms_gain_error_db": -0.07199613253724413, "output_peak": 0.23107775924566304}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7665530636108987, "全帯域_10ms_rms_envelope_nrmse": 0.07059309264752715, "全帯域_10ms_envelope_correlation": 0.9952761987922987, "全帯域_10ms_modulation_mae_db": 0.40994228593807575, "全帯域_40ms_log_rms_mae_db": 0.4762010029971076, "全帯域_40ms_rms_envelope_nrmse": 0.04164017413838583, "全帯域_40ms_envelope_correlation": 0.9984702561399876, "全帯域_40ms_modulation_mae_db": 0.584472854819955, "300–1000 Hz_10ms_log_rms_mae_db": 1.4301009893008312, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09840963931453857, "300–1000 Hz_10ms_envelope_correlation": 0.9927448408717702, "300–1000 Hz_10ms_modulation_mae_db": 0.46150494007820075, "300–1000 Hz_40ms_log_rms_mae_db": 0.8751238301954385, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05889411196410377, "300–1000 Hz_40ms_envelope_correlation": 0.9975746856735026, "300–1000 Hz_40ms_modulation_mae_db": 0.7659147854650638, "1000–3000 Hz_10ms_log_rms_mae_db": 1.641102570458401, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16881836516597606, "1000–3000 Hz_10ms_envelope_correlation": 0.9793972546500174, "1000–3000 Hz_10ms_modulation_mae_db": 0.6561512283496865, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9760906165576884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09965278295260233, "1000–3000 Hz_40ms_envelope_correlation": 0.9927797301788951, "1000–3000 Hz_40ms_modulation_mae_db": 0.33554852615517733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6249233012412536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18876995955004333, "3000–6000 Hz_10ms_envelope_correlation": 0.9719497141875661, "3000–6000 Hz_10ms_modulation_mae_db": 1.137281721897365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0757900531957132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11092256707392051, "3000–6000 Hz_40ms_envelope_correlation": 0.9896978983769218, "3000–6000 Hz_40ms_modulation_mae_db": 0.9071507105491746, "mr_stft_db": 5.432783575597273, "band_mae_db": 0.17601870008898987, "rms_gain_error_db": -0.15665398106637257, "output_peak": 0.2435083495004095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7598943781834182, "全帯域_10ms_rms_envelope_nrmse": 0.06535004541661378, "全帯域_10ms_envelope_correlation": 0.9958086195916993, "全帯域_10ms_modulation_mae_db": 0.34752162249065083, "全帯域_40ms_log_rms_mae_db": 0.43063445060810746, "全帯域_40ms_rms_envelope_nrmse": 0.036426676538807615, "全帯域_40ms_envelope_correlation": 0.9986856520345999, "全帯域_40ms_modulation_mae_db": 0.5111021861504268, "300–1000 Hz_10ms_log_rms_mae_db": 1.391954867945808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08841148259421007, "300–1000 Hz_10ms_envelope_correlation": 0.9940341414584992, "300–1000 Hz_10ms_modulation_mae_db": 0.1646863377571622, "300–1000 Hz_40ms_log_rms_mae_db": 0.8242650975328749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0525698807364542, "300–1000 Hz_40ms_envelope_correlation": 0.9979227535345732, "300–1000 Hz_40ms_modulation_mae_db": 0.37937039440713144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6395934857604306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1664212541718406, "1000–3000 Hz_10ms_envelope_correlation": 0.9804012454226068, "1000–3000 Hz_10ms_modulation_mae_db": 1.1227143573942113, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9912019881959423, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10345949957484381, "1000–3000 Hz_40ms_envelope_correlation": 0.9928079321340546, "1000–3000 Hz_40ms_modulation_mae_db": 0.9765849851955409, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6305673525913666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18701736366307867, "3000–6000 Hz_10ms_envelope_correlation": 0.9725532734311668, "3000–6000 Hz_10ms_modulation_mae_db": 1.022237355902318, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0713447504510367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11102861091467157, "3000–6000 Hz_40ms_envelope_correlation": 0.9901268756629159, "3000–6000 Hz_40ms_modulation_mae_db": 0.6024417964690044, "mr_stft_db": 5.420403008112931, "band_mae_db": 0.18460597880797644, "rms_gain_error_db": -0.0878641100181797, "output_peak": 0.2344372950229772}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7367578199950942, "全帯域_10ms_rms_envelope_nrmse": 0.06481373978456849, "全帯域_10ms_envelope_correlation": 0.9958870180552358, "全帯域_10ms_modulation_mae_db": 0.10184924392296006, "全帯域_40ms_log_rms_mae_db": 0.40674973620651117, "全帯域_40ms_rms_envelope_nrmse": 0.03497424204837593, "全帯域_40ms_envelope_correlation": 0.9988050187661918, "全帯域_40ms_modulation_mae_db": 0.19273398227660668, "300–1000 Hz_10ms_log_rms_mae_db": 1.3322215030719335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08674826162921612, "300–1000 Hz_10ms_envelope_correlation": 0.994302382197861, "300–1000 Hz_10ms_modulation_mae_db": 0.398262655045287, "300–1000 Hz_40ms_log_rms_mae_db": 0.7938689808335136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046493605784555435, "300–1000 Hz_40ms_envelope_correlation": 0.998470065662441, "300–1000 Hz_40ms_modulation_mae_db": 0.7346624730414234, "1000–3000 Hz_10ms_log_rms_mae_db": 1.608304748992737, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16700485231122805, "1000–3000 Hz_10ms_envelope_correlation": 0.980131614957271, "1000–3000 Hz_10ms_modulation_mae_db": 1.2448828006122683, "1000–3000 Hz_40ms_log_rms_mae_db": 0.938468112895004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10227954219910844, "1000–3000 Hz_40ms_envelope_correlation": 0.9928024777565212, "1000–3000 Hz_40ms_modulation_mae_db": 1.393953938990533, "3000–6000 Hz_10ms_log_rms_mae_db": 1.619821944767306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19636026726052666, "3000–6000 Hz_10ms_envelope_correlation": 0.9696547362240779, "3000–6000 Hz_10ms_modulation_mae_db": 0.4081891057831228, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0456151571742178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11421673720912612, "3000–6000 Hz_40ms_envelope_correlation": 0.9891446489570169, "3000–6000 Hz_40ms_modulation_mae_db": 0.6112749324470513, "mr_stft_db": 5.426167262327861, "band_mae_db": 0.19606256707400072, "rms_gain_error_db": -0.0861346787159625, "output_peak": 0.23526848183466945}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.254206405456745, "spectral_convergence": 0.18834472659933918, "spectral_convergence_status": "ok", "amplitude_floor": 7.733796592545008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.25538915373075, "spectral_convergence": 0.1958863411895917, "spectral_convergence_status": "ok", "amplitude_floor": 4.871370963739547e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.750434348266957, "spectral_convergence": 0.20033125628182907, "spectral_convergence_status": "ok", "amplitude_floor": 4.7483837933045734e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001233024392514864, "output_power": 0.00012712485917522787, "delta_db": 0.13258816648602537, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000393494821723947, "output_power": 0.00038921076665065796, "delta_db": -0.047541759912082746, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000358215617892152, "output_power": 0.00034352152383952206, "delta_db": -0.18190563191829923, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.947495410559829e-05, "output_power": 5.360542867582745e-05, "delta_db": -0.45125342221908815, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.0477761937191392e-05, "output_power": 9.443126446663979e-06, "delta_db": -0.45152721644005145, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.39686924640271e-06, "output_power": 9.552166293507342e-06, "delta_db": 0.07118690609595177, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22266718384524978, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10084130567494962, "pitch": null, "output_rms_dbfs": -30.303707807205473, "output_peak": 0.24340173916946145, "reference_rms_dbfs": -30.202866501530522, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.420009969151484}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.933633747331572e-05, "log_rms_mae_db": 0.7575882295279549, "rms_envelope_nrmse": 0.0628143741687676, "envelope_correlation": 0.9961601538895931, "rms_gain_error_db": -0.1008412165177105, "reference_mean_power": 0.0009551583992700257, "output_mean_power": 0.0009332355569956513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1194144575506277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002166069295810575, "output_power": 0.00020554671633415446, "reference_active": true, "delta_db": -0.22761803047513107}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.38201712334803e-05, "output_power": 2.1279841019538768e-05, "reference_active": true, "delta_db": -0.48976500077418317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.302931519380675e-06, "output_power": 6.89716790399437e-06, "reference_active": true, "delta_db": -0.8056065743877615}], "active_band_count": 3, "active_band_mae_db": 0.5076632018790252}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.652181544798992e-05, "log_rms_mae_db": 0.4457909469791468, "rms_envelope_nrmse": 0.03636263007566104, "envelope_correlation": 0.9987197486770477, "rms_gain_error_db": -0.10084101832227153, "reference_mean_power": 0.0009599620519447541, "output_mean_power": 0.0009379289988028574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.73215395386014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001986653950491428, "output_power": 0.00018844552491443394, "reference_active": true, "delta_db": -0.22936396640794207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916927878865493e-05, "output_power": 9.718695112781882e-06, "reference_active": true, "delta_db": -0.5049248321441006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.81959177520348e-07, "output_power": 4.5763057768905345e-07, "reference_active": true, "delta_db": -1.0437748623449516}], "active_band_count": 3, "active_band_mae_db": 0.5926878869656648}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.027389787958093e-05, "log_rms_mae_db": 1.2992026161025856, "rms_envelope_nrmse": 0.08490046061840407, "envelope_correlation": 0.9944915991523962, "rms_gain_error_db": -0.13549881744972078, "reference_mean_power": 0.0006127358569067375, "output_mean_power": 0.0005939137964920905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.366962347376557e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001936423107817767, "output_power": 0.0001819140691547657, "reference_active": true, "delta_db": -0.27133968035456657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9545968551344243e-05, "output_power": 1.6673677720670207e-05, "reference_active": true, "delta_db": -0.6902579274771505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.863910891753386e-06, "output_power": 4.852247012321381e-06, "reference_active": true, "delta_db": -1.5062873551039835}], "active_band_count": 3, "active_band_mae_db": 0.8226283209785669}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.943169310162246e-05, "log_rms_mae_db": 0.7886057799384689, "rms_envelope_nrmse": 0.05243379275998184, "envelope_correlation": 0.9978997923319162, "rms_gain_error_db": -0.13549882428118398, "reference_mean_power": 0.000615817475654304, "output_mean_power": 0.0005969007529330503}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.097820742933078e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018026482035346189, "output_power": 0.0001690299109686915, "reference_active": true, "delta_db": -0.27947417378385914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.739141870245847e-06, "output_power": 7.6075238746787915e-06, "reference_active": true, "delta_db": -0.6022546571436037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.763892611677605e-07, "output_power": 3.377125223428344e-07, "reference_active": true, "delta_db": -2.3216871625831788}], "active_band_count": 3, "active_band_mae_db": 1.0678053311702138}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.05554473589972e-05, "log_rms_mae_db": 1.6452500193984234, "rms_envelope_nrmse": 0.16020950295461092, "envelope_correlation": 0.9823581868097605, "rms_gain_error_db": -0.43340992706684783, "reference_mean_power": 5.214627438247735e-05, "output_mean_power": 4.719351149238118e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.666296423769008e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.72355480524535e-05, "output_power": 2.342121673128681e-05, "reference_active": true, "delta_db": -0.6552666596608627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.163900077209868e-06, "output_power": 3.7340274640649626e-06, "reference_active": true, "delta_db": -0.4732279112884742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1939677607224186e-06, "output_power": 7.212925058060566e-07, "reference_active": true, "delta_db": -2.18881180304298}], "active_band_count": 3, "active_band_mae_db": 1.1057687913307723}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.291047760123656e-05, "log_rms_mae_db": 0.9466121028798098, "rms_envelope_nrmse": 0.09836749084067528, "envelope_correlation": 0.9941733625276111, "rms_gain_error_db": -0.43341069003163873, "reference_mean_power": 5.2408520364582374e-05, "output_mean_power": 4.743084147237104e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.33304878770606e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.583453225376265e-05, "output_power": 2.2304855754992634e-05, "reference_active": true, "delta_db": -0.6380118383508397}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0765400661096766e-06, "output_power": 1.8932667105176666e-06, "reference_active": true, "delta_db": -0.4012851636406687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.119613733354675e-08, "output_power": 3.679406026833081e-08, "reference_active": true, "delta_db": -2.20946295193927}], "active_band_count": 3, "active_band_mae_db": 1.0829199846435928}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.3449414786854306e-05, "log_rms_mae_db": 1.7141199465354235, "rms_envelope_nrmse": 0.20447129238019202, "envelope_correlation": 0.9673183480083613, "rms_gain_error_db": -0.3658498073665046, "reference_mean_power": 6.886139457761324e-06, "output_mean_power": 6.32981231603525e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.997833564030985e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1829042042252175e-06, "output_power": 1.987022292710126e-06, "reference_active": true, "delta_db": -0.4083193775415914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.111376843686143e-07, "output_power": 6.354811839577495e-07, "reference_active": true, "delta_db": -1.0599188260810746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6537364416157584e-07, "output_power": 1.9607960783016886e-07, "reference_active": true, "delta_db": 0.7396613304492807}], "active_band_count": 3, "active_band_mae_db": 0.7359665113573156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.0291390653962812e-05, "log_rms_mae_db": 1.1373838739212847, "rms_envelope_nrmse": 0.12156005101372765, "envelope_correlation": 0.9885434107919246, "rms_gain_error_db": -0.3658499049162993, "reference_mean_power": 6.9207717017349184e-06, "output_mean_power": 6.361646498569321e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.443349031906855e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9318994555375575e-06, "output_power": 1.7900151874369599e-06, "reference_active": true, "delta_db": -0.33127804311186787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.056544990756342e-07, "output_power": 3.443412818854521e-07, "reference_active": true, "delta_db": -0.7116720517010349}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.969546554060793e-09, "output_power": 9.579823042316517e-09, "reference_active": true, "delta_db": 0.799238750000159}], "active_band_count": 3, "active_band_mae_db": 0.6140629482710206}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9102567077314064, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "140d7b10bb881bb625326055822560ea3f3de31486a2cc218587bc706344166c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.979975607018233, "standard_deviation": 0.001725105591047608, "minimum": 0.9768242532032547, "maximum": 0.9823581868097605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.63209941363648, "standard_deviation": 0.018325681932833567, "minimum": 1.6061806428635543, "maximum": 1.6543173540471612, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8390242650481833, "standard_deviation": 0.2849695709969899, "minimum": 0.4830702229423863, "maximum": 1.2448828006122683, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1679960397800777, "standard_deviation": 0.006178151025602292, "minimum": 0.16005707445719913, "maximum": 0.17864463442606981, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929057347872186, "standard_deviation": 0.0007763918286505665, "minimum": 0.9917819747091985, "maximum": 0.9941733625276111, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9781571360556996, "standard_deviation": 0.023401600969912173, "minimum": 0.938468112895004, "maximum": 1.0000018123032695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.804333176150589, "standard_deviation": 0.3418179977242453, "minimum": 0.33554852615517733, "maximum": 1.393953938990533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10222484884270322, "standard_deviation": 0.0053807663992578416, "minimum": 0.09425719922798922, "maximum": 0.11276968096290876, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9716556314277309, "standard_deviation": 0.0024129483278112944, "minimum": 0.9673183480083613, "maximum": 0.9754531608890878, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6502668052286138, "standard_deviation": 0.030111977975730866, "minimum": 1.619821944767306, "maximum": 1.7141199465354235, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7555492126039043, "standard_deviation": 0.3493977609859698, "minimum": 0.21121548330450465, "maximum": 1.137281721897365, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19017882261569813, "standard_deviation": 0.008011187234626176, "minimum": 0.1771622565056952, "maximum": 0.20447129238019202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9897325565809141, "standard_deviation": 0.000721182679473745, "minimum": 0.9885434107919246, "maximum": 0.9907766647575109, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0889093481967014, "standard_deviation": 0.03097635123616496, "minimum": 1.0456151571742178, "maximum": 1.1373838739212847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.635396309811663, "standard_deviation": 0.23911006126091353, "minimum": 0.12034164135587144, "maximum": 0.9071507105491746, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11289827207917699, "standard_deviation": 0.004066891352513779, "minimum": 0.10750259490478149, "maximum": 0.12156005101372765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937289741982414, "standard_deviation": 0.0007039573825366293, "minimum": 0.9927448408717702, "maximum": 0.9944915991523962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3948597824723177, "standard_deviation": 0.051764890857289615, "minimum": 1.2992026161025856, "maximum": 1.4443172772527377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.46656555565696683, "standard_deviation": 0.20348533834979274, "minimum": 0.1646863377571622, "maximum": 0.8226283209785669, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09093475602815218, "standard_deviation": 0.0051013794901307394, "minimum": 0.08490046061840407, "maximum": 0.09840963931453857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979020126590008, "standard_deviation": 0.0003602185935433992, "minimum": 0.9974841546707222, "maximum": 0.998470065662441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8441708044391938, "standard_deviation": 0.04722214287280177, "minimum": 0.7886057799384689, "maximum": 0.9038685787970328, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7082242308611965, "standard_deviation": 0.23617788828558367, "minimum": 0.35924814636473235, "maximum": 1.0678053311702138, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.053635785833135066, "standard_deviation": 0.004424335800226525, "minimum": 0.046493605784555435, "maximum": 0.05889411196410377, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1943376419067823, "standard_deviation": 0.026868211108513715, "minimum": 0.16529042847888, "maximum": 0.24541916394665073, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.427535659915224, "standard_deviation": 0.005371625388676272, "minimum": 5.420009969151484, "maximum": 5.433414300437801, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.24039870944411645, "standard_deviation": 0.007900585043431685, "minimum": 0.23107775924566304, "maximum": 0.2534014914718217, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1132546818938707, "standard_deviation": 0.033799133971598984, "minimum": -0.16399716845361345, "maximum": -0.07199613253724413, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9956771014278264, "standard_deviation": 0.00036348735292978523, "minimum": 0.9950995920749718, "maximum": 0.9961601538895931, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7652401406829574, "standard_deviation": 0.01508500294208198, "minimum": 0.7367578199950942, "maximum": 0.7895614040309127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.336841787766177, "standard_deviation": 0.14015138676673625, "minimum": 0.10184924392296006, "maximum": 0.5076632018790252, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0667853391055756, "standard_deviation": 0.0031247854843347964, "minimum": 0.0628143741687676, "maximum": 0.07120136426154455, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986718528122281, "standard_deviation": 0.0001902598111457232, "minimum": 0.9984233533655005, "maximum": 0.999025952825018, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4400577603658139, "standard_deviation": 0.021744850516338107, "minimum": 0.40674973620651117, "maximum": 0.4762010029971076, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43957712734103815, "standard_deviation": 0.16116615328372388, "minimum": 0.19273398227660668, "maximum": 0.5926878869656648, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037506478664452156, "standard_deviation": 0.0031429219217900704, "minimum": 0.03259810516205753, "maximum": 0.04164017413838583, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0023", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.9340453147888184, "predicted_mos": 2.9340453147888184, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8977b2bb0879b5004667501f3bf402b2e3065413143655051048c654129ceea2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.222339800093323, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7520397894723878, "全帯域_10ms_rms_envelope_nrmse": 0.06547839483164691, "全帯域_10ms_envelope_correlation": 0.99584663350445, "全帯域_10ms_modulation_mae_db": 0.29165237552312756, "全帯域_40ms_log_rms_mae_db": 0.4304980818638673, "全帯域_40ms_rms_envelope_nrmse": 0.03415518012332344, "全帯域_40ms_envelope_correlation": 0.9989250082751707, "全帯域_40ms_modulation_mae_db": 0.26189074036608156, "300–1000 Hz_10ms_log_rms_mae_db": 1.3688417795389816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08633612554487756, "300–1000 Hz_10ms_envelope_correlation": 0.9943808964637634, "300–1000 Hz_10ms_modulation_mae_db": 0.32252002153566134, "300–1000 Hz_40ms_log_rms_mae_db": 0.783581181194565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049076317131529504, "300–1000 Hz_40ms_envelope_correlation": 0.9982880914327253, "300–1000 Hz_40ms_modulation_mae_db": 0.4190312478336254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5662657108365288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15023849744393689, "1000–3000 Hz_10ms_envelope_correlation": 0.9837735205696667, "1000–3000 Hz_10ms_modulation_mae_db": 0.20648989632007217, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9193024987269044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08962725378212157, "1000–3000 Hz_40ms_envelope_correlation": 0.9942137166746474, "1000–3000 Hz_40ms_modulation_mae_db": 0.48553708146172125, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6290553412908286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1892964853678217, "3000–6000 Hz_10ms_envelope_correlation": 0.9719452268881708, "3000–6000 Hz_10ms_modulation_mae_db": 0.5288820126814153, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0133816671684452, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1006174498510173, "3000–6000 Hz_40ms_envelope_correlation": 0.9915068767694085, "3000–6000 Hz_40ms_modulation_mae_db": 0.3048145769496649, "mr_stft_db": 5.404485572731119, "band_mae_db": 0.14419250996815092, "rms_gain_error_db": -0.11012348080008101, "output_peak": 0.25377700094514666}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7345539157840753, "全帯域_10ms_rms_envelope_nrmse": 0.06449238914221966, "全帯域_10ms_envelope_correlation": 0.9961307211709215, "全帯域_10ms_modulation_mae_db": 0.3625750925776288, "全帯域_40ms_log_rms_mae_db": 0.4092656205149147, "全帯域_40ms_rms_envelope_nrmse": 0.0372955040123906, "全帯域_40ms_envelope_correlation": 0.9988637274664381, "全帯域_40ms_modulation_mae_db": 0.4277056365602377, "300–1000 Hz_10ms_log_rms_mae_db": 1.3353629359244699, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08766068840719145, "300–1000 Hz_10ms_envelope_correlation": 0.9944733146716438, "300–1000 Hz_10ms_modulation_mae_db": 0.44828983717113885, "300–1000 Hz_40ms_log_rms_mae_db": 0.808517763589143, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053483849919898276, "300–1000 Hz_40ms_envelope_correlation": 0.9982414571062679, "300–1000 Hz_40ms_modulation_mae_db": 0.7167005644295713, "1000–3000 Hz_10ms_log_rms_mae_db": 1.606834491933811, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17771428645280607, "1000–3000 Hz_10ms_envelope_correlation": 0.9775617623768843, "1000–3000 Hz_10ms_modulation_mae_db": 0.6053476921460597, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9569666006037573, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11242833092705834, "1000–3000 Hz_40ms_envelope_correlation": 0.9913815105209992, "1000–3000 Hz_40ms_modulation_mae_db": 0.6822289485883384, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6010560953341975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19208046108204746, "3000–6000 Hz_10ms_envelope_correlation": 0.9708877996683726, "3000–6000 Hz_10ms_modulation_mae_db": 0.5345801612336935, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0128489848770081, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09556480105412507, "3000–6000 Hz_40ms_envelope_correlation": 0.9925528119248922, "3000–6000 Hz_40ms_modulation_mae_db": 0.29777376845189824, "mr_stft_db": 5.417892889509166, "band_mae_db": 0.24169183696002608, "rms_gain_error_db": -0.16204390107782762, "output_peak": 0.23992380305382585}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7434944067330712, "全帯域_10ms_rms_envelope_nrmse": 0.062083587502506365, "全帯域_10ms_envelope_correlation": 0.9962975743565503, "全帯域_10ms_modulation_mae_db": 0.6395482198294066, "全帯域_40ms_log_rms_mae_db": 0.3951799851486064, "全帯域_40ms_rms_envelope_nrmse": 0.032239192365576194, "全帯域_40ms_envelope_correlation": 0.99907849517265, "全帯域_40ms_modulation_mae_db": 0.9464323723694629, "300–1000 Hz_10ms_log_rms_mae_db": 1.3505678089839421, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08910242307750003, "300–1000 Hz_10ms_envelope_correlation": 0.9940340023011373, "300–1000 Hz_10ms_modulation_mae_db": 0.7085661296394375, "300–1000 Hz_40ms_log_rms_mae_db": 0.7902221083351275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049521975489263306, "300–1000 Hz_40ms_envelope_correlation": 0.9983047629819491, "300–1000 Hz_40ms_modulation_mae_db": 1.2188413984527249, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6274636926596477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16365557876418482, "1000–3000 Hz_10ms_envelope_correlation": 0.9807729977845806, "1000–3000 Hz_10ms_modulation_mae_db": 0.8009338598556921, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9398086832146963, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0920973089406946, "1000–3000 Hz_40ms_envelope_correlation": 0.9940722751299237, "1000–3000 Hz_40ms_modulation_mae_db": 0.446239590275417, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6201317316796933, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18881297529727192, "3000–6000 Hz_10ms_envelope_correlation": 0.9719613813023359, "3000–6000 Hz_10ms_modulation_mae_db": 0.6229024297616393, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0882328676234516, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12121965021413232, "3000–6000 Hz_40ms_envelope_correlation": 0.9877502765921885, "3000–6000 Hz_40ms_modulation_mae_db": 0.39314971111752356, "mr_stft_db": 5.424605477754781, "band_mae_db": 0.18263005079934433, "rms_gain_error_db": -0.11705933634168482, "output_peak": 0.24886225706077314}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7554385027092465, "全帯域_10ms_rms_envelope_nrmse": 0.06947203654223726, "全帯域_10ms_envelope_correlation": 0.9953497454970716, "全帯域_10ms_modulation_mae_db": 0.390524720859086, "全帯域_40ms_log_rms_mae_db": 0.43676014144617253, "全帯域_40ms_rms_envelope_nrmse": 0.04189954212752419, "全帯域_40ms_envelope_correlation": 0.9983516150054117, "全帯域_40ms_modulation_mae_db": 0.4640687725622589, "300–1000 Hz_10ms_log_rms_mae_db": 1.3818343618989444, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09435426690350396, "300–1000 Hz_10ms_envelope_correlation": 0.9933473411716635, "300–1000 Hz_10ms_modulation_mae_db": 0.3869033176448597, "300–1000 Hz_40ms_log_rms_mae_db": 0.8650027708556294, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05805405167867105, "300–1000 Hz_40ms_envelope_correlation": 0.9976304332324258, "300–1000 Hz_40ms_modulation_mae_db": 0.5072439896995856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6348520641349904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14817735021836853, "1000–3000 Hz_10ms_envelope_correlation": 0.9848448386761075, "1000–3000 Hz_10ms_modulation_mae_db": 0.9350617030841636, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9274809853511609, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0944367219740687, "1000–3000 Hz_40ms_envelope_correlation": 0.9943515604981981, "1000–3000 Hz_40ms_modulation_mae_db": 0.9289736998048556, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6121971363486922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1817437199564422, "3000–6000 Hz_10ms_envelope_correlation": 0.9743307094434239, "3000–6000 Hz_10ms_modulation_mae_db": 0.6898731441687834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.086913467022869, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10972500282926877, "3000–6000 Hz_40ms_envelope_correlation": 0.9907440295573572, "3000–6000 Hz_40ms_modulation_mae_db": 0.9112111049227019, "mr_stft_db": 5.418020319362333, "band_mae_db": 0.20615269598544325, "rms_gain_error_db": -0.1296765673294214, "output_peak": 0.23089661065542225}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.740799822716751, "全帯域_10ms_rms_envelope_nrmse": 0.062297351046717644, "全帯域_10ms_envelope_correlation": 0.9963096824452252, "全帯域_10ms_modulation_mae_db": 0.7572029118741588, "全帯域_40ms_log_rms_mae_db": 0.3899076081919454, "全帯域_40ms_rms_envelope_nrmse": 0.03347590304040051, "全帯域_40ms_envelope_correlation": 0.9990306120510674, "全帯域_40ms_modulation_mae_db": 0.8922173544599344, "300–1000 Hz_10ms_log_rms_mae_db": 1.3902546844842156, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0871024408558374, "300–1000 Hz_10ms_envelope_correlation": 0.9943803853249806, "300–1000 Hz_10ms_modulation_mae_db": 0.8305664314578608, "300–1000 Hz_40ms_log_rms_mae_db": 0.8279864749697489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05186903296801016, "300–1000 Hz_40ms_envelope_correlation": 0.9981803792070635, "300–1000 Hz_40ms_modulation_mae_db": 1.0602500539500106, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6445228361955515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16904698631165344, "1000–3000 Hz_10ms_envelope_correlation": 0.9800959255183912, "1000–3000 Hz_10ms_modulation_mae_db": 1.1471676191721336, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9566310858449965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10542327730233655, "1000–3000 Hz_40ms_envelope_correlation": 0.9929677314030421, "1000–3000 Hz_40ms_modulation_mae_db": 1.123816759109957, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6207221526193352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18473439009704398, "3000–6000 Hz_10ms_envelope_correlation": 0.9731440008011075, "3000–6000 Hz_10ms_modulation_mae_db": 0.3727137842555801, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0025707201849745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10126599733405542, "3000–6000 Hz_40ms_envelope_correlation": 0.9916716389768861, "3000–6000 Hz_40ms_modulation_mae_db": 0.5822980332752001, "mr_stft_db": 5.429754036596744, "band_mae_db": 0.21292830879386723, "rms_gain_error_db": -0.12695817447119906, "output_peak": 0.23225531837949367}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7638728375460868, "全帯域_10ms_rms_envelope_nrmse": 0.06624211464197616, "全帯域_10ms_envelope_correlation": 0.9956281695233947, "全帯域_10ms_modulation_mae_db": 0.18016598041088658, "全帯域_40ms_log_rms_mae_db": 0.4019513608916201, "全帯域_40ms_rms_envelope_nrmse": 0.03536315537537624, "全帯域_40ms_envelope_correlation": 0.998668745812005, "全帯域_40ms_modulation_mae_db": 0.3329106446284007, "300–1000 Hz_10ms_log_rms_mae_db": 1.376567386960532, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08514609023038205, "300–1000 Hz_10ms_envelope_correlation": 0.9943448869838827, "300–1000 Hz_10ms_modulation_mae_db": 0.21749679017926393, "300–1000 Hz_40ms_log_rms_mae_db": 0.7817115299055313, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048015544232515785, "300–1000 Hz_40ms_envelope_correlation": 0.9981078513051888, "300–1000 Hz_40ms_modulation_mae_db": 0.4500454682323489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6065886816149715, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15520321419288977, "1000–3000 Hz_10ms_envelope_correlation": 0.9825457008497411, "1000–3000 Hz_10ms_modulation_mae_db": 0.7639654153568177, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8956230786914355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08611607544105086, "1000–3000 Hz_40ms_envelope_correlation": 0.9944016912372406, "1000–3000 Hz_40ms_modulation_mae_db": 0.8073615254283671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6107627966882765, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18100496852026798, "3000–6000 Hz_10ms_envelope_correlation": 0.9743011296362359, "3000–6000 Hz_10ms_modulation_mae_db": 0.480921161922929, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0658261966742941, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10030265803109796, "3000–6000 Hz_40ms_envelope_correlation": 0.9919379562569353, "3000–6000 Hz_40ms_modulation_mae_db": 0.3320492564553037, "mr_stft_db": 5.423929760854127, "band_mae_db": 0.1508369371473731, "rms_gain_error_db": -0.010288134116065996, "output_peak": 0.24531562978327826}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7656945226602351, "全帯域_10ms_rms_envelope_nrmse": 0.06966180837382169, "全帯域_10ms_envelope_correlation": 0.9953869283209841, "全帯域_10ms_modulation_mae_db": 0.8372118550422966, "全帯域_40ms_log_rms_mae_db": 0.4321573565883213, "全帯域_40ms_rms_envelope_nrmse": 0.039513173966215384, "全帯域_40ms_envelope_correlation": 0.9986300596209662, "全帯域_40ms_modulation_mae_db": 1.1280300306060698, "300–1000 Hz_10ms_log_rms_mae_db": 1.445542980306959, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09357860152051209, "300–1000 Hz_10ms_envelope_correlation": 0.9936112784111327, "300–1000 Hz_10ms_modulation_mae_db": 0.8611746928620573, "300–1000 Hz_40ms_log_rms_mae_db": 0.895903258363978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05530291167688452, "300–1000 Hz_40ms_envelope_correlation": 0.9980664856408151, "300–1000 Hz_40ms_modulation_mae_db": 1.3343682477148884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.58866247770095, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15919662359991757, "1000–3000 Hz_10ms_envelope_correlation": 0.9816799432925496, "1000–3000 Hz_10ms_modulation_mae_db": 0.40116107043828975, "1000–3000 Hz_40ms_log_rms_mae_db": 0.913896865652143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09108325680890797, "1000–3000 Hz_40ms_envelope_correlation": 0.993923798970591, "1000–3000 Hz_40ms_modulation_mae_db": 0.39980294652814946, "3000–6000 Hz_10ms_log_rms_mae_db": 1.628197340732184, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17541073059104084, "3000–6000 Hz_10ms_envelope_correlation": 0.9759782350581128, "3000–6000 Hz_10ms_modulation_mae_db": 0.585448283766253, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0804401625434137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10063778470488156, "3000–6000 Hz_40ms_envelope_correlation": 0.9920686714605884, "3000–6000 Hz_40ms_modulation_mae_db": 0.5190229695198497, "mr_stft_db": 5.430806224113296, "band_mae_db": 0.188984722674453, "rms_gain_error_db": -0.14974964311920028, "output_peak": 0.24858469474172118}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.691020126846843, "全帯域_10ms_rms_envelope_nrmse": 0.0569341872390151, "全帯域_10ms_envelope_correlation": 0.9969735496537557, "全帯域_10ms_modulation_mae_db": 0.3820291567660299, "全帯域_40ms_log_rms_mae_db": 0.3988527724501137, "全帯域_40ms_rms_envelope_nrmse": 0.03341339885782718, "全帯域_40ms_envelope_correlation": 0.9990598534225872, "全帯域_40ms_modulation_mae_db": 0.5194640149666956, "300–1000 Hz_10ms_log_rms_mae_db": 1.3012133448158754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07780476202147985, "300–1000 Hz_10ms_envelope_correlation": 0.9956612676107869, "300–1000 Hz_10ms_modulation_mae_db": 0.5121300716461633, "300–1000 Hz_40ms_log_rms_mae_db": 0.7773916379271375, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04786014574497929, "300–1000 Hz_40ms_envelope_correlation": 0.9985872258558871, "300–1000 Hz_40ms_modulation_mae_db": 0.6486664142965914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5833708201405843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1607202629317592, "1000–3000 Hz_10ms_envelope_correlation": 0.9814618587059261, "1000–3000 Hz_10ms_modulation_mae_db": 0.7216970099586009, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9319528362002735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10161072542775507, "1000–3000 Hz_40ms_envelope_correlation": 0.9925889320464878, "1000–3000 Hz_40ms_modulation_mae_db": 0.747828917215287, "3000–6000 Hz_10ms_log_rms_mae_db": 1.670557948097289, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19839713102687617, "3000–6000 Hz_10ms_envelope_correlation": 0.968966219931587, "3000–6000 Hz_10ms_modulation_mae_db": 0.49566048927579626, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0598131495584824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10945597633756401, "3000–6000 Hz_40ms_envelope_correlation": 0.9902690036952392, "3000–6000 Hz_40ms_modulation_mae_db": 0.5081445441767536, "mr_stft_db": 5.40874024161947, "band_mae_db": 0.22395697727214134, "rms_gain_error_db": -0.1365107392637728, "output_peak": 0.26284497024287273}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.2413988222125685, "spectral_convergence": 0.19101806107276825, "spectral_convergence_status": "ok", "amplitude_floor": 7.733796592545008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.234554577278102, "spectral_convergence": 0.19463283680614216, "spectral_convergence_status": "ok", "amplitude_floor": 4.871370963739547e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.737503318702687, "spectral_convergence": 0.19780354735977054, "spectral_convergence_status": "ok", "amplitude_floor": 4.7483837933045734e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001233024392514864, "output_power": 0.00012711801657289725, "delta_db": 0.13235439754158482, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000393494821723947, "output_power": 0.00038856579019654427, "delta_db": -0.05474459411222826, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000358215617892152, "output_power": 0.00033993561219150797, "delta_db": -0.22747852448727396, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.947495410559829e-05, "output_power": 5.507281945593717e-05, "delta_db": -0.3339680480239918, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.0477761937191392e-05, "output_power": 1.0292705391266244e-05, "delta_db": -0.07738984603112298, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.39686924640271e-06, "output_power": 9.482113485819166e-06, "delta_db": 0.039219649612703675, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14419250996815092, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11012348080008101, "pitch": null, "output_rms_dbfs": -30.312989982330603, "output_peak": 0.25377700094514666, "reference_rms_dbfs": -30.202866501530522, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.404485572731119}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.933633747331572e-05, "log_rms_mae_db": 0.7520397894723878, "rms_envelope_nrmse": 0.06547839483164691, "envelope_correlation": 0.99584663350445, "rms_gain_error_db": -0.11012340554719972, "reference_mean_power": 0.0009551583992700257, "output_mean_power": 0.0009312430798537865}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1194144575506277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002166069295810575, "output_power": 0.00020333022355081316, "reference_active": true, "delta_db": -0.27470408109283884}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.38201712334803e-05, "output_power": 2.3077052603734407e-05, "reference_active": true, "delta_db": -0.13764539097172024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.302931519380675e-06, "output_power": 7.463983841115306e-06, "reference_active": true, "delta_db": -0.46260765450482355}], "active_band_count": 3, "active_band_mae_db": 0.29165237552312756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.652181544798992e-05, "log_rms_mae_db": 0.4304980818638673, "rms_envelope_nrmse": 0.03415518012332344, "envelope_correlation": 0.9989250082751707, "rms_gain_error_db": -0.11012340954909539, "reference_mean_power": 0.0009599620519447541, "output_mean_power": 0.0009359264574914901}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.73215395386014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001986653950491428, "output_power": 0.00018707702327374072, "reference_active": true, "delta_db": -0.26101774298963176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916927878865493e-05, "output_power": 1.0782371985401904e-05, "reference_active": true, "delta_db": -0.05386130558876413}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.81959177520348e-07, "output_power": 5.221715477660893e-07, "reference_active": true, "delta_db": -0.47079317251984887}], "active_band_count": 3, "active_band_mae_db": 0.26189074036608156}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.027389787958093e-05, "log_rms_mae_db": 1.3688417795389816, "rms_envelope_nrmse": 0.08633612554487756, "envelope_correlation": 0.9943808964637634, "rms_gain_error_db": -0.16601833992997347, "reference_mean_power": 0.0006127358569067375, "output_mean_power": 0.0005897547693522266}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.366962347376557e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001936423107817767, "output_power": 0.00017899748066325125, "reference_active": true, "delta_db": -0.34153338087501783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9545968551344243e-05, "output_power": 1.8166982335761175e-05, "reference_active": true, "delta_db": -0.31774401758837917}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.863910891753386e-06, "output_power": 6.393569201780889e-06, "reference_active": true, "delta_db": -0.3082826661435869}], "active_band_count": 3, "active_band_mae_db": 0.32252002153566134}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.943169310162246e-05, "log_rms_mae_db": 0.783581181194565, "rms_envelope_nrmse": 0.049076317131529504, "envelope_correlation": 0.9982880914327253, "rms_gain_error_db": -0.16601833749911196, "reference_mean_power": 0.000615817475654304, "output_mean_power": 0.0005927208101616539}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.097820742933078e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018026482035346189, "output_power": 0.0001674525689740611, "reference_active": true, "delta_db": -0.32019165437022623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.739141870245847e-06, "output_power": 8.358323287049581e-06, "reference_active": true, "delta_db": -0.1934962471930589}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.763892611677605e-07, "output_power": 4.857080446007181e-07, "reference_active": true, "delta_db": -0.743405841937591}], "active_band_count": 3, "active_band_mae_db": 0.4190312478336254}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.05554473589972e-05, "log_rms_mae_db": 1.5662657108365288, "rms_envelope_nrmse": 0.15023849744393689, "envelope_correlation": 0.9837735205696667, "rms_gain_error_db": -0.2174260910214309, "reference_mean_power": 5.214627438247735e-05, "output_mean_power": 4.9599885778762286e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.666296423769008e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.72355480524535e-05, "output_power": 2.4952951704329775e-05, "reference_active": true, "delta_db": -0.3801419283493407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.163900077209868e-06, "output_power": 3.956761673441396e-06, "reference_active": true, "delta_db": -0.22160406712224606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1939677607224186e-06, "output_power": 1.1988503366517562e-06, "reference_active": true, "delta_db": 0.017723693488629837}], "active_band_count": 3, "active_band_mae_db": 0.20648989632007217}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.291047760123656e-05, "log_rms_mae_db": 0.9193024987269044, "rms_envelope_nrmse": 0.08962725378212157, "envelope_correlation": 0.9942137166746474, "rms_gain_error_db": -0.21742617301140515, "reference_mean_power": 5.2408520364582374e-05, "output_mean_power": 4.984932491583448e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.33304878770606e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.583453225376265e-05, "output_power": 2.3801208218470953e-05, "reference_active": true, "delta_db": -0.35601599183415333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0765400661096766e-06, "output_power": 2.0226015809165986e-06, "reference_active": true, "delta_db": -0.11429972725594417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.119613733354675e-08, "output_power": 7.679864597348366e-08, "reference_active": true, "delta_db": 0.9862955252950663}], "active_band_count": 3, "active_band_mae_db": 0.48553708146172125}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.3449414786854306e-05, "log_rms_mae_db": 1.6290553412908286, "rms_envelope_nrmse": 0.1892964853678217, "envelope_correlation": 0.9719452268881708, "rms_gain_error_db": -0.03895172061588277, "reference_mean_power": 6.886139457761324e-06, "output_mean_power": 6.8246540555107625e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.997833564030985e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1829042042252175e-06, "output_power": 2.1956576340705365e-06, "reference_active": true, "delta_db": 0.025299447969513347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.111376843686143e-07, "output_power": 7.179557258756592e-07, "reference_active": true, "delta_db": -0.5299691514728506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6537364416157584e-07, "output_power": 1.3041530239917553e-07, "reference_active": true, "delta_db": -1.031377438601882}], "active_band_count": 3, "active_band_mae_db": 0.5288820126814153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.0291390653962812e-05, "log_rms_mae_db": 1.0133816671684452, "rms_envelope_nrmse": 0.1006174498510173, "envelope_correlation": 0.9915068767694085, "rms_gain_error_db": -0.03895170230114303, "reference_mean_power": 6.9207717017349184e-06, "output_mean_power": 6.858977101815341e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.443349031906855e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9318994555375575e-06, "output_power": 1.969881301777976e-06, "reference_active": true, "delta_db": 0.08455537778121824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.056544990756342e-07, "output_power": 4.038664896787927e-07, "reference_active": true, "delta_db": -0.019184774408000496}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.969546554060793e-09, "output_power": 6.612462985960534e-09, "reference_active": true, "delta_db": -0.8107035786597759}], "active_band_count": 3, "active_band_mae_db": 0.3048145769496649}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9126205219541386, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "25ef7bb2bb3314c5cb5bfe0c040b6de3ede1151eb663623f55d2d29ac6e332e4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.981592068471731, "standard_deviation": 0.0022496357255354428, "minimum": 0.9775617623768843, "maximum": 0.9848448386761075, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6073200969021295, "standard_deviation": 0.0271391269679464, "minimum": 1.5662657108365288, "maximum": 1.6445228361955515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6977280332914786, "standard_deviation": 0.29561982234359874, "minimum": 0.20648989632007217, "maximum": 1.1471676191721336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16049409998943953, "standard_deviation": 0.009756239102539655, "minimum": 0.14817735021836853, "maximum": 0.17771428645280607, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934876520601412, "standard_deviation": 0.0010793404618292426, "minimum": 0.9913815105209992, "maximum": 0.9944016912372406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9302078292856709, "standard_deviation": 0.021037232964813836, "minimum": 0.8956230786914355, "maximum": 0.9569666006037573, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7027236835515116, "standard_deviation": 0.2525942894686983, "minimum": 0.39980294652814946, "maximum": 1.123816759109957, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0966028688254992, "standard_deviation": 0.00900328435126104, "minimum": 0.08611607544105086, "maximum": 0.11242833092705834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9726893378411683, "standard_deviation": 0.0022202414742413207, "minimum": 0.968966219931587, "maximum": 0.9759782350581128, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.624085067848812, "standard_deviation": 0.02096739124708028, "minimum": 1.6010560953341975, "maximum": 1.670557948097289, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5388726833832613, "standard_deviation": 0.09635063610942769, "minimum": 0.3727137842555801, "maximum": 0.6898731441687834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18643510774235156, "standard_deviation": 0.00721829634456973, "minimum": 0.17541073059104084, "maximum": 0.19839713102687617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9910626581541868, "standard_deviation": 0.0015245720157003352, "minimum": 0.9877502765921885, "maximum": 0.9925528119248922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0512534019566173, "standard_deviation": 0.03596783243706461, "minimum": 1.0025707201849745, "maximum": 1.0882328676234516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.48105799560861195, "standard_deviation": 0.2042556466811622, "minimum": 0.29777376845189824, "maximum": 0.9112111049227019, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10484866504451781, "standard_deviation": 0.008179605792945972, "minimum": 0.09556480105412507, "maximum": 0.12121965021413232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9942791716173739, "standard_deviation": 0.0006915085824305016, "minimum": 0.9933473411716635, "maximum": 0.9956612676107869, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.36877316036424, "standard_deviation": 0.042423742321345125, "minimum": 1.3012133448158754, "maximum": 1.445542980306959, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5359559115170554, "standard_deviation": 0.23898525945929056, "minimum": 0.21749679017926393, "maximum": 0.8611746928620573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08763567482016055, "standard_deviation": 0.005174496223133825, "minimum": 0.07780476202147985, "maximum": 0.09435426690350396, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981758358452903, "standard_deviation": 0.00027172344957846053, "minimum": 0.9976304332324258, "maximum": 0.9985872258558871, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8162895906426075, "standard_deviation": 0.043697948800323856, "minimum": 0.7773916379271375, "maximum": 0.895903258363978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7943934230761683, "standard_deviation": 0.3608735944687192, "minimum": 0.4190312478336254, "maximum": 1.3343682477148884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05164797860521898, "standard_deviation": 0.003712742262024366, "minimum": 0.04786014574497929, "maximum": 0.05805405167867105, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1939217549500999, "standard_deviation": 0.03417625851937787, "minimum": 0.14419250996815092, "maximum": 0.24169183696002608, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.41977931531763, "standard_deviation": 0.009441687226055494, "minimum": 5.404485572731119, "maximum": 5.430806224113296, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2453075356078167, "standard_deviation": 0.010754992512208282, "minimum": 0.23089661065542225, "maximum": 0.26284497024287273, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1178012470649066, "standard_deviation": 0.04655503360246431, "minimum": -0.16204390107782762, "maximum": -0.010288134116065996, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9959903755590441, "standard_deviation": 0.0005489036626354386, "minimum": 0.9953497454970716, "maximum": 0.9969735496537557, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7433642405585872, "standard_deviation": 0.023798013871432745, "minimum": 0.691020126846843, "maximum": 0.7656945226602351, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.48011378911032765, "standard_deviation": 0.2350274397763646, "minimum": 0.18016598041088658, "maximum": 0.8372118550422966, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0645827336650176, "standard_deviation": 0.004195834098790964, "minimum": 0.0569341872390151, "maximum": 0.06966180837382169, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998826014603287, "standard_deviation": 0.00025626324644446104, "minimum": 0.9983516150054117, "maximum": 0.99907849517265, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4118216158869452, "standard_deviation": 0.018568776947647234, "minimum": 0.3899076081919454, "maximum": 0.43676014144617253, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6215899458148927, "standard_deviation": 0.3208908057194101, "minimum": 0.26189074036608156, "maximum": 1.1280300306060698, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03591938123357922, "standard_deviation": 0.0033761241649483058, "minimum": 0.032239192365576194, "maximum": 0.04189954212752419, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5639-40744-0030", "language": "en", "style": "normal", "speaker": "librispeech_5639", "source_utterance_id": "5639-40744-0030", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5639-40744-0030.wav", "source_file_sha256": "8ab8d49e12b48a5efe855c8d6a251bf76cd0cdab358f1b68200b4bc02f4a5c46", "decoded_pcm_sha256": "76c45c30d066507ae3cbb52fafb188ebe36da864c7c54942db2ea8356726ce3a", "original_source_sha256": "c7f95b2f012ab8f4629d15c8c78e78bf17b3445c96e5a9b0c525b0743e50d9c5", "corpus_entry_id": "librispeech_5639-40744-0030"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.415320873260498, "predicted_mos": 4.415320873260498, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4a3fdb4ec69b25e35e202c4a0e41f6c2cf1817cd33274c0a0ab218548852a9e5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.712686303002556, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5639-40744-0030", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "4a3fdb4ec69b25e35e202c4a0e41f6c2cf1817cd33274c0a0ab218548852a9e5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "275d5d566a2394fbcd42dc2039695c9ffa82180022ac0926fe455c46d70e2ac4"}, "rows": [{"case": "librispeech_5639-40744-0030", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.2283947467803955, "predicted_mos": 3.2283947467803955, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "27d6ed117def03fdbd0bb258270b493058f147020578a4b34b3896695706b90a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.410599729999376, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5639-40744-0030", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "27d6ed117def03fdbd0bb258270b493058f147020578a4b34b3896695706b90a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7449464810470854, "全帯域_10ms_rms_envelope_nrmse": 0.07035956348028814, "全帯域_10ms_envelope_correlation": 0.995631687553773, "全帯域_10ms_modulation_mae_db": 0.10949772805371687, "全帯域_40ms_log_rms_mae_db": 0.4229592712896925, "全帯域_40ms_rms_envelope_nrmse": 0.036664083528842366, "全帯域_40ms_envelope_correlation": 0.99883925591872, "全帯域_40ms_modulation_mae_db": 0.10932370484554672, "300–1000 Hz_10ms_log_rms_mae_db": 1.4561505381929138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09513865134845939, "300–1000 Hz_10ms_envelope_correlation": 0.9934701551512134, "300–1000 Hz_10ms_modulation_mae_db": 0.20261602932180112, "300–1000 Hz_40ms_log_rms_mae_db": 0.844771017180306, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052480652366031826, "300–1000 Hz_40ms_envelope_correlation": 0.9979788992444684, "300–1000 Hz_40ms_modulation_mae_db": 0.2901371518017328, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5283500607939047, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1723943979986338, "1000–3000 Hz_10ms_envelope_correlation": 0.9786606511836707, "1000–3000 Hz_10ms_modulation_mae_db": 0.898971281750975, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8909698018284962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10687937549980997, "1000–3000 Hz_40ms_envelope_correlation": 0.992261824652916, "1000–3000 Hz_40ms_modulation_mae_db": 0.6781773586812601, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6194726018306795, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17735883287829024, "3000–6000 Hz_10ms_envelope_correlation": 0.975269092526899, "3000–6000 Hz_10ms_modulation_mae_db": 0.8585834903032549, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9583471949831924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09539956377784031, "3000–6000 Hz_40ms_envelope_correlation": 0.9928373058349851, "3000–6000 Hz_40ms_modulation_mae_db": 0.8916941647146848, "mr_stft_db": 5.368286111658769, "band_mae_db": 0.19985620822515934, "rms_gain_error_db": -0.10593736587821083, "output_peak": 0.20488277407153202}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7240741049021188, "全帯域_10ms_rms_envelope_nrmse": 0.07866617820632789, "全帯域_10ms_envelope_correlation": 0.9946264004360604, "全帯域_10ms_modulation_mae_db": 0.47709891166906776, "全帯域_40ms_log_rms_mae_db": 0.386730453805353, "全帯域_40ms_rms_envelope_nrmse": 0.03949787123558693, "全帯域_40ms_envelope_correlation": 0.9987842138017551, "全帯域_40ms_modulation_mae_db": 0.36593628851029925, "300–1000 Hz_10ms_log_rms_mae_db": 1.5677470858183766, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12222760883174456, "300–1000 Hz_10ms_envelope_correlation": 0.9895822163537573, "300–1000 Hz_10ms_modulation_mae_db": 0.6020416249997207, "300–1000 Hz_40ms_log_rms_mae_db": 0.9627487707474547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07063081283234324, "300–1000 Hz_40ms_envelope_correlation": 0.996865403123908, "300–1000 Hz_40ms_modulation_mae_db": 0.7580813476572489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4742319584464016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15074740118592747, "1000–3000 Hz_10ms_envelope_correlation": 0.9846483651054995, "1000–3000 Hz_10ms_modulation_mae_db": 0.9161877530551698, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8702613094368326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09403459133909244, "1000–3000 Hz_40ms_envelope_correlation": 0.9951076430626532, "1000–3000 Hz_40ms_modulation_mae_db": 1.2057952253143718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5644230545698252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17970205650725968, "3000–6000 Hz_10ms_envelope_correlation": 0.9743366398525498, "3000–6000 Hz_10ms_modulation_mae_db": 0.6235855132868336, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9185811400788263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10085648437265081, "3000–6000 Hz_40ms_envelope_correlation": 0.9913011824272311, "3000–6000 Hz_40ms_modulation_mae_db": 0.6725532164699217, "mr_stft_db": 5.358999871894323, "band_mae_db": 0.24172004459211652, "rms_gain_error_db": -0.1547999270531519, "output_peak": 0.2123755036722194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7051332359595015, "全帯域_10ms_rms_envelope_nrmse": 0.07467141181022639, "全帯域_10ms_envelope_correlation": 0.9950472227253597, "全帯域_10ms_modulation_mae_db": 0.5063614843347496, "全帯域_40ms_log_rms_mae_db": 0.4108535053487665, "全帯域_40ms_rms_envelope_nrmse": 0.04304368792401826, "全帯域_40ms_envelope_correlation": 0.9983278573479736, "全帯域_40ms_modulation_mae_db": 0.4636949151080953, "300–1000 Hz_10ms_log_rms_mae_db": 1.4862121566911803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10877934486389636, "300–1000 Hz_10ms_envelope_correlation": 0.991522765605698, "300–1000 Hz_10ms_modulation_mae_db": 0.7442938027667548, "300–1000 Hz_40ms_log_rms_mae_db": 0.9220000819333705, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06366135442843274, "300–1000 Hz_40ms_envelope_correlation": 0.9971133756338381, "300–1000 Hz_40ms_modulation_mae_db": 0.7182782946520024, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5145671861330605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17736440146924892, "1000–3000 Hz_10ms_envelope_correlation": 0.9774145667899803, "1000–3000 Hz_10ms_modulation_mae_db": 0.9587861233042015, "1000–3000 Hz_40ms_log_rms_mae_db": 0.940305171144165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10929069083412818, "1000–3000 Hz_40ms_envelope_correlation": 0.9919281042144137, "1000–3000 Hz_40ms_modulation_mae_db": 1.1247652168608955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5377444673390674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16723851663159778, "3000–6000 Hz_10ms_envelope_correlation": 0.9786225582655843, "3000–6000 Hz_10ms_modulation_mae_db": 1.0826496101566994, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8942242151184441, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09680548373531504, "3000–6000 Hz_40ms_envelope_correlation": 0.9932785168675626, "3000–6000 Hz_40ms_modulation_mae_db": 0.9151350174973855, "mr_stft_db": 5.367390755681427, "band_mae_db": 0.2365314833492569, "rms_gain_error_db": -0.09645312430139644, "output_peak": 0.20570582244540364}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.720407305381758, "全帯域_10ms_rms_envelope_nrmse": 0.07334673059603508, "全帯域_10ms_envelope_correlation": 0.9953056482596151, "全帯域_10ms_modulation_mae_db": 0.5516214290887801, "全帯域_40ms_log_rms_mae_db": 0.39894224688163477, "全帯域_40ms_rms_envelope_nrmse": 0.03959543270917777, "全帯域_40ms_envelope_correlation": 0.9987034601778617, "全帯域_40ms_modulation_mae_db": 0.44787115570343045, "300–1000 Hz_10ms_log_rms_mae_db": 1.516007660866926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09855210936649686, "300–1000 Hz_10ms_envelope_correlation": 0.9931642708024075, "300–1000 Hz_10ms_modulation_mae_db": 0.7238067313405115, "300–1000 Hz_40ms_log_rms_mae_db": 0.8912325355826334, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05382346240719233, "300–1000 Hz_40ms_envelope_correlation": 0.998105829851932, "300–1000 Hz_40ms_modulation_mae_db": 0.8366003293571275, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5230439383309857, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18441547949732653, "1000–3000 Hz_10ms_envelope_correlation": 0.9768287473954718, "1000–3000 Hz_10ms_modulation_mae_db": 0.7774390121116594, "1000–3000 Hz_40ms_log_rms_mae_db": 0.89102501601441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1112203681958734, "1000–3000 Hz_40ms_envelope_correlation": 0.993558079077707, "1000–3000 Hz_40ms_modulation_mae_db": 0.7865960458367504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.523413013739954, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16901321812714087, "3000–6000 Hz_10ms_envelope_correlation": 0.9774156083901709, "3000–6000 Hz_10ms_modulation_mae_db": 1.2606385041481394, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8836935307916118, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09750392575017773, "3000–6000 Hz_40ms_envelope_correlation": 0.9919924318014223, "3000–6000 Hz_40ms_modulation_mae_db": 1.2905471353846663, "mr_stft_db": 5.381316579732407, "band_mae_db": 0.3234759022221763, "rms_gain_error_db": -0.1341544581637375, "output_peak": 0.23592805527716842}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7277831926909888, "全帯域_10ms_rms_envelope_nrmse": 0.07897587653180188, "全帯域_10ms_envelope_correlation": 0.9946023460446516, "全帯域_10ms_modulation_mae_db": 0.2484351560489285, "全帯域_40ms_log_rms_mae_db": 0.3992588352196389, "全帯域_40ms_rms_envelope_nrmse": 0.03874901510182449, "全帯域_40ms_envelope_correlation": 0.9988679319483023, "全帯域_40ms_modulation_mae_db": 0.43943365163346737, "300–1000 Hz_10ms_log_rms_mae_db": 1.5568119519026846, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12099861792085363, "300–1000 Hz_10ms_envelope_correlation": 0.9896429299640328, "300–1000 Hz_10ms_modulation_mae_db": 0.3648133808570299, "300–1000 Hz_40ms_log_rms_mae_db": 0.9363680583878422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06128188584039994, "300–1000 Hz_40ms_envelope_correlation": 0.997636434954602, "300–1000 Hz_40ms_modulation_mae_db": 0.6918289017897342, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4727864850646202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14502795904449695, "1000–3000 Hz_10ms_envelope_correlation": 0.9848941626438875, "1000–3000 Hz_10ms_modulation_mae_db": 0.6834422388290209, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8804682519146293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07940137217730103, "1000–3000 Hz_40ms_envelope_correlation": 0.9958099728210555, "1000–3000 Hz_40ms_modulation_mae_db": 0.423636751187945, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6012220229268797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16094184654154028, "3000–6000 Hz_10ms_envelope_correlation": 0.979476616379347, "3000–6000 Hz_10ms_modulation_mae_db": 0.5141844085360704, "3000–6000 Hz_40ms_log_rms_mae_db": 0.919410924828555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08611881366700898, "3000–6000 Hz_40ms_envelope_correlation": 0.9936939321540089, "3000–6000 Hz_40ms_modulation_mae_db": 0.38761055819225404, "mr_stft_db": 5.362633150362993, "band_mae_db": 0.22733227429732994, "rms_gain_error_db": -0.16126343728930653, "output_peak": 0.21890489732765428}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6726568221247804, "全帯域_10ms_rms_envelope_nrmse": 0.07086743609289785, "全帯域_10ms_envelope_correlation": 0.995625696908332, "全帯域_10ms_modulation_mae_db": 0.30861432988423937, "全帯域_40ms_log_rms_mae_db": 0.35942299452497, "全帯域_40ms_rms_envelope_nrmse": 0.03919863583968536, "全帯域_40ms_envelope_correlation": 0.9987289420356519, "全帯域_40ms_modulation_mae_db": 0.35560252998845227, "300–1000 Hz_10ms_log_rms_mae_db": 1.4444409411695787, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09949688120841838, "300–1000 Hz_10ms_envelope_correlation": 0.9932268399442842, "300–1000 Hz_10ms_modulation_mae_db": 0.3473685034189269, "300–1000 Hz_40ms_log_rms_mae_db": 0.824060512932152, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05787181688020837, "300–1000 Hz_40ms_envelope_correlation": 0.9980029850929728, "300–1000 Hz_40ms_modulation_mae_db": 0.47803974726407006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5346345069372525, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15393700453701506, "1000–3000 Hz_10ms_envelope_correlation": 0.9827844146652951, "1000–3000 Hz_10ms_modulation_mae_db": 0.5171872203626225, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8783780419626953, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09089830621486314, "1000–3000 Hz_40ms_envelope_correlation": 0.9940956437430439, "1000–3000 Hz_40ms_modulation_mae_db": 0.31527488803245335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.576869935294547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1690864078926524, "3000–6000 Hz_10ms_envelope_correlation": 0.9777956097559338, "3000–6000 Hz_10ms_modulation_mae_db": 0.7614154304596363, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9603512262815143, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10148060724145887, "3000–6000 Hz_40ms_envelope_correlation": 0.9919813787759714, "3000–6000 Hz_40ms_modulation_mae_db": 0.733872887375928, "mr_stft_db": 5.388563709191104, "band_mae_db": 0.2079550981109105, "rms_gain_error_db": -0.1239131302431668, "output_peak": 0.23127239192597585}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7148899765980822, "全帯域_10ms_rms_envelope_nrmse": 0.07539874016941496, "全帯域_10ms_envelope_correlation": 0.9950220854195384, "全帯域_10ms_modulation_mae_db": 0.13278705920267278, "全帯域_40ms_log_rms_mae_db": 0.37035998367449163, "全帯域_40ms_rms_envelope_nrmse": 0.038523703489613946, "全帯域_40ms_envelope_correlation": 0.9987813286755439, "全帯域_40ms_modulation_mae_db": 0.19952227153154256, "300–1000 Hz_10ms_log_rms_mae_db": 1.4957757399790184, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10656185244357634, "300–1000 Hz_10ms_envelope_correlation": 0.9919753798809867, "300–1000 Hz_10ms_modulation_mae_db": 0.32168169555563303, "300–1000 Hz_40ms_log_rms_mae_db": 0.8553931958751404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05834095711640107, "300–1000 Hz_40ms_envelope_correlation": 0.9977551235838835, "300–1000 Hz_40ms_modulation_mae_db": 0.5491968550156536, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5082279084514536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1671875373178347, "1000–3000 Hz_10ms_envelope_correlation": 0.9801109852195103, "1000–3000 Hz_10ms_modulation_mae_db": 0.4556077851100448, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8704970191884981, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10819849749870065, "1000–3000 Hz_40ms_envelope_correlation": 0.9921299890902641, "1000–3000 Hz_40ms_modulation_mae_db": 0.8682400617320821, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6127175592472511, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17812738263785696, "3000–6000 Hz_10ms_envelope_correlation": 0.9748355561159314, "3000–6000 Hz_10ms_modulation_mae_db": 1.3773483564379456, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0046188207971047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10350994101415603, "3000–6000 Hz_40ms_envelope_correlation": 0.9909565944241058, "3000–6000 Hz_40ms_modulation_mae_db": 1.6609743861221833, "mr_stft_db": 5.369481249576983, "band_mae_db": 0.22279936575473744, "rms_gain_error_db": -0.1304615561510195, "output_peak": 0.23089180503570159}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6966411484979671, "全帯域_10ms_rms_envelope_nrmse": 0.07450892181774982, "全帯域_10ms_envelope_correlation": 0.9950412428493719, "全帯域_10ms_modulation_mae_db": 0.40858464603757144, "全帯域_40ms_log_rms_mae_db": 0.372742440706617, "全帯域_40ms_rms_envelope_nrmse": 0.03667268284795041, "全帯域_40ms_envelope_correlation": 0.9987706766785056, "全帯域_40ms_modulation_mae_db": 0.1425190427990927, "300–1000 Hz_10ms_log_rms_mae_db": 1.4844789374717553, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11039370709049666, "300–1000 Hz_10ms_envelope_correlation": 0.9911595650306892, "300–1000 Hz_10ms_modulation_mae_db": 0.16645544041200824, "300–1000 Hz_40ms_log_rms_mae_db": 0.8820796154997792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058373193997121965, "300–1000 Hz_40ms_envelope_correlation": 0.9974427335128473, "300–1000 Hz_40ms_modulation_mae_db": 0.22464448662960432, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5198551935055293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17462797707849131, "1000–3000 Hz_10ms_envelope_correlation": 0.9780160246689195, "1000–3000 Hz_10ms_modulation_mae_db": 0.9735195709685717, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8782214937600912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1005908063682462, "1000–3000 Hz_40ms_envelope_correlation": 0.9931997445831289, "1000–3000 Hz_40ms_modulation_mae_db": 1.058838151484556, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5568285518833385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16372336562700376, "3000–6000 Hz_10ms_envelope_correlation": 0.979319315857624, "3000–6000 Hz_10ms_modulation_mae_db": 1.232996776180974, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8958361266772779, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0911083903327372, "3000–6000 Hz_40ms_envelope_correlation": 0.9938832590993364, "3000–6000 Hz_40ms_modulation_mae_db": 1.3080687465397287, "mr_stft_db": 5.379286664400946, "band_mae_db": 0.22873554886759076, "rms_gain_error_db": -0.07059561034318367, "output_peak": 0.21793959092259188}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.391574618423584, "spectral_convergence": 0.23842055796588488, "spectral_convergence_status": "ok", "amplitude_floor": 6.716189184049609e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.157520663899622, "spectral_convergence": 0.23017480306530555, "spectral_convergence_status": "ok", "amplitude_floor": 4.979433430044932e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5557630526531, "spectral_convergence": 0.24484532087689484, "spectral_convergence_status": "ok", "amplitude_floor": 3.630042469105695e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.462171010234058e-05, "output_power": 7.752861408666324e-05, "delta_db": 0.1659682299869495, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003077822779549395, "output_power": 0.0002983456865069713, "delta_db": -0.13523846441913331, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 5.338393512143839e-05, "output_power": 5.254816462776801e-05, "delta_db": -0.06853032196336978, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.1515857114073014e-05, "output_power": 2.044525572610081e-05, "delta_db": -0.22166104474865997, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.940885952831811e-05, "output_power": 2.6080337372905302e-05, "delta_db": -0.5216497786038119, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.017666424897867e-06, "output_power": 7.86029856388538e-06, "delta_db": -0.08608940962903175, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19985620822515934, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10593736587821083, "pitch": null, "output_rms_dbfs": -33.16225221347421, "output_peak": 0.20488277407153202, "reference_rms_dbfs": -33.056314847596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.368286111658769}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.650394450462669e-05, "log_rms_mae_db": 0.7449464810470854, "rms_envelope_nrmse": 0.07035956348028814, "envelope_correlation": 0.995631687553773, "rms_gain_error_db": -0.10589178841995753, "reference_mean_power": 0.0004951134756771775, "output_mean_power": 0.00048318736357652424}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.645854738282274e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010486873457313227, "output_power": 0.00010056184329275461, "reference_active": true, "delta_db": -0.1821280216094566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5071002541192514e-05, "output_power": 2.585642153883089e-05, "reference_active": true, "delta_db": 0.13396718575009567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059292105678353e-06, "output_power": 5.044869739714287e-06, "reference_active": true, "delta_db": -0.012397976801598301}], "active_band_count": 3, "active_band_mae_db": 0.10949772805371687}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.020301765617229e-05, "log_rms_mae_db": 0.4229592712896925, "rms_envelope_nrmse": 0.036664083528842366, "envelope_correlation": 0.99883925591872, "rms_gain_error_db": -0.10568362154592974, "reference_mean_power": 0.0004970274687625529, "output_mean_power": 0.0004850785034170835}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.472129124053938e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.182773781582673e-05, "output_power": 8.868751421219909e-05, "reference_active": true, "delta_db": -0.15111403169713072}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2533326164459841e-05, "output_power": 1.2954364895705328e-05, "reference_active": true, "delta_db": 0.1434978420357578}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4723702388808576e-07, "output_power": 2.491434254443304e-07, "reference_active": true, "delta_db": 0.033359240803751626}], "active_band_count": 3, "active_band_mae_db": 0.10932370484554672}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.432622539351727e-05, "log_rms_mae_db": 1.4561505381929138, "rms_envelope_nrmse": 0.09513865134845939, "envelope_correlation": 0.9934701551512134, "rms_gain_error_db": -0.10779372716698213, "reference_mean_power": 0.00026104952136561866, "output_mean_power": 0.0002546499108974741}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.392840173310503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.633554932680893e-05, "output_power": 9.218302170055134e-05, "reference_active": true, "delta_db": -0.1913563825094723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.160826072139944e-05, "output_power": 2.231609470963718e-05, "reference_active": true, "delta_db": 0.13998384560221583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.769645916048805e-06, "output_power": 4.475436026587492e-06, "reference_active": true, "delta_db": -0.27650785985371523}], "active_band_count": 3, "active_band_mae_db": 0.20261602932180112}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.77451695930137e-05, "log_rms_mae_db": 0.844771017180306, "rms_envelope_nrmse": 0.052480652366031826, "envelope_correlation": 0.9979788992444684, "rms_gain_error_db": -0.10718320810399726, "reference_mean_power": 0.00026200159904973233, "output_mean_power": 0.00025561457952346224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2678124187221727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604417147270571e-05, "output_power": 8.307581723460548e-05, "reference_active": true, "delta_db": -0.15246834861502706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1040946271380912e-05, "output_power": 1.1360090833830459e-05, "reference_active": true, "delta_db": 0.12375507476495717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9369342076961927e-07, "output_power": 2.5613882592482767e-07, "reference_active": true, "delta_db": -0.594188032025214}], "active_band_count": 3, "active_band_mae_db": 0.2901371518017328}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.588832427367551e-05, "log_rms_mae_db": 1.5283500607939047, "rms_envelope_nrmse": 0.1723943979986338, "envelope_correlation": 0.9786606511836707, "rms_gain_error_db": -0.3913856814793255, "reference_mean_power": 4.0575597065913316e-05, "output_mean_power": 3.707885810687845e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.7364679678483124e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6458578452566544e-05, "output_power": 1.4807612509513113e-05, "reference_active": true, "delta_db": -0.45907280821264335}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.144001855335748e-06, "output_power": 2.3013080715971443e-06, "reference_active": true, "delta_db": -1.3550803294218072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.079905260013679e-07, "output_power": 7.409778849829135e-07, "reference_active": true, "delta_db": -0.8827607076184741}], "active_band_count": 3, "active_band_mae_db": 0.898971281750975}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.068450322528164e-05, "log_rms_mae_db": 0.8909698018284962, "rms_envelope_nrmse": 0.10687937549980997, "envelope_correlation": 0.992261824652916, "rms_gain_error_db": -0.3945172131363261, "reference_mean_power": 4.071142209359612e-05, "output_mean_power": 3.7176161977080475e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.516451938680711e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5218014411010514e-05, "output_power": 1.369520589059267e-05, "reference_active": true, "delta_db": -0.4578942530582629}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2444589918408885e-06, "output_power": 1.0538216265403004e-06, "reference_active": true, "delta_db": -0.7221348312326001}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2414094224167626e-08, "output_power": 4.305244156203459e-08, "reference_active": true, "delta_db": -0.8545029917529173}], "active_band_count": 3, "active_band_mae_db": 0.6781773586812601}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.7354442023779843e-05, "log_rms_mae_db": 1.6194726018306795, "rms_envelope_nrmse": 0.17735883287829024, "envelope_correlation": 0.975269092526899, "rms_gain_error_db": -0.30156118497957385, "reference_mean_power": 9.786447369332928e-06, "output_mean_power": 9.129961503788601e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.593718108385407e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.378843387910109e-06, "output_power": 2.1910571171850554e-06, "reference_active": true, "delta_db": -0.35712151993224284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.926849454092717e-07, "output_power": 7.843398510550857e-07, "reference_active": true, "delta_db": -0.5619392934267482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6983089704559518e-07, "output_power": 1.1597055559452966e-07, "reference_active": true, "delta_db": -1.6566896575507735}], "active_band_count": 3, "active_band_mae_db": 0.8585834903032549}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.2351945663576383e-05, "log_rms_mae_db": 0.9583471949831924, "rms_envelope_nrmse": 0.09539956377784031, "envelope_correlation": 0.9928373058349851, "rms_gain_error_db": -0.3038701237870529, "reference_mean_power": 9.83170860207206e-06, "output_mean_power": 9.167311439583758e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.2206731147886266e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.414723334288e-06, "output_power": 2.2039840466982227e-06, "reference_active": true, "delta_db": -0.3965893226878745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.701511911331394e-07, "output_power": 3.6528396837522754e-07, "reference_active": true, "delta_db": -0.057485390003502126}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1816168225696604e-08, "output_power": 7.085587854053692e-09, "reference_active": true, "delta_db": -2.221007781452678}], "active_band_count": 3, "active_band_mae_db": 0.8916941647146848}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9008182763617703, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fb2d0a8fa4ee2a9314eba52c398d1ed4239c06c6bbc3f524d04f77e8f5f24caa", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9804197397090293, "standard_deviation": 0.003259870261019595, "minimum": 0.9768287473954718, "maximum": 0.9848941626438875, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.509462154707901, "standard_deviation": 0.02359399620056456, "minimum": 1.4727864850646202, "maximum": 1.5346345069372525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7726426231865331, "standard_deviation": 0.2018887070967354, "minimum": 0.4556077851100448, "maximum": 0.9735195709685717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16571276976612184, "standard_deviation": 0.014153970605392398, "minimum": 0.14502795904449695, "maximum": 0.18441547949732653, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9935113751556477, "standard_deviation": 0.0014267164051217835, "minimum": 0.9919281042144137, "maximum": 0.9958099728210555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8875157631562272, "standard_deviation": 0.02273408222032071, "minimum": 0.8702613094368326, "maximum": 0.940305171144165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8076654623912892, "standard_deviation": 0.3235857542122575, "minimum": 0.31527488803245335, "maximum": 1.2057952253143718, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10006425101600187, "standard_deviation": 0.011147076440409371, "minimum": 0.07940137217730103, "maximum": 0.1112203681958734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.977133874643005, "standard_deviation": 0.0020557320977221756, "minimum": 0.9743366398525498, "maximum": 0.979476616379347, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5740864008539428, "standard_deviation": 0.03500225623927621, "minimum": 1.523413013739954, "maximum": 1.6194726018306795, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9639252611886941, "standard_deviation": 0.3197448657541337, "minimum": 0.5141844085360704, "maximum": 1.3773483564379456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17064895335541774, "standard_deviation": 0.006988235105305786, "minimum": 0.16094184654154028, "maximum": 0.17970205650725968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9924905751730779, "standard_deviation": 0.0010956987750465172, "minimum": 0.9909565944241058, "maximum": 0.9938832590993364, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9293828974445658, "standard_deviation": 0.04161656140559828, "minimum": 0.8836935307916118, "maximum": 1.0046188207971047, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9825570140370941, "standard_deviation": 0.4115780001196469, "minimum": 0.38761055819225404, "maximum": 1.6609743861221833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09659790123641812, "standard_deviation": 0.005756966363532184, "minimum": 0.08611881366700898, "maximum": 0.10350994101415603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9917180153416336, "standard_deviation": 0.0015455947911003473, "minimum": 0.9895822163537573, "maximum": 0.9934701551512134, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5009531265115543, "standard_deviation": 0.04395562668006739, "minimum": 1.4444409411695787, "maximum": 1.5677470858183766, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4341346510840483, "standard_deviation": 0.22631642434068372, "minimum": 0.16645544041200824, "maximum": 0.7442938027667548, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10776859663424276, "standard_deviation": 0.010045226136748428, "minimum": 0.09513865134845939, "maximum": 0.12222760883174456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976125981248065, "standard_deviation": 0.0004451508867812683, "minimum": 0.996865403123908, "maximum": 0.998105829851932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8898317235173347, "standard_deviation": 0.04799065170854305, "minimum": 0.824060512932152, "maximum": 0.9627487707474547, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5683508892708966, "standard_deviation": 0.2235238593010955, "minimum": 0.22464448662960432, "maximum": 0.8366003293571275, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05955801698351643, "standard_deviation": 0.005746034994021923, "minimum": 0.052480652366031826, "maximum": 0.07063081283234324, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23605074067740972, "standard_deviation": 0.037947450369387924, "minimum": 0.19985620822515934, "maximum": 0.3234759022221763, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.371994761562369, "standard_deviation": 0.010087997005063076, "minimum": 5.358999871894323, "maximum": 5.388563709191104, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2197376050847809, "standard_deviation": 0.011924056621954657, "minimum": 0.20488277407153202, "maximum": 0.23592805527716842, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12219732617789664, "standard_deviation": 0.03022107217408296, "minimum": -0.16126343728930653, "maximum": -0.07059561034318367, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951127912745878, "standard_deviation": 0.00039357259247992224, "minimum": 0.9946023460446516, "maximum": 0.995631687553773, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7133165334002853, "standard_deviation": 0.021962516360614522, "minimum": 0.6726568221247804, "maximum": 0.7449464810470854, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3428750930399658, "standard_deviation": 0.1695924904989045, "minimum": 0.10949772805371687, "maximum": 0.5516214290887801, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07459935733809275, "standard_deviation": 0.003158152657589036, "minimum": 0.07035956348028814, "maximum": 0.07897587653180188, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987254583230393, "standard_deviation": 0.00016922074295998507, "minimum": 0.9983278573479736, "maximum": 0.9988679319483023, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.39015871643139555, "standard_deviation": 0.02176454299057518, "minimum": 0.35942299452497, "maximum": 0.4229592712896925, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3154879450149908, "standard_deviation": 0.14386369859767176, "minimum": 0.10932370484554672, "maximum": 0.4636949151080953, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038993139084587444, "standard_deviation": 0.0020055519438415994, "minimum": 0.036664083528842366, "maximum": 0.04304368792401826, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0030", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.105706214904785, "predicted_mos": 3.105706214904785, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c77f5b60639f9b7478718cbffb5973c2239ac26b341ea44eeaced8df39719a80", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.038266899995506, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7468635055756105, "全帯域_10ms_rms_envelope_nrmse": 0.0708003648678254, "全帯域_10ms_envelope_correlation": 0.9955778533668297, "全帯域_10ms_modulation_mae_db": 0.09669236906639982, "全帯域_40ms_log_rms_mae_db": 0.42897350447581245, "全帯域_40ms_rms_envelope_nrmse": 0.03730466677731802, "全帯域_40ms_envelope_correlation": 0.9987952974821563, "全帯域_40ms_modulation_mae_db": 0.11130691882899413, "300–1000 Hz_10ms_log_rms_mae_db": 1.4518244033004477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09704228037095074, "300–1000 Hz_10ms_envelope_correlation": 0.9931995140498945, "300–1000 Hz_10ms_modulation_mae_db": 0.163858968893129, "300–1000 Hz_40ms_log_rms_mae_db": 0.8269426131653078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053182516394315246, "300–1000 Hz_40ms_envelope_correlation": 0.997918695625601, "300–1000 Hz_40ms_modulation_mae_db": 0.30004929160083943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.540699799867102, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1633036065714083, "1000–3000 Hz_10ms_envelope_correlation": 0.9810307383880753, "1000–3000 Hz_10ms_modulation_mae_db": 1.040290797740665, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9104036096142047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09866373615780609, "1000–3000 Hz_40ms_envelope_correlation": 0.9936451178942992, "1000–3000 Hz_40ms_modulation_mae_db": 0.9476383623523161, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5471416920698593, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16863607620049093, "3000–6000 Hz_10ms_envelope_correlation": 0.9780968915302889, "3000–6000 Hz_10ms_modulation_mae_db": 0.9023057468707423, "3000–6000 Hz_40ms_log_rms_mae_db": 0.968014401039022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09266104936598833, "3000–6000 Hz_40ms_envelope_correlation": 0.9938459325244196, "3000–6000 Hz_40ms_modulation_mae_db": 0.9265265013084427, "mr_stft_db": 5.2769257192680845, "band_mae_db": 0.21224910073529604, "rms_gain_error_db": -0.10678594316586254, "output_peak": 0.20864468790604737}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7288555439874477, "全帯域_10ms_rms_envelope_nrmse": 0.07898360864323509, "全帯域_10ms_envelope_correlation": 0.9945656822608433, "全帯域_10ms_modulation_mae_db": 0.5271371419778861, "全帯域_40ms_log_rms_mae_db": 0.390401278618347, "全帯域_40ms_rms_envelope_nrmse": 0.04015240226594507, "全帯域_40ms_envelope_correlation": 0.9987155987275419, "全帯域_40ms_modulation_mae_db": 0.3430809592086909, "300–1000 Hz_10ms_log_rms_mae_db": 1.568008411469022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12177324219658434, "300–1000 Hz_10ms_envelope_correlation": 0.9896136990736082, "300–1000 Hz_10ms_modulation_mae_db": 0.6029368939338667, "300–1000 Hz_40ms_log_rms_mae_db": 0.9841434983868296, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06941452031618865, "300–1000 Hz_40ms_envelope_correlation": 0.9969343023155078, "300–1000 Hz_40ms_modulation_mae_db": 0.7396375964398078, "1000–3000 Hz_10ms_log_rms_mae_db": 1.492279116296806, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15253681389939208, "1000–3000 Hz_10ms_envelope_correlation": 0.9845534389071003, "1000–3000 Hz_10ms_modulation_mae_db": 0.8649555118161928, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9025251511645721, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09307705131295188, "1000–3000 Hz_40ms_envelope_correlation": 0.9956657255408803, "1000–3000 Hz_40ms_modulation_mae_db": 1.2125087869689015, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4817867362337642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17874663983089298, "3000–6000 Hz_10ms_envelope_correlation": 0.974715493509813, "3000–6000 Hz_10ms_modulation_mae_db": 0.6152881225586183, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9010576584905221, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1064933274828672, "3000–6000 Hz_40ms_envelope_correlation": 0.9904014425217696, "3000–6000 Hz_40ms_modulation_mae_db": 0.6923007075547473, "mr_stft_db": 5.277002771691417, "band_mae_db": 0.2526562099069738, "rms_gain_error_db": -0.15006251254094344, "output_peak": 0.21714757610925844}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6945272717106591, "全帯域_10ms_rms_envelope_nrmse": 0.07408836428336535, "全帯域_10ms_envelope_correlation": 0.9951525959062494, "全帯域_10ms_modulation_mae_db": 0.5130888713839256, "全帯域_40ms_log_rms_mae_db": 0.40577344333100457, "全帯域_40ms_rms_envelope_nrmse": 0.04243092913359172, "全帯域_40ms_envelope_correlation": 0.9984128097512145, "全帯域_40ms_modulation_mae_db": 0.5235387171323455, "300–1000 Hz_10ms_log_rms_mae_db": 1.4684406855651224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10633494211409435, "300–1000 Hz_10ms_envelope_correlation": 0.991907936159327, "300–1000 Hz_10ms_modulation_mae_db": 0.7752738966413081, "300–1000 Hz_40ms_log_rms_mae_db": 0.8712650053328377, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062356092677039066, "300–1000 Hz_40ms_envelope_correlation": 0.9972374191922035, "300–1000 Hz_40ms_modulation_mae_db": 0.8099608490422163, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5049040012734416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17352684939897747, "1000–3000 Hz_10ms_envelope_correlation": 0.9787545487026297, "1000–3000 Hz_10ms_modulation_mae_db": 0.9376385337892533, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9454178626178061, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10648193672703275, "1000–3000 Hz_40ms_envelope_correlation": 0.9928204701092639, "1000–3000 Hz_40ms_modulation_mae_db": 1.1807568010672067, "3000–6000 Hz_10ms_log_rms_mae_db": 1.491591446418174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16713900171375085, "3000–6000 Hz_10ms_envelope_correlation": 0.9788684533753865, "3000–6000 Hz_10ms_modulation_mae_db": 0.9762350173210317, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9336782006305233, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09976636149267784, "3000–6000 Hz_40ms_envelope_correlation": 0.9931107042591446, "3000–6000 Hz_40ms_modulation_mae_db": 0.8630043021063608, "mr_stft_db": 5.284585928364578, "band_mae_db": 0.24581305625858807, "rms_gain_error_db": -0.11053003631896084, "output_peak": 0.20473091281755915}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7280102061669311, "全帯域_10ms_rms_envelope_nrmse": 0.07301933568619086, "全帯域_10ms_envelope_correlation": 0.9953623803997498, "全帯域_10ms_modulation_mae_db": 0.5620554727321302, "全帯域_40ms_log_rms_mae_db": 0.4099548992336973, "全帯域_40ms_rms_envelope_nrmse": 0.04001931722701743, "全帯域_40ms_envelope_correlation": 0.9986859538502149, "全帯域_40ms_modulation_mae_db": 0.5199358756534703, "300–1000 Hz_10ms_log_rms_mae_db": 1.5238242304047227, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10108459166866479, "300–1000 Hz_10ms_envelope_correlation": 0.992792858021827, "300–1000 Hz_10ms_modulation_mae_db": 0.7518528621449435, "300–1000 Hz_40ms_log_rms_mae_db": 0.85824798024816, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05520315829898252, "300–1000 Hz_40ms_envelope_correlation": 0.9979915126485662, "300–1000 Hz_40ms_modulation_mae_db": 0.996926299972316, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5208728884447398, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18091893100820924, "1000–3000 Hz_10ms_envelope_correlation": 0.977794138671054, "1000–3000 Hz_10ms_modulation_mae_db": 0.7995031140424832, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9001778149195432, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10885081102949086, "1000–3000 Hz_40ms_envelope_correlation": 0.9939012167059357, "1000–3000 Hz_40ms_modulation_mae_db": 0.8973597203945012, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4435889981664287, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16057806517475218, "3000–6000 Hz_10ms_envelope_correlation": 0.9798378055676498, "3000–6000 Hz_10ms_modulation_mae_db": 1.3591219192664108, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8721819460300227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09125406441401299, "3000–6000 Hz_40ms_envelope_correlation": 0.9934016422255086, "3000–6000 Hz_40ms_modulation_mae_db": 1.4463595393593234, "mr_stft_db": 5.2877253665305615, "band_mae_db": 0.32230022389576757, "rms_gain_error_db": -0.13896422204450012, "output_peak": 0.2339912642267531}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7140549206706057, "全帯域_10ms_rms_envelope_nrmse": 0.07836907849345141, "全帯域_10ms_envelope_correlation": 0.9947096019759467, "全帯域_10ms_modulation_mae_db": 0.2711115802622656, "全帯域_40ms_log_rms_mae_db": 0.4013451753853049, "全帯域_40ms_rms_envelope_nrmse": 0.03971361334239257, "全帯域_40ms_envelope_correlation": 0.9988181614343963, "全帯域_40ms_modulation_mae_db": 0.48306087498464506, "300–1000 Hz_10ms_log_rms_mae_db": 1.5528750484304932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12010464877180267, "300–1000 Hz_10ms_envelope_correlation": 0.9897725604733757, "300–1000 Hz_10ms_modulation_mae_db": 0.3894701111736251, "300–1000 Hz_40ms_log_rms_mae_db": 0.9097874153249995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06067602816182769, "300–1000 Hz_40ms_envelope_correlation": 0.99764968086806, "300–1000 Hz_40ms_modulation_mae_db": 0.7995700907528688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4515888928834138, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1389971946218549, "1000–3000 Hz_10ms_envelope_correlation": 0.9865313166110956, "1000–3000 Hz_10ms_modulation_mae_db": 0.8612254820329633, "1000–3000 Hz_40ms_log_rms_mae_db": 0.877880120011192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07611245867199341, "1000–3000 Hz_40ms_envelope_correlation": 0.9966640655593135, "1000–3000 Hz_40ms_modulation_mae_db": 0.6522005620662589, "3000–6000 Hz_10ms_log_rms_mae_db": 1.510479640712301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16358782180304146, "3000–6000 Hz_10ms_envelope_correlation": 0.9789750953378177, "3000–6000 Hz_10ms_modulation_mae_db": 0.5060362424180652, "3000–6000 Hz_40ms_log_rms_mae_db": 0.918695030665732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09312400951079723, "3000–6000 Hz_40ms_envelope_correlation": 0.9929470854402802, "3000–6000 Hz_40ms_modulation_mae_db": 0.44493636630676064, "mr_stft_db": 5.276234836776896, "band_mae_db": 0.2494616762900527, "rms_gain_error_db": -0.16764291751395483, "output_peak": 0.22529296157434697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6884891906024492, "全帯域_10ms_rms_envelope_nrmse": 0.07042143398781932, "全帯域_10ms_envelope_correlation": 0.9956807121455563, "全帯域_10ms_modulation_mae_db": 0.3269796218584593, "全帯域_40ms_log_rms_mae_db": 0.3671766301332125, "全帯域_40ms_rms_envelope_nrmse": 0.03873509894200209, "全帯域_40ms_envelope_correlation": 0.9987601887860259, "全帯域_40ms_modulation_mae_db": 0.38464221109213886, "300–1000 Hz_10ms_log_rms_mae_db": 1.4218972494072066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09510558697062889, "300–1000 Hz_10ms_envelope_correlation": 0.9938159317839318, "300–1000 Hz_10ms_modulation_mae_db": 0.2725598704300118, "300–1000 Hz_40ms_log_rms_mae_db": 0.7874558677205279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05530689607628018, "300–1000 Hz_40ms_envelope_correlation": 0.9981674244387132, "300–1000 Hz_40ms_modulation_mae_db": 0.541262638293315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5486357160618476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15629814299651995, "1000–3000 Hz_10ms_envelope_correlation": 0.9824382202380318, "1000–3000 Hz_10ms_modulation_mae_db": 0.60584480492097, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8933073055734033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09001092513160792, "1000–3000 Hz_40ms_envelope_correlation": 0.994522868139609, "1000–3000 Hz_40ms_modulation_mae_db": 0.38838161645287217, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5297652082496285, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17165228323415785, "3000–6000 Hz_10ms_envelope_correlation": 0.9772653393650361, "3000–6000 Hz_10ms_modulation_mae_db": 0.949816222203754, "3000–6000 Hz_40ms_log_rms_mae_db": 0.953465648802737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10932727652902101, "3000–6000 Hz_40ms_envelope_correlation": 0.9908035504428935, "3000–6000 Hz_40ms_modulation_mae_db": 1.0156862103627924, "mr_stft_db": 5.302957242756016, "band_mae_db": 0.22021742342526815, "rms_gain_error_db": -0.12409377034662396, "output_peak": 0.22963278392758538}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7156243108094537, "全帯域_10ms_rms_envelope_nrmse": 0.07571336217749597, "全帯域_10ms_envelope_correlation": 0.9949449438376987, "全帯域_10ms_modulation_mae_db": 0.15150083345266763, "全帯域_40ms_log_rms_mae_db": 0.37907869756599755, "全帯域_40ms_rms_envelope_nrmse": 0.038454290768819496, "全帯域_40ms_envelope_correlation": 0.9987437157323985, "全帯域_40ms_modulation_mae_db": 0.23764566318178357, "300–1000 Hz_10ms_log_rms_mae_db": 1.4976935722526785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10728429944251201, "300–1000 Hz_10ms_envelope_correlation": 0.991775920931202, "300–1000 Hz_10ms_modulation_mae_db": 0.3222512690104596, "300–1000 Hz_40ms_log_rms_mae_db": 0.8461295530626838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05617322096862797, "300–1000 Hz_40ms_envelope_correlation": 0.997836485841287, "300–1000 Hz_40ms_modulation_mae_db": 0.5670776166251414, "1000–3000 Hz_10ms_log_rms_mae_db": 1.51930141766244, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16895731608925882, "1000–3000 Hz_10ms_envelope_correlation": 0.9801497529252108, "1000–3000 Hz_10ms_modulation_mae_db": 0.45954915091704956, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9004797870256491, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10748425751942572, "1000–3000 Hz_40ms_envelope_correlation": 0.9929146514677514, "1000–3000 Hz_40ms_modulation_mae_db": 0.8594036710890712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.539700469167985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1709615832300583, "3000–6000 Hz_10ms_envelope_correlation": 0.9771159006347416, "3000–6000 Hz_10ms_modulation_mae_db": 1.4237672906957648, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0243228122584438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10307630159119645, "3000–6000 Hz_40ms_envelope_correlation": 0.9914865996868952, "3000–6000 Hz_40ms_modulation_mae_db": 1.6796023863106297, "mr_stft_db": 5.280092616513483, "band_mae_db": 0.22961721451089753, "rms_gain_error_db": -0.11647992360620703, "output_peak": 0.2303198523984184}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6997486188248776, "全帯域_10ms_rms_envelope_nrmse": 0.07580597711227237, "全帯域_10ms_envelope_correlation": 0.9948539770719708, "全帯域_10ms_modulation_mae_db": 0.4226429792104172, "全帯域_40ms_log_rms_mae_db": 0.3742687669207251, "全帯域_40ms_rms_envelope_nrmse": 0.038383410614717374, "全帯域_40ms_envelope_correlation": 0.998628478322262, "全帯域_40ms_modulation_mae_db": 0.11113854850395337, "300–1000 Hz_10ms_log_rms_mae_db": 1.4765905684322385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11384053298116024, "300–1000 Hz_10ms_envelope_correlation": 0.990582188292303, "300–1000 Hz_10ms_modulation_mae_db": 0.15267709565198154, "300–1000 Hz_40ms_log_rms_mae_db": 0.8718809449402564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06014602946778254, "300–1000 Hz_40ms_envelope_correlation": 0.9972484160051803, "300–1000 Hz_40ms_modulation_mae_db": 0.22749907595200317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5316197554854947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1738461302898426, "1000–3000 Hz_10ms_envelope_correlation": 0.9783180657197664, "1000–3000 Hz_10ms_modulation_mae_db": 0.9899317858378156, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8987524977909273, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09830879316947576, "1000–3000 Hz_40ms_envelope_correlation": 0.993641977100817, "1000–3000 Hz_40ms_modulation_mae_db": 0.9980677613055683, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5094443329909342, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15619525313328012, "3000–6000 Hz_10ms_envelope_correlation": 0.9814221498978553, "3000–6000 Hz_10ms_modulation_mae_db": 1.1867125368351557, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9243327018984694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09048997367308542, "3000–6000 Hz_40ms_envelope_correlation": 0.9941970699903857, "3000–6000 Hz_40ms_modulation_mae_db": 1.248811611398478, "mr_stft_db": 5.288741270951143, "band_mae_db": 0.23140195154379847, "rms_gain_error_db": -0.057756335291799715, "output_peak": 0.21285650836262426}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.310244724149571, "spectral_convergence": 0.23873923142750436, "spectral_convergence_status": "ok", "amplitude_floor": 6.716189184049609e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.069795803456566, "spectral_convergence": 0.2294479278389487, "spectral_convergence_status": "ok", "amplitude_floor": 4.979433430044932e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.450736630198115, "spectral_convergence": 0.24502113062630748, "spectral_convergence_status": "ok", "amplitude_floor": 3.630042469105695e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.462171010234058e-05, "output_power": 7.757713628164128e-05, "delta_db": 0.16868546286402675, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003077822779549395, "output_power": 0.0002985381869637642, "delta_db": -0.13243718622203948, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 5.338393512143839e-05, "output_power": 5.2477733221621045e-05, "delta_db": -0.07435516635965275, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.1515857114073014e-05, "output_power": 2.0302933003958614e-05, "delta_db": -0.25199869957375376, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.940885952831811e-05, "output_power": 2.6003779087280778e-05, "delta_db": -0.5344171507687729, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.017666424897867e-06, "output_power": 7.81426058309877e-06, "delta_db": -0.11160093862353052, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21224910073529604, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10678594316586254, "pitch": null, "output_rms_dbfs": -33.16310079076186, "output_peak": 0.20864468790604737, "reference_rms_dbfs": -33.056314847596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.2769257192680845}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.650394450462669e-05, "log_rms_mae_db": 0.7468635055756105, "rms_envelope_nrmse": 0.0708003648678254, "envelope_correlation": 0.9955778533668297, "rms_gain_error_db": -0.10674579025153273, "reference_mean_power": 0.0004951134756771775, "output_mean_power": 0.00048309235838027014}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.645854738282274e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010486873457313227, "output_power": 0.00010113873924055319, "reference_active": true, "delta_db": -0.157284919342977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5071002541192514e-05, "output_power": 2.5458557217497904e-05, "reference_active": true, "delta_db": 0.06662086854068522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059292105678353e-06, "output_power": 5.1369683115254984e-06, "reference_active": true, "delta_db": 0.06617131931553723}], "active_band_count": 3, "active_band_mae_db": 0.09669236906639982}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.020301765617229e-05, "log_rms_mae_db": 0.42897350447581245, "rms_envelope_nrmse": 0.03730466677731802, "envelope_correlation": 0.9987952974821563, "rms_gain_error_db": -0.10644434255442753, "reference_mean_power": 0.0004970274687625529, "output_mean_power": 0.00048499354330193596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.472129124053938e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.182773781582673e-05, "output_power": 8.918013016518313e-05, "reference_active": true, "delta_db": -0.12705783667200343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2533326164459841e-05, "output_power": 1.2800829636905324e-05, "reference_active": true, "delta_db": 0.09171775900765748}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4723702388808576e-07, "output_power": 2.5387972733835754e-07, "reference_active": true, "delta_db": 0.11514516080732146}], "active_band_count": 3, "active_band_mae_db": 0.11130691882899413}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.432622539351727e-05, "log_rms_mae_db": 1.4518244033004477, "rms_envelope_nrmse": 0.09704228037095074, "envelope_correlation": 0.9931995140498945, "rms_gain_error_db": -0.1061799480920258, "reference_mean_power": 0.00026104952136561866, "output_mean_power": 0.00025474455291474513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.392840173310503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.633554932680893e-05, "output_power": 9.249925621142329e-05, "reference_active": true, "delta_db": -0.17648337586918028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.160826072139944e-05, "output_power": 2.178415759841679e-05, "reference_active": true, "delta_db": 0.035209589348600724}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.769645916048805e-06, "output_power": 4.4719583020055754e-06, "reference_active": true, "delta_db": -0.279883941461606}], "active_band_count": 3, "active_band_mae_db": 0.163858968893129}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.77451695930137e-05, "log_rms_mae_db": 0.8269426131653078, "rms_envelope_nrmse": 0.053182516394315246, "envelope_correlation": 0.997918695625601, "rms_gain_error_db": -0.10547428407285091, "reference_mean_power": 0.00026200159904973233, "output_mean_power": 0.0002557151821954715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2678124187221727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604417147270571e-05, "output_power": 8.326294491587913e-05, "reference_active": true, "delta_db": -0.1426968975567457}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1040946271380912e-05, "output_power": 1.1128615956904889e-05, "reference_active": true, "delta_db": 0.03434858905566543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9369342076961927e-07, "output_power": 2.4864743235997285e-07, "reference_active": true, "delta_db": -0.7231023881901072}], "active_band_count": 3, "active_band_mae_db": 0.30004929160083943}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.588832427367551e-05, "log_rms_mae_db": 1.540699799867102, "rms_envelope_nrmse": 0.1633036065714083, "envelope_correlation": 0.9810307383880753, "rms_gain_error_db": -0.3969063549854467, "reference_mean_power": 4.0575597065913316e-05, "output_mean_power": 3.703175407319402e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.7364679678483124e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6458578452566544e-05, "output_power": 1.5143389304693678e-05, "reference_active": true, "delta_db": -0.3616923466330051}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.144001855335748e-06, "output_power": 2.2265836179356154e-06, "reference_active": true, "delta_db": -1.498437842256167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.079905260013679e-07, "output_power": 6.792147313625159e-07, "reference_active": true, "delta_db": -1.2607422043328227}], "active_band_count": 3, "active_band_mae_db": 1.040290797740665}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.068450322528164e-05, "log_rms_mae_db": 0.9104036096142047, "rms_envelope_nrmse": 0.09866373615780609, "envelope_correlation": 0.9936451178942992, "rms_gain_error_db": -0.3981856811662997, "reference_mean_power": 4.071142209359612e-05, "output_mean_power": 3.714477268157907e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.516451938680711e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5218014411010514e-05, "output_power": 1.397183310697308e-05, "reference_active": true, "delta_db": -0.371046016636294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2444589918408885e-06, "output_power": 9.853939155444564e-07, "reference_active": true, "delta_db": -1.0137071371431834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2414094224167626e-08, "output_power": 3.746553095826009e-08, "reference_active": true, "delta_db": -1.458161933277471}], "active_band_count": 3, "active_band_mae_db": 0.9476383623523161}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.7354442023779843e-05, "log_rms_mae_db": 1.5471416920698593, "rms_envelope_nrmse": 0.16863607620049093, "envelope_correlation": 0.9780968915302889, "rms_gain_error_db": -0.37767990297631293, "reference_mean_power": 9.786447369332928e-06, "output_mean_power": 8.971335010600772e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.593718108385407e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.378843387910109e-06, "output_power": 2.184159013210011e-06, "reference_active": true, "delta_db": -0.3708159792913696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.926849454092717e-07, "output_power": 8.009638718996138e-07, "reference_active": true, "delta_db": -0.47085283387762605}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6983089704559518e-07, "output_power": 1.1053297539983002e-07, "reference_active": true, "delta_db": -1.8652484274432313}], "active_band_count": 3, "active_band_mae_db": 0.9023057468707423}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.2351945663576383e-05, "log_rms_mae_db": 0.968014401039022, "rms_envelope_nrmse": 0.09266104936598833, "envelope_correlation": 0.9938459325244196, "rms_gain_error_db": -0.37936740733428226, "reference_mean_power": 9.83170860207206e-06, "output_mean_power": 9.009325075444158e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.2206731147886266e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.414723334288e-06, "output_power": 2.2026790538567567e-06, "reference_active": true, "delta_db": -0.3991615692772202}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.701511911331394e-07, "output_power": 3.7172644018613386e-07, "reference_active": true, "delta_db": 0.018443014604731415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1816168225696604e-08, "output_power": 6.859290139292959e-09, "reference_active": true, "delta_db": -2.3619749200433766}], "active_band_count": 3, "active_band_mae_db": 0.9265265013084427}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9028903322308243, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3b0d0b82c27dcb641449938861efe2c2008891f11033dd44c4fdd3322ad36d98", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9811962775203705, "standard_deviation": 0.0031217876028513144, "minimum": 0.977794138671054, "maximum": 0.9865313166110956, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5137376984969109, "standard_deviation": 0.031047059683200175, "minimum": 1.4515888928834138, "maximum": 1.5486357160618476, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8198673976371741, "standard_deviation": 0.19682169389056497, "minimum": 0.45954915091704956, "maximum": 1.040290797740665, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16354812310943292, "standard_deviation": 0.013696961530077217, "minimum": 0.1389971946218549, "maximum": 0.18091893100820924, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942220115647338, "standard_deviation": 0.0013407949101901012, "minimum": 0.9928204701092639, "maximum": 0.9966640655593135, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9036180185896623, "standard_deviation": 0.019304228193258582, "minimum": 0.877880120011192, "maximum": 0.9454178626178061, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.892039660212087, "standard_deviation": 0.27062859404360884, "minimum": 0.38838161645287217, "maximum": 1.2125087869689015, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09737374621497305, "standard_deviation": 0.010981530248751956, "minimum": 0.07611245867199341, "maximum": 0.10885081102949086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9782871411523236, "standard_deviation": 0.002009007923933934, "minimum": 0.974715493509813, "maximum": 0.9814221498978553, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5066873155011344, "standard_deviation": 0.034038462844981444, "minimum": 1.4435889981664287, "maximum": 1.5471416920698593, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9899103872711927, "standard_deviation": 0.3267207663590152, "minimum": 0.5060362424180652, "maximum": 1.4237672906957648, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1671870905400531, "standard_deviation": 0.007044392237258855, "minimum": 0.15619525313328012, "maximum": 0.17874663983089298, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9925242533864121, "standard_deviation": 0.0014333548891748862, "minimum": 0.9904014425217696, "maximum": 0.9941970699903857, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.936968549976934, "standard_deviation": 0.046071489508011335, "minimum": 0.8721819460300227, "maximum": 1.0243228122584438, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0396534530884418, "standard_deviation": 0.40313255787911745, "minimum": 0.44493636630676064, "maximum": 1.6796023863106297, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09827404550745582, "standard_deviation": 0.0073961349754635645, "minimum": 0.09048997367308542, "maximum": 0.10932727652902101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916825760981837, "standard_deviation": 0.0015713414991833801, "minimum": 0.9896136990736082, "maximum": 0.9938159317839318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4951442711577414, "standard_deviation": 0.05038441241541246, "minimum": 1.4218972494072066, "maximum": 1.568008411469022, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.42886012098491566, "standard_deviation": 0.25027955806449254, "minimum": 0.15267709565198154, "maximum": 0.7752738966413081, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10782126556454974, "standard_deviation": 0.010053613453398894, "minimum": 0.09510558697062889, "maximum": 0.12177324219658434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99762299211689, "standard_deviation": 0.00043576103320435014, "minimum": 0.9969343023155078, "maximum": 0.9981674244387132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8694816097727003, "standard_deviation": 0.05854198194117334, "minimum": 0.7874558677205279, "maximum": 0.9841434983868296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6227479323348135, "standard_deviation": 0.2645974043366572, "minimum": 0.22749907595200317, "maximum": 0.996926299972316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05905730779513048, "standard_deviation": 0.005260507724477072, "minimum": 0.053182516394315246, "maximum": 0.06941452031618865, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2454646070708303, "standard_deviation": 0.034144895912190665, "minimum": 0.21224910073529604, "maximum": 0.32230022389576757, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.2842832191065225, "standard_deviation": 0.009023847167749257, "minimum": 5.276234836776896, "maximum": 5.302957242756016, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.22032706841532412, "standard_deviation": 0.010976982005128337, "minimum": 0.20473091281755915, "maximum": 0.2339912642267531, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12153945760360657, "standard_deviation": 0.033127458894126924, "minimum": -0.16764291751395483, "maximum": -0.057756335291799715, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951059683706056, "standard_deviation": 0.0004069698174625614, "minimum": 0.9945656822608433, "maximum": 0.9956807121455563, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7145216960435044, "standard_deviation": 0.019750361801603587, "minimum": 0.6884891906024492, "maximum": 0.7468635055756105, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35890110874301895, "standard_deviation": 0.17643543160294542, "minimum": 0.09669236906639982, "maximum": 0.5620554727321302, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07465019065645698, "standard_deviation": 0.00317925626028153, "minimum": 0.07042143398781932, "maximum": 0.07898360864323509, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986950255107763, "standard_deviation": 0.00012895942305940123, "minimum": 0.9984128097512145, "maximum": 0.9988181614343963, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3946215494580127, "standard_deviation": 0.02074191958366031, "minimum": 0.3671766301332125, "maximum": 0.42897350447581245, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33929372107325273, "standard_deviation": 0.1705799863861378, "minimum": 0.11113854850395337, "maximum": 0.5235387171323455, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03939921613397547, "standard_deviation": 0.0015581526370538413, "minimum": 0.03730466677731802, "maximum": 0.04243092913359172, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0030", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.4089086055755615, "predicted_mos": 3.4089086055755615, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5f04abe68317dfa84103c7067851b4007df37550cf70b0592ba3150f2c51f4cb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.151034700218588, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7170228229012423, "全帯域_10ms_rms_envelope_nrmse": 0.06964538587432968, "全帯域_10ms_envelope_correlation": 0.9957125417085809, "全帯域_10ms_modulation_mae_db": 0.5116765886679863, "全帯域_40ms_log_rms_mae_db": 0.3735730533959415, "全帯域_40ms_rms_envelope_nrmse": 0.0371867112541865, "全帯域_40ms_envelope_correlation": 0.9987844983720314, "全帯域_40ms_modulation_mae_db": 0.26344606388172226, "300–1000 Hz_10ms_log_rms_mae_db": 1.7242413084219301, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10143432163436154, "300–1000 Hz_10ms_envelope_correlation": 0.9927547311666187, "300–1000 Hz_10ms_modulation_mae_db": 0.7977658599454337, "300–1000 Hz_40ms_log_rms_mae_db": 1.060695746790144, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05908837922870022, "300–1000 Hz_40ms_envelope_correlation": 0.9976512486778645, "300–1000 Hz_40ms_modulation_mae_db": 0.7643750600261913, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7599280188002098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17132932468687476, "1000–3000 Hz_10ms_envelope_correlation": 0.9784059781145509, "1000–3000 Hz_10ms_modulation_mae_db": 1.042705202005337, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0105394920688524, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09668494884779628, "1000–3000 Hz_40ms_envelope_correlation": 0.993025778857002, "1000–3000 Hz_40ms_modulation_mae_db": 0.8254866093524477, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0865452988548867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21774575748363728, "3000–6000 Hz_10ms_envelope_correlation": 0.9621776111395793, "3000–6000 Hz_10ms_modulation_mae_db": 0.8347668273280546, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2265859625030635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12127891789827333, "3000–6000 Hz_40ms_envelope_correlation": 0.9871574960022462, "3000–6000 Hz_40ms_modulation_mae_db": 0.8658183325310583, "mr_stft_db": 5.756637469140724, "band_mae_db": 0.16597917290221684, "rms_gain_error_db": -0.09734343476168786, "output_peak": 0.21551545337538242}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6852049010314784, "全帯域_10ms_rms_envelope_nrmse": 0.06915083122846186, "全帯域_10ms_envelope_correlation": 0.9958088198220216, "全帯域_10ms_modulation_mae_db": 0.42504422032708017, "全帯域_40ms_log_rms_mae_db": 0.3388881221321729, "全帯域_40ms_rms_envelope_nrmse": 0.033807342515871315, "全帯域_40ms_envelope_correlation": 0.999064846112857, "全帯域_40ms_modulation_mae_db": 0.28355538342616454, "300–1000 Hz_10ms_log_rms_mae_db": 1.7121227135693928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10037054653749934, "300–1000 Hz_10ms_envelope_correlation": 0.9928229156881068, "300–1000 Hz_10ms_modulation_mae_db": 0.8404552328859497, "300–1000 Hz_40ms_log_rms_mae_db": 0.9760216859487522, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05078395620280557, "300–1000 Hz_40ms_envelope_correlation": 0.9982571260686843, "300–1000 Hz_40ms_modulation_mae_db": 0.864650512132124, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7254284475404527, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17417297878370494, "1000–3000 Hz_10ms_envelope_correlation": 0.977669533047408, "1000–3000 Hz_10ms_modulation_mae_db": 0.9377383124660127, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9526129943501652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09269663617727597, "1000–3000 Hz_40ms_envelope_correlation": 0.9936988452746937, "1000–3000 Hz_40ms_modulation_mae_db": 0.7667841143815494, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0559799214068932, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19626789138876952, "3000–6000 Hz_10ms_envelope_correlation": 0.9695241297795335, "3000–6000 Hz_10ms_modulation_mae_db": 0.7761326088391547, "3000–6000 Hz_40ms_log_rms_mae_db": 1.203529396266449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10362697989097179, "3000–6000 Hz_40ms_envelope_correlation": 0.9912505468481958, "3000–6000 Hz_40ms_modulation_mae_db": 0.9892521455234285, "mr_stft_db": 5.743799121845481, "band_mae_db": 0.18086213695542974, "rms_gain_error_db": -0.11235293722453112, "output_peak": 0.20683211017668351}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7213628645353697, "全帯域_10ms_rms_envelope_nrmse": 0.07321320894929365, "全帯域_10ms_envelope_correlation": 0.9952873613984327, "全帯域_10ms_modulation_mae_db": 0.18929061918599024, "全帯域_40ms_log_rms_mae_db": 0.37445891867603553, "全帯域_40ms_rms_envelope_nrmse": 0.03787329079004852, "全帯域_40ms_envelope_correlation": 0.9987881334283802, "全帯域_40ms_modulation_mae_db": 0.23339068855176404, "300–1000 Hz_10ms_log_rms_mae_db": 1.7694978017569427, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10240679833930202, "300–1000 Hz_10ms_envelope_correlation": 0.9925887706365052, "300–1000 Hz_10ms_modulation_mae_db": 0.3762737410952797, "300–1000 Hz_40ms_log_rms_mae_db": 1.0551905813782836, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057675638888756826, "300–1000 Hz_40ms_envelope_correlation": 0.9977674318941753, "300–1000 Hz_40ms_modulation_mae_db": 0.5053899932092781, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7742900886725315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16638935762226548, "1000–3000 Hz_10ms_envelope_correlation": 0.9802265598450318, "1000–3000 Hz_10ms_modulation_mae_db": 1.299643734808426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0031738030363913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1023265900736954, "1000–3000 Hz_40ms_envelope_correlation": 0.992956274607532, "1000–3000 Hz_40ms_modulation_mae_db": 1.2118749169563767, "3000–6000 Hz_10ms_log_rms_mae_db": 2.019913862403254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20063884247285999, "3000–6000 Hz_10ms_envelope_correlation": 0.9681119990398462, "3000–6000 Hz_10ms_modulation_mae_db": 0.9029090442434385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1683279073168318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10583815551711909, "3000–6000 Hz_40ms_envelope_correlation": 0.9902293574250459, "3000–6000 Hz_40ms_modulation_mae_db": 1.2201199817108226, "mr_stft_db": 5.729217716538258, "band_mae_db": 0.18232661129178718, "rms_gain_error_db": -0.11777610128803101, "output_peak": 0.21638496350981828}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6837318412942056, "全帯域_10ms_rms_envelope_nrmse": 0.07102436665451965, "全帯域_10ms_envelope_correlation": 0.9955311479234127, "全帯域_10ms_modulation_mae_db": 0.543224378925479, "全帯域_40ms_log_rms_mae_db": 0.3633299906062321, "全帯域_40ms_rms_envelope_nrmse": 0.036744701652200076, "全帯域_40ms_envelope_correlation": 0.9988151368972453, "全帯域_40ms_modulation_mae_db": 0.2493154146474583, "300–1000 Hz_10ms_log_rms_mae_db": 1.732111351951322, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10408520333723958, "300–1000 Hz_10ms_envelope_correlation": 0.9922902650295751, "300–1000 Hz_10ms_modulation_mae_db": 0.667935590848578, "300–1000 Hz_40ms_log_rms_mae_db": 1.0588046113913625, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059229291587449856, "300–1000 Hz_40ms_envelope_correlation": 0.9975675569289292, "300–1000 Hz_40ms_modulation_mae_db": 0.8011384998571667, "1000–3000 Hz_10ms_log_rms_mae_db": 1.828835176559998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.173865388901408, "1000–3000 Hz_10ms_envelope_correlation": 0.9782018707695701, "1000–3000 Hz_10ms_modulation_mae_db": 0.6684700497476417, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0516462251906777, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10111574565964654, "1000–3000 Hz_40ms_envelope_correlation": 0.9931231850349574, "1000–3000 Hz_40ms_modulation_mae_db": 0.8798161764213805, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9982431359954695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1881576954804593, "3000–6000 Hz_10ms_envelope_correlation": 0.9721769997698534, "3000–6000 Hz_10ms_modulation_mae_db": 1.2022193056619797, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1571129727322873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09565716159527067, "3000–6000 Hz_40ms_envelope_correlation": 0.9928028277393776, "3000–6000 Hz_40ms_modulation_mae_db": 1.345053850947232, "mr_stft_db": 5.737300606269307, "band_mae_db": 0.2702309146092888, "rms_gain_error_db": -0.0965103356553814, "output_peak": 0.23176328704991736}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6795982849349179, "全帯域_10ms_rms_envelope_nrmse": 0.06775403385182503, "全帯域_10ms_envelope_correlation": 0.995945627136705, "全帯域_10ms_modulation_mae_db": 0.10905057153095216, "全帯域_40ms_log_rms_mae_db": 0.3668461728321441, "全帯域_40ms_rms_envelope_nrmse": 0.03519417873180804, "全帯域_40ms_envelope_correlation": 0.9989254377345373, "全帯域_40ms_modulation_mae_db": 0.27717249884064493, "300–1000 Hz_10ms_log_rms_mae_db": 1.7384793924951103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09709601661728313, "300–1000 Hz_10ms_envelope_correlation": 0.9932977364744162, "300–1000 Hz_10ms_modulation_mae_db": 0.12358415725177989, "300–1000 Hz_40ms_log_rms_mae_db": 1.0445822117836057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055099291009368376, "300–1000 Hz_40ms_envelope_correlation": 0.997901022620415, "300–1000 Hz_40ms_modulation_mae_db": 0.09896811565336318, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7837176949713733, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17413334261723662, "1000–3000 Hz_10ms_envelope_correlation": 0.9782428153468775, "1000–3000 Hz_10ms_modulation_mae_db": 1.4462212641402417, "1000–3000 Hz_40ms_log_rms_mae_db": 1.046813242162298, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10398428331406102, "1000–3000 Hz_40ms_envelope_correlation": 0.9927696868639299, "1000–3000 Hz_40ms_modulation_mae_db": 1.407148752034085, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0683855813063166, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1972386151264516, "3000–6000 Hz_10ms_envelope_correlation": 0.9690977584953264, "3000–6000 Hz_10ms_modulation_mae_db": 1.1894011221757224, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2100687734145703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10101233615868656, "3000–6000 Hz_40ms_envelope_correlation": 0.9912774786279216, "3000–6000 Hz_40ms_modulation_mae_db": 1.2944103848190354, "mr_stft_db": 5.764532718367651, "band_mae_db": 0.2326885616898922, "rms_gain_error_db": -0.09841680322655749, "output_peak": 0.21347256323587657}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.690427622838139, "全帯域_10ms_rms_envelope_nrmse": 0.06674586821070998, "全帯域_10ms_envelope_correlation": 0.996022833930966, "全帯域_10ms_modulation_mae_db": 0.5556022435724989, "全帯域_40ms_log_rms_mae_db": 0.3458626879437643, "全帯域_40ms_rms_envelope_nrmse": 0.035508151358594024, "全帯域_40ms_envelope_correlation": 0.9988417092519268, "全帯域_40ms_modulation_mae_db": 0.35788240543311106, "300–1000 Hz_10ms_log_rms_mae_db": 1.6849758320656216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09734538040192908, "300–1000 Hz_10ms_envelope_correlation": 0.9932243843284486, "300–1000 Hz_10ms_modulation_mae_db": 0.7231183772835369, "300–1000 Hz_40ms_log_rms_mae_db": 0.9519678833909446, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055067048252334154, "300–1000 Hz_40ms_envelope_correlation": 0.9978420289634023, "300–1000 Hz_40ms_modulation_mae_db": 0.7672844458274642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7931855569401347, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17617965436909772, "1000–3000 Hz_10ms_envelope_correlation": 0.9772694756198932, "1000–3000 Hz_10ms_modulation_mae_db": 0.8350869980963019, "1000–3000 Hz_40ms_log_rms_mae_db": 1.018719274069919, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10472386597677372, "1000–3000 Hz_40ms_envelope_correlation": 0.9920202520441451, "1000–3000 Hz_40ms_modulation_mae_db": 0.786809890505117, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0784463800324997, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20124958077452298, "3000–6000 Hz_10ms_envelope_correlation": 0.9681783055453544, "3000–6000 Hz_10ms_modulation_mae_db": 1.3203377983173434, "3000–6000 Hz_40ms_log_rms_mae_db": 1.21008093103618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10667880850054438, "3000–6000 Hz_40ms_envelope_correlation": 0.9911447010327036, "3000–6000 Hz_40ms_modulation_mae_db": 1.2659870688230468, "mr_stft_db": 5.7533771511291105, "band_mae_db": 0.1554728515625777, "rms_gain_error_db": -0.06840484944953767, "output_peak": 0.2433254631466708}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7124825950616891, "全帯域_10ms_rms_envelope_nrmse": 0.07834179725636688, "全帯域_10ms_envelope_correlation": 0.9946108017239846, "全帯域_10ms_modulation_mae_db": 0.27992512837225647, "全帯域_40ms_log_rms_mae_db": 0.38687517942735306, "全帯域_40ms_rms_envelope_nrmse": 0.04620156959893058, "全帯域_40ms_envelope_correlation": 0.9981474462279882, "全帯域_40ms_modulation_mae_db": 0.28057630197529787, "300–1000 Hz_10ms_log_rms_mae_db": 1.7759090039857912, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1152365477553783, "300–1000 Hz_10ms_envelope_correlation": 0.9905098426129658, "300–1000 Hz_10ms_modulation_mae_db": 0.5511842094657682, "300–1000 Hz_40ms_log_rms_mae_db": 1.0985412505565177, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06674474664075, "300–1000 Hz_40ms_envelope_correlation": 0.9968698249753795, "300–1000 Hz_40ms_modulation_mae_db": 0.8136790081418906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7250766048704247, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1767445102678759, "1000–3000 Hz_10ms_envelope_correlation": 0.9790414481573542, "1000–3000 Hz_10ms_modulation_mae_db": 1.1921102215558255, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9710447590921959, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1110583296176421, "1000–3000 Hz_40ms_envelope_correlation": 0.9935364654513101, "1000–3000 Hz_40ms_modulation_mae_db": 1.3740676824356086, "3000–6000 Hz_10ms_log_rms_mae_db": 2.042404551879249, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1956265392072509, "3000–6000 Hz_10ms_envelope_correlation": 0.9696599331445489, "3000–6000 Hz_10ms_modulation_mae_db": 1.4605696426269494, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2165806175552898, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10906256015470087, "3000–6000 Hz_40ms_envelope_correlation": 0.9897140075769354, "3000–6000 Hz_40ms_modulation_mae_db": 1.6948387898347328, "mr_stft_db": 5.739779597709066, "band_mae_db": 0.30766617544294655, "rms_gain_error_db": -0.12690156124524063, "output_peak": 0.2179272130489841}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7103485283782626, "全帯域_10ms_rms_envelope_nrmse": 0.07659381373703222, "全帯域_10ms_envelope_correlation": 0.9948091502729154, "全帯域_10ms_modulation_mae_db": 0.16121155280436475, "全帯域_40ms_log_rms_mae_db": 0.3685871105893623, "全帯域_40ms_rms_envelope_nrmse": 0.04217806881278045, "全帯域_40ms_envelope_correlation": 0.9984341650493087, "全帯域_40ms_modulation_mae_db": 0.1435721132749436, "300–1000 Hz_10ms_log_rms_mae_db": 1.7095596710714402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11282393034193629, "300–1000 Hz_10ms_envelope_correlation": 0.9908695465371338, "300–1000 Hz_10ms_modulation_mae_db": 0.5594890191712129, "300–1000 Hz_40ms_log_rms_mae_db": 1.0036927089325458, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06536369778468563, "300–1000 Hz_40ms_envelope_correlation": 0.9969492946024803, "300–1000 Hz_40ms_modulation_mae_db": 0.46269524436900417, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7747609957758133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17359116627569582, "1000–3000 Hz_10ms_envelope_correlation": 0.9782465636466976, "1000–3000 Hz_10ms_modulation_mae_db": 1.5146522750159825, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9730594564843558, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09733433673695681, "1000–3000 Hz_40ms_envelope_correlation": 0.9936720184056232, "1000–3000 Hz_40ms_modulation_mae_db": 1.0597631212842937, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0642489646587783, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18761168948384752, "3000–6000 Hz_10ms_envelope_correlation": 0.9721043097246097, "3000–6000 Hz_10ms_modulation_mae_db": 0.8593179971102191, "3000–6000 Hz_40ms_log_rms_mae_db": 1.221342249903857, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09876838444823925, "3000–6000 Hz_40ms_envelope_correlation": 0.9916317199410811, "3000–6000 Hz_40ms_modulation_mae_db": 0.8511011398779534, "mr_stft_db": 5.739554901629998, "band_mae_db": 0.19500538617375215, "rms_gain_error_db": -0.1117296981518378, "output_peak": 0.22280141240409887}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.834684443620648, "spectral_convergence": 0.24446009180266498, "spectral_convergence_status": "ok", "amplitude_floor": 6.716189184049609e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.614067340518994, "spectral_convergence": 0.23878534792823092, "spectral_convergence_status": "ok", "amplitude_floor": 4.979433430044932e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8211606232825295, "spectral_convergence": 0.2504095472579316, "spectral_convergence_status": "ok", "amplitude_floor": 3.630042469105695e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.462171010234058e-05, "output_power": 7.968067833833241e-05, "delta_db": 0.28487825145655365, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003077822779549395, "output_power": 0.0002952858065869557, "delta_db": -0.1800103725564942, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 5.338393512143839e-05, "output_power": 5.2022848896401494e-05, "delta_db": -0.11216452514217148, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.1515857114073014e-05, "output_power": 2.1177399737303904e-05, "delta_db": -0.06886017201711728, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.940885952831811e-05, "output_power": 2.776232988584936e-05, "delta_db": -0.2502227253324372, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.017666424897867e-06, "output_power": 7.835632959594211e-06, "delta_db": -0.09973899090852709, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16597917290221684, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09734343476168786, "pitch": null, "output_rms_dbfs": -33.15365828235769, "output_peak": 0.21551545337538242, "reference_rms_dbfs": -33.056314847596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.756637469140724}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.650394450462669e-05, "log_rms_mae_db": 0.7170228229012423, "rms_envelope_nrmse": 0.06964538587432968, "envelope_correlation": 0.9957125417085809, "rms_gain_error_db": -0.09726446685226091, "reference_mean_power": 0.0004951134756771775, "output_mean_power": 0.0004841481761551351}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.645854738282274e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010486873457313227, "output_power": 0.00010358766276390548, "reference_active": true, "delta_db": -0.05337993259465608}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5071002541192514e-05, "output_power": 2.2337549687764375e-05, "reference_active": true, "delta_db": -0.5013616934512127}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059292105678353e-06, "output_power": 4.037020369895291e-06, "reference_active": true, "delta_db": -0.9802881399580898}], "active_band_count": 3, "active_band_mae_db": 0.5116765886679863}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.020301765617229e-05, "log_rms_mae_db": 0.3735730533959415, "rms_envelope_nrmse": 0.0371867112541865, "envelope_correlation": 0.9987844983720314, "rms_gain_error_db": -0.09685008222561886, "reference_mean_power": 0.0004970274687625529, "output_mean_power": 0.0004860661560277879}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.472129124053938e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.182773781582673e-05, "output_power": 9.017732079138895e-05, "reference_active": true, "delta_db": -0.07876557488528294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2533326164459841e-05, "output_power": 1.0908192664954301e-05, "reference_active": true, "delta_db": -0.6031354177431616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4723702388808576e-07, "output_power": 2.4114029464706637e-07, "reference_active": true, "delta_db": -0.10843719901672227}], "active_band_count": 3, "active_band_mae_db": 0.26344606388172226}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.432622539351727e-05, "log_rms_mae_db": 1.7242413084219301, "rms_envelope_nrmse": 0.10143432163436154, "envelope_correlation": 0.9927547311666187, "rms_gain_error_db": -0.1874087271274016, "reference_mean_power": 0.00026104952136561866, "output_mean_power": 0.00025002418975961845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.392840173310503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.633554932680893e-05, "output_power": 8.9960248829518e-05, "reference_active": true, "delta_db": -0.2973593014716869}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.160826072139944e-05, "output_power": 1.860365654724529e-05, "reference_active": true, "delta_db": -0.6502149817594607}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.769645916048805e-06, "output_power": 3.4191157997839902e-06, "reference_active": true, "delta_db": -1.4457232966051539}], "active_band_count": 3, "active_band_mae_db": 0.7977658599454337}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.77451695930137e-05, "log_rms_mae_db": 1.060695746790144, "rms_envelope_nrmse": 0.05908837922870022, "envelope_correlation": 0.9976512486778645, "rms_gain_error_db": -0.1868254789536819, "reference_mean_power": 0.00026200159904973233, "output_mean_power": 0.000250969759226169}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2678124187221727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604417147270571e-05, "output_power": 8.029422743424371e-05, "reference_active": true, "delta_db": -0.3003713326653761}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1040946271380912e-05, "output_power": 9.05682026353065e-06, "reference_active": true, "delta_db": -0.8603054740675435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9369342076961927e-07, "output_power": 2.262816726635556e-07, "reference_active": true, "delta_db": -1.132448373345654}], "active_band_count": 3, "active_band_mae_db": 0.7643750600261913}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.588832427367551e-05, "log_rms_mae_db": 1.7599280188002098, "rms_envelope_nrmse": 0.17132932468687476, "envelope_correlation": 0.9784059781145509, "rms_gain_error_db": -0.2220455518640116, "reference_mean_power": 4.0575597065913316e-05, "output_mean_power": 3.855319402418644e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.7364679678483124e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6458578452566544e-05, "output_power": 1.5835389226446622e-05, "reference_active": true, "delta_db": -0.1676357935962377}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.144001855335748e-06, "output_power": 2.308515365213736e-06, "reference_active": true, "delta_db": -1.3415002414185762}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.079905260013679e-07, "output_power": 6.254366976718157e-07, "reference_active": true, "delta_db": -1.6189795710011967}], "active_band_count": 3, "active_band_mae_db": 1.042705202005337}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.068450322528164e-05, "log_rms_mae_db": 1.0105394920688524, "rms_envelope_nrmse": 0.09668494884779628, "envelope_correlation": 0.993025778857002, "rms_gain_error_db": -0.2206731147886813, "reference_mean_power": 4.071142209359612e-05, "output_mean_power": 3.869447526145135e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.516451938680711e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5218014411010514e-05, "output_power": 1.4478029563436854e-05, "reference_active": true, "delta_db": -0.21648531927100506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2444589918408885e-06, "output_power": 9.890799049146884e-07, "reference_active": true, "delta_db": -0.9974921163424497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2414094224167626e-08, "output_power": 3.919221833401006e-08, "reference_active": true, "delta_db": -1.2624823924438884}], "active_band_count": 3, "active_band_mae_db": 0.8254866093524477}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.7354442023779843e-05, "log_rms_mae_db": 2.0865452988548867, "rms_envelope_nrmse": 0.21774575748363728, "envelope_correlation": 0.9621776111395793, "rms_gain_error_db": -0.12230334329441869, "reference_mean_power": 9.786447369332928e-06, "output_mean_power": 9.514691916938013e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.593718108385407e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.378843387910109e-06, "output_power": 2.2599807212722743e-06, "reference_active": true, "delta_db": -0.22261116609017678}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.926849454092717e-07, "output_power": 5.985432292328396e-07, "reference_active": true, "delta_db": -1.736026883286161}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6983089704559518e-07, "output_power": 1.4977883914799268e-07, "reference_active": true, "delta_db": -0.545662432607826}], "active_band_count": 3, "active_band_mae_db": 0.8347668273280546}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.2351945663576383e-05, "log_rms_mae_db": 1.2265859625030635, "rms_envelope_nrmse": 0.12127891789827333, "envelope_correlation": 0.9871574960022462, "rms_gain_error_db": -0.12439502152611996, "reference_mean_power": 9.83170860207206e-06, "output_mean_power": 9.554093695865093e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.2206731147886266e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.414723334288e-06, "output_power": 2.211558925625666e-06, "reference_active": true, "delta_db": -0.3816886349714613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.701511911331394e-07, "output_power": 2.8212124389901843e-07, "reference_active": true, "delta_db": -1.1794336124251799}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1816168225696604e-08, "output_power": 9.307721719157877e-09, "reference_active": true, "delta_db": -1.0363327501965338}], "active_band_count": 3, "active_band_mae_db": 0.8658183325310583}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8847637210957341, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0cceca86488c471c5d3201da67e78456ed1a8ef3bfee046312632d4e277e774f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9784130305684229, "standard_deviation": 0.0008980634914469781, "minimum": 0.9772694756198932, "maximum": 0.9802265598450318, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7706528230163672, "standard_deviation": 0.03449878931453137, "minimum": 1.7250766048704247, "maximum": 1.828835176559998, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.117078507229471, "standard_deviation": 0.2985255981054888, "minimum": 0.6684700497476417, "maximum": 1.5146522750159825, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1733007154405199, "standard_deviation": 0.0032438643120131154, "minimum": 0.16638935762226548, "maximum": 0.1767445102678759, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9931003133173992, "standard_deviation": 0.0005578002397974896, "minimum": 0.9920202520441451, "maximum": 0.9936988452746937, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0034511558068568, "standard_deviation": 0.03594603239154744, "minimum": 0.9526129943501652, "maximum": 1.0516462251906777, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0389689079213573, "standard_deviation": 0.2636961154723443, "minimum": 0.7667841143815494, "maximum": 1.407148752034085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10124059205048098, "standard_deviation": 0.005685267750808607, "minimum": 0.09269663617727597, "maximum": 0.1110583296176421, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9688788808298315, "standard_deviation": 0.003124584665569722, "minimum": 0.9621776111395793, "maximum": 0.9721769997698534, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0517709620671685, "standard_deviation": 0.0300906404178141, "minimum": 1.9982431359954695, "maximum": 2.0865452988548867, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0682067932878578, "standard_deviation": 0.2566057259138611, "minimum": 0.7761326088391547, "maximum": 1.4605696426269494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19806707642722488, "standard_deviation": 0.009421733131204703, "minimum": 0.18761168948384752, "maximum": 0.21774575748363728, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906510168991884, "standard_deviation": 0.0016838796304468543, "minimum": 0.9871574960022462, "maximum": 0.9928028277393776, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.201703601341066, "standard_deviation": 0.02527529686382689, "minimum": 1.1571129727322873, "maximum": 1.2265859625030635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1908227117584138, "standard_deviation": 0.28201498287320403, "minimum": 0.8511011398779534, "maximum": 1.6948387898347328, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10524041302047574, "standard_deviation": 0.007823722051841495, "minimum": 0.09565716159527067, "maximum": 0.12127891789827333, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922947740592213, "standard_deviation": 0.0010463754486050632, "minimum": 0.9905098426129658, "maximum": 0.9932977364744162, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.730862134414694, "standard_deviation": 0.030556428672349676, "minimum": 1.6849758320656216, "maximum": 1.7759090039857912, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5799757734934423, "standard_deviation": 0.23741022112181798, "minimum": 0.12358415725177989, "maximum": 0.8404552328859497, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10384984312061617, "standard_deviation": 0.006742101336562886, "minimum": 0.09709601661728313, "maximum": 0.1152365477553783, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976006918414163, "standard_deviation": 0.000473480875777123, "minimum": 0.9968698249753795, "maximum": 0.9982571260686843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0311870850215195, "standard_deviation": 0.0492812824997872, "minimum": 0.9519678833909446, "maximum": 1.0985412505565177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6347726099020603, "standard_deviation": 0.2619493074274562, "minimum": 0.09896811565336318, "maximum": 0.864650512132124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05863150619935632, "standard_deviation": 0.00533710438085394, "minimum": 0.05078395620280557, "maximum": 0.06674474664075, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21127897632848638, "standard_deviation": 0.0540090502635268, "minimum": 0.1554728515625777, "maximum": 0.30766617544294655, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.745524910328699, "standard_deviation": 0.01165966875793283, "minimum": 5.729217716538258, "maximum": 5.764532718367651, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.221002808243429, "standard_deviation": 0.011561833244437278, "minimum": 0.20683211017668351, "maximum": 0.2433254631466708, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10367946512535062, "standard_deviation": 0.017894740204094137, "minimum": -0.12690156124524063, "maximum": -0.06840484944953767, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954660354896273, "standard_deviation": 0.0005233663164647849, "minimum": 0.9946108017239846, "maximum": 0.996022833930966, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7000224326219131, "standard_deviation": 0.01690502882603787, "minimum": 0.6795982849349179, "maximum": 0.7213628645353697, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.346878162923326, "standard_deviation": 0.18351201913012188, "minimum": 0.10905057153095216, "maximum": 0.5556022435724989, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07155866322031737, "standard_deviation": 0.004165901171694047, "minimum": 0.06674586821070998, "maximum": 0.07834179725636688, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987251716342844, "standard_deviation": 0.00029352423946540695, "minimum": 0.9981474462279882, "maximum": 0.999064846112857, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.36480265445037574, "standard_deviation": 0.015621925189135183, "minimum": 0.3388881221321729, "maximum": 0.38687517942735306, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2611138587538884, "standard_deviation": 0.06004012805020095, "minimum": 0.1435721132749436, "maximum": 0.35788240543311106, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038086751839302435, "standard_deviation": 0.004115742998489895, "minimum": 0.033807342515871315, "maximum": 0.04620156959893058, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0030", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.298860549926758, "predicted_mos": 3.298860549926758, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "27bb09f2738dd537f100f5e8a907427d42a265ac866b9fd3d2284cd77e5d0db5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.080157799646258, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.716795600357177, "全帯域_10ms_rms_envelope_nrmse": 0.0783424793381591, "全帯域_10ms_envelope_correlation": 0.9945915643625846, "全帯域_10ms_modulation_mae_db": 0.3572116021309569, "全帯域_40ms_log_rms_mae_db": 0.39201637603451545, "全帯域_40ms_rms_envelope_nrmse": 0.043936500977798595, "全帯域_40ms_envelope_correlation": 0.9983275175953141, "全帯域_40ms_modulation_mae_db": 0.20724989103279923, "300–1000 Hz_10ms_log_rms_mae_db": 1.4599146080922876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10949365807495758, "300–1000 Hz_10ms_envelope_correlation": 0.9914246590598647, "300–1000 Hz_10ms_modulation_mae_db": 0.6194582445261501, "300–1000 Hz_40ms_log_rms_mae_db": 0.9065996378784458, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06606749320577592, "300–1000 Hz_40ms_envelope_correlation": 0.9968960222108503, "300–1000 Hz_40ms_modulation_mae_db": 0.5208959035421105, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3813521555140313, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15478599267433826, "1000–3000 Hz_10ms_envelope_correlation": 0.9828032146093916, "1000–3000 Hz_10ms_modulation_mae_db": 0.5886286774265853, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8351202714686572, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09560627300863052, "1000–3000 Hz_40ms_envelope_correlation": 0.9937230409350668, "1000–3000 Hz_40ms_modulation_mae_db": 0.9023789831519782, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4251255287708633, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1508177562700964, "3000–6000 Hz_10ms_envelope_correlation": 0.9820853996233702, "3000–6000 Hz_10ms_modulation_mae_db": 0.5489740217895877, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8829924353260394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08581412998629015, "3000–6000 Hz_40ms_envelope_correlation": 0.993895556371717, "3000–6000 Hz_40ms_modulation_mae_db": 0.5306718386661918, "mr_stft_db": 5.266199768156878, "band_mae_db": 0.16967013157451769, "rms_gain_error_db": -0.11961175577444655, "output_peak": 0.2025860216287537}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7482369314751899, "全帯域_10ms_rms_envelope_nrmse": 0.07143752463716789, "全帯域_10ms_envelope_correlation": 0.9955480795114405, "全帯域_10ms_modulation_mae_db": 0.40725009819575425, "全帯域_40ms_log_rms_mae_db": 0.4238590940825517, "全帯域_40ms_rms_envelope_nrmse": 0.042088968098824425, "全帯域_40ms_envelope_correlation": 0.9984964920215776, "全帯域_40ms_modulation_mae_db": 0.3273250464150519, "300–1000 Hz_10ms_log_rms_mae_db": 1.3880260837124943, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10102506317431788, "300–1000 Hz_10ms_envelope_correlation": 0.9928576304068968, "300–1000 Hz_10ms_modulation_mae_db": 0.3996293369200968, "300–1000 Hz_40ms_log_rms_mae_db": 0.8140300374280487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06209224527473825, "300–1000 Hz_40ms_envelope_correlation": 0.9974328886142924, "300–1000 Hz_40ms_modulation_mae_db": 0.6346368231645697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4194548480456253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17240238372061067, "1000–3000 Hz_10ms_envelope_correlation": 0.9791118480644512, "1000–3000 Hz_10ms_modulation_mae_db": 1.0780000610450253, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8360916415070762, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10595089045486512, "1000–3000 Hz_40ms_envelope_correlation": 0.9930559832738095, "1000–3000 Hz_40ms_modulation_mae_db": 0.806099912730223, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3926177196677232, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16445330675466774, "3000–6000 Hz_10ms_envelope_correlation": 0.979085693831691, "3000–6000 Hz_10ms_modulation_mae_db": 1.1730364518685674, "3000–6000 Hz_40ms_log_rms_mae_db": 0.910928715123711, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1031807834010842, "3000–6000 Hz_40ms_envelope_correlation": 0.9917441314045271, "3000–6000 Hz_40ms_modulation_mae_db": 1.2797995302818772, "mr_stft_db": 5.2555753143407875, "band_mae_db": 0.31268302728448044, "rms_gain_error_db": -0.13016607827091248, "output_peak": 0.2184492517328369}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7184595781778947, "全帯域_10ms_rms_envelope_nrmse": 0.07346972823542237, "全帯域_10ms_envelope_correlation": 0.995208955419899, "全帯域_10ms_modulation_mae_db": 0.3890590931010469, "全帯域_40ms_log_rms_mae_db": 0.4080872155004704, "全帯域_40ms_rms_envelope_nrmse": 0.040760115802796135, "全帯域_40ms_envelope_correlation": 0.998510684439649, "全帯域_40ms_modulation_mae_db": 0.38046038179135216, "300–1000 Hz_10ms_log_rms_mae_db": 1.4256295622438089, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10573905484905098, "300–1000 Hz_10ms_envelope_correlation": 0.9919446224187084, "300–1000 Hz_10ms_modulation_mae_db": 0.54781405979966, "300–1000 Hz_40ms_log_rms_mae_db": 0.9015247138038793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059255943777000344, "300–1000 Hz_40ms_envelope_correlation": 0.9974480847923369, "300–1000 Hz_40ms_modulation_mae_db": 0.4653411809716001, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3886779835867271, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16378277314010029, "1000–3000 Hz_10ms_envelope_correlation": 0.9818373653658401, "1000–3000 Hz_10ms_modulation_mae_db": 0.6903053714814401, "1000–3000 Hz_40ms_log_rms_mae_db": 0.879669314628699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10091246776188063, "1000–3000 Hz_40ms_envelope_correlation": 0.9944930649722069, "1000–3000 Hz_40ms_modulation_mae_db": 0.8755380793212554, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3882348334388979, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15818694538290928, "3000–6000 Hz_10ms_envelope_correlation": 0.9801986857305276, "3000–6000 Hz_10ms_modulation_mae_db": 0.682582021511701, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8228396375891247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09039424966373749, "3000–6000 Hz_40ms_envelope_correlation": 0.9930614985559452, "3000–6000 Hz_40ms_modulation_mae_db": 0.8645781986957138, "mr_stft_db": 5.2600357276182805, "band_mae_db": 0.24888939979457214, "rms_gain_error_db": -0.09539360154695847, "output_peak": 0.22683547654888198}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.72805175759953, "全帯域_10ms_rms_envelope_nrmse": 0.07623987226554559, "全帯域_10ms_envelope_correlation": 0.9948674896755699, "全帯域_10ms_modulation_mae_db": 0.4046206893033706, "全帯域_40ms_log_rms_mae_db": 0.40518417791943456, "全帯域_40ms_rms_envelope_nrmse": 0.05112508447957477, "全帯域_40ms_envelope_correlation": 0.9976241056405833, "全帯域_40ms_modulation_mae_db": 0.5793339493106465, "300–1000 Hz_10ms_log_rms_mae_db": 1.4006211875725667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10268874369088886, "300–1000 Hz_10ms_envelope_correlation": 0.9923740034712941, "300–1000 Hz_10ms_modulation_mae_db": 0.138070726061667, "300–1000 Hz_40ms_log_rms_mae_db": 0.8792581691059871, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06800886310010627, "300–1000 Hz_40ms_envelope_correlation": 0.9965112458933713, "300–1000 Hz_40ms_modulation_mae_db": 0.15922879454773273, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3892098944909617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17089186529309186, "1000–3000 Hz_10ms_envelope_correlation": 0.9829549806751533, "1000–3000 Hz_10ms_modulation_mae_db": 1.4644052230181241, "1000–3000 Hz_40ms_log_rms_mae_db": 0.841494914869547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12093074346601553, "1000–3000 Hz_40ms_envelope_correlation": 0.9944153914840831, "1000–3000 Hz_40ms_modulation_mae_db": 1.3420105484272387, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3890233228998012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16073621763314103, "3000–6000 Hz_10ms_envelope_correlation": 0.9802260643780434, "3000–6000 Hz_10ms_modulation_mae_db": 1.1456279696614222, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9071698093312611, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0957245957879859, "3000–6000 Hz_40ms_envelope_correlation": 0.9932968864214374, "3000–6000 Hz_40ms_modulation_mae_db": 1.0402947332538661, "mr_stft_db": 5.2538527990762445, "band_mae_db": 0.3811724822566071, "rms_gain_error_db": -0.11056742400745809, "output_peak": 0.2201718034611977}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7355505387984252, "全帯域_10ms_rms_envelope_nrmse": 0.07731750378185405, "全帯域_10ms_envelope_correlation": 0.9946372287283034, "全帯域_10ms_modulation_mae_db": 0.3862837502047811, "全帯域_40ms_log_rms_mae_db": 0.3892231396245287, "全帯域_40ms_rms_envelope_nrmse": 0.037963343378269086, "全帯域_40ms_envelope_correlation": 0.9986545522373388, "全帯域_40ms_modulation_mae_db": 0.46242678438159945, "300–1000 Hz_10ms_log_rms_mae_db": 1.4806057207859633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10761816804587579, "300–1000 Hz_10ms_envelope_correlation": 0.9916076852342838, "300–1000 Hz_10ms_modulation_mae_db": 0.3564990096184763, "300–1000 Hz_40ms_log_rms_mae_db": 0.8812801038316633, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05708492382177641, "300–1000 Hz_40ms_envelope_correlation": 0.9975805349032099, "300–1000 Hz_40ms_modulation_mae_db": 0.4604117499522928, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4032154813334625, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15209901400746323, "1000–3000 Hz_10ms_envelope_correlation": 0.9830088887428565, "1000–3000 Hz_10ms_modulation_mae_db": 0.5448825916036943, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8243942569521452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09658361909326611, "1000–3000 Hz_40ms_envelope_correlation": 0.9928546991729309, "1000–3000 Hz_40ms_modulation_mae_db": 0.3586322289290367, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3952740451022891, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16252429129889442, "3000–6000 Hz_10ms_envelope_correlation": 0.9797316637819918, "3000–6000 Hz_10ms_modulation_mae_db": 1.2195542467804197, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8598469879625419, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10343595031190761, "3000–6000 Hz_40ms_envelope_correlation": 0.9918452395705266, "3000–6000 Hz_40ms_modulation_mae_db": 1.0622312706872805, "mr_stft_db": 5.260509450647283, "band_mae_db": 0.21065224444762473, "rms_gain_error_db": -0.05672079342069479, "output_peak": 0.21404477585774068}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7193418753706105, "全帯域_10ms_rms_envelope_nrmse": 0.07628412528439624, "全帯域_10ms_envelope_correlation": 0.9949587712125579, "全帯域_10ms_modulation_mae_db": 0.6010416733140329, "全帯域_40ms_log_rms_mae_db": 0.40279180501508616, "全帯域_40ms_rms_envelope_nrmse": 0.04378040701741579, "全帯域_40ms_envelope_correlation": 0.9984310473564326, "全帯域_40ms_modulation_mae_db": 0.6516522788819974, "300–1000 Hz_10ms_log_rms_mae_db": 1.3725737698420237, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10486458983905675, "300–1000 Hz_10ms_envelope_correlation": 0.9922331642700531, "300–1000 Hz_10ms_modulation_mae_db": 0.656180892701542, "300–1000 Hz_40ms_log_rms_mae_db": 0.8155718149380641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06231716509820166, "300–1000 Hz_40ms_envelope_correlation": 0.997361458844427, "300–1000 Hz_40ms_modulation_mae_db": 0.7140682773691923, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3876874712794736, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1614600297250283, "1000–3000 Hz_10ms_envelope_correlation": 0.9818194736462141, "1000–3000 Hz_10ms_modulation_mae_db": 0.7349755435484132, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8611884485067969, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09876099206641681, "1000–3000 Hz_40ms_envelope_correlation": 0.9940720354370283, "1000–3000 Hz_40ms_modulation_mae_db": 0.5386546329736069, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4339283401844705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1704497374195604, "3000–6000 Hz_10ms_envelope_correlation": 0.9779032355757954, "3000–6000 Hz_10ms_modulation_mae_db": 1.1829637056565658, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9090234266511876, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09903576571276843, "3000–6000 Hz_40ms_envelope_correlation": 0.9932046870261679, "3000–6000 Hz_40ms_modulation_mae_db": 1.3085806308631887, "mr_stft_db": 5.262232495827674, "band_mae_db": 0.2224901031411468, "rms_gain_error_db": -0.1532947936776181, "output_peak": 0.2356215414663413}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.738083957273852, "全帯域_10ms_rms_envelope_nrmse": 0.0743429966073349, "全帯域_10ms_envelope_correlation": 0.9951028983006298, "全帯域_10ms_modulation_mae_db": 0.40441495197479643, "全帯域_40ms_log_rms_mae_db": 0.4063930494656705, "全帯域_40ms_rms_envelope_nrmse": 0.03883635391632217, "全帯域_40ms_envelope_correlation": 0.9986753699298648, "全帯域_40ms_modulation_mae_db": 0.3693942286984373, "300–1000 Hz_10ms_log_rms_mae_db": 1.4173376460653793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10123178238650278, "300–1000 Hz_10ms_envelope_correlation": 0.9926418626439928, "300–1000 Hz_10ms_modulation_mae_db": 0.6160302524333708, "300–1000 Hz_40ms_log_rms_mae_db": 0.8283448339956901, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05465172374992496, "300–1000 Hz_40ms_envelope_correlation": 0.9978732658033314, "300–1000 Hz_40ms_modulation_mae_db": 0.5999590383159837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.396858873677, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1457990211979368, "1000–3000 Hz_10ms_envelope_correlation": 0.9856449783725703, "1000–3000 Hz_10ms_modulation_mae_db": 0.9353600009184229, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8485803751035704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09271441393947207, "1000–3000 Hz_40ms_envelope_correlation": 0.9951260367065414, "1000–3000 Hz_40ms_modulation_mae_db": 1.05747963683791, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4720300863163773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1659651352502199, "3000–6000 Hz_10ms_envelope_correlation": 0.9783511881341906, "3000–6000 Hz_10ms_modulation_mae_db": 0.6177839587417132, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9584179970397751, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10419436131695228, "3000–6000 Hz_40ms_envelope_correlation": 0.9910443552453034, "3000–6000 Hz_40ms_modulation_mae_db": 0.45701603695623, "mr_stft_db": 5.254990434563548, "band_mae_db": 0.20889097817919489, "rms_gain_error_db": -0.10366560660479752, "output_peak": 0.22829429375682325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.764110135636261, "全帯域_10ms_rms_envelope_nrmse": 0.07690011856527963, "全帯域_10ms_envelope_correlation": 0.9947114562514503, "全帯域_10ms_modulation_mae_db": 0.3362750418004096, "全帯域_40ms_log_rms_mae_db": 0.4072985464004471, "全帯域_40ms_rms_envelope_nrmse": 0.03911396456720802, "全帯域_40ms_envelope_correlation": 0.9986011724287723, "全帯域_40ms_modulation_mae_db": 0.31078092810778696, "300–1000 Hz_10ms_log_rms_mae_db": 1.451905411885693, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10811823341241936, "300–1000 Hz_10ms_envelope_correlation": 0.9915260161820442, "300–1000 Hz_10ms_modulation_mae_db": 0.45877244505419545, "300–1000 Hz_40ms_log_rms_mae_db": 0.8266245711890506, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057082635573893155, "300–1000 Hz_40ms_envelope_correlation": 0.997578247347171, "300–1000 Hz_40ms_modulation_mae_db": 0.7262633003066808, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3896358833811615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1473136190187829, "1000–3000 Hz_10ms_envelope_correlation": 0.9854394267648515, "1000–3000 Hz_10ms_modulation_mae_db": 0.8627826561265941, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8428081200129398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09874148713397338, "1000–3000 Hz_40ms_envelope_correlation": 0.9943413298448456, "1000–3000 Hz_40ms_modulation_mae_db": 0.6498923732368334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4642640475721804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1747474259502968, "3000–6000 Hz_10ms_envelope_correlation": 0.9765413509195466, "3000–6000 Hz_10ms_modulation_mae_db": 0.5227862101252728, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9297009290188284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1049228600857728, "3000–6000 Hz_40ms_envelope_correlation": 0.9919162562804552, "3000–6000 Hz_40ms_modulation_mae_db": 0.3988960203850145, "mr_stft_db": 5.265021952736148, "band_mae_db": 0.24462226307881688, "rms_gain_error_db": -0.07790921866765427, "output_peak": 0.21941804459810227}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.224027774674207, "spectral_convergence": 0.24535874809426841, "spectral_convergence_status": "ok", "amplitude_floor": 6.716189184049609e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.033193783778544, "spectral_convergence": 0.23292811403084568, "spectral_convergence_status": "ok", "amplitude_floor": 4.979433430044932e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.541377746017884, "spectral_convergence": 0.24351014841763677, "spectral_convergence_status": "ok", "amplitude_floor": 3.630042469105695e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.462171010234058e-05, "output_power": 7.689560081216991e-05, "delta_db": 0.13036297052654142, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003077822779549395, "output_power": 0.00029478795232472344, "delta_db": -0.18733879233177742, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 5.338393512143839e-05, "output_power": 5.4327266838050316e-05, "delta_db": 0.07607272534139864, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.1515857114073014e-05, "output_power": 2.092811708832279e-05, "delta_db": -0.12028495052121702, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.940885952831811e-05, "output_power": 2.6398150892440246e-05, "delta_db": -0.46904675980284916, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.017666424897867e-06, "output_power": 7.95346774909539e-06, "delta_db": -0.034914590923322476, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16967013157451769, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11961175577444655, "pitch": null, "output_rms_dbfs": -33.17592660337044, "output_peak": 0.2025860216287537, "reference_rms_dbfs": -33.056314847596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.266199768156878}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.650394450462669e-05, "log_rms_mae_db": 0.716795600357177, "rms_envelope_nrmse": 0.0783424793381591, "envelope_correlation": 0.9945915643625846, "rms_gain_error_db": -0.11957214484852678, "reference_mean_power": 0.0004951134756771775, "output_mean_power": 0.00048166770917925197}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.645854738282274e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010486873457313227, "output_power": 9.93036815163207e-05, "reference_active": true, "delta_db": -0.23680677976004486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5071002541192514e-05, "output_power": 2.354436965647259e-05, "reference_active": true, "delta_db": -0.27284633156362986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059292105678353e-06, "output_power": 4.445202861639943e-06, "reference_active": true, "delta_db": -0.561981695069196}], "active_band_count": 3, "active_band_mae_db": 0.3572116021309569}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.020301765617229e-05, "log_rms_mae_db": 0.39201637603451545, "rms_envelope_nrmse": 0.043936500977798595, "envelope_correlation": 0.9983275175953141, "rms_gain_error_db": -0.12005765210602118, "reference_mean_power": 0.0004970274687625529, "output_mean_power": 0.0004834756722516705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.472129124053938e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.182773781582673e-05, "output_power": 8.674394530668122e-05, "reference_active": true, "delta_db": -0.24734714630573498}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2533326164459841e-05, "output_power": 1.1564920448328164e-05, "reference_active": true, "delta_db": -0.34923692064728373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4723702388808576e-07, "output_power": 2.486738211084528e-07, "reference_active": true, "delta_db": 0.02516560614537899}], "active_band_count": 3, "active_band_mae_db": 0.20724989103279923}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.432622539351727e-05, "log_rms_mae_db": 1.4599146080922876, "rms_envelope_nrmse": 0.10949365807495758, "envelope_correlation": 0.9914246590598647, "rms_gain_error_db": -0.17472027547714503, "reference_mean_power": 0.00026104952136561866, "output_mean_power": 0.00025075573455260935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.392840173310503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.633554932680893e-05, "output_power": 9.235929319565684e-05, "reference_active": true, "delta_db": -0.18305977465928192}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.160826072139944e-05, "output_power": 1.9747188868379868e-05, "reference_active": true, "delta_db": -0.3911453141064824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.769645916048805e-06, "output_power": 3.548699089293235e-06, "reference_active": true, "delta_db": -1.2841696448126858}], "active_band_count": 3, "active_band_mae_db": 0.6194582445261501}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.77451695930137e-05, "log_rms_mae_db": 0.9065996378784458, "rms_envelope_nrmse": 0.06606749320577592, "envelope_correlation": 0.9968960222108503, "rms_gain_error_db": -0.17502613095687816, "reference_mean_power": 0.00026200159904973233, "output_mean_power": 0.0002516525461473817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2678124187221727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604417147270571e-05, "output_power": 8.221575914040298e-05, "reference_active": true, "delta_db": -0.19766385870306283}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1040946271380912e-05, "output_power": 9.964045947529757e-06, "reference_active": true, "delta_db": -0.44570574930025647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9369342076961927e-07, "output_power": 2.3766345560552994e-07, "reference_active": true, "delta_db": -0.9193181026230123}], "active_band_count": 3, "active_band_mae_db": 0.5208959035421105}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.588832427367551e-05, "log_rms_mae_db": 1.3813521555140313, "rms_envelope_nrmse": 0.15478599267433826, "envelope_correlation": 0.9828032146093916, "rms_gain_error_db": -0.32859624485976047, "reference_mean_power": 4.0575597065913316e-05, "output_mean_power": 3.761883090273245e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.7364679678483124e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6458578452566544e-05, "output_power": 1.5432781488747332e-05, "reference_active": true, "delta_db": -0.27948114856827494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.144001855335748e-06, "output_power": 2.3662904224284275e-06, "reference_active": true, "delta_db": -1.23414747732261}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.079905260013679e-07, "output_power": 8.567528868395023e-07, "reference_active": true, "delta_db": -0.2522574063888709}], "active_band_count": 3, "active_band_mae_db": 0.5886286774265853}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.068450322528164e-05, "log_rms_mae_db": 0.8351202714686572, "rms_envelope_nrmse": 0.09560627300863052, "envelope_correlation": 0.9937230409350668, "rms_gain_error_db": -0.330850931374026, "reference_mean_power": 4.071142209359612e-05, "output_mean_power": 3.772516777416494e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.516451938680711e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5218014411010514e-05, "output_power": 1.4295663370207001e-05, "reference_active": true, "delta_db": -0.27153678040907064}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2444589918408885e-06, "output_power": 9.683238049941294e-07, "reference_active": true, "delta_db": -1.0895998149292205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2414094224167626e-08, "output_power": 3.844572451684423e-08, "reference_active": true, "delta_db": -1.3460003541176435}], "active_band_count": 3, "active_band_mae_db": 0.9023789831519782}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.7354442023779843e-05, "log_rms_mae_db": 1.4251255287708633, "rms_envelope_nrmse": 0.1508177562700964, "envelope_correlation": 0.9820853996233702, "rms_gain_error_db": -0.19613989101154566, "reference_mean_power": 9.786447369332928e-06, "output_mean_power": 9.354295299583713e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.593718108385407e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.378843387910109e-06, "output_power": 2.2690311834558807e-06, "reference_active": true, "delta_db": -0.20525386564708675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.926849454092717e-07, "output_power": 7.511864953616611e-07, "reference_active": true, "delta_db": -0.7495043902900158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6983089704559518e-07, "output_power": 1.4481058987493305e-07, "reference_active": true, "delta_db": -0.6921638094316604}], "active_band_count": 3, "active_band_mae_db": 0.5489740217895877}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.2351945663576383e-05, "log_rms_mae_db": 0.8829924353260394, "rms_envelope_nrmse": 0.08581412998629015, "envelope_correlation": 0.993895556371717, "rms_gain_error_db": -0.19794633705873277, "reference_mean_power": 9.83170860207206e-06, "output_mean_power": 9.393649779618186e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.2206731147886266e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.414723334288e-06, "output_power": 2.2566968223959075e-06, "reference_active": true, "delta_db": -0.29394161482761216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.701511911331394e-07, "output_power": 3.3685019174486855e-07, "reference_active": true, "delta_db": -0.4094235236265397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1816168225696604e-08, "output_power": 9.629674946823482e-09, "reference_active": true, "delta_db": -0.8886503775444234}], "active_band_count": 3, "active_band_mae_db": 0.5306718386661918}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9046216565482368, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f24c45a4213262254da2d38ed967a658f622f2831942c653426ee188cfb01da2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9828275220301661, "standard_deviation": 0.0020912279991011913, "minimum": 0.9791118480644512, "maximum": 0.9856449783725703, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3945115739135554, "standard_deviation": 0.012006452749158774, "minimum": 1.3813521555140313, "maximum": 1.4194548480456253, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8624175156460374, "standard_deviation": 0.3013869310689905, "minimum": 0.5448825916036943, "maximum": 1.4644052230181241, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15856683734716903, "standard_deviation": 0.010174792260507295, "minimum": 0.1457990211979368, "maximum": 0.17240238372061067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994010197728314, "standard_deviation": 0.0007638317804402908, "minimum": 0.9928546991729309, "maximum": 0.9951260367065414, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.846168417881179, "standard_deviation": 0.01725679666557013, "minimum": 0.8243942569521452, "maximum": 0.879669314628699, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8163357994510103, "standard_deviation": 0.30701330999900667, "minimum": 0.3586322289290367, "maximum": 1.3420105484272387, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10127511086556502, "standard_deviation": 0.008852035357717393, "minimum": 0.09271441393947207, "maximum": 0.12093074346601553, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792654102468946, "standard_deviation": 0.0016948540200608729, "minimum": 0.9765413509195466, "maximum": 0.9820853996233702, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4200622404940753, "standard_deviation": 0.034260383111501404, "minimum": 1.3882348334388979, "maximum": 1.4720300863163773, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8866635732669063, "standard_deviation": 0.31804502121170514, "minimum": 0.5227862101252728, "maximum": 1.2195542467804197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16348510199497324, "standard_deviation": 0.0073625083701512615, "minimum": 0.1508177562700964, "maximum": 0.1747474259502968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9925010763595099, "standard_deviation": 0.0009895285786460414, "minimum": 0.9910443552453034, "maximum": 0.993895556371717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8976149922553087, "standard_deviation": 0.04201498004115541, "minimum": 0.8228396375891247, "maximum": 0.9584179970397751, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8677585324736703, "standard_deviation": 0.36525319971655273, "minimum": 0.3988960203850145, "maximum": 1.3085806308631887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09833783703331235, "standard_deviation": 0.007115036859199668, "minimum": 0.08581412998629015, "maximum": 0.1049228600857728, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920762054608923, "standard_deviation": 0.0005355849902925833, "minimum": 0.9914246590598647, "maximum": 0.9928576304068968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4245767487750272, "standard_deviation": 0.03742136270010678, "minimum": 1.3725737698420237, "maximum": 1.4806057207859633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4740568708893948, "standard_deviation": 0.17418663820664024, "minimum": 0.138070726061667, "maximum": 0.656180892701542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10509741168413375, "standard_deviation": 0.0032204982887433474, "minimum": 0.10102506317431788, "maximum": 0.10949365807495758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973352185511238, "standard_deviation": 0.0004315303502963278, "minimum": 0.9965112458933713, "maximum": 0.9978732658033314, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8566542352713536, "standard_deviation": 0.03933930194053723, "minimum": 0.8140300374280487, "maximum": 0.9065996378784458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5351006335212704, "standard_deviation": 0.18307409611339187, "minimum": 0.15922879454773273, "maximum": 0.7262633003066808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06082012420017712, "standard_deviation": 0.004652013996114272, "minimum": 0.05465172374992496, "maximum": 0.06800886310010627, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2498838287196201, "standard_deviation": 0.06720257994675337, "minimum": 0.16967013157451769, "maximum": 0.3811724822566071, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.2598022428708555, "standard_deviation": 0.004645286300778172, "minimum": 5.2538527990762445, "maximum": 5.266199768156878, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.22067765113133472, "standard_deviation": 0.009981141245361449, "minimum": 0.2025860216287537, "maximum": 0.2356215414663413, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10591615899631754, "standard_deviation": 0.03014810856051451, "minimum": -0.1532947936776181, "maximum": -0.05672079342069479, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949533054328044, "standard_deviation": 0.0003247279030412196, "minimum": 0.9945915643625846, "maximum": 0.9955480795114405, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7335787968361176, "standard_deviation": 0.01653875254142583, "minimum": 0.716795600357177, "maximum": 0.764110135636261, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4107696125031436, "standard_deviation": 0.0808727754982611, "minimum": 0.3362750418004096, "maximum": 0.6010416733140329, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07554179358939497, "standard_deviation": 0.0022821041963845055, "minimum": 0.07143752463716789, "maximum": 0.0783424793381591, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984151177061916, "standard_deviation": 0.0003399703506601228, "minimum": 0.9976241056405833, "maximum": 0.9986753699298648, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.40435667550533805, "standard_deviation": 0.010633526235905704, "minimum": 0.3892231396245287, "maximum": 0.4238590940825517, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.41107793607745885, "standard_deviation": 0.14644242115669606, "minimum": 0.20724989103279923, "maximum": 0.6516522788819974, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04220059227977612, "standard_deviation": 0.004243361802886239, "minimum": 0.037963343378269086, "maximum": 0.05112508447957477, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0030", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.2204713821411133, "predicted_mos": 3.2204713821411133, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a43f69d46584aa27434ae0914edfd8b1aeba57f1b9539947ce49cd1649efad0a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9624169003218412, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7445658418615377, "全帯域_10ms_rms_envelope_nrmse": 0.07036898277458258, "全帯域_10ms_envelope_correlation": 0.9956306128228241, "全帯域_10ms_modulation_mae_db": 0.10692206366880865, "全帯域_40ms_log_rms_mae_db": 0.42282471814415096, "全帯域_40ms_rms_envelope_nrmse": 0.03667787310219672, "全帯域_40ms_envelope_correlation": 0.9988384324735743, "全帯域_40ms_modulation_mae_db": 0.10968498440645462, "300–1000 Hz_10ms_log_rms_mae_db": 1.4575487460671745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09514465145105656, "300–1000 Hz_10ms_envelope_correlation": 0.9934688260044046, "300–1000 Hz_10ms_modulation_mae_db": 0.2023548612750918, "300–1000 Hz_40ms_log_rms_mae_db": 0.84533684471152, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052463095055989416, "300–1000 Hz_40ms_envelope_correlation": 0.9979796402264043, "300–1000 Hz_40ms_modulation_mae_db": 0.2898805145546259, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5298183700899999, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17244780278692243, "1000–3000 Hz_10ms_envelope_correlation": 0.9786425383347113, "1000–3000 Hz_10ms_modulation_mae_db": 0.8986266680013988, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8915599378075334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10693544885849729, "1000–3000 Hz_40ms_envelope_correlation": 0.9922464744022931, "1000–3000 Hz_40ms_modulation_mae_db": 0.6772865676091692, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6216910312944908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17736454487514267, "3000–6000 Hz_10ms_envelope_correlation": 0.9752559862867722, "3000–6000 Hz_10ms_modulation_mae_db": 0.8565878740683841, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9599711968742617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09535500652629425, "3000–6000 Hz_40ms_envelope_correlation": 0.9928241083132254, "3000–6000 Hz_40ms_modulation_mae_db": 0.8897344211706639, "mr_stft_db": 5.3661778584957815, "band_mae_db": 0.19926089124521726, "rms_gain_error_db": -0.10600006406840527, "output_peak": 0.2048143821334081}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7247831894889405, "全帯域_10ms_rms_envelope_nrmse": 0.0786931077225992, "全帯域_10ms_envelope_correlation": 0.9946223219951839, "全帯域_10ms_modulation_mae_db": 0.47702471743225017, "全帯域_40ms_log_rms_mae_db": 0.38682877561305073, "全帯域_40ms_rms_envelope_nrmse": 0.03951876511806552, "全帯域_40ms_envelope_correlation": 0.9987823967875452, "全帯域_40ms_modulation_mae_db": 0.3649147552570338, "300–1000 Hz_10ms_log_rms_mae_db": 1.569300507878896, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12225174364537379, "300–1000 Hz_10ms_envelope_correlation": 0.989576350563214, "300–1000 Hz_10ms_modulation_mae_db": 0.6027777694829984, "300–1000 Hz_40ms_log_rms_mae_db": 0.9641401842196251, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07064092753863899, "300–1000 Hz_40ms_envelope_correlation": 0.9968624477505428, "300–1000 Hz_40ms_modulation_mae_db": 0.7583656209110087, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4756070857945447, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15077369837454685, "1000–3000 Hz_10ms_envelope_correlation": 0.9846368968990176, "1000–3000 Hz_10ms_modulation_mae_db": 0.9164202361357517, "1000–3000 Hz_40ms_log_rms_mae_db": 0.870877635298059, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09403824991965595, "1000–3000 Hz_40ms_envelope_correlation": 0.9951002092076343, "1000–3000 Hz_40ms_modulation_mae_db": 1.2061435541689816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5663986516413455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.179707499127006, "3000–6000 Hz_10ms_envelope_correlation": 0.9743325033337641, "3000–6000 Hz_10ms_modulation_mae_db": 0.6211018936081759, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9195997787692549, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10080000921997552, "3000–6000 Hz_40ms_envelope_correlation": 0.9913007935350832, "3000–6000 Hz_40ms_modulation_mae_db": 0.6698573461901445, "mr_stft_db": 5.356603837156285, "band_mae_db": 0.24157891673281098, "rms_gain_error_db": -0.15475253464899047, "output_peak": 0.2124214904237969}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7060726908311616, "全帯域_10ms_rms_envelope_nrmse": 0.07471782973987659, "全帯域_10ms_envelope_correlation": 0.9950408137051526, "全帯域_10ms_modulation_mae_db": 0.506496469438055, "全帯域_40ms_log_rms_mae_db": 0.41107254666072107, "全帯域_40ms_rms_envelope_nrmse": 0.04307944307170986, "全帯域_40ms_envelope_correlation": 0.9983247338525967, "全帯域_40ms_modulation_mae_db": 0.46465525433166915, "300–1000 Hz_10ms_log_rms_mae_db": 1.4880593864598506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10878436202931878, "300–1000 Hz_10ms_envelope_correlation": 0.9915211501279511, "300–1000 Hz_10ms_modulation_mae_db": 0.744346438914054, "300–1000 Hz_40ms_log_rms_mae_db": 0.9235048061465805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06365014990529885, "300–1000 Hz_40ms_envelope_correlation": 0.9971134338243179, "300–1000 Hz_40ms_modulation_mae_db": 0.7183118678153683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5158515842821694, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1773998116575086, "1000–3000 Hz_10ms_envelope_correlation": 0.9774003079565893, "1000–3000 Hz_10ms_modulation_mae_db": 0.9581188859623958, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9404282057966864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10929497706546916, "1000–3000 Hz_40ms_envelope_correlation": 0.991921083094024, "1000–3000 Hz_40ms_modulation_mae_db": 1.1239724211854656, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5400926261169234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16720345178263857, "3000–6000 Hz_10ms_envelope_correlation": 0.9786140865619525, "3000–6000 Hz_10ms_modulation_mae_db": 1.0795050245876003, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8954402944094824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09668270682243658, "3000–6000 Hz_40ms_envelope_correlation": 0.9932728469530099, "3000–6000 Hz_40ms_modulation_mae_db": 0.9121215949046692, "mr_stft_db": 5.365149346702972, "band_mae_db": 0.23529829987810236, "rms_gain_error_db": -0.09640270907252683, "output_peak": 0.2057331405054409}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7209546101745447, "全帯域_10ms_rms_envelope_nrmse": 0.07334757115242738, "全帯域_10ms_envelope_correlation": 0.9953049517672284, "全帯域_10ms_modulation_mae_db": 0.552408438289205, "全帯域_40ms_log_rms_mae_db": 0.3992562312143555, "全帯域_40ms_rms_envelope_nrmse": 0.03957043848570807, "全帯域_40ms_envelope_correlation": 0.9987046416008735, "全帯域_40ms_modulation_mae_db": 0.45080167891854517, "300–1000 Hz_10ms_log_rms_mae_db": 1.5178756324212286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09854943298596167, "300–1000 Hz_10ms_envelope_correlation": 0.9931638278689318, "300–1000 Hz_10ms_modulation_mae_db": 0.7237923894748874, "300–1000 Hz_40ms_log_rms_mae_db": 0.8925327109864459, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0537888029443817, "300–1000 Hz_40ms_envelope_correlation": 0.9981075958966689, "300–1000 Hz_40ms_modulation_mae_db": 0.8368186781228696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5242405324773076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18443278352713907, "1000–3000 Hz_10ms_envelope_correlation": 0.9768173928131773, "1000–3000 Hz_10ms_modulation_mae_db": 0.7775276792867691, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8913338234282793, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11121834710319435, "1000–3000 Hz_40ms_envelope_correlation": 0.9935503818895498, "1000–3000 Hz_40ms_modulation_mae_db": 0.7868099475627299, "3000–6000 Hz_10ms_log_rms_mae_db": 1.525097774810762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.169022624314103, "3000–6000 Hz_10ms_envelope_correlation": 0.977408935326948, "3000–6000 Hz_10ms_modulation_mae_db": 1.2579549566489345, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8846710683487083, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09746345141898737, "3000–6000 Hz_40ms_envelope_correlation": 0.9919879755561015, "3000–6000 Hz_40ms_modulation_mae_db": 1.2875538774585653, "mr_stft_db": 5.379029812086053, "band_mae_db": 0.3217116810006939, "rms_gain_error_db": -0.13393371851611346, "output_peak": 0.23590545529297874}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7273683568622367, "全帯域_10ms_rms_envelope_nrmse": 0.07899395136059795, "全帯域_10ms_envelope_correlation": 0.994599618712125, "全帯域_10ms_modulation_mae_db": 0.24763894905339057, "全帯域_40ms_log_rms_mae_db": 0.39986703958684877, "全帯域_40ms_rms_envelope_nrmse": 0.03879888965350399, "全帯域_40ms_envelope_correlation": 0.9988640284007434, "全帯域_40ms_modulation_mae_db": 0.43872509547820426, "300–1000 Hz_10ms_log_rms_mae_db": 1.5588416040884174, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12103748044629169, "300–1000 Hz_10ms_envelope_correlation": 0.9896348656282192, "300–1000 Hz_10ms_modulation_mae_db": 0.3654077181826982, "300–1000 Hz_40ms_log_rms_mae_db": 0.9384130868665249, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06129162996403789, "300–1000 Hz_40ms_envelope_correlation": 0.9976340737804321, "300–1000 Hz_40ms_modulation_mae_db": 0.6915031933367187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.474687008737427, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14506237666207425, "1000–3000 Hz_10ms_envelope_correlation": 0.9848832624044102, "1000–3000 Hz_10ms_modulation_mae_db": 0.6833191774896833, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8813084638985401, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07942062550225158, "1000–3000 Hz_40ms_envelope_correlation": 0.9958028932527062, "1000–3000 Hz_40ms_modulation_mae_db": 0.42371984842407423, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6027460142149772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16103731083890235, "3000–6000 Hz_10ms_envelope_correlation": 0.9794487076167381, "3000–6000 Hz_10ms_modulation_mae_db": 0.5117815531538669, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9202756349787838, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08621749143812452, "3000–6000 Hz_40ms_envelope_correlation": 0.9936682987214686, "3000–6000 Hz_40ms_modulation_mae_db": 0.3847361091254155, "mr_stft_db": 5.360146978355091, "band_mae_db": 0.22645078599340218, "rms_gain_error_db": -0.16122898826382526, "output_peak": 0.21891581825414608}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6727121094454726, "全帯域_10ms_rms_envelope_nrmse": 0.07091478282215279, "全帯域_10ms_envelope_correlation": 0.9956190868890915, "全帯域_10ms_modulation_mae_db": 0.3095563003268011, "全帯域_40ms_log_rms_mae_db": 0.35968512025777305, "全帯域_40ms_rms_envelope_nrmse": 0.039213884420649665, "全帯域_40ms_envelope_correlation": 0.9987272729720545, "全帯域_40ms_modulation_mae_db": 0.3572460614654847, "300–1000 Hz_10ms_log_rms_mae_db": 1.4457936706710015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09952100374456252, "300–1000 Hz_10ms_envelope_correlation": 0.9932222099688031, "300–1000 Hz_10ms_modulation_mae_db": 0.34770138995085964, "300–1000 Hz_40ms_log_rms_mae_db": 0.8247969560581764, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05787398554405975, "300–1000 Hz_40ms_envelope_correlation": 0.9980014304989413, "300–1000 Hz_40ms_modulation_mae_db": 0.4780264236411804, "1000–3000 Hz_10ms_log_rms_mae_db": 1.536050061923955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15398912197344647, "1000–3000 Hz_10ms_envelope_correlation": 0.9827695569445583, "1000–3000 Hz_10ms_modulation_mae_db": 0.5168248978655405, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8789052996015535, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09093875843186171, "1000–3000 Hz_40ms_envelope_correlation": 0.9940857038987534, "1000–3000 Hz_40ms_modulation_mae_db": 0.3154484733540834, "3000–6000 Hz_10ms_log_rms_mae_db": 1.57819476383108, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1691385725679206, "3000–6000 Hz_10ms_envelope_correlation": 0.9777655130134681, "3000–6000 Hz_10ms_modulation_mae_db": 0.7588446125609766, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9605817613065816, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10148822685862505, "3000–6000 Hz_40ms_envelope_correlation": 0.9919551856234986, "3000–6000 Hz_40ms_modulation_mae_db": 0.7303549156275172, "mr_stft_db": 5.387312380599643, "band_mae_db": 0.2069797245198187, "rms_gain_error_db": -0.12370036424162237, "output_peak": 0.23121116334269606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7160615565978768, "全帯域_10ms_rms_envelope_nrmse": 0.07545014968687891, "全帯域_10ms_envelope_correlation": 0.9950150441114562, "全帯域_10ms_modulation_mae_db": 0.13056218877366893, "全帯域_40ms_log_rms_mae_db": 0.3702454679189533, "全帯域_40ms_rms_envelope_nrmse": 0.03856058053840895, "全帯域_40ms_envelope_correlation": 0.998778661837203, "全帯域_40ms_modulation_mae_db": 0.20021939685080328, "300–1000 Hz_10ms_log_rms_mae_db": 1.4975389831707828, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10658573237296315, "300–1000 Hz_10ms_envelope_correlation": 0.9919705805180578, "300–1000 Hz_10ms_modulation_mae_db": 0.3216179798228743, "300–1000 Hz_40ms_log_rms_mae_db": 0.8565386689705604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058342709567828256, "300–1000 Hz_40ms_envelope_correlation": 0.9977536626225154, "300–1000 Hz_40ms_modulation_mae_db": 0.5491804512057133, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5099018737992485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16724168527459757, "1000–3000 Hz_10ms_envelope_correlation": 0.9800929632313332, "1000–3000 Hz_10ms_modulation_mae_db": 0.4550925776085875, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8718104834364572, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10824592332324127, "1000–3000 Hz_40ms_envelope_correlation": 0.9921160018323805, "1000–3000 Hz_40ms_modulation_mae_db": 0.8681498080689005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.614413124097001, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17822646572159923, "3000–6000 Hz_10ms_envelope_correlation": 0.9748013396319051, "3000–6000 Hz_10ms_modulation_mae_db": 1.3741953760591041, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0055289841738468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10357322411104988, "3000–6000 Hz_40ms_envelope_correlation": 0.9909305318515865, "3000–6000 Hz_40ms_modulation_mae_db": 1.6575131524932694, "mr_stft_db": 5.3671760075976325, "band_mae_db": 0.22272619046189615, "rms_gain_error_db": -0.13052080065438668, "output_peak": 0.23092841490269486}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6955544705937021, "全帯域_10ms_rms_envelope_nrmse": 0.07454467423814264, "全帯域_10ms_envelope_correlation": 0.9950370546767768, "全帯域_10ms_modulation_mae_db": 0.4111297574061296, "全帯域_40ms_log_rms_mae_db": 0.373032575832467, "全帯域_40ms_rms_envelope_nrmse": 0.036707869346126326, "全帯域_40ms_envelope_correlation": 0.9987688433137195, "全帯域_40ms_modulation_mae_db": 0.1411985523708652, "300–1000 Hz_10ms_log_rms_mae_db": 1.485822511828572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11041317298813282, "300–1000 Hz_10ms_envelope_correlation": 0.9911559626366331, "300–1000 Hz_10ms_modulation_mae_db": 0.16672885267450147, "300–1000 Hz_40ms_log_rms_mae_db": 0.8834861278443426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05838452785251283, "300–1000 Hz_40ms_envelope_correlation": 0.9974410505181567, "300–1000 Hz_40ms_modulation_mae_db": 0.2246086495910067, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5208762861223615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1746423517843591, "1000–3000 Hz_10ms_envelope_correlation": 0.9780081181843008, "1000–3000 Hz_10ms_modulation_mae_db": 0.97389965843201, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8775031140569505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10058315485595155, "1000–3000 Hz_40ms_envelope_correlation": 0.9931957408825336, "1000–3000 Hz_40ms_modulation_mae_db": 1.0589031187612816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5582973895454044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1637556014109906, "3000–6000 Hz_10ms_envelope_correlation": 0.9792942860942421, "3000–6000 Hz_10ms_modulation_mae_db": 1.2295912043847632, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8965341430943427, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09104729125261275, "3000–6000 Hz_40ms_envelope_correlation": 0.9938682371629178, "3000–6000 Hz_40ms_modulation_mae_db": 1.3043868042794846, "mr_stft_db": 5.376601992008673, "band_mae_db": 0.23103531346457798, "rms_gain_error_db": -0.0708501118064436, "output_peak": 0.21799281253674307}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.392399039975449, "spectral_convergence": 0.23823481838258642, "spectral_convergence_status": "ok", "amplitude_floor": 6.716189184049609e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.154069469730455, "spectral_convergence": 0.22995778652153784, "spectral_convergence_status": "ok", "amplitude_floor": 4.979433430044932e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.55206506578144, "spectral_convergence": 0.24470003766568962, "spectral_convergence_status": "ok", "amplitude_floor": 3.630042469105695e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.462171010234058e-05, "output_power": 7.748956021680024e-05, "delta_db": 0.16377998585189735, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003077822779549395, "output_power": 0.00029835716516061876, "delta_db": -0.1350713756939683, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 5.338393512143839e-05, "output_power": 5.2561712692172734e-05, "delta_db": -0.06741076026392677, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.1515857114073014e-05, "output_power": 2.0451161379590116e-05, "delta_db": -0.22040675749232752, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.940885952831811e-05, "output_power": 2.6084982304491982e-05, "delta_db": -0.5208763650889635, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.017666424897867e-06, "output_power": 7.856804977201332e-06, "delta_db": -0.08802010308022007, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19926089124521726, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10600006406840527, "pitch": null, "output_rms_dbfs": -33.1623149116644, "output_peak": 0.2048143821334081, "reference_rms_dbfs": -33.056314847596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.3661778584957815}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.650394450462669e-05, "log_rms_mae_db": 0.7445658418615377, "rms_envelope_nrmse": 0.07036898277458258, "envelope_correlation": 0.9956306128228241, "rms_gain_error_db": -0.1059552420323927, "reference_mean_power": 0.0004951134756771775, "output_mean_power": 0.0004831803039059555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.645854738282274e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010486873457313227, "output_power": 0.00010056963997700609, "reference_active": true, "delta_db": -0.18179132077135726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5071002541192514e-05, "output_power": 2.585526077790792e-05, "reference_active": true, "delta_db": 0.1337722154682733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059292105678353e-06, "output_power": 5.053234927556805e-06, "reference_active": true, "delta_db": -0.005202654766795383}], "active_band_count": 3, "active_band_mae_db": 0.10692206366880865}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.020301765617229e-05, "log_rms_mae_db": 0.42282471814415096, "rms_envelope_nrmse": 0.03667787310219672, "envelope_correlation": 0.9988384324735743, "rms_gain_error_db": -0.10574779420791872, "reference_mean_power": 0.0004970274687625529, "output_mean_power": 0.0004850713358038275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.472129124053938e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.182773781582673e-05, "output_power": 8.86977806611716e-05, "reference_active": true, "delta_db": -0.15061132236649932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2533326164459841e-05, "output_power": 1.2959333226806307e-05, "reference_active": true, "delta_db": 0.14516315340711097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4723702388808576e-07, "output_power": 2.4913890703673235e-07, "reference_active": true, "delta_db": 0.03328047744575355}], "active_band_count": 3, "active_band_mae_db": 0.10968498440645462}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.432622539351727e-05, "log_rms_mae_db": 1.4575487460671745, "rms_envelope_nrmse": 0.09514465145105656, "envelope_correlation": 0.9934688260044046, "rms_gain_error_db": -0.10748147993537059, "reference_mean_power": 0.00026104952136561866, "output_mean_power": 0.00025466822026853735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.392840173310503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.633554932680893e-05, "output_power": 9.217604979745531e-05, "reference_active": true, "delta_db": -0.19168485661381812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.160826072139944e-05, "output_power": 2.231360871330358e-05, "reference_active": true, "delta_db": 0.13950001783370047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.769645916048805e-06, "output_power": 4.476083387008209e-06, "reference_active": true, "delta_db": -0.2758797093777568}], "active_band_count": 3, "active_band_mae_db": 0.2023548612750918}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.77451695930137e-05, "log_rms_mae_db": 0.84533684471152, "rms_envelope_nrmse": 0.052463095055989416, "envelope_correlation": 0.9979796402264043, "rms_gain_error_db": -0.10686933377601682, "reference_mean_power": 0.00026200159904973233, "output_mean_power": 0.0002556330540279781}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2678124187221727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604417147270571e-05, "output_power": 8.307363743519055e-05, "reference_active": true, "delta_db": -0.15258230323010463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1040946271380912e-05, "output_power": 1.1359051054214385e-05, "reference_active": true, "delta_db": 0.1233575504639094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9369342076961927e-07, "output_power": 2.561675110828289e-07, "reference_active": true, "delta_db": -0.5937016899698635}], "active_band_count": 3, "active_band_mae_db": 0.2898805145546259}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.588832427367551e-05, "log_rms_mae_db": 1.5298183700899999, "rms_envelope_nrmse": 0.17244780278692243, "envelope_correlation": 0.9786425383347113, "rms_gain_error_db": -0.3905205113500142, "reference_mean_power": 4.0575597065913316e-05, "output_mean_power": 3.708624542523901e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.7364679678483124e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6458578452566544e-05, "output_power": 1.4807993598492901e-05, "reference_active": true, "delta_db": -0.4589610395455147}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.144001855335748e-06, "output_power": 2.3013916634764167e-06, "reference_active": true, "delta_db": -1.354922580735195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.079905260013679e-07, "output_power": 7.411083026985367e-07, "reference_active": true, "delta_db": -0.8819963837234867}], "active_band_count": 3, "active_band_mae_db": 0.8986266680013988}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.068450322528164e-05, "log_rms_mae_db": 0.8915599378075334, "rms_envelope_nrmse": 0.10693544885849729, "envelope_correlation": 0.9922464744022931, "rms_gain_error_db": -0.3936578522247087, "reference_mean_power": 4.071142209359612e-05, "output_mean_power": 3.718351894402781e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.516451938680711e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5218014411010514e-05, "output_power": 1.3696053506769666e-05, "reference_active": true, "delta_db": -0.457625470215372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2444589918408885e-06, "output_power": 1.0539757978952892e-06, "reference_active": true, "delta_db": -0.7214995162058103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2414094224167626e-08, "output_power": 4.30699743824071e-08, "reference_active": true, "delta_db": -0.8527347164063253}], "active_band_count": 3, "active_band_mae_db": 0.6772865676091692}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.7354442023779843e-05, "log_rms_mae_db": 1.6216910312944908, "rms_envelope_nrmse": 0.17736454487514267, "envelope_correlation": 0.9752559862867722, "rms_gain_error_db": -0.2981038260051599, "reference_mean_power": 9.786447369332928e-06, "output_mean_power": 9.137232635116505e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.593718108385407e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.378843387910109e-06, "output_power": 2.192344873473103e-06, "reference_active": true, "delta_db": -0.3545697783068267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.926849454092717e-07, "output_power": 7.849479886013066e-07, "reference_active": true, "delta_db": -0.558573297882671}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6983089704559518e-07, "output_power": 1.1597240110885984e-07, "reference_active": true, "delta_db": -1.6566205460156547}], "active_band_count": 3, "active_band_mae_db": 0.8565878740683841}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.2351945663576383e-05, "log_rms_mae_db": 0.9599711968742617, "rms_envelope_nrmse": 0.09535500652629425, "envelope_correlation": 0.9928241083132254, "rms_gain_error_db": -0.3004106040522598, "reference_mean_power": 9.83170860207206e-06, "output_mean_power": 9.174616881200981e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.2206731147886266e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.414723334288e-06, "output_power": 2.2052371858213096e-06, "reference_active": true, "delta_db": -0.39412071704453255}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.701511911331394e-07, "output_power": 3.6554896636837087e-07, "reference_active": true, "delta_db": -0.05433591039714808}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1816168225696604e-08, "output_power": 7.0860139299675865e-09, "reference_active": true, "delta_db": -2.220746636070311}], "active_band_count": 3, "active_band_mae_db": 0.8897344211706639}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9007770630205475, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "db21328a9987e0e9521a5c14b493f1345ddbe3b26b4e238b49b6328f574758e3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9804063795960123, "standard_deviation": 0.0032602077048680105, "minimum": 0.9768173928131773, "maximum": 0.9848832624044102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5108791004033768, "standard_deviation": 0.02345332115069069, "minimum": 1.474687008737427, "maximum": 1.536050061923955, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7724787225977671, "standard_deviation": 0.20203708523592115, "minimum": 0.4550925776085875, "maximum": 0.97389965843201, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1657487040050743, "standard_deviation": 0.014149758845154528, "minimum": 0.14506237666207425, "maximum": 0.18443278352713907, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9935023110574843, "standard_deviation": 0.0014283787544971708, "minimum": 0.991921083094024, "maximum": 0.9958028932527062, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8879658704155073, "standard_deviation": 0.022568769967936376, "minimum": 0.870877635298059, "maximum": 0.9404282057966864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8075542173918357, "standard_deviation": 0.3235382305385328, "minimum": 0.3154484733540834, "maximum": 1.2061435541689816, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10008443563251536, "standard_deviation": 0.011146978224548338, "minimum": 0.07942062550225158, "maximum": 0.11121834710319435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9771151697332238, "standard_deviation": 0.0020529960180298576, "minimum": 0.9743325033337641, "maximum": 0.9794487076167381, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.575866421943998, "standard_deviation": 0.034988174122812445, "minimum": 1.525097774810762, "maximum": 1.6216910312944908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9611953118839758, "standard_deviation": 0.3194176988240332, "minimum": 0.5117815531538669, "maximum": 1.3741953760591041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17068200882978787, "standard_deviation": 0.006982283944355368, "minimum": 0.16103731083890235, "maximum": 0.179707499127006, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9924759972146113, "standard_deviation": 0.0010951218150622487, "minimum": 0.9909305318515865, "maximum": 0.9938682371629178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9303253577444078, "standard_deviation": 0.04159167860800639, "minimum": 0.8846710683487083, "maximum": 1.0055289841738468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9795322776562162, "standard_deviation": 0.41134693286822815, "minimum": 0.3847361091254155, "maximum": 1.6575131524932694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09657842595601324, "standard_deviation": 0.0057456620437748625, "minimum": 0.08621749143812452, "maximum": 0.10357322411104988, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9917142216645268, "standard_deviation": 0.0015474812372277046, "minimum": 0.989576350563214, "maximum": 0.9934688260044046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5025976303232405, "standard_deviation": 0.044111097823325694, "minimum": 1.4457936706710015, "maximum": 1.569300507878896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4343409249722457, "standard_deviation": 0.22635463734879832, "minimum": 0.16672885267450147, "maximum": 0.744346438914054, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10778594745795761, "standard_deviation": 0.010054330315702652, "minimum": 0.09514465145105656, "maximum": 0.12225174364537379, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976116668897475, "standard_deviation": 0.00044603105859431863, "minimum": 0.9968624477505428, "maximum": 0.9981075958966689, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.891093673225472, "standard_deviation": 0.048356322043847086, "minimum": 0.8247969560581764, "maximum": 0.9641401842196251, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5683369248973115, "standard_deviation": 0.22362782446853513, "minimum": 0.2246086495910067, "maximum": 0.8368186781228696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05955447854659346, "standard_deviation": 0.005755662187988325, "minimum": 0.052463095055989416, "maximum": 0.07064092753863899, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23563022541206494, "standard_deviation": 0.037532810808305526, "minimum": 0.19926089124521726, "maximum": 0.3217116810006939, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.369774776625266, "standard_deviation": 0.010330871024790125, "minimum": 5.356603837156285, "maximum": 5.387312380599643, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2197403346739881, "standard_deviation": 0.011918466248411594, "minimum": 0.2048143821334081, "maximum": 0.23590545529297874, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12217366140903924, "standard_deviation": 0.030135029485583043, "minimum": -0.16122898826382526, "maximum": -0.0708501118064436, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9951086880849798, "standard_deviation": 0.00039381701230201864, "minimum": 0.994599618712125, "maximum": 0.9956306128228241, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7135091032319341, "standard_deviation": 0.021999088626660276, "minimum": 0.6727121094454726, "maximum": 0.7445658418615377, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34271736054853863, "standard_deviation": 0.17082215728582015, "minimum": 0.10692206366880865, "maximum": 0.552408438289205, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07462888118715726, "standard_deviation": 0.00315875825760388, "minimum": 0.07036898277458258, "maximum": 0.07899395136059795, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987236264047887, "standard_deviation": 0.00016941477751722842, "minimum": 0.9983247338525967, "maximum": 0.9988640284007434, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.39035155940354005, "standard_deviation": 0.021747727293147143, "minimum": 0.35968512025777305, "maximum": 0.42282471814415096, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3159307223848825, "standard_deviation": 0.14439616072813133, "minimum": 0.10968498440645462, "maximum": 0.46465525433166915, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.039015967967046136, "standard_deviation": 0.0020056972754511635, "minimum": 0.03667787310219672, "maximum": 0.04307944307170986, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5639-40744-0030", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.1740145683288574, "predicted_mos": 3.1740145683288574, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1c8087902628016e6aa80d934711402b062350ce88e700dca5257fda98d9ae5c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9358124998398125, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7008601488055463, "全帯域_10ms_rms_envelope_nrmse": 0.07292246281388946, "全帯域_10ms_envelope_correlation": 0.9952484691930203, "全帯域_10ms_modulation_mae_db": 0.20436865132742002, "全帯域_40ms_log_rms_mae_db": 0.39253362027226035, "全帯域_40ms_rms_envelope_nrmse": 0.04139706169904237, "全帯域_40ms_envelope_correlation": 0.9984186726753977, "全帯域_40ms_modulation_mae_db": 0.3797821823255416, "300–1000 Hz_10ms_log_rms_mae_db": 1.4857258076217825, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10490593146648272, "300–1000 Hz_10ms_envelope_correlation": 0.9920289943828049, "300–1000 Hz_10ms_modulation_mae_db": 0.12921678812450807, "300–1000 Hz_40ms_log_rms_mae_db": 0.8911697330204829, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06648025753680015, "300–1000 Hz_40ms_envelope_correlation": 0.996670659575477, "300–1000 Hz_40ms_modulation_mae_db": 0.4740044150008717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6083144162787806, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17406726509079415, "1000–3000 Hz_10ms_envelope_correlation": 0.9779833590923628, "1000–3000 Hz_10ms_modulation_mae_db": 0.649566557375382, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9554792319185538, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09762577299428866, "1000–3000 Hz_40ms_envelope_correlation": 0.9933781531396317, "1000–3000 Hz_40ms_modulation_mae_db": 0.7275718278399403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5889924009815524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19586501776323656, "3000–6000 Hz_10ms_envelope_correlation": 0.9696371446185398, "3000–6000 Hz_10ms_modulation_mae_db": 1.3299429402987135, "3000–6000 Hz_40ms_log_rms_mae_db": 0.95500797597472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1105580726479028, "3000–6000 Hz_40ms_envelope_correlation": 0.9900918723784622, "3000–6000 Hz_40ms_modulation_mae_db": 1.3102392549475292, "mr_stft_db": 5.360266764664204, "band_mae_db": 0.20196749115708698, "rms_gain_error_db": -0.07255123247993747, "output_peak": 0.2143198386211967}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7221092577148943, "全帯域_10ms_rms_envelope_nrmse": 0.08091001635140253, "全帯域_10ms_envelope_correlation": 0.9941905794858606, "全帯域_10ms_modulation_mae_db": 0.6356246412082306, "全帯域_40ms_log_rms_mae_db": 0.4164267545195609, "全帯域_40ms_rms_envelope_nrmse": 0.04905839469524439, "全帯域_40ms_envelope_correlation": 0.9978227697967136, "全帯域_40ms_modulation_mae_db": 0.4229994042106891, "300–1000 Hz_10ms_log_rms_mae_db": 1.517238533980602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11248502818503744, "300–1000 Hz_10ms_envelope_correlation": 0.9909408793172129, "300–1000 Hz_10ms_modulation_mae_db": 0.739456200693299, "300–1000 Hz_40ms_log_rms_mae_db": 0.848244144398574, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0672788111183726, "300–1000 Hz_40ms_envelope_correlation": 0.9967743211582104, "300–1000 Hz_40ms_modulation_mae_db": 0.709321939657969, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4431523405580093, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1449276646140013, "1000–3000 Hz_10ms_envelope_correlation": 0.9851396648081131, "1000–3000 Hz_10ms_modulation_mae_db": 1.0067293275768197, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8166742340930097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08120835386374564, "1000–3000 Hz_40ms_envelope_correlation": 0.9958695933366714, "1000–3000 Hz_40ms_modulation_mae_db": 1.0383351557588065, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5875854876672002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18194666685625296, "3000–6000 Hz_10ms_envelope_correlation": 0.9746521182631652, "3000–6000 Hz_10ms_modulation_mae_db": 1.5031448578635622, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9286587499537647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11427446158257254, "3000–6000 Hz_40ms_envelope_correlation": 0.990330999945502, "3000–6000 Hz_40ms_modulation_mae_db": 1.5869416801580527, "mr_stft_db": 5.36031067378152, "band_mae_db": 0.2485860892799735, "rms_gain_error_db": -0.11175794339492565, "output_peak": 0.21935939343032615}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7215756156854998, "全帯域_10ms_rms_envelope_nrmse": 0.07376717072716668, "全帯域_10ms_envelope_correlation": 0.99520559343496, "全帯域_10ms_modulation_mae_db": 0.4638667625653798, "全帯域_40ms_log_rms_mae_db": 0.3731894573767665, "全帯域_40ms_rms_envelope_nrmse": 0.041089249676159914, "全帯域_40ms_envelope_correlation": 0.9985316179420183, "全帯域_40ms_modulation_mae_db": 0.4592452900260204, "300–1000 Hz_10ms_log_rms_mae_db": 1.4968364217413233, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10847178367378983, "300–1000 Hz_10ms_envelope_correlation": 0.9915120039787241, "300–1000 Hz_10ms_modulation_mae_db": 0.39727627995311526, "300–1000 Hz_40ms_log_rms_mae_db": 0.8807214979333848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06185946786275621, "300–1000 Hz_40ms_envelope_correlation": 0.9972037688829263, "300–1000 Hz_40ms_modulation_mae_db": 0.7097025912457218, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5113811599059874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.172114524459035, "1000–3000 Hz_10ms_envelope_correlation": 0.9804326155248256, "1000–3000 Hz_10ms_modulation_mae_db": 1.1116560506184936, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9178936286762146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11488986960445136, "1000–3000 Hz_40ms_envelope_correlation": 0.9929564054886427, "1000–3000 Hz_40ms_modulation_mae_db": 0.731672283306139, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5664073964009397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1611181033134683, "3000–6000 Hz_10ms_envelope_correlation": 0.9804490317801708, "3000–6000 Hz_10ms_modulation_mae_db": 1.218142465799751, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9054834147740444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08846827520566211, "3000–6000 Hz_40ms_envelope_correlation": 0.9949328972932237, "3000–6000 Hz_40ms_modulation_mae_db": 1.324719264208662, "mr_stft_db": 5.358545430452092, "band_mae_db": 0.289514648357123, "rms_gain_error_db": -0.11548932519424253, "output_peak": 0.21086528226170653}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.700071874421697, "全帯域_10ms_rms_envelope_nrmse": 0.07381154460385914, "全帯域_10ms_envelope_correlation": 0.9951895802718382, "全帯域_10ms_modulation_mae_db": 0.3016848306819145, "全帯域_40ms_log_rms_mae_db": 0.3609190435695395, "全帯域_40ms_rms_envelope_nrmse": 0.03660303740648551, "全帯域_40ms_envelope_correlation": 0.998858442203961, "全帯域_40ms_modulation_mae_db": 0.3566872265449698, "300–1000 Hz_10ms_log_rms_mae_db": 1.4612207800319126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10019126942064117, "300–1000 Hz_10ms_envelope_correlation": 0.9927786145233036, "300–1000 Hz_10ms_modulation_mae_db": 0.4264999148446087, "300–1000 Hz_40ms_log_rms_mae_db": 0.8845533124070621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05312267518297159, "300–1000 Hz_40ms_envelope_correlation": 0.9979766670423577, "300–1000 Hz_40ms_modulation_mae_db": 0.47438829481196815, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4547410283757667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16012472328858612, "1000–3000 Hz_10ms_envelope_correlation": 0.9830561329395916, "1000–3000 Hz_10ms_modulation_mae_db": 0.7891948544886489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8194716802057269, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10325740208748885, "1000–3000 Hz_40ms_envelope_correlation": 0.9944764416389121, "1000–3000 Hz_40ms_modulation_mae_db": 0.6846985265243976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5975598211361057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18759756293761579, "3000–6000 Hz_10ms_envelope_correlation": 0.9720157174544125, "3000–6000 Hz_10ms_modulation_mae_db": 0.8414453722938099, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9435679452499677, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12347069394730631, "3000–6000 Hz_40ms_envelope_correlation": 0.9867744329676451, "3000–6000 Hz_40ms_modulation_mae_db": 0.8837352659083108, "mr_stft_db": 5.35855805614204, "band_mae_db": 0.24889242162610062, "rms_gain_error_db": -0.11142822672824883, "output_peak": 0.20985361094023783}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.735012515474326, "全帯域_10ms_rms_envelope_nrmse": 0.08392442838632252, "全帯域_10ms_envelope_correlation": 0.9936641055298889, "全帯域_10ms_modulation_mae_db": 0.2557346117512657, "全帯域_40ms_log_rms_mae_db": 0.4147901221736672, "全帯域_40ms_rms_envelope_nrmse": 0.047704787765492696, "全帯域_40ms_envelope_correlation": 0.9978411315986061, "全帯域_40ms_modulation_mae_db": 0.3362969190229443, "300–1000 Hz_10ms_log_rms_mae_db": 1.5347469930454476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1163453305616534, "300–1000 Hz_10ms_envelope_correlation": 0.9901802856706369, "300–1000 Hz_10ms_modulation_mae_db": 0.2977980864486383, "300–1000 Hz_40ms_log_rms_mae_db": 0.8947646178261456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0705909118068525, "300–1000 Hz_40ms_envelope_correlation": 0.9962501685674123, "300–1000 Hz_40ms_modulation_mae_db": 0.5923215423805472, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4724096222554606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16253707033142142, "1000–3000 Hz_10ms_envelope_correlation": 0.9807028534991825, "1000–3000 Hz_10ms_modulation_mae_db": 0.7223316615965386, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8605308952213332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10202963111795178, "1000–3000 Hz_40ms_envelope_correlation": 0.992310574063641, "1000–3000 Hz_40ms_modulation_mae_db": 0.9430500298748375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6112309933834552, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1725858568154526, "3000–6000 Hz_10ms_envelope_correlation": 0.9767699432517923, "3000–6000 Hz_10ms_modulation_mae_db": 0.7822141422176255, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9175091905631406, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09238169090046038, "3000–6000 Hz_40ms_envelope_correlation": 0.9935406586614336, "3000–6000 Hz_40ms_modulation_mae_db": 0.7423850067867552, "mr_stft_db": 5.3558530026420135, "band_mae_db": 0.23051724059326473, "rms_gain_error_db": -0.05155981435354945, "output_peak": 0.21599588804748213}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7113160556165637, "全帯域_10ms_rms_envelope_nrmse": 0.07218010128366995, "全帯域_10ms_envelope_correlation": 0.9955021294494439, "全帯域_10ms_modulation_mae_db": 0.35723039515904603, "全帯域_40ms_log_rms_mae_db": 0.3807974254955951, "全帯域_40ms_rms_envelope_nrmse": 0.04044321204156997, "全帯域_40ms_envelope_correlation": 0.9986835083592738, "全帯域_40ms_modulation_mae_db": 0.3299600319684083, "300–1000 Hz_10ms_log_rms_mae_db": 1.4584361653416587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10873114142057161, "300–1000 Hz_10ms_envelope_correlation": 0.9916983465624547, "300–1000 Hz_10ms_modulation_mae_db": 0.33302069731239636, "300–1000 Hz_40ms_log_rms_mae_db": 0.8842628146927454, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06382829008183609, "300–1000 Hz_40ms_envelope_correlation": 0.9973121160837076, "300–1000 Hz_40ms_modulation_mae_db": 0.4900709173984814, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4848864291695891, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15307813997341824, "1000–3000 Hz_10ms_envelope_correlation": 0.9832126142644797, "1000–3000 Hz_10ms_modulation_mae_db": 0.7688884985841108, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8385370124320121, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0830837409790031, "1000–3000 Hz_40ms_envelope_correlation": 0.9955325769710115, "1000–3000 Hz_40ms_modulation_mae_db": 0.7711066720246382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5561539283965427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18314083824157315, "3000–6000 Hz_10ms_envelope_correlation": 0.9750208728053702, "3000–6000 Hz_10ms_modulation_mae_db": 1.4095345567560156, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9531957076059794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11566103053216928, "3000–6000 Hz_40ms_envelope_correlation": 0.9910680649818181, "3000–6000 Hz_40ms_modulation_mae_db": 1.3518186092559568, "mr_stft_db": 5.375086070485021, "band_mae_db": 0.22742195049711744, "rms_gain_error_db": -0.14733521799960678, "output_peak": 0.2126997828341928}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7167289180743102, "全帯域_10ms_rms_envelope_nrmse": 0.07939489902506285, "全帯域_10ms_envelope_correlation": 0.9943644991296547, "全帯域_10ms_modulation_mae_db": 0.2991990071531084, "全帯域_40ms_log_rms_mae_db": 0.38340903091864675, "全帯域_40ms_rms_envelope_nrmse": 0.044000300813079284, "全帯域_40ms_envelope_correlation": 0.9982187117648024, "全帯域_40ms_modulation_mae_db": 0.36776536294756035, "300–1000 Hz_10ms_log_rms_mae_db": 1.477225923146944, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10792281209473985, "300–1000 Hz_10ms_envelope_correlation": 0.9916672092934496, "300–1000 Hz_10ms_modulation_mae_db": 0.51417636378969, "300–1000 Hz_40ms_log_rms_mae_db": 0.833172114982562, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05940775343731919, "300–1000 Hz_40ms_envelope_correlation": 0.9975328540510852, "300–1000 Hz_40ms_modulation_mae_db": 0.7666891959773304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4698329291069634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15812504270323102, "1000–3000 Hz_10ms_envelope_correlation": 0.9815622728035106, "1000–3000 Hz_10ms_modulation_mae_db": 0.4242288586219643, "1000–3000 Hz_40ms_log_rms_mae_db": 0.817597895406216, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09117382637557571, "1000–3000 Hz_40ms_envelope_correlation": 0.9936546425473664, "1000–3000 Hz_40ms_modulation_mae_db": 0.6648126981448952, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5920067801893736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1589389316349766, "3000–6000 Hz_10ms_envelope_correlation": 0.9802819007184831, "3000–6000 Hz_10ms_modulation_mae_db": 0.8372583242464078, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9060899480303553, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08161828076136214, "3000–6000 Hz_40ms_envelope_correlation": 0.9949144670925274, "3000–6000 Hz_40ms_modulation_mae_db": 0.8478300910716404, "mr_stft_db": 5.362929196487285, "band_mae_db": 0.17108501734491535, "rms_gain_error_db": -0.08332662240651509, "output_peak": 0.21144207020488914}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7079325134970944, "全帯域_10ms_rms_envelope_nrmse": 0.07777004803238204, "全帯域_10ms_envelope_correlation": 0.9945986882063153, "全帯域_10ms_modulation_mae_db": 0.24672490896809596, "全帯域_40ms_log_rms_mae_db": 0.3875969901795992, "全帯域_40ms_rms_envelope_nrmse": 0.0431215321880266, "全帯域_40ms_envelope_correlation": 0.9982941922960281, "全帯域_40ms_modulation_mae_db": 0.13629833971972322, "300–1000 Hz_10ms_log_rms_mae_db": 1.4526846368198052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10958931786905167, "300–1000 Hz_10ms_envelope_correlation": 0.9913210610440519, "300–1000 Hz_10ms_modulation_mae_db": 0.1881705612268345, "300–1000 Hz_40ms_log_rms_mae_db": 0.832266447346219, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05986309254488762, "300–1000 Hz_40ms_envelope_correlation": 0.9973787388959517, "300–1000 Hz_40ms_modulation_mae_db": 0.3094704250852536, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5222300014525292, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15494188217938656, "1000–3000 Hz_10ms_envelope_correlation": 0.9832600823157899, "1000–3000 Hz_10ms_modulation_mae_db": 0.8531557986897883, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8980994984702413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09337330290371226, "1000–3000 Hz_40ms_envelope_correlation": 0.9947186859592692, "1000–3000 Hz_40ms_modulation_mae_db": 0.6890129180711208, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5981865599480558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17719507348252844, "3000–6000 Hz_10ms_envelope_correlation": 0.9751172303051396, "3000–6000 Hz_10ms_modulation_mae_db": 0.519247094828174, "3000–6000 Hz_40ms_log_rms_mae_db": 0.928615486096624, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09274985472856091, "3000–6000 Hz_40ms_envelope_correlation": 0.992879471352394, "3000–6000 Hz_40ms_modulation_mae_db": 0.8030474385301609, "mr_stft_db": 5.358739988369976, "band_mae_db": 0.26579004693198155, "rms_gain_error_db": -0.08518480775568431, "output_peak": 0.2173422144759559}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.397773421082889, "spectral_convergence": 0.2468053569817378, "spectral_convergence_status": "ok", "amplitude_floor": 6.716189184049609e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.137321405405449, "spectral_convergence": 0.23653280797789303, "spectral_convergence_status": "ok", "amplitude_floor": 4.979433430044932e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.545705467504275, "spectral_convergence": 0.24897235953119126, "spectral_convergence_status": "ok", "amplitude_floor": 3.630042469105695e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.462171010234058e-05, "output_power": 7.832669991035596e-05, "delta_db": 0.2104463154570328, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003077822779549395, "output_power": 0.00029775906942404776, "delta_db": -0.14378611118118018, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 5.338393512143839e-05, "output_power": 5.5212771596650335e-05, "delta_db": 0.14628964569513248, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.1515857114073014e-05, "output_power": 2.0319443812716253e-05, "delta_db": -0.24846835295596886, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.940885952831811e-05, "output_power": 2.7092969798810275e-05, "delta_db": -0.35621570072051334, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.017666424897867e-06, "output_power": 7.82326607860557e-06, "delta_db": -0.10659882093269417, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20196749115708698, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07255123247993747, "pitch": null, "output_rms_dbfs": -33.12886608007594, "output_peak": 0.2143198386211967, "reference_rms_dbfs": -33.056314847596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.360266764664204}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.650394450462669e-05, "log_rms_mae_db": 0.7008601488055463, "rms_envelope_nrmse": 0.07292246281388946, "envelope_correlation": 0.9952484691930203, "rms_gain_error_db": -0.07243628338932478, "reference_mean_power": 0.0004951134756771775, "output_mean_power": 0.0004869239299057003}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.645854738282274e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010486873457313227, "output_power": 0.00010295241262209858, "reference_active": true, "delta_db": -0.08009499025531956}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5071002541192514e-05, "output_power": 2.399763234143996e-05, "reference_active": true, "delta_db": -0.19003305519047575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059292105678353e-06, "output_power": 4.675111400808301e-06, "reference_active": true, "delta_db": -0.3429779085364647}], "active_band_count": 3, "active_band_mae_db": 0.20436865132742002}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.020301765617229e-05, "log_rms_mae_db": 0.39253362027226035, "rms_envelope_nrmse": 0.04139706169904237, "envelope_correlation": 0.9984186726753977, "rms_gain_error_db": -0.07233173597370844, "reference_mean_power": 0.0004970274687625529, "output_mean_power": 0.0004888180312610513}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.472129124053938e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.182773781582673e-05, "output_power": 9.060962714107346e-05, "reference_active": true, "delta_db": -0.057995422818404724}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2533326164459841e-05, "output_power": 1.2114810749649839e-05, "reference_active": true, "delta_db": -0.14749707561114397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4723702388808576e-07, "output_power": 1.9940135367722374e-07, "reference_active": true, "delta_db": -0.9338540485470761}], "active_band_count": 3, "active_band_mae_db": 0.3797821823255416}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.432622539351727e-05, "log_rms_mae_db": 1.4857258076217825, "rms_envelope_nrmse": 0.10490593146648272, "envelope_correlation": 0.9920289943828049, "rms_gain_error_db": -0.09538603056304788, "reference_mean_power": 0.00026104952136561866, "output_mean_power": 0.00025537848027796855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.392840173310503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.633554932680893e-05, "output_power": 9.39460867943104e-05, "reference_active": true, "delta_db": -0.10907883343693066}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.160826072139944e-05, "output_power": 2.1016063672189496e-05, "reference_active": true, "delta_db": -0.1206843577204836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.769645916048805e-06, "output_power": 4.599360194027147e-06, "reference_active": true, "delta_db": -0.1578871732161099}], "active_band_count": 3, "active_band_mae_db": 0.12921678812450807}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.77451695930137e-05, "log_rms_mae_db": 0.8911697330204829, "rms_envelope_nrmse": 0.06648025753680015, "envelope_correlation": 0.996670659575477, "rms_gain_error_db": -0.0941528544301202, "reference_mean_power": 0.00026200159904973233, "output_mean_power": 0.0002563826643646377}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2678124187221727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604417147270571e-05, "output_power": 8.441927152624255e-05, "reference_active": true, "delta_db": -0.08279856875804016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1040946271380912e-05, "output_power": 1.0948190325142163e-05, "reference_active": true, "delta_db": -0.03663957824959738}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9369342076961927e-07, "output_power": 2.1758888559443098e-07, "reference_active": true, "delta_db": -1.3025750979949775}], "active_band_count": 3, "active_band_mae_db": 0.4740044150008717}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.588832427367551e-05, "log_rms_mae_db": 1.6083144162787806, "rms_envelope_nrmse": 0.17406726509079415, "envelope_correlation": 0.9779833590923628, "rms_gain_error_db": -0.3249914629000762, "reference_mean_power": 4.0575597065913316e-05, "output_mean_power": 3.765006868804847e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.7364679678483124e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6458578452566544e-05, "output_power": 1.584500572398069e-05, "reference_active": true, "delta_db": -0.16499921539126994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.144001855335748e-06, "output_power": 2.287319436119168e-06, "reference_active": true, "delta_db": -1.3815597331541802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.079905260013679e-07, "output_power": 8.276891168065099e-07, "reference_active": true, "delta_db": -0.4021407235806959}], "active_band_count": 3, "active_band_mae_db": 0.649566557375382}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.068450322528164e-05, "log_rms_mae_db": 0.9554792319185538, "rms_envelope_nrmse": 0.09762577299428866, "envelope_correlation": 0.9933781531396317, "rms_gain_error_db": -0.32651454717866596, "reference_mean_power": 4.071142209359612e-05, "output_mean_power": 3.7762854764711724e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.516451938680711e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5218014411010514e-05, "output_power": 1.46560526454364e-05, "reference_active": true, "delta_db": -0.16340974689898363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2444589918408885e-06, "output_power": 9.136863586355489e-07, "reference_active": true, "delta_db": -1.3418344911748004}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2414094224167626e-08, "output_power": 4.4843650646095386e-08, "reference_active": true, "delta_db": -0.6774712454460369}], "active_band_count": 3, "active_band_mae_db": 0.7275718278399403}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.7354442023779843e-05, "log_rms_mae_db": 1.5889924009815524, "rms_envelope_nrmse": 0.19586501776323656, "envelope_correlation": 0.9696371446185398, "rms_gain_error_db": -0.3032325721366135, "reference_mean_power": 9.786447369332928e-06, "output_mean_power": 9.126448503956061e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.593718108385407e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.378843387910109e-06, "output_power": 2.08169363739739e-06, "reference_active": true, "delta_db": -0.5794903621991122}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.926849454092717e-07, "output_power": 6.400508728111238e-07, "reference_active": true, "delta_db": -1.4448371657734995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6983089704559518e-07, "output_power": 1.0801064866232544e-07, "reference_active": true, "delta_db": -1.9655012929235285}], "active_band_count": 3, "active_band_mae_db": 1.3299429402987135}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.2351945663576383e-05, "log_rms_mae_db": 0.95500797597472, "rms_envelope_nrmse": 0.1105580726479028, "envelope_correlation": 0.9900918723784622, "rms_gain_error_db": -0.3066802726283688, "reference_mean_power": 9.83170860207206e-06, "output_mean_power": 9.16138155148699e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.2206731147886266e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.414723334288e-06, "output_power": 2.058925558953857e-06, "reference_active": true, "delta_db": -0.6922673400605811}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.701511911331394e-07, "output_power": 2.7932640185116035e-07, "reference_active": true, "delta_db": -1.2226716423376844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1816168225696604e-08, "output_power": 7.428459965097609e-09, "reference_active": true, "delta_db": -2.015778782444322}], "active_band_count": 3, "active_band_mae_db": 1.3102392549475292}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8987828583784998, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0e3ba5de4a24d89b3b8177828ae2c11062e1e8c5f6743ee468f88693019978cd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.981918699405982, "standard_deviation": 0.0022181899575596927, "minimum": 0.9779833590923628, "maximum": 0.9851396648081131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4958684908878856, "standard_deviation": 0.05264285819010378, "minimum": 1.4431523405580093, "maximum": 1.6083144162787806, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7907189509439683, "standard_deviation": 0.21144204324517443, "minimum": 0.4242288586219643, "maximum": 1.1116560506184936, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15998953907998423, "standard_deviation": 0.009675362800796672, "minimum": 0.1449276646140013, "maximum": 0.17406726509079415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941121341431433, "standard_deviation": 0.0012499314162836468, "minimum": 0.992310574063641, "maximum": 0.9958695933366714, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8655355095529134, "standard_deviation": 0.05271613968157757, "minimum": 0.8166742340930097, "maximum": 0.9554792319185538, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7812825139430969, "standard_deviation": 0.13580538247253215, "minimum": 0.6648126981448952, "maximum": 1.0383351557588065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09583023749077717, "standard_deviation": 0.01111035633677156, "minimum": 0.08120835386374564, "maximum": 0.11488986960445136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9754929948996341, "standard_deviation": 0.0037144530526597216, "minimum": 0.9696371446185398, "maximum": 0.9804490317801708, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5872654210129031, "standard_deviation": 0.017841132897374313, "minimum": 1.5561539283965427, "maximum": 1.6112309933834552, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0551162192880073, "standard_deviation": 0.35520574682780837, "minimum": 0.519247094828174, "maximum": 1.5031448578635622, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17729850638063804, "standard_deviation": 0.012692086065230975, "minimum": 0.1589389316349766, "maximum": 0.19586501776323656, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9918166080841258, "standard_deviation": 0.0027922241280289545, "minimum": 0.9867744329676451, "maximum": 0.9949328972932237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9297660522810745, "standard_deviation": 0.01956520485407931, "minimum": 0.9054834147740444, "maximum": 0.95500797597472, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1063395763583836, "standard_deviation": 0.32101001038196164, "minimum": 0.7423850067867552, "maximum": 1.5869416801580527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10239779503824956, "standard_deviation": 0.015338330388457323, "minimum": 0.08161828076136214, "maximum": 0.12347069394730631, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915159243465799, "standard_deviation": 0.0007621761025315586, "minimum": 0.9901802856706369, "maximum": 0.9927786145233036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4855144077161846, "standard_deviation": 0.0293462207999314, "minimum": 1.4526846368198052, "maximum": 1.5347469930454476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.37820186154913626, "standard_deviation": 0.1920944205117761, "minimum": 0.12921678812450807, "maximum": 0.739456200693299, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10858032683649596, "standard_deviation": 0.004794406805718486, "minimum": 0.10019126942064117, "maximum": 0.1163453305616534, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997137411782141, "standard_deviation": 0.0005465189265097895, "minimum": 0.9962501685674123, "maximum": 0.9979766670423577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.868644335325897, "standard_deviation": 0.02626880264050548, "minimum": 0.832266447346219, "maximum": 0.8947646178261456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.565746165194768, "standard_deviation": 0.15610830270495288, "minimum": 0.3094704250852536, "maximum": 0.7666891959773304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06280390744647449, "standard_deviation": 0.00547758534077173, "minimum": 0.05312267518297159, "maximum": 0.0705909118068525, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2354718632234454, "standard_deviation": 0.03693520862140263, "minimum": 0.17108501734491535, "maximum": 0.289514648357123, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.361286147878019, "standard_deviation": 0.0059291552815102875, "minimum": 5.3558530026420135, "maximum": 5.375086070485021, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2139847601019984, "standard_deviation": 0.0033643099568163868, "minimum": 0.20985361094023783, "maximum": 0.21935939343032615, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09732914878908876, "standard_deviation": 0.029975480344039317, "minimum": -0.14733521799960678, "maximum": -0.05155981435354945, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947454555876227, "standard_deviation": 0.0006412516239813618, "minimum": 0.9936641055298889, "maximum": 0.9955021294494439, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7144508624112414, "standard_deviation": 0.011839177820505049, "minimum": 0.700071874421697, "maximum": 0.735012515474326, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34555422610180764, "standard_deviation": 0.14148019986564936, "minimum": 0.20436865132742002, "maximum": 0.6356246412082306, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07683508390296939, "standard_deviation": 0.004306507354586452, "minimum": 0.07218010128366995, "maximum": 0.08392442838632252, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983336308296001, "standard_deviation": 0.0003708231294769153, "minimum": 0.9978227697967136, "maximum": 0.998858442203961, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3887078055632044, "standard_deviation": 0.01915420017027938, "minimum": 0.3609190435695395, "maximum": 0.4164267545195609, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.34862934459573214, "standard_deviation": 0.09617951719511561, "minimum": 0.13629833971972322, "maximum": 0.4592452900260204, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04292719703563759, "standard_deviation": 0.004026533725569451, "minimum": 0.03660303740648551, "maximum": 0.04905839469524439, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5683-32865-0017", "language": "en", "style": "normal", "speaker": "librispeech_5683", "source_utterance_id": "5683-32865-0017", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 87280, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5683-32865-0017.wav", "source_file_sha256": "538e548c7f1ff5ba24244bcd5357ec3a7990e6d822c94fa1c0bc64f943f526bf", "decoded_pcm_sha256": "44477e160959a9526d3545567baa9b096405cf9d914595d737ffdf3d46ce1e6f", "original_source_sha256": "d0027b49e5e0c29b3b59d1d782b8d4d9753f90e4343f243a15698e787c8e64ca", "corpus_entry_id": "librispeech_5683-32865-0017"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 2.519167900085449, "predicted_mos": 2.519167900085449, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "a2cb55f48070d07f4f7a09d098a56a574637adb9ebd5453423db08b1cff77bb5", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8094930660008686, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5683-32865-0017", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "a2cb55f48070d07f4f7a09d098a56a574637adb9ebd5453423db08b1cff77bb5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "6b805ad37580bd662b327382203a4d8dcbce78d1e6ea6d584ccb141d9858eb07"}, "rows": [{"case": "librispeech_5683-32865-0017", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.330758571624756, "predicted_mos": 2.330758571624756, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "eadb689453903b203b86d994f7046d0e6d55f06db0f64940deb53bdc3f134667", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2804284260018903, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5683-32865-0017", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "eadb689453903b203b86d994f7046d0e6d55f06db0f64940deb53bdc3f134667", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1609930940325506, "全帯域_10ms_rms_envelope_nrmse": 0.0995744471248366, "全帯域_10ms_envelope_correlation": 0.9904088355800083, "全帯域_10ms_modulation_mae_db": 0.15755925822027358, "全帯域_40ms_log_rms_mae_db": 0.6376161174769903, "全帯域_40ms_rms_envelope_nrmse": 0.05820113742784233, "全帯域_40ms_envelope_correlation": 0.9966191405238479, "全帯域_40ms_modulation_mae_db": 0.34332561985809545, "300–1000 Hz_10ms_log_rms_mae_db": 1.7543243261589079, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1248254916165632, "300–1000 Hz_10ms_envelope_correlation": 0.9875977343332044, "300–1000 Hz_10ms_modulation_mae_db": 0.14007614037047525, "300–1000 Hz_40ms_log_rms_mae_db": 1.1346933233274452, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06524232164969458, "300–1000 Hz_40ms_envelope_correlation": 0.9964178891697607, "300–1000 Hz_40ms_modulation_mae_db": 0.5309755810376385, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7233635401274316, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2216973963320271, "1000–3000 Hz_10ms_envelope_correlation": 0.9683809270608147, "1000–3000 Hz_10ms_modulation_mae_db": 0.9631006852177234, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2026001211305102, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14578657230000533, "1000–3000 Hz_40ms_envelope_correlation": 0.988870064892573, "1000–3000 Hz_40ms_modulation_mae_db": 0.9122334477934785, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5756187436905371, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22376511626948642, "3000–6000 Hz_10ms_envelope_correlation": 0.9656772053265837, "3000–6000 Hz_10ms_modulation_mae_db": 0.8018775860517979, "3000–6000 Hz_40ms_log_rms_mae_db": 1.002183790714111, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1273428895646251, "3000–6000 Hz_40ms_envelope_correlation": 0.9877896036002339, "3000–6000 Hz_40ms_modulation_mae_db": 1.0637304309490276, "mr_stft_db": 5.82519412849859, "band_mae_db": 0.31257141290425194, "rms_gain_error_db": -0.15584174251532823, "output_peak": 0.13516089384274924}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.293773065335117, "全帯域_10ms_rms_envelope_nrmse": 0.09926670999732186, "全帯域_10ms_envelope_correlation": 0.9903551888822876, "全帯域_10ms_modulation_mae_db": 0.18664242398476025, "全帯域_40ms_log_rms_mae_db": 0.699780274933148, "全帯域_40ms_rms_envelope_nrmse": 0.05021451814274032, "全帯域_40ms_envelope_correlation": 0.9974019770818953, "全帯域_40ms_modulation_mae_db": 0.4706349105748198, "300–1000 Hz_10ms_log_rms_mae_db": 1.7169302401413307, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12895833515823024, "300–1000 Hz_10ms_envelope_correlation": 0.9867746782162086, "300–1000 Hz_10ms_modulation_mae_db": 0.12936469940939457, "300–1000 Hz_40ms_log_rms_mae_db": 1.0595374817254835, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0714954232821385, "300–1000 Hz_40ms_envelope_correlation": 0.9956382544451017, "300–1000 Hz_40ms_modulation_mae_db": 0.2974669945686252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7771513785660438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23123226811011294, "1000–3000 Hz_10ms_envelope_correlation": 0.9667282616042704, "1000–3000 Hz_10ms_modulation_mae_db": 0.8614299893812859, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2816859964045062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1618991075157932, "1000–3000 Hz_40ms_envelope_correlation": 0.9869514525015242, "1000–3000 Hz_40ms_modulation_mae_db": 0.7934771789229749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5067776770996346, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21319321940179906, "3000–6000 Hz_10ms_envelope_correlation": 0.968840879899029, "3000–6000 Hz_10ms_modulation_mae_db": 1.2476246743642492, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9696307791913775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11547852656491014, "3000–6000 Hz_40ms_envelope_correlation": 0.9904675570403567, "3000–6000 Hz_40ms_modulation_mae_db": 1.6011549365645916, "mr_stft_db": 5.817742102399631, "band_mae_db": 0.3409430313811337, "rms_gain_error_db": -0.10916139098456984, "output_peak": 0.15407326148085806}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2102722064037996, "全帯域_10ms_rms_envelope_nrmse": 0.10129810279532114, "全帯域_10ms_envelope_correlation": 0.9899558715873038, "全帯域_10ms_modulation_mae_db": 0.19761506852707478, "全帯域_40ms_log_rms_mae_db": 0.7162283894394382, "全帯域_40ms_rms_envelope_nrmse": 0.05986819312737611, "全帯域_40ms_envelope_correlation": 0.9961551382298999, "全帯域_40ms_modulation_mae_db": 0.29139769127261816, "300–1000 Hz_10ms_log_rms_mae_db": 1.7384568342649345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11831311514930484, "300–1000 Hz_10ms_envelope_correlation": 0.9889508139948612, "300–1000 Hz_10ms_modulation_mae_db": 0.31339223304982006, "300–1000 Hz_40ms_log_rms_mae_db": 1.0622887691851213, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06406992837902623, "300–1000 Hz_40ms_envelope_correlation": 0.9965191181526146, "300–1000 Hz_40ms_modulation_mae_db": 0.6867034186523194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.754727228720894, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23896501960812752, "1000–3000 Hz_10ms_envelope_correlation": 0.961022079889474, "1000–3000 Hz_10ms_modulation_mae_db": 0.35455354002974654, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2454967656058815, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15073137010993973, "1000–3000 Hz_40ms_envelope_correlation": 0.9834881285225837, "1000–3000 Hz_40ms_modulation_mae_db": 0.06477690588756822, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5358860383205981, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24580410057360338, "3000–6000 Hz_10ms_envelope_correlation": 0.9585031187027021, "3000–6000 Hz_10ms_modulation_mae_db": 0.728197877268645, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9894809826918911, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15176713299998573, "3000–6000 Hz_40ms_envelope_correlation": 0.9826052406979885, "3000–6000 Hz_40ms_modulation_mae_db": 0.9617060961186974, "mr_stft_db": 5.834919948020615, "band_mae_db": 0.22119282631858064, "rms_gain_error_db": -0.0075506380331525486, "output_peak": 0.15242971001724406}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1918169193351955, "全帯域_10ms_rms_envelope_nrmse": 0.09934802289183838, "全帯域_10ms_envelope_correlation": 0.9905331532319539, "全帯域_10ms_modulation_mae_db": 0.4620551580936139, "全帯域_40ms_log_rms_mae_db": 0.6628878326125548, "全帯域_40ms_rms_envelope_nrmse": 0.053031503630141946, "全帯域_40ms_envelope_correlation": 0.9973608412276596, "全帯域_40ms_modulation_mae_db": 0.8063824278888688, "300–1000 Hz_10ms_log_rms_mae_db": 1.73167498737819, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12724612920169284, "300–1000 Hz_10ms_envelope_correlation": 0.9871535995306424, "300–1000 Hz_10ms_modulation_mae_db": 0.15283125138830658, "300–1000 Hz_40ms_log_rms_mae_db": 1.1559800732787193, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0739369380935038, "300–1000 Hz_40ms_envelope_correlation": 0.9954510483397732, "300–1000 Hz_40ms_modulation_mae_db": 0.47844630546738526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6384004426909913, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.208127635666715, "1000–3000 Hz_10ms_envelope_correlation": 0.9702117997573373, "1000–3000 Hz_10ms_modulation_mae_db": 0.7473201039301486, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1225154708751488, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1226537264484087, "1000–3000 Hz_40ms_envelope_correlation": 0.9899145288409872, "1000–3000 Hz_40ms_modulation_mae_db": 0.8231165961808418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.506778117806088, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23979346935061654, "3000–6000 Hz_10ms_envelope_correlation": 0.9602816736960063, "3000–6000 Hz_10ms_modulation_mae_db": 0.9431264206836466, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9970600486841757, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14059565661085022, "3000–6000 Hz_40ms_envelope_correlation": 0.9851380784427676, "3000–6000 Hz_40ms_modulation_mae_db": 1.503979015110039, "mr_stft_db": 5.813760681571972, "band_mae_db": 0.2900094767039863, "rms_gain_error_db": -0.1826206506119385, "output_peak": 0.14709655668589028}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2146639780503867, "全帯域_10ms_rms_envelope_nrmse": 0.09795737404893094, "全帯域_10ms_envelope_correlation": 0.9905748645266559, "全帯域_10ms_modulation_mae_db": 0.3214887229726469, "全帯域_40ms_log_rms_mae_db": 0.6893841950750369, "全帯域_40ms_rms_envelope_nrmse": 0.05446004107011929, "全帯域_40ms_envelope_correlation": 0.9968485175378506, "全帯域_40ms_modulation_mae_db": 0.27329675789775876, "300–1000 Hz_10ms_log_rms_mae_db": 1.7730044731534738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.133467758138951, "300–1000 Hz_10ms_envelope_correlation": 0.9860794949243523, "300–1000 Hz_10ms_modulation_mae_db": 0.6131906046466304, "300–1000 Hz_40ms_log_rms_mae_db": 1.1990663607763157, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07476645442435671, "300–1000 Hz_40ms_envelope_correlation": 0.9953479756711051, "300–1000 Hz_40ms_modulation_mae_db": 0.20147083581782657, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7300075034993834, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24014393210478094, "1000–3000 Hz_10ms_envelope_correlation": 0.9608686465242303, "1000–3000 Hz_10ms_modulation_mae_db": 0.8226579555823449, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2090413150298331, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13319218492396045, "1000–3000 Hz_40ms_envelope_correlation": 0.9899699562278479, "1000–3000 Hz_40ms_modulation_mae_db": 0.867523413177724, "3000–6000 Hz_10ms_log_rms_mae_db": 1.516037609990664, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2144768471158371, "3000–6000 Hz_10ms_envelope_correlation": 0.9687341569893253, "3000–6000 Hz_10ms_modulation_mae_db": 1.367898416825607, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9606078042784382, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11673446009343465, "3000–6000 Hz_40ms_envelope_correlation": 0.9909227844716352, "3000–6000 Hz_40ms_modulation_mae_db": 1.8881918700632827, "mr_stft_db": 5.810387714991286, "band_mae_db": 0.3157512148888217, "rms_gain_error_db": -0.08524633728236834, "output_peak": 0.15201456642975342}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.259230744418896, "全帯域_10ms_rms_envelope_nrmse": 0.10764424087867107, "全帯域_10ms_envelope_correlation": 0.9886154549021504, "全帯域_10ms_modulation_mae_db": 0.36498180153900533, "全帯域_40ms_log_rms_mae_db": 0.7209031003121942, "全帯域_40ms_rms_envelope_nrmse": 0.05945143875743819, "全帯域_40ms_envelope_correlation": 0.9961695719437809, "全帯域_40ms_modulation_mae_db": 0.48076409703972867, "300–1000 Hz_10ms_log_rms_mae_db": 1.7521027697298102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14083779041123876, "300–1000 Hz_10ms_envelope_correlation": 0.9842200603519266, "300–1000 Hz_10ms_modulation_mae_db": 0.326777089456942, "300–1000 Hz_40ms_log_rms_mae_db": 1.1600747420031248, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07405244786126759, "300–1000 Hz_40ms_envelope_correlation": 0.9953505812352366, "300–1000 Hz_40ms_modulation_mae_db": 0.8474569175201773, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7116680407605567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2550373694533245, "1000–3000 Hz_10ms_envelope_correlation": 0.9589640627966941, "1000–3000 Hz_10ms_modulation_mae_db": 1.0586730641763327, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2015145770834446, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.158945839430713, "1000–3000 Hz_40ms_envelope_correlation": 0.9838747055859978, "1000–3000 Hz_40ms_modulation_mae_db": 0.975737519660329, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4935489611028436, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.217354301032025, "3000–6000 Hz_10ms_envelope_correlation": 0.9675104266565492, "3000–6000 Hz_10ms_modulation_mae_db": 0.6298480937399004, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9570662375478602, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11841741783324948, "3000–6000 Hz_40ms_envelope_correlation": 0.9896866839592091, "3000–6000 Hz_40ms_modulation_mae_db": 0.7014182884709667, "mr_stft_db": 5.820565169708806, "band_mae_db": 0.2075395424568988, "rms_gain_error_db": -0.015833442724540653, "output_peak": 0.1407427038880446}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2643057267687712, "全帯域_10ms_rms_envelope_nrmse": 0.10619227498266821, "全帯域_10ms_envelope_correlation": 0.9889190359824872, "全帯域_10ms_modulation_mae_db": 0.31743361779214935, "全帯域_40ms_log_rms_mae_db": 0.7379791586707457, "全帯域_40ms_rms_envelope_nrmse": 0.05394042381517232, "全帯域_40ms_envelope_correlation": 0.9969324801558873, "全帯域_40ms_modulation_mae_db": 0.5046141891756725, "300–1000 Hz_10ms_log_rms_mae_db": 1.7884114241396631, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12558312715398862, "300–1000 Hz_10ms_envelope_correlation": 0.9876032500426749, "300–1000 Hz_10ms_modulation_mae_db": 0.2715970674507355, "300–1000 Hz_40ms_log_rms_mae_db": 1.1635568305871156, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07254897607460063, "300–1000 Hz_40ms_envelope_correlation": 0.9955666838177484, "300–1000 Hz_40ms_modulation_mae_db": 0.5881149686932697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.788421900403596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2471119902715897, "1000–3000 Hz_10ms_envelope_correlation": 0.958781682987121, "1000–3000 Hz_10ms_modulation_mae_db": 1.133458552274053, "1000–3000 Hz_40ms_log_rms_mae_db": 1.239834331419297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1475752605626326, "1000–3000 Hz_40ms_envelope_correlation": 0.9873378731181344, "1000–3000 Hz_40ms_modulation_mae_db": 1.0861359550786418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.540576465750307, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21355302229860623, "3000–6000 Hz_10ms_envelope_correlation": 0.9700415220224234, "3000–6000 Hz_10ms_modulation_mae_db": 1.1738208571908746, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9456905610432302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.112024942114884, "3000–6000 Hz_40ms_envelope_correlation": 0.9912638025590664, "3000–6000 Hz_40ms_modulation_mae_db": 1.1816728756761472, "mr_stft_db": 5.82118236554392, "band_mae_db": 0.25548006829788866, "rms_gain_error_db": -0.09470860915362332, "output_peak": 0.16286859525788822}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2609968126436326, "全帯域_10ms_rms_envelope_nrmse": 0.09992032351211429, "全帯域_10ms_envelope_correlation": 0.9904381373522256, "全帯域_10ms_modulation_mae_db": 0.5492677004540748, "全帯域_40ms_log_rms_mae_db": 0.6990858259702857, "全帯域_40ms_rms_envelope_nrmse": 0.05676328483068656, "全帯域_40ms_envelope_correlation": 0.9969222670056452, "全帯域_40ms_modulation_mae_db": 0.7666917568616038, "300–1000 Hz_10ms_log_rms_mae_db": 1.6865968730137995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1303161319779116, "300–1000 Hz_10ms_envelope_correlation": 0.98655167612859, "300–1000 Hz_10ms_modulation_mae_db": 0.27005955716062163, "300–1000 Hz_40ms_log_rms_mae_db": 0.9899575446179268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07119642905118446, "300–1000 Hz_40ms_envelope_correlation": 0.9958809654627738, "300–1000 Hz_40ms_modulation_mae_db": 0.6083190195753546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7533849384342342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23172750607533485, "1000–3000 Hz_10ms_envelope_correlation": 0.9642042014933045, "1000–3000 Hz_10ms_modulation_mae_db": 1.413328311883297, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2158834410946744, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1503909362302956, "1000–3000 Hz_40ms_envelope_correlation": 0.9865875582145732, "1000–3000 Hz_40ms_modulation_mae_db": 1.7599419493387918, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5777747387083698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21952671178746255, "3000–6000 Hz_10ms_envelope_correlation": 0.9702996126320377, "3000–6000 Hz_10ms_modulation_mae_db": 0.7636331753641058, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0058296987902327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1484435677491432, "3000–6000 Hz_40ms_envelope_correlation": 0.9860189876472333, "3000–6000 Hz_40ms_modulation_mae_db": 0.7313502927012226, "mr_stft_db": 5.819133206588631, "band_mae_db": 0.23908637257777532, "rms_gain_error_db": -0.1837052512493323, "output_peak": 0.13136778704597002}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87280, "output_samples": 87280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.900283461281246, "spectral_convergence": 0.3464674558582401, "spectral_convergence_status": "ok", "amplitude_floor": 4.3919411473735467e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.693874589461047, "spectral_convergence": 0.3407599248227673, "spectral_convergence_status": "ok", "amplitude_floor": 3.1709236310047874e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.881424334753475, "spectral_convergence": 0.3563317492367505, "spectral_convergence_status": "ok", "amplitude_floor": 2.989229099554426e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001235363438035874, "output_power": 0.00012401831698922815, "delta_db": 0.016910896264873816, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 7.224896060742283e-05, "output_power": 7.295228448660635e-05, "delta_db": 0.04207292690964508, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.329659782840476e-05, "output_power": 4.222724642819961e-05, "delta_db": -0.10861008893124852, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.747692740945149e-05, "output_power": 3.606992288269391e-05, "delta_db": -1.1933739091598503, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.909876260278539e-06, "output_power": 9.712019187198848e-06, "delta_db": 0.374378591019994, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1191996093186054e-05, "output_power": 2.0519352805615165e-05, "delta_db": -0.14008206513989988, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.31257141290425194, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15584174251532823, "pitch": null, "output_rms_dbfs": -35.14990004035623, "output_peak": 0.13516089384274924, "reference_rms_dbfs": -34.9940582978409, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.82519412849859}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 5.579882181126775e-05, "log_rms_mae_db": 1.1609930940325506, "rms_envelope_nrmse": 0.0995744471248366, "envelope_correlation": 0.9904088355800083, "rms_gain_error_db": -0.1558417243400365, "reference_mean_power": 0.0003169510345489487, "output_mean_power": 0.00030577924305336726}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8938855835285014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.978895728357633e-05, "output_power": 5.529794942869859e-05, "reference_active": true, "delta_db": -0.3391195241691067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1930505918996815e-05, "output_power": 2.1393576792132035e-05, "reference_active": true, "delta_db": -0.10765250277775176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0863861439295844e-06, "output_power": 3.0680305924316252e-06, "reference_active": true, "delta_db": -0.025905747713962243}], "active_band_count": 3, "active_band_mae_db": 0.15755925822027358}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 5.258028604719465e-05, "log_rms_mae_db": 0.6376161174769903, "rms_envelope_nrmse": 0.05820113742784233, "envelope_correlation": 0.9966191405238479, "rms_gain_error_db": -0.15583826974509613, "reference_mean_power": 0.00031870366172679027, "output_mean_power": 0.0003074703387506498}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 1.711395707762808e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.23137186055323e-05, "output_power": 4.875761646882374e-05, "reference_active": true, "delta_db": -0.30573125146887586}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1932036565608145e-05, "output_power": 1.1372850657964112e-05, "reference_active": true, "delta_db": -0.2084523931656183}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.569030812932547e-07, "output_power": 2.2813398763209924e-07, "reference_active": true, "delta_db": -0.5157932149397922}], "active_band_count": 3, "active_band_mae_db": 0.34332561985809545}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 4.2439491809696055e-05, "log_rms_mae_db": 1.7543243261589079, "rms_envelope_nrmse": 0.1248254916165632, "envelope_correlation": 0.9875977343332044, "rms_gain_error_db": -0.07508895140732114, "reference_mean_power": 0.000101477720422738, "output_mean_power": 9.973826464176369e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.361033224981284e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.73677738976729e-05, "output_power": 2.7450293457283145e-05, "reference_active": true, "delta_db": 0.013075184218253435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1422451661963211e-05, "output_power": 1.0819483745768622e-05, "reference_active": true, "delta_db": -0.23552790064227563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.10199254010919e-06, "output_power": 1.1464132270655135e-06, "reference_active": true, "delta_db": 0.1716253362508967}], "active_band_count": 3, "active_band_mae_db": 0.14007614037047525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.7895186903062846e-05, "log_rms_mae_db": 1.1346933233274452, "rms_envelope_nrmse": 0.06524232164969458, "envelope_correlation": 0.9964178891697607, "rms_gain_error_db": -0.0750896201665068, "reference_mean_power": 0.00010203935243697765, "output_mean_power": 0.00010029025413333363}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 7.765913024651851e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5071474654741866e-05, "output_power": 2.5192197933412287e-05, "reference_active": true, "delta_db": 0.020861808093237046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067872704197783e-06, "output_power": 5.6715233817466865e-06, "reference_active": true, "delta_db": -0.2933673442739291}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0845180908068795e-07, "output_power": 8.079175492281462e-08, "reference_active": true, "delta_db": -1.2786975907457494}], "active_band_count": 3, "active_band_mae_db": 0.5309755810376385}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 3.774086107813168e-05, "log_rms_mae_db": 1.7233635401274316, "rms_envelope_nrmse": 0.2216973963320271, "envelope_correlation": 0.9683809270608147, "rms_gain_error_db": -0.7415908089135302, "reference_mean_power": 4.068228265632881e-05, "output_mean_power": 3.429621807646189e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.809144227115543e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.649297963098168e-06, "output_power": 6.251406209048118e-06, "reference_active": true, "delta_db": -1.410031387824581}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.584218334576925e-06, "output_power": 2.2528252786868546e-06, "reference_active": true, "delta_db": -0.5960169274399599}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3501889520111653e-06, "output_power": 1.6547025038745589e-06, "reference_active": true, "delta_db": 0.8832537403886294}], "active_band_count": 3, "active_band_mae_db": 0.9631006852177234}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.332899304538301e-05, "log_rms_mae_db": 1.2026001211305102, "rms_envelope_nrmse": 0.14578657230000533, "envelope_correlation": 0.988870064892573, "rms_gain_error_db": -0.7415936153719358, "reference_mean_power": 4.0907280378011884e-05, "output_mean_power": 3.4485874698822554e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 2.6173037083412648e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.99444690144364e-06, "output_power": 5.897830968256539e-06, "reference_active": true, "delta_db": -1.3209610078754968}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1581434874555947e-06, "output_power": 1.0325219994848712e-06, "reference_active": true, "delta_db": -0.49863055619800967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.731926648355553e-08, "output_power": 1.0785020047976397e-07, "reference_active": true, "delta_db": 0.9171087793069289}], "active_band_count": 3, "active_band_mae_db": 0.9122334477934785}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 1.8891386922351935e-05, "log_rms_mae_db": 1.5756187436905371, "rms_envelope_nrmse": 0.22376511626948642, "envelope_correlation": 0.9656772053265837, "rms_gain_error_db": -0.054005590209901824, "reference_mean_power": 6.700122358829966e-06, "output_mean_power": 6.617320581674452e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 5.001365064934297e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.184828883572986e-06, "output_power": 2.3478139420674096e-06, "reference_active": true, "delta_db": 0.31246248701476753}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0698658461156927e-06, "output_power": 1.0817214454091666e-06, "reference_active": true, "delta_db": 0.04786116262530605}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.241288004170751e-07, "output_power": 3.2727069112239966e-07, "reference_active": true, "delta_db": -2.0453091085153203}], "active_band_count": 3, "active_band_mae_db": 0.8018775860517979}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 1.2324616345992443e-05, "log_rms_mae_db": 1.002183790714111, "rms_envelope_nrmse": 0.1273428895646251, "envelope_correlation": 0.9877896036002339, "rms_gain_error_db": -0.05400329739886932, "reference_mean_power": 6.7368909993733856e-06, "output_mean_power": 6.653638338936584e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 4.500371991965458e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.395916806498352e-06, "output_power": 2.5361617520359478e-06, "reference_active": true, "delta_db": 0.24705214691506408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.187368503320664e-07, "output_power": 5.505975473896582e-07, "reference_active": true, "delta_db": -0.5067170970448022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.878047962583857e-08, "output_power": 2.7829401163825234e-08, "reference_active": true, "delta_db": -2.4374220488872167}], "active_band_count": 3, "active_band_mae_db": 1.0637304309490276}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8305758832893982, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "53106a3e998aa835ef5e1e6cf5568edf0deae6a66f75081d5b3af86e7386a3f0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9636452077641557, "standard_deviation": 0.004400031595035079, "minimum": 0.958781682987121, "maximum": 0.9702117997573373, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7346406216503913, "standard_deviation": 0.046893437651648676, "minimum": 1.6384004426909913, "maximum": 1.788421900403596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9193152753093665, "standard_deviation": 0.3099482627818724, "minimum": 0.35455354002974654, "maximum": 1.413328311883297, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23425538970275156, "standard_deviation": 0.014686877496937806, "minimum": 0.208127635666715, "maximum": 0.2550373694533245, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9871242834880276, "standard_deviation": 0.0024806686816708405, "minimum": 0.9834881285225837, "maximum": 0.9899699562278479, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.214821502330412, "standard_deviation": 0.04616562125958522, "minimum": 1.1225154708751488, "maximum": 1.2816859964045062, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9103678707550438, "standard_deviation": 0.4621507410244757, "minimum": 0.06477690588756822, "maximum": 1.7599419493387918, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14639687469021856, "standard_deviation": 0.012941616697932779, "minimum": 0.1226537264484087, "maximum": 0.1618991075157932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.966236074490582, "standard_deviation": 0.004491404533044811, "minimum": 0.9585031187027021, "maximum": 0.9702996126320377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5316247940586303, "standard_deviation": 0.031827754125118055, "minimum": 1.4935489611028436, "maximum": 1.5777747387083698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9570033876861033, "standard_deviation": 0.27290407625030205, "minimum": 0.6298480937399004, "maximum": 1.367898416825607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22343334847867954, "standard_deviation": 0.012553490168066518, "minimum": 0.21319321940179906, "maximum": 0.24580410057360338, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9879865923023113, "standard_deviation": 0.003148470788000544, "minimum": 0.9826052406979885, "maximum": 0.9912638025590664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9784437378676646, "standard_deviation": 0.023014050510120027, "minimum": 0.9456905610432302, "maximum": 1.0058296987902327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2041504757067467, "standard_deviation": 0.42606241963466734, "minimum": 0.7014182884709667, "maximum": 1.8881918700632827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12885057419138532, "standard_deviation": 0.015887275515251438, "minimum": 0.112024942114884, "maximum": 0.15176713299998573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9868664134403076, "standard_deviation": 0.0013750252099063271, "minimum": 0.9842200603519266, "maximum": 0.9889508139948612, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7426877409975137, "standard_deviation": 0.03203263506517486, "minimum": 1.6865968730137995, "maximum": 1.7884114241396631, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2771610803666158, "standard_deviation": 0.15714671634849683, "minimum": 0.12936469940939457, "maximum": 0.6131906046466304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12869348485098514, "standard_deviation": 0.006627470779551489, "minimum": 0.11831311514930484, "maximum": 0.14083779041123876, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9957715645367642, "standard_deviation": 0.00046409255415787306, "minimum": 0.9953479756711051, "maximum": 0.9965191181526146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1156443906876565, "standard_deviation": 0.07072148904336728, "minimum": 0.9899575446179268, "maximum": 1.1990663607763157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5298692551665746, "standard_deviation": 0.20684973257692507, "minimum": 0.20147083581782657, "maximum": 0.8474569175201773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07091361485197156, "standard_deviation": 0.004069567039605582, "minimum": 0.06406992837902623, "maximum": 0.07476645442435671, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27282174319116714, "standard_deviation": 0.04888035477305791, "minimum": 0.2075395424568988, "maximum": 0.3409430313811337, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.820360664665431, "standard_deviation": 0.007439615416295027, "minimum": 5.810387714991286, "maximum": 5.834919948020615, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.14696925933104973, "standard_deviation": 0.010557035603563091, "minimum": 0.13136778704597002, "maximum": 0.16286859525788822, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10433350781935671, "standard_deviation": 0.06841188413715218, "minimum": -0.1837052512493323, "maximum": -0.0075506380331525486, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.989975067755634, "standard_deviation": 0.0007729733245459828, "minimum": 0.9886154549021504, "maximum": 0.9905748645266559, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2320065683735437, "standard_deviation": 0.04451515819083426, "minimum": 1.1609930940325506, "maximum": 1.293773065335117, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3196304689479499, "standard_deviation": 0.1383183327341963, "minimum": 0.15755925822027358, "maximum": 0.5492677004540748, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10140018702896281, "standard_deviation": 0.00354767775063786, "minimum": 0.09795737404893094, "maximum": 0.10764424087867107, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968012417133083, "standard_deviation": 0.000471738048607135, "minimum": 0.9961551382298999, "maximum": 0.9974019770818953, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6954831118112992, "standard_deviation": 0.03243045209770001, "minimum": 0.6376161174769903, "maximum": 0.7379791586707457, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49213843132114576, "standard_deviation": 0.2018521435395487, "minimum": 0.27329675789775876, "maximum": 0.8063824278888688, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05574131760018963, "standard_deviation": 0.003394632805508207, "minimum": 0.05021451814274032, "maximum": 0.05986819312737611, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32865-0017", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1925547122955322, "predicted_mos": 2.1925547122955322, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "d0322a698a16ca28eae90e5e744f751ada8c35f28dc250e4c71785f95dc150a1", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.0902467998676, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1848001381272621, "全帯域_10ms_rms_envelope_nrmse": 0.10572527460577458, "全帯域_10ms_envelope_correlation": 0.9891259999120482, "全帯域_10ms_modulation_mae_db": 0.2842798258434625, "全帯域_40ms_log_rms_mae_db": 0.6651192485258486, "全帯域_40ms_rms_envelope_nrmse": 0.06344280735050624, "全帯域_40ms_envelope_correlation": 0.9958959509078663, "全帯域_40ms_modulation_mae_db": 0.29197452599130297, "300–1000 Hz_10ms_log_rms_mae_db": 1.6410407502824547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12900789485138298, "300–1000 Hz_10ms_envelope_correlation": 0.9867427957570861, "300–1000 Hz_10ms_modulation_mae_db": 0.324379059118336, "300–1000 Hz_40ms_log_rms_mae_db": 0.9886736669401909, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06715524764165481, "300–1000 Hz_40ms_envelope_correlation": 0.9961990467839831, "300–1000 Hz_40ms_modulation_mae_db": 0.4375476982975495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6720717682929966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23166002648503645, "1000–3000 Hz_10ms_envelope_correlation": 0.963961795946677, "1000–3000 Hz_10ms_modulation_mae_db": 1.0421231518453948, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0825918003690835, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14841403905696093, "1000–3000 Hz_40ms_envelope_correlation": 0.9866761720978561, "1000–3000 Hz_40ms_modulation_mae_db": 0.9290033390870995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5314847945407017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22203028388991514, "3000–6000 Hz_10ms_envelope_correlation": 0.9659700876609312, "3000–6000 Hz_10ms_modulation_mae_db": 0.8347484316257584, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8837099084444816, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12961731451361416, "3000–6000 Hz_40ms_envelope_correlation": 0.9876612939280931, "3000–6000 Hz_40ms_modulation_mae_db": 1.2179290157839069, "mr_stft_db": 5.770086772578747, "band_mae_db": 0.3602850508640252, "rms_gain_error_db": -0.15200035496641562, "output_peak": 0.13528658388145823}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3064242559418846, "全帯域_10ms_rms_envelope_nrmse": 0.09926085277276483, "全帯域_10ms_envelope_correlation": 0.9903298243818665, "全帯域_10ms_modulation_mae_db": 0.12362667333600004, "全帯域_40ms_log_rms_mae_db": 0.7157390206845596, "全帯域_40ms_rms_envelope_nrmse": 0.05113047197662974, "全帯域_40ms_envelope_correlation": 0.99725671864625, "全帯域_40ms_modulation_mae_db": 0.4031718792627436, "300–1000 Hz_10ms_log_rms_mae_db": 1.6341985889787443, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12631230258257703, "300–1000 Hz_10ms_envelope_correlation": 0.9873121350674503, "300–1000 Hz_10ms_modulation_mae_db": 0.15968208193059782, "300–1000 Hz_40ms_log_rms_mae_db": 0.9563318808382826, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07235328282679275, "300–1000 Hz_40ms_envelope_correlation": 0.9955210189103793, "300–1000 Hz_40ms_modulation_mae_db": 0.25656987936773534, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6960242406356658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2328200340695983, "1000–3000 Hz_10ms_envelope_correlation": 0.9648650864193976, "1000–3000 Hz_10ms_modulation_mae_db": 0.7865417547997983, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1002818240083188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15475347602441508, "1000–3000 Hz_40ms_envelope_correlation": 0.9870453687257114, "1000–3000 Hz_40ms_modulation_mae_db": 0.779327737341601, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4320830900065784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20658755057564795, "3000–6000 Hz_10ms_envelope_correlation": 0.9711432771629515, "3000–6000 Hz_10ms_modulation_mae_db": 1.2155049359255417, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8567630457478528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11232056989273918, "3000–6000 Hz_40ms_envelope_correlation": 0.9918578307537841, "3000–6000 Hz_40ms_modulation_mae_db": 1.7085476529424308, "mr_stft_db": 5.768415750584391, "band_mae_db": 0.32957351796738305, "rms_gain_error_db": -0.09003687714356286, "output_peak": 0.1299100516144924}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2523050735486843, "全帯域_10ms_rms_envelope_nrmse": 0.10211196033823648, "全帯域_10ms_envelope_correlation": 0.9898210016931073, "全帯域_10ms_modulation_mae_db": 0.3008684562618951, "全帯域_40ms_log_rms_mae_db": 0.7395110115515195, "全帯域_40ms_rms_envelope_nrmse": 0.06004936699857849, "全帯域_40ms_envelope_correlation": 0.9961355428698414, "全帯域_40ms_modulation_mae_db": 0.2273431733143819, "300–1000 Hz_10ms_log_rms_mae_db": 1.6394562015838696, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11916959708895081, "300–1000 Hz_10ms_envelope_correlation": 0.9888450033142593, "300–1000 Hz_10ms_modulation_mae_db": 0.2966099120703182, "300–1000 Hz_40ms_log_rms_mae_db": 0.9461262162228754, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06702518739882882, "300–1000 Hz_40ms_envelope_correlation": 0.9962266888258456, "300–1000 Hz_40ms_modulation_mae_db": 0.6615396050841911, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7055397652336641, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2357988776901651, "1000–3000 Hz_10ms_envelope_correlation": 0.962593192394392, "1000–3000 Hz_10ms_modulation_mae_db": 0.37388674012369716, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1460664440463135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14678861607867105, "1000–3000 Hz_40ms_envelope_correlation": 0.9846428777284395, "1000–3000 Hz_40ms_modulation_mae_db": 0.27323999581841024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4584132410775377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2437646494910712, "3000–6000 Hz_10ms_envelope_correlation": 0.9588966904018744, "3000–6000 Hz_10ms_modulation_mae_db": 0.7739504502924768, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8302091195240646, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1466873724591146, "3000–6000 Hz_40ms_envelope_correlation": 0.9839026289195982, "3000–6000 Hz_40ms_modulation_mae_db": 0.9914366566052163, "mr_stft_db": 5.78383476680699, "band_mae_db": 0.1918310247835394, "rms_gain_error_db": 0.009068260962228866, "output_peak": 0.1416313903748135}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2305034476691108, "全帯域_10ms_rms_envelope_nrmse": 0.1003220196277097, "全帯域_10ms_envelope_correlation": 0.9902924882202903, "全帯域_10ms_modulation_mae_db": 0.41097899955052547, "全帯域_40ms_log_rms_mae_db": 0.6856296048025916, "全帯域_40ms_rms_envelope_nrmse": 0.054503414003059324, "全帯域_40ms_envelope_correlation": 0.9971215747859608, "全帯域_40ms_modulation_mae_db": 0.7474093357547074, "300–1000 Hz_10ms_log_rms_mae_db": 1.6710040426330663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1291375202903779, "300–1000 Hz_10ms_envelope_correlation": 0.9867421900352651, "300–1000 Hz_10ms_modulation_mae_db": 0.1725560898228017, "300–1000 Hz_40ms_log_rms_mae_db": 1.0224803382617904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07477899650478784, "300–1000 Hz_40ms_envelope_correlation": 0.9952992612322975, "300–1000 Hz_40ms_modulation_mae_db": 0.5154272356366993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5926717856086379, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2156161883112844, "1000–3000 Hz_10ms_envelope_correlation": 0.9678634742884049, "1000–3000 Hz_10ms_modulation_mae_db": 0.6418254587890684, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0447162465788886, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12308501791201405, "1000–3000 Hz_40ms_envelope_correlation": 0.9896176395806066, "1000–3000 Hz_40ms_modulation_mae_db": 0.5614420710194652, "3000–6000 Hz_10ms_log_rms_mae_db": 1.479018667798228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23218067824816888, "3000–6000 Hz_10ms_envelope_correlation": 0.9627236796500444, "3000–6000 Hz_10ms_modulation_mae_db": 1.0711539190744779, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9012548574657718, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1371528250966404, "3000–6000 Hz_40ms_envelope_correlation": 0.9860977131646103, "3000–6000 Hz_40ms_modulation_mae_db": 1.7640915104217836, "mr_stft_db": 5.7598757966960035, "band_mae_db": 0.27737944433151895, "rms_gain_error_db": -0.16918411291995836, "output_peak": 0.16047905871141036}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2366839160922425, "全帯域_10ms_rms_envelope_nrmse": 0.1006272620772431, "全帯域_10ms_envelope_correlation": 0.9900889360028318, "全帯域_10ms_modulation_mae_db": 0.34167380285294086, "全帯域_40ms_log_rms_mae_db": 0.6886364316704212, "全帯域_40ms_rms_envelope_nrmse": 0.054994214906717694, "全帯域_40ms_envelope_correlation": 0.9968564713719439, "全帯域_40ms_modulation_mae_db": 0.3396605703489828, "300–1000 Hz_10ms_log_rms_mae_db": 1.6653823396911913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13354451102135748, "300–1000 Hz_10ms_envelope_correlation": 0.986001931139032, "300–1000 Hz_10ms_modulation_mae_db": 0.6772355941989573, "300–1000 Hz_40ms_log_rms_mae_db": 1.0183024093791613, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07267940726403171, "300–1000 Hz_40ms_envelope_correlation": 0.995563713026853, "300–1000 Hz_40ms_modulation_mae_db": 0.2038992797547511, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7018988506543136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2423067928581093, "1000–3000 Hz_10ms_envelope_correlation": 0.9602990373400363, "1000–3000 Hz_10ms_modulation_mae_db": 0.8752423952860621, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1079333319472011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1406388416894306, "1000–3000 Hz_40ms_envelope_correlation": 0.9886412181367064, "1000–3000 Hz_40ms_modulation_mae_db": 1.0360073777167114, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4697831698739785, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22801345636217313, "3000–6000 Hz_10ms_envelope_correlation": 0.964897594675106, "3000–6000 Hz_10ms_modulation_mae_db": 1.5581021679093692, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8837251373554409, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1323228199219285, "3000–6000 Hz_40ms_envelope_correlation": 0.988897151402312, "3000–6000 Hz_40ms_modulation_mae_db": 2.1588971248754425, "mr_stft_db": 5.761481857873378, "band_mae_db": 0.32180267445500155, "rms_gain_error_db": -0.11397650086014606, "output_peak": 0.14850937289953914}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.262945996893493, "全帯域_10ms_rms_envelope_nrmse": 0.11187334386673538, "全帯域_10ms_envelope_correlation": 0.9877417914582578, "全帯域_10ms_modulation_mae_db": 0.42497774708879543, "全帯域_40ms_log_rms_mae_db": 0.7100560279998964, "全帯域_40ms_rms_envelope_nrmse": 0.06265906168256437, "全帯域_40ms_envelope_correlation": 0.9957642009035412, "全帯域_40ms_modulation_mae_db": 0.46388710544520473, "300–1000 Hz_10ms_log_rms_mae_db": 1.6154443820209479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13934564159212082, "300–1000 Hz_10ms_envelope_correlation": 0.9845466879133001, "300–1000 Hz_10ms_modulation_mae_db": 0.2272994288853746, "300–1000 Hz_40ms_log_rms_mae_db": 0.969392712646998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07172322335638948, "300–1000 Hz_40ms_envelope_correlation": 0.9956070007607865, "300–1000 Hz_40ms_modulation_mae_db": 0.8828189721794546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6413819640161846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26493036234444145, "1000–3000 Hz_10ms_envelope_correlation": 0.9572869369905889, "1000–3000 Hz_10ms_modulation_mae_db": 1.1110315212353774, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0176917656646347, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16521358432111385, "1000–3000 Hz_40ms_envelope_correlation": 0.9838333653633354, "1000–3000 Hz_40ms_modulation_mae_db": 1.0357618650735172, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4127621730280846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21671959692623588, "3000–6000 Hz_10ms_envelope_correlation": 0.9677566407848577, "3000–6000 Hz_10ms_modulation_mae_db": 0.721878326721625, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8262862701605315, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12476326033072792, "3000–6000 Hz_40ms_envelope_correlation": 0.9889814392108571, "3000–6000 Hz_40ms_modulation_mae_db": 0.8154426857597891, "mr_stft_db": 5.766821678885836, "band_mae_db": 0.19107799791385371, "rms_gain_error_db": 0.0051003101469970355, "output_peak": 0.14860170895044247}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2914840095822695, "全帯域_10ms_rms_envelope_nrmse": 0.10485569677765973, "全帯域_10ms_envelope_correlation": 0.9891866713349301, "全帯域_10ms_modulation_mae_db": 0.3320543164223253, "全帯域_40ms_log_rms_mae_db": 0.7563376699649401, "全帯域_40ms_rms_envelope_nrmse": 0.05446744260581643, "全帯域_40ms_envelope_correlation": 0.9968347224859103, "全帯域_40ms_modulation_mae_db": 0.5196230667711402, "300–1000 Hz_10ms_log_rms_mae_db": 1.6943093409129513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12610771402623183, "300–1000 Hz_10ms_envelope_correlation": 0.9875219093365218, "300–1000 Hz_10ms_modulation_mae_db": 0.23614184640465474, "300–1000 Hz_40ms_log_rms_mae_db": 1.0424220291676147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07246969350282872, "300–1000 Hz_40ms_envelope_correlation": 0.9955945726070068, "300–1000 Hz_40ms_modulation_mae_db": 0.659234019180532, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7131636175763107, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24825357927424221, "1000–3000 Hz_10ms_envelope_correlation": 0.9578806422044174, "1000–3000 Hz_10ms_modulation_mae_db": 1.3099774770848789, "1000–3000 Hz_40ms_log_rms_mae_db": 1.102796382409632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14748667628649373, "1000–3000 Hz_40ms_envelope_correlation": 0.9864621803414165, "1000–3000 Hz_40ms_modulation_mae_db": 1.2956900669047011, "3000–6000 Hz_10ms_log_rms_mae_db": 1.45547457307079, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19921042810921966, "3000–6000 Hz_10ms_envelope_correlation": 0.9728835689975965, "3000–6000 Hz_10ms_modulation_mae_db": 1.105071500576768, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8396553886622842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10156791556060542, "3000–6000 Hz_40ms_envelope_correlation": 0.9923155329602371, "3000–6000 Hz_40ms_modulation_mae_db": 1.38145356876934, "mr_stft_db": 5.76316004781313, "band_mae_db": 0.27543018537302866, "rms_gain_error_db": -0.0777429267246828, "output_peak": 0.16135023498153106}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.26390653260042, "全帯域_10ms_rms_envelope_nrmse": 0.10120609955918469, "全帯域_10ms_envelope_correlation": 0.990089112870512, "全帯域_10ms_modulation_mae_db": 0.48976069504689806, "全帯域_40ms_log_rms_mae_db": 0.6855507806253902, "全帯域_40ms_rms_envelope_nrmse": 0.05385273252000444, "全帯域_40ms_envelope_correlation": 0.9971633177475892, "全帯域_40ms_modulation_mae_db": 0.7072943015951401, "300–1000 Hz_10ms_log_rms_mae_db": 1.606901945667377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1357884713778966, "300–1000 Hz_10ms_envelope_correlation": 0.985300363677158, "300–1000 Hz_10ms_modulation_mae_db": 0.32373698360340547, "300–1000 Hz_40ms_log_rms_mae_db": 0.8850978287669069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06881699083668143, "300–1000 Hz_40ms_envelope_correlation": 0.9960025393106718, "300–1000 Hz_40ms_modulation_mae_db": 0.23801930564421095, "1000–3000 Hz_10ms_log_rms_mae_db": 1.683756806846184, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2357623624758007, "1000–3000 Hz_10ms_envelope_correlation": 0.9625788858592293, "1000–3000 Hz_10ms_modulation_mae_db": 1.2801473631737468, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0734245267909295, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1531481850851932, "1000–3000 Hz_40ms_envelope_correlation": 0.9855566563137015, "1000–3000 Hz_40ms_modulation_mae_db": 1.5764874599397078, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4704918754059404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21558954782899242, "3000–6000 Hz_10ms_envelope_correlation": 0.9706358550784754, "3000–6000 Hz_10ms_modulation_mae_db": 0.648826697175523, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8249137227269453, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13750108914102516, "3000–6000 Hz_40ms_envelope_correlation": 0.9875936114178352, "3000–6000 Hz_40ms_modulation_mae_db": 0.6591837218817256, "mr_stft_db": 5.767122162164742, "band_mae_db": 0.2334102660549676, "rms_gain_error_db": -0.15848791002458623, "output_peak": 0.14337465503053193}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87280, "output_samples": 87280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.862672279694852, "spectral_convergence": 0.34457158701952223, "spectral_convergence_status": "ok", "amplitude_floor": 4.3919411473735467e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.636430405635509, "spectral_convergence": 0.3373620017965635, "spectral_convergence_status": "ok", "amplitude_floor": 3.1709236310047874e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.811157632405879, "spectral_convergence": 0.3528161094506674, "spectral_convergence_status": "ok", "amplitude_floor": 2.989229099554426e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001235363438035874, "output_power": 0.00012367489188201683, "delta_db": 0.004867956108317707, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 7.224896060742283e-05, "output_power": 7.289693119590607e-05, "delta_db": 0.038776422400103046, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.329659782840476e-05, "output_power": 4.182493776390986e-05, "delta_db": -0.15018468187943995, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.747692740945149e-05, "output_power": 3.63773557268908e-05, "delta_db": -1.1565147774438695, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.909876260278539e-06, "output_power": 1.0425037531047437e-05, "delta_db": 0.6820595451856121, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1191996093186054e-05, "output_power": 2.057032593035302e-05, "delta_db": -0.1293069221668086, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3602850508640252, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15200035496641562, "pitch": null, "output_rms_dbfs": -35.14605865280731, "output_peak": 0.13528658388145823, "reference_rms_dbfs": -34.9940582978409, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.770086772578747}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 5.579882181126775e-05, "log_rms_mae_db": 1.1848001381272621, "rms_envelope_nrmse": 0.10572527460577458, "envelope_correlation": 0.9891259999120482, "rms_gain_error_db": -0.15200054113332728, "reference_mean_power": 0.0003169510345489487, "output_mean_power": 0.00030604981376553984}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8938855835285014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.978895728357633e-05, "output_power": 5.513409663630701e-05, "reference_active": true, "delta_db": -0.3520071629180954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1930505918996815e-05, "output_power": 2.1416377754672567e-05, "reference_active": true, "delta_db": -0.10302632026735052}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0863861439295844e-06, "output_power": 3.3824459073204633e-06, "reference_active": true, "delta_db": 0.3978059943449417}], "active_band_count": 3, "active_band_mae_db": 0.2842798258434625}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 5.258028604719465e-05, "log_rms_mae_db": 0.6651192485258486, "rms_envelope_nrmse": 0.06344280735050624, "envelope_correlation": 0.9958959509078663, "rms_gain_error_db": -0.15199762610723558, "reference_mean_power": 0.00031870366172679027, "output_mean_power": 0.0003077423676056165}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 1.711395707762808e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.23137186055323e-05, "output_power": 4.8810906819868174e-05, "reference_active": true, "delta_db": -0.30098715841128315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1932036565608145e-05, "output_power": 1.1353082971583354e-05, "reference_active": true, "delta_db": -0.21600763768955866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.569030812932547e-07, "output_power": 2.3652466444921056e-07, "reference_active": true, "delta_db": -0.358928781873067}], "active_band_count": 3, "active_band_mae_db": 0.29197452599130297}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 4.2439491809696055e-05, "log_rms_mae_db": 1.6410407502824547, "rms_envelope_nrmse": 0.12900789485138298, "envelope_correlation": 0.9867427957570861, "rms_gain_error_db": -0.08680967992089073, "reference_mean_power": 0.000101477720422738, "output_mean_power": 9.946945416062483e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.361033224981284e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.73677738976729e-05, "output_power": 2.7348143267836244e-05, "reference_active": true, "delta_db": -0.0031162685455317874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1422451661963211e-05, "output_power": 1.1014786629295217e-05, "reference_active": true, "delta_db": -0.15783240144198116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.10199254010919e-06, "output_power": 1.3286099645698594e-06, "reference_active": true, "delta_db": 0.812188507367495}], "active_band_count": 3, "active_band_mae_db": 0.324379059118336}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.7895186903062846e-05, "log_rms_mae_db": 0.9886736669401909, "rms_envelope_nrmse": 0.06715524764165481, "envelope_correlation": 0.9961990467839831, "rms_gain_error_db": -0.08680982972151566, "reference_mean_power": 0.00010203935243697765, "output_mean_power": 0.00010001996790459898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 7.765913024651851e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5071474654741866e-05, "output_power": 2.508041714317599e-05, "reference_active": true, "delta_db": 0.0015487644766149803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067872704197783e-06, "output_power": 5.842226677484815e-06, "reference_active": true, "delta_db": -0.16458057773902002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0845180908068795e-07, "output_power": 8.328857382469407e-08, "reference_active": true, "delta_db": -1.1465137526770137}], "active_band_count": 3, "active_band_mae_db": 0.4375476982975495}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 3.774086107813168e-05, "log_rms_mae_db": 1.6720717682929966, "rms_envelope_nrmse": 0.23166002648503645, "envelope_correlation": 0.963961795946677, "rms_gain_error_db": -0.6308283874540717, "reference_mean_power": 4.068228265632881e-05, "output_mean_power": 3.518215796071874e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.809144227115543e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.649297963098168e-06, "output_power": 6.214256905095991e-06, "reference_active": true, "delta_db": -1.435916549853554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.584218334576925e-06, "output_power": 2.3532280960960646e-06, "reference_active": true, "delta_db": -0.4066517844518002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3501889520111653e-06, "output_power": 1.8145739094460876e-06, "reference_active": true, "delta_db": 1.2838011212308302}], "active_band_count": 3, "active_band_mae_db": 1.0421231518453948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.332899304538301e-05, "log_rms_mae_db": 1.0825918003690835, "rms_envelope_nrmse": 0.14841403905696093, "envelope_correlation": 0.9866761720978561, "rms_gain_error_db": -0.630829513404021, "reference_mean_power": 4.0907280378011884e-05, "output_mean_power": 3.537672748190216e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 2.6173037083412648e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.99444690144364e-06, "output_power": 5.97929383259004e-06, "reference_active": true, "delta_db": -1.2613852640027283}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1581434874555947e-06, "output_power": 1.0872009595477075e-06, "reference_active": true, "delta_db": -0.2745254233870349}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.731926648355553e-08, "output_power": 1.164715920262489e-07, "reference_active": true, "delta_db": 1.2510993298715354}], "active_band_count": 3, "active_band_mae_db": 0.9290033390870995}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 1.8891386922351935e-05, "log_rms_mae_db": 1.5314847945407017, "rms_envelope_nrmse": 0.22203028388991514, "envelope_correlation": 0.9659700876609312, "rms_gain_error_db": -0.25609945197676737, "reference_mean_power": 6.700122358829966e-06, "output_mean_power": 6.31644601569229e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 5.001365064934297e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.184828883572986e-06, "output_power": 2.272072008065644e-06, "reference_active": true, "delta_db": 0.17004662646374574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0698658461156927e-06, "output_power": 1.0153290434689933e-06, "reference_active": true, "delta_db": -0.227225141839387}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.241288004170751e-07, "output_power": 3.2265668818280624e-07, "reference_active": true, "delta_db": -2.1069735265741425}], "active_band_count": 3, "active_band_mae_db": 0.8347484316257584}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 1.2324616345992443e-05, "log_rms_mae_db": 0.8837099084444816, "rms_envelope_nrmse": 0.12961731451361416, "envelope_correlation": 0.9876612939280931, "rms_gain_error_db": -0.2560968783645006, "reference_mean_power": 6.7368909993733856e-06, "output_mean_power": 6.351112897503871e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 4.500371991965458e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.395916806498352e-06, "output_power": 2.444205801054113e-06, "reference_active": true, "delta_db": 0.08666036557714739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.187368503320664e-07, "output_power": 5.189727008816125e-07, "reference_active": true, "delta_db": -0.7636146862862271}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.878047962583857e-08, "output_power": 2.5579665828454928e-08, "reference_active": true, "delta_db": -2.8035119954883463}], "active_band_count": 3, "active_band_mae_db": 1.2179290157839069}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8338270558567218, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d565db4c40452c4a0daf68f68bf227f3c9b591a3f5017a03cdd666edc5bfe8c5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.962166131430393, "standard_deviation": 0.003565293589284719, "minimum": 0.9572869369905889, "maximum": 0.9678634742884049, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6758135998579946, "standard_deviation": 0.0406074829581704, "minimum": 1.5926717856086379, "maximum": 1.7131636175763107, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.927596982792253, "standard_deviation": 0.32252775370247716, "minimum": 0.37388674012369716, "maximum": 1.3099774770848789, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23839352793858476, "standard_deviation": 0.014270870194661606, "minimum": 0.2156161883112844, "maximum": 0.26493036234444145, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9865594347859716, "standard_deviation": 0.0019293176596308329, "minimum": 0.9838333653633354, "maximum": 0.9896176395806066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0844377902268751, "standard_deviation": 0.03981337758025341, "minimum": 1.0176917656646347, "maximum": 1.1460664440463135, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9358699891126516, "standard_deviation": 0.4077279410032768, "minimum": 0.27323999581841024, "maximum": 1.5764874599397078, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14744105455678655, "standard_deviation": 0.012205190569673076, "minimum": 0.12308501791201405, "maximum": 0.16521358432111385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9668634243014796, "standard_deviation": 0.004703451978690249, "minimum": 0.9588966904018744, "maximum": 0.9728835689975965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4636889481002298, "standard_deviation": 0.0350560208530618, "minimum": 1.4127621730280846, "maximum": 1.5314847945407017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9911545536626925, "standard_deviation": 0.3053050478821391, "minimum": 0.648826697175523, "maximum": 1.5581021679093692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22051202392892802, "standard_deviation": 0.014251220033850036, "minimum": 0.19921042810921966, "maximum": 0.2437646494910712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9884134002196658, "standard_deviation": 0.002792710075063428, "minimum": 0.9839026289195982, "maximum": 0.9923155329602371, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8558146812609215, "standard_deviation": 0.03015770476614786, "minimum": 0.8249137227269453, "maximum": 0.9012548574657718, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3371227421299543, "standard_deviation": 0.5158896009129884, "minimum": 0.6591837218817256, "maximum": 2.1588971248754425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1277416458645494, "standard_deviation": 0.014633656876072687, "minimum": 0.10156791556060542, "maximum": 0.1466873724591146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.986626627030009, "standard_deviation": 0.0013466816976842944, "minimum": 0.9845466879133001, "maximum": 0.9888450033142593, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6459671989713254, "standard_deviation": 0.02928591434297352, "minimum": 1.606901945667377, "maximum": 1.6943093409129513, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3022051245043057, "standard_deviation": 0.16419995377667862, "minimum": 0.15968208193059782, "maximum": 0.6772355941989573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12980170660386192, "standard_deviation": 0.0063343867366841555, "minimum": 0.11916959708895081, "maximum": 0.13934564159212082, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.995751730182228, "standard_deviation": 0.0003438302063245712, "minimum": 0.9952992612322975, "maximum": 0.9962266888258456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9786033852779775, "standard_deviation": 0.05078574238476198, "minimum": 0.8850978287669069, "maximum": 1.0424220291676147, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48188199939314047, "standard_deviation": 0.24351028747145506, "minimum": 0.2038992797547511, "maximum": 0.8828189721794546, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07087525366649944, "standard_deviation": 0.002849509744882088, "minimum": 0.06702518739882882, "maximum": 0.07477899650478784, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2725987702179148, "standard_deviation": 0.06334903936776383, "minimum": 0.19107799791385371, "maximum": 0.3602850508640252, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.767599854175403, "standard_deviation": 0.007445606642461117, "minimum": 5.7598757966960035, "maximum": 5.78383476680699, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.1461428820555274, "standard_deviation": 0.011070012807822122, "minimum": 0.1299100516144924, "maximum": 0.16135023498153106, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09340751394126576, "standard_deviation": 0.06999548095940686, "minimum": -0.16918411291995836, "maximum": 0.009068260962228866, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9895844782342305, "standard_deviation": 0.000877158022787152, "minimum": 0.9877417914582578, "maximum": 0.9903298243818665, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2536316713069209, "standard_deviation": 0.03774702807226093, "minimum": 1.1848001381272621, "maximum": 1.3064242559418846, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3385275645503554, "standard_deviation": 0.11096300840994519, "minimum": 0.12362667333600004, "maximum": 0.48976069504689806, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10324781370316356, "standard_deviation": 0.0041387549866535155, "minimum": 0.09926085277276483, "maximum": 0.11187334386673538, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.996628562464863, "standard_deviation": 0.0006029638578114903, "minimum": 0.9957642009035412, "maximum": 0.99725671864625, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.705822474478146, "standard_deviation": 0.030644900757373878, "minimum": 0.6651192485258486, "maximum": 0.7563376699649401, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46254549481045043, "standard_deviation": 0.18797779267175507, "minimum": 0.2273431733143819, "maximum": 0.7474093357547074, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0568874390054846, "standard_deviation": 0.004531771006667241, "minimum": 0.05113047197662974, "maximum": 0.06344280735050624, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32865-0017", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.995405912399292, "predicted_mos": 1.995405912399292, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "ba55a46d8d12583febc3cb1da24ec1423685f9a5d6ad24221941c564859381bb", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.066608800087124, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1975886021578244, "全帯域_10ms_rms_envelope_nrmse": 0.10010556195000675, "全帯域_10ms_envelope_correlation": 0.9901627349530129, "全帯域_10ms_modulation_mae_db": 0.1727390104167895, "全帯域_40ms_log_rms_mae_db": 0.6731133546084827, "全帯域_40ms_rms_envelope_nrmse": 0.05823794488769202, "全帯域_40ms_envelope_correlation": 0.9964035331417344, "全帯域_40ms_modulation_mae_db": 0.2866312700844738, "300–1000 Hz_10ms_log_rms_mae_db": 2.0568590231686534, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12891517376616946, "300–1000 Hz_10ms_envelope_correlation": 0.9869563835247501, "300–1000 Hz_10ms_modulation_mae_db": 0.6243739992347448, "300–1000 Hz_40ms_log_rms_mae_db": 1.4725362005435585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07108888755542073, "300–1000 Hz_40ms_envelope_correlation": 0.9958006207606441, "300–1000 Hz_40ms_modulation_mae_db": 0.4234111809399513, "1000–3000 Hz_10ms_log_rms_mae_db": 2.123262899334253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24133090063641044, "1000–3000 Hz_10ms_envelope_correlation": 0.9608500144328669, "1000–3000 Hz_10ms_modulation_mae_db": 1.792079973875466, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4946302166140133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1614543006786663, "1000–3000 Hz_40ms_envelope_correlation": 0.9838896132763347, "1000–3000 Hz_40ms_modulation_mae_db": 1.7944204574909453, "3000–6000 Hz_10ms_log_rms_mae_db": 2.034295210092159, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24734084810792087, "3000–6000 Hz_10ms_envelope_correlation": 0.9584515410929231, "3000–6000 Hz_10ms_modulation_mae_db": 1.3156076513558375, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3973669830904212, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14097337408527372, "3000–6000 Hz_40ms_envelope_correlation": 0.9852046438142157, "3000–6000 Hz_40ms_modulation_mae_db": 1.7079438911460565, "mr_stft_db": 6.117489339976626, "band_mae_db": 0.20284876489574002, "rms_gain_error_db": -0.09080159524125647, "output_peak": 0.13523628882272692}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2584965381195312, "全帯域_10ms_rms_envelope_nrmse": 0.10568919734136037, "全帯域_10ms_envelope_correlation": 0.9890125119903017, "全帯域_10ms_modulation_mae_db": 0.16916472447115768, "全帯域_40ms_log_rms_mae_db": 0.7303249643678528, "全帯域_40ms_rms_envelope_nrmse": 0.050927274698808825, "全帯域_40ms_envelope_correlation": 0.9971829908636536, "全帯域_40ms_modulation_mae_db": 0.334904622396862, "300–1000 Hz_10ms_log_rms_mae_db": 2.0120851361059575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1307164035243594, "300–1000 Hz_10ms_envelope_correlation": 0.9866648793956462, "300–1000 Hz_10ms_modulation_mae_db": 0.05217374763266031, "300–1000 Hz_40ms_log_rms_mae_db": 1.3675318716335474, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07331235222774557, "300–1000 Hz_40ms_envelope_correlation": 0.9955633107927906, "300–1000 Hz_40ms_modulation_mae_db": 0.26898317411411843, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2458053505054822, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25655741672744903, "1000–3000 Hz_10ms_envelope_correlation": 0.9539674718832901, "1000–3000 Hz_10ms_modulation_mae_db": 0.6732814837132851, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5642973030174614, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14115394116485044, "1000–3000 Hz_40ms_envelope_correlation": 0.9862632823886753, "1000–3000 Hz_40ms_modulation_mae_db": 0.6254831784178129, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0562527913211017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24644617445440828, "3000–6000 Hz_10ms_envelope_correlation": 0.9628332458384078, "3000–6000 Hz_10ms_modulation_mae_db": 1.0551867032798554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.461867819184835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14830733835141535, "3000–6000 Hz_40ms_envelope_correlation": 0.9866290485723777, "3000–6000 Hz_40ms_modulation_mae_db": 1.0781834520078233, "mr_stft_db": 6.099600061893565, "band_mae_db": 0.26664080586812655, "rms_gain_error_db": -0.027252580407785533, "output_peak": 0.1744540075508135}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2318949382208533, "全帯域_10ms_rms_envelope_nrmse": 0.10316034370397525, "全帯域_10ms_envelope_correlation": 0.9895436653402447, "全帯域_10ms_modulation_mae_db": 0.3477936651827578, "全帯域_40ms_log_rms_mae_db": 0.6166273195967783, "全帯域_40ms_rms_envelope_nrmse": 0.05104653001072472, "全帯域_40ms_envelope_correlation": 0.9972627177130956, "全帯域_40ms_modulation_mae_db": 0.286910859371119, "300–1000 Hz_10ms_log_rms_mae_db": 2.0125295942559123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13844553029069076, "300–1000 Hz_10ms_envelope_correlation": 0.9847569773721039, "300–1000 Hz_10ms_modulation_mae_db": 0.9156835869177646, "300–1000 Hz_40ms_log_rms_mae_db": 1.3926719738892228, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07356901803819121, "300–1000 Hz_40ms_envelope_correlation": 0.9954733430699545, "300–1000 Hz_40ms_modulation_mae_db": 0.4607301208520311, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1639147483689074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23622957663366392, "1000–3000 Hz_10ms_envelope_correlation": 0.9611141793862606, "1000–3000 Hz_10ms_modulation_mae_db": 0.9347440739158864, "1000–3000 Hz_40ms_log_rms_mae_db": 1.486510262333597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1318198209348538, "1000–3000 Hz_40ms_envelope_correlation": 0.9877836576701355, "1000–3000 Hz_40ms_modulation_mae_db": 0.9261517422057305, "3000–6000 Hz_10ms_log_rms_mae_db": 2.035318357892816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2289266739508595, "3000–6000 Hz_10ms_envelope_correlation": 0.9649369868374036, "3000–6000 Hz_10ms_modulation_mae_db": 1.2525019912327242, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4186771270048262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12895426998635182, "3000–6000 Hz_40ms_envelope_correlation": 0.9879353997684003, "3000–6000 Hz_40ms_modulation_mae_db": 1.2676765889496457, "mr_stft_db": 6.1050898373307625, "band_mae_db": 0.29173447880766806, "rms_gain_error_db": -0.09311575253604709, "output_peak": 0.15977551775081217}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.244459417701068, "全帯域_10ms_rms_envelope_nrmse": 0.08727322055310785, "全帯域_10ms_envelope_correlation": 0.9925414068945082, "全帯域_10ms_modulation_mae_db": 0.3767359613616906, "全帯域_40ms_log_rms_mae_db": 0.7708607196207845, "全帯域_40ms_rms_envelope_nrmse": 0.04484014895649624, "全帯域_40ms_envelope_correlation": 0.9978915662400482, "全帯域_40ms_modulation_mae_db": 0.5474700957431565, "300–1000 Hz_10ms_log_rms_mae_db": 2.0349265813926847, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12748552360258325, "300–1000 Hz_10ms_envelope_correlation": 0.9872451817743165, "300–1000 Hz_10ms_modulation_mae_db": 0.15995181138971884, "300–1000 Hz_40ms_log_rms_mae_db": 1.40534501833336, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06708463475761117, "300–1000 Hz_40ms_envelope_correlation": 0.9962817480848978, "300–1000 Hz_40ms_modulation_mae_db": 0.2935448804890561, "1000–3000 Hz_10ms_log_rms_mae_db": 2.197130338686686, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21544172528837108, "1000–3000 Hz_10ms_envelope_correlation": 0.9683968423057615, "1000–3000 Hz_10ms_modulation_mae_db": 0.9005428815916261, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5205965633753176, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12852467204035872, "1000–3000 Hz_40ms_envelope_correlation": 0.9895175461050659, "1000–3000 Hz_40ms_modulation_mae_db": 0.9778355026340911, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0405610006901234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23400748396093668, "3000–6000 Hz_10ms_envelope_correlation": 0.9638980357394246, "3000–6000 Hz_10ms_modulation_mae_db": 1.028918728670447, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3866227222017193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13421002012750047, "3000–6000 Hz_40ms_envelope_correlation": 0.9872133708659201, "3000–6000 Hz_40ms_modulation_mae_db": 1.255635994300436, "mr_stft_db": 6.1080849794113545, "band_mae_db": 0.24022973624427013, "rms_gain_error_db": -0.0794177149196254, "output_peak": 0.14998915851172914}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.177431270219609, "全帯域_10ms_rms_envelope_nrmse": 0.09730701363162628, "全帯域_10ms_envelope_correlation": 0.9906780658637843, "全帯域_10ms_modulation_mae_db": 0.20056877890496838, "全帯域_40ms_log_rms_mae_db": 0.6946466132563843, "全帯域_40ms_rms_envelope_nrmse": 0.051527709526195294, "全帯域_40ms_envelope_correlation": 0.9971162877567535, "全帯域_40ms_modulation_mae_db": 0.296809268825506, "300–1000 Hz_10ms_log_rms_mae_db": 2.1975132060546496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1356464045516847, "300–1000 Hz_10ms_envelope_correlation": 0.9855718978906693, "300–1000 Hz_10ms_modulation_mae_db": 0.2931327856167106, "300–1000 Hz_40ms_log_rms_mae_db": 1.5213200825648534, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07482936813399282, "300–1000 Hz_40ms_envelope_correlation": 0.995466426446908, "300–1000 Hz_40ms_modulation_mae_db": 0.4930516482960951, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1544183251073328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22595001343370377, "1000–3000 Hz_10ms_envelope_correlation": 0.964775698817445, "1000–3000 Hz_10ms_modulation_mae_db": 0.33295669752019824, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5939265870355637, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14723609862281353, "1000–3000 Hz_40ms_envelope_correlation": 0.984065757367498, "1000–3000 Hz_40ms_modulation_mae_db": 0.4310620747894782, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9992825910257475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22901815404578796, "3000–6000 Hz_10ms_envelope_correlation": 0.9649660023180464, "3000–6000 Hz_10ms_modulation_mae_db": 1.2821030312531978, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3617281212007932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12108775740030056, "3000–6000 Hz_40ms_envelope_correlation": 0.9893967936699479, "3000–6000 Hz_40ms_modulation_mae_db": 1.660456402042276, "mr_stft_db": 6.115959149773594, "band_mae_db": 0.21902065076930674, "rms_gain_error_db": -0.029214023180842923, "output_peak": 0.16391032446096335}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1388219392504835, "全帯域_10ms_rms_envelope_nrmse": 0.09613525099172496, "全帯域_10ms_envelope_correlation": 0.9910383435589596, "全帯域_10ms_modulation_mae_db": 0.27932621882274977, "全帯域_40ms_log_rms_mae_db": 0.6546279649260773, "全帯域_40ms_rms_envelope_nrmse": 0.04917953578224488, "全帯域_40ms_envelope_correlation": 0.9976243742381246, "全帯域_40ms_modulation_mae_db": 0.44071593557051636, "300–1000 Hz_10ms_log_rms_mae_db": 2.0305141639417044, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12730497140913857, "300–1000 Hz_10ms_envelope_correlation": 0.9871826168491591, "300–1000 Hz_10ms_modulation_mae_db": 0.7640066266594929, "300–1000 Hz_40ms_log_rms_mae_db": 1.4127323245881542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07131858516465438, "300–1000 Hz_40ms_envelope_correlation": 0.9958681159660606, "300–1000 Hz_40ms_modulation_mae_db": 0.7461897866497492, "1000–3000 Hz_10ms_log_rms_mae_db": 2.155698501000464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2458023532778704, "1000–3000 Hz_10ms_envelope_correlation": 0.9578560233223098, "1000–3000 Hz_10ms_modulation_mae_db": 1.2996914313592336, "1000–3000 Hz_40ms_log_rms_mae_db": 1.532987913051706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1426921233460533, "1000–3000 Hz_40ms_envelope_correlation": 0.985789989659511, "1000–3000 Hz_40ms_modulation_mae_db": 1.362104200699206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9572130737612021, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20670720577270216, "3000–6000 Hz_10ms_envelope_correlation": 0.9709489098527008, "3000–6000 Hz_10ms_modulation_mae_db": 1.055236776219559, "3000–6000 Hz_40ms_log_rms_mae_db": 1.312673997185054, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11069822633344757, "3000–6000 Hz_40ms_envelope_correlation": 0.9910968522034458, "3000–6000 Hz_40ms_modulation_mae_db": 1.3110289383430187, "mr_stft_db": 6.095681933497899, "band_mae_db": 0.20741666960519967, "rms_gain_error_db": -0.14271639554096846, "output_peak": 0.15058311646228953}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1897876586982898, "全帯域_10ms_rms_envelope_nrmse": 0.09288857626836024, "全帯域_10ms_envelope_correlation": 0.9915653404216348, "全帯域_10ms_modulation_mae_db": 0.21179019776828914, "全帯域_40ms_log_rms_mae_db": 0.7326074839631386, "全帯域_40ms_rms_envelope_nrmse": 0.04833649971886581, "全帯域_40ms_envelope_correlation": 0.9975850044871705, "全帯域_40ms_modulation_mae_db": 0.40976490731178905, "300–1000 Hz_10ms_log_rms_mae_db": 2.0314293296844173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.124940110537982, "300–1000 Hz_10ms_envelope_correlation": 0.9876951235381419, "300–1000 Hz_10ms_modulation_mae_db": 0.24726232036444354, "300–1000 Hz_40ms_log_rms_mae_db": 1.448906412241239, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06831522269047464, "300–1000 Hz_40ms_envelope_correlation": 0.996212480504476, "300–1000 Hz_40ms_modulation_mae_db": 0.3326554935025337, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0449287605377187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22890925128220774, "1000–3000 Hz_10ms_envelope_correlation": 0.9637420467803409, "1000–3000 Hz_10ms_modulation_mae_db": 1.077819289029262, "1000–3000 Hz_40ms_log_rms_mae_db": 1.50235436344098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14951584770298365, "1000–3000 Hz_40ms_envelope_correlation": 0.9844150026225668, "1000–3000 Hz_40ms_modulation_mae_db": 1.1715524454305373, "3000–6000 Hz_10ms_log_rms_mae_db": 2.073518288157028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21783398943463245, "3000–6000 Hz_10ms_envelope_correlation": 0.9680810202638339, "3000–6000 Hz_10ms_modulation_mae_db": 0.7127502190293032, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4705453048292363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11701440991439106, "3000–6000 Hz_40ms_envelope_correlation": 0.9900171165318552, "3000–6000 Hz_40ms_modulation_mae_db": 0.8394340318525834, "mr_stft_db": 6.125638563346786, "band_mae_db": 0.23295434961203676, "rms_gain_error_db": -0.10365054536535487, "output_peak": 0.14004925378363772}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1917841893563232, "全帯域_10ms_rms_envelope_nrmse": 0.09338165792858585, "全帯域_10ms_envelope_correlation": 0.9915231623014299, "全帯域_10ms_modulation_mae_db": 0.3548635119365198, "全帯域_40ms_log_rms_mae_db": 0.6674657281714705, "全帯域_40ms_rms_envelope_nrmse": 0.04826814231563987, "全帯域_40ms_envelope_correlation": 0.9976676521505548, "全帯域_40ms_modulation_mae_db": 0.18173748271609333, "300–1000 Hz_10ms_log_rms_mae_db": 2.177528240656392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13302105750448398, "300–1000 Hz_10ms_envelope_correlation": 0.9862188151020708, "300–1000 Hz_10ms_modulation_mae_db": 0.43475150373973076, "300–1000 Hz_40ms_log_rms_mae_db": 1.4917395858663736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07245306877543382, "300–1000 Hz_40ms_envelope_correlation": 0.9960707126539815, "300–1000 Hz_40ms_modulation_mae_db": 0.620531594461133, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1548624607997247, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23728196984835573, "1000–3000 Hz_10ms_envelope_correlation": 0.9609317461280126, "1000–3000 Hz_10ms_modulation_mae_db": 0.29730652316479106, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5190675726439031, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1328039330655008, "1000–3000 Hz_40ms_envelope_correlation": 0.9870909282943937, "1000–3000 Hz_40ms_modulation_mae_db": 0.4055141192431205, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0720800877113086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2273145793703988, "3000–6000 Hz_10ms_envelope_correlation": 0.9646225098400989, "3000–6000 Hz_10ms_modulation_mae_db": 1.2638552528057094, "3000–6000 Hz_40ms_log_rms_mae_db": 1.425984777181776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12329315131647268, "3000–6000 Hz_40ms_envelope_correlation": 0.9888614590659022, "3000–6000 Hz_40ms_modulation_mae_db": 1.5738751025493223, "mr_stft_db": 6.101141666165277, "band_mae_db": 0.19783906818500774, "rms_gain_error_db": -0.12659772864097185, "output_peak": 0.14433144815884785}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87280, "output_samples": 87280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.232629774141657, "spectral_convergence": 0.3515764638018016, "spectral_convergence_status": "ok", "amplitude_floor": 4.3919411473735467e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.057710137232545, "spectral_convergence": 0.347814547893665, "spectral_convergence_status": "ok", "amplitude_floor": 3.1709236310047874e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.062128108555678, "spectral_convergence": 0.34776668767080604, "spectral_convergence_status": "ok", "amplitude_floor": 2.989229099554426e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001235363438035874, "output_power": 0.00012353205993545624, "delta_db": -0.0001506028502709314, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 7.224896060742283e-05, "output_power": 7.460026455737251e-05, "delta_db": 0.13908764029488918, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.329659782840476e-05, "output_power": 4.359073554671096e-05, "delta_db": 0.029404257445014875, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.747692740945149e-05, "output_power": 3.894049891165725e-05, "delta_db": -0.8608109324870263, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.909876260278539e-06, "output_power": 8.982918228224121e-06, "delta_db": 0.03545773675056869, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1191996093186054e-05, "output_power": 2.0462265871693363e-05, "delta_db": -0.15218141954667022, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20284876489574002, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09080159524125647, "pitch": null, "output_rms_dbfs": -35.084859893082154, "output_peak": 0.13523628882272692, "reference_rms_dbfs": -34.9940582978409, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.117489339976626}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 5.579882181126775e-05, "log_rms_mae_db": 1.1975886021578244, "rms_envelope_nrmse": 0.10010556195000675, "envelope_correlation": 0.9901627349530129, "rms_gain_error_db": -0.09080187259459344, "reference_mean_power": 0.0003169510345489487, "output_mean_power": 0.00031039304856639934}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8938855835285014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.978895728357633e-05, "output_power": 5.717105129746184e-05, "reference_active": true, "delta_db": -0.19444801042759519}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1930505918996815e-05, "output_power": 2.0668437770177553e-05, "reference_active": true, "delta_db": -0.2574099902048507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0863861439295844e-06, "output_power": 3.133907427076781e-06, "reference_active": true, "delta_db": 0.0663590306179225}], "active_band_count": 3, "active_band_mae_db": 0.1727390104167895}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 5.258028604719465e-05, "log_rms_mae_db": 0.6731133546084827, "rms_envelope_nrmse": 0.05823794488769202, "envelope_correlation": 0.9964035331417344, "rms_gain_error_db": -0.09079797318421555, "reference_mean_power": 0.00031870366172679027, "output_mean_power": 0.00031210969263458605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 1.711395707762808e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.23137186055323e-05, "output_power": 5.0261172429077424e-05, "reference_active": true, "delta_db": -0.17382976944699186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1932036565608145e-05, "output_power": 1.1075133706545381e-05, "reference_active": true, "delta_db": -0.32365597558142944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.569030812932547e-07, "output_power": 2.3633525223969682e-07, "reference_active": true, "delta_db": -0.3624080652250001}], "active_band_count": 3, "active_band_mae_db": 0.2866312700844738}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 4.2439491809696055e-05, "log_rms_mae_db": 2.0568590231686534, "rms_envelope_nrmse": 0.12891517376616946, "envelope_correlation": 0.9869563835247501, "rms_gain_error_db": 0.048713724602761875, "reference_mean_power": 0.000101477720422738, "output_mean_power": 0.00010262237827262288}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.361033224981284e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.73677738976729e-05, "output_power": 2.684290855914875e-05, "reference_active": true, "delta_db": -0.08409901255893634}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1422451661963211e-05, "output_power": 1.093665233560298e-05, "reference_active": true, "delta_db": -0.18874922284814064}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.10199254010919e-06, "output_power": 1.5929642556259934e-06, "reference_active": true, "delta_db": 1.6002737622971575}], "active_band_count": 3, "active_band_mae_db": 0.6243739992347448}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.7895186903062846e-05, "log_rms_mae_db": 1.4725362005435585, "rms_envelope_nrmse": 0.07108888755542073, "envelope_correlation": 0.9958006207606441, "rms_gain_error_db": 0.048713378493153985, "reference_mean_power": 0.00010203935243697765, "output_mean_power": 0.00010319033721228677}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 7.765913024651851e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5071474654741866e-05, "output_power": 2.4509888648498133e-05, "reference_active": true, "delta_db": -0.09838540853038857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067872704197783e-06, "output_power": 5.850491143730971e-06, "reference_active": true, "delta_db": -0.15844135033414825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0845180908068795e-07, "output_power": 1.369548697620822e-07, "reference_active": true, "delta_db": 1.0134067839553171}], "active_band_count": 3, "active_band_mae_db": 0.4234111809399513}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 3.774086107813168e-05, "log_rms_mae_db": 2.123262899334253, "rms_envelope_nrmse": 0.24133090063641044, "envelope_correlation": 0.9608500144328669, "rms_gain_error_db": -0.6456731109737434, "reference_mean_power": 4.068228265632881e-05, "output_mean_power": 3.506210627808636e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.809144227115543e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.649297963098168e-06, "output_power": 6.125294190027854e-06, "reference_active": true, "delta_db": -1.4985390640381273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.584218334576925e-06, "output_power": 1.64997161926144e-06, "reference_active": true, "delta_db": -1.94852729310358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3501889520111653e-06, "output_power": 8.659188495486724e-07, "reference_active": true, "delta_db": -1.929173564484691}], "active_band_count": 3, "active_band_mae_db": 1.792079973875466}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.332899304538301e-05, "log_rms_mae_db": 1.4946302166140133, "rms_envelope_nrmse": 0.1614543006786663, "envelope_correlation": 0.9838896132763347, "rms_gain_error_db": -0.6456792565711988, "reference_mean_power": 4.0907280378011884e-05, "output_mean_power": 3.525597112241848e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 2.6173037083412648e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.99444690144364e-06, "output_power": 5.632460439502134e-06, "reference_active": true, "delta_db": -1.520902722396499}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1581434874555947e-06, "output_power": 7.903642109257096e-07, "reference_active": true, "delta_db": -1.6593510295446146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.731926648355553e-08, "output_power": 5.2578636162951246e-08, "reference_active": true, "delta_db": -2.2030076205317224}], "active_band_count": 3, "active_band_mae_db": 1.7944204574909453}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 1.8891386922351935e-05, "log_rms_mae_db": 2.034295210092159, "rms_envelope_nrmse": 0.24734084810792087, "envelope_correlation": 0.9584515410929231, "rms_gain_error_db": 0.03571549273725845, "reference_mean_power": 6.700122358829966e-06, "output_mean_power": 6.755449988818665e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 5.001365064934297e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.184828883572986e-06, "output_power": 2.115040815175032e-06, "reference_active": true, "delta_db": -0.14098675943937444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0698658461156927e-06, "output_power": 1.1074070672628268e-06, "reference_active": true, "delta_db": 0.14977967238559767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.241288004170751e-07, "output_power": 2.2585638478227942e-07, "reference_active": true, "delta_db": -3.6560565222425403}], "active_band_count": 3, "active_band_mae_db": 1.3156076513558375}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 1.2324616345992443e-05, "log_rms_mae_db": 1.3973669830904212, "rms_envelope_nrmse": 0.14097337408527372, "envelope_correlation": 0.9852046438142157, "rms_gain_error_db": 0.035681604740282785, "reference_mean_power": 6.7368909993733856e-06, "output_mean_power": 6.792469252085365e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 4.500371991965458e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.395916806498352e-06, "output_power": 2.245563885885532e-06, "reference_active": true, "delta_db": -0.281463187830668}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.187368503320664e-07, "output_power": 5.773179840745444e-07, "reference_active": true, "delta_db": -0.30090895773809817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.878047962583857e-08, "output_power": 1.7143524536018303e-08, "reference_active": true, "delta_db": -4.541459527869403}], "active_band_count": 3, "active_band_mae_db": 1.7079438911460565}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.809282030506831, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d0ad5c8b87a24e2a93786cf2097167c3ca34fcf6ba40403f789bf5bb36cb999e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9614542528820359, "standard_deviation": 0.004378820209845249, "minimum": 0.9539674718832901, "maximum": 0.9683968423057615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.155002673042571, "standard_deviation": 0.05759937534555722, "minimum": 2.0449287605377187, "maximum": 2.2458053505054822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9135527942712186, "standard_deviation": 0.49614175087208273, "minimum": 0.29730652316479106, "maximum": 1.792079973875466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.235937900891004, "standard_deviation": 0.012664353359466314, "minimum": 0.21544172528837108, "maximum": 0.25655741672744903, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9861019721730226, "standard_deviation": 0.0019814197134889666, "minimum": 0.9838896132763347, "maximum": 0.9895175461050659, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.5267963476890678, "standard_deviation": 0.036458071243798545, "minimum": 1.486510262333597, "maximum": 1.5939265870355637, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9617654651138653, "standard_deviation": 0.4785797477427427, "minimum": 0.4055141192431205, "maximum": 1.7944204574909453, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14190009219451008, "standard_deviation": 0.01091053006819071, "minimum": 0.12852467204035872, "maximum": 0.1614543006786663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9648422814728549, "standard_deviation": 0.003657132003539219, "minimum": 0.9584515410929231, "maximum": 0.9709489098527008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.033565175081436, "standard_deviation": 0.03900890885891302, "minimum": 1.9572130737612021, "maximum": 2.073518288157028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.120770044230829, "standard_deviation": 0.201894362186824, "minimum": 0.7127502190293032, "maximum": 1.3156076513558375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22969938863720585, "standard_deviation": 0.013556296732450726, "minimum": 0.20670720577270216, "maximum": 0.24734084810792087, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.988294335561508, "standard_deviation": 0.0019263515543402746, "minimum": 0.9852046438142157, "maximum": 0.9910968522034458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4044333564848328, "standard_deviation": 0.05202681371658413, "minimum": 1.312673997185054, "maximum": 1.4705453048292363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3367793001488952, "standard_deviation": 0.29880501762397743, "minimum": 0.8394340318525834, "maximum": 1.7079438911460565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12806731843939417, "standard_deviation": 0.012593100360981282, "minimum": 0.11069822633344757, "maximum": 0.14830733835141535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9865364844308572, "standard_deviation": 0.0009748971001882238, "minimum": 0.9847569773721039, "maximum": 0.9876951235381419, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0691731594075464, "standard_deviation": 0.07456838448510515, "minimum": 2.0120851361059575, "maximum": 2.1975132060546496, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4364170476944083, "standard_deviation": 0.30536342953752427, "minimum": 0.05217374763266031, "maximum": 0.9156835869177646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13080939689838653, "standard_deviation": 0.004597716855424889, "minimum": 0.124940110537982, "maximum": 0.13844553029069076, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958420947849641, "standard_deviation": 0.0003251305174819333, "minimum": 0.995466426446908, "maximum": 0.9962817480848978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4390979337075387, "standard_deviation": 0.053261383586974693, "minimum": 1.3675318716335474, "maximum": 1.5213200825648534, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45488723491308347, "standard_deviation": 0.1648847684197968, "minimum": 0.26898317411411843, "maximum": 0.7461897866497492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07149639216794054, "standard_deviation": 0.0026564897862202006, "minimum": 0.06708463475761117, "maximum": 0.07482936813399282, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23233556549841944, "standard_deviation": 0.03300540931454373, "minimum": 0.19783906818500774, "maximum": 0.29173447880766806, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.108585691424483, "standard_deviation": 0.010278806305651908, "minimum": 6.095681933497899, "maximum": 6.125638563346786, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.15229113943772754, "standard_deviation": 0.013063137558613202, "minimum": 0.13523628882272692, "maximum": 0.1744540075508135, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08659579197910658, "standard_deviation": 0.04134618534994677, "minimum": -0.14271639554096846, "maximum": -0.027252580407785533, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9907581539154845, "standard_deviation": 0.001157489362927071, "minimum": 0.9890125119903017, "maximum": 0.9925414068945082, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2037830692154978, "standard_deviation": 0.03915352193026524, "minimum": 1.1388219392504835, "maximum": 1.2584965381195312, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26412275860811535, "standard_deviation": 0.0864319769689751, "minimum": 0.16916472447115768, "maximum": 0.3767359613616906, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09699260279609345, "standard_deviation": 0.0059587217389685315, "minimum": 0.08727322055310785, "maximum": 0.10568919734136037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973417658238919, "standard_deviation": 0.00046446497254801146, "minimum": 0.9964035331417344, "maximum": 0.9978915662400482, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6925342685638711, "standard_deviation": 0.049825195276368354, "minimum": 0.6166273195967783, "maximum": 0.7708607196207845, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3481180552524395, "standard_deviation": 0.11347281910153165, "minimum": 0.18173748271609333, "maximum": 0.5474700957431565, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05029547323708346, "standard_deviation": 0.003864084605639679, "minimum": 0.04484014895649624, "maximum": 0.05823794488769202, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32865-0017", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.6840808391571045, "predicted_mos": 1.6840808391571045, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "fb38feeef85234b0ee63043a5f274bbea346885bf1a806ad6678c5a838a6f658", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.878049500286579, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2434215250927665, "全帯域_10ms_rms_envelope_nrmse": 0.11640569464191401, "全帯域_10ms_envelope_correlation": 0.9867323606089164, "全帯域_10ms_modulation_mae_db": 0.5446500528049136, "全帯域_40ms_log_rms_mae_db": 0.7088053523157092, "全帯域_40ms_rms_envelope_nrmse": 0.06077009169660212, "全帯域_40ms_envelope_correlation": 0.9960024868357361, "全帯域_40ms_modulation_mae_db": 0.27061029132661324, "300–1000 Hz_10ms_log_rms_mae_db": 1.6012604674140585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12196300019597803, "300–1000 Hz_10ms_envelope_correlation": 0.9882124467765915, "300–1000 Hz_10ms_modulation_mae_db": 0.2566442117351175, "300–1000 Hz_40ms_log_rms_mae_db": 0.9718776875015834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06529649199940177, "300–1000 Hz_40ms_envelope_correlation": 0.9963763801614436, "300–1000 Hz_40ms_modulation_mae_db": 0.36226352894723296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6175879701093479, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.28266425728068084, "1000–3000 Hz_10ms_envelope_correlation": 0.94631472887984, "1000–3000 Hz_10ms_modulation_mae_db": 1.2440422495729733, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0968914333679065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.175906943914808, "1000–3000 Hz_40ms_envelope_correlation": 0.9779472829792537, "1000–3000 Hz_40ms_modulation_mae_db": 1.400026754851597, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3639966383151994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23496515901322293, "3000–6000 Hz_10ms_envelope_correlation": 0.9618092459694575, "3000–6000 Hz_10ms_modulation_mae_db": 0.9769040516276671, "3000–6000 Hz_40ms_log_rms_mae_db": 0.833118446896877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13063590910379652, "3000–6000 Hz_40ms_envelope_correlation": 0.9874752076610531, "3000–6000 Hz_40ms_modulation_mae_db": 1.411930766856312, "mr_stft_db": 5.737036355573687, "band_mae_db": 0.21826745351268081, "rms_gain_error_db": 0.0199380909007325, "output_peak": 0.19614316602165052}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2459573572534157, "全帯域_10ms_rms_envelope_nrmse": 0.10161904681478365, "全帯域_10ms_envelope_correlation": 0.9898396677371114, "全帯域_10ms_modulation_mae_db": 0.2131479141515381, "全帯域_40ms_log_rms_mae_db": 0.6777727351347024, "全帯域_40ms_rms_envelope_nrmse": 0.05431350203105979, "全帯域_40ms_envelope_correlation": 0.996855416896693, "全帯域_40ms_modulation_mae_db": 0.4081584214865124, "300–1000 Hz_10ms_log_rms_mae_db": 1.5667580078225896, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11170804155966943, "300–1000 Hz_10ms_envelope_correlation": 0.9900819650616847, "300–1000 Hz_10ms_modulation_mae_db": 0.4919741732020559, "300–1000 Hz_40ms_log_rms_mae_db": 0.9853144878216097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06598381684266248, "300–1000 Hz_40ms_envelope_correlation": 0.9962943084003879, "300–1000 Hz_40ms_modulation_mae_db": 0.44770743138877817, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5674772337654574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21569252213405318, "1000–3000 Hz_10ms_envelope_correlation": 0.9677099026214996, "1000–3000 Hz_10ms_modulation_mae_db": 0.3624305535153984, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0683791274921441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11633011408062542, "1000–3000 Hz_40ms_envelope_correlation": 0.9905769346928747, "1000–3000 Hz_40ms_modulation_mae_db": 0.43596716090092075, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3519892953334678, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20969410095559615, "3000–6000 Hz_10ms_envelope_correlation": 0.9697170973402508, "3000–6000 Hz_10ms_modulation_mae_db": 0.7670391909882044, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8258164627040159, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11640582399748547, "3000–6000 Hz_40ms_envelope_correlation": 0.9899367423659128, "3000–6000 Hz_40ms_modulation_mae_db": 1.101575155335625, "mr_stft_db": 5.711201271432266, "band_mae_db": 0.2235210501001568, "rms_gain_error_db": -0.07607061819492937, "output_peak": 0.15609772790620768}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.280174851566866, "全帯域_10ms_rms_envelope_nrmse": 0.10506952775032477, "全帯域_10ms_envelope_correlation": 0.9891360960699527, "全帯域_10ms_modulation_mae_db": 0.37675161301930177, "全帯域_40ms_log_rms_mae_db": 0.7472621761460659, "全帯域_40ms_rms_envelope_nrmse": 0.0482755151041127, "全帯域_40ms_envelope_correlation": 0.9974684766336079, "全帯域_40ms_modulation_mae_db": 0.2288694539872955, "300–1000 Hz_10ms_log_rms_mae_db": 1.5133634288911046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11265854043025504, "300–1000 Hz_10ms_envelope_correlation": 0.9899059537969288, "300–1000 Hz_10ms_modulation_mae_db": 0.4534982903067121, "300–1000 Hz_40ms_log_rms_mae_db": 0.9180502622105877, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05636425309928521, "300–1000 Hz_40ms_envelope_correlation": 0.9973028622745442, "300–1000 Hz_40ms_modulation_mae_db": 0.3015876357201937, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5951988557016974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2165567953965065, "1000–3000 Hz_10ms_envelope_correlation": 0.9674553292961683, "1000–3000 Hz_10ms_modulation_mae_db": 0.2500212732274418, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0514758006797975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10417316547774588, "1000–3000 Hz_40ms_envelope_correlation": 0.9922499512882298, "1000–3000 Hz_40ms_modulation_mae_db": 0.252481128450647, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3427434024943747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2248738229054203, "3000–6000 Hz_10ms_envelope_correlation": 0.9650868007245809, "3000–6000 Hz_10ms_modulation_mae_db": 1.427928143742539, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8250870848747182, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12580977221001832, "3000–6000 Hz_40ms_envelope_correlation": 0.9881934505178036, "3000–6000 Hz_40ms_modulation_mae_db": 1.9929066648348772, "mr_stft_db": 5.711782527750518, "band_mae_db": 0.21816318883908659, "rms_gain_error_db": -0.024195295328563494, "output_peak": 0.15063075645546453}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1731724392352771, "全帯域_10ms_rms_envelope_nrmse": 0.09078888329370179, "全帯域_10ms_envelope_correlation": 0.9918900881991053, "全帯域_10ms_modulation_mae_db": 0.40385236657324325, "全帯域_40ms_log_rms_mae_db": 0.642276209078465, "全帯域_40ms_rms_envelope_nrmse": 0.04884816675524447, "全帯域_40ms_envelope_correlation": 0.9974186961853548, "全帯域_40ms_modulation_mae_db": 0.2827682861549073, "300–1000 Hz_10ms_log_rms_mae_db": 1.59844728194999, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12320929426603454, "300–1000 Hz_10ms_envelope_correlation": 0.988170975608121, "300–1000 Hz_10ms_modulation_mae_db": 0.6369829379104787, "300–1000 Hz_40ms_log_rms_mae_db": 0.9454971276270595, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06712559759865586, "300–1000 Hz_40ms_envelope_correlation": 0.9962944280989403, "300–1000 Hz_40ms_modulation_mae_db": 0.09948559947576263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5787188001590926, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19407177264760725, "1000–3000 Hz_10ms_envelope_correlation": 0.9748322828477017, "1000–3000 Hz_10ms_modulation_mae_db": 0.3853982105789897, "1000–3000 Hz_40ms_log_rms_mae_db": 1.057134960064159, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11587553392054738, "1000–3000 Hz_40ms_envelope_correlation": 0.9919959915132969, "1000–3000 Hz_40ms_modulation_mae_db": 0.43303842538670595, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2961326265918218, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22243493465546965, "3000–6000 Hz_10ms_envelope_correlation": 0.9660833511544042, "3000–6000 Hz_10ms_modulation_mae_db": 0.169749013050127, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7763721377597121, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11410089722894177, "3000–6000 Hz_40ms_envelope_correlation": 0.9902004856528042, "3000–6000 Hz_40ms_modulation_mae_db": 0.473100922897115, "mr_stft_db": 5.710601911053101, "band_mae_db": 0.25879254190752127, "rms_gain_error_db": -0.04100016384977406, "output_peak": 0.1669076874236488}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2560536166403018, "全帯域_10ms_rms_envelope_nrmse": 0.0977008582841445, "全帯域_10ms_envelope_correlation": 0.9908850466287523, "全帯域_10ms_modulation_mae_db": 0.500521372838207, "全帯域_40ms_log_rms_mae_db": 0.7490529107160413, "全帯域_40ms_rms_envelope_nrmse": 0.058992947982423695, "全帯域_40ms_envelope_correlation": 0.996653786170183, "全帯域_40ms_modulation_mae_db": 0.7019434379475826, "300–1000 Hz_10ms_log_rms_mae_db": 1.5604146244786061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11013456233058037, "300–1000 Hz_10ms_envelope_correlation": 0.9903772780997833, "300–1000 Hz_10ms_modulation_mae_db": 0.2902231582896924, "300–1000 Hz_40ms_log_rms_mae_db": 1.0045171253341045, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06535934275020983, "300–1000 Hz_40ms_envelope_correlation": 0.9964176978216813, "300–1000 Hz_40ms_modulation_mae_db": 0.36875745873816174, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6316265205287246, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23669986134265297, "1000–3000 Hz_10ms_envelope_correlation": 0.9634693733401769, "1000–3000 Hz_10ms_modulation_mae_db": 1.3370614596402763, "1000–3000 Hz_40ms_log_rms_mae_db": 1.138525245138097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16043743266541938, "1000–3000 Hz_40ms_envelope_correlation": 0.9855678623249227, "1000–3000 Hz_40ms_modulation_mae_db": 1.3505557213594817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3552743517493804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23870872696833106, "3000–6000 Hz_10ms_envelope_correlation": 0.9611775960870046, "3000–6000 Hz_10ms_modulation_mae_db": 0.570125371696366, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8276880163267415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1400695389915593, "3000–6000 Hz_40ms_envelope_correlation": 0.985284781975031, "3000–6000 Hz_40ms_modulation_mae_db": 0.6463048192146773, "mr_stft_db": 5.711431742531935, "band_mae_db": 0.2572752885589403, "rms_gain_error_db": -0.18491346331306102, "output_peak": 0.12784366664571173}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2072288781216463, "全帯域_10ms_rms_envelope_nrmse": 0.10832689341393505, "全帯域_10ms_envelope_correlation": 0.9885055339250699, "全帯域_10ms_modulation_mae_db": 0.21938887216162276, "全帯域_40ms_log_rms_mae_db": 0.6827121310975918, "全帯域_40ms_rms_envelope_nrmse": 0.05907033814779441, "全帯域_40ms_envelope_correlation": 0.9962567209819146, "全帯域_40ms_modulation_mae_db": 0.49532945803969236, "300–1000 Hz_10ms_log_rms_mae_db": 1.5934885383008872, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1314679830708255, "300–1000 Hz_10ms_envelope_correlation": 0.9867922370652282, "300–1000 Hz_10ms_modulation_mae_db": 0.49270857869078527, "300–1000 Hz_40ms_log_rms_mae_db": 0.9350366687259914, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07982780780451128, "300–1000 Hz_40ms_envelope_correlation": 0.9949058840175822, "300–1000 Hz_40ms_modulation_mae_db": 0.22507225857423932, "1000–3000 Hz_10ms_log_rms_mae_db": 1.603079854145333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21030085228622217, "1000–3000 Hz_10ms_envelope_correlation": 0.9695937993458951, "1000–3000 Hz_10ms_modulation_mae_db": 1.17961269596399, "1000–3000 Hz_40ms_log_rms_mae_db": 1.156575315621312, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12490999275620912, "1000–3000 Hz_40ms_envelope_correlation": 0.9893091205392806, "1000–3000 Hz_40ms_modulation_mae_db": 1.3384479262615028, "3000–6000 Hz_10ms_log_rms_mae_db": 1.351102370287387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2190748277800015, "3000–6000 Hz_10ms_envelope_correlation": 0.9674911977437863, "3000–6000 Hz_10ms_modulation_mae_db": 1.3436473588170568, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8440721982485679, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12566293912886983, "3000–6000 Hz_40ms_envelope_correlation": 0.9897025671879648, "3000–6000 Hz_40ms_modulation_mae_db": 2.049765054400181, "mr_stft_db": 5.736756195636112, "band_mae_db": 0.31431957157076784, "rms_gain_error_db": -0.020733047944913614, "output_peak": 0.1649958630840109}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2709820619410046, "全帯域_10ms_rms_envelope_nrmse": 0.1034310554670004, "全帯域_10ms_envelope_correlation": 0.9894752395453632, "全帯域_10ms_modulation_mae_db": 0.3428709476701997, "全帯域_40ms_log_rms_mae_db": 0.7636686107268301, "全帯域_40ms_rms_envelope_nrmse": 0.0626767367559941, "全帯域_40ms_envelope_correlation": 0.9957852105542362, "全帯域_40ms_modulation_mae_db": 0.18284338151711563, "300–1000 Hz_10ms_log_rms_mae_db": 1.647228446448428, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13397213717514897, "300–1000 Hz_10ms_envelope_correlation": 0.9858199567586914, "300–1000 Hz_10ms_modulation_mae_db": 0.16675846671337724, "300–1000 Hz_40ms_log_rms_mae_db": 0.9833807441792893, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07264283536256862, "300–1000 Hz_40ms_envelope_correlation": 0.9955120356540209, "300–1000 Hz_40ms_modulation_mae_db": 0.4250759039846576, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6390104451577405, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22581790095255375, "1000–3000 Hz_10ms_envelope_correlation": 0.9645629432212173, "1000–3000 Hz_10ms_modulation_mae_db": 1.1860378082316412, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1312915749796586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15180680680259517, "1000–3000 Hz_40ms_envelope_correlation": 0.9834234339366077, "1000–3000 Hz_40ms_modulation_mae_db": 1.0612110839172522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3301228279961448, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21419629770922202, "3000–6000 Hz_10ms_envelope_correlation": 0.9683984662920302, "3000–6000 Hz_10ms_modulation_mae_db": 0.7773489254780875, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8365886946608112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12275976739515634, "3000–6000 Hz_40ms_envelope_correlation": 0.9886951594228881, "3000–6000 Hz_40ms_modulation_mae_db": 1.0139846576700176, "mr_stft_db": 5.712455421388092, "band_mae_db": 0.286166809550092, "rms_gain_error_db": -0.07956640325529028, "output_peak": 0.14971412245696147}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2650528049668157, "全帯域_10ms_rms_envelope_nrmse": 0.10724970021454468, "全帯域_10ms_envelope_correlation": 0.9890848233055337, "全帯域_10ms_modulation_mae_db": 0.3422624487435888, "全帯域_40ms_log_rms_mae_db": 0.7614668964499391, "全帯域_40ms_rms_envelope_nrmse": 0.06614179439139366, "全帯域_40ms_envelope_correlation": 0.9958796147163963, "全帯域_40ms_modulation_mae_db": 0.6560456256896537, "300–1000 Hz_10ms_log_rms_mae_db": 1.6115519438495984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12033563917332335, "300–1000 Hz_10ms_envelope_correlation": 0.9884991565607172, "300–1000 Hz_10ms_modulation_mae_db": 0.495451200780362, "300–1000 Hz_40ms_log_rms_mae_db": 0.9808721770029523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06519320903617495, "300–1000 Hz_40ms_envelope_correlation": 0.9964407088529375, "300–1000 Hz_40ms_modulation_mae_db": 0.176409272664513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6134190510310784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24785036443820013, "1000–3000 Hz_10ms_envelope_correlation": 0.9616826222926049, "1000–3000 Hz_10ms_modulation_mae_db": 1.2504421425279164, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1502714290877678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17701328121005896, "1000–3000 Hz_40ms_envelope_correlation": 0.9841920970692444, "1000–3000 Hz_40ms_modulation_mae_db": 1.3881679208997353, "3000–6000 Hz_10ms_log_rms_mae_db": 1.377692014030254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18632893448434784, "3000–6000 Hz_10ms_envelope_correlation": 0.9765126206602832, "3000–6000 Hz_10ms_modulation_mae_db": 0.7049747020847535, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8503833753917851, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09965598828925086, "3000–6000 Hz_40ms_envelope_correlation": 0.9926268638904934, "3000–6000 Hz_40ms_modulation_mae_db": 0.7384787773259546, "mr_stft_db": 5.7220450805703384, "band_mae_db": 0.28460932212060247, "rms_gain_error_db": -0.222019933948376, "output_peak": 0.14957361260112015}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87280, "output_samples": 87280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.734800771397301, "spectral_convergence": 0.35284744025465903, "spectral_convergence_status": "ok", "amplitude_floor": 4.3919411473735467e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.604348898020904, "spectral_convergence": 0.3497662517967441, "spectral_convergence_status": "ok", "amplitude_floor": 3.1709236310047874e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.871959397302857, "spectral_convergence": 0.36328495987302517, "spectral_convergence_status": "ok", "amplitude_floor": 2.989229099554426e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001235363438035874, "output_power": 0.00012231628751698562, "delta_db": -0.04310452645854375, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 7.224896060742283e-05, "output_power": 7.110810386880903e-05, "delta_db": -0.06912505471634009, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.329659782840476e-05, "output_power": 4.582645473753378e-05, "delta_db": 0.24662488724246173, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.747692740945149e-05, "output_power": 4.714320045760976e-05, "delta_db": -0.0306354214293526, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.909876260278539e-06, "output_power": 8.052208996098782e-06, "delta_db": -0.43956634097301545, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1191996093186054e-05, "output_power": 2.367155098266305e-05, "delta_db": 0.48054849025637136, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21826745351268081, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.0199380909007325, "pitch": null, "output_rms_dbfs": -34.97412020694017, "output_peak": 0.19614316602165052, "reference_rms_dbfs": -34.9940582978409, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.737036355573687}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 5.579882181126775e-05, "log_rms_mae_db": 1.2434215250927665, "rms_envelope_nrmse": 0.11640569464191401, "envelope_correlation": 0.9867323606089164, "rms_gain_error_db": 0.019938386753623263, "reference_mean_power": 0.0003169510345489487, "output_mean_power": 0.0003184094967535679}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8938855835285014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.978895728357633e-05, "output_power": 5.7772758889025665e-05, "reference_active": true, "delta_db": -0.14897872011463634}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1930505918996815e-05, "output_power": 2.2921284670762904e-05, "reference_active": true, "delta_db": 0.191903042943536}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0863861439295844e-06, "output_power": 4.156780577008869e-06, "reference_active": true, "delta_db": 1.2930683953565687}], "active_band_count": 3, "active_band_mae_db": 0.5446500528049136}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 5.258028604719465e-05, "log_rms_mae_db": 0.7088053523157092, "rms_envelope_nrmse": 0.06077009169660212, "envelope_correlation": 0.9960024868357361, "rms_gain_error_db": 0.019940213529334446, "reference_mean_power": 0.00031870366172679027, "output_mean_power": 0.00032017032338525346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 1.711395707762808e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.23137186055323e-05, "output_power": 5.0850696093135514e-05, "reference_active": true, "delta_db": -0.12318689652777581}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1932036565608145e-05, "output_power": 1.251857877762349e-05, "reference_active": true, "delta_db": 0.2084045104910062}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.569030812932547e-07, "output_power": 2.869414310530474e-07, "reference_active": true, "delta_db": 0.48023946696105774}], "active_band_count": 3, "active_band_mae_db": 0.27061029132661324}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 4.2439491809696055e-05, "log_rms_mae_db": 1.6012604674140585, "rms_envelope_nrmse": 0.12196300019597803, "envelope_correlation": 0.9882124467765915, "rms_gain_error_db": 0.0025163272022947127, "reference_mean_power": 0.000101477720422738, "output_mean_power": 0.00010153653423438454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.361033224981284e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.73677738976729e-05, "output_power": 2.7241769042128314e-05, "reference_active": true, "delta_db": -0.020041665842909433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1422451661963211e-05, "output_power": 1.0989439913422536e-05, "reference_active": true, "delta_db": -0.16783770081957253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.10199254010919e-06, "output_power": 9.637698317253178e-07, "reference_active": true, "delta_db": -0.5820532685428705}], "active_band_count": 3, "active_band_mae_db": 0.2566442117351175}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.7895186903062846e-05, "log_rms_mae_db": 0.9718776875015834, "rms_envelope_nrmse": 0.06529649199940177, "envelope_correlation": 0.9963763801614436, "rms_gain_error_db": 0.0025162115757780975, "reference_mean_power": 0.00010203935243697765, "output_mean_power": 0.00010209848903746503}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 7.765913024651851e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5071474654741866e-05, "output_power": 2.477347005335446e-05, "reference_active": true, "delta_db": -0.05193035994897702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067872704197783e-06, "output_power": 5.929639243675951e-06, "reference_active": true, "delta_db": -0.10008189390445564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0845180908068795e-07, "output_power": 8.744982959787844e-08, "reference_active": true, "delta_db": -0.9347783329882664}], "active_band_count": 3, "active_band_mae_db": 0.36226352894723296}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 3.774086107813168e-05, "log_rms_mae_db": 1.6175879701093479, "rms_envelope_nrmse": 0.28266425728068084, "envelope_correlation": 0.94631472887984, "rms_gain_error_db": 0.09149993449968408, "reference_mean_power": 4.068228265632881e-05, "output_mean_power": 4.154849588894354e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.809144227115543e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.649297963098168e-06, "output_power": 9.37229255870986e-06, "reference_active": true, "delta_db": 0.34864978210595743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.584218334576925e-06, "output_power": 3.221425679393375e-06, "reference_active": true, "delta_db": 0.9571891291145178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3501889520111653e-06, "output_power": 2.3606050603995167e-06, "reference_active": true, "delta_db": 2.4262878374984447}], "active_band_count": 3, "active_band_mae_db": 1.2440422495729733}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.332899304538301e-05, "log_rms_mae_db": 1.0968914333679065, "rms_envelope_nrmse": 0.175906943914808, "envelope_correlation": 0.9779472829792537, "rms_gain_error_db": 0.09150144429906108, "reference_mean_power": 4.0907280378011884e-05, "output_mean_power": 4.1778298819667936e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 2.6173037083412648e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.99444690144364e-06, "output_power": 9.300203267833494e-06, "reference_active": true, "delta_db": 0.6570401841023157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1581434874555947e-06, "output_power": 1.633828768517791e-06, "reference_active": true, "delta_db": 1.4944416949905701}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.731926648355553e-08, "output_power": 1.3994903795186149e-07, "reference_active": true, "delta_db": 2.048598385461905}], "active_band_count": 3, "active_band_mae_db": 1.400026754851597}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 1.8891386922351935e-05, "log_rms_mae_db": 1.3639966383151994, "rms_envelope_nrmse": 0.23496515901322293, "envelope_correlation": 0.9618092459694575, "rms_gain_error_db": -0.245212375478781, "reference_mean_power": 6.700122358829966e-06, "output_mean_power": 6.332300211566524e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 5.001365064934297e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.184828883572986e-06, "output_power": 2.0009494634167916e-06, "reference_active": true, "delta_db": -0.38181308467253766}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0698658461156927e-06, "output_power": 9.451978978112575e-07, "reference_active": true, "delta_db": -0.5380657647263851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.241288004170751e-07, "output_power": 3.298790186426849e-07, "reference_active": true, "delta_db": -2.0108333054840783}], "active_band_count": 3, "active_band_mae_db": 0.9769040516276671}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 1.2324616345992443e-05, "log_rms_mae_db": 0.833118446896877, "rms_envelope_nrmse": 0.13063590910379652, "envelope_correlation": 0.9874752076610531, "rms_gain_error_db": -0.24522127673828553, "reference_mean_power": 6.7368909993733856e-06, "output_mean_power": 6.367037283867961e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 4.500371991965458e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.395916806498352e-06, "output_power": 2.1330724827293456e-06, "reference_active": true, "delta_db": -0.5046612074157801}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.187368503320664e-07, "output_power": 4.742248316962445e-07, "reference_active": true, "delta_db": -1.1552169114959425}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.878047962583857e-08, "output_power": 2.6955950361431467e-08, "reference_active": true, "delta_db": -2.5759141816572133}], "active_band_count": 3, "active_band_mae_db": 1.411930766856312}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8368900515998569, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0cf2ef46968f7a4892223c982d9105292efd1a254b51010db4b8d204226f8122", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.964452622730638, "standard_deviation": 0.008378801532823532, "minimum": 0.94631472887984, "maximum": 0.9748322828477017, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6057648413248091, "standard_deviation": 0.024752340061509056, "minimum": 1.5674772337654574, "maximum": 1.6390104451577405, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8993807991573284, "standard_deviation": 0.4733563383185416, "minimum": 0.2500212732274418, "maximum": 1.3370614596402763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2287067908098096, "standard_deviation": 0.02726975324449468, "minimum": 0.19407177264760725, "maximum": 0.28266425728068084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9869078342929638, "standard_deviation": 0.005003703064499203, "minimum": 0.9779472829792537, "maximum": 0.9922499512882298, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1063181108038553, "standard_deviation": 0.04321630683934922, "minimum": 1.0514758006797975, "maximum": 1.156575315621312, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9574870152534803, "standard_deviation": 0.4979648512828039, "minimum": 0.252481128450647, "maximum": 1.400026754851597, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1408066588535012, "standard_deviation": 0.02894501959028403, "minimum": 0.10417316547774588, "maximum": 0.17701328121005896, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9670345469964747, "standard_deviation": 0.004860994847057264, "minimum": 0.9611775960870046, "maximum": 0.9765126206602832, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3461316908497538, "standard_deviation": 0.024569110671733764, "minimum": 1.2961326265918218, "maximum": 1.377692014030254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8422145946856001, "standard_deviation": 0.40840676791297426, "minimum": 0.169749013050127, "maximum": 1.427928143742539, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21878460055895144, "standard_deviation": 0.016328454791451927, "minimum": 0.18632893448434784, "maximum": 0.23870872696833106, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.989014407334244, "standard_deviation": 0.002165839339802889, "minimum": 0.985284781975031, "maximum": 0.9926268638904934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8273908021079036, "standard_deviation": 0.02247256311744972, "minimum": 0.7763721377597121, "maximum": 0.8503833753917851, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.178505852316845, "standard_deviation": 0.5957314720826614, "minimum": 0.473100922897115, "maximum": 2.049765054400181, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1218875795431348, "standard_deviation": 0.012085648404245327, "minimum": 0.09965598828925086, "maximum": 0.1400695389915593, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9884824962159682, "standard_deviation": 0.001616480841412252, "minimum": 0.9858199567586914, "maximum": 0.9903772780997833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5865640923944078, "standard_deviation": 0.039900073312949876, "minimum": 1.5133634288911046, "maximum": 1.647228446448428, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4105301272035726, "standard_deviation": 0.1564354828124482, "minimum": 0.16675846671337724, "maximum": 0.6369829379104787, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1206811497752269, "standard_deviation": 0.00892244032708894, "minimum": 0.11013456233058037, "maximum": 0.13397213717514897, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961930381601922, "standard_deviation": 0.0007089431139020377, "minimum": 0.9949058840175822, "maximum": 0.9973028622745442, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9655682850503973, "standard_deviation": 0.029498715473317838, "minimum": 0.9180502622105877, "maximum": 1.0045171253341045, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.30079488618669237, "standard_deviation": 0.12372673067519123, "minimum": 0.09948559947576263, "maximum": 0.44770743138877817, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06722416931168375, "standard_deviation": 0.006750851093058978, "minimum": 0.05636425309928521, "maximum": 0.07982780780451128, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.257639403269981, "standard_deviation": 0.03591234800620515, "minimum": 0.21816318883908659, "maximum": 0.31431957157076784, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.719163813242006, "standard_deviation": 0.0115461833181717, "minimum": 5.710601911053101, "maximum": 5.737036355573687, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.15773832532434695, "standard_deviation": 0.01960186187498713, "minimum": 0.12784366664571173, "maximum": 0.19614316602165052, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07857010436677192, "standard_deviation": 0.08393119083116957, "minimum": -0.222019933948376, "maximum": 0.0199380909007325, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9894436070024756, "standard_deviation": 0.0015434196718838574, "minimum": 0.9867323606089164, "maximum": 0.9918900881991053, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2427554418522617, "standard_deviation": 0.0358281514861456, "minimum": 1.1731724392352771, "maximum": 1.280174851566866, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3679306984953269, "standard_deviation": 0.11801175881928003, "minimum": 0.2131479141515381, "maximum": 0.5446500528049136, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10382395748504361, "standard_deviation": 0.007599672985974014, "minimum": 0.09078888329370179, "maximum": 0.11640569464191401, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9965400511217652, "standard_deviation": 0.0006677889069513652, "minimum": 0.9957852105542362, "maximum": 0.9974684766336079, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.716627127708168, "standard_deviation": 0.04545417420812148, "minimum": 0.642276209078465, "maximum": 0.7636686107268301, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40332104451867157, "standard_deviation": 0.19738391345776382, "minimum": 0.18284338151711563, "maximum": 0.7019434379475826, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05738613660807812, "standard_deviation": 0.006401595330677023, "minimum": 0.0482755151041127, "maximum": 0.06614179439139366, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32865-0017", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.302715539932251, "predicted_mos": 2.302715539932251, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "6c88bd68e5c7ccc959ac18ba2ee70f5aa613480b24f59b6120c9c404d4f2fe2b", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.831088799983263, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1278709263183142, "全帯域_10ms_rms_envelope_nrmse": 0.09962546310864644, "全帯域_10ms_envelope_correlation": 0.9903986765306055, "全帯域_10ms_modulation_mae_db": 0.1579044839025301, "全帯域_40ms_log_rms_mae_db": 0.6200352004023109, "全帯域_40ms_rms_envelope_nrmse": 0.05830490836707898, "全帯域_40ms_envelope_correlation": 0.9966058288510359, "全帯域_40ms_modulation_mae_db": 0.33826487217448276, "300–1000 Hz_10ms_log_rms_mae_db": 1.807406360609351, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1250581155746613, "300–1000 Hz_10ms_envelope_correlation": 0.9875525857257076, "300–1000 Hz_10ms_modulation_mae_db": 0.1417448785664063, "300–1000 Hz_40ms_log_rms_mae_db": 1.206568814442824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06546579998302066, "300–1000 Hz_40ms_envelope_correlation": 0.9963933783534016, "300–1000 Hz_40ms_modulation_mae_db": 0.530210013423703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7642098155562962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22161014192785708, "1000–3000 Hz_10ms_envelope_correlation": 0.9683771781173227, "1000–3000 Hz_10ms_modulation_mae_db": 0.9618237798587881, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2532508877570403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14562622612581652, "1000–3000 Hz_40ms_envelope_correlation": 0.9888724902939601, "1000–3000 Hz_40ms_modulation_mae_db": 0.9107513640957552, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6283848963954044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22404917572979907, "3000–6000 Hz_10ms_envelope_correlation": 0.9656341788251901, "3000–6000 Hz_10ms_modulation_mae_db": 0.8020404883916116, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0659375795758916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12724011489136253, "3000–6000 Hz_40ms_envelope_correlation": 0.9878256819358339, "3000–6000 Hz_40ms_modulation_mae_db": 1.0632363767002138, "mr_stft_db": 5.844634811747731, "band_mae_db": 0.31365206969125453, "rms_gain_error_db": -0.15535031442758687, "output_peak": 0.13522516010487368}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.260422048223653, "全帯域_10ms_rms_envelope_nrmse": 0.09935033217114923, "全帯域_10ms_envelope_correlation": 0.9903387476681155, "全帯域_10ms_modulation_mae_db": 0.18782128956192853, "全帯域_40ms_log_rms_mae_db": 0.6670042948092666, "全帯域_40ms_rms_envelope_nrmse": 0.05025538034901816, "全帯域_40ms_envelope_correlation": 0.9973972402131156, "全帯域_40ms_modulation_mae_db": 0.4754091234159012, "300–1000 Hz_10ms_log_rms_mae_db": 1.7620581803330733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12924646125690217, "300–1000 Hz_10ms_envelope_correlation": 0.9867186018257857, "300–1000 Hz_10ms_modulation_mae_db": 0.12670568226501555, "300–1000 Hz_40ms_log_rms_mae_db": 1.1347040292462862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0716518524005131, "300–1000 Hz_40ms_envelope_correlation": 0.995621214252405, "300–1000 Hz_40ms_modulation_mae_db": 0.2971539451959866, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8161225068856317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23132768353751818, "1000–3000 Hz_10ms_envelope_correlation": 0.9666491610364799, "1000–3000 Hz_10ms_modulation_mae_db": 0.8574005827933918, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3301796931228247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1618375612858687, "1000–3000 Hz_40ms_envelope_correlation": 0.9869161788389788, "1000–3000 Hz_40ms_modulation_mae_db": 0.7902115470827574, "3000–6000 Hz_10ms_log_rms_mae_db": 1.561136449356386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21334398197932303, "3000–6000 Hz_10ms_envelope_correlation": 0.9688228986250416, "3000–6000 Hz_10ms_modulation_mae_db": 1.2492713035134695, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0340703843660397, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11531855324240611, "3000–6000 Hz_40ms_envelope_correlation": 0.9904924081804066, "3000–6000 Hz_40ms_modulation_mae_db": 1.6007944152700386, "mr_stft_db": 5.837387811120972, "band_mae_db": 0.33906925543893635, "rms_gain_error_db": -0.10884978289130057, "output_peak": 0.15478305586799715}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.174362629304106, "全帯域_10ms_rms_envelope_nrmse": 0.10139287383235289, "全帯域_10ms_envelope_correlation": 0.9899358108640675, "全帯域_10ms_modulation_mae_db": 0.19658741478565292, "全帯域_40ms_log_rms_mae_db": 0.694912734217169, "全帯域_40ms_rms_envelope_nrmse": 0.0599304164234376, "全帯域_40ms_envelope_correlation": 0.9961452829921533, "全帯域_40ms_modulation_mae_db": 0.28981330223271895, "300–1000 Hz_10ms_log_rms_mae_db": 1.7781283865233348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11853009164260364, "300–1000 Hz_10ms_envelope_correlation": 0.9889165397873254, "300–1000 Hz_10ms_modulation_mae_db": 0.3114209346762715, "300–1000 Hz_40ms_log_rms_mae_db": 1.11883325306138, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06416379552047767, "300–1000 Hz_40ms_envelope_correlation": 0.996513238285832, "300–1000 Hz_40ms_modulation_mae_db": 0.685587109080891, "1000–3000 Hz_10ms_log_rms_mae_db": 1.792887524676522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23903129143360644, "1000–3000 Hz_10ms_envelope_correlation": 0.960997672779986, "1000–3000 Hz_10ms_modulation_mae_db": 0.35672525855415455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2946616742049784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1505866197624331, "1000–3000 Hz_40ms_envelope_correlation": 0.9835130849316851, "1000–3000 Hz_40ms_modulation_mae_db": 0.0640136023302242, "3000–6000 Hz_10ms_log_rms_mae_db": 1.58228499477583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24597948518766313, "3000–6000 Hz_10ms_envelope_correlation": 0.9584748288238357, "3000–6000 Hz_10ms_modulation_mae_db": 0.7238463624597768, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0454564422803492, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15164793479825614, "3000–6000 Hz_40ms_envelope_correlation": 0.9826422436431055, "3000–6000 Hz_40ms_modulation_mae_db": 0.9590750314377102, "mr_stft_db": 5.855047304799221, "band_mae_db": 0.22319085295983995, "rms_gain_error_db": -0.007082835426637329, "output_peak": 0.15246715984438877}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1595807867534977, "全帯域_10ms_rms_envelope_nrmse": 0.09938755017334476, "全帯域_10ms_envelope_correlation": 0.9905252353547984, "全帯域_10ms_modulation_mae_db": 0.46282594523165227, "全帯域_40ms_log_rms_mae_db": 0.6416878757821091, "全帯域_40ms_rms_envelope_nrmse": 0.053061166481329684, "全帯域_40ms_envelope_correlation": 0.9973573669045364, "全帯域_40ms_modulation_mae_db": 0.8102585580440494, "300–1000 Hz_10ms_log_rms_mae_db": 1.7728938275218027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1274366901401769, "300–1000 Hz_10ms_envelope_correlation": 0.9871181463210784, "300–1000 Hz_10ms_modulation_mae_db": 0.15432189776864566, "300–1000 Hz_40ms_log_rms_mae_db": 1.2122641207242422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07405217164614818, "300–1000 Hz_40ms_envelope_correlation": 0.9954383219495916, "300–1000 Hz_40ms_modulation_mae_db": 0.4768644247085394, "1000–3000 Hz_10ms_log_rms_mae_db": 1.683215479737395, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20816997588828015, "1000–3000 Hz_10ms_envelope_correlation": 0.9701831988484821, "1000–3000 Hz_10ms_modulation_mae_db": 0.7459820096209909, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1698741187496176, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12251667092895642, "1000–3000 Hz_40ms_envelope_correlation": 0.9899173247630908, "1000–3000 Hz_40ms_modulation_mae_db": 0.8244953077817682, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5588272119805477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23994971279290828, "3000–6000 Hz_10ms_envelope_correlation": 0.9602598583686253, "3000–6000 Hz_10ms_modulation_mae_db": 0.9362875778377978, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0615668342418094, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1405389769692375, "3000–6000 Hz_40ms_envelope_correlation": 0.9851512195968489, "3000–6000 Hz_40ms_modulation_mae_db": 1.4985594013182533, "mr_stft_db": 5.833419805436455, "band_mae_db": 0.29250787191192884, "rms_gain_error_db": -0.18274102929262775, "output_peak": 0.1468240705612893}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1958480892419685, "全帯域_10ms_rms_envelope_nrmse": 0.0980405416924683, "全帯域_10ms_envelope_correlation": 0.9905579307360671, "全帯域_10ms_modulation_mae_db": 0.3277385718861649, "全帯域_40ms_log_rms_mae_db": 0.6793836994748432, "全帯域_40ms_rms_envelope_nrmse": 0.0544947541740397, "全帯域_40ms_envelope_correlation": 0.9968430747547111, "全帯域_40ms_modulation_mae_db": 0.27249548630524073, "300–1000 Hz_10ms_log_rms_mae_db": 1.8237067944610155, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13367513115316484, "300–1000 Hz_10ms_envelope_correlation": 0.9860432795845889, "300–1000 Hz_10ms_modulation_mae_db": 0.6159066176940495, "300–1000 Hz_40ms_log_rms_mae_db": 1.2789125797644874, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0748818439161727, "300–1000 Hz_40ms_envelope_correlation": 0.9953400422323225, "300–1000 Hz_40ms_modulation_mae_db": 0.20174087922329528, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7803078044724567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24009749733421096, "1000–3000 Hz_10ms_envelope_correlation": 0.9608611969350098, "1000–3000 Hz_10ms_modulation_mae_db": 0.8222203094185669, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2670064878163445, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13296108506863305, "1000–3000 Hz_40ms_envelope_correlation": 0.9899875350554733, "1000–3000 Hz_40ms_modulation_mae_db": 0.8676053010117419, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5734336524102308, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21482190223412972, "3000–6000 Hz_10ms_envelope_correlation": 0.9686457750195436, "3000–6000 Hz_10ms_modulation_mae_db": 1.358941760682855, "3000–6000 Hz_40ms_log_rms_mae_db": 1.031393717108394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1166954976618413, "3000–6000 Hz_40ms_envelope_correlation": 0.9909204012696399, "3000–6000 Hz_40ms_modulation_mae_db": 1.8793920017678556, "mr_stft_db": 5.829910669014424, "band_mae_db": 0.3117574310972713, "rms_gain_error_db": -0.08461412814298891, "output_peak": 0.15214566975022328}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.23333727329875, "全帯域_10ms_rms_envelope_nrmse": 0.10780005714818182, "全帯域_10ms_envelope_correlation": 0.9885798501201088, "全帯域_10ms_modulation_mae_db": 0.368818015586651, "全帯域_40ms_log_rms_mae_db": 0.7032435340782113, "全帯域_40ms_rms_envelope_nrmse": 0.05948708944505335, "全帯域_40ms_envelope_correlation": 0.9961636901074996, "全帯域_40ms_modulation_mae_db": 0.4806569685113568, "300–1000 Hz_10ms_log_rms_mae_db": 1.798649588521514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14102937333511179, "300–1000 Hz_10ms_envelope_correlation": 0.9841826426465955, "300–1000 Hz_10ms_modulation_mae_db": 0.32696594440221133, "300–1000 Hz_40ms_log_rms_mae_db": 1.2308799180347763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0741246680433858, "300–1000 Hz_40ms_envelope_correlation": 0.9953458726451854, "300–1000 Hz_40ms_modulation_mae_db": 0.8481562480437752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7616101374897504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25507180493777515, "1000–3000 Hz_10ms_envelope_correlation": 0.9589666907635465, "1000–3000 Hz_10ms_modulation_mae_db": 1.0586768721553506, "1000–3000 Hz_40ms_log_rms_mae_db": 1.254958318396361, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15892118483285295, "1000–3000 Hz_40ms_envelope_correlation": 0.9838957961627383, "1000–3000 Hz_40ms_modulation_mae_db": 0.9734889067647203, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5568983746636893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21774190866451115, "3000–6000 Hz_10ms_envelope_correlation": 0.9674278891248468, "3000–6000 Hz_10ms_modulation_mae_db": 0.6240295404122592, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0340655965422756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11867600196208593, "3000–6000 Hz_40ms_envelope_correlation": 0.9896442738985343, "3000–6000 Hz_40ms_modulation_mae_db": 0.6956249885909281, "mr_stft_db": 5.841884272578561, "band_mae_db": 0.2117979325284145, "rms_gain_error_db": -0.01693362378405197, "output_peak": 0.1405207393236831}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2417823715643002, "全帯域_10ms_rms_envelope_nrmse": 0.10606346257853916, "全帯域_10ms_envelope_correlation": 0.9889459046496211, "全帯域_10ms_modulation_mae_db": 0.31647029533141713, "全帯域_40ms_log_rms_mae_db": 0.7151442365941674, "全帯域_40ms_rms_envelope_nrmse": 0.05385605919605424, "全帯域_40ms_envelope_correlation": 0.9969429759203308, "全帯域_40ms_modulation_mae_db": 0.5092523460094395, "300–1000 Hz_10ms_log_rms_mae_db": 1.8327117406586102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12573333024124575, "300–1000 Hz_10ms_envelope_correlation": 0.9875793959502189, "300–1000 Hz_10ms_modulation_mae_db": 0.275413605509997, "300–1000 Hz_40ms_log_rms_mae_db": 1.2268989790995992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07256435392139587, "300–1000 Hz_40ms_envelope_correlation": 0.9955694211894184, "300–1000 Hz_40ms_modulation_mae_db": 0.5878607933579033, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8389609986706028, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24705012931303127, "1000–3000 Hz_10ms_envelope_correlation": 0.9587751109548115, "1000–3000 Hz_10ms_modulation_mae_db": 1.1333573054341561, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3075791651662578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14733380664001694, "1000–3000 Hz_40ms_envelope_correlation": 0.987357732564412, "1000–3000 Hz_40ms_modulation_mae_db": 1.0872538100104114, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5979892946014231, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21382619156593574, "3000–6000 Hz_10ms_envelope_correlation": 0.9700272868732536, "3000–6000 Hz_10ms_modulation_mae_db": 1.173128452378034, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0131466446117534, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11190688982573857, "3000–6000 Hz_40ms_envelope_correlation": 0.9913202839948998, "3000–6000 Hz_40ms_modulation_mae_db": 1.1826095731355097, "mr_stft_db": 5.842058372171628, "band_mae_db": 0.2543784501057795, "rms_gain_error_db": -0.09497500944710267, "output_peak": 0.16305487386362474}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2426669261085834, "全帯域_10ms_rms_envelope_nrmse": 0.10009961350124046, "全帯域_10ms_envelope_correlation": 0.9903991774356132, "全帯域_10ms_modulation_mae_db": 0.5528066248940053, "全帯域_40ms_log_rms_mae_db": 0.6838882066041562, "全帯域_40ms_rms_envelope_nrmse": 0.056826279745846725, "全帯域_40ms_envelope_correlation": 0.9969100127216577, "全帯域_40ms_modulation_mae_db": 0.7704961037935373, "300–1000 Hz_10ms_log_rms_mae_db": 1.7388092358015979, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1306043474614935, "300–1000 Hz_10ms_envelope_correlation": 0.9864900463486911, "300–1000 Hz_10ms_modulation_mae_db": 0.2683218128115806, "300–1000 Hz_40ms_log_rms_mae_db": 1.0599048362107153, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07132589297677325, "300–1000 Hz_40ms_envelope_correlation": 0.995861001380338, "300–1000 Hz_40ms_modulation_mae_db": 0.6088183068283547, "1000–3000 Hz_10ms_log_rms_mae_db": 1.803766122941613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2318118320931234, "1000–3000 Hz_10ms_envelope_correlation": 0.9641489047003727, "1000–3000 Hz_10ms_modulation_mae_db": 1.40989719820491, "1000–3000 Hz_40ms_log_rms_mae_db": 1.281999084546249, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15030863658442664, "1000–3000 Hz_40ms_envelope_correlation": 0.9865717793991438, "1000–3000 Hz_40ms_modulation_mae_db": 1.7594442609499348, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6291740068036458, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21970466948065392, "3000–6000 Hz_10ms_envelope_correlation": 0.9703233259442012, "3000–6000 Hz_10ms_modulation_mae_db": 0.7643714638786546, "3000–6000 Hz_40ms_log_rms_mae_db": 1.072634900202092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14841005250898465, "3000–6000 Hz_40ms_envelope_correlation": 0.9860874988694839, "3000–6000 Hz_40ms_modulation_mae_db": 0.7327484268297205, "mr_stft_db": 5.837977733433824, "band_mae_db": 0.242127769019941, "rms_gain_error_db": -0.18276823756073268, "output_peak": 0.13133465962732147}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87280, "output_samples": 87280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9162872287059995, "spectral_convergence": 0.3460509842088498, "spectral_convergence_status": "ok", "amplitude_floor": 4.3919411473735467e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.712700102998662, "spectral_convergence": 0.3408853524838566, "spectral_convergence_status": "ok", "amplitude_floor": 3.1709236310047874e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.904917103538533, "spectral_convergence": 0.3568013974107628, "spectral_convergence_status": "ok", "amplitude_floor": 2.989229099554426e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001235363438035874, "output_power": 0.00012390320475122284, "delta_db": 0.012877957695700316, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 7.224896060742283e-05, "output_power": 7.299676446661409e-05, "delta_db": 0.044720071321911206, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.329659782840476e-05, "output_power": 4.225199498685534e-05, "delta_db": -0.10606551997284347, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.747692740945149e-05, "output_power": 3.610186357980362e-05, "delta_db": -1.189529839312487, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.909876260278539e-06, "output_power": 9.74929419265041e-06, "delta_db": 0.3910150313879708, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1191996093186054e-05, "output_power": 2.0530591666218716e-05, "delta_db": -0.13770399845661452, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.31365206969125453, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15535031442758687, "pitch": null, "output_rms_dbfs": -35.14940861226849, "output_peak": 0.13522516010487368, "reference_rms_dbfs": -34.9940582978409, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.844634811747731}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 5.579882181126775e-05, "log_rms_mae_db": 1.1278709263183142, "rms_envelope_nrmse": 0.09962546310864644, "envelope_correlation": 0.9903986765306055, "rms_gain_error_db": -0.15535037459984086, "reference_mean_power": 0.0003169510345489487, "output_mean_power": 0.00030581384009693017}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8938855835285014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.978895728357633e-05, "output_power": 5.532361984295237e-05, "reference_active": true, "delta_db": -0.3371039101198569}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1930505918996815e-05, "output_power": 2.1417516070143534e-05, "reference_active": true, "delta_db": -0.10279549178846226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0863861439295844e-06, "output_power": 3.0624489338284106e-06, "reference_active": true, "delta_db": -0.03381404979927109}], "active_band_count": 3, "active_band_mae_db": 0.1579044839025301}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 5.258028604719465e-05, "log_rms_mae_db": 0.6200352004023109, "rms_envelope_nrmse": 0.05830490836707898, "envelope_correlation": 0.9966058288510359, "rms_gain_error_db": -0.15534795424843456, "reference_mean_power": 0.00031870366172679027, "output_mean_power": 0.0003075050539010059}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 1.711395707762808e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.23137186055323e-05, "output_power": 4.878894960817776e-05, "reference_active": true, "delta_db": -0.3029412383423409}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1932036565608145e-05, "output_power": 1.1390473565175162e-05, "reference_active": true, "delta_db": -0.20172795097880242}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.569030812932547e-07, "output_power": 2.284319097028163e-07, "reference_active": true, "delta_db": -0.510125427202305}], "active_band_count": 3, "active_band_mae_db": 0.33826487217448276}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 4.2439491809696055e-05, "log_rms_mae_db": 1.807406360609351, "rms_envelope_nrmse": 0.1250581155746613, "envelope_correlation": 0.9875525857257076, "rms_gain_error_db": -0.07308604550905394, "reference_mean_power": 0.000101477720422738, "output_mean_power": 9.978427315411621e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.361033224981284e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.73677738976729e-05, "output_power": 2.7450876436509267e-05, "reference_active": true, "delta_db": 0.013167417101985112}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1422451661963211e-05, "output_power": 1.0816509191846695e-05, "reference_active": true, "delta_db": -0.23672205185713419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.10199254010919e-06, "output_power": 1.1473955765723658e-06, "reference_active": true, "delta_db": 0.1753451667400996}], "active_band_count": 3, "active_band_mae_db": 0.1417448785664063}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.7895186903062846e-05, "log_rms_mae_db": 1.206568814442824, "rms_envelope_nrmse": 0.06546579998302066, "envelope_correlation": 0.9963933783534016, "rms_gain_error_db": -0.0730869120359501, "reference_mean_power": 0.00010203935243697765, "output_mean_power": 0.00010033651270519873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 7.765913024651851e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5071474654741866e-05, "output_power": 2.519539692250533e-05, "reference_active": true, "delta_db": 0.02141325466123372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067872704197783e-06, "output_power": 5.669684337092255e-06, "reference_active": true, "delta_db": -0.29477581323454427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0845180908068795e-07, "output_power": 8.087097956634858e-08, "reference_active": true, "delta_db": -1.2744409723753312}], "active_band_count": 3, "active_band_mae_db": 0.530210013423703}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 3.774086107813168e-05, "log_rms_mae_db": 1.7642098155562962, "rms_envelope_nrmse": 0.22161014192785708, "envelope_correlation": 0.9683771781173227, "rms_gain_error_db": -0.7358389458805575, "reference_mean_power": 4.068228265632881e-05, "output_mean_power": 3.434167060847998e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.809144227115543e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.649297963098168e-06, "output_power": 6.256976997739458e-06, "reference_active": true, "delta_db": -1.406163001465671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.584218334576925e-06, "output_power": 2.2521084653868007e-06, "reference_active": true, "delta_db": -0.5973990035417673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3501889520111653e-06, "output_power": 1.6541903519866075e-06, "reference_active": true, "delta_db": 0.8819093345689256}], "active_band_count": 3, "active_band_mae_db": 0.9618237798587881}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.332899304538301e-05, "log_rms_mae_db": 1.2532508877570403, "rms_envelope_nrmse": 0.14562622612581652, "envelope_correlation": 0.9888724902939601, "rms_gain_error_db": -0.7358428920766649, "reference_mean_power": 4.0907280378011884e-05, "output_mean_power": 3.453156951913339e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 2.6173037083412648e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.99444690144364e-06, "output_power": 5.903042292977668e-06, "reference_active": true, "delta_db": -1.3171252752702673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1581434874555947e-06, "output_power": 1.0316054092118506e-06, "reference_active": true, "delta_db": -0.5024875867497726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.731926648355553e-08, "output_power": 1.0773931297382055e-07, "reference_active": true, "delta_db": 0.9126412302672255}], "active_band_count": 3, "active_band_mae_db": 0.9107513640957552}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 1.8891386922351935e-05, "log_rms_mae_db": 1.6283848963954044, "rms_envelope_nrmse": 0.22404917572979907, "envelope_correlation": 0.9656341788251901, "rms_gain_error_db": -0.04283994514782775, "reference_mean_power": 6.700122358829966e-06, "output_mean_power": 6.634355501099017e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 5.001365064934297e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.184828883572986e-06, "output_power": 2.351697238596321e-06, "reference_active": true, "delta_db": 0.3196398064129544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0698658461156927e-06, "output_power": 1.0824144008250373e-06, "reference_active": true, "delta_db": 0.05064238104212963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.241288004170751e-07, "output_power": 3.2798508643955174e-07, "reference_active": true, "delta_db": -2.035839277719751}], "active_band_count": 3, "active_band_mae_db": 0.8020404883916116}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 1.2324616345992443e-05, "log_rms_mae_db": 1.0659375795758916, "rms_envelope_nrmse": 0.12724011489136253, "envelope_correlation": 0.9878256819358339, "rms_gain_error_db": -0.04284567510066904, "reference_mean_power": 6.7368909993733856e-06, "output_mean_power": 6.670754427934614e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 4.500371991965458e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.395916806498352e-06, "output_power": 2.5397488211063712e-06, "reference_active": true, "delta_db": 0.25319033449884365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.187368503320664e-07, "output_power": 5.50980662068492e-07, "reference_active": true, "delta_db": -0.5036962566499313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.878047962583857e-08, "output_power": 2.7858890235907377e-08, "reference_active": true, "delta_db": -2.432822538951866}], "active_band_count": 3, "active_band_mae_db": 1.0632363767002138}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.830147405884757, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c4e357bdcb0a27ad1e425247a7f35a9d77e20baf3ccd4bbe4f35b45c3f8839fc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9636198892670014, "standard_deviation": 0.004387900743506238, "minimum": 0.9587751109548115, "maximum": 0.9701831988484821, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7801350488037835, "standard_deviation": 0.04700586730616549, "minimum": 1.683215479737395, "maximum": 1.8389609986706028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9182604145050386, "standard_deviation": 0.30880348559591353, "minimum": 0.35672525855415455, "maximum": 1.40989719820491, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23427129455817533, "standard_deviation": 0.014681667297682704, "minimum": 0.20816997588828015, "maximum": 0.25507180493777515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9871289902511853, "standard_deviation": 0.0024762382506356637, "minimum": 0.9835130849316851, "maximum": 0.9899875350554733, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2699386787199591, "standard_deviation": 0.048295035827190906, "minimum": 1.1698741187496176, "maximum": 1.3301796931228247, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9096580125034142, "standard_deviation": 0.4623164589015406, "minimum": 0.0640136023302242, "maximum": 1.7594442609499348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14626147390362554, "standard_deviation": 0.012984818828816433, "minimum": 0.12251667092895642, "maximum": 0.1618375612858687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9662020052005673, "standard_deviation": 0.004492895021757718, "minimum": 0.9584748288238357, "maximum": 0.9703233259442012, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5860161101233947, "standard_deviation": 0.029711447834166262, "minimum": 1.5568983746636893, "maximum": 1.6291740068036458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9539896186943073, "standard_deviation": 0.27265778023403836, "minimum": 0.6240295404122592, "maximum": 1.358941760682855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2236771284543655, "standard_deviation": 0.012510382576125502, "minimum": 0.21334398197932303, "maximum": 0.24597948518766313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.988010501423594, "standard_deviation": 0.0031390861422991544, "minimum": 0.9826422436431055, "maximum": 0.9913202839948998, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0447840123660757, "standard_deviation": 0.02040710639270388, "minimum": 1.0131466446117534, "maximum": 1.072634900202092, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2015050268812788, "standard_deviation": 0.4244490532260364, "minimum": 0.6956249885909281, "maximum": 1.8793920017678556, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1288042527324891, "standard_deviation": 0.01586982695136915, "minimum": 0.11190688982573857, "maximum": 0.15164793479825614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.986825154773749, "standard_deviation": 0.0013771409931350271, "minimum": 0.9841826426465955, "maximum": 0.9889165397873254, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7892955143037874, "standard_deviation": 0.0320023729529079, "minimum": 1.7388092358015979, "maximum": 1.8327117406586102, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.27760017171177215, "standard_deviation": 0.15790878215739762, "minimum": 0.12670568226501555, "maximum": 0.6159066176940495, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12891419260067, "standard_deviation": 0.006626903474803002, "minimum": 0.11853009164260364, "maximum": 0.14102937333511179, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9957603112860618, "standard_deviation": 0.00046069219426085064, "minimum": 0.9953400422323225, "maximum": 0.996513238285832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1836208163230388, "standard_deviation": 0.0721406521999807, "minimum": 1.0599048362107153, "maximum": 1.2789125797644874, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.529548964982806, "standard_deviation": 0.2069451567093229, "minimum": 0.20174087922329528, "maximum": 0.8481562480437752, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0710287973009859, "standard_deviation": 0.00404405670692205, "minimum": 0.06416379552047767, "maximum": 0.0748818439161727, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27356020409417076, "standard_deviation": 0.04694133594943926, "minimum": 0.2117979325284145, "maximum": 0.33906925543893635, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.840290097537853, "standard_deviation": 0.0076627451364943595, "minimum": 5.829910669014424, "maximum": 5.855047304799221, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.1470444236179252, "standard_deviation": 0.010695449054406881, "minimum": 0.13133465962732147, "maximum": 0.16305487386362474, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1041643701216286, "standard_deviation": 0.0681337939294547, "minimum": -0.18276823756073268, "maximum": -0.007082835426637329, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899601666698746, "standard_deviation": 0.000769003422683895, "minimum": 0.9885798501201088, "maximum": 0.9905579307360671, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2044838813516465, "standard_deviation": 0.047326251447660055, "minimum": 1.1278709263183142, "maximum": 1.260422048223653, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3213715801475003, "standard_deviation": 0.13939435443887416, "minimum": 0.1579044839025301, "maximum": 0.5528066248940053, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10146998677574039, "standard_deviation": 0.0035263603083849683, "minimum": 0.0980405416924683, "maximum": 0.10780005714818182, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.99679568405813, "standard_deviation": 0.0004740181041694807, "minimum": 0.9961452829921533, "maximum": 0.9973972402131156, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6756624727452791, "standard_deviation": 0.03183717083131565, "minimum": 0.6200352004023109, "maximum": 0.7151442365941674, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49333084506084085, "standard_deviation": 0.20432193337748708, "minimum": 0.27249548630524073, "maximum": 0.8102585580440494, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055777006772732306, "standard_deviation": 0.003416454795241511, "minimum": 0.05025538034901816, "maximum": 0.0599304164234376, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32865-0017", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.6192288398742676, "predicted_mos": 1.6192288398742676, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "d9464d8967fbad459b7835b15e946a0bfdafb7185a2b3e033966efc72faaf29c", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.809514000080526, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1986852387653162, "全帯域_10ms_rms_envelope_nrmse": 0.11004872382321389, "全帯域_10ms_envelope_correlation": 0.9881232721079415, "全帯域_10ms_modulation_mae_db": 0.20306650590033737, "全帯域_40ms_log_rms_mae_db": 0.6628234205083897, "全帯域_40ms_rms_envelope_nrmse": 0.061305079789369096, "全帯域_40ms_envelope_correlation": 0.9959505049379858, "全帯域_40ms_modulation_mae_db": 0.36595437554124716, "300–1000 Hz_10ms_log_rms_mae_db": 1.675211068270181, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11912607139719786, "300–1000 Hz_10ms_envelope_correlation": 0.9887251501207569, "300–1000 Hz_10ms_modulation_mae_db": 0.5054798251408498, "300–1000 Hz_40ms_log_rms_mae_db": 1.150327523699216, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06995747436635648, "300–1000 Hz_40ms_envelope_correlation": 0.9958679703485211, "300–1000 Hz_40ms_modulation_mae_db": 0.9454515384871608, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7265195454617621, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2684037485990322, "1000–3000 Hz_10ms_envelope_correlation": 0.9505497120187678, "1000–3000 Hz_10ms_modulation_mae_db": 0.7892272631076663, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1720017154760627, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15949525089242123, "1000–3000 Hz_40ms_envelope_correlation": 0.9814238415167001, "1000–3000 Hz_40ms_modulation_mae_db": 0.7959767562548091, "3000–6000 Hz_10ms_log_rms_mae_db": 1.540683503229357, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2015839426087963, "3000–6000 Hz_10ms_envelope_correlation": 0.9726043984763217, "3000–6000 Hz_10ms_modulation_mae_db": 0.5737791453893707, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9534256032371117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10345399823793064, "3000–6000 Hz_40ms_envelope_correlation": 0.9920959722654883, "3000–6000 Hz_40ms_modulation_mae_db": 0.8520302967915773, "mr_stft_db": 5.798320872593405, "band_mae_db": 0.13255383598435264, "rms_gain_error_db": -0.0071360111331841, "output_peak": 0.15511349312736902}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.17376102543136, "全帯域_10ms_rms_envelope_nrmse": 0.10130913030891042, "全帯域_10ms_envelope_correlation": 0.9899746795066908, "全帯域_10ms_modulation_mae_db": 0.2354627200040765, "全帯域_40ms_log_rms_mae_db": 0.6804169298316265, "全帯域_40ms_rms_envelope_nrmse": 0.050671833629084576, "全帯域_40ms_envelope_correlation": 0.9973976638437853, "全帯域_40ms_modulation_mae_db": 0.703840874128946, "300–1000 Hz_10ms_log_rms_mae_db": 1.6965670313756873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1294289041107707, "300–1000 Hz_10ms_envelope_correlation": 0.9866961634237527, "300–1000 Hz_10ms_modulation_mae_db": 0.4450639110086468, "300–1000 Hz_40ms_log_rms_mae_db": 1.0832971546172254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07313638728445364, "300–1000 Hz_40ms_envelope_correlation": 0.9954993922119133, "300–1000 Hz_40ms_modulation_mae_db": 1.157998204197731, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6967556482727781, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22552519490390213, "1000–3000 Hz_10ms_envelope_correlation": 0.9652720770819248, "1000–3000 Hz_10ms_modulation_mae_db": 1.2271061069962537, "1000–3000 Hz_40ms_log_rms_mae_db": 1.194032248045171, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13029035996834848, "1000–3000 Hz_40ms_envelope_correlation": 0.9894071726603055, "1000–3000 Hz_40ms_modulation_mae_db": 1.2637029267715465, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5303397690812788, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22274163460493693, "3000–6000 Hz_10ms_envelope_correlation": 0.9660636262952146, "3000–6000 Hz_10ms_modulation_mae_db": 0.8630124266483837, "3000–6000 Hz_40ms_log_rms_mae_db": 0.972334780656354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12488864091987102, "3000–6000 Hz_40ms_envelope_correlation": 0.988276077355013, "3000–6000 Hz_40ms_modulation_mae_db": 1.3368993537552132, "mr_stft_db": 5.7974852450751255, "band_mae_db": 0.2392325769886229, "rms_gain_error_db": -0.1252560935566998, "output_peak": 0.13977720017876039}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1517101767871307, "全帯域_10ms_rms_envelope_nrmse": 0.10172139200129897, "全帯域_10ms_envelope_correlation": 0.9899226688593279, "全帯域_10ms_modulation_mae_db": 0.3389244154749223, "全帯域_40ms_log_rms_mae_db": 0.6484181582248116, "全帯域_40ms_rms_envelope_nrmse": 0.05716603731178805, "全帯域_40ms_envelope_correlation": 0.9965121581960008, "全帯域_40ms_modulation_mae_db": 0.43766456374804297, "300–1000 Hz_10ms_log_rms_mae_db": 1.655654382525554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12489408998823501, "300–1000 Hz_10ms_envelope_correlation": 0.9877413734026478, "300–1000 Hz_10ms_modulation_mae_db": 0.29099686443325035, "300–1000 Hz_40ms_log_rms_mae_db": 1.0639159129514542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06348179589239218, "300–1000 Hz_40ms_envelope_correlation": 0.9966356187674077, "300–1000 Hz_40ms_modulation_mae_db": 0.2821427226427525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.722401969037679, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.229646846578785, "1000–3000 Hz_10ms_envelope_correlation": 0.9650855305367596, "1000–3000 Hz_10ms_modulation_mae_db": 0.5515974051795388, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1788513443245363, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13556410139468747, "1000–3000 Hz_40ms_envelope_correlation": 0.9873224897019534, "1000–3000 Hz_40ms_modulation_mae_db": 0.7318765430175898, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4816385767838283, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20143665102214536, "3000–6000 Hz_10ms_envelope_correlation": 0.9726500328337688, "3000–6000 Hz_10ms_modulation_mae_db": 1.3616397714626363, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9452978192313888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10877313910556662, "3000–6000 Hz_40ms_envelope_correlation": 0.9923604687756478, "3000–6000 Hz_40ms_modulation_mae_db": 1.7786925566553735, "mr_stft_db": 5.794291528239728, "band_mae_db": 0.13755079604387985, "rms_gain_error_db": 0.036324582279961576, "output_peak": 0.1527400197226232}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2697302748712211, "全帯域_10ms_rms_envelope_nrmse": 0.0956472566962789, "全帯域_10ms_envelope_correlation": 0.9910301726895997, "全帯域_10ms_modulation_mae_db": 0.3737175432607757, "全帯域_40ms_log_rms_mae_db": 0.7235401555528739, "全帯域_40ms_rms_envelope_nrmse": 0.046762829108229195, "全帯域_40ms_envelope_correlation": 0.9977233374371214, "全帯域_40ms_modulation_mae_db": 0.4607843187232195, "300–1000 Hz_10ms_log_rms_mae_db": 1.644734034826233, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1111108523030687, "300–1000 Hz_10ms_envelope_correlation": 0.9902086232878394, "300–1000 Hz_10ms_modulation_mae_db": 0.23256840278106017, "300–1000 Hz_40ms_log_rms_mae_db": 0.9987684172657948, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05241795557378091, "300–1000 Hz_40ms_envelope_correlation": 0.9976639017236126, "300–1000 Hz_40ms_modulation_mae_db": 0.4825982305847432, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7630230353896836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20778337481273493, "1000–3000 Hz_10ms_envelope_correlation": 0.9705337869771211, "1000–3000 Hz_10ms_modulation_mae_db": 0.4340575621905101, "1000–3000 Hz_40ms_log_rms_mae_db": 1.161472792144096, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12808574404324413, "1000–3000 Hz_40ms_envelope_correlation": 0.9892082837441676, "1000–3000 Hz_40ms_modulation_mae_db": 0.5806165138615189, "3000–6000 Hz_10ms_log_rms_mae_db": 1.531130094086852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2153877166269603, "3000–6000 Hz_10ms_envelope_correlation": 0.9681974112135384, "3000–6000 Hz_10ms_modulation_mae_db": 0.93386175204656, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9616900596712066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10107104041665968, "3000–6000 Hz_40ms_envelope_correlation": 0.992358745784026, "3000–6000 Hz_40ms_modulation_mae_db": 1.2964515490906119, "mr_stft_db": 5.802755620570241, "band_mae_db": 0.2184397390521301, "rms_gain_error_db": -0.0934571659416529, "output_peak": 0.15159485380167537}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2161754070176916, "全帯域_10ms_rms_envelope_nrmse": 0.10628932315669103, "全帯域_10ms_envelope_correlation": 0.9889998660865864, "全帯域_10ms_modulation_mae_db": 0.27827668662702304, "全帯域_40ms_log_rms_mae_db": 0.7051787633132107, "全帯域_40ms_rms_envelope_nrmse": 0.058956052287953926, "全帯域_40ms_envelope_correlation": 0.996485503321321, "全帯域_40ms_modulation_mae_db": 0.640177792939208, "300–1000 Hz_10ms_log_rms_mae_db": 1.7195234798450751, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12056003891594139, "300–1000 Hz_10ms_envelope_correlation": 0.9885450965124659, "300–1000 Hz_10ms_modulation_mae_db": 0.5971129380603316, "300–1000 Hz_40ms_log_rms_mae_db": 1.0924618225972114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06653401765145488, "300–1000 Hz_40ms_envelope_correlation": 0.996443839801357, "300–1000 Hz_40ms_modulation_mae_db": 1.023314713070928, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7359305298808032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24979639690070948, "1000–3000 Hz_10ms_envelope_correlation": 0.9572058951975636, "1000–3000 Hz_10ms_modulation_mae_db": 0.6357213302431234, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2003539876565108, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1503456553687033, "1000–3000 Hz_40ms_envelope_correlation": 0.9858223671231957, "1000–3000 Hz_40ms_modulation_mae_db": 0.7824230666844979, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5443834988282081, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21159457543003887, "3000–6000 Hz_10ms_envelope_correlation": 0.9698509908925347, "3000–6000 Hz_10ms_modulation_mae_db": 0.7258902057822268, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9673162214601163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11699608962024058, "3000–6000 Hz_40ms_envelope_correlation": 0.9899019349692518, "3000–6000 Hz_40ms_modulation_mae_db": 1.112739221236964, "mr_stft_db": 5.804862084603836, "band_mae_db": 0.2864905764833368, "rms_gain_error_db": -0.15004319548449202, "output_peak": 0.14993962914171644}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.231622853618676, "全帯域_10ms_rms_envelope_nrmse": 0.10398329333166749, "全帯域_10ms_envelope_correlation": 0.9894453917776916, "全帯域_10ms_modulation_mae_db": 0.3658407264547177, "全帯域_40ms_log_rms_mae_db": 0.7085532550462088, "全帯域_40ms_rms_envelope_nrmse": 0.05415992306055856, "全帯域_40ms_envelope_correlation": 0.9970220443689078, "全帯域_40ms_modulation_mae_db": 0.5471449598445969, "300–1000 Hz_10ms_log_rms_mae_db": 1.695439664398786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1307819049901209, "300–1000 Hz_10ms_envelope_correlation": 0.986744241706109, "300–1000 Hz_10ms_modulation_mae_db": 0.13895859579432362, "300–1000 Hz_40ms_log_rms_mae_db": 1.091571508324799, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07124527847631185, "300–1000 Hz_40ms_envelope_correlation": 0.9959189088820847, "300–1000 Hz_40ms_modulation_mae_db": 0.5562544686546486, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7854202826123609, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2505129746490984, "1000–3000 Hz_10ms_envelope_correlation": 0.957846089715796, "1000–3000 Hz_10ms_modulation_mae_db": 1.3941752631686029, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2316541709163193, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16229283424927282, "1000–3000 Hz_40ms_envelope_correlation": 0.9843044639242664, "1000–3000 Hz_40ms_modulation_mae_db": 1.4986750520585639, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5092423968017814, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2325018165169811, "3000–6000 Hz_10ms_envelope_correlation": 0.9628664269915026, "3000–6000 Hz_10ms_modulation_mae_db": 1.195243017162263, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9777878265145128, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12843665273952423, "3000–6000 Hz_40ms_envelope_correlation": 0.9884634492216662, "3000–6000 Hz_40ms_modulation_mae_db": 1.595749144101554, "mr_stft_db": 5.796929265811922, "band_mae_db": 0.36695508884446165, "rms_gain_error_db": -0.1342249218407267, "output_peak": 0.14055501930930256}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2288016520275187, "全帯域_10ms_rms_envelope_nrmse": 0.09872016755330369, "全帯域_10ms_envelope_correlation": 0.9904220511100401, "全帯域_10ms_modulation_mae_db": 0.31863649382348364, "全帯域_40ms_log_rms_mae_db": 0.6838603284018678, "全帯域_40ms_rms_envelope_nrmse": 0.05419072937212153, "全帯域_40ms_envelope_correlation": 0.9968762098594993, "全帯域_40ms_modulation_mae_db": 0.21427613075592558, "300–1000 Hz_10ms_log_rms_mae_db": 1.6845328545901002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13668058648463832, "300–1000 Hz_10ms_envelope_correlation": 0.9851169317233774, "300–1000 Hz_10ms_modulation_mae_db": 0.2412687672444088, "300–1000 Hz_40ms_log_rms_mae_db": 1.0620631023540148, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07837204627757156, "300–1000 Hz_40ms_envelope_correlation": 0.9948326444253899, "300–1000 Hz_40ms_modulation_mae_db": 0.5435351622108641, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7505379650362247, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20606638891382217, "1000–3000 Hz_10ms_envelope_correlation": 0.9710895986346388, "1000–3000 Hz_10ms_modulation_mae_db": 0.3427923643331186, "1000–3000 Hz_40ms_log_rms_mae_db": 1.203081272759668, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1113043233052089, "1000–3000 Hz_40ms_envelope_correlation": 0.9910163050563635, "1000–3000 Hz_40ms_modulation_mae_db": 0.46590541573909666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4862215506759118, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22529661620715213, "3000–6000 Hz_10ms_envelope_correlation": 0.965377894478213, "3000–6000 Hz_10ms_modulation_mae_db": 0.58778012569128, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9978032699594926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14070006701143092, "3000–6000 Hz_40ms_envelope_correlation": 0.985153067569146, "3000–6000 Hz_40ms_modulation_mae_db": 0.5891987120967729, "mr_stft_db": 5.8068391506473525, "band_mae_db": 0.23100528470999815, "rms_gain_error_db": -0.0818863432820836, "output_peak": 0.14353734616312716}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1884595462569216, "全帯域_10ms_rms_envelope_nrmse": 0.09911809149619093, "全帯域_10ms_envelope_correlation": 0.9903326286940423, "全帯域_10ms_modulation_mae_db": 0.4898359764768246, "全帯域_40ms_log_rms_mae_db": 0.6835492066846274, "全帯域_40ms_rms_envelope_nrmse": 0.05720247316467801, "全帯域_40ms_envelope_correlation": 0.9964783286153017, "全帯域_40ms_modulation_mae_db": 0.7650293313249148, "300–1000 Hz_10ms_log_rms_mae_db": 1.6569898754784942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11971118807103076, "300–1000 Hz_10ms_envelope_correlation": 0.9886932469811797, "300–1000 Hz_10ms_modulation_mae_db": 0.2989126489436103, "300–1000 Hz_40ms_log_rms_mae_db": 1.055572482460772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06302208048841744, "300–1000 Hz_40ms_envelope_correlation": 0.9966887178093271, "300–1000 Hz_40ms_modulation_mae_db": 0.26974068507037835, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7526678382363592, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21922714710447105, "1000–3000 Hz_10ms_envelope_correlation": 0.9666359292904394, "1000–3000 Hz_10ms_modulation_mae_db": 0.553408073387751, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2613470627263512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12094666315861884, "1000–3000 Hz_40ms_envelope_correlation": 0.9893176783700919, "1000–3000 Hz_40ms_modulation_mae_db": 0.9374599664370193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5429742945901697, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24828256926256337, "3000–6000 Hz_10ms_envelope_correlation": 0.9573933927372971, "3000–6000 Hz_10ms_modulation_mae_db": 1.5090048200680943, "3000–6000 Hz_40ms_log_rms_mae_db": 0.991294879395916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15560955181487998, "3000–6000 Hz_40ms_envelope_correlation": 0.9823918949810037, "3000–6000 Hz_40ms_modulation_mae_db": 2.324505299121714, "mr_stft_db": 5.7957155632910835, "band_mae_db": 0.12016698597216639, "rms_gain_error_db": -0.0652579112266032, "output_peak": 0.14757259306790194}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87280, "output_samples": 87280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.883624077561946, "spectral_convergence": 0.3434391786226405, "spectral_convergence_status": "ok", "amplitude_floor": 4.3919411473735467e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.667700042934169, "spectral_convergence": 0.33845975804567735, "spectral_convergence_status": "ok", "amplitude_floor": 3.1709236310047874e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.843638497284102, "spectral_convergence": 0.34757056947862836, "spectral_convergence_status": "ok", "amplitude_floor": 2.989229099554426e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001235363438035874, "output_power": 0.00012376040917681238, "delta_db": 0.007869927981922535, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 7.224896060742283e-05, "output_power": 7.129691512285404e-05, "delta_db": -0.057608644075204836, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.329659782840476e-05, "output_power": 4.3316024515892744e-05, "delta_db": 0.0019481925591939477, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.747692740945149e-05, "output_power": 4.866885283814532e-05, "delta_db": 0.10768504997086392, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.909876260278539e-06, "output_power": 7.772889273169924e-06, "delta_db": -0.592891915165286, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1191996093186054e-05, "output_power": 2.1325724472533786e-05, "delta_db": 0.02731928615364452, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13255383598435264, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0071360111331841, "pitch": null, "output_rms_dbfs": -35.001194308974085, "output_peak": 0.15511349312736902, "reference_rms_dbfs": -34.9940582978409, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.798320872593405}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 5.579882181126775e-05, "log_rms_mae_db": 1.1986852387653162, "rms_envelope_nrmse": 0.11004872382321389, "envelope_correlation": 0.9881232721079415, "rms_gain_error_db": -0.007135407229613781, "reference_mean_power": 0.0003169510345489487, "output_mean_power": 0.0003164307152853449}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8938855835285014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.978895728357633e-05, "output_power": 6.0981441488171567e-05, "reference_active": true, "delta_db": 0.08576706654583496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1930505918996815e-05, "output_power": 2.1231357490919073e-05, "reference_active": true, "delta_db": -0.14070887632603477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0863861439295844e-06, "output_power": 3.3707195354527196e-06, "reference_active": true, "delta_db": 0.3827235748291423}], "active_band_count": 3, "active_band_mae_db": 0.20306650590033737}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 5.258028604719465e-05, "log_rms_mae_db": 0.6628234205083897, "rms_envelope_nrmse": 0.061305079789369096, "envelope_correlation": 0.9959505049379858, "rms_gain_error_db": -0.0071271727513008195, "reference_mean_power": 0.00031870366172679027, "output_mean_power": 0.00031818106857100495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 1.711395707762808e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.23137186055323e-05, "output_power": 5.439542427398952e-05, "reference_active": true, "delta_db": 0.16946776527255195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1932036565608145e-05, "output_power": 1.1088837780573849e-05, "reference_active": true, "delta_db": -0.31828545388404417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.569030812932547e-07, "output_power": 2.2323296234857777e-07, "reference_active": true, "delta_db": -0.6101099074671456}], "active_band_count": 3, "active_band_mae_db": 0.36595437554124716}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 4.2439491809696055e-05, "log_rms_mae_db": 1.675211068270181, "rms_envelope_nrmse": 0.11912607139719786, "envelope_correlation": 0.9887251501207569, "rms_gain_error_db": -0.07924982085403731, "reference_mean_power": 0.000101477720422738, "output_mean_power": 9.964275360510661e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.361033224981284e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.73677738976729e-05, "output_power": 2.782729333504271e-05, "reference_active": true, "delta_db": 0.0723149284717777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1422451661963211e-05, "output_power": 9.452946287631055e-06, "reference_active": true, "delta_db": -0.8219213865903607}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.10199254010919e-06, "output_power": 9.549009789037519e-07, "reference_active": true, "delta_db": -0.6222031603604108}], "active_band_count": 3, "active_band_mae_db": 0.5054798251408498}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.7895186903062846e-05, "log_rms_mae_db": 1.150327523699216, "rms_envelope_nrmse": 0.06995747436635648, "envelope_correlation": 0.9958679703485211, "rms_gain_error_db": -0.07925039669382432, "reference_mean_power": 0.00010203935243697765, "output_mean_power": 0.00010019421664597602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 7.765913024651851e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5071474654741866e-05, "output_power": 2.5383871627076892e-05, "reference_active": true, "delta_db": 0.0537798370112005}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067872704197783e-06, "output_power": 4.9302492814048065e-06, "reference_active": true, "delta_db": -0.901675828303838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0845180908068795e-07, "output_power": 7.033101836131385e-08, "reference_active": true, "delta_db": -1.880898950146444}], "active_band_count": 3, "active_band_mae_db": 0.9454515384871608}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 3.774086107813168e-05, "log_rms_mae_db": 1.7265195454617621, "rms_envelope_nrmse": 0.2684037485990322, "envelope_correlation": 0.9505497120187678, "rms_gain_error_db": -0.0313894858541586, "reference_mean_power": 4.068228265632881e-05, "output_mean_power": 4.03893035332094e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.809144227115543e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.649297963098168e-06, "output_power": 8.507514043494567e-06, "reference_active": true, "delta_db": -0.07178183882998329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.584218334576925e-06, "output_power": 3.193067177323371e-06, "reference_active": true, "delta_db": 0.9187885213670195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3501889520111653e-06, "output_power": 1.8539827667329513e-06, "reference_active": true, "delta_db": 1.3771114291259965}], "active_band_count": 3, "active_band_mae_db": 0.7892272631076663}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.332899304538301e-05, "log_rms_mae_db": 1.1720017154760627, "rms_envelope_nrmse": 0.15949525089242123, "envelope_correlation": 0.9814238415167001, "rms_gain_error_db": -0.031390363643301764, "reference_mean_power": 4.0907280378011884e-05, "output_mean_power": 4.061267269390121e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 2.6173037083412648e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.99444690144364e-06, "output_power": 8.425784907035724e-06, "reference_active": true, "delta_db": 0.22821945978554575}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1581434874555947e-06, "output_power": 1.5383266833469782e-06, "reference_active": true, "delta_db": 1.2328620388243015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.731926648355553e-08, "output_power": 1.080923492594672e-07, "reference_active": true, "delta_db": 0.9268487701545801}], "active_band_count": 3, "active_band_mae_db": 0.7959767562548091}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 1.8891386922351935e-05, "log_rms_mae_db": 1.540683503229357, "rms_envelope_nrmse": 0.2015839426087963, "envelope_correlation": 0.9726043984763217, "rms_gain_error_db": 0.0745296407644962, "reference_mean_power": 6.700122358829966e-06, "output_mean_power": 6.816095991740798e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 5.001365064934297e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.184828883572986e-06, "output_power": 2.2127958277061316e-06, "reference_active": true, "delta_db": 0.05523915321003567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0698658461156927e-06, "output_power": 1.1193592198406086e-06, "reference_active": true, "delta_db": 0.19640157241012704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.241288004170751e-07, "output_power": 3.7365289668426163e-07, "reference_active": true, "delta_db": -1.4696967105479495}], "active_band_count": 3, "active_band_mae_db": 0.5737791453893707}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 1.2324616345992443e-05, "log_rms_mae_db": 0.9534256032371117, "rms_envelope_nrmse": 0.10345399823793064, "envelope_correlation": 0.9920959722654883, "rms_gain_error_db": 0.0745600272977409, "reference_mean_power": 6.7368909993733856e-06, "output_mean_power": 6.853549019806152e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 4.500371991965458e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.395916806498352e-06, "output_power": 2.403683830445339e-06, "reference_active": true, "delta_db": 0.01405608024785599}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.187368503320664e-07, "output_power": 5.977903968467603e-07, "reference_active": true, "delta_db": -0.14957048149831353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.878047962583857e-08, "output_power": 2.8118984559101544e-08, "reference_active": true, "delta_db": -2.3924643286285625}], "active_band_count": 3, "active_band_mae_db": 0.8520302967915773}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8355861494537891, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1ad9a763321749b41d110e69e46779e1a91d82f5c0884bdaa11f3a3b397a97dc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9630273274316264, "standard_deviation": 0.007172063303082221, "minimum": 0.9505497120187678, "maximum": 0.9710895986346388, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7416571017409563, "standard_deviation": 0.02730950924405162, "minimum": 1.6967556482727781, "maximum": 1.7854202826123609, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7410106710758206, "standard_deviation": 0.37800105688868596, "minimum": 0.3427923643331186, "maximum": 1.3941752631686029, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23212025905781944, "standard_deviation": 0.02221377471530427, "minimum": 0.20606638891382217, "maximum": 0.2684037485990322, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9872278252621305, "standard_deviation": 0.0031984009131925347, "minimum": 0.9814238415167001, "maximum": 0.9910163050563635, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2003493242560894, "standard_deviation": 0.032787674846487216, "minimum": 1.161472792144096, "maximum": 1.2613470627263512, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8820795301030802, "standard_deviation": 0.3449389856257907, "minimum": 0.46590541573909666, "maximum": 1.4986750520585639, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13729061654756314, "standard_deviation": 0.01839144633650708, "minimum": 0.1113043233052089, "maximum": 0.16229283424927282, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9668755217397988, "standard_deviation": 0.005149916618531735, "minimum": 0.9573933927372971, "maximum": 0.9726500328337688, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5208267105096733, "standard_deviation": 0.025365362226701483, "minimum": 1.4816385767838283, "maximum": 1.5443834988282081, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9687764080313519, "standard_deviation": 0.35255298475237623, "minimum": 0.5737791453893707, "maximum": 1.5090048200680943, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21985319028494682, "standard_deviation": 0.015889218421637753, "minimum": 0.20143665102214536, "maximum": 0.24828256926256337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9888752013651553, "standard_deviation": 0.0036277846211166275, "minimum": 0.9823918949810037, "maximum": 0.9923604687756478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9708688075157623, "standard_deviation": 0.017919353774074005, "minimum": 0.9452978192313888, "maximum": 0.9978032699594926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3607832666062225, "standard_deviation": 0.5449802700124399, "minimum": 0.5891987120967729, "maximum": 2.324505299121714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12249114748326295, "standard_deviation": 0.018932601950410795, "minimum": 0.10107104041665968, "maximum": 0.15560955181487998, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9878088533947661, "standard_deviation": 0.0015831782693258834, "minimum": 0.9851169317233774, "maximum": 0.9902086232878394, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6785815489137637, "standard_deviation": 0.025264891096240676, "minimum": 1.644734034826233, "maximum": 1.7195234798450751, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3437952441758102, "standard_deviation": 0.15593513217997707, "minimum": 0.13895859579432362, "maximum": 0.5971129380603316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12403670453262547, "standard_deviation": 0.008081422048298986, "minimum": 0.1111108523030687, "maximum": 0.13668058648463832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961938742462018, "standard_deviation": 0.0008602776262922768, "minimum": 0.9948326444253899, "maximum": 0.9976639017236126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.074747240533811, "standard_deviation": 0.04276347004076679, "minimum": 0.9987684172657948, "maximum": 1.150327523699216, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6576294656149009, "standard_deviation": 0.340653668686019, "minimum": 0.26974068507037835, "maximum": 1.157998204197731, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06727087950134236, "standard_deviation": 0.007875962406134853, "minimum": 0.05241795557378091, "maximum": 0.07837204627757156, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21654936050986856, "standard_deviation": 0.08521092807395333, "minimum": 0.12016698597216639, "maximum": 0.36695508884446165, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.799649916354086, "standard_deviation": 0.004575442865531644, "minimum": 5.794291528239728, "maximum": 5.8068391506473525, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.1476037693140595, "standard_deviation": 0.005752500658637188, "minimum": 0.13977720017876039, "maximum": 0.15511349312736902, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0776171325231851, "standard_deviation": 0.06446663924717821, "minimum": -0.15004319548449202, "maximum": 0.036324582279961576, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9897813413539901, "standard_deviation": 0.0009105821425820831, "minimum": 0.9881232721079415, "maximum": 0.9910301726895997, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2073682718469794, "standard_deviation": 0.03717615422026571, "minimum": 1.1517101767871307, "maximum": 1.2697302748712211, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3254701335027701, "standard_deviation": 0.0898511848700367, "minimum": 0.20306650590033737, "maximum": 0.4898359764768246, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10210467229594442, "standard_deviation": 0.004582591574655668, "minimum": 0.0956472566962789, "maximum": 0.11004872382321389, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968057188224904, "standard_deviation": 0.0005696464446242233, "minimum": 0.9959505049379858, "maximum": 0.9977233374371214, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6870425271954521, "standard_deviation": 0.024693395802420953, "minimum": 0.6484181582248116, "maximum": 0.7235401555528739, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5168590433757626, "standard_deviation": 0.18377790389292467, "minimum": 0.21427613075592558, "maximum": 0.7650293313249148, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055051869715472865, "standard_deviation": 0.004673177558323712, "minimum": 0.046762829108229195, "maximum": 0.061305079789369096, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5683-32879-0009", "language": "en", "style": "normal", "speaker": "librispeech_5683", "source_utterance_id": "5683-32879-0009", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 82160, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5683-32879-0009.wav", "source_file_sha256": "139752a62d60c24aa2e108a04cfacc51815cb9121ea79d6a4c2920159b0f48df", "decoded_pcm_sha256": "c0eff412ac130cf1ccdb29b7cbb8149ed148614513d6a355a1677d911d2c6b5f", "original_source_sha256": "c66196c0436cce295ac3ae7695f4904f87672f1d63d89b8810937a146378411a", "corpus_entry_id": "librispeech_5683-32879-0009"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.7318692207336426, "predicted_mos": 3.7318692207336426, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "c3dbf96155daa7a77fd55580c3af7272825ffcce906d44f210d49fe8d41cf34b", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6447827349984436, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5683-32879-0009", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c3dbf96155daa7a77fd55580c3af7272825ffcce906d44f210d49fe8d41cf34b", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "b74b0dd3e4e57aa921833f8a2e77bec268aa0de7494ce4760b2cad80bb0ffb8c"}, "rows": [{"case": "librispeech_5683-32879-0009", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.4208288192749023, "predicted_mos": 2.4208288192749023, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "d19e8e5ce89e82f4fec0871a0f45796ac3ae3b853f38113d2737cabb125cf82c", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.12484861700068, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5683-32879-0009", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d19e8e5ce89e82f4fec0871a0f45796ac3ae3b853f38113d2737cabb125cf82c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8367733185734033, "全帯域_10ms_rms_envelope_nrmse": 0.061177904091500056, "全帯域_10ms_envelope_correlation": 0.9969145235177326, "全帯域_10ms_modulation_mae_db": 0.22620243409623045, "全帯域_40ms_log_rms_mae_db": 0.4597462445688198, "全帯域_40ms_rms_envelope_nrmse": 0.032866757436945435, "全帯域_40ms_envelope_correlation": 0.9991452231227037, "全帯域_40ms_modulation_mae_db": 0.40481500912474483, "300–1000 Hz_10ms_log_rms_mae_db": 1.2393327887469985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07882146520885722, "300–1000 Hz_10ms_envelope_correlation": 0.9952473935399357, "300–1000 Hz_10ms_modulation_mae_db": 0.691026604682488, "300–1000 Hz_40ms_log_rms_mae_db": 0.7895648919761545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04279395117146625, "300–1000 Hz_40ms_envelope_correlation": 0.9985466681576834, "300–1000 Hz_40ms_modulation_mae_db": 0.9223737475029433, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2430948020372925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06946552279116931, "1000–3000 Hz_10ms_envelope_correlation": 0.9972585230008831, "1000–3000 Hz_10ms_modulation_mae_db": 0.4160871515755202, "1000–3000 Hz_40ms_log_rms_mae_db": 0.766459279927913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.043955428427017634, "1000–3000 Hz_40ms_envelope_correlation": 0.9990402716955993, "1000–3000 Hz_40ms_modulation_mae_db": 0.26656703388331165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.435683203889365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17705121753217762, "3000–6000 Hz_10ms_envelope_correlation": 0.9792790929766595, "3000–6000 Hz_10ms_modulation_mae_db": 0.5382362419318246, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8851275674899456, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09472504147823176, "3000–6000 Hz_40ms_envelope_correlation": 0.9941785659436999, "3000–6000 Hz_40ms_modulation_mae_db": 0.49291096213390045, "mr_stft_db": 5.481321814444171, "band_mae_db": 0.15671249406637036, "rms_gain_error_db": -0.09158494072404917, "output_peak": 0.744133231351029}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8123871379212411, "全帯域_10ms_rms_envelope_nrmse": 0.05912590319412075, "全帯域_10ms_envelope_correlation": 0.9970780061192257, "全帯域_10ms_modulation_mae_db": 0.38417847911536035, "全帯域_40ms_log_rms_mae_db": 0.4610697141661643, "全帯域_40ms_rms_envelope_nrmse": 0.03282649443823875, "全帯域_40ms_envelope_correlation": 0.9990843065644035, "全帯域_40ms_modulation_mae_db": 0.4265018641654242, "300–1000 Hz_10ms_log_rms_mae_db": 1.2504997002108613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06870775425123696, "300–1000 Hz_10ms_envelope_correlation": 0.9964011512735433, "300–1000 Hz_10ms_modulation_mae_db": 0.5968667140280512, "300–1000 Hz_40ms_log_rms_mae_db": 0.7974064854216035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04155007625677973, "300–1000 Hz_40ms_envelope_correlation": 0.9986143251222642, "300–1000 Hz_40ms_modulation_mae_db": 0.586401844349483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.268286512654656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08031122322641464, "1000–3000 Hz_10ms_envelope_correlation": 0.9961739575351924, "1000–3000 Hz_10ms_modulation_mae_db": 0.3306698885269511, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7564294890856127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04756319318855349, "1000–3000 Hz_40ms_envelope_correlation": 0.9987490926185317, "1000–3000 Hz_40ms_modulation_mae_db": 0.5961672360255509, "3000–6000 Hz_10ms_log_rms_mae_db": 1.420798010843189, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16793605973368889, "3000–6000 Hz_10ms_envelope_correlation": 0.9800823227235731, "3000–6000 Hz_10ms_modulation_mae_db": 0.8540417343946146, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8931017104496473, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09074334720473544, "3000–6000 Hz_40ms_envelope_correlation": 0.994235023000386, "3000–6000 Hz_40ms_modulation_mae_db": 1.079564483481321, "mr_stft_db": 5.477174831091396, "band_mae_db": 0.18638997520461495, "rms_gain_error_db": -0.04865898250918866, "output_peak": 0.7408399778334058}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7893524973037485, "全帯域_10ms_rms_envelope_nrmse": 0.05424191837654407, "全帯域_10ms_envelope_correlation": 0.9975266862053306, "全帯域_10ms_modulation_mae_db": 0.19068027651471567, "全帯域_40ms_log_rms_mae_db": 0.4187325031449032, "全帯域_40ms_rms_envelope_nrmse": 0.026483593283513056, "全帯域_40ms_envelope_correlation": 0.9993948450325122, "全帯域_40ms_modulation_mae_db": 0.13705354909466816, "300–1000 Hz_10ms_log_rms_mae_db": 1.2767676913561394, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06762624108581124, "300–1000 Hz_10ms_envelope_correlation": 0.9964976397234371, "300–1000 Hz_10ms_modulation_mae_db": 0.11152805235411556, "300–1000 Hz_40ms_log_rms_mae_db": 0.8547495386026146, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04364167217107307, "300–1000 Hz_40ms_envelope_correlation": 0.9984589709020687, "300–1000 Hz_40ms_modulation_mae_db": 0.2815045303904254, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2248431187000859, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07006737119412061, "1000–3000 Hz_10ms_envelope_correlation": 0.9970213595443914, "1000–3000 Hz_10ms_modulation_mae_db": 0.12778689752063055, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7233063700453229, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04143955016759764, "1000–3000 Hz_40ms_envelope_correlation": 0.9989653862942949, "1000–3000 Hz_40ms_modulation_mae_db": 0.04234656712081309, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4073561661703216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18234123777490535, "3000–6000 Hz_10ms_envelope_correlation": 0.9767571091806897, "3000–6000 Hz_10ms_modulation_mae_db": 0.6334280732330472, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8820534325989853, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09600580391566542, "3000–6000 Hz_40ms_envelope_correlation": 0.9931165375101142, "3000–6000 Hz_40ms_modulation_mae_db": 0.6164518299883447, "mr_stft_db": 5.483059451078326, "band_mae_db": 0.11660684398489923, "rms_gain_error_db": -0.03792800174824139, "output_peak": 0.8100131315571926}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8195581052570133, "全帯域_10ms_rms_envelope_nrmse": 0.056902357742437466, "全帯域_10ms_envelope_correlation": 0.9972788185048111, "全帯域_10ms_modulation_mae_db": 0.46601162510808697, "全帯域_40ms_log_rms_mae_db": 0.3991052883903448, "全帯域_40ms_rms_envelope_nrmse": 0.02772527262023023, "全帯域_40ms_envelope_correlation": 0.9993383105615197, "全帯域_40ms_modulation_mae_db": 0.4125560623828441, "300–1000 Hz_10ms_log_rms_mae_db": 1.2374092232903888, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07217035603833015, "300–1000 Hz_10ms_envelope_correlation": 0.9960437772463849, "300–1000 Hz_10ms_modulation_mae_db": 0.7435642599724864, "300–1000 Hz_40ms_log_rms_mae_db": 0.733188134092406, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03845068560029613, "300–1000 Hz_40ms_envelope_correlation": 0.9988533707595428, "300–1000 Hz_40ms_modulation_mae_db": 1.0756034733171267, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2240308721375939, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07005850412675108, "1000–3000 Hz_10ms_envelope_correlation": 0.9969966825945519, "1000–3000 Hz_10ms_modulation_mae_db": 0.4212682403423751, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6826732764759096, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.036140181828705896, "1000–3000 Hz_40ms_envelope_correlation": 0.9991841825962571, "1000–3000 Hz_40ms_modulation_mae_db": 0.36917618382596545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4354303680150748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17552511544711075, "3000–6000 Hz_10ms_envelope_correlation": 0.978119647153981, "3000–6000 Hz_10ms_modulation_mae_db": 0.5636653236395556, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8827400989110544, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09546726557445309, "3000–6000 Hz_40ms_envelope_correlation": 0.9934212511833717, "3000–6000 Hz_40ms_modulation_mae_db": 0.5692078184538277, "mr_stft_db": 5.472307391151798, "band_mae_db": 0.1784467179685557, "rms_gain_error_db": -0.04179658236305207, "output_peak": 0.8287433054018715}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8189608843685834, "全帯域_10ms_rms_envelope_nrmse": 0.06691954981529799, "全帯域_10ms_envelope_correlation": 0.99638500702184, "全帯域_10ms_modulation_mae_db": 0.40496817402469376, "全帯域_40ms_log_rms_mae_db": 0.438249847952693, "全帯域_40ms_rms_envelope_nrmse": 0.03765627381719919, "全帯域_40ms_envelope_correlation": 0.9989607144701271, "全帯域_40ms_modulation_mae_db": 0.35689091211372564, "300–1000 Hz_10ms_log_rms_mae_db": 1.222348298321971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07414953810294744, "300–1000 Hz_10ms_envelope_correlation": 0.9957994123460377, "300–1000 Hz_10ms_modulation_mae_db": 0.6730490607206213, "300–1000 Hz_40ms_log_rms_mae_db": 0.807353925130422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039922641948970196, "300–1000 Hz_40ms_envelope_correlation": 0.9987362451013446, "300–1000 Hz_40ms_modulation_mae_db": 0.822547648148103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.269459955210987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07466037184700083, "1000–3000 Hz_10ms_envelope_correlation": 0.9969639171618727, "1000–3000 Hz_10ms_modulation_mae_db": 0.42234898846889574, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7280056873799519, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04549410984841416, "1000–3000 Hz_40ms_envelope_correlation": 0.999146725367085, "1000–3000 Hz_40ms_modulation_mae_db": 0.6139942054605713, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5116981768503148, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19993512108752134, "3000–6000 Hz_10ms_envelope_correlation": 0.9729789648850803, "3000–6000 Hz_10ms_modulation_mae_db": 1.030584190281256, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9718078338797318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1269645793987286, "3000–6000 Hz_40ms_envelope_correlation": 0.9904024531725062, "3000–6000 Hz_40ms_modulation_mae_db": 1.0679497851061812, "mr_stft_db": 5.468088643506743, "band_mae_db": 0.21422521545231854, "rms_gain_error_db": -0.1364943791710586, "output_peak": 0.7595746201366471}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8173543380946731, "全帯域_10ms_rms_envelope_nrmse": 0.056554396048651055, "全帯域_10ms_envelope_correlation": 0.9974971393701864, "全帯域_10ms_modulation_mae_db": 0.15607505684206874, "全帯域_40ms_log_rms_mae_db": 0.426179104801167, "全帯域_40ms_rms_envelope_nrmse": 0.03257427836646122, "全帯域_40ms_envelope_correlation": 0.9992947701036997, "全帯域_40ms_modulation_mae_db": 0.2461306050142065, "300–1000 Hz_10ms_log_rms_mae_db": 1.230178163386853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07132963520299432, "300–1000 Hz_10ms_envelope_correlation": 0.9961055661784712, "300–1000 Hz_10ms_modulation_mae_db": 0.44496441438265383, "300–1000 Hz_40ms_log_rms_mae_db": 0.7843200308904152, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042464579104552584, "300–1000 Hz_40ms_envelope_correlation": 0.9985522497872119, "300–1000 Hz_40ms_modulation_mae_db": 0.6925191874275857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2627472361037615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.071619349773685, "1000–3000 Hz_10ms_envelope_correlation": 0.997235464352019, "1000–3000 Hz_10ms_modulation_mae_db": 0.36229785721321167, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7547603614613645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.046560228524145576, "1000–3000 Hz_40ms_envelope_correlation": 0.99907416983574, "1000–3000 Hz_40ms_modulation_mae_db": 0.35548400730765106, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4206299516884873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20119475789129987, "3000–6000 Hz_10ms_envelope_correlation": 0.9727237696197223, "3000–6000 Hz_10ms_modulation_mae_db": 0.7961731357102569, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9377464541515522, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12666296939280952, "3000–6000 Hz_40ms_envelope_correlation": 0.9906475805159873, "3000–6000 Hz_40ms_modulation_mae_db": 0.8606576984462406, "mr_stft_db": 5.478935129020684, "band_mae_db": 0.288500948304924, "rms_gain_error_db": -0.14552456060989596, "output_peak": 0.8252666548585595}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8170477508550263, "全帯域_10ms_rms_envelope_nrmse": 0.061339761513749265, "全帯域_10ms_envelope_correlation": 0.9968683055059644, "全帯域_10ms_modulation_mae_db": 0.5014883869752699, "全帯域_40ms_log_rms_mae_db": 0.4331493815289677, "全帯域_40ms_rms_envelope_nrmse": 0.03312887687728281, "全帯域_40ms_envelope_correlation": 0.9990882795422847, "全帯域_40ms_modulation_mae_db": 0.6695351221422924, "300–1000 Hz_10ms_log_rms_mae_db": 1.241894853462802, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07989644954780016, "300–1000 Hz_10ms_envelope_correlation": 0.9951247909963309, "300–1000 Hz_10ms_modulation_mae_db": 0.3948539504090171, "300–1000 Hz_40ms_log_rms_mae_db": 0.8092123687873757, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05141986274339445, "300–1000 Hz_40ms_envelope_correlation": 0.9978895776021852, "300–1000 Hz_40ms_modulation_mae_db": 0.8493998967577067, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2577942752712896, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07554711583935403, "1000–3000 Hz_10ms_envelope_correlation": 0.9966673457272388, "1000–3000 Hz_10ms_modulation_mae_db": 0.7066973173188825, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7563919574032972, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048763874978338485, "1000–3000 Hz_40ms_envelope_correlation": 0.9987089290442926, "1000–3000 Hz_40ms_modulation_mae_db": 0.9404609202123494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.440267403116887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17391296166204878, "3000–6000 Hz_10ms_envelope_correlation": 0.9786651312904309, "3000–6000 Hz_10ms_modulation_mae_db": 0.6548399332494451, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8998209276511208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09874602769637524, "3000–6000 Hz_40ms_envelope_correlation": 0.9926039170571632, "3000–6000 Hz_40ms_modulation_mae_db": 0.7358810726071893, "mr_stft_db": 5.468734942670981, "band_mae_db": 0.16256720426979318, "rms_gain_error_db": -0.07231784502999913, "output_peak": 0.7455078600007947}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8286427657499994, "全帯域_10ms_rms_envelope_nrmse": 0.05380681323621988, "全帯域_10ms_envelope_correlation": 0.9975790632949573, "全帯域_10ms_modulation_mae_db": 0.23099172170193794, "全帯域_40ms_log_rms_mae_db": 0.4200546247401616, "全帯域_40ms_rms_envelope_nrmse": 0.027745184028215197, "全帯域_40ms_envelope_correlation": 0.9993522592344718, "全帯域_40ms_modulation_mae_db": 0.4080688621510498, "300–1000 Hz_10ms_log_rms_mae_db": 1.1682982940397217, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07956420751529553, "300–1000 Hz_10ms_envelope_correlation": 0.9951584685768929, "300–1000 Hz_10ms_modulation_mae_db": 0.9152286355267517, "300–1000 Hz_40ms_log_rms_mae_db": 0.7846576229730158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0521437702919197, "300–1000 Hz_40ms_envelope_correlation": 0.9978050742597284, "300–1000 Hz_40ms_modulation_mae_db": 1.4893420770937602, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2604093238438394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06704489846721576, "1000–3000 Hz_10ms_envelope_correlation": 0.9973047875190117, "1000–3000 Hz_10ms_modulation_mae_db": 0.28571721163507446, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6800485113090462, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03430859799842217, "1000–3000 Hz_40ms_envelope_correlation": 0.9993421178389437, "1000–3000 Hz_40ms_modulation_mae_db": 0.482060386724779, "3000–6000 Hz_10ms_log_rms_mae_db": 1.443429137045378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17396754560555613, "3000–6000 Hz_10ms_envelope_correlation": 0.9785220456109179, "3000–6000 Hz_10ms_modulation_mae_db": 0.8159499648787079, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9034207252774723, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09968337933679411, "3000–6000 Hz_40ms_envelope_correlation": 0.992416756708772, "3000–6000 Hz_40ms_modulation_mae_db": 0.8003201790699102, "mr_stft_db": 5.47030053788476, "band_mae_db": 0.22627933362214214, "rms_gain_error_db": -0.05469087074692959, "output_peak": 0.8192428564388724}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 82160, "output_samples": 82160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.434442932398886, "spectral_convergence": 0.22345494583418812, "spectral_convergence_status": "ok", "amplitude_floor": 2.2676184880983433e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.357331976334417, "spectral_convergence": 0.21629748323974832, "spectral_convergence_status": "ok", "amplitude_floor": 1.9761500134157035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.652190534599209, "spectral_convergence": 0.2195645196854807, "spectral_convergence_status": "ok", "amplitude_floor": 1.298774161363914e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.367365175011865e-05, "output_power": 9.320603182599666e-05, "delta_db": 0.46855295405427605, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016048781148481498, "output_power": 0.0015921101468546828, "delta_db": -0.034689445642745446, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001327868811244639, "output_power": 0.0012967052175267106, "delta_db": -0.10313912077650741, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0026813637139198997, "output_power": 0.0025824449224967575, "delta_db": -0.16324660102497363, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003087131182449005, "output_power": 0.0003157842929911294, "delta_db": 0.09835440054042, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00020800689998023804, "output_power": 0.00020457308635472783, "delta_db": -0.0722924423592995, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15671249406637036, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09158494072404917, "pitch": null, "output_rms_dbfs": -22.157520005190847, "output_peak": 0.744133231351029, "reference_rms_dbfs": -22.065935064466796, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.481321814444171}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 82160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003747956436237944, "log_rms_mae_db": 0.8367733185734033, "rms_envelope_nrmse": 0.061177904091500056, "envelope_correlation": 0.9969145235177326, "rms_gain_error_db": -0.09158476308303576, "reference_mean_power": 0.006220557891926456, "output_mean_power": 0.0060907512065419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.814448626587233e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017308539978626828, "output_power": 0.0016443886193359169, "reference_active": true, "delta_db": -0.22255973119886666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0006400481816254359, "output_power": 0.0005974610835174595, "reference_active": true, "delta_db": -0.29903045938739786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4706618104713675e-05, "output_power": 4.635253915765371e-05, "reference_active": true, "delta_db": 0.15701711170242683}], "active_band_count": 3, "active_band_mae_db": 0.22620243409623045}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.00031163820648484283, "log_rms_mae_db": 0.4597462445688198, "rms_envelope_nrmse": 0.032866757436945435, "envelope_correlation": 0.9991452231227037, "rms_gain_error_db": -0.09157543149302244, "reference_mean_power": 0.006257088451954998, "output_mean_power": 0.006126532633776601}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 6.546091034740867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001677353066756132, "output_power": 0.0015907626912116273, "reference_active": true, "delta_db": -0.23019090304866907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00035132010859926165, "output_power": 0.0003267832427020622, "reference_active": true, "delta_db": -0.3144323032285778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3827329510863976e-06, "output_power": 2.780081660885648e-06, "reference_active": true, "delta_db": 0.6698218210969875}], "active_band_count": 3, "active_band_mae_db": 0.40481500912474483}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.00015152624573583172, "log_rms_mae_db": 1.2393327887469985, "rms_envelope_nrmse": 0.07882146520885722, "envelope_correlation": 0.9952473935399357, "rms_gain_error_db": -0.04804480759912587, "reference_mean_power": 0.001672933915423699, "output_mean_power": 0.0016545286999079221}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 1.623496524404657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006256720367801861, "output_power": 0.00061184590397818, "reference_active": true, "delta_db": -0.0970468846237488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015725509039345338, "output_power": 0.00012645635360356437, "reference_active": true, "delta_db": -0.9466405790276646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4941700779965039e-05, "output_power": 1.1788561133439034e-05, "reference_active": true, "delta_db": -1.0293923503960503}], "active_band_count": 3, "active_band_mae_db": 0.691026604682488}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0001385613366624899, "log_rms_mae_db": 0.7895648919761545, "rms_envelope_nrmse": 0.04279395117146625, "envelope_correlation": 0.9985466681576834, "rms_gain_error_db": -0.04804514651193765, "reference_mean_power": 0.0016827742497549935, "output_mean_power": 0.0016642606433790203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 1.5301470087365327e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005852793815004822, "output_power": 0.0005697229869208883, "reference_active": true, "delta_db": -0.11699482327686067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.151264176170612e-05, "output_power": 6.756980173473518e-05, "reference_active": true, "delta_db": -0.8147232390664148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3333886770336671e-06, "output_power": 8.738088617768529e-07, "reference_active": true, "delta_db": -1.8354031801655544}], "active_band_count": 3, "active_band_mae_db": 0.9223737475029433}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003048971769130071, "log_rms_mae_db": 1.2430948020372925, "rms_envelope_nrmse": 0.06946552279116931, "envelope_correlation": 0.9972585230008831, "rms_gain_error_db": -0.18187097268496502, "reference_mean_power": 0.0025898505670984154, "output_mean_power": 0.0024836340712074616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 4.739471399174337e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014413951986732973, "output_power": 0.0013594063311377792, "reference_active": true, "delta_db": -0.2543378285652151}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0004245147064825657, "output_power": 0.00038012128772441144, "reference_active": true, "delta_db": -0.47970548284046466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3891971817365186e-05, "output_power": 2.6895139926855444e-05, "reference_active": true, "delta_db": 0.5142181433208808}], "active_band_count": 3, "active_band_mae_db": 0.4160871515755202}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0002474836728453422, "log_rms_mae_db": 0.766459279927913, "rms_envelope_nrmse": 0.043955428427017634, "envelope_correlation": 0.9990402716955993, "rms_gain_error_db": -0.1818477300826445, "reference_mean_power": 0.002605032676611572, "output_mean_power": 0.0024982068928768226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 4.638343817264169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014427582782272192, "output_power": 0.0013539574256449284, "reference_active": true, "delta_db": -0.27588566431273365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020918188683306615, "output_power": 0.00018771751307407073, "reference_active": true, "delta_db": -0.4701928409847927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0082267328987986e-06, "output_power": 2.0331761359747043e-06, "reference_active": true, "delta_db": 0.05362259635240858}], "active_band_count": 3, "active_band_mae_db": 0.26656703388331165}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 8.624763428459013e-05, "log_rms_mae_db": 1.435683203889365, "rms_envelope_nrmse": 0.17705121753217762, "envelope_correlation": 0.9792790929766595, "rms_gain_error_db": 0.27730784709181067, "reference_mean_power": 0.00014375800668184303, "output_mean_power": 0.00015323671279798685}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.640530401894145e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.056434359733932e-05, "output_power": 2.2918986612761334e-05, "reference_active": true, "delta_db": 0.4708055930710332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.844428212989113e-05, "output_power": 1.722526226195025e-05, "reference_active": true, "delta_db": -0.29695913763066745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.64438001039378e-06, "output_power": 2.998674015278292e-06, "reference_active": true, "delta_db": -0.8469439950937732}], "active_band_count": 3, "active_band_mae_db": 0.5382362419318246}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 6.581694648038094e-05, "log_rms_mae_db": 0.8851275674899456, "rms_envelope_nrmse": 0.09472504147823176, "envelope_correlation": 0.9941785659436999, "rms_gain_error_db": 0.277291699406391, "reference_mean_power": 0.00014459942402382801, "output_mean_power": 0.00015413303603128715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 5.765239137842014e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0474191785118657e-05, "output_power": 2.2462602479847176e-05, "reference_active": true, "delta_db": 0.402533044008513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1083544144974636e-05, "output_power": 1.051665094955201e-05, "reference_active": true, "delta_db": -0.2280119557347728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.842442602727836e-07, "output_power": 2.3381529393546373e-07, "reference_active": true, "delta_db": -0.8481878866584157}], "active_band_count": 3, "active_band_mae_db": 0.49291096213390045}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8792984399017004, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "81c634c8472fe44aa705e6f13715380c7ac0da0f0656ea70bbfa88f7aee361a9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9969527546793951, "standard_deviation": 0.0003764667121738119, "minimum": 0.9961739575351924, "maximum": 0.9973047875190117, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2513332619949382, "standard_deviation": 0.01846002831315386, "minimum": 1.2240308721375939, "maximum": 1.269459955210987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3841091940751926, "standard_deviation": 0.16328868359495996, "minimum": 0.12778689752063055, "maximum": 0.7066973173188825, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0723467946582139, "standard_deviation": 0.0042493675029503485, "minimum": 0.06704489846721576, "maximum": 0.08031122322641464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9990263594113431, "standard_deviation": 0.00021478744551544398, "minimum": 0.9987089290442926, "maximum": 0.9993421178389437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7310093666360522, "standard_deviation": 0.03404410853202131, "minimum": 0.6800485113090462, "maximum": 0.766459279927913, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4582820675701239, "standard_deviation": 0.2684837686902127, "minimum": 0.04234656712081309, "maximum": 0.9404609202123494, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04302814562014938, "standard_deviation": 0.005330235612179833, "minimum": 0.03430859799842217, "maximum": 0.048763874978338485, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9771410104301319, "standard_deviation": 0.0028130478400660774, "minimum": 0.9727237696197223, "maximum": 0.9800823227235731, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4394115522023772, "standard_deviation": 0.03161993086168558, "minimum": 1.4073561661703216, "maximum": 1.5116981768503148, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7358648246648385, "standard_deviation": 0.16760670565601754, "minimum": 0.5382362419318246, "maximum": 1.030584190281256, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18148300209178858, "standard_deviation": 0.01243391073884919, "minimum": 0.16793605973368889, "maximum": 0.20119475789129987, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926277606365, "standard_deviation": 0.0014528306006910859, "minimum": 0.9904024531725062, "maximum": 0.994235023000386, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9069773438011887, "standard_deviation": 0.03183032188305123, "minimum": 0.8820534325989853, "maximum": 0.9718078338797318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7778679786608644, "standard_deviation": 0.21879021638958526, "minimum": 0.49291096213390045, "maximum": 1.079564483481321, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10362480174972415, "standard_deviation": 0.014563024433169014, "minimum": 0.09074334720473544, "maximum": 0.1269645793987286, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9957972749851292, "standard_deviation": 0.0005572466339760985, "minimum": 0.9951247909963309, "maximum": 0.9964976397234371, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.233341126601967, "standard_deviation": 0.03086217176274082, "minimum": 1.1682982940397217, "maximum": 1.2767676913561394, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5713852115095231, "standard_deviation": 0.24836233440430755, "minimum": 0.11152805235411556, "maximum": 0.9152286355267517, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07403320586915912, "standard_deviation": 0.004900432966972963, "minimum": 0.06762624108581124, "maximum": 0.07989644954780016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984320602115037, "standard_deviation": 0.00038152981322018404, "minimum": 0.9978050742597284, "maximum": 0.9988533707595428, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.795056624734251, "standard_deviation": 0.03380967200174943, "minimum": 0.733188134092406, "maximum": 0.8547495386026146, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8399615506233917, "standard_deviation": 0.35502762448241276, "minimum": 0.2815045303904254, "maximum": 1.4893420770937602, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.044048404911056514, "standard_deviation": 0.005052848541475989, "minimum": 0.03845068560029613, "maximum": 0.0521437702919197, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19121609160920228, "standard_deviation": 0.052081446087239816, "minimum": 0.11660684398489923, "maximum": 0.288500948304924, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.474990342606107, "standard_deviation": 0.005873516257729788, "minimum": 5.468088643506743, "maximum": 5.483059451078326, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7841652046972966, "standard_deviation": 0.039917440805575376, "minimum": 0.7408399778334058, "maximum": 0.8287433054018715, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07862452036280183, "standard_deviation": 0.04228397075673399, "minimum": -0.14552456060989596, "maximum": -0.03792800174824139, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.997140943692506, "standard_deviation": 0.0004118997730745083, "minimum": 0.99638500702184, "maximum": 0.9975790632949573, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.817509599765461, "standard_deviation": 0.013735461372906713, "minimum": 0.7893524973037485, "maximum": 0.8367733185734033, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32007451929729547, "standard_deviation": 0.13411684003879096, "minimum": 0.15607505684206874, "maximum": 0.5014883869752699, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.058758575502315065, "standard_deviation": 0.004350306984014339, "minimum": 0.05380681323621988, "maximum": 0.06691954981529799, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992073385789653, "standard_deviation": 0.0001580923374103231, "minimum": 0.9989607144701271, "maximum": 0.9993948450325122, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4320358386616526, "standard_deviation": 0.021009987785151008, "minimum": 0.3991052883903448, "maximum": 0.4610697141661643, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38269399827361944, "standard_deviation": 0.15393714629851452, "minimum": 0.13705354909466816, "maximum": 0.6695351221422924, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031375841358510737, "standard_deviation": 0.0037554314525034818, "minimum": 0.026483593283513056, "maximum": 0.03765627381719919, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0009", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.676558017730713, "predicted_mos": 2.676558017730713, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "5dd98c80db08e08eb064c3f085f5d409831a27d69397b8a06dcd4cda6fc2faf0", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6718712002038956, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8414619325297441, "全帯域_10ms_rms_envelope_nrmse": 0.06092958489200662, "全帯域_10ms_envelope_correlation": 0.9969540205927653, "全帯域_10ms_modulation_mae_db": 0.1791227538702589, "全帯域_40ms_log_rms_mae_db": 0.47193738004333213, "全帯域_40ms_rms_envelope_nrmse": 0.03305237162628369, "全帯域_40ms_envelope_correlation": 0.9991509625424244, "全帯域_40ms_modulation_mae_db": 0.3115330429949319, "300–1000 Hz_10ms_log_rms_mae_db": 1.1573939142666116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07917108689879519, "300–1000 Hz_10ms_envelope_correlation": 0.9951914735214773, "300–1000 Hz_10ms_modulation_mae_db": 0.7788327671701261, "300–1000 Hz_40ms_log_rms_mae_db": 0.7197271144688063, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04272971413265729, "300–1000 Hz_40ms_envelope_correlation": 0.9985339988032297, "300–1000 Hz_40ms_modulation_mae_db": 0.971801263094593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2783151241476374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06958016834162746, "1000–3000 Hz_10ms_envelope_correlation": 0.9972844827155779, "1000–3000 Hz_10ms_modulation_mae_db": 0.3682235760223705, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7663368452191973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.044755925585139626, "1000–3000 Hz_40ms_envelope_correlation": 0.9990344201517956, "1000–3000 Hz_40ms_modulation_mae_db": 0.28981046300511487, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4626505938706629, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1823809548195034, "3000–6000 Hz_10ms_envelope_correlation": 0.9777199378439595, "3000–6000 Hz_10ms_modulation_mae_db": 0.5579237333098033, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8964431558392216, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09772826057223247, "3000–6000 Hz_40ms_envelope_correlation": 0.993552545926919, "3000–6000 Hz_40ms_modulation_mae_db": 0.5620148026614796, "mr_stft_db": 5.458775621878374, "band_mae_db": 0.1466480291368909, "rms_gain_error_db": -0.09878187004866154, "output_peak": 0.7507033365936112}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8256704045170004, "全帯域_10ms_rms_envelope_nrmse": 0.06020776189105224, "全帯域_10ms_envelope_correlation": 0.996963840015331, "全帯域_10ms_modulation_mae_db": 0.36372822060882753, "全帯域_40ms_log_rms_mae_db": 0.470957258019218, "全帯域_40ms_rms_envelope_nrmse": 0.033238139654475266, "全帯域_40ms_envelope_correlation": 0.9990547116880721, "全帯域_40ms_modulation_mae_db": 0.38928467349356016, "300–1000 Hz_10ms_log_rms_mae_db": 1.1750482027538882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06958926333880715, "300–1000 Hz_10ms_envelope_correlation": 0.9962994341310395, "300–1000 Hz_10ms_modulation_mae_db": 0.6524191997103445, "300–1000 Hz_40ms_log_rms_mae_db": 0.6963967083764436, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04129401882447553, "300–1000 Hz_40ms_envelope_correlation": 0.9986250111991599, "300–1000 Hz_40ms_modulation_mae_db": 0.6275787987197818, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2981201270370417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.080653332423106, "1000–3000 Hz_10ms_envelope_correlation": 0.9961201311447034, "1000–3000 Hz_10ms_modulation_mae_db": 0.3253972468922089, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7650879431864365, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04742998847639582, "1000–3000 Hz_40ms_envelope_correlation": 0.9987337360026054, "1000–3000 Hz_40ms_modulation_mae_db": 0.6165766278805896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4803091823921757, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1690306915365717, "3000–6000 Hz_10ms_envelope_correlation": 0.9802776013792704, "3000–6000 Hz_10ms_modulation_mae_db": 0.9508530593814121, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9307049142993647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09476700642975058, "3000–6000 Hz_40ms_envelope_correlation": 0.9943807962469994, "3000–6000 Hz_40ms_modulation_mae_db": 1.2053382083543451, "mr_stft_db": 5.466153344183709, "band_mae_db": 0.18927000551686715, "rms_gain_error_db": -0.04610360445607932, "output_peak": 0.7371150916112285}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7765428039056224, "全帯域_10ms_rms_envelope_nrmse": 0.05512526734901039, "全帯域_10ms_envelope_correlation": 0.9974557203996535, "全帯域_10ms_modulation_mae_db": 0.21235291278629642, "全帯域_40ms_log_rms_mae_db": 0.40397931375584323, "全帯域_40ms_rms_envelope_nrmse": 0.026184201947814307, "全帯域_40ms_envelope_correlation": 0.9994232067840906, "全帯域_40ms_modulation_mae_db": 0.16692399481662556, "300–1000 Hz_10ms_log_rms_mae_db": 1.1844744400704714, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06589124010818724, "300–1000 Hz_10ms_envelope_correlation": 0.99667483474918, "300–1000 Hz_10ms_modulation_mae_db": 0.14202959407421129, "300–1000 Hz_40ms_log_rms_mae_db": 0.7332939661583218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041265162350018196, "300–1000 Hz_40ms_envelope_correlation": 0.9986218518644884, "300–1000 Hz_40ms_modulation_mae_db": 0.23790637375425283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2751542471878872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07010612016768077, "1000–3000 Hz_10ms_envelope_correlation": 0.9970816117556036, "1000–3000 Hz_10ms_modulation_mae_db": 0.14939761860779452, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7135469925300382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03987353343096642, "1000–3000 Hz_40ms_envelope_correlation": 0.9991232291883803, "1000–3000 Hz_40ms_modulation_mae_db": 0.1475539230544705, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4791388292748617, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1832906500741742, "3000–6000 Hz_10ms_envelope_correlation": 0.9765144310900716, "3000–6000 Hz_10ms_modulation_mae_db": 0.7245650833530545, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9191847668586469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09600847939658567, "3000–6000 Hz_40ms_envelope_correlation": 0.9931104439853853, "3000–6000 Hz_40ms_modulation_mae_db": 0.7257341217332813, "mr_stft_db": 5.471364377809334, "band_mae_db": 0.10527969161709155, "rms_gain_error_db": -0.05257789058263101, "output_peak": 0.7943064517412686}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8216546160675248, "全帯域_10ms_rms_envelope_nrmse": 0.05815607340300797, "全帯域_10ms_envelope_correlation": 0.9971608490967143, "全帯域_10ms_modulation_mae_db": 0.48570790122016444, "全帯域_40ms_log_rms_mae_db": 0.40153933715840534, "全帯域_40ms_rms_envelope_nrmse": 0.028905102743418898, "全帯域_40ms_envelope_correlation": 0.9992852373025884, "全帯域_40ms_modulation_mae_db": 0.4252081113143695, "300–1000 Hz_10ms_log_rms_mae_db": 1.150681287819471, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07444952542354658, "300–1000 Hz_10ms_envelope_correlation": 0.9957680773987702, "300–1000 Hz_10ms_modulation_mae_db": 0.7027608408075453, "300–1000 Hz_40ms_log_rms_mae_db": 0.6670117603413879, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03947867735267376, "300–1000 Hz_40ms_envelope_correlation": 0.9987689943219068, "300–1000 Hz_40ms_modulation_mae_db": 1.1059126496470448, "1000–3000 Hz_10ms_log_rms_mae_db": 1.215722501381306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06991621502578371, "1000–3000 Hz_10ms_envelope_correlation": 0.9970153028173615, "1000–3000 Hz_10ms_modulation_mae_db": 0.4619823922503897, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6580007834734398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03570715022074519, "1000–3000 Hz_40ms_envelope_correlation": 0.9992164516204368, "1000–3000 Hz_40ms_modulation_mae_db": 0.48837621459724817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4837199578997875, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18460856948401672, "3000–6000 Hz_10ms_envelope_correlation": 0.9760245131946129, "3000–6000 Hz_10ms_modulation_mae_db": 0.6660366019257736, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9012286557076421, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10458679565920705, "3000–6000 Hz_40ms_envelope_correlation": 0.9925569588341705, "3000–6000 Hz_40ms_modulation_mae_db": 0.7378535029189863, "mr_stft_db": 5.452192261343623, "band_mae_db": 0.17741000404112683, "rms_gain_error_db": -0.048313786436655196, "output_peak": 0.8261564835231769}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8326214843406542, "全帯域_10ms_rms_envelope_nrmse": 0.06891791483379407, "全帯域_10ms_envelope_correlation": 0.9961840990548816, "全帯域_10ms_modulation_mae_db": 0.3293971230231055, "全帯域_40ms_log_rms_mae_db": 0.4592655073314143, "全帯域_40ms_rms_envelope_nrmse": 0.03950003168347521, "全帯域_40ms_envelope_correlation": 0.9988725294161224, "全帯域_40ms_modulation_mae_db": 0.4261889732276447, "300–1000 Hz_10ms_log_rms_mae_db": 1.1357011677166982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07257658647394871, "300–1000 Hz_10ms_envelope_correlation": 0.9959678438147556, "300–1000 Hz_10ms_modulation_mae_db": 0.7632474348100589, "300–1000 Hz_40ms_log_rms_mae_db": 0.7062486370859714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03881760746727678, "300–1000 Hz_40ms_envelope_correlation": 0.9987946681581411, "300–1000 Hz_40ms_modulation_mae_db": 0.8477213893734387, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3090217178018102, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07896101337162353, "1000–3000 Hz_10ms_envelope_correlation": 0.9966575805849666, "1000–3000 Hz_10ms_modulation_mae_db": 0.48361732901284066, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7661767242469365, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04902132600124788, "1000–3000 Hz_40ms_envelope_correlation": 0.9990519839995108, "1000–3000 Hz_40ms_modulation_mae_db": 0.6968647014011351, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5658982696616168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2132832553719784, "3000–6000 Hz_10ms_envelope_correlation": 0.968666679963658, "3000–6000 Hz_10ms_modulation_mae_db": 1.0914345119513789, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0012273225976147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12996799868503778, "3000–6000 Hz_40ms_envelope_correlation": 0.9896439453864919, "3000–6000 Hz_40ms_modulation_mae_db": 1.1717732193795112, "mr_stft_db": 5.455358604154057, "band_mae_db": 0.20368783933384205, "rms_gain_error_db": -0.14713156297141416, "output_peak": 0.7616560937087792}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8109757934850522, "全帯域_10ms_rms_envelope_nrmse": 0.05607599707248997, "全帯域_10ms_envelope_correlation": 0.9975177727688028, "全帯域_10ms_modulation_mae_db": 0.1609040178043507, "全帯域_40ms_log_rms_mae_db": 0.4339954004509658, "全帯域_40ms_rms_envelope_nrmse": 0.032538366983210734, "全帯域_40ms_envelope_correlation": 0.9992688262477817, "全帯域_40ms_modulation_mae_db": 0.24109122490557064, "300–1000 Hz_10ms_log_rms_mae_db": 1.1623942496701403, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06928277454526356, "300–1000 Hz_10ms_envelope_correlation": 0.996315763675564, "300–1000 Hz_10ms_modulation_mae_db": 0.4364237031973854, "300–1000 Hz_40ms_log_rms_mae_db": 0.6940450165915173, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039714947854783064, "300–1000 Hz_40ms_envelope_correlation": 0.9987214219482372, "300–1000 Hz_40ms_modulation_mae_db": 0.676420779083621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2909905176920111, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07023337022070339, "1000–3000 Hz_10ms_envelope_correlation": 0.9973674222695599, "1000–3000 Hz_10ms_modulation_mae_db": 0.46841476994926706, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7639555144248876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04585166267172124, "1000–3000 Hz_40ms_envelope_correlation": 0.9991266147859463, "1000–3000 Hz_40ms_modulation_mae_db": 0.3424153070498668, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4549473506411867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20361086764540642, "3000–6000 Hz_10ms_envelope_correlation": 0.9724676840304225, "3000–6000 Hz_10ms_modulation_mae_db": 0.8633958285586928, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9320703419675327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1264994114593528, "3000–6000 Hz_40ms_envelope_correlation": 0.991370728053713, "3000–6000 Hz_40ms_modulation_mae_db": 0.9698568222335888, "mr_stft_db": 5.45664425484586, "band_mae_db": 0.2690414043114879, "rms_gain_error_db": -0.13670594588452464, "output_peak": 0.8061853936968842}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8209181531144619, "全帯域_10ms_rms_envelope_nrmse": 0.059339902740020885, "全帯域_10ms_envelope_correlation": 0.997057513791474, "全帯域_10ms_modulation_mae_db": 0.483934786103875, "全帯域_40ms_log_rms_mae_db": 0.4310237535181666, "全帯域_40ms_rms_envelope_nrmse": 0.031531771927838244, "全帯域_40ms_envelope_correlation": 0.9991605208792269, "全帯域_40ms_modulation_mae_db": 0.593610203630893, "300–1000 Hz_10ms_log_rms_mae_db": 1.1675355912481813, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07623630125804264, "300–1000 Hz_10ms_envelope_correlation": 0.9955565186062896, "300–1000 Hz_10ms_modulation_mae_db": 0.4028036788978069, "300–1000 Hz_40ms_log_rms_mae_db": 0.7245316916040715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04859690763944925, "300–1000 Hz_40ms_envelope_correlation": 0.9981057708019294, "300–1000 Hz_40ms_modulation_mae_db": 0.7851115585981979, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2788244821144235, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07402500565435458, "1000–3000 Hz_10ms_envelope_correlation": 0.9967592132217188, "1000–3000 Hz_10ms_modulation_mae_db": 0.6557818173673342, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7586017362068407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.046232978376637986, "1000–3000 Hz_40ms_envelope_correlation": 0.9988069739840206, "1000–3000 Hz_40ms_modulation_mae_db": 0.8535672676046054, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4756179227105792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1749144651985191, "3000–6000 Hz_10ms_envelope_correlation": 0.978344333363103, "3000–6000 Hz_10ms_modulation_mae_db": 0.7025682002456676, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9397818433591018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10055077718173738, "3000–6000 Hz_40ms_envelope_correlation": 0.9923182650675865, "3000–6000 Hz_40ms_modulation_mae_db": 0.8051394469579228, "mr_stft_db": 5.451221128672663, "band_mae_db": 0.16678125693564202, "rms_gain_error_db": -0.06027727913487344, "output_peak": 0.7635530175921409}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8410348718373537, "全帯域_10ms_rms_envelope_nrmse": 0.053861296410652806, "全帯域_10ms_envelope_correlation": 0.9975754282912289, "全帯域_10ms_modulation_mae_db": 0.21033180976015098, "全帯域_40ms_log_rms_mae_db": 0.4268945594046844, "全帯域_40ms_rms_envelope_nrmse": 0.027551225211611224, "全帯域_40ms_envelope_correlation": 0.9993632453345489, "全帯域_40ms_modulation_mae_db": 0.39863909587940977, "300–1000 Hz_10ms_log_rms_mae_db": 1.1384173970381986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07968551603811683, "300–1000 Hz_10ms_envelope_correlation": 0.9951378107652352, "300–1000 Hz_10ms_modulation_mae_db": 1.0065165667742022, "300–1000 Hz_40ms_log_rms_mae_db": 0.7461538689399501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050989244133799586, "300–1000 Hz_40ms_envelope_correlation": 0.9979003916468536, "300–1000 Hz_40ms_modulation_mae_db": 1.5882314554824333, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2984813029808497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06632576836855548, "1000–3000 Hz_10ms_envelope_correlation": 0.9973635585428089, "1000–3000 Hz_10ms_modulation_mae_db": 0.24797899749126825, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6826497438205247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03362709891999569, "1000–3000 Hz_40ms_envelope_correlation": 0.9993715390571412, "1000–3000 Hz_40ms_modulation_mae_db": 0.5460758856225653, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5053735172899383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1825202497106748, "3000–6000 Hz_10ms_envelope_correlation": 0.9764285574485482, "3000–6000 Hz_10ms_modulation_mae_db": 0.8325156783708078, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9110366399986626, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10577083293114789, "3000–6000 Hz_40ms_envelope_correlation": 0.9914815353190147, "3000–6000 Hz_40ms_modulation_mae_db": 0.8331845860265218, "mr_stft_db": 5.458657877516394, "band_mae_db": 0.20585205511193835, "rms_gain_error_db": -0.056051446234994556, "output_peak": 0.822518272590954}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 82160, "output_samples": 82160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.415500811383032, "spectral_convergence": 0.2235301782039697, "spectral_convergence_status": "ok", "amplitude_floor": 2.2676184880983433e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.341328761255847, "spectral_convergence": 0.21526544968418762, "spectral_convergence_status": "ok", "amplitude_floor": 1.9761500134157035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.619497292996244, "spectral_convergence": 0.21884524421404664, "spectral_convergence_status": "ok", "amplitude_floor": 1.298774161363914e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.367365175011865e-05, "output_power": 9.245108312421334e-05, "delta_db": 0.4332328035120247, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016048781148481498, "output_power": 0.0015888979163260094, "delta_db": -0.043460592648268924, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001327868811244639, "output_power": 0.0013001229492376003, "delta_db": -0.09170746042498942, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0026813637139198997, "output_power": 0.0025740747583132945, "delta_db": -0.17734572033479656, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003087131182449005, "output_power": 0.0003140408974482481, "delta_db": 0.07431125303387526, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00020800689998023804, "output_power": 0.00020516095268152985, "delta_db": -0.05983034486739044, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1466480291368909, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09878187004866154, "pitch": null, "output_rms_dbfs": -22.164716934515457, "output_peak": 0.7507033365936112, "reference_rms_dbfs": -22.065935064466796, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.458775621878374}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 82160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003747956436237944, "log_rms_mae_db": 0.8414619325297441, "rms_envelope_nrmse": 0.06092958489200662, "envelope_correlation": 0.9969540205927653, "rms_gain_error_db": -0.0987817005122424, "reference_mean_power": 0.006220557891926456, "output_mean_power": 0.006080666239614097}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.814448626587233e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017308539978626828, "output_power": 0.0016462599020901972, "reference_active": true, "delta_db": -0.21762035310013655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0006400481816254359, "output_power": 0.0005995037084488589, "reference_active": true, "delta_db": -0.28420794157592694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4706618104713675e-05, "output_power": 4.434225974365762e-05, "reference_active": true, "delta_db": -0.035539966934713174}], "active_band_count": 3, "active_band_mae_db": 0.1791227538702589}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.00031163820648484283, "log_rms_mae_db": 0.47193738004333213, "rms_envelope_nrmse": 0.03305237162628369, "envelope_correlation": 0.9991509625424244, "rms_gain_error_db": -0.09877249372304395, "reference_mean_power": 0.006257088451954998, "output_mean_power": 0.0061163882447794245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 6.546091034740867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001677353066756132, "output_power": 0.0015926223994338793, "reference_active": true, "delta_db": -0.22511667490626133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00035132010859926165, "output_power": 0.00032890700780387547, "reference_active": true, "delta_db": -0.28629881521969935}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3827329510863976e-06, "output_power": 2.626598758519494e-06, "reference_active": true, "delta_db": 0.42318363885883503}], "active_band_count": 3, "active_band_mae_db": 0.3115330429949319}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.00015152624573583172, "log_rms_mae_db": 1.1573939142666116, "rms_envelope_nrmse": 0.07917108689879519, "envelope_correlation": 0.9951914735214773, "rms_gain_error_db": -0.034514397220014065, "reference_mean_power": 0.001672933915423699, "output_mean_power": 0.0016596914090555367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 1.623496524404657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006256720367801861, "output_power": 0.0006159811228132958, "reference_active": true, "delta_db": -0.0677934234661544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015725509039345338, "output_power": 0.00012758403711014642, "reference_active": true, "delta_db": -0.9080837227995673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4941700779965039e-05, "output_power": 1.0922899040595542e-05, "reference_active": true, "delta_db": -1.3606211552446565}], "active_band_count": 3, "active_band_mae_db": 0.7788327671701261}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0001385613366624899, "log_rms_mae_db": 0.7197271144688063, "rms_envelope_nrmse": 0.04272971413265729, "envelope_correlation": 0.9985339988032297, "rms_gain_error_db": -0.03451476659887811, "reference_mean_power": 0.0016827742497549935, "output_mean_power": 0.001669453707888166}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 1.5301470087365327e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005852793815004822, "output_power": 0.0005733189655015486, "reference_active": true, "delta_db": -0.08966916294546584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.151264176170612e-05, "output_power": 6.836206545357083e-05, "reference_active": true, "delta_db": -0.7640979273427773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3333886770336671e-06, "output_power": 8.294555032380228e-07, "reference_active": true, "delta_db": -2.061636698995536}], "active_band_count": 3, "active_band_mae_db": 0.971801263094593}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003048971769130071, "log_rms_mae_db": 1.2783151241476374, "rms_envelope_nrmse": 0.06958016834162746, "envelope_correlation": 0.9972844827155779, "rms_gain_error_db": -0.19574589194615533, "reference_mean_power": 0.0025898505670984154, "output_mean_power": 0.002475711973380296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 4.739471399174337e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014413951986732973, "output_power": 0.0013553968499425265, "reference_active": true, "delta_db": -0.2671659911118047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0004245147064825657, "output_power": 0.00038092708435418204, "reference_active": true, "delta_db": -0.47050887528024027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3891971817365186e-05, "output_power": 2.5998696842242123e-05, "reference_active": true, "delta_db": 0.3669958616750666}], "active_band_count": 3, "active_band_mae_db": 0.3682235760223705}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0002474836728453422, "log_rms_mae_db": 0.7663368452191973, "rms_envelope_nrmse": 0.044755925585139626, "envelope_correlation": 0.9990344201517956, "rms_gain_error_db": -0.19572457002469607, "reference_mean_power": 0.002605032676611572, "output_mean_power": 0.002490237210510251}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 4.638343817264169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014427582782272192, "output_power": 0.0013515642382728973, "reference_active": true, "delta_db": -0.283568828375111}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020918188683306615, "output_power": 0.00018843702245663154, "reference_active": true, "delta_db": -0.453578428231028}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0082267328987986e-06, "output_power": 1.9479792692848486e-06, "reference_active": true, "delta_db": -0.13228413240920553}], "active_band_count": 3, "active_band_mae_db": 0.28981046300511487}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 8.624763428459013e-05, "log_rms_mae_db": 1.4626505938706629, "rms_envelope_nrmse": 0.1823809548195034, "envelope_correlation": 0.9777199378439595, "rms_gain_error_db": 0.24233221239868835, "reference_mean_power": 0.00014375800668184303, "output_mean_power": 0.000152007586496934}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.640530401894145e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.056434359733932e-05, "output_power": 2.2957378489448778e-05, "reference_active": true, "delta_db": 0.47807442705576275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.844428212989113e-05, "output_power": 1.7060734308262205e-05, "reference_active": true, "delta_db": -0.33864037080199366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.64438001039378e-06, "output_power": 2.991699822265307e-06, "reference_active": true, "delta_db": -0.8570564020716533}], "active_band_count": 3, "active_band_mae_db": 0.5579237333098033}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 6.581694648038094e-05, "log_rms_mae_db": 0.8964431558392216, "rms_envelope_nrmse": 0.09772826057223247, "envelope_correlation": 0.993552545926919, "rms_gain_error_db": 0.24233775815713618, "reference_mean_power": 0.00014459942402382801, "output_mean_power": 0.00015289748397234046}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 5.765239137842014e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0474191785118657e-05, "output_power": 2.2614696565706466e-05, "reference_active": true, "delta_db": 0.43183997866461366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1083544144974636e-05, "output_power": 1.0296682622728004e-05, "reference_active": true, "delta_db": -0.31981328803565584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.842442602727836e-07, "output_power": 2.2922004298599149e-07, "reference_active": true, "delta_db": -0.9343911412841693}], "active_band_count": 3, "active_band_mae_db": 0.5620148026614796}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8808614470803755, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e43d2864e563acd91e37fca39e9a89e965755cff0ad3806aa47149d062dc2234", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9969561628815375, "standard_deviation": 0.00042912269238186087, "minimum": 0.9961201311447034, "maximum": 0.9973674222695599, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.280578752542871, "standard_deviation": 0.028759019228501593, "minimum": 1.215722501381306, "maximum": 1.3090217178018102, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3950992184491842, "standard_deviation": 0.15731392767333188, "minimum": 0.14939761860779452, "maximum": 0.6557818173673342, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07247512419667937, "standard_deviation": 0.004996184034469074, "minimum": 0.06632576836855548, "maximum": 0.080653332423106, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9990581185987296, "standard_deviation": 0.00020740911328716953, "minimum": 0.9987337360026054, "maximum": 0.9993715390571412, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7342945353885376, "standard_deviation": 0.0437205620970298, "minimum": 0.6580007834734398, "maximum": 0.7663368452191973, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4976550487769495, "standard_deviation": 0.23077526386039524, "minimum": 0.1475539230544705, "maximum": 0.8535672676046054, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04281245796035624, "standard_deviation": 0.005708126014532591, "minimum": 0.03362709891999569, "maximum": 0.04902132600124788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9758054672892058, "standard_deviation": 0.0036479535447782826, "minimum": 0.968666679963658, "maximum": 0.9802776013792704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.488456952967601, "standard_deviation": 0.0346671575999231, "minimum": 1.4549473506411867, "maximum": 1.5658982696616168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7986615871370738, "standard_deviation": 0.17091484223723433, "minimum": 0.5579237333098033, "maximum": 1.0914345119513789, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18670496298010558, "standard_deviation": 0.014618345704958725, "minimum": 0.1690306915365717, "maximum": 0.2132832553719784, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992301902352535, "standard_deviation": 0.0014740514224273224, "minimum": 0.9896439453864919, "maximum": 0.9943807962469994, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9289597050784734, "standard_deviation": 0.03294057807765572, "minimum": 0.8964431558392216, "maximum": 1.0012273225976147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8763618387832046, "standard_deviation": 0.2241998085482168, "minimum": 0.5620148026614796, "maximum": 1.2053382083543451, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10698494528938146, "standard_deviation": 0.013696224132517996, "minimum": 0.09476700642975058, "maximum": 0.12996799868503778, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9958639695827889, "standard_deviation": 0.0005536009965772707, "minimum": 0.9951378107652352, "maximum": 0.99667483474918, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1589557813229576, "standard_deviation": 0.017025666041748714, "minimum": 1.1357011677166982, "maximum": 1.1844744400704714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.61062922318021, "standard_deviation": 0.2703624046935706, "minimum": 0.14202959407421129, "maximum": 1.0065165667742022, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07336028676058848, "standard_deviation": 0.0049349973435035095, "minimum": 0.06589124010818724, "maximum": 0.07968551603811683, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9985090135929933, "standard_deviation": 0.0003282297588411372, "minimum": 0.9979003916468536, "maximum": 0.9987946681581411, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7109260954458088, "standard_deviation": 0.02521425108516085, "minimum": 0.6670117603413879, "maximum": 0.7461538689399501, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8550855334691705, "standard_deviation": 0.39344283638099853, "minimum": 0.23790637375425283, "maximum": 1.5882314554824333, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.042860784969391684, "standard_deviation": 0.004498811372070052, "minimum": 0.03881760746727678, "maximum": 0.050989244133799586, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18299628575061083, "standard_deviation": 0.04788206797662329, "minimum": 0.10527969161709155, "maximum": 0.2690414043114879, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.458795933800502, "standard_deviation": 0.006861360387046302, "minimum": 5.451221128672663, "maximum": 5.471364377809334, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7827742676322554, "standard_deviation": 0.0339567232460737, "minimum": 0.7371150916112285, "maximum": 0.8261564835231769, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08074292321872924, "standard_deviation": 0.04129688158480529, "minimum": -0.14713156297141416, "maximum": -0.04610360445607932, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9971086555013564, "standard_deviation": 0.00044892499661017553, "minimum": 0.9961840990548816, "maximum": 0.9975754282912289, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8213600074746767, "standard_deviation": 0.02088500394304041, "minimum": 0.7765428039056224, "maximum": 0.8414619325297441, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30318494064712864, "standard_deviation": 0.13247379979026172, "minimum": 0.1609040178043507, "maximum": 0.48570790122016444, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05907672482400437, "standard_deviation": 0.004694940962538734, "minimum": 0.053861296410652806, "maximum": 0.06891791483379407, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991974050243569, "standard_deviation": 0.00017757406373281216, "minimum": 0.9988725294161224, "maximum": 0.9994232067840906, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4374490637102537, "standard_deviation": 0.02767810596007953, "minimum": 0.40153933715840534, "maximum": 0.47193738004333213, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3690599150328756, "standard_deviation": 0.13020171550368775, "minimum": 0.16692399481662556, "maximum": 0.593610203630893, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031562651472265946, "standard_deviation": 0.004158697098264748, "minimum": 0.026184201947814307, "maximum": 0.03950003168347521, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0009", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.9020411968231201, "predicted_mos": 1.9020411968231201, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "d7f628ea9f43bf724e246d94ea8a8bdd5fa2b1b2a62a18ea9ca43fdc94018d2a", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.824011799879372, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8092730417961956, "全帯域_10ms_rms_envelope_nrmse": 0.06211961623186359, "全帯域_10ms_envelope_correlation": 0.9967672407700799, "全帯域_10ms_modulation_mae_db": 0.22241239285579337, "全帯域_40ms_log_rms_mae_db": 0.4343852692905401, "全帯域_40ms_rms_envelope_nrmse": 0.032017089265194476, "全帯域_40ms_envelope_correlation": 0.9991291336126479, "全帯域_40ms_modulation_mae_db": 0.2751632329928459, "300–1000 Hz_10ms_log_rms_mae_db": 1.5347426514600946, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.086032447148445, "300–1000 Hz_10ms_envelope_correlation": 0.9943951964216794, "300–1000 Hz_10ms_modulation_mae_db": 0.48646169686369145, "300–1000 Hz_40ms_log_rms_mae_db": 1.0671600783345263, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04977334502678403, "300–1000 Hz_40ms_envelope_correlation": 0.9980532489740885, "300–1000 Hz_40ms_modulation_mae_db": 0.21735457206310257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4165797088803573, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07405170718784027, "1000–3000 Hz_10ms_envelope_correlation": 0.9968044107240911, "1000–3000 Hz_10ms_modulation_mae_db": 0.4281699126827106, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8347455973107787, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.040859578624107766, "1000–3000 Hz_40ms_envelope_correlation": 0.999157823627774, "1000–3000 Hz_40ms_modulation_mae_db": 0.9005648399915428, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8347043730635286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1865159443738014, "3000–6000 Hz_10ms_envelope_correlation": 0.9751806902735338, "3000–6000 Hz_10ms_modulation_mae_db": 1.0735282282301533, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1684601663982248, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.102547353506989, "3000–6000 Hz_40ms_envelope_correlation": 0.9920349560595484, "3000–6000 Hz_40ms_modulation_mae_db": 1.1586366736611353, "mr_stft_db": 5.72311743616833, "band_mae_db": 0.15553754769321748, "rms_gain_error_db": -0.053189463336020176, "output_peak": 0.7473434478089619}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7504497564392771, "全帯域_10ms_rms_envelope_nrmse": 0.05653241948984097, "全帯域_10ms_envelope_correlation": 0.9973149919740159, "全帯域_10ms_modulation_mae_db": 0.15857497366097037, "全帯域_40ms_log_rms_mae_db": 0.3799692595968951, "全帯域_40ms_rms_envelope_nrmse": 0.02889780459821059, "全帯域_40ms_envelope_correlation": 0.9992802232778109, "全帯域_40ms_modulation_mae_db": 0.17937349390979365, "300–1000 Hz_10ms_log_rms_mae_db": 1.3913543656388168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08666753957619827, "300–1000 Hz_10ms_envelope_correlation": 0.9942770030960753, "300–1000 Hz_10ms_modulation_mae_db": 0.4931023463596202, "300–1000 Hz_40ms_log_rms_mae_db": 0.9720701773558558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04983585143431084, "300–1000 Hz_40ms_envelope_correlation": 0.9980474019067075, "300–1000 Hz_40ms_modulation_mae_db": 0.9187852644809101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3721815081899194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07425260977657354, "1000–3000 Hz_10ms_envelope_correlation": 0.9966257205865602, "1000–3000 Hz_10ms_modulation_mae_db": 0.537666140930686, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7472698516585612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03580115884898977, "1000–3000 Hz_40ms_envelope_correlation": 0.9992044745923663, "1000–3000 Hz_40ms_modulation_mae_db": 0.36173130801260855, "3000–6000 Hz_10ms_log_rms_mae_db": 1.845998268622014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18724839170062427, "3000–6000 Hz_10ms_envelope_correlation": 0.9751648474718319, "3000–6000 Hz_10ms_modulation_mae_db": 0.8595396443831067, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1570850780300268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10561927682346625, "3000–6000 Hz_40ms_envelope_correlation": 0.9919993336759275, "3000–6000 Hz_40ms_modulation_mae_db": 0.9781694850796875, "mr_stft_db": 5.730765350663025, "band_mae_db": 0.16647629130929673, "rms_gain_error_db": -0.04272119765136284, "output_peak": 0.8188442768934697}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7856299837049603, "全帯域_10ms_rms_envelope_nrmse": 0.05339822230429285, "全帯域_10ms_envelope_correlation": 0.9976394324801815, "全帯域_10ms_modulation_mae_db": 0.46422815905553083, "全帯域_40ms_log_rms_mae_db": 0.4180913075755437, "全帯域_40ms_rms_envelope_nrmse": 0.027074531037468128, "全帯域_40ms_envelope_correlation": 0.9994120628829906, "全帯域_40ms_modulation_mae_db": 0.47922084793078495, "300–1000 Hz_10ms_log_rms_mae_db": 1.3885941676867397, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07883161756738133, "300–1000 Hz_10ms_envelope_correlation": 0.9952947071217999, "300–1000 Hz_10ms_modulation_mae_db": 0.6790128576492637, "300–1000 Hz_40ms_log_rms_mae_db": 0.9494961341609298, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046834759343002776, "300–1000 Hz_40ms_envelope_correlation": 0.9982636322558132, "300–1000 Hz_40ms_modulation_mae_db": 1.0170311646474193, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3392623262559635, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06644658800274238, "1000–3000 Hz_10ms_envelope_correlation": 0.9974844680940999, "1000–3000 Hz_10ms_modulation_mae_db": 0.20671578983757097, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7776237114962914, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04041361586181017, "1000–3000 Hz_40ms_envelope_correlation": 0.9992011735338062, "1000–3000 Hz_40ms_modulation_mae_db": 0.5315547217269421, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8508877172465106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17822594838159433, "3000–6000 Hz_10ms_envelope_correlation": 0.9775027294930472, "3000–6000 Hz_10ms_modulation_mae_db": 0.9613257526352799, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1695218992036154, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09584309707451048, "3000–6000 Hz_40ms_envelope_correlation": 0.993012182898867, "3000–6000 Hz_40ms_modulation_mae_db": 0.9411303404788876, "mr_stft_db": 5.7060791027465045, "band_mae_db": 0.2113625629354274, "rms_gain_error_db": -0.07460976619260819, "output_peak": 0.7655483301216577}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.798477717893839, "全帯域_10ms_rms_envelope_nrmse": 0.05815977594927909, "全帯域_10ms_envelope_correlation": 0.9971725564846247, "全帯域_10ms_modulation_mae_db": 0.20032490262065272, "全帯域_40ms_log_rms_mae_db": 0.42064872545542215, "全帯域_40ms_rms_envelope_nrmse": 0.0297279248936536, "全帯域_40ms_envelope_correlation": 0.9992583653511337, "全帯域_40ms_modulation_mae_db": 0.2608772054579674, "300–1000 Hz_10ms_log_rms_mae_db": 1.4698143596648794, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07614303882972524, "300–1000 Hz_10ms_envelope_correlation": 0.9956744371140535, "300–1000 Hz_10ms_modulation_mae_db": 0.443462507783224, "300–1000 Hz_40ms_log_rms_mae_db": 0.9959444634423765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04384750603981022, "300–1000 Hz_40ms_envelope_correlation": 0.9985312549609177, "300–1000 Hz_40ms_modulation_mae_db": 0.8825628845840415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4417334120464522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08494365133270294, "1000–3000 Hz_10ms_envelope_correlation": 0.9957501872390089, "1000–3000 Hz_10ms_modulation_mae_db": 0.4307040632861215, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8212051945338327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05110224750346669, "1000–3000 Hz_40ms_envelope_correlation": 0.998591766803821, "1000–3000 Hz_40ms_modulation_mae_db": 0.39040471305716234, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8021881970774511, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16934974436278058, "3000–6000 Hz_10ms_envelope_correlation": 0.9798867294518951, "3000–6000 Hz_10ms_modulation_mae_db": 1.0477056205290973, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1043126424179834, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09410162668226872, "3000–6000 Hz_40ms_envelope_correlation": 0.9939756281972143, "3000–6000 Hz_40ms_modulation_mae_db": 1.2899937131484303, "mr_stft_db": 5.722885194217576, "band_mae_db": 0.24752213035862503, "rms_gain_error_db": -0.06125835695465085, "output_peak": 0.7621700883071939}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7904821744943771, "全帯域_10ms_rms_envelope_nrmse": 0.057377410391175235, "全帯域_10ms_envelope_correlation": 0.9972257672453424, "全帯域_10ms_modulation_mae_db": 0.36528019840168907, "全帯域_40ms_log_rms_mae_db": 0.43041619079041055, "全帯域_40ms_rms_envelope_nrmse": 0.031132603768496615, "全帯域_40ms_envelope_correlation": 0.9991494655958497, "全帯域_40ms_modulation_mae_db": 0.1356063214647098, "300–1000 Hz_10ms_log_rms_mae_db": 1.5153813549994617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08332959526512486, "300–1000 Hz_10ms_envelope_correlation": 0.9947200884437729, "300–1000 Hz_10ms_modulation_mae_db": 0.4834213237894242, "300–1000 Hz_40ms_log_rms_mae_db": 1.0415020432027293, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048733199826234974, "300–1000 Hz_40ms_envelope_correlation": 0.9981065781645468, "300–1000 Hz_40ms_modulation_mae_db": 0.8935540335077549, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3352397079152791, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07515415170001329, "1000–3000 Hz_10ms_envelope_correlation": 0.9965915359736398, "1000–3000 Hz_10ms_modulation_mae_db": 0.31829332759828793, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7821422810137153, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04340831490567108, "1000–3000 Hz_40ms_envelope_correlation": 0.998896136480019, "1000–3000 Hz_40ms_modulation_mae_db": 0.16277961901209, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8714706332513606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18937579165251295, "3000–6000 Hz_10ms_envelope_correlation": 0.974409298230437, "3000–6000 Hz_10ms_modulation_mae_db": 1.5247095493513187, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2068945691815924, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11684285308323059, "3000–6000 Hz_40ms_envelope_correlation": 0.989724684991745, "3000–6000 Hz_40ms_modulation_mae_db": 1.6895869356433078, "mr_stft_db": 5.697723689906296, "band_mae_db": 0.2177370932211381, "rms_gain_error_db": -0.0234720223824004, "output_peak": 0.7811616849889481}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7734999347997715, "全帯域_10ms_rms_envelope_nrmse": 0.0587969878612831, "全帯域_10ms_envelope_correlation": 0.9970963288332226, "全帯域_10ms_modulation_mae_db": 0.13234144209872797, "全帯域_40ms_log_rms_mae_db": 0.3628427576003352, "全帯域_40ms_rms_envelope_nrmse": 0.02996616881188016, "全帯域_40ms_envelope_correlation": 0.9992257056353122, "全帯域_40ms_modulation_mae_db": 0.28878189636548635, "300–1000 Hz_10ms_log_rms_mae_db": 1.391060221026399, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07932309223058924, "300–1000 Hz_10ms_envelope_correlation": 0.9952727291495802, "300–1000 Hz_10ms_modulation_mae_db": 0.25133770057138205, "300–1000 Hz_40ms_log_rms_mae_db": 0.931138217092122, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043713254499544336, "300–1000 Hz_40ms_envelope_correlation": 0.9985313272521238, "300–1000 Hz_40ms_modulation_mae_db": 0.2740020111499817, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3788530812952453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07385131896346028, "1000–3000 Hz_10ms_envelope_correlation": 0.9967346792668281, "1000–3000 Hz_10ms_modulation_mae_db": 0.32599168483897983, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8347256581439763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04139391110733716, "1000–3000 Hz_40ms_envelope_correlation": 0.9990225578128635, "1000–3000 Hz_40ms_modulation_mae_db": 0.4366003658919184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.822727870178898, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1721157848551462, "3000–6000 Hz_10ms_envelope_correlation": 0.9789644655812594, "3000–6000 Hz_10ms_modulation_mae_db": 1.044443751448115, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1109968979262155, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08609313966153842, "3000–6000 Hz_40ms_envelope_correlation": 0.9945246026354554, "3000–6000 Hz_40ms_modulation_mae_db": 1.057741705082549, "mr_stft_db": 5.695979979048104, "band_mae_db": 0.16482620653846825, "rms_gain_error_db": -0.03521419574452746, "output_peak": 0.8547237609799357}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7889861430651426, "全帯域_10ms_rms_envelope_nrmse": 0.06425361606515605, "全帯域_10ms_envelope_correlation": 0.9965650336746172, "全帯域_10ms_modulation_mae_db": 0.6777585339855609, "全帯域_40ms_log_rms_mae_db": 0.4102448578265984, "全帯域_40ms_rms_envelope_nrmse": 0.035762004370558376, "全帯域_40ms_envelope_correlation": 0.9989389239715074, "全帯域_40ms_modulation_mae_db": 0.8844509380841984, "300–1000 Hz_10ms_log_rms_mae_db": 1.5386336953156643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0842697925268484, "300–1000 Hz_10ms_envelope_correlation": 0.9946633862036128, "300–1000 Hz_10ms_modulation_mae_db": 0.8263944360720679, "300–1000 Hz_40ms_log_rms_mae_db": 1.0782265034907013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048558531858351024, "300–1000 Hz_40ms_envelope_correlation": 0.9981738350332883, "300–1000 Hz_40ms_modulation_mae_db": 1.4443470296107546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.352105332812706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08689413068639593, "1000–3000 Hz_10ms_envelope_correlation": 0.9955454954788796, "1000–3000 Hz_10ms_modulation_mae_db": 0.3738210249987645, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7742017351769468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05238772485574188, "1000–3000 Hz_40ms_envelope_correlation": 0.9985079226609533, "1000–3000 Hz_40ms_modulation_mae_db": 0.9257240812218317, "3000–6000 Hz_10ms_log_rms_mae_db": 1.831100934384301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18646308536583178, "3000–6000 Hz_10ms_envelope_correlation": 0.9753703785428813, "3000–6000 Hz_10ms_modulation_mae_db": 1.6051058508736542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1554503837956824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1115843390290496, "3000–6000 Hz_40ms_envelope_correlation": 0.9909276103374943, "3000–6000 Hz_40ms_modulation_mae_db": 1.7080087753402962, "mr_stft_db": 5.710060136539641, "band_mae_db": 0.22934006213851502, "rms_gain_error_db": -0.0807362321221739, "output_peak": 0.7559889626332914}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8020562226938216, "全帯域_10ms_rms_envelope_nrmse": 0.05660096384128035, "全帯域_10ms_envelope_correlation": 0.997315361061978, "全帯域_10ms_modulation_mae_db": 0.3656776665094701, "全帯域_40ms_log_rms_mae_db": 0.43233902096859195, "全帯域_40ms_rms_envelope_nrmse": 0.027770435544220318, "全帯域_40ms_envelope_correlation": 0.9993461644749135, "全帯域_40ms_modulation_mae_db": 0.5518314010285212, "300–1000 Hz_10ms_log_rms_mae_db": 1.3900839526896809, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07992245210703282, "300–1000 Hz_10ms_envelope_correlation": 0.9951667854355148, "300–1000 Hz_10ms_modulation_mae_db": 0.21753835866018964, "300–1000 Hz_40ms_log_rms_mae_db": 0.9789994729471974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04907284889359986, "300–1000 Hz_40ms_envelope_correlation": 0.9981353680757838, "300–1000 Hz_40ms_modulation_mae_db": 0.34897930792258713, "1000–3000 Hz_10ms_log_rms_mae_db": 1.346355705252138, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07026585970513487, "1000–3000 Hz_10ms_envelope_correlation": 0.9969841642556402, "1000–3000 Hz_10ms_modulation_mae_db": 0.5387648857781899, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7689874024715249, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.036013522858577944, "1000–3000 Hz_40ms_envelope_correlation": 0.9992023838056941, "1000–3000 Hz_40ms_modulation_mae_db": 0.6655296209186589, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7390339749095818, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15701594926067192, "3000–6000 Hz_10ms_envelope_correlation": 0.9828363925919263, "3000–6000 Hz_10ms_modulation_mae_db": 0.9805925383583464, "3000–6000 Hz_40ms_log_rms_mae_db": 1.073762099251749, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08596424705979097, "3000–6000 Hz_40ms_envelope_correlation": 0.9950310903642143, "3000–6000 Hz_40ms_modulation_mae_db": 1.182017664446999, "mr_stft_db": 5.702969255269963, "band_mae_db": 0.21370511259803493, "rms_gain_error_db": -0.05070489842687931, "output_peak": 0.7793710623110425}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 82160, "output_samples": 82160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.704448310398184, "spectral_convergence": 0.22496636829436523, "spectral_convergence_status": "ok", "amplitude_floor": 2.2676184880983433e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.690846896378113, "spectral_convergence": 0.21885495465468813, "spectral_convergence_status": "ok", "amplitude_floor": 1.9761500134157035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.774057101728694, "spectral_convergence": 0.21856751460210447, "spectral_convergence_status": "ok", "amplitude_floor": 1.298774161363914e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.367365175011865e-05, "output_power": 9.504805860123361e-05, "delta_db": 0.5535452742013588, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016048781148481498, "output_power": 0.0015974728274893205, "delta_db": -0.020085750920911485, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001327868811244639, "output_power": 0.001345785716130702, "delta_db": 0.058207440980934194, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0026813637139198997, "output_power": 0.002587317729729303, "delta_db": -0.1550596336087356, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003087131182449005, "output_power": 0.00030245766941199134, "delta_db": -0.08890483110077052, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00020800689998023804, "output_power": 0.00021077542666527683, "delta_db": 0.05742235534659426, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15553754769321748, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.053189463336020176, "pitch": null, "output_rms_dbfs": -22.119124527802818, "output_peak": 0.7473434478089619, "reference_rms_dbfs": -22.065935064466796, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.72311743616833}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 82160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003747956436237944, "log_rms_mae_db": 0.8092730417961956, "rms_envelope_nrmse": 0.06211961623186359, "envelope_correlation": 0.9967672407700799, "rms_gain_error_db": -0.05318887557019959, "reference_mean_power": 0.006220557891926456, "output_mean_power": 0.0061448381535639425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.814448626587233e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017308539978626828, "output_power": 0.001681209849496533, "reference_active": true, "delta_db": -0.12638509837685438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0006400481816254359, "output_power": 0.0005865873264470344, "reference_active": true, "delta_db": -0.37879992521202754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4706618104713675e-05, "output_power": 4.306917995475055e-05, "reference_active": true, "delta_db": -0.16205215497849815}], "active_band_count": 3, "active_band_mae_db": 0.22241239285579337}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.00031163820648484283, "log_rms_mae_db": 0.4343852692905401, "rms_envelope_nrmse": 0.032017089265194476, "envelope_correlation": 0.9991291336126479, "rms_gain_error_db": -0.053181158926588215, "reference_mean_power": 0.006257088451954998, "output_mean_power": 0.006180935027782534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 6.546091034740867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001677353066756132, "output_power": 0.001621301887409307, "reference_active": true, "delta_db": -0.14760598779587578}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00035132010859926165, "output_power": 0.00032500156695316443, "reference_active": true, "delta_db": -0.33817553449431914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3827329510863976e-06, "output_power": 2.2034568647225983e-06, "reference_active": true, "delta_db": -0.33970817668834286}], "active_band_count": 3, "active_band_mae_db": 0.2751632329928459}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.00015152624573583172, "log_rms_mae_db": 1.5347426514600946, "rms_envelope_nrmse": 0.086032447148445, "envelope_correlation": 0.9943951964216794, "rms_gain_error_db": 0.029916949247203364, "reference_mean_power": 0.001672933915423699, "output_mean_power": 0.0016844979261344188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 1.623496524404657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006256720367801861, "output_power": 0.0005959033447041566, "reference_active": true, "delta_db": -0.21170922393977262}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015725509039345338, "output_power": 0.00014103686202228258, "reference_active": true, "delta_db": -0.47272075953497633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4941700779965039e-05, "output_power": 1.786057949869934e-05, "reference_active": true, "delta_db": 0.7749551071163253}], "active_band_count": 3, "active_band_mae_db": 0.48646169686369145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0001385613366624899, "log_rms_mae_db": 1.0671600783345263, "rms_envelope_nrmse": 0.04977334502678403, "envelope_correlation": 0.9980532489740885, "rms_gain_error_db": 0.029916452206132094, "reference_mean_power": 0.0016827742497549935, "output_mean_power": 0.0016944060870027055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 1.5301470087365327e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005852793815004822, "output_power": 0.0005573038146884877, "reference_active": true, "delta_db": -0.21271209083400408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.151264176170612e-05, "output_power": 7.511657783126576e-05, "reference_active": true, "delta_db": -0.35489174437746074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3333886770336671e-06, "output_power": 1.3595736876647602e-06, "reference_active": true, "delta_db": 0.08445988097784284}], "active_band_count": 3, "active_band_mae_db": 0.21735457206310257}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003048971769130071, "log_rms_mae_db": 1.4165797088803573, "rms_envelope_nrmse": 0.07405170718784027, "envelope_correlation": 0.9968044107240911, "rms_gain_error_db": -0.1630776592396433, "reference_mean_power": 0.0025898505670984154, "output_mean_power": 0.0024944048389112907}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 4.739471399174337e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014413951986732973, "output_power": 0.0014036269787448218, "reference_active": true, "delta_db": -0.11531363995521267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0004245147064825657, "output_power": 0.0003483331218562067, "reference_active": true, "delta_db": -0.8589796806214846}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3891971817365186e-05, "output_power": 2.2244894585190047e-05, "reference_active": true, "delta_db": -0.31021641747143447}], "active_band_count": 3, "active_band_mae_db": 0.4281699126827106}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0002474836728453422, "log_rms_mae_db": 0.8347455973107787, "rms_envelope_nrmse": 0.040859578624107766, "envelope_correlation": 0.999157823627774, "rms_gain_error_db": -0.16306304097732385, "reference_mean_power": 0.002605032676611572, "output_mean_power": 0.0025090358759706105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 4.638343817264169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014427582782272192, "output_power": 0.001384692862279136, "reference_active": true, "delta_db": -0.17840121335411105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020918188683306615, "output_power": 0.00017363063361957676, "reference_active": true, "delta_db": -0.808977259480921}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0082267328987986e-06, "output_power": 1.3532596704510258e-06, "reference_active": true, "delta_db": -1.7143160471395964}], "active_band_count": 3, "active_band_mae_db": 0.9005648399915428}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 8.624763428459013e-05, "log_rms_mae_db": 1.8347043730635286, "rms_envelope_nrmse": 0.1865159443738014, "envelope_correlation": 0.9751806902735338, "rms_gain_error_db": -0.11590027962866864, "reference_mean_power": 0.00014375800668184303, "output_mean_power": 0.00013997227274429192}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.640530401894145e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.056434359733932e-05, "output_power": 1.949017196873677e-05, "reference_active": true, "delta_db": -0.23299180553228357}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.844428212989113e-05, "output_power": 1.4204990328728907e-05, "reference_active": true, "delta_db": -1.1342081433575515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.64438001039378e-06, "output_power": 2.378401566646016e-06, "reference_active": true, "delta_db": -1.8533847358006246}], "active_band_count": 3, "active_band_mae_db": 1.0735282282301533}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 6.581694648038094e-05, "log_rms_mae_db": 1.1684601663982248, "rms_envelope_nrmse": 0.102547353506989, "envelope_correlation": 0.9920349560595484, "rms_gain_error_db": -0.11588803225942378, "reference_mean_power": 0.00014459942402382801, "output_mean_power": 0.00014079192918127874}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 5.765239137842014e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0474191785118657e-05, "output_power": 1.898130077542357e-05, "reference_active": true, "delta_db": -0.32880796065794116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1083544144974636e-05, "output_power": 8.440045835832778e-06, "reference_active": true, "delta_db": -1.183338502025952}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.842442602727836e-07, "output_power": 1.8084868196026282e-07, "reference_active": true, "delta_db": -1.963763558299513}], "active_band_count": 3, "active_band_mae_db": 1.1586366736611353}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8617835238847705, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "aafafbd39dedb4e93960a816c202a50bcfebb6e6ea3ca39145d801e93df3dd5f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9965650827023435, "standard_deviation": 0.0006341393711779613, "minimum": 0.9955454954788796, "maximum": 0.9974844680940999, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3727888478310075, "standard_deviation": 0.038486770663417484, "minimum": 1.3352397079152791, "maximum": 1.4417334120464522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3950158537439139, "standard_deviation": 0.11335314307417313, "minimum": 0.20671578983757097, "maximum": 0.5387648857781899, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07573250216935794, "standard_deviation": 0.0069119781599647125, "minimum": 0.06644658800274238, "maximum": 0.08689413068639593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9989730299146622, "standard_deviation": 0.0002834531968580198, "minimum": 0.9985079226609533, "maximum": 0.9992044745923663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7926126789757034, "standard_deviation": 0.03306617753612268, "minimum": 0.7472698516585612, "maximum": 0.8347455973107787, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5468611587290944, "standard_deviation": 0.2676318504741575, "minimum": 0.16277961901209, "maximum": 0.9257240812218317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04267250932071281, "standard_deviation": 0.006182567628299309, "minimum": 0.03580115884898977, "maximum": 0.05238772485574188, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9774144414546014, "standard_deviation": 0.0029563376971975933, "minimum": 0.974409298230437, "maximum": 0.9828363925919263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8247639960917057, "standard_deviation": 0.04020886543083931, "minimum": 1.7390339749095818, "maximum": 1.8714706332513606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.137118866976134, "standard_deviation": 0.27316124609875725, "minimum": 0.8595396443831067, "maximum": 1.6051058508736542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17828882999412043, "standard_deviation": 0.011390794991035397, "minimum": 0.15701594926067192, "maximum": 0.18937579165251295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926537611450583, "standard_deviation": 0.0018282630134400838, "minimum": 0.989724684991745, "maximum": 0.9950310903642143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1433104670256364, "standard_deviation": 0.0432626110795149, "minimum": 1.073762099251749, "maximum": 1.2068945691815924, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2506606616101617, "standard_deviation": 0.29852712733981235, "minimum": 0.9411303404788876, "maximum": 1.7080087753402962, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0998244916151055, "standard_deviation": 0.011327408238967171, "minimum": 0.08596424705979097, "maximum": 0.11684285308323059, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949330416232611, "standard_deviation": 0.0004910950015629392, "minimum": 0.9942770030960753, "maximum": 0.9956744371140535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4524580960602171, "standard_deviation": 0.06962148519450533, "minimum": 1.3885941676867397, "maximum": 1.5386336953156643, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4850914034686079, "standard_deviation": 0.20075590113809846, "minimum": 0.21753835866018964, "maximum": 0.8263944360720679, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08181494690641813, "standard_deviation": 0.003789403685612626, "minimum": 0.07614303882972524, "maximum": 0.08666753957619827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982303308279088, "standard_deviation": 0.00019806512647259605, "minimum": 0.9980474019067075, "maximum": 0.9985313272521238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0018171362533046, "standard_deviation": 0.05458413663212131, "minimum": 0.931138217092122, "maximum": 1.0782265034907013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.749577033495819, "standard_deviation": 0.42948374098511644, "minimum": 0.21735457206310257, "maximum": 1.4443470296107546, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04754616211520475, "standard_deviation": 0.0025027149801128477, "minimum": 0.043713254499544336, "maximum": 0.04983585143431084, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20081337584909037, "standard_deviation": 0.03398189929453463, "minimum": 0.15553754769321748, "maximum": 0.24752213035862503, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.71119751806993, "standard_deviation": 0.012927821348515636, "minimum": 5.695979979048104, "maximum": 5.730765350663025, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7831439517555626, "standard_deviation": 0.03622336409702629, "minimum": 0.7473434478089619, "maximum": 0.8547237609799357, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.052738266601327896, "standard_deviation": 0.019284066155238568, "minimum": -0.0807362321221739, "maximum": -0.0234720223824004, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9971370890655078, "standard_deviation": 0.00033606919815897494, "minimum": 0.9965650336746172, "maximum": 0.9976394324801815, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7873568718609232, "standard_deviation": 0.01847630410399219, "minimum": 0.7504497564392771, "maximum": 0.8092730417961956, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32332478364854944, "standard_deviation": 0.1843755817740533, "minimum": 0.13234144209872797, "maximum": 0.6777585339855609, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.058404876516771405, "standard_deviation": 0.003403849508832966, "minimum": 0.05339822230429285, "maximum": 0.06425361606515605, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992175056002708, "standard_deviation": 0.00014644486210438926, "minimum": 0.9989389239715074, "maximum": 0.9994120628829906, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4111171736380421, "standard_deviation": 0.02620001334366939, "minimum": 0.3628427576003352, "maximum": 0.4343852692905401, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38191316715428847, "standard_deviation": 0.2470665814511607, "minimum": 0.1356063214647098, "maximum": 0.8844509380841984, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.030293570286210283, "standard_deviation": 0.002743355533770038, "minimum": 0.027074531037468128, "maximum": 0.035762004370558376, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0009", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.619720697402954, "predicted_mos": 2.619720697402954, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "321789a14eec1371d05e916b590bf791ddcc20b53a4d6745458f6e647d84db18", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.859212300274521, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8207043387967365, "全帯域_10ms_rms_envelope_nrmse": 0.06110483596081876, "全帯域_10ms_envelope_correlation": 0.9968555113048211, "全帯域_10ms_modulation_mae_db": 0.4108678794571579, "全帯域_40ms_log_rms_mae_db": 0.4237697496998479, "全帯域_40ms_rms_envelope_nrmse": 0.032909791513578715, "全帯域_40ms_envelope_correlation": 0.9990519435466677, "全帯域_40ms_modulation_mae_db": 0.4785031097623131, "300–1000 Hz_10ms_log_rms_mae_db": 1.186942420196567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07117816911363824, "300–1000 Hz_10ms_envelope_correlation": 0.9961394334900393, "300–1000 Hz_10ms_modulation_mae_db": 0.48808148443529525, "300–1000 Hz_40ms_log_rms_mae_db": 0.7451591359872429, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040711158858329266, "300–1000 Hz_40ms_envelope_correlation": 0.9986761021754442, "300–1000 Hz_40ms_modulation_mae_db": 0.8356149507703323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1759026102388503, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07093749981531541, "1000–3000 Hz_10ms_envelope_correlation": 0.996946531408541, "1000–3000 Hz_10ms_modulation_mae_db": 0.47384506354350703, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7367110075818212, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.039043270556941664, "1000–3000 Hz_40ms_envelope_correlation": 0.999089736407152, "1000–3000 Hz_40ms_modulation_mae_db": 0.3746969512976283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3882741155729306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19860297806296104, "3000–6000 Hz_10ms_envelope_correlation": 0.9740725723664707, "3000–6000 Hz_10ms_modulation_mae_db": 0.4102368863273309, "3000–6000 Hz_40ms_log_rms_mae_db": 0.899416268229275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12989927971598744, "3000–6000 Hz_40ms_envelope_correlation": 0.9886454741576759, "3000–6000 Hz_40ms_modulation_mae_db": 0.38807404417288477, "mr_stft_db": 5.388117445921079, "band_mae_db": 0.15020943327964298, "rms_gain_error_db": -0.02928159402576826, "output_peak": 0.777062475254475}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.862422816405605, "全帯域_10ms_rms_envelope_nrmse": 0.06861119891722006, "全帯域_10ms_envelope_correlation": 0.9960531265748513, "全帯域_10ms_modulation_mae_db": 0.6473649143732096, "全帯域_40ms_log_rms_mae_db": 0.4670475434038357, "全帯域_40ms_rms_envelope_nrmse": 0.036936737787015596, "全帯域_40ms_envelope_correlation": 0.9988403997725315, "全帯域_40ms_modulation_mae_db": 0.7190117218364391, "300–1000 Hz_10ms_log_rms_mae_db": 1.1579515225833332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07242277262815715, "300–1000 Hz_10ms_envelope_correlation": 0.9960083336090828, "300–1000 Hz_10ms_modulation_mae_db": 0.3170203675416298, "300–1000 Hz_40ms_log_rms_mae_db": 0.7394826973925336, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04595637451302799, "300–1000 Hz_40ms_envelope_correlation": 0.9983093407423238, "300–1000 Hz_40ms_modulation_mae_db": 0.4184265186900529, "1000–3000 Hz_10ms_log_rms_mae_db": 1.226180429597951, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09098515708127829, "1000–3000 Hz_10ms_envelope_correlation": 0.9951159617000873, "1000–3000 Hz_10ms_modulation_mae_db": 0.6013086503644565, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7135461368760775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0514889621237349, "1000–3000 Hz_40ms_envelope_correlation": 0.9985985620850665, "1000–3000 Hz_40ms_modulation_mae_db": 0.7564832608534499, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3210760313421408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18149169111331642, "3000–6000 Hz_10ms_envelope_correlation": 0.9765151105353475, "3000–6000 Hz_10ms_modulation_mae_db": 0.5676375854886945, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8460407750878793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10404558024524142, "3000–6000 Hz_40ms_envelope_correlation": 0.9919728152737259, "3000–6000 Hz_40ms_modulation_mae_db": 0.7106081255644056, "mr_stft_db": 5.405710917158558, "band_mae_db": 0.17882527781887328, "rms_gain_error_db": -0.0629296884499736, "output_peak": 0.8124108652195424}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7934920801870241, "全帯域_10ms_rms_envelope_nrmse": 0.05359072854444007, "全帯域_10ms_envelope_correlation": 0.9976926185284141, "全帯域_10ms_modulation_mae_db": 0.2897767079534654, "全帯域_40ms_log_rms_mae_db": 0.41953106343273355, "全帯域_40ms_rms_envelope_nrmse": 0.028484809379281228, "全帯域_40ms_envelope_correlation": 0.9994227695992808, "全帯域_40ms_modulation_mae_db": 0.3009598530102046, "300–1000 Hz_10ms_log_rms_mae_db": 1.107735747307806, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0699923351812543, "300–1000 Hz_10ms_envelope_correlation": 0.9962844265707356, "300–1000 Hz_10ms_modulation_mae_db": 0.29073528480177685, "300–1000 Hz_40ms_log_rms_mae_db": 0.675494269875995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03861091127683368, "300–1000 Hz_40ms_envelope_correlation": 0.9988414657498641, "300–1000 Hz_40ms_modulation_mae_db": 0.49044875061165855, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2138004770627704, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0705695873463203, "1000–3000 Hz_10ms_envelope_correlation": 0.9971493959334362, "1000–3000 Hz_10ms_modulation_mae_db": 0.43359023536828323, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7299546617094517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.041801068531608435, "1000–3000 Hz_40ms_envelope_correlation": 0.9991456475727054, "1000–3000 Hz_40ms_modulation_mae_db": 0.23975731806758901, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3177798516235768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15923629558971647, "3000–6000 Hz_10ms_envelope_correlation": 0.9820244028243046, "3000–6000 Hz_10ms_modulation_mae_db": 0.5413696586242523, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8247238320563177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09077572441570182, "3000–6000 Hz_40ms_envelope_correlation": 0.9939887955401623, "3000–6000 Hz_40ms_modulation_mae_db": 0.6473187627453149, "mr_stft_db": 5.3909673219678425, "band_mae_db": 0.22223103175383715, "rms_gain_error_db": -0.11147167989744572, "output_peak": 0.779070824328812}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8375260284122039, "全帯域_10ms_rms_envelope_nrmse": 0.0603336913623059, "全帯域_10ms_envelope_correlation": 0.9970070939486521, "全帯域_10ms_modulation_mae_db": 0.25825334451896137, "全帯域_40ms_log_rms_mae_db": 0.4801077427666948, "全帯域_40ms_rms_envelope_nrmse": 0.035520206247746725, "全帯域_40ms_envelope_correlation": 0.9989851281246641, "全帯域_40ms_modulation_mae_db": 0.38326234274197524, "300–1000 Hz_10ms_log_rms_mae_db": 1.1631525387229829, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08600437607643202, "300–1000 Hz_10ms_envelope_correlation": 0.9943531674790412, "300–1000 Hz_10ms_modulation_mae_db": 0.6499438145305616, "300–1000 Hz_40ms_log_rms_mae_db": 0.7394395693706932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04959699015952661, "300–1000 Hz_40ms_envelope_correlation": 0.9980316691125048, "300–1000 Hz_40ms_modulation_mae_db": 1.1995699986823212, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1419958503191283, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06875155208810152, "1000–3000 Hz_10ms_envelope_correlation": 0.9973456019402408, "1000–3000 Hz_10ms_modulation_mae_db": 0.24747325662364794, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7000306345462498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04378564440812106, "1000–3000 Hz_40ms_envelope_correlation": 0.9990790579968863, "1000–3000 Hz_40ms_modulation_mae_db": 0.3551554242380852, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2874871793143674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17003936404021228, "3000–6000 Hz_10ms_envelope_correlation": 0.9795030161869215, "3000–6000 Hz_10ms_modulation_mae_db": 0.5158246773106331, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8233652162105898, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10213183213456664, "3000–6000 Hz_40ms_envelope_correlation": 0.9920400969525763, "3000–6000 Hz_40ms_modulation_mae_db": 0.5917454778198478, "mr_stft_db": 5.38895974548124, "band_mae_db": 0.2047723393229025, "rms_gain_error_db": -0.09514043134185304, "output_peak": 0.7733420263664141}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8156183869126223, "全帯域_10ms_rms_envelope_nrmse": 0.05847133953378011, "全帯域_10ms_envelope_correlation": 0.9971173791619802, "全帯域_10ms_modulation_mae_db": 0.2647915606848387, "全帯域_40ms_log_rms_mae_db": 0.4404029044276615, "全帯域_40ms_rms_envelope_nrmse": 0.02957933004928339, "全帯域_40ms_envelope_correlation": 0.9992290544317189, "全帯域_40ms_modulation_mae_db": 0.33515048158259725, "300–1000 Hz_10ms_log_rms_mae_db": 1.1451765779602538, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07501203181787709, "300–1000 Hz_10ms_envelope_correlation": 0.9957180558382092, "300–1000 Hz_10ms_modulation_mae_db": 0.6401375951136435, "300–1000 Hz_40ms_log_rms_mae_db": 0.714705582681409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044335757769260215, "300–1000 Hz_40ms_envelope_correlation": 0.9984266884217753, "300–1000 Hz_40ms_modulation_mae_db": 0.732800033669544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2331743019449133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07480653512465327, "1000–3000 Hz_10ms_envelope_correlation": 0.9966006856178516, "1000–3000 Hz_10ms_modulation_mae_db": 0.2080273834377402, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7659892077798798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04171826613101544, "1000–3000 Hz_40ms_envelope_correlation": 0.9989487704122006, "1000–3000 Hz_40ms_modulation_mae_db": 0.5542091696161344, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2874653012213289, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16942458515326, "3000–6000 Hz_10ms_envelope_correlation": 0.9796041324800377, "3000–6000 Hz_10ms_modulation_mae_db": 0.6488843561027883, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8298110231557374, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09281304710106598, "3000–6000 Hz_40ms_envelope_correlation": 0.9934310362881305, "3000–6000 Hz_40ms_modulation_mae_db": 0.7781416348258237, "mr_stft_db": 5.401260744363313, "band_mae_db": 0.1927347623600282, "rms_gain_error_db": -0.015948752381327563, "output_peak": 0.777967086815879}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8555133487341643, "全帯域_10ms_rms_envelope_nrmse": 0.06206736740529035, "全帯域_10ms_envelope_correlation": 0.9967753539396079, "全帯域_10ms_modulation_mae_db": 0.22918387701011692, "全帯域_40ms_log_rms_mae_db": 0.4915191469175656, "全帯域_40ms_rms_envelope_nrmse": 0.03290497909567934, "全帯域_40ms_envelope_correlation": 0.9990857304309264, "全帯域_40ms_modulation_mae_db": 0.3420412687210333, "300–1000 Hz_10ms_log_rms_mae_db": 1.1914883533228866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07726513554477567, "300–1000 Hz_10ms_envelope_correlation": 0.9954249914974858, "300–1000 Hz_10ms_modulation_mae_db": 0.5991574236058602, "300–1000 Hz_40ms_log_rms_mae_db": 0.72694545953506, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041507149854662036, "300–1000 Hz_40ms_envelope_correlation": 0.9986193065414434, "300–1000 Hz_40ms_modulation_mae_db": 0.9785809473284152, "1000–3000 Hz_10ms_log_rms_mae_db": 1.201335653187128, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07749216529531158, "1000–3000 Hz_10ms_envelope_correlation": 0.9963968148037999, "1000–3000 Hz_10ms_modulation_mae_db": 0.11340252396844375, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7374507931319849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04312388396997562, "1000–3000 Hz_40ms_envelope_correlation": 0.9989380710218491, "1000–3000 Hz_40ms_modulation_mae_db": 0.48945691388486795, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3070894395773656, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.181210354817433, "3000–6000 Hz_10ms_envelope_correlation": 0.9781549053316737, "3000–6000 Hz_10ms_modulation_mae_db": 0.34624543490189635, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8295897240634694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11331102385870964, "3000–6000 Hz_40ms_envelope_correlation": 0.9911873888542082, "3000–6000 Hz_40ms_modulation_mae_db": 0.37921043484630373, "mr_stft_db": 5.398731731119825, "band_mae_db": 0.19843528282765321, "rms_gain_error_db": -0.062127949433436754, "output_peak": 0.798051072298783}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8424298495828823, "全帯域_10ms_rms_envelope_nrmse": 0.06366286536968781, "全帯域_10ms_envelope_correlation": 0.9965854066023083, "全帯域_10ms_modulation_mae_db": 0.48289580147435984, "全帯域_40ms_log_rms_mae_db": 0.44494517978536763, "全帯域_40ms_rms_envelope_nrmse": 0.030251554425028613, "全帯域_40ms_envelope_correlation": 0.9992049483282494, "全帯域_40ms_modulation_mae_db": 0.40143999185910345, "300–1000 Hz_10ms_log_rms_mae_db": 1.116746063798896, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08233708833094527, "300–1000 Hz_10ms_envelope_correlation": 0.994829779771433, "300–1000 Hz_10ms_modulation_mae_db": 0.981144166408515, "300–1000 Hz_40ms_log_rms_mae_db": 0.7171297745977409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05046230211694122, "300–1000 Hz_40ms_envelope_correlation": 0.9979839311294479, "300–1000 Hz_40ms_modulation_mae_db": 1.4284390001642142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2148973391084534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07080889155881617, "1000–3000 Hz_10ms_envelope_correlation": 0.9969359686200953, "1000–3000 Hz_10ms_modulation_mae_db": 0.5683176749759589, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7402591677079898, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.037687093950742684, "1000–3000 Hz_40ms_envelope_correlation": 0.9991223421092673, "1000–3000 Hz_40ms_modulation_mae_db": 0.5446453029737226, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2998633613673705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18402384632091104, "3000–6000 Hz_10ms_envelope_correlation": 0.9777254097157485, "3000–6000 Hz_10ms_modulation_mae_db": 0.19251821936726454, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8529423112856799, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10543016299405752, "3000–6000 Hz_40ms_envelope_correlation": 0.9927152841266955, "3000–6000 Hz_40ms_modulation_mae_db": 0.1819140945536941, "mr_stft_db": 5.379883528141732, "band_mae_db": 0.18852938225872662, "rms_gain_error_db": -0.03390868424092932, "output_peak": 0.7868244559348512}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8411636053575793, "全帯域_10ms_rms_envelope_nrmse": 0.05986918506094223, "全帯域_10ms_envelope_correlation": 0.996979779941216, "全帯域_10ms_modulation_mae_db": 0.31613770023756316, "全帯域_40ms_log_rms_mae_db": 0.40205622778312056, "全帯域_40ms_rms_envelope_nrmse": 0.03195628139466901, "全帯域_40ms_envelope_correlation": 0.999098509962949, "全帯域_40ms_modulation_mae_db": 0.41144702540125216, "300–1000 Hz_10ms_log_rms_mae_db": 1.2500991412242226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09024535173260272, "300–1000 Hz_10ms_envelope_correlation": 0.9937395767019894, "300–1000 Hz_10ms_modulation_mae_db": 0.5008034171073187, "300–1000 Hz_40ms_log_rms_mae_db": 0.7718170592944259, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056055967270459216, "300–1000 Hz_40ms_envelope_correlation": 0.9974537579240274, "300–1000 Hz_40ms_modulation_mae_db": 0.8596634472722466, "1000–3000 Hz_10ms_log_rms_mae_db": 1.155137018260536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06564857076631667, "1000–3000 Hz_10ms_envelope_correlation": 0.9973799825571338, "1000–3000 Hz_10ms_modulation_mae_db": 0.4555666414796004, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6684095466895731, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03526625324998556, "1000–3000 Hz_40ms_envelope_correlation": 0.9992305373681776, "1000–3000 Hz_40ms_modulation_mae_db": 0.502810615536027, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3045199743870992, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16389649830533998, "3000–6000 Hz_10ms_envelope_correlation": 0.9808806854142262, "3000–6000 Hz_10ms_modulation_mae_db": 0.47017127220927374, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8516848437644994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09853661786552635, "3000–6000 Hz_40ms_envelope_correlation": 0.992708133835296, "3000–6000 Hz_40ms_modulation_mae_db": 0.46418593708524813, "mr_stft_db": 5.39733697000532, "band_mae_db": 0.13023283413646333, "rms_gain_error_db": -0.0031496981294917886, "output_peak": 0.7840631635745992}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 82160, "output_samples": 82160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.302600965848613, "spectral_convergence": 0.21942125189163633, "spectral_convergence_status": "ok", "amplitude_floor": 2.2676184880983433e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.256483085496017, "spectral_convergence": 0.21708093889339222, "spectral_convergence_status": "ok", "amplitude_floor": 1.9761500134157035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.605268286418608, "spectral_convergence": 0.22092748372581802, "spectral_convergence_status": "ok", "amplitude_floor": 1.298774161363914e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.367365175011865e-05, "output_power": 9.409057377106476e-05, "delta_db": 0.5095739368500799, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016048781148481498, "output_power": 0.001583729434581957, "delta_db": -0.057610663480391656, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001327868811244639, "output_power": 0.001336294614531529, "delta_db": 0.027470477510513725, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0026813637139198997, "output_power": 0.0026432164961638373, "delta_db": -0.06222991852646125, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003087131182449005, "output_power": 0.0003009211456672118, "delta_db": -0.11102377891431234, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00020800689998023804, "output_power": 0.0002144927039127478, "delta_db": 0.13334782439609905, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15020943327964298, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.02928159402576826, "pitch": null, "output_rms_dbfs": -22.09521665849256, "output_peak": 0.777062475254475, "reference_rms_dbfs": -22.065935064466796, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.388117445921079}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 82160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003747956436237944, "log_rms_mae_db": 0.8207043387967365, "rms_envelope_nrmse": 0.06110483596081876, "envelope_correlation": 0.9968555113048211, "rms_gain_error_db": -0.029280500548534837, "reference_mean_power": 0.006220557891926456, "output_mean_power": 0.0061787594282716575}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.814448626587233e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017308539978626828, "output_power": 0.001721874538308342, "reference_active": true, "delta_db": -0.022589314428445126}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0006400481816254359, "output_power": 0.0005924307142415387, "reference_active": true, "delta_db": -0.3357510189849758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4706618104713675e-05, "output_power": 3.655489054038665e-05, "reference_active": true, "delta_db": -0.8742633049580528}], "active_band_count": 3, "active_band_mae_db": 0.4108678794571579}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.00031163820648484283, "log_rms_mae_db": 0.4237697496998479, "rms_envelope_nrmse": 0.032909791513578715, "envelope_correlation": 0.9990519435466677, "rms_gain_error_db": -0.0292734823964438, "reference_mean_power": 0.006257088451954998, "output_mean_power": 0.006215054568032577}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 6.546091034740867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001677353066756132, "output_power": 0.0016542696620457395, "reference_active": true, "delta_db": -0.06018181845410941}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00035132010859926165, "output_power": 0.000327667998743981, "reference_active": true, "delta_db": -0.3026897966340424}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3827329510863976e-06, "output_power": 1.8612795225675327e-06, "reference_active": true, "delta_db": -1.0726377141987873}], "active_band_count": 3, "active_band_mae_db": 0.4785031097623131}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.00015152624573583172, "log_rms_mae_db": 1.186942420196567, "rms_envelope_nrmse": 0.07117816911363824, "envelope_correlation": 0.9961394334900393, "rms_gain_error_db": -0.0003685464252156255, "reference_mean_power": 0.001672933915423699, "output_mean_power": 0.0016727919546850965}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 1.623496524404657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006256720367801861, "output_power": 0.0006121998455409155, "reference_active": true, "delta_db": -0.09453529758614458}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015725509039345338, "output_power": 0.00013114644927451576, "reference_active": true, "delta_db": -0.7884817580828372}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4941700779965039e-05, "output_power": 1.3070051393159243e-05, "reference_active": true, "delta_db": -0.5812273976369041}], "active_band_count": 3, "active_band_mae_db": 0.48808148443529525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0001385613366624899, "log_rms_mae_db": 0.7451591359872429, "rms_envelope_nrmse": 0.040711158858329266, "envelope_correlation": 0.9986761021754442, "rms_gain_error_db": -0.0003684158337074962, "reference_mean_power": 0.0016827742497549935, "output_mean_power": 0.0016826315045880998}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 1.5301470087365327e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005852793815004822, "output_power": 0.0005673091545300608, "reference_active": true, "delta_db": -0.13543433125930282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.151264176170612e-05, "output_power": 7.053303321765966e-05, "reference_active": true, "delta_db": -0.6283240777499998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3333886770336671e-06, "output_power": 8.925819876368566e-07, "reference_active": true, "delta_db": -1.7430864433016944}], "active_band_count": 3, "active_band_mae_db": 0.8356149507703323}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003048971769130071, "log_rms_mae_db": 1.1759026102388503, "rms_envelope_nrmse": 0.07093749981531541, "envelope_correlation": 0.996946531408541, "rms_gain_error_db": -0.08173730983025906, "reference_mean_power": 0.0025898505670984154, "output_mean_power": 0.0025415635609789132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 4.739471399174337e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014413951986732973, "output_power": 0.0014185919849209166, "reference_active": true, "delta_db": -0.06925569333974196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0004245147064825657, "output_power": 0.0003903602595345821, "reference_active": true, "delta_db": -0.36427142050136296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3891971817365186e-05, "output_power": 1.9030543257914364e-05, "reference_active": true, "delta_db": -0.9880080767894162}], "active_band_count": 3, "active_band_mae_db": 0.47384506354350703}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0002474836728453422, "log_rms_mae_db": 0.7367110075818212, "rms_envelope_nrmse": 0.039043270556941664, "envelope_correlation": 0.999089736407152, "rms_gain_error_db": -0.08172229706023916, "reference_mean_power": 0.002605032676611572, "output_mean_power": 0.0025564714417433145}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 4.638343817264169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014427582782272192, "output_power": 0.0014067894089321092, "reference_active": true, "delta_db": -0.10964484818738081}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020918188683306615, "output_power": 0.00019224356506270443, "reference_active": true, "delta_db": -0.3667226439236436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0082267328987986e-06, "output_power": 1.7299773129518518e-06, "reference_active": true, "delta_db": -0.6477233617818605}], "active_band_count": 3, "active_band_mae_db": 0.3746969512976283}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 8.624763428459013e-05, "log_rms_mae_db": 1.3882741155729306, "rms_envelope_nrmse": 0.19860297806296104, "envelope_correlation": 0.9740725723664707, "rms_gain_error_db": 0.2869998600206746, "reference_mean_power": 0.00014375800668184303, "output_mean_power": 0.00015357906820497632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.640530401894145e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.056434359733932e-05, "output_power": 2.0752945954243143e-05, "reference_active": true, "delta_db": 0.03964903442586592}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.844428212989113e-05, "output_power": 1.718786677336837e-05, "reference_active": true, "delta_db": -0.30639777990876255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.64438001039378e-06, "output_power": 2.9727423582217313e-06, "reference_active": true, "delta_db": -0.8846638446473643}], "active_band_count": 3, "active_band_mae_db": 0.4102368863273309}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 6.581694648038094e-05, "log_rms_mae_db": 0.899416268229275, "rms_envelope_nrmse": 0.12989927971598744, "envelope_correlation": 0.9886454741576759, "rms_gain_error_db": 0.2870175676652782, "reference_mean_power": 0.00014459942402382801, "output_mean_power": 0.00015447859819800957}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 5.765239137842014e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0474191785118657e-05, "output_power": 2.027758030920573e-05, "reference_active": true, "delta_db": -0.04190636981178385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1083544144974636e-05, "output_power": 1.0360153832557716e-05, "reference_active": true, "delta_db": -0.29312451311294785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.842442602727836e-07, "output_power": 2.3484027420948855e-07, "reference_active": true, "delta_db": -0.8291912495939225}], "active_band_count": 3, "active_band_mae_db": 0.38807404417288477}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8722106092415065, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "295f258d156503d2296ccc729decc3fbc8f81060c48f2f5b14f3c578001300de", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9967338678226483, "standard_deviation": 0.0007367282386435626, "minimum": 0.9951159617000873, "maximum": 0.9973799825571338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1953029599649665, "standard_deviation": 0.03376700053498063, "minimum": 1.1419958503191283, "maximum": 1.2331743019449133, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.38769142872020473, "standard_deviation": 0.17710733637900059, "minimum": 0.11340252396844375, "maximum": 0.6013086503644565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07374999488451414, "standard_deviation": 0.007832748515403307, "minimum": 0.06564857076631667, "maximum": 0.09098515708127829, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9990190906216632, "standard_deviation": 0.00019578688441577302, "minimum": 0.9985985620850665, "maximum": 0.9992305373681776, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7240438945028784, "standard_deviation": 0.02971171599156964, "minimum": 0.6684095466895731, "maximum": 0.7659892077798798, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.477151869558438, "standard_deviation": 0.1563262698855746, "minimum": 0.23975731806758901, "maximum": 0.7564832608534499, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04173930536526567, "standard_deviation": 0.004880779363406912, "minimum": 0.03526625324998556, "maximum": 0.0514889621237349, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9785600293568413, "standard_deviation": 0.002522651162122493, "minimum": 0.9740725723664707, "maximum": 0.9820244028243046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3141944068007725, "standard_deviation": 0.0323489690114975, "minimum": 1.2874653012213289, "maximum": 1.3882741155729306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4616110112915167, "standard_deviation": 0.14345045395759726, "minimum": 0.19251821936726454, "maximum": 0.6488843561027883, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17599070167539377, "standard_deviation": 0.012755191124250094, "minimum": 0.15923629558971647, "maximum": 0.19860297806296104, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920861281285589, "standard_deviation": 0.0016419692144717393, "minimum": 0.9886454741576759, "maximum": 0.9939887955401623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.844696749231681, "standard_deviation": 0.025118614882709042, "minimum": 0.8233652162105898, "maximum": 0.899416268229275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5176498139516903, "standard_deviation": 0.1996158108747161, "minimum": 0.1819140945536941, "maximum": 0.7781416348258237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10461790854135711, "standard_deviation": 0.012477572947046387, "minimum": 0.09077572441570182, "maximum": 0.12989927971598744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.995312220619752, "standard_deviation": 0.0009194096960337749, "minimum": 0.9937395767019894, "maximum": 0.9962844265707356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1649115456396184, "standard_deviation": 0.045479356562345814, "minimum": 1.107735747307806, "maximum": 1.2500991412242226, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5583779441930751, "standard_deviation": 0.21830702006693545, "minimum": 0.29073528480177685, "maximum": 0.981144166408515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07805715755321031, "standard_deviation": 0.007406911752884062, "minimum": 0.0699923351812543, "maximum": 0.09024535173260272, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9982927827246039, "standard_deviation": 0.0004536111948238601, "minimum": 0.9974537579240274, "maximum": 0.9988414657498641, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7287716935918875, "standard_deviation": 0.02809471153951794, "minimum": 0.675494269875995, "maximum": 0.7718170592944259, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8679429558985982, "standard_deviation": 0.3381287240605402, "minimum": 0.4184265186900529, "maximum": 1.4284390001642142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04590457647738003, "standard_deviation": 0.005848417265021244, "minimum": 0.03861091127683368, "maximum": 0.056055967270459216, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1832462929697659, "standard_deviation": 0.029888445729763898, "minimum": 0.13023283413646333, "maximum": 0.22223103175383715, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.393871050519864, "standard_deviation": 0.008381409153090928, "minimum": 5.379883528141732, "maximum": 5.405710917158558, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7860989962241695, "standard_deviation": 0.013091761972949891, "minimum": 0.7733420263664141, "maximum": 0.8124108652195424, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.051744809737528256, "standard_deviation": 0.038076907367236304, "minimum": -0.11147167989744572, "maximum": -0.0031496981294917886, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968832837502314, "standard_deviation": 0.0004667173829632391, "minimum": 0.9960531265748513, "maximum": 0.9976926185284141, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8336088067986022, "standard_deviation": 0.02255095425726013, "minimum": 0.7934920801870241, "maximum": 0.862422816405605, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36240897321370913, "standard_deviation": 0.14323217262117438, "minimum": 0.22918387701011692, "maximum": 0.6473649143732096, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.060963901519310665, "standard_deviation": 0.004295764970481766, "minimum": 0.05359072854444007, "maximum": 0.06861119891722006, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991148105246235, "standard_deviation": 0.00017451880308652008, "minimum": 0.9988403997725315, "maximum": 0.9994227695992808, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4461724447771034, "standard_deviation": 0.03125100206635336, "minimum": 0.40205622778312056, "maximum": 0.4915191469175656, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4214769743643648, "standard_deviation": 0.13203247214897978, "minimum": 0.3009598530102046, "maximum": 0.7190117218364391, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.032317961236535325, "standard_deviation": 0.002900155816949133, "minimum": 0.028484809379281228, "maximum": 0.036936737787015596, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0009", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.412008285522461, "predicted_mos": 2.412008285522461, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "e835e6249f1b0271cfeb7b0fcbc529db7773d420a82d0b46dfdb1437c469a69d", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8714484004303813, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8400189585346443, "全帯域_10ms_rms_envelope_nrmse": 0.061219325498311085, "全帯域_10ms_envelope_correlation": 0.9969107056399371, "全帯域_10ms_modulation_mae_db": 0.22662868647968945, "全帯域_40ms_log_rms_mae_db": 0.4565886307367638, "全帯域_40ms_rms_envelope_nrmse": 0.03289436897488695, "全帯域_40ms_envelope_correlation": 0.9991442139940993, "全帯域_40ms_modulation_mae_db": 0.40558891114229406, "300–1000 Hz_10ms_log_rms_mae_db": 1.2526784316055304, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07887497324384134, "300–1000 Hz_10ms_envelope_correlation": 0.9952414968494389, "300–1000 Hz_10ms_modulation_mae_db": 0.6917400518687457, "300–1000 Hz_40ms_log_rms_mae_db": 0.8104644872215749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04282155905752378, "300–1000 Hz_40ms_envelope_correlation": 0.9985451188789447, "300–1000 Hz_40ms_modulation_mae_db": 0.923732655054789, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2532136110329695, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06951683965502421, "1000–3000 Hz_10ms_envelope_correlation": 0.9972549466563925, "1000–3000 Hz_10ms_modulation_mae_db": 0.4181033449147041, "1000–3000 Hz_40ms_log_rms_mae_db": 0.779054078102239, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.043989732504000914, "1000–3000 Hz_40ms_envelope_correlation": 0.9990391896231052, "1000–3000 Hz_40ms_modulation_mae_db": 0.26785175952703333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.453198808331552, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17707195048502233, "3000–6000 Hz_10ms_envelope_correlation": 0.9792859566440462, "3000–6000 Hz_10ms_modulation_mae_db": 0.5377247707424915, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9112733637339774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09472365336939592, "3000–6000 Hz_40ms_envelope_correlation": 0.9941889863571908, "3000–6000 Hz_40ms_modulation_mae_db": 0.4924251336331631, "mr_stft_db": 5.491103460741676, "band_mae_db": 0.15250035937180523, "rms_gain_error_db": -0.09180329203235867, "output_peak": 0.7434766822581035}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7991112383830314, "全帯域_10ms_rms_envelope_nrmse": 0.05910019492598124, "全帯域_10ms_envelope_correlation": 0.9970808275226727, "全帯域_10ms_modulation_mae_db": 0.383059692528492, "全帯域_40ms_log_rms_mae_db": 0.44580492079312933, "全帯域_40ms_rms_envelope_nrmse": 0.03282852290831117, "全帯域_40ms_envelope_correlation": 0.9990845775626369, "全帯域_40ms_modulation_mae_db": 0.42523951946680505, "300–1000 Hz_10ms_log_rms_mae_db": 1.2718672377168918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06878667306822107, "300–1000 Hz_10ms_envelope_correlation": 0.9963937442594979, "300–1000 Hz_10ms_modulation_mae_db": 0.5977593622090752, "300–1000 Hz_40ms_log_rms_mae_db": 0.823491485782891, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041589089026645785, "300–1000 Hz_40ms_envelope_correlation": 0.9986122953490472, "300–1000 Hz_40ms_modulation_mae_db": 0.5873136076683552, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2765285050103838, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08035916824931123, "1000–3000 Hz_10ms_envelope_correlation": 0.9961693543808046, "1000–3000 Hz_10ms_modulation_mae_db": 0.3303209622788499, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7662257863393974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.047586091775957795, "1000–3000 Hz_40ms_envelope_correlation": 0.9987478607921167, "1000–3000 Hz_40ms_modulation_mae_db": 0.5961218109924098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.430590568028834, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16794169860254018, "3000–6000 Hz_10ms_envelope_correlation": 0.9800777295294869, "3000–6000 Hz_10ms_modulation_mae_db": 0.8512323436870138, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9017871013138585, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.090669494422701, "3000–6000 Hz_40ms_envelope_correlation": 0.9942390347014399, "3000–6000 Hz_40ms_modulation_mae_db": 1.0771785961481288, "mr_stft_db": 5.48722124770499, "band_mae_db": 0.1795497057643427, "rms_gain_error_db": -0.04903433712651845, "output_peak": 0.7395390005543027}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7739423927357666, "全帯域_10ms_rms_envelope_nrmse": 0.05427872402131083, "全帯域_10ms_envelope_correlation": 0.9975233363789555, "全帯域_10ms_modulation_mae_db": 0.18811947778094876, "全帯域_40ms_log_rms_mae_db": 0.40112259904524294, "全帯域_40ms_rms_envelope_nrmse": 0.026508127040115834, "全帯域_40ms_envelope_correlation": 0.9993937071276907, "全帯域_40ms_modulation_mae_db": 0.13561049634669195, "300–1000 Hz_10ms_log_rms_mae_db": 1.2946608561120247, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0676960248211073, "300–1000 Hz_10ms_envelope_correlation": 0.9964909599900676, "300–1000 Hz_10ms_modulation_mae_db": 0.11182179515552955, "300–1000 Hz_40ms_log_rms_mae_db": 0.8750445024187846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04368999209309558, "300–1000 Hz_40ms_envelope_correlation": 0.9984558097738306, "300–1000 Hz_40ms_modulation_mae_db": 0.28315328099247256, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2283975185094897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07010802498518291, "1000–3000 Hz_10ms_envelope_correlation": 0.9970176676041275, "1000–3000 Hz_10ms_modulation_mae_db": 0.12799149255305867, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7214694515943222, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04146945666424644, "1000–3000 Hz_40ms_envelope_correlation": 0.9989636334647196, "1000–3000 Hz_40ms_modulation_mae_db": 0.041978085083119114, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4162541333205256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18234409346026378, "3000–6000 Hz_10ms_envelope_correlation": 0.9767644819749208, "3000–6000 Hz_10ms_modulation_mae_db": 0.6330572395914448, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8974310009684042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0959826128852866, "3000–6000 Hz_40ms_envelope_correlation": 0.9931259056625448, "3000–6000 Hz_40ms_modulation_mae_db": 0.6162980453728111, "mr_stft_db": 5.492005919772342, "band_mae_db": 0.11325997213354815, "rms_gain_error_db": -0.03801380726365636, "output_peak": 0.811065652729015}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8084836698577751, "全帯域_10ms_rms_envelope_nrmse": 0.056938028374684486, "全帯域_10ms_envelope_correlation": 0.9972752892610414, "全帯域_10ms_modulation_mae_db": 0.466009872538558, "全帯域_40ms_log_rms_mae_db": 0.39474826800348656, "全帯域_40ms_rms_envelope_nrmse": 0.027728446716621245, "全帯域_40ms_envelope_correlation": 0.9993379969405023, "全帯域_40ms_modulation_mae_db": 0.41176238728356435, "300–1000 Hz_10ms_log_rms_mae_db": 1.2501055377655146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0722505728406971, "300–1000 Hz_10ms_envelope_correlation": 0.9960356675729095, "300–1000 Hz_10ms_modulation_mae_db": 0.7446503194402, "300–1000 Hz_40ms_log_rms_mae_db": 0.748236436899601, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038487367311658784, "300–1000 Hz_40ms_envelope_correlation": 0.9988515788643445, "300–1000 Hz_40ms_modulation_mae_db": 1.0776755121932355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2301334245343334, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0701079484672785, "1000–3000 Hz_10ms_envelope_correlation": 0.996992044690154, "1000–3000 Hz_10ms_modulation_mae_db": 0.4201305119904697, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6884029484786197, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.036156202369940556, "1000–3000 Hz_40ms_envelope_correlation": 0.9991828236105027, "1000–3000 Hz_40ms_modulation_mae_db": 0.3682365525778877, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4534783833843636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17552741408001352, "3000–6000 Hz_10ms_envelope_correlation": 0.9781194782327134, "3000–6000 Hz_10ms_modulation_mae_db": 0.5630307385861048, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9066391782426612, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09542547335128813, "3000–6000 Hz_40ms_envelope_correlation": 0.9934251480759025, "3000–6000 Hz_40ms_modulation_mae_db": 0.5687315931846816, "mr_stft_db": 5.480469931312283, "band_mae_db": 0.17306534987769032, "rms_gain_error_db": -0.04160100351685804, "output_peak": 0.8284459507705485}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8165260550624122, "全帯域_10ms_rms_envelope_nrmse": 0.06696806032863861, "全帯域_10ms_envelope_correlation": 0.9963803489026246, "全帯域_10ms_modulation_mae_db": 0.4073651626631798, "全帯域_40ms_log_rms_mae_db": 0.4401921731842637, "全帯域_40ms_rms_envelope_nrmse": 0.03770830428393598, "全帯域_40ms_envelope_correlation": 0.9989584484117245, "全帯域_40ms_modulation_mae_db": 0.3555760060502662, "300–1000 Hz_10ms_log_rms_mae_db": 1.2415875551861943, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07421007310618637, "300–1000 Hz_10ms_envelope_correlation": 0.9957933563488426, "300–1000 Hz_10ms_modulation_mae_db": 0.6730410296602553, "300–1000 Hz_40ms_log_rms_mae_db": 0.8331669425841401, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039955031184608386, "300–1000 Hz_40ms_envelope_correlation": 0.9987348251336198, "300–1000 Hz_40ms_modulation_mae_db": 0.8234800864200146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2750041545680215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0746983745741974, "1000–3000 Hz_10ms_envelope_correlation": 0.9969603913387656, "1000–3000 Hz_10ms_modulation_mae_db": 0.423908837077485, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7343899660860413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04549214909050875, "1000–3000 Hz_40ms_envelope_correlation": 0.9991470067375398, "1000–3000 Hz_40ms_modulation_mae_db": 0.6141154138997673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5276579228031044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19991695870754111, "3000–6000 Hz_10ms_envelope_correlation": 0.9729762461801785, "3000–6000 Hz_10ms_modulation_mae_db": 1.0284193550827874, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9897606993232099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.126895596858175, "3000–6000 Hz_40ms_envelope_correlation": 0.990404224443675, "3000–6000 Hz_40ms_modulation_mae_db": 1.0659675776605502, "mr_stft_db": 5.477281398073772, "band_mae_db": 0.20745568034802034, "rms_gain_error_db": -0.13683162617558067, "output_peak": 0.7590222362618498}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8132048665059489, "全帯域_10ms_rms_envelope_nrmse": 0.056569241185701634, "全帯域_10ms_envelope_correlation": 0.9974954659312658, "全帯域_10ms_modulation_mae_db": 0.15394026515305512, "全帯域_40ms_log_rms_mae_db": 0.41444669663262546, "全帯域_40ms_rms_envelope_nrmse": 0.03258257729023641, "全帯域_40ms_envelope_correlation": 0.9992940363454412, "全帯域_40ms_modulation_mae_db": 0.2466846465328345, "300–1000 Hz_10ms_log_rms_mae_db": 1.2459756038791985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07139682780201835, "300–1000 Hz_10ms_envelope_correlation": 0.9960987142718714, "300–1000 Hz_10ms_modulation_mae_db": 0.44520768006116773, "300–1000 Hz_40ms_log_rms_mae_db": 0.8025067972918203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042493182001350716, "300–1000 Hz_40ms_envelope_correlation": 0.9985504260299031, "300–1000 Hz_40ms_modulation_mae_db": 0.6933251733760651, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2686007313248333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07163073250619181, "1000–3000 Hz_10ms_envelope_correlation": 0.9972321664837801, "1000–3000 Hz_10ms_modulation_mae_db": 0.3628253581696958, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7608894796429639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.046531920282540364, "1000–3000 Hz_40ms_envelope_correlation": 0.9990733310334147, "1000–3000 Hz_40ms_modulation_mae_db": 0.35591512872252434, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4346838946908902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2011875894420828, "3000–6000 Hz_10ms_envelope_correlation": 0.9727197511697311, "3000–6000 Hz_10ms_modulation_mae_db": 0.7936251960137195, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9600432308413398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12658621986877985, "3000–6000 Hz_40ms_envelope_correlation": 0.9906528879687879, "3000–6000 Hz_40ms_modulation_mae_db": 0.8588818470698948, "mr_stft_db": 5.489273632408151, "band_mae_db": 0.2838490428558507, "rms_gain_error_db": -0.14544918233715806, "output_peak": 0.8255040678164379}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8074901465622166, "全帯域_10ms_rms_envelope_nrmse": 0.0613908674616894, "全帯域_10ms_envelope_correlation": 0.9968628066694005, "全帯域_10ms_modulation_mae_db": 0.5030115197323534, "全帯域_40ms_log_rms_mae_db": 0.42487045242659827, "全帯域_40ms_rms_envelope_nrmse": 0.03315261684749791, "全帯域_40ms_envelope_correlation": 0.9990865603988364, "全帯域_40ms_modulation_mae_db": 0.6709354723597146, "300–1000 Hz_10ms_log_rms_mae_db": 1.2576945150852685, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07997953603246924, "300–1000 Hz_10ms_envelope_correlation": 0.9951155546653574, "300–1000 Hz_10ms_modulation_mae_db": 0.39593173394603687, "300–1000 Hz_40ms_log_rms_mae_db": 0.8322894715335262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05145855850220614, "300–1000 Hz_40ms_envelope_correlation": 0.9978871044761033, "300–1000 Hz_40ms_modulation_mae_db": 0.8523115437414471, "1000–3000 Hz_10ms_log_rms_mae_db": 1.262840201624509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07560212171159553, "1000–3000 Hz_10ms_envelope_correlation": 0.9966604949293596, "1000–3000 Hz_10ms_modulation_mae_db": 0.7064268648326109, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7593661543959641, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.048772080010850684, "1000–3000 Hz_40ms_envelope_correlation": 0.998706364858704, "1000–3000 Hz_40ms_modulation_mae_db": 0.9407489794554662, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4563443397694313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17396419773472258, "3000–6000 Hz_10ms_envelope_correlation": 0.9786542353143081, "3000–6000 Hz_10ms_modulation_mae_db": 0.6532189973168522, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9190275539282099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0987610935656197, "3000–6000 Hz_40ms_envelope_correlation": 0.992601026857844, "3000–6000 Hz_40ms_modulation_mae_db": 0.7349034834380431, "mr_stft_db": 5.477822397058077, "band_mae_db": 0.15918059611087348, "rms_gain_error_db": -0.0720895496226559, "output_peak": 0.7460260713572759}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8297219695746548, "全帯域_10ms_rms_envelope_nrmse": 0.05386296034281287, "全帯域_10ms_envelope_correlation": 0.9975739796796259, "全帯域_10ms_modulation_mae_db": 0.23314786716384714, "全帯域_40ms_log_rms_mae_db": 0.42328201476582666, "全帯域_40ms_rms_envelope_nrmse": 0.027777672936839175, "全帯域_40ms_envelope_correlation": 0.999350644463111, "全帯域_40ms_modulation_mae_db": 0.40797850953019205, "300–1000 Hz_10ms_log_rms_mae_db": 1.1830802868413701, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07962004045985954, "300–1000 Hz_10ms_envelope_correlation": 0.9951524485787102, "300–1000 Hz_10ms_modulation_mae_db": 0.916079013812848, "300–1000 Hz_40ms_log_rms_mae_db": 0.8088504414579256, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05216637941725933, "300–1000 Hz_40ms_envelope_correlation": 0.9978033916137243, "300–1000 Hz_40ms_modulation_mae_db": 1.4905492123516912, "1000–3000 Hz_10ms_log_rms_mae_db": 1.268151251456172, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06709893699069557, "1000–3000 Hz_10ms_envelope_correlation": 0.9973003073121676, "1000–3000 Hz_10ms_modulation_mae_db": 0.2883266043672381, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6879652481857652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03431084612524914, "1000–3000 Hz_40ms_envelope_correlation": 0.9993418894290695, "1000–3000 Hz_40ms_modulation_mae_db": 0.48167378697614854, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4631302651615428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1740158225159278, "3000–6000 Hz_10ms_envelope_correlation": 0.9785140578558489, "3000–6000 Hz_10ms_modulation_mae_db": 0.8149535788334122, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9355246186108785, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09971901357097335, "3000–6000 Hz_40ms_envelope_correlation": 0.9924125804634214, "3000–6000 Hz_40ms_modulation_mae_db": 0.7998355973808792, "mr_stft_db": 5.478148527395807, "band_mae_db": 0.22304082320400495, "rms_gain_error_db": -0.054732922419045565, "output_peak": 0.8186848913679181}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 82160, "output_samples": 82160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.440793654397106, "spectral_convergence": 0.22324182396197847, "spectral_convergence_status": "ok", "amplitude_floor": 2.2676184880983433e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3693427300538, "spectral_convergence": 0.21638316866028462, "spectral_convergence_status": "ok", "amplitude_floor": 1.9761500134157035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.663173997774121, "spectral_convergence": 0.21968048348385344, "spectral_convergence_status": "ok", "amplitude_floor": 1.298774161363914e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.367365175011865e-05, "output_power": 9.266209262831308e-05, "delta_db": 0.4431338062449061, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016048781148481498, "output_power": 0.0015921983333197515, "delta_db": -0.034448897998531944, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001327868811244639, "output_power": 0.0012967669111394848, "delta_db": -0.10293250050628786, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0026813637139198997, "output_power": 0.00258239217272184, "delta_db": -0.16333531219145087, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003087131182449005, "output_power": 0.0003158842264092003, "delta_db": 0.09972855584672592, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00020800689998023804, "output_power": 0.00020461404133937077, "delta_db": -0.07142308344292861, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15250035937180523, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09180329203235867, "pitch": null, "output_rms_dbfs": -22.157738356499156, "output_peak": 0.7434766822581035, "reference_rms_dbfs": -22.065935064466796, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.491103460741676}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 82160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003747956436237944, "log_rms_mae_db": 0.8400189585346443, "rms_envelope_nrmse": 0.061219325498311085, "envelope_correlation": 0.9969107056399371, "rms_gain_error_db": -0.0918030931683225, "reference_mean_power": 0.006220557891926456, "output_mean_power": 0.006090445017801216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.814448626587233e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017308539978626828, "output_power": 0.0016442439715863605, "reference_active": true, "delta_db": -0.22294177277849458}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0006400481816254359, "output_power": 0.0005976530849688277, "reference_active": true, "delta_db": -0.2976350249801626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4706618104713675e-05, "output_power": 4.637700988405219e-05, "reference_active": true, "delta_db": 0.1593092616804112}], "active_band_count": 3, "active_band_mae_db": 0.22662868647968945}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.00031163820648484283, "log_rms_mae_db": 0.4565886307367638, "rms_envelope_nrmse": 0.03289436897488695, "envelope_correlation": 0.9991442139940993, "rms_gain_error_db": -0.09179382937444777, "reference_mean_power": 0.006257088451954998, "output_mean_power": 0.006126224550630069}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 6.546091034740867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001677353066756132, "output_power": 0.0015906620894116355, "reference_active": true, "delta_db": -0.23046556493529716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00035132010859926165, "output_power": 0.0003268778501122669, "reference_active": true, "delta_db": -0.31317515401355944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3827329510863976e-06, "output_power": 2.782197603662807e-06, "reference_active": true, "delta_db": 0.6731260144780256}], "active_band_count": 3, "active_band_mae_db": 0.40558891114229406}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.00015152624573583172, "log_rms_mae_db": 1.2526784316055304, "rms_envelope_nrmse": 0.07887497324384134, "envelope_correlation": 0.9952414968494389, "rms_gain_error_db": -0.04775150320931541, "reference_mean_power": 0.001672933915423699, "output_mean_power": 0.0016546404436528356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 1.623496524404657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006256720367801861, "output_power": 0.0006117479296203938, "reference_active": true, "delta_db": -0.09774237232362308}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015725509039345338, "output_power": 0.00012640775569142107, "reference_active": true, "delta_db": -0.9483099187686472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4941700779965039e-05, "output_power": 1.178917049737319e-05, "reference_active": true, "delta_db": -1.0291678645139666}], "active_band_count": 3, "active_band_mae_db": 0.6917400518687457}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0001385613366624899, "log_rms_mae_db": 0.8104644872215749, "rms_envelope_nrmse": 0.04282155905752378, "envelope_correlation": 0.9985451188789447, "rms_gain_error_db": -0.04775185264286546, "reference_mean_power": 0.0016827742497549935, "output_mean_power": 0.0016643730403690974}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 1.5301470087365327e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005852793815004822, "output_power": 0.0005696284033309567, "reference_active": true, "delta_db": -0.1177158849551398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.151264176170612e-05, "output_power": 6.75509767995819e-05, "reference_active": true, "delta_db": -0.8159333513231427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3333886770336671e-06, "output_power": 8.733772798582845e-07, "reference_active": true, "delta_db": -1.8375487288860843}], "active_band_count": 3, "active_band_mae_db": 0.923732655054789}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003048971769130071, "log_rms_mae_db": 1.2532136110329695, "rms_envelope_nrmse": 0.06951683965502421, "envelope_correlation": 0.9972549466563925, "rms_gain_error_db": -0.18197289989234178, "reference_mean_power": 0.0025898505670984154, "output_mean_power": 0.002483575781976322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 4.739471399174337e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014413951986732973, "output_power": 0.00135900744687458, "reference_active": true, "delta_db": -0.255612345616812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0004245147064825657, "output_power": 0.000380043183110379, "reference_active": true, "delta_db": -0.48059793189080235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3891971817365186e-05, "output_power": 2.6919188866261962e-05, "reference_active": true, "delta_db": 0.5180997572364978}], "active_band_count": 3, "active_band_mae_db": 0.4181033449147041}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0002474836728453422, "log_rms_mae_db": 0.779054078102239, "rms_envelope_nrmse": 0.043989732504000914, "envelope_correlation": 0.9990391896231052, "rms_gain_error_db": -0.1819499671955356, "reference_mean_power": 0.002605032676611572, "output_mean_power": 0.0024981480833674904}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 4.638343817264169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014427582782272192, "output_power": 0.0013535760725152228, "reference_active": true, "delta_db": -0.27710906236927074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020918188683306615, "output_power": 0.00018769939091072354, "reference_active": true, "delta_db": -0.47061212714183115}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0082267328987986e-06, "output_power": 2.0342117234402558e-06, "reference_active": true, "delta_db": 0.05583408906999812}], "active_band_count": 3, "active_band_mae_db": 0.26785175952703333}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 8.624763428459013e-05, "log_rms_mae_db": 1.453198808331552, "rms_envelope_nrmse": 0.17707195048502233, "envelope_correlation": 0.9792859566440462, "rms_gain_error_db": 0.27928974471076, "reference_mean_power": 0.00014375800668184303, "output_mean_power": 0.00015330665814518202}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.640530401894145e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.056434359733932e-05, "output_power": 2.2928548292510225e-05, "reference_active": true, "delta_db": 0.4726170687900463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.844428212989113e-05, "output_power": 1.7230345491068886e-05, "reference_active": true, "delta_db": -0.2956777100103431}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.64438001039378e-06, "output_power": 3.0000998034103838e-06, "reference_active": true, "delta_db": -0.8448795334270851}], "active_band_count": 3, "active_band_mae_db": 0.5377247707424915}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 6.581694648038094e-05, "log_rms_mae_db": 0.9112733637339774, "rms_envelope_nrmse": 0.09472365336939592, "envelope_correlation": 0.9941889863571908, "rms_gain_error_db": 0.2792728344437965, "reference_mean_power": 0.00014459942402382801, "output_mean_power": 0.00015420336343114412}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 5.765239137842014e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0474191785118657e-05, "output_power": 2.2470738556277882e-05, "reference_active": true, "delta_db": 0.40410579736062857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1083544144974636e-05, "output_power": 1.052083053862744e-05, "reference_active": true, "delta_db": -0.22628630003118083}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.842442602727836e-07, "output_power": 2.338855405825124e-07, "reference_active": true, "delta_db": -0.84688330350768}], "active_band_count": 3, "active_band_mae_db": 0.4924251336331631}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8790942771540569, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4549c904c78bd8b201a55a74c51a0b25ea11a30b4f41159bb4e5a4e204e2d790", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.996948421674444, "standard_deviation": 0.0003770147934033369, "minimum": 0.9961693543808046, "maximum": 0.9973003073121676, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.257858674757589, "standard_deviation": 0.01907466685100157, "minimum": 1.2283975185094897, "maximum": 1.2765285050103838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.384754247023014, "standard_deviation": 0.1630246512602044, "minimum": 0.12799149255305867, "maximum": 0.7064268648326109, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07239026839243465, "standard_deviation": 0.004249302724384479, "minimum": 0.06709893699069557, "maximum": 0.08035916824931123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9990252624436465, "standard_deviation": 0.00021542313012231498, "minimum": 0.998706364858704, "maximum": 0.9993418894290695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7372203891031641, "standard_deviation": 0.03526905992160244, "minimum": 0.6879652481857652, "maximum": 0.779054078102239, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.45833018965429456, "standard_deviation": 0.26853177379575377, "minimum": 0.041978085083119114, "maximum": 0.9407489794554662, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04303855985291183, "standard_deviation": 0.005327604834324994, "minimum": 0.03431084612524914, "maximum": 0.048772080010850684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9771389921126542, "standard_deviation": 0.002813034892558417, "minimum": 0.9727197511697311, "maximum": 0.9800777295294869, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4544172894362806, "standard_deviation": 0.03350848952482443, "minimum": 1.4162541333205256, "maximum": 1.5276579228031044, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7344077774817284, "standard_deviation": 0.16690585882099773, "minimum": 0.5377247707424915, "maximum": 1.0284193550827874, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18149621562851426, "standard_deviation": 0.012417764323343584, "minimum": 0.16794169860254018, "maximum": 0.2011875894420828, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926312243163509, "standard_deviation": 0.0014544886945710895, "minimum": 0.990404224443675, "maximum": 0.9942390347014399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9276858433703175, "standard_deviation": 0.032370914123943655, "minimum": 0.8974310009684042, "maximum": 0.9897606993232099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7767777342360189, "standard_deviation": 0.21804028968059103, "minimum": 0.4924251336331631, "maximum": 1.0771785961481288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10359539473652744, "standard_deviation": 0.01454236674924568, "minimum": 0.090669494422701, "maximum": 0.126895596858175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.995790242817087, "standard_deviation": 0.0005572539713849851, "minimum": 0.9951155546653574, "maximum": 0.9964909599900676, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2497062530239993, "standard_deviation": 0.03185610852688647, "minimum": 1.1830802868413701, "maximum": 1.2946608561120247, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5720288732692322, "standard_deviation": 0.248495275964943, "minimum": 0.11182179515552955, "maximum": 0.916079013812848, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07410184017180005, "standard_deviation": 0.004896385124896993, "minimum": 0.0676960248211073, "maximum": 0.07997953603246924, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984300687649397, "standard_deviation": 0.00038166425155674097, "minimum": 0.9978033916137243, "maximum": 0.9988515788643445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.816756320648783, "standard_deviation": 0.03576425859651362, "minimum": 0.748236436899601, "maximum": 0.8750445024187846, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8414426339747587, "standard_deviation": 0.3550784827785351, "minimum": 0.28315328099247256, "maximum": 1.4905492123516912, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04408264482429357, "standard_deviation": 0.005050939352740066, "minimum": 0.038487367311658784, "maximum": 0.05216637941725933, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18648769120826697, "standard_deviation": 0.05174999064130699, "minimum": 0.11325997213354815, "maximum": 0.2838490428558507, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.484165814308387, "standard_deviation": 0.006353009330936562, "minimum": 5.477281398073772, "maximum": 5.492005919772342, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7839705691394314, "standard_deviation": 0.04021094767970452, "minimum": 0.7395390005543027, "maximum": 0.8284459507705485, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07869446506172897, "standard_deviation": 0.042319039208338995, "minimum": -0.14544918233715806, "maximum": -0.03801380726365636, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9971378449981905, "standard_deviation": 0.00041229157848758516, "minimum": 0.9963803489026246, "maximum": 0.9975739796796259, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8110624121520562, "standard_deviation": 0.019861087422798934, "minimum": 0.7739423927357666, "maximum": 0.8400189585346443, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32016031800501543, "standard_deviation": 0.13504014902053899, "minimum": 0.15394026515305512, "maximum": 0.5030115197323534, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05879092526739127, "standard_deviation": 0.004352863065413196, "minimum": 0.05386296034281287, "maximum": 0.06696806032863861, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992062731555054, "standard_deviation": 0.00015831140470058864, "minimum": 0.9989584484117245, "maximum": 0.9993937071276907, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4251319694484921, "standard_deviation": 0.021583681509601983, "minimum": 0.39474826800348656, "maximum": 0.4565886307367638, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3824219935890454, "standard_deviation": 0.15454325230755087, "minimum": 0.13561049634669195, "maximum": 0.6709354723597146, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031397579624805584, "standard_deviation": 0.003762039818807074, "minimum": 0.026508127040115834, "maximum": 0.03770830428393598, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0009", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.25632643699646, "predicted_mos": 2.25632643699646, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "355da5a676cfa99b4e228f23d4e28c4d78ca9e8fbcab58fb0c21d3fe2298164b", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.004158199764788, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8347144717732644, "全帯域_10ms_rms_envelope_nrmse": 0.055046469525508825, "全帯域_10ms_envelope_correlation": 0.9974555878623497, "全帯域_10ms_modulation_mae_db": 0.5914334188090365, "全帯域_40ms_log_rms_mae_db": 0.4439642362685156, "全帯域_40ms_rms_envelope_nrmse": 0.02762015051198449, "全帯域_40ms_envelope_correlation": 0.9993397203987706, "全帯域_40ms_modulation_mae_db": 0.6245088023737911, "300–1000 Hz_10ms_log_rms_mae_db": 1.1625977564025087, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06924032370995103, "300–1000 Hz_10ms_envelope_correlation": 0.9963478058543674, "300–1000 Hz_10ms_modulation_mae_db": 0.4137919014932477, "300–1000 Hz_40ms_log_rms_mae_db": 0.7300427234040013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.040070173221208046, "300–1000 Hz_40ms_envelope_correlation": 0.9987125823711558, "300–1000 Hz_40ms_modulation_mae_db": 0.6131642872259933, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1995877012835643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07549856982655702, "1000–3000 Hz_10ms_envelope_correlation": 0.9965981648307219, "1000–3000 Hz_10ms_modulation_mae_db": 0.7168114204683773, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7158852531348593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04174246122046672, "1000–3000 Hz_40ms_envelope_correlation": 0.9990318957261519, "1000–3000 Hz_40ms_modulation_mae_db": 0.9609947609516917, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4716025131449897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16628491304997156, "3000–6000 Hz_10ms_envelope_correlation": 0.9804673886409804, "3000–6000 Hz_10ms_modulation_mae_db": 0.9318916373811948, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8983196481077018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0869494264363567, "3000–6000 Hz_40ms_envelope_correlation": 0.994263067245984, "3000–6000 Hz_40ms_modulation_mae_db": 0.9235978662188248, "mr_stft_db": 5.468686754028668, "band_mae_db": 0.21820146216113223, "rms_gain_error_db": -0.03411758622743856, "output_peak": 0.7527774881333261}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8004652389777515, "全帯域_10ms_rms_envelope_nrmse": 0.0579657841109329, "全帯域_10ms_envelope_correlation": 0.9971672257572137, "全帯域_10ms_modulation_mae_db": 0.6502747718151901, "全帯域_40ms_log_rms_mae_db": 0.42805841074223405, "全帯域_40ms_rms_envelope_nrmse": 0.028332149513561163, "全帯域_40ms_envelope_correlation": 0.9992939969228075, "全帯域_40ms_modulation_mae_db": 0.39289178344114245, "300–1000 Hz_10ms_log_rms_mae_db": 1.221205850574489, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08026922486930259, "300–1000 Hz_10ms_envelope_correlation": 0.9950757146231876, "300–1000 Hz_10ms_modulation_mae_db": 0.36924852156607324, "300–1000 Hz_40ms_log_rms_mae_db": 0.7765030776772093, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04487617886682003, "300–1000 Hz_40ms_envelope_correlation": 0.9983867462668344, "300–1000 Hz_40ms_modulation_mae_db": 0.7305959146354865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2163981337689456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.08403480280936307, "1000–3000 Hz_10ms_envelope_correlation": 0.9956795330137826, "1000–3000 Hz_10ms_modulation_mae_db": 0.739205774411202, "1000–3000 Hz_40ms_log_rms_mae_db": 0.686993056343181, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.039646827687744204, "1000–3000 Hz_40ms_envelope_correlation": 0.9990315945028408, "1000–3000 Hz_40ms_modulation_mae_db": 0.16772116371815604, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4079130835681946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16593139997779, "3000–6000 Hz_10ms_envelope_correlation": 0.9803965120155713, "3000–6000 Hz_10ms_modulation_mae_db": 0.41025574402465576, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8396681285519358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08606641882894422, "3000–6000 Hz_40ms_envelope_correlation": 0.9944553754386211, "3000–6000 Hz_40ms_modulation_mae_db": 0.3497036797497379, "mr_stft_db": 5.459964484276212, "band_mae_db": 0.14343417225121055, "rms_gain_error_db": -0.01921881223505176, "output_peak": 0.7481068342365434}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8256763628518423, "全帯域_10ms_rms_envelope_nrmse": 0.05917843326068095, "全帯域_10ms_envelope_correlation": 0.9970499237826971, "全帯域_10ms_modulation_mae_db": 0.9216390626301051, "全帯域_40ms_log_rms_mae_db": 0.4726786295874565, "全帯域_40ms_rms_envelope_nrmse": 0.029510320501526095, "全帯域_40ms_envelope_correlation": 0.9992397550637568, "全帯域_40ms_modulation_mae_db": 0.7383578457364764, "300–1000 Hz_10ms_log_rms_mae_db": 1.184377893733401, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07663756201579411, "300–1000 Hz_10ms_envelope_correlation": 0.9955241863065794, "300–1000 Hz_10ms_modulation_mae_db": 0.8280173234245908, "300–1000 Hz_40ms_log_rms_mae_db": 0.7856513671383577, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04796562687562858, "300–1000 Hz_40ms_envelope_correlation": 0.9981543093800653, "300–1000 Hz_40ms_modulation_mae_db": 1.2072222085597986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.221058605825457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.06784494308063738, "1000–3000 Hz_10ms_envelope_correlation": 0.9971934695433528, "1000–3000 Hz_10ms_modulation_mae_db": 0.947233475647212, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7265216762299413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03569974240986528, "1000–3000 Hz_40ms_envelope_correlation": 0.9992241370494096, "1000–3000 Hz_40ms_modulation_mae_db": 0.7756784502961972, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4396425581746082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17350886736574386, "3000–6000 Hz_10ms_envelope_correlation": 0.9787442062143518, "3000–6000 Hz_10ms_modulation_mae_db": 0.8923141091733751, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8879628544222103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09869471398192228, "3000–6000 Hz_40ms_envelope_correlation": 0.9931091782263384, "3000–6000 Hz_40ms_modulation_mae_db": 0.984067837268034, "mr_stft_db": 5.446628903800987, "band_mae_db": 0.14108887788597493, "rms_gain_error_db": -0.03145695124080896, "output_peak": 0.7796277727819548}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8210519652148472, "全帯域_10ms_rms_envelope_nrmse": 0.059846049187444336, "全帯域_10ms_envelope_correlation": 0.9969826239779561, "全帯域_10ms_modulation_mae_db": 0.15562347145344602, "全帯域_40ms_log_rms_mae_db": 0.40043604154506957, "全帯域_40ms_rms_envelope_nrmse": 0.0309960655955468, "全帯域_40ms_envelope_correlation": 0.9991514490494515, "全帯域_40ms_modulation_mae_db": 0.12443432382506907, "300–1000 Hz_10ms_log_rms_mae_db": 1.203820233574565, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07853961751544651, "300–1000 Hz_10ms_envelope_correlation": 0.9953526153382529, "300–1000 Hz_10ms_modulation_mae_db": 0.3165015520390456, "300–1000 Hz_40ms_log_rms_mae_db": 0.7374135073768535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044197380456000204, "300–1000 Hz_40ms_envelope_correlation": 0.9984768927458934, "300–1000 Hz_40ms_modulation_mae_db": 0.5458504371272583, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2635522460897923, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0736474886126376, "1000–3000 Hz_10ms_envelope_correlation": 0.9966811184585327, "1000–3000 Hz_10ms_modulation_mae_db": 0.21728015374978535, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6839148438309804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.03535550966456633, "1000–3000 Hz_40ms_envelope_correlation": 0.9992281886562584, "1000–3000 Hz_40ms_modulation_mae_db": 0.21790380242608784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4000950343528498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1746639848165052, "3000–6000 Hz_10ms_envelope_correlation": 0.9782790480568835, "3000–6000 Hz_10ms_modulation_mae_db": 0.29685680868647707, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8516819408767484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09490932616235986, "3000–6000 Hz_40ms_envelope_correlation": 0.9932745193079633, "3000–6000 Hz_40ms_modulation_mae_db": 0.32811355197207687, "mr_stft_db": 5.460243258447143, "band_mae_db": 0.17093534662358112, "rms_gain_error_db": 0.00036302392096401915, "output_peak": 0.7413454142315021}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8118470121316801, "全帯域_10ms_rms_envelope_nrmse": 0.058749181194805905, "全帯域_10ms_envelope_correlation": 0.9970906847905613, "全帯域_10ms_modulation_mae_db": 0.280030950075483, "全帯域_40ms_log_rms_mae_db": 0.42536399472289144, "全帯域_40ms_rms_envelope_nrmse": 0.028485765526249224, "全帯域_40ms_envelope_correlation": 0.9992881434876679, "全帯域_40ms_modulation_mae_db": 0.40441968455953986, "300–1000 Hz_10ms_log_rms_mae_db": 1.2379517106085947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08121859086694203, "300–1000 Hz_10ms_envelope_correlation": 0.9950748732013899, "300–1000 Hz_10ms_modulation_mae_db": 0.9375726273643665, "300–1000 Hz_40ms_log_rms_mae_db": 0.767583856983243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0435512586731122, "300–1000 Hz_40ms_envelope_correlation": 0.9985685122982035, "300–1000 Hz_40ms_modulation_mae_db": 1.238380375977733, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2740657939353315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.0778714581477268, "1000–3000 Hz_10ms_envelope_correlation": 0.9963168513462407, "1000–3000 Hz_10ms_modulation_mae_db": 0.17877820637743017, "1000–3000 Hz_40ms_log_rms_mae_db": 0.697754271746084, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04184187187145071, "1000–3000 Hz_40ms_envelope_correlation": 0.9989560352353788, "1000–3000 Hz_40ms_modulation_mae_db": 0.2740511966232631, "3000–6000 Hz_10ms_log_rms_mae_db": 1.417164127829179, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18743348766313148, "3000–6000 Hz_10ms_envelope_correlation": 0.9751818592112869, "3000–6000 Hz_10ms_modulation_mae_db": 0.5652796968681778, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8838102128671111, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10822911698416549, "3000–6000 Hz_40ms_envelope_correlation": 0.9918339360477144, "3000–6000 Hz_40ms_modulation_mae_db": 0.639371783895173, "mr_stft_db": 5.463123960547861, "band_mae_db": 0.16426217803948984, "rms_gain_error_db": -0.02286026013061365, "output_peak": 0.7679103396376962}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8029381361775703, "全帯域_10ms_rms_envelope_nrmse": 0.06099266703293123, "全帯域_10ms_envelope_correlation": 0.9968727711860373, "全帯域_10ms_modulation_mae_db": 0.26036404754574033, "全帯域_40ms_log_rms_mae_db": 0.43008305341244973, "全帯域_40ms_rms_envelope_nrmse": 0.03439806866810821, "全帯域_40ms_envelope_correlation": 0.9989635339817959, "全帯域_40ms_modulation_mae_db": 0.34790985028701993, "300–1000 Hz_10ms_log_rms_mae_db": 1.1860718694989456, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08401757578880296, "300–1000 Hz_10ms_envelope_correlation": 0.9946277776205001, "300–1000 Hz_10ms_modulation_mae_db": 0.25393432209536604, "300–1000 Hz_40ms_log_rms_mae_db": 0.711345874531809, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05195518755032416, "300–1000 Hz_40ms_envelope_correlation": 0.9978399264983322, "300–1000 Hz_40ms_modulation_mae_db": 0.5612677299228143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.204130707062224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07500746997949895, "1000–3000 Hz_10ms_envelope_correlation": 0.9965840945790114, "1000–3000 Hz_10ms_modulation_mae_db": 0.227772352296799, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6887094637119965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04172048580045421, "1000–3000 Hz_40ms_envelope_correlation": 0.9989245814589466, "1000–3000 Hz_40ms_modulation_mae_db": 0.634563895780013, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4017963358688479, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17190570013890247, "3000–6000 Hz_10ms_envelope_correlation": 0.9789793417717241, "3000–6000 Hz_10ms_modulation_mae_db": 0.8119730934561865, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8802165283437705, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09788398741598944, "3000–6000 Hz_40ms_envelope_correlation": 0.992778095043391, "3000–6000 Hz_40ms_modulation_mae_db": 0.7596672685588698, "mr_stft_db": 5.45592286227405, "band_mae_db": 0.1757506245907993, "rms_gain_error_db": 0.0016442224182365269, "output_peak": 0.8321295177007704}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7697886482188762, "全帯域_10ms_rms_envelope_nrmse": 0.059064279915946996, "全帯域_10ms_envelope_correlation": 0.997058816599296, "全帯域_10ms_modulation_mae_db": 0.16117385588425134, "全帯域_40ms_log_rms_mae_db": 0.4271688146460013, "全帯域_40ms_rms_envelope_nrmse": 0.03093112511666118, "全帯域_40ms_envelope_correlation": 0.9991579711410901, "全帯域_40ms_modulation_mae_db": 0.09022148836424859, "300–1000 Hz_10ms_log_rms_mae_db": 1.2307827142166445, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08389574872254059, "300–1000 Hz_10ms_envelope_correlation": 0.9946329859154839, "300–1000 Hz_10ms_modulation_mae_db": 0.47297609796916346, "300–1000 Hz_40ms_log_rms_mae_db": 0.7806008673020662, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05000541855588928, "300–1000 Hz_40ms_envelope_correlation": 0.9979965552214497, "300–1000 Hz_40ms_modulation_mae_db": 0.6322132288756707, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1852462002015787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07521278303082243, "1000–3000 Hz_10ms_envelope_correlation": 0.9965829453674863, "1000–3000 Hz_10ms_modulation_mae_db": 0.21570819963011176, "1000–3000 Hz_40ms_log_rms_mae_db": 0.697820610483156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04357623169341334, "1000–3000 Hz_40ms_envelope_correlation": 0.9988802767504502, "1000–3000 Hz_40ms_modulation_mae_db": 0.0957342887047462, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4215064280427476, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17462887688074208, "3000–6000 Hz_10ms_envelope_correlation": 0.9785491413191543, "3000–6000 Hz_10ms_modulation_mae_db": 0.7106076800101339, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9170074421563507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1060051722574169, "3000–6000 Hz_40ms_envelope_correlation": 0.9920791031478339, "3000–6000 Hz_40ms_modulation_mae_db": 0.7539310412151936, "mr_stft_db": 5.460954206420063, "band_mae_db": 0.14380176223855354, "rms_gain_error_db": -0.021083594374770063, "output_peak": 0.8453049798050059}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7852985141827914, "全帯域_10ms_rms_envelope_nrmse": 0.055636659058289926, "全帯域_10ms_envelope_correlation": 0.9974600310055128, "全帯域_10ms_modulation_mae_db": 0.5273785019340738, "全帯域_40ms_log_rms_mae_db": 0.39847054122308373, "全帯域_40ms_rms_envelope_nrmse": 0.02704756464035131, "全帯域_40ms_envelope_correlation": 0.9994458863398271, "全帯域_40ms_modulation_mae_db": 0.32386363767789056, "300–1000 Hz_10ms_log_rms_mae_db": 1.2075036704507844, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06697395730225829, "300–1000 Hz_10ms_envelope_correlation": 0.996595222257019, "300–1000 Hz_10ms_modulation_mae_db": 0.5008342892993426, "300–1000 Hz_40ms_log_rms_mae_db": 0.7278567649125648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03615270623136276, "300–1000 Hz_40ms_envelope_correlation": 0.9989624258983368, "300–1000 Hz_40ms_modulation_mae_db": 0.5715055866724404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1980501158444947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07212204137959899, "1000–3000 Hz_10ms_envelope_correlation": 0.9969446720250345, "1000–3000 Hz_10ms_modulation_mae_db": 0.6852303422922721, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6733208926015884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04176686654953003, "1000–3000 Hz_40ms_envelope_correlation": 0.9990727613247506, "1000–3000 Hz_40ms_modulation_mae_db": 0.5602517762474059, "3000–6000 Hz_10ms_log_rms_mae_db": 1.409481196943489, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17955305181927625, "3000–6000 Hz_10ms_envelope_correlation": 0.9770463674655316, "3000–6000 Hz_10ms_modulation_mae_db": 0.8931669876420746, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8549378203919622, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09409837587454581, "3000–6000 Hz_40ms_envelope_correlation": 0.9934221872722953, "3000–6000 Hz_40ms_modulation_mae_db": 0.7432935208400341, "mr_stft_db": 5.453284629379151, "band_mae_db": 0.16696681359728008, "rms_gain_error_db": -0.08615042140450897, "output_peak": 0.7644251062587818}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 82160, "output_samples": 82160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.440472285525868, "spectral_convergence": 0.2205266263292065, "spectral_convergence_status": "ok", "amplitude_floor": 2.2676184880983433e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.341092212541601, "spectral_convergence": 0.21184685932073521, "spectral_convergence_status": "ok", "amplitude_floor": 1.9761500134157035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.624495764018536, "spectral_convergence": 0.2188833926490949, "spectral_convergence_status": "ok", "amplitude_floor": 1.298774161363914e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.367365175011865e-05, "output_power": 9.581856030710018e-05, "delta_db": 0.5886091799585517, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016048781148481498, "output_power": 0.0016423525636586169, "delta_db": 0.10024337971436315, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001327868811244639, "output_power": 0.0013195992214151738, "delta_db": -0.027131197556234166, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0026813637139198997, "output_power": 0.0026079627659939897, "delta_db": -0.12054341246979676, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003087131182449005, "output_power": 0.0002855897375441005, "delta_db": -0.3381248715433598, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00020800689998023804, "output_power": 0.0002145524186698036, "delta_db": 0.13455673172448776, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21820146216113223, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03411758622743856, "pitch": null, "output_rms_dbfs": -22.100052650694234, "output_peak": 0.7527774881333261, "reference_rms_dbfs": -22.065935064466796, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.468686754028668}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 82160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003747956436237944, "log_rms_mae_db": 0.8347144717732644, "rms_envelope_nrmse": 0.055046469525508825, "envelope_correlation": 0.9974555878623497, "rms_gain_error_db": -0.034116991405763686, "reference_mean_power": 0.006220557891926456, "output_mean_power": 0.006171882325034528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.814448626587233e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017308539978626828, "output_power": 0.0017200922148050476, "reference_active": true, "delta_db": -0.027087053001496292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0006400481816254359, "output_power": 0.0006120160723541988, "reference_active": true, "delta_db": -0.1944984064297709}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4706618104713675e-05, "output_power": 3.126801694140406e-05, "reference_active": true, "delta_db": -1.5527147969958424}], "active_band_count": 3, "active_band_mae_db": 0.5914334188090365}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.00031163820648484283, "log_rms_mae_db": 0.4439642362685156, "rms_envelope_nrmse": 0.02762015051198449, "envelope_correlation": 0.9993397203987706, "rms_gain_error_db": -0.0341112243279135, "reference_mean_power": 0.006257088451954998, "output_mean_power": 0.006208135279078222}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 6.546091034740867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001677353066756132, "output_power": 0.0016543399343236843, "reference_active": true, "delta_db": -0.0599973369572277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00035132010859926165, "output_power": 0.0003357564875168455, "reference_active": true, "delta_db": -0.19678595429620654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3827329510863976e-06, "output_power": 1.6421055131513564e-06, "reference_active": true, "delta_db": -1.616743115867939}], "active_band_count": 3, "active_band_mae_db": 0.6245088023737911}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.00015152624573583172, "log_rms_mae_db": 1.1625977564025087, "rms_envelope_nrmse": 0.06924032370995103, "envelope_correlation": 0.9963478058543674, "rms_gain_error_db": 0.0004505772616337955, "reference_mean_power": 0.001672933915423699, "output_mean_power": 0.0016731074900643251}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 1.623496524404657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006256720367801861, "output_power": 0.0006191830637330084, "reference_active": true, "delta_db": -0.045276766928687255}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015725509039345338, "output_power": 0.0001425599773282236, "reference_active": true, "delta_db": -0.42607094959211744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4941700779965039e-05, "output_power": 1.2514031281493533e-05, "reference_active": true, "delta_db": -0.7700279879589385}], "active_band_count": 3, "active_band_mae_db": 0.4137919014932477}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0001385613366624899, "log_rms_mae_db": 0.7300427234040013, "rms_envelope_nrmse": 0.040070173221208046, "envelope_correlation": 0.9987125823711558, "rms_gain_error_db": 0.00045026379759218843, "reference_mean_power": 0.0016827742497549935, "output_mean_power": 0.0016829487239043961}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 1.5301470087365327e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005852793815004822, "output_power": 0.0005773017466087093, "reference_active": true, "delta_db": -0.05960353406527736}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.151264176170612e-05, "output_power": 7.610091451972379e-05, "reference_active": true, "delta_db": -0.29835092709550026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3333886770336671e-06, "output_power": 9.479881809024548e-07, "reference_active": true, "delta_db": -1.481538400517202}], "active_band_count": 3, "active_band_mae_db": 0.6131642872259933}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003048971769130071, "log_rms_mae_db": 1.1995877012835643, "rms_envelope_nrmse": 0.07549856982655702, "envelope_correlation": 0.9965981648307219, "rms_gain_error_db": -0.13042322355892413, "reference_mean_power": 0.0025898505670984154, "output_mean_power": 0.002513230860688609}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 4.739471399174337e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014413951986732973, "output_power": 0.0014352803773137563, "reference_active": true, "delta_db": -0.018463237190868478}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0004245147064825657, "output_power": 0.00037995434859688736, "reference_active": true, "delta_db": -0.48161320727205303}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3891971817365186e-05, "output_power": 1.633865159898866e-05, "reference_active": true, "delta_db": -1.6503578169422104}], "active_band_count": 3, "active_band_mae_db": 0.7168114204683773}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0002474836728453422, "log_rms_mae_db": 0.7158852531348593, "rms_envelope_nrmse": 0.04174246122046672, "envelope_correlation": 0.9990318957261519, "rms_gain_error_db": -0.13040742406512562, "reference_mean_power": 0.002605032676611572, "output_mean_power": 0.002527973010154473}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 4.638343817264169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014427582782272192, "output_power": 0.0014184524483545158, "reference_active": true, "delta_db": -0.07378793615893874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020918188683306615, "output_power": 0.00018431909744314102, "reference_active": true, "delta_db": -0.5495374090630547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0082267328987986e-06, "output_power": 1.1935671302286441e-06, "reference_active": true, "delta_db": -2.2596589376330813}], "active_band_count": 3, "active_band_mae_db": 0.9609947609516917}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 8.624763428459013e-05, "log_rms_mae_db": 1.4716025131449897, "rms_envelope_nrmse": 0.16628491304997156, "envelope_correlation": 0.9804673886409804, "rms_gain_error_db": 0.008897012423547259, "reference_mean_power": 0.00014375800668184303, "output_mean_power": 0.00014405281304594625}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.640530401894145e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.056434359733932e-05, "output_power": 2.1863027371848007e-05, "reference_active": true, "delta_db": 0.2659544688088607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.844428212989113e-05, "output_power": 1.5391977541904285e-05, "reference_active": true, "delta_db": -0.7856733568674378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.64438001039378e-06, "output_power": 2.4390401372727756e-06, "reference_active": true, "delta_db": -1.7440470864672861}], "active_band_count": 3, "active_band_mae_db": 0.9318916373811948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 6.581694648038094e-05, "log_rms_mae_db": 0.8983196481077018, "rms_envelope_nrmse": 0.0869494264363567, "envelope_correlation": 0.994263067245984, "rms_gain_error_db": 0.008845538353428373, "reference_mean_power": 0.00014459942402382801, "output_mean_power": 0.00014489423854678477}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 5.765239137842014e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0474191785118657e-05, "output_power": 2.136373243632084e-05, "reference_active": true, "delta_db": 0.1847036306218916}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1083544144974636e-05, "output_power": 8.974203202702478e-06, "reference_active": true, "delta_db": -0.9168275635258205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.842442602727836e-07, "output_power": 1.9353764996454608e-07, "reference_active": true, "delta_db": -1.6692624045087623}], "active_band_count": 3, "active_band_mae_db": 0.9235978662188248}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8689689050087748, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dc072fcc25eab66cf15734b80ba5dbfc6f8c06624ed4258d97506b67dfd85bb3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9965726061455203, "standard_deviation": 0.0004471435636399464, "minimum": 0.9956795330137826, "maximum": 0.9971934695433528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2202611880014236, "standard_deviation": 0.032052583935709145, "minimum": 1.1852462002015787, "maximum": 1.2740657939353315, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4910024906091487, "standard_deviation": 0.3107481034751654, "minimum": 0.17877820637743017, "maximum": 0.947233475647212, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07515494460835528, "standard_deviation": 0.004648833152042588, "minimum": 0.06784494308063738, "maximum": 0.08403480280936307, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9990436838380233, "standard_deviation": 0.00012889948011033793, "minimum": 0.9988802767504502, "maximum": 0.9992281886562584, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6963650085102233, "standard_deviation": 0.0174314662924555, "minimum": 0.6733208926015884, "maximum": 0.7265216762299413, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4608624168434451, "standard_deviation": 0.3167851028359096, "minimum": 0.0957342887047462, "maximum": 0.9609947609516917, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.040168749612186354, "standard_deviation": 0.003053345005557635, "minimum": 0.03535550966456633, "maximum": 0.04357623169341334, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9784554830869354, "standard_deviation": 0.0017285529672501126, "minimum": 0.9751818592112869, "maximum": 0.9804673886409804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.421150159740613, "standard_deviation": 0.02400207676518439, "minimum": 1.4000950343528498, "maximum": 1.4716025131449897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6890432196552845, "standard_deviation": 0.24035323650453896, "minimum": 0.29685680868647707, "maximum": 0.9318916373811948, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17423878521400787, "standard_deviation": 0.006978818630898733, "minimum": 0.16593139997779, "maximum": 0.18743348766313148, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9931519327162677, "standard_deviation": 0.0009297702423099311, "minimum": 0.9918339360477144, "maximum": 0.9944553754386211, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.876700571964724, "standard_deviation": 0.026056880938626487, "minimum": 0.8396681285519358, "maximum": 0.9170074421563507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6852183187147429, "standard_deviation": 0.23963318665383707, "minimum": 0.32811355197207687, "maximum": 0.984067837268034, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09660456724271259, "standard_deviation": 0.007954660037458454, "minimum": 0.08606641882894422, "maximum": 0.10822911698416549, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9954038976395976, "standard_deviation": 0.0007309430936846944, "minimum": 0.9946277776205001, "maximum": 0.996595222257019, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2042889623824915, "standard_deviation": 0.02563007287511475, "minimum": 1.1625977564025087, "maximum": 1.2379517106085947, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5116095794063995, "standard_deviation": 0.24420883362412832, "minimum": 0.25393432209536604, "maximum": 0.9375726273643665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07759907509887976, "standard_deviation": 0.006388011365513387, "minimum": 0.06697395730225829, "maximum": 0.08401757578880296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983872438350339, "standard_deviation": 0.0003749337693349001, "minimum": 0.9978399264983322, "maximum": 0.9989624258983368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.752124754915763, "standard_deviation": 0.028595950025827304, "minimum": 0.711345874531809, "maximum": 0.7856513671383577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7625249711246493, "standard_deviation": 0.2899369301064662, "minimum": 0.5458504371272583, "maximum": 1.238380375977733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04484674130379315, "standard_deviation": 0.0051737257376188485, "minimum": 0.03615270623136276, "maximum": 0.05195518755032416, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16555515467350268, "standard_deviation": 0.025268102363679195, "minimum": 0.14108887788597493, "maximum": 0.21820146216113223, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.458601132396767, "standard_deviation": 0.006665911089737386, "minimum": 5.446628903800987, "maximum": 5.468686754028668, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7789534315981976, "standard_deviation": 0.03894090117674395, "minimum": 0.7413454142315021, "maximum": 0.8453049798050059, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.026610047409248928, "standard_deviation": 0.027386114930748523, "minimum": -0.08615042140450897, "maximum": 0.0016442224182365269, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.997142208120203, "standard_deviation": 0.00021247947448128766, "minimum": 0.9968727711860373, "maximum": 0.9974600310055128, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8064725436910779, "standard_deviation": 0.021548831834990897, "minimum": 0.7697886482188762, "maximum": 0.8347144717732644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4434897600184158, "standard_deviation": 0.2733742915136761, "minimum": 0.15562347145344602, "maximum": 0.9216390626301051, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.058309940410817634, "standard_deviation": 0.0020349802265093643, "minimum": 0.055046469525508825, "maximum": 0.06099266703293123, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992350570481459, "standard_deviation": 0.0001455898679285238, "minimum": 0.9989635339817959, "maximum": 0.9994458863398271, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42827796526846273, "standard_deviation": 0.023586198511407534, "minimum": 0.39847054122308373, "maximum": 0.4726786295874565, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3808259270331472, "standard_deviation": 0.2211749247598456, "minimum": 0.09022148836424859, "maximum": 0.7383578457364764, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02966515125924856, "standard_deviation": 0.002390461242575482, "minimum": 0.02704756464035131, "maximum": 0.03439806866810821, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_5683-32879-0019", "language": "en", "style": "normal", "speaker": "librispeech_5683", "source_utterance_id": "5683-32879-0019", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_5683-32879-0019.wav", "source_file_sha256": "c8c3389dc76d8e23574147d00809d2cbaff8e1d274e629ef465ddbbdf3f708b9", "decoded_pcm_sha256": "cf57296e7013c02425f574c6c3f30623e8057ab1e199fc8fcd761f1b2ff63d52", "original_source_sha256": "0c9375f8c28b1576aaa73e07006a7fc838f49b250df910885e24a16abef0c19f", "corpus_entry_id": "librispeech_5683-32879-0019"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.435281991958618, "predicted_mos": 3.435281991958618, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "94acd51133660839055d5b2df725cf443deb1a9a8b39c3a85c4f6c4a7158a4c8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8136900309982593, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_5683-32879-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "94acd51133660839055d5b2df725cf443deb1a9a8b39c3a85c4f6c4a7158a4c8", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "ba9e262f1a4e5d33d19cd97573eda08b9fced12e76deaf30d03c322a36da8b82"}, "rows": [{"case": "librispeech_5683-32879-0019", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.27073335647583, "predicted_mos": 2.27073335647583, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5d9956a668bbb108f0e600bfae1c4e3b303ebba7cbfd70acd1120b623f020545", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.12090445099966, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_5683-32879-0019", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5d9956a668bbb108f0e600bfae1c4e3b303ebba7cbfd70acd1120b623f020545", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8103555096433547, "全帯域_10ms_rms_envelope_nrmse": 0.07704562879786725, "全帯域_10ms_envelope_correlation": 0.9947607726404084, "全帯域_10ms_modulation_mae_db": 0.35244732989136845, "全帯域_40ms_log_rms_mae_db": 0.457666852215418, "全帯域_40ms_rms_envelope_nrmse": 0.03759829435352359, "全帯域_40ms_envelope_correlation": 0.9987745178770134, "全帯域_40ms_modulation_mae_db": 0.6547187004384797, "300–1000 Hz_10ms_log_rms_mae_db": 1.2827855092713094, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07902066006401759, "300–1000 Hz_10ms_envelope_correlation": 0.9951420144567047, "300–1000 Hz_10ms_modulation_mae_db": 0.5929168948080571, "300–1000 Hz_40ms_log_rms_mae_db": 0.8241624801312325, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04809771608457601, "300–1000 Hz_40ms_envelope_correlation": 0.9982144844247807, "300–1000 Hz_40ms_modulation_mae_db": 0.7464310490122327, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2872860848918397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11624529200150487, "1000–3000 Hz_10ms_envelope_correlation": 0.9906374325586705, "1000–3000 Hz_10ms_modulation_mae_db": 0.7333107163018013, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7628344758863421, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05763046963586352, "1000–3000 Hz_40ms_envelope_correlation": 0.9977156370255218, "1000–3000 Hz_40ms_modulation_mae_db": 0.7984971266243034, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4778603101309224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20645748684049697, "3000–6000 Hz_10ms_envelope_correlation": 0.9655529856456824, "3000–6000 Hz_10ms_modulation_mae_db": 0.681235734983631, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8937690926730725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11812630854833626, "3000–6000 Hz_40ms_envelope_correlation": 0.9878073988427507, "3000–6000 Hz_40ms_modulation_mae_db": 0.4687802117001903, "mr_stft_db": 5.535459827773445, "band_mae_db": 0.31978599424316423, "rms_gain_error_db": -0.10334722441678948, "output_peak": 0.7555893248712029}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7910739452668658, "全帯域_10ms_rms_envelope_nrmse": 0.0840651426339448, "全帯域_10ms_envelope_correlation": 0.9937294540929997, "全帯域_10ms_modulation_mae_db": 0.2575120688394032, "全帯域_40ms_log_rms_mae_db": 0.44050073333826306, "全帯域_40ms_rms_envelope_nrmse": 0.04996826197186397, "全帯域_40ms_envelope_correlation": 0.9976368815695036, "全帯域_40ms_modulation_mae_db": 0.17164953510403133, "300–1000 Hz_10ms_log_rms_mae_db": 1.2830145353858868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09152939595441585, "300–1000 Hz_10ms_envelope_correlation": 0.9933536753241529, "300–1000 Hz_10ms_modulation_mae_db": 1.0925998840491837, "300–1000 Hz_40ms_log_rms_mae_db": 0.8611384577010626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06174509880138809, "300–1000 Hz_40ms_envelope_correlation": 0.9968212195035533, "300–1000 Hz_40ms_modulation_mae_db": 1.365279863480974, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2860137691462135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13516073775133594, "1000–3000 Hz_10ms_envelope_correlation": 0.988230189444292, "1000–3000 Hz_10ms_modulation_mae_db": 0.27728824452291234, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8030719473788027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08683381982383698, "1000–3000 Hz_40ms_envelope_correlation": 0.9952072603778311, "1000–3000 Hz_40ms_modulation_mae_db": 0.34649670224530144, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4279121462790316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17232831429467127, "3000–6000 Hz_10ms_envelope_correlation": 0.9750005558127277, "3000–6000 Hz_10ms_modulation_mae_db": 0.6360039393787674, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8689040524832102, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09234827527046921, "3000–6000 Hz_40ms_envelope_correlation": 0.992521488395628, "3000–6000 Hz_40ms_modulation_mae_db": 0.6770301130877217, "mr_stft_db": 5.544032342536828, "band_mae_db": 0.3288614245825741, "rms_gain_error_db": 0.04459335861491698, "output_peak": 0.826416400449523}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8344745907218258, "全帯域_10ms_rms_envelope_nrmse": 0.08039376827771674, "全帯域_10ms_envelope_correlation": 0.9943878484953147, "全帯域_10ms_modulation_mae_db": 0.8273313823765683, "全帯域_40ms_log_rms_mae_db": 0.4568011909421028, "全帯域_40ms_rms_envelope_nrmse": 0.05018236405374289, "全帯域_40ms_envelope_correlation": 0.9978117849562731, "全帯域_40ms_modulation_mae_db": 1.2674225915381996, "300–1000 Hz_10ms_log_rms_mae_db": 1.2842027401497758, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08366011815464405, "300–1000 Hz_10ms_envelope_correlation": 0.9945572187087924, "300–1000 Hz_10ms_modulation_mae_db": 0.65558882613719, "300–1000 Hz_40ms_log_rms_mae_db": 0.8476492335883107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05101744119128939, "300–1000 Hz_40ms_envelope_correlation": 0.9979910820406045, "300–1000 Hz_40ms_modulation_mae_db": 0.9147499243442822, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3060302150567284, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13991146092052864, "1000–3000 Hz_10ms_envelope_correlation": 0.9866124086706862, "1000–3000 Hz_10ms_modulation_mae_db": 1.2935460939343273, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8175744933960128, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09003045720697717, "1000–3000 Hz_40ms_envelope_correlation": 0.9945208396449594, "1000–3000 Hz_40ms_modulation_mae_db": 1.6426970454895613, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4871352051225788, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19306943935324614, "3000–6000 Hz_10ms_envelope_correlation": 0.9693869976436431, "3000–6000 Hz_10ms_modulation_mae_db": 0.6375466400040214, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9070919015225711, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1211802282172051, "3000–6000 Hz_40ms_envelope_correlation": 0.9867392203191706, "3000–6000 Hz_40ms_modulation_mae_db": 0.7138958413479073, "mr_stft_db": 5.562012022016425, "band_mae_db": 0.4324997561337793, "rms_gain_error_db": -0.14090668281720956, "output_peak": 0.8576683644380241}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.763875677899942, "全帯域_10ms_rms_envelope_nrmse": 0.0730727331477826, "全帯域_10ms_envelope_correlation": 0.9953264261792855, "全帯域_10ms_modulation_mae_db": 0.23534373226843744, "全帯域_40ms_log_rms_mae_db": 0.42949625983300216, "全帯域_40ms_rms_envelope_nrmse": 0.037759984616421, "全帯域_40ms_envelope_correlation": 0.9987901364215384, "全帯域_40ms_modulation_mae_db": 0.38185780817525017, "300–1000 Hz_10ms_log_rms_mae_db": 1.3535039893640577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09894663524658927, "300–1000 Hz_10ms_envelope_correlation": 0.9922861750388736, "300–1000 Hz_10ms_modulation_mae_db": 0.6561530775243362, "300–1000 Hz_40ms_log_rms_mae_db": 0.9022615927635632, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.067019695001408, "300–1000 Hz_40ms_envelope_correlation": 0.9963615594789621, "300–1000 Hz_40ms_modulation_mae_db": 0.7436848984011079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3199248582317185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11295876160564751, "1000–3000 Hz_10ms_envelope_correlation": 0.9910919000317029, "1000–3000 Hz_10ms_modulation_mae_db": 0.33586326365632274, "1000–3000 Hz_40ms_log_rms_mae_db": 0.793510839782101, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05737092408366975, "1000–3000 Hz_40ms_envelope_correlation": 0.9975942261305046, "1000–3000 Hz_40ms_modulation_mae_db": 0.4163703948611975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4136371879974454, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16800155466283356, "3000–6000 Hz_10ms_envelope_correlation": 0.9762173342849292, "3000–6000 Hz_10ms_modulation_mae_db": 0.42247983791218197, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8468663772416393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09067027095574938, "3000–6000 Hz_40ms_envelope_correlation": 0.9922867535627478, "3000–6000 Hz_40ms_modulation_mae_db": 0.47802506265033307, "mr_stft_db": 5.526170679405108, "band_mae_db": 0.3267537878277954, "rms_gain_error_db": -0.11832995504046555, "output_peak": 0.8202852117840677}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7523677099940609, "全帯域_10ms_rms_envelope_nrmse": 0.07486799671122268, "全帯域_10ms_envelope_correlation": 0.9951045463410498, "全帯域_10ms_modulation_mae_db": 0.5805281254425555, "全帯域_40ms_log_rms_mae_db": 0.4195286922638613, "全帯域_40ms_rms_envelope_nrmse": 0.03907974268454705, "全帯域_40ms_envelope_correlation": 0.9987103834663702, "全帯域_40ms_modulation_mae_db": 0.8480347109489191, "300–1000 Hz_10ms_log_rms_mae_db": 1.3050202927630759, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09686109528788898, "300–1000 Hz_10ms_envelope_correlation": 0.992641858215521, "300–1000 Hz_10ms_modulation_mae_db": 0.9637450934165052, "300–1000 Hz_40ms_log_rms_mae_db": 0.854659678230314, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06527225646548812, "300–1000 Hz_40ms_envelope_correlation": 0.9965780455396809, "300–1000 Hz_40ms_modulation_mae_db": 1.1213178347045372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.312741275749283, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11438583704440818, "1000–3000 Hz_10ms_envelope_correlation": 0.9909511001846502, "1000–3000 Hz_10ms_modulation_mae_db": 0.8408096246043265, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7982427827536686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06482966666780293, "1000–3000 Hz_40ms_envelope_correlation": 0.9970698251697541, "1000–3000 Hz_40ms_modulation_mae_db": 0.9674452370343296, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4436919334671594, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20198912132469748, "3000–6000 Hz_10ms_envelope_correlation": 0.9654994160482117, "3000–6000 Hz_10ms_modulation_mae_db": 0.682971047791598, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9162058440024156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11620982297661442, "3000–6000 Hz_40ms_envelope_correlation": 0.9872953105929715, "3000–6000 Hz_40ms_modulation_mae_db": 0.6959185012578343, "mr_stft_db": 5.53960896874477, "band_mae_db": 0.3244032990133036, "rms_gain_error_db": -0.12150889013205146, "output_peak": 0.895570526910964}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8202564555807739, "全帯域_10ms_rms_envelope_nrmse": 0.08716553047148369, "全帯域_10ms_envelope_correlation": 0.9934566943005495, "全帯域_10ms_modulation_mae_db": 0.5323591207143744, "全帯域_40ms_log_rms_mae_db": 0.47234390070219734, "全帯域_40ms_rms_envelope_nrmse": 0.05530942760123261, "全帯域_40ms_envelope_correlation": 0.9974190574952503, "全帯域_40ms_modulation_mae_db": 1.0096602386877727, "300–1000 Hz_10ms_log_rms_mae_db": 1.2765176710125736, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10913758001063582, "300–1000 Hz_10ms_envelope_correlation": 0.9905663151071359, "300–1000 Hz_10ms_modulation_mae_db": 0.5525845981529584, "300–1000 Hz_40ms_log_rms_mae_db": 0.8732234671223653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0810284547324046, "300–1000 Hz_40ms_envelope_correlation": 0.9945678184982214, "300–1000 Hz_40ms_modulation_mae_db": 0.6616067332066757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3240223839506438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1253075525864963, "1000–3000 Hz_10ms_envelope_correlation": 0.9897316137997393, "1000–3000 Hz_10ms_modulation_mae_db": 0.9510986067350425, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8352641021375268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08135671541261837, "1000–3000 Hz_40ms_envelope_correlation": 0.9960596990896883, "1000–3000 Hz_40ms_modulation_mae_db": 1.399219664613361, "3000–6000 Hz_10ms_log_rms_mae_db": 1.440566803739767, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19595779991272994, "3000–6000 Hz_10ms_envelope_correlation": 0.967427189826512, "3000–6000 Hz_10ms_modulation_mae_db": 0.4173284696948378, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8901888743991033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10933486660392944, "3000–6000 Hz_40ms_envelope_correlation": 0.9892972010147204, "3000–6000 Hz_40ms_modulation_mae_db": 0.46761925192848297, "mr_stft_db": 5.542557759711232, "band_mae_db": 0.3687116552847631, "rms_gain_error_db": -0.17226303329563378, "output_peak": 0.7988039404566406}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8017396993500671, "全帯域_10ms_rms_envelope_nrmse": 0.07987589771755926, "全帯域_10ms_envelope_correlation": 0.9942806839358718, "全帯域_10ms_modulation_mae_db": 0.2230588886366062, "全帯域_40ms_log_rms_mae_db": 0.44264954959319763, "全帯域_40ms_rms_envelope_nrmse": 0.047749757743858375, "全帯域_40ms_envelope_correlation": 0.9978097409770028, "全帯域_40ms_modulation_mae_db": 0.19566363147822433, "300–1000 Hz_10ms_log_rms_mae_db": 1.2860521706673496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09102450383464912, "300–1000 Hz_10ms_envelope_correlation": 0.9935679543919474, "300–1000 Hz_10ms_modulation_mae_db": 0.32285185725262394, "300–1000 Hz_40ms_log_rms_mae_db": 0.8232501865374291, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06578063662550378, "300–1000 Hz_40ms_envelope_correlation": 0.9965652230527265, "300–1000 Hz_40ms_modulation_mae_db": 0.4166654739967905, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2985411701488487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12486588558644254, "1000–3000 Hz_10ms_envelope_correlation": 0.9897947180521776, "1000–3000 Hz_10ms_modulation_mae_db": 0.15458907529876134, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7918850563702611, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08435784495772175, "1000–3000 Hz_40ms_envelope_correlation": 0.9952865198215856, "1000–3000 Hz_40ms_modulation_mae_db": 0.3216624647430764, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4380343342650554, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.195138132077531, "3000–6000 Hz_10ms_envelope_correlation": 0.9676918628150127, "3000–6000 Hz_10ms_modulation_mae_db": 0.38717552724368715, "3000–6000 Hz_40ms_log_rms_mae_db": 0.851529431506477, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10074480232628155, "3000–6000 Hz_40ms_envelope_correlation": 0.9905137401564216, "3000–6000 Hz_40ms_modulation_mae_db": 0.5270650265708676, "mr_stft_db": 5.544800782713466, "band_mae_db": 0.310046163290941, "rms_gain_error_db": -0.018923556597858437, "output_peak": 0.8935858623427444}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7653788777503185, "全帯域_10ms_rms_envelope_nrmse": 0.07553020519215647, "全帯域_10ms_envelope_correlation": 0.9948940015303095, "全帯域_10ms_modulation_mae_db": 0.3776407731746229, "全帯域_40ms_log_rms_mae_db": 0.4507570584337418, "全帯域_40ms_rms_envelope_nrmse": 0.04634003976318648, "全帯域_40ms_envelope_correlation": 0.9979528223539647, "全帯域_40ms_modulation_mae_db": 0.7264858712335228, "300–1000 Hz_10ms_log_rms_mae_db": 1.2876799059733783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08529066549878177, "300–1000 Hz_10ms_envelope_correlation": 0.9942121952253886, "300–1000 Hz_10ms_modulation_mae_db": 1.0051413843261534, "300–1000 Hz_40ms_log_rms_mae_db": 0.8359990532976025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052058541859730266, "300–1000 Hz_40ms_envelope_correlation": 0.9977604211104749, "300–1000 Hz_40ms_modulation_mae_db": 1.234505597521395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3092069318988353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11324823119146807, "1000–3000 Hz_10ms_envelope_correlation": 0.9912138316151301, "1000–3000 Hz_10ms_modulation_mae_db": 0.6437566336385511, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8094374647627458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06944196419898907, "1000–3000 Hz_40ms_envelope_correlation": 0.9965461864641209, "1000–3000 Hz_40ms_modulation_mae_db": 0.8177079566942133, "3000–6000 Hz_10ms_log_rms_mae_db": 1.458011343007558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17302164290383776, "3000–6000 Hz_10ms_envelope_correlation": 0.9756420117444812, "3000–6000 Hz_10ms_modulation_mae_db": 0.3189859686300276, "3000–6000 Hz_40ms_log_rms_mae_db": 0.846450775952587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09212930237939893, "3000–6000 Hz_40ms_envelope_correlation": 0.9925430726507231, "3000–6000 Hz_40ms_modulation_mae_db": 0.18352383790930635, "mr_stft_db": 5.533275619706444, "band_mae_db": 0.2945751737101076, "rms_gain_error_db": -0.03439224120303485, "output_peak": 0.8945723610755761}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.439977316249001, "spectral_convergence": 0.26422623770749254, "spectral_convergence_status": "ok", "amplitude_floor": 2.005860984254184e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.447016132518041, "spectral_convergence": 0.2640650694445915, "spectral_convergence_status": "ok", "amplitude_floor": 1.6673734947378395e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7193860345532945, "spectral_convergence": 0.2772256526438215, "spectral_convergence_status": "ok", "amplitude_floor": 1.455049241978443e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.2140486659240935e-05, "output_power": 4.420034620570938e-05, "delta_db": 1.3837322263588865, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002187325069648938, "output_power": 0.0021463076853514163, "delta_db": -0.08221349910491985, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018573993391017246, "output_power": 0.0017975586806664888, "delta_db": -0.14222210005862654, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.002022873257906857, "output_power": 0.0019627863281581736, "delta_db": -0.13095648894419218, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00023141111513331255, "output_power": 0.0002290927482250829, "delta_db": -0.043728730282650556, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.19984427950361e-05, "output_power": 4.3333072168088325e-05, "delta_db": 0.13586292070970973, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.31978599424316423, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10334722441678948, "pitch": null, "output_rms_dbfs": -22.059807381474002, "output_peak": 0.7555893248712029, "reference_rms_dbfs": -21.95646015705721, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.535459827773445}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003217977906320387, "log_rms_mae_db": 0.8103555096433547, "rms_envelope_nrmse": 0.07704562879786725, "envelope_correlation": 0.9947607726404084, "rms_gain_error_db": -0.10328135185529838, "reference_mean_power": 0.006378025340653872, "output_mean_power": 0.0062281362553477555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.934428693189062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0026409373563640536, "output_power": 0.0025576492217585247, "reference_active": true, "delta_db": -0.139171184577425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022353254716419522, "output_power": 0.00019941340766805506, "reference_active": true, "delta_db": -0.49586411970959443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.171045723198585e-05, "output_power": 2.877211818900417e-05, "reference_active": true, "delta_db": -0.42230668538708593}], "active_band_count": 3, "active_band_mae_db": 0.35244732989136845}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00029448546471938644, "log_rms_mae_db": 0.457666852215418, "rms_envelope_nrmse": 0.03759829435352359, "envelope_correlation": 0.9987745178770134, "rms_gain_error_db": -0.1032280194844627, "reference_mean_power": 0.006407121641291136, "output_mean_power": 0.006256625600219225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7183045790225306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0025278777543949896, "output_power": 0.0024445158503151186, "reference_active": true, "delta_db": -0.14563210489753708}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010431651037251879, "output_power": 9.214548930171502e-05, "reference_active": true, "delta_db": -0.5387896995210066}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4474914310652027e-06, "output_power": 1.822836643039528e-06, "reference_active": true, "delta_db": -1.2797342968968952}], "active_band_count": 3, "active_band_mae_db": 0.6547187004384797}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002169046697405618, "log_rms_mae_db": 1.2827855092713094, "rms_envelope_nrmse": 0.07902066006401759, "envelope_correlation": 0.9951420144567047, "rms_gain_error_db": -0.13570003334337105, "reference_mean_power": 0.00266513165861851, "output_mean_power": 0.0025831442846057386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9887000298118345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014141653900070267, "output_power": 0.0013512634214310456, "reference_active": true, "delta_db": -0.19760183589162206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010731466350302803, "output_power": 8.820456867158363e-05, "reference_active": true, "delta_db": -0.8516798756149523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0173327828420217e-05, "output_power": 1.7054172395788664e-05, "reference_active": true, "delta_db": -0.7294689729175972}], "active_band_count": 3, "active_band_mae_db": 0.5929168948080571}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020963310569053025, "log_rms_mae_db": 0.8241624801312325, "rms_envelope_nrmse": 0.04809771608457601, "envelope_correlation": 0.9982144844247807, "rms_gain_error_db": -0.13572530559258325, "reference_mean_power": 0.0026772178601939312, "output_mean_power": 0.002594843578855926}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.883939001913369e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013521293519634116, "output_power": 0.0012893852346769173, "reference_active": true, "delta_db": -0.20635547933598997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.792726117783695e-05, "output_power": 3.9027001505244935e-05, "reference_active": true, "delta_db": -0.8921742661312883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6850826685161704e-06, "output_power": 1.2958205757729294e-06, "reference_active": true, "delta_db": -1.1407634015694197}], "active_band_count": 3, "active_band_mae_db": 0.7464310490122327}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002744641361828464, "log_rms_mae_db": 1.2872860848918397, "rms_envelope_nrmse": 0.11624529200150487, "envelope_correlation": 0.9906374325586705, "rms_gain_error_db": -0.13920552038364808, "reference_mean_power": 0.0019381811578355007, "output_mean_power": 0.0018770412429877307}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1196235236571854e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006735796402683569, "output_power": 0.0006609379383759047, "reference_active": true, "delta_db": -0.08228270188971278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013823323993445422, "output_power": 0.00013272295430129496, "reference_active": true, "delta_db": -0.17666447054725756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8792319908524634e-05, "output_power": 1.2019375390985084e-05, "reference_active": true, "delta_db": -1.9409849764684337}], "active_band_count": 3, "active_band_mae_db": 0.7333107163018013}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002445180716567873, "log_rms_mae_db": 0.7628344758863421, "rms_envelope_nrmse": 0.05763046963586352, "envelope_correlation": 0.9977156370255218, "rms_gain_error_db": -0.13895934037892602, "reference_mean_power": 0.0019467823981069799, "output_mean_power": 0.001885478032548488}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0571732642454036e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006649824803295284, "output_power": 0.0006489934773038983, "reference_active": true, "delta_db": -0.1056987154974092}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927173921912536e-05, "output_power": 6.616573247390724e-05, "reference_active": true, "delta_db": -0.19922967063391275}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7481239585012505e-06, "output_power": 1.0802292322076736e-06, "reference_active": true, "delta_db": -2.0905629937415884}], "active_band_count": 3, "active_band_mae_db": 0.7984971266243034}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.6882736857616036e-05, "log_rms_mae_db": 1.4778603101309224, "rms_envelope_nrmse": 0.20645748684049697, "envelope_correlation": 0.9655529856456824, "rms_gain_error_db": 0.19141187970875576, "reference_mean_power": 6.564824297759055e-05, "output_mean_power": 6.860634690927137e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.001422573299629e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.567613726314846e-05, "output_power": 1.7162343038105797e-05, "reference_active": true, "delta_db": 0.39337520744238325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.79643479102731e-06, "output_power": 6.5433370677661654e-06, "reference_active": true, "delta_db": 0.5263833825820006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.958485457614687e-07, "output_power": 7.68773548565068e-07, "reference_active": true, "delta_db": -1.1239486149265092}], "active_band_count": 3, "active_band_mae_db": 0.681235734983631}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.766393520730197e-05, "log_rms_mae_db": 0.8937690926730725, "rms_envelope_nrmse": 0.11812630854833626, "envelope_correlation": 0.9878073988427507, "rms_gain_error_db": 0.19211357094201587, "reference_mean_power": 6.585532886735065e-05, "output_mean_power": 6.883388468894128e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6817222442955538e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5480772645031266e-05, "output_power": 1.7385391374814336e-05, "reference_active": true, "delta_db": 0.5039183932279324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.979992150286795e-06, "output_power": 3.1612824964716353e-06, "reference_active": true, "delta_db": 0.2564818662912577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.719812912358681e-08, "output_power": 7.514725643656129e-08, "reference_active": true, "delta_db": -0.6459403755813808}], "active_band_count": 3, "active_band_mae_db": 0.4687802117001903}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.869102276345456, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5a816584321cc11765d785e8b666897492c6e6f6af0c606a303c3518d7e71d94", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897828992946311, "standard_deviation": 0.0016178318478389138, "minimum": 0.9866124086706862, "maximum": 0.9912138316151301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3054708361342637, "standard_deviation": 0.01402799341067438, "minimum": 1.2860137691462135, "maximum": 1.3240223839506438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6537827823365057, "standard_deviation": 0.3837993085080801, "minimum": 0.15458907529876134, "maximum": 1.2935460939343273, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.122760469835979, "standard_deviation": 0.010405916795241105, "minimum": 0.11295876160564751, "maximum": 0.13991146092052864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962500242154957, "standard_deviation": 0.0011805982412113527, "minimum": 0.9945208396449594, "maximum": 0.9977156370255218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8014776453084326, "standard_deviation": 0.02113149423362285, "minimum": 0.7628344758863421, "maximum": 0.8352641021375268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.838762074038168, "standard_deviation": 0.4880251721083102, "minimum": 0.3216624647430764, "maximum": 1.6426970454895613, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07398148274843494, "standard_deviation": 0.013268891407118996, "minimum": 0.05737092408366975, "maximum": 0.09003045720697717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.97030229422765, "standard_deviation": 0.004584381249542305, "minimum": 0.9654994160482117, "maximum": 0.9762173342849292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4483561580011899, "standard_deviation": 0.024726693658631872, "minimum": 1.4136371879974454, "maximum": 1.4871352051225788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5229658957048441, "standard_deviation": 0.1501717518626178, "minimum": 0.3189859686300276, "maximum": 0.682971047791598, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18824543642125552, "standard_deviation": 0.014860118621056784, "minimum": 0.16800155466283356, "maximum": 0.20645748684049697, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9898755231918916, "standard_deviation": 0.0024332163428865245, "minimum": 0.9867392203191706, "maximum": 0.9925430726507231, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8776257937226344, "standard_deviation": 0.027903459109510993, "minimum": 0.846450775952587, "maximum": 0.9162058440024156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5264822308065804, "standard_deviation": 0.17454448690666113, "minimum": 0.18352383790930635, "maximum": 0.7138958413479073, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10509298465974803, "standard_deviation": 0.012692239638732566, "minimum": 0.09067027095574938, "maximum": 0.1211802282172051, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932909258085646, "standard_deviation": 0.0014547342856248648, "minimum": 0.9905663151071359, "maximum": 0.9951420144567047, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.294847101823426, "standard_deviation": 0.025092986763303585, "minimum": 1.2765176710125736, "maximum": 1.3535039893640577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.730197701958376, "standard_deviation": 0.264263374233301, "minimum": 0.32285185725262394, "maximum": 1.0925998840491837, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09193383175645281, "standard_deviation": 0.009634408100510392, "minimum": 0.07902066006401759, "maximum": 0.10913758001063582, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968574817061255, "standard_deviation": 0.001170782202794287, "minimum": 0.9945678184982214, "maximum": 0.9982144844247807, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8527930186714849, "standard_deviation": 0.026561670242984708, "minimum": 0.8232501865374291, "maximum": 0.9022615927635632, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9005301718334994, "standard_deviation": 0.31982979604729883, "minimum": 0.4166654739967905, "maximum": 1.365279863480974, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06150248009522353, "standard_deviation": 0.010843614021051436, "minimum": 0.04809771608457601, "maximum": 0.0810284547324046, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3382046567608035, "standard_deviation": 0.04353458110174993, "minimum": 0.2945751737101076, "maximum": 0.4324997561337793, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.540989750325965, "standard_deviation": 0.010554451809979553, "minimum": 5.526170679405108, "maximum": 5.562012022016425, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8428114990410929, "standard_deviation": 0.05152263030294442, "minimum": 0.7555893248712029, "maximum": 0.895570526910964, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08313477811101577, "standard_deviation": 0.07291349822878593, "minimum": -0.17226303329563378, "maximum": 0.04459335861491698, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944925534394736, "standard_deviation": 0.0006561965893733099, "minimum": 0.9934566943005495, "maximum": 0.9953264261792855, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7924403082759011, "standard_deviation": 0.02952962502193625, "minimum": 0.7523677099940609, "maximum": 0.8344745907218258, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42327767766799207, "standard_deviation": 0.21039775067848163, "minimum": 0.2230588886366062, "maximum": 0.8273313823765683, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0790021128687167, "standard_deviation": 0.0048271005081865, "minimum": 0.0730727331477826, "maximum": 0.08716553047148369, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981131656396145, "standard_deviation": 0.0005567231675521974, "minimum": 0.9974190574952503, "maximum": 0.9987901364215384, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.446218029665223, "standard_deviation": 0.016833650841439868, "minimum": 0.4195286922638613, "maximum": 0.47234390070219734, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.65693663595055, "standard_deviation": 0.38958077246579786, "minimum": 0.17164953510403133, "maximum": 1.2674225915381996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045498484098546996, "standard_deviation": 0.006627284734114229, "minimum": 0.03759829435352359, "maximum": 0.05530942760123261, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0019", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.2443623542785645, "predicted_mos": 2.2443623542785645, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4497d7b474a6a239aa1f794eef49cc83478a5add995a14dee6c24a1ba08ebeba", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.991725299973041, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8152213809522483, "全帯域_10ms_rms_envelope_nrmse": 0.07639396111513191, "全帯域_10ms_envelope_correlation": 0.9948450736931527, "全帯域_10ms_modulation_mae_db": 0.3081918592625977, "全帯域_40ms_log_rms_mae_db": 0.45447236359953763, "全帯域_40ms_rms_envelope_nrmse": 0.03830712983657974, "全帯域_40ms_envelope_correlation": 0.9987139172021928, "全帯域_40ms_modulation_mae_db": 0.5962728231297459, "300–1000 Hz_10ms_log_rms_mae_db": 1.2531639908856436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07328348252552369, "300–1000 Hz_10ms_envelope_correlation": 0.995824352909102, "300–1000 Hz_10ms_modulation_mae_db": 0.6418071571895272, "300–1000 Hz_40ms_log_rms_mae_db": 0.7752449306495748, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045101886094130794, "300–1000 Hz_40ms_envelope_correlation": 0.9984207564254725, "300–1000 Hz_40ms_modulation_mae_db": 0.8461795449471629, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3348788446318554, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11873394418372596, "1000–3000 Hz_10ms_envelope_correlation": 0.9902240440628255, "1000–3000 Hz_10ms_modulation_mae_db": 0.7901118163885431, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7735038607883944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059469133786650946, "1000–3000 Hz_40ms_envelope_correlation": 0.9975557919640151, "1000–3000 Hz_40ms_modulation_mae_db": 0.8460438667035474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.513833207968128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18725003487888325, "3000–6000 Hz_10ms_envelope_correlation": 0.9704821864863109, "3000–6000 Hz_10ms_modulation_mae_db": 0.5761746128934847, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9094742566162329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09576364664677942, "3000–6000 Hz_40ms_envelope_correlation": 0.9914574084058816, "3000–6000 Hz_40ms_modulation_mae_db": 0.54479503269233, "mr_stft_db": 5.512110670486173, "band_mae_db": 0.2660556062672809, "rms_gain_error_db": -0.10165415561431185, "output_peak": 0.7526208169354647}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7975607948000597, "全帯域_10ms_rms_envelope_nrmse": 0.08755588336730241, "全帯域_10ms_envelope_correlation": 0.9931793413162596, "全帯域_10ms_modulation_mae_db": 0.15809312867199407, "全帯域_40ms_log_rms_mae_db": 0.443781065197153, "全帯域_40ms_rms_envelope_nrmse": 0.05178077228601389, "全帯域_40ms_envelope_correlation": 0.99744873925627, "全帯域_40ms_modulation_mae_db": 0.13895706624384005, "300–1000 Hz_10ms_log_rms_mae_db": 1.2561152632334192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09818695362436063, "300–1000 Hz_10ms_envelope_correlation": 0.9923330827849376, "300–1000 Hz_10ms_modulation_mae_db": 1.3697953795772702, "300–1000 Hz_40ms_log_rms_mae_db": 0.8233644694247295, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06450066026903517, "300–1000 Hz_40ms_envelope_correlation": 0.9965357080979805, "300–1000 Hz_40ms_modulation_mae_db": 1.664914670383218, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3407909247200755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1387825062782337, "1000–3000 Hz_10ms_envelope_correlation": 0.987603677913176, "1000–3000 Hz_10ms_modulation_mae_db": 0.30012301156962173, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8422000331586215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08924372834195558, "1000–3000 Hz_40ms_envelope_correlation": 0.9949422553900492, "1000–3000 Hz_40ms_modulation_mae_db": 0.31495575752802174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4780496920443265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1756269514571801, "3000–6000 Hz_10ms_envelope_correlation": 0.9740881893538116, "3000–6000 Hz_10ms_modulation_mae_db": 1.0311628304924767, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9312278104075751, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09695673595829318, "3000–6000 Hz_40ms_envelope_correlation": 0.9918624452942588, "3000–6000 Hz_40ms_modulation_mae_db": 1.2044733345967191, "mr_stft_db": 5.523486211796912, "band_mae_db": 0.3108217328901323, "rms_gain_error_db": 0.035922411603926264, "output_peak": 0.8242333658158097}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8241869514754357, "全帯域_10ms_rms_envelope_nrmse": 0.07580995419800181, "全帯域_10ms_envelope_correlation": 0.9949797161378766, "全帯域_10ms_modulation_mae_db": 0.720512698401869, "全帯域_40ms_log_rms_mae_db": 0.451785783309647, "全帯域_40ms_rms_envelope_nrmse": 0.04434377018977015, "全帯域_40ms_envelope_correlation": 0.9982841329274311, "全帯域_40ms_modulation_mae_db": 1.1366254119750063, "300–1000 Hz_10ms_log_rms_mae_db": 1.231744885682617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08042873033317186, "300–1000 Hz_10ms_envelope_correlation": 0.9949148578417851, "300–1000 Hz_10ms_modulation_mae_db": 0.6872881764082068, "300–1000 Hz_40ms_log_rms_mae_db": 0.793895557655528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04218066743600631, "300–1000 Hz_40ms_envelope_correlation": 0.9986267395524879, "300–1000 Hz_40ms_modulation_mae_db": 0.9101844179621265, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3561582495702795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12520338775389886, "1000–3000 Hz_10ms_envelope_correlation": 0.9893513526762272, "1000–3000 Hz_10ms_modulation_mae_db": 1.1322832846353628, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8646809512386533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0753338626907659, "1000–3000 Hz_40ms_envelope_correlation": 0.9962896945574383, "1000–3000 Hz_40ms_modulation_mae_db": 1.4981723365409765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5256915964461453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17892851193940598, "3000–6000 Hz_10ms_envelope_correlation": 0.9731321175273272, "3000–6000 Hz_10ms_modulation_mae_db": 1.0155853712062113, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9243716304509157, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10700876888093407, "3000–6000 Hz_40ms_envelope_correlation": 0.9894194239743825, "3000–6000 Hz_40ms_modulation_mae_db": 1.2116790035077063, "mr_stft_db": 5.533189768076926, "band_mae_db": 0.3936833599512555, "rms_gain_error_db": -0.12043903456553168, "output_peak": 0.8598513921222959}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7745457050359198, "全帯域_10ms_rms_envelope_nrmse": 0.07528423412802225, "全帯域_10ms_envelope_correlation": 0.9950526185898017, "全帯域_10ms_modulation_mae_db": 0.32654530049722985, "全帯域_40ms_log_rms_mae_db": 0.44054486860371456, "全帯域_40ms_rms_envelope_nrmse": 0.03778501842558806, "全帯域_40ms_envelope_correlation": 0.998816620181503, "全帯域_40ms_modulation_mae_db": 0.4743181200453253, "300–1000 Hz_10ms_log_rms_mae_db": 1.3122717716541972, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09243000556638413, "300–1000 Hz_10ms_envelope_correlation": 0.9932743674082694, "300–1000 Hz_10ms_modulation_mae_db": 0.5472651563747163, "300–1000 Hz_40ms_log_rms_mae_db": 0.8611541394105612, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0600725511647996, "300–1000 Hz_40ms_envelope_correlation": 0.9970920617746449, "300–1000 Hz_40ms_modulation_mae_db": 0.6785137690523008, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3539838954392565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11407262035993153, "1000–3000 Hz_10ms_envelope_correlation": 0.990947459791049, "1000–3000 Hz_10ms_modulation_mae_db": 0.5101828070491395, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8207143244584668, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05546476791855631, "1000–3000 Hz_40ms_envelope_correlation": 0.9978276798470574, "1000–3000 Hz_40ms_modulation_mae_db": 0.6881612361729809, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4638629530895009, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15893317731377118, "3000–6000 Hz_10ms_envelope_correlation": 0.9788724306295288, "3000–6000 Hz_10ms_modulation_mae_db": 0.8519689131312935, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9209869362067501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08315739346539404, "3000–6000 Hz_40ms_envelope_correlation": 0.9941193025961478, "3000–6000 Hz_40ms_modulation_mae_db": 1.000414338087287, "mr_stft_db": 5.496508222117957, "band_mae_db": 0.32840386698404506, "rms_gain_error_db": -0.12714224147999612, "output_peak": 0.8256012994573447}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7643043786143396, "全帯域_10ms_rms_envelope_nrmse": 0.07562843760473402, "全帯域_10ms_envelope_correlation": 0.9950155501667152, "全帯域_10ms_modulation_mae_db": 0.537422163431919, "全帯域_40ms_log_rms_mae_db": 0.4312016606402566, "全帯域_40ms_rms_envelope_nrmse": 0.04090584083911387, "全帯域_40ms_envelope_correlation": 0.9985860335769845, "全帯域_40ms_modulation_mae_db": 0.7534904181268218, "300–1000 Hz_10ms_log_rms_mae_db": 1.2496853990893988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0979591230979947, "300–1000 Hz_10ms_envelope_correlation": 0.9924684523796579, "300–1000 Hz_10ms_modulation_mae_db": 0.954790532794146, "300–1000 Hz_40ms_log_rms_mae_db": 0.8004523671851992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0674353775806301, "300–1000 Hz_40ms_envelope_correlation": 0.9963317120036559, "300–1000 Hz_40ms_modulation_mae_db": 1.1545262794059334, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3909275042040914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11372725108290917, "1000–3000 Hz_10ms_envelope_correlation": 0.9910366397893544, "1000–3000 Hz_10ms_modulation_mae_db": 0.7483684192114964, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8553109278567151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06495638364170603, "1000–3000 Hz_40ms_envelope_correlation": 0.9970263924354997, "1000–3000 Hz_40ms_modulation_mae_db": 0.8570154080861836, "3000–6000 Hz_10ms_log_rms_mae_db": 1.516975707817901, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19424248446088324, "3000–6000 Hz_10ms_envelope_correlation": 0.9680058963084857, "3000–6000 Hz_10ms_modulation_mae_db": 1.0203333901873324, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9475695583866648, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10316131022490777, "3000–6000 Hz_40ms_envelope_correlation": 0.990447675656636, "3000–6000 Hz_40ms_modulation_mae_db": 1.1372334891191167, "mr_stft_db": 5.516145346689103, "band_mae_db": 0.3398632955137204, "rms_gain_error_db": -0.1267122271465069, "output_peak": 0.8538160556850327}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8296899135498094, "全帯域_10ms_rms_envelope_nrmse": 0.09016188485605678, "全帯域_10ms_envelope_correlation": 0.9929865352712254, "全帯域_10ms_modulation_mae_db": 0.554347074470989, "全帯域_40ms_log_rms_mae_db": 0.48793975433655756, "全帯域_40ms_rms_envelope_nrmse": 0.0570710963496719, "全帯域_40ms_envelope_correlation": 0.9972485858062422, "全帯域_40ms_modulation_mae_db": 1.1037551801097767, "300–1000 Hz_10ms_log_rms_mae_db": 1.2416812113214688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11809982420687684, "300–1000 Hz_10ms_envelope_correlation": 0.9889254178240542, "300–1000 Hz_10ms_modulation_mae_db": 0.644377226285561, "300–1000 Hz_40ms_log_rms_mae_db": 0.8319440337572038, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08563557326327655, "300–1000 Hz_40ms_envelope_correlation": 0.9939393028329069, "300–1000 Hz_40ms_modulation_mae_db": 0.7899648959155985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3858860341973933, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1267660050493347, "1000–3000 Hz_10ms_envelope_correlation": 0.9893989937709922, "1000–3000 Hz_10ms_modulation_mae_db": 1.0180518001700938, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8931698961359429, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0810329430804632, "1000–3000 Hz_40ms_envelope_correlation": 0.9960224544971973, "1000–3000 Hz_40ms_modulation_mae_db": 1.4254728869483426, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4551745080831457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1812486859887552, "3000–6000 Hz_10ms_envelope_correlation": 0.973182721201212, "3000–6000 Hz_10ms_modulation_mae_db": 0.93837863470086, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8919449237136143, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10184170923604362, "3000–6000 Hz_40ms_envelope_correlation": 0.9923686335713036, "3000–6000 Hz_40ms_modulation_mae_db": 1.1773466356280355, "mr_stft_db": 5.51941359273739, "band_mae_db": 0.35053440732061597, "rms_gain_error_db": -0.17742340298329862, "output_peak": 0.7956841879385098}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8021824010838792, "全帯域_10ms_rms_envelope_nrmse": 0.08043904964142183, "全帯域_10ms_envelope_correlation": 0.9942044776767908, "全帯域_10ms_modulation_mae_db": 0.31478095957917, "全帯域_40ms_log_rms_mae_db": 0.4501970592607327, "全帯域_40ms_rms_envelope_nrmse": 0.04974448535968199, "全帯域_40ms_envelope_correlation": 0.9976197235930797, "全帯域_40ms_modulation_mae_db": 0.3315005891590257, "300–1000 Hz_10ms_log_rms_mae_db": 1.226414897348661, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09233884894418504, "300–1000 Hz_10ms_envelope_correlation": 0.9933393165675599, "300–1000 Hz_10ms_modulation_mae_db": 0.17462986161557945, "300–1000 Hz_40ms_log_rms_mae_db": 0.7731725775120043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06917979471608382, "300–1000 Hz_40ms_envelope_correlation": 0.9961333435904163, "300–1000 Hz_40ms_modulation_mae_db": 0.1744451230355473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.360017316670995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12823252404152663, "1000–3000 Hz_10ms_envelope_correlation": 0.9895400953568317, "1000–3000 Hz_10ms_modulation_mae_db": 0.2042251357306748, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8426222835602034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09214457409517285, "1000–3000 Hz_40ms_envelope_correlation": 0.9946091072840735, "1000–3000 Hz_40ms_modulation_mae_db": 0.3699352775811408, "3000–6000 Hz_10ms_log_rms_mae_db": 1.479814621649903, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1859889352757573, "3000–6000 Hz_10ms_envelope_correlation": 0.9710103408526675, "3000–6000 Hz_10ms_modulation_mae_db": 0.7136130538269515, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9068641290919759, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0934628809313835, "3000–6000 Hz_40ms_envelope_correlation": 0.9928964516432282, "3000–6000 Hz_40ms_modulation_mae_db": 0.9993863742361765, "mr_stft_db": 5.517796451786961, "band_mae_db": 0.308386705120908, "rms_gain_error_db": -0.004315464780845278, "output_peak": 0.8832986947639528}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7898837087132405, "全帯域_10ms_rms_envelope_nrmse": 0.08213487541949098, "全帯域_10ms_envelope_correlation": 0.9939593528039232, "全帯域_10ms_modulation_mae_db": 0.26915752881562904, "全帯域_40ms_log_rms_mae_db": 0.4612849274251323, "全帯域_40ms_rms_envelope_nrmse": 0.052762761225201925, "全帯域_40ms_envelope_correlation": 0.9973422936812296, "全帯域_40ms_modulation_mae_db": 0.7105368385338414, "300–1000 Hz_10ms_log_rms_mae_db": 1.2438860131142224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08597528338330032, "300–1000 Hz_10ms_envelope_correlation": 0.9940918663083561, "300–1000 Hz_10ms_modulation_mae_db": 1.029062785776832, "300–1000 Hz_40ms_log_rms_mae_db": 0.7855858063072116, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04833581757082284, "300–1000 Hz_40ms_envelope_correlation": 0.9980457511086663, "300–1000 Hz_40ms_modulation_mae_db": 1.2716490243463487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3918079496028855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12442139768076889, "1000–3000 Hz_10ms_envelope_correlation": 0.9892844133247355, "1000–3000 Hz_10ms_modulation_mae_db": 0.6514988239939222, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8580712858088724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0814888693954159, "1000–3000 Hz_40ms_envelope_correlation": 0.9951437674520691, "1000–3000 Hz_40ms_modulation_mae_db": 0.9107229124360842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5004740146424973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17342963876689865, "3000–6000 Hz_10ms_envelope_correlation": 0.974841822072133, "3000–6000 Hz_10ms_modulation_mae_db": 0.43697752155218694, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8955394826564291, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08414191410752349, "3000–6000 Hz_40ms_envelope_correlation": 0.9934565392499569, "3000–6000 Hz_40ms_modulation_mae_db": 0.44536980958735767, "mr_stft_db": 5.515600960944258, "band_mae_db": 0.26319223382667684, "rms_gain_error_db": -0.042176702805545516, "output_peak": 0.9536484665381167}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.432140268456261, "spectral_convergence": 0.2593929293211395, "spectral_convergence_status": "ok", "amplitude_floor": 2.005860984254184e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.430630039821537, "spectral_convergence": 0.2613987845245924, "spectral_convergence_status": "ok", "amplitude_floor": 1.6673734947378395e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.673561703180719, "spectral_convergence": 0.27404758548627045, "spectral_convergence_status": "ok", "amplitude_floor": 1.455049241978443e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.2140486659240935e-05, "output_power": 4.185190741225308e-05, "delta_db": 1.1466280746953408, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002187325069648938, "output_power": 0.002144681395140266, "delta_db": -0.08550546226420443, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018573993391017246, "output_power": 0.0018046129296452398, "delta_db": -0.125212220167961, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.002022873257906857, "output_power": 0.0019664251049153233, "delta_db": -0.12291263004172977, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00023141111513331255, "output_power": 0.00022632423063241767, "delta_db": -0.0965316243672338, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.19984427950361e-05, "output_power": 4.18098707953564e-05, "delta_db": -0.019543626067215776, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2660556062672809, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10165415561431185, "pitch": null, "output_rms_dbfs": -22.058114312671524, "output_peak": 0.7526208169354647, "reference_rms_dbfs": -21.95646015705721, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.512110670486173}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003217977906320387, "log_rms_mae_db": 0.8152213809522483, "rms_envelope_nrmse": 0.07639396111513191, "envelope_correlation": 0.9948450736931527, "rms_gain_error_db": -0.10158890436110811, "reference_mean_power": 0.006378025340653872, "output_mean_power": 0.006230563835752507}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.934428693189062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0026409373563640536, "output_power": 0.0025502614863212175, "reference_active": true, "delta_db": -0.15173387470566238}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022353254716419522, "output_power": 0.0001986234439929609, "reference_active": true, "delta_db": -0.5131025896657172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.171045723198585e-05, "output_power": 2.986954401790517e-05, "reference_active": true, "delta_db": -0.25973911341641354}], "active_band_count": 3, "active_band_mae_db": 0.3081918592625977}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00029448546471938644, "log_rms_mae_db": 0.45447236359953763, "rms_envelope_nrmse": 0.03830712983657974, "envelope_correlation": 0.9987139172021928, "rms_gain_error_db": -0.101522766114608, "reference_mean_power": 0.006407121641291136, "output_mean_power": 0.0062590827409889195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7183045790225306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0025278777543949896, "output_power": 0.002437630091212036, "reference_active": true, "delta_db": -0.1578826570149261}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010431651037251879, "output_power": 9.217367047104857e-05, "reference_active": true, "delta_db": -0.5374616849569649}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4474914310652027e-06, "output_power": 1.9027151107680613e-06, "reference_active": true, "delta_db": -1.0934741274173467}], "active_band_count": 3, "active_band_mae_db": 0.5962728231297459}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002169046697405618, "log_rms_mae_db": 1.2531639908856436, "rms_envelope_nrmse": 0.07328348252552369, "envelope_correlation": 0.995824352909102, "rms_gain_error_db": -0.1234915652133531, "reference_mean_power": 0.00266513165861851, "output_mean_power": 0.002590415986956647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9887000298118345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014141653900070267, "output_power": 0.0013523165310018175, "reference_active": true, "delta_db": -0.19421847230257228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010731466350302803, "output_power": 8.828578953346329e-05, "reference_active": true, "delta_db": -0.8476826289688865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0173327828420217e-05, "output_power": 1.6459836202781114e-05, "reference_active": true, "delta_db": -0.8835203702971227}], "active_band_count": 3, "active_band_mae_db": 0.6418071571895272}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020963310569053025, "log_rms_mae_db": 0.7752449306495748, "rms_envelope_nrmse": 0.045101886094130794, "envelope_correlation": 0.9984207564254725, "rms_gain_error_db": -0.12348362660737992, "reference_mean_power": 0.0026772178601939312, "output_mean_power": 0.0026021681143217844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.883939001913369e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013521293519634116, "output_power": 0.0012902760764160926, "reference_active": true, "delta_db": -0.20335595637312176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.792726117783695e-05, "output_power": 3.9335991045968484e-05, "reference_active": true, "delta_db": -0.8579251534825639}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6850826685161704e-06, "output_power": 1.19921060178309e-06, "reference_active": true, "delta_db": -1.4772575249858033}], "active_band_count": 3, "active_band_mae_db": 0.8461795449471629}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002744641361828464, "log_rms_mae_db": 1.3348788446318554, "rms_envelope_nrmse": 0.11873394418372596, "envelope_correlation": 0.9902240440628255, "rms_gain_error_db": -0.1422064311101099, "reference_mean_power": 0.0019381811578355007, "output_mean_power": 0.0018757446832153597}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1196235236571854e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006735796402683569, "output_power": 0.0006616557935240389, "reference_active": true, "delta_db": -0.0775683205534765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013823323993445422, "output_power": 0.00013252057896990773, "reference_active": true, "delta_db": -0.1832916260270287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8792319908524634e-05, "output_power": 1.1561996850252141e-05, "reference_active": true, "delta_db": -2.1094755025851244}], "active_band_count": 3, "active_band_mae_db": 0.7901118163885431}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002445180716567873, "log_rms_mae_db": 0.7735038607883944, "rms_envelope_nrmse": 0.059469133786650946, "envelope_correlation": 0.9975557919640151, "rms_gain_error_db": -0.141944169082731, "reference_mean_power": 0.0019467823981069799, "output_mean_power": 0.0018841826222506374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0571732642454036e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006649824803295284, "output_power": 0.0006495615085644665, "reference_active": true, "delta_db": -0.10189921743328906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927173921912536e-05, "output_power": 6.6127185324283e-05, "reference_active": true, "delta_db": -0.20176054154178885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7481239585012505e-06, "output_power": 1.0450210554628414e-06, "reference_active": true, "delta_db": -2.234471841135564}], "active_band_count": 3, "active_band_mae_db": 0.8460438667035474}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.6882736857616036e-05, "log_rms_mae_db": 1.513833207968128, "rms_envelope_nrmse": 0.18725003487888325, "envelope_correlation": 0.9704821864863109, "rms_gain_error_db": -0.050839981305082785, "reference_mean_power": 6.564824297759055e-05, "output_mean_power": 6.4884223098533e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.001422573299629e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.567613726314846e-05, "output_power": 1.6165200651905585e-05, "reference_active": true, "delta_db": 0.1334204211234558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.79643479102731e-06, "output_power": 5.852419326739837e-06, "reference_active": true, "delta_db": 0.0417448135009926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.958485457614687e-07, "output_power": 6.963978683660856e-07, "reference_active": true, "delta_db": -1.553358604056006}], "active_band_count": 3, "active_band_mae_db": 0.5761746128934847}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.766393520730197e-05, "log_rms_mae_db": 0.9094742566162329, "rms_envelope_nrmse": 0.09576364664677942, "envelope_correlation": 0.9914574084058816, "rms_gain_error_db": -0.050537862498749725, "reference_mean_power": 6.585532886735065e-05, "output_mean_power": 6.509342699943324e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6817222442955538e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5480772645031266e-05, "output_power": 1.6232585998487978e-05, "reference_active": true, "delta_db": 0.20595079880062236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.979992150286795e-06, "output_power": 2.8211771347105705e-06, "reference_active": true, "delta_db": -0.23784764801655145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.719812912358681e-08, "output_power": 6.629007370894453e-08, "reference_active": true, "delta_db": -1.1905866512598162}], "active_band_count": 3, "active_band_mae_db": 0.54479503269233}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8719497018968906, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e846ae6bfc72e7d1738ce9ca7d312a0a0eb365516f835eccca30e29524275eab", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.989673334585649, "standard_deviation": 0.0010965337242887314, "minimum": 0.987603677913176, "maximum": 0.9910366397893544, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.364306339879604, "standard_deviation": 0.022491269150147014, "minimum": 1.3348788446318554, "maximum": 1.3918079496028855, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6693556373436068, "standard_deviation": 0.3240480709988738, "minimum": 0.2042251357306748, "maximum": 1.1322832846353628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12374245455379118, "standard_deviation": 0.008254835786975358, "minimum": 0.11372725108290917, "maximum": 0.1387825062782337, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996177142928425, "standard_deviation": 0.001220907992195306, "minimum": 0.9946091072840735, "maximum": 0.9978276798470574, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8437841953757337, "standard_deviation": 0.03522988249189498, "minimum": 0.7735038607883944, "maximum": 0.8931698961359429, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8638099602496597, "standard_deviation": 0.43035684613044983, "minimum": 0.31495575752802174, "maximum": 1.4981723365409765, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07489178286883584, "standard_deviation": 0.013622707672480664, "minimum": 0.05546476791855631, "maximum": 0.09214457409517285, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9729519630539346, "standard_deviation": 0.003258691409151633, "minimum": 0.9680058963084857, "maximum": 0.9788724306295288, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4917345377176936, "standard_deviation": 0.026180004872869516, "minimum": 1.4551745080831457, "maximum": 1.5256915964461453, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8230242909988497, "standard_deviation": 0.2253297292596089, "minimum": 0.43697752155218694, "maximum": 1.0311628304924767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17945605251019187, "standard_deviation": 0.010675226366781579, "minimum": 0.15893317731377118, "maximum": 0.19424248446088324, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920034850489744, "standard_deviation": 0.0015553952001608024, "minimum": 0.9894194239743825, "maximum": 0.9941193025961478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9159973409412697, "standard_deviation": 0.018696087551675834, "minimum": 0.8919449237136143, "maximum": 0.9475695583866648, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9650872521818412, "standard_deviation": 0.3028384207794083, "minimum": 0.44536980958735767, "maximum": 1.2116790035077063, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09568679493140739, "standard_deviation": 0.008608741962974883, "minimum": 0.08315739346539404, "maximum": 0.10700876888093407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931464642529653, "standard_deviation": 0.0020765411586943956, "minimum": 0.9889254178240542, "maximum": 0.995824352909102, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2518704290412037, "standard_deviation": 0.026439637747601847, "minimum": 1.226414897348661, "maximum": 1.3122717716541972, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7561270345027298, "standard_deviation": 0.35908774021392725, "minimum": 0.17462986161557945, "maximum": 1.3697953795772702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09233778146022464, "standard_deviation": 0.013495379305199316, "minimum": 0.07328348252552369, "maximum": 0.11809982420687684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968906719232788, "standard_deviation": 0.0015330835600264753, "minimum": 0.9939393028329069, "maximum": 0.9986267395524879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8056017352377516, "standard_deviation": 0.03078235783680121, "minimum": 0.7731725775120043, "maximum": 0.8611541394105612, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9362972156310295, "standard_deviation": 0.4420819723828328, "minimum": 0.1744451230355473, "maximum": 1.664914670383218, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060305291011848144, "standard_deviation": 0.014598371904163267, "minimum": 0.04218066743600631, "maximum": 0.08563557326327655, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.32011765098432937, "standard_deviation": 0.04339664476320716, "minimum": 0.26319223382667684, "maximum": 0.3936833599512555, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5167814030794595, "standard_deviation": 0.010397549633884843, "minimum": 5.496508222117957, "maximum": 5.533189768076926, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8435942849070659, "standard_deviation": 0.06015133672564941, "minimum": 0.7526208169354647, "maximum": 0.9536484665381167, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08299260222151371, "standard_deviation": 0.0722348069712066, "minimum": -0.17742340298329862, "maximum": 0.035922411603926264, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9942778332069682, "standard_deviation": 0.0008400566064052337, "minimum": 0.9929865352712254, "maximum": 0.9950526185898017, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7996969042781166, "standard_deviation": 0.023091759675429122, "minimum": 0.7643043786143396, "maximum": 0.8296899135498094, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3986313391414247, "standard_deviation": 0.18599023533233003, "minimum": 0.15809312867199407, "maximum": 0.720512698401869, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08042603504127024, "standard_deviation": 0.0058010183081239975, "minimum": 0.07528423412802225, "maximum": 0.09016188485605678, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980075057781166, "standard_deviation": 0.0006596494651551855, "minimum": 0.9972485858062422, "maximum": 0.998816620181503, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4526509352965914, "standard_deviation": 0.016967350982612005, "minimum": 0.4312016606402566, "maximum": 0.48793975433655756, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6556820559154228, "standard_deviation": 0.34911566478167094, "minimum": 0.13895706624384005, "maximum": 1.1366254119750063, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.046587609313952685, "standard_deviation": 0.007253965056806459, "minimum": 0.03778501842558806, "maximum": 0.0570710963496719, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0019", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.291226625442505, "predicted_mos": 2.291226625442505, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7e732232abc89c5644051370af4b406884c0b4c0062030492d0b5a9262b5a344", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.009242999833077, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7762941912686905, "全帯域_10ms_rms_envelope_nrmse": 0.06802287443343843, "全帯域_10ms_envelope_correlation": 0.9958885245571858, "全帯域_10ms_modulation_mae_db": 0.2632948430064597, "全帯域_40ms_log_rms_mae_db": 0.40519255120625547, "全帯域_40ms_rms_envelope_nrmse": 0.033558028389472104, "全帯域_40ms_envelope_correlation": 0.9989699074639321, "全帯域_40ms_modulation_mae_db": 0.6422349514432287, "300–1000 Hz_10ms_log_rms_mae_db": 1.4631853456710555, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.091376931911443, "300–1000 Hz_10ms_envelope_correlation": 0.993328293110384, "300–1000 Hz_10ms_modulation_mae_db": 0.6946362659285511, "300–1000 Hz_40ms_log_rms_mae_db": 1.0018500778625974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0521014309307506, "300–1000 Hz_40ms_envelope_correlation": 0.9977258970726887, "300–1000 Hz_40ms_modulation_mae_db": 0.7953206577971228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4380292448315224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09247516116047323, "1000–3000 Hz_10ms_envelope_correlation": 0.9941946814723978, "1000–3000 Hz_10ms_modulation_mae_db": 0.4763795301249139, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8154456779646829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.050168727127433546, "1000–3000 Hz_40ms_envelope_correlation": 0.9983569936890991, "1000–3000 Hz_40ms_modulation_mae_db": 0.7731089276719262, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6666073400894434, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22776411468999475, "3000–6000 Hz_10ms_envelope_correlation": 0.9605831418299401, "3000–6000 Hz_10ms_modulation_mae_db": 0.18792728363445302, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9935341825882298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1378846302459916, "3000–6000 Hz_40ms_envelope_correlation": 0.9853408693577281, "3000–6000 Hz_40ms_modulation_mae_db": 0.23918381717066808, "mr_stft_db": 5.774791115075151, "band_mae_db": 0.465217369754997, "rms_gain_error_db": -0.06195924295640953, "output_peak": 0.9214698202026926}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7738210426202623, "全帯域_10ms_rms_envelope_nrmse": 0.08666655235862818, "全帯域_10ms_envelope_correlation": 0.993269351167803, "全帯域_10ms_modulation_mae_db": 0.09947581000011722, "全帯域_40ms_log_rms_mae_db": 0.436582933011688, "全帯域_40ms_rms_envelope_nrmse": 0.044543802539367953, "全帯域_40ms_envelope_correlation": 0.9981260530695614, "全帯域_40ms_modulation_mae_db": 0.4126455261081543, "300–1000 Hz_10ms_log_rms_mae_db": 1.4704257730018333, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13392132781814747, "300–1000 Hz_10ms_envelope_correlation": 0.9856291930369945, "300–1000 Hz_10ms_modulation_mae_db": 0.12790808647853277, "300–1000 Hz_40ms_log_rms_mae_db": 0.9845878069732713, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0876019075977768, "300–1000 Hz_40ms_envelope_correlation": 0.9936021672595958, "300–1000 Hz_40ms_modulation_mae_db": 0.4388471553693413, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4131766042598073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12703263548735821, "1000–3000 Hz_10ms_envelope_correlation": 0.9889745043969136, "1000–3000 Hz_10ms_modulation_mae_db": 0.582719156811393, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8079439989260571, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08986484169996538, "1000–3000 Hz_40ms_envelope_correlation": 0.9941879404750341, "1000–3000 Hz_40ms_modulation_mae_db": 0.7317086406144119, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7238636058620391, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23604979804427814, "3000–6000 Hz_10ms_envelope_correlation": 0.9539438621639806, "3000–6000 Hz_10ms_modulation_mae_db": 0.12715157079285658, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0730306170339694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14394415973098057, "3000–6000 Hz_40ms_envelope_correlation": 0.9810557308638992, "3000–6000 Hz_40ms_modulation_mae_db": 0.08190457766320387, "mr_stft_db": 5.768582017291404, "band_mae_db": 0.35484112888841257, "rms_gain_error_db": -0.058322157545270065, "output_peak": 0.8486555442057502}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7580271582323781, "全帯域_10ms_rms_envelope_nrmse": 0.07695274975158084, "全帯域_10ms_envelope_correlation": 0.9948699842708042, "全帯域_10ms_modulation_mae_db": 0.25771627223897003, "全帯域_40ms_log_rms_mae_db": 0.42232408324273324, "全帯域_40ms_rms_envelope_nrmse": 0.0459626728398129, "全帯域_40ms_envelope_correlation": 0.9982064120473887, "全帯域_40ms_modulation_mae_db": 0.5936821609582834, "300–1000 Hz_10ms_log_rms_mae_db": 1.423677439653208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08075090486747197, "300–1000 Hz_10ms_envelope_correlation": 0.9949509219890502, "300–1000 Hz_10ms_modulation_mae_db": 0.3792593683395036, "300–1000 Hz_40ms_log_rms_mae_db": 0.976041888224641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05336850393363222, "300–1000 Hz_40ms_envelope_correlation": 0.9977812025496334, "300–1000 Hz_40ms_modulation_mae_db": 0.4217807720165019, "1000–3000 Hz_10ms_log_rms_mae_db": 1.46711777821605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1206923875437481, "1000–3000 Hz_10ms_envelope_correlation": 0.9901378094277616, "1000–3000 Hz_10ms_modulation_mae_db": 0.6998703181504699, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8642738899489213, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07788104838439182, "1000–3000 Hz_40ms_envelope_correlation": 0.9959653295487662, "1000–3000 Hz_40ms_modulation_mae_db": 0.9569574012805592, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7141991301904838, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2042646976190972, "3000–6000 Hz_10ms_envelope_correlation": 0.966961634786814, "3000–6000 Hz_10ms_modulation_mae_db": 0.4068550183061557, "3000–6000 Hz_40ms_log_rms_mae_db": 1.051864506083176, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1293060936876937, "3000–6000 Hz_40ms_envelope_correlation": 0.9858867549975195, "3000–6000 Hz_40ms_modulation_mae_db": 0.5635028635216363, "mr_stft_db": 5.782032820912423, "band_mae_db": 0.4296415295121753, "rms_gain_error_db": -0.1432037551165286, "output_peak": 0.8478505270438985}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7930213102483664, "全帯域_10ms_rms_envelope_nrmse": 0.0729469384170754, "全帯域_10ms_envelope_correlation": 0.9953535255310572, "全帯域_10ms_modulation_mae_db": 0.8309082667553763, "全帯域_40ms_log_rms_mae_db": 0.45487663928188243, "全帯域_40ms_rms_envelope_nrmse": 0.04183795259021201, "全帯域_40ms_envelope_correlation": 0.9984766837514812, "全帯域_40ms_modulation_mae_db": 1.197602410872592, "300–1000 Hz_10ms_log_rms_mae_db": 1.5165049438999825, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09422125277136509, "300–1000 Hz_10ms_envelope_correlation": 0.9931096577705764, "300–1000 Hz_10ms_modulation_mae_db": 1.172565021036768, "300–1000 Hz_40ms_log_rms_mae_db": 0.9980892162974709, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06504474042553468, "300–1000 Hz_40ms_envelope_correlation": 0.9966570105550939, "300–1000 Hz_40ms_modulation_mae_db": 1.206218333588289, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4495440858127473, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10721897006508037, "1000–3000 Hz_10ms_envelope_correlation": 0.9920396775477114, "1000–3000 Hz_10ms_modulation_mae_db": 1.1160915187976368, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8775500534308871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05681735960780967, "1000–3000 Hz_40ms_envelope_correlation": 0.9977309718779129, "1000–3000 Hz_40ms_modulation_mae_db": 1.3797403841954916, "3000–6000 Hz_10ms_log_rms_mae_db": 1.671272909588339, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.210424529638004, "3000–6000 Hz_10ms_envelope_correlation": 0.9627358269827475, "3000–6000 Hz_10ms_modulation_mae_db": 0.4081054863651354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0056534230839644, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1167944911879202, "3000–6000 Hz_40ms_envelope_correlation": 0.9872222946356993, "3000–6000 Hz_40ms_modulation_mae_db": 0.3914194539337408, "mr_stft_db": 5.797022846632809, "band_mae_db": 0.4098773335893346, "rms_gain_error_db": -0.11217470207747486, "output_peak": 0.7853014498825822}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7760253711395717, "全帯域_10ms_rms_envelope_nrmse": 0.08616948491176768, "全帯域_10ms_envelope_correlation": 0.993365950484085, "全帯域_10ms_modulation_mae_db": 0.781302174172552, "全帯域_40ms_log_rms_mae_db": 0.43982302592818173, "全帯域_40ms_rms_envelope_nrmse": 0.052101946633396984, "全帯域_40ms_envelope_correlation": 0.9974363898112654, "全帯域_40ms_modulation_mae_db": 1.0523628164374463, "300–1000 Hz_10ms_log_rms_mae_db": 1.4948390711857316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09028108208726468, "300–1000 Hz_10ms_envelope_correlation": 0.9935253754570372, "300–1000 Hz_10ms_modulation_mae_db": 0.6732554525766684, "300–1000 Hz_40ms_log_rms_mae_db": 0.944801358268344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060063978263658904, "300–1000 Hz_40ms_envelope_correlation": 0.9969462949373786, "300–1000 Hz_40ms_modulation_mae_db": 0.9310352414660233, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4821865634741165, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1488210525575095, "1000–3000 Hz_10ms_envelope_correlation": 0.9848677674457571, "1000–3000 Hz_10ms_modulation_mae_db": 1.3720248866802451, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8687372981136866, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0968037960910547, "1000–3000 Hz_40ms_envelope_correlation": 0.9936852529313412, "1000–3000 Hz_40ms_modulation_mae_db": 1.4366603506598488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7197378875500917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20014545416740326, "3000–6000 Hz_10ms_envelope_correlation": 0.9665287595260085, "3000–6000 Hz_10ms_modulation_mae_db": 0.14768340267906413, "3000–6000 Hz_40ms_log_rms_mae_db": 1.077197272445438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11840270968759854, "3000–6000 Hz_40ms_envelope_correlation": 0.9869955087445097, "3000–6000 Hz_40ms_modulation_mae_db": 0.2664476540121646, "mr_stft_db": 5.792957170957089, "band_mae_db": 0.3942625731395266, "rms_gain_error_db": -0.06796911467676245, "output_peak": 0.7980212263342888}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7821071343091269, "全帯域_10ms_rms_envelope_nrmse": 0.08033829986238211, "全帯域_10ms_envelope_correlation": 0.9943752469686377, "全帯域_10ms_modulation_mae_db": 0.28648254471007367, "全帯域_40ms_log_rms_mae_db": 0.4178799879494373, "全帯域_40ms_rms_envelope_nrmse": 0.04603866164900781, "全帯域_40ms_envelope_correlation": 0.9981837354175666, "全帯域_40ms_modulation_mae_db": 0.8257393474767212, "300–1000 Hz_10ms_log_rms_mae_db": 1.4665712112207212, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10890709200896695, "300–1000 Hz_10ms_envelope_correlation": 0.9905390986269559, "300–1000 Hz_10ms_modulation_mae_db": 0.13672336687850442, "300–1000 Hz_40ms_log_rms_mae_db": 0.9814387476127339, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07866618963822418, "300–1000 Hz_40ms_envelope_correlation": 0.9948045841763878, "300–1000 Hz_40ms_modulation_mae_db": 0.26642284058023247, "1000–3000 Hz_10ms_log_rms_mae_db": 1.418726782491548, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13195205804124477, "1000–3000 Hz_10ms_envelope_correlation": 0.9881863572711239, "1000–3000 Hz_10ms_modulation_mae_db": 1.224641401067, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8287398108429198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08625621348795146, "1000–3000 Hz_40ms_envelope_correlation": 0.9950408368629705, "1000–3000 Hz_40ms_modulation_mae_db": 1.4588519304421428, "3000–6000 Hz_10ms_log_rms_mae_db": 1.690613544148744, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22517680735497878, "3000–6000 Hz_10ms_envelope_correlation": 0.9572338545615534, "3000–6000 Hz_10ms_modulation_mae_db": 0.359805246613435, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0167234080824032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13219400381077037, "3000–6000 Hz_40ms_envelope_correlation": 0.9835815202288299, "3000–6000 Hz_40ms_modulation_mae_db": 0.23429991149689042, "mr_stft_db": 5.780132322091776, "band_mae_db": 0.41569100831416717, "rms_gain_error_db": -0.12986223859980492, "output_peak": 0.8379324361703995}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7909355997602947, "全帯域_10ms_rms_envelope_nrmse": 0.07941047585472896, "全帯域_10ms_envelope_correlation": 0.9943824104351119, "全帯域_10ms_modulation_mae_db": 0.31309865973328105, "全帯域_40ms_log_rms_mae_db": 0.4329623174443144, "全帯域_40ms_rms_envelope_nrmse": 0.046896003557479214, "全帯域_40ms_envelope_correlation": 0.9979536191476321, "全帯域_40ms_modulation_mae_db": 0.5554750700190362, "300–1000 Hz_10ms_log_rms_mae_db": 1.433693960394567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08908552497379896, "300–1000 Hz_10ms_envelope_correlation": 0.9936473242280749, "300–1000 Hz_10ms_modulation_mae_db": 0.3688032962371315, "300–1000 Hz_40ms_log_rms_mae_db": 0.932864297400303, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04884622613016216, "300–1000 Hz_40ms_envelope_correlation": 0.9979955950832313, "300–1000 Hz_40ms_modulation_mae_db": 0.1388192199298092, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3906056423547963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12152457397317948, "1000–3000 Hz_10ms_envelope_correlation": 0.9896963782707209, "1000–3000 Hz_10ms_modulation_mae_db": 1.2916286022965566, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8351626341231058, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07938100500531112, "1000–3000 Hz_40ms_envelope_correlation": 0.9953911343159458, "1000–3000 Hz_40ms_modulation_mae_db": 1.772548470312936, "3000–6000 Hz_10ms_log_rms_mae_db": 1.705018982640621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2339378976255002, "3000–6000 Hz_10ms_envelope_correlation": 0.9545157437089059, "3000–6000 Hz_10ms_modulation_mae_db": 0.3086821012354349, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0396520061054777, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1504198086434923, "3000–6000 Hz_40ms_envelope_correlation": 0.9791099796772637, "3000–6000 Hz_40ms_modulation_mae_db": 0.37694787073703656, "mr_stft_db": 5.779828293836881, "band_mae_db": 0.3559416942284002, "rms_gain_error_db": -0.08328477253941599, "output_peak": 0.941025076881832}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7850105629358628, "全帯域_10ms_rms_envelope_nrmse": 0.06951756906076702, "全帯域_10ms_envelope_correlation": 0.9956730512475086, "全帯域_10ms_modulation_mae_db": 0.3599703925470569, "全帯域_40ms_log_rms_mae_db": 0.42116318724399776, "全帯域_40ms_rms_envelope_nrmse": 0.03272446734688274, "全帯域_40ms_envelope_correlation": 0.9989725924173543, "全帯域_40ms_modulation_mae_db": 0.6193215336935616, "300–1000 Hz_10ms_log_rms_mae_db": 1.441772774223036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10596210633586055, "300–1000 Hz_10ms_envelope_correlation": 0.9910365567486559, "300–1000 Hz_10ms_modulation_mae_db": 0.3823788643220228, "300–1000 Hz_40ms_log_rms_mae_db": 0.9351651172808967, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058255751154863505, "300–1000 Hz_40ms_envelope_correlation": 0.9971281830821485, "300–1000 Hz_40ms_modulation_mae_db": 0.573901438234898, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4071573783671651, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10134144301565269, "1000–3000 Hz_10ms_envelope_correlation": 0.9928429937243084, "1000–3000 Hz_10ms_modulation_mae_db": 0.7860835428708143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8248286950029761, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05296492645955684, "1000–3000 Hz_40ms_envelope_correlation": 0.9979474033018676, "1000–3000 Hz_40ms_modulation_mae_db": 1.0407345168709392, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7503703668886106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22978512074996318, "3000–6000 Hz_10ms_envelope_correlation": 0.9571418095776278, "3000–6000 Hz_10ms_modulation_mae_db": 0.20618992003926895, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1110316361311139, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14248554873956062, "3000–6000 Hz_40ms_envelope_correlation": 0.9819792891964834, "3000–6000 Hz_40ms_modulation_mae_db": 0.17846832580209196, "mr_stft_db": 5.78911745168077, "band_mae_db": 0.3607218210237846, "rms_gain_error_db": -0.01170530858173615, "output_peak": 0.8471509770534306}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.715158001544399, "spectral_convergence": 0.2584114398171235, "spectral_convergence_status": "ok", "amplitude_floor": 2.005860984254184e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7432330533256275, "spectral_convergence": 0.2653631984899701, "spectral_convergence_status": "ok", "amplitude_floor": 1.6673734947378395e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.865982290355424, "spectral_convergence": 0.2603278433869525, "spectral_convergence_status": "ok", "amplitude_floor": 1.455049241978443e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.2140486659240935e-05, "output_power": 5.178717834805848e-05, "delta_db": 2.0716980049490057, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002187325069648938, "output_power": 0.0021903103148733557, "delta_db": 0.005923178223765346, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018573993391017246, "output_power": 0.001831351237661556, "delta_db": -0.06133640386972439, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.002022873257906857, "output_power": 0.0019227282706723197, "delta_db": -0.22050761160931143, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00023141111513331255, "output_power": 0.00024240804079541604, "delta_db": 0.2016280635819266, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.19984427950361e-05, "output_power": 4.428475880891327e-05, "delta_db": 0.23021095629624894, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.465217369754997, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06195924295640953, "pitch": null, "output_rms_dbfs": -22.018419400013624, "output_peak": 0.9214698202026926, "reference_rms_dbfs": -21.95646015705721, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.774791115075151}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003217977906320387, "log_rms_mae_db": 0.7762941912686905, "rms_envelope_nrmse": 0.06802287443343843, "envelope_correlation": 0.9958885245571858, "rms_gain_error_db": -0.06188212507687447, "reference_mean_power": 0.006378025340653872, "output_mean_power": 0.006287789988805602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.934428693189062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0026409373563640536, "output_power": 0.002602173195769143, "reference_active": true, "delta_db": -0.06421900750982489}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022353254716419522, "output_power": 0.00022887544085651058, "reference_active": true, "delta_db": 0.10258426890119263}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.171045723198585e-05, "output_power": 2.747225942114747e-05, "reference_active": true, "delta_db": -0.6230812526083617}], "active_band_count": 3, "active_band_mae_db": 0.2632948430064597}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00029448546471938644, "log_rms_mae_db": 0.40519255120625547, "rms_envelope_nrmse": 0.033558028389472104, "envelope_correlation": 0.9989699074639321, "rms_gain_error_db": -0.061926766027095834, "reference_mean_power": 0.006407121641291136, "output_mean_power": 0.006316409712870421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7183045790225306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0025278777543949896, "output_power": 0.002492595466311253, "reference_active": true, "delta_db": -0.06104267313048437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010431651037251879, "output_power": 0.00010546186964024099, "reference_active": true, "delta_db": 0.04742415806201859}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4474914310652027e-06, "output_power": 1.6102650899094666e-06, "reference_active": true, "delta_db": -1.818238023137183}], "active_band_count": 3, "active_band_mae_db": 0.6422349514432287}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002169046697405618, "log_rms_mae_db": 1.4631853456710555, "rms_envelope_nrmse": 0.091376931911443, "envelope_correlation": 0.993328293110384, "rms_gain_error_db": -0.056077640783556634, "reference_mean_power": 0.00266513165861851, "output_mean_power": 0.002630939759277925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9887000298118345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014141653900070267, "output_power": 0.001392132259076968, "reference_active": true, "delta_db": -0.06819706962166555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010731466350302803, "output_power": 0.00010692416657151591, "reference_active": true, "delta_db": -0.015831943285687072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0173327828420217e-05, "output_power": 1.2728861693024527e-05, "reference_active": true, "delta_db": -1.9998797848783005}], "active_band_count": 3, "active_band_mae_db": 0.6946362659285511}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020963310569053025, "log_rms_mae_db": 1.0018500778625974, "rms_envelope_nrmse": 0.0521014309307506, "envelope_correlation": 0.9977258970726887, "rms_gain_error_db": -0.056323275177269, "reference_mean_power": 0.0026772178601939312, "output_mean_power": 0.0026427214277871492}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.883939001913369e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013521293519634116, "output_power": 0.0013297018206226295, "reference_active": true, "delta_db": -0.07263977147544147}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.792726117783695e-05, "output_power": 4.722883012926911e-05, "reference_active": true, "delta_db": -0.06375423652469556}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6850826685161704e-06, "output_power": 1.0038398110130093e-06, "reference_active": true, "delta_db": -2.2495679653912313}], "active_band_count": 3, "active_band_mae_db": 0.7953206577971228}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002744641361828464, "log_rms_mae_db": 1.4380292448315224, "rms_envelope_nrmse": 0.09247516116047323, "envelope_correlation": 0.9941946814723978, "rms_gain_error_db": -0.15789137153919092, "reference_mean_power": 0.0019381811578355007, "output_mean_power": 0.0018689824791041085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1196235236571854e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006735796402683569, "output_power": 0.0006707791102659171, "reference_active": true, "delta_db": -0.018094225002663104}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013823323993445422, "output_power": 0.00012667756486172622, "reference_active": true, "delta_db": -0.379127809240499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8792319908524634e-05, "output_power": 1.4817971029890058e-05, "reference_active": true, "delta_db": -1.0319165561315797}], "active_band_count": 3, "active_band_mae_db": 0.4763795301249139}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002445180716567873, "log_rms_mae_db": 0.8154456779646829, "rms_envelope_nrmse": 0.050168727127433546, "envelope_correlation": 0.9983569936890991, "rms_gain_error_db": -0.15774112757627778, "reference_mean_power": 0.0019467823981069799, "output_mean_power": 0.00187734157561199}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0571732642454036e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006649824803295284, "output_power": 0.0006596127365381139, "reference_active": true, "delta_db": -0.03521170608456025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927173921912536e-05, "output_power": 6.287140156996728e-05, "reference_active": true, "delta_db": -0.4210294965862882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7481239585012505e-06, "output_power": 1.1383181667492875e-06, "reference_active": true, "delta_db": -1.86308558034493}], "active_band_count": 3, "active_band_mae_db": 0.7731089276719262}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.6882736857616036e-05, "log_rms_mae_db": 1.6666073400894434, "rms_envelope_nrmse": 0.22776411468999475, "envelope_correlation": 0.9605831418299401, "rms_gain_error_db": 0.3920933622881585, "reference_mean_power": 6.564824297759055e-05, "output_mean_power": 7.185093818700341e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.001422573299629e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.567613726314846e-05, "output_power": 1.626711039913248e-05, "reference_active": true, "delta_db": 0.16071356496926392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.79643479102731e-06, "output_power": 6.261284934224978e-06, "reference_active": true, "delta_db": 0.3350251332372695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.958485457614687e-07, "output_power": 9.803676639721123e-07, "reference_active": true, "delta_db": -0.06804315269682568}], "active_band_count": 3, "active_band_mae_db": 0.18792728363445302}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.766393520730197e-05, "log_rms_mae_db": 0.9935341825882298, "rms_envelope_nrmse": 0.1378846302459916, "envelope_correlation": 0.9853408693577281, "rms_gain_error_db": 0.3932480773992067, "reference_mean_power": 6.585532886735065e-05, "output_mean_power": 7.209675709022815e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6817222442955538e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5480772645031266e-05, "output_power": 1.6349934317518907e-05, "reference_active": true, "delta_db": 0.237233797954098}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.979992150286795e-06, "output_power": 3.0967717762543875e-06, "reference_active": true, "delta_db": 0.16694080115834495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.719812912358681e-08, "output_power": 9.372271313232524e-08, "reference_active": true, "delta_db": 0.3133768523995614}], "active_band_count": 3, "active_band_mae_db": 0.23918381717066808}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8516851559784903, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0c897066f4d67db7605f3ce57c8f31d3ac040d7dced40b6bbf7db30d5cfef9e6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9901175211945867, "standard_deviation": 0.002940352029075299, "minimum": 0.9848677674457571, "maximum": 0.9941946814723978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4333180099759693, "standard_deviation": 0.03149454887689174, "minimum": 1.3906056423547963, "maximum": 1.4821865634741165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9436798695998787, "standard_deviation": 0.3476769593033349, "minimum": 0.4763795301249139, "maximum": 1.3720248866802451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1188822852305308, "standard_deviation": 0.018066881254236375, "minimum": 0.09247516116047323, "maximum": 0.1488210525575095, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960382328753672, "standard_deviation": 0.001783491583117641, "minimum": 0.9936852529313412, "maximum": 0.9983569936890991, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8403352572941546, "standard_deviation": 0.026281508350992508, "minimum": 0.8079439989260571, "maximum": 0.8775500534308871, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.193788827756032, "standard_deviation": 0.37201961441542025, "minimum": 0.7317086406144119, "maximum": 1.772548470312936, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07376723973293431, "standard_deviation": 0.018012505324711367, "minimum": 0.050168727127433546, "maximum": 0.0968037960910547, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9599555791421973, "standard_deviation": 0.005091628471722209, "minimum": 0.9539438621639806, "maximum": 0.966961634786814, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7052104708697966, "standard_deviation": 0.0281177732162281, "minimum": 1.6666073400894434, "maximum": 1.7503703668886106, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.26905000370822546, "standard_deviation": 0.11559236845466714, "minimum": 0.12715157079285658, "maximum": 0.4081054863651354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22094355248615244, "standard_deviation": 0.01394407421718196, "minimum": 0.20014545416740326, "maximum": 0.23604979804427814, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9838964934627417, "standard_deviation": 0.0029632930176628517, "minimum": 0.9791099796772637, "maximum": 0.9872222946356993, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0460858814442215, "standard_deviation": 0.04008080287922436, "minimum": 0.9935341825882298, "maximum": 1.1110316361311139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.2915218092921791, "standard_deviation": 0.14861918902893595, "minimum": 0.08190457766320387, "maximum": 0.5635028635216363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13392893071675097, "standard_deviation": 0.012072106507716921, "minimum": 0.1167944911879202, "maximum": 0.1504198086434923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991970802620966, "standard_deviation": 0.002937124641607136, "minimum": 0.9856291930369945, "maximum": 0.9949509219890502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4638338149062669, "standard_deviation": 0.031155316740419638, "minimum": 1.423677439653208, "maximum": 1.5165049438999825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4919412152247103, "standard_deviation": 0.3455374270276583, "minimum": 0.12790808647853277, "maximum": 1.172565021036768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09931327784678984, "standard_deviation": 0.016705981816370764, "minimum": 0.08075090486747197, "maximum": 0.13392132781814747, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965801168395197, "standard_deviation": 0.0015684420601236096, "minimum": 0.9936021672595958, "maximum": 0.9979955950832313, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9693548137400323, "standard_deviation": 0.027794541830270768, "minimum": 0.932864297400303, "maximum": 1.0018500778625974, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5965432073727772, "standard_deviation": 0.3580731098105316, "minimum": 0.1388192199298092, "maximum": 1.206218333588289, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06299359100932538, "standard_deviation": 0.013616512820678342, "minimum": 0.04884622613016216, "maximum": 0.0876019075977768, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3982743073063497, "standard_deviation": 0.03965570088065472, "minimum": 0.35484112888841257, "maximum": 0.465217369754997, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.783058004809789, "standard_deviation": 0.009469232347914892, "minimum": 5.768582017291404, "maximum": 5.797022846632809, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8534258822218592, "standard_deviation": 0.05391457872300635, "minimum": 0.7853014498825822, "maximum": 0.941025076881832, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08356016151167533, "standard_deviation": 0.04314405360332318, "minimum": -0.1432037551165286, "maximum": -0.01170530858173615, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946472555827742, "standard_deviation": 0.0009874878438363925, "minimum": 0.993269351167803, "maximum": 0.9958885245571858, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7794052963143192, "standard_deviation": 0.011125095321705126, "minimum": 0.7580271582323781, "maximum": 0.7930213102483664, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3990311203954859, "standard_deviation": 0.26250277428278507, "minimum": 0.09947581000011722, "maximum": 0.8309082667553763, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07750311808129608, "standard_deviation": 0.007025020614265531, "minimum": 0.06802287443343843, "maximum": 0.08666655235862818, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982906741407728, "standard_deviation": 0.000514089193280425, "minimum": 0.9974363898112654, "maximum": 0.9989725924173543, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4288505906635613, "standard_deviation": 0.015381776708409216, "minimum": 0.40519255120625547, "maximum": 0.45487663928188243, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.737382977126128, "standard_deviation": 0.2675303852162815, "minimum": 0.4126455261081543, "maximum": 1.197602410872592, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04295794194320396, "standard_deviation": 0.0067027936588445755, "minimum": 0.03272446734688274, "maximum": 0.052101946633396984, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0019", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.441006660461426, "predicted_mos": 2.441006660461426, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bfaa8e904a7e0ede39689b60ff1ffbf4ae252f50dff83e6ecd1823b908b2ce43", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.823421800043434, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8072578875922407, "全帯域_10ms_rms_envelope_nrmse": 0.1005579315993242, "全帯域_10ms_envelope_correlation": 0.990968916432693, "全帯域_10ms_modulation_mae_db": 0.2202392844143113, "全帯域_40ms_log_rms_mae_db": 0.4225258424681938, "全帯域_40ms_rms_envelope_nrmse": 0.04791216645143365, "全帯域_40ms_envelope_correlation": 0.9977999540356484, "全帯域_40ms_modulation_mae_db": 0.1210574543893634, "300–1000 Hz_10ms_log_rms_mae_db": 1.248913134084275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11455165070596358, "300–1000 Hz_10ms_envelope_correlation": 0.9894544831618631, "300–1000 Hz_10ms_modulation_mae_db": 0.7576413027432151, "300–1000 Hz_40ms_log_rms_mae_db": 0.7797551057051124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05103243900311958, "300–1000 Hz_40ms_envelope_correlation": 0.9977871502158164, "300–1000 Hz_40ms_modulation_mae_db": 0.7661358536225898, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2887379837162474, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1431002095785628, "1000–3000 Hz_10ms_envelope_correlation": 0.9862482999753586, "1000–3000 Hz_10ms_modulation_mae_db": 0.13789468440259814, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8252698662133248, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08515241870160072, "1000–3000 Hz_40ms_envelope_correlation": 0.9949859490991403, "1000–3000 Hz_40ms_modulation_mae_db": 0.296787966996586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3406254378218019, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15734013914673758, "3000–6000 Hz_10ms_envelope_correlation": 0.9792300093183931, "3000–6000 Hz_10ms_modulation_mae_db": 0.6310418249267995, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8588228714720194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08693349823112248, "3000–6000 Hz_40ms_envelope_correlation": 0.9933624466493715, "3000–6000 Hz_40ms_modulation_mae_db": 0.6675113152624679, "mr_stft_db": 5.466502618033805, "band_mae_db": 0.27664950271722416, "rms_gain_error_db": 0.01323998105556057, "output_peak": 0.8242956265583292}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8167447486246618, "全帯域_10ms_rms_envelope_nrmse": 0.08038431802449746, "全帯域_10ms_envelope_correlation": 0.9942354132069559, "全帯域_10ms_modulation_mae_db": 0.38553761734826536, "全帯域_40ms_log_rms_mae_db": 0.46166285247320543, "全帯域_40ms_rms_envelope_nrmse": 0.042584873517540145, "全帯域_40ms_envelope_correlation": 0.9983153045366903, "全帯域_40ms_modulation_mae_db": 0.7183156439773302, "300–1000 Hz_10ms_log_rms_mae_db": 1.1592687770320942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10615555717417617, "300–1000 Hz_10ms_envelope_correlation": 0.9909749659505613, "300–1000 Hz_10ms_modulation_mae_db": 0.22204759945087682, "300–1000 Hz_40ms_log_rms_mae_db": 0.7599177589795058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07380615379594727, "300–1000 Hz_40ms_envelope_correlation": 0.9953919460633663, "300–1000 Hz_40ms_modulation_mae_db": 0.39114936477117634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2768413216336598, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10663094122100952, "1000–3000 Hz_10ms_envelope_correlation": 0.9920719535142235, "1000–3000 Hz_10ms_modulation_mae_db": 0.466840584762689, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7830733992664852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056750244959075644, "1000–3000 Hz_40ms_envelope_correlation": 0.9976542454536598, "1000–3000 Hz_40ms_modulation_mae_db": 0.7056169410824253, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3294725132913985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19636608767607092, "3000–6000 Hz_10ms_envelope_correlation": 0.969278265419588, "3000–6000 Hz_10ms_modulation_mae_db": 0.29965253881647014, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8242220716803474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1258613192379261, "3000–6000 Hz_40ms_envelope_correlation": 0.9864879939715756, "3000–6000 Hz_40ms_modulation_mae_db": 0.2789225697476116, "mr_stft_db": 5.471943901303355, "band_mae_db": 0.38361332054700203, "rms_gain_error_db": -0.07078823243190092, "output_peak": 0.8348592443007175}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7844730459182174, "全帯域_10ms_rms_envelope_nrmse": 0.09121938355295593, "全帯域_10ms_envelope_correlation": 0.9927332231822108, "全帯域_10ms_modulation_mae_db": 0.3380099233686978, "全帯域_40ms_log_rms_mae_db": 0.43954155128647165, "全帯域_40ms_rms_envelope_nrmse": 0.05982154579805719, "全帯域_40ms_envelope_correlation": 0.9968412401308797, "全帯域_40ms_modulation_mae_db": 0.7544400397465636, "300–1000 Hz_10ms_log_rms_mae_db": 1.236804588833965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1206794139092793, "300–1000 Hz_10ms_envelope_correlation": 0.9885931538006052, "300–1000 Hz_10ms_modulation_mae_db": 0.704010422801678, "300–1000 Hz_40ms_log_rms_mae_db": 0.8045231944488515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08958612656243355, "300–1000 Hz_40ms_envelope_correlation": 0.9935545805086551, "300–1000 Hz_40ms_modulation_mae_db": 0.8846424256913772, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2321210687659363, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11583534637273489, "1000–3000 Hz_10ms_envelope_correlation": 0.9906437023051243, "1000–3000 Hz_10ms_modulation_mae_db": 0.447511510374854, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7322436574657384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06808376107371385, "1000–3000 Hz_40ms_envelope_correlation": 0.9966212275773111, "1000–3000 Hz_40ms_modulation_mae_db": 0.7125642485295943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3486678357594006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1751137967386879, "3000–6000 Hz_10ms_envelope_correlation": 0.9740677614204727, "3000–6000 Hz_10ms_modulation_mae_db": 0.25177297067212995, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8666667827861779, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10756925284632289, "3000–6000 Hz_40ms_envelope_correlation": 0.989203343064976, "3000–6000 Hz_40ms_modulation_mae_db": 0.34167953702789533, "mr_stft_db": 5.458620093445312, "band_mae_db": 0.37468254105952314, "rms_gain_error_db": -0.16355132123013652, "output_peak": 0.8897013375446929}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.809675864377399, "全帯域_10ms_rms_envelope_nrmse": 0.08063312876563071, "全帯域_10ms_envelope_correlation": 0.9942588932918686, "全帯域_10ms_modulation_mae_db": 0.6052024913757678, "全帯域_40ms_log_rms_mae_db": 0.45138473630586706, "全帯域_40ms_rms_envelope_nrmse": 0.04720683208535763, "全帯域_40ms_envelope_correlation": 0.9979890111139578, "全帯域_40ms_modulation_mae_db": 0.832349296581631, "300–1000 Hz_10ms_log_rms_mae_db": 1.1715167327340537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.083804970316349, "300–1000 Hz_10ms_envelope_correlation": 0.9943932407956586, "300–1000 Hz_10ms_modulation_mae_db": 0.5477677095637786, "300–1000 Hz_40ms_log_rms_mae_db": 0.7354722767030613, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04646613774495207, "300–1000 Hz_40ms_envelope_correlation": 0.9981586495346596, "300–1000 Hz_40ms_modulation_mae_db": 0.6061065488709906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.247047600213272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.132932100901176, "1000–3000 Hz_10ms_envelope_correlation": 0.9883556389754713, "1000–3000 Hz_10ms_modulation_mae_db": 0.9601105836264296, "1000–3000 Hz_40ms_log_rms_mae_db": 0.763083726217937, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09071208499761478, "1000–3000 Hz_40ms_envelope_correlation": 0.9948822527646696, "1000–3000 Hz_40ms_modulation_mae_db": 1.0462889846384724, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2935414641802296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17253490196196677, "3000–6000 Hz_10ms_envelope_correlation": 0.9749448891115591, "3000–6000 Hz_10ms_modulation_mae_db": 0.4845740935605971, "3000–6000 Hz_40ms_log_rms_mae_db": 0.821814289082716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09990681346357097, "3000–6000 Hz_40ms_envelope_correlation": 0.9906404001582109, "3000–6000 Hz_40ms_modulation_mae_db": 0.571923000704493, "mr_stft_db": 5.469889823510719, "band_mae_db": 0.33228072303893036, "rms_gain_error_db": -0.1081679867732803, "output_peak": 0.8912042906342541}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8632574909146574, "全帯域_10ms_rms_envelope_nrmse": 0.08556859137492005, "全帯域_10ms_envelope_correlation": 0.9935071242290161, "全帯域_10ms_modulation_mae_db": 0.21017517562015509, "全帯域_40ms_log_rms_mae_db": 0.47258324357493436, "全帯域_40ms_rms_envelope_nrmse": 0.043129031222341255, "全帯域_40ms_envelope_correlation": 0.9983451682071797, "全帯域_40ms_modulation_mae_db": 0.7360420187605065, "300–1000 Hz_10ms_log_rms_mae_db": 1.2374491348367822, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08873258690685962, "300–1000 Hz_10ms_envelope_correlation": 0.9937142585047057, "300–1000 Hz_10ms_modulation_mae_db": 0.8511254071471009, "300–1000 Hz_40ms_log_rms_mae_db": 0.7962872980914282, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05313598554545215, "300–1000 Hz_40ms_envelope_correlation": 0.997643650881511, "300–1000 Hz_40ms_modulation_mae_db": 0.9799797066599379, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2239063018961005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13400924909266748, "1000–3000 Hz_10ms_envelope_correlation": 0.9876368009550114, "1000–3000 Hz_10ms_modulation_mae_db": 0.31545819378419915, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7453647188951289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07212559758202014, "1000–3000 Hz_40ms_envelope_correlation": 0.9965478400724259, "1000–3000 Hz_40ms_modulation_mae_db": 0.7517516302400046, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3320181992695224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18484513966690572, "3000–6000 Hz_10ms_envelope_correlation": 0.9724719219467496, "3000–6000 Hz_10ms_modulation_mae_db": 0.3082545021241091, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8449308207316027, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12097891043445264, "3000–6000 Hz_40ms_envelope_correlation": 0.9872036096188395, "3000–6000 Hz_40ms_modulation_mae_db": 0.33180088316124406, "mr_stft_db": 5.4562218774010605, "band_mae_db": 0.4061217902986391, "rms_gain_error_db": -0.10820951711205243, "output_peak": 0.8486603203209381}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7918534934595941, "全帯域_10ms_rms_envelope_nrmse": 0.08072008313678827, "全帯域_10ms_envelope_correlation": 0.9941662233104919, "全帯域_10ms_modulation_mae_db": 0.20519964238103594, "全帯域_40ms_log_rms_mae_db": 0.41614881729165615, "全帯域_40ms_rms_envelope_nrmse": 0.0412684539017029, "全帯域_40ms_envelope_correlation": 0.998392361351458, "全帯域_40ms_modulation_mae_db": 0.3477207123959228, "300–1000 Hz_10ms_log_rms_mae_db": 1.241952583400915, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10890806245835731, "300–1000 Hz_10ms_envelope_correlation": 0.9904765256163712, "300–1000 Hz_10ms_modulation_mae_db": 0.42206216961477033, "300–1000 Hz_40ms_log_rms_mae_db": 0.7922077504372934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06372467514754014, "300–1000 Hz_40ms_envelope_correlation": 0.9965573427010029, "300–1000 Hz_40ms_modulation_mae_db": 0.520186045472039, "1000–3000 Hz_10ms_log_rms_mae_db": 1.254682163579205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10681163470521383, "1000–3000 Hz_10ms_envelope_correlation": 0.9920709204192549, "1000–3000 Hz_10ms_modulation_mae_db": 0.35921897151576765, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7628197282389427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05027873591908693, "1000–3000 Hz_40ms_envelope_correlation": 0.9982234192900415, "1000–3000 Hz_40ms_modulation_mae_db": 0.5024026404384437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3294442358935823, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1720021181462103, "3000–6000 Hz_10ms_envelope_correlation": 0.9750145914951286, "3000–6000 Hz_10ms_modulation_mae_db": 0.33113608504727626, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8180801901535412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.096486770951011, "3000–6000 Hz_40ms_envelope_correlation": 0.9912919472927179, "3000–6000 Hz_40ms_modulation_mae_db": 0.3976463620679614, "mr_stft_db": 5.471132306451021, "band_mae_db": 0.3555262034433902, "rms_gain_error_db": -0.053970483235915495, "output_peak": 0.8552804965378794}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8184017459406587, "全帯域_10ms_rms_envelope_nrmse": 0.08527986691478845, "全帯域_10ms_envelope_correlation": 0.9935393154436573, "全帯域_10ms_modulation_mae_db": 0.22091249370154264, "全帯域_40ms_log_rms_mae_db": 0.464990086943362, "全帯域_40ms_rms_envelope_nrmse": 0.044066179927709594, "全帯域_40ms_envelope_correlation": 0.9982437742189665, "全帯域_40ms_modulation_mae_db": 0.6358262442844548, "300–1000 Hz_10ms_log_rms_mae_db": 1.2472049765829942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10446560815218588, "300–1000 Hz_10ms_envelope_correlation": 0.9912803914733963, "300–1000 Hz_10ms_modulation_mae_db": 0.34582340078874374, "300–1000 Hz_40ms_log_rms_mae_db": 0.7871282210879138, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07306012664550844, "300–1000 Hz_40ms_envelope_correlation": 0.9955048902596282, "300–1000 Hz_40ms_modulation_mae_db": 0.43521463610730665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2845201028308046, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11831840146709284, "1000–3000 Hz_10ms_envelope_correlation": 0.9903254719192071, "1000–3000 Hz_10ms_modulation_mae_db": 0.628482814247009, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7906980638978093, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05842144133879775, "1000–3000 Hz_40ms_envelope_correlation": 0.9976955160010621, "1000–3000 Hz_40ms_modulation_mae_db": 0.9854535590388217, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3053783328925672, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15896783126535444, "3000–6000 Hz_10ms_envelope_correlation": 0.9786530729444191, "3000–6000 Hz_10ms_modulation_mae_db": 0.21666096352212919, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8076211458627232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08427743893327212, "3000–6000 Hz_40ms_envelope_correlation": 0.9934876312204901, "3000–6000 Hz_40ms_modulation_mae_db": 0.29923376413428665, "mr_stft_db": 5.466220994147907, "band_mae_db": 0.4170629929346265, "rms_gain_error_db": -0.09985938505838735, "output_peak": 0.8035016332649396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8019801487105017, "全帯域_10ms_rms_envelope_nrmse": 0.08366628844554426, "全帯域_10ms_envelope_correlation": 0.9937533193314875, "全帯域_10ms_modulation_mae_db": 0.6527354498879201, "全帯域_40ms_log_rms_mae_db": 0.4491524985168492, "全帯域_40ms_rms_envelope_nrmse": 0.04209553504120068, "全帯域_40ms_envelope_correlation": 0.9983655031108278, "全帯域_40ms_modulation_mae_db": 0.966946245899341, "300–1000 Hz_10ms_log_rms_mae_db": 1.2585193124586722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10595879160781553, "300–1000 Hz_10ms_envelope_correlation": 0.9910569246803196, "300–1000 Hz_10ms_modulation_mae_db": 0.39624849852486727, "300–1000 Hz_40ms_log_rms_mae_db": 0.7969889750582496, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06301660655535775, "300–1000 Hz_40ms_envelope_correlation": 0.996739779089777, "300–1000 Hz_40ms_modulation_mae_db": 0.4704480701243841, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2136169909889116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12714903252285406, "1000–3000 Hz_10ms_envelope_correlation": 0.9887046131498318, "1000–3000 Hz_10ms_modulation_mae_db": 0.7319300271535255, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7637222282831628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07327338752201235, "1000–3000 Hz_40ms_envelope_correlation": 0.9960619100116732, "1000–3000 Hz_40ms_modulation_mae_db": 1.058682085213898, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3512750615568825, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20289662798567076, "3000–6000 Hz_10ms_envelope_correlation": 0.9660503081547233, "3000–6000 Hz_10ms_modulation_mae_db": 0.5632376976686407, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8573742086840725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1081877158618502, "3000–6000 Hz_40ms_envelope_correlation": 0.9893426076823123, "3000–6000 Hz_40ms_modulation_mae_db": 0.5336430197664686, "mr_stft_db": 5.470613384350912, "band_mae_db": 0.3872370564248991, "rms_gain_error_db": -0.0801769279268465, "output_peak": 0.8961594484096383}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.331057663597207, "spectral_convergence": 0.26688270788686097, "spectral_convergence_status": "ok", "amplitude_floor": 2.005860984254184e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.358223345499739, "spectral_convergence": 0.2708305944308789, "spectral_convergence_status": "ok", "amplitude_floor": 1.6673734947378395e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.71022684500447, "spectral_convergence": 0.2681576831313387, "spectral_convergence_status": "ok", "amplitude_floor": 1.455049241978443e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.2140486659240935e-05, "output_power": 4.4467131513280836e-05, "delta_db": 1.4098666647240803, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002187325069648938, "output_power": 0.00216781728090874, "delta_db": -0.03890656600177645, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018573993391017246, "output_power": 0.0018334869168389505, "delta_db": -0.056274713569909186, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.002022873257906857, "output_power": 0.0020736372603362965, "delta_db": 0.10764114796046309, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00023141111513331255, "output_power": 0.00023084455371644378, "delta_db": -0.010645823806810121, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.19984427950361e-05, "output_power": 4.2353509049853965e-05, "delta_db": 0.03656210024030575, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27664950271722416, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.01323998105556057, "pitch": null, "output_rms_dbfs": -21.943220176001653, "output_peak": 0.8242956265583292, "reference_rms_dbfs": -21.95646015705721, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.466502618033805}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003217977906320387, "log_rms_mae_db": 0.8072578875922407, "rms_envelope_nrmse": 0.1005579315993242, "envelope_correlation": 0.990968916432693, "rms_gain_error_db": 0.01336793487290609, "reference_mean_power": 0.006378025340653872, "output_mean_power": 0.006397687663301923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.934428693189062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0026409373563640536, "output_power": 0.00266325263402812, "reference_active": true, "delta_db": 0.0365426545737561}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022353254716419522, "output_power": 0.00021676394868211182, "reference_active": true, "delta_db": -0.13353713210958568}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.171045723198585e-05, "output_power": 2.832296381426301e-05, "reference_active": true, "delta_db": -0.49063806655959213}], "active_band_count": 3, "active_band_mae_db": 0.2202392844143113}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00029448546471938644, "log_rms_mae_db": 0.4225258424681938, "rms_envelope_nrmse": 0.04791216645143365, "envelope_correlation": 0.9977999540356484, "rms_gain_error_db": 0.01347939343432087, "reference_mean_power": 0.006407121641291136, "output_mean_power": 0.00642703860589733}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7183045790225306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0025278777543949896, "output_power": 0.0025471840691569827, "reference_active": true, "delta_db": 0.03304261782481627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010431651037251879, "output_power": 0.00010013112184743834, "reference_active": true, "delta_db": -0.17783968546268503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4474914310652027e-06, "output_power": 2.534837865179564e-06, "reference_active": true, "delta_db": 0.1522900598805889}], "active_band_count": 3, "active_band_mae_db": 0.1210574543893634}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002169046697405618, "log_rms_mae_db": 1.248913134084275, "rms_envelope_nrmse": 0.11455165070596358, "envelope_correlation": 0.9894544831618631, "rms_gain_error_db": -0.0447458442641203, "reference_mean_power": 0.00266513165861851, "output_mean_power": 0.0026378134830146514}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9887000298118345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014141653900070267, "output_power": 0.0014104984783932038, "reference_active": true, "delta_db": -0.011275822174318322}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010731466350302803, "output_power": 9.40665302106915e-05, "reference_active": true, "delta_db": -0.5722394343211145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0173327828420217e-05, "output_power": 1.367214579641766e-05, "reference_active": true, "delta_db": -1.6894086517342124}], "active_band_count": 3, "active_band_mae_db": 0.7576413027432151}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020963310569053025, "log_rms_mae_db": 0.7797551057051124, "rms_envelope_nrmse": 0.05103243900311958, "envelope_correlation": 0.9977871502158164, "rms_gain_error_db": -0.04464057429969998, "reference_mean_power": 0.0026772178601939312, "output_mean_power": 0.0026498400279137456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.883939001913369e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013521293519634116, "output_power": 0.0013446252946888346, "reference_active": true, "delta_db": -0.024169637246945933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.792726117783695e-05, "output_power": 4.110425843551693e-05, "reference_active": true, "delta_db": -0.6669579429405914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6850826685161704e-06, "output_power": 1.1638398240740614e-06, "reference_active": true, "delta_db": -1.6072799806802323}], "active_band_count": 3, "active_band_mae_db": 0.7661358536225898}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002744641361828464, "log_rms_mae_db": 1.2887379837162474, "rms_envelope_nrmse": 0.1431002095785628, "envelope_correlation": 0.9862482999753586, "rms_gain_error_db": 0.14981186494923932, "reference_mean_power": 0.0019381811578355007, "output_mean_power": 0.0020062061330160053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1196235236571854e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006735796402683569, "output_power": 0.0006901990835630467, "reference_active": true, "delta_db": 0.10585426673715537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013823323993445422, "output_power": 0.00013081677071789645, "reference_active": true, "delta_db": -0.23949063072191779}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8792319908524634e-05, "output_power": 1.9090368540942682e-05, "reference_active": true, "delta_db": 0.06833915574872122}], "active_band_count": 3, "active_band_mae_db": 0.13789468440259814}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002445180716567873, "log_rms_mae_db": 0.8252698662133248, "rms_envelope_nrmse": 0.08515241870160072, "envelope_correlation": 0.9949859490991403, "rms_gain_error_db": 0.1500744007042128, "reference_mean_power": 0.0019467823981069799, "output_mean_power": 0.002015231073053968}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0571732642454036e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006649824803295284, "output_power": 0.0006782768090417247, "reference_active": true, "delta_db": 0.08596764854915089}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927173921912536e-05, "output_power": 6.537814456333148e-05, "reference_active": true, "delta_db": -0.2512350058122435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7481239585012505e-06, "output_power": 1.9855843888295385e-06, "reference_active": true, "delta_db": 0.5531612466283635}], "active_band_count": 3, "active_band_mae_db": 0.296787966996586}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.6882736857616036e-05, "log_rms_mae_db": 1.3406254378218019, "rms_envelope_nrmse": 0.15734013914673758, "envelope_correlation": 0.9792300093183931, "rms_gain_error_db": -0.22125717488372637, "reference_mean_power": 6.564824297759055e-05, "output_mean_power": 6.23874722285353e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.001422573299629e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.567613726314846e-05, "output_power": 1.557705661745733e-05, "reference_active": true, "delta_db": -0.027536591627071366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.79643479102731e-06, "output_power": 5.26787935998529e-06, "reference_active": true, "delta_db": -0.41525133883871085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.958485457614687e-07, "output_power": 7.131149464219656e-07, "reference_active": true, "delta_db": -1.4503375443146163}], "active_band_count": 3, "active_band_mae_db": 0.6310418249267995}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.766393520730197e-05, "log_rms_mae_db": 0.8588228714720194, "rms_envelope_nrmse": 0.08693349823112248, "envelope_correlation": 0.9933624466493715, "rms_gain_error_db": -0.22240692711563068, "reference_mean_power": 6.585532886735065e-05, "output_mean_power": 6.256770570920216e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6817222442955538e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5480772645031266e-05, "output_power": 1.5458139811348798e-05, "reference_active": true, "delta_db": -0.006354015542946505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.979992150286795e-06, "output_power": 2.5897355808610753e-06, "reference_active": true, "delta_db": -0.6095969640424503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.719812912358681e-08, "output_power": 6.336491701271248e-08, "reference_active": true, "delta_db": -1.3865829662020066}], "active_band_count": 3, "active_band_mae_db": 0.6675113152624679}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8611912749983707, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4ee364d30a85cf1dd59d4c27f7e0802bfeee6ba9fd1f981dc5c971a215704f2b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895071751516854, "standard_deviation": 0.0021112247562080923, "minimum": 0.9862482999753586, "maximum": 0.9920719535142235, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2526841917030174, "standard_deviation": 0.02856067795287942, "minimum": 1.2136169909889116, "maximum": 1.2887379837162474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.505930921233384, "standard_deviation": 0.2592763384533077, "minimum": 0.13789468440259814, "maximum": 0.9601105836264296, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12309836448266392, "standard_deviation": 0.013332369674438838, "minimum": 0.10663094122100952, "maximum": 0.1431002095785628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996584045033748, "standard_deviation": 0.001241748218527382, "minimum": 0.9948822527646696, "maximum": 0.9982234192900415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7707844235598161, "standard_deviation": 0.028835492034664293, "minimum": 0.7322436574657384, "maximum": 0.8252698662133248, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7574435070222808, "standard_deviation": 0.26942979026515923, "minimum": 0.296787966996586, "maximum": 1.058682085213898, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06934970901174027, "standard_deviation": 0.014003345107197107, "minimum": 0.05027873591908693, "maximum": 0.09071208499761478, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9737138524763791, "standard_deviation": 0.0044417378423178226, "minimum": 0.9660503081547233, "maximum": 0.9792300093183931, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.328802885083173, "standard_deviation": 0.020151792411003715, "minimum": 1.2935414641802296, "maximum": 1.3512750615568825, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.385791334542269, "standard_deviation": 0.1532402881395442, "minimum": 0.21666096352212919, "maximum": 0.6310418249267995, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17750833032345054, "standard_deviation": 0.016323964136279942, "minimum": 0.15734013914673758, "maximum": 0.20289662798567076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9901274974573118, "standard_deviation": 0.0025812407535870246, "minimum": 0.9864879939715756, "maximum": 0.9934876312204901, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8374415475566501, "standard_deviation": 0.022193177350702933, "minimum": 0.8076211458627232, "maximum": 0.8666667827861779, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4277950564840535, "standard_deviation": 0.14425734511372773, "minimum": 0.2789225697476116, "maximum": 0.6675113152624679, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10377521499494105, "standard_deviation": 0.014878175130125279, "minimum": 0.08427743893327212, "maximum": 0.1258613192379261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9912429929979352, "standard_deviation": 0.001962031032496708, "minimum": 0.9885931538006052, "maximum": 0.9943932407956586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.225203654995469, "standard_deviation": 0.03770351325353046, "minimum": 1.1592687770320942, "maximum": 1.2585193124586722, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5308408138293789, "standard_deviation": 0.22162629192588718, "minimum": 0.22204759945087682, "maximum": 0.8511254071471009, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1041570801538733, "standard_deviation": 0.01232341020404686, "minimum": 0.083804970316349, "maximum": 0.1206794139092793, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996417248656802, "standard_deviation": 0.0015406554317837988, "minimum": 0.9935545805086551, "maximum": 0.9981586495346596, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.781535072563927, "standard_deviation": 0.02306002125714521, "minimum": 0.7354722767030613, "maximum": 0.8045231944488515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6317328314149752, "standard_deviation": 0.22003742958174594, "minimum": 0.39114936477117634, "maximum": 0.9799797066599379, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06422853137503887, "standard_deviation": 0.014278566590488265, "minimum": 0.04646613774495207, "maximum": 0.08958612656243355, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.36664676630802934, "standard_deviation": 0.045167419380177876, "minimum": 0.27664950271722416, "maximum": 0.4170629929346265, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.466393124830511, "standard_deviation": 0.005942875240670533, "minimum": 5.4562218774010605, "maximum": 5.471943901303355, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8554577996964237, "standard_deviation": 0.03432641136066883, "minimum": 0.8035016332649396, "maximum": 0.8961594484096383, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08393548408911987, "standard_deviation": 0.05112539805511952, "minimum": -0.16355132123013652, "maximum": 0.01323998105556057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933953035535477, "standard_deviation": 0.0011040841195370632, "minimum": 0.990968916432693, "maximum": 0.9942588932918686, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8117055531922414, "standard_deviation": 0.023850826152903796, "minimum": 0.7844730459182174, "maximum": 0.8632574909146574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35475150976221204, "standard_deviation": 0.18206069879442166, "minimum": 0.20519964238103594, "maximum": 0.6527354498879201, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08600369897680617, "standard_deviation": 0.0069028240113775795, "minimum": 0.08038431802449746, "maximum": 0.1005579315993242, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998036539588201, "standard_deviation": 0.0005261322907768808, "minimum": 0.9968412401308797, "maximum": 0.998392361351458, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44724870360756747, "standard_deviation": 0.02008168133173129, "minimum": 0.41614881729165615, "maximum": 0.47258324357493436, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6390872070043891, "standard_deviation": 0.2744046371107139, "minimum": 0.1210574543893634, "maximum": 0.966946245899341, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.046010577243167886, "standard_deviation": 0.006064235238837747, "minimum": 0.0412684539017029, "maximum": 0.05982154579805719, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0019", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2465453147888184, "predicted_mos": 2.2465453147888184, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ed3a3e054046d4e06f44af9006bb812bf7fe83629a96cdf3130883526f089674", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.229588099755347, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8085385706359092, "全帯域_10ms_rms_envelope_nrmse": 0.07724769665265467, "全帯域_10ms_envelope_correlation": 0.9947322837591313, "全帯域_10ms_modulation_mae_db": 0.35363085384026877, "全帯域_40ms_log_rms_mae_db": 0.45799751908102626, "全帯域_40ms_rms_envelope_nrmse": 0.03770515226839409, "全帯域_40ms_envelope_correlation": 0.998766364964389, "全帯域_40ms_modulation_mae_db": 0.6578303463308645, "300–1000 Hz_10ms_log_rms_mae_db": 1.2912949969071668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07905545615143718, "300–1000 Hz_10ms_envelope_correlation": 0.9951379706568557, "300–1000 Hz_10ms_modulation_mae_db": 0.5933447777433055, "300–1000 Hz_40ms_log_rms_mae_db": 0.8330877035242945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04810534457856316, "300–1000 Hz_40ms_envelope_correlation": 0.9982140433966434, "300–1000 Hz_40ms_modulation_mae_db": 0.7470240492658307, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2942885000981794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11627098203470466, "1000–3000 Hz_10ms_envelope_correlation": 0.9906335851978246, "1000–3000 Hz_10ms_modulation_mae_db": 0.7337785898438539, "1000–3000 Hz_40ms_log_rms_mae_db": 0.768382230825778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05764677558683337, "1000–3000 Hz_40ms_envelope_correlation": 0.9977143941833714, "1000–3000 Hz_40ms_modulation_mae_db": 0.7993834468770066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4874600967537508, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20649353738399065, "3000–6000 Hz_10ms_envelope_correlation": 0.9655514681083132, "3000–6000 Hz_10ms_modulation_mae_db": 0.6807687356210338, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9016466573957704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11812955418883972, "3000–6000 Hz_40ms_envelope_correlation": 0.9878151795032716, "3000–6000 Hz_40ms_modulation_mae_db": 0.4697344880879791, "mr_stft_db": 5.5406355839980135, "band_mae_db": 0.3189070039756868, "rms_gain_error_db": -0.10312405546448383, "output_peak": 0.7551253835662786}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7898173066623402, "全帯域_10ms_rms_envelope_nrmse": 0.08409651282999568, "全帯域_10ms_envelope_correlation": 0.9937247387162709, "全帯域_10ms_modulation_mae_db": 0.2580044600691642, "全帯域_40ms_log_rms_mae_db": 0.44324113756961825, "全帯域_40ms_rms_envelope_nrmse": 0.04998772625418853, "全帯域_40ms_envelope_correlation": 0.9976350267886036, "全帯域_40ms_modulation_mae_db": 0.1713771134925212, "300–1000 Hz_10ms_log_rms_mae_db": 1.2957050839079831, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09158238515507693, "300–1000 Hz_10ms_envelope_correlation": 0.9933467301061338, "300–1000 Hz_10ms_modulation_mae_db": 1.0927775286706265, "300–1000 Hz_40ms_log_rms_mae_db": 0.8738780072335274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061774870863760566, "300–1000 Hz_40ms_envelope_correlation": 0.9968186504709589, "300–1000 Hz_40ms_modulation_mae_db": 1.365595799543729, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2915441928014462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13517563313058062, "1000–3000 Hz_10ms_envelope_correlation": 0.9882272480833777, "1000–3000 Hz_10ms_modulation_mae_db": 0.27726806136210663, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8058156966628875, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08684271507530279, "1000–3000 Hz_40ms_envelope_correlation": 0.9952059550715643, "1000–3000 Hz_40ms_modulation_mae_db": 0.3461589816870623, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4344499714367989, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17245936513658697, "3000–6000 Hz_10ms_envelope_correlation": 0.9749583251947309, "3000–6000 Hz_10ms_modulation_mae_db": 0.6352511122524644, "3000–6000 Hz_40ms_log_rms_mae_db": 0.870453677834967, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09236224497888611, "3000–6000 Hz_40ms_envelope_correlation": 0.9925086889241375, "3000–6000 Hz_40ms_modulation_mae_db": 0.6784422449996028, "mr_stft_db": 5.550952861331663, "band_mae_db": 0.3227072392710674, "rms_gain_error_db": 0.04460837429221104, "output_peak": 0.827162690127598}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8309894812591883, "全帯域_10ms_rms_envelope_nrmse": 0.0804207050042362, "全帯域_10ms_envelope_correlation": 0.9943838871917859, "全帯域_10ms_modulation_mae_db": 0.829166240432173, "全帯域_40ms_log_rms_mae_db": 0.4568801970944229, "全帯域_40ms_rms_envelope_nrmse": 0.05017259787640022, "全帯域_40ms_envelope_correlation": 0.9978127691916058, "全帯域_40ms_modulation_mae_db": 1.2695665394876725, "300–1000 Hz_10ms_log_rms_mae_db": 1.2938893524302104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08370304647428292, "300–1000 Hz_10ms_envelope_correlation": 0.9945524426932163, "300–1000 Hz_10ms_modulation_mae_db": 0.6561156267177384, "300–1000 Hz_40ms_log_rms_mae_db": 0.8576956488160739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05104263260873581, "300–1000 Hz_40ms_envelope_correlation": 0.9979897024143728, "300–1000 Hz_40ms_modulation_mae_db": 0.915447411134731, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3121388928692892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13989813761279224, "1000–3000 Hz_10ms_envelope_correlation": 0.9866146451573929, "1000–3000 Hz_10ms_modulation_mae_db": 1.2939822330605184, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8205022504352051, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09000947304647108, "1000–3000 Hz_40ms_envelope_correlation": 0.9945225681019041, "1000–3000 Hz_40ms_modulation_mae_db": 1.6438714594557464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.494425576517197, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19322237289888167, "3000–6000 Hz_10ms_envelope_correlation": 0.9693472067111284, "3000–6000 Hz_10ms_modulation_mae_db": 0.637328455477712, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9135666086486461, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12124207191506788, "3000–6000 Hz_40ms_envelope_correlation": 0.9867305946954612, "3000–6000 Hz_40ms_modulation_mae_db": 0.7149419827335363, "mr_stft_db": 5.567519463682184, "band_mae_db": 0.42763774267396854, "rms_gain_error_db": -0.1409211747257387, "output_peak": 0.857224457908677}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7626902669597991, "全帯域_10ms_rms_envelope_nrmse": 0.07308596388358563, "全帯域_10ms_envelope_correlation": 0.9953247234590713, "全帯域_10ms_modulation_mae_db": 0.23449751023201817, "全帯域_40ms_log_rms_mae_db": 0.42869033161827635, "全帯域_40ms_rms_envelope_nrmse": 0.03778062938369419, "全帯域_40ms_envelope_correlation": 0.9987887739338436, "全帯域_40ms_modulation_mae_db": 0.3817108151651543, "300–1000 Hz_10ms_log_rms_mae_db": 1.3639010118660202, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09896088394748462, "300–1000 Hz_10ms_envelope_correlation": 0.9922842052589395, "300–1000 Hz_10ms_modulation_mae_db": 0.6563048962762484, "300–1000 Hz_40ms_log_rms_mae_db": 0.9120073153611269, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06702441741901824, "300–1000 Hz_40ms_envelope_correlation": 0.9963610876525203, "300–1000 Hz_40ms_modulation_mae_db": 0.7436615770251457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3246313667328826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11297963685491626, "1000–3000 Hz_10ms_envelope_correlation": 0.991088399878135, "1000–3000 Hz_10ms_modulation_mae_db": 0.3359769053049701, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7971439085872507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05738029515787735, "1000–3000 Hz_40ms_envelope_correlation": 0.9975928373042622, "1000–3000 Hz_40ms_modulation_mae_db": 0.41706123737636397, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4197188929980469, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16809582578720067, "3000–6000 Hz_10ms_envelope_correlation": 0.9761941755273467, "3000–6000 Hz_10ms_modulation_mae_db": 0.4224027113281245, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8525512577070908, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09069376374809515, "3000–6000 Hz_40ms_envelope_correlation": 0.9922816423925058, "3000–6000 Hz_40ms_modulation_mae_db": 0.4805009518184657, "mr_stft_db": 5.5323454222879045, "band_mae_db": 0.31976232249265785, "rms_gain_error_db": -0.11841078192904197, "output_peak": 0.8201479292160746}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7538965715956056, "全帯域_10ms_rms_envelope_nrmse": 0.07487014028377229, "全帯域_10ms_envelope_correlation": 0.9951044727873363, "全帯域_10ms_modulation_mae_db": 0.5791567809142105, "全帯域_40ms_log_rms_mae_db": 0.4193088377067705, "全帯域_40ms_rms_envelope_nrmse": 0.039088633629455534, "全帯域_40ms_envelope_correlation": 0.9987100230666596, "全帯域_40ms_modulation_mae_db": 0.8458550373611994, "300–1000 Hz_10ms_log_rms_mae_db": 1.3134984909846354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09688809916722273, "300–1000 Hz_10ms_envelope_correlation": 0.9926380853723985, "300–1000 Hz_10ms_modulation_mae_db": 0.9639399208719505, "300–1000 Hz_40ms_log_rms_mae_db": 0.8620470010003008, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06528786855371746, "300–1000 Hz_40ms_envelope_correlation": 0.9965764147220783, "300–1000 Hz_40ms_modulation_mae_db": 1.1215651735414203, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3190306415332305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11440252301580811, "1000–3000 Hz_10ms_envelope_correlation": 0.9909480071271369, "1000–3000 Hz_10ms_modulation_mae_db": 0.8409810721346176, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8017979008378092, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06482904716684278, "1000–3000 Hz_40ms_envelope_correlation": 0.9970691680080551, "1000–3000 Hz_40ms_modulation_mae_db": 0.9685785607296369, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4497182101758994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20207574457419755, "3000–6000 Hz_10ms_envelope_correlation": 0.9654745891695277, "3000–6000 Hz_10ms_modulation_mae_db": 0.6833422709508586, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9198342006973972, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11622861335505091, "3000–6000 Hz_40ms_envelope_correlation": 0.9872904677876964, "3000–6000 Hz_40ms_modulation_mae_db": 0.6987211748011605, "mr_stft_db": 5.545947733453024, "band_mae_db": 0.3158841665128536, "rms_gain_error_db": -0.12164484322130331, "output_peak": 0.89543745360491}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8200303095281993, "全帯域_10ms_rms_envelope_nrmse": 0.08718913210817528, "全帯域_10ms_envelope_correlation": 0.993452803497452, "全帯域_10ms_modulation_mae_db": 0.5328741647084431, "全帯域_40ms_log_rms_mae_db": 0.47051820507958786, "全帯域_40ms_rms_envelope_nrmse": 0.05532057532131933, "全帯域_40ms_envelope_correlation": 0.9974176462904925, "全帯域_40ms_modulation_mae_db": 1.0115119964972934, "300–1000 Hz_10ms_log_rms_mae_db": 1.2861228852452684, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10915986288548277, "300–1000 Hz_10ms_envelope_correlation": 0.9905626583286827, "300–1000 Hz_10ms_modulation_mae_db": 0.5527177335807939, "300–1000 Hz_40ms_log_rms_mae_db": 0.8825191280750044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08103629430401482, "300–1000 Hz_40ms_envelope_correlation": 0.9945667732482835, "300–1000 Hz_40ms_modulation_mae_db": 0.6616777478866066, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3297749231939444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12533035005630055, "1000–3000 Hz_10ms_envelope_correlation": 0.9897277271800731, "1000–3000 Hz_10ms_modulation_mae_db": 0.9515063009713924, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8372208566356726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08134997469448106, "1000–3000 Hz_40ms_envelope_correlation": 0.9960603141250959, "1000–3000 Hz_40ms_modulation_mae_db": 1.4010311609444364, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4461115808661085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19607056185934477, "3000–6000 Hz_10ms_envelope_correlation": 0.9673874411989075, "3000–6000 Hz_10ms_modulation_mae_db": 0.4174201960969171, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8945506933266384, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10937522849402267, "3000–6000 Hz_40ms_envelope_correlation": 0.9892823134970294, "3000–6000 Hz_40ms_modulation_mae_db": 0.46835825708129936, "mr_stft_db": 5.548556712162349, "band_mae_db": 0.36228091950029917, "rms_gain_error_db": -0.17221612555355592, "output_peak": 0.7991243208396143}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8013581459911545, "全帯域_10ms_rms_envelope_nrmse": 0.0799427451183992, "全帯域_10ms_envelope_correlation": 0.9942710874986901, "全帯域_10ms_modulation_mae_db": 0.22180915937718526, "全帯域_40ms_log_rms_mae_db": 0.44290613351254343, "全帯域_40ms_rms_envelope_nrmse": 0.04780897603650604, "全帯域_40ms_envelope_correlation": 0.9978041912329038, "全帯域_40ms_modulation_mae_db": 0.19269404450795669, "300–1000 Hz_10ms_log_rms_mae_db": 1.2955427507239736, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09105625231240908, "300–1000 Hz_10ms_envelope_correlation": 0.9935638956107026, "300–1000 Hz_10ms_modulation_mae_db": 0.3231482409216702, "300–1000 Hz_40ms_log_rms_mae_db": 0.8330123565358617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06579290973303378, "300–1000 Hz_40ms_envelope_correlation": 0.9965642131011291, "300–1000 Hz_40ms_modulation_mae_db": 0.41739922448493516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3009609943749747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12487707328491193, "1000–3000 Hz_10ms_envelope_correlation": 0.9897925780612382, "1000–3000 Hz_10ms_modulation_mae_db": 0.15400693258190595, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7925346958691482, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0843638562403676, "1000–3000 Hz_40ms_envelope_correlation": 0.9952854740679882, "1000–3000 Hz_40ms_modulation_mae_db": 0.32160566888046443, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4477445863892606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19519923780898396, "3000–6000 Hz_10ms_envelope_correlation": 0.9676731183574976, "3000–6000 Hz_10ms_modulation_mae_db": 0.38784963071889655, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8589620734448563, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10073680758321986, "3000–6000 Hz_40ms_envelope_correlation": 0.99051116870546, "3000–6000 Hz_40ms_modulation_mae_db": 0.5287978442843904, "mr_stft_db": 5.550583902218381, "band_mae_db": 0.30657627819237815, "rms_gain_error_db": -0.018787966600397735, "output_peak": 0.8935380111824561}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7656893760455356, "全帯域_10ms_rms_envelope_nrmse": 0.07559370973948581, "全帯域_10ms_envelope_correlation": 0.9948852721485361, "全帯域_10ms_modulation_mae_db": 0.377849161692079, "全帯域_40ms_log_rms_mae_db": 0.44965150664244224, "全帯域_40ms_rms_envelope_nrmse": 0.04634458565302193, "全帯域_40ms_envelope_correlation": 0.997952247244742, "全帯域_40ms_modulation_mae_db": 0.7279809930004579, "300–1000 Hz_10ms_log_rms_mae_db": 1.298923425295222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08532441913488761, "300–1000 Hz_10ms_envelope_correlation": 0.9942074684525642, "300–1000 Hz_10ms_modulation_mae_db": 1.005271296722219, "300–1000 Hz_40ms_log_rms_mae_db": 0.8471687107291455, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05205670729407247, "300–1000 Hz_40ms_envelope_correlation": 0.997760332108016, "300–1000 Hz_40ms_modulation_mae_db": 1.2352610753823006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3176160138850934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11328000374641786, "1000–3000 Hz_10ms_envelope_correlation": 0.9912084835024781, "1000–3000 Hz_10ms_modulation_mae_db": 0.6438213872133945, "1000–3000 Hz_40ms_log_rms_mae_db": 0.816023330086517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06944410549575183, "1000–3000 Hz_40ms_envelope_correlation": 0.9965452439927797, "1000–3000 Hz_40ms_modulation_mae_db": 0.8184700628548619, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4661826914638243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17313442358647066, "3000–6000 Hz_10ms_envelope_correlation": 0.9756192061656384, "3000–6000 Hz_10ms_modulation_mae_db": 0.31772299504633444, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8528745446622861, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0921938416002165, "3000–6000 Hz_40ms_envelope_correlation": 0.992539099684099, "3000–6000 Hz_40ms_modulation_mae_db": 0.1855101367451707, "mr_stft_db": 5.5384143790845615, "band_mae_db": 0.29061458401573376, "rms_gain_error_db": -0.03429469575133502, "output_peak": 0.8939386280745922}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.443585136553117, "spectral_convergence": 0.2642987846010118, "spectral_convergence_status": "ok", "amplitude_floor": 2.005860984254184e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.453436826947406, "spectral_convergence": 0.26418996078438434, "spectral_convergence_status": "ok", "amplitude_floor": 1.6673734947378395e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.724884788493518, "spectral_convergence": 0.27735183592589896, "spectral_convergence_status": "ok", "amplitude_floor": 1.455049241978443e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.2140486659240935e-05, "output_power": 4.4113164781533575e-05, "delta_db": 1.3751576779528558, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002187325069648938, "output_power": 0.0021464181801068008, "delta_db": -0.08198992432256802, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018573993391017246, "output_power": 0.001797624896172728, "delta_db": -0.14206212475630997, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.002022873257906857, "output_power": 0.0019628825863377717, "delta_db": -0.13074350920924022, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00023141111513331255, "output_power": 0.0002291727196787447, "delta_db": -0.04221296425530745, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.19984427950361e-05, "output_power": 4.33871147461302e-05, "delta_db": 0.14127582335783948, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3189070039756868, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10312405546448383, "pitch": null, "output_rms_dbfs": -22.0595842125217, "output_peak": 0.7551253835662786, "reference_rms_dbfs": -21.95646015705721, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5406355839980135}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003217977906320387, "log_rms_mae_db": 0.8085385706359092, "rms_envelope_nrmse": 0.07724769665265467, "envelope_correlation": 0.9947322837591313, "rms_gain_error_db": -0.10305817588845265, "reference_mean_power": 0.006378025340653872, "output_mean_power": 0.006228456316067608}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.934428693189062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0026409373563640536, "output_power": 0.002557351679877502, "reference_active": true, "delta_db": -0.13967644664753606}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022353254716419522, "output_power": 0.00019944004330567843, "reference_active": true, "delta_db": -0.4952840715525197}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.171045723198585e-05, "output_power": 2.8748110123896242e-05, "reference_active": true, "delta_db": -0.4259320433207506}], "active_band_count": 3, "active_band_mae_db": 0.35363085384026877}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00029448546471938644, "log_rms_mae_db": 0.45799751908102626, "rms_envelope_nrmse": 0.03770515226839409, "envelope_correlation": 0.998766364964389, "rms_gain_error_db": -0.10300496395866979, "reference_mean_power": 0.006407121641291136, "output_mean_power": 0.006256946951470594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7183045790225306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0025278777543949896, "output_power": 0.0024442316148043674, "reference_active": true, "delta_db": -0.14613710915435354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010431651037251879, "output_power": 9.215354909709601e-05, "reference_active": true, "delta_db": -0.5384098467890692}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4474914310652027e-06, "output_power": 1.8189751738212587e-06, "reference_active": true, "delta_db": -1.2889440830491707}], "active_band_count": 3, "active_band_mae_db": 0.6578303463308645}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002169046697405618, "log_rms_mae_db": 1.2912949969071668, "rms_envelope_nrmse": 0.07905545615143718, "envelope_correlation": 0.9951379706568557, "rms_gain_error_db": -0.13554566251270658, "reference_mean_power": 0.00266513165861851, "output_mean_power": 0.002583236104610984}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9887000298118345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014141653900070267, "output_power": 0.001351057157126036, "reference_active": true, "delta_db": -0.19826481755533767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010731466350302803, "output_power": 8.8191391364777e-05, "reference_active": true, "delta_db": -0.8523287376004457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0173327828420217e-05, "output_power": 1.7054283113585455e-05, "reference_active": true, "delta_db": -0.7294407780741331}], "active_band_count": 3, "active_band_mae_db": 0.5933447777433055}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020963310569053025, "log_rms_mae_db": 0.8330877035242945, "rms_envelope_nrmse": 0.04810534457856316, "envelope_correlation": 0.9982140433966434, "rms_gain_error_db": -0.13557092214196628, "reference_mean_power": 0.0026772178601939312, "output_mean_power": 0.0025949358222627835}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.883939001913369e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013521293519634116, "output_power": 0.0012891896603536477, "reference_active": true, "delta_db": -0.20701426841925064}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.792726117783695e-05, "output_power": 3.9021503176265876e-05, "reference_active": true, "delta_db": -0.8927861661155323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6850826685161704e-06, "output_power": 1.2956689178005157e-06, "reference_active": true, "delta_db": -1.141271713262709}], "active_band_count": 3, "active_band_mae_db": 0.7470240492658307}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002744641361828464, "log_rms_mae_db": 1.2942885000981794, "rms_envelope_nrmse": 0.11627098203470466, "envelope_correlation": 0.9906335851978246, "rms_gain_error_db": -0.1388793516559109, "reference_mean_power": 0.0019381811578355007, "output_mean_power": 0.0018771822199447259}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1196235236571854e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006735796402683569, "output_power": 0.0006608886172732295, "reference_active": true, "delta_db": -0.08260679709950562}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013823323993445422, "output_power": 0.00013269835246670182, "reference_active": true, "delta_db": -0.1774695634894335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8792319908524634e-05, "output_power": 1.2018615905693562e-05, "reference_active": true, "delta_db": -1.9412594089426227}], "active_band_count": 3, "active_band_mae_db": 0.7337785898438539}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002445180716567873, "log_rms_mae_db": 0.768382230825778, "rms_envelope_nrmse": 0.05764677558683337, "envelope_correlation": 0.9977143941833714, "rms_gain_error_db": -0.1386337054426111, "reference_mean_power": 0.0019467823981069799, "output_mean_power": 0.0018856194113970103}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0571732642454036e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006649824803295284, "output_power": 0.0006489455255756451, "reference_active": true, "delta_db": -0.10601961148622469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927173921912536e-05, "output_power": 6.615575582244477e-05, "reference_active": true, "delta_db": -0.1998845611950783}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7481239585012505e-06, "output_power": 1.0798106540848646e-06, "reference_active": true, "delta_db": -2.0922461679497166}], "active_band_count": 3, "active_band_mae_db": 0.7993834468770066}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.6882736857616036e-05, "log_rms_mae_db": 1.4874600967537508, "rms_envelope_nrmse": 0.20649353738399065, "envelope_correlation": 0.9655514681083132, "rms_gain_error_db": 0.19397716799998993, "reference_mean_power": 6.564824297759055e-05, "output_mean_power": 6.864688323987998e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.001422573299629e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.567613726314846e-05, "output_power": 1.7159809680436365e-05, "reference_active": true, "delta_db": 0.39273409191123265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.79643479102731e-06, "output_power": 6.538382585518353e-06, "reference_active": true, "delta_db": 0.5230937471811192}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.958485457614687e-07, "output_power": 7.683258705862356e-07, "reference_active": true, "delta_db": -1.1264783677707493}], "active_band_count": 3, "active_band_mae_db": 0.6807687356210338}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.766393520730197e-05, "log_rms_mae_db": 0.9016466573957704, "rms_envelope_nrmse": 0.11812955418883972, "envelope_correlation": 0.9878151795032716, "rms_gain_error_db": 0.19467807292321238, "reference_mean_power": 6.585532886735065e-05, "output_mean_power": 6.887454299109189e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6817222442955538e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5480772645031266e-05, "output_power": 1.7382953860782827e-05, "reference_active": true, "delta_db": 0.5033094491507797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.979992150286795e-06, "output_power": 3.159404656559505e-06, "reference_active": true, "delta_db": 0.2539013382063031}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.719812912358681e-08, "output_power": 7.504260461781895e-08, "reference_active": true, "delta_db": -0.6519926769068546}], "active_band_count": 3, "active_band_mae_db": 0.4697344880879791}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8689430652650397, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8ab4aa0d79c518c7caf26e12a1748db87a005db95d4183323c63b6a618fe3d7d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897800842734571, "standard_deviation": 0.0016159351158139988, "minimum": 0.9866146451573929, "maximum": 0.9912084835024781, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.31124819068613, "standard_deviation": 0.01417222297210411, "minimum": 1.2915441928014462, "maximum": 1.3297749231939444, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.653915185309095, "standard_deviation": 0.3840714328853651, "minimum": 0.15400693258190595, "maximum": 1.2939822330605184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12277679246705403, "standard_deviation": 0.010395265114402915, "minimum": 0.11297963685491626, "maximum": 0.13989813761279224, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962494943568776, "standard_deviation": 0.0011799641232510157, "minimum": 0.9945225681019041, "maximum": 0.9977143941833714, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8049276087425336, "standard_deviation": 0.020594233075868915, "minimum": 0.768382230825778, "maximum": 0.8372208566356726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8395200723506973, "standard_deviation": 0.48859828137365113, "minimum": 0.32160566888046443, "maximum": 1.6438714594557464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07398328030799098, "standard_deviation": 0.013262046027643281, "minimum": 0.05738029515787735, "maximum": 0.09000947304647108, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9702756913041363, "standard_deviation": 0.004580314890798735, "minimum": 0.9654745891695277, "maximum": 0.9761941755273467, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4557264508251109, "standard_deviation": 0.02551041024837038, "minimum": 1.4197188929980469, "maximum": 1.494425576517197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5227607634365428, "standard_deviation": 0.1502106370657163, "minimum": 0.31772299504633444, "maximum": 0.6833422709508586, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18834388362945712, "standard_deviation": 0.014842508838250962, "minimum": 0.16809582578720067, "maximum": 0.20649353738399065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9898698943987077, "standard_deviation": 0.0024316779454004915, "minimum": 0.9867305946954612, "maximum": 0.992539099684099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8830549642147065, "standard_deviation": 0.027625219235333294, "minimum": 0.8525512577070908, "maximum": 0.9198342006973972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5281258850689505, "standard_deviation": 0.17453361390560918, "minimum": 0.1855101367451707, "maximum": 0.7149419827335363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10512026573292485, "standard_deviation": 0.012693375747011659, "minimum": 0.09069376374809515, "maximum": 0.12124207191506788, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932866820599366, "standard_deviation": 0.001454238280697842, "minimum": 0.9905626583286827, "maximum": 0.9951379706568557, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.30485974967006, "standard_deviation": 0.02513328764931885, "minimum": 1.2861228852452684, "maximum": 1.3639010118660202, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.730452502688069, "standard_deviation": 0.26420499215724447, "minimum": 0.3231482409216702, "maximum": 1.0927775286706265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09196630065353548, "standard_deviation": 0.009627552150531063, "minimum": 0.07905545615143718, "maximum": 0.10915986288548277, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968564021392503, "standard_deviation": 0.0011709323611762007, "minimum": 0.9945667732482835, "maximum": 0.9982140433966434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8626769839094168, "standard_deviation": 0.026656245463476926, "minimum": 0.8330123565358617, "maximum": 0.9120073153611269, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9009540072830874, "standard_deviation": 0.3198317086930754, "minimum": 0.41739922448493516, "maximum": 1.365595799543729, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06151513066936454, "standard_deviation": 0.010842942611592906, "minimum": 0.04810534457856316, "maximum": 0.08103629430401482, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3330462820793306, "standard_deviation": 0.04322885169718204, "minimum": 0.29061458401573376, "maximum": 0.42763774267396854, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.546869507277259, "standard_deviation": 0.010568738091515411, "minimum": 5.5323454222879045, "maximum": 5.567519463682184, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8427123593150251, "standard_deviation": 0.05143686172486486, "minimum": 0.7551253835662786, "maximum": 0.89543745360491, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08309890861920569, "standard_deviation": 0.07294412279723664, "minimum": -0.17221612555355592, "maximum": 0.04460837429221104, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944849086322842, "standard_deviation": 0.000655706620286789, "minimum": 0.993452803497452, "maximum": 0.9953247234590713, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7916262535847165, "standard_deviation": 0.02847004544943592, "minimum": 0.7538965715956056, "maximum": 0.8309894812591883, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4233735414081928, "standard_deviation": 0.21095440222689718, "minimum": 0.22180915937718526, "maximum": 0.829166240432173, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07905582570253811, "standard_deviation": 0.00481995126721422, "minimum": 0.07308596388358563, "maximum": 0.08718913210817528, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998110880339155, "standard_deviation": 0.0005559129939871535, "minimum": 0.9974176462904925, "maximum": 0.9987887739338436, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.446149233538086, "standard_deviation": 0.016500402439279322, "minimum": 0.4193088377067705, "maximum": 0.47051820507958786, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.65731586073039, "standard_deviation": 0.3907528781152976, "minimum": 0.1713771134925212, "maximum": 1.2695665394876725, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04552610955287248, "standard_deviation": 0.006610702920150181, "minimum": 0.03770515226839409, "maximum": 0.05532057532131933, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_5683-32879-0019", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.255751848220825, "predicted_mos": 2.255751848220825, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3a1b31052d580bd4cad7721166b8f027bd3f5ada9a36a124680c25d3fae48a29", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.12491670018062, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7846872711277941, "全帯域_10ms_rms_envelope_nrmse": 0.07402783459926014, "全帯域_10ms_envelope_correlation": 0.9952781617949155, "全帯域_10ms_modulation_mae_db": 0.38185211037230454, "全帯域_40ms_log_rms_mae_db": 0.4190543228343498, "全帯域_40ms_rms_envelope_nrmse": 0.0381408376836781, "全帯域_40ms_envelope_correlation": 0.9988560289523021, "全帯域_40ms_modulation_mae_db": 0.9626111204310334, "300–1000 Hz_10ms_log_rms_mae_db": 1.3215970561784163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10420216634496271, "300–1000 Hz_10ms_envelope_correlation": 0.9916411990732316, "300–1000 Hz_10ms_modulation_mae_db": 0.8085953787874782, "300–1000 Hz_40ms_log_rms_mae_db": 0.8536989044772133, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07664150943560077, "300–1000 Hz_40ms_envelope_correlation": 0.9954257355789884, "300–1000 Hz_40ms_modulation_mae_db": 1.033546129113686, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2950250768025502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11112813399592147, "1000–3000 Hz_10ms_envelope_correlation": 0.9913782188462534, "1000–3000 Hz_10ms_modulation_mae_db": 0.8014093067382277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7689901420872768, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06295115637278549, "1000–3000 Hz_40ms_envelope_correlation": 0.9970852931526135, "1000–3000 Hz_40ms_modulation_mae_db": 1.193709966415932, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4002533765082361, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16270658801322085, "3000–6000 Hz_10ms_envelope_correlation": 0.9776945898892437, "3000–6000 Hz_10ms_modulation_mae_db": 0.400161461442295, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8563473679788309, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09020711340118989, "3000–6000 Hz_40ms_envelope_correlation": 0.9926047096578412, "3000–6000 Hz_40ms_modulation_mae_db": 0.46751978151314627, "mr_stft_db": 5.524449593466066, "band_mae_db": 0.3996422742730639, "rms_gain_error_db": -0.1414565961901103, "output_peak": 0.8378483724101903}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7904205374592503, "全帯域_10ms_rms_envelope_nrmse": 0.08990190391791647, "全帯域_10ms_envelope_correlation": 0.9927976551791929, "全帯域_10ms_modulation_mae_db": 0.45863562679608033, "全帯域_40ms_log_rms_mae_db": 0.44418053129566515, "全帯域_40ms_rms_envelope_nrmse": 0.048959930820474055, "全帯域_40ms_envelope_correlation": 0.9977956171050925, "全帯域_40ms_modulation_mae_db": 0.5890693321249452, "300–1000 Hz_10ms_log_rms_mae_db": 1.286946288674884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11236966261576968, "300–1000 Hz_10ms_envelope_correlation": 0.9900252426781332, "300–1000 Hz_10ms_modulation_mae_db": 0.5079809789261839, "300–1000 Hz_40ms_log_rms_mae_db": 0.834733374371256, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08733680170190114, "300–1000 Hz_40ms_envelope_correlation": 0.9936990742782452, "300–1000 Hz_40ms_modulation_mae_db": 0.5570135734757148, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2940921620677748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14093213939641075, "1000–3000 Hz_10ms_envelope_correlation": 0.986168372192215, "1000–3000 Hz_10ms_modulation_mae_db": 0.7544344556941511, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8317811982525319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08427932420605198, "1000–3000 Hz_40ms_envelope_correlation": 0.9947534597136388, "1000–3000 Hz_40ms_modulation_mae_db": 0.8290257411195955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4400911830700462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1775068589102052, "3000–6000 Hz_10ms_envelope_correlation": 0.9745088575254325, "3000–6000 Hz_10ms_modulation_mae_db": 0.35090280610819713, "3000–6000 Hz_40ms_log_rms_mae_db": 0.871831126918333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09819439150498485, "3000–6000 Hz_40ms_envelope_correlation": 0.9916106986027878, "3000–6000 Hz_40ms_modulation_mae_db": 0.33306644005268243, "mr_stft_db": 5.532218977195804, "band_mae_db": 0.2892172519392558, "rms_gain_error_db": -0.09845382157844396, "output_peak": 0.8061637256706705}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7927459334817472, "全帯域_10ms_rms_envelope_nrmse": 0.08051514202088211, "全帯域_10ms_envelope_correlation": 0.9941912008030748, "全帯域_10ms_modulation_mae_db": 0.16798626042531947, "全帯域_40ms_log_rms_mae_db": 0.42739731760578825, "全帯域_40ms_rms_envelope_nrmse": 0.039640729734843186, "全帯域_40ms_envelope_correlation": 0.998509477965353, "全帯域_40ms_modulation_mae_db": 0.42594504107907616, "300–1000 Hz_10ms_log_rms_mae_db": 1.2122901904692949, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08470302904546473, "300–1000 Hz_10ms_envelope_correlation": 0.9942672340960276, "300–1000 Hz_10ms_modulation_mae_db": 0.566098666817378, "300–1000 Hz_40ms_log_rms_mae_db": 0.7608176851730333, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058002709396248055, "300–1000 Hz_40ms_envelope_correlation": 0.9971479305362613, "300–1000 Hz_40ms_modulation_mae_db": 0.7580900250647359, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2600221984405897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11969326947068243, "1000–3000 Hz_10ms_envelope_correlation": 0.9899976786050558, "1000–3000 Hz_10ms_modulation_mae_db": 0.3869706242358973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7391134659072727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06063828831692496, "1000–3000 Hz_40ms_envelope_correlation": 0.9973280781297469, "1000–3000 Hz_40ms_modulation_mae_db": 0.5781793275820415, "3000–6000 Hz_10ms_log_rms_mae_db": 1.405977956580322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19821141539515807, "3000–6000 Hz_10ms_envelope_correlation": 0.9676709122220496, "3000–6000 Hz_10ms_modulation_mae_db": 0.2516901806880618, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8675476360112304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11625010127801474, "3000–6000 Hz_40ms_envelope_correlation": 0.9877611578407183, "3000–6000 Hz_40ms_modulation_mae_db": 0.21924692846836114, "mr_stft_db": 5.530043467933569, "band_mae_db": 0.2908889077538264, "rms_gain_error_db": -0.04734983755161345, "output_peak": 0.7919925760234228}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7827142142692508, "全帯域_10ms_rms_envelope_nrmse": 0.09665487681148728, "全帯域_10ms_envelope_correlation": 0.9916188571486961, "全帯域_10ms_modulation_mae_db": 0.45303849647522115, "全帯域_40ms_log_rms_mae_db": 0.4477227264462132, "全帯域_40ms_rms_envelope_nrmse": 0.04669134599337576, "全帯域_40ms_envelope_correlation": 0.9979078919649921, "全帯域_40ms_modulation_mae_db": 0.2941115892284928, "300–1000 Hz_10ms_log_rms_mae_db": 1.2926510212794158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1032306027565018, "300–1000 Hz_10ms_envelope_correlation": 0.9915128836103136, "300–1000 Hz_10ms_modulation_mae_db": 0.5857898304617954, "300–1000 Hz_40ms_log_rms_mae_db": 0.8119794072103237, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05079566929618337, "300–1000 Hz_40ms_envelope_correlation": 0.9978116610096261, "300–1000 Hz_40ms_modulation_mae_db": 0.8237104244670816, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3301771417799313, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13373273869119312, "1000–3000 Hz_10ms_envelope_correlation": 0.9874883307857499, "1000–3000 Hz_10ms_modulation_mae_db": 0.46489833044251466, "1000–3000 Hz_40ms_log_rms_mae_db": 0.746821534333913, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06408393063615386, "1000–3000 Hz_40ms_envelope_correlation": 0.9970045089759796, "1000–3000 Hz_40ms_modulation_mae_db": 0.3248302209037332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.400994488445746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19755236856345532, "3000–6000 Hz_10ms_envelope_correlation": 0.967319648302622, "3000–6000 Hz_10ms_modulation_mae_db": 0.3971014929018229, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8734261268218467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10251223131830839, "3000–6000 Hz_40ms_envelope_correlation": 0.990197464067393, "3000–6000 Hz_40ms_modulation_mae_db": 0.550848200364399, "mr_stft_db": 5.517389149955467, "band_mae_db": 0.3029053298441763, "rms_gain_error_db": -0.028746493804844994, "output_peak": 0.870577113320157}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7964877895590435, "全帯域_10ms_rms_envelope_nrmse": 0.07504624512111305, "全帯域_10ms_envelope_correlation": 0.9949691369053338, "全帯域_10ms_modulation_mae_db": 0.2600336836083526, "全帯域_40ms_log_rms_mae_db": 0.4266983983202241, "全帯域_40ms_rms_envelope_nrmse": 0.037691045294370194, "全帯域_40ms_envelope_correlation": 0.9986758033809879, "全帯域_40ms_modulation_mae_db": 0.209817148263291, "300–1000 Hz_10ms_log_rms_mae_db": 1.2485346196513756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0931273264655504, "300–1000 Hz_10ms_envelope_correlation": 0.993081783698544, "300–1000 Hz_10ms_modulation_mae_db": 0.3713449248427749, "300–1000 Hz_40ms_log_rms_mae_db": 0.7814959138992621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05681818242602515, "300–1000 Hz_40ms_envelope_correlation": 0.9973144801535633, "300–1000 Hz_40ms_modulation_mae_db": 0.4334843684676919, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3242327361228583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11327130777192729, "1000–3000 Hz_10ms_envelope_correlation": 0.9910415035639628, "1000–3000 Hz_10ms_modulation_mae_db": 0.18506001583568044, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7687370331512196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05763658414279015, "1000–3000 Hz_40ms_envelope_correlation": 0.9975554244820289, "1000–3000 Hz_40ms_modulation_mae_db": 0.1802192177747349, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4411000220571537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20304121057351499, "3000–6000 Hz_10ms_envelope_correlation": 0.965023140862034, "3000–6000 Hz_10ms_modulation_mae_db": 0.35330713392016716, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8790559604691879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10538576522462051, "3000–6000 Hz_40ms_envelope_correlation": 0.9896213659982427, "3000–6000 Hz_40ms_modulation_mae_db": 0.45420026660796103, "mr_stft_db": 5.537199350386953, "band_mae_db": 0.3285378587577442, "rms_gain_error_db": -0.06304578505003544, "output_peak": 0.7809266152228085}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8124807977063135, "全帯域_10ms_rms_envelope_nrmse": 0.09100502769331135, "全帯域_10ms_envelope_correlation": 0.9926177469239421, "全帯域_10ms_modulation_mae_db": 0.5170529761284011, "全帯域_40ms_log_rms_mae_db": 0.4403760957560337, "全帯域_40ms_rms_envelope_nrmse": 0.0449323614148283, "全帯域_40ms_envelope_correlation": 0.9981570583372456, "全帯域_40ms_modulation_mae_db": 0.663964454725889, "300–1000 Hz_10ms_log_rms_mae_db": 1.2835873053450728, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09525955453988948, "300–1000 Hz_10ms_envelope_correlation": 0.9927502682254805, "300–1000 Hz_10ms_modulation_mae_db": 0.9253343942430218, "300–1000 Hz_40ms_log_rms_mae_db": 0.8097123262554127, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04457663814281864, "300–1000 Hz_40ms_envelope_correlation": 0.9983638442174506, "300–1000 Hz_40ms_modulation_mae_db": 1.0441136584137523, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2934033971107293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13857533903660052, "1000–3000 Hz_10ms_envelope_correlation": 0.9865829403070371, "1000–3000 Hz_10ms_modulation_mae_db": 0.6556190848807358, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7492476425951431, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06907678418737659, "1000–3000 Hz_40ms_envelope_correlation": 0.9965933077975598, "1000–3000 Hz_40ms_modulation_mae_db": 0.7541271169004524, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4552165352036044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23479948360460828, "3000–6000 Hz_10ms_envelope_correlation": 0.9549833375747727, "3000–6000 Hz_10ms_modulation_mae_db": 0.4029250489655967, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9062254175437122, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14005764918184016, "3000–6000 Hz_40ms_envelope_correlation": 0.9823706069086897, "3000–6000 Hz_40ms_modulation_mae_db": 0.41983658040561256, "mr_stft_db": 5.537672760477762, "band_mae_db": 0.33908260336698764, "rms_gain_error_db": -0.08367646500102704, "output_peak": 0.810419677115868}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7846685827204025, "全帯域_10ms_rms_envelope_nrmse": 0.07156620662502414, "全帯域_10ms_envelope_correlation": 0.9954141133825596, "全帯域_10ms_modulation_mae_db": 0.5772410744150213, "全帯域_40ms_log_rms_mae_db": 0.40783809687853095, "全帯域_40ms_rms_envelope_nrmse": 0.03611078611068796, "全帯域_40ms_envelope_correlation": 0.9987556427032384, "全帯域_40ms_modulation_mae_db": 0.35553216100754126, "300–1000 Hz_10ms_log_rms_mae_db": 1.2288373561489507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06654237035064613, "300–1000 Hz_10ms_envelope_correlation": 0.9965176912168555, "300–1000 Hz_10ms_modulation_mae_db": 0.3568019708424252, "300–1000 Hz_40ms_log_rms_mae_db": 0.7690162556650711, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03857706997139341, "300–1000 Hz_40ms_envelope_correlation": 0.9988053645102012, "300–1000 Hz_40ms_modulation_mae_db": 0.535219487361937, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3164117412468639, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1082757586870279, "1000–3000 Hz_10ms_envelope_correlation": 0.9918398912016134, "1000–3000 Hz_10ms_modulation_mae_db": 0.28419895018003255, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7726886050569144, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05472405805479809, "1000–3000 Hz_40ms_envelope_correlation": 0.9977845185240292, "1000–3000 Hz_40ms_modulation_mae_db": 0.25200669919942836, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4553074790845213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19802794743924634, "3000–6000 Hz_10ms_envelope_correlation": 0.9696812896408389, "3000–6000 Hz_10ms_modulation_mae_db": 0.6035888836140562, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9083130191577019, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13401587709231905, "3000–6000 Hz_40ms_envelope_correlation": 0.9854117790543868, "3000–6000 Hz_40ms_modulation_mae_db": 0.48312498163298856, "mr_stft_db": 5.523403664662793, "band_mae_db": 0.38108545441916003, "rms_gain_error_db": -0.02731507403503118, "output_peak": 0.8575237386642185}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7624213183485976, "全帯域_10ms_rms_envelope_nrmse": 0.07845665715364204, "全帯域_10ms_envelope_correlation": 0.9945193860198667, "全帯域_10ms_modulation_mae_db": 0.25750499851395964, "全帯域_40ms_log_rms_mae_db": 0.4271677420569835, "全帯域_40ms_rms_envelope_nrmse": 0.040896080831450785, "全帯域_40ms_envelope_correlation": 0.9984519292903966, "全帯域_40ms_modulation_mae_db": 0.4050710052712574, "300–1000 Hz_10ms_log_rms_mae_db": 1.250641865133888, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07385289775905524, "300–1000 Hz_10ms_envelope_correlation": 0.9956831927380209, "300–1000 Hz_10ms_modulation_mae_db": 0.5568575512563544, "300–1000 Hz_40ms_log_rms_mae_db": 0.7452877290590666, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03998668417787243, "300–1000 Hz_40ms_envelope_correlation": 0.9987096463456714, "300–1000 Hz_40ms_modulation_mae_db": 0.6587267209640767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.305294710864112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12123827740519288, "1000–3000 Hz_10ms_envelope_correlation": 0.9897338030091205, "1000–3000 Hz_10ms_modulation_mae_db": 0.2316808169187573, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7707677683204494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059956227802707096, "1000–3000 Hz_40ms_envelope_correlation": 0.9973622683333996, "1000–3000 Hz_40ms_modulation_mae_db": 0.3610997359202994, "3000–6000 Hz_10ms_log_rms_mae_db": 1.440616490404251, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21121806683910613, "3000–6000 Hz_10ms_envelope_correlation": 0.9625361693203983, "3000–6000 Hz_10ms_modulation_mae_db": 0.5498808162984455, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8889727280025994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11823606084783962, "3000–6000 Hz_40ms_envelope_correlation": 0.9869173508848595, "3000–6000 Hz_40ms_modulation_mae_db": 0.5806166111249224, "mr_stft_db": 5.530019080962302, "band_mae_db": 0.25679939184691875, "rms_gain_error_db": -0.06271000019221529, "output_peak": 0.8186615660490968}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.449484788466413, "spectral_convergence": 0.25347778457014125, "spectral_convergence_status": "ok", "amplitude_floor": 2.005860984254184e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.426621191546312, "spectral_convergence": 0.2741925765993926, "spectral_convergence_status": "ok", "amplitude_floor": 1.6673734947378395e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6972428003854745, "spectral_convergence": 0.26745873846164564, "spectral_convergence_status": "ok", "amplitude_floor": 1.455049241978443e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.2140486659240935e-05, "output_power": 4.3730951822021404e-05, "delta_db": 1.3373648164254706, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002187325069648938, "output_power": 0.002152208732036492, "delta_db": -0.07028941472841646, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018573993391017246, "output_power": 0.0017298215082039226, "delta_db": -0.3090399395833244, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.002022873257906857, "output_power": 0.0019859052632948187, "delta_db": -0.08010146290490537, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00023141111513331255, "output_power": 0.00021287335263973792, "delta_db": -0.36262915024629366, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.19984427950361e-05, "output_power": 4.4368635651332344e-05, "delta_db": 0.23842886174997335, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3996422742730639, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1414565961901103, "pitch": null, "output_rms_dbfs": -22.097916753247322, "output_peak": 0.8378483724101903, "reference_rms_dbfs": -21.95646015705721, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.524449593466066}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003217977906320387, "log_rms_mae_db": 0.7846872711277941, "rms_envelope_nrmse": 0.07402783459926014, "envelope_correlation": 0.9952781617949155, "rms_gain_error_db": -0.14143888847900954, "reference_mean_power": 0.006378025340653872, "output_mean_power": 0.00617365493253834}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.934428693189062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0026409373563640536, "output_power": 0.002552549037631357, "reference_active": true, "delta_db": -0.14784005590540647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022353254716419522, "output_power": 0.00021622591410928182, "reference_active": true, "delta_db": -0.14433025159795676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.171045723198585e-05, "output_power": 2.60533696751272e-05, "reference_active": true, "delta_db": -0.8533860236135504}], "active_band_count": 3, "active_band_mae_db": 0.38185211037230454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00029448546471938644, "log_rms_mae_db": 0.4190543228343498, "rms_envelope_nrmse": 0.0381408376836781, "envelope_correlation": 0.9988560289523021, "rms_gain_error_db": -0.14143697560809954, "reference_mean_power": 0.006407121641291136, "output_mean_power": 0.006201821635037016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7183045790225306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0025278777543949896, "output_power": 0.0024473131509581674, "reference_active": true, "delta_db": -0.14066524131502198}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010431651037251879, "output_power": 9.707766423325219e-05, "reference_active": true, "delta_db": -0.3123373216960158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4474914310652027e-06, "output_power": 1.3971341240762938e-06, "reference_active": true, "delta_db": -2.434830798282062}], "active_band_count": 3, "active_band_mae_db": 0.9626111204310334}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002169046697405618, "log_rms_mae_db": 1.3215970561784163, "rms_envelope_nrmse": 0.10420216634496271, "envelope_correlation": 0.9916411990732316, "rms_gain_error_db": -0.21859779780811012, "reference_mean_power": 0.00266513165861851, "output_mean_power": 0.002534305046830357}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9887000298118345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014141653900070267, "output_power": 0.0013460676996961162, "reference_active": true, "delta_db": -0.2143330117723956}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010731466350302803, "output_power": 0.00010520250209329569, "reference_active": true, "delta_db": -0.0863299910656079}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0173327828420217e-05, "output_power": 1.2367024830591204e-05, "reference_active": true, "delta_db": -2.1251231335244314}], "active_band_count": 3, "active_band_mae_db": 0.8085953787874782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020963310569053025, "log_rms_mae_db": 0.8536989044772133, "rms_envelope_nrmse": 0.07664150943560077, "envelope_correlation": 0.9954257355789884, "rms_gain_error_db": -0.21879904810272, "reference_mean_power": 0.0026772178601939312, "output_mean_power": 0.0025456799895832018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.883939001913369e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013521293519634116, "output_power": 0.0012870196648641506, "reference_active": true, "delta_db": -0.21433057832079228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.792726117783695e-05, "output_power": 4.553933272028248e-05, "reference_active": true, "delta_db": -0.22195949085955238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6850826685161704e-06, "output_power": 9.124023180579027e-07, "reference_active": true, "delta_db": -2.6643483181607133}], "active_band_count": 3, "active_band_mae_db": 1.033546129113686}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002744641361828464, "log_rms_mae_db": 1.2950250768025502, "rms_envelope_nrmse": 0.11112813399592147, "envelope_correlation": 0.9913782188462534, "rms_gain_error_db": -0.06370554857709511, "reference_mean_power": 0.0019381811578355007, "output_mean_power": 0.0019099579787188146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1196235236571854e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006735796402683569, "output_power": 0.0006681715436060414, "reference_active": true, "delta_db": -0.035009758670289666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013823323993445422, "output_power": 0.00012754606440547022, "reference_active": true, "delta_db": -0.3494542475930393}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8792319908524634e-05, "output_power": 1.1803315220516471e-05, "reference_active": true, "delta_db": -2.0197639139513544}], "active_band_count": 3, "active_band_mae_db": 0.8014093067382277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002445180716567873, "log_rms_mae_db": 0.7689901420872768, "rms_envelope_nrmse": 0.06295115637278549, "envelope_correlation": 0.9970852931526135, "rms_gain_error_db": -0.06352289617215881, "reference_mean_power": 0.0019467823981069799, "output_mean_power": 0.0019185146562513689}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0571732642454036e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006649824803295284, "output_power": 0.0006573775184850998, "reference_active": true, "delta_db": -0.0499535584963465}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927173921912536e-05, "output_power": 6.22384995150975e-05, "reference_active": true, "delta_db": -0.46496977827106073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7481239585012505e-06, "output_power": 8.628823162369535e-07, "reference_active": true, "delta_db": -3.066206562480389}], "active_band_count": 3, "active_band_mae_db": 1.193709966415932}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.6882736857616036e-05, "log_rms_mae_db": 1.4002533765082361, "rms_envelope_nrmse": 0.16270658801322085, "envelope_correlation": 0.9776945898892437, "rms_gain_error_db": -0.14438280080335492, "reference_mean_power": 6.564824297759055e-05, "output_mean_power": 6.350162324148823e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.001422573299629e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.567613726314846e-05, "output_power": 1.5863351917444137e-05, "reference_active": true, "delta_db": 0.05155901275412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.79643479102731e-06, "output_power": 5.379481338114385e-06, "reference_active": true, "delta_db": -0.3242054923650757}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.958485457614687e-07, "output_power": 8.23609420723297e-07, "reference_active": true, "delta_db": -0.8247198792076891}], "active_band_count": 3, "active_band_mae_db": 0.400161461442295}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.766393520730197e-05, "log_rms_mae_db": 0.8563473679788309, "rms_envelope_nrmse": 0.09020711340118989, "envelope_correlation": 0.9926047096578412, "rms_gain_error_db": -0.14435380687025176, "reference_mean_power": 6.585532886735065e-05, "output_mean_power": 6.370236295001463e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6817222442955538e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5480772645031266e-05, "output_power": 1.579596241720997e-05, "reference_active": true, "delta_db": 0.08753459247713585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.979992150286795e-06, "output_power": 2.6419817135790266e-06, "reference_active": true, "delta_db": -0.5228531275700807}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.719812912358681e-08, "output_power": 7.265909840559029e-08, "reference_active": true, "delta_db": -0.7921716244922223}], "active_band_count": 3, "active_band_mae_db": 0.46751978151314627}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8603795286989049, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b049839720e9e33d64793e6f9d798b148c5377e483cdeb479d9e877d8a5bb3a2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9892788423138761, "standard_deviation": 0.0022343387101251707, "minimum": 0.986168372192215, "maximum": 0.9918398912016134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3023323955544264, "standard_deviation": 0.022210474506967413, "minimum": 1.2600221984405897, "maximum": 1.3301771417799313, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4705339481157496, "standard_deviation": 0.24042106338443156, "minimum": 0.18506001583568044, "maximum": 0.8014093067382277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12335587055686956, "standard_deviation": 0.012788172039871095, "minimum": 0.1082757586870279, "maximum": 0.14093213939641075, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969333573886245, "standard_deviation": 0.0009522114434825125, "minimum": 0.9947534597136388, "maximum": 0.9977845185240292, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7685184237130901, "standard_deviation": 0.02861485246593915, "minimum": 0.7391134659072727, "maximum": 0.8317811982525319, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5591497532270271, "standard_deviation": 0.34763130488811067, "minimum": 0.1802192177747349, "maximum": 1.193709966415932, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06416829421494853, "standard_deviation": 0.009197007440850247, "minimum": 0.05472405805479809, "maximum": 0.08427932420605198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.967427243167174, "standard_deviation": 0.0070231499349315205, "minimum": 0.9549833375747727, "maximum": 0.9776945898892437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.429944691419235, "standard_deviation": 0.023656197879030443, "minimum": 1.4002533765082361, "maximum": 1.4553074790845213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4136947279923303, "standard_deviation": 0.11276769550529192, "minimum": 0.2516901806880618, "maximum": 0.6035888836140562, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19788299241731439, "standard_deviation": 0.021447820985580655, "minimum": 0.16270658801322085, "maximum": 0.23479948360460828, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.988311891626865, "standard_deviation": 0.003391967176581226, "minimum": 0.9823706069086897, "maximum": 0.9926047096578412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8814649228629303, "standard_deviation": 0.018431678134902842, "minimum": 0.8563473679788309, "maximum": 0.9083130191577019, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4385574737712592, "standard_deviation": 0.1167633296785985, "minimum": 0.21924692846836114, "maximum": 0.5806166111249224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11310739873113965, "standard_deviation": 0.01739606798912648, "minimum": 0.09020711340118989, "maximum": 0.14005764918184016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931849369170758, "standard_deviation": 0.002200066053744683, "minimum": 0.9900252426781332, "maximum": 0.9965176912168555, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.265635712860162, "standard_deviation": 0.03655724634762109, "minimum": 1.2122901904692949, "maximum": 1.3215970561784163, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5848504620221765, "standard_deviation": 0.1966239548462145, "minimum": 0.3568019708424252, "maximum": 0.9253343942430218, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09166095123473003, "standard_deviation": 0.015727920566030982, "minimum": 0.06654237035064613, "maximum": 0.11236966261576968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997159717078751, "standard_deviation": 0.001772497036966031, "minimum": 0.9936990742782452, "maximum": 0.9988053645102012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7958426995138299, "standard_deviation": 0.03783541413460591, "minimum": 0.7452877290590666, "maximum": 0.8536989044772133, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7304880484160845, "standard_deviation": 0.22716940892099044, "minimum": 0.4334843684676919, "maximum": 1.0441136584137523, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056591908068505364, "standard_deviation": 0.017433578058265974, "minimum": 0.03857706997139341, "maximum": 0.08733680170190114, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.32351988402514165, "standard_deviation": 0.04854133230939648, "minimum": 0.25679939184691875, "maximum": 0.3996422742730639, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5290495056300895, "standard_deviation": 0.00698726073164342, "minimum": 5.517389149955467, "maximum": 5.537672760477762, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8217641730595541, "standard_deviation": 0.03129560209913752, "minimum": 0.7809266152228085, "maximum": 0.870577113320157, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0690942591754152, "standard_deviation": 0.03823711287035681, "minimum": -0.1414565961901103, "maximum": -0.02731507403503118, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939257822696976, "standard_deviation": 0.0014071750779235384, "minimum": 0.9916188571486961, "maximum": 0.9954141133825596, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.78832830558405, "standard_deviation": 0.014158054936907515, "minimum": 0.7624213183485976, "maximum": 0.8124807977063135, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3841681533418325, "standard_deviation": 0.14313677474543943, "minimum": 0.16798626042531947, "maximum": 0.5772410744150213, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08214673674282957, "standard_deviation": 0.009209141609607736, "minimum": 0.07156620662502414, "maximum": 0.09665487681148728, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983886812124511, "standard_deviation": 0.0003947220455013114, "minimum": 0.9977956171050925, "maximum": 0.9988560289523021, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.43005440389922356, "standard_deviation": 0.013413618869917527, "minimum": 0.40783809687853095, "maximum": 0.4477227264462132, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.48826523151644075, "standard_deviation": 0.24192319659951766, "minimum": 0.209817148263291, "maximum": 0.9626111204310334, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041632889735463544, "standard_deviation": 0.004673600730846685, "minimum": 0.03611078611068796, "maximum": 0.048959930820474055, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_61-70968-0042", "language": "en", "style": "normal", "speaker": "librispeech_61", "source_utterance_id": "61-70968-0042", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 44560, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_61-70968-0042.wav", "source_file_sha256": "29ed325d24673e45ebdd02201503a449d9a648e3bad5be2867769f84d9f2a081", "decoded_pcm_sha256": "4d1c295afd61c34ca0f59a70b2baace0777ff6e94127ed717b67dc664c9d9ee4", "original_source_sha256": "9aa0aa0746140383c3458b53991ccd9ebfc2c0ded6c6817ec0748aa70e6b8088", "corpus_entry_id": "librispeech_61-70968-0042"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.2773261070251465, "predicted_mos": 4.2773261070251465, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "f1b413a3333e8660deda8c827a58772b2c14c5003bd4ad6cf9e98a5a5592cd40", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.048219653999695, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_61-70968-0042", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f1b413a3333e8660deda8c827a58772b2c14c5003bd4ad6cf9e98a5a5592cd40", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "961c69b03b318b9af140f68a462f11556afbc167c633fa7b9f84f744bc866984"}, "rows": [{"case": "librispeech_61-70968-0042", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.0413060188293457, "predicted_mos": 2.0413060188293457, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "239110f21915a7debe6ed6c0c4824d4c8b1af4141295932183a5c16d7c6b8b58", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1827971499988053, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_61-70968-0042", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "239110f21915a7debe6ed6c0c4824d4c8b1af4141295932183a5c16d7c6b8b58", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1659703270163158, "全帯域_10ms_rms_envelope_nrmse": 0.09254188985824197, "全帯域_10ms_envelope_correlation": 0.9882836817540562, "全帯域_10ms_modulation_mae_db": 0.41296364450044964, "全帯域_40ms_log_rms_mae_db": 0.5352862920441215, "全帯域_40ms_rms_envelope_nrmse": 0.04329474229913915, "全帯域_40ms_envelope_correlation": 0.9973887157519706, "全帯域_40ms_modulation_mae_db": 0.44380815105646493, "300–1000 Hz_10ms_log_rms_mae_db": 1.7334352303183933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1250183555394948, "300–1000 Hz_10ms_envelope_correlation": 0.9858361627762572, "300–1000 Hz_10ms_modulation_mae_db": 0.686175841983828, "300–1000 Hz_40ms_log_rms_mae_db": 1.0518831042623829, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0657059426796576, "300–1000 Hz_40ms_envelope_correlation": 0.9963347345856329, "300–1000 Hz_40ms_modulation_mae_db": 0.7356672428574097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8813377322404599, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.222064384752126, "1000–3000 Hz_10ms_envelope_correlation": 0.9470903458440432, "1000–3000 Hz_10ms_modulation_mae_db": 0.4149706960313911, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0425292803167745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1288290471726013, "1000–3000 Hz_40ms_envelope_correlation": 0.9794647467225637, "1000–3000 Hz_40ms_modulation_mae_db": 0.42793358639081075, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0239388902969515, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19558540184862278, "3000–6000 Hz_10ms_envelope_correlation": 0.9775401742900329, "3000–6000 Hz_10ms_modulation_mae_db": 1.7843128390052663, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4320382825796782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13074572453181418, "3000–6000 Hz_40ms_envelope_correlation": 0.9912710530788151, "3000–6000 Hz_40ms_modulation_mae_db": 2.2997825075867167, "mr_stft_db": 5.736429058894196, "band_mae_db": 0.26455144373922285, "rms_gain_error_db": -0.16082654625159654, "output_peak": 0.3477708812056554}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0860967208714358, "全帯域_10ms_rms_envelope_nrmse": 0.09362406305050688, "全帯域_10ms_envelope_correlation": 0.9877728255076762, "全帯域_10ms_modulation_mae_db": 0.4087957521001571, "全帯域_40ms_log_rms_mae_db": 0.5302588513734029, "全帯域_40ms_rms_envelope_nrmse": 0.04870629271541662, "全帯域_40ms_envelope_correlation": 0.9960834654305186, "全帯域_40ms_modulation_mae_db": 0.2091570387024866, "300–1000 Hz_10ms_log_rms_mae_db": 1.6440229464978025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11230970274859209, "300–1000 Hz_10ms_envelope_correlation": 0.9881585556421306, "300–1000 Hz_10ms_modulation_mae_db": 0.8277021171469281, "300–1000 Hz_40ms_log_rms_mae_db": 0.9521883892342207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06501200912746032, "300–1000 Hz_40ms_envelope_correlation": 0.995556541097638, "300–1000 Hz_40ms_modulation_mae_db": 0.49639774488607474, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9270303341873272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.216316590996922, "1000–3000 Hz_10ms_envelope_correlation": 0.9498471429190936, "1000–3000 Hz_10ms_modulation_mae_db": 0.6149744407859518, "1000–3000 Hz_40ms_log_rms_mae_db": 1.030239192103113, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11623032334374951, "1000–3000 Hz_40ms_envelope_correlation": 0.9842108268686779, "1000–3000 Hz_40ms_modulation_mae_db": 0.7572624645089227, "3000–6000 Hz_10ms_log_rms_mae_db": 2.050787753340365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22178190121296545, "3000–6000 Hz_10ms_envelope_correlation": 0.9706229983789092, "3000–6000 Hz_10ms_modulation_mae_db": 1.0110097180586974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4693179656234219, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14191982541986867, "3000–6000 Hz_40ms_envelope_correlation": 0.9898857560336737, "3000–6000 Hz_40ms_modulation_mae_db": 1.3656985219410842, "mr_stft_db": 5.743704266736125, "band_mae_db": 0.27453609258850364, "rms_gain_error_db": -0.03971593771059505, "output_peak": 0.3629139983898995}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2379283858478607, "全帯域_10ms_rms_envelope_nrmse": 0.09454035335788491, "全帯域_10ms_envelope_correlation": 0.9875038690534338, "全帯域_10ms_modulation_mae_db": 0.2615631823540772, "全帯域_40ms_log_rms_mae_db": 0.6380161702467229, "全帯域_40ms_rms_envelope_nrmse": 0.052728757116868974, "全帯域_40ms_envelope_correlation": 0.9954785428380535, "全帯域_40ms_modulation_mae_db": 0.19128467690441978, "300–1000 Hz_10ms_log_rms_mae_db": 1.7156593438828176, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12378358599325018, "300–1000 Hz_10ms_envelope_correlation": 0.9857969291704907, "300–1000 Hz_10ms_modulation_mae_db": 0.3448534424497706, "300–1000 Hz_40ms_log_rms_mae_db": 1.0197167538104899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07839474370625525, "300–1000 Hz_40ms_envelope_correlation": 0.9938029190281774, "300–1000 Hz_40ms_modulation_mae_db": 0.24174061262743227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8641123957790324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20449319393521648, "1000–3000 Hz_10ms_envelope_correlation": 0.9552782794687082, "1000–3000 Hz_10ms_modulation_mae_db": 0.7018835547312253, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0110613977687994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10803977631313165, "1000–3000 Hz_40ms_envelope_correlation": 0.9864504590627116, "1000–3000 Hz_40ms_modulation_mae_db": 1.0394956586289645, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0769871456197153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20830136929142298, "3000–6000 Hz_10ms_envelope_correlation": 0.9783417050008759, "3000–6000 Hz_10ms_modulation_mae_db": 1.0776223221573966, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5111401981803279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13759575062175444, "3000–6000 Hz_40ms_envelope_correlation": 0.9921024121392739, "3000–6000 Hz_40ms_modulation_mae_db": 1.1891788997862711, "mr_stft_db": 5.7410229081873, "band_mae_db": 0.2158154910396064, "rms_gain_error_db": -0.08074467395148024, "output_peak": 0.33659749984325077}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1449084331712485, "全帯域_10ms_rms_envelope_nrmse": 0.09912116800075539, "全帯域_10ms_envelope_correlation": 0.9863789060997525, "全帯域_10ms_modulation_mae_db": 0.26060280321376905, "全帯域_40ms_log_rms_mae_db": 0.5606108125830459, "全帯域_40ms_rms_envelope_nrmse": 0.06346412798221497, "全帯域_40ms_envelope_correlation": 0.9933788954377306, "全帯域_40ms_modulation_mae_db": 0.16373683509621922, "300–1000 Hz_10ms_log_rms_mae_db": 1.6921231757110842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14853758490380703, "300–1000 Hz_10ms_envelope_correlation": 0.9793064421341985, "300–1000 Hz_10ms_modulation_mae_db": 0.6978793257661215, "300–1000 Hz_40ms_log_rms_mae_db": 1.0016416486312165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10295831867929904, "300–1000 Hz_40ms_envelope_correlation": 0.9886096820707406, "300–1000 Hz_40ms_modulation_mae_db": 0.2951395641871714, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8090139462953194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2113446667795055, "1000–3000 Hz_10ms_envelope_correlation": 0.9530138665109936, "1000–3000 Hz_10ms_modulation_mae_db": 0.5219197559160287, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0307044282354674, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1253743435595552, "1000–3000 Hz_40ms_envelope_correlation": 0.9808401703926246, "1000–3000 Hz_40ms_modulation_mae_db": 0.5993444549324322, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0398147417283963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18667171320218084, "3000–6000 Hz_10ms_envelope_correlation": 0.9795215582927781, "3000–6000 Hz_10ms_modulation_mae_db": 0.9435224838274677, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4131224903847412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12887444562918624, "3000–6000 Hz_40ms_envelope_correlation": 0.9900995017596119, "3000–6000 Hz_40ms_modulation_mae_db": 0.8719836610212659, "mr_stft_db": 5.726260032657461, "band_mae_db": 0.12027125331528248, "rms_gain_error_db": 0.021398769429544304, "output_peak": 0.3506145353477776}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1637861711036006, "全帯域_10ms_rms_envelope_nrmse": 0.0997177438264969, "全帯域_10ms_envelope_correlation": 0.9862313163517497, "全帯域_10ms_modulation_mae_db": 0.05517027395879656, "全帯域_40ms_log_rms_mae_db": 0.5487768039889196, "全帯域_40ms_rms_envelope_nrmse": 0.04949406850533643, "全帯域_40ms_envelope_correlation": 0.9959961666082381, "全帯域_40ms_modulation_mae_db": 0.30986113056747844, "300–1000 Hz_10ms_log_rms_mae_db": 1.6681022617989612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1306909214568448, "300–1000 Hz_10ms_envelope_correlation": 0.9839863403131704, "300–1000 Hz_10ms_modulation_mae_db": 0.13159613595621, "300–1000 Hz_40ms_log_rms_mae_db": 0.9302487241592156, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07077683058335121, "300–1000 Hz_40ms_envelope_correlation": 0.9946294716756298, "300–1000 Hz_40ms_modulation_mae_db": 0.38691024753042674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8496636909160529, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2108309721966167, "1000–3000 Hz_10ms_envelope_correlation": 0.952281655079462, "1000–3000 Hz_10ms_modulation_mae_db": 0.40605963823234514, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0085120383404091, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11456693355556499, "1000–3000 Hz_40ms_envelope_correlation": 0.9839536573136772, "1000–3000 Hz_40ms_modulation_mae_db": 0.3286742482468626, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0441644165757396, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22327066276721236, "3000–6000 Hz_10ms_envelope_correlation": 0.9697053353309218, "3000–6000 Hz_10ms_modulation_mae_db": 0.8321503653684941, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4218058472308597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1396228328513446, "3000–6000 Hz_40ms_envelope_correlation": 0.9877381918294923, "3000–6000 Hz_40ms_modulation_mae_db": 1.2911258543705257, "mr_stft_db": 5.746841749621823, "band_mae_db": 0.12228709652194303, "rms_gain_error_db": 0.010832397935948583, "output_peak": 0.35550524089491065}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1669953663411752, "全帯域_10ms_rms_envelope_nrmse": 0.08595582327314821, "全帯域_10ms_envelope_correlation": 0.9896607352084416, "全帯域_10ms_modulation_mae_db": 0.4240580562845864, "全帯域_40ms_log_rms_mae_db": 0.5390503005049732, "全帯域_40ms_rms_envelope_nrmse": 0.0418361490115655, "全帯域_40ms_envelope_correlation": 0.9971289317083099, "全帯域_40ms_modulation_mae_db": 0.6144315966832238, "300–1000 Hz_10ms_log_rms_mae_db": 1.703275311470201, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1157302204604965, "300–1000 Hz_10ms_envelope_correlation": 0.9874248279762583, "300–1000 Hz_10ms_modulation_mae_db": 0.4554456067303863, "300–1000 Hz_40ms_log_rms_mae_db": 1.0526577538040225, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06213494799770611, "300–1000 Hz_40ms_envelope_correlation": 0.9959523539817277, "300–1000 Hz_40ms_modulation_mae_db": 0.6220285616155851, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8564221621241714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2121132555479697, "1000–3000 Hz_10ms_envelope_correlation": 0.9522495196243782, "1000–3000 Hz_10ms_modulation_mae_db": 0.26974086417213816, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0580785735865463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12096139457097187, "1000–3000 Hz_40ms_envelope_correlation": 0.9825794604536355, "1000–3000 Hz_40ms_modulation_mae_db": 0.497588386279263, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0639779959031164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2184902361731218, "3000–6000 Hz_10ms_envelope_correlation": 0.9723281087786424, "3000–6000 Hz_10ms_modulation_mae_db": 1.2954314201367059, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4378611339839766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14257339704238528, "3000–6000 Hz_40ms_envelope_correlation": 0.9906915568036598, "3000–6000 Hz_40ms_modulation_mae_db": 1.4158484933326847, "mr_stft_db": 5.762544581899353, "band_mae_db": 0.21415821143872413, "rms_gain_error_db": -0.05081714447362233, "output_peak": 0.3893390932065087}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1769810399026706, "全帯域_10ms_rms_envelope_nrmse": 0.0931334076382274, "全帯域_10ms_envelope_correlation": 0.9879197991356018, "全帯域_10ms_modulation_mae_db": 0.4473868702211978, "全帯域_40ms_log_rms_mae_db": 0.5514546974513741, "全帯域_40ms_rms_envelope_nrmse": 0.04594981794842393, "全帯域_40ms_envelope_correlation": 0.9964897117057149, "全帯域_40ms_modulation_mae_db": 0.2949294892877874, "300–1000 Hz_10ms_log_rms_mae_db": 1.6080765218474333, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11346827908810393, "300–1000 Hz_10ms_envelope_correlation": 0.9879116648841273, "300–1000 Hz_10ms_modulation_mae_db": 0.5306357496753995, "300–1000 Hz_40ms_log_rms_mae_db": 0.9242380555970119, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06666654042886903, "300–1000 Hz_40ms_envelope_correlation": 0.9952134281540486, "300–1000 Hz_40ms_modulation_mae_db": 0.5536685661858953, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8550728327579136, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2242212897238225, "1000–3000 Hz_10ms_envelope_correlation": 0.945854521565376, "1000–3000 Hz_10ms_modulation_mae_db": 0.8048611443886804, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9495081566158354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10830279798454727, "1000–3000 Hz_40ms_envelope_correlation": 0.986221567819425, "1000–3000 Hz_40ms_modulation_mae_db": 0.9522876709404349, "3000–6000 Hz_10ms_log_rms_mae_db": 2.040904755331352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2381004078430915, "3000–6000 Hz_10ms_envelope_correlation": 0.9644823200897568, "3000–6000 Hz_10ms_modulation_mae_db": 0.7525500955411712, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4518813459045004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12865107500585457, "3000–6000 Hz_40ms_envelope_correlation": 0.9892986033728098, "3000–6000 Hz_40ms_modulation_mae_db": 1.151366572336063, "mr_stft_db": 5.7374260245924384, "band_mae_db": 0.10491148248739107, "rms_gain_error_db": -0.0013674089218198148, "output_peak": 0.3627047699038109}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1732926423554708, "全帯域_10ms_rms_envelope_nrmse": 0.10109895853673959, "全帯域_10ms_envelope_correlation": 0.9858946949125781, "全帯域_10ms_modulation_mae_db": 0.7618066354157428, "全帯域_40ms_log_rms_mae_db": 0.5443441027900058, "全帯域_40ms_rms_envelope_nrmse": 0.051961201043586494, "全帯域_40ms_envelope_correlation": 0.995612482810009, "全帯域_40ms_modulation_mae_db": 0.6805094864365154, "300–1000 Hz_10ms_log_rms_mae_db": 1.62950827446351, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.123250711527852, "300–1000 Hz_10ms_envelope_correlation": 0.9857753592287171, "300–1000 Hz_10ms_modulation_mae_db": 1.513265277563444, "300–1000 Hz_40ms_log_rms_mae_db": 0.9902435057694827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07447173231627638, "300–1000 Hz_40ms_envelope_correlation": 0.9942950730647544, "300–1000 Hz_40ms_modulation_mae_db": 1.8384356462186198, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9215814443502868, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22609125481823605, "1000–3000 Hz_10ms_envelope_correlation": 0.9460666137731356, "1000–3000 Hz_10ms_modulation_mae_db": 0.48763335519940626, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0644347140158552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12449642906996197, "1000–3000 Hz_40ms_envelope_correlation": 0.9813622367937535, "1000–3000 Hz_40ms_modulation_mae_db": 0.3605126905094287, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0125838460004193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23302737755532102, "3000–6000 Hz_10ms_envelope_correlation": 0.9668645381279397, "3000–6000 Hz_10ms_modulation_mae_db": 0.5789145881111836, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3458356187084433, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1176051494982317, "3000–6000 Hz_40ms_envelope_correlation": 0.9913531913003286, "3000–6000 Hz_40ms_modulation_mae_db": 0.6039109740242702, "mr_stft_db": 5.744618695487184, "band_mae_db": 0.16696459583432777, "rms_gain_error_db": 0.04246272719742195, "output_peak": 0.34204129396009053}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44560, "output_samples": 44560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9745668385270925, "spectral_convergence": 0.3231405455213443, "spectral_convergence_status": "ok", "amplitude_floor": 9.278908430047933e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.553903766841823, "spectral_convergence": 0.3023956420686867, "spectral_convergence_status": "ok", "amplitude_floor": 4.74996132181228e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.680816571313674, "spectral_convergence": 0.30818531597615106, "spectral_convergence_status": "ok", "amplitude_floor": 3.301834166389807e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005909300513360037, "output_power": 0.0006083434494963771, "delta_db": 0.12612759745442656, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000598279616407713, "output_power": 0.0005646610800305618, "delta_db": -0.25116352052294943, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005543526355345429, "output_power": 0.000519052237841474, "delta_db": -0.28575048817232307, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011395641530102642, "output_power": 0.00010713973726141874, "delta_db": -0.2678820226884906, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011253089498164014, "output_power": 9.901097698491384e-05, "delta_db": -0.5558842701088929, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6482875672349115e-05, "output_power": 1.6105822425607355e-05, "delta_db": -0.10050076348825443, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.26455144373922285, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16082654625159654, "pitch": null, "output_rms_dbfs": -27.17986982399975, "output_peak": 0.3477708812056554, "reference_rms_dbfs": -27.019043277748153, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.736429058894196}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 0.00011061650326899306, "log_rms_mae_db": 1.1659703270163158, "rms_envelope_nrmse": 0.09254188985824197, "envelope_correlation": 0.9882836817540562, "rms_gain_error_db": -0.1608032299432459, "reference_mean_power": 0.00199005080866686, "output_mean_power": 0.0019177140128562864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 8.439637043999617e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970864143567599, "output_power": 0.00057462906834697, "reference_active": true, "delta_db": -0.1664959813898862}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010123102438186569, "output_power": 9.613499901522453e-05, "reference_active": true, "delta_db": -0.224321054134482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240115070482855e-05, "output_power": 2.6653426227311577e-05, "reference_active": true, "delta_db": -0.8480738979769807}], "active_band_count": 3, "active_band_mae_db": 0.41296364450044964}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 9.74750993349889e-05, "log_rms_mae_db": 0.5352862920441215, "rms_envelope_nrmse": 0.04329474229913915, "envelope_correlation": 0.9973887157519706, "rms_gain_error_db": -0.16076921879938855, "reference_mean_power": 0.0020114637230104894, "output_mean_power": 0.0019383637644090374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 6.839204647135984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005172552173341314, "output_power": 0.000497138180945016, "reference_active": true, "delta_db": -0.17227761028015548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6641602045504645e-05, "output_power": 5.4416783757336275e-05, "reference_active": true, "delta_db": -0.1740265868140659}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.529860845744804e-06, "output_power": 2.0164367687319447e-06, "reference_active": true, "delta_db": -0.9851202560751735}], "active_band_count": 3, "active_band_mae_db": 0.44380815105646493}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 8.308955229004579e-05, "log_rms_mae_db": 1.7334352303183933, "rms_envelope_nrmse": 0.1250183555394948, "envelope_correlation": 0.9858361627762572, "rms_gain_error_db": -0.2812250377427206, "reference_mean_power": 0.000839400894204417, "output_mean_power": 0.0007867684276915211}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.694834819211473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042315114528166547, "output_power": 0.00039629941390123335, "reference_active": true, "delta_db": -0.2847209061046912}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.939611213191948e-05, "output_power": 5.8644714545927753e-05, "reference_active": true, "delta_db": -0.7310626340465811}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4277196064029695e-05, "output_power": 1.9095196292534564e-05, "reference_active": true, "delta_db": -1.0427439858002114}], "active_band_count": 3, "active_band_mae_db": 0.686175841983828}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 8.008999734930271e-05, "log_rms_mae_db": 1.0518831042623829, "rms_envelope_nrmse": 0.0657059426796576, "envelope_correlation": 0.9963347345856329, "rms_gain_error_db": -0.2812371848788326, "reference_mean_power": 0.0008485551352947845, "output_mean_power": 0.0007953464511497325}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 5.733703583211585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003764127527272362, "output_power": 0.00035398300583703055, "reference_active": true, "delta_db": -0.2668191600702323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8963437868750806e-05, "output_power": 3.3380206113088676e-05, "reference_active": true, "delta_db": -0.6716825509451554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.296630500499659e-06, "output_power": 1.714909185066843e-06, "reference_active": true, "delta_db": -1.2685000175568413}], "active_band_count": 3, "active_band_mae_db": 0.7356672428574097}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.2528786526254916e-05, "log_rms_mae_db": 1.8813377322404599, "rms_envelope_nrmse": 0.222064384752126, "envelope_correlation": 0.9470903458440432, "rms_gain_error_db": -0.1939736574286649, "reference_mean_power": 0.0001368277838750774, "output_mean_power": 0.00013085096422277466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.073610600879068e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.932287229550182e-05, "output_power": 3.93347821801961e-05, "reference_active": true, "delta_db": 0.0013151669158802178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8616503346595646e-06, "output_power": 8.820762215754196e-06, "reference_active": true, "delta_db": 1.090775319003783}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.317401673918684e-06, "output_power": 4.472029047459364e-06, "reference_active": true, "delta_db": 0.1528216021745102}], "active_band_count": 3, "active_band_mae_db": 0.4149706960313911}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 3.268734932259802e-05, "log_rms_mae_db": 1.0425292803167745, "rms_envelope_nrmse": 0.1288290471726013, "envelope_correlation": 0.9794647467225637, "rms_gain_error_db": -0.1940384032374815, "reference_mean_power": 0.00013831808192096785, "output_mean_power": 0.0001322741920726219}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 4.85870788941741e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5019378943818874e-05, "output_power": 3.6120439489478765e-05, "reference_active": true, "delta_db": 0.1344458612179832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.168829067205367e-06, "output_power": 5.234723925449731e-06, "reference_active": true, "delta_db": 0.9887969403300896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163313597471507e-07, "output_power": 3.282449429450945e-07, "reference_active": true, "delta_db": 0.16055795762435962}], "active_band_count": 3, "active_band_mae_db": 0.42793358639081075}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.9606629692154165e-05, "log_rms_mae_db": 2.0239388902969515, "rms_envelope_nrmse": 0.19558540184862278, "envelope_correlation": 0.9775401742900329, "rms_gain_error_db": -0.5414445888878501, "reference_mean_power": 5.1065139870557587e-05, "output_mean_power": 4.5079601360141805e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 2.3044643706212192e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0452875124351046e-05, "output_power": 9.052204619664009e-06, "reference_active": true, "delta_db": -0.624813997518724}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.855628387226494e-06, "output_power": 2.3791551931132483e-06, "reference_active": true, "delta_db": -0.792789184019592}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6522678437727886e-06, "output_power": 6.67647023558085e-07, "reference_active": true, "delta_db": -3.9353353354774825}], "active_band_count": 3, "active_band_mae_db": 1.7843128390052663}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 4.1718716562101114e-05, "log_rms_mae_db": 1.4320382825796782, "rms_envelope_nrmse": 0.13074572453181418, "envelope_correlation": 0.9912710530788151, "rms_gain_error_db": -0.5414760149739919, "reference_mean_power": 5.162171435861935e-05, "output_mean_power": 4.5570607888060004e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 2.066669918587524e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0052154991285675e-05, "output_power": 8.429462120678999e-06, "reference_active": true, "delta_db": -0.7645931281593672}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.890809551942203e-06, "output_power": 1.322794004340627e-06, "reference_active": true, "delta_db": -1.5515556981889709}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8144313206293327e-07, "output_power": 6.315687956859686e-08, "reference_active": true, "delta_db": -4.583198696411811}], "active_band_count": 3, "active_band_mae_db": 2.2997825075867167}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8411974415967339, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "982e87708e3e06d13e684c173b1055fff305323abed0ab144c21d18307f7d6ac", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9502102430981487, "standard_deviation": 0.003545666729796473, "minimum": 0.945854521565376, "maximum": 0.9552782794687082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8705293173313202, "standard_deviation": 0.03893972132262284, "minimum": 1.8090139462953194, "maximum": 1.9270303341873272, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5277554311821459, "standard_deviation": 0.17339506943053448, "minimum": 0.26974086417213816, "maximum": 0.8048611443886804, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21593445109380185, "standard_deviation": 0.007579766219650072, "minimum": 0.20449319393521648, "maximum": 0.22609125481823605, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9831353906783836, "standard_deviation": 0.002522331329442534, "minimum": 0.9794647467225637, "maximum": 0.9864504590627116, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.02438347262285, "standard_deviation": 0.036211026507674667, "minimum": 0.9495081566158354, "maximum": 1.0644347140158552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6203873950546399, "standard_deviation": 0.26985091302019754, "minimum": 0.3286742482468626, "maximum": 1.0394956586289645, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11835013069626046, "standard_deviation": 0.007835828021991885, "minimum": 0.10803977631313165, "maximum": 0.1288290471726013, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9724258422862322, "standard_deviation": 0.005557383943427654, "minimum": 0.9644823200897568, "maximum": 0.9795215582927781, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.044144943099507, "standard_deviation": 0.02052776669241108, "minimum": 2.0125838460004193, "maximum": 2.0769871456197153, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0344392290257978, "standard_deviation": 0.37205834743172933, "minimum": 0.5789145881111836, "maximum": 1.7843128390052663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21565363373674235, "standard_deviation": 0.017752533784582424, "minimum": 0.18667171320218084, "maximum": 0.2381004078430915, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9903050332897081, "standard_deviation": 0.0013748461467524473, "minimum": 0.9877381918294923, "maximum": 0.9921024121392739, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4353753603244936, "standard_deviation": 0.047627814597221436, "minimum": 1.3458356187084433, "maximum": 1.5111401981803279, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.27361193554986, "standard_deviation": 0.4948190828336938, "minimum": 0.6039109740242702, "maximum": 2.2997825075867167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13344852507505497, "standard_deviation": 0.008557396004060474, "minimum": 0.1176051494982317, "maximum": 0.14257339704238528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9855245352656687, "standard_deviation": 0.00286636933752105, "minimum": 0.9793064421341985, "maximum": 0.9881585556421306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6742753832487753, "standard_deviation": 0.04425859056776866, "minimum": 1.6080765218474333, "maximum": 1.7334352303183933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.648444187159011, "standard_deviation": 0.41306225108015376, "minimum": 0.13159613595621, "maximum": 1.513265277563444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12409867021480517, "standard_deviation": 0.011737597596909644, "minimum": 0.11230970274859209, "maximum": 0.14853758490380703, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942992754572937, "standard_deviation": 0.002450045023673732, "minimum": 0.9886096820707406, "maximum": 0.9963347345856329, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9903522419085052, "standard_deviation": 0.05084677128382789, "minimum": 0.9242380555970119, "maximum": 1.0526577538040225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6462485232635768, "standard_deviation": 0.5092121711175694, "minimum": 0.24174061262743227, "maximum": 1.8384356462186198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07326513318985937, "standard_deviation": 0.013138403722726701, "minimum": 0.06213494799770611, "maximum": 0.10295831867929904, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18543695837062518, "standard_deviation": 0.06655349915583166, "minimum": 0.10491148248739107, "maximum": 0.27453609258850364, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.742355914759485, "standard_deviation": 0.010384940949291556, "minimum": 5.726260032657461, "maximum": 5.762544581899353, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.355935914093988, "standard_deviation": 0.016355376974601012, "minimum": 0.33659749984325077, "maximum": 0.3893390932065087, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.032347227093274894, "standard_deviation": 0.06606714616075414, "minimum": -0.16082654625159654, "maximum": 0.04246272719742195, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9874557285029112, "standard_deviation": 0.0012513620126361512, "minimum": 0.9858946949125781, "maximum": 0.9896607352084416, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.164494885826222, "standard_deviation": 0.041686956663324545, "minimum": 1.0860967208714358, "maximum": 1.2379283858478607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37904340225609706, "standard_deviation": 0.20282441643225646, "minimum": 0.05517027395879656, "maximum": 0.7618066354157428, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09496667594275016, "standard_deviation": 0.004925989769993914, "minimum": 0.08595582327314821, "maximum": 0.10109895853673959, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959446140363182, "standard_deviation": 0.0012373677360975634, "minimum": 0.9933788954377306, "maximum": 0.9973887157519706, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5559747538728207, "standard_deviation": 0.034507882524333865, "minimum": 0.5302588513734029, "maximum": 0.6380161702467229, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3634648005918244, "standard_deviation": 0.19671418804069352, "minimum": 0.16373683509621922, "maximum": 0.6805094864365154, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049679394577819, "standard_deviation": 0.006779121605831282, "minimum": 0.0418361490115655, "maximum": 0.06346412798221497, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0042", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.960693120956421, "predicted_mos": 1.960693120956421, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "751925bea8d97959289633b1700aeaa586c607fe518c3cecdcee43c30886b4c0", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.145208100322634, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1837051652656083, "全帯域_10ms_rms_envelope_nrmse": 0.0945276802582679, "全帯域_10ms_envelope_correlation": 0.9877477250970732, "全帯域_10ms_modulation_mae_db": 0.46137966428868715, "全帯域_40ms_log_rms_mae_db": 0.5664079600050417, "全帯域_40ms_rms_envelope_nrmse": 0.0444995147359249, "全帯域_40ms_envelope_correlation": 0.9972008123974117, "全帯域_40ms_modulation_mae_db": 0.5266851332224548, "300–1000 Hz_10ms_log_rms_mae_db": 1.7346043124398636, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12879549068357843, "300–1000 Hz_10ms_envelope_correlation": 0.9846887405606293, "300–1000 Hz_10ms_modulation_mae_db": 0.7118411720193643, "300–1000 Hz_40ms_log_rms_mae_db": 0.9430998581371984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0666744743887637, "300–1000 Hz_40ms_envelope_correlation": 0.9958774393299843, "300–1000 Hz_40ms_modulation_mae_db": 0.7877852783374757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.796454801914969, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21792918716477655, "1000–3000 Hz_10ms_envelope_correlation": 0.9492589934148529, "1000–3000 Hz_10ms_modulation_mae_db": 0.3226608916433924, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9739918893587508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12884774538595375, "1000–3000 Hz_40ms_envelope_correlation": 0.980170888804828, "1000–3000 Hz_40ms_modulation_mae_db": 0.25193965968817866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9905328869508303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19550112807467004, "3000–6000 Hz_10ms_envelope_correlation": 0.977458452637765, "3000–6000 Hz_10ms_modulation_mae_db": 1.6085017687149648, "3000–6000 Hz_40ms_log_rms_mae_db": 1.436652345736418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13032913169706192, "3000–6000 Hz_40ms_envelope_correlation": 0.9912803746621378, "3000–6000 Hz_40ms_modulation_mae_db": 2.090204344047368, "mr_stft_db": 5.7000628149465955, "band_mae_db": 0.3186542290427143, "rms_gain_error_db": -0.15955999829807985, "output_peak": 0.347733347966685}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.134589877359661, "全帯域_10ms_rms_envelope_nrmse": 0.09091350567911298, "全帯域_10ms_envelope_correlation": 0.9884983266803838, "全帯域_10ms_modulation_mae_db": 0.24882371206327206, "全帯域_40ms_log_rms_mae_db": 0.5656485461835562, "全帯域_40ms_rms_envelope_nrmse": 0.0455480155128569, "全帯域_40ms_envelope_correlation": 0.9966206405809787, "全帯域_40ms_modulation_mae_db": 0.10971862024865763, "300–1000 Hz_10ms_log_rms_mae_db": 1.606258439731709, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.110462190039584, "300–1000 Hz_10ms_envelope_correlation": 0.9885957179392494, "300–1000 Hz_10ms_modulation_mae_db": 0.6211332809645826, "300–1000 Hz_40ms_log_rms_mae_db": 0.855361530982162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05948780036249176, "300–1000 Hz_40ms_envelope_correlation": 0.9963884804682099, "300–1000 Hz_40ms_modulation_mae_db": 0.35552081536461383, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9096820840498623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21826482353080687, "1000–3000 Hz_10ms_envelope_correlation": 0.9491614303545917, "1000–3000 Hz_10ms_modulation_mae_db": 0.5585304017585048, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0146650217468964, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11874165053514839, "1000–3000 Hz_40ms_envelope_correlation": 0.9838804099524059, "1000–3000 Hz_40ms_modulation_mae_db": 0.6655007782989376, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0122963243492817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22823398780839346, "3000–6000 Hz_10ms_envelope_correlation": 0.96857774363735, "3000–6000 Hz_10ms_modulation_mae_db": 1.024196915284662, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4800444575897, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14544831750543905, "3000–6000 Hz_40ms_envelope_correlation": 0.9890762078565343, "3000–6000 Hz_40ms_modulation_mae_db": 1.2996997196822513, "mr_stft_db": 5.693974016298372, "band_mae_db": 0.25570537865204573, "rms_gain_error_db": -0.04851579317041819, "output_peak": 0.36411909811374876}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2122554120537412, "全帯域_10ms_rms_envelope_nrmse": 0.09267720972647972, "全帯域_10ms_envelope_correlation": 0.9880649600658777, "全帯域_10ms_modulation_mae_db": 0.1855219696633719, "全帯域_40ms_log_rms_mae_db": 0.5703359863174617, "全帯域_40ms_rms_envelope_nrmse": 0.04780755125469314, "全帯域_40ms_envelope_correlation": 0.9964287631079622, "全帯域_40ms_modulation_mae_db": 0.38619754417370333, "300–1000 Hz_10ms_log_rms_mae_db": 1.6369983761141829, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11890354430130146, "300–1000 Hz_10ms_envelope_correlation": 0.9870359997296407, "300–1000 Hz_10ms_modulation_mae_db": 0.30553311104980185, "300–1000 Hz_40ms_log_rms_mae_db": 0.8534255316456771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07084304258065131, "300–1000 Hz_40ms_envelope_correlation": 0.9951620257337989, "300–1000 Hz_40ms_modulation_mae_db": 0.4242947580602827, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8192473722934557, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2065719475221004, "1000–3000 Hz_10ms_envelope_correlation": 0.9549870659500553, "1000–3000 Hz_10ms_modulation_mae_db": 0.7593760661102321, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9880361207282002, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11394254057575487, "1000–3000 Hz_40ms_envelope_correlation": 0.9860337189614021, "1000–3000 Hz_40ms_modulation_mae_db": 1.1326877191799687, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0304941088667525, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21114816157166252, "3000–6000 Hz_10ms_envelope_correlation": 0.9777722861887872, "3000–6000 Hz_10ms_modulation_mae_db": 1.0329847043968148, "3000–6000 Hz_40ms_log_rms_mae_db": 1.532752981176913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14007363456633087, "3000–6000 Hz_40ms_envelope_correlation": 0.9917471168388853, "3000–6000 Hz_40ms_modulation_mae_db": 1.1556585086858764, "mr_stft_db": 5.694172990016469, "band_mae_db": 0.242076218907488, "rms_gain_error_db": -0.11087033147117853, "output_peak": 0.3350911550826036}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1516305645087743, "全帯域_10ms_rms_envelope_nrmse": 0.09541839012111376, "全帯域_10ms_envelope_correlation": 0.9873356042611277, "全帯域_10ms_modulation_mae_db": 0.29423330434392625, "全帯域_40ms_log_rms_mae_db": 0.5585929743707392, "全帯域_40ms_rms_envelope_nrmse": 0.058760687954639776, "全帯域_40ms_envelope_correlation": 0.9942971377089928, "全帯域_40ms_modulation_mae_db": 0.1959149647671092, "300–1000 Hz_10ms_log_rms_mae_db": 1.6400122709977614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1424182131084223, "300–1000 Hz_10ms_envelope_correlation": 0.9809156246570411, "300–1000 Hz_10ms_modulation_mae_db": 0.67673254029966, "300–1000 Hz_40ms_log_rms_mae_db": 0.8808707782716774, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09466786138676382, "300–1000 Hz_40ms_envelope_correlation": 0.9903351526995674, "300–1000 Hz_40ms_modulation_mae_db": 0.23751306221829804, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7877587302673124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21005880301399327, "1000–3000 Hz_10ms_envelope_correlation": 0.9535659854948496, "1000–3000 Hz_10ms_modulation_mae_db": 0.38493288573709655, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0036441880024416, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12607365339105148, "1000–3000 Hz_40ms_envelope_correlation": 0.9807366291764681, "1000–3000 Hz_40ms_modulation_mae_db": 0.44045910668043603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9971404524573264, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.186920703511005, "3000–6000 Hz_10ms_envelope_correlation": 0.9796326568504102, "3000–6000 Hz_10ms_modulation_mae_db": 0.971809387313086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4353953663222212, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12860871584934597, "3000–6000 Hz_40ms_envelope_correlation": 0.9902673973526377, "3000–6000 Hz_40ms_modulation_mae_db": 0.9397119200433351, "mr_stft_db": 5.6917088521993415, "band_mae_db": 0.09450002368678506, "rms_gain_error_db": 0.006171385716153165, "output_peak": 0.34988878106569504}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1749092265833345, "全帯域_10ms_rms_envelope_nrmse": 0.09412345156858647, "全帯域_10ms_envelope_correlation": 0.9877398472043131, "全帯域_10ms_modulation_mae_db": 0.09364793931402149, "全帯域_40ms_log_rms_mae_db": 0.5526121921835594, "全帯域_40ms_rms_envelope_nrmse": 0.04667325502365904, "全帯域_40ms_envelope_correlation": 0.9964633128817703, "全帯域_40ms_modulation_mae_db": 0.23821843984195015, "300–1000 Hz_10ms_log_rms_mae_db": 1.673084406427591, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12325935622000249, "300–1000 Hz_10ms_envelope_correlation": 0.9857527444651992, "300–1000 Hz_10ms_modulation_mae_db": 0.15484158359584352, "300–1000 Hz_40ms_log_rms_mae_db": 0.8900298249069131, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0653749477399268, "300–1000 Hz_40ms_envelope_correlation": 0.99541180272316, "300–1000 Hz_40ms_modulation_mae_db": 0.40355756279358784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8055548922603466, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20673778462116285, "1000–3000 Hz_10ms_envelope_correlation": 0.9544673395360274, "1000–3000 Hz_10ms_modulation_mae_db": 0.45684333848189734, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9728373825253125, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11813197539760753, "1000–3000 Hz_40ms_envelope_correlation": 0.9838184029520077, "1000–3000 Hz_40ms_modulation_mae_db": 0.42096818900701355, "3000–6000 Hz_10ms_log_rms_mae_db": 1.991815274453156, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2233586254473387, "3000–6000 Hz_10ms_envelope_correlation": 0.9692146073494745, "3000–6000 Hz_10ms_modulation_mae_db": 0.8238526642964423, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4756074728206972, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14120402632615264, "3000–6000 Hz_40ms_envelope_correlation": 0.9871572319585744, "3000–6000 Hz_40ms_modulation_mae_db": 1.24231821187429, "mr_stft_db": 5.704256506304365, "band_mae_db": 0.0968249876562523, "rms_gain_error_db": -0.005167021978693038, "output_peak": 0.3564253410698049}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1744727427534525, "全帯域_10ms_rms_envelope_nrmse": 0.08608201509643272, "全帯域_10ms_envelope_correlation": 0.989624742175538, "全帯域_10ms_modulation_mae_db": 0.3518365699101693, "全帯域_40ms_log_rms_mae_db": 0.5266512923637305, "全帯域_40ms_rms_envelope_nrmse": 0.04281514019616447, "全帯域_40ms_envelope_correlation": 0.9969747776722451, "全帯域_40ms_modulation_mae_db": 0.5535085536729177, "300–1000 Hz_10ms_log_rms_mae_db": 1.6256970776536093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11648656867820542, "300–1000 Hz_10ms_envelope_correlation": 0.9872413117540069, "300–1000 Hz_10ms_modulation_mae_db": 0.4299702573457491, "300–1000 Hz_40ms_log_rms_mae_db": 0.9331429703628977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06332153538797125, "300–1000 Hz_40ms_envelope_correlation": 0.9957412784911089, "300–1000 Hz_40ms_modulation_mae_db": 0.6315811450079031, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8081570454338116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21317221428947108, "1000–3000 Hz_10ms_envelope_correlation": 0.9520043273226252, "1000–3000 Hz_10ms_modulation_mae_db": 0.27544766227375644, "1000–3000 Hz_40ms_log_rms_mae_db": 1.022631702729419, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12458564321064824, "1000–3000 Hz_40ms_envelope_correlation": 0.981885346993473, "1000–3000 Hz_40ms_modulation_mae_db": 0.5546673637625242, "3000–6000 Hz_10ms_log_rms_mae_db": 2.034996032594276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21711611697226055, "3000–6000 Hz_10ms_envelope_correlation": 0.9726086687816167, "3000–6000 Hz_10ms_modulation_mae_db": 1.2532859901509923, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4542098668564694, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14248049715981984, "3000–6000 Hz_40ms_envelope_correlation": 0.9905661761966215, "3000–6000 Hz_40ms_modulation_mae_db": 1.4278991397052083, "mr_stft_db": 5.71098913844881, "band_mae_db": 0.24697352219431298, "rms_gain_error_db": -0.03956488956646481, "output_peak": 0.38371611951700096}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2458848819313952, "全帯域_10ms_rms_envelope_nrmse": 0.09218932937134146, "全帯域_10ms_envelope_correlation": 0.9881694625270234, "全帯域_10ms_modulation_mae_db": 0.4438175220942393, "全帯域_40ms_log_rms_mae_db": 0.613733439279229, "全帯域_40ms_rms_envelope_nrmse": 0.04586882329845287, "全帯域_40ms_envelope_correlation": 0.9965061660658929, "全帯域_40ms_modulation_mae_db": 0.2382176318915156, "300–1000 Hz_10ms_log_rms_mae_db": 1.6532064629699708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11408369907675212, "300–1000 Hz_10ms_envelope_correlation": 0.9877935299561512, "300–1000 Hz_10ms_modulation_mae_db": 0.5063070665001972, "300–1000 Hz_40ms_log_rms_mae_db": 0.8855167229489384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06713555876971983, "300–1000 Hz_40ms_envelope_correlation": 0.9951532885134535, "300–1000 Hz_40ms_modulation_mae_db": 0.5711162696057549, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7979538117915619, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22532109160221014, "1000–3000 Hz_10ms_envelope_correlation": 0.9455854847647234, "1000–3000 Hz_10ms_modulation_mae_db": 0.8339486673230407, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8971804417302887, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11407601231352513, "1000–3000 Hz_40ms_envelope_correlation": 0.9853272461596457, "1000–3000 Hz_40ms_modulation_mae_db": 1.0034270787351722, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0287458107854035, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24555064025852555, "3000–6000 Hz_10ms_envelope_correlation": 0.9625034158120166, "3000–6000 Hz_10ms_modulation_mae_db": 0.5848456199910828, "3000–6000 Hz_40ms_log_rms_mae_db": 1.492871918443716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12837370588729044, "3000–6000 Hz_40ms_envelope_correlation": 0.9893055758068703, "3000–6000 Hz_40ms_modulation_mae_db": 1.0182095915464489, "mr_stft_db": 5.70261532787674, "band_mae_db": 0.12968797063001422, "rms_gain_error_db": -0.01362975197996389, "output_peak": 0.36299778041463315}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2067824289633786, "全帯域_10ms_rms_envelope_nrmse": 0.1064907959795129, "全帯域_10ms_envelope_correlation": 0.98435856265519, "全帯域_10ms_modulation_mae_db": 0.737558282862727, "全帯域_40ms_log_rms_mae_db": 0.5554796554186772, "全帯域_40ms_rms_envelope_nrmse": 0.055784450534938565, "全帯域_40ms_envelope_correlation": 0.9949453591343765, "全帯域_40ms_modulation_mae_db": 0.7292412942444352, "300–1000 Hz_10ms_log_rms_mae_db": 1.6549697399712981, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13580384470881593, "300–1000 Hz_10ms_envelope_correlation": 0.9826350058644425, "300–1000 Hz_10ms_modulation_mae_db": 1.4369940571682271, "300–1000 Hz_40ms_log_rms_mae_db": 0.9241265230409772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07901527823454578, "300–1000 Hz_40ms_envelope_correlation": 0.9934741274536725, "300–1000 Hz_40ms_modulation_mae_db": 1.8973354626099443, "1000–3000 Hz_10ms_log_rms_mae_db": 1.886585411679777, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22306612271399837, "1000–3000 Hz_10ms_envelope_correlation": 0.9472063616571271, "1000–3000 Hz_10ms_modulation_mae_db": 0.5986597630139592, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0510992030719635, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12704330145754522, "1000–3000 Hz_40ms_envelope_correlation": 0.9808972184546288, "1000–3000 Hz_40ms_modulation_mae_db": 0.40489513889097584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.979940296730284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23440173186535537, "3000–6000 Hz_10ms_envelope_correlation": 0.9662160300689399, "3000–6000 Hz_10ms_modulation_mae_db": 0.6155156009434893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.405868136955645, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12080850821463111, "3000–6000 Hz_40ms_envelope_correlation": 0.9906685370543864, "3000–6000 Hz_40ms_modulation_mae_db": 0.6442428607422408, "mr_stft_db": 5.712961608775423, "band_mae_db": 0.15893221749921424, "rms_gain_error_db": 0.03671458475892243, "output_peak": 0.34574709057081243}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44560, "output_samples": 44560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.934172389368523, "spectral_convergence": 0.31983590083214813, "spectral_convergence_status": "ok", "amplitude_floor": 9.278908430047933e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.511922359416564, "spectral_convergence": 0.2993421581763378, "spectral_convergence_status": "ok", "amplitude_floor": 4.74996132181228e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6540936960547015, "spectral_convergence": 0.3070957544641737, "spectral_convergence_status": "ok", "amplitude_floor": 3.301834166389807e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005909300513360037, "output_power": 0.0006055671616776633, "delta_db": 0.10626240192858875, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000598279616407713, "output_power": 0.0005629026365968981, "delta_db": -0.2647092366155478, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005543526355345429, "output_power": 0.0005274539113501943, "delta_db": -0.21601599416349485, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011395641530102642, "output_power": 0.0001038029358089989, "delta_db": -0.40529142657289624, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011253089498164014, "output_power": 0.00010016046476774877, "delta_db": -0.5057544182058114, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6482875672349115e-05, "output_power": 1.4984553034240428e-05, "delta_db": -0.4138918967699468, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3186542290427143, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15955999829807985, "pitch": null, "output_rms_dbfs": -27.178603276046232, "output_peak": 0.347733347966685, "reference_rms_dbfs": -27.019043277748153, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7000628149465955}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 0.00011061650326899306, "log_rms_mae_db": 1.1837051652656083, "rms_envelope_nrmse": 0.0945276802582679, "envelope_correlation": 0.9877477250970732, "rms_gain_error_db": -0.1595599260159945, "reference_mean_power": 0.00199005080866686, "output_mean_power": 0.0019182630971265047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 8.439637043999617e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970864143567599, "output_power": 0.0005757590412874584, "reference_active": true, "delta_db": -0.157964231687847}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010123102438186569, "output_power": 9.553904791091538e-05, "reference_active": true, "delta_db": -0.2513272257310507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240115070482855e-05, "output_power": 2.588663979132642e-05, "reference_active": true, "delta_db": -0.9748475354471636}], "active_band_count": 3, "active_band_mae_db": 0.46137966428868715}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 9.74750993349889e-05, "log_rms_mae_db": 0.5664079600050417, "rms_envelope_nrmse": 0.0444995147359249, "envelope_correlation": 0.9972008123974117, "rms_gain_error_db": -0.15950069068288641, "reference_mean_power": 0.0020114637230104894, "output_mean_power": 0.00193893002259959}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 6.839204647135984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005172552173341314, "output_power": 0.000498746307172728, "reference_active": true, "delta_db": -0.15825186809305286}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6641602045504645e-05, "output_power": 5.4609047529618254e-05, "reference_active": true, "delta_db": -0.15870926596589965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.529860845744804e-06, "output_power": 1.8914166467423586e-06, "reference_active": true, "delta_db": -1.2630942656084119}], "active_band_count": 3, "active_band_mae_db": 0.5266851332224548}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 8.308955229004579e-05, "log_rms_mae_db": 1.7346043124398636, "rms_envelope_nrmse": 0.12879549068357843, "envelope_correlation": 0.9846887405606293, "rms_gain_error_db": -0.2354299319094708, "reference_mean_power": 0.000839400894204417, "output_mean_power": 0.0007951085697258481}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.694834819211473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042315114528166547, "output_power": 0.0003982250725884519, "reference_active": true, "delta_db": -0.2636692055307278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.939611213191948e-05, "output_power": 6.059342210850254e-05, "reference_active": true, "delta_db": -0.5890965944447919}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4277196064029695e-05, "output_power": 1.806852721157218e-05, "reference_active": true, "delta_db": -1.282757716082573}], "active_band_count": 3, "active_band_mae_db": 0.7118411720193643}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 8.008999734930271e-05, "log_rms_mae_db": 0.9430998581371984, "rms_envelope_nrmse": 0.0666744743887637, "envelope_correlation": 0.9958774393299843, "rms_gain_error_db": -0.23543563694369762, "reference_mean_power": 0.0008485551352947845, "output_mean_power": 0.0008037787168411808}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 5.733703583211585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003764127527272362, "output_power": 0.0003561307971298019, "reference_active": true, "delta_db": -0.24054796953144245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8963437868750806e-05, "output_power": 3.428918281089904e-05, "reference_active": true, "delta_db": -0.5550013407401175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.296630500499659e-06, "output_power": 1.6007019563186505e-06, "reference_active": true, "delta_db": -1.5678065247408672}], "active_band_count": 3, "active_band_mae_db": 0.7877852783374757}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.2528786526254916e-05, "log_rms_mae_db": 1.796454801914969, "rms_envelope_nrmse": 0.21792918716477655, "envelope_correlation": 0.9492589934148529, "rms_gain_error_db": -0.3055144892120947, "reference_mean_power": 0.0001368277838750774, "output_mean_power": 0.00012753307876116094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.073610600879068e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.932287229550182e-05, "output_power": 3.826487821414137e-05, "reference_active": true, "delta_db": -0.11844897440457879}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8616503346595646e-06, "output_power": 8.3334495450326e-06, "reference_active": true, "delta_db": 0.8439622740887832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.317401673918684e-06, "output_power": 4.31186656713995e-06, "reference_active": true, "delta_db": -0.005571426436815146}], "active_band_count": 3, "active_band_mae_db": 0.3226608916433924}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 3.268734932259802e-05, "log_rms_mae_db": 0.9739918893587508, "rms_envelope_nrmse": 0.12884774538595375, "envelope_correlation": 0.980170888804828, "rms_gain_error_db": -0.305561235747327, "reference_mean_power": 0.00013831808192096785, "output_mean_power": 0.00012892075324574725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 4.85870788941741e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5019378943818874e-05, "output_power": 3.506311306968746e-05, "reference_active": true, "delta_db": 0.005420324379285172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.168829067205367e-06, "output_power": 4.939235261762189e-06, "reference_active": true, "delta_db": 0.7364562433458589}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163313597471507e-07, "output_power": 3.153174509428872e-07, "reference_active": true, "delta_db": -0.013942411339391931}], "active_band_count": 3, "active_band_mae_db": 0.25193965968817866}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.9606629692154165e-05, "log_rms_mae_db": 1.9905328869508303, "rms_envelope_nrmse": 0.19550112807467004, "envelope_correlation": 0.977458452637765, "rms_gain_error_db": -0.5687153116862624, "reference_mean_power": 5.1065139870557587e-05, "output_mean_power": 4.479741918172002e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 2.3044643706212192e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0452875124351046e-05, "output_power": 9.097308339715805e-06, "reference_active": true, "delta_db": -0.6032284738736481}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.855628387226494e-06, "output_power": 2.460368498801579e-06, "reference_active": true, "delta_db": -0.647015326211722}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6522678437727886e-06, "output_power": 7.25361164852605e-07, "reference_active": true, "delta_db": -3.575261506059524}], "active_band_count": 3, "active_band_mae_db": 1.6085017687149648}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 4.1718716562101114e-05, "log_rms_mae_db": 1.436652345736418, "rms_envelope_nrmse": 0.13032913169706192, "envelope_correlation": 0.9912803746621378, "rms_gain_error_db": -0.5687306275190311, "reference_mean_power": 5.162171435861935e-05, "output_mean_power": 4.52855201717824e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 2.066669918587524e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0052154991285675e-05, "output_power": 8.493162924690522e-06, "reference_active": true, "delta_db": -0.7318972098040684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.890809551942203e-06, "output_power": 1.390936396400426e-06, "reference_active": true, "delta_db": -1.333405162099687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8144313206293327e-07, "output_power": 6.8898452582288e-08, "reference_active": true, "delta_db": -4.205310660238347}], "active_band_count": 3, "active_band_mae_db": 2.090204344047368}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8377175992578889, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "24334eba2153d766d707baaf571bca28aee5b68bb1715684aeda83b8fb8cb000", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9507796235618566, "standard_deviation": 0.003489651568016691, "minimum": 0.9455854847647234, "maximum": 0.9549870659500553, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8264242687113872, "standard_deviation": 0.04563162565679933, "minimum": 1.7877587302673124, "maximum": 1.9096820840498623, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.523799959542735, "standard_deviation": 0.20156585531460863, "minimum": 0.27544766227375644, "maximum": 0.8339486673230407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21514024680731494, "standard_deviation": 0.007148642469438887, "minimum": 0.2065719475221004, "maximum": 0.22532109160221014, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9828437326818575, "standard_deviation": 0.002225178448533547, "minimum": 0.980170888804828, "maximum": 0.9860337189614021, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9905107437366591, "standard_deviation": 0.04591999750158378, "minimum": 0.8971804417302887, "maximum": 1.0510992030719635, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6093181292804009, "standard_deviation": 0.30911809738966695, "minimum": 0.25193965968817866, "maximum": 1.1326877191799687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12143031528340432, "standard_deviation": 0.005931767655161221, "minimum": 0.11394254057575487, "maximum": 0.12884774538595375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9717479826657951, "standard_deviation": 0.0061451896363167256, "minimum": 0.9625034158120166, "maximum": 0.9796326568504102, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.008245148398414, "standard_deviation": 0.021231577208810644, "minimum": 1.979940296730284, "maximum": 2.034996032594276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9893740813864417, "standard_deviation": 0.33534681460838883, "minimum": 0.5848456199910828, "maximum": 1.6085017687149648, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21777888693865138, "standard_deviation": 0.019570253226027023, "minimum": 0.186920703511005, "maximum": 0.24555064025852555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900085772158309, "standard_deviation": 0.0014612682172736747, "minimum": 0.9871572319585744, "maximum": 0.9917471168388853, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4641753182377224, "standard_deviation": 0.0396299881360147, "minimum": 1.405868136955645, "maximum": 1.532752981176913, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2272430370408773, "standard_deviation": 0.42457121151372335, "minimum": 0.6442428607422408, "maximum": 2.090204344047368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13466581715075898, "standard_deviation": 0.008754061897746988, "minimum": 0.12080850821463111, "maximum": 0.14544831750543905, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.985582334365795, "standard_deviation": 0.002676022895361452, "minimum": 0.9809156246570411, "maximum": 0.9885957179392494, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6531038857882483, "standard_deviation": 0.03858467395847752, "minimum": 1.606258439731709, "maximum": 1.7346043124398636, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6054191336179282, "standard_deviation": 0.38577052862397426, "minimum": 0.15484158359584352, "maximum": 1.4369940571682271, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12377661335208276, "standard_deviation": 0.011121011187370647, "minimum": 0.110462190039584, "maximum": 0.1424182131084223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946929494266195, "standard_deviation": 0.0019567264830556382, "minimum": 0.9903351526995674, "maximum": 0.9963884804682099, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8956967175370552, "standard_deviation": 0.034282165914494483, "minimum": 0.8534255316456771, "maximum": 0.9430998581371984, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6635880442497325, "standard_deviation": 0.5275896067288504, "minimum": 0.23751306221829804, "maximum": 1.8973354626099443, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07081506235635429, "standard_deviation": 0.011218213621681534, "minimum": 0.05948780036249176, "maximum": 0.09466786138676382, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19291931853360333, "standard_deviation": 0.08380540322934313, "minimum": 0.09450002368678506, "maximum": 0.3186542290427143, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.701342656858264, "standard_deviation": 0.007913045600184243, "minimum": 5.6917088521993415, "maximum": 5.712961608775423, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.355714839225123, "standard_deviation": 0.014787006013601427, "minimum": 0.3350911550826036, "maximum": 0.38371611951700096, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04180272699871534, "standard_deviation": 0.0646855661150614, "minimum": -0.15955999829807985, "maximum": 0.03671458475892243, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9876924038333159, "standard_deviation": 0.0015117133373034278, "minimum": 0.98435856265519, "maximum": 0.989624742175538, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1855287874274183, "standard_deviation": 0.035438249626850084, "minimum": 1.134589877359661, "maximum": 1.2458848819313952, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3521023705675518, "standard_deviation": 0.19910247183795832, "minimum": 0.09364793931402149, "maximum": 0.737558282862727, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.094052797225106, "standard_deviation": 0.00580089199379351, "minimum": 0.08608201509643272, "maximum": 0.1064907959795129, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9961796211937037, "standard_deviation": 0.001013193996515456, "minimum": 0.9942971377089928, "maximum": 0.9972008123974117, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5636827557652493, "standard_deviation": 0.02434194163146879, "minimum": 0.5266512923637305, "maximum": 0.613733439279229, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3722127727578429, "standard_deviation": 0.21390521450425745, "minimum": 0.10971862024865763, "maximum": 0.7292412942444352, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04846967981391621, "standard_deviation": 0.005683470788760939, "minimum": 0.04281514019616447, "maximum": 0.058760687954639776, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0042", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.8270602226257324, "predicted_mos": 1.8270602226257324, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "6210befcb5bd38dafe6330ee0d98af0b7a6cb4cdea258c1a5a354011a67ead94", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9899735003709793, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.042029240111435, "全帯域_10ms_rms_envelope_nrmse": 0.07890297574117722, "全帯域_10ms_envelope_correlation": 0.9912926981685153, "全帯域_10ms_modulation_mae_db": 0.4744933852457803, "全帯域_40ms_log_rms_mae_db": 0.4603243975187606, "全帯域_40ms_rms_envelope_nrmse": 0.03826708468643728, "全帯域_40ms_envelope_correlation": 0.9975618880547729, "全帯域_40ms_modulation_mae_db": 0.6007229075662229, "300–1000 Hz_10ms_log_rms_mae_db": 1.8545856119274147, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1061200061865148, "300–1000 Hz_10ms_envelope_correlation": 0.9896210903830044, "300–1000 Hz_10ms_modulation_mae_db": 0.47370590883124003, "300–1000 Hz_40ms_log_rms_mae_db": 1.1122655907884644, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06115906233482488, "300–1000 Hz_40ms_envelope_correlation": 0.9962182748344582, "300–1000 Hz_40ms_modulation_mae_db": 0.7675198380999886, "1000–3000 Hz_10ms_log_rms_mae_db": 2.214323704835938, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22550466281427628, "1000–3000 Hz_10ms_envelope_correlation": 0.9451329493060774, "1000–3000 Hz_10ms_modulation_mae_db": 0.8678535062466048, "1000–3000 Hz_40ms_log_rms_mae_db": 1.132597358342936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1145155648627802, "1000–3000 Hz_40ms_envelope_correlation": 0.9838803938842332, "1000–3000 Hz_40ms_modulation_mae_db": 0.8824771798113525, "3000–6000 Hz_10ms_log_rms_mae_db": 2.544580562390857, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22428518212672338, "3000–6000 Hz_10ms_envelope_correlation": 0.9693300205320278, "3000–6000 Hz_10ms_modulation_mae_db": 1.3929742854070344, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7116587725017698, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13531797434836665, "3000–6000 Hz_40ms_envelope_correlation": 0.9885120934875014, "3000–6000 Hz_40ms_modulation_mae_db": 2.168167269044121, "mr_stft_db": 5.9846427279421635, "band_mae_db": 0.21111810787411, "rms_gain_error_db": -0.014151412540300862, "output_peak": 0.32842808907421556}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0955312670503767, "全帯域_10ms_rms_envelope_nrmse": 0.09138917760340984, "全帯域_10ms_envelope_correlation": 0.9883321885357702, "全帯域_10ms_modulation_mae_db": 0.4658750626758869, "全帯域_40ms_log_rms_mae_db": 0.532897684352495, "全帯域_40ms_rms_envelope_nrmse": 0.048801494332141315, "全帯域_40ms_envelope_correlation": 0.9960327365480535, "全帯域_40ms_modulation_mae_db": 0.46621550083278257, "300–1000 Hz_10ms_log_rms_mae_db": 1.916965396826514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12347627259741575, "300–1000 Hz_10ms_envelope_correlation": 0.9856383533987719, "300–1000 Hz_10ms_modulation_mae_db": 0.19228531998463097, "300–1000 Hz_40ms_log_rms_mae_db": 1.1625326995739342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07069122108077913, "300–1000 Hz_40ms_envelope_correlation": 0.994651890912818, "300–1000 Hz_40ms_modulation_mae_db": 0.5087805406481453, "1000–3000 Hz_10ms_log_rms_mae_db": 2.288612311188321, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24021291342637002, "1000–3000 Hz_10ms_envelope_correlation": 0.9375118208183114, "1000–3000 Hz_10ms_modulation_mae_db": 0.5569107183396694, "1000–3000 Hz_40ms_log_rms_mae_db": 1.239421190097441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13590793108535582, "1000–3000 Hz_40ms_envelope_correlation": 0.9769757627481784, "1000–3000 Hz_40ms_modulation_mae_db": 0.8834926607910706, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4877821918387966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2062716312676854, "3000–6000 Hz_10ms_envelope_correlation": 0.9743080224690094, "3000–6000 Hz_10ms_modulation_mae_db": 1.6455670849874517, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6447588061338978, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13396292632379236, "3000–6000 Hz_40ms_envelope_correlation": 0.9899081042856195, "3000–6000 Hz_40ms_modulation_mae_db": 2.1396181423609635, "mr_stft_db": 5.9904712869185515, "band_mae_db": 0.12651133579602744, "rms_gain_error_db": -0.006871150130695271, "output_peak": 0.3402426801891043}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9957104947479752, "全帯域_10ms_rms_envelope_nrmse": 0.08568195035149484, "全帯域_10ms_envelope_correlation": 0.9897636950888998, "全帯域_10ms_modulation_mae_db": 0.34348827682126676, "全帯域_40ms_log_rms_mae_db": 0.4560719040316165, "全帯域_40ms_rms_envelope_nrmse": 0.04292795310436376, "全帯域_40ms_envelope_correlation": 0.997020708295145, "全帯域_40ms_modulation_mae_db": 0.6348690299435747, "300–1000 Hz_10ms_log_rms_mae_db": 1.9875584497782988, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12021811876913745, "300–1000 Hz_10ms_envelope_correlation": 0.9866149186896733, "300–1000 Hz_10ms_modulation_mae_db": 0.6033381737155602, "300–1000 Hz_40ms_log_rms_mae_db": 1.270823694209037, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06740033595885316, "300–1000 Hz_40ms_envelope_correlation": 0.9955121796821936, "300–1000 Hz_40ms_modulation_mae_db": 0.9105241474631951, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1807579712716487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2104026984086858, "1000–3000 Hz_10ms_envelope_correlation": 0.9526455830726795, "1000–3000 Hz_10ms_modulation_mae_db": 0.8042015293290848, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0744354375272238, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10844295647065945, "1000–3000 Hz_40ms_envelope_correlation": 0.9853335324510233, "1000–3000 Hz_40ms_modulation_mae_db": 0.917293221763277, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5564587427421364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20212676262352203, "3000–6000 Hz_10ms_envelope_correlation": 0.9750436157910495, "3000–6000 Hz_10ms_modulation_mae_db": 1.0946380558725528, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6733802653408285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12228486175244406, "3000–6000 Hz_40ms_envelope_correlation": 0.9907253849182461, "3000–6000 Hz_40ms_modulation_mae_db": 1.0187081447502784, "mr_stft_db": 5.988407698851163, "band_mae_db": 0.11539051117591893, "rms_gain_error_db": 0.00027823711375590923, "output_peak": 0.33928595374487885}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1753980620448539, "全帯域_10ms_rms_envelope_nrmse": 0.08831470926482918, "全帯域_10ms_envelope_correlation": 0.989124453772588, "全帯域_10ms_modulation_mae_db": 0.5508225265011365, "全帯域_40ms_log_rms_mae_db": 0.584386950197269, "全帯域_40ms_rms_envelope_nrmse": 0.04281411227013189, "全帯域_40ms_envelope_correlation": 0.997055123160202, "全帯域_40ms_modulation_mae_db": 0.46120350049932735, "300–1000 Hz_10ms_log_rms_mae_db": 1.9032646261186525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11614308085593865, "300–1000 Hz_10ms_envelope_correlation": 0.9876802774251907, "300–1000 Hz_10ms_modulation_mae_db": 0.7434438683591885, "300–1000 Hz_40ms_log_rms_mae_db": 1.1313867626144793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06015525218194838, "300–1000 Hz_40ms_envelope_correlation": 0.996707429668714, "300–1000 Hz_40ms_modulation_mae_db": 0.7690136559671892, "1000–3000 Hz_10ms_log_rms_mae_db": 2.324920813292407, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2378001587977351, "1000–3000 Hz_10ms_envelope_correlation": 0.9391011848262366, "1000–3000 Hz_10ms_modulation_mae_db": 0.8030986315590823, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3263050353315948, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13590846197738954, "1000–3000 Hz_40ms_envelope_correlation": 0.9770954312678064, "1000–3000 Hz_40ms_modulation_mae_db": 1.288974612871536, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5031466330175207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23263897360354072, "3000–6000 Hz_10ms_envelope_correlation": 0.966265128984766, "3000–6000 Hz_10ms_modulation_mae_db": 1.3054617794791745, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6550375068987815, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13256877137535886, "3000–6000 Hz_40ms_envelope_correlation": 0.9889095202619985, "3000–6000 Hz_40ms_modulation_mae_db": 1.8514257607927027, "mr_stft_db": 6.005302298090457, "band_mae_db": 0.19037394684740763, "rms_gain_error_db": -0.05754242367801018, "output_peak": 0.3784236856815011}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1446970985958613, "全帯域_10ms_rms_envelope_nrmse": 0.09216652398761588, "全帯域_10ms_envelope_correlation": 0.9881810197518324, "全帯域_10ms_modulation_mae_db": 0.1691411195406288, "全帯域_40ms_log_rms_mae_db": 0.597410397671876, "全帯域_40ms_rms_envelope_nrmse": 0.04770477817395967, "全帯域_40ms_envelope_correlation": 0.9962468121594954, "全帯域_40ms_modulation_mae_db": 0.07306072301650568, "300–1000 Hz_10ms_log_rms_mae_db": 1.87789246962429, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11483863860749305, "300–1000 Hz_10ms_envelope_correlation": 0.9877691583443851, "300–1000 Hz_10ms_modulation_mae_db": 0.5521750263572319, "300–1000 Hz_40ms_log_rms_mae_db": 1.1270416593373263, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06597514953113583, "300–1000 Hz_40ms_envelope_correlation": 0.995590877952936, "300–1000 Hz_40ms_modulation_mae_db": 0.5299884687331711, "1000–3000 Hz_10ms_log_rms_mae_db": 2.159991004620407, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23280441882373984, "1000–3000 Hz_10ms_envelope_correlation": 0.9426843538508844, "1000–3000 Hz_10ms_modulation_mae_db": 0.6350061781955898, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1661271596362737, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13123291174153381, "1000–3000 Hz_40ms_envelope_correlation": 0.9787260585946629, "1000–3000 Hz_40ms_modulation_mae_db": 0.7659561636911084, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5192104429661675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20481807287165182, "3000–6000 Hz_10ms_envelope_correlation": 0.9741601916693851, "3000–6000 Hz_10ms_modulation_mae_db": 1.3941193983648914, "3000–6000 Hz_40ms_log_rms_mae_db": 1.655979560646034, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.127955340247499, "3000–6000 Hz_40ms_envelope_correlation": 0.9897856437534579, "3000–6000 Hz_40ms_modulation_mae_db": 1.946401330915098, "mr_stft_db": 5.979897321342901, "band_mae_db": 0.11973287864689695, "rms_gain_error_db": 0.0005508942457521302, "output_peak": 0.35989286215459826}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1735851976736693, "全帯域_10ms_rms_envelope_nrmse": 0.09266944038676721, "全帯域_10ms_envelope_correlation": 0.9880023321912573, "全帯域_10ms_modulation_mae_db": 0.7919711318298365, "全帯域_40ms_log_rms_mae_db": 0.6359318042739585, "全帯域_40ms_rms_envelope_nrmse": 0.05209848615077413, "全帯域_40ms_envelope_correlation": 0.9955696715115996, "全帯域_40ms_modulation_mae_db": 0.7629166764773657, "300–1000 Hz_10ms_log_rms_mae_db": 1.9210586409051544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13416717002934672, "300–1000 Hz_10ms_envelope_correlation": 0.9834190378553924, "300–1000 Hz_10ms_modulation_mae_db": 1.3012527168689927, "300–1000 Hz_40ms_log_rms_mae_db": 1.1883534042302006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08654953930164769, "300–1000 Hz_40ms_envelope_correlation": 0.9925337083595027, "300–1000 Hz_40ms_modulation_mae_db": 1.4411608768414048, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2311175346799685, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23328612216647351, "1000–3000 Hz_10ms_envelope_correlation": 0.9415164737274793, "1000–3000 Hz_10ms_modulation_mae_db": 0.4637508212717883, "1000–3000 Hz_40ms_log_rms_mae_db": 1.157460305864127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11467422125346453, "1000–3000 Hz_40ms_envelope_correlation": 0.9835499023178835, "1000–3000 Hz_40ms_modulation_mae_db": 0.7879740959620234, "3000–6000 Hz_10ms_log_rms_mae_db": 2.514877744170537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21619960989061585, "3000–6000 Hz_10ms_envelope_correlation": 0.9712905273765218, "3000–6000 Hz_10ms_modulation_mae_db": 1.6982550308585382, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6423278303610567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13019927896050199, "3000–6000 Hz_40ms_envelope_correlation": 0.9902050224048624, "3000–6000 Hz_40ms_modulation_mae_db": 2.2086174820525653, "mr_stft_db": 5.9681415412321535, "band_mae_db": 0.22748460802205236, "rms_gain_error_db": -0.07159739530213849, "output_peak": 0.3568328465892686}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.159871826465319, "全帯域_10ms_rms_envelope_nrmse": 0.09541882365417773, "全帯域_10ms_envelope_correlation": 0.9873940089663509, "全帯域_10ms_modulation_mae_db": 0.906583604447865, "全帯域_40ms_log_rms_mae_db": 0.6592293864476825, "全帯域_40ms_rms_envelope_nrmse": 0.049088985332501285, "全帯域_40ms_envelope_correlation": 0.9963074680582713, "全帯域_40ms_modulation_mae_db": 1.177986815775387, "300–1000 Hz_10ms_log_rms_mae_db": 2.0009106279862525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13725025017718817, "300–1000 Hz_10ms_envelope_correlation": 0.9829025557553688, "300–1000 Hz_10ms_modulation_mae_db": 1.3090775369141479, "300–1000 Hz_40ms_log_rms_mae_db": 1.2423636290883284, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08365873002719332, "300–1000 Hz_40ms_envelope_correlation": 0.9935726034723431, "300–1000 Hz_40ms_modulation_mae_db": 1.6191899817352384, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1853828335765755, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22495122171714693, "1000–3000 Hz_10ms_envelope_correlation": 0.9462723015985295, "1000–3000 Hz_10ms_modulation_mae_db": 0.543715003356203, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2917731256287899, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13141555923073955, "1000–3000 Hz_40ms_envelope_correlation": 0.9810859841921649, "1000–3000 Hz_40ms_modulation_mae_db": 0.8961572090111606, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5357645279895764, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23875457825518007, "3000–6000 Hz_10ms_envelope_correlation": 0.9658784730693455, "3000–6000 Hz_10ms_modulation_mae_db": 1.3368785403299557, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7137178041799983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14414770638565697, "3000–6000 Hz_40ms_envelope_correlation": 0.987370134450549, "3000–6000 Hz_40ms_modulation_mae_db": 1.6978715952245704, "mr_stft_db": 5.993904813365916, "band_mae_db": 0.34901683978867776, "rms_gain_error_db": -0.13212263948868685, "output_peak": 0.34585324848730475}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1725821498181392, "全帯域_10ms_rms_envelope_nrmse": 0.09800766458690789, "全帯域_10ms_envelope_correlation": 0.9866317229451608, "全帯域_10ms_modulation_mae_db": 0.2455621017088708, "全帯域_40ms_log_rms_mae_db": 0.5338848338121456, "全帯域_40ms_rms_envelope_nrmse": 0.053377188343278106, "全帯域_40ms_envelope_correlation": 0.9952764091528228, "全帯域_40ms_modulation_mae_db": 0.285189675448573, "300–1000 Hz_10ms_log_rms_mae_db": 1.9590792213728547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13326807979899463, "300–1000 Hz_10ms_envelope_correlation": 0.9833396827355976, "300–1000 Hz_10ms_modulation_mae_db": 0.317348782437759, "300–1000 Hz_40ms_log_rms_mae_db": 1.208752274544867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07958601366437405, "300–1000 Hz_40ms_envelope_correlation": 0.9934418952107374, "300–1000 Hz_40ms_modulation_mae_db": 0.2992181250495565, "1000–3000 Hz_10ms_log_rms_mae_db": 2.26831020322174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24494013459658218, "1000–3000 Hz_10ms_envelope_correlation": 0.9373689396372936, "1000–3000 Hz_10ms_modulation_mae_db": 0.2402813761873248, "1000–3000 Hz_40ms_log_rms_mae_db": 1.210159666411991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13315566072659477, "1000–3000 Hz_40ms_envelope_correlation": 0.9785252510679199, "1000–3000 Hz_40ms_modulation_mae_db": 0.6600610063183864, "3000–6000 Hz_10ms_log_rms_mae_db": 2.53599137717459, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2753231310814182, "3000–6000 Hz_10ms_envelope_correlation": 0.9521929618116783, "3000–6000 Hz_10ms_modulation_mae_db": 0.48509826015523744, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6358245660132575, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1251238923391984, "3000–6000 Hz_40ms_envelope_correlation": 0.9901268751695974, "3000–6000 Hz_40ms_modulation_mae_db": 0.9169802071923078, "mr_stft_db": 6.000231496667123, "band_mae_db": 0.14754525168477867, "rms_gain_error_db": 0.004133617424210173, "output_peak": 0.337897978604562}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44560, "output_samples": 44560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.278483182954355, "spectral_convergence": 0.32205025983424457, "spectral_convergence_status": "ok", "amplitude_floor": 9.278908430047933e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.875416418249237, "spectral_convergence": 0.30073980499119785, "spectral_convergence_status": "ok", "amplitude_floor": 4.74996132181228e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.800028582622899, "spectral_convergence": 0.3032795606831897, "spectral_convergence_status": "ok", "amplitude_floor": 3.301834166389807e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005909300513360037, "output_power": 0.0006070450826913421, "delta_db": 0.11684869217469633, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000598279616407713, "output_power": 0.00060345479936008, "delta_db": 0.03740539113444263, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005543526355345429, "output_power": 0.0005340606381853687, "delta_db": -0.16195546228404425, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011395641530102642, "output_power": 0.00010757305613738633, "delta_db": -0.25035272233831984, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011253089498164014, "output_power": 0.00010915159305135934, "delta_db": -0.132416943634153, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6482875672349115e-05, "output_power": 1.8784771922540402e-05, "delta_db": 0.5677294356790039, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21111810787411, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.014151412540300862, "pitch": null, "output_rms_dbfs": -27.033194690288454, "output_peak": 0.32842808907421556, "reference_rms_dbfs": -27.019043277748153, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9846427279421635}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 0.00011061650326899306, "log_rms_mae_db": 1.042029240111435, "rms_envelope_nrmse": 0.07890297574117722, "envelope_correlation": 0.9912926981685153, "rms_gain_error_db": -0.014105522075994055, "reference_mean_power": 0.00199005080866686, "output_mean_power": 0.0019835977749672}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 8.439637043999617e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970864143567599, "output_power": 0.0005774595104696196, "reference_active": true, "delta_db": -0.14515651338377325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010123102438186569, "output_power": 9.164656847042767e-05, "reference_active": true, "delta_db": -0.4319742332373528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240115070482855e-05, "output_power": 2.666401182425417e-05, "reference_active": true, "delta_db": -0.8463494091162149}], "active_band_count": 3, "active_band_mae_db": 0.4744933852457803}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 9.74750993349889e-05, "log_rms_mae_db": 0.4603243975187606, "rms_envelope_nrmse": 0.03826708468643728, "envelope_correlation": 0.9975618880547729, "rms_gain_error_db": -0.013910607991956387, "reference_mean_power": 0.0020114637230104894, "output_mean_power": 0.002005031239811896}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 6.839204647135984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005172552173341314, "output_power": 0.0005018811326639506, "reference_active": true, "delta_db": -0.13104010450400894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6641602045504645e-05, "output_power": 5.1605335288770826e-05, "reference_active": true, "delta_db": -0.4044092431540651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.529860845744804e-06, "output_power": 1.8898385167483155e-06, "reference_active": true, "delta_db": -1.2667193750405947}], "active_band_count": 3, "active_band_mae_db": 0.6007229075662229}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 8.308955229004579e-05, "log_rms_mae_db": 1.8545856119274147, "rms_envelope_nrmse": 0.1061200061865148, "envelope_correlation": 0.9896210903830044, "rms_gain_error_db": -0.10610443358197448, "reference_mean_power": 0.000839400894204417, "output_mean_power": 0.0008191416039324629}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.694834819211473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042315114528166547, "output_power": 0.00040196288336372414, "reference_active": true, "delta_db": -0.2230956780502103}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.939611213191948e-05, "output_power": 6.00342054132548e-05, "reference_active": true, "delta_db": -0.6293637322122374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4277196064029695e-05, "output_power": 2.1297699452703652e-05, "reference_active": true, "delta_db": -0.5686583162312724}], "active_band_count": 3, "active_band_mae_db": 0.47370590883124003}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 8.008999734930271e-05, "log_rms_mae_db": 1.1122655907884644, "rms_envelope_nrmse": 0.06115906233482488, "envelope_correlation": 0.9962182748344582, "rms_gain_error_db": -0.10611589402758713, "reference_mean_power": 0.0008485551352947845, "output_mean_power": 0.000828072718425403}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 5.733703583211585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003764127527272362, "output_power": 0.0003607645897258485, "reference_active": true, "delta_db": -0.18440425265566085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8963437868750806e-05, "output_power": 3.428843654487259e-05, "reference_active": true, "delta_db": -0.5550958611449884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.296630500499659e-06, "output_power": 1.6024525854072178e-06, "reference_active": true, "delta_db": -1.5630594004993164}], "active_band_count": 3, "active_band_mae_db": 0.7675198380999886}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.2528786526254916e-05, "log_rms_mae_db": 2.214323704835938, "rms_envelope_nrmse": 0.22550466281427628, "envelope_correlation": 0.9451329493060774, "rms_gain_error_db": -0.164680501405569, "reference_mean_power": 0.0001368277838750774, "output_mean_power": 0.00013173653700824245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.073610600879068e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.932287229550182e-05, "output_power": 3.7389080225460366e-05, "reference_active": true, "delta_db": -0.21900451271865853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8616503346595646e-06, "output_power": 5.722388533829385e-06, "reference_active": true, "delta_db": -0.7885124108859266}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.317401673918684e-06, "output_power": 2.989634496449762e-06, "reference_active": true, "delta_db": -1.5960435951352292}], "active_band_count": 3, "active_band_mae_db": 0.8678535062466048}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 3.268734932259802e-05, "log_rms_mae_db": 1.132597358342936, "rms_envelope_nrmse": 0.1145155648627802, "envelope_correlation": 0.9838803938842332, "rms_gain_error_db": -0.16473051810416514, "reference_mean_power": 0.00013831808192096785, "output_mean_power": 0.00013316984862067561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 4.85870788941741e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5019378943818874e-05, "output_power": 3.363041142370168e-05, "reference_active": true, "delta_db": -0.1757625939117906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.168829067205367e-06, "output_power": 3.390252602987447e-06, "reference_active": true, "delta_db": -0.8978203020813427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163313597471507e-07, "output_power": 2.2016961541859717e-07, "reference_active": true, "delta_db": -1.573848643440924}], "active_band_count": 3, "active_band_mae_db": 0.8824771798113525}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.9606629692154165e-05, "log_rms_mae_db": 2.544580562390857, "rms_envelope_nrmse": 0.22428518212672338, "envelope_correlation": 0.9693300205320278, "rms_gain_error_db": 0.12203937714628055, "reference_mean_power": 5.1065139870557587e-05, "output_mean_power": 5.252045302467518e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 2.3044643706212192e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0452875124351046e-05, "output_power": 1.0498637628092722e-05, "reference_active": true, "delta_db": 0.01897183720663146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.855628387226494e-06, "output_power": 2.557405578583236e-06, "reference_active": true, "delta_db": -0.4790208235415661}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6522678437727886e-06, "output_power": 7.079252962307716e-07, "reference_active": true, "delta_db": -3.680930195472905}], "active_band_count": 3, "active_band_mae_db": 1.3929742854070344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 4.1718716562101114e-05, "log_rms_mae_db": 1.7116587725017698, "rms_envelope_nrmse": 0.13531797434836665, "envelope_correlation": 0.9885120934875014, "rms_gain_error_db": 0.1220015444586923, "reference_mean_power": 5.162171435861935e-05, "output_mean_power": 5.3092426907376006e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 2.066669918587524e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0052154991285675e-05, "output_power": 9.899542306224305e-06, "reference_active": true, "delta_db": -0.06644060270232072}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.890809551942203e-06, "output_power": 1.3056985351964082e-06, "reference_active": true, "delta_db": -1.6080487069598168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8144313206293327e-07, "output_power": 5.9667698187211e-08, "reference_active": true, "delta_db": -4.830012497470227}], "active_band_count": 3, "active_band_mae_db": 2.168167269044121}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8291746650209392, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8f7c72282ef64524dbc2301bb144b2b52b9b8c7393bf24a9ea043eb6b3600992", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9427792008546865, "standard_deviation": 0.005171831385994004, "minimum": 0.9373689396372936, "maximum": 0.9526455830726795, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.231677047085876, "standard_deviation": 0.05788444039046929, "minimum": 2.159991004620407, "maximum": 2.324920813292407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6143522205606684, "standard_deviation": 0.20950839608978047, "minimum": 0.2402813761873248, "maximum": 0.8678535062466048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2312377913438762, "standard_deviation": 0.010852994928696779, "minimum": 0.2104026984086858, "maximum": 0.24494013459658218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.980646539565484, "standard_deviation": 0.003280274978661196, "minimum": 0.9769757627481784, "maximum": 0.9853335324510233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.199784909855047, "standard_deviation": 0.0839206409317256, "minimum": 1.0744354375272238, "maximum": 1.3263050353315948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8852982687774894, "standard_deviation": 0.18460527782448466, "minimum": 0.6600610063183864, "maximum": 1.288974612871536, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12565665841856472, "standard_deviation": 0.011159442313608547, "minimum": 0.10844295647065945, "maximum": 0.13590846197738954, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.968558617712973, "standard_deviation": 0.007510229571974656, "minimum": 0.9521929618116783, "maximum": 0.9750436157910495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.5247265277862727, "standard_deviation": 0.022702586333459677, "minimum": 2.4877821918387966, "maximum": 2.5564587427421364, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2941240544318546, "standard_deviation": 0.37850804413146555, "minimum": 0.48509826015523744, "maximum": 1.6982550308585382, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22505224271504218, "standard_deviation": 0.024316786749834157, "minimum": 0.20212676262352203, "maximum": 0.2753231310814182, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9894428473414791, "standard_deviation": 0.0011002960754444352, "minimum": 0.987370134450549, "maximum": 0.9907253849182461, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.666585639009453, "standard_deviation": 0.030603779137868853, "minimum": 1.6358245660132575, "maximum": 1.7137178041799983, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.743473741541576, "standard_deviation": 0.5094735674190866, "minimum": 0.9169802071923078, "maximum": 2.2086174820525653, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13144509396660228, "standard_deviation": 0.006777647882130709, "minimum": 0.12228486175244406, "maximum": 0.14414770638565697, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.985873134323423, "standard_deviation": 0.002472087535514307, "minimum": 0.9829025557553688, "maximum": 0.9896210903830044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.927664380567429, "standard_deviation": 0.05142318282366623, "minimum": 1.8545856119274147, "maximum": 2.0009106279862525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.686578416683594, "standard_deviation": 0.41739454302340234, "minimum": 0.19228531998463097, "maximum": 1.3090775369141479, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12318520212775366, "standard_deviation": 0.010950925796157196, "minimum": 0.1061200061865148, "maximum": 0.13725025017718817, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947786075117129, "standard_deviation": 0.001477825774145069, "minimum": 0.9925337083595027, "maximum": 0.996707429668714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1804399642983296, "standard_deviation": 0.0574381396910305, "minimum": 1.1122655907884644, "maximum": 1.270823694209037, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8556744543172361, "standard_deviation": 0.4598276891038962, "minimum": 0.2992181250495565, "maximum": 1.6191899817352384, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07189691301009456, "standard_deviation": 0.010156421492194537, "minimum": 0.06015525218194838, "maximum": 0.08654953930164769, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18589668497948372, "standard_deviation": 0.07868574770282405, "minimum": 0.11539051117591893, "maximum": 0.34901683978867776, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.988874898051304, "standard_deviation": 0.011681969116827621, "minimum": 5.9681415412321535, "maximum": 6.005302298090457, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3483571680656792, "standard_deviation": 0.01590089456481871, "minimum": 0.32842808907421556, "maximum": 0.3784236856815011, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03466528404451418, "standard_deviation": 0.048616864994903294, "minimum": -0.13212263948868685, "maximum": 0.004133617424210173, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9885902649275469, "standard_deviation": 0.0014546529507232493, "minimum": 0.9866317229451608, "maximum": 0.9912926981685153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1199256670634536, "standard_deviation": 0.06869280019951438, "minimum": 0.9957104947479752, "maximum": 1.1753980620448539, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4934921510964089, "standard_deviation": 0.2544795087879133, "minimum": 0.1691411195406288, "maximum": 0.906583604447865, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09031890819704746, "standard_deviation": 0.005988703413660065, "minimum": 0.07890297574117722, "maximum": 0.09800766458690789, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963838521175452, "standard_deviation": 0.0007820463879208606, "minimum": 0.9952764091528228, "maximum": 0.9975618880547729, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5575171697882254, "standard_deviation": 0.07531824664351933, "minimum": 0.4560719040316165, "maximum": 0.6592293864476825, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5577706036949673, "standard_deviation": 0.3298254304052336, "minimum": 0.07306072301650568, "maximum": 1.177986815775387, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.046885010299198435, "standard_deviation": 0.005141293567753525, "minimum": 0.03826708468643728, "maximum": 0.053377188343278106, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0042", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.7993838787078857, "predicted_mos": 2.7993838787078857, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "c23ac457bb60ef5fc4a6874a179d4839b8e37711f9a8fea6645d16f5c373ff25", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9333664998412132, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.197728320452365, "全帯域_10ms_rms_envelope_nrmse": 0.09295714954815444, "全帯域_10ms_envelope_correlation": 0.9879001167789475, "全帯域_10ms_modulation_mae_db": 0.7867202871171566, "全帯域_40ms_log_rms_mae_db": 0.6229081523308913, "全帯域_40ms_rms_envelope_nrmse": 0.05266334560645114, "全帯域_40ms_envelope_correlation": 0.995386675133555, "全帯域_40ms_modulation_mae_db": 0.9079867747759813, "300–1000 Hz_10ms_log_rms_mae_db": 1.5773007217486719, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12516379818072726, "300–1000 Hz_10ms_envelope_correlation": 0.9858153219447258, "300–1000 Hz_10ms_modulation_mae_db": 1.323573492163879, "300–1000 Hz_40ms_log_rms_mae_db": 0.9030189700169687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08130116330963938, "300–1000 Hz_40ms_envelope_correlation": 0.9937719276591622, "300–1000 Hz_40ms_modulation_mae_db": 1.6921426217445938, "1000–3000 Hz_10ms_log_rms_mae_db": 1.688639705099921, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21774682621881244, "1000–3000 Hz_10ms_envelope_correlation": 0.949827928337275, "1000–3000 Hz_10ms_modulation_mae_db": 0.5871769243632897, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9731279963326208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1273853187153259, "1000–3000 Hz_40ms_envelope_correlation": 0.980113395200871, "1000–3000 Hz_40ms_modulation_mae_db": 0.5424475946690747, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8270249928852584, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17402670700807046, "3000–6000 Hz_10ms_envelope_correlation": 0.9812085052481022, "3000–6000 Hz_10ms_modulation_mae_db": 0.322517504337098, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3429701437779196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10932726907972358, "3000–6000 Hz_40ms_envelope_correlation": 0.9922373541408983, "3000–6000 Hz_40ms_modulation_mae_db": 0.4626077860314535, "mr_stft_db": 5.634968023933957, "band_mae_db": 0.19820490801219823, "rms_gain_error_db": -0.033710482581080684, "output_peak": 0.37051862731257545}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2530437312097265, "全帯域_10ms_rms_envelope_nrmse": 0.10067324423623857, "全帯域_10ms_envelope_correlation": 0.9857918689073412, "全帯域_10ms_modulation_mae_db": 0.8204768507491184, "全帯域_40ms_log_rms_mae_db": 0.543597003814144, "全帯域_40ms_rms_envelope_nrmse": 0.056495784105970503, "全帯域_40ms_envelope_correlation": 0.9947093543890191, "全帯域_40ms_modulation_mae_db": 0.788632309293765, "300–1000 Hz_10ms_log_rms_mae_db": 1.5706502487706933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12422466548883021, "300–1000 Hz_10ms_envelope_correlation": 0.9858026156306514, "300–1000 Hz_10ms_modulation_mae_db": 1.0480294808228174, "300–1000 Hz_40ms_log_rms_mae_db": 0.8787391621832111, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0671733341987639, "300–1000 Hz_40ms_envelope_correlation": 0.9957720319828458, "300–1000 Hz_40ms_modulation_mae_db": 1.0024507670301583, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6851350665643698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2074914118958135, "1000–3000 Hz_10ms_envelope_correlation": 0.9547522308796854, "1000–3000 Hz_10ms_modulation_mae_db": 0.688438599279355, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9589004415523804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12608356595077327, "1000–3000 Hz_40ms_envelope_correlation": 0.9808330098661658, "1000–3000 Hz_40ms_modulation_mae_db": 0.9582392554411018, "3000–6000 Hz_10ms_log_rms_mae_db": 1.829104065641713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26349654387307, "3000–6000 Hz_10ms_envelope_correlation": 0.969803663531539, "3000–6000 Hz_10ms_modulation_mae_db": 0.8549193534368006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3436639507256256, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19109817449895627, "3000–6000 Hz_40ms_envelope_correlation": 0.9875007025338585, "3000–6000 Hz_40ms_modulation_mae_db": 1.13435030344378, "mr_stft_db": 5.637551790275172, "band_mae_db": 0.2839136083070033, "rms_gain_error_db": -0.05343828266150241, "output_peak": 0.326303668348972}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2073086925937313, "全帯域_10ms_rms_envelope_nrmse": 0.09496532967340628, "全帯域_10ms_envelope_correlation": 0.9873765597039106, "全帯域_10ms_modulation_mae_db": 0.6282798531794062, "全帯域_40ms_log_rms_mae_db": 0.6147510498688192, "全帯域_40ms_rms_envelope_nrmse": 0.04940589854067188, "全帯域_40ms_envelope_correlation": 0.996002616286872, "全帯域_40ms_modulation_mae_db": 0.5863133341230888, "300–1000 Hz_10ms_log_rms_mae_db": 1.598426446610979, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11713211530161222, "300–1000 Hz_10ms_envelope_correlation": 0.9871141188560153, "300–1000 Hz_10ms_modulation_mae_db": 0.7728322295164037, "300–1000 Hz_40ms_log_rms_mae_db": 0.9605379435985772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06719207720919995, "300–1000 Hz_40ms_envelope_correlation": 0.9952383462151779, "300–1000 Hz_40ms_modulation_mae_db": 0.9871488423991218, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7573307460944014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21199053505404677, "1000–3000 Hz_10ms_envelope_correlation": 0.9529971685711529, "1000–3000 Hz_10ms_modulation_mae_db": 0.5979991707008776, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9953688365079554, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12344862930675957, "1000–3000 Hz_40ms_envelope_correlation": 0.9842328102492761, "1000–3000 Hz_40ms_modulation_mae_db": 0.7020236287635715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7547313793811476, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1931792485063245, "3000–6000 Hz_10ms_envelope_correlation": 0.9780601972151594, "3000–6000 Hz_10ms_modulation_mae_db": 0.753942597143018, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3278064158789986, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12471127438516173, "3000–6000 Hz_40ms_envelope_correlation": 0.9907573120925757, "3000–6000 Hz_40ms_modulation_mae_db": 0.8166143969217247, "mr_stft_db": 5.6205073828097225, "band_mae_db": 0.16837912099860367, "rms_gain_error_db": -0.06842861215038504, "output_peak": 0.3637941582727017}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2341749249559586, "全帯域_10ms_rms_envelope_nrmse": 0.08549792721583357, "全帯域_10ms_envelope_correlation": 0.989798139555876, "全帯域_10ms_modulation_mae_db": 0.4567647900721987, "全帯域_40ms_log_rms_mae_db": 0.5282802945134951, "全帯域_40ms_rms_envelope_nrmse": 0.04022607581296078, "全帯域_40ms_envelope_correlation": 0.9973200953403273, "全帯域_40ms_modulation_mae_db": 0.537545450790401, "300–1000 Hz_10ms_log_rms_mae_db": 1.634551384990824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11607791417927793, "300–1000 Hz_10ms_envelope_correlation": 0.9874282288559371, "300–1000 Hz_10ms_modulation_mae_db": 0.6234848337535158, "300–1000 Hz_40ms_log_rms_mae_db": 1.0408693541760123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07002564800767401, "300–1000 Hz_40ms_envelope_correlation": 0.9949586862965757, "300–1000 Hz_40ms_modulation_mae_db": 0.9071078062452118, "1000–3000 Hz_10ms_log_rms_mae_db": 1.689326443766342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2086904001250315, "1000–3000 Hz_10ms_envelope_correlation": 0.953840372653679, "1000–3000 Hz_10ms_modulation_mae_db": 0.4601503781154439, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9874341080111517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11172382468853485, "1000–3000 Hz_40ms_envelope_correlation": 0.9848934675269595, "1000–3000 Hz_40ms_modulation_mae_db": 0.44403264886351484, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8354371951057329, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1810573051168645, "3000–6000 Hz_10ms_envelope_correlation": 0.9805358590690846, "3000–6000 Hz_10ms_modulation_mae_db": 0.7341519402144684, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3928292998525715, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11419084721068942, "3000–6000 Hz_40ms_envelope_correlation": 0.9921503873586774, "3000–6000 Hz_40ms_modulation_mae_db": 0.753734090800173, "mr_stft_db": 5.634687533885437, "band_mae_db": 0.17073534901426368, "rms_gain_error_db": -0.025785255703439404, "output_peak": 0.3409963898347532}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.162556098662642, "全帯域_10ms_rms_envelope_nrmse": 0.08984671835353421, "全帯域_10ms_envelope_correlation": 0.9890431190591676, "全帯域_10ms_modulation_mae_db": 0.393978176521012, "全帯域_40ms_log_rms_mae_db": 0.6996432256012539, "全帯域_40ms_rms_envelope_nrmse": 0.05287237833024779, "全帯域_40ms_envelope_correlation": 0.9958676416403299, "全帯域_40ms_modulation_mae_db": 0.3292139723627039, "300–1000 Hz_10ms_log_rms_mae_db": 1.5620645488397684, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12046927312185142, "300–1000 Hz_10ms_envelope_correlation": 0.987244857242247, "300–1000 Hz_10ms_modulation_mae_db": 0.7143481683625872, "300–1000 Hz_40ms_log_rms_mae_db": 0.8709087481409097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07784413655650317, "300–1000 Hz_40ms_envelope_correlation": 0.9948185294602099, "300–1000 Hz_40ms_modulation_mae_db": 0.8604663713136885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.766625429666833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22557565277492694, "1000–3000 Hz_10ms_envelope_correlation": 0.9455100503816778, "1000–3000 Hz_10ms_modulation_mae_db": 0.1708273036632756, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0164288602181217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12109636925964265, "1000–3000 Hz_40ms_envelope_correlation": 0.9819949713441238, "1000–3000 Hz_40ms_modulation_mae_db": 0.24353165898999116, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8039039099065697, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20517868895287394, "3000–6000 Hz_10ms_envelope_correlation": 0.9760160903324475, "3000–6000 Hz_10ms_modulation_mae_db": 1.0459029800629933, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2961543517834642, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13838617643561518, "3000–6000 Hz_40ms_envelope_correlation": 0.9913504780785467, "3000–6000 Hz_40ms_modulation_mae_db": 1.4374569689381584, "mr_stft_db": 5.6270528014918435, "band_mae_db": 0.27058543116269956, "rms_gain_error_db": -0.15587472359494614, "output_peak": 0.3484533868092553}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2019314092495794, "全帯域_10ms_rms_envelope_nrmse": 0.09646998525068579, "全帯域_10ms_envelope_correlation": 0.9869874446465138, "全帯域_10ms_modulation_mae_db": 0.36448463290103583, "全帯域_40ms_log_rms_mae_db": 0.6100755747475198, "全帯域_40ms_rms_envelope_nrmse": 0.054829097681980944, "全帯域_40ms_envelope_correlation": 0.995089874797384, "全帯域_40ms_modulation_mae_db": 0.5948371088979302, "300–1000 Hz_10ms_log_rms_mae_db": 1.5091699872948032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10813075398610834, "300–1000 Hz_10ms_envelope_correlation": 0.9892873708967966, "300–1000 Hz_10ms_modulation_mae_db": 0.5446680522701833, "300–1000 Hz_40ms_log_rms_mae_db": 0.8951558551775519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06086354930692449, "300–1000 Hz_40ms_envelope_correlation": 0.9964533824051263, "300–1000 Hz_40ms_modulation_mae_db": 0.7711217968788864, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7326655596996676, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21836792322815018, "1000–3000 Hz_10ms_envelope_correlation": 0.9490377454478026, "1000–3000 Hz_10ms_modulation_mae_db": 0.3740949429455889, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0112447347985145, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13374504140658702, "1000–3000 Hz_40ms_envelope_correlation": 0.9788783703579468, "1000–3000 Hz_40ms_modulation_mae_db": 0.4950663792860264, "3000–6000 Hz_10ms_log_rms_mae_db": 1.802132784882309, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2188793665745957, "3000–6000 Hz_10ms_envelope_correlation": 0.9739311786526907, "3000–6000 Hz_10ms_modulation_mae_db": 0.8256293175121843, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3444430672790932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15961390097235276, "3000–6000 Hz_40ms_envelope_correlation": 0.9893774643205394, "3000–6000 Hz_40ms_modulation_mae_db": 0.7708480662682152, "mr_stft_db": 5.6387874864791, "band_mae_db": 0.32335963506303256, "rms_gain_error_db": -0.08301877753617609, "output_peak": 0.35969171781142295}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1081841167255866, "全帯域_10ms_rms_envelope_nrmse": 0.09081380515625975, "全帯域_10ms_envelope_correlation": 0.9885076027170775, "全帯域_10ms_modulation_mae_db": 0.6096398411312366, "全帯域_40ms_log_rms_mae_db": 0.5125781470379639, "全帯域_40ms_rms_envelope_nrmse": 0.04301947775340593, "全帯域_40ms_envelope_correlation": 0.9970361269980261, "全帯域_40ms_modulation_mae_db": 0.6159194386741343, "300–1000 Hz_10ms_log_rms_mae_db": 1.554385341131219, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11397461780295494, "300–1000 Hz_10ms_envelope_correlation": 0.9880171495659932, "300–1000 Hz_10ms_modulation_mae_db": 0.7438835074255659, "300–1000 Hz_40ms_log_rms_mae_db": 0.9233295127574678, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06383259498683753, "300–1000 Hz_40ms_envelope_correlation": 0.9960701838936001, "300–1000 Hz_40ms_modulation_mae_db": 0.8220804476689981, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7333597060740131, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2186512538449771, "1000–3000 Hz_10ms_envelope_correlation": 0.9493055774231797, "1000–3000 Hz_10ms_modulation_mae_db": 0.4458450391986648, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9511515154951402, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11465298435182085, "1000–3000 Hz_40ms_envelope_correlation": 0.9858803484862488, "1000–3000 Hz_40ms_modulation_mae_db": 0.7029141757187204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8597795822567005, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21351270846474593, "3000–6000 Hz_10ms_envelope_correlation": 0.9714989335664501, "3000–6000 Hz_10ms_modulation_mae_db": 0.8155440288820982, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3458353711579436, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12550801375988396, "3000–6000 Hz_40ms_envelope_correlation": 0.989736252415265, "3000–6000 Hz_40ms_modulation_mae_db": 0.8171866203943732, "mr_stft_db": 5.627995887484748, "band_mae_db": 0.2631048423036019, "rms_gain_error_db": -0.08190148991144187, "output_peak": 0.34366888581952604}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1954994845692177, "全帯域_10ms_rms_envelope_nrmse": 0.09583053127611263, "全帯域_10ms_envelope_correlation": 0.9873437369071888, "全帯域_10ms_modulation_mae_db": 0.7287752312530501, "全帯域_40ms_log_rms_mae_db": 0.5836993735476957, "全帯域_40ms_rms_envelope_nrmse": 0.04748074021758877, "全帯域_40ms_envelope_correlation": 0.9963975705826964, "全帯域_40ms_modulation_mae_db": 0.7155531486004033, "300–1000 Hz_10ms_log_rms_mae_db": 1.72966703506926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12164471179139302, "300–1000 Hz_10ms_envelope_correlation": 0.9860966614821911, "300–1000 Hz_10ms_modulation_mae_db": 0.508648324520041, "300–1000 Hz_40ms_log_rms_mae_db": 1.0455226065644065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06586161615797138, "300–1000 Hz_40ms_envelope_correlation": 0.9954160351501259, "300–1000 Hz_40ms_modulation_mae_db": 0.6503430671986493, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6595450854756202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21813315434540692, "1000–3000 Hz_10ms_envelope_correlation": 0.9490540085060386, "1000–3000 Hz_10ms_modulation_mae_db": 0.5493835984261848, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8969849496561175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13051054515940708, "1000–3000 Hz_40ms_envelope_correlation": 0.9790146057597169, "1000–3000 Hz_40ms_modulation_mae_db": 0.5093335863864535, "3000–6000 Hz_10ms_log_rms_mae_db": 1.821847953145177, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1967329696151489, "3000–6000 Hz_10ms_envelope_correlation": 0.9823335314219479, "3000–6000 Hz_10ms_modulation_mae_db": 0.7639389886119554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.379808875233617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14185729304390945, "3000–6000 Hz_40ms_envelope_correlation": 0.9925820545240672, "3000–6000 Hz_40ms_modulation_mae_db": 0.7542910643118258, "mr_stft_db": 5.625591061369637, "band_mae_db": 0.2670622783409307, "rms_gain_error_db": 0.06623525352833876, "output_peak": 0.3557594562848806}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44560, "output_samples": 44560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.807175365952109, "spectral_convergence": 0.32097510985035826, "spectral_convergence_status": "ok", "amplitude_floor": 9.278908430047933e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.409389535176486, "spectral_convergence": 0.30270334167353663, "spectral_convergence_status": "ok", "amplitude_floor": 4.74996132181228e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.688339170673273, "spectral_convergence": 0.3110708926787253, "spectral_convergence_status": "ok", "amplitude_floor": 3.301834166389807e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005909300513360037, "output_power": 0.0006167118348046465, "delta_db": 0.18546206406470273, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000598279616407713, "output_power": 0.0005926169121750697, "delta_db": -0.04130165260975442, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005543526355345429, "output_power": 0.0005223633771210944, "delta_db": -0.25813395654139143, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011395641530102642, "output_power": 0.00010322414882394565, "delta_db": -0.42957468874989185, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011253089498164014, "output_power": 0.00011978623683412556, "delta_db": 0.2713514855403274, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6482875672349115e-05, "output_power": 1.6469955386797623e-05, "delta_db": -0.0034056005671215034, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19820490801219823, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.033710482581080684, "pitch": null, "output_rms_dbfs": -27.05275376032923, "output_peak": 0.37051862731257545, "reference_rms_dbfs": -27.019043277748153, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.634968023933957}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 0.00011061650326899306, "log_rms_mae_db": 1.197728320452365, "rms_envelope_nrmse": 0.09295714954815444, "envelope_correlation": 0.9879001167789475, "rms_gain_error_db": -0.03364259169662461, "reference_mean_power": 0.00199005080866686, "output_mean_power": 0.001974694449737669}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 8.439637043999617e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970864143567599, "output_power": 0.0005841557269431937, "reference_active": true, "delta_db": -0.09508550908279291}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010123102438186569, "output_power": 9.605480246571385e-05, "reference_active": true, "delta_db": -0.22794548380486768}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240115070482855e-05, "output_power": 2.026970589503631e-05, "reference_active": true, "delta_db": -2.037129868463809}], "active_band_count": 3, "active_band_mae_db": 0.7867202871171566}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 9.74750993349889e-05, "log_rms_mae_db": 0.6229081523308913, "rms_envelope_nrmse": 0.05266334560645114, "envelope_correlation": 0.995386675133555, "rms_gain_error_db": -0.033492682429582976, "reference_mean_power": 0.0020114637230104894, "output_mean_power": 0.001996011026797618}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 6.839204647135984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005172552173341314, "output_power": 0.0005086187525180187, "reference_active": true, "delta_db": -0.07312511549862588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6641602045504645e-05, "output_power": 5.109807128259348e-05, "reference_active": true, "delta_db": -0.4473102057499237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.529860845744804e-06, "output_power": 1.5231551151208606e-06, "reference_active": true, "delta_db": -2.2035250030793945}], "active_band_count": 3, "active_band_mae_db": 0.9079867747759813}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 8.308955229004579e-05, "log_rms_mae_db": 1.5773007217486719, "rms_envelope_nrmse": 0.12516379818072726, "envelope_correlation": 0.9858153219447258, "rms_gain_error_db": -0.27974897743323895, "reference_mean_power": 0.000839400894204417, "output_mean_power": 0.0007870358764092242}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.694834819211473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042315114528166547, "output_power": 0.0003992363667077419, "reference_active": true, "delta_db": -0.2526542606221827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.939611213191948e-05, "output_power": 5.48055945726137e-05, "reference_active": true, "delta_db": -1.0251024642850717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4277196064029695e-05, "output_power": 1.3058766322876779e-05, "reference_active": true, "delta_db": -2.692963751584383}], "active_band_count": 3, "active_band_mae_db": 1.323573492163879}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 8.008999734930271e-05, "log_rms_mae_db": 0.9030189700169687, "rms_envelope_nrmse": 0.08130116330963938, "envelope_correlation": 0.9937719276591622, "rms_gain_error_db": -0.27976140059416366, "reference_mean_power": 0.0008485551352947845, "output_mean_power": 0.0007956167652554257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 5.733703583211585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003764127527272362, "output_power": 0.0003564925801220403, "reference_active": true, "delta_db": -0.23613833673830967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8963437868750806e-05, "output_power": 2.94405673648519e-05, "reference_active": true, "delta_db": -1.2171109382210543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.296630500499659e-06, "output_power": 9.9717933344516e-07, "reference_active": true, "delta_db": -3.623178590274417}], "active_band_count": 3, "active_band_mae_db": 1.6921426217445938}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.2528786526254916e-05, "log_rms_mae_db": 1.688639705099921, "rms_envelope_nrmse": 0.21774682621881244, "envelope_correlation": 0.949827928337275, "rms_gain_error_db": -0.07269919432189322, "reference_mean_power": 0.0001368277838750774, "output_mean_power": 0.00013455640443425814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.073610600879068e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.932287229550182e-05, "output_power": 3.7906846741491714e-05, "reference_active": true, "delta_db": -0.15927573475792114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8616503346595646e-06, "output_power": 8.078864474030252e-06, "reference_active": true, "delta_db": 0.7092173982686295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.317401673918684e-06, "output_power": 3.5149462768736507e-06, "reference_active": true, "delta_db": -0.8930376400633184}], "active_band_count": 3, "active_band_mae_db": 0.5871769243632897}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 3.268734932259802e-05, "log_rms_mae_db": 0.9731279963326208, "rms_envelope_nrmse": 0.1273853187153259, "envelope_correlation": 0.980113395200871, "rms_gain_error_db": -0.0727157676241578, "reference_mean_power": 0.00013831808192096785, "output_mean_power": 0.0001360214440378197}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 4.85870788941741e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5019378943818874e-05, "output_power": 3.434886872415254e-05, "reference_active": true, "delta_db": -0.08396001563422623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.168829067205367e-06, "output_power": 4.600851014774619e-06, "reference_active": true, "delta_db": 0.4282408183222649}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163313597471507e-07, "output_power": 2.4469666658547546e-07, "reference_active": true, "delta_db": -1.115141950050733}], "active_band_count": 3, "active_band_mae_db": 0.5424475946690747}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.9606629692154165e-05, "log_rms_mae_db": 1.8270249928852584, "rms_envelope_nrmse": 0.17402670700807046, "envelope_correlation": 0.9812085052481022, "rms_gain_error_db": 0.005177639908723513, "reference_mean_power": 5.1065139870557587e-05, "output_mean_power": 5.112605581267375e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 2.3044643706212192e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0452875124351046e-05, "output_power": 1.0336088102304646e-05, "reference_active": true, "delta_db": -0.04879559624708446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.855628387226494e-06, "output_power": 3.077576384237786e-06, "reference_active": true, "delta_db": 0.3250715021371913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6522678437727886e-06, "output_power": 1.4411590366835466e-06, "reference_active": true, "delta_db": -0.5936854146270183}], "active_band_count": 3, "active_band_mae_db": 0.322517504337098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 4.1718716562101114e-05, "log_rms_mae_db": 1.3429701437779196, "rms_envelope_nrmse": 0.10932726907972358, "envelope_correlation": 0.9922373541408983, "rms_gain_error_db": 0.0051691118025405415, "reference_mean_power": 5.162171435861935e-05, "output_mean_power": 5.1683192753328875e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 2.066669918587524e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0052154991285675e-05, "output_power": 9.933990309141262e-06, "reference_active": true, "delta_db": -0.05135444239664542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.890809551942203e-06, "output_power": 1.9003787806820154e-06, "reference_active": true, "delta_db": 0.02192384919324457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8144313206293327e-07, "output_power": 1.34055929234991e-07, "reference_active": true, "delta_db": -1.3145450665044707}], "active_band_count": 3, "active_band_mae_db": 0.4626077860314535}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8421616969169508, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6212cefb836104b5cb49accdd3c9b13872e29db7f765d5300e71e5194e0de8d3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9505406352750614, "standard_deviation": 0.003081493157754092, "minimum": 0.9455100503816778, "maximum": 0.9547522308796854, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.714078467805146, "standard_deviation": 0.03858310606998515, "minimum": 1.6595450854756202, "maximum": 1.766625429666833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4842394945865851, "standard_deviation": 0.16119723960482005, "minimum": 0.1708273036632756, "maximum": 0.688438599279355, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21583089468589567, "standard_deviation": 0.006019070180738037, "minimum": 0.2074914118958135, "maximum": 0.22557565277492694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9819801223489136, "standard_deviation": 0.002725274787337365, "minimum": 0.9788783703579468, "maximum": 0.9858803484862488, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9738301803215003, "standard_deviation": 0.0387415953114673, "minimum": 0.8969849496561175, "maximum": 1.0164288602181217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5746986160148068, "standard_deviation": 0.21319121789485393, "minimum": 0.24353165898999116, "maximum": 0.9582392554411018, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1235807848548564, "standard_deviation": 0.007545623067938457, "minimum": 0.11172382468853485, "maximum": 0.13374504140658702, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9766734948796777, "standard_deviation": 0.004650134863852726, "minimum": 0.969803663531539, "maximum": 0.9823335314219479, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.816745232900576, "standard_deviation": 0.030960661406733522, "minimum": 1.7547313793811476, "maximum": 1.8597795822567005, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.764568338775077, "standard_deviation": 0.2035705740526206, "minimum": 0.322517504337098, "maximum": 1.0459029800629933, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20575794226396174, "standard_deviation": 0.02783383413668583, "minimum": 0.17402670700807046, "maximum": 0.26349654387307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9907115006830536, "standard_deviation": 0.0017463420162744498, "minimum": 0.9875007025338585, "maximum": 0.9925820545240672, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3466889344611541, "standard_deviation": 0.029676613616248314, "minimum": 1.2961543517834642, "maximum": 1.3928292998525715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.868386162138713, "standard_deviation": 0.2928219292706889, "minimum": 0.4626077860314535, "maximum": 1.4374569689381584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13808661867328656, "standard_deviation": 0.026788019373099346, "minimum": 0.10932726907972358, "maximum": 0.19109817449895627, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9871007905593197, "standard_deviation": 0.0012008900865685657, "minimum": 0.9858026156306514, "maximum": 0.9892873708967966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5920269643070273, "standard_deviation": 0.06616841258727482, "minimum": 1.5091699872948032, "maximum": 1.72966703506926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7849335111043743, "standard_deviation": 0.2741983180523163, "minimum": 0.508648324520041, "maximum": 1.323573492163879, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11835223123159441, "standard_deviation": 0.005688440915669979, "minimum": 0.10813075398610834, "maximum": 0.12516379818072726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9953123903828529, "standard_deviation": 0.0008320061445050737, "minimum": 0.9937719276591622, "maximum": 0.9964533824051263, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.939760269076888, "standard_deviation": 0.0696049924828917, "minimum": 0.8709087481409097, "maximum": 1.0455226065644065, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9616077150599135, "standard_deviation": 0.3166431045964708, "minimum": 0.6503430671986493, "maximum": 1.6921426217445938, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06926176496668923, "standard_deviation": 0.006963166430906935, "minimum": 0.06086354930692449, "maximum": 0.08130116330963938, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2431681466502917, "standard_deviation": 0.05691168304921824, "minimum": 0.16837912099860367, "maximum": 0.32335963506303256, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.630892745966202, "standard_deviation": 0.0065112945259141524, "minimum": 5.6205073828097225, "maximum": 5.6387874864791, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.35114828631176087, "standard_deviation": 0.014207147382816183, "minimum": 0.326303668348972, "maximum": 0.37051862731257545, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05449029632632911, "standard_deviation": 0.06309841706347043, "minimum": -0.15587472359494614, "maximum": 0.06623525352833876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9878435735345029, "standard_deviation": 0.0012600315649894443, "minimum": 0.9857918689073412, "maximum": 0.989798139555876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.195053347302351, "standard_deviation": 0.04424053183314598, "minimum": 1.1081841167255866, "maximum": 1.2530437312097265, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5986399578655268, "standard_deviation": 0.17702471647044546, "minimum": 0.36448463290103583, "maximum": 0.8204768507491184, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09338183633877815, "standard_deviation": 0.004669552993583076, "minimum": 0.08549792721583357, "maximum": 0.10067324423623857, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959762443960263, "standard_deviation": 0.0009147634785458519, "minimum": 0.9947093543890191, "maximum": 0.9973200953403273, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5894416026827229, "standard_deviation": 0.061078559987180425, "minimum": 0.5125781470379639, "maximum": 0.6996432256012539, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.634500192189801, "standard_deviation": 0.17401210051702448, "minimum": 0.3292139723627039, "maximum": 0.9079867747759813, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049624099756159716, "standard_deviation": 0.005740655337936533, "minimum": 0.04022607581296078, "maximum": 0.056495784105970503, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0042", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.0247488021850586, "predicted_mos": 2.0247488021850586, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "f0f467c95c9091a36109e21abdb3acdd611b40bdde085653856a8e41771c95d4", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.897640500217676, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.150225836775212, "全帯域_10ms_rms_envelope_nrmse": 0.09257385833585892, "全帯域_10ms_envelope_correlation": 0.9882894663811007, "全帯域_10ms_modulation_mae_db": 0.4340615884252421, "全帯域_40ms_log_rms_mae_db": 0.4872721637716255, "全帯域_40ms_rms_envelope_nrmse": 0.04253219374164949, "全帯域_40ms_envelope_correlation": 0.9975161059574734, "全帯域_40ms_modulation_mae_db": 0.38484689896861396, "300–1000 Hz_10ms_log_rms_mae_db": 1.780914675981198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1253241636398398, "300–1000 Hz_10ms_envelope_correlation": 0.9857392718381357, "300–1000 Hz_10ms_modulation_mae_db": 0.686708003775992, "300–1000 Hz_40ms_log_rms_mae_db": 1.0972013010086437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06558659000936612, "300–1000 Hz_40ms_envelope_correlation": 0.9963157382921948, "300–1000 Hz_40ms_modulation_mae_db": 0.7384900626393694, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9263491561533472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22244518734948537, "1000–3000 Hz_10ms_envelope_correlation": 0.9468526174853362, "1000–3000 Hz_10ms_modulation_mae_db": 0.4338834150637332, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0833685541757314, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.128555763560691, "1000–3000 Hz_40ms_envelope_correlation": 0.9794321113412443, "1000–3000 Hz_40ms_modulation_mae_db": 0.4350268977215079, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0730167522507053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1959306078241507, "3000–6000 Hz_10ms_envelope_correlation": 0.9774259630607509, "3000–6000 Hz_10ms_modulation_mae_db": 1.7607076627546043, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4772441465357269, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13043458041100214, "3000–6000 Hz_40ms_envelope_correlation": 0.991281228729007, "3000–6000 Hz_40ms_modulation_mae_db": 2.277364023804221, "mr_stft_db": 5.756883914650672, "band_mae_db": 0.2470550602728593, "rms_gain_error_db": -0.1629762434108378, "output_peak": 0.3479316517959658}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0701173233112833, "全帯域_10ms_rms_envelope_nrmse": 0.0933032716486894, "全帯域_10ms_envelope_correlation": 0.9878411215460575, "全帯域_10ms_modulation_mae_db": 0.389706619771855, "全帯域_40ms_log_rms_mae_db": 0.5022523530248133, "全帯域_40ms_rms_envelope_nrmse": 0.048286647109360946, "全帯域_40ms_envelope_correlation": 0.9961421198710042, "全帯域_40ms_modulation_mae_db": 0.21092476382697045, "300–1000 Hz_10ms_log_rms_mae_db": 1.694320046745145, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11284092440264444, "300–1000 Hz_10ms_envelope_correlation": 0.9880381674202583, "300–1000 Hz_10ms_modulation_mae_db": 0.8340267731388485, "300–1000 Hz_40ms_log_rms_mae_db": 1.0055630523420112, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06523974435147789, "300–1000 Hz_40ms_envelope_correlation": 0.9955069826275648, "300–1000 Hz_40ms_modulation_mae_db": 0.4998821443293213, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9710351913173578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2170635653440173, "1000–3000 Hz_10ms_envelope_correlation": 0.9493764265524399, "1000–3000 Hz_10ms_modulation_mae_db": 0.6102346206987191, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0611312241054454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11637041954964346, "1000–3000 Hz_40ms_envelope_correlation": 0.9839584124939557, "1000–3000 Hz_40ms_modulation_mae_db": 0.7572124662614025, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0961444883907867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2220229902917091, "3000–6000 Hz_10ms_envelope_correlation": 0.9705065694218575, "3000–6000 Hz_10ms_modulation_mae_db": 1.0210744126119087, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5097777672020671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14176550643581434, "3000–6000 Hz_40ms_envelope_correlation": 0.9898530848202933, "3000–6000 Hz_40ms_modulation_mae_db": 1.3476373041482914, "mr_stft_db": 5.767073697595954, "band_mae_db": 0.2673986058378777, "rms_gain_error_db": -0.038891674382400876, "output_peak": 0.36338638086275754}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2268607728200367, "全帯域_10ms_rms_envelope_nrmse": 0.09486602645172805, "全帯域_10ms_envelope_correlation": 0.987419625824585, "全帯域_10ms_modulation_mae_db": 0.23387798837026844, "全帯域_40ms_log_rms_mae_db": 0.6398746421033996, "全帯域_40ms_rms_envelope_nrmse": 0.05287114947694102, "全帯域_40ms_envelope_correlation": 0.995461046328152, "全帯域_40ms_modulation_mae_db": 0.20905954036151883, "300–1000 Hz_10ms_log_rms_mae_db": 1.7654685235973566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1244829263424135, "300–1000 Hz_10ms_envelope_correlation": 0.9856228221189346, "300–1000 Hz_10ms_modulation_mae_db": 0.3505152074672199, "300–1000 Hz_40ms_log_rms_mae_db": 1.0696534784420104, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07874343952102963, "300–1000 Hz_40ms_envelope_correlation": 0.9937237976452907, "300–1000 Hz_40ms_modulation_mae_db": 0.2457805878048603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9041321143934429, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20501320761284517, "1000–3000 Hz_10ms_envelope_correlation": 0.9549519237819931, "1000–3000 Hz_10ms_modulation_mae_db": 0.6938184899620987, "1000–3000 Hz_40ms_log_rms_mae_db": 1.052849507573489, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10778135179698291, "1000–3000 Hz_40ms_envelope_correlation": 0.9863260453853456, "1000–3000 Hz_40ms_modulation_mae_db": 1.0365556293321172, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1265911242445847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20932063148896932, "3000–6000 Hz_10ms_envelope_correlation": 0.9782855904249231, "3000–6000 Hz_10ms_modulation_mae_db": 1.084495027137949, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5615986058629208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13850029936747846, "3000–6000 Hz_40ms_envelope_correlation": 0.9921287931322428, "3000–6000 Hz_40ms_modulation_mae_db": 1.1998956119477822, "mr_stft_db": 5.7618129059965995, "band_mae_db": 0.20517146382805918, "rms_gain_error_db": -0.08220060863844636, "output_peak": 0.3365769603352119}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1471431391275948, "全帯域_10ms_rms_envelope_nrmse": 0.09930722457378874, "全帯域_10ms_envelope_correlation": 0.9863144864726752, "全帯域_10ms_modulation_mae_db": 0.26428412073981516, "全帯域_40ms_log_rms_mae_db": 0.5441942862086084, "全帯域_40ms_rms_envelope_nrmse": 0.06386979177259045, "全帯域_40ms_envelope_correlation": 0.9932754892128586, "全帯域_40ms_modulation_mae_db": 0.16151676784489635, "300–1000 Hz_10ms_log_rms_mae_db": 1.741971331287649, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14959749006993692, "300–1000 Hz_10ms_envelope_correlation": 0.9790307757579569, "300–1000 Hz_10ms_modulation_mae_db": 0.7015143637554684, "300–1000 Hz_40ms_log_rms_mae_db": 1.0560582095603683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10376861774492187, "300–1000 Hz_40ms_envelope_correlation": 0.9884413191235839, "300–1000 Hz_40ms_modulation_mae_db": 0.29729299952818083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.85348738491687, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2117606545951427, "1000–3000 Hz_10ms_envelope_correlation": 0.9528647282639733, "1000–3000 Hz_10ms_modulation_mae_db": 0.5177835235133058, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0645692622393585, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12557813650070532, "1000–3000 Hz_40ms_envelope_correlation": 0.9807401983470242, "1000–3000 Hz_40ms_modulation_mae_db": 0.5991446166920679, "3000–6000 Hz_10ms_log_rms_mae_db": 2.083058877319877, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18721831386003476, "3000–6000 Hz_10ms_envelope_correlation": 0.9795137726499809, "3000–6000 Hz_10ms_modulation_mae_db": 0.9523472629428366, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4493310548425502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12922946084677858, "3000–6000 Hz_40ms_envelope_correlation": 0.9901459406074009, "3000–6000 Hz_40ms_modulation_mae_db": 0.882282213716577, "mr_stft_db": 5.746693592376481, "band_mae_db": 0.12380914716012535, "rms_gain_error_db": 0.025228293559787104, "output_peak": 0.35024359591557364}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1549260909874413, "全帯域_10ms_rms_envelope_nrmse": 0.09973709246401068, "全帯域_10ms_envelope_correlation": 0.986207892440775, "全帯域_10ms_modulation_mae_db": 0.06218644621932498, "全帯域_40ms_log_rms_mae_db": 0.5519778349233219, "全帯域_40ms_rms_envelope_nrmse": 0.0498512275969124, "全帯域_40ms_envelope_correlation": 0.9959307626656878, "全帯域_40ms_modulation_mae_db": 0.3204637611933643, "300–1000 Hz_10ms_log_rms_mae_db": 1.6977723055585972, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1315711619010092, "300–1000 Hz_10ms_envelope_correlation": 0.9837877392298309, "300–1000 Hz_10ms_modulation_mae_db": 0.13948777197120193, "300–1000 Hz_40ms_log_rms_mae_db": 0.9768660243624855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0714152922870018, "300–1000 Hz_40ms_envelope_correlation": 0.9945467667556825, "300–1000 Hz_40ms_modulation_mae_db": 0.38774919653290824, "1000–3000 Hz_10ms_log_rms_mae_db": 1.888531048585839, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21125184710701894, "1000–3000 Hz_10ms_envelope_correlation": 0.9520383634837664, "1000–3000 Hz_10ms_modulation_mae_db": 0.42774657310667097, "1000–3000 Hz_40ms_log_rms_mae_db": 1.061157833946124, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11451072201759625, "1000–3000 Hz_40ms_envelope_correlation": 0.9838531311538287, "1000–3000 Hz_40ms_modulation_mae_db": 0.3434285528874142, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0836070783466027, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2236559150423053, "3000–6000 Hz_10ms_envelope_correlation": 0.9696951944055623, "3000–6000 Hz_10ms_modulation_mae_db": 0.8142637496856148, "3000–6000 Hz_40ms_log_rms_mae_db": 1.449750494421262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1395311040505641, "3000–6000 Hz_40ms_envelope_correlation": 0.9878280646373188, "3000–6000 Hz_40ms_modulation_mae_db": 1.2720244876447215, "mr_stft_db": 5.767647438198552, "band_mae_db": 0.11698711333813754, "rms_gain_error_db": 0.011092954824636575, "output_peak": 0.35595819768923126}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.154536833727896, "全帯域_10ms_rms_envelope_nrmse": 0.08508394482228886, "全帯域_10ms_envelope_correlation": 0.9898755045203806, "全帯域_10ms_modulation_mae_db": 0.43874891169857255, "全帯域_40ms_log_rms_mae_db": 0.5214654787441884, "全帯域_40ms_rms_envelope_nrmse": 0.04105299512047417, "全帯域_40ms_envelope_correlation": 0.9972421595931145, "全帯域_40ms_modulation_mae_db": 0.5802907068127969, "300–1000 Hz_10ms_log_rms_mae_db": 1.7617297839970738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11632875220641446, "300–1000 Hz_10ms_envelope_correlation": 0.9872953842699694, "300–1000 Hz_10ms_modulation_mae_db": 0.455341578066178, "300–1000 Hz_40ms_log_rms_mae_db": 1.1090343093392536, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06244099734227685, "300–1000 Hz_40ms_envelope_correlation": 0.9959036990957276, "300–1000 Hz_40ms_modulation_mae_db": 0.6253601142770346, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9066390539905786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21220525309138627, "1000–3000 Hz_10ms_envelope_correlation": 0.9520817290562337, "1000–3000 Hz_10ms_modulation_mae_db": 0.2612161959376927, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1002662083281343, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12028617943668175, "1000–3000 Hz_40ms_envelope_correlation": 0.9825600321935682, "1000–3000 Hz_40ms_modulation_mae_db": 0.492971096422498, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1118315541908843, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21882839730542494, "3000–6000 Hz_10ms_envelope_correlation": 0.9721684485214465, "3000–6000 Hz_10ms_modulation_mae_db": 1.2711875854016474, "3000–6000 Hz_40ms_log_rms_mae_db": 1.475943058271084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14238073773224383, "3000–6000 Hz_40ms_envelope_correlation": 0.9906494777467879, "3000–6000 Hz_40ms_modulation_mae_db": 1.3893474067486824, "mr_stft_db": 5.783191058728697, "band_mae_db": 0.19942489331822577, "rms_gain_error_db": -0.05511590095612703, "output_peak": 0.3894904821538813}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.163864403714647, "全帯域_10ms_rms_envelope_nrmse": 0.09300118820813845, "全帯域_10ms_envelope_correlation": 0.9879617197103109, "全帯域_10ms_modulation_mae_db": 0.46100542814262396, "全帯域_40ms_log_rms_mae_db": 0.5363034911783691, "全帯域_40ms_rms_envelope_nrmse": 0.04591196031339158, "全帯域_40ms_envelope_correlation": 0.9964985592770439, "全帯域_40ms_modulation_mae_db": 0.31886455378797934, "300–1000 Hz_10ms_log_rms_mae_db": 1.6486341659688961, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11425664902993096, "300–1000 Hz_10ms_envelope_correlation": 0.987755774603121, "300–1000 Hz_10ms_modulation_mae_db": 0.5332573830086499, "300–1000 Hz_40ms_log_rms_mae_db": 0.9736703926353754, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06718165451849067, "300–1000 Hz_40ms_envelope_correlation": 0.9951436349943896, "300–1000 Hz_40ms_modulation_mae_db": 0.5579626813861052, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8940189378398289, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22440691380280636, "1000–3000 Hz_10ms_envelope_correlation": 0.9456857873836356, "1000–3000 Hz_10ms_modulation_mae_db": 0.8058921485248375, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9858866421798612, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1078633140031288, "1000–3000 Hz_40ms_envelope_correlation": 0.9861769401011239, "1000–3000 Hz_40ms_modulation_mae_db": 0.9513204218930379, "3000–6000 Hz_10ms_log_rms_mae_db": 2.087864123827551, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2387230243840525, "3000–6000 Hz_10ms_envelope_correlation": 0.9643752715345487, "3000–6000 Hz_10ms_modulation_mae_db": 0.7266462251051706, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4962961717802756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12905054920166373, "3000–6000 Hz_40ms_envelope_correlation": 0.9892874580909606, "3000–6000 Hz_40ms_modulation_mae_db": 1.1243674938243922, "mr_stft_db": 5.759091025186433, "band_mae_db": 0.09136416177135365, "rms_gain_error_db": 0.0008627455115641137, "output_peak": 0.3631958711625635}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1387544073026998, "全帯域_10ms_rms_envelope_nrmse": 0.10064850805516208, "全帯域_10ms_envelope_correlation": 0.9860236579806053, "全帯域_10ms_modulation_mae_db": 0.783761265678479, "全帯域_40ms_log_rms_mae_db": 0.5216290937041681, "全帯域_40ms_rms_envelope_nrmse": 0.05156074740515208, "全帯域_40ms_envelope_correlation": 0.9956922028512589, "全帯域_40ms_modulation_mae_db": 0.7224248995096708, "300–1000 Hz_10ms_log_rms_mae_db": 1.683793888058901, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12350274589736807, "300–1000 Hz_10ms_envelope_correlation": 0.985700373124521, "300–1000 Hz_10ms_modulation_mae_db": 1.5157201312316664, "300–1000 Hz_40ms_log_rms_mae_db": 1.0369922645408907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07425597798591421, "300–1000 Hz_40ms_envelope_correlation": 0.9942956660076961, "300–1000 Hz_40ms_modulation_mae_db": 1.8439556520135387, "1000–3000 Hz_10ms_log_rms_mae_db": 1.967028836530529, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22650929958511015, "1000–3000 Hz_10ms_envelope_correlation": 0.9458154026642275, "1000–3000 Hz_10ms_modulation_mae_db": 0.4901395622630313, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0944808153744936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12423089368249883, "1000–3000 Hz_40ms_envelope_correlation": 0.9812979743415778, "1000–3000 Hz_40ms_modulation_mae_db": 0.35477031992905544, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0571264236974462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23359847621997576, "3000–6000 Hz_10ms_envelope_correlation": 0.9668120880185964, "3000–6000 Hz_10ms_modulation_mae_db": 0.5883557734186037, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3913388143840408, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11820659714302496, "3000–6000 Hz_40ms_envelope_correlation": 0.9913555543899457, "3000–6000 Hz_40ms_modulation_mae_db": 0.6145393441261957, "mr_stft_db": 5.765365220061532, "band_mae_db": 0.1678488473244901, "rms_gain_error_db": 0.044047188945342375, "output_peak": 0.3430998918164184}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44560, "output_samples": 44560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.995885943582178, "spectral_convergence": 0.3216829854189462, "spectral_convergence_status": "ok", "amplitude_floor": 9.278908430047933e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.575508161314728, "spectral_convergence": 0.30194082003508055, "spectral_convergence_status": "ok", "amplitude_floor": 4.74996132181228e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.699257639055107, "spectral_convergence": 0.3076854860611755, "spectral_convergence_status": "ok", "amplitude_floor": 3.301834166389807e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005909300513360037, "output_power": 0.0006048701395318235, "delta_db": 0.10126069048331879, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000598279616407713, "output_power": 0.0005654144664958376, "delta_db": -0.24537290530100614, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005543526355345429, "output_power": 0.0005197738168176318, "delta_db": -0.2797171813141233, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011395641530102642, "output_power": 0.00010760359688498982, "delta_db": -0.2491199047384397, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011253089498164014, "output_power": 9.949221883709383e-05, "delta_db": -0.5348265647691895, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6482875672349115e-05, "output_power": 1.621174154203895e-05, "delta_db": -0.07203311503107852, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2470550602728593, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1629762434108378, "pitch": null, "output_rms_dbfs": -27.182019521158992, "output_peak": 0.3479316517959658, "reference_rms_dbfs": -27.019043277748153, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.756883914650672}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 0.00011061650326899306, "log_rms_mae_db": 1.150225836775212, "rms_envelope_nrmse": 0.09257385833585892, "envelope_correlation": 0.9882894663811007, "rms_gain_error_db": -0.16295617543744273, "reference_mean_power": 0.00199005080866686, "output_mean_power": 0.0019167635723816006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 8.439637043999617e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970864143567599, "output_power": 0.0005782300648805981, "reference_active": true, "delta_db": -0.13936520773524122}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010123102438186569, "output_power": 9.598399381393126e-05, "reference_active": true, "delta_db": -0.2311481500087345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240115070482855e-05, "output_power": 2.6145279770083024e-05, "reference_active": true, "delta_db": -0.9316714075317506}], "active_band_count": 3, "active_band_mae_db": 0.4340615884252421}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 9.74750993349889e-05, "log_rms_mae_db": 0.4872721637716255, "rms_envelope_nrmse": 0.04253219374164949, "envelope_correlation": 0.9975161059574734, "rms_gain_error_db": -0.16296276861054954, "reference_mean_power": 0.0020114637230104894, "output_mean_power": 0.0019373849760417837}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 6.839204647135984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005172552173341314, "output_power": 0.0005004526954422182, "reference_active": true, "delta_db": -0.14341847198442423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6641602045504645e-05, "output_power": 5.4403833073556313e-05, "reference_active": true, "delta_db": -0.17506028992894632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.529860845744804e-06, "output_power": 2.0868461867069423e-06, "reference_active": true, "delta_db": -0.8360619349924714}], "active_band_count": 3, "active_band_mae_db": 0.38484689896861396}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 8.308955229004579e-05, "log_rms_mae_db": 1.780914675981198, "rms_envelope_nrmse": 0.1253241636398398, "envelope_correlation": 0.9857392718381357, "rms_gain_error_db": -0.2755906628140905, "reference_mean_power": 0.000839400894204417, "output_mean_power": 0.0007877898141708493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.694834819211473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042315114528166547, "output_power": 0.00039550414748340003, "reference_active": true, "delta_db": -0.293444785356759}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.939611213191948e-05, "output_power": 5.864091384213786e-05, "reference_active": true, "delta_db": -0.7313441049661751}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4277196064029695e-05, "output_power": 1.9127799624137116e-05, "reference_active": true, "delta_db": -1.0353351210050419}], "active_band_count": 3, "active_band_mae_db": 0.686708003775992}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 8.008999734930271e-05, "log_rms_mae_db": 1.0972013010086437, "rms_envelope_nrmse": 0.06558659000936612, "envelope_correlation": 0.9963157382921948, "rms_gain_error_db": -0.2756047756343545, "reference_mean_power": 0.0008485551352947845, "output_mean_power": 0.0007963786132061378}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 5.733703583211585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003764127527272362, "output_power": 0.00035330208391619566, "reference_active": true, "delta_db": -0.2751812965968444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8963437868750806e-05, "output_power": 3.338098602731653e-05, "reference_active": true, "delta_db": -0.6715810810837974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.296630500499659e-06, "output_power": 1.7148271354284716e-06, "reference_active": true, "delta_db": -1.2687078102374665}], "active_band_count": 3, "active_band_mae_db": 0.7384900626393694}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.2528786526254916e-05, "log_rms_mae_db": 1.9263491561533472, "rms_envelope_nrmse": 0.22244518734948537, "envelope_correlation": 0.9468526174853362, "rms_gain_error_db": -0.16997366349709098, "reference_mean_power": 0.0001368277838750774, "output_mean_power": 0.00013157607490002808}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.073610600879068e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.932287229550182e-05, "output_power": 3.9259685795300896e-05, "reference_active": true, "delta_db": -0.006984133616613738}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8616503346595646e-06, "output_power": 8.862089913090034e-06, "reference_active": true, "delta_db": 1.1110756907447787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.317401673918684e-06, "output_power": 4.50382490012389e-06, "reference_active": true, "delta_db": 0.18359042082980712}], "active_band_count": 3, "active_band_mae_db": 0.4338834150637332}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 3.268734932259802e-05, "log_rms_mae_db": 1.0833685541757314, "rms_envelope_nrmse": 0.128555763560691, "envelope_correlation": 0.9794321113412443, "rms_gain_error_db": -0.17004884435127565, "reference_mean_power": 0.00013831808192096785, "output_mean_power": 0.00013300686998394347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 4.85870788941741e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5019378943818874e-05, "output_power": 3.60571655723021e-05, "reference_active": true, "delta_db": 0.12683144394989654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.168829067205367e-06, "output_power": 5.25563496151777e-06, "reference_active": true, "delta_db": 1.0061110469331966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163313597471507e-07, "output_power": 3.2912135957244306e-07, "reference_active": true, "delta_db": 0.17213820228143048}], "active_band_count": 3, "active_band_mae_db": 0.4350268977215079}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.9606629692154165e-05, "log_rms_mae_db": 2.0730167522507053, "rms_envelope_nrmse": 0.1959306078241507, "envelope_correlation": 0.9774259630607509, "rms_gain_error_db": -0.5218283878678408, "reference_mean_power": 5.1065139870557587e-05, "output_mean_power": 4.528367731546564e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 2.3044643706212192e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0452875124351046e-05, "output_power": 9.063061586549562e-06, "reference_active": true, "delta_db": -0.6196083093223994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.855628387226494e-06, "output_power": 2.4044413618610414e-06, "reference_active": true, "delta_db": -0.7468750032810701}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6522678437727886e-06, "output_power": 6.706817412434045e-07, "reference_active": true, "delta_db": -3.915639675660344}], "active_band_count": 3, "active_band_mae_db": 1.7607076627546043}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 4.1718716562101114e-05, "log_rms_mae_db": 1.4772441465357269, "rms_envelope_nrmse": 0.13043458041100214, "envelope_correlation": 0.991281228729007, "rms_gain_error_db": -0.5218647520361055, "reference_mean_power": 5.162171435861935e-05, "output_mean_power": 4.57768545865355e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 2.066669918587524e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0052154991285675e-05, "output_power": 8.436939072604471e-06, "reference_active": true, "delta_db": -0.7607426337495772}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.890809551942203e-06, "output_power": 1.3382621603729703e-06, "reference_active": true, "delta_db": -1.5010658922132578}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8144313206293327e-07, "output_power": 6.334497652036428e-08, "reference_active": true, "delta_db": -4.570283545449828}], "active_band_count": 3, "active_band_mae_db": 2.277364023804221}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8403637274048065, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d13c5fbd646d5acda092272ca07a6ae853324d7731d245878542f1cee12de1d3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9499583723339506, "standard_deviation": 0.003537978069136092, "minimum": 0.9456857873836356, "maximum": 0.9549519237819931, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9139027154659742, "standard_deviation": 0.039782878772023675, "minimum": 1.85348738491687, "maximum": 1.9710351913173578, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5300893161337611, "standard_deviation": 0.17032217251121645, "minimum": 0.2612161959376927, "maximum": 0.8058921485248375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21633199106097653, "standard_deviation": 0.007545463732482989, "minimum": 0.20501320761284517, "maximum": 0.22650929958511015, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9830431056697085, "standard_deviation": 0.002498972497428168, "minimum": 0.9794321113412443, "maximum": 0.9863260453853456, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0629637559903298, "standard_deviation": 0.03556388706834521, "minimum": 0.9858866421798612, "maximum": 1.1002662083281343, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6213037501423876, "standard_deviation": 0.26718916882072286, "minimum": 0.3434285528874142, "maximum": 1.0365556293321172, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11814709756849104, "standard_deviation": 0.007880660011277678, "minimum": 0.10778135179698291, "maximum": 0.128555763560691, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9723478622547083, "standard_deviation": 0.0055685364825063186, "minimum": 0.9643752715345487, "maximum": 0.9795137726499809, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0899050527835548, "standard_deviation": 0.021764725954266192, "minimum": 2.0571264236974462, "maximum": 2.1265911242445847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.027384712382292, "standard_deviation": 0.36542441252923935, "minimum": 0.5883557734186037, "maximum": 1.7607076627546043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2161622945520778, "standard_deviation": 0.017746229731693246, "minimum": 0.18721831386003476, "maximum": 0.2387230243840525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9903162002692446, "standard_deviation": 0.001357555436602726, "minimum": 0.9878280646373188, "maximum": 0.9921287931322428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.476410014162491, "standard_deviation": 0.04992563791803853, "minimum": 1.3913388143840408, "maximum": 1.5615986058629208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.263432235745108, "standard_deviation": 0.4841472400827256, "minimum": 0.6145393441261957, "maximum": 2.277364023804221, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13363735439857127, "standard_deviation": 0.008364315910584721, "minimum": 0.11820659714302496, "maximum": 0.14238073773224383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.985371288545341, "standard_deviation": 0.002915889214901915, "minimum": 0.9790307757579569, "maximum": 0.9880381674202583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.721825590149352, "standard_deviation": 0.04710615918488619, "minimum": 1.6486341659688961, "maximum": 1.780914675981198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6520714015519031, "standard_deviation": 0.4121613235143996, "minimum": 0.13948777197120193, "maximum": 1.5157201312316664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12473810168619467, "standard_deviation": 0.011884776730172766, "minimum": 0.11284092440264444, "maximum": 0.14959749006993692, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942347005677663, "standard_deviation": 0.002492460917387099, "minimum": 0.9884413191235839, "maximum": 0.9963157382921948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.04062987902888, "standard_deviation": 0.051757968509801335, "minimum": 0.9736703926353754, "maximum": 1.1090343093392536, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6495591798139149, "standard_deviation": 0.510118115657693, "minimum": 0.2457805878048603, "maximum": 1.8439556520135387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07357903922005987, "standard_deviation": 0.013318233044762874, "minimum": 0.06244099734227685, "maximum": 0.10376861774492187, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1773824116063911, "standard_deviation": 0.06349691835993516, "minimum": 0.09136416177135365, "maximum": 0.2673986058378777, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7634698565993645, "standard_deviation": 0.010486013790464729, "minimum": 5.746693592376481, "maximum": 5.783191058728697, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3562353789664504, "standard_deviation": 0.016341934629310297, "minimum": 0.3365769603352119, "maximum": 0.3894904821538813, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.032244155568310236, "standard_deviation": 0.06787746936448845, "minimum": -0.1629762434108378, "maximum": 0.044047188945342375, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9874916843595614, "standard_deviation": 0.0013014445122184527, "minimum": 0.9860236579806053, "maximum": 0.9898755045203806, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1508036009708513, "standard_deviation": 0.04251957729218934, "minimum": 1.0701173233112833, "maximum": 1.2268607728200367, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3834540461307727, "standard_deviation": 0.2110196896608975, "minimum": 0.06218644621932498, "maximum": 0.783761265678479, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09481513931995814, "standard_deviation": 0.0051232088476101815, "minimum": 0.08508394482228886, "maximum": 0.10064850805516208, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.995969805719574, "standard_deviation": 0.0013047148876196336, "minimum": 0.9932754892128586, "maximum": 0.9975161059574734, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5381211679573118, "standard_deviation": 0.04630266428934971, "minimum": 0.4872721637716255, "maximum": 0.6398746421033996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36354898653822637, "standard_deviation": 0.19568695488828206, "minimum": 0.16151676784489635, "maximum": 0.7224248995096708, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049492089067059014, "standard_deviation": 0.007136161172225209, "minimum": 0.04105299512047417, "maximum": 0.06386979177259045, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0042", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.095690965652466, "predicted_mos": 2.095690965652466, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "489a96b6df1800353022e07207d7d5c1169e4dc774c84c0d7a915bee4500b082", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.039803999941796, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0954404676139857, "全帯域_10ms_rms_envelope_nrmse": 0.09403599127792524, "全帯域_10ms_envelope_correlation": 0.9876406175849994, "全帯域_10ms_modulation_mae_db": 0.13233923791907498, "全帯域_40ms_log_rms_mae_db": 0.5252962067047782, "全帯域_40ms_rms_envelope_nrmse": 0.04841050819083398, "全帯域_40ms_envelope_correlation": 0.9960945660929981, "全帯域_40ms_modulation_mae_db": 0.22620031502683688, "300–1000 Hz_10ms_log_rms_mae_db": 1.7209661752993801, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12743106108217972, "300–1000 Hz_10ms_envelope_correlation": 0.9846948116515722, "300–1000 Hz_10ms_modulation_mae_db": 0.3095204324800034, "300–1000 Hz_40ms_log_rms_mae_db": 1.0054282635837688, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06812082250247481, "300–1000 Hz_40ms_envelope_correlation": 0.9950304417592333, "300–1000 Hz_40ms_modulation_mae_db": 0.33898207821277054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.886970352763164, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2082322914524617, "1000–3000 Hz_10ms_envelope_correlation": 0.9533875471060467, "1000–3000 Hz_10ms_modulation_mae_db": 0.7430389408025778, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9985701552776163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10607802498881384, "1000–3000 Hz_40ms_envelope_correlation": 0.9861094835264287, "1000–3000 Hz_40ms_modulation_mae_db": 0.8487493988330456, "3000–6000 Hz_10ms_log_rms_mae_db": 2.063320788521023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.214012366981898, "3000–6000 Hz_10ms_envelope_correlation": 0.9779689616469415, "3000–6000 Hz_10ms_modulation_mae_db": 1.0118046253785618, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4611715242502572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14091610054023251, "3000–6000 Hz_40ms_envelope_correlation": 0.9924896273475271, "3000–6000 Hz_40ms_modulation_mae_db": 0.9387578861842275, "mr_stft_db": 5.740890164382791, "band_mae_db": 0.13343063689907775, "rms_gain_error_db": 0.0006782751467122318, "output_peak": 0.3454309876247169}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1468847609866453, "全帯域_10ms_rms_envelope_nrmse": 0.08721537663646578, "全帯域_10ms_envelope_correlation": 0.9893524199424595, "全帯域_10ms_modulation_mae_db": 0.5822579631519734, "全帯域_40ms_log_rms_mae_db": 0.6030159438935415, "全帯域_40ms_rms_envelope_nrmse": 0.04308888971210899, "全帯域_40ms_envelope_correlation": 0.9969106726707513, "全帯域_40ms_modulation_mae_db": 0.5985494836172105, "300–1000 Hz_10ms_log_rms_mae_db": 1.7268294281209287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11784317030140995, "300–1000 Hz_10ms_envelope_correlation": 0.9869198055663503, "300–1000 Hz_10ms_modulation_mae_db": 0.7246355929868059, "300–1000 Hz_40ms_log_rms_mae_db": 1.0138755264296475, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06739287041998855, "300–1000 Hz_40ms_envelope_correlation": 0.9951704579220868, "300–1000 Hz_40ms_modulation_mae_db": 0.9502080194811834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8889729921332474, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20877608216952137, "1000–3000 Hz_10ms_envelope_correlation": 0.9533453118217953, "1000–3000 Hz_10ms_modulation_mae_db": 0.48824762024890417, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9544227582363475, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09918318624873407, "1000–3000 Hz_40ms_envelope_correlation": 0.9879238204122731, "1000–3000 Hz_40ms_modulation_mae_db": 0.4644175575914771, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0764102535608973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2266678908023198, "3000–6000 Hz_10ms_envelope_correlation": 0.9682713752143275, "3000–6000 Hz_10ms_modulation_mae_db": 1.3877703477708934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4662795531960484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13716414535750876, "3000–6000 Hz_40ms_envelope_correlation": 0.987965403252656, "3000–6000 Hz_40ms_modulation_mae_db": 1.1938544203822288, "mr_stft_db": 5.723763713806119, "band_mae_db": 0.17231103879845486, "rms_gain_error_db": -0.022136814580412265, "output_peak": 0.3485922469597333}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.217537006359276, "全帯域_10ms_rms_envelope_nrmse": 0.0936710215044459, "全帯域_10ms_envelope_correlation": 0.9878167657832244, "全帯域_10ms_modulation_mae_db": 0.6946177904897125, "全帯域_40ms_log_rms_mae_db": 0.5888362600907799, "全帯域_40ms_rms_envelope_nrmse": 0.04657506597709319, "全帯域_40ms_envelope_correlation": 0.9965136208452524, "全帯域_40ms_modulation_mae_db": 0.6207770273069341, "300–1000 Hz_10ms_log_rms_mae_db": 1.747419591509525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11574560212100646, "300–1000 Hz_10ms_envelope_correlation": 0.9875761020782001, "300–1000 Hz_10ms_modulation_mae_db": 0.7062530373664483, "300–1000 Hz_40ms_log_rms_mae_db": 1.0423277669206956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06452038250792504, "300–1000 Hz_40ms_envelope_correlation": 0.9958788110678755, "300–1000 Hz_40ms_modulation_mae_db": 0.841383242172598, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9836819079593833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24979009397367505, "1000–3000 Hz_10ms_envelope_correlation": 0.9325535383175255, "1000–3000 Hz_10ms_modulation_mae_db": 0.4962335139581733, "1000–3000 Hz_40ms_log_rms_mae_db": 1.065442754609217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13094402127945515, "1000–3000 Hz_40ms_envelope_correlation": 0.9788788653582663, "1000–3000 Hz_40ms_modulation_mae_db": 0.49876744341530094, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9748040996660574, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21160479935850224, "3000–6000 Hz_10ms_envelope_correlation": 0.9759677489073347, "3000–6000 Hz_10ms_modulation_mae_db": 1.091934473250402, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3772010313755971, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13976959769450514, "3000–6000 Hz_40ms_envelope_correlation": 0.9903357134367126, "3000–6000 Hz_40ms_modulation_mae_db": 1.329180449377554, "mr_stft_db": 5.736869821634362, "band_mae_db": 0.1798078123230292, "rms_gain_error_db": 0.023434068343808134, "output_peak": 0.35264815173101205}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.195122773117007, "全帯域_10ms_rms_envelope_nrmse": 0.1039740718409299, "全帯域_10ms_envelope_correlation": 0.9848295536299522, "全帯域_10ms_modulation_mae_db": 0.8186390529405934, "全帯域_40ms_log_rms_mae_db": 0.5763849697916319, "全帯域_40ms_rms_envelope_nrmse": 0.05236669763981951, "全帯域_40ms_envelope_correlation": 0.9954846343305234, "全帯域_40ms_modulation_mae_db": 0.832363206958691, "300–1000 Hz_10ms_log_rms_mae_db": 1.6934181023241917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14077255799951888, "300–1000 Hz_10ms_envelope_correlation": 0.9813049963483381, "300–1000 Hz_10ms_modulation_mae_db": 1.0925497850157828, "300–1000 Hz_40ms_log_rms_mae_db": 0.9991320725077306, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08248807068634265, "300–1000 Hz_40ms_envelope_correlation": 0.9928088977361138, "300–1000 Hz_40ms_modulation_mae_db": 1.2099151156122985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.898224811084382, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20936109388329666, "1000–3000 Hz_10ms_envelope_correlation": 0.9529005620455586, "1000–3000 Hz_10ms_modulation_mae_db": 0.7502028882208517, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0422646257814778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11036822797339196, "1000–3000 Hz_40ms_envelope_correlation": 0.9849538412073542, "1000–3000 Hz_40ms_modulation_mae_db": 1.0528511119808126, "3000–6000 Hz_10ms_log_rms_mae_db": 2.075766596405919, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2178324393578526, "3000–6000 Hz_10ms_envelope_correlation": 0.970204386626176, "3000–6000 Hz_10ms_modulation_mae_db": 0.868611802428149, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4157018297584216, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13316635641790472, "3000–6000 Hz_40ms_envelope_correlation": 0.9885273045568962, "3000–6000 Hz_40ms_modulation_mae_db": 1.128770926069415, "mr_stft_db": 5.755046931193839, "band_mae_db": 0.14477726253633313, "rms_gain_error_db": -0.05710613012839573, "output_peak": 0.3562786117151335}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0831673449454167, "全帯域_10ms_rms_envelope_nrmse": 0.08907869146221455, "全帯域_10ms_envelope_correlation": 0.9888931119365438, "全帯域_10ms_modulation_mae_db": 0.375022724198472, "全帯域_40ms_log_rms_mae_db": 0.47213092643925164, "全帯域_40ms_rms_envelope_nrmse": 0.0439803621724485, "全帯域_40ms_envelope_correlation": 0.9968186658618899, "全帯域_40ms_modulation_mae_db": 0.39377023801614364, "300–1000 Hz_10ms_log_rms_mae_db": 1.7472167958107832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12301713858828316, "300–1000 Hz_10ms_envelope_correlation": 0.9857812028042008, "300–1000 Hz_10ms_modulation_mae_db": 0.24391251839432282, "300–1000 Hz_40ms_log_rms_mae_db": 1.089195779739847, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07259675871581381, "300–1000 Hz_40ms_envelope_correlation": 0.994404536606386, "300–1000 Hz_40ms_modulation_mae_db": 0.4184834848614405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7484727374009459, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2006216328640982, "1000–3000 Hz_10ms_envelope_correlation": 0.9570429765334897, "1000–3000 Hz_10ms_modulation_mae_db": 0.4428019412096163, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9203555775407404, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09955494398811572, "1000–3000 Hz_40ms_envelope_correlation": 0.9876260238068417, "1000–3000 Hz_40ms_modulation_mae_db": 0.5457840431504545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9897898926784972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19144799818836444, "3000–6000 Hz_10ms_envelope_correlation": 0.9770584976285556, "3000–6000 Hz_10ms_modulation_mae_db": 1.5034995913855782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4350261652664065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1232653433396172, "3000–6000 Hz_40ms_envelope_correlation": 0.9903059590400088, "3000–6000 Hz_40ms_modulation_mae_db": 2.113023866453075, "mr_stft_db": 5.722484360943279, "band_mae_db": 0.11930631096646976, "rms_gain_error_db": -0.052794550686543026, "output_peak": 0.35177889734275286}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0950127131074252, "全帯域_10ms_rms_envelope_nrmse": 0.09224634335433984, "全帯域_10ms_envelope_correlation": 0.9880935521907878, "全帯域_10ms_modulation_mae_db": 0.3676671986183753, "全帯域_40ms_log_rms_mae_db": 0.5632998904119478, "全帯域_40ms_rms_envelope_nrmse": 0.04782966507598808, "全帯域_40ms_envelope_correlation": 0.99624842414455, "全帯域_40ms_modulation_mae_db": 0.24891281555141034, "300–1000 Hz_10ms_log_rms_mae_db": 1.803604176089171, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1244633511041015, "300–1000 Hz_10ms_envelope_correlation": 0.9855825934390279, "300–1000 Hz_10ms_modulation_mae_db": 0.5275275883705388, "300–1000 Hz_40ms_log_rms_mae_db": 1.1184341931805384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07351764292192796, "300–1000 Hz_40ms_envelope_correlation": 0.9945297359468444, "300–1000 Hz_40ms_modulation_mae_db": 0.5191350521357395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8155335115578848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20656996738472067, "1000–3000 Hz_10ms_envelope_correlation": 0.954489758890303, "1000–3000 Hz_10ms_modulation_mae_db": 0.2592074617757091, "1000–3000 Hz_40ms_log_rms_mae_db": 0.941272243187533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10721500382616637, "1000–3000 Hz_40ms_envelope_correlation": 0.9856461715472707, "1000–3000 Hz_40ms_modulation_mae_db": 0.3888236323914618, "3000–6000 Hz_10ms_log_rms_mae_db": 2.061783799904545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21746932870196098, "3000–6000 Hz_10ms_envelope_correlation": 0.9722952562901446, "3000–6000 Hz_10ms_modulation_mae_db": 1.3875049448085068, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4378483898525907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1505289600888512, "3000–6000 Hz_40ms_envelope_correlation": 0.9885959975643105, "3000–6000 Hz_40ms_modulation_mae_db": 1.676405851429654, "mr_stft_db": 5.732020349124487, "band_mae_db": 0.20272382041590745, "rms_gain_error_db": -0.06368062580328787, "output_peak": 0.33788278044692455}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.082249681274776, "全帯域_10ms_rms_envelope_nrmse": 0.08158102169814031, "全帯域_10ms_envelope_correlation": 0.9907343238067966, "全帯域_10ms_modulation_mae_db": 0.6147634182726575, "全帯域_40ms_log_rms_mae_db": 0.5086755142129369, "全帯域_40ms_rms_envelope_nrmse": 0.03758553747904505, "全帯域_40ms_envelope_correlation": 0.9976664932573921, "全帯域_40ms_modulation_mae_db": 0.7154897145016043, "300–1000 Hz_10ms_log_rms_mae_db": 1.6930645716801536, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11362892408159388, "300–1000 Hz_10ms_envelope_correlation": 0.987858695624207, "300–1000 Hz_10ms_modulation_mae_db": 0.09721659250907388, "300–1000 Hz_40ms_log_rms_mae_db": 0.9567774428895177, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058254927496857116, "300–1000 Hz_40ms_envelope_correlation": 0.9963580892154491, "300–1000 Hz_40ms_modulation_mae_db": 0.38548546372475984, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8477016992786968, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2102001346118825, "1000–3000 Hz_10ms_envelope_correlation": 0.9526153714606049, "1000–3000 Hz_10ms_modulation_mae_db": 1.02080905347172, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0574802418102895, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11746210531121488, "1000–3000 Hz_40ms_envelope_correlation": 0.9836767456536772, "1000–3000 Hz_40ms_modulation_mae_db": 1.0260723719134612, "3000–6000 Hz_10ms_log_rms_mae_db": 2.02281793206462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2338746434632929, "3000–6000 Hz_10ms_envelope_correlation": 0.9669208226158994, "3000–6000 Hz_10ms_modulation_mae_db": 0.7666389262559333, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3469067503281933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11798195566890204, "3000–6000 Hz_40ms_envelope_correlation": 0.9914234387467192, "3000–6000 Hz_40ms_modulation_mae_db": 1.2922433822505377, "mr_stft_db": 5.733839712149373, "band_mae_db": 0.09440804942699237, "rms_gain_error_db": 0.022088797456782643, "output_peak": 0.36672991447345427}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1664556807149977, "全帯域_10ms_rms_envelope_nrmse": 0.0963041349732368, "全帯域_10ms_envelope_correlation": 0.9870528314567925, "全帯域_10ms_modulation_mae_db": 0.3659352386258508, "全帯域_40ms_log_rms_mae_db": 0.5916302229846249, "全帯域_40ms_rms_envelope_nrmse": 0.04888790627167666, "全帯域_40ms_envelope_correlation": 0.9960449365795047, "全帯域_40ms_modulation_mae_db": 0.14756937026320696, "300–1000 Hz_10ms_log_rms_mae_db": 1.7324627763461136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11796591169632384, "300–1000 Hz_10ms_envelope_correlation": 0.9869573810229042, "300–1000 Hz_10ms_modulation_mae_db": 0.7684898157583347, "300–1000 Hz_40ms_log_rms_mae_db": 1.0267408811647931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06788396688244697, "300–1000 Hz_40ms_envelope_correlation": 0.9952143446022724, "300–1000 Hz_40ms_modulation_mae_db": 0.37349492474510537, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9009830592830042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23016292336093766, "1000–3000 Hz_10ms_envelope_correlation": 0.9428961925848519, "1000–3000 Hz_10ms_modulation_mae_db": 0.7312128927928634, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0547687594032662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12309813165803896, "1000–3000 Hz_40ms_envelope_correlation": 0.9819056853734447, "1000–3000 Hz_40ms_modulation_mae_db": 0.9202780520109094, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9561973753516004, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18734728709773832, "3000–6000 Hz_10ms_envelope_correlation": 0.9785200010416133, "3000–6000 Hz_10ms_modulation_mae_db": 0.7632287303545048, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3476565631984403, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10874158828727924, "3000–6000 Hz_40ms_envelope_correlation": 0.9925062102149076, "3000–6000 Hz_40ms_modulation_mae_db": 1.0571459522344326, "mr_stft_db": 5.726694669864379, "band_mae_db": 0.15024457128823604, "rms_gain_error_db": -0.03119090304562478, "output_peak": 0.3476547312267887}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44560, "output_samples": 44560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.976990466002318, "spectral_convergence": 0.32467339590980343, "spectral_convergence_status": "ok", "amplitude_floor": 9.278908430047933e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.540546210642357, "spectral_convergence": 0.297133312640461, "spectral_convergence_status": "ok", "amplitude_floor": 4.74996132181228e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7051338165037, "spectral_convergence": 0.3146070696711838, "spectral_convergence_status": "ok", "amplitude_floor": 3.301834166389807e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005909300513360037, "output_power": 0.0006076259426278427, "delta_db": 0.1210023153228857, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000598279616407713, "output_power": 0.0005857825357058106, "delta_db": -0.09167786881041265, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005543526355345429, "output_power": 0.0005470889605204798, "delta_db": -0.05728165197852629, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011395641530102642, "output_power": 0.00011015339350765847, "delta_db": -0.14740898236882824, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011253089498164014, "output_power": 0.00011919511848719214, "delta_db": 0.24986696768478917, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6482875672349115e-05, "output_power": 1.6996816524118302e-05, "delta_db": 0.1333460352290243, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13343063689907775, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.0006782751467122318, "pitch": null, "output_rms_dbfs": -27.018365002601442, "output_peak": 0.3454309876247169, "reference_rms_dbfs": -27.019043277748153, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.740890164382791}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 0.00011061650326899306, "log_rms_mae_db": 1.0954404676139857, "rms_envelope_nrmse": 0.09403599127792524, "envelope_correlation": 0.9876406175849994, "rms_gain_error_db": 0.0007276596189668814, "reference_mean_power": 0.00199005080866686, "output_mean_power": 0.0019903842692547834}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 8.439637043999617e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970864143567599, "output_power": 0.0005910755116350541, "reference_active": true, "delta_db": -0.0439422286992462}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010123102438186569, "output_power": 0.00010195282671989838, "reference_active": true, "delta_db": 0.030856398224856925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240115070482855e-05, "output_power": 3.0084204035872692e-05, "reference_active": true, "delta_db": -0.32221908683312184}], "active_band_count": 3, "active_band_mae_db": 0.13233923791907498}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 9.74750993349889e-05, "log_rms_mae_db": 0.5252962067047782, "rms_envelope_nrmse": 0.04841050819083398, "envelope_correlation": 0.9960945660929981, "rms_gain_error_db": 0.0008219880066336182, "reference_mean_power": 0.0020114637230104894, "output_mean_power": 0.0020118444682430603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 6.839204647135984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005172552173341314, "output_power": 0.0005147377428912564, "reference_active": true, "delta_db": -0.021188660577075892}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6641602045504645e-05, "output_power": 5.7438497837100484e-05, "reference_active": true, "delta_db": 0.06067545078912356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.529860845744804e-06, "output_power": 2.9024872253421686e-06, "reference_active": true, "delta_db": 0.5967368337143112}], "active_band_count": 3, "active_band_mae_db": 0.22620031502683688}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 8.308955229004579e-05, "log_rms_mae_db": 1.7209661752993801, "rms_envelope_nrmse": 0.12743106108217972, "envelope_correlation": 0.9846948116515722, "rms_gain_error_db": -0.05621362959707976, "reference_mean_power": 0.000839400894204417, "output_mean_power": 0.0008286059825071987}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.694834819211473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042315114528166547, "output_power": 0.00041346641526198786, "reference_active": true, "delta_db": -0.1005528187765037}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.939611213191948e-05, "output_power": 6.129861484357178e-05, "reference_active": true, "delta_db": -0.538844791591378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4277196064029695e-05, "output_power": 2.5948657539498127e-05, "reference_active": true, "delta_db": 0.2891636870721284}], "active_band_count": 3, "active_band_mae_db": 0.3095204324800034}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 8.008999734930271e-05, "log_rms_mae_db": 1.0054282635837688, "rms_envelope_nrmse": 0.06812082250247481, "envelope_correlation": 0.9950304417592333, "rms_gain_error_db": -0.05622512861070962, "reference_mean_power": 0.0008485551352947845, "output_mean_power": 0.0008376402798379436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 5.733703583211585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003764127527272362, "output_power": 0.0003710408993019824, "reference_active": true, "delta_db": -0.06242544837044192}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8963437868750806e-05, "output_power": 3.570945977909883e-05, "reference_active": true, "delta_db": -0.37873988952547255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.296630500499659e-06, "output_power": 2.622220557800165e-06, "reference_active": true, "delta_db": 0.5757808967423971}], "active_band_count": 3, "active_band_mae_db": 0.33898207821277054}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.2528786526254916e-05, "log_rms_mae_db": 1.886970352763164, "rms_envelope_nrmse": 0.2082322914524617, "envelope_correlation": 0.9533875471060467, "rms_gain_error_db": -0.1552644752307889, "reference_mean_power": 0.0001368277838750774, "output_mean_power": 0.0001320224675025169}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.073610600879068e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.932287229550182e-05, "output_power": 4.0372275349241426e-05, "reference_active": true, "delta_db": 0.11437993714261933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8616503346595646e-06, "output_power": 7.99459449358073e-06, "reference_active": true, "delta_db": 0.6636785719321583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.317401673918684e-06, "output_power": 3.0911254033274484e-06, "reference_active": true, "delta_db": -1.4510583133329558}], "active_band_count": 3, "active_band_mae_db": 0.7430389408025778}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 3.268734932259802e-05, "log_rms_mae_db": 0.9985701552776163, "rms_envelope_nrmse": 0.10607802498881384, "envelope_correlation": 0.9861094835264287, "rms_gain_error_db": -0.15527349241378383, "reference_mean_power": 0.00013831808192096785, "output_mean_power": 0.00013346015000090104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 4.85870788941741e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5019378943818874e-05, "output_power": 3.661785095362284e-05, "reference_active": true, "delta_db": 0.1938441286808571}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.168829067205367e-06, "output_power": 4.684699174805739e-06, "reference_active": true, "delta_db": 0.506676198629818}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163313597471507e-07, "output_power": 2.068090126668768e-07, "reference_active": true, "delta_db": -1.8457278691884615}], "active_band_count": 3, "active_band_mae_db": 0.8487493988330456}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.9606629692154165e-05, "log_rms_mae_db": 2.063320788521023, "rms_envelope_nrmse": 0.214012366981898, "envelope_correlation": 0.9779689616469415, "rms_gain_error_db": 0.7055725736639308, "reference_mean_power": 5.1065139870557587e-05, "output_mean_power": 6.007334060045975e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 2.3044643706212192e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0452875124351046e-05, "output_power": 1.2341397828095185e-05, "reference_active": true, "delta_db": 0.7212859002003074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.855628387226494e-06, "output_power": 3.611156557855241e-06, "reference_active": true, "delta_db": 1.0194462666878334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6522678437727886e-06, "output_power": 1.2263439131380725e-06, "reference_active": true, "delta_db": -1.294681709247544}], "active_band_count": 3, "active_band_mae_db": 1.0118046253785618}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 4.1718716562101114e-05, "log_rms_mae_db": 1.4611715242502572, "rms_envelope_nrmse": 0.14091610054023251, "envelope_correlation": 0.9924896273475271, "rms_gain_error_db": 0.7055555512811617, "reference_mean_power": 5.162171435861935e-05, "output_mean_power": 6.072786018135964e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 2.066669918587524e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0052154991285675e-05, "output_power": 1.1798035228996067e-05, "reference_active": true, "delta_db": 0.6955051240687533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.890809551942203e-06, "output_power": 2.0977943622520424e-06, "reference_active": true, "delta_db": 0.45115126332131683}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8144313206293327e-07, "output_power": 1.2353181060913683e-07, "reference_active": true, "delta_db": -1.6696172711626125}], "active_band_count": 3, "active_band_mae_db": 0.9387578861842275}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8511050374660342, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f742b50901fb7683f6d6d7ddd8f9f320b39711e237ede161648febd423cc4799", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.949903907345022, "standard_deviation": 0.008128000128523874, "minimum": 0.9325535383175255, "maximum": 0.9570429765334897, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8713176339325885, "standard_deviation": 0.06922729878927111, "minimum": 1.7484727374009459, "maximum": 1.9836819079593833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6164692890600519, "standard_deviation": 0.23882303396921284, "minimum": 0.2592074617757091, "maximum": 1.02080905347172, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21546427746257424, "standard_deviation": 0.01628197227663706, "minimum": 0.2006216328640982, "maximum": 0.24979009397367505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9845900796106946, "standard_deviation": 0.0030365663269612377, "minimum": 0.9788788653582663, "maximum": 0.9879238204122731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0043221394808108, "standard_deviation": 0.05866942947287177, "minimum": 0.9203555775407404, "maximum": 1.065442754609217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7182179514108653, "standard_deviation": 0.27135313704239844, "minimum": 0.3888236323914618, "maximum": 1.0528511119808126, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11173795565924137, "standard_deviation": 0.011282784074555748, "minimum": 0.09918318624873407, "maximum": 0.13094402127945515, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9734008812463741, "standard_deviation": 0.004579187226698801, "minimum": 0.9669208226158994, "maximum": 0.9785200010416133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.027611342269145, "standard_deviation": 0.04852861016152496, "minimum": 1.9561973753516004, "maximum": 2.0764102535608973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0976241802040663, "standard_deviation": 0.29620603321350986, "minimum": 0.7632287303545048, "maximum": 1.5034995913855782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21253209424399117, "standard_deviation": 0.015989549241372178, "minimum": 0.18734728709773832, "maximum": 0.2338746434632929, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9902687067699673, "standard_deviation": 0.0017894068565680679, "minimum": 0.987965403252656, "maximum": 0.9925062102149076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4109739759032445, "standard_deviation": 0.04804151657931124, "minimum": 1.3469067503281933, "maximum": 1.4662795531960484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3411728417976405, "standard_deviation": 0.38174401409578795, "minimum": 0.9387578861842275, "maximum": 2.113023866453075, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1314417559243501, "standard_deviation": 0.013744999409536647, "minimum": 0.10874158828727924, "maximum": 0.1505289600888512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.98583444856685, "standard_deviation": 0.0021196659452152686, "minimum": 0.9813049963483381, "maximum": 0.987858695624207, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7331227021475308, "standard_deviation": 0.03533434933263291, "minimum": 1.6930645716801536, "maximum": 1.803604176089171, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5587631703601639, "standard_deviation": 0.32807536542051596, "minimum": 0.09721659250907388, "maximum": 1.0925497850157828, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12260846462180217, "standard_deviation": 0.00869147119529958, "minimum": 0.11362892408159388, "maximum": 0.14077255799951888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949244143570326, "standard_deviation": 0.0010706199786518831, "minimum": 0.9928088977361138, "maximum": 0.9963580892154491, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0314889908020675, "standard_deviation": 0.051613600785563596, "minimum": 0.9567774428895177, "maximum": 1.1184341931805384, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6296359226182369, "standard_deviation": 0.32750582272521633, "minimum": 0.33898207821277054, "maximum": 1.2099151156122985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06934693026672212, "standard_deviation": 0.007115735492971495, "minimum": 0.058254927496857116, "maximum": 0.08248807068634265, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14962618783181258, "standard_deviation": 0.034845412883193656, "minimum": 0.09440804942699237, "maximum": 0.20272382041590745, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.733951215387329, "standard_deviation": 0.010646049372605902, "minimum": 5.722484360943279, "maximum": 5.755046931193839, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3508745401900645, "standard_deviation": 0.008439317511138147, "minimum": 0.33788278044692455, "maximum": 0.36672991447345427, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.02258848541212008, "standard_deviation": 0.03489191674585536, "minimum": -0.06368062580328787, "maximum": 0.023434068343808134, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9880516470414445, "standard_deviation": 0.0017403701134551656, "minimum": 0.9848295536299522, "maximum": 0.9907343238067966, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1352338035149412, "standard_deviation": 0.05370625837110128, "minimum": 1.082249681274776, "maximum": 1.217537006359276, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.49390532802708875, "standard_deviation": 0.22210868346772644, "minimum": 0.13233923791907498, "maximum": 0.8186390529405934, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09226333159346228, "standard_deviation": 0.006636794166435669, "minimum": 0.08158102169814031, "maximum": 0.1039740718409299, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9964727517228578, "standard_deviation": 0.0006644943830313837, "minimum": 0.9954846343305234, "maximum": 0.9976664932573921, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5536587418161866, "standard_deviation": 0.04659792545798265, "minimum": 0.47213092643925164, "maximum": 0.6030159438935415, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4729540214052547, "standard_deviation": 0.25322251157276415, "minimum": 0.14756937026320696, "maximum": 0.832363206958691, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04609057906487675, "standard_deviation": 0.004498488784669752, "minimum": 0.03758553747904505, "maximum": 0.05236669763981951, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_61-70968-0043", "language": "en", "style": "normal", "speaker": "librispeech_61", "source_utterance_id": "61-70968-0043", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_61-70968-0043.wav", "source_file_sha256": "6169375c760230a6f9ee7e3b36c6e8ade864500ce7909f7f994b6b265ba180f0", "decoded_pcm_sha256": "1731c4205d3ab140aea21c6e2df47e27d59a45b1eaa6b80c407cddcad99de7af", "original_source_sha256": "92cde1bfb6bab32cd1428505868f1ca27db04ed17bec9e47e7cbb24e34dc09c3", "corpus_entry_id": "librispeech_61-70968-0043"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.17936897277832, "predicted_mos": 4.17936897277832, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a9abc1caa4fa3514a44723d671d8741bec80f3d35c8bd852c7d5b2bf6c44cca1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9442526339989854, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_61-70968-0043", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "a9abc1caa4fa3514a44723d671d8741bec80f3d35c8bd852c7d5b2bf6c44cca1", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "11395567141c5a03808579bb9837eae0b644cb0d8d9e75b3474c4e55832698e7"}, "rows": [{"case": "librispeech_61-70968-0043", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.8848605155944824, "predicted_mos": 1.8848605155944824, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "886c8de149f43b94b9e4faef65e0e0866552b76104582d8e585cc49418569f25", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0118752569978824, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_61-70968-0043", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "886c8de149f43b94b9e4faef65e0e0866552b76104582d8e585cc49418569f25", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.419497530899699, "全帯域_10ms_rms_envelope_nrmse": 0.13987841916100108, "全帯域_10ms_envelope_correlation": 0.9786911870586884, "全帯域_10ms_modulation_mae_db": 0.17010260107924235, "全帯域_40ms_log_rms_mae_db": 0.727904377586821, "全帯域_40ms_rms_envelope_nrmse": 0.06831564911099561, "全帯域_40ms_envelope_correlation": 0.9946322449870223, "全帯域_40ms_modulation_mae_db": 0.11416023333225346, "300–1000 Hz_10ms_log_rms_mae_db": 2.0569283577627906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1853995002361604, "300–1000 Hz_10ms_envelope_correlation": 0.9740916862034111, "300–1000 Hz_10ms_modulation_mae_db": 0.596267854081782, "300–1000 Hz_40ms_log_rms_mae_db": 1.2851928237919081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0938592214035021, "300–1000 Hz_40ms_envelope_correlation": 0.9935328562445669, "300–1000 Hz_40ms_modulation_mae_db": 0.4388607197727967, "1000–3000 Hz_10ms_log_rms_mae_db": 1.887696331166828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22627218436320148, "1000–3000 Hz_10ms_envelope_correlation": 0.9595162077148502, "1000–3000 Hz_10ms_modulation_mae_db": 1.010179090198638, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0359158561634512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11723575260354836, "1000–3000 Hz_40ms_envelope_correlation": 0.990057514656664, "1000–3000 Hz_40ms_modulation_mae_db": 1.4267466369423467, "3000–6000 Hz_10ms_log_rms_mae_db": 1.883238217342043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23644258866636042, "3000–6000 Hz_10ms_envelope_correlation": 0.9631957430035922, "3000–6000 Hz_10ms_modulation_mae_db": 0.46871178395843227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4450366901273293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17223322146977318, "3000–6000 Hz_40ms_envelope_correlation": 0.97956088852406, "3000–6000 Hz_40ms_modulation_mae_db": 0.62785034352457, "mr_stft_db": 6.028329950756409, "band_mae_db": 0.2931898895997806, "rms_gain_error_db": -0.12482884896172308, "output_peak": 0.49875163455022326}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4254958660102666, "全帯域_10ms_rms_envelope_nrmse": 0.14436246033363404, "全帯域_10ms_envelope_correlation": 0.9773057452585832, "全帯域_10ms_modulation_mae_db": 0.09979093113356226, "全帯域_40ms_log_rms_mae_db": 0.6994350990463858, "全帯域_40ms_rms_envelope_nrmse": 0.07631785453037179, "全帯域_40ms_envelope_correlation": 0.9931732963356786, "全帯域_40ms_modulation_mae_db": 0.06143928882848248, "300–1000 Hz_10ms_log_rms_mae_db": 2.155353096455948, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2004572758589064, "300–1000 Hz_10ms_envelope_correlation": 0.9695482960211499, "300–1000 Hz_10ms_modulation_mae_db": 0.4269762953818626, "300–1000 Hz_40ms_log_rms_mae_db": 1.404293234136925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1183090829732036, "300–1000 Hz_40ms_envelope_correlation": 0.9892489702888078, "300–1000 Hz_40ms_modulation_mae_db": 0.42509463125706765, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8895020685073212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2254384397126815, "1000–3000 Hz_10ms_envelope_correlation": 0.9593207745577791, "1000–3000 Hz_10ms_modulation_mae_db": 0.6460570962894192, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0906903278495563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1055247646720525, "1000–3000 Hz_40ms_envelope_correlation": 0.9908249970655838, "1000–3000 Hz_40ms_modulation_mae_db": 0.7738552747714049, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8120300587353875, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21753246708674484, "3000–6000 Hz_10ms_envelope_correlation": 0.968246089932891, "3000–6000 Hz_10ms_modulation_mae_db": 0.5380247488352785, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3092418368114391, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1596833604569636, "3000–6000 Hz_40ms_envelope_correlation": 0.9819866965786935, "3000–6000 Hz_40ms_modulation_mae_db": 0.5588631460195219, "mr_stft_db": 6.039050560791499, "band_mae_db": 0.16491017390964033, "rms_gain_error_db": -0.09282948834995496, "output_peak": 0.5524524421440485}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4379588027237054, "全帯域_10ms_rms_envelope_nrmse": 0.14577420666502358, "全帯域_10ms_envelope_correlation": 0.9769877482429692, "全帯域_10ms_modulation_mae_db": 0.07117275881285523, "全帯域_40ms_log_rms_mae_db": 0.6993779602612052, "全帯域_40ms_rms_envelope_nrmse": 0.062197215068128545, "全帯域_40ms_envelope_correlation": 0.9954446548209633, "全帯域_40ms_modulation_mae_db": 0.20119283309687155, "300–1000 Hz_10ms_log_rms_mae_db": 2.083554547919881, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1839304524574803, "300–1000 Hz_10ms_envelope_correlation": 0.9743996718286018, "300–1000 Hz_10ms_modulation_mae_db": 0.4668311944628359, "300–1000 Hz_40ms_log_rms_mae_db": 1.2856772937454735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08704222595300003, "300–1000 Hz_40ms_envelope_correlation": 0.9940081471251795, "300–1000 Hz_40ms_modulation_mae_db": 0.4065903793299492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8726344188211848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.233894754985437, "1000–3000 Hz_10ms_envelope_correlation": 0.9564992523776834, "1000–3000 Hz_10ms_modulation_mae_db": 0.8474458249771114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0595481374350335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11290314131762416, "1000–3000 Hz_40ms_envelope_correlation": 0.9891895994757759, "1000–3000 Hz_40ms_modulation_mae_db": 1.1329171512507934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8384495224259707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.222292703303622, "3000–6000 Hz_10ms_envelope_correlation": 0.9677845009537975, "3000–6000 Hz_10ms_modulation_mae_db": 0.19113135201480835, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3749336083914667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16240542777670147, "3000–6000 Hz_40ms_envelope_correlation": 0.9821027433608062, "3000–6000 Hz_40ms_modulation_mae_db": 0.10994145379895015, "mr_stft_db": 6.058462473718066, "band_mae_db": 0.19774874624607264, "rms_gain_error_db": -0.02592077594113482, "output_peak": 0.5078148551207295}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4562844038879537, "全帯域_10ms_rms_envelope_nrmse": 0.15022240209052415, "全帯域_10ms_envelope_correlation": 0.9758614458023745, "全帯域_10ms_modulation_mae_db": 0.09916545136658854, "全帯域_40ms_log_rms_mae_db": 0.6998903943231276, "全帯域_40ms_rms_envelope_nrmse": 0.06970149520741034, "全帯域_40ms_envelope_correlation": 0.9943996995549758, "全帯域_40ms_modulation_mae_db": 0.1405708080529908, "300–1000 Hz_10ms_log_rms_mae_db": 2.174685742414133, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1796606127289921, "300–1000 Hz_10ms_envelope_correlation": 0.9756074268125492, "300–1000 Hz_10ms_modulation_mae_db": 0.33478102600365967, "300–1000 Hz_40ms_log_rms_mae_db": 1.3002853393278668, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08814733356606129, "300–1000 Hz_40ms_envelope_correlation": 0.9938536029310281, "300–1000 Hz_40ms_modulation_mae_db": 0.49836279155772906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9133477923646276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2525284092092009, "1000–3000 Hz_10ms_envelope_correlation": 0.9496373757820644, "1000–3000 Hz_10ms_modulation_mae_db": 0.6482528370038126, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1162743467146805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1219684725473162, "1000–3000 Hz_40ms_envelope_correlation": 0.9874284187848537, "1000–3000 Hz_40ms_modulation_mae_db": 1.3359884160256474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7715695748869105, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2196620369354975, "3000–6000 Hz_10ms_envelope_correlation": 0.9690115273697014, "3000–6000 Hz_10ms_modulation_mae_db": 0.3746093575421559, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3421828007684056, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16104680381014383, "3000–6000 Hz_40ms_envelope_correlation": 0.9827131745363724, "3000–6000 Hz_40ms_modulation_mae_db": 0.23797478284772144, "mr_stft_db": 6.04333355380341, "band_mae_db": 0.14717945051091236, "rms_gain_error_db": 0.0317173230496032, "output_peak": 0.5117972869830796}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3812873295267452, "全帯域_10ms_rms_envelope_nrmse": 0.1501537373108217, "全帯域_10ms_envelope_correlation": 0.9757873897121669, "全帯域_10ms_modulation_mae_db": 0.27443604580055586, "全帯域_40ms_log_rms_mae_db": 0.6185447068104694, "全帯域_40ms_rms_envelope_nrmse": 0.07247641882578076, "全帯域_40ms_envelope_correlation": 0.9938839386496653, "全帯域_40ms_modulation_mae_db": 0.2612552195885822, "300–1000 Hz_10ms_log_rms_mae_db": 2.1419105988246274, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18867905772228988, "300–1000 Hz_10ms_envelope_correlation": 0.9731828390464328, "300–1000 Hz_10ms_modulation_mae_db": 0.14786744334846358, "300–1000 Hz_40ms_log_rms_mae_db": 1.3694036870087904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09625388837393982, "300–1000 Hz_40ms_envelope_correlation": 0.9926193520722634, "300–1000 Hz_40ms_modulation_mae_db": 0.4119023761809379, "1000–3000 Hz_10ms_log_rms_mae_db": 1.86907328313938, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23759959305607573, "1000–3000 Hz_10ms_envelope_correlation": 0.9553368853381576, "1000–3000 Hz_10ms_modulation_mae_db": 0.3584369951638649, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0416451160801772, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11481852142043462, "1000–3000 Hz_40ms_envelope_correlation": 0.9888352802797649, "1000–3000 Hz_40ms_modulation_mae_db": 0.8672202166285988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7844201150158019, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23196111855528, "3000–6000 Hz_10ms_envelope_correlation": 0.9635429862105607, "3000–6000 Hz_10ms_modulation_mae_db": 0.5560583287888857, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3500063721196844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17293311148386856, "3000–6000 Hz_40ms_envelope_correlation": 0.9787429918634252, "3000–6000 Hz_40ms_modulation_mae_db": 0.29022761781569095, "mr_stft_db": 6.039894786390854, "band_mae_db": 0.14889446010611623, "rms_gain_error_db": 0.04057733239281801, "output_peak": 0.5353399173756587}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4148015128870663, "全帯域_10ms_rms_envelope_nrmse": 0.14908303717427893, "全帯域_10ms_envelope_correlation": 0.9757731802774241, "全帯域_10ms_modulation_mae_db": 0.14642219741667095, "全帯域_40ms_log_rms_mae_db": 0.6708067830113672, "全帯域_40ms_rms_envelope_nrmse": 0.07413009262121505, "全帯域_40ms_envelope_correlation": 0.9937595453982631, "全帯域_40ms_modulation_mae_db": 0.43635538702131077, "300–1000 Hz_10ms_log_rms_mae_db": 2.184126941794237, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21421562012405657, "300–1000 Hz_10ms_envelope_correlation": 0.9651203405439819, "300–1000 Hz_10ms_modulation_mae_db": 0.5864339106611488, "300–1000 Hz_40ms_log_rms_mae_db": 1.4374781594540051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11588693046707908, "300–1000 Hz_40ms_envelope_correlation": 0.9898223454130806, "300–1000 Hz_40ms_modulation_mae_db": 0.12422451434260508, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8366040275870836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21894335525970232, "1000–3000 Hz_10ms_envelope_correlation": 0.9616943816048544, "1000–3000 Hz_10ms_modulation_mae_db": 0.744987171334793, "1000–3000 Hz_40ms_log_rms_mae_db": 1.021485064758748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10649975183815429, "1000–3000 Hz_40ms_envelope_correlation": 0.9905368487745493, "1000–3000 Hz_40ms_modulation_mae_db": 0.8707096118562836, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7509965030967836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22565312746902622, "3000–6000 Hz_10ms_envelope_correlation": 0.9664257662163294, "3000–6000 Hz_10ms_modulation_mae_db": 0.46655456203873696, "3000–6000 Hz_40ms_log_rms_mae_db": 1.31877184495423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16306406947470875, "3000–6000 Hz_40ms_envelope_correlation": 0.9816612648897414, "3000–6000 Hz_40ms_modulation_mae_db": 0.45295828964683865, "mr_stft_db": 6.028852926308809, "band_mae_db": 0.16268968476558734, "rms_gain_error_db": -0.15395668018911443, "output_peak": 0.5203265894065922}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4225334013506221, "全帯域_10ms_rms_envelope_nrmse": 0.13849424155145723, "全帯域_10ms_envelope_correlation": 0.9791193935875444, "全帯域_10ms_modulation_mae_db": 0.23494220724120654, "全帯域_40ms_log_rms_mae_db": 0.6119049153280494, "全帯域_40ms_rms_envelope_nrmse": 0.053096079921365426, "全帯域_40ms_envelope_correlation": 0.9967256940076484, "全帯域_40ms_modulation_mae_db": 0.26476657249422997, "300–1000 Hz_10ms_log_rms_mae_db": 2.1880645104949235, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17999281079413948, "300–1000 Hz_10ms_envelope_correlation": 0.9755197999475299, "300–1000 Hz_10ms_modulation_mae_db": 0.5437177711430897, "300–1000 Hz_40ms_log_rms_mae_db": 1.366747090443644, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07994132979301001, "300–1000 Hz_40ms_envelope_correlation": 0.995246817752765, "300–1000 Hz_40ms_modulation_mae_db": 0.1730396693358894, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8446543337572214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2243821485203147, "1000–3000 Hz_10ms_envelope_correlation": 0.9597544677351751, "1000–3000 Hz_10ms_modulation_mae_db": 0.6440940478112732, "1000–3000 Hz_40ms_log_rms_mae_db": 1.076598079504484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10972065226976548, "1000–3000 Hz_40ms_envelope_correlation": 0.990456453843352, "1000–3000 Hz_40ms_modulation_mae_db": 0.893639959259918, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9107535805634048, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2388171161504621, "3000–6000 Hz_10ms_envelope_correlation": 0.9616906436991238, "3000–6000 Hz_10ms_modulation_mae_db": 0.7268554610733463, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4435858038872258, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17116746252225806, "3000–6000 Hz_40ms_envelope_correlation": 0.979359813070421, "3000–6000 Hz_40ms_modulation_mae_db": 0.7544076598302029, "mr_stft_db": 6.043528995097714, "band_mae_db": 0.1681921922429104, "rms_gain_error_db": -0.06927867138436362, "output_peak": 0.5610767834000107}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.380752107760703, "全帯域_10ms_rms_envelope_nrmse": 0.13115994913453902, "全帯域_10ms_envelope_correlation": 0.9813337830776551, "全帯域_10ms_modulation_mae_db": 0.2148517225185873, "全帯域_40ms_log_rms_mae_db": 0.5963613729410608, "全帯域_40ms_rms_envelope_nrmse": 0.06069726996249515, "全帯域_40ms_envelope_correlation": 0.995860678440044, "全帯域_40ms_modulation_mae_db": 0.11858943572950954, "300–1000 Hz_10ms_log_rms_mae_db": 2.114869028654191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18278094898657227, "300–1000 Hz_10ms_envelope_correlation": 0.9749159591011326, "300–1000 Hz_10ms_modulation_mae_db": 0.5515407072235364, "300–1000 Hz_40ms_log_rms_mae_db": 1.3117523590460105, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10765872645876085, "300–1000 Hz_40ms_envelope_correlation": 0.9913816556723299, "300–1000 Hz_40ms_modulation_mae_db": 0.4455804344981802, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8846260013527931, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23165409885512991, "1000–3000 Hz_10ms_envelope_correlation": 0.9573795136627357, "1000–3000 Hz_10ms_modulation_mae_db": 0.9200958355813335, "1000–3000 Hz_40ms_log_rms_mae_db": 1.044867440932652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1156749162600593, "1000–3000 Hz_40ms_envelope_correlation": 0.9886812623271067, "1000–3000 Hz_40ms_modulation_mae_db": 1.2207252795079475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.834500730532216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23614970648091285, "3000–6000 Hz_10ms_envelope_correlation": 0.9622776153252842, "3000–6000 Hz_10ms_modulation_mae_db": 0.45856640421179157, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3947744220923814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1718320590640183, "3000–6000 Hz_40ms_envelope_correlation": 0.9790058061463088, "3000–6000 Hz_40ms_modulation_mae_db": 0.2981098519857404, "mr_stft_db": 6.040553596542144, "band_mae_db": 0.12724566711136157, "rms_gain_error_db": -0.1419640298112981, "output_peak": 0.4994290324167706}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.255415776521359, "spectral_convergence": 0.5027998922621562, "spectral_convergence_status": "ok", "amplitude_floor": 1.4385235391470026e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.759514319478462, "spectral_convergence": 0.4257615718993632, "spectral_convergence_status": "ok", "amplitude_floor": 9.243439915833013e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.070059756269408, "spectral_convergence": 0.46452689857941354, "spectral_convergence_status": "ok", "amplitude_floor": 6.315989368278398e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00029756845932849967, "output_power": 0.00032962547399650947, "delta_db": 0.44433870899164474, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006886429842704082, "output_power": 0.0006510330003823987, "delta_db": -0.24391124163223538, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003825307900717409, "output_power": 0.00035660466831201364, "delta_db": -0.3047937322465828, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019304685978074288, "output_power": 0.00017007636962488787, "delta_db": -0.5501876449583722, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010892090215881134, "output_power": 0.00011184058130603258, "delta_db": 0.11488185897426213, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.08275993922433e-05, "output_power": 8.272986113947134e-05, "delta_db": 0.1010261507955862, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2931898895997806, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12482884896172308, "pitch": null, "output_rms_dbfs": -27.69063421445782, "output_peak": 0.49875163455022326, "reference_rms_dbfs": -27.565805365496097, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.028329950756409}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017313921740261034, "log_rms_mae_db": 1.419497530899699, "rms_envelope_nrmse": 0.13987841916100108, "envelope_correlation": 0.9786911870586884, "rms_gain_error_db": -0.1182174219822836, "reference_mean_power": 0.0017487207983142375, "output_mean_power": 0.0017017616528550356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.577112847045013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002769005691966373, "output_power": 0.00027449041162860925, "reference_active": true, "delta_db": -0.03796670003734999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.377497445156709e-05, "output_power": 3.1709078450128274e-05, "reference_active": true, "delta_db": -0.27411408718787134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4953137492438862e-05, "output_power": 1.4285966422085793e-05, "reference_active": true, "delta_db": -0.19822701601250578}], "active_band_count": 3, "active_band_mae_db": 0.17010260107924235}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015508669364693045, "log_rms_mae_db": 0.727904377586821, "rms_envelope_nrmse": 0.06831564911099561, "envelope_correlation": 0.9946322449870223, "rms_gain_error_db": -0.11629687830531238, "reference_mean_power": 0.0017487025727625826, "output_mean_power": 0.0017024966309313258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.09499315514524e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025296787730251264, "output_power": 0.0002512741163014449, "reference_active": true, "delta_db": -0.029176222224426278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6687984786311574e-05, "output_power": 1.575784233051344e-05, "reference_active": true, "delta_db": -0.24907144445377566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.96993861330795e-07, "output_power": 6.867610402637176e-07, "reference_active": true, "delta_db": -0.06423303331855845}], "active_band_count": 3, "active_band_mae_db": 0.11416023333225346}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014635391485956695, "log_rms_mae_db": 2.0569283577627906, "rms_envelope_nrmse": 0.1853995002361604, "envelope_correlation": 0.9740916862034111, "rms_gain_error_db": -0.25548331513383427, "reference_mean_power": 0.0008142561216173605, "output_mean_power": 0.0007677374056397972}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2354521678013805e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001985819153870851, "output_power": 0.00018236061209095654, "reference_active": true, "delta_db": -0.3700865405521811}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.365393218996536e-05, "output_power": 2.110711364240935e-05, "reference_active": true, "delta_db": -0.49474499036069985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.646479803369054e-06, "output_power": 5.3727234444003765e-06, "reference_active": true, "delta_db": -0.9239720313324652}], "active_band_count": 3, "active_band_mae_db": 0.596267854081782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012544182666407365, "log_rms_mae_db": 1.2851928237919081, "rms_envelope_nrmse": 0.0938592214035021, "envelope_correlation": 0.9935328562445669, "rms_gain_error_db": -0.25164568350557476, "reference_mean_power": 0.0008126713020357493, "output_mean_power": 0.0007669205152558623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0135181965386276e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018492598004057708, "output_power": 0.00016926448370241368, "reference_active": true, "delta_db": -0.3843208812284981}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0725232186915858e-05, "output_power": 9.916506011289311e-06, "reference_active": true, "delta_db": -0.34048023370740876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.105637817013119e-07, "output_power": 3.5826347211779277e-07, "reference_active": true, "delta_db": -0.5917810443824834}], "active_band_count": 3, "active_band_mae_db": 0.4388607197727967}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.111876173726916e-05, "log_rms_mae_db": 1.887696331166828, "rms_envelope_nrmse": 0.22627218436320148, "envelope_correlation": 0.9595162077148502, "rms_gain_error_db": -0.4376876879445085, "reference_mean_power": 0.0002425197498003479, "output_mean_power": 0.00021926955878792095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1105776313023488e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8957411226134796e-05, "output_power": 3.122281793114943e-05, "reference_active": true, "delta_db": -0.9611799312493359}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.645635870343935e-06, "output_power": 4.854049116332242e-06, "reference_active": true, "delta_db": -0.6560869730230976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.039589927108413e-06, "output_power": 2.917495829918212e-06, "reference_active": true, "delta_db": -1.4132703663234805}], "active_band_count": 3, "active_band_mae_db": 1.010179090198638}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.178958171420995e-05, "log_rms_mae_db": 1.0359158561634512, "rms_envelope_nrmse": 0.11723575260354836, "envelope_correlation": 0.990057514656664, "rms_gain_error_db": -0.43930990419321847, "reference_mean_power": 0.00024219152470715313, "output_mean_power": 0.00021889102295639184}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0514876259433796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7077906526859964e-05, "output_power": 2.934477658522729e-05, "reference_active": true, "delta_db": -1.015843978733677}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.355351339185207e-06, "output_power": 2.6120361877641684e-06, "reference_active": true, "delta_db": -1.0875881244337766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8063796091292985e-07, "output_power": 1.7000654265035505e-07, "reference_active": true, "delta_db": -2.1768078076595865}], "active_band_count": 3, "active_band_mae_db": 1.4267466369423467}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3400740982798506e-05, "log_rms_mae_db": 1.883238217342043, "rms_envelope_nrmse": 0.23644258866636042, "envelope_correlation": 0.9631957430035922, "rms_gain_error_db": 0.17292848695603522, "reference_mean_power": 0.0001023804664102342, "output_mean_power": 0.00010653932782307717}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5067174911166034e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7037143173191167e-05, "output_power": 2.7430139976008283e-05, "reference_active": true, "delta_db": 0.06267222997490902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.303852533122454e-06, "output_power": 4.467334776177626e-06, "reference_active": true, "delta_db": 0.16191116867187583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8801329619345497e-06, "output_power": 2.4679920632735633e-06, "reference_active": true, "delta_db": 1.181551953228512}], "active_band_count": 3, "active_band_mae_db": 0.46871178395843227}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.152733550445532e-05, "log_rms_mae_db": 1.4450366901273293, "rms_envelope_nrmse": 0.17223322146977318, "envelope_correlation": 0.97956088852406, "rms_gain_error_db": 0.17283196080102753, "reference_mean_power": 0.0001028923086971688, "output_mean_power": 0.00010706958222633111}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.218443109204273e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6964722520221083e-05, "output_power": 2.725055720045413e-05, "reference_active": true, "delta_db": 0.04579431304121051}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9782668045768803e-06, "output_power": 2.3856010943514755e-06, "reference_active": true, "delta_db": 0.8131296176242987}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5223765481589483e-07, "output_power": 1.9274574307046228e-07, "reference_active": true, "delta_db": 1.024627099908201}], "active_band_count": 3, "active_band_mae_db": 0.62785034352457}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7549530772099197, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "986daf52d2c1de52254600d55377335544baa8479b3dadc275b58c79593f9b83", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9573923573466625, "standard_deviation": 0.003731698526006779, "minimum": 0.9496373757820644, "maximum": 0.9616943816048544, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.874767282087055, "standard_deviation": 0.02498604965252295, "minimum": 1.8366040275870836, "maximum": 1.9133477923646276, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7274436122950307, "standard_deviation": 0.2027561922182432, "minimum": 0.3584369951638649, "maximum": 1.010179090198638, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23133912299521792, "standard_deviation": 0.010401622690725257, "minimum": 0.21894335525970232, "maximum": 0.2525284092092009, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9895012969009562, "standard_deviation": 0.001168409578150776, "minimum": 0.9874284187848537, "maximum": 0.9908249970655838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0608780461798477, "standard_deviation": 0.03163987843699539, "minimum": 1.021485064758748, "maximum": 1.1162743467146805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0652253182803675, "standard_deviation": 0.24615708084495566, "minimum": 0.7738552747714049, "maximum": 1.4267466369423467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11304324661611936, "standard_deviation": 0.005576970599333513, "minimum": 0.1055247646720525, "maximum": 0.1219684725473162, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.96527185908891, "standard_deviation": 0.002917089977964835, "minimum": 0.9616906436991238, "maximum": 0.9690115273697014, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8232447878248148, "standard_deviation": 0.054952119727637964, "minimum": 1.7509965030967836, "maximum": 1.9107535805634048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4725639998079294, "standard_deviation": 0.15351449102594808, "minimum": 0.19113135201480835, "maximum": 0.7268554610733463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22856385808098825, "standard_deviation": 0.008324009278365555, "minimum": 0.21753246708674484, "maximum": 0.2388171161504621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9806416723712286, "standard_deviation": 0.001619824137154935, "minimum": 0.9787429918634252, "maximum": 0.9827131745363724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3723166723940203, "standard_deviation": 0.05227669803932611, "minimum": 1.3092418368114391, "maximum": 1.4450366901273293, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4162916431836546, "standard_deviation": 0.21917280079733267, "minimum": 0.10994145379895015, "maximum": 0.7544076598302029, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16679568950730447, "standard_deviation": 0.005713973065809297, "minimum": 0.1596833604569636, "maximum": 0.17293311148386856, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9727982524380987, "standard_deviation": 0.0036582769679092174, "minimum": 0.9651203405439819, "maximum": 0.9756074268125492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1374366030400913, "standard_deviation": 0.04835262932754643, "minimum": 2.0569283577627906, "maximum": 2.1880645104949235, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.45680202528829733, "standard_deviation": 0.15311282787643862, "minimum": 0.14786744334846358, "maximum": 0.596267854081782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18938953486357468, "standard_deviation": 0.012021884160294835, "minimum": 0.1796606127289921, "maximum": 0.21421562012405657, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9924642184375027, "standard_deviation": 0.0021284559492570337, "minimum": 0.9892489702888078, "maximum": 0.995246817752765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.345103748369328, "standard_deviation": 0.05773978038751168, "minimum": 1.2851928237919081, "maximum": 1.4374781594540051, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3654569395343944, "standard_deviation": 0.13737078940060413, "minimum": 0.12422451434260508, "maximum": 0.49836279155772906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0983873423735696, "standard_deviation": 0.014081143558477319, "minimum": 0.07994132979301001, "maximum": 0.1183090829732036, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17625628306154767, "standard_deviation": 0.05142176120387347, "minimum": 0.12724566711136157, "maximum": 0.2931898895997806, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.0402508554261125, "standard_deviation": 0.00945358715486245, "minimum": 6.028329950756409, "maximum": 6.058462473718066, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5233735676746392, "standard_deviation": 0.023830518214670172, "minimum": 0.49875163455022326, "maximum": 0.5610767834000107, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06706047989939598, "standard_deviation": 0.07578453129061219, "minimum": -0.15395668018911443, "maximum": 0.04057733239281801, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9776074841271758, "standard_deviation": 0.0019837471456057233, "minimum": 0.9757731802774241, "maximum": 0.9813337830776551, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.417326369380845, "standard_deviation": 0.025870163356754582, "minimum": 1.380752107760703, "maximum": 1.4562844038879537, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.16386048942115863, "standard_deviation": 0.07280512125728092, "minimum": 0.07117275881285523, "maximum": 0.27443604580055586, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14364105667765997, "standard_deviation": 0.006727769879176061, "minimum": 0.13115994913453902, "maximum": 0.15022240209052415, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9947349690242826, "standard_deviation": 0.0011935149145080094, "minimum": 0.9931732963356786, "maximum": 0.9967256940076484, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6655282011635608, "standard_deviation": 0.04965862131599061, "minimum": 0.5963613729410608, "maximum": 0.727904377586821, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.19979122226802887, "standard_deviation": 0.11984448197047433, "minimum": 0.06143928882848248, "maximum": 0.43635538702131077, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06711650940597033, "standard_deviation": 0.00786204583496528, "minimum": 0.053096079921365426, "maximum": 0.07631785453037179, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0043", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.9207508563995361, "predicted_mos": 1.9207508563995361, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5b0f88017632c276be1a340665a41f87d31a4123105d843a9d3b8f82ed141054", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.114008800126612, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4648517086877608, "全帯域_10ms_rms_envelope_nrmse": 0.14296068468807885, "全帯域_10ms_envelope_correlation": 0.9777810027215333, "全帯域_10ms_modulation_mae_db": 0.09965572264687207, "全帯域_40ms_log_rms_mae_db": 0.7251952782049389, "全帯域_40ms_rms_envelope_nrmse": 0.06888527567292964, "全帯域_40ms_envelope_correlation": 0.9946945468971256, "全帯域_40ms_modulation_mae_db": 0.08276749087762324, "300–1000 Hz_10ms_log_rms_mae_db": 2.032268927541266, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19097155334672578, "300–1000 Hz_10ms_envelope_correlation": 0.9724498716108391, "300–1000 Hz_10ms_modulation_mae_db": 0.37590111707039703, "300–1000 Hz_40ms_log_rms_mae_db": 1.2206103509484512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09819157568731433, "300–1000 Hz_40ms_envelope_correlation": 0.9928115902888524, "300–1000 Hz_40ms_modulation_mae_db": 0.2736577408018182, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8995866244929105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23187962631704503, "1000–3000 Hz_10ms_envelope_correlation": 0.9575995593606291, "1000–3000 Hz_10ms_modulation_mae_db": 1.02169527652839, "1000–3000 Hz_40ms_log_rms_mae_db": 1.007486175531512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11989214945291635, "1000–3000 Hz_40ms_envelope_correlation": 0.9900294160921788, "1000–3000 Hz_40ms_modulation_mae_db": 1.4179920814890188, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7774019177945468, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2356729356468206, "3000–6000 Hz_10ms_envelope_correlation": 0.963335171357586, "3000–6000 Hz_10ms_modulation_mae_db": 0.35649598489885537, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3396705554866732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1691943895192144, "3000–6000 Hz_40ms_envelope_correlation": 0.9801298147092224, "3000–6000 Hz_40ms_modulation_mae_db": 0.4745375003797491, "mr_stft_db": 5.988034123493182, "band_mae_db": 0.2500367497832576, "rms_gain_error_db": -0.16582802689570153, "output_peak": 0.49848040870293114}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4743682902063344, "全帯域_10ms_rms_envelope_nrmse": 0.14473612307891387, "全帯域_10ms_envelope_correlation": 0.9771883672934484, "全帯域_10ms_modulation_mae_db": 0.08586681680166193, "全帯域_40ms_log_rms_mae_db": 0.7188331376398837, "全帯域_40ms_rms_envelope_nrmse": 0.07511333713547427, "全帯域_40ms_envelope_correlation": 0.9934309946433899, "全帯域_40ms_modulation_mae_db": 0.13037594791428614, "300–1000 Hz_10ms_log_rms_mae_db": 2.074589822708604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20027549873460274, "300–1000 Hz_10ms_envelope_correlation": 0.9696535792132196, "300–1000 Hz_10ms_modulation_mae_db": 0.5685422944293085, "300–1000 Hz_40ms_log_rms_mae_db": 1.3051117213315502, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11495392969665531, "300–1000 Hz_40ms_envelope_correlation": 0.990031582156576, "300–1000 Hz_40ms_modulation_mae_db": 0.47569594431691326, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8669120043788232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22835878463688233, "1000–3000 Hz_10ms_envelope_correlation": 0.9583988670700726, "1000–3000 Hz_10ms_modulation_mae_db": 0.6919626682557922, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0710535699026706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10404479177003954, "1000–3000 Hz_40ms_envelope_correlation": 0.9908980939793584, "1000–3000 Hz_40ms_modulation_mae_db": 0.7759717477787063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.747450073740001, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21665852435779742, "3000–6000 Hz_10ms_envelope_correlation": 0.9686650547325966, "3000–6000 Hz_10ms_modulation_mae_db": 0.4355336056856185, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2677708020782956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.158220896211601, "3000–6000 Hz_40ms_envelope_correlation": 0.9823298017626276, "3000–6000 Hz_40ms_modulation_mae_db": 0.45307538688774335, "mr_stft_db": 5.9982963076044475, "band_mae_db": 0.1283553860766102, "rms_gain_error_db": -0.11103104567985285, "output_peak": 0.5505273698389656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4690860777723014, "全帯域_10ms_rms_envelope_nrmse": 0.14873805978526594, "全帯域_10ms_envelope_correlation": 0.9760827988735903, "全帯域_10ms_modulation_mae_db": 0.2065633182880182, "全帯域_40ms_log_rms_mae_db": 0.7132914209353421, "全帯域_40ms_rms_envelope_nrmse": 0.0649547132305954, "全帯域_40ms_envelope_correlation": 0.9950496408173883, "全帯域_40ms_modulation_mae_db": 0.1165558768906391, "300–1000 Hz_10ms_log_rms_mae_db": 2.0651929585191713, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1830779363363986, "300–1000 Hz_10ms_envelope_correlation": 0.9747028007166633, "300–1000 Hz_10ms_modulation_mae_db": 0.33156550113412414, "300–1000 Hz_40ms_log_rms_mae_db": 1.2422273919405704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08585406883727742, "300–1000 Hz_40ms_envelope_correlation": 0.9941126633931966, "300–1000 Hz_40ms_modulation_mae_db": 0.3138516400879264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8886057040908926, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2347690627527241, "1000–3000 Hz_10ms_envelope_correlation": 0.9561295681978942, "1000–3000 Hz_10ms_modulation_mae_db": 1.0291468093438654, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0581349175711001, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11135937050832174, "1000–3000 Hz_40ms_envelope_correlation": 0.9895399154776943, "1000–3000 Hz_40ms_modulation_mae_db": 1.479505175814829, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7399663191419268, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22142441791590697, "3000–6000 Hz_10ms_envelope_correlation": 0.9684389891303543, "3000–6000 Hz_10ms_modulation_mae_db": 0.14625908825261544, "3000–6000 Hz_40ms_log_rms_mae_db": 1.30531549197263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16265139375002013, "3000–6000 Hz_40ms_envelope_correlation": 0.9822707277325474, "3000–6000 Hz_40ms_modulation_mae_db": 0.09821544207287318, "mr_stft_db": 6.018315622365424, "band_mae_db": 0.1679170834022582, "rms_gain_error_db": -0.01913880393811218, "output_peak": 0.50556051967468}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4807546105630307, "全帯域_10ms_rms_envelope_nrmse": 0.15114361551147165, "全帯域_10ms_envelope_correlation": 0.9755347300517849, "全帯域_10ms_modulation_mae_db": 0.09478108493911741, "全帯域_40ms_log_rms_mae_db": 0.7367552975476758, "全帯域_40ms_rms_envelope_nrmse": 0.07420422287873657, "全帯域_40ms_envelope_correlation": 0.9936281376927395, "全帯域_40ms_modulation_mae_db": 0.0860469515656075, "300–1000 Hz_10ms_log_rms_mae_db": 2.0996253190896037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1770835872965916, "300–1000 Hz_10ms_envelope_correlation": 0.9763620223021281, "300–1000 Hz_10ms_modulation_mae_db": 0.3245753837649215, "300–1000 Hz_40ms_log_rms_mae_db": 1.1999435716242262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08996778522833583, "300–1000 Hz_40ms_envelope_correlation": 0.9935731393845655, "300–1000 Hz_40ms_modulation_mae_db": 0.46152657657804513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9117044620690773, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2609900432849634, "1000–3000 Hz_10ms_envelope_correlation": 0.9458907671979543, "1000–3000 Hz_10ms_modulation_mae_db": 0.7139977012329486, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1097530828586024, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1309614145208952, "1000–3000 Hz_40ms_envelope_correlation": 0.9854749187529301, "1000–3000 Hz_40ms_modulation_mae_db": 1.5372287685267043, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6817314586575458, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2185184954863077, "3000–6000 Hz_10ms_envelope_correlation": 0.9688639035108698, "3000–6000 Hz_10ms_modulation_mae_db": 0.2711347202215999, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2379242468206195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15802132828656568, "3000–6000 Hz_40ms_envelope_correlation": 0.9829369483370992, "3000–6000 Hz_40ms_modulation_mae_db": 0.22143678796392238, "mr_stft_db": 6.002209254366821, "band_mae_db": 0.09532041622522384, "rms_gain_error_db": 0.022724458617896653, "output_peak": 0.5120324000144294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3749665101470385, "全帯域_10ms_rms_envelope_nrmse": 0.1486093964828747, "全帯域_10ms_envelope_correlation": 0.9761584073506347, "全帯域_10ms_modulation_mae_db": 0.25766649892342114, "全帯域_40ms_log_rms_mae_db": 0.6331293770798235, "全帯域_40ms_rms_envelope_nrmse": 0.07152115744780212, "全帯域_40ms_envelope_correlation": 0.9939855837637339, "全帯域_40ms_modulation_mae_db": 0.23979730553650613, "300–1000 Hz_10ms_log_rms_mae_db": 2.1047175785805265, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1860301621865098, "300–1000 Hz_10ms_envelope_correlation": 0.9740226154260565, "300–1000 Hz_10ms_modulation_mae_db": 0.29994788020626073, "300–1000 Hz_40ms_log_rms_mae_db": 1.3047815817032482, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09375449759857026, "300–1000 Hz_40ms_envelope_correlation": 0.9930237737559082, "300–1000 Hz_40ms_modulation_mae_db": 0.6050855272069696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8683066593405964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24262199138811194, "1000–3000 Hz_10ms_envelope_correlation": 0.9530754459377337, "1000–3000 Hz_10ms_modulation_mae_db": 0.4397606458436974, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0272137653331477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12170387697969674, "1000–3000 Hz_40ms_envelope_correlation": 0.9874588495923736, "1000–3000 Hz_40ms_modulation_mae_db": 0.8943197134403116, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7094945092536278, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23408543584655864, "3000–6000 Hz_10ms_envelope_correlation": 0.9629772899463954, "3000–6000 Hz_10ms_modulation_mae_db": 0.5501352664749912, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2553079487928704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17127437223289788, "3000–6000 Hz_40ms_envelope_correlation": 0.9790848833165456, "3000–6000 Hz_40ms_modulation_mae_db": 0.29206174364015247, "mr_stft_db": 6.002611644390097, "band_mae_db": 0.14706459148923098, "rms_gain_error_db": 0.010467864972008814, "output_peak": 0.5276000847769026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.425105973200045, "全帯域_10ms_rms_envelope_nrmse": 0.14664711424595261, "全帯域_10ms_envelope_correlation": 0.9765700346279147, "全帯域_10ms_modulation_mae_db": 0.14248195006905126, "全帯域_40ms_log_rms_mae_db": 0.6675033289957889, "全帯域_40ms_rms_envelope_nrmse": 0.07183373328687291, "全帯域_40ms_envelope_correlation": 0.9941574725242087, "全帯域_40ms_modulation_mae_db": 0.3886271601152051, "300–1000 Hz_10ms_log_rms_mae_db": 2.1356690647874093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21152093365742441, "300–1000 Hz_10ms_envelope_correlation": 0.965981318700925, "300–1000 Hz_10ms_modulation_mae_db": 0.6058282354225946, "300–1000 Hz_40ms_log_rms_mae_db": 1.3288651334897885, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1111970669470024, "300–1000 Hz_40ms_envelope_correlation": 0.9905103595075649, "300–1000 Hz_40ms_modulation_mae_db": 0.1996661516799291, "1000–3000 Hz_10ms_log_rms_mae_db": 1.833451300872556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2175027431266967, "1000–3000 Hz_10ms_envelope_correlation": 0.9622362891476499, "1000–3000 Hz_10ms_modulation_mae_db": 0.6947586247824424, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9782314497002091, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10507378975412143, "1000–3000 Hz_40ms_envelope_correlation": 0.9908591260902615, "1000–3000 Hz_40ms_modulation_mae_db": 0.8901238151844066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6583825083756647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22809115481976763, "3000–6000 Hz_10ms_envelope_correlation": 0.9658258619037076, "3000–6000 Hz_10ms_modulation_mae_db": 0.29978616865818397, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2389378140915135, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1642881912649423, "3000–6000 Hz_40ms_envelope_correlation": 0.9813839525949841, "3000–6000 Hz_40ms_modulation_mae_db": 0.185512806755741, "mr_stft_db": 5.9961413420641305, "band_mae_db": 0.12301490810049477, "rms_gain_error_db": -0.1513089041268938, "output_peak": 0.5256905187108657}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4679569505281012, "全帯域_10ms_rms_envelope_nrmse": 0.14288924281441542, "全帯域_10ms_envelope_correlation": 0.9777576900150499, "全帯域_10ms_modulation_mae_db": 0.17660946841238265, "全帯域_40ms_log_rms_mae_db": 0.6339178396473081, "全帯域_40ms_rms_envelope_nrmse": 0.05360491364142644, "全帯域_40ms_envelope_correlation": 0.9966859883123801, "全帯域_40ms_modulation_mae_db": 0.24761672589319503, "300–1000 Hz_10ms_log_rms_mae_db": 2.1456700655816796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1855436785066869, "300–1000 Hz_10ms_envelope_correlation": 0.9739552135038141, "300–1000 Hz_10ms_modulation_mae_db": 0.6665612495797189, "300–1000 Hz_40ms_log_rms_mae_db": 1.2884559177511534, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0792824708124403, "300–1000 Hz_40ms_envelope_correlation": 0.995345707960059, "300–1000 Hz_40ms_modulation_mae_db": 0.3102843597611299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8782293429216748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22832713732475202, "1000–3000 Hz_10ms_envelope_correlation": 0.958343538917163, "1000–3000 Hz_10ms_modulation_mae_db": 0.7028097086077961, "1000–3000 Hz_40ms_log_rms_mae_db": 1.054361663145133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10793411353839684, "1000–3000 Hz_40ms_envelope_correlation": 0.9910077997358561, "1000–3000 Hz_40ms_modulation_mae_db": 1.0772881102065257, "3000–6000 Hz_10ms_log_rms_mae_db": 1.754950876849572, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24141081512767829, "3000–6000 Hz_10ms_envelope_correlation": 0.9609951196457063, "3000–6000 Hz_10ms_modulation_mae_db": 0.5771237367154534, "3000–6000 Hz_40ms_log_rms_mae_db": 1.297843735873232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1679844718618265, "3000–6000 Hz_40ms_envelope_correlation": 0.980074578510767, "3000–6000 Hz_40ms_modulation_mae_db": 0.5278940292345203, "mr_stft_db": 5.9933147464540175, "band_mae_db": 0.16193238596293358, "rms_gain_error_db": -0.07958682465509018, "output_peak": 0.5615576950770886}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.403601593737651, "全帯域_10ms_rms_envelope_nrmse": 0.13583195480948207, "全帯域_10ms_envelope_correlation": 0.9799732058048669, "全帯域_10ms_modulation_mae_db": 0.1744811629477043, "全帯域_40ms_log_rms_mae_db": 0.610152947414894, "全帯域_40ms_rms_envelope_nrmse": 0.06508425683734202, "全帯域_40ms_envelope_correlation": 0.9952175789232484, "全帯域_40ms_modulation_mae_db": 0.18633307324803136, "300–1000 Hz_10ms_log_rms_mae_db": 2.0777865405840164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1835112233571179, "300–1000 Hz_10ms_envelope_correlation": 0.9745896094740286, "300–1000 Hz_10ms_modulation_mae_db": 0.26022652223263654, "300–1000 Hz_40ms_log_rms_mae_db": 1.229832782443873, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10806995192329698, "300–1000 Hz_40ms_envelope_correlation": 0.9910344780790252, "300–1000 Hz_40ms_modulation_mae_db": 0.3668009549702635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.847785384218793, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23389612245796404, "1000–3000 Hz_10ms_envelope_correlation": 0.9563121435516173, "1000–3000 Hz_10ms_modulation_mae_db": 1.0883064638444764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0023755405920707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11267868549545393, "1000–3000 Hz_40ms_envelope_correlation": 0.9893866960593138, "1000–3000 Hz_40ms_modulation_mae_db": 1.3628481474208645, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7490463581479916, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23886876303029336, "3000–6000 Hz_10ms_envelope_correlation": 0.9615509544417029, "3000–6000 Hz_10ms_modulation_mae_db": 0.3698617820218457, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3058684215969245, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17313999613151546, "3000–6000 Hz_40ms_envelope_correlation": 0.9786648873183682, "3000–6000 Hz_40ms_modulation_mae_db": 0.3310705210813731, "mr_stft_db": 6.003646912828823, "band_mae_db": 0.11384917655175324, "rms_gain_error_db": -0.14957987667651554, "output_peak": 0.5049101916955601}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.221040167273979, "spectral_convergence": 0.4995923696782566, "spectral_convergence_status": "ok", "amplitude_floor": 1.4385235391470026e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.71972051195061, "spectral_convergence": 0.4209664548105648, "spectral_convergence_status": "ok", "amplitude_floor": 9.243439915833013e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.023341691254954, "spectral_convergence": 0.4604801611314742, "spectral_convergence_status": "ok", "amplitude_floor": 6.315989368278398e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00029756845932849967, "output_power": 0.0003146961575689071, "delta_db": 0.24304543810621845, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006886429842704082, "output_power": 0.0006485250926016723, "delta_db": -0.2606734297678365, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003825307900717409, "output_power": 0.00036177518655878665, "delta_db": -0.2422762130094749, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019304685978074288, "output_power": 0.00016796493615192317, "delta_db": -0.6044411232019912, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010892090215881134, "output_power": 0.00011045718590045795, "delta_db": 0.06082744767222574, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.08275993922433e-05, "output_power": 8.250026907153759e-05, "delta_db": 0.08895684694179867, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2500367497832576, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16582802689570153, "pitch": null, "output_rms_dbfs": -27.731633392391796, "output_peak": 0.49848040870293114, "reference_rms_dbfs": -27.565805365496097, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.988034123493182}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017313921740261034, "log_rms_mae_db": 1.4648517086877608, "rms_envelope_nrmse": 0.14296068468807885, "envelope_correlation": 0.9777810027215333, "rms_gain_error_db": -0.1591214536198682, "reference_mean_power": 0.0017487207983142375, "output_mean_power": 0.0016858088520977908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.577112847045013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002769005691966373, "output_power": 0.0002705323022541364, "reference_active": true, "delta_db": -0.10104720062807157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.377497445156709e-05, "output_power": 3.2778966739465366e-05, "reference_active": true, "delta_db": -0.12999769586632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4953137492438862e-05, "output_power": 1.4721094461804296e-05, "reference_active": true, "delta_db": -0.06792227144622463}], "active_band_count": 3, "active_band_mae_db": 0.09965572264687207}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015508669364693045, "log_rms_mae_db": 0.7251952782049389, "rms_envelope_nrmse": 0.06888527567292964, "envelope_correlation": 0.9946945468971256, "rms_gain_error_db": -0.15475605078792515, "reference_mean_power": 0.0017487025727625826, "output_mean_power": 0.0016874866429087267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.09499315514524e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025296787730251264, "output_power": 0.0002480977933244479, "reference_active": true, "delta_db": -0.08442475017573647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6687984786311574e-05, "output_power": 1.639866686035259e-05, "reference_active": true, "delta_db": -0.0759535190846756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.96993861330795e-07, "output_power": 7.11248516339871e-07, "reference_active": true, "delta_db": 0.08792420337245761}], "active_band_count": 3, "active_band_mae_db": 0.08276749087762324}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014635391485956695, "log_rms_mae_db": 2.032268927541266, "rms_envelope_nrmse": 0.19097155334672578, "envelope_correlation": 0.9724498716108391, "rms_gain_error_db": -0.24870166760698664, "reference_mean_power": 0.0008142561216173605, "output_mean_power": 0.0007689371887108495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2354521678013805e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001985819153870851, "output_power": 0.00017940363215772045, "reference_active": true, "delta_db": -0.4410846388477047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.365393218996536e-05, "output_power": 2.1489271748424328e-05, "reference_active": true, "delta_db": -0.4168164955888764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.646479803369054e-06, "output_power": 6.246136454936568e-06, "reference_active": true, "delta_db": -0.26980221677461014}], "active_band_count": 3, "active_band_mae_db": 0.37590111707039703}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012544182666407365, "log_rms_mae_db": 1.2206103509484512, "rms_envelope_nrmse": 0.09819157568731433, "envelope_correlation": 0.9928115902888524, "rms_gain_error_db": -0.23833318984694682, "reference_mean_power": 0.0008126713020357493, "output_mean_power": 0.000769274974909431}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0135181965386276e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018492598004057708, "output_power": 0.0001669701004482274, "reference_active": true, "delta_db": -0.44359220547664174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0725232186915858e-05, "output_power": 1.0256432366749291e-05, "reference_active": true, "delta_db": -0.19410382318267355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.105637817013119e-07, "output_power": 4.2826083001497553e-07, "reference_active": true, "delta_db": 0.1832771937461393}], "active_band_count": 3, "active_band_mae_db": 0.2736577408018182}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.111876173726916e-05, "log_rms_mae_db": 1.8995866244929105, "rms_envelope_nrmse": 0.23187962631704503, "envelope_correlation": 0.9575995593606291, "rms_gain_error_db": -0.4958168081472953, "reference_mean_power": 0.0002425197498003479, "output_mean_power": 0.00021635424996412604}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1105776313023488e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8957411226134796e-05, "output_power": 3.082646667641804e-05, "reference_active": true, "delta_db": -1.0166634086581414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.645635870343935e-06, "output_power": 5.05385897299167e-06, "reference_active": true, "delta_db": -0.4808974494103626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.039589927108413e-06, "output_power": 2.8156896435070752e-06, "reference_active": true, "delta_db": -1.5675249715166661}], "active_band_count": 3, "active_band_mae_db": 1.02169527652839}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.178958171420995e-05, "log_rms_mae_db": 1.007486175531512, "rms_envelope_nrmse": 0.11989214945291635, "envelope_correlation": 0.9900294160921788, "rms_gain_error_db": -0.5016838022502264, "reference_mean_power": 0.00024219152470715313, "output_mean_power": 0.00021576975133776806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0514876259433796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7077906526859964e-05, "output_power": 2.9178409910097477e-05, "reference_active": true, "delta_db": -1.0405358412031112}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.355351339185207e-06, "output_power": 2.7458115789742928e-06, "reference_active": true, "delta_db": -0.8706726981353539}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8063796091292985e-07, "output_power": 1.6363253014070412e-07, "reference_active": true, "delta_db": -2.3427677051285913}], "active_band_count": 3, "active_band_mae_db": 1.4179920814890188}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3400740982798506e-05, "log_rms_mae_db": 1.7774019177945468, "rms_envelope_nrmse": 0.2356729356468206, "envelope_correlation": 0.963335171357586, "rms_gain_error_db": 0.15437646916823333, "reference_mean_power": 0.0001023804664102342, "output_mean_power": 0.00010608518806688955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5067174911166034e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7037143173191167e-05, "output_power": 2.7496070895035315e-05, "reference_active": true, "delta_db": 0.07309838084836204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.303852533122454e-06, "output_power": 4.383410034408166e-06, "reference_active": true, "delta_db": 0.07954715193903422}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8801329619345497e-06, "output_power": 2.3220566222652375e-06, "reference_active": true, "delta_db": 0.9168424219091699}], "active_band_count": 3, "active_band_mae_db": 0.35649598489885537}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.152733550445532e-05, "log_rms_mae_db": 1.3396705554866732, "rms_envelope_nrmse": 0.1691943895192144, "envelope_correlation": 0.9801298147092224, "rms_gain_error_db": 0.15427315478522638, "reference_mean_power": 0.0001028923086971688, "output_mean_power": 0.00010661301554044467}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.218443109204273e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6964722520221083e-05, "output_power": 2.7273397904752365e-05, "reference_active": true, "delta_db": 0.04943293111307221}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9782668045768803e-06, "output_power": 2.310811880359693e-06, "reference_active": true, "delta_db": 0.6747972810117532}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5223765481589483e-07, "output_power": 1.7883820987107422e-07, "reference_active": true, "delta_db": 0.6993822890144219}], "active_band_count": 3, "active_band_mae_db": 0.4745375003797491}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7528486522916031, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a0c6100c51a3c78aa0a08e8abc78b66cfce2b0ec6c455c96a5b44c782a6766e8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9559982724225893, "standard_deviation": 0.004833815888464688, "minimum": 0.9458907671979543, "maximum": 0.9622362891476499, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8743226852981656, "standard_deviation": 0.02596904649567152, "minimum": 1.833451300872556, "maximum": 1.9117044620690773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7978047373049261, "standard_deviation": 0.224913212487692, "minimum": 0.4397606458436974, "maximum": 1.0883064638444764, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23479318891114243, "standard_deviation": 0.012756121116079266, "minimum": 0.2175027431266967, "maximum": 0.2609900432849634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893318519724958, "standard_deviation": 0.0019461113950508006, "minimum": 0.9854749187529301, "maximum": 0.9910077997358561, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0385762705793056, "standard_deviation": 0.04276233934245904, "minimum": 0.9782314497002091, "maximum": 1.1097530828586024, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.179409694982671, "standard_deviation": 0.3039775887273828, "minimum": 0.7759717477787063, "maximum": 1.5372287685267043, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11420602400248023, "standard_deviation": 0.009302183584509755, "minimum": 0.10404479177003954, "maximum": 0.1309614145208952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9650815430836148, "standard_deviation": 0.003286680234479331, "minimum": 0.9609951196457063, "maximum": 0.9688639035108698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7273030027451095, "standard_deviation": 0.04045683131920045, "minimum": 1.6583825083756647, "maximum": 1.7774019177945468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3757912941161454, "standard_deviation": 0.14358770146688335, "minimum": 0.14625908825261544, "maximum": 0.5771237367154534, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22934131777889133, "standard_deviation": 0.0095712960947987, "minimum": 0.21665852435779742, "maximum": 0.24141081512767829, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9808594492852702, "standard_deviation": 0.0015972442628671174, "minimum": 0.9786648873183682, "maximum": 0.9829369483370992, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2810798770890948, "standard_deviation": 0.03663608133156809, "minimum": 1.2379242468206195, "maximum": 1.3396705554866732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3229755272520094, "standard_deviation": 0.1524246884599772, "minimum": 0.09821544207287318, "maximum": 0.5278940292345203, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1655968799073229, "standard_deviation": 0.005731589965537771, "minimum": 0.15802132828656568, "maximum": 0.17313999613151546, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9727146288684593, "standard_deviation": 0.0033551522152202714, "minimum": 0.965981318700925, "maximum": 0.9763620223021281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0919400346740344, "standard_deviation": 0.03741442021526187, "minimum": 2.032268927541266, "maximum": 2.1456700655816796, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.42914352297999525, "standard_deviation": 0.1583539746998943, "minimum": 0.26022652223263654, "maximum": 0.6665612495797189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18975182167775723, "standard_deviation": 0.011094779687033535, "minimum": 0.1770835872965916, "maximum": 0.21152093365742441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9925554118157185, "standard_deviation": 0.0018668820183491927, "minimum": 0.990031582156576, "maximum": 0.995345707960059, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2649785564041078, "standard_deviation": 0.047477153588387444, "minimum": 1.1999435716242262, "maximum": 1.3288651334897885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3758211119253744, "standard_deviation": 0.13069438172695025, "minimum": 0.1996661516799291, "maximum": 0.6050855272069696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0976589183413616, "standard_deviation": 0.01277563839633202, "minimum": 0.0792824708124403, "maximum": 0.11495392969665531, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14843633719897031, "standard_deviation": 0.047739224253645365, "minimum": 0.09532041622522384, "maximum": 0.2500367497832576, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.0003212441958675, "standard_deviation": 0.008986533805165576, "minimum": 5.988034123493182, "maximum": 6.018315622365424, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5232948985614279, "standard_deviation": 0.02275657247237214, "minimum": 0.49848040870293114, "maximum": 0.5615576950770886, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08041014479778258, "standard_deviation": 0.07624711159281812, "minimum": -0.16582802689570153, "maximum": 0.022724458617896653, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9771307795923528, "standard_deviation": 0.0014042837690013126, "minimum": 0.9755347300517849, "maximum": 0.9799732058048669, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4450864643552828, "standard_deviation": 0.039017157287661884, "minimum": 1.3749665101470385, "maximum": 1.4807546105630307, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.1547632528785286, "standard_deviation": 0.06055767922920064, "minimum": 0.08586681680166193, "maximum": 0.25766649892342114, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1451945239270569, "standard_deviation": 0.0047792991243193585, "minimum": 0.13583195480948207, "maximum": 0.15114361551147165, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9946062429467768, "standard_deviation": 0.0010574891351987205, "minimum": 0.9934309946433899, "maximum": 0.9966859883123801, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6798473284332068, "standard_deviation": 0.04962015737645841, "minimum": 0.610152947414894, "maximum": 0.7367552975476758, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.18476506650513672, "standard_deviation": 0.10440342885510777, "minimum": 0.08276749087762324, "maximum": 0.3886271601152051, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06815020126639743, "standard_deviation": 0.006987591839846189, "minimum": 0.05360491364142644, "maximum": 0.07511333713547427, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0043", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.310030937194824, "predicted_mos": 2.310030937194824, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f55ac3d0dee4822eaa155bd61311fc8066bd78ae4ab8f6695e6ce8d700dd3524", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.870200399775058, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.402351975763042, "全帯域_10ms_rms_envelope_nrmse": 0.14593088400400986, "全帯域_10ms_envelope_correlation": 0.9773493211003871, "全帯域_10ms_modulation_mae_db": 0.08181582559841867, "全帯域_40ms_log_rms_mae_db": 0.6649150185252223, "全帯域_40ms_rms_envelope_nrmse": 0.07015367032267969, "全帯域_40ms_envelope_correlation": 0.9943702479314293, "全帯域_40ms_modulation_mae_db": 0.12395327493940345, "300–1000 Hz_10ms_log_rms_mae_db": 2.5260113479335518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19355885858304786, "300–1000 Hz_10ms_envelope_correlation": 0.9717599861827354, "300–1000 Hz_10ms_modulation_mae_db": 0.9783587328177165, "300–1000 Hz_40ms_log_rms_mae_db": 1.6330387664356092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0875217694136889, "300–1000 Hz_40ms_envelope_correlation": 0.9940139983157343, "300–1000 Hz_40ms_modulation_mae_db": 0.9823516568250933, "1000–3000 Hz_10ms_log_rms_mae_db": 2.223003872828331, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23086548329595588, "1000–3000 Hz_10ms_envelope_correlation": 0.9574682883668002, "1000–3000 Hz_10ms_modulation_mae_db": 0.8234661668776795, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2058308398128652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11613252709365111, "1000–3000 Hz_40ms_envelope_correlation": 0.9885718785332486, "1000–3000 Hz_40ms_modulation_mae_db": 1.0739158386406997, "3000–6000 Hz_10ms_log_rms_mae_db": 2.343783183206868, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24350379256643756, "3000–6000 Hz_10ms_envelope_correlation": 0.9604619196706998, "3000–6000 Hz_10ms_modulation_mae_db": 0.4956514203641788, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7013395290598234, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17290267849542076, "3000–6000 Hz_40ms_envelope_correlation": 0.9792518107578128, "3000–6000 Hz_40ms_modulation_mae_db": 0.3351155515178211, "mr_stft_db": 6.301028437485452, "band_mae_db": 0.1723089708896688, "rms_gain_error_db": 0.07980449802165648, "output_peak": 0.5288699632832311}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.347150584000134, "全帯域_10ms_rms_envelope_nrmse": 0.14806624192743695, "全帯域_10ms_envelope_correlation": 0.9762452244076575, "全帯域_10ms_modulation_mae_db": 0.2095036670466688, "全帯域_40ms_log_rms_mae_db": 0.6154422733789172, "全帯域_40ms_rms_envelope_nrmse": 0.06689037924419881, "全帯域_40ms_envelope_correlation": 0.9947328701281806, "全帯域_40ms_modulation_mae_db": 0.25891650280580686, "300–1000 Hz_10ms_log_rms_mae_db": 2.3795709989464484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20060444655868068, "300–1000 Hz_10ms_envelope_correlation": 0.9695092362941541, "300–1000 Hz_10ms_modulation_mae_db": 0.7577642703384303, "300–1000 Hz_40ms_log_rms_mae_db": 1.5188928860961635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10834589030650617, "300–1000 Hz_40ms_envelope_correlation": 0.9906333218809487, "300–1000 Hz_40ms_modulation_mae_db": 0.4231447013182203, "1000–3000 Hz_10ms_log_rms_mae_db": 2.243688360825962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25034667347235506, "1000–3000 Hz_10ms_envelope_correlation": 0.9499064993026746, "1000–3000 Hz_10ms_modulation_mae_db": 0.4368890412779021, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3218085448081922, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12294509791491058, "1000–3000 Hz_40ms_envelope_correlation": 0.9888772458935849, "1000–3000 Hz_40ms_modulation_mae_db": 1.067878331191234, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4155826609025395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2290588863873802, "3000–6000 Hz_10ms_envelope_correlation": 0.9648121456367975, "3000–6000 Hz_10ms_modulation_mae_db": 0.8261107359852824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7644696265764364, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1639835569499167, "3000–6000 Hz_40ms_envelope_correlation": 0.9814025374809513, "3000–6000 Hz_40ms_modulation_mae_db": 0.4122542619695973, "mr_stft_db": 6.3088940754926215, "band_mae_db": 0.26190279024709456, "rms_gain_error_db": -0.038794378858368006, "output_peak": 0.5387450848690405}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3847001650398443, "全帯域_10ms_rms_envelope_nrmse": 0.14884485693775854, "全帯域_10ms_envelope_correlation": 0.9758605855801716, "全帯域_10ms_modulation_mae_db": 0.2546633795324213, "全帯域_40ms_log_rms_mae_db": 0.6092143962482002, "全帯域_40ms_rms_envelope_nrmse": 0.06904307564510997, "全帯域_40ms_envelope_correlation": 0.9945143472308208, "全帯域_40ms_modulation_mae_db": 0.3614693833425047, "300–1000 Hz_10ms_log_rms_mae_db": 2.5083386440544913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.208835055679918, "300–1000 Hz_10ms_envelope_correlation": 0.9668899525532436, "300–1000 Hz_10ms_modulation_mae_db": 0.689502890196052, "300–1000 Hz_40ms_log_rms_mae_db": 1.535459128297616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10219871159457296, "300–1000 Hz_40ms_envelope_correlation": 0.9921626093792485, "300–1000 Hz_40ms_modulation_mae_db": 0.2807814714285305, "1000–3000 Hz_10ms_log_rms_mae_db": 2.253259300891214, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26387523640861743, "1000–3000 Hz_10ms_envelope_correlation": 0.9445510506107822, "1000–3000 Hz_10ms_modulation_mae_db": 0.9760432496998858, "1000–3000 Hz_40ms_log_rms_mae_db": 1.275637745472764, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14345391185596776, "1000–3000 Hz_40ms_envelope_correlation": 0.9825605863738628, "1000–3000 Hz_40ms_modulation_mae_db": 1.619435326782192, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3557653435887196, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24073185362342378, "3000–6000 Hz_10ms_envelope_correlation": 0.9617409443987474, "3000–6000 Hz_10ms_modulation_mae_db": 0.621464027161783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7105664313470605, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1687338515122943, "3000–6000 Hz_40ms_envelope_correlation": 0.9804883003048969, "3000–6000 Hz_40ms_modulation_mae_db": 0.44760383407110327, "mr_stft_db": 6.310863696142029, "band_mae_db": 0.175862678109055, "rms_gain_error_db": -0.11909182624040032, "output_peak": 0.5592998624112647}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.437619166940666, "全帯域_10ms_rms_envelope_nrmse": 0.136220554734838, "全帯域_10ms_envelope_correlation": 0.9798783419948447, "全帯域_10ms_modulation_mae_db": 0.4587744239184111, "全帯域_40ms_log_rms_mae_db": 0.6399100533916631, "全帯域_40ms_rms_envelope_nrmse": 0.0588148186822167, "全帯域_40ms_envelope_correlation": 0.9959409925782372, "全帯域_40ms_modulation_mae_db": 0.5986975873760069, "300–1000 Hz_10ms_log_rms_mae_db": 2.4803388423589916, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18400196243877637, "300–1000 Hz_10ms_envelope_correlation": 0.9745007888670096, "300–1000 Hz_10ms_modulation_mae_db": 0.5449118518320892, "300–1000 Hz_40ms_log_rms_mae_db": 1.5566064078947086, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09045561324269785, "300–1000 Hz_40ms_envelope_correlation": 0.9939928750180166, "300–1000 Hz_40ms_modulation_mae_db": 0.5069044642639018, "1000–3000 Hz_10ms_log_rms_mae_db": 2.290051846611681, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2620227490880111, "1000–3000 Hz_10ms_envelope_correlation": 0.9469609579686177, "1000–3000 Hz_10ms_modulation_mae_db": 0.4845642424135163, "1000–3000 Hz_40ms_log_rms_mae_db": 1.295181890169337, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13841970759599445, "1000–3000 Hz_40ms_envelope_correlation": 0.9843688855829262, "1000–3000 Hz_40ms_modulation_mae_db": 0.46530673523946825, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3202720098513483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2374586256886131, "3000–6000 Hz_10ms_envelope_correlation": 0.9617642216444108, "3000–6000 Hz_10ms_modulation_mae_db": 0.6739655684761162, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7541253124652707, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1775027858465547, "3000–6000 Hz_40ms_envelope_correlation": 0.9778889823620972, "3000–6000 Hz_40ms_modulation_mae_db": 0.40239244543021546, "mr_stft_db": 6.305051032245154, "band_mae_db": 0.19367167072020997, "rms_gain_error_db": -0.007220174530838795, "output_peak": 0.5402623236248422}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3761680336069155, "全帯域_10ms_rms_envelope_nrmse": 0.14267627314178352, "全帯域_10ms_envelope_correlation": 0.9780511573695386, "全帯域_10ms_modulation_mae_db": 0.16116638654238793, "全帯域_40ms_log_rms_mae_db": 0.6880992433003618, "全帯域_40ms_rms_envelope_nrmse": 0.06830547759073505, "全帯域_40ms_envelope_correlation": 0.9945246214662951, "全帯域_40ms_modulation_mae_db": 0.180656514415774, "300–1000 Hz_10ms_log_rms_mae_db": 2.5494288382664183, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.194236437150253, "300–1000 Hz_10ms_envelope_correlation": 0.9714083130785427, "300–1000 Hz_10ms_modulation_mae_db": 0.5236266423499348, "300–1000 Hz_40ms_log_rms_mae_db": 1.5582511769411977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09040331356241331, "300–1000 Hz_40ms_envelope_correlation": 0.9937884105746352, "300–1000 Hz_40ms_modulation_mae_db": 0.5766803116983134, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2478201104135294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22644844102607892, "1000–3000 Hz_10ms_envelope_correlation": 0.9596052039579912, "1000–3000 Hz_10ms_modulation_mae_db": 0.5074838417363982, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2899455510804303, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1026826726455097, "1000–3000 Hz_40ms_envelope_correlation": 0.991125486610009, "1000–3000 Hz_40ms_modulation_mae_db": 0.927152243974644, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4200503689870385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22570904650235668, "3000–6000 Hz_10ms_envelope_correlation": 0.9675793916223538, "3000–6000 Hz_10ms_modulation_mae_db": 0.4641381136873894, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7837833500243792, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1677467731560342, "3000–6000 Hz_40ms_envelope_correlation": 0.9816109769516029, "3000–6000 Hz_40ms_modulation_mae_db": 0.1836106035678382, "mr_stft_db": 6.326549496500957, "band_mae_db": 0.17026162623451055, "rms_gain_error_db": 0.0066100624020879436, "output_peak": 0.5677358502045198}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.354586767701435, "全帯域_10ms_rms_envelope_nrmse": 0.13878107366923811, "全帯域_10ms_envelope_correlation": 0.979080277689531, "全帯域_10ms_modulation_mae_db": 0.32121926887718094, "全帯域_40ms_log_rms_mae_db": 0.6283113069199793, "全帯域_40ms_rms_envelope_nrmse": 0.059489170642879244, "全帯域_40ms_envelope_correlation": 0.9958367214481922, "全帯域_40ms_modulation_mae_db": 0.35508198430470417, "300–1000 Hz_10ms_log_rms_mae_db": 2.5219560431136983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18768235721129414, "300–1000 Hz_10ms_envelope_correlation": 0.9733312537512631, "300–1000 Hz_10ms_modulation_mae_db": 0.16151326328894247, "300–1000 Hz_40ms_log_rms_mae_db": 1.6140910839425162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09605889721962031, "300–1000 Hz_40ms_envelope_correlation": 0.9928213418914394, "300–1000 Hz_40ms_modulation_mae_db": 0.40598295373777443, "1000–3000 Hz_10ms_log_rms_mae_db": 2.196517527743928, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2462315337994523, "1000–3000 Hz_10ms_envelope_correlation": 0.9514672523788551, "1000–3000 Hz_10ms_modulation_mae_db": 0.7452933870932953, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2548864941671414, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11615833539510635, "1000–3000 Hz_40ms_envelope_correlation": 0.988652500270509, "1000–3000 Hz_40ms_modulation_mae_db": 1.1456977646480906, "3000–6000 Hz_10ms_log_rms_mae_db": 2.379991375919343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23836793458921401, "3000–6000 Hz_10ms_envelope_correlation": 0.9618797742520121, "3000–6000 Hz_10ms_modulation_mae_db": 0.5223274346361849, "3000–6000 Hz_40ms_log_rms_mae_db": 1.752720399294349, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1684005943395261, "3000–6000 Hz_40ms_envelope_correlation": 0.98027774234835, "3000–6000 Hz_40ms_modulation_mae_db": 0.3059644051787614, "mr_stft_db": 6.318103827691469, "band_mae_db": 0.2560661820959496, "rms_gain_error_db": -0.04453000222991194, "output_peak": 0.4956131566971522}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4259814955226788, "全帯域_10ms_rms_envelope_nrmse": 0.1493300252023486, "全帯域_10ms_envelope_correlation": 0.9757703146207086, "全帯域_10ms_modulation_mae_db": 0.18552903186510364, "全帯域_40ms_log_rms_mae_db": 0.69830303159167, "全帯域_40ms_rms_envelope_nrmse": 0.07030792955005581, "全帯域_40ms_envelope_correlation": 0.9941508855109008, "全帯域_40ms_modulation_mae_db": 0.1824098059785448, "300–1000 Hz_10ms_log_rms_mae_db": 2.64430135671888, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20479015009884177, "300–1000 Hz_10ms_envelope_correlation": 0.9681587445890298, "300–1000 Hz_10ms_modulation_mae_db": 0.5254792767811771, "300–1000 Hz_40ms_log_rms_mae_db": 1.682107464134627, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11435725108155705, "300–1000 Hz_40ms_envelope_correlation": 0.9898162329165989, "300–1000 Hz_40ms_modulation_mae_db": 0.4645436269220709, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1615226443619764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2563544141389233, "1000–3000 Hz_10ms_envelope_correlation": 0.9491070165338585, "1000–3000 Hz_10ms_modulation_mae_db": 0.6422648989384064, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2132555050388307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13213262643632814, "1000–3000 Hz_40ms_envelope_correlation": 0.9857952073577211, "1000–3000 Hz_40ms_modulation_mae_db": 1.0241176137220922, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4595337270851356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24528735091865742, "3000–6000 Hz_10ms_envelope_correlation": 0.9600437059700386, "3000–6000 Hz_10ms_modulation_mae_db": 0.4723013113038437, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8446815148957902, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17301991897630523, "3000–6000 Hz_40ms_envelope_correlation": 0.9794477269360665, "3000–6000 Hz_40ms_modulation_mae_db": 0.1136861347724672, "mr_stft_db": 6.331005610500576, "band_mae_db": 0.16909056532302327, "rms_gain_error_db": -0.026289148903126443, "output_peak": 0.5464134978747501}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3508296246271116, "全帯域_10ms_rms_envelope_nrmse": 0.14295468028443317, "全帯域_10ms_envelope_correlation": 0.9778020916935132, "全帯域_10ms_modulation_mae_db": 0.18277773401207523, "全帯域_40ms_log_rms_mae_db": 0.6138885573989077, "全帯域_40ms_rms_envelope_nrmse": 0.06100559722923387, "全帯域_40ms_envelope_correlation": 0.9956016827238856, "全帯域_40ms_modulation_mae_db": 0.07240149696472178, "300–1000 Hz_10ms_log_rms_mae_db": 2.535224074357837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18624069707490487, "300–1000 Hz_10ms_envelope_correlation": 0.9738733740434573, "300–1000 Hz_10ms_modulation_mae_db": 0.35652173937380044, "300–1000 Hz_40ms_log_rms_mae_db": 1.5639442674311341, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08870368029525039, "300–1000 Hz_40ms_envelope_correlation": 0.9938614551311379, "300–1000 Hz_40ms_modulation_mae_db": 0.35649826155522185, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1714496940502324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24196150930535593, "1000–3000 Hz_10ms_envelope_correlation": 0.9531751534349361, "1000–3000 Hz_10ms_modulation_mae_db": 0.5636279800534161, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1967132667717666, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11542766803630758, "1000–3000 Hz_40ms_envelope_correlation": 0.988743491702611, "1000–3000 Hz_40ms_modulation_mae_db": 0.7832415575229273, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3894134801489684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22797603831786803, "3000–6000 Hz_10ms_envelope_correlation": 0.9651263826911055, "3000–6000 Hz_10ms_modulation_mae_db": 0.6986408208022848, "3000–6000 Hz_40ms_log_rms_mae_db": 1.758445293997393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17223363723199583, "3000–6000 Hz_40ms_envelope_correlation": 0.9792833735137714, "3000–6000 Hz_40ms_modulation_mae_db": 0.31162076160218616, "mr_stft_db": 6.309095250045957, "band_mae_db": 0.11439346259069404, "rms_gain_error_db": -0.024044121908092065, "output_peak": 0.5491741383254988}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.5846690677542155, "spectral_convergence": 0.5235729572072978, "spectral_convergence_status": "ok", "amplitude_floor": 1.4385235391470026e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.126540175144726, "spectral_convergence": 0.44673749177176253, "spectral_convergence_status": "ok", "amplitude_floor": 9.243439915833013e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.191876069557411, "spectral_convergence": 0.48453327993889717, "spectral_convergence_status": "ok", "amplitude_floor": 6.315989368278398e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00029756845932849967, "output_power": 0.0003382282359193303, "delta_db": 0.5562296407080093, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006886429842704082, "output_power": 0.00068277170778046, "delta_db": -0.03718610533591967, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003825307900717409, "output_power": 0.00038462194762302663, "delta_db": 0.023676652381277384, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019304685978074288, "output_power": 0.00018434527994088561, "delta_db": -0.20030719301628885, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010892090215881134, "output_power": 0.00010969500389722527, "delta_db": 0.030756180971150664, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.08275993922433e-05, "output_power": 8.435862368419888e-05, "delta_db": 0.18569805292536687, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1723089708896688, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.07980449802165648, "pitch": null, "output_rms_dbfs": -27.48600086747444, "output_peak": 0.5288699632832311, "reference_rms_dbfs": -27.565805365496097, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.301028437485452}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017313921740261034, "log_rms_mae_db": 1.402351975763042, "rms_envelope_nrmse": 0.14593088400400986, "envelope_correlation": 0.9773493211003871, "rms_gain_error_db": 0.08416586309451816, "reference_mean_power": 0.0017487207983142375, "output_mean_power": 0.0017829413679754595}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.577112847045013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002769005691966373, "output_power": 0.00027865648378877766, "reference_active": true, "delta_db": 0.027453042069692346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.377497445156709e-05, "output_power": 3.495999282689364e-05, "reference_active": true, "delta_db": 0.14976305673201273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4953137492438862e-05, "output_power": 1.5189919224496056e-05, "reference_active": true, "delta_db": 0.06823137799355093}], "active_band_count": 3, "active_band_mae_db": 0.08181582559841867}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015508669364693045, "log_rms_mae_db": 0.6649150185252223, "rms_envelope_nrmse": 0.07015367032267969, "envelope_correlation": 0.9943702479314293, "rms_gain_error_db": 0.07961692833329956, "reference_mean_power": 0.0017487025727625826, "output_mean_power": 0.0017810562749766149}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.09499315514524e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025296787730251264, "output_power": 0.00025656443879713565, "reference_active": true, "delta_db": 0.06131084145534625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6687984786311574e-05, "output_power": 1.6524563301549018e-05, "reference_active": true, "delta_db": -0.04273904047055124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.96993861330795e-07, "output_power": 7.41327221044758e-07, "reference_active": true, "delta_db": 0.26780994289231286}], "active_band_count": 3, "active_band_mae_db": 0.12395327493940345}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014635391485956695, "log_rms_mae_db": 2.5260113479335518, "rms_envelope_nrmse": 0.19355885858304786, "envelope_correlation": 0.9717599861827354, "rms_gain_error_db": -0.0417176302836791, "reference_mean_power": 0.0008142561216173605, "output_mean_power": 0.0008064719546982496}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2354521678013805e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001985819153870851, "output_power": 0.00018705496756809498, "reference_active": true, "delta_db": -0.25970449162947257}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.365393218996536e-05, "output_power": 2.0383522169995683e-05, "reference_active": true, "delta_db": -0.6462411740629406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.646479803369054e-06, "output_power": 1.0604855192523758e-05, "reference_active": true, "delta_db": 2.0291305327607363}], "active_band_count": 3, "active_band_mae_db": 0.9783587328177165}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012544182666407365, "log_rms_mae_db": 1.6330387664356092, "rms_envelope_nrmse": 0.0875217694136889, "envelope_correlation": 0.9940139983157343, "rms_gain_error_db": -0.05265885177077253, "reference_mean_power": 0.0008126713020357493, "output_mean_power": 0.000802877040082545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0135181965386276e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018492598004057708, "output_power": 0.00017633207752730968, "reference_active": true, "delta_db": -0.20666604625507468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0725232186915858e-05, "output_power": 9.32617640197569e-06, "reference_active": true, "delta_db": -0.6070307709970951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.105637817013119e-07, "output_power": 6.709906219392484e-07, "reference_active": true, "delta_db": 2.13335815322311}], "active_band_count": 3, "active_band_mae_db": 0.9823516568250933}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.111876173726916e-05, "log_rms_mae_db": 2.223003872828331, "rms_envelope_nrmse": 0.23086548329595588, "envelope_correlation": 0.9574682883668002, "rms_gain_error_db": -0.10322958205565877, "reference_mean_power": 0.0002425197498003479, "output_mean_power": 0.00023682314995296817}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1105776313023488e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8957411226134796e-05, "output_power": 3.6783156627133216e-05, "reference_active": true, "delta_db": -0.249410932630082}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.645635870343935e-06, "output_power": 6.35824768326477e-06, "reference_active": true, "delta_db": 0.5162457790841273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.039589927108413e-06, "output_power": 2.728117690074137e-06, "reference_active": true, "delta_db": -1.7047417889188292}], "active_band_count": 3, "active_band_mae_db": 0.8234661668776795}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.178958171420995e-05, "log_rms_mae_db": 1.2058308398128652, "rms_envelope_nrmse": 0.11613252709365111, "envelope_correlation": 0.9885718785332486, "rms_gain_error_db": -0.09954145219410257, "reference_mean_power": 0.00024219152470715313, "output_mean_power": 0.00023670356346001668}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0514876259433796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7077906526859964e-05, "output_power": 3.456795309453283e-05, "reference_active": true, "delta_db": -0.30441541252696447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.355351339185207e-06, "output_power": 3.1870230488368193e-06, "reference_active": true, "delta_db": -0.22352797564296203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8063796091292985e-07, "output_power": 1.509266746471257e-07, "reference_active": true, "delta_db": -2.693804127752173}], "active_band_count": 3, "active_band_mae_db": 1.0739158386406997}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3400740982798506e-05, "log_rms_mae_db": 2.343783183206868, "rms_envelope_nrmse": 0.24350379256643756, "envelope_correlation": 0.9604619196706998, "rms_gain_error_db": 0.08443634650650624, "reference_mean_power": 0.0001023804664102342, "output_mean_power": 0.00010439044246352225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5067174911166034e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7037143173191167e-05, "output_power": 2.7179687148639077e-05, "reference_active": true, "delta_db": 0.022836526679528547}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.303852533122454e-06, "output_power": 3.3854111669074208e-06, "reference_active": true, "delta_db": -1.0424595987187315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8801329619345497e-06, "output_power": 2.071831238610783e-06, "reference_active": true, "delta_db": 0.42165813569427624}], "active_band_count": 3, "active_band_mae_db": 0.4956514203641788}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.152733550445532e-05, "log_rms_mae_db": 1.7013395290598234, "rms_envelope_nrmse": 0.17290267849542076, "envelope_correlation": 0.9792518107578128, "rms_gain_error_db": 0.08439805139571727, "reference_mean_power": 0.0001028923086971688, "output_mean_power": 0.00010491140836260702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.218443109204273e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6964722520221083e-05, "output_power": 2.6634969890781647e-05, "reference_active": true, "delta_db": -0.05343745373505241}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9782668045768803e-06, "output_power": 1.7285645675785864e-06, "reference_active": true, "delta_db": -0.5859925706089733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5223765481589483e-07, "output_power": 1.3993631140258264e-07, "reference_active": true, "delta_db": -0.36591663020943754}], "active_band_count": 3, "active_band_mae_db": 0.3351155515178211}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7299008316979423, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "45fda7b3dee389c22dd746baab00005d7a53578ae619989a052941037564caee", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9515301778193145, "standard_deviation": 0.005088799150783844, "minimum": 0.9445510506107822, "maximum": 0.9596052039579912, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2234141697158565, "standard_deviation": 0.04406564897674058, "minimum": 2.1615226443619764, "maximum": 2.290051846611681, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6474541010113124, "standard_deviation": 0.18747943587668964, "minimum": 0.4368890412779021, "maximum": 0.9760432496998858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24726325506684377, "standard_deviation": 0.0137179423159827, "minimum": 0.22644844102607892, "maximum": 0.26387523640861743, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9873369102905591, "standard_deviation": 0.002827407390884982, "minimum": 0.9825605863738628, "maximum": 0.991125486610009, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.256657479665166, "standard_deviation": 0.04669148201886123, "minimum": 1.1967132667717666, "maximum": 1.3218085448081922, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0133431764651684, "standard_deviation": 0.3278379877504766, "minimum": 0.46530673523946825, "maximum": 1.619435326782192, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12341906837172195, "standard_deviation": 0.01363942492977032, "minimum": 0.1026826726455097, "maximum": 0.14345391185596776, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9629260607357707, "standard_deviation": 0.0026262756818009425, "minimum": 0.9600437059700386, "maximum": 0.9675793916223538, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.385549018711245, "standard_deviation": 0.0455199509913451, "minimum": 2.3202720098513483, "maximum": 2.4595337270851356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5968249290521329, "standard_deviation": 0.13005169212031886, "minimum": 0.4641381136873894, "maximum": 0.8261107359852824, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23601169107424386, "standard_deviation": 0.007475221750162683, "minimum": 0.22570904650235668, "maximum": 0.24528735091865742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9799564313319437, "standard_deviation": 0.0012362091802176664, "minimum": 0.9778889823620972, "maximum": 0.9816109769516029, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7587664322075627, "standard_deviation": 0.044270372163848715, "minimum": 1.7013395290598234, "maximum": 1.8446815148957902, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.31403099976374876, "standard_deviation": 0.11523903020285649, "minimum": 0.1136861347724672, "maximum": 0.44760383407110327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.170565474563506, "standard_deviation": 0.00416968893330712, "minimum": 0.1639835569499167, "maximum": 0.1775027858465547, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9711789561699294, "standard_deviation": 0.002767728309627782, "minimum": 0.9668899525532436, "maximum": 0.9745007888670096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.5181462682187896, "standard_deviation": 0.07366955861454853, "minimum": 2.3795709989464484, "maximum": 2.64430135671888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5672098333722678, "standard_deviation": 0.24883365542751626, "minimum": 0.16151326328894247, "maximum": 0.9783587328177165, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19499374559946459, "standard_deviation": 0.009043982558562789, "minimum": 0.18400196243877637, "maximum": 0.208835055679918, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9926362806384699, "standard_deviation": 0.001638380528940558, "minimum": 0.9898162329165989, "maximum": 0.9940139983157343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5827988976466967, "standard_deviation": 0.055206887488424485, "minimum": 1.5188928860961635, "maximum": 1.682107464134627, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4996109309686408, "standard_deviation": 0.2148893603010094, "minimum": 0.2807814714285305, "maximum": 0.9823516568250933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09725564083953836, "standard_deviation": 0.010017155972253961, "minimum": 0.0875217694136889, "maximum": 0.11435725108155705, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18919474327627572, "standard_deviation": 0.048695904871052485, "minimum": 0.11439346259069404, "maximum": 0.26190279024709456, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.313823928263027, "standard_deviation": 0.010498863473293228, "minimum": 6.301028437485452, "maximum": 6.331005610500576, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5407642346612874, "standard_deviation": 0.02190270892934173, "minimum": 0.4956131566971522, "maximum": 0.5677358502045198, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.021694386530874143, "standard_deviation": 0.05559989420513349, "minimum": -0.11909182624040032, "maximum": 0.07980449802165648, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.977504664307044, "standard_deviation": 0.0015035816800882298, "minimum": 0.9757703146207086, "maximum": 0.9798783419948447, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3849234766502283, "standard_deviation": 0.03454304766928013, "minimum": 1.347150584000134, "maximum": 1.437619166940666, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.23193121467408345, "standard_deviation": 0.11499159135956899, "minimum": 0.08181582559841867, "maximum": 0.4587744239184111, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14410057373773083, "standard_deviation": 0.004820572432266939, "minimum": 0.136220554734838, "maximum": 0.1493300252023486, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9949590461272426, "standard_deviation": 0.0007156374751864595, "minimum": 0.9941508855109008, "maximum": 0.9959409925782372, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6447604850943651, "standard_deviation": 0.034890681331572776, "minimum": 0.6092143962482002, "maximum": 0.69830303159167, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.26669831876593336, "standard_deviation": 0.1689092935001918, "minimum": 0.07240149696472178, "maximum": 0.5986975873760069, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06550126486338864, "standard_deviation": 0.004901187032475984, "minimum": 0.0588148186822167, "maximum": 0.07030792955005581, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0043", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.22686505317688, "predicted_mos": 2.22686505317688, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "077ab843e0b461af2b33298e7f43176fa3f256150dffcac62a1dcbe04dadb694", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9722974998876452, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4425495709714473, "全帯域_10ms_rms_envelope_nrmse": 0.1419540917689548, "全帯域_10ms_envelope_correlation": 0.9783912591818081, "全帯域_10ms_modulation_mae_db": 0.3469154596466595, "全帯域_40ms_log_rms_mae_db": 0.6951620646531429, "全帯域_40ms_rms_envelope_nrmse": 0.0701148845290689, "全帯域_40ms_envelope_correlation": 0.9943045945408376, "全帯域_40ms_modulation_mae_db": 0.24071130418696163, "300–1000 Hz_10ms_log_rms_mae_db": 1.9536927183606219, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18462632835962847, "300–1000 Hz_10ms_envelope_correlation": 0.9745438916712853, "300–1000 Hz_10ms_modulation_mae_db": 0.4172002057087902, "300–1000 Hz_40ms_log_rms_mae_db": 1.2265425722516266, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09489365566082748, "300–1000 Hz_40ms_envelope_correlation": 0.992860521036545, "300–1000 Hz_40ms_modulation_mae_db": 0.6687457655672405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7163766193972978, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2125480562878737, "1000–3000 Hz_10ms_envelope_correlation": 0.963922800931863, "1000–3000 Hz_10ms_modulation_mae_db": 0.6791731586365503, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9689835979259003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1024129798753429, "1000–3000 Hz_40ms_envelope_correlation": 0.9915405649329473, "1000–3000 Hz_40ms_modulation_mae_db": 0.36562300685464516, "3000–6000 Hz_10ms_log_rms_mae_db": 1.591085533744132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2215119936952819, "3000–6000 Hz_10ms_envelope_correlation": 0.970074272645874, "3000–6000 Hz_10ms_modulation_mae_db": 0.7442730297127391, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2334491645814754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16466812529906064, "3000–6000 Hz_40ms_envelope_correlation": 0.9832436873003827, "3000–6000 Hz_40ms_modulation_mae_db": 0.6748620793012279, "mr_stft_db": 5.924336322096024, "band_mae_db": 0.20725368051879137, "rms_gain_error_db": 0.04176259835803594, "output_peak": 0.5063149919821897}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4681776939079383, "全帯域_10ms_rms_envelope_nrmse": 0.1452400919233493, "全帯域_10ms_envelope_correlation": 0.977109847190372, "全帯域_10ms_modulation_mae_db": 0.4260320882280631, "全帯域_40ms_log_rms_mae_db": 0.6783806157957621, "全帯域_40ms_rms_envelope_nrmse": 0.06542842336165061, "全帯域_40ms_envelope_correlation": 0.9949510603510935, "全帯域_40ms_modulation_mae_db": 0.48175065633072295, "300–1000 Hz_10ms_log_rms_mae_db": 1.9275014963179384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19342537167835927, "300–1000 Hz_10ms_envelope_correlation": 0.9716768832790579, "300–1000 Hz_10ms_modulation_mae_db": 0.5818334724803144, "300–1000 Hz_40ms_log_rms_mae_db": 1.1927157820944552, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09714992208243917, "300–1000 Hz_40ms_envelope_correlation": 0.992511412127076, "300–1000 Hz_40ms_modulation_mae_db": 0.6543075689257812, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7229840135928975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22235823499638524, "1000–3000 Hz_10ms_envelope_correlation": 0.9604761335970988, "1000–3000 Hz_10ms_modulation_mae_db": 0.6376726561112798, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9606537458202258, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10339848034976984, "1000–3000 Hz_40ms_envelope_correlation": 0.9913002333948724, "1000–3000 Hz_40ms_modulation_mae_db": 0.7934447723107061, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5492262181076428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21854108360579147, "3000–6000 Hz_10ms_envelope_correlation": 0.9675836075912351, "3000–6000 Hz_10ms_modulation_mae_db": 0.43411024084504074, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1900338504220855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16419828516735152, "3000–6000 Hz_40ms_envelope_correlation": 0.9807298043015862, "3000–6000 Hz_40ms_modulation_mae_db": 0.5183138278530136, "mr_stft_db": 5.918022700482472, "band_mae_db": 0.20516729176273263, "rms_gain_error_db": -0.03269865308256656, "output_peak": 0.5307667913628481}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.483435664667223, "全帯域_10ms_rms_envelope_nrmse": 0.1445339158208169, "全帯域_10ms_envelope_correlation": 0.9773620693244238, "全帯域_10ms_modulation_mae_db": 0.1549920678736374, "全帯域_40ms_log_rms_mae_db": 0.6602297129809235, "全帯域_40ms_rms_envelope_nrmse": 0.06312792204004714, "全帯域_40ms_envelope_correlation": 0.9957582914148799, "全帯域_40ms_modulation_mae_db": 0.20510622913144985, "300–1000 Hz_10ms_log_rms_mae_db": 1.943161653245667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1783611471998568, "300–1000 Hz_10ms_envelope_correlation": 0.976374167604421, "300–1000 Hz_10ms_modulation_mae_db": 0.4232286054945648, "300–1000 Hz_40ms_log_rms_mae_db": 1.2975938763387196, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10292554868785438, "300–1000 Hz_40ms_envelope_correlation": 0.9925451673418809, "300–1000 Hz_40ms_modulation_mae_db": 0.4553641105699033, "1000–3000 Hz_10ms_log_rms_mae_db": 1.665515106871379, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23020057470105917, "1000–3000 Hz_10ms_envelope_correlation": 0.9577256145142775, "1000–3000 Hz_10ms_modulation_mae_db": 0.3972780415357917, "1000–3000 Hz_40ms_log_rms_mae_db": 0.951139164707299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11300166463649848, "1000–3000 Hz_40ms_envelope_correlation": 0.9896498295313454, "1000–3000 Hz_40ms_modulation_mae_db": 0.6944957271987073, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5487749665736978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22275911637085427, "3000–6000 Hz_10ms_envelope_correlation": 0.9660490703271433, "3000–6000 Hz_10ms_modulation_mae_db": 0.5495741093799812, "3000–6000 Hz_40ms_log_rms_mae_db": 1.202251240855398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16475424005349207, "3000–6000 Hz_40ms_envelope_correlation": 0.9808632251972956, "3000–6000 Hz_40ms_modulation_mae_db": 0.5383301795875789, "mr_stft_db": 5.91207517865453, "band_mae_db": 0.19458507563789304, "rms_gain_error_db": -0.19698223972374462, "output_peak": 0.5310501147914294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4310691057522964, "全帯域_10ms_rms_envelope_nrmse": 0.14668071042322037, "全帯域_10ms_envelope_correlation": 0.9767036751918864, "全帯域_10ms_modulation_mae_db": 0.10911882864839512, "全帯域_40ms_log_rms_mae_db": 0.6591950321111242, "全帯域_40ms_rms_envelope_nrmse": 0.05923002510302686, "全帯域_40ms_envelope_correlation": 0.995871355254675, "全帯域_40ms_modulation_mae_db": 0.1974961963791698, "300–1000 Hz_10ms_log_rms_mae_db": 2.0002972197008346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18569049800226745, "300–1000 Hz_10ms_envelope_correlation": 0.9739094921428239, "300–1000 Hz_10ms_modulation_mae_db": 0.45672779493504406, "300–1000 Hz_40ms_log_rms_mae_db": 1.2571229789146743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08643609077301614, "300–1000 Hz_40ms_envelope_correlation": 0.9941001166572915, "300–1000 Hz_40ms_modulation_mae_db": 0.4646215560209095, "1000–3000 Hz_10ms_log_rms_mae_db": 1.716190966215779, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23647156215721132, "1000–3000 Hz_10ms_envelope_correlation": 0.9557965524530627, "1000–3000 Hz_10ms_modulation_mae_db": 0.5227560826786244, "1000–3000 Hz_40ms_log_rms_mae_db": 1.010325395939742, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11296540315661852, "1000–3000 Hz_40ms_envelope_correlation": 0.9892858520878319, "1000–3000 Hz_40ms_modulation_mae_db": 0.8436770741921582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6048830440355646, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22575961101529912, "3000–6000 Hz_10ms_envelope_correlation": 0.966230425630434, "3000–6000 Hz_10ms_modulation_mae_db": 0.7459474721079333, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2224129697925465, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16137388496741153, "3000–6000 Hz_40ms_envelope_correlation": 0.9818304393389178, "3000–6000 Hz_40ms_modulation_mae_db": 0.3085751925868286, "mr_stft_db": 5.942091319715452, "band_mae_db": 0.19224430436836412, "rms_gain_error_db": -0.02597501434615803, "output_peak": 0.5227064843888665}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4438977963802033, "全帯域_10ms_rms_envelope_nrmse": 0.14463043449720722, "全帯域_10ms_envelope_correlation": 0.9772958454727058, "全帯域_10ms_modulation_mae_db": 0.35716158019078154, "全帯域_40ms_log_rms_mae_db": 0.7390112882780486, "全帯域_40ms_rms_envelope_nrmse": 0.07447548301969595, "全帯域_40ms_envelope_correlation": 0.9938538720300448, "全帯域_40ms_modulation_mae_db": 0.4926840666255113, "300–1000 Hz_10ms_log_rms_mae_db": 1.9959087023645243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19967265567828504, "300–1000 Hz_10ms_envelope_correlation": 0.9699810852995011, "300–1000 Hz_10ms_modulation_mae_db": 1.1016420979984185, "300–1000 Hz_40ms_log_rms_mae_db": 1.3325403736972061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11157292027094924, "300–1000 Hz_40ms_envelope_correlation": 0.9910115125848534, "300–1000 Hz_40ms_modulation_mae_db": 0.8018605737700536, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7907759706372177, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23630683702922306, "1000–3000 Hz_10ms_envelope_correlation": 0.9552279387846833, "1000–3000 Hz_10ms_modulation_mae_db": 0.7843710031599525, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0343236841430494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10838693453803962, "1000–3000 Hz_40ms_envelope_correlation": 0.9906386334923852, "1000–3000 Hz_40ms_modulation_mae_db": 0.72940249279393, "3000–6000 Hz_10ms_log_rms_mae_db": 1.532148808119974, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.212983604559599, "3000–6000 Hz_10ms_envelope_correlation": 0.9696886529029364, "3000–6000 Hz_10ms_modulation_mae_db": 0.4334799821613597, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1573483401309326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15757844925335177, "3000–6000 Hz_40ms_envelope_correlation": 0.9824761617715334, "3000–6000 Hz_40ms_modulation_mae_db": 0.18271335771613276, "mr_stft_db": 5.928024060426338, "band_mae_db": 0.1817049609643365, "rms_gain_error_db": -0.18934171810478687, "output_peak": 0.5162134085444333}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4288030440766781, "全帯域_10ms_rms_envelope_nrmse": 0.1444194389139937, "全帯域_10ms_envelope_correlation": 0.977274821229691, "全帯域_10ms_modulation_mae_db": 0.13975092960440608, "全帯域_40ms_log_rms_mae_db": 0.6692886293407988, "全帯域_40ms_rms_envelope_nrmse": 0.07381708078229599, "全帯域_40ms_envelope_correlation": 0.9935940750754715, "全帯域_40ms_modulation_mae_db": 0.13996433573466227, "300–1000 Hz_10ms_log_rms_mae_db": 2.0442742848919786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19505148822405993, "300–1000 Hz_10ms_envelope_correlation": 0.9712275513657054, "300–1000 Hz_10ms_modulation_mae_db": 0.9082353480883393, "300–1000 Hz_40ms_log_rms_mae_db": 1.3192459117188715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10536926740548742, "300–1000 Hz_40ms_envelope_correlation": 0.9916046575451707, "300–1000 Hz_40ms_modulation_mae_db": 0.7536020650939316, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6648391428240399, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20002583170746926, "1000–3000 Hz_10ms_envelope_correlation": 0.9682291071608548, "1000–3000 Hz_10ms_modulation_mae_db": 0.28913864058635064, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9562131550486553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09383422845797068, "1000–3000 Hz_40ms_envelope_correlation": 0.9925547932476438, "1000–3000 Hz_40ms_modulation_mae_db": 0.3288991597905058, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5665016422192548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22166757764710393, "3000–6000 Hz_10ms_envelope_correlation": 0.9678143636862838, "3000–6000 Hz_10ms_modulation_mae_db": 0.4032625050855691, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1798519551664115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15890447752084746, "3000–6000 Hz_40ms_envelope_correlation": 0.9826770093900997, "3000–6000 Hz_40ms_modulation_mae_db": 0.42462541773624585, "mr_stft_db": 5.911040186271799, "band_mae_db": 0.2306102435684312, "rms_gain_error_db": -0.08738284932851331, "output_peak": 0.5213780534422017}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4680409730171364, "全帯域_10ms_rms_envelope_nrmse": 0.14205326181304614, "全帯域_10ms_envelope_correlation": 0.9781396690449441, "全帯域_10ms_modulation_mae_db": 0.3150167362078562, "全帯域_40ms_log_rms_mae_db": 0.6947863894588027, "全帯域_40ms_rms_envelope_nrmse": 0.0655431525309043, "全帯域_40ms_envelope_correlation": 0.9950836183671878, "全帯域_40ms_modulation_mae_db": 0.5213847138644919, "300–1000 Hz_10ms_log_rms_mae_db": 1.953016954229807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18767443690450247, "300–1000 Hz_10ms_envelope_correlation": 0.9735732232314093, "300–1000 Hz_10ms_modulation_mae_db": 0.5721493242649155, "300–1000 Hz_40ms_log_rms_mae_db": 1.2377120138636364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09455065543100986, "300–1000 Hz_40ms_envelope_correlation": 0.9936624226492928, "300–1000 Hz_40ms_modulation_mae_db": 0.3633713054304761, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7560779424787714, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2387550515721642, "1000–3000 Hz_10ms_envelope_correlation": 0.9544816551350692, "1000–3000 Hz_10ms_modulation_mae_db": 0.48541064089277436, "1000–3000 Hz_40ms_log_rms_mae_db": 1.047773204844465, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12657412785215305, "1000–3000 Hz_40ms_envelope_correlation": 0.9866261404599497, "1000–3000 Hz_40ms_modulation_mae_db": 1.1321020513199955, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6166892920052338, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24211652912688625, "3000–6000 Hz_10ms_envelope_correlation": 0.9597397327984593, "3000–6000 Hz_10ms_modulation_mae_db": 0.5053687535715264, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2553697292013442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16956907082506012, "3000–6000 Hz_40ms_envelope_correlation": 0.9795074025942784, "3000–6000 Hz_40ms_modulation_mae_db": 0.36951526556050346, "mr_stft_db": 5.919773057844623, "band_mae_db": 0.27462181653333234, "rms_gain_error_db": -0.09773215211832847, "output_peak": 0.5339278301208583}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4156515794743525, "全帯域_10ms_rms_envelope_nrmse": 0.13748187498717693, "全帯域_10ms_envelope_correlation": 0.9795378833514788, "全帯域_10ms_modulation_mae_db": 0.6391886410831152, "全帯域_40ms_log_rms_mae_db": 0.6382794988168349, "全帯域_40ms_rms_envelope_nrmse": 0.06766978164517906, "全帯域_40ms_envelope_correlation": 0.9945955638605269, "全帯域_40ms_modulation_mae_db": 0.7058376214571095, "300–1000 Hz_10ms_log_rms_mae_db": 2.037935265162657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.180865523587598, "300–1000 Hz_10ms_envelope_correlation": 0.9757241698188714, "300–1000 Hz_10ms_modulation_mae_db": 1.171467802696341, "300–1000 Hz_40ms_log_rms_mae_db": 1.3342765974406694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09238662145594254, "300–1000 Hz_40ms_envelope_correlation": 0.9933054706657735, "300–1000 Hz_40ms_modulation_mae_db": 1.1330634214769209, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7002059873971198, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22019852805797613, "1000–3000 Hz_10ms_envelope_correlation": 0.9612825461286295, "1000–3000 Hz_10ms_modulation_mae_db": 0.6469601209657445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9778308772257281, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10116173757248204, "1000–3000 Hz_40ms_envelope_correlation": 0.9918450181143885, "1000–3000 Hz_40ms_modulation_mae_db": 0.6245773333825754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5472673170340425, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23001634895075773, "3000–6000 Hz_10ms_envelope_correlation": 0.9641991664898981, "3000–6000 Hz_10ms_modulation_mae_db": 0.6285419762078265, "3000–6000 Hz_40ms_log_rms_mae_db": 1.164593384481527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16588769822325794, "3000–6000 Hz_40ms_envelope_correlation": 0.9804008254355748, "3000–6000 Hz_40ms_modulation_mae_db": 0.5059207750800193, "mr_stft_db": 5.929660135994561, "band_mae_db": 0.13333015546163587, "rms_gain_error_db": -0.01875672234293211, "output_peak": 0.5257566312607198}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.094061182967709, "spectral_convergence": 0.5142301926640518, "spectral_convergence_status": "ok", "amplitude_floor": 1.4385235391470026e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6372522903032225, "spectral_convergence": 0.43354161875227354, "spectral_convergence_status": "ok", "amplitude_floor": 9.243439915833013e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.041695493017139, "spectral_convergence": 0.47012878117306967, "spectral_convergence_status": "ok", "amplitude_floor": 6.315989368278398e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00029756845932849967, "output_power": 0.00031242616308912546, "delta_db": 0.21160498862222976, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006886429842704082, "output_power": 0.0006814791636183251, "delta_db": -0.045415456358695844, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003825307900717409, "output_power": 0.00039214663019155574, "delta_db": 0.1078208991663683, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019304685978074288, "output_power": 0.00017990439364045207, "delta_db": -0.3062097166372698, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010892090215881134, "output_power": 0.00011567047832495913, "delta_db": 0.26111301684405047, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.08275993922433e-05, "output_power": 8.683513440695027e-05, "delta_db": 0.3113580054841341, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20725368051879137, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.04176259835803594, "pitch": null, "output_rms_dbfs": -27.52404276713806, "output_peak": 0.5063149919821897, "reference_rms_dbfs": -27.565805365496097, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.924336322096024}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017313921740261034, "log_rms_mae_db": 1.4425495709714473, "rms_envelope_nrmse": 0.1419540917689548, "envelope_correlation": 0.9783912591818081, "rms_gain_error_db": 0.04964858643156106, "reference_mean_power": 0.0017487207983142375, "output_mean_power": 0.0017688268986864987}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.577112847045013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002769005691966373, "output_power": 0.0002779175486608212, "reference_active": true, "delta_db": 0.01592121994360727}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.377497445156709e-05, "output_power": 3.8788734026072565e-05, "reference_active": true, "delta_db": 0.6011057629063496}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4953137492438862e-05, "output_power": 1.3563144447362488e-05, "reference_active": true, "delta_db": -0.42371939609002146}], "active_band_count": 3, "active_band_mae_db": 0.3469154596466595}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015508669364693045, "log_rms_mae_db": 0.6951620646531429, "rms_envelope_nrmse": 0.0701148845290689, "envelope_correlation": 0.9943045945408376, "rms_gain_error_db": 0.052943591514854275, "reference_mean_power": 0.0017487025727625826, "output_mean_power": 0.0017701509730177385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.09499315514524e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025296787730251264, "output_power": 0.00025505522282651745, "reference_active": true, "delta_db": 0.03568844582076147}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6687984786311574e-05, "output_power": 1.938205845963068e-05, "reference_active": true, "delta_db": 0.6499600398747442}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.96993861330795e-07, "output_power": 7.028740278104409e-07, "reference_active": true, "delta_db": 0.03648542686537922}], "active_band_count": 3, "active_band_mae_db": 0.24071130418696163}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014635391485956695, "log_rms_mae_db": 1.9536927183606219, "rms_envelope_nrmse": 0.18462632835962847, "envelope_correlation": 0.9745438916712853, "rms_gain_error_db": 0.023379666684300388, "reference_mean_power": 0.0008142561216173605, "output_mean_power": 0.0008186513813475184}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2354521678013805e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001985819153870851, "output_power": 0.00019346817817358485, "reference_active": true, "delta_db": -0.11330153223727019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.365393218996536e-05, "output_power": 2.5628790440423045e-05, "reference_active": true, "delta_db": 0.34824762529740794}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.646479803369054e-06, "output_power": 7.972542358703078e-06, "reference_active": true, "delta_db": 0.7900514595916924}], "active_band_count": 3, "active_band_mae_db": 0.4172002057087902}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012544182666407365, "log_rms_mae_db": 1.2265425722516266, "rms_envelope_nrmse": 0.09489365566082748, "envelope_correlation": 0.992860521036545, "rms_gain_error_db": 0.03172338205392298, "reference_mean_power": 0.0008126713020357493, "output_mean_power": 0.000818629257172068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0135181965386276e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018492598004057708, "output_power": 0.00018108230542346316, "reference_active": true, "delta_db": -0.09121913966703926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0725232186915858e-05, "output_power": 1.2010605076400836e-05, "reference_active": true, "delta_db": 0.4915818421085918}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.105637817013119e-07, "output_power": 5.698023899509466e-07, "reference_active": true, "delta_db": 1.4234363149260907}], "active_band_count": 3, "active_band_mae_db": 0.6687457655672405}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.111876173726916e-05, "log_rms_mae_db": 1.7163766193972978, "rms_envelope_nrmse": 0.2125480562878737, "envelope_correlation": 0.963922800931863, "rms_gain_error_db": -0.23019575152458033, "reference_mean_power": 0.0002425197498003479, "output_mean_power": 0.00022999984130480186}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1105776313023488e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8957411226134796e-05, "output_power": 3.6410141026418854e-05, "reference_active": true, "delta_db": -0.29367728618011096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.645635870343935e-06, "output_power": 6.565385724207059e-06, "reference_active": true, "delta_db": 0.6554738270544929}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.039589927108413e-06, "output_power": 3.1441296370399087e-06, "reference_active": true, "delta_db": -1.0883683626750473}], "active_band_count": 3, "active_band_mae_db": 0.6791731586365503}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.178958171420995e-05, "log_rms_mae_db": 0.9689835979259003, "rms_envelope_nrmse": 0.1024129798753429, "envelope_correlation": 0.9915405649329473, "rms_gain_error_db": -0.23920871196969726, "reference_mean_power": 0.00024219152470715313, "output_mean_power": 0.00022921237972086268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0514876259433796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7077906526859964e-05, "output_power": 3.488197777231575e-05, "reference_active": true, "delta_db": -0.2651410424969203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.355351339185207e-06, "output_power": 3.433174551096699e-06, "reference_active": true, "delta_db": 0.09957882688038852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8063796091292985e-07, "output_power": 2.3709997109818308e-07, "reference_active": true, "delta_db": -0.7321491511866267}], "active_band_count": 3, "active_band_mae_db": 0.36562300685464516}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3400740982798506e-05, "log_rms_mae_db": 1.591085533744132, "rms_envelope_nrmse": 0.2215119936952819, "envelope_correlation": 0.970074272645874, "rms_gain_error_db": 0.42346865772962955, "reference_mean_power": 0.0001023804664102342, "output_mean_power": 0.00011286621717935778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5067174911166034e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7037143173191167e-05, "output_power": 3.0084147456954008e-05, "reference_active": true, "delta_db": 0.46376907880821544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.303852533122454e-06, "output_power": 4.622539430052963e-06, "reference_active": true, "delta_db": 0.310232420793697}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8801329619345497e-06, "output_power": 2.630693927849922e-06, "reference_active": true, "delta_db": 1.458817589536305}], "active_band_count": 3, "active_band_mae_db": 0.7442730297127391}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.152733550445532e-05, "log_rms_mae_db": 1.2334491645814754, "rms_envelope_nrmse": 0.16466812529906064, "envelope_correlation": 0.9832436873003827, "rms_gain_error_db": 0.42342523313457003, "reference_mean_power": 0.0001028923086971688, "output_mean_power": 0.00011342934789795558}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.218443109204273e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6964722520221083e-05, "output_power": 2.993242917430651e-05, "reference_active": true, "delta_db": 0.45346008004381555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9782668045768803e-06, "output_power": 2.288103233233817e-06, "reference_active": true, "delta_db": 0.6319075122293931}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5223765481589483e-07, "output_power": 1.8899223009355205e-07, "reference_active": true, "delta_db": 0.9392186456304751}], "active_band_count": 3, "active_band_mae_db": 0.6748620793012279}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7628265881035574, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d8308a9e674122ad2f191662c01120dcb21805c51403258d6c0b87c0c1f42215", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9596427935881924, "standard_deviation": 0.004781271903907539, "minimum": 0.9544816551350692, "maximum": 0.9682291071608548, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7166207186768128, "standard_deviation": 0.04252590514041796, "minimum": 1.6648391428240399, "maximum": 1.7907759706372177, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5553450430708835, "standard_deviation": 0.1624637707355899, "minimum": 0.28913864058635064, "maximum": 0.7843710031599525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22460808456367026, "standard_deviation": 0.013539640111752672, "minimum": 0.20002583170746926, "maximum": 0.2387550515721642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9904301331576706, "standard_deviation": 0.0018874888396616949, "minimum": 0.9866261404599497, "maximum": 0.9925547932476438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9884053532068832, "standard_deviation": 0.037412384233020846, "minimum": 0.951139164707299, "maximum": 1.047773204844465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.689027702230403, "standard_deviation": 0.2594585760274052, "minimum": 0.3288991597905058, "maximum": 1.1321020513199955, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10771694455485939, "standard_deviation": 0.009957040235372383, "minimum": 0.09383422845797068, "maximum": 0.12657412785215305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.966422411509033, "standard_deviation": 0.0033180800212220833, "minimum": 0.9597397327984593, "maximum": 0.970074272645874, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5695721027299427, "standard_deviation": 0.030899907259714136, "minimum": 1.532148808119974, "maximum": 1.6166892920052338, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.555569758633997, "standard_deviation": 0.1376275818620737, "minimum": 0.4032625050855691, "maximum": 0.7459474721079333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22441948312144672, "standard_deviation": 0.00870487944282047, "minimum": 0.212983604559599, "maximum": 0.24211652912688625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9814660694162086, "standard_deviation": 0.001290403087536006, "minimum": 0.9795074025942784, "maximum": 0.9832436873003827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.200663829328965, "standard_deviation": 0.03436484463684345, "minimum": 1.1573483401309326, "maximum": 1.2553697292013442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4403570119276938, "standard_deviation": 0.15305460134097743, "minimum": 0.18271335771613276, "maximum": 0.6748620793012279, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16336677891372914, "standard_deviation": 0.003900951035263894, "minimum": 0.15757844925335177, "maximum": 0.16956907082506012, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9733763080516344, "standard_deviation": 0.002243042147806747, "minimum": 0.9699810852995011, "maximum": 0.976374167604421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9819735367842535, "standard_deviation": 0.04416672683233869, "minimum": 1.9275014963179384, "maximum": 2.0442742848919786, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7040605814583409, "standard_deviation": 0.3100453734582128, "minimum": 0.4172002057087902, "maximum": 1.171467802696341, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18817093120431969, "standard_deviation": 0.007327102164654516, "minimum": 0.1783611471998568, "maximum": 0.19967265567828504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9927001600759855, "standard_deviation": 0.001027427363958989, "minimum": 0.9910115125848534, "maximum": 0.9941001166572915, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2747187632899823, "standard_deviation": 0.05361364088392215, "minimum": 1.1927157820944552, "maximum": 1.3342765974406694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6618670458569021, "standard_deviation": 0.24518009155990483, "minimum": 0.3633713054304761, "maximum": 1.1330634214769209, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09816058522094079, "standard_deviation": 0.008018630062992893, "minimum": 0.08643609077301614, "maximum": 0.11157292027094924, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20243969110193966, "standard_deviation": 0.04036644894421928, "minimum": 0.13333015546163587, "maximum": 0.27462181653333234, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.923127870185724, "standard_deviation": 0.010235146489183621, "minimum": 5.911040186271799, "maximum": 5.942091319715452, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5235142882366934, "standard_deviation": 0.009080529581420804, "minimum": 0.5063149919821897, "maximum": 0.5339278301208583, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07588834383612425, "standard_deviation": 0.08418273626271788, "minimum": -0.19698223972374462, "maximum": 0.04176259835803594, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9777268837484138, "standard_deviation": 0.00091358320630149, "minimum": 0.9767036751918864, "maximum": 0.9795378833514788, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4477031785309094, "standard_deviation": 0.02333573377533875, "minimum": 1.4156515794743525, "maximum": 1.483435664667223, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31102204143536427, "standard_deviation": 0.17679602201915828, "minimum": 0.10911882864839512, "maximum": 0.6391886410831152, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14337422751847068, "standard_deviation": 0.0028513463803502646, "minimum": 0.13748187498717693, "maximum": 0.14668071042322037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9947515538618397, "standard_deviation": 0.000826977341392038, "minimum": 0.9935940750754715, "maximum": 0.995871355254675, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6792916539294298, "standard_deviation": 0.030725770133346954, "minimum": 0.6382794988168349, "maximum": 0.7390112882780486, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37311689046375984, "standard_deviation": 0.203450801549014, "minimum": 0.13996433573466227, "maximum": 0.7058376214571095, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0674258441264836, "standard_deviation": 0.005223758354141478, "minimum": 0.05923002510302686, "maximum": 0.07447548301969595, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0043", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.981201171875, "predicted_mos": 1.981201171875, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c2f2094c773e822dbe969f002d415515755f9baea2edadea665a03ce088bc36a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.134516099933535, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3992768279909946, "全帯域_10ms_rms_envelope_nrmse": 0.13968269928934884, "全帯域_10ms_envelope_correlation": 0.9787539968419531, "全帯域_10ms_modulation_mae_db": 0.1652456593784802, "全帯域_40ms_log_rms_mae_db": 0.7186029747858198, "全帯域_40ms_rms_envelope_nrmse": 0.06811785727643539, "全帯域_40ms_envelope_correlation": 0.9946886830438042, "全帯域_40ms_modulation_mae_db": 0.11502793358944326, "300–1000 Hz_10ms_log_rms_mae_db": 2.1086187563352032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18565156168070865, "300–1000 Hz_10ms_envelope_correlation": 0.9740064660180813, "300–1000 Hz_10ms_modulation_mae_db": 0.5865390070035411, "300–1000 Hz_40ms_log_rms_mae_db": 1.3486290488761983, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09386244525716962, "300–1000 Hz_40ms_envelope_correlation": 0.993503115756249, "300–1000 Hz_40ms_modulation_mae_db": 0.4298773279687236, "1000–3000 Hz_10ms_log_rms_mae_db": 1.932223045188516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2265394153479176, "1000–3000 Hz_10ms_envelope_correlation": 0.9593539613795368, "1000–3000 Hz_10ms_modulation_mae_db": 0.9917738473993265, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0868891050940692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11677176358165303, "1000–3000 Hz_40ms_envelope_correlation": 0.9900108732117371, "1000–3000 Hz_40ms_modulation_mae_db": 1.4089755063268228, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9430184320732697, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23703917168644473, "3000–6000 Hz_10ms_envelope_correlation": 0.9631965566172674, "3000–6000 Hz_10ms_modulation_mae_db": 0.483912937545156, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5097171203293536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17259082217501964, "3000–6000 Hz_40ms_envelope_correlation": 0.9796503464786794, "3000–6000 Hz_40ms_modulation_mae_db": 0.6432965173050883, "mr_stft_db": 6.0477778524130175, "band_mae_db": 0.27884876380647344, "rms_gain_error_db": -0.13025289006720653, "output_peak": 0.49863895985609785}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.408456749774109, "全帯域_10ms_rms_envelope_nrmse": 0.1443097064617326, "全帯域_10ms_envelope_correlation": 0.9773167936363025, "全帯域_10ms_modulation_mae_db": 0.1212297497301072, "全帯域_40ms_log_rms_mae_db": 0.6786619770827312, "全帯域_40ms_rms_envelope_nrmse": 0.07626619784486441, "全帯域_40ms_envelope_correlation": 0.993187292378003, "全帯域_40ms_modulation_mae_db": 0.0892441212855981, "300–1000 Hz_10ms_log_rms_mae_db": 2.2075829445501642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20107748453032348, "300–1000 Hz_10ms_envelope_correlation": 0.9693408726990617, "300–1000 Hz_10ms_modulation_mae_db": 0.43161482168042026, "300–1000 Hz_40ms_log_rms_mae_db": 1.4641381374378002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11887425495950055, "300–1000 Hz_40ms_envelope_correlation": 0.9891004663780774, "300–1000 Hz_40ms_modulation_mae_db": 0.41608724400438285, "1000–3000 Hz_10ms_log_rms_mae_db": 1.928829386194782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22610044492866016, "1000–3000 Hz_10ms_envelope_correlation": 0.9591056505556976, "1000–3000 Hz_10ms_modulation_mae_db": 0.6403325751206763, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1301506556405998, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10553101846047792, "1000–3000 Hz_40ms_envelope_correlation": 0.9907654412188451, "1000–3000 Hz_40ms_modulation_mae_db": 0.7678858779623781, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8602718695096083, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2180090383097265, "3000–6000 Hz_10ms_envelope_correlation": 0.9682281963700048, "3000–6000 Hz_10ms_modulation_mae_db": 0.5354191744680625, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3656544461037423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15984863886779738, "3000–6000 Hz_40ms_envelope_correlation": 0.9820637482175497, "3000–6000 Hz_40ms_modulation_mae_db": 0.5603176905960829, "mr_stft_db": 6.059177274353851, "band_mae_db": 0.14272063612941643, "rms_gain_error_db": -0.0956587943422675, "output_peak": 0.551554910533876}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4067759062385197, "全帯域_10ms_rms_envelope_nrmse": 0.1453394411003539, "全帯域_10ms_envelope_correlation": 0.9771354737245934, "全帯域_10ms_modulation_mae_db": 0.0973265090753513, "全帯域_40ms_log_rms_mae_db": 0.6776948744723211, "全帯域_40ms_rms_envelope_nrmse": 0.06169338943144506, "全帯域_40ms_envelope_correlation": 0.9955207874578861, "全帯域_40ms_modulation_mae_db": 0.20010639109223924, "300–1000 Hz_10ms_log_rms_mae_db": 2.1104880147059095, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18419375744970856, "300–1000 Hz_10ms_envelope_correlation": 0.9743291663264013, "300–1000 Hz_10ms_modulation_mae_db": 0.4736520626349847, "300–1000 Hz_40ms_log_rms_mae_db": 1.3277727916081223, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08694821817911576, "300–1000 Hz_40ms_envelope_correlation": 0.994006358412022, "300–1000 Hz_40ms_modulation_mae_db": 0.41061242709426776, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9090665877304183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23449106417280885, "1000–3000 Hz_10ms_envelope_correlation": 0.9563475033247302, "1000–3000 Hz_10ms_modulation_mae_db": 0.8261619038684739, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1035200502897657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11296867016280872, "1000–3000 Hz_40ms_envelope_correlation": 0.9891719907654867, "1000–3000 Hz_40ms_modulation_mae_db": 1.109923729550988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.89130427824945, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22301430846339804, "3000–6000 Hz_10ms_envelope_correlation": 0.9677706681976247, "3000–6000 Hz_10ms_modulation_mae_db": 0.19573938828651302, "3000–6000 Hz_40ms_log_rms_mae_db": 1.445094302020188, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16284715562924035, "3000–6000 Hz_40ms_envelope_correlation": 0.9821945382017738, "3000–6000 Hz_40ms_modulation_mae_db": 0.10513250058774735, "mr_stft_db": 6.075844943601695, "band_mae_db": 0.19159743136854612, "rms_gain_error_db": -0.020507979859767763, "output_peak": 0.5066645674374095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.417021094698338, "全帯域_10ms_rms_envelope_nrmse": 0.15018714774333036, "全帯域_10ms_envelope_correlation": 0.9758646740022071, "全帯域_10ms_modulation_mae_db": 0.10613689891635426, "全帯域_40ms_log_rms_mae_db": 0.6755625517880026, "全帯域_40ms_rms_envelope_nrmse": 0.06983745115901241, "全帯域_40ms_envelope_correlation": 0.994371209769343, "全帯域_40ms_modulation_mae_db": 0.147081010085045, "300–1000 Hz_10ms_log_rms_mae_db": 2.2263724950015926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18020179865628888, "300–1000 Hz_10ms_envelope_correlation": 0.9754779092450064, "300–1000 Hz_10ms_modulation_mae_db": 0.3306474638515284, "300–1000 Hz_40ms_log_rms_mae_db": 1.3646223638914479, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08859367448449468, "300–1000 Hz_40ms_envelope_correlation": 0.9937921197558932, "300–1000 Hz_40ms_modulation_mae_db": 0.49831067883030294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9572867789723887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25386180446509726, "1000–3000 Hz_10ms_envelope_correlation": 0.9492337224755732, "1000–3000 Hz_10ms_modulation_mae_db": 0.6305239592293131, "1000–3000 Hz_40ms_log_rms_mae_db": 1.167452567155163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1230710318145697, "1000–3000 Hz_40ms_envelope_correlation": 0.9872314284606061, "1000–3000 Hz_40ms_modulation_mae_db": 1.3121759759136298, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8321984378887584, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22009413632240882, "3000–6000 Hz_10ms_envelope_correlation": 0.9690829900630494, "3000–6000 Hz_10ms_modulation_mae_db": 0.3708860148032707, "3000–6000 Hz_40ms_log_rms_mae_db": 1.410083387259336, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1613346731856282, "3000–6000 Hz_40ms_envelope_correlation": 0.9828444387869041, "3000–6000 Hz_40ms_modulation_mae_db": 0.23213147004356113, "mr_stft_db": 6.06271906409237, "band_mae_db": 0.14697868901434435, "rms_gain_error_db": 0.034684009186777096, "output_peak": 0.5121729562347067}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.364725392225852, "全帯域_10ms_rms_envelope_nrmse": 0.14974895704194519, "全帯域_10ms_envelope_correlation": 0.9759136951959009, "全帯域_10ms_modulation_mae_db": 0.3179215485204777, "全帯域_40ms_log_rms_mae_db": 0.6153921442094966, "全帯域_40ms_rms_envelope_nrmse": 0.0724383522277212, "全帯域_40ms_envelope_correlation": 0.9938907219879163, "全帯域_40ms_modulation_mae_db": 0.2855714009327557, "300–1000 Hz_10ms_log_rms_mae_db": 2.190032139276168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1894010051043928, "300–1000 Hz_10ms_envelope_correlation": 0.9730072734296855, "300–1000 Hz_10ms_modulation_mae_db": 0.13735718926559137, "300–1000 Hz_40ms_log_rms_mae_db": 1.425500949078587, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09702839180943892, "300–1000 Hz_40ms_envelope_correlation": 0.9925134720138666, "300–1000 Hz_40ms_modulation_mae_db": 0.4065816655234411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9154634366985694, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23835326129534276, "1000–3000 Hz_10ms_envelope_correlation": 0.9551460360695089, "1000–3000 Hz_10ms_modulation_mae_db": 0.3520063849868109, "1000–3000 Hz_40ms_log_rms_mae_db": 1.094660418761862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11503719218680078, "1000–3000 Hz_40ms_envelope_correlation": 0.9888111598086555, "1000–3000 Hz_40ms_modulation_mae_db": 0.8424642920947285, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8483207536329982, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23234652449904095, "3000–6000 Hz_10ms_envelope_correlation": 0.9635351711455694, "3000–6000 Hz_10ms_modulation_mae_db": 0.5514949587936814, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4256307290837378, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17310752255454612, "3000–6000 Hz_40ms_envelope_correlation": 0.978804266860908, "3000–6000 Hz_40ms_modulation_mae_db": 0.28333455118710976, "mr_stft_db": 6.0592954425594305, "band_mae_db": 0.14169747832480942, "rms_gain_error_db": 0.04021923895399911, "output_peak": 0.5349499731307197}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3920662593945647, "全帯域_10ms_rms_envelope_nrmse": 0.148632604842575, "全帯域_10ms_envelope_correlation": 0.9759168958965934, "全帯域_10ms_modulation_mae_db": 0.13287201216253688, "全帯域_40ms_log_rms_mae_db": 0.6589461895854251, "全帯域_40ms_rms_envelope_nrmse": 0.07394414580711146, "全帯域_40ms_envelope_correlation": 0.9937818188783534, "全帯域_40ms_modulation_mae_db": 0.41250019639708574, "300–1000 Hz_10ms_log_rms_mae_db": 2.2324378399156153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21457774998850299, "300–1000 Hz_10ms_envelope_correlation": 0.964990520290053, "300–1000 Hz_10ms_modulation_mae_db": 0.5935692127277137, "300–1000 Hz_40ms_log_rms_mae_db": 1.4964949673657242, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11602516110099767, "300–1000 Hz_40ms_envelope_correlation": 0.9897615517544832, "300–1000 Hz_40ms_modulation_mae_db": 0.13134285144648647, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8780453506033352, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2198452352825658, "1000–3000 Hz_10ms_envelope_correlation": 0.9613939432001346, "1000–3000 Hz_10ms_modulation_mae_db": 0.7410031826116827, "1000–3000 Hz_40ms_log_rms_mae_db": 1.068605096374491, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10692968126614545, "1000–3000 Hz_40ms_envelope_correlation": 0.9903971170711575, "1000–3000 Hz_40ms_modulation_mae_db": 0.8466935866304975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8058412611455235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22613445334200372, "3000–6000 Hz_10ms_envelope_correlation": 0.9664515759080172, "3000–6000 Hz_10ms_modulation_mae_db": 0.4639014852299505, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3841559804256813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16339882231617223, "3000–6000 Hz_40ms_envelope_correlation": 0.9817483121749652, "3000–6000 Hz_40ms_modulation_mae_db": 0.4481465612014288, "mr_stft_db": 6.049540687498815, "band_mae_db": 0.14679141989826783, "rms_gain_error_db": -0.15062815349757375, "output_peak": 0.5206360869902201}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3977448358358815, "全帯域_10ms_rms_envelope_nrmse": 0.13820538634872523, "全帯域_10ms_envelope_correlation": 0.9792064898397033, "全帯域_10ms_modulation_mae_db": 0.21028829828162157, "全帯域_40ms_log_rms_mae_db": 0.5896347215687315, "全帯域_40ms_rms_envelope_nrmse": 0.05239883807888766, "全帯域_40ms_envelope_correlation": 0.99681121268752, "全帯域_40ms_modulation_mae_db": 0.22822161183785883, "300–1000 Hz_10ms_log_rms_mae_db": 2.24066969772133, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1804428816184763, "300–1000 Hz_10ms_envelope_correlation": 0.9754068781335814, "300–1000 Hz_10ms_modulation_mae_db": 0.5452147025526172, "300–1000 Hz_40ms_log_rms_mae_db": 1.4293736135895025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08022189591838973, "300–1000 Hz_40ms_envelope_correlation": 0.9951947425594466, "300–1000 Hz_40ms_modulation_mae_db": 0.16098130546590403, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8834967558852351, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.224947974401407, "1000–3000 Hz_10ms_envelope_correlation": 0.9595196952361899, "1000–3000 Hz_10ms_modulation_mae_db": 0.6419966189049667, "1000–3000 Hz_40ms_log_rms_mae_db": 1.116530273411249, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1094786548270734, "1000–3000 Hz_40ms_envelope_correlation": 0.9903907193610995, "1000–3000 Hz_40ms_modulation_mae_db": 0.8680499598417194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9673365267589762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23928511270342503, "3000–6000 Hz_10ms_envelope_correlation": 0.9616834524639106, "3000–6000 Hz_10ms_modulation_mae_db": 0.7417291293063423, "3000–6000 Hz_40ms_log_rms_mae_db": 1.509981648692931, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17142050514763404, "3000–6000 Hz_40ms_envelope_correlation": 0.9794312734749551, "3000–6000 Hz_40ms_modulation_mae_db": 0.7668730213104523, "mr_stft_db": 6.062270151979362, "band_mae_db": 0.14797540718216418, "rms_gain_error_db": -0.06812671661741651, "output_peak": 0.5627032879954337}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3566745230896562, "全帯域_10ms_rms_envelope_nrmse": 0.1312920591228156, "全帯域_10ms_envelope_correlation": 0.9812906713547344, "全帯域_10ms_modulation_mae_db": 0.19634409891764107, "全帯域_40ms_log_rms_mae_db": 0.5904888630793326, "全帯域_40ms_rms_envelope_nrmse": 0.06067803320214956, "全帯域_40ms_envelope_correlation": 0.9958620810907055, "全帯域_40ms_modulation_mae_db": 0.11260489832959975, "300–1000 Hz_10ms_log_rms_mae_db": 2.164126163106558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18321798159126812, "300–1000 Hz_10ms_envelope_correlation": 0.974770977111769, "300–1000 Hz_10ms_modulation_mae_db": 0.5653299742983976, "300–1000 Hz_40ms_log_rms_mae_db": 1.3706771440440197, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1077641182182794, "300–1000 Hz_40ms_envelope_correlation": 0.9913249510685339, "300–1000 Hz_40ms_modulation_mae_db": 0.4612802801930825, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9269526912246508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23228046875718059, "1000–3000 Hz_10ms_envelope_correlation": 0.9572106409572808, "1000–3000 Hz_10ms_modulation_mae_db": 0.9016300263540976, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0914393887021796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11568583140570977, "1000–3000 Hz_40ms_envelope_correlation": 0.9886622835869306, "1000–3000 Hz_40ms_modulation_mae_db": 1.1990026219721006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8919328941681395, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23647552155234047, "3000–6000 Hz_10ms_envelope_correlation": 0.9622995565997936, "3000–6000 Hz_10ms_modulation_mae_db": 0.45931990363808667, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4581424871423947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17188417885884075, "3000–6000 Hz_40ms_envelope_correlation": 0.979098536505325, "3000–6000 Hz_40ms_modulation_mae_db": 0.28945723198427437, "mr_stft_db": 6.0591961388781455, "band_mae_db": 0.11189032971035501, "rms_gain_error_db": -0.1412549491839481, "output_peak": 0.49970231265162135}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.275687321928733, "spectral_convergence": 0.49994200944133854, "spectral_convergence_status": "ok", "amplitude_floor": 1.4385235391470026e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.780329722497736, "spectral_convergence": 0.42387657836330545, "spectral_convergence_status": "ok", "amplitude_floor": 9.243439915833013e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.087316512812587, "spectral_convergence": 0.46329161790696516, "spectral_convergence_status": "ok", "amplitude_floor": 6.315989368278398e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00029756845932849967, "output_power": 0.0003228943135469158, "delta_db": 0.35473500364645255, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006886429842704082, "output_power": 0.0006525963313760851, "delta_db": -0.2334949931960599, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003825307900717409, "output_power": 0.0003576746380840472, "delta_db": -0.29178251069203925, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019304685978074288, "output_power": 0.00017095081793754594, "delta_db": -0.5279155843681703, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010892090215881134, "output_power": 0.00011244836659579097, "delta_db": 0.13841921523665207, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.08275993922433e-05, "output_power": 8.322124485802839e-05, "delta_db": 0.12674527569946673, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27884876380647344, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13025289006720653, "pitch": null, "output_rms_dbfs": -27.6960582555633, "output_peak": 0.49863895985609785, "reference_rms_dbfs": -27.565805365496097, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0477778524130175}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017313921740261034, "log_rms_mae_db": 1.3992768279909946, "rms_envelope_nrmse": 0.13968269928934884, "envelope_correlation": 0.9787539968419531, "rms_gain_error_db": -0.12366431988333303, "reference_mean_power": 0.0017487207983142375, "output_mean_power": 0.0016996286504744767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.577112847045013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002769005691966373, "output_power": 0.000274240562885023, "reference_active": true, "delta_db": -0.041921568385893865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.377497445156709e-05, "output_power": 3.168221011691159e-05, "reference_active": true, "delta_db": -0.277795592901503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4953137492438862e-05, "output_power": 1.4359203318936497e-05, "reference_active": true, "delta_db": -0.1760198168480437}], "active_band_count": 3, "active_band_mae_db": 0.1652456593784802}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015508669364693045, "log_rms_mae_db": 0.7186029747858198, "rms_envelope_nrmse": 0.06811785727643539, "envelope_correlation": 0.9946886830438042, "rms_gain_error_db": -0.12189869441262668, "reference_mean_power": 0.0017487025727625826, "output_mean_power": 0.0017003020543667259}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.09499315514524e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025296787730251264, "output_power": 0.00025100502903927733, "reference_active": true, "delta_db": -0.03382953603903115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6687984786311574e-05, "output_power": 1.5789522987122017e-05, "reference_active": true, "delta_db": -0.24034885298869993}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.96993861330795e-07, "output_power": 6.857067300070187e-07, "reference_active": true, "delta_db": -0.07090541174059871}], "active_band_count": 3, "active_band_mae_db": 0.11502793358944326}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014635391485956695, "log_rms_mae_db": 2.1086187563352032, "rms_envelope_nrmse": 0.18565156168070865, "envelope_correlation": 0.9740064660180813, "rms_gain_error_db": -0.24413914271196682, "reference_mean_power": 0.0008142561216173605, "output_mean_power": 0.0007697454279806357}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2354521678013805e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001985819153870851, "output_power": 0.00018244922848057772, "reference_active": true, "delta_db": -0.3679766407909617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.365393218996536e-05, "output_power": 2.1142677678842992e-05, "reference_active": true, "delta_db": -0.48743358457985997}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.646479803369054e-06, "output_power": 5.397231045848686e-06, "reference_active": true, "delta_db": -0.9042067956398019}], "active_band_count": 3, "active_band_mae_db": 0.5865390070035411}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012544182666407365, "log_rms_mae_db": 1.3486290488761983, "rms_envelope_nrmse": 0.09386244525716962, "envelope_correlation": 0.993503115756249, "rms_gain_error_db": -0.24051976420677076, "reference_mean_power": 0.0008126713020357493, "output_mean_power": 0.0007688877598668506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0135181965386276e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018492598004057708, "output_power": 0.00016936090015851093, "reference_active": true, "delta_db": -0.3818477567685516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0725232186915858e-05, "output_power": 9.937887987676307e-06, "reference_active": true, "delta_db": -0.33112605452920274}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.105637817013119e-07, "output_power": 3.5951318098715185e-07, "reference_active": true, "delta_db": -0.5766581726084162}], "active_band_count": 3, "active_band_mae_db": 0.4298773279687236}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.111876173726916e-05, "log_rms_mae_db": 1.932223045188516, "rms_envelope_nrmse": 0.2265394153479176, "envelope_correlation": 0.9593539613795368, "rms_gain_error_db": -0.41440711598694335, "reference_mean_power": 0.0002425197498003479, "output_mean_power": 0.0002204481202226583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1105776313023488e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8957411226134796e-05, "output_power": 3.1276768537008044e-05, "reference_active": true, "delta_db": -0.9536821356434934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.645635870343935e-06, "output_power": 4.874251281110009e-06, "reference_active": true, "delta_db": -0.6380494928075582}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.039589927108413e-06, "output_power": 2.9375027998469634e-06, "reference_active": true, "delta_db": -1.383589913746928}], "active_band_count": 3, "active_band_mae_db": 0.9917738473993265}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.178958171420995e-05, "log_rms_mae_db": 1.0868891050940692, "rms_envelope_nrmse": 0.11677176358165303, "envelope_correlation": 0.9900108732117371, "rms_gain_error_db": -0.416063569794638, "reference_mean_power": 0.00024219152470715313, "output_mean_power": 0.00022006581488918285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0514876259433796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7077906526859964e-05, "output_power": 2.9396291814522664e-05, "reference_active": true, "delta_db": -1.0082265532553356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.355351339185207e-06, "output_power": 2.620916471013529e-06, "reference_active": true, "delta_db": -1.0728482176678038}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8063796091292985e-07, "output_power": 1.712226606616228e-07, "reference_active": true, "delta_db": -2.1458517480573285}], "active_band_count": 3, "active_band_mae_db": 1.4089755063268228}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3400740982798506e-05, "log_rms_mae_db": 1.9430184320732697, "rms_envelope_nrmse": 0.23703917168644473, "envelope_correlation": 0.9631965566172674, "rms_gain_error_db": 0.19754770053836873, "reference_mean_power": 0.0001023804664102342, "output_mean_power": 0.00010714499126368941}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5067174911166034e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7037143173191167e-05, "output_power": 2.756806725499547e-05, "reference_active": true, "delta_db": 0.08445518791598897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.303852533122454e-06, "output_power": 4.468764646331351e-06, "reference_active": true, "delta_db": 0.16330100263307115}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8801329619345497e-06, "output_power": 2.4807718523930466e-06, "reference_active": true, "delta_db": 1.2039826220864078}], "active_band_count": 3, "active_band_mae_db": 0.483912937545156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.152733550445532e-05, "log_rms_mae_db": 1.5097171203293536, "rms_envelope_nrmse": 0.17259082217501964, "envelope_correlation": 0.9796503464786794, "rms_gain_error_db": 0.19745018057085262, "reference_mean_power": 0.0001028923086971688, "output_mean_power": 0.00010767823545990405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.218443109204273e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6964722520221083e-05, "output_power": 2.7380089406563464e-05, "reference_active": true, "delta_db": 0.0663890638839821}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9782668045768803e-06, "output_power": 2.3902663676993878e-06, "reference_active": true, "delta_db": 0.821614371994554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5223765481589483e-07, "output_power": 1.9351324549308412e-07, "reference_active": true, "delta_db": 1.0418861160367285}], "active_band_count": 3, "active_band_mae_db": 0.6432965173050883}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7541467391368754, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c87437558a90da959314b7e25476359738df1d15a7b6522d3a8f062c321efbcb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9571638941498315, "standard_deviation": 0.003772796980559951, "minimum": 0.9492337224755732, "maximum": 0.9613939432001346, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9164205040622369, "standard_deviation": 0.026176056981399264, "minimum": 1.8780453506033352, "maximum": 1.9572867789723887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7156785623094184, "standard_deviation": 0.19794272769752122, "minimum": 0.3520063849868109, "maximum": 0.9917738473993265, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23205245858137252, "standard_deviation": 0.010600920289790879, "minimum": 0.2198452352825658, "maximum": 0.25386180446509726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9894301266855647, "standard_deviation": 0.001186335966692198, "minimum": 0.9872314284606061, "maximum": 0.9907654412188451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1074059444286724, "standard_deviation": 0.03062468232816851, "minimum": 1.068605096374491, "maximum": 1.167452567155163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0443964437866082, "standard_deviation": 0.2450189074944456, "minimum": 0.7678858779623781, "maximum": 1.4089755063268228, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11318423046315484, "standard_deviation": 0.00575424695231125, "minimum": 0.10553101846047792, "maximum": 0.1230710318145697, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9652810209206546, "standard_deviation": 0.0029260887155266244, "minimum": 0.9616834524639106, "maximum": 0.9690829900630494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8800280566783405, "standard_deviation": 0.05483279484244358, "minimum": 1.8058412611455235, "maximum": 1.9673365267589762, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4753003740088829, "standard_deviation": 0.1558022354801324, "minimum": 0.19573938828651302, "maximum": 0.7417291293063423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22904978335984855, "standard_deviation": 0.008294836206998517, "minimum": 0.2180090383097265, "maximum": 0.23928511270342503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9807294325876326, "standard_deviation": 0.0016324402643314484, "minimum": 0.978804266860908, "maximum": 0.9828444387869041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4385575126321706, "standard_deviation": 0.05324291929587645, "minimum": 1.3656544461037423, "maximum": 1.509981648692931, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.41608619302696814, "standard_deviation": 0.22699778274237592, "minimum": 0.10513250058774735, "maximum": 0.7668730213104523, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16705403984185985, "standard_deviation": 0.005674307029199864, "minimum": 0.15984863886779738, "maximum": 0.17310752255454612, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9726662579067049, "standard_deviation": 0.003674216610929456, "minimum": 0.964990520290053, "maximum": 0.9754779092450064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1850410063265677, "standard_deviation": 0.05266917369486964, "minimum": 2.1086187563352032, "maximum": 2.24066969772133, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4579905542518493, "standard_deviation": 0.15766105598055344, "minimum": 0.13735718926559137, "maximum": 0.5935692127277137, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1898455275774587, "standard_deviation": 0.01202915958626909, "minimum": 0.18020179865628888, "maximum": 0.21457774998850299, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9923995972123215, "standard_deviation": 0.0021568088589869303, "minimum": 0.9891004663780774, "maximum": 0.9951947425594466, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4034011269864253, "standard_deviation": 0.05953713958696723, "minimum": 1.3277727916081223, "maximum": 1.4964949673657242, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3643842225658239, "standard_deviation": 0.13829904935965007, "minimum": 0.13134285144648647, "maximum": 0.49831067883030294, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0986647699909233, "standard_deviation": 0.014127929844650462, "minimum": 0.08022189591838973, "maximum": 0.11887425495950055, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1635625194292971, "standard_deviation": 0.05135117890048724, "minimum": 0.11189032971035501, "maximum": 0.27884876380647344, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.059477694422086, "standard_deviation": 0.008635675946489943, "minimum": 6.0477778524130175, "maximum": 6.075844943601695, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5233778818537607, "standard_deviation": 0.024081099063087565, "minimum": 0.49863895985609785, "maximum": 0.5627032879954337, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0664407794284255, "standard_deviation": 0.07683525851993495, "minimum": -0.15062815349757375, "maximum": 0.04021923895399911, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9776748363114984, "standard_deviation": 0.001944684319429484, "minimum": 0.9758646740022071, "maximum": 0.9812906713547344, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3928426986559894, "standard_deviation": 0.021338360403199644, "minimum": 1.3566745230896562, "maximum": 1.417021094698338, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.16842059687282127, "standard_deviation": 0.07301423289741865, "minimum": 0.0973265090753513, "maximum": 0.3179215485204777, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14342475024385334, "standard_deviation": 0.006608905769871692, "minimum": 0.1312920591228156, "maximum": 0.15018714774333036, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9947642259116914, "standard_deviation": 0.0012155449509461422, "minimum": 0.993187292378003, "maximum": 0.99681121268752, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6506230370714825, "standard_deviation": 0.04691710326637959, "minimum": 0.5896347215687315, "maximum": 0.7186029747858198, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.1987946954437032, "standard_deviation": 0.10901447268077781, "minimum": 0.0892441212855981, "maximum": 0.41250019639708574, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0669217831284534, "standard_deviation": 0.008055453864553249, "minimum": 0.05239883807888766, "maximum": 0.07626619784486441, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0043", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.5447473526000977, "predicted_mos": 2.5447473526000977, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7c511e8e147fd342dce35c524aed6182085632fe3399082861c71303adc7b34f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9368663001805544, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3974611162711725, "全帯域_10ms_rms_envelope_nrmse": 0.14576181456118642, "全帯域_10ms_envelope_correlation": 0.9768232008585833, "全帯域_10ms_modulation_mae_db": 0.381680933869274, "全帯域_40ms_log_rms_mae_db": 0.6401452787711608, "全帯域_40ms_rms_envelope_nrmse": 0.0647094558796994, "全帯域_40ms_envelope_correlation": 0.9951671696589047, "全帯域_40ms_modulation_mae_db": 0.563650504749396, "300–1000 Hz_10ms_log_rms_mae_db": 2.149873233057337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18805678866575593, "300–1000 Hz_10ms_envelope_correlation": 0.9732171555608057, "300–1000 Hz_10ms_modulation_mae_db": 0.44109373513887934, "300–1000 Hz_40ms_log_rms_mae_db": 1.334552705732181, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09343499384598818, "300–1000 Hz_40ms_envelope_correlation": 0.993275502026902, "300–1000 Hz_40ms_modulation_mae_db": 0.38831949957889983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8741056452004519, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2330704621650075, "1000–3000 Hz_10ms_envelope_correlation": 0.9565542470998942, "1000–3000 Hz_10ms_modulation_mae_db": 0.3703773644457145, "1000–3000 Hz_40ms_log_rms_mae_db": 1.014584780221927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10554729873812223, "1000–3000 Hz_40ms_envelope_correlation": 0.991602529085704, "1000–3000 Hz_40ms_modulation_mae_db": 0.4700665309601571, "3000–6000 Hz_10ms_log_rms_mae_db": 1.821161597721064, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2165726970135195, "3000–6000 Hz_10ms_envelope_correlation": 0.9683653642785744, "3000–6000 Hz_10ms_modulation_mae_db": 0.23947627757331294, "3000–6000 Hz_40ms_log_rms_mae_db": 1.373185725240691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16351015204825345, "3000–6000 Hz_40ms_envelope_correlation": 0.9811022055291778, "3000–6000 Hz_40ms_modulation_mae_db": 0.3305803167406454, "mr_stft_db": 6.027498146668946, "band_mae_db": 0.20127524803007815, "rms_gain_error_db": -0.09824451115140424, "output_peak": 0.5660082994486727}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3794164621669247, "全帯域_10ms_rms_envelope_nrmse": 0.1428244858354116, "全帯域_10ms_envelope_correlation": 0.9778656604399025, "全帯域_10ms_modulation_mae_db": 0.580479847015375, "全帯域_40ms_log_rms_mae_db": 0.6818717506913685, "全帯域_40ms_rms_envelope_nrmse": 0.06454062440654461, "全帯域_40ms_envelope_correlation": 0.995075196766964, "全帯域_40ms_modulation_mae_db": 0.29147289443348695, "300–1000 Hz_10ms_log_rms_mae_db": 2.1960073905095663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1962466633910721, "300–1000 Hz_10ms_envelope_correlation": 0.9709031930278058, "300–1000 Hz_10ms_modulation_mae_db": 0.34645492218643065, "300–1000 Hz_40ms_log_rms_mae_db": 1.3640565209676196, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09652763336931952, "300–1000 Hz_40ms_envelope_correlation": 0.992551777604311, "300–1000 Hz_40ms_modulation_mae_db": 0.30922391715866065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8131149320233795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22418637072533923, "1000–3000 Hz_10ms_envelope_correlation": 0.9600407682110617, "1000–3000 Hz_10ms_modulation_mae_db": 0.5659797757254322, "1000–3000 Hz_40ms_log_rms_mae_db": 1.032500398961578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10315377243546431, "1000–3000 Hz_40ms_envelope_correlation": 0.9909996403086083, "1000–3000 Hz_40ms_modulation_mae_db": 1.3394021054640521, "3000–6000 Hz_10ms_log_rms_mae_db": 1.763956943627243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21206438886663032, "3000–6000 Hz_10ms_envelope_correlation": 0.9700274701262424, "3000–6000 Hz_10ms_modulation_mae_db": 0.4510252785076206, "3000–6000 Hz_40ms_log_rms_mae_db": 1.322836588936279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16035128991217945, "3000–6000 Hz_40ms_envelope_correlation": 0.9819995511159197, "3000–6000 Hz_40ms_modulation_mae_db": 0.22885565522298248, "mr_stft_db": 6.0126200182131315, "band_mae_db": 0.16502567384951813, "rms_gain_error_db": -0.023053707517594085, "output_peak": 0.5750501313145686}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4483741218172959, "全帯域_10ms_rms_envelope_nrmse": 0.14381349017992284, "全帯域_10ms_envelope_correlation": 0.9776755650499819, "全帯域_10ms_modulation_mae_db": 0.23946704988578305, "全帯域_40ms_log_rms_mae_db": 0.6771522117631582, "全帯域_40ms_rms_envelope_nrmse": 0.06417916188662068, "全帯域_40ms_envelope_correlation": 0.9951537692614333, "全帯域_40ms_modulation_mae_db": 0.1383091522405334, "300–1000 Hz_10ms_log_rms_mae_db": 2.1332924382109, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18529751369645167, "300–1000 Hz_10ms_envelope_correlation": 0.9740011733174774, "300–1000 Hz_10ms_modulation_mae_db": 0.6017653260793562, "300–1000 Hz_40ms_log_rms_mae_db": 1.3832410191548108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09168342984746082, "300–1000 Hz_40ms_envelope_correlation": 0.9935378960976617, "300–1000 Hz_40ms_modulation_mae_db": 0.5094520110634048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8261616060042045, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21192700586226368, "1000–3000 Hz_10ms_envelope_correlation": 0.9641527469389939, "1000–3000 Hz_10ms_modulation_mae_db": 0.36178324291362207, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0398290332887072, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10602122991476758, "1000–3000 Hz_40ms_envelope_correlation": 0.9905990246249329, "1000–3000 Hz_40ms_modulation_mae_db": 0.5018311177995697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8038674369896828, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22089748179431765, "3000–6000 Hz_10ms_envelope_correlation": 0.9669239332950811, "3000–6000 Hz_10ms_modulation_mae_db": 0.38327446708981067, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3530352324488273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16130938916528195, "3000–6000 Hz_40ms_envelope_correlation": 0.981495737704657, "3000–6000 Hz_40ms_modulation_mae_db": 0.2904540055538724, "mr_stft_db": 6.01768045322451, "band_mae_db": 0.15689300908486983, "rms_gain_error_db": 0.01463184457331599, "output_peak": 0.5293850860126641}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.344329816710561, "全帯域_10ms_rms_envelope_nrmse": 0.13667636224444435, "全帯域_10ms_envelope_correlation": 0.979668569980674, "全帯域_10ms_modulation_mae_db": 0.13907748961663097, "全帯域_40ms_log_rms_mae_db": 0.5923870557437254, "全帯域_40ms_rms_envelope_nrmse": 0.06010595423534963, "全帯域_40ms_envelope_correlation": 0.9958665151343886, "全帯域_40ms_modulation_mae_db": 0.21976223168448916, "300–1000 Hz_10ms_log_rms_mae_db": 2.1344238719139783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18686938626461178, "300–1000 Hz_10ms_envelope_correlation": 0.9735973816207809, "300–1000 Hz_10ms_modulation_mae_db": 0.21348610277566146, "300–1000 Hz_40ms_log_rms_mae_db": 1.3633147711560953, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09199968112068654, "300–1000 Hz_40ms_envelope_correlation": 0.9936453814862403, "300–1000 Hz_40ms_modulation_mae_db": 0.20020008008980428, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8345112160676178, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21940512658046635, "1000–3000 Hz_10ms_envelope_correlation": 0.9616482048123952, "1000–3000 Hz_10ms_modulation_mae_db": 0.44826918541752897, "1000–3000 Hz_40ms_log_rms_mae_db": 1.045164971642243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09849211403213487, "1000–3000 Hz_40ms_envelope_correlation": 0.991842320381619, "1000–3000 Hz_40ms_modulation_mae_db": 0.7284142913449848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7443484757985899, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21631775206994205, "3000–6000 Hz_10ms_envelope_correlation": 0.969457303512437, "3000–6000 Hz_10ms_modulation_mae_db": 0.10350878353514155, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3159630802571443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15733909803095297, "3000–6000 Hz_40ms_envelope_correlation": 0.9831592737342082, "3000–6000 Hz_40ms_modulation_mae_db": 0.1858277660262431, "mr_stft_db": 6.010925327515539, "band_mae_db": 0.11044995498860392, "rms_gain_error_db": -0.11321418666582944, "output_peak": 0.5107387227715735}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3814804610781724, "全帯域_10ms_rms_envelope_nrmse": 0.138998127766629, "全帯域_10ms_envelope_correlation": 0.9789674874503005, "全帯域_10ms_modulation_mae_db": 0.294728514385984, "全帯域_40ms_log_rms_mae_db": 0.6555791010337987, "全帯域_40ms_rms_envelope_nrmse": 0.06880281650746609, "全帯域_40ms_envelope_correlation": 0.9944734281686816, "全帯域_40ms_modulation_mae_db": 0.26794174141860316, "300–1000 Hz_10ms_log_rms_mae_db": 2.046892192798663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17362204185695504, "300–1000 Hz_10ms_envelope_correlation": 0.9775769082142206, "300–1000 Hz_10ms_modulation_mae_db": 0.8066378001278277, "300–1000 Hz_40ms_log_rms_mae_db": 1.2720857915772494, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09120398892078022, "300–1000 Hz_40ms_envelope_correlation": 0.9942697389556164, "300–1000 Hz_40ms_modulation_mae_db": 0.7187169276671401, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8556348141834054, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24231369489289634, "1000–3000 Hz_10ms_envelope_correlation": 0.9534181092020404, "1000–3000 Hz_10ms_modulation_mae_db": 0.7714392055114483, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0587318085593065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1122949607653081, "1000–3000 Hz_40ms_envelope_correlation": 0.9893186584622742, "1000–3000 Hz_40ms_modulation_mae_db": 0.6453132705437474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8035922132155435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22736868785691045, "3000–6000 Hz_10ms_envelope_correlation": 0.9671600460614491, "3000–6000 Hz_10ms_modulation_mae_db": 0.44028013920288367, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3483979951533367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1648014169327651, "3000–6000 Hz_40ms_envelope_correlation": 0.9821914900486971, "3000–6000 Hz_40ms_modulation_mae_db": 0.254865213192969, "mr_stft_db": 6.028695931968193, "band_mae_db": 0.30918950723816524, "rms_gain_error_db": -0.09790650074515, "output_peak": 0.5536033863178674}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4394082027423345, "全帯域_10ms_rms_envelope_nrmse": 0.13669783207740338, "全帯域_10ms_envelope_correlation": 0.9800372857386461, "全帯域_10ms_modulation_mae_db": 0.3096946792473428, "全帯域_40ms_log_rms_mae_db": 0.7150458921644853, "全帯域_40ms_rms_envelope_nrmse": 0.0627872947949436, "全帯域_40ms_envelope_correlation": 0.9955170782633285, "全帯域_40ms_modulation_mae_db": 0.4263531869158257, "300–1000 Hz_10ms_log_rms_mae_db": 2.09871351789056, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17138655471743952, "300–1000 Hz_10ms_envelope_correlation": 0.9778425943389726, "300–1000 Hz_10ms_modulation_mae_db": 0.5275611910804784, "300–1000 Hz_40ms_log_rms_mae_db": 1.3282403575782578, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08809218143138106, "300–1000 Hz_40ms_envelope_correlation": 0.993817828387461, "300–1000 Hz_40ms_modulation_mae_db": 0.3570290913680933, "1000–3000 Hz_10ms_log_rms_mae_db": 1.865392082835448, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22802833357730104, "1000–3000 Hz_10ms_envelope_correlation": 0.9588264961090002, "1000–3000 Hz_10ms_modulation_mae_db": 0.2381058508209537, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0679662013618154, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11330761917476782, "1000–3000 Hz_40ms_envelope_correlation": 0.9891472744716745, "1000–3000 Hz_40ms_modulation_mae_db": 0.37384695142918045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8028438410444187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2145507170382665, "3000–6000 Hz_10ms_envelope_correlation": 0.9686495314623866, "3000–6000 Hz_10ms_modulation_mae_db": 0.45682663706727994, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3609139152138936, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16129474472634886, "3000–6000 Hz_40ms_envelope_correlation": 0.9817633596492185, "3000–6000 Hz_40ms_modulation_mae_db": 0.287034280420708, "mr_stft_db": 6.016568708983308, "band_mae_db": 0.11735178816401608, "rms_gain_error_db": 0.02858075086420282, "output_peak": 0.5455077624946231}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3979052045643423, "全帯域_10ms_rms_envelope_nrmse": 0.13706275068220994, "全帯域_10ms_envelope_correlation": 0.9795375576496087, "全帯域_10ms_modulation_mae_db": 0.2815738603301412, "全帯域_40ms_log_rms_mae_db": 0.6872853291626659, "全帯域_40ms_rms_envelope_nrmse": 0.06756836928761495, "全帯域_40ms_envelope_correlation": 0.9947130363704615, "全帯域_40ms_modulation_mae_db": 0.2219264595219217, "300–1000 Hz_10ms_log_rms_mae_db": 2.1607348265683335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18793569236504107, "300–1000 Hz_10ms_envelope_correlation": 0.9733539167808121, "300–1000 Hz_10ms_modulation_mae_db": 0.7183818922806271, "300–1000 Hz_40ms_log_rms_mae_db": 1.3437308371610195, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10226994415359326, "300–1000 Hz_40ms_envelope_correlation": 0.9921015291151627, "300–1000 Hz_40ms_modulation_mae_db": 0.5580039436824824, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8626036101094599, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24039345391736197, "1000–3000 Hz_10ms_envelope_correlation": 0.9543944319142041, "1000–3000 Hz_10ms_modulation_mae_db": 0.41306361401689123, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0644239004526919, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12028129398420384, "1000–3000 Hz_40ms_envelope_correlation": 0.9878689641702866, "1000–3000 Hz_40ms_modulation_mae_db": 0.19268601714999659, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7881710304949512, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22200020314030264, "3000–6000 Hz_10ms_envelope_correlation": 0.968690961168079, "3000–6000 Hz_10ms_modulation_mae_db": 0.2843295512487814, "3000–6000 Hz_40ms_log_rms_mae_db": 1.323910181193047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1620932056990699, "3000–6000 Hz_40ms_envelope_correlation": 0.9827866870386159, "3000–6000 Hz_40ms_modulation_mae_db": 0.17347150213390497, "mr_stft_db": 6.014595529822645, "band_mae_db": 0.17457730261957505, "rms_gain_error_db": -0.11316750266968016, "output_peak": 0.5637801927369221}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4450571912921621, "全帯域_10ms_rms_envelope_nrmse": 0.14641548528767687, "全帯域_10ms_envelope_correlation": 0.9767297849687259, "全帯域_10ms_modulation_mae_db": 0.49605644626566137, "全帯域_40ms_log_rms_mae_db": 0.6871558961278725, "全帯域_40ms_rms_envelope_nrmse": 0.07213069055301755, "全帯域_40ms_envelope_correlation": 0.9941121220994114, "全帯域_40ms_modulation_mae_db": 0.5421949799788698, "300–1000 Hz_10ms_log_rms_mae_db": 2.1206759101339596, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19047992826716073, "300–1000 Hz_10ms_envelope_correlation": 0.9725198626284897, "300–1000 Hz_10ms_modulation_mae_db": 0.6126930887137249, "300–1000 Hz_40ms_log_rms_mae_db": 1.3108321212258305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09269593177049454, "300–1000 Hz_40ms_envelope_correlation": 0.9934479559174032, "300–1000 Hz_40ms_modulation_mae_db": 0.2911124557127797, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8616475796078604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21405673295111977, "1000–3000 Hz_10ms_envelope_correlation": 0.9633921447526181, "1000–3000 Hz_10ms_modulation_mae_db": 0.7266345138766646, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9767279247653469, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0939160130640464, "1000–3000 Hz_40ms_envelope_correlation": 0.9927824597548104, "1000–3000 Hz_40ms_modulation_mae_db": 0.6637219180000977, "3000–6000 Hz_10ms_log_rms_mae_db": 1.777630330583529, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22456394755917528, "3000–6000 Hz_10ms_envelope_correlation": 0.9658812845967768, "3000–6000 Hz_10ms_modulation_mae_db": 0.5809192788814022, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3403432826834727, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16661598363342675, "3000–6000 Hz_40ms_envelope_correlation": 0.9802471750531724, "3000–6000 Hz_40ms_modulation_mae_db": 0.18324385424640197, "mr_stft_db": 6.014547078018668, "band_mae_db": 0.19348794802803404, "rms_gain_error_db": -0.15123760040879328, "output_peak": 0.5606002573106934}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.286023630222864, "spectral_convergence": 0.4999325724234788, "spectral_convergence_status": "ok", "amplitude_floor": 1.4385235391470026e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.738307357791737, "spectral_convergence": 0.42176748572246353, "spectral_convergence_status": "ok", "amplitude_floor": 9.243439915833013e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.058163451992235, "spectral_convergence": 0.47511765135007344, "spectral_convergence_status": "ok", "amplitude_floor": 6.315989368278398e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00029756845932849967, "output_power": 0.0003141918097439198, "delta_db": 0.2360796338391884, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006886429842704082, "output_power": 0.0006470042630236335, "delta_db": -0.2708698517043378, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003825307900717409, "output_power": 0.0003850570768294959, "delta_db": 0.02858712096398315, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019304685978074288, "output_power": 0.00017411375101804034, "delta_db": -0.4482966967794639, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010892090215881134, "output_power": 0.00011261891146191309, "delta_db": 0.14500095462488924, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.08275993922433e-05, "output_power": 7.937394342085587e-05, "delta_db": -0.07881723026860663, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20127524803007815, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09824451115140424, "pitch": null, "output_rms_dbfs": -27.6640498766475, "output_peak": 0.5660082994486727, "reference_rms_dbfs": -27.565805365496097, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.027498146668946}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017313921740261034, "log_rms_mae_db": 1.3974611162711725, "rms_envelope_nrmse": 0.14576181456118642, "envelope_correlation": 0.9768232008585833, "rms_gain_error_db": -0.09324189763339687, "reference_mean_power": 0.0017487207983142375, "output_mean_power": 0.0017115763840865922}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.577112847045013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002769005691966373, "output_power": 0.00026125178598871074, "reference_active": true, "delta_db": -0.25264580396759473}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.377497445156709e-05, "output_power": 3.563727944981882e-05, "reference_active": true, "delta_db": 0.23309513405803658}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4953137492438862e-05, "output_power": 1.7404538539629766e-05, "reference_active": true, "delta_db": 0.6593018635821907}], "active_band_count": 3, "active_band_mae_db": 0.381680933869274}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015508669364693045, "log_rms_mae_db": 0.6401452787711608, "rms_envelope_nrmse": 0.0647094558796994, "envelope_correlation": 0.9951671696589047, "rms_gain_error_db": -0.09834359774599345, "reference_mean_power": 0.0017487025727625826, "output_mean_power": 0.0017095491414223479}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.09499315514524e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025296787730251264, "output_power": 0.00023963813451488523, "reference_active": true, "delta_db": -0.23509446401561518}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6687984786311574e-05, "output_power": 1.7947710919361556e-05, "reference_active": true, "delta_db": 0.3160517065635865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.96993861330795e-07, "output_power": 9.06169597989814e-07, "reference_active": true, "delta_db": 1.1398053436689863}], "active_band_count": 3, "active_band_mae_db": 0.563650504749396}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014635391485956695, "log_rms_mae_db": 2.149873233057337, "rms_envelope_nrmse": 0.18805678866575593, "envelope_correlation": 0.9732171555608057, "rms_gain_error_db": -0.1522403800291711, "reference_mean_power": 0.0008142561216173605, "output_mean_power": 0.0007862071593939257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2354521678013805e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001985819153870851, "output_power": 0.00018234066807111643, "reference_active": true, "delta_db": -0.37056153629490596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.365393218996536e-05, "output_power": 2.180241612680411e-05, "reference_active": true, "delta_db": -0.35398723012369815}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.646479803369054e-06, "output_power": 7.6289528941145745e-06, "reference_active": true, "delta_db": 0.5987324389980339}], "active_band_count": 3, "active_band_mae_db": 0.44109373513887934}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012544182666407365, "log_rms_mae_db": 1.334552705732181, "rms_envelope_nrmse": 0.09343499384598818, "envelope_correlation": 0.993275502026902, "rms_gain_error_db": -0.1632860666932473, "reference_mean_power": 0.0008126713020357493, "output_mean_power": 0.0007826837498604841}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0135181965386276e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018492598004057708, "output_power": 0.0001706834800386038, "reference_active": true, "delta_db": -0.3480643994381006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0725232186915858e-05, "output_power": 9.899493704675111e-06, "reference_active": true, "delta_db": -0.3479371905181477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.105637817013119e-07, "output_power": 4.5737911084444964e-07, "reference_active": true, "delta_db": 0.46895690878045115}], "active_band_count": 3, "active_band_mae_db": 0.38831949957889983}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.111876173726916e-05, "log_rms_mae_db": 1.8741056452004519, "rms_envelope_nrmse": 0.2330704621650075, "envelope_correlation": 0.9565542470998942, "rms_gain_error_db": -0.34559614267826566, "reference_mean_power": 0.0002425197498003479, "output_mean_power": 0.000223968786665933}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1105776313023488e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8957411226134796e-05, "output_power": 3.382891177099492e-05, "reference_active": true, "delta_db": -0.6130206233263817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.645635870343935e-06, "output_power": 6.204165191054947e-06, "reference_active": true, "delta_db": 0.40970488960967466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.039589927108413e-06, "output_power": 3.958189865992311e-06, "reference_active": true, "delta_db": -0.08840658040108725}], "active_band_count": 3, "active_band_mae_db": 0.3703773644457145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.178958171420995e-05, "log_rms_mae_db": 1.014584780221927, "rms_envelope_nrmse": 0.10554729873812223, "envelope_correlation": 0.991602529085704, "rms_gain_error_db": -0.34978563731476864, "reference_mean_power": 0.00024219152470715313, "output_mean_power": 0.00022345000955323614}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0514876259433796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7077906526859964e-05, "output_power": 3.250643714952583e-05, "reference_active": true, "delta_db": -0.571458337668276}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.355351339185207e-06, "output_power": 3.31901910124209e-06, "reference_active": true, "delta_db": -0.047282500879556384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8063796091292985e-07, "output_power": 2.338840150662647e-07, "reference_active": true, "delta_db": -0.7914587543326389}], "active_band_count": 3, "active_band_mae_db": 0.4700665309601571}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3400740982798506e-05, "log_rms_mae_db": 1.821161597721064, "rms_envelope_nrmse": 0.2165726970135195, "envelope_correlation": 0.9683653642785744, "rms_gain_error_db": 0.01050456986615959, "reference_mean_power": 0.0001023804664102342, "output_mean_power": 0.00010262840058927535}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5067174911166034e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7037143173191167e-05, "output_power": 2.6000479246232764e-05, "reference_active": true, "delta_db": -0.1697944776682694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.303852533122454e-06, "output_power": 4.597833126738937e-06, "reference_active": true, "delta_db": 0.2869582270207233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8801329619345497e-06, "output_power": 1.99689935273834e-06, "reference_active": true, "delta_db": 0.2616761280309461}], "active_band_count": 3, "active_band_mae_db": 0.23947627757331294}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.152733550445532e-05, "log_rms_mae_db": 1.373185725240691, "rms_envelope_nrmse": 0.16351015204825345, "envelope_correlation": 0.9811022055291778, "rms_gain_error_db": 0.010422889664217606, "reference_mean_power": 0.0001028923086971688, "output_mean_power": 0.00010313954258015853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.218443109204273e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6964722520221083e-05, "output_power": 2.582943712844206e-05, "reference_active": true, "delta_db": -0.18681013360514934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9782668045768803e-06, "output_power": 2.2499427828282447e-06, "reference_active": true, "delta_db": 0.5588661035432293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5223765481589483e-07, "output_power": 1.4385190377072523e-07, "reference_active": true, "delta_db": -0.2460647130735577}], "active_band_count": 3, "active_band_mae_db": 0.3305803167406454}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7647906568835959, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7c32ba7f57c00bee4ffd4b377b265fcd69362ceef7a824871a425b6ab973a9d2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.959053393630026, "standard_deviation": 0.004005276695039897, "minimum": 0.9534181092020404, "maximum": 0.9641527469389939, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8491464357539784, "standard_deviation": 0.02173580881601214, "minimum": 1.8131149320233795, "maximum": 1.8741056452004519, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.486956594091032, "standard_deviation": 0.18622689303213985, "minimum": 0.2381058508209537, "maximum": 0.7714392055114483, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2266726475839695, "standard_deviation": 0.011409685667679716, "minimum": 0.21192700586226368, "maximum": 0.24231369489289634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9905201089074887, "standard_deviation": 0.0016314328502482595, "minimum": 0.9878689641702866, "maximum": 0.9927824597548104, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.037491127406702, "standard_deviation": 0.030264074715279896, "minimum": 0.9767279247653469, "maximum": 1.0679662013618154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6144102753364733, "standard_deviation": 0.3404257730375143, "minimum": 0.19268601714999659, "maximum": 1.3394021054640521, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1066267877636019, "standard_deviation": 0.00848980713363572, "minimum": 0.0939160130640464, "maximum": 0.12028129398420384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9681444868126283, "standard_deviation": 0.0013862881463326687, "minimum": 0.9658812845967768, "maximum": 0.9700274701262424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7881964836843778, "standard_deviation": 0.0250875564582675, "minimum": 1.7443484757985899, "maximum": 1.821161597721064, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.36745505163827913, "standard_deviation": 0.1507353568199009, "minimum": 0.10350878353514155, "maximum": 0.5809192788814022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21929198441738304, "standard_deviation": 0.005263040133718429, "minimum": 0.21206438886663032, "maximum": 0.22736868785691045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9818431849842084, "standard_deviation": 0.0009251596355347104, "minimum": 0.9802471750531724, "maximum": 0.9831592737342082, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3423232501408364, "standard_deviation": 0.020238944740587635, "minimum": 1.3159630802571443, "maximum": 1.373185725240691, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.24179157419221592, "standard_deviation": 0.058387925750206654, "minimum": 0.17347150213390497, "maximum": 0.3305803167406454, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1621644100185348, "standard_deviation": 0.0028446647670347805, "minimum": 0.15733909803095297, "maximum": 0.16661598363342675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9741265231861707, "standard_deviation": 0.002402532313841782, "minimum": 0.9709031930278058, "maximum": 0.9778425943389726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.130076672635412, "standard_deviation": 0.04423896884807683, "minimum": 2.046892192798663, "maximum": 2.1960073905095663, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5335092572978732, "standard_deviation": 0.1948208752744995, "minimum": 0.21348610277566146, "maximum": 0.8066378001278277, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18498682115306098, "standard_deviation": 0.008390852602435215, "minimum": 0.17138655471743952, "maximum": 0.1962466633910721, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9933309511988448, "standard_deviation": 0.0006964624193466301, "minimum": 0.9921015291151627, "maximum": 0.9942697389556164, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.337506765569133, "standard_deviation": 0.03502516841557999, "minimum": 1.2720857915772494, "maximum": 1.3832410191548108, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.41650724079015816, "standard_deviation": 0.168330157102203, "minimum": 0.20020008008980428, "maximum": 0.7187169276671401, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09348847305746302, "standard_deviation": 0.004256007942332943, "minimum": 0.08809218143138106, "maximum": 0.10226994415359326, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17853130400035755, "standard_deviation": 0.06194099667621418, "minimum": 0.11044995498860392, "maximum": 0.30918950723816524, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.017891399301868, "standard_deviation": 0.00664596960810748, "minimum": 6.010925327515539, "maximum": 6.028695931968193, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5505842298009481, "standard_deviation": 0.021323663784717596, "minimum": 0.5107387227715735, "maximum": 0.5750501313145686, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06920142671511656, "standard_deviation": 0.06653306193214932, "minimum": -0.15123760040879328, "maximum": 0.02858075086420282, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9784131390170528, "standard_deviation": 0.0013088114168140397, "minimum": 0.9767297849687259, "maximum": 0.9800372857386461, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4041790720803706, "standard_deviation": 0.03714864773943183, "minimum": 1.344329816710561, "maximum": 1.4483741218172959, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34034485257702407, "standard_deviation": 0.1418141555563816, "minimum": 0.13907748961663097, "maximum": 0.580479847015375, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14103129357936056, "standard_deviation": 0.004139546506772098, "minimum": 0.13667636224444435, "maximum": 0.14641548528767687, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9950097894654466, "standard_deviation": 0.0005634904180213366, "minimum": 0.9941121220994114, "maximum": 0.9958665151343886, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6670778144322794, "standard_deviation": 0.03754816877696603, "minimum": 0.5923870557437254, "maximum": 0.7150458921644853, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33395139386789074, "standard_deviation": 0.15797445479281305, "minimum": 0.1383091522405334, "maximum": 0.563650504749396, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06560304594390706, "standard_deviation": 0.003757132872802794, "minimum": 0.06010595423534963, "maximum": 0.07213069055301755, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_61-70968-0054", "language": "en", "style": "normal", "speaker": "librispeech_61", "source_utterance_id": "61-70968-0054", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_61-70968-0054.wav", "source_file_sha256": "d28a6714446c34fbec58a8cc1ee0a6722edfb7cc6e258ce8f93c981fb2774aed", "decoded_pcm_sha256": "0c30dfc3b23081f52641ba6b1793c8bc5c9ea236aea529b533de133f00169e95", "original_source_sha256": "7b58da65dd63a6b09d43ee20c6aa4b5a8853199af7c350a87ab692fb4d0d874c", "corpus_entry_id": "librispeech_61-70968-0054"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.157610893249512, "predicted_mos": 4.157610893249512, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5915f0a66f2b1aa7ffa48d1b641ac26c677d0e3db4129aa851a7c04743878710", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.07169669400173, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_61-70968-0054", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5915f0a66f2b1aa7ffa48d1b641ac26c677d0e3db4129aa851a7c04743878710", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "e07ca663384d7b0ea8ac49ce7905c02e33e1c111573b428d6692890acf3f6ecb"}, "rows": [{"case": "librispeech_61-70968-0054", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.1780574321746826, "predicted_mos": 2.1780574321746826, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "894c700aab8d6e82791190efdd2db81688bfa747456094cfa3dced961a4b2805", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1952321930002654, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_61-70968-0054", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "894c700aab8d6e82791190efdd2db81688bfa747456094cfa3dced961a4b2805", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2168975693427384, "全帯域_10ms_rms_envelope_nrmse": 0.11413996827001506, "全帯域_10ms_envelope_correlation": 0.9808761379763448, "全帯域_10ms_modulation_mae_db": 0.22160177824544577, "全帯域_40ms_log_rms_mae_db": 0.5829950159391767, "全帯域_40ms_rms_envelope_nrmse": 0.05581080383712584, "全帯域_40ms_envelope_correlation": 0.9944870516402433, "全帯域_40ms_modulation_mae_db": 0.15989800978803018, "300–1000 Hz_10ms_log_rms_mae_db": 2.047373137453323, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15210395121438886, "300–1000 Hz_10ms_envelope_correlation": 0.97880340578762, "300–1000 Hz_10ms_modulation_mae_db": 0.7219790008232273, "300–1000 Hz_40ms_log_rms_mae_db": 1.2315291142089937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08012319880632976, "300–1000 Hz_40ms_envelope_correlation": 0.9933008381613281, "300–1000 Hz_40ms_modulation_mae_db": 0.28256083106569313, "1000–3000 Hz_10ms_log_rms_mae_db": 2.072743830098464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2424312777035911, "1000–3000 Hz_10ms_envelope_correlation": 0.9427291723510844, "1000–3000 Hz_10ms_modulation_mae_db": 0.6041715341571692, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2599485253110299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14736599733560204, "1000–3000 Hz_40ms_envelope_correlation": 0.975845153059082, "1000–3000 Hz_40ms_modulation_mae_db": 0.9311242361643416, "3000–6000 Hz_10ms_log_rms_mae_db": 2.279147012393972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20478169378666083, "3000–6000 Hz_10ms_envelope_correlation": 0.9727451885521443, "3000–6000 Hz_10ms_modulation_mae_db": 0.45256028477735527, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6766155241560599, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13841876507261422, "3000–6000 Hz_40ms_envelope_correlation": 0.9873031381162095, "3000–6000 Hz_40ms_modulation_mae_db": 0.4098979819546919, "mr_stft_db": 5.969903747701622, "band_mae_db": 0.09059527958446463, "rms_gain_error_db": -0.03425377386460978, "output_peak": 0.43589706197216144}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.287995280216142, "全帯域_10ms_rms_envelope_nrmse": 0.12077502146651362, "全帯域_10ms_envelope_correlation": 0.9785690970513872, "全帯域_10ms_modulation_mae_db": 0.2842808365479237, "全帯域_40ms_log_rms_mae_db": 0.5924123115453108, "全帯域_40ms_rms_envelope_nrmse": 0.05590902923970845, "全帯域_40ms_envelope_correlation": 0.994472345736726, "全帯域_40ms_modulation_mae_db": 0.23277300407831536, "300–1000 Hz_10ms_log_rms_mae_db": 2.0958368022532463, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16678159479984084, "300–1000 Hz_10ms_envelope_correlation": 0.9744492178770663, "300–1000 Hz_10ms_modulation_mae_db": 0.9602552146830291, "300–1000 Hz_40ms_log_rms_mae_db": 1.3169033723814192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09879242686336845, "300–1000 Hz_40ms_envelope_correlation": 0.9899376389503007, "300–1000 Hz_40ms_modulation_mae_db": 0.7221847695246377, "1000–3000 Hz_10ms_log_rms_mae_db": 2.022514857510623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24021584529808512, "1000–3000 Hz_10ms_envelope_correlation": 0.943556863891435, "1000–3000 Hz_10ms_modulation_mae_db": 0.4835821389883049, "1000–3000 Hz_40ms_log_rms_mae_db": 1.220877649648655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14119029294760574, "1000–3000 Hz_40ms_envelope_correlation": 0.977938958990347, "1000–3000 Hz_40ms_modulation_mae_db": 0.42123795720800494, "3000–6000 Hz_10ms_log_rms_mae_db": 2.261898602314151, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22072809356903952, "3000–6000 Hz_10ms_envelope_correlation": 0.9681581906152619, "3000–6000 Hz_10ms_modulation_mae_db": 0.4126285951170983, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6809152716316442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15294619821370745, "3000–6000 Hz_40ms_envelope_correlation": 0.9842960326377942, "3000–6000 Hz_40ms_modulation_mae_db": 0.44771417594873225, "mr_stft_db": 5.960004881146975, "band_mae_db": 0.17530279848629682, "rms_gain_error_db": -0.04571892347731459, "output_peak": 0.40979678770337674}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2384222409713728, "全帯域_10ms_rms_envelope_nrmse": 0.11062733298954557, "全帯域_10ms_envelope_correlation": 0.9819247389635939, "全帯域_10ms_modulation_mae_db": 0.36994261023069214, "全帯域_40ms_log_rms_mae_db": 0.5672823839141349, "全帯域_40ms_rms_envelope_nrmse": 0.05537157557988606, "全帯域_40ms_envelope_correlation": 0.9945594073615341, "全帯域_40ms_modulation_mae_db": 0.420916751105997, "300–1000 Hz_10ms_log_rms_mae_db": 2.0697069027152, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1435350134677409, "300–1000 Hz_10ms_envelope_correlation": 0.9813548042321142, "300–1000 Hz_10ms_modulation_mae_db": 0.4650840137848357, "300–1000 Hz_40ms_log_rms_mae_db": 1.2769538071111326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08310966387618886, "300–1000 Hz_40ms_envelope_correlation": 0.9932766975362997, "300–1000 Hz_40ms_modulation_mae_db": 0.354346670510721, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9516943304761578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21527913839709284, "1000–3000 Hz_10ms_envelope_correlation": 0.9548378803724505, "1000–3000 Hz_10ms_modulation_mae_db": 0.4575824545045073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1463810220648407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12503778098321547, "1000–3000 Hz_40ms_envelope_correlation": 0.9830191762134557, "1000–3000 Hz_40ms_modulation_mae_db": 0.7215463310184386, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3171821738812635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2079913027571745, "3000–6000 Hz_10ms_envelope_correlation": 0.9723015832444982, "3000–6000 Hz_10ms_modulation_mae_db": 0.21466867945678725, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7359297115885663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13909376986773506, "3000–6000 Hz_40ms_envelope_correlation": 0.9872017086708927, "3000–6000 Hz_40ms_modulation_mae_db": 0.3095652390262153, "mr_stft_db": 5.9442756311436264, "band_mae_db": 0.18039561690162298, "rms_gain_error_db": -0.053503360049927534, "output_peak": 0.39191835054560836}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.217058929993732, "全帯域_10ms_rms_envelope_nrmse": 0.11791119066404938, "全帯域_10ms_envelope_correlation": 0.9795183193746226, "全帯域_10ms_modulation_mae_db": 0.3261652297404085, "全帯域_40ms_log_rms_mae_db": 0.5601400064011247, "全帯域_40ms_rms_envelope_nrmse": 0.06074615872762748, "全帯域_40ms_envelope_correlation": 0.9935889297823391, "全帯域_40ms_modulation_mae_db": 0.21861227219231086, "300–1000 Hz_10ms_log_rms_mae_db": 2.1158576160921996, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.169203802627274, "300–1000 Hz_10ms_envelope_correlation": 0.9737520745277259, "300–1000 Hz_10ms_modulation_mae_db": 0.8128979215826807, "300–1000 Hz_40ms_log_rms_mae_db": 1.3223389856322834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09604071823945425, "300–1000 Hz_40ms_envelope_correlation": 0.9906968438278287, "300–1000 Hz_40ms_modulation_mae_db": 0.3554168415052564, "1000–3000 Hz_10ms_log_rms_mae_db": 2.011488333517074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23210721028460476, "1000–3000 Hz_10ms_envelope_correlation": 0.9478520097321754, "1000–3000 Hz_10ms_modulation_mae_db": 0.40565958608683367, "1000–3000 Hz_40ms_log_rms_mae_db": 1.236903331896569, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13826242572566308, "1000–3000 Hz_40ms_envelope_correlation": 0.9787945343804757, "1000–3000 Hz_40ms_modulation_mae_db": 0.7347344696814995, "3000–6000 Hz_10ms_log_rms_mae_db": 2.248526518778577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2081112938319747, "3000–6000 Hz_10ms_envelope_correlation": 0.9718440436154908, "3000–6000 Hz_10ms_modulation_mae_db": 0.67319064762494, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6933577060474583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13809562641549833, "3000–6000 Hz_40ms_envelope_correlation": 0.9871002231269773, "3000–6000 Hz_40ms_modulation_mae_db": 0.7301281429068122, "mr_stft_db": 5.948941339820682, "band_mae_db": 0.08428158065511222, "rms_gain_error_db": -0.10622885874423359, "output_peak": 0.4097918788163948}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2676277882502616, "全帯域_10ms_rms_envelope_nrmse": 0.11391624715404164, "全帯域_10ms_envelope_correlation": 0.9808147945315642, "全帯域_10ms_modulation_mae_db": 0.3973877263654943, "全帯域_40ms_log_rms_mae_db": 0.5752612280836455, "全帯域_40ms_rms_envelope_nrmse": 0.05842774974592677, "全帯域_40ms_envelope_correlation": 0.9940020180065586, "全帯域_40ms_modulation_mae_db": 0.29617421405175753, "300–1000 Hz_10ms_log_rms_mae_db": 2.061960121504162, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16093818439338664, "300–1000 Hz_10ms_envelope_correlation": 0.9762452544698785, "300–1000 Hz_10ms_modulation_mae_db": 0.7876999155733766, "300–1000 Hz_40ms_log_rms_mae_db": 1.2974047958885129, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08891616524225279, "300–1000 Hz_40ms_envelope_correlation": 0.9919424958043427, "300–1000 Hz_40ms_modulation_mae_db": 0.33880850116645567, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9830754995636577, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21685617121095005, "1000–3000 Hz_10ms_envelope_correlation": 0.9543937254128103, "1000–3000 Hz_10ms_modulation_mae_db": 0.3253615269953975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2079953753363595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12881855212552384, "1000–3000 Hz_40ms_envelope_correlation": 0.9816709702897378, "1000–3000 Hz_40ms_modulation_mae_db": 0.6084037668445839, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2801664230159218, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2173516384996381, "3000–6000 Hz_10ms_envelope_correlation": 0.9717390192396718, "3000–6000 Hz_10ms_modulation_mae_db": 0.43361798750120145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7348789262999167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15346157751396552, "3000–6000 Hz_40ms_envelope_correlation": 0.9857434020510899, "3000–6000 Hz_40ms_modulation_mae_db": 0.6678691080019511, "mr_stft_db": 5.951026971742574, "band_mae_db": 0.16900516324554293, "rms_gain_error_db": -0.06620723296458401, "output_peak": 0.468462759017482}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2676287160715212, "全帯域_10ms_rms_envelope_nrmse": 0.11729630209092885, "全帯域_10ms_envelope_correlation": 0.9796654394608609, "全帯域_10ms_modulation_mae_db": 0.1567872136257697, "全帯域_40ms_log_rms_mae_db": 0.5796313260012723, "全帯域_40ms_rms_envelope_nrmse": 0.05494830413259869, "全帯域_40ms_envelope_correlation": 0.9947030336607209, "全帯域_40ms_modulation_mae_db": 0.22022875253416865, "300–1000 Hz_10ms_log_rms_mae_db": 2.0748474432363526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16183480060571542, "300–1000 Hz_10ms_envelope_correlation": 0.9759574361085708, "300–1000 Hz_10ms_modulation_mae_db": 0.5976858751999591, "300–1000 Hz_40ms_log_rms_mae_db": 1.2994195446993702, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09067890581945343, "300–1000 Hz_40ms_envelope_correlation": 0.9915148586177563, "300–1000 Hz_40ms_modulation_mae_db": 0.5306110739586568, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9951297302890398, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22750116268392198, "1000–3000 Hz_10ms_envelope_correlation": 0.9495026281517469, "1000–3000 Hz_10ms_modulation_mae_db": 0.7530895851874965, "1000–3000 Hz_40ms_log_rms_mae_db": 1.259252770827032, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13877468005783147, "1000–3000 Hz_40ms_envelope_correlation": 0.9795409322253525, "1000–3000 Hz_40ms_modulation_mae_db": 1.02485383028918, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2917346571710695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2118817975561488, "3000–6000 Hz_10ms_envelope_correlation": 0.9714053521755658, "3000–6000 Hz_10ms_modulation_mae_db": 0.3192996348398143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7122222061645171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14225842840223218, "3000–6000 Hz_40ms_envelope_correlation": 0.9866036999682716, "3000–6000 Hz_40ms_modulation_mae_db": 0.4226447839280345, "mr_stft_db": 5.9679567406163185, "band_mae_db": 0.1919975565671893, "rms_gain_error_db": -0.0756079860837448, "output_peak": 0.4172507536906317}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2908643131380588, "全帯域_10ms_rms_envelope_nrmse": 0.11998571411654184, "全帯域_10ms_envelope_correlation": 0.9788610315807357, "全帯域_10ms_modulation_mae_db": 0.19580128939451316, "全帯域_40ms_log_rms_mae_db": 0.6546114885024665, "全帯域_40ms_rms_envelope_nrmse": 0.06459020636173654, "全帯域_40ms_envelope_correlation": 0.9926760239274935, "全帯域_40ms_modulation_mae_db": 0.21029709075210345, "300–1000 Hz_10ms_log_rms_mae_db": 2.1298544944894133, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16613844349935683, "300–1000 Hz_10ms_envelope_correlation": 0.9746852000731955, "300–1000 Hz_10ms_modulation_mae_db": 0.6732440777645475, "300–1000 Hz_40ms_log_rms_mae_db": 1.3799398781125762, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09628473424318405, "300–1000 Hz_40ms_envelope_correlation": 0.9902760053119193, "300–1000 Hz_40ms_modulation_mae_db": 0.49835957351694393, "1000–3000 Hz_10ms_log_rms_mae_db": 2.073161619905568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24208007641905413, "1000–3000 Hz_10ms_envelope_correlation": 0.9426769196517694, "1000–3000 Hz_10ms_modulation_mae_db": 0.5520390623629043, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3152425329217399, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1411038281466378, "1000–3000 Hz_40ms_envelope_correlation": 0.9779616618574004, "1000–3000 Hz_40ms_modulation_mae_db": 0.8469641718651966, "3000–6000 Hz_10ms_log_rms_mae_db": 2.280245418657717, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23301027595153584, "3000–6000 Hz_10ms_envelope_correlation": 0.9645262519251699, "3000–6000 Hz_10ms_modulation_mae_db": 0.7777880335547677, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7040630735129099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15653526023813924, "3000–6000 Hz_40ms_envelope_correlation": 0.9834997392717129, "3000–6000 Hz_40ms_modulation_mae_db": 0.821691076219427, "mr_stft_db": 5.971489189772399, "band_mae_db": 0.14135454138762046, "rms_gain_error_db": -0.07100341612186532, "output_peak": 0.3954426173074973}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2378507317066552, "全帯域_10ms_rms_envelope_nrmse": 0.11637161825452308, "全帯域_10ms_envelope_correlation": 0.9800399363591644, "全帯域_10ms_modulation_mae_db": 0.1424420566571671, "全帯域_40ms_log_rms_mae_db": 0.5854059229752263, "全帯域_40ms_rms_envelope_nrmse": 0.05756923652312311, "全帯域_40ms_envelope_correlation": 0.994119906392263, "全帯域_40ms_modulation_mae_db": 0.34399093252173557, "300–1000 Hz_10ms_log_rms_mae_db": 2.107617050514273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1527282270193391, "300–1000 Hz_10ms_envelope_correlation": 0.9786694342694122, "300–1000 Hz_10ms_modulation_mae_db": 0.4762802451952661, "300–1000 Hz_40ms_log_rms_mae_db": 1.3145806556426127, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08557240022203345, "300–1000 Hz_40ms_envelope_correlation": 0.9925902320950775, "300–1000 Hz_40ms_modulation_mae_db": 0.23970187311724053, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0513138903659724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2461570767323827, "1000–3000 Hz_10ms_envelope_correlation": 0.9406029127735714, "1000–3000 Hz_10ms_modulation_mae_db": 0.8283410902330176, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2766843267288344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1468449368157017, "1000–3000 Hz_40ms_envelope_correlation": 0.9761645671708867, "1000–3000 Hz_40ms_modulation_mae_db": 0.977819817947086, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2314502991605623, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2085078441993144, "3000–6000 Hz_10ms_envelope_correlation": 0.973478233731205, "3000–6000 Hz_10ms_modulation_mae_db": 0.3463082382239799, "3000–6000 Hz_40ms_log_rms_mae_db": 1.685554776264923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14159235378205706, "3000–6000 Hz_40ms_envelope_correlation": 0.98755554643218, "3000–6000 Hz_40ms_modulation_mae_db": 0.19214178516273459, "mr_stft_db": 5.9617001039601005, "band_mae_db": 0.17331866171763444, "rms_gain_error_db": -0.049940152862969316, "output_peak": 0.3907169020354549}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2630523458932945, "spectral_convergence": 0.4114852278843785, "spectral_convergence_status": "ok", "amplitude_floor": 1.3149451373049662e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.759099795850027, "spectral_convergence": 0.3555911110949848, "spectral_convergence_status": "ok", "amplitude_floor": 7.412465414213296e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.887559101361545, "spectral_convergence": 0.3899767622211072, "spectral_convergence_status": "ok", "amplitude_floor": 4.876012126794439e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006137697087039973, "output_power": 0.0006262381666336349, "delta_db": 0.08734081599092405, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009615845769051578, "output_power": 0.0009370127765596808, "delta_db": -0.11241976242629086, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000542394172591287, "output_power": 0.0005436196521135882, "delta_db": 0.009801334163728173, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013703842807085808, "output_power": 0.00013413516018034687, "delta_db": -0.09299736096397901, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.620351571877408e-05, "output_power": 8.403898124955429e-05, "delta_db": -0.11044199331278134, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011238415255857088, "output_power": 0.00010905561570544777, "delta_db": -0.13057041064908434, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09059527958446463, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03425377386460978, "pitch": null, "output_rms_dbfs": -26.136615207594048, "output_peak": 0.43589706197216144, "reference_rms_dbfs": -26.102361433729442, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.969903747701622}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013602681282863616, "log_rms_mae_db": 1.2168975693427384, "rms_envelope_nrmse": 0.11413996827001506, "envelope_correlation": 0.9808761379763448, "rms_gain_error_db": -0.033895339090007345, "reference_mean_power": 0.0024543453947075094, "output_mean_power": 0.0024352645460456426}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.153929939453048e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000440684439069062, "output_power": 0.0004330732649268219, "reference_active": true, "delta_db": -0.0756634139004449}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001235413948928314, "output_power": 0.00011448090198504801, "reference_active": true, "delta_db": -0.3307945806888069}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6241583124510384e-05, "output_power": 2.7849972200014673e-05, "reference_active": true, "delta_db": 0.2583473401470855}], "active_band_count": 3, "active_band_mae_db": 0.22160177824544577}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011635584990380617, "log_rms_mae_db": 0.5829950159391767, "rms_envelope_nrmse": 0.05581080383712584, "envelope_correlation": 0.9944870516402433, "rms_gain_error_db": -0.033369570290403085, "reference_mean_power": 0.0024645680451820984, "output_mean_power": 0.0024457037881187222}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.594353848464586e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003918219819660693, "output_power": 0.0003889802369105075, "reference_active": true, "delta_db": -0.0316126060110371}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.98714927012752e-05, "output_power": 5.520976147028642e-05, "reference_active": true, "delta_db": -0.3520421519664392}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.328965955891304e-06, "output_power": 1.3586818781660372e-06, "reference_active": true, "delta_db": 0.09603927138661428}], "active_band_count": 3, "active_band_mae_db": 0.15989800978803018}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010777091024361473, "log_rms_mae_db": 2.047373137453323, "rms_envelope_nrmse": 0.15210395121438886, "envelope_correlation": 0.97880340578762, "rms_gain_error_db": -0.06602506746070069, "reference_mean_power": 0.0010415487104655132, "output_mean_power": 0.0010258339756457314}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.64672159263247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003552137892650671, "output_power": 0.0003338359112528319, "reference_active": true, "delta_db": -0.26956763986400467}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.132330818484316e-05, "output_power": 6.363715194762747e-05, "reference_active": true, "delta_db": -0.4952074407262175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2237174874701015e-05, "output_power": 1.6896523572033604e-05, "reference_active": true, "delta_db": 1.4011619218794595}], "active_band_count": 3, "active_band_mae_db": 0.7219790008232273}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.988831023316543e-05, "log_rms_mae_db": 1.2315291142089937, "rms_envelope_nrmse": 0.08012319880632976, "envelope_correlation": 0.9933008381613281, "rms_gain_error_db": -0.06607024748518282, "reference_mean_power": 0.0010464471526710588, "output_mean_power": 0.0010306477888983095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.816907801707229e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003300382500563707, "output_power": 0.0003105237463872515, "reference_active": true, "delta_db": -0.26469458541941066}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.204099572239431e-05, "output_power": 2.938823666349428e-05, "reference_active": true, "delta_db": -0.37532475419432465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.619198216109287e-07, "output_power": 8.216759071621739e-07, "reference_active": true, "delta_db": -0.2076631535833442}], "active_band_count": 3, "active_band_mae_db": 0.28256083106569313}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5701896659278796e-05, "log_rms_mae_db": 2.072743830098464, "rms_envelope_nrmse": 0.2424312777035911, "envelope_correlation": 0.9427291723510844, "rms_gain_error_db": -0.08974294394847623, "reference_mean_power": 0.0001796467459408233, "output_mean_power": 0.00017597260395490862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2519806248605502e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7558987061199216e-05, "output_power": 4.459154762017928e-05, "reference_active": true, "delta_db": -0.2798005070222614}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1481328556544434e-05, "output_power": 9.11692695766774e-06, "reference_active": true, "delta_db": -1.001436697872558}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593622469144843e-06, "output_power": 8.488967760828679e-06, "reference_active": true, "delta_db": -0.5312773975766883}], "active_band_count": 3, "active_band_mae_db": 0.6041715341571692}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.308273569685666e-05, "log_rms_mae_db": 1.2599485253110299, "rms_envelope_nrmse": 0.14736599733560204, "envelope_correlation": 0.975845153059082, "rms_gain_error_db": -0.08979409059358928, "reference_mean_power": 0.0001804894756897876, "output_mean_power": 0.00017679601603329178}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.07434863630674e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.420432318446027e-05, "output_power": 4.210427911520685e-05, "reference_active": true, "delta_db": -0.21138509424616259}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8496256649417895e-06, "output_power": 4.277019849613678e-06, "reference_active": true, "delta_db": -1.3598680934876726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.792037562074947e-07, "output_power": 4.371388622386663e-07, "reference_active": true, "delta_db": -1.2221195207591895}], "active_band_count": 3, "active_band_mae_db": 0.9311242361643416}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.825215268339244e-05, "log_rms_mae_db": 2.279147012393972, "rms_envelope_nrmse": 0.20478169378666083, "envelope_correlation": 0.9727451885521443, "rms_gain_error_db": -0.15917669099711007, "reference_mean_power": 0.0001267678324533094, "output_mean_power": 0.00012220568099260883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.277972501807574e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1540328186660286e-05, "output_power": 4.582409023517051e-05, "reference_active": true, "delta_db": -0.5105332813446242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.007459575479445e-05, "output_power": 9.139878723546843e-06, "reference_active": true, "delta_db": -0.4228719586127619}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.777776179485975e-06, "output_power": 3.426168062449998e-06, "reference_active": true, "delta_db": -0.42427561437467975}], "active_band_count": 3, "active_band_mae_db": 0.45256028477735527}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.580061124327403e-05, "log_rms_mae_db": 1.6766155241560599, "rms_envelope_nrmse": 0.13841876507261422, "envelope_correlation": 0.9873031381162095, "rms_gain_error_db": -0.15932507677923924, "reference_mean_power": 0.0001273148740012305, "output_mean_power": 0.00012272884211788823}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.770926544449125e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2932423481689434e-05, "output_power": 4.6013868643105024e-05, "reference_active": true, "delta_db": -0.6083303001459797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.69146274773148e-06, "output_power": 4.408079588077468e-06, "reference_active": true, "delta_db": -0.27058845065961207}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.856201910676535e-07, "output_power": 2.634579970969747e-07, "reference_active": true, "delta_db": -0.350775195058484}], "active_band_count": 3, "active_band_mae_db": 0.4098979819546919}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7956083589440337, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "24417352feb748db4297982124afa771bb02344e52ef8d1e8236c4ab663c3c4a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9470190140421304, "standard_deviation": 0.00551352394365468, "minimum": 0.9406029127735714, "maximum": 0.9548378803724505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0201402614658197, "standard_deviation": 0.04361333069429007, "minimum": 1.9516943304761578, "maximum": 2.073161619905568, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5512283723144539, "standard_deviation": 0.17150373483767306, "minimum": 0.3253615269953975, "maximum": 0.8283410902330176, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23282849484121032, "standard_deviation": 0.01196438163570786, "minimum": 0.21527913839709284, "maximum": 0.2461570767323827, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9788669942733422, "standard_deviation": 0.002497937778067763, "minimum": 0.975845153059082, "maximum": 0.9830191762134557, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2404106918418827, "standard_deviation": 0.05062255956460634, "minimum": 1.1463810220648407, "maximum": 1.3152425329217399, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7833355726272913, "standard_deviation": 0.20351941835242543, "minimum": 0.42123795720800494, "maximum": 1.02485383028918, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13842481176722263, "standard_deviation": 0.007903976527974496, "minimum": 0.12503778098321547, "maximum": 0.14736599733560204, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9707747328873759, "standard_deviation": 0.002973301350312237, "minimum": 0.9645262519251699, "maximum": 0.973478233731205, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2737938881716544, "standard_deviation": 0.02642568887065557, "minimum": 2.2314502991605623, "maximum": 2.3171821738812635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.45375776263699297, "standard_deviation": 0.1859466277207601, "minimum": 0.21466867945678725, "maximum": 0.7777880335547677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21404549251893584, "standard_deviation": 0.009321892727565144, "minimum": 0.20478169378666083, "maximum": 0.23301027595153584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.986162936284391, "standard_deviation": 0.0015187027871182385, "minimum": 0.9834997392717129, "maximum": 0.98755554643218, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7029421494582495, "standard_deviation": 0.023200892046251087, "minimum": 1.6766155241560599, "maximum": 1.7359297115885663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5002065366435748, "standard_deviation": 0.21794556855694464, "minimum": 0.19214178516273459, "maximum": 0.821691076219427, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14530024743824363, "standard_deviation": 0.007672550799124527, "minimum": 0.13809562641549833, "maximum": 0.15653526023813924, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.976739603418198, "standard_deviation": 0.0026330737164127313, "minimum": 0.9737520745277259, "maximum": 0.9813548042321142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0878816960322713, "standard_deviation": 0.028810271103170405, "minimum": 2.047373137453323, "maximum": 2.1298544944894133, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6868907830758653, "standard_deviation": 0.17058653663457432, "minimum": 0.4650840137848357, "maximum": 0.9602552146830291, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15915800220338033, "standard_deviation": 0.008890424628642113, "minimum": 0.1435350134677409, "maximum": 0.169203802627274, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9916919512881066, "standard_deviation": 0.001313323704098331, "minimum": 0.9899376389503007, "maximum": 0.9933008381613281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3048837692096127, "standard_deviation": 0.04212801157600559, "minimum": 1.2315291142089937, "maximum": 1.3799398781125762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.41524876679570066, "standard_deviation": 0.15855617160255728, "minimum": 0.23970187311724053, "maximum": 0.7221847695246377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08993977666403313, "standard_deviation": 0.006756299717491916, "minimum": 0.08012319880632976, "maximum": 0.09879242686336845, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15078139981818547, "standard_deviation": 0.04166287629468534, "minimum": 0.08428158065511222, "maximum": 0.1919975565671893, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.959412325738037, "standard_deviation": 0.010306640629057584, "minimum": 5.9442756311436264, "maximum": 5.971489189772399, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4149096388860759, "standard_deviation": 0.026346794119147702, "minimum": 0.3907169020354549, "maximum": 0.468462759017482, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06280796302115611, "standard_deviation": 0.022320613243974167, "minimum": -0.10622885874423359, "maximum": -0.03425377386460978, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9800336869122842, "standard_deviation": 0.0011223269001873828, "minimum": 0.9785690970513872, "maximum": 0.9819247389635939, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2530431962113102, "standard_deviation": 0.029576752075129848, "minimum": 1.2168975693427384, "maximum": 1.2908643131380588, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2618010926009268, "standard_deviation": 0.0970808626611836, "minimum": 0.1424420566571671, "maximum": 0.3973877263654943, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11637792437576988, "standard_deviation": 0.0033739035491181612, "minimum": 0.11062733298954557, "maximum": 0.12077502146651362, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9940760895634848, "standard_deviation": 0.0006716162515917448, "minimum": 0.9926760239274935, "maximum": 0.9947030336607209, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5872174604202947, "standard_deviation": 0.029083989024991486, "minimum": 0.5601400064011247, "maximum": 0.6546114885024665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.26286137837805235, "standard_deviation": 0.08512054014493342, "minimum": 0.15989800978803018, "maximum": 0.420916751105997, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05792163301846662, "standard_deviation": 0.003309678550526239, "minimum": 0.05494830413259869, "maximum": 0.06459020636173654, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0054", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.248202085494995, "predicted_mos": 2.248202085494995, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7c841f6ac761b5553b38a432d29e8d2e1337b9f01696b82a82c6e6e8770c123a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.09032790036872, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2405106003882456, "全帯域_10ms_rms_envelope_nrmse": 0.11413749778138703, "全帯域_10ms_envelope_correlation": 0.980903241853137, "全帯域_10ms_modulation_mae_db": 0.22529412090462853, "全帯域_40ms_log_rms_mae_db": 0.5849561022686642, "全帯域_40ms_rms_envelope_nrmse": 0.05588011545927678, "全帯域_40ms_envelope_correlation": 0.9945037091344081, "全帯域_40ms_modulation_mae_db": 0.20237324504430546, "300–1000 Hz_10ms_log_rms_mae_db": 2.0750682939956455, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15393097040253662, "300–1000 Hz_10ms_envelope_correlation": 0.9783385158944172, "300–1000 Hz_10ms_modulation_mae_db": 0.7447073855648755, "300–1000 Hz_40ms_log_rms_mae_db": 1.192389688400729, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08204471644343321, "300–1000 Hz_40ms_envelope_correlation": 0.992957412706406, "300–1000 Hz_40ms_modulation_mae_db": 0.33073868032579995, "1000–3000 Hz_10ms_log_rms_mae_db": 2.060368322933565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23491854919122657, "1000–3000 Hz_10ms_envelope_correlation": 0.9459743513639006, "1000–3000 Hz_10ms_modulation_mae_db": 0.6329066306342902, "1000–3000 Hz_40ms_log_rms_mae_db": 1.205140313131723, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14295315655022464, "1000–3000 Hz_40ms_envelope_correlation": 0.977702843823769, "1000–3000 Hz_40ms_modulation_mae_db": 0.9309964478434836, "3000–6000 Hz_10ms_log_rms_mae_db": 2.190636383023665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20204588104209295, "3000–6000 Hz_10ms_envelope_correlation": 0.9733803342590789, "3000–6000 Hz_10ms_modulation_mae_db": 0.5558442242044968, "3000–6000 Hz_40ms_log_rms_mae_db": 1.597773131551692, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1370016812283804, "3000–6000 Hz_40ms_envelope_correlation": 0.9874018608785907, "3000–6000 Hz_40ms_modulation_mae_db": 0.49669866661293743, "mr_stft_db": 5.9237761787363725, "band_mae_db": 0.12301708240555544, "rms_gain_error_db": -0.041554199010778545, "output_peak": 0.4388816822393134}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3131504989470288, "全帯域_10ms_rms_envelope_nrmse": 0.12348696911030284, "全帯域_10ms_envelope_correlation": 0.9776261887942648, "全帯域_10ms_modulation_mae_db": 0.31064729930603246, "全帯域_40ms_log_rms_mae_db": 0.6063194567320029, "全帯域_40ms_rms_envelope_nrmse": 0.05895974443094451, "全帯域_40ms_envelope_correlation": 0.9938439985358863, "全帯域_40ms_modulation_mae_db": 0.21023174817065693, "300–1000 Hz_10ms_log_rms_mae_db": 2.083430767350422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1741711943352052, "300–1000 Hz_10ms_envelope_correlation": 0.9720926862048126, "300–1000 Hz_10ms_modulation_mae_db": 0.981739155631358, "300–1000 Hz_40ms_log_rms_mae_db": 1.2682686926253997, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10660285857047896, "300–1000 Hz_40ms_envelope_correlation": 0.9881448281173035, "300–1000 Hz_40ms_modulation_mae_db": 0.7398939960254397, "1000–3000 Hz_10ms_log_rms_mae_db": 1.977190800719156, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2359776515965111, "1000–3000 Hz_10ms_envelope_correlation": 0.9454752539226378, "1000–3000 Hz_10ms_modulation_mae_db": 0.6981585630979242, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1705035118945575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13461288812544292, "1000–3000 Hz_40ms_envelope_correlation": 0.9804564431743042, "1000–3000 Hz_40ms_modulation_mae_db": 0.7790566371388192, "3000–6000 Hz_10ms_log_rms_mae_db": 2.184468634674239, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2213650045761969, "3000–6000 Hz_10ms_envelope_correlation": 0.9679092586305789, "3000–6000 Hz_10ms_modulation_mae_db": 0.3286788595341922, "3000–6000 Hz_40ms_log_rms_mae_db": 1.624953520174604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15307234302448225, "3000–6000 Hz_40ms_envelope_correlation": 0.9841441241103867, "3000–6000 Hz_40ms_modulation_mae_db": 0.3659547383168207, "mr_stft_db": 5.91331022694493, "band_mae_db": 0.21518604190630075, "rms_gain_error_db": -0.044476110438978964, "output_peak": 0.4108711623840878}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2840776643464544, "全帯域_10ms_rms_envelope_nrmse": 0.11333188762785895, "全帯域_10ms_envelope_correlation": 0.9810263977466354, "全帯域_10ms_modulation_mae_db": 0.4454697895275353, "全帯域_40ms_log_rms_mae_db": 0.5976177714885623, "全帯域_40ms_rms_envelope_nrmse": 0.057047219148120644, "全帯域_40ms_envelope_correlation": 0.9942074016331189, "全帯域_40ms_modulation_mae_db": 0.5177062052421328, "300–1000 Hz_10ms_log_rms_mae_db": 2.0693871264331345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14700974186921253, "300–1000 Hz_10ms_envelope_correlation": 0.9802914308249369, "300–1000 Hz_10ms_modulation_mae_db": 0.5213750174139086, "300–1000 Hz_40ms_log_rms_mae_db": 1.2107096524275924, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08348484996302714, "300–1000 Hz_40ms_envelope_correlation": 0.9929560679834407, "300–1000 Hz_40ms_modulation_mae_db": 0.3985581549363982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9556072899360963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21419497269449514, "1000–3000 Hz_10ms_envelope_correlation": 0.9553568285657866, "1000–3000 Hz_10ms_modulation_mae_db": 0.5433933503212883, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1511726523346901, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12234099658156734, "1000–3000 Hz_40ms_envelope_correlation": 0.9841165603504778, "1000–3000 Hz_40ms_modulation_mae_db": 0.8934394017319386, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2426363116638264, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20852191549039512, "3000–6000 Hz_10ms_envelope_correlation": 0.972530784503166, "3000–6000 Hz_10ms_modulation_mae_db": 0.1884614179687325, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6589313487316553, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13907069666852462, "3000–6000 Hz_40ms_envelope_correlation": 0.9873231999624354, "3000–6000 Hz_40ms_modulation_mae_db": 0.3709311526921038, "mr_stft_db": 5.895363941513505, "band_mae_db": 0.1618325503573228, "rms_gain_error_db": -0.045841274321440605, "output_peak": 0.3847461077017335}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2079832051253196, "全帯域_10ms_rms_envelope_nrmse": 0.11885213614413058, "全帯域_10ms_envelope_correlation": 0.9792004858209896, "全帯域_10ms_modulation_mae_db": 0.3780422980335649, "全帯域_40ms_log_rms_mae_db": 0.5704101464854434, "全帯域_40ms_rms_envelope_nrmse": 0.06010981653763901, "全帯域_40ms_envelope_correlation": 0.9937614404445997, "全帯域_40ms_modulation_mae_db": 0.2625790237552027, "300–1000 Hz_10ms_log_rms_mae_db": 2.1346701714525373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1670602922090835, "300–1000 Hz_10ms_envelope_correlation": 0.9744031105591809, "300–1000 Hz_10ms_modulation_mae_db": 0.6985242194370868, "300–1000 Hz_40ms_log_rms_mae_db": 1.275964578200367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09040477601141093, "300–1000 Hz_40ms_envelope_correlation": 0.9917386100625495, "300–1000 Hz_40ms_modulation_mae_db": 0.21063498927936752, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0098795686960393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2308478245419594, "1000–3000 Hz_10ms_envelope_correlation": 0.9481065405036662, "1000–3000 Hz_10ms_modulation_mae_db": 0.37732402764046286, "1000–3000 Hz_40ms_log_rms_mae_db": 1.191759873823937, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13444290044370907, "1000–3000 Hz_40ms_envelope_correlation": 0.9799571227450535, "1000–3000 Hz_40ms_modulation_mae_db": 0.7366974367244535, "3000–6000 Hz_10ms_log_rms_mae_db": 2.217855565277296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21235569133924503, "3000–6000 Hz_10ms_envelope_correlation": 0.9705939091971537, "3000–6000 Hz_10ms_modulation_mae_db": 0.5555439349938615, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6679449868773981, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13891757414741993, "3000–6000 Hz_40ms_envelope_correlation": 0.9868430395861538, "3000–6000 Hz_40ms_modulation_mae_db": 0.6120832835518785, "mr_stft_db": 5.900246822339032, "band_mae_db": 0.10805482499682205, "rms_gain_error_db": -0.11363764035687042, "output_peak": 0.41124427867907504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2619007908053264, "全帯域_10ms_rms_envelope_nrmse": 0.11137844519226334, "全帯域_10ms_envelope_correlation": 0.981667859671199, "全帯域_10ms_modulation_mae_db": 0.3413417796505554, "全帯域_40ms_log_rms_mae_db": 0.5692805477541998, "全帯域_40ms_rms_envelope_nrmse": 0.054115544077957635, "全帯域_40ms_envelope_correlation": 0.9948528954659503, "全帯域_40ms_modulation_mae_db": 0.3034690418812431, "300–1000 Hz_10ms_log_rms_mae_db": 2.0527942094608655, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16377752957606728, "300–1000 Hz_10ms_envelope_correlation": 0.9753649379563492, "300–1000 Hz_10ms_modulation_mae_db": 0.6722950413923532, "300–1000 Hz_40ms_log_rms_mae_db": 1.2246762948840657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08838126884085425, "300–1000 Hz_40ms_envelope_correlation": 0.9919372761416996, "300–1000 Hz_40ms_modulation_mae_db": 0.2663672230622874, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9593926928608447, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2069938563633919, "1000–3000 Hz_10ms_envelope_correlation": 0.9583115738877386, "1000–3000 Hz_10ms_modulation_mae_db": 0.41924465203676875, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1300700756745383, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12266761711712139, "1000–3000 Hz_40ms_envelope_correlation": 0.9834974530827933, "1000–3000 Hz_40ms_modulation_mae_db": 0.8259129506237942, "3000–6000 Hz_10ms_log_rms_mae_db": 2.182442194585407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21235373460400128, "3000–6000 Hz_10ms_envelope_correlation": 0.9731821569588129, "3000–6000 Hz_10ms_modulation_mae_db": 0.40568721138738756, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6317761151760741, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14868716463800327, "3000–6000 Hz_40ms_envelope_correlation": 0.9867444731083871, "3000–6000 Hz_40ms_modulation_mae_db": 0.6598274153469982, "mr_stft_db": 5.8994421014817675, "band_mae_db": 0.1860440479851905, "rms_gain_error_db": -0.04711369462215156, "output_peak": 0.46808773581666235}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.271713632801968, "全帯域_10ms_rms_envelope_nrmse": 0.121854432995633, "全帯域_10ms_envelope_correlation": 0.9780551984018082, "全帯域_10ms_modulation_mae_db": 0.3079425456915606, "全帯域_40ms_log_rms_mae_db": 0.5785659488034534, "全帯域_40ms_rms_envelope_nrmse": 0.05717564514608635, "全帯域_40ms_envelope_correlation": 0.9942553264549283, "全帯域_40ms_modulation_mae_db": 0.42552685518944716, "300–1000 Hz_10ms_log_rms_mae_db": 2.068060258088378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1652638595090743, "300–1000 Hz_10ms_envelope_correlation": 0.9749375100462424, "300–1000 Hz_10ms_modulation_mae_db": 0.6689951476993783, "300–1000 Hz_40ms_log_rms_mae_db": 1.2515559211595055, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08818576404578662, "300–1000 Hz_40ms_envelope_correlation": 0.991875010199815, "300–1000 Hz_40ms_modulation_mae_db": 0.5088204761422043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.981239698102768, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2178312466991165, "1000–3000 Hz_10ms_envelope_correlation": 0.9545104180234101, "1000–3000 Hz_10ms_modulation_mae_db": 0.8234722115407179, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1948063559676403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13163752797267744, "1000–3000 Hz_40ms_envelope_correlation": 0.9830020118962324, "1000–3000 Hz_40ms_modulation_mae_db": 1.1123856552139662, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2169941921619905, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2156297202394054, "3000–6000 Hz_10ms_envelope_correlation": 0.970727528524641, "3000–6000 Hz_10ms_modulation_mae_db": 0.0675393347737079, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6193500007492538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14229268890299515, "3000–6000 Hz_40ms_envelope_correlation": 0.986751436187628, "3000–6000 Hz_40ms_modulation_mae_db": 0.22450816407063634, "mr_stft_db": 5.909962424848962, "band_mae_db": 0.20827662219794071, "rms_gain_error_db": -0.07847122045682235, "output_peak": 0.40667029193538473}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3514679557186378, "全帯域_10ms_rms_envelope_nrmse": 0.12169288062542602, "全帯域_10ms_envelope_correlation": 0.9782368945210246, "全帯域_10ms_modulation_mae_db": 0.2848153814563989, "全帯域_40ms_log_rms_mae_db": 0.6698967690469361, "全帯域_40ms_rms_envelope_nrmse": 0.06474139695181447, "全帯域_40ms_envelope_correlation": 0.9926082949962556, "全帯域_40ms_modulation_mae_db": 0.18218795384869824, "300–1000 Hz_10ms_log_rms_mae_db": 2.097324680010189, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1669367857245379, "300–1000 Hz_10ms_envelope_correlation": 0.9744969657060786, "300–1000 Hz_10ms_modulation_mae_db": 0.7857498200373468, "300–1000 Hz_40ms_log_rms_mae_db": 1.3094469437511838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09601496257910293, "300–1000 Hz_40ms_envelope_correlation": 0.9903294407361879, "300–1000 Hz_40ms_modulation_mae_db": 0.4405253762938672, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0436271711783673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23663726468392526, "1000–3000 Hz_10ms_envelope_correlation": 0.9452823789761832, "1000–3000 Hz_10ms_modulation_mae_db": 0.6455974662568198, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2485627736720057, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13524087342909216, "1000–3000 Hz_40ms_envelope_correlation": 0.9803503367000531, "1000–3000 Hz_40ms_modulation_mae_db": 0.9627718154455328, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2336121506417213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23245598120464078, "3000–6000 Hz_10ms_envelope_correlation": 0.9646026672235151, "3000–6000 Hz_10ms_modulation_mae_db": 0.8920655034884373, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6417223950773014, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15656902531104872, "3000–6000 Hz_40ms_envelope_correlation": 0.983369949927588, "3000–6000 Hz_40ms_modulation_mae_db": 0.8734502048917664, "mr_stft_db": 5.921932064064518, "band_mae_db": 0.1822912736710546, "rms_gain_error_db": -0.06625173818513605, "output_peak": 0.3944590264994322}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2476220006627559, "全帯域_10ms_rms_envelope_nrmse": 0.11670596267258075, "全帯域_10ms_envelope_correlation": 0.9799272005395179, "全帯域_10ms_modulation_mae_db": 0.20980330081906892, "全帯域_40ms_log_rms_mae_db": 0.6160371004785362, "全帯域_40ms_rms_envelope_nrmse": 0.05818533775011024, "全帯域_40ms_envelope_correlation": 0.9939933955846705, "全帯域_40ms_modulation_mae_db": 0.2820626655968085, "300–1000 Hz_10ms_log_rms_mae_db": 2.0856006468085933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15583050903369514, "300–1000 Hz_10ms_envelope_correlation": 0.9777254785142105, "300–1000 Hz_10ms_modulation_mae_db": 0.467012701173749, "300–1000 Hz_40ms_log_rms_mae_db": 1.2474138034987179, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08534579898493376, "300–1000 Hz_40ms_envelope_correlation": 0.9924772032383612, "300–1000 Hz_40ms_modulation_mae_db": 0.2360788509027463, "1000–3000 Hz_10ms_log_rms_mae_db": 2.029273335849512, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2416791629526875, "1000–3000 Hz_10ms_envelope_correlation": 0.9428713010975136, "1000–3000 Hz_10ms_modulation_mae_db": 0.9055569884579201, "1000–3000 Hz_40ms_log_rms_mae_db": 1.254599740524282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14301252769325612, "1000–3000 Hz_40ms_envelope_correlation": 0.9781689185866612, "1000–3000 Hz_40ms_modulation_mae_db": 1.1053107979980665, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1738663118967723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21040752589189382, "3000–6000 Hz_10ms_envelope_correlation": 0.9733540992241403, "3000–6000 Hz_10ms_modulation_mae_db": 0.3804770109072521, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6549561498639644, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14210267896335113, "3000–6000 Hz_40ms_envelope_correlation": 0.9877067182507157, "3000–6000 Hz_40ms_modulation_mae_db": 0.2770006567312211, "mr_stft_db": 5.915554245517323, "band_mae_db": 0.2047572760732669, "rms_gain_error_db": -0.05085509032935537, "output_peak": 0.3883422096853653}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2066057618317085, "spectral_convergence": 0.41109344358226857, "spectral_convergence_status": "ok", "amplitude_floor": 1.3149451373049662e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.721945867074296, "spectral_convergence": 0.35386126938568785, "spectral_convergence_status": "ok", "amplitude_floor": 7.412465414213296e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.842776907303113, "spectral_convergence": 0.39097831760327556, "spectral_convergence_status": "ok", "amplitude_floor": 4.876012126794439e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006137697087039973, "output_power": 0.0006174981134923844, "delta_db": 0.02630184182264548, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009615845769051578, "output_power": 0.0009449970310922175, "delta_db": -0.07557044873464444, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000542394172591287, "output_power": 0.0005475173398583812, "delta_db": 0.04082862940055351, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013703842807085808, "output_power": 0.00012849226331219775, "delta_db": -0.2796538930274193, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.620351571877408e-05, "output_power": 8.309632108657368e-05, "delta_db": -0.15943181682421476, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011238415255857088, "output_power": 0.00010841103449719547, "delta_db": -0.15631586462385527, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12301708240555544, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.041554199010778545, "pitch": null, "output_rms_dbfs": -26.14391563274022, "output_peak": 0.4388816822393134, "reference_rms_dbfs": -26.102361433729442, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9237761787363725}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013602681282863616, "log_rms_mae_db": 1.2405106003882456, "rms_envelope_nrmse": 0.11413749778138703, "envelope_correlation": 0.980903241853137, "rms_gain_error_db": -0.04116656730216068, "reference_mean_power": 0.0024543453947075094, "output_mean_power": 0.002431190686051393}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.153929939453048e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000440684439069062, "output_power": 0.0004353402083458142, "reference_active": true, "delta_db": -0.052989347127525135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001235413948928314, "output_power": 0.00011545600588489366, "reference_active": true, "delta_db": -0.29395971276267785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6241583124510384e-05, "output_power": 2.8306316757245565e-05, "reference_active": true, "delta_db": 0.3289333028236826}], "active_band_count": 3, "active_band_mae_db": 0.22529412090462853}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011635584990380617, "log_rms_mae_db": 0.5849561022686642, "rms_envelope_nrmse": 0.05588011545927678, "envelope_correlation": 0.9945037091344081, "rms_gain_error_db": -0.040622697300606025, "reference_mean_power": 0.0024645680451820984, "output_mean_power": 0.0024416226412736224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.594353848464586e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003918219819660693, "output_power": 0.0003919234855016443, "reference_active": true, "delta_db": 0.0011249168887129445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.98714927012752e-05, "output_power": 5.609323147938043e-05, "reference_active": true, "delta_db": -0.2830962596831882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.328965955891304e-06, "output_power": 1.4315407385810777e-06, "reference_active": true, "delta_db": 0.32289855856101524}], "active_band_count": 3, "active_band_mae_db": 0.20237324504430546}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010777091024361473, "log_rms_mae_db": 2.0750682939956455, "rms_envelope_nrmse": 0.15393097040253662, "envelope_correlation": 0.9783385158944172, "rms_gain_error_db": -0.03119717103032152, "reference_mean_power": 0.0010415487104655132, "output_mean_power": 0.0010340936432523993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.64672159263247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003552137892650671, "output_power": 0.0003358863231175778, "reference_active": true, "delta_db": -0.24297496322707254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.132330818484316e-05, "output_power": 6.17676005498516e-05, "reference_active": true, "delta_db": -0.6247074794656101}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2237174874701015e-05, "output_power": 1.6761973037836544e-05, "reference_active": true, "delta_db": 1.3664397140019438}], "active_band_count": 3, "active_band_mae_db": 0.7447073855648755}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.988831023316543e-05, "log_rms_mae_db": 1.192389688400729, "rms_envelope_nrmse": 0.08204471644343321, "envelope_correlation": 0.992957412706406, "rms_gain_error_db": -0.031234219951503238, "reference_mean_power": 0.0010464471526710588, "output_mean_power": 0.0010389481608731289}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.816907801707229e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003300382500563707, "output_power": 0.0003130078265208552, "reference_active": true, "delta_db": -0.2300907883057052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.204099572239431e-05, "output_power": 2.9077926031596053e-05, "reference_active": true, "delta_db": -0.4214257652287541}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.619198216109287e-07, "output_power": 7.968873257758359e-07, "reference_active": true, "delta_db": -0.34069948744294054}], "active_band_count": 3, "active_band_mae_db": 0.33073868032579995}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5701896659278796e-05, "log_rms_mae_db": 2.060368322933565, "rms_envelope_nrmse": 0.23491854919122657, "envelope_correlation": 0.9459743513639006, "rms_gain_error_db": -0.25949507519683823, "reference_mean_power": 0.0001796467459408233, "output_mean_power": 0.0001692270747938942}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2519806248605502e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7558987061199216e-05, "output_power": 4.331075909678059e-05, "reference_active": true, "delta_db": -0.4063680062351848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1481328556544434e-05, "output_power": 9.872646409221934e-06, "reference_active": true, "delta_db": -0.6555856220180196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593622469144843e-06, "output_power": 7.912359250618863e-06, "reference_active": true, "delta_db": -0.836766263649666}], "active_band_count": 3, "active_band_mae_db": 0.6329066306342902}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.308273569685666e-05, "log_rms_mae_db": 1.205140313131723, "rms_envelope_nrmse": 0.14295315655022464, "envelope_correlation": 0.977702843823769, "rms_gain_error_db": -0.2596004546184963, "reference_mean_power": 0.0001804894756897876, "output_mean_power": 0.00017001680004325692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.07434863630674e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.420432318446027e-05, "output_power": 4.110512408181919e-05, "reference_active": true, "delta_db": -0.31568781918321215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8496256649417895e-06, "output_power": 4.76275414377971e-06, "reference_active": true, "delta_db": -0.8926991161830843}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.792037562074947e-07, "output_power": 4.021343102344709e-07, "reference_active": true, "delta_db": -1.5846024081641543}], "active_band_count": 3, "active_band_mae_db": 0.9309964478434836}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.825215268339244e-05, "log_rms_mae_db": 2.190636383023665, "rms_envelope_nrmse": 0.20204588104209295, "envelope_correlation": 0.9733803342590789, "rms_gain_error_db": -0.1686257381263181, "reference_mean_power": 0.0001267678324533094, "output_mean_power": 0.00012194008425717643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.277972501807574e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1540328186660286e-05, "output_power": 4.648122817700111e-05, "reference_active": true, "delta_db": -0.44869584694545905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.007459575479445e-05, "output_power": 8.739313734828944e-06, "reference_active": true, "delta_db": -0.6175029853387886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.777776179485975e-06, "output_power": 3.28929511327208e-06, "reference_active": true, "delta_db": -0.6013338403292425}], "active_band_count": 3, "active_band_mae_db": 0.5558442242044968}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.580061124327403e-05, "log_rms_mae_db": 1.597773131551692, "rms_envelope_nrmse": 0.1370016812283804, "envelope_correlation": 0.9874018608785907, "rms_gain_error_db": -0.16857383702878004, "reference_mean_power": 0.0001273148740012305, "output_mean_power": 0.00012246775617401448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.770926544449125e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2932423481689434e-05, "output_power": 4.647527984063635e-05, "reference_active": true, "delta_db": -0.5649976491861943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.69146274773148e-06, "output_power": 4.20213784057741e-06, "reference_active": true, "delta_db": -0.4783797794384488}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.856201910676535e-07, "output_power": 2.577015603884426e-07, "reference_active": true, "delta_db": -0.44671857121416936}], "active_band_count": 3, "active_band_mae_db": 0.49669866661293743}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7968457669300468, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1056bd01e4c31736079627e9145e425b60924e3cbaf51f85fe6ea7b56bfe8246", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9494860807926047, "standard_deviation": 0.005723978373517699, "minimum": 0.9428713010975136, "maximum": 0.9583115738877386, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0020723600345436, "standard_deviation": 0.03956859457833722, "minimum": 1.9556072899360963, "maximum": 2.060368322933565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.630706736248274, "standard_deviation": 0.18284269170434048, "minimum": 0.37732402764046286, "maximum": 0.9055569884579201, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22738506609041417, "standard_deviation": 0.012612671837772583, "minimum": 0.2069938563633919, "maximum": 0.2416791629526875, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9809064612949181, "standard_deviation": 0.0024077151662088934, "minimum": 0.977702843823769, "maximum": 0.9841165603504778, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1933269121279217, "standard_deviation": 0.043505134806336145, "minimum": 1.1300700756745383, "maximum": 1.254599740524282, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9183213928400069, "standard_deviation": 0.13954151453506072, "minimum": 0.7366974367244535, "maximum": 1.1123856552139662, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1333635609891364, "standard_deviation": 0.007839390709174006, "minimum": 0.12234099658156734, "maximum": 0.14301252769325612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9707850923151358, "standard_deviation": 0.0031318890086322938, "minimum": 0.9646026672235151, "maximum": 0.9733803342590789, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2053139679906146, "standard_deviation": 0.02577234479004911, "minimum": 2.1738663118967723, "maximum": 2.2426363116638264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4217871871572585, "standard_deviation": 0.252866765489862, "minimum": 0.0675393347737079, "maximum": 0.8920655034884373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21439193179848393, "standard_deviation": 0.009161439525941291, "minimum": 0.20204588104209295, "maximum": 0.23245598120464078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9862856002514857, "standard_deviation": 0.0016110098970921602, "minimum": 0.983369949927588, "maximum": 0.9877067182507157, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.637175956025243, "standard_deviation": 0.023316852426952753, "minimum": 1.597773131551692, "maximum": 1.6679449868773981, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4850567852767953, "standard_deviation": 0.2191047956836885, "minimum": 0.22450816407063634, "maximum": 0.8734502048917664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14471423161052568, "standard_deviation": 0.00720908743013733, "minimum": 0.1370016812283804, "maximum": 0.15656902531104872, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9759563294632786, "standard_deviation": 0.00263101045392203, "minimum": 0.9720926862048126, "maximum": 0.9802914308249369, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0832920191999706, "standard_deviation": 0.02468393854694026, "minimum": 2.0527942094608655, "maximum": 2.1346701714525373, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.692549811043757, "standard_deviation": 0.15865762806236197, "minimum": 0.467012701173749, "maximum": 0.981739155631358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16174761033242657, "standard_deviation": 0.008777396817692969, "minimum": 0.14700974186921253, "maximum": 0.1741711943352052, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9915519811482204, "standard_deviation": 0.0016121707109468504, "minimum": 0.9881448281173035, "maximum": 0.992957412706406, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.247553196868445, "standard_deviation": 0.037809725821589396, "minimum": 1.192389688400729, "maximum": 1.3094469437511838, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.39145221837101385, "standard_deviation": 0.1747707857731604, "minimum": 0.21063498927936752, "maximum": 0.7398939960254397, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09005812442987848, "standard_deviation": 0.007975664323297127, "minimum": 0.08204471644343321, "maximum": 0.10660285857047896, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17368246494918171, "standard_deviation": 0.03986809575694359, "minimum": 0.10805482499682205, "maximum": 0.21518604190630075, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.909948500680802, "standard_deviation": 0.010655331309393896, "minimum": 5.895363941513505, "maximum": 5.9237761787363725, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4129128118676318, "standard_deviation": 0.02803619028904806, "minimum": 0.3847461077017335, "maximum": 0.46808773581666235, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.061025120965191734, "standard_deviation": 0.024727815939197022, "minimum": -0.11363764035687042, "maximum": -0.041554199010778545, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.979580433418572, "standard_deviation": 0.0015312507326198206, "minimum": 0.9776261887942648, "maximum": 0.981667859671199, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.272303293599467, "standard_deviation": 0.04462502313380243, "minimum": 1.2079832051253196, "maximum": 1.3514679557186378, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3129195644236681, "standard_deviation": 0.07722815558227346, "minimum": 0.20980330081906892, "maximum": 0.4454697895275353, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11768002651869781, "standard_deviation": 0.004483901999876427, "minimum": 0.11137844519226334, "maximum": 0.12348696911030284, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9940033077812273, "standard_deviation": 0.0006662696393916778, "minimum": 0.9926082949962556, "maximum": 0.9948528954659503, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5991354803822247, "standard_deviation": 0.03313901432841787, "minimum": 0.5692805477541998, "maximum": 0.6698967690469361, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.29826709234106186, "standard_deviation": 0.11731037686487825, "minimum": 0.18218795384869824, "maximum": 0.5177062052421328, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.058276852437743706, "standard_deviation": 0.003193974248792842, "minimum": 0.054115544077957635, "maximum": 0.06474139695181447, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0054", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.490394353866577, "predicted_mos": 2.490394353866577, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9caa57a9f6015e308974ac0ba6b55897e876de745f268cbddd9326eb4645ce3c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.822487900033593, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.238462179994891, "全帯域_10ms_rms_envelope_nrmse": 0.11239223391976183, "全帯域_10ms_envelope_correlation": 0.981386833257572, "全帯域_10ms_modulation_mae_db": 0.25793024767015255, "全帯域_40ms_log_rms_mae_db": 0.5778293273430287, "全帯域_40ms_rms_envelope_nrmse": 0.05533631819678982, "全帯域_40ms_envelope_correlation": 0.9946370910720633, "全帯域_40ms_modulation_mae_db": 0.31952522150335255, "300–1000 Hz_10ms_log_rms_mae_db": 2.490469317405261, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16047260976776387, "300–1000 Hz_10ms_envelope_correlation": 0.9764360933141791, "300–1000 Hz_10ms_modulation_mae_db": 0.3405512709199421, "300–1000 Hz_40ms_log_rms_mae_db": 1.6068148140783585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09043791991541307, "300–1000 Hz_40ms_envelope_correlation": 0.9916925155233888, "300–1000 Hz_40ms_modulation_mae_db": 0.45555968146785136, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4761378835723002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.27120842695756886, "1000–3000 Hz_10ms_envelope_correlation": 0.9287806372446331, "1000–3000 Hz_10ms_modulation_mae_db": 1.0651379239346814, "1000–3000 Hz_40ms_log_rms_mae_db": 1.487692972795287, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15782839481063973, "1000–3000 Hz_40ms_envelope_correlation": 0.9724152560153224, "1000–3000 Hz_40ms_modulation_mae_db": 1.5390736978186144, "3000–6000 Hz_10ms_log_rms_mae_db": 2.840295575707004, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21107547085503744, "3000–6000 Hz_10ms_envelope_correlation": 0.971567489481914, "3000–6000 Hz_10ms_modulation_mae_db": 0.366599955373637, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0509622529341747, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14858590275098088, "3000–6000 Hz_40ms_envelope_correlation": 0.9855186031365087, "3000–6000 Hz_40ms_modulation_mae_db": 0.47190346935558675, "mr_stft_db": 6.271183426829126, "band_mae_db": 0.09459401233487717, "rms_gain_error_db": -0.07914752823171858, "output_peak": 0.42641258449384867}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2398515223389124, "全帯域_10ms_rms_envelope_nrmse": 0.11759248861180985, "全帯域_10ms_envelope_correlation": 0.9796313383865012, "全帯域_10ms_modulation_mae_db": 0.2791544288544752, "全帯域_40ms_log_rms_mae_db": 0.5679436713689267, "全帯域_40ms_rms_envelope_nrmse": 0.059037149154901256, "全帯域_40ms_envelope_correlation": 0.9938952369275975, "全帯域_40ms_modulation_mae_db": 0.2986086805430354, "300–1000 Hz_10ms_log_rms_mae_db": 2.3714675095543853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15713964990029466, "300–1000 Hz_10ms_envelope_correlation": 0.977469901729083, "300–1000 Hz_10ms_modulation_mae_db": 0.5049668549579058, "300–1000 Hz_40ms_log_rms_mae_db": 1.4497510412255197, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0862157264317691, "300–1000 Hz_40ms_envelope_correlation": 0.9925578018119808, "300–1000 Hz_40ms_modulation_mae_db": 0.79584243855109, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4820775127449095, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25516616177998275, "1000–3000 Hz_10ms_envelope_correlation": 0.9366571773321789, "1000–3000 Hz_10ms_modulation_mae_db": 1.467009344040088, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5054964827438062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1524113433035588, "1000–3000 Hz_40ms_envelope_correlation": 0.9743633817997891, "1000–3000 Hz_40ms_modulation_mae_db": 2.0411044388551947, "3000–6000 Hz_10ms_log_rms_mae_db": 2.851804670061565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23866833097792986, "3000–6000 Hz_10ms_envelope_correlation": 0.9634735901092005, "3000–6000 Hz_10ms_modulation_mae_db": 0.7690315519053398, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0985178356916414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1669439389980806, "3000–6000 Hz_40ms_envelope_correlation": 0.9825534419512575, "3000–6000 Hz_40ms_modulation_mae_db": 0.6705454450415472, "mr_stft_db": 6.259967486011973, "band_mae_db": 0.16333454296255828, "rms_gain_error_db": -0.08293166473053282, "output_peak": 0.3886256112719968}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1883147316736633, "全帯域_10ms_rms_envelope_nrmse": 0.11603520750587228, "全帯域_10ms_envelope_correlation": 0.9801088091394866, "全帯域_10ms_modulation_mae_db": 0.17039445565779324, "全帯域_40ms_log_rms_mae_db": 0.5440063692112601, "全帯域_40ms_rms_envelope_nrmse": 0.06006740322957909, "全帯域_40ms_envelope_correlation": 0.9936762816312068, "全帯域_40ms_modulation_mae_db": 0.13746916278006724, "300–1000 Hz_10ms_log_rms_mae_db": 2.403476129248525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16788386904102304, "300–1000 Hz_10ms_envelope_correlation": 0.9743728746674909, "300–1000 Hz_10ms_modulation_mae_db": 0.5548287234692745, "300–1000 Hz_40ms_log_rms_mae_db": 1.5354675168101573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08872164922842163, "300–1000 Hz_40ms_envelope_correlation": 0.992525512711574, "300–1000 Hz_40ms_modulation_mae_db": 0.30838692749452884, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4720511987157145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26243153999686525, "1000–3000 Hz_10ms_envelope_correlation": 0.9322952152214843, "1000–3000 Hz_10ms_modulation_mae_db": 1.182623465423445, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5165988691755992, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15887356848003986, "1000–3000 Hz_40ms_envelope_correlation": 0.9719591568581767, "1000–3000 Hz_40ms_modulation_mae_db": 1.3369275091021187, "3000–6000 Hz_10ms_log_rms_mae_db": 2.916981210787838, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21324164483763217, "3000–6000 Hz_10ms_envelope_correlation": 0.9707635953918938, "3000–6000 Hz_10ms_modulation_mae_db": 0.1989942911591299, "3000–6000 Hz_40ms_log_rms_mae_db": 2.157713280346264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14791907587830752, "3000–6000 Hz_40ms_envelope_correlation": 0.985594627461222, "3000–6000 Hz_40ms_modulation_mae_db": 0.28458476179675146, "mr_stft_db": 6.26714273486454, "band_mae_db": 0.15695307556954477, "rms_gain_error_db": -0.08783757465691311, "output_peak": 0.40026704429072685}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2406753407603954, "全帯域_10ms_rms_envelope_nrmse": 0.11747546516921865, "全帯域_10ms_envelope_correlation": 0.9796182665798235, "全帯域_10ms_modulation_mae_db": 0.29496626302345286, "全帯域_40ms_log_rms_mae_db": 0.5845306105403476, "全帯域_40ms_rms_envelope_nrmse": 0.05856724641631812, "全帯域_40ms_envelope_correlation": 0.993977661321381, "全帯域_40ms_modulation_mae_db": 0.13956375242683544, "300–1000 Hz_10ms_log_rms_mae_db": 2.4026237749397104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.164311200231524, "300–1000 Hz_10ms_envelope_correlation": 0.9754817852419344, "300–1000 Hz_10ms_modulation_mae_db": 1.0389106398843782, "300–1000 Hz_40ms_log_rms_mae_db": 1.5104292441665845, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08793980837004829, "300–1000 Hz_40ms_envelope_correlation": 0.9926909977139381, "300–1000 Hz_40ms_modulation_mae_db": 0.5148925563215314, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5057459823474892, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2626098358120831, "1000–3000 Hz_10ms_envelope_correlation": 0.93218634703603, "1000–3000 Hz_10ms_modulation_mae_db": 1.239081458095033, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4887045020085756, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15410938133827737, "1000–3000 Hz_40ms_envelope_correlation": 0.9741971906376619, "1000–3000 Hz_40ms_modulation_mae_db": 1.6555459144683375, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8213458692227635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2232626486421921, "3000–6000 Hz_10ms_envelope_correlation": 0.9696930651151984, "3000–6000 Hz_10ms_modulation_mae_db": 0.49808840754556377, "3000–6000 Hz_40ms_log_rms_mae_db": 2.056450276259647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14888432776033453, "3000–6000 Hz_40ms_envelope_correlation": 0.9864492229391946, "3000–6000 Hz_40ms_modulation_mae_db": 0.4588824694070701, "mr_stft_db": 6.2697307996344, "band_mae_db": 0.2013882258724126, "rms_gain_error_db": -0.08205721020145801, "output_peak": 0.39179189686444765}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2146338272128716, "全帯域_10ms_rms_envelope_nrmse": 0.11523027408910336, "全帯域_10ms_envelope_correlation": 0.9803711509376898, "全帯域_10ms_modulation_mae_db": 0.42261937436350977, "全帯域_40ms_log_rms_mae_db": 0.5729151991846625, "全帯域_40ms_rms_envelope_nrmse": 0.05652956750500534, "全帯域_40ms_envelope_correlation": 0.994327145653728, "全帯域_40ms_modulation_mae_db": 0.4954531347882654, "300–1000 Hz_10ms_log_rms_mae_db": 2.414467379311914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15581357445546065, "300–1000 Hz_10ms_envelope_correlation": 0.9778652500949786, "300–1000 Hz_10ms_modulation_mae_db": 1.111896050810503, "300–1000 Hz_40ms_log_rms_mae_db": 1.5191951568396973, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0878188302840497, "300–1000 Hz_40ms_envelope_correlation": 0.99224200364651, "300–1000 Hz_40ms_modulation_mae_db": 0.7418479083133439, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5057496976651294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24651398997096352, "1000–3000 Hz_10ms_envelope_correlation": 0.9410662851888165, "1000–3000 Hz_10ms_modulation_mae_db": 0.5862050053188396, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4517425913505113, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13661849696688066, "1000–3000 Hz_40ms_envelope_correlation": 0.9794585738171703, "1000–3000 Hz_40ms_modulation_mae_db": 0.8183003649289158, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8412125088718674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22905576769748562, "3000–6000 Hz_10ms_envelope_correlation": 0.9684222929985788, "3000–6000 Hz_10ms_modulation_mae_db": 0.43170875089746263, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0890373647591574, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15973615590263815, "3000–6000 Hz_40ms_envelope_correlation": 0.9845309856215372, "3000–6000 Hz_40ms_modulation_mae_db": 0.5117097408561472, "mr_stft_db": 6.2635426154430265, "band_mae_db": 0.21249753833870103, "rms_gain_error_db": -0.04767560672916658, "output_peak": 0.4275342331582395}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2329369005405642, "全帯域_10ms_rms_envelope_nrmse": 0.11554434646128209, "全帯域_10ms_envelope_correlation": 0.980323281901051, "全帯域_10ms_modulation_mae_db": 0.2908450214959913, "全帯域_40ms_log_rms_mae_db": 0.574004132937968, "全帯域_40ms_rms_envelope_nrmse": 0.054947223017273505, "全帯域_40ms_envelope_correlation": 0.9948492283346024, "全帯域_40ms_modulation_mae_db": 0.24636249373655872, "300–1000 Hz_10ms_log_rms_mae_db": 2.427678174963585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17284521124579477, "300–1000 Hz_10ms_envelope_correlation": 0.9727143156507284, "300–1000 Hz_10ms_modulation_mae_db": 1.1809360804061888, "300–1000 Hz_40ms_log_rms_mae_db": 1.5357456234205655, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09218152464884909, "300–1000 Hz_40ms_envelope_correlation": 0.9917746107292547, "300–1000 Hz_40ms_modulation_mae_db": 0.696792114498364, "1000–3000 Hz_10ms_log_rms_mae_db": 2.561776555264465, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26726044841584345, "1000–3000 Hz_10ms_envelope_correlation": 0.9309915740214403, "1000–3000 Hz_10ms_modulation_mae_db": 1.2026422660848133, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5446926672907642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15011665698364898, "1000–3000 Hz_40ms_envelope_correlation": 0.9751164147168575, "1000–3000 Hz_40ms_modulation_mae_db": 1.8045935171881018, "3000–6000 Hz_10ms_log_rms_mae_db": 2.856900754894065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2115767821218148, "3000–6000 Hz_10ms_envelope_correlation": 0.9716540169273473, "3000–6000 Hz_10ms_modulation_mae_db": 0.5113074081244137, "3000–6000 Hz_40ms_log_rms_mae_db": 2.105776689533177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1481455611333401, "3000–6000 Hz_40ms_envelope_correlation": 0.9857199240407983, "3000–6000 Hz_40ms_modulation_mae_db": 0.5701348381817296, "mr_stft_db": 6.2807507233895095, "band_mae_db": 0.1404731899110757, "rms_gain_error_db": -0.11418465656747638, "output_peak": 0.4277805630476217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2113505009334111, "全帯域_10ms_rms_envelope_nrmse": 0.11218175898310093, "全帯域_10ms_envelope_correlation": 0.981438571438706, "全帯域_10ms_modulation_mae_db": 0.18493504736459845, "全帯域_40ms_log_rms_mae_db": 0.6197318154996102, "全帯域_40ms_rms_envelope_nrmse": 0.061042271360865374, "全帯域_40ms_envelope_correlation": 0.9933831929676297, "全帯域_40ms_modulation_mae_db": 0.4288285847809736, "300–1000 Hz_10ms_log_rms_mae_db": 2.3945042920487656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1525841047726126, "300–1000 Hz_10ms_envelope_correlation": 0.9789124879910932, "300–1000 Hz_10ms_modulation_mae_db": 0.46550696670237596, "300–1000 Hz_40ms_log_rms_mae_db": 1.4856779934928361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08910296354497277, "300–1000 Hz_40ms_envelope_correlation": 0.9922798821776775, "300–1000 Hz_40ms_modulation_mae_db": 0.4161599567435834, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3983606012257828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25751299136224376, "1000–3000 Hz_10ms_envelope_correlation": 0.935526008926009, "1000–3000 Hz_10ms_modulation_mae_db": 1.1594749862133915, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4753833232864288, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15530261130019407, "1000–3000 Hz_40ms_envelope_correlation": 0.9733656793722961, "1000–3000 Hz_40ms_modulation_mae_db": 1.4921769743860496, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8485311301194645, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21696754579800506, "3000–6000 Hz_10ms_envelope_correlation": 0.969908710736, "3000–6000 Hz_10ms_modulation_mae_db": 0.8089373933447987, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0878736428732037, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1500743187064749, "3000–6000 Hz_40ms_envelope_correlation": 0.9852750513687516, "3000–6000 Hz_40ms_modulation_mae_db": 0.830042596948782, "mr_stft_db": 6.2640334094479115, "band_mae_db": 0.19390825663370334, "rms_gain_error_db": -0.05524662436936249, "output_peak": 0.43621100147806424}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1940219319425625, "全帯域_10ms_rms_envelope_nrmse": 0.11258306959327248, "全帯域_10ms_envelope_correlation": 0.9814156984797029, "全帯域_10ms_modulation_mae_db": 0.10785502631041681, "全帯域_40ms_log_rms_mae_db": 0.5601219723830408, "全帯域_40ms_rms_envelope_nrmse": 0.05865428531355723, "全帯域_40ms_envelope_correlation": 0.9940973885655258, "全帯域_40ms_modulation_mae_db": 0.08643567126548302, "300–1000 Hz_10ms_log_rms_mae_db": 2.34550174148276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.159105765393695, "300–1000 Hz_10ms_envelope_correlation": 0.9767660640722907, "300–1000 Hz_10ms_modulation_mae_db": 0.6166570759647854, "300–1000 Hz_40ms_log_rms_mae_db": 1.4771581886634026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08858620458782915, "300–1000 Hz_40ms_envelope_correlation": 0.9918629116484549, "300–1000 Hz_40ms_modulation_mae_db": 0.29157986083706516, "1000–3000 Hz_10ms_log_rms_mae_db": 2.462606037519851, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25590257359022656, "1000–3000 Hz_10ms_envelope_correlation": 0.9356723869755325, "1000–3000 Hz_10ms_modulation_mae_db": 1.0659036493329133, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4566685021578132, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14163232635349504, "1000–3000 Hz_40ms_envelope_correlation": 0.9779952211518196, "1000–3000 Hz_40ms_modulation_mae_db": 1.2731158977919421, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8984883499030962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23552035031918972, "3000–6000 Hz_10ms_envelope_correlation": 0.9640067348791603, "3000–6000 Hz_10ms_modulation_mae_db": 0.5802125694826032, "3000–6000 Hz_40ms_log_rms_mae_db": 2.076553793191641, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16078005772140383, "3000–6000 Hz_40ms_envelope_correlation": 0.9828418705792197, "3000–6000 Hz_40ms_modulation_mae_db": 0.5444411693129244, "mr_stft_db": 6.270142134962481, "band_mae_db": 0.14060540304879712, "rms_gain_error_db": -0.09906090559014231, "output_peak": 0.41527987480117606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.617887072337924, "spectral_convergence": 0.4174234821698809, "spectral_convergence_status": "ok", "amplitude_floor": 1.3149451373049662e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.1484699204230395, "spectral_convergence": 0.3640445502287812, "spectral_convergence_status": "ok", "amplitude_floor": 7.412465414213296e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.047193287726416, "spectral_convergence": 0.3949314435054129, "spectral_convergence_status": "ok", "amplitude_floor": 4.876012126794439e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006137697087039973, "output_power": 0.0006283354749644999, "delta_db": 0.10186129136607502, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009615845769051578, "output_power": 0.000906563275106412, "delta_db": -0.25589367154807424, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000542394172591287, "output_power": 0.0005376186299175973, "delta_db": -0.038407051164646676, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013703842807085808, "output_power": 0.0001383111247996905, "delta_db": 0.040147448716245086, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.620351571877408e-05, "output_power": 8.45837335910968e-05, "delta_db": -0.08238127098147398, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011238415255857088, "output_power": 0.00011365601090464231, "delta_db": 0.04887334023274799, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09459401233487717, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07914752823171858, "pitch": null, "output_rms_dbfs": -26.18150896196116, "output_peak": 0.42641258449384867, "reference_rms_dbfs": -26.102361433729442, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.271183426829126}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013602681282863616, "log_rms_mae_db": 1.238462179994891, "rms_envelope_nrmse": 0.11239223391976183, "envelope_correlation": 0.981386833257572, "rms_gain_error_db": -0.07841930919307222, "reference_mean_power": 0.0024543453947075094, "output_mean_power": 0.0024104256997138944}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.153929939453048e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000440684439069062, "output_power": 0.00042695944716451665, "reference_active": true, "delta_db": -0.13741088021517622}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001235413948928314, "output_power": 0.0001248278691967091, "reference_active": true, "delta_db": 0.044990565959786205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6241583124510384e-05, "output_power": 2.290082364261813e-05, "reference_active": true, "delta_db": -0.5913892968354951}], "active_band_count": 3, "active_band_mae_db": 0.25793024767015255}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011635584990380617, "log_rms_mae_db": 0.5778293273430287, "rms_envelope_nrmse": 0.05533631819678982, "envelope_correlation": 0.9946370910720633, "rms_gain_error_db": -0.07726076809005758, "reference_mean_power": 0.0024645680451820984, "output_mean_power": 0.0024211111982974337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.594353848464586e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003918219819660693, "output_power": 0.00038222609148610563, "reference_active": true, "delta_db": -0.10768467591943787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.98714927012752e-05, "output_power": 6.0799013896534834e-05, "reference_active": true, "delta_db": 0.0667644947376361}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.328965955891304e-06, "output_power": 1.1094333233847823e-06, "reference_active": true, "delta_db": -0.7841264938529836}], "active_band_count": 3, "active_band_mae_db": 0.31952522150335255}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010777091024361473, "log_rms_mae_db": 2.490469317405261, "rms_envelope_nrmse": 0.16047260976776387, "envelope_correlation": 0.9764360933141791, "rms_gain_error_db": -0.16465803522833858, "reference_mean_power": 0.0010415487104655132, "output_mean_power": 0.0010027987456157127}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.64672159263247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003552137892650671, "output_power": 0.00033395759255389285, "reference_active": true, "delta_db": -0.26798494946988805}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.132330818484316e-05, "output_power": 6.415254140215442e-05, "reference_active": true, "delta_db": -0.46017613203716445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2237174874701015e-05, "output_power": 1.3092736860782827e-05, "reference_active": true, "delta_db": 0.29349273125277386}], "active_band_count": 3, "active_band_mae_db": 0.3405512709199421}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.988831023316543e-05, "log_rms_mae_db": 1.6068148140783585, "rms_envelope_nrmse": 0.09043791991541307, "envelope_correlation": 0.9916925155233888, "rms_gain_error_db": -0.16470404153470672, "reference_mean_power": 0.0010464471526710588, "output_mean_power": 0.001007504272404347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.816907801707229e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003300382500563707, "output_power": 0.00031101627208464914, "reference_active": true, "delta_db": -0.25781164205322843}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.204099572239431e-05, "output_power": 3.107814466913207e-05, "reference_active": true, "delta_db": -0.13250919977218298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.619198216109287e-07, "output_power": 6.883844446486653e-07, "reference_active": true, "delta_db": -0.9763582025781428}], "active_band_count": 3, "active_band_mae_db": 0.45555968146785136}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5701896659278796e-05, "log_rms_mae_db": 2.4761378835723002, "rms_envelope_nrmse": 0.27120842695756886, "envelope_correlation": 0.9287806372446331, "rms_gain_error_db": 0.008119038141004342, "reference_mean_power": 0.0001796467459408233, "output_mean_power": 0.00017998290563534612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2519806248605502e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7558987061199216e-05, "output_power": 4.4888913335164234e-05, "reference_active": true, "delta_db": -0.25093504004286626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1481328556544434e-05, "output_power": 9.12078803109496e-06, "reference_active": true, "delta_db": -0.9995978239678499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593622469144843e-06, "output_power": 6.13048075208094e-06, "reference_active": true, "delta_db": -1.9448809077933282}], "active_band_count": 3, "active_band_mae_db": 1.0651379239346814}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.308273569685666e-05, "log_rms_mae_db": 1.487692972795287, "rms_envelope_nrmse": 0.15782839481063973, "envelope_correlation": 0.9724152560153224, "rms_gain_error_db": 0.008244619711511943, "reference_mean_power": 0.0001804894756897876, "output_mean_power": 0.0001808324412382785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.07434863630674e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.420432318446027e-05, "output_power": 4.185004754627138e-05, "reference_active": true, "delta_db": -0.2376878970676949}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8496256649417895e-06, "output_power": 4.116759718360958e-06, "reference_active": true, "delta_db": -1.5257255576161426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.792037562074947e-07, "output_power": 3.002275939365044e-07, "reference_active": true, "delta_db": -2.853807638772005}], "active_band_count": 3, "active_band_mae_db": 1.5390736978186144}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.825215268339244e-05, "log_rms_mae_db": 2.840295575707004, "rms_envelope_nrmse": 0.21107547085503744, "envelope_correlation": 0.971567489481914, "rms_gain_error_db": 0.010752829590138098, "reference_mean_power": 0.0001267678324533094, "output_mean_power": 0.00012708208967662774}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.277972501807574e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1540328186660286e-05, "output_power": 4.542398165168275e-05, "reference_active": true, "delta_db": -0.5486197986905648}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.007459575479445e-05, "output_power": 9.614752494839696e-06, "reference_active": true, "delta_db": -0.20289519997862143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.777776179485975e-06, "output_power": 3.4866450488528344e-06, "reference_active": true, "delta_db": -0.34828486745172466}], "active_band_count": 3, "active_band_mae_db": 0.366599955373637}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.580061124327403e-05, "log_rms_mae_db": 2.0509622529341747, "rms_envelope_nrmse": 0.14858590275098088, "envelope_correlation": 0.9855186031365087, "rms_gain_error_db": 0.011339853003414406, "reference_mean_power": 0.0001273148740012305, "output_mean_power": 0.0001276477399529544}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.770926544449125e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2932423481689434e-05, "output_power": 4.5548600208520904e-05, "reference_active": true, "delta_db": -0.6524674352543002}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.69146274773148e-06, "output_power": 4.4263543965334925e-06, "reference_active": true, "delta_db": -0.25262089370099977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.856201910676535e-07, "output_power": 2.5393741683106225e-07, "reference_active": true, "delta_db": -0.5106220791114603}], "active_band_count": 3, "active_band_mae_db": 0.47190346935558675}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7691849290235019, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4191c23e4f0290b36d7eb42c0db8e9c0527d3875f0ac650fa769cb31453f41a8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9341469539932655, "standard_deviation": 0.003862526893642655, "minimum": 0.9287806372446331, "maximum": 0.9410662851888165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.4830631836319554, "standard_deviation": 0.0462727929680378, "minimum": 2.3983606012257828, "maximum": 2.561776555264465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1210097623054005, "standard_deviation": 0.25007735875925596, "minimum": 0.5862050053188396, "maximum": 1.467009344040088, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25982574598572217, "standard_deviation": 0.007737074324142043, "minimum": 0.24651398997096352, "maximum": 0.27120842695756886, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9748588592961367, "standard_deviation": 0.002628221425500123, "minimum": 0.9719591568581767, "maximum": 0.9794585738171703, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4908724888510982, "standard_deviation": 0.031002545454391833, "minimum": 1.4517425913505113, "maximum": 1.5446926672907642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4951047893174094, "standard_deviation": 0.36915839233857556, "minimum": 0.8183003649289158, "maximum": 2.0411044388551947, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15086159744209182, "standard_deviation": 0.007874013809515, "minimum": 0.13661849696688066, "maximum": 0.15887356848003986, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9686861869549117, "standard_deviation": 0.0032309218158686325, "minimum": 0.9634735901092005, "maximum": 0.9716540169273473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.8594450086959577, "standard_deviation": 0.0320031852286291, "minimum": 2.8213458692227635, "maximum": 2.916981210787838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5206100409791186, "standard_deviation": 0.20144172143621067, "minimum": 0.1989942911591299, "maximum": 0.8089373933447987, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22242106765616085, "standard_deviation": 0.010958467530961944, "minimum": 0.21107547085503744, "maximum": 0.23866833097792986, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9848104658873111, "standard_deviation": 0.0014087725137875277, "minimum": 0.9825534419512575, "maximum": 0.9864492229391946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.090360641948613, "standard_deviation": 0.03325503933805078, "minimum": 2.0509622529341747, "maximum": 2.157713280346264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5427805613625674, "standard_deviation": 0.16003895752294425, "minimum": 0.28458476179675146, "maximum": 0.830042596948782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15388366735644504, "standard_deviation": 0.0074494062609082295, "minimum": 0.14791907587830752, "maximum": 0.1669439389980806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9762523465952223, "standard_deviation": 0.002003631041135895, "minimum": 0.9727143156507284, "maximum": 0.9789124879910932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.406273539869363, "standard_deviation": 0.04255846243382069, "minimum": 2.34550174148276, "maximum": 2.490469317405261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7267817078894192, "standard_deviation": 0.32957315567704965, "minimum": 0.3405512709199421, "maximum": 1.1809360804061888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1612694981010211, "standard_deviation": 0.006703672406051135, "minimum": 0.1525841047726126, "maximum": 0.17284521124579477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9922032794953474, "standard_deviation": 0.00038456727582946234, "minimum": 0.9916925155233888, "maximum": 0.9926909977139381, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5150299473371402, "standard_deviation": 0.04760995329126416, "minimum": 1.4497510412255197, "maximum": 1.6068148140783585, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5276326805284197, "standard_deviation": 0.19567180329338546, "minimum": 0.29157986083706516, "maximum": 0.79584243855109, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0888755783764191, "standard_deviation": 0.0017966843124676737, "minimum": 0.0862157264317691, "maximum": 0.09218152464884909, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16296928058395876, "standard_deviation": 0.038921177624580264, "minimum": 0.09459401233487717, "maximum": 0.21249753833870103, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.26831166632287, "standard_deviation": 0.006328854307360788, "minimum": 6.259967486011973, "maximum": 6.2807507233895095, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41423785117576517, "standard_deviation": 0.018312053954902246, "minimum": 0.3886256112719968, "maximum": 0.43621100147806424, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08101772138459629, "standard_deviation": 0.02156783277403737, "minimum": -0.11418465656747638, "maximum": -0.04767560672916658, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9805367437650666, "standard_deviation": 0.0007771726158303946, "minimum": 0.9796182665798235, "maximum": 0.981438571438706, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2200308669246591, "standard_deviation": 0.021085494037871338, "minimum": 1.1883147316736633, "maximum": 1.2406753407603954, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2510874830925488, "standard_deviation": 0.09648598853701745, "minimum": 0.10785502631041681, "maximum": 0.42261937436350977, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11487935554167769, "standard_deviation": 0.002227681596880537, "minimum": 0.11218175898310093, "maximum": 0.11759248861180985, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9941054033092169, "standard_deviation": 0.00048571340233709813, "minimum": 0.9933831929676297, "maximum": 0.9948492283346024, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5751353873086056, "standard_deviation": 0.021832862142142563, "minimum": 0.5440063692112601, "maximum": 0.6197318154996102, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2690308377280714, "standard_deviation": 0.14548046372229104, "minimum": 0.08643567126548302, "maximum": 0.4954531347882654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05802268302428622, "standard_deviation": 0.0022015517548609176, "minimum": 0.054947223017273505, "maximum": 0.061042271360865374, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0054", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.191538095474243, "predicted_mos": 2.191538095474243, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "33dc7146e6ef7332211f3b060cd0dc19e0d90ae59dce10db88d052c27f134f4e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8692736998200417, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2841913813871917, "全帯域_10ms_rms_envelope_nrmse": 0.12627285410344463, "全帯域_10ms_envelope_correlation": 0.9764607191436675, "全帯域_10ms_modulation_mae_db": 0.264991482935139, "全帯域_40ms_log_rms_mae_db": 0.5618175917998657, "全帯域_40ms_rms_envelope_nrmse": 0.06165854950904302, "全帯域_40ms_envelope_correlation": 0.993376342615863, "全帯域_40ms_modulation_mae_db": 0.14453697090755022, "300–1000 Hz_10ms_log_rms_mae_db": 2.0184813946845797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16490901560510846, "300–1000 Hz_10ms_envelope_correlation": 0.9750707650725033, "300–1000 Hz_10ms_modulation_mae_db": 0.5973282459298564, "300–1000 Hz_40ms_log_rms_mae_db": 1.24270897797848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08882352745370468, "300–1000 Hz_40ms_envelope_correlation": 0.9920333699960733, "300–1000 Hz_40ms_modulation_mae_db": 0.2257875348973876, "1000–3000 Hz_10ms_log_rms_mae_db": 1.842147056745297, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22459701290684475, "1000–3000 Hz_10ms_envelope_correlation": 0.9511530198933307, "1000–3000 Hz_10ms_modulation_mae_db": 0.8489511793253012, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1810092986609422, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14019864260706266, "1000–3000 Hz_40ms_envelope_correlation": 0.9798708830230719, "1000–3000 Hz_40ms_modulation_mae_db": 1.0919099813063324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9806751831145268, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22162197277192525, "3000–6000 Hz_10ms_envelope_correlation": 0.968271877671202, "3000–6000 Hz_10ms_modulation_mae_db": 0.4914156885496855, "3000–6000 Hz_40ms_log_rms_mae_db": 1.514065005450702, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14402083534598345, "3000–6000 Hz_40ms_envelope_correlation": 0.9860375013592141, "3000–6000 Hz_40ms_modulation_mae_db": 0.38119681805640404, "mr_stft_db": 5.829816947903669, "band_mae_db": 0.19578623743677767, "rms_gain_error_db": -0.10329251800039724, "output_peak": 0.3913277104467354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3502538522388898, "全帯域_10ms_rms_envelope_nrmse": 0.1227353561764013, "全帯域_10ms_envelope_correlation": 0.9777592447578056, "全帯域_10ms_modulation_mae_db": 0.08058180116112605, "全帯域_40ms_log_rms_mae_db": 0.6381812458030982, "全帯域_40ms_rms_envelope_nrmse": 0.06315116830325637, "全帯域_40ms_envelope_correlation": 0.9929151406194285, "全帯域_40ms_modulation_mae_db": 0.17739012307290283, "300–1000 Hz_10ms_log_rms_mae_db": 2.0560723801697502, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15998304449712275, "300–1000 Hz_10ms_envelope_correlation": 0.9766539975194303, "300–1000 Hz_10ms_modulation_mae_db": 0.2404953838618478, "300–1000 Hz_40ms_log_rms_mae_db": 1.2715424356028024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08690526973040787, "300–1000 Hz_40ms_envelope_correlation": 0.9925157863454444, "300–1000 Hz_40ms_modulation_mae_db": 0.1564293306283275, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8751182539099622, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22230612838550098, "1000–3000 Hz_10ms_envelope_correlation": 0.9518231643923668, "1000–3000 Hz_10ms_modulation_mae_db": 0.502965208013024, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2038451068000537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1376466621990175, "1000–3000 Hz_40ms_envelope_correlation": 0.9790590614749428, "1000–3000 Hz_40ms_modulation_mae_db": 0.96202223641427, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9955127725110162, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20803295677116804, "3000–6000 Hz_10ms_envelope_correlation": 0.9718798111916825, "3000–6000 Hz_10ms_modulation_mae_db": 0.29488692697610736, "3000–6000 Hz_40ms_log_rms_mae_db": 1.584491991996638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14687718606069206, "3000–6000 Hz_40ms_envelope_correlation": 0.9858946050029838, "3000–6000 Hz_40ms_modulation_mae_db": 0.3405513364390607, "mr_stft_db": 5.84396916110569, "band_mae_db": 0.2289490604471008, "rms_gain_error_db": -0.06371430395338547, "output_peak": 0.44500515227674475}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.342650364517313, "全帯域_10ms_rms_envelope_nrmse": 0.12553650115847845, "全帯域_10ms_envelope_correlation": 0.976834734223828, "全帯域_10ms_modulation_mae_db": 0.2701445291144229, "全帯域_40ms_log_rms_mae_db": 0.6544206373299727, "全帯域_40ms_rms_envelope_nrmse": 0.06537093653223235, "全帯域_40ms_envelope_correlation": 0.9924915351819218, "全帯域_40ms_modulation_mae_db": 0.17346945801364658, "300–1000 Hz_10ms_log_rms_mae_db": 2.0338531650920304, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1734434626377397, "300–1000 Hz_10ms_envelope_correlation": 0.9724538581485792, "300–1000 Hz_10ms_modulation_mae_db": 0.38975224708773, "300–1000 Hz_40ms_log_rms_mae_db": 1.287684527365619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09657373583468545, "300–1000 Hz_40ms_envelope_correlation": 0.9906314767899955, "300–1000 Hz_40ms_modulation_mae_db": 0.16912687508037794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8345140820953456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21815243339339657, "1000–3000 Hz_10ms_envelope_correlation": 0.9536532350103815, "1000–3000 Hz_10ms_modulation_mae_db": 0.39935970428043105, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1777635030988363, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1271524446272831, "1000–3000 Hz_40ms_envelope_correlation": 0.9824661124687415, "1000–3000 Hz_40ms_modulation_mae_db": 0.6130064537852508, "3000–6000 Hz_10ms_log_rms_mae_db": 2.01604957249875, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22687848511287922, "3000–6000 Hz_10ms_envelope_correlation": 0.9665354935725257, "3000–6000 Hz_10ms_modulation_mae_db": 0.5560447770689578, "3000–6000 Hz_40ms_log_rms_mae_db": 1.588755511654279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16300193990000858, "3000–6000 Hz_40ms_envelope_correlation": 0.9819394831531943, "3000–6000 Hz_40ms_modulation_mae_db": 0.705160144037345, "mr_stft_db": 5.827754697736804, "band_mae_db": 0.1495369945902061, "rms_gain_error_db": -0.07658877062516185, "output_peak": 0.400422103614401}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2385466192760142, "全帯域_10ms_rms_envelope_nrmse": 0.11496579895683572, "全帯域_10ms_envelope_correlation": 0.9806143898324192, "全帯域_10ms_modulation_mae_db": 0.3810138304236483, "全帯域_40ms_log_rms_mae_db": 0.5900172347001282, "全帯域_40ms_rms_envelope_nrmse": 0.06329496423196142, "全帯域_40ms_envelope_correlation": 0.992962153028897, "全帯域_40ms_modulation_mae_db": 0.1349939071999098, "300–1000 Hz_10ms_log_rms_mae_db": 1.8778076301352429, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15397856688587375, "300–1000 Hz_10ms_envelope_correlation": 0.9783094028219771, "300–1000 Hz_10ms_modulation_mae_db": 1.1510686012980382, "300–1000 Hz_40ms_log_rms_mae_db": 1.1551697056051022, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09190771185407826, "300–1000 Hz_40ms_envelope_correlation": 0.9914225887987208, "300–1000 Hz_40ms_modulation_mae_db": 0.8629263677263924, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8444857894735043, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21722576244051234, "1000–3000 Hz_10ms_envelope_correlation": 0.9542964599823459, "1000–3000 Hz_10ms_modulation_mae_db": 0.48549617600531086, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1621182920453828, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12899268524838778, "1000–3000 Hz_40ms_envelope_correlation": 0.9829607057195924, "1000–3000 Hz_40ms_modulation_mae_db": 0.7064828909572093, "3000–6000 Hz_10ms_log_rms_mae_db": 2.010387800389455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2173886652572194, "3000–6000 Hz_10ms_envelope_correlation": 0.9698111335112467, "3000–6000 Hz_10ms_modulation_mae_db": 0.4824641171484915, "3000–6000 Hz_40ms_log_rms_mae_db": 1.579587565647105, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16301663641240696, "3000–6000 Hz_40ms_envelope_correlation": 0.9821041055690668, "3000–6000 Hz_40ms_modulation_mae_db": 0.49441845300981685, "mr_stft_db": 5.832128275940126, "band_mae_db": 0.21714663724148586, "rms_gain_error_db": -0.08213942511873516, "output_peak": 0.39871105389817946}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2636807008741633, "全帯域_10ms_rms_envelope_nrmse": 0.11398945745777589, "全帯域_10ms_envelope_correlation": 0.9808339982655884, "全帯域_10ms_modulation_mae_db": 0.23832097853578924, "全帯域_40ms_log_rms_mae_db": 0.6327364936179919, "全帯域_40ms_rms_envelope_nrmse": 0.05654209941880302, "全帯域_40ms_envelope_correlation": 0.9943533082501628, "全帯域_40ms_modulation_mae_db": 0.49201165006221115, "300–1000 Hz_10ms_log_rms_mae_db": 1.9866613519985727, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14938488604290237, "300–1000 Hz_10ms_envelope_correlation": 0.9795495638789209, "300–1000 Hz_10ms_modulation_mae_db": 0.7518100481304657, "300–1000 Hz_40ms_log_rms_mae_db": 1.228268156313187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07590609509991139, "300–1000 Hz_40ms_envelope_correlation": 0.9940606137652316, "300–1000 Hz_40ms_modulation_mae_db": 0.30354667417724834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9071047826869385, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23004502669437854, "1000–3000 Hz_10ms_envelope_correlation": 0.9491080454524679, "1000–3000 Hz_10ms_modulation_mae_db": 1.3026326781136044, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2118546123315377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1379530349315139, "1000–3000 Hz_40ms_envelope_correlation": 0.9815983861689023, "1000–3000 Hz_40ms_modulation_mae_db": 1.6568350476348215, "3000–6000 Hz_10ms_log_rms_mae_db": 2.012244471967851, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2261854022469626, "3000–6000 Hz_10ms_envelope_correlation": 0.9672458255290834, "3000–6000 Hz_10ms_modulation_mae_db": 0.7165386236760757, "3000–6000 Hz_40ms_log_rms_mae_db": 1.60139277517199, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16342731165638544, "3000–6000 Hz_40ms_envelope_correlation": 0.981991466246584, "3000–6000 Hz_40ms_modulation_mae_db": 0.6889246952004041, "mr_stft_db": 5.829665668135749, "band_mae_db": 0.23832888954654183, "rms_gain_error_db": -0.06632318993604058, "output_peak": 0.4380267856140009}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2713616532431884, "全帯域_10ms_rms_envelope_nrmse": 0.12348411956868728, "全帯域_10ms_envelope_correlation": 0.9775999880454108, "全帯域_10ms_modulation_mae_db": 0.6772900843899624, "全帯域_40ms_log_rms_mae_db": 0.6135372229333356, "全帯域_40ms_rms_envelope_nrmse": 0.06409725914338293, "全帯域_40ms_envelope_correlation": 0.992650535680265, "全帯域_40ms_modulation_mae_db": 0.6812333266808145, "300–1000 Hz_10ms_log_rms_mae_db": 2.035218258332718, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14863253601827733, "300–1000 Hz_10ms_envelope_correlation": 0.9797575767645867, "300–1000 Hz_10ms_modulation_mae_db": 1.0201637266850154, "300–1000 Hz_40ms_log_rms_mae_db": 1.3102741446939652, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08687611503351353, "300–1000 Hz_40ms_envelope_correlation": 0.9921896863432036, "300–1000 Hz_40ms_modulation_mae_db": 0.6793410792941946, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8231480907852597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22527523022992002, "1000–3000 Hz_10ms_envelope_correlation": 0.9507572665619975, "1000–3000 Hz_10ms_modulation_mae_db": 0.5282494930456642, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1917315070943546, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14481322820832213, "1000–3000 Hz_40ms_envelope_correlation": 0.9783296463841835, "1000–3000 Hz_40ms_modulation_mae_db": 0.9831198009260808, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9406423796909038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19633441113501823, "3000–6000 Hz_10ms_envelope_correlation": 0.9749575667417382, "3000–6000 Hz_10ms_modulation_mae_db": 0.47069557444896987, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5592087223736142, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14176065769108376, "3000–6000 Hz_40ms_envelope_correlation": 0.9863229151689537, "3000–6000 Hz_40ms_modulation_mae_db": 0.5790778572768402, "mr_stft_db": 5.814088193320264, "band_mae_db": 0.14398068066942932, "rms_gain_error_db": -0.008001183717277077, "output_peak": 0.41137926559602794}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2916550440887626, "全帯域_10ms_rms_envelope_nrmse": 0.11702586589484631, "全帯域_10ms_envelope_correlation": 0.9799831042419975, "全帯域_10ms_modulation_mae_db": 0.2032757405141122, "全帯域_40ms_log_rms_mae_db": 0.6002161036901392, "全帯域_40ms_rms_envelope_nrmse": 0.06360249783188002, "全帯域_40ms_envelope_correlation": 0.9932467223700984, "全帯域_40ms_modulation_mae_db": 0.1598585977858041, "300–1000 Hz_10ms_log_rms_mae_db": 1.9931616273042057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1703320464629551, "300–1000 Hz_10ms_envelope_correlation": 0.9735809373414415, "300–1000 Hz_10ms_modulation_mae_db": 0.8582557505114911, "300–1000 Hz_40ms_log_rms_mae_db": 1.248735130356953, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1048045254154567, "300–1000 Hz_40ms_envelope_correlation": 0.9891802710843619, "300–1000 Hz_40ms_modulation_mae_db": 0.8174074262175958, "1000–3000 Hz_10ms_log_rms_mae_db": 1.904048741710965, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2234087629117139, "1000–3000 Hz_10ms_envelope_correlation": 0.9513564588419156, "1000–3000 Hz_10ms_modulation_mae_db": 0.6836274093558551, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1649045226173558, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13395218335305198, "1000–3000 Hz_40ms_envelope_correlation": 0.9808614677378433, "1000–3000 Hz_40ms_modulation_mae_db": 0.83085474952349, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9563950475072425, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1937416034269441, "3000–6000 Hz_10ms_envelope_correlation": 0.9762162948664417, "3000–6000 Hz_10ms_modulation_mae_db": 0.19183514263251292, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5845317997964075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1360217545005914, "3000–6000 Hz_40ms_envelope_correlation": 0.9877490860351368, "3000–6000 Hz_40ms_modulation_mae_db": 0.14085245650402886, "mr_stft_db": 5.838703088754399, "band_mae_db": 0.19206387232271038, "rms_gain_error_db": -0.16149433827550938, "output_peak": 0.3839282813681957}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3498703472231928, "全帯域_10ms_rms_envelope_nrmse": 0.11942230084798486, "全帯域_10ms_envelope_correlation": 0.9791378060666813, "全帯域_10ms_modulation_mae_db": 0.4357875071518575, "全帯域_40ms_log_rms_mae_db": 0.6736727980756172, "全帯域_40ms_rms_envelope_nrmse": 0.05939483719071028, "全帯域_40ms_envelope_correlation": 0.993740999389418, "全帯域_40ms_modulation_mae_db": 0.46260853619423625, "300–1000 Hz_10ms_log_rms_mae_db": 2.027826146997278, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15693787829285774, "300–1000 Hz_10ms_envelope_correlation": 0.977425875469411, "300–1000 Hz_10ms_modulation_mae_db": 0.7611473379289794, "300–1000 Hz_40ms_log_rms_mae_db": 1.3161348841819478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08818179220493863, "300–1000 Hz_40ms_envelope_correlation": 0.9918756367423036, "300–1000 Hz_40ms_modulation_mae_db": 0.2800009445180851, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8987880264435333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2351743951770873, "1000–3000 Hz_10ms_envelope_correlation": 0.9463153501075476, "1000–3000 Hz_10ms_modulation_mae_db": 0.7219745479817355, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1814535580425007, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1367185916419142, "1000–3000 Hz_40ms_envelope_correlation": 0.9811342527816971, "1000–3000 Hz_40ms_modulation_mae_db": 0.8475772612079711, "3000–6000 Hz_10ms_log_rms_mae_db": 2.033920975706729, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22545886916155278, "3000–6000 Hz_10ms_envelope_correlation": 0.9681449752225871, "3000–6000 Hz_10ms_modulation_mae_db": 0.8800844491893987, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5787043755263381, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1509077399485646, "3000–6000 Hz_40ms_envelope_correlation": 0.9851824022382497, "3000–6000 Hz_40ms_modulation_mae_db": 0.9217088333033733, "mr_stft_db": 5.837727483564103, "band_mae_db": 0.17420430008561297, "rms_gain_error_db": -0.007403109231555523, "output_peak": 0.43200126800283556}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.036258072263417, "spectral_convergence": 0.42077680565747627, "spectral_convergence_status": "ok", "amplitude_floor": 1.3149451373049662e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.606320680555159, "spectral_convergence": 0.3586601431129179, "spectral_convergence_status": "ok", "amplitude_floor": 7.412465414213296e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.846872090892432, "spectral_convergence": 0.38732069899093813, "spectral_convergence_status": "ok", "amplitude_floor": 4.876012126794439e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006137697087039973, "output_power": 0.0006335769433186784, "delta_db": 0.13793913387185994, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009615845769051578, "output_power": 0.0009027535697272659, "delta_db": -0.2741827453908905, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000542394172591287, "output_power": 0.0005388978400613463, "delta_db": -0.028085719655035254, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013703842807085808, "output_power": 0.00012467630141777133, "delta_db": -0.41058458074593734, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.620351571877408e-05, "output_power": 8.142812556128414e-05, "delta_db": -0.24750540779948416, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011238415255857088, "output_power": 0.0001143792010979716, "delta_db": 0.07641983715745904, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19578623743677767, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10329251800039724, "pitch": null, "output_rms_dbfs": -26.205653951729836, "output_peak": 0.3913277104467354, "reference_rms_dbfs": -26.102361433729442, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.829816947903669}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013602681282863616, "log_rms_mae_db": 1.2841913813871917, "rms_envelope_nrmse": 0.12627285410344463, "envelope_correlation": 0.9764607191436675, "rms_gain_error_db": -0.10306482921253123, "reference_mean_power": 0.0024543453947075094, "output_mean_power": 0.0023967856569873233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.153929939453048e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000440684439069062, "output_power": 0.0004173144354279668, "reference_active": true, "delta_db": -0.23664307723839384}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001235413948928314, "output_power": 0.0001254440594008905, "reference_active": true, "delta_db": 0.06637598842515664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6241583124510384e-05, "output_power": 2.3431197984528436e-05, "reference_active": true, "delta_db": -0.4919553831418667}], "active_band_count": 3, "active_band_mae_db": 0.264991482935139}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011635584990380617, "log_rms_mae_db": 0.5618175917998657, "rms_envelope_nrmse": 0.06165854950904302, "envelope_correlation": 0.993376342615863, "rms_gain_error_db": -0.10308592040973144, "reference_mean_power": 0.0024645680451820984, "output_mean_power": 0.002406756875805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.594353848464586e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003918219819660693, "output_power": 0.0003746501511846455, "reference_active": true, "delta_db": -0.19462884909717013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.98714927012752e-05, "output_power": 6.0649540961666746e-05, "reference_active": true, "delta_db": 0.056074321658231535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.328965955891304e-06, "output_power": 1.3861321285568232e-06, "reference_active": true, "delta_db": 0.182907741967249}], "active_band_count": 3, "active_band_mae_db": 0.14453697090755022}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010777091024361473, "log_rms_mae_db": 2.0184813946845797, "rms_envelope_nrmse": 0.16490901560510846, "envelope_correlation": 0.9750707650725033, "rms_gain_error_db": -0.1811442336833079, "reference_mean_power": 0.0010415487104655132, "output_mean_power": 0.0009989992500237376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.64672159263247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003552137892650671, "output_power": 0.00032448436839019974, "reference_active": true, "delta_db": -0.39296036509481946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.132330818484316e-05, "output_power": 6.370333453860344e-05, "reference_active": true, "delta_db": -0.490693128359425}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2237174874701015e-05, "output_power": 1.5083922444718604e-05, "reference_active": true, "delta_db": 0.9083312443353249}], "active_band_count": 3, "active_band_mae_db": 0.5973282459298564}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.988831023316543e-05, "log_rms_mae_db": 1.24270897797848, "rms_envelope_nrmse": 0.08882352745370468, "envelope_correlation": 0.9920333699960733, "rms_gain_error_db": -0.18132386142384246, "reference_mean_power": 0.0010464471526710588, "output_mean_power": 0.0010036560676469056}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.816907801707229e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003300382500563707, "output_power": 0.0003028555261006664, "reference_active": true, "delta_db": -0.3732877321979544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.204099572239431e-05, "output_power": 2.989140716450703e-05, "reference_active": true, "delta_db": -0.3015964366118258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.619198216109287e-07, "output_power": 8.614280807137945e-07, "reference_active": true, "delta_db": -0.002478435882382648}], "active_band_count": 3, "active_band_mae_db": 0.2257875348973876}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5701896659278796e-05, "log_rms_mae_db": 1.842147056745297, "rms_envelope_nrmse": 0.22459701290684475, "envelope_correlation": 0.9511530198933307, "rms_gain_error_db": -0.40987219684770426, "reference_mean_power": 0.0001796467459408233, "output_mean_power": 0.00016346776893678546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2519806248605502e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7558987061199216e-05, "output_power": 3.93601064289675e-05, "reference_active": true, "delta_db": -0.8217633210542935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1481328556544434e-05, "output_power": 1.0468775141067812e-05, "reference_active": true, "delta_db": -0.400962734628443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593622469144843e-06, "output_power": 7.072449836292473e-06, "reference_active": true, "delta_db": -1.3241274822931668}], "active_band_count": 3, "active_band_mae_db": 0.8489511793253012}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.308273569685666e-05, "log_rms_mae_db": 1.1810092986609422, "rms_envelope_nrmse": 0.14019864260706266, "envelope_correlation": 0.9798708830230719, "rms_gain_error_db": -0.40978457689817627, "reference_mean_power": 0.0001804894756897876, "output_mean_power": 0.00016423791599225077}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.07434863630674e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.420432318446027e-05, "output_power": 3.704939286086605e-05, "reference_active": true, "delta_db": -0.7668364997860576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8496256649417895e-06, "output_power": 5.165585169672584e-06, "reference_active": true, "delta_db": -0.5400854862154756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.792037562074947e-07, "output_power": 3.680870800431852e-07, "reference_active": true, "delta_db": -1.968807957917464}], "active_band_count": 3, "active_band_mae_db": 1.0919099813063324}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.825215268339244e-05, "log_rms_mae_db": 1.9806751831145268, "rms_envelope_nrmse": 0.22162197277192525, "envelope_correlation": 0.968271877671202, "rms_gain_error_db": 0.008485561902087484, "reference_mean_power": 0.0001267678324533094, "output_mean_power": 0.00012701576281070278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.277972501807574e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1540328186660286e-05, "output_power": 5.0475576427908024e-05, "reference_active": true, "delta_db": -0.09065892364648179}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.007459575479445e-05, "output_power": 9.490051062528047e-06, "reference_active": true, "delta_db": -0.259590797933445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.777776179485975e-06, "output_power": 2.9163288184000533e-06, "reference_active": true, "delta_db": -1.1239973440691295}], "active_band_count": 3, "active_band_mae_db": 0.4914156885496855}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.580061124327403e-05, "log_rms_mae_db": 1.514065005450702, "rms_envelope_nrmse": 0.14402083534598345, "envelope_correlation": 0.9860375013592141, "rms_gain_error_db": 0.007268196865936123, "reference_mean_power": 0.0001273148740012305, "output_mean_power": 0.00012752812200572376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.770926544449125e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2932423481689434e-05, "output_power": 5.017403716366051e-05, "reference_active": true, "delta_db": -0.23242731220663287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.69146274773148e-06, "output_power": 4.5414638447887e-06, "reference_active": true, "delta_db": -0.14112411126372595}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.856201910676535e-07, "output_power": 2.3921312373182147e-07, "reference_active": true, "delta_db": -0.7700390306988534}], "active_band_count": 3, "active_band_mae_db": 0.38119681805640404}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8012984904774804, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dccb29d36a57c262f3c8c1312eb6599e43064a3d62f17fe191e9149fcfa99e90", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9510578750302943, "standard_deviation": 0.0025154364017791887, "minimum": 0.9463153501075476, "maximum": 0.9542964599823459, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8661693529813508, "standard_deviation": 0.03413157524437211, "minimum": 1.8231480907852597, "maximum": 1.9071047826869385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6841570495151158, "standard_deviation": 0.29002782433187346, "minimum": 0.39935970428043105, "maximum": 1.3026326781136044, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22452309401741932, "standard_deviation": 0.005910114394871009, "minimum": 0.21722576244051234, "maximum": 0.2351743951770873, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9807850644698719, "standard_deviation": 0.0016121188974353657, "minimum": 0.9783296463841835, "maximum": 0.9829607057195924, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1843350500863705, "standard_deviation": 0.017428854481395615, "minimum": 1.1621182920453828, "maximum": 1.2118546123315377, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9614760527194283, "standard_deviation": 0.32015472267832273, "minimum": 0.6130064537852508, "maximum": 1.6568350476348215, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13592843410206915, "standard_deviation": 0.005780238288242112, "minimum": 0.1271524446272831, "maximum": 0.14481322820832213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9703828722883134, "standard_deviation": 0.003618054460013892, "minimum": 0.9665354935725257, "maximum": 0.9762162948664417, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.993228525423309, "standard_deviation": 0.03188715515308643, "minimum": 1.9406423796909038, "maximum": 2.033920975706729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5104956624612749, "standard_deviation": 0.21774487320216906, "minimum": 0.19183514263251292, "maximum": 0.8800844491893987, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2144552957354587, "standard_deviation": 0.013478265696216678, "minimum": 0.1937416034269441, "maximum": 0.22687848511287922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.984652695596673, "standard_deviation": 0.0023008339113291835, "minimum": 0.9819394831531943, "maximum": 0.9877490860351368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5738422184521343, "standard_deviation": 0.026868006207797775, "minimum": 1.514065005450702, "maximum": 1.60139277517199, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5314863242284091, "standard_deviation": 0.24567689229832315, "minimum": 0.14085245650402886, "maximum": 0.9217088333033733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15112925768946453, "standard_deviation": 0.010811228667712967, "minimum": 0.1360217545005914, "maximum": 0.16342731165638544, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9766002471271062, "standard_deviation": 0.0026971162561484708, "minimum": 0.9724538581485792, "maximum": 0.9797575767645867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0036352443392973, "standard_deviation": 0.055672856741590594, "minimum": 1.8778076301352429, "maximum": 2.0560723801697502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.721252667679178, "standard_deviation": 0.3051374627021669, "minimum": 0.2404953838618478, "maximum": 1.1510686012980382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15970017955535465, "standard_deviation": 0.00924578346296456, "minimum": 0.14863253601827733, "maximum": 0.1734434626377397, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9917386787331668, "standard_deviation": 0.0014230523292398298, "minimum": 0.9891802710843619, "maximum": 0.9940606137652316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2575647452622571, "standard_deviation": 0.05203434203092662, "minimum": 1.1551697056051022, "maximum": 1.3161348841819478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.43682077906745115, "standard_deviation": 0.29818429894814774, "minimum": 0.1564293306283275, "maximum": 0.8629263677263924, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08999734657833706, "standard_deviation": 0.008355420987004016, "minimum": 0.07590609509991139, "maximum": 0.1048045254154567, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19249958404248313, "standard_deviation": 0.03501416903364718, "minimum": 0.14398068066942932, "maximum": 0.23832888954654183, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.8317316895576, "standard_deviation": 0.009019875250286262, "minimum": 5.814088193320264, "maximum": 5.84396916110569, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41260020260214003, "standard_deviation": 0.022960514085057396, "minimum": 0.3839282813681957, "maximum": 0.44500515227674475, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07111960485725778, "standard_deviation": 0.04996514209630205, "minimum": -0.16149433827550938, "maximum": -0.007403109231555523, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9786529980721748, "standard_deviation": 0.0017166344199107358, "minimum": 0.9764607191436675, "maximum": 0.9808339982655884, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2990262453560895, "standard_deviation": 0.04319273138848647, "minimum": 1.2385466192760142, "maximum": 1.3502538522388898, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3189257442782572, "standard_deviation": 0.18043795953100025, "minimum": 0.08058180116112605, "maximum": 0.6772900843899624, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1204290317705568, "standard_deviation": 0.0047629413475439196, "minimum": 0.11398945745777589, "maximum": 0.12627285410344463, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9932170921420067, "standard_deviation": 0.0006088974007185925, "minimum": 0.9924915351819218, "maximum": 0.9943533082501628, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6205749159937686, "standard_deviation": 0.036396644563768635, "minimum": 0.5618175917998657, "maximum": 0.6736727980756172, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.30326282123963444, "standard_deviation": 0.21065958586618316, "minimum": 0.1349939071999098, "maximum": 0.6812333266808145, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06213903902015867, "standard_deviation": 0.00287524694110664, "minimum": 0.05654209941880302, "maximum": 0.06537093653223235, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0054", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.226691484451294, "predicted_mos": 2.226691484451294, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d994160fd611883538c8428ff97969119d9c3a6b39c1c292ea3a7159d7196b5b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.899645899888128, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2025302542458276, "全帯域_10ms_rms_envelope_nrmse": 0.11419248044553693, "全帯域_10ms_envelope_correlation": 0.980849681537581, "全帯域_10ms_modulation_mae_db": 0.225223060243957, "全帯域_40ms_log_rms_mae_db": 0.5661019735127862, "全帯域_40ms_rms_envelope_nrmse": 0.05565039710952711, "全帯域_40ms_envelope_correlation": 0.9945084335280017, "全帯域_40ms_modulation_mae_db": 0.16375851009474088, "300–1000 Hz_10ms_log_rms_mae_db": 2.1006719690859126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15266145835010317, "300–1000 Hz_10ms_envelope_correlation": 0.9786615198762373, "300–1000 Hz_10ms_modulation_mae_db": 0.7296205525384479, "300–1000 Hz_40ms_log_rms_mae_db": 1.285188412249265, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08032970089994526, "300–1000 Hz_40ms_envelope_correlation": 0.9932621230078192, "300–1000 Hz_40ms_modulation_mae_db": 0.2768785632901323, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1175362152776342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24334055300349744, "1000–3000 Hz_10ms_envelope_correlation": 0.9424344035428042, "1000–3000 Hz_10ms_modulation_mae_db": 0.590246425459953, "1000–3000 Hz_40ms_log_rms_mae_db": 1.307823189723148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14780749933339524, "1000–3000 Hz_40ms_envelope_correlation": 0.9757375471013222, "1000–3000 Hz_40ms_modulation_mae_db": 0.9220551228574426, "3000–6000 Hz_10ms_log_rms_mae_db": 2.343369061574162, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20518367416778058, "3000–6000 Hz_10ms_envelope_correlation": 0.9726872314509267, "3000–6000 Hz_10ms_modulation_mae_db": 0.44087382694030475, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7325245134017657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13840728603557412, "3000–6000 Hz_40ms_envelope_correlation": 0.9873287081138807, "3000–6000 Hz_40ms_modulation_mae_db": 0.40393623825214336, "mr_stft_db": 5.993495207491577, "band_mae_db": 0.0717997521187135, "rms_gain_error_db": -0.031029316485022638, "output_peak": 0.43652958742844233}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2704700077253537, "全帯域_10ms_rms_envelope_nrmse": 0.120539706852796, "全帯域_10ms_envelope_correlation": 0.9786460121646492, "全帯域_10ms_modulation_mae_db": 0.2876189955057888, "全帯域_40ms_log_rms_mae_db": 0.5784289166544956, "全帯域_40ms_rms_envelope_nrmse": 0.055843999877979174, "全帯域_40ms_envelope_correlation": 0.9944857279487588, "全帯域_40ms_modulation_mae_db": 0.24596372306262915, "300–1000 Hz_10ms_log_rms_mae_db": 2.1504646623409593, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16738031430948047, "300–1000 Hz_10ms_envelope_correlation": 0.9742581268441798, "300–1000 Hz_10ms_modulation_mae_db": 0.972363485545849, "300–1000 Hz_40ms_log_rms_mae_db": 1.3763416032704219, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09917522217660768, "300–1000 Hz_40ms_envelope_correlation": 0.9898323557633651, "300–1000 Hz_40ms_modulation_mae_db": 0.7328847367735641, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0743167512366836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24114562445185506, "1000–3000 Hz_10ms_envelope_correlation": 0.943206959838185, "1000–3000 Hz_10ms_modulation_mae_db": 0.46784010159615547, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2606722421326786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1418298681115719, "1000–3000 Hz_40ms_envelope_correlation": 0.9777146474133492, "1000–3000 Hz_40ms_modulation_mae_db": 0.4071611320956147, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3172948638421578, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22091771980032146, "3000–6000 Hz_10ms_envelope_correlation": 0.9681392482420763, "3000–6000 Hz_10ms_modulation_mae_db": 0.40702566265575985, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7356613084744354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15274402753183505, "3000–6000 Hz_40ms_envelope_correlation": 0.9843454885653108, "3000–6000 Hz_40ms_modulation_mae_db": 0.44702221381109486, "mr_stft_db": 5.9829680449581915, "band_mae_db": 0.15412942586508369, "rms_gain_error_db": -0.04734645881182509, "output_peak": 0.4092558518469701}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2140445665640403, "全帯域_10ms_rms_envelope_nrmse": 0.11053156268313603, "全帯域_10ms_envelope_correlation": 0.9819608304108947, "全帯域_10ms_modulation_mae_db": 0.3636727966818933, "全帯域_40ms_log_rms_mae_db": 0.5653081663593896, "全帯域_40ms_rms_envelope_nrmse": 0.05525140589495412, "全帯域_40ms_envelope_correlation": 0.9945784423377921, "全帯域_40ms_modulation_mae_db": 0.4142740738441704, "300–1000 Hz_10ms_log_rms_mae_db": 2.1199376305355457, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14376575492750698, "300–1000 Hz_10ms_envelope_correlation": 0.9812757433330913, "300–1000 Hz_10ms_modulation_mae_db": 0.47580186873977887, "300–1000 Hz_40ms_log_rms_mae_db": 1.33327815643048, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0828011986185101, "300–1000 Hz_40ms_envelope_correlation": 0.9932936677969187, "300–1000 Hz_40ms_modulation_mae_db": 0.35018451686286234, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9970852152653327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21629843667426374, "1000–3000 Hz_10ms_envelope_correlation": 0.9544435138004048, "1000–3000 Hz_10ms_modulation_mae_db": 0.4391048403361341, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1849053002001848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1255853414072195, "1000–3000 Hz_40ms_envelope_correlation": 0.9828073257704939, "1000–3000 Hz_40ms_modulation_mae_db": 0.7067860085986389, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3771755939699317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20849327173416196, "3000–6000 Hz_10ms_envelope_correlation": 0.9722962287602516, "3000–6000 Hz_10ms_modulation_mae_db": 0.20830366597505479, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7988184321195166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13935556927617665, "3000–6000 Hz_40ms_envelope_correlation": 0.987261052270677, "3000–6000 Hz_40ms_modulation_mae_db": 0.3003049111207385, "mr_stft_db": 5.96872849310564, "band_mae_db": 0.16793481700539167, "rms_gain_error_db": -0.05251813195645171, "output_peak": 0.39141992812280096}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.218457588341954, "全帯域_10ms_rms_envelope_nrmse": 0.11800651525141717, "全帯域_10ms_envelope_correlation": 0.9794721206427318, "全帯域_10ms_modulation_mae_db": 0.3351660567072112, "全帯域_40ms_log_rms_mae_db": 0.5531475103220018, "全帯域_40ms_rms_envelope_nrmse": 0.060714127568689526, "全帯域_40ms_envelope_correlation": 0.9935922950346079, "全帯域_40ms_modulation_mae_db": 0.2223613502001761, "300–1000 Hz_10ms_log_rms_mae_db": 2.162226281431572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1698661881422117, "300–1000 Hz_10ms_envelope_correlation": 0.9735325496175834, "300–1000 Hz_10ms_modulation_mae_db": 0.8259598456150264, "300–1000 Hz_40ms_log_rms_mae_db": 1.3692325725889016, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09625979684583837, "300–1000 Hz_40ms_envelope_correlation": 0.9906190261460189, "300–1000 Hz_40ms_modulation_mae_db": 0.3650549597225566, "1000–3000 Hz_10ms_log_rms_mae_db": 2.049712787110113, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2330954349941214, "1000–3000 Hz_10ms_envelope_correlation": 0.9475673480646859, "1000–3000 Hz_10ms_modulation_mae_db": 0.38912207778618263, "1000–3000 Hz_40ms_log_rms_mae_db": 1.26644164414345, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13898560169668073, "1000–3000 Hz_40ms_envelope_correlation": 0.9786483104478206, "1000–3000 Hz_40ms_modulation_mae_db": 0.7286802410535417, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2987457483333507, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2082083614438559, "3000–6000 Hz_10ms_envelope_correlation": 0.9718901634255971, "3000–6000 Hz_10ms_modulation_mae_db": 0.6688712173858908, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7373222085673357, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13779191366422727, "3000–6000 Hz_40ms_envelope_correlation": 0.9872005156290203, "3000–6000 Hz_40ms_modulation_mae_db": 0.7310929288482724, "mr_stft_db": 5.9716487115223265, "band_mae_db": 0.06723202375335356, "rms_gain_error_db": -0.10579955734912273, "output_peak": 0.4098672915038995}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.253117000806126, "全帯域_10ms_rms_envelope_nrmse": 0.11378011640604635, "全帯域_10ms_envelope_correlation": 0.9808606076466273, "全帯域_10ms_modulation_mae_db": 0.4197533010542404, "全帯域_40ms_log_rms_mae_db": 0.5656195799510167, "全帯域_40ms_rms_envelope_nrmse": 0.05826004708211898, "全帯域_40ms_envelope_correlation": 0.9940268039882078, "全帯域_40ms_modulation_mae_db": 0.3213552918481034, "300–1000 Hz_10ms_log_rms_mae_db": 2.1177731322316182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16150693880427386, "300–1000 Hz_10ms_envelope_correlation": 0.9760687735279944, "300–1000 Hz_10ms_modulation_mae_db": 0.7970439121031409, "300–1000 Hz_40ms_log_rms_mae_db": 1.350342529676164, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08920308797367169, "300–1000 Hz_40ms_envelope_correlation": 0.9918623789677641, "300–1000 Hz_40ms_modulation_mae_db": 0.3453520299842054, "1000–3000 Hz_10ms_log_rms_mae_db": 2.030496302875841, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21771328380411858, "1000–3000 Hz_10ms_envelope_correlation": 0.9541589195956217, "1000–3000 Hz_10ms_modulation_mae_db": 0.3104786900711966, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2563841507940112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12935897568314092, "1000–3000 Hz_40ms_envelope_correlation": 0.9815692799830461, "1000–3000 Hz_40ms_modulation_mae_db": 0.5982822504939526, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3397218850100203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2178878935578735, "3000–6000 Hz_10ms_envelope_correlation": 0.9718145489950205, "3000–6000 Hz_10ms_modulation_mae_db": 0.43867262854144573, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7905628001629335, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15394402621914557, "3000–6000 Hz_40ms_envelope_correlation": 0.9858515817609451, "3000–6000 Hz_40ms_modulation_mae_db": 0.6681907295017995, "mr_stft_db": 5.975817529381651, "band_mae_db": 0.15892958209354122, "rms_gain_error_db": -0.06360888267932131, "output_peak": 0.46817686241750367}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2518003581973927, "全帯域_10ms_rms_envelope_nrmse": 0.117522466950736, "全帯域_10ms_envelope_correlation": 0.979583320626379, "全帯域_10ms_modulation_mae_db": 0.19088835735506338, "全帯域_40ms_log_rms_mae_db": 0.5673262600309201, "全帯域_40ms_rms_envelope_nrmse": 0.0546257454349723, "全帯域_40ms_envelope_correlation": 0.9947638030342917, "全帯域_40ms_modulation_mae_db": 0.23658865146901886, "300–1000 Hz_10ms_log_rms_mae_db": 2.1188733627532996, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16226385589242415, "300–1000 Hz_10ms_envelope_correlation": 0.9758225226591243, "300–1000 Hz_10ms_modulation_mae_db": 0.6061278475164306, "300–1000 Hz_40ms_log_rms_mae_db": 1.351326454243879, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09084227707460275, "300–1000 Hz_40ms_envelope_correlation": 0.9914611312152278, "300–1000 Hz_40ms_modulation_mae_db": 0.535584693046966, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0313650764109137, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22819310479151125, "1000–3000 Hz_10ms_envelope_correlation": 0.9491412710244549, "1000–3000 Hz_10ms_modulation_mae_db": 0.7337050477108998, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2882734376646348, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13901337967570052, "1000–3000 Hz_40ms_envelope_correlation": 0.9793131908122343, "1000–3000 Hz_40ms_modulation_mae_db": 1.0103817253103, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3424938685763372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2124159693508712, "3000–6000 Hz_10ms_envelope_correlation": 0.9714091961329949, "3000–6000 Hz_10ms_modulation_mae_db": 0.30685045805085576, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7679064016258623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1425103837415986, "3000–6000 Hz_40ms_envelope_correlation": 0.9866792360835184, "3000–6000 Hz_40ms_modulation_mae_db": 0.4127153501605094, "mr_stft_db": 5.991131112291832, "band_mae_db": 0.18210329016505442, "rms_gain_error_db": -0.07371196051003061, "output_peak": 0.4176746237191964}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2703138672543464, "全帯域_10ms_rms_envelope_nrmse": 0.12011610001092858, "全帯域_10ms_envelope_correlation": 0.9787832485888882, "全帯域_10ms_modulation_mae_db": 0.21458878495882075, "全帯域_40ms_log_rms_mae_db": 0.6339151268354424, "全帯域_40ms_rms_envelope_nrmse": 0.06436671017414437, "全帯域_40ms_envelope_correlation": 0.9927054525274636, "全帯域_40ms_modulation_mae_db": 0.23883878700795144, "300–1000 Hz_10ms_log_rms_mae_db": 2.185761458600604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16685285606133524, "300–1000 Hz_10ms_envelope_correlation": 0.9744845198702253, "300–1000 Hz_10ms_modulation_mae_db": 0.6828167827338136, "300–1000 Hz_40ms_log_rms_mae_db": 1.4301285317069354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09675044435626337, "300–1000 Hz_40ms_envelope_correlation": 0.9901792779723787, "300–1000 Hz_40ms_modulation_mae_db": 0.4892684455476461, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1209821769797217, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2428655359127154, "1000–3000 Hz_10ms_envelope_correlation": 0.94235563715477, "1000–3000 Hz_10ms_modulation_mae_db": 0.5334718777317519, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3586750488841477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14123074309075806, "1000–3000 Hz_40ms_envelope_correlation": 0.977868832416861, "1000–3000 Hz_40ms_modulation_mae_db": 0.8350416828048607, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3307754197056383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23320088014166, "3000–6000 Hz_10ms_envelope_correlation": 0.9645390366983725, "3000–6000 Hz_10ms_modulation_mae_db": 0.7689120458724901, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7528040206053512, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15636901941615358, "3000–6000 Hz_40ms_envelope_correlation": 0.9835707400137499, "3000–6000 Hz_40ms_modulation_mae_db": 0.8176303400004357, "mr_stft_db": 5.995297995274772, "band_mae_db": 0.13352751464465487, "rms_gain_error_db": -0.07011557807984732, "output_peak": 0.3939046852231277}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2304351681786094, "全帯域_10ms_rms_envelope_nrmse": 0.11660233431360356, "全帯域_10ms_envelope_correlation": 0.9799549663058266, "全帯域_10ms_modulation_mae_db": 0.1266848413236884, "全帯域_40ms_log_rms_mae_db": 0.5869143235131479, "全帯域_40ms_rms_envelope_nrmse": 0.05775826057152265, "全帯域_40ms_envelope_correlation": 0.9940695162848275, "全帯域_40ms_modulation_mae_db": 0.3491052541090849, "300–1000 Hz_10ms_log_rms_mae_db": 2.1646103990547982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15301883394694302, "300–1000 Hz_10ms_envelope_correlation": 0.9785749840475229, "300–1000 Hz_10ms_modulation_mae_db": 0.4864142788829318, "300–1000 Hz_40ms_log_rms_mae_db": 1.3639697276066667, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08546138942420957, "300–1000 Hz_40ms_envelope_correlation": 0.9925732332456951, "300–1000 Hz_40ms_modulation_mae_db": 0.24495064429766789, "1000–3000 Hz_10ms_log_rms_mae_db": 2.09523621265089, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24677039719396168, "1000–3000 Hz_10ms_envelope_correlation": 0.9403474764314664, "1000–3000 Hz_10ms_modulation_mae_db": 0.8099297363724451, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3119021663011445, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14693662147293812, "1000–3000 Hz_40ms_envelope_correlation": 0.9760683288266806, "1000–3000 Hz_40ms_modulation_mae_db": 0.9635662151867282, "3000–6000 Hz_10ms_log_rms_mae_db": 2.282202532651523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20909315369604725, "3000–6000 Hz_10ms_envelope_correlation": 0.9735211324373464, "3000–6000 Hz_10ms_modulation_mae_db": 0.3488965671903781, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7321824602601774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14208604637915748, "3000–6000 Hz_40ms_envelope_correlation": 0.9876411696606351, "3000–6000 Hz_40ms_modulation_mae_db": 0.194246460097843, "mr_stft_db": 5.98514684728529, "band_mae_db": 0.1596081606946832, "rms_gain_error_db": -0.04591393457178961, "output_peak": 0.3881522796175974}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2894833966510735, "spectral_convergence": 0.41045804597071195, "spectral_convergence_status": "ok", "amplitude_floor": 1.3149451373049662e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.783694828167179, "spectral_convergence": 0.3551074525372866, "spectral_convergence_status": "ok", "amplitude_floor": 7.412465414213296e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.907307397656477, "spectral_convergence": 0.3905459773221126, "spectral_convergence_status": "ok", "amplitude_floor": 4.876012126794439e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006137697087039973, "output_power": 0.0006222613487550557, "delta_db": 0.05967375036483224, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009615845769051578, "output_power": 0.0009392650353297066, "delta_db": -0.1019933317639002, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000542394172591287, "output_power": 0.0005450068294484089, "delta_db": 0.020869295456246988, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013703842807085808, "output_power": 0.00013503319428087986, "delta_db": -0.06401827073652026, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.620351571877408e-05, "output_power": 8.465376045720158e-05, "delta_db": -0.0787872348677911, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011238415255857088, "output_power": 0.00010968807433377428, "delta_db": -0.10545662952299023, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.0717997521187135, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.031029316485022638, "pitch": null, "output_rms_dbfs": -26.133390750214463, "output_peak": 0.43652958742844233, "reference_rms_dbfs": -26.102361433729442, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.993495207491577}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013602681282863616, "log_rms_mae_db": 1.2025302542458276, "rms_envelope_nrmse": 0.11419248044553693, "envelope_correlation": 0.980849681537581, "rms_gain_error_db": -0.03067735548091814, "reference_mean_power": 0.0024543453947075094, "output_mean_power": 0.002437069668097049}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.153929939453048e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000440684439069062, "output_power": 0.0004318978420521392, "reference_active": true, "delta_db": -0.08746681353969829}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001235413948928314, "output_power": 0.00011473737276702972, "reference_active": true, "delta_db": -0.32107599328683184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6241583124510384e-05, "output_power": 2.790632638838907e-05, "reference_active": true, "delta_db": 0.2671263739053409}], "active_band_count": 3, "active_band_mae_db": 0.225223060243957}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011635584990380617, "log_rms_mae_db": 0.5661019735127862, "rms_envelope_nrmse": 0.05565039710952711, "envelope_correlation": 0.9945084335280017, "rms_gain_error_db": -0.03018384384493939, "reference_mean_power": 0.0024645680451820984, "output_mean_power": 0.002447498469352908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.594353848464586e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003918219819660693, "output_power": 0.00038791629178965866, "reference_active": true, "delta_db": -0.04350777523442407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.98714927012752e-05, "output_power": 5.5370669164119556e-05, "reference_active": true, "delta_db": -0.33940313935847016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.328965955891304e-06, "output_power": 1.362543315085364e-06, "reference_active": true, "delta_db": 0.10836461569132838}], "active_band_count": 3, "active_band_mae_db": 0.16375851009474088}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010777091024361473, "log_rms_mae_db": 2.1006719690859126, "rms_envelope_nrmse": 0.15266145835010317, "envelope_correlation": 0.9786615198762373, "rms_gain_error_db": -0.05477062171188818, "reference_mean_power": 0.0010415487104655132, "output_mean_power": 0.0010284958020167227}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.64672159263247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003552137892650671, "output_power": 0.0003338758076565365, "reference_active": true, "delta_db": -0.2690486497817097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.132330818484316e-05, "output_power": 6.36068807280719e-05, "reference_active": true, "delta_db": -0.49727380463893445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2237174874701015e-05, "output_power": 1.697989837529583e-05, "reference_active": true, "delta_db": 1.4225392031946995}], "active_band_count": 3, "active_band_mae_db": 0.7296205525384479}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.988831023316543e-05, "log_rms_mae_db": 1.285188412249265, "rms_envelope_nrmse": 0.08032970089994526, "envelope_correlation": 0.9932621230078192, "rms_gain_error_db": -0.05483512330987626, "reference_mean_power": 0.0010464471526710588, "output_mean_power": 0.0010333175089210346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.816907801707229e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003300382500563707, "output_power": 0.0003106011577107413, "reference_active": true, "delta_db": -0.2636120556228841}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.204099572239431e-05, "output_power": 2.9388166870821503e-05, "reference_active": true, "delta_db": -0.3753350680518735}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.619198216109287e-07, "output_power": 8.247038304301909e-07, "reference_active": true, "delta_db": -0.19168856619563923}], "active_band_count": 3, "active_band_mae_db": 0.2768785632901323}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5701896659278796e-05, "log_rms_mae_db": 2.1175362152776342, "rms_envelope_nrmse": 0.24334055300349744, "envelope_correlation": 0.9424344035428042, "rms_gain_error_db": -0.05800962738008256, "reference_mean_power": 0.0001796467459408233, "output_mean_power": 0.0001772631212982752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2519806248605502e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7558987061199216e-05, "output_power": 4.4745654143761266e-05, "reference_active": true, "delta_db": -0.26481734706665583}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1481328556544434e-05, "output_power": 9.098957723835152e-06, "reference_active": true, "delta_db": -1.010004979639067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593622469144843e-06, "output_power": 8.55836745432848e-06, "reference_active": true, "delta_db": -0.49591694967413624}], "active_band_count": 3, "active_band_mae_db": 0.590246425459953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.308273569685666e-05, "log_rms_mae_db": 1.307823189723148, "rms_envelope_nrmse": 0.14780749933339524, "envelope_correlation": 0.9757375471013222, "rms_gain_error_db": -0.0581148305704526, "reference_mean_power": 0.0001804894756897876, "output_mean_power": 0.00017809035527245617}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.07434863630674e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.420432318446027e-05, "output_power": 4.225424355115735e-05, "reference_active": true, "delta_db": -0.19594414209009176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8496256649417895e-06, "output_power": 4.259697582016911e-06, "reference_active": true, "delta_db": -1.3774930778372942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.792037562074947e-07, "output_power": 4.4010728234736755e-07, "reference_active": true, "delta_db": -1.1927281486449417}], "active_band_count": 3, "active_band_mae_db": 0.9220551228574426}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.825215268339244e-05, "log_rms_mae_db": 2.343369061574162, "rms_envelope_nrmse": 0.20518367416778058, "envelope_correlation": 0.9726872314509267, "rms_gain_error_db": -0.13411506204794377, "reference_mean_power": 0.0001267678324533094, "output_mean_power": 0.00012291292628271213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.277972501807574e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1540328186660286e-05, "output_power": 4.600782384708948e-05, "reference_active": true, "delta_db": -0.4931548784075224}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.007459575479445e-05, "output_power": 9.136973645862507e-06, "reference_active": true, "delta_db": -0.4242525674757453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.777776179485975e-06, "output_power": 3.4412388747870634e-06, "reference_active": true, "delta_db": -0.4052140349376465}], "active_band_count": 3, "active_band_mae_db": 0.44087382694030475}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.580061124327403e-05, "log_rms_mae_db": 1.7325245134017657, "rms_envelope_nrmse": 0.13840728603557412, "envelope_correlation": 0.9873287081138807, "rms_gain_error_db": -0.13436340838284577, "reference_mean_power": 0.0001273148740012305, "output_mean_power": 0.0001234362739802431}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.770926544449125e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2932423481689434e-05, "output_power": 4.6181396465270895e-05, "reference_active": true, "delta_db": -0.5925471709964881}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.69146274773148e-06, "output_power": 4.4067092262701975e-06, "reference_active": true, "delta_db": -0.27193877359629864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.856201910676535e-07, "output_power": 2.636675163546835e-07, "reference_active": true, "delta_db": -0.34732277016364327}], "active_band_count": 3, "active_band_mae_db": 0.40393623825214336}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7944348340454123, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bf08d7e70dda9cda4d126bb0d8f6a05a4ab79fc5a256d0754b4cc4800c7ff67a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9467069411815492, "standard_deviation": 0.005502387610050228, "minimum": 0.9403474764314664, "maximum": 0.9544435138004048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0645913422258912, "standard_deviation": 0.04481977006113083, "minimum": 1.9970852152653327, "maximum": 2.1209821769797217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5342373496330898, "standard_deviation": 0.1705103373191634, "minimum": 0.3104786900711966, "maximum": 0.8099297363724451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23367779635325556, "standard_deviation": 0.011906162075892516, "minimum": 0.21629843667426374, "maximum": 0.24677039719396168, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.978715932846476, "standard_deviation": 0.0024739533474050384, "minimum": 0.9757375471013222, "maximum": 0.9828073257704939, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2793846474804251, "standard_deviation": 0.05099742769486465, "minimum": 1.1849053002001848, "maximum": 1.3586750488841477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.771494297300135, "standard_deviation": 0.20333889086400433, "minimum": 0.4071611320956147, "maximum": 1.0103817253103, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13884350380892563, "standard_deviation": 0.007803417255845015, "minimum": 0.1255853414072195, "maximum": 0.14780749933339524, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9707870982678233, "standard_deviation": 0.0029778160843210374, "minimum": 0.9645390366983725, "maximum": 0.9735211324373464, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.32897237170789, "standard_deviation": 0.029446998409652744, "minimum": 2.282202532651523, "maximum": 2.3771755939699317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4485507590765225, "standard_deviation": 0.18546540063342365, "minimum": 0.20830366597505479, "maximum": 0.7689120458724901, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2144251154865715, "standard_deviation": 0.00924569092402951, "minimum": 0.20518367416778058, "maximum": 0.23320088014166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9862348115122173, "standard_deviation": 0.0015199343942973462, "minimum": 0.9835707400137499, "maximum": 0.9876411696606351, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7559727681521724, "standard_deviation": 0.026888319215684083, "minimum": 1.7321824602601774, "maximum": 1.7988184321195166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4968923964741046, "standard_deviation": 0.2189327021447602, "minimum": 0.194246460097843, "maximum": 0.8176303400004357, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14540103403298354, "standard_deviation": 0.00765229043767588, "minimum": 0.13779191366422727, "maximum": 0.15636901941615358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9765848424719947, "standard_deviation": 0.0026797311801730747, "minimum": 0.9735325496175834, "maximum": 0.9812757433330913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1400398620042886, "standard_deviation": 0.02974378207168763, "minimum": 2.1006719690859126, "maximum": 2.185761458600604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6970185717094274, "standard_deviation": 0.17124020053337963, "minimum": 0.47580186873977887, "maximum": 0.972363485545849, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1596645250542848, "standard_deviation": 0.009034928180526303, "minimum": 0.14376575492750698, "maximum": 0.1698661881422117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9916353992643985, "standard_deviation": 0.001350226212686951, "minimum": 0.9898323557633651, "maximum": 0.9932936677969187, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3574759984715894, "standard_deviation": 0.04090829663589866, "minimum": 1.285188412249265, "maximum": 1.4301285317069354, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.41751982369070006, "standard_deviation": 0.16062248920414443, "minimum": 0.24495064429766789, "maximum": 0.7328847367735641, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0901028896712061, "standard_deviation": 0.006929589543546546, "minimum": 0.08032970089994526, "maximum": 0.09917522217660768, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1369080707925595, "standard_deviation": 0.04376455806467638, "minimum": 0.06723202375335356, "maximum": 0.18210329016505442, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.9830292426639105, "standard_deviation": 0.010115220016097607, "minimum": 5.96872849310564, "maximum": 5.995297995274772, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4143726387349423, "standard_deviation": 0.02692974697696625, "minimum": 0.3881522796175974, "maximum": 0.46817686241750367, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.061255477555426374, "standard_deviation": 0.022810288010782987, "minimum": -0.10579955734912273, "maximum": -0.031029316485022638, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9800138484904473, "standard_deviation": 0.0011385453937929415, "minimum": 0.9786460121646492, "maximum": 0.9819608304108947, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2388961014142064, "standard_deviation": 0.026137371234299895, "minimum": 1.2025302542458276, "maximum": 1.2704700077253537, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2704495242288329, "standard_deviation": 0.09829208308668895, "minimum": 0.1266848413236884, "maximum": 0.4197533010542404, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11641141036427508, "standard_deviation": 0.0034017386773145513, "minimum": 0.11053156268313603, "maximum": 0.120539706852796, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9940913093354939, "standard_deviation": 0.0006754210552425282, "minimum": 0.9927054525274636, "maximum": 0.9947638030342917, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5770952321474, "standard_deviation": 0.025034539418980904, "minimum": 0.5531475103220018, "maximum": 0.6339151268354424, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2740307052044844, "standard_deviation": 0.08088953674464268, "minimum": 0.16375851009474088, "maximum": 0.4142740738441704, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05780883671423853, "standard_deviation": 0.0033132749404104654, "minimum": 0.0546257454349723, "maximum": 0.06436671017414437, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_61-70968-0054", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.4520106315612793, "predicted_mos": 2.4520106315612793, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7bc39c2e4def6fb770c7aeb2a4aba65e9b7afc2fa349b7a427a0a17e33b8ecf9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7501696003600955, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2183319487767388, "全帯域_10ms_rms_envelope_nrmse": 0.11673754589002083, "全帯域_10ms_envelope_correlation": 0.9799861095934443, "全帯域_10ms_modulation_mae_db": 0.3197480570007074, "全帯域_40ms_log_rms_mae_db": 0.5691098790596083, "全帯域_40ms_rms_envelope_nrmse": 0.054907248615661375, "全帯域_40ms_envelope_correlation": 0.9947291278806056, "全帯域_40ms_modulation_mae_db": 0.22002424000875406, "300–1000 Hz_10ms_log_rms_mae_db": 2.071073902273783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16611414749352163, "300–1000 Hz_10ms_envelope_correlation": 0.9747155812252132, "300–1000 Hz_10ms_modulation_mae_db": 0.7190555426260716, "300–1000 Hz_40ms_log_rms_mae_db": 1.3251388643039674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09680328312167195, "300–1000 Hz_40ms_envelope_correlation": 0.9904894875982739, "300–1000 Hz_40ms_modulation_mae_db": 0.57849032217534, "1000–3000 Hz_10ms_log_rms_mae_db": 2.066793987191947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23495712116812356, "1000–3000 Hz_10ms_envelope_correlation": 0.9459614145529898, "1000–3000 Hz_10ms_modulation_mae_db": 0.46858106873736016, "1000–3000 Hz_40ms_log_rms_mae_db": 1.26151788524736, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1372915667224915, "1000–3000 Hz_40ms_envelope_correlation": 0.9798377509157546, "1000–3000 Hz_40ms_modulation_mae_db": 0.8520894381016509, "3000–6000 Hz_10ms_log_rms_mae_db": 2.244689492973014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1892910444204064, "3000–6000 Hz_10ms_envelope_correlation": 0.9768891150227144, "3000–6000 Hz_10ms_modulation_mae_db": 0.8688995542362777, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7041282011171144, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13402792148010886, "3000–6000 Hz_40ms_envelope_correlation": 0.9879909368034673, "3000–6000 Hz_40ms_modulation_mae_db": 0.766864285100035, "mr_stft_db": 5.954434093248466, "band_mae_db": 0.16499091643130426, "rms_gain_error_db": -0.058450570918493754, "output_peak": 0.37082049530104516}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2451235763126463, "全帯域_10ms_rms_envelope_nrmse": 0.12006314338895654, "全帯域_10ms_envelope_correlation": 0.9787162152323772, "全帯域_10ms_modulation_mae_db": 0.5056754749919317, "全帯域_40ms_log_rms_mae_db": 0.635814764741552, "全帯域_40ms_rms_envelope_nrmse": 0.059758386708423025, "全帯域_40ms_envelope_correlation": 0.9937866749958074, "全帯域_40ms_modulation_mae_db": 0.5116015526846769, "300–1000 Hz_10ms_log_rms_mae_db": 2.070326262676577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1538656002566207, "300–1000 Hz_10ms_envelope_correlation": 0.9785330023059617, "300–1000 Hz_10ms_modulation_mae_db": 0.6528567279372303, "300–1000 Hz_40ms_log_rms_mae_db": 1.2882506031371488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08558724240271469, "300–1000 Hz_40ms_envelope_correlation": 0.9929353560582941, "300–1000 Hz_40ms_modulation_mae_db": 0.6027647279951442, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0754390874323856, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23203742679181313, "1000–3000 Hz_10ms_envelope_correlation": 0.9473710832677752, "1000–3000 Hz_10ms_modulation_mae_db": 0.7699344792779805, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2566463312514577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.132786478359509, "1000–3000 Hz_40ms_envelope_correlation": 0.9812917275792177, "1000–3000 Hz_40ms_modulation_mae_db": 1.0708386284248104, "3000–6000 Hz_10ms_log_rms_mae_db": 2.267742891918385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22391682413515981, "3000–6000 Hz_10ms_envelope_correlation": 0.9673696550133424, "3000–6000 Hz_10ms_modulation_mae_db": 0.18710944544508223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7768910534064508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16058945585987935, "3000–6000 Hz_40ms_envelope_correlation": 0.9825471865226718, "3000–6000 Hz_40ms_modulation_mae_db": 0.31591168806479725, "mr_stft_db": 5.937262001716043, "band_mae_db": 0.208101159915668, "rms_gain_error_db": -0.10272110521642233, "output_peak": 0.3869529106948561}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2182656814847648, "全帯域_10ms_rms_envelope_nrmse": 0.11266574459165468, "全帯域_10ms_envelope_correlation": 0.9813318490727418, "全帯域_10ms_modulation_mae_db": 0.4681898455243178, "全帯域_40ms_log_rms_mae_db": 0.5684263992897256, "全帯域_40ms_rms_envelope_nrmse": 0.055373147152283496, "全帯域_40ms_envelope_correlation": 0.9945822310707583, "全帯域_40ms_modulation_mae_db": 0.09848819873641816, "300–1000 Hz_10ms_log_rms_mae_db": 2.0809704886293834, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14976679216472735, "300–1000 Hz_10ms_envelope_correlation": 0.9794898365112873, "300–1000 Hz_10ms_modulation_mae_db": 0.7493847157705531, "300–1000 Hz_40ms_log_rms_mae_db": 1.291471674654614, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0744044715521356, "300–1000 Hz_40ms_envelope_correlation": 0.9944857734005702, "300–1000 Hz_40ms_modulation_mae_db": 0.45708336890531515, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9969743581219797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2241043430328991, "1000–3000 Hz_10ms_envelope_correlation": 0.9509793002793726, "1000–3000 Hz_10ms_modulation_mae_db": 1.0656966241038752, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2260699530106798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13365587082242156, "1000–3000 Hz_40ms_envelope_correlation": 0.9806886097021109, "1000–3000 Hz_40ms_modulation_mae_db": 1.5318257137302915, "3000–6000 Hz_10ms_log_rms_mae_db": 2.207721597978667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18740993373101625, "3000–6000 Hz_10ms_envelope_correlation": 0.9775095853099927, "3000–6000 Hz_10ms_modulation_mae_db": 0.5263679348157476, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6318111499567287, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13406908692832503, "3000–6000 Hz_40ms_envelope_correlation": 0.9880915166361139, "3000–6000 Hz_40ms_modulation_mae_db": 0.4992391460501416, "mr_stft_db": 5.941737383219874, "band_mae_db": 0.12429247351151214, "rms_gain_error_db": -0.06303817424986691, "output_peak": 0.4313641971940303}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.183045986995994, "全帯域_10ms_rms_envelope_nrmse": 0.1135253797669462, "全帯域_10ms_envelope_correlation": 0.9810261454361506, "全帯域_10ms_modulation_mae_db": 0.1382126151622006, "全帯域_40ms_log_rms_mae_db": 0.5360797696522162, "全帯域_40ms_rms_envelope_nrmse": 0.056270679545412434, "全帯域_40ms_envelope_correlation": 0.9945240370082186, "全帯域_40ms_modulation_mae_db": 0.14314168138618735, "300–1000 Hz_10ms_log_rms_mae_db": 1.9187138611835017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14575271891459615, "300–1000 Hz_10ms_envelope_correlation": 0.980631104239641, "300–1000 Hz_10ms_modulation_mae_db": 0.49786270307344394, "300–1000 Hz_40ms_log_rms_mae_db": 1.1620973464695514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07973716605049457, "300–1000 Hz_40ms_envelope_correlation": 0.9936652629958547, "300–1000 Hz_40ms_modulation_mae_db": 0.5588161647119317, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0615085274622325, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23375197606823667, "1000–3000 Hz_10ms_envelope_correlation": 0.9465195412173161, "1000–3000 Hz_10ms_modulation_mae_db": 0.7820257274086967, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2699339591134886, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.136991952795694, "1000–3000 Hz_40ms_envelope_correlation": 0.9798025685623987, "1000–3000 Hz_40ms_modulation_mae_db": 1.1385148751195235, "3000–6000 Hz_10ms_log_rms_mae_db": 2.247177042630242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21373704677920527, "3000–6000 Hz_10ms_envelope_correlation": 0.970935449531557, "3000–6000 Hz_10ms_modulation_mae_db": 0.4961666748195193, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6746661461705825, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14660798466603256, "3000–6000 Hz_40ms_envelope_correlation": 0.9856991394160095, "3000–6000 Hz_40ms_modulation_mae_db": 0.2745426139254676, "mr_stft_db": 5.95457524923435, "band_mae_db": 0.17694313434003014, "rms_gain_error_db": -0.10218247965560487, "output_peak": 0.39178452215540227}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1978128471602631, "全帯域_10ms_rms_envelope_nrmse": 0.11714173103237209, "全帯域_10ms_envelope_correlation": 0.9798077440872794, "全帯域_10ms_modulation_mae_db": 0.2923314658641048, "全帯域_40ms_log_rms_mae_db": 0.5884670030796246, "全帯域_40ms_rms_envelope_nrmse": 0.05754911631203935, "全帯域_40ms_envelope_correlation": 0.9942727158695117, "全帯域_40ms_modulation_mae_db": 0.5141362113442449, "300–1000 Hz_10ms_log_rms_mae_db": 2.14410073359075, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17968142662361966, "300–1000 Hz_10ms_envelope_correlation": 0.9703293417024099, "300–1000 Hz_10ms_modulation_mae_db": 0.8004461884858903, "300–1000 Hz_40ms_log_rms_mae_db": 1.404058189253771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10677814590598642, "300–1000 Hz_40ms_envelope_correlation": 0.9883832264444509, "300–1000 Hz_40ms_modulation_mae_db": 0.44086463869263165, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0162428525956058, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22833891593873873, "1000–3000 Hz_10ms_envelope_correlation": 0.9491317955002049, "1000–3000 Hz_10ms_modulation_mae_db": 0.43062121532520686, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2384237138594796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13974593763215656, "1000–3000 Hz_40ms_envelope_correlation": 0.9792736175757814, "1000–3000 Hz_40ms_modulation_mae_db": 0.6561797026594712, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2923041505868147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21144049284047067, "3000–6000 Hz_10ms_envelope_correlation": 0.9710832009745045, "3000–6000 Hz_10ms_modulation_mae_db": 0.1759224656980679, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6833713557981962, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14266057603700746, "3000–6000 Hz_40ms_envelope_correlation": 0.9862720174367219, "3000–6000 Hz_40ms_modulation_mae_db": 0.18856547490760225, "mr_stft_db": 5.94249599731966, "band_mae_db": 0.1907710229636113, "rms_gain_error_db": -0.10137913931014977, "output_peak": 0.4127536803420025}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2606647464612526, "全帯域_10ms_rms_envelope_nrmse": 0.11184470125403478, "全帯域_10ms_envelope_correlation": 0.9816404713410632, "全帯域_10ms_modulation_mae_db": 0.16209337220813433, "全帯域_40ms_log_rms_mae_db": 0.595893120800117, "全帯域_40ms_rms_envelope_nrmse": 0.05609507279576036, "全帯域_40ms_envelope_correlation": 0.9947362939161708, "全帯域_40ms_modulation_mae_db": 0.1857595757943311, "300–1000 Hz_10ms_log_rms_mae_db": 2.0746625706299815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15378121908778059, "300–1000 Hz_10ms_envelope_correlation": 0.9785095486450637, "300–1000 Hz_10ms_modulation_mae_db": 0.5187554690407831, "300–1000 Hz_40ms_log_rms_mae_db": 1.3151467776990244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0866221195127408, "300–1000 Hz_40ms_envelope_correlation": 0.9926791779175133, "300–1000 Hz_40ms_modulation_mae_db": 0.7451204690231606, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0358123434653534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.225782286605404, "1000–3000 Hz_10ms_envelope_correlation": 0.9502381995039775, "1000–3000 Hz_10ms_modulation_mae_db": 0.40452913379863453, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2507014942303076, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13877462470393673, "1000–3000 Hz_40ms_envelope_correlation": 0.9787173723613758, "1000–3000 Hz_40ms_modulation_mae_db": 0.6902059518580145, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2938505355438417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1955843779101323, "3000–6000 Hz_10ms_envelope_correlation": 0.975531337813041, "3000–6000 Hz_10ms_modulation_mae_db": 0.18247341784385526, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7289443691955775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1315874085082824, "3000–6000 Hz_40ms_envelope_correlation": 0.9885482673934894, "3000–6000 Hz_40ms_modulation_mae_db": 0.26829759394738656, "mr_stft_db": 5.95021835897143, "band_mae_db": 0.15133698144194155, "rms_gain_error_db": -0.13554998568455626, "output_peak": 0.4163765241197209}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2052039572602016, "全帯域_10ms_rms_envelope_nrmse": 0.11451671162119983, "全帯域_10ms_envelope_correlation": 0.9806669082836136, "全帯域_10ms_modulation_mae_db": 0.5131476341192668, "全帯域_40ms_log_rms_mae_db": 0.5689861753004919, "全帯域_40ms_rms_envelope_nrmse": 0.05393920196661872, "全帯域_40ms_envelope_correlation": 0.9950018256511289, "全帯域_40ms_modulation_mae_db": 0.12566522083071432, "300–1000 Hz_10ms_log_rms_mae_db": 2.038793299904222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16005533453690038, "300–1000 Hz_10ms_envelope_correlation": 0.9767227688405936, "300–1000 Hz_10ms_modulation_mae_db": 1.1355291605673044, "300–1000 Hz_40ms_log_rms_mae_db": 1.2903256370130378, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07958082543941364, "300–1000 Hz_40ms_envelope_correlation": 0.9941109335676835, "300–1000 Hz_40ms_modulation_mae_db": 0.5100585275683084, "1000–3000 Hz_10ms_log_rms_mae_db": 2.061738561346956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23839073712835565, "1000–3000 Hz_10ms_envelope_correlation": 0.9443469746854036, "1000–3000 Hz_10ms_modulation_mae_db": 0.6829471532975709, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2617079143740235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13964528206346669, "1000–3000 Hz_40ms_envelope_correlation": 0.9786032375884768, "1000–3000 Hz_40ms_modulation_mae_db": 1.130505632595163, "3000–6000 Hz_10ms_log_rms_mae_db": 2.256362606008402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21213738189545564, "3000–6000 Hz_10ms_envelope_correlation": 0.9713251607686525, "3000–6000 Hz_10ms_modulation_mae_db": 0.9402342779088158, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6980589704640818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1412371898927907, "3000–6000 Hz_40ms_envelope_correlation": 0.9869069197807472, "3000–6000 Hz_40ms_modulation_mae_db": 0.9183088438194802, "mr_stft_db": 5.9427660242404405, "band_mae_db": 0.13425171943558512, "rms_gain_error_db": -0.10221410543716805, "output_peak": 0.40043364316072355}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.233789534585515, "全帯域_10ms_rms_envelope_nrmse": 0.1154178739359378, "全帯域_10ms_envelope_correlation": 0.9803524691613139, "全帯域_10ms_modulation_mae_db": 0.49936434353544074, "全帯域_40ms_log_rms_mae_db": 0.568910301377579, "全帯域_40ms_rms_envelope_nrmse": 0.05553292732668284, "全帯域_40ms_envelope_correlation": 0.9946269059557161, "全帯域_40ms_modulation_mae_db": 0.12819462432986103, "300–1000 Hz_10ms_log_rms_mae_db": 2.1144855087678405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1663532360354424, "300–1000 Hz_10ms_envelope_correlation": 0.9747382188254086, "300–1000 Hz_10ms_modulation_mae_db": 1.4275183364825583, "300–1000 Hz_40ms_log_rms_mae_db": 1.3356694578428385, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0961283549609386, "300–1000 Hz_40ms_envelope_correlation": 0.9908656168134784, "300–1000 Hz_40ms_modulation_mae_db": 0.864759298680621, "1000–3000 Hz_10ms_log_rms_mae_db": 2.065380890395905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23036910132366478, "1000–3000 Hz_10ms_envelope_correlation": 0.9480835874103625, "1000–3000 Hz_10ms_modulation_mae_db": 0.9822635768984656, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2806599908961063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13589432004537091, "1000–3000 Hz_40ms_envelope_correlation": 0.9799344735385452, "1000–3000 Hz_40ms_modulation_mae_db": 1.4695115220126278, "3000–6000 Hz_10ms_log_rms_mae_db": 2.233535710406638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21221417676863513, "3000–6000 Hz_10ms_envelope_correlation": 0.9714871338236764, "3000–6000 Hz_10ms_modulation_mae_db": 0.6940732524641291, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7190229845045444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15243739842896123, "3000–6000 Hz_40ms_envelope_correlation": 0.9846521079682016, "3000–6000 Hz_40ms_modulation_mae_db": 0.6437711386902625, "mr_stft_db": 5.95001040951744, "band_mae_db": 0.20342795624990173, "rms_gain_error_db": -0.09037800030659253, "output_peak": 0.40379842907235497}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.258553454190378, "spectral_convergence": 0.41820074825695874, "spectral_convergence_status": "ok", "amplitude_floor": 1.3149451373049662e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.738920473809087, "spectral_convergence": 0.35820748969372557, "spectral_convergence_status": "ok", "amplitude_floor": 7.412465414213296e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8658283517459315, "spectral_convergence": 0.39387994402916504, "spectral_convergence_status": "ok", "amplitude_floor": 4.876012126794439e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006137697087039973, "output_power": 0.0006384405272691399, "delta_db": 0.17114996648993588, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009615845769051578, "output_power": 0.0009388030760114241, "delta_db": -0.10412985051557466, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000542394172591287, "output_power": 0.0005226995800120776, "delta_db": -0.1606286374036905, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013703842807085808, "output_power": 0.00012526540928425463, "delta_db": -0.39011206587715613, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.620351571877408e-05, "output_power": 8.308615120147163e-05, "delta_db": -0.15996336806738898, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011238415255857088, "output_power": 0.00011228168309957912, "delta_db": -0.003961610234079477, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16499091643130426, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.058450570918493754, "pitch": null, "output_rms_dbfs": -26.160812004647937, "output_peak": 0.37082049530104516, "reference_rms_dbfs": -26.102361433729442, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.954434093248466}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013602681282863616, "log_rms_mae_db": 1.2183319487767388, "rms_envelope_nrmse": 0.11673754589002083, "envelope_correlation": 0.9799861095934443, "rms_gain_error_db": -0.05830841486949466, "reference_mean_power": 0.0024543453947075094, "output_mean_power": 0.0024216135527921853}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.153929939453048e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000440684439069062, "output_power": 0.0004330289652275291, "reference_active": true, "delta_db": -0.07610768281697139}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001235413948928314, "output_power": 0.00011881262197386679, "reference_active": true, "delta_db": -0.16949920542955443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6241583124510384e-05, "output_power": 2.2265184779414965e-05, "reference_active": true, "delta_db": -0.7136372827555965}], "active_band_count": 3, "active_band_mae_db": 0.3197480570007074}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011635584990380617, "log_rms_mae_db": 0.5691098790596083, "rms_envelope_nrmse": 0.054907248615661375, "envelope_correlation": 0.9947291278806056, "rms_gain_error_db": -0.05821711986317643, "reference_mean_power": 0.0024645680451820984, "output_mean_power": 0.002431750989518456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.594353848464586e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003918219819660693, "output_power": 0.00039013844028493676, "reference_active": true, "delta_db": -0.01870053703673636}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.98714927012752e-05, "output_power": 5.7309257292555937e-05, "reference_active": true, "delta_db": -0.18995305810154522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.328965955891304e-06, "output_power": 1.197765896070182e-06, "reference_active": true, "delta_db": -0.4514191248879806}], "active_band_count": 3, "active_band_mae_db": 0.22002424000875406}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010777091024361473, "log_rms_mae_db": 2.071073902273783, "rms_envelope_nrmse": 0.16611414749352163, "envelope_correlation": 0.9747155812252132, "rms_gain_error_db": -0.1887937255944695, "reference_mean_power": 0.0010415487104655132, "output_mean_power": 0.0009972412008347627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.64672159263247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003552137892650671, "output_power": 0.0003347837002448, "reference_active": true, "delta_db": -0.2572551142412323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.132330818484316e-05, "output_power": 5.5465145145695456e-05, "reference_active": true, "delta_db": -1.0921132500258743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2237174874701015e-05, "output_power": 1.4738761549846185e-05, "reference_active": true, "delta_db": 0.8077982636111082}], "active_band_count": 3, "active_band_mae_db": 0.7190555426260716}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.988831023316543e-05, "log_rms_mae_db": 1.3251388643039674, "rms_envelope_nrmse": 0.09680328312167195, "envelope_correlation": 0.9904894875982739, "rms_gain_error_db": -0.188876040176829, "reference_mean_power": 0.0010464471526710588, "output_mean_power": 0.001001912273122094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.816907801707229e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003300382500563707, "output_power": 0.00031188302190488166, "reference_active": true, "delta_db": -0.24572542167393174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.204099572239431e-05, "output_power": 2.524261246198687e-05, "reference_active": true, "delta_db": -1.0357170415591717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.619198216109287e-07, "output_power": 9.569069715467228e-07, "reference_active": true, "delta_db": 0.45402850329291655}], "active_band_count": 3, "active_band_mae_db": 0.57849032217534}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5701896659278796e-05, "log_rms_mae_db": 2.066793987191947, "rms_envelope_nrmse": 0.23495712116812356, "envelope_correlation": 0.9459614145529898, "rms_gain_error_db": -0.2920526517558133, "reference_mean_power": 0.0001796467459408233, "output_mean_power": 0.00016796318052442874}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2519806248605502e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7558987061199216e-05, "output_power": 4.2652911450478e-05, "reference_active": true, "delta_db": -0.4728391511289174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1481328556544434e-05, "output_power": 1.14322557719896e-05, "reference_active": true, "delta_db": -0.01860212860800146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593622469144843e-06, "output_power": 7.772351468867639e-06, "reference_active": true, "delta_db": -0.9143019264751617}], "active_band_count": 3, "active_band_mae_db": 0.46858106873736016}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.308273569685666e-05, "log_rms_mae_db": 1.26151788524736, "rms_envelope_nrmse": 0.1372915667224915, "envelope_correlation": 0.9798377509157546, "rms_gain_error_db": -0.2917568085522447, "reference_mean_power": 0.0001804894756897876, "output_mean_power": 0.00016876259800829748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.07434863630674e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.420432318446027e-05, "output_power": 4.030286872436756e-05, "reference_active": true, "delta_db": -0.4012878548474744}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8496256649417895e-06, "output_power": 5.570360215821392e-06, "reference_active": true, "delta_db": -0.2124479477185704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.792037562074947e-07, "output_power": 3.703208106586626e-07, "reference_active": true, "delta_db": -1.9425325117389078}], "active_band_count": 3, "active_band_mae_db": 0.8520894381016509}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.825215268339244e-05, "log_rms_mae_db": 2.244689492973014, "rms_envelope_nrmse": 0.1892910444204064, "envelope_correlation": 0.9768891150227144, "rms_gain_error_db": -0.047199022876020715, "reference_mean_power": 0.0001267678324533094, "output_mean_power": 0.00012539758202235093}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.277972501807574e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1540328186660286e-05, "output_power": 4.936165528094869e-05, "reference_active": true, "delta_db": -0.1875746483639315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.007459575479445e-05, "output_power": 9.641911194398778e-06, "reference_active": true, "delta_db": -0.1906450185713832}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.777776179485975e-06, "output_power": 2.261456947544557e-06, "reference_active": true, "delta_db": -2.2284789957735187}], "active_band_count": 3, "active_band_mae_db": 0.8688995542362777}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.580061124327403e-05, "log_rms_mae_db": 1.7041282011171144, "rms_envelope_nrmse": 0.13402792148010886, "envelope_correlation": 0.9879909368034673, "rms_gain_error_db": -0.046738023368456054, "reference_mean_power": 0.0001273148740012305, "output_mean_power": 0.0001259520795103269}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.770926544449125e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2932423481689434e-05, "output_power": 4.8923141706886174e-05, "reference_active": true, "delta_db": -0.3420743994596182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.69146274773148e-06, "output_power": 4.366390126597986e-06, "reference_active": true, "delta_db": -0.3118573589885444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.856201910676535e-07, "output_power": 1.9548921184018229e-07, "reference_active": true, "delta_db": -1.6466610968519424}], "active_band_count": 3, "active_band_mae_db": 0.766864285100035}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8037607808610739, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "457a86377cdd9a3f8746701fc9c305d9fe37a2d6a4cf8230f905000842aa58c3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9478289870521752, "standard_deviation": 0.0022369453245212486, "minimum": 0.9443469746854036, "maximum": 0.9509793002793726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0474863260015455, "standard_deviation": 0.028127266463728248, "minimum": 1.9969743581219797, "maximum": 2.0754390874323856, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6983248723559738, "standard_deviation": 0.2503474255584933, "minimum": 0.40452913379863453, "maximum": 1.0656966241038752, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23096648850715445, "standard_deviation": 0.004798459057774058, "minimum": 0.2241043430328991, "maximum": 0.23839073712835565, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9797686697279576, "standard_deviation": 0.0009186902071902204, "minimum": 0.9786032375884768, "maximum": 0.9812917275792177, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2557076552478628, "standard_deviation": 0.017311595795049626, "minimum": 1.2260699530106798, "maximum": 1.2806599908961063, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.067458933062694, "standard_deviation": 0.326052687105683, "minimum": 0.6561797026594712, "maximum": 1.5318257137302915, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13684825414313087, "standard_deviation": 0.0026117136607065033, "minimum": 0.132786478359509, "maximum": 0.13974593763215656, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9727663297821851, "standard_deviation": 0.0035084297385863644, "minimum": 0.9673696550133424, "maximum": 0.9775095853099927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2554230035057508, "standard_deviation": 0.029098940922060324, "minimum": 2.207721597978667, "maximum": 2.2938505355438417, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5089058779039369, "standard_deviation": 0.30969636067485784, "minimum": 0.1759224656980679, "maximum": 0.9402342779088158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20571640981006017, "standard_deviation": 0.013198368127169133, "minimum": 0.18740993373101625, "maximum": 0.22391682413515981, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9863385114946779, "standard_deviation": 0.0020227620985469514, "minimum": 0.9825471865226718, "maximum": 0.9885482673934894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7021117788266595, "standard_deviation": 0.04253578750589865, "minimum": 1.6318111499567287, "maximum": 1.7768910534064508, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.48443759806314657, "standard_deviation": 0.26734917130519675, "minimum": 0.18856547490760225, "maximum": 0.9183088438194802, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14290212772517347, "standard_deviation": 0.01003653128728786, "minimum": 0.1315874085082824, "maximum": 0.16058945585987935, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9767086752869474, "standard_deviation": 0.003343284073776683, "minimum": 0.9703293417024099, "maximum": 0.980631104239641, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.064140828457005, "standard_deviation": 0.06676177788825656, "minimum": 1.9187138611835017, "maximum": 2.14410073359075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8126761054979794, "standard_deviation": 0.31763057993554095, "minimum": 0.49786270307344394, "maximum": 1.4275183364825583, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1594213093891511, "standard_deviation": 0.01099634106765876, "minimum": 0.14575271891459615, "maximum": 0.17968142662361966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9922018543495148, "standard_deviation": 0.0021067221264103823, "minimum": 0.9883832264444509, "maximum": 0.9944857734005702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3015198187967443, "standard_deviation": 0.06789155361216367, "minimum": 1.1620973464695514, "maximum": 1.404058189253771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5947446897190567, "standard_deviation": 0.14488498091011398, "minimum": 0.44086463869263165, "maximum": 0.864759298680621, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08820520111826204, "standard_deviation": 0.010873067148839427, "minimum": 0.0744044715521356, "maximum": 0.10677814590598642, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16926442053619428, "standard_deviation": 0.031133255667274585, "minimum": 0.12429247351151214, "maximum": 0.208101159915668, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.946687439683464, "standard_deviation": 0.00645893985705251, "minimum": 5.937262001716043, "maximum": 5.95457524923435, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40178555025501694, "standard_deviation": 0.018871121066958478, "minimum": 0.37082049530104516, "maximum": 0.4313641971940303, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09448919509735682, "standard_deviation": 0.024567505059163313, "minimum": -0.13554998568455626, "maximum": -0.058450570918493754, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.980440989025998, "standard_deviation": 0.000942370251661878, "minimum": 0.9787162152323772, "maximum": 0.9816404713410632, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.220279784879672, "standard_deviation": 0.025514485195492462, "minimum": 1.183045986995994, "maximum": 1.2606647464612526, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36234535105076304, "standard_deviation": 0.15598302390514013, "minimum": 0.1382126151622006, "maximum": 0.5131476341192668, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11523910393514034, "standard_deviation": 0.0026931522311363157, "minimum": 0.11184470125403478, "maximum": 0.12006314338895654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9945324765434898, "standard_deviation": 0.00036560342752796706, "minimum": 0.9937866749958074, "maximum": 0.9950018256511289, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5789609266626143, "standard_deviation": 0.028925390125078165, "minimum": 0.5360797696522162, "maximum": 0.635814764741552, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24087641313939848, "standard_deviation": 0.17206019840833278, "minimum": 0.09848819873641816, "maximum": 0.5141362113442449, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0561782225528602, "standard_deviation": 0.0017890750811324519, "minimum": 0.05393920196661872, "maximum": 0.059758386708423025, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_672-122797-0003", "language": "en", "style": "normal", "speaker": "librispeech_672", "source_utterance_id": "672-122797-0003", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 76160, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_672-122797-0003.wav", "source_file_sha256": "7d997da2aa6341e9b9347500f0716467dece28b400bc5c5052a4320302b2f66d", "decoded_pcm_sha256": "4d9f339e6dbd2372c553d9d5f6718849a37e2770b2e7e3a43beccd5922c086ec", "original_source_sha256": "b00cbf4705fcd94f9ea9130da79993f83b7e3a83313328314b3492d19295af94", "corpus_entry_id": "librispeech_672-122797-0003"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.56923770904541, "predicted_mos": 3.56923770904541, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "84117bdcd6486d7427c042e289684276fa8699885b4a95eb49400d64454c0426", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0358501499977137, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_672-122797-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "84117bdcd6486d7427c042e289684276fa8699885b4a95eb49400d64454c0426", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "cdb8e35cf9303bf26bc3f9cba71892434576732b577ad73f2aca4b6dcec6f2ac"}, "rows": [{"case": "librispeech_672-122797-0003", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.465224504470825, "predicted_mos": 2.465224504470825, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "991f8d93e3b23f55c9006e7b8880430604b3ce2afc5b76e495895b9660d51d47", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1316829659990617, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_672-122797-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "991f8d93e3b23f55c9006e7b8880430604b3ce2afc5b76e495895b9660d51d47", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6315914712365436, "全帯域_10ms_rms_envelope_nrmse": 0.1438352602539658, "全帯域_10ms_envelope_correlation": 0.9818948517306347, "全帯域_10ms_modulation_mae_db": 0.5700798863549013, "全帯域_40ms_log_rms_mae_db": 0.9003726918324706, "全帯域_40ms_rms_envelope_nrmse": 0.07865436480017308, "全帯域_40ms_envelope_correlation": 0.9941898567549331, "全帯域_40ms_modulation_mae_db": 0.49542069446551656, "300–1000 Hz_10ms_log_rms_mae_db": 2.0556545030530815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16072054924783905, "300–1000 Hz_10ms_envelope_correlation": 0.9821933904972974, "300–1000 Hz_10ms_modulation_mae_db": 0.735587828017357, "300–1000 Hz_40ms_log_rms_mae_db": 1.2132512638279833, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07606185103410809, "300–1000 Hz_40ms_envelope_correlation": 0.9962253374887305, "300–1000 Hz_40ms_modulation_mae_db": 0.7327760823060044, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7214574370885334, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1917360382456516, "1000–3000 Hz_10ms_envelope_correlation": 0.9735287783752048, "1000–3000 Hz_10ms_modulation_mae_db": 1.7393057724839214, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0525708120832535, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11879267816563359, "1000–3000 Hz_40ms_envelope_correlation": 0.9907123911134101, "1000–3000 Hz_40ms_modulation_mae_db": 1.9192620181240008, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5997828508612388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18022061763233865, "3000–6000 Hz_10ms_envelope_correlation": 0.9820346350300168, "3000–6000 Hz_10ms_modulation_mae_db": 0.4223402215956233, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9815108632953092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10248023940154441, "3000–6000 Hz_40ms_envelope_correlation": 0.9941522492947722, "3000–6000 Hz_40ms_modulation_mae_db": 0.5343434784976032, "mr_stft_db": 6.024687947632135, "band_mae_db": 0.23715282331964693, "rms_gain_error_db": -0.04445928695913867, "output_peak": 0.3392525616066561}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.584736693061089, "全帯域_10ms_rms_envelope_nrmse": 0.16505098619393366, "全帯域_10ms_envelope_correlation": 0.9766392445451211, "全帯域_10ms_modulation_mae_db": 1.5953598879306916, "全帯域_40ms_log_rms_mae_db": 0.9025658370175549, "全帯域_40ms_rms_envelope_nrmse": 0.11168238592658163, "全帯域_40ms_envelope_correlation": 0.9895622072682472, "全帯域_40ms_modulation_mae_db": 1.954916665946305, "300–1000 Hz_10ms_log_rms_mae_db": 2.1988928152999696, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1612010409480875, "300–1000 Hz_10ms_envelope_correlation": 0.9819337470993366, "300–1000 Hz_10ms_modulation_mae_db": 0.5420117833663003, "300–1000 Hz_40ms_log_rms_mae_db": 1.3521685656363895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07621936354763016, "300–1000 Hz_40ms_envelope_correlation": 0.9958069685337293, "300–1000 Hz_40ms_modulation_mae_db": 1.0341695479845388, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7045882897481883, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20885077822182022, "1000–3000 Hz_10ms_envelope_correlation": 0.9677178134151136, "1000–3000 Hz_10ms_modulation_mae_db": 1.503507562206978, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0050812363362132, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11641323650614667, "1000–3000 Hz_40ms_envelope_correlation": 0.9907362640687568, "1000–3000 Hz_40ms_modulation_mae_db": 1.8173128971586643, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6558496097847286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.215583618898896, "3000–6000 Hz_10ms_envelope_correlation": 0.9806475682124608, "3000–6000 Hz_10ms_modulation_mae_db": 2.453813923139268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1686657084451408, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16492732780241912, "3000–6000 Hz_40ms_envelope_correlation": 0.9929797714913321, "3000–6000 Hz_40ms_modulation_mae_db": 2.5895576610171727, "mr_stft_db": 6.030730965463047, "band_mae_db": 0.510791151851329, "rms_gain_error_db": -0.38992140791501323, "output_peak": 0.35927395382099764}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6343313565443758, "全帯域_10ms_rms_envelope_nrmse": 0.15446890289590678, "全帯域_10ms_envelope_correlation": 0.9793136116241496, "全帯域_10ms_modulation_mae_db": 1.906955713596645, "全帯域_40ms_log_rms_mae_db": 0.9442810539633502, "全帯域_40ms_rms_envelope_nrmse": 0.10678192253631022, "全帯域_40ms_envelope_correlation": 0.9900241000984372, "全帯域_40ms_modulation_mae_db": 2.4028216153724964, "300–1000 Hz_10ms_log_rms_mae_db": 2.14104576503884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17837960288913374, "300–1000 Hz_10ms_envelope_correlation": 0.9782442509697836, "300–1000 Hz_10ms_modulation_mae_db": 0.40741947063465006, "300–1000 Hz_40ms_log_rms_mae_db": 1.246333200703171, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09287310377686128, "300–1000 Hz_40ms_envelope_correlation": 0.9946915851158058, "300–1000 Hz_40ms_modulation_mae_db": 0.6360032670530277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7267062695454907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20019763779872526, "1000–3000 Hz_10ms_envelope_correlation": 0.9700876141226242, "1000–3000 Hz_10ms_modulation_mae_db": 1.29985742570408, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0750362102533215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10929341768494154, "1000–3000 Hz_40ms_envelope_correlation": 0.9913066883781542, "1000–3000 Hz_40ms_modulation_mae_db": 1.6958330045567072, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5695784922119065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21460785423867165, "3000–6000 Hz_10ms_envelope_correlation": 0.9754200601397714, "3000–6000 Hz_10ms_modulation_mae_db": 3.181462616718394, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0420736946068838, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16359680672476734, "3000–6000 Hz_40ms_envelope_correlation": 0.9869387416701454, "3000–6000 Hz_40ms_modulation_mae_db": 3.762697988763197, "mr_stft_db": 6.026838911520581, "band_mae_db": 0.28989486057214703, "rms_gain_error_db": -0.3177835888454022, "output_peak": 0.4081113121311869}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6588123214935042, "全帯域_10ms_rms_envelope_nrmse": 0.17140954194609564, "全帯域_10ms_envelope_correlation": 0.9738836196576995, "全帯域_10ms_modulation_mae_db": 1.6207242565739233, "全帯域_40ms_log_rms_mae_db": 0.8748278545308733, "全帯域_40ms_rms_envelope_nrmse": 0.0954107907929628, "全帯域_40ms_envelope_correlation": 0.9914519482789877, "全帯域_40ms_modulation_mae_db": 1.8938875456477484, "300–1000 Hz_10ms_log_rms_mae_db": 2.116012245088963, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18401765198983208, "300–1000 Hz_10ms_envelope_correlation": 0.9764216150147719, "300–1000 Hz_10ms_modulation_mae_db": 0.5746400801379589, "300–1000 Hz_40ms_log_rms_mae_db": 1.210000895603217, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08931950752188811, "300–1000 Hz_40ms_envelope_correlation": 0.9944543384880222, "300–1000 Hz_40ms_modulation_mae_db": 0.5096010020162026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7101492483098795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19225196290038835, "1000–3000 Hz_10ms_envelope_correlation": 0.9724101161940666, "1000–3000 Hz_10ms_modulation_mae_db": 1.046985516271364, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0130754876732386, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10443426535262192, "1000–3000 Hz_40ms_envelope_correlation": 0.9919243123489971, "1000–3000 Hz_40ms_modulation_mae_db": 1.3602914454667372, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6178000391369878, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22420164895717057, "3000–6000 Hz_10ms_envelope_correlation": 0.9711635330593477, "3000–6000 Hz_10ms_modulation_mae_db": 2.5333802758491415, "3000–6000 Hz_40ms_log_rms_mae_db": 1.103116838921171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13848841088652855, "3000–6000 Hz_40ms_envelope_correlation": 0.9891513270174416, "3000–6000 Hz_40ms_modulation_mae_db": 2.8663033821928767, "mr_stft_db": 6.023348625635691, "band_mae_db": 0.18942262139701838, "rms_gain_error_db": -0.1541326377292775, "output_peak": 0.45775674959782897}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6002931286087567, "全帯域_10ms_rms_envelope_nrmse": 0.1678230136198193, "全帯域_10ms_envelope_correlation": 0.9764054819383579, "全帯域_10ms_modulation_mae_db": 1.149749805948004, "全帯域_40ms_log_rms_mae_db": 0.8963800411069562, "全帯域_40ms_rms_envelope_nrmse": 0.11553783318274859, "全帯域_40ms_envelope_correlation": 0.9880815102292525, "全帯域_40ms_modulation_mae_db": 1.2397968443133573, "300–1000 Hz_10ms_log_rms_mae_db": 2.1029838760709363, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15645781792738622, "300–1000 Hz_10ms_envelope_correlation": 0.9831244511517168, "300–1000 Hz_10ms_modulation_mae_db": 0.407577335861969, "300–1000 Hz_40ms_log_rms_mae_db": 1.2225890694313037, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.089402632239207, "300–1000 Hz_40ms_envelope_correlation": 0.9945105084168364, "300–1000 Hz_40ms_modulation_mae_db": 0.5430100106395571, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8074867757275432, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22040184672752763, "1000–3000 Hz_10ms_envelope_correlation": 0.9639413320115585, "1000–3000 Hz_10ms_modulation_mae_db": 1.0617781724619646, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0944468285003184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1272008933147036, "1000–3000 Hz_40ms_envelope_correlation": 0.9887570295046364, "1000–3000 Hz_40ms_modulation_mae_db": 1.4040632773481068, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5904078232081278, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2248322126762109, "3000–6000 Hz_10ms_envelope_correlation": 0.9749380439032505, "3000–6000 Hz_10ms_modulation_mae_db": 1.6156588887185759, "3000–6000 Hz_40ms_log_rms_mae_db": 1.038294676104233, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17133951644328899, "3000–6000 Hz_40ms_envelope_correlation": 0.9858774376350274, "3000–6000 Hz_40ms_modulation_mae_db": 1.6066118584498046, "mr_stft_db": 6.020979796379401, "band_mae_db": 0.2527667327342704, "rms_gain_error_db": 0.15007018984212367, "output_peak": 0.38657290990549353}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6106300901710682, "全帯域_10ms_rms_envelope_nrmse": 0.14677177871410368, "全帯域_10ms_envelope_correlation": 0.981230432981462, "全帯域_10ms_modulation_mae_db": 0.6392865759988756, "全帯域_40ms_log_rms_mae_db": 0.8980695848615928, "全帯域_40ms_rms_envelope_nrmse": 0.08950076264382288, "全帯域_40ms_envelope_correlation": 0.9924788574599216, "全帯域_40ms_modulation_mae_db": 0.6590386079260212, "300–1000 Hz_10ms_log_rms_mae_db": 2.099276836798368, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1538128190861688, "300–1000 Hz_10ms_envelope_correlation": 0.9835681885213954, "300–1000 Hz_10ms_modulation_mae_db": 0.2596685532989133, "300–1000 Hz_40ms_log_rms_mae_db": 1.2794707937882628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08589515375107923, "300–1000 Hz_40ms_envelope_correlation": 0.9946253133705737, "300–1000 Hz_40ms_modulation_mae_db": 0.3725198421944634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.650659891341565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22934540017637411, "1000–3000 Hz_10ms_envelope_correlation": 0.9608733577954475, "1000–3000 Hz_10ms_modulation_mae_db": 2.052921997894264, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0208798115760536, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14056175268769883, "1000–3000 Hz_40ms_envelope_correlation": 0.9859571820345674, "1000–3000 Hz_40ms_modulation_mae_db": 2.6196263581196164, "3000–6000 Hz_10ms_log_rms_mae_db": 1.552136411214463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1789697809055497, "3000–6000 Hz_10ms_envelope_correlation": 0.9827010351395397, "3000–6000 Hz_10ms_modulation_mae_db": 0.7739581373173028, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0493063461144283, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11468618626682654, "3000–6000 Hz_40ms_envelope_correlation": 0.9929376559782879, "3000–6000 Hz_40ms_modulation_mae_db": 0.7324199415363825, "mr_stft_db": 6.005712937784956, "band_mae_db": 0.23615991508356413, "rms_gain_error_db": 0.00048017613335094456, "output_peak": 0.4181847222144214}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6828184844384153, "全帯域_10ms_rms_envelope_nrmse": 0.14691171497644717, "全帯域_10ms_envelope_correlation": 0.9811515578047391, "全帯域_10ms_modulation_mae_db": 0.36645166157003545, "全帯域_40ms_log_rms_mae_db": 0.8702176961989501, "全帯域_40ms_rms_envelope_nrmse": 0.08929870441569993, "全帯域_40ms_envelope_correlation": 0.992876006568651, "全帯域_40ms_modulation_mae_db": 0.5678509471435578, "300–1000 Hz_10ms_log_rms_mae_db": 2.145425080689364, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17289241483560694, "300–1000 Hz_10ms_envelope_correlation": 0.9793843483588753, "300–1000 Hz_10ms_modulation_mae_db": 0.2651536262910174, "300–1000 Hz_40ms_log_rms_mae_db": 1.2610033258257294, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09190041058295666, "300–1000 Hz_40ms_envelope_correlation": 0.994422756748848, "300–1000 Hz_40ms_modulation_mae_db": 0.18718014850400663, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6960847907996541, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18350907536078545, "1000–3000 Hz_10ms_envelope_correlation": 0.9747582570408191, "1000–3000 Hz_10ms_modulation_mae_db": 0.33409942912684093, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0151084568920246, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10603520282997209, "1000–3000 Hz_40ms_envelope_correlation": 0.991041812281703, "1000–3000 Hz_40ms_modulation_mae_db": 0.49963266362613945, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6061343811390314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18281382698032855, "3000–6000 Hz_10ms_envelope_correlation": 0.9813770420175636, "3000–6000 Hz_10ms_modulation_mae_db": 0.7663583494581939, "3000–6000 Hz_40ms_log_rms_mae_db": 1.049102726677774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12202961660431627, "3000–6000 Hz_40ms_envelope_correlation": 0.9921999481507816, "3000–6000 Hz_40ms_modulation_mae_db": 0.9080838901086893, "mr_stft_db": 6.026839838284111, "band_mae_db": 0.1888722837848197, "rms_gain_error_db": -0.23637586933481286, "output_peak": 0.4195296244661182}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5625705804912675, "全帯域_10ms_rms_envelope_nrmse": 0.1281720311907005, "全帯域_10ms_envelope_correlation": 0.9856319914024694, "全帯域_10ms_modulation_mae_db": 0.18467601689074917, "全帯域_40ms_log_rms_mae_db": 0.8734097791128625, "全帯域_40ms_rms_envelope_nrmse": 0.06356719850555666, "全帯域_40ms_envelope_correlation": 0.9961776650395402, "全帯域_40ms_modulation_mae_db": 0.2861947443191312, "300–1000 Hz_10ms_log_rms_mae_db": 2.1295798496171865, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18084302980473335, "300–1000 Hz_10ms_envelope_correlation": 0.9772020241267715, "300–1000 Hz_10ms_modulation_mae_db": 0.7163893286650681, "300–1000 Hz_40ms_log_rms_mae_db": 1.2800402538104867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09110433635405267, "300–1000 Hz_40ms_envelope_correlation": 0.9940641532486127, "300–1000 Hz_40ms_modulation_mae_db": 0.9999752820459045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6993291083452484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22111376029730412, "1000–3000 Hz_10ms_envelope_correlation": 0.9632431742407586, "1000–3000 Hz_10ms_modulation_mae_db": 0.7451892022031955, "1000–3000 Hz_40ms_log_rms_mae_db": 1.00939275645093, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14089869645738307, "1000–3000 Hz_40ms_envelope_correlation": 0.9840415946124283, "1000–3000 Hz_40ms_modulation_mae_db": 0.9792947486644635, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6529377908712604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15482255824267202, "3000–6000 Hz_10ms_envelope_correlation": 0.9867805533818057, "3000–6000 Hz_10ms_modulation_mae_db": 0.3590192063051029, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1223083517431065, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07721749081892133, "3000–6000 Hz_40ms_envelope_correlation": 0.9967389728600377, "3000–6000 Hz_40ms_modulation_mae_db": 0.3394655442826942, "mr_stft_db": 6.0109062168647425, "band_mae_db": 0.2198261937470486, "rms_gain_error_db": -0.0010653445755139798, "output_peak": 0.381656884562236}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76160, "output_samples": 76160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.1236690984150695, "spectral_convergence": 0.4897027819557284, "spectral_convergence_status": "ok", "amplitude_floor": 1.3356779608968942e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.779114731953467, "spectral_convergence": 0.46236058516418704, "spectral_convergence_status": "ok", "amplitude_floor": 9.962006407647762e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.171280012527867, "spectral_convergence": 0.515292068385318, "spectral_convergence_status": "ok", "amplitude_floor": 4.291398708237255e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021802020252140593, "output_power": 0.0002106378097635212, "delta_db": -0.14960408451788437, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019115120810434436, "output_power": 0.00018486684513572069, "delta_db": -0.14518017383082618, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.836069030265609e-05, "output_power": 4.829665000941432e-05, "delta_db": -0.005754834543178476, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.9699402895957116e-05, "output_power": 4.216979364100141e-05, "delta_db": -0.7134969522750472, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001234531768639194, "output_power": 0.00011754804700062475, "delta_db": -0.21286852319018185, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00036807758285640346, "output_power": 0.00038507081614340226, "delta_db": 0.1960123715607634, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23715282331964693, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04445928695913867, "pitch": null, "output_rms_dbfs": -30.049838037509996, "output_peak": 0.3392525616066561, "reference_rms_dbfs": -30.00537875055086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.024687947632135}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.00018341452353274373, "log_rms_mae_db": 1.6315914712365436, "rms_envelope_nrmse": 0.1438352602539658, "envelope_correlation": 0.9818948517306347, "rms_gain_error_db": -0.04453502837229849, "reference_mean_power": 0.0009998027766238483, "output_mean_power": 0.0009896026183188185}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 5.754282242130778e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018894414816227902, "output_power": 0.00020012386475359474, "reference_active": true, "delta_db": 0.2496543548845787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013084274193993133, "output_power": 0.0001520662884209327, "reference_active": true, "delta_db": 0.6528330934792155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.690431882761885e-05, "output_power": 3.064093210763905e-05, "reference_active": true, "delta_db": -0.8077522107009095}], "active_band_count": 3, "active_band_mae_db": 0.5700798863549013}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.0001230241515816465, "log_rms_mae_db": 0.9003726918324706, "rms_envelope_nrmse": 0.07865436480017308, "envelope_correlation": 0.9941898567549331, "rms_gain_error_db": -0.044616643682118234, "reference_mean_power": 0.001006100998286567, "output_mean_power": 0.0009958178702485988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 5.061730772699165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019045538591950462, "output_power": 0.00020818260018325697, "reference_active": true, "delta_db": 0.38651169879367353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.588390935994399e-05, "output_power": 8.450614243391395e-05, "reference_active": true, "delta_db": 0.4673858071092835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5269059669432984e-06, "output_power": 2.1845028879474246e-06, "reference_active": true, "delta_db": -0.6323645774935927}], "active_band_count": 3, "active_band_mae_db": 0.49542069446551656}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 5.489840871101384e-05, "log_rms_mae_db": 2.0556545030530815, "rms_envelope_nrmse": 0.16072054924783905, "envelope_correlation": 0.9821933904972974, "rms_gain_error_db": -0.21968035047126527, "reference_mean_power": 0.0001499835837122031, "output_mean_power": 0.00014258560784969525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.245105068973029e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.560041827470393e-05, "output_power": 6.032595642440282e-05, "reference_active": true, "delta_db": -0.36402392428447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1382200906061474e-05, "output_power": 1.7266801555507705e-05, "reference_active": true, "delta_db": -0.9284050807619236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4492416985862286e-06, "output_power": 1.984258259936843e-06, "reference_active": true, "delta_db": -0.9143344790056771}], "active_band_count": 3, "active_band_mae_db": 0.735587828017357}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 4.550505877934166e-05, "log_rms_mae_db": 1.2132512638279833, "rms_envelope_nrmse": 0.07606185103410809, "envelope_correlation": 0.9962253374887305, "rms_gain_error_db": -0.21968487927855507, "reference_mean_power": 0.00015093488949896852, "output_mean_power": 0.00014348984062198368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 1.1354747567630587e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.429099822482352e-05, "output_power": 5.840861150577299e-05, "reference_active": true, "delta_db": -0.4167328668005811}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916365304226932e-05, "output_power": 8.941138604153824e-06, "reference_active": true, "delta_db": -0.8668523313227366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3162457402511593e-07, "output_power": 1.87633477984284e-07, "reference_active": true, "delta_db": -0.9147430487946957}], "active_band_count": 3, "active_band_mae_db": 0.7327760823060044}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 4.066436664003698e-05, "log_rms_mae_db": 1.7214574370885334, "rms_envelope_nrmse": 0.1917360382456516, "envelope_correlation": 0.9735287783752048, "rms_gain_error_db": -0.4763575465615356, "reference_mean_power": 0.0001020893317881212, "output_mean_power": 9.148388652972454e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 9.602503308503003e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.081225810114696e-05, "output_power": 2.6076301121639192e-05, "reference_active": true, "delta_db": -0.7247753959897927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3389822486480084e-05, "output_power": 1.0992809670308695e-05, "reference_active": true, "delta_db": -0.856661108050021}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310273392307641e-06, "output_power": 2.2986285650960346e-06, "reference_active": true, "delta_db": -3.6364808134119504}], "active_band_count": 3, "active_band_mae_db": 1.7393057724839214}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 3.8037796302268945e-05, "log_rms_mae_db": 1.0525708120832535, "rms_envelope_nrmse": 0.11879267816563359, "envelope_correlation": 0.9907123911134101, "rms_gain_error_db": -0.4763849068981229, "reference_mean_power": 0.00010273559922291735, "output_mean_power": 9.206243714598108e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 8.714815910547825e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9375203264151156e-05, "output_power": 2.4613046110948003e-05, "reference_active": true, "delta_db": -0.7681551510633831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.005890574709935e-06, "output_power": 5.478532551073854e-06, "reference_active": true, "delta_db": -1.0679910338853644}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.15357493378345e-07, "output_power": 2.0890296699423545e-07, "reference_active": true, "delta_db": -3.921639869423255}], "active_band_count": 3, "active_band_mae_db": 1.9192620181240008}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001808188641270182, "log_rms_mae_db": 1.5997828508612388, "rms_envelope_nrmse": 0.18022061763233865, "envelope_correlation": 0.9820346350300168, "rms_gain_error_db": 0.14418043920450746, "reference_mean_power": 0.0004025082939999733, "output_mean_power": 0.0004160953646176825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 3.5966286763117237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001367199427375967, "output_power": 0.00016091968543337776, "reference_active": true, "delta_db": 0.7077730712179073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.299777768885521e-05, "output_power": 0.0001053549734405647, "reference_active": true, "delta_db": 0.5418247134695305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495937881252574e-05, "output_power": 2.5059711130141378e-05, "reference_active": true, "delta_db": 0.01742288009943214}], "active_band_count": 3, "active_band_mae_db": 0.4223402215956233}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.00011978792663441872, "log_rms_mae_db": 0.9815108632953092, "rms_envelope_nrmse": 0.10248023940154441, "envelope_correlation": 0.9941522492947722, "rms_gain_error_db": 0.1441833041701765, "reference_mean_power": 0.0004050621320160614, "output_mean_power": 0.00041873568622654406}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 3.450538172499259e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016854041012359937, "output_power": 0.00019685675223406483, "reference_active": true, "delta_db": 0.6744626939525844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.211674623356507e-05, "output_power": 5.8618564201567885e-05, "reference_active": true, "delta_db": 0.51057882758958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0452544406168747e-06, "output_power": 1.857583962299889e-06, "reference_active": true, "delta_db": -0.4179889139506452}], "active_band_count": 3, "active_band_mae_db": 0.5343434784976032}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7380773012073407, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "842af0259ac45a2eb8648241b810ff8885786a0502308f0ebfa7aad5508ff640", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9683200553994491, "standard_deviation": 0.005198874026128995, "minimum": 0.9608733577954475, "maximum": 0.9747582570408191, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7145577263632628, "standard_deviation": 0.0440874441892746, "minimum": 1.650659891341565, "maximum": 1.8074867757275432, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.222955634794076, "standard_deviation": 0.5497213155552837, "minimum": 0.33409942912684093, "maximum": 2.052921997894264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2059258124660721, "standard_deviation": 0.01656548814465319, "minimum": 0.18350907536078545, "maximum": 0.22934540017637411, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893096592928317, "standard_deviation": 0.0028566903032966524, "minimum": 0.9840415946124283, "maximum": 0.9919243123489971, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0356989499706692, "standard_deviation": 0.03395006738774899, "minimum": 1.0050812363362132, "maximum": 1.0944468285003184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5369145516330545, "standard_deviation": 0.6389803809252317, "minimum": 0.49963266362613945, "maximum": 2.6196263581196164, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12045376787488767, "standard_deviation": 0.014520959569914957, "minimum": 0.10443426535262192, "maximum": 0.14089869645738307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793828088604695, "standard_deviation": 0.0050900718690139995, "minimum": 0.9711635330593477, "maximum": 0.9867805533818057, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6055784248034681, "standard_deviation": 0.03651918934176226, "minimum": 1.552136411214463, "maximum": 1.6558496097847286, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5132489523877002, "standard_deviation": 1.0918843249750245, "minimum": 0.3590192063051029, "maximum": 3.181462616718394, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19700651481647974, "standard_deviation": 0.02606795904811917, "minimum": 0.15482255824267202, "maximum": 0.2248322126762109, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9913720130122282, "standard_deviation": 0.0037231183793857915, "minimum": 0.9858774376350274, "maximum": 0.9967389728600377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0692974007385059, "standard_deviation": 0.05860792034948499, "minimum": 0.9815108632953092, "maximum": 1.1686657084451408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6674354681060526, "standard_deviation": 1.263763573360819, "minimum": 0.3394655442826942, "maximum": 3.762697988763197, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13184569936857657, "standard_deviation": 0.03368719540708241, "minimum": 0.07721749081892133, "maximum": 0.17133951644328899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9802590019674936, "standard_deviation": 0.0027929256953490014, "minimum": 0.9764216150147719, "maximum": 0.9835681885213954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1236088714570887, "standard_deviation": 0.041706902968300395, "minimum": 2.0556545030530815, "maximum": 2.1988928152999696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4885560007841543, "standard_deviation": 0.18463845119270655, "minimum": 0.2596685532989133, "maximum": 0.735587828017357, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16854061584109845, "standard_deviation": 0.01185963539377588, "minimum": 0.1538128190861688, "maximum": 0.18401765198983208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9948501201763948, "standard_deviation": 0.0007516473055243645, "minimum": 0.9940641532486127, "maximum": 0.9962253374887305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.258107171078318, "standard_deviation": 0.04708628285286276, "minimum": 1.210000895603217, "maximum": 1.3521685656363895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6269043978429631, "standard_deviation": 0.29159797731947573, "minimum": 0.18718014850400663, "maximum": 1.0341695479845388, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0865970448509729, "standard_deviation": 0.00678541116891283, "minimum": 0.07606185103410809, "maximum": 0.09287310377686128, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2656108228112305, "standard_deviation": 0.10442698765461607, "minimum": 0.1888722837848197, "maximum": 0.510791151851329, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.021255654945583, "standard_deviation": 0.008594529402987092, "minimum": 6.005712937784956, "maximum": 6.030730965463047, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.39629233978811734, "standard_deviation": 0.037548769524296914, "minimum": 0.3392525616066561, "maximum": 0.45775674959782897, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12414847117296048, "standard_deviation": 0.18277693186533825, "minimum": -0.38992140791501323, "maximum": 0.15007018984212367, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9795188489605791, "standard_deviation": 0.003749176389048079, "minimum": 0.9738836196576995, "maximum": 0.9856319914024694, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6207230157556276, "standard_deviation": 0.03928998652879864, "minimum": 1.5625705804912675, "maximum": 1.6828184844384153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0041604756079783, "standard_deviation": 0.6508993082252338, "minimum": 0.18467601689074917, "maximum": 1.906955713596645, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15305540372387155, "standard_deviation": 0.014551734307750953, "minimum": 0.1281720311907005, "maximum": 0.17140954194609564, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9918552689622463, "standard_deviation": 0.0026343719711046003, "minimum": 0.9880815102292525, "maximum": 0.9961776650395402, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8950155673280764, "standard_deviation": 0.023944192810598567, "minimum": 0.8702176961989501, "maximum": 0.9442810539633502, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1874909581417668, "standard_deviation": 0.8037603243119896, "minimum": 0.2861947443191312, "maximum": 2.4028216153724964, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09380424535048197, "standard_deviation": 0.017507531395831978, "minimum": 0.06356719850555666, "maximum": 0.11553783318274859, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0003", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.3079757690429688, "predicted_mos": 2.3079757690429688, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "09e0195610d1992e849709e2e4aa04aafbb3db13e4253078eb506dce7021137a", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9071350996382535, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6829596798171707, "全帯域_10ms_rms_envelope_nrmse": 0.1464051741179687, "全帯域_10ms_envelope_correlation": 0.9813815983130018, "全帯域_10ms_modulation_mae_db": 0.5467400667258073, "全帯域_40ms_log_rms_mae_db": 0.8940176088507154, "全帯域_40ms_rms_envelope_nrmse": 0.08264728642369286, "全帯域_40ms_envelope_correlation": 0.9936241584854909, "全帯域_40ms_modulation_mae_db": 0.4922536589598791, "300–1000 Hz_10ms_log_rms_mae_db": 1.9855033631829384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1576439145468621, "300–1000 Hz_10ms_envelope_correlation": 0.9827581321003422, "300–1000 Hz_10ms_modulation_mae_db": 0.744090821018939, "300–1000 Hz_40ms_log_rms_mae_db": 1.0343220220780298, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07245080691324268, "300–1000 Hz_40ms_envelope_correlation": 0.9963697966050282, "300–1000 Hz_40ms_modulation_mae_db": 0.7630973994732555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.61184565864298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19454699047383492, "1000–3000 Hz_10ms_envelope_correlation": 0.9730921414414438, "1000–3000 Hz_10ms_modulation_mae_db": 1.7093920651620922, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8604413023823425, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12145427881794991, "1000–3000 Hz_40ms_envelope_correlation": 0.9907963509001481, "1000–3000 Hz_40ms_modulation_mae_db": 1.8715740367086156, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5152687011260955, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19222044483424447, "3000–6000 Hz_10ms_envelope_correlation": 0.9801200323178352, "3000–6000 Hz_10ms_modulation_mae_db": 0.6294579056240477, "3000–6000 Hz_40ms_log_rms_mae_db": 0.811864827722417, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11551456592486854, "3000–6000 Hz_40ms_envelope_correlation": 0.992935219560118, "3000–6000 Hz_40ms_modulation_mae_db": 0.7333650747151784, "mr_stft_db": 5.967526268564654, "band_mae_db": 0.3084503144582676, "rms_gain_error_db": -0.008409586387205922, "output_peak": 0.35396860719409906}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6394405655557038, "全帯域_10ms_rms_envelope_nrmse": 0.16214919908094322, "全帯域_10ms_envelope_correlation": 0.9772942701328972, "全帯域_10ms_modulation_mae_db": 1.1999424285464704, "全帯域_40ms_log_rms_mae_db": 0.9626755981367604, "全帯域_40ms_rms_envelope_nrmse": 0.1069805271692007, "全帯域_40ms_envelope_correlation": 0.9903035863480305, "全帯域_40ms_modulation_mae_db": 1.5587138892850942, "300–1000 Hz_10ms_log_rms_mae_db": 2.024113169027451, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1644757499500064, "300–1000 Hz_10ms_envelope_correlation": 0.981223651039483, "300–1000 Hz_10ms_modulation_mae_db": 0.5207158953164504, "300–1000 Hz_40ms_log_rms_mae_db": 1.0879668595407697, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08075235191198261, "300–1000 Hz_40ms_envelope_correlation": 0.9952335955123284, "300–1000 Hz_40ms_modulation_mae_db": 0.9392340707002439, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6232791321162452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21272139667758316, "1000–3000 Hz_10ms_envelope_correlation": 0.9670767758895924, "1000–3000 Hz_10ms_modulation_mae_db": 1.5713858736359156, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8278097758599755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1199964121413064, "1000–3000 Hz_40ms_envelope_correlation": 0.9910529164749657, "1000–3000 Hz_40ms_modulation_mae_db": 1.8624041401300344, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5553109172283077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20349846702052407, "3000–6000 Hz_10ms_envelope_correlation": 0.9826887993884756, "3000–6000 Hz_10ms_modulation_mae_db": 1.944315168889638, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9730634445308215, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1524310465896308, "3000–6000 Hz_40ms_envelope_correlation": 0.9942314612189423, "3000–6000 Hz_40ms_modulation_mae_db": 2.135791768272965, "mr_stft_db": 5.969472813293759, "band_mae_db": 0.55788739298061, "rms_gain_error_db": -0.35336164232824385, "output_peak": 0.34743403015437935}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7087881375967353, "全帯域_10ms_rms_envelope_nrmse": 0.15574077410007084, "全帯域_10ms_envelope_correlation": 0.9788061118074411, "全帯域_10ms_modulation_mae_db": 1.8044136032845877, "全帯域_40ms_log_rms_mae_db": 0.9792604517857998, "全帯域_40ms_rms_envelope_nrmse": 0.10371519739760461, "全帯域_40ms_envelope_correlation": 0.9904661883035167, "全帯域_40ms_modulation_mae_db": 2.2531625557124144, "300–1000 Hz_10ms_log_rms_mae_db": 1.9927769164396385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17492507651581235, "300–1000 Hz_10ms_envelope_correlation": 0.9789079675274619, "300–1000 Hz_10ms_modulation_mae_db": 0.38141970947871445, "300–1000 Hz_40ms_log_rms_mae_db": 1.0138139561986517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09055915177643273, "300–1000 Hz_40ms_envelope_correlation": 0.9946250839246321, "300–1000 Hz_40ms_modulation_mae_db": 0.6987983560676603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6137257425949543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2001758177881409, "1000–3000 Hz_10ms_envelope_correlation": 0.9706409689411802, "1000–3000 Hz_10ms_modulation_mae_db": 1.341539076398571, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8894659528053291, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10728249077251821, "1000–3000 Hz_40ms_envelope_correlation": 0.9926077397436848, "1000–3000 Hz_40ms_modulation_mae_db": 1.6640797823016626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4946540137101576, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2127900446227216, "3000–6000 Hz_10ms_envelope_correlation": 0.9754066887038431, "3000–6000 Hz_10ms_modulation_mae_db": 3.1395835495181714, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8849988336762276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15232445348325838, "3000–6000 Hz_40ms_envelope_correlation": 0.9885831316860038, "3000–6000 Hz_40ms_modulation_mae_db": 3.6717509290290433, "mr_stft_db": 5.969623747600071, "band_mae_db": 0.3035705125426151, "rms_gain_error_db": -0.28834888471033077, "output_peak": 0.404531947718739}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6741197796314602, "全帯域_10ms_rms_envelope_nrmse": 0.16984662377398932, "全帯域_10ms_envelope_correlation": 0.9743961576038168, "全帯域_10ms_modulation_mae_db": 1.3510601810500222, "全帯域_40ms_log_rms_mae_db": 0.8615819471748761, "全帯域_40ms_rms_envelope_nrmse": 0.09113081307573781, "全帯域_40ms_envelope_correlation": 0.9921930077109801, "全帯域_40ms_modulation_mae_db": 1.5070766202126464, "300–1000 Hz_10ms_log_rms_mae_db": 1.9939675264753796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17872434393243058, "300–1000 Hz_10ms_envelope_correlation": 0.9778021091549536, "300–1000 Hz_10ms_modulation_mae_db": 0.5781439780328292, "300–1000 Hz_40ms_log_rms_mae_db": 1.0405413026195902, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08507992953315416, "300–1000 Hz_40ms_envelope_correlation": 0.9949784164739424, "300–1000 Hz_40ms_modulation_mae_db": 0.5103605693334368, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6023862546546668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19932609279465358, "1000–3000 Hz_10ms_envelope_correlation": 0.9703889242014189, "1000–3000 Hz_10ms_modulation_mae_db": 0.8597307776063161, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8269470795211031, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11164583565385976, "1000–3000 Hz_40ms_envelope_correlation": 0.9909470262688159, "1000–3000 Hz_40ms_modulation_mae_db": 1.1627327519696742, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5282738091096812, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.220373080221982, "3000–6000 Hz_10ms_envelope_correlation": 0.9720902395350577, "3000–6000 Hz_10ms_modulation_mae_db": 2.103557114104403, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9167913873428862, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1292830546689536, "3000–6000 Hz_40ms_envelope_correlation": 0.9903891161218231, "3000–6000 Hz_40ms_modulation_mae_db": 2.234527734106395, "mr_stft_db": 5.963096339674256, "band_mae_db": 0.20568778434494073, "rms_gain_error_db": -0.1380084460064027, "output_peak": 0.4415771875310789}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6354132024558878, "全帯域_10ms_rms_envelope_nrmse": 0.18164844794531076, "全帯域_10ms_envelope_correlation": 0.9731166605364693, "全帯域_10ms_modulation_mae_db": 1.3003528245721707, "全帯域_40ms_log_rms_mae_db": 0.9193612996128938, "全帯域_40ms_rms_envelope_nrmse": 0.13289669646896488, "全帯域_40ms_envelope_correlation": 0.9848267424170045, "全帯域_40ms_modulation_mae_db": 1.3352099907469004, "300–1000 Hz_10ms_log_rms_mae_db": 1.9622087069570529, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15609911573358473, "300–1000 Hz_10ms_envelope_correlation": 0.9830995847778661, "300–1000 Hz_10ms_modulation_mae_db": 0.5314199485101497, "300–1000 Hz_40ms_log_rms_mae_db": 0.9713278060233204, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08206469050626151, "300–1000 Hz_40ms_envelope_correlation": 0.9952644489651743, "300–1000 Hz_40ms_modulation_mae_db": 0.43723860795325775, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7065068071479799, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23235462645226188, "1000–3000 Hz_10ms_envelope_correlation": 0.9602352988198076, "1000–3000 Hz_10ms_modulation_mae_db": 0.9452481396186019, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9203036156597865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14021491586852006, "1000–3000 Hz_40ms_envelope_correlation": 0.9867872980274014, "1000–3000 Hz_40ms_modulation_mae_db": 1.2397675390501253, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5561331784715244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24991103524020497, "3000–6000 Hz_10ms_envelope_correlation": 0.9711772861450113, "3000–6000 Hz_10ms_modulation_mae_db": 1.7839855173544557, "3000–6000 Hz_40ms_log_rms_mae_db": 0.950684822164007, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20075191806398057, "3000–6000 Hz_40ms_envelope_correlation": 0.9822636758931959, "3000–6000 Hz_40ms_modulation_mae_db": 1.6807648775886317, "mr_stft_db": 5.9633078955628065, "band_mae_db": 0.28347031613135837, "rms_gain_error_db": 0.22374823234738803, "output_peak": 0.4013078355862445}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6786723209587424, "全帯域_10ms_rms_envelope_nrmse": 0.14832064498568182, "全帯域_10ms_envelope_correlation": 0.9806686117417248, "全帯域_10ms_modulation_mae_db": 0.7467727382072967, "全帯域_40ms_log_rms_mae_db": 0.9262730111122408, "全帯域_40ms_rms_envelope_nrmse": 0.08748496839507014, "全帯域_40ms_envelope_correlation": 0.992747920855373, "全帯域_40ms_modulation_mae_db": 0.7742933514958527, "300–1000 Hz_10ms_log_rms_mae_db": 1.9916895210187653, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15477022897063433, "300–1000 Hz_10ms_envelope_correlation": 0.9834703266924876, "300–1000 Hz_10ms_modulation_mae_db": 0.35279075423305645, "300–1000 Hz_40ms_log_rms_mae_db": 1.0809088988278976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08578881418845288, "300–1000 Hz_40ms_envelope_correlation": 0.9946256175516155, "300–1000 Hz_40ms_modulation_mae_db": 0.42400430612100926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5864602178412364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22789867150758372, "1000–3000 Hz_10ms_envelope_correlation": 0.9621819837540994, "1000–3000 Hz_10ms_modulation_mae_db": 2.082537476485483, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8875852879646158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14002143824329377, "1000–3000 Hz_40ms_envelope_correlation": 0.9873229530482626, "1000–3000 Hz_40ms_modulation_mae_db": 2.6202948718244268, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4711712944077366, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1802789379060655, "3000–6000 Hz_10ms_envelope_correlation": 0.9819081423167668, "3000–6000 Hz_10ms_modulation_mae_db": 0.8698933401092152, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8660966971313608, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11208790415765703, "3000–6000 Hz_40ms_envelope_correlation": 0.9928790912772972, "3000–6000 Hz_40ms_modulation_mae_db": 0.749129035329802, "mr_stft_db": 5.9417588949262, "band_mae_db": 0.2574849226507682, "rms_gain_error_db": -0.041676192175366625, "output_peak": 0.4055318073345108}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7329845056734752, "全帯域_10ms_rms_envelope_nrmse": 0.14729121285421046, "全帯域_10ms_envelope_correlation": 0.9809942302568899, "全帯域_10ms_modulation_mae_db": 0.4181061633417036, "全帯域_40ms_log_rms_mae_db": 0.8690660599303512, "全帯域_40ms_rms_envelope_nrmse": 0.08982832586336949, "全帯域_40ms_envelope_correlation": 0.9926792835022271, "全帯域_40ms_modulation_mae_db": 0.562143111524876, "300–1000 Hz_10ms_log_rms_mae_db": 2.0370728907257614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15926576673832982, "300–1000 Hz_10ms_envelope_correlation": 0.9825021255850576, "300–1000 Hz_10ms_modulation_mae_db": 0.16034195516541283, "300–1000 Hz_40ms_log_rms_mae_db": 1.0659345469157002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08442245566031696, "300–1000 Hz_40ms_envelope_correlation": 0.9951859296877535, "300–1000 Hz_40ms_modulation_mae_db": 0.10915909657989026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6007888379877282, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1920903024998443, "1000–3000 Hz_10ms_envelope_correlation": 0.9723531486301683, "1000–3000 Hz_10ms_modulation_mae_db": 0.3391393277549523, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8438973423348378, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12052219566161063, "1000–3000 Hz_40ms_envelope_correlation": 0.9887384051937993, "1000–3000 Hz_40ms_modulation_mae_db": 0.3712007105823067, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5258074989996142, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18192714675804644, "3000–6000 Hz_10ms_envelope_correlation": 0.9812323948525807, "3000–6000 Hz_10ms_modulation_mae_db": 0.8414696683137438, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8964309807635167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11931666250442582, "3000–6000 Hz_40ms_envelope_correlation": 0.9920843254226976, "3000–6000 Hz_40ms_modulation_mae_db": 1.0225779886055932, "mr_stft_db": 5.973630344522969, "band_mae_db": 0.18639581746463474, "rms_gain_error_db": -0.21454236742028948, "output_peak": 0.40881521992031444}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.667498899131978, "全帯域_10ms_rms_envelope_nrmse": 0.1316533946523366, "全帯域_10ms_envelope_correlation": 0.9848354173251487, "全帯域_10ms_modulation_mae_db": 0.3596453746623636, "全帯域_40ms_log_rms_mae_db": 0.9130707335354403, "全帯域_40ms_rms_envelope_nrmse": 0.0650278044454112, "全帯域_40ms_envelope_correlation": 0.9959932853781239, "全帯域_40ms_modulation_mae_db": 0.22716697715005943, "300–1000 Hz_10ms_log_rms_mae_db": 2.008508479803477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18761758641113183, "300–1000 Hz_10ms_envelope_correlation": 0.9754483057394038, "300–1000 Hz_10ms_modulation_mae_db": 0.7846194135516179, "300–1000 Hz_40ms_log_rms_mae_db": 1.101197674435977, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09401690564375964, "300–1000 Hz_40ms_envelope_correlation": 0.9936056217139546, "300–1000 Hz_40ms_modulation_mae_db": 0.9531275276805525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.583471906111453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2167558209224049, "1000–3000 Hz_10ms_envelope_correlation": 0.9645800432615245, "1000–3000 Hz_10ms_modulation_mae_db": 0.8043304324548887, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8193679056219471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14259923077585918, "1000–3000 Hz_40ms_envelope_correlation": 0.9837719668467882, "1000–3000 Hz_40ms_modulation_mae_db": 1.0767051943885841, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5431965551615232, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15533221999691224, "3000–6000 Hz_10ms_envelope_correlation": 0.9866036246110046, "3000–6000 Hz_10ms_modulation_mae_db": 0.5247900486707667, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8933834747044401, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07844787045613791, "3000–6000 Hz_40ms_envelope_correlation": 0.9965543592887616, "3000–6000 Hz_40ms_modulation_mae_db": 0.5153230338282092, "mr_stft_db": 5.954207901362781, "band_mae_db": 0.19895939541265442, "rms_gain_error_db": -0.016205430513584338, "output_peak": 0.37287554528424316}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76160, "output_samples": 76160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.07579534947718, "spectral_convergence": 0.48699308371534406, "spectral_convergence_status": "ok", "amplitude_floor": 1.3356779608968942e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.734130911114449, "spectral_convergence": 0.46183335652319923, "spectral_convergence_status": "ok", "amplitude_floor": 9.962006407647762e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.092652545102333, "spectral_convergence": 0.520055967953226, "spectral_convergence_status": "ok", "amplitude_floor": 4.291398708237255e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021802020252140593, "output_power": 0.00020719912712609625, "delta_db": -0.22108817216859503, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019115120810434436, "output_power": 0.00018685240757809242, "delta_db": -0.09878349099246904, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.836069030265609e-05, "output_power": 4.9971079737979565e-05, "delta_db": 0.14226242084766727, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.9699402895957116e-05, "output_power": 4.116458815894135e-05, "delta_db": -0.8182739625075517, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001234531768639194, "output_power": 0.00011628506145565961, "delta_db": -0.2597834353283896, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00036807758285640346, "output_power": 0.00039535788864221726, "delta_db": 0.31051040490493276, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3084503144582676, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.008409586387205922, "pitch": null, "output_rms_dbfs": -30.013788336938067, "output_peak": 0.35396860719409906, "reference_rms_dbfs": -30.00537875055086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.967526268564654}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.00018341452353274373, "log_rms_mae_db": 1.6829596798171707, "rms_envelope_nrmse": 0.1464051741179687, "envelope_correlation": 0.9813815983130018, "rms_gain_error_db": -0.00849276519882745, "reference_mean_power": 0.0009998027766238483, "output_mean_power": 0.0009978495412748813}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 5.754282242130778e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018894414816227902, "output_power": 0.00020620138075135337, "reference_active": true, "delta_db": 0.3795812323433363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013084274193993133, "output_power": 0.00015716476283340918, "reference_active": true, "delta_db": 0.7960554488828957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.690431882761885e-05, "output_power": 3.316032983788287e-05, "reference_active": true, "delta_db": -0.4645835189511899}], "active_band_count": 3, "active_band_mae_db": 0.5467400667258073}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.0001230241515816465, "log_rms_mae_db": 0.8940176088507154, "rms_envelope_nrmse": 0.08264728642369286, "envelope_correlation": 0.9936241584854909, "rms_gain_error_db": -0.008566178031520544, "reference_mean_power": 0.001006100998286567, "output_mean_power": 0.0010041184849102517}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 5.061730772699165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019045538591950462, "output_power": 0.00021561289280968822, "reference_active": true, "delta_db": 0.5388146775015852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.588390935994399e-05, "output_power": 8.767451181599996e-05, "reference_active": true, "delta_db": 0.6272365972772505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5269059669432984e-06, "output_power": 2.352437591509411e-06, "reference_active": true, "delta_db": -0.31070970210080157}], "active_band_count": 3, "active_band_mae_db": 0.4922536589598791}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 5.489840871101384e-05, "log_rms_mae_db": 1.9855033631829384, "rms_envelope_nrmse": 0.1576439145468621, "envelope_correlation": 0.9827581321003422, "rms_gain_error_db": -0.16174495847662398, "reference_mean_power": 0.0001499835837122031, "output_mean_power": 0.00014450046032143566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.245105068973029e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.560041827470393e-05, "output_power": 6.110116900278541e-05, "reference_active": true, "delta_db": -0.30857089136470695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1382200906061474e-05, "output_power": 1.721812009137089e-05, "reference_active": true, "delta_db": -0.9406667326840299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4492416985862286e-06, "output_power": 1.9531165533486297e-06, "reference_active": true, "delta_db": -0.9830348390080802}], "active_band_count": 3, "active_band_mae_db": 0.744090821018939}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 4.550505877934166e-05, "log_rms_mae_db": 1.0343220220780298, "rms_envelope_nrmse": 0.07245080691324268, "envelope_correlation": 0.9963697966050282, "rms_gain_error_db": -0.16174708961269493, "reference_mean_power": 0.00015093488949896852, "output_mean_power": 0.00014541691676408215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 1.1354747567630587e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.429099822482352e-05, "output_power": 5.9263959640338024e-05, "reference_active": true, "delta_db": -0.3535950405424129}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916365304226932e-05, "output_power": 8.853386536199946e-06, "reference_active": true, "delta_db": -0.9096863463059723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3162457402511593e-07, "output_power": 1.82887301719109e-07, "reference_active": true, "delta_db": -1.0260108115713815}], "active_band_count": 3, "active_band_mae_db": 0.7630973994732555}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 4.066436664003698e-05, "log_rms_mae_db": 1.61184565864298, "rms_envelope_nrmse": 0.19454699047383492, "envelope_correlation": 0.9730921414414438, "rms_gain_error_db": -0.5576331459470298, "reference_mean_power": 0.0001020893317881212, "output_mean_power": 8.978774130560567e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 9.602503308503003e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.081225810114696e-05, "output_power": 2.664445305222654e-05, "reference_active": true, "delta_db": -0.6311671747697019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3389822486480084e-05, "output_power": 1.0813791300276648e-05, "reference_active": true, "delta_db": -0.9279683576165283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310273392307641e-06, "output_power": 2.334601791311164e-06, "reference_active": true, "delta_db": -3.5690406631000466}], "active_band_count": 3, "active_band_mae_db": 1.7093920651620922}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 3.8037796302268945e-05, "log_rms_mae_db": 0.8604413023823425, "rms_envelope_nrmse": 0.12145427881794991, "envelope_correlation": 0.9907963509001481, "rms_gain_error_db": -0.5576501460506499, "reference_mean_power": 0.00010273559922291735, "output_mean_power": 9.035578092495319e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 8.714815910547825e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9375203264151156e-05, "output_power": 2.5251702510821474e-05, "reference_active": true, "delta_db": -0.6569021614523982}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.005890574709935e-06, "output_power": 5.412792687332996e-06, "reference_active": true, "delta_db": -1.120419560577223}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.15357493378345e-07, "output_power": 2.1299458200645762e-07, "reference_active": true, "delta_db": -3.8374003880962255}], "active_band_count": 3, "active_band_mae_db": 1.8715740367086156}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001808188641270182, "log_rms_mae_db": 1.5152687011260955, "rms_envelope_nrmse": 0.19222044483424447, "envelope_correlation": 0.9801200323178352, "rms_gain_error_db": 0.2515438838718581, "reference_mean_power": 0.0004025082939999733, "output_mean_power": 0.00042651000395179734}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 3.5966286763117237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001367199427375967, "output_power": 0.00017192134526059855, "reference_active": true, "delta_db": 0.9949793297336471}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.299777768885521e-05, "output_power": 0.00011314290182742134, "reference_active": true, "delta_db": 0.8515474248746323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495937881252574e-05, "output_power": 2.5201040290199225e-05, "reference_active": true, "delta_db": 0.04184696226386379}], "active_band_count": 3, "active_band_mae_db": 0.6294579056240477}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.00011978792663441872, "log_rms_mae_db": 0.811864827722417, "rms_envelope_nrmse": 0.11551456592486854, "envelope_correlation": 0.992935219560118, "rms_gain_error_db": 0.25154964131415947, "reference_mean_power": 0.0004050621320160614, "output_mean_power": 0.0004292166972317062}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 3.450538172499259e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016854041012359937, "output_power": 0.00020972638226807844, "reference_active": true, "delta_db": 0.9494901904131314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.211674623356507e-05, "output_power": 6.241528299163741e-05, "reference_active": true, "delta_db": 0.7831365005008871}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0452544406168747e-06, "output_power": 1.8365404194422252e-06, "reference_active": true, "delta_db": -0.4674685332315167}], "active_band_count": 3, "active_band_mae_db": 0.7333650747151784}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7393105875146004, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "80168a341cb04eaa9deedfc04a836684b8ad23d7707e93787008caff53da7dee", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9675686606174043, "standard_deviation": 0.004821994226560988, "minimum": 0.9602352988198076, "maximum": 0.9730921414414438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6160580696371554, "standard_deviation": 0.03893999204046592, "minimum": 1.583471906111453, "maximum": 1.7065068071479799, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2066628961396026, "standard_deviation": 0.5699969241811144, "minimum": 0.3391393277549523, "maximum": 2.082537476485483, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20948371488953843, "standard_deviation": 0.015311088877811182, "minimum": 0.1920903024998443, "maximum": 0.23235462645226188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9890030820629833, "standard_deviation": 0.002911976818931441, "minimum": 0.9837719668467882, "maximum": 0.9926077397436848, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8594772827687422, "standard_deviation": 0.03645485307613757, "minimum": 0.8193679056219471, "maximum": 0.9203036156597865, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4835948783694288, "standard_deviation": 0.6744384155707788, "minimum": 0.3712007105823067, "maximum": 2.6202948718244268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12546709974186473, "standard_deviation": 0.01370624122068592, "minimum": 0.10728249077251821, "maximum": 0.14259923077585918, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9789034009838219, "standard_deviation": 0.00544969954493529, "minimum": 0.9711772861450113, "maximum": 0.9866036246110046, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5237269960268303, "standard_deviation": 0.029594046939781543, "minimum": 1.4711712944077366, "maximum": 1.5561331784715244, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.4796315390730552, "standard_deviation": 0.9154885302742898, "minimum": 0.5247900486707667, "maximum": 3.1395835495181714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19954142207508768, "standard_deviation": 0.028912667424303727, "minimum": 0.15533221999691224, "maximum": 0.24991103524020497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991240047558605, "standard_deviation": 0.00433955028826345, "minimum": 0.9822636758931959, "maximum": 0.9965543592887616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8991643085044596, "standard_deviation": 0.04983168184050198, "minimum": 0.811864827722417, "maximum": 0.9730634445308215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5929038051844773, "standard_deviation": 1.0672359550503496, "minimum": 0.5153230338282092, "maximum": 3.6717509290290433, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13251968448111406, "standard_deviation": 0.03637015506086409, "minimum": 0.07844787045613791, "maximum": 0.20075191806398057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.980651525327132, "standard_deviation": 0.0029364978620104207, "minimum": 0.9754483057394038, "maximum": 0.9834703266924876, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.999480071703808, "standard_deviation": 0.023366758486356016, "minimum": 1.9622087069570529, "maximum": 2.0370728907257614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5066928094133962, "standard_deviation": 0.20662335210287183, "minimum": 0.16034195516541283, "maximum": 0.7846194135516179, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16669022284984902, "standard_deviation": 0.012228230613003507, "minimum": 0.15477022897063433, "maximum": 0.18761758641113183, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949860638043035, "standard_deviation": 0.0007800494398720032, "minimum": 0.9936056217139546, "maximum": 0.9963697966050282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.049501633329992, "standard_deviation": 0.043266577281271805, "minimum": 0.9713278060233204, "maximum": 1.101197674435977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6043774917386633, "standard_deviation": 0.28844736037231855, "minimum": 0.10915909657989026, "maximum": 0.9531275276805525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0843918882667004, "standard_deviation": 0.006484633859818454, "minimum": 0.07245080691324268, "maximum": 0.09401690564375964, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.28773830699823116, "standard_deviation": 0.11921046289646059, "minimum": 0.18639581746463474, "maximum": 0.55788739298061, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.962828025688436, "standard_deviation": 0.010334517052122308, "minimum": 5.9417588949262, "maximum": 5.973630344522969, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.39200527259045115, "standard_deviation": 0.0315452091085958, "minimum": 0.34743403015437935, "maximum": 0.4415771875310789, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10460053964925445, "standard_deviation": 0.1842698404145559, "minimum": -0.35336164232824385, "maximum": 0.22374823234738803, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9789366322146736, "standard_deviation": 0.0038768425407114506, "minimum": 0.9731166605364693, "maximum": 0.9848354173251487, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.677484636352644, "standard_deviation": 0.032557477661944695, "minimum": 1.6354132024558878, "maximum": 1.7329845056734752, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9658791725488027, "standard_deviation": 0.5222783247411626, "minimum": 0.3596453746623636, "maximum": 1.8044136032845877, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15538193393881394, "standard_deviation": 0.0156023304886804, "minimum": 0.1316533946523366, "maximum": 0.18164844794531076, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9916042716250935, "standard_deviation": 0.00327491541729679, "minimum": 0.9848267424170045, "maximum": 0.9959932853781239, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9156633387673847, "standard_deviation": 0.04132164412799257, "minimum": 0.8615819471748761, "maximum": 0.9792604517857998, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0887525193859653, "standard_deviation": 0.6852312783927997, "minimum": 0.22716697715005943, "maximum": 2.2531625557124144, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09496395240488147, "standard_deviation": 0.020016962323680023, "minimum": 0.0650278044454112, "maximum": 0.13289669646896488, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0003", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.341621160507202, "predicted_mos": 2.341621160507202, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "ea49bf3f5e41ef3f36ff3e230cede7d01499d381530e9c5993b3f5db64797dfe", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8862016000784934, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5400242046474837, "全帯域_10ms_rms_envelope_nrmse": 0.14055973479993014, "全帯域_10ms_envelope_correlation": 0.9824876695582943, "全帯域_10ms_modulation_mae_db": 1.378000964202446, "全帯域_40ms_log_rms_mae_db": 0.843715110375738, "全帯域_40ms_rms_envelope_nrmse": 0.07898976527922698, "全帯域_40ms_envelope_correlation": 0.9941911714981233, "全帯域_40ms_modulation_mae_db": 1.9951524397423681, "300–1000 Hz_10ms_log_rms_mae_db": 2.639061953885179, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20930192529570169, "300–1000 Hz_10ms_envelope_correlation": 0.9696301361155416, "300–1000 Hz_10ms_modulation_mae_db": 0.5260209603257626, "300–1000 Hz_40ms_log_rms_mae_db": 1.652049848538113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11398748161247883, "300–1000 Hz_40ms_envelope_correlation": 0.990524329436297, "300–1000 Hz_40ms_modulation_mae_db": 0.6135673812010952, "1000–3000 Hz_10ms_log_rms_mae_db": 2.195425346838845, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19155659560675123, "1000–3000 Hz_10ms_envelope_correlation": 0.9727277669728903, "1000–3000 Hz_10ms_modulation_mae_db": 1.0074685213549082, "1000–3000 Hz_40ms_log_rms_mae_db": 1.396944325292114, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11706907056371779, "1000–3000 Hz_40ms_envelope_correlation": 0.9897477382040352, "1000–3000 Hz_40ms_modulation_mae_db": 1.309494676668601, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1026627955569803, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1845784964412195, "3000–6000 Hz_10ms_envelope_correlation": 0.9812194991406256, "3000–6000 Hz_10ms_modulation_mae_db": 2.4737545341758804, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4491183575902262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11662190136439363, "3000–6000 Hz_40ms_envelope_correlation": 0.9932002387748137, "3000–6000 Hz_40ms_modulation_mae_db": 2.932953225974573, "mr_stft_db": 6.311129230644681, "band_mae_db": 0.24321504836686061, "rms_gain_error_db": -0.13776404514057056, "output_peak": 0.3799772695177661}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5062183354976086, "全帯域_10ms_rms_envelope_nrmse": 0.13913822571017845, "全帯域_10ms_envelope_correlation": 0.9829358152681344, "全帯域_10ms_modulation_mae_db": 0.6374381810429619, "全帯域_40ms_log_rms_mae_db": 0.8134210329094785, "全帯域_40ms_rms_envelope_nrmse": 0.06539817909131784, "全帯域_40ms_envelope_correlation": 0.9959563332792203, "全帯域_40ms_modulation_mae_db": 0.7419037277678701, "300–1000 Hz_10ms_log_rms_mae_db": 2.717514958303387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18593481808174736, "300–1000 Hz_10ms_envelope_correlation": 0.9759169205659568, "300–1000 Hz_10ms_modulation_mae_db": 0.5591622266913677, "300–1000 Hz_40ms_log_rms_mae_db": 1.6178152420665521, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09066957375836306, "300–1000 Hz_40ms_envelope_correlation": 0.9942520148237777, "300–1000 Hz_40ms_modulation_mae_db": 0.9699223202743923, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3058581144835446, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2069269211948332, "1000–3000 Hz_10ms_envelope_correlation": 0.9683082447168894, "1000–3000 Hz_10ms_modulation_mae_db": 1.6609968375121669, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5232314864240606, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12998068093700976, "1000–3000 Hz_40ms_envelope_correlation": 0.987302593662058, "1000–3000 Hz_40ms_modulation_mae_db": 1.9961032032934893, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0815012099257353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1804760771881318, "3000–6000 Hz_10ms_envelope_correlation": 0.9816095721777951, "3000–6000 Hz_10ms_modulation_mae_db": 0.8703052364831425, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4416365984549195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08535751125174895, "3000–6000 Hz_40ms_envelope_correlation": 0.9958233330214344, "3000–6000 Hz_40ms_modulation_mae_db": 0.9007805723378853, "mr_stft_db": 6.331535968054631, "band_mae_db": 0.1679155349752893, "rms_gain_error_db": -0.051456302847292, "output_peak": 0.3943131523603217}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.434599677927054, "全帯域_10ms_rms_envelope_nrmse": 0.13816904397627416, "全帯域_10ms_envelope_correlation": 0.9830970257638184, "全帯域_10ms_modulation_mae_db": 0.534118658528561, "全帯域_40ms_log_rms_mae_db": 0.7353745165249568, "全帯域_40ms_rms_envelope_nrmse": 0.0662222431964043, "全帯域_40ms_envelope_correlation": 0.9958642803692219, "全帯域_40ms_modulation_mae_db": 0.5617661314327475, "300–1000 Hz_10ms_log_rms_mae_db": 2.6825599730908536, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17065447605435755, "300–1000 Hz_10ms_envelope_correlation": 0.9802108966715508, "300–1000 Hz_10ms_modulation_mae_db": 1.0484339225110013, "300–1000 Hz_40ms_log_rms_mae_db": 1.641473112924776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10638236884895323, "300–1000 Hz_40ms_envelope_correlation": 0.9925634211044012, "300–1000 Hz_40ms_modulation_mae_db": 1.5644390898470044, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2761668576917993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22085261370581916, "1000–3000 Hz_10ms_envelope_correlation": 0.9633034123681021, "1000–3000 Hz_10ms_modulation_mae_db": 1.9167739784192728, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4071767954108374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13079115303040043, "1000–3000 Hz_40ms_envelope_correlation": 0.986697153817022, "1000–3000 Hz_40ms_modulation_mae_db": 2.529394368046273, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0747150881356773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1795239430719399, "3000–6000 Hz_10ms_envelope_correlation": 0.9815886339441595, "3000–6000 Hz_10ms_modulation_mae_db": 0.7125008421009961, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4811380226001223, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08928117856739456, "3000–6000 Hz_40ms_envelope_correlation": 0.9954790908433082, "3000–6000 Hz_40ms_modulation_mae_db": 0.7998446526722777, "mr_stft_db": 6.3155162256506285, "band_mae_db": 0.261559842965752, "rms_gain_error_db": -0.08643756359423271, "output_peak": 0.3485395426739418}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5748843260853918, "全帯域_10ms_rms_envelope_nrmse": 0.14456873300287176, "全帯域_10ms_envelope_correlation": 0.9818887223179841, "全帯域_10ms_modulation_mae_db": 0.7142434451088482, "全帯域_40ms_log_rms_mae_db": 0.8599736639978639, "全帯域_40ms_rms_envelope_nrmse": 0.0971844395360949, "全帯域_40ms_envelope_correlation": 0.991139092924791, "全帯域_40ms_modulation_mae_db": 0.7147484556084773, "300–1000 Hz_10ms_log_rms_mae_db": 2.7154327124734516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17144876394096736, "300–1000 Hz_10ms_envelope_correlation": 0.9797839094874113, "300–1000 Hz_10ms_modulation_mae_db": 0.738877182173696, "300–1000 Hz_40ms_log_rms_mae_db": 1.655913579477952, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08837817633523458, "300–1000 Hz_40ms_envelope_correlation": 0.9948672136229197, "300–1000 Hz_40ms_modulation_mae_db": 0.9857564122512876, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2472318639797657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20936499749276, "1000–3000 Hz_10ms_envelope_correlation": 0.9672027889101881, "1000–3000 Hz_10ms_modulation_mae_db": 1.3342188845147216, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3701643116499302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11625417844425665, "1000–3000 Hz_40ms_envelope_correlation": 0.9898119764241528, "1000–3000 Hz_40ms_modulation_mae_db": 1.5657079598150034, "3000–6000 Hz_10ms_log_rms_mae_db": 2.177558042715289, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17327665120822608, "3000–6000 Hz_10ms_envelope_correlation": 0.9831797754349956, "3000–6000 Hz_10ms_modulation_mae_db": 1.0697275251736449, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5113538223614607, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13259667175409734, "3000–6000 Hz_40ms_envelope_correlation": 0.9898989938187022, "3000–6000 Hz_40ms_modulation_mae_db": 0.9825061752781453, "mr_stft_db": 6.310357538205401, "band_mae_db": 0.11670859660361725, "rms_gain_error_db": 0.011014542155950165, "output_peak": 0.44118121790177667}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4749815827307582, "全帯域_10ms_rms_envelope_nrmse": 0.13038490431082944, "全帯域_10ms_envelope_correlation": 0.9849923676685416, "全帯域_10ms_modulation_mae_db": 0.11167520384792545, "全帯域_40ms_log_rms_mae_db": 0.8510956797359398, "全帯域_40ms_rms_envelope_nrmse": 0.0714131460210987, "全帯域_40ms_envelope_correlation": 0.9952385843033743, "全帯域_40ms_modulation_mae_db": 0.16514402086743288, "300–1000 Hz_10ms_log_rms_mae_db": 2.6985406109556473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16861473256041581, "300–1000 Hz_10ms_envelope_correlation": 0.9805977273700962, "300–1000 Hz_10ms_modulation_mae_db": 0.4770196661835837, "300–1000 Hz_40ms_log_rms_mae_db": 1.6957622565454455, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0835593489204207, "300–1000 Hz_40ms_envelope_correlation": 0.995794437070595, "300–1000 Hz_40ms_modulation_mae_db": 0.46351656106109157, "1000–3000 Hz_10ms_log_rms_mae_db": 2.288559546199125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2124872409741615, "1000–3000 Hz_10ms_envelope_correlation": 0.9661804542511192, "1000–3000 Hz_10ms_modulation_mae_db": 1.0768823895162074, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4150675419122376, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12135555170710614, "1000–3000 Hz_40ms_envelope_correlation": 0.9887104234290456, "1000–3000 Hz_40ms_modulation_mae_db": 1.3835030688321888, "3000–6000 Hz_10ms_log_rms_mae_db": 2.125961430590244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16663997597117464, "3000–6000 Hz_10ms_envelope_correlation": 0.9841707155639008, "3000–6000 Hz_10ms_modulation_mae_db": 0.34907514190451644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4817487606732047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09570345179808909, "3000–6000 Hz_40ms_envelope_correlation": 0.9947655632414247, "3000–6000 Hz_40ms_modulation_mae_db": 0.2753121725603596, "mr_stft_db": 6.313513575363269, "band_mae_db": 0.2469984848542259, "rms_gain_error_db": -0.11761688767136194, "output_peak": 0.3810761270732385}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4566820986276723, "全帯域_10ms_rms_envelope_nrmse": 0.13667315681206993, "全帯域_10ms_envelope_correlation": 0.9834533180304746, "全帯域_10ms_modulation_mae_db": 0.44413689538016793, "全帯域_40ms_log_rms_mae_db": 0.7158174319192017, "全帯域_40ms_rms_envelope_nrmse": 0.06905251485511836, "全帯域_40ms_envelope_correlation": 0.9956286264547001, "全帯域_40ms_modulation_mae_db": 0.5164001784335367, "300–1000 Hz_10ms_log_rms_mae_db": 2.770969397668642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16818634621619782, "300–1000 Hz_10ms_envelope_correlation": 0.9806610728974791, "300–1000 Hz_10ms_modulation_mae_db": 0.5083829597046048, "300–1000 Hz_40ms_log_rms_mae_db": 1.750927939797489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09056686396856685, "300–1000 Hz_40ms_envelope_correlation": 0.9947654402910213, "300–1000 Hz_40ms_modulation_mae_db": 0.7428933761355067, "1000–3000 Hz_10ms_log_rms_mae_db": 2.342468609575455, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2147835674037899, "1000–3000 Hz_10ms_envelope_correlation": 0.9655438107822119, "1000–3000 Hz_10ms_modulation_mae_db": 1.5137647006157728, "1000–3000 Hz_40ms_log_rms_mae_db": 1.430527972408139, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11889793579401915, "1000–3000 Hz_40ms_envelope_correlation": 0.9895567570614678, "1000–3000 Hz_40ms_modulation_mae_db": 1.8107091593861406, "3000–6000 Hz_10ms_log_rms_mae_db": 2.101849153109032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17962045782476455, "3000–6000 Hz_10ms_envelope_correlation": 0.9816648563703729, "3000–6000 Hz_10ms_modulation_mae_db": 0.5177912971597312, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4758722967690003, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09748215487859475, "3000–6000 Hz_40ms_envelope_correlation": 0.9947976682941782, "3000–6000 Hz_40ms_modulation_mae_db": 0.673260813257269, "mr_stft_db": 6.321015909863337, "band_mae_db": 0.18474791938344493, "rms_gain_error_db": -0.13401666795527264, "output_peak": 0.4014346382287152}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5672093304578056, "全帯域_10ms_rms_envelope_nrmse": 0.14878199809342835, "全帯域_10ms_envelope_correlation": 0.9813585938649136, "全帯域_10ms_modulation_mae_db": 0.7704612787925057, "全帯域_40ms_log_rms_mae_db": 0.9027773938723221, "全帯域_40ms_rms_envelope_nrmse": 0.0896126199865517, "全帯域_40ms_envelope_correlation": 0.9928622112318826, "全帯域_40ms_modulation_mae_db": 0.7979452447858328, "300–1000 Hz_10ms_log_rms_mae_db": 2.6703357159626884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16889898569267567, "300–1000 Hz_10ms_envelope_correlation": 0.9803165650445692, "300–1000 Hz_10ms_modulation_mae_db": 0.3863524383596868, "300–1000 Hz_40ms_log_rms_mae_db": 1.6988265703658885, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09497142133921188, "300–1000 Hz_40ms_envelope_correlation": 0.9937115105005646, "300–1000 Hz_40ms_modulation_mae_db": 0.6304698077415948, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3253212574568605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2271741104748798, "1000–3000 Hz_10ms_envelope_correlation": 0.9619586555103362, "1000–3000 Hz_10ms_modulation_mae_db": 1.9489949367669632, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4894768485379293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14662256480153582, "1000–3000 Hz_40ms_envelope_correlation": 0.9849057105546531, "1000–3000 Hz_40ms_modulation_mae_db": 2.3553684645128565, "3000–6000 Hz_10ms_log_rms_mae_db": 2.115558093019488, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18434903631420457, "3000–6000 Hz_10ms_envelope_correlation": 0.9834109994771297, "3000–6000 Hz_10ms_modulation_mae_db": 1.176164917757706, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4876240085809478, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11623360441128504, "3000–6000 Hz_40ms_envelope_correlation": 0.9942497685653656, "3000–6000 Hz_40ms_modulation_mae_db": 1.0955157655395251, "mr_stft_db": 6.31463611664753, "band_mae_db": 0.2510564208065383, "rms_gain_error_db": 0.13205958684201136, "output_peak": 0.45683225791231075}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5830710745299714, "全帯域_10ms_rms_envelope_nrmse": 0.1501088641058838, "全帯域_10ms_envelope_correlation": 0.980307715651274, "全帯域_10ms_modulation_mae_db": 0.7764550600569007, "全帯域_40ms_log_rms_mae_db": 0.955601542524067, "全帯域_40ms_rms_envelope_nrmse": 0.08643184580093259, "全帯域_40ms_envelope_correlation": 0.992941608915771, "全帯域_40ms_modulation_mae_db": 0.9659398159051295, "300–1000 Hz_10ms_log_rms_mae_db": 2.5144221526135224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15698475134022505, "300–1000 Hz_10ms_envelope_correlation": 0.9830094677048575, "300–1000 Hz_10ms_modulation_mae_db": 0.5828283274972006, "300–1000 Hz_40ms_log_rms_mae_db": 1.510228821626273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07623848950672713, "300–1000 Hz_40ms_envelope_correlation": 0.996058259143359, "300–1000 Hz_40ms_modulation_mae_db": 0.5077456971187614, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2866349137897664, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20511354678882535, "1000–3000 Hz_10ms_envelope_correlation": 0.9684354962859566, "1000–3000 Hz_10ms_modulation_mae_db": 1.7825085819577025, "1000–3000 Hz_40ms_log_rms_mae_db": 1.499152732962451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1275311990348691, "1000–3000 Hz_40ms_envelope_correlation": 0.9871485442315442, "1000–3000 Hz_40ms_modulation_mae_db": 1.8856202796154067, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1067888108923922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19709211008316563, "3000–6000 Hz_10ms_envelope_correlation": 0.9780430670161318, "3000–6000 Hz_10ms_modulation_mae_db": 1.2518578565204177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4634155274667056, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11376103387377252, "3000–6000 Hz_40ms_envelope_correlation": 0.9925706557316266, "3000–6000 Hz_40ms_modulation_mae_db": 1.4190053459648284, "mr_stft_db": 6.301733955606281, "band_mae_db": 0.23003928959414, "rms_gain_error_db": 0.030783147634826768, "output_peak": 0.34935702013652625}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76160, "output_samples": 76160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.4443052895079, "spectral_convergence": 0.5036993714694205, "spectral_convergence_status": "ok", "amplitude_floor": 1.3356779608968942e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.142474965663901, "spectral_convergence": 0.4925456415391898, "spectral_convergence_status": "ok", "amplitude_floor": 9.962006407647762e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.346607436762242, "spectral_convergence": 0.4974420877478936, "spectral_convergence_status": "ok", "amplitude_floor": 4.291398708237255e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021802020252140593, "output_power": 0.00022743023463688155, "delta_db": 0.18351460686892873, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019115120810434436, "output_power": 0.0001930887295075495, "delta_db": 0.043798777976809394, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.836069030265609e-05, "output_power": 4.9204151552571096e-05, "delta_db": 0.07509256117228578, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.9699402895957116e-05, "output_power": 4.360426313932554e-05, "delta_db": -0.5682221919301297, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001234531768639194, "output_power": 0.00011801301257638229, "delta_db": -0.19572373309274593, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00036807758285640346, "output_power": 0.0003362370187712082, "delta_db": -0.3929384191602642, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24321504836686061, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13776404514057056, "pitch": null, "output_rms_dbfs": -30.143142795691432, "output_peak": 0.3799772695177661, "reference_rms_dbfs": -30.00537875055086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.311129230644681}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.00018341452353274373, "log_rms_mae_db": 1.5400242046474837, "rms_envelope_nrmse": 0.14055973479993014, "envelope_correlation": 0.9824876695582943, "rms_gain_error_db": -0.13779388554840788, "reference_mean_power": 0.0009998027766238483, "output_mean_power": 0.0009685787796678964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 5.754282242130778e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018894414816227902, "output_power": 0.0001804578566624863, "reference_active": true, "delta_db": -0.19957650963894347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013084274193993133, "output_power": 9.920199914071706e-05, "reference_active": true, "delta_db": -1.2022921238036968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.690431882761885e-05, "output_power": 1.9672690872343262e-05, "reference_active": true, "delta_db": -2.7321342591646975}], "active_band_count": 3, "active_band_mae_db": 1.378000964202446}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.0001230241515816465, "log_rms_mae_db": 0.843715110375738, "rms_envelope_nrmse": 0.07898976527922698, "envelope_correlation": 0.9941911714981233, "rms_gain_error_db": -0.13779743741986952, "reference_mean_power": 0.001006100998286567, "output_mean_power": 0.0009746795097432219}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 5.061730772699165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019045538591950462, "output_power": 0.0001721200383681887, "reference_active": true, "delta_db": -0.43961824435029656}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.588390935994399e-05, "output_power": 5.300787659592214e-05, "reference_active": true, "delta_db": -1.558092891326115}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5269059669432984e-06, "output_power": 1.0088218094992378e-06, "reference_active": true, "delta_db": -3.987746183550693}], "active_band_count": 3, "active_band_mae_db": 1.9951524397423681}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 5.489840871101384e-05, "log_rms_mae_db": 2.639061953885179, "rms_envelope_nrmse": 0.20930192529570169, "envelope_correlation": 0.9696301361155416, "rms_gain_error_db": -0.014934322194366186, "reference_mean_power": 0.0001499835837122031, "output_mean_power": 0.00014946871271382382}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.245105068973029e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.560041827470393e-05, "output_power": 5.825228683660863e-05, "reference_active": true, "delta_db": -0.5159362915932587}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1382200906061474e-05, "output_power": 1.7005063438903508e-05, "reference_active": true, "delta_db": -0.9947414939227617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4492416985862286e-06, "output_power": 2.4875404656021935e-06, "reference_active": true, "delta_db": 0.06738509546126732}], "active_band_count": 3, "active_band_mae_db": 0.5260209603257626}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 4.550505877934166e-05, "log_rms_mae_db": 1.652049848538113, "rms_envelope_nrmse": 0.11398748161247883, "envelope_correlation": 0.990524329436297, "rms_gain_error_db": -0.01493891795266253, "reference_mean_power": 0.00015093488949896852, "output_mean_power": 0.00015041659363865435}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 1.1354747567630587e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.429099822482352e-05, "output_power": 5.680816813528113e-05, "reference_active": true, "delta_db": -0.5373938393390154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916365304226932e-05, "output_power": 9.00495494732661e-06, "reference_active": true, "delta_db": -0.8359651608786106}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3162457402511593e-07, "output_power": 2.0799376639663374e-07, "reference_active": true, "delta_db": -0.4673431433856594}], "active_band_count": 3, "active_band_mae_db": 0.6135673812010952}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 4.066436664003698e-05, "log_rms_mae_db": 2.195425346838845, "rms_envelope_nrmse": 0.19155659560675123, "envelope_correlation": 0.9727277669728903, "rms_gain_error_db": -0.2687325030779187, "reference_mean_power": 0.0001020893317881212, "output_mean_power": 9.596372795599224e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 9.602503308503003e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.081225810114696e-05, "output_power": 2.6620439765435207e-05, "reference_active": true, "delta_db": -0.6350830143032855}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3389822486480084e-05, "output_power": 1.0246986045727155e-05, "reference_active": true, "delta_db": -1.1617867466984029}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310273392307641e-06, "output_power": 4.004638335242195e-06, "reference_active": true, "delta_db": -1.2255358030630361}], "active_band_count": 3, "active_band_mae_db": 1.0074685213549082}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 3.8037796302268945e-05, "log_rms_mae_db": 1.396944325292114, "rms_envelope_nrmse": 0.11706907056371779, "envelope_correlation": 0.9897477382040352, "rms_gain_error_db": -0.2687393432027919, "reference_mean_power": 0.00010273559922291735, "output_mean_power": 9.657106570130173e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 8.714815910547825e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9375203264151156e-05, "output_power": 2.5537509539972356e-05, "reference_active": true, "delta_db": -0.6080233854816854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.005890574709935e-06, "output_power": 5.198856794498697e-06, "reference_active": true, "delta_db": -1.295554950644706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.15357493378345e-07, "output_power": 3.233091723025977e-07, "reference_active": true, "delta_db": -2.0249056938794117}], "active_band_count": 3, "active_band_mae_db": 1.309494676668601}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001808188641270182, "log_rms_mae_db": 2.1026627955569803, "rms_envelope_nrmse": 0.1845784964412195, "envelope_correlation": 0.9812194991406256, "rms_gain_error_db": -0.4171317013553422, "reference_mean_power": 0.0004025082939999733, "output_mean_power": 0.0003656467045594377}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 3.5966286763117237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001367199427375967, "output_power": 0.00011515161170943575, "reference_active": true, "delta_db": -0.7456184690977097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.299777768885521e-05, "output_power": 5.8985066604444906e-05, "reference_active": true, "delta_db": -1.9773049646031857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495937881252574e-05, "output_power": 8.460572582016513e-06, "reference_active": true, "delta_db": -4.698340168826746}], "active_band_count": 3, "active_band_mae_db": 2.4737545341758804}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.00011978792663441872, "log_rms_mae_db": 1.4491183575902262, "rms_envelope_nrmse": 0.11662190136439363, "envelope_correlation": 0.9932002387748137, "rms_gain_error_db": -0.41713519601135557, "reference_mean_power": 0.0004050621320160614, "output_mean_power": 0.00036796636675898425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 3.450538172499259e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016854041012359937, "output_power": 0.0001284347836289513, "reference_active": true, "delta_db": -1.1802138798364008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.211674623356507e-05, "output_power": 3.085931470956305e-05, "reference_active": true, "delta_db": -2.275910164094293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0452544406168747e-06, "output_power": 5.976869601706955e-07, "reference_active": true, "delta_db": -5.3427356339930245}], "active_band_count": 3, "active_band_mae_db": 2.932953225974573}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.6910768293664734, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9d424250f3e022d3b7de438c0797baf31b11bef1357254980549419d14089aac", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9667075787247117, "standard_deviation": 0.0033339930307245153, "minimum": 0.9619586555103362, "maximum": 0.9727277669728903, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.283458313751895, "standard_deviation": 0.0461177216634795, "minimum": 2.195425346838845, "maximum": 2.342468609575455, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.5302011038322143, "standard_deviation": 0.363125292133527, "minimum": 1.0074685213549082, "maximum": 1.9489949367669632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2110324492052275, "standard_deviation": 0.010738144298681439, "minimum": 0.19155659560675123, "maximum": 0.2271741104748798, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9879851121729973, "standard_deviation": 0.0017626139713212046, "minimum": 0.9849057105546531, "maximum": 0.9898119764241528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4414677518247125, "standard_deviation": 0.055261423320121426, "minimum": 1.3701643116499302, "maximum": 1.5232314864240606, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.8544876475212448, "standard_deviation": 0.43614528319203666, "minimum": 1.309494676668601, "maximum": 2.529394368046273, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12606279178911434, "standard_deviation": 0.010093540575282773, "minimum": 0.11625417844425665, "maximum": 0.14662256480153582, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9818608898906388, "standard_deviation": 0.0018761838197966223, "minimum": 0.9780430670161318, "maximum": 0.9841707155639008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1108243279931047, "standard_deviation": 0.03169520314638222, "minimum": 2.0747150881356773, "maximum": 2.177558042715289, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0526471689095043, "standard_deviation": 0.6551784837753112, "minimum": 0.34907514190451644, "maximum": 2.4737545341758804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18069459351285333, "standard_deviation": 0.008883759543700541, "minimum": 0.16663997597117464, "maximum": 0.19709211008316563, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9938481640363568, "standard_deviation": 0.00193144877618762, "minimum": 0.9898989938187022, "maximum": 0.9958233330214344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4739884243120733, "standard_deviation": 0.022277271197467457, "minimum": 1.4416365984549195, "maximum": 1.5113538223614607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.134897340448108, "standard_deviation": 0.7979941112139326, "minimum": 0.2753121725603596, "maximum": 2.932953225974573, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10587968848742199, "standard_deviation": 0.01634125439471564, "minimum": 0.08535751125174895, "maximum": 0.13259667175409734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9787658369821828, "standard_deviation": 0.004175830541939946, "minimum": 0.9696301361155416, "maximum": 0.9830094677048575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.6761046843691716, "standard_deviation": 0.07592528838897554, "minimum": 2.5144221526135224, "maximum": 2.770969397668642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.603384710430863, "standard_deviation": 0.20594271672999337, "minimum": 0.3863524383596868, "maximum": 1.0484339225110013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17500309989778606, "standard_deviation": 0.015931353554364956, "minimum": 0.15698475134022505, "maximum": 0.20930192529570169, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.994067078249117, "standard_deviation": 0.0018145908602124454, "minimum": 0.990524329436297, "maximum": 0.996058259143359, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.652874671417811, "standard_deviation": 0.07099524963503696, "minimum": 1.510228821626273, "maximum": 1.750927939797489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8097888307038418, "standard_deviation": 0.36102474157696, "minimum": 0.46351656106109157, "maximum": 1.5644390898470044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09309421553624453, "standard_deviation": 0.01210763683576801, "minimum": 0.07623848950672713, "maximum": 0.11398748161247883, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21278014219373353, "standard_deviation": 0.05108532042107082, "minimum": 0.11670859660361725, "maximum": 0.261559842965752, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.3149298150044695, "standard_deviation": 0.008657957549549006, "minimum": 6.301733955606281, "maximum": 6.331535968054631, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.39408890322557466, "standard_deviation": 0.039010220143971625, "minimum": 0.3485395426739418, "maximum": 0.45683225791231075, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.044179273821992694, "standard_deviation": 0.0954586867275432, "minimum": -0.13776404514057056, "maximum": 0.13205958684201136, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9825651535154294, "standard_deviation": 0.0014200532084419795, "minimum": 0.980307715651274, "maximum": 0.9849923676685416, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5172088288129681, "standard_deviation": 0.05741649597176713, "minimum": 1.434599677927054, "maximum": 1.5830710745299714, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6708162108700395, "standard_deviation": 0.3599419392731864, "minimum": 0.11167520384792545, "maximum": 1.378000964202446, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14104808260143326, "standard_deviation": 0.006539161553975831, "minimum": 0.13038490431082944, "maximum": 0.1501088641058838, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9942277386221355, "standard_deviation": 0.0017611063147340506, "minimum": 0.991139092924791, "maximum": 0.9959563332792203, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.834722046482446, "standard_deviation": 0.07990348676033246, "minimum": 0.7158174319192017, "maximum": 0.955601542524067, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8073750018179243, "standard_deviation": 0.5353722103577055, "minimum": 0.16514402086743288, "maximum": 1.9951524397423681, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07803809422084318, "standard_deviation": 0.011927910211671645, "minimum": 0.06539817909131784, "maximum": 0.0971844395360949, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0003", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.373413324356079, "predicted_mos": 2.373413324356079, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "a8a5ded4ada5c86087173b5283858c20cad2bfa8758bd06034309bc505e34247", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.721267099957913, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6376230128931069, "全帯域_10ms_rms_envelope_nrmse": 0.14616358374575888, "全帯域_10ms_envelope_correlation": 0.9815122502823587, "全帯域_10ms_modulation_mae_db": 0.18399304764419866, "全帯域_40ms_log_rms_mae_db": 0.9866471971905011, "全帯域_40ms_rms_envelope_nrmse": 0.0807812678276315, "全帯域_40ms_envelope_correlation": 0.9938929276796439, "全帯域_40ms_modulation_mae_db": 0.34915547324639634, "300–1000 Hz_10ms_log_rms_mae_db": 1.7807692811154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1822036062759021, "300–1000 Hz_10ms_envelope_correlation": 0.9768597272688718, "300–1000 Hz_10ms_modulation_mae_db": 0.7717154413070433, "300–1000 Hz_40ms_log_rms_mae_db": 0.9869242473408026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09097461797237143, "300–1000 Hz_40ms_envelope_correlation": 0.994066830452925, "300–1000 Hz_40ms_modulation_mae_db": 0.509927677226715, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4632337932450707, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20918162992401776, "1000–3000 Hz_10ms_envelope_correlation": 0.9681883456053632, "1000–3000 Hz_10ms_modulation_mae_db": 0.6529623921209375, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8583319436150753, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13609867627902483, "1000–3000 Hz_40ms_envelope_correlation": 0.985618110672227, "1000–3000 Hz_40ms_modulation_mae_db": 0.9499809802537259, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3903241590913353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1708610358183916, "3000–6000 Hz_10ms_envelope_correlation": 0.9838188755699225, "3000–6000 Hz_10ms_modulation_mae_db": 0.4794802795906552, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8334635175469507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09662946942974505, "3000–6000 Hz_40ms_envelope_correlation": 0.9947640452771237, "3000–6000 Hz_40ms_modulation_mae_db": 0.5874252652526929, "mr_stft_db": 5.895118236430927, "band_mae_db": 0.17270259320862955, "rms_gain_error_db": 0.0728764634369162, "output_peak": 0.3782117321057057}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7549676488490773, "全帯域_10ms_rms_envelope_nrmse": 0.15810507483828173, "全帯域_10ms_envelope_correlation": 0.977756513707868, "全帯域_10ms_modulation_mae_db": 0.9912932530454541, "全帯域_40ms_log_rms_mae_db": 0.9780432499447552, "全帯域_40ms_rms_envelope_nrmse": 0.08404932910727747, "全帯域_40ms_envelope_correlation": 0.9934436737366792, "全帯域_40ms_modulation_mae_db": 1.289144871437312, "300–1000 Hz_10ms_log_rms_mae_db": 1.9096392885189164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18654972315235374, "300–1000 Hz_10ms_envelope_correlation": 0.9757627870407888, "300–1000 Hz_10ms_modulation_mae_db": 0.42524843075613167, "300–1000 Hz_40ms_log_rms_mae_db": 1.0411731834522393, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11070691528984153, "300–1000 Hz_40ms_envelope_correlation": 0.99125429755761, "300–1000 Hz_40ms_modulation_mae_db": 0.5942081117370592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4860263614798526, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19787902212988967, "1000–3000 Hz_10ms_envelope_correlation": 0.9706692269808451, "1000–3000 Hz_10ms_modulation_mae_db": 0.9277096870226541, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8793693987679077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.123562368291425, "1000–3000 Hz_40ms_envelope_correlation": 0.9877499000498722, "1000–3000 Hz_40ms_modulation_mae_db": 1.297630565331082, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3780169277201975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1903600968864766, "3000–6000 Hz_10ms_envelope_correlation": 0.9796356662337614, "3000–6000 Hz_10ms_modulation_mae_db": 1.6898448591644648, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8279959139061354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10694239178610593, "3000–6000 Hz_40ms_envelope_correlation": 0.994230655997462, "3000–6000 Hz_40ms_modulation_mae_db": 1.6440501371789777, "mr_stft_db": 5.8865456591716905, "band_mae_db": 0.29429080516640355, "rms_gain_error_db": -0.1297053409223685, "output_peak": 0.415959788757825}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7039390673114363, "全帯域_10ms_rms_envelope_nrmse": 0.15393876520097516, "全帯域_10ms_envelope_correlation": 0.9795385170281927, "全帯域_10ms_modulation_mae_db": 0.3628564786650994, "全帯域_40ms_log_rms_mae_db": 1.0276575475182024, "全帯域_40ms_rms_envelope_nrmse": 0.1023250328692267, "全帯域_40ms_envelope_correlation": 0.9909698530140082, "全帯域_40ms_modulation_mae_db": 0.44516779461360195, "300–1000 Hz_10ms_log_rms_mae_db": 1.8205209329209264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18587353746403676, "300–1000 Hz_10ms_envelope_correlation": 0.9759396100293009, "300–1000 Hz_10ms_modulation_mae_db": 0.4004037697938451, "300–1000 Hz_40ms_log_rms_mae_db": 1.0092554695287568, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10774508784720997, "300–1000 Hz_40ms_envelope_correlation": 0.9917570514993654, "300–1000 Hz_40ms_modulation_mae_db": 0.32236241854957176, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4380374409140808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1905025127684907, "1000–3000 Hz_10ms_envelope_correlation": 0.9737883181270821, "1000–3000 Hz_10ms_modulation_mae_db": 1.1086704645819696, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8411795004465865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11564197398631461, "1000–3000 Hz_40ms_envelope_correlation": 0.9912530362947841, "1000–3000 Hz_40ms_modulation_mae_db": 1.1653296040370922, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4202460394101628, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1997727846229318, "3000–6000 Hz_10ms_envelope_correlation": 0.9786181189050107, "3000–6000 Hz_10ms_modulation_mae_db": 0.43823475604850826, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8647976933402846, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14488220240533317, "3000–6000 Hz_40ms_envelope_correlation": 0.9899098715843122, "3000–6000 Hz_40ms_modulation_mae_db": 0.4263399119540238, "mr_stft_db": 5.882682164557434, "band_mae_db": 0.33741754248409017, "rms_gain_error_db": -0.32868123608500494, "output_peak": 0.3602539295670434}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7164594280612873, "全帯域_10ms_rms_envelope_nrmse": 0.15260753072118957, "全帯域_10ms_envelope_correlation": 0.9796262856816992, "全帯域_10ms_modulation_mae_db": 0.29609286206995633, "全帯域_40ms_log_rms_mae_db": 0.9846611322455542, "全帯域_40ms_rms_envelope_nrmse": 0.08319773521397525, "全帯域_40ms_envelope_correlation": 0.9934403426463646, "全帯域_40ms_modulation_mae_db": 0.3577195803803436, "300–1000 Hz_10ms_log_rms_mae_db": 1.7838272434003721, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16094236697982311, "300–1000 Hz_10ms_envelope_correlation": 0.9823336627437413, "300–1000 Hz_10ms_modulation_mae_db": 0.5510659190300177, "300–1000 Hz_40ms_log_rms_mae_db": 1.023840185081808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09092886004945155, "300–1000 Hz_40ms_envelope_correlation": 0.9946865353366883, "300–1000 Hz_40ms_modulation_mae_db": 0.550612325111425, "1000–3000 Hz_10ms_log_rms_mae_db": 1.464975250375498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1839926712471113, "1000–3000 Hz_10ms_envelope_correlation": 0.9751566886098066, "1000–3000 Hz_10ms_modulation_mae_db": 1.2884263348160607, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7889734426758962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10520871181061499, "1000–3000 Hz_40ms_envelope_correlation": 0.9923824803303242, "1000–3000 Hz_40ms_modulation_mae_db": 1.6762522673332085, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4089869659741916, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19795064469388043, "3000–6000 Hz_10ms_envelope_correlation": 0.9779135439607995, "3000–6000 Hz_10ms_modulation_mae_db": 0.47216355037138324, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8475575316557303, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11506184658889856, "3000–6000 Hz_40ms_envelope_correlation": 0.9923454043599264, "3000–6000 Hz_40ms_modulation_mae_db": 0.4660058855859257, "mr_stft_db": 5.898361524401015, "band_mae_db": 0.20075953354636697, "rms_gain_error_db": 0.012836886043043625, "output_peak": 0.3873686190055012}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7486804475171365, "全帯域_10ms_rms_envelope_nrmse": 0.1628564832180869, "全帯域_10ms_envelope_correlation": 0.9764420666297781, "全帯域_10ms_modulation_mae_db": 0.6683006002066034, "全帯域_40ms_log_rms_mae_db": 1.1017665749330114, "全帯域_40ms_rms_envelope_nrmse": 0.11093944331209445, "全帯域_40ms_envelope_correlation": 0.9882733514916483, "全帯域_40ms_modulation_mae_db": 0.7137357714978441, "300–1000 Hz_10ms_log_rms_mae_db": 1.8101183366816564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16578950354135422, "300–1000 Hz_10ms_envelope_correlation": 0.9812722026285436, "300–1000 Hz_10ms_modulation_mae_db": 0.5349457880559316, "300–1000 Hz_40ms_log_rms_mae_db": 1.076204239347208, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0998775842670378, "300–1000 Hz_40ms_envelope_correlation": 0.9935129124528186, "300–1000 Hz_40ms_modulation_mae_db": 0.44044178128506156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4939565740449576, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.201986365442797, "1000–3000 Hz_10ms_envelope_correlation": 0.9703898367923913, "1000–3000 Hz_10ms_modulation_mae_db": 0.9969141964785226, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8340739590429876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12122310324694698, "1000–3000 Hz_40ms_envelope_correlation": 0.990419889503193, "1000–3000 Hz_40ms_modulation_mae_db": 1.270558169689321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3704177657309233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20613015615477176, "3000–6000 Hz_10ms_envelope_correlation": 0.9757241969694601, "3000–6000 Hz_10ms_modulation_mae_db": 0.891616242093996, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8601440839888508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15379239330596053, "3000–6000 Hz_40ms_envelope_correlation": 0.9861332593715487, "3000–6000 Hz_40ms_modulation_mae_db": 0.7465493176682824, "mr_stft_db": 5.900488085273774, "band_mae_db": 0.21812067246860453, "rms_gain_error_db": -0.13719329173886205, "output_peak": 0.3948401648083354}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5446203179432563, "全帯域_10ms_rms_envelope_nrmse": 0.14504339503314825, "全帯域_10ms_envelope_correlation": 0.9814547442324183, "全帯域_10ms_modulation_mae_db": 0.45933839162855966, "全帯域_40ms_log_rms_mae_db": 0.8352447789796592, "全帯域_40ms_rms_envelope_nrmse": 0.08208479343118708, "全帯域_40ms_envelope_correlation": 0.9936359600063778, "全帯域_40ms_modulation_mae_db": 0.3639954433341941, "300–1000 Hz_10ms_log_rms_mae_db": 1.8576407383490836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18393054079314936, "300–1000 Hz_10ms_envelope_correlation": 0.9764403495025705, "300–1000 Hz_10ms_modulation_mae_db": 0.5294790901058154, "300–1000 Hz_40ms_log_rms_mae_db": 1.02880994979092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09221009109798092, "300–1000 Hz_40ms_envelope_correlation": 0.9938928765058099, "300–1000 Hz_40ms_modulation_mae_db": 0.8742379797042318, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4223521307972227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18380791228336554, "1000–3000 Hz_10ms_envelope_correlation": 0.9752176251081406, "1000–3000 Hz_10ms_modulation_mae_db": 1.6480206968921711, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8416081530041817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11111156348489261, "1000–3000 Hz_40ms_envelope_correlation": 0.9913248246440053, "1000–3000 Hz_40ms_modulation_mae_db": 1.7702262897197258, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4035797342547416, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17681722491482194, "3000–6000 Hz_10ms_envelope_correlation": 0.9821772651849817, "3000–6000 Hz_10ms_modulation_mae_db": 0.5694986909977917, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8648110754718538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1114628247564278, "3000–6000 Hz_40ms_envelope_correlation": 0.9927457390104114, "3000–6000 Hz_40ms_modulation_mae_db": 0.4578899924979655, "mr_stft_db": 5.891551999094699, "band_mae_db": 0.24886476677246197, "rms_gain_error_db": -0.05801052360092401, "output_peak": 0.4042603084740907}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5652566587441408, "全帯域_10ms_rms_envelope_nrmse": 0.14984580739732664, "全帯域_10ms_envelope_correlation": 0.9801671717568939, "全帯域_10ms_modulation_mae_db": 0.7333519796726122, "全帯域_40ms_log_rms_mae_db": 0.8807548989537989, "全帯域_40ms_rms_envelope_nrmse": 0.07818936451327446, "全帯域_40ms_envelope_correlation": 0.9941620817909463, "全帯域_40ms_modulation_mae_db": 0.6074247705359427, "300–1000 Hz_10ms_log_rms_mae_db": 1.8450985341847923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15693730444126663, "300–1000 Hz_10ms_envelope_correlation": 0.9832606528437238, "300–1000 Hz_10ms_modulation_mae_db": 0.777315085704042, "300–1000 Hz_40ms_log_rms_mae_db": 1.0031863032885708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09580563706545026, "300–1000 Hz_40ms_envelope_correlation": 0.9940068158631601, "300–1000 Hz_40ms_modulation_mae_db": 0.649283814800799, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4174289281051582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18106635691221482, "1000–3000 Hz_10ms_envelope_correlation": 0.9756960873572932, "1000–3000 Hz_10ms_modulation_mae_db": 1.097525077545513, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8123121390280938, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09919724795189823, "1000–3000 Hz_40ms_envelope_correlation": 0.9929500124011593, "1000–3000 Hz_40ms_modulation_mae_db": 1.339210728195375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4279564711697974, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2022621673669608, "3000–6000 Hz_10ms_envelope_correlation": 0.9770115753524, "3000–6000 Hz_10ms_modulation_mae_db": 0.8209848640780498, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8911675144664264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10919947154260314, "3000–6000 Hz_40ms_envelope_correlation": 0.9931561110173808, "3000–6000 Hz_40ms_modulation_mae_db": 0.8495734914445289, "mr_stft_db": 5.89563208519553, "band_mae_db": 0.166438320702659, "rms_gain_error_db": -0.03396907039040114, "output_peak": 0.3347145990698354}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.678485575069679, "全帯域_10ms_rms_envelope_nrmse": 0.1528985546540982, "全帯域_10ms_envelope_correlation": 0.9793799493814241, "全帯域_10ms_modulation_mae_db": 1.2965438675947194, "全帯域_40ms_log_rms_mae_db": 0.8734340714092257, "全帯域_40ms_rms_envelope_nrmse": 0.08109886529211224, "全帯域_40ms_envelope_correlation": 0.9941701300024169, "全帯域_40ms_modulation_mae_db": 1.6197807337968477, "300–1000 Hz_10ms_log_rms_mae_db": 1.9567632566758073, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17566484026021387, "300–1000 Hz_10ms_envelope_correlation": 0.9788910797463367, "300–1000 Hz_10ms_modulation_mae_db": 0.5686556753332113, "300–1000 Hz_40ms_log_rms_mae_db": 1.0968054529472495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09849414222867507, "300–1000 Hz_40ms_envelope_correlation": 0.9937965774842477, "300–1000 Hz_40ms_modulation_mae_db": 0.7865275212793179, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4843737032490076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22648847765703212, "1000–3000 Hz_10ms_envelope_correlation": 0.9629647782808687, "1000–3000 Hz_10ms_modulation_mae_db": 0.9483143435586929, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8445437507592212, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14772739566207735, "1000–3000 Hz_40ms_envelope_correlation": 0.9857042096964356, "1000–3000 Hz_40ms_modulation_mae_db": 1.4492808981888092, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4541622977135447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19229102075727647, "3000–6000 Hz_10ms_envelope_correlation": 0.9789236366346491, "3000–6000 Hz_10ms_modulation_mae_db": 1.8825104818619598, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8398346793000652, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10373886164163626, "3000–6000 Hz_40ms_envelope_correlation": 0.9941260671119666, "3000–6000 Hz_40ms_modulation_mae_db": 2.068624670116923, "mr_stft_db": 5.8983407639165355, "band_mae_db": 0.288383562456522, "rms_gain_error_db": -0.2209452357529329, "output_peak": 0.40165627687955957}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76160, "output_samples": 76160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.940110744958393, "spectral_convergence": 0.47579435957660654, "spectral_convergence_status": "ok", "amplitude_floor": 1.3356779608968942e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.631156113793007, "spectral_convergence": 0.45854382830288454, "spectral_convergence_status": "ok", "amplitude_floor": 9.962006407647762e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.11408785054138, "spectral_convergence": 0.47971116491365573, "spectral_convergence_status": "ok", "amplitude_floor": 4.291398708237255e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021802020252140593, "output_power": 0.0002197883757214077, "delta_db": 0.03507980780334052, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019115120810434436, "output_power": 0.00018938878800356997, "delta_db": -0.040227823883080535, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.836069030265609e-05, "output_power": 4.771840764618357e-05, "delta_db": -0.05806548452617432, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.9699402895957116e-05, "output_power": 5.372421736290973e-05, "delta_db": 0.33818926531412397, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001234531768639194, "output_power": 0.0001395914147921035, "delta_db": 0.5335643868879113, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00036807758285640346, "output_power": 0.00036545212303961604, "delta_db": -0.031088790837146375, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17270259320862955, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.0728764634369162, "pitch": null, "output_rms_dbfs": -29.932502287113945, "output_peak": 0.3782117321057057, "reference_rms_dbfs": -30.00537875055086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.895118236430927}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.00018341452353274373, "log_rms_mae_db": 1.6376230128931069, "rms_envelope_nrmse": 0.14616358374575888, "envelope_correlation": 0.9815122502823587, "rms_gain_error_db": 0.07281433300144499, "reference_mean_power": 0.0009998027766238483, "output_mean_power": 0.001016706902040147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 5.754282242130778e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018894414816227902, "output_power": 0.0001814865509192872, "reference_active": true, "delta_db": -0.1748899869097153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013084274193993133, "output_power": 0.00013113860031619358, "reference_active": true, "delta_db": 0.009809073711270579}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.690431882761885e-05, "output_power": 3.3911670624442095e-05, "reference_active": true, "delta_db": -0.36728008231161013}], "active_band_count": 3, "active_band_mae_db": 0.18399304764419866}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.0001230241515816465, "log_rms_mae_db": 0.9866471971905011, "rms_envelope_nrmse": 0.0807812678276315, "envelope_correlation": 0.9938929276796439, "rms_gain_error_db": 0.0727891885527521, "reference_mean_power": 0.001006100998286567, "output_mean_power": 0.0010231056871175084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 5.061730772699165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019045538591950462, "output_power": 0.00018141623747365063, "reference_active": true, "delta_db": -0.21117103112004904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.588390935994399e-05, "output_power": 7.522426146741496e-05, "reference_active": true, "delta_db": -0.03791763979099903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5269059669432984e-06, "output_power": 2.1025740791196698e-06, "reference_active": true, "delta_db": -0.7983777488281409}], "active_band_count": 3, "active_band_mae_db": 0.34915547324639634}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 5.489840871101384e-05, "log_rms_mae_db": 1.7807692811154, "rms_envelope_nrmse": 0.1822036062759021, "envelope_correlation": 0.9768597272688718, "rms_gain_error_db": -0.1507292949067413, "reference_mean_power": 0.0001499835837122031, "output_mean_power": 0.0001448674437742889}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.245105068973029e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.560041827470393e-05, "output_power": 5.953661549969413e-05, "reference_active": true, "delta_db": -0.42122465977563317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1382200906061474e-05, "output_power": 1.669035546320506e-05, "reference_active": true, "delta_db": -1.0758681966445147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4492416985862286e-06, "output_power": 2.956902411593887e-06, "reference_active": true, "delta_db": 0.8180534675009818}], "active_band_count": 3, "active_band_mae_db": 0.7717154413070433}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 4.550505877934166e-05, "log_rms_mae_db": 0.9869242473408026, "rms_envelope_nrmse": 0.09097461797237143, "envelope_correlation": 0.994066830452925, "rms_gain_error_db": -0.15074471843820675, "reference_mean_power": 0.00015093488949896852, "output_mean_power": 0.0001457857815083204}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 1.1354747567630587e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.429099822482352e-05, "output_power": 5.8124806857996175e-05, "reference_active": true, "delta_db": -0.43788646168861195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916365304226932e-05, "output_power": 8.845215366849045e-06, "reference_active": true, "delta_db": -0.9136964868033631}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3162457402511593e-07, "output_power": 2.4132629309758095e-07, "reference_active": true, "delta_db": 0.17820008318816985}], "active_band_count": 3, "active_band_mae_db": 0.509927677226715}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 4.066436664003698e-05, "log_rms_mae_db": 1.4632337932450707, "rms_envelope_nrmse": 0.20918162992401776, "envelope_correlation": 0.9681883456053632, "rms_gain_error_db": 0.1275161916539076, "reference_mean_power": 0.0001020893317881212, "output_mean_power": 0.0001051312868465906}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 9.602503308503003e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.081225810114696e-05, "output_power": 3.3089195588833664e-05, "reference_active": true, "delta_db": 0.3096268231696396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3389822486480084e-05, "output_power": 1.7901248682741727e-05, "reference_active": true, "delta_db": 1.2610850633289812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310273392307641e-06, "output_power": 4.856231165138643e-06, "reference_active": true, "delta_db": -0.38817528986419186}], "active_band_count": 3, "active_band_mae_db": 0.6529623921209375}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 3.8037796302268945e-05, "log_rms_mae_db": 0.8583319436150753, "rms_envelope_nrmse": 0.13609867627902483, "envelope_correlation": 0.985618110672227, "rms_gain_error_db": 0.1275274793221966, "reference_mean_power": 0.00010273559922291735, "output_mean_power": 0.00010579708608223527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 8.714815910547825e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9375203264151156e-05, "output_power": 3.258198088832556e-05, "reference_active": true, "delta_db": 0.4499660417173438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.005890574709935e-06, "output_power": 9.770857730958385e-06, "reference_active": true, "delta_db": 1.4446933999352773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.15357493378345e-07, "output_power": 4.1359972871456617e-07, "reference_active": true, "delta_db": -0.9552834991085569}], "active_band_count": 3, "active_band_mae_db": 0.9499809802537259}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001808188641270182, "log_rms_mae_db": 1.3903241590913353, "rms_envelope_nrmse": 0.1708610358183916, "envelope_correlation": 0.9838188755699225, "rms_gain_error_db": 0.1277651452544268, "reference_mean_power": 0.0004025082939999733, "output_mean_power": 0.0004145255920899825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 3.5966286763117237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001367199427375967, "output_power": 0.00012858313861917426, "reference_active": true, "delta_db": -0.2664784545571169}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.299777768885521e-05, "output_power": 8.334518564246486e-05, "reference_active": true, "delta_db": -0.47592052338087143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495937881252574e-05, "output_power": 2.126324711072301e-05, "reference_active": true, "delta_db": -0.6960418608339773}], "active_band_count": 3, "active_band_mae_db": 0.4794802795906552}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.00011978792663441872, "log_rms_mae_db": 0.8334635175469507, "rms_envelope_nrmse": 0.09662946942974505, "envelope_correlation": 0.9947640452771237, "rms_gain_error_db": 0.12776586175598995, "reference_mean_power": 0.0004050621320160614, "output_mean_power": 0.00041715574638286534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 3.450538172499259e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016854041012359937, "output_power": 0.0001548911127196043, "reference_active": true, "delta_db": -0.366775466931499}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.211674623356507e-05, "output_power": 4.614719745973853e-05, "reference_active": true, "delta_db": -0.5283196263772514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0452544406168747e-06, "output_power": 1.6750557809052076e-06, "reference_active": true, "delta_db": -0.8671807024493283}], "active_band_count": 3, "active_band_mae_db": 0.5874252652526929}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7405149713729582, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a3aabcd66d80c035c8f26285494a9112ac1586afba77cf03cd1b204e389e15eb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9715088633577238, "standard_deviation": 0.004404133436780665, "minimum": 0.9629647782808687, "maximum": 0.9756960873572932, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.458798022776356, "standard_deviation": 0.029665185423057128, "minimum": 1.4174289281051582, "maximum": 1.4939565740449576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0835678991270652, "standard_deviation": 0.2920161821887363, "minimum": 0.6529623921209375, "maximum": 1.6480206968921711, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19686311854561486, "standard_deviation": 0.015504695842090552, "minimum": 0.18106635691221482, "maximum": 0.22648847765703212, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9896753079490002, "standard_deviation": 0.002921708696597904, "minimum": 0.985618110672227, "maximum": 0.9929500124011593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8375490359174937, "standard_deviation": 0.027433087748470728, "minimum": 0.7889734426758962, "maximum": 0.8793693987679077, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3648086878435424, "standard_deviation": 0.2657660994054351, "minimum": 0.9499809802537259, "maximum": 1.7702262897197258, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11997138008914932, "standard_deviation": 0.015998612164075734, "minimum": 0.09919724795189823, "maximum": 0.14772739566207735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792278598513732, "standard_deviation": 0.0026544317609119208, "minimum": 0.9757241969694601, "maximum": 0.9838188755699225, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4067112951331118, "standard_deviation": 0.027542836844298632, "minimum": 1.3704177657309233, "maximum": 1.4541622977135447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9055417155258512, "standard_deviation": 0.5704119891282234, "minimum": 0.43823475604850826, "maximum": 1.8825104818619598, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19205564140193893, "standard_deviation": 0.012426477599002058, "minimum": 0.1708610358183916, "maximum": 0.20613015615477176, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921763942162665, "standard_deviation": 0.0028706972872440386, "minimum": 0.9861332593715487, "maximum": 0.9947640452771237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8537215012095372, "standard_deviation": 0.02067484411899482, "minimum": 0.8279959139061354, "maximum": 0.8911675144664264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.905807333962415, "standard_deviation": 0.6153957337452145, "minimum": 0.4263399119540238, "maximum": 2.068624670116923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1177136826820888, "standard_deviation": 0.02040398929545028, "minimum": 0.09662946942974505, "maximum": 0.15379239330596053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9788450089754847, "standard_deviation": 0.0030525673774653445, "minimum": 0.9757627870407888, "maximum": 0.9832606528437238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8455472014808694, "standard_deviation": 0.061540128907509546, "minimum": 1.7807692811154, "maximum": 1.9567632566758073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5698536500107547, "standard_deviation": 0.13970143639663227, "minimum": 0.4004037697938451, "maximum": 0.777315085704042, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17473642786351248, "standard_deviation": 0.011903624517100832, "minimum": 0.15693730444126663, "maximum": 0.18654972315235374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9933717371440781, "standard_deviation": 0.0012058156768101956, "minimum": 0.99125429755761, "maximum": 0.9946865353366883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0332748788471944, "standard_deviation": 0.03718876853524451, "minimum": 0.9869242473408026, "maximum": 1.0968054529472495, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5909502037117726, "standard_deviation": 0.17924438749084182, "minimum": 0.32236241854957176, "maximum": 0.8742379797042318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09834286697725231, "standard_deviation": 0.007530478613691819, "minimum": 0.09092886004945155, "maximum": 0.11070691528984153, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24087222460071722, "standard_deviation": 0.061886897888406386, "minimum": 0.166438320702659, "maximum": 0.33741754248409017, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.893590064755201, "standard_deviation": 0.006235651658956721, "minimum": 5.882682164557434, "maximum": 5.900488085273774, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.38465817733348706, "standard_deviation": 0.02644529163895443, "minimum": 0.3347145990698354, "maximum": 0.415959788757825, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10284891862631672, "standard_deviation": 0.12976093929959492, "minimum": -0.32868123608500494, "maximum": 0.0728764634369162, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9794846873375791, "standard_deviation": 0.0017211276734675767, "minimum": 0.9764420666297781, "maximum": 0.9815122502823587, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.66875401954864, "standard_deviation": 0.07974919584760924, "minimum": 1.5446203179432563, "maximum": 1.7549676488490773, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6239713100659003, "standard_deviation": 0.37750658894110445, "minimum": 0.18399304764419866, "maximum": 1.2965438675947194, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15268239935110817, "standard_deviation": 0.005897260542290467, "minimum": 0.14504339503314825, "maximum": 0.1628564832180869, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9927485400460107, "standard_deviation": 0.0020794572558231962, "minimum": 0.9882733514916483, "maximum": 0.9941701300024169, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9585261813968385, "standard_deviation": 0.08915399438929536, "minimum": 0.8352447789796592, "maximum": 1.1017665749330114, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7182655548553103, "standard_deviation": 0.4806638803319824, "minimum": 0.34915547324639634, "maximum": 1.6197807337968477, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08783322894584739, "standard_deviation": 0.011957338839825643, "minimum": 0.07818936451327446, "maximum": 0.11093944331209445, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0003", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.5229334831237793, "predicted_mos": 2.5229334831237793, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "6a3f847620d14a007469bbce0d855456dd249a2fbd85288a80be08c05b14488a", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.694130700081587, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5861523233698906, "全帯域_10ms_rms_envelope_nrmse": 0.14261759862200782, "全帯域_10ms_envelope_correlation": 0.982180133878601, "全帯域_10ms_modulation_mae_db": 0.5352180133195482, "全帯域_40ms_log_rms_mae_db": 0.8604684605397938, "全帯域_40ms_rms_envelope_nrmse": 0.0774447133504398, "全帯域_40ms_envelope_correlation": 0.9943756220449913, "全帯域_40ms_modulation_mae_db": 0.48139413872444714, "300–1000 Hz_10ms_log_rms_mae_db": 2.1500811891413836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16208941451799566, "300–1000 Hz_10ms_envelope_correlation": 0.981871188100034, "300–1000 Hz_10ms_modulation_mae_db": 0.7209477834925578, "300–1000 Hz_40ms_log_rms_mae_db": 1.3340709056961166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07686684483834205, "300–1000 Hz_40ms_envelope_correlation": 0.9960962250474665, "300–1000 Hz_40ms_modulation_mae_db": 0.7252394274144601, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8236457480187813, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19225220916839966, "1000–3000 Hz_10ms_envelope_correlation": 0.97321332358037, "1000–3000 Hz_10ms_modulation_mae_db": 1.7254215752890145, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1836706153323215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11806775568783034, "1000–3000 Hz_40ms_envelope_correlation": 0.9905963291335175, "1000–3000 Hz_40ms_modulation_mae_db": 1.914555677429724, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7068763572558685, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18066259628450915, "3000–6000 Hz_10ms_envelope_correlation": 0.98197883860522, "3000–6000 Hz_10ms_modulation_mae_db": 0.419475592088349, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1371320524968458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10285784172538802, "3000–6000 Hz_40ms_envelope_correlation": 0.994136468172486, "3000–6000 Hz_40ms_modulation_mae_db": 0.5370903212952073, "mr_stft_db": 6.068907757104405, "band_mae_db": 0.23940320602285428, "rms_gain_error_db": -0.050458848242134, "output_peak": 0.3374406162831328}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5347962934209567, "全帯域_10ms_rms_envelope_nrmse": 0.16489436171556118, "全帯域_10ms_envelope_correlation": 0.9766627787109023, "全帯域_10ms_modulation_mae_db": 1.6037248852167663, "全帯域_40ms_log_rms_mae_db": 0.8720060092645582, "全帯域_40ms_rms_envelope_nrmse": 0.11181429534999528, "全帯域_40ms_envelope_correlation": 0.9894938888600959, "全帯域_40ms_modulation_mae_db": 1.9659577491000755, "300–1000 Hz_10ms_log_rms_mae_db": 2.3124894298631027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16222318175805855, "300–1000 Hz_10ms_envelope_correlation": 0.9817229718386848, "300–1000 Hz_10ms_modulation_mae_db": 0.5302485331355121, "300–1000 Hz_40ms_log_rms_mae_db": 1.4898186468598367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07638721462388046, "300–1000 Hz_40ms_envelope_correlation": 0.9957787157299831, "300–1000 Hz_40ms_modulation_mae_db": 1.0330008362434364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7994264367409116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20950570422180043, "1000–3000 Hz_10ms_envelope_correlation": 0.967343395976025, "1000–3000 Hz_10ms_modulation_mae_db": 1.4645486781498311, "1000–3000 Hz_40ms_log_rms_mae_db": 1.134153546592381, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1154901384323482, "1000–3000 Hz_40ms_envelope_correlation": 0.9906405128114443, "1000–3000 Hz_40ms_modulation_mae_db": 1.7927608667826007, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7625049876939425, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21565425867592258, "3000–6000 Hz_10ms_envelope_correlation": 0.9805645695575192, "3000–6000 Hz_10ms_modulation_mae_db": 2.4567700245981317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3096779692358798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16464484737333157, "3000–6000 Hz_40ms_envelope_correlation": 0.9929625951895092, "3000–6000 Hz_40ms_modulation_mae_db": 2.5992564892572125, "mr_stft_db": 6.076341660630962, "band_mae_db": 0.4968729212814249, "rms_gain_error_db": -0.38634331226227697, "output_peak": 0.36164879315799625}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.580336993875596, "全帯域_10ms_rms_envelope_nrmse": 0.15429659848970895, "全帯域_10ms_envelope_correlation": 0.979334578521771, "全帯域_10ms_modulation_mae_db": 1.9121843720241494, "全帯域_40ms_log_rms_mae_db": 0.9142875111924664, "全帯域_40ms_rms_envelope_nrmse": 0.1065843000318241, "全帯域_40ms_envelope_correlation": 0.9900212485624668, "全帯域_40ms_modulation_mae_db": 2.377102490378811, "300–1000 Hz_10ms_log_rms_mae_db": 2.233160669241521, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17901883825365297, "300–1000 Hz_10ms_envelope_correlation": 0.9780207232800697, "300–1000 Hz_10ms_modulation_mae_db": 0.397146412576111, "300–1000 Hz_40ms_log_rms_mae_db": 1.3701541054779627, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0926161278589968, "300–1000 Hz_40ms_envelope_correlation": 0.9946412986900932, "300–1000 Hz_40ms_modulation_mae_db": 0.6364359207843188, "1000–3000 Hz_10ms_log_rms_mae_db": 1.821317351344574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.200896783363797, "1000–3000 Hz_10ms_envelope_correlation": 0.9698025207577207, "1000–3000 Hz_10ms_modulation_mae_db": 1.26054979657854, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1833793647830142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10844693461478014, "1000–3000 Hz_40ms_envelope_correlation": 0.991287006635355, "1000–3000 Hz_40ms_modulation_mae_db": 1.6649306194132238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6820107094641579, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21473926383610126, "3000–6000 Hz_10ms_envelope_correlation": 0.9753603670307276, "3000–6000 Hz_10ms_modulation_mae_db": 3.183735536918005, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1878671630631596, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1634199189380935, "3000–6000 Hz_40ms_envelope_correlation": 0.9869387886933606, "3000–6000 Hz_40ms_modulation_mae_db": 3.7695679961390915, "mr_stft_db": 6.069389586494519, "band_mae_db": 0.2649565665338239, "rms_gain_error_db": -0.3113802581792111, "output_peak": 0.40781615105420654}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.596880650693672, "全帯域_10ms_rms_envelope_nrmse": 0.1701785728708887, "全帯域_10ms_envelope_correlation": 0.9742336581800588, "全帯域_10ms_modulation_mae_db": 1.6200806308687359, "全帯域_40ms_log_rms_mae_db": 0.8266048925443332, "全帯域_40ms_rms_envelope_nrmse": 0.09516953266156183, "全帯域_40ms_envelope_correlation": 0.9914787668269089, "全帯域_40ms_modulation_mae_db": 1.9202441440050297, "300–1000 Hz_10ms_log_rms_mae_db": 2.2104461309916794, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18506092786809075, "300–1000 Hz_10ms_envelope_correlation": 0.9761272169231867, "300–1000 Hz_10ms_modulation_mae_db": 0.5931468629527117, "300–1000 Hz_40ms_log_rms_mae_db": 1.3371559164112978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08958219134876391, "300–1000 Hz_40ms_envelope_correlation": 0.9943765440200029, "300–1000 Hz_40ms_modulation_mae_db": 0.5194575958414941, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8184191042730171, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19345911311723052, "1000–3000 Hz_10ms_envelope_correlation": 0.972049425271687, "1000–3000 Hz_10ms_modulation_mae_db": 1.0214002318311861, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1552442831073768, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10473591470629004, "1000–3000 Hz_40ms_envelope_correlation": 0.9917785303082612, "1000–3000 Hz_40ms_modulation_mae_db": 1.3447738385633958, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7226241106735614, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2244906765839888, "3000–6000 Hz_10ms_envelope_correlation": 0.9710955377172454, "3000–6000 Hz_10ms_modulation_mae_db": 2.5370055103100673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2488733272404546, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13848014293841004, "3000–6000 Hz_40ms_envelope_correlation": 0.9891473056305771, "3000–6000 Hz_40ms_modulation_mae_db": 2.875263337603732, "mr_stft_db": 6.065365575820401, "band_mae_db": 0.1704770117452681, "rms_gain_error_db": -0.15123210026986478, "output_peak": 0.4571607050056804}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5518411928564881, "全帯域_10ms_rms_envelope_nrmse": 0.1674579388752842, "全帯域_10ms_envelope_correlation": 0.9764655335104784, "全帯域_10ms_modulation_mae_db": 1.1401255239697297, "全帯域_40ms_log_rms_mae_db": 0.8637071070809135, "全帯域_40ms_rms_envelope_nrmse": 0.11547960257501275, "全帯域_40ms_envelope_correlation": 0.9880694799871086, "全帯域_40ms_modulation_mae_db": 1.2406573875996363, "300–1000 Hz_10ms_log_rms_mae_db": 2.205027748963336, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15742052302423643, "300–1000 Hz_10ms_envelope_correlation": 0.9828977481564335, "300–1000 Hz_10ms_modulation_mae_db": 0.42727225001654684, "300–1000 Hz_40ms_log_rms_mae_db": 1.3552670431989302, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08958628417279427, "300–1000 Hz_40ms_envelope_correlation": 0.9944497604685999, "300–1000 Hz_40ms_modulation_mae_db": 0.5452029376292277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9051358824320348, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2211359469809433, "1000–3000 Hz_10ms_envelope_correlation": 0.9635616507301048, "1000–3000 Hz_10ms_modulation_mae_db": 1.0387964119168624, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2163988724282784, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1263818808866267, "1000–3000 Hz_40ms_envelope_correlation": 0.9887020464355777, "1000–3000 Hz_40ms_modulation_mae_db": 1.3967736362908345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6979956211765759, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22525157581041605, "3000–6000 Hz_10ms_envelope_correlation": 0.9748899526469573, "3000–6000 Hz_10ms_modulation_mae_db": 1.6129639054385863, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1781330943731867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1717337344568351, "3000–6000 Hz_40ms_envelope_correlation": 0.9858545877626581, "3000–6000 Hz_40ms_modulation_mae_db": 1.6000016500839827, "mr_stft_db": 6.064925347406619, "band_mae_db": 0.23576130178441423, "rms_gain_error_db": 0.14967745011248615, "output_peak": 0.38635092588036124}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.58640762012291, "全帯域_10ms_rms_envelope_nrmse": 0.14671245147619746, "全帯域_10ms_envelope_correlation": 0.9812418193216608, "全帯域_10ms_modulation_mae_db": 0.5958163928225415, "全帯域_40ms_log_rms_mae_db": 0.8823929349360085, "全帯域_40ms_rms_envelope_nrmse": 0.08930526280146006, "全帯域_40ms_envelope_correlation": 0.9925097109802293, "全帯域_40ms_modulation_mae_db": 0.6318504176765344, "300–1000 Hz_10ms_log_rms_mae_db": 2.2088946354141457, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1547176576161113, "300–1000 Hz_10ms_envelope_correlation": 0.983393682998014, "300–1000 Hz_10ms_modulation_mae_db": 0.24944843825160898, "300–1000 Hz_40ms_log_rms_mae_db": 1.419735225580663, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08665702753905104, "300–1000 Hz_40ms_envelope_correlation": 0.9945299775338281, "300–1000 Hz_40ms_modulation_mae_db": 0.3675560690909321, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7373000505225757, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23011494730385654, "1000–3000 Hz_10ms_envelope_correlation": 0.9604393550624837, "1000–3000 Hz_10ms_modulation_mae_db": 2.0177442110794526, "1000–3000 Hz_40ms_log_rms_mae_db": 1.135832888071148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.139815874014753, "1000–3000 Hz_40ms_envelope_correlation": 0.9858585532743099, "1000–3000 Hz_40ms_modulation_mae_db": 2.598746506259188, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6561013799190252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17934266334044896, "3000–6000 Hz_10ms_envelope_correlation": 0.9826553539059087, "3000–6000 Hz_10ms_modulation_mae_db": 0.7740933920543237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1877703693845376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11503676976325963, "3000–6000 Hz_40ms_envelope_correlation": 0.992915913166637, "3000–6000 Hz_40ms_modulation_mae_db": 0.7272699244370694, "mr_stft_db": 6.053246820096618, "band_mae_db": 0.2380238258405268, "rms_gain_error_db": -0.0024979955923799995, "output_peak": 0.41781721503933944}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6373958838604457, "全帯域_10ms_rms_envelope_nrmse": 0.1465692255388523, "全帯域_10ms_envelope_correlation": 0.9812338779656982, "全帯域_10ms_modulation_mae_db": 0.3699253629043667, "全帯域_40ms_log_rms_mae_db": 0.8511692307660336, "全帯域_40ms_rms_envelope_nrmse": 0.08912746285408, "全帯域_40ms_envelope_correlation": 0.9929046708422608, "全帯域_40ms_modulation_mae_db": 0.5500904109131914, "300–1000 Hz_10ms_log_rms_mae_db": 2.241996296664064, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17390084390666233, "300–1000 Hz_10ms_envelope_correlation": 0.979090061052352, "300–1000 Hz_10ms_modulation_mae_db": 0.2915140676242199, "300–1000 Hz_40ms_log_rms_mae_db": 1.3828009659258502, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09205768874784856, "300–1000 Hz_40ms_envelope_correlation": 0.9943343646880809, "300–1000 Hz_40ms_modulation_mae_db": 0.16789882011356483, "1000–3000 Hz_10ms_log_rms_mae_db": 1.800431841482944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18419545209513105, "1000–3000 Hz_10ms_envelope_correlation": 0.9746423165117394, "1000–3000 Hz_10ms_modulation_mae_db": 0.3229105944503084, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1400493416527215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10570794376782475, "1000–3000 Hz_40ms_envelope_correlation": 0.9911120585369853, "1000–3000 Hz_40ms_modulation_mae_db": 0.49292541042564214, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7160401050910332, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18295446943746096, "3000–6000 Hz_10ms_envelope_correlation": 0.9813223117195984, "3000–6000 Hz_10ms_modulation_mae_db": 0.7673325976852324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.208806172059725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1219911564841866, "3000–6000 Hz_40ms_envelope_correlation": 0.9921784574916468, "3000–6000 Hz_40ms_modulation_mae_db": 0.9139575566146698, "mr_stft_db": 6.072318276386651, "band_mae_db": 0.18672526587073168, "rms_gain_error_db": -0.2362381834057931, "output_peak": 0.4204683285337112}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5319707071447561, "全帯域_10ms_rms_envelope_nrmse": 0.1276372615219352, "全帯域_10ms_envelope_correlation": 0.9857461396723833, "全帯域_10ms_modulation_mae_db": 0.1854814111624716, "全帯域_40ms_log_rms_mae_db": 0.8473249996886101, "全帯域_40ms_rms_envelope_nrmse": 0.06317892412630875, "全帯域_40ms_envelope_correlation": 0.9962226869349062, "全帯域_40ms_modulation_mae_db": 0.2758884919473184, "300–1000 Hz_10ms_log_rms_mae_db": 2.226885469896311, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1818776084449655, "300–1000 Hz_10ms_envelope_correlation": 0.976946431430504, "300–1000 Hz_10ms_modulation_mae_db": 0.7366505768570312, "300–1000 Hz_40ms_log_rms_mae_db": 1.4058678726710625, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09199374160545383, "300–1000 Hz_40ms_envelope_correlation": 0.993921075375175, "300–1000 Hz_40ms_modulation_mae_db": 1.0088062594911145, "1000–3000 Hz_10ms_log_rms_mae_db": 1.799983971289317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2222352925972871, "1000–3000 Hz_10ms_envelope_correlation": 0.9629805632593836, "1000–3000 Hz_10ms_modulation_mae_db": 0.727701344736543, "1000–3000 Hz_40ms_log_rms_mae_db": 1.143464795353505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1412703361595603, "1000–3000 Hz_40ms_envelope_correlation": 0.9839841808897403, "1000–3000 Hz_40ms_modulation_mae_db": 0.9739018850145276, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7604617702450365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15509414658997542, "3000–6000 Hz_10ms_envelope_correlation": 0.9867542477484831, "3000–6000 Hz_10ms_modulation_mae_db": 0.3573630434681962, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2896842424507826, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07756364708506726, "3000–6000 Hz_40ms_envelope_correlation": 0.9967301139425698, "3000–6000 Hz_40ms_modulation_mae_db": 0.34068112080245444, "mr_stft_db": 6.054225579715459, "band_mae_db": 0.22412523163344147, "rms_gain_error_db": 0.005124638869445304, "output_peak": 0.37979908618473857}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76160, "output_samples": 76160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.160186172166214, "spectral_convergence": 0.4875585876734439, "spectral_convergence_status": "ok", "amplitude_floor": 1.3356779608968942e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.824233712832519, "spectral_convergence": 0.4622370196017725, "spectral_convergence_status": "ok", "amplitude_floor": 9.962006407647762e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.222303386314482, "spectral_convergence": 0.5161545792233538, "spectral_convergence_status": "ok", "amplitude_floor": 4.291398708237255e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021802020252140593, "output_power": 0.00020664110175033608, "delta_db": -0.23280030025201925, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019115120810434436, "output_power": 0.00018555658808912544, "delta_db": -0.12900668814235736, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.836069030265609e-05, "output_power": 4.861213896230324e-05, "delta_db": 0.02252239332167729, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.9699402895957116e-05, "output_power": 4.258649993086587e-05, "delta_db": -0.6707922296330465, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001234531768639194, "output_power": 0.00011840535991561771, "delta_db": -0.18130908003080948, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00036807758285640346, "output_power": 0.00038542352828619905, "delta_db": 0.19998854475721572, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23940320602285428, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.050458848242134, "pitch": null, "output_rms_dbfs": -30.055837598792998, "output_peak": 0.3374406162831328, "reference_rms_dbfs": -30.00537875055086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.068907757104405}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.00018341452353274373, "log_rms_mae_db": 1.5861523233698906, "rms_envelope_nrmse": 0.14261759862200782, "envelope_correlation": 0.982180133878601, "rms_gain_error_db": -0.05056316506815689, "reference_mean_power": 0.0009998027766238483, "output_mean_power": 0.0009882299732811436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 5.754282242130778e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018894414816227902, "output_power": 0.00019925008676611477, "reference_active": true, "delta_db": 0.23065073372107026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013084274193993133, "output_power": 0.00015088530916983499, "reference_active": true, "delta_db": 0.618973204696868}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.690431882761885e-05, "output_power": 3.10080305652816e-05, "reference_active": true, "delta_db": -0.7560301015407063}], "active_band_count": 3, "active_band_mae_db": 0.5352180133195482}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.0001230241515816465, "log_rms_mae_db": 0.8604684605397938, "rms_envelope_nrmse": 0.0774447133504398, "envelope_correlation": 0.9943756220449913, "rms_gain_error_db": -0.050698680104415625, "reference_mean_power": 0.001006100998286567, "output_mean_power": 0.0009944242624931965}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 5.061730772699165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019045538591950462, "output_power": 0.00020718244463987953, "reference_active": true, "delta_db": 0.365596945449764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.588390935994399e-05, "output_power": 8.422783580380287e-05, "reference_active": true, "delta_db": 0.45305945214492793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5269059669432984e-06, "output_power": 2.1879453953076664e-06, "reference_active": true, "delta_db": -0.6255260185786495}], "active_band_count": 3, "active_band_mae_db": 0.48139413872444714}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 5.489840871101384e-05, "log_rms_mae_db": 2.1500811891413836, "rms_envelope_nrmse": 0.16208941451799566, "envelope_correlation": 0.981871188100034, "rms_gain_error_db": -0.1993724887731298, "reference_mean_power": 0.0001499835837122031, "output_mean_power": 0.00014325390770319947}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.245105068973029e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.560041827470393e-05, "output_power": 6.0330175258774615e-05, "reference_active": true, "delta_db": -0.36372021547781047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1382200906061474e-05, "output_power": 1.717208417993875e-05, "reference_active": true, "delta_db": -0.952293970538643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4492416985862286e-06, "output_power": 2.015341872596671e-06, "reference_active": true, "delta_db": -0.8468291644612203}], "active_band_count": 3, "active_band_mae_db": 0.7209477834925578}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 4.550505877934166e-05, "log_rms_mae_db": 1.3340709056961166, "rms_envelope_nrmse": 0.07686684483834205, "envelope_correlation": 0.9960962250474665, "rms_gain_error_db": -0.19937909165327689, "reference_mean_power": 0.00015093488949896852, "output_mean_power": 0.00014416230977349552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 1.1354747567630587e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.429099822482352e-05, "output_power": 5.8392234675836245e-05, "reference_active": true, "delta_db": -0.4179507290253366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916365304226932e-05, "output_power": 8.898803108178335e-06, "reference_active": true, "delta_db": -0.8874646260274778}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3162457402511593e-07, "output_power": 1.8956333510615436e-07, "reference_active": true, "delta_db": -0.8703029271905665}], "active_band_count": 3, "active_band_mae_db": 0.7252394274144601}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 4.066436664003698e-05, "log_rms_mae_db": 1.8236457480187813, "rms_envelope_nrmse": 0.19225220916839966, "envelope_correlation": 0.97321332358037, "rms_gain_error_db": -0.43574557192541696, "reference_mean_power": 0.0001020893317881212, "output_mean_power": 9.234338791828984e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 9.602503308503003e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.081225810114696e-05, "output_power": 2.6152635531768507e-05, "reference_active": true, "delta_db": -0.7120806567024875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3389822486480084e-05, "output_power": 1.0992402058373166e-05, "reference_active": true, "delta_db": -0.8568221468578918}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310273392307641e-06, "output_power": 2.3140923599085016e-06, "reference_active": true, "delta_db": -3.607361922306664}], "active_band_count": 3, "active_band_mae_db": 1.7254215752890145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 3.8037796302268945e-05, "log_rms_mae_db": 1.1836706153323215, "rms_envelope_nrmse": 0.11806775568783034, "envelope_correlation": 0.9905963291335175, "rms_gain_error_db": -0.43577742355342447, "reference_mean_power": 0.00010273559922291735, "output_mean_power": 9.292727798101963e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 8.714815910547825e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9375203264151156e-05, "output_power": 2.4674424639297315e-05, "reference_active": true, "delta_db": -0.7573384588925767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.005890574709935e-06, "output_power": 5.481274566743656e-06, "reference_active": true, "delta_db": -1.0658179257517406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.15357493378345e-07, "output_power": 2.0895729152761712e-07, "reference_active": true, "delta_db": -3.920510647644855}], "active_band_count": 3, "active_band_mae_db": 1.914555677429724}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001808188641270182, "log_rms_mae_db": 1.7068763572558685, "rms_envelope_nrmse": 0.18066259628450915, "envelope_correlation": 0.98197883860522, "rms_gain_error_db": 0.15074242461062592, "reference_mean_power": 0.0004025082939999733, "output_mean_power": 0.0004167245403556188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 3.5966286763117237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001367199427375967, "output_power": 0.0001612533262236795, "reference_active": true, "delta_db": 0.7167681391300054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.299777768885521e-05, "output_power": 0.00010509944477495115, "reference_active": true, "delta_db": 0.5312785111214022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495937881252574e-05, "output_power": 2.5019105880648306e-05, "reference_active": true, "delta_db": 0.010380126013639286}], "active_band_count": 3, "active_band_mae_db": 0.419475592088349}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.00011978792663441872, "log_rms_mae_db": 1.1371320524968458, "rms_envelope_nrmse": 0.10285784172538802, "envelope_correlation": 0.994136468172486, "rms_gain_error_db": 0.15074389655106163, "reference_mean_power": 0.0004050621320160614, "output_mean_power": 0.0004193687198666765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 3.450538172499259e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016854041012359937, "output_power": 0.00019706332686206599, "reference_active": true, "delta_db": 0.6790176396904033}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.211674623356507e-05, "output_power": 5.84712674853692e-05, "reference_active": true, "delta_db": 0.49965214268952873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0452544406168747e-06, "output_power": 1.851344439892103e-06, "reference_active": true, "delta_db": -0.4326011815056899}], "active_band_count": 3, "active_band_mae_db": 0.5370903212952073}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7348918440312608, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ffc6efb929c246e17a45848d2fbb5faf7ab4f63155e17e4a97626c6d783ac725", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9680040688936893, "standard_deviation": 0.005256751783750012, "minimum": 0.9604393550624837, "maximum": 0.9746423165117394, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8132075482630192, "standard_deviation": 0.04620243156713169, "minimum": 1.7373000505225757, "maximum": 1.9051358824320348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1973841055039673, "standard_deviation": 0.5436132137794784, "minimum": 0.3229105944503084, "maximum": 2.0177442110794526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2067244311060557, "standard_deviation": 0.016604907126665076, "minimum": 0.18419545209513105, "maximum": 0.23011494730385654, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9892449022531489, "standard_deviation": 0.0028607456213305643, "minimum": 0.9839841808897403, "maximum": 0.9917785303082612, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1615242134150934, "standard_deviation": 0.02979761923341996, "minimum": 1.134153546592381, "maximum": 1.2163988724282784, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.522421055022392, "standard_deviation": 0.6340030417114247, "minimum": 0.49292541042564214, "maximum": 2.598746506259188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11998959728375168, "standard_deviation": 0.014543651102123481, "minimum": 0.10473591470629004, "maximum": 0.1412703361595603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9793276473664575, "standard_deviation": 0.005098537302604277, "minimum": 0.9710955377172454, "maximum": 0.9867542477484831, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7130768801899001, "standard_deviation": 0.036382101921237586, "minimum": 1.6561013799190252, "maximum": 1.7625049876939425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5135924503201115, "standard_deviation": 1.093744003043071, "minimum": 0.3573630434681962, "maximum": 3.183735536918005, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19727370631985292, "standard_deviation": 0.02604397966817263, "minimum": 0.15509414658997542, "maximum": 0.22525157581041605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9913580287561806, "standard_deviation": 0.003721724018041797, "minimum": 0.9858545877626581, "maximum": 0.9967301139425698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2184930487880714, "standard_deviation": 0.0591970537585268, "minimum": 1.1371320524968458, "maximum": 1.3096779692358798, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6703860495291774, "standard_deviation": 1.2671776782165423, "minimum": 0.34068112080245444, "maximum": 3.7695679961390915, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13196600734557146, "standard_deviation": 0.033539230898113705, "minimum": 0.07756364708506726, "maximum": 0.1717337344568351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9800087529724099, "standard_deviation": 0.0028145450687124536, "minimum": 0.9761272169231867, "maximum": 0.983393682998014, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.223622696271943, "standard_deviation": 0.04548338336498567, "minimum": 2.1500811891413836, "maximum": 2.3124894298631027, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.49329686561328745, "standard_deviation": 0.18367826275612087, "minimum": 0.24944843825160898, "maximum": 0.7366505768570312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1695386244237217, "standard_deviation": 0.011817733117340566, "minimum": 0.1547176576161113, "maximum": 0.18506092786809075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947659951941538, "standard_deviation": 0.00075765040123255, "minimum": 0.993921075375175, "maximum": 0.9960962250474665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.386858835227715, "standard_deviation": 0.05153836937447893, "minimum": 1.3340709056961166, "maximum": 1.4898186468598367, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6254497333260686, "standard_deviation": 0.2968103455710392, "minimum": 0.16789882011356483, "maximum": 1.0330008362434364, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08696839009189136, "standard_deviation": 0.006663727232383734, "minimum": 0.07638721462388046, "maximum": 0.0926161278589968, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25704316633906066, "standard_deviation": 0.10157942086504769, "minimum": 0.1704770117452681, "maximum": 0.4968729212814249, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.065590075456955, "standard_deviation": 0.008179316658144102, "minimum": 6.053246820096618, "maximum": 6.076341660630962, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3960627276423958, "standard_deviation": 0.037644375088060913, "minimum": 0.3374406162831328, "maximum": 0.4571607050056804, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12291857612121605, "standard_deviation": 0.18087429457369628, "minimum": -0.38634331226227697, "maximum": 0.14967745011248615, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9796373149701942, "standard_deviation": 0.00372455816179842, "minimum": 0.9742336581800588, "maximum": 0.9857461396723833, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5757227081680893, "standard_deviation": 0.03516290816106579, "minimum": 1.5319707071447561, "maximum": 1.6373958838604457, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9953195740360387, "standard_deviation": 0.659040994722147, "minimum": 0.1854814111624716, "maximum": 1.9121843720241494, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15254550113880447, "standard_deviation": 0.014528743456812728, "minimum": 0.1276372615219352, "maximum": 0.1701785728708887, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9918845093798709, "standard_deviation": 0.0026822942442708284, "minimum": 0.9880694799871086, "maximum": 0.9962226869349062, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8647451432515897, "standard_deviation": 0.02613269437608162, "minimum": 0.8266048925443332, "maximum": 0.9142875111924664, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1803981537931305, "standard_deviation": 0.8110967198324721, "minimum": 0.2758884919473184, "maximum": 2.377102490378811, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09351301171883532, "standard_deviation": 0.017753695826541564, "minimum": 0.06317892412630875, "maximum": 0.11547960257501275, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0003", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6156392097473145, "predicted_mos": 2.6156392097473145, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "a541b12a09203e460416b36eab723cfab2b1c9bec477580dd1976ecc1f9553f0", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7905751997604966, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5429745014361789, "全帯域_10ms_rms_envelope_nrmse": 0.12549571164452525, "全帯域_10ms_envelope_correlation": 0.9860795739162044, "全帯域_10ms_modulation_mae_db": 0.5365075791330364, "全帯域_40ms_log_rms_mae_db": 0.9198327750474069, "全帯域_40ms_rms_envelope_nrmse": 0.06758467784761805, "全帯域_40ms_envelope_correlation": 0.9957828293979979, "全帯域_40ms_modulation_mae_db": 0.6975884387330528, "300–1000 Hz_10ms_log_rms_mae_db": 2.0938585297050816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16552831025821288, "300–1000 Hz_10ms_envelope_correlation": 0.9811890048003548, "300–1000 Hz_10ms_modulation_mae_db": 0.4468734974067318, "300–1000 Hz_40ms_log_rms_mae_db": 1.1752145698300172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09456800279539164, "300–1000 Hz_40ms_envelope_correlation": 0.9940637194785973, "300–1000 Hz_40ms_modulation_mae_db": 0.5690856126338516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.689542057828744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1973916074611252, "1000–3000 Hz_10ms_envelope_correlation": 0.9718565680522324, "1000–3000 Hz_10ms_modulation_mae_db": 1.7870446915723137, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0607206679038848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1236226234683364, "1000–3000 Hz_40ms_envelope_correlation": 0.9899293104316261, "1000–3000 Hz_40ms_modulation_mae_db": 2.136819729024006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6162122849641454, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14689940025404963, "3000–6000 Hz_10ms_envelope_correlation": 0.987753038863059, "3000–6000 Hz_10ms_modulation_mae_db": 0.5943475920055201, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0872346960581438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08016582873748246, "3000–6000 Hz_40ms_envelope_correlation": 0.9964304939608664, "3000–6000 Hz_40ms_modulation_mae_db": 0.7459568063651781, "mr_stft_db": 6.011247648097007, "band_mae_db": 0.15285180831900633, "rms_gain_error_db": -0.12380193486635702, "output_peak": 0.36661366653459904}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6014162157307117, "全帯域_10ms_rms_envelope_nrmse": 0.1621186668559784, "全帯域_10ms_envelope_correlation": 0.9767302671008128, "全帯域_10ms_modulation_mae_db": 0.4961003557311827, "全帯域_40ms_log_rms_mae_db": 0.8661845885794738, "全帯域_40ms_rms_envelope_nrmse": 0.08401926296395175, "全帯域_40ms_envelope_correlation": 0.9936016022503051, "全帯域_40ms_modulation_mae_db": 0.4908373206782029, "300–1000 Hz_10ms_log_rms_mae_db": 2.110470382666231, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2026484125168877, "300–1000 Hz_10ms_envelope_correlation": 0.9715134193073056, "300–1000 Hz_10ms_modulation_mae_db": 0.627815268548367, "300–1000 Hz_40ms_log_rms_mae_db": 1.2587774671372722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10923201108862807, "300–1000 Hz_40ms_envelope_correlation": 0.9912498492389267, "300–1000 Hz_40ms_modulation_mae_db": 0.5604790929178893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.733098017097568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20333425449779002, "1000–3000 Hz_10ms_envelope_correlation": 0.9688905208132379, "1000–3000 Hz_10ms_modulation_mae_db": 0.5773373739377353, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0495310552589108, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11374353659236261, "1000–3000 Hz_40ms_envelope_correlation": 0.9899802557596006, "1000–3000 Hz_40ms_modulation_mae_db": 0.8263261001818988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6224687195743073, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22231595220006245, "3000–6000 Hz_10ms_envelope_correlation": 0.9717736651330737, "3000–6000 Hz_10ms_modulation_mae_db": 0.7673283916107478, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0322372851267518, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12047429504664339, "3000–6000 Hz_40ms_envelope_correlation": 0.9923164193186157, "3000–6000 Hz_40ms_modulation_mae_db": 0.3622688650858838, "mr_stft_db": 5.99442963076838, "band_mae_db": 0.22719524163440263, "rms_gain_error_db": -0.20010305278841137, "output_peak": 0.3972693970848761}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5403817635785462, "全帯域_10ms_rms_envelope_nrmse": 0.15541885293010188, "全帯域_10ms_envelope_correlation": 0.9785735615549996, "全帯域_10ms_modulation_mae_db": 0.2837821620148011, "全帯域_40ms_log_rms_mae_db": 0.7644518453458067, "全帯域_40ms_rms_envelope_nrmse": 0.09234909633384321, "全帯域_40ms_envelope_correlation": 0.9920171103613072, "全帯域_40ms_modulation_mae_db": 0.4334895475446184, "300–1000 Hz_10ms_log_rms_mae_db": 2.0634047140318574, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.149458104900255, "300–1000 Hz_10ms_envelope_correlation": 0.9845007804452065, "300–1000 Hz_10ms_modulation_mae_db": 0.5681634882072384, "300–1000 Hz_40ms_log_rms_mae_db": 1.1964028793376436, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07092106926932304, "300–1000 Hz_40ms_envelope_correlation": 0.9963359707243545, "300–1000 Hz_40ms_modulation_mae_db": 1.0195476660664393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6588649376721027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18833578987510147, "1000–3000 Hz_10ms_envelope_correlation": 0.9737203990336761, "1000–3000 Hz_10ms_modulation_mae_db": 1.4729263903279222, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9611729059579917, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10326812680477702, "1000–3000 Hz_40ms_envelope_correlation": 0.9923809450899038, "1000–3000 Hz_40ms_modulation_mae_db": 1.9930226767026682, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5943923823156982, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2130900205703629, "3000–6000 Hz_10ms_envelope_correlation": 0.9741235518574981, "3000–6000 Hz_10ms_modulation_mae_db": 0.6009630039325148, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0444144979113965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14022012636716868, "3000–6000 Hz_40ms_envelope_correlation": 0.9891109443252296, "3000–6000 Hz_40ms_modulation_mae_db": 0.7120213715155798, "mr_stft_db": 5.999345300793102, "band_mae_db": 0.17757162196618234, "rms_gain_error_db": -0.16001606897096063, "output_peak": 0.4266925023653742}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5462059481667327, "全帯域_10ms_rms_envelope_nrmse": 0.13876401438089067, "全帯域_10ms_envelope_correlation": 0.983170963087187, "全帯域_10ms_modulation_mae_db": 0.566056012502632, "全帯域_40ms_log_rms_mae_db": 0.8562900213655462, "全帯域_40ms_rms_envelope_nrmse": 0.07709739001102954, "全帯域_40ms_envelope_correlation": 0.9948525135749021, "全帯域_40ms_modulation_mae_db": 0.5536727153147106, "300–1000 Hz_10ms_log_rms_mae_db": 2.01123794478485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17407965932509992, "300–1000 Hz_10ms_envelope_correlation": 0.9788971340847221, "300–1000 Hz_10ms_modulation_mae_db": 0.6232396781398568, "300–1000 Hz_40ms_log_rms_mae_db": 1.1769698278088168, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09638268319465561, "300–1000 Hz_40ms_envelope_correlation": 0.9933093728415797, "300–1000 Hz_40ms_modulation_mae_db": 0.3767089487936485, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7436919657067762, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18703332290519423, "1000–3000 Hz_10ms_envelope_correlation": 0.9748713097581058, "1000–3000 Hz_10ms_modulation_mae_db": 1.6123047461366244, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0213311807179826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11301824191214267, "1000–3000 Hz_40ms_envelope_correlation": 0.9918066658824619, "1000–3000 Hz_40ms_modulation_mae_db": 1.9133524945708695, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5654182686586895, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17036890750163813, "3000–6000 Hz_10ms_envelope_correlation": 0.9838602497616049, "3000–6000 Hz_10ms_modulation_mae_db": 0.7276011258899101, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0534324016317305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10315722417201915, "3000–6000 Hz_40ms_envelope_correlation": 0.9946064632185312, "3000–6000 Hz_40ms_modulation_mae_db": 0.690696907728808, "mr_stft_db": 6.007589410481795, "band_mae_db": 0.3012909981000819, "rms_gain_error_db": -0.23278965995226167, "output_peak": 0.4817345138804685}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.527317940727005, "全帯域_10ms_rms_envelope_nrmse": 0.1587764256269968, "全帯域_10ms_envelope_correlation": 0.9780974619259144, "全帯域_10ms_modulation_mae_db": 1.265997970212813, "全帯域_40ms_log_rms_mae_db": 0.8799020193975665, "全帯域_40ms_rms_envelope_nrmse": 0.09556944972089619, "全帯域_40ms_envelope_correlation": 0.9923303750363672, "全帯域_40ms_modulation_mae_db": 1.5771562740397442, "300–1000 Hz_10ms_log_rms_mae_db": 2.0192184053141062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18393739118843758, "300–1000 Hz_10ms_envelope_correlation": 0.9764305643709452, "300–1000 Hz_10ms_modulation_mae_db": 0.45044352474967714, "300–1000 Hz_40ms_log_rms_mae_db": 1.2049641980794423, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09614002680542895, "300–1000 Hz_40ms_envelope_correlation": 0.9935140015066085, "300–1000 Hz_40ms_modulation_mae_db": 0.764030934145778, "1000–3000 Hz_10ms_log_rms_mae_db": 1.734929228615105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20135865280937038, "1000–3000 Hz_10ms_envelope_correlation": 0.9695916067648493, "1000–3000 Hz_10ms_modulation_mae_db": 0.44036025172055343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0246417107007726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12348291655312561, "1000–3000 Hz_40ms_envelope_correlation": 0.9882928158687476, "1000–3000 Hz_40ms_modulation_mae_db": 0.6546132435850643, "3000–6000 Hz_10ms_log_rms_mae_db": 1.619026182824952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22140014382081533, "3000–6000 Hz_10ms_envelope_correlation": 0.973706174236367, "3000–6000 Hz_10ms_modulation_mae_db": 2.3767418269744645, "3000–6000 Hz_40ms_log_rms_mae_db": 1.036488587446089, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14295660401754856, "3000–6000 Hz_40ms_envelope_correlation": 0.9911260075836653, "3000–6000 Hz_40ms_modulation_mae_db": 2.7249769695366908, "mr_stft_db": 5.9925028263421, "band_mae_db": 0.2511618267430187, "rms_gain_error_db": -0.3257794902268857, "output_peak": 0.3137530889762652}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6064000845814403, "全帯域_10ms_rms_envelope_nrmse": 0.15107533239324955, "全帯域_10ms_envelope_correlation": 0.9801316331374516, "全帯域_10ms_modulation_mae_db": 1.0072660870153574, "全帯域_40ms_log_rms_mae_db": 0.858824630910295, "全帯域_40ms_rms_envelope_nrmse": 0.0813103061326085, "全帯域_40ms_envelope_correlation": 0.9945782097783513, "全帯域_40ms_modulation_mae_db": 1.3911461660934237, "300–1000 Hz_10ms_log_rms_mae_db": 2.040699700552425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1677542945185347, "300–1000 Hz_10ms_envelope_correlation": 0.9807628555760467, "300–1000 Hz_10ms_modulation_mae_db": 0.6781458464694085, "300–1000 Hz_40ms_log_rms_mae_db": 1.165949222640523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07803714777615804, "300–1000 Hz_40ms_envelope_correlation": 0.9964103520011266, "300–1000 Hz_40ms_modulation_mae_db": 1.2367980544945525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.726604849957543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19948789239905507, "1000–3000 Hz_10ms_envelope_correlation": 0.9708328331423375, "1000–3000 Hz_10ms_modulation_mae_db": 1.5106845240129159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0632658296487474, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11437353071848959, "1000–3000 Hz_40ms_envelope_correlation": 0.9912125405012384, "1000–3000 Hz_40ms_modulation_mae_db": 1.7107513500486966, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6387065567680998, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1960452261489662, "3000–6000 Hz_10ms_envelope_correlation": 0.9786575079616471, "3000–6000 Hz_10ms_modulation_mae_db": 1.314369371584628, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0719571072956653, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11270798088860397, "3000–6000 Hz_40ms_envelope_correlation": 0.9939883110798486, "3000–6000 Hz_40ms_modulation_mae_db": 1.5616279696927562, "mr_stft_db": 6.010606425844248, "band_mae_db": 0.3143644828513405, "rms_gain_error_db": -0.2913916491458857, "output_peak": 0.49639083961281016}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6288973164710765, "全帯域_10ms_rms_envelope_nrmse": 0.15283389740107511, "全帯域_10ms_envelope_correlation": 0.9794750973590137, "全帯域_10ms_modulation_mae_db": 1.017626342503994, "全帯域_40ms_log_rms_mae_db": 0.9528308980333425, "全帯域_40ms_rms_envelope_nrmse": 0.08301347509416632, "全帯域_40ms_envelope_correlation": 0.9939603273377053, "全帯域_40ms_modulation_mae_db": 1.283503937278523, "300–1000 Hz_10ms_log_rms_mae_db": 2.1672953657675933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17477154140518616, "300–1000 Hz_10ms_envelope_correlation": 0.9787581701285379, "300–1000 Hz_10ms_modulation_mae_db": 0.38786016182221905, "300–1000 Hz_40ms_log_rms_mae_db": 1.2607549490543033, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08800481725778996, "300–1000 Hz_40ms_envelope_correlation": 0.9945126263306172, "300–1000 Hz_40ms_modulation_mae_db": 0.7216412245771404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6785223249867396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1896299899987104, "1000–3000 Hz_10ms_envelope_correlation": 0.974575472177558, "1000–3000 Hz_10ms_modulation_mae_db": 1.3643934411934264, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0269556732588494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11066173217001815, "1000–3000 Hz_40ms_envelope_correlation": 0.9929938332197048, "1000–3000 Hz_40ms_modulation_mae_db": 1.6870879543893835, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5808381450817661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1995532421273791, "3000–6000 Hz_10ms_envelope_correlation": 0.9772450185352862, "3000–6000 Hz_10ms_modulation_mae_db": 1.4091116714095613, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0227475910610173, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1118924315544864, "3000–6000 Hz_40ms_envelope_correlation": 0.9930697932350471, "3000–6000 Hz_40ms_modulation_mae_db": 1.647393218822027, "mr_stft_db": 6.008818410395542, "band_mae_db": 0.2389720949720849, "rms_gain_error_db": -0.23952635096641783, "output_peak": 0.38403675152024097}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5254793449947144, "全帯域_10ms_rms_envelope_nrmse": 0.15875441766906562, "全帯域_10ms_envelope_correlation": 0.9777158467204018, "全帯域_10ms_modulation_mae_db": 0.1700849136116236, "全帯域_40ms_log_rms_mae_db": 0.8537785694069873, "全帯域_40ms_rms_envelope_nrmse": 0.08810336347318991, "全帯域_40ms_envelope_correlation": 0.992905339596756, "全帯域_40ms_modulation_mae_db": 0.1341849134443394, "300–1000 Hz_10ms_log_rms_mae_db": 2.1136578941315185, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15819445714107822, "300–1000 Hz_10ms_envelope_correlation": 0.9827132995026883, "300–1000 Hz_10ms_modulation_mae_db": 0.7554266480727975, "300–1000 Hz_40ms_log_rms_mae_db": 1.2304667640348956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08633565583303576, "300–1000 Hz_40ms_envelope_correlation": 0.9948897302985394, "300–1000 Hz_40ms_modulation_mae_db": 1.1606978492859277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.66979880110881, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20749565307021856, "1000–3000 Hz_10ms_envelope_correlation": 0.9677247747851461, "1000–3000 Hz_10ms_modulation_mae_db": 1.630754617639101, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9983831965390954, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12395134507564369, "1000–3000 Hz_40ms_envelope_correlation": 0.9876795534234111, "1000–3000 Hz_40ms_modulation_mae_db": 2.220930115341585, "3000–6000 Hz_10ms_log_rms_mae_db": 1.615845367183164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20587982855569897, "3000–6000 Hz_10ms_envelope_correlation": 0.9764767098435582, "3000–6000 Hz_10ms_modulation_mae_db": 0.20732009180470964, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0439339648693113, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11989797975348902, "3000–6000 Hz_40ms_envelope_correlation": 0.993220808738528, "3000–6000 Hz_40ms_modulation_mae_db": 0.13340035469723466, "mr_stft_db": 6.016890829737149, "band_mae_db": 0.22161967860755669, "rms_gain_error_db": -0.19806595139591598, "output_peak": 0.39973822158255684}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76160, "output_samples": 76160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0995908577795515, "spectral_convergence": 0.47527714258491627, "spectral_convergence_status": "ok", "amplitude_floor": 1.3356779608968942e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.762716589255058, "spectral_convergence": 0.451607038528895, "spectral_convergence_status": "ok", "amplitude_floor": 9.962006407647762e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.171435497256412, "spectral_convergence": 0.48409346429847905, "spectral_convergence_status": "ok", "amplitude_floor": 4.291398708237255e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021802020252140593, "output_power": 0.00021859725609405846, "delta_db": 0.011479675094923373, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019115120810434436, "output_power": 0.00018593457208187418, "delta_db": -0.12016898557784744, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.836069030265609e-05, "output_power": 4.820173915430623e-05, "delta_db": -0.01429783185013533, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.9699402895957116e-05, "output_power": 4.4250558439601166e-05, "delta_db": -0.5043241517527406, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001234531768639194, "output_power": 0.00012453443621492978, "delta_db": 0.037871887194554664, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00036807758285640346, "output_power": 0.000349174513269504, "delta_db": -0.22896831844383653, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15285180831900633, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12380193486635702, "pitch": null, "output_rms_dbfs": -30.129180685417218, "output_peak": 0.36661366653459904, "reference_rms_dbfs": -30.00537875055086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.011247648097007}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.00018341452353274373, "log_rms_mae_db": 1.5429745014361789, "rms_envelope_nrmse": 0.12549571164452525, "envelope_correlation": 0.9860795739162044, "rms_gain_error_db": -0.12378509010566482, "reference_mean_power": 0.0009998027766238483, "output_mean_power": 0.000971708114689196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 5.754282242130778e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018894414816227902, "output_power": 0.00018129208164426193, "reference_active": true, "delta_db": -0.17954610133057253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013084274193993133, "output_power": 0.00012102346727344444, "reference_active": true, "delta_db": -0.33880045485481025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.690431882761885e-05, "output_power": 2.8705133797407392e-05, "reference_active": true, "delta_db": -1.0911761812137262}], "active_band_count": 3, "active_band_mae_db": 0.5365075791330364}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.0001230241515816465, "log_rms_mae_db": 0.9198327750474069, "rms_envelope_nrmse": 0.06758467784761805, "envelope_correlation": 0.9957828293979979, "rms_gain_error_db": -0.1236771311805483, "reference_mean_power": 0.001006100998286567, "output_mean_power": 0.000977853662673829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 5.061730772699165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019045538591950462, "output_power": 0.000182173764221155, "reference_active": true, "delta_db": -0.19307426493662005}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.588390935994399e-05, "output_power": 6.729313063747605e-05, "reference_active": true, "delta_db": -0.5217896342862465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5269059669432984e-06, "output_power": 1.8399199958126654e-06, "reference_active": true, "delta_db": -1.3779014169762918}], "active_band_count": 3, "active_band_mae_db": 0.6975884387330528}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 5.489840871101384e-05, "log_rms_mae_db": 2.0938585297050816, "rms_envelope_nrmse": 0.16552831025821288, "envelope_correlation": 0.9811890048003548, "rms_gain_error_db": -0.2861968213448852, "reference_mean_power": 0.0001499835837122031, "output_mean_power": 0.0001404184078077382}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.245105068973029e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.560041827470393e-05, "output_power": 6.398351550375268e-05, "reference_active": true, "delta_db": -0.10838510253339509}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1382200906061474e-05, "output_power": 1.9091283631704014e-05, "reference_active": true, "delta_db": -0.49217276008042643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4492416985862286e-06, "output_power": 2.0655008677809977e-06, "reference_active": true, "delta_db": -0.7400626296063739}], "active_band_count": 3, "active_band_mae_db": 0.4468734974067318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 4.550505877934166e-05, "log_rms_mae_db": 1.1752145698300172, "rms_envelope_nrmse": 0.09456800279539164, "envelope_correlation": 0.9940637194785973, "rms_gain_error_db": -0.28620468671072735, "reference_mean_power": 0.00015093488949896852, "output_mean_power": 0.00014130878832016385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 1.1354747567630587e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.429099822482352e-05, "output_power": 6.252690371206474e-05, "reference_active": true, "delta_db": -0.12083245700649661}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916365304226932e-05, "output_power": 1.021197497085981e-05, "reference_active": true, "delta_db": -0.2896831873573377}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3162457402511593e-07, "output_power": 1.7183455963875498e-07, "reference_active": true, "delta_db": -1.2967411935377204}], "active_band_count": 3, "active_band_mae_db": 0.5690856126338516}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 4.066436664003698e-05, "log_rms_mae_db": 1.689542057828744, "rms_envelope_nrmse": 0.1973916074611252, "envelope_correlation": 0.9718565680522324, "rms_gain_error_db": -0.49809487888254433, "reference_mean_power": 0.0001020893317881212, "output_mean_power": 9.102713487762213e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 9.602503308503003e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.081225810114696e-05, "output_power": 2.4062984009427595e-05, "reference_active": true, "delta_db": -1.0737404464591491}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3389822486480084e-05, "output_power": 1.088958476145049e-05, "reference_active": true, "delta_db": -0.8976349977980703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310273392307641e-06, "output_power": 2.4329938866157115e-06, "reference_active": true, "delta_db": -3.3897586304597214}], "active_band_count": 3, "active_band_mae_db": 1.7870446915723137}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 3.8037796302268945e-05, "log_rms_mae_db": 1.0607206679038848, "rms_envelope_nrmse": 0.1236226234683364, "envelope_correlation": 0.9899293104316261, "rms_gain_error_db": -0.49811603191022824, "reference_mean_power": 0.00010273559922291735, "output_mean_power": 9.160292789055001e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 8.714815910547825e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9375203264151156e-05, "output_power": 2.304979942166526e-05, "reference_active": true, "delta_db": -1.0531372989607182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.005890574709935e-06, "output_power": 5.146610739187039e-06, "reference_active": true, "delta_db": -1.339420278815101}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.15357493378345e-07, "output_power": 2.043235540530942e-07, "reference_active": true, "delta_db": -4.0179016092961986}], "active_band_count": 3, "active_band_mae_db": 2.136819729024006}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001808188641270182, "log_rms_mae_db": 1.6162122849641454, "rms_envelope_nrmse": 0.14689940025404963, "envelope_correlation": 0.987753038863059, "rms_gain_error_db": -0.14746292683341194, "reference_mean_power": 0.0004025082939999733, "output_mean_power": 0.00038907071321735143}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 3.5966286763117237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001367199427375967, "output_power": 0.0001324199295597591, "reference_active": true, "delta_db": -0.13878515293009816}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.299777768885521e-05, "output_power": 7.973341660217882e-05, "reference_active": true, "delta_db": -0.6683219645939648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495937881252574e-05, "output_power": 1.9936100146179666e-05, "reference_active": true, "delta_db": -0.9759356584924974}], "active_band_count": 3, "active_band_mae_db": 0.5943475920055201}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.00011978792663441872, "log_rms_mae_db": 1.0872346960581438, "rms_envelope_nrmse": 0.08016582873748246, "envelope_correlation": 0.9964304939608664, "rms_gain_error_db": -0.1474656075955342, "reference_mean_power": 0.0004050621320160614, "output_mean_power": 0.0003915390506729239}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 3.450538172499259e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016854041012359937, "output_power": 0.00015806792840579624, "reference_active": true, "delta_db": -0.27860284804571245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.211674623356507e-05, "output_power": 4.418921355293648e-05, "reference_active": true, "delta_db": -0.7166102152148567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0452544406168747e-06, "output_power": 1.5363197325416822e-06, "reference_active": true, "delta_db": -1.242657355834965}], "active_band_count": 3, "active_band_mae_db": 0.7459568063651781}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7431609445295443, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "670af82b7fb617a365b67a80178413419926af994e5d1a51b84d8f94aa5552fd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9715079355658929, "standard_deviation": 0.0026999960545990376, "minimum": 0.9677247747851461, "maximum": 0.9748713097581058, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7043815228716737, "standard_deviation": 0.03370802905676172, "minimum": 1.6588649376721027, "maximum": 1.7436919657067762, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.299475754567574, "standard_deviation": 0.5048059022747422, "minimum": 0.44036025172055343, "maximum": 1.7870446915723137, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19675839537707068, "standard_deviation": 0.0075958220394761475, "minimum": 0.18703332290519423, "maximum": 0.20749565307021856, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9905344900220867, "standard_deviation": 0.001904323393145444, "minimum": 0.9876795534234111, "maximum": 0.9929938332197048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0257502774982794, "standard_deviation": 0.03409498203567738, "minimum": 0.9611729059579917, "maximum": 1.0632658296487474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.6428629579805216, "standard_deviation": 0.5884441920778356, "minimum": 0.6546132435850643, "maximum": 2.220930115341585, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11576525666186196, "standard_deviation": 0.0074075935067305695, "minimum": 0.10326812680477702, "maximum": 0.12395134507564369, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9779494895240118, "standard_deviation": 0.0054115966255639135, "minimum": 0.9717736651330737, "maximum": 0.987753038863059, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6066134884213528, "standard_deviation": 0.02427097838906085, "minimum": 1.5654182686586895, "maximum": 1.6387065567680998, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9997228844015069, "standard_deviation": 0.6805544972862988, "minimum": 0.20732009180470964, "maximum": 2.3767418269744645, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19694409014737158, "standard_deviation": 0.026188043578103483, "minimum": 0.14689940025404963, "maximum": 0.22231595220006245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9929836551825415, "standard_deviation": 0.0022194319162053898, "minimum": 0.9891109443252296, "maximum": 0.9964304939608664, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0490557664250133, "standard_deviation": 0.021320501982768133, "minimum": 1.0227475910610173, "maximum": 1.0872346960581438, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0722928079305198, "standard_deviation": 0.8509753634325647, "minimum": 0.13340035469723466, "maximum": 2.7249769695366908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1164340588171802, "standard_deviation": 0.02008451271386193, "minimum": 0.08016582873748246, "maximum": 0.14295660401754856, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.979345653526976, "standard_deviation": 0.0040304328484075305, "minimum": 0.9715134193073056, "maximum": 0.9845007804452065, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.077480367119208, "standard_deviation": 0.05353982353812234, "minimum": 2.01123794478485, "maximum": 2.1672953657675933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.567246014177037, "standard_deviation": 0.1280776209227787, "minimum": 0.38786016182221905, "maximum": 0.7554266480727975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1720465214067115, "standard_deviation": 0.016267647298491016, "minimum": 0.149458104900255, "maximum": 0.2026484125168877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942857028025438, "standard_deviation": 0.0016881309533842523, "minimum": 0.9912498492389267, "maximum": 0.9964103520011266, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2086874847403641, "standard_deviation": 0.03743974858316154, "minimum": 1.165949222640523, "maximum": 1.2607549490543033, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8011236728644034, "standard_deviation": 0.30854206938831935, "minimum": 0.3767089487936485, "maximum": 1.2367980544945525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08995267675255139, "standard_deviation": 0.011914463968356031, "minimum": 0.07092106926932304, "maximum": 0.10923201108862807, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23562846914920926, "standard_deviation": 0.05510163494527327, "minimum": 0.15285180831900633, "maximum": 0.3143644828513405, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.005178810307416, "standard_deviation": 0.008724138493379476, "minimum": 5.9925028263421, "maximum": 6.016890829737149, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4082786226946489, "standard_deviation": 0.05970338181250359, "minimum": 0.3137530889762652, "maximum": 0.49639083961281016, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.221434269789137, "standard_deviation": 0.06605581576993727, "minimum": -0.3257794902268857, "maximum": -0.12380193486635702, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9799968006002482, "standard_deviation": 0.0031366925546428095, "minimum": 0.9767302671008128, "maximum": 0.9860795739162044, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5648841394608008, "standard_deviation": 0.04061858078213544, "minimum": 1.5254793449947144, "maximum": 1.6288973164710765, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.66792767784068, "standard_deviation": 0.386876999467226, "minimum": 0.1700849136116236, "maximum": 1.265997970212813, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1504046648627354, "standard_deviation": 0.012331587399976043, "minimum": 0.12549571164452525, "maximum": 0.1621186668559784, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9937535384167115, "standard_deviation": 0.0013000942291777607, "minimum": 0.9920171103613072, "maximum": 0.9957828293979979, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8690119185108032, "standard_deviation": 0.05498760925361344, "minimum": 0.7644518453458067, "maximum": 0.9528308980333425, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8201974141408268, "standard_deviation": 0.5248570825360096, "minimum": 0.1341849134443394, "maximum": 1.5771562740397442, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08363087769716293, "standard_deviation": 0.00882062550107777, "minimum": 0.06758467784761805, "maximum": 0.09556944972089619, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_672-122797-0044", "language": "en", "style": "normal", "speaker": "librispeech_672", "source_utterance_id": "672-122797-0044", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 59840, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_672-122797-0044.wav", "source_file_sha256": "f32a0e0ccbf08855c14b02c0ab8ca3a1d3f116c352cc6482f952d9a78e319735", "decoded_pcm_sha256": "432f9a1ae86511f6d52017a356d8d76c16009ab41a0b0e212b50e3c7fe63423f", "original_source_sha256": "30dfd88ccc8b641d250263094e829de48032160f0e8f954c0487a8ecd0da85cc", "corpus_entry_id": "librispeech_672-122797-0044"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.247800350189209, "predicted_mos": 4.247800350189209, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "91ccd5d317df4e0ef2e9c356c4e56d5cd86286976b50858f9ffa08eb513b65c5", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8591661999998905, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_672-122797-0044", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "91ccd5d317df4e0ef2e9c356c4e56d5cd86286976b50858f9ffa08eb513b65c5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "8287d82eeca7b7a411e5e89ca1c7537c5badcb42acb25504ce71209865ab8804"}, "rows": [{"case": "librispeech_672-122797-0044", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.115241527557373, "predicted_mos": 3.115241527557373, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "c7a083d5b3dabaf7ee8e418e80eabf8ab685d868e97306686bd30085aef75c32", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.036896956000419, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_672-122797-0044", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c7a083d5b3dabaf7ee8e418e80eabf8ab685d868e97306686bd30085aef75c32", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1060756306678667, "全帯域_10ms_rms_envelope_nrmse": 0.0755824567714418, "全帯域_10ms_envelope_correlation": 0.9941803091532488, "全帯域_10ms_modulation_mae_db": 0.25030263427266247, "全帯域_40ms_log_rms_mae_db": 0.6065468549954548, "全帯域_40ms_rms_envelope_nrmse": 0.03999015282296533, "全帯域_40ms_envelope_correlation": 0.9983064484015449, "全帯域_40ms_modulation_mae_db": 0.43552362151282464, "300–1000 Hz_10ms_log_rms_mae_db": 1.2583293448547308, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07563530624155536, "300–1000 Hz_10ms_envelope_correlation": 0.9961021344953505, "300–1000 Hz_10ms_modulation_mae_db": 0.29882895694722095, "300–1000 Hz_40ms_log_rms_mae_db": 0.7692624739173772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04380221457623418, "300–1000 Hz_40ms_envelope_correlation": 0.9986720711711662, "300–1000 Hz_40ms_modulation_mae_db": 0.19433286302009847, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3411108684694752, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20199400784198737, "1000–3000 Hz_10ms_envelope_correlation": 0.9651340130135757, "1000–3000 Hz_10ms_modulation_mae_db": 0.5802035969762208, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8305027341028871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1322096932225741, "1000–3000 Hz_40ms_envelope_correlation": 0.9842736172728259, "1000–3000 Hz_40ms_modulation_mae_db": 0.3291475483728669, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3517945956323683, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1818558240796969, "3000–6000 Hz_10ms_envelope_correlation": 0.9790605278789906, "3000–6000 Hz_10ms_modulation_mae_db": 1.547436720271617, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9219677623898153, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1112975261789839, "3000–6000 Hz_40ms_envelope_correlation": 0.9925397915718712, "3000–6000 Hz_40ms_modulation_mae_db": 1.9176627003714126, "mr_stft_db": 5.431173637171477, "band_mae_db": 0.11057357499851893, "rms_gain_error_db": -0.04774318212445325, "output_peak": 0.3054346935108263}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0156648510458461, "全帯域_10ms_rms_envelope_nrmse": 0.07149599223088421, "全帯域_10ms_envelope_correlation": 0.994873164266457, "全帯域_10ms_modulation_mae_db": 0.8010227707262995, "全帯域_40ms_log_rms_mae_db": 0.5655118411740557, "全帯域_40ms_rms_envelope_nrmse": 0.03508819353880657, "全帯域_40ms_envelope_correlation": 0.9988218455573282, "全帯域_40ms_modulation_mae_db": 0.7422898155969392, "300–1000 Hz_10ms_log_rms_mae_db": 1.2954219247230154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07291701430408191, "300–1000 Hz_10ms_envelope_correlation": 0.99647245254701, "300–1000 Hz_10ms_modulation_mae_db": 0.6901812976677415, "300–1000 Hz_40ms_log_rms_mae_db": 0.8114443491907554, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038247380391177845, "300–1000 Hz_40ms_envelope_correlation": 0.9991141655993993, "300–1000 Hz_40ms_modulation_mae_db": 0.6694744377615253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3787731203527067, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18558423958394002, "1000–3000 Hz_10ms_envelope_correlation": 0.9706660945979332, "1000–3000 Hz_10ms_modulation_mae_db": 0.7575155720822022, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9284365490487786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13702140136747368, "1000–3000 Hz_40ms_envelope_correlation": 0.982981100108874, "1000–3000 Hz_40ms_modulation_mae_db": 0.3340323298295689, "3000–6000 Hz_10ms_log_rms_mae_db": 1.303578526954009, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1728927659250678, "3000–6000 Hz_10ms_envelope_correlation": 0.9810389357908841, "3000–6000 Hz_10ms_modulation_mae_db": 0.646588479543995, "3000–6000 Hz_40ms_log_rms_mae_db": 0.886355254142445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09244628983009216, "3000–6000 Hz_40ms_envelope_correlation": 0.9949946631938884, "3000–6000 Hz_40ms_modulation_mae_db": 0.8819412791612921, "mr_stft_db": 5.425643366107036, "band_mae_db": 0.14285634175906722, "rms_gain_error_db": -0.10681198531582495, "output_peak": 0.29264315497752036}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.006570191887003, "全帯域_10ms_rms_envelope_nrmse": 0.06816354877761789, "全帯域_10ms_envelope_correlation": 0.9952620337158574, "全帯域_10ms_modulation_mae_db": 0.7316532401926898, "全帯域_40ms_log_rms_mae_db": 0.5736241305710486, "全帯域_40ms_rms_envelope_nrmse": 0.03757824012264564, "全帯域_40ms_envelope_correlation": 0.9984808566882455, "全帯域_40ms_modulation_mae_db": 0.5573715631594326, "300–1000 Hz_10ms_log_rms_mae_db": 1.2427272978217216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07279985080270272, "300–1000 Hz_10ms_envelope_correlation": 0.9964636812787383, "300–1000 Hz_10ms_modulation_mae_db": 0.5813852069457734, "300–1000 Hz_40ms_log_rms_mae_db": 0.7847931960070333, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03708007246753789, "300–1000 Hz_40ms_envelope_correlation": 0.9991524275331228, "300–1000 Hz_40ms_modulation_mae_db": 0.32649647689015543, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3468757267989155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.183206798924857, "1000–3000 Hz_10ms_envelope_correlation": 0.9714782367377041, "1000–3000 Hz_10ms_modulation_mae_db": 0.5267368449216988, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8939389386187196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12488727698513666, "1000–3000 Hz_40ms_envelope_correlation": 0.9858955369169236, "1000–3000 Hz_40ms_modulation_mae_db": 0.48440150186802855, "3000–6000 Hz_10ms_log_rms_mae_db": 1.336113419426515, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1826062978349288, "3000–6000 Hz_10ms_envelope_correlation": 0.9820400542700274, "3000–6000 Hz_10ms_modulation_mae_db": 1.2608596187037728, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9205249665888271, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12046799471518173, "3000–6000 Hz_40ms_envelope_correlation": 0.9931616971692101, "3000–6000 Hz_40ms_modulation_mae_db": 1.321749913119439, "mr_stft_db": 5.4283992251845845, "band_mae_db": 0.230943119691242, "rms_gain_error_db": 7.831703480406806e-05, "output_peak": 0.29457459123247176}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1179872883653874, "全帯域_10ms_rms_envelope_nrmse": 0.06958255908340234, "全帯域_10ms_envelope_correlation": 0.9950732700324336, "全帯域_10ms_modulation_mae_db": 0.5007580337238488, "全帯域_40ms_log_rms_mae_db": 0.6625787761994665, "全帯域_40ms_rms_envelope_nrmse": 0.03685707551282418, "全帯域_40ms_envelope_correlation": 0.9985662912399785, "全帯域_40ms_modulation_mae_db": 0.6416183777895128, "300–1000 Hz_10ms_log_rms_mae_db": 1.3332660643703584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06829439575192754, "300–1000 Hz_10ms_envelope_correlation": 0.9969348070709397, "300–1000 Hz_10ms_modulation_mae_db": 1.0025147127834855, "300–1000 Hz_40ms_log_rms_mae_db": 0.8017605068076622, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04078995055888598, "300–1000 Hz_40ms_envelope_correlation": 0.9989840899777146, "300–1000 Hz_40ms_modulation_mae_db": 1.0552418518660283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4090381124928457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19549807549337206, "1000–3000 Hz_10ms_envelope_correlation": 0.9675398803285127, "1000–3000 Hz_10ms_modulation_mae_db": 0.6924943313033923, "1000–3000 Hz_40ms_log_rms_mae_db": 0.914591669381724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13632819268729138, "1000–3000 Hz_40ms_envelope_correlation": 0.9836377234191859, "1000–3000 Hz_40ms_modulation_mae_db": 0.2889479657326686, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3944828181953974, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16006328694365995, "3000–6000 Hz_10ms_envelope_correlation": 0.9837877180004887, "3000–6000 Hz_10ms_modulation_mae_db": 0.4801947076695905, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9570491035821569, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08646015937677454, "3000–6000 Hz_40ms_envelope_correlation": 0.9951281971630849, "3000–6000 Hz_40ms_modulation_mae_db": 0.482661685556338, "mr_stft_db": 5.438301518825074, "band_mae_db": 0.13788850425728225, "rms_gain_error_db": -0.049520555236792066, "output_peak": 0.2924205122384721}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0757808969322349, "全帯域_10ms_rms_envelope_nrmse": 0.07494785491368498, "全帯域_10ms_envelope_correlation": 0.9942758239148557, "全帯域_10ms_modulation_mae_db": 0.2940998640519206, "全帯域_40ms_log_rms_mae_db": 0.5107867365111697, "全帯域_40ms_rms_envelope_nrmse": 0.044093026456327546, "全帯域_40ms_envelope_correlation": 0.9979502243149492, "全帯域_40ms_modulation_mae_db": 0.28722631365235846, "300–1000 Hz_10ms_log_rms_mae_db": 1.2732864348484871, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07934709126624043, "300–1000 Hz_10ms_envelope_correlation": 0.995860740133095, "300–1000 Hz_10ms_modulation_mae_db": 0.7130062763922522, "300–1000 Hz_40ms_log_rms_mae_db": 0.8340403323081045, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05393562995081843, "300–1000 Hz_40ms_envelope_correlation": 0.9981664057324707, "300–1000 Hz_40ms_modulation_mae_db": 0.9546616996396677, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3283699415384234, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1793119534113434, "1000–3000 Hz_10ms_envelope_correlation": 0.9731877230827776, "1000–3000 Hz_10ms_modulation_mae_db": 0.5786672226384842, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8937674126730919, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1281340328051008, "1000–3000 Hz_40ms_envelope_correlation": 0.9861959285378926, "1000–3000 Hz_40ms_modulation_mae_db": 0.5685134572873604, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3661973952369126, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18453911587496177, "3000–6000 Hz_10ms_envelope_correlation": 0.9782459781786115, "3000–6000 Hz_10ms_modulation_mae_db": 0.14621989705855917, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8961728135171898, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10984326291294257, "3000–6000 Hz_40ms_envelope_correlation": 0.9919734528156608, "3000–6000 Hz_40ms_modulation_mae_db": 0.15948893232942654, "mr_stft_db": 5.435183436669859, "band_mae_db": 0.21059568106491142, "rms_gain_error_db": -0.05977427474854198, "output_peak": 0.2794625939595916}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0667652494925786, "全帯域_10ms_rms_envelope_nrmse": 0.07175299983282256, "全帯域_10ms_envelope_correlation": 0.9948175992803209, "全帯域_10ms_modulation_mae_db": 0.1785783447712789, "全帯域_40ms_log_rms_mae_db": 0.5399007972063952, "全帯域_40ms_rms_envelope_nrmse": 0.042391001610666244, "全帯域_40ms_envelope_correlation": 0.9981833076615289, "全帯域_40ms_modulation_mae_db": 0.1601689792051477, "300–1000 Hz_10ms_log_rms_mae_db": 1.323946274034906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07071321283485732, "300–1000 Hz_10ms_envelope_correlation": 0.9967249411334522, "300–1000 Hz_10ms_modulation_mae_db": 0.3464850175042344, "300–1000 Hz_40ms_log_rms_mae_db": 0.8563762655961545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04322548628922003, "300–1000 Hz_40ms_envelope_correlation": 0.9988648357122043, "300–1000 Hz_40ms_modulation_mae_db": 0.2535344863113538, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2933361017586762, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17474059111636311, "1000–3000 Hz_10ms_envelope_correlation": 0.9740951984307491, "1000–3000 Hz_10ms_modulation_mae_db": 0.7288848450358305, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8098561171449308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1220494051473137, "1000–3000 Hz_40ms_envelope_correlation": 0.9865805716249023, "1000–3000 Hz_40ms_modulation_mae_db": 0.2739910207853132, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3881728142695904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1857771776090767, "3000–6000 Hz_10ms_envelope_correlation": 0.977726826474866, "3000–6000 Hz_10ms_modulation_mae_db": 0.4436548374832206, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9570294563420364, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11548167855732355, "3000–6000 Hz_40ms_envelope_correlation": 0.9911346575367213, "3000–6000 Hz_40ms_modulation_mae_db": 0.7684601916358432, "mr_stft_db": 5.419190865442197, "band_mae_db": 0.2122070836989932, "rms_gain_error_db": -0.09879727042275202, "output_peak": 0.28376844671068246}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9976821767983635, "全帯域_10ms_rms_envelope_nrmse": 0.06612583442543415, "全帯域_10ms_envelope_correlation": 0.9955350634171588, "全帯域_10ms_modulation_mae_db": 0.19040570456744524, "全帯域_40ms_log_rms_mae_db": 0.5226749278687288, "全帯域_40ms_rms_envelope_nrmse": 0.03370820790040408, "全帯域_40ms_envelope_correlation": 0.9987864833040077, "全帯域_40ms_modulation_mae_db": 0.2220319227351046, "300–1000 Hz_10ms_log_rms_mae_db": 1.2803695951221377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06348092050763719, "300–1000 Hz_10ms_envelope_correlation": 0.9973088566439503, "300–1000 Hz_10ms_modulation_mae_db": 0.2838666937313525, "300–1000 Hz_40ms_log_rms_mae_db": 0.7856784894496167, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03313058026442098, "300–1000 Hz_40ms_envelope_correlation": 0.9993129210831445, "300–1000 Hz_40ms_modulation_mae_db": 0.5012439128861661, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3357336462734481, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18908823616357182, "1000–3000 Hz_10ms_envelope_correlation": 0.9696182350212482, "1000–3000 Hz_10ms_modulation_mae_db": 0.45852922741155594, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8308424902352922, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13416936140778848, "1000–3000 Hz_40ms_envelope_correlation": 0.9835662189463761, "1000–3000 Hz_40ms_modulation_mae_db": 0.4483834296124776, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3804753614862746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13956084216619372, "3000–6000 Hz_10ms_envelope_correlation": 0.9875168728333965, "3000–6000 Hz_10ms_modulation_mae_db": 0.3854075405770929, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9093162211649555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08324040332900545, "3000–6000 Hz_40ms_envelope_correlation": 0.9954695662431509, "3000–6000 Hz_40ms_modulation_mae_db": 0.5276522761171374, "mr_stft_db": 5.418931517829304, "band_mae_db": 0.11410028805221241, "rms_gain_error_db": -0.02410896819095333, "output_peak": 0.2871383378533637}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.030065391227042, "全帯域_10ms_rms_envelope_nrmse": 0.06706258236626463, "全帯域_10ms_envelope_correlation": 0.9955651614681492, "全帯域_10ms_modulation_mae_db": 0.33122037927065295, "全帯域_40ms_log_rms_mae_db": 0.5272127388455236, "全帯域_40ms_rms_envelope_nrmse": 0.034091301903420426, "全帯域_40ms_envelope_correlation": 0.9989669725407205, "全帯域_40ms_modulation_mae_db": 0.5038470024607871, "300–1000 Hz_10ms_log_rms_mae_db": 1.2016366087734722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06505708758218935, "300–1000 Hz_10ms_envelope_correlation": 0.9973320738189518, "300–1000 Hz_10ms_modulation_mae_db": 0.35121757497703404, "300–1000 Hz_40ms_log_rms_mae_db": 0.6911957305792421, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03956233101679013, "300–1000 Hz_40ms_envelope_correlation": 0.9991632705392975, "300–1000 Hz_40ms_modulation_mae_db": 0.4444383940554688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3359089685633694, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1876415553286045, "1000–3000 Hz_10ms_envelope_correlation": 0.9700421820940425, "1000–3000 Hz_10ms_modulation_mae_db": 0.9219708862959423, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8564442094486684, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.130374706722334, "1000–3000 Hz_40ms_envelope_correlation": 0.9845209473473168, "1000–3000 Hz_40ms_modulation_mae_db": 0.48437273978714973, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3226611933394972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16176998746815877, "3000–6000 Hz_10ms_envelope_correlation": 0.9843007575401533, "3000–6000 Hz_10ms_modulation_mae_db": 0.9997071321824412, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8645575453470857, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09691294004447643, "3000–6000 Hz_40ms_envelope_correlation": 0.9954403008708539, "3000–6000 Hz_40ms_modulation_mae_db": 1.1735712795067137, "mr_stft_db": 5.421974734535994, "band_mae_db": 0.21640628151743999, "rms_gain_error_db": -0.1282850971903249, "output_peak": 0.2943835490815716}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59840, "output_samples": 59840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.491529087982484, "spectral_convergence": 0.2756426835783954, "spectral_convergence_status": "ok", "amplitude_floor": 9.91771656423115e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.180590047057182, "spectral_convergence": 0.25292792480593784, "spectral_convergence_status": "ok", "amplitude_floor": 5.25712991724442e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.621401776474767, "spectral_convergence": 0.25927275230676256, "spectral_convergence_status": "ok", "amplitude_floor": 4.1776466342690114e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00025386719019706026, "output_power": 0.00025555804813765973, "delta_db": 0.02882986215391626, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004894966727833924, "output_power": 0.00048307640328928893, "delta_db": -0.05733920021957439, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005421484972995941, "output_power": 0.0005425144454083417, "delta_db": 0.0029304817941963957, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.5605939839496278e-05, "output_power": 1.5744887142909925e-05, "delta_db": 0.03849623634747547, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000107892990107672, "output_power": 9.758336191956295e-05, "delta_db": -0.4361745289813001, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.406384293716839e-05, "output_power": 7.238342490604149e-05, "delta_db": -0.09967114049465095, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11057357499851893, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04774318212445325, "pitch": null, "output_rms_dbfs": -28.336111650989903, "output_peak": 0.3054346935108263, "reference_rms_dbfs": -28.28836846886545, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.431173637171477}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.000110539633783132, "log_rms_mae_db": 1.1060756306678667, "rms_envelope_nrmse": 0.0755824567714418, "envelope_correlation": 0.9941803091532488, "rms_gain_error_db": -0.04774514139284664, "reference_mean_power": 0.00148505750597689, "output_mean_power": 0.0014688206075568434}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.0996976338580339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028388367321682654, "output_power": 0.00027998714661439146, "reference_active": true, "delta_db": -0.06002321381097304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6121837601531416e-05, "output_power": 4.453908426110008e-05, "reference_active": true, "delta_db": -0.15165318726171162}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3216272322760606e-05, "output_power": 1.1673089855275908e-05, "reference_active": true, "delta_db": -0.5392315017453029}], "active_band_count": 3, "active_band_mae_db": 0.25030263427266247}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 0.00010014294879668002, "log_rms_mae_db": 0.6065468549954548, "rms_envelope_nrmse": 0.03999015282296533, "envelope_correlation": 0.9983064484015449, "rms_gain_error_db": -0.04775567802231425, "reference_mean_power": 0.0014970488171277764, "output_mean_power": 0.0014806772191808667}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.0259449632693504e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002533237576934496, "output_power": 0.00024990064595811485, "reference_active": true, "delta_db": -0.05908542818598725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.399735365000864e-05, "output_power": 2.279447580545773e-05, "reference_active": true, "delta_db": -0.22333742397859252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.21988770898296e-07, "output_power": 4.91323733091051e-07, "reference_active": true, "delta_db": -1.0241480123738942}], "active_band_count": 3, "active_band_mae_db": 0.43552362151282464}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.00010594974849403162, "log_rms_mae_db": 1.2583293448547308, "rms_envelope_nrmse": 0.07563530624155536, "envelope_correlation": 0.9961021344953505, "rms_gain_error_db": -0.04439335362437625, "reference_mean_power": 0.0008075829245111437, "output_mean_power": 0.0007993699023665923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.2363198879746262e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030947682521494476, "output_power": 0.00031032178704435727, "reference_active": true, "delta_db": 0.01184134588278261}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.465844834587486e-05, "output_power": 3.21945503735239e-05, "reference_active": true, "delta_db": -0.3202675135787026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.693327955984243e-06, "output_power": 7.622192380918046e-06, "reference_active": true, "delta_db": 0.5643780113801777}], "active_band_count": 3, "active_band_mae_db": 0.29882895694722095}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 9.562043532359368e-05, "log_rms_mae_db": 0.7692624739173772, "rms_envelope_nrmse": 0.04380221457623418, "envelope_correlation": 0.9986720711711662, "rms_gain_error_db": -0.044393596386165746, "reference_mean_power": 0.0008141211968758593, "output_mean_power": 0.0008058416362352742}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.204812078547339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002902058778585621, "output_power": 0.00029114450396786935, "reference_active": true, "delta_db": 0.01402391765103086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7747321249654287e-05, "output_power": 1.652904017028277e-05, "reference_active": true, "delta_db": -0.3088517554163108}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.319074590753967e-07, "output_power": 3.126113349185307e-07, "reference_active": true, "delta_db": -0.2601229159929538}], "active_band_count": 3, "active_band_mae_db": 0.19433286302009847}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 2.8704944305868236e-05, "log_rms_mae_db": 1.3411108684694752, "rms_envelope_nrmse": 0.20199400784198737, "envelope_correlation": 0.9651340130135757, "rms_gain_error_db": -0.206020756579055, "reference_mean_power": 4.5961113590202676e-05, "output_mean_power": 4.383171534312278e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 2.8606659711814982e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3098445296769517e-05, "output_power": 1.2057316743826151e-05, "reference_active": true, "delta_db": -0.359690807763344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2173629500624627e-06, "output_power": 3.0934392083459874e-06, "reference_active": true, "delta_db": -0.17058470567281994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2063471779341703e-06, "output_power": 1.5940643173229463e-06, "reference_active": true, "delta_db": 1.2103352774924985}], "active_band_count": 3, "active_band_mae_db": 0.5802035969762208}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 2.1311353675739154e-05, "log_rms_mae_db": 0.8305027341028871, "rms_envelope_nrmse": 0.1322096932225741, "envelope_correlation": 0.9842736172728259, "rms_gain_error_db": -0.20604922939544645, "reference_mean_power": 4.633078215253765e-05, "output_mean_power": 4.418396732545141e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 2.5755466765658237e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2064714839118615e-05, "output_power": 1.1266732678644957e-05, "reference_active": true, "delta_db": -0.29719071303406525}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.637842198880982e-06, "output_power": 1.5567656057096433e-06, "reference_active": true, "delta_db": -0.22048828510358198}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.209021145285469e-07, "output_power": 1.085069439300466e-07, "reference_active": true, "delta_db": -0.4697636469809535}], "active_band_count": 3, "active_band_mae_db": 0.3291475483728669}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 6.327125594838897e-05, "log_rms_mae_db": 1.3517945956323683, "rms_envelope_nrmse": 0.1818558240796969, "envelope_correlation": 0.9790605278789906, "rms_gain_error_db": -0.37205094937575867, "reference_mean_power": 0.0001246130887459015, "output_mean_power": 0.0001143822317445475}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.481947168400202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.119164755141333e-05, "output_power": 5.678536619964829e-05, "reference_active": true, "delta_db": -0.32455715669684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1929332977586818e-05, "output_power": 1.628097803079422e-05, "reference_active": true, "delta_db": -1.2934493171312318}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.860956874662985e-06, "output_power": 3.4194347494592507e-06, "reference_active": true, "delta_db": -3.0243036869867796}], "active_band_count": 3, "active_band_mae_db": 1.547436720271617}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 5.46294552803286e-05, "log_rms_mae_db": 0.9219677623898153, "rms_envelope_nrmse": 0.1112975261789839, "envelope_correlation": 0.9925397915718712, "rms_gain_error_db": -0.372061673882288, "reference_mean_power": 0.00012561601350824365, "output_mean_power": 0.00011530253066921955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.3825123163038373e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.737559737198788e-05, "output_power": 5.2039817963360863e-05, "reference_active": true, "delta_db": -0.42391451552144493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1774546568382367e-05, "output_power": 8.218402436934608e-06, "reference_active": true, "delta_db": -1.5615678778621374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.943011691356486e-07, "output_power": 2.4960657082760987e-07, "reference_active": true, "delta_db": -3.7675057077306557}], "active_band_count": 3, "active_band_mae_db": 1.9176627003714126}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8240644187889328, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0dab873bcd2f4db4b435b5c09eeaaa887c1f477c4bc8f301b02446d62bea815e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9702201954133178, "standard_deviation": 0.0029044485172024176, "minimum": 0.9651340130135757, "maximum": 0.9740951984307491, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3461433107809824, "standard_deviation": 0.034566538978464796, "minimum": 1.2933361017586762, "maximum": 1.4090381124928457, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.655625315833166, "standard_deviation": 0.14886137363949842, "minimum": 0.45852922741155594, "maximum": 0.9219708862959423, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1871331822330049, "standard_deviation": 0.008680519242133353, "minimum": 0.17474059111636311, "maximum": 0.20199400784198737, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.984706455521787, "standard_deviation": 0.0013514910753040317, "minimum": 0.982981100108874, "maximum": 0.9865805716249023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8697975150817616, "standard_deviation": 0.043810097687768175, "minimum": 0.8098561171449308, "maximum": 0.9284365490487786, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.40147374915942924, "standard_deviation": 0.1085918698344692, "minimum": 0.2739910207853132, "maximum": 0.5685134572873604, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1306467587931266, "standard_deviation": 0.005361771787170462, "minimum": 0.1220494051473137, "maximum": 0.13702140136747368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9817147088709273, "standard_deviation": 0.0033846843165994856, "minimum": 0.977726826474866, "maximum": 0.9875168728333965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3554345155675707, "standard_deviation": 0.03270897138029694, "minimum": 1.303578526954009, "maximum": 1.3944828181953974, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7387586166862861, "standard_deviation": 0.4830024913035459, "minimum": 0.14621989705855917, "maximum": 1.547436720271617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17113316223771805, "standard_deviation": 0.016260793677504326, "minimum": 0.13956084216619372, "maximum": 0.1857771776090767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937302908205552, "standard_deviation": 0.0017347857887422954, "minimum": 0.9911346575367213, "maximum": 0.9954695662431509, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.914121640384314, "standard_deviation": 0.03242405664547406, "minimum": 0.8645575453470857, "maximum": 0.9570491035821569, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9041485322247003, "standard_deviation": 0.5563931677030349, "minimum": 0.15948893232942654, "maximum": 1.9176627003714126, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10201878186809754, "standard_deviation": 0.014047577831275424, "minimum": 0.08324040332900545, "maximum": 0.12046799471518173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.996649960890186, "standard_deviation": 0.0005312214379990407, "minimum": 0.995860740133095, "maximum": 0.9973320738189518, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2761229430686036, "standard_deviation": 0.042980116794765606, "minimum": 1.2016366087734722, "maximum": 1.3332660643703584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5334357171186368, "standard_deviation": 0.2576343104071028, "minimum": 0.2838666937313525, "maximum": 1.0025147127834855, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07103060991139898, "standard_deviation": 0.005309347729969205, "minimum": 0.06348092050763719, "maximum": 0.07934709126624043, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998928773418565, "standard_deviation": 0.00036668363868171263, "minimum": 0.9981664057324707, "maximum": 0.9993129210831445, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7918189179819932, "standard_deviation": 0.04950494396802536, "minimum": 0.6911957305792421, "maximum": 0.8563762655961545, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.549928015303808, "standard_deviation": 0.3187385104961757, "minimum": 0.19433286302009847, "maximum": 1.0552418518660283, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.041221705689385685, "standard_deviation": 0.006172797665055148, "minimum": 0.03313058026442098, "maximum": 0.05393562995081843, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17194635937995845, "standard_deviation": 0.0502720479628087, "minimum": 0.11057357499851893, "maximum": 0.230943119691242, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.42734978772069, "standard_deviation": 0.007233532685911022, "minimum": 5.418931517829304, "maximum": 5.438301518825074, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2912282349455625, "standard_deviation": 0.00790304736023847, "minimum": 0.2794625939595916, "maximum": 0.3054346935108263, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0643703770243548, "standard_deviation": 0.04369425058288273, "minimum": -0.1282850971903249, "maximum": 7.831703480406806e-05, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949478031560601, "standard_deviation": 0.000521483575910156, "minimum": 0.9941803091532488, "maximum": 0.9955651614681492, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0520739595520403, "standard_deviation": 0.046101842918528195, "minimum": 0.9976821767983635, "maximum": 1.1179872883653874, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4097551214470998, "standard_deviation": 0.24246240090988996, "minimum": 0.1785783447712789, "maximum": 0.8010227707262995, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07058922855019407, "standard_deviation": 0.003492092332499114, "minimum": 0.06612583442543415, "maximum": 0.0755824567714418, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998507803713538, "standard_deviation": 0.00034805664024715425, "minimum": 0.9979502243149492, "maximum": 0.9989669725407205, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5636046004214803, "standard_deviation": 0.05084311439499911, "minimum": 0.5107867365111697, "maximum": 0.6625787761994665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4437596995140134, "standard_deviation": 0.20665494492185682, "minimum": 0.1601689792051477, "maximum": 0.7422898155969392, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037974649983507504, "standard_deviation": 0.0038548968616143107, "minimum": 0.03370820790040408, "maximum": 0.044093026456327546, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0044", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.0566306114196777, "predicted_mos": 3.0566306114196777, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "e59249fbf55714f7dcbc584d86b7d7797a3cc10bbd40239992b516e9a878e90d", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6952078002505004, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1499086579177162, "全帯域_10ms_rms_envelope_nrmse": 0.07459685976570875, "全帯域_10ms_envelope_correlation": 0.9943445123115592, "全帯域_10ms_modulation_mae_db": 0.2645995406216027, "全帯域_40ms_log_rms_mae_db": 0.6159507699035921, "全帯域_40ms_rms_envelope_nrmse": 0.04049281105173389, "全帯域_40ms_envelope_correlation": 0.998272870291184, "全帯域_40ms_modulation_mae_db": 0.4610771389393067, "300–1000 Hz_10ms_log_rms_mae_db": 1.210323893996754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07195475938283058, "300–1000 Hz_10ms_envelope_correlation": 0.9964721810923098, "300–1000 Hz_10ms_modulation_mae_db": 0.22866822139406676, "300–1000 Hz_40ms_log_rms_mae_db": 0.7061371847685146, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04191672301959967, "300–1000 Hz_40ms_envelope_correlation": 0.9987818800368309, "300–1000 Hz_40ms_modulation_mae_db": 0.2787726472865057, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3475955807653581, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19926690555110782, "1000–3000 Hz_10ms_envelope_correlation": 0.9662539371857413, "1000–3000 Hz_10ms_modulation_mae_db": 0.4922946434410323, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7534890767825798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12643143018154143, "1000–3000 Hz_40ms_envelope_correlation": 0.9858932917822921, "1000–3000 Hz_40ms_modulation_mae_db": 0.590615824990148, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2582146028462755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18438613063048356, "3000–6000 Hz_10ms_envelope_correlation": 0.9785157307303636, "3000–6000 Hz_10ms_modulation_mae_db": 1.5040862248344358, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7934891311925828, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11143904543661906, "3000–6000 Hz_40ms_envelope_correlation": 0.9926300942671622, "3000–6000 Hz_40ms_modulation_mae_db": 1.8082258015809423, "mr_stft_db": 5.384503415362313, "band_mae_db": 0.11979039928918843, "rms_gain_error_db": -0.05165430007574089, "output_peak": 0.30551111349016524}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0406971140147483, "全帯域_10ms_rms_envelope_nrmse": 0.07191556019076233, "全帯域_10ms_envelope_correlation": 0.9948031808844432, "全帯域_10ms_modulation_mae_db": 0.7570757273274515, "全帯域_40ms_log_rms_mae_db": 0.6153001931420254, "全帯域_40ms_rms_envelope_nrmse": 0.03623540872955353, "全帯域_40ms_envelope_correlation": 0.9987219766086086, "全帯域_40ms_modulation_mae_db": 0.8891718273787595, "300–1000 Hz_10ms_log_rms_mae_db": 1.2058717931930576, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06892868514618264, "300–1000 Hz_10ms_envelope_correlation": 0.9968256662513512, "300–1000 Hz_10ms_modulation_mae_db": 0.6273922843105582, "300–1000 Hz_40ms_log_rms_mae_db": 0.6573784019115153, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03594681027963657, "300–1000 Hz_40ms_envelope_correlation": 0.9991898625951798, "300–1000 Hz_40ms_modulation_mae_db": 0.6672336116598844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.318254937426108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1698625621081357, "1000–3000 Hz_10ms_envelope_correlation": 0.9758303813630904, "1000–3000 Hz_10ms_modulation_mae_db": 0.5372969949124163, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8203237664317706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11857018184965748, "1000–3000 Hz_40ms_envelope_correlation": 0.9878888097526084, "1000–3000 Hz_40ms_modulation_mae_db": 0.3187458920910368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2615041176546002, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17964212357864756, "3000–6000 Hz_10ms_envelope_correlation": 0.979566654234839, "3000–6000 Hz_10ms_modulation_mae_db": 0.7272302707233583, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7771757936818714, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09497499165550007, "3000–6000 Hz_40ms_envelope_correlation": 0.9948396991867842, "3000–6000 Hz_40ms_modulation_mae_db": 1.1988571301489965, "mr_stft_db": 5.384896170392668, "band_mae_db": 0.16376103113865823, "rms_gain_error_db": -0.10309757336917387, "output_peak": 0.2927046929726001}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0282992967589628, "全帯域_10ms_rms_envelope_nrmse": 0.06961580026455029, "全帯域_10ms_envelope_correlation": 0.995056894306552, "全帯域_10ms_modulation_mae_db": 0.6837617611647634, "全帯域_40ms_log_rms_mae_db": 0.5789976887688744, "全帯域_40ms_rms_envelope_nrmse": 0.03844602723650314, "全帯域_40ms_envelope_correlation": 0.998408948374074, "全帯域_40ms_modulation_mae_db": 0.3845699977411327, "300–1000 Hz_10ms_log_rms_mae_db": 1.1920351609639597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07371978803476772, "300–1000 Hz_10ms_envelope_correlation": 0.9963606891633581, "300–1000 Hz_10ms_modulation_mae_db": 0.4241272210578779, "300–1000 Hz_40ms_log_rms_mae_db": 0.6529422387186814, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03983505999318618, "300–1000 Hz_40ms_envelope_correlation": 0.998991443048416, "300–1000 Hz_40ms_modulation_mae_db": 0.18346957360220564, "1000–3000 Hz_10ms_log_rms_mae_db": 1.308345084026744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1788561138922847, "1000–3000 Hz_10ms_envelope_correlation": 0.9731305864877385, "1000–3000 Hz_10ms_modulation_mae_db": 0.3755977562650681, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8277022504116707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11874890246879727, "1000–3000 Hz_40ms_envelope_correlation": 0.9876352691485192, "1000–3000 Hz_40ms_modulation_mae_db": 0.35845204711907747, "3000–6000 Hz_10ms_log_rms_mae_db": 1.303711945139176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18595089697924086, "3000–6000 Hz_10ms_envelope_correlation": 0.9817021652624915, "3000–6000 Hz_10ms_modulation_mae_db": 1.3039825656938235, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8225728942903766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12111646407441125, "3000–6000 Hz_40ms_envelope_correlation": 0.9934322599870352, "3000–6000 Hz_40ms_modulation_mae_db": 1.1468163687450008, "mr_stft_db": 5.389711035391024, "band_mae_db": 0.20414307220211025, "rms_gain_error_db": -0.006775265211383944, "output_peak": 0.29474865074298046}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1001170023180493, "全帯域_10ms_rms_envelope_nrmse": 0.06427241465908051, "全帯域_10ms_envelope_correlation": 0.9958362752329663, "全帯域_10ms_modulation_mae_db": 0.18030614977875745, "全帯域_40ms_log_rms_mae_db": 0.681882022491339, "全帯域_40ms_rms_envelope_nrmse": 0.03136817949181265, "全帯域_40ms_envelope_correlation": 0.999023050897763, "全帯域_40ms_modulation_mae_db": 0.38538921464728343, "300–1000 Hz_10ms_log_rms_mae_db": 1.2987528228648577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0633063790411768, "300–1000 Hz_10ms_envelope_correlation": 0.9974441272868882, "300–1000 Hz_10ms_modulation_mae_db": 0.7144811132750911, "300–1000 Hz_40ms_log_rms_mae_db": 0.7093329454318057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037307767770656065, "300–1000 Hz_40ms_envelope_correlation": 0.9992357588101555, "300–1000 Hz_40ms_modulation_mae_db": 0.6849351244747366, "1000–3000 Hz_10ms_log_rms_mae_db": 1.392780627942069, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1914047819343387, "1000–3000 Hz_10ms_envelope_correlation": 0.9693204228310801, "1000–3000 Hz_10ms_modulation_mae_db": 0.4718966513934304, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8196495904473112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13052302065233043, "1000–3000 Hz_40ms_envelope_correlation": 0.985609259348598, "1000–3000 Hz_40ms_modulation_mae_db": 0.35735385477635545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3376630251371044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16237652155776583, "3000–6000 Hz_10ms_envelope_correlation": 0.9832551328893959, "3000–6000 Hz_10ms_modulation_mae_db": 0.5074532889832555, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8428599287943702, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08526878531786883, "3000–6000 Hz_40ms_envelope_correlation": 0.9952522324411087, "3000–6000 Hz_40ms_modulation_mae_db": 0.488273346909597, "mr_stft_db": 5.392294408004587, "band_mae_db": 0.14733587204117962, "rms_gain_error_db": -0.07960651622341147, "output_peak": 0.29232853435234024}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0919116058380995, "全帯域_10ms_rms_envelope_nrmse": 0.07355568165353944, "全帯域_10ms_envelope_correlation": 0.9944796597437001, "全帯域_10ms_modulation_mae_db": 0.19281874517445943, "全帯域_40ms_log_rms_mae_db": 0.5381775569812349, "全帯域_40ms_rms_envelope_nrmse": 0.043972819452270676, "全帯域_40ms_envelope_correlation": 0.9979481451375168, "全帯域_40ms_modulation_mae_db": 0.4135500445696092, "300–1000 Hz_10ms_log_rms_mae_db": 1.2031316668567278, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07943497283324305, "300–1000 Hz_10ms_envelope_correlation": 0.9958399153137302, "300–1000 Hz_10ms_modulation_mae_db": 0.6488357484196969, "300–1000 Hz_40ms_log_rms_mae_db": 0.7271139870460795, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054605627249765226, "300–1000 Hz_40ms_envelope_correlation": 0.9981035573789224, "300–1000 Hz_40ms_modulation_mae_db": 0.891967482465986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2449950374517509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16822446859826662, "1000–3000 Hz_10ms_envelope_correlation": 0.9769822121224317, "1000–3000 Hz_10ms_modulation_mae_db": 0.7033225752398197, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7712572688390782, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11346346339054336, "1000–3000 Hz_40ms_envelope_correlation": 0.9899978602626782, "1000–3000 Hz_40ms_modulation_mae_db": 0.6401506892071438, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3135485577813861, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19309216662523482, "3000–6000 Hz_10ms_envelope_correlation": 0.9766514045193264, "3000–6000 Hz_10ms_modulation_mae_db": 0.2285977270509819, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7876560421040526, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11496639651881085, "3000–6000 Hz_40ms_envelope_correlation": 0.9914600219046094, "3000–6000 Hz_40ms_modulation_mae_db": 0.23179543564491664, "mr_stft_db": 5.391009949137389, "band_mae_db": 0.22157144017483588, "rms_gain_error_db": -0.04858452263935378, "output_peak": 0.279566173748477}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0482053745545439, "全帯域_10ms_rms_envelope_nrmse": 0.07164345895090382, "全帯域_10ms_envelope_correlation": 0.9947987620747598, "全帯域_10ms_modulation_mae_db": 0.15169528876963273, "全帯域_40ms_log_rms_mae_db": 0.5313558665435024, "全帯域_40ms_rms_envelope_nrmse": 0.04110671214710897, "全帯域_40ms_envelope_correlation": 0.9982535484272721, "全帯域_40ms_modulation_mae_db": 0.13083026561614433, "300–1000 Hz_10ms_log_rms_mae_db": 1.2681452361042476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0707006000760686, "300–1000 Hz_10ms_envelope_correlation": 0.9966517285458198, "300–1000 Hz_10ms_modulation_mae_db": 0.4553864514845462, "300–1000 Hz_40ms_log_rms_mae_db": 0.7609142083885015, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04344865022385012, "300–1000 Hz_40ms_envelope_correlation": 0.9987649972762744, "300–1000 Hz_40ms_modulation_mae_db": 0.1104071251783094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.256882638143486, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16553395188859277, "1000–3000 Hz_10ms_envelope_correlation": 0.9770053463397657, "1000–3000 Hz_10ms_modulation_mae_db": 0.5341583466128621, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7849785824754267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11339128227883147, "1000–3000 Hz_40ms_envelope_correlation": 0.9887592780193951, "1000–3000 Hz_40ms_modulation_mae_db": 0.2220795786060814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.348230950105428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.189314653065538, "3000–6000 Hz_10ms_envelope_correlation": 0.9769216504756043, "3000–6000 Hz_10ms_modulation_mae_db": 0.4861182611449357, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8492987659895437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10659062543387228, "3000–6000 Hz_40ms_envelope_correlation": 0.9924182895053646, "3000–6000 Hz_40ms_modulation_mae_db": 0.7426696449146829, "mr_stft_db": 5.373439303661549, "band_mae_db": 0.17739026524512533, "rms_gain_error_db": -0.07794881157515107, "output_peak": 0.284245443976902}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.03572810515101, "全帯域_10ms_rms_envelope_nrmse": 0.06715039048839168, "全帯域_10ms_envelope_correlation": 0.9953970260778351, "全帯域_10ms_modulation_mae_db": 0.22553276006551845, "全帯域_40ms_log_rms_mae_db": 0.5438684783229031, "全帯域_40ms_rms_envelope_nrmse": 0.033595216311333165, "全帯域_40ms_envelope_correlation": 0.9987977144425493, "全帯域_40ms_modulation_mae_db": 0.352282103712984, "300–1000 Hz_10ms_log_rms_mae_db": 1.2745981386746927, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06774580146528553, "300–1000 Hz_10ms_envelope_correlation": 0.9969372722663258, "300–1000 Hz_10ms_modulation_mae_db": 0.32805337149326264, "300–1000 Hz_40ms_log_rms_mae_db": 0.6909385255842077, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0340350049645689, "300–1000 Hz_40ms_envelope_correlation": 0.9992871026143265, "300–1000 Hz_40ms_modulation_mae_db": 0.61464940527844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3000275143232844, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1822399639102043, "1000–3000 Hz_10ms_envelope_correlation": 0.971890705757796, "1000–3000 Hz_10ms_modulation_mae_db": 0.3317370361577206, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7449081140513062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12521436589354004, "1000–3000 Hz_40ms_envelope_correlation": 0.9858548241740175, "1000–3000 Hz_40ms_modulation_mae_db": 0.6001683503797325, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3312585696820762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13829992679943148, "3000–6000 Hz_10ms_envelope_correlation": 0.987781667376504, "3000–6000 Hz_10ms_modulation_mae_db": 0.3254353363128683, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7974628052132794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08325660684399813, "3000–6000 Hz_40ms_envelope_correlation": 0.995404264459176, "3000–6000 Hz_40ms_modulation_mae_db": 0.4085883921198503, "mr_stft_db": 5.37708023449794, "band_mae_db": 0.10139293301435097, "rms_gain_error_db": -0.02824716904000542, "output_peak": 0.28696869737620595}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.09138692161726, "全帯域_10ms_rms_envelope_nrmse": 0.07206973133631196, "全帯域_10ms_envelope_correlation": 0.9947861751253122, "全帯域_10ms_modulation_mae_db": 0.18611979370607742, "全帯域_40ms_log_rms_mae_db": 0.5637868608522977, "全帯域_40ms_rms_envelope_nrmse": 0.034565310481789686, "全帯域_40ms_envelope_correlation": 0.9988596437386116, "全帯域_40ms_modulation_mae_db": 0.2511052393655835, "300–1000 Hz_10ms_log_rms_mae_db": 1.1864440270860999, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06684147968335345, "300–1000 Hz_10ms_envelope_correlation": 0.9970937427839108, "300–1000 Hz_10ms_modulation_mae_db": 0.5519457705683737, "300–1000 Hz_40ms_log_rms_mae_db": 0.6098689164485805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03552880394139986, "300–1000 Hz_40ms_envelope_correlation": 0.9992962939721229, "300–1000 Hz_40ms_modulation_mae_db": 0.4888840791116749, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3093634030359225, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18201115151034428, "1000–3000 Hz_10ms_envelope_correlation": 0.9720600681434829, "1000–3000 Hz_10ms_modulation_mae_db": 0.6388501193760021, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7954536430245484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1251402805354401, "1000–3000 Hz_40ms_envelope_correlation": 0.9861652544321344, "1000–3000 Hz_40ms_modulation_mae_db": 0.2118480952686748, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3112669267668153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1757693706206145, "3000–6000 Hz_10ms_envelope_correlation": 0.9805548397469128, "3000–6000 Hz_10ms_modulation_mae_db": 0.8219955372369888, "3000–6000 Hz_40ms_log_rms_mae_db": 0.79913953331489, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09837569542333345, "3000–6000 Hz_40ms_envelope_correlation": 0.9944867131056823, "3000–6000 Hz_40ms_modulation_mae_db": 0.8958011514992762, "mr_stft_db": 5.382241900931511, "band_mae_db": 0.17239816187619683, "rms_gain_error_db": -0.10605505829069553, "output_peak": 0.2949076784816365}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59840, "output_samples": 59840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.456066499580124, "spectral_convergence": 0.2739836316197804, "spectral_convergence_status": "ok", "amplitude_floor": 9.91771656423115e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.139075302309881, "spectral_convergence": 0.24945513189605098, "spectral_convergence_status": "ok", "amplitude_floor": 5.25712991724442e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.558368444196933, "spectral_convergence": 0.2562638791067274, "spectral_convergence_status": "ok", "amplitude_floor": 4.1776466342690114e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00025386719019706026, "output_power": 0.00025484572673582166, "delta_db": 0.016707794479834358, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004894966727833924, "output_power": 0.0004835609202886819, "delta_db": -0.0529854871626825, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005421484972995941, "output_power": 0.0005432671018392666, "delta_db": 0.008951482757134317, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.5605939839496278e-05, "output_power": 1.5444279143817314e-05, "delta_db": -0.04522285816954316, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000107892990107672, "output_power": 9.532149301968052e-05, "delta_db": -0.5380239306760043, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.406384293716839e-05, "output_power": 7.31006364588969e-05, "delta_db": -0.056850842489931865, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11979039928918843, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05165430007574089, "pitch": null, "output_rms_dbfs": -28.34002276894119, "output_peak": 0.30551111349016524, "reference_rms_dbfs": -28.28836846886545, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.384503415362313}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.000110539633783132, "log_rms_mae_db": 1.1499086579177162, "rms_envelope_nrmse": 0.07459685976570875, "envelope_correlation": 0.9943445123115592, "rms_gain_error_db": -0.05165750465200215, "reference_mean_power": 0.00148505750597689, "output_mean_power": 0.0014674980090925826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.0996976338580339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028388367321682654, "output_power": 0.00027997714900790177, "reference_active": true, "delta_db": -0.06017829174594563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6121837601531416e-05, "output_power": 4.43659374477314e-05, "reference_active": true, "delta_db": -0.16856939721978023}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3216272322760606e-05, "output_power": 1.1603897541255855e-05, "reference_active": true, "delta_db": -0.5650509328990823}], "active_band_count": 3, "active_band_mae_db": 0.2645995406216027}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 0.00010014294879668002, "log_rms_mae_db": 0.6159507699035921, "rms_envelope_nrmse": 0.04049281105173389, "envelope_correlation": 0.998272870291184, "rms_gain_error_db": -0.0516703642205075, "reference_mean_power": 0.0014970488171277764, "output_mean_power": 0.0014793431531730894}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.0259449632693504e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002533237576934496, "output_power": 0.00024994605355338576, "reference_active": true, "delta_db": -0.05829637553699864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.399735365000864e-05, "output_power": 2.277127406748671e-05, "reference_active": true, "delta_db": -0.22776021419727022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.21988770898296e-07, "output_power": 4.831311743244111e-07, "reference_active": true, "delta_db": -1.0971748270836512}], "active_band_count": 3, "active_band_mae_db": 0.4610771389393067}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.00010594974849403162, "log_rms_mae_db": 1.210323893996754, "rms_envelope_nrmse": 0.07195475938283058, "envelope_correlation": 0.9964721810923098, "rms_gain_error_db": -0.0386804959595605, "reference_mean_power": 0.0008075829245111437, "output_mean_power": 0.0008004221126914172}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.2363198879746262e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030947682521494476, "output_power": 0.0003112526918531859, "reference_active": true, "delta_db": 0.024849831193736387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.465844834587486e-05, "output_power": 3.1868329152601246e-05, "reference_active": true, "delta_db": -0.36449821371968677}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.693327955984243e-06, "output_power": 7.166510916822215e-06, "reference_active": true, "delta_db": 0.2966566192687771}], "active_band_count": 3, "active_band_mae_db": 0.22866822139406676}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 9.562043532359368e-05, "log_rms_mae_db": 0.7061371847685146, "rms_envelope_nrmse": 0.04191672301959967, "envelope_correlation": 0.9987818800368309, "rms_gain_error_db": -0.03868047740698934, "reference_mean_power": 0.0008141211968758593, "output_mean_power": 0.0008069024138523423}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.204812078547339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002902058778585621, "output_power": 0.00029177446909300843, "reference_active": true, "delta_db": 0.023410830645620065}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7747321249654287e-05, "output_power": 1.63109069320706e-05, "reference_active": true, "delta_db": -0.3665470097050447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.319074590753967e-07, "output_power": 2.9948908737647397e-07, "reference_active": true, "delta_db": -0.4463601015088523}], "active_band_count": 3, "active_band_mae_db": 0.2787726472865057}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 2.8704944305868236e-05, "log_rms_mae_db": 1.3475955807653581, "rms_envelope_nrmse": 0.19926690555110782, "envelope_correlation": 0.9662539371857413, "rms_gain_error_db": -0.26905134973224454, "reference_mean_power": 4.5961113590202676e-05, "output_mean_power": 4.320016520897268e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 2.8606659711814982e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3098445296769517e-05, "output_power": 1.2129756932001751e-05, "reference_active": true, "delta_db": -0.33367652552202387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2173629500624627e-06, "output_power": 2.9209426150079824e-06, "reference_active": true, "delta_db": -0.4197703145294635}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2063471779341703e-06, "output_power": 1.4250037972044179e-06, "reference_active": true, "delta_db": 0.7234370902716096}], "active_band_count": 3, "active_band_mae_db": 0.4922946434410323}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 2.1311353675739154e-05, "log_rms_mae_db": 0.7534890767825798, "rms_envelope_nrmse": 0.12643143018154143, "envelope_correlation": 0.9858932917822921, "rms_gain_error_db": -0.2690553152669893, "reference_mean_power": 4.633078215253765e-05, "output_mean_power": 4.3547587500457844e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 2.5755466765658237e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2064714839118615e-05, "output_power": 1.1318614217300743e-05, "reference_active": true, "delta_db": -0.27723803691447135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.637842198880982e-06, "output_power": 1.4438413721347687e-06, "reference_active": true, "delta_db": -0.5475257446516315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.209021145285469e-07, "output_power": 9.721326447420032e-08, "reference_active": true, "delta_db": -0.947083693404341}], "active_band_count": 3, "active_band_mae_db": 0.590615824990148}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 6.327125594838897e-05, "log_rms_mae_db": 1.2582146028462755, "rms_envelope_nrmse": 0.18438613063048356, "envelope_correlation": 0.9785157307303636, "rms_gain_error_db": -0.3877055243395613, "reference_mean_power": 0.0001246130887459015, "output_mean_power": 0.00011397067185535368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.481947168400202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.119164755141333e-05, "output_power": 5.7091422347928657e-05, "reference_active": true, "delta_db": -0.30121283550877553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1929332977586818e-05, "output_power": 1.6435772498853776e-05, "reference_active": true, "delta_db": -1.2523530078326486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.860956874662985e-06, "output_power": 3.4714859061677795e-06, "reference_active": true, "delta_db": -2.958692831161884}], "active_band_count": 3, "active_band_mae_db": 1.5040862248344358}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 5.46294552803286e-05, "log_rms_mae_db": 0.7934891311925828, "rms_envelope_nrmse": 0.11143904543661906, "envelope_correlation": 0.9926300942671622, "rms_gain_error_db": -0.38771332655796686, "reference_mean_power": 0.00012561601350824365, "output_mean_power": 0.00011488773674879192}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.3825123163038373e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.737559737198788e-05, "output_power": 5.248122305723672e-05, "reference_active": true, "delta_db": -0.3872327269869178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1774546568382367e-05, "output_power": 8.2320883760409e-06, "reference_active": true, "delta_db": -1.554341674981761}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.943011691356486e-07, "output_power": 2.6649942991926633e-07, "reference_active": true, "delta_db": -3.4831030027741483}], "active_band_count": 3, "active_band_mae_db": 1.8082258015809423}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8296574403809203, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "09783ae77350ba4f4e8b01a3ae2f4e73393e4449b683c71bc4097fef58d9b951", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9728092075288908, "standard_deviation": 0.003794740105748144, "minimum": 0.9662539371857413, "maximum": 0.9770053463397657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3097806028893404, "standard_deviation": 0.04703116402432397, "minimum": 1.2449950374517509, "maximum": 1.392780627942069, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.510644265424794, "standard_deviation": 0.12358247367886559, "minimum": 0.3317370361577206, "maximum": 0.7033225752398197, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17967498742415935, "standard_deviation": 0.01171998058262444, "minimum": 0.16553395188859277, "maximum": 0.19926690555110782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9872254808650304, "standard_deviation": 0.0016051805631230138, "minimum": 0.985609259348598, "maximum": 0.9899978602626782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7897202865579616, "standard_deviation": 0.03160478108879185, "minimum": 0.7449081140513062, "maximum": 0.8277022504116707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4124267915547813, "standard_deviation": 0.17322268299027477, "minimum": 0.2118480952686748, "maximum": 0.6401506892071438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1214353659063352, "standard_deviation": 0.006315946224145465, "minimum": 0.11339128227883147, "maximum": 0.13052302065233043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9806186556544297, "standard_deviation": 0.003671779322776871, "minimum": 0.9766514045193264, "maximum": 0.987781667376504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.308174836889108, "standard_deviation": 0.03325108051970956, "minimum": 1.2582146028462755, "maximum": 1.348230950105428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7381124014975811, "standard_deviation": 0.45680442494733864, "minimum": 0.2285977270509819, "maximum": 1.5040862248344358, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17610397373211956, "standard_deviation": 0.017963837236493783, "minimum": 0.13829992679943148, "maximum": 0.19309216662523482, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937404468571154, "standard_deviation": 0.0014684955231526358, "minimum": 0.9914600219046094, "maximum": 0.995404264459176, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8087068618226209, "standard_deviation": 0.026439388101110118, "minimum": 0.7771757936818714, "maximum": 0.8492987659895437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8651284089454079, "standard_deviation": 0.5137608954543832, "minimum": 0.23179543564491664, "maximum": 1.8082258015809423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10199857633805173, "standard_deviation": 0.013822707505355157, "minimum": 0.08325660684399813, "maximum": 0.12111646407441125, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9967031653379618, "standard_deviation": 0.0004913360616817553, "minimum": 0.9958399153137302, "maximum": 0.9974441272868882, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2299128424675496, "standard_deviation": 0.043425834160381975, "minimum": 1.1864440270860999, "maximum": 1.2987528228648577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4973612727504342, "standard_deviation": 0.16816628952193682, "minimum": 0.22866822139406676, "maximum": 0.7144811132750911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07032905820786355, "standard_deviation": 0.004888925741578365, "minimum": 0.0633063790411768, "maximum": 0.07943497283324305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989563619665286, "standard_deviation": 0.0004058675047278362, "minimum": 0.9981035573789224, "maximum": 0.9992962939721229, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6893283010372357, "standard_deviation": 0.047653740335287, "minimum": 0.6098689164485805, "maximum": 0.7609142083885015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.49003988113221786, "standard_deviation": 0.2749536758554718, "minimum": 0.1104071251783094, "maximum": 0.891967482465986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04032805593033282, "standard_deviation": 0.00662578107278123, "minimum": 0.0340350049645689, "maximum": 0.054605627249765226, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1634728968727057, "standard_deviation": 0.04022335230134586, "minimum": 0.10139293301435097, "maximum": 0.22157144017483588, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.384397052172373, "standard_deviation": 0.006685274172847439, "minimum": 5.373439303661549, "maximum": 5.392294408004587, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.29137262314266343, "standard_deviation": 0.007892697859428019, "minimum": 0.279566173748477, "maximum": 0.30551111349016524, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0627461520531145, "standard_deviation": 0.03518480237860207, "minimum": -0.10605505829069553, "maximum": -0.006775265211383944, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949378107196409, "standard_deviation": 0.00048618875185346694, "minimum": 0.9943445123115592, "maximum": 0.9958362752329663, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0732817597712987, "standard_deviation": 0.04207300288213432, "minimum": 1.0282992967589628, "maximum": 1.1499086579177162, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3302387208260329, "standard_deviation": 0.2439122377939991, "minimum": 0.15169528876963273, "maximum": 0.7570757273274515, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07060248716365611, "standard_deviation": 0.0034350566450534483, "minimum": 0.06427241465908051, "maximum": 0.07459685976570875, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985357372396975, "standard_deviation": 0.00036951254225619883, "minimum": 0.9979481451375168, "maximum": 0.999023050897763, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5836649296257211, "standard_deviation": 0.051402862835059855, "minimum": 0.5313558665435024, "maximum": 0.681882022491339, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4084969789963504, "standard_deviation": 0.22038150732324396, "minimum": 0.13083026561614433, "maximum": 0.8891718273787595, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03747281061276321, "standard_deviation": 0.004273003086566456, "minimum": 0.03136817949181265, "maximum": 0.043972819452270676, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0044", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5913405418395996, "predicted_mos": 2.5913405418395996, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "e10c259a97ecc08c24afc6fd58502beecf3c29df8ca7d0084c390df04596dddd", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8504746998660266, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0084502925944405, "全帯域_10ms_rms_envelope_nrmse": 0.0723598526104339, "全帯域_10ms_envelope_correlation": 0.9947482960024748, "全帯域_10ms_modulation_mae_db": 0.15975249184269513, "全帯域_40ms_log_rms_mae_db": 0.5469655784548304, "全帯域_40ms_rms_envelope_nrmse": 0.0352340069471188, "全帯域_40ms_envelope_correlation": 0.9988125878993325, "全帯域_40ms_modulation_mae_db": 0.17618705519361968, "300–1000 Hz_10ms_log_rms_mae_db": 1.5184849013830126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06677987777426193, "300–1000 Hz_10ms_envelope_correlation": 0.9970760960403368, "300–1000 Hz_10ms_modulation_mae_db": 0.47215729062118356, "300–1000 Hz_40ms_log_rms_mae_db": 1.0853107442756196, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036094468211404204, "300–1000 Hz_40ms_envelope_correlation": 0.9992393903755026, "300–1000 Hz_40ms_modulation_mae_db": 0.41615252998152735, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7128568025129332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2188428506049008, "1000–3000 Hz_10ms_envelope_correlation": 0.9590937906435886, "1000–3000 Hz_10ms_modulation_mae_db": 0.770171340486769, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1355103706882625, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15532746356356839, "1000–3000 Hz_40ms_envelope_correlation": 0.9786677000023275, "1000–3000 Hz_40ms_modulation_mae_db": 0.4036808901043715, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7254204729941351, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1880572884044948, "3000–6000 Hz_10ms_envelope_correlation": 0.9787606123060721, "3000–6000 Hz_10ms_modulation_mae_db": 1.0617390394447395, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1862804263862743, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11539645640930657, "3000–6000 Hz_40ms_envelope_correlation": 0.9935961189600825, "3000–6000 Hz_40ms_modulation_mae_db": 1.2182146167723193, "mr_stft_db": 5.632351009384003, "band_mae_db": 0.29668481881172043, "rms_gain_error_db": -0.10745046248185233, "output_peak": 0.3027475319445152}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9853049238884886, "全帯域_10ms_rms_envelope_nrmse": 0.06732207942936157, "全帯域_10ms_envelope_correlation": 0.9954194800104453, "全帯域_10ms_modulation_mae_db": 0.08502557396834425, "全帯域_40ms_log_rms_mae_db": 0.4884514455079751, "全帯域_40ms_rms_envelope_nrmse": 0.03736632951028834, "全帯域_40ms_envelope_correlation": 0.9985815174498777, "全帯域_40ms_modulation_mae_db": 0.20479751652346326, "300–1000 Hz_10ms_log_rms_mae_db": 1.5301809666754793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06353805779187177, "300–1000 Hz_10ms_envelope_correlation": 0.9974683034205656, "300–1000 Hz_10ms_modulation_mae_db": 0.404325287582612, "300–1000 Hz_40ms_log_rms_mae_db": 1.0417331563673393, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04272311627385169, "300–1000 Hz_40ms_envelope_correlation": 0.9989768209664158, "300–1000 Hz_40ms_modulation_mae_db": 0.32044876893654434, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6501856931677477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19411805891362358, "1000–3000 Hz_10ms_envelope_correlation": 0.9679016136311291, "1000–3000 Hz_10ms_modulation_mae_db": 0.721313175433593, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0777610188294415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13636226323379425, "1000–3000 Hz_40ms_envelope_correlation": 0.9832218025287535, "1000–3000 Hz_40ms_modulation_mae_db": 0.5263394947274377, "3000–6000 Hz_10ms_log_rms_mae_db": 1.802713658628846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18762790228791965, "3000–6000 Hz_10ms_envelope_correlation": 0.9774185608260137, "3000–6000 Hz_10ms_modulation_mae_db": 0.5259803216477547, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2820960828443597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10771266378036218, "3000–6000 Hz_40ms_envelope_correlation": 0.9922526141384584, "3000–6000 Hz_40ms_modulation_mae_db": 0.9220174786000538, "mr_stft_db": 5.617917616600845, "band_mae_db": 0.09666023971409358, "rms_gain_error_db": -0.0801548340281248, "output_peak": 0.2828029237620916}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0345991884456305, "全帯域_10ms_rms_envelope_nrmse": 0.06539019111666818, "全帯域_10ms_envelope_correlation": 0.9957329266712006, "全帯域_10ms_modulation_mae_db": 0.2171186773922357, "全帯域_40ms_log_rms_mae_db": 0.5269243174147605, "全帯域_40ms_rms_envelope_nrmse": 0.036600832067442184, "全帯域_40ms_envelope_correlation": 0.9987104708985384, "全帯域_40ms_modulation_mae_db": 0.44541428831938273, "300–1000 Hz_10ms_log_rms_mae_db": 1.48420538378256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06103804814004716, "300–1000 Hz_10ms_envelope_correlation": 0.9977360232117835, "300–1000 Hz_10ms_modulation_mae_db": 0.3943915305296571, "300–1000 Hz_40ms_log_rms_mae_db": 0.9879642366218496, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03917431444120884, "300–1000 Hz_40ms_envelope_correlation": 0.999236627647707, "300–1000 Hz_40ms_modulation_mae_db": 0.35202181705310065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.668027699922895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2055079573112316, "1000–3000 Hz_10ms_envelope_correlation": 0.9639054135387298, "1000–3000 Hz_10ms_modulation_mae_db": 0.5683869991266165, "1000–3000 Hz_40ms_log_rms_mae_db": 1.094217345664559, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1428684283850836, "1000–3000 Hz_40ms_envelope_correlation": 0.9817881616596758, "1000–3000 Hz_40ms_modulation_mae_db": 0.4441527044761427, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7129624926885907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15369296557944534, "3000–6000 Hz_10ms_envelope_correlation": 0.9847909138228904, "3000–6000 Hz_10ms_modulation_mae_db": 0.8854270897730618, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2226620555371965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08812173368570717, "3000–6000 Hz_40ms_envelope_correlation": 0.9948574168129739, "3000–6000 Hz_40ms_modulation_mae_db": 1.0457724417752121, "mr_stft_db": 5.628368241863211, "band_mae_db": 0.16577462716198252, "rms_gain_error_db": -0.09794760064980815, "output_peak": 0.2854080647777078}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0616896415286443, "全帯域_10ms_rms_envelope_nrmse": 0.0753579734508597, "全帯域_10ms_envelope_correlation": 0.9942295533546789, "全帯域_10ms_modulation_mae_db": 0.4325765615290595, "全帯域_40ms_log_rms_mae_db": 0.6257132222930915, "全帯域_40ms_rms_envelope_nrmse": 0.04287197754308715, "全帯域_40ms_envelope_correlation": 0.9980916684226716, "全帯域_40ms_modulation_mae_db": 0.534876070550463, "300–1000 Hz_10ms_log_rms_mae_db": 1.6230330433165072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0701426697011263, "300–1000 Hz_10ms_envelope_correlation": 0.9967902481260231, "300–1000 Hz_10ms_modulation_mae_db": 1.0149124051120713, "300–1000 Hz_40ms_log_rms_mae_db": 1.108793052913528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04584008289151129, "300–1000 Hz_40ms_envelope_correlation": 0.9987053205763499, "300–1000 Hz_40ms_modulation_mae_db": 0.7457773565539411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6179632969029274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19235884917480744, "1000–3000 Hz_10ms_envelope_correlation": 0.9686262510406898, "1000–3000 Hz_10ms_modulation_mae_db": 0.325207366718292, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1027711013973371, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13492283567902078, "1000–3000 Hz_40ms_envelope_correlation": 0.9840488569691558, "1000–3000 Hz_40ms_modulation_mae_db": 0.9236521334513504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.811438412454687, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20857424753021575, "3000–6000 Hz_10ms_envelope_correlation": 0.97191261373255, "3000–6000 Hz_10ms_modulation_mae_db": 1.3025828995545126, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2589751396735573, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12739892011357737, "3000–6000 Hz_40ms_envelope_correlation": 0.9891779495196085, "3000–6000 Hz_40ms_modulation_mae_db": 1.6631947347848455, "mr_stft_db": 5.636176453221037, "band_mae_db": 0.19694801021932762, "rms_gain_error_db": -0.07133020054430443, "output_peak": 0.28993208256241276}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9966020454159159, "全帯域_10ms_rms_envelope_nrmse": 0.0697657770539881, "全帯域_10ms_envelope_correlation": 0.9950509466595595, "全帯域_10ms_modulation_mae_db": 0.24051769789855282, "全帯域_40ms_log_rms_mae_db": 0.5840716166199527, "全帯域_40ms_rms_envelope_nrmse": 0.04097972349879978, "全帯域_40ms_envelope_correlation": 0.9981991297506879, "全帯域_40ms_modulation_mae_db": 0.47034016545628926, "300–1000 Hz_10ms_log_rms_mae_db": 1.6457970152050918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06514141496446016, "300–1000 Hz_10ms_envelope_correlation": 0.9971417837691481, "300–1000 Hz_10ms_modulation_mae_db": 0.38890128823511255, "300–1000 Hz_40ms_log_rms_mae_db": 1.0920452648910426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03827757451263631, "300–1000 Hz_40ms_envelope_correlation": 0.99902581370329, "300–1000 Hz_40ms_modulation_mae_db": 0.3836941015012498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6255723276525949, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2173039025184255, "1000–3000 Hz_10ms_envelope_correlation": 0.9597403660071445, "1000–3000 Hz_10ms_modulation_mae_db": 0.7545926863893122, "1000–3000 Hz_40ms_log_rms_mae_db": 1.138926314332256, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1543310038276581, "1000–3000 Hz_40ms_envelope_correlation": 0.9782390160050299, "1000–3000 Hz_40ms_modulation_mae_db": 0.43251583760189466, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7053432637151975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1730123772301929, "3000–6000 Hz_10ms_envelope_correlation": 0.9816308404155584, "3000–6000 Hz_10ms_modulation_mae_db": 0.1438554986845767, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1450452525318473, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09781808640728455, "3000–6000 Hz_40ms_envelope_correlation": 0.9941068372707952, "3000–6000 Hz_40ms_modulation_mae_db": 0.18172070882264432, "mr_stft_db": 5.618176802226113, "band_mae_db": 0.2476657203902072, "rms_gain_error_db": 0.014901909054817059, "output_peak": 0.2909696658275891}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9687223144874351, "全帯域_10ms_rms_envelope_nrmse": 0.06483725798462842, "全帯域_10ms_envelope_correlation": 0.9957664482895818, "全帯域_10ms_modulation_mae_db": 0.19766236216707878, "全帯域_40ms_log_rms_mae_db": 0.5141480632948002, "全帯域_40ms_rms_envelope_nrmse": 0.032941049331521, "全帯域_40ms_envelope_correlation": 0.9989178472934106, "全帯域_40ms_modulation_mae_db": 0.15669174880411585, "300–1000 Hz_10ms_log_rms_mae_db": 1.4456761505865523, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06749372450397285, "300–1000 Hz_10ms_envelope_correlation": 0.9969793720988379, "300–1000 Hz_10ms_modulation_mae_db": 0.38165550016346367, "300–1000 Hz_40ms_log_rms_mae_db": 0.9837893862609698, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04015228217517395, "300–1000 Hz_40ms_envelope_correlation": 0.9989859843619306, "300–1000 Hz_40ms_modulation_mae_db": 0.3424614890118798, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7249647233121777, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21015235955757497, "1000–3000 Hz_10ms_envelope_correlation": 0.9622786087415446, "1000–3000 Hz_10ms_modulation_mae_db": 0.7450834691579092, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1079595190042375, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1383152025437383, "1000–3000 Hz_40ms_envelope_correlation": 0.9832479561571607, "1000–3000 Hz_40ms_modulation_mae_db": 0.8127571279282343, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7167295391296558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1874432982978711, "3000–6000 Hz_10ms_envelope_correlation": 0.9775651482367613, "3000–6000 Hz_10ms_modulation_mae_db": 1.03986104778602, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1915765578191986, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11779156343980944, "3000–6000 Hz_40ms_envelope_correlation": 0.9913209303772555, "3000–6000 Hz_40ms_modulation_mae_db": 1.2734040328816647, "mr_stft_db": 5.62795470085063, "band_mae_db": 0.2008455372579011, "rms_gain_error_db": -0.08157070291065394, "output_peak": 0.2857131013275735}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0925527384154552, "全帯域_10ms_rms_envelope_nrmse": 0.07188530936050432, "全帯域_10ms_envelope_correlation": 0.9947482715046936, "全帯域_10ms_modulation_mae_db": 0.19846659576037914, "全帯域_40ms_log_rms_mae_db": 0.6207808408937805, "全帯域_40ms_rms_envelope_nrmse": 0.03397595515963472, "全帯域_40ms_envelope_correlation": 0.9988203141817631, "全帯域_40ms_modulation_mae_db": 0.2496765296966388, "300–1000 Hz_10ms_log_rms_mae_db": 1.528049119895593, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07263040214137809, "300–1000 Hz_10ms_envelope_correlation": 0.996501330484504, "300–1000 Hz_10ms_modulation_mae_db": 0.39357763382923067, "300–1000 Hz_40ms_log_rms_mae_db": 1.0580551408062915, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038187278812542066, "300–1000 Hz_40ms_envelope_correlation": 0.9991195223357493, "300–1000 Hz_40ms_modulation_mae_db": 0.6614984714335973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5721857149706748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21066618669030995, "1000–3000 Hz_10ms_envelope_correlation": 0.9619878612529587, "1000–3000 Hz_10ms_modulation_mae_db": 0.47056991648551466, "1000–3000 Hz_40ms_log_rms_mae_db": 1.073416157170651, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1495727750025288, "1000–3000 Hz_40ms_envelope_correlation": 0.9796752544477343, "1000–3000 Hz_40ms_modulation_mae_db": 0.5523824174303283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7331962231777307, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.184393817200993, "3000–6000 Hz_10ms_envelope_correlation": 0.9784205984774316, "3000–6000 Hz_10ms_modulation_mae_db": 0.9791810137636303, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1733168432963355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10660141690861309, "3000–6000 Hz_40ms_envelope_correlation": 0.9932175330726193, "3000–6000 Hz_40ms_modulation_mae_db": 1.552862591407879, "mr_stft_db": 5.608368693672194, "band_mae_db": 0.18098497425466778, "rms_gain_error_db": -0.06901319695192464, "output_peak": 0.2981708537282487}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0105276898408502, "全帯域_10ms_rms_envelope_nrmse": 0.07142694886303545, "全帯域_10ms_envelope_correlation": 0.9949357599819936, "全帯域_10ms_modulation_mae_db": 0.6703320500449506, "全帯域_40ms_log_rms_mae_db": 0.5785687607501092, "全帯域_40ms_rms_envelope_nrmse": 0.03852904660865184, "全帯域_40ms_envelope_correlation": 0.9986105400119465, "全帯域_40ms_modulation_mae_db": 0.9189259766618146, "300–1000 Hz_10ms_log_rms_mae_db": 1.553553171420522, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0661135843755043, "300–1000 Hz_10ms_envelope_correlation": 0.9971370792046912, "300–1000 Hz_10ms_modulation_mae_db": 0.40632751511373705, "300–1000 Hz_40ms_log_rms_mae_db": 1.0119816638728123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03923451865419954, "300–1000 Hz_40ms_envelope_correlation": 0.9990708605003515, "300–1000 Hz_40ms_modulation_mae_db": 0.5392165222752484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6835125751318036, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21272185358758383, "1000–3000 Hz_10ms_envelope_correlation": 0.961792105389222, "1000–3000 Hz_10ms_modulation_mae_db": 0.6458219292049738, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0642996533256432, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14679974923589698, "1000–3000 Hz_40ms_envelope_correlation": 0.980462722875825, "1000–3000 Hz_40ms_modulation_mae_db": 0.4273079902809765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7888017296327692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20496249520961138, "3000–6000 Hz_10ms_envelope_correlation": 0.9769356822085894, "3000–6000 Hz_10ms_modulation_mae_db": 2.1015950339243634, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2754041301754704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13604067008932463, "3000–6000 Hz_40ms_envelope_correlation": 0.9932283727753317, "3000–6000 Hz_40ms_modulation_mae_db": 2.5165380791802434, "mr_stft_db": 5.626802771793442, "band_mae_db": 0.34110978327149805, "rms_gain_error_db": -0.1273659744164498, "output_peak": 0.2900802710614004}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59840, "output_samples": 59840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.71246177305055, "spectral_convergence": 0.28464917031854364, "spectral_convergence_status": "ok", "amplitude_floor": 9.91771656423115e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.437950648501042, "spectral_convergence": 0.2551177005155103, "spectral_convergence_status": "ok", "amplitude_floor": 5.25712991724442e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.746640606600415, "spectral_convergence": 0.25440780009598624, "spectral_convergence_status": "ok", "amplitude_floor": 4.1776466342690114e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00025386719019706026, "output_power": 0.0002581268849920496, "delta_db": 0.07226664212830117, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004894966727833924, "output_power": 0.00048110209703406114, "delta_db": -0.07512494238910411, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005421484972995941, "output_power": 0.0005268835130511713, "delta_db": -0.12403649273606351, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.5605939839496278e-05, "output_power": 1.790922277804566e-05, "delta_db": 0.5978681041809554, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000107892990107672, "output_power": 9.355606439014583e-05, "delta_db": -0.6192128502957845, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.406384293716839e-05, "output_power": 6.92542219299699e-05, "delta_db": -0.29159988114011365, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29668481881172043, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10745046248185233, "pitch": null, "output_rms_dbfs": -28.3958189313473, "output_peak": 0.3027475319445152, "reference_rms_dbfs": -28.28836846886545, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.632351009384003}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.000110539633783132, "log_rms_mae_db": 1.0084502925944405, "rms_envelope_nrmse": 0.0723598526104339, "envelope_correlation": 0.9947482960024748, "rms_gain_error_db": -0.10745002706220934, "reference_mean_power": 0.00148505750597689, "output_mean_power": 0.0014487660773621254}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.0996976338580339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028388367321682654, "output_power": 0.0002737026687153307, "reference_active": true, "delta_db": -0.1586138401649817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6121837601531416e-05, "output_power": 4.6806809353463934e-05, "reference_active": true, "delta_db": 0.06402435620863949}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3216272322760606e-05, "output_power": 1.245796340585656e-05, "reference_active": true, "delta_db": -0.2566192791544642}], "active_band_count": 3, "active_band_mae_db": 0.15975249184269513}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 0.00010014294879668002, "log_rms_mae_db": 0.5469655784548304, "rms_envelope_nrmse": 0.0352340069471188, "envelope_correlation": 0.9988125878993325, "rms_gain_error_db": -0.10745857512739262, "reference_mean_power": 0.0014970488171277764, "output_mean_power": 0.001460461473557535}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.0259449632693504e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002533237576934496, "output_power": 0.00024631005334883904, "reference_active": true, "delta_db": -0.12193783246956275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.399735365000864e-05, "output_power": 2.3027475122154265e-05, "reference_active": true, "delta_db": -0.17917029987287628}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.21988770898296e-07, "output_power": 5.90251692262938e-07, "reference_active": true, "delta_db": -0.22745303323841998}], "active_band_count": 3, "active_band_mae_db": 0.17618705519361968}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.00010594974849403162, "log_rms_mae_db": 1.5184849013830126, "rms_envelope_nrmse": 0.06677987777426193, "envelope_correlation": 0.9970760960403368, "rms_gain_error_db": -0.12599167208863124, "reference_mean_power": 0.0008075829245111437, "output_mean_power": 0.000784490990810552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.2363198879746262e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030947682521494476, "output_power": 0.00030032046155860495, "reference_active": true, "delta_db": -0.130432102352384}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.465844834587486e-05, "output_power": 2.878926597548249e-05, "reference_active": true, "delta_db": -0.8057852346179559}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.693327955984243e-06, "output_power": 7.4759701443677396e-06, "reference_active": true, "delta_db": 0.4802545348932109}], "active_band_count": 3, "active_band_mae_db": 0.47215729062118356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 9.562043532359368e-05, "log_rms_mae_db": 1.0853107442756196, "rms_envelope_nrmse": 0.036094468211404204, "envelope_correlation": 0.9992393903755026, "rms_gain_error_db": -0.12599293646117118, "reference_mean_power": 0.0008141211968758593, "output_mean_power": 0.0007908420783238042}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.204812078547339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002902058778585621, "output_power": 0.00028074887211253935, "reference_active": true, "delta_db": -0.14388184242592228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7747321249654287e-05, "output_power": 1.4145340115120115e-05, "reference_active": true, "delta_db": -0.9851941643554663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.319074590753967e-07, "output_power": 3.229080368214064e-07, "reference_active": true, "delta_db": -0.11938158316319337}], "active_band_count": 3, "active_band_mae_db": 0.41615252998152735}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 2.8704944305868236e-05, "log_rms_mae_db": 1.7128568025129332, "rms_envelope_nrmse": 0.2188428506049008, "envelope_correlation": 0.9590937906435886, "rms_gain_error_db": -0.3419240067227135, "reference_mean_power": 4.5961113590202676e-05, "output_mean_power": 4.248133360506859e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 2.8606659711814982e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3098445296769517e-05, "output_power": 1.1577217972290677e-05, "reference_active": true, "delta_db": -0.5361554055539731}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2173629500624627e-06, "output_power": 3.083396246575935e-06, "reference_active": true, "delta_db": -0.1847071693646022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2063471779341703e-06, "output_power": 1.7395521670313847e-06, "reference_active": true, "delta_db": 1.589651446541732}], "active_band_count": 3, "active_band_mae_db": 0.770171340486769}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 2.1311353675739154e-05, "log_rms_mae_db": 1.1355103706882625, "rms_envelope_nrmse": 0.15532746356356839, "envelope_correlation": 0.9786677000023275, "rms_gain_error_db": -0.34201205941445095, "reference_mean_power": 4.633078215253765e-05, "output_mean_power": 4.2822145822452215e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 2.5755466765658237e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2064714839118615e-05, "output_power": 1.0830469311857627e-05, "reference_active": true, "delta_db": -0.4686978528699542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.637842198880982e-06, "output_power": 1.735457778282844e-06, "reference_active": true, "delta_db": 0.25141995758015345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.209021145285469e-07, "output_power": 1.0797952447489854e-07, "reference_active": true, "delta_db": -0.49092485986300693}], "active_band_count": 3, "active_band_mae_db": 0.4036808901043715}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 6.327125594838897e-05, "log_rms_mae_db": 1.7254204729941351, "rms_envelope_nrmse": 0.1880572884044948, "envelope_correlation": 0.9787606123060721, "rms_gain_error_db": -0.5909033267476174, "reference_mean_power": 0.0001246130887459015, "output_mean_power": 0.00010876103409825239}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.481947168400202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.119164755141333e-05, "output_power": 5.146057956751968e-05, "reference_active": true, "delta_db": -0.7521747341161221}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1929332977586818e-05, "output_power": 2.0590097967982333e-05, "reference_active": true, "delta_db": -0.2736700898257234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.860956874662985e-06, "output_power": 4.172991115585419e-06, "reference_active": true, "delta_db": -2.1593722943923725}], "active_band_count": 3, "active_band_mae_db": 1.0617390394447395}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 5.46294552803286e-05, "log_rms_mae_db": 1.1862804263862743, "rms_envelope_nrmse": 0.11539645640930657, "envelope_correlation": 0.9935961189600825, "rms_gain_error_db": -0.5909142274196839, "reference_mean_power": 0.00012561601350824365, "output_mean_power": 0.00010963610142746425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.3825123163038373e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.737559737198788e-05, "output_power": 4.854646614258349e-05, "reference_active": true, "delta_db": -0.7256959860433183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1774546568382367e-05, "output_power": 1.0422251452589061e-05, "reference_active": true, "delta_db": -0.5298264470128488}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.943011691356486e-07, "output_power": 3.4205381840933397e-07, "reference_active": true, "delta_db": -2.3991214172607913}], "active_band_count": 3, "active_band_mae_db": 1.2182146167723193}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8009588327010908, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c31fb3eb8ed9faa400def113131ece2332834dcdb3d80622490af5311136b9ce", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9631657512806259, "standard_deviation": 0.0034881027922864594, "minimum": 0.9590937906435886, "maximum": 0.9686262510406898, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6569086041967191, "standard_deviation": 0.051151185550484014, "minimum": 1.5721857149706748, "maximum": 1.7249647233121777, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6251433603753725, "standard_deviation": 0.1598785001367445, "minimum": 0.325207366718292, "maximum": 0.770171340486769, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2077090022948072, "standard_deviation": 0.009863741992462541, "minimum": 0.19235884917480744, "maximum": 0.2188428506049008, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9811689338307078, "standard_deviation": 0.002229170417577451, "minimum": 0.9782390160050299, "maximum": 0.9840488569691558, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0993576850515485, "standard_deviation": 0.027653993766369858, "minimum": 1.0642996533256432, "maximum": 1.138926314332256, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.565348574500092, "standard_deviation": 0.19596168728043598, "minimum": 0.4036808901043715, "maximum": 0.9236521334513504, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14481246518391114, "standard_deviation": 0.007955070278414017, "minimum": 0.13492283567902078, "maximum": 0.15532746356356839, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9784293712532334, "standard_deviation": 0.0037281991238016223, "minimum": 0.97191261373255, "maximum": 0.9847909138228904, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7495757240527015, "standard_deviation": 0.04378027466165961, "minimum": 1.7053432637151975, "maximum": 1.811438412454687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0050277430723324, "standard_deviation": 0.5709021508520107, "minimum": 0.1438554986845767, "maximum": 2.1015950339243634, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.185970548967593, "standard_deviation": 0.017292170772115215, "minimum": 0.15369296557944534, "maximum": 0.20857424753021575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927197216158906, "standard_deviation": 0.001792340889763602, "minimum": 0.9891779495196085, "maximum": 0.9948574168129739, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.21691956103303, "standard_deviation": 0.05087549137911794, "minimum": 1.1450452525318473, "maximum": 1.2820960828443597, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2967155855281078, "standard_deviation": 0.6698801289222999, "minimum": 0.18172070882264432, "maximum": 2.5165380791802434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11211018885424813, "standard_deviation": 0.015486795186957107, "minimum": 0.08812173368570717, "maximum": 0.13604067008932463, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9971037795444863, "standard_deviation": 0.0003803482067607851, "minimum": 0.996501330484504, "maximum": 0.9977360232117835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5411224690331646, "standard_deviation": 0.06651881351127768, "minimum": 1.4456761505865523, "maximum": 1.6457970152050918, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48203105639838345, "standard_deviation": 0.21715114426662582, "minimum": 0.38165550016346367, "maximum": 1.0149124051120713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06660972242407782, "standard_deviation": 0.003633602520530819, "minimum": 0.06103804814004716, "maximum": 0.07263040214137809, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9990450425584121, "standard_deviation": 0.00017096890024542936, "minimum": 0.9987053205763499, "maximum": 0.9992393903755026, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0462090807511817, "standard_deviation": 0.048058310817597004, "minimum": 0.9837893862609698, "maximum": 1.108793052913528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4701588820933861, "standard_deviation": 0.16054117951887784, "minimum": 0.32044876893654434, "maximum": 0.7457773565539411, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03996045449656599, "standard_deviation": 0.003030940779320971, "minimum": 0.036094468211404204, "maximum": 0.04584008289151129, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2158342138851748, "standard_deviation": 0.07722716987640282, "minimum": 0.09666023971409358, "maximum": 0.34110978327149805, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.624514536201435, "standard_deviation": 0.009053072674248269, "minimum": 5.608368693672194, "maximum": 5.636176453221037, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2907280618739424, "standard_deviation": 0.0067319537846753106, "minimum": 0.2828029237620916, "maximum": 0.3027475319445152, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07749138286603763, "standard_deviation": 0.04221301965685743, "minimum": -0.1273659744164498, "maximum": 0.014901909054817059, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9950789603093285, "standard_deviation": 0.0005315398969797871, "minimum": 0.9942295533546789, "maximum": 0.9957664482895818, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0198061043271074, "standard_deviation": 0.041089615683139534, "minimum": 0.9687223144874351, "maximum": 1.0925527384154552, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27518150132541197, "standard_deviation": 0.1877945528389475, "minimum": 0.08502557396834425, "maximum": 0.6703320500449506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06979317373368496, "standard_deviation": 0.003677256420588685, "minimum": 0.06483725798462842, "maximum": 0.0753579734508597, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985930094885285, "standard_deviation": 0.0002990753327888544, "minimum": 0.9980916684226716, "maximum": 0.9989178472934106, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5607029806536625, "standard_deviation": 0.049883176415366874, "minimum": 0.4884514455079751, "maximum": 0.6257132222930915, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3946136689007234, "standard_deviation": 0.2575283935142448, "minimum": 0.15669174880411585, "maximum": 0.9189259766618146, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03731236508331798, "standard_deviation": 0.003399746185896823, "minimum": 0.032941049331521, "maximum": 0.04287197754308715, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0044", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.2938530445098877, "predicted_mos": 3.2938530445098877, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "3509acc6975997ae0f53f87725504803117c94dda65f70b14050afa037e34cad", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.07047659996897, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0886325638202003, "全帯域_10ms_rms_envelope_nrmse": 0.0695318076976679, "全帯域_10ms_envelope_correlation": 0.9950830547615073, "全帯域_10ms_modulation_mae_db": 0.21083373805599245, "全帯域_40ms_log_rms_mae_db": 0.5702748877988298, "全帯域_40ms_rms_envelope_nrmse": 0.032920516321403234, "全帯域_40ms_envelope_correlation": 0.9988834833566931, "全帯域_40ms_modulation_mae_db": 0.1164320766530353, "300–1000 Hz_10ms_log_rms_mae_db": 1.1795890757061915, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07655669968454638, "300–1000 Hz_10ms_envelope_correlation": 0.9961245232025331, "300–1000 Hz_10ms_modulation_mae_db": 0.4760595890994961, "300–1000 Hz_40ms_log_rms_mae_db": 0.6953373263209226, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04086243516777063, "300–1000 Hz_40ms_envelope_correlation": 0.9990011890210567, "300–1000 Hz_40ms_modulation_mae_db": 0.701524073932196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2229205899939868, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17449730758091275, "1000–3000 Hz_10ms_envelope_correlation": 0.9745754432779078, "1000–3000 Hz_10ms_modulation_mae_db": 0.2888450483237663, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8100977802364481, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12595807381924218, "1000–3000 Hz_40ms_envelope_correlation": 0.9864976416610243, "1000–3000 Hz_40ms_modulation_mae_db": 0.5271721846738124, "3000–6000 Hz_10ms_log_rms_mae_db": 1.272220571202078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15712554065308332, "3000–6000 Hz_10ms_envelope_correlation": 0.9843616067061222, "3000–6000 Hz_10ms_modulation_mae_db": 0.39552134419732093, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8330188907653883, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09008568211036899, "3000–6000 Hz_40ms_envelope_correlation": 0.9950616218139224, "3000–6000 Hz_40ms_modulation_mae_db": 0.37659491858791383, "mr_stft_db": 5.374460415314196, "band_mae_db": 0.167254412038612, "rms_gain_error_db": -0.06229697968879205, "output_peak": 0.2862277975682654}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1557716048221423, "全帯域_10ms_rms_envelope_nrmse": 0.07306783185893832, "全帯域_10ms_envelope_correlation": 0.9946295983826586, "全帯域_10ms_modulation_mae_db": 0.3833670634459438, "全帯域_40ms_log_rms_mae_db": 0.6929288168296553, "全帯域_40ms_rms_envelope_nrmse": 0.03597585320054835, "全帯域_40ms_envelope_correlation": 0.9987366929635872, "全帯域_40ms_modulation_mae_db": 0.6792605015503884, "300–1000 Hz_10ms_log_rms_mae_db": 1.1557239280023022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07170524918602167, "300–1000 Hz_10ms_envelope_correlation": 0.9965233021031834, "300–1000 Hz_10ms_modulation_mae_db": 0.40389799148024513, "300–1000 Hz_40ms_log_rms_mae_db": 0.6107489842375123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03392862962626888, "300–1000 Hz_40ms_envelope_correlation": 0.9992478672682658, "300–1000 Hz_40ms_modulation_mae_db": 0.34567840784334697, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2210045866171018, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1703820279057029, "1000–3000 Hz_10ms_envelope_correlation": 0.9755908464332802, "1000–3000 Hz_10ms_modulation_mae_db": 0.1955095362948277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8000171131243807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11874034129819415, "1000–3000 Hz_40ms_envelope_correlation": 0.9876628428535643, "1000–3000 Hz_40ms_modulation_mae_db": 0.529157141613626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2503757814394025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16721457850669394, "3000–6000 Hz_10ms_envelope_correlation": 0.982843823017332, "3000–6000 Hz_10ms_modulation_mae_db": 1.1450260696412398, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7940818285311733, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09362359820471909, "3000–6000 Hz_40ms_envelope_correlation": 0.9955707595407505, "3000–6000 Hz_40ms_modulation_mae_db": 1.5551249804209506, "mr_stft_db": 5.3554441323908355, "band_mae_db": 0.2422471154382205, "rms_gain_error_db": -0.1008469897489811, "output_peak": 0.28847287343748823}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0364128012098222, "全帯域_10ms_rms_envelope_nrmse": 0.06433529090357751, "全帯域_10ms_envelope_correlation": 0.9958020040898727, "全帯域_10ms_modulation_mae_db": 0.14628133584208644, "全帯域_40ms_log_rms_mae_db": 0.5374520416208788, "全帯域_40ms_rms_envelope_nrmse": 0.031225385230019546, "全帯域_40ms_envelope_correlation": 0.9990045822126268, "全帯域_40ms_modulation_mae_db": 0.09129350845125073, "300–1000 Hz_10ms_log_rms_mae_db": 1.1483321489928622, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.058020411990609456, "300–1000 Hz_10ms_envelope_correlation": 0.9978785864459763, "300–1000 Hz_10ms_modulation_mae_db": 0.3321767162178616, "300–1000 Hz_40ms_log_rms_mae_db": 0.7101203198027988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034223578762341285, "300–1000 Hz_40ms_envelope_correlation": 0.9993840441131098, "300–1000 Hz_40ms_modulation_mae_db": 0.4878495967417776, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2645065394291146, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18621528371971444, "1000–3000 Hz_10ms_envelope_correlation": 0.9714456374626875, "1000–3000 Hz_10ms_modulation_mae_db": 0.9424976408302687, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7982294076244151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12519483096759906, "1000–3000 Hz_40ms_envelope_correlation": 0.987501659829296, "1000–3000 Hz_40ms_modulation_mae_db": 0.5685923693923345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2032792999074138, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1742841890151287, "3000–6000 Hz_10ms_envelope_correlation": 0.9808882146624728, "3000–6000 Hz_10ms_modulation_mae_db": 0.420886388129096, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7642277150411444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09254093562587963, "3000–6000 Hz_40ms_envelope_correlation": 0.994447239735753, "3000–6000 Hz_40ms_modulation_mae_db": 0.6202323545426207, "mr_stft_db": 5.34424166847942, "band_mae_db": 0.13292941045303563, "rms_gain_error_db": -0.06435744258265508, "output_peak": 0.2916746885206436}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1412677541117342, "全帯域_10ms_rms_envelope_nrmse": 0.08427507675156251, "全帯域_10ms_envelope_correlation": 0.9930449965493563, "全帯域_10ms_modulation_mae_db": 0.8487456776945557, "全帯域_40ms_log_rms_mae_db": 0.7278790143066315, "全帯域_40ms_rms_envelope_nrmse": 0.05239838461011472, "全帯域_40ms_envelope_correlation": 0.9974497107145617, "全帯域_40ms_modulation_mae_db": 0.9685118192251548, "300–1000 Hz_10ms_log_rms_mae_db": 1.1851377149251514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0804918119198505, "300–1000 Hz_10ms_envelope_correlation": 0.9958717054454878, "300–1000 Hz_10ms_modulation_mae_db": 0.6757996713881096, "300–1000 Hz_40ms_log_rms_mae_db": 0.7059427476640397, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05380961413365083, "300–1000 Hz_40ms_envelope_correlation": 0.9983304689806961, "300–1000 Hz_40ms_modulation_mae_db": 1.072517608094738, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2202992556731105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18325430818736638, "1000–3000 Hz_10ms_envelope_correlation": 0.9725958325784116, "1000–3000 Hz_10ms_modulation_mae_db": 0.40443705279029324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7723495050314242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12871250796208664, "1000–3000 Hz_40ms_envelope_correlation": 0.986939991050695, "1000–3000 Hz_40ms_modulation_mae_db": 0.9965232630180029, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2626295944502008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22844429672637923, "3000–6000 Hz_10ms_envelope_correlation": 0.9691798471266783, "3000–6000 Hz_10ms_modulation_mae_db": 1.286958520972767, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8824755732455115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15804788696105654, "3000–6000 Hz_40ms_envelope_correlation": 0.9878395058706988, "3000–6000 Hz_40ms_modulation_mae_db": 1.4552852445940356, "mr_stft_db": 5.369385529885267, "band_mae_db": 0.29824585382374, "rms_gain_error_db": -0.1804066917339952, "output_peak": 0.2931806509389691}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1381548923506848, "全帯域_10ms_rms_envelope_nrmse": 0.06989569430417943, "全帯域_10ms_envelope_correlation": 0.9950304564260603, "全帯域_10ms_modulation_mae_db": 0.08033308580779336, "全帯域_40ms_log_rms_mae_db": 0.693624927888778, "全帯域_40ms_rms_envelope_nrmse": 0.03936634266627007, "全帯域_40ms_envelope_correlation": 0.9983555662483957, "全帯域_40ms_modulation_mae_db": 0.3364784075652953, "300–1000 Hz_10ms_log_rms_mae_db": 1.1776972110085815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.053176584922942624, "300–1000 Hz_10ms_envelope_correlation": 0.9980860317045912, "300–1000 Hz_10ms_modulation_mae_db": 0.14167768792977442, "300–1000 Hz_40ms_log_rms_mae_db": 0.6933078010357565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.029080062500346098, "300–1000 Hz_40ms_envelope_correlation": 0.9994280962281059, "300–1000 Hz_40ms_modulation_mae_db": 0.45698576325125484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2034414168496144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18233848753045614, "1000–3000 Hz_10ms_envelope_correlation": 0.9733123981003375, "1000–3000 Hz_10ms_modulation_mae_db": 0.5710409317084982, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8300303374792687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13123392947516402, "1000–3000 Hz_40ms_envelope_correlation": 0.9868287356042936, "1000–3000 Hz_40ms_modulation_mae_db": 0.9914527610802134, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2304159171402147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2004899300581256, "3000–6000 Hz_10ms_envelope_correlation": 0.9742040452220089, "3000–6000 Hz_10ms_modulation_mae_db": 0.7058904629377357, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8199463785092778, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13934739867309692, "3000–6000 Hz_40ms_envelope_correlation": 0.9873490624315411, "3000–6000 Hz_40ms_modulation_mae_db": 1.1188872501884248, "mr_stft_db": 5.343663925003969, "band_mae_db": 0.16203841186854254, "rms_gain_error_db": -0.04198065371590691, "output_peak": 0.28173570002347825}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0555240725853356, "全帯域_10ms_rms_envelope_nrmse": 0.06903241138343788, "全帯域_10ms_envelope_correlation": 0.9951783324631467, "全帯域_10ms_modulation_mae_db": 0.2680020465738157, "全帯域_40ms_log_rms_mae_db": 0.5213145994343756, "全帯域_40ms_rms_envelope_nrmse": 0.035621070097940834, "全帯域_40ms_envelope_correlation": 0.9986745675013018, "全帯域_40ms_modulation_mae_db": 0.5879810210039694, "300–1000 Hz_10ms_log_rms_mae_db": 1.1278512677231014, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05646428109323584, "300–1000 Hz_10ms_envelope_correlation": 0.9978290434045612, "300–1000 Hz_10ms_modulation_mae_db": 0.4563912382250887, "300–1000 Hz_40ms_log_rms_mae_db": 0.6454185727760943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033539117819448606, "300–1000 Hz_40ms_envelope_correlation": 0.999218021503439, "300–1000 Hz_40ms_modulation_mae_db": 0.6812715287223886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2025753444642007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1820245982769886, "1000–3000 Hz_10ms_envelope_correlation": 0.9720758060852852, "1000–3000 Hz_10ms_modulation_mae_db": 0.581910034241735, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7392489604392807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11659524973722882, "1000–3000 Hz_40ms_envelope_correlation": 0.9883232077260468, "1000–3000 Hz_40ms_modulation_mae_db": 0.36980119510210363, "3000–6000 Hz_10ms_log_rms_mae_db": 1.223133395273006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17827464463021164, "3000–6000 Hz_10ms_envelope_correlation": 0.9797507316987206, "3000–6000 Hz_10ms_modulation_mae_db": 0.26748140990607455, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7498389418549973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09062507840029141, "3000–6000 Hz_40ms_envelope_correlation": 0.9951292374003674, "3000–6000 Hz_40ms_modulation_mae_db": 0.4316323502033265, "mr_stft_db": 5.343952425386974, "band_mae_db": 0.18016133227067754, "rms_gain_error_db": -0.0434454217084786, "output_peak": 0.29160442481878546}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1314250332895568, "全帯域_10ms_rms_envelope_nrmse": 0.08108266305403644, "全帯域_10ms_envelope_correlation": 0.9936844197825636, "全帯域_10ms_modulation_mae_db": 0.6140602272269061, "全帯域_40ms_log_rms_mae_db": 0.6228543470037272, "全帯域_40ms_rms_envelope_nrmse": 0.0464181444895552, "全帯域_40ms_envelope_correlation": 0.9981998695543476, "全帯域_40ms_modulation_mae_db": 0.5974948699694497, "300–1000 Hz_10ms_log_rms_mae_db": 1.1078658035135347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08001307293910878, "300–1000 Hz_10ms_envelope_correlation": 0.9959300416366709, "300–1000 Hz_10ms_modulation_mae_db": 0.42273729831301626, "300–1000 Hz_40ms_log_rms_mae_db": 0.6822236486473815, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050359128491262665, "300–1000 Hz_40ms_envelope_correlation": 0.9985923548813426, "300–1000 Hz_40ms_modulation_mae_db": 0.88512542803482, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2314238353115226, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17317943880477724, "1000–3000 Hz_10ms_envelope_correlation": 0.9751694610758244, "1000–3000 Hz_10ms_modulation_mae_db": 0.4520305632132733, "1000–3000 Hz_40ms_log_rms_mae_db": 0.74105270932346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10708799566975782, "1000–3000 Hz_40ms_envelope_correlation": 0.990901407972782, "1000–3000 Hz_40ms_modulation_mae_db": 0.48624301046801605, "3000–6000 Hz_10ms_log_rms_mae_db": 1.282774063732843, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20053811678258177, "3000–6000 Hz_10ms_envelope_correlation": 0.9783789508147315, "3000–6000 Hz_10ms_modulation_mae_db": 1.4947176728606035, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8303183645474148, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13249765453876705, "3000–6000 Hz_40ms_envelope_correlation": 0.9942334724173403, "3000–6000 Hz_40ms_modulation_mae_db": 1.5642250356750766, "mr_stft_db": 5.3426852737277315, "band_mae_db": 0.37670202871556685, "rms_gain_error_db": -0.19908401486211658, "output_peak": 0.28391931866547787}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1415058955565183, "全帯域_10ms_rms_envelope_nrmse": 0.07065814833531009, "全帯域_10ms_envelope_correlation": 0.9949328692938073, "全帯域_10ms_modulation_mae_db": 0.6511898883102997, "全帯域_40ms_log_rms_mae_db": 0.60536170009252, "全帯域_40ms_rms_envelope_nrmse": 0.03403110598852469, "全帯域_40ms_envelope_correlation": 0.9988182840251404, "全帯域_40ms_modulation_mae_db": 0.6374748258658276, "300–1000 Hz_10ms_log_rms_mae_db": 1.1278681958703507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07223513919036939, "300–1000 Hz_10ms_envelope_correlation": 0.9965557306557881, "300–1000 Hz_10ms_modulation_mae_db": 0.66987328114635, "300–1000 Hz_40ms_log_rms_mae_db": 0.6541092923144307, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04456842275723133, "300–1000 Hz_40ms_envelope_correlation": 0.9987593063146524, "300–1000 Hz_40ms_modulation_mae_db": 0.6588929021676573, "1000–3000 Hz_10ms_log_rms_mae_db": 1.266990017866061, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19247541987280384, "1000–3000 Hz_10ms_envelope_correlation": 0.9685359467128547, "1000–3000 Hz_10ms_modulation_mae_db": 0.7181928529435705, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7904131442474825, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12261169438561616, "1000–3000 Hz_40ms_envelope_correlation": 0.9864537752517394, "1000–3000 Hz_40ms_modulation_mae_db": 0.47057925482641405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2389127733854912, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17431480924587514, "3000–6000 Hz_10ms_envelope_correlation": 0.980804948575261, "3000–6000 Hz_10ms_modulation_mae_db": 0.43191732645470227, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7948740215259592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09257335560665622, "3000–6000 Hz_40ms_envelope_correlation": 0.9950979774936801, "3000–6000 Hz_40ms_modulation_mae_db": 0.6573754583543704, "mr_stft_db": 5.34263076905309, "band_mae_db": 0.16970889326628802, "rms_gain_error_db": -0.07224815930065338, "output_peak": 0.28464781207372775}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59840, "output_samples": 59840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.3750434284745054, "spectral_convergence": 0.28220510904644525, "spectral_convergence_status": "ok", "amplitude_floor": 9.91771656423115e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.125179143208592, "spectral_convergence": 0.24946868740159936, "spectral_convergence_status": "ok", "amplitude_floor": 5.25712991724442e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.623158674259494, "spectral_convergence": 0.2557981512159712, "spectral_convergence_status": "ok", "amplitude_floor": 4.1776466342690114e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00025386719019706026, "output_power": 0.0002634475616569861, "delta_db": 0.16087607038937132, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004894966727833924, "output_power": 0.00048764674984544875, "delta_db": -0.016444101554602568, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005421484972995941, "output_power": 0.0005218260611474572, "delta_db": -0.16592493422515736, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.5605939839496278e-05, "output_power": 1.5295836061854437e-05, "delta_db": -0.08716708125236913, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000107892990107672, "output_power": 9.742476756562187e-05, "delta_db": -0.4432385081235717, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.406384293716839e-05, "output_power": 7.631217296447927e-05, "delta_db": 0.12987577668659986, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.167254412038612, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06229697968879205, "pitch": null, "output_rms_dbfs": -28.350665448554242, "output_peak": 0.2862277975682654, "reference_rms_dbfs": -28.28836846886545, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.374460415314196}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.000110539633783132, "log_rms_mae_db": 1.0886325638202003, "rms_envelope_nrmse": 0.0695318076976679, "envelope_correlation": 0.9950830547615073, "rms_gain_error_db": -0.06230225704490289, "reference_mean_power": 0.00148505750597689, "output_mean_power": 0.0014639055101886317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.0996976338580339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028388367321682654, "output_power": 0.00028393632092846017, "reference_active": true, "delta_db": 0.0008053471504561849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6121837601531416e-05, "output_power": 4.4009990733769856e-05, "reference_active": true, "delta_db": -0.20355325149003134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3216272322760606e-05, "output_power": 1.1975529287371535e-05, "reference_active": true, "delta_db": -0.42814261552748983}], "active_band_count": 3, "active_band_mae_db": 0.21083373805599245}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 0.00010014294879668002, "log_rms_mae_db": 0.5702748877988298, "rms_envelope_nrmse": 0.032920516321403234, "envelope_correlation": 0.9988834833566931, "rms_gain_error_db": -0.06227807206701717, "reference_mean_power": 0.0014970488171277764, "output_mean_power": 0.0014757342445384212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.0259449632693504e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002533237576934496, "output_power": 0.0002545149492098715, "reference_active": true, "delta_db": 0.02037374639634689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.399735365000864e-05, "output_power": 2.2317246966954066e-05, "reference_active": true, "delta_db": -0.31522732330306175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.21988770898296e-07, "output_power": 6.200304639107964e-07, "reference_active": true, "delta_db": -0.013695160259697312}], "active_band_count": 3, "active_band_mae_db": 0.1164320766530353}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.00010594974849403162, "log_rms_mae_db": 1.1795890757061915, "rms_envelope_nrmse": 0.07655669968454638, "envelope_correlation": 0.9961245232025331, "rms_gain_error_db": -0.1417522015454552, "reference_mean_power": 0.0008075829245111437, "output_mean_power": 0.000781649235606248}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.2363198879746262e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030947682521494476, "output_power": 0.00030182307710616466, "reference_active": true, "delta_db": -0.10875690521365858}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.465844834587486e-05, "output_power": 2.9737522733407602e-05, "reference_active": true, "delta_db": -0.6650432845687971}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.693327955984243e-06, "output_power": 5.757106282663904e-06, "reference_active": true, "delta_db": -0.6543785775160327}], "active_band_count": 3, "active_band_mae_db": 0.4760595890994961}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 9.562043532359368e-05, "log_rms_mae_db": 0.6953373263209226, "rms_envelope_nrmse": 0.04086243516777063, "envelope_correlation": 0.9990011890210567, "rms_gain_error_db": -0.1417535188280163, "reference_mean_power": 0.0008141211968758593, "output_mean_power": 0.0007879773072184239}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.204812078547339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002902058778585621, "output_power": 0.0002832471432671758, "reference_active": true, "delta_db": -0.10540666020773538}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7747321249654287e-05, "output_power": 1.5213398428809193e-05, "reference_active": true, "delta_db": -0.6690657136993872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.319074590753967e-07, "output_power": 2.443470092785528e-07, "reference_active": true, "delta_db": -1.3300998478894654}], "active_band_count": 3, "active_band_mae_db": 0.701524073932196}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 2.8704944305868236e-05, "log_rms_mae_db": 1.2229205899939868, "rms_envelope_nrmse": 0.17449730758091275, "envelope_correlation": 0.9745754432779078, "rms_gain_error_db": -0.346467644736819, "reference_mean_power": 4.5961113590202676e-05, "output_mean_power": 4.243691239417397e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 2.8606659711814982e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3098445296769517e-05, "output_power": 1.1438569511156534e-05, "reference_active": true, "delta_db": -0.5884803499267282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2173629500624627e-06, "output_power": 3.2005407755264897e-06, "reference_active": true, "delta_db": -0.022766917729137217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2063471779341703e-06, "output_power": 1.2793845455593417e-06, "reference_active": true, "delta_db": 0.2552878773154335}], "active_band_count": 3, "active_band_mae_db": 0.2888450483237663}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 2.1311353675739154e-05, "log_rms_mae_db": 0.8100977802364481, "rms_envelope_nrmse": 0.12595807381924218, "envelope_correlation": 0.9864976416610243, "rms_gain_error_db": -0.34650205618556945, "reference_mean_power": 4.633078215253765e-05, "output_mean_power": 4.277789659803692e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 2.5755466765658237e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2064714839118615e-05, "output_power": 1.0440505364406726e-05, "reference_active": true, "delta_db": -0.6279554055158209}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.637842198880982e-06, "output_power": 1.5572994648264594e-06, "reference_active": true, "delta_db": -0.2189992213520757}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.209021145285469e-07, "output_power": 1.0208870712505194e-07, "reference_active": true, "delta_db": -0.7345619271535407}], "active_band_count": 3, "active_band_mae_db": 0.5271721846738124}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 6.327125594838897e-05, "log_rms_mae_db": 1.272220571202078, "rms_envelope_nrmse": 0.15712554065308332, "envelope_correlation": 0.9843616067061222, "rms_gain_error_db": -0.27513339990972985, "reference_mean_power": 0.0001246130887459015, "output_mean_power": 0.00011696349065638877}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.481947168400202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.119164755141333e-05, "output_power": 5.596492637731708e-05, "reference_active": true, "delta_db": -0.38776209717721133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1929332977586818e-05, "output_power": 2.2213918162778398e-05, "reference_active": true, "delta_db": 0.055997455862974825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.860956874662985e-06, "output_power": 5.782348216149344e-06, "reference_active": true, "delta_db": -0.7428044795517768}], "active_band_count": 3, "active_band_mae_db": 0.39552134419732093}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 5.46294552803286e-05, "log_rms_mae_db": 0.8330188907653883, "rms_envelope_nrmse": 0.09008568211036899, "envelope_correlation": 0.9950616218139224, "rms_gain_error_db": -0.2750961393834235, "reference_mean_power": 0.00012561601350824365, "output_mean_power": 0.00011790586065756563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.3825123163038373e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.737559737198788e-05, "output_power": 5.317498108819679e-05, "reference_active": true, "delta_db": -0.33019876153965144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1774546568382367e-05, "output_power": 1.1511443313829297e-05, "reference_active": true, "delta_db": -0.09814412442149338}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.943011691356486e-07, "output_power": 5.056644213283968e-07, "reference_active": true, "delta_db": -0.7014418698025966}], "active_band_count": 3, "active_band_mae_db": 0.37659491858791383}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.829991364192937, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4810cf87500fa427aa81f76d7856b60bcce2068d0e122ed8803030c10b051e78", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9729126714658236, "standard_deviation": 0.0023083373997269856, "minimum": 0.9685359467128547, "maximum": 0.9755908464332802, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.229145198275589, "standard_deviation": 0.02460474554362511, "minimum": 1.2025753444642007, "maximum": 1.266990017866061, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5193079575432791, "standard_deviation": 0.2393205181811269, "minimum": 0.1955095362948277, "maximum": 0.9424976408302687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1805458589848403, "standard_deviation": 0.007380914252737354, "minimum": 0.1703820279057029, "maximum": 0.19247541987280384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9876386577436802, "standard_deviation": 0.0014630435167969271, "minimum": 0.9864537752517394, "maximum": 0.990901407972782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7851798696882699, "standard_deviation": 0.03223077740835393, "minimum": 0.7392489604392807, "maximum": 0.8300303374792687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6174401475218153, "standard_deviation": 0.23964962655692523, "minimum": 0.36980119510210363, "maximum": 0.9965232630180029, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12201682791436111, "standard_deviation": 0.007734643661157665, "minimum": 0.10708799566975782, "maximum": 0.13123392947516402, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.978801520977916, "standard_deviation": 0.004934075819020376, "minimum": 0.9691798471266783, "maximum": 0.9843616067061222, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2454676745663313, "standard_deviation": 0.026657366382732646, "minimum": 1.2032792999074138, "maximum": 1.282774063732843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7685498993874424, "standard_deviation": 0.4730054343191741, "minimum": 0.26748140990607455, "maximum": 1.4947176728606035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1850857632022599, "standard_deviation": 0.023103055131093717, "minimum": 0.15712554065308332, "maximum": 0.22844429672637923, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9930911095880067, "standard_deviation": 0.0034206848740556605, "minimum": 0.9873490624315411, "maximum": 0.9955707595407505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8085977142526083, "standard_deviation": 0.04221585955422835, "minimum": 0.7498389418549973, "maximum": 0.8824755732455115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.97241969907084, "standard_deviation": 0.5092203851992766, "minimum": 0.37659491858791383, "maximum": 1.5642250356750766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11116769876510449, "standard_deviation": 0.027551942089167825, "minimum": 0.09008568211036899, "maximum": 0.15804788696105654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.996849870574849, "standard_deviation": 0.0009309991926992299, "minimum": 0.9958717054454878, "maximum": 0.9980860317045912, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1512581682177596, "standard_deviation": 0.028424995787593346, "minimum": 1.1078658035135347, "maximum": 1.1851377149251514, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4473266842249927, "standard_deviation": 0.17394833097693374, "minimum": 0.14167768792977442, "maximum": 0.6757996713881096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06858290636583558, "standard_deviation": 0.011051322189120475, "minimum": 0.053176584922942624, "maximum": 0.0804918119198505, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989951685388335, "standard_deviation": 0.0003985549376629644, "minimum": 0.9983304689806961, "maximum": 0.9994280962281059, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6746510865998671, "standard_deviation": 0.03470396359148309, "minimum": 0.6107489842375123, "maximum": 0.7101203198027988, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6612306635985223, "standard_deviation": 0.23684433596447893, "minimum": 0.34567840784334697, "maximum": 1.072517608094738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04004637365729004, "standard_deviation": 0.008870743860536327, "minimum": 0.029080062500346098, "maximum": 0.05380961413365083, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2161609322343354, "standard_deviation": 0.08357563831143375, "minimum": 0.13292941045303563, "maximum": 0.37670202871556685, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.352058017405186, "standard_deviation": 0.013024997673636067, "minimum": 5.34263076905309, "maximum": 5.374460415314196, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.28768290825585446, "standard_deviation": 0.004192252543946257, "minimum": 0.28173570002347825, "maximum": 0.2931806509389691, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09558329416769737, "standard_deviation": 0.061125487221157605, "minimum": -0.19908401486211658, "maximum": -0.04198065371590691, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9946732164686216, "standard_deviation": 0.0008882593406868972, "minimum": 0.9930449965493563, "maximum": 0.9958020040898727, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1110868272182493, "standard_deviation": 0.044955349276323775, "minimum": 1.0364128012098222, "maximum": 1.1557716048221423, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4003516328696741, "standard_deviation": 0.2752489575613182, "minimum": 0.08033308580779336, "maximum": 0.8487456776945557, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07273486553608877, "standard_deviation": 0.006653326909605243, "minimum": 0.06433529090357751, "maximum": 0.08427507675156251, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985153445720818, "standard_deviation": 0.0005073218357666525, "minimum": 0.9974497107145617, "maximum": 0.9990045822126268, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6214612918719246, "standard_deviation": 0.07709736764903086, "minimum": 0.5213145994343756, "maximum": 0.7278790143066315, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5018658787855463, "standard_deviation": 0.299680710295354, "minimum": 0.09129350845125073, "maximum": 0.9685118192251548, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03849460032554708, "standard_deviation": 0.0073221157129321115, "minimum": 0.031225385230019546, "maximum": 0.05239838461011472, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0044", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.1010258197784424, "predicted_mos": 3.1010258197784424, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "c708c18116955dedd25bbec877e6149197242002011e4aef7247dbff5c9877de", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.921617299783975, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0738340862649929, "全帯域_10ms_rms_envelope_nrmse": 0.07479130043177419, "全帯域_10ms_envelope_correlation": 0.9943043269308764, "全帯域_10ms_modulation_mae_db": 0.2616117380940381, "全帯域_40ms_log_rms_mae_db": 0.5679763209136658, "全帯域_40ms_rms_envelope_nrmse": 0.03971428926953835, "全帯域_40ms_envelope_correlation": 0.998333191171471, "全帯域_40ms_modulation_mae_db": 0.47858931474289657, "300–1000 Hz_10ms_log_rms_mae_db": 1.3176332140887121, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07584440672807806, "300–1000 Hz_10ms_envelope_correlation": 0.9960800148729603, "300–1000 Hz_10ms_modulation_mae_db": 0.3028747986648908, "300–1000 Hz_40ms_log_rms_mae_db": 0.8460229603740039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04387878326538493, "300–1000 Hz_40ms_envelope_correlation": 0.9986663997795345, "300–1000 Hz_40ms_modulation_mae_db": 0.19442299370022476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3881981288404865, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20249854370248327, "1000–3000 Hz_10ms_envelope_correlation": 0.9649321349021118, "1000–3000 Hz_10ms_modulation_mae_db": 0.5823176667230338, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8888676413813553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13146095270384153, "1000–3000 Hz_40ms_envelope_correlation": 0.9844065006550352, "1000–3000 Hz_40ms_modulation_mae_db": 0.33220056323112673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4321293405544735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18172178545146275, "3000–6000 Hz_10ms_envelope_correlation": 0.9790231683200247, "3000–6000 Hz_10ms_modulation_mae_db": 1.5463486378938447, "3000–6000 Hz_40ms_log_rms_mae_db": 1.004424709869887, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11043690038394142, "3000–6000 Hz_40ms_envelope_correlation": 0.992563598646436, "3000–6000 Hz_40ms_modulation_mae_db": 1.9196034898062593, "mr_stft_db": 5.465124544124525, "band_mae_db": 0.10426208909478552, "rms_gain_error_db": -0.04826631952755105, "output_peak": 0.3051923604250357}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0192476809511495, "全帯域_10ms_rms_envelope_nrmse": 0.0719523121602392, "全帯域_10ms_envelope_correlation": 0.9948022817253931, "全帯域_10ms_modulation_mae_db": 0.7921460806675563, "全帯域_40ms_log_rms_mae_db": 0.5615251066977706, "全帯域_40ms_rms_envelope_nrmse": 0.0352186659530818, "全帯域_40ms_envelope_correlation": 0.9988081977206781, "全帯域_40ms_modulation_mae_db": 0.7309038513308589, "300–1000 Hz_10ms_log_rms_mae_db": 1.3558968486182514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07318271570047724, "300–1000 Hz_10ms_envelope_correlation": 0.9964429254667219, "300–1000 Hz_10ms_modulation_mae_db": 0.6806065349223821, "300–1000 Hz_40ms_log_rms_mae_db": 0.9146215120390918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038370519045776964, "300–1000 Hz_40ms_envelope_correlation": 0.9991044616184186, "300–1000 Hz_40ms_modulation_mae_db": 0.6554693634855284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4299240634757442, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18563795865670496, "1000–3000 Hz_10ms_envelope_correlation": 0.9706158168700522, "1000–3000 Hz_10ms_modulation_mae_db": 0.7607625894829976, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9913812399534819, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13619705059699166, "1000–3000 Hz_40ms_envelope_correlation": 0.9831478371432504, "1000–3000 Hz_40ms_modulation_mae_db": 0.3295149319627256, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3716325157443385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17283363260473608, "3000–6000 Hz_10ms_envelope_correlation": 0.9809927781952122, "3000–6000 Hz_10ms_modulation_mae_db": 0.6435471904979487, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9647732170864853, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09161273306710348, "3000–6000 Hz_40ms_envelope_correlation": 0.9950134312751917, "3000–6000 Hz_40ms_modulation_mae_db": 0.8856469148596159, "mr_stft_db": 5.462918295694678, "band_mae_db": 0.1415006842000973, "rms_gain_error_db": -0.10542642667911108, "output_peak": 0.2917216819867174}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9835045156649458, "全帯域_10ms_rms_envelope_nrmse": 0.06814272742100148, "全帯域_10ms_envelope_correlation": 0.9952652619269983, "全帯域_10ms_modulation_mae_db": 0.7262526887379197, "全帯域_40ms_log_rms_mae_db": 0.5563837992243573, "全帯域_40ms_rms_envelope_nrmse": 0.03767914235117917, "全帯域_40ms_envelope_correlation": 0.9984725789087868, "全帯域_40ms_modulation_mae_db": 0.5539048839635652, "300–1000 Hz_10ms_log_rms_mae_db": 1.3019703458178353, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07305620177647651, "300–1000 Hz_10ms_envelope_correlation": 0.9964380273966714, "300–1000 Hz_10ms_modulation_mae_db": 0.5769051660868614, "300–1000 Hz_40ms_log_rms_mae_db": 0.878094904849176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03721707912620042, "300–1000 Hz_40ms_envelope_correlation": 0.9991452361101838, "300–1000 Hz_40ms_modulation_mae_db": 0.32214797847670706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3962976025772804, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18337704584646916, "1000–3000 Hz_10ms_envelope_correlation": 0.9713815349955421, "1000–3000 Hz_10ms_modulation_mae_db": 0.5253062729501475, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9509075842893218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12397743864646724, "1000–3000 Hz_40ms_envelope_correlation": 0.9860384831482396, "1000–3000 Hz_40ms_modulation_mae_db": 0.486240645037705, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4126966595462955, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18322655902726542, "3000–6000 Hz_10ms_envelope_correlation": 0.9820148454356931, "3000–6000 Hz_10ms_modulation_mae_db": 1.259805619022421, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0053288058446188, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12098512954937188, "3000–6000 Hz_40ms_envelope_correlation": 0.9931792315387458, "3000–6000 Hz_40ms_modulation_mae_db": 1.329213454125372, "mr_stft_db": 5.463249253432338, "band_mae_db": 0.24313442805932217, "rms_gain_error_db": 0.00047749363215513824, "output_peak": 0.2946576392495131}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0707388968032834, "全帯域_10ms_rms_envelope_nrmse": 0.06919138735341303, "全帯域_10ms_envelope_correlation": 0.9951272285371491, "全帯域_10ms_modulation_mae_db": 0.5042270277914412, "全帯域_40ms_log_rms_mae_db": 0.6043972612247446, "全帯域_40ms_rms_envelope_nrmse": 0.036675054263103246, "全帯域_40ms_envelope_correlation": 0.9985810281783671, "全帯域_40ms_modulation_mae_db": 0.650276842046461, "300–1000 Hz_10ms_log_rms_mae_db": 1.3864598208519923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06871538406416162, "300–1000 Hz_10ms_envelope_correlation": 0.9968980370677123, "300–1000 Hz_10ms_modulation_mae_db": 1.0076494666665967, "300–1000 Hz_40ms_log_rms_mae_db": 0.887804903577271, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04102725279900558, "300–1000 Hz_40ms_envelope_correlation": 0.9989735697789326, "300–1000 Hz_40ms_modulation_mae_db": 1.0601226655709455, "1000–3000 Hz_10ms_log_rms_mae_db": 1.459100081839449, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19573982283701086, "1000–3000 Hz_10ms_envelope_correlation": 0.9673739192101525, "1000–3000 Hz_10ms_modulation_mae_db": 0.6967926207547869, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9909085304657869, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13564755297577966, "1000–3000 Hz_40ms_envelope_correlation": 0.9837077575286827, "1000–3000 Hz_40ms_modulation_mae_db": 0.2891541191066147, "3000–6000 Hz_10ms_log_rms_mae_db": 1.468329679806729, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16056937960718232, "3000–6000 Hz_10ms_envelope_correlation": 0.9836985572736544, "3000–6000 Hz_10ms_modulation_mae_db": 0.47813832025666, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0363385636694682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08674779701385156, "3000–6000 Hz_40ms_envelope_correlation": 0.9950876384780335, "3000–6000 Hz_40ms_modulation_mae_db": 0.48458819598083397, "mr_stft_db": 5.472284273184061, "band_mae_db": 0.1311685295702351, "rms_gain_error_db": -0.050467934734290704, "output_peak": 0.2921083256165085}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0428417835102144, "全帯域_10ms_rms_envelope_nrmse": 0.0750751811774666, "全帯域_10ms_envelope_correlation": 0.9942576186225072, "全帯域_10ms_modulation_mae_db": 0.3087974446654142, "全帯域_40ms_log_rms_mae_db": 0.5134577174737807, "全帯域_40ms_rms_envelope_nrmse": 0.04400567267734704, "全帯域_40ms_envelope_correlation": 0.9979601226183137, "全帯域_40ms_modulation_mae_db": 0.28428498276032393, "300–1000 Hz_10ms_log_rms_mae_db": 1.3246236231787882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07947292215313809, "300–1000 Hz_10ms_envelope_correlation": 0.99584397799801, "300–1000 Hz_10ms_modulation_mae_db": 0.7137539660951245, "300–1000 Hz_40ms_log_rms_mae_db": 0.9117209197511904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05392788411893555, "300–1000 Hz_40ms_envelope_correlation": 0.998163016368309, "300–1000 Hz_40ms_modulation_mae_db": 0.9549582090850915, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3897300737683647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17970165714899122, "1000–3000 Hz_10ms_envelope_correlation": 0.9729515649310226, "1000–3000 Hz_10ms_modulation_mae_db": 0.5771436330195098, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9602288057892766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12762103495804691, "1000–3000 Hz_40ms_envelope_correlation": 0.9861813600140366, "1000–3000 Hz_40ms_modulation_mae_db": 0.579128452298022, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4372809698244862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18458352870248027, "3000–6000 Hz_10ms_envelope_correlation": 0.9782335077784197, "3000–6000 Hz_10ms_modulation_mae_db": 0.14443864744272492, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9754416681751467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10946012591466522, "3000–6000 Hz_40ms_envelope_correlation": 0.9920076461012822, "3000–6000 Hz_40ms_modulation_mae_db": 0.16035326288170892, "mr_stft_db": 5.470413193314514, "band_mae_db": 0.20932966534289466, "rms_gain_error_db": -0.06059339117317968, "output_peak": 0.2799768080005254}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0433668316489055, "全帯域_10ms_rms_envelope_nrmse": 0.07057265073856019, "全帯域_10ms_envelope_correlation": 0.9949962631810371, "全帯域_10ms_modulation_mae_db": 0.19064694718871575, "全帯域_40ms_log_rms_mae_db": 0.5241292249134225, "全帯域_40ms_rms_envelope_nrmse": 0.04179811173490439, "全帯域_40ms_envelope_correlation": 0.9982440994803965, "全帯域_40ms_modulation_mae_db": 0.19022801984159532, "300–1000 Hz_10ms_log_rms_mae_db": 1.3810082714649152, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07101528503710158, "300–1000 Hz_10ms_envelope_correlation": 0.996694643635777, "300–1000 Hz_10ms_modulation_mae_db": 0.3519412337027228, "300–1000 Hz_40ms_log_rms_mae_db": 0.9455654396431166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04337075896459628, "300–1000 Hz_40ms_envelope_correlation": 0.9988549500767377, "300–1000 Hz_40ms_modulation_mae_db": 0.25112912822993566, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3459199958493382, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17490292235237923, "1000–3000 Hz_10ms_envelope_correlation": 0.9739895811140762, "1000–3000 Hz_10ms_modulation_mae_db": 0.7284464864222602, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8735645854490265, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12145465245694198, "1000–3000 Hz_40ms_envelope_correlation": 0.986637506513916, "1000–3000 Hz_40ms_modulation_mae_db": 0.262984281366305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.462369089274626, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.185599053074946, "3000–6000 Hz_10ms_envelope_correlation": 0.9777537827991165, "3000–6000 Hz_10ms_modulation_mae_db": 0.4426969742368765, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0405554534706911, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1146322014806278, "3000–6000 Hz_40ms_envelope_correlation": 0.9912224122038071, "3000–6000 Hz_40ms_modulation_mae_db": 0.7710368840491357, "mr_stft_db": 5.456361110206527, "band_mae_db": 0.20901352120077177, "rms_gain_error_db": -0.1012357380361564, "output_peak": 0.2836421191262799}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9673905433995369, "全帯域_10ms_rms_envelope_nrmse": 0.06619165273804127, "全帯域_10ms_envelope_correlation": 0.9955251397954715, "全帯域_10ms_modulation_mae_db": 0.19396529793066808, "全帯域_40ms_log_rms_mae_db": 0.5130545636083027, "全帯域_40ms_rms_envelope_nrmse": 0.033961036864411105, "全帯域_40ms_envelope_correlation": 0.9987678271650159, "全帯域_40ms_modulation_mae_db": 0.22021243107335656, "300–1000 Hz_10ms_log_rms_mae_db": 1.3399115516729816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0637506541089581, "300–1000 Hz_10ms_envelope_correlation": 0.9972864792603742, "300–1000 Hz_10ms_modulation_mae_db": 0.28843011824131753, "300–1000 Hz_40ms_log_rms_mae_db": 0.8797663977641355, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03330760574499997, "300–1000 Hz_40ms_envelope_correlation": 0.9993062017517714, "300–1000 Hz_40ms_modulation_mae_db": 0.5076176773384106, "1000–3000 Hz_10ms_log_rms_mae_db": 1.395245206469878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18954856507314552, "1000–3000 Hz_10ms_envelope_correlation": 0.9694907328782011, "1000–3000 Hz_10ms_modulation_mae_db": 0.46481380250662957, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9086129613949234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13394215375617666, "1000–3000 Hz_40ms_envelope_correlation": 0.9836252678679003, "1000–3000 Hz_40ms_modulation_mae_db": 0.4552530805172383, "3000–6000 Hz_10ms_log_rms_mae_db": 1.449113527623722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1397539696573783, "3000–6000 Hz_10ms_envelope_correlation": 0.987458289472793, "3000–6000 Hz_10ms_modulation_mae_db": 0.3830157766693045, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9848741827414798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08280706093657132, "3000–6000 Hz_40ms_envelope_correlation": 0.995476033713012, "3000–6000 Hz_40ms_modulation_mae_db": 0.5347293522688147, "mr_stft_db": 5.4536503121627105, "band_mae_db": 0.11091649049616302, "rms_gain_error_db": -0.02236312360966078, "output_peak": 0.2872916951648806}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0265507063219588, "全帯域_10ms_rms_envelope_nrmse": 0.0671274698922161, "全帯域_10ms_envelope_correlation": 0.9955516766373314, "全帯域_10ms_modulation_mae_db": 0.32620106060286264, "全帯域_40ms_log_rms_mae_db": 0.536143599118898, "全帯域_40ms_rms_envelope_nrmse": 0.034315679340637284, "全帯域_40ms_envelope_correlation": 0.9989447459175651, "全帯域_40ms_modulation_mae_db": 0.509258132885113, "300–1000 Hz_10ms_log_rms_mae_db": 1.260797806444134, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06536372810640079, "300–1000 Hz_10ms_envelope_correlation": 0.9973035600663811, "300–1000 Hz_10ms_modulation_mae_db": 0.3497416039105483, "300–1000 Hz_40ms_log_rms_mae_db": 0.7825519379256581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039724464074367576, "300–1000 Hz_40ms_envelope_correlation": 0.9991531087387941, "300–1000 Hz_40ms_modulation_mae_db": 0.4389712333602682, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3926395319593046, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18850114286849662, "1000–3000 Hz_10ms_envelope_correlation": 0.969759054391077, "1000–3000 Hz_10ms_modulation_mae_db": 0.9297027904262983, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9294816529293732, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12994758597751485, "1000–3000 Hz_40ms_envelope_correlation": 0.9846003252625757, "1000–3000 Hz_40ms_modulation_mae_db": 0.4821374931659765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3933205878802903, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16162986731119128, "3000–6000 Hz_10ms_envelope_correlation": 0.984243901299254, "3000–6000 Hz_10ms_modulation_mae_db": 1.00027166932896, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9475211977241434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09617557665430362, "3000–6000 Hz_40ms_envelope_correlation": 0.9954251647856086, "3000–6000 Hz_40ms_modulation_mae_db": 1.1799875544472875, "mr_stft_db": 5.456003995186034, "band_mae_db": 0.21593780374550178, "rms_gain_error_db": -0.12674431975796982, "output_peak": 0.2941378592565553}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59840, "output_samples": 59840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.522197786653836, "spectral_convergence": 0.27404907817396096, "spectral_convergence_status": "ok", "amplitude_floor": 9.91771656423115e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.214976247644681, "spectral_convergence": 0.2531255457040616, "spectral_convergence_status": "ok", "amplitude_floor": 5.25712991724442e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.658199598075061, "spectral_convergence": 0.2601553478585203, "spectral_convergence_status": "ok", "amplitude_floor": 4.1776466342690114e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00025386719019706026, "output_power": 0.00025394582317242155, "delta_db": 0.0013449780399047036, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004894966727833924, "output_power": 0.00048330983211131954, "delta_db": -0.05524113947361058, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005421484972995941, "output_power": 0.0005427629371012558, "delta_db": 0.004919255810064768, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.5605939839496278e-05, "output_power": 1.585624404922106e-05, "delta_db": 0.06910393188529594, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000107892990107672, "output_power": 9.819942478434782e-05, "delta_db": -0.40884285286727395, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.406384293716839e-05, "output_power": 7.260962683975473e-05, "delta_db": -0.08612037649256307, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10426208909478552, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04826631952755105, "pitch": null, "output_rms_dbfs": -28.336634788393, "output_peak": 0.3051923604250357, "reference_rms_dbfs": -28.28836846886545, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.465124544124525}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.000110539633783132, "log_rms_mae_db": 1.0738340862649929, "rms_envelope_nrmse": 0.07479130043177419, "envelope_correlation": 0.9943043269308764, "rms_gain_error_db": -0.04826971959626662, "reference_mean_power": 0.00148505750597689, "output_mean_power": 0.0014686432014936882}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.0996976338580339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028388367321682654, "output_power": 0.0002808781433915212, "reference_active": true, "delta_db": -0.046224702256980235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6121837601531416e-05, "output_power": 4.4264540725426174e-05, "reference_active": true, "delta_db": -0.17850639792196382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3216272322760606e-05, "output_power": 1.1617122457822678e-05, "reference_active": true, "delta_db": -0.5601041141031703}], "active_band_count": 3, "active_band_mae_db": 0.2616117380940381}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 0.00010014294879668002, "log_rms_mae_db": 0.5679763209136658, "rms_envelope_nrmse": 0.03971428926953835, "envelope_correlation": 0.998333191171471, "rms_gain_error_db": -0.048286011255491014, "reference_mean_power": 0.0014970488171277764, "output_mean_power": 0.0014804964191876455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.0259449632693504e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002533237576934496, "output_power": 0.0002507355170239541, "reference_active": true, "delta_db": -0.04460064822355097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.399735365000864e-05, "output_power": 2.2734807737821e-05, "reference_active": true, "delta_db": -0.23472065926781976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.21988770898296e-07, "output_power": 4.765822689196195e-07, "reference_active": true, "delta_db": -1.156446636737319}], "active_band_count": 3, "active_band_mae_db": 0.47858931474289657}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.00010594974849403162, "log_rms_mae_db": 1.3176332140887121, "rms_envelope_nrmse": 0.07584440672807806, "envelope_correlation": 0.9960800148729603, "rms_gain_error_db": -0.042359558384976355, "reference_mean_power": 0.0008075829245111437, "output_mean_power": 0.0007997443338867972}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.2363198879746262e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030947682521494476, "output_power": 0.0003103518546177126, "reference_active": true, "delta_db": 0.012262120355101753}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.465844834587486e-05, "output_power": 3.216489622051163e-05, "reference_active": true, "delta_db": -0.32426961011982147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.693327955984243e-06, "output_power": 7.635744207802556e-06, "reference_active": true, "delta_db": 0.5720926655197491}], "active_band_count": 3, "active_band_mae_db": 0.3028747986648908}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 9.562043532359368e-05, "log_rms_mae_db": 0.8460229603740039, "rms_envelope_nrmse": 0.04387878326538493, "envelope_correlation": 0.9986663997795345, "rms_gain_error_db": -0.04236008178658325, "reference_mean_power": 0.0008141211968758593, "output_mean_power": 0.0008062190470718866}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.204812078547339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002902058778585621, "output_power": 0.00029120095372273824, "reference_active": true, "delta_db": 0.01486588582635954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7747321249654287e-05, "output_power": 1.6508872914766534e-05, "reference_active": true, "delta_db": -0.3141538632408341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.319074590753967e-07, "output_power": 3.130344170679392e-07, "reference_active": true, "delta_db": -0.2542492320334806}], "active_band_count": 3, "active_band_mae_db": 0.19442299370022476}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 2.8704944305868236e-05, "log_rms_mae_db": 1.3881981288404865, "rms_envelope_nrmse": 0.20249854370248327, "envelope_correlation": 0.9649321349021118, "rms_gain_error_db": -0.17407697025926477, "reference_mean_power": 4.5961113590202676e-05, "output_mean_power": 4.415530059249302e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 2.8606659711814982e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3098445296769517e-05, "output_power": 1.2071995304765484e-05, "reference_active": true, "delta_db": -0.3544069283126915}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2173629500624627e-06, "output_power": 3.0941061713503292e-06, "reference_active": true, "delta_db": -0.16964844311386837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2063471779341703e-06, "output_power": 1.5986819643283627e-06, "reference_active": true, "delta_db": 1.2228976287425413}], "active_band_count": 3, "active_band_mae_db": 0.5823176667230338}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 2.1311353675739154e-05, "log_rms_mae_db": 0.8888676413813553, "rms_envelope_nrmse": 0.13146095270384153, "envelope_correlation": 0.9844065006550352, "rms_gain_error_db": -0.17411976792994124, "reference_mean_power": 4.633078215253765e-05, "output_mean_power": 4.4510006242962045e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 2.5755466765658237e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2064714839118615e-05, "output_power": 1.1283289740178405e-05, "reference_active": true, "delta_db": -0.2908132100594561}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.637842198880982e-06, "output_power": 1.556756259081047e-06, "reference_active": true, "delta_db": -0.22051435968664318}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.209021145285469e-07, "output_power": 1.0812011141727508e-07, "reference_active": true, "delta_db": -0.4852741199472809}], "active_band_count": 3, "active_band_mae_db": 0.33220056323112673}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 6.327125594838897e-05, "log_rms_mae_db": 1.4321293405544735, "rms_envelope_nrmse": 0.18172178545146275, "envelope_correlation": 0.9790231683200247, "rms_gain_error_db": -0.3531822043475179, "reference_mean_power": 0.0001246130887459015, "output_mean_power": 0.00011488026810348006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.481947168400202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.119164755141333e-05, "output_power": 5.679222460608282e-05, "reference_active": true, "delta_db": -0.3240326574445393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1929332977586818e-05, "output_power": 1.626864314655677e-05, "reference_active": true, "delta_db": -1.296740889949536}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.860956874662985e-06, "output_power": 3.424186832618522e-06, "reference_active": true, "delta_db": -3.0182723662874587}], "active_band_count": 3, "active_band_mae_db": 1.5463486378938447}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 5.46294552803286e-05, "log_rms_mae_db": 1.004424709869887, "rms_envelope_nrmse": 0.11043690038394142, "envelope_correlation": 0.992563598646436, "rms_gain_error_db": -0.3531988497446318, "reference_mean_power": 0.00012561601350824365, "output_mean_power": 0.00011580441625901179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.3825123163038373e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.737559737198788e-05, "output_power": 5.203922727064191e-05, "reference_active": true, "delta_db": -0.42396381162860464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1774546568382367e-05, "output_power": 8.210215039310171e-06, "reference_active": true, "delta_db": -1.5658965950033765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.943011691356486e-07, "output_power": 2.495235717142838e-07, "reference_active": true, "delta_db": -3.7689500627867965}], "active_band_count": 3, "active_band_mae_db": 1.9196034898062593}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8218131314279403, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c6f118f5c4012d5408bcb7871abe5e50e7d6e966d4bdd648b1cd25337e00ecdd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9700617924115295, "standard_deviation": 0.002922395805431061, "minimum": 0.9649321349021118, "maximum": 0.9739895811140762, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3996318355974808, "standard_deviation": 0.033052079214225843, "minimum": 1.3459199958493382, "maximum": 1.459100081839449, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.658160732785708, "standard_deviation": 0.1502588357604423, "minimum": 0.46481380250662957, "maximum": 0.9297027904262983, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18748845731071012, "standard_deviation": 0.008766945216563626, "minimum": 0.17490292235237923, "maximum": 0.20249854370248327, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9847931297667045, "standard_deviation": 0.0013264017719004891, "minimum": 0.9831478371432504, "maximum": 0.986637506513916, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9367441252065682, "standard_deviation": 0.04439924183494941, "minimum": 0.8735645854490265, "maximum": 0.9913812399534819, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4020766958357142, "standard_deviation": 0.11333543485148155, "minimum": 0.262984281366305, "maximum": 0.579128452298022, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1300310527589701, "standard_deviation": 0.005384129686332082, "minimum": 0.12145465245694198, "maximum": 0.13619705059699166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9816773538217709, "standard_deviation": 0.0033588555990116654, "minimum": 0.9777537827991165, "maximum": 0.987458289472793, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4283590462818703, "standard_deviation": 0.03374230514472616, "minimum": 1.3716325157443385, "maximum": 1.468329679806729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7372828544185925, "standard_deviation": 0.48350995254767165, "minimum": 0.14443864744272492, "maximum": 1.5463486378938447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17123972192958029, "standard_deviation": 0.01620363340148728, "minimum": 0.1397539696573783, "maximum": 0.185599053074946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937468945927647, "standard_deviation": 0.0017041398985979408, "minimum": 0.9912224122038071, "maximum": 0.995476033713012, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9949072248227401, "standard_deviation": 0.033020712775444605, "minimum": 0.9475211977241434, "maximum": 1.0405554534706911, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9081448885523785, "standard_deviation": 0.5569793773647558, "minimum": 0.16035326288170892, "maximum": 1.9196034898062593, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10160719062505455, "standard_deviation": 0.014083556058816352, "minimum": 0.08280706093657132, "maximum": 0.12098512954937188, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9966234582205761, "standard_deviation": 0.0005281359513315157, "minimum": 0.99584397799801, "maximum": 0.9973035600663811, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3335376852672014, "standard_deviation": 0.04177279864254923, "minimum": 1.260797806444134, "maximum": 1.3864598208519923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5339878610363056, "standard_deviation": 0.2565734676781804, "minimum": 0.28843011824131753, "maximum": 1.0076494666665967, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07130016220934901, "standard_deviation": 0.005251807871481692, "minimum": 0.0637506541089581, "maximum": 0.07947292215313809, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989208680278352, "standard_deviation": 0.0003650208110691156, "minimum": 0.998163016368309, "maximum": 0.9993062017517714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8807686219904555, "standard_deviation": 0.0495899341519973, "minimum": 0.7825519379256581, "maximum": 0.9455654396431166, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5481049061558889, "standard_deviation": 0.32006926438616845, "minimum": 0.19442299370022476, "maximum": 1.0601226655709455, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04135304339240841, "standard_deviation": 0.006118663325871503, "minimum": 0.03330760574499997, "maximum": 0.05392788411893555, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1706579014637214, "standard_deviation": 0.05432280399949919, "minimum": 0.10426208909478552, "maximum": 0.24313442805932217, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.462500622163174, "standard_deviation": 0.006805210526701532, "minimum": 5.4536503121627105, "maximum": 5.472284273184061, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.291091061103252, "standard_deviation": 0.0077065014403694355, "minimum": 0.2799768080005254, "maximum": 0.3051923604250357, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06432746998572056, "standard_deviation": 0.0437066322517819, "minimum": -0.12674431975796982, "maximum": 0.00047749363215513824, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9949787246695955, "standard_deviation": 0.0004981747253375353, "minimum": 0.9942576186225072, "maximum": 0.9955516766373314, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0284343805706233, "standard_deviation": 0.03801192147785419, "minimum": 0.9673905433995369, "maximum": 1.0738340862649929, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.412981035709827, "standard_deviation": 0.2357144399105298, "minimum": 0.19064694718871575, "maximum": 0.7921460806675563, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07038058523908901, "standard_deviation": 0.0033499033757928096, "minimum": 0.06619165273804127, "maximum": 0.0750751811774666, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985139738950743, "standard_deviation": 0.0003284268409493763, "minimum": 0.9979601226183137, "maximum": 0.9989447459175651, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5471334491468678, "standard_deviation": 0.03151338658929481, "minimum": 0.5130545636083027, "maximum": 0.6043972612247446, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4522073073305213, "standard_deviation": 0.20065933643749942, "minimum": 0.19022801984159532, "maximum": 0.7309038513308589, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0379209565567753, "standard_deviation": 0.0036422962474496013, "minimum": 0.033961036864411105, "maximum": 0.04400567267734704, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0044", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.3585760593414307, "predicted_mos": 3.3585760593414307, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "a17e95fa68c7400c327402499fce232c7958e5201eeb6bc5bf2a7c6c6cf5eac3", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7482255999930203, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0478852336515354, "全帯域_10ms_rms_envelope_nrmse": 0.0754472730241431, "全帯域_10ms_envelope_correlation": 0.99425628591946, "全帯域_10ms_modulation_mae_db": 0.7584771898937547, "全帯域_40ms_log_rms_mae_db": 0.6426331207619295, "全帯域_40ms_rms_envelope_nrmse": 0.041847002531518775, "全帯域_40ms_envelope_correlation": 0.9982347134163458, "全帯域_40ms_modulation_mae_db": 0.4984110858231607, "300–1000 Hz_10ms_log_rms_mae_db": 1.348143011878034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08477873760244115, "300–1000 Hz_10ms_envelope_correlation": 0.9951907318686366, "300–1000 Hz_10ms_modulation_mae_db": 0.33709298148169653, "300–1000 Hz_40ms_log_rms_mae_db": 0.834665370583464, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053377904903860615, "300–1000 Hz_40ms_envelope_correlation": 0.9981459212018771, "300–1000 Hz_40ms_modulation_mae_db": 0.28657697976223934, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3122883868005197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1715340891336996, "1000–3000 Hz_10ms_envelope_correlation": 0.9750680271938182, "1000–3000 Hz_10ms_modulation_mae_db": 0.2475958165234636, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8582861122509019, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12282917945627113, "1000–3000 Hz_40ms_envelope_correlation": 0.986599112899363, "1000–3000 Hz_40ms_modulation_mae_db": 0.6093990079000196, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4275205146609493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1855535539570722, "3000–6000 Hz_10ms_envelope_correlation": 0.9785828872657031, "3000–6000 Hz_10ms_modulation_mae_db": 1.5290643750555306, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9655224363274345, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1092886083194496, "3000–6000 Hz_40ms_envelope_correlation": 0.9934755011690591, "3000–6000 Hz_40ms_modulation_mae_db": 1.686841805627715, "mr_stft_db": 5.428020548712106, "band_mae_db": 0.1882212493483533, "rms_gain_error_db": -0.0992686828982936, "output_peak": 0.289354488072027}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.02940533147346, "全帯域_10ms_rms_envelope_nrmse": 0.06439550063231975, "全帯域_10ms_envelope_correlation": 0.9957934372957644, "全帯域_10ms_modulation_mae_db": 0.3756003614844254, "全帯域_40ms_log_rms_mae_db": 0.5434919710505332, "全帯域_40ms_rms_envelope_nrmse": 0.02984204400871222, "全帯域_40ms_envelope_correlation": 0.9991002367277562, "全帯域_40ms_modulation_mae_db": 0.3574313578583399, "300–1000 Hz_10ms_log_rms_mae_db": 1.2664889309482195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05700787581588119, "300–1000 Hz_10ms_envelope_correlation": 0.9978641823027969, "300–1000 Hz_10ms_modulation_mae_db": 0.8993988855214803, "300–1000 Hz_40ms_log_rms_mae_db": 0.7618878472139144, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03186601993617399, "300–1000 Hz_40ms_envelope_correlation": 0.9993888407243281, "300–1000 Hz_40ms_modulation_mae_db": 0.6378889154838863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.374380262196956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21420527163156927, "1000–3000 Hz_10ms_envelope_correlation": 0.9607829739376138, "1000–3000 Hz_10ms_modulation_mae_db": 1.2201695373809458, "1000–3000 Hz_40ms_log_rms_mae_db": 0.878523417361393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14641286021481695, "1000–3000 Hz_40ms_envelope_correlation": 0.9806667114317043, "1000–3000 Hz_40ms_modulation_mae_db": 0.7765861583186419, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3784182656804091, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1691460513324396, "3000–6000 Hz_10ms_envelope_correlation": 0.981663655361272, "3000–6000 Hz_10ms_modulation_mae_db": 0.9783177537800803, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9478324398298332, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09301496366111857, "3000–6000 Hz_40ms_envelope_correlation": 0.9945465711759054, "3000–6000 Hz_40ms_modulation_mae_db": 0.9412685432669986, "mr_stft_db": 5.426860758324721, "band_mae_db": 0.14160567731674442, "rms_gain_error_db": -0.06162231849176288, "output_peak": 0.2916320775886155}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0068128431158243, "全帯域_10ms_rms_envelope_nrmse": 0.06775986220666529, "全帯域_10ms_envelope_correlation": 0.9953452892805563, "全帯域_10ms_modulation_mae_db": 0.3781322997006038, "全帯域_40ms_log_rms_mae_db": 0.5544649319676364, "全帯域_40ms_rms_envelope_nrmse": 0.035742994697140706, "全帯域_40ms_envelope_correlation": 0.9986713936516517, "全帯域_40ms_modulation_mae_db": 0.5860876465385944, "300–1000 Hz_10ms_log_rms_mae_db": 1.268805360396642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05909949201548391, "300–1000 Hz_10ms_envelope_correlation": 0.9976296425711073, "300–1000 Hz_10ms_modulation_mae_db": 0.44853042746989136, "300–1000 Hz_40ms_log_rms_mae_db": 0.796574019118227, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0338601196033563, "300–1000 Hz_40ms_envelope_correlation": 0.999216182580869, "300–1000 Hz_40ms_modulation_mae_db": 0.42708082624979843, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3036259534245025, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18560294032085964, "1000–3000 Hz_10ms_envelope_correlation": 0.9707497320362943, "1000–3000 Hz_10ms_modulation_mae_db": 0.6984342932259394, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8578929526588315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13372425533934762, "1000–3000 Hz_40ms_envelope_correlation": 0.9840441066508359, "1000–3000 Hz_40ms_modulation_mae_db": 0.33967006891765245, "3000–6000 Hz_10ms_log_rms_mae_db": 1.339561047038194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18676274969024387, "3000–6000 Hz_10ms_envelope_correlation": 0.9777061100240269, "3000–6000 Hz_10ms_modulation_mae_db": 0.9389135447270882, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9036614811096881, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.109977904801838, "3000–6000 Hz_40ms_envelope_correlation": 0.992486635341821, "3000–6000 Hz_40ms_modulation_mae_db": 1.2925655318670646, "mr_stft_db": 5.406249322391712, "band_mae_db": 0.2623473246521039, "rms_gain_error_db": -0.057818366419131836, "output_peak": 0.28173349697433375}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9879539135082759, "全帯域_10ms_rms_envelope_nrmse": 0.07455851490260927, "全帯域_10ms_envelope_correlation": 0.9944840078213713, "全帯域_10ms_modulation_mae_db": 0.7454420169501116, "全帯域_40ms_log_rms_mae_db": 0.5147139921994948, "全帯域_40ms_rms_envelope_nrmse": 0.03801418438922815, "全帯域_40ms_envelope_correlation": 0.9986844769165424, "全帯域_40ms_modulation_mae_db": 0.5222747089085312, "300–1000 Hz_10ms_log_rms_mae_db": 1.3081473186640709, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07879347477075051, "300–1000 Hz_10ms_envelope_correlation": 0.9960813638333366, "300–1000 Hz_10ms_modulation_mae_db": 0.6419768490667952, "300–1000 Hz_40ms_log_rms_mae_db": 0.7772586826157775, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043994856424882184, "300–1000 Hz_40ms_envelope_correlation": 0.999039398755792, "300–1000 Hz_40ms_modulation_mae_db": 0.6592493371669673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3591289454283468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18438327069888225, "1000–3000 Hz_10ms_envelope_correlation": 0.9719161287057769, "1000–3000 Hz_10ms_modulation_mae_db": 0.5709145366289687, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8883406138774739, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1317200136150027, "1000–3000 Hz_40ms_envelope_correlation": 0.9858050499769565, "1000–3000 Hz_40ms_modulation_mae_db": 0.9100620355589383, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3473634009778537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19444583277274435, "3000–6000 Hz_10ms_envelope_correlation": 0.9755858154106343, "3000–6000 Hz_10ms_modulation_mae_db": 0.7252557631529889, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8783776422477818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10590424937568588, "3000–6000 Hz_40ms_envelope_correlation": 0.9926909834935553, "3000–6000 Hz_40ms_modulation_mae_db": 0.7330381778419577, "mr_stft_db": 5.425940693967488, "band_mae_db": 0.19032787579513175, "rms_gain_error_db": -0.1349868750600474, "output_peak": 0.2901032312285657}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.077972775136482, "全帯域_10ms_rms_envelope_nrmse": 0.0712815745616064, "全帯域_10ms_envelope_correlation": 0.9949956404820285, "全帯域_10ms_modulation_mae_db": 0.4034981750109256, "全帯域_40ms_log_rms_mae_db": 0.6748627479063731, "全帯域_40ms_rms_envelope_nrmse": 0.04546553511029581, "全帯域_40ms_envelope_correlation": 0.9980160381767911, "全帯域_40ms_modulation_mae_db": 0.46237331967261025, "300–1000 Hz_10ms_log_rms_mae_db": 1.2432300760293191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06677672583406798, "300–1000 Hz_10ms_envelope_correlation": 0.9970954880345732, "300–1000 Hz_10ms_modulation_mae_db": 0.3848721865171905, "300–1000 Hz_40ms_log_rms_mae_db": 0.7962423400029709, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03704392903864179, "300–1000 Hz_40ms_envelope_correlation": 0.9992128826157136, "300–1000 Hz_40ms_modulation_mae_db": 0.18872497531095236, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3088234241225818, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1796771538657017, "1000–3000 Hz_10ms_envelope_correlation": 0.9727018332259191, "1000–3000 Hz_10ms_modulation_mae_db": 0.5890513698638443, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8535998439355346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12403491153692317, "1000–3000 Hz_40ms_envelope_correlation": 0.9865754862657689, "1000–3000 Hz_40ms_modulation_mae_db": 0.2732925926530349, "3000–6000 Hz_10ms_log_rms_mae_db": 1.250762492360836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18240236863526854, "3000–6000 Hz_10ms_envelope_correlation": 0.9806892465317374, "3000–6000 Hz_10ms_modulation_mae_db": 0.7171794127502551, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8639511760677383, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13643733860245968, "3000–6000 Hz_40ms_envelope_correlation": 0.9903485198888119, "3000–6000 Hz_40ms_modulation_mae_db": 0.9450675055531629, "mr_stft_db": 5.424724218430249, "band_mae_db": 0.27032196305087336, "rms_gain_error_db": -0.13891784480838296, "output_peak": 0.29210838132177935}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0899329190217169, "全帯域_10ms_rms_envelope_nrmse": 0.0712936014668138, "全帯域_10ms_envelope_correlation": 0.9948257715330447, "全帯域_10ms_modulation_mae_db": 0.3149160409385169, "全帯域_40ms_log_rms_mae_db": 0.6267609815790086, "全帯域_40ms_rms_envelope_nrmse": 0.0391537195701688, "全帯域_40ms_envelope_correlation": 0.9983868944501827, "全帯域_40ms_modulation_mae_db": 0.30392387352753797, "300–1000 Hz_10ms_log_rms_mae_db": 1.338474810524126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07933574665851746, "300–1000 Hz_10ms_envelope_correlation": 0.9958168754862656, "300–1000 Hz_10ms_modulation_mae_db": 0.7800910169909993, "300–1000 Hz_40ms_log_rms_mae_db": 0.8344264599114329, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0447892200581969, "300–1000 Hz_40ms_envelope_correlation": 0.9987533988414949, "300–1000 Hz_40ms_modulation_mae_db": 0.888819579143901, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2757565459527669, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.175563481249481, "1000–3000 Hz_10ms_envelope_correlation": 0.9737915671530509, "1000–3000 Hz_10ms_modulation_mae_db": 0.4815257870189676, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7820355999599523, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12189004838445672, "1000–3000 Hz_40ms_envelope_correlation": 0.9867254789051004, "1000–3000 Hz_40ms_modulation_mae_db": 0.6478713915916358, "3000–6000 Hz_10ms_log_rms_mae_db": 1.272481335358433, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1598568802414373, "3000–6000 Hz_10ms_envelope_correlation": 0.9837558603490981, "3000–6000 Hz_10ms_modulation_mae_db": 0.27123872879463407, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8581120726127942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09909059333832772, "3000–6000 Hz_40ms_envelope_correlation": 0.9934867465492478, "3000–6000 Hz_40ms_modulation_mae_db": 0.35718828203371206, "mr_stft_db": 5.40454141508832, "band_mae_db": 0.18690804532288996, "rms_gain_error_db": -0.05771771146442804, "output_peak": 0.291954212848755}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0354966803018488, "全帯域_10ms_rms_envelope_nrmse": 0.06394719356869763, "全帯域_10ms_envelope_correlation": 0.9959429122401509, "全帯域_10ms_modulation_mae_db": 0.36366187998651145, "全帯域_40ms_log_rms_mae_db": 0.6012397418100744, "全帯域_40ms_rms_envelope_nrmse": 0.0321521766050953, "全帯域_40ms_envelope_correlation": 0.9990525752379015, "全帯域_40ms_modulation_mae_db": 0.6011524120406163, "300–1000 Hz_10ms_log_rms_mae_db": 1.280035257147737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05895857002356997, "300–1000 Hz_10ms_envelope_correlation": 0.9977651373993153, "300–1000 Hz_10ms_modulation_mae_db": 0.4010470940132458, "300–1000 Hz_40ms_log_rms_mae_db": 0.7796171820349015, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0365270763069133, "300–1000 Hz_40ms_envelope_correlation": 0.9992283222614179, "300–1000 Hz_40ms_modulation_mae_db": 0.42938239808047013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.36434878202936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19634314261633087, "1000–3000 Hz_10ms_envelope_correlation": 0.9673320873355488, "1000–3000 Hz_10ms_modulation_mae_db": 0.4870044373771842, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9343852741726907, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13937962777806184, "1000–3000 Hz_40ms_envelope_correlation": 0.9830469820280526, "1000–3000 Hz_40ms_modulation_mae_db": 0.6408479264194015, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3349357074168897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1651243699792603, "3000–6000 Hz_10ms_envelope_correlation": 0.9824830111869388, "3000–6000 Hz_10ms_modulation_mae_db": 0.7090587749960443, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8982501286093151, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08796118967227103, "3000–6000 Hz_40ms_envelope_correlation": 0.9950378487030147, "3000–6000 Hz_40ms_modulation_mae_db": 1.122725393593509, "mr_stft_db": 5.417804343916528, "band_mae_db": 0.2055606957113195, "rms_gain_error_db": -0.11156650349272601, "output_peak": 0.28982875389940466}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0589859631025298, "全帯域_10ms_rms_envelope_nrmse": 0.07452502781732204, "全帯域_10ms_envelope_correlation": 0.994436894671396, "全帯域_10ms_modulation_mae_db": 0.42008496550142677, "全帯域_40ms_log_rms_mae_db": 0.5911901496619053, "全帯域_40ms_rms_envelope_nrmse": 0.03971911889677891, "全帯域_40ms_envelope_correlation": 0.9984267849354096, "全帯域_40ms_modulation_mae_db": 0.1927253413414779, "300–1000 Hz_10ms_log_rms_mae_db": 1.22598902370399, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08151195511948045, "300–1000 Hz_10ms_envelope_correlation": 0.9955171072575995, "300–1000 Hz_10ms_modulation_mae_db": 0.26779438781838105, "300–1000 Hz_40ms_log_rms_mae_db": 0.7683828659421272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04688498674797432, "300–1000 Hz_40ms_envelope_correlation": 0.9985442703475096, "300–1000 Hz_40ms_modulation_mae_db": 0.26058572177845735, "1000–3000 Hz_10ms_log_rms_mae_db": 1.364756217385898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18210946030349998, "1000–3000 Hz_10ms_envelope_correlation": 0.9717691453744839, "1000–3000 Hz_10ms_modulation_mae_db": 0.7128516793341927, "1000–3000 Hz_40ms_log_rms_mae_db": 0.945895624287685, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1341020464728652, "1000–3000 Hz_40ms_envelope_correlation": 0.983695603156746, "1000–3000 Hz_40ms_modulation_mae_db": 0.38611042322887384, "3000–6000 Hz_10ms_log_rms_mae_db": 1.353550106778565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18477291458299647, "3000–6000 Hz_10ms_envelope_correlation": 0.9781519285279958, "3000–6000 Hz_10ms_modulation_mae_db": 0.6387159000736155, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8793564844469013, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1087028922123513, "3000–6000 Hz_40ms_envelope_correlation": 0.9925976108127009, "3000–6000 Hz_40ms_modulation_mae_db": 0.7910856681647357, "mr_stft_db": 5.423193180337594, "band_mae_db": 0.15171364789794164, "rms_gain_error_db": -0.08563745566149733, "output_peak": 0.3038217189077282}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59840, "output_samples": 59840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.495097139880453, "spectral_convergence": 0.2732686952124712, "spectral_convergence_status": "ok", "amplitude_floor": 9.91771656423115e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.175225504674116, "spectral_convergence": 0.2496477566662587, "spectral_convergence_status": "ok", "amplitude_floor": 5.25712991724442e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.61373900158175, "spectral_convergence": 0.2602692303969858, "spectral_convergence_status": "ok", "amplitude_floor": 4.1776466342690114e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00025386719019706026, "output_power": 0.000258198483818963, "delta_db": 0.07347111425060464, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004894966727833924, "output_power": 0.0004864161638024571, "delta_db": -0.02741745871299691, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005421484972995941, "output_power": 0.0005225549637871789, "delta_db": -0.15986280865763125, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.5605939839496278e-05, "output_power": 1.605505426662332e-05, "delta_db": 0.12321849446465288, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000107892990107672, "output_power": 9.681081883820323e-05, "delta_db": -0.4706933568882582, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.406384293716839e-05, "output_power": 6.952481087467438e-05, "delta_db": -0.2746642631159759, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1882212493483533, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0992686828982936, "pitch": null, "output_rms_dbfs": -28.38763715176374, "output_peak": 0.289354488072027, "reference_rms_dbfs": -28.28836846886545, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.428020548712106}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.000110539633783132, "log_rms_mae_db": 1.0478852336515354, "rms_envelope_nrmse": 0.0754472730241431, "envelope_correlation": 0.99425628591946, "rms_gain_error_db": -0.09926853797137924, "reference_mean_power": 0.00148505750597689, "output_mean_power": 0.0014514979185683168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.0996976338580339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028388367321682654, "output_power": 0.0002786889926282916, "reference_active": true, "delta_db": -0.08020600265366504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6121837601531416e-05, "output_power": 4.0736146224465375e-05, "reference_active": true, "delta_db": -0.5392666137352438}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3216272322760606e-05, "output_power": 9.02636899629992e-06, "reference_active": true, "delta_db": -1.6559589532923553}], "active_band_count": 3, "active_band_mae_db": 0.7584771898937547}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 0.00010014294879668002, "log_rms_mae_db": 0.6426331207619295, "rms_envelope_nrmse": 0.041847002531518775, "envelope_correlation": 0.9982347134163458, "rms_gain_error_db": -0.09925700656413848, "reference_mean_power": 0.0014970488171277764, "output_mean_power": 0.001463222133136499}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.0259449632693504e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002533237576934496, "output_power": 0.0002475867083869784, "reference_active": true, "delta_db": -0.09948595522881135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.399735365000864e-05, "output_power": 2.0044272567200764e-05, "reference_active": true, "delta_db": -0.7817305201209733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.21988770898296e-07, "output_power": 5.399839681045615e-07, "reference_active": true, "delta_db": -0.6140167821196972}], "active_band_count": 3, "active_band_mae_db": 0.4984110858231607}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.00010594974849403162, "log_rms_mae_db": 1.348143011878034, "rms_envelope_nrmse": 0.08477873760244115, "envelope_correlation": 0.9951907318686366, "rms_gain_error_db": -0.1325635245495398, "reference_mean_power": 0.0008075829245111437, "output_mean_power": 0.0007833047771998057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.2363198879746262e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030947682521494476, "output_power": 0.0002969185877890987, "reference_active": true, "delta_db": -0.17990746690229684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.465844834587486e-05, "output_power": 2.9497790574026412e-05, "reference_active": true, "delta_db": -0.7001962757765618}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.693327955984243e-06, "output_power": 6.494183930022964e-06, "reference_active": true, "delta_db": -0.13117520176623085}], "active_band_count": 3, "active_band_mae_db": 0.33709298148169653}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 9.562043532359368e-05, "log_rms_mae_db": 0.834665370583464, "rms_envelope_nrmse": 0.053377904903860615, "envelope_correlation": 0.9981459212018771, "rms_gain_error_db": -0.13256541295960844, "reference_mean_power": 0.0008141211968758593, "output_mean_power": 0.0007896461478925941}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.204812078547339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002902058778585621, "output_power": 0.0002772593818636772, "reference_active": true, "delta_db": -0.19819953769223994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7747321249654287e-05, "output_power": 1.538533883682187e-05, "reference_active": true, "delta_db": -0.620257453591351}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.319074590753967e-07, "output_power": 3.287680590968082e-07, "reference_active": true, "delta_db": -0.04127394800312696}], "active_band_count": 3, "active_band_mae_db": 0.28657697976223934}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 2.8704944305868236e-05, "log_rms_mae_db": 1.3122883868005197, "rms_envelope_nrmse": 0.1715340891336996, "envelope_correlation": 0.9750680271938182, "rms_gain_error_db": -0.22812762674308495, "reference_mean_power": 4.5961113590202676e-05, "output_mean_power": 4.360916588475409e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 2.8606659711814982e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3098445296769517e-05, "output_power": 1.1915580568922915e-05, "reference_active": true, "delta_db": -0.411045431360087}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2173629500624627e-06, "output_power": 3.0615001020033475e-06, "reference_active": true, "delta_db": -0.21565778096490623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2063471779341703e-06, "output_power": 1.2390269029696182e-06, "reference_active": true, "delta_db": 0.11608423724539765}], "active_band_count": 3, "active_band_mae_db": 0.2475958165234636}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 2.1311353675739154e-05, "log_rms_mae_db": 0.8582861122509019, "rms_envelope_nrmse": 0.12282917945627113, "envelope_correlation": 0.986599112899363, "rms_gain_error_db": -0.22816731744948798, "reference_mean_power": 4.633078215253765e-05, "output_mean_power": 4.395951580825512e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 2.5755466765658237e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2064714839118615e-05, "output_power": 1.092986519425989e-05, "reference_active": true, "delta_db": -0.42902255885175766}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.637842198880982e-06, "output_power": 1.5380585359108344e-06, "reference_active": true, "delta_db": -0.2729919219172217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.209021145285469e-07, "output_power": 9.328582190104917e-08, "reference_active": true, "delta_db": -1.1261825429310792}], "active_band_count": 3, "active_band_mae_db": 0.6093990079000196}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 6.327125594838897e-05, "log_rms_mae_db": 1.4275205146609493, "rms_envelope_nrmse": 0.1855535539570722, "envelope_correlation": 0.9785828872657031, "rms_gain_error_db": -0.46887098901499835, "reference_mean_power": 0.0001246130887459015, "output_mean_power": 0.0001118604499659658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.481947168400202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.119164755141333e-05, "output_power": 5.594939795105294e-05, "reference_active": true, "delta_db": -0.3889672887240326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1929332977586818e-05, "output_power": 1.777929705233631e-05, "reference_active": true, "delta_db": -0.9111083589776146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.860956874662985e-06, "output_power": 3.2186439839424677e-06, "reference_active": true, "delta_db": -3.2871174774649448}], "active_band_count": 3, "active_band_mae_db": 1.5290643750555306}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 5.46294552803286e-05, "log_rms_mae_db": 0.9655224363274345, "rms_envelope_nrmse": 0.1092886083194496, "envelope_correlation": 0.9934755011690591, "rms_gain_error_db": -0.4688816177216819, "reference_mean_power": 0.00012561601350824365, "output_mean_power": 0.00011276046157402614}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.3825123163038373e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.737559737198788e-05, "output_power": 5.139638279398202e-05, "reference_active": true, "delta_db": -0.4779466524243645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1774546568382367e-05, "output_power": 8.697535021897452e-06, "reference_active": true, "delta_db": -1.3154800549662629}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.943011691356486e-07, "output_power": 2.800894260803852e-07, "reference_active": true, "delta_db": -3.267098709492518}], "active_band_count": 3, "active_band_mae_db": 1.686841805627715}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.827435370225298, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "db836e1794a182e56697bc0b34d4600e55fb54fd41ffa66246b48d33b5ae389f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9705139368703133, "standard_deviation": 0.0045463106731874845, "minimum": 0.9607829739376138, "maximum": 0.9750680271938182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3328885646676165, "standard_deviation": 0.03691549695755604, "minimum": 1.2757565459527669, "maximum": 1.374380262196956, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6259434321691884, "standard_deviation": 0.28122417464975413, "minimum": 0.2475958165234636, "maximum": 1.2201695373809458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18617735122750306, "standard_deviation": 0.013513675467787749, "minimum": 0.1715340891336996, "maximum": 0.21420527163156927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.984644816414316, "standard_deviation": 0.0021667062281623883, "minimum": 0.9806667114317043, "maximum": 0.9867254789051004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8748699298130579, "standard_deviation": 0.05129105032072266, "minimum": 0.7820355999599523, "maximum": 0.945895624287685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5729799505735247, "standard_deviation": 0.22225699130763393, "minimum": 0.2732925926530349, "maximum": 0.9100620355589383, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13176161784971818, "standard_deviation": 0.008610314337066863, "minimum": 0.12189004838445672, "maximum": 0.14641286021481695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9798273143321758, "standard_deviation": 0.0027631920521181607, "minimum": 0.9755858154106343, "maximum": 0.9837558603490981, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3380741087840162, "standard_deviation": 0.05592328778906863, "minimum": 1.250762492360836, "maximum": 1.4275205146609493, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8134680316662797, "standard_deviation": 0.3600585773804137, "minimum": 0.27123872879463407, "maximum": 1.5290643750555306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17850809014893282, "standard_deviation": 0.012193923556260545, "minimum": 0.1598568802414373, "maximum": 0.19444583277274435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9930838021417645, "standard_deviation": 0.0014393872318947088, "minimum": 0.9903485198888119, "maximum": 0.9950378487030147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8993829826564359, "standard_deviation": 0.03880786376457881, "minimum": 0.8581120726127942, "maximum": 0.9655224363274345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.983722613493607, "standard_deviation": 0.3972900761736535, "minimum": 0.35718828203371206, "maximum": 1.686841805627715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10629721749793772, "standard_deviation": 0.01464546779351033, "minimum": 0.08796118967227103, "maximum": 0.13643733860245968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.996620066094204, "standard_deviation": 0.0010888469602907134, "minimum": 0.9951907318686366, "maximum": 0.9978641823027969, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2849142236615174, "standard_deviation": 0.04348880214352565, "minimum": 1.22598902370399, "maximum": 1.348143011878034, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.52010047860996, "standard_deviation": 0.22710884434282796, "minimum": 0.26779438781838105, "maximum": 0.8993988855214803, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07078282223002408, "standard_deviation": 0.011529057155535703, "minimum": 0.05700787581588119, "maximum": 0.08477873760244115, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989411521661253, "standard_deviation": 0.0004254715103407338, "minimum": 0.9981459212018771, "maximum": 0.9993888407243281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.793631845927852, "standard_deviation": 0.02796197791817801, "minimum": 0.7618878472139144, "maximum": 0.834665370583464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.47228859162208403, "standard_deviation": 0.23888902596124129, "minimum": 0.18872497531095236, "maximum": 0.888819579143901, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04104301412749993, "standard_deviation": 0.007379732655502785, "minimum": 0.03186601993617399, "maximum": 0.053377904903860615, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19962580988691972, "standard_deviation": 0.04630230713358061, "minimum": 0.14160567731674442, "maximum": 0.27032196305087336, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.41966681014609, "standard_deviation": 0.009344115456566517, "minimum": 5.40454141508832, "maximum": 5.428020548712106, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.29131704510515116, "standard_deviation": 0.006057848366726816, "minimum": 0.28173349697433375, "maximum": 0.3038217189077282, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09344196978703376, "standard_deviation": 0.033327697768271915, "minimum": -0.13891784480838296, "maximum": -0.05771771146442804, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9950100299054715, "standard_deviation": 0.0006328920995116077, "minimum": 0.99425628591946, "maximum": 0.9959429122401509, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.041805707413959, "standard_deviation": 0.03438125842207061, "minimum": 0.9879539135082759, "maximum": 1.0899329190217169, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4699766161832845, "standard_deviation": 0.17676101260836774, "minimum": 0.3149160409385169, "maximum": 0.7584771898937547, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07040106852252216, "standard_deviation": 0.004567318764697589, "minimum": 0.06394719356869763, "maximum": 0.0754472730241431, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985716391890727, "standard_deviation": 0.00037998552247571287, "minimum": 0.9980160381767911, "maximum": 0.9991002367277562, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5936697046171193, "standard_deviation": 0.054033106694929174, "minimum": 0.5147139921994948, "maximum": 0.6748627479063731, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4405474682138586, "standard_deviation": 0.14366125851554165, "minimum": 0.1927253413414779, "maximum": 0.6011524120406163, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037742096976117336, "standard_deviation": 0.005072777886191256, "minimum": 0.02984204400871222, "maximum": 0.04546553511029581, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_672-122797-0056", "language": "en", "style": "normal", "speaker": "librispeech_672", "source_utterance_id": "672-122797-0056", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 83600, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_672-122797-0056.wav", "source_file_sha256": "0c532e31b3d09dc1b5730a25c396cfe39e9696c75d4a2f749ee597cf1f8df6e9", "decoded_pcm_sha256": "1ee86311a9a7b6ee7bf0ef8ccad1d6c7f831fa6001565687e3d9e0eb90cd44f4", "original_source_sha256": "e58678d63f84c218fb28da622351b90e344380dbfc13763dd1123e71faf708d1", "corpus_entry_id": "librispeech_672-122797-0056"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.408585786819458, "predicted_mos": 3.408585786819458, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "6fe1ac355644383e56a818b8e2903dd662eda2633844085342aa222e77256ed5", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.071710598000209, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_672-122797-0056", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6fe1ac355644383e56a818b8e2903dd662eda2633844085342aa222e77256ed5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "22780bba4503a457a8d49959a4601aca0ceb4e387b94d1e3c1ba07185cf8e7cb"}, "rows": [{"case": "librispeech_672-122797-0056", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.328829050064087, "predicted_mos": 2.328829050064087, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "637e639d45c41aa3e16fc244af0118155bdee35d73d1e548bf1ea36ec67961df", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3275432939990424, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_672-122797-0056", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "637e639d45c41aa3e16fc244af0118155bdee35d73d1e548bf1ea36ec67961df", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.036614684459299, "全帯域_10ms_rms_envelope_nrmse": 0.08005827084874267, "全帯域_10ms_envelope_correlation": 0.9925948852017841, "全帯域_10ms_modulation_mae_db": 0.26533251223624565, "全帯域_40ms_log_rms_mae_db": 0.5887157370706319, "全帯域_40ms_rms_envelope_nrmse": 0.0389515921386556, "全帯域_40ms_envelope_correlation": 0.998154790713246, "全帯域_40ms_modulation_mae_db": 0.4061885581938165, "300–1000 Hz_10ms_log_rms_mae_db": 1.3759597177503151, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0805117769739711, "300–1000 Hz_10ms_envelope_correlation": 0.9946971057741069, "300–1000 Hz_10ms_modulation_mae_db": 0.20192076236890366, "300–1000 Hz_40ms_log_rms_mae_db": 0.7572195566770952, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043139614122683356, "300–1000 Hz_40ms_envelope_correlation": 0.9984704958505534, "300–1000 Hz_40ms_modulation_mae_db": 0.03940678333824279, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3347106696323336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15784689293513643, "1000–3000 Hz_10ms_envelope_correlation": 0.9812213325250108, "1000–3000 Hz_10ms_modulation_mae_db": 0.8506461801732638, "1000–3000 Hz_40ms_log_rms_mae_db": 0.750461150665438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09105980437764441, "1000–3000 Hz_40ms_envelope_correlation": 0.9948119328165823, "1000–3000 Hz_40ms_modulation_mae_db": 0.44312939001092394, "3000–6000 Hz_10ms_log_rms_mae_db": 1.482357526115425, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13802659383699792, "3000–6000 Hz_10ms_envelope_correlation": 0.9891660014942504, "3000–6000 Hz_10ms_modulation_mae_db": 0.2937416673474827, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9290120694778714, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07449779518213222, "3000–6000 Hz_40ms_envelope_correlation": 0.9967873233281326, "3000–6000 Hz_40ms_modulation_mae_db": 0.524496812971389, "mr_stft_db": 5.487150704705012, "band_mae_db": 0.24711104143605814, "rms_gain_error_db": -0.05686124421887585, "output_peak": 0.31474974041951725}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9833384419227807, "全帯域_10ms_rms_envelope_nrmse": 0.09368905201416525, "全帯域_10ms_envelope_correlation": 0.9898097585279398, "全帯域_10ms_modulation_mae_db": 0.36506146162669767, "全帯域_40ms_log_rms_mae_db": 0.489042433165572, "全帯域_40ms_rms_envelope_nrmse": 0.051765829739053595, "全帯域_40ms_envelope_correlation": 0.9967036667840576, "全帯域_40ms_modulation_mae_db": 0.5398993094965067, "300–1000 Hz_10ms_log_rms_mae_db": 1.4260447623751051, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09561657597874987, "300–1000 Hz_10ms_envelope_correlation": 0.9927490672617072, "300–1000 Hz_10ms_modulation_mae_db": 0.5497773292670296, "300–1000 Hz_40ms_log_rms_mae_db": 0.8332561913347398, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05791873026044409, "300–1000 Hz_40ms_envelope_correlation": 0.9975098297061021, "300–1000 Hz_40ms_modulation_mae_db": 0.39545780982439666, "1000–3000 Hz_10ms_log_rms_mae_db": 1.336145682688358, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16624931715654312, "1000–3000 Hz_10ms_envelope_correlation": 0.9780697928589719, "1000–3000 Hz_10ms_modulation_mae_db": 0.17864602517315764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7007349894221309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08796068119693312, "1000–3000 Hz_40ms_envelope_correlation": 0.9936463867353237, "1000–3000 Hz_40ms_modulation_mae_db": 0.49050647746027015, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4775260110860076, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18089527541490605, "3000–6000 Hz_10ms_envelope_correlation": 0.9810919228220889, "3000–6000 Hz_10ms_modulation_mae_db": 0.6425742581042413, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9118614846886166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11678024311820623, "3000–6000 Hz_40ms_envelope_correlation": 0.9921615459762811, "3000–6000 Hz_40ms_modulation_mae_db": 0.9543907139102137, "mr_stft_db": 5.498229925971611, "band_mae_db": 0.15605595131924213, "rms_gain_error_db": -0.0795670650634358, "output_peak": 0.37561053755456014}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0675325217667488, "全帯域_10ms_rms_envelope_nrmse": 0.09263658610520809, "全帯域_10ms_envelope_correlation": 0.990272331175493, "全帯域_10ms_modulation_mae_db": 0.24986534082395093, "全帯域_40ms_log_rms_mae_db": 0.5791104161222855, "全帯域_40ms_rms_envelope_nrmse": 0.04905413913310198, "全帯域_40ms_envelope_correlation": 0.9973854086630576, "全帯域_40ms_modulation_mae_db": 0.3371186483684956, "300–1000 Hz_10ms_log_rms_mae_db": 1.3988098490907022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11353743639127056, "300–1000 Hz_10ms_envelope_correlation": 0.989645899055066, "300–1000 Hz_10ms_modulation_mae_db": 1.1621346692113617, "300–1000 Hz_40ms_log_rms_mae_db": 0.7687002122362914, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06711816772423841, "300–1000 Hz_40ms_envelope_correlation": 0.9965836316504562, "300–1000 Hz_40ms_modulation_mae_db": 0.7703872341361144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3112022005070485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1558854676663315, "1000–3000 Hz_10ms_envelope_correlation": 0.9822935550471104, "1000–3000 Hz_10ms_modulation_mae_db": 1.448357922903045, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6596205609595357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.095022115979829, "1000–3000 Hz_40ms_envelope_correlation": 0.9948028830038862, "1000–3000 Hz_40ms_modulation_mae_db": 1.138999175963366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.396446169374972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1601304978937095, "3000–6000 Hz_10ms_envelope_correlation": 0.9852366621728136, "3000–6000 Hz_10ms_modulation_mae_db": 0.34088942873764877, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8884415268292899, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08913495934759351, "3000–6000 Hz_40ms_envelope_correlation": 0.9955438199926757, "3000–6000 Hz_40ms_modulation_mae_db": 0.4037314639573051, "mr_stft_db": 5.48806839967628, "band_mae_db": 0.2594718421268054, "rms_gain_error_db": -0.1687738761811839, "output_peak": 0.3345583423593539}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.986122127547168, "全帯域_10ms_rms_envelope_nrmse": 0.08316986905254631, "全帯域_10ms_envelope_correlation": 0.9920006730046238, "全帯域_10ms_modulation_mae_db": 0.2909434206903356, "全帯域_40ms_log_rms_mae_db": 0.5412104439099041, "全帯域_40ms_rms_envelope_nrmse": 0.04202978657100833, "全帯域_40ms_envelope_correlation": 0.997910768695739, "全帯域_40ms_modulation_mae_db": 0.2938882686204903, "300–1000 Hz_10ms_log_rms_mae_db": 1.405089007032959, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08972501755560447, "300–1000 Hz_10ms_envelope_correlation": 0.9935254461077991, "300–1000 Hz_10ms_modulation_mae_db": 0.5008171285935169, "300–1000 Hz_40ms_log_rms_mae_db": 0.7515553903497761, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04872970912048151, "300–1000 Hz_40ms_envelope_correlation": 0.9981991938282786, "300–1000 Hz_40ms_modulation_mae_db": 0.30902360647984256, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3560093372739594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15827379888087573, "1000–3000 Hz_10ms_envelope_correlation": 0.9800219592384473, "1000–3000 Hz_10ms_modulation_mae_db": 0.22641809683453126, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7577753989126197, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08365508789021021, "1000–3000 Hz_40ms_envelope_correlation": 0.9943264892966035, "1000–3000 Hz_40ms_modulation_mae_db": 0.5715349808446758, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4453668377247009, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13716768788343864, "3000–6000 Hz_10ms_envelope_correlation": 0.9896159051511147, "3000–6000 Hz_10ms_modulation_mae_db": 0.8724343470849482, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8938824952819467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07864424099442008, "3000–6000 Hz_40ms_envelope_correlation": 0.9970657572362173, "3000–6000 Hz_40ms_modulation_mae_db": 0.9918418770936323, "mr_stft_db": 5.5029554166452, "band_mae_db": 0.19294817226647365, "rms_gain_error_db": -0.0948755855602308, "output_peak": 0.3304744945972161}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9657648329498932, "全帯域_10ms_rms_envelope_nrmse": 0.0966683714884231, "全帯域_10ms_envelope_correlation": 0.9893169491911548, "全帯域_10ms_modulation_mae_db": 0.3296858495344961, "全帯域_40ms_log_rms_mae_db": 0.4800037223195203, "全帯域_40ms_rms_envelope_nrmse": 0.05654237340229667, "全帯域_40ms_envelope_correlation": 0.9960625654469613, "全帯域_40ms_modulation_mae_db": 0.20021568607033238, "300–1000 Hz_10ms_log_rms_mae_db": 1.4127759297175975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10405216142115338, "300–1000 Hz_10ms_envelope_correlation": 0.9912987440695167, "300–1000 Hz_10ms_modulation_mae_db": 0.5671083635292083, "300–1000 Hz_40ms_log_rms_mae_db": 0.7876818408127928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06759762048167434, "300–1000 Hz_40ms_envelope_correlation": 0.9963913300101078, "300–1000 Hz_40ms_modulation_mae_db": 0.3502907593548135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2983647161074057, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1369288753559483, "1000–3000 Hz_10ms_envelope_correlation": 0.9851430958817753, "1000–3000 Hz_10ms_modulation_mae_db": 0.4846958957695251, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7465564275619148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07784760339549072, "1000–3000 Hz_40ms_envelope_correlation": 0.9951230562317621, "1000–3000 Hz_40ms_modulation_mae_db": 0.5152180574359094, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4912586226572397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19172630001094676, "3000–6000 Hz_10ms_envelope_correlation": 0.980825281254152, "3000–6000 Hz_10ms_modulation_mae_db": 0.584795540480299, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9651728359957347, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12119736393697395, "3000–6000 Hz_40ms_envelope_correlation": 0.9927839498949593, "3000–6000 Hz_40ms_modulation_mae_db": 0.6702053649658705, "mr_stft_db": 5.49665281998485, "band_mae_db": 0.25791962864434803, "rms_gain_error_db": 0.044130216128788605, "output_peak": 0.36161490109467154}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.045178735818875, "全帯域_10ms_rms_envelope_nrmse": 0.08463656432526531, "全帯域_10ms_envelope_correlation": 0.9917197968410306, "全帯域_10ms_modulation_mae_db": 0.24873087427377952, "全帯域_40ms_log_rms_mae_db": 0.6045642882542528, "全帯域_40ms_rms_envelope_nrmse": 0.044103556322745145, "全帯域_40ms_envelope_correlation": 0.9976488067610166, "全帯域_40ms_modulation_mae_db": 0.40846154236057974, "300–1000 Hz_10ms_log_rms_mae_db": 1.4390819196262192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0926890862522347, "300–1000 Hz_10ms_envelope_correlation": 0.9930824102682828, "300–1000 Hz_10ms_modulation_mae_db": 0.37013601806994184, "300–1000 Hz_40ms_log_rms_mae_db": 0.8191448140992758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05202720484419596, "300–1000 Hz_40ms_envelope_correlation": 0.9979224556929748, "300–1000 Hz_40ms_modulation_mae_db": 0.16946238213423612, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3275772598238051, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1477230428303638, "1000–3000 Hz_10ms_envelope_correlation": 0.9825820892870615, "1000–3000 Hz_10ms_modulation_mae_db": 0.16313372943395, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6823963941447974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0790398220203104, "1000–3000 Hz_40ms_envelope_correlation": 0.994755526946136, "1000–3000 Hz_40ms_modulation_mae_db": 0.10203239193719023, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5309714725490335, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15360284928941936, "3000–6000 Hz_10ms_envelope_correlation": 0.9865471308815968, "3000–6000 Hz_10ms_modulation_mae_db": 0.49806081974401933, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9794429243309297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0907395412767351, "3000–6000 Hz_40ms_envelope_correlation": 0.9955300258050327, "3000–6000 Hz_40ms_modulation_mae_db": 0.4167902403335961, "mr_stft_db": 5.498978317338872, "band_mae_db": 0.12063558810492618, "rms_gain_error_db": -0.08435154139892076, "output_peak": 0.3448445290776071}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.123998866846082, "全帯域_10ms_rms_envelope_nrmse": 0.11470730999381017, "全帯域_10ms_envelope_correlation": 0.9847547400539743, "全帯域_10ms_modulation_mae_db": 0.5272437001415609, "全帯域_40ms_log_rms_mae_db": 0.6096077572579415, "全帯域_40ms_rms_envelope_nrmse": 0.06399217777543252, "全帯域_40ms_envelope_correlation": 0.9949054101460038, "全帯域_40ms_modulation_mae_db": 1.147093211491815, "300–1000 Hz_10ms_log_rms_mae_db": 1.4277607253810498, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.095644763054704, "300–1000 Hz_10ms_envelope_correlation": 0.9925362388020795, "300–1000 Hz_10ms_modulation_mae_db": 0.25385503489513167, "300–1000 Hz_40ms_log_rms_mae_db": 0.7950985731784275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05333855506761226, "300–1000 Hz_40ms_envelope_correlation": 0.9976984977180896, "300–1000 Hz_40ms_modulation_mae_db": 0.516128982904718, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3314912664152956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18692982374984257, "1000–3000 Hz_10ms_envelope_correlation": 0.9721400962830072, "1000–3000 Hz_10ms_modulation_mae_db": 0.4029522608008726, "1000–3000 Hz_40ms_log_rms_mae_db": 0.774310980375049, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11058939002123436, "1000–3000 Hz_40ms_envelope_correlation": 0.9896369954701502, "1000–3000 Hz_40ms_modulation_mae_db": 0.5961967607584536, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5079547182586153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2173963186363845, "3000–6000 Hz_10ms_envelope_correlation": 0.9725080771115578, "3000–6000 Hz_10ms_modulation_mae_db": 1.043349897868539, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9661009176135882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12992820312919662, "3000–6000 Hz_40ms_envelope_correlation": 0.990159884870506, "3000–6000 Hz_40ms_modulation_mae_db": 1.6418828586208571, "mr_stft_db": 5.497813629758807, "band_mae_db": 0.12288767377714904, "rms_gain_error_db": -0.07550840201084653, "output_peak": 0.3257981925416191}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.018420035741866, "全帯域_10ms_rms_envelope_nrmse": 0.09559953382514463, "全帯域_10ms_envelope_correlation": 0.9894057552394848, "全帯域_10ms_modulation_mae_db": 0.26216841944859764, "全帯域_40ms_log_rms_mae_db": 0.5467104782491853, "全帯域_40ms_rms_envelope_nrmse": 0.04872969066219383, "全帯域_40ms_envelope_correlation": 0.9970744151442728, "全帯域_40ms_modulation_mae_db": 0.45377265046831905, "300–1000 Hz_10ms_log_rms_mae_db": 1.4135968230316325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09224462733266414, "300–1000 Hz_10ms_envelope_correlation": 0.993252450715327, "300–1000 Hz_10ms_modulation_mae_db": 0.31092694789210323, "300–1000 Hz_40ms_log_rms_mae_db": 0.7718738363525125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04902680652882379, "300–1000 Hz_40ms_envelope_correlation": 0.9983246213404536, "300–1000 Hz_40ms_modulation_mae_db": 0.29359261457233043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3391446878286477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16508046481627486, "1000–3000 Hz_10ms_envelope_correlation": 0.9786588366263967, "1000–3000 Hz_10ms_modulation_mae_db": 0.8431367252696546, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6939674847345041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0940476436358795, "1000–3000 Hz_40ms_envelope_correlation": 0.9934531671059995, "1000–3000 Hz_40ms_modulation_mae_db": 0.5653511520878411, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4391362205202394, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17317025588507462, "3000–6000 Hz_10ms_envelope_correlation": 0.9826517405876183, "3000–6000 Hz_10ms_modulation_mae_db": 0.35176848380628406, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8897813877838774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09307933709142531, "3000–6000 Hz_40ms_envelope_correlation": 0.9948981506978812, "3000–6000 Hz_40ms_modulation_mae_db": 0.5714869714404679, "mr_stft_db": 5.482609208516543, "band_mae_db": 0.16740209843727663, "rms_gain_error_db": -0.06760037946454796, "output_peak": 0.34820672115116474}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83600, "output_samples": 83600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6020939457787, "spectral_convergence": 0.35144360750287795, "spectral_convergence_status": "ok", "amplitude_floor": 9.102684030105167e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.234853226650117, "spectral_convergence": 0.33435465434764944, "spectral_convergence_status": "ok", "amplitude_floor": 4.6649711782186323e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.62450494168622, "spectral_convergence": 0.3536996702455223, "spectral_convergence_status": "ok", "amplitude_floor": 4.1057122515023654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021084017487574348, "output_power": 0.0002136127992023482, "delta_db": 0.05673903335552691, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00047622157098200406, "output_power": 0.00047063620869119364, "delta_db": -0.051237265195161905, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 9.587972951197478e-05, "output_power": 9.428812017100099e-05, "delta_db": -0.07269822903998245, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.411215555329107e-05, "output_power": 6.309509908327315e-05, "delta_db": -0.6989381821621677, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013227809814888984, "output_power": 0.00012290344746057657, "delta_db": -0.31923877058706807, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00014809096583739848, "output_power": 0.00015809205413091568, "delta_db": 0.2838147682764418, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24711104143605814, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05686124421887585, "pitch": null, "output_rms_dbfs": -29.497642349742232, "output_peak": 0.31474974041951725, "reference_rms_dbfs": -29.44078110552336, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.487150704705012}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011323786448874551, "log_rms_mae_db": 1.036614684459299, "rms_envelope_nrmse": 0.08005827084874267, "envelope_correlation": 0.9925948852017841, "rms_gain_error_db": -0.056949034624732844, "reference_mean_power": 0.0011384943575262435, "output_mean_power": 0.0011236627372643723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 5.62759551802579e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002928658523286973, "output_power": 0.0002918648893534324, "reference_active": true, "delta_db": -0.014868830741544027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4610498849864094e-05, "output_power": 4.4368497603060725e-05, "reference_active": true, "delta_db": -0.023623567231618806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2101686887119984e-05, "output_power": 1.01647194898413e-05, "reference_active": true, "delta_db": -0.7575051387355741}], "active_band_count": 3, "active_band_mae_db": 0.26533251223624565}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.563924945576787e-05, "log_rms_mae_db": 0.5887157370706319, "rms_envelope_nrmse": 0.0389515921386556, "envelope_correlation": 0.998154790713246, "rms_gain_error_db": -0.057036917696008445, "reference_mean_power": 0.0011449801518647084, "output_mean_power": 0.0011300411710253842}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.095010942049215e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002727818973842863, "output_power": 0.00027293722923619405, "reference_active": true, "delta_db": 0.002472325585762029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1267992599082604e-05, "output_power": 2.0825298048135286e-05, "reference_active": true, "delta_db": -0.09135274842407215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.757168583210849e-07, "output_power": 4.4435992697991033e-07, "reference_active": true, "delta_db": -1.1247406005716154}], "active_band_count": 3, "active_band_mae_db": 0.4061885581938165}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.979916609312644e-05, "log_rms_mae_db": 1.3759597177503151, "rms_envelope_nrmse": 0.0805117769739711, "envelope_correlation": 0.9946971057741069, "rms_gain_error_db": -0.10297191664502081, "reference_mean_power": 0.00038382289364078325, "output_mean_power": 0.0003748294309491366}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 2.8601579771498626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014700129208541836, "output_power": 0.00014605239726903167, "reference_active": true, "delta_db": -0.028124622746123883}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.957621157356578e-05, "output_power": 1.85409045046023e-05, "reference_active": true, "delta_db": -0.23597733020741415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3113428408893907e-06, "output_power": 2.5005204026357644e-06, "reference_active": true, "delta_db": 0.341660334153173}], "active_band_count": 3, "active_band_mae_db": 0.20192076236890366}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.59156861629162e-05, "log_rms_mae_db": 0.7572195566770952, "rms_envelope_nrmse": 0.043139614122683356, "envelope_correlation": 0.9984704958505534, "rms_gain_error_db": -0.1029684439197799, "reference_mean_power": 0.00038603855676420007, "output_mean_power": 0.0003769934796968646}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 2.667606930039739e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013762626674571402, "output_power": 0.00013744833117913875, "reference_active": true, "delta_db": -0.005618581000153637}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.746730510582489e-06, "output_power": 9.502572785217277e-06, "reference_active": true, "delta_db": -0.11017753552081516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5949673646594045e-07, "output_power": 1.5940773016935285e-07, "reference_active": true, "delta_db": -0.0024242334937595537}], "active_band_count": 3, "active_band_mae_db": 0.03940678333824279}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.341026161507385e-05, "log_rms_mae_db": 1.3347106696323336, "rms_envelope_nrmse": 0.15784689293513643, "envelope_correlation": 0.9812213325250108, "rms_gain_error_db": -0.442470040316013, "reference_mean_power": 0.0001433729237288462, "output_mean_power": 0.00012948520170333612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.1455083323281239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.589033152322262e-05, "output_power": 4.605608081555387e-05, "reference_active": true, "delta_db": -0.8404970720358202}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4083924473542424e-05, "output_power": 1.3406890184963811e-05, "reference_active": true, "delta_db": -0.21395635342226915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.499188380560451e-06, "output_power": 3.895382291834947e-06, "reference_active": true, "delta_db": -1.497485115061702}], "active_band_count": 3, "active_band_mae_db": 0.8506461801732638}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 5.482542724448801e-05, "log_rms_mae_db": 0.750461150665438, "rms_envelope_nrmse": 0.09105980437764441, "envelope_correlation": 0.9948119328165823, "rms_gain_error_db": -0.44236288091811765, "reference_mean_power": 0.00014418079784839864, "output_mean_power": 0.00013021803464598547}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.042775196620521e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2049448137923216e-05, "output_power": 4.3358179012158735e-05, "reference_active": true, "delta_db": -0.79345095095823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.197217517258297e-06, "output_power": 5.804848846575327e-06, "reference_active": true, "delta_db": -0.2840582424218409}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5366430302416474e-07, "output_power": 2.3937095785691066e-07, "reference_active": true, "delta_db": -0.25187897665270104}], "active_band_count": 3, "active_band_mae_db": 0.44312939001092394}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011141137328067575, "log_rms_mae_db": 1.482357526115425, "rms_envelope_nrmse": 0.13802659383699792, "envelope_correlation": 0.9891660014942504, "rms_gain_error_db": 0.06286705371687866, "reference_mean_power": 0.00018590462562984253, "output_mean_power": 0.00018861529242990665}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.861645592580093e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.396584441083375e-05, "output_power": 9.234177774106067e-05, "reference_active": true, "delta_db": -0.07571789613532325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7617019162087818e-05, "output_power": 1.703149069800127e-05, "reference_active": true, "delta_db": -0.1467976504377269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.489954892102086e-06, "output_power": 5.576615900905792e-06, "reference_active": true, "delta_db": -0.658709455469398}], "active_band_count": 3, "active_band_mae_db": 0.2937416673474827}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.054702623382987e-05, "log_rms_mae_db": 0.9290120694778714, "rms_envelope_nrmse": 0.07449779518213222, "envelope_correlation": 0.9967873233281326, "rms_gain_error_db": 0.0628696564781442, "reference_mean_power": 0.00018697687803543306, "output_mean_power": 0.00018970329298918404}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.7928912528322647e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.176227386223464e-05, "output_power": 9.079886375163674e-05, "reference_active": true, "delta_db": -0.045837529759394785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.193273720945643e-06, "output_power": 7.588067236147007e-06, "reference_active": true, "delta_db": -0.33326293596019463}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.787097063305561e-07, "output_power": 2.876520271281904e-07, "reference_active": true, "delta_db": -1.1943899731945775}], "active_band_count": 3, "active_band_mae_db": 0.524496812971389}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8429657338158211, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5203a3ccc03da612bc8f6a232599e2d5b51dcacbbee0a6674c0ca8269252ba15", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9800163447184727, "standard_deviation": 0.003912827229185917, "minimum": 0.9721400962830072, "maximum": 0.9851430958817753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3293307275346067, "standard_deviation": 0.01764219519126348, "minimum": 1.2983647161074057, "maximum": 1.3560093372739594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.57474835454475, "standard_deviation": 0.44667446954231654, "minimum": 0.16313372943395, "maximum": 1.448357922903045, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15936471042391454, "standard_deviation": 0.014611136840750128, "minimum": 0.1369288753559483, "maximum": 0.18692982374984257, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938195547008054, "standard_deviation": 0.001790527479256221, "minimum": 0.9896369954701502, "maximum": 0.9951230562317621, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7207279233469988, "standard_deviation": 0.04160029229566896, "minimum": 0.6596205609595357, "maximum": 0.774310980375049, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5528710483123288, "standard_deviation": 0.28448781239746784, "minimum": 0.10203239193719023, "maximum": 1.138999175963366, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08990276856469147, "standard_deviation": 0.010547840527184917, "minimum": 0.07784760339549072, "maximum": 0.11058939002123436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9834553401843991, "standard_deviation": 0.00555861986660134, "minimum": 0.9725080771115578, "maximum": 0.9896159051511147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4713771972857792, "standard_deviation": 0.042717874517234945, "minimum": 1.396446169374972, "maximum": 1.5309714725490335, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5784518053966827, "standard_deviation": 0.26780901710276767, "minimum": 0.2937416673474827, "maximum": 1.043349897868539, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1690144723563597, "standard_deviation": 0.027531705377968962, "minimum": 0.13716768788343864, "maximum": 0.2173963186363845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943663072252108, "standard_deviation": 0.0024269557643287506, "minimum": 0.990159884870506, "maximum": 0.9970657572362173, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9279619552502318, "standard_deviation": 0.03766559201342298, "minimum": 0.8884415268292899, "maximum": 0.9794429243309297, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7718532879116664, "standard_deviation": 0.4160896155030084, "minimum": 0.4037314639573051, "maximum": 1.6418828586208571, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09925021050958538, "standard_deviation": 0.020633662986905447, "minimum": 0.07449779518213222, "maximum": 0.12992820312919662, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925984202567356, "standard_deviation": 0.0015287231223077942, "minimum": 0.989645899055066, "maximum": 0.9946971057741069, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4123898417506975, "standard_deviation": 0.019645668853046156, "minimum": 1.3759597177503151, "maximum": 1.4390819196262192, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48958453172839966, "standard_deviation": 0.30392393211603574, "minimum": 0.20192076236890366, "maximum": 1.1621346692113617, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09550268062004402, "standard_deviation": 0.009819943296092516, "minimum": 0.0805117769739711, "maximum": 0.11353743639127056, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997637506974627, "standard_deviation": 0.0007788559461778346, "minimum": 0.9963913300101078, "maximum": 0.9984704958505534, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7855663018801138, "standard_deviation": 0.029104271652492412, "minimum": 0.7515553903497761, "maximum": 0.8332561913347398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.35546877159308676, "standard_deviation": 0.22041335795543962, "minimum": 0.03940678333824279, "maximum": 0.7703872341361144, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05486205101876922, "standard_deviation": 0.008792683883883897, "minimum": 0.043139614122683356, "maximum": 0.06759762048167434, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1905539995140349, "standard_deviation": 0.058169297029178405, "minimum": 0.12063558810492618, "maximum": 0.2594718421268054, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.494057302824647, "standard_deviation": 0.007134195524758823, "minimum": 5.482609208516543, "maximum": 5.5029554166452, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.34198218234946376, "standard_deviation": 0.0198361701674523, "minimum": 0.31474974041951725, "maximum": 0.37561053755456014, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07292598472115662, "standard_deviation": 0.05835667615506865, "minimum": -0.1687738761811839, "maximum": 0.044130216128788605, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899843611544357, "standard_deviation": 0.0024534133510986738, "minimum": 0.9847547400539743, "maximum": 0.9925948852017841, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.028371280881589, "standard_deviation": 0.05190681812316086, "minimum": 0.9657648329498932, "maximum": 1.123998866846082, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.317378947346958, "standard_deviation": 0.0942601357532873, "minimum": 0.24873087427377952, "maximum": 0.5272437001415609, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0926456947066632, "standard_deviation": 0.01085680379970218, "minimum": 0.08005827084874267, "maximum": 0.11470730999381017, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969807290442944, "standard_deviation": 0.0010749751816158631, "minimum": 0.9949054101460038, "maximum": 0.998154790713246, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5548706595436617, "standard_deviation": 0.04984298976319005, "minimum": 0.4800037223195203, "maximum": 0.6096077572579415, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4733297343837944, "standard_deviation": 0.2910157123392154, "minimum": 0.20021568607033238, "maximum": 1.147093211491815, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04939614321806096, "standard_deviation": 0.008120984319329274, "minimum": 0.0389515921386556, "maximum": 0.06399217777543252, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0056", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.3888368606567383, "predicted_mos": 2.3888368606567383, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "eb5c6d8b5919487aa5eedc449168967ab8443e1a6f02b1e5f41c4f96325ad3f4", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9967220001854002, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0775576084109346, "全帯域_10ms_rms_envelope_nrmse": 0.08706444711883025, "全帯域_10ms_envelope_correlation": 0.991248682693231, "全帯域_10ms_modulation_mae_db": 0.13623114759477137, "全帯域_40ms_log_rms_mae_db": 0.6315606000532192, "全帯域_40ms_rms_envelope_nrmse": 0.04166563376420752, "全帯域_40ms_envelope_correlation": 0.9978911731113008, "全帯域_40ms_modulation_mae_db": 0.2685292582026593, "300–1000 Hz_10ms_log_rms_mae_db": 1.3260411552766664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08518190457897826, "300–1000 Hz_10ms_envelope_correlation": 0.9940484030234515, "300–1000 Hz_10ms_modulation_mae_db": 0.4787737488682316, "300–1000 Hz_40ms_log_rms_mae_db": 0.6930002338238425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0421066250984882, "300–1000 Hz_40ms_envelope_correlation": 0.9985527746314488, "300–1000 Hz_40ms_modulation_mae_db": 0.27661687930176143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3389580437590778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16578294229106144, "1000–3000 Hz_10ms_envelope_correlation": 0.979014838781364, "1000–3000 Hz_10ms_modulation_mae_db": 0.6982006260978895, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7101327998871642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0976189565414837, "1000–3000 Hz_40ms_envelope_correlation": 0.9936029010384281, "1000–3000 Hz_40ms_modulation_mae_db": 0.43659399667772997, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4832772751080223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15685430308440176, "3000–6000 Hz_10ms_envelope_correlation": 0.9859892009119263, "3000–6000 Hz_10ms_modulation_mae_db": 0.33714566177649813, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8802823260971696, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08113431355582701, "3000–6000 Hz_40ms_envelope_correlation": 0.9961897654224277, "3000–6000 Hz_40ms_modulation_mae_db": 0.6307024567928715, "mr_stft_db": 5.450445444952286, "band_mae_db": 0.2476613577930303, "rms_gain_error_db": -0.06419722830166434, "output_peak": 0.3298703619418707}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.031455933953256, "全帯域_10ms_rms_envelope_nrmse": 0.098179416516251, "全帯域_10ms_envelope_correlation": 0.9888170959861314, "全帯域_10ms_modulation_mae_db": 0.386897218191432, "全帯域_40ms_log_rms_mae_db": 0.5150799259201784, "全帯域_40ms_rms_envelope_nrmse": 0.05675017831148725, "全帯域_40ms_envelope_correlation": 0.9960426865667832, "全帯域_40ms_modulation_mae_db": 0.5794968123584242, "300–1000 Hz_10ms_log_rms_mae_db": 1.3867844075006759, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.091265419271215, "300–1000 Hz_10ms_envelope_correlation": 0.9933564591721075, "300–1000 Hz_10ms_modulation_mae_db": 0.3353048030460415, "300–1000 Hz_40ms_log_rms_mae_db": 0.7300655435357004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05331059741023925, "300–1000 Hz_40ms_envelope_correlation": 0.9978687560571692, "300–1000 Hz_40ms_modulation_mae_db": 0.49388498016344656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3164292069674135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1685755792287438, "1000–3000 Hz_10ms_envelope_correlation": 0.9775729901665439, "1000–3000 Hz_10ms_modulation_mae_db": 0.2369918552796871, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6629336945162613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08749402486448572, "1000–3000 Hz_40ms_envelope_correlation": 0.9939665901825429, "1000–3000 Hz_40ms_modulation_mae_db": 0.39755224988464394, "3000–6000 Hz_10ms_log_rms_mae_db": 1.458549628589049, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18864206626487606, "3000–6000 Hz_10ms_envelope_correlation": 0.979439104001045, "3000–6000 Hz_10ms_modulation_mae_db": 0.6358477870478466, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8903947192867943, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1248218368172987, "3000–6000 Hz_40ms_envelope_correlation": 0.9910509986698128, "3000–6000 Hz_40ms_modulation_mae_db": 1.0373043997543028, "mr_stft_db": 5.470873284624709, "band_mae_db": 0.17146733373004788, "rms_gain_error_db": -0.09018565246226172, "output_peak": 0.3773659200002907}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1082541772062424, "全帯域_10ms_rms_envelope_nrmse": 0.0945981263571299, "全帯域_10ms_envelope_correlation": 0.9897790757133501, "全帯域_10ms_modulation_mae_db": 0.22659204052945803, "全帯域_40ms_log_rms_mae_db": 0.5742748016302247, "全帯域_40ms_rms_envelope_nrmse": 0.04955513999878042, "全帯域_40ms_envelope_correlation": 0.9972438447044392, "全帯域_40ms_modulation_mae_db": 0.29452191380209913, "300–1000 Hz_10ms_log_rms_mae_db": 1.3803083749245089, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10646324685482368, "300–1000 Hz_10ms_envelope_correlation": 0.99089418758907, "300–1000 Hz_10ms_modulation_mae_db": 0.7725733266605276, "300–1000 Hz_40ms_log_rms_mae_db": 0.723582915689329, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06318381703602478, "300–1000 Hz_40ms_envelope_correlation": 0.9969436397609961, "300–1000 Hz_40ms_modulation_mae_db": 0.4780439924106639, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3209862291978263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15813389815749662, "1000–3000 Hz_10ms_envelope_correlation": 0.9817905951630624, "1000–3000 Hz_10ms_modulation_mae_db": 1.4409720157941457, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6411535266003608, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09894897603284039, "1000–3000 Hz_40ms_envelope_correlation": 0.9942516936022882, "1000–3000 Hz_40ms_modulation_mae_db": 1.0691413298641488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3716908786013524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1678482383327557, "3000–6000 Hz_10ms_envelope_correlation": 0.9837180185022514, "3000–6000 Hz_10ms_modulation_mae_db": 0.24802295996382903, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8152651835599865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09354336869402731, "3000–6000 Hz_40ms_envelope_correlation": 0.994962137631633, "3000–6000 Hz_40ms_modulation_mae_db": 0.3240785245113925, "mr_stft_db": 5.450277591825015, "band_mae_db": 0.2557856881246125, "rms_gain_error_db": -0.15347047107506426, "output_peak": 0.33407900827437675}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0199795167021442, "全帯域_10ms_rms_envelope_nrmse": 0.08301225360398078, "全帯域_10ms_envelope_correlation": 0.9920232105481056, "全帯域_10ms_modulation_mae_db": 0.31732517551346395, "全帯域_40ms_log_rms_mae_db": 0.5769560066941152, "全帯域_40ms_rms_envelope_nrmse": 0.0427203257806126, "全帯域_40ms_envelope_correlation": 0.9978279273580842, "全帯域_40ms_modulation_mae_db": 0.28747082824938414, "300–1000 Hz_10ms_log_rms_mae_db": 1.3588512292209642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08965605813884864, "300–1000 Hz_10ms_envelope_correlation": 0.9935412414862874, "300–1000 Hz_10ms_modulation_mae_db": 0.40235742709852734, "300–1000 Hz_40ms_log_rms_mae_db": 0.6752671634930332, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04719523396817334, "300–1000 Hz_40ms_envelope_correlation": 0.9983423873699141, "300–1000 Hz_40ms_modulation_mae_db": 0.37128129763181406, "1000–3000 Hz_10ms_log_rms_mae_db": 1.345174336752759, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16254469335348506, "1000–3000 Hz_10ms_envelope_correlation": 0.9790118838731816, "1000–3000 Hz_10ms_modulation_mae_db": 0.19964590276991298, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6925993606724973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08899794353650836, "1000–3000 Hz_40ms_envelope_correlation": 0.9936553362549033, "1000–3000 Hz_40ms_modulation_mae_db": 0.7310397944090457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3925105537968265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12559797617550567, "3000–6000 Hz_10ms_envelope_correlation": 0.9910946426634107, "3000–6000 Hz_10ms_modulation_mae_db": 0.8245281582158405, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8376546349799948, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06761893859649219, "3000–6000 Hz_40ms_envelope_correlation": 0.9976358423956799, "3000–6000 Hz_40ms_modulation_mae_db": 1.062349546744336, "mr_stft_db": 5.462896209531107, "band_mae_db": 0.1898248191442543, "rms_gain_error_db": -0.09055897878133556, "output_peak": 0.3169079196434241}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0284670217796708, "全帯域_10ms_rms_envelope_nrmse": 0.09437689277641399, "全帯域_10ms_envelope_correlation": 0.9897467114435228, "全帯域_10ms_modulation_mae_db": 0.2722379048350552, "全帯域_40ms_log_rms_mae_db": 0.5203884618237439, "全帯域_40ms_rms_envelope_nrmse": 0.05059511019775505, "全帯域_40ms_envelope_correlation": 0.996806656494853, "全帯域_40ms_modulation_mae_db": 0.14237970425894, "300–1000 Hz_10ms_log_rms_mae_db": 1.3788962457300635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10681762323767376, "300–1000 Hz_10ms_envelope_correlation": 0.9908113089970233, "300–1000 Hz_10ms_modulation_mae_db": 0.6695339243868362, "300–1000 Hz_40ms_log_rms_mae_db": 0.712519699776379, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06796853281004687, "300–1000 Hz_40ms_envelope_correlation": 0.996356583217098, "300–1000 Hz_40ms_modulation_mae_db": 0.319539263096556, "1000–3000 Hz_10ms_log_rms_mae_db": 1.279115750610876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14573076539313412, "1000–3000 Hz_10ms_envelope_correlation": 0.9831218771989955, "1000–3000 Hz_10ms_modulation_mae_db": 0.46473069153456087, "1000–3000 Hz_40ms_log_rms_mae_db": 0.673957352868704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08306010423753443, "1000–3000 Hz_40ms_envelope_correlation": 0.9943440396082248, "1000–3000 Hz_40ms_modulation_mae_db": 0.4744912417723059, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4827678718757356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17646301015126392, "3000–6000 Hz_10ms_envelope_correlation": 0.9831764804837073, "3000–6000 Hz_10ms_modulation_mae_db": 0.3651714567837185, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8993995593613474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09677284242492756, "3000–6000 Hz_40ms_envelope_correlation": 0.9951841136620379, "3000–6000 Hz_40ms_modulation_mae_db": 0.4482054476066428, "mr_stft_db": 5.456179588370482, "band_mae_db": 0.23279274732446753, "rms_gain_error_db": 0.020113810246611335, "output_peak": 0.33800577986730795}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0582666971144667, "全帯域_10ms_rms_envelope_nrmse": 0.08551182949168194, "全帯域_10ms_envelope_correlation": 0.9915410233643076, "全帯域_10ms_modulation_mae_db": 0.23102863437206522, "全帯域_40ms_log_rms_mae_db": 0.6198519033111459, "全帯域_40ms_rms_envelope_nrmse": 0.045437602064023044, "全帯域_40ms_envelope_correlation": 0.9974942830716413, "全帯域_40ms_modulation_mae_db": 0.4124559830196861, "300–1000 Hz_10ms_log_rms_mae_db": 1.386550789031869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08664004887867598, "300–1000 Hz_10ms_envelope_correlation": 0.99389482273288, "300–1000 Hz_10ms_modulation_mae_db": 0.2417422050144137, "300–1000 Hz_40ms_log_rms_mae_db": 0.7438563822949514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04663143017614319, "300–1000 Hz_40ms_envelope_correlation": 0.9982658157992853, "300–1000 Hz_40ms_modulation_mae_db": 0.3578837994993715, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3278164260717416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14582851190134732, "1000–3000 Hz_10ms_envelope_correlation": 0.9831247433745628, "1000–3000 Hz_10ms_modulation_mae_db": 0.31846031717472384, "1000–3000 Hz_40ms_log_rms_mae_db": 0.628539584577338, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07702677725863902, "1000–3000 Hz_40ms_envelope_correlation": 0.9952360778649072, "1000–3000 Hz_40ms_modulation_mae_db": 0.23293507247610332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.49255293545136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15287085013543983, "3000–6000 Hz_10ms_envelope_correlation": 0.9866849848285102, "3000–6000 Hz_10ms_modulation_mae_db": 0.4126376229145214, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9154026388823043, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08418316406931298, "3000–6000 Hz_40ms_envelope_correlation": 0.9962248674624329, "3000–6000 Hz_40ms_modulation_mae_db": 0.4447139016952551, "mr_stft_db": 5.4581435833467395, "band_mae_db": 0.13350616472679439, "rms_gain_error_db": -0.08403685782364013, "output_peak": 0.3397172489310938}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1912192756121125, "全帯域_10ms_rms_envelope_nrmse": 0.12394687631275231, "全帯域_10ms_envelope_correlation": 0.9822784855408895, "全帯域_10ms_modulation_mae_db": 0.6731430369204916, "全帯域_40ms_log_rms_mae_db": 0.6543432408025095, "全帯域_40ms_rms_envelope_nrmse": 0.07383818431823225, "全帯域_40ms_envelope_correlation": 0.993199959204797, "全帯域_40ms_modulation_mae_db": 1.427203120679814, "300–1000 Hz_10ms_log_rms_mae_db": 1.3756379133677326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.098691321906587, "300–1000 Hz_10ms_envelope_correlation": 0.9919963082086036, "300–1000 Hz_10ms_modulation_mae_db": 0.33852260337652357, "300–1000 Hz_40ms_log_rms_mae_db": 0.7298400465335758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055019594901037215, "300–1000 Hz_40ms_envelope_correlation": 0.9974811850019227, "300–1000 Hz_40ms_modulation_mae_db": 0.640925022300308, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3173603234119122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18091244309377044, "1000–3000 Hz_10ms_envelope_correlation": 0.9739575894319064, "1000–3000 Hz_10ms_modulation_mae_db": 0.48758392971136194, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7099845227234888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.103749891383959, "1000–3000 Hz_40ms_envelope_correlation": 0.9911863959399803, "1000–3000 Hz_40ms_modulation_mae_db": 0.3468153202290627, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4986327024543227, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2454050066735976, "3000–6000 Hz_10ms_envelope_correlation": 0.9654149978717557, "3000–6000 Hz_10ms_modulation_mae_db": 1.3948889659655634, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9439682865020396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15609199505948781, "3000–6000 Hz_40ms_envelope_correlation": 0.985651534622802, "3000–6000 Hz_40ms_modulation_mae_db": 2.0764303190281823, "mr_stft_db": 5.457518485338272, "band_mae_db": 0.10987090297589425, "rms_gain_error_db": -0.05860232213589716, "output_peak": 0.31990175779147173}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0281763570034108, "全帯域_10ms_rms_envelope_nrmse": 0.09018589979501798, "全帯域_10ms_envelope_correlation": 0.9905905309069937, "全帯域_10ms_modulation_mae_db": 0.2703116053904114, "全帯域_40ms_log_rms_mae_db": 0.5533022408314253, "全帯域_40ms_rms_envelope_nrmse": 0.04690125002511997, "全帯域_40ms_envelope_correlation": 0.9972998852796701, "全帯域_40ms_modulation_mae_db": 0.39597740093623707, "300–1000 Hz_10ms_log_rms_mae_db": 1.3509795447643467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09109269453873421, "300–1000 Hz_10ms_envelope_correlation": 0.9933617274003503, "300–1000 Hz_10ms_modulation_mae_db": 0.444731686553894, "300–1000 Hz_40ms_log_rms_mae_db": 0.6589335557463167, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04738503180518306, "300–1000 Hz_40ms_envelope_correlation": 0.9983821303261753, "300–1000 Hz_40ms_modulation_mae_db": 0.11556896174134114, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3455997840851723, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16069102381921577, "1000–3000 Hz_10ms_envelope_correlation": 0.9799828837831577, "1000–3000 Hz_10ms_modulation_mae_db": 0.912023112010495, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6778428316409596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08838897131073555, "1000–3000 Hz_40ms_envelope_correlation": 0.9945601314214955, "1000–3000 Hz_40ms_modulation_mae_db": 0.6531295828286386, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4299230035770982, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16161644467521927, "3000–6000 Hz_10ms_envelope_correlation": 0.9849211152146097, "3000–6000 Hz_10ms_modulation_mae_db": 0.2541097262963784, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8544280951634324, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08554357313043057, "3000–6000 Hz_40ms_envelope_correlation": 0.9956981050350223, "3000–6000 Hz_40ms_modulation_mae_db": 0.46961197729371235, "mr_stft_db": 5.4500238551071325, "band_mae_db": 0.15097979966876807, "rms_gain_error_db": -0.06800060683703982, "output_peak": 0.3515332808339596}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83600, "output_samples": 83600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.571993300413566, "spectral_convergence": 0.3552332770384163, "spectral_convergence_status": "ok", "amplitude_floor": 9.102684030105167e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.19339124150198, "spectral_convergence": 0.33612500431884085, "spectral_convergence_status": "ok", "amplitude_floor": 4.6649711782186323e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5859517929413105, "spectral_convergence": 0.3541650951808382, "spectral_convergence_status": "ok", "amplitude_floor": 4.1057122515023654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021084017487574348, "output_power": 0.00021346909443329865, "delta_db": 0.053816399944615775, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00047622157098200406, "output_power": 0.0004694660091304909, "delta_db": -0.06204909958471404, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 9.587972951197478e-05, "output_power": 9.473448637430096e-05, "delta_db": -0.05218695339068141, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.411215555329107e-05, "output_power": 6.2634867066123e-05, "delta_db": -0.7307328488265317, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013227809814888984, "output_power": 0.00012290771196881174, "delta_db": -0.31908808154601265, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00014809096583739848, "output_power": 0.0001575208482862585, "delta_db": 0.26809476346562633, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2476613577930303, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06419722830166434, "pitch": null, "output_rms_dbfs": -29.504978333825026, "output_peak": 0.3298703619418707, "reference_rms_dbfs": -29.44078110552336, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.450445444952286}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011323786448874551, "log_rms_mae_db": 1.0775576084109346, "rms_envelope_nrmse": 0.08706444711883025, "envelope_correlation": 0.991248682693231, "rms_gain_error_db": -0.06426241113926637, "reference_mean_power": 0.0011384943575262435, "output_mean_power": 0.0011217721184167832}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 5.62759551802579e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002928658523286973, "output_power": 0.00029015609128061584, "reference_active": true, "delta_db": -0.04037044428342587}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4610498849864094e-05, "output_power": 4.52811321656786e-05, "reference_active": true, "delta_db": 0.06480197297298192}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2101686887119984e-05, "output_power": 1.1284798518232957e-05, "reference_active": true, "delta_db": -0.3035210255279063}], "active_band_count": 3, "active_band_mae_db": 0.13623114759477137}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.563924945576787e-05, "log_rms_mae_db": 0.6315606000532192, "rms_envelope_nrmse": 0.04166563376420752, "envelope_correlation": 0.9978911731113008, "rms_gain_error_db": -0.06430226987116665, "reference_mean_power": 0.0011449801518647084, "output_mean_power": 0.0011281522951969915}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.095010942049215e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002727818973842863, "output_power": 0.00027115189291142234, "reference_active": true, "delta_db": -0.026029052789688555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1267992599082604e-05, "output_power": 2.1191622661599013e-05, "reference_active": true, "delta_db": -0.015622882117752732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.757168583210849e-07, "output_power": 4.82853476706565e-07, "reference_active": true, "delta_db": -0.7639358397005367}], "active_band_count": 3, "active_band_mae_db": 0.2685292582026593}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.979916609312644e-05, "log_rms_mae_db": 1.3260411552766664, "rms_envelope_nrmse": 0.08518190457897826, "envelope_correlation": 0.9940484030234515, "rms_gain_error_db": -0.10507892734708524, "reference_mean_power": 0.00038382289364078325, "output_mean_power": 0.0003746476238792448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 2.8601579771498626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014700129208541836, "output_power": 0.00014573535336115856, "reference_active": true, "delta_db": -0.0375623372217324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.957621157356578e-05, "output_power": 1.9174578032133307e-05, "reference_active": true, "delta_db": -0.09002834720193842}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3113428408893907e-06, "output_power": 3.1241899086775657e-06, "reference_active": true, "delta_db": 1.308730562181024}], "active_band_count": 3, "active_band_mae_db": 0.4787737488682316}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.59156861629162e-05, "log_rms_mae_db": 0.6930002338238425, "rms_envelope_nrmse": 0.0421066250984882, "envelope_correlation": 0.9985527746314488, "rms_gain_error_db": -0.10507478624699032, "reference_mean_power": 0.00038603855676420007, "output_mean_power": 0.0003768106809686888}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 2.667606930039739e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013762626674571402, "output_power": 0.0001374409874135021, "reference_active": true, "delta_db": -0.005850627622370477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.746730510582489e-06, "output_power": 9.862701377917714e-06, "reference_active": true, "delta_db": 0.05136925458668347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5949673646594045e-07, "output_power": 1.9055259906496118e-07, "reference_active": true, "delta_db": 0.7726307556962303}], "active_band_count": 3, "active_band_mae_db": 0.27661687930176143}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.341026161507385e-05, "log_rms_mae_db": 1.3389580437590778, "rms_envelope_nrmse": 0.16578294229106144, "envelope_correlation": 0.979014838781364, "rms_gain_error_db": -0.43736135978942214, "reference_mean_power": 0.0001433729237288462, "output_mean_power": 0.00012963760698929755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.1455083323281239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.589033152322262e-05, "output_power": 4.527127650366152e-05, "reference_active": true, "delta_db": -0.9151394534833646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4083924473542424e-05, "output_power": 1.3748272831094066e-05, "reference_active": true, "delta_db": -0.10475545438754645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.499188380560451e-06, "output_power": 4.293662460976831e-06, "reference_active": true, "delta_db": -1.0747069704227574}], "active_band_count": 3, "active_band_mae_db": 0.6982006260978895}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 5.482542724448801e-05, "log_rms_mae_db": 0.7101327998871642, "rms_envelope_nrmse": 0.0976189565414837, "envelope_correlation": 0.9936029010384281, "rms_gain_error_db": -0.43719768638487644, "reference_mean_power": 0.00014418079784839864, "output_mean_power": 0.000130372998993887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.042775196620521e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2049448137923216e-05, "output_power": 4.238564503394998e-05, "reference_active": true, "delta_db": -0.8919733246726567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.197217517258297e-06, "output_power": 5.818860185937732e-06, "reference_active": true, "delta_db": -0.27358817594070656}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5366430302416474e-07, "output_power": 2.6222941388685014e-07, "reference_active": true, "delta_db": 0.14422048941982674}], "active_band_count": 3, "active_band_mae_db": 0.43659399667772997}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011141137328067575, "log_rms_mae_db": 1.4832772751080223, "rms_envelope_nrmse": 0.15685430308440176, "envelope_correlation": 0.9859892009119263, "rms_gain_error_db": 0.04862927703596932, "reference_mean_power": 0.00018590462562984253, "output_mean_power": 0.00018799795434209994}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.861645592580093e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.396584441083375e-05, "output_power": 9.355538842622389e-05, "reference_active": true, "delta_db": -0.019012146109257097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7617019162087818e-05, "output_power": 1.621404873835499e-05, "reference_active": true, "delta_db": -0.3604095257872476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.489954892102086e-06, "output_power": 5.610998426833239e-06, "reference_active": true, "delta_db": -0.6320153134329898}], "active_band_count": 3, "active_band_mae_db": 0.33714566177649813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.054702623382987e-05, "log_rms_mae_db": 0.8802823260971696, "rms_envelope_nrmse": 0.08113431355582701, "envelope_correlation": 0.9961897654224277, "rms_gain_error_db": 0.04863912977715806, "reference_mean_power": 0.00018697687803543306, "output_mean_power": 0.00018908270952284527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.7928912528322647e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.176227386223464e-05, "output_power": 9.183654562666705e-05, "reference_active": true, "delta_db": 0.0035137283892709347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.193273720945643e-06, "output_power": 7.08245909551551e-06, "reference_active": true, "delta_db": -0.6327338906708089}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.787097063305561e-07, "output_power": 2.836092955259251e-07, "reference_active": true, "delta_db": -1.255859751318535}], "active_band_count": 3, "active_band_mae_db": 0.6307024567928715}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8439319521360492, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d2d72d0eaa49523b2ce8a66d232fb9d2f0167da8ae77f0a073a764167f5bd83d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9796971752215968, "standard_deviation": 0.003078289366002517, "minimum": 0.9739575894319064, "maximum": 0.9831247433745628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3239300126070974, "standard_deviation": 0.021640745546416845, "minimum": 1.279115750610876, "maximum": 1.3455997840851723, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.594826056296597, "standard_deviation": 0.41683837689641245, "minimum": 0.19964590276991298, "maximum": 1.4409720157941457, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16102498215478184, "standard_deviation": 0.011654465697896046, "minimum": 0.14573076539313412, "maximum": 0.18091244309377044, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938503957390963, "standard_deviation": 0.001197513008864836, "minimum": 0.9911863959399803, "maximum": 0.9952360778649072, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6746429591858467, "standard_deviation": 0.029822969634359847, "minimum": 0.628539584577338, "maximum": 0.7101327998871642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5427123235177098, "standard_deviation": 0.26621982044713044, "minimum": 0.23293507247610332, "maximum": 1.0691413298641488, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09066070564577328, "standard_deviation": 0.008864423530926573, "minimum": 0.07702677725863902, "maximum": 0.103749891383959, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.982554818059652, "standard_deviation": 0.00767744227508342, "minimum": 0.9654149978717557, "maximum": 0.9910946426634107, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4512381061817208, "standard_deviation": 0.04819347336878095, "minimum": 1.3716908786013524, "maximum": 1.4986327024543227, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5590440423705245, "standard_deviation": 0.39151659853402243, "minimum": 0.24802295996382903, "maximum": 1.3948889659655634, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1719122369366325, "standard_deviation": 0.034966575643884897, "minimum": 0.12559797617550567, "maximum": 0.2454050066735976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994074670612731, "standard_deviation": 0.003901002249425284, "minimum": 0.985651534622802, "maximum": 0.9976358423956799, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8795994304791337, "standard_deviation": 0.04220041703580414, "minimum": 0.8152651835599865, "maximum": 0.9439682865020396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8116745716783369, "standard_deviation": 0.5815780785832826, "minimum": 0.3240785245113925, "maximum": 2.0764303190281823, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09871375404347552, "standard_deviation": 0.028447722822773646, "minimum": 0.06761893859649219, "maximum": 0.15609199505948781, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927380573262217, "standard_deviation": 0.0013165978091152213, "minimum": 0.9908113089970233, "maximum": 0.9940484030234515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3680062074771033, "standard_deviation": 0.021245127068888082, "minimum": 1.3260411552766664, "maximum": 1.3867844075006759, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4604424656256244, "standard_deviation": 0.17866782433190292, "minimum": 0.2417422050144137, "maximum": 0.7725733266605276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09447603967569207, "standard_deviation": 0.008506698652395908, "minimum": 0.08518190457897826, "maximum": 0.10681762323767376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977741590205012, "standard_deviation": 0.0007869059661599412, "minimum": 0.996356583217098, "maximum": 0.9985527746314488, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.708383192611641, "standard_deviation": 0.029798331144679387, "minimum": 0.6589335557463167, "maximum": 0.7438563822949514, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.38171802451815784, "standard_deviation": 0.15832748790918985, "minimum": 0.11556896174134114, "maximum": 0.640925022300308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05285010790066699, "standard_deviation": 0.008913954324957205, "minimum": 0.0421066250984882, "maximum": 0.06796853281004687, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18648610168598367, "standard_deviation": 0.05459372069114614, "minimum": 0.10987090297589425, "maximum": 0.2557856881246125, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.457044755386968, "standard_deviation": 0.007233716047838505, "minimum": 5.4500238551071325, "maximum": 5.470873284624709, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3384226596604744, "standard_deviation": 0.019228819929860398, "minimum": 0.3169079196434241, "maximum": 0.3773659200002907, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07361728839628645, "standard_deviation": 0.04807417257357457, "minimum": -0.15347047107506426, "maximum": 0.020113810246611335, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9895031020245665, "standard_deviation": 0.0031060958142113907, "minimum": 0.9822784855408895, "maximum": 0.9920232105481056, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0679220734727797, "standard_deviation": 0.05828457237193058, "minimum": 1.0199795167021442, "maximum": 1.1912192756121125, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31422084541839357, "standard_deviation": 0.1621671554704752, "minimum": 0.13623114759477137, "maximum": 0.6731430369204916, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09460946774650728, "standard_deviation": 0.012913116159947065, "minimum": 0.08301225360398078, "maximum": 0.12394687631275231, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9967258019739461, "standard_deviation": 0.0015430037823302959, "minimum": 0.993199959204797, "maximum": 0.9978911731113008, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5807196476333203, "standard_deviation": 0.051119481364979284, "minimum": 0.5150799259201784, "maximum": 0.6543432408025095, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47600437768840553, "standard_deviation": 0.40510396902951956, "minimum": 0.14237970425894, "maximum": 1.427203120679814, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05093292805752726, "standard_deviation": 0.010421723198181687, "minimum": 0.04166563376420752, "maximum": 0.07383818431823225, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0056", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.0131614208221436, "predicted_mos": 3.0131614208221436, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "24b8ac76b7982f67d8ca0caad14e1963dab1c30ea1924bc6d4b98d2720bcae34", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9086198997683823, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9698899213993668, "全帯域_10ms_rms_envelope_nrmse": 0.08435571442057342, "全帯域_10ms_envelope_correlation": 0.991765479202302, "全帯域_10ms_modulation_mae_db": 0.5541652991922956, "全帯域_40ms_log_rms_mae_db": 0.5166416264671377, "全帯域_40ms_rms_envelope_nrmse": 0.04112016730125204, "全帯域_40ms_envelope_correlation": 0.9978856913812815, "全帯域_40ms_modulation_mae_db": 0.40484473603935306, "300–1000 Hz_10ms_log_rms_mae_db": 1.7348601622205737, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09525761498246653, "300–1000 Hz_10ms_envelope_correlation": 0.9926716780887339, "300–1000 Hz_10ms_modulation_mae_db": 0.6101615211389014, "300–1000 Hz_40ms_log_rms_mae_db": 1.0728075092073608, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05153572616969835, "300–1000 Hz_40ms_envelope_correlation": 0.9979443636127813, "300–1000 Hz_40ms_modulation_mae_db": 0.3735875341566524, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5625098253154948, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1524908798336197, "1000–3000 Hz_10ms_envelope_correlation": 0.9813897092867022, "1000–3000 Hz_10ms_modulation_mae_db": 0.6576288678682035, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8397813929253369, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08019036980905815, "1000–3000 Hz_40ms_envelope_correlation": 0.9945752617912341, "1000–3000 Hz_40ms_modulation_mae_db": 0.29762094387925775, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9007019321294485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17807500137235616, "3000–6000 Hz_10ms_envelope_correlation": 0.9817726745386531, "3000–6000 Hz_10ms_modulation_mae_db": 0.5022561765133827, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2463439482956788, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10250324730330847, "3000–6000 Hz_40ms_envelope_correlation": 0.9938186293579554, "3000–6000 Hz_40ms_modulation_mae_db": 0.366302740303371, "mr_stft_db": 5.699313031915058, "band_mae_db": 0.12721563117548942, "rms_gain_error_db": -0.007296662155227755, "output_peak": 0.34340288775570926}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9818042194125489, "全帯域_10ms_rms_envelope_nrmse": 0.07826573138032955, "全帯域_10ms_envelope_correlation": 0.992990866765819, "全帯域_10ms_modulation_mae_db": 0.43026408177469655, "全帯域_40ms_log_rms_mae_db": 0.5337030549236286, "全帯域_40ms_rms_envelope_nrmse": 0.044705552697211964, "全帯域_40ms_envelope_correlation": 0.9976485840469147, "全帯域_40ms_modulation_mae_db": 0.2330539699293442, "300–1000 Hz_10ms_log_rms_mae_db": 1.6878729061827864, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09468228794425596, "300–1000 Hz_10ms_envelope_correlation": 0.9930387514168547, "300–1000 Hz_10ms_modulation_mae_db": 0.7959022253777315, "300–1000 Hz_40ms_log_rms_mae_db": 1.0350598366245762, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061149092704495894, "300–1000 Hz_40ms_envelope_correlation": 0.997297889595754, "300–1000 Hz_40ms_modulation_mae_db": 0.4541502027595194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6333910658834745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14556803591283057, "1000–3000 Hz_10ms_envelope_correlation": 0.9843077143157151, "1000–3000 Hz_10ms_modulation_mae_db": 0.5247876413056892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9993209283414809, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09029786237667886, "1000–3000 Hz_40ms_envelope_correlation": 0.9948407605740305, "1000–3000 Hz_40ms_modulation_mae_db": 0.5802094599852139, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8379001574554392, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16745489542040642, "3000–6000 Hz_10ms_envelope_correlation": 0.9839178713033753, "3000–6000 Hz_10ms_modulation_mae_db": 0.37494433488411055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1983317692930064, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10731341194337035, "3000–6000 Hz_40ms_envelope_correlation": 0.9932253489139452, "3000–6000 Hz_40ms_modulation_mae_db": 0.30236100717746, "mr_stft_db": 5.704438929035356, "band_mae_db": 0.25822000992446514, "rms_gain_error_db": -0.10137615436401948, "output_peak": 0.35679254452974274}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0416610285256562, "全帯域_10ms_rms_envelope_nrmse": 0.09540502146011165, "全帯域_10ms_envelope_correlation": 0.9894296418786086, "全帯域_10ms_modulation_mae_db": 0.19953736625521445, "全帯域_40ms_log_rms_mae_db": 0.5852537900660825, "全帯域_40ms_rms_envelope_nrmse": 0.055829348475443366, "全帯域_40ms_envelope_correlation": 0.9961339453300949, "全帯域_40ms_modulation_mae_db": 0.1650358750389857, "300–1000 Hz_10ms_log_rms_mae_db": 1.7911229345691293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1147728136140414, "300–1000 Hz_10ms_envelope_correlation": 0.9894951643353423, "300–1000 Hz_10ms_modulation_mae_db": 1.0737303348887481, "300–1000 Hz_40ms_log_rms_mae_db": 1.1061694110356528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07426963609294687, "300–1000 Hz_40ms_envelope_correlation": 0.9957703329498223, "300–1000 Hz_40ms_modulation_mae_db": 0.7811838747168814, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6367378647241966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16764738865137407, "1000–3000 Hz_10ms_envelope_correlation": 0.9782085880225388, "1000–3000 Hz_10ms_modulation_mae_db": 1.4107200607914152, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9476194736432593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10344741327324938, "1000–3000 Hz_40ms_envelope_correlation": 0.9921397398366066, "1000–3000 Hz_40ms_modulation_mae_db": 1.315623324343973, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8004582970182235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18836749243445, "3000–6000 Hz_10ms_envelope_correlation": 0.9797229581265678, "3000–6000 Hz_10ms_modulation_mae_db": 0.1273271487375039, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1756657050219617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11629019563889517, "3000–6000 Hz_40ms_envelope_correlation": 0.9920833945431676, "3000–6000 Hz_40ms_modulation_mae_db": 0.08561864230962211, "mr_stft_db": 5.698089620564564, "band_mae_db": 0.23336230069063787, "rms_gain_error_db": -0.06977763614653917, "output_peak": 0.33158470471976026}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0140325816981524, "全帯域_10ms_rms_envelope_nrmse": 0.09519797166495245, "全帯域_10ms_envelope_correlation": 0.9895123448998118, "全帯域_10ms_modulation_mae_db": 0.13138251642312734, "全帯域_40ms_log_rms_mae_db": 0.5060431571640773, "全帯域_40ms_rms_envelope_nrmse": 0.04781195375638573, "全帯域_40ms_envelope_correlation": 0.9971426963007154, "全帯域_40ms_modulation_mae_db": 0.16527636400178827, "300–1000 Hz_10ms_log_rms_mae_db": 1.7270802181691014, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10317656322860802, "300–1000 Hz_10ms_envelope_correlation": 0.9912680976298448, "300–1000 Hz_10ms_modulation_mae_db": 0.557832409687878, "300–1000 Hz_40ms_log_rms_mae_db": 1.0868351545249384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06792321262735014, "300–1000 Hz_40ms_envelope_correlation": 0.9961073957609038, "300–1000 Hz_40ms_modulation_mae_db": 0.16932789382592106, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6572343816375916, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16460969029720687, "1000–3000 Hz_10ms_envelope_correlation": 0.9795086870344586, "1000–3000 Hz_10ms_modulation_mae_db": 1.28395837983818, "1000–3000 Hz_40ms_log_rms_mae_db": 0.993058353646547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09435360722934134, "1000–3000 Hz_40ms_envelope_correlation": 0.9943987491507124, "1000–3000 Hz_40ms_modulation_mae_db": 1.1294653850685343, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8465346677424455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18863961611118818, "3000–6000 Hz_10ms_envelope_correlation": 0.9798326936289238, "3000–6000 Hz_10ms_modulation_mae_db": 0.44734694900344985, "3000–6000 Hz_40ms_log_rms_mae_db": 1.181367950661148, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09294199926768065, "3000–6000 Hz_40ms_envelope_correlation": 0.995033742970282, "3000–6000 Hz_40ms_modulation_mae_db": 0.3609666131541304, "mr_stft_db": 5.716816024027755, "band_mae_db": 0.20024636359077086, "rms_gain_error_db": -0.016403970794022744, "output_peak": 0.3482001833422786}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9932906145676053, "全帯域_10ms_rms_envelope_nrmse": 0.09146451330379649, "全帯域_10ms_envelope_correlation": 0.9903278967962567, "全帯域_10ms_modulation_mae_db": 0.4337346945976602, "全帯域_40ms_log_rms_mae_db": 0.5722402458398014, "全帯域_40ms_rms_envelope_nrmse": 0.0483387934196649, "全帯域_40ms_envelope_correlation": 0.9971221207361939, "全帯域_40ms_modulation_mae_db": 0.2982281703739938, "300–1000 Hz_10ms_log_rms_mae_db": 1.7355780630190818, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09379058421273091, "300–1000 Hz_10ms_envelope_correlation": 0.9928295012637853, "300–1000 Hz_10ms_modulation_mae_db": 0.6345466508672412, "300–1000 Hz_40ms_log_rms_mae_db": 1.064566373079552, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0501852144197678, "300–1000 Hz_40ms_envelope_correlation": 0.9979798524016695, "300–1000 Hz_40ms_modulation_mae_db": 0.34613635736353593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6052553067894308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1757067334505763, "1000–3000 Hz_10ms_envelope_correlation": 0.9755469585242612, "1000–3000 Hz_10ms_modulation_mae_db": 0.992884252834496, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9242328873213097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10533310304745364, "1000–3000 Hz_40ms_envelope_correlation": 0.9913116847405747, "1000–3000 Hz_40ms_modulation_mae_db": 0.8914871534891202, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7950280599799064, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1849146525583566, "3000–6000 Hz_10ms_envelope_correlation": 0.9805445431850323, "3000–6000 Hz_10ms_modulation_mae_db": 0.5817330761483432, "3000–6000 Hz_40ms_log_rms_mae_db": 1.113574867100846, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09853198669327776, "3000–6000 Hz_40ms_envelope_correlation": 0.9943636608205316, "3000–6000 Hz_40ms_modulation_mae_db": 0.6878590330076914, "mr_stft_db": 5.700954720199935, "band_mae_db": 0.11533196048574575, "rms_gain_error_db": -0.06354340811354627, "output_peak": 0.33890629248244086}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0146370182033846, "全帯域_10ms_rms_envelope_nrmse": 0.0953916444417923, "全帯域_10ms_envelope_correlation": 0.989483781983041, "全帯域_10ms_modulation_mae_db": 0.4570101187895146, "全帯域_40ms_log_rms_mae_db": 0.6056356998590958, "全帯域_40ms_rms_envelope_nrmse": 0.056286491128537654, "全帯域_40ms_envelope_correlation": 0.9960947403655316, "全帯域_40ms_modulation_mae_db": 0.5078071887358658, "300–1000 Hz_10ms_log_rms_mae_db": 1.7467818831385409, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.096165065458866, "300–1000 Hz_10ms_envelope_correlation": 0.9926145440959991, "300–1000 Hz_10ms_modulation_mae_db": 0.5216944776126642, "300–1000 Hz_40ms_log_rms_mae_db": 1.0808348628422024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06151228062881325, "300–1000 Hz_40ms_envelope_correlation": 0.9970430967386024, "300–1000 Hz_40ms_modulation_mae_db": 0.39788908535808803, "1000–3000 Hz_10ms_log_rms_mae_db": 1.708445759971783, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17570298057542857, "1000–3000 Hz_10ms_envelope_correlation": 0.9756986764668962, "1000–3000 Hz_10ms_modulation_mae_db": 0.7209670467535121, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9881433648083835, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10952064066583299, "1000–3000 Hz_40ms_envelope_correlation": 0.9907269706414736, "1000–3000 Hz_40ms_modulation_mae_db": 0.20605375630977965, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8225139249106046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1771136112579024, "3000–6000 Hz_10ms_envelope_correlation": 0.9825606504858795, "3000–6000 Hz_10ms_modulation_mae_db": 0.7736985464397733, "3000–6000 Hz_40ms_log_rms_mae_db": 1.178179788844682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11329819453773407, "3000–6000 Hz_40ms_envelope_correlation": 0.9928236864296679, "3000–6000 Hz_40ms_modulation_mae_db": 0.8725689798524642, "mr_stft_db": 5.712318853822999, "band_mae_db": 0.2314889369879518, "rms_gain_error_db": -0.04833774993339416, "output_peak": 0.3172731929201499}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0080601342413407, "全帯域_10ms_rms_envelope_nrmse": 0.106896195791767, "全帯域_10ms_envelope_correlation": 0.9867198074437747, "全帯域_10ms_modulation_mae_db": 0.15536195206946554, "全帯域_40ms_log_rms_mae_db": 0.5558261124211787, "全帯域_40ms_rms_envelope_nrmse": 0.04646982739074608, "全帯域_40ms_envelope_correlation": 0.9972869210385574, "全帯域_40ms_modulation_mae_db": 0.21070564768645117, "300–1000 Hz_10ms_log_rms_mae_db": 1.7128490984907676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10743585413478737, "300–1000 Hz_10ms_envelope_correlation": 0.990626250014311, "300–1000 Hz_10ms_modulation_mae_db": 0.8047242781443312, "300–1000 Hz_40ms_log_rms_mae_db": 1.0232761751712058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059993862677190424, "300–1000 Hz_40ms_envelope_correlation": 0.9971645659277035, "300–1000 Hz_40ms_modulation_mae_db": 0.3154847451035545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6455709412901924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15234125664386047, "1000–3000 Hz_10ms_envelope_correlation": 0.9815388447513391, "1000–3000 Hz_10ms_modulation_mae_db": 0.828242264028127, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9570966305198041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08011043128386863, "1000–3000 Hz_40ms_envelope_correlation": 0.9948667295414655, "1000–3000 Hz_40ms_modulation_mae_db": 0.5308473310822235, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8338510270013006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22296530243958915, "3000–6000 Hz_10ms_envelope_correlation": 0.9718334528317432, "3000–6000 Hz_10ms_modulation_mae_db": 0.14359579498481853, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1463804831072284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09787191978576582, "3000–6000 Hz_40ms_envelope_correlation": 0.9945024446709783, "3000–6000 Hz_40ms_modulation_mae_db": 0.536551614723456, "mr_stft_db": 5.699048833499987, "band_mae_db": 0.20273799254815417, "rms_gain_error_db": -0.030423401932588426, "output_peak": 0.3576080150366527}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.032947007605496, "全帯域_10ms_rms_envelope_nrmse": 0.08883352601330248, "全帯域_10ms_envelope_correlation": 0.99091648656011, "全帯域_10ms_modulation_mae_db": 0.4105004881910004, "全帯域_40ms_log_rms_mae_db": 0.5568700108416061, "全帯域_40ms_rms_envelope_nrmse": 0.0444497112814246, "全帯域_40ms_envelope_correlation": 0.9976508863714034, "全帯域_40ms_modulation_mae_db": 0.4819964847647664, "300–1000 Hz_10ms_log_rms_mae_db": 1.7387763184797254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10872796654695416, "300–1000 Hz_10ms_envelope_correlation": 0.9904669999125075, "300–1000 Hz_10ms_modulation_mae_db": 0.65166885364736, "300–1000 Hz_40ms_log_rms_mae_db": 1.0769266278891303, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07089451371241298, "300–1000 Hz_40ms_envelope_correlation": 0.9959810617248002, "300–1000 Hz_40ms_modulation_mae_db": 0.20809336440414264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6063401573895177, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15758628857185727, "1000–3000 Hz_10ms_envelope_correlation": 0.9805802052768668, "1000–3000 Hz_10ms_modulation_mae_db": 1.5248718208283016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8978778226062761, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08382269000488916, "1000–3000 Hz_40ms_envelope_correlation": 0.9949133447942994, "1000–3000 Hz_40ms_modulation_mae_db": 0.8250587597800401, "3000–6000 Hz_10ms_log_rms_mae_db": 1.87112414752021, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17851310250975638, "3000–6000 Hz_10ms_envelope_correlation": 0.9815909081000297, "3000–6000 Hz_10ms_modulation_mae_db": 0.5028143325656252, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1460684032876287, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09413644939563581, "3000–6000 Hz_40ms_envelope_correlation": 0.9948018371356357, "3000–6000 Hz_40ms_modulation_mae_db": 0.831688099425265, "mr_stft_db": 5.71210198089133, "band_mae_db": 0.19175689315157937, "rms_gain_error_db": -0.09025959831277183, "output_peak": 0.3076100081878034}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83600, "output_samples": 83600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.831214881480257, "spectral_convergence": 0.3661608291834828, "spectral_convergence_status": "ok", "amplitude_floor": 9.102684030105167e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.50956078815162, "spectral_convergence": 0.35268255555708067, "spectral_convergence_status": "ok", "amplitude_floor": 4.6649711782186323e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.757163426113297, "spectral_convergence": 0.370844327470466, "spectral_convergence_status": "ok", "amplitude_floor": 4.1057122515023654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021084017487574348, "output_power": 0.00021712173106204095, "delta_db": 0.1274992510330519, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00047622157098200406, "output_power": 0.00047278277983439556, "delta_db": -0.031474135077938274, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 9.587972951197478e-05, "output_power": 9.128378509008971e-05, "delta_db": -0.21333160325395661, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.411215555329107e-05, "output_power": 6.983187741868511e-05, "delta_db": -0.258357267781782, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013227809814888984, "output_power": 0.00013206069608150464, "delta_db": -0.0071436006797653065, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00014809096583739848, "output_power": 0.00015243242491161816, "delta_db": 0.1254879292264424, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12721563117548942, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.007296662155227755, "pitch": null, "output_rms_dbfs": -29.448077767678583, "output_peak": 0.34340288775570926, "reference_rms_dbfs": -29.44078110552336, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.699313031915058}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011323786448874551, "log_rms_mae_db": 0.9698899213993668, "rms_envelope_nrmse": 0.08435571442057342, "envelope_correlation": 0.991765479202302, "rms_gain_error_db": -0.007368655854211719, "reference_mean_power": 0.0011384943575262435, "output_mean_power": 0.0011365643168399794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 5.62759551802579e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002928658523286973, "output_power": 0.00028672861394262936, "reference_active": true, "delta_db": -0.0919770130336271}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4610498849864094e-05, "output_power": 5.167569911154423e-05, "reference_active": true, "delta_db": 0.6384928113528553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2101686887119984e-05, "output_power": 9.764349247660462e-06, "reference_active": true, "delta_db": -0.9320260731904044}], "active_band_count": 3, "active_band_mae_db": 0.5541652991922956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.563924945576787e-05, "log_rms_mae_db": 0.5166416264671377, "rms_envelope_nrmse": 0.04112016730125204, "envelope_correlation": 0.9978856913812815, "rms_gain_error_db": -0.007324151026205024, "reference_mean_power": 0.0011449801518647084, "output_mean_power": 0.0011430508295745092}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.095010942049215e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002727818973842863, "output_power": 0.0002690666030272736, "reference_active": true, "delta_db": -0.05955750080636099}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1267992599082604e-05, "output_power": 2.442247552431778e-05, "reference_active": true, "delta_db": 0.6006318252008951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.757168583210849e-07, "output_power": 5.067274819991483e-07, "reference_active": true, "delta_db": -0.554344882110803}], "active_band_count": 3, "active_band_mae_db": 0.40484473603935306}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.979916609312644e-05, "log_rms_mae_db": 1.7348601622205737, "rms_envelope_nrmse": 0.09525761498246653, "envelope_correlation": 0.9926716780887339, "rms_gain_error_db": -0.14625123556542302, "reference_mean_power": 0.00038382289364078325, "output_mean_power": 0.0003711126443967491}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 2.8601579771498626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014700129208541836, "output_power": 0.00013825324464326132, "reference_active": true, "delta_db": -0.2664581956418858}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.957621157356578e-05, "output_power": 1.7329565027994376e-05, "reference_active": true, "delta_db": -0.5294098802674586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3113428408893907e-06, "output_power": 2.933094245115445e-06, "reference_active": true, "delta_db": 1.03461648750736}], "active_band_count": 3, "active_band_mae_db": 0.6101615211389014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.59156861629162e-05, "log_rms_mae_db": 1.0728075092073608, "rms_envelope_nrmse": 0.05153572616969835, "envelope_correlation": 0.9979443636127813, "rms_gain_error_db": -0.14626031958274258, "reference_mean_power": 0.00038603855676420007, "output_mean_power": 0.0003732541553744862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 2.667606930039739e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013762626674571402, "output_power": 0.00012987765458284797, "reference_active": true, "delta_db": -0.251668921046082}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.746730510582489e-06, "output_power": 8.870547398263328e-06, "reference_active": true, "delta_db": -0.4090853751680685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5949673646594045e-07, "output_power": 1.7731792146252073e-07, "reference_active": true, "delta_db": 0.46000830625580685}], "active_band_count": 3, "active_band_mae_db": 0.3735875341566524}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.341026161507385e-05, "log_rms_mae_db": 1.5625098253154948, "rms_envelope_nrmse": 0.1524908798336197, "envelope_correlation": 0.9813897092867022, "rms_gain_error_db": -0.10732610614275169, "reference_mean_power": 0.0001433729237288462, "output_mean_power": 0.0001398732065482714}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.1455083323281239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.589033152322262e-05, "output_power": 5.4087045859753345e-05, "reference_active": true, "delta_db": -0.14243424035211205}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4083924473542424e-05, "output_power": 1.4874109837910558e-05, "reference_active": true, "delta_db": 0.23707296755591256}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.499188380560451e-06, "output_power": 3.8103130448921134e-06, "reference_active": true, "delta_db": -1.5933793956965858}], "active_band_count": 3, "active_band_mae_db": 0.6576288678682035}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 5.482542724448801e-05, "log_rms_mae_db": 0.8397813929253369, "rms_envelope_nrmse": 0.08019036980905815, "envelope_correlation": 0.9945752617912341, "rms_gain_error_db": -0.10746592621979767, "reference_mean_power": 0.00014418079784839864, "output_mean_power": 0.0001406568320638887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.042775196620521e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2049448137923216e-05, "output_power": 5.142576539791076e-05, "reference_active": true, "delta_db": -0.05235364940450454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.197217517258297e-06, "output_power": 6.517753200633184e-06, "reference_active": true, "delta_db": 0.21901171772853695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5366430302416474e-07, "output_power": 2.198414646885702e-07, "reference_active": true, "delta_db": -0.6214974645047318}], "active_band_count": 3, "active_band_mae_db": 0.29762094387925775}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011141137328067575, "log_rms_mae_db": 1.9007019321294485, "rms_envelope_nrmse": 0.17807500137235616, "envelope_correlation": 0.9817726745386531, "rms_gain_error_db": -0.005370083666643065, "reference_mean_power": 0.00018590462562984253, "output_mean_power": 0.0001856748952343668}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.861645592580093e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.396584441083375e-05, "output_power": 8.925574699610032e-05, "reference_active": true, "delta_db": -0.2233383181573306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7617019162087818e-05, "output_power": 2.073099322610048e-05, "reference_active": true, "delta_db": 0.7068768300689305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.489954892102086e-06, "output_power": 5.68311363780467e-06, "reference_active": true, "delta_db": -0.5765533813138872}], "active_band_count": 3, "active_band_mae_db": 0.5022561765133827}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.054702623382987e-05, "log_rms_mae_db": 1.2463439482956788, "rms_envelope_nrmse": 0.10250324730330847, "envelope_correlation": 0.9938186293579554, "rms_gain_error_db": -0.005389932647489985, "reference_mean_power": 0.00018697687803543306, "output_mean_power": 0.00018674496911068309}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.7928912528322647e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.176227386223464e-05, "output_power": 8.920772243882705e-05, "reference_active": true, "delta_db": -0.12261715266810178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.193273720945643e-06, "output_power": 9.451842245793735e-06, "reference_active": true, "delta_db": 0.6205900049510872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.787097063305561e-07, "output_power": 4.1103285966679846e-07, "reference_active": true, "delta_db": 0.35570106329092416}], "active_band_count": 3, "active_band_mae_db": 0.366302740303371}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8266551366572847, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "40422b71bff3de575af9ad142e782f0a6a25639d72855b60fe74e60cfe7a8235", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9795974229598472, "standard_deviation": 0.003016083967605025, "minimum": 0.9755469585242612, "maximum": 0.9843077143157151, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6319356628752102, "standard_deviation": 0.0429149299853195, "minimum": 1.5625098253154948, "maximum": 1.708445759971783, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9930075417809906, "standard_deviation": 0.3731866869152125, "minimum": 0.5247876413056892, "maximum": 1.5248718208283016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16145665674209422, "standard_deviation": 0.0112452714334693, "minimum": 0.14556803591283057, "maximum": 0.1757067334505763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934716551337996, "standard_deviation": 0.0017707692347046135, "minimum": 0.9907269706414736, "maximum": 0.9949133447942994, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9433913567265497, "standard_deviation": 0.05474216790790929, "minimum": 0.8397813929253369, "maximum": 0.9993209283414809, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7220457642422677, "standard_deviation": 0.3892677534934039, "minimum": 0.20605375630977965, "maximum": 1.315623324343973, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09338451471129652, "standard_deviation": 0.011689965518186587, "minimum": 0.08011043128386863, "maximum": 0.10952064066583299, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9802219690250256, "standard_deviation": 0.0036699476763879142, "minimum": 0.9718334528317432, "maximum": 0.9839178713033753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.838514026719697, "standard_deviation": 0.035076401848468115, "minimum": 1.7950280599799064, "maximum": 1.9007019321294485, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4317145449096259, "standard_deviation": 0.21667000861165217, "minimum": 0.1273271487375039, "maximum": 0.7736985464397733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18575545926300066, "standard_deviation": 0.016567842353728057, "minimum": 0.16745489542040642, "maximum": 0.22296530243958915, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9938315931052705, "standard_deviation": 0.0010391926681505702, "minimum": 0.9920833945431676, "maximum": 0.995033742970282, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1732391144515226, "standard_deviation": 0.03980271110026369, "minimum": 1.113574867100846, "maximum": 1.2463439482956788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5054895912441826, "standard_deviation": 0.2759888488863665, "minimum": 0.08561864230962211, "maximum": 0.8725689798524642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10286092557070851, "standard_deviation": 0.008679745887961325, "minimum": 0.09294199926768065, "maximum": 0.11629019563889517, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916263733446723, "standard_deviation": 0.001337710133596815, "minimum": 0.9894951643353423, "maximum": 0.9930387514168547, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7343651980337134, "standard_deviation": 0.02945780375128826, "minimum": 1.6878729061827864, "maximum": 1.7911229345691293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.706282593920607, "standard_deviation": 0.1797690909964564, "minimum": 0.5216944776126642, "maximum": 1.0737303348887481, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10175109376533878, "standard_deviation": 0.007921914562017825, "minimum": 0.09379058421273091, "maximum": 0.1147728136140414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969110698390046, "standard_deviation": 0.0008663058556913263, "minimum": 0.9957703329498223, "maximum": 0.9979798524016695, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0683094937968274, "standard_deviation": 0.02719111171025727, "minimum": 1.0232761751712058, "maximum": 1.1061694110356528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.38073163221103695, "standard_deviation": 0.1874822668941573, "minimum": 0.16932789382592106, "maximum": 0.7811838747168814, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06218294237908446, "standard_deviation": 0.008606888719598537, "minimum": 0.0501852144197678, "maximum": 0.07426963609294687, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1950450110693493, "standard_deviation": 0.050482667269190214, "minimum": 0.11533196048574575, "maximum": 0.25822000992446514, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.705385249244623, "standard_deviation": 0.007315592582029931, "minimum": 5.698089620564564, "maximum": 5.716816024027755, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3376722286218172, "standard_deviation": 0.01798864308768612, "minimum": 0.3076100081878034, "maximum": 0.3576080150366527, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05342732271901373, "standard_deviation": 0.03397315214973786, "minimum": -0.10137615436401948, "maximum": -0.007296662155227755, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901432881912154, "standard_deviation": 0.0018683328256913307, "minimum": 0.9867198074437747, "maximum": 0.992990866765819, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0070403157066938, "standard_deviation": 0.024446614660071506, "minimum": 0.9698899213993668, "maximum": 1.0416610285256562, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3464945646616218, "standard_deviation": 0.15967245010480133, "minimum": 0.13138251642312734, "maximum": 0.5541652991922956, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09197628980957817, "standard_deviation": 0.008557328836084264, "minimum": 0.07826573138032955, "maximum": 0.106896195791767, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971206981963365, "standard_deviation": 0.0006760562326940748, "minimum": 0.9960947403655316, "maximum": 0.9978856913812815, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5540267121978261, "standard_deviation": 0.034005861321347204, "minimum": 0.5060431571640773, "maximum": 0.6056356998590958, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.30836855457131856, "standard_deviation": 0.1391622229054505, "minimum": 0.1650358750389857, "maximum": 0.5078071887358658, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048126480681333295, "standard_deviation": 0.00538673311326959, "minimum": 0.04112016730125204, "maximum": 0.056286491128537654, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0056", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.550478935241699, "predicted_mos": 2.550478935241699, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "c61f5539d91fbff9c28ac35e0683c23b68d360cb61b113d0d7894bb4ed5a65a8", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.730574599932879, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0571128390969573, "全帯域_10ms_rms_envelope_nrmse": 0.10801027897355288, "全帯域_10ms_envelope_correlation": 0.9864218707693515, "全帯域_10ms_modulation_mae_db": 0.28941653344199986, "全帯域_40ms_log_rms_mae_db": 0.5713562714125923, "全帯域_40ms_rms_envelope_nrmse": 0.04650125858732039, "全帯域_40ms_envelope_correlation": 0.9974090389252529, "全帯域_40ms_modulation_mae_db": 0.2934144533582374, "300–1000 Hz_10ms_log_rms_mae_db": 1.3519620674469104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09891594739852819, "300–1000 Hz_10ms_envelope_correlation": 0.9919156710000535, "300–1000 Hz_10ms_modulation_mae_db": 0.6547875016249812, "300–1000 Hz_40ms_log_rms_mae_db": 0.6968701969442621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05140042116264579, "300–1000 Hz_40ms_envelope_correlation": 0.9977609295039629, "300–1000 Hz_40ms_modulation_mae_db": 0.4876102734583226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2105292490266406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15900746485756645, "1000–3000 Hz_10ms_envelope_correlation": 0.9811016685523298, "1000–3000 Hz_10ms_modulation_mae_db": 0.7858955211751963, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6416296225003673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09448466336013234, "1000–3000 Hz_40ms_envelope_correlation": 0.9944907991048793, "1000–3000 Hz_40ms_modulation_mae_db": 0.7495641049322211, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3398339928323428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20990304617953018, "3000–6000 Hz_10ms_envelope_correlation": 0.974471040109458, "3000–6000 Hz_10ms_modulation_mae_db": 0.44260488364210976, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8066921867319421, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08701880730133889, "3000–6000 Hz_40ms_envelope_correlation": 0.9961038936382057, "3000–6000 Hz_40ms_modulation_mae_db": 0.5571629385639792, "mr_stft_db": 5.419821418533165, "band_mae_db": 0.21835914745688934, "rms_gain_error_db": -0.09885289822448685, "output_peak": 0.3177117532249673}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1117304704831936, "全帯域_10ms_rms_envelope_nrmse": 0.10496582044069032, "全帯域_10ms_envelope_correlation": 0.9872521174120864, "全帯域_10ms_modulation_mae_db": 0.2316712571895743, "全帯域_40ms_log_rms_mae_db": 0.5591931317067723, "全帯域_40ms_rms_envelope_nrmse": 0.06182842260590746, "全帯域_40ms_envelope_correlation": 0.9952648400081017, "全帯域_40ms_modulation_mae_db": 0.12370545654273575, "300–1000 Hz_10ms_log_rms_mae_db": 1.2799608050875289, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09892626343913039, "300–1000 Hz_10ms_envelope_correlation": 0.9920537438774791, "300–1000 Hz_10ms_modulation_mae_db": 0.5408353401486391, "300–1000 Hz_40ms_log_rms_mae_db": 0.7114302288411566, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06521159970770199, "300–1000 Hz_40ms_envelope_correlation": 0.9965180215409044, "300–1000 Hz_40ms_modulation_mae_db": 0.5246826300780819, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1581501977442692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1562022447570387, "1000–3000 Hz_10ms_envelope_correlation": 0.9806373503151325, "1000–3000 Hz_10ms_modulation_mae_db": 0.23691445607873682, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6522666002830787, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08981961921407366, "1000–3000 Hz_40ms_envelope_correlation": 0.9933788837349568, "1000–3000 Hz_40ms_modulation_mae_db": 0.581300488689844, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3063122153867306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20645273760406135, "3000–6000 Hz_10ms_envelope_correlation": 0.9752688525527666, "3000–6000 Hz_10ms_modulation_mae_db": 0.5196284671146719, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8270450797898816, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13819750041186848, "3000–6000 Hz_40ms_envelope_correlation": 0.9887601525285151, "3000–6000 Hz_40ms_modulation_mae_db": 0.42726392631468174, "mr_stft_db": 5.403401408616889, "band_mae_db": 0.19526144681377064, "rms_gain_error_db": -0.07477656763686191, "output_peak": 0.34531264691959446}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0873614328269694, "全帯域_10ms_rms_envelope_nrmse": 0.11997785581463026, "全帯域_10ms_envelope_correlation": 0.9834344861238062, "全帯域_10ms_modulation_mae_db": 0.687591781664883, "全帯域_40ms_log_rms_mae_db": 0.633152959199932, "全帯域_40ms_rms_envelope_nrmse": 0.0819590177918588, "全帯域_40ms_envelope_correlation": 0.9916295329899169, "全帯域_40ms_modulation_mae_db": 0.9296645570959159, "300–1000 Hz_10ms_log_rms_mae_db": 1.312047568966821, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09406480096868136, "300–1000 Hz_10ms_envelope_correlation": 0.993102204835075, "300–1000 Hz_10ms_modulation_mae_db": 0.9676117533588449, "300–1000 Hz_40ms_log_rms_mae_db": 0.7058024718097091, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054386075625816606, "300–1000 Hz_40ms_envelope_correlation": 0.9979940459688387, "300–1000 Hz_40ms_modulation_mae_db": 0.49846902477666916, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2656890393593843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1596101186482097, "1000–3000 Hz_10ms_envelope_correlation": 0.979886814168987, "1000–3000 Hz_10ms_modulation_mae_db": 0.5765157430253388, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6959721821825459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0975730960494985, "1000–3000 Hz_40ms_envelope_correlation": 0.992229897980237, "1000–3000 Hz_40ms_modulation_mae_db": 0.2909422521213045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2892130470494072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2461741142866958, "3000–6000 Hz_10ms_envelope_correlation": 0.9659285900504281, "3000–6000 Hz_10ms_modulation_mae_db": 1.1476061922543948, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8566449276881406, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18689890377594914, "3000–6000 Hz_40ms_envelope_correlation": 0.9799267604448243, "3000–6000 Hz_40ms_modulation_mae_db": 1.2353712732319664, "mr_stft_db": 5.413589410233418, "band_mae_db": 0.20859154003023994, "rms_gain_error_db": -0.04842661475210272, "output_peak": 0.3455463404708387}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0720627463650314, "全帯域_10ms_rms_envelope_nrmse": 0.101166669226228, "全帯域_10ms_envelope_correlation": 0.9883808450581125, "全帯域_10ms_modulation_mae_db": 0.37248538354618366, "全帯域_40ms_log_rms_mae_db": 0.574187305850186, "全帯域_40ms_rms_envelope_nrmse": 0.05765245502672345, "全帯域_40ms_envelope_correlation": 0.9964065112236965, "全帯域_40ms_modulation_mae_db": 0.6121752198529941, "300–1000 Hz_10ms_log_rms_mae_db": 1.354546594884022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0977536603471713, "300–1000 Hz_10ms_envelope_correlation": 0.9924384447465944, "300–1000 Hz_10ms_modulation_mae_db": 0.3975655522665272, "300–1000 Hz_40ms_log_rms_mae_db": 0.7946739251833728, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05764292451201447, "300–1000 Hz_40ms_envelope_correlation": 0.9975839691785509, "300–1000 Hz_40ms_modulation_mae_db": 0.5365049502066728, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2271616121284559, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14077463633933482, "1000–3000 Hz_10ms_envelope_correlation": 0.9859217719602784, "1000–3000 Hz_10ms_modulation_mae_db": 1.189378466390475, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6905182000362206, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08282940430884177, "1000–3000 Hz_40ms_envelope_correlation": 0.9965940597679597, "1000–3000 Hz_40ms_modulation_mae_db": 0.9353477406350383, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3419479093624216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20849944696598002, "3000–6000 Hz_10ms_envelope_correlation": 0.9771510026566894, "3000–6000 Hz_10ms_modulation_mae_db": 0.7433164209124042, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8797394021977823, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13649523398991434, "3000–6000 Hz_40ms_envelope_correlation": 0.9926210209385938, "3000–6000 Hz_40ms_modulation_mae_db": 0.9783713057331843, "mr_stft_db": 5.416662521839796, "band_mae_db": 0.3895333196666413, "rms_gain_error_db": -0.1861309130733131, "output_peak": 0.3278286579776677}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0787597648111351, "全帯域_10ms_rms_envelope_nrmse": 0.09486978372697327, "全帯域_10ms_envelope_correlation": 0.9898285334949642, "全帯域_10ms_modulation_mae_db": 0.42278474864943183, "全帯域_40ms_log_rms_mae_db": 0.5809345351839188, "全帯域_40ms_rms_envelope_nrmse": 0.04881395855636685, "全帯域_40ms_envelope_correlation": 0.9975120043091515, "全帯域_40ms_modulation_mae_db": 0.274795168411945, "300–1000 Hz_10ms_log_rms_mae_db": 1.3129402639605157, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09848297684682368, "300–1000 Hz_10ms_envelope_correlation": 0.9921833560618499, "300–1000 Hz_10ms_modulation_mae_db": 0.5064623105331699, "300–1000 Hz_40ms_log_rms_mae_db": 0.7377770923681163, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061085990980450755, "300–1000 Hz_40ms_envelope_correlation": 0.9970562473231274, "300–1000 Hz_40ms_modulation_mae_db": 0.4095535192159397, "1000–3000 Hz_10ms_log_rms_mae_db": 1.297843116948777, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15652952173046525, "1000–3000 Hz_10ms_envelope_correlation": 0.9821034377423477, "1000–3000 Hz_10ms_modulation_mae_db": 0.831146912191878, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7251757534015824, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08588564438017886, "1000–3000 Hz_40ms_envelope_correlation": 0.9963029952580452, "1000–3000 Hz_40ms_modulation_mae_db": 0.6097066975526405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3072436264197802, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17867161719292282, "3000–6000 Hz_10ms_envelope_correlation": 0.9828288800560054, "3000–6000 Hz_10ms_modulation_mae_db": 0.8747013447156563, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7942061291707114, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1018840745328397, "3000–6000 Hz_40ms_envelope_correlation": 0.995961127571839, "3000–6000 Hz_40ms_modulation_mae_db": 1.0114149821637366, "mr_stft_db": 5.41054167119329, "band_mae_db": 0.33240863018186323, "rms_gain_error_db": -0.1858679344282198, "output_peak": 0.3396597420123721}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0489151407026336, "全帯域_10ms_rms_envelope_nrmse": 0.10521161404961613, "全帯域_10ms_envelope_correlation": 0.9872429666182775, "全帯域_10ms_modulation_mae_db": 0.6782824459100668, "全帯域_40ms_log_rms_mae_db": 0.5961721061178882, "全帯域_40ms_rms_envelope_nrmse": 0.05897451875351858, "全帯域_40ms_envelope_correlation": 0.9959360076647582, "全帯域_40ms_modulation_mae_db": 0.7535761280412405, "300–1000 Hz_10ms_log_rms_mae_db": 1.3755565637158227, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11306647994547316, "300–1000 Hz_10ms_envelope_correlation": 0.9897939196034353, "300–1000 Hz_10ms_modulation_mae_db": 1.1393662838699419, "300–1000 Hz_40ms_log_rms_mae_db": 0.7311381458532762, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06917625022566927, "300–1000 Hz_40ms_envelope_correlation": 0.9964091782486384, "300–1000 Hz_40ms_modulation_mae_db": 0.7522763747940154, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2134264138803175, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16457796480852951, "1000–3000 Hz_10ms_envelope_correlation": 0.9825272089801892, "1000–3000 Hz_10ms_modulation_mae_db": 1.5343796600492905, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7254324662220161, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10705408387656164, "1000–3000 Hz_40ms_envelope_correlation": 0.9957825200484143, "1000–3000 Hz_40ms_modulation_mae_db": 1.4910787410400033, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2973927855370184, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20130835357370366, "3000–6000 Hz_10ms_envelope_correlation": 0.9767197764713021, "3000–6000 Hz_10ms_modulation_mae_db": 1.1689201052293645, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8394861272105354, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1227982669978094, "3000–6000 Hz_40ms_envelope_correlation": 0.9911644837096851, "3000–6000 Hz_40ms_modulation_mae_db": 1.5870867101226214, "mr_stft_db": 5.415233354078744, "band_mae_db": 0.3906416455582056, "rms_gain_error_db": -0.15459811551155395, "output_peak": 0.30488621403827526}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0048150793129658, "全帯域_10ms_rms_envelope_nrmse": 0.10307407874779304, "全帯域_10ms_envelope_correlation": 0.987697440468469, "全帯域_10ms_modulation_mae_db": 0.2628123040700077, "全帯域_40ms_log_rms_mae_db": 0.5557207965037249, "全帯域_40ms_rms_envelope_nrmse": 0.05465685274308491, "全帯域_40ms_envelope_correlation": 0.996419848031841, "全帯域_40ms_modulation_mae_db": 0.33545970724008245, "300–1000 Hz_10ms_log_rms_mae_db": 1.3080842221281659, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09994160247496571, "300–1000 Hz_10ms_envelope_correlation": 0.9920543412186688, "300–1000 Hz_10ms_modulation_mae_db": 0.9704207831623856, "300–1000 Hz_40ms_log_rms_mae_db": 0.6644208251364827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06149927761015123, "300–1000 Hz_40ms_envelope_correlation": 0.9971589070849219, "300–1000 Hz_40ms_modulation_mae_db": 0.41268532331639235, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1853184787223277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13876820347562077, "1000–3000 Hz_10ms_envelope_correlation": 0.9851169398110826, "1000–3000 Hz_10ms_modulation_mae_db": 0.4465903720703874, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6176940402182065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07544773874994758, "1000–3000 Hz_40ms_envelope_correlation": 0.9959843049229221, "1000–3000 Hz_40ms_modulation_mae_db": 0.25277611876314127, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3104059158403663, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20579619286537257, "3000–6000 Hz_10ms_envelope_correlation": 0.9754618756311618, "3000–6000 Hz_10ms_modulation_mae_db": 0.5294200736065852, "3000–6000 Hz_40ms_log_rms_mae_db": 0.807155235636474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10680254816392215, "3000–6000 Hz_40ms_envelope_correlation": 0.9935988258505977, "3000–6000 Hz_40ms_modulation_mae_db": 0.6589078470395742, "mr_stft_db": 5.410631402464657, "band_mae_db": 0.19205855696993757, "rms_gain_error_db": -0.11898611480024507, "output_peak": 0.37808461852578257}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1015593843986085, "全帯域_10ms_rms_envelope_nrmse": 0.09483724028831764, "全帯域_10ms_envelope_correlation": 0.9895927516344655, "全帯域_10ms_modulation_mae_db": 0.07200501655417209, "全帯域_40ms_log_rms_mae_db": 0.5359253303956307, "全帯域_40ms_rms_envelope_nrmse": 0.048005642194768584, "全帯域_40ms_envelope_correlation": 0.9971288104716429, "全帯域_40ms_modulation_mae_db": 0.25710317698392265, "300–1000 Hz_10ms_log_rms_mae_db": 1.3745563114718022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11005677965672009, "300–1000 Hz_10ms_envelope_correlation": 0.9900025103930301, "300–1000 Hz_10ms_modulation_mae_db": 0.7539572071576807, "300–1000 Hz_40ms_log_rms_mae_db": 0.7707924531454642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07123290360574736, "300–1000 Hz_40ms_envelope_correlation": 0.9956646259557256, "300–1000 Hz_40ms_modulation_mae_db": 0.652099033619476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2404506155490758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14675392657466654, "1000–3000 Hz_10ms_envelope_correlation": 0.9831570663513214, "1000–3000 Hz_10ms_modulation_mae_db": 0.3679642043002493, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6525337114343396, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07999202936244484, "1000–3000 Hz_40ms_envelope_correlation": 0.9951884841945318, "1000–3000 Hz_40ms_modulation_mae_db": 0.18946615908427508, "3000–6000 Hz_10ms_log_rms_mae_db": 1.302328570714582, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15984889495814453, "3000–6000 Hz_10ms_envelope_correlation": 0.9854839420740195, "3000–6000 Hz_10ms_modulation_mae_db": 0.3204586417125222, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8079008247509585, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07942303196173507, "3000–6000 Hz_40ms_envelope_correlation": 0.9963575984300842, "3000–6000 Hz_40ms_modulation_mae_db": 0.15903131403212747, "mr_stft_db": 5.409085703285975, "band_mae_db": 0.1563063027400461, "rms_gain_error_db": -0.041409638105616015, "output_peak": 0.3288039404489226}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83600, "output_samples": 83600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.472915304254914, "spectral_convergence": 0.3695472342729787, "spectral_convergence_status": "ok", "amplitude_floor": 9.102684030105167e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.138947665778128, "spectral_convergence": 0.34699002822696146, "spectral_convergence_status": "ok", "amplitude_floor": 4.6649711782186323e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.647601285566455, "spectral_convergence": 0.35735507196739075, "spectral_convergence_status": "ok", "amplitude_floor": 4.1057122515023654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021084017487574348, "output_power": 0.00021798536935363754, "delta_db": 0.14473978039291796, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00047622157098200406, "output_power": 0.00046540619156673987, "delta_db": -0.09976906703720353, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 9.587972951197478e-05, "output_power": 9.897344189054182e-05, "delta_db": 0.13791873341811384, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.411215555329107e-05, "output_power": 7.046546643612289e-05, "delta_db": -0.21913113888202473, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013227809814888984, "output_power": 0.0001181194485568117, "delta_db": -0.491665313582976, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00014809096583739848, "output_power": 0.0001408755032288865, "delta_db": -0.21693085142809976, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21835914745688934, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09885289822448685, "pitch": null, "output_rms_dbfs": -29.539634003747842, "output_peak": 0.3177117532249673, "reference_rms_dbfs": -29.44078110552336, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.419821418533165}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011323786448874551, "log_rms_mae_db": 1.0571128390969573, "rms_envelope_nrmse": 0.10801027897355288, "envelope_correlation": 0.9864218707693515, "rms_gain_error_db": -0.09888743078648128, "reference_mean_power": 0.0011384943575262435, "output_mean_power": 0.0011128641174008491}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 5.62759551802579e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002928658523286973, "output_power": 0.00027998547756718996, "reference_active": true, "delta_db": -0.1953323090557416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4610498849864094e-05, "output_power": 4.228598123745216e-05, "reference_active": true, "delta_db": -0.2324066694040961}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2101686887119984e-05, "output_power": 1.3393589194613104e-05, "reference_active": true, "delta_db": 0.44051062186616197}], "active_band_count": 3, "active_band_mae_db": 0.28941653344199986}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.563924945576787e-05, "log_rms_mae_db": 0.5713562714125923, "rms_envelope_nrmse": 0.04650125858732039, "envelope_correlation": 0.9974090389252529, "rms_gain_error_db": -0.09891702427384416, "reference_mean_power": 0.0011449801518647084, "output_mean_power": 0.0011191962745444317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.095010942049215e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002727818973842863, "output_power": 0.00026162799484100147, "reference_active": true, "delta_db": -0.18131333151093096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1267992599082604e-05, "output_power": 2.033740854958453e-05, "reference_active": true, "delta_db": -0.19430887440115951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.757168583210849e-07, "output_power": 6.466526508051029e-07, "reference_active": true, "delta_db": 0.5046211541626217}], "active_band_count": 3, "active_band_mae_db": 0.2934144533582374}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.979916609312644e-05, "log_rms_mae_db": 1.3519620674469104, "rms_envelope_nrmse": 0.09891594739852819, "envelope_correlation": 0.9919156710000535, "rms_gain_error_db": -0.09479637836131678, "reference_mean_power": 0.00038382289364078325, "output_mean_power": 0.00037553570714619205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 2.8601579771498626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014700129208541836, "output_power": 0.00014571149624797633, "reference_active": true, "delta_db": -0.03827334251454735}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.957621157356578e-05, "output_power": 1.7417673513365098e-05, "reference_active": true, "delta_db": -0.5073850445546582}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3113428408893907e-06, "output_power": 3.2043118865599075e-06, "reference_active": true, "delta_db": 1.418704117805738}], "active_band_count": 3, "active_band_mae_db": 0.6547875016249812}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.59156861629162e-05, "log_rms_mae_db": 0.6968701969442621, "rms_envelope_nrmse": 0.05140042116264579, "envelope_correlation": 0.9977609295039629, "rms_gain_error_db": -0.09480195259705246, "reference_mean_power": 0.00038603855676420007, "output_mean_power": 0.0003777030467159041}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 2.667606930039739e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013762626674571402, "output_power": 0.00013678548695054385, "reference_active": true, "delta_db": -0.02661308513597021}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.746730510582489e-06, "output_power": 8.81112764960319e-06, "reference_active": true, "delta_db": -0.4382746528041984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5949673646594045e-07, "output_power": 2.0069941630740307e-07, "reference_active": true, "delta_db": 0.9979430824347991}], "active_band_count": 3, "active_band_mae_db": 0.4876102734583226}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.341026161507385e-05, "log_rms_mae_db": 1.2105292490266406, "rms_envelope_nrmse": 0.15900746485756645, "envelope_correlation": 0.9811016685523298, "rms_gain_error_db": -0.47037232843790744, "reference_mean_power": 0.0001433729237288462, "output_mean_power": 0.00012865595972734505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.1455083323281239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.589033152322262e-05, "output_power": 4.5593720216081376e-05, "reference_active": true, "delta_db": -0.8843165584432902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4083924473542424e-05, "output_power": 1.1328566961308785e-05, "reference_active": true, "delta_db": -0.9454871141127434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.499188380560451e-06, "output_power": 4.869790835677094e-06, "reference_active": true, "delta_db": -0.5278828909695557}], "active_band_count": 3, "active_band_mae_db": 0.7858955211751963}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 5.482542724448801e-05, "log_rms_mae_db": 0.6416296225003673, "rms_envelope_nrmse": 0.09448466336013234, "envelope_correlation": 0.9944907991048793, "rms_gain_error_db": -0.47046272388555144, "reference_mean_power": 0.00014418079784839864, "output_mean_power": 0.00012937821412315706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.042775196620521e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2049448137923216e-05, "output_power": 4.2670397088275405e-05, "reference_active": true, "delta_db": -0.8628944480088194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.197217517258297e-06, "output_power": 5.104023516575166e-06, "reference_active": true, "delta_db": -0.8428407308635383}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5366430302416474e-07, "output_power": 2.238533725102008e-07, "reference_active": true, "delta_db": -0.5429571359243062}], "active_band_count": 3, "active_band_mae_db": 0.7495641049322211}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011141137328067575, "log_rms_mae_db": 1.3398339928323428, "rms_envelope_nrmse": 0.20990304617953018, "envelope_correlation": 0.974471040109458, "rms_gain_error_db": -0.3041928176358653, "reference_mean_power": 0.00018590462562984253, "output_mean_power": 0.00017332887337451373}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.861645592580093e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.396584441083375e-05, "output_power": 8.60838249164458e-05, "reference_active": true, "delta_db": -0.38048465314130975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7617019162087818e-05, "output_power": 1.6262762235128783e-05, "reference_active": true, "delta_db": -0.3473811414879108}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.489954892102086e-06, "output_power": 5.652580977421856e-06, "reference_active": true, "delta_db": -0.5999488562971088}], "active_band_count": 3, "active_band_mae_db": 0.44260488364210976}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.054702623382987e-05, "log_rms_mae_db": 0.8066921867319421, "rms_envelope_nrmse": 0.08701880730133889, "envelope_correlation": 0.9961038936382057, "rms_gain_error_db": -0.3041980604964914, "reference_mean_power": 0.00018697687803543306, "output_mean_power": 0.00017432838146564912}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.7928912528322647e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.176227386223464e-05, "output_power": 8.509630424480189e-05, "reference_active": true, "delta_db": -0.3275346784216914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.193273720945643e-06, "output_power": 6.9104206122968086e-06, "reference_active": true, "delta_db": -0.7395298190255296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.787097063305561e-07, "output_power": 3.2950651388147187e-07, "reference_active": true, "delta_db": -0.6044243182447169}], "active_band_count": 3, "active_band_mae_db": 0.5571629385639792}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8542426166031799, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "050edeb53def66846a0f9569e12b1b9c87e3daeaf91b05621aa4a5af9cbec55e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9825565322352086, "standard_deviation": 0.0021170938028465336, "minimum": 0.979886814168987, "maximum": 0.9859217719602784, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2248210904199062, "standard_deviation": 0.04406444201634452, "minimum": 1.1581501977442692, "maximum": 1.297843116948777, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.746098166910194, "standard_deviation": 0.43873498631684127, "minimum": 0.23691445607873682, "maximum": 1.5343796600492905, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15277801014892897, "standard_deviation": 0.009465480238555068, "minimum": 0.13876820347562077, "maximum": 0.16457796480852951, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949939931264933, "standard_deviation": 0.0015313870146935137, "minimum": 0.992229897980237, "maximum": 0.9965940597679597, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6751528220347947, "standard_deviation": 0.03992957497806797, "minimum": 0.6176940402182065, "maximum": 0.7254324662220161, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6375227878523084, "standard_deviation": 0.4313845178961801, "minimum": 0.18946615908427508, "maximum": 1.4910787410400033, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0891357849127099, "standard_deviation": 0.010300045146176579, "minimum": 0.07544773874994758, "maximum": 0.10705408387656164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9766642449502289, "standard_deviation": 0.005849139688644465, "minimum": 0.9659285900504281, "maximum": 0.9854839420740195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.311834757892831, "standard_deviation": 0.01910518838973681, "minimum": 1.2892130470494072, "maximum": 1.3419479093624216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7183320161484636, "standard_deviation": 0.32113215521912397, "minimum": 0.3204586417125222, "maximum": 1.1689201052293645, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20208180045330137, "standard_deviation": 0.025104482297970947, "minimum": 0.15984889495814453, "maximum": 0.2461741142866958, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9918117328890431, "standard_deviation": 0.005492037174006111, "minimum": 0.9799267604448243, "maximum": 0.9963575984300842, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8273587391470532, "standard_deviation": 0.02942685981026392, "minimum": 0.7942061291707114, "maximum": 0.8797394021977823, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8268262871502339, "standard_deviation": 0.46421345548803805, "minimum": 0.15903131403212747, "maximum": 1.5870867101226214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11993979589192214, "standard_deviation": 0.03448211812242391, "minimum": 0.07942303196173507, "maximum": 0.18689890377594914, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916930239670233, "standard_deviation": 0.0011684264546709007, "minimum": 0.9897939196034353, "maximum": 0.993102204835075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3337067997076986, "standard_deviation": 0.035118487042091494, "minimum": 1.2799608050875289, "maximum": 1.3755565637158227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7413758415152712, "standard_deviation": 0.26276085580239195, "minimum": 0.3975655522665272, "maximum": 1.1393662838699419, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10140106388468673, "standard_deviation": 0.0065582375060227955, "minimum": 0.09406480096868136, "maximum": 0.11306647994547316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970182406005839, "standard_deviation": 0.0007836281900437857, "minimum": 0.9956646259557256, "maximum": 0.9979940459688387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.72661316741023, "standard_deviation": 0.04165388524445811, "minimum": 0.6644208251364827, "maximum": 0.7946739251833728, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5342351411831963, "standard_deviation": 0.1166822533018511, "minimum": 0.4095535192159397, "maximum": 0.7522763747940154, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06145443042877468, "standard_deviation": 0.006918623292381278, "minimum": 0.05140042116264579, "maximum": 0.07123290360574736, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2603950736771992, "standard_deviation": 0.09488652973436863, "minimum": 0.1563063027400461, "maximum": 0.3906416455582056, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.412370861280742, "standard_deviation": 0.0050881217104577275, "minimum": 5.403401408616889, "maximum": 5.419821418533165, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.33597923920230255, "standard_deviation": 0.02198520651805696, "minimum": 0.30488621403827526, "maximum": 0.37808461852578257, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11363109956654993, "standard_deviation": 0.057789012888995316, "minimum": -0.1861309130733131, "maximum": -0.041409638105616015, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9874813764474416, "standard_deviation": 0.0020153713360162983, "minimum": 0.9834344861238062, "maximum": 0.9898285334949642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0702896072496868, "standard_deviation": 0.0337615330053917, "minimum": 1.0048150793129658, "maximum": 1.1117304704831936, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3771311838782899, "standard_deviation": 0.21520984154287065, "minimum": 0.07200501655417209, "maximum": 0.687591781664883, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1040141676584752, "standard_deviation": 0.008017619293891842, "minimum": 0.09483724028831764, "maximum": 0.11997785581463026, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959633242030452, "standard_deviation": 0.00190945245538827, "minimum": 0.9916295329899169, "maximum": 0.9975120043091515, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5758303045463307, "standard_deviation": 0.029330725019261362, "minimum": 0.5359253303956307, "maximum": 0.633152959199932, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44748673344088424, "standard_deviation": 0.28296879717650564, "minimum": 0.12370545654273575, "maximum": 0.9296645570959159, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05729901578244363, "standard_deviation": 0.01142365359070857, "minimum": 0.04650125858732039, "maximum": 0.0819590177918588, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0056", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.4533469676971436, "predicted_mos": 2.4533469676971436, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "4c310820c8d1a99d6e31e4329277489f420a42350a067b21ddbefcdd2927ec57", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7850677999667823, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.017972281424754, "全帯域_10ms_rms_envelope_nrmse": 0.07992713591685557, "全帯域_10ms_envelope_correlation": 0.9926179328482382, "全帯域_10ms_modulation_mae_db": 0.27274148106727736, "全帯域_40ms_log_rms_mae_db": 0.5716956963338636, "全帯域_40ms_rms_envelope_nrmse": 0.03892116863302411, "全帯域_40ms_envelope_correlation": 0.9981591359489046, "全帯域_40ms_modulation_mae_db": 0.4122796196700311, "300–1000 Hz_10ms_log_rms_mae_db": 1.4295202768146675, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08097189624611358, "300–1000 Hz_10ms_envelope_correlation": 0.9946324170066806, "300–1000 Hz_10ms_modulation_mae_db": 0.20850488953969623, "300–1000 Hz_40ms_log_rms_mae_db": 0.8278661884717184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04326027609078852, "300–1000 Hz_40ms_envelope_correlation": 0.9984546039753157, "300–1000 Hz_40ms_modulation_mae_db": 0.04565605999990638, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3887585488523657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15797233745553227, "1000–3000 Hz_10ms_envelope_correlation": 0.9810947700871018, "1000–3000 Hz_10ms_modulation_mae_db": 0.8568535989808167, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8113124058000147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09057800875807998, "1000–3000 Hz_40ms_envelope_correlation": 0.9947725686400795, "1000–3000 Hz_40ms_modulation_mae_db": 0.45638292181974505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.547510790384609, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13829708841724672, "3000–6000 Hz_10ms_envelope_correlation": 0.9891363948297596, "3000–6000 Hz_10ms_modulation_mae_db": 0.29779989221046377, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0077213244508487, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0747254238904819, "3000–6000 Hz_40ms_envelope_correlation": 0.9967730055452302, "3000–6000 Hz_40ms_modulation_mae_db": 0.5291873786768563, "mr_stft_db": 5.518753650962346, "band_mae_db": 0.23554902658435903, "rms_gain_error_db": -0.05734119406776289, "output_peak": 0.31592758749225236}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9551548337372888, "全帯域_10ms_rms_envelope_nrmse": 0.09297385814695833, "全帯域_10ms_envelope_correlation": 0.989972693999592, "全帯域_10ms_modulation_mae_db": 0.3733710530235475, "全帯域_40ms_log_rms_mae_db": 0.4482896552264272, "全帯域_40ms_rms_envelope_nrmse": 0.051049238139969866, "全帯域_40ms_envelope_correlation": 0.9968114909039969, "全帯域_40ms_modulation_mae_db": 0.5561200537077138, "300–1000 Hz_10ms_log_rms_mae_db": 1.4814931579992712, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09633171802610105, "300–1000 Hz_10ms_envelope_correlation": 0.9926286034764782, "300–1000 Hz_10ms_modulation_mae_db": 0.5584625402903985, "300–1000 Hz_40ms_log_rms_mae_db": 0.9113767141889484, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058330574186258616, "300–1000 Hz_40ms_envelope_correlation": 0.9974605129892001, "300–1000 Hz_40ms_modulation_mae_db": 0.40162873180435854, "1000–3000 Hz_10ms_log_rms_mae_db": 1.389252372899274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16650656626705426, "1000–3000 Hz_10ms_envelope_correlation": 0.977927629166116, "1000–3000 Hz_10ms_modulation_mae_db": 0.17819229720292185, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7635929183414433, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0877193442245514, "1000–3000 Hz_40ms_envelope_correlation": 0.9935864844912581, "1000–3000 Hz_40ms_modulation_mae_db": 0.4877808711391778, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5353465042154228, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18101831096440513, "3000–6000 Hz_10ms_envelope_correlation": 0.9810563179294154, "3000–6000 Hz_10ms_modulation_mae_db": 0.6463058771568263, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9766556009881118, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11676511383302603, "3000–6000 Hz_40ms_envelope_correlation": 0.9921463887006897, "3000–6000 Hz_40ms_modulation_mae_db": 0.9546776068957393, "mr_stft_db": 5.526906868070654, "band_mae_db": 0.137534135741371, "rms_gain_error_db": -0.08509893453505299, "output_peak": 0.37553707823956356}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0270324326191835, "全帯域_10ms_rms_envelope_nrmse": 0.0924271644502929, "全帯域_10ms_envelope_correlation": 0.9903078108679716, "全帯域_10ms_modulation_mae_db": 0.24194480974850788, "全帯域_40ms_log_rms_mae_db": 0.5653210994232754, "全帯域_40ms_rms_envelope_nrmse": 0.048911966028538444, "全帯域_40ms_envelope_correlation": 0.9973960127619823, "全帯域_40ms_modulation_mae_db": 0.32641306336446557, "300–1000 Hz_10ms_log_rms_mae_db": 1.4458661256008434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11408920895691087, "300–1000 Hz_10ms_envelope_correlation": 0.9895301968514003, "300–1000 Hz_10ms_modulation_mae_db": 1.1685246229393524, "300–1000 Hz_40ms_log_rms_mae_db": 0.8390737058815992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06718393339336783, "300–1000 Hz_40ms_envelope_correlation": 0.9965649032591806, "300–1000 Hz_40ms_modulation_mae_db": 0.7758190409120429, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3598751686314878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15626915589377793, "1000–3000 Hz_10ms_envelope_correlation": 0.9821065089665529, "1000–3000 Hz_10ms_modulation_mae_db": 1.4558906156164466, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7236169182180582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09470123644927278, "1000–3000 Hz_40ms_envelope_correlation": 0.994734573467241, "1000–3000 Hz_40ms_modulation_mae_db": 1.1511171466333125, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4634606148257308, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16034581703466103, "3000–6000 Hz_10ms_envelope_correlation": 0.9851929022909823, "3000–6000 Hz_10ms_modulation_mae_db": 0.3410282795726482, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9646071231810148, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08925013768274415, "3000–6000 Hz_40ms_envelope_correlation": 0.9955237146953443, "3000–6000 Hz_40ms_modulation_mae_db": 0.4064363071347091, "mr_stft_db": 5.517991868418459, "band_mae_db": 0.2492732902614211, "rms_gain_error_db": -0.16776195383690795, "output_peak": 0.33440698029336546}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9682931159304766, "全帯域_10ms_rms_envelope_nrmse": 0.08240761397183449, "全帯域_10ms_envelope_correlation": 0.9921468889152228, "全帯域_10ms_modulation_mae_db": 0.2950258635558737, "全帯域_40ms_log_rms_mae_db": 0.5079284142230738, "全帯域_40ms_rms_envelope_nrmse": 0.041324838716754325, "全帯域_40ms_envelope_correlation": 0.9979845403444405, "全帯域_40ms_modulation_mae_db": 0.28194426414592705, "300–1000 Hz_10ms_log_rms_mae_db": 1.4535753906185487, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0903782512677478, "300–1000 Hz_10ms_envelope_correlation": 0.9934223949423506, "300–1000 Hz_10ms_modulation_mae_db": 0.511113228205864, "300–1000 Hz_40ms_log_rms_mae_db": 0.8182064153107793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048927316349869736, "300–1000 Hz_40ms_envelope_correlation": 0.9981745422772038, "300–1000 Hz_40ms_modulation_mae_db": 0.31205022322137405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4116533701657514, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1587609180412885, "1000–3000 Hz_10ms_envelope_correlation": 0.9798570955000548, "1000–3000 Hz_10ms_modulation_mae_db": 0.22184521117947373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8243045100240143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08366043821250418, "1000–3000 Hz_40ms_envelope_correlation": 0.9942711577399795, "1000–3000 Hz_40ms_modulation_mae_db": 0.5730108416244358, "3000–6000 Hz_10ms_log_rms_mae_db": 1.512082595432424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1373052699538762, "3000–6000 Hz_10ms_envelope_correlation": 0.9895931399901778, "3000–6000 Hz_10ms_modulation_mae_db": 0.8759072934991473, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9684969352119933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0786269350054301, "3000–6000 Hz_40ms_envelope_correlation": 0.9970591467400305, "3000–6000 Hz_40ms_modulation_mae_db": 0.9941442941078215, "mr_stft_db": 5.532787191159365, "band_mae_db": 0.18061219643710422, "rms_gain_error_db": -0.09359253791455169, "output_peak": 0.3318462590088716}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9516869257373813, "全帯域_10ms_rms_envelope_nrmse": 0.09672972131284521, "全帯域_10ms_envelope_correlation": 0.98929890056353, "全帯域_10ms_modulation_mae_db": 0.3414549701622718, "全帯域_40ms_log_rms_mae_db": 0.48764174419181067, "全帯域_40ms_rms_envelope_nrmse": 0.05693540303872099, "全帯域_40ms_envelope_correlation": 0.9960050978586675, "全帯域_40ms_modulation_mae_db": 0.2219015020954794, "300–1000 Hz_10ms_log_rms_mae_db": 1.4650569746023319, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10452146354402457, "300–1000 Hz_10ms_envelope_correlation": 0.9912085445792879, "300–1000 Hz_10ms_modulation_mae_db": 0.5774606633799143, "300–1000 Hz_40ms_log_rms_mae_db": 0.8570583709750922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06769888278904089, "300–1000 Hz_40ms_envelope_correlation": 0.996369065101673, "300–1000 Hz_40ms_modulation_mae_db": 0.34604398685104604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3477656919483871, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13737458214618, "1000–3000 Hz_10ms_envelope_correlation": 0.9850188335963191, "1000–3000 Hz_10ms_modulation_mae_db": 0.4896609830497969, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8008359949247312, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07798492233903348, "1000–3000 Hz_40ms_envelope_correlation": 0.9950592803245907, "1000–3000 Hz_40ms_modulation_mae_db": 0.5234403682952508, "3000–6000 Hz_10ms_log_rms_mae_db": 1.551928083156048, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19205268082861723, "3000–6000 Hz_10ms_envelope_correlation": 0.9807856019160862, "3000–6000 Hz_10ms_modulation_mae_db": 0.5829844176367746, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0404915220009552, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12152007812009114, "3000–6000 Hz_40ms_envelope_correlation": 0.9927656963596858, "3000–6000 Hz_40ms_modulation_mae_db": 0.669105195822194, "mr_stft_db": 5.526731726801885, "band_mae_db": 0.25409325423612733, "rms_gain_error_db": 0.047681653767261106, "output_peak": 0.3608305336422242}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0110994649487497, "全帯域_10ms_rms_envelope_nrmse": 0.08355744318559025, "全帯域_10ms_envelope_correlation": 0.9919315873293453, "全帯域_10ms_modulation_mae_db": 0.2604049850605818, "全帯域_40ms_log_rms_mae_db": 0.5626528444841307, "全帯域_40ms_rms_envelope_nrmse": 0.042175171082387775, "全帯域_40ms_envelope_correlation": 0.9978633778372789, "全帯域_40ms_modulation_mae_db": 0.3990579905246754, "300–1000 Hz_10ms_log_rms_mae_db": 1.4797317912747798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09321504024259218, "300–1000 Hz_10ms_envelope_correlation": 0.9929932924824886, "300–1000 Hz_10ms_modulation_mae_db": 0.3789416871518169, "300–1000 Hz_40ms_log_rms_mae_db": 0.8863600475868658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052105960939978296, "300–1000 Hz_40ms_envelope_correlation": 0.9979045766867766, "300–1000 Hz_40ms_modulation_mae_db": 0.17597354227096604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.377686200569042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.148358392269641, "1000–3000 Hz_10ms_envelope_correlation": 0.9824055305643512, "1000–3000 Hz_10ms_modulation_mae_db": 0.16107226201707903, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7448569471935834, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0792714085270717, "1000–3000 Hz_40ms_envelope_correlation": 0.9946717464195616, "1000–3000 Hz_40ms_modulation_mae_db": 0.10253257531171896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.599003133029519, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1537424066619982, "3000–6000 Hz_10ms_envelope_correlation": 0.9865103475985151, "3000–6000 Hz_10ms_modulation_mae_db": 0.5021093718138544, "3000–6000 Hz_40ms_log_rms_mae_db": 1.048069659742603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09073746614193352, "3000–6000 Hz_40ms_envelope_correlation": 0.9955134578736504, "3000–6000 Hz_40ms_modulation_mae_db": 0.41852616832726697, "mr_stft_db": 5.528624518352385, "band_mae_db": 0.10664996847056073, "rms_gain_error_db": -0.08556177334514296, "output_peak": 0.3452679435633505}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0861660465549239, "全帯域_10ms_rms_envelope_nrmse": 0.11458376111513341, "全帯域_10ms_envelope_correlation": 0.9847863176344219, "全帯域_10ms_modulation_mae_db": 0.5176038143964055, "全帯域_40ms_log_rms_mae_db": 0.5890718912449256, "全帯域_40ms_rms_envelope_nrmse": 0.06380727549716714, "全帯域_40ms_envelope_correlation": 0.9949333679527336, "全帯域_40ms_modulation_mae_db": 1.1318653711258175, "300–1000 Hz_10ms_log_rms_mae_db": 1.4802442450541984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09630458659586776, "300–1000 Hz_10ms_envelope_correlation": 0.9924259056549013, "300–1000 Hz_10ms_modulation_mae_db": 0.2636529720921329, "300–1000 Hz_40ms_log_rms_mae_db": 0.8743599120053162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05346735892954778, "300–1000 Hz_40ms_envelope_correlation": 0.9976788615897919, "300–1000 Hz_40ms_modulation_mae_db": 0.5150471726760802, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3844437575251107, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1874024866378477, "1000–3000 Hz_10ms_envelope_correlation": 0.9719531645818904, "1000–3000 Hz_10ms_modulation_mae_db": 0.3921292598668771, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8394501567374639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.110555698250069, "1000–3000 Hz_40ms_envelope_correlation": 0.9895808046206768, "1000–3000 Hz_40ms_modulation_mae_db": 0.5918721755793223, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5704627787259096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21748454283395408, "3000–6000 Hz_10ms_envelope_correlation": 0.9724737614158908, "3000–6000 Hz_10ms_modulation_mae_db": 1.0429328660200143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.032926877117491, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1299584244989523, "3000–6000 Hz_40ms_envelope_correlation": 0.9901431338575994, "3000–6000 Hz_40ms_modulation_mae_db": 1.6419162993364045, "mr_stft_db": 5.529188096444034, "band_mae_db": 0.11854067540033975, "rms_gain_error_db": -0.07397722280724327, "output_peak": 0.3249685017487501}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9945795096338738, "全帯域_10ms_rms_envelope_nrmse": 0.09514126098251018, "全帯域_10ms_envelope_correlation": 0.9894984023413512, "全帯域_10ms_modulation_mae_db": 0.25014582484836, "全帯域_40ms_log_rms_mae_db": 0.5281305762635619, "全帯域_40ms_rms_envelope_nrmse": 0.048127313035645894, "全帯域_40ms_envelope_correlation": 0.9971439027722933, "全帯域_40ms_modulation_mae_db": 0.46585586232802073, "300–1000 Hz_10ms_log_rms_mae_db": 1.4683895938928606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09279492162574557, "300–1000 Hz_10ms_envelope_correlation": 0.9931589366013556, "300–1000 Hz_10ms_modulation_mae_db": 0.31891938233199424, "300–1000 Hz_40ms_log_rms_mae_db": 0.8537237590927265, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049101928035454274, "300–1000 Hz_40ms_envelope_correlation": 0.9983083568342392, "300–1000 Hz_40ms_modulation_mae_db": 0.29530830833234395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3897997953290044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16547392861441032, "1000–3000 Hz_10ms_envelope_correlation": 0.9784806086337251, "1000–3000 Hz_10ms_modulation_mae_db": 0.8458405879164209, "1000–3000 Hz_40ms_log_rms_mae_db": 0.750975084906055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0938450841983772, "1000–3000 Hz_40ms_envelope_correlation": 0.9933830871321438, "1000–3000 Hz_40ms_modulation_mae_db": 0.5674611385952896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4911784114584918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17333946315375287, "3000–6000 Hz_10ms_envelope_correlation": 0.9826221167954465, "3000–6000 Hz_10ms_modulation_mae_db": 0.35566596010567464, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9531507796551669, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09313429931522843, "3000–6000 Hz_40ms_envelope_correlation": 0.9948899717528717, "3000–6000 Hz_40ms_modulation_mae_db": 0.5714458286679472, "mr_stft_db": 5.51337855006059, "band_mae_db": 0.1613330625959641, "rms_gain_error_db": -0.06657403288426436, "output_peak": 0.34791356691154035}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83600, "output_samples": 83600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.635394729659673, "spectral_convergence": 0.35057431520021637, "spectral_convergence_status": "ok", "amplitude_floor": 9.102684030105167e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.265443154467397, "spectral_convergence": 0.33570483182655103, "spectral_convergence_status": "ok", "amplitude_floor": 4.6649711782186323e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.655423068759964, "spectral_convergence": 0.3559282520333712, "spectral_convergence_status": "ok", "amplitude_floor": 4.1057122515023654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021084017487574348, "output_power": 0.00021217881230552394, "delta_db": 0.02748646312504046, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00047622157098200406, "output_power": 0.00047087070526729863, "delta_db": -0.04907391260319269, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 9.587972951197478e-05, "output_power": 9.446294937102098e-05, "delta_db": -0.06465298948273916, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.411215555329107e-05, "output_power": 6.33486750101576e-05, "delta_db": -0.6815190922599033, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013227809814888984, "output_power": 0.0001233858336397803, "delta_db": -0.3022264245580997, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00014809096583739848, "output_power": 0.0001582566955624909, "delta_db": 0.28833527747717874, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23554902658435903, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05734119406776289, "pitch": null, "output_rms_dbfs": -29.498122299591124, "output_peak": 0.31592758749225236, "reference_rms_dbfs": -29.44078110552336, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.518753650962346}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011323786448874551, "log_rms_mae_db": 1.017972281424754, "rms_envelope_nrmse": 0.07992713591685557, "envelope_correlation": 0.9926179328482382, "rms_gain_error_db": -0.057414841845522084, "reference_mean_power": 0.0011384943575262435, "output_mean_power": 0.0011235422240711132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 5.62759551802579e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002928658523286973, "output_power": 0.0002913476852378085, "reference_active": true, "delta_db": -0.02257164656214845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4610498849864094e-05, "output_power": 4.428976671218503e-05, "reference_active": true, "delta_db": -0.03133686970029916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2101686887119984e-05, "output_power": 1.0148791243376063e-05, "reference_active": true, "delta_db": -0.7643159269393844}], "active_band_count": 3, "active_band_mae_db": 0.27274148106727736}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.563924945576787e-05, "log_rms_mae_db": 0.5716956963338636, "rms_envelope_nrmse": 0.03892116863302411, "envelope_correlation": 0.9981591359489046, "rms_gain_error_db": -0.057482840876672325, "reference_mean_power": 0.0011449801518647084, "output_mean_power": 0.0011299251470589442}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.095010942049215e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002727818973842863, "output_power": 0.0002724346888742275, "reference_active": true, "delta_db": -0.005531407670755815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1267992599082604e-05, "output_power": 2.0792188533287245e-05, "reference_active": true, "delta_db": -0.09826296036229074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.757168583210849e-07, "output_power": 4.435111041422238e-07, "reference_active": true, "delta_db": -1.1330444909770467}], "active_band_count": 3, "active_band_mae_db": 0.4122796196700311}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.979916609312644e-05, "log_rms_mae_db": 1.4295202768146675, "rms_envelope_nrmse": 0.08097189624611358, "envelope_correlation": 0.9946324170066806, "rms_gain_error_db": -0.09897037735586021, "reference_mean_power": 0.00038382289364078325, "output_mean_power": 0.00037517495362149794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 2.8601579771498626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014700129208541836, "output_power": 0.00014589001193488475, "reference_active": true, "delta_db": -0.03295592231834109}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.957621157356578e-05, "output_power": 1.8517658260592792e-05, "reference_active": true, "delta_db": -0.2414258506405699}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3113428408893907e-06, "output_power": 2.5059803347030833e-06, "reference_active": true, "delta_db": 0.35113289566017775}], "active_band_count": 3, "active_band_mae_db": 0.20850488953969623}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.59156861629162e-05, "log_rms_mae_db": 0.8278661884717184, "rms_envelope_nrmse": 0.04326027609078852, "envelope_correlation": 0.9984546039753157, "rms_gain_error_db": -0.09896875451192973, "reference_mean_power": 0.00038603855676420007, "output_mean_power": 0.0003773408364889602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 2.667606930039739e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013762626674571402, "output_power": 0.0001373110712222782, "reference_active": true, "delta_db": -0.009957740736909245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.746730510582489e-06, "output_power": 9.492731055454703e-06, "reference_active": true, "delta_db": -0.11467781553378965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5949673646594045e-07, "output_power": 1.5995030168950526e-07, "reference_active": true, "delta_db": 0.012332623729020224}], "active_band_count": 3, "active_band_mae_db": 0.04565605999990638}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.341026161507385e-05, "log_rms_mae_db": 1.3887585488523657, "rms_envelope_nrmse": 0.15797233745553227, "envelope_correlation": 0.9810947700871018, "rms_gain_error_db": -0.4239731245297393, "reference_mean_power": 0.0001433729237288462, "output_mean_power": 0.00013003786461362557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.1455083323281239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.589033152322262e-05, "output_power": 4.596186811452497e-05, "reference_active": true, "delta_db": -0.8493901349555766}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4083924473542424e-05, "output_power": 1.33547715398153e-05, "reference_active": true, "delta_db": -0.23087224514947235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.499188380560451e-06, "output_power": 3.901833700714851e-06, "reference_active": true, "delta_db": -1.4902984168374012}], "active_band_count": 3, "active_band_mae_db": 0.8568535989808167}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 5.482542724448801e-05, "log_rms_mae_db": 0.8113124058000147, "rms_envelope_nrmse": 0.09057800875807998, "envelope_correlation": 0.9947725686400795, "rms_gain_error_db": -0.4238962146155921, "reference_mean_power": 0.00014418079784839864, "output_mean_power": 0.00013077291453804696}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.042775196620521e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2049448137923216e-05, "output_power": 4.326307254284575e-05, "reference_active": true, "delta_db": -0.8029876941181407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.197217517258297e-06, "output_power": 5.790093803849952e-06, "reference_active": true, "delta_db": -0.29511140117915324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5366430302416474e-07, "output_power": 2.383166515894363e-07, "reference_active": true, "delta_db": -0.2710496701619412}], "active_band_count": 3, "active_band_mae_db": 0.45638292181974505}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011141137328067575, "log_rms_mae_db": 1.547510790384609, "rms_envelope_nrmse": 0.13829708841724672, "envelope_correlation": 0.9891363948297596, "rms_gain_error_db": 0.06956051385093638, "reference_mean_power": 0.00018590462562984253, "output_mean_power": 0.00018890621538280946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.861645592580093e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.396584441083375e-05, "output_power": 9.227531075610225e-05, "reference_active": true, "delta_db": -0.0788450439284526}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7617019162087818e-05, "output_power": 1.699292737231353e-05, "reference_active": true, "delta_db": -0.15664225479618343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.489954892102086e-06, "output_power": 5.577639492851795e-06, "reference_active": true, "delta_db": -0.6579123779067553}], "active_band_count": 3, "active_band_mae_db": 0.29779989221046377}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.054702623382987e-05, "log_rms_mae_db": 1.0077213244508487, "rms_envelope_nrmse": 0.0747254238904819, "envelope_correlation": 0.9967730055452302, "rms_gain_error_db": 0.06955990973993761, "reference_mean_power": 0.00018697687803543306, "output_mean_power": 0.0001899957537951277}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.7928912528322647e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.176227386223464e-05, "output_power": 9.072325018688345e-05, "reference_active": true, "delta_db": -0.04945566257156017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.193273720945643e-06, "output_power": 7.580257865774272e-06, "reference_active": true, "delta_db": -0.33773484269596593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.787097063305561e-07, "output_power": 2.8725610878222985e-07, "reference_active": true, "delta_db": -1.2003716307630425}], "active_band_count": 3, "active_band_mae_db": 0.5291873786768563}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8406397644982715, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "616b104d91d5cc1619d914b467b215437c36f5d499548ec66a963cc8c55a0c06", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9798555176370138, "standard_deviation": 0.003924636857576388, "minimum": 0.9719531645818904, "maximum": 0.9850188335963191, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.381154363240053, "standard_deviation": 0.019707830494548825, "minimum": 1.3477656919483871, "maximum": 1.4116533701657514, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5751856019787291, "standard_deviation": 0.45087206959184006, "minimum": 0.16107226201707903, "maximum": 1.4558906156164466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15976479591571652, "standard_deviation": 0.014588427439767971, "minimum": 0.13737458214618, "maximum": 0.1874024866378477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937574628544414, "standard_deviation": 0.0017885978262291247, "minimum": 0.9895808046206768, "maximum": 0.9950592803245907, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7823681170181706, "standard_deviation": 0.042075911766568205, "minimum": 0.7236169182180582, "maximum": 0.8394501567374639, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5566997548747816, "standard_deviation": 0.2871381666055645, "minimum": 0.10253257531171896, "maximum": 1.1511171466333125, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08978951761986997, "standard_deviation": 0.0104489158649781, "minimum": 0.07798492233903348, "maximum": 0.110555698250069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9834213228457842, "standard_deviation": 0.005560858707396166, "minimum": 0.9724737614158908, "maximum": 0.9895931399901778, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5338716139035196, "standard_deviation": 0.043671576727693805, "minimum": 1.4634606148257308, "maximum": 1.599003133029519, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5805917447519254, "standard_deviation": 0.2671028787890608, "minimum": 0.29779989221046377, "maximum": 1.0429328660200143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16919819748106393, "standard_deviation": 0.027516351496602292, "minimum": 0.1373052699538762, "maximum": 0.21748454283395408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9943518144406378, "standard_deviation": 0.002428899192774901, "minimum": 0.9901431338575994, "maximum": 0.9970591467400305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9990149777935231, "standard_deviation": 0.03793076533527377, "minimum": 0.9531507796551669, "maximum": 1.048069659742603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7731798848711173, "standard_deviation": 0.4153845507390012, "minimum": 0.4064363071347091, "maximum": 1.6419162993364045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09933973481098594, "standard_deviation": 0.020640827868269006, "minimum": 0.0747254238904819, "maximum": 0.1299584244989523, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925000364493679, "standard_deviation": 0.0015392561421752305, "minimum": 0.9895301968514003, "maximum": 0.9946324170066806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4629846944821876, "standard_deviation": 0.01873070622876148, "minimum": 1.4295202768146675, "maximum": 1.4814931579992712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4981974982413962, "standard_deviation": 0.30348342586726584, "minimum": 0.20850488953969623, "maximum": 1.1685246229393524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09607588581313792, "standard_deviation": 0.009823209045714942, "minimum": 0.08097189624611358, "maximum": 0.11408920895691087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976144278391726, "standard_deviation": 0.0007806242564783252, "minimum": 0.996369065101673, "maximum": 0.9984546039753157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8585031391891307, "standard_deviation": 0.03114010031964877, "minimum": 0.8182064153107793, "maximum": 0.9113767141889484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.35844088325851475, "standard_deviation": 0.2197451189828224, "minimum": 0.04565605999990638, "maximum": 0.7758190409120429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05500952883928824, "standard_deviation": 0.008791328217978449, "minimum": 0.04326027609078852, "maximum": 0.06769888278904089, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1804482012159059, "standard_deviation": 0.05938054156533224, "minimum": 0.10664996847056073, "maximum": 0.25409325423612733, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.524295308783715, "standard_deviation": 0.006731257793683337, "minimum": 5.51337855006059, "maximum": 5.532787191159365, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3420873063624898, "standard_deviation": 0.01947712355808127, "minimum": 0.31592758749225236, "maximum": 0.37553707823956356, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07277824945295812, "standard_deviation": 0.05923220319794669, "minimum": -0.16776195383690795, "maximum": 0.047681653767261106, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900700668124591, "standard_deviation": 0.0024835336790731116, "minimum": 0.9847863176344219, "maximum": 0.9926179328482382, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.001498076323329, "standard_deviation": 0.04466308390550925, "minimum": 0.9516869257373813, "maximum": 1.0861660465549239, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3190866002328532, "standard_deviation": 0.09237364925455496, "minimum": 0.24194480974850788, "maximum": 0.5176038143964055, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09221849488525255, "standard_deviation": 0.011032793208364481, "minimum": 0.07992713591685557, "maximum": 0.11458376111513341, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970371157975372, "standard_deviation": 0.001103318666101939, "minimum": 0.9949333679527336, "maximum": 0.9981591359489046, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5325914901738835, "standard_deviation": 0.04847611060654198, "minimum": 0.4482896552264272, "maximum": 0.5890718912449256, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4744297158702663, "standard_deviation": 0.2856676570017175, "minimum": 0.2219015020954794, "maximum": 1.1318653711258175, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048906546771526074, "standard_deviation": 0.008402330481677512, "minimum": 0.03892116863302411, "maximum": 0.06380727549716714, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_672-122797-0056", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.9422268867492676, "predicted_mos": 2.9422268867492676, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "8e9f96cb0c04eaaa35d3b23d976422ab6c90fa4ec5d569bbef8ae15e9ef392e0", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.65878379996866, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.010915720086726, "全帯域_10ms_rms_envelope_nrmse": 0.08961195780237552, "全帯域_10ms_envelope_correlation": 0.9906833939786739, "全帯域_10ms_modulation_mae_db": 0.35408627342479987, "全帯域_40ms_log_rms_mae_db": 0.5363989002844632, "全帯域_40ms_rms_envelope_nrmse": 0.04197291152330846, "全帯域_40ms_envelope_correlation": 0.9977919196534131, "全帯域_40ms_modulation_mae_db": 0.30261085144537314, "300–1000 Hz_10ms_log_rms_mae_db": 1.3082829651709507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08790683166517956, "300–1000 Hz_10ms_envelope_correlation": 0.9936898471991775, "300–1000 Hz_10ms_modulation_mae_db": 0.7276651166677534, "300–1000 Hz_40ms_log_rms_mae_db": 0.7379871005187504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04730213211560194, "300–1000 Hz_40ms_envelope_correlation": 0.9981911554444747, "300–1000 Hz_40ms_modulation_mae_db": 0.38885701495669683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3161711535916016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1388371251719541, "1000–3000 Hz_10ms_envelope_correlation": 0.984878700770451, "1000–3000 Hz_10ms_modulation_mae_db": 0.9297288499953962, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7223028171817781, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08878355375442346, "1000–3000 Hz_40ms_envelope_correlation": 0.9937793287410532, "1000–3000 Hz_40ms_modulation_mae_db": 0.8492610092422165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4490442786014448, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18019577414495613, "3000–6000 Hz_10ms_envelope_correlation": 0.9813039168228399, "3000–6000 Hz_10ms_modulation_mae_db": 0.7672508676288398, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9501715948440929, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09523805062336702, "3000–6000 Hz_40ms_envelope_correlation": 0.9946456195390352, "3000–6000 Hz_40ms_modulation_mae_db": 0.40747995559315564, "mr_stft_db": 5.491137263327865, "band_mae_db": 0.20331854797025872, "rms_gain_error_db": -0.033126882186354484, "output_peak": 0.3094807598788179}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0703810759547103, "全帯域_10ms_rms_envelope_nrmse": 0.09913673229982965, "全帯域_10ms_envelope_correlation": 0.9885594067692434, "全帯域_10ms_modulation_mae_db": 0.39971974223307205, "全帯域_40ms_log_rms_mae_db": 0.6379491141304405, "全帯域_40ms_rms_envelope_nrmse": 0.05451303709467455, "全帯域_40ms_envelope_correlation": 0.9962783398408062, "全帯域_40ms_modulation_mae_db": 0.4247394424674731, "300–1000 Hz_10ms_log_rms_mae_db": 1.4270205717033837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09264608103932027, "300–1000 Hz_10ms_envelope_correlation": 0.9929777226684084, "300–1000 Hz_10ms_modulation_mae_db": 0.36822511307972183, "300–1000 Hz_40ms_log_rms_mae_db": 0.7780761454221928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051020973542986854, "300–1000 Hz_40ms_envelope_correlation": 0.9978637367151711, "300–1000 Hz_40ms_modulation_mae_db": 0.6975590773147621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3606681820311464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17495035033329612, "1000–3000 Hz_10ms_envelope_correlation": 0.9756083127386274, "1000–3000 Hz_10ms_modulation_mae_db": 0.530150048839691, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7383476830670841, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11813236953146812, "1000–3000 Hz_40ms_envelope_correlation": 0.988214957187589, "1000–3000 Hz_40ms_modulation_mae_db": 0.40485803294536904, "3000–6000 Hz_10ms_log_rms_mae_db": 1.458088270249665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1762265496455923, "3000–6000 Hz_10ms_envelope_correlation": 0.9821270202173635, "3000–6000 Hz_10ms_modulation_mae_db": 0.8032760148212755, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9148255795221583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09422279417105725, "3000–6000 Hz_40ms_envelope_correlation": 0.9947632196793235, "3000–6000 Hz_40ms_modulation_mae_db": 1.2199991902143954, "mr_stft_db": 5.477406414913832, "band_mae_db": 0.09630429821592908, "rms_gain_error_db": -0.04710152845632487, "output_peak": 0.3532460369889068}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.060392282050638, "全帯域_10ms_rms_envelope_nrmse": 0.09862294045687575, "全帯域_10ms_envelope_correlation": 0.9887179636539888, "全帯域_10ms_modulation_mae_db": 0.19960354692932158, "全帯域_40ms_log_rms_mae_db": 0.5522644162887791, "全帯域_40ms_rms_envelope_nrmse": 0.05086633715958834, "全帯域_40ms_envelope_correlation": 0.99675909296815, "全帯域_40ms_modulation_mae_db": 0.2577066114083611, "300–1000 Hz_10ms_log_rms_mae_db": 1.408044537524487, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11325373157599425, "300–1000 Hz_10ms_envelope_correlation": 0.9895013277224817, "300–1000 Hz_10ms_modulation_mae_db": 1.4714393216444057, "300–1000 Hz_40ms_log_rms_mae_db": 0.7270627546355537, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0631771626279493, "300–1000 Hz_40ms_envelope_correlation": 0.9967796326015262, "300–1000 Hz_40ms_modulation_mae_db": 1.0038451085046376, "1000–3000 Hz_10ms_log_rms_mae_db": 1.306602371756519, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1486623966978531, "1000–3000 Hz_10ms_envelope_correlation": 0.9826318912175473, "1000–3000 Hz_10ms_modulation_mae_db": 1.0240400575912925, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7005128228980334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07386293449193787, "1000–3000 Hz_40ms_envelope_correlation": 0.9960191869357169, "1000–3000 Hz_40ms_modulation_mae_db": 0.5011299287937785, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4420316737289423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1849591513479797, "3000–6000 Hz_10ms_envelope_correlation": 0.9802833607209941, "3000–6000 Hz_10ms_modulation_mae_db": 0.47993583466062617, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8827336695326439, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10548934987480041, "3000–6000 Hz_40ms_envelope_correlation": 0.9934298400127056, "3000–6000 Hz_40ms_modulation_mae_db": 0.44509176183453264, "mr_stft_db": 5.486390706208115, "band_mae_db": 0.1899473086392679, "rms_gain_error_db": -0.03347834532535822, "output_peak": 0.3682052622192241}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0373141747153083, "全帯域_10ms_rms_envelope_nrmse": 0.10185215812504649, "全帯域_10ms_envelope_correlation": 0.9879859113407558, "全帯域_10ms_modulation_mae_db": 0.6105823026979044, "全帯域_40ms_log_rms_mae_db": 0.5634882316494939, "全帯域_40ms_rms_envelope_nrmse": 0.05715629121105417, "全帯域_40ms_envelope_correlation": 0.9959161600703867, "全帯域_40ms_modulation_mae_db": 0.39757000534605647, "300–1000 Hz_10ms_log_rms_mae_db": 1.4002527208269269, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0954775760129786, "300–1000 Hz_10ms_envelope_correlation": 0.9928360290018086, "300–1000 Hz_10ms_modulation_mae_db": 0.46217150383936206, "300–1000 Hz_40ms_log_rms_mae_db": 0.7778077310427827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05675422719642251, "300–1000 Hz_40ms_envelope_correlation": 0.9976869403916302, "300–1000 Hz_40ms_modulation_mae_db": 0.868098364315261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3615402062983493, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17600455332010406, "1000–3000 Hz_10ms_envelope_correlation": 0.9754013282550926, "1000–3000 Hz_10ms_modulation_mae_db": 0.8120766377925906, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7822378669855529, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09956107454056659, "1000–3000 Hz_40ms_envelope_correlation": 0.9921769410185296, "1000–3000 Hz_40ms_modulation_mae_db": 0.6177728366986859, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4756732342383778, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19879622937257269, "3000–6000 Hz_10ms_envelope_correlation": 0.9781694091381972, "3000–6000 Hz_10ms_modulation_mae_db": 1.1355174359296274, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9285610954843959, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11876870539047917, "3000–6000 Hz_40ms_envelope_correlation": 0.9922148807326782, "3000–6000 Hz_40ms_modulation_mae_db": 0.8582025147399704, "mr_stft_db": 5.482372186205801, "band_mae_db": 0.23401851394541773, "rms_gain_error_db": -0.03617928878299492, "output_peak": 0.4069268143802684}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9567722465580041, "全帯域_10ms_rms_envelope_nrmse": 0.09776719850107683, "全帯域_10ms_envelope_correlation": 0.9894289599675969, "全帯域_10ms_modulation_mae_db": 0.46525202188850506, "全帯域_40ms_log_rms_mae_db": 0.48650613703088846, "全帯域_40ms_rms_envelope_nrmse": 0.05664593215735574, "全帯域_40ms_envelope_correlation": 0.9963116118536923, "全帯域_40ms_modulation_mae_db": 0.5621293239052254, "300–1000 Hz_10ms_log_rms_mae_db": 1.393272893762825, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0971573225729704, "300–1000 Hz_10ms_envelope_correlation": 0.9923303884005802, "300–1000 Hz_10ms_modulation_mae_db": 0.5870777361448725, "300–1000 Hz_40ms_log_rms_mae_db": 0.7768311680055563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04959080442759516, "300–1000 Hz_40ms_envelope_correlation": 0.9981064656655404, "300–1000 Hz_40ms_modulation_mae_db": 0.31181071981099495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2703418487243299, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15200520320667305, "1000–3000 Hz_10ms_envelope_correlation": 0.9815418494181591, "1000–3000 Hz_10ms_modulation_mae_db": 0.5244228311191902, "1000–3000 Hz_40ms_log_rms_mae_db": 0.712554463322035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08431482841938263, "1000–3000 Hz_40ms_envelope_correlation": 0.9940710191125525, "1000–3000 Hz_40ms_modulation_mae_db": 0.6395758662624408, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4568817770142288, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1966123450887227, "3000–6000 Hz_10ms_envelope_correlation": 0.9830802203946949, "3000–6000 Hz_10ms_modulation_mae_db": 1.0510056715526517, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9197866487242764, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1321482905359611, "3000–6000 Hz_40ms_envelope_correlation": 0.9941791703699627, "3000–6000 Hz_40ms_modulation_mae_db": 1.3423251883174718, "mr_stft_db": 5.492083506162761, "band_mae_db": 0.24646179713449434, "rms_gain_error_db": 0.10023346389101298, "output_peak": 0.3503129417425863}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9854826082342456, "全帯域_10ms_rms_envelope_nrmse": 0.10047121917556318, "全帯域_10ms_envelope_correlation": 0.9883595839078941, "全帯域_10ms_modulation_mae_db": 0.15589289282679508, "全帯域_40ms_log_rms_mae_db": 0.5647530624581575, "全帯域_40ms_rms_envelope_nrmse": 0.05427491200253178, "全帯域_40ms_envelope_correlation": 0.9964502160349502, "全帯域_40ms_modulation_mae_db": 0.30784506786201765, "300–1000 Hz_10ms_log_rms_mae_db": 1.3792705620415677, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10208056331646745, "300–1000 Hz_10ms_envelope_correlation": 0.9914723213519795, "300–1000 Hz_10ms_modulation_mae_db": 0.8823262532624992, "300–1000 Hz_40ms_log_rms_mae_db": 0.7742508940621272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06181750788392162, "300–1000 Hz_40ms_envelope_correlation": 0.996840889160431, "300–1000 Hz_40ms_modulation_mae_db": 0.6891338754049968, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3025761625083179, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15821163599143515, "1000–3000 Hz_10ms_envelope_correlation": 0.9812874914659886, "1000–3000 Hz_10ms_modulation_mae_db": 0.4951259469684531, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7309315923014803, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09049156747944742, "1000–3000 Hz_40ms_envelope_correlation": 0.995142914120757, "1000–3000 Hz_40ms_modulation_mae_db": 0.6941428290677539, "3000–6000 Hz_10ms_log_rms_mae_db": 1.469021960576061, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20493957360800755, "3000–6000 Hz_10ms_envelope_correlation": 0.975646753397959, "3000–6000 Hz_10ms_modulation_mae_db": 0.2330875584746254, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8724341948243587, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11325764812148885, "3000–6000 Hz_40ms_envelope_correlation": 0.9924297602151215, "3000–6000 Hz_40ms_modulation_mae_db": 0.2635235461082717, "mr_stft_db": 5.490566548636855, "band_mae_db": 0.19146043834733084, "rms_gain_error_db": -0.10168440237443394, "output_peak": 0.33804853996976325}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.087271389740852, "全帯域_10ms_rms_envelope_nrmse": 0.1043374853210042, "全帯域_10ms_envelope_correlation": 0.9873452013141654, "全帯域_10ms_modulation_mae_db": 0.16474701219934465, "全帯域_40ms_log_rms_mae_db": 0.5695908157318125, "全帯域_40ms_rms_envelope_nrmse": 0.05021383082437555, "全帯域_40ms_envelope_correlation": 0.9969218541672951, "全帯域_40ms_modulation_mae_db": 0.3096035201514496, "300–1000 Hz_10ms_log_rms_mae_db": 1.3666162160613438, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1014848830583924, "300–1000 Hz_10ms_envelope_correlation": 0.9915619863120452, "300–1000 Hz_10ms_modulation_mae_db": 0.6104299770095598, "300–1000 Hz_40ms_log_rms_mae_db": 0.7573411435911761, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06026829935713278, "300–1000 Hz_40ms_envelope_correlation": 0.9969954526143977, "300–1000 Hz_40ms_modulation_mae_db": 0.23043495992644777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3456465590206088, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14388283080736047, "1000–3000 Hz_10ms_envelope_correlation": 0.9850601782993152, "1000–3000 Hz_10ms_modulation_mae_db": 1.3554939715077066, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8023974947173022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0874853501207622, "1000–3000 Hz_40ms_envelope_correlation": 0.9957030826960317, "1000–3000 Hz_40ms_modulation_mae_db": 1.3212936128536101, "3000–6000 Hz_10ms_log_rms_mae_db": 1.410660416846604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20514752206819534, "3000–6000 Hz_10ms_envelope_correlation": 0.9755666653224907, "3000–6000 Hz_10ms_modulation_mae_db": 0.2281247554431046, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8615984429856245, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0935726832912817, "3000–6000 Hz_40ms_envelope_correlation": 0.9951779107531087, "3000–6000 Hz_40ms_modulation_mae_db": 0.40432481047200525, "mr_stft_db": 5.469500795553238, "band_mae_db": 0.2505364158708466, "rms_gain_error_db": -0.08868247936442028, "output_peak": 0.36988451080531326}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0491350826271326, "全帯域_10ms_rms_envelope_nrmse": 0.10484563944491199, "全帯域_10ms_envelope_correlation": 0.9877215039409942, "全帯域_10ms_modulation_mae_db": 0.2502511547879393, "全帯域_40ms_log_rms_mae_db": 0.5550276947975039, "全帯域_40ms_rms_envelope_nrmse": 0.06686006768431645, "全帯域_40ms_envelope_correlation": 0.9947471814898886, "全帯域_40ms_modulation_mae_db": 0.4836707871449802, "300–1000 Hz_10ms_log_rms_mae_db": 1.4025712658734242, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0991401457133811, "300–1000 Hz_10ms_envelope_correlation": 0.9921290266793998, "300–1000 Hz_10ms_modulation_mae_db": 0.668444666558064, "300–1000 Hz_40ms_log_rms_mae_db": 0.7992660403353878, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06016574670540344, "300–1000 Hz_40ms_envelope_correlation": 0.9972195690920855, "300–1000 Hz_40ms_modulation_mae_db": 0.36377952179586054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3440516613182698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1531097018371666, "1000–3000 Hz_10ms_envelope_correlation": 0.981811664111686, "1000–3000 Hz_10ms_modulation_mae_db": 0.7561608459398483, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7485893324406293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08093566044551247, "1000–3000 Hz_40ms_envelope_correlation": 0.9954470033723302, "1000–3000 Hz_40ms_modulation_mae_db": 0.684113716351023, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4411743859878188, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20977558779919864, "3000–6000 Hz_10ms_envelope_correlation": 0.9811140740001891, "3000–6000 Hz_10ms_modulation_mae_db": 0.9964501791260948, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8948417995117608, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15322768373034956, "3000–6000 Hz_40ms_envelope_correlation": 0.9917265964972737, "3000–6000 Hz_40ms_modulation_mae_db": 0.7077829567278742, "mr_stft_db": 5.485155464544806, "band_mae_db": 0.3612636788050339, "rms_gain_error_db": 0.04020439252520161, "output_peak": 0.3746325677364236}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83600, "output_samples": 83600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.611039549777963, "spectral_convergence": 0.36379111517034846, "spectral_convergence_status": "ok", "amplitude_floor": 9.102684030105167e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.216455593193679, "spectral_convergence": 0.3315198272383218, "spectral_convergence_status": "ok", "amplitude_floor": 4.6649711782186323e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.64591664701195, "spectral_convergence": 0.3483137284305067, "spectral_convergence_status": "ok", "amplitude_floor": 4.1057122515023654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021084017487574348, "output_power": 0.00021656940408865458, "delta_db": 0.11643733774999247, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00047622157098200406, "output_power": 0.00047187989715963567, "delta_db": -0.039775872669643476, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 9.587972951197478e-05, "output_power": 9.477791660230292e-05, "delta_db": -0.05019642312427283, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.411215555329107e-05, "output_power": 6.925115543680124e-05, "delta_db": -0.2946242102985275, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013227809814888984, "output_power": 0.0001190401133434041, "delta_db": -0.4579461042977648, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00014809096583739848, "output_power": 0.0001572612413731323, "delta_db": 0.2609313396813513, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20331854797025872, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.033126882186354484, "pitch": null, "output_rms_dbfs": -29.473907987709712, "output_peak": 0.3094807598788179, "reference_rms_dbfs": -29.44078110552336, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.491137263327865}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011323786448874551, "log_rms_mae_db": 1.010915720086726, "rms_envelope_nrmse": 0.08961195780237552, "envelope_correlation": 0.9906833939786739, "rms_gain_error_db": -0.03321578588252701, "reference_mean_power": 0.0011384943575262435, "output_mean_power": 0.0011298201187517891}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 5.62759551802579e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002928658523286973, "output_power": 0.0002871482640390085, "reference_active": true, "delta_db": -0.08562541530716027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4610498849864094e-05, "output_power": 4.325909138113568e-05, "reference_active": true, "delta_db": -0.133596864699945}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2101686887119984e-05, "output_power": 1.4694322461798698e-05, "reference_active": true, "delta_db": 0.8430365402672945}], "active_band_count": 3, "active_band_mae_db": 0.35408627342479987}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.563924945576787e-05, "log_rms_mae_db": 0.5363989002844632, "rms_envelope_nrmse": 0.04197291152330846, "envelope_correlation": 0.9977919196534131, "rms_gain_error_db": -0.033209281068986775, "reference_mean_power": 0.0011449801518647084, "output_mean_power": 0.0011362581994102234}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.095010942049215e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002727818973842863, "output_power": 0.0002689364352753049, "reference_active": true, "delta_db": -0.061659018213518824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1267992599082604e-05, "output_power": 2.03536420076767e-05, "reference_active": true, "delta_db": -0.1908436890352819}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.757168583210849e-07, "output_power": 6.694866656890567e-07, "reference_active": true, "delta_db": 0.6553298470873187}], "active_band_count": 3, "active_band_mae_db": 0.30261085144537314}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.979916609312644e-05, "log_rms_mae_db": 1.3082829651709507, "rms_envelope_nrmse": 0.08790683166517956, "envelope_correlation": 0.9936898471991775, "rms_gain_error_db": -0.11689906299565762, "reference_mean_power": 0.00038382289364078325, "output_mean_power": 0.00037362933673930506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 2.8601579771498626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014700129208541836, "output_power": 0.00014316680576933327, "reference_active": true, "delta_db": -0.1147881663440744}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.957621157356578e-05, "output_power": 1.7880257809483307e-05, "reference_active": true, "delta_db": -0.3935487363532869}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3113428408893907e-06, "output_power": 3.3988360327465204e-06, "reference_active": true, "delta_db": 1.6746584473058987}], "active_band_count": 3, "active_band_mae_db": 0.7276651166677534}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.59156861629162e-05, "log_rms_mae_db": 0.7379871005187504, "rms_envelope_nrmse": 0.04730213211560194, "envelope_correlation": 0.9981911554444747, "rms_gain_error_db": -0.11690297707182953, "reference_mean_power": 0.00038603855676420007, "output_mean_power": 0.000375785817671046}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 2.667606930039739e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013762626674571402, "output_power": 0.00013449018473533805, "reference_active": true, "delta_db": -0.10010739218963868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.746730510582489e-06, "output_power": 9.229653431324908e-06, "reference_active": true, "delta_db": -0.23673564529807906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5949673646594045e-07, "output_power": 1.9307435911476013e-07, "reference_active": true, "delta_db": 0.8297280073823727}], "active_band_count": 3, "active_band_mae_db": 0.38885701495669683}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.341026161507385e-05, "log_rms_mae_db": 1.3161711535916016, "rms_envelope_nrmse": 0.1388371251719541, "envelope_correlation": 0.984878700770451, "rms_gain_error_db": -0.24348376955658008, "reference_mean_power": 0.0001433729237288462, "output_mean_power": 0.00013555600610690055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.1455083323281239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.589033152322262e-05, "output_power": 5.0308384686549935e-05, "reference_active": true, "delta_db": -0.45696312596582567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4083924473542424e-05, "output_power": 1.2287904287884671e-05, "reference_active": true, "delta_db": -0.5924586753334976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.499188380560451e-06, "output_power": 3.684021213347758e-06, "reference_active": true, "delta_db": -1.7397647486868653}], "active_band_count": 3, "active_band_mae_db": 0.9297288499953962}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 5.482542724448801e-05, "log_rms_mae_db": 0.7223028171817781, "rms_envelope_nrmse": 0.08878355375442346, "envelope_correlation": 0.9937793287410532, "rms_gain_error_db": -0.24344022667087808, "reference_mean_power": 0.00014418079784839864, "output_mean_power": 0.00013632120042286457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.042775196620521e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2049448137923216e-05, "output_power": 4.6693411197373245e-05, "reference_active": true, "delta_db": -0.47160526630408683}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.197217517258297e-06, "output_power": 5.4616189277079706e-06, "reference_active": true, "delta_db": -0.5487534486681857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5366430302416474e-07, "output_power": 1.7845015275043483e-07, "reference_active": true, "delta_db": -1.5274243127543772}], "active_band_count": 3, "active_band_mae_db": 0.8492610092422165}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011141137328067575, "log_rms_mae_db": 1.4490442786014448, "rms_envelope_nrmse": 0.18019577414495613, "envelope_correlation": 0.9813039168228399, "rms_gain_error_db": -0.02016588712809731, "reference_mean_power": 0.00018590462562984253, "output_mean_power": 0.00018504340323906816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.861645592580093e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.396584441083375e-05, "output_power": 8.827133959471108e-05, "reference_active": true, "delta_db": -0.2715030340345762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7617019162087818e-05, "output_power": 1.4144603118117718e-05, "reference_active": true, "delta_db": -0.9534166058007044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.489954892102086e-06, "output_power": 8.316200943293199e-06, "reference_active": true, "delta_db": 1.0768329630512388}], "active_band_count": 3, "active_band_mae_db": 0.7672508676288398}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.054702623382987e-05, "log_rms_mae_db": 0.9501715948440929, "rms_envelope_nrmse": 0.09523805062336702, "envelope_correlation": 0.9946456195390352, "rms_gain_error_db": -0.020166304862075483, "reference_mean_power": 0.00018697687803543306, "output_mean_power": 0.00018611067042318527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.7928912528322647e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.176227386223464e-05, "output_power": 8.774515549941484e-05, "reference_active": true, "delta_db": -0.19441018857866446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.193273720945643e-06, "output_power": 6.686259221574306e-06, "reference_active": true, "delta_db": -0.882742541398356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.787097063305561e-07, "output_power": 3.6625007521548226e-07, "reference_active": true, "delta_db": -0.14528713680244648}], "active_band_count": 3, "active_band_mae_db": 0.40747995559315564}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8526888344772439, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a800e839dcb651cd38f149991ab6b44879e5aa03fa15aeebf9aff900ea493bef", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9810276770346085, "standard_deviation": 0.003696267922527207, "minimum": 0.9754013282550926, "maximum": 0.9850601782993152, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3259497681561427, "standard_deviation": 0.03228488998703109, "minimum": 1.2703418487243299, "maximum": 1.3615402062983493, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.803399898719271, "standard_deviation": 0.2974372146924891, "minimum": 0.4951259469684531, "maximum": 1.3554939715077066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15570797467073033, "standard_deviation": 0.013538567889143477, "minimum": 0.1388371251719541, "maximum": 0.17600455332010406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938193041480701, "standard_deviation": 0.0025901040790129344, "minimum": 0.988214957187589, "maximum": 0.9960191869357169, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7422342591142368, "standard_deviation": 0.03468630269820966, "minimum": 0.7005128228980334, "maximum": 0.8023974947173022, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7140184790268598, "standard_deviation": 0.2787906559561825, "minimum": 0.40485803294536904, "maximum": 1.3212936128536101, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0904459173479376, "standard_deviation": 0.013438679666851768, "minimum": 0.07386293449193787, "maximum": 0.11813236953146812, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9796614275018412, "standard_deviation": 0.002880298855877366, "minimum": 0.9755666653224907, "maximum": 0.9830802203946949, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4503219996553929, "standard_deviation": 0.020081268219146395, "minimum": 1.410660416846604, "maximum": 1.4756732342383778, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7118310397046057, "standard_deviation": 0.35935133324571905, "minimum": 0.2281247554431046, "maximum": 1.1355174359296274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19458159163440314, "standard_deviation": 0.012581857659377063, "minimum": 0.1762265496455923, "maximum": 0.20977558779919864, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935708747249011, "standard_deviation": 0.001314287764528296, "minimum": 0.9917265964972737, "maximum": 0.9951779107531087, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9031191281786639, "standard_deviation": 0.030297781841979832, "minimum": 0.8615984429856245, "maximum": 0.9501715948440929, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7060912405009596, "standard_deviation": 0.40289209624878386, "minimum": 0.2635235461082717, "maximum": 1.3423251883174718, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11324065071734814, "standard_deviation": 0.021110949817089632, "minimum": 0.0935726832912817, "maximum": 0.15322768373034956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920623311669852, "standard_deviation": 0.0012716050906414705, "minimum": 0.9895013277224817, "maximum": 0.9936898471991775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3856664666206135, "standard_deviation": 0.0361729077514405, "minimum": 1.3082829651709507, "maximum": 1.4270205717033837, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7222224610257798, "standard_deviation": 0.34091957179582166, "minimum": 0.36822511307972183, "maximum": 1.4714393216444057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0986433918693355, "standard_deviation": 0.007531545702284997, "minimum": 0.08790683166517956, "maximum": 0.11325373157599425, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974604802106571, "standard_deviation": 0.000571566631029898, "minimum": 0.9967796326015262, "maximum": 0.9981911554444747, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7660778722016908, "standard_deviation": 0.02376382481543045, "minimum": 0.7270627546355537, "maximum": 0.7992660403353878, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5691898302537072, "standard_deviation": 0.2840646173875915, "minimum": 0.23043495992644777, "maximum": 1.0038451085046376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0562621067321267, "standard_deviation": 0.006123345766283449, "minimum": 0.04730213211560194, "maximum": 0.0631771626279493, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2216638748660724, "standard_deviation": 0.07469850406520753, "minimum": 0.09630429821592908, "maximum": 0.3612636788050339, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.4843266106941595, "standard_deviation": 0.007763949753687369, "minimum": 5.469500795553238, "maximum": 5.492083506162761, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3588421792151629, "standard_deviation": 0.02864576451002797, "minimum": 0.3094807598788179, "maximum": 0.4069268143802684, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.024976883759209017, "standard_deviation": 0.06614093973660409, "minimum": -0.10168440237443394, "maximum": 0.10023346389101298, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9886002406091641, "standard_deviation": 0.0010572503228100224, "minimum": 0.9873452013141654, "maximum": 0.9906833939786739, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0322080724959521, "standard_deviation": 0.04459867044091909, "minimum": 0.9567722465580041, "maximum": 1.087271389740852, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3250168683734602, "standard_deviation": 0.16179653195188623, "minimum": 0.15589289282679508, "maximum": 0.6105823026979044, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09958066639083546, "standard_deviation": 0.004782082028723706, "minimum": 0.08961195780237552, "maximum": 0.10484563944491199, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963970470098228, "standard_deviation": 0.0008725738205211048, "minimum": 0.9947471814898886, "maximum": 0.9977919196534131, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5582472965464423, "standard_deviation": 0.04178371970039823, "minimum": 0.48650613703088846, "maximum": 0.6379491141304405, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38073445121636706, "standard_deviation": 0.10510639643375061, "minimum": 0.2577066114083611, "maximum": 0.5621293239052254, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.054062914957150626, "standard_deviation": 0.007085774142810784, "minimum": 0.04197291152330846, "maximum": 0.06686006768431645, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_6829-68769-0014", "language": "en", "style": "normal", "speaker": "librispeech_6829", "source_utterance_id": "6829-68769-0014", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 58480, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_6829-68769-0014.wav", "source_file_sha256": "cf6129ce3ecac87c2a17b4f9a0bbc5b076e0b7fe8f8333e69944d7f38fbc7b48", "decoded_pcm_sha256": "f8c838b07c2dd578b611a15c028e07c92a592c03766bbc5d9020f61b8ed5195e", "original_source_sha256": "c228d6cd297ffb4916b952d7d5861d8ac8225058219ec05b587d8dd4bd041179", "corpus_entry_id": "librispeech_6829-68769-0014"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.7640280723571777, "predicted_mos": 3.7640280723571777, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "6c2326dfad893a99b40c2c32714bcb701c5566c86e84cebc17cbca7a7608aa1c", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5475116380002873, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_6829-68769-0014", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6c2326dfad893a99b40c2c32714bcb701c5566c86e84cebc17cbca7a7608aa1c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "3e63bfef2739048aec1a1dd1664bc947427270be279a01e2399186b1cc779f6b"}, "rows": [{"case": "librispeech_6829-68769-0014", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.5310842990875244, "predicted_mos": 2.5310842990875244, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "2d105515376a025772532d182aa6e5952427ba8cc6119a1c68977780d6231098", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4908762679988286, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_6829-68769-0014", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "2d105515376a025772532d182aa6e5952427ba8cc6119a1c68977780d6231098", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8776758901019865, "全帯域_10ms_rms_envelope_nrmse": 0.07688606302471769, "全帯域_10ms_envelope_correlation": 0.9940986776335241, "全帯域_10ms_modulation_mae_db": 0.425445271726826, "全帯域_40ms_log_rms_mae_db": 0.46873946605454236, "全帯域_40ms_rms_envelope_nrmse": 0.04167408276123258, "全帯域_40ms_envelope_correlation": 0.9980748848599552, "全帯域_40ms_modulation_mae_db": 0.4190726033797684, "300–1000 Hz_10ms_log_rms_mae_db": 1.622780381298416, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0865979884039179, "300–1000 Hz_10ms_envelope_correlation": 0.9939835358751608, "300–1000 Hz_10ms_modulation_mae_db": 0.786895294771773, "300–1000 Hz_40ms_log_rms_mae_db": 1.030285911717904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053930212661356475, "300–1000 Hz_40ms_envelope_correlation": 0.9974598487162218, "300–1000 Hz_40ms_modulation_mae_db": 1.0226346654149703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4965961220906847, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1803004598571424, "1000–3000 Hz_10ms_envelope_correlation": 0.9742627666698808, "1000–3000 Hz_10ms_modulation_mae_db": 0.4880879401032468, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8568275649865142, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11071135074323345, "1000–3000 Hz_40ms_envelope_correlation": 0.9894371151439172, "1000–3000 Hz_40ms_modulation_mae_db": 0.4847305787516592, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6528226125663859, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16434627576092048, "3000–6000 Hz_10ms_envelope_correlation": 0.9811253685804717, "3000–6000 Hz_10ms_modulation_mae_db": 0.2713518896151272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0214834032161206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0990304758382139, "3000–6000 Hz_40ms_envelope_correlation": 0.9935655171100543, "3000–6000 Hz_40ms_modulation_mae_db": 0.48239859545200997, "mr_stft_db": 5.663306440812789, "band_mae_db": 0.15768184099593394, "rms_gain_error_db": -0.006251521424782695, "output_peak": 0.5099358754557584}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8822438995600415, "全帯域_10ms_rms_envelope_nrmse": 0.08318237283657642, "全帯域_10ms_envelope_correlation": 0.9930670958882822, "全帯域_10ms_modulation_mae_db": 0.6603533877150648, "全帯域_40ms_log_rms_mae_db": 0.4891153649701937, "全帯域_40ms_rms_envelope_nrmse": 0.043944717408436, "全帯域_40ms_envelope_correlation": 0.9978830826148251, "全帯域_40ms_modulation_mae_db": 0.9184255188375308, "300–1000 Hz_10ms_log_rms_mae_db": 1.6109961605935814, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09412146450053455, "300–1000 Hz_10ms_envelope_correlation": 0.9928912641616006, "300–1000 Hz_10ms_modulation_mae_db": 0.7626386434435846, "300–1000 Hz_40ms_log_rms_mae_db": 0.9552822987028081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0563941985117382, "300–1000 Hz_40ms_envelope_correlation": 0.9972874872334646, "300–1000 Hz_40ms_modulation_mae_db": 0.9503579877067679, "1000–3000 Hz_10ms_log_rms_mae_db": 1.557039546360496, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19000781591093716, "1000–3000 Hz_10ms_envelope_correlation": 0.9718661657546023, "1000–3000 Hz_10ms_modulation_mae_db": 0.5436447933569395, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9060945045412305, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11461937683804792, "1000–3000 Hz_40ms_envelope_correlation": 0.9887719571498623, "1000–3000 Hz_40ms_modulation_mae_db": 0.5238968663174074, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6168115232979077, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16082515808503087, "3000–6000 Hz_10ms_envelope_correlation": 0.9794909833488983, "3000–6000 Hz_10ms_modulation_mae_db": 0.25952589090571726, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9288406510792367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08057664116658354, "3000–6000 Hz_40ms_envelope_correlation": 0.9945121723369365, "3000–6000 Hz_40ms_modulation_mae_db": 0.24153964875179504, "mr_stft_db": 5.6818810066323975, "band_mae_db": 0.0789339308368052, "rms_gain_error_db": -0.06845554557571559, "output_peak": 0.47619972821122597}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.886148183068175, "全帯域_10ms_rms_envelope_nrmse": 0.08555334190559805, "全帯域_10ms_envelope_correlation": 0.9927029798559924, "全帯域_10ms_modulation_mae_db": 0.22987311142460634, "全帯域_40ms_log_rms_mae_db": 0.49841991967500127, "全帯域_40ms_rms_envelope_nrmse": 0.054148901052059356, "全帯域_40ms_envelope_correlation": 0.996809072346727, "全帯域_40ms_modulation_mae_db": 0.3166798051906835, "300–1000 Hz_10ms_log_rms_mae_db": 1.6691236704451722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09439952785884516, "300–1000 Hz_10ms_envelope_correlation": 0.9928849584392656, "300–1000 Hz_10ms_modulation_mae_db": 0.5883813718276046, "300–1000 Hz_40ms_log_rms_mae_db": 1.0539975348022295, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06074551037626734, "300–1000 Hz_40ms_envelope_correlation": 0.9968620206691562, "300–1000 Hz_40ms_modulation_mae_db": 0.9133099164427051, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5797102777464145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19145021464189896, "1000–3000 Hz_10ms_envelope_correlation": 0.9712644347749647, "1000–3000 Hz_10ms_modulation_mae_db": 0.41222380980553114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9156023687244946, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11452876004035019, "1000–3000 Hz_40ms_envelope_correlation": 0.9886947237924051, "1000–3000 Hz_40ms_modulation_mae_db": 0.7215051456262933, "3000–6000 Hz_10ms_log_rms_mae_db": 1.65700020750409, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16872920372860964, "3000–6000 Hz_10ms_envelope_correlation": 0.9774012204939813, "3000–6000 Hz_10ms_modulation_mae_db": 0.8982540136493631, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0234401057073366, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09628216556790752, "3000–6000 Hz_40ms_envelope_correlation": 0.992069919527324, "3000–6000 Hz_40ms_modulation_mae_db": 1.0042251457885867, "mr_stft_db": 5.677427731268775, "band_mae_db": 0.1191774772741599, "rms_gain_error_db": -0.09652708377297958, "output_peak": 0.4991402066995412}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8739981613988046, "全帯域_10ms_rms_envelope_nrmse": 0.0935810954106365, "全帯域_10ms_envelope_correlation": 0.9914844097085951, "全帯域_10ms_modulation_mae_db": 1.222294674733136, "全帯域_40ms_log_rms_mae_db": 0.4695966228021611, "全帯域_40ms_rms_envelope_nrmse": 0.055278635519580864, "全帯域_40ms_envelope_correlation": 0.9969960113344383, "全帯域_40ms_modulation_mae_db": 1.5405369120529986, "300–1000 Hz_10ms_log_rms_mae_db": 1.6812503204395566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10336566755735768, "300–1000 Hz_10ms_envelope_correlation": 0.9917858041252834, "300–1000 Hz_10ms_modulation_mae_db": 1.4737910617124748, "300–1000 Hz_40ms_log_rms_mae_db": 0.9799160790150742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06557810866824508, "300–1000 Hz_40ms_envelope_correlation": 0.9967603761457455, "300–1000 Hz_40ms_modulation_mae_db": 1.855196824718063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5530222166190473, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19111309835059703, "1000–3000 Hz_10ms_envelope_correlation": 0.9710321349627795, "1000–3000 Hz_10ms_modulation_mae_db": 0.3194410818300042, "1000–3000 Hz_40ms_log_rms_mae_db": 0.925866921569705, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11898254707560696, "1000–3000 Hz_40ms_envelope_correlation": 0.9878778609550596, "1000–3000 Hz_40ms_modulation_mae_db": 0.49625488587210986, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7420113591880144, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16983061098467936, "3000–6000 Hz_10ms_envelope_correlation": 0.9772877915917643, "3000–6000 Hz_10ms_modulation_mae_db": 0.6286443803940238, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0815209112126123, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09103892244873442, "3000–6000 Hz_40ms_envelope_correlation": 0.9934394538504453, "3000–6000 Hz_40ms_modulation_mae_db": 0.6256510028698082, "mr_stft_db": 5.685274530611458, "band_mae_db": 0.20254103253234643, "rms_gain_error_db": -0.18072583329238212, "output_peak": 0.47660239348126343}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8604923350456894, "全帯域_10ms_rms_envelope_nrmse": 0.08762934614843594, "全帯域_10ms_envelope_correlation": 0.9924830195445826, "全帯域_10ms_modulation_mae_db": 1.0513846456248077, "全帯域_40ms_log_rms_mae_db": 0.45787752823631045, "全帯域_40ms_rms_envelope_nrmse": 0.04691776036842991, "全帯域_40ms_envelope_correlation": 0.9978300554345548, "全帯域_40ms_modulation_mae_db": 1.4168950690184452, "300–1000 Hz_10ms_log_rms_mae_db": 1.6575096875513131, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10456672004455768, "300–1000 Hz_10ms_envelope_correlation": 0.9914781408055958, "300–1000 Hz_10ms_modulation_mae_db": 1.4160343331215204, "300–1000 Hz_40ms_log_rms_mae_db": 0.9757320424680938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060432438527076665, "300–1000 Hz_40ms_envelope_correlation": 0.9972296840989556, "300–1000 Hz_40ms_modulation_mae_db": 1.9449706529791593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5687550737885263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1816138711548793, "1000–3000 Hz_10ms_envelope_correlation": 0.9741970559075968, "1000–3000 Hz_10ms_modulation_mae_db": 0.2975937942893883, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8907489181664756, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11230413136986503, "1000–3000 Hz_40ms_envelope_correlation": 0.9891565083129402, "1000–3000 Hz_40ms_modulation_mae_db": 0.27814245062813797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5956215873867279, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1363154991574689, "3000–6000 Hz_10ms_envelope_correlation": 0.9853443858409718, "3000–6000 Hz_10ms_modulation_mae_db": 0.3574903674456181, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9497631016797174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07701184138696099, "3000–6000 Hz_40ms_envelope_correlation": 0.9949793291585198, "3000–6000 Hz_40ms_modulation_mae_db": 0.21731357094916168, "mr_stft_db": 5.666598354790711, "band_mae_db": 0.19372051198453322, "rms_gain_error_db": -0.15034983508812128, "output_peak": 0.4883511987470048}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8720482054047881, "全帯域_10ms_rms_envelope_nrmse": 0.08365031139268192, "全帯域_10ms_envelope_correlation": 0.9929812165668482, "全帯域_10ms_modulation_mae_db": 0.4855919739631118, "全帯域_40ms_log_rms_mae_db": 0.47792324445056633, "全帯域_40ms_rms_envelope_nrmse": 0.046470501036977224, "全帯域_40ms_envelope_correlation": 0.9975746989340943, "全帯域_40ms_modulation_mae_db": 0.5119728803158402, "300–1000 Hz_10ms_log_rms_mae_db": 1.6033167257882073, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08427859543790477, "300–1000 Hz_10ms_envelope_correlation": 0.9943017114981402, "300–1000 Hz_10ms_modulation_mae_db": 0.5372112305843827, "300–1000 Hz_40ms_log_rms_mae_db": 0.9985096921412424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04842123551262075, "300–1000 Hz_40ms_envelope_correlation": 0.9979735411167447, "300–1000 Hz_40ms_modulation_mae_db": 0.656547848097027, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5860594513486874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19038945465255383, "1000–3000 Hz_10ms_envelope_correlation": 0.9719234949100473, "1000–3000 Hz_10ms_modulation_mae_db": 0.3069777192030021, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9278017689770944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.117034147570028, "1000–3000 Hz_40ms_envelope_correlation": 0.9883938653159031, "1000–3000 Hz_40ms_modulation_mae_db": 0.2989835139080876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6451642555519488, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1684962047724219, "3000–6000 Hz_10ms_envelope_correlation": 0.9776508415466975, "3000–6000 Hz_10ms_modulation_mae_db": 0.7572901215305601, "3000–6000 Hz_40ms_log_rms_mae_db": 1.004679859683095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09254550842478898, "3000–6000 Hz_40ms_envelope_correlation": 0.993212729804028, "3000–6000 Hz_40ms_modulation_mae_db": 1.0209392547641583, "mr_stft_db": 5.666829644485197, "band_mae_db": 0.16684970246396316, "rms_gain_error_db": -0.004457020665103788, "output_peak": 0.5055640877535157}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.94900245667572, "全帯域_10ms_rms_envelope_nrmse": 0.08891804637552284, "全帯域_10ms_envelope_correlation": 0.9920526578326146, "全帯域_10ms_modulation_mae_db": 0.11074626830926015, "全帯域_40ms_log_rms_mae_db": 0.5252937834175252, "全帯域_40ms_rms_envelope_nrmse": 0.05330486576783528, "全帯域_40ms_envelope_correlation": 0.9968205041139341, "全帯域_40ms_modulation_mae_db": 0.08618853340196936, "300–1000 Hz_10ms_log_rms_mae_db": 1.7658015217593375, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10498910660182911, "300–1000 Hz_10ms_envelope_correlation": 0.991119107535645, "300–1000 Hz_10ms_modulation_mae_db": 0.3451298407470871, "300–1000 Hz_40ms_log_rms_mae_db": 1.0578428209935824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06135033125576319, "300–1000 Hz_40ms_envelope_correlation": 0.9967592645716288, "300–1000 Hz_40ms_modulation_mae_db": 0.6677470704415717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.50891755282554, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1818451688518529, "1000–3000 Hz_10ms_envelope_correlation": 0.9741221397919207, "1000–3000 Hz_10ms_modulation_mae_db": 0.4642086702087151, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8471661947758339, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10796697904500147, "1000–3000 Hz_40ms_envelope_correlation": 0.9899889997999456, "1000–3000 Hz_40ms_modulation_mae_db": 0.6491920163203849, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6420660932107478, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1561973710999383, "3000–6000 Hz_10ms_envelope_correlation": 0.9806929781786469, "3000–6000 Hz_10ms_modulation_mae_db": 0.3005860408876609, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9918737433679033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08973709945970403, "3000–6000 Hz_40ms_envelope_correlation": 0.9930963011107109, "3000–6000 Hz_40ms_modulation_mae_db": 0.2934956622277171, "mr_stft_db": 5.681343214111709, "band_mae_db": 0.1429626135234691, "rms_gain_error_db": -0.04476817693985337, "output_peak": 0.7178587561593586}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8904833058375085, "全帯域_10ms_rms_envelope_nrmse": 0.09793732062132444, "全帯域_10ms_envelope_correlation": 0.9903781063848457, "全帯域_10ms_modulation_mae_db": 0.261073856127191, "全帯域_40ms_log_rms_mae_db": 0.4719846185262206, "全帯域_40ms_rms_envelope_nrmse": 0.05368013940847178, "全帯域_40ms_envelope_correlation": 0.9968315895447474, "全帯域_40ms_modulation_mae_db": 0.3549261683657061, "300–1000 Hz_10ms_log_rms_mae_db": 1.6287133871255928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1157028678405587, "300–1000 Hz_10ms_envelope_correlation": 0.9891592217364622, "300–1000 Hz_10ms_modulation_mae_db": 0.4391504089483937, "300–1000 Hz_40ms_log_rms_mae_db": 0.9754325612514443, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06752306827970127, "300–1000 Hz_40ms_envelope_correlation": 0.996025347845341, "300–1000 Hz_40ms_modulation_mae_db": 0.6415739876097882, "1000–3000 Hz_10ms_log_rms_mae_db": 1.528058882534368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18799688350884286, "1000–3000 Hz_10ms_envelope_correlation": 0.9722736107699833, "1000–3000 Hz_10ms_modulation_mae_db": 0.09380195090060241, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8734312328605961, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10755399221570353, "1000–3000 Hz_40ms_envelope_correlation": 0.990015137092239, "1000–3000 Hz_40ms_modulation_mae_db": 0.2501851387036276, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6581505144317568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1726312707869313, "3000–6000 Hz_10ms_envelope_correlation": 0.9768289270359212, "3000–6000 Hz_10ms_modulation_mae_db": 0.7891508721681832, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9913741452860284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09243378668650269, "3000–6000 Hz_40ms_envelope_correlation": 0.9927813235840508, "3000–6000 Hz_40ms_modulation_mae_db": 0.429817335731771, "mr_stft_db": 5.686769572362799, "band_mae_db": 0.07588021944948732, "rms_gain_error_db": -0.0814644008403009, "output_peak": 0.4888989384048927}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 58480, "output_samples": 58480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8248364240200585, "spectral_convergence": 0.28171910072740125, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267125480811954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.468660492340433, "spectral_convergence": 0.25975687020848964, "spectral_convergence_status": "ok", "amplitude_floor": 1.0684354905475814e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.696422406077875, "spectral_convergence": 0.2737754054563702, "spectral_convergence_status": "ok", "amplitude_floor": 7.412945744843815e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002955479965644196, "output_power": 0.0003135144806895856, "delta_db": 0.25629585201713784, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00146613378566028, "output_power": 0.001483204151774824, "delta_db": 0.050273307093314644, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0021080578166581473, "output_power": 0.0020554665672958042, "delta_db": -0.10972100623739914, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001787602167832836, "output_power": 0.0001751181865538113, "delta_db": -0.08939621437096022, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014763682412872298, "output_power": 0.00016057914401888273, "delta_db": 0.36494444189020364, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001169621027158164, "output_power": 0.00011901211952698932, "delta_db": 0.07546022436658806, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15768184099593394, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.006251521424782695, "pitch": null, "output_rms_dbfs": -23.65835751303127, "output_peak": 0.5099358754557584, "reference_rms_dbfs": -23.65210599160649, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.663306440812789}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 58480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.0002136685864137289, "log_rms_mae_db": 0.8776758901019865, "rms_envelope_nrmse": 0.07688606302471769, "envelope_correlation": 0.9940986776335241, "rms_gain_error_db": -0.00625068033062286, "reference_mean_power": 0.004318997149502425, "output_mean_power": 0.0043127854076564455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.9590597233654063e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007496792113024279, "output_power": 0.0007182834585834028, "reference_active": true, "delta_db": -0.18579602832775235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031885669741703006, "output_power": 0.00029607951304004587, "reference_active": true, "delta_db": -0.3218718573951277}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.908796664605564e-05, "output_power": 5.7880945663380524e-05, "reference_active": true, "delta_db": -0.768667929457598}], "active_band_count": 3, "active_band_mae_db": 0.425445271726826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00017810705079567896, "log_rms_mae_db": 0.46873946605454236, "rms_envelope_nrmse": 0.04167408276123258, "envelope_correlation": 0.9980748848599552, "rms_gain_error_db": -0.006245948617279137, "reference_mean_power": 0.0043546897567722925, "output_mean_power": 0.0043484314181835865}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.684242583835521e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006964729892166045, "output_power": 0.0006572748461815761, "reference_active": true, "delta_db": -0.2515726596527149}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017819877477929594, "output_power": 0.00016336599082079103, "reference_active": true, "delta_db": -0.3774306256442481}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2119939135052175e-06, "output_power": 4.510061327180574e-06, "reference_active": true, "delta_db": -0.6282145248423423}], "active_band_count": 3, "active_band_mae_db": 0.4190726033797684}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.00018825872292702335, "log_rms_mae_db": 1.622780381298416, "rms_envelope_nrmse": 0.0865979884039179, "envelope_correlation": 0.9939835358751608, "rms_gain_error_db": -0.058408809681677734, "reference_mean_power": 0.003284629583795886, "output_mean_power": 0.003240749921424592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 2.043578979045215e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008901806972697356, "output_power": 0.000860680655878081, "reference_active": true, "delta_db": -0.1463613063838717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002718123373610413, "output_power": 0.00022863894104907962, "reference_active": true, "delta_db": -0.751189651510093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0768894288562865e-05, "output_power": 4.338782118631113e-05, "reference_active": true, "delta_db": -1.4631349264213545}], "active_band_count": 3, "active_band_mae_db": 0.786895294771773}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00015953911997549416, "log_rms_mae_db": 1.030285911717904, "rms_envelope_nrmse": 0.053930212661356475, "envelope_correlation": 0.9974598487162218, "rms_gain_error_db": -0.0584093693929533, "reference_mean_power": 0.0033118454961761057, "output_mean_power": 0.003267601832857749}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.8056141501282585e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008648498098004977, "output_power": 0.0008088644787533487, "reference_active": true, "delta_db": -0.2906493042875181}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013850439438479803, "output_power": 0.00011548187341011149, "reference_active": true, "delta_db": -0.7894973202684853}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.4525026177578845e-06, "output_power": 4.082746848438845e-06, "reference_active": true, "delta_db": -1.9877573716889076}], "active_band_count": 3, "active_band_mae_db": 1.0226346654149703}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 6.604508172479527e-05, "log_rms_mae_db": 1.4965961220906847, "rms_envelope_nrmse": 0.1803004598571424, "envelope_correlation": 0.9742627666698808, "rms_gain_error_db": -0.10305570536526906, "reference_mean_power": 0.00023215995390588967, "output_mean_power": 0.00022671577433151498}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.444500480390906e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.349295087979117e-05, "output_power": 6.74739646013779e-05, "reference_active": true, "delta_db": -0.3710945650516162}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7764968189144465e-05, "output_power": 2.3133242983995975e-05, "reference_active": true, "delta_db": -0.7926066067408801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.892866752425807e-06, "output_power": 5.498830795702895e-06, "reference_active": true, "delta_db": -0.30056264851724424}], "active_band_count": 3, "active_band_mae_db": 0.4880879401032468}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.621650617144381e-05, "log_rms_mae_db": 0.8568275649865142, "rms_envelope_nrmse": 0.11071135074323345, "envelope_correlation": 0.9894371151439172, "rms_gain_error_db": -0.10317330145557127, "reference_mean_power": 0.00023404438013250112, "output_mean_power": 0.00022854982188882776}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.2566052761381804e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.967443215044184e-05, "output_power": 6.324329989086201e-05, "reference_active": true, "delta_db": -0.4205891553646056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5348913604759878e-05, "output_power": 1.295046228363257e-05, "reference_active": true, "delta_db": -0.737923701604385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.927931708651043e-07, "output_power": 4.60359164090513e-07, "reference_active": true, "delta_db": -0.295678879285987}], "active_band_count": 3, "active_band_mae_db": 0.4847305787516592}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 5.916637942334853e-05, "log_rms_mae_db": 1.6528226125663859, "rms_envelope_nrmse": 0.16434627576092048, "envelope_correlation": 0.9811253685804717, "rms_gain_error_db": 0.35549071357819784, "reference_mean_power": 0.00012513898539343656, "output_mean_power": 0.0001358131120441038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 8.947237162820932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.53948723828488e-05, "output_power": 5.1227253770764455e-05, "reference_active": true, "delta_db": 0.5249427499651749}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2730550619395128e-05, "output_power": 1.296756081832395e-05, "reference_active": true, "delta_db": 0.08011105443430525}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.684612122113266e-06, "output_power": 1.6054609072227035e-06, "reference_active": true, "delta_db": -0.20900186444590155}], "active_band_count": 3, "active_band_mae_db": 0.2713518896151272}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.1723901891273096e-05, "log_rms_mae_db": 1.0214834032161206, "rms_envelope_nrmse": 0.0990304758382139, "envelope_correlation": 0.9935655171100543, "rms_gain_error_db": 0.3559340162143704, "reference_mean_power": 0.0001261310627116738, "output_mean_power": 0.00013690378537543302}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 8.199333001241026e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1138157243798736e-05, "output_power": 4.6738170846447224e-05, "reference_active": true, "delta_db": 0.5542687789701924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.547615689423209e-06, "output_power": 7.176619390540868e-06, "reference_active": true, "delta_db": 0.3983673350170252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.061187881770097e-07, "output_power": 1.1891817871001648e-07, "reference_active": true, "delta_db": 0.49455967236881226}], "active_band_count": 3, "active_band_mae_db": 0.48239859545200997}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8891050887195014, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c1b1ae4bacf97b400a35c8702e2e0623553874b598934209a6129c606d5d3a88", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.972617725442722, "standard_deviation": 0.0013614285353066439, "minimum": 0.9710321349627795, "maximum": 0.9742627666698808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5472698904142206, "standard_deviation": 0.03285238135495033, "minimum": 1.4965961220906847, "maximum": 1.5860594513486874, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.36574746996217866, "standard_deviation": 0.14276191992069795, "minimum": 0.09380195090060241, "maximum": 0.5436447933569395, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18683962086608807, "standard_deviation": 0.004758629139661044, "minimum": 0.1803004598571424, "maximum": 0.19145021464189896, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.989042020945284, "standard_deviation": 0.000754212431094274, "minimum": 0.9878778609550596, "maximum": 0.990015137092239, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.892942434325243, "standard_deviation": 0.031100967006087752, "minimum": 0.8471661947758339, "maximum": 0.9278017689770944, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4628613245159635, "standard_deviation": 0.174496272445834, "minimum": 0.2501851387036276, "maximum": 0.7215051456262933, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11296266061222957, "standard_deviation": 0.0041009754969334065, "minimum": 0.10755399221570353, "maximum": 0.11898254707560696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9794778120771691, "standard_deviation": 0.0028802472092590375, "minimum": 0.9768289270359212, "maximum": 0.9853443858409718, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6512060191421976, "standard_deviation": 0.04263243791830941, "minimum": 1.5956215873867279, "maximum": 1.7420113591880144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5327866970745316, "standard_deviation": 0.2636407207475649, "minimum": 0.25952589090571726, "maximum": 0.8982540136493631, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1621714492970001, "standard_deviation": 0.011722014919839258, "minimum": 0.1363154991574689, "maximum": 0.1726312707869313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9934570933102587, "standard_deviation": 0.0009268426405601393, "minimum": 0.992069919527324, "maximum": 0.9949793291585198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9991219901540063, "standard_deviation": 0.04690447483777898, "minimum": 0.9288406510792367, "maximum": 1.0815209112126123, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.539422527066876, "standard_deviation": 0.3215073031520471, "minimum": 0.21731357094916168, "maximum": 1.0209392547641583, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08983205512242451, "standard_deviation": 0.007481718838202184, "minimum": 0.07701184138696099, "maximum": 0.0990304758382139, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922004680221441, "standard_deviation": 0.0016755597690817983, "minimum": 0.9891592217364622, "maximum": 0.9943017114981402, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.654936481875147, "standard_deviation": 0.05279121873791071, "minimum": 1.6033167257882073, "maximum": 1.7658015217593375, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7936540231446026, "standard_deviation": 0.4285686805393013, "minimum": 0.3451298407470871, "maximum": 1.4737910617124748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0985027422806882, "standard_deviation": 0.010545266993836533, "minimum": 0.08427859543790477, "maximum": 0.1157028678405587, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970446962996573, "standard_deviation": 0.000581011805291863, "minimum": 0.996025347845341, "maximum": 0.9979735411167447, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0033748676365473, "standard_deviation": 0.03907025690007771, "minimum": 0.9552822987028081, "maximum": 1.0578428209935824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0815423691762565, "standard_deviation": 0.5262332867366675, "minimum": 0.6415739876097882, "maximum": 1.9449706529791593, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05929688797409612, "standard_deviation": 0.006213292358811409, "minimum": 0.04842123551262075, "maximum": 0.06752306827970127, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14221841613258726, "standard_deviation": 0.04791990760313387, "minimum": 0.07588021944948732, "maximum": 0.20254103253234643, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.67617881188448, "standard_deviation": 0.009262306159902181, "minimum": 5.663306440812789, "maximum": 5.686769572362799, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5203188981140701, "standard_deviation": 0.08076670607908677, "minimum": 0.47619972821122597, "maximum": 0.7178587561593586, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07912492719990491, "standard_deviation": 0.06311451196157522, "minimum": -0.18072583329238212, "maximum": -0.004457020665103788, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924060204269106, "standard_deviation": 0.0011220988212106329, "minimum": 0.9903781063848457, "maximum": 0.9940986776335241, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8865115546365891, "standard_deviation": 0.026886201456263067, "minimum": 0.8604923350456894, "maximum": 0.94900245667572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5558453987030005, "standard_deviation": 0.39889792278129954, "minimum": 0.11074626830926015, "maximum": 1.222294674733136, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08716723721443673, "standard_deviation": 0.006513351815097137, "minimum": 0.07688606302471769, "maximum": 0.09793732062132444, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973524873979095, "standard_deviation": 0.0005421589729116705, "minimum": 0.996809072346727, "maximum": 0.9980748848599552, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4823688185165651, "standard_deviation": 0.021454312316796457, "minimum": 0.45787752823631045, "maximum": 0.5252937834175252, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6955871863203678, "standard_deviation": 0.5378587177546349, "minimum": 0.08618853340196936, "maximum": 1.5405369120529986, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04942745041537787, "standard_deviation": 0.005276163943213336, "minimum": 0.04167408276123258, "maximum": 0.055278635519580864, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0014", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.360330104827881, "predicted_mos": 2.360330104827881, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "0633f1d730ba5c47c365761171ba0396856fbd42862bab7119bda28d56ea9987", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7666521999053657, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8937470019782506, "全帯域_10ms_rms_envelope_nrmse": 0.08304034381186555, "全帯域_10ms_envelope_correlation": 0.9931296487159772, "全帯域_10ms_modulation_mae_db": 0.3168649603316586, "全帯域_40ms_log_rms_mae_db": 0.5074447927934356, "全帯域_40ms_rms_envelope_nrmse": 0.04692484605779504, "全帯域_40ms_envelope_correlation": 0.9975635774614285, "全帯域_40ms_modulation_mae_db": 0.27850421321763275, "300–1000 Hz_10ms_log_rms_mae_db": 1.6338768343172374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09822182489697186, "300–1000 Hz_10ms_envelope_correlation": 0.992237552733628, "300–1000 Hz_10ms_modulation_mae_db": 0.5531679099745097, "300–1000 Hz_40ms_log_rms_mae_db": 0.9909865569011546, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05756883967869456, "300–1000 Hz_40ms_envelope_correlation": 0.9970772017865495, "300–1000 Hz_40ms_modulation_mae_db": 0.7915648889379749, "1000–3000 Hz_10ms_log_rms_mae_db": 1.53711456055667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17716171393038332, "1000–3000 Hz_10ms_envelope_correlation": 0.9751895581134682, "1000–3000 Hz_10ms_modulation_mae_db": 0.7960158441034384, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8495685982000195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11061540000041462, "1000–3000 Hz_40ms_envelope_correlation": 0.9896743185803487, "1000–3000 Hz_40ms_modulation_mae_db": 0.8605178227648533, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6522152180663052, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1599268725891985, "3000–6000 Hz_10ms_envelope_correlation": 0.982129113983801, "3000–6000 Hz_10ms_modulation_mae_db": 0.23035909323494497, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0382562252017435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10186016037719815, "3000–6000 Hz_40ms_envelope_correlation": 0.9931005190826075, "3000–6000 Hz_40ms_modulation_mae_db": 0.4724196138179357, "mr_stft_db": 5.645018408134786, "band_mae_db": 0.14306252014865223, "rms_gain_error_db": 0.02204042177877331, "output_peak": 0.5447145187583732}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.888770397379353, "全帯域_10ms_rms_envelope_nrmse": 0.08819075436056256, "全帯域_10ms_envelope_correlation": 0.9921916635031351, "全帯域_10ms_modulation_mae_db": 0.36459347202154396, "全帯域_40ms_log_rms_mae_db": 0.5035031034976338, "全帯域_40ms_rms_envelope_nrmse": 0.05134733740285042, "全帯域_40ms_envelope_correlation": 0.9970752062779902, "全帯域_40ms_modulation_mae_db": 0.5352012257008428, "300–1000 Hz_10ms_log_rms_mae_db": 1.575845114002317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10669657994065433, "300–1000 Hz_10ms_envelope_correlation": 0.9907816511454292, "300–1000 Hz_10ms_modulation_mae_db": 0.5726241569007019, "300–1000 Hz_40ms_log_rms_mae_db": 0.9443006353407414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06542909805715412, "300–1000 Hz_40ms_envelope_correlation": 0.9962444409370481, "300–1000 Hz_40ms_modulation_mae_db": 0.7555715003318683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6095212098801848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.197047698666503, "1000–3000 Hz_10ms_envelope_correlation": 0.9694865658303738, "1000–3000 Hz_10ms_modulation_mae_db": 0.7004831510489608, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9182515786441512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11891401458562267, "1000–3000 Hz_40ms_envelope_correlation": 0.9878090524477293, "1000–3000 Hz_40ms_modulation_mae_db": 0.6894183745414976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.674787553761308, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15766514092279793, "3000–6000 Hz_10ms_envelope_correlation": 0.9805413952604797, "3000–6000 Hz_10ms_modulation_mae_db": 0.279291611952668, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0277569207414254, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08499433823378716, "3000–6000 Hz_40ms_envelope_correlation": 0.9943083113539498, "3000–6000 Hz_40ms_modulation_mae_db": 0.21205183401220584, "mr_stft_db": 5.657920603498286, "band_mae_db": 0.07938460834138646, "rms_gain_error_db": -0.06003425257740802, "output_peak": 0.504565365166045}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8971623955681562, "全帯域_10ms_rms_envelope_nrmse": 0.09125393041341945, "全帯域_10ms_envelope_correlation": 0.9916654499295906, "全帯域_10ms_modulation_mae_db": 0.08509218464371689, "全帯域_40ms_log_rms_mae_db": 0.5041543961611576, "全帯域_40ms_rms_envelope_nrmse": 0.058791183350731815, "全帯域_40ms_envelope_correlation": 0.9961883396530794, "全帯域_40ms_modulation_mae_db": 0.10463582671452587, "300–1000 Hz_10ms_log_rms_mae_db": 1.6820045764792075, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.106177706894718, "300–1000 Hz_10ms_envelope_correlation": 0.9908895810760349, "300–1000 Hz_10ms_modulation_mae_db": 0.3389516948243729, "300–1000 Hz_40ms_log_rms_mae_db": 1.0158913970642196, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06808617711584651, "300–1000 Hz_40ms_envelope_correlation": 0.9959306645803876, "300–1000 Hz_40ms_modulation_mae_db": 0.7548919926304887, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5883501944055056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19004094749001632, "1000–3000 Hz_10ms_envelope_correlation": 0.9715429275995393, "1000–3000 Hz_10ms_modulation_mae_db": 0.4974006599449035, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9152421691766653, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11456609359707248, "1000–3000 Hz_40ms_envelope_correlation": 0.9886393701322774, "1000–3000 Hz_40ms_modulation_mae_db": 0.7352051861778528, "3000–6000 Hz_10ms_log_rms_mae_db": 1.700950031231266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16260587707628266, "3000–6000 Hz_10ms_envelope_correlation": 0.9791322892879599, "3000–6000 Hz_10ms_modulation_mae_db": 0.8095686802708966, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0386036422461213, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09388579979013974, "3000–6000 Hz_40ms_envelope_correlation": 0.9926099842453117, "3000–6000 Hz_40ms_modulation_mae_db": 0.9686195921912438, "mr_stft_db": 5.652205700537192, "band_mae_db": 0.12224093788505837, "rms_gain_error_db": -0.08608759690413686, "output_peak": 0.5222335565273518}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.87500323594767, "全帯域_10ms_rms_envelope_nrmse": 0.09748127907810797, "全帯域_10ms_envelope_correlation": 0.990767414961592, "全帯域_10ms_modulation_mae_db": 1.2439142678922652, "全帯域_40ms_log_rms_mae_db": 0.4667048066217808, "全帯域_40ms_rms_envelope_nrmse": 0.05991159013493497, "全帯域_40ms_envelope_correlation": 0.9964485782779681, "全帯域_40ms_modulation_mae_db": 1.5469495117745644, "300–1000 Hz_10ms_log_rms_mae_db": 1.6959120677052668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10875440156142951, "300–1000 Hz_10ms_envelope_correlation": 0.9908333022206385, "300–1000 Hz_10ms_modulation_mae_db": 1.4822069097866812, "300–1000 Hz_40ms_log_rms_mae_db": 0.99179676004524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06819623000046583, "300–1000 Hz_40ms_envelope_correlation": 0.9964442135995951, "300–1000 Hz_40ms_modulation_mae_db": 1.8707358106144933, "1000–3000 Hz_10ms_log_rms_mae_db": 1.556365723623791, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19214039017036405, "1000–3000 Hz_10ms_envelope_correlation": 0.9707849307856536, "1000–3000 Hz_10ms_modulation_mae_db": 0.46707016047825195, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8914457987407636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12093478796281193, "1000–3000 Hz_40ms_envelope_correlation": 0.987627736248398, "1000–3000 Hz_40ms_modulation_mae_db": 0.6137177052981388, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7945962629170475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1639432024526691, "3000–6000 Hz_10ms_envelope_correlation": 0.9791639111592334, "3000–6000 Hz_10ms_modulation_mae_db": 0.6632109002346503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.166939434234069, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09079923235752352, "3000–6000 Hz_40ms_envelope_correlation": 0.9938858362811764, "3000–6000 Hz_40ms_modulation_mae_db": 0.7484819473449411, "mr_stft_db": 5.6682570367788045, "band_mae_db": 0.20474522046114715, "rms_gain_error_db": -0.19348905983791423, "output_peak": 0.4789869690775039}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8817746178072159, "全帯域_10ms_rms_envelope_nrmse": 0.09642697820568466, "全帯域_10ms_envelope_correlation": 0.9909516043256399, "全帯域_10ms_modulation_mae_db": 1.213492755691991, "全帯域_40ms_log_rms_mae_db": 0.49732895663551163, "全帯域_40ms_rms_envelope_nrmse": 0.05470476577640034, "全帯域_40ms_envelope_correlation": 0.9971003302474851, "全帯域_40ms_modulation_mae_db": 1.5415144789302995, "300–1000 Hz_10ms_log_rms_mae_db": 1.6643746914088877, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11623047420648455, "300–1000 Hz_10ms_envelope_correlation": 0.9894499429597426, "300–1000 Hz_10ms_modulation_mae_db": 1.6730096447743492, "300–1000 Hz_40ms_log_rms_mae_db": 0.9754717551041642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.068734898218317, "300–1000 Hz_40ms_envelope_correlation": 0.9964021786346817, "300–1000 Hz_40ms_modulation_mae_db": 2.260911966100743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5868596814845393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18105088835903227, "1000–3000 Hz_10ms_envelope_correlation": 0.9744845493081559, "1000–3000 Hz_10ms_modulation_mae_db": 0.2585615084599118, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9031360344608447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11185320195551285, "1000–3000 Hz_40ms_envelope_correlation": 0.9893275895641165, "1000–3000 Hz_40ms_modulation_mae_db": 0.21977204613580423, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6498227327475319, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13092527860434292, "3000–6000 Hz_10ms_envelope_correlation": 0.9865778892729794, "3000–6000 Hz_10ms_modulation_mae_db": 0.4329253188171183, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0107915275043322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07694370581374495, "3000–6000 Hz_40ms_envelope_correlation": 0.9951704440222174, "3000–6000 Hz_40ms_modulation_mae_db": 0.22313145999575282, "mr_stft_db": 5.646988617370009, "band_mae_db": 0.20116389092119602, "rms_gain_error_db": -0.18901334447262047, "output_peak": 0.49154710817208935}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8857053542553754, "全帯域_10ms_rms_envelope_nrmse": 0.0881927612345051, "全帯域_10ms_envelope_correlation": 0.9921978499306969, "全帯域_10ms_modulation_mae_db": 0.13858060120387708, "全帯域_40ms_log_rms_mae_db": 0.5009489085727303, "全帯域_40ms_rms_envelope_nrmse": 0.05105829481532557, "全帯域_40ms_envelope_correlation": 0.9970714045883554, "全帯域_40ms_modulation_mae_db": 0.16615638181520037, "300–1000 Hz_10ms_log_rms_mae_db": 1.620702139597885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09350158870056417, "300–1000 Hz_10ms_envelope_correlation": 0.9929418043501103, "300–1000 Hz_10ms_modulation_mae_db": 0.322759354966025, "300–1000 Hz_40ms_log_rms_mae_db": 0.9678513138814249, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053737950331913996, "300–1000 Hz_40ms_envelope_correlation": 0.9974483582489475, "300–1000 Hz_40ms_modulation_mae_db": 0.5177566633816454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5894492880411597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1882581943033482, "1000–3000 Hz_10ms_envelope_correlation": 0.972342969013751, "1000–3000 Hz_10ms_modulation_mae_db": 0.15887549839474835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9260081068799476, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11524939421390502, "1000–3000 Hz_40ms_envelope_correlation": 0.9886464093544618, "1000–3000 Hz_40ms_modulation_mae_db": 0.15059023407251185, "3000–6000 Hz_10ms_log_rms_mae_db": 1.691481520273939, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15986300652544405, "3000–6000 Hz_10ms_envelope_correlation": 0.9800252958888377, "3000–6000 Hz_10ms_modulation_mae_db": 0.6891116344406175, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0635398692446734, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0893429320090986, "3000–6000 Hz_40ms_envelope_correlation": 0.993811744290708, "3000–6000 Hz_40ms_modulation_mae_db": 1.0300468249402144, "mr_stft_db": 5.655204570402177, "band_mae_db": 0.18103621117768667, "rms_gain_error_db": 0.0022332400475700317, "output_peak": 0.5248553433663705}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9749886742723223, "全帯域_10ms_rms_envelope_nrmse": 0.0911672853047862, "全帯域_10ms_envelope_correlation": 0.9916453168141406, "全帯域_10ms_modulation_mae_db": 0.27157653254536657, "全帯域_40ms_log_rms_mae_db": 0.5386511435863222, "全帯域_40ms_rms_envelope_nrmse": 0.05251959124374859, "全帯域_40ms_envelope_correlation": 0.9969219092817605, "全帯域_40ms_modulation_mae_db": 0.28043591341248736, "300–1000 Hz_10ms_log_rms_mae_db": 1.7532441790596003, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10916510991242646, "300–1000 Hz_10ms_envelope_correlation": 0.9903787434497704, "300–1000 Hz_10ms_modulation_mae_db": 0.6874629062430401, "300–1000 Hz_40ms_log_rms_mae_db": 1.0603184508817671, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059003711424426286, "300–1000 Hz_40ms_envelope_correlation": 0.9969814275820286, "300–1000 Hz_40ms_modulation_mae_db": 1.12435530534408, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5332606396090778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18314935627895046, "1000–3000 Hz_10ms_envelope_correlation": 0.9736439924162749, "1000–3000 Hz_10ms_modulation_mae_db": 0.450582039989533, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8762524367648545, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11169312446118634, "1000–3000 Hz_40ms_envelope_correlation": 0.9892282579072986, "1000–3000 Hz_40ms_modulation_mae_db": 0.5935782429447272, "3000–6000 Hz_10ms_log_rms_mae_db": 1.654284260184303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14884222269295905, "3000–6000 Hz_10ms_envelope_correlation": 0.9825426429186691, "3000–6000 Hz_10ms_modulation_mae_db": 0.18940754956462758, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0485522082576422, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09278100306626054, "3000–6000 Hz_40ms_envelope_correlation": 0.9928099434299503, "3000–6000 Hz_40ms_modulation_mae_db": 0.3838140894783367, "mr_stft_db": 5.669141605287712, "band_mae_db": 0.08277314930427564, "rms_gain_error_db": -0.05187179244332817, "output_peak": 0.7079217474575495}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.88791852323162, "全帯域_10ms_rms_envelope_nrmse": 0.09698768175560588, "全帯域_10ms_envelope_correlation": 0.9905426216030503, "全帯域_10ms_modulation_mae_db": 0.3197731395710773, "全帯域_40ms_log_rms_mae_db": 0.4896454175890781, "全帯域_40ms_rms_envelope_nrmse": 0.05490733827034503, "全帯域_40ms_envelope_correlation": 0.9966317234248679, "全帯域_40ms_modulation_mae_db": 0.38377953714258917, "300–1000 Hz_10ms_log_rms_mae_db": 1.6047484904599048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11248783908735352, "300–1000 Hz_10ms_envelope_correlation": 0.9897341639647628, "300–1000 Hz_10ms_modulation_mae_db": 0.4968896135896921, "300–1000 Hz_40ms_log_rms_mae_db": 0.9835787331195086, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06550866716390341, "300–1000 Hz_40ms_envelope_correlation": 0.9961837400861783, "300–1000 Hz_40ms_modulation_mae_db": 0.7800200660376352, "1000–3000 Hz_10ms_log_rms_mae_db": 1.555471270722617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.186328767697952, "1000–3000 Hz_10ms_envelope_correlation": 0.9726730760995687, "1000–3000 Hz_10ms_modulation_mae_db": 0.23197621211314526, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9017609331749702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10707244883670197, "1000–3000 Hz_40ms_envelope_correlation": 0.9901166651883352, "1000–3000 Hz_40ms_modulation_mae_db": 0.3542554802412578, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6907574615799403, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16612733265189356, "3000–6000 Hz_10ms_envelope_correlation": 0.9785338815585221, "3000–6000 Hz_10ms_modulation_mae_db": 0.7345005366725453, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0453492905834505, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09006883133159972, "3000–6000 Hz_40ms_envelope_correlation": 0.9931841545217253, "3000–6000 Hz_40ms_modulation_mae_db": 0.4334748003510001, "mr_stft_db": 5.6585988016673, "band_mae_db": 0.10544726895761347, "rms_gain_error_db": -0.05183626937115816, "output_peak": 0.4959338743665871}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 58480, "output_samples": 58480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8072029051695, "spectral_convergence": 0.28568552765138183, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267125480811954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.457960003287359, "spectral_convergence": 0.26422101646659757, "spectral_convergence_status": "ok", "amplitude_floor": 1.0684354905475814e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.669892315947502, "spectral_convergence": 0.27515693269369945, "spectral_convergence_status": "ok", "amplitude_floor": 7.412945744843815e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002955479965644196, "output_power": 0.0003032229872343309, "delta_db": 0.11134102192570278, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00146613378566028, "output_power": 0.0015164118153649204, "delta_db": 0.14643557875510485, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0021080578166581473, "output_power": 0.002068554312783593, "delta_db": -0.08215589493973627, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001787602167832836, "output_power": 0.00017168630155289992, "delta_db": -0.1753522745374982, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014763682412872298, "output_power": 0.00015886823726528045, "delta_db": 0.3184238246957449, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001169621027158164, "output_power": 0.00011629967864918297, "delta_db": -0.024666526038126378, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14306252014865223, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.02204042177877331, "pitch": null, "output_rms_dbfs": -23.630065569827714, "output_peak": 0.5447145187583732, "reference_rms_dbfs": -23.65210599160649, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.645018408134786}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 58480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.0002136685864137289, "log_rms_mae_db": 0.8937470019782506, "rms_envelope_nrmse": 0.08304034381186555, "envelope_correlation": 0.9931296487159772, "rms_gain_error_db": 0.022042148204951265, "reference_mean_power": 0.004318997149502425, "output_mean_power": 0.004340973475955225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.9590597233654063e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007496792113024279, "output_power": 0.0007226447298318386, "reference_active": true, "delta_db": -0.15950628208782888}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031885669741703006, "output_power": 0.0003046575690456193, "reference_active": true, "delta_db": -0.1978357114979706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.908796664605564e-05, "output_power": 6.026665984298121e-05, "reference_active": true, "delta_db": -0.5932528874091763}], "active_band_count": 3, "active_band_mae_db": 0.3168649603316586}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00017810705079567896, "log_rms_mae_db": 0.5074447927934356, "rms_envelope_nrmse": 0.04692484605779504, "envelope_correlation": 0.9975635774614285, "rms_gain_error_db": 0.022048301907768467, "reference_mean_power": 0.0043546897567722925, "output_mean_power": 0.004376853899443295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.684242583835521e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006964729892166045, "output_power": 0.0006664410876828288, "reference_active": true, "delta_db": -0.19142513658316407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017819877477929594, "output_power": 0.0001669432000517338, "reference_active": true, "delta_db": -0.2833597966138696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2119939135052175e-06, "output_power": 4.796573781527924e-06, "reference_active": true, "delta_db": -0.3607277064558645}], "active_band_count": 3, "active_band_mae_db": 0.27850421321763275}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.00018825872292702335, "log_rms_mae_db": 1.6338768343172374, "rms_envelope_nrmse": 0.09822182489697186, "envelope_correlation": 0.992237552733628, "rms_gain_error_db": 0.0012204096004190166, "reference_mean_power": 0.003284629583795886, "output_mean_power": 0.003285552726254407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 2.043578979045215e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008901806972697356, "output_power": 0.0008816094483398611, "reference_active": true, "delta_db": -0.04201936850332683}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002718123373610413, "output_power": 0.00024371643590935405, "reference_active": true, "delta_db": -0.47384346739955296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0768894288562865e-05, "output_power": 4.6700039174529394e-05, "reference_active": true, "delta_db": -1.1436408940206493}], "active_band_count": 3, "active_band_mae_db": 0.5531679099745097}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00015953911997549416, "log_rms_mae_db": 0.9909865569011546, "rms_envelope_nrmse": 0.05756883967869456, "envelope_correlation": 0.9970772017865495, "rms_gain_error_db": 0.0012197298610660632, "reference_mean_power": 0.0033118454961761057, "output_mean_power": 0.003312775769142854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.8056141501282585e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008648498098004977, "output_power": 0.0008376342068684219, "reference_active": true, "delta_db": -0.138862897172088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013850439438479803, "output_power": 0.00012228356156920376, "reference_active": true, "delta_db": -0.5409547337547772}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.4525026177578845e-06, "output_power": 4.3675761504372905e-06, "reference_active": true, "delta_db": -1.6948770358870595}], "active_band_count": 3, "active_band_mae_db": 0.7915648889379749}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 6.604508172479527e-05, "log_rms_mae_db": 1.53711456055667, "rms_envelope_nrmse": 0.17716171393038332, "envelope_correlation": 0.9751895581134682, "rms_gain_error_db": -0.2040033336688707, "reference_mean_power": 0.00023215995390588967, "output_mean_power": 0.0002215067553221591}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.444500480390906e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.349295087979117e-05, "output_power": 6.609353666020074e-05, "reference_active": true, "delta_db": -0.46086693910129367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7764968189144465e-05, "output_power": 2.2362692780452278e-05, "reference_active": true, "delta_db": -0.9397308274669471}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.892866752425807e-06, "output_power": 4.694416779713947e-06, "reference_active": true, "delta_db": -0.9874497657420742}], "active_band_count": 3, "active_band_mae_db": 0.7960158441034384}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.621650617144381e-05, "log_rms_mae_db": 0.8495685982000195, "rms_envelope_nrmse": 0.11061540000041462, "envelope_correlation": 0.9896743185803487, "rms_gain_error_db": -0.2041499329680081, "reference_mean_power": 0.00023404438013250112, "output_mean_power": 0.0002232971725878898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.2566052761381804e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.967443215044184e-05, "output_power": 6.164029189778167e-05, "reference_active": true, "delta_db": -0.5320875122740211}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5348913604759878e-05, "output_power": 1.2446782234455016e-05, "reference_active": true, "delta_db": -0.9102055024607851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.927931708651043e-07, "output_power": 3.7908677647888886e-07, "reference_active": true, "delta_db": -1.1392604535597541}], "active_band_count": 3, "active_band_mae_db": 0.8605178227648533}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 5.916637942334853e-05, "log_rms_mae_db": 1.6522152180663052, "rms_envelope_nrmse": 0.1599268725891985, "envelope_correlation": 0.982129113983801, "rms_gain_error_db": 0.3127661997479946, "reference_mean_power": 0.00012513898539343656, "output_mean_power": 0.0001344835762174933}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 8.947237162820932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.53948723828488e-05, "output_power": 5.167094574616634e-05, "reference_active": true, "delta_db": 0.5623961110038388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2730550619395128e-05, "output_power": 1.272348252175434e-05, "reference_active": true, "delta_db": -0.0024119053077538556}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.684612122113266e-06, "output_power": 1.6363379215315142e-06, "reference_active": true, "delta_db": -0.1262692633932423}], "active_band_count": 3, "active_band_mae_db": 0.23035909323494497}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.1723901891273096e-05, "log_rms_mae_db": 1.0382562252017435, "rms_envelope_nrmse": 0.10186016037719815, "envelope_correlation": 0.9931005190826075, "rms_gain_error_db": 0.313236246256167, "reference_mean_power": 0.0001261310627116738, "output_mean_power": 0.00013556440725912437}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 8.199333001241026e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1138157243798736e-05, "output_power": 4.722953508521377e-05, "reference_active": true, "delta_db": 0.5996883665620941}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.547615689423209e-06, "output_power": 6.844766498989309e-06, "reference_active": true, "delta_db": 0.1927545659880412}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.061187881770097e-07, "output_power": 1.2253887046981174e-07, "reference_active": true, "delta_db": 0.6248159089036717}], "active_band_count": 3, "active_band_mae_db": 0.4724196138179357}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8901639683633182, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a9debc2232003ecb4bf874c14a3227b5b50a456479cc25f1cf21b2f7e6bec85f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9725185711458482, "standard_deviation": 0.001907540222467927, "minimum": 0.9694865658303738, "maximum": 0.9751895581134682, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5695490710404432, "standard_deviation": 0.027743114540374467, "minimum": 1.5332606396090778, "maximum": 1.6095212098801848, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.44512063431661164, "standard_deviation": 0.22480258098713812, "minimum": 0.15887549839474835, "maximum": 0.7960158441034384, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1868972446120687, "standard_deviation": 0.006387349290976066, "minimum": 0.17716171393038332, "maximum": 0.197047698666503, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9888836749278707, "standard_deviation": 0.0008709071812512192, "minimum": 0.987627736248398, "maximum": 0.9901166651883352, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8977082070052771, "standard_deviation": 0.02505203142813931, "minimum": 0.8495685982000195, "maximum": 0.9260081068799476, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5271318865220804, "standard_deviation": 0.25602169170461603, "minimum": 0.15059023407251185, "maximum": 0.8605178227648533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1138623082016535, "standard_deviation": 0.0045252544983096195, "minimum": 0.10707244883670197, "maximum": 0.12093478796281193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9810808024163102, "standard_deviation": 0.0026408043735244927, "minimum": 0.9785338815585221, "maximum": 0.9865778892729794, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6886118800952052, "standard_deviation": 0.047202557703680606, "minimum": 1.6498227327475319, "maximum": 1.7945962629170475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5035469156485086, "standard_deviation": 0.24943041604377916, "minimum": 0.18940754956462758, "maximum": 0.8095686802708966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15623736668944846, "standard_deviation": 0.011476282818542948, "minimum": 0.13092527860434292, "maximum": 0.16612733265189356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9936101171534557, "standard_deviation": 0.0008552733730079722, "minimum": 0.9926099842453117, "maximum": 0.9951704440222174, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0549736397516822, "standard_deviation": 0.04778499451416381, "minimum": 1.0107915275043322, "maximum": 1.166939434234069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5590050202664538, "standard_deviation": 0.3190835712455558, "minimum": 0.21205183401220584, "maximum": 1.0300468249402144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09008450037241904, "standard_deviation": 0.00717175427619553, "minimum": 0.07694370581374495, "maximum": 0.10186016037719815, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9909058427375146, "standard_deviation": 0.0011777612109454025, "minimum": 0.9894499429597426, "maximum": 0.9929418043501103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6538385116287884, "standard_deviation": 0.0567517112340432, "minimum": 1.575845114002317, "maximum": 1.7532441790596003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7658840238824215, "standard_deviation": 0.5176501080443633, "minimum": 0.322759354966025, "maximum": 1.6730096447743492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1064044406500753, "standard_deviation": 0.007370576887813571, "minimum": 0.09350158870056417, "maximum": 0.11623047420648455, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965890281819271, "standard_deviation": 0.0005214080359699305, "minimum": 0.9959306645803876, "maximum": 0.9974483582489475, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9912744502922776, "standard_deviation": 0.03471836818147462, "minimum": 0.9443006353407414, "maximum": 1.0603184508817671, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1069760241723663, "standard_deviation": 0.6229206956446521, "minimum": 0.5177566633816454, "maximum": 2.260911966100743, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0632831964988402, "standard_deviation": 0.005714202563661721, "minimum": 0.053737950331913996, "maximum": 0.068734898218317, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13998172589962699, "standard_deviation": 0.050828459080222335, "minimum": 0.07938460834138646, "maximum": 0.20474522046114715, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.656666917959534, "standard_deviation": 0.00883702264806905, "minimum": 5.645018408134786, "maximum": 5.669141605287712, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5338448103614838, "standard_deviation": 0.0734161967937796, "minimum": 0.4789869690775039, "maximum": 0.7079217474575495, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07600733172252783, "standard_deviation": 0.07912989313348277, "minimum": -0.19348905983791423, "maximum": 0.02204042177877331, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9916364462229778, "standard_deviation": 0.0008680594116834285, "minimum": 0.9905426216030503, "maximum": 0.9931296487159772, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8981337750549955, "standard_deviation": 0.03179397204767186, "minimum": 0.87500323594767, "maximum": 0.9749886742723223, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4942359892376871, "standard_deviation": 0.4631444404727221, "minimum": 0.08509218464371689, "maximum": 1.2439142678922652, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09159262677056718, "standard_deviation": 0.005123096078917349, "minimum": 0.08304034381186555, "maximum": 0.09748127907810797, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968751336516168, "standard_deviation": 0.0004339670336011131, "minimum": 0.9961883396530794, "maximum": 0.9975635774614285, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5010476906822063, "standard_deviation": 0.019978229275367174, "minimum": 0.4667048066217808, "maximum": 0.5386511435863222, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6046471360885177, "standard_deviation": 0.5943816703304401, "minimum": 0.10463582671452587, "maximum": 1.5469495117745644, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05377061838151647, "standard_deviation": 0.004255098246043279, "minimum": 0.04692484605779504, "maximum": 0.05991159013493497, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0014", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.848719358444214, "predicted_mos": 2.848719358444214, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "f9f067936ced1fccf6007399364bd983d2c7c59d8cd97c8b6c63cfb2fae8d6a7", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7358301002532244, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8933313578716789, "全帯域_10ms_rms_envelope_nrmse": 0.08297667319200447, "全帯域_10ms_envelope_correlation": 0.9930923589425912, "全帯域_10ms_modulation_mae_db": 0.8430542993681857, "全帯域_40ms_log_rms_mae_db": 0.4990059974799265, "全帯域_40ms_rms_envelope_nrmse": 0.048762247508452615, "全帯域_40ms_envelope_correlation": 0.9973428947208763, "全帯域_40ms_modulation_mae_db": 0.7727669801253692, "300–1000 Hz_10ms_log_rms_mae_db": 2.0052305072172625, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10126347813471258, "300–1000 Hz_10ms_envelope_correlation": 0.9918364824716531, "300–1000 Hz_10ms_modulation_mae_db": 1.4529829033210975, "300–1000 Hz_40ms_log_rms_mae_db": 1.3863234686873278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06938863805094041, "300–1000 Hz_40ms_envelope_correlation": 0.9958663058638463, "300–1000 Hz_40ms_modulation_mae_db": 1.5744528311678987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.840812828483306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2141388759299602, "1000–3000 Hz_10ms_envelope_correlation": 0.9634852691393464, "1000–3000 Hz_10ms_modulation_mae_db": 1.0018690585324839, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0906491947282526, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1356898775871416, "1000–3000 Hz_40ms_envelope_correlation": 0.9841766811373716, "1000–3000 Hz_40ms_modulation_mae_db": 1.3039380190060015, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1198102274372657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2083504327567644, "3000–6000 Hz_10ms_envelope_correlation": 0.9658284927821806, "3000–6000 Hz_10ms_modulation_mae_db": 0.6723833771142349, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2933360411353667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12299518344427333, "3000–6000 Hz_40ms_envelope_correlation": 0.9871606044700978, "3000–6000 Hz_40ms_modulation_mae_db": 0.42468599115311534, "mr_stft_db": 5.978649310504409, "band_mae_db": 0.2098955455973939, "rms_gain_error_db": -0.040865781311761204, "output_peak": 0.5574198493035282}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8452555177582781, "全帯域_10ms_rms_envelope_nrmse": 0.07248475947904136, "全帯域_10ms_envelope_correlation": 0.9947270447930824, "全帯域_10ms_modulation_mae_db": 0.3526757469363278, "全帯域_40ms_log_rms_mae_db": 0.4448347434722566, "全帯域_40ms_rms_envelope_nrmse": 0.0389493156928813, "全帯域_40ms_envelope_correlation": 0.9983044248242969, "全帯域_40ms_modulation_mae_db": 0.493337078466263, "300–1000 Hz_10ms_log_rms_mae_db": 1.928127493816443, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0940706518552739, "300–1000 Hz_10ms_envelope_correlation": 0.9928568007601154, "300–1000 Hz_10ms_modulation_mae_db": 0.77268766190373, "300–1000 Hz_40ms_log_rms_mae_db": 1.206713269549227, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058843745832380744, "300–1000 Hz_40ms_envelope_correlation": 0.9969511972269861, "300–1000 Hz_40ms_modulation_mae_db": 1.1272665019087298, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7968811502204756, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1995610862698549, "1000–3000 Hz_10ms_envelope_correlation": 0.9687811241188318, "1000–3000 Hz_10ms_modulation_mae_db": 0.5885347040311456, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0481421333904577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1187946751563503, "1000–3000 Hz_40ms_envelope_correlation": 0.9878737221896666, "1000–3000 Hz_40ms_modulation_mae_db": 0.6880775929460236, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1285356448707202, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21210502123237637, "3000–6000 Hz_10ms_envelope_correlation": 0.9657252170747372, "3000–6000 Hz_10ms_modulation_mae_db": 1.1031522608993396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3139301069053437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11371918216873413, "3000–6000 Hz_40ms_envelope_correlation": 0.9897821119765069, "3000–6000 Hz_40ms_modulation_mae_db": 1.0051771663927525, "mr_stft_db": 5.95171742661848, "band_mae_db": 0.14705694304428749, "rms_gain_error_db": -0.026899521459762862, "output_peak": 0.6140029057446208}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8851357264141403, "全帯域_10ms_rms_envelope_nrmse": 0.08658004115261721, "全帯域_10ms_envelope_correlation": 0.9925304101850576, "全帯域_10ms_modulation_mae_db": 0.7004784344339489, "全帯域_40ms_log_rms_mae_db": 0.44483344073028336, "全帯域_40ms_rms_envelope_nrmse": 0.04937432620862916, "全帯域_40ms_envelope_correlation": 0.9973744801330019, "全帯域_40ms_modulation_mae_db": 0.7245894651099835, "300–1000 Hz_10ms_log_rms_mae_db": 1.8914036809346035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09827068078660012, "300–1000 Hz_10ms_envelope_correlation": 0.9924627364672043, "300–1000 Hz_10ms_modulation_mae_db": 0.9584437920954576, "300–1000 Hz_40ms_log_rms_mae_db": 1.251823463756065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06162965492764927, "300–1000 Hz_40ms_envelope_correlation": 0.9969805428258726, "300–1000 Hz_40ms_modulation_mae_db": 1.2027002675752776, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7292993881881071, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20292560120288777, "1000–3000 Hz_10ms_envelope_correlation": 0.9678517223838432, "1000–3000 Hz_10ms_modulation_mae_db": 0.3335460680236055, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0213646092545399, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11739947691690338, "1000–3000 Hz_40ms_envelope_correlation": 0.9881941125458866, "1000–3000 Hz_40ms_modulation_mae_db": 0.5694611593761504, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0529239897790585, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19371027164393545, "3000–6000 Hz_10ms_envelope_correlation": 0.9713657622417807, "3000–6000 Hz_10ms_modulation_mae_db": 0.38445742073413447, "3000–6000 Hz_40ms_log_rms_mae_db": 1.276379751579918, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11030904887791233, "3000–6000 Hz_40ms_envelope_correlation": 0.9901231960722375, "3000–6000 Hz_40ms_modulation_mae_db": 0.26715001730377974, "mr_stft_db": 5.971813338702819, "band_mae_db": 0.13068151025008554, "rms_gain_error_db": -0.0950098581644506, "output_peak": 0.47098763393070386}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8864028208769416, "全帯域_10ms_rms_envelope_nrmse": 0.08671277857917498, "全帯域_10ms_envelope_correlation": 0.992565107773546, "全帯域_10ms_modulation_mae_db": 0.8007360314977455, "全帯域_40ms_log_rms_mae_db": 0.47562026168857846, "全帯域_40ms_rms_envelope_nrmse": 0.04631637433655043, "全帯域_40ms_envelope_correlation": 0.9977934852551267, "全帯域_40ms_modulation_mae_db": 1.025460022797022, "300–1000 Hz_10ms_log_rms_mae_db": 2.1002897796754847, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11873050420835896, "300–1000 Hz_10ms_envelope_correlation": 0.988908589385204, "300–1000 Hz_10ms_modulation_mae_db": 1.2282095833637132, "300–1000 Hz_40ms_log_rms_mae_db": 1.4653240539186172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07268859645061478, "300–1000 Hz_40ms_envelope_correlation": 0.9957995898510368, "300–1000 Hz_40ms_modulation_mae_db": 1.801219254543798, "1000–3000 Hz_10ms_log_rms_mae_db": 1.817858844776516, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19256150488643656, "1000–3000 Hz_10ms_envelope_correlation": 0.9710960344989497, "1000–3000 Hz_10ms_modulation_mae_db": 0.1650724509133846, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0042409217359323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11247878384579513, "1000–3000 Hz_40ms_envelope_correlation": 0.9891961826307695, "1000–3000 Hz_40ms_modulation_mae_db": 0.12118769465888168, "3000–6000 Hz_10ms_log_rms_mae_db": 2.031922482676358, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21228406649618098, "3000–6000 Hz_10ms_envelope_correlation": 0.9644812784199602, "3000–6000 Hz_10ms_modulation_mae_db": 0.7750605921525091, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2368950692874947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1311884848423935, "3000–6000 Hz_40ms_envelope_correlation": 0.9855638735228952, "3000–6000 Hz_40ms_modulation_mae_db": 0.7758637619158565, "mr_stft_db": 5.987934473745045, "band_mae_db": 0.22553971060381506, "rms_gain_error_db": -0.1261338224101762, "output_peak": 0.49023481732860286}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8569591224063372, "全帯域_10ms_rms_envelope_nrmse": 0.08769265672331798, "全帯域_10ms_envelope_correlation": 0.9923911988757482, "全帯域_10ms_modulation_mae_db": 0.8684961971377737, "全帯域_40ms_log_rms_mae_db": 0.4590391362781404, "全帯域_40ms_rms_envelope_nrmse": 0.04948537217818411, "全帯域_40ms_envelope_correlation": 0.9974582267971835, "全帯域_40ms_modulation_mae_db": 0.9848850548432329, "300–1000 Hz_10ms_log_rms_mae_db": 2.0510587875655273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10503349625164206, "300–1000 Hz_10ms_envelope_correlation": 0.9914683139728138, "300–1000 Hz_10ms_modulation_mae_db": 1.9003335648567914, "300–1000 Hz_40ms_log_rms_mae_db": 1.31709862331459, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06481132824873072, "300–1000 Hz_40ms_envelope_correlation": 0.9967825702852726, "300–1000 Hz_40ms_modulation_mae_db": 2.208004633809121, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7552109133914733, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19125409250852524, "1000–3000 Hz_10ms_envelope_correlation": 0.9721143978810398, "1000–3000 Hz_10ms_modulation_mae_db": 0.11898460490990669, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9464823480434527, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11683554151683159, "1000–3000 Hz_40ms_envelope_correlation": 0.9887685602160662, "1000–3000 Hz_40ms_modulation_mae_db": 0.1560412774316978, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0439287987674266, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20800221817268255, "3000–6000 Hz_10ms_envelope_correlation": 0.9678526973216144, "3000–6000 Hz_10ms_modulation_mae_db": 0.950073716851327, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2597745351542178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11099757240933589, "3000–6000 Hz_40ms_envelope_correlation": 0.9906964043503428, "3000–6000 Hz_40ms_modulation_mae_db": 1.0443790307755485, "mr_stft_db": 5.961641724378051, "band_mae_db": 0.24181142453067336, "rms_gain_error_db": -0.12154700545865912, "output_peak": 0.5293664963033374}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8721895971614463, "全帯域_10ms_rms_envelope_nrmse": 0.08721991720168741, "全帯域_10ms_envelope_correlation": 0.9923776294832808, "全帯域_10ms_modulation_mae_db": 0.5713110060621652, "全帯域_40ms_log_rms_mae_db": 0.4700035282646674, "全帯域_40ms_rms_envelope_nrmse": 0.04384296434692109, "全帯域_40ms_envelope_correlation": 0.9978797310083364, "全帯域_40ms_modulation_mae_db": 0.8263601028703548, "300–1000 Hz_10ms_log_rms_mae_db": 2.121114274670937, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09713987205776511, "300–1000 Hz_10ms_envelope_correlation": 0.9924790205858945, "300–1000 Hz_10ms_modulation_mae_db": 1.1034889778902326, "300–1000 Hz_40ms_log_rms_mae_db": 1.3705604497941943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05475800432325692, "300–1000 Hz_40ms_envelope_correlation": 0.9974475818060422, "300–1000 Hz_40ms_modulation_mae_db": 1.4794288846641, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7248898141595708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1957669780880321, "1000–3000 Hz_10ms_envelope_correlation": 0.9700320269872903, "1000–3000 Hz_10ms_modulation_mae_db": 0.7064342678305119, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0542969121034713, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12168016128327616, "1000–3000 Hz_40ms_envelope_correlation": 0.9872908426037773, "1000–3000 Hz_40ms_modulation_mae_db": 0.8175501914708289, "3000–6000 Hz_10ms_log_rms_mae_db": 2.067445346682182, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21663995505119366, "3000–6000 Hz_10ms_envelope_correlation": 0.9632970310308614, "3000–6000 Hz_10ms_modulation_mae_db": 0.7875572795895246, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2726868600031194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1150420867540817, "3000–6000 Hz_40ms_envelope_correlation": 0.9889212007476098, "3000–6000 Hz_40ms_modulation_mae_db": 0.5972768844011695, "mr_stft_db": 5.973382957807627, "band_mae_db": 0.16984678391300576, "rms_gain_error_db": -0.023780126975905123, "output_peak": 0.5067823599771106}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.90320209208927, "全帯域_10ms_rms_envelope_nrmse": 0.09020574849178481, "全帯域_10ms_envelope_correlation": 0.9918238413253339, "全帯域_10ms_modulation_mae_db": 0.9001852263252818, "全帯域_40ms_log_rms_mae_db": 0.5027533358782433, "全帯域_40ms_rms_envelope_nrmse": 0.05029306326773625, "全帯域_40ms_envelope_correlation": 0.9971620677877436, "全帯域_40ms_modulation_mae_db": 1.038787859665953, "300–1000 Hz_10ms_log_rms_mae_db": 1.9271821934616016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10030505138575355, "300–1000 Hz_10ms_envelope_correlation": 0.9918978995631212, "300–1000 Hz_10ms_modulation_mae_db": 0.9404720638392087, "300–1000 Hz_40ms_log_rms_mae_db": 1.2959523454761264, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06589739621197827, "300–1000 Hz_40ms_envelope_correlation": 0.9961623280877158, "300–1000 Hz_40ms_modulation_mae_db": 1.2279259800695765, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8457445433238515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1959810663854233, "1000–3000 Hz_10ms_envelope_correlation": 0.9698404898786639, "1000–3000 Hz_10ms_modulation_mae_db": 0.7594929522736339, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0605514025068363, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11323831993673868, "1000–3000 Hz_40ms_envelope_correlation": 0.9889497261141129, "1000–3000 Hz_40ms_modulation_mae_db": 0.9854599747613374, "3000–6000 Hz_10ms_log_rms_mae_db": 2.059961806282063, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2002290852283672, "3000–6000 Hz_10ms_envelope_correlation": 0.9689018267558317, "3000–6000 Hz_10ms_modulation_mae_db": 0.40755225706986903, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2872948521781284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11849584986239799, "3000–6000 Hz_40ms_envelope_correlation": 0.9882331074881107, "3000–6000 Hz_40ms_modulation_mae_db": 0.17163541253219042, "mr_stft_db": 5.970914117764386, "band_mae_db": 0.0651776568530608, "rms_gain_error_db": -0.015165108683345006, "output_peak": 0.5522538129001243}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8776335817647014, "全帯域_10ms_rms_envelope_nrmse": 0.08576389976465737, "全帯域_10ms_envelope_correlation": 0.9926654854611858, "全帯域_10ms_modulation_mae_db": 0.7581019891605307, "全帯域_40ms_log_rms_mae_db": 0.4318457010729955, "全帯域_40ms_rms_envelope_nrmse": 0.0396984919154403, "全帯域_40ms_envelope_correlation": 0.9983384349991735, "全帯域_40ms_modulation_mae_db": 0.9517166005829841, "300–1000 Hz_10ms_log_rms_mae_db": 2.074229665737621, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09799608902777486, "300–1000 Hz_10ms_envelope_correlation": 0.9924389456865509, "300–1000 Hz_10ms_modulation_mae_db": 0.9140454379522264, "300–1000 Hz_40ms_log_rms_mae_db": 1.2854162800221887, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05173804142689889, "300–1000 Hz_40ms_envelope_correlation": 0.9978985722737176, "300–1000 Hz_40ms_modulation_mae_db": 1.1857125771834542, "1000–3000 Hz_10ms_log_rms_mae_db": 1.787373032535935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18520861170838004, "1000–3000 Hz_10ms_envelope_correlation": 0.9730417881512032, "1000–3000 Hz_10ms_modulation_mae_db": 0.3762014178062363, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0241720474732203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10618662972567693, "1000–3000 Hz_40ms_envelope_correlation": 0.9902653045549693, "1000–3000 Hz_40ms_modulation_mae_db": 0.4882199089562162, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0587597975720917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20786744744587507, "3000–6000 Hz_10ms_envelope_correlation": 0.9661072113753606, "3000–6000 Hz_10ms_modulation_mae_db": 0.7946927575224807, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2894719520258966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11826286139550257, "3000–6000 Hz_40ms_envelope_correlation": 0.9884264268180364, "3000–6000 Hz_40ms_modulation_mae_db": 0.7093095297422521, "mr_stft_db": 5.966958470467792, "band_mae_db": 0.10836892352673687, "rms_gain_error_db": -0.08397599015376207, "output_peak": 0.51511835564038}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 58480, "output_samples": 58480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.170605292925121, "spectral_convergence": 0.29692941849279125, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267125480811954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.862695438520571, "spectral_convergence": 0.2726748584256728, "spectral_convergence_status": "ok", "amplitude_floor": 1.0684354905475814e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.902647200067534, "spectral_convergence": 0.2777314848621424, "spectral_convergence_status": "ok", "amplitude_floor": 7.412945744843815e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002955479965644196, "output_power": 0.00032936804366418503, "delta_db": 0.47053440508486194, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00146613378566028, "output_power": 0.0014846023664319655, "delta_db": 0.054365467086327605, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0021080578166581473, "output_power": 0.0020225721678636924, "delta_db": -0.1797849116132716, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001787602167832836, "output_power": 0.0001715727741048888, "delta_db": -0.1782249936881295, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014763682412872298, "output_power": 0.00014194934445319106, "delta_db": -0.17061303133173816, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001169621027158164, "output_power": 0.0001226394555423453, "delta_db": 0.2058504647800347, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2098955455973939, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.040865781311761204, "pitch": null, "output_rms_dbfs": -23.692971772918252, "output_peak": 0.5574198493035282, "reference_rms_dbfs": -23.65210599160649, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.978649310504409}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 58480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.0002136685864137289, "log_rms_mae_db": 0.8933313578716789, "rms_envelope_nrmse": 0.08297667319200447, "envelope_correlation": 0.9930923589425912, "rms_gain_error_db": -0.0408655422175216, "reference_mean_power": 0.004318997149502425, "output_mean_power": 0.004278547552709004}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.9590597233654063e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007496792113024279, "output_power": 0.0007226565110457218, "reference_active": true, "delta_db": -0.1594354800150845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031885669741703006, "output_power": 0.0002738639460732016, "reference_active": true, "delta_db": -0.6606068198240764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.908796664605564e-05, "output_power": 4.661120721478946e-05, "reference_active": true, "delta_db": -1.7091205982653959}], "active_band_count": 3, "active_band_mae_db": 0.8430542993681857}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00017810705079567896, "log_rms_mae_db": 0.4990059974799265, "rms_envelope_nrmse": 0.048762247508452615, "envelope_correlation": 0.9973428947208763, "rms_gain_error_db": -0.0408608834672236, "reference_mean_power": 0.0043546897567722925, "output_mean_power": 0.004313910508228132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.684242583835521e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006964729892166045, "output_power": 0.0006574721695807544, "reference_active": true, "delta_db": -0.250269040583349}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017819877477929594, "output_power": 0.0001542290050575536, "reference_active": true, "delta_db": -0.6273865675210957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2119939135052175e-06, "output_power": 3.7405836104932343e-06, "reference_active": true, "delta_db": -1.4406453322716626}], "active_band_count": 3, "active_band_mae_db": 0.7727669801253692}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.00018825872292702335, "log_rms_mae_db": 2.0052305072172625, "rms_envelope_nrmse": 0.10126347813471258, "envelope_correlation": 0.9918364824716531, "rms_gain_error_db": -0.13029498287006736, "reference_mean_power": 0.003284629583795886, "output_mean_power": 0.0031875492367040605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 2.043578979045215e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008901806972697356, "output_power": 0.0008155031549351013, "reference_active": true, "delta_db": -0.38052527248111434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002718123373610413, "output_power": 0.00020704865233185413, "reference_active": true, "delta_db": -1.1819675710030801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0768894288562865e-05, "output_power": 3.191800567650527e-05, "reference_active": true, "delta_db": -2.796455866479098}], "active_band_count": 3, "active_band_mae_db": 1.4529829033210975}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00015953911997549416, "log_rms_mae_db": 1.3863234686873278, "rms_envelope_nrmse": 0.06938863805094041, "envelope_correlation": 0.9958663058638463, "rms_gain_error_db": -0.13029566501622408, "reference_mean_power": 0.0033118454961761057, "output_mean_power": 0.003213960252127557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.8056141501282585e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008648498098004977, "output_power": 0.0007644219839930565, "reference_active": true, "delta_db": -0.5360752581454188}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013850439438479803, "output_power": 0.00010608409436354283, "reference_active": true, "delta_db": -1.1581327947941342}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.4525026177578845e-06, "output_power": 3.2122781626832417e-06, "reference_active": true, "delta_db": -3.0291504405641434}], "active_band_count": 3, "active_band_mae_db": 1.5744528311678987}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 6.604508172479527e-05, "log_rms_mae_db": 1.840812828483306, "rms_envelope_nrmse": 0.2141388759299602, "envelope_correlation": 0.9634852691393464, "rms_gain_error_db": -0.17936160885066418, "reference_mean_power": 0.00023215995390588967, "output_mean_power": 0.0002227671496393275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.444500480390906e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.349295087979117e-05, "output_power": 6.045611708873638e-05, "reference_active": true, "delta_db": -0.8480543513467196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7764968189144465e-05, "output_power": 2.1759870660887692e-05, "reference_active": true, "delta_db": -1.0584087061084213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.892866752425807e-06, "output_power": 4.575222394154037e-06, "reference_active": true, "delta_db": -1.0991441181423107}], "active_band_count": 3, "active_band_mae_db": 1.0018690585324839}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.621650617144381e-05, "log_rms_mae_db": 1.0906491947282526, "rms_envelope_nrmse": 0.1356898775871416, "envelope_correlation": 0.9841766811373716, "rms_gain_error_db": -0.17926319501955618, "reference_mean_power": 0.00023404438013250112, "output_mean_power": 0.0002245804242003155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.2566052761381804e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.967443215044184e-05, "output_power": 5.5913715333825545e-05, "reference_active": true, "delta_db": -0.9555508682524201}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5348913604759878e-05, "output_power": 1.2708364944619832e-05, "reference_active": true, "delta_db": -0.8198796364437042}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.927931708651043e-07, "output_power": 3.013188668261863e-07, "reference_active": true, "delta_db": -2.13638355232188}], "active_band_count": 3, "active_band_mae_db": 1.3039380190060015}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 5.916637942334853e-05, "log_rms_mae_db": 2.1198102274372657, "rms_envelope_nrmse": 0.2083504327567644, "envelope_correlation": 0.9658284927821806, "rms_gain_error_db": 0.008170276698896797, "reference_mean_power": 0.00012513898539343656, "output_mean_power": 0.00012537462791472066}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 8.947237162820932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.53948723828488e-05, "output_power": 4.619737298744964e-05, "reference_active": true, "delta_db": 0.07610480594665367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2730550619395128e-05, "output_power": 9.181508478561511e-06, "reference_active": true, "delta_db": -1.419331484841737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.684612122113266e-06, "output_power": 1.8996399988318647e-06, "reference_active": true, "delta_db": 0.5217138405543139}], "active_band_count": 3, "active_band_mae_db": 0.6723833771142349}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.1723901891273096e-05, "log_rms_mae_db": 1.2933360411353667, "rms_envelope_nrmse": 0.12299518344427333, "envelope_correlation": 0.9871606044700978, "rms_gain_error_db": 0.008179937534356859, "reference_mean_power": 0.0001261310627116738, "output_mean_power": 0.00012636885446650915}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 8.199333001241026e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1138157243798736e-05, "output_power": 4.151198678508195e-05, "reference_active": true, "delta_db": 0.03928685569710325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.547615689423209e-06, "output_power": 5.279182390892182e-06, "reference_active": true, "delta_db": -0.9351651395309024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.061187881770097e-07, "output_power": 1.1369801424772827e-07, "reference_active": true, "delta_db": 0.2996059782313405}], "active_band_count": 3, "active_band_mae_db": 0.42468599115311534}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8655872836680765, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2307bd73a561c86c49351dc7dc6b310fce0a0bf3346b05562d2a56642f474388", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9695303566298961, "standard_deviation": 0.0029695807434148703, "minimum": 0.9634852691393464, "maximum": 0.9730417881512032, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7872588143849044, "standard_deviation": 0.0471984146230605, "minimum": 1.7248898141595708, "maximum": 1.8457445433238515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5062669405401135, "standard_deviation": 0.30931505978096624, "minimum": 0.11898460490990669, "maximum": 1.0018690585324839, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1971747271224375, "standard_deviation": 0.008697124668246256, "minimum": 0.18520861170838004, "maximum": 0.2141388759299602, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9880893914990776, "standard_deviation": 0.0018191635634233262, "minimum": 0.9841766811373716, "maximum": 0.9902653045549693, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0312374461545204, "standard_deviation": 0.043533571740651276, "minimum": 0.9464823480434527, "maximum": 1.0906491947282526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6412419773258922, "standard_deviation": 0.40095861026406443, "minimum": 0.12118769465888168, "maximum": 1.3039380190060015, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11778793324608922, "standard_deviation": 0.00862866336176876, "minimum": 0.10618662972567693, "maximum": 0.1356898775871416, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9666949396252908, "standard_deviation": 0.0025776493390544204, "minimum": 0.9632970310308614, "maximum": 0.9713657622417807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.070411011758396, "standard_deviation": 0.034971330613354036, "minimum": 2.031922482676358, "maximum": 2.1285356448707202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7343662077416775, "standard_deviation": 0.2460380014250546, "minimum": 0.38445742073413447, "maximum": 1.1031522608993396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20739856225342196, "standard_deviation": 0.007286810595075598, "minimum": 0.19371027164393545, "maximum": 0.21663995505119366, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9886133656807297, "standard_deviation": 0.0016762555307149785, "minimum": 0.9855638735228952, "maximum": 0.9906964043503428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2787211460336858, "standard_deviation": 0.023276732240929698, "minimum": 1.2368950692874947, "maximum": 1.3139301069053437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.624434724277083, "standard_deviation": 0.321773921751779, "minimum": 0.17163541253219042, "maximum": 1.0443790307755485, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11762628371932893, "standard_deviation": 0.006900776517501085, "minimum": 0.11030904887791233, "maximum": 0.1311884848423935, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9917935986115697, "standard_deviation": 0.0012482116610212323, "minimum": 0.988908589385204, "maximum": 0.9928568007601154, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.012329547884935, "standard_deviation": 0.08776284383471991, "minimum": 1.8914036809346035, "maximum": 2.121114274670937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1588329981528072, "standard_deviation": 0.3666625210037276, "minimum": 0.77268766190373, "maximum": 1.9003335648567914, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10160122796348514, "standard_deviation": 0.007631603119593062, "minimum": 0.0940706518552739, "maximum": 0.11873050420835896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967360860275613, "standard_deviation": 0.000749145613467692, "minimum": 0.9957995898510368, "maximum": 0.9978985722737176, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.322401494314792, "standard_deviation": 0.08219555912805933, "minimum": 1.206713269549227, "maximum": 1.4653240539186172, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.4758388663652444, "standard_deviation": 0.37671856880757404, "minimum": 1.1272665019087298, "maximum": 2.208004633809121, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.062469425684056244, "standard_deviation": 0.007151641542686714, "minimum": 0.05173804142689889, "maximum": 0.07268859645061478, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16229731228988237, "standard_deviation": 0.06119874847674755, "minimum": 0.0651776568530608, "maximum": 0.24181142453067336, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.970376477498576, "standard_deviation": 0.010851296255963708, "minimum": 5.95171742661848, "maximum": 5.987934473745045, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.529520778891051, "standard_deviation": 0.04492276522939924, "minimum": 0.47098763393070386, "maximum": 0.6140029057446208, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06667215182722777, "standard_deviation": 0.04533986014483506, "minimum": -0.1261338224101762, "maximum": -0.015165108683345006, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9927716346049782, "standard_deviation": 0.0008649431673329541, "minimum": 0.9918238413253339, "maximum": 0.9947270447930824, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8775137270428492, "standard_deviation": 0.01904372776218492, "minimum": 0.8452555177582781, "maximum": 0.90320209208927, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.724379866365245, "standard_deviation": 0.1831835203120565, "minimum": 0.3526757469363278, "maximum": 0.9001852263252818, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0849545593230357, "standard_deviation": 0.005424817457834739, "minimum": 0.07248475947904136, "maximum": 0.09020574849178481, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977067181907173, "standard_deviation": 0.0004464019621627492, "minimum": 0.9971620677877436, "maximum": 0.9983384349991735, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46599201810813645, "standard_deviation": 0.025800724812119178, "minimum": 0.4318457010729955, "maximum": 0.5027533358782433, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8522378955576453, "standard_deviation": 0.18684878567319155, "minimum": 0.493337078466263, "maximum": 1.038787859665953, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045840269431849406, "standard_deviation": 0.004528705872134353, "minimum": 0.0389493156928813, "maximum": 0.05029306326773625, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0014", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.798811435699463, "predicted_mos": 2.798811435699463, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "494af445708c28544b76d61f2fca23c1e267b3f093cbed67350cd84c8ab47fda", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8194551998749375, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8896030106726398, "全帯域_10ms_rms_envelope_nrmse": 0.0935593443658246, "全帯域_10ms_envelope_correlation": 0.9912784378303295, "全帯域_10ms_modulation_mae_db": 0.8766730168030593, "全帯域_40ms_log_rms_mae_db": 0.5111635122479694, "全帯域_40ms_rms_envelope_nrmse": 0.04989061440184852, "全帯域_40ms_envelope_correlation": 0.9973693550984503, "全帯域_40ms_modulation_mae_db": 1.2371667751019724, "300–1000 Hz_10ms_log_rms_mae_db": 1.4572144570978816, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10642396026267179, "300–1000 Hz_10ms_envelope_correlation": 0.9909659803758089, "300–1000 Hz_10ms_modulation_mae_db": 1.1290650676801495, "300–1000 Hz_40ms_log_rms_mae_db": 0.8927544406995462, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06541390282848644, "300–1000 Hz_40ms_envelope_correlation": 0.9964417218141959, "300–1000 Hz_40ms_modulation_mae_db": 1.3700275014939187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.430978239546954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18654447831710358, "1000–3000 Hz_10ms_envelope_correlation": 0.9724569303488646, "1000–3000 Hz_10ms_modulation_mae_db": 1.1623581799422753, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8136875572598892, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11352668107074734, "1000–3000 Hz_40ms_envelope_correlation": 0.9891688228639614, "1000–3000 Hz_40ms_modulation_mae_db": 1.2740567983553748, "3000–6000 Hz_10ms_log_rms_mae_db": 1.456230391183502, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13227553904996409, "3000–6000 Hz_10ms_envelope_correlation": 0.9861791287576117, "3000–6000 Hz_10ms_modulation_mae_db": 0.403191092951571, "3000–6000 Hz_40ms_log_rms_mae_db": 0.859302725302152, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07125672355952095, "3000–6000 Hz_40ms_envelope_correlation": 0.9956371491633135, "3000–6000 Hz_40ms_modulation_mae_db": 0.2456052667563052, "mr_stft_db": 5.566866730404542, "band_mae_db": 0.12971421577774786, "rms_gain_error_db": -0.11428475123753652, "output_peak": 0.5221102346044515}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9529201068911738, "全帯域_10ms_rms_envelope_nrmse": 0.09269983534045655, "全帯域_10ms_envelope_correlation": 0.9914375092156706, "全帯域_10ms_modulation_mae_db": 1.061709448440107, "全帯域_40ms_log_rms_mae_db": 0.5417551570042368, "全帯域_40ms_rms_envelope_nrmse": 0.05101336796154936, "全帯域_40ms_envelope_correlation": 0.9972303054995574, "全帯域_40ms_modulation_mae_db": 1.310856892268748, "300–1000 Hz_10ms_log_rms_mae_db": 1.520051029669072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09715104123104835, "300–1000 Hz_10ms_envelope_correlation": 0.9925071810448847, "300–1000 Hz_10ms_modulation_mae_db": 1.1472613509548528, "300–1000 Hz_40ms_log_rms_mae_db": 0.9898090971094775, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05805566601051909, "300–1000 Hz_40ms_envelope_correlation": 0.9972318973562243, "300–1000 Hz_40ms_modulation_mae_db": 1.5042097621714146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4971068110172618, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18904318056969688, "1000–3000 Hz_10ms_envelope_correlation": 0.971722708340945, "1000–3000 Hz_10ms_modulation_mae_db": 0.4567698943459167, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8453041910320459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10941776648556881, "1000–3000 Hz_40ms_envelope_correlation": 0.9896571741603505, "1000–3000 Hz_40ms_modulation_mae_db": 0.5113082851130419, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5112793636782387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18642425981156785, "3000–6000 Hz_10ms_envelope_correlation": 0.9755025034465938, "3000–6000 Hz_10ms_modulation_mae_db": 0.6549564722612286, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9491560094800037, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13224358969406652, "3000–6000 Hz_40ms_envelope_correlation": 0.9874725503797417, "3000–6000 Hz_40ms_modulation_mae_db": 0.9733748381039365, "mr_stft_db": 5.582271345866722, "band_mae_db": 0.09274933864430318, "rms_gain_error_db": -0.11217052233888426, "output_peak": 0.4937716115955716}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8880682126037746, "全帯域_10ms_rms_envelope_nrmse": 0.09269034566932983, "全帯域_10ms_envelope_correlation": 0.9913928549593779, "全帯域_10ms_modulation_mae_db": 0.5178663970686818, "全帯域_40ms_log_rms_mae_db": 0.5020300101451135, "全帯域_40ms_rms_envelope_nrmse": 0.05124440105464012, "全帯域_40ms_envelope_correlation": 0.9971412344294285, "全帯域_40ms_modulation_mae_db": 0.7954635516516787, "300–1000 Hz_10ms_log_rms_mae_db": 1.5467087113093851, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1024193358753653, "300–1000 Hz_10ms_envelope_correlation": 0.9915500486313384, "300–1000 Hz_10ms_modulation_mae_db": 0.8310926510022666, "300–1000 Hz_40ms_log_rms_mae_db": 0.9322183758157895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05915300467583989, "300–1000 Hz_40ms_envelope_correlation": 0.9969851132199623, "300–1000 Hz_40ms_modulation_mae_db": 1.136383324996838, "1000–3000 Hz_10ms_log_rms_mae_db": 1.460717486227528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1718656220644013, "1000–3000 Hz_10ms_envelope_correlation": 0.9766622015330784, "1000–3000 Hz_10ms_modulation_mae_db": 0.7345209609550123, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8254955589117022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10094095331564333, "1000–3000 Hz_40ms_envelope_correlation": 0.9912353246573439, "1000–3000 Hz_40ms_modulation_mae_db": 0.9327682488242584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4830569579093662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14948616646214227, "3000–6000 Hz_10ms_envelope_correlation": 0.9827353808265521, "3000–6000 Hz_10ms_modulation_mae_db": 0.14345104189199492, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8910619274830243, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09341638664583939, "3000–6000 Hz_40ms_envelope_correlation": 0.9927089062242485, "3000–6000 Hz_40ms_modulation_mae_db": 0.1823512598659018, "mr_stft_db": 5.543692179728356, "band_mae_db": 0.10217443448485551, "rms_gain_error_db": -0.08766371689859809, "output_peak": 0.5101318346585367}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9488618841029852, "全帯域_10ms_rms_envelope_nrmse": 0.09492913427574903, "全帯域_10ms_envelope_correlation": 0.990985656289719, "全帯域_10ms_modulation_mae_db": 0.8053530998108406, "全帯域_40ms_log_rms_mae_db": 0.515317402110273, "全帯域_40ms_rms_envelope_nrmse": 0.05346395642041984, "全帯域_40ms_envelope_correlation": 0.9968954873915333, "全帯域_40ms_modulation_mae_db": 1.019201093495213, "300–1000 Hz_10ms_log_rms_mae_db": 1.493883937303627, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10231144119269382, "300–1000 Hz_10ms_envelope_correlation": 0.9916231807454341, "300–1000 Hz_10ms_modulation_mae_db": 1.0054631459400407, "300–1000 Hz_40ms_log_rms_mae_db": 0.8971045272574103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06086854751968198, "300–1000 Hz_40ms_envelope_correlation": 0.9968700507105355, "300–1000 Hz_40ms_modulation_mae_db": 1.181941802722464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4850171235787861, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17027805364083462, "1000–3000 Hz_10ms_envelope_correlation": 0.9771342845925932, "1000–3000 Hz_10ms_modulation_mae_db": 0.34348044835969355, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8757682126915746, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10484148879324534, "1000–3000 Hz_40ms_envelope_correlation": 0.9904889131436794, "1000–3000 Hz_40ms_modulation_mae_db": 0.37931727397643916, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4955301269981063, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15906651495269108, "3000–6000 Hz_10ms_envelope_correlation": 0.979970953159627, "3000–6000 Hz_10ms_modulation_mae_db": 0.6904187449736762, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9560228821417736, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09296915778849336, "3000–6000 Hz_40ms_envelope_correlation": 0.9926399882371097, "3000–6000 Hz_40ms_modulation_mae_db": 0.19442754527253825, "mr_stft_db": 5.54924453172259, "band_mae_db": 0.1005813158666035, "rms_gain_error_db": -0.07828754380187633, "output_peak": 0.45760246132573934}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9650425696622763, "全帯域_10ms_rms_envelope_nrmse": 0.08887282845894749, "全帯域_10ms_envelope_correlation": 0.9921091690739369, "全帯域_10ms_modulation_mae_db": 0.05239321809863443, "全帯域_40ms_log_rms_mae_db": 0.5780888523902533, "全帯域_40ms_rms_envelope_nrmse": 0.04839319752961637, "全帯域_40ms_envelope_correlation": 0.9974765943356835, "全帯域_40ms_modulation_mae_db": 0.21170038338276695, "300–1000 Hz_10ms_log_rms_mae_db": 1.5029961265398157, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09127191055861356, "300–1000 Hz_10ms_envelope_correlation": 0.9933289958687969, "300–1000 Hz_10ms_modulation_mae_db": 0.09930028432834298, "300–1000 Hz_40ms_log_rms_mae_db": 0.9302577931674796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052342377926482234, "300–1000 Hz_40ms_envelope_correlation": 0.9976843731838184, "300–1000 Hz_40ms_modulation_mae_db": 0.4420058579741779, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4183568102571984, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1754157389275539, "1000–3000 Hz_10ms_envelope_correlation": 0.9756782399683079, "1000–3000 Hz_10ms_modulation_mae_db": 0.12822264941428047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8317771233312935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10683843454443707, "1000–3000 Hz_40ms_envelope_correlation": 0.9901378398298195, "1000–3000 Hz_40ms_modulation_mae_db": 0.19344638314312504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.485913931199387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16165065283615726, "3000–6000 Hz_10ms_envelope_correlation": 0.979310537094227, "3000–6000 Hz_10ms_modulation_mae_db": 0.8207471121524655, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9652447474885718, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09342549121766257, "3000–6000 Hz_40ms_envelope_correlation": 0.9925568300377722, "3000–6000 Hz_40ms_modulation_mae_db": 0.7115552769041159, "mr_stft_db": 5.547396306700636, "band_mae_db": 0.15092339591336443, "rms_gain_error_db": -0.09594298049805028, "output_peak": 0.6050469359782362}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9545590663906365, "全帯域_10ms_rms_envelope_nrmse": 0.10181163403302124, "全帯域_10ms_envelope_correlation": 0.9899721011917825, "全帯域_10ms_modulation_mae_db": 1.063593508425908, "全帯域_40ms_log_rms_mae_db": 0.5249048657364601, "全帯域_40ms_rms_envelope_nrmse": 0.06211799531790459, "全帯域_40ms_envelope_correlation": 0.9962591476908338, "全帯域_40ms_modulation_mae_db": 1.2098911014605511, "300–1000 Hz_10ms_log_rms_mae_db": 1.5966607276557074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.118812958874804, "300–1000 Hz_10ms_envelope_correlation": 0.9892759464410206, "300–1000 Hz_10ms_modulation_mae_db": 1.5236792682146358, "300–1000 Hz_40ms_log_rms_mae_db": 1.042227925749822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08003037341134202, "300–1000 Hz_40ms_envelope_correlation": 0.9952989907270999, "300–1000 Hz_40ms_modulation_mae_db": 1.9579095102407997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.472861012674553, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17115160101976867, "1000–3000 Hz_10ms_envelope_correlation": 0.9770663289185011, "1000–3000 Hz_10ms_modulation_mae_db": 0.10412197653761524, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8066318925603002, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09968435139077508, "1000–3000 Hz_40ms_envelope_correlation": 0.9914484915404509, "1000–3000 Hz_40ms_modulation_mae_db": 0.12453384453125003, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4826092880664767, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14925245230797812, "3000–6000 Hz_10ms_envelope_correlation": 0.9823700704606894, "3000–6000 Hz_10ms_modulation_mae_db": 0.4213283623966079, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8954659359511467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08427523131184475, "3000–6000 Hz_40ms_envelope_correlation": 0.9939342601457074, "3000–6000 Hz_40ms_modulation_mae_db": 0.35258637730765324, "mr_stft_db": 5.573609016312535, "band_mae_db": 0.22537305015554665, "rms_gain_error_db": -0.21183385197894658, "output_peak": 0.4531584460399002}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9030385478733621, "全帯域_10ms_rms_envelope_nrmse": 0.08705056823496565, "全帯域_10ms_envelope_correlation": 0.9924054321489538, "全帯域_10ms_modulation_mae_db": 0.3140220007553576, "全帯域_40ms_log_rms_mae_db": 0.4655617364835314, "全帯域_40ms_rms_envelope_nrmse": 0.04562695640528592, "全帯域_40ms_envelope_correlation": 0.99772037448722, "全帯域_40ms_modulation_mae_db": 0.5740659044170888, "300–1000 Hz_10ms_log_rms_mae_db": 1.477286628545792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09034888144312637, "300–1000 Hz_10ms_envelope_correlation": 0.9934558141815031, "300–1000 Hz_10ms_modulation_mae_db": 0.2875212088088775, "300–1000 Hz_40ms_log_rms_mae_db": 0.9014350617232832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051566388158296464, "300–1000 Hz_40ms_envelope_correlation": 0.9977449302996828, "300–1000 Hz_40ms_modulation_mae_db": 0.6148565801117715, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3993552220472112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16570663975749508, "1000–3000 Hz_10ms_envelope_correlation": 0.9783571830659784, "1000–3000 Hz_10ms_modulation_mae_db": 0.6500926232885301, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8094308765763563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1008662442813078, "1000–3000 Hz_40ms_envelope_correlation": 0.991448487656496, "1000–3000 Hz_40ms_modulation_mae_db": 0.7498609302147649, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4381860228969023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14208459609701532, "3000–6000 Hz_10ms_envelope_correlation": 0.9847121129128694, "3000–6000 Hz_10ms_modulation_mae_db": 0.22402998440955835, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9191133759532014, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0855978012272192, "3000–6000 Hz_40ms_envelope_correlation": 0.9941166436920812, "3000–6000 Hz_40ms_modulation_mae_db": 0.12308904292856872, "mr_stft_db": 5.567924840725773, "band_mae_db": 0.11683009295404838, "rms_gain_error_db": -0.0685902522221008, "output_peak": 0.4962143609026114}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9337244482691148, "全帯域_10ms_rms_envelope_nrmse": 0.10289706196728267, "全帯域_10ms_envelope_correlation": 0.9898100570039075, "全帯域_10ms_modulation_mae_db": 0.8818634586086973, "全帯域_40ms_log_rms_mae_db": 0.5318405707071937, "全帯域_40ms_rms_envelope_nrmse": 0.06196229251121417, "全帯域_40ms_envelope_correlation": 0.9963975217449347, "全帯域_40ms_modulation_mae_db": 1.0411970483155308, "300–1000 Hz_10ms_log_rms_mae_db": 1.489646049449065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11356591587649727, "300–1000 Hz_10ms_envelope_correlation": 0.9900586743204138, "300–1000 Hz_10ms_modulation_mae_db": 0.8374855088237187, "300–1000 Hz_40ms_log_rms_mae_db": 0.8988070444501144, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07440860558373572, "300–1000 Hz_40ms_envelope_correlation": 0.9958152026128883, "300–1000 Hz_40ms_modulation_mae_db": 1.0781636525375926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4496672143265485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18274322908418242, "1000–3000 Hz_10ms_envelope_correlation": 0.973547229176861, "1000–3000 Hz_10ms_modulation_mae_db": 1.0141769821243913, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9153575076363692, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1109902490849609, "1000–3000 Hz_40ms_envelope_correlation": 0.9895548874109096, "1000–3000 Hz_40ms_modulation_mae_db": 1.1494402622046989, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4331441133043568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13185019565781358, "3000–6000 Hz_10ms_envelope_correlation": 0.9864170330595264, "3000–6000 Hz_10ms_modulation_mae_db": 0.5323981655806325, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8665064080249129, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07315241900878891, "3000–6000 Hz_40ms_envelope_correlation": 0.9957254478715251, "3000–6000 Hz_40ms_modulation_mae_db": 0.3298579615407933, "mr_stft_db": 5.553223023127555, "band_mae_db": 0.1826959494974432, "rms_gain_error_db": -0.23230706751699537, "output_peak": 0.4538406933305263}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 58480, "output_samples": 58480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.661322953473425, "spectral_convergence": 0.2844351310497025, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267125480811954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.361976215121314, "spectral_convergence": 0.2596833100703939, "spectral_convergence_status": "ok", "amplitude_floor": 1.0684354905475814e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.677301022618889, "spectral_convergence": 0.25927738237169773, "spectral_convergence_status": "ok", "amplitude_floor": 7.412945744843815e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002955479965644196, "output_power": 0.000292888136043982, "delta_db": -0.03926239570676365, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00146613378566028, "output_power": 0.0014722203315856261, "delta_db": 0.01799209342660749, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0021080578166581473, "output_power": 0.0020096010279454187, "delta_db": -0.20772673677370812, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001787602167832836, "output_power": 0.0001741834775886392, "delta_db": -0.112639156295518, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014763682412872298, "output_power": 0.00013848003790228815, "delta_db": -0.2780752061738314, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001169621027158164, "output_power": 0.00011370673693981376, "delta_db": -0.12258970629005839, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12971421577774786, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11428475123753652, "pitch": null, "output_rms_dbfs": -23.766390742844024, "output_peak": 0.5221102346044515, "reference_rms_dbfs": -23.65210599160649, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.566866730404542}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 58480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.0002136685864137289, "log_rms_mae_db": 0.8896030106726398, "rms_envelope_nrmse": 0.0935593443658246, "envelope_correlation": 0.9912784378303295, "rms_gain_error_db": -0.1142893735122375, "reference_mean_power": 0.004318997149502425, "output_mean_power": 0.004206820485275939}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.9590597233654063e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007496792113024279, "output_power": 0.000702254611439752, "reference_active": true, "delta_db": -0.2838086824303748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031885669741703006, "output_power": 0.00024145256486727228, "reference_active": true, "delta_db": -1.2076372042470906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.908796664605564e-05, "output_power": 5.315511898269741e-05, "reference_active": true, "delta_db": -1.1385731637317127}], "active_band_count": 3, "active_band_mae_db": 0.8766730168030593}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00017810705079567896, "log_rms_mae_db": 0.5111635122479694, "rms_envelope_nrmse": 0.04989061440184852, "envelope_correlation": 0.9973693550984503, "rms_gain_error_db": -0.11428587249358879, "reference_mean_power": 0.0043546897567722925, "output_mean_power": 0.004241589473129155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.684242583835521e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006964729892166045, "output_power": 0.0006395585695247668, "reference_active": true, "delta_db": -0.3702395569687423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017819877477929594, "output_power": 0.00013100846385877136, "reference_active": true, "delta_db": -1.3360535933680073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2119939135052175e-06, "output_power": 3.2846052475167726e-06, "reference_active": true, "delta_db": -2.0052071749691676}], "active_band_count": 3, "active_band_mae_db": 1.2371667751019724}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.00018825872292702335, "log_rms_mae_db": 1.4572144570978816, "rms_envelope_nrmse": 0.10642396026267179, "envelope_correlation": 0.9909659803758089, "rms_gain_error_db": -0.1664479396868303, "reference_mean_power": 0.003284629583795886, "output_mean_power": 0.0031611245395233426}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 2.043578979045215e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008901806972697356, "output_power": 0.0008101952287372674, "reference_active": true, "delta_db": -0.40888491499126456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002718123373610413, "output_power": 0.00019643318904586792, "reference_active": true, "delta_db": -1.410542977791174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0768894288562865e-05, "output_power": 4.235499192242329e-05, "reference_active": true, "delta_db": -1.5677673102580103}], "active_band_count": 3, "active_band_mae_db": 1.1290650676801495}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00015953911997549416, "log_rms_mae_db": 0.8927544406995462, "rms_envelope_nrmse": 0.06541390282848644, "envelope_correlation": 0.9964417218141959, "rms_gain_error_db": -0.16644814917242945, "reference_mean_power": 0.0033118454961761057, "output_mean_power": 0.00318731695523028}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.8056141501282585e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008648498098004977, "output_power": 0.0007639600772709535, "reference_active": true, "delta_db": -0.538700302651698}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013850439438479803, "output_power": 9.569847997284508e-05, "reference_active": true, "delta_db": -1.6055851294165269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.4525026177578845e-06, "output_power": 4.103443728025625e-06, "reference_active": true, "delta_db": -1.9657970724135314}], "active_band_count": 3, "active_band_mae_db": 1.3700275014939187}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 6.604508172479527e-05, "log_rms_mae_db": 1.430978239546954, "rms_envelope_nrmse": 0.18654447831710358, "envelope_correlation": 0.9724569303488646, "rms_gain_error_db": -0.2247332476516269, "reference_mean_power": 0.00023215995390588967, "output_mean_power": 0.0002204519699925602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.444500480390906e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.349295087979117e-05, "output_power": 6.480210875458274e-05, "reference_active": true, "delta_db": -0.5465654679485017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7764968189144465e-05, "output_power": 2.0767545838699817e-05, "reference_active": true, "delta_db": -1.261120025042811}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.892866752425807e-06, "output_power": 4.003019598616687e-06, "reference_active": true, "delta_db": -1.6793890468355128}], "active_band_count": 3, "active_band_mae_db": 1.1623581799422753}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.621650617144381e-05, "log_rms_mae_db": 0.8136875572598892, "rms_envelope_nrmse": 0.11352668107074734, "envelope_correlation": 0.9891688228639614, "rms_gain_error_db": -0.22473860654199648, "reference_mean_power": 0.00023404438013250112, "output_mean_power": 0.00022224108909651126}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.2566052761381804e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.967443215044184e-05, "output_power": 6.0453458847961205e-05, "reference_active": true, "delta_db": -0.6165228379118938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5348913604759878e-05, "output_power": 1.1707431973020146e-05, "reference_active": true, "delta_db": -1.1761599860380358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.927931708651043e-07, "output_power": 3.0882747002653614e-07, "reference_active": true, "delta_db": -2.029487571116195}], "active_band_count": 3, "active_band_mae_db": 1.2740567983553748}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 5.916637942334853e-05, "log_rms_mae_db": 1.456230391183502, "rms_envelope_nrmse": 0.13227553904996409, "envelope_correlation": 0.9861791287576117, "rms_gain_error_db": -0.06244348622807659, "reference_mean_power": 0.00012513898539343656, "output_mean_power": 0.0001233525923182562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 8.947237162820932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.53948723828488e-05, "output_power": 4.369990418302351e-05, "reference_active": true, "delta_db": -0.16526314786146853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2730550619395128e-05, "output_power": 1.2497417892704625e-05, "reference_active": true, "delta_db": -0.08026895938642371}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.684612122113266e-06, "output_power": 1.3492604933198245e-06, "reference_active": true, "delta_db": -0.9640411716068208}], "active_band_count": 3, "active_band_mae_db": 0.403191092951571}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.1723901891273096e-05, "log_rms_mae_db": 0.859302725302152, "rms_envelope_nrmse": 0.07125672355952095, "envelope_correlation": 0.9956371491633135, "rms_gain_error_db": -0.0622955739200206, "reference_mean_power": 0.0001261310627116738, "output_mean_power": 0.00012433474199169747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 8.199333001241026e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1138157243798736e-05, "output_power": 3.963091006756252e-05, "reference_active": true, "delta_db": -0.16210788459305148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.547615689423209e-06, "output_power": 6.500453374894529e-06, "reference_active": true, "delta_db": -0.03139533068300975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.061187881770097e-07, "output_power": 9.363991771976564e-08, "reference_active": true, "delta_db": -0.5433125849928543}], "active_band_count": 3, "active_band_mae_db": 0.2456052667563052}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8947750305397991, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1be32354719289f89ebe6f655d02cff81b4cfad6d6349b5b97f5787e33949924", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9753281382431412, "standard_deviation": 0.0024429337378298853, "minimum": 0.971722708340945, "maximum": 0.9783571830659784, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4517574899595052, "standard_deviation": 0.03377690237817423, "minimum": 1.3993552220472112, "maximum": 1.4971068110172618, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5742179643709644, "standard_deviation": 0.3887457439808358, "minimum": 0.10412197653761524, "maximum": 1.1623581799422753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17659356792262956, "standard_deviation": 0.008481900794472792, "minimum": 0.16570663975749508, "maximum": 0.18904318056969688, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9903924926578763, "standard_deviation": 0.000906674727975373, "minimum": 0.9891688228639614, "maximum": 0.9914484915404509, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8404316149999413, "standard_deviation": 0.03780212420844663, "minimum": 0.8066318925603002, "maximum": 0.9153575076363692, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6643415032953691, "standard_deviation": 0.43164676974698063, "minimum": 0.12453384453125003, "maximum": 1.2740567983553748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10588827112083571, "standard_deviation": 0.005168817412017867, "minimum": 0.09968435139077508, "maximum": 0.11352668107074734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.982149714964712, "standard_deviation": 0.0037492280820370506, "minimum": 0.9755025034465938, "maximum": 0.9864170330595264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.473243774404542, "standard_deviation": 0.027811651779492665, "minimum": 1.4331441133043568, "maximum": 1.5112793636782387, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.48631512207721683, "standard_deviation": 0.23346726623724326, "minimum": 0.14345104189199492, "maximum": 0.8207471121524655, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1515112971469162, "standard_deviation": 0.017862340033150622, "minimum": 0.13185019565781358, "maximum": 0.18642425981156785, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9930989719689375, "standard_deviation": 0.002601851986989817, "minimum": 0.9874725503797417, "maximum": 0.9957254478715251, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9127342514780983, "standard_deviation": 0.04098487747124216, "minimum": 0.859302725302152, "maximum": 0.9652447474885718, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3891059460849766, "standard_deviation": 0.29813523524260777, "minimum": 0.12308904292856872, "maximum": 0.9733748381039365, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09079210005667945, "standard_deviation": 0.01889454495561776, "minimum": 0.07125672355952095, "maximum": 0.13224358969406652, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915957277011501, "standard_deviation": 0.001485079487246829, "minimum": 0.9892759464410206, "maximum": 0.9934558141815031, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.510555958446293, "standard_deviation": 0.043973538987426425, "minimum": 1.4572144570978816, "maximum": 1.5966607276557074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8576085607191106, "standard_deviation": 0.4666350059359238, "minimum": 0.09930028432834298, "maximum": 1.5236792682146358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10278818066435257, "standard_deviation": 0.010040441769884343, "minimum": 0.09034888144312637, "maximum": 0.118812958874804, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967590349905509, "standard_deviation": 0.0008650773264780241, "minimum": 0.9952989907270999, "maximum": 0.9977449302996828, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9355767832466153, "standard_deviation": 0.053701182431714206, "minimum": 0.8927544406995462, "maximum": 1.042227925749822, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1606872490311222, "standard_deviation": 0.4810228735046184, "minimum": 0.4420058579741779, "maximum": 1.9579095102407997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06272985826429799, "standard_deviation": 0.01009621077158298, "minimum": 0.051566388158296464, "maximum": 0.08003037341134202, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1376302241617391, "standard_deviation": 0.046361573817618545, "minimum": 0.09274933864430318, "maximum": 0.22537305015554665, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.560528496823588, "standard_deviation": 0.014017052817091296, "minimum": 5.543692179728356, "maximum": 5.582271345866722, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4989845723044467, "standard_deviation": 0.05041770451469728, "minimum": 0.4531584460399002, "maximum": 0.6050469359782362, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12513508581162353, "standard_deviation": 0.06203295470137244, "minimum": -0.23230706751699537, "maximum": -0.0685902522221008, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9911739022142096, "standard_deviation": 0.0009149225348322536, "minimum": 0.9898100570039075, "maximum": 0.9924054321489538, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9294772308082454, "standard_deviation": 0.03126163339566129, "minimum": 0.8880682126037746, "maximum": 0.9650425696622763, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6966842685014107, "standard_deviation": 0.36642659937813876, "minimum": 0.05239321809863443, "maximum": 1.063593508425908, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09431384404319713, "standard_deviation": 0.005586168295943613, "minimum": 0.08705056823496565, "maximum": 0.10289706196728267, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970612525847051, "standard_deviation": 0.0005138430253163763, "minimum": 0.9962591476908338, "maximum": 0.99772037448722, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5213327633531288, "standard_deviation": 0.032478792409110606, "minimum": 0.4655617364835314, "maximum": 0.5780888523902533, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9249428437616938, "standard_deviation": 0.37770035825053994, "minimum": 0.21170038338276695, "maximum": 1.310856892268748, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05296409770030986, "standard_deviation": 0.006046805338296841, "minimum": 0.04562695640528592, "maximum": 0.06211799531790459, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0014", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.503192901611328, "predicted_mos": 2.503192901611328, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "1fa44d2fba22b6b3e0c90d7499922da90050e74afda73fbf3855b2d0e3e5e3f7", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.16044559981674, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8730500784300516, "全帯域_10ms_rms_envelope_nrmse": 0.07713792833248427, "全帯域_10ms_envelope_correlation": 0.9940600349580182, "全帯域_10ms_modulation_mae_db": 0.41463425445865365, "全帯域_40ms_log_rms_mae_db": 0.46422932711893694, "全帯域_40ms_rms_envelope_nrmse": 0.041935284715552544, "全帯域_40ms_envelope_correlation": 0.9980499472162578, "全帯域_40ms_modulation_mae_db": 0.41469233556643, "300–1000 Hz_10ms_log_rms_mae_db": 1.6328632563435022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0869180643961941, "300–1000 Hz_10ms_envelope_correlation": 0.9939406749961567, "300–1000 Hz_10ms_modulation_mae_db": 0.7869945169452862, "300–1000 Hz_40ms_log_rms_mae_db": 1.033121089676325, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053979721398566424, "300–1000 Hz_40ms_envelope_correlation": 0.9974555927612692, "300–1000 Hz_40ms_modulation_mae_db": 1.0254193991109464, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5081332417165068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18096191202382994, "1000–3000 Hz_10ms_envelope_correlation": 0.9740835266018468, "1000–3000 Hz_10ms_modulation_mae_db": 0.4824290971206833, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8645434414391882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11114378726963825, "1000–3000 Hz_40ms_envelope_correlation": 0.9893464751353922, "1000–3000 Hz_40ms_modulation_mae_db": 0.48189311126940243, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6679032068519706, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1650325166353046, "3000–6000 Hz_10ms_envelope_correlation": 0.9812076447286038, "3000–6000 Hz_10ms_modulation_mae_db": 0.2895617315157298, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0321001670197794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09995887219180212, "3000–6000 Hz_40ms_envelope_correlation": 0.9936503440639001, "3000–6000 Hz_40ms_modulation_mae_db": 0.4887831208303184, "mr_stft_db": 5.6720594887866085, "band_mae_db": 0.160977109239592, "rms_gain_error_db": -0.00448185328975999, "output_peak": 0.5123225870076662}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8763065044953985, "全帯域_10ms_rms_envelope_nrmse": 0.08360164713077234, "全帯域_10ms_envelope_correlation": 0.9929930651283967, "全帯域_10ms_modulation_mae_db": 0.6602177871882463, "全帯域_40ms_log_rms_mae_db": 0.48503374544023203, "全帯域_40ms_rms_envelope_nrmse": 0.04437550776117007, "全帯域_40ms_envelope_correlation": 0.9978341408028414, "全帯域_40ms_modulation_mae_db": 0.9086755143399557, "300–1000 Hz_10ms_log_rms_mae_db": 1.6231897572183298, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09441543502509715, "300–1000 Hz_10ms_envelope_correlation": 0.9928410681091563, "300–1000 Hz_10ms_modulation_mae_db": 0.7538156255264239, "300–1000 Hz_40ms_log_rms_mae_db": 0.9657218746227495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0564436000190558, "300–1000 Hz_40ms_envelope_correlation": 0.997273895007019, "300–1000 Hz_40ms_modulation_mae_db": 0.9369426932393226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5647237170747164, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19097145945279131, "1000–3000 Hz_10ms_envelope_correlation": 0.9716238536876619, "1000–3000 Hz_10ms_modulation_mae_db": 0.5354260357143131, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9101626422958853, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11554665203768899, "1000–3000 Hz_40ms_envelope_correlation": 0.9886160063618001, "1000–3000 Hz_40ms_modulation_mae_db": 0.5119856049285407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6330002289486987, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16236226574856275, "3000–6000 Hz_10ms_envelope_correlation": 0.979098485721937, "3000–6000 Hz_10ms_modulation_mae_db": 0.23869781699405815, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9405547740520667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08228310581829536, "3000–6000 Hz_40ms_envelope_correlation": 0.994207187186835, "3000–6000 Hz_40ms_modulation_mae_db": 0.23244701223365127, "mr_stft_db": 5.692753091453512, "band_mae_db": 0.07528083374225768, "rms_gain_error_db": -0.0650910556316067, "output_peak": 0.47672076868745783}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.874035887017898, "全帯域_10ms_rms_envelope_nrmse": 0.08611758492062443, "全帯域_10ms_envelope_correlation": 0.9925994454774949, "全帯域_10ms_modulation_mae_db": 0.2256571328571455, "全帯域_40ms_log_rms_mae_db": 0.49299679142386876, "全帯域_40ms_rms_envelope_nrmse": 0.05497864110250299, "全帯域_40ms_envelope_correlation": 0.9966994601123029, "全帯域_40ms_modulation_mae_db": 0.31165763843573363, "300–1000 Hz_10ms_log_rms_mae_db": 1.6815300621298133, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09469624216045926, "300–1000 Hz_10ms_envelope_correlation": 0.9928368113049716, "300–1000 Hz_10ms_modulation_mae_db": 0.5790187235433765, "300–1000 Hz_40ms_log_rms_mae_db": 1.0577564767918288, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06081253128692099, "300–1000 Hz_40ms_envelope_correlation": 0.9968498189080115, "300–1000 Hz_40ms_modulation_mae_db": 0.9062762204460059, "1000–3000 Hz_10ms_log_rms_mae_db": 1.591752592208126, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19208184618217208, "1000–3000 Hz_10ms_envelope_correlation": 0.9711066117505485, "1000–3000 Hz_10ms_modulation_mae_db": 0.41626241961882027, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9276518913028926, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11516438977573101, "1000–3000 Hz_40ms_envelope_correlation": 0.9885861594342887, "1000–3000 Hz_40ms_modulation_mae_db": 0.7305764692904756, "3000–6000 Hz_10ms_log_rms_mae_db": 1.670552515517926, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1690166372167951, "3000–6000 Hz_10ms_envelope_correlation": 0.9773369508409652, "3000–6000 Hz_10ms_modulation_mae_db": 0.8854931725180086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.029387635496311, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09574768088339977, "3000–6000 Hz_40ms_envelope_correlation": 0.9921207464027708, "3000–6000 Hz_40ms_modulation_mae_db": 1.0013903744825565, "mr_stft_db": 5.686517848639656, "band_mae_db": 0.1071361722155435, "rms_gain_error_db": -0.09404435095570313, "output_peak": 0.4989703646328863}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8640549691233991, "全帯域_10ms_rms_envelope_nrmse": 0.09228462522700845, "全帯域_10ms_envelope_correlation": 0.9917331260207307, "全帯域_10ms_modulation_mae_db": 1.2128119847805658, "全帯域_40ms_log_rms_mae_db": 0.4614736929825299, "全帯域_40ms_rms_envelope_nrmse": 0.054120177018347454, "全帯域_40ms_envelope_correlation": 0.9971427924732564, "全帯域_40ms_modulation_mae_db": 1.538651728634017, "300–1000 Hz_10ms_log_rms_mae_db": 1.694409877970027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10346064138816952, "300–1000 Hz_10ms_envelope_correlation": 0.9917639374323853, "300–1000 Hz_10ms_modulation_mae_db": 1.4703947692063775, "300–1000 Hz_40ms_log_rms_mae_db": 0.9953486805580342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06539942928986425, "300–1000 Hz_40ms_envelope_correlation": 0.9967723855131043, "300–1000 Hz_40ms_modulation_mae_db": 1.8516731595362337, "1000–3000 Hz_10ms_log_rms_mae_db": 1.567656836851236, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1915841127047528, "1000–3000 Hz_10ms_envelope_correlation": 0.9708815236155064, "1000–3000 Hz_10ms_modulation_mae_db": 0.3205465858803932, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9350305083949932, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1192597324844719, "1000–3000 Hz_40ms_envelope_correlation": 0.9877992893902386, "1000–3000 Hz_40ms_modulation_mae_db": 0.49978906976768184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7540422151947141, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16974247823610067, "3000–6000 Hz_10ms_envelope_correlation": 0.9772524073137564, "3000–6000 Hz_10ms_modulation_mae_db": 0.6096164635159304, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0878500828723217, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09020322079286466, "3000–6000 Hz_40ms_envelope_correlation": 0.9934503032656281, "3000–6000 Hz_40ms_modulation_mae_db": 0.6112521344922897, "mr_stft_db": 5.695807440951834, "band_mae_db": 0.18348915845026223, "rms_gain_error_db": -0.18089620015670738, "output_peak": 0.47612190588774567}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8576334168249804, "全帯域_10ms_rms_envelope_nrmse": 0.08747957106999914, "全帯域_10ms_envelope_correlation": 0.9925073279315046, "全帯域_10ms_modulation_mae_db": 1.0709262802634731, "全帯域_40ms_log_rms_mae_db": 0.45647750638149276, "全帯域_40ms_rms_envelope_nrmse": 0.04678823579004758, "全帯域_40ms_envelope_correlation": 0.9978401634947929, "全帯域_40ms_modulation_mae_db": 1.457117726249382, "300–1000 Hz_10ms_log_rms_mae_db": 1.6703580733237668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10471787839719193, "300–1000 Hz_10ms_envelope_correlation": 0.9914507694051024, "300–1000 Hz_10ms_modulation_mae_db": 1.4156742241874973, "300–1000 Hz_40ms_log_rms_mae_db": 0.9880658190567234, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060310619647354874, "300–1000 Hz_40ms_envelope_correlation": 0.9972391601385688, "300–1000 Hz_40ms_modulation_mae_db": 1.9452277710287345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5834786533895062, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18241391038924992, "1000–3000 Hz_10ms_envelope_correlation": 0.9739989279138009, "1000–3000 Hz_10ms_modulation_mae_db": 0.3005492646215207, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9025590698788305, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11307810727743357, "1000–3000 Hz_40ms_envelope_correlation": 0.9890232144698687, "1000–3000 Hz_40ms_modulation_mae_db": 0.2774094756780617, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6126906985529237, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13812810108227055, "3000–6000 Hz_10ms_envelope_correlation": 0.9849466496306245, "3000–6000 Hz_10ms_modulation_mae_db": 0.3667149359856785, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9574643744728095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07859210385772536, "3000–6000 Hz_40ms_envelope_correlation": 0.9947198402928565, "3000–6000 Hz_40ms_modulation_mae_db": 0.2117192265066498, "mr_stft_db": 5.676030833457798, "band_mae_db": 0.17974000090898853, "rms_gain_error_db": -0.1493770062931788, "output_peak": 0.4880550070322944}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.876973872667954, "全帯域_10ms_rms_envelope_nrmse": 0.08182389595807418, "全帯域_10ms_envelope_correlation": 0.9932821665150878, "全帯域_10ms_modulation_mae_db": 0.4949030102843395, "全帯域_40ms_log_rms_mae_db": 0.4786633073753573, "全帯域_40ms_rms_envelope_nrmse": 0.04460218878300335, "全帯域_40ms_envelope_correlation": 0.9977657188469379, "全帯域_40ms_modulation_mae_db": 0.5286144718078605, "300–1000 Hz_10ms_log_rms_mae_db": 1.6119255082796349, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08439809139880523, "300–1000 Hz_10ms_envelope_correlation": 0.9942855902716662, "300–1000 Hz_10ms_modulation_mae_db": 0.5348515456089111, "300–1000 Hz_40ms_log_rms_mae_db": 1.007206400293391, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04836881858899888, "300–1000 Hz_40ms_envelope_correlation": 0.9979759836675799, "300–1000 Hz_40ms_modulation_mae_db": 0.6561985910287441, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6045099482766163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19151999223968547, "1000–3000 Hz_10ms_envelope_correlation": 0.97163963293086, "1000–3000 Hz_10ms_modulation_mae_db": 0.325284474342782, "1000–3000 Hz_40ms_log_rms_mae_db": 0.940129961680918, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1181117374171353, "1000–3000 Hz_40ms_envelope_correlation": 0.988212717832795, "1000–3000 Hz_40ms_modulation_mae_db": 0.31795153779462076, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6618360942991306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17019730865591487, "3000–6000 Hz_10ms_envelope_correlation": 0.9771369825624097, "3000–6000 Hz_10ms_modulation_mae_db": 0.7572592676683376, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0164347319704163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09399982944182737, "3000–6000 Hz_40ms_envelope_correlation": 0.9928501855798064, "3000–6000 Hz_40ms_modulation_mae_db": 1.0210386546573522, "mr_stft_db": 5.67570409463213, "band_mae_db": 0.14857765031275347, "rms_gain_error_db": -0.009778704011929968, "output_peak": 0.5056667916766242}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9453672995678262, "全帯域_10ms_rms_envelope_nrmse": 0.08813534198860472, "全帯域_10ms_envelope_correlation": 0.9921936164790548, "全帯域_10ms_modulation_mae_db": 0.10024581954913599, "全帯域_40ms_log_rms_mae_db": 0.5236762284219472, "全帯域_40ms_rms_envelope_nrmse": 0.05256102383501278, "全帯域_40ms_envelope_correlation": 0.9969131420411035, "全帯域_40ms_modulation_mae_db": 0.061525194250210226, "300–1000 Hz_10ms_log_rms_mae_db": 1.776822756168326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10540849086749866, "300–1000 Hz_10ms_envelope_correlation": 0.9910436617443164, "300–1000 Hz_10ms_modulation_mae_db": 0.3398589079232304, "300–1000 Hz_40ms_log_rms_mae_db": 1.0661750051850987, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06144900945456307, "300–1000 Hz_40ms_envelope_correlation": 0.9967414209320028, "300–1000 Hz_40ms_modulation_mae_db": 0.6616890802896521, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5223510527914748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18245744467154634, "1000–3000 Hz_10ms_envelope_correlation": 0.9739870461532373, "1000–3000 Hz_10ms_modulation_mae_db": 0.4639747271524204, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8596017362496581, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1085760944808048, "1000–3000 Hz_40ms_envelope_correlation": 0.9898984255493284, "1000–3000 Hz_40ms_modulation_mae_db": 0.6540606213870984, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6593485053017316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1580071277926075, "3000–6000 Hz_10ms_envelope_correlation": 0.9802738825591728, "3000–6000 Hz_10ms_modulation_mae_db": 0.2828031970363158, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0047689662966917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09156886004817728, "3000–6000 Hz_40ms_envelope_correlation": 0.9927883408919276, "3000–6000 Hz_40ms_modulation_mae_db": 0.27800584574812287, "mr_stft_db": 5.689719389491948, "band_mae_db": 0.12460181723309942, "rms_gain_error_db": -0.04849104606206093, "output_peak": 0.7032268590439197}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8819527624245915, "全帯域_10ms_rms_envelope_nrmse": 0.09839541634199699, "全帯域_10ms_envelope_correlation": 0.9902861718915029, "全帯域_10ms_modulation_mae_db": 0.25463762686625707, "全帯域_40ms_log_rms_mae_db": 0.4691078836817394, "全帯域_40ms_rms_envelope_nrmse": 0.053809972087327664, "全帯域_40ms_envelope_correlation": 0.9968148840355702, "全帯域_40ms_modulation_mae_db": 0.3496460188192863, "300–1000 Hz_10ms_log_rms_mae_db": 1.633718321485036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11583921399226457, "300–1000 Hz_10ms_envelope_correlation": 0.9891305514003416, "300–1000 Hz_10ms_modulation_mae_db": 0.43941483050740815, "300–1000 Hz_40ms_log_rms_mae_db": 0.9814463497479838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06742661863001391, "300–1000 Hz_40ms_envelope_correlation": 0.996031622764652, "300–1000 Hz_40ms_modulation_mae_db": 0.6379704696807761, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5418976170004857, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18836053681311946, "1000–3000 Hz_10ms_envelope_correlation": 0.9721851065403043, "1000–3000 Hz_10ms_modulation_mae_db": 0.09132929020726066, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8841986275633095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10782982743801896, "1000–3000 Hz_40ms_envelope_correlation": 0.9899705194487284, "1000–3000 Hz_40ms_modulation_mae_db": 0.2527457132956418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6698183394691306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17336646190611182, "3000–6000 Hz_10ms_envelope_correlation": 0.9767624843774044, "3000–6000 Hz_10ms_modulation_mae_db": 0.7802192454152893, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9992633126002061, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0928406833350951, "3000–6000 Hz_40ms_envelope_correlation": 0.9928049383305355, "3000–6000 Hz_40ms_modulation_mae_db": 0.4258314214659309, "mr_stft_db": 5.69446722493841, "band_mae_db": 0.08486582626448974, "rms_gain_error_db": -0.08116093027157682, "output_peak": 0.48900954124112084}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 58480, "output_samples": 58480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.834792649759768, "spectral_convergence": 0.28129816608444597, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267125480811954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.480370207781517, "spectral_convergence": 0.2602870619510084, "spectral_convergence_status": "ok", "amplitude_floor": 1.0684354905475814e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.701015608818539, "spectral_convergence": 0.27349410670781205, "spectral_convergence_status": "ok", "amplitude_floor": 7.412945744843815e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002955479965644196, "output_power": 0.000311747595784235, "delta_db": 0.2317509312249894, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00146613378566028, "output_power": 0.0014839976055347115, "delta_db": 0.05259598432231802, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0021080578166581473, "output_power": 0.0020563510418360313, "delta_db": -0.10785262366480666, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001787602167832836, "output_power": 0.0001753739098884444, "delta_db": -0.083058881577482, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014763682412872298, "output_power": 0.00016138603513608676, "delta_db": 0.38671257807641296, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001169621027158164, "output_power": 0.00011979379766000723, "delta_db": 0.1038916565715431, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.160977109239592, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.00448185328975999, "pitch": null, "output_rms_dbfs": -23.65658784489625, "output_peak": 0.5123225870076662, "reference_rms_dbfs": -23.65210599160649, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6720594887866085}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 58480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.0002136685864137289, "log_rms_mae_db": 0.8730500784300516, "rms_envelope_nrmse": 0.07713792833248427, "envelope_correlation": 0.9940600349580182, "rms_gain_error_db": -0.004481029249677665, "reference_mean_power": 0.004318997149502425, "output_mean_power": 0.004314543127555378}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.9590597233654063e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007496792113024279, "output_power": 0.0007165744937434855, "reference_active": true, "delta_db": -0.19614122451234953}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031885669741703006, "output_power": 0.0002973680990477381, "reference_active": true, "delta_db": -0.30301166954678194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.908796664605564e-05, "output_power": 5.8200594929859214e-05, "reference_active": true, "delta_db": -0.7447498693168295}], "active_band_count": 3, "active_band_mae_db": 0.41463425445865365}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00017810705079567896, "log_rms_mae_db": 0.46422932711893694, "rms_envelope_nrmse": 0.041935284715552544, "envelope_correlation": 0.9980499472162578, "rms_gain_error_db": -0.0044762936725074535, "reference_mean_power": 0.0043546897567722925, "output_mean_power": 0.0043502036698497655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.684242583835521e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006964729892166045, "output_power": 0.0006563355134201168, "reference_active": true, "delta_db": -0.2577837419804908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017819877477929594, "output_power": 0.0001637327916084635, "reference_active": true, "delta_db": -0.3676904716985103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2119939135052175e-06, "output_power": 4.5200539721217005e-06, "reference_active": true, "delta_db": -0.6186027930202888}], "active_band_count": 3, "active_band_mae_db": 0.41469233556643}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.00018825872292702335, "log_rms_mae_db": 1.6328632563435022, "rms_envelope_nrmse": 0.0869180643961941, "envelope_correlation": 0.9939406749961567, "rms_gain_error_db": -0.05664504093148861, "reference_mean_power": 0.003284629583795886, "output_mean_power": 0.0032420663310312754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 2.043578979045215e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008901806972697356, "output_power": 0.0008610922418187183, "reference_active": true, "delta_db": -0.1442849639522811}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002718123373610413, "output_power": 0.00022850571033999727, "reference_active": true, "delta_db": -0.7537210762237769}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0768894288562865e-05, "output_power": 4.338939386360273e-05, "reference_active": true, "delta_db": -1.4629775106598006}], "active_band_count": 3, "active_band_mae_db": 0.7869945169452862}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00015953911997549416, "log_rms_mae_db": 1.033121089676325, "rms_envelope_nrmse": 0.053979721398566424, "envelope_correlation": 0.9974555927612692, "rms_gain_error_db": -0.056645601150307395, "reference_mean_power": 0.0033118454961761057, "output_mean_power": 0.003268929149469592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.8056141501282585e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008648498098004977, "output_power": 0.000809158803244727, "reference_active": true, "delta_db": -0.2890693084329892}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013850439438479803, "output_power": 0.00011545642835097278, "reference_active": true, "delta_db": -0.7904543420129161}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.4525026177578845e-06, "output_power": 4.074316237252286e-06, "reference_active": true, "delta_db": -1.9967345468869337}], "active_band_count": 3, "active_band_mae_db": 1.0254193991109464}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 6.604508172479527e-05, "log_rms_mae_db": 1.5081332417165068, "rms_envelope_nrmse": 0.18096191202382994, "envelope_correlation": 0.9740835266018468, "rms_gain_error_db": -0.09213345649656207, "reference_mean_power": 0.00023215995390588967, "output_mean_power": 0.00022728666865118213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.444500480390906e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.349295087979117e-05, "output_power": 6.765030656783049e-05, "reference_active": true, "delta_db": -0.3597591645683302}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7764968189144465e-05, "output_power": 2.3140654867476034e-05, "reference_active": true, "delta_db": -0.7912153515093633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.892866752425807e-06, "output_power": 5.504214417363798e-06, "reference_active": true, "delta_db": -0.2963127752843567}], "active_band_count": 3, "active_band_mae_db": 0.4824290971206833}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.621650617144381e-05, "log_rms_mae_db": 0.8645434414391882, "rms_envelope_nrmse": 0.11114378726963825, "envelope_correlation": 0.9893464751353922, "rms_gain_error_db": -0.09224932938024502, "reference_mean_power": 0.00023404438013250112, "output_mean_power": 0.00022912542544744737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.2566052761381804e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.967443215044184e-05, "output_power": 6.340507949348736e-05, "reference_active": true, "delta_db": -0.4094938650401061}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5348913604759878e-05, "output_power": 1.297245708971795e-05, "reference_active": true, "delta_db": -0.7305539875162027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.927931708651043e-07, "output_power": 4.5930538013441484e-07, "reference_active": true, "delta_db": -0.3056314812518986}], "active_band_count": 3, "active_band_mae_db": 0.48189311126940243}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 5.916637942334853e-05, "log_rms_mae_db": 1.6679032068519706, "rms_envelope_nrmse": 0.1650325166353046, "envelope_correlation": 0.9812076447286038, "rms_gain_error_db": 0.3824185056311825, "reference_mean_power": 0.00012513898539343656, "output_mean_power": 0.00013665781734514156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 8.947237162820932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.53948723828488e-05, "output_power": 5.158620945345109e-05, "reference_active": true, "delta_db": 0.5552681763763538}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2730550619395128e-05, "output_power": 1.2950128632359153e-05, "reference_active": true, "delta_db": 0.07426894159622927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.684612122113266e-06, "output_power": 1.5943553146701496e-06, "reference_active": true, "delta_db": -0.23914807657460627}], "active_band_count": 3, "active_band_mae_db": 0.2895617315157298}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.1723901891273096e-05, "log_rms_mae_db": 1.0321001670197794, "rms_envelope_nrmse": 0.09995887219180212, "envelope_correlation": 0.9936503440639001, "rms_gain_error_db": 0.3828680178647983, "reference_mean_power": 0.0001261310627116738, "output_mean_power": 0.00013775547120931542}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 8.199333001241026e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1138157243798736e-05, "output_power": 4.703328057457935e-05, "reference_active": true, "delta_db": 0.5816043800881477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.547615689423209e-06, "output_power": 7.1765622824755415e-06, "reference_active": true, "delta_db": 0.39833277582560894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.061187881770097e-07, "output_power": 1.1869529456595974e-07, "reference_active": true, "delta_db": 0.4864122065771986}], "active_band_count": 3, "active_band_mae_db": 0.4887831208303184}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8884223641856454, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ebabc66cb748a5fc877bd5322ac0961d1a522e0dd187e13d4c74e03517978987", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9724382786492207, "standard_deviation": 0.0013681682131525, "minimum": 0.9708815236155064, "maximum": 0.9740835266018468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5605629574135835, "standard_deviation": 0.03392073779666199, "minimum": 1.5081332417165068, "maximum": 1.6045099482766163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3669752368322742, "standard_deviation": 0.14032117750259612, "minimum": 0.09132929020726066, "maximum": 0.5354260357143131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18754390180964342, "standard_deviation": 0.004790646096756284, "minimum": 0.18096191202382994, "maximum": 0.19208184618217208, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9889316009528051, "standard_deviation": 0.0007757454727130611, "minimum": 0.9877992893902386, "maximum": 0.9899705194487284, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9029847348507094, "standard_deviation": 0.031120978375867848, "minimum": 0.8596017362496581, "maximum": 0.940129961680918, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4658014504264404, "standard_deviation": 0.17396152363943124, "minimum": 0.2527457132956418, "maximum": 0.7305764692904756, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11358879102261535, "standard_deviation": 0.004199727572207535, "minimum": 0.10782982743801896, "maximum": 0.1192597324844719, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792519359668592, "standard_deviation": 0.0028197812041376013, "minimum": 0.9767624843774044, "maximum": 0.9849466496306245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6661489755170282, "standard_deviation": 0.041027218177238686, "minimum": 1.6126906985529237, "maximum": 1.7540422151947141, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5262957288311685, "standard_deviation": 0.2611230262290935, "minimum": 0.23869781699405815, "maximum": 0.8854931725180086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1632316121592085, "standard_deviation": 0.011266730571942421, "minimum": 0.13812810108227055, "maximum": 0.17336646190611182, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933239857517825, "standard_deviation": 0.0008518482185623479, "minimum": 0.9921207464027708, "maximum": 0.9947198402928565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.008478005597575, "standard_deviation": 0.04578377560716249, "minimum": 0.9405547740520667, "maximum": 1.0878500828723217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5338084738021089, "standard_deviation": 0.3241924314054565, "minimum": 0.2117192265066498, "maximum": 1.0210386546573522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09064929454614838, "standard_deviation": 0.007021432411132886, "minimum": 0.07859210385772536, "maximum": 0.09995887219180212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921616330830121, "standard_deviation": 0.0016773485426361486, "minimum": 0.9891305514003416, "maximum": 0.9942855902716662, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6656022016148047, "standard_deviation": 0.05377815837197222, "minimum": 1.6119255082796349, "maximum": 1.776822756168326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7900028929310638, "standard_deviation": 0.4294305889748746, "minimum": 0.3398589079232304, "maximum": 1.4703947692063775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09873175720321005, "standard_deviation": 0.010524618586461615, "minimum": 0.08439809139880523, "maximum": 0.11583921399226457, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997042484961526, "standard_deviation": 0.000580254312963493, "minimum": 0.996031622764652, "maximum": 0.9979759836675799, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0118552119915167, "standard_deviation": 0.03668942931197512, "minimum": 0.9657218746227495, "maximum": 1.0661750051850987, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0776746730450517, "standard_deviation": 0.5274820857902098, "minimum": 0.6379704696807761, "maximum": 1.9452277710287345, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05927379353941728, "standard_deviation": 0.006177400408871674, "minimum": 0.04836881858899888, "maximum": 0.06742661863001391, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13308357104587332, "standard_deviation": 0.04165672070599385, "minimum": 0.07528083374225768, "maximum": 0.18348915845026223, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.685382426543987, "standard_deviation": 0.009443536654087545, "minimum": 5.6720594887866085, "maximum": 5.695807440951834, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5187617281512145, "standard_deviation": 0.07563607845487051, "minimum": 0.47612190588774567, "maximum": 0.7032268590439197, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07916514333406546, "standard_deviation": 0.062146531415844454, "minimum": -0.18089620015670738, "maximum": -0.00448185328975999, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924568693002238, "standard_deviation": 0.00112546303155838, "minimum": 0.9902861718915029, "maximum": 0.9940600349580182, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8811718488190123, "standard_deviation": 0.027063129262587565, "minimum": 0.8576334168249804, "maximum": 0.9453672995678262, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5542542370309771, "standard_deviation": 0.4033365666702125, "minimum": 0.10024581954913599, "maximum": 1.2128119847805658, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08687200137119556, "standard_deviation": 0.006497809161248154, "minimum": 0.07713792833248427, "maximum": 0.09839541634199699, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973825311278829, "standard_deviation": 0.0005440937046216356, "minimum": 0.9966994601123029, "maximum": 0.9980499472162578, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47895731035326305, "standard_deviation": 0.021905124390576736, "minimum": 0.45647750638149276, "maximum": 0.5236762284219472, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6963225785128594, "standard_deviation": 0.5492829739561135, "minimum": 0.061525194250210226, "maximum": 1.538651728634017, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04914637888662056, "standard_deviation": 0.005252678841405422, "minimum": 0.041935284715552544, "maximum": 0.05497864110250299, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0014", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.609426975250244, "predicted_mos": 2.609426975250244, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "433292994aebfd7e4dece1c0c63a97081fd0e1c94625333b7a64c0b93f3532c2", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.904349799733609, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8624018314278176, "全帯域_10ms_rms_envelope_nrmse": 0.08107825906581471, "全帯域_10ms_envelope_correlation": 0.9934017247298148, "全帯域_10ms_modulation_mae_db": 0.3924463342867213, "全帯域_40ms_log_rms_mae_db": 0.4564847776445283, "全帯域_40ms_rms_envelope_nrmse": 0.04473345994955123, "全帯域_40ms_envelope_correlation": 0.9977511762616036, "全帯域_40ms_modulation_mae_db": 0.5416852895726302, "300–1000 Hz_10ms_log_rms_mae_db": 1.6846340425153858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09612532804740824, "300–1000 Hz_10ms_envelope_correlation": 0.9925193292709935, "300–1000 Hz_10ms_modulation_mae_db": 0.5821346400294596, "300–1000 Hz_40ms_log_rms_mae_db": 0.9711565902422187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05931361815025533, "300–1000 Hz_40ms_envelope_correlation": 0.9968833003911777, "300–1000 Hz_40ms_modulation_mae_db": 0.8135227494126355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.529791202806242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18897149146282216, "1000–3000 Hz_10ms_envelope_correlation": 0.9732323104067578, "1000–3000 Hz_10ms_modulation_mae_db": 0.42299364470980755, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8534343539010133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12324364267705304, "1000–3000 Hz_40ms_envelope_correlation": 0.9877923120499338, "1000–3000 Hz_40ms_modulation_mae_db": 0.3390361715545371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6147501875406398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15870969413563354, "3000–6000 Hz_10ms_envelope_correlation": 0.9802382510920957, "3000–6000 Hz_10ms_modulation_mae_db": 0.2930613382008798, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9781028661662214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08507310005128148, "3000–6000 Hz_40ms_envelope_correlation": 0.9937915909095087, "3000–6000 Hz_40ms_modulation_mae_db": 0.1836493557971636, "mr_stft_db": 5.639003165042205, "band_mae_db": 0.13823190483956074, "rms_gain_error_db": -0.006545427232662654, "output_peak": 0.5922666395333801}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9095632781347776, "全帯域_10ms_rms_envelope_nrmse": 0.10335121194110433, "全帯域_10ms_envelope_correlation": 0.9893078100734771, "全帯域_10ms_modulation_mae_db": 0.4008485896462783, "全帯域_40ms_log_rms_mae_db": 0.4852054518588301, "全帯域_40ms_rms_envelope_nrmse": 0.054717494729806956, "全帯域_40ms_envelope_correlation": 0.9967725795565953, "全帯域_40ms_modulation_mae_db": 0.5320092851573187, "300–1000 Hz_10ms_log_rms_mae_db": 1.6409794054462858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11213889261954355, "300–1000 Hz_10ms_envelope_correlation": 0.9900211326676828, "300–1000 Hz_10ms_modulation_mae_db": 0.7291313965191938, "300–1000 Hz_40ms_log_rms_mae_db": 1.0758875436599584, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06238653499823108, "300–1000 Hz_40ms_envelope_correlation": 0.9969109381738436, "300–1000 Hz_40ms_modulation_mae_db": 0.7609737915268799, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4964145056315592, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18054278681458585, "1000–3000 Hz_10ms_envelope_correlation": 0.9747683523873008, "1000–3000 Hz_10ms_modulation_mae_db": 0.1397562265819621, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8552607429551162, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11474823959052369, "1000–3000 Hz_40ms_envelope_correlation": 0.9888585848339123, "1000–3000 Hz_40ms_modulation_mae_db": 0.2068288226819416, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6501245474789095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16996597775454445, "3000–6000 Hz_10ms_envelope_correlation": 0.9775958776437305, "3000–6000 Hz_10ms_modulation_mae_db": 0.2914427908890123, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0467854687189293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10610876160526456, "3000–6000 Hz_40ms_envelope_correlation": 0.9905026602448475, "3000–6000 Hz_40ms_modulation_mae_db": 0.48639040515376, "mr_stft_db": 5.659075085187723, "band_mae_db": 0.13893483753271038, "rms_gain_error_db": -0.10093229962627562, "output_peak": 0.4630193737713418}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8379337684082068, "全帯域_10ms_rms_envelope_nrmse": 0.0824564588008422, "全帯域_10ms_envelope_correlation": 0.9931796073892663, "全帯域_10ms_modulation_mae_db": 0.11539838148537727, "全帯域_40ms_log_rms_mae_db": 0.45320490603443603, "全帯域_40ms_rms_envelope_nrmse": 0.045690326640681476, "全帯域_40ms_envelope_correlation": 0.9976544884816694, "全帯域_40ms_modulation_mae_db": 0.2999317673364081, "300–1000 Hz_10ms_log_rms_mae_db": 1.6641239399017669, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09391019137200235, "300–1000 Hz_10ms_envelope_correlation": 0.9928708809222306, "300–1000 Hz_10ms_modulation_mae_db": 0.07797027049837561, "300–1000 Hz_40ms_log_rms_mae_db": 0.9764898367546196, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05676946152790616, "300–1000 Hz_40ms_envelope_correlation": 0.9971288283103443, "300–1000 Hz_40ms_modulation_mae_db": 0.3969497759988451, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5278987980205354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19032329989786928, "1000–3000 Hz_10ms_envelope_correlation": 0.9721844502528321, "1000–3000 Hz_10ms_modulation_mae_db": 0.5144839484821709, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8542906910045595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11033043033871161, "1000–3000 Hz_40ms_envelope_correlation": 0.9898055837380899, "1000–3000 Hz_40ms_modulation_mae_db": 0.23648828536980393, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6460034333754994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15971394489693946, "3000–6000 Hz_10ms_envelope_correlation": 0.9799974277258338, "3000–6000 Hz_10ms_modulation_mae_db": 0.8610192511379378, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9887756254329665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08249856140456777, "3000–6000 Hz_40ms_envelope_correlation": 0.9946888228644686, "3000–6000 Hz_40ms_modulation_mae_db": 0.9204496136753589, "mr_stft_db": 5.652726823405769, "band_mae_db": 0.113670537966295, "rms_gain_error_db": -0.0006649113627880336, "output_peak": 0.4646972154471452}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.858248148088796, "全帯域_10ms_rms_envelope_nrmse": 0.08846586230108791, "全帯域_10ms_envelope_correlation": 0.9921528019883592, "全帯域_10ms_modulation_mae_db": 0.269922005596285, "全帯域_40ms_log_rms_mae_db": 0.4465343174387637, "全帯域_40ms_rms_envelope_nrmse": 0.047809033033067154, "全帯域_40ms_envelope_correlation": 0.9974907111750166, "全帯域_40ms_modulation_mae_db": 0.4327082330085498, "300–1000 Hz_10ms_log_rms_mae_db": 1.6371537460965337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10292064071669405, "300–1000 Hz_10ms_envelope_correlation": 0.9915199044618515, "300–1000 Hz_10ms_modulation_mae_db": 0.6283256825639363, "300–1000 Hz_40ms_log_rms_mae_db": 1.012820840104229, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06582288220884851, "300–1000 Hz_40ms_envelope_correlation": 0.9963094667768657, "300–1000 Hz_40ms_modulation_mae_db": 0.9292607135379196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5598655271772959, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1829767873241771, "1000–3000 Hz_10ms_envelope_correlation": 0.9735822190231591, "1000–3000 Hz_10ms_modulation_mae_db": 0.3989712774450302, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9137101993083007, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11358893306244114, "1000–3000 Hz_40ms_envelope_correlation": 0.9888248437268118, "1000–3000 Hz_40ms_modulation_mae_db": 0.7114419380768547, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6545451678596568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1580587913459139, "3000–6000 Hz_10ms_envelope_correlation": 0.9809938211280189, "3000–6000 Hz_10ms_modulation_mae_db": 0.2344230927085581, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9567379874155926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08577576549980609, "3000–6000 Hz_40ms_envelope_correlation": 0.994109891349301, "3000–6000 Hz_40ms_modulation_mae_db": 0.16981452850434478, "mr_stft_db": 5.652564299701207, "band_mae_db": 0.0468702299507688, "rms_gain_error_db": -0.07360967333726529, "output_peak": 0.49140444265629535}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8862536438205936, "全帯域_10ms_rms_envelope_nrmse": 0.08865891015305538, "全帯域_10ms_envelope_correlation": 0.9921150310746376, "全帯域_10ms_modulation_mae_db": 0.4285341290572296, "全帯域_40ms_log_rms_mae_db": 0.4802546865762535, "全帯域_40ms_rms_envelope_nrmse": 0.04543790794317692, "全帯域_40ms_envelope_correlation": 0.9977375143048096, "全帯域_40ms_modulation_mae_db": 0.4786406071236624, "300–1000 Hz_10ms_log_rms_mae_db": 1.5948702591491744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09613123055527531, "300–1000 Hz_10ms_envelope_correlation": 0.9925644398325407, "300–1000 Hz_10ms_modulation_mae_db": 0.612618061864353, "300–1000 Hz_40ms_log_rms_mae_db": 0.9331420878617923, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05897610777511419, "300–1000 Hz_40ms_envelope_correlation": 0.9970051361152023, "300–1000 Hz_40ms_modulation_mae_db": 0.8795862161615129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5186259306704182, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17862072046596508, "1000–3000 Hz_10ms_envelope_correlation": 0.9748485544962998, "1000–3000 Hz_10ms_modulation_mae_db": 0.32044619964712623, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8568283894911835, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10304908461563024, "1000–3000 Hz_40ms_envelope_correlation": 0.9908283467525737, "1000–3000 Hz_40ms_modulation_mae_db": 0.4773822260338485, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6254616630801513, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16457413378863867, "3000–6000 Hz_10ms_envelope_correlation": 0.9785647206776058, "3000–6000 Hz_10ms_modulation_mae_db": 0.4092324131629323, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9519301284499988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09364773629666327, "3000–6000 Hz_40ms_envelope_correlation": 0.9924212659988754, "3000–6000 Hz_40ms_modulation_mae_db": 0.3555713134072576, "mr_stft_db": 5.652006301762128, "band_mae_db": 0.06832249093964955, "rms_gain_error_db": -0.06542392762903296, "output_peak": 0.5301257881057376}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8846803447568315, "全帯域_10ms_rms_envelope_nrmse": 0.0882442769635869, "全帯域_10ms_envelope_correlation": 0.9926026387004788, "全帯域_10ms_modulation_mae_db": 1.1724695802344909, "全帯域_40ms_log_rms_mae_db": 0.47491394642631707, "全帯域_40ms_rms_envelope_nrmse": 0.05208383994322519, "全帯域_40ms_envelope_correlation": 0.9975563120636808, "全帯域_40ms_modulation_mae_db": 1.5385438454702525, "300–1000 Hz_10ms_log_rms_mae_db": 1.7294690050710588, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11320356997986367, "300–1000 Hz_10ms_envelope_correlation": 0.9904858572152682, "300–1000 Hz_10ms_modulation_mae_db": 1.635324440126185, "300–1000 Hz_40ms_log_rms_mae_db": 1.0432986662678891, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0739038579012372, "300–1000 Hz_40ms_envelope_correlation": 0.9963258609252352, "300–1000 Hz_40ms_modulation_mae_db": 2.0070728253177, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5378764166521035, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18285326845191385, "1000–3000 Hz_10ms_envelope_correlation": 0.974539131874897, "1000–3000 Hz_10ms_modulation_mae_db": 0.32200659945716326, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8506122386921899, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11349445253433539, "1000–3000 Hz_40ms_envelope_correlation": 0.9894056134551295, "1000–3000 Hz_40ms_modulation_mae_db": 0.32864097332581854, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6155257646632442, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15408618000988872, "3000–6000 Hz_10ms_envelope_correlation": 0.9814283547061301, "3000–6000 Hz_10ms_modulation_mae_db": 0.09718365422652485, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9896577435671756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0805098925385035, "3000–6000 Hz_40ms_envelope_correlation": 0.9944734625055478, "3000–6000 Hz_40ms_modulation_mae_db": 0.5219175036699629, "mr_stft_db": 5.658884613344216, "band_mae_db": 0.17604729409320896, "rms_gain_error_db": -0.21272460851047517, "output_peak": 0.4926102381962249}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9498188069013893, "全帯域_10ms_rms_envelope_nrmse": 0.10484006812059754, "全帯域_10ms_envelope_correlation": 0.9890726419633801, "全帯域_10ms_modulation_mae_db": 1.0203867434327691, "全帯域_40ms_log_rms_mae_db": 0.50461222052712, "全帯域_40ms_rms_envelope_nrmse": 0.060361008293425736, "全帯域_40ms_envelope_correlation": 0.9961766753672019, "全帯域_40ms_modulation_mae_db": 1.3783433289737748, "300–1000 Hz_10ms_log_rms_mae_db": 1.7424258647607687, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11513860527963178, "300–1000 Hz_10ms_envelope_correlation": 0.9896580113390547, "300–1000 Hz_10ms_modulation_mae_db": 1.1801499786058953, "300–1000 Hz_40ms_log_rms_mae_db": 1.0250092992417643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07211748449242379, "300–1000 Hz_40ms_envelope_correlation": 0.9959750852982806, "300–1000 Hz_40ms_modulation_mae_db": 1.4358517179336199, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5277970323009833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17654436370314433, "1000–3000 Hz_10ms_envelope_correlation": 0.975562080656852, "1000–3000 Hz_10ms_modulation_mae_db": 0.20061848059165813, "1000–3000 Hz_40ms_log_rms_mae_db": 0.877276346446991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10124725349819992, "1000–3000 Hz_40ms_envelope_correlation": 0.9911674760936348, "1000–3000 Hz_40ms_modulation_mae_db": 0.36458553628409823, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6542997546312301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18014600454600768, "3000–6000 Hz_10ms_envelope_correlation": 0.9756007295039413, "3000–6000 Hz_10ms_modulation_mae_db": 0.9058166208223096, "3000–6000 Hz_40ms_log_rms_mae_db": 1.010661484938177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09795316343016876, "3000–6000 Hz_40ms_envelope_correlation": 0.9924645162065046, "3000–6000 Hz_40ms_modulation_mae_db": 1.2206080704631666, "mr_stft_db": 5.682225651754393, "band_mae_db": 0.16279622539013008, "rms_gain_error_db": -0.13674394051627334, "output_peak": 0.46682730437952447}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8477014493228179, "全帯域_10ms_rms_envelope_nrmse": 0.08207475131872843, "全帯域_10ms_envelope_correlation": 0.9932439716328855, "全帯域_10ms_modulation_mae_db": 0.24391639699838008, "全帯域_40ms_log_rms_mae_db": 0.48694765926789346, "全帯域_40ms_rms_envelope_nrmse": 0.04419963236517543, "全帯域_40ms_envelope_correlation": 0.9978163198766038, "全帯域_40ms_modulation_mae_db": 0.6349514012413716, "300–1000 Hz_10ms_log_rms_mae_db": 1.5816062960937904, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09606159536887132, "300–1000 Hz_10ms_envelope_correlation": 0.9925499000304935, "300–1000 Hz_10ms_modulation_mae_db": 0.6352805762904458, "300–1000 Hz_40ms_log_rms_mae_db": 0.9255382611048862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05422183009058947, "300–1000 Hz_40ms_envelope_correlation": 0.9974212298937112, "300–1000 Hz_40ms_modulation_mae_db": 1.059858860281463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6140930761606005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19440376859583364, "1000–3000 Hz_10ms_envelope_correlation": 0.9707789730800411, "1000–3000 Hz_10ms_modulation_mae_db": 0.1661176442368854, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9591730805946265, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11851349287894346, "1000–3000 Hz_40ms_envelope_correlation": 0.988148299441478, "1000–3000 Hz_40ms_modulation_mae_db": 0.298947306326767, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6088872692076353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16559371786433827, "3000–6000 Hz_10ms_envelope_correlation": 0.9793799289223274, "3000–6000 Hz_10ms_modulation_mae_db": 0.31246974301041874, "3000–6000 Hz_40ms_log_rms_mae_db": 0.954334575218182, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09790359153542656, "3000–6000 Hz_40ms_envelope_correlation": 0.9924468911797609, "3000–6000 Hz_40ms_modulation_mae_db": 0.6759469280369265, "mr_stft_db": 5.650964780953867, "band_mae_db": 0.1257207020561641, "rms_gain_error_db": -0.011036367084312214, "output_peak": 0.4750332807178351}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 58480, "output_samples": 58480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.826149537225763, "spectral_convergence": 0.2829838260412932, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267125480811954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.423177675382141, "spectral_convergence": 0.2543542981113626, "spectral_convergence_status": "ok", "amplitude_floor": 1.0684354905475814e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.66768228251871, "spectral_convergence": 0.25409566797061367, "spectral_convergence_status": "ok", "amplitude_floor": 7.412945744843815e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002955479965644196, "output_power": 0.00030349421136479835, "delta_db": 0.1152239232622561, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00146613378566028, "output_power": 0.0014794140234455579, "delta_db": 0.03916129319379833, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0021080578166581473, "output_power": 0.002066785328952816, "delta_db": -0.085871478324423, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001787602167832836, "output_power": 0.00018792575807183553, "delta_db": 0.21715438158405528, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014763682412872298, "output_power": 0.00015573427819672974, "delta_db": 0.23189519970832184, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001169621027158164, "output_power": 0.00011324959355288249, "delta_db": -0.14008515296450993, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13823190483956074, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.006545427232662654, "pitch": null, "output_rms_dbfs": -23.658651418839153, "output_peak": 0.5922666395333801, "reference_rms_dbfs": -23.65210599160649, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.639003165042205}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 58480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.0002136685864137289, "log_rms_mae_db": 0.8624018314278176, "rms_envelope_nrmse": 0.08107825906581471, "envelope_correlation": 0.9934017247298148, "rms_gain_error_db": -0.006545960593921144, "reference_mean_power": 0.004318997149502425, "output_mean_power": 0.004312492187923685}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.9590597233654063e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007496792113024279, "output_power": 0.0007565194514848842, "reference_active": true, "delta_db": 0.039446310270419754}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031885669741703006, "output_power": 0.00029845002834079576, "reference_active": true, "delta_db": -0.2872391907771398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.908796664605564e-05, "output_power": 5.679852541007072e-05, "reference_active": true, "delta_db": -0.8506535018126045}], "active_band_count": 3, "active_band_mae_db": 0.3924463342867213}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00017810705079567896, "log_rms_mae_db": 0.4564847776445283, "rms_envelope_nrmse": 0.04473345994955123, "envelope_correlation": 0.9977511762616036, "rms_gain_error_db": -0.006541089867874645, "reference_mean_power": 0.0043546897567722925, "output_mean_power": 0.004348135914111545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.684242583835521e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006964729892166045, "output_power": 0.0006954948796561875, "reference_active": true, "delta_db": -0.0061034117841305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017819877477929594, "output_power": 0.0001683108030606076, "reference_active": true, "delta_db": -0.247927215391828}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2119939135052175e-06, "output_power": 3.8010306861313865e-06, "reference_active": true, "delta_db": -1.3710252415419322}], "active_band_count": 3, "active_band_mae_db": 0.5416852895726302}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.00018825872292702335, "log_rms_mae_db": 1.6846340425153858, "rms_envelope_nrmse": 0.09612532804740824, "envelope_correlation": 0.9925193292709935, "rms_gain_error_db": -0.0576714508542902, "reference_mean_power": 0.003284629583795886, "output_mean_power": 0.003241300192848592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 2.043578979045215e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008901806972697356, "output_power": 0.0008803142148596608, "reference_active": true, "delta_db": -0.04840458105600652}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002718123373610413, "output_power": 0.00024343271686004607, "reference_active": true, "delta_db": -0.4789021904110386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0768894288562865e-05, "output_power": 4.589565998204141e-05, "reference_active": true, "delta_db": -1.2190971486213336}], "active_band_count": 3, "active_band_mae_db": 0.5821346400294596}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00015953911997549416, "log_rms_mae_db": 0.9711565902422187, "rms_envelope_nrmse": 0.05931361815025533, "envelope_correlation": 0.9968833003911777, "rms_gain_error_db": -0.05767134269027072, "reference_mean_power": 0.0033118454961761057, "output_mean_power": 0.003268157166260893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.8056141501282585e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008648498098004977, "output_power": 0.0008431184951731887, "reference_active": true, "delta_db": -0.11052077880818609}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013850439438479803, "output_power": 0.00012288157331533756, "reference_active": true, "delta_db": -0.5197678943892612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.4525026177578845e-06, "output_power": 4.253047487867138e-06, "reference_active": true, "delta_db": -1.8102795750404592}], "active_band_count": 3, "active_band_mae_db": 0.8135227494126355}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 6.604508172479527e-05, "log_rms_mae_db": 1.529791202806242, "rms_envelope_nrmse": 0.18897149146282216, "envelope_correlation": 0.9732323104067578, "rms_gain_error_db": 0.22116312245210434, "reference_mean_power": 0.00023215995390588967, "output_mean_power": 0.0002442888365305924}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.444500480390906e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.349295087979117e-05, "output_power": 7.400126944267967e-05, "reference_active": true, "delta_db": 0.029934843610036014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7764968189144465e-05, "output_power": 3.143956119836569e-05, "reference_active": true, "delta_db": 0.5397929573665535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.892866752425807e-06, "output_power": 6.922324187612846e-06, "reference_active": true, "delta_db": 0.6992531331528331}], "active_band_count": 3, "active_band_mae_db": 0.42299364470980755}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.621650617144381e-05, "log_rms_mae_db": 0.8534343539010133, "rms_envelope_nrmse": 0.12324364267705304, "envelope_correlation": 0.9877923120499338, "rms_gain_error_db": 0.2212954054014432, "reference_mean_power": 0.00023404438013250112, "output_mean_power": 0.00024627921342731653}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.2566052761381804e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.967443215044184e-05, "output_power": 7.112687312202087e-05, "reference_active": true, "delta_db": 0.08960278791933007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5348913604759878e-05, "output_power": 1.815143227700744e-05, "reference_active": true, "delta_db": 0.7283325808329242}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.927931708651043e-07, "output_power": 5.159195621227618e-07, "reference_active": true, "delta_db": 0.1991731459113569}], "active_band_count": 3, "active_band_mae_db": 0.3390361715545371}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 5.916637942334853e-05, "log_rms_mae_db": 1.6147501875406398, "rms_envelope_nrmse": 0.15870969413563354, "envelope_correlation": 0.9802382510920957, "rms_gain_error_db": 0.030287295117079185, "reference_mean_power": 0.00012513898539343656, "output_mean_power": 0.00012601474327053155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 8.947237162820932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.53948723828488e-05, "output_power": 4.351592800592988e-05, "reference_active": true, "delta_db": -0.183585499438926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2730550619395128e-05, "output_power": 1.1318329009165604e-05, "reference_active": true, "delta_db": -0.5106487392269717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.684612122113266e-06, "output_power": 1.7579029666217011e-06, "reference_active": true, "delta_db": 0.18494977593674183}], "active_band_count": 3, "active_band_mae_db": 0.2930613382008798}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.1723901891273096e-05, "log_rms_mae_db": 0.9781028661662214, "rms_envelope_nrmse": 0.08507310005128148, "envelope_correlation": 0.9937915909095087, "rms_gain_error_db": 0.030260318348357424, "reference_mean_power": 0.0001261310627116738, "output_mean_power": 0.0001270129744652319}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 8.199333001241026e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1138157243798736e-05, "output_power": 3.9435664163298676e-05, "reference_active": true, "delta_db": -0.1835567438654151}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.547615689423209e-06, "output_power": 6.060927536451208e-06, "reference_active": true, "delta_db": -0.3354408900706419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.061187881770097e-07, "output_power": 1.0690236796678764e-07, "reference_active": true, "delta_db": 0.03195043345543386}], "active_band_count": 3, "active_band_mae_db": 0.1836493557971636}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8942643449101514, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9f06fdaec19074163f2950e24c6223ccc97f754e697e765598bc2df12e6203ee", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9736870090222676, "standard_deviation": 0.0015898151745178531, "minimum": 0.9707789730800411, "maximum": 0.975562080656852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5390453111774671, "standard_deviation": 0.03512784502850454, "minimum": 1.4964145056315592, "maximum": 1.6140930761606005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.31067425264397547, "standard_deviation": 0.1333074712310219, "minimum": 0.1397562265819621, "maximum": 0.5144839484821709, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1844045608395389, "standard_deviation": 0.006215951088069132, "minimum": 0.17654436370314433, "maximum": 0.19440376859583364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893538825114454, "standard_deviation": 0.0012004172630670099, "minimum": 0.9877923120499338, "maximum": 0.9911674760936348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8775732552992476, "standard_deviation": 0.03917790340641, "minimum": 0.8506122386921899, "maximum": 0.9591730805946265, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3704189074567087, "standard_deviation": 0.16049453759456378, "minimum": 0.2068288226819416, "maximum": 0.7114419380768547, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11227694114947981, "standard_deviation": 0.007358865817466434, "minimum": 0.10124725349819992, "maximum": 0.12324364267705304, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792248889249604, "standard_deviation": 0.0019219101527748815, "minimum": 0.9756007295039413, "maximum": 0.9814283547061301, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6336997234796207, "standard_deviation": 0.019467743184798082, "minimum": 1.6088872692076353, "maximum": 1.6545451678596568, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4255811130198217, "standard_deviation": 0.29598137898664406, "minimum": 0.09718365422652485, "maximum": 0.9058166208223096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16385605554273808, "standard_deviation": 0.008271548059499891, "minimum": 0.15408618000988872, "maximum": 0.18014600454600768, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9931123876573518, "standard_deviation": 0.0014114305139883768, "minimum": 0.9905026602448475, "maximum": 0.9946888228644686, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9846232349884054, "standard_deviation": 0.032494102219658226, "minimum": 0.9519301284499988, "maximum": 1.0467854687189293, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5667934648384926, "standard_deviation": 0.36291393588272897, "minimum": 0.16981452850434478, "maximum": 1.2206080704631666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09118382154521025, "standard_deviation": 0.00906924553085829, "minimum": 0.0805098925385035, "maximum": 0.10610876160526456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915236819675144, "standard_deviation": 0.0012960398363122535, "minimum": 0.9896580113390547, "maximum": 0.9928708809222306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6594078198793456, "standard_deviation": 0.05795588673385591, "minimum": 1.5816062960937904, "maximum": 1.7424258647607687, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7601168808122306, "standard_deviation": 0.462151301107802, "minimum": 0.07797027049837561, "maximum": 1.635324440126185, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10320375674241128, "standard_deviation": 0.008943148119848698, "minimum": 0.09391019137200235, "maximum": 0.11513860527963178, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967449807355826, "standard_deviation": 0.0004894719768054002, "minimum": 0.9959750852982806, "maximum": 0.9974212298937112, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9954178906546698, "standard_deviation": 0.0529889045009641, "minimum": 0.9255382611048862, "maximum": 1.0758875436599584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0353845812713218, "standard_deviation": 0.48884840213918657, "minimum": 0.3969497759988451, "maximum": 2.0070728253177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06293897214307571, "standard_deviation": 0.007131367781740942, "minimum": 0.05422183009058947, "maximum": 0.0739038579012372, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12132427784606097, "standard_deviation": 0.04430580991660029, "minimum": 0.0468702299507688, "maximum": 0.17604729409320896, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.655931340143939, "standard_deviation": 0.01229539644509181, "minimum": 5.639003165042205, "maximum": 5.682225651754393, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.49699803535093556, "standard_deviation": 0.04442443319229736, "minimum": 0.4630193737713418, "maximum": 0.5922666395333801, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07596014441238566, "standard_deviation": 0.0735257054825273, "minimum": -0.21272460851047517, "maximum": -0.0006649113627880336, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918845284440374, "standard_deviation": 0.001732634965538511, "minimum": 0.9890726419633801, "maximum": 0.9934017247298148, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8795751588576538, "standard_deviation": 0.03665247174311381, "minimum": 0.8379337684082068, "maximum": 0.9498188069013893, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5054902700921915, "standard_deviation": 0.38092290533239936, "minimum": 0.11539838148537727, "maximum": 1.1724695802344909, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08989622483310217, "standard_deviation": 0.009296259105188826, "minimum": 0.08107825906581471, "maximum": 0.10484006812059754, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973694721358977, "standard_deviation": 0.0005844060775409287, "minimum": 0.9961766753672019, "maximum": 0.9978163198766038, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4735197457217678, "standard_deviation": 0.019863843650840166, "minimum": 0.4465343174387637, "maximum": 0.50461222052712, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.729601719735496, "standard_deviation": 0.4620313172074179, "minimum": 0.2999317673364081, "maximum": 1.5385438454702525, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04937908786226376, "standard_deviation": 0.005811692606659549, "minimum": 0.04419963236517543, "maximum": 0.060361008293425736, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_6829-68769-0038", "language": "en", "style": "normal", "speaker": "librispeech_6829", "source_utterance_id": "6829-68769-0038", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 40080, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_6829-68769-0038.wav", "source_file_sha256": "be19277bf4caa99a71243f647a62637857fd2b67531e9d756a10ad84a04757df", "decoded_pcm_sha256": "2491d28a0087423aef64b260b9a3bef96a9ffd38bdd5460aeb1d8818a5a4e0a8", "original_source_sha256": "0f141cbbe137a2927bbf38106a3b14d2b2f186b86e1a54bcbd354d8c2a8981c4", "corpus_entry_id": "librispeech_6829-68769-0038"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.2226054668426514, "predicted_mos": 3.2226054668426514, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "6934a0611fddcb9e50971df510c2041882a867c4b3899e450e1acf3ed2970bc2", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3184462179997354, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_6829-68769-0038", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6934a0611fddcb9e50971df510c2041882a867c4b3899e450e1acf3ed2970bc2", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0d340ad2987da76914c8379223b9ebe6ae6f6ad2c3070d2beb52a071272d0c7a"}, "rows": [{"case": "librispeech_6829-68769-0038", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.606018543243408, "predicted_mos": 2.606018543243408, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "0a85217cb47fc352c37c17f20758bc49862a5905386b2aa8f66a8ba2bd2cb96a", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.088366218998999, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_6829-68769-0038", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "0a85217cb47fc352c37c17f20758bc49862a5905386b2aa8f66a8ba2bd2cb96a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9172785265511405, "全帯域_10ms_rms_envelope_nrmse": 0.08763766156440823, "全帯域_10ms_envelope_correlation": 0.993091107741212, "全帯域_10ms_modulation_mae_db": 0.19494954225664393, "全帯域_40ms_log_rms_mae_db": 0.47012856586758256, "全帯域_40ms_rms_envelope_nrmse": 0.04738486070682242, "全帯域_40ms_envelope_correlation": 0.9980569472174338, "全帯域_40ms_modulation_mae_db": 0.12173018495932254, "300–1000 Hz_10ms_log_rms_mae_db": 1.5549098031278359, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10915238327530558, "300–1000 Hz_10ms_envelope_correlation": 0.9915809569454597, "300–1000 Hz_10ms_modulation_mae_db": 0.4020103203820347, "300–1000 Hz_40ms_log_rms_mae_db": 0.8920527617798834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06808932308003177, "300–1000 Hz_40ms_envelope_correlation": 0.9971727331085494, "300–1000 Hz_40ms_modulation_mae_db": 0.6766719627637748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5177126309761366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20152531326865608, "1000–3000 Hz_10ms_envelope_correlation": 0.9628935487749539, "1000–3000 Hz_10ms_modulation_mae_db": 0.8101763745679121, "1000–3000 Hz_40ms_log_rms_mae_db": 0.77949508822923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10021579215058393, "1000–3000 Hz_40ms_envelope_correlation": 0.9897536545614792, "1000–3000 Hz_40ms_modulation_mae_db": 0.8243441577082485, "3000–6000 Hz_10ms_log_rms_mae_db": 1.723612219362208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19453276139082717, "3000–6000 Hz_10ms_envelope_correlation": 0.9717122548850029, "3000–6000 Hz_10ms_modulation_mae_db": 0.5619921490238143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0935917973658975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13060278701506053, "3000–6000 Hz_40ms_envelope_correlation": 0.986399041736525, "3000–6000 Hz_40ms_modulation_mae_db": 0.6550696005796911, "mr_stft_db": 5.847254266451807, "band_mae_db": 0.36775348394342017, "rms_gain_error_db": -0.1673328068222579, "output_peak": 0.39309567358754116}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9182753453863264, "全帯域_10ms_rms_envelope_nrmse": 0.09443927709973934, "全帯域_10ms_envelope_correlation": 0.9918330108083747, "全帯域_10ms_modulation_mae_db": 0.5267000279791526, "全帯域_40ms_log_rms_mae_db": 0.4870306094871119, "全帯域_40ms_rms_envelope_nrmse": 0.04998896913214998, "全帯域_40ms_envelope_correlation": 0.9976698910503075, "全帯域_40ms_modulation_mae_db": 0.5123436299536877, "300–1000 Hz_10ms_log_rms_mae_db": 1.6323063783855796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13228681083528274, "300–1000 Hz_10ms_envelope_correlation": 0.9868738703821521, "300–1000 Hz_10ms_modulation_mae_db": 0.7176620036988014, "300–1000 Hz_40ms_log_rms_mae_db": 0.9749655374589227, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08571329986864029, "300–1000 Hz_40ms_envelope_correlation": 0.99451802883709, "300–1000 Hz_40ms_modulation_mae_db": 0.6733110765752318, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5054503044654466, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20020797478251998, "1000–3000 Hz_10ms_envelope_correlation": 0.9632775550776287, "1000–3000 Hz_10ms_modulation_mae_db": 0.2753630290352819, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8305485530520418, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1019418276500099, "1000–3000 Hz_40ms_envelope_correlation": 0.9894374992830383, "1000–3000 Hz_40ms_modulation_mae_db": 0.293212779254788, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7402838938103105, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2022698814182919, "3000–6000 Hz_10ms_envelope_correlation": 0.9701371062419899, "3000–6000 Hz_10ms_modulation_mae_db": 1.075415665912727, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0858286831123631, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13368916817570134, "3000–6000 Hz_40ms_envelope_correlation": 0.9859521451210698, "3000–6000 Hz_40ms_modulation_mae_db": 0.8743017925860218, "mr_stft_db": 5.862860607419837, "band_mae_db": 0.47482267285509555, "rms_gain_error_db": -0.1361417676939145, "output_peak": 0.40421148769920645}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9043822758484396, "全帯域_10ms_rms_envelope_nrmse": 0.07765413120785462, "全帯域_10ms_envelope_correlation": 0.9944159461058758, "全帯域_10ms_modulation_mae_db": 0.13721628501323407, "全帯域_40ms_log_rms_mae_db": 0.47357253320602316, "全帯域_40ms_rms_envelope_nrmse": 0.04108301276389569, "全帯域_40ms_envelope_correlation": 0.9982859898361223, "全帯域_40ms_modulation_mae_db": 0.21196748524393935, "300–1000 Hz_10ms_log_rms_mae_db": 1.6702491555944254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10086821159177174, "300–1000 Hz_10ms_envelope_correlation": 0.9923192189039866, "300–1000 Hz_10ms_modulation_mae_db": 0.2378955556070539, "300–1000 Hz_40ms_log_rms_mae_db": 0.9143033662757858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060544072991783285, "300–1000 Hz_40ms_envelope_correlation": 0.9972052558207187, "300–1000 Hz_40ms_modulation_mae_db": 0.4001817065218627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5705617495198678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19765579472071876, "1000–3000 Hz_10ms_envelope_correlation": 0.9641814699977073, "1000–3000 Hz_10ms_modulation_mae_db": 0.5019568943972607, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9411591717421219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11228106846096737, "1000–3000 Hz_40ms_envelope_correlation": 0.9872345035672124, "1000–3000 Hz_40ms_modulation_mae_db": 0.8874715991195946, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7239182102669262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18886415166064185, "3000–6000 Hz_10ms_envelope_correlation": 0.9739866576874755, "3000–6000 Hz_10ms_modulation_mae_db": 0.4937184482067754, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0760642608884665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1284067668390032, "3000–6000 Hz_40ms_envelope_correlation": 0.9870991242274707, "3000–6000 Hz_40ms_modulation_mae_db": 0.5064580705429075, "mr_stft_db": 5.867119931848386, "band_mae_db": 0.29976187976282476, "rms_gain_error_db": 0.0008522110674246174, "output_peak": 0.4335247802020952}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.986359832933446, "全帯域_10ms_rms_envelope_nrmse": 0.10557659031965078, "全帯域_10ms_envelope_correlation": 0.9896618735270999, "全帯域_10ms_modulation_mae_db": 0.5392628495892858, "全帯域_40ms_log_rms_mae_db": 0.5569384092641254, "全帯域_40ms_rms_envelope_nrmse": 0.050639608860752944, "全帯域_40ms_envelope_correlation": 0.9974726773063561, "全帯域_40ms_modulation_mae_db": 0.37508783327019146, "300–1000 Hz_10ms_log_rms_mae_db": 1.6646802612929554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13934227177611594, "300–1000 Hz_10ms_envelope_correlation": 0.9852236111615057, "300–1000 Hz_10ms_modulation_mae_db": 0.6920003832048076, "300–1000 Hz_40ms_log_rms_mae_db": 0.9883711237469843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08261094198042475, "300–1000 Hz_40ms_envelope_correlation": 0.9948037526675594, "300–1000 Hz_40ms_modulation_mae_db": 1.0074741540607584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4877896584834183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19027350440351695, "1000–3000 Hz_10ms_envelope_correlation": 0.967191472938669, "1000–3000 Hz_10ms_modulation_mae_db": 0.3488843213548109, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8565831489234005, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11157843206798952, "1000–3000 Hz_40ms_envelope_correlation": 0.9887838215010236, "1000–3000 Hz_40ms_modulation_mae_db": 0.5609760531748781, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7589187841633402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19062236316871872, "3000–6000 Hz_10ms_envelope_correlation": 0.9740152908455666, "3000–6000 Hz_10ms_modulation_mae_db": 0.6617586759545669, "3000–6000 Hz_40ms_log_rms_mae_db": 1.090886436486867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13189347170007493, "3000–6000 Hz_40ms_envelope_correlation": 0.9867258847289605, "3000–6000 Hz_40ms_modulation_mae_db": 0.32239085146343793, "mr_stft_db": 5.875772823486121, "band_mae_db": 0.4131817896506192, "rms_gain_error_db": -0.0851012997632907, "output_peak": 0.42116899173988503}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9629963410831827, "全帯域_10ms_rms_envelope_nrmse": 0.11923202076931634, "全帯域_10ms_envelope_correlation": 0.987162990074526, "全帯域_10ms_modulation_mae_db": 0.4518984734576294, "全帯域_40ms_log_rms_mae_db": 0.47446234464631537, "全帯域_40ms_rms_envelope_nrmse": 0.06973209733316973, "全帯域_40ms_envelope_correlation": 0.9952732439472262, "全帯域_40ms_modulation_mae_db": 0.5180825419087245, "300–1000 Hz_10ms_log_rms_mae_db": 1.716826299057055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14381213770190918, "300–1000 Hz_10ms_envelope_correlation": 0.9841919909506606, "300–1000 Hz_10ms_modulation_mae_db": 0.5704092361013847, "300–1000 Hz_40ms_log_rms_mae_db": 0.9615631530530724, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08119901430915749, "300–1000 Hz_40ms_envelope_correlation": 0.9944833187481816, "300–1000 Hz_40ms_modulation_mae_db": 0.08382979684944707, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5344587949584712, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20281999713129825, "1000–3000 Hz_10ms_envelope_correlation": 0.9630999012745994, "1000–3000 Hz_10ms_modulation_mae_db": 0.05840849982995861, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8061075786557681, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1018243082568114, "1000–3000 Hz_40ms_envelope_correlation": 0.9897220791914547, "1000–3000 Hz_40ms_modulation_mae_db": 0.08497732524680879, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7335060129214932, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1922902532460253, "3000–6000 Hz_10ms_envelope_correlation": 0.9725310532042913, "3000–6000 Hz_10ms_modulation_mae_db": 0.6033412617901658, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1112175370282382, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12575262887276492, "3000–6000 Hz_40ms_envelope_correlation": 0.9873441836102137, "3000–6000 Hz_40ms_modulation_mae_db": 0.2977935003193561, "mr_stft_db": 5.837320910208676, "band_mae_db": 0.3705455817335958, "rms_gain_error_db": 0.10359703372670585, "output_peak": 0.43558839168522523}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0085823901443718, "全帯域_10ms_rms_envelope_nrmse": 0.1192810626977186, "全帯域_10ms_envelope_correlation": 0.9868681391876173, "全帯域_10ms_modulation_mae_db": 0.32962340499830534, "全帯域_40ms_log_rms_mae_db": 0.5579366758433214, "全帯域_40ms_rms_envelope_nrmse": 0.07662310343525512, "全帯域_40ms_envelope_correlation": 0.9940623537078664, "全帯域_40ms_modulation_mae_db": 0.4839359729352643, "300–1000 Hz_10ms_log_rms_mae_db": 1.7984707640401432, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1600785344591368, "300–1000 Hz_10ms_envelope_correlation": 0.9801005624490757, "300–1000 Hz_10ms_modulation_mae_db": 0.17317292312496346, "300–1000 Hz_40ms_log_rms_mae_db": 1.0034848772709202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11744913822262337, "300–1000 Hz_40ms_envelope_correlation": 0.9882856191261934, "300–1000 Hz_40ms_modulation_mae_db": 0.532693879820823, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5942145593107293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18816008664012318, "1000–3000 Hz_10ms_envelope_correlation": 0.9673711389480785, "1000–3000 Hz_10ms_modulation_mae_db": 0.26133755219167304, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9160209228999439, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.098582800428528, "1000–3000 Hz_40ms_envelope_correlation": 0.9904161410909403, "1000–3000 Hz_40ms_modulation_mae_db": 0.15719986757706392, "3000–6000 Hz_10ms_log_rms_mae_db": 1.733594708318374, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19049943774559647, "3000–6000 Hz_10ms_envelope_correlation": 0.9730306745120798, "3000–6000 Hz_10ms_modulation_mae_db": 0.8329981046475271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.039045262886562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12284384172120763, "3000–6000 Hz_40ms_envelope_correlation": 0.9880656000572373, "3000–6000 Hz_40ms_modulation_mae_db": 0.4650335490566275, "mr_stft_db": 5.855505319920655, "band_mae_db": 0.41104229018915683, "rms_gain_error_db": 0.015074480694332185, "output_peak": 0.38664216320960737}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9440394964421356, "全帯域_10ms_rms_envelope_nrmse": 0.11484925266605549, "全帯域_10ms_envelope_correlation": 0.9877555054534123, "全帯域_10ms_modulation_mae_db": 0.7838624116702748, "全帯域_40ms_log_rms_mae_db": 0.5191713374411375, "全帯域_40ms_rms_envelope_nrmse": 0.06178631721462392, "全帯域_40ms_envelope_correlation": 0.9962125493433582, "全帯域_40ms_modulation_mae_db": 1.0222896909761137, "300–1000 Hz_10ms_log_rms_mae_db": 1.7075636319894363, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15354238263325035, "300–1000 Hz_10ms_envelope_correlation": 0.9819176321092992, "300–1000 Hz_10ms_modulation_mae_db": 0.7978101302205975, "300–1000 Hz_40ms_log_rms_mae_db": 0.9373350476365027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09280771555721584, "300–1000 Hz_40ms_envelope_correlation": 0.9932560953303075, "300–1000 Hz_40ms_modulation_mae_db": 1.1830719329416393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.543589781780485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22340041423006213, "1000–3000 Hz_10ms_envelope_correlation": 0.957538558191405, "1000–3000 Hz_10ms_modulation_mae_db": 0.4424326436789492, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8647581449399593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12375089666910129, "1000–3000 Hz_40ms_envelope_correlation": 0.986287610211111, "1000–3000 Hz_40ms_modulation_mae_db": 0.47679592581895297, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8308530229016773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18030104431755084, "3000–6000 Hz_10ms_envelope_correlation": 0.9771615653934035, "3000–6000 Hz_10ms_modulation_mae_db": 0.6940651106628821, "3000–6000 Hz_40ms_log_rms_mae_db": 1.213557383130838, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12732479666499097, "3000–6000 Hz_40ms_envelope_correlation": 0.9879784108650884, "3000–6000 Hz_40ms_modulation_mae_db": 0.3241446948477517, "mr_stft_db": 5.867619919439881, "band_mae_db": 0.33841270452943273, "rms_gain_error_db": -0.0990413681564365, "output_peak": 0.568664784627302}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9918293306546534, "全帯域_10ms_rms_envelope_nrmse": 0.10330167782686775, "全帯域_10ms_envelope_correlation": 0.9901662618855901, "全帯域_10ms_modulation_mae_db": 0.26965691440759926, "全帯域_40ms_log_rms_mae_db": 0.519940431907855, "全帯域_40ms_rms_envelope_nrmse": 0.05198194705477098, "全帯域_40ms_envelope_correlation": 0.9972844368369235, "全帯域_40ms_modulation_mae_db": 0.3501089997700631, "300–1000 Hz_10ms_log_rms_mae_db": 1.628657305252182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12197784126314659, "300–1000 Hz_10ms_envelope_correlation": 0.9884560697239728, "300–1000 Hz_10ms_modulation_mae_db": 0.27298059970417987, "300–1000 Hz_40ms_log_rms_mae_db": 0.87477545184273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06030852105287203, "300–1000 Hz_40ms_envelope_correlation": 0.996937474623891, "300–1000 Hz_40ms_modulation_mae_db": 0.4887765123876675, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6235913452050732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22504931411490145, "1000–3000 Hz_10ms_envelope_correlation": 0.9533283834883439, "1000–3000 Hz_10ms_modulation_mae_db": 0.26008670268144446, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0166266011099034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12278768858369682, "1000–3000 Hz_40ms_envelope_correlation": 0.98494117177055, "1000–3000 Hz_40ms_modulation_mae_db": 0.3457242970029198, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8309979444156292, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19754750806657487, "3000–6000 Hz_10ms_envelope_correlation": 0.9719369055255708, "3000–6000 Hz_10ms_modulation_mae_db": 0.8456601895834209, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2036009777585397, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13743426945755127, "3000–6000 Hz_40ms_envelope_correlation": 0.9854748809613512, "3000–6000 Hz_40ms_modulation_mae_db": 0.9070105317660734, "mr_stft_db": 5.870239930592227, "band_mae_db": 0.286706827541319, "rms_gain_error_db": 0.024885896880425633, "output_peak": 0.3993328200095866}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 40080, "output_samples": 40080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.933762041569281, "spectral_convergence": 0.3045300135895497, "spectral_convergence_status": "ok", "amplitude_floor": 1.4347408585118857e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.681271064305966, "spectral_convergence": 0.2991000076047986, "spectral_convergence_status": "ok", "amplitude_floor": 9.356428468567288e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.926729693480174, "spectral_convergence": 0.30587684663837, "spectral_convergence_status": "ok", "amplitude_floor": 6.30553407186755e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020558767458207216, "output_power": 0.0002659634988409415, "delta_db": 1.1182496338292176, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002656133227213937, "output_power": 0.0025060289334507676, "delta_db": -0.25263773879136386, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000658897280424305, "output_power": 0.0006097576542729542, "delta_db": -0.3366045430779018, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.296049498403337e-05, "output_power": 6.533112651613472e-05, "delta_db": 0.16052013161549197, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018191885630194454, "output_power": 0.0001785903378057265, "delta_db": -0.08019758259094519, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012525230408717123, "output_power": 0.00011801972445219579, "delta_db": -0.2583112737556006, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.36775348394342017, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1673328068222579, "pitch": null, "output_rms_dbfs": -24.266999726863464, "output_peak": 0.39309567358754116, "reference_rms_dbfs": -24.099666920041205, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.847254266451807}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 40080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.0001593748762509214, "log_rms_mae_db": 0.9172785265511405, "rms_envelope_nrmse": 0.08763766156440823, "envelope_correlation": 0.993091107741212, "rms_gain_error_db": -0.1673219510561802, "reference_mean_power": 0.003898496329889168, "output_mean_power": 0.003751154350178515}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.3730898670333144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011160072986886684, "output_power": 0.0010956063366187087, "reference_active": true, "delta_db": -0.08012499494661963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003924601048841299, "output_power": 0.0003690215767086593, "reference_active": true, "delta_db": -0.26743755459996943}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.521600297301381e-05, "output_power": 9.000150414730553e-05, "reference_active": true, "delta_db": 0.23728607722334272}], "active_band_count": 3, "active_band_mae_db": 0.19494954225664393}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00014354118626608257, "log_rms_mae_db": 0.47012856586758256, "rms_envelope_nrmse": 0.04738486070682242, "envelope_correlation": 0.9980569472174338, "rms_gain_error_db": -0.1673092576291596, "reference_mean_power": 0.00394559944963645, "output_mean_power": 0.0037964883241674097}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 2.0339309551628037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010452646972319804, "output_power": 0.0010349234975391179, "reference_active": true, "delta_db": -0.04318035235517721}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020675655639960477, "output_power": 0.0001959230239067728, "reference_active": true, "delta_db": -0.23373815096625128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.033282667633223e-06, "output_power": 6.15716622285706e-06, "reference_active": true, "delta_db": 0.08827205155653917}], "active_band_count": 3, "active_band_mae_db": 0.12173018495932254}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.00014555177137634006, "log_rms_mae_db": 1.5549098031278359, "rms_envelope_nrmse": 0.10915238327530558, "envelope_correlation": 0.9915809569454597, "rms_gain_error_db": -0.3329706026046523, "reference_mean_power": 0.0025948122953778335, "output_mean_power": 0.002403304982792609}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.091667884405576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010721637846984168, "output_power": 0.0010256899394161156, "reference_active": true, "delta_db": -0.19245037922328773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00045198280410970506, "output_power": 0.00037915495220286434, "reference_active": true, "delta_db": -0.7630517944699434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.681659745338991e-05, "output_power": 5.363179629177407e-05, "reference_active": true, "delta_db": -0.2505287874528729}], "active_band_count": 3, "active_band_mae_db": 0.4020103203820347}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00013266563413733552, "log_rms_mae_db": 0.8920527617798834, "rms_envelope_nrmse": 0.06808932308003177, "envelope_correlation": 0.9971727331085494, "rms_gain_error_db": -0.3329697119323926, "reference_mean_power": 0.002626297782108314, "output_mean_power": 0.002432467216120038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 1.815151890487515e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010442677388382842, "output_power": 0.000997754162102882, "reference_active": true, "delta_db": -0.19788313196055535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002292871431701824, "output_power": 0.00019620769015722456, "reference_active": true, "delta_db": -0.6766367809577846}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.238038641621491e-06, "output_power": 5.547167754792333e-06, "reference_active": true, "delta_db": -1.1554959753729845}], "active_band_count": 3, "active_band_mae_db": 0.6766719627637748}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.977119603351369e-05, "log_rms_mae_db": 1.5177126309761366, "rms_envelope_nrmse": 0.20152531326865608, "envelope_correlation": 0.9628935487749539, "rms_gain_error_db": -0.054700142138533095, "reference_mean_power": 0.0001519473982974274, "output_mean_power": 0.00015004559640412012}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 8.081548208249742e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5556653827814076e-05, "output_power": 4.605071020820045e-05, "reference_active": true, "delta_db": 0.046845151014470066}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1818343242486535e-05, "output_power": 1.2279779513138331e-05, "reference_active": true, "delta_db": 0.16633969869178897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7000790662746577e-06, "output_power": 6.165175330180856e-06, "reference_active": true, "delta_db": 2.2173442739974774}], "active_band_count": 3, "active_band_mae_db": 0.8101763745679121}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.109627865595404e-05, "log_rms_mae_db": 0.77949508822923, "rms_envelope_nrmse": 0.10021579215058393, "envelope_correlation": 0.9897536545614792, "rms_gain_error_db": -0.05464805385390352, "reference_mean_power": 0.0001537098951112271, "output_mean_power": 0.00015178785397785256}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 6.838335178235257e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3201238570188224e-05, "output_power": 4.5078647968358376e-05, "reference_active": true, "delta_db": 0.1847468377050647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.064245816738516e-06, "output_power": 6.551934350290369e-06, "reference_active": true, "delta_db": 0.33592739898907475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0299327724151498e-07, "output_power": 3.182126781970874e-07, "reference_active": true, "delta_db": 1.9523582364306058}], "active_band_count": 3, "active_band_mae_db": 0.8243441577082485}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.882145378793849e-05, "log_rms_mae_db": 1.723612219362208, "rms_envelope_nrmse": 0.19453276139082717, "envelope_correlation": 0.9717122548850029, "rms_gain_error_db": -0.1431396334318197, "reference_mean_power": 0.00013710154501992013, "output_mean_power": 0.0001326564544546502}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 1.0763544842397476e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5103901227352814e-05, "output_power": 4.367815820819201e-05, "reference_active": true, "delta_db": -0.1394979043170182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.11132679073469e-05, "output_power": 2.351254400310787e-05, "reference_active": true, "delta_db": -1.2164600725204706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.435692144349255e-06, "output_power": 5.037940164046684e-06, "reference_active": true, "delta_db": -0.3300184702339545}], "active_band_count": 3, "active_band_mae_db": 0.5619921490238143}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.3950747306848176e-05, "log_rms_mae_db": 1.0935917973658975, "rms_envelope_nrmse": 0.13060278701506053, "envelope_correlation": 0.986399041736525, "rms_gain_error_db": -0.14269708810801282, "reference_mean_power": 0.00013864857515666164, "output_mean_power": 0.00013416699785503142}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 9.21368112936246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.381631867120006e-05, "output_power": 4.265867476698333e-05, "reference_active": true, "delta_db": -0.11628526962973786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.911163628753255e-05, "output_power": 1.4082734860458592e-05, "reference_active": true, "delta_db": -1.3261086900699117}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5242776229186804e-07, "output_power": 3.975132384237524e-07, "reference_active": true, "delta_db": 0.5228148420394237}], "active_band_count": 3, "active_band_mae_db": 0.6550696005796911}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.87529772713814, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cc3bf7458d8d3a439ccc62984d69b0afa7106c7bd1fee72d223433d1d045a1e6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9623602535864232, "standard_deviation": 0.00475488966660812, "minimum": 0.9533283834883439, "maximum": 0.9673711389480785, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5471711030874535, "standard_deviation": 0.04615079900190002, "minimum": 1.4877896584834183, "maximum": 1.6235913452050732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.36983075221716133, "standard_deviation": 0.2225802818393918, "minimum": 0.05840849982995861, "maximum": 0.8101763745679121, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2036365499114746, "standard_deviation": 0.013731389186530486, "minimum": 0.18816008664012318, "maximum": 0.22504931411490145, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9883220601471012, "standard_deviation": 0.0019495160929371524, "minimum": 0.98494117177055, "maximum": 0.9904161410909403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8764124011940462, "standard_deviation": 0.0778177800757975, "minimum": 0.77949508822923, "maximum": 1.0166266011099034, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.45383775061290677, "standard_deviation": 0.2924917467169172, "minimum": 0.08497732524680879, "maximum": 0.8874715991195946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10912035178346102, "standard_deviation": 0.010085179070250205, "minimum": 0.098582800428528, "maximum": 0.12375089666910129, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9730639385369225, "standard_deviation": 0.0020872298921981937, "minimum": 0.9701371062419899, "maximum": 0.9771615653934035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7594605995199948, "standard_deviation": 0.04547143863960879, "minimum": 1.723612219362208, "maximum": 1.8309979444156292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.721118700722735, "standard_deviation": 0.18849520331527658, "minimum": 0.4937184482067754, "maximum": 1.075415665912727, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1921159251267784, "standard_deviation": 0.0064750687520855514, "minimum": 0.18030104431755084, "maximum": 0.2022698814182919, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9868799089134896, "standard_deviation": 0.0009231918440193972, "minimum": 0.9854748809613512, "maximum": 0.9880656000572373, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1142240423322216, "standard_deviation": 0.0618165903037824, "minimum": 1.039045262886562, "maximum": 1.213557383130838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5440253238952334, "standard_deviation": 0.24463444409006851, "minimum": 0.2977935003193561, "maximum": 0.9070105317660734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12974346630579436, "standard_deviation": 0.004648291587286008, "minimum": 0.12284384172120763, "maximum": 0.13743426945755127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.986332989078264, "standard_deviation": 0.004347893569974208, "minimum": 0.9801005624490757, "maximum": 0.9923192189039866, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6717079498424514, "standard_deviation": 0.07219939833275163, "minimum": 1.5549098031278359, "maximum": 1.7984707640401432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4829926440054779, "standard_deviation": 0.24267252811247092, "minimum": 0.17317292312496346, "maximum": 0.7978101302205975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13263257169198986, "standard_deviation": 0.020820458650583535, "minimum": 0.10086821159177174, "maximum": 0.1600785344591368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9945827847828114, "standard_deviation": 0.002942473807706101, "minimum": 0.9882856191261934, "maximum": 0.9972052558207187, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9433564148831002, "standard_deviation": 0.04658038789969786, "minimum": 0.87477545184273, "maximum": 1.0034848772709202, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6307513777401506, "standard_deviation": 0.34495461362130825, "minimum": 0.08382979684944707, "maximum": 1.1830719329416393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0810902533828436, "standard_deviation": 0.018920942381374675, "minimum": 0.06030852105287203, "maximum": 0.11744913822262337, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.370278403775683, "standard_deviation": 0.06257630434487624, "minimum": 0.286706827541319, "maximum": 0.47482267285509555, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.860461713670949, "standard_deviation": 0.012905532402882374, "minimum": 5.837320910208676, "maximum": 5.875772823486121, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4302786365950561, "standard_deviation": 0.05879297444333467, "minimum": 0.38664216320960737, "maximum": 0.568664784627302, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.042900952508376416, "standard_deviation": 0.09292236287887555, "minimum": -0.1673328068222579, "maximum": 0.10359703372670585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901193543479635, "standard_deviation": 0.002811669503533212, "minimum": 0.9868681391876173, "maximum": 0.9944159461058758, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.954217942380462, "standard_deviation": 0.03914050927034828, "minimum": 0.9043822758484396, "maximum": 1.0085823901443718, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40414623867151567, "standard_deviation": 0.21327261789751453, "minimum": 0.13721628501323407, "maximum": 0.7838624116702748, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1027464592689514, "standard_deviation": 0.015245010130475372, "minimum": 0.07765413120785462, "maximum": 0.1192810626977186, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9967897611556993, "standard_deviation": 0.0014833320221416203, "minimum": 0.9940623537078664, "maximum": 0.9982859898361223, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.507397613457934, "standard_deviation": 0.03652219460583684, "minimum": 0.47012856586758256, "maximum": 0.5579366758433214, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4494432923771633, "standard_deviation": 0.2718185361536948, "minimum": 0.12173018495932254, "maximum": 1.0222896909761137, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0561524895626801, "standard_deviation": 0.012100333542137824, "minimum": 0.04108301276389569, "maximum": 0.07662310343525512, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0038", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.3632683753967285, "predicted_mos": 2.3632683753967285, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "60a1c1b6a0b75a6cf6a2aa8a22642c4edd06c0e04a1b4fe974f220702e161db1", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.729571199975908, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9535205909850029, "全帯域_10ms_rms_envelope_nrmse": 0.09054468396892704, "全帯域_10ms_envelope_correlation": 0.9925729797247242, "全帯域_10ms_modulation_mae_db": 0.14069423376525203, "全帯域_40ms_log_rms_mae_db": 0.4832044547238493, "全帯域_40ms_rms_envelope_nrmse": 0.05042921752322914, "全帯域_40ms_envelope_correlation": 0.9977108835509563, "全帯域_40ms_modulation_mae_db": 0.12595977947376827, "300–1000 Hz_10ms_log_rms_mae_db": 1.6371056347546455, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10838861838545848, "300–1000 Hz_10ms_envelope_correlation": 0.991587726092633, "300–1000 Hz_10ms_modulation_mae_db": 0.5375732479998281, "300–1000 Hz_40ms_log_rms_mae_db": 0.9325956310604061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06708780601027409, "300–1000 Hz_40ms_envelope_correlation": 0.9971260300167607, "300–1000 Hz_40ms_modulation_mae_db": 0.8491233123747065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5165172083265683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20241336716767105, "1000–3000 Hz_10ms_envelope_correlation": 0.9625835839532476, "1000–3000 Hz_10ms_modulation_mae_db": 0.7217413783325713, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7964431655133459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1049016532132692, "1000–3000 Hz_40ms_envelope_correlation": 0.9892372355340973, "1000–3000 Hz_40ms_modulation_mae_db": 0.7126439643616681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5643270261912978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18474912270555172, "3000–6000 Hz_10ms_envelope_correlation": 0.9746140182700935, "3000–6000 Hz_10ms_modulation_mae_db": 0.492013188605596, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9342736079783915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1263907885331295, "3000–6000 Hz_40ms_envelope_correlation": 0.987464156260105, "3000–6000 Hz_40ms_modulation_mae_db": 0.4469533443910552, "mr_stft_db": 5.786886411425058, "band_mae_db": 0.360132687966904, "rms_gain_error_db": -0.15794175645610378, "output_peak": 0.38075248012536583}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9041422425365068, "全帯域_10ms_rms_envelope_nrmse": 0.09470853482710775, "全帯域_10ms_envelope_correlation": 0.9917786524025932, "全帯域_10ms_modulation_mae_db": 0.5290930374311125, "全帯域_40ms_log_rms_mae_db": 0.46473505220900285, "全帯域_40ms_rms_envelope_nrmse": 0.04769518906783087, "全帯域_40ms_envelope_correlation": 0.9978911403012578, "全帯域_40ms_modulation_mae_db": 0.5015658545988654, "300–1000 Hz_10ms_log_rms_mae_db": 1.6588644871015765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12700942947643662, "300–1000 Hz_10ms_envelope_correlation": 0.9878370005498419, "300–1000 Hz_10ms_modulation_mae_db": 0.8627448799337573, "300–1000 Hz_40ms_log_rms_mae_db": 0.9398224449352267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07592809365975381, "300–1000 Hz_40ms_envelope_correlation": 0.9957085970716675, "300–1000 Hz_40ms_modulation_mae_db": 0.8260904589698201, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4018430031928908, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17951983246089215, "1000–3000 Hz_10ms_envelope_correlation": 0.9711766731186771, "1000–3000 Hz_10ms_modulation_mae_db": 0.19559431751322423, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7500042049143393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08824187226847453, "1000–3000 Hz_40ms_envelope_correlation": 0.9937425741063025, "1000–3000 Hz_40ms_modulation_mae_db": 0.20863165302593714, "3000–6000 Hz_10ms_log_rms_mae_db": 1.658673410809543, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18803577872614535, "3000–6000 Hz_10ms_envelope_correlation": 0.9740873981655385, "3000–6000 Hz_10ms_modulation_mae_db": 0.7813448425777093, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0399928421541302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12155876089594683, "3000–6000 Hz_40ms_envelope_correlation": 0.9883074442333029, "3000–6000 Hz_40ms_modulation_mae_db": 0.8864898106479661, "mr_stft_db": 5.789735859370809, "band_mae_db": 0.4496351512354309, "rms_gain_error_db": -0.13319432052439156, "output_peak": 0.39351499243084354}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9615595873258924, "全帯域_10ms_rms_envelope_nrmse": 0.08626112585497622, "全帯域_10ms_envelope_correlation": 0.9931423536910411, "全帯域_10ms_modulation_mae_db": 0.2774744655541178, "全帯域_40ms_log_rms_mae_db": 0.5037341853566695, "全帯域_40ms_rms_envelope_nrmse": 0.044929461121290426, "全帯域_40ms_envelope_correlation": 0.9979696690149579, "全帯域_40ms_modulation_mae_db": 0.4697523531570493, "300–1000 Hz_10ms_log_rms_mae_db": 1.7261566019342918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10244148708387822, "300–1000 Hz_10ms_envelope_correlation": 0.9919395216989707, "300–1000 Hz_10ms_modulation_mae_db": 0.2074809382988688, "300–1000 Hz_40ms_log_rms_mae_db": 0.9450264114200362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056286876881892, "300–1000 Hz_40ms_envelope_correlation": 0.9974459350478835, "300–1000 Hz_40ms_modulation_mae_db": 0.353303604616497, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5369697483179074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18682133818447605, "1000–3000 Hz_10ms_envelope_correlation": 0.9682943539416368, "1000–3000 Hz_10ms_modulation_mae_db": 0.6916644651804074, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8985964025362917, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10382020512702272, "1000–3000 Hz_40ms_envelope_correlation": 0.9898836944271657, "1000–3000 Hz_40ms_modulation_mae_db": 1.063829991679018, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6515366080638978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17609781263244118, "3000–6000 Hz_10ms_envelope_correlation": 0.9773279375341559, "3000–6000 Hz_10ms_modulation_mae_db": 0.3290059864912451, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9341833202045335, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12020436770292633, "3000–6000 Hz_40ms_envelope_correlation": 0.9885816849904259, "3000–6000 Hz_40ms_modulation_mae_db": 0.5041333839108595, "mr_stft_db": 5.81087046129392, "band_mae_db": 0.2980646870422979, "rms_gain_error_db": 0.030181151514928683, "output_peak": 0.45926200435249726}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0046410238545433, "全帯域_10ms_rms_envelope_nrmse": 0.1080419125876229, "全帯域_10ms_envelope_correlation": 0.9891563678027689, "全帯域_10ms_modulation_mae_db": 0.41425335655085327, "全帯域_40ms_log_rms_mae_db": 0.5739645829412711, "全帯域_40ms_rms_envelope_nrmse": 0.054385400515190245, "全帯域_40ms_envelope_correlation": 0.9970351869830762, "全帯域_40ms_modulation_mae_db": 0.18537201901860678, "300–1000 Hz_10ms_log_rms_mae_db": 1.7451976952633022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14771397156015004, "300–1000 Hz_10ms_envelope_correlation": 0.9831160207692372, "300–1000 Hz_10ms_modulation_mae_db": 0.41942082182965246, "300–1000 Hz_40ms_log_rms_mae_db": 1.0472894409054827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08443287098978637, "300–1000 Hz_40ms_envelope_correlation": 0.9942389012220729, "300–1000 Hz_40ms_modulation_mae_db": 0.726505649771891, "1000–3000 Hz_10ms_log_rms_mae_db": 1.432254891841909, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18448821426201104, "1000–3000 Hz_10ms_envelope_correlation": 0.9713215624624876, "1000–3000 Hz_10ms_modulation_mae_db": 0.6438120386668764, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8168907163624852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11044249198876799, "1000–3000 Hz_40ms_envelope_correlation": 0.9919373685270191, "1000–3000 Hz_40ms_modulation_mae_db": 0.9004793296860448, "3000–6000 Hz_10ms_log_rms_mae_db": 1.674413296393849, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18530130664715055, "3000–6000 Hz_10ms_envelope_correlation": 0.9755550664283579, "3000–6000 Hz_10ms_modulation_mae_db": 0.5875777874772602, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9675581821658528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1269198989176028, "3000–6000 Hz_40ms_envelope_correlation": 0.9877895656537646, "3000–6000 Hz_40ms_modulation_mae_db": 0.3638025379194729, "mr_stft_db": 5.810484752631136, "band_mae_db": 0.3770034774662812, "rms_gain_error_db": -0.059636294248879874, "output_peak": 0.42569009729192975}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9638608500657033, "全帯域_10ms_rms_envelope_nrmse": 0.11634087868022042, "全帯域_10ms_envelope_correlation": 0.9876973091502202, "全帯域_10ms_modulation_mae_db": 0.38969621317919395, "全帯域_40ms_log_rms_mae_db": 0.47945123724453764, "全帯域_40ms_rms_envelope_nrmse": 0.06684038811781672, "全帯域_40ms_envelope_correlation": 0.9956027489349293, "全帯域_40ms_modulation_mae_db": 0.48539291192999795, "300–1000 Hz_10ms_log_rms_mae_db": 1.7501590584284155, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14643086770954664, "300–1000 Hz_10ms_envelope_correlation": 0.9835765324760057, "300–1000 Hz_10ms_modulation_mae_db": 0.571961577741427, "300–1000 Hz_40ms_log_rms_mae_db": 0.9493394075692453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08594511825609098, "300–1000 Hz_40ms_envelope_correlation": 0.9938027977538864, "300–1000 Hz_40ms_modulation_mae_db": 0.10224168243642523, "1000–3000 Hz_10ms_log_rms_mae_db": 1.460858145706105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18033533348392858, "1000–3000 Hz_10ms_envelope_correlation": 0.9702865461523654, "1000–3000 Hz_10ms_modulation_mae_db": 0.5550409932240218, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7436245141849823, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08262839072705376, "1000–3000 Hz_40ms_envelope_correlation": 0.9936376989910305, "1000–3000 Hz_40ms_modulation_mae_db": 0.7515546313173543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5963767046707524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17879259593218214, "3000–6000 Hz_10ms_envelope_correlation": 0.97625931659858, "3000–6000 Hz_10ms_modulation_mae_db": 0.5362815844746004, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9093873670382543, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11647280140939538, "3000–6000 Hz_40ms_envelope_correlation": 0.9891386415054727, "3000–6000 Hz_40ms_modulation_mae_db": 0.3949231214803491, "mr_stft_db": 5.7746311158754535, "band_mae_db": 0.34372755759524337, "rms_gain_error_db": 0.08454653625669424, "output_peak": 0.44591184439577614}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9797507207431827, "全帯域_10ms_rms_envelope_nrmse": 0.11501688195733768, "全帯域_10ms_envelope_correlation": 0.9877240570540969, "全帯域_10ms_modulation_mae_db": 0.3741978526414436, "全帯域_40ms_log_rms_mae_db": 0.5211246283055423, "全帯域_40ms_rms_envelope_nrmse": 0.06529408295355416, "全帯域_40ms_envelope_correlation": 0.9956668745512319, "全帯域_40ms_modulation_mae_db": 0.5275480345561258, "300–1000 Hz_10ms_log_rms_mae_db": 1.7546050716872665, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14723860421595192, "300–1000 Hz_10ms_envelope_correlation": 0.9831497707313602, "300–1000 Hz_10ms_modulation_mae_db": 0.4471976227418905, "300–1000 Hz_40ms_log_rms_mae_db": 0.966647526058281, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10179130503196647, "300–1000 Hz_40ms_envelope_correlation": 0.991216479033701, "300–1000 Hz_40ms_modulation_mae_db": 0.7735831323781142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.531384765038216, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18172333366964893, "1000–3000 Hz_10ms_envelope_correlation": 0.9703129988909538, "1000–3000 Hz_10ms_modulation_mae_db": 0.39785778807833455, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7995439633407269, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09524643205701797, "1000–3000 Hz_40ms_envelope_correlation": 0.9924507242881524, "1000–3000 Hz_40ms_modulation_mae_db": 0.5532253734814754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5848274306432215, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1760535690418768, "3000–6000 Hz_10ms_envelope_correlation": 0.9768927276264843, "3000–6000 Hz_10ms_modulation_mae_db": 0.7853232891711718, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8934707584894789, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11349146264524693, "3000–6000 Hz_40ms_envelope_correlation": 0.9898381910994927, "3000–6000 Hz_40ms_modulation_mae_db": 0.5318997160453728, "mr_stft_db": 5.778720872401617, "band_mae_db": 0.39934138251173756, "rms_gain_error_db": -0.024566003262275322, "output_peak": 0.4019286216950522}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9520754060510022, "全帯域_10ms_rms_envelope_nrmse": 0.11760285220997589, "全帯域_10ms_envelope_correlation": 0.9871491895226977, "全帯域_10ms_modulation_mae_db": 0.7221188046348207, "全帯域_40ms_log_rms_mae_db": 0.5475305691769722, "全帯域_40ms_rms_envelope_nrmse": 0.06046278220192138, "全帯域_40ms_envelope_correlation": 0.9963318451486236, "全帯域_40ms_modulation_mae_db": 0.9061983288610288, "300–1000 Hz_10ms_log_rms_mae_db": 1.6973033792133738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14738037600762394, "300–1000 Hz_10ms_envelope_correlation": 0.9832336864976422, "300–1000 Hz_10ms_modulation_mae_db": 0.7992184288790941, "300–1000 Hz_40ms_log_rms_mae_db": 0.9063022806954165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08397003500483136, "300–1000 Hz_40ms_envelope_correlation": 0.9943598159845943, "300–1000 Hz_40ms_modulation_mae_db": 1.1421496057147487, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4867492564156384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18531069889983337, "1000–3000 Hz_10ms_envelope_correlation": 0.968874794501304, "1000–3000 Hz_10ms_modulation_mae_db": 0.31787180540958926, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7962895435277346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0855540342530303, "1000–3000 Hz_40ms_envelope_correlation": 0.9928790029988791, "1000–3000 Hz_40ms_modulation_mae_db": 0.3659950941143068, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7138703705387597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1738718586970072, "3000–6000 Hz_10ms_envelope_correlation": 0.9784820468572464, "3000–6000 Hz_10ms_modulation_mae_db": 0.5865852890907312, "3000–6000 Hz_40ms_log_rms_mae_db": 1.081244974723017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12531419701655783, "3000–6000 Hz_40ms_envelope_correlation": 0.9880521761971107, "3000–6000 Hz_40ms_modulation_mae_db": 0.3259496519610217, "mr_stft_db": 5.800068343110965, "band_mae_db": 0.25975316320323366, "rms_gain_error_db": -0.0705454364400827, "output_peak": 0.5592162351542744}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9940579257438052, "全帯域_10ms_rms_envelope_nrmse": 0.10663930928263057, "全帯域_10ms_envelope_correlation": 0.9896351380744947, "全帯域_10ms_modulation_mae_db": 0.16880858604085155, "全帯域_40ms_log_rms_mae_db": 0.5238622803301357, "全帯域_40ms_rms_envelope_nrmse": 0.05691947364594579, "全帯域_40ms_envelope_correlation": 0.9968117523979392, "全帯域_40ms_modulation_mae_db": 0.24835065142894477, "300–1000 Hz_10ms_log_rms_mae_db": 1.6795705335890052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12590902438742266, "300–1000 Hz_10ms_envelope_correlation": 0.9877448988049754, "300–1000 Hz_10ms_modulation_mae_db": 0.16609613916640747, "300–1000 Hz_40ms_log_rms_mae_db": 0.9264647491152619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06145263883636035, "300–1000 Hz_40ms_envelope_correlation": 0.9968010308470485, "300–1000 Hz_40ms_modulation_mae_db": 0.4596877372089488, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5624397486940165, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1955923930012047, "1000–3000 Hz_10ms_envelope_correlation": 0.9671789147854856, "1000–3000 Hz_10ms_modulation_mae_db": 0.937518526701968, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9166926759955425, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10667639476682189, "1000–3000 Hz_40ms_envelope_correlation": 0.9926991537252532, "1000–3000 Hz_40ms_modulation_mae_db": 1.1821238180898936, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7007694287535815, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18489192790330766, "3000–6000 Hz_10ms_envelope_correlation": 0.9751599759175346, "3000–6000 Hz_10ms_modulation_mae_db": 0.6315322805948329, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0584189198276968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12934203082066956, "3000–6000 Hz_40ms_envelope_correlation": 0.986940027896531, "3000–6000 Hz_40ms_modulation_mae_db": 0.9201847003819562, "mr_stft_db": 5.805370667034798, "band_mae_db": 0.31887874705181274, "rms_gain_error_db": 0.06659555522581125, "output_peak": 0.39400486742232715}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 40080, "output_samples": 40080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.875356955046423, "spectral_convergence": 0.3044717112667844, "spectral_convergence_status": "ok", "amplitude_floor": 1.4347408585118857e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.624036199582029, "spectral_convergence": 0.3012865323338434, "spectral_convergence_status": "ok", "amplitude_floor": 9.356428468567288e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.861266079646722, "spectral_convergence": 0.3099648912887839, "spectral_convergence_status": "ok", "amplitude_floor": 6.30553407186755e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020558767458207216, "output_power": 0.0002606184180577101, "delta_db": 1.030080300822581, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002656133227213937, "output_power": 0.0025194564433283734, "delta_db": -0.22943000108912096, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000658897280424305, "output_power": 0.0006197870978933752, "delta_db": -0.26575183725700663, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.296049498403337e-05, "output_power": 6.056712307459102e-05, "delta_db": -0.16831188451066267, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018191885630194454, "output_power": 0.0001732854086509913, "delta_db": -0.21115722144788285, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012525230408717123, "output_power": 0.00011808078602347162, "delta_db": -0.2560648826741701, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.360132687966904, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15794175645610378, "pitch": null, "output_rms_dbfs": -24.257608676497306, "output_peak": 0.38075248012536583, "reference_rms_dbfs": -24.099666920041205, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.786886411425058}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 40080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.0001593748762509214, "log_rms_mae_db": 0.9535205909850029, "rms_envelope_nrmse": 0.09054468396892704, "envelope_correlation": 0.9925729797247242, "rms_gain_error_db": -0.15792896183368957, "reference_mean_power": 0.003898496329889168, "output_mean_power": 0.0037592761855365755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.3730898670333144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011160072986886684, "output_power": 0.0010930292748994822, "reference_active": true, "delta_db": -0.09035240964765705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003924601048841299, "output_power": 0.0003711787140328744, "reference_active": true, "delta_db": -0.24212452782884433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.521600297301381e-05, "output_power": 8.699248487019176e-05, "reference_active": true, "delta_db": 0.08960576381925474}], "active_band_count": 3, "active_band_mae_db": 0.14069423376525203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00014354118626608257, "log_rms_mae_db": 0.4832044547238493, "rms_envelope_nrmse": 0.05042921752322914, "envelope_correlation": 0.9977108835509563, "rms_gain_error_db": -0.15791543316025802, "reference_mean_power": 0.00394559944963645, "output_mean_power": 0.0038047090463949012}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 2.0339309551628037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010452646972319804, "output_power": 0.0010292333597974144, "reference_active": true, "delta_db": -0.06712428542720332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020675655639960477, "output_power": 0.0001967975593293234, "reference_active": true, "delta_db": -0.21439582197883597}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.033282667633223e-06, "output_power": 5.900893139597556e-06, "reference_active": true, "delta_db": -0.09635923101526558}], "active_band_count": 3, "active_band_mae_db": 0.12595977947376827}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.00014555177137634006, "log_rms_mae_db": 1.6371056347546455, "rms_envelope_nrmse": 0.10838861838545848, "envelope_correlation": 0.991587726092633, "rms_gain_error_db": -0.30796694484761944, "reference_mean_power": 0.0025948122953778335, "output_mean_power": 0.0024171814496917678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.091667884405576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010721637846984168, "output_power": 0.001027072707092832, "reference_active": true, "delta_db": -0.186599449856425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00045198280410970506, "output_power": 0.00037546166810023166, "reference_active": true, "delta_db": -0.8055630689352631}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.681659745338991e-05, "output_power": 4.925150424107114e-05, "reference_active": true, "delta_db": -0.6205572252077962}], "active_band_count": 3, "active_band_mae_db": 0.5375732479998281}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00013266563413733552, "log_rms_mae_db": 0.9325956310604061, "rms_envelope_nrmse": 0.06708780601027409, "envelope_correlation": 0.9971260300167607, "rms_gain_error_db": -0.30796626669071775, "reference_mean_power": 0.002626297782108314, "output_mean_power": 0.00244651194341625}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 1.815151890487515e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010442677388382842, "output_power": 0.000994404900613925, "reference_active": true, "delta_db": -0.2124860537932185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002292871431701824, "output_power": 0.00019412412380019278, "reference_active": true, "delta_db": -0.7230019468889192}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.238038641621491e-06, "output_power": 4.99381792274483e-06, "reference_active": true, "delta_db": -1.6118819364419816}], "active_band_count": 3, "active_band_mae_db": 0.8491233123747065}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.977119603351369e-05, "log_rms_mae_db": 1.5165172083265683, "rms_envelope_nrmse": 0.20241336716767105, "envelope_correlation": 0.9625835839532476, "rms_gain_error_db": -0.22472356017007392, "reference_mean_power": 0.0001519473982974274, "output_mean_power": 0.0001442849098042644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 8.081548208249742e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5556653827814076e-05, "output_power": 4.462256753233177e-05, "reference_active": true, "delta_db": -0.0899726182335977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1818343242486535e-05, "output_power": 1.1687923540563664e-05, "reference_active": true, "delta_db": -0.04819237236582244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7000790662746577e-06, "output_power": 5.90088184818096e-06, "reference_active": true, "delta_db": 2.027059144398294}], "active_band_count": 3, "active_band_mae_db": 0.7217413783325713}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.109627865595404e-05, "log_rms_mae_db": 0.7964431655133459, "rms_envelope_nrmse": 0.1049016532132692, "envelope_correlation": 0.9892372355340973, "rms_gain_error_db": -0.22482634909562632, "reference_mean_power": 0.0001537098951112271, "output_mean_power": 0.00014595507193258643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 6.838335178235257e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3201238570188224e-05, "output_power": 4.367792539853596e-05, "reference_active": true, "delta_db": 0.0476580407041257}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.064245816738516e-06, "output_power": 6.363387612994566e-06, "reference_active": true, "delta_db": 0.2091158093021139}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0299327724151498e-07, "output_power": 3.1303828777301226e-07, "reference_active": true, "delta_db": 1.8811580430787644}], "active_band_count": 3, "active_band_mae_db": 0.7126439643616681}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.882145378793849e-05, "log_rms_mae_db": 1.5643270261912978, "rms_envelope_nrmse": 0.18474912270555172, "envelope_correlation": 0.9746140182700935, "rms_gain_error_db": -0.2333603652152319, "reference_mean_power": 0.00013710154501992013, "output_mean_power": 0.00012992906477736122}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 1.0763544842397476e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5103901227352814e-05, "output_power": 4.272558474426843e-05, "reference_active": true, "delta_db": -0.23526092263601017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.11132679073469e-05, "output_power": 2.507630007120296e-05, "reference_active": true, "delta_db": -0.9368217047420944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.435692144349255e-06, "output_power": 5.068263177613775e-06, "reference_active": true, "delta_db": -0.3039569384386833}], "active_band_count": 3, "active_band_mae_db": 0.492013188605596}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.3950747306848176e-05, "log_rms_mae_db": 0.9342736079783915, "rms_envelope_nrmse": 0.1263907885331295, "envelope_correlation": 0.987464156260105, "rms_gain_error_db": -0.23291202372194125, "reference_mean_power": 0.00013864857515666164, "output_mean_power": 0.0001314087270934253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 9.21368112936246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.381631867120006e-05, "output_power": 4.209747069741794e-05, "reference_active": true, "delta_db": -0.17379883193330803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.911163628753255e-05, "output_power": 1.5007863882791681e-05, "reference_active": true, "delta_db": -1.049789896657448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5242776229186804e-07, "output_power": 3.6207391908924296e-07, "reference_active": true, "delta_db": 0.11727130458240963}], "active_band_count": 3, "active_band_mae_db": 0.4469533443910552}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8758678798109187, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a9132e5cc3d2b818a8493006629dace4c511bb4c80adef43b23022e261dd7797", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9687536784757698, "standard_deviation": 0.0028790298439467852, "minimum": 0.9625835839532476, "maximum": 0.9713215624624876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4911270959416565, "standard_deviation": 0.05583821196518869, "minimum": 1.4018430031928908, "maximum": 1.5624397486940165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5576376641383741, "standard_deviation": 0.24223014685519323, "minimum": 0.19559431751322423, "maximum": 0.937518526701968, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18702556389120822, "standard_deviation": 0.00800717978194449, "minimum": 0.17951983246089215, "maximum": 0.20241336716767105, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9920584315747374, "standard_deviation": 0.0016592365959915876, "minimum": 0.9892372355340973, "maximum": 0.9937425741063025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8147606482969311, "standard_deviation": 0.0627752397429919, "minimum": 0.7436245141849823, "maximum": 0.9166926759955425, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7173104819694622, "standard_deviation": 0.33380341703215727, "minimum": 0.20863165302593714, "maximum": 1.1821238180898936, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0971889343001823, "standard_deviation": 0.010692668018907888, "minimum": 0.08262839072705376, "maximum": 0.11044249198876799, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.976047310924749, "standard_deviation": 0.00147613749339695, "minimum": 0.9740873981655385, "maximum": 0.9784820468572464, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6430992845081127, "standard_deviation": 0.05530742450070605, "minimum": 1.5643270261912978, "maximum": 1.7138703705387597, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5912080310603933, "standard_deviation": 0.1498092966398604, "minimum": 0.3290059864912451, "maximum": 0.7853232891711718, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18097424653570782, "standard_deviation": 0.005363671323406447, "minimum": 0.1738718586970072, "maximum": 0.18803577872614535, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9882639859795257, "standard_deviation": 0.0009264753442996185, "minimum": 0.986940027896531, "maximum": 0.9898381910994927, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9773162465726695, "standard_deviation": 0.07246665855969936, "minimum": 0.8934707584894789, "maximum": 1.081244974723017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5467920333422567, "standard_deviation": 0.23057278003208678, "minimum": 0.3259496519610217, "maximum": 0.9201847003819562, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1224617884926844, "standard_deviation": 0.005516622650977643, "minimum": 0.11349146264524693, "maximum": 0.12934203082066956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9865231447025833, "standard_deviation": 0.003792798232746897, "minimum": 0.9831160207692372, "maximum": 0.9919395216989707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7061203077464846, "standard_deviation": 0.044717279506926055, "minimum": 1.6371056347546455, "maximum": 1.7546050716872665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5014617070738657, "standard_deviation": 0.2490062895853863, "minimum": 0.16609613916640747, "maximum": 0.8627448799337573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13156404735330857, "standard_deviation": 0.018576629855710115, "minimum": 0.10244148708387822, "maximum": 0.14771397156015004, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9950874483722019, "standard_deviation": 0.002101858943006974, "minimum": 0.991216479033701, "maximum": 0.9974459350478835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9516859864699195, "standard_deviation": 0.04245115073349028, "minimum": 0.9063022806954165, "maximum": 1.0472894409054827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6540856479338939, "standard_deviation": 0.32910012055426874, "minimum": 0.10224168243642523, "maximum": 1.1421496057147487, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07711184308386942, "standard_deviation": 0.014969050524667479, "minimum": 0.056286876881892, "maximum": 0.10179130503196647, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.35081710675911765, "standard_deviation": 0.05980670942480296, "minimum": 0.25975316320323366, "maximum": 0.4496351512354309, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.794596060392969, "standard_deviation": 0.014127104327135242, "minimum": 5.7746311158754535, "maximum": 5.81087046129392, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.43253514285850825, "standard_deviation": 0.05808069832500541, "minimum": 0.38075248012536583, "maximum": 0.5592162351542744, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03307007099178738, "standard_deviation": 0.08909698573972011, "minimum": -0.15794175645610378, "maximum": 0.08454653625669424, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9898570059278297, "standard_deviation": 0.0023588404731482424, "minimum": 0.9871491895226977, "maximum": 0.9931423536910411, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9642010434132049, "standard_deviation": 0.030807869207391227, "minimum": 0.9041422425365068, "maximum": 1.0046410238545433, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3770420687247057, "standard_deviation": 0.19006973819079576, "minimum": 0.14069423376525203, "maximum": 0.7221188046348207, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10439452242109981, "standard_deviation": 0.012321925294401901, "minimum": 0.08626112585497622, "maximum": 0.11760285220997589, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968775126103715, "standard_deviation": 0.0009511165891461643, "minimum": 0.9956027489349293, "maximum": 0.9979696690149579, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5122008737859975, "standard_deviation": 0.03687156276534782, "minimum": 0.46473505220900285, "maximum": 0.5739645829412711, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4312674916280484, "standard_deviation": 0.24822734441481453, "minimum": 0.12595977947376827, "maximum": 0.9061983288610288, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055869499393347344, "standard_deviation": 0.00801504719331147, "minimum": 0.044929461121290426, "maximum": 0.06684038811781672, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0038", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.137228012084961, "predicted_mos": 2.137228012084961, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "97df57d22c49411b3eb4b8f9b9c673867ce480688e8d9599648aff497bdce7db", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.933474799618125, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9447734787646227, "全帯域_10ms_rms_envelope_nrmse": 0.10513102502811372, "全帯域_10ms_envelope_correlation": 0.9897308023214231, "全帯域_10ms_modulation_mae_db": 0.1658937617206494, "全帯域_40ms_log_rms_mae_db": 0.4753386749549219, "全帯域_40ms_rms_envelope_nrmse": 0.055798584268347357, "全帯域_40ms_envelope_correlation": 0.9968445808675301, "全帯域_40ms_modulation_mae_db": 0.14049558419449384, "300–1000 Hz_10ms_log_rms_mae_db": 1.89739763000243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1280885792508515, "300–1000 Hz_10ms_envelope_correlation": 0.9874298164721235, "300–1000 Hz_10ms_modulation_mae_db": 0.5012993144230385, "300–1000 Hz_40ms_log_rms_mae_db": 1.0682243184673308, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06674141893785417, "300–1000 Hz_40ms_envelope_correlation": 0.9966129576160185, "300–1000 Hz_40ms_modulation_mae_db": 0.22693381739638505, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6887383317272946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20866458561595266, "1000–3000 Hz_10ms_envelope_correlation": 0.9615146362635535, "1000–3000 Hz_10ms_modulation_mae_db": 0.3099717473809092, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9819023408760469, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11606015856809121, "1000–3000 Hz_40ms_envelope_correlation": 0.9871007288669109, "1000–3000 Hz_40ms_modulation_mae_db": 0.1300138479706061, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2942704724630323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22170681046709975, "3000–6000 Hz_10ms_envelope_correlation": 0.966275843049108, "3000–6000 Hz_10ms_modulation_mae_db": 0.46019507179157665, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6119321306144574, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15661268142745213, "3000–6000 Hz_40ms_envelope_correlation": 0.9827255507072018, "3000–6000 Hz_40ms_modulation_mae_db": 0.8188419687052573, "mr_stft_db": 6.218563943755605, "band_mae_db": 0.44161673071575286, "rms_gain_error_db": -0.026138094525600883, "output_peak": 0.43370876484813903}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0093851841950874, "全帯域_10ms_rms_envelope_nrmse": 0.100892240129625, "全帯域_10ms_envelope_correlation": 0.9906244115001162, "全帯域_10ms_modulation_mae_db": 0.22544324346098776, "全帯域_40ms_log_rms_mae_db": 0.4905849675163826, "全帯域_40ms_rms_envelope_nrmse": 0.05939260645851494, "全帯域_40ms_envelope_correlation": 0.9964439879658261, "全帯域_40ms_modulation_mae_db": 0.11443142039506377, "300–1000 Hz_10ms_log_rms_mae_db": 1.8088526657708048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12788920433246534, "300–1000 Hz_10ms_envelope_correlation": 0.9874450411192567, "300–1000 Hz_10ms_modulation_mae_db": 0.28094378798777625, "300–1000 Hz_40ms_log_rms_mae_db": 1.0381433122058883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08385281477955182, "300–1000 Hz_40ms_envelope_correlation": 0.994316527785042, "300–1000 Hz_40ms_modulation_mae_db": 0.43493719426934535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8229499171429655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2321176145530152, "1000–3000 Hz_10ms_envelope_correlation": 0.9509315504073812, "1000–3000 Hz_10ms_modulation_mae_db": 0.5142891374343034, "1000–3000 Hz_40ms_log_rms_mae_db": 1.188877088829366, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14689221710997324, "1000–3000 Hz_40ms_envelope_correlation": 0.9781389406150586, "1000–3000 Hz_40ms_modulation_mae_db": 0.506137891770546, "3000–6000 Hz_10ms_log_rms_mae_db": 2.325250042920493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2486971630420767, "3000–6000 Hz_10ms_envelope_correlation": 0.9549980572642287, "3000–6000 Hz_10ms_modulation_mae_db": 0.58665993699517, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6167802082915472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17083588594276958, "3000–6000 Hz_40ms_envelope_correlation": 0.9776280120741251, "3000–6000 Hz_40ms_modulation_mae_db": 0.5696576958305525, "mr_stft_db": 6.17647272143914, "band_mae_db": 0.29095123151866614, "rms_gain_error_db": 0.02968308326863713, "output_peak": 0.4389800682308165}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9635802180377616, "全帯域_10ms_rms_envelope_nrmse": 0.08841365574786593, "全帯域_10ms_envelope_correlation": 0.9927508133228676, "全帯域_10ms_modulation_mae_db": 0.340276794934671, "全帯域_40ms_log_rms_mae_db": 0.5141665342109404, "全帯域_40ms_rms_envelope_nrmse": 0.042634132222019894, "全帯域_40ms_envelope_correlation": 0.9981600928673546, "全帯域_40ms_modulation_mae_db": 0.38704637852057394, "300–1000 Hz_10ms_log_rms_mae_db": 1.8991905542030545, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10666115117007847, "300–1000 Hz_10ms_envelope_correlation": 0.9913069027242699, "300–1000 Hz_10ms_modulation_mae_db": 0.5499231378877439, "300–1000 Hz_40ms_log_rms_mae_db": 1.1327963292902414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057721358368943806, "300–1000 Hz_40ms_envelope_correlation": 0.99738052596692, "300–1000 Hz_40ms_modulation_mae_db": 0.8254926490052434, "1000–3000 Hz_10ms_log_rms_mae_db": 1.725037735731089, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22099476617250227, "1000–3000 Hz_10ms_envelope_correlation": 0.9573811037113837, "1000–3000 Hz_10ms_modulation_mae_db": 0.3230162282629983, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9860376415385071, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.130133715409839, "1000–3000 Hz_40ms_envelope_correlation": 0.9839712015171795, "1000–3000 Hz_40ms_modulation_mae_db": 0.3419055627426304, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3332096526188075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23894104574917208, "3000–6000 Hz_10ms_envelope_correlation": 0.9580691094023167, "3000–6000 Hz_10ms_modulation_mae_db": 0.5472364330129731, "3000–6000 Hz_40ms_log_rms_mae_db": 1.636087993825493, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1617380107551084, "3000–6000 Hz_40ms_envelope_correlation": 0.9795563922150452, "3000–6000 Hz_40ms_modulation_mae_db": 0.9319887959984653, "mr_stft_db": 6.200549869475598, "band_mae_db": 0.3355366783253149, "rms_gain_error_db": -0.013101483896203955, "output_peak": 0.37990685588800877}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9693536942726929, "全帯域_10ms_rms_envelope_nrmse": 0.10584458665797876, "全帯域_10ms_envelope_correlation": 0.9899797393050556, "全帯域_10ms_modulation_mae_db": 0.47260950443124505, "全帯域_40ms_log_rms_mae_db": 0.5368205901988142, "全帯域_40ms_rms_envelope_nrmse": 0.06574141980514465, "全帯域_40ms_envelope_correlation": 0.9958792114333229, "全帯域_40ms_modulation_mae_db": 0.7753473017705627, "300–1000 Hz_10ms_log_rms_mae_db": 1.9230485440704532, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13879083985465293, "300–1000 Hz_10ms_envelope_correlation": 0.9851312342452483, "300–1000 Hz_10ms_modulation_mae_db": 0.3351949759467747, "300–1000 Hz_40ms_log_rms_mae_db": 1.1354556221203673, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08959035405602753, "300–1000 Hz_40ms_envelope_correlation": 0.9932112540901149, "300–1000 Hz_40ms_modulation_mae_db": 0.07041416496025814, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8149213811561744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.267169859418737, "1000–3000 Hz_10ms_envelope_correlation": 0.938162105912066, "1000–3000 Hz_10ms_modulation_mae_db": 0.40003223923846987, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0475753013531026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17802334062446304, "1000–3000 Hz_40ms_envelope_correlation": 0.969907764654, "1000–3000 Hz_40ms_modulation_mae_db": 0.41012218124354244, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2758032425644186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23828446963564942, "3000–6000 Hz_10ms_envelope_correlation": 0.9590475176889812, "3000–6000 Hz_10ms_modulation_mae_db": 0.3499061198162207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.603166730825888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16669073425075376, "3000–6000 Hz_40ms_envelope_correlation": 0.9786653007804055, "3000–6000 Hz_40ms_modulation_mae_db": 1.0451384663670842, "mr_stft_db": 6.193491534397489, "band_mae_db": 0.40694646185256095, "rms_gain_error_db": 0.11959253785865054, "output_peak": 0.4027163456230203}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.955373648773673, "全帯域_10ms_rms_envelope_nrmse": 0.0971034372299311, "全帯域_10ms_envelope_correlation": 0.9912405292901391, "全帯域_10ms_modulation_mae_db": 0.1906704044732935, "全帯域_40ms_log_rms_mae_db": 0.49742411271490244, "全帯域_40ms_rms_envelope_nrmse": 0.05354035629096612, "全帯域_40ms_envelope_correlation": 0.9970974455534368, "全帯域_40ms_modulation_mae_db": 0.08494685941581154, "300–1000 Hz_10ms_log_rms_mae_db": 1.9233770224491675, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12617673663917373, "300–1000 Hz_10ms_envelope_correlation": 0.9882704132827979, "300–1000 Hz_10ms_modulation_mae_db": 0.7079175706785305, "300–1000 Hz_40ms_log_rms_mae_db": 1.0662170872006038, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07174319942997726, "300–1000 Hz_40ms_envelope_correlation": 0.996601819680318, "300–1000 Hz_40ms_modulation_mae_db": 1.0820206822679515, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7472860547505011, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21756083951700117, "1000–3000 Hz_10ms_envelope_correlation": 0.9587383954127031, "1000–3000 Hz_10ms_modulation_mae_db": 0.3472373880476456, "1000–3000 Hz_40ms_log_rms_mae_db": 1.079031035762538, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14648258790589014, "1000–3000 Hz_40ms_envelope_correlation": 0.9795567356230649, "1000–3000 Hz_40ms_modulation_mae_db": 0.3520888793151549, "3000–6000 Hz_10ms_log_rms_mae_db": 2.349952240013707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23200148166257495, "3000–6000 Hz_10ms_envelope_correlation": 0.9620621834888029, "3000–6000 Hz_10ms_modulation_mae_db": 0.3189805011185189, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5995621200690988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1543118631382957, "3000–6000 Hz_40ms_envelope_correlation": 0.9825014504327716, "3000–6000 Hz_40ms_modulation_mae_db": 0.6549861970535753, "mr_stft_db": 6.194093874058375, "band_mae_db": 0.3776186930809102, "rms_gain_error_db": -0.04294773517881136, "output_peak": 0.3939923907170641}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8741770063241998, "全帯域_10ms_rms_envelope_nrmse": 0.11668736900788056, "全帯域_10ms_envelope_correlation": 0.9875057192726466, "全帯域_10ms_modulation_mae_db": 0.5835509012282142, "全帯域_40ms_log_rms_mae_db": 0.49188698301290307, "全帯域_40ms_rms_envelope_nrmse": 0.06940965486016298, "全帯域_40ms_envelope_correlation": 0.9951891631766734, "全帯域_40ms_modulation_mae_db": 0.8014156046513863, "300–1000 Hz_10ms_log_rms_mae_db": 1.875891481068414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13577834188489496, "300–1000 Hz_10ms_envelope_correlation": 0.9856737254515672, "300–1000 Hz_10ms_modulation_mae_db": 0.7074271375410416, "300–1000 Hz_40ms_log_rms_mae_db": 1.1094610789742045, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07714591957919484, "300–1000 Hz_40ms_envelope_correlation": 0.9949798692855027, "300–1000 Hz_40ms_modulation_mae_db": 0.9476921186055005, "1000–3000 Hz_10ms_log_rms_mae_db": 1.612562060948776, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2519123459121095, "1000–3000 Hz_10ms_envelope_correlation": 0.9493614685496472, "1000–3000 Hz_10ms_modulation_mae_db": 1.1670825075917992, "1000–3000 Hz_40ms_log_rms_mae_db": 0.951754093472076, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17188692450922036, "1000–3000 Hz_40ms_envelope_correlation": 0.9756167041501747, "1000–3000 Hz_40ms_modulation_mae_db": 1.503438049042799, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3381354443119147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2430812916203248, "3000–6000 Hz_10ms_envelope_correlation": 0.9576314995611488, "3000–6000 Hz_10ms_modulation_mae_db": 0.43524861240934754, "3000–6000 Hz_40ms_log_rms_mae_db": 1.631016964249572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17156045699804315, "3000–6000 Hz_40ms_envelope_correlation": 0.9776066310367535, "3000–6000 Hz_40ms_modulation_mae_db": 0.7744580202646022, "mr_stft_db": 6.1840854127830776, "band_mae_db": 0.5913881428206352, "rms_gain_error_db": 0.047710234257964695, "output_peak": 0.45125862295375685}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9925045624567116, "全帯域_10ms_rms_envelope_nrmse": 0.10170754854078681, "全帯域_10ms_envelope_correlation": 0.9904171001446018, "全帯域_10ms_modulation_mae_db": 0.418216761707864, "全帯域_40ms_log_rms_mae_db": 0.49883471017574404, "全帯域_40ms_rms_envelope_nrmse": 0.05534815756697653, "全帯域_40ms_envelope_correlation": 0.9968861499309081, "全帯域_40ms_modulation_mae_db": 0.5088386362084641, "300–1000 Hz_10ms_log_rms_mae_db": 1.8291427161335578, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12654981967957807, "300–1000 Hz_10ms_envelope_correlation": 0.9878977405456385, "300–1000 Hz_10ms_modulation_mae_db": 0.9105902521734408, "300–1000 Hz_40ms_log_rms_mae_db": 1.0110641715419744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07544715321989291, "300–1000 Hz_40ms_envelope_correlation": 0.9957343822834996, "300–1000 Hz_40ms_modulation_mae_db": 1.2451786082945981, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8041564953431262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2351405367725123, "1000–3000 Hz_10ms_envelope_correlation": 0.9511291545087552, "1000–3000 Hz_10ms_modulation_mae_db": 0.5799844259731489, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1739508099702307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14973808509836853, "1000–3000 Hz_40ms_envelope_correlation": 0.9782405429389137, "1000–3000 Hz_40ms_modulation_mae_db": 0.5820238271276231, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3225902238276808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2227569779239005, "3000–6000 Hz_10ms_envelope_correlation": 0.9634238791420059, "3000–6000 Hz_10ms_modulation_mae_db": 1.0652154961540006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6254602390005806, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15913290313601752, "3000–6000 Hz_40ms_envelope_correlation": 0.9801616172996391, "3000–6000 Hz_40ms_modulation_mae_db": 0.7406064687087204, "mr_stft_db": 6.175270906245989, "band_mae_db": 0.5444166334589221, "rms_gain_error_db": -0.0007225571631212055, "output_peak": 0.4172355644603508}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9972102387052418, "全帯域_10ms_rms_envelope_nrmse": 0.10300688424495481, "全帯域_10ms_envelope_correlation": 0.9902848595068245, "全帯域_10ms_modulation_mae_db": 0.3800406957221097, "全帯域_40ms_log_rms_mae_db": 0.5063518404034986, "全帯域_40ms_rms_envelope_nrmse": 0.06779785759680554, "全帯域_40ms_envelope_correlation": 0.9954080076006233, "全帯域_40ms_modulation_mae_db": 0.3489241453747123, "300–1000 Hz_10ms_log_rms_mae_db": 1.9071231840717142, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13538996092098138, "300–1000 Hz_10ms_envelope_correlation": 0.9859315610345774, "300–1000 Hz_10ms_modulation_mae_db": 0.9837725605610385, "300–1000 Hz_40ms_log_rms_mae_db": 1.1632027385750197, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08336771624828032, "300–1000 Hz_40ms_envelope_correlation": 0.9944641549465995, "300–1000 Hz_40ms_modulation_mae_db": 1.140795530325973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7991326010758686, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.27291698587609925, "1000–3000 Hz_10ms_envelope_correlation": 0.9400903590705497, "1000–3000 Hz_10ms_modulation_mae_db": 0.8813639278644819, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1044169522856486, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18791857930468886, "1000–3000 Hz_40ms_envelope_correlation": 0.9701102400245379, "1000–3000 Hz_40ms_modulation_mae_db": 0.7827636818133105, "3000–6000 Hz_10ms_log_rms_mae_db": 2.400657186185733, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22912788591098127, "3000–6000 Hz_10ms_envelope_correlation": 0.9612362224318736, "3000–6000 Hz_10ms_modulation_mae_db": 0.5642276012894419, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6226086091397351, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16279169810135233, "3000–6000 Hz_40ms_envelope_correlation": 0.9791551823062712, "3000–6000 Hz_40ms_modulation_mae_db": 0.42489640752992397, "mr_stft_db": 6.214761762656518, "band_mae_db": 0.42173989544944623, "rms_gain_error_db": 0.06011481077506511, "output_peak": 0.4276612786126575}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 40080, "output_samples": 40080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.339342551226097, "spectral_convergence": 0.33448246403137377, "spectral_convergence_status": "ok", "amplitude_floor": 1.4347408585118857e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.133020800203831, "spectral_convergence": 0.3221779942567439, "spectral_convergence_status": "ok", "amplitude_floor": 9.356428468567288e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.183328479836884, "spectral_convergence": 0.3236810891985162, "spectral_convergence_status": "ok", "amplitude_floor": 6.30553407186755e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020558767458207216, "output_power": 0.0002786554152757785, "delta_db": 1.3207041314949717, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002656133227213937, "output_power": 0.002562867924082897, "delta_db": -0.15523629083570267, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000658897280424305, "output_power": 0.0006338227018906138, "delta_db": -0.1684992457734844, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.296049498403337e-05, "output_power": 7.334859713330216e-05, "delta_db": 0.6632367822485074, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018191885630194454, "output_power": 0.00018657720295125926, "delta_db": 0.10980861112903738, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012525230408717123, "output_power": 0.00013213177099552148, "delta_db": 0.2322153228128137, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.44161673071575286, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.026138094525600883, "pitch": null, "output_rms_dbfs": -24.125805014566804, "output_peak": 0.43370876484813903, "reference_rms_dbfs": -24.099666920041205, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.218563943755605}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 40080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.0001593748762509214, "log_rms_mae_db": 0.9447734787646227, "rms_envelope_nrmse": 0.10513102502811372, "envelope_correlation": 0.9897308023214231, "rms_gain_error_db": -0.026134874606267705, "reference_mean_power": 0.003898496329889168, "output_mean_power": 0.003875106495290825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.3730898670333144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011160072986886684, "output_power": 0.0010869807808286578, "reference_active": true, "delta_db": -0.11445169609219663}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003924601048841299, "output_power": 0.0003771438371691104, "reference_active": true, "delta_db": -0.17288500187932263}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.521600297301381e-05, "output_power": 8.944490736740576e-05, "reference_active": true, "delta_db": 0.2103445871904289}], "active_band_count": 3, "active_band_mae_db": 0.1658937617206494}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00014354118626608257, "log_rms_mae_db": 0.4753386749549219, "rms_envelope_nrmse": 0.055798584268347357, "envelope_correlation": 0.9968445808675301, "rms_gain_error_db": -0.0261458389270228, "reference_mean_power": 0.00394559944963645, "output_mean_power": 0.003921917108739545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 2.0339309551628037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010452646972319804, "output_power": 0.0010255993882943505, "reference_active": true, "delta_db": -0.08248529637059182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020675655639960477, "output_power": 0.00019763410268149026, "reference_active": true, "delta_db": -0.1959740398834162}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.033282667633223e-06, "output_power": 6.235286520292963e-06, "reference_active": true, "delta_db": 0.14302741632947358}], "active_band_count": 3, "active_band_mae_db": 0.14049558419449384}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.00014555177137634006, "log_rms_mae_db": 1.89739763000243, "rms_envelope_nrmse": 0.1280885792508515, "envelope_correlation": 0.9874298164721235, "rms_gain_error_db": -0.19431741672494424, "reference_mean_power": 0.0025948122953778335, "output_mean_power": 0.0024812710327883395}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.091667884405576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010721637846984168, "output_power": 0.0010313255437733244, "reference_active": true, "delta_db": -0.16865359190504056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00045198280410970506, "output_power": 0.0003967464682889139, "reference_active": true, "delta_db": -0.5660884270310097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.681659745338991e-05, "output_power": 6.782515801477558e-05, "reference_active": true, "delta_db": 0.7691559243330655}], "active_band_count": 3, "active_band_mae_db": 0.5012993144230385}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00013266563413733552, "log_rms_mae_db": 1.0682243184673308, "rms_envelope_nrmse": 0.06674141893785417, "envelope_correlation": 0.9966129576160185, "rms_gain_error_db": -0.19432991012883644, "reference_mean_power": 0.002626297782108314, "output_mean_power": 0.002511371583831454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 1.815151890487515e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010442677388382842, "output_power": 0.0010043714344990426, "reference_active": true, "delta_db": -0.1691750893050411}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002292871431701824, "output_power": 0.00020572865912550653, "reference_active": true, "delta_db": -0.47084907694078293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.238038641621491e-06, "output_power": 7.306318932448372e-06, "reference_active": true, "delta_db": 0.040777285943331035}], "active_band_count": 3, "active_band_mae_db": 0.22693381739638505}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.977119603351369e-05, "log_rms_mae_db": 1.6887383317272946, "rms_envelope_nrmse": 0.20866458561595266, "envelope_correlation": 0.9615146362635535, "rms_gain_error_db": 0.19274350443652477, "reference_mean_power": 0.0001519473982974274, "output_mean_power": 0.00015884283100915282}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 8.081548208249742e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5556653827814076e-05, "output_power": 4.4319971145744246e-05, "reference_active": true, "delta_db": -0.11952348330387363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1818343242486535e-05, "output_power": 1.2318099647109713e-05, "reference_active": true, "delta_db": 0.17987113860527104}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7000790662746577e-06, "output_power": 4.278219536005248e-06, "reference_active": true, "delta_db": 0.6305206202335829}], "active_band_count": 3, "active_band_mae_db": 0.3099717473809092}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.109627865595404e-05, "log_rms_mae_db": 0.9819023408760469, "rms_envelope_nrmse": 0.11606015856809121, "envelope_correlation": 0.9871007288669109, "rms_gain_error_db": 0.19314772884277953, "reference_mean_power": 0.0001537098951112271, "output_mean_power": 0.00016070026728130161}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 6.838335178235257e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3201238570188224e-05, "output_power": 4.293710863680215e-05, "reference_active": true, "delta_db": -0.02663402208571712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.064245816738516e-06, "output_power": 6.238366799168019e-06, "reference_active": true, "delta_db": 0.1229410922872298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0299327724151498e-07, "output_power": 2.1454989092735004e-07, "reference_active": true, "delta_db": 0.24046642953887143}], "active_band_count": 3, "active_band_mae_db": 0.1300138479706061}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.882145378793849e-05, "log_rms_mae_db": 2.2942704724630323, "rms_envelope_nrmse": 0.22170681046709975, "envelope_correlation": 0.966275843049108, "rms_gain_error_db": 0.35313588242464083, "reference_mean_power": 0.00013710154501992013, "output_mean_power": 0.0001487153978193756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 1.0763544842397476e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5103901227352814e-05, "output_power": 4.6196197636326856e-05, "reference_active": true, "delta_db": 0.1039212320394238}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.11132679073469e-05, "output_power": 3.016333523788191e-05, "reference_active": true, "delta_db": -0.13466267666565485}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.435692144349255e-06, "output_power": 7.070579060349486e-06, "reference_active": true, "delta_db": 1.1420013066696513}], "active_band_count": 3, "active_band_mae_db": 0.46019507179157665}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.3950747306848176e-05, "log_rms_mae_db": 1.6119321306144574, "rms_envelope_nrmse": 0.15661268142745213, "envelope_correlation": 0.9827255507072018, "rms_gain_error_db": 0.35324645508622476, "reference_mean_power": 0.00013864857515666164, "output_mean_power": 0.00015039730577492546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 9.21368112936246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.381631867120006e-05, "output_power": 4.554440660431087e-05, "reference_active": true, "delta_db": 0.167991615403635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.911163628753255e-05, "output_power": 1.859714800116944e-05, "reference_active": true, "delta_db": -0.11851524468012706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5242776229186804e-07, "output_power": 5.80860730904452e-07, "reference_active": true, "delta_db": 2.1700190460320097}], "active_band_count": 3, "active_band_mae_db": 0.8188419687052573}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8359419866042311, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "024f5d2f434abe1cf0a5476af7816ccd83362842f4ddf1d46d120e8f2c075d41", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.950913596729505, "standard_deviation": 0.00843154824845864, "minimum": 0.938162105912066, "maximum": 0.9615146362635535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7518480722344743, "standard_deviation": 0.07380355283482708, "minimum": 1.612562060948776, "maximum": 1.8229499171429655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5653722002242195, "standard_deviation": 0.3078995591827949, "minimum": 0.3099717473809092, "maximum": 1.1670825075917992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23830969172974115, "standard_deviation": 0.023522467929852884, "minimum": 0.20866458561595266, "maximum": 0.27291698587609925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.97783035729873, "standard_deviation": 0.006025677910556536, "minimum": 0.969907764654, "maximum": 0.9871007288669109, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0641931580109394, "standard_deviation": 0.08882543131981058, "minimum": 0.951754093472076, "maximum": 1.188877088829366, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5760617401282766, "standard_deviation": 0.42074169530428435, "minimum": 0.1300138479706061, "maximum": 1.503438049042799, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1533919510663168, "standard_deviation": 0.024433005446409924, "minimum": 0.11606015856809121, "maximum": 0.18791857930468886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9603430390035582, "standard_deviation": 0.0036097760407134026, "minimum": 0.9549980572642287, "maximum": 0.966275843049108, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.329983563113223, "standard_deviation": 0.037337016358133725, "minimum": 2.2758032425644186, "maximum": 2.400657186185733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5409587215734062, "standard_deviation": 0.23329547130161538, "minimum": 0.3189805011185189, "maximum": 1.0652154961540006, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23432464075147244, "standard_deviation": 0.00960404352886068, "minimum": 0.22170681046709975, "maximum": 0.2486971630420767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9797500171065265, "standard_deviation": 0.001973341546879069, "minimum": 0.9776066310367535, "maximum": 0.9827255507072018, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6183268745020465, "standard_deviation": 0.012933592854951943, "minimum": 1.5995621200690988, "maximum": 1.636087993825493, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7450717525572726, "standard_deviation": 0.19737610953878598, "minimum": 0.42489640752992397, "maximum": 1.0451384663670842, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16295927921872408, "standard_deviation": 0.006337674108540476, "minimum": 0.1543118631382957, "maximum": 0.17156045699804315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9873858043594349, "standard_deviation": 0.0019476797289877685, "minimum": 0.9851312342452483, "maximum": 0.9913069027242699, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8830029747211996, "standard_deviation": 0.04265226304203034, "minimum": 1.8088526657708048, "maximum": 1.9233770224491675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6221335921499231, "standard_deviation": 0.25275154156779345, "minimum": 0.28094378798777625, "maximum": 0.9837725605610385, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12816557921658456, "standard_deviation": 0.009943633436522737, "minimum": 0.10666115117007847, "maximum": 0.13879083985465293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954126864567518, "standard_deviation": 0.0014124689756291183, "minimum": 0.9932112540901149, "maximum": 0.99738052596692, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0905705822969538, "standard_deviation": 0.052904628112829144, "minimum": 1.0110641715419744, "maximum": 1.1632027385750197, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7466830956406569, "standard_deviation": 0.4452212520903657, "minimum": 0.07041416496025814, "maximum": 1.2451786082945981, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07570124182746533, "standard_deviation": 0.010278290574786146, "minimum": 0.057721358368943806, "maximum": 0.08959035405602753, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.42627680840277604, "standard_deviation": 0.10054332773427747, "minimum": 0.29095123151866614, "maximum": 0.5913881428206352, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.194661253101474, "standard_deviation": 0.016170483260532147, "minimum": 6.175270906245989, "maximum": 6.218563943755605, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41818248641672673, "standard_deviation": 0.024342102959761403, "minimum": 0.37990685588800877, "maximum": 0.45125862295375685, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": 0.021773849424572507, "standard_deviation": 0.05342630736781371, "minimum": -0.04294773517881136, "maximum": 0.11959253785865054, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9903167468329593, "standard_deviation": 0.0014756393205721634, "minimum": 0.9875057192726466, "maximum": 0.9927508133228676, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9632947539412489, "standard_deviation": 0.04227857868698512, "minimum": 0.8741770063241998, "maximum": 1.0093851841950874, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3470877584598793, "standard_deviation": 0.14636411938957314, "minimum": 0.1658937617206494, "maximum": 0.5835509012282142, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10234834332339209, "standard_deviation": 0.008019830608037331, "minimum": 0.08841365574786593, "maximum": 0.11668736900788056, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9964885799244594, "standard_deviation": 0.000976795803061009, "minimum": 0.9951891631766734, "maximum": 0.9981600928673546, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5014260516485134, "standard_deviation": 0.01832703457493479, "minimum": 0.4753386749549219, "maximum": 0.5368205901988142, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.39518074131638353, "standard_deviation": 0.28400379962472216, "minimum": 0.08494685941581154, "maximum": 0.8014156046513863, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05870784613361725, "standard_deviation": 0.008879475658322023, "minimum": 0.042634132222019894, "maximum": 0.06940965486016298, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0038", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.4659316539764404, "predicted_mos": 2.4659316539764404, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "ac4eaeecab843748bea4545bfa286e4b04be4a1cdc60f1364486d65fc50b0016", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8628002000041306, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9320150359409302, "全帯域_10ms_rms_envelope_nrmse": 0.10215665026523217, "全帯域_10ms_envelope_correlation": 0.9903373888715449, "全帯域_10ms_modulation_mae_db": 0.37311998558978826, "全帯域_40ms_log_rms_mae_db": 0.5220720920402584, "全帯域_40ms_rms_envelope_nrmse": 0.061584941804164225, "全帯域_40ms_envelope_correlation": 0.9961725635889591, "全帯域_40ms_modulation_mae_db": 0.6455067131011957, "300–1000 Hz_10ms_log_rms_mae_db": 1.5594988149239297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1411079345432263, "300–1000 Hz_10ms_envelope_correlation": 0.9847123069168086, "300–1000 Hz_10ms_modulation_mae_db": 0.6319654215028346, "300–1000 Hz_40ms_log_rms_mae_db": 0.9453777074747345, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08803362701104542, "300–1000 Hz_40ms_envelope_correlation": 0.9938015706440676, "300–1000 Hz_40ms_modulation_mae_db": 0.4101467338322586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3110129905663104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18665307080455423, "1000–3000 Hz_10ms_envelope_correlation": 0.9683139623706921, "1000–3000 Hz_10ms_modulation_mae_db": 0.7129122742515747, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7526417635812707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10080386370967728, "1000–3000 Hz_40ms_envelope_correlation": 0.9908017554034189, "1000–3000 Hz_40ms_modulation_mae_db": 0.46834054783931817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5263698483752846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1751376546446317, "3000–6000 Hz_10ms_envelope_correlation": 0.9771592731549558, "3000–6000 Hz_10ms_modulation_mae_db": 0.6392135240489706, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9405019122249756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11640745514118854, "3000–6000 Hz_40ms_envelope_correlation": 0.9892252503940268, "3000–6000 Hz_40ms_modulation_mae_db": 0.5217642546748061, "mr_stft_db": 5.709535958568004, "band_mae_db": 0.35477427435259695, "rms_gain_error_db": -0.05599672767827704, "output_peak": 0.5104088482105881}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9081641661714501, "全帯域_10ms_rms_envelope_nrmse": 0.10171043850930173, "全帯域_10ms_envelope_correlation": 0.9904785161114059, "全帯域_10ms_modulation_mae_db": 0.4211544515851764, "全帯域_40ms_log_rms_mae_db": 0.4908505807837052, "全帯域_40ms_rms_envelope_nrmse": 0.06627594083216232, "全帯域_40ms_envelope_correlation": 0.9957043875625314, "全帯域_40ms_modulation_mae_db": 0.512622615100093, "300–1000 Hz_10ms_log_rms_mae_db": 1.5865910448545413, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14364733544819178, "300–1000 Hz_10ms_envelope_correlation": 0.9845382499855624, "300–1000 Hz_10ms_modulation_mae_db": 0.8455094224871393, "300–1000 Hz_40ms_log_rms_mae_db": 0.9891927964855941, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10114493189703436, "300–1000 Hz_40ms_envelope_correlation": 0.9922257213361719, "300–1000 Hz_40ms_modulation_mae_db": 0.9187150569390555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.472692572376444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19927649427314822, "1000–3000 Hz_10ms_envelope_correlation": 0.9638315532785592, "1000–3000 Hz_10ms_modulation_mae_db": 0.40170930346586947, "1000–3000 Hz_40ms_log_rms_mae_db": 0.784639129382423, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10795614577235513, "1000–3000 Hz_40ms_envelope_correlation": 0.9894733437465337, "1000–3000 Hz_40ms_modulation_mae_db": 0.5006322213024319, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4956051864907736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18308327534257546, "3000–6000 Hz_10ms_envelope_correlation": 0.9760466355741882, "3000–6000 Hz_10ms_modulation_mae_db": 0.17392094143331804, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8892948743235073, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12245951001960435, "3000–6000 Hz_40ms_envelope_correlation": 0.9885424738031057, "3000–6000 Hz_40ms_modulation_mae_db": 0.4901240670580056, "mr_stft_db": 5.720578652433428, "band_mae_db": 0.45902634280917337, "rms_gain_error_db": -0.13176649856654063, "output_peak": 0.45590672714092867}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9044355124062522, "全帯域_10ms_rms_envelope_nrmse": 0.15041455126554706, "全帯域_10ms_envelope_correlation": 0.9793874752377542, "全帯域_10ms_modulation_mae_db": 0.26396861094015606, "全帯域_40ms_log_rms_mae_db": 0.5048663741423709, "全帯域_40ms_rms_envelope_nrmse": 0.10200956898446029, "全帯域_40ms_envelope_correlation": 0.9896220362126427, "全帯域_40ms_modulation_mae_db": 0.33061517766799076, "300–1000 Hz_10ms_log_rms_mae_db": 1.5653581788376785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17048007682188032, "300–1000 Hz_10ms_envelope_correlation": 0.9776012087118152, "300–1000 Hz_10ms_modulation_mae_db": 0.5391555843764614, "300–1000 Hz_40ms_log_rms_mae_db": 0.9589514922847108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11815586983496273, "300–1000 Hz_40ms_envelope_correlation": 0.9882041869938037, "300–1000 Hz_40ms_modulation_mae_db": 0.8648758099142105, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3931448576802263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18111381232502052, "1000–3000 Hz_10ms_envelope_correlation": 0.9700349820511137, "1000–3000 Hz_10ms_modulation_mae_db": 1.0076434263558447, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7945682470446357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11015442840523469, "1000–3000 Hz_40ms_envelope_correlation": 0.9881419421052673, "1000–3000 Hz_40ms_modulation_mae_db": 0.9274845404240807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.546231906665682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1767823904777622, "3000–6000 Hz_10ms_envelope_correlation": 0.9770091257206672, "3000–6000 Hz_10ms_modulation_mae_db": 0.3117915038817289, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9778043660706808, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11662933256969904, "3000–6000 Hz_40ms_envelope_correlation": 0.9891756702399221, "3000–6000 Hz_40ms_modulation_mae_db": 0.3311802752547774, "mr_stft_db": 5.740832721388682, "band_mae_db": 0.2809551208369749, "rms_gain_error_db": 0.06751604287165927, "output_peak": 0.5904487547843892}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9659160350658519, "全帯域_10ms_rms_envelope_nrmse": 0.1103714585187199, "全帯域_10ms_envelope_correlation": 0.988680744751843, "全帯域_10ms_modulation_mae_db": 0.2116475348068517, "全帯域_40ms_log_rms_mae_db": 0.5059993741747568, "全帯域_40ms_rms_envelope_nrmse": 0.0575013769440839, "全帯域_40ms_envelope_correlation": 0.9966440916948172, "全帯域_40ms_modulation_mae_db": 0.2480827283834178, "300–1000 Hz_10ms_log_rms_mae_db": 1.5905675750476456, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14186313516779406, "300–1000 Hz_10ms_envelope_correlation": 0.9847584186625516, "300–1000 Hz_10ms_modulation_mae_db": 0.4992046918606441, "300–1000 Hz_40ms_log_rms_mae_db": 0.9646425711955503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09133080646205645, "300–1000 Hz_40ms_envelope_correlation": 0.9936200706740598, "300–1000 Hz_40ms_modulation_mae_db": 0.5153847571697937, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3404995880598485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1698081316374532, "1000–3000 Hz_10ms_envelope_correlation": 0.9736563726527904, "1000–3000 Hz_10ms_modulation_mae_db": 0.35545526969302926, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7312923117323459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08030586680196826, "1000–3000 Hz_40ms_envelope_correlation": 0.9940599313334469, "1000–3000 Hz_40ms_modulation_mae_db": 0.6528464161925123, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4465293872605902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16236016445623702, "3000–6000 Hz_10ms_envelope_correlation": 0.9811331224174724, "3000–6000 Hz_10ms_modulation_mae_db": 0.5393129330928416, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8652900477318285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11214432205768048, "3000–6000 Hz_40ms_envelope_correlation": 0.9903825029623962, "3000–6000 Hz_40ms_modulation_mae_db": 0.3342883755673703, "mr_stft_db": 5.734200014713693, "band_mae_db": 0.2718191658538543, "rms_gain_error_db": -0.036941294386512065, "output_peak": 0.47390872515738286}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9628151400515688, "全帯域_10ms_rms_envelope_nrmse": 0.10848714537425097, "全帯域_10ms_envelope_correlation": 0.9891748856718237, "全帯域_10ms_modulation_mae_db": 1.0041777364807645, "全帯域_40ms_log_rms_mae_db": 0.5059359246838884, "全帯域_40ms_rms_envelope_nrmse": 0.06352763110455811, "全帯域_40ms_envelope_correlation": 0.9961304164654314, "全帯域_40ms_modulation_mae_db": 1.17952052267851, "300–1000 Hz_10ms_log_rms_mae_db": 1.5276651312825256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13692690931737092, "300–1000 Hz_10ms_envelope_correlation": 0.9857182465958935, "300–1000 Hz_10ms_modulation_mae_db": 1.173134640850709, "300–1000 Hz_40ms_log_rms_mae_db": 0.8500700445181303, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08391557251112165, "300–1000 Hz_40ms_envelope_correlation": 0.9945642348449695, "300–1000 Hz_40ms_modulation_mae_db": 1.3530579282205306, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4144905374916412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18627167117670404, "1000–3000 Hz_10ms_envelope_correlation": 0.9696799030501523, "1000–3000 Hz_10ms_modulation_mae_db": 0.9924918237240234, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8190532222179296, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11097449242493448, "1000–3000 Hz_40ms_envelope_correlation": 0.990337339081891, "1000–3000 Hz_40ms_modulation_mae_db": 0.9151524712966088, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5306245339189037, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17881980908720055, "3000–6000 Hz_10ms_envelope_correlation": 0.9761565012387289, "3000–6000 Hz_10ms_modulation_mae_db": 0.379888898208322, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8902018760653883, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11642278289749045, "3000–6000 Hz_40ms_envelope_correlation": 0.9892906213347862, "3000–6000 Hz_40ms_modulation_mae_db": 0.3144377953721646, "mr_stft_db": 5.71371165326586, "band_mae_db": 0.4544244437317788, "rms_gain_error_db": -0.15054962653787532, "output_peak": 0.386158750011785}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9594146251973391, "全帯域_10ms_rms_envelope_nrmse": 0.10044679005646598, "全帯域_10ms_envelope_correlation": 0.9907030078336693, "全帯域_10ms_modulation_mae_db": 0.42609142601101874, "全帯域_40ms_log_rms_mae_db": 0.4851607234485811, "全帯域_40ms_rms_envelope_nrmse": 0.048746245067979053, "全帯域_40ms_envelope_correlation": 0.9977479849993958, "全帯域_40ms_modulation_mae_db": 0.5115433719149168, "300–1000 Hz_10ms_log_rms_mae_db": 1.6311518334755386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12285591526664845, "300–1000 Hz_10ms_envelope_correlation": 0.9886057677126052, "300–1000 Hz_10ms_modulation_mae_db": 0.6845014025238457, "300–1000 Hz_40ms_log_rms_mae_db": 0.9348622933014304, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07058554933005816, "300–1000 Hz_40ms_envelope_correlation": 0.9963275097447257, "300–1000 Hz_40ms_modulation_mae_db": 0.8674957963764065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3860855663707927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22051517903314052, "1000–3000 Hz_10ms_envelope_correlation": 0.955783785726795, "1000–3000 Hz_10ms_modulation_mae_db": 0.42480261170332634, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7759827387662359, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14510725468475555, "1000–3000 Hz_40ms_envelope_correlation": 0.9787099327982932, "1000–3000 Hz_40ms_modulation_mae_db": 0.3655960895454157, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4760076597758012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17829145496063248, "3000–6000 Hz_10ms_envelope_correlation": 0.9766760451477816, "3000–6000 Hz_10ms_modulation_mae_db": 0.9498248227488721, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9654603431299297, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12961634428842222, "3000–6000 Hz_40ms_envelope_correlation": 0.9873317640366353, "3000–6000 Hz_40ms_modulation_mae_db": 0.6065085148230118, "mr_stft_db": 5.720591143626881, "band_mae_db": 0.4202178831316927, "rms_gain_error_db": -0.11614173615404942, "output_peak": 0.410879562039368}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9784872676341269, "全帯域_10ms_rms_envelope_nrmse": 0.12007159504970234, "全帯域_10ms_envelope_correlation": 0.9866151169870172, "全帯域_10ms_modulation_mae_db": 0.2323504908985979, "全帯域_40ms_log_rms_mae_db": 0.5618096346242966, "全帯域_40ms_rms_envelope_nrmse": 0.07316662482257745, "全帯域_40ms_envelope_correlation": 0.9945891789135664, "全帯域_40ms_modulation_mae_db": 0.44691583972118565, "300–1000 Hz_10ms_log_rms_mae_db": 1.6889575297122097, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15468630944508063, "300–1000 Hz_10ms_envelope_correlation": 0.9816000398322154, "300–1000 Hz_10ms_modulation_mae_db": 0.15190795016599098, "300–1000 Hz_40ms_log_rms_mae_db": 1.147257094588845, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10893903770219276, "300–1000 Hz_40ms_envelope_correlation": 0.9903448835300018, "300–1000 Hz_40ms_modulation_mae_db": 0.4580677109918028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3670731569555525, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1758077175495806, "1000–3000 Hz_10ms_envelope_correlation": 0.9718080592824264, "1000–3000 Hz_10ms_modulation_mae_db": 0.7944597649863523, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7291700749305959, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08955700953780876, "1000–3000 Hz_40ms_envelope_correlation": 0.9919366407469791, "1000–3000 Hz_40ms_modulation_mae_db": 0.5720009171924143, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4509910884957098, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18629985453607947, "3000–6000 Hz_10ms_envelope_correlation": 0.9741248110631588, "3000–6000 Hz_10ms_modulation_mae_db": 1.0659680520355836, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9019804152091863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13338258502334052, "3000–6000 Hz_40ms_envelope_correlation": 0.9859832420493106, "3000–6000 Hz_40ms_modulation_mae_db": 1.1036731556373947, "mr_stft_db": 5.7190416700597355, "band_mae_db": 0.42121506740204334, "rms_gain_error_db": -0.07293131071119031, "output_peak": 0.40634445850960293}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0169961761744795, "全帯域_10ms_rms_envelope_nrmse": 0.12157350059586461, "全帯域_10ms_envelope_correlation": 0.9867793420199013, "全帯域_10ms_modulation_mae_db": 0.11533895509955093, "全帯域_40ms_log_rms_mae_db": 0.4718231298095083, "全帯域_40ms_rms_envelope_nrmse": 0.0717948995926075, "全帯域_40ms_envelope_correlation": 0.9951032957088973, "全帯域_40ms_modulation_mae_db": 0.23283187438986133, "300–1000 Hz_10ms_log_rms_mae_db": 1.6529840726864864, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13812529572576146, "300–1000 Hz_10ms_envelope_correlation": 0.9853995948384778, "300–1000 Hz_10ms_modulation_mae_db": 0.22779598946516766, "300–1000 Hz_40ms_log_rms_mae_db": 0.8899633561350982, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08387224103924347, "300–1000 Hz_40ms_envelope_correlation": 0.9941297043620018, "300–1000 Hz_40ms_modulation_mae_db": 0.19158215500455889, "1000–3000 Hz_10ms_log_rms_mae_db": 1.381473851900762, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19092652796530574, "1000–3000 Hz_10ms_envelope_correlation": 0.9665780349348585, "1000–3000 Hz_10ms_modulation_mae_db": 0.2549783592534305, "1000–3000 Hz_40ms_log_rms_mae_db": 0.697759803426412, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09516856618141183, "1000–3000 Hz_40ms_envelope_correlation": 0.9911587141664278, "1000–3000 Hz_40ms_modulation_mae_db": 0.306817386448357, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5956250395623863, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19629308263508577, "3000–6000 Hz_10ms_envelope_correlation": 0.971479989623581, "3000–6000 Hz_10ms_modulation_mae_db": 0.7441730626467011, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9723671562386921, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13036246443164654, "3000–6000 Hz_40ms_envelope_correlation": 0.9863990830649325, "3000–6000 Hz_40ms_modulation_mae_db": 0.6496418710372721, "mr_stft_db": 5.74213964516143, "band_mae_db": 0.3803984477382707, "rms_gain_error_db": 0.12745210237808074, "output_peak": 0.40980378531985934}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 40080, "output_samples": 40080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7500078257548415, "spectral_convergence": 0.33040072988412605, "spectral_convergence_status": "ok", "amplitude_floor": 1.4347408585118857e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.506676243958292, "spectral_convergence": 0.31886965758651603, "spectral_convergence_status": "ok", "amplitude_floor": 9.356428468567288e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.871923805990877, "spectral_convergence": 0.3210522217081813, "spectral_convergence_status": "ok", "amplitude_floor": 6.30553407186755e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020558767458207216, "output_power": 0.0002741387914549513, "delta_db": 1.249734196230104, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002656133227213937, "output_power": 0.002582925925020022, "delta_db": -0.121379033823844, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000658897280424305, "output_power": 0.0006321545993515356, "delta_db": -0.17994412944582555, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.296049498403337e-05, "output_power": 6.386943983632453e-05, "delta_db": 0.06224973507849664, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018191885630194454, "output_power": 0.00016566635475462538, "delta_db": -0.4064340066763971, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012525230408717123, "output_power": 0.00012215050568274417, "delta_db": -0.1089045448609139, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.35477427435259695, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05599672767827704, "pitch": null, "output_rms_dbfs": -24.15566364771948, "output_peak": 0.5104088482105881, "reference_rms_dbfs": -24.099666920041205, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.709535958568004}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 40080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.0001593748762509214, "log_rms_mae_db": 0.9320150359409302, "rms_envelope_nrmse": 0.10215665026523217, "envelope_correlation": 0.9903373888715449, "rms_gain_error_db": -0.05600497390433632, "reference_mean_power": 0.003898496329889168, "output_mean_power": 0.003848545560945265}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.3730898670333144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011160072986886684, "output_power": 0.0011303214847418655, "reference_active": true, "delta_db": 0.055349476945978235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003924601048841299, "output_power": 0.00041313916685973445, "reference_active": true, "delta_db": 0.22300853796908182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.521600297301381e-05, "output_power": 0.00010342400563766296, "reference_active": true, "delta_db": 0.8410019418543047}], "active_band_count": 3, "active_band_mae_db": 0.37311998558978826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00014354118626608257, "log_rms_mae_db": 0.5220720920402584, "rms_envelope_nrmse": 0.061584941804164225, "envelope_correlation": 0.9961725635889591, "rms_gain_error_db": -0.05601323152933733, "reference_mean_power": 0.00394559944963645, "output_mean_power": 0.003895037750461615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 2.0339309551628037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010452646972319804, "output_power": 0.0010740936445179658, "reference_active": true, "delta_db": 0.11815864060094612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020675655639960477, "output_power": 0.00022289309460987226, "reference_active": true, "delta_db": 0.3263732369547901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.033282667633223e-06, "output_power": 8.506531200544748e-06, "reference_active": true, "delta_db": 1.4919882617478508}], "active_band_count": 3, "active_band_mae_db": 0.6455067131011957}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.00014555177137634006, "log_rms_mae_db": 1.5594988149239297, "rms_envelope_nrmse": 0.1411079345432263, "envelope_correlation": 0.9847123069168086, "rms_gain_error_db": -0.2173722136409184, "reference_mean_power": 0.0025948122953778335, "output_mean_power": 0.002468133949139082}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.091667884405576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010721637846984168, "output_power": 0.001075442852436911, "reference_active": true, "delta_db": 0.013262038835491836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00045198280410970506, "output_power": 0.00042540967379496196, "reference_active": true, "delta_db": -0.2631455068155309}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.681659745338991e-05, "output_power": 8.24943288660453e-05, "reference_active": true, "delta_db": 1.6194887188574811}], "active_band_count": 3, "active_band_mae_db": 0.6319654215028346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00013266563413733552, "log_rms_mae_db": 0.9453777074747345, "rms_envelope_nrmse": 0.08803362701104542, "envelope_correlation": 0.9938015706440676, "rms_gain_error_db": -0.21737156971735577, "reference_mean_power": 0.002626297782108314, "output_mean_power": 0.0024980826895278016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 1.815151890487515e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010442677388382842, "output_power": 0.0010615679066092564, "reference_active": true, "delta_db": 0.07135919132638827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002292871431701824, "output_power": 0.00021972585953027976, "reference_active": true, "delta_db": -0.18498531161857745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.238038641621491e-06, "output_power": 9.057961432676991e-06, "reference_active": true, "delta_db": 0.97409569855181}], "active_band_count": 3, "active_band_mae_db": 0.4101467338322586}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.977119603351369e-05, "log_rms_mae_db": 1.3110129905663104, "rms_envelope_nrmse": 0.18665307080455423, "envelope_correlation": 0.9683139623706921, "rms_gain_error_db": -0.3385172677270475, "reference_mean_power": 0.0001519473982974274, "output_mean_power": 0.00014055346014861663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 8.081548208249742e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5556653827814076e-05, "output_power": 3.972143462510896e-05, "reference_active": true, "delta_db": -0.5952689171938357}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1818343242486535e-05, "output_power": 1.0336593209142046e-05, "reference_active": true, "delta_db": -0.5817917393474339}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7000790662746577e-06, "output_power": 2.9651276038171823e-06, "reference_active": true, "delta_db": -0.9616761662134544}], "active_band_count": 3, "active_band_mae_db": 0.7129122742515747}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.109627865595404e-05, "log_rms_mae_db": 0.7526417635812707, "rms_envelope_nrmse": 0.10080386370967728, "envelope_correlation": 0.9908017554034189, "rms_gain_error_db": -0.3389781807937628, "reference_mean_power": 0.0001537098951112271, "output_mean_power": 0.00014216870520875378}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 6.838335178235257e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3201238570188224e-05, "output_power": 3.717327034782082e-05, "reference_active": true, "delta_db": -0.6526542794592073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.064245816738516e-06, "output_power": 5.3438342307146394e-06, "reference_active": true, "delta_db": -0.5492381971582241}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0299327724151498e-07, "output_power": 1.9371744770849964e-07, "reference_active": true, "delta_db": -0.20312916690052338}], "active_band_count": 3, "active_band_mae_db": 0.46834054783931817}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.882145378793849e-05, "log_rms_mae_db": 1.5263698483752846, "rms_envelope_nrmse": 0.1751376546446317, "envelope_correlation": 0.9771592731549558, "rms_gain_error_db": -0.14886596887092576, "reference_mean_power": 0.00013710154501992013, "output_mean_power": 0.00013248165721356102}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 1.0763544842397476e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5103901227352814e-05, "output_power": 4.194721830790635e-05, "reference_active": true, "delta_db": -0.3151094119957103}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.11132679073469e-05, "output_power": 2.8353251317424283e-05, "reference_active": true, "delta_db": -0.40342761181830955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.435692144349255e-06, "output_power": 4.124245386297096e-06, "reference_active": true, "delta_db": -1.1991035483328918}], "active_band_count": 3, "active_band_mae_db": 0.6392135240489706}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.3950747306848176e-05, "log_rms_mae_db": 0.9405019122249756, "rms_envelope_nrmse": 0.11640745514118854, "envelope_correlation": 0.9892252503940268, "rms_gain_error_db": -0.14870583294184808, "reference_mean_power": 0.00013864857515666164, "output_mean_power": 0.00013398149749798495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 9.21368112936246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.381631867120006e-05, "output_power": 4.087724642876382e-05, "reference_active": true, "delta_db": -0.30154253320975183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.911163628753255e-05, "output_power": 1.7145075510977682e-05, "reference_active": true, "delta_db": -0.4715846959573986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5242776229186804e-07, "output_power": 2.93665926375117e-07, "reference_active": true, "delta_db": -0.7921655348572677}], "active_band_count": 3, "active_band_mae_db": 0.5217642546748061}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8844765177083529, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "21664f47700a38a460f7410f9e4a65a8394bc74bc2aafcfd991d8b9d52b7adf9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9674608316684234, "standard_deviation": 0.005604409866076328, "minimum": 0.955783785726795, "maximum": 0.9736563726527904, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3833091401751971, "standard_deviation": 0.0483919913779798, "minimum": 1.3110129905663104, "maximum": 1.472692572376444, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6180566041791813, "standard_deviation": 0.29689676702349255, "minimum": 0.2549783592534305, "maximum": 1.0076434263558447, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1887965755956134, "standard_deviation": 0.015673914819989854, "minimum": 0.1698081316374532, "maximum": 0.22051517903314052, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893274499227822, "standard_deviation": 0.004627705729552801, "minimum": 0.9787099327982932, "maximum": 0.9940599313334469, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.760638411385231, "standard_deviation": 0.04007871219532347, "minimum": 0.697759803426412, "maximum": 0.8190532222179296, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5886088237801423, "standard_deviation": 0.23213765327314098, "minimum": 0.306817386448357, "maximum": 0.9274845404240807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10500345343976825, "standard_deviation": 0.019428050752159087, "minimum": 0.08030586680196826, "maximum": 0.14510725468475555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9762231879925668, "standard_deviation": 0.0027449492211955194, "minimum": 0.971479989623581, "maximum": 0.9811331224174724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5084980813181414, "standard_deviation": 0.050980326578561906, "minimum": 1.4465293872605902, "maximum": 1.5956250395623863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6005117172620422, "standard_deviation": 0.31134831857572104, "minimum": 0.17392094143331804, "maximum": 1.0659680520355836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1796334607675256, "standard_deviation": 0.009732535096569923, "minimum": 0.16236016445623702, "maximum": 0.19629308263508577, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9882913259856394, "standard_deviation": 0.0015546042671417504, "minimum": 0.9859832420493106, "maximum": 0.9903825029623962, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9253626238742736, "standard_deviation": 0.04389791787976086, "minimum": 0.8652900477318285, "maximum": 0.9778043660706808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5439522886781003, "standard_deviation": 0.2601287025544761, "minimum": 0.3144377953721646, "maximum": 1.1036731556373947, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12217809955363401, "standard_deviation": 0.007980516563212412, "minimum": 0.11214432205768048, "maximum": 0.13338258502334052, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9841167291569912, "standard_deviation": 0.003253392944187757, "minimum": 0.9776012087118152, "maximum": 0.9886057677126052, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6003467726025693, "standard_deviation": 0.053494638565650995, "minimum": 1.5276651312825256, "maximum": 1.6889575297122097, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5941468879040992, "standard_deviation": 0.3272898332282148, "minimum": 0.15190795016599098, "maximum": 1.173134640850709, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14371161396699425, "standard_deviation": 0.013935535427437327, "minimum": 0.12285591526664845, "maximum": 0.17048007682188032, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9929022352662252, "standard_deviation": 0.0025744596270397273, "minimum": 0.9882041869938037, "maximum": 0.9963275097447257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9600396694980117, "standard_deviation": 0.08764825922788509, "minimum": 0.8500700445181303, "maximum": 1.147257094588845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6974157435560772, "standard_deviation": 0.37121884622708046, "minimum": 0.19158215500455889, "maximum": 1.3530579282205306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09324720447346438, "standard_deviation": 0.015345651865469802, "minimum": 0.07058554933005816, "maximum": 0.11815586983496273, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3803538432320481, "standard_deviation": 0.07294694344607745, "minimum": 0.2718191658538543, "maximum": 0.45902634280917337, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.725078932402214, "standard_deviation": 0.012364027920148624, "minimum": 5.709535958568004, "maximum": 5.74213964516143, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45548245139673804, "standard_deviation": 0.06852402694776316, "minimum": 0.386158750011785, "maximum": 0.5904487547843892, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04616988109808809, "standard_deviation": 0.09789569009664079, "minimum": -0.15054962653787532, "maximum": 0.12745210237808074, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.98776955968562, "standard_deviation": 0.003740036625023274, "minimum": 0.9793874752377542, "maximum": 0.9907030078336693, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9535304948302499, "standard_deviation": 0.03752389428783228, "minimum": 0.9044355124062522, "maximum": 1.0169961761744795, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38098114892648804, "standard_deviation": 0.27441034719591106, "minimum": 0.11533895509955093, "maximum": 1.0041777364807645, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1144040162043856, "standard_deviation": 0.016636798810383477, "minimum": 0.10044679005646598, "maximum": 0.15041455126554706, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9952142443932801, "standard_deviation": 0.0024535135400318452, "minimum": 0.9896220362126427, "maximum": 0.9977479849993958, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5060647292134207, "standard_deviation": 0.027271444929757617, "minimum": 0.4718231298095083, "maximum": 0.5618096346242966, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5134548553696463, "standard_deviation": 0.3042825107108538, "minimum": 0.23283187438986133, "maximum": 1.17952052267851, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0680759036440741, "standard_deviation": 0.01577862734767759, "minimum": 0.048746245067979053, "maximum": 0.10200956898446029, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0038", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.636467695236206, "predicted_mos": 2.636467695236206, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "dd1e148d8a43f0c0f64c63fbc3f617dad21c6c4915a01cf7c3b072f0cdd8b75d", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.904184500221163, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.913934290426132, "全帯域_10ms_rms_envelope_nrmse": 0.0875505793279052, "全帯域_10ms_envelope_correlation": 0.9931073540048588, "全帯域_10ms_modulation_mae_db": 0.19573719790811142, "全帯域_40ms_log_rms_mae_db": 0.47007641836219555, "全帯域_40ms_rms_envelope_nrmse": 0.047383690585783816, "全帯域_40ms_envelope_correlation": 0.9980595536733371, "全帯域_40ms_modulation_mae_db": 0.12198225687349828, "300–1000 Hz_10ms_log_rms_mae_db": 1.5556305302874152, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10913948466427127, "300–1000 Hz_10ms_envelope_correlation": 0.9915791721408826, "300–1000 Hz_10ms_modulation_mae_db": 0.4024064353334185, "300–1000 Hz_40ms_log_rms_mae_db": 0.8911321192604691, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06802945771770486, "300–1000 Hz_40ms_envelope_correlation": 0.9971747658784741, "300–1000 Hz_40ms_modulation_mae_db": 0.6772274320030927, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5223030011191958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20171144304993013, "1000–3000 Hz_10ms_envelope_correlation": 0.9628350521038813, "1000–3000 Hz_10ms_modulation_mae_db": 0.8119790114381721, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7838416498782848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10033424566016322, "1000–3000 Hz_40ms_envelope_correlation": 0.989728023646444, "1000–3000 Hz_40ms_modulation_mae_db": 0.826333905821366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.72944209663637, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1944704611804575, "3000–6000 Hz_10ms_envelope_correlation": 0.9717356313012231, "3000–6000 Hz_10ms_modulation_mae_db": 0.5543615424602449, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0968901142184686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13038033758063935, "3000–6000 Hz_40ms_envelope_correlation": 0.9864358246240715, "3000–6000 Hz_40ms_modulation_mae_db": 0.6511961099161956, "mr_stft_db": 5.849931267270388, "band_mae_db": 0.3613467001591834, "rms_gain_error_db": -0.16784183950268122, "output_peak": 0.3920935910531185}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9123196703001134, "全帯域_10ms_rms_envelope_nrmse": 0.09446021074641414, "全帯域_10ms_envelope_correlation": 0.9918280197551874, "全帯域_10ms_modulation_mae_db": 0.5238655509434652, "全帯域_40ms_log_rms_mae_db": 0.484488438127962, "全帯域_40ms_rms_envelope_nrmse": 0.04990510168202523, "全帯域_40ms_envelope_correlation": 0.9976770581495247, "全帯域_40ms_modulation_mae_db": 0.5093343870248959, "300–1000 Hz_10ms_log_rms_mae_db": 1.6339641450632671, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13229624736647702, "300–1000 Hz_10ms_envelope_correlation": 0.9868684490383606, "300–1000 Hz_10ms_modulation_mae_db": 0.7174875347574693, "300–1000 Hz_40ms_log_rms_mae_db": 0.9759414776896498, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08565385239789555, "300–1000 Hz_40ms_envelope_correlation": 0.99452212697916, "300–1000 Hz_40ms_modulation_mae_db": 0.6733978128628711, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5082494425722284, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2005500178688402, "1000–3000 Hz_10ms_envelope_correlation": 0.9631613543603318, "1000–3000 Hz_10ms_modulation_mae_db": 0.27565004896907336, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8341576304769417, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10217018381319955, "1000–3000 Hz_40ms_envelope_correlation": 0.9893830640298031, "1000–3000 Hz_40ms_modulation_mae_db": 0.2931777999115866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.744493394839235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2024747691435689, "3000–6000 Hz_10ms_envelope_correlation": 0.9701264579579439, "3000–6000 Hz_10ms_modulation_mae_db": 1.0747178702393079, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0885762015409022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13384874485047737, "3000–6000 Hz_40ms_envelope_correlation": 0.9859533744218898, "3000–6000 Hz_40ms_modulation_mae_db": 0.8723333809045403, "mr_stft_db": 5.869231420591401, "band_mae_db": 0.4720124772922758, "rms_gain_error_db": -0.13567126048202768, "output_peak": 0.4036908394689761}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9028240067708355, "全帯域_10ms_rms_envelope_nrmse": 0.07778637220922252, "全帯域_10ms_envelope_correlation": 0.994395914542328, "全帯域_10ms_modulation_mae_db": 0.13794682793273935, "全帯域_40ms_log_rms_mae_db": 0.47284524475160666, "全帯域_40ms_rms_envelope_nrmse": 0.041099331471404925, "全帯域_40ms_envelope_correlation": 0.9982845299789382, "全帯域_40ms_modulation_mae_db": 0.2110158790024735, "300–1000 Hz_10ms_log_rms_mae_db": 1.668547462273108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10088660345172243, "300–1000 Hz_10ms_envelope_correlation": 0.9923143951195108, "300–1000 Hz_10ms_modulation_mae_db": 0.2373850856764431, "300–1000 Hz_40ms_log_rms_mae_db": 0.9131009346373304, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060521760269313316, "300–1000 Hz_40ms_envelope_correlation": 0.9972049156730055, "300–1000 Hz_40ms_modulation_mae_db": 0.39980062128177923, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5741033174764183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19792186664968406, "1000–3000 Hz_10ms_envelope_correlation": 0.9640865837580844, "1000–3000 Hz_10ms_modulation_mae_db": 0.5009150558497371, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9447108306534333, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11243121149332169, "1000–3000 Hz_40ms_envelope_correlation": 0.9871900252812176, "1000–3000 Hz_40ms_modulation_mae_db": 0.8858436106311913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7308124195356784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18920238553587299, "3000–6000 Hz_10ms_envelope_correlation": 0.9739417915241387, "3000–6000 Hz_10ms_modulation_mae_db": 0.4845672375284318, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0804841575911321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12865518946939314, "3000–6000 Hz_40ms_envelope_correlation": 0.9870853928216066, "3000–6000 Hz_40ms_modulation_mae_db": 0.4978374436931983, "mr_stft_db": 5.874405908730626, "band_mae_db": 0.29199870842624376, "rms_gain_error_db": -0.0006335515690642785, "output_peak": 0.43381579849007884}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9849809469784886, "全帯域_10ms_rms_envelope_nrmse": 0.1055622795996395, "全帯域_10ms_envelope_correlation": 0.9896648627193612, "全帯域_10ms_modulation_mae_db": 0.5400547564718107, "全帯域_40ms_log_rms_mae_db": 0.5547683561892511, "全帯域_40ms_rms_envelope_nrmse": 0.05063001008429266, "全帯域_40ms_envelope_correlation": 0.9974737076466202, "全帯域_40ms_modulation_mae_db": 0.3767917239869287, "300–1000 Hz_10ms_log_rms_mae_db": 1.6635180951925155, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13937163176463851, "300–1000 Hz_10ms_envelope_correlation": 0.9852148605396429, "300–1000 Hz_10ms_modulation_mae_db": 0.6925333972417395, "300–1000 Hz_40ms_log_rms_mae_db": 0.9859649106047542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08259749919052482, "300–1000 Hz_40ms_envelope_correlation": 0.9948022813368748, "300–1000 Hz_40ms_modulation_mae_db": 1.0081629487419332, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4915225440673907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19052709517671315, "1000–3000 Hz_10ms_envelope_correlation": 0.9670695555472401, "1000–3000 Hz_10ms_modulation_mae_db": 0.349259508564613, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8601174887447077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11168446171349937, "1000–3000 Hz_40ms_envelope_correlation": 0.9887101555698996, "1000–3000 Hz_40ms_modulation_mae_db": 0.5588507271289155, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7657634709491363, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19076590438673666, "3000–6000 Hz_10ms_envelope_correlation": 0.9740301552374137, "3000–6000 Hz_10ms_modulation_mae_db": 0.6603295658603633, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0935267053089144, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1319772032356663, "3000–6000 Hz_40ms_envelope_correlation": 0.9867536972525178, "3000–6000 Hz_40ms_modulation_mae_db": 0.3208250860463157, "mr_stft_db": 5.880732966730456, "band_mae_db": 0.4094588999291197, "rms_gain_error_db": -0.08503918477742894, "output_peak": 0.42113952098213747}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9672962010508724, "全帯域_10ms_rms_envelope_nrmse": 0.11928611186923659, "全帯域_10ms_envelope_correlation": 0.9871536838887351, "全帯域_10ms_modulation_mae_db": 0.4529912539568374, "全帯域_40ms_log_rms_mae_db": 0.47876052362441684, "全帯域_40ms_rms_envelope_nrmse": 0.0698031570008599, "全帯域_40ms_envelope_correlation": 0.995265325346262, "全帯域_40ms_modulation_mae_db": 0.519468240809892, "300–1000 Hz_10ms_log_rms_mae_db": 1.7196595188739292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14388073428318837, "300–1000 Hz_10ms_envelope_correlation": 0.9841785415957716, "300–1000 Hz_10ms_modulation_mae_db": 0.5703644466372229, "300–1000 Hz_40ms_log_rms_mae_db": 0.9608497107818674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08127018657562396, "300–1000 Hz_40ms_envelope_correlation": 0.9944747524262681, "300–1000 Hz_40ms_modulation_mae_db": 0.08363565068503187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5395921140086966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20335805286312844, "1000–3000 Hz_10ms_envelope_correlation": 0.9629421890753037, "1000–3000 Hz_10ms_modulation_mae_db": 0.059124332174102824, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8100353261011903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10231528228991212, "1000–3000 Hz_40ms_envelope_correlation": 0.9896454983781602, "1000–3000 Hz_40ms_modulation_mae_db": 0.08171200967739271, "3000–6000 Hz_10ms_log_rms_mae_db": 1.739964184305638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1927231004023594, "3000–6000 Hz_10ms_envelope_correlation": 0.9724324572388795, "3000–6000 Hz_10ms_modulation_mae_db": 0.5930053464276267, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1163165371165165, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1259897888250304, "3000–6000 Hz_40ms_envelope_correlation": 0.9873051965121439, "3000–6000 Hz_40ms_modulation_mae_db": 0.28656688993263585, "mr_stft_db": 5.842920284623438, "band_mae_db": 0.369069969683683, "rms_gain_error_db": 0.10426442031434668, "output_peak": 0.43654624713493667}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.005569279117364, "全帯域_10ms_rms_envelope_nrmse": 0.11925528352615666, "全帯域_10ms_envelope_correlation": 0.986870915364858, "全帯域_10ms_modulation_mae_db": 0.33348025660668207, "全帯域_40ms_log_rms_mae_db": 0.5563876438910037, "全帯域_40ms_rms_envelope_nrmse": 0.0766076626706861, "全帯域_40ms_envelope_correlation": 0.9940630553932104, "全帯域_40ms_modulation_mae_db": 0.4886753613478856, "300–1000 Hz_10ms_log_rms_mae_db": 1.7986010484727613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1600639768865344, "300–1000 Hz_10ms_envelope_correlation": 0.9801038330019053, "300–1000 Hz_10ms_modulation_mae_db": 0.17307701752922744, "300–1000 Hz_40ms_log_rms_mae_db": 1.001050054588703, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11742421046909453, "300–1000 Hz_40ms_envelope_correlation": 0.9882896990250251, "300–1000 Hz_40ms_modulation_mae_db": 0.5326532630768589, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6010472082736846, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.188392114221574, "1000–3000 Hz_10ms_envelope_correlation": 0.9672891065279183, "1000–3000 Hz_10ms_modulation_mae_db": 0.26196686205185665, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9225806911984901, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09873019141323129, "1000–3000 Hz_40ms_envelope_correlation": 0.990370613862301, "1000–3000 Hz_40ms_modulation_mae_db": 0.1576282006655933, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7405846227359854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19065448749069894, "3000–6000 Hz_10ms_envelope_correlation": 0.9730088148846027, "3000–6000 Hz_10ms_modulation_mae_db": 0.8242474432094347, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0424717271872517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12281932753395769, "3000–6000 Hz_40ms_envelope_correlation": 0.98807677975474, "3000–6000 Hz_40ms_modulation_mae_db": 0.45549955770284917, "mr_stft_db": 5.861500541239443, "band_mae_db": 0.40347087604297865, "rms_gain_error_db": 0.0136298314389149, "output_peak": 0.38788909078273115}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9422303734628173, "全帯域_10ms_rms_envelope_nrmse": 0.11477411382410455, "全帯域_10ms_envelope_correlation": 0.9877721951540653, "全帯域_10ms_modulation_mae_db": 0.788937455242065, "全帯域_40ms_log_rms_mae_db": 0.5160710067916343, "全帯域_40ms_rms_envelope_nrmse": 0.06172833326626714, "全帯域_40ms_envelope_correlation": 0.9962214937175303, "全帯域_40ms_modulation_mae_db": 1.0271272156309632, "300–1000 Hz_10ms_log_rms_mae_db": 1.7064982661921886, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15354615150178436, "300–1000 Hz_10ms_envelope_correlation": 0.9819143300526001, "300–1000 Hz_10ms_modulation_mae_db": 0.79821828791057, "300–1000 Hz_40ms_log_rms_mae_db": 0.9383077285631151, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09279595706530731, "300–1000 Hz_40ms_envelope_correlation": 0.9932548160733201, "300–1000 Hz_40ms_modulation_mae_db": 1.1836698046633465, "1000–3000 Hz_10ms_log_rms_mae_db": 1.549403041963614, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22390349037384785, "1000–3000 Hz_10ms_envelope_correlation": 0.9574095479014324, "1000–3000 Hz_10ms_modulation_mae_db": 0.4448347693558916, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8708856655026203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12424074028590323, "1000–3000 Hz_40ms_envelope_correlation": 0.9862241370826059, "1000–3000 Hz_40ms_modulation_mae_db": 0.48199502898441704, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8370055235060168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18060229170256192, "3000–6000 Hz_10ms_envelope_correlation": 0.9771504212922771, "3000–6000 Hz_10ms_modulation_mae_db": 0.6942736880977675, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2188344684059458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12769194656085897, "3000–6000 Hz_40ms_envelope_correlation": 0.9879656130161028, "3000–6000 Hz_40ms_modulation_mae_db": 0.32582765963241356, "mr_stft_db": 5.869917793858616, "band_mae_db": 0.3364984887313227, "rms_gain_error_db": -0.0998025323649384, "output_peak": 0.568673144851044}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9902066771392091, "全帯域_10ms_rms_envelope_nrmse": 0.10337004318813436, "全帯域_10ms_envelope_correlation": 0.9901549453670245, "全帯域_10ms_modulation_mae_db": 0.2688019031348072, "全帯域_40ms_log_rms_mae_db": 0.5182848695774034, "全帯域_40ms_rms_envelope_nrmse": 0.05204149480132116, "全帯域_40ms_envelope_correlation": 0.9972790916534972, "全帯域_40ms_modulation_mae_db": 0.34804693708955875, "300–1000 Hz_10ms_log_rms_mae_db": 1.6299422452794803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12205476463916781, "300–1000 Hz_10ms_envelope_correlation": 0.9884414338150769, "300–1000 Hz_10ms_modulation_mae_db": 0.27284324074618205, "300–1000 Hz_40ms_log_rms_mae_db": 0.8739774181554477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060262120641350586, "300–1000 Hz_40ms_envelope_correlation": 0.9969411524894193, "300–1000 Hz_40ms_modulation_mae_db": 0.4886140101577823, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6280399172931532, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22553198420422182, "1000–3000 Hz_10ms_envelope_correlation": 0.9531224987215121, "1000–3000 Hz_10ms_modulation_mae_db": 0.2618590344491249, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0196915184856434, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12302312055329576, "1000–3000 Hz_40ms_envelope_correlation": 0.9848555899118533, "1000–3000 Hz_40ms_modulation_mae_db": 0.34147216646973183, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8365297690104185, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1977098355373808, "3000–6000 Hz_10ms_envelope_correlation": 0.9719452673448238, "3000–6000 Hz_10ms_modulation_mae_db": 0.8445395957493557, "3000–6000 Hz_40ms_log_rms_mae_db": 1.206773833647921, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13751291939715374, "3000–6000 Hz_40ms_envelope_correlation": 0.9855025384690816, "3000–6000 Hz_40ms_modulation_mae_db": 0.9056888524039605, "mr_stft_db": 5.875552312238425, "band_mae_db": 0.284727034259367, "rms_gain_error_db": 0.025859643787416065, "output_peak": 0.400508758361743}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 40080, "output_samples": 40080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9347295964530735, "spectral_convergence": 0.30415234637465854, "spectral_convergence_status": "ok", "amplitude_floor": 1.4347408585118857e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.685881277936386, "spectral_convergence": 0.298666767393146, "spectral_convergence_status": "ok", "amplitude_floor": 9.356428468567288e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.929182927421703, "spectral_convergence": 0.30561145407565643, "spectral_convergence_status": "ok", "amplitude_floor": 6.30553407186755e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020558767458207216, "output_power": 0.00026450861626878096, "delta_db": 1.0944274935447702, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002656133227213937, "output_power": 0.00250620273700984, "delta_db": -0.2523365478982321, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000658897280424305, "output_power": 0.000610006711628791, "delta_db": -0.3348310163239773, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.296049498403337e-05, "output_power": 6.538833742794834e-05, "delta_db": 0.16432161368129378, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018191885630194454, "output_power": 0.0001788598924940608, "delta_db": -0.07364751626102044, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012525230408717123, "output_power": 0.00011828621151547096, "delta_db": -0.24851601324580647, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3613467001591834, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16784183950268122, "pitch": null, "output_rms_dbfs": -24.267508759543883, "output_peak": 0.3920935910531185, "reference_rms_dbfs": -24.099666920041205, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.849931267270388}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 40080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.0001593748762509214, "log_rms_mae_db": 0.913934290426132, "rms_envelope_nrmse": 0.0875505793279052, "envelope_correlation": 0.9931073540048588, "rms_gain_error_db": -0.167831021552311, "reference_mean_power": 0.003898496329889168, "output_mean_power": 0.0037507146738366866}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.3730898670333144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011160072986886684, "output_power": 0.0010949494964658517, "reference_active": true, "delta_db": -0.08272946690865986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003924601048841299, "output_power": 0.0003689323259473651, "reference_active": true, "delta_db": -0.2684880568924627}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.521600297301381e-05, "output_power": 8.997473307142143e-05, "reference_active": true, "delta_db": 0.23599406992321176}], "active_band_count": 3, "active_band_mae_db": 0.19573719790811142}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00014354118626608257, "log_rms_mae_db": 0.47007641836219555, "rms_envelope_nrmse": 0.047383690585783816, "envelope_correlation": 0.9980595536733371, "rms_gain_error_db": -0.16781842830287577, "reference_mean_power": 0.00394559944963645, "output_mean_power": 0.0037960432466253536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 2.0339309551628037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010452646972319804, "output_power": 0.0010343032226456186, "reference_active": true, "delta_db": -0.04578404936866319}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020675655639960477, "output_power": 0.00019591414063318237, "reference_active": true, "delta_db": -0.2339350672902429}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.033282667633223e-06, "output_power": 6.154268480877515e-06, "reference_active": true, "delta_db": 0.0862276539615888}], "active_band_count": 3, "active_band_mae_db": 0.12198225687349828}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.00014555177137634006, "log_rms_mae_db": 1.5556305302874152, "rms_envelope_nrmse": 0.10913948466427127, "envelope_correlation": 0.9915791721408826, "rms_gain_error_db": -0.3321868938653606, "reference_mean_power": 0.0025948122953778335, "output_mean_power": 0.0024037387117830005}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.091667884405576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010721637846984168, "output_power": 0.0010256055647286616, "reference_active": true, "delta_db": -0.19280765062684915}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00045198280410970506, "output_power": 0.0003791552490963144, "reference_active": true, "delta_db": -0.7630483937722647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.681659745338991e-05, "output_power": 5.362149221239619e-05, "reference_active": true, "delta_db": -0.25136326160114175}], "active_band_count": 3, "active_band_mae_db": 0.4024064353334185}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00013266563413733552, "log_rms_mae_db": 0.8911321192604691, "rms_envelope_nrmse": 0.06802945771770486, "envelope_correlation": 0.9971747658784741, "rms_gain_error_db": -0.3321860034493556, "reference_mean_power": 0.002626297782108314, "output_mean_power": 0.0024329062079302163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 1.815151890487515e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010442677388382842, "output_power": 0.000997675643378976, "reference_active": true, "delta_db": -0.198224915454441}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002292871431701824, "output_power": 0.0001962170087679876, "reference_active": true, "delta_db": -0.67643052374486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.238038641621491e-06, "output_power": 5.545212731195463e-06, "reference_active": true, "delta_db": -1.157026856809977}], "active_band_count": 3, "active_band_mae_db": 0.6772274320030927}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.977119603351369e-05, "log_rms_mae_db": 1.5223030011191958, "rms_envelope_nrmse": 0.20171144304993013, "envelope_correlation": 0.9628350521038813, "rms_gain_error_db": -0.049928087849628794, "reference_mean_power": 0.0001519473982974274, "output_mean_power": 0.00015021055803558274}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 8.081548208249742e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5556653827814076e-05, "output_power": 4.6074141871035115e-05, "reference_active": true, "delta_db": 0.049054379438869124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1818343242486535e-05, "output_power": 1.227973267843268e-05, "reference_active": true, "delta_db": 0.16632313480072553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7000790662746577e-06, "output_power": 6.169741332312568e-06, "reference_active": true, "delta_db": 2.2205595200749215}], "active_band_count": 3, "active_band_mae_db": 0.8119790114381721}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.109627865595404e-05, "log_rms_mae_db": 0.7838416498782848, "rms_envelope_nrmse": 0.10033424566016322, "envelope_correlation": 0.989728023646444, "rms_gain_error_db": -0.04987387808529683, "reference_mean_power": 0.0001537098951112271, "output_mean_power": 0.0001519548052929366}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 6.838335178235257e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3201238570188224e-05, "output_power": 4.51013687882774e-05, "reference_active": true, "delta_db": 0.18693524424541227}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.064245816738516e-06, "output_power": 6.552729393538806e-06, "reference_active": true, "delta_db": 0.33645436084455443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0299327724151498e-07, "output_power": 3.184511828536049e-07, "reference_active": true, "delta_db": 1.9556121123741317}], "active_band_count": 3, "active_band_mae_db": 0.826333905821366}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.882145378793849e-05, "log_rms_mae_db": 1.72944209663637, "rms_envelope_nrmse": 0.1944704611804575, "envelope_correlation": 0.9717356313012231, "rms_gain_error_db": -0.13420265356318006, "reference_mean_power": 0.00013710154501992013, "output_mean_power": 0.0001329297180509417}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 1.0763544842397476e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5103901227352814e-05, "output_power": 4.377887387087235e-05, "reference_active": true, "delta_db": -0.12949521503305372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.11132679073469e-05, "output_power": 2.3560001682290596e-05, "reference_active": true, "delta_db": -1.2077031145629271}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.435692144349255e-06, "output_power": 5.042735883359921e-06, "reference_active": true, "delta_db": -0.32588629778475386}], "active_band_count": 3, "active_band_mae_db": 0.5543615424602449}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.3950747306848176e-05, "log_rms_mae_db": 1.0968901142184686, "rms_envelope_nrmse": 0.13038033758063935, "envelope_correlation": 0.9864358246240715, "rms_gain_error_db": -0.13375399545234334, "reference_mean_power": 0.00013864857515666164, "output_mean_power": 0.00013444356230369803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 9.21368112936246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.381631867120006e-05, "output_power": 4.2757217041702404e-05, "reference_active": true, "delta_db": -0.10626456172038679}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.911163628753255e-05, "output_power": 1.4110525834618805e-05, "reference_active": true, "delta_db": -1.3175467356622974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5242776229186804e-07, "output_power": 3.981510043796692e-07, "reference_active": true, "delta_db": 0.5297770323659026}], "active_band_count": 3, "active_band_mae_db": 0.6511961099161956}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8752288624879712, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "64bb53a63c32fedc8b83d92d4c3c2db3bf454661eca927ca598190d7fe99ea9a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.962239485999463, "standard_deviation": 0.004786679439221929, "minimum": 0.9531224987215121, "maximum": 0.9672891065279183, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5517825733467978, "standard_deviation": 0.04673652843291209, "minimum": 1.4915225440673907, "maximum": 1.6280399172931532, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.37069857785657145, "standard_deviation": 0.22278174022132544, "minimum": 0.059124332174102824, "maximum": 0.8119790114381721, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20398700805099246, "standard_deviation": 0.013832646973871704, "minimum": 0.188392114221574, "maximum": 0.22553198420422182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9882633884702856, "standard_deviation": 0.0019593294341968617, "minimum": 0.9848555899118533, "maximum": 0.990370613862301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.880752600130164, "standard_deviation": 0.07767256183171277, "minimum": 0.7838416498782848, "maximum": 1.0196915184856434, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4533766811612743, "standard_deviation": 0.2932226395524967, "minimum": 0.08171200967739271, "maximum": 0.8858436106311913, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10936617965281578, "standard_deviation": 0.010132909805670606, "minimum": 0.09873019141323129, "maximum": 0.12424074028590323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9730463745976627, "standard_deviation": 0.002085579696758779, "minimum": 0.9701264579579439, "maximum": 0.9771504212922771, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.76557443518981, "standard_deviation": 0.04531764883404173, "minimum": 1.72944209663637, "maximum": 1.8370055235060168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7162552861965665, "standard_deviation": 0.19097679547604154, "minimum": 0.4845672375284318, "maximum": 1.0747178702393079, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19232540442245466, "standard_deviation": 0.006427282872084755, "minimum": 0.18060229170256192, "maximum": 0.2024747691435689, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9868848021090193, "standard_deviation": 0.0009104873565417312, "minimum": 0.9855025384690816, "maximum": 0.98807677975474, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1179842181271313, "standard_deviation": 0.06219037980982266, "minimum": 1.0424717271872517, "maximum": 1.2188344684059458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5394718725290136, "standard_deviation": 0.24599359393916478, "minimum": 0.28656688993263585, "maximum": 0.9056888524039605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12985943218164714, "standard_deviation": 0.0046280539937540565, "minimum": 0.12281932753395769, "maximum": 0.13751291939715374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9863268769129688, "standard_deviation": 0.004346598030278212, "minimum": 0.9801038330019053, "maximum": 0.9923143951195108, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.672045163954333, "standard_deviation": 0.07204231540828568, "minimum": 1.5556305302874152, "maximum": 1.7986010484727613, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4830394307290341, "standard_deviation": 0.24287677033397015, "minimum": 0.17307701752922744, "maximum": 0.79821828791057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13265494931972302, "standard_deviation": 0.020817323319724582, "minimum": 0.10088660345172243, "maximum": 0.1600639768865344, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9945830637351933, "standard_deviation": 0.002941957374832744, "minimum": 0.9882896990250251, "maximum": 0.9972049156730055, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9425405442851671, "standard_deviation": 0.04627282770955251, "minimum": 0.8739774181554477, "maximum": 1.001050054588703, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6308951929340869, "standard_deviation": 0.34530255871932414, "minimum": 0.08363565068503187, "maximum": 1.1836698046633465, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08106938054085186, "standard_deviation": 0.018927551568157498, "minimum": 0.060262120641350586, "maximum": 0.11742421046909453, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.36607289431552176, "standard_deviation": 0.06269455269426556, "minimum": 0.284727034259367, "maximum": 0.4720124772922758, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.8655240619103495, "standard_deviation": 0.013169596442136673, "minimum": 5.842920284623438, "maximum": 5.880732966730456, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4305446238905957, "standard_deviation": 0.05871957536227596, "minimum": 0.38788909078273115, "maximum": 0.568673144851044, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.043154309144432856, "standard_deviation": 0.09304176593150192, "minimum": -0.16784183950268122, "maximum": 0.10426442031434668, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901184863495522, "standard_deviation": 0.0028081814968412356, "minimum": 0.986870915364858, "maximum": 0.994395914542328, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9524201806557291, "standard_deviation": 0.03997673034278359, "minimum": 0.9028240067708355, "maximum": 1.005569279117364, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40522690027456476, "standard_deviation": 0.21409371676449832, "minimum": 0.13794682793273935, "maximum": 0.788937455242065, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1027556242863517, "standard_deviation": 0.01522059468257832, "minimum": 0.07778637220922252, "maximum": 0.11928611186923659, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.996790476944865, "standard_deviation": 0.0014843460609825507, "minimum": 0.9940630553932104, "maximum": 0.9982845299789382, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5064603126644343, "standard_deviation": 0.0353803682434326, "minimum": 0.47007641836219555, "maximum": 0.5563876438910037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.45030525022076195, "standard_deviation": 0.2734384746020813, "minimum": 0.12198225687349828, "maximum": 1.0271272156309632, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.056149847695330116, "standard_deviation": 0.01210526522057964, "minimum": 0.041099331471404925, "maximum": 0.0766076626706861, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0038", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.3890063762664795, "predicted_mos": 2.3890063762664795, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "e3982edfb8a2cf6bce281926d09a97a6fcaee2a5b2683317b5672550628f055e", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.159047699999064, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.007687489491291, "全帯域_10ms_rms_envelope_nrmse": 0.11346671476323436, "全帯域_10ms_envelope_correlation": 0.9881931076807452, "全帯域_10ms_modulation_mae_db": 0.18874801601922178, "全帯域_40ms_log_rms_mae_db": 0.508842082424781, "全帯域_40ms_rms_envelope_nrmse": 0.06255691072720598, "全帯域_40ms_envelope_correlation": 0.996102550632068, "全帯域_40ms_modulation_mae_db": 0.13452948885732033, "300–1000 Hz_10ms_log_rms_mae_db": 1.6972759415750074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14332821619838906, "300–1000 Hz_10ms_envelope_correlation": 0.9840444960413663, "300–1000 Hz_10ms_modulation_mae_db": 0.5315429972261495, "300–1000 Hz_40ms_log_rms_mae_db": 0.8794730092798639, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08115374115723464, "300–1000 Hz_40ms_envelope_correlation": 0.994427903615812, "300–1000 Hz_40ms_modulation_mae_db": 0.441582357890765, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6041270327228594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21455043390362794, "1000–3000 Hz_10ms_envelope_correlation": 0.9593765787085083, "1000–3000 Hz_10ms_modulation_mae_db": 0.2594553683532083, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8921744221054906, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11983256780360667, "1000–3000 Hz_40ms_envelope_correlation": 0.9861210164490547, "1000–3000 Hz_40ms_modulation_mae_db": 0.16254310723543125, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7503468126138164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20179215726045102, "3000–6000 Hz_10ms_envelope_correlation": 0.9719561368628278, "3000–6000 Hz_10ms_modulation_mae_db": 0.8435095584656157, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1403605376169348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14413193939722624, "3000–6000 Hz_40ms_envelope_correlation": 0.9850490275916234, "3000–6000 Hz_40ms_modulation_mae_db": 0.5887463963971435, "mr_stft_db": 5.838815558757896, "band_mae_db": 0.4502364254888858, "rms_gain_error_db": 0.061517814736902696, "output_peak": 0.4792435695116444}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.944174485666508, "全帯域_10ms_rms_envelope_nrmse": 0.1193362604555847, "全帯域_10ms_envelope_correlation": 0.986751933560063, "全帯域_10ms_modulation_mae_db": 0.9510212510767282, "全帯域_40ms_log_rms_mae_db": 0.48500124850639154, "全帯域_40ms_rms_envelope_nrmse": 0.066212523761611, "全帯域_40ms_envelope_correlation": 0.9956113637063335, "全帯域_40ms_modulation_mae_db": 1.0120441004068343, "300–1000 Hz_10ms_log_rms_mae_db": 1.676428483621803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16657890292896224, "300–1000 Hz_10ms_envelope_correlation": 0.9785898533834549, "300–1000 Hz_10ms_modulation_mae_db": 1.0852086348844672, "300–1000 Hz_40ms_log_rms_mae_db": 0.9652660291124823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10764974765008865, "300–1000 Hz_40ms_envelope_correlation": 0.9906917789360135, "300–1000 Hz_40ms_modulation_mae_db": 1.4124774674891143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5900587473497065, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20762038022113993, "1000–3000 Hz_10ms_envelope_correlation": 0.960958462761134, "1000–3000 Hz_10ms_modulation_mae_db": 0.2569308521502577, "1000–3000 Hz_40ms_log_rms_mae_db": 0.782763696268648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10588950334449203, "1000–3000 Hz_40ms_envelope_correlation": 0.9887077997284289, "1000–3000 Hz_40ms_modulation_mae_db": 0.3027749882975703, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7224623850208973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18702775380102438, "3000–6000 Hz_10ms_envelope_correlation": 0.9740577374290045, "3000–6000 Hz_10ms_modulation_mae_db": 1.264827599361637, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1031604029528637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12575498427378382, "3000–6000 Hz_40ms_envelope_correlation": 0.9873462124582624, "3000–6000 Hz_40ms_modulation_mae_db": 1.113769629946496, "mr_stft_db": 5.855056154085802, "band_mae_db": 0.2935848676498779, "rms_gain_error_db": -0.09284413093627866, "output_peak": 0.5413028230303593}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9426108436295023, "全帯域_10ms_rms_envelope_nrmse": 0.10573112519189348, "全帯域_10ms_envelope_correlation": 0.9896261928665673, "全帯域_10ms_modulation_mae_db": 1.0437705795046786, "全帯域_40ms_log_rms_mae_db": 0.45993172084286943, "全帯域_40ms_rms_envelope_nrmse": 0.054070609907100685, "全帯域_40ms_envelope_correlation": 0.9971049578503557, "全帯域_40ms_modulation_mae_db": 1.0309053149333, "300–1000 Hz_10ms_log_rms_mae_db": 1.577315231550591, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1533740892738362, "300–1000 Hz_10ms_envelope_correlation": 0.9819169719629641, "300–1000 Hz_10ms_modulation_mae_db": 1.1418813017346563, "300–1000 Hz_40ms_log_rms_mae_db": 0.8993648590356207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09082751565854998, "300–1000 Hz_40ms_envelope_correlation": 0.9935303539717618, "300–1000 Hz_40ms_modulation_mae_db": 1.3125982204018742, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5126710469482565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21643815233317149, "1000–3000 Hz_10ms_envelope_correlation": 0.9588161908308159, "1000–3000 Hz_10ms_modulation_mae_db": 0.5301918846578609, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8375343066129054, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1203256062457268, "1000–3000 Hz_40ms_envelope_correlation": 0.9861020846324671, "1000–3000 Hz_40ms_modulation_mae_db": 0.6003746456857647, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7367479987648053, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19315792712310642, "3000–6000 Hz_10ms_envelope_correlation": 0.9723294548979646, "3000–6000 Hz_10ms_modulation_mae_db": 1.4306544080597912, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1285670036465671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13425518304638673, "3000–6000 Hz_40ms_envelope_correlation": 0.9861530303459473, "3000–6000 Hz_40ms_modulation_mae_db": 1.1458911524036584, "mr_stft_db": 5.8513250256810885, "band_mae_db": 0.4219150062853512, "rms_gain_error_db": -0.0872453538368769, "output_peak": 0.4711833546493738}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9645211895926309, "全帯域_10ms_rms_envelope_nrmse": 0.11769257556597279, "全帯域_10ms_envelope_correlation": 0.9872717803232093, "全帯域_10ms_modulation_mae_db": 0.5660926149999287, "全帯域_40ms_log_rms_mae_db": 0.5211034648406225, "全帯域_40ms_rms_envelope_nrmse": 0.06108151134372733, "全帯域_40ms_envelope_correlation": 0.9965347956186997, "全帯域_40ms_modulation_mae_db": 0.6945078443611408, "300–1000 Hz_10ms_log_rms_mae_db": 1.5961364682349273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13464978585887577, "300–1000 Hz_10ms_envelope_correlation": 0.9864781129375798, "300–1000 Hz_10ms_modulation_mae_db": 0.6901092562510013, "300–1000 Hz_40ms_log_rms_mae_db": 0.970123959847036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08508578068293396, "300–1000 Hz_40ms_envelope_correlation": 0.994752618370807, "300–1000 Hz_40ms_modulation_mae_db": 1.0401968106244672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4832295877653543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19678257290160264, "1000–3000 Hz_10ms_envelope_correlation": 0.9646411858156125, "1000–3000 Hz_10ms_modulation_mae_db": 0.2261192688506489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8481128887193276, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09985714401873783, "1000–3000 Hz_40ms_envelope_correlation": 0.9898538634425721, "1000–3000 Hz_40ms_modulation_mae_db": 0.37447761120130263, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7252995968181082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1921940137903102, "3000–6000 Hz_10ms_envelope_correlation": 0.9726830957259714, "3000–6000 Hz_10ms_modulation_mae_db": 0.7082426318650294, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1112723238381281, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1251198931930374, "3000–6000 Hz_40ms_envelope_correlation": 0.98753612100649, "3000–6000 Hz_40ms_modulation_mae_db": 0.4331074999499516, "mr_stft_db": 5.817310504248593, "band_mae_db": 0.30429903738472586, "rms_gain_error_db": -0.1711671736404119, "output_peak": 0.4107872590618493}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8840968096485385, "全帯域_10ms_rms_envelope_nrmse": 0.08801907116740222, "全帯域_10ms_envelope_correlation": 0.9928319112640219, "全帯域_10ms_modulation_mae_db": 0.23286252133386146, "全帯域_40ms_log_rms_mae_db": 0.4506320062841193, "全帯域_40ms_rms_envelope_nrmse": 0.051253652325153894, "全帯域_40ms_envelope_correlation": 0.9973390936546583, "全帯域_40ms_modulation_mae_db": 0.16747583669427382, "300–1000 Hz_10ms_log_rms_mae_db": 1.5328217533610387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12200603060306728, "300–1000 Hz_10ms_envelope_correlation": 0.9884674554692111, "300–1000 Hz_10ms_modulation_mae_db": 0.12750212788792623, "300–1000 Hz_40ms_log_rms_mae_db": 0.8506720964665867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07503682700657337, "300–1000 Hz_40ms_envelope_correlation": 0.9952545120517804, "300–1000 Hz_40ms_modulation_mae_db": 0.02810628939292865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4423899806267613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19226651779644444, "1000–3000 Hz_10ms_envelope_correlation": 0.9661968571434781, "1000–3000 Hz_10ms_modulation_mae_db": 0.2009589177819007, "1000–3000 Hz_40ms_log_rms_mae_db": 0.77268858895183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0949343885727186, "1000–3000 Hz_40ms_envelope_correlation": 0.9908101419663734, "1000–3000 Hz_40ms_modulation_mae_db": 0.2802823965846031, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6603122317567314, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18679217908116919, "3000–6000 Hz_10ms_envelope_correlation": 0.9744703713015703, "3000–6000 Hz_10ms_modulation_mae_db": 0.9646848070600686, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0547884676944606, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12553504147343106, "3000–6000 Hz_40ms_envelope_correlation": 0.9876103341072652, "3000–6000 Hz_40ms_modulation_mae_db": 0.7179293529453233, "mr_stft_db": 5.837030155848325, "band_mae_db": 0.2355807116850497, "rms_gain_error_db": 0.006859357593481142, "output_peak": 0.4800274722900792}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9493478672781667, "全帯域_10ms_rms_envelope_nrmse": 0.09757037181268598, "全帯域_10ms_envelope_correlation": 0.9911817321407184, "全帯域_10ms_modulation_mae_db": 0.5141421039700331, "全帯域_40ms_log_rms_mae_db": 0.4946208216784995, "全帯域_40ms_rms_envelope_nrmse": 0.05174500632028685, "全帯域_40ms_envelope_correlation": 0.9972813242122293, "全帯域_40ms_modulation_mae_db": 0.50517209647002, "300–1000 Hz_10ms_log_rms_mae_db": 1.6062448590291711, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13345642432608798, "300–1000 Hz_10ms_envelope_correlation": 0.9863284234288421, "300–1000 Hz_10ms_modulation_mae_db": 0.6898212250944443, "300–1000 Hz_40ms_log_rms_mae_db": 0.8781623091414053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07339031079863333, "300–1000 Hz_40ms_envelope_correlation": 0.9957945999234955, "300–1000 Hz_40ms_modulation_mae_db": 0.7190367896295559, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4703258414315346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22035854933179763, "1000–3000 Hz_10ms_envelope_correlation": 0.9567353378409876, "1000–3000 Hz_10ms_modulation_mae_db": 0.5715736188309929, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8185407404091174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11532753942750297, "1000–3000 Hz_40ms_envelope_correlation": 0.9869502675966807, "1000–3000 Hz_40ms_modulation_mae_db": 0.4251945162056419, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7186916522126876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20913530551824758, "3000–6000 Hz_10ms_envelope_correlation": 0.9673881683127753, "3000–6000 Hz_10ms_modulation_mae_db": 0.669610544575287, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0856286734244207, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15383188706653816, "3000–6000 Hz_40ms_envelope_correlation": 0.9816194467846896, "3000–6000 Hz_40ms_modulation_mae_db": 0.40416207967835954, "mr_stft_db": 5.868291134120902, "band_mae_db": 0.3642238064813119, "rms_gain_error_db": -0.002121429198438623, "output_peak": 0.38167779541005614}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9244153167653546, "全帯域_10ms_rms_envelope_nrmse": 0.0875224103254128, "全帯域_10ms_envelope_correlation": 0.9929558937658717, "全帯域_10ms_modulation_mae_db": 0.4825663660581146, "全帯域_40ms_log_rms_mae_db": 0.49821028600310663, "全帯域_40ms_rms_envelope_nrmse": 0.046535185656989636, "全帯域_40ms_envelope_correlation": 0.9979246565866392, "全帯域_40ms_modulation_mae_db": 0.5881470952567883, "300–1000 Hz_10ms_log_rms_mae_db": 1.6561141324153168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12313240002796326, "300–1000 Hz_10ms_envelope_correlation": 0.9884412115264406, "300–1000 Hz_10ms_modulation_mae_db": 0.6295381091157889, "300–1000 Hz_40ms_log_rms_mae_db": 0.8928880122992067, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07380214583558685, "300–1000 Hz_40ms_envelope_correlation": 0.995760219008986, "300–1000 Hz_40ms_modulation_mae_db": 0.8610111464191187, "1000–3000 Hz_10ms_log_rms_mae_db": 1.549493304035292, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21710378170710634, "1000–3000 Hz_10ms_envelope_correlation": 0.9565385595587126, "1000–3000 Hz_10ms_modulation_mae_db": 0.32856688226880465, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8951417301063925, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13063341365944264, "1000–3000 Hz_40ms_envelope_correlation": 0.982621767139278, "1000–3000 Hz_40ms_modulation_mae_db": 0.36545099091006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7554033406875282, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20037448486228074, "3000–6000 Hz_10ms_envelope_correlation": 0.9702434887813668, "3000–6000 Hz_10ms_modulation_mae_db": 1.4113602123871811, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1103974732303559, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1410000338174377, "3000–6000 Hz_40ms_envelope_correlation": 0.9848189121358797, "3000–6000 Hz_40ms_modulation_mae_db": 1.1613450629637139, "mr_stft_db": 5.870566314908711, "band_mae_db": 0.3836482326187179, "rms_gain_error_db": -0.10651539040124963, "output_peak": 0.36428471453810574}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9110286592692163, "全帯域_10ms_rms_envelope_nrmse": 0.10329124261238859, "全帯域_10ms_envelope_correlation": 0.9902687224958093, "全帯域_10ms_modulation_mae_db": 0.12396661868406515, "全帯域_40ms_log_rms_mae_db": 0.5057909998407986, "全帯域_40ms_rms_envelope_nrmse": 0.06090721107108066, "全帯域_40ms_envelope_correlation": 0.9965316031482521, "全帯域_40ms_modulation_mae_db": 0.14722436297477737, "300–1000 Hz_10ms_log_rms_mae_db": 1.6944954637106395, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13339961991404042, "300–1000 Hz_10ms_envelope_correlation": 0.9864895349617675, "300–1000 Hz_10ms_modulation_mae_db": 0.5640418564523352, "300–1000 Hz_40ms_log_rms_mae_db": 1.0043024546822883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07412519380957669, "300–1000 Hz_40ms_envelope_correlation": 0.9959260873812645, "300–1000 Hz_40ms_modulation_mae_db": 0.2776547462056968, "1000–3000 Hz_10ms_log_rms_mae_db": 1.431061309350689, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17239521309454428, "1000–3000 Hz_10ms_envelope_correlation": 0.9727132860227253, "1000–3000 Hz_10ms_modulation_mae_db": 0.38151405260039173, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7498711214286813, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08812595962580688, "1000–3000 Hz_40ms_envelope_correlation": 0.9923770409298884, "1000–3000 Hz_40ms_modulation_mae_db": 0.47157183029695976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8238456695798455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20397030282035108, "3000–6000 Hz_10ms_envelope_correlation": 0.9695138291411921, "3000–6000 Hz_10ms_modulation_mae_db": 1.306234238962537, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2057843898098477, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15021849667102732, "3000–6000 Hz_40ms_envelope_correlation": 0.9832060108350346, "3000–6000 Hz_40ms_modulation_mae_db": 0.9262464442024259, "mr_stft_db": 5.847058328916286, "band_mae_db": 0.38038307031215496, "rms_gain_error_db": -0.1642926438093003, "output_peak": 0.4012020661447918}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 40080, "output_samples": 40080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.940642496442655, "spectral_convergence": 0.32919892326394123, "spectral_convergence_status": "ok", "amplitude_floor": 1.4347408585118857e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.646051909540671, "spectral_convergence": 0.31846312407898736, "spectral_convergence_status": "ok", "amplitude_floor": 9.356428468567288e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9297522702903604, "spectral_convergence": 0.33749087471443495, "spectral_convergence_status": "ok", "amplitude_floor": 6.30553407186755e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020558767458207216, "output_power": 0.0002843468011613521, "delta_db": 1.4085127267206023, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002656133227213937, "output_power": 0.0025745765423458335, "delta_db": -0.13544046768013918, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000658897280424305, "output_power": 0.0006952726354457931, "delta_db": 0.23337421735318345, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.296049498403337e-05, "output_power": 6.899388779685614e-05, "delta_db": 0.39742484366127073, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018191885630194454, "output_power": 0.00020067674407958278, "delta_db": 0.426193291098135, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012525230408717123, "output_power": 0.00012238788311484718, "delta_db": -0.1004730064199845, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4502364254888858, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.061517814736902696, "pitch": null, "output_rms_dbfs": -24.0381491053043, "output_peak": 0.4792435695116444, "reference_rms_dbfs": -24.099666920041205, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.838815558757896}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 40080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.0001593748762509214, "log_rms_mae_db": 1.007687489491291, "rms_envelope_nrmse": 0.11346671476323436, "envelope_correlation": 0.9881931076807452, "rms_gain_error_db": 0.061530211301253995, "reference_mean_power": 0.003898496329889168, "output_mean_power": 0.003954122782833434}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.3730898670333144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011160072986886684, "output_power": 0.0010567528100754772, "reference_active": true, "delta_db": -0.23693623535739505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003924601048841299, "output_power": 0.00039001763544367293, "reference_active": true, "delta_db": -0.0271127059773095}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.521600297301381e-05, "output_power": 7.94880145349334e-05, "reference_active": true, "delta_db": -0.3021951067229608}], "active_band_count": 3, "active_band_mae_db": 0.18874801601922178}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00014354118626608257, "log_rms_mae_db": 0.508842082424781, "rms_envelope_nrmse": 0.06255691072720598, "envelope_correlation": 0.996102550632068, "rms_gain_error_db": 0.06153953700489486, "reference_mean_power": 0.00394559944963645, "output_mean_power": 0.004001906595979339}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 2.0339309551628037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010452646972319804, "output_power": 0.0009813129169400844, "reference_active": true, "delta_db": -0.2741876731817146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020675655639960477, "output_power": 0.0002057801825462515, "reference_active": true, "delta_db": -0.02055742087934956}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.033282667633223e-06, "output_power": 5.883954908126351e-06, "reference_active": true, "delta_db": -0.10884337251089682}], "active_band_count": 3, "active_band_mae_db": 0.13452948885732033}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.00014555177137634006, "log_rms_mae_db": 1.6972759415750074, "rms_envelope_nrmse": 0.14332821619838906, "envelope_correlation": 0.9840444960413663, "rms_gain_error_db": -0.06928226958613615, "reference_mean_power": 0.0025948122953778335, "output_mean_power": 0.0025537461222573496}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.091667884405576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010721637846984168, "output_power": 0.0009841536226044406, "reference_active": true, "delta_db": -0.3719823821276027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00045198280410970506, "output_power": 0.0004014273810183115, "reference_active": true, "delta_db": -0.5151492017867375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.681659745338991e-05, "output_power": 6.686902063637206e-05, "reference_active": true, "delta_db": 0.7074974077641082}], "active_band_count": 3, "active_band_mae_db": 0.5315429972261495}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00013266563413733552, "log_rms_mae_db": 0.8794730092798639, "rms_envelope_nrmse": 0.08115374115723464, "envelope_correlation": 0.994427903615812, "rms_gain_error_db": -0.06927709335226993, "reference_mean_power": 0.002626297782108314, "output_mean_power": 0.002584736392172561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 1.815151890487515e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010442677388382842, "output_power": 0.0009455980439064697, "reference_active": true, "delta_db": -0.431052961598416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002292871431701824, "output_power": 0.00020663088676922082, "reference_active": true, "delta_db": -0.45184463694435084}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.238038641621491e-06, "output_power": 6.537865018254382e-06, "reference_active": true, "delta_db": -0.44184947512952827}], "active_band_count": 3, "active_band_mae_db": 0.441582357890765}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.977119603351369e-05, "log_rms_mae_db": 1.6041270327228594, "rms_envelope_nrmse": 0.21455043390362794, "envelope_correlation": 0.9593765787085083, "rms_gain_error_db": 0.17700011898857915, "reference_mean_power": 0.0001519473982974274, "output_mean_power": 0.0001582680604634576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 8.081548208249742e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5556653827814076e-05, "output_power": 4.2582612923259794e-05, "reference_active": true, "delta_db": -0.2931951052456839}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1818343242486535e-05, "output_power": 1.1309738889858752e-05, "reference_active": true, "delta_db": -0.1910402072754863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7000790662746577e-06, "output_power": 3.959351726810166e-06, "reference_active": true, "delta_db": 0.29413079253845487}], "active_band_count": 3, "active_band_mae_db": 0.2594553683532083}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.109627865595404e-05, "log_rms_mae_db": 0.8921744221054906, "rms_envelope_nrmse": 0.11983256780360667, "envelope_correlation": 0.9861210164490547, "rms_gain_error_db": 0.1774275739225267, "reference_mean_power": 0.0001537098951112271, "output_mean_power": 0.0001601196321105224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 6.838335178235257e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3201238570188224e-05, "output_power": 4.0867372736504246e-05, "reference_active": true, "delta_db": -0.2411947924412702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.064245816738516e-06, "output_power": 5.901539729195096e-06, "reference_active": true, "delta_db": -0.11811462146458149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0299327724151498e-07, "output_power": 2.0908055579713674e-07, "reference_active": true, "delta_db": 0.12831990780044206}], "active_band_count": 3, "active_band_mae_db": 0.16254310723543125}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.882145378793849e-05, "log_rms_mae_db": 1.7503468126138164, "rms_envelope_nrmse": 0.20179215726045102, "envelope_correlation": 0.9719561368628278, "rms_gain_error_db": 0.2935843698634038, "reference_mean_power": 0.00013710154501992013, "output_mean_power": 0.00014669009368438414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 1.0763544842397476e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5103901227352814e-05, "output_power": 4.713482162434362e-05, "reference_active": true, "delta_db": 0.19127760418268872}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.11132679073469e-05, "output_power": 2.5688880484648613e-05, "reference_active": true, "delta_db": -0.8320045035427641}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.435692144349255e-06, "output_power": 3.841759689517473e-06, "reference_active": true, "delta_db": -1.5072465676713942}], "active_band_count": 3, "active_band_mae_db": 0.8435095584656157}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.3950747306848176e-05, "log_rms_mae_db": 1.1403605376169348, "rms_envelope_nrmse": 0.14413193939722624, "envelope_correlation": 0.9850490275916234, "rms_gain_error_db": 0.293578499578016, "reference_mean_power": 0.00013864857515666164, "output_mean_power": 0.0001483451188311916}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 9.21368112936246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.381631867120006e-05, "output_power": 4.586148493692074e-05, "reference_active": true, "delta_db": 0.19812226020467472}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.911163628753255e-05, "output_power": 1.5159544042168598e-05, "reference_active": true, "delta_db": -1.00611732690492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5242776229186804e-07, "output_power": 3.096493353669061e-07, "reference_active": true, "delta_db": -0.5619996020818362}], "active_band_count": 3, "active_band_mae_db": 0.5887463963971435}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8795836802585153, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fdbeaa6deb01a81d00327fb722587396982a11434b912d5073038c4abfe13abd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9619970573352468, "standard_deviation": 0.005542197601603215, "minimum": 0.9565385595587126, "maximum": 0.9727132860227253, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5104196062788064, "standard_deviation": 0.06538662351933026, "minimum": 1.431061309350689, "maximum": 1.6041270327228594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3444138556867582, "standard_deviation": 0.1399932315605993, "minimum": 0.2009589177819007, "maximum": 0.5715736188309929, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20468945016117934, "standard_deviation": 0.016489231788893853, "minimum": 0.17239521309454428, "maximum": 0.22035854933179763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9879429977355929, "standard_deviation": 0.003119687941741934, "minimum": 0.982621767139278, "maximum": 0.9923770409298884, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8246034368252991, "standard_deviation": 0.05388980975036631, "minimum": 0.7498711214286813, "maximum": 0.8951417301063925, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3728337608021667, "standard_deviation": 0.13202762833367987, "minimum": 0.16254310723543125, "maximum": 0.6003746456857647, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10936576533725431, "standard_deviation": 0.014537733861867356, "minimum": 0.08812595962580688, "maximum": 0.13063341365944264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9715802853065841, "standard_deviation": 0.0023916577421914556, "minimum": 0.9673881683127753, "maximum": 0.9744703713015703, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7366387109318024, "standard_deviation": 0.045704737214946695, "minimum": 1.6603122317567314, "maximum": 1.8238456695798455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0748905000921434, "standard_deviation": 0.31481077973049404, "minimum": 0.669610544575287, "maximum": 1.4306544080597912, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19680551553211756, "standard_deviation": 0.008206307823517512, "minimum": 0.18679217908116919, "maximum": 0.20913530551824758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.985417386908149, "standard_deviation": 0.0021833128440570076, "minimum": 0.9816194467846896, "maximum": 0.9876103341072652, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1174949090266972, "standard_deviation": 0.04421148468887489, "minimum": 1.0547884676944606, "maximum": 1.2057843898098477, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8113997023108841, "standard_deviation": 0.3176301320242925, "minimum": 0.40416207967835954, "maximum": 1.1613450629637139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13748093236735856, "standard_deviation": 0.011524025521775307, "minimum": 0.1251198931930374, "maximum": 0.15383188706653816, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9850945074639534, "standard_deviation": 0.0034115350821526825, "minimum": 0.9785898533834549, "maximum": 0.9884674554692111, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6296040416873119, "standard_deviation": 0.06029967643354856, "minimum": 1.5328217533610387, "maximum": 1.6972759415750074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6824556885808462, "standard_deviation": 0.3209283094839729, "minimum": 0.12750212788792623, "maximum": 1.1418813017346563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13874068364140277, "standard_deviation": 0.015151621323222658, "minimum": 0.12200603060306728, "maximum": 0.16657890292896224, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.99451725915749, "standard_deviation": 0.0017475103335815871, "minimum": 0.9906917789360135, "maximum": 0.9959260873812645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9175315912330613, "standard_deviation": 0.05474569879301467, "minimum": 0.8506720964665867, "maximum": 1.0043024546822883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7615829785066901, "standard_deviation": 0.491525064538069, "minimum": 0.02810628939292865, "maximum": 1.4124774674891143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08263390782489718, "standard_deviation": 0.011920498445605636, "minimum": 0.07339031079863333, "maximum": 0.10764974765008865, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3542338947382594, "standard_deviation": 0.0714106555453261, "minimum": 0.2355807116850497, "maximum": 0.4502364254888858, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.84818164707095, "standard_deviation": 0.017439582109238773, "minimum": 5.817310504248593, "maximum": 5.870566314908711, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44121363182953244, "standard_deviation": 0.06077838494384569, "minimum": 0.36428471453810574, "maximum": 0.5413028230303593, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06947611868652154, "standard_deviation": 0.08374120950554068, "minimum": -0.1711671736404119, "maximum": 0.061517814736902696, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9898851592621257, "standard_deviation": 0.0023760090988110593, "minimum": 0.986751933560063, "maximum": 0.9929558937658717, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.940985332667651, "standard_deviation": 0.03682803864558868, "minimum": 0.8840968096485385, "maximum": 1.007687489491291, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.512896258955829, "standard_deviation": 0.34064845776754327, "minimum": 0.12396661868406515, "maximum": 1.0437705795046786, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10407872148682186, "standard_deviation": 0.012444143316368136, "minimum": 0.0875224103254128, "maximum": 0.1193362604555847, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968037931761545, "standard_deviation": 0.0007484931225938278, "minimum": 0.9956113637063335, "maximum": 0.9979246565866392, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49051657880264854, "standard_deviation": 0.024311546605054985, "minimum": 0.4506320062841193, "maximum": 0.5211034648406225, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5350007674943069, "standard_deviation": 0.36809758229929146, "minimum": 0.13452948885732033, "maximum": 1.0309053149333, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05679532638914451, "standard_deviation": 0.006825047247555726, "minimum": 0.046535185656989636, "maximum": 0.066212523761611, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_6829-68769-0041", "language": "en", "style": "normal", "speaker": "librispeech_6829", "source_utterance_id": "6829-68769-0041", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 42560, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_6829-68769-0041.wav", "source_file_sha256": "34382d3f8904c22b2a772d7ac182bbba66481ef0f41e5042d10685687b7fdfda", "decoded_pcm_sha256": "d14d8bc17d20f03134a1360b706d2d290d10240225382c5adf8b0919adb152e4", "original_source_sha256": "e0e3964567d7a515d2443c6e1eff105f88da0e87ef44cc2954b36f1b9d93096c", "corpus_entry_id": "librispeech_6829-68769-0041"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.9850518703460693, "predicted_mos": 3.9850518703460693, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "00d848fcbd4f0e18f07397e09d5c2748448fdd93e0274aa0188591272042e61a", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6510771560024295, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_6829-68769-0041", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "00d848fcbd4f0e18f07397e09d5c2748448fdd93e0274aa0188591272042e61a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0d776c98d75836f78dcbf836cd371840458d509810ee9e94758fbd26584e9d1f"}, "rows": [{"case": "librispeech_6829-68769-0041", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.648960828781128, "predicted_mos": 2.648960828781128, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "fdf9ffa113f094ad8d0d9b43d0869afe9e38f96bc37769399fcabf506327e882", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2057805959993857, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_6829-68769-0041", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "fdf9ffa113f094ad8d0d9b43d0869afe9e38f96bc37769399fcabf506327e882", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8926628607658864, "全帯域_10ms_rms_envelope_nrmse": 0.10120697100522309, "全帯域_10ms_envelope_correlation": 0.9900878636523057, "全帯域_10ms_modulation_mae_db": 0.8654450913733713, "全帯域_40ms_log_rms_mae_db": 0.4895863075435125, "全帯域_40ms_rms_envelope_nrmse": 0.06690084149848115, "全帯域_40ms_envelope_correlation": 0.9954766212781037, "全帯域_40ms_modulation_mae_db": 0.9320266724738717, "300–1000 Hz_10ms_log_rms_mae_db": 1.3156708054214208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12330600208921773, "300–1000 Hz_10ms_envelope_correlation": 0.9877427067450785, "300–1000 Hz_10ms_modulation_mae_db": 1.0676522143670133, "300–1000 Hz_40ms_log_rms_mae_db": 0.8371745317657151, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09350923861102382, "300–1000 Hz_40ms_envelope_correlation": 0.9926375418439535, "300–1000 Hz_40ms_modulation_mae_db": 1.1146482368587283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4740122814180365, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21359777195626795, "1000–3000 Hz_10ms_envelope_correlation": 0.9636287172740816, "1000–3000 Hz_10ms_modulation_mae_db": 0.6636067158302192, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8166430740260922, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11764212166476995, "1000–3000 Hz_40ms_envelope_correlation": 0.9887675789105086, "1000–3000 Hz_40ms_modulation_mae_db": 0.7872304942527982, "3000–6000 Hz_10ms_log_rms_mae_db": 1.584682030368444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1877864905626322, "3000–6000 Hz_10ms_envelope_correlation": 0.979116962677872, "3000–6000 Hz_10ms_modulation_mae_db": 0.5956850733187948, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0207675871769166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13263038986288417, "3000–6000 Hz_40ms_envelope_correlation": 0.9898837483825969, "3000–6000 Hz_40ms_modulation_mae_db": 0.4488246842233883, "mr_stft_db": 5.5834811700686275, "band_mae_db": 0.27349626122588677, "rms_gain_error_db": -0.19279714925526867, "output_peak": 0.4914528355333998}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8447007993023435, "全帯域_10ms_rms_envelope_nrmse": 0.09381506871235909, "全帯域_10ms_envelope_correlation": 0.9912165081190606, "全帯域_10ms_modulation_mae_db": 0.7458616205772075, "全帯域_40ms_log_rms_mae_db": 0.4419522065092465, "全帯域_40ms_rms_envelope_nrmse": 0.04730785606335744, "全帯域_40ms_envelope_correlation": 0.9975874572422763, "全帯域_40ms_modulation_mae_db": 1.068120302855159, "300–1000 Hz_10ms_log_rms_mae_db": 1.2939739263335148, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10590242659523516, "300–1000 Hz_10ms_envelope_correlation": 0.9908410298530012, "300–1000 Hz_10ms_modulation_mae_db": 0.8841868715068611, "300–1000 Hz_40ms_log_rms_mae_db": 0.6923728141309003, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05961959272072854, "300–1000 Hz_40ms_envelope_correlation": 0.997121019215108, "300–1000 Hz_40ms_modulation_mae_db": 1.226681856048032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5003331301481213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1908355970229226, "1000–3000 Hz_10ms_envelope_correlation": 0.9731487521216571, "1000–3000 Hz_10ms_modulation_mae_db": 0.3749941036195543, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8817497275666919, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12660496391792656, "1000–3000 Hz_40ms_envelope_correlation": 0.9875438369322248, "1000–3000 Hz_40ms_modulation_mae_db": 0.47361018054676024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.56313636165799, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16287468433702, "3000–6000 Hz_10ms_envelope_correlation": 0.9819726587482277, "3000–6000 Hz_10ms_modulation_mae_db": 0.6874831741043878, "3000–6000 Hz_40ms_log_rms_mae_db": 1.053719081659529, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10932982981963864, "3000–6000 Hz_40ms_envelope_correlation": 0.9914209641537165, "3000–6000 Hz_40ms_modulation_mae_db": 0.5124753607279001, "mr_stft_db": 5.610939080780141, "band_mae_db": 0.10804044848720935, "rms_gain_error_db": -0.08186836395143413, "output_peak": 0.6032725604308117}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9018713047142792, "全帯域_10ms_rms_envelope_nrmse": 0.0964108179070497, "全帯域_10ms_envelope_correlation": 0.9907266366217661, "全帯域_10ms_modulation_mae_db": 0.6746753923426551, "全帯域_40ms_log_rms_mae_db": 0.47532467239118964, "全帯域_40ms_rms_envelope_nrmse": 0.04354449090380165, "全帯域_40ms_envelope_correlation": 0.997967088279761, "全帯域_40ms_modulation_mae_db": 1.0249975187331477, "300–1000 Hz_10ms_log_rms_mae_db": 1.4401682901619426, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1113809722930202, "300–1000 Hz_10ms_envelope_correlation": 0.989628498924027, "300–1000 Hz_10ms_modulation_mae_db": 0.9128561660948766, "300–1000 Hz_40ms_log_rms_mae_db": 0.7578717386136248, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055532078720276316, "300–1000 Hz_40ms_envelope_correlation": 0.9972507537077244, "300–1000 Hz_40ms_modulation_mae_db": 1.0962162401648146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5961414713477746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19845518218092403, "1000–3000 Hz_10ms_envelope_correlation": 0.9688294149043923, "1000–3000 Hz_10ms_modulation_mae_db": 0.6435918602994496, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9580495942506163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11491310260913012, "1000–3000 Hz_40ms_envelope_correlation": 0.9894535946278491, "1000–3000 Hz_40ms_modulation_mae_db": 0.7038570916256934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5253264011509686, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1571778562481844, "3000–6000 Hz_10ms_envelope_correlation": 0.9836241591584834, "3000–6000 Hz_10ms_modulation_mae_db": 0.31638439177047323, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9744586615538137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10757091179014072, "3000–6000 Hz_40ms_envelope_correlation": 0.9919422823805485, "3000–6000 Hz_40ms_modulation_mae_db": 0.3377417873855713, "mr_stft_db": 5.570134396060507, "band_mae_db": 0.17419977208980222, "rms_gain_error_db": -0.07630921603211745, "output_peak": 0.5469001401798544}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9338859914035756, "全帯域_10ms_rms_envelope_nrmse": 0.0977601314478857, "全帯域_10ms_envelope_correlation": 0.9905028814311361, "全帯域_10ms_modulation_mae_db": 0.776480721491959, "全帯域_40ms_log_rms_mae_db": 0.5087811843167898, "全帯域_40ms_rms_envelope_nrmse": 0.050923682479515946, "全帯域_40ms_envelope_correlation": 0.9972688133332102, "全帯域_40ms_modulation_mae_db": 0.9287078471955196, "300–1000 Hz_10ms_log_rms_mae_db": 1.4417073611688547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11726037272119295, "300–1000 Hz_10ms_envelope_correlation": 0.9886045284100928, "300–1000 Hz_10ms_modulation_mae_db": 0.8685304480840174, "300–1000 Hz_40ms_log_rms_mae_db": 0.8131205314352772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06609499167752902, "300–1000 Hz_40ms_envelope_correlation": 0.9962724668189388, "300–1000 Hz_40ms_modulation_mae_db": 1.3797147016420712, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4940191446440696, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20803136904073433, "1000–3000 Hz_10ms_envelope_correlation": 0.9672813969356124, "1000–3000 Hz_10ms_modulation_mae_db": 0.5009430019106972, "1000–3000 Hz_40ms_log_rms_mae_db": 0.911521093806499, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14182047308098264, "1000–3000 Hz_40ms_envelope_correlation": 0.9856064789280627, "1000–3000 Hz_40ms_modulation_mae_db": 0.6937120748893105, "3000–6000 Hz_10ms_log_rms_mae_db": 1.595715417714754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1805336384328483, "3000–6000 Hz_10ms_envelope_correlation": 0.9787122096376508, "3000–6000 Hz_10ms_modulation_mae_db": 0.7254831202097097, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9975581179241717, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11996603204424895, "3000–6000 Hz_40ms_envelope_correlation": 0.9902152544257938, "3000–6000 Hz_40ms_modulation_mae_db": 0.7158813835170319, "mr_stft_db": 5.5751991030413315, "band_mae_db": 0.18178879806445822, "rms_gain_error_db": -0.11334599342792387, "output_peak": 0.5097679390048061}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9118616855894339, "全帯域_10ms_rms_envelope_nrmse": 0.09361677995087453, "全帯域_10ms_envelope_correlation": 0.9912180946366795, "全帯域_10ms_modulation_mae_db": 0.6035587079373846, "全帯域_40ms_log_rms_mae_db": 0.5195451024966429, "全帯域_40ms_rms_envelope_nrmse": 0.047679897046727655, "全帯域_40ms_envelope_correlation": 0.9974833881606313, "全帯域_40ms_modulation_mae_db": 0.6367254083029257, "300–1000 Hz_10ms_log_rms_mae_db": 1.386345050844654, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11359732234999982, "300–1000 Hz_10ms_envelope_correlation": 0.9892140928057624, "300–1000 Hz_10ms_modulation_mae_db": 0.7507426046100275, "300–1000 Hz_40ms_log_rms_mae_db": 0.8567039881708671, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07145510147840561, "300–1000 Hz_40ms_envelope_correlation": 0.9954153751817791, "300–1000 Hz_40ms_modulation_mae_db": 1.1332597943389355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5779912942478322, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21766545373759258, "1000–3000 Hz_10ms_envelope_correlation": 0.9625961277164617, "1000–3000 Hz_10ms_modulation_mae_db": 0.6062595712777658, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8924495560330985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10474041162824857, "1000–3000 Hz_40ms_envelope_correlation": 0.9904613986272555, "1000–3000 Hz_40ms_modulation_mae_db": 0.5521551950824366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4726849464712333, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1715822599063031, "3000–6000 Hz_10ms_envelope_correlation": 0.9800201772455159, "3000–6000 Hz_10ms_modulation_mae_db": 1.0046688204254663, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9237784112350738, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12049556277989684, "3000–6000 Hz_40ms_envelope_correlation": 0.9895899265045167, "3000–6000 Hz_40ms_modulation_mae_db": 0.9057353531577919, "mr_stft_db": 5.552761044695184, "band_mae_db": 0.1334569462599324, "rms_gain_error_db": -0.03496300675237808, "output_peak": 0.5319343369601074}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9209714128669032, "全帯域_10ms_rms_envelope_nrmse": 0.10800911590124138, "全帯域_10ms_envelope_correlation": 0.9883035612265342, "全帯域_10ms_modulation_mae_db": 0.12258211806603292, "全帯域_40ms_log_rms_mae_db": 0.5337871154844719, "全帯域_40ms_rms_envelope_nrmse": 0.04720236377440556, "全帯域_40ms_envelope_correlation": 0.9975232681111719, "全帯域_40ms_modulation_mae_db": 0.059544529795590724, "300–1000 Hz_10ms_log_rms_mae_db": 1.364042827571485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1049126609719027, "300–1000 Hz_10ms_envelope_correlation": 0.9907745660071031, "300–1000 Hz_10ms_modulation_mae_db": 0.13850552001395103, "300–1000 Hz_40ms_log_rms_mae_db": 0.8068491238809404, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05150307899378331, "300–1000 Hz_40ms_envelope_correlation": 0.9976041073588005, "300–1000 Hz_40ms_modulation_mae_db": 0.1267321928050761, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5318356755894187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2134501874334658, "1000–3000 Hz_10ms_envelope_correlation": 0.9635350897726402, "1000–3000 Hz_10ms_modulation_mae_db": 0.47364067498462187, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9148952392462566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11042942415466352, "1000–3000 Hz_40ms_envelope_correlation": 0.9897752699092613, "1000–3000 Hz_40ms_modulation_mae_db": 0.6720995305553589, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5709382209402598, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1757652135736539, "3000–6000 Hz_10ms_envelope_correlation": 0.9802971051925858, "3000–6000 Hz_10ms_modulation_mae_db": 0.4455713633083656, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9665167798758847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12172209582549823, "3000–6000 Hz_40ms_envelope_correlation": 0.9902913022919594, "3000–6000 Hz_40ms_modulation_mae_db": 0.5955743600508737, "mr_stft_db": 5.576531699552565, "band_mae_db": 0.12603260724177098, "rms_gain_error_db": -0.02316240542345338, "output_peak": 0.5117375098873023}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9314021357484971, "全帯域_10ms_rms_envelope_nrmse": 0.12126543423658837, "全帯域_10ms_envelope_correlation": 0.9854426602923244, "全帯域_10ms_modulation_mae_db": 0.8666087700452793, "全帯域_40ms_log_rms_mae_db": 0.5146709559292909, "全帯域_40ms_rms_envelope_nrmse": 0.05555449572656173, "全帯域_40ms_envelope_correlation": 0.9970233932563212, "全帯域_40ms_modulation_mae_db": 0.8338370148496118, "300–1000 Hz_10ms_log_rms_mae_db": 1.4815982820576337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13668317893261062, "300–1000 Hz_10ms_envelope_correlation": 0.9846545336202247, "300–1000 Hz_10ms_modulation_mae_db": 0.9873272175896372, "300–1000 Hz_40ms_log_rms_mae_db": 0.8881161792362987, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07040864959845489, "300–1000 Hz_40ms_envelope_correlation": 0.9961660859024903, "300–1000 Hz_40ms_modulation_mae_db": 1.1289773780931982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5512566049711174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22215572321966123, "1000–3000 Hz_10ms_envelope_correlation": 0.9608163053875743, "1000–3000 Hz_10ms_modulation_mae_db": 0.5800888665195084, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9106693971727537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12637633911019752, "1000–3000 Hz_40ms_envelope_correlation": 0.9874344761015754, "1000–3000 Hz_40ms_modulation_mae_db": 0.7871506386169077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5755186507437289, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1666593420134657, "3000–6000 Hz_10ms_envelope_correlation": 0.9812354735158697, "3000–6000 Hz_10ms_modulation_mae_db": 0.5094965251918341, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9845690240131699, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12452591131492724, "3000–6000 Hz_40ms_envelope_correlation": 0.9891704467506431, "3000–6000 Hz_40ms_modulation_mae_db": 0.18641597496919923, "mr_stft_db": 5.594796352515001, "band_mae_db": 0.2645454057093694, "rms_gain_error_db": -0.18901792552919022, "output_peak": 0.4836639725170707}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8675543218590185, "全帯域_10ms_rms_envelope_nrmse": 0.09039230512165813, "全帯域_10ms_envelope_correlation": 0.9919302171545379, "全帯域_10ms_modulation_mae_db": 0.4941647010306951, "全帯域_40ms_log_rms_mae_db": 0.4787913618274442, "全帯域_40ms_rms_envelope_nrmse": 0.043340770536630394, "全帯域_40ms_envelope_correlation": 0.9981162712220105, "全帯域_40ms_modulation_mae_db": 0.7122557840909262, "300–1000 Hz_10ms_log_rms_mae_db": 1.4638015176601527, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10152506507796769, "300–1000 Hz_10ms_envelope_correlation": 0.9916779933743493, "300–1000 Hz_10ms_modulation_mae_db": 0.7278785465793284, "300–1000 Hz_40ms_log_rms_mae_db": 0.9098667039952346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0575513623170004, "300–1000 Hz_40ms_envelope_correlation": 0.997444149142602, "300–1000 Hz_40ms_modulation_mae_db": 0.9515921055576259, "1000–3000 Hz_10ms_log_rms_mae_db": 1.51325337334826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.225597031458575, "1000–3000 Hz_10ms_envelope_correlation": 0.9595146468464101, "1000–3000 Hz_10ms_modulation_mae_db": 0.8753285489228197, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8993355573784352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10328548461062018, "1000–3000 Hz_40ms_envelope_correlation": 0.9906894773494, "1000–3000 Hz_40ms_modulation_mae_db": 0.8027320458768005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5545234138801016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15923946308798073, "3000–6000 Hz_10ms_envelope_correlation": 0.9840389689581727, "3000–6000 Hz_10ms_modulation_mae_db": 0.3370965131682544, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9119367483521365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11189438560842702, "3000–6000 Hz_40ms_envelope_correlation": 0.9919969366137037, "3000–6000 Hz_40ms_modulation_mae_db": 0.36150923584087397, "mr_stft_db": 5.575122865198264, "band_mae_db": 0.13527580525588548, "rms_gain_error_db": -0.12752316976166422, "output_peak": 0.5448447316473705}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 42560, "output_samples": 42560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.731415745279375, "spectral_convergence": 0.3075174592277055, "spectral_convergence_status": "ok", "amplitude_floor": 1.6028753096638316e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3966763194635865, "spectral_convergence": 0.296329462844767, "spectral_convergence_status": "ok", "amplitude_floor": 1.0331614165752405e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.622351445462923, "spectral_convergence": 0.3014017964996289, "spectral_convergence_status": "ok", "amplitude_floor": 7.831529815953997e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042273796895084704, "output_power": 0.00044950043159426947, "delta_db": 0.26658856563354, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0017345552824327432, "output_power": 0.0016789288231711587, "delta_db": -0.14155860938395076, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017389853785610774, "output_power": 0.001583243281067756, "delta_db": -0.4074827680349977, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00033443608489121105, "output_power": 0.0003061512725580487, "delta_db": -0.3837706170212858, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00017248781948922851, "output_power": 0.00018360977412985462, "delta_db": 0.2713736427777047, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012677719609633915, "output_power": 0.00013184433643992664, "delta_db": 0.17020336450384171, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27349626122588677, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.19279714925526867, "pitch": null, "output_rms_dbfs": -23.63183456176671, "output_peak": 0.4914528355333998, "reference_rms_dbfs": -23.43903741251144, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5834811700686275}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 42560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00020029577381909937, "log_rms_mae_db": 0.8926628607658864, "rms_envelope_nrmse": 0.10120697100522309, "envelope_correlation": 0.9900878636523057, "rms_gain_error_db": -0.1927961263718624, "reference_mean_power": 0.004538506592765079, "output_mean_power": 0.004341435548473961}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.8045928558777289e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008527253032911584, "output_power": 0.0008056047951558378, "reference_active": true, "delta_db": -0.24687107605029665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002893325022312121, "output_power": 0.00026198235357151556, "reference_active": true, "delta_db": -0.431251835622306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.722708282094668e-05, "output_power": 6.251237789192515e-05, "reference_active": true, "delta_db": -1.9182123624475114}], "active_band_count": 3, "active_band_mae_db": 0.8654450913733713}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00017887965336918488, "log_rms_mae_db": 0.4895863075435125, "rms_envelope_nrmse": 0.06690084149848115, "envelope_correlation": 0.9954766212781037, "rms_gain_error_db": -0.19279731883992768, "reference_mean_power": 0.004590346759257262, "output_mean_power": 0.0043910235054613654}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.4591224648550268e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007420820227306516, "output_power": 0.0006867766811385704, "reference_active": true, "delta_db": -0.3363637006273767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015779167985385838, "output_power": 0.00014397389718997218, "reference_active": true, "delta_db": -0.39800339095178877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.56670863086132e-06, "output_power": 4.084853205393568e-06, "reference_active": true, "delta_db": -2.06171292584245}], "active_band_count": 3, "active_band_mae_db": 0.9320266724738717}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00017771570730226456, "log_rms_mae_db": 1.3156708054214208, "rms_envelope_nrmse": 0.12330600208921773, "envelope_correlation": 0.9877427067450785, "rms_gain_error_db": -0.27948160450574755, "reference_mean_power": 0.003163634497093627, "output_mean_power": 0.0029664575782737644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.820001870942524e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009255585566014781, "output_power": 0.0008910591275302072, "reference_active": true, "delta_db": -0.16497376895257637}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002598152375676334, "output_power": 0.000226280865045288, "reference_active": true, "delta_db": -0.6001678753039618}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.760488032618478e-05, "output_power": 4.426979731855921e-05, "reference_active": true, "delta_db": -2.4378149988445017}], "active_band_count": 3, "active_band_mae_db": 1.0676522143670133}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00016184521030355344, "log_rms_mae_db": 0.8371745317657151, "rms_envelope_nrmse": 0.09350923861102382, "envelope_correlation": 0.9926375418439535, "rms_gain_error_db": -0.27948273213846525, "reference_mean_power": 0.0031997824540876, "output_mean_power": 0.0030003517959807115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.5556561049407657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008585656786559776, "output_power": 0.0008147415835275917, "reference_active": true, "delta_db": -0.22753640680844894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015662214176101382, "output_power": 0.00013794884870970076, "reference_active": true, "delta_db": -0.5513507821386628}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.505375823689367e-06, "output_power": 4.1578296487296615e-06, "reference_active": true, "delta_db": -2.565057521629073}], "active_band_count": 3, "active_band_mae_db": 1.1146482368587283}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00011375403588925618, "log_rms_mae_db": 1.4740122814180365, "rms_envelope_nrmse": 0.21359777195626795, "envelope_correlation": 0.9636287172740816, "rms_gain_error_db": -0.3116215139817853, "reference_mean_power": 0.0003949518827357503, "output_mean_power": 0.0003676055309017686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.6254954818630519e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.926011413944617e-05, "output_power": 8.03247906031633e-05, "reference_active": true, "delta_db": 0.057949086633782836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0135009616582686e-05, "output_power": 1.9500941141707406e-05, "reference_active": true, "delta_db": -0.13896269859302252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.142336744382203e-06, "output_power": 4.725518699618528e-06, "reference_active": true, "delta_db": -1.7939083622638523}], "active_band_count": 3, "active_band_mae_db": 0.6636067158302192}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 8.70060383669589e-05, "log_rms_mae_db": 0.8166430740260922, "rms_envelope_nrmse": 0.11764212166476995, "envelope_correlation": 0.9887675789105086, "rms_gain_error_db": -0.311636874902387, "reference_mean_power": 0.00039945208139827864, "output_mean_power": 0.0003717928221095448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.3661376304062792e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.270190105592004e-05, "output_power": 7.36633239855976e-05, "reference_active": true, "delta_db": 0.05705544857604988}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.983801499648517e-06, "output_power": 9.538591180677504e-06, "reference_active": true, "delta_db": -0.19811702137890752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.054433356878425e-07, "output_power": 3.1118637490454924e-07, "reference_active": true, "delta_db": -2.106519012803437}], "active_band_count": 3, "active_band_mae_db": 0.7872304942527982}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 6.751108516993363e-05, "log_rms_mae_db": 1.584682030368444, "rms_envelope_nrmse": 0.1877864905626322, "envelope_correlation": 0.979116962677872, "rms_gain_error_db": 0.44375754951199287, "reference_mean_power": 0.00015129473773616968, "output_mean_power": 0.0001675712764338901}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 2.0948865520528105e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010579126587837114, "output_power": 0.00011391378268734167, "reference_active": true, "delta_db": 0.3212645968569134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588508061007275e-05, "output_power": 1.8565210510103783e-05, "reference_active": true, "delta_db": 0.6771045500632028}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6030646339801794e-06, "output_power": 4.672583004377082e-06, "reference_active": true, "delta_db": -0.7886860730362684}], "active_band_count": 3, "active_band_mae_db": 0.5956850733187948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 5.7873331101151596e-05, "log_rms_mae_db": 1.0207675871769166, "rms_envelope_nrmse": 0.13263038986288417, "envelope_correlation": 0.9898837483825969, "rms_gain_error_db": 0.4437983860000243, "reference_mean_power": 0.0001530185286286403, "output_mean_power": 0.00016948210923233692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.9708112174484802e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.70769534187325e-05, "output_power": 0.00010677396623048756, "reference_active": true, "delta_db": 0.4134923700337387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.36755214697274e-06, "output_power": 8.665983744758464e-06, "reference_active": true, "delta_db": 0.7049465163402275}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2457588916034453e-07, "output_power": 3.079730675847962e-07, "reference_active": true, "delta_db": -0.22803516629619866}], "active_band_count": 3, "active_band_mae_db": 0.4488246842233883}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9083713719403268, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4c025c153ef98d478987dfb065bfa94350ee85f1b4073edf7b0879d6ae40bc42", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9649188063698537, "standard_deviation": 0.004529522088590919, "minimum": 0.9595146468464101, "maximum": 0.9731487521216571, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5298553719643289, "standard_deviation": 0.04263252184563063, "minimum": 1.4740122814180365, "maximum": 1.5961414713477746, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5898066679205796, "standard_deviation": 0.15012513782451206, "minimum": 0.3749941036195543, "maximum": 0.8753285489228197, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21122353950626793, "standard_deviation": 0.011753979332861661, "minimum": 0.1908355970229226, "maximum": 0.225597031458575, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9887165139232672, "standard_deviation": 0.0017441812291118244, "minimum": 0.9856064789280627, "maximum": 0.9906894773494, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8981641549350554, "standard_deviation": 0.03991085209810353, "minimum": 0.8166430740260922, "maximum": 0.9580495942506163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6840684064307583, "standard_deviation": 0.11812283488333669, "minimum": 0.47361018054676024, "maximum": 0.8027320458768005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11822654009706739, "standard_deviation": 0.012931230818460877, "minimum": 0.10328548461062018, "maximum": 0.14182047308098264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9811272143917973, "standard_deviation": 0.0019722183471630626, "minimum": 0.9787122096376508, "maximum": 0.9840389689581727, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.555315680365935, "standard_deviation": 0.03952527377744261, "minimum": 1.4726849464712333, "maximum": 1.595715417714754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5777336226871608, "standard_deviation": 0.22828666157115682, "minimum": 0.31638439177047323, "maximum": 1.0046688204254663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17020236852026105, "standard_deviation": 0.010722853025816513, "minimum": 0.1571778562481844, "maximum": 0.1877864905626322, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905638576879349, "standard_deviation": 0.0010847578182664257, "minimum": 0.9891704467506431, "maximum": 0.9919969366137037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9791630514738369, "standard_deviation": 0.04694378965858101, "minimum": 0.9119367483521365, "maximum": 1.053719081659529, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5080197674840787, "standard_deviation": 0.22880521382961264, "minimum": 0.18641597496919923, "maximum": 0.9057353531577919, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11851688988070772, "standard_deviation": 0.008448683724240317, "minimum": 0.10757091179014072, "maximum": 0.13263038986288417, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9891422437174548, "standard_deviation": 0.002223554636068033, "minimum": 0.9846545336202247, "maximum": 0.9916779933743493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3984135076524573, "standard_deviation": 0.06958457380138275, "minimum": 1.2939739263335148, "maximum": 1.4815982820576337, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.792209948605714, "standard_deviation": 0.2868466048680043, "minimum": 0.13850552001395103, "maximum": 1.0676522143670133, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11432100012889336, "standard_deviation": 0.011475003052463158, "minimum": 0.10152506507796769, "maximum": 0.13668317893261062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962389373964247, "standard_deviation": 0.0016376289991088012, "minimum": 0.9926375418439535, "maximum": 0.9976041073588005, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8202594514036072, "standard_deviation": 0.07043369647677986, "minimum": 0.6923728141309003, "maximum": 0.9098667039952346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.01972781318856, "standard_deviation": 0.3806365970055166, "minimum": 0.1267321928050761, "maximum": 1.3797147016420712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06570926176465025, "standard_deviation": 0.013293997014515927, "minimum": 0.05150307899378331, "maximum": 0.09350923861102382, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17460450554178936, "standard_deviation": 0.06318973021275048, "minimum": 0.10804044848720935, "maximum": 0.27349626122588677, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.579870713988953, "standard_deviation": 0.01726391192218275, "minimum": 5.552761044695184, "maximum": 5.610939080780141, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5279467532700903, "standard_deviation": 0.03821112433163899, "minimum": 0.4836639725170707, "maximum": 0.6032725604308117, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10487340376667875, "standard_deviation": 0.06358905765859801, "minimum": -0.19279714925526867, "maximum": -0.02316240542345338, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.989928552891793, "standard_deviation": 0.0021072801547999346, "minimum": 0.9854426602923244, "maximum": 0.9919302171545379, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9006138140312422, "standard_deviation": 0.03135032277256863, "minimum": 0.8447007993023435, "maximum": 0.9338859914035756, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6436721403580732, "standard_deviation": 0.24594493187044775, "minimum": 0.12258211806603292, "maximum": 0.8666087700452793, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10030957803536, "standard_deviation": 0.010044619132711458, "minimum": 0.09039230512165813, "maximum": 0.12126543423658837, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973057876104358, "standard_deviation": 0.0008176047687135583, "minimum": 0.9954766212781037, "maximum": 0.9981162712220105, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49530486331232354, "standard_deviation": 0.029703874993922935, "minimum": 0.4419522065092465, "maximum": 0.5337871154844719, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7745268847870941, "standard_deviation": 0.32405797150535404, "minimum": 0.059544529795590724, "maximum": 1.068120302855159, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05030679975368519, "standard_deviation": 0.007774199698106749, "minimum": 0.043340770536630394, "maximum": 0.06690084149848115, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0041", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.685678720474243, "predicted_mos": 2.685678720474243, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "1212a79467b4de113759991b9edb4c0dcb9588e753e203bbbb575ef2ec895ca7", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.005720400251448, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.897451884238631, "全帯域_10ms_rms_envelope_nrmse": 0.10434930721243893, "全帯域_10ms_envelope_correlation": 0.9893383208039611, "全帯域_10ms_modulation_mae_db": 0.9652530878112756, "全帯域_40ms_log_rms_mae_db": 0.4758736365209674, "全帯域_40ms_rms_envelope_nrmse": 0.06311064273104526, "全帯域_40ms_envelope_correlation": 0.9959190489949858, "全帯域_40ms_modulation_mae_db": 1.0866858307351512, "300–1000 Hz_10ms_log_rms_mae_db": 1.2996381501003527, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11708283449759635, "300–1000 Hz_10ms_envelope_correlation": 0.9888319283789899, "300–1000 Hz_10ms_modulation_mae_db": 1.1432810086377658, "300–1000 Hz_40ms_log_rms_mae_db": 0.8248986560069247, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08130444297836216, "300–1000 Hz_40ms_envelope_correlation": 0.9943868202808975, "300–1000 Hz_40ms_modulation_mae_db": 1.2926290963938456, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5094341806358904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2195732345408971, "1000–3000 Hz_10ms_envelope_correlation": 0.9614478846085093, "1000–3000 Hz_10ms_modulation_mae_db": 0.8009253520250658, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8515022049274577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10698315941844799, "1000–3000 Hz_40ms_envelope_correlation": 0.9908452417214192, "1000–3000 Hz_40ms_modulation_mae_db": 0.9769030339349429, "3000–6000 Hz_10ms_log_rms_mae_db": 1.507331175236643, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16259668717929104, "3000–6000 Hz_10ms_envelope_correlation": 0.9836074521493554, "3000–6000 Hz_10ms_modulation_mae_db": 0.6988150509349973, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9110465439936482, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1001382606623906, "3000–6000 Hz_40ms_envelope_correlation": 0.9939931604287806, "3000–6000 Hz_40ms_modulation_mae_db": 0.7232295440822448, "mr_stft_db": 5.5087071345550305, "band_mae_db": 0.20552863970918192, "rms_gain_error_db": -0.16922589771750896, "output_peak": 0.48526052267948283}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8395728036908363, "全帯域_10ms_rms_envelope_nrmse": 0.0935229913077538, "全帯域_10ms_envelope_correlation": 0.9912575172797057, "全帯域_10ms_modulation_mae_db": 0.6764231226316234, "全帯域_40ms_log_rms_mae_db": 0.446700772272365, "全帯域_40ms_rms_envelope_nrmse": 0.048287717025483505, "全帯域_40ms_envelope_correlation": 0.9974611644901087, "全帯域_40ms_modulation_mae_db": 1.0739938734727272, "300–1000 Hz_10ms_log_rms_mae_db": 1.299836280442075, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10518875224243696, "300–1000 Hz_10ms_envelope_correlation": 0.9908045577968557, "300–1000 Hz_10ms_modulation_mae_db": 0.8521981151080954, "300–1000 Hz_40ms_log_rms_mae_db": 0.7214890479376488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05966537813577684, "300–1000 Hz_40ms_envelope_correlation": 0.9968883793448743, "300–1000 Hz_40ms_modulation_mae_db": 1.2746493217162318, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5061004918441085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20186515874350902, "1000–3000 Hz_10ms_envelope_correlation": 0.9689327491220082, "1000–3000 Hz_10ms_modulation_mae_db": 0.3558849828183477, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8642746466764327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1241168671522642, "1000–3000 Hz_40ms_envelope_correlation": 0.987307591689482, "1000–3000 Hz_40ms_modulation_mae_db": 0.4924441023999779, "3000–6000 Hz_10ms_log_rms_mae_db": 1.492404850113054, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14304856866654103, "3000–6000 Hz_10ms_envelope_correlation": 0.9862709161385206, "3000–6000 Hz_10ms_modulation_mae_db": 0.8310344304324017, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0110656353684815, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08225474184243546, "3000–6000 Hz_40ms_envelope_correlation": 0.9955176378274777, "3000–6000 Hz_40ms_modulation_mae_db": 0.8199608124106647, "mr_stft_db": 5.538767136560554, "band_mae_db": 0.11067446766088727, "rms_gain_error_db": -0.07302493476383712, "output_peak": 0.5648850624538878}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8832647077092526, "全帯域_10ms_rms_envelope_nrmse": 0.10090957252321082, "全帯域_10ms_envelope_correlation": 0.9898257437496574, "全帯域_10ms_modulation_mae_db": 0.6355021940922708, "全帯域_40ms_log_rms_mae_db": 0.4699255527858465, "全帯域_40ms_rms_envelope_nrmse": 0.042264447449701835, "全帯域_40ms_envelope_correlation": 0.9981026893367009, "全帯域_40ms_modulation_mae_db": 1.0482041033232847, "300–1000 Hz_10ms_log_rms_mae_db": 1.4465333898078236, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1161446104393956, "300–1000 Hz_10ms_envelope_correlation": 0.9886872455664091, "300–1000 Hz_10ms_modulation_mae_db": 0.8965220590629839, "300–1000 Hz_40ms_log_rms_mae_db": 0.8111994729304985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055192875336186595, "300–1000 Hz_40ms_envelope_correlation": 0.9972299467208748, "300–1000 Hz_40ms_modulation_mae_db": 1.1459107219015336, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6186882245795735, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23432233608648323, "1000–3000 Hz_10ms_envelope_correlation": 0.9571120398851971, "1000–3000 Hz_10ms_modulation_mae_db": 0.8255084778404228, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9491251623682863, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13320504721623397, "1000–3000 Hz_40ms_envelope_correlation": 0.9877219657907884, "1000–3000 Hz_40ms_modulation_mae_db": 0.8509719019460582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.503734809241268, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1518071488501076, "3000–6000 Hz_10ms_envelope_correlation": 0.9847288101078143, "3000–6000 Hz_10ms_modulation_mae_db": 0.3311365513773569, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9760287940153557, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09829306000120135, "3000–6000 Hz_40ms_envelope_correlation": 0.9932974182023601, "3000–6000 Hz_40ms_modulation_mae_db": 0.5170686427020531, "mr_stft_db": 5.507946949334577, "band_mae_db": 0.20864360631959425, "rms_gain_error_db": -0.08249125776590337, "output_peak": 0.5432715874044811}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9390966725076578, "全帯域_10ms_rms_envelope_nrmse": 0.09592758888592244, "全帯域_10ms_envelope_correlation": 0.9908517941414767, "全帯域_10ms_modulation_mae_db": 0.7485074862645318, "全帯域_40ms_log_rms_mae_db": 0.5077777302200391, "全帯域_40ms_rms_envelope_nrmse": 0.050442890926355184, "全帯域_40ms_envelope_correlation": 0.9973103754888281, "全帯域_40ms_modulation_mae_db": 0.8850502958804141, "300–1000 Hz_10ms_log_rms_mae_db": 1.4886924331080884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11047583402480102, "300–1000 Hz_10ms_envelope_correlation": 0.9898276063438182, "300–1000 Hz_10ms_modulation_mae_db": 0.8597221867414548, "300–1000 Hz_40ms_log_rms_mae_db": 0.8771784297503005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06053484753699177, "300–1000 Hz_40ms_envelope_correlation": 0.9967696788738794, "300–1000 Hz_40ms_modulation_mae_db": 1.3830452833591806, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5220622673163469, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2192231071282181, "1000–3000 Hz_10ms_envelope_correlation": 0.9635489560090372, "1000–3000 Hz_10ms_modulation_mae_db": 0.6671281246465489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9233580813840657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14132144685203082, "1000–3000 Hz_40ms_envelope_correlation": 0.9863520567805482, "1000–3000 Hz_40ms_modulation_mae_db": 0.8775188180667634, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5382821841391963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1623327509304112, "3000–6000 Hz_10ms_envelope_correlation": 0.9824029988192796, "3000–6000 Hz_10ms_modulation_mae_db": 0.94956473672225, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9171600160928178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09261175246723423, "3000–6000 Hz_40ms_envelope_correlation": 0.9940189192308749, "3000–6000 Hz_40ms_modulation_mae_db": 1.0822796648942241, "mr_stft_db": 5.499039314874399, "band_mae_db": 0.16919315177936597, "rms_gain_error_db": -0.10942630575411123, "output_peak": 0.5496769432142191}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.934184964619756, "全帯域_10ms_rms_envelope_nrmse": 0.09918139153198856, "全帯域_10ms_envelope_correlation": 0.9901395312768023, "全帯域_10ms_modulation_mae_db": 0.5274514441491707, "全帯域_40ms_log_rms_mae_db": 0.5246934073995208, "全帯域_40ms_rms_envelope_nrmse": 0.04807645188813667, "全帯域_40ms_envelope_correlation": 0.9974302969507614, "全帯域_40ms_modulation_mae_db": 0.5155784597652918, "300–1000 Hz_10ms_log_rms_mae_db": 1.4404984134993317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.109371572581379, "300–1000 Hz_10ms_envelope_correlation": 0.9899633734468483, "300–1000 Hz_10ms_modulation_mae_db": 0.6595255845227056, "300–1000 Hz_40ms_log_rms_mae_db": 0.891280360149705, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06280785426245951, "300–1000 Hz_40ms_envelope_correlation": 0.9963909178897828, "300–1000 Hz_40ms_modulation_mae_db": 1.0789828105085986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5701100762129947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23637985084674074, "1000–3000 Hz_10ms_envelope_correlation": 0.9551122014197392, "1000–3000 Hz_10ms_modulation_mae_db": 0.7072627780967693, "1000–3000 Hz_40ms_log_rms_mae_db": 0.891321071852378, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10063352629591699, "1000–3000 Hz_40ms_envelope_correlation": 0.9914149180126915, "1000–3000 Hz_40ms_modulation_mae_db": 0.8418925314188964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4276682832240526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1504646588302885, "3000–6000 Hz_10ms_envelope_correlation": 0.9846777295223632, "3000–6000 Hz_10ms_modulation_mae_db": 1.1163479681454198, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8568182573326482, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08793818095557884, "3000–6000 Hz_40ms_envelope_correlation": 0.9946669034998549, "3000–6000 Hz_40ms_modulation_mae_db": 1.1009614342087144, "mr_stft_db": 5.492356094377541, "band_mae_db": 0.18032331920345657, "rms_gain_error_db": -0.030867991200681247, "output_peak": 0.5739037198812823}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9031973078136687, "全帯域_10ms_rms_envelope_nrmse": 0.11069871173400084, "全帯域_10ms_envelope_correlation": 0.9877625747159462, "全帯域_10ms_modulation_mae_db": 0.164906162971612, "全帯域_40ms_log_rms_mae_db": 0.5287914301904478, "全帯域_40ms_rms_envelope_nrmse": 0.052945576226208664, "全帯域_40ms_envelope_correlation": 0.9968836355731864, "全帯域_40ms_modulation_mae_db": 0.11891729659380841, "300–1000 Hz_10ms_log_rms_mae_db": 1.3971021636942782, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11130325721044848, "300–1000 Hz_10ms_envelope_correlation": 0.9896443087920179, "300–1000 Hz_10ms_modulation_mae_db": 0.1794537428967172, "300–1000 Hz_40ms_log_rms_mae_db": 0.8315578381758557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058245145516651144, "300–1000 Hz_40ms_envelope_correlation": 0.9968866988696233, "300–1000 Hz_40ms_modulation_mae_db": 0.2414880672901317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5265157042195197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20351118330948162, "1000–3000 Hz_10ms_envelope_correlation": 0.9670240166344748, "1000–3000 Hz_10ms_modulation_mae_db": 0.3869584957246297, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9135901695626208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09895104529467859, "1000–3000 Hz_40ms_envelope_correlation": 0.9915741530952512, "1000–3000 Hz_40ms_modulation_mae_db": 0.5174920557761329, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5185035589298852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15301026940245277, "3000–6000 Hz_10ms_envelope_correlation": 0.9846703611231948, "3000–6000 Hz_10ms_modulation_mae_db": 0.5864711397627428, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9104096003094858, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09107005352354056, "3000–6000 Hz_40ms_envelope_correlation": 0.9944471591748656, "3000–6000 Hz_40ms_modulation_mae_db": 0.9081801980682767, "mr_stft_db": 5.499139854785604, "band_mae_db": 0.10636361746865741, "rms_gain_error_db": 0.013888572105005641, "output_peak": 0.5262194625810203}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9558209465010933, "全帯域_10ms_rms_envelope_nrmse": 0.1246573836960428, "全帯域_10ms_envelope_correlation": 0.9846521526015177, "全帯域_10ms_modulation_mae_db": 0.9236780312006898, "全帯域_40ms_log_rms_mae_db": 0.524074978230744, "全帯域_40ms_rms_envelope_nrmse": 0.058123991284309104, "全帯域_40ms_envelope_correlation": 0.9968007770412642, "全帯域_40ms_modulation_mae_db": 0.8528579535525974, "300–1000 Hz_10ms_log_rms_mae_db": 1.4548688769646891, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1426503358826096, "300–1000 Hz_10ms_envelope_correlation": 0.9831253489006331, "300–1000 Hz_10ms_modulation_mae_db": 1.0963753752478567, "300–1000 Hz_40ms_log_rms_mae_db": 0.8721187611752411, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0694385247379385, "300–1000 Hz_40ms_envelope_correlation": 0.9961664495172743, "300–1000 Hz_40ms_modulation_mae_db": 1.2306012880403745, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5163854432109731, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2097404439736987, "1000–3000 Hz_10ms_envelope_correlation": 0.9651826282836709, "1000–3000 Hz_10ms_modulation_mae_db": 0.6942351441379034, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8732957443346636, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10258972934859353, "1000–3000 Hz_40ms_envelope_correlation": 0.9922004412448564, "1000–3000 Hz_40ms_modulation_mae_db": 1.0409592833739418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4834359505536947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13995590578702255, "3000–6000 Hz_10ms_envelope_correlation": 0.9871410229468471, "3000–6000 Hz_10ms_modulation_mae_db": 0.5905923546417898, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8919908962638773, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08670002631170401, "3000–6000 Hz_40ms_envelope_correlation": 0.9953503377057672, "3000–6000 Hz_40ms_modulation_mae_db": 0.34364455650921105, "mr_stft_db": 5.522198919020532, "band_mae_db": 0.24073219666033038, "rms_gain_error_db": -0.2101221227048279, "output_peak": 0.4890201927887944}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9058074719085334, "全帯域_10ms_rms_envelope_nrmse": 0.10129268987072144, "全帯域_10ms_envelope_correlation": 0.9897367324491214, "全帯域_10ms_modulation_mae_db": 0.466897072449195, "全帯域_40ms_log_rms_mae_db": 0.48466387012325024, "全帯域_40ms_rms_envelope_nrmse": 0.04223816245715015, "全帯域_40ms_envelope_correlation": 0.9981094777563289, "全帯域_40ms_modulation_mae_db": 0.6430760480507521, "300–1000 Hz_10ms_log_rms_mae_db": 1.4649642637337263, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10002586721584703, "300–1000 Hz_10ms_envelope_correlation": 0.9917426339949743, "300–1000 Hz_10ms_modulation_mae_db": 0.7113638942601557, "300–1000 Hz_40ms_log_rms_mae_db": 0.855344645057701, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05334430394936963, "300–1000 Hz_40ms_envelope_correlation": 0.9976229708526068, "300–1000 Hz_40ms_modulation_mae_db": 0.905452640642714, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5102606375723828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26048465220233885, "1000–3000 Hz_10ms_envelope_correlation": 0.9466184683633247, "1000–3000 Hz_10ms_modulation_mae_db": 1.0092718712127724, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8906568210368813, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11226006055151455, "1000–3000 Hz_40ms_envelope_correlation": 0.9891715893041848, "1000–3000 Hz_40ms_modulation_mae_db": 0.9614792517098811, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4800378840677009, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13065561032010664, "3000–6000 Hz_10ms_envelope_correlation": 0.9888235114592915, "3000–6000 Hz_10ms_modulation_mae_db": 0.22498895635251415, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8688051579532804, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07613634096529262, "3000–6000 Hz_40ms_envelope_correlation": 0.9961100914505189, "3000–6000 Hz_40ms_modulation_mae_db": 0.4765348145672408, "mr_stft_db": 5.502190753034164, "band_mae_db": 0.06105237777834934, "rms_gain_error_db": -0.08893893267049124, "output_peak": 0.5518427456167323}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 42560, "output_samples": 42560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.661256157347398, "spectral_convergence": 0.306731202346581, "spectral_convergence_status": "ok", "amplitude_floor": 1.6028753096638316e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.327383594398393, "spectral_convergence": 0.29589407453021144, "spectral_convergence_status": "ok", "amplitude_floor": 1.0331614165752405e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5374816519192995, "spectral_convergence": 0.29977132247629534, "spectral_convergence_status": "ok", "amplitude_floor": 7.831529815953997e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042273796895084704, "output_power": 0.0004430070999895309, "delta_db": 0.203394301327708, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0017345552824327432, "output_power": 0.001687061828493399, "delta_db": -0.12057146727936097, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017389853785610774, "output_power": 0.0016138260993360499, "delta_db": -0.3243919571450699, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00033443608489121105, "output_power": 0.0003076070227812587, "delta_db": -0.3631688433475417, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00017248781948922851, "output_power": 0.0001797895421397607, "delta_db": 0.18005994349736457, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012677719609633915, "output_power": 0.00012556905034562742, "delta_db": -0.04158532565804638, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20552863970918192, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16922589771750896, "pitch": null, "output_rms_dbfs": -23.608263310228953, "output_peak": 0.48526052267948283, "reference_rms_dbfs": -23.43903741251144, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.5087071345550305}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 42560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00020029577381909937, "log_rms_mae_db": 0.897451884238631, "rms_envelope_nrmse": 0.10434930721243893, "envelope_correlation": 0.9893383208039611, "rms_gain_error_db": -0.16922471805521494, "reference_mean_power": 0.004538506592765079, "output_mean_power": 0.004365062825926637}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.8045928558777289e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008527253032911584, "output_power": 0.0008106866523758108, "reference_active": true, "delta_db": -0.21956127589647778}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002893325022312121, "output_power": 0.00025559206709815214, "reference_active": true, "delta_db": -0.538498526126484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.722708282094668e-05, "output_power": 5.943158453238819e-05, "reference_active": true, "delta_db": -2.137699461410865}], "active_band_count": 3, "active_band_mae_db": 0.9652530878112756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00017887965336918488, "log_rms_mae_db": 0.4758736365209674, "rms_envelope_nrmse": 0.06311064273104526, "envelope_correlation": 0.9959190489949858, "rms_gain_error_db": -0.1692259959157813, "reference_mean_power": 0.004590346759257262, "output_mean_power": 0.004414920567305605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.4591224648550268e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007420820227306516, "output_power": 0.0006910978821765009, "reference_active": true, "delta_db": -0.3091234863357847}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015779167985385838, "output_power": 0.0001410235942964702, "reference_active": true, "delta_db": -0.48792320272550804}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.56670863086132e-06, "output_power": 3.7243172320491392e-06, "reference_active": true, "delta_db": -2.463010803144161}], "active_band_count": 3, "active_band_mae_db": 1.0866858307351512}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00017771570730226456, "log_rms_mae_db": 1.2996381501003527, "rms_envelope_nrmse": 0.11708283449759635, "envelope_correlation": 0.9888319283789899, "rms_gain_error_db": -0.2263681115180524, "reference_mean_power": 0.003163634497093627, "output_mean_power": 0.0030029596131362375}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.820001870942524e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009255585566014781, "output_power": 0.0008938551730449817, "reference_active": true, "delta_db": -0.15136742235002326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002598152375676334, "output_power": 0.00022474156863913806, "reference_active": true, "delta_db": -0.6298121005517475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.760488032618478e-05, "output_power": 4.217185292368178e-05, "reference_active": true, "delta_db": -2.648663503011526}], "active_band_count": 3, "active_band_mae_db": 1.1432810086377658}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00016184521030355344, "log_rms_mae_db": 0.8248986560069247, "rms_envelope_nrmse": 0.08130444297836216, "envelope_correlation": 0.9943868202808975, "rms_gain_error_db": -0.22636941513610392, "reference_mean_power": 0.0031997824540876, "output_mean_power": 0.003037270773537312}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.5556561049407657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008585656786559776, "output_power": 0.0008163649527980539, "reference_active": true, "delta_db": -0.21889171643005306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015662214176101382, "output_power": 0.0001353786434617229, "reference_active": true, "delta_db": -0.6330300050907269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.505375823689367e-06, "output_power": 3.7391758249134036e-06, "reference_active": true, "delta_db": -3.025965567660757}], "active_band_count": 3, "active_band_mae_db": 1.2926290963938456}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00011375403588925618, "log_rms_mae_db": 1.5094341806358904, "rms_envelope_nrmse": 0.2195732345408971, "envelope_correlation": 0.9614478846085093, "rms_gain_error_db": -0.3085516782804451, "reference_mean_power": 0.0003949518827357503, "output_mean_power": 0.00036786546685839443}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.6254954818630519e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.926011413944617e-05, "output_power": 8.001244199616267e-05, "reference_active": true, "delta_db": 0.041028319591704554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0135009616582686e-05, "output_power": 1.940181449859206e-05, "reference_active": true, "delta_db": -0.1610949339968319}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.142336744382203e-06, "output_power": 4.3030366199321065e-06, "reference_active": true, "delta_db": -2.2006528024866605}], "active_band_count": 3, "active_band_mae_db": 0.8009253520250658}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 8.70060383669589e-05, "log_rms_mae_db": 0.8515022049274577, "rms_envelope_nrmse": 0.10698315941844799, "envelope_correlation": 0.9908452417214192, "rms_gain_error_db": -0.30856822578301696, "reference_mean_power": 0.00039945208139827864, "output_mean_power": 0.00037205561727027037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.3661376304062792e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.270190105592004e-05, "output_power": 7.346144773504988e-05, "reference_active": true, "delta_db": 0.04513715696026523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.983801499648517e-06, "output_power": 9.376733071119388e-06, "reference_active": true, "delta_db": -0.272443846192675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.054433356878425e-07, "output_power": 2.7692340618971266e-07, "reference_active": true, "delta_db": -2.6131280986518886}], "active_band_count": 3, "active_band_mae_db": 0.9769030339349429}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 6.751108516993363e-05, "log_rms_mae_db": 1.507331175236643, "rms_envelope_nrmse": 0.16259668717929104, "envelope_correlation": 0.9836074521493554, "rms_gain_error_db": 0.28746981338652866, "reference_mean_power": 0.00015129473773616968, "output_mean_power": 0.00016164817439786935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 2.0948865520528105e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010579126587837114, "output_power": 0.00011375521775331325, "reference_active": true, "delta_db": 0.315215122675249}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588508061007275e-05, "output_power": 1.7732474626686588e-05, "reference_active": true, "delta_db": 0.47779924061342155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6030646339801794e-06, "output_power": 4.150329271187694e-06, "reference_active": true, "delta_db": -1.303430789516321}], "active_band_count": 3, "active_band_mae_db": 0.6988150509349973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 5.7873331101151596e-05, "log_rms_mae_db": 0.9110465439936482, "rms_envelope_nrmse": 0.1001382606623906, "envelope_correlation": 0.9939931604287806, "rms_gain_error_db": 0.2875092156028589, "reference_mean_power": 0.0001530185286286403, "output_mean_power": 0.0001634914114504739}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.9708112174484802e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.70769534187325e-05, "output_power": 0.00010563665385466733, "reference_active": true, "delta_db": 0.36698497704293337}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.36755214697274e-06, "output_power": 8.309945447667505e-06, "reference_active": true, "delta_db": 0.5227495434263294}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2457588916034453e-07, "output_power": 2.417245967024266e-07, "reference_active": true, "delta_db": -1.2799541117774715}], "active_band_count": 3, "active_band_mae_db": 0.7232295440822448}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9080619529775452, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "44bfb1372b30495f8fcd91b6340a9abaf89b64c67b69e16d00d776e31cfee8ee", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9606223680407452, "standard_deviation": 0.007357012716571531, "minimum": 0.9466184683633247, "maximum": 0.9689327491220082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5349446281989738, "standard_deviation": 0.03950246019055486, "minimum": 1.5061004918441085, "maximum": 1.6186882245795735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6808969033128075, "standard_deviation": 0.21916876691500067, "minimum": 0.3558849828183477, "maximum": 1.0092718712127724, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22313749585392093, "standard_deviation": 0.0197862960195499, "minimum": 0.20186515874350902, "maximum": 0.26048465220233885, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9895734947049027, "standard_deviation": 0.002236493698302352, "minimum": 0.9863520567805482, "maximum": 0.9922004412448564, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8946404877678482, "standard_deviation": 0.03255292906729718, "minimum": 0.8515022049274577, "maximum": 0.9491251623682863, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8199576223283245, "standard_deviation": 0.20598502362673804, "minimum": 0.4924441023999779, "maximum": 1.0409592833739418, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11500761026621008, "standard_deviation": 0.01602282053121433, "minimum": 0.09895104529467859, "maximum": 0.14132144685203082, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9852903502833333, "standard_deviation": 0.0020393867891261765, "minimum": 0.9824029988192796, "maximum": 0.9888235114592915, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.493924836938187, "standard_deviation": 0.03282585817886429, "minimum": 1.4276682832240526, "maximum": 1.5382821841391963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6661188985461841, "standard_deviation": 0.30004065259269863, "minimum": 0.22498895635251415, "maximum": 1.1163479681454198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14923394999577766, "standard_deviation": 0.010967703675751181, "minimum": 0.13065561032010664, "maximum": 0.16259668717929104, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9946752034400625, "standard_deviation": 0.0009312558683208813, "minimum": 0.9932974182023601, "maximum": 0.9961100914505189, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9179156126661994, "standard_deviation": 0.05206449069522114, "minimum": 0.8568182573326482, "maximum": 1.0110656353684815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7464824584303287, "standard_deviation": 0.2824084777586379, "minimum": 0.34364455650921105, "maximum": 1.1009614342087144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0893928020911722, "standard_deviation": 0.007963370736466673, "minimum": 0.07613634096529262, "maximum": 0.1001382606623906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9890783754025683, "standard_deviation": 0.0026020615965440135, "minimum": 0.9831253489006331, "maximum": 0.9917426339949743, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4115167464187957, "standard_deviation": 0.07363555068913752, "minimum": 1.2996381501003527, "maximum": 1.4886924331080884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7998052458097169, "standard_deviation": 0.3009401646877892, "minimum": 0.1794537428967172, "maximum": 1.1432810086377658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11403038301181426, "standard_deviation": 0.01281163849017713, "minimum": 0.10002586721584703, "maximum": 0.1426503358826096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965427327937266, "standard_deviation": 0.0009813301177974537, "minimum": 0.9943868202808975, "maximum": 0.9976229708526068, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8356334013979845, "standard_deviation": 0.0538517585138256, "minimum": 0.7214890479376488, "maximum": 0.891280360149705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0690949037315765, "standard_deviation": 0.3653920817224018, "minimum": 0.2414880672901317, "maximum": 1.3830452833591806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06256667155671702, "standard_deviation": 0.009016104224862504, "minimum": 0.05334430394936963, "maximum": 0.08130444297836216, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16031392207247788, "standard_deviation": 0.061590826808396934, "minimum": 0.06105237777834934, "maximum": 0.24073219666033038, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5087932695678, "standard_deviation": 0.015028400821951692, "minimum": 5.492356094377541, "maximum": 5.538767136560554, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5355100295774875, "standard_deviation": 0.03303094079796802, "minimum": 0.48526052267948283, "maximum": 0.5739037198812823, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09377610880904444, "standard_deviation": 0.07131866756012031, "minimum": -0.2101221227048279, "maximum": 0.013888572105005641, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9891955458772737, "standard_deviation": 0.0021143130786760007, "minimum": 0.9846521526015177, "maximum": 0.9912575172797057, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9072995948736786, "standard_deviation": 0.03658419927459764, "minimum": 0.8395728036908363, "maximum": 0.9558209465010933, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6385773251962962, "standard_deviation": 0.258488272217159, "minimum": 0.164906162971612, "maximum": 0.9652530878112756, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10381745459525996, "standard_deviation": 0.009903004327811467, "minimum": 0.0935229913077538, "maximum": 0.1246573836960428, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972521832040206, "standard_deviation": 0.0007224897781911625, "minimum": 0.9959190489949858, "maximum": 0.9981094777563289, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4953126722178976, "standard_deviation": 0.030391724666945175, "minimum": 0.446700772272365, "maximum": 0.5287914301904478, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7780454826717533, "standard_deviation": 0.3368833165333431, "minimum": 0.11891729659380841, "maximum": 1.0866858307351512, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.050686234998548796, "standard_deviation": 0.007267444596312649, "minimum": 0.04223816245715015, "maximum": 0.06311064273104526, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0041", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.4421677589416504, "predicted_mos": 2.4421677589416504, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "c1c470f92cf45d623594751dbe4178dac3de1b177a584b1a687f6438a9161a11", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.955659800209105, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8367768907438949, "全帯域_10ms_rms_envelope_nrmse": 0.10315119181053008, "全帯域_10ms_envelope_correlation": 0.9893792668995671, "全帯域_10ms_modulation_mae_db": 0.5092869967943888, "全帯域_40ms_log_rms_mae_db": 0.45995990436454814, "全帯域_40ms_rms_envelope_nrmse": 0.05199851543850492, "全帯域_40ms_envelope_correlation": 0.9969958930892383, "全帯域_40ms_modulation_mae_db": 0.44151271229921224, "300–1000 Hz_10ms_log_rms_mae_db": 1.5914192335186026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11775627185450666, "300–1000 Hz_10ms_envelope_correlation": 0.9883879316187965, "300–1000 Hz_10ms_modulation_mae_db": 0.700934001569311, "300–1000 Hz_40ms_log_rms_mae_db": 0.9284418789238664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06454045150042965, "300–1000 Hz_40ms_envelope_correlation": 0.9962173469621124, "300–1000 Hz_40ms_modulation_mae_db": 0.7305181747315421, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7112656123798673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2722642385871431, "1000–3000 Hz_10ms_envelope_correlation": 0.944840210662262, "1000–3000 Hz_10ms_modulation_mae_db": 0.6963894628590799, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0402183530638118, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14760469594500192, "1000–3000 Hz_40ms_envelope_correlation": 0.983035896473478, "1000–3000 Hz_40ms_modulation_mae_db": 0.7511594347647845, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8065400550537805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2154304310307992, "3000–6000 Hz_10ms_envelope_correlation": 0.9707497725005658, "3000–6000 Hz_10ms_modulation_mae_db": 0.7848212291664911, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1730941017085477, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15957179347733944, "3000–6000 Hz_40ms_envelope_correlation": 0.9835216324484486, "3000–6000 Hz_40ms_modulation_mae_db": 0.8996024612698971, "mr_stft_db": 5.856151398317728, "band_mae_db": 0.2028597393550131, "rms_gain_error_db": 0.01582156349720467, "output_peak": 0.5867407182835288}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8954465759166391, "全帯域_10ms_rms_envelope_nrmse": 0.09665448979205475, "全帯域_10ms_envelope_correlation": 0.9907968586087712, "全帯域_10ms_modulation_mae_db": 0.8045130359123002, "全帯域_40ms_log_rms_mae_db": 0.499025122375746, "全帯域_40ms_rms_envelope_nrmse": 0.042329409867941106, "全帯域_40ms_envelope_correlation": 0.9982565576533343, "全帯域_40ms_modulation_mae_db": 1.0827052693900672, "300–1000 Hz_10ms_log_rms_mae_db": 1.5602112232249448, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1083064913681849, "300–1000 Hz_10ms_envelope_correlation": 0.9906308827078217, "300–1000 Hz_10ms_modulation_mae_db": 1.0555365654409588, "300–1000 Hz_40ms_log_rms_mae_db": 0.8958779682444147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06586054332166716, "300–1000 Hz_40ms_envelope_correlation": 0.9966320823428797, "300–1000 Hz_40ms_modulation_mae_db": 1.2295904812700027, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6322076508584678, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21346553384105427, "1000–3000 Hz_10ms_envelope_correlation": 0.9635382592150747, "1000–3000 Hz_10ms_modulation_mae_db": 0.7858065155402253, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9116968673397451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11139539878190934, "1000–3000 Hz_40ms_envelope_correlation": 0.9896025072976766, "1000–3000 Hz_40ms_modulation_mae_db": 0.8829162718390151, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9056731070820176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21001058332638212, "3000–6000 Hz_10ms_envelope_correlation": 0.9718470292813344, "3000–6000 Hz_10ms_modulation_mae_db": 0.9392200489144832, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2888680840356141, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16289288545926744, "3000–6000 Hz_40ms_envelope_correlation": 0.9823855148485827, "3000–6000 Hz_40ms_modulation_mae_db": 0.6869218977231331, "mr_stft_db": 5.870912867846055, "band_mae_db": 0.23987666908756236, "rms_gain_error_db": -0.13327680746298154, "output_peak": 0.5189255771202246}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8981731166782638, "全帯域_10ms_rms_envelope_nrmse": 0.09904641495841264, "全帯域_10ms_envelope_correlation": 0.9901997603839315, "全帯域_10ms_modulation_mae_db": 0.40958490784011253, "全帯域_40ms_log_rms_mae_db": 0.49235737800750357, "全帯域_40ms_rms_envelope_nrmse": 0.04491782187215439, "全帯域_40ms_envelope_correlation": 0.997819995319061, "全帯域_40ms_modulation_mae_db": 0.5129977148339645, "300–1000 Hz_10ms_log_rms_mae_db": 1.5759533252520757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10536578098127855, "300–1000 Hz_10ms_envelope_correlation": 0.9909651471711872, "300–1000 Hz_10ms_modulation_mae_db": 0.5923067940736556, "300–1000 Hz_40ms_log_rms_mae_db": 0.9112763207060733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0629873085938303, "300–1000 Hz_40ms_envelope_correlation": 0.9967349050794263, "300–1000 Hz_40ms_modulation_mae_db": 0.8323882863472855, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7656286915531543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2311581351545446, "1000–3000 Hz_10ms_envelope_correlation": 0.9574188864010177, "1000–3000 Hz_10ms_modulation_mae_db": 0.92976004241619, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0180818854323797, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.133188838487729, "1000–3000 Hz_40ms_envelope_correlation": 0.9844841907290978, "1000–3000 Hz_40ms_modulation_mae_db": 1.0322033711363199, "3000–6000 Hz_10ms_log_rms_mae_db": 1.907765894358807, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2284524326336894, "3000–6000 Hz_10ms_envelope_correlation": 0.9705237062739152, "3000–6000 Hz_10ms_modulation_mae_db": 1.244705202203694, "3000–6000 Hz_40ms_log_rms_mae_db": 1.248855720990911, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1732015602666215, "3000–6000 Hz_40ms_envelope_correlation": 0.9836166051350617, "3000–6000 Hz_40ms_modulation_mae_db": 0.7754619270980981, "mr_stft_db": 5.873037309390582, "band_mae_db": 0.16146429138318194, "rms_gain_error_db": -0.0642689320186921, "output_peak": 0.5221254974490135}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.895006145423218, "全帯域_10ms_rms_envelope_nrmse": 0.09875184984758587, "全帯域_10ms_envelope_correlation": 0.9904471277739773, "全帯域_10ms_modulation_mae_db": 0.69226719443727, "全帯域_40ms_log_rms_mae_db": 0.48725977895393835, "全帯域_40ms_rms_envelope_nrmse": 0.05122548844747226, "全帯域_40ms_envelope_correlation": 0.9974298442032735, "全帯域_40ms_modulation_mae_db": 0.7835598844182282, "300–1000 Hz_10ms_log_rms_mae_db": 1.6302193897634485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12540862350352433, "300–1000 Hz_10ms_envelope_correlation": 0.9876308421829008, "300–1000 Hz_10ms_modulation_mae_db": 0.7382059679079737, "300–1000 Hz_40ms_log_rms_mae_db": 0.9479377339062958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08226738476562301, "300–1000 Hz_40ms_envelope_correlation": 0.9949531049650191, "300–1000 Hz_40ms_modulation_mae_db": 0.9434744959677182, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7724837434051843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2245104015386631, "1000–3000 Hz_10ms_envelope_correlation": 0.9609354862560645, "1000–3000 Hz_10ms_modulation_mae_db": 0.5719015985569466, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0149792896582097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13446350687516428, "1000–3000 Hz_40ms_envelope_correlation": 0.9846904632267298, "1000–3000 Hz_40ms_modulation_mae_db": 0.629001480953389, "3000–6000 Hz_10ms_log_rms_mae_db": 1.957126554428304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23924556751063905, "3000–6000 Hz_10ms_envelope_correlation": 0.9634610091324741, "3000–6000 Hz_10ms_modulation_mae_db": 0.7011432449729736, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2815129206802793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1815658519203632, "3000–6000 Hz_40ms_envelope_correlation": 0.9781134631033386, "3000–6000 Hz_40ms_modulation_mae_db": 0.4903240913435584, "mr_stft_db": 5.891933174443846, "band_mae_db": 0.26122403844206266, "rms_gain_error_db": -0.1639462538129616, "output_peak": 0.4996641911442009}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8493889739764516, "全帯域_10ms_rms_envelope_nrmse": 0.0989993243031824, "全帯域_10ms_envelope_correlation": 0.9902448700384782, "全帯域_10ms_modulation_mae_db": 0.3230401361358877, "全帯域_40ms_log_rms_mae_db": 0.49730400667757996, "全帯域_40ms_rms_envelope_nrmse": 0.05980584609117517, "全帯域_40ms_envelope_correlation": 0.9961472736062305, "全帯域_40ms_modulation_mae_db": 0.5137573553037845, "300–1000 Hz_10ms_log_rms_mae_db": 1.5626258468015013, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12314609855397698, "300–1000 Hz_10ms_envelope_correlation": 0.9874275661255982, "300–1000 Hz_10ms_modulation_mae_db": 0.6468501590384611, "300–1000 Hz_40ms_log_rms_mae_db": 0.9262377963446362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07757541359785922, "300–1000 Hz_40ms_envelope_correlation": 0.9947820123306439, "300–1000 Hz_40ms_modulation_mae_db": 0.93896055739949, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6825796104010389, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18932014643601286, "1000–3000 Hz_10ms_envelope_correlation": 0.9715225927985982, "1000–3000 Hz_10ms_modulation_mae_db": 0.7617493466186321, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9634430761617311, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11341839853052199, "1000–3000 Hz_40ms_envelope_correlation": 0.9889902166605486, "1000–3000 Hz_40ms_modulation_mae_db": 0.8493357897653159, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9320784835671891, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23177959087304575, "3000–6000 Hz_10ms_envelope_correlation": 0.9665125406132838, "3000–6000 Hz_10ms_modulation_mae_db": 0.9585837339833073, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2731062283302486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18067158582922943, "3000–6000 Hz_40ms_envelope_correlation": 0.9790093038200778, "3000–6000 Hz_40ms_modulation_mae_db": 0.7879428769246314, "mr_stft_db": 5.856950695669606, "band_mae_db": 0.23159795908159453, "rms_gain_error_db": -0.11073731419305935, "output_peak": 0.5638074231033281}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8838405140303158, "全帯域_10ms_rms_envelope_nrmse": 0.10262664292740822, "全帯域_10ms_envelope_correlation": 0.9894464998649943, "全帯域_10ms_modulation_mae_db": 0.9037531362842053, "全帯域_40ms_log_rms_mae_db": 0.520451570293854, "全帯域_40ms_rms_envelope_nrmse": 0.04960109789950841, "全帯域_40ms_envelope_correlation": 0.9972821700830655, "全帯域_40ms_modulation_mae_db": 1.07577801157367, "300–1000 Hz_10ms_log_rms_mae_db": 1.5716428817607029, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.126424269528979, "300–1000 Hz_10ms_envelope_correlation": 0.9866832160606411, "300–1000 Hz_10ms_modulation_mae_db": 1.2798992395927586, "300–1000 Hz_40ms_log_rms_mae_db": 0.9437866355384317, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07406721566502156, "300–1000 Hz_40ms_envelope_correlation": 0.9951622459337206, "300–1000 Hz_40ms_modulation_mae_db": 1.5909561563361407, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6880440434949038, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22236384335607315, "1000–3000 Hz_10ms_envelope_correlation": 0.9604874911465479, "1000–3000 Hz_10ms_modulation_mae_db": 0.4820696881942596, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0354833458304578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1336246613178486, "1000–3000 Hz_40ms_envelope_correlation": 0.9849247393929074, "1000–3000 Hz_40ms_modulation_mae_db": 0.45609808559149534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9339136241790866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2280900614193218, "3000–6000 Hz_10ms_envelope_correlation": 0.9656796563590155, "3000–6000 Hz_10ms_modulation_mae_db": 0.960728445086794, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2504865104499334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1638300661169143, "3000–6000 Hz_40ms_envelope_correlation": 0.981507244435408, "3000–6000 Hz_40ms_modulation_mae_db": 0.6505623102946992, "mr_stft_db": 5.859588811931325, "band_mae_db": 0.18482760481657903, "rms_gain_error_db": -0.05242230468396881, "output_peak": 0.5491582683712989}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8965635049794815, "全帯域_10ms_rms_envelope_nrmse": 0.09923045624909432, "全帯域_10ms_envelope_correlation": 0.9901796812294743, "全帯域_10ms_modulation_mae_db": 0.669776462451427, "全帯域_40ms_log_rms_mae_db": 0.5177553366392917, "全帯域_40ms_rms_envelope_nrmse": 0.0567675034625404, "全帯域_40ms_envelope_correlation": 0.9964553038531626, "全帯域_40ms_modulation_mae_db": 0.6757372685631315, "300–1000 Hz_10ms_log_rms_mae_db": 1.6321352110169716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12883145455916326, "300–1000 Hz_10ms_envelope_correlation": 0.9862311511875249, "300–1000 Hz_10ms_modulation_mae_db": 1.1943244035632048, "300–1000 Hz_40ms_log_rms_mae_db": 1.0345546188353802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08687627654283255, "300–1000 Hz_40ms_envelope_correlation": 0.9932844162842903, "300–1000 Hz_40ms_modulation_mae_db": 1.1704488093106258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6784674501257213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.244050200117625, "1000–3000 Hz_10ms_envelope_correlation": 0.9521062210225972, "1000–3000 Hz_10ms_modulation_mae_db": 0.5150754733727614, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9904297684518787, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16504420075735843, "1000–3000 Hz_40ms_envelope_correlation": 0.9761621682789634, "1000–3000 Hz_40ms_modulation_mae_db": 0.7275720106416005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9691893949313106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2315474697328712, "3000–6000 Hz_10ms_envelope_correlation": 0.9652531226645272, "3000–6000 Hz_10ms_modulation_mae_db": 0.503812791205815, "3000–6000 Hz_40ms_log_rms_mae_db": 1.28598634419099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18203767011434538, "3000–6000 Hz_40ms_envelope_correlation": 0.9776182056289923, "3000–6000 Hz_40ms_modulation_mae_db": 0.5364388028577932, "mr_stft_db": 5.865286322554076, "band_mae_db": 0.2589951473217415, "rms_gain_error_db": -0.054314471865884485, "output_peak": 0.529126294990639}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8758802698829027, "全帯域_10ms_rms_envelope_nrmse": 0.09340699500124475, "全帯域_10ms_envelope_correlation": 0.9915055591368703, "全帯域_10ms_modulation_mae_db": 1.0077093443795, "全帯域_40ms_log_rms_mae_db": 0.45294990576178373, "全帯域_40ms_rms_envelope_nrmse": 0.04274453121495753, "全帯域_40ms_envelope_correlation": 0.9983205177967405, "全帯域_40ms_modulation_mae_db": 1.1508085302376145, "300–1000 Hz_10ms_log_rms_mae_db": 1.4724249647103518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10914984273113484, "300–1000 Hz_10ms_envelope_correlation": 0.9907703804452468, "300–1000 Hz_10ms_modulation_mae_db": 1.1375524469929867, "300–1000 Hz_40ms_log_rms_mae_db": 0.8680324072460172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06261674406995499, "300–1000 Hz_40ms_envelope_correlation": 0.9974237744819447, "300–1000 Hz_40ms_modulation_mae_db": 1.4971688865159196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7505973759588784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2357511542868672, "1000–3000 Hz_10ms_envelope_correlation": 0.9559675862290318, "1000–3000 Hz_10ms_modulation_mae_db": 0.7999698704824191, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0488664800640155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1531561318144288, "1000–3000 Hz_40ms_envelope_correlation": 0.9795179400296561, "1000–3000 Hz_40ms_modulation_mae_db": 0.8454625076439228, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8607567637388915, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23116429911403932, "3000–6000 Hz_10ms_envelope_correlation": 0.9655928765532711, "3000–6000 Hz_10ms_modulation_mae_db": 1.152003255372114, "3000–6000 Hz_40ms_log_rms_mae_db": 1.306159411317458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19183545772212868, "3000–6000 Hz_40ms_envelope_correlation": 0.9750936314338579, "3000–6000 Hz_40ms_modulation_mae_db": 1.0858401142595353, "mr_stft_db": 5.867568122991571, "band_mae_db": 0.19404058108790134, "rms_gain_error_db": -0.15473899349827044, "output_peak": 0.5138698377463609}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 42560, "output_samples": 42560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.032358554114669, "spectral_convergence": 0.3244450543755761, "spectral_convergence_status": "ok", "amplitude_floor": 1.6028753096638316e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.727041291632558, "spectral_convergence": 0.30952466346036145, "spectral_convergence_status": "ok", "amplitude_floor": 1.0331614165752405e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.809054349205958, "spectral_convergence": 0.3161703502843224, "spectral_convergence_status": "ok", "amplitude_floor": 7.831529815953997e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042273796895084704, "output_power": 0.00043574399214362094, "delta_db": 0.131601514520198, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0017345552824327432, "output_power": 0.001783292463659126, "delta_db": 0.12034428307866106, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017389853785610774, "output_power": 0.0016572119482961313, "delta_db": -0.20917874609790224, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00033443608489121105, "output_power": 0.0003576402703607727, "delta_db": 0.2913328383667078, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00017248781948922851, "output_power": 0.00017196053467139174, "delta_db": -0.01329645288691064, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012677719609633915, "output_power": 0.00014066355930522742, "delta_db": 0.4514046011796991, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2028597393550131, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.01582156349720467, "pitch": null, "output_rms_dbfs": -23.423215849014237, "output_peak": 0.5867407182835288, "reference_rms_dbfs": -23.43903741251144, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.856151398317728}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 42560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00020029577381909937, "log_rms_mae_db": 0.8367768907438949, "rms_envelope_nrmse": 0.10315119181053008, "envelope_correlation": 0.9893792668995671, "rms_gain_error_db": 0.015822224011951283, "reference_mean_power": 0.004538506592765079, "output_mean_power": 0.004555071443991648}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.8045928558777289e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008527253032911584, "output_power": 0.0008378349954123692, "reference_active": true, "delta_db": -0.0765065379542062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002893325022312121, "output_power": 0.0002987920653997328, "reference_active": true, "delta_db": 0.13971837393332281}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.722708282094668e-05, "output_power": 7.188257855496191e-05, "reference_active": true, "delta_db": -1.3116360784956373}], "active_band_count": 3, "active_band_mae_db": 0.5092869967943888}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00017887965336918488, "log_rms_mae_db": 0.45995990436454814, "rms_envelope_nrmse": 0.05199851543850492, "envelope_correlation": 0.9969958930892383, "rms_gain_error_db": 0.015825217207530995, "reference_mean_power": 0.004590346759257262, "output_mean_power": 0.0046071039943759275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.4591224648550268e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007420820227306516, "output_power": 0.0007291015522952595, "reference_active": true, "delta_db": -0.07663887866080932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015779167985385838, "output_power": 0.00015755116140334422, "reference_active": true, "delta_db": -0.006624907398770134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.56670863086132e-06, "output_power": 4.934240478062646e-06, "reference_active": true, "delta_db": -1.2412743508380573}], "active_band_count": 3, "active_band_mae_db": 0.44151271229921224}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00017771570730226456, "log_rms_mae_db": 1.5914192335186026, "rms_envelope_nrmse": 0.11775627185450666, "envelope_correlation": 0.9883879316187965, "rms_gain_error_db": -0.051524530924013344, "reference_mean_power": 0.003163634497093627, "output_mean_power": 0.003126323027208462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.820001870942524e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009255585566014781, "output_power": 0.0009127867586791054, "reference_active": true, "delta_db": -0.06034568760573178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002598152375676334, "output_power": 0.00024744170686910637, "reference_active": true, "delta_db": -0.21191715061717079}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.760488032618478e-05, "output_power": 5.0913753651712406e-05, "reference_active": true, "delta_db": -1.8305391664850301}], "active_band_count": 3, "active_band_mae_db": 0.700934001569311}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00016184521030355344, "log_rms_mae_db": 0.9284418789238664, "rms_envelope_nrmse": 0.06454045150042965, "envelope_correlation": 0.9962173469621124, "rms_gain_error_db": -0.051524558463630105, "reference_mean_power": 0.0031997824540876, "output_mean_power": 0.0031620446401210076}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.5556561049407657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008585656786559776, "output_power": 0.0008357977332690634, "reference_active": true, "delta_db": -0.11672334499040138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015662214176101382, "output_power": 0.0001466316877097735, "reference_active": true, "delta_db": -0.28625325251107503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.505375823689367e-06, "output_power": 4.97181146263966e-06, "reference_active": true, "delta_db": -1.78857792669315}], "active_band_count": 3, "active_band_mae_db": 0.7305181747315421}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00011375403588925618, "log_rms_mae_db": 1.7112656123798673, "rms_envelope_nrmse": 0.2722642385871431, "envelope_correlation": 0.944840210662262, "rms_gain_error_db": 0.2978765790307284, "reference_mean_power": 0.0003949518827357503, "output_mean_power": 0.00042299170066010795}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.6254954818630519e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.926011413944617e-05, "output_power": 8.115845459231817e-05, "reference_active": true, "delta_db": 0.10279075257224715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0135009616582686e-05, "output_power": 2.2666326680517184e-05, "reference_active": true, "delta_db": 0.5142930242112075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.142336744382203e-06, "output_power": 5.0889930894037795e-06, "reference_active": true, "delta_db": -1.472084611793785}], "active_band_count": 3, "active_band_mae_db": 0.6963894628590799}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 8.70060383669589e-05, "log_rms_mae_db": 1.0402183530638118, "rms_envelope_nrmse": 0.14760469594500192, "envelope_correlation": 0.983035896473478, "rms_gain_error_db": 0.2979116528100479, "reference_mean_power": 0.00039945208139827864, "output_mean_power": 0.00042781484834242457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.3661376304062792e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.270190105592004e-05, "output_power": 7.524631853630693e-05, "reference_active": true, "delta_db": 0.1493948949328171}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.983801499648517e-06, "output_power": 1.08873890447602e-05, "reference_active": true, "delta_db": 0.37627804392630637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.054433356878425e-07, "output_power": 3.395407770100717e-07, "reference_active": true, "delta_db": -1.7278053654352301}], "active_band_count": 3, "active_band_mae_db": 0.7511594347647845}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 6.751108516993363e-05, "log_rms_mae_db": 1.8065400550537805, "rms_envelope_nrmse": 0.2154304310307992, "envelope_correlation": 0.9707497725005658, "rms_gain_error_db": 0.3328486294716098, "reference_mean_power": 0.00015129473773616968, "output_mean_power": 0.0001633460683855604}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 2.0948865520528105e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010579126587837114, "output_power": 0.00010469662726757027, "reference_active": true, "delta_db": -0.04517122453047005}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588508061007275e-05, "output_power": 2.317470549006442e-05, "reference_active": true, "delta_db": 1.640248006992767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6030646339801794e-06, "output_power": 4.80309511162009e-06, "reference_active": true, "delta_db": -0.6690444559762364}], "active_band_count": 3, "active_band_mae_db": 0.7848212291664911}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 5.7873331101151596e-05, "log_rms_mae_db": 1.1730941017085477, "rms_envelope_nrmse": 0.15957179347733944, "envelope_correlation": 0.9835216324484486, "rms_gain_error_db": 0.33287692238928146, "reference_mean_power": 0.0001530185286286403, "output_mean_power": 0.00016520824352886094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.9708112174484802e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.70769534187325e-05, "output_power": 0.00010079093564369935, "reference_active": true, "delta_db": 0.16305338440938916}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.36755214697274e-06, "output_power": 1.0020239455516551e-05, "reference_active": true, "delta_db": 1.3355488163474432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2457588916034453e-07, "output_power": 2.462043586770634e-07, "reference_active": true, "delta_db": -1.2002051830528586}], "active_band_count": 3, "active_band_mae_db": 0.8996024612698971}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8948447652697236, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "45a3e1554a860734f3483edf4a5bc17a234472d9e119a0d496706dc4a99ea830", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9583520917163992, "standard_deviation": 0.007925824509422743, "minimum": 0.944840210662262, "maximum": 0.9715225927985982, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.710159272272152, "standard_deviation": 0.049176371155805094, "minimum": 1.6322076508584678, "maximum": 1.7724837434051843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6928402497550642, "standard_deviation": 0.15659877467434882, "minimum": 0.4820696881942596, "maximum": 0.92976004241619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2291104566647479, "standard_deviation": 0.02398942141060018, "minimum": 0.18932014643601286, "maximum": 0.2722642385871431, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9839260152611322, "standard_deviation": 0.004476332152178013, "minimum": 0.9761621682789634, "maximum": 0.9896025072976766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0028998832502787, "standard_deviation": 0.046250894964135744, "minimum": 0.9116968673397451, "maximum": 1.0488664800640155, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7717186190419804, "standard_deviation": 0.1748615436320231, "minimum": 0.45609808559149534, "maximum": 1.0322033711363199, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1364869790637453, "standard_deviation": 0.018523695387371107, "minimum": 0.11139539878190934, "maximum": 0.16504420075735843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9674524641722984, "standard_deviation": 0.0031137978444147176, "minimum": 0.9634610091324741, "maximum": 0.9718470292813344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9091304846674233, "standard_deviation": 0.05335881071093393, "minimum": 1.8065400550537805, "maximum": 1.9691893949313106, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.905627243863209, "standard_deviation": 0.23919706848601674, "minimum": 0.503812791205815, "maximum": 1.244705202203694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22696505445509846, "standard_deviation": 0.009536415249087653, "minimum": 0.21001058332638212, "maximum": 0.23924556751063905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.980108200106721, "standard_deviation": 0.0031084317106176004, "minimum": 0.9750936314338579, "maximum": 0.9836166051350617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2635086652129979, "standard_deviation": 0.04130509711541222, "minimum": 1.1730941017085477, "maximum": 1.306159411317458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7391368102214182, "standard_deviation": 0.19391457723498146, "minimum": 0.4903240913435584, "maximum": 1.0858401142595353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1744508588632762, "standard_deviation": 0.011454505502461408, "minimum": 0.15957179347733944, "maximum": 0.19183545772212868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9885908896874647, "standard_deviation": 0.001929938542256215, "minimum": 0.9862311511875249, "maximum": 0.9909651471711872, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.574579009506075, "standard_deviation": 0.0499820413915065, "minimum": 1.4724249647103518, "maximum": 1.6321352110169716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9182011972724138, "standard_deviation": 0.2760660733241255, "minimum": 0.5923067940736556, "maximum": 1.2798992395927586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11804860413509358, "standard_deviation": 0.00927063062667247, "minimum": 0.10536578098127855, "maximum": 0.12883145455916326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956487360475046, "standard_deviation": 0.0013460315267114715, "minimum": 0.9932844162842903, "maximum": 0.9974237744819447, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9320181699681394, "standard_deviation": 0.04896408529816034, "minimum": 0.8680324072460172, "maximum": 1.0345546188353802, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1166882309848405, "standard_deviation": 0.3109608627133457, "minimum": 0.7305181747315421, "maximum": 1.5909561563361407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07209891725715231, "standard_deviation": 0.009446201235425796, "minimum": 0.06261674406995499, "maximum": 0.08687627654283255, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21686075382195458, "standard_deviation": 0.036453502845414446, "minimum": 0.16146429138318194, "maximum": 0.26122403844206266, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.867678587893099, "standard_deviation": 0.01162881159493359, "minimum": 5.856151398317728, "maximum": 5.891933174443846, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5354272260260743, "standard_deviation": 0.028932703648178914, "minimum": 0.4996641911442009, "maximum": 0.5867407182835288, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0897354392548267, "standard_deviation": 0.061536719956570195, "minimum": -0.1639462538129616, "maximum": 0.01582156349720467, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9902749529920081, "standard_deviation": 0.0006887509287644751, "minimum": 0.9893792668995671, "maximum": 0.9915055591368703, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.878884498953896, "standard_deviation": 0.023590046468136294, "minimum": 0.8367768907438949, "maximum": 0.8981731166782638, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6649914017793864, "standard_deviation": 0.23937505331859132, "minimum": 0.3230401361358877, "maximum": 1.0077093443795, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09898342061118913, "standard_deviation": 0.0031002926082909, "minimum": 0.09340699500124475, "maximum": 0.10315119181053008, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973384444505133, "standard_deviation": 0.0007898443426098203, "minimum": 0.9961472736062305, "maximum": 0.9983205177967405, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49088287538428066, "standard_deviation": 0.02424401343634358, "minimum": 0.45294990576178373, "maximum": 0.520451570293854, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.779607093327459, "standard_deviation": 0.2889006856927639, "minimum": 0.44151271229921224, "maximum": 1.1508085302376145, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04992377678678177, "standard_deviation": 0.006379325584819957, "minimum": 0.042329409867941106, "maximum": 0.05980584609117517, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0041", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.047847270965576, "predicted_mos": 2.047847270965576, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "043343b806d66cda00163ed3e569bb50a95cdbdeb811d3a087c67fb8268f01ac", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.077344300225377, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9032009780077146, "全帯域_10ms_rms_envelope_nrmse": 0.10884219869837478, "全帯域_10ms_envelope_correlation": 0.9881450989168559, "全帯域_10ms_modulation_mae_db": 0.5740020381605272, "全帯域_40ms_log_rms_mae_db": 0.4936575791636094, "全帯域_40ms_rms_envelope_nrmse": 0.05998542388989604, "全帯域_40ms_envelope_correlation": 0.9960788019299723, "全帯域_40ms_modulation_mae_db": 0.48144887063296643, "300–1000 Hz_10ms_log_rms_mae_db": 1.392485068761405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11242261893782303, "300–1000 Hz_10ms_envelope_correlation": 0.9894398730841297, "300–1000 Hz_10ms_modulation_mae_db": 0.5981050834249473, "300–1000 Hz_40ms_log_rms_mae_db": 0.8951438335734178, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06664504757668384, "300–1000 Hz_40ms_envelope_correlation": 0.9960244334145958, "300–1000 Hz_40ms_modulation_mae_db": 0.597966055516237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4538672818529264, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20025802567507814, "1000–3000 Hz_10ms_envelope_correlation": 0.9682364867265664, "1000–3000 Hz_10ms_modulation_mae_db": 0.4807201733301358, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8552931667056762, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12263007597882968, "1000–3000 Hz_40ms_envelope_correlation": 0.9876808835096509, "1000–3000 Hz_40ms_modulation_mae_db": 0.4568316287814318, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3584682897310014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14719778125854335, "3000–6000 Hz_10ms_envelope_correlation": 0.985289980009236, "3000–6000 Hz_10ms_modulation_mae_db": 0.602240446205831, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8247972034799822, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08049622120732156, "3000–6000 Hz_40ms_envelope_correlation": 0.9954457808531028, "3000–6000 Hz_40ms_modulation_mae_db": 0.7209445624028809, "mr_stft_db": 5.486019986242844, "band_mae_db": 0.21483572867698897, "rms_gain_error_db": -0.09535117374735805, "output_peak": 0.5351369585224263}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8682384030736312, "全帯域_10ms_rms_envelope_nrmse": 0.09576487938786954, "全帯域_10ms_envelope_correlation": 0.9908842544603298, "全帯域_10ms_modulation_mae_db": 0.7301198464330613, "全帯域_40ms_log_rms_mae_db": 0.43203235297564213, "全帯域_40ms_rms_envelope_nrmse": 0.042031904341997446, "全帯域_40ms_envelope_correlation": 0.9981743951287934, "全帯域_40ms_modulation_mae_db": 1.0011621511761295, "300–1000 Hz_10ms_log_rms_mae_db": 1.291100497140297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11589899967380979, "300–1000 Hz_10ms_envelope_correlation": 0.9888933375955937, "300–1000 Hz_10ms_modulation_mae_db": 1.0022171026934459, "300–1000 Hz_40ms_log_rms_mae_db": 0.7260942546114418, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06141257155540716, "300–1000 Hz_40ms_envelope_correlation": 0.9968478055752229, "300–1000 Hz_40ms_modulation_mae_db": 1.024510196869311, "1000–3000 Hz_10ms_log_rms_mae_db": 1.459109500511153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14749321078384467, "1000–3000 Hz_10ms_envelope_correlation": 0.9829326796077356, "1000–3000 Hz_10ms_modulation_mae_db": 0.10210999025963835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.848392617140605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08391362403692675, "1000–3000 Hz_40ms_envelope_correlation": 0.9939160095749534, "1000–3000 Hz_40ms_modulation_mae_db": 0.14389802006366673, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4059083691455347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1562961171528784, "3000–6000 Hz_10ms_envelope_correlation": 0.9835750944766405, "3000–6000 Hz_10ms_modulation_mae_db": 0.7662757621765032, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8947595329969965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10284502710471076, "3000–6000 Hz_40ms_envelope_correlation": 0.9924711446653068, "3000–6000 Hz_40ms_modulation_mae_db": 0.538410062198396, "mr_stft_db": 5.4591321206941785, "band_mae_db": 0.179337226679543, "rms_gain_error_db": -0.10332624032586718, "output_peak": 0.5205474732272001}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9086713801824758, "全帯域_10ms_rms_envelope_nrmse": 0.10976570924322483, "全帯域_10ms_envelope_correlation": 0.9882326210738563, "全帯域_10ms_modulation_mae_db": 1.0876444488455717, "全帯域_40ms_log_rms_mae_db": 0.5230095995425508, "全帯域_40ms_rms_envelope_nrmse": 0.06623848117512063, "全帯域_40ms_envelope_correlation": 0.9956205113679432, "全帯域_40ms_modulation_mae_db": 1.1448847531531026, "300–1000 Hz_10ms_log_rms_mae_db": 1.3784811499858338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12265493735953309, "300–1000 Hz_10ms_envelope_correlation": 0.9877369934616578, "300–1000 Hz_10ms_modulation_mae_db": 1.2809144001331438, "300–1000 Hz_40ms_log_rms_mae_db": 0.9428749940959401, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08273652935435169, "300–1000 Hz_40ms_envelope_correlation": 0.9942754818448662, "300–1000 Hz_40ms_modulation_mae_db": 1.4474665306123462, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4101394209512297, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23447297572987355, "1000–3000 Hz_10ms_envelope_correlation": 0.9583339146584413, "1000–3000 Hz_10ms_modulation_mae_db": 0.42120350489469666, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8988310905748624, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16817378377035064, "1000–3000 Hz_40ms_envelope_correlation": 0.9794826524343926, "1000–3000 Hz_40ms_modulation_mae_db": 0.35643422593462476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4180408223223124, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1575967907386367, "3000–6000 Hz_10ms_envelope_correlation": 0.9831471083226875, "3000–6000 Hz_10ms_modulation_mae_db": 0.2913521443836345, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8792382407925735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09634620115215096, "3000–6000 Hz_40ms_envelope_correlation": 0.9933527400881881, "3000–6000 Hz_40ms_modulation_mae_db": 0.32832061983950966, "mr_stft_db": 5.472310202421511, "band_mae_db": 0.27363557039671926, "rms_gain_error_db": -0.20558247888490994, "output_peak": 0.5285412149744593}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9310236232146624, "全帯域_10ms_rms_envelope_nrmse": 0.10061123020259607, "全帯域_10ms_envelope_correlation": 0.9899372008373936, "全帯域_10ms_modulation_mae_db": 0.7296516767519899, "全帯域_40ms_log_rms_mae_db": 0.48917007967816606, "全帯域_40ms_rms_envelope_nrmse": 0.044700620943837116, "全帯域_40ms_envelope_correlation": 0.9979344176096657, "全帯域_40ms_modulation_mae_db": 1.2854916373231065, "300–1000 Hz_10ms_log_rms_mae_db": 1.3397235924464983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10476273012165084, "300–1000 Hz_10ms_envelope_correlation": 0.9910927454160987, "300–1000 Hz_10ms_modulation_mae_db": 1.256736310386519, "300–1000 Hz_40ms_log_rms_mae_db": 0.7235576966109271, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05855361684846711, "300–1000 Hz_40ms_envelope_correlation": 0.9972722040894071, "300–1000 Hz_40ms_modulation_mae_db": 1.402264953396692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4413610674294375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22063051853514348, "1000–3000 Hz_10ms_envelope_correlation": 0.9610228579700945, "1000–3000 Hz_10ms_modulation_mae_db": 0.7109583408170016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8266141122942876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0839261686583865, "1000–3000 Hz_40ms_envelope_correlation": 0.9940709657200084, "1000–3000 Hz_40ms_modulation_mae_db": 0.7301565925555128, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3655245687819149, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1565044111246986, "3000–6000 Hz_10ms_envelope_correlation": 0.9846733467353388, "3000–6000 Hz_10ms_modulation_mae_db": 1.042366973953148, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8647925767647482, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09309543566297819, "3000–6000 Hz_40ms_envelope_correlation": 0.9946911175315108, "3000–6000 Hz_40ms_modulation_mae_db": 0.8433655531919461, "mr_stft_db": 5.4697008567780285, "band_mae_db": 0.16377848904764122, "rms_gain_error_db": -0.10835700764951386, "output_peak": 0.5220881109504815}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8790615898413298, "全帯域_10ms_rms_envelope_nrmse": 0.10696686854443099, "全帯域_10ms_envelope_correlation": 0.9886289181785044, "全帯域_10ms_modulation_mae_db": 0.4034069877105709, "全帯域_40ms_log_rms_mae_db": 0.5067813408871789, "全帯域_40ms_rms_envelope_nrmse": 0.052760713463759455, "全帯域_40ms_envelope_correlation": 0.9971401621442334, "全帯域_40ms_modulation_mae_db": 0.40644326370248823, "300–1000 Hz_10ms_log_rms_mae_db": 1.3916392280372183, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1186403997527814, "300–1000 Hz_10ms_envelope_correlation": 0.9883288300788987, "300–1000 Hz_10ms_modulation_mae_db": 0.5740804651859978, "300–1000 Hz_40ms_log_rms_mae_db": 0.8338350211116828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06575366959664736, "300–1000 Hz_40ms_envelope_correlation": 0.9963304716651222, "300–1000 Hz_40ms_modulation_mae_db": 0.6775482699018546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.439003926432335, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1815785326766435, "1000–3000 Hz_10ms_envelope_correlation": 0.9742909844374738, "1000–3000 Hz_10ms_modulation_mae_db": 0.5102220854058728, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8954214554457809, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10151180591056803, "1000–3000 Hz_40ms_envelope_correlation": 0.9922823655268715, "1000–3000 Hz_40ms_modulation_mae_db": 0.5450790889970407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3644614446595358, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15465794650876255, "3000–6000 Hz_10ms_envelope_correlation": 0.9837310698517671, "3000–6000 Hz_10ms_modulation_mae_db": 0.28456869014886355, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8745674869837254, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09760654170992232, "3000–6000 Hz_40ms_envelope_correlation": 0.9932115892272003, "3000–6000 Hz_40ms_modulation_mae_db": 0.29149044714140854, "mr_stft_db": 5.4891307981189135, "band_mae_db": 0.2278879158135285, "rms_gain_error_db": -0.14000251946118017, "output_peak": 0.49973226790319597}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.846819954103772, "全帯域_10ms_rms_envelope_nrmse": 0.09262375911105329, "全帯域_10ms_envelope_correlation": 0.9914094620968151, "全帯域_10ms_modulation_mae_db": 0.1680644307895601, "全帯域_40ms_log_rms_mae_db": 0.5016321763604467, "全帯域_40ms_rms_envelope_nrmse": 0.04412190833578787, "全帯域_40ms_envelope_correlation": 0.997832594100395, "全帯域_40ms_modulation_mae_db": 0.22262425129719618, "300–1000 Hz_10ms_log_rms_mae_db": 1.2479820756318343, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09260959227012037, "300–1000 Hz_10ms_envelope_correlation": 0.9928611820537281, "300–1000 Hz_10ms_modulation_mae_db": 0.06116065388563327, "300–1000 Hz_40ms_log_rms_mae_db": 0.781824257711186, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0483713847321848, "300–1000 Hz_40ms_envelope_correlation": 0.9979529762428114, "300–1000 Hz_40ms_modulation_mae_db": 0.1535862306648991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4278387235286503, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20209049450701874, "1000–3000 Hz_10ms_envelope_correlation": 0.9684962648654863, "1000–3000 Hz_10ms_modulation_mae_db": 0.3560597707404294, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8449085057079481, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10554309416458792, "1000–3000 Hz_40ms_envelope_correlation": 0.9906972909518137, "1000–3000 Hz_40ms_modulation_mae_db": 0.49427581803208404, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3777846511873195, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14528304872790954, "3000–6000 Hz_10ms_envelope_correlation": 0.9858595780087354, "3000–6000 Hz_10ms_modulation_mae_db": 0.46028071170091583, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9092302855496905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09132287482434398, "3000–6000 Hz_40ms_envelope_correlation": 0.9940899506001013, "3000–6000 Hz_40ms_modulation_mae_db": 0.240150926610165, "mr_stft_db": 5.499768957021897, "band_mae_db": 0.13677296250219756, "rms_gain_error_db": -0.017241378702317133, "output_peak": 0.5405040060466771}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8440290883012538, "全帯域_10ms_rms_envelope_nrmse": 0.09552689279246748, "全帯域_10ms_envelope_correlation": 0.9908725107819941, "全帯域_10ms_modulation_mae_db": 0.32415948577083986, "全帯域_40ms_log_rms_mae_db": 0.45986135345951423, "全帯域_40ms_rms_envelope_nrmse": 0.043431518811923836, "全帯域_40ms_envelope_correlation": 0.9979670334214931, "全帯域_40ms_modulation_mae_db": 0.20937869965305245, "300–1000 Hz_10ms_log_rms_mae_db": 1.251704720090108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10550587580232275, "300–1000 Hz_10ms_envelope_correlation": 0.9907523254737217, "300–1000 Hz_10ms_modulation_mae_db": 0.39819599843706666, "300–1000 Hz_40ms_log_rms_mae_db": 0.7482113281277948, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06220038887758438, "300–1000 Hz_40ms_envelope_correlation": 0.9966207235113936, "300–1000 Hz_40ms_modulation_mae_db": 0.6291001397665896, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5233563972295912, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21599998161730743, "1000–3000 Hz_10ms_envelope_correlation": 0.9638891517616289, "1000–3000 Hz_10ms_modulation_mae_db": 0.5613116103882687, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9095299087971088, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11701168167350254, "1000–3000 Hz_40ms_envelope_correlation": 0.9885192536033113, "1000–3000 Hz_40ms_modulation_mae_db": 0.6021359072158451, "3000–6000 Hz_10ms_log_rms_mae_db": 1.36998919525725, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15250629076183417, "3000–6000 Hz_10ms_envelope_correlation": 0.9842659068470512, "3000–6000 Hz_10ms_modulation_mae_db": 0.16948995744576764, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8440352537638086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08602817615326448, "3000–6000 Hz_40ms_envelope_correlation": 0.9946926721482964, "3000–6000 Hz_40ms_modulation_mae_db": 0.09107061424367728, "mr_stft_db": 5.463636393712103, "band_mae_db": 0.09723144974856011, "rms_gain_error_db": -0.07697725886442564, "output_peak": 0.5265488910647662}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8954525395893171, "全帯域_10ms_rms_envelope_nrmse": 0.1011632016379673, "全帯域_10ms_envelope_correlation": 0.9899190024345433, "全帯域_10ms_modulation_mae_db": 0.9256390199515704, "全帯域_40ms_log_rms_mae_db": 0.4773722324729124, "全帯域_40ms_rms_envelope_nrmse": 0.051364665456233916, "全帯域_40ms_envelope_correlation": 0.997365622743695, "全帯域_40ms_modulation_mae_db": 1.2983250888266566, "300–1000 Hz_10ms_log_rms_mae_db": 1.3930012838630286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11099901453044442, "300–1000 Hz_10ms_envelope_correlation": 0.989988842984038, "300–1000 Hz_10ms_modulation_mae_db": 1.3425093608691956, "300–1000 Hz_40ms_log_rms_mae_db": 0.7482103747680053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06280391837141379, "300–1000 Hz_40ms_envelope_correlation": 0.9968658864674307, "300–1000 Hz_40ms_modulation_mae_db": 1.5064794083491113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4410812308382086, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19252559274545092, "1000–3000 Hz_10ms_envelope_correlation": 0.9713624437303794, "1000–3000 Hz_10ms_modulation_mae_db": 0.4551882885864702, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9224895732739833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1334170040796236, "1000–3000 Hz_40ms_envelope_correlation": 0.9861161321400085, "1000–3000 Hz_40ms_modulation_mae_db": 0.537622083515076, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4549900965449236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16790248242810837, "3000–6000 Hz_10ms_envelope_correlation": 0.9829869069564272, "3000–6000 Hz_10ms_modulation_mae_db": 1.1820251456634334, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9432114005944469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11412485463439033, "3000–6000 Hz_40ms_envelope_correlation": 0.9923074776350364, "3000–6000 Hz_40ms_modulation_mae_db": 0.8963514663577415, "mr_stft_db": 5.490623818120237, "band_mae_db": 0.18286785956061055, "rms_gain_error_db": -0.1546195690123258, "output_peak": 0.508889265482854}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 42560, "output_samples": 42560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.581226393034491, "spectral_convergence": 0.2974924524777082, "spectral_convergence_status": "ok", "amplitude_floor": 1.6028753096638316e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.281248963039557, "spectral_convergence": 0.297706787760242, "spectral_convergence_status": "ok", "amplitude_floor": 1.0331614165752405e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.595584602654481, "spectral_convergence": 0.3050140842642432, "spectral_convergence_status": "ok", "amplitude_floor": 7.831529815953997e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042273796895084704, "output_power": 0.0004259459955317531, "delta_db": 0.03283283144073903, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0017345552824327432, "output_power": 0.001789331936084802, "delta_db": 0.13502767407159477, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017389853785610774, "output_power": 0.0016138236540086679, "delta_db": -0.32439853773633376, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00033443608489121105, "output_power": 0.000300277575006131, "delta_db": -0.4679023076434551, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00017248781948922851, "output_power": 0.0001797479288272063, "delta_db": 0.1790546278595706, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012677719609633915, "output_power": 0.00012247890726644708, "delta_db": -0.14979839331024075, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21483572867698897, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09535117374735805, "pitch": null, "output_rms_dbfs": -23.5343885862588, "output_peak": 0.5351369585224263, "reference_rms_dbfs": -23.43903741251144, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.486019986242844}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 42560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00020029577381909937, "log_rms_mae_db": 0.9032009780077146, "rms_envelope_nrmse": 0.10884219869837478, "envelope_correlation": 0.9881450989168559, "rms_gain_error_db": -0.09535201564948906, "reference_mean_power": 0.004538506592765079, "output_mean_power": 0.004439946827713351}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.8045928558777289e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008527253032911584, "output_power": 0.0008269704913781484, "reference_active": true, "delta_db": -0.13319137198449346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002893325022312121, "output_power": 0.00029463619616456254, "reference_active": true, "delta_db": 0.07888876064241418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.722708282094668e-05, "output_power": 6.867436577459362e-05, "reference_active": true, "delta_db": -1.5099259818546737}], "active_band_count": 3, "active_band_mae_db": 0.5740020381605272}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00017887965336918488, "log_rms_mae_db": 0.4936575791636094, "rms_envelope_nrmse": 0.05998542388989604, "envelope_correlation": 0.9960788019299723, "rms_gain_error_db": -0.09535330436880042, "reference_mean_power": 0.004590346759257262, "output_mean_power": 0.004490659882825963}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.4591224648550268e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007420820227306516, "output_power": 0.000719085195826726, "reference_active": true, "delta_db": -0.13671562934433212}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015779167985385838, "output_power": 0.00016486939670320292, "reference_active": true, "delta_db": 0.19055949048468895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.56670863086132e-06, "output_power": 5.0773908680516514e-06, "reference_active": true, "delta_db": -1.117071492069878}], "active_band_count": 3, "active_band_mae_db": 0.48144887063296643}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00017771570730226456, "log_rms_mae_db": 1.392485068761405, "rms_envelope_nrmse": 0.11242261893782303, "envelope_correlation": 0.9894398730841297, "rms_gain_error_db": -0.11660138250827888, "reference_mean_power": 0.003163634497093627, "output_mean_power": 0.003079825881676062}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.820001870942524e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009255585566014781, "output_power": 0.0008985776210548206, "reference_active": true, "delta_db": -0.12848301782467905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002598152375676334, "output_power": 0.0002570788459236028, "reference_active": true, "delta_db": -0.0459827624770547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.760488032618478e-05, "output_power": 5.34446314467758e-05, "reference_active": true, "delta_db": -1.619849469973108}], "active_band_count": 3, "active_band_mae_db": 0.5981050834249473}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00016184521030355344, "log_rms_mae_db": 0.8951438335734178, "rms_envelope_nrmse": 0.06664504757668384, "envelope_correlation": 0.9960244334145958, "rms_gain_error_db": -0.11660078508332512, "reference_mean_power": 0.0031997824540876, "output_mean_power": 0.0031150166628064824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.5556561049407657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008585656786559776, "output_power": 0.0008337453531854793, "reference_active": true, "delta_db": -0.1274009705191514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015662214176101382, "output_power": 0.00015723455405062953, "reference_active": true, "delta_db": 0.01694834739071092}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.505375823689367e-06, "output_power": 5.133547195622016e-06, "reference_active": true, "delta_db": -1.649548848638849}], "active_band_count": 3, "active_band_mae_db": 0.597966055516237}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00011375403588925618, "log_rms_mae_db": 1.4538672818529264, "rms_envelope_nrmse": 0.20025802567507814, "envelope_correlation": 0.9682364867265664, "rms_gain_error_db": -0.2952197774073399, "reference_mean_power": 0.0003949518827357503, "output_mean_power": 0.000368996469321595}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.6254954818630519e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.926011413944617e-05, "output_power": 8.19466660305549e-05, "reference_active": true, "delta_db": 0.14476595748020127}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0135009616582686e-05, "output_power": 2.1365097030967015e-05, "reference_active": true, "delta_db": 0.25753028169365483}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.142336744382203e-06, "output_power": 5.621521735853047e-06, "reference_active": true, "delta_db": -1.0398642808165512}], "active_band_count": 3, "active_band_mae_db": 0.4807201733301358}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 8.70060383669589e-05, "log_rms_mae_db": 0.8552931667056762, "rms_envelope_nrmse": 0.12263007597882968, "envelope_correlation": 0.9876808835096509, "rms_gain_error_db": -0.2952444937430398, "reference_mean_power": 0.00039945208139827864, "output_mean_power": 0.00037319880038542654}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.3661376304062792e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.270190105592004e-05, "output_power": 7.58329135890453e-05, "reference_active": true, "delta_db": 0.18311975189917476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.983801499648517e-06, "output_power": 1.0349419958170224e-05, "reference_active": true, "delta_db": 0.1562007240391285}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.054433356878425e-07, "output_power": 3.9861628165090513e-07, "reference_active": true, "delta_db": -1.0311744104059921}], "active_band_count": 3, "active_band_mae_db": 0.4568316287814318}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 6.751108516993363e-05, "log_rms_mae_db": 1.3584682897310014, "rms_envelope_nrmse": 0.14719778125854335, "envelope_correlation": 0.985289980009236, "rms_gain_error_db": -0.12937064586617777, "reference_mean_power": 0.00015129473773616968, "output_mean_power": 0.00014685433069256516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 2.0948865520528105e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010579126587837114, "output_power": 0.00010495508339563595, "reference_active": true, "delta_db": -0.03446335797020386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588508061007275e-05, "output_power": 1.4671016927007173e-05, "reference_active": true, "delta_db": -0.3452920485922203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6030646339801794e-06, "output_power": 4.033936328858799e-06, "reference_active": true, "delta_db": -1.4269659320550687}], "active_band_count": 3, "active_band_mae_db": 0.602240446205831}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 5.7873331101151596e-05, "log_rms_mae_db": 0.8247972034799822, "rms_envelope_nrmse": 0.08049622120732156, "envelope_correlation": 0.9954457808531028, "rms_gain_error_db": -0.12937187306879958, "reference_mean_power": 0.0001530185286286403, "output_mean_power": 0.0001485274874177555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.9708112174484802e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.70769534187325e-05, "output_power": 9.79878206949239e-05, "reference_active": true, "delta_db": 0.040559604761902866}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.36755214697274e-06, "output_power": 6.550226396998425e-06, "reference_active": true, "delta_db": -0.5106690756370993}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2457588916034453e-07, "output_power": 2.2395241746420274e-07, "reference_active": true, "delta_db": -1.6116050068096404}], "active_band_count": 3, "active_band_mae_db": 0.7209445624028809}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9047711748274841, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a212254f5e42b4b4f85c5f7a9f9fe167921a5ca29c8793a0b6a715f4eeb6d924", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9685705979697258, "standard_deviation": 0.007842901902071268, "minimum": 0.9583339146584413, "maximum": 0.9829326796077356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4494696935966915, "standard_deviation": 0.033449166665074526, "minimum": 1.4101394209512297, "maximum": 1.5233563972295912, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4497217205528142, "standard_deviation": 0.17556753742117426, "minimum": 0.10210999025963835, "maximum": 0.7109583408170016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19938116653379506, "standard_deviation": 0.02678900895105878, "minimum": 0.14749321078384467, "maximum": 0.23447297572987355, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9890956941826263, "standard_deviation": 0.004845190375449435, "minimum": 0.9794826524343926, "maximum": 0.9940709657200084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8751850537425316, "standard_deviation": 0.035406337591710274, "minimum": 0.8266141122942876, "maximum": 0.9224895732739833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4833041706369102, "standard_deviation": 0.174821282807757, "minimum": 0.14389802006366673, "maximum": 0.7301565925555128, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11451590478409696, "standard_deviation": 0.027852513786107695, "minimum": 0.08391362403692675, "maximum": 0.16817378377035064, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9841911239009854, "standard_deviation": 0.0010254339962357465, "minimum": 0.9829869069564272, "maximum": 0.9858595780087354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3893959297037242, "standard_deviation": 0.03392284884150042, "minimum": 1.3584682897310014, "maximum": 1.4549900965449236, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5998249789597621, "standard_deviation": 0.37065868006072306, "minimum": 0.16948995744576764, "maximum": 1.1820251456634334, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15474310858767146, "standard_deviation": 0.006951484399971101, "minimum": 0.14528304872790954, "maximum": 0.16790248242810837, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937828090935928, "standard_deviation": 0.0011289061674827348, "minimum": 0.9923074776350364, "maximum": 0.9954457808531028, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8793289976157466, "standard_deviation": 0.03716085215615722, "minimum": 0.8247972034799822, "maximum": 0.9432114005944469, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4937630314982156, "standard_deviation": 0.30063500628067474, "minimum": 0.09107061424367728, "maximum": 0.8963514663577415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09523316655613531, "standard_deviation": 0.010305877216560356, "minimum": 0.08049622120732156, "maximum": 0.11412485463439033, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9898867662684834, "standard_deviation": 0.0016603119199093972, "minimum": 0.9877369934616578, "maximum": 0.9928611820537281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.335764701994528, "standard_deviation": 0.0634961393173103, "minimum": 1.2479820756318343, "maximum": 1.3930012838630286, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8142399218769936, "standard_deviation": 0.4739806967620031, "minimum": 0.06116065388563327, "maximum": 1.3425093608691956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1104367710560607, "standard_deviation": 0.009449078070025718, "minimum": 0.09260959227012037, "maximum": 0.12265493735953309, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965237478513562, "standard_deviation": 0.0010807224886109756, "minimum": 0.9942754818448662, "maximum": 0.9979529762428114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7999689700762994, "standard_deviation": 0.08247119698789397, "minimum": 0.7235576966109271, "maximum": 0.9428749940959401, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9298652231346303, "standard_deviation": 0.49271896077868144, "minimum": 0.1535862306648991, "maximum": 1.5064794083491113, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06355964086409252, "standard_deviation": 0.009598054406029607, "minimum": 0.0483713847321848, "maximum": 0.08273652935435169, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18454340030322364, "standard_deviation": 0.05490286437035793, "minimum": 0.09723144974856011, "maximum": 0.27363557039671926, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.478790391638714, "standard_deviation": 0.01454070639917351, "minimum": 5.4591321206941785, "maximum": 5.499768957021897, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5227485235215075, "standard_deviation": 0.01333007540669931, "minimum": 0.49973226790319597, "maximum": 0.5405040060466771, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11268220333098722, "standard_deviation": 0.05596683140774978, "minimum": -0.20558247888490994, "maximum": -0.017241378702317133, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9897536335975365, "standard_deviation": 0.0012816757547423768, "minimum": 0.9881450989168559, "maximum": 0.9914094620968151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8845621945392697, "standard_deviation": 0.030636237380125282, "minimum": 0.8440290883012538, "maximum": 0.9310236232146624, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6178359918017113, "standard_deviation": 0.3109925728183124, "minimum": 0.1680644307895601, "maximum": 1.0876444488455717, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10140809245224805, "standard_deviation": 0.006549912999201732, "minimum": 0.09262375911105329, "maximum": 0.10976570924322483, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972641923057739, "standard_deviation": 0.0009431896002564243, "minimum": 0.9956205113679432, "maximum": 0.9981743951287934, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48543958931750253, "standard_deviation": 0.028740934244426714, "minimum": 0.43203235297564213, "maximum": 0.5230095995425508, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7562198394705872, "standard_deviation": 0.4730850287150149, "minimum": 0.20937869965305245, "maximum": 1.2983250888266566, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05057940455231954, "standard_deviation": 0.008778756893913634, "minimum": 0.042031904341997446, "maximum": 0.06623848117512063, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0041", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.706559658050537, "predicted_mos": 2.706559658050537, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "c17d5fe64175c6cea9a77e4642cbfccaec1b17112b485eaf9d4eeb3836111d9d", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8292641998268664, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8948987306249896, "全帯域_10ms_rms_envelope_nrmse": 0.10124083875265645, "全帯域_10ms_envelope_correlation": 0.9900876917900702, "全帯域_10ms_modulation_mae_db": 0.8716897017800607, "全帯域_40ms_log_rms_mae_db": 0.4905347820723073, "全帯域_40ms_rms_envelope_nrmse": 0.06698615082637238, "全帯域_40ms_envelope_correlation": 0.9954710479888094, "全帯域_40ms_modulation_mae_db": 0.9413575901095476, "300–1000 Hz_10ms_log_rms_mae_db": 1.319193682906995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12336278305650793, "300–1000 Hz_10ms_envelope_correlation": 0.9877292776019053, "300–1000 Hz_10ms_modulation_mae_db": 1.0679367406786235, "300–1000 Hz_40ms_log_rms_mae_db": 0.8421950868104454, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09351903956368163, "300–1000 Hz_40ms_envelope_correlation": 0.9926329635949539, "300–1000 Hz_40ms_modulation_mae_db": 1.1158719627651958, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4798588596789246, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21362392010115558, "1000–3000 Hz_10ms_envelope_correlation": 0.9636154208217352, "1000–3000 Hz_10ms_modulation_mae_db": 0.6685586485877479, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8174610953828777, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11759384144825662, "1000–3000 Hz_40ms_envelope_correlation": 0.9887632542640421, "1000–3000 Hz_40ms_modulation_mae_db": 0.7940802762790735, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5921166658437558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18805696687443035, "3000–6000 Hz_10ms_envelope_correlation": 0.979216864803998, "3000–6000 Hz_10ms_modulation_mae_db": 0.6021299671968353, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0247661540292974, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13279021898652368, "3000–6000 Hz_40ms_envelope_correlation": 0.9900106510521605, "3000–6000 Hz_40ms_modulation_mae_db": 0.45505024009170314, "mr_stft_db": 5.590756161999575, "band_mae_db": 0.27235442197692616, "rms_gain_error_db": -0.19427005075310239, "output_peak": 0.4920198815481416}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8460926264562173, "全帯域_10ms_rms_envelope_nrmse": 0.09362001947285577, "全帯域_10ms_envelope_correlation": 0.9912530984617043, "全帯域_10ms_modulation_mae_db": 0.7470611453481308, "全帯域_40ms_log_rms_mae_db": 0.4443361330006819, "全帯域_40ms_rms_envelope_nrmse": 0.047296558630449306, "全帯域_40ms_envelope_correlation": 0.9975875297128883, "全帯域_40ms_modulation_mae_db": 1.0652179311051182, "300–1000 Hz_10ms_log_rms_mae_db": 1.2993832178313809, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10591697813103292, "300–1000 Hz_10ms_envelope_correlation": 0.9908375003389487, "300–1000 Hz_10ms_modulation_mae_db": 0.8849126040209514, "300–1000 Hz_40ms_log_rms_mae_db": 0.6934188426072799, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05959414744467816, "300–1000 Hz_40ms_envelope_correlation": 0.9971218450492054, "300–1000 Hz_40ms_modulation_mae_db": 1.2282568778799534, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5083705956080227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19096809130500023, "1000–3000 Hz_10ms_envelope_correlation": 0.9731312414352775, "1000–3000 Hz_10ms_modulation_mae_db": 0.3796567279666019, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8840370830534566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12673507039068763, "1000–3000 Hz_40ms_envelope_correlation": 0.9875359878744547, "1000–3000 Hz_40ms_modulation_mae_db": 0.4791164940399981, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5689229651494183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1630826786390062, "3000–6000 Hz_10ms_envelope_correlation": 0.9819469716308197, "3000–6000 Hz_10ms_modulation_mae_db": 0.687600871411039, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0558748038042636, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10918681318500231, "3000–6000 Hz_40ms_envelope_correlation": 0.991444109142036, "3000–6000 Hz_40ms_modulation_mae_db": 0.5049108462346547, "mr_stft_db": 5.618884645784561, "band_mae_db": 0.10864917638708409, "rms_gain_error_db": -0.0814757361774106, "output_peak": 0.6050408795497529}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8927109424073978, "全帯域_10ms_rms_envelope_nrmse": 0.09607471811123484, "全帯域_10ms_envelope_correlation": 0.9907897470538969, "全帯域_10ms_modulation_mae_db": 0.6587518128771735, "全帯域_40ms_log_rms_mae_db": 0.4726258738812123, "全帯域_40ms_rms_envelope_nrmse": 0.043332659550399194, "全帯域_40ms_envelope_correlation": 0.9979881196125565, "全帯域_40ms_modulation_mae_db": 1.023442338400588, "300–1000 Hz_10ms_log_rms_mae_db": 1.4443139760718036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11148954109993812, "300–1000 Hz_10ms_envelope_correlation": 0.9896087279172545, "300–1000 Hz_10ms_modulation_mae_db": 0.9144555567744274, "300–1000 Hz_40ms_log_rms_mae_db": 0.7587186883781518, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05559626238440382, "300–1000 Hz_40ms_envelope_correlation": 0.997243564156414, "300–1000 Hz_40ms_modulation_mae_db": 1.0979178875527011, "1000–3000 Hz_10ms_log_rms_mae_db": 1.601138871541124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1984241320857775, "1000–3000 Hz_10ms_envelope_correlation": 0.9688308443704065, "1000–3000 Hz_10ms_modulation_mae_db": 0.6425197401321464, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9622955327847407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11483467086875877, "1000–3000 Hz_40ms_envelope_correlation": 0.9894533417508772, "1000–3000 Hz_40ms_modulation_mae_db": 0.7054740976878137, "3000–6000 Hz_10ms_log_rms_mae_db": 1.533831615002238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15733736235668552, "3000–6000 Hz_10ms_envelope_correlation": 0.9836575109235327, "3000–6000 Hz_10ms_modulation_mae_db": 0.31686992082626286, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9798764831835576, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10758590595379507, "3000–6000 Hz_40ms_envelope_correlation": 0.9919934696153527, "3000–6000 Hz_40ms_modulation_mae_db": 0.34320429576312367, "mr_stft_db": 5.57701743773407, "band_mae_db": 0.17124169112994236, "rms_gain_error_db": -0.07691721533776322, "output_peak": 0.5465289082083045}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9353139511543869, "全帯域_10ms_rms_envelope_nrmse": 0.09793975134713023, "全帯域_10ms_envelope_correlation": 0.9904686674718802, "全帯域_10ms_modulation_mae_db": 0.7832269099410221, "全帯域_40ms_log_rms_mae_db": 0.5091306566483467, "全帯域_40ms_rms_envelope_nrmse": 0.050923001187426015, "全帯域_40ms_envelope_correlation": 0.9972686284231013, "全帯域_40ms_modulation_mae_db": 0.9329611238453296, "300–1000 Hz_10ms_log_rms_mae_db": 1.4475454666778618, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11730846470304712, "300–1000 Hz_10ms_envelope_correlation": 0.9885947902122536, "300–1000 Hz_10ms_modulation_mae_db": 0.8699665649725453, "300–1000 Hz_40ms_log_rms_mae_db": 0.8179441525988354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.066089891632412, "300–1000 Hz_40ms_envelope_correlation": 0.9962720099319904, "300–1000 Hz_40ms_modulation_mae_db": 1.3819098258692966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5010752244787025, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20805649168719142, "1000–3000 Hz_10ms_envelope_correlation": 0.9672562689287679, "1000–3000 Hz_10ms_modulation_mae_db": 0.5031572577380785, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9121770525534653, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14175783384202637, "1000–3000 Hz_40ms_envelope_correlation": 0.9855937231408163, "1000–3000 Hz_40ms_modulation_mae_db": 0.699492912716476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6030958592384064, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18074362119368004, "3000–6000 Hz_10ms_envelope_correlation": 0.9787628805459111, "3000–6000 Hz_10ms_modulation_mae_db": 0.7343309993672733, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0025834966618756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12011917353472311, "3000–6000 Hz_40ms_envelope_correlation": 0.9902718643291246, "3000–6000 Hz_40ms_modulation_mae_db": 0.7232590921183896, "mr_stft_db": 5.5819038280369915, "band_mae_db": 0.18347780737132605, "rms_gain_error_db": -0.11281445503828666, "output_peak": 0.5109489375790385}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9136990012802272, "全帯域_10ms_rms_envelope_nrmse": 0.09389278825416009, "全帯域_10ms_envelope_correlation": 0.9911661112172347, "全帯域_10ms_modulation_mae_db": 0.6082252130831122, "全帯域_40ms_log_rms_mae_db": 0.5191685890926919, "全帯域_40ms_rms_envelope_nrmse": 0.047737035188272735, "全帯域_40ms_envelope_correlation": 0.9974785381575393, "全帯域_40ms_modulation_mae_db": 0.6412708661990821, "300–1000 Hz_10ms_log_rms_mae_db": 1.3913352768383906, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11364838957354308, "300–1000 Hz_10ms_envelope_correlation": 0.9892036889375129, "300–1000 Hz_10ms_modulation_mae_db": 0.7530957389460338, "300–1000 Hz_40ms_log_rms_mae_db": 0.8624875235291954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0714660586205796, "300–1000 Hz_40ms_envelope_correlation": 0.9954124748834398, "300–1000 Hz_40ms_modulation_mae_db": 1.1359031169824378, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5893777207974489, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21790048798852404, "1000–3000 Hz_10ms_envelope_correlation": 0.9625302683557619, "1000–3000 Hz_10ms_modulation_mae_db": 0.6051745603807307, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8991444907596088, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10496630274360867, "1000–3000 Hz_40ms_envelope_correlation": 0.9904246350717164, "1000–3000 Hz_40ms_modulation_mae_db": 0.5618939918737941, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4782420647250836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1719513143858646, "3000–6000 Hz_10ms_envelope_correlation": 0.9799644985659081, "3000–6000 Hz_10ms_modulation_mae_db": 1.0067049860995574, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9286624600583218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12073028724363546, "3000–6000 Hz_40ms_envelope_correlation": 0.9895582086021758, "3000–6000 Hz_40ms_modulation_mae_db": 0.9119485548389327, "mr_stft_db": 5.560056867946095, "band_mae_db": 0.13140010870405294, "rms_gain_error_db": -0.035453216965234804, "output_peak": 0.5315518587818099}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.915664345455519, "全帯域_10ms_rms_envelope_nrmse": 0.10800503552107697, "全帯域_10ms_envelope_correlation": 0.9883047461975295, "全帯域_10ms_modulation_mae_db": 0.12210836439427857, "全帯域_40ms_log_rms_mae_db": 0.5334746954247209, "全帯域_40ms_rms_envelope_nrmse": 0.04720678190954016, "全帯域_40ms_envelope_correlation": 0.9975225501560172, "全帯域_40ms_modulation_mae_db": 0.05740811067635177, "300–1000 Hz_10ms_log_rms_mae_db": 1.3637753084763782, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10497801061474384, "300–1000 Hz_10ms_envelope_correlation": 0.9907635927647208, "300–1000 Hz_10ms_modulation_mae_db": 0.14110356655696069, "300–1000 Hz_40ms_log_rms_mae_db": 0.8086897802423679, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05153704560830836, "300–1000 Hz_40ms_envelope_correlation": 0.9976012518486195, "300–1000 Hz_40ms_modulation_mae_db": 0.13050422571099513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5402079200345593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21352329289039354, "1000–3000 Hz_10ms_envelope_correlation": 0.9635062431376882, "1000–3000 Hz_10ms_modulation_mae_db": 0.47342438755796407, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9157703627993629, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.110418036130466, "1000–3000 Hz_40ms_envelope_correlation": 0.989766952861435, "1000–3000 Hz_40ms_modulation_mae_db": 0.6734435351307347, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5787436333228495, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17616981762250994, "3000–6000 Hz_10ms_envelope_correlation": 0.9803175921668417, "3000–6000 Hz_10ms_modulation_mae_db": 0.45347130828938703, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9691508604132401, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12216487690135963, "3000–6000 Hz_40ms_envelope_correlation": 0.9903172804299145, "3000–6000 Hz_40ms_modulation_mae_db": 0.6038338753185086, "mr_stft_db": 5.584771515152911, "band_mae_db": 0.12476564321227897, "rms_gain_error_db": -0.02274313991812152, "output_peak": 0.5130325638535599}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9315548024250668, "全帯域_10ms_rms_envelope_nrmse": 0.12146864019192186, "全帯域_10ms_envelope_correlation": 0.9853922417041389, "全帯域_10ms_modulation_mae_db": 0.8539398369179857, "全帯域_40ms_log_rms_mae_db": 0.5150490881702264, "全帯域_40ms_rms_envelope_nrmse": 0.055594706759211396, "全帯域_40ms_envelope_correlation": 0.9970186080157022, "全帯域_40ms_modulation_mae_db": 0.8391870664907087, "300–1000 Hz_10ms_log_rms_mae_db": 1.484822268224346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1367252438041013, "300–1000 Hz_10ms_envelope_correlation": 0.9846429783342958, "300–1000 Hz_10ms_modulation_mae_db": 0.9892561556757228, "300–1000 Hz_40ms_log_rms_mae_db": 0.8880166680581618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07038796454332237, "300–1000 Hz_40ms_envelope_correlation": 0.9961661248122448, "300–1000 Hz_40ms_modulation_mae_db": 1.132092538459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5562260898607647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2222710483996833, "1000–3000 Hz_10ms_envelope_correlation": 0.9607696315888865, "1000–3000 Hz_10ms_modulation_mae_db": 0.5767427321393965, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9121600872350666, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12637689051407677, "1000–3000 Hz_40ms_envelope_correlation": 0.9874195779194651, "1000–3000 Hz_40ms_modulation_mae_db": 0.786595613426066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5831866308132485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16678186031479686, "3000–6000 Hz_10ms_envelope_correlation": 0.9812078648463372, "3000–6000 Hz_10ms_modulation_mae_db": 0.5082117851024037, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9879119260908097, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12433420649025358, "3000–6000 Hz_40ms_envelope_correlation": 0.989185810363377, "3000–6000 Hz_40ms_modulation_mae_db": 0.1826996273450795, "mr_stft_db": 5.602659453945378, "band_mae_db": 0.259836408113584, "rms_gain_error_db": -0.18893766332688564, "output_peak": 0.48407025473165455}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8629630042693837, "全帯域_10ms_rms_envelope_nrmse": 0.09045851590865543, "全帯域_10ms_envelope_correlation": 0.9919150275096044, "全帯域_10ms_modulation_mae_db": 0.49906452229322973, "全帯域_40ms_log_rms_mae_db": 0.4751579347222972, "全帯域_40ms_rms_envelope_nrmse": 0.043276721007946314, "全帯域_40ms_envelope_correlation": 0.9981169842842923, "全帯域_40ms_modulation_mae_db": 0.7211080346629548, "300–1000 Hz_10ms_log_rms_mae_db": 1.4723193144200966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1015858546641993, "300–1000 Hz_10ms_envelope_correlation": 0.9916660866746752, "300–1000 Hz_10ms_modulation_mae_db": 0.7302827143276772, "300–1000 Hz_40ms_log_rms_mae_db": 0.9142740010202476, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057528878120291546, "300–1000 Hz_40ms_envelope_correlation": 0.9974445425388693, "300–1000 Hz_40ms_modulation_mae_db": 0.9553541304839636, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5220754232447977, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22573229940761141, "1000–3000 Hz_10ms_envelope_correlation": 0.9594728895270415, "1000–3000 Hz_10ms_modulation_mae_db": 0.8743032534560315, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9030985678050857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10339884132257322, "1000–3000 Hz_40ms_envelope_correlation": 0.9906658355499092, "1000–3000 Hz_40ms_modulation_mae_db": 0.8062101340913926, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5647280830824337, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15961748538529874, "3000–6000 Hz_10ms_envelope_correlation": 0.984076921862992, "3000–6000 Hz_10ms_modulation_mae_db": 0.33784516125575575, "3000–6000 Hz_40ms_log_rms_mae_db": 0.918345887765704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11208502604000667, "3000–6000 Hz_40ms_envelope_correlation": 0.9920745892526305, "3000–6000 Hz_40ms_modulation_mae_db": 0.3661716020033973, "mr_stft_db": 5.580244036541515, "band_mae_db": 0.1390621816816516, "rms_gain_error_db": -0.12590020482158376, "output_peak": 0.5445434512211016}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 42560, "output_samples": 42560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7382138929017845, "spectral_convergence": 0.30714784795078665, "spectral_convergence_status": "ok", "amplitude_floor": 1.6028753096638316e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.406456878301321, "spectral_convergence": 0.29552391827576224, "spectral_convergence_status": "ok", "amplitude_floor": 1.0331614165752405e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.627597714795621, "spectral_convergence": 0.3007004384689765, "spectral_convergence_status": "ok", "amplitude_floor": 7.831529815953997e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042273796895084704, "output_power": 0.00044652903694589955, "delta_db": 0.23778448998421672, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0017345552824327432, "output_power": 0.0016791602460432295, "delta_db": -0.14096002087474768, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017389853785610774, "output_power": 0.0015834041504884124, "delta_db": -0.40704151460676974, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00033443608489121105, "output_power": 0.0003063550851051161, "delta_db": -0.38088037215464976, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00017248781948922851, "output_power": 0.00018403155959962117, "delta_db": 0.2813387454813395, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012677719609633915, "output_power": 0.0001323284669734783, "delta_db": 0.18612138875983336, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27235442197692616, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.19427005075310239, "pitch": null, "output_rms_dbfs": -23.633307463264543, "output_peak": 0.4920198815481416, "reference_rms_dbfs": -23.43903741251144, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.590756161999575}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 42560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00020029577381909937, "log_rms_mae_db": 0.8948987306249896, "rms_envelope_nrmse": 0.10124083875265645, "envelope_correlation": 0.9900876917900702, "rms_gain_error_db": -0.1942690315570789, "reference_mean_power": 0.004538506592765079, "output_mean_power": 0.004339963404808375}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.8045928558777289e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008527253032911584, "output_power": 0.0008061746274286077, "reference_active": true, "delta_db": -0.24380024615238982}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002893325022312121, "output_power": 0.0002606331458340713, "reference_active": true, "delta_db": -0.45367576976727375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.722708282094668e-05, "output_power": 6.252129234773113e-05, "reference_active": true, "delta_db": -1.9175930894205186}], "active_band_count": 3, "active_band_mae_db": 0.8716897017800607}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00017887965336918488, "log_rms_mae_db": 0.4905347820723073, "rms_envelope_nrmse": 0.06698615082637238, "envelope_correlation": 0.9954710479888094, "rms_gain_error_db": -0.19427023770951712, "reference_mean_power": 0.004590346759257262, "output_mean_power": 0.004389534533112477}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.4591224648550268e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007420820227306516, "output_power": 0.0006873436712863752, "reference_active": true, "delta_db": -0.3327797247984008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015779167985385838, "output_power": 0.0001437793178840385, "reference_active": true, "delta_db": -0.4038768079850428}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.56670863086132e-06, "output_power": 4.060748784942615e-06, "reference_active": true, "delta_db": -2.0874162375451992}], "active_band_count": 3, "active_band_mae_db": 0.9413575901095476}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00017771570730226456, "log_rms_mae_db": 1.319193682906995, "rms_envelope_nrmse": 0.12336278305650793, "envelope_correlation": 0.9877292776019053, "rms_gain_error_db": -0.27895317242919127, "reference_mean_power": 0.003163634497093627, "output_mean_power": 0.0029668185468734553}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.820001870942524e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009255585566014781, "output_power": 0.0008907905036812184, "reference_active": true, "delta_db": -0.16628321565519094}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002598152375676334, "output_power": 0.00022614327646852897, "reference_active": true, "delta_db": -0.6028093772053215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.760488032618478e-05, "output_power": 4.430138161020661e-05, "reference_active": true, "delta_db": -2.4347176291753585}], "active_band_count": 3, "active_band_mae_db": 1.0679367406786235}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00016184521030355344, "log_rms_mae_db": 0.8421950868104454, "rms_envelope_nrmse": 0.09351903956368163, "envelope_correlation": 0.9926329635949539, "rms_gain_error_db": -0.27895430108075553, "reference_mean_power": 0.0031997824540876, "output_mean_power": 0.003000716888239584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.5556561049407657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008585656786559776, "output_power": 0.0008145448937720026, "reference_active": true, "delta_db": -0.22858497962979596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015662214176101382, "output_power": 0.0001378675643332347, "reference_active": true, "delta_db": -0.5539105541838925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.505375823689367e-06, "output_power": 4.1577694945210675e-06, "reference_active": true, "delta_db": -2.5651203544818992}], "active_band_count": 3, "active_band_mae_db": 1.1158719627651958}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00011375403588925618, "log_rms_mae_db": 1.4798588596789246, "rms_envelope_nrmse": 0.21362392010115558, "envelope_correlation": 0.9636154208217352, "rms_gain_error_db": -0.3079716066250251, "reference_mean_power": 0.0003949518827357503, "output_mean_power": 0.00036791460461874337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.6254954818630519e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.926011413944617e-05, "output_power": 8.038192406637117e-05, "reference_active": true, "delta_db": 0.06103704084896164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0135009616582686e-05, "output_power": 1.9458009485419526e-05, "reference_active": true, "delta_db": -0.1485343059755581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.142336744382203e-06, "output_power": 4.7231295986348585e-06, "reference_active": true, "delta_db": -1.7961045989387239}], "active_band_count": 3, "active_band_mae_db": 0.6685586485877479}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 8.70060383669589e-05, "log_rms_mae_db": 0.8174610953828777, "rms_envelope_nrmse": 0.11759384144825662, "envelope_correlation": 0.9887632542640421, "rms_gain_error_db": -0.3079868531513021, "reference_mean_power": 0.00039945208139827864, "output_mean_power": 0.0003721054261996539}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.3661376304062792e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.270190105592004e-05, "output_power": 7.372517558732733e-05, "reference_active": true, "delta_db": 0.060700483430570815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.983801499648517e-06, "output_power": 9.5181269653043e-06, "reference_active": true, "delta_db": -0.20744443983706168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.054433356878425e-07, "output_power": 3.106439387688514e-07, "reference_active": true, "delta_db": -2.114095905569588}], "active_band_count": 3, "active_band_mae_db": 0.7940802762790735}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 6.751108516993363e-05, "log_rms_mae_db": 1.5921166658437558, "rms_envelope_nrmse": 0.18805696687443035, "envelope_correlation": 0.979216864803998, "rms_gain_error_db": 0.45938491676873966, "reference_mean_power": 0.00015129473773616968, "output_mean_power": 0.00016817534006335257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 2.0948865520528105e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010579126587837114, "output_power": 0.00011406738538014378, "reference_active": true, "delta_db": 0.32711673039485933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588508061007275e-05, "output_power": 1.8566319441142867e-05, "reference_active": true, "delta_db": 0.6773639536654037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6030646339801794e-06, "output_power": 4.658377833318431e-06, "reference_active": true, "delta_db": -0.8019092175302428}], "active_band_count": 3, "active_band_mae_db": 0.6021299671968353}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 5.7873331101151596e-05, "log_rms_mae_db": 1.0247661540292974, "rms_envelope_nrmse": 0.13279021898652368, "envelope_correlation": 0.9900106510521605, "rms_gain_error_db": 0.45942612893414897, "reference_mean_power": 0.0001530185286286403, "output_mean_power": 0.00017009307577611624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.9708112174484802e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.70769534187325e-05, "output_power": 0.00010694423109367794, "reference_active": true, "delta_db": 0.42041223994801674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.36755214697274e-06, "output_power": 8.668052858211422e-06, "reference_active": true, "delta_db": 0.7059833256881041}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2457588916034453e-07, "output_power": 3.0721381399931446e-07, "reference_active": true, "delta_db": -0.2387551546389885}], "active_band_count": 3, "active_band_mae_db": 0.45505024009170314}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9079053226846009, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3cd19d941db89a154c31a4926241779cfdf83b1909aa3bc62ff64f964a35797e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9648891010206957, "standard_deviation": 0.004543090942963518, "minimum": 0.9594728895270415, "maximum": 0.9731312414352775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.537291338155543, "standard_deviation": 0.04281948811669317, "minimum": 1.4798588596789246, "maximum": 1.601138871541124, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5904421634948372, "standard_deviation": 0.1490629828783387, "minimum": 0.3796567279666019, "maximum": 0.8743032534560315, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21131247048316712, "standard_deviation": 0.011785334922065406, "minimum": 0.19096809130500023, "maximum": 0.22573229940761141, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9887029135540895, "standard_deviation": 0.0017399540959543416, "minimum": 0.9855937231408163, "maximum": 0.9906658355499092, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.900768034046708, "standard_deviation": 0.04052985774036056, "minimum": 0.8174610953828777, "maximum": 0.9622955327847407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6882883819056685, "standard_deviation": 0.11657704787643633, "minimum": 0.4791164940399981, "maximum": 0.8062101340913926, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11826018590755674, "standard_deviation": 0.012879154768852081, "minimum": 0.10339884132257322, "maximum": 0.14175783384202637, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9811438881682926, "standard_deviation": 0.0019649196033579832, "minimum": 0.9787628805459111, "maximum": 0.984076921862992, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5628584396471792, "standard_deviation": 0.03997039969026266, "minimum": 1.4782420647250836, "maximum": 1.6030958592384064, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5808956249435643, "standard_deviation": 0.22897159227092134, "minimum": 0.31686992082626286, "maximum": 1.0067049860995574, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17046763834653403, "standard_deviation": 0.010743429892962491, "minimum": 0.15733736235668552, "maximum": 0.18805696687443035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906069978483465, "standard_deviation": 0.0010986027956677786, "minimum": 0.989185810363377, "maximum": 0.9920745892526305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9833965090008838, "standard_deviation": 0.045974851443166334, "minimum": 0.918345887765704, "maximum": 1.0558748038042636, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5113847667142237, "standard_deviation": 0.23130371536666258, "minimum": 0.1826996273450795, "maximum": 0.9119485548389327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11862456354191243, "standard_deviation": 0.008503412849981086, "minimum": 0.10758590595379507, "maximum": 0.13279021898652368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9891308303476958, "standard_deviation": 0.002224293129908196, "minimum": 0.9846429783342958, "maximum": 0.9916660866746752, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4028360639309065, "standard_deviation": 0.07033141715418682, "minimum": 1.2993832178313809, "maximum": 1.484822268224346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7938762052441177, "standard_deviation": 0.28628620724514053, "minimum": 0.14110356655696069, "maximum": 1.0679367406786235, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1143769082058892, "standard_deviation": 0.011471557660119988, "minimum": 0.1015858546641993, "maximum": 0.1367252438041013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962368471019671, "standard_deviation": 0.0016384065921765255, "minimum": 0.9926329635949539, "maximum": 0.9976012518486195, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8232180929055857, "standard_deviation": 0.07135339925945319, "minimum": 0.6934188426072799, "maximum": 0.9142740010202476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.022226320712943, "standard_deviation": 0.3800287053606702, "minimum": 0.13050422571099513, "maximum": 1.3819098258692966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06571491098970969, "standard_deviation": 0.013288002653906238, "minimum": 0.05153704560830836, "maximum": 0.09351903956368163, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17384842982210577, "standard_deviation": 0.06195492429979866, "minimum": 0.10864917638708409, "maximum": 0.27235442197692616, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.587036743392638, "standard_deviation": 0.017639692080792368, "minimum": 5.560056867946095, "maximum": 5.618884645784561, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5284670919341704, "standard_deviation": 0.03836353916143227, "minimum": 0.48407025473165455, "maximum": 0.6050408795497529, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10481396029229857, "standard_deviation": 0.06375988496245749, "minimum": -0.19427005075310239, "maximum": -0.02274313991812152, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899221664257574, "standard_deviation": 0.002121498742205066, "minimum": 0.9853922417041389, "maximum": 0.9919150275096044, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8991121755091485, "standard_deviation": 0.03168236662154922, "minimum": 0.8460926264562173, "maximum": 0.9353139511543869, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6430084383293742, "standard_deviation": 0.24517918385571483, "minimum": 0.12210836439427857, "maximum": 0.8716897017800607, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10033753844496146, "standard_deviation": 0.010101668154305365, "minimum": 0.09045851590865543, "maximum": 0.12146864019192186, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973065007938633, "standard_deviation": 0.0008220134431239628, "minimum": 0.9954710479888094, "maximum": 0.9981169842842923, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4949347191265606, "standard_deviation": 0.029633728014636887, "minimum": 0.4443361330006819, "maximum": 0.5334746954247209, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7777441326862101, "standard_deviation": 0.3247773500300136, "minimum": 0.05740811067635177, "maximum": 1.0652179311051182, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.050294201882452184, "standard_deviation": 0.00783651834301778, "minimum": 0.043276721007946314, "maximum": 0.06698615082637238, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6829-68769-0041", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.735257863998413, "predicted_mos": 2.735257863998413, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "ff88101b0fe8a19f85cc5ca71dd057afd3150046d2e16e8d8c9c9de001a5a4ea", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6327173998579383, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8078739483851356, "全帯域_10ms_rms_envelope_nrmse": 0.09885162118206989, "全帯域_10ms_envelope_correlation": 0.9903820680227875, "全帯域_10ms_modulation_mae_db": 0.303136859266905, "全帯域_40ms_log_rms_mae_db": 0.436844870641145, "全帯域_40ms_rms_envelope_nrmse": 0.04625806612760529, "全帯域_40ms_envelope_correlation": 0.9979431158938752, "全帯域_40ms_modulation_mae_db": 0.1990110945891684, "300–1000 Hz_10ms_log_rms_mae_db": 1.3179750648126345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1099243723537964, "300–1000 Hz_10ms_envelope_correlation": 0.9904710495226734, "300–1000 Hz_10ms_modulation_mae_db": 0.473134637554248, "300–1000 Hz_40ms_log_rms_mae_db": 0.7299056440769341, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06535178620531307, "300–1000 Hz_40ms_envelope_correlation": 0.9969494634773515, "300–1000 Hz_40ms_modulation_mae_db": 0.3968974285062843, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4896787477754188, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22723719510964316, "1000–3000 Hz_10ms_envelope_correlation": 0.9595076094615744, "1000–3000 Hz_10ms_modulation_mae_db": 0.284779225590089, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8567817168154772, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11662024920537531, "1000–3000 Hz_40ms_envelope_correlation": 0.9882611489933276, "1000–3000 Hz_40ms_modulation_mae_db": 0.32324565807639155, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5114728181325898, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17108093721461093, "3000–6000 Hz_10ms_envelope_correlation": 0.9815885652619243, "3000–6000 Hz_10ms_modulation_mae_db": 0.47281433519130345, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9621689869168744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11505849739924724, "3000–6000 Hz_40ms_envelope_correlation": 0.9915946279160555, "3000–6000 Hz_40ms_modulation_mae_db": 0.4492840976565679, "mr_stft_db": 5.572981237508046, "band_mae_db": 0.186723899345164, "rms_gain_error_db": -0.15822621420004346, "output_peak": 0.5329771950217072}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9083669944998494, "全帯域_10ms_rms_envelope_nrmse": 0.10480745086104297, "全帯域_10ms_envelope_correlation": 0.9893200276271706, "全帯域_10ms_modulation_mae_db": 0.47088285262036367, "全帯域_40ms_log_rms_mae_db": 0.5246554754645628, "全帯域_40ms_rms_envelope_nrmse": 0.05459258359374629, "全帯域_40ms_envelope_correlation": 0.9972127137031581, "全帯域_40ms_modulation_mae_db": 0.7145223642140772, "300–1000 Hz_10ms_log_rms_mae_db": 1.409559374211687, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11116998310715896, "300–1000 Hz_10ms_envelope_correlation": 0.9901768431492849, "300–1000 Hz_10ms_modulation_mae_db": 0.8749094938643206, "300–1000 Hz_40ms_log_rms_mae_db": 0.739420808447159, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06369436704813548, "300–1000 Hz_40ms_envelope_correlation": 0.9970712660368695, "300–1000 Hz_40ms_modulation_mae_db": 1.162489853516844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5608532313029857, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2072618380676797, "1000–3000 Hz_10ms_envelope_correlation": 0.9657311127130328, "1000–3000 Hz_10ms_modulation_mae_db": 0.5685241580212659, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9172211828963704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1054279440154176, "1000–3000 Hz_40ms_envelope_correlation": 0.9909469027983142, "1000–3000 Hz_40ms_modulation_mae_db": 0.6075860499614896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4507028412749334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15911278296260878, "3000–6000 Hz_10ms_envelope_correlation": 0.9828617280690587, "3000–6000 Hz_10ms_modulation_mae_db": 0.3011546574809496, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8704184548571287, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10562302822473281, "3000–6000 Hz_40ms_envelope_correlation": 0.9921879827996904, "3000–6000 Hz_40ms_modulation_mae_db": 0.15456087668614857, "mr_stft_db": 5.542616419869336, "band_mae_db": 0.21914338683251885, "rms_gain_error_db": -0.20006984617717852, "output_peak": 0.5601875701681265}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9296164472707314, "全帯域_10ms_rms_envelope_nrmse": 0.09550451534053638, "全帯域_10ms_envelope_correlation": 0.9911378585910476, "全帯域_10ms_modulation_mae_db": 0.9997611627804831, "全帯域_40ms_log_rms_mae_db": 0.514508099764703, "全帯域_40ms_rms_envelope_nrmse": 0.044976905387389896, "全帯域_40ms_envelope_correlation": 0.9981732765127374, "全帯域_40ms_modulation_mae_db": 1.1204785162154385, "300–1000 Hz_10ms_log_rms_mae_db": 1.3814671281200188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10704888728178516, "300–1000 Hz_10ms_envelope_correlation": 0.9909805221451925, "300–1000 Hz_10ms_modulation_mae_db": 1.2087478896028576, "300–1000 Hz_40ms_log_rms_mae_db": 0.775454720221544, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05972662542728085, "300–1000 Hz_40ms_envelope_correlation": 0.9975635712252696, "300–1000 Hz_40ms_modulation_mae_db": 1.3987385862081065, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5143820902031353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21198310371520102, "1000–3000 Hz_10ms_envelope_correlation": 0.9643356009889384, "1000–3000 Hz_10ms_modulation_mae_db": 0.7553412937783436, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8660512533300857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10015857988586266, "1000–3000 Hz_40ms_envelope_correlation": 0.9912516701953777, "1000–3000 Hz_40ms_modulation_mae_db": 0.529423843392301, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5183951309128918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1786123646138582, "3000–6000 Hz_10ms_envelope_correlation": 0.9783352336135861, "3000–6000 Hz_10ms_modulation_mae_db": 0.5494760227888422, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9811892269231699, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12399822115777819, "3000–6000 Hz_40ms_envelope_correlation": 0.9890519565779153, "3000–6000 Hz_40ms_modulation_mae_db": 0.6311553278734829, "mr_stft_db": 5.565640863458462, "band_mae_db": 0.20643271095092733, "rms_gain_error_db": -0.177897067199064, "output_peak": 0.5157350981758506}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9025485993754936, "全帯域_10ms_rms_envelope_nrmse": 0.11687646173764017, "全帯域_10ms_envelope_correlation": 0.9865428394680437, "全帯域_10ms_modulation_mae_db": 1.041351886937382, "全帯域_40ms_log_rms_mae_db": 0.4895328621464259, "全帯域_40ms_rms_envelope_nrmse": 0.05699929774954578, "全帯域_40ms_envelope_correlation": 0.9968823904278489, "全帯域_40ms_modulation_mae_db": 1.263115594928668, "300–1000 Hz_10ms_log_rms_mae_db": 1.2836105842270482, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11644497651847463, "300–1000 Hz_10ms_envelope_correlation": 0.989383729843713, "300–1000 Hz_10ms_modulation_mae_db": 1.3003565380235111, "300–1000 Hz_40ms_log_rms_mae_db": 0.8132148746794352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07114543304237696, "300–1000 Hz_40ms_envelope_correlation": 0.9964204440099518, "300–1000 Hz_40ms_modulation_mae_db": 1.6314682306224044, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4692504979573007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26878753223749524, "1000–3000 Hz_10ms_envelope_correlation": 0.9417617088155851, "1000–3000 Hz_10ms_modulation_mae_db": 0.5239484314714467, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8355522232963707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1184638618875928, "1000–3000 Hz_40ms_envelope_correlation": 0.9879508816873456, "1000–3000 Hz_40ms_modulation_mae_db": 0.5120295773412024, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4866549572737524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15843544920535546, "3000–6000 Hz_10ms_envelope_correlation": 0.9842649767053493, "3000–6000 Hz_10ms_modulation_mae_db": 0.3539671997248351, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9014382443279465, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10747640449591317, "3000–6000 Hz_40ms_envelope_correlation": 0.9927197457620455, "3000–6000 Hz_40ms_modulation_mae_db": 0.42329192929016507, "mr_stft_db": 5.555397910806957, "band_mae_db": 0.18393655137612294, "rms_gain_error_db": -0.19899168005197693, "output_peak": 0.5031977783823034}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8719235678021409, "全帯域_10ms_rms_envelope_nrmse": 0.10169475755863353, "全帯域_10ms_envelope_correlation": 0.9896288831535469, "全帯域_10ms_modulation_mae_db": 0.830471379932177, "全帯域_40ms_log_rms_mae_db": 0.46795520080570824, "全帯域_40ms_rms_envelope_nrmse": 0.048993520474294255, "全帯域_40ms_envelope_correlation": 0.9973462894841438, "全帯域_40ms_modulation_mae_db": 1.0442400391667213, "300–1000 Hz_10ms_log_rms_mae_db": 1.332577638248483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1107200242488146, "300–1000 Hz_10ms_envelope_correlation": 0.9897201421080238, "300–1000 Hz_10ms_modulation_mae_db": 0.8707230580663738, "300–1000 Hz_40ms_log_rms_mae_db": 0.7516896633872185, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06173404792992143, "300–1000 Hz_40ms_envelope_correlation": 0.9965440594053581, "300–1000 Hz_40ms_modulation_mae_db": 1.1211405161156893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5457555911130378, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2352240865619595, "1000–3000 Hz_10ms_envelope_correlation": 0.9555997856040293, "1000–3000 Hz_10ms_modulation_mae_db": 0.5492103709907908, "1000–3000 Hz_40ms_log_rms_mae_db": 0.858239426767554, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12289197610219456, "1000–3000 Hz_40ms_envelope_correlation": 0.9876773280751868, "1000–3000 Hz_40ms_modulation_mae_db": 0.5798520798957245, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4993782160146558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15992527312402105, "3000–6000 Hz_10ms_envelope_correlation": 0.9829035850725764, "3000–6000 Hz_10ms_modulation_mae_db": 0.7299036393886972, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9126745567785864, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10317628644052527, "3000–6000 Hz_40ms_envelope_correlation": 0.9925589999405837, "3000–6000 Hz_40ms_modulation_mae_db": 0.25178803527675603, "mr_stft_db": 5.565207235833487, "band_mae_db": 0.15331946428881618, "rms_gain_error_db": -0.048994714217743714, "output_peak": 0.5288765651287919}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8337378550534086, "全帯域_10ms_rms_envelope_nrmse": 0.10553119334719363, "全帯域_10ms_envelope_correlation": 0.9888770998428742, "全帯域_10ms_modulation_mae_db": 0.7980459362469853, "全帯域_40ms_log_rms_mae_db": 0.42090878664368897, "全帯域_40ms_rms_envelope_nrmse": 0.04912669288832496, "全帯域_40ms_envelope_correlation": 0.9974296525733604, "全帯域_40ms_modulation_mae_db": 1.010730224349839, "300–1000 Hz_10ms_log_rms_mae_db": 1.3769966796216224, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11588492119519796, "300–1000 Hz_10ms_envelope_correlation": 0.9890417438533866, "300–1000 Hz_10ms_modulation_mae_db": 0.92242517063757, "300–1000 Hz_40ms_log_rms_mae_db": 0.7487348679369753, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06629054066515201, "300–1000 Hz_40ms_envelope_correlation": 0.996466804696645, "300–1000 Hz_40ms_modulation_mae_db": 1.1567226060400249, "1000–3000 Hz_10ms_log_rms_mae_db": 1.540966223959441, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23022931090672236, "1000–3000 Hz_10ms_envelope_correlation": 0.9592156510993416, "1000–3000 Hz_10ms_modulation_mae_db": 0.6412608505996108, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8658271813981925, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12311826607285616, "1000–3000 Hz_40ms_envelope_correlation": 0.9873540306899764, "1000–3000 Hz_40ms_modulation_mae_db": 0.6017783609985291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.506064017937316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17101916121307492, "3000–6000 Hz_10ms_envelope_correlation": 0.9801668405374826, "3000–6000 Hz_10ms_modulation_mae_db": 0.4552465513581974, "3000–6000 Hz_40ms_log_rms_mae_db": 0.965925857457243, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11705132964718373, "3000–6000 Hz_40ms_envelope_correlation": 0.9903032458854975, "3000–6000 Hz_40ms_modulation_mae_db": 0.27541179298445867, "mr_stft_db": 5.557895114734503, "band_mae_db": 0.16600846995600332, "rms_gain_error_db": -0.09838337766723212, "output_peak": 0.556736936684653}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8805652619535962, "全帯域_10ms_rms_envelope_nrmse": 0.09791200808159244, "全帯域_10ms_envelope_correlation": 0.990422289906596, "全帯域_10ms_modulation_mae_db": 0.4357706134025055, "全帯域_40ms_log_rms_mae_db": 0.44827492010863984, "全帯域_40ms_rms_envelope_nrmse": 0.04482692469068335, "全帯域_40ms_envelope_correlation": 0.997838152277122, "全帯域_40ms_modulation_mae_db": 0.47250113191758464, "300–1000 Hz_10ms_log_rms_mae_db": 1.4264788901854066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1134083687643166, "300–1000 Hz_10ms_envelope_correlation": 0.9893298567878588, "300–1000 Hz_10ms_modulation_mae_db": 0.5704837598455118, "300–1000 Hz_40ms_log_rms_mae_db": 0.7554462293500834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05548614881156855, "300–1000 Hz_40ms_envelope_correlation": 0.9974501296909919, "300–1000 Hz_40ms_modulation_mae_db": 0.7136707101977199, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4490345807009646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18740202905655612, "1000–3000 Hz_10ms_envelope_correlation": 0.9727243725841412, "1000–3000 Hz_10ms_modulation_mae_db": 0.6850073899410738, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8749074406137045, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10661903626803827, "1000–3000 Hz_40ms_envelope_correlation": 0.9903498496633312, "1000–3000 Hz_40ms_modulation_mae_db": 0.9308170331825852, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5625339667710971, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1631479250317427, "3000–6000 Hz_10ms_envelope_correlation": 0.9820958030650889, "3000–6000 Hz_10ms_modulation_mae_db": 0.7164957637661878, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9747900491476604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10834274239415034, "3000–6000 Hz_40ms_envelope_correlation": 0.9916606108221917, "3000–6000 Hz_40ms_modulation_mae_db": 0.4010427888291607, "mr_stft_db": 5.571590593934121, "band_mae_db": 0.11529718475639333, "rms_gain_error_db": -0.06709629825879442, "output_peak": 0.50087769534629}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9163208798230997, "全帯域_10ms_rms_envelope_nrmse": 0.09358952856263664, "全帯域_10ms_envelope_correlation": 0.991513034195654, "全帯域_10ms_modulation_mae_db": 0.39099020518635874, "全帯域_40ms_log_rms_mae_db": 0.4833657733377693, "全帯域_40ms_rms_envelope_nrmse": 0.05040962555299114, "全帯域_40ms_envelope_correlation": 0.9976160742006416, "全帯域_40ms_modulation_mae_db": 0.3970179918135712, "300–1000 Hz_10ms_log_rms_mae_db": 1.3766562287237083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11276451806015778, "300–1000 Hz_10ms_envelope_correlation": 0.9897767564997819, "300–1000 Hz_10ms_modulation_mae_db": 0.401955780987518, "300–1000 Hz_40ms_log_rms_mae_db": 0.7244249012027931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06475592248319656, "300–1000 Hz_40ms_envelope_correlation": 0.9968321071349628, "300–1000 Hz_40ms_modulation_mae_db": 0.5210890051086357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5000297135086147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19141528471823718, "1000–3000 Hz_10ms_envelope_correlation": 0.9715309996205993, "1000–3000 Hz_10ms_modulation_mae_db": 0.6386047085699793, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8624611441511587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1207686333315346, "1000–3000 Hz_40ms_envelope_correlation": 0.9888334623333355, "1000–3000 Hz_40ms_modulation_mae_db": 0.5828568066654213, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5199735888593544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17117535069257098, "3000–6000 Hz_10ms_envelope_correlation": 0.9805044796236085, "3000–6000 Hz_10ms_modulation_mae_db": 0.6061193747936334, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9895170764021695, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11312852549352875, "3000–6000 Hz_40ms_envelope_correlation": 0.9917111802870321, "3000–6000 Hz_40ms_modulation_mae_db": 0.6469927377324007, "mr_stft_db": 5.561149216884869, "band_mae_db": 0.26928150041242965, "rms_gain_error_db": -0.18555432590773352, "output_peak": 0.48957162920889297}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 42560, "output_samples": 42560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.744667874949125, "spectral_convergence": 0.29605258658807043, "spectral_convergence_status": "ok", "amplitude_floor": 1.6028753096638316e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.387917530483079, "spectral_convergence": 0.27976844811292556, "spectral_convergence_status": "ok", "amplitude_floor": 1.0331614165752405e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.586358307091935, "spectral_convergence": 0.2853717429148534, "spectral_convergence_status": "ok", "amplitude_floor": 7.831529815953997e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042273796895084704, "output_power": 0.0004371314642579412, "delta_db": 0.14540811169554443, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0017345552824327432, "output_power": 0.0017194114581956794, "delta_db": -0.03808329343746997, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017389853785610774, "output_power": 0.0015691276408261188, "delta_db": -0.44637657710850864, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00033443608489121105, "output_power": 0.00033183077788016986, "delta_db": -0.033964655568127416, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00017248781948922851, "output_power": 0.0001708865481365743, "delta_db": -0.04050554841495763, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012677719609633915, "output_power": 0.000139521669176692, "delta_db": 0.4160052098463758, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.186723899345164, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15822621420004346, "pitch": null, "output_rms_dbfs": -23.597263626711484, "output_peak": 0.5329771950217072, "reference_rms_dbfs": -23.43903741251144, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.572981237508046}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 42560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00020029577381909937, "log_rms_mae_db": 0.8078739483851356, "rms_envelope_nrmse": 0.09885162118206989, "envelope_correlation": 0.9903820680227875, "rms_gain_error_db": -0.15822489802761164, "reference_mean_power": 0.004538506592765079, "output_mean_power": 0.00437613267944655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.8045928558777289e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008527253032911584, "output_power": 0.0008081573057549691, "reference_active": true, "delta_db": -0.23313246870495158}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002893325022312121, "output_power": 0.00025944155236124123, "reference_active": true, "delta_db": -0.4735768869444013}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.722708282094668e-05, "output_power": 9.279340971383577e-05, "reference_active": true, "delta_db": -0.20270122215136194}], "active_band_count": 3, "active_band_mae_db": 0.303136859266905}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00017887965336918488, "log_rms_mae_db": 0.436844870641145, "rms_envelope_nrmse": 0.04625806612760529, "envelope_correlation": 0.9979431158938752, "rms_gain_error_db": -0.15822383806836846, "reference_mean_power": 0.004590346759257262, "output_mean_power": 0.004426119243268854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.4591224648550268e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007420820227306516, "output_power": 0.0007017081604671108, "reference_active": true, "delta_db": -0.2429538358183964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015779167985385838, "output_power": 0.0001499402459139495, "reference_active": true, "delta_db": -0.2216588091647101}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.56670863086132e-06, "output_power": 6.7700178017702e-06, "reference_active": true, "delta_db": 0.13242063878439875}], "active_band_count": 3, "active_band_mae_db": 0.1990110945891684}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00017771570730226456, "log_rms_mae_db": 1.3179750648126345, "rms_envelope_nrmse": 0.1099243723537964, "envelope_correlation": 0.9904710495226734, "rms_gain_error_db": -0.28801051637920855, "reference_mean_power": 0.003163634497093627, "output_mean_power": 0.0029606376037711233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.820001870942524e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009255585566014781, "output_power": 0.0008736361798572747, "reference_active": true, "delta_db": -0.2507328898149127}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002598152375676334, "output_power": 0.00022728788607030836, "reference_active": true, "delta_db": -0.5808832842971099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.760488032618478e-05, "output_power": 6.778135706210584e-05, "reference_active": true, "delta_db": -0.5877877385507214}], "active_band_count": 3, "active_band_mae_db": 0.473134637554248}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00016184521030355344, "log_rms_mae_db": 0.7299056440769341, "rms_envelope_nrmse": 0.06535178620531307, "envelope_correlation": 0.9969494634773515, "rms_gain_error_db": -0.2880110381551746, "reference_mean_power": 0.0031997824540876, "output_mean_power": 0.0029944657412214588}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.5556561049407657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008585656786559776, "output_power": 0.000810588985777376, "reference_active": true, "delta_db": -0.24972825192167636}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015662214176101382, "output_power": 0.00014378749918945487, "reference_active": true, "delta_db": -0.37132028168151066}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.505375823689367e-06, "output_power": 6.582760716505704e-06, "reference_active": true, "delta_db": -0.569643751915666}], "active_band_count": 3, "active_band_mae_db": 0.3968974285062843}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00011375403588925618, "log_rms_mae_db": 1.4896787477754188, "rms_envelope_nrmse": 0.22723719510964316, "envelope_correlation": 0.9595076094615744, "rms_gain_error_db": 0.03814216016847131, "reference_mean_power": 0.0003949518827357503, "output_mean_power": 0.0003984358467974358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.6254954818630519e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.926011413944617e-05, "output_power": 7.9566400257201e-05, "reference_active": true, "delta_db": 0.016750167466930786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0135009616582686e-05, "output_power": 2.0173557483116204e-05, "reference_active": true, "delta_db": 0.008306487693675863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.142336744382203e-06, "output_power": 5.900818004347911e-06, "reference_active": true, "delta_db": -0.8292810216096603}], "active_band_count": 3, "active_band_mae_db": 0.284779225590089}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 8.70060383669589e-05, "log_rms_mae_db": 0.8567817168154772, "rms_envelope_nrmse": 0.11662024920537531, "envelope_correlation": 0.9882611489933276, "rms_gain_error_db": 0.03817040597043843, "reference_mean_power": 0.00039945208139827864, "output_mean_power": 0.00040297836367454807}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.3661376304062792e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.270190105592004e-05, "output_power": 7.334060596493855e-05, "reference_active": true, "delta_db": 0.037987267095151496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.983801499648517e-06, "output_power": 9.978291103411299e-06, "reference_active": true, "delta_db": -0.0023976792286220477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.054433356878425e-07, "output_power": 4.080724466844926e-07, "reference_active": true, "delta_db": -0.9293520279054011}], "active_band_count": 3, "active_band_mae_db": 0.32324565807639155}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 6.751108516993363e-05, "log_rms_mae_db": 1.5114728181325898, "rms_envelope_nrmse": 0.17108093721461093, "envelope_correlation": 0.9815885652619243, "rms_gain_error_db": 0.2865159782622716, "reference_mean_power": 0.00015129473773616968, "output_mean_power": 0.00016161267572533799}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 2.0948865520528105e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010579126587837114, "output_power": 0.00011293378489669366, "reference_active": true, "delta_db": 0.2837406961136684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588508061007275e-05, "output_power": 1.682338959648613e-05, "reference_active": true, "delta_db": 0.2492407941555817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6030646339801794e-06, "output_power": 4.5696136258358504e-06, "reference_active": true, "delta_db": -0.8854615153046602}], "active_band_count": 3, "active_band_mae_db": 0.47281433519130345}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 5.7873331101151596e-05, "log_rms_mae_db": 0.9621689869168744, "rms_envelope_nrmse": 0.11505849739924724, "envelope_correlation": 0.9915946279160555, "rms_gain_error_db": 0.28651463165134794, "reference_mean_power": 0.0001530185286286403, "output_mean_power": 0.00016345397433742575}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.9708112174484802e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.70769534187325e-05, "output_power": 0.0001061544138420463, "reference_active": true, "delta_db": 0.3882191826088076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.36755214697274e-06, "output_power": 7.733804792384982e-06, "reference_active": true, "delta_db": 0.21069987488184844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2457588916034453e-07, "output_power": 2.7316366562524367e-07, "reference_active": true, "delta_db": -0.7489332354790478}], "active_band_count": 3, "active_band_mae_db": 0.4492840976565679}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9163827262588702, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1d5243db43ed0dd1da2b08857fb42a90e57a5656252b1500ff25dd4665084396", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9613008551109052, "standard_deviation": 0.009898295021415414, "minimum": 0.9417617088155851, "maximum": 0.9727243725841412, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5087438345651123, "standard_deviation": 0.03913285745761198, "minimum": 1.4490345807009646, "maximum": 1.5608532313029857, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.580834553620325, "standard_deviation": 0.14158779874402155, "minimum": 0.284779225590089, "maximum": 0.7553412937783436, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2199425475466868, "standard_deviation": 0.026390557171353288, "minimum": 0.18740202905655612, "maximum": 0.26878753223749524, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9890781593045244, "standard_deviation": 0.0015476541281203551, "minimum": 0.9873540306899764, "maximum": 0.9912516701953777, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8671301961586142, "standard_deviation": 0.023230397790661737, "minimum": 0.8355522232963707, "maximum": 0.9172211828963704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5834486761892056, "standard_deviation": 0.1678952191468389, "minimum": 0.32324565807639155, "maximum": 0.9308170331825852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.114258568346109, "standard_deviation": 0.008894821934124308, "minimum": 0.10015857988586266, "maximum": 0.12311826607285616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9815901514935843, "standard_deviation": 0.0018727893096225857, "minimum": 0.9783352336135861, "maximum": 0.9842649767053493, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.506896942147074, "standard_deviation": 0.0317228266485434, "minimum": 1.4507028412749334, "maximum": 1.5625339667710971, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5231471930615808, "standard_deviation": 0.15709684812314073, "minimum": 0.3011546574809496, "maximum": 0.7299036393886972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16656365550723037, "standard_deviation": 0.007406706068018869, "minimum": 0.15843544920535546, "maximum": 0.1786123646138582, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9914735437488764, "standard_deviation": 0.0012294868769641744, "minimum": 0.9890519565779153, "maximum": 0.9927197457620455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9447653066013473, "standard_deviation": 0.04377877821117974, "minimum": 0.8704184548571287, "maximum": 0.9895170764021695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4041909482911426, "standard_deviation": 0.17525474537866198, "minimum": 0.15456087668614857, "maximum": 0.6469927377324007, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11173187940663244, "standard_deviation": 0.006885681297839016, "minimum": 0.10317628644052527, "maximum": 0.12399822115777819, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9898600804887394, "standard_deviation": 0.0006473306557751099, "minimum": 0.9890417438533866, "maximum": 0.9909805221451925, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.363165198518826, "standard_deviation": 0.04807138318030331, "minimum": 1.2836105842270482, "maximum": 1.4264788901854066, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.827842041072739, "standard_deviation": 0.3283702274564279, "minimum": 0.401955780987518, "maximum": 1.3003565380235111, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11217075644121277, "standard_deviation": 0.003124423844618662, "minimum": 0.10704888728178516, "maximum": 0.11644497651847463, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969122307096749, "standard_deviation": 0.00043465695558962886, "minimum": 0.9964204440099518, "maximum": 0.9975635712252696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7547864636627678, "standard_deviation": 0.028447159876133166, "minimum": 0.7244249012027931, "maximum": 0.8132148746794352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0127771170394635, "standard_deviation": 0.4304733965583593, "minimum": 0.3968974285062843, "maximum": 1.6314682306224044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06352310895161811, "standard_deviation": 0.004672999597287546, "minimum": 0.05548614881156855, "maximum": 0.07114543304237696, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18751789598979696, "standard_deviation": 0.04614372675877558, "minimum": 0.11529718475639333, "maximum": 0.26928150041242965, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5615598241287225, "standard_deviation": 0.00979390447893466, "minimum": 5.542616419869336, "maximum": 5.572981237508046, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.523520058514577, "standard_deviation": 0.02592508476269171, "minimum": 0.48957162920889297, "maximum": 0.5601875701681265, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14190169045997084, "standard_deviation": 0.06121613645786265, "minimum": -0.20006984617717852, "maximum": -0.048994714217743714, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9897280126009651, "standard_deviation": 0.0015643088878360772, "minimum": 0.9865428394680437, "maximum": 0.991513034195654, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8813691942704318, "standard_deviation": 0.042235780316786875, "minimum": 0.8078739483851356, "maximum": 0.9296164472707314, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6588013620466451, "standard_deviation": 0.2914824845504333, "minimum": 0.303136859266905, "maximum": 1.041351886937382, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1018459420839182, "standard_deviation": 0.007378683977764675, "minimum": 0.09358952856263664, "maximum": 0.11687646173764017, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975552081341109, "standard_deviation": 0.000421839563120406, "minimum": 0.9968823904278489, "maximum": 0.9981732765127374, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4732557486140804, "standard_deviation": 0.03670138082803972, "minimum": 0.42090878664368897, "maximum": 0.5246554754645628, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7777021196493836, "standard_deviation": 0.3882947135924916, "minimum": 0.1990110945891684, "maximum": 1.263115594928668, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049522952058072625, "standard_deviation": 0.004411010757674909, "minimum": 0.04482692469068335, "maximum": 0.05699929774954578, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_6930-75918-0013", "language": "en", "style": "normal", "speaker": "librispeech_6930", "source_utterance_id": "6930-75918-0013", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 47040, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_6930-75918-0013.wav", "source_file_sha256": "253c14a9b1fe1e647a8f64e464ea5983fae7b4c279bbebf98e347961f64df121", "decoded_pcm_sha256": "3f54b1539585ee3c865737cd0781dd64a40895a5d1930b99c37db78e450618a7", "original_source_sha256": "83f99c357f446d1fddffc61c09f59eb8a41c82c5e4261db42c71396c537c409e", "corpus_entry_id": "librispeech_6930-75918-0013"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.062913417816162, "predicted_mos": 4.062913417816162, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "f38f497c4ae6fc9108020c809f36529449591a97f64f030035a5b56ed251477a", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.080563588999212, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_6930-75918-0013", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f38f497c4ae6fc9108020c809f36529449591a97f64f030035a5b56ed251477a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "321a0414ac42f02e98ce930c29201d21e63bf4d633ddb31c75f7bbcc7654bddb"}, "rows": [{"case": "librispeech_6930-75918-0013", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.922168493270874, "predicted_mos": 1.922168493270874, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "bff5ffd19058ad3474597bf90bdee552f2dfe5882ba99ee460f781008ed14218", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8934949079994112, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_6930-75918-0013", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "bff5ffd19058ad3474597bf90bdee552f2dfe5882ba99ee460f781008ed14218", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.233982016555169, "全帯域_10ms_rms_envelope_nrmse": 0.144818867627008, "全帯域_10ms_envelope_correlation": 0.9777152115715965, "全帯域_10ms_modulation_mae_db": 0.8468328905865197, "全帯域_40ms_log_rms_mae_db": 0.7609038265219712, "全帯域_40ms_rms_envelope_nrmse": 0.07986724228287136, "全帯域_40ms_envelope_correlation": 0.9930489063835506, "全帯域_40ms_modulation_mae_db": 1.750933646287815, "300–1000 Hz_10ms_log_rms_mae_db": 2.0201963522381856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.25580340892632103, "300–1000 Hz_10ms_envelope_correlation": 0.9492796120486827, "300–1000 Hz_10ms_modulation_mae_db": 2.3842635105371395, "300–1000 Hz_40ms_log_rms_mae_db": 1.2984791455613374, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14672707232643858, "300–1000 Hz_40ms_envelope_correlation": 0.9885919674771058, "300–1000 Hz_40ms_modulation_mae_db": 1.8589868187746952, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5708856209026119, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19167437459369768, "1000–3000 Hz_10ms_envelope_correlation": 0.966833859101778, "1000–3000 Hz_10ms_modulation_mae_db": 0.5971661868503494, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0390833597835745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0982273576042246, "1000–3000 Hz_40ms_envelope_correlation": 0.990608162448071, "1000–3000 Hz_40ms_modulation_mae_db": 1.0359847121788048, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6915601213721485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18957766267687143, "3000–6000 Hz_10ms_envelope_correlation": 0.9721612366922758, "3000–6000 Hz_10ms_modulation_mae_db": 0.8855047399337271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0639225588122896, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11067119627064811, "3000–6000 Hz_40ms_envelope_correlation": 0.9900180051469993, "3000–6000 Hz_40ms_modulation_mae_db": 1.640696455834681, "mr_stft_db": 6.162981039733211, "band_mae_db": 0.26635057531250306, "rms_gain_error_db": -0.24218281855410084, "output_peak": 0.2770707343303237}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.210593254627184, "全帯域_10ms_rms_envelope_nrmse": 0.15474773094270083, "全帯域_10ms_envelope_correlation": 0.9745107561813166, "全帯域_10ms_modulation_mae_db": 0.863676934590928, "全帯域_40ms_log_rms_mae_db": 0.6937082480942774, "全帯域_40ms_rms_envelope_nrmse": 0.10208359486705218, "全帯域_40ms_envelope_correlation": 0.9880675069178543, "全帯域_40ms_modulation_mae_db": 1.1963512143095092, "300–1000 Hz_10ms_log_rms_mae_db": 2.0264957848872225, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20550321936569726, "300–1000 Hz_10ms_envelope_correlation": 0.9658981242837017, "300–1000 Hz_10ms_modulation_mae_db": 1.73943094372391, "300–1000 Hz_40ms_log_rms_mae_db": 1.366711341849986, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1085827063238629, "300–1000 Hz_40ms_envelope_correlation": 0.991038420943503, "300–1000 Hz_40ms_modulation_mae_db": 1.1706757100729195, "1000–3000 Hz_10ms_log_rms_mae_db": 1.596422868823293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17437420954738872, "1000–3000 Hz_10ms_envelope_correlation": 0.9730644465054088, "1000–3000 Hz_10ms_modulation_mae_db": 0.4499622978728695, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0539845874183935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1008696530215931, "1000–3000 Hz_40ms_envelope_correlation": 0.9910049474305809, "1000–3000 Hz_40ms_modulation_mae_db": 0.6152174087188019, "3000–6000 Hz_10ms_log_rms_mae_db": 1.648536808202847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19762093809310696, "3000–6000 Hz_10ms_envelope_correlation": 0.970201085179867, "3000–6000 Hz_10ms_modulation_mae_db": 1.2367208603852207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0498970707612771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13457980308116485, "3000–6000 Hz_40ms_envelope_correlation": 0.9856779225250937, "3000–6000 Hz_40ms_modulation_mae_db": 1.5948462956208358, "mr_stft_db": 6.180593400871093, "band_mae_db": 0.15832260505506054, "rms_gain_error_db": -0.26801808353750506, "output_peak": 0.3277127246314766}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.286217802877545, "全帯域_10ms_rms_envelope_nrmse": 0.1455890065338626, "全帯域_10ms_envelope_correlation": 0.9774800196491923, "全帯域_10ms_modulation_mae_db": 0.5998641741171656, "全帯域_40ms_log_rms_mae_db": 0.6516258418901365, "全帯域_40ms_rms_envelope_nrmse": 0.07193238208418357, "全帯域_40ms_envelope_correlation": 0.9944787421057523, "全帯域_40ms_modulation_mae_db": 0.49416156605413636, "300–1000 Hz_10ms_log_rms_mae_db": 1.9243692343264251, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2042129298820148, "300–1000 Hz_10ms_envelope_correlation": 0.9662710896842731, "300–1000 Hz_10ms_modulation_mae_db": 1.879761669525191, "300–1000 Hz_40ms_log_rms_mae_db": 1.2060172704345828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08965228010836368, "300–1000 Hz_40ms_envelope_correlation": 0.9941028374478598, "300–1000 Hz_40ms_modulation_mae_db": 0.9595109908467251, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6375498853544508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20900855088918033, "1000–3000 Hz_10ms_envelope_correlation": 0.9612382749511904, "1000–3000 Hz_10ms_modulation_mae_db": 0.6261936552054879, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0363016971081123, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11662109843152135, "1000–3000 Hz_40ms_envelope_correlation": 0.9886725505714727, "1000–3000 Hz_40ms_modulation_mae_db": 0.9287635269057871, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6467404320351313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18186137584396533, "3000–6000 Hz_10ms_envelope_correlation": 0.9745234086375456, "3000–6000 Hz_10ms_modulation_mae_db": 0.6269125986593359, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0016001968471087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09360804565839093, "3000–6000 Hz_40ms_envelope_correlation": 0.9932682408466506, "3000–6000 Hz_40ms_modulation_mae_db": 0.5131715346762897, "mr_stft_db": 6.1487148045861035, "band_mae_db": 0.43937013974049344, "rms_gain_error_db": -0.2441498707684331, "output_peak": 0.2884508970707578}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2476169952200205, "全帯域_10ms_rms_envelope_nrmse": 0.15030139073897258, "全帯域_10ms_envelope_correlation": 0.9760126943216879, "全帯域_10ms_modulation_mae_db": 0.18634584435778942, "全帯域_40ms_log_rms_mae_db": 0.664987595868874, "全帯域_40ms_rms_envelope_nrmse": 0.07770707413352367, "全帯域_40ms_envelope_correlation": 0.9926992494870557, "全帯域_40ms_modulation_mae_db": 0.3268546802987505, "300–1000 Hz_10ms_log_rms_mae_db": 1.9820644990658984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23174208635903168, "300–1000 Hz_10ms_envelope_correlation": 0.9571743416917327, "300–1000 Hz_10ms_modulation_mae_db": 2.3985420253545047, "300–1000 Hz_40ms_log_rms_mae_db": 1.2284455392374938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09964872682462025, "300–1000 Hz_40ms_envelope_correlation": 0.9915143043075372, "300–1000 Hz_40ms_modulation_mae_db": 1.9514338959085566, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6537038768146768, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2157345923777302, "1000–3000 Hz_10ms_envelope_correlation": 0.9619603480777198, "1000–3000 Hz_10ms_modulation_mae_db": 0.6136273028365129, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0693065828503503, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12190300318038853, "1000–3000 Hz_40ms_envelope_correlation": 0.987631029828577, "1000–3000 Hz_40ms_modulation_mae_db": 0.7415601400656607, "3000–6000 Hz_10ms_log_rms_mae_db": 1.728508539310423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20638500148640934, "3000–6000 Hz_10ms_envelope_correlation": 0.9668547124453275, "3000–6000 Hz_10ms_modulation_mae_db": 0.4505132031129088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0597042600178483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11930088721013052, "3000–6000 Hz_40ms_envelope_correlation": 0.9876853000590387, "3000–6000 Hz_40ms_modulation_mae_db": 0.559872900358483, "mr_stft_db": 6.1597162776696806, "band_mae_db": 0.3783552660045291, "rms_gain_error_db": 0.016177175493858123, "output_peak": 0.3314710336625447}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2191089999199836, "全帯域_10ms_rms_envelope_nrmse": 0.1327891895298487, "全帯域_10ms_envelope_correlation": 0.981102119531493, "全帯域_10ms_modulation_mae_db": 0.3368940606546182, "全帯域_40ms_log_rms_mae_db": 0.5911968053781608, "全帯域_40ms_rms_envelope_nrmse": 0.06677429061549144, "全帯域_40ms_envelope_correlation": 0.9948192675099861, "全帯域_40ms_modulation_mae_db": 0.7739983222901315, "300–1000 Hz_10ms_log_rms_mae_db": 1.9210435753191022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.24008021046205308, "300–1000 Hz_10ms_envelope_correlation": 0.9548020452802597, "300–1000 Hz_10ms_modulation_mae_db": 1.4176394295127566, "300–1000 Hz_40ms_log_rms_mae_db": 1.2707112759457715, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15279941886404988, "300–1000 Hz_40ms_envelope_correlation": 0.9847810217583173, "300–1000 Hz_40ms_modulation_mae_db": 0.8745034727775768, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5493053888105193, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17298854354736568, "1000–3000 Hz_10ms_envelope_correlation": 0.9732186351249853, "1000–3000 Hz_10ms_modulation_mae_db": 0.8277938157591641, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9720837415199848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09951724101137698, "1000–3000 Hz_40ms_envelope_correlation": 0.9907296795329509, "1000–3000 Hz_40ms_modulation_mae_db": 0.7859831987488978, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6867146003291305, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16473797479898297, "3000–6000 Hz_10ms_envelope_correlation": 0.9789584015213286, "3000–6000 Hz_10ms_modulation_mae_db": 0.19320437081491657, "3000–6000 Hz_40ms_log_rms_mae_db": 0.984605841422036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07748492752782961, "3000–6000 Hz_40ms_envelope_correlation": 0.9950094243109656, "3000–6000 Hz_40ms_modulation_mae_db": 0.6834492462914333, "mr_stft_db": 6.147818804758468, "band_mae_db": 0.2890306576658382, "rms_gain_error_db": -0.14677525627646817, "output_peak": 0.3548741245987411}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2494839433069518, "全帯域_10ms_rms_envelope_nrmse": 0.1395288421467423, "全帯域_10ms_envelope_correlation": 0.9799492516219732, "全帯域_10ms_modulation_mae_db": 1.2018694636631004, "全帯域_40ms_log_rms_mae_db": 0.7125202305019928, "全帯域_40ms_rms_envelope_nrmse": 0.08336946755556568, "全帯域_40ms_envelope_correlation": 0.9931945025923223, "全帯域_40ms_modulation_mae_db": 1.6240101101173334, "300–1000 Hz_10ms_log_rms_mae_db": 1.9418452607610597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2037545773030243, "300–1000 Hz_10ms_envelope_correlation": 0.9674745356777089, "300–1000 Hz_10ms_modulation_mae_db": 1.36917622367828, "300–1000 Hz_40ms_log_rms_mae_db": 1.3998646074443089, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1338667454611393, "300–1000 Hz_40ms_envelope_correlation": 0.9871853312753505, "300–1000 Hz_40ms_modulation_mae_db": 0.7102016155093409, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6343878879378113, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2043800208108819, "1000–3000 Hz_10ms_envelope_correlation": 0.9633373756925326, "1000–3000 Hz_10ms_modulation_mae_db": 0.6318685411198008, "1000–3000 Hz_40ms_log_rms_mae_db": 1.000494512316435, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0986453696260841, "1000–3000 Hz_40ms_envelope_correlation": 0.9905727535867276, "1000–3000 Hz_40ms_modulation_mae_db": 0.5197946086242137, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7822585628481875, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18977243608169475, "3000–6000 Hz_10ms_envelope_correlation": 0.9733759322444303, "3000–6000 Hz_10ms_modulation_mae_db": 1.3014824917604848, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1200734228896292, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12214568270324395, "3000–6000 Hz_40ms_envelope_correlation": 0.9895843161340759, "3000–6000 Hz_40ms_modulation_mae_db": 1.6909859589216785, "mr_stft_db": 6.177827689204847, "band_mae_db": 0.4871919007252467, "rms_gain_error_db": -0.3293761784636797, "output_peak": 0.2759594497091529}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3167418711059868, "全帯域_10ms_rms_envelope_nrmse": 0.14085439043234796, "全帯域_10ms_envelope_correlation": 0.9786750803343436, "全帯域_10ms_modulation_mae_db": 0.33483920277809176, "全帯域_40ms_log_rms_mae_db": 0.6755144228302035, "全帯域_40ms_rms_envelope_nrmse": 0.07940210809816327, "全帯域_40ms_envelope_correlation": 0.9923235578649368, "全帯域_40ms_modulation_mae_db": 0.9679406982356457, "300–1000 Hz_10ms_log_rms_mae_db": 1.9885479121009613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.25472056699329926, "300–1000 Hz_10ms_envelope_correlation": 0.9534949355569432, "300–1000 Hz_10ms_modulation_mae_db": 2.6734109709921747, "300–1000 Hz_40ms_log_rms_mae_db": 1.3611851065361134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14575343094467966, "300–1000 Hz_40ms_envelope_correlation": 0.9851602247053279, "300–1000 Hz_40ms_modulation_mae_db": 2.7229813909420257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5563026281548569, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19072847850766314, "1000–3000 Hz_10ms_envelope_correlation": 0.9673382487331333, "1000–3000 Hz_10ms_modulation_mae_db": 0.4110425373957534, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9919587998397225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10321036688484334, "1000–3000 Hz_40ms_envelope_correlation": 0.9896355435191575, "1000–3000 Hz_40ms_modulation_mae_db": 0.140140421571744, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6377947276251918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1833965586963495, "3000–6000 Hz_10ms_envelope_correlation": 0.9739389152803386, "3000–6000 Hz_10ms_modulation_mae_db": 0.7518364019725335, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0124818868667407, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10822125446998376, "3000–6000 Hz_40ms_envelope_correlation": 0.9903365900097354, "3000–6000 Hz_40ms_modulation_mae_db": 1.348806201365749, "mr_stft_db": 6.1715160833036045, "band_mae_db": 0.32311663389787104, "rms_gain_error_db": -0.07423329746865925, "output_peak": 0.29726299527090894}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.154733020450866, "全帯域_10ms_rms_envelope_nrmse": 0.1379043728684358, "全帯域_10ms_envelope_correlation": 0.9796041069140085, "全帯域_10ms_modulation_mae_db": 0.3490715221981411, "全帯域_40ms_log_rms_mae_db": 0.6258933874198276, "全帯域_40ms_rms_envelope_nrmse": 0.06814492917782478, "全帯域_40ms_envelope_correlation": 0.9943527068392404, "全帯域_40ms_modulation_mae_db": 0.6500676563789329, "300–1000 Hz_10ms_log_rms_mae_db": 1.9072811143962813, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1831100313781005, "300–1000 Hz_10ms_envelope_correlation": 0.9731362181162718, "300–1000 Hz_10ms_modulation_mae_db": 1.3913819050778364, "300–1000 Hz_40ms_log_rms_mae_db": 1.3217212736199992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09679405222940458, "300–1000 Hz_40ms_envelope_correlation": 0.9931586693132919, "300–1000 Hz_40ms_modulation_mae_db": 0.82562725820122, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5621626279842022, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18348740182623857, "1000–3000 Hz_10ms_envelope_correlation": 0.9720754786791298, "1000–3000 Hz_10ms_modulation_mae_db": 0.37867661206087827, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9935787362216569, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09590131922440132, "1000–3000 Hz_40ms_envelope_correlation": 0.9922985146806219, "1000–3000 Hz_40ms_modulation_mae_db": 0.12683658836030953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6367198824131384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17493581412698708, "3000–6000 Hz_10ms_envelope_correlation": 0.9762798819645905, "3000–6000 Hz_10ms_modulation_mae_db": 0.4033353010567206, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9338037703544676, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09164187249782617, "3000–6000 Hz_40ms_envelope_correlation": 0.9928564629277362, "3000–6000 Hz_40ms_modulation_mae_db": 0.8234558165083575, "mr_stft_db": 6.155139202844737, "band_mae_db": 0.32438194713748225, "rms_gain_error_db": -0.006076678255114506, "output_peak": 0.3441506539388677}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 47040, "output_samples": 47040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.1445879572726865, "spectral_convergence": 0.5704463605456489, "spectral_convergence_status": "ok", "amplitude_floor": 8.904239335448439e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.930822423408422, "spectral_convergence": 0.5695028877585314, "spectral_convergence_status": "ok", "amplitude_floor": 4.6801685305432975e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.413532738518523, "spectral_convergence": 0.5858859739252406, "spectral_convergence_status": "ok", "amplitude_floor": 2.078367899935203e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.736571532646871e-05, "output_power": 5.043021191498815e-05, "delta_db": 0.272266926381921, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00015972486768670453, "output_power": 0.00014036327622719534, "delta_db": -0.5611904048853049, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 3.942775929744416e-05, "output_power": 4.079982625094721e-05, "delta_db": 0.14856217053085383, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.533823001233142e-05, "output_power": 6.41865806400518e-05, "delta_db": -0.07723124946509596, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005191422513822901, "output_power": 0.00047456192093739045, "delta_db": -0.38993488907796625, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017487193556096138, "output_power": 0.0001809722134313364, "delta_db": 0.14891781153387645, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.26635057531250306, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.24218281855410084, "pitch": null, "output_rms_dbfs": -30.216760984945964, "output_peak": 0.2770707343303237, "reference_rms_dbfs": -29.974578166391858, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.162981039733211}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 47040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001280716783512989, "log_rms_mae_db": 1.233982016555169, "rms_envelope_nrmse": 0.144818867627008, "envelope_correlation": 0.9777152115715965, "rms_gain_error_db": -0.24212893176721778, "reference_mean_power": 0.0010075555569028208, "output_mean_power": 0.0009529192701076471}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 5.126669358054452e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020945501880404742, "output_power": 0.00019022618747185685, "reference_active": true, "delta_db": -0.41820467230684616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010240530560263733, "output_power": 8.426672174717327e-05, "reference_active": true, "delta_db": -0.8466635918580262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.779418868232472e-05, "output_power": 2.0720057708998965e-05, "reference_active": true, "delta_db": -1.2756304075946867}], "active_band_count": 3, "active_band_mae_db": 0.8468328905865197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 9.465799671406843e-05, "log_rms_mae_db": 0.7609038265219712, "rms_envelope_nrmse": 0.07986724228287136, "envelope_correlation": 0.9930489063835506, "rms_gain_error_db": -0.2422425293540667, "reference_mean_power": 0.001017821671515535, "output_mean_power": 0.0009626035095207955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 4.336511567866126e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018218362838522706, "output_power": 0.0001632309932378753, "reference_active": true, "delta_db": -0.4770672401738029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6513405903536415e-05, "output_power": 4.4466327167442766e-05, "reference_active": true, "delta_db": -1.0412022274309611}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.038210274571303e-06, "output_power": 8.625729593517204e-07, "reference_active": true, "delta_db": -3.7345314712586806}], "active_band_count": 3, "active_band_mae_db": 1.750933646287815}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.812061629421404e-05, "log_rms_mae_db": 2.0201963522381856, "rms_envelope_nrmse": 0.25580340892632103, "envelope_correlation": 0.9492796120486827, "rms_gain_error_db": -0.83109800060941, "reference_mean_power": 0.00014923818668565985, "output_mean_power": 0.00012324524243604668}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 7.990988889403065e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7339937941838396e-05, "output_power": 3.4503816815952656e-05, "reference_active": true, "delta_db": -1.373605448413491}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462900675807451e-06, "output_power": 4.9766562122273175e-06, "reference_active": true, "delta_db": -1.1348984089360257}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2961437099516358e-06, "output_power": 3.7764333479847155e-06, "reference_active": true, "delta_db": 4.644286674261902}], "active_band_count": 3, "active_band_mae_db": 2.3842635105371395}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.488253980944051e-05, "log_rms_mae_db": 1.2984791455613374, "rms_envelope_nrmse": 0.14672707232643858, "envelope_correlation": 0.9885919674771058, "rms_gain_error_db": -0.831104444895432, "reference_mean_power": 0.00015077911609921791, "output_mean_power": 0.00012451760207305457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 7.46020344966024e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.50390789043777e-05, "output_power": 3.4075222081534696e-05, "reference_active": true, "delta_db": -1.211508051812289}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3028613860270352e-06, "output_power": 2.656767647933657e-06, "reference_active": true, "delta_db": -0.9453677355114545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.84375886973483e-08, "output_power": 1.9437721693795477e-07, "reference_active": true, "delta_db": 3.420084669000343}], "active_band_count": 3, "active_band_mae_db": 1.8589868187746952}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.5882621361725546e-05, "log_rms_mae_db": 1.5708856209026119, "rms_envelope_nrmse": 0.19167437459369768, "envelope_correlation": 0.966833859101778, "rms_gain_error_db": -0.17304115835825018, "reference_mean_power": 0.0001648439551682589, "output_mean_power": 0.00015840500894301752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 1.0813916284336453e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.944970033970189e-05, "output_power": 4.886613093890337e-05, "reference_active": true, "delta_db": -0.05155709792768661}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1861966099880283e-05, "output_power": 1.0766458016944262e-05, "reference_active": true, "delta_db": -0.42083827299240784}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.815426369634166e-06, "output_power": 3.5540573187637785e-06, "reference_active": true, "delta_db": -1.3191031896309537}], "active_band_count": 3, "active_band_mae_db": 0.5971661868503494}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.018253092677299e-05, "log_rms_mae_db": 1.0390833597835745, "rms_envelope_nrmse": 0.0982273576042246, "envelope_correlation": 0.990608162448071, "rms_gain_error_db": -0.17308810540565797, "reference_mean_power": 0.00016654362747028494, "output_mean_power": 0.0001600365605899227}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 9.377377349368422e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3529785361500865e-05, "output_power": 4.31488505274933e-05, "reference_active": true, "delta_db": -0.03817294995272}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.594238028592104e-06, "output_power": 4.753589432063343e-06, "reference_active": true, "delta_db": -0.7071927221532819}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1488780558631956e-07, "output_power": 1.8276666867520014e-07, "reference_active": true, "delta_db": -2.3625884644304125}], "active_band_count": 3, "active_band_mae_db": 1.0359847121788048}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001195123420429498, "log_rms_mae_db": 1.6915601213721485, "rms_envelope_nrmse": 0.18957766267687143, "envelope_correlation": 0.9721612366922758, "rms_gain_error_db": -0.269505884392918, "reference_mean_power": 0.0005186016960204273, "output_mean_power": 0.000487397552820946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 3.8692936095348015e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017018269823121574, "output_power": 0.00015578347632159547, "reference_active": true, "delta_db": -0.3839401414476169}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010605303684284744, "output_power": 8.47858569265234e-05, "reference_active": true, "delta_db": -0.9719969519195683}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2588117400162252e-05, "output_power": 1.6742577720381872e-05, "reference_active": true, "delta_db": -1.3005771264339963}], "active_band_count": 3, "active_band_mae_db": 0.8855047399337271}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 8.500195655960054e-05, "log_rms_mae_db": 1.0639225588122896, "rms_envelope_nrmse": 0.11067119627064811, "envelope_correlation": 0.9900180051469993, "rms_gain_error_db": -0.2696347046721871, "reference_mean_power": 0.0005239212873916433, "output_mean_power": 0.0004923824603708358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 3.3612492298744645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016844288389443048, "output_power": 0.00015054894074971098, "reference_active": true, "delta_db": -0.4877496412763787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.87289448022482e-05, "output_power": 4.554819512467102e-05, "reference_active": true, "delta_db": -1.1038102537142827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9018720843055598e-06, "output_power": 8.833409351172724e-07, "reference_active": true, "delta_db": -3.330529472513381}], "active_band_count": 3, "active_band_mae_db": 1.640696455834681}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.783835805956347, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4d1395c9921374212a93d6d97eee5ca52efd4569c4a93161486f00cbf8b05a47", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9673833333582347, "standard_deviation": 0.00495964084537356, "minimum": 0.9612382749511904, "maximum": 0.9732186351249853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5950900980978027, "standard_deviation": 0.041499197215879545, "minimum": 1.5493053888105193, "maximum": 1.6537038768146768, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5670413686376021, "standard_deviation": 0.14750088309461057, "minimum": 0.37867661206087827, "maximum": 0.8277938157591641, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1927970215125183, "standard_deviation": 0.01579826274193185, "minimum": 0.17298854354736568, "maximum": 0.2157345923777302, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99014414769977, "standard_deviation": 0.0014572072722678318, "minimum": 0.987631029828577, "maximum": 0.9922985146806219, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0195990021322787, "standard_deviation": 0.03459056190817374, "minimum": 0.9720837415199848, "maximum": 1.0693065828503503, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6117850756467775, "standard_deviation": 0.33672900626171814, "minimum": 0.12683658836030953, "maximum": 1.0359847121788048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10436192612305417, "standard_deviation": 0.00953837019721947, "minimum": 0.09590131922440132, "maximum": 0.12190300318038853, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.973286696745713, "standard_deviation": 0.003688752832566258, "minimum": 0.9668547124453275, "maximum": 0.9789584015213286, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6823542092670247, "standard_deviation": 0.05167876160775917, "minimum": 1.6367198824131384, "maximum": 1.7822585628481875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.731188745961981, "standard_deviation": 0.39489687105674753, "minimum": 0.19320437081491657, "maximum": 1.3014824917604848, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1860359702255459, "standard_deviation": 0.01296154269001412, "minimum": 0.16473797479898297, "maximum": 0.20638500148640934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905545327450369, "standard_deviation": 0.0030659339687730477, "minimum": 0.9856779225250937, "maximum": 0.9950094243109656, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0282611259964247, "standard_deviation": 0.057253553690082296, "minimum": 0.9338037703544676, "maximum": 1.1200734228896292, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1069105511971884, "standard_deviation": 0.5118745025742232, "minimum": 0.5131715346762897, "maximum": 1.6909859589216785, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10720670867740223, "standard_deviation": 0.01867898899802103, "minimum": 0.07748492752782961, "maximum": 0.13457980308116485, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9609413627924467, "standard_deviation": 0.008346938651919862, "minimum": 0.9492796120486827, "maximum": 0.9731362181162718, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.963980466636892, "standard_deviation": 0.04647025570998691, "minimum": 1.9072811143962813, "maximum": 2.0264957848872225, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.9067008348002241, "standard_deviation": 0.5182650451323846, "minimum": 1.36917622367828, "maximum": 2.6734109709921747, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22236587883369274, "standard_deviation": 0.026906945522078087, "minimum": 0.1831100313781005, "maximum": 0.25580340892632103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9894415971535366, "standard_deviation": 0.003550723609142241, "minimum": 0.9847810217583173, "maximum": 0.9941028374478598, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3066419450786992, "standard_deviation": 0.06876334897109287, "minimum": 1.2060172704345828, "maximum": 1.3998646074443089, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3842401441291323, "standard_deviation": 0.7162934028448846, "minimum": 0.7102016155093409, "maximum": 2.7229813909420257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12172805413531985, "standard_deviation": 0.025706596620352486, "minimum": 0.08965228010836368, "maximum": 0.15279941886404988, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.33326496569237807, "standard_deviation": 0.10296038095122262, "minimum": 0.15832260505506054, "maximum": 0.4871919007252467, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.163038412871468, "standard_deviation": 0.012586902560860338, "minimum": 6.147818804758468, "maximum": 6.180593400871093, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31211907665159666, "standard_deviation": 0.03114130689223221, "minimum": 0.2759594497091529, "maximum": 0.3548741245987411, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1618293759787628, "standard_deviation": 0.1289663560433874, "minimum": -0.3293761784636797, "maximum": 0.016177175493858123, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9781311550157015, "standard_deviation": 0.0021643110786814922, "minimum": 0.9745107561813166, "maximum": 0.981102119531493, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2398097380079633, "standard_deviation": 0.04896199512660155, "minimum": 1.154733020450866, "maximum": 1.3167418711059868, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5899242616182943, "standard_deviation": 0.35162816248581463, "minimum": 0.18634584435778942, "maximum": 1.2018694636631004, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14331672385248984, "standard_deviation": 0.007044824390923307, "minimum": 0.1327891895298487, "maximum": 0.15474773094270083, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9928730549625873, "standard_deviation": 0.0021419298331933627, "minimum": 0.9880675069178543, "maximum": 0.9948192675099861, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6720437948131806, "standard_deviation": 0.05230217070686062, "minimum": 0.5911968053781608, "maximum": 0.7609038265219712, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9730397367465318, "standard_deviation": 0.5168010471878122, "minimum": 0.3268546802987505, "maximum": 1.750933646287815, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0786601361018345, "standard_deviation": 0.011153844640558512, "minimum": 0.06677429061549144, "maximum": 0.10208359486705218, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-75918-0013", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.828000783920288, "predicted_mos": 1.828000783920288, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "fb9fcc5b146766145aaac2853bef71b8d5e425fb50f7c4f2077bbe15926a045b", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.1121796001680195, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2340484612091422, "全帯域_10ms_rms_envelope_nrmse": 0.1388340496436939, "全帯域_10ms_envelope_correlation": 0.9795164631219342, "全帯域_10ms_modulation_mae_db": 0.8128723723563204, "全帯域_40ms_log_rms_mae_db": 0.7694445030611621, "全帯域_40ms_rms_envelope_nrmse": 0.0745883155856091, "全帯域_40ms_envelope_correlation": 0.9939194790949523, "全帯域_40ms_modulation_mae_db": 1.6073370072412898, "300–1000 Hz_10ms_log_rms_mae_db": 1.8220743959512764, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.24913169192888865, "300–1000 Hz_10ms_envelope_correlation": 0.9513689186230114, "300–1000 Hz_10ms_modulation_mae_db": 2.180862676792003, "300–1000 Hz_40ms_log_rms_mae_db": 1.0555034186482695, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1389667528202002, "300–1000 Hz_40ms_envelope_correlation": 0.9890218095315725, "300–1000 Hz_40ms_modulation_mae_db": 1.7646198837018447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4748117764775812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1902025711572014, "1000–3000 Hz_10ms_envelope_correlation": 0.967373409101125, "1000–3000 Hz_10ms_modulation_mae_db": 0.7407627310390218, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8711619293184847, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0976446496400329, "1000–3000 Hz_40ms_envelope_correlation": 0.9907407845321438, "1000–3000 Hz_40ms_modulation_mae_db": 1.1096369508168316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7337553589324102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18579596341024496, "3000–6000 Hz_10ms_envelope_correlation": 0.9731957415432991, "3000–6000 Hz_10ms_modulation_mae_db": 0.8524646281015232, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0174534971902658, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10361093185060793, "3000–6000 Hz_40ms_envelope_correlation": 0.9911252113296143, "3000–6000 Hz_40ms_modulation_mae_db": 1.588465440140893, "mr_stft_db": 6.11138234813746, "band_mae_db": 0.21370670310184126, "rms_gain_error_db": -0.22436286315797893, "output_peak": 0.29441429740026365}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2454794841242272, "全帯域_10ms_rms_envelope_nrmse": 0.15254635381673395, "全帯域_10ms_envelope_correlation": 0.9752735126847832, "全帯域_10ms_modulation_mae_db": 0.7708634973539721, "全帯域_40ms_log_rms_mae_db": 0.7082263686614595, "全帯域_40ms_rms_envelope_nrmse": 0.09971985055311482, "全帯域_40ms_envelope_correlation": 0.988700581889829, "全帯域_40ms_modulation_mae_db": 1.1679161763533892, "300–1000 Hz_10ms_log_rms_mae_db": 1.8233755614631317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19659837678420353, "300–1000 Hz_10ms_envelope_correlation": 0.9689674647145468, "300–1000 Hz_10ms_modulation_mae_db": 1.5108025809245358, "300–1000 Hz_40ms_log_rms_mae_db": 1.0866801296859003, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10541764567935945, "300–1000 Hz_40ms_envelope_correlation": 0.9915400894345265, "300–1000 Hz_40ms_modulation_mae_db": 1.019000830673715, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4736287933190875, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17289896342223235, "1000–3000 Hz_10ms_envelope_correlation": 0.9735330800674429, "1000–3000 Hz_10ms_modulation_mae_db": 0.4132334620873852, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8731995188416304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0996898607500192, "1000–3000 Hz_40ms_envelope_correlation": 0.9911917985305225, "1000–3000 Hz_40ms_modulation_mae_db": 0.533607280832893, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6466331885381185, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19554018125084097, "3000–6000 Hz_10ms_envelope_correlation": 0.9707283524034075, "3000–6000 Hz_10ms_modulation_mae_db": 1.1248829840556596, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0131524874836269, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12955043746977524, "3000–6000 Hz_40ms_envelope_correlation": 0.9866762273520822, "3000–6000 Hz_40ms_modulation_mae_db": 1.565624378201145, "mr_stft_db": 6.146883039582192, "band_mae_db": 0.14095900808548323, "rms_gain_error_db": -0.26986971350086014, "output_peak": 0.34196294094979907}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2858046278931168, "全帯域_10ms_rms_envelope_nrmse": 0.14741384835999682, "全帯域_10ms_envelope_correlation": 0.9769178233189004, "全帯域_10ms_modulation_mae_db": 0.6050404029171087, "全帯域_40ms_log_rms_mae_db": 0.6636343120487814, "全帯域_40ms_rms_envelope_nrmse": 0.07164078474767088, "全帯域_40ms_envelope_correlation": 0.9945588295356167, "全帯域_40ms_modulation_mae_db": 0.4295045843949776, "300–1000 Hz_10ms_log_rms_mae_db": 1.7538962292675726, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19986871678032828, "300–1000 Hz_10ms_envelope_correlation": 0.9677268472152738, "300–1000 Hz_10ms_modulation_mae_db": 1.8058299185996383, "300–1000 Hz_40ms_log_rms_mae_db": 1.0139395207493573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08375840184460312, "300–1000 Hz_40ms_envelope_correlation": 0.994457604042528, "300–1000 Hz_40ms_modulation_mae_db": 0.9794524404454649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5408599842665196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21001645281917902, "1000–3000 Hz_10ms_envelope_correlation": 0.9611268577171639, "1000–3000 Hz_10ms_modulation_mae_db": 0.760364267904099, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8982024894974342, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11807852158281286, "1000–3000 Hz_40ms_envelope_correlation": 0.9887510587095857, "1000–3000 Hz_40ms_modulation_mae_db": 1.051343217045513, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7022904727097241, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18665094122839737, "3000–6000 Hz_10ms_envelope_correlation": 0.9730705752209041, "3000–6000 Hz_10ms_modulation_mae_db": 0.6319101599408221, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9758598898129894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09267827377505798, "3000–6000 Hz_40ms_envelope_correlation": 0.9933597489334756, "3000–6000 Hz_40ms_modulation_mae_db": 0.3538863599016418, "mr_stft_db": 6.106860377699575, "band_mae_db": 0.4111760798740267, "rms_gain_error_db": -0.24838477189456437, "output_peak": 0.3101777571934217}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2617179636876799, "全帯域_10ms_rms_envelope_nrmse": 0.14669403814250526, "全帯域_10ms_envelope_correlation": 0.97704597330841, "全帯域_10ms_modulation_mae_db": 0.3467006369703863, "全帯域_40ms_log_rms_mae_db": 0.6766458713566129, "全帯域_40ms_rms_envelope_nrmse": 0.0756060207467155, "全帯域_40ms_envelope_correlation": 0.99305617060852, "全帯域_40ms_modulation_mae_db": 0.5378712345512471, "300–1000 Hz_10ms_log_rms_mae_db": 1.7501933780888828, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22354650919834557, "300–1000 Hz_10ms_envelope_correlation": 0.9609187369596032, "300–1000 Hz_10ms_modulation_mae_db": 2.3176272272199028, "300–1000 Hz_40ms_log_rms_mae_db": 0.9978016479115557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09845978507642324, "300–1000 Hz_40ms_envelope_correlation": 0.9919452039895353, "300–1000 Hz_40ms_modulation_mae_db": 1.9338652108661234, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6166751547948262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21673883187547707, "1000–3000 Hz_10ms_envelope_correlation": 0.9608973342322987, "1000–3000 Hz_10ms_modulation_mae_db": 0.2921748146533207, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9675614199553508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12115366347180931, "1000–3000 Hz_40ms_envelope_correlation": 0.9872878542505058, "1000–3000 Hz_40ms_modulation_mae_db": 0.2687954912080515, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7382094149868854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.199669729910958, "3000–6000 Hz_10ms_envelope_correlation": 0.9689798054932326, "3000–6000 Hz_10ms_modulation_mae_db": 0.5445979456758271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.038650877450657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11477713311960641, "3000–6000 Hz_40ms_envelope_correlation": 0.9886091609935771, "3000–6000 Hz_40ms_modulation_mae_db": 0.6214177992608705, "mr_stft_db": 6.121115899327471, "band_mae_db": 0.3757538006793224, "rms_gain_error_db": -0.0010409205696458505, "output_peak": 0.3437317217849048}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2504668061426514, "全帯域_10ms_rms_envelope_nrmse": 0.13001472877170261, "全帯域_10ms_envelope_correlation": 0.9818746847439219, "全帯域_10ms_modulation_mae_db": 0.31668772845501075, "全帯域_40ms_log_rms_mae_db": 0.642260254495605, "全帯域_40ms_rms_envelope_nrmse": 0.06601896658429895, "全帯域_40ms_envelope_correlation": 0.9948672839351311, "全帯域_40ms_modulation_mae_db": 0.7542336816388047, "300–1000 Hz_10ms_log_rms_mae_db": 1.7679838245659305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2319978613697616, "300–1000 Hz_10ms_envelope_correlation": 0.9581320721939516, "300–1000 Hz_10ms_modulation_mae_db": 1.312954904413511, "300–1000 Hz_40ms_log_rms_mae_db": 1.0533279036392262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14916596739883464, "300–1000 Hz_40ms_envelope_correlation": 0.9856517023105286, "300–1000 Hz_40ms_modulation_mae_db": 0.8881379156915276, "1000–3000 Hz_10ms_log_rms_mae_db": 1.474164780851031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17441191638034476, "1000–3000 Hz_10ms_envelope_correlation": 0.9727779494608307, "1000–3000 Hz_10ms_modulation_mae_db": 0.8511795569093964, "1000–3000 Hz_40ms_log_rms_mae_db": 0.81590083352617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09783864308225516, "1000–3000 Hz_40ms_envelope_correlation": 0.9910767579607827, "1000–3000 Hz_40ms_modulation_mae_db": 0.8293372231161653, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6982421377969095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16365857809007142, "3000–6000 Hz_10ms_envelope_correlation": 0.97923392446868, "3000–6000 Hz_10ms_modulation_mae_db": 0.20263656676365382, "3000–6000 Hz_40ms_log_rms_mae_db": 0.960767828025685, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0767617400520273, "3000–6000 Hz_40ms_envelope_correlation": 0.9949677119000271, "3000–6000 Hz_40ms_modulation_mae_db": 0.6965157722701106, "mr_stft_db": 6.125494339616601, "band_mae_db": 0.24797605236663436, "rms_gain_error_db": -0.12810628694042422, "output_peak": 0.3334682801373965}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2693309900261065, "全帯域_10ms_rms_envelope_nrmse": 0.1426376520193138, "全帯域_10ms_envelope_correlation": 0.9788447297212558, "全帯域_10ms_modulation_mae_db": 1.2171842872260166, "全帯域_40ms_log_rms_mae_db": 0.711643567173731, "全帯域_40ms_rms_envelope_nrmse": 0.08515117849953009, "全帯域_40ms_envelope_correlation": 0.9926292723016771, "全帯域_40ms_modulation_mae_db": 1.6293946904066476, "300–1000 Hz_10ms_log_rms_mae_db": 1.769006012753664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20923897712638567, "300–1000 Hz_10ms_envelope_correlation": 0.9654500681511917, "300–1000 Hz_10ms_modulation_mae_db": 1.3737271824785775, "300–1000 Hz_40ms_log_rms_mae_db": 1.1398634688357345, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13760225932767872, "300–1000 Hz_40ms_envelope_correlation": 0.9859795968528083, "300–1000 Hz_40ms_modulation_mae_db": 0.7567346678476792, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5577113849425621, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20224145165917837, "1000–3000 Hz_10ms_envelope_correlation": 0.9641930100378683, "1000–3000 Hz_10ms_modulation_mae_db": 0.6612710084936039, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8458433646449988, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09609278244648443, "1000–3000 Hz_40ms_envelope_correlation": 0.9911092934931512, "1000–3000 Hz_40ms_modulation_mae_db": 0.5832292380068598, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8040909767350584, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19560101616186706, "3000–6000 Hz_10ms_envelope_correlation": 0.9713209929225752, "3000–6000 Hz_10ms_modulation_mae_db": 1.4282504045485354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1242829926093025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12472556793274364, "3000–6000 Hz_40ms_envelope_correlation": 0.9886986991269959, "3000–6000 Hz_40ms_modulation_mae_db": 1.8941492020509942, "mr_stft_db": 6.137590723127862, "band_mae_db": 0.4675796098232752, "rms_gain_error_db": -0.3162179735329287, "output_peak": 0.2840064734115351}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3230092486479303, "全帯域_10ms_rms_envelope_nrmse": 0.14197493234514358, "全帯域_10ms_envelope_correlation": 0.978342113435108, "全帯域_10ms_modulation_mae_db": 0.5749743557367551, "全帯域_40ms_log_rms_mae_db": 0.6622465395737425, "全帯域_40ms_rms_envelope_nrmse": 0.08040492798856551, "全帯域_40ms_envelope_correlation": 0.9921846278477929, "全帯域_40ms_modulation_mae_db": 1.2260233766841413, "300–1000 Hz_10ms_log_rms_mae_db": 1.8233618879908244, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.25099421741462885, "300–1000 Hz_10ms_envelope_correlation": 0.9547703378481417, "300–1000 Hz_10ms_modulation_mae_db": 2.6803238396630307, "300–1000 Hz_40ms_log_rms_mae_db": 1.1376747260811675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14402565070056464, "300–1000 Hz_40ms_envelope_correlation": 0.9854012872593972, "300–1000 Hz_40ms_modulation_mae_db": 2.7480920099860904, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5395024105934942, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18800177134072368, "1000–3000 Hz_10ms_envelope_correlation": 0.9683463933470662, "1000–3000 Hz_10ms_modulation_mae_db": 0.1868253479637878, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9247310878844655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10051571739048165, "1000–3000 Hz_40ms_envelope_correlation": 0.9903260877252645, "1000–3000 Hz_40ms_modulation_mae_db": 0.1622965044916752, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6647023594850006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18988389446762655, "3000–6000 Hz_10ms_envelope_correlation": 0.9720241068984531, "3000–6000 Hz_10ms_modulation_mae_db": 0.9510701431712271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0025371198529442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11143349638066219, "3000–6000 Hz_40ms_envelope_correlation": 0.9897244527276313, "3000–6000 Hz_40ms_modulation_mae_db": 1.5751939238119148, "mr_stft_db": 6.131794057601471, "band_mae_db": 0.32012649290848016, "rms_gain_error_db": -0.10486485932920075, "output_peak": 0.2952006614374435}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1599878494221905, "全帯域_10ms_rms_envelope_nrmse": 0.1290625146488234, "全帯域_10ms_envelope_correlation": 0.9821853948433337, "全帯域_10ms_modulation_mae_db": 0.33146088693575065, "全帯域_40ms_log_rms_mae_db": 0.6216649362229496, "全帯域_40ms_rms_envelope_nrmse": 0.06479053239762689, "全帯域_40ms_envelope_correlation": 0.9949019953845558, "全帯域_40ms_modulation_mae_db": 0.7310782072666936, "300–1000 Hz_10ms_log_rms_mae_db": 1.7337863075485491, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18749395329461988, "300–1000 Hz_10ms_envelope_correlation": 0.9714807870738723, "300–1000 Hz_10ms_modulation_mae_db": 1.272302606759457, "300–1000 Hz_40ms_log_rms_mae_db": 1.0875010294157845, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09673921009097376, "300–1000 Hz_40ms_envelope_correlation": 0.9923124978675336, "300–1000 Hz_40ms_modulation_mae_db": 0.7866205690407805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4752289071554325, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17997193760351932, "1000–3000 Hz_10ms_envelope_correlation": 0.9726649458149064, "1000–3000 Hz_10ms_modulation_mae_db": 0.20473289585087281, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8363330976891483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09237098641130136, "1000–3000 Hz_40ms_envelope_correlation": 0.9925253362044314, "1000–3000 Hz_40ms_modulation_mae_db": 0.34559769392204814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6719712282314936, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1693936538568172, "3000–6000 Hz_10ms_envelope_correlation": 0.9778041288510287, "3000–6000 Hz_10ms_modulation_mae_db": 0.3126488341031998, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9194229994052762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08935402993138618, "3000–6000 Hz_40ms_envelope_correlation": 0.9931367883462008, "3000–6000 Hz_40ms_modulation_mae_db": 0.8248911168198837, "mr_stft_db": 6.124692720654752, "band_mae_db": 0.2975995970525468, "rms_gain_error_db": 0.01144974436322539, "output_peak": 0.34925573758983636}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 47040, "output_samples": 47040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.1053592200484905, "spectral_convergence": 0.5679907110345436, "spectral_convergence_status": "ok", "amplitude_floor": 8.904239335448439e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.8866671244336235, "spectral_convergence": 0.5666433667764911, "spectral_convergence_status": "ok", "amplitude_floor": 4.6801685305432975e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.342120699930267, "spectral_convergence": 0.5850563393561309, "spectral_convergence_status": "ok", "amplitude_floor": 2.078367899935203e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.736571532646871e-05, "output_power": 4.810332673028071e-05, "delta_db": 0.06711011861427396, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00015972486768670453, "output_power": 0.0001422021594280482, "delta_db": -0.5046634556727609, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 3.942775929744416e-05, "output_power": 4.08956304202039e-05, "delta_db": 0.15874810811415596, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.533823001233142e-05, "output_power": 6.490544927905128e-05, "delta_db": -0.028862049524660406, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005191422513822901, "output_power": 0.0004776418427704163, "delta_db": -0.36184010994618476, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017487193556096138, "output_power": 0.00018147706827235105, "delta_db": 0.16101637673901134, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21370670310184126, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.22436286315797893, "pitch": null, "output_rms_dbfs": -30.19894102954984, "output_peak": 0.29441429740026365, "reference_rms_dbfs": -29.974578166391858, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.11138234813746}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 47040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001280716783512989, "log_rms_mae_db": 1.2340484612091422, "rms_envelope_nrmse": 0.1388340496436939, "envelope_correlation": 0.9795164631219342, "rms_gain_error_db": -0.22431435419148849, "reference_mean_power": 0.0010075555569028208, "output_mean_power": 0.0009568361329152103}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 5.126669358054452e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020945501880404742, "output_power": 0.00018826373765177916, "reference_active": true, "delta_db": -0.46324094455090425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010240530560263733, "output_power": 8.760853890953248e-05, "reference_active": true, "delta_db": -0.6777602049787619}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.779418868232472e-05, "output_power": 2.061543032515397e-05, "reference_active": true, "delta_db": -1.2976159675392949}], "active_band_count": 3, "active_band_mae_db": 0.8128723723563204}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 9.465799671406843e-05, "log_rms_mae_db": 0.7694445030611621, "rms_envelope_nrmse": 0.0745883155856091, "envelope_correlation": 0.9939194790949523, "rms_gain_error_db": -0.22441950651451803, "reference_mean_power": 0.001017821671515535, "output_mean_power": 0.0009665620578265583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 4.336511567866126e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018218362838522706, "output_power": 0.0001611739739299269, "reference_active": true, "delta_db": -0.5321443319769228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6513405903536415e-05, "output_power": 4.6584409068513356e-05, "reference_active": true, "delta_db": -0.8391089103241081}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.038210274571303e-06, "output_power": 9.208167455756641e-07, "reference_active": true, "delta_db": -3.4507577794228386}], "active_band_count": 3, "active_band_mae_db": 1.6073370072412898}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.812061629421404e-05, "log_rms_mae_db": 1.8220743959512764, "rms_envelope_nrmse": 0.24913169192888865, "envelope_correlation": 0.9513689186230114, "rms_gain_error_db": -0.7478514210022147, "reference_mean_power": 0.00014923818668565985, "output_mean_power": 0.00012563042281620208}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 7.990988889403065e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7339937941838396e-05, "output_power": 3.5239279248712904e-05, "reference_active": true, "delta_db": -1.2820066709726934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462900675807451e-06, "output_power": 5.175433329842643e-06, "reference_active": true, "delta_db": -0.9648076322233474}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2961437099516358e-06, "output_power": 3.4852226623214998e-06, "reference_active": true, "delta_db": 4.295773727179968}], "active_band_count": 3, "active_band_mae_db": 2.180862676792003}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.488253980944051e-05, "log_rms_mae_db": 1.0555034186482695, "rms_envelope_nrmse": 0.1389667528202002, "envelope_correlation": 0.9890218095315725, "rms_gain_error_db": -0.7478534322347975, "reference_mean_power": 0.00015077911609921791, "output_mean_power": 0.0001269275361472184}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 7.46020344966024e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.50390789043777e-05, "output_power": 3.465134363808213e-05, "reference_active": true, "delta_db": -1.1386942071712518}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3028613860270352e-06, "output_power": 2.701547494778728e-06, "reference_active": true, "delta_db": -0.8727774057827499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.84375886973483e-08, "output_power": 1.8831100301417158e-07, "reference_active": true, "delta_db": 3.2823880381515327}], "active_band_count": 3, "active_band_mae_db": 1.7646198837018447}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.5882621361725546e-05, "log_rms_mae_db": 1.4748117764775812, "rms_envelope_nrmse": 0.1902025711572014, "envelope_correlation": 0.967373409101125, "rms_gain_error_db": -0.17731670093729904, "reference_mean_power": 0.0001648439551682589, "output_mean_power": 0.0001582491391081598}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 1.0813916284336453e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.944970033970189e-05, "output_power": 4.7926427812638825e-05, "reference_active": true, "delta_db": -0.13588604019681855}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1861966099880283e-05, "output_power": 1.0774803804861535e-05, "reference_active": true, "delta_db": -0.417473075678581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.815426369634166e-06, "output_power": 3.279003188181792e-06, "reference_active": true, "delta_db": -1.6689290772416658}], "active_band_count": 3, "active_band_mae_db": 0.7407627310390218}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.018253092677299e-05, "log_rms_mae_db": 0.8711619293184847, "rms_envelope_nrmse": 0.0976446496400329, "envelope_correlation": 0.9907407845321438, "rms_gain_error_db": -0.17735127582999183, "reference_mean_power": 0.00016654362747028494, "output_mean_power": 0.0001598795407786039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 9.377377349368422e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3529785361500865e-05, "output_power": 4.210871326722386e-05, "reference_active": true, "delta_db": -0.1441455501213974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.594238028592104e-06, "output_power": 4.763473900889409e-06, "reference_active": true, "delta_db": -0.6981715111745134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1488780558631956e-07, "output_power": 1.7762188033513903e-07, "reference_active": true, "delta_db": -2.4865937911545837}], "active_band_count": 3, "active_band_mae_db": 1.1096369508168316}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001195123420429498, "log_rms_mae_db": 1.7337553589324102, "rms_envelope_nrmse": 0.18579596341024496, "envelope_correlation": 0.9731957415432991, "rms_gain_error_db": -0.22224108471687126, "reference_mean_power": 0.0005186016960204273, "output_mean_power": 0.0004927309292573797}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 3.8692936095348015e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017018269823121574, "output_power": 0.00015728055214225708, "reference_active": true, "delta_db": -0.3424037996886483}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010605303684284744, "output_power": 8.640907920833503e-05, "reference_active": true, "delta_db": -0.8896373187611148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2588117400162252e-05, "output_power": 1.6647336557146093e-05, "reference_active": true, "delta_db": -1.3253527658548068}], "active_band_count": 3, "active_band_mae_db": 0.8524646281015232}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 8.500195655960054e-05, "log_rms_mae_db": 1.0174534971902658, "rms_envelope_nrmse": 0.10361093185060793, "envelope_correlation": 0.9911252113296143, "rms_gain_error_db": -0.22231115823401218, "reference_mean_power": 0.0005239212873916433, "output_mean_power": 0.0004977771178084143}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 3.3612492298744645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016844288389443048, "output_power": 0.00015185378852976245, "reference_active": true, "delta_db": -0.45027036924863606}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.87289448022482e-05, "output_power": 4.6436585088837826e-05, "reference_active": true, "delta_db": -1.0199192327984572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9018720843055598e-06, "output_power": 8.905547656980952e-07, "reference_active": true, "delta_db": -3.295206718375586}], "active_band_count": 3, "active_band_mae_db": 1.588465440140893}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.783301175270525, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "907b26ab048207311bed30491e552c60eb30577a83c31fde44f3ea79ccfb9e86", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9676141224723378, "standard_deviation": 0.005164549582533969, "minimum": 0.9608973342322987, "maximum": 0.9735330800674429, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5190728990500668, "standard_deviation": 0.053284175304496674, "minimum": 1.4736287933190875, "maximum": 1.6166751547948262, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.513818010612686, "standard_deviation": 0.26980017206608625, "minimum": 0.1868253479637878, "maximum": 0.8511795569093964, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19181048703223197, "standard_deviation": 0.016385025991278536, "minimum": 0.17289896342223235, "maximum": 0.21673883187547707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9903761214257985, "standard_deviation": 0.0016312393039955725, "minimum": 0.9872878542505058, "maximum": 0.9925253362044314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8791167176697103, "standard_deviation": 0.0497127418082065, "minimum": 0.81590083352617, "maximum": 0.9675614199553508, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6104804499300046, "standard_deviation": 0.35573825169838336, "minimum": 0.1622965044916752, "maximum": 1.1096369508168316, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10292310309689962, "standard_deviation": 0.010623448823578302, "minimum": 0.09237098641130136, "maximum": 0.12115366347180931, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9732947034751975, "standard_deviation": 0.0035129096932957943, "minimum": 0.9689798054932326, "maximum": 0.97923392446868, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.70748689217695, "standard_deviation": 0.05053114320516229, "minimum": 1.6466331885381185, "maximum": 1.8040909767350584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7560577082950559, "standard_deviation": 0.41373101742465274, "minimum": 0.20263656676365382, "maximum": 1.4282504045485354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18577424479710294, "standard_deviation": 0.012873606548826246, "minimum": 0.16365857809007142, "maximum": 0.199669729910958, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9907872500887005, "standard_deviation": 0.0028501810088826587, "minimum": 0.9866762273520822, "maximum": 0.9949677119000271, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0065159614788435, "standard_deviation": 0.060591255001397504, "minimum": 0.9194229994052762, "maximum": 1.1242829926093025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1400179990571817, "standard_deviation": 0.5761059119618094, "minimum": 0.3538863599016418, "maximum": 1.8941492020509942, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10536145131398336, "standard_deviation": 0.01821972711607395, "minimum": 0.0767617400520273, "maximum": 0.12955043746977524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.962351904097449, "standard_deviation": 0.0072067215328330825, "minimum": 0.9513689186230114, "maximum": 0.9714807870738723, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.780459699703729, "standard_deviation": 0.03684147511458536, "minimum": 1.7337863075485491, "maximum": 1.8233755614631317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.806803867106332, "standard_deviation": 0.5303165039896927, "minimum": 1.272302606759457, "maximum": 2.6803238396630307, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21860878798714525, "standard_deviation": 0.024129670072674254, "minimum": 0.18749395329461988, "maximum": 0.25099421741462885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9895387239110538, "standard_deviation": 0.003522331186221539, "minimum": 0.9854012872593972, "maximum": 0.994457604042528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0715364806208745, "standard_deviation": 0.051920115079852246, "minimum": 0.9978016479115557, "maximum": 1.1398634688357345, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.359565441031653, "standard_deviation": 0.7167665674980539, "minimum": 0.7567346678476792, "maximum": 2.7480920099860904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11926695911732972, "standard_deviation": 0.025702439297598724, "minimum": 0.08375840184460312, "maximum": 0.14916596739883464, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.30935966798645126, "standard_deviation": 0.10779527989976344, "minimum": 0.14095900808548323, "maximum": 0.4675796098232752, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.125726688218423, "standard_deviation": 0.013143383520313339, "minimum": 6.106860377699575, "maximum": 6.146883039582192, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.31902723373807507, "standard_deviation": 0.026013550680121882, "minimum": 0.2840064734115351, "maximum": 0.34925573758983636, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1601747055702972, "standard_deviation": 0.12372186983312253, "minimum": -0.3162179735329287, "maximum": 0.01144974436322539, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9787500868972059, "standard_deviation": 0.002410540713336379, "minimum": 0.9752735126847832, "maximum": 0.9821853948433337, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2537306788941307, "standard_deviation": 0.04688781094756569, "minimum": 1.1599878494221905, "maximum": 1.3230092486479303, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.621973020993915, "standard_deviation": 0.30908950879349156, "minimum": 0.31668772845501075, "maximum": 1.2171842872260166, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14114726471848915, "standard_deviation": 0.008265312984004023, "minimum": 0.1290625146488234, "maximum": 0.15254635381673395, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9931022800747593, "standard_deviation": 0.0020558786564775565, "minimum": 0.988700581889829, "maximum": 0.9949019953845558, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6819707940742554, "standard_deviation": 0.04658146390951341, "minimum": 0.6216649362229496, "maximum": 0.7694445030611621, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.010419869817149, "standard_deviation": 0.46524306476736693, "minimum": 0.4295045843949776, "maximum": 1.6293946904066476, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07724007213789147, "standard_deviation": 0.011338307757631698, "minimum": 0.06479053239762689, "maximum": 0.09971985055311482, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-75918-0013", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.6714372634887695, "predicted_mos": 1.6714372634887695, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "0b3b52b392c4d7f69338c53dc93052ce359bbbbf52ac6f6adea8595b2d33e7a1", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.736786600202322, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.139648441222821, "全帯域_10ms_rms_envelope_nrmse": 0.14360218019479112, "全帯域_10ms_envelope_correlation": 0.9784859372236733, "全帯域_10ms_modulation_mae_db": 0.27881771916860876, "全帯域_40ms_log_rms_mae_db": 0.5896840767358067, "全帯域_40ms_rms_envelope_nrmse": 0.07439443070911243, "全帯域_40ms_envelope_correlation": 0.9935613021938255, "全帯域_40ms_modulation_mae_db": 0.2916796524605685, "300–1000 Hz_10ms_log_rms_mae_db": 2.4422966603161185, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22161327579941584, "300–1000 Hz_10ms_envelope_correlation": 0.9601376523061914, "300–1000 Hz_10ms_modulation_mae_db": 1.1802253187643772, "300–1000 Hz_40ms_log_rms_mae_db": 1.7817369333988347, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14041567650076908, "300–1000 Hz_40ms_envelope_correlation": 0.984211481500445, "300–1000 Hz_40ms_modulation_mae_db": 0.8167596479499064, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8793791325045077, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1950693460593805, "1000–3000 Hz_10ms_envelope_correlation": 0.9686749269359588, "1000–3000 Hz_10ms_modulation_mae_db": 0.691012560253926, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2295954297056242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10424922867164012, "1000–3000 Hz_40ms_envelope_correlation": 0.991088781344866, "1000–3000 Hz_40ms_modulation_mae_db": 0.7226158976655079, "3000–6000 Hz_10ms_log_rms_mae_db": 2.032605288148469, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18300047201913044, "3000–6000 Hz_10ms_envelope_correlation": 0.9740874033499847, "3000–6000 Hz_10ms_modulation_mae_db": 0.554423931101759, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2628705816538042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10150723747801978, "3000–6000 Hz_40ms_envelope_correlation": 0.9911225575631302, "3000–6000 Hz_40ms_modulation_mae_db": 0.6816780671940377, "mr_stft_db": 6.454080744303464, "band_mae_db": 0.4897346026077825, "rms_gain_error_db": 0.09020248387313051, "output_peak": 0.3444424377554597}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.178689241018744, "全帯域_10ms_rms_envelope_nrmse": 0.13800800250778053, "全帯域_10ms_envelope_correlation": 0.9795511939643577, "全帯域_10ms_modulation_mae_db": 0.7776427109428994, "全帯域_40ms_log_rms_mae_db": 0.6329628791278428, "全帯域_40ms_rms_envelope_nrmse": 0.07457537600991916, "全帯域_40ms_envelope_correlation": 0.9932032698161289, "全帯域_40ms_modulation_mae_db": 1.417183519328787, "300–1000 Hz_10ms_log_rms_mae_db": 2.4101446993301705, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21720904732977592, "300–1000 Hz_10ms_envelope_correlation": 0.961587002204437, "300–1000 Hz_10ms_modulation_mae_db": 1.8626407475428952, "300–1000 Hz_40ms_log_rms_mae_db": 1.6727908762114052, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10829214063325611, "300–1000 Hz_40ms_envelope_correlation": 0.9907100741640473, "300–1000 Hz_40ms_modulation_mae_db": 1.3094318066237578, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9954767337404586, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2030038397457975, "1000–3000 Hz_10ms_envelope_correlation": 0.9643810584137112, "1000–3000 Hz_10ms_modulation_mae_db": 1.0800301887899375, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2440697100836071, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09857258839271657, "1000–3000 Hz_40ms_envelope_correlation": 0.990971826353901, "1000–3000 Hz_40ms_modulation_mae_db": 1.305499086199493, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0395596710543624, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18187000080092974, "3000–6000 Hz_10ms_envelope_correlation": 0.9742881813041825, "3000–6000 Hz_10ms_modulation_mae_db": 0.953188413691025, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2367947138523132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1028801377618169, "3000–6000 Hz_40ms_envelope_correlation": 0.9909531976915815, "3000–6000 Hz_40ms_modulation_mae_db": 1.5405047375600098, "mr_stft_db": 6.465961607980923, "band_mae_db": 0.40394525561332956, "rms_gain_error_db": -0.04083978616749539, "output_peak": 0.3682626378437223}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1309426246031398, "全帯域_10ms_rms_envelope_nrmse": 0.1248784291536584, "全帯域_10ms_envelope_correlation": 0.9834084970789186, "全帯域_10ms_modulation_mae_db": 0.10068365128014582, "全帯域_40ms_log_rms_mae_db": 0.5616062919659053, "全帯域_40ms_rms_envelope_nrmse": 0.058177720987306496, "全帯域_40ms_envelope_correlation": 0.9958908002750739, "全帯域_40ms_modulation_mae_db": 0.3092799198931913, "300–1000 Hz_10ms_log_rms_mae_db": 2.2677443561847803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19505577374934024, "300–1000 Hz_10ms_envelope_correlation": 0.9703878229940236, "300–1000 Hz_10ms_modulation_mae_db": 0.7900052345533367, "300–1000 Hz_40ms_log_rms_mae_db": 1.626160138381764, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12566994474229193, "300–1000 Hz_40ms_envelope_correlation": 0.9889195000657012, "300–1000 Hz_40ms_modulation_mae_db": 0.6547047031721628, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9001647166458202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2182436322465739, "1000–3000 Hz_10ms_envelope_correlation": 0.9583040202743087, "1000–3000 Hz_10ms_modulation_mae_db": 0.956936514927392, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2407548819777974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1162688419326329, "1000–3000 Hz_40ms_envelope_correlation": 0.9868969470701972, "1000–3000 Hz_40ms_modulation_mae_db": 0.7241905379924968, "3000–6000 Hz_10ms_log_rms_mae_db": 2.042294285379419, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17974197796432212, "3000–6000 Hz_10ms_envelope_correlation": 0.9748898218814028, "3000–6000 Hz_10ms_modulation_mae_db": 0.44057009806226244, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2266678637552912, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08349591879299882, "3000–6000 Hz_40ms_envelope_correlation": 0.9941365807763355, "3000–6000 Hz_40ms_modulation_mae_db": 0.7141264722554261, "mr_stft_db": 6.453147060299469, "band_mae_db": 0.4634369337028455, "rms_gain_error_db": -0.005519677046541025, "output_peak": 0.3653706959292403}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1538995245821428, "全帯域_10ms_rms_envelope_nrmse": 0.11942525516180647, "全帯域_10ms_envelope_correlation": 0.9847453317083523, "全帯域_10ms_modulation_mae_db": 0.4993092053299793, "全帯域_40ms_log_rms_mae_db": 0.6226812908554742, "全帯域_40ms_rms_envelope_nrmse": 0.05703311323300576, "全帯域_40ms_envelope_correlation": 0.9960293372567054, "全帯域_40ms_modulation_mae_db": 1.0457951167761739, "300–1000 Hz_10ms_log_rms_mae_db": 2.2774543648881402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20866525609043118, "300–1000 Hz_10ms_envelope_correlation": 0.9646083261569237, "300–1000 Hz_10ms_modulation_mae_db": 1.7079501004661726, "300–1000 Hz_40ms_log_rms_mae_db": 1.6227602506630008, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12926422400435375, "300–1000 Hz_40ms_envelope_correlation": 0.9862976427381638, "300–1000 Hz_40ms_modulation_mae_db": 1.2629594318619428, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9302421737034676, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1870719306115253, "1000–3000 Hz_10ms_envelope_correlation": 0.9698140273305004, "1000–3000 Hz_10ms_modulation_mae_db": 0.22139034989623343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.281574102157923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10884555455783751, "1000–3000 Hz_40ms_envelope_correlation": 0.988914698970787, "1000–3000 Hz_40ms_modulation_mae_db": 0.2969829906303119, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9707486237886345, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.161783227608435, "3000–6000 Hz_10ms_envelope_correlation": 0.9797738514862555, "3000–6000 Hz_10ms_modulation_mae_db": 0.7537702230217649, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1653821362009325, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08747271754663624, "3000–6000 Hz_40ms_envelope_correlation": 0.9937093231835968, "3000–6000 Hz_40ms_modulation_mae_db": 1.464532313563793, "mr_stft_db": 6.47026152370817, "band_mae_db": 0.3608689068886644, "rms_gain_error_db": -0.0032627101140620864, "output_peak": 0.402970528424159}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2330255613862247, "全帯域_10ms_rms_envelope_nrmse": 0.13168550288975672, "全帯域_10ms_envelope_correlation": 0.9814373961863955, "全帯域_10ms_modulation_mae_db": 0.4834991303344185, "全帯域_40ms_log_rms_mae_db": 0.6626850369534187, "全帯域_40ms_rms_envelope_nrmse": 0.061854842961606364, "全帯域_40ms_envelope_correlation": 0.9953759540800555, "全帯域_40ms_modulation_mae_db": 1.0334260363726522, "300–1000 Hz_10ms_log_rms_mae_db": 2.1939119022297264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.25148023126282415, "300–1000 Hz_10ms_envelope_correlation": 0.9483616229551474, "300–1000 Hz_10ms_modulation_mae_db": 2.567953059395748, "300–1000 Hz_40ms_log_rms_mae_db": 1.4652294224714235, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1293950485190833, "300–1000 Hz_40ms_envelope_correlation": 0.9870511229094153, "300–1000 Hz_40ms_modulation_mae_db": 1.9302749017141971, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9987354717695571, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22294979103539864, "1000–3000 Hz_10ms_envelope_correlation": 0.9568085848084961, "1000–3000 Hz_10ms_modulation_mae_db": 0.372253071560926, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2995423864947682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11369836576786042, "1000–3000 Hz_40ms_envelope_correlation": 0.9876962026025624, "1000–3000 Hz_40ms_modulation_mae_db": 0.22612539940622334, "3000–6000 Hz_10ms_log_rms_mae_db": 2.06710110891209, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17314371472138776, "3000–6000 Hz_10ms_envelope_correlation": 0.976704538913589, "3000–6000 Hz_10ms_modulation_mae_db": 0.7248234236181483, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2345448953232092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09128593262013454, "3000–6000 Hz_40ms_envelope_correlation": 0.9928896725341309, "3000–6000 Hz_40ms_modulation_mae_db": 1.2906543115886524, "mr_stft_db": 6.478055892822706, "band_mae_db": 0.4311458048770102, "rms_gain_error_db": -0.05154144196756256, "output_peak": 0.34862337140029404}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2600780287128062, "全帯域_10ms_rms_envelope_nrmse": 0.14720520760197683, "全帯域_10ms_envelope_correlation": 0.9766579747787227, "全帯域_10ms_modulation_mae_db": 0.5369820885432075, "全帯域_40ms_log_rms_mae_db": 0.6922400904197702, "全帯域_40ms_rms_envelope_nrmse": 0.08006570566081078, "全帯域_40ms_envelope_correlation": 0.9922609271867261, "全帯域_40ms_modulation_mae_db": 0.7841257869181829, "300–1000 Hz_10ms_log_rms_mae_db": 2.3881505026597436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22619982148248444, "300–1000 Hz_10ms_envelope_correlation": 0.9589860539854215, "300–1000 Hz_10ms_modulation_mae_db": 1.5559270214425684, "300–1000 Hz_40ms_log_rms_mae_db": 1.6246577413291623, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1278782600788414, "300–1000 Hz_40ms_envelope_correlation": 0.9887896234686216, "300–1000 Hz_40ms_modulation_mae_db": 1.1312469455245837, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0428006768098945, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2206239046832088, "1000–3000 Hz_10ms_envelope_correlation": 0.959929433496659, "1000–3000 Hz_10ms_modulation_mae_db": 0.45388143117267793, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3036892743033663, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11866375400676711, "1000–3000 Hz_40ms_envelope_correlation": 0.9882631403724987, "1000–3000 Hz_40ms_modulation_mae_db": 0.4503965215807793, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0751598404678244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19023726450443312, "3000–6000 Hz_10ms_envelope_correlation": 0.9718751528613337, "3000–6000 Hz_10ms_modulation_mae_db": 0.47763428587601314, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2808162527067926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10123854652202423, "3000–6000 Hz_40ms_envelope_correlation": 0.9915794455089728, "3000–6000 Hz_40ms_modulation_mae_db": 0.8111924696546279, "mr_stft_db": 6.467139375558602, "band_mae_db": 0.4569862265574853, "rms_gain_error_db": -0.1071329106965467, "output_peak": 0.3100354053498706}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2177274672980116, "全帯域_10ms_rms_envelope_nrmse": 0.12991187363983384, "全帯域_10ms_envelope_correlation": 0.9819212075402028, "全帯域_10ms_modulation_mae_db": 0.2226017065026784, "全帯域_40ms_log_rms_mae_db": 0.7316005473847177, "全帯域_40ms_rms_envelope_nrmse": 0.07680256849787952, "全帯域_40ms_envelope_correlation": 0.9927857313699681, "全帯域_40ms_modulation_mae_db": 0.5115610075279492, "300–1000 Hz_10ms_log_rms_mae_db": 2.381307036000109, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23120461046937743, "300–1000 Hz_10ms_envelope_correlation": 0.9582025195142578, "300–1000 Hz_10ms_modulation_mae_db": 2.099638496855858, "300–1000 Hz_40ms_log_rms_mae_db": 1.6967657151410558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13886415480369993, "300–1000 Hz_40ms_envelope_correlation": 0.988168774621093, "300–1000 Hz_40ms_modulation_mae_db": 1.5569393876696875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9300278964376971, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21734870594110392, "1000–3000 Hz_10ms_envelope_correlation": 0.9584521808625256, "1000–3000 Hz_10ms_modulation_mae_db": 0.14726635221604994, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3628533163943262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11844457492218849, "1000–3000 Hz_40ms_envelope_correlation": 0.9863509819312526, "1000–3000 Hz_40ms_modulation_mae_db": 0.12393704521347954, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0195275271768525, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17371199252419847, "3000–6000 Hz_10ms_envelope_correlation": 0.9765822885441053, "3000–6000 Hz_10ms_modulation_mae_db": 0.3086552589464811, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2978593194627044, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11126647862121403, "3000–6000 Hz_40ms_envelope_correlation": 0.9892687026866878, "3000–6000 Hz_40ms_modulation_mae_db": 0.615353440784535, "mr_stft_db": 6.444550519428373, "band_mae_db": 0.4001622680526549, "rms_gain_error_db": -0.017488991298647047, "output_peak": 0.4111057883299917}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.167521530411872, "全帯域_10ms_rms_envelope_nrmse": 0.1317461191554822, "全帯域_10ms_envelope_correlation": 0.9816163146639262, "全帯域_10ms_modulation_mae_db": 0.33511401390696266, "全帯域_40ms_log_rms_mae_db": 0.6032305652008797, "全帯域_40ms_rms_envelope_nrmse": 0.06220246948175597, "全帯域_40ms_envelope_correlation": 0.9953892964300729, "全帯域_40ms_modulation_mae_db": 0.09664900801554366, "300–1000 Hz_10ms_log_rms_mae_db": 2.364490400534911, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2520788377161614, "300–1000 Hz_10ms_envelope_correlation": 0.9479967184045259, "300–1000 Hz_10ms_modulation_mae_db": 2.103587418168224, "300–1000 Hz_40ms_log_rms_mae_db": 1.6327416333592937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13719968467273155, "300–1000 Hz_40ms_envelope_correlation": 0.9840861721542461, "300–1000 Hz_40ms_modulation_mae_db": 1.197862337230907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9457121531730777, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19927994100101268, "1000–3000 Hz_10ms_envelope_correlation": 0.9643495400583734, "1000–3000 Hz_10ms_modulation_mae_db": 0.43858907976636735, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2517350671038592, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09505988481886221, "1000–3000 Hz_40ms_envelope_correlation": 0.99099646233479, "1000–3000 Hz_40ms_modulation_mae_db": 0.5962430030804143, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0103336094023017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15646604390850577, "3000–6000 Hz_10ms_envelope_correlation": 0.9811475185433681, "3000–6000 Hz_10ms_modulation_mae_db": 0.33581844678930817, "3000–6000 Hz_40ms_log_rms_mae_db": 1.137362424742364, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07435757825440335, "3000–6000 Hz_40ms_envelope_correlation": 0.9952308908287838, "3000–6000 Hz_40ms_modulation_mae_db": 0.07503608279949199, "mr_stft_db": 6.4574236452871085, "band_mae_db": 0.37689806780956464, "rms_gain_error_db": 0.06476074978359236, "output_peak": 0.4320790802148561}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 47040, "output_samples": 47040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.470259224414901, "spectral_convergence": 0.586613394728917, "spectral_convergence_status": "ok", "amplitude_floor": 8.904239335448439e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.300083952279028, "spectral_convergence": 0.585503237342263, "spectral_convergence_status": "ok", "amplitude_floor": 4.6801685305432975e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.591899056216461, "spectral_convergence": 0.6051483422544173, "spectral_convergence_status": "ok", "amplitude_floor": 2.078367899935203e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.736571532646871e-05, "output_power": 5.5956780336799786e-05, "delta_db": 0.7238861743970735, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00015972486768670453, "output_power": 0.00014354211607413907, "delta_db": -0.4639319280142411, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 3.942775929744416e-05, "output_power": 5.062061225367252e-05, "delta_db": 1.0852529705734777, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.533823001233142e-05, "output_power": 7.406036284353495e-05, "delta_db": 0.544184700279324, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005191422513822901, "output_power": 0.0005293848297041408, "delta_db": 0.08485115598639559, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017487193556096138, "output_power": 0.00017341635290300757, "delta_db": -0.03630068639618303, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4897346026077825, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.09020248387313051, "pitch": null, "output_rms_dbfs": -29.88437568251873, "output_peak": 0.3444424377554597, "reference_rms_dbfs": -29.974578166391858, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.454080744303464}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 47040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001280716783512989, "log_rms_mae_db": 1.139648441222821, "rms_envelope_nrmse": 0.14360218019479112, "envelope_correlation": 0.9784859372236733, "rms_gain_error_db": 0.09027469862648752, "reference_mean_power": 0.0010075555569028208, "output_mean_power": 0.0010287183161918308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 5.126669358054452e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020945501880404742, "output_power": 0.00018982806591147524, "reference_active": true, "delta_db": -0.42730348167446786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010240530560263733, "output_power": 0.00010230100022507511, "reference_active": true, "delta_db": -0.0044257800115399815}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.779418868232472e-05, "output_power": 2.532104422935139e-05, "reference_active": true, "delta_db": -0.4047238958198185}], "active_band_count": 3, "active_band_mae_db": 0.27881771916860876}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 9.465799671406843e-05, "log_rms_mae_db": 0.5896840767358067, "rms_envelope_nrmse": 0.07439443070911243, "envelope_correlation": 0.9935613021938255, "rms_gain_error_db": 0.0904853418091371, "reference_mean_power": 0.001017821671515535, "output_mean_power": 0.0010392504658154405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 4.336511567866126e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018218362838522706, "output_power": 0.00016355285824130118, "reference_active": true, "delta_db": -0.4685120903721529}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6513405903536415e-05, "output_power": 5.675563006766863e-05, "reference_active": true, "delta_db": 0.018574675298256346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.038210274571303e-06, "output_power": 1.8640337327329232e-06, "reference_active": true, "delta_db": -0.38795219171129625}], "active_band_count": 3, "active_band_mae_db": 0.2916796524605685}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.812061629421404e-05, "log_rms_mae_db": 2.4422966603161185, "rms_envelope_nrmse": 0.22161327579941584, "envelope_correlation": 0.9601376523061914, "rms_gain_error_db": -0.3609900009291123, "reference_mean_power": 0.00014923818668565985, "output_mean_power": 0.0001373349151186459}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 7.990988889403065e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7339937941838396e-05, "output_power": 4.064618244666673e-05, "reference_active": true, "delta_db": -0.6620792204940705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462900675807451e-06, "output_power": 7.068110085469557e-06, "reference_active": true, "delta_db": 0.38875823423301714}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2961437099516358e-06, "output_power": 2.299519023520075e-06, "reference_active": true, "delta_db": 2.489838501566044}], "active_band_count": 3, "active_band_mae_db": 1.1802253187643772}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.488253980944051e-05, "log_rms_mae_db": 1.7817369333988347, "rms_envelope_nrmse": 0.14041567650076908, "envelope_correlation": 0.984211481500445, "rms_gain_error_db": -0.36099364302403214, "reference_mean_power": 0.00015077911609921791, "output_mean_power": 0.00013875282329180882}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 7.46020344966024e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.50390789043777e-05, "output_power": 3.9798969161578844e-05, "reference_active": true, "delta_db": -0.5371767666404687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3028613860270352e-06, "output_power": 3.4453881792251334e-06, "reference_active": true, "delta_db": 0.18347811693003227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.84375886973483e-08, "output_power": 1.31704101174765e-07, "reference_active": true, "delta_db": 1.7296240602792181}], "active_band_count": 3, "active_band_mae_db": 0.8167596479499064}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.5882621361725546e-05, "log_rms_mae_db": 1.8793791325045077, "rms_envelope_nrmse": 0.1950693460593805, "envelope_correlation": 0.9686749269359588, "rms_gain_error_db": 0.32513461512598585, "reference_mean_power": 0.0001648439551682589, "output_mean_power": 0.0001776587031978361}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 1.0813916284336453e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.944970033970189e-05, "output_power": 5.4820936019863133e-05, "reference_active": true, "delta_db": 0.4478278230829619}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1861966099880283e-05, "output_power": 1.361145445198655e-05, "reference_active": true, "delta_db": 0.5974785568446994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.815426369634166e-06, "output_power": 3.8006827221818543e-06, "reference_active": true, "delta_db": -1.0277313008341167}], "active_band_count": 3, "active_band_mae_db": 0.691012560253926}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.018253092677299e-05, "log_rms_mae_db": 1.2295954297056242, "rms_envelope_nrmse": 0.10424922867164012, "envelope_correlation": 0.991088781344866, "rms_gain_error_db": 0.32513320854403976, "reference_mean_power": 0.00016654362747028494, "output_mean_power": 0.0001794904476125497}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 9.377377349368422e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3529785361500865e-05, "output_power": 4.903929927911804e-05, "reference_active": true, "delta_db": 0.5175773021043747}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.594238028592104e-06, "output_power": 6.409195638442935e-06, "reference_active": true, "delta_db": 0.5906258729799744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1488780558631956e-07, "output_power": 2.4671263432937876e-07, "reference_active": true, "delta_db": -1.0596445179121747}], "active_band_count": 3, "active_band_mae_db": 0.7226158976655079}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001195123420429498, "log_rms_mae_db": 2.032605288148469, "rms_envelope_nrmse": 0.18300047201913044, "envelope_correlation": 0.9740874033499847, "rms_gain_error_db": -0.05119695089324836, "reference_mean_power": 0.0005186016960204273, "output_mean_power": 0.0005125240362375408}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 3.8692936095348015e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017018269823121574, "output_power": 0.00015016437062524013, "reference_active": true, "delta_db": -0.5434850487289659}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010605303684284744, "output_power": 8.83305472938897e-05, "reference_active": true, "delta_db": -0.7941218778789434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2588117400162252e-05, "output_power": 2.0956248979114487e-05, "reference_active": true, "delta_db": -0.32566486669736777}], "active_band_count": 3, "active_band_mae_db": 0.554423931101759}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 8.500195655960054e-05, "log_rms_mae_db": 1.2628705816538042, "rms_envelope_nrmse": 0.10150723747801978, "envelope_correlation": 0.9911225575631302, "rms_gain_error_db": -0.051243741948076155, "reference_mean_power": 0.0005239212873916433, "output_mean_power": 0.0005177757070440476}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 3.3612492298744645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016844288389443048, "output_power": 0.00014622535331617237, "reference_active": true, "delta_db": -0.6142998892779357}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.87289448022482e-05, "output_power": 4.9689742127126765e-05, "reference_active": true, "delta_db": -0.7258545492402463}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9018720843055598e-06, "output_power": 1.6169378510416521e-06, "reference_active": true, "delta_db": -0.7048797630639314}], "active_band_count": 3, "active_band_mae_db": 0.6816780671940377}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7664045570228579, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2bf1e59cfa4cb4530dff751bab9357be4c02dab0aabcb7fb91127aed4db343f2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9625892215225667, "standard_deviation": 0.0049499107994364225, "minimum": 0.9568085848084961, "maximum": 0.9698140273305004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9528173693480602, "standard_deviation": 0.055123561491065516, "minimum": 1.8793791325045077, "maximum": 2.0428006768098945, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5451699435729387, "standard_deviation": 0.33586797384591804, "minimum": 0.14726635221604994, "maximum": 1.0800301887899375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20794888641550016, "standard_deviation": 0.013529626596754812, "minimum": 0.1870719306115253, "maximum": 0.22294979103539864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9888973801226069, "standard_deviation": 0.0019214197206692983, "minimum": 0.9863509819312526, "maximum": 0.991088781344866, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.276726771027659, "standard_deviation": 0.0445536856885947, "minimum": 1.2295954297056242, "maximum": 1.3628533163943262, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5557488102210882, "standard_deviation": 0.37666854394400395, "minimum": 0.12393704521347954, "maximum": 1.305499086199493, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10922534913381317, "standard_deviation": 0.009124096736018232, "minimum": 0.09505988481886221, "maximum": 0.11866375400676711, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9761685946105277, "standard_deviation": 0.00307483222882096, "minimum": 0.9718751528613337, "maximum": 0.9811475185433681, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0321662442912443, "standard_deviation": 0.033025643798331, "minimum": 1.9707486237886345, "maximum": 2.0751598404678244, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5686105101383453, "standard_deviation": 0.2246355506862568, "minimum": 0.3086552589464811, "maximum": 0.953188413691025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17499433675641782, "standard_deviation": 0.011270442700725771, "minimum": 0.15646604390850577, "maximum": 0.19023726450443312, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9923612963466524, "standard_deviation": 0.0019703229220837007, "minimum": 0.9892687026866878, "maximum": 0.9952308908287838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2302872734621766, "standard_deviation": 0.054905255694801716, "minimum": 1.137362424742364, "maximum": 1.2978593194627044, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8991347369250717, "standard_deviation": 0.4975995512857125, "minimum": 0.07503608279949199, "maximum": 1.5405047375600098, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09418806844965599, "standard_deviation": 0.012139618580841215, "minimum": 0.07435757825440335, "maximum": 0.11126647862121403, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.958783464815116, "standard_deviation": 0.007593521766658907, "minimum": 0.9479967184045259, "maximum": 0.9703878229940236, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3406874902679626, "standard_deviation": 0.084920292969095, "minimum": 2.1939119022297264, "maximum": 2.4422966603161185, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.7334909246486476, "standard_deviation": 0.5622012275384062, "minimum": 0.7900052345533367, "maximum": 2.567953059395748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2254383567374763, "standard_deviation": 0.01967055449173226, "minimum": 0.19505577374934024, "maximum": 0.2520788377161614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9872792989527166, "standard_deviation": 0.0023347071168892757, "minimum": 0.9840861721542461, "maximum": 0.9907100741640473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6403553388694925, "standard_deviation": 0.08920883238278272, "minimum": 1.4652294224714235, "maximum": 1.7817369333988347, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.232522395218393, "standard_deviation": 0.39948831552069414, "minimum": 0.6547047031721628, "maximum": 1.9302749017141971, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1296223917443784, "standard_deviation": 0.010234004679818949, "minimum": 0.10829214063325611, "maximum": 0.14041567650076908, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.42289725826366714, "standard_deviation": 0.0450083996055086, "minimum": 0.3608689068886644, "maximum": 0.4897346026077825, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.461327546173601, "standard_deviation": 0.010895942762979033, "minimum": 6.444550519428373, "maximum": 6.478055892822706, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3728612431559492, "standard_deviation": 0.040170015137645045, "minimum": 0.3100354053498706, "maximum": 0.4320790802148561, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.008852785454266491, "standard_deviation": 0.06307853855570908, "minimum": -0.1071329106965467, "maximum": 0.09020248387313051, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9809779816430686, "standard_deviation": 0.002632679391142268, "minimum": 0.9766579747787227, "maximum": 0.9847453317083523, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1851915524044703, "standard_deviation": 0.04676000181261833, "minimum": 1.1309426246031398, "maximum": 1.2600780287128062, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40433127825111254, "standard_deviation": 0.21314589922500113, "minimum": 0.10068365128014582, "maximum": 0.7776427109428994, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.13330782128813576, "standard_deviation": 0.009270432723204523, "minimum": 0.11942525516180647, "maximum": 0.14720520760197683, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9943120773260695, "standard_deviation": 0.0015148620319032715, "minimum": 0.9922609271867261, "maximum": 0.9960293372567054, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6370863473304769, "standard_deviation": 0.05599119693577677, "minimum": 0.5616062919659053, "maximum": 0.7316005473847177, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6862125059116311, "standard_deviation": 0.4581928997421533, "minimum": 0.09664900801554366, "maximum": 1.417183519328787, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06813827844267456, "standard_deviation": 0.009220798672386071, "minimum": 0.05703311323300576, "maximum": 0.08006570566081078, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-75918-0013", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.7190160751342773, "predicted_mos": 1.7190160751342773, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "62aef1dd75649e3b8831f20b172d87aa7fe02a51a04b16cd19cd3cb491e563dc", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.694197999779135, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.213108646352567, "全帯域_10ms_rms_envelope_nrmse": 0.14751747054158929, "全帯域_10ms_envelope_correlation": 0.9766809052326647, "全帯域_10ms_modulation_mae_db": 0.17719664255571646, "全帯域_40ms_log_rms_mae_db": 0.59831786146803, "全帯域_40ms_rms_envelope_nrmse": 0.07584080712389721, "全帯域_40ms_envelope_correlation": 0.9929679800516302, "全帯域_40ms_modulation_mae_db": 0.32169296525181806, "300–1000 Hz_10ms_log_rms_mae_db": 1.6761737513684696, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20620853640576559, "300–1000 Hz_10ms_envelope_correlation": 0.9655301011139853, "300–1000 Hz_10ms_modulation_mae_db": 1.5278543680713674, "300–1000 Hz_40ms_log_rms_mae_db": 1.09079014220951, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11966221742544683, "300–1000 Hz_40ms_envelope_correlation": 0.9879816234878603, "300–1000 Hz_40ms_modulation_mae_db": 0.8647009287305355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.442022360668286, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20189842705417663, "1000–3000 Hz_10ms_envelope_correlation": 0.9632814177549135, "1000–3000 Hz_10ms_modulation_mae_db": 0.6868896139379421, "1000–3000 Hz_40ms_log_rms_mae_db": 0.871923133573545, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10307840701307092, "1000–3000 Hz_40ms_envelope_correlation": 0.9895591044302606, "1000–3000 Hz_40ms_modulation_mae_db": 0.8246619867519672, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5130181469959054, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18323295108757762, "3000–6000 Hz_10ms_envelope_correlation": 0.9741482484867248, "3000–6000 Hz_10ms_modulation_mae_db": 0.23239495287561526, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8946165353044095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08803303358136855, "3000–6000 Hz_40ms_envelope_correlation": 0.9933582646919594, "3000–6000 Hz_40ms_modulation_mae_db": 0.2761351122280543, "mr_stft_db": 6.0573801270350645, "band_mae_db": 0.34197167464976297, "rms_gain_error_db": -0.021619812417730692, "output_peak": 0.34500406743108475}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.295237926327935, "全帯域_10ms_rms_envelope_nrmse": 0.14561758703382577, "全帯域_10ms_envelope_correlation": 0.9774003176791667, "全帯域_10ms_modulation_mae_db": 0.8450771727019998, "全帯域_40ms_log_rms_mae_db": 0.600492471045894, "全帯域_40ms_rms_envelope_nrmse": 0.08395097393469446, "全帯域_40ms_envelope_correlation": 0.9921183971052618, "全帯域_40ms_modulation_mae_db": 1.6225808335540541, "300–1000 Hz_10ms_log_rms_mae_db": 1.8959339374748292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22833087991288056, "300–1000 Hz_10ms_envelope_correlation": 0.9581121915190842, "300–1000 Hz_10ms_modulation_mae_db": 1.3983477106369477, "300–1000 Hz_40ms_log_rms_mae_db": 1.2064948837475828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14182852024946108, "300–1000 Hz_40ms_envelope_correlation": 0.9844972336095286, "300–1000 Hz_40ms_modulation_mae_db": 1.0007214012673018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.377400314936179, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18317619899001214, "1000–3000 Hz_10ms_envelope_correlation": 0.9717898687918415, "1000–3000 Hz_10ms_modulation_mae_db": 0.7715184836988324, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8359838771662403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1016768506499864, "1000–3000 Hz_40ms_envelope_correlation": 0.9908000009088476, "1000–3000 Hz_40ms_modulation_mae_db": 0.5431404198081814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.528847765314901, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19539780131605258, "3000–6000 Hz_10ms_envelope_correlation": 0.9711025938232172, "3000–6000 Hz_10ms_modulation_mae_db": 1.2608395700637347, "3000–6000 Hz_40ms_log_rms_mae_db": 0.866460838675554, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1232350558430175, "3000–6000 Hz_40ms_envelope_correlation": 0.9886838037973195, "3000–6000 Hz_40ms_modulation_mae_db": 2.0517985848971843, "mr_stft_db": 6.052483258057158, "band_mae_db": 0.5603713331407779, "rms_gain_error_db": -0.20547020609921526, "output_peak": 0.271198578699836}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2946321202140045, "全帯域_10ms_rms_envelope_nrmse": 0.15118748752597727, "全帯域_10ms_envelope_correlation": 0.9755882468377315, "全帯域_10ms_modulation_mae_db": 0.31238176430134285, "全帯域_40ms_log_rms_mae_db": 0.6979019607230367, "全帯域_40ms_rms_envelope_nrmse": 0.07220541732964109, "全帯域_40ms_envelope_correlation": 0.9936547947680787, "全帯域_40ms_modulation_mae_db": 0.7691795665096807, "300–1000 Hz_10ms_log_rms_mae_db": 1.754979388438879, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22613170634563415, "300–1000 Hz_10ms_envelope_correlation": 0.9586182555933555, "300–1000 Hz_10ms_modulation_mae_db": 1.571635549961372, "300–1000 Hz_40ms_log_rms_mae_db": 1.141211711202812, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09935756598129585, "300–1000 Hz_40ms_envelope_correlation": 0.9916026151157132, "300–1000 Hz_40ms_modulation_mae_db": 1.3584945306387193, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4511866515874854, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21811333848081277, "1000–3000 Hz_10ms_envelope_correlation": 0.9573256021520524, "1000–3000 Hz_10ms_modulation_mae_db": 1.024735996356167, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9275372251831911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12533275832774834, "1000–3000 Hz_40ms_envelope_correlation": 0.9845282773365921, "1000–3000 Hz_40ms_modulation_mae_db": 0.8737481626008835, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5542942990052655, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17734581791882725, "3000–6000 Hz_10ms_envelope_correlation": 0.9756671854961321, "3000–6000 Hz_10ms_modulation_mae_db": 0.3631831715636686, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9288062698785882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08613360850050389, "3000–6000 Hz_40ms_envelope_correlation": 0.9936003645621999, "3000–6000 Hz_40ms_modulation_mae_db": 0.8819177693567705, "mr_stft_db": 6.058844987249668, "band_mae_db": 0.17389938246759287, "rms_gain_error_db": -0.01077231239007045, "output_peak": 0.3440731426160447}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.22371094552193, "全帯域_10ms_rms_envelope_nrmse": 0.14636945745131563, "全帯域_10ms_envelope_correlation": 0.9772071971067035, "全帯域_10ms_modulation_mae_db": 0.8844529101572189, "全帯域_40ms_log_rms_mae_db": 0.6858898652266963, "全帯域_40ms_rms_envelope_nrmse": 0.08648453237218393, "全帯域_40ms_envelope_correlation": 0.9915166543055816, "全帯域_40ms_modulation_mae_db": 1.6879387584649352, "300–1000 Hz_10ms_log_rms_mae_db": 1.8825356554421442, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23002858778434676, "300–1000 Hz_10ms_envelope_correlation": 0.9592705760811854, "300–1000 Hz_10ms_modulation_mae_db": 2.210002682972624, "300–1000 Hz_40ms_log_rms_mae_db": 1.2381461223451202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13514633968993714, "300–1000 Hz_40ms_envelope_correlation": 0.9892878178699605, "300–1000 Hz_40ms_modulation_mae_db": 1.8833949178532527, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3325159306678955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17016564515989988, "1000–3000 Hz_10ms_envelope_correlation": 0.9744372871444692, "1000–3000 Hz_10ms_modulation_mae_db": 0.46048825777062047, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7026891316018861, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09009057004423743, "1000–3000 Hz_40ms_envelope_correlation": 0.9931995249239951, "1000–3000 Hz_40ms_modulation_mae_db": 0.2918537304999623, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5114777699497977, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19067968046619171, "3000–6000 Hz_10ms_envelope_correlation": 0.9719520481506054, "3000–6000 Hz_10ms_modulation_mae_db": 1.1998536892328386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0102707466944496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11861404699278366, "3000–6000 Hz_40ms_envelope_correlation": 0.9885750842084446, "3000–6000 Hz_40ms_modulation_mae_db": 1.7749593171124785, "mr_stft_db": 6.08190550081365, "band_mae_db": 0.45678476687932584, "rms_gain_error_db": -0.2368937147682325, "output_peak": 0.3466421493829751}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.277264303941201, "全帯域_10ms_rms_envelope_nrmse": 0.13393946101055762, "全帯域_10ms_envelope_correlation": 0.9809802393508592, "全帯域_10ms_modulation_mae_db": 0.5407264169208971, "全帯域_40ms_log_rms_mae_db": 0.7185875670402976, "全帯域_40ms_rms_envelope_nrmse": 0.06641659823564211, "全帯域_40ms_envelope_correlation": 0.9947009018152844, "全帯域_40ms_modulation_mae_db": 0.7820356022642226, "300–1000 Hz_10ms_log_rms_mae_db": 1.7013891365752192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1843695183660461, "300–1000 Hz_10ms_envelope_correlation": 0.9724517506275542, "300–1000 Hz_10ms_modulation_mae_db": 1.6290382067290388, "300–1000 Hz_40ms_log_rms_mae_db": 0.9899939250489357, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07423665801881124, "300–1000 Hz_40ms_envelope_correlation": 0.9956939753631316, "300–1000 Hz_40ms_modulation_mae_db": 0.9074144493126299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3591180834427772, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18093465312175783, "1000–3000 Hz_10ms_envelope_correlation": 0.9708472278405793, "1000–3000 Hz_10ms_modulation_mae_db": 0.4438529358641132, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8561958820158849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09369238691845676, "1000–3000 Hz_40ms_envelope_correlation": 0.9915449421170225, "1000–3000 Hz_40ms_modulation_mae_db": 0.48548170636003984, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4790728303983072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16876863742828838, "3000–6000 Hz_10ms_envelope_correlation": 0.9780847823289696, "3000–6000 Hz_10ms_modulation_mae_db": 0.4683377034185152, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8753071919468207, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08741953971109609, "3000–6000 Hz_40ms_envelope_correlation": 0.9934104635914223, "3000–6000 Hz_40ms_modulation_mae_db": 0.6319757583100912, "mr_stft_db": 6.03444946954698, "band_mae_db": 0.15115237909325774, "rms_gain_error_db": -0.003399697767816889, "output_peak": 0.2973537810377894}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.219744063742198, "全帯域_10ms_rms_envelope_nrmse": 0.15386147720905838, "全帯域_10ms_envelope_correlation": 0.9744778978642232, "全帯域_10ms_modulation_mae_db": 0.5083524859104439, "全帯域_40ms_log_rms_mae_db": 0.6792280709316163, "全帯域_40ms_rms_envelope_nrmse": 0.07801747651809035, "全帯域_40ms_envelope_correlation": 0.9928208672530018, "全帯域_40ms_modulation_mae_db": 0.856285836246002, "300–1000 Hz_10ms_log_rms_mae_db": 1.8724056106433662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23635475547044194, "300–1000 Hz_10ms_envelope_correlation": 0.9544225448208702, "300–1000 Hz_10ms_modulation_mae_db": 1.9122727122214362, "300–1000 Hz_40ms_log_rms_mae_db": 1.1801876805290856, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11060047770874723, "300–1000 Hz_40ms_envelope_correlation": 0.9902370868700331, "300–1000 Hz_40ms_modulation_mae_db": 0.8822980469251389, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4116079651990567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18325585726272484, "1000–3000 Hz_10ms_envelope_correlation": 0.9699141148126394, "1000–3000 Hz_10ms_modulation_mae_db": 0.570387251023012, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8786415502686749, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1046915739418525, "1000–3000 Hz_40ms_envelope_correlation": 0.9898596550236246, "1000–3000 Hz_40ms_modulation_mae_db": 0.4411029814005117, "3000–6000 Hz_10ms_log_rms_mae_db": 1.549772215993766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1968731693003056, "3000–6000 Hz_10ms_envelope_correlation": 0.9698386537318181, "3000–6000 Hz_10ms_modulation_mae_db": 0.7298016261647069, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9044338377893555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10761196533844361, "3000–6000 Hz_40ms_envelope_correlation": 0.9903972937103878, "3000–6000 Hz_40ms_modulation_mae_db": 1.118873276874662, "mr_stft_db": 6.075926035170499, "band_mae_db": 0.4477582161641845, "rms_gain_error_db": -0.14223951715640043, "output_peak": 0.30488292030484176}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.269877811034483, "全帯域_10ms_rms_envelope_nrmse": 0.1352035761876597, "全帯域_10ms_envelope_correlation": 0.9803579812445581, "全帯域_10ms_modulation_mae_db": 0.3407807752773331, "全帯域_40ms_log_rms_mae_db": 0.6750541533152555, "全帯域_40ms_rms_envelope_nrmse": 0.06450141332085028, "全帯域_40ms_envelope_correlation": 0.9950091206279124, "全帯域_40ms_modulation_mae_db": 0.6522759772776817, "300–1000 Hz_10ms_log_rms_mae_db": 1.6660484144931804, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21576538287810978, "300–1000 Hz_10ms_envelope_correlation": 0.9673988266729671, "300–1000 Hz_10ms_modulation_mae_db": 1.9572375133908704, "300–1000 Hz_40ms_log_rms_mae_db": 1.0401203989631325, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14104369957378265, "300–1000 Hz_40ms_envelope_correlation": 0.9913307941928603, "300–1000 Hz_40ms_modulation_mae_db": 1.285864859112225, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3924828938494604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.192976766605251, "1000–3000 Hz_10ms_envelope_correlation": 0.9664263918529695, "1000–3000 Hz_10ms_modulation_mae_db": 0.6929453304448977, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9175197207998976, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12082131451426464, "1000–3000 Hz_40ms_envelope_correlation": 0.9855538933333349, "1000–3000 Hz_40ms_modulation_mae_db": 0.7557973812540842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4966087185517294, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17206810770610542, "3000–6000 Hz_10ms_envelope_correlation": 0.9770502604629704, "3000–6000 Hz_10ms_modulation_mae_db": 0.7872197034767595, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8970887031094351, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08572563527913067, "3000–6000 Hz_40ms_envelope_correlation": 0.9936571294204747, "3000–6000 Hz_40ms_modulation_mae_db": 1.0169305716931492, "mr_stft_db": 6.05592360657183, "band_mae_db": 0.3435270349962483, "rms_gain_error_db": -0.09393487920950121, "output_peak": 0.29663560029202773}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2651070145957943, "全帯域_10ms_rms_envelope_nrmse": 0.14536448824375844, "全帯域_10ms_envelope_correlation": 0.9775371267237174, "全帯域_10ms_modulation_mae_db": 0.8815118786179742, "全帯域_40ms_log_rms_mae_db": 0.6532740249726906, "全帯域_40ms_rms_envelope_nrmse": 0.0719963380811584, "全帯域_40ms_envelope_correlation": 0.9937382744090257, "全帯域_40ms_modulation_mae_db": 1.346870185143268, "300–1000 Hz_10ms_log_rms_mae_db": 1.7316485244870674, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20810321524933834, "300–1000 Hz_10ms_envelope_correlation": 0.9648512257265608, "300–1000 Hz_10ms_modulation_mae_db": 2.3652796686641477, "300–1000 Hz_40ms_log_rms_mae_db": 1.0003245553803726, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11213709026798241, "300–1000 Hz_40ms_envelope_correlation": 0.9893479401740447, "300–1000 Hz_40ms_modulation_mae_db": 1.8090648108976592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4630392217543378, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19415040793979543, "1000–3000 Hz_10ms_envelope_correlation": 0.9682584163343233, "1000–3000 Hz_10ms_modulation_mae_db": 0.7113441226065236, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8786583597997726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10374412417988775, "1000–3000 Hz_40ms_envelope_correlation": 0.9903990734086379, "1000–3000 Hz_40ms_modulation_mae_db": 0.6676773920296873, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5874950744919047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19832602521065687, "3000–6000 Hz_10ms_envelope_correlation": 0.9695008120676465, "3000–6000 Hz_10ms_modulation_mae_db": 1.2409828857144285, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9453802093332012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11051672499075874, "3000–6000 Hz_40ms_envelope_correlation": 0.989410940519432, "3000–6000 Hz_40ms_modulation_mae_db": 1.5915554075990954, "mr_stft_db": 6.053006783100732, "band_mae_db": 0.5052799049273525, "rms_gain_error_db": 0.045179931000589396, "output_peak": 0.2902567725506395}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 47040, "output_samples": 47040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.970904266403743, "spectral_convergence": 0.5669627174015488, "spectral_convergence_status": "ok", "amplitude_floor": 8.904239335448439e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7838634894515675, "spectral_convergence": 0.5537994654603722, "spectral_convergence_status": "ok", "amplitude_floor": 4.6801685305432975e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.417372625249886, "spectral_convergence": 0.5980043884702847, "spectral_convergence_status": "ok", "amplitude_floor": 2.078367899935203e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.736571532646871e-05, "output_power": 5.416671161509562e-05, "delta_db": 0.5826837048757237, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00015972486768670453, "output_power": 0.00014995970149691526, "delta_db": -0.27397969779177, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 3.942775929744416e-05, "output_power": 4.9579243276712354e-05, "delta_db": 0.9949779732107694, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.533823001233142e-05, "output_power": 6.397518755438e-05, "delta_db": -0.09155797656473648, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005191422513822901, "output_power": 0.0005093713033939785, "delta_db": -0.08251901865539807, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017487193556096138, "output_power": 0.00017382368371076152, "delta_db": -0.026111676800179943, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.34197167464976297, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.021619812417730692, "pitch": null, "output_rms_dbfs": -29.99619797880959, "output_peak": 0.34500406743108475, "reference_rms_dbfs": -29.974578166391858, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0573801270350645}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 47040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001280716783512989, "log_rms_mae_db": 1.213108646352567, "rms_envelope_nrmse": 0.14751747054158929, "envelope_correlation": 0.9766809052326647, "rms_gain_error_db": -0.021567124749837213, "reference_mean_power": 0.0010075555569028208, "output_mean_power": 0.0010025644251790364}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 5.126669358054452e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020945501880404742, "output_power": 0.00019125557158508575, "reference_active": true, "delta_db": -0.39476675347741463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010240530560263733, "output_power": 0.00010319948530548057, "reference_active": true, "delta_db": 0.033550733562883205}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.779418868232472e-05, "output_power": 2.8463037533081235e-05, "reference_active": true, "delta_db": 0.1032724406268516}], "active_band_count": 3, "active_band_mae_db": 0.17719664255571646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 9.465799671406843e-05, "log_rms_mae_db": 0.59831786146803, "rms_envelope_nrmse": 0.07584080712389721, "envelope_correlation": 0.9929679800516302, "rms_gain_error_db": -0.02138094673142953, "reference_mean_power": 0.001017821671515535, "output_mean_power": 0.001012823102358312}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 4.336511567866126e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018218362838522706, "output_power": 0.0001683557282064154, "reference_active": true, "delta_db": -0.34281449678056014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6513405903536415e-05, "output_power": 5.680447456731859e-05, "reference_active": true, "delta_db": 0.02231065247891137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.038210274571303e-06, "output_power": 1.7752258443472334e-06, "reference_active": true, "delta_db": -0.5999537464959827}], "active_band_count": 3, "active_band_mae_db": 0.32169296525181806}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.812061629421404e-05, "log_rms_mae_db": 1.6761737513684696, "rms_envelope_nrmse": 0.20620853640576559, "envelope_correlation": 0.9655301011139853, "rms_gain_error_db": -0.21896849915524932, "reference_mean_power": 0.00014923818668565985, "output_mean_power": 0.0001419002346446462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 7.990988889403065e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7339937941838396e-05, "output_power": 4.227860468366948e-05, "reference_active": true, "delta_db": -0.49107038377702367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462900675807451e-06, "output_power": 5.462606020656516e-06, "reference_active": true, "delta_db": -0.730276025411583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2961437099516358e-06, "output_power": 2.811090491213871e-06, "reference_active": true, "delta_db": 3.3622166950254955}], "active_band_count": 3, "active_band_mae_db": 1.5278543680713674}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.488253980944051e-05, "log_rms_mae_db": 1.09079014220951, "rms_envelope_nrmse": 0.11966221742544683, "envelope_correlation": 0.9879816234878603, "rms_gain_error_db": -0.21897170600955673, "reference_mean_power": 0.00015077911609921791, "output_mean_power": 0.00014336529162231154}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 7.46020344966024e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.50390789043777e-05, "output_power": 4.121211995026742e-05, "reference_active": true, "delta_db": -0.3856454496909154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3028613860270352e-06, "output_power": 2.581861892851681e-06, "reference_active": true, "delta_db": -1.069573400951797}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.84375886973483e-08, "output_power": 1.1495431723874841e-07, "reference_active": true, "delta_db": 1.138883935548894}], "active_band_count": 3, "active_band_mae_db": 0.8647009287305355}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.5882621361725546e-05, "log_rms_mae_db": 1.442022360668286, "rms_envelope_nrmse": 0.20189842705417663, "envelope_correlation": 0.9632814177549135, "rms_gain_error_db": -0.1444621072413457, "reference_mean_power": 0.0001648439551682589, "output_mean_power": 0.0001594508414592018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 1.0813916284336453e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.944970033970189e-05, "output_power": 4.710315190961212e-05, "reference_active": true, "delta_db": -0.21113695222719966}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1861966099880283e-05, "output_power": 1.2806296398173182e-05, "reference_active": true, "delta_db": 0.3326687069366667}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.815426369634166e-06, "output_power": 3.3958494515933454e-06, "reference_active": true, "delta_db": -1.5168631826499601}], "active_band_count": 3, "active_band_mae_db": 0.6868896139379421}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.018253092677299e-05, "log_rms_mae_db": 0.871923133573545, "rms_envelope_nrmse": 0.10307840701307092, "envelope_correlation": 0.9895591044302606, "rms_gain_error_db": -0.1444644184771151, "reference_mean_power": 0.00016654362747028494, "output_mean_power": 0.0001610948207372252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 9.377377349368422e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3529785361500865e-05, "output_power": 4.193347923052499e-05, "reference_active": true, "delta_db": -0.16225628312473275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.594238028592104e-06, "output_power": 5.619364172441768e-06, "reference_active": true, "delta_db": 0.0194623702391506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1488780558631956e-07, "output_power": 1.8575012371708798e-07, "reference_active": true, "delta_db": -2.2922673068920183}], "active_band_count": 3, "active_band_mae_db": 0.8246619867519672}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001195123420429498, "log_rms_mae_db": 1.5130181469959054, "rms_envelope_nrmse": 0.18323295108757762, "envelope_correlation": 0.9741482484867248, "rms_gain_error_db": -0.032871391334951955, "reference_mean_power": 0.0005186016960204273, "output_mean_power": 0.0005146912600851374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 3.8692936095348015e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017018269823121574, "output_power": 0.00016269436670177714, "reference_active": true, "delta_db": -0.19542889339561514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010605303684284744, "output_power": 0.00010530857039677409, "reference_active": true, "delta_db": -0.030593920132414534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2588117400162252e-05, "output_power": 2.5176551351249105e-05, "reference_active": true, "delta_db": 0.4711620450988161}], "active_band_count": 3, "active_band_mae_db": 0.23239495287561526}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 8.500195655960054e-05, "log_rms_mae_db": 0.8946165353044095, "rms_envelope_nrmse": 0.08803303358136855, "envelope_correlation": 0.9933582646919594, "rms_gain_error_db": -0.03283118945137301, "reference_mean_power": 0.0005239212873916433, "output_mean_power": 0.000519975553200629}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 3.3612492298744645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016844288389443048, "output_power": 0.00016435452742470723, "reference_active": true, "delta_db": -0.10670996364154069}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.87289448022482e-05, "output_power": 6.0091819172285605e-05, "reference_active": true, "delta_db": 0.09963153865829306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9018720843055598e-06, "output_power": 1.6480672165720145e-06, "reference_active": true, "delta_db": -0.6220638343843292}], "active_band_count": 3, "active_band_mae_db": 0.2761351122280543}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7855706445650342, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bd14438d767a55bb6ad1d35f398cb1f9c1522edf927ec424abece9818d058dfa", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9677850408354736, "standard_deviation": 0.005424197939586565, "minimum": 0.9573256021520524, "maximum": 0.9744372871444692, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4036716777631848, "standard_deviation": 0.04656711601308109, "minimum": 1.3325159306678955, "maximum": 1.4630392217543378, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6702702489627635, "standard_deviation": 0.18655695935926553, "minimum": 0.4438529358641132, "maximum": 1.024735996356167, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19058391182680381, "standard_deviation": 0.014749923174849928, "minimum": 0.17016564515989988, "maximum": 0.21811333848081277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9894305589352894, "standard_deviation": 0.0029459575432604023, "minimum": 0.9845282773365921, "maximum": 0.9931995249239951, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8586436100511365, "standard_deviation": 0.0697248002646508, "minimum": 0.7026891316018861, "maximum": 0.9275372251831911, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6104329700881648, "standard_deviation": 0.20352015718413352, "minimum": 0.2918537304999623, "maximum": 0.8737481626008835, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1053909981986881, "standard_deviation": 0.012124788213581053, "minimum": 0.09009057004423743, "maximum": 0.12533275832774834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9734180730685105, "standard_deviation": 0.0033000102418813816, "minimum": 0.9695008120676465, "maximum": 0.9780847823289696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.527573352587697, "standard_deviation": 0.035012378222635185, "minimum": 1.4790728303983072, "maximum": 1.5874950744919047, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7853266628137834, "standard_deviation": 0.4128429294533762, "minimum": 0.23239495287561526, "maximum": 1.2608395700637347, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18533652380425067, "standard_deviation": 0.011652235149251702, "minimum": 0.16876863742828838, "maximum": 0.19832602521065687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991386668062705, "standard_deviation": 0.0023339490911466858, "minimum": 0.9885750842084446, "maximum": 0.9936571294204747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9152955415914767, "standard_deviation": 0.04624299142346762, "minimum": 0.866460838675554, "maximum": 1.0102707466944496, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1680182247589357, "standard_deviation": 0.5998478554062948, "minimum": 0.2761351122280543, "maximum": 2.0517985848971843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10091120127963783, "standard_deviation": 0.015792232954812878, "minimum": 0.08572563527913067, "maximum": 0.1232350558430175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9625819340194453, "standard_deviation": 0.005948435291938434, "minimum": 0.9544225448208702, "maximum": 0.9724517506275542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7726393023653944, "standard_deviation": 0.09633869879334987, "minimum": 1.6660484144931804, "maximum": 1.8959339374748292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.8214585515809754, "standard_deviation": 0.34600875488869126, "minimum": 1.3983477106369477, "maximum": 2.3652796686641477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21691157280157042, "standard_deviation": 0.01697875176679635, "minimum": 0.1843695183660461, "maximum": 0.23635475547044194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9899973858353914, "standard_deviation": 0.0032094406819732466, "minimum": 0.9844972336095286, "maximum": 0.9956939753631316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.110908677428319, "standard_deviation": 0.09510062758808763, "minimum": 0.9899939250489357, "maximum": 1.2381461223451202, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.2489942430921828, "standard_deviation": 0.4123030479791783, "minimum": 0.8647009287305355, "maximum": 1.8833949178532527, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11675157111443304, "standard_deviation": 0.023099161671764214, "minimum": 0.07423665801881124, "maximum": 0.14182852024946108, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3725930865398128, "standard_deviation": 0.14923719308107342, "minimum": 0.15115237909325774, "maximum": 0.5603713331407779, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.058739970943197, "standard_deviation": 0.014657221754919231, "minimum": 6.03444946954698, "maximum": 6.08190550081365, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3120058765394048, "standard_deviation": 0.029171381360989655, "minimum": 0.271198578699836, "maximum": 0.3466421493829751, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08364377610104726, "standard_deviation": 0.10285806610077297, "minimum": -0.2368937147682325, "maximum": 0.045179931000589396, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9775287390049531, "standard_deviation": 0.0021983559276903446, "minimum": 0.9744778978642232, "maximum": 0.9809802393508592, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2573353539662642, "standard_deviation": 0.033683901114015376, "minimum": 1.213108646352567, "maximum": 1.295237926327935, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5613100058053657, "standard_deviation": 0.28001123722387317, "minimum": 0.17719664255571646, "maximum": 0.8844529101572189, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14488262565046778, "standard_deviation": 0.007007479123033423, "minimum": 0.13393946101055762, "maximum": 0.15386147720905838, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9933158737919721, "standard_deviation": 0.0012021338103394548, "minimum": 0.9915166543055816, "maximum": 0.9950091206279124, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6635932468404396, "standard_deviation": 0.04379085451963232, "minimum": 0.59831786146803, "maximum": 0.7185875670402976, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.004857465588958, "standard_deviation": 0.49040064330831024, "minimum": 0.32169296525181806, "maximum": 1.6879387584649352, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07492669461451973, "standard_deviation": 0.007769080982973389, "minimum": 0.06450141332085028, "maximum": 0.08648453237218393, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-75918-0013", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.9538862705230713, "predicted_mos": 1.9538862705230713, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "0fdd31b8739c060275289901ae3e62b5889ccab808310d0e6da9a0c8c5b83aec", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8942670999094844, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.227742849045206, "全帯域_10ms_rms_envelope_nrmse": 0.14485479226386794, "全帯域_10ms_envelope_correlation": 0.9777123522089809, "全帯域_10ms_modulation_mae_db": 0.8755927440096066, "全帯域_40ms_log_rms_mae_db": 0.7289907829586745, "全帯域_40ms_rms_envelope_nrmse": 0.08005336244048435, "全帯域_40ms_envelope_correlation": 0.993033390635596, "全帯域_40ms_modulation_mae_db": 1.776286499650684, "300–1000 Hz_10ms_log_rms_mae_db": 2.081164414589326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.25584381095794734, "300–1000 Hz_10ms_envelope_correlation": 0.9491972413190808, "300–1000 Hz_10ms_modulation_mae_db": 2.383918393435401, "300–1000 Hz_40ms_log_rms_mae_db": 1.3743939010311597, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14657219213184297, "300–1000 Hz_40ms_envelope_correlation": 0.9885711911484073, "300–1000 Hz_40ms_modulation_mae_db": 1.8586997813017827, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6331828627102591, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1921319767565315, "1000–3000 Hz_10ms_envelope_correlation": 0.966657582348528, "1000–3000 Hz_10ms_modulation_mae_db": 0.5974239051746187, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0910919632974019, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09850565122147692, "1000–3000 Hz_40ms_envelope_correlation": 0.990499883730537, "1000–3000 Hz_40ms_modulation_mae_db": 1.0463759324584687, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7308024858492497, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.189831449193823, "3000–6000 Hz_10ms_envelope_correlation": 0.9720797508250361, "3000–6000 Hz_10ms_modulation_mae_db": 0.8853346603187454, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1000537240490966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11050130136621113, "3000–6000 Hz_40ms_envelope_correlation": 0.9900323723697857, "3000–6000 Hz_40ms_modulation_mae_db": 1.6451169301433837, "mr_stft_db": 6.191418554089988, "band_mae_db": 0.23302295608997392, "rms_gain_error_db": -0.2419110092738435, "output_peak": 0.2772638659027342}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.171179064108051, "全帯域_10ms_rms_envelope_nrmse": 0.15437431325057202, "全帯域_10ms_envelope_correlation": 0.9746222235933484, "全帯域_10ms_modulation_mae_db": 0.8781841007942495, "全帯域_40ms_log_rms_mae_db": 0.6649001898011526, "全帯域_40ms_rms_envelope_nrmse": 0.10177163132465032, "全帯域_40ms_envelope_correlation": 0.9881217329730011, "全帯域_40ms_modulation_mae_db": 1.2080797408369295, "300–1000 Hz_10ms_log_rms_mae_db": 2.096784267964681, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2055536454181122, "300–1000 Hz_10ms_envelope_correlation": 0.9658550960699253, "300–1000 Hz_10ms_modulation_mae_db": 1.7382536548088316, "300–1000 Hz_40ms_log_rms_mae_db": 1.4374931013176824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10861329770751592, "300–1000 Hz_40ms_envelope_correlation": 0.9910040794303454, "300–1000 Hz_40ms_modulation_mae_db": 1.1705947769551135, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6625243919568027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17440101166043107, "1000–3000 Hz_10ms_envelope_correlation": 0.9729811080988955, "1000–3000 Hz_10ms_modulation_mae_db": 0.4513694797084953, "1000–3000 Hz_40ms_log_rms_mae_db": 1.10304417656865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10053579161506923, "1000–3000 Hz_40ms_envelope_correlation": 0.9909705412427103, "1000–3000 Hz_40ms_modulation_mae_db": 0.6200763332482473, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6820284659278226, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19765096001861993, "3000–6000 Hz_10ms_envelope_correlation": 0.9701564212154532, "3000–6000 Hz_10ms_modulation_mae_db": 1.2400879445253004, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0758273712497375, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13433564335110565, "3000–6000 Hz_40ms_envelope_correlation": 0.985684548768927, "3000–6000 Hz_40ms_modulation_mae_db": 1.6039732100210482, "mr_stft_db": 6.208416517633371, "band_mae_db": 0.1345472658329405, "rms_gain_error_db": -0.26427690999621645, "output_peak": 0.3276358370742942}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.24973554943219, "全帯域_10ms_rms_envelope_nrmse": 0.14575013628012887, "全帯域_10ms_envelope_correlation": 0.977422840133882, "全帯域_10ms_modulation_mae_db": 0.5997983039180697, "全帯域_40ms_log_rms_mae_db": 0.629616724958392, "全帯域_40ms_rms_envelope_nrmse": 0.07236140468891533, "全帯域_40ms_envelope_correlation": 0.994403737921196, "全帯域_40ms_modulation_mae_db": 0.5116329443151147, "300–1000 Hz_10ms_log_rms_mae_db": 1.977254356453506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20432243673360667, "300–1000 Hz_10ms_envelope_correlation": 0.9662079848483105, "300–1000 Hz_10ms_modulation_mae_db": 1.879557394762406, "300–1000 Hz_40ms_log_rms_mae_db": 1.2818413547303964, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08958815480009147, "300–1000 Hz_40ms_envelope_correlation": 0.9940815985119785, "300–1000 Hz_40ms_modulation_mae_db": 0.9594118584465964, "1000–3000 Hz_10ms_log_rms_mae_db": 1.703566822529742, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20896664787758687, "1000–3000 Hz_10ms_envelope_correlation": 0.961146825568188, "1000–3000 Hz_10ms_modulation_mae_db": 0.6208285963926415, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0873043270656337, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11606724844078697, "1000–3000 Hz_40ms_envelope_correlation": 0.9886581509739576, "1000–3000 Hz_40ms_modulation_mae_db": 0.9281170628988641, "3000–6000 Hz_10ms_log_rms_mae_db": 1.687172767880675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18202885445745673, "3000–6000 Hz_10ms_envelope_correlation": 0.9744571151255098, "3000–6000 Hz_10ms_modulation_mae_db": 0.6332117923708435, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0326286557673139, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09332575028270687, "3000–6000 Hz_40ms_envelope_correlation": 0.9932813254877085, "3000–6000 Hz_40ms_modulation_mae_db": 0.5151565653843814, "mr_stft_db": 6.1755047325237, "band_mae_db": 0.42459129631679543, "rms_gain_error_db": -0.24395334191828816, "output_peak": 0.28881262908826766}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.223880662048049, "全帯域_10ms_rms_envelope_nrmse": 0.15048950889506774, "全帯域_10ms_envelope_correlation": 0.9759509377637212, "全帯域_10ms_modulation_mae_db": 0.198887664504338, "全帯域_40ms_log_rms_mae_db": 0.66368043198197, "全帯域_40ms_rms_envelope_nrmse": 0.07780731787585123, "全帯域_40ms_envelope_correlation": 0.9926799985239113, "全帯域_40ms_modulation_mae_db": 0.33268516535171955, "300–1000 Hz_10ms_log_rms_mae_db": 2.0481963287449423, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.231877862717139, "300–1000 Hz_10ms_envelope_correlation": 0.9571173897173089, "300–1000 Hz_10ms_modulation_mae_db": 2.401950817343535, "300–1000 Hz_40ms_log_rms_mae_db": 1.298630249097126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09973356630028349, "300–1000 Hz_40ms_envelope_correlation": 0.9914952888145353, "300–1000 Hz_40ms_modulation_mae_db": 1.9566207199240757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.720432348130242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21677760217536451, "1000–3000 Hz_10ms_envelope_correlation": 0.9617243962278247, "1000–3000 Hz_10ms_modulation_mae_db": 0.6175032963274191, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1282582938869656, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12320105706503294, "1000–3000 Hz_40ms_envelope_correlation": 0.98746324163523, "1000–3000 Hz_40ms_modulation_mae_db": 0.7452246573672848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7669383489077726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2065512647151612, "3000–6000 Hz_10ms_envelope_correlation": 0.9667992642540716, "3000–6000 Hz_10ms_modulation_mae_db": 0.4511093331640667, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0945260662135434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11916527569375955, "3000–6000 Hz_40ms_envelope_correlation": 0.9876974717729441, "3000–6000 Hz_40ms_modulation_mae_db": 0.561156307738809, "mr_stft_db": 6.189131394190507, "band_mae_db": 0.37759325654294135, "rms_gain_error_db": 0.01818770240976628, "output_peak": 0.331756291422835}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1821982388621344, "全帯域_10ms_rms_envelope_nrmse": 0.13172184547057783, "全帯域_10ms_envelope_correlation": 0.981420629449119, "全帯域_10ms_modulation_mae_db": 0.3680517055621868, "全帯域_40ms_log_rms_mae_db": 0.5777139744823454, "全帯域_40ms_rms_envelope_nrmse": 0.06648687345505237, "全帯域_40ms_envelope_correlation": 0.9948843773560755, "全帯域_40ms_modulation_mae_db": 0.7782895411460334, "300–1000 Hz_10ms_log_rms_mae_db": 1.9848193309347542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.24005324499874436, "300–1000 Hz_10ms_envelope_correlation": 0.9547669086409819, "300–1000 Hz_10ms_modulation_mae_db": 1.4176010966148187, "300–1000 Hz_40ms_log_rms_mae_db": 1.3405398846236745, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15257294807088584, "300–1000 Hz_40ms_envelope_correlation": 0.9847926484894586, "300–1000 Hz_40ms_modulation_mae_db": 0.8727223780696597, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6128123800863758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17330456349817475, "1000–3000 Hz_10ms_envelope_correlation": 0.9730664513879113, "1000–3000 Hz_10ms_modulation_mae_db": 0.8216516050044215, "1000–3000 Hz_40ms_log_rms_mae_db": 1.026911794398089, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09946952275012416, "1000–3000 Hz_40ms_envelope_correlation": 0.9906536244116162, "1000–3000 Hz_40ms_modulation_mae_db": 0.7867612157385269, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7251012074365935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16496509728306227, "3000–6000 Hz_10ms_envelope_correlation": 0.9788873627819497, "3000–6000 Hz_10ms_modulation_mae_db": 0.19204595788009168, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0112866458220595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07735078983889414, "3000–6000 Hz_40ms_envelope_correlation": 0.9950033547464215, "3000–6000 Hz_40ms_modulation_mae_db": 0.6870948382448215, "mr_stft_db": 6.174669485875701, "band_mae_db": 0.2416117408783529, "rms_gain_error_db": -0.1512260279693476, "output_peak": 0.35136765509599677}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.23280677327959, "全帯域_10ms_rms_envelope_nrmse": 0.1394558372477274, "全帯域_10ms_envelope_correlation": 0.9799514135431195, "全帯域_10ms_modulation_mae_db": 1.2063311400751813, "全帯域_40ms_log_rms_mae_db": 0.7042074722107189, "全帯域_40ms_rms_envelope_nrmse": 0.08313608728642316, "全帯域_40ms_envelope_correlation": 0.9932059005121726, "全帯域_40ms_modulation_mae_db": 1.6293602076949683, "300–1000 Hz_10ms_log_rms_mae_db": 2.008367088417692, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20377225888163544, "300–1000 Hz_10ms_envelope_correlation": 0.9674200181554994, "300–1000 Hz_10ms_modulation_mae_db": 1.3719606907494513, "300–1000 Hz_40ms_log_rms_mae_db": 1.4782812783596735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13373392934593567, "300–1000 Hz_40ms_envelope_correlation": 0.9871667907006968, "300–1000 Hz_40ms_modulation_mae_db": 0.7134655648487525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7008796393804952, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20471998474913652, "1000–3000 Hz_10ms_envelope_correlation": 0.963266340860588, "1000–3000 Hz_10ms_modulation_mae_db": 0.6338238564191055, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0600342582791038, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0990420949938981, "1000–3000 Hz_40ms_envelope_correlation": 0.9905282872655728, "1000–3000 Hz_40ms_modulation_mae_db": 0.5256260818070221, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8203615464114442, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18983089805836104, "3000–6000 Hz_10ms_envelope_correlation": 0.9732974643110824, "3000–6000 Hz_10ms_modulation_mae_db": 1.296758846625359, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1437186914393989, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12176171167160751, "3000–6000 Hz_40ms_envelope_correlation": 0.9895900245385476, "3000–6000 Hz_40ms_modulation_mae_db": 1.6923746877518637, "mr_stft_db": 6.208500520633116, "band_mae_db": 0.5139403099210657, "rms_gain_error_db": -0.327660739642443, "output_peak": 0.2772779619359492}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2581970811280148, "全帯域_10ms_rms_envelope_nrmse": 0.14002567098626467, "全帯域_10ms_envelope_correlation": 0.9789313273332758, "全帯域_10ms_modulation_mae_db": 0.3406825639457831, "全帯域_40ms_log_rms_mae_db": 0.654180522758724, "全帯域_40ms_rms_envelope_nrmse": 0.07869521259696137, "全帯域_40ms_envelope_correlation": 0.9924557960330687, "全帯域_40ms_modulation_mae_db": 0.9789482689545469, "300–1000 Hz_10ms_log_rms_mae_db": 2.05218443020858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2551368156604947, "300–1000 Hz_10ms_envelope_correlation": 0.9534238909009933, "300–1000 Hz_10ms_modulation_mae_db": 2.681759725430192, "300–1000 Hz_40ms_log_rms_mae_db": 1.4388573349292737, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1461377113372503, "300–1000 Hz_40ms_envelope_correlation": 0.9851564265677305, "300–1000 Hz_40ms_modulation_mae_db": 2.732092011371647, "1000–3000 Hz_10ms_log_rms_mae_db": 1.598736244224971, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1908707204512054, "1000–3000 Hz_10ms_envelope_correlation": 0.9672518883600744, "1000–3000 Hz_10ms_modulation_mae_db": 0.3967395581434818, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0339129761838013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10302028372136748, "1000–3000 Hz_40ms_envelope_correlation": 0.9896170504615022, "1000–3000 Hz_40ms_modulation_mae_db": 0.14672263326410367, "3000–6000 Hz_10ms_log_rms_mae_db": 1.674149716020978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18347132655664827, "3000–6000 Hz_10ms_envelope_correlation": 0.9738988413080755, "3000–6000 Hz_10ms_modulation_mae_db": 0.7500253817657608, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0340464148518012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10790138811847545, "3000–6000 Hz_40ms_envelope_correlation": 0.9903684539657128, "3000–6000 Hz_40ms_modulation_mae_db": 1.3529675288752265, "mr_stft_db": 6.198716312305259, "band_mae_db": 0.298092874886006, "rms_gain_error_db": -0.07101985150760401, "output_peak": 0.29653124994562197}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1554431511153969, "全帯域_10ms_rms_envelope_nrmse": 0.13841631260117873, "全帯域_10ms_envelope_correlation": 0.9794583648351867, "全帯域_10ms_modulation_mae_db": 0.3559694214638891, "全帯域_40ms_log_rms_mae_db": 0.6115360656905763, "全帯域_40ms_rms_envelope_nrmse": 0.06845621062718078, "全帯域_40ms_envelope_correlation": 0.9943051849554071, "全帯域_40ms_modulation_mae_db": 0.6699111702597044, "300–1000 Hz_10ms_log_rms_mae_db": 1.9833152239815857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18345342372150006, "300–1000 Hz_10ms_envelope_correlation": 0.9730133631729369, "300–1000 Hz_10ms_modulation_mae_db": 1.3968011736440153, "300–1000 Hz_40ms_log_rms_mae_db": 1.4064072460342611, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09692549658213538, "300–1000 Hz_40ms_envelope_correlation": 0.9931139617714587, "300–1000 Hz_40ms_modulation_mae_db": 0.8296820953908037, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6223108126054413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1844371318997473, "1000–3000 Hz_10ms_envelope_correlation": 0.971913351730644, "1000–3000 Hz_10ms_modulation_mae_db": 0.38315204877574643, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0491118142384819, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09710989265072814, "1000–3000 Hz_40ms_envelope_correlation": 0.992203310172264, "1000–3000 Hz_40ms_modulation_mae_db": 0.14145142563790142, "3000–6000 Hz_10ms_log_rms_mae_db": 1.682081630033614, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.175284648253162, "3000–6000 Hz_10ms_envelope_correlation": 0.9762133441263233, "3000–6000 Hz_10ms_modulation_mae_db": 0.4050777519186591, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9709156178640134, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09179721149299022, "3000–6000 Hz_40ms_envelope_correlation": 0.9928497337404509, "3000–6000 Hz_40ms_modulation_mae_db": 0.8296373950856178, "mr_stft_db": 6.185411016121232, "band_mae_db": 0.305240049200505, "rms_gain_error_db": -0.0025056433945405622, "output_peak": 0.34292407299648786}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 47040, "output_samples": 47040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.177519427120053, "spectral_convergence": 0.5697612034518823, "spectral_convergence_status": "ok", "amplitude_floor": 8.904239335448439e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.957742158967163, "spectral_convergence": 0.5698050907810357, "spectral_convergence_status": "ok", "amplitude_floor": 4.6801685305432975e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.438994076182747, "spectral_convergence": 0.5869418070100196, "spectral_convergence_status": "ok", "amplitude_floor": 2.078367899935203e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.736571532646871e-05, "output_power": 4.829805774174302e-05, "delta_db": 0.08465565890387247, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00015972486768670453, "output_power": 0.00014052602462013224, "delta_db": -0.5561577648604061, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 3.942775929744416e-05, "output_power": 4.089092231496508e-05, "delta_db": 0.1582480982569567, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.533823001233142e-05, "output_power": 6.44122715968702e-05, "delta_db": -0.06198749944900702, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005191422513822901, "output_power": 0.00047585475517637006, "delta_db": -0.37811962664781906, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017487193556096138, "output_power": 0.0001813915391351143, "delta_db": 0.15896908842178203, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23302295608997392, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2419110092738435, "pitch": null, "output_rms_dbfs": -30.2164891756657, "output_peak": 0.2772638659027342, "reference_rms_dbfs": -29.974578166391858, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.191418554089988}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 47040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001280716783512989, "log_rms_mae_db": 1.227742849045206, "rms_envelope_nrmse": 0.14485479226386794, "envelope_correlation": 0.9777123522089809, "rms_gain_error_db": -0.24185650948662682, "reference_mean_power": 0.0010075555569028208, "output_mean_power": 0.0009529790462719074}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 5.126669358054452e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020945501880404742, "output_power": 0.0001897288387187065, "reference_active": true, "delta_db": -0.429574225414353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010240530560263733, "output_power": 8.349895478393699e-05, "reference_active": true, "delta_db": -0.8864141880755421}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.779418868232472e-05, "output_power": 2.055299040853402e-05, "reference_active": true, "delta_db": -1.3107898185389248}], "active_band_count": 3, "active_band_mae_db": 0.8755927440096066}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 9.465799671406843e-05, "log_rms_mae_db": 0.7289907829586745, "rms_envelope_nrmse": 0.08005336244048435, "envelope_correlation": 0.993033390635596, "rms_gain_error_db": -0.24199465423789648, "reference_mean_power": 0.001017821671515535, "output_mean_power": 0.0009626584520254936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 4.336511567866126e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018218362838522706, "output_power": 0.00016287334397268902, "reference_active": true, "delta_db": -0.4865933425208001}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6513405903536415e-05, "output_power": 4.4060014356033386e-05, "reference_active": true, "delta_db": -1.0810684746927883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.038210274571303e-06, "output_power": 8.572928830011687e-07, "reference_active": true, "delta_db": -3.761197681738463}], "active_band_count": 3, "active_band_mae_db": 1.776286499650684}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.812061629421404e-05, "log_rms_mae_db": 2.081164414589326, "rms_envelope_nrmse": 0.25584381095794734, "envelope_correlation": 0.9491972413190808, "rms_gain_error_db": -0.8244678851499108, "reference_mean_power": 0.00014923818668565985, "output_mean_power": 0.00012343353730756955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 7.990988889403065e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7339937941838396e-05, "output_power": 3.449283437235286e-05, "reference_active": true, "delta_db": -1.3749880125267748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462900675807451e-06, "output_power": 4.982338970834469e-06, "reference_active": true, "delta_db": -1.1299421037514765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2961437099516358e-06, "output_power": 3.7786412650009917e-06, "reference_active": true, "delta_db": 4.646825064027952}], "active_band_count": 3, "active_band_mae_db": 2.383918393435401}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.488253980944051e-05, "log_rms_mae_db": 1.3743939010311597, "rms_envelope_nrmse": 0.14657219213184297, "envelope_correlation": 0.9885711911484073, "rms_gain_error_db": -0.8244753031352643, "reference_mean_power": 0.00015077911609921791, "output_mean_power": 0.00012470781290397438}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 7.46020344966024e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.50390789043777e-05, "output_power": 3.406438293180092e-05, "reference_active": true, "delta_db": -1.2128897395859963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3028613860270352e-06, "output_power": 2.6601850230397986e-06, "reference_active": true, "delta_db": -0.9397850363700371}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.84375886973483e-08, "output_power": 1.945267583169398e-07, "reference_active": true, "delta_db": 3.423424567949314}], "active_band_count": 3, "active_band_mae_db": 1.8586997813017827}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.5882621361725546e-05, "log_rms_mae_db": 1.6331828627102591, "rms_envelope_nrmse": 0.1921319767565315, "envelope_correlation": 0.966657582348528, "rms_gain_error_db": -0.15589193422636052, "reference_mean_power": 0.0001648439551682589, "output_mean_power": 0.0001590317480851459}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 1.0813916284336453e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.944970033970189e-05, "output_power": 4.9059851600166984e-05, "reference_active": true, "delta_db": -0.034374339729325444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1861966099880283e-05, "output_power": 1.0698268122830741e-05, "reference_active": true, "delta_db": -0.44843200437175046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.815426369634166e-06, "output_power": 3.561953203986493e-06, "reference_active": true, "delta_db": -1.3094653714227804}], "active_band_count": 3, "active_band_mae_db": 0.5974239051746187}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.018253092677299e-05, "log_rms_mae_db": 1.0910919632974019, "rms_envelope_nrmse": 0.09850565122147692, "envelope_correlation": 0.990499883730537, "rms_gain_error_db": -0.1559456613840564, "reference_mean_power": 0.00016654362747028494, "output_mean_power": 0.00016066950423239537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 9.377377349368422e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3529785361500865e-05, "output_power": 4.33291233008048e-05, "reference_active": true, "delta_db": -0.020066238930557433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.594238028592104e-06, "output_power": 4.723303238345955e-06, "reference_active": true, "delta_db": -0.7349511278777936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1488780558631956e-07, "output_power": 1.8186318777802807e-07, "reference_active": true, "delta_db": -2.384110430567055}], "active_band_count": 3, "active_band_mae_db": 1.0463759324584687}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001195123420429498, "log_rms_mae_db": 1.7308024858492497, "rms_envelope_nrmse": 0.189831449193823, "envelope_correlation": 0.9720797508250361, "rms_gain_error_db": -0.2602539765506668, "reference_mean_power": 0.0005186016960204273, "output_mean_power": 0.0004884369774631226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 3.8692936095348015e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017018269823121574, "output_power": 0.00015600789358277993, "reference_active": true, "delta_db": -0.3776883202354451}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010605303684284744, "output_power": 8.449447662586988e-05, "reference_active": true, "delta_db": -0.9869478891053505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2588117400162252e-05, "output_power": 1.6778118566865054e-05, "reference_active": true, "delta_db": -1.291367771615441}], "active_band_count": 3, "active_band_mae_db": 0.8853346603187454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 8.500195655960054e-05, "log_rms_mae_db": 1.1000537240490966, "rms_envelope_nrmse": 0.11050130136621113, "envelope_correlation": 0.9900323723697857, "rms_gain_error_db": -0.2604082180078944, "reference_mean_power": 0.0005239212873916433, "output_mean_power": 0.0004934296275627786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 3.3612492298744645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016844288389443048, "output_power": 0.00015067104661668353, "reference_active": true, "delta_db": -0.4842286327535006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.87289448022482e-05, "output_power": 4.540785199496851e-05, "reference_active": true, "delta_db": -1.1172123962001865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9018720843055598e-06, "output_power": 8.826536627714926e-07, "reference_active": true, "delta_db": -3.3339097614764643}], "active_band_count": 3, "active_band_mae_db": 1.6451169301433837}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7826110077986957, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "11d2c98880da9a998b6ba8f6d186555f1bfc64ed20aa5a1363a6c262a9478123", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9672509930728318, "standard_deviation": 0.004963079197350499, "minimum": 0.961146825568188, "maximum": 0.9730664513879113, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.656805687703041, "standard_deviation": 0.04669900288282962, "minimum": 1.598736244224971, "maximum": 1.720432348130242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5653115432432412, "standard_deviation": 0.1472717991709407, "minimum": 0.38315204877574643, "maximum": 0.8216516050044215, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19320120488352224, "standard_deviation": 0.015900516510203636, "minimum": 0.17330456349817475, "maximum": 0.21677760217536451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9900742612366737, "standard_deviation": 0.0014682816184849305, "minimum": 0.98746324163523, "maximum": 0.992203310172264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0724587004897659, "standard_deviation": 0.03560977887407835, "minimum": 1.026911794398089, "maximum": 1.1282582938869656, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6175444178025524, "standard_deviation": 0.3342542841258711, "minimum": 0.14145142563790142, "maximum": 1.0463759324584687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10461894280731049, "standard_deviation": 0.009613698808197148, "minimum": 0.09710989265072814, "maximum": 0.12320105706503294, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9732236954934377, "standard_deviation": 0.0036837056546640792, "minimum": 0.9667992642540716, "maximum": 0.9788873627819497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7210795210585186, "standard_deviation": 0.051365020864213524, "minimum": 1.674149716020978, "maximum": 1.8203615464114442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7317064585711033, "standard_deviation": 0.3942495326382147, "minimum": 0.19204595788009168, "maximum": 1.296758846625359, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1862018123170368, "standard_deviation": 0.012909461588798906, "minimum": 0.16496509728306227, "maximum": 0.2065512647151612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905634106738123, "standard_deviation": 0.0030615548595372107, "minimum": 0.985684548768927, "maximum": 0.9950033547464215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0578753984071205, "standard_deviation": 0.05574349502462495, "minimum": 0.9709156178640134, "maximum": 1.1437186914393989, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.110934682905644, "standard_deviation": 0.5128444828911273, "minimum": 0.5151565653843814, "maximum": 1.6923746877518637, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1070173839769688, "standard_deviation": 0.018606428238928604, "minimum": 0.07735078983889414, "maximum": 0.13433564335110565, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9608752366031297, "standard_deviation": 0.008338693771344816, "minimum": 0.9491972413190808, "maximum": 0.9730133631729369, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0290106801618837, "standard_deviation": 0.046849619615955365, "minimum": 1.977254356453506, "maximum": 2.096784267964681, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.9089753683485813, "standard_deviation": 0.5193345327276813, "minimum": 1.3719606907494513, "maximum": 2.681759725430192, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22250168738614745, "standard_deviation": 0.02690193383487969, "minimum": 0.18345342372150006, "maximum": 0.25584381095794734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9894227481793264, "standard_deviation": 0.003537145624224902, "minimum": 0.9847926484894586, "maximum": 0.9940815985119785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3820555437654058, "standard_deviation": 0.07064708857227958, "minimum": 1.2818413547303964, "maximum": 1.4782812783596735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.386661148288554, "standard_deviation": 0.7185934915462048, "minimum": 0.7134655648487525, "maximum": 2.732092011371647, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12173466203449262, "standard_deviation": 0.025669618093336972, "minimum": 0.08958815480009147, "maximum": 0.15257294807088584, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3160799687085726, "standard_deviation": 0.11984898434397255, "minimum": 0.1345472658329405, "maximum": 0.5139403099210657, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.191471066671609, "standard_deviation": 0.013142418649677908, "minimum": 6.174669485875701, "maximum": 6.208500520633116, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3116961954327734, "standard_deviation": 0.03006972366710998, "minimum": 0.2772638659027342, "maximum": 0.35136765509599677, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.16054572766156464, "standard_deviation": 0.12944318728938112, "minimum": -0.327660739642443, "maximum": 0.01818770240976628, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9781837611075792, "standard_deviation": 0.002212195365488062, "minimum": 0.9746222235933484, "maximum": 0.981420629449119, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.212647921127329, "standard_deviation": 0.03804107507913915, "minimum": 1.1554431511153969, "maximum": 1.2581970811280148, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.602937205534163, "standard_deviation": 0.35099731420678293, "minimum": 0.198887664504338, "maximum": 1.2063311400751813, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14313605212442315, "standard_deviation": 0.007230272177494179, "minimum": 0.13172184547057783, "maximum": 0.15437431325057202, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9928862648638035, "standard_deviation": 0.0021166327211653724, "minimum": 0.9881217329730011, "maximum": 0.9948843773560755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6543532706053192, "standard_deviation": 0.04865806993878756, "minimum": 0.5777139744823454, "maximum": 0.7289907829586745, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9856491922762126, "standard_deviation": 0.5186153848390757, "minimum": 0.33268516535171955, "maximum": 1.776286499650684, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07859601253693986, "standard_deviation": 0.011011646436375384, "minimum": 0.06648687345505237, "maximum": 0.10177163132465032, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-75918-0013", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.5080924034118652, "predicted_mos": 1.5080924034118652, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "7509d6d828c18ebd188a2789ab30bf5aed91c74df61f3d50edeb6eba59f7c65b", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.691932999994606, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1829399904923, "全帯域_10ms_rms_envelope_nrmse": 0.12007481900199161, "全帯域_10ms_envelope_correlation": 0.9845640235870303, "全帯域_10ms_modulation_mae_db": 0.32715507776849206, "全帯域_40ms_log_rms_mae_db": 0.5537086419486448, "全帯域_40ms_rms_envelope_nrmse": 0.05862968542225614, "全帯域_40ms_envelope_correlation": 0.9958052685317653, "全帯域_40ms_modulation_mae_db": 0.35022530997551654, "300–1000 Hz_10ms_log_rms_mae_db": 2.03215941399629, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2183642186395641, "300–1000 Hz_10ms_envelope_correlation": 0.96343003233493, "300–1000 Hz_10ms_modulation_mae_db": 1.8174295960922542, "300–1000 Hz_40ms_log_rms_mae_db": 1.375420978376713, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13988930137588793, "300–1000 Hz_40ms_envelope_correlation": 0.9878009637738802, "300–1000 Hz_40ms_modulation_mae_db": 1.3206916902673063, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5324870160367374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17527714971961922, "1000–3000 Hz_10ms_envelope_correlation": 0.9724391490044428, "1000–3000 Hz_10ms_modulation_mae_db": 0.4961987159316384, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0152266018255531, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0889220328285109, "1000–3000 Hz_40ms_envelope_correlation": 0.9925264693903103, "1000–3000 Hz_40ms_modulation_mae_db": 0.8352356698605141, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7228366033341125, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16002152643915288, "3000–6000 Hz_10ms_envelope_correlation": 0.9807732872622739, "3000–6000 Hz_10ms_modulation_mae_db": 0.4496112019001386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0109646028276227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08097530594276384, "3000–6000 Hz_40ms_envelope_correlation": 0.9946475480379067, "3000–6000 Hz_40ms_modulation_mae_db": 0.4499074875568388, "mr_stft_db": 6.163683466630736, "band_mae_db": 0.2190901134342269, "rms_gain_error_db": -0.006094589558985739, "output_peak": 0.30861593406942595}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1946902817563887, "全帯域_10ms_rms_envelope_nrmse": 0.12637930879499137, "全帯域_10ms_envelope_correlation": 0.9829308531118196, "全帯域_10ms_modulation_mae_db": 0.27754867334313765, "全帯域_40ms_log_rms_mae_db": 0.6472324912877235, "全帯域_40ms_rms_envelope_nrmse": 0.05541068607744073, "全帯域_40ms_envelope_correlation": 0.9962756065961579, "全帯域_40ms_modulation_mae_db": 0.6153763023684342, "300–1000 Hz_10ms_log_rms_mae_db": 2.0013254088065735, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20796485823051897, "300–1000 Hz_10ms_envelope_correlation": 0.9662134674332064, "300–1000 Hz_10ms_modulation_mae_db": 1.6179035075984147, "300–1000 Hz_40ms_log_rms_mae_db": 1.261148757184981, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0970021708120747, "300–1000 Hz_40ms_envelope_correlation": 0.992290250536856, "300–1000 Hz_40ms_modulation_mae_db": 0.7868792291936814, "1000–3000 Hz_10ms_log_rms_mae_db": 1.573104355151817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20532263849629917, "1000–3000 Hz_10ms_envelope_correlation": 0.9625101478774614, "1000–3000 Hz_10ms_modulation_mae_db": 0.2841335388335451, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9497390980682202, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09910863781609518, "1000–3000 Hz_40ms_envelope_correlation": 0.9902804215525477, "1000–3000 Hz_40ms_modulation_mae_db": 0.2641953574410733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6503465895612204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15807794110861834, "3000–6000 Hz_10ms_envelope_correlation": 0.9806755450286057, "3000–6000 Hz_10ms_modulation_mae_db": 0.5833374477457429, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9917794169690373, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08088649800265899, "3000–6000 Hz_40ms_envelope_correlation": 0.9946181408464926, "3000–6000 Hz_40ms_modulation_mae_db": 0.75330375112901, "mr_stft_db": 6.151345202886141, "band_mae_db": 0.3783156877637112, "rms_gain_error_db": -0.03607531079633588, "output_peak": 0.334592933898205}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1745257561825981, "全帯域_10ms_rms_envelope_nrmse": 0.12270561164962818, "全帯域_10ms_envelope_correlation": 0.9838777002375595, "全帯域_10ms_modulation_mae_db": 0.4208013257141509, "全帯域_40ms_log_rms_mae_db": 0.6931362622561188, "全帯域_40ms_rms_envelope_nrmse": 0.06570841527115233, "全帯域_40ms_envelope_correlation": 0.9947632297872692, "全帯域_40ms_modulation_mae_db": 0.7453974194162125, "300–1000 Hz_10ms_log_rms_mae_db": 1.9088207834062034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1978398377037216, "300–1000 Hz_10ms_envelope_correlation": 0.970532950493991, "300–1000 Hz_10ms_modulation_mae_db": 1.8307841407468954, "300–1000 Hz_40ms_log_rms_mae_db": 1.2268011120516915, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0852758636227568, "300–1000 Hz_40ms_envelope_correlation": 0.9947378525532604, "300–1000 Hz_40ms_modulation_mae_db": 1.4271444244439848, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4655766861993123, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19261790382797891, "1000–3000 Hz_10ms_envelope_correlation": 0.9681521017109396, "1000–3000 Hz_10ms_modulation_mae_db": 0.15659451992139803, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8698203035462555, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08954064548475489, "1000–3000 Hz_40ms_envelope_correlation": 0.9927277454755103, "1000–3000 Hz_40ms_modulation_mae_db": 0.491951090880902, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6304883204315055, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1427289866474257, "3000–6000 Hz_10ms_envelope_correlation": 0.9844425225966004, "3000–6000 Hz_10ms_modulation_mae_db": 0.7560315596365066, "3000–6000 Hz_40ms_log_rms_mae_db": 1.006482448148485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0881687768402182, "3000–6000 Hz_40ms_envelope_correlation": 0.9937248478988191, "3000–6000 Hz_40ms_modulation_mae_db": 1.0921471385462522, "mr_stft_db": 6.140796807995852, "band_mae_db": 0.30754963635775723, "rms_gain_error_db": -0.00118617440827528, "output_peak": 0.3270971457834527}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.259641600689345, "全帯域_10ms_rms_envelope_nrmse": 0.13366881703596592, "全帯域_10ms_envelope_correlation": 0.9812399945779641, "全帯域_10ms_modulation_mae_db": 0.3354195479030843, "全帯域_40ms_log_rms_mae_db": 0.6615358354313807, "全帯域_40ms_rms_envelope_nrmse": 0.06783758146703608, "全帯域_40ms_envelope_correlation": 0.9945750490766799, "全帯域_40ms_modulation_mae_db": 0.20582643403330078, "300–1000 Hz_10ms_log_rms_mae_db": 1.9202357937917318, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21558997942183256, "300–1000 Hz_10ms_envelope_correlation": 0.964474419420557, "300–1000 Hz_10ms_modulation_mae_db": 1.63184637667006, "300–1000 Hz_40ms_log_rms_mae_db": 1.3208802775379243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11495168406310939, "300–1000 Hz_40ms_envelope_correlation": 0.9895328897177424, "300–1000 Hz_40ms_modulation_mae_db": 1.4064449444106335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5479300475661044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2120927482268072, "1000–3000 Hz_10ms_envelope_correlation": 0.9608564843951231, "1000–3000 Hz_10ms_modulation_mae_db": 0.6876530175510908, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9932436843509826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10558585077823014, "1000–3000 Hz_40ms_envelope_correlation": 0.9893816876219121, "1000–3000 Hz_40ms_modulation_mae_db": 0.31214142263130157, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6367961256341605, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.168128871232091, "3000–6000 Hz_10ms_envelope_correlation": 0.9781250723971683, "3000–6000 Hz_10ms_modulation_mae_db": 0.40119216643434674, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9691780515164073, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09158263945773194, "3000–6000 Hz_40ms_envelope_correlation": 0.9927596513265379, "3000–6000 Hz_40ms_modulation_mae_db": 0.17474308385412654, "mr_stft_db": 6.150031815106935, "band_mae_db": 0.41957163135913084, "rms_gain_error_db": 0.0714664945134075, "output_peak": 0.3355801351734133}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1913008203765156, "全帯域_10ms_rms_envelope_nrmse": 0.1274142876953616, "全帯域_10ms_envelope_correlation": 0.9827133828157357, "全帯域_10ms_modulation_mae_db": 0.6584718919360685, "全帯域_40ms_log_rms_mae_db": 0.6494868149383817, "全帯域_40ms_rms_envelope_nrmse": 0.06688432636806783, "全帯域_40ms_envelope_correlation": 0.9949365896683628, "全帯域_40ms_modulation_mae_db": 0.9729065955171098, "300–1000 Hz_10ms_log_rms_mae_db": 1.9060847399403125, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21014467864699762, "300–1000 Hz_10ms_envelope_correlation": 0.9643012755610798, "300–1000 Hz_10ms_modulation_mae_db": 1.7809671972688843, "300–1000 Hz_40ms_log_rms_mae_db": 1.2712412797432002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11644546781324325, "300–1000 Hz_40ms_envelope_correlation": 0.9890852850098241, "300–1000 Hz_40ms_modulation_mae_db": 1.4497082463275255, "1000–3000 Hz_10ms_log_rms_mae_db": 1.537340654497488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2088668301248909, "1000–3000 Hz_10ms_envelope_correlation": 0.9626004444081016, "1000–3000 Hz_10ms_modulation_mae_db": 0.3789338387278656, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9806455372647713, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11333639729970647, "1000–3000 Hz_40ms_envelope_correlation": 0.9881036675606237, "1000–3000 Hz_40ms_modulation_mae_db": 0.5567004885124774, "3000–6000 Hz_10ms_log_rms_mae_db": 1.679458048956132, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16123095885850824, "3000–6000 Hz_10ms_envelope_correlation": 0.9805841745319559, "3000–6000 Hz_10ms_modulation_mae_db": 0.7137443531905716, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9956426176350631, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08916170419870749, "3000–6000 Hz_40ms_envelope_correlation": 0.994604849510565, "3000–6000 Hz_40ms_modulation_mae_db": 1.0583269164551152, "mr_stft_db": 6.157385814258156, "band_mae_db": 0.3228845501568729, "rms_gain_error_db": -0.17711427687433784, "output_peak": 0.30397496045998074}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2010604145105686, "全帯域_10ms_rms_envelope_nrmse": 0.14529046278955082, "全帯域_10ms_envelope_correlation": 0.9774981491859455, "全帯域_10ms_modulation_mae_db": 0.8709011660031766, "全帯域_40ms_log_rms_mae_db": 0.6732595530402027, "全帯域_40ms_rms_envelope_nrmse": 0.07769222597489164, "全帯域_40ms_envelope_correlation": 0.9926765798983003, "全帯域_40ms_modulation_mae_db": 1.3855330033782136, "300–1000 Hz_10ms_log_rms_mae_db": 1.8990481718814964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1796969371118355, "300–1000 Hz_10ms_envelope_correlation": 0.9746094140764282, "300–1000 Hz_10ms_modulation_mae_db": 1.4124576205831225, "300–1000 Hz_40ms_log_rms_mae_db": 1.2546845698042708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09552883669578008, "300–1000 Hz_40ms_envelope_correlation": 0.9939284244917751, "300–1000 Hz_40ms_modulation_mae_db": 0.8017865805791192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5268969908133418, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1780276290336703, "1000–3000 Hz_10ms_envelope_correlation": 0.972027625375629, "1000–3000 Hz_10ms_modulation_mae_db": 0.22048607051862423, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9858458461558026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0923399715392631, "1000–3000 Hz_40ms_envelope_correlation": 0.9916790017240783, "1000–3000 Hz_40ms_modulation_mae_db": 0.37318401471823476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6634074199921414, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20064993337048068, "3000–6000 Hz_10ms_envelope_correlation": 0.9688527116444675, "3000–6000 Hz_10ms_modulation_mae_db": 0.9821685178115452, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9795353352098625, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11373170244746825, "3000–6000 Hz_40ms_envelope_correlation": 0.9887933490540515, "3000–6000 Hz_40ms_modulation_mae_db": 1.5657904766256205, "mr_stft_db": 6.135976903918594, "band_mae_db": 0.45605682255519114, "rms_gain_error_db": 0.0030869569903088628, "output_peak": 0.30552023113399346}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2746914227133816, "全帯域_10ms_rms_envelope_nrmse": 0.14097559953993122, "全帯域_10ms_envelope_correlation": 0.978661701949556, "全帯域_10ms_modulation_mae_db": 0.5567607742654319, "全帯域_40ms_log_rms_mae_db": 0.6727222200503034, "全帯域_40ms_rms_envelope_nrmse": 0.06848147076209789, "全帯域_40ms_envelope_correlation": 0.9943013535328888, "全帯域_40ms_modulation_mae_db": 1.177254658899691, "300–1000 Hz_10ms_log_rms_mae_db": 1.9435269675874673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1921385303010246, "300–1000 Hz_10ms_envelope_correlation": 0.9702667718849792, "300–1000 Hz_10ms_modulation_mae_db": 1.5666125274525282, "300–1000 Hz_40ms_log_rms_mae_db": 1.255667014511824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11026641332643763, "300–1000 Hz_40ms_envelope_correlation": 0.9895686709557151, "300–1000 Hz_40ms_modulation_mae_db": 1.126342551803132, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5117149861568713, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17189496636199794, "1000–3000 Hz_10ms_envelope_correlation": 0.9735175887579112, "1000–3000 Hz_10ms_modulation_mae_db": 0.10857930482039063, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8543680536517758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0766399402402154, "1000–3000 Hz_40ms_envelope_correlation": 0.9941752409056923, "1000–3000 Hz_40ms_modulation_mae_db": 0.27935035618457194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6708289995303125, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17393219398494017, "3000–6000 Hz_10ms_envelope_correlation": 0.9764927887105194, "3000–6000 Hz_10ms_modulation_mae_db": 0.6196846501658569, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9697325270601468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08383044834028554, "3000–6000 Hz_40ms_envelope_correlation": 0.9941158861347335, "3000–6000 Hz_40ms_modulation_mae_db": 1.0769665312226335, "mr_stft_db": 6.1545979165004505, "band_mae_db": 0.42282831975460183, "rms_gain_error_db": -0.014104033718952145, "output_peak": 0.3153050666195435}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.253396252781523, "全帯域_10ms_rms_envelope_nrmse": 0.14459434887725145, "全帯域_10ms_envelope_correlation": 0.9775070649641148, "全帯域_10ms_modulation_mae_db": 0.7711925251507692, "全帯域_40ms_log_rms_mae_db": 0.671293587105696, "全帯域_40ms_rms_envelope_nrmse": 0.06645127203902634, "全帯域_40ms_envelope_correlation": 0.9946996526140017, "全帯域_40ms_modulation_mae_db": 1.136504329196211, "300–1000 Hz_10ms_log_rms_mae_db": 1.9327483981874294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22261941041510366, "300–1000 Hz_10ms_envelope_correlation": 0.9598932847164763, "300–1000 Hz_10ms_modulation_mae_db": 1.827438561895562, "300–1000 Hz_40ms_log_rms_mae_db": 1.2349867454096581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13331753070934493, "300–1000 Hz_40ms_envelope_correlation": 0.9851350229873399, "300–1000 Hz_40ms_modulation_mae_db": 1.383943152126512, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5322413570401527, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20057005383994067, "1000–3000 Hz_10ms_envelope_correlation": 0.9640944235204439, "1000–3000 Hz_10ms_modulation_mae_db": 0.560277404536457, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9455110786755422, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0994654242109208, "1000–3000 Hz_40ms_envelope_correlation": 0.9902783891855318, "1000–3000 Hz_40ms_modulation_mae_db": 0.5986921016550938, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6733622201591056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18097248128474788, "3000–6000 Hz_10ms_envelope_correlation": 0.9745947557173332, "3000–6000 Hz_10ms_modulation_mae_db": 0.6644428077867788, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9444319981438171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08701267706383088, "3000–6000 Hz_40ms_envelope_correlation": 0.993868915904879, "3000–6000 Hz_40ms_modulation_mae_db": 1.034512024654247, "mr_stft_db": 6.148667617507454, "band_mae_db": 0.24367603128974433, "rms_gain_error_db": -0.09637089630944373, "output_peak": 0.37343688282340615}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 47040, "output_samples": 47040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.128240868647503, "spectral_convergence": 0.5680389793259327, "spectral_convergence_status": "ok", "amplitude_floor": 8.904239335448439e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.9314725214927275, "spectral_convergence": 0.5676674148268529, "spectral_convergence_status": "ok", "amplitude_floor": 4.6801685305432975e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.4313370097519735, "spectral_convergence": 0.5979963246985975, "spectral_convergence_status": "ok", "amplitude_floor": 2.078367899935203e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.736571532646871e-05, "output_power": 4.9489925753084e-05, "delta_db": 0.19052701771858874, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00015972486768670453, "output_power": 0.00014340715266805516, "delta_db": -0.46801724017413615, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 3.942775929744416e-05, "output_power": 4.2902771062907795e-05, "delta_db": 0.3668324733134037, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.533823001233142e-05, "output_power": 6.443808457595419e-05, "delta_db": -0.06024742905097184, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005191422513822901, "output_power": 0.0005202934107000104, "delta_db": 0.009619495695825164, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017487193556096138, "output_power": 0.0001839288347671524, "delta_db": 0.21929702465243606, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2190901134342269, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.006094589558985739, "pitch": null, "output_rms_dbfs": -29.98067275595085, "output_peak": 0.30861593406942595, "reference_rms_dbfs": -29.974578166391858, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.163683466630736}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 47040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001280716783512989, "log_rms_mae_db": 1.1829399904923, "rms_envelope_nrmse": 0.12007481900199161, "envelope_correlation": 0.9845640235870303, "rms_gain_error_db": -0.006047460401347822, "reference_mean_power": 0.0010075555569028208, "output_mean_power": 0.0010061535331008452}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 5.126669358054452e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020945501880404742, "output_power": 0.00020633545295697863, "reference_active": true, "delta_db": -0.0651691535720137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010240530560263733, "output_power": 9.983612316594974e-05, "reference_active": true, "delta_db": -0.11034749798523015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.779418868232472e-05, "output_power": 2.3086555604576272e-05, "reference_active": true, "delta_db": -0.8059485817482324}], "active_band_count": 3, "active_band_mae_db": 0.32715507776849206}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 9.465799671406843e-05, "log_rms_mae_db": 0.5537086419486448, "rms_envelope_nrmse": 0.05862968542225614, "envelope_correlation": 0.9958052685317653, "rms_gain_error_db": -0.006010937477308585, "reference_mean_power": 0.001017821671515535, "output_mean_power": 0.001016413910025085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 4.336511567866126e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018218362838522706, "output_power": 0.00017899974149104705, "reference_active": true, "delta_db": -0.07656943494207329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6513405903536415e-05, "output_power": 5.4624319341899865e-05, "reference_active": true, "delta_db": -0.14765443397584221}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.038210274571303e-06, "output_power": 1.685015035360263e-06, "reference_active": true, "delta_db": -0.8264520610086341}], "active_band_count": 3, "active_band_mae_db": 0.35022530997551654}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.812061629421404e-05, "log_rms_mae_db": 2.03215941399629, "rms_envelope_nrmse": 0.2183642186395641, "envelope_correlation": 0.96343003233493, "rms_gain_error_db": -0.676792310928871, "reference_mean_power": 0.00014923818668565985, "output_mean_power": 0.00012770289196486754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 7.990988889403065e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7339937941838396e-05, "output_power": 3.394250223543137e-05, "reference_active": true, "delta_db": -1.4448382899823349}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462900675807451e-06, "output_power": 5.263575726686172e-06, "reference_active": true, "delta_db": -0.8914660609988433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2961437099516358e-06, "output_power": 2.656143903344914e-06, "reference_active": true, "delta_db": 3.1159844372955847}], "active_band_count": 3, "active_band_mae_db": 1.8174295960922542}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.488253980944051e-05, "log_rms_mae_db": 1.375420978376713, "rms_envelope_nrmse": 0.13988930137588793, "envelope_correlation": 0.9878009637738802, "rms_gain_error_db": -0.6768038913563796, "reference_mean_power": 0.00015077911609921791, "output_mean_power": 0.00012902111891299975}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 7.46020344966024e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.50390789043777e-05, "output_power": 3.308669900180492e-05, "reference_active": true, "delta_db": -1.3393605959259123}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3028613860270352e-06, "output_power": 2.780864120465029e-06, "reference_active": true, "delta_db": -0.7471057893370554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.84375886973483e-08, "output_power": 1.3620647742029198e-07, "reference_active": true, "delta_db": 1.875608685538951}], "active_band_count": 3, "active_band_mae_db": 1.3206916902673063}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.5882621361725546e-05, "log_rms_mae_db": 1.5324870160367374, "rms_envelope_nrmse": 0.17527714971961922, "envelope_correlation": 0.9724391490044428, "rms_gain_error_db": -0.20764945260785228, "reference_mean_power": 0.0001648439551682589, "output_mean_power": 0.00015714771874991576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 1.0813916284336453e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.944970033970189e-05, "output_power": 4.374247386194861e-05, "reference_active": true, "delta_db": -0.5326032317597269}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1861966099880283e-05, "output_power": 9.804016315237423e-06, "reference_active": true, "delta_db": -0.8275265322931247}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.815426369634166e-06, "output_power": 4.95999667132483e-06, "reference_active": true, "delta_db": 0.1284663837420636}], "active_band_count": 3, "active_band_mae_db": 0.4961987159316384}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.018253092677299e-05, "log_rms_mae_db": 1.0152266018255531, "rms_envelope_nrmse": 0.0889220328285109, "envelope_correlation": 0.9925264693903103, "rms_gain_error_db": -0.20765163677771342, "reference_mean_power": 0.00016654362747028494, "output_mean_power": 0.00015876795688435084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 9.377377349368422e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3529785361500865e-05, "output_power": 3.8827302685613505e-05, "reference_active": true, "delta_db": -0.4964930436381854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.594238028592104e-06, "output_power": 4.6252019469299645e-06, "reference_active": true, "delta_db": -0.8261024139209099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1488780558631956e-07, "output_power": 2.397975199394059e-07, "reference_active": true, "delta_db": -1.183111552022447}], "active_band_count": 3, "active_band_mae_db": 0.8352356698605141}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001195123420429498, "log_rms_mae_db": 1.7228366033341125, "rms_envelope_nrmse": 0.16002152643915288, "envelope_correlation": 0.9807732872622739, "rms_gain_error_db": 0.13848221194019372, "reference_mean_power": 0.0005186016960204273, "output_mean_power": 0.0005354046687519474}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 3.8692936095348015e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017018269823121574, "output_power": 0.00017853695743848827, "reference_active": true, "delta_db": 0.20812724338332228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010605303684284744, "output_power": 9.856178237691441e-05, "reference_active": true, "delta_db": -0.3181456050151326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2588117400162252e-05, "output_power": 1.8690630872948858e-05, "reference_active": true, "delta_db": -0.8225607573019609}], "active_band_count": 3, "active_band_mae_db": 0.4496112019001386}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 8.500195655960054e-05, "log_rms_mae_db": 1.0109646028276227, "rms_envelope_nrmse": 0.08097530594276384, "envelope_correlation": 0.9946475480379067, "rms_gain_error_db": 0.13849555819429674, "reference_mean_power": 0.0005239212873916433, "output_mean_power": 0.0005408982799587123}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 3.3612492298744645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016844288389443048, "output_power": 0.000174457398257419, "reference_active": true, "delta_db": 0.152367230806483}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.87289448022482e-05, "output_power": 5.411659003990342e-05, "reference_active": true, "delta_db": -0.3552177453416001}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9018720843055598e-06, "output_power": 1.5666340275199686e-06, "reference_active": true, "delta_db": -0.8421374865224334}], "active_band_count": 3, "active_band_mae_db": 0.4499074875568388}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7829098201559233, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "98a4f98c1fadb6d9401aa83cc95d15f78b088bed888f3c5d132c2f5808a3965e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9670247456312566, "standard_deviation": 0.005132011326176897, "minimum": 0.9608564843951231, "maximum": 0.9735175887579112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5284115116827282, "standard_deviation": 0.03098090097053242, "minimum": 1.4655766861993123, "maximum": 1.573104355151817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.36160705135512616, "standard_deviation": 0.20569098527923282, "minimum": 0.10857930482039063, "maximum": 0.6876530175510908, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19308373995390055, "standard_deviation": 0.01607671841708417, "minimum": 0.17189496636199794, "maximum": 0.2120927482268072, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9911440779270259, "standard_deviation": 0.001990640085420709, "minimum": 0.9881036675606237, "maximum": 0.9941752409056923, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9493000254423629, "standard_deviation": 0.058479729232831894, "minimum": 0.8543680536517758, "maximum": 1.0152266018255531, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.46393131273552113, "standard_deviation": 0.19663543060548094, "minimum": 0.2641953574410733, "maximum": 0.8352356698605141, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09561736252471212, "standard_deviation": 0.011294558179427006, "minimum": 0.0766399402402154, "maximum": 0.11333639729970647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9780676072361155, "standard_deviation": 0.004789170145995584, "minimum": 0.9688527116444675, "maximum": 0.9844425225966004, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6659405409498362, "standard_deviation": 0.0289203541585305, "minimum": 1.6304883204315055, "maximum": 1.7228366033341125, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6462765880839358, "standard_deviation": 0.18250775619929316, "minimum": 0.40119216643434674, "maximum": 0.9821685178115452, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1682178616157456, "standard_deviation": 0.017385045176106468, "minimum": 0.1427289866474257, "maximum": 0.20064993337048068, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933916485892482, "standard_deviation": 0.0019625249117955663, "minimum": 0.9887933490540515, "maximum": 0.9946475480379067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9834683746888053, "standard_deviation": 0.02215262027043325, "minimum": 0.9444319981438171, "maximum": 1.0109646028276227, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9007121762554805, "standard_deviation": 0.43173605165459816, "minimum": 0.17474308385412654, "maximum": 1.5657904766256205, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08941871903670814, "standard_deviation": 0.010546089009963715, "minimum": 0.08088649800265899, "maximum": 0.11373170244746825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9667152019902059, "standard_deviation": 0.004750010776932011, "minimum": 0.9598932847164763, "maximum": 0.9746094140764282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.942993709699688, "standard_deviation": 0.04845845744571259, "minimum": 1.8990481718814964, "maximum": 2.03215941399629, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.6856799410384653, "standard_deviation": 0.15302079742988786, "minimum": 1.4124576205831225, "maximum": 1.8307841407468954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20554480630882482, "standard_deviation": 0.014594556407983848, "minimum": 0.1796969371118355, "maximum": 0.22261941041510366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9902599200032991, "standard_deviation": 0.003214459474482882, "minimum": 0.9851350229873399, "maximum": 0.9947378525532604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.275103841827533, "standard_deviation": 0.049433860257736925, "minimum": 1.2268011120516915, "maximum": 1.375420978376713, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.2128676023939868, "standard_deviation": 0.2773119432410463, "minimum": 0.7868792291936814, "maximum": 1.4497082463275255, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11158465855232934, "standard_deviation": 0.018794998158056575, "minimum": 0.0852758636227568, "maximum": 0.13988930137588793, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.34624659908390454, "standard_deviation": 0.08706382338861354, "minimum": 0.2190901134342269, "maximum": 0.45605682255519114, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.15031069310054, "standard_deviation": 0.008840555078724081, "minimum": 6.135976903918594, "maximum": 6.163683466630736, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3255154112451776, "standard_deviation": 0.023103219075446463, "minimum": 0.30397496045998074, "maximum": 0.37343688282340615, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03204897877032678, "standard_deviation": 0.07473344288273599, "minimum": -0.17711427687433784, "maximum": 0.0714664945134075, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9811241088037157, "standard_deviation": 0.002866378070381147, "minimum": 0.9774981491859455, "maximum": 0.9845640235870303, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2165308174378275, "standard_deviation": 0.03936263620834943, "minimum": 1.1745257561825981, "maximum": 1.2746914227133816, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5272813727605389, "standard_deviation": 0.22239383193054305, "minimum": 0.27754867334313765, "maximum": 0.8709011660031766, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.13263790692308403, "standard_deviation": 0.009975408934465626, "minimum": 0.12007481900199161, "maximum": 0.14529046278955082, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9947541662131782, "standard_deviation": 0.0010696739348818127, "minimum": 0.9926765798983003, "maximum": 0.9962756065961579, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6527969257573064, "standard_deviation": 0.04263114440543614, "minimum": 0.5537086419486448, "maximum": 0.6931362622561188, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8236280065980861, "standard_deviation": 0.41683200081004085, "minimum": 0.20582643403330078, "maximum": 1.3855330033782136, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06588695792274613, "standard_deviation": 0.006693985898948063, "minimum": 0.05541068607744073, "maximum": 0.07769222597489164, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_6930-76324-0024", "language": "en", "style": "normal", "speaker": "librispeech_6930", "source_utterance_id": "6930-76324-0024", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 64800, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_6930-76324-0024.wav", "source_file_sha256": "ce247c6b9acfed89fe4f762aa8c9b38612ad5ce6dc2651004c20ebc8a73b73ff", "decoded_pcm_sha256": "d8dd1ce759e26c72605c90e837ae65e87caae1e3095da47a3f97063f55664b19", "original_source_sha256": "5d8bd7d094fe646881a4848026dbf1a3748ec90f170b87d2f0836adcfd9b7f7f", "corpus_entry_id": "librispeech_6930-76324-0024"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.9916303157806396, "predicted_mos": 3.9916303157806396, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "04931a10cacc15e36e68c2278ab832d51edf639fd1cddad7eb239ab5af58a2b6", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9135296450003807, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_6930-76324-0024", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "04931a10cacc15e36e68c2278ab832d51edf639fd1cddad7eb239ab5af58a2b6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "22e1e852f27049dc0fb233b38864a80ef568f82169bbc07c6fb7e15c8d9f1003"}, "rows": [{"case": "librispeech_6930-76324-0024", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.855567216873169, "predicted_mos": 1.855567216873169, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "96a64d2e2ec53b1ee2851c1b7282af525645c05aae1295d0436ebf4f88bf31e6", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1338270060005016, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_6930-76324-0024", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "96a64d2e2ec53b1ee2851c1b7282af525645c05aae1295d0436ebf4f88bf31e6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2209794156661782, "全帯域_10ms_rms_envelope_nrmse": 0.15536834987904388, "全帯域_10ms_envelope_correlation": 0.9733668690746303, "全帯域_10ms_modulation_mae_db": 0.5165447580069441, "全帯域_40ms_log_rms_mae_db": 0.6257251944574986, "全帯域_40ms_rms_envelope_nrmse": 0.09182144707742174, "全帯域_40ms_envelope_correlation": 0.9899288768640606, "全帯域_40ms_modulation_mae_db": 1.181387321009663, "300–1000 Hz_10ms_log_rms_mae_db": 1.8699948824132673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17843516525926065, "300–1000 Hz_10ms_envelope_correlation": 0.9700249118418872, "300–1000 Hz_10ms_modulation_mae_db": 0.6246991608451999, "300–1000 Hz_40ms_log_rms_mae_db": 1.1907512529703403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09936432351353894, "300–1000 Hz_40ms_envelope_correlation": 0.9905968790115872, "300–1000 Hz_40ms_modulation_mae_db": 0.691712967005579, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7405949662317135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19953199337395833, "1000–3000 Hz_10ms_envelope_correlation": 0.9670638845296743, "1000–3000 Hz_10ms_modulation_mae_db": 0.843361074961023, "1000–3000 Hz_40ms_log_rms_mae_db": 1.073707274452048, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11755006271950087, "1000–3000 Hz_40ms_envelope_correlation": 0.9884416104338507, "1000–3000 Hz_40ms_modulation_mae_db": 0.6949337502417249, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5164777112746501, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14537806804888206, "3000–6000 Hz_10ms_envelope_correlation": 0.9842330794964084, "3000–6000 Hz_10ms_modulation_mae_db": 0.22257671867800202, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8312879939047189, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06573854220344093, "3000–6000 Hz_40ms_envelope_correlation": 0.996644172936936, "3000–6000 Hz_40ms_modulation_mae_db": 0.42646989717879485, "mr_stft_db": 6.1176185654550865, "band_mae_db": 0.5679905502772128, "rms_gain_error_db": -0.1324789194490673, "output_peak": 0.4548762932183803}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2146099007887554, "全帯域_10ms_rms_envelope_nrmse": 0.15909920856431306, "全帯域_10ms_envelope_correlation": 0.9719861806631062, "全帯域_10ms_modulation_mae_db": 0.721012940436336, "全帯域_40ms_log_rms_mae_db": 0.6040283425189674, "全帯域_40ms_rms_envelope_nrmse": 0.09469884061402624, "全帯域_40ms_envelope_correlation": 0.9893253434711303, "全帯域_40ms_modulation_mae_db": 1.0508605332528806, "300–1000 Hz_10ms_log_rms_mae_db": 1.8579117363833164, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18975950256076318, "300–1000 Hz_10ms_envelope_correlation": 0.9658064073923633, "300–1000 Hz_10ms_modulation_mae_db": 0.9847306008868966, "300–1000 Hz_40ms_log_rms_mae_db": 1.259206926326005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10876983735978174, "300–1000 Hz_40ms_envelope_correlation": 0.9882766134051609, "300–1000 Hz_40ms_modulation_mae_db": 0.47646430734980294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7288066103013824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20569949382262154, "1000–3000 Hz_10ms_envelope_correlation": 0.9629264301008531, "1000–3000 Hz_10ms_modulation_mae_db": 0.5976541430311645, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0610990719017253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13097144191751744, "1000–3000 Hz_40ms_envelope_correlation": 0.983629093880765, "1000–3000 Hz_40ms_modulation_mae_db": 0.41163885553174956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5592967240951674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15540671873179193, "3000–6000 Hz_10ms_envelope_correlation": 0.981691627381798, "3000–6000 Hz_10ms_modulation_mae_db": 0.7455609733403072, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8359044309497002, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06306915390040425, "3000–6000 Hz_40ms_envelope_correlation": 0.9967962790797184, "3000–6000 Hz_40ms_modulation_mae_db": 0.9148737268350805, "mr_stft_db": 6.122034501503662, "band_mae_db": 0.3765359206027474, "rms_gain_error_db": -0.17427835079708176, "output_peak": 0.4628948438859862}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1507470389982613, "全帯域_10ms_rms_envelope_nrmse": 0.1591569729029106, "全帯域_10ms_envelope_correlation": 0.9723235481093803, "全帯域_10ms_modulation_mae_db": 0.983164731324719, "全帯域_40ms_log_rms_mae_db": 0.707258539831495, "全帯域_40ms_rms_envelope_nrmse": 0.09941391353094765, "全帯域_40ms_envelope_correlation": 0.989392126144371, "全帯域_40ms_modulation_mae_db": 1.3918884498346282, "300–1000 Hz_10ms_log_rms_mae_db": 1.7816510855112588, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16814740921373525, "300–1000 Hz_10ms_envelope_correlation": 0.9732846402209508, "300–1000 Hz_10ms_modulation_mae_db": 1.235489260589482, "300–1000 Hz_40ms_log_rms_mae_db": 1.2074879082774388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09434420348161095, "300–1000 Hz_40ms_envelope_correlation": 0.990837197444671, "300–1000 Hz_40ms_modulation_mae_db": 0.637361820256987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7148076083130301, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18438754984194522, "1000–3000 Hz_10ms_envelope_correlation": 0.9690616430736361, "1000–3000 Hz_10ms_modulation_mae_db": 0.1457167615598188, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9628942501963371, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08618088658824724, "1000–3000 Hz_40ms_envelope_correlation": 0.9926358809895591, "1000–3000 Hz_40ms_modulation_mae_db": 0.5175410953362766, "3000–6000 Hz_10ms_log_rms_mae_db": 1.533325298803003, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1576311858401375, "3000–6000 Hz_10ms_envelope_correlation": 0.9813251407514574, "3000–6000 Hz_10ms_modulation_mae_db": 0.6728370937913374, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8771964323271655, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08023876027457574, "3000–6000 Hz_40ms_envelope_correlation": 0.9955003037338693, "3000–6000 Hz_40ms_modulation_mae_db": 0.6796767906436019, "mr_stft_db": 6.120835684959297, "band_mae_db": 0.3701550429790292, "rms_gain_error_db": -0.28699140290321606, "output_peak": 0.4067670696033654}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1740854017533713, "全帯域_10ms_rms_envelope_nrmse": 0.14722122428719864, "全帯域_10ms_envelope_correlation": 0.9764566317274828, "全帯域_10ms_modulation_mae_db": 0.30708589854641577, "全帯域_40ms_log_rms_mae_db": 0.6451950643135306, "全帯域_40ms_rms_envelope_nrmse": 0.07606789725868227, "全帯域_40ms_envelope_correlation": 0.9937250267793952, "全帯域_40ms_modulation_mae_db": 0.5600773820227941, "300–1000 Hz_10ms_log_rms_mae_db": 1.849776582475899, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17451464130043462, "300–1000 Hz_10ms_envelope_correlation": 0.9713327446171623, "300–1000 Hz_10ms_modulation_mae_db": 0.7496663276608274, "300–1000 Hz_40ms_log_rms_mae_db": 1.2874230908759112, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10419895094754256, "300–1000 Hz_40ms_envelope_correlation": 0.9894209450023641, "300–1000 Hz_40ms_modulation_mae_db": 1.4151000256245727, "1000–3000 Hz_10ms_log_rms_mae_db": 1.735155802162306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19515567335584091, "1000–3000 Hz_10ms_envelope_correlation": 0.965906277543792, "1000–3000 Hz_10ms_modulation_mae_db": 1.1324644955318603, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9895520955647494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10240306163847854, "1000–3000 Hz_40ms_envelope_correlation": 0.9896549748264158, "1000–3000 Hz_40ms_modulation_mae_db": 1.0495721605211097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5454517839426554, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14043644916423248, "3000–6000 Hz_10ms_envelope_correlation": 0.9850812325421932, "3000–6000 Hz_10ms_modulation_mae_db": 0.5173436310540388, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8564979701363857, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06911974252373475, "3000–6000 Hz_40ms_envelope_correlation": 0.9961917556884119, "3000–6000 Hz_40ms_modulation_mae_db": 0.3876152540967807, "mr_stft_db": 6.103294790114281, "band_mae_db": 0.48350285711806246, "rms_gain_error_db": -0.22486540352802734, "output_peak": 0.43924728023414605}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1675816390867049, "全帯域_10ms_rms_envelope_nrmse": 0.14716649852594854, "全帯域_10ms_envelope_correlation": 0.9763302044451807, "全帯域_10ms_modulation_mae_db": 0.7703214186477775, "全帯域_40ms_log_rms_mae_db": 0.594066087378139, "全帯域_40ms_rms_envelope_nrmse": 0.08124311013724698, "全帯域_40ms_envelope_correlation": 0.9930986642910545, "全帯域_40ms_modulation_mae_db": 0.4728682755495339, "300–1000 Hz_10ms_log_rms_mae_db": 1.868356767108995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1837007407348021, "300–1000 Hz_10ms_envelope_correlation": 0.967984868656682, "300–1000 Hz_10ms_modulation_mae_db": 0.925360521156137, "300–1000 Hz_40ms_log_rms_mae_db": 1.1280929374658868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08990873145900932, "300–1000 Hz_40ms_envelope_correlation": 0.9917674386702484, "300–1000 Hz_40ms_modulation_mae_db": 0.5598389003972216, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7239686958741482, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17691082441375075, "1000–3000 Hz_10ms_envelope_correlation": 0.971500904927175, "1000–3000 Hz_10ms_modulation_mae_db": 0.46585001236987816, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9429774425030212, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0886768408385803, "1000–3000 Hz_40ms_envelope_correlation": 0.9924501315847929, "1000–3000 Hz_40ms_modulation_mae_db": 0.8870655475419841, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5340026780686868, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16348367006752848, "3000–6000 Hz_10ms_envelope_correlation": 0.9810660172316463, "3000–6000 Hz_10ms_modulation_mae_db": 1.1245464746130345, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8456675355671814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09217765585493538, "3000–6000 Hz_40ms_envelope_correlation": 0.9955199840769873, "3000–6000 Hz_40ms_modulation_mae_db": 0.6704381154539693, "mr_stft_db": 6.1252797937077075, "band_mae_db": 0.2764922007003826, "rms_gain_error_db": -0.25077732730357466, "output_peak": 0.45213443522733626}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1550635905884288, "全帯域_10ms_rms_envelope_nrmse": 0.16350762945534938, "全帯域_10ms_envelope_correlation": 0.9702717921467477, "全帯域_10ms_modulation_mae_db": 1.266882217929938, "全帯域_40ms_log_rms_mae_db": 0.601464717010265, "全帯域_40ms_rms_envelope_nrmse": 0.07711664527841329, "全帯域_40ms_envelope_correlation": 0.9928151918670775, "全帯域_40ms_modulation_mae_db": 0.9147098185896461, "300–1000 Hz_10ms_log_rms_mae_db": 1.9057071098950058, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16455348778662898, "300–1000 Hz_10ms_envelope_correlation": 0.9745257850625249, "300–1000 Hz_10ms_modulation_mae_db": 0.45630376685737534, "300–1000 Hz_40ms_log_rms_mae_db": 1.3180728283038434, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.095594789521526, "300–1000 Hz_40ms_envelope_correlation": 0.991185191234667, "300–1000 Hz_40ms_modulation_mae_db": 0.09050206863399361, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7463883787421828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20323735287021452, "1000–3000 Hz_10ms_envelope_correlation": 0.9632440578351749, "1000–3000 Hz_10ms_modulation_mae_db": 0.7197236447011207, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9829877649837626, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10507465245667252, "1000–3000 Hz_40ms_envelope_correlation": 0.9892261497242567, "1000–3000 Hz_40ms_modulation_mae_db": 0.5082820471749866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.584849735759381, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1564750215595404, "3000–6000 Hz_10ms_envelope_correlation": 0.9817564300667823, "3000–6000 Hz_10ms_modulation_mae_db": 0.6372292394270899, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8702857853591995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08053126702555073, "3000–6000 Hz_40ms_envelope_correlation": 0.9956945913819785, "3000–6000 Hz_40ms_modulation_mae_db": 0.6009977044768543, "mr_stft_db": 6.128288982314339, "band_mae_db": 0.27859360203168126, "rms_gain_error_db": -0.09266370867017323, "output_peak": 0.44900898005887524}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2345432052860572, "全帯域_10ms_rms_envelope_nrmse": 0.15473768935009294, "全帯域_10ms_envelope_correlation": 0.9735283196540777, "全帯域_10ms_modulation_mae_db": 0.25065819342927836, "全帯域_40ms_log_rms_mae_db": 0.60925403439842, "全帯域_40ms_rms_envelope_nrmse": 0.08403471902522368, "全帯域_40ms_envelope_correlation": 0.9914865257772285, "全帯域_40ms_modulation_mae_db": 0.6278643726316707, "300–1000 Hz_10ms_log_rms_mae_db": 1.840529875577204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17072891412553515, "300–1000 Hz_10ms_envelope_correlation": 0.9725762915170015, "300–1000 Hz_10ms_modulation_mae_db": 0.5969065776256399, "300–1000 Hz_40ms_log_rms_mae_db": 1.1772684712590964, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0898968026248077, "300–1000 Hz_40ms_envelope_correlation": 0.9923929688942751, "300–1000 Hz_40ms_modulation_mae_db": 1.0133761990121686, "1000–3000 Hz_10ms_log_rms_mae_db": 1.749045183700209, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19707754624088708, "1000–3000 Hz_10ms_envelope_correlation": 0.9646897466945304, "1000–3000 Hz_10ms_modulation_mae_db": 0.6643441140625445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9858999678467231, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08963469391482069, "1000–3000 Hz_40ms_envelope_correlation": 0.9920689747903504, "1000–3000 Hz_40ms_modulation_mae_db": 0.6450180454392472, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5375308202889562, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16334128879480822, "3000–6000 Hz_10ms_envelope_correlation": 0.9800134976661826, "3000–6000 Hz_10ms_modulation_mae_db": 0.3421331990207636, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8497840719183708, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07219246438558635, "3000–6000 Hz_40ms_envelope_correlation": 0.9958973100289052, "3000–6000 Hz_40ms_modulation_mae_db": 0.6018179222476906, "mr_stft_db": 6.132769207518812, "band_mae_db": 0.5112610229125174, "rms_gain_error_db": -0.10418767691442246, "output_peak": 0.539135609958476}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1710423199162936, "全帯域_10ms_rms_envelope_nrmse": 0.19357266302710283, "全帯域_10ms_envelope_correlation": 0.9600998252454167, "全帯域_10ms_modulation_mae_db": 0.5661578164844364, "全帯域_40ms_log_rms_mae_db": 0.6317292265086848, "全帯域_40ms_rms_envelope_nrmse": 0.09000493700722483, "全帯域_40ms_envelope_correlation": 0.990724795929412, "全帯域_40ms_modulation_mae_db": 0.3347480726599537, "300–1000 Hz_10ms_log_rms_mae_db": 1.933142861189325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19577995671339785, "300–1000 Hz_10ms_envelope_correlation": 0.9636376574298823, "300–1000 Hz_10ms_modulation_mae_db": 0.8503108251182362, "300–1000 Hz_40ms_log_rms_mae_db": 1.3040696189119974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10609244629169644, "300–1000 Hz_40ms_envelope_correlation": 0.9884200710537103, "300–1000 Hz_40ms_modulation_mae_db": 0.6033700911380526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7314113241239708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19807066434808313, "1000–3000 Hz_10ms_envelope_correlation": 0.9646115504652937, "1000–3000 Hz_10ms_modulation_mae_db": 0.8514407757934978, "1000–3000 Hz_40ms_log_rms_mae_db": 1.036078765005137, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10925231861921877, "1000–3000 Hz_40ms_envelope_correlation": 0.9881503450245261, "1000–3000 Hz_40ms_modulation_mae_db": 0.6539723175746388, "3000–6000 Hz_10ms_log_rms_mae_db": 1.621174883147234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2281511194972015, "3000–6000 Hz_10ms_envelope_correlation": 0.9621645513654679, "3000–6000 Hz_10ms_modulation_mae_db": 0.7349174887023374, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9439825471579181, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10967824148710692, "3000–6000 Hz_40ms_envelope_correlation": 0.9911384367314863, "3000–6000 Hz_40ms_modulation_mae_db": 0.7261653347779337, "mr_stft_db": 6.114667846591064, "band_mae_db": 0.35961752649407375, "rms_gain_error_db": 0.08775412687628271, "output_peak": 0.5887349657319398}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64800, "output_samples": 64800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.124511323467292, "spectral_convergence": 0.6181545361195239, "spectral_convergence_status": "ok", "amplitude_floor": 1.7331847370499167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.805405715447264, "spectral_convergence": 0.5726136277835685, "spectral_convergence_status": "ok", "amplitude_floor": 1.0846095570828337e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.422938657450704, "spectral_convergence": 0.6048960038351461, "spectral_convergence_status": "ok", "amplitude_floor": 4.5533844599041527e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024118742205032635, "output_power": 0.00016917965751833745, "delta_db": -1.540065141020552, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 9.191122965591724e-05, "output_power": 0.00010505010598608813, "delta_db": 0.5802791845311046, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 7.435432994216677e-05, "output_power": 6.502081343657657e-05, "delta_db": -0.5825386586013965, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.4154339925686723e-05, "output_power": 2.5773452006237905e-05, "delta_db": 0.2817741647072917, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006210431709364446, "output_power": 0.0006285654341787459, "delta_db": 0.052287041964374364, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023011417890304625, "output_power": 0.00025063590741430654, "delta_db": 0.3709991108385577, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5679905502772128, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1324789194490673, "pitch": null, "output_rms_dbfs": -29.051008978514155, "output_peak": 0.4548762932183803, "reference_rms_dbfs": -28.918530059065088, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.1176185654550865}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.0001629307662014554, "log_rms_mae_db": 1.2209794156661782, "rms_envelope_nrmse": 0.15536834987904388, "envelope_correlation": 0.9733668690746303, "rms_gain_error_db": -0.13203041901568455, "reference_mean_power": 0.0012840985730840112, "output_mean_power": 0.0012456479610490339}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 7.3397819245367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003947037290166488, "output_power": 0.00037703649467982914, "reference_active": true, "delta_db": -0.19887840349651192}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980542214464163e-05, "output_power": 5.6737387167298265e-05, "reference_active": true, "delta_db": -0.9001982573246472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6036857037353872e-05, "output_power": 2.347106505365935e-05, "reference_active": true, "delta_db": -0.4505576131996733}], "active_band_count": 3, "active_band_mae_db": 0.5165447580069441}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.000144561944924798, "log_rms_mae_db": 0.6257251944574986, "rms_envelope_nrmse": 0.09182144707742174, "envelope_correlation": 0.9899288768640606, "rms_gain_error_db": -0.13193990682761758, "reference_mean_power": 0.0012928029739092497, "output_mean_power": 0.001254117857299578}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 6.614408925582191e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037454976712341086, "output_power": 0.0003507186225417459, "reference_active": true, "delta_db": -0.2855070461762885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9110632304140427e-05, "output_power": 2.3188984881283677e-05, "reference_active": true, "delta_db": -0.9876990131211452}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1315453531682465e-06, "output_power": 1.2635690719439924e-06, "reference_active": true, "delta_db": -2.270955903731555}], "active_band_count": 3, "active_band_mae_db": 1.181387321009663}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 3.733123977436472e-05, "log_rms_mae_db": 1.8699948824132673, "rms_envelope_nrmse": 0.17843516525926065, "envelope_correlation": 0.9700249118418872, "rms_gain_error_db": -0.2658024073072721, "reference_mean_power": 0.00013419941652038683, "output_mean_power": 0.00012623226848108545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.2377163404532544e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0020888510118017e-05, "output_power": 1.729390291650998e-05, "reference_active": true, "delta_db": -0.6359033056343809}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.868636044401026e-06, "output_power": 6.412904005203282e-06, "reference_active": true, "delta_db": -0.8884471892444519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3701828329102724e-06, "output_power": 1.4850915283749071e-06, "reference_active": true, "delta_db": 0.34974698765676726}], "active_band_count": 3, "active_band_mae_db": 0.6246991608451999}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.100732475570721e-05, "log_rms_mae_db": 1.1907512529703403, "rms_envelope_nrmse": 0.09936432351353894, "envelope_correlation": 0.9905968790115872, "rms_gain_error_db": -0.2658027850482604, "reference_mean_power": 0.00013520032575819994, "output_mean_power": 0.00012717374469972337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 4.6212903051863556e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.815754516308902e-05, "output_power": 1.5475569234453248e-05, "reference_active": true, "delta_db": -0.6941050050710973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8683197262456875e-06, "output_power": 2.9750534097987775e-06, "reference_active": true, "delta_db": -1.1402759558497666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2646920446881671e-07, "output_power": 1.1964897781687864e-07, "reference_active": true, "delta_db": -0.2407579400958733}], "active_band_count": 3, "active_band_mae_db": 0.691712967005579}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 4.662083951365485e-05, "log_rms_mae_db": 1.7405949662317135, "rms_envelope_nrmse": 0.19953199337395833, "envelope_correlation": 0.9670638845296743, "rms_gain_error_db": 0.33693811727130457, "reference_mean_power": 0.00017506320579060826, "output_mean_power": 0.0001891858666546159}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 9.028764248682407e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2773698500231605e-05, "output_power": 4.922792345161917e-05, "reference_active": true, "delta_db": 0.6103471271934486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.603092418254862e-06, "output_power": 7.69501490164451e-06, "reference_active": true, "delta_db": 0.6646208940213532}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4970630790605404e-06, "output_power": 4.668904453512709e-06, "reference_active": true, "delta_db": 1.255115203668267}], "active_band_count": 3, "active_band_mae_db": 0.843361074961023}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.913562930844258e-05, "log_rms_mae_db": 1.073707274452048, "rms_envelope_nrmse": 0.11755006271950087, "envelope_correlation": 0.9884416104338507, "rms_gain_error_db": 0.3370653648410076, "reference_mean_power": 0.00017624479575510055, "output_mean_power": 0.00019046835819759385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 7.899096535479963e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7672515224694995e-05, "output_power": 4.471318159233419e-05, "reference_active": true, "delta_db": 0.7441095637904707}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.050253319949336e-06, "output_power": 3.8066698726605424e-06, "reference_active": true, "delta_db": 0.9620930646472328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55667878122912e-07, "output_power": 2.7895620704349536e-07, "reference_active": true, "delta_db": 0.3785986222874713}], "active_band_count": 3, "active_band_mae_db": 0.6949337502417249}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.00015276264453937597, "log_rms_mae_db": 1.5164777112746501, "rms_envelope_nrmse": 0.14537806804888206, "envelope_correlation": 0.9842330794964084, "rms_gain_error_db": 0.07597647006132588, "reference_mean_power": 0.0006070254152411961, "output_mean_power": 0.0006177382902282253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.389907119886337e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002863305566547835, "output_power": 0.00029595291356181825, "reference_active": true, "delta_db": 0.14354922239750284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.608243104449444e-05, "output_power": 3.5908758779124374e-05, "reference_active": true, "delta_db": -0.020953967565334332}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0810295137794261e-05, "output_power": 1.213836653921096e-05, "reference_active": true, "delta_db": 0.5032269660711689}], "active_band_count": 3, "active_band_mae_db": 0.22257671867800202}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.00013215266595876854, "log_rms_mae_db": 0.8312879939047189, "rms_envelope_nrmse": 0.06573854220344093, "envelope_correlation": 0.996644172936936, "rms_gain_error_db": 0.07670200989702343, "reference_mean_power": 0.0006109513084873906, "output_mean_power": 0.0006218373447633278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 5.083168365591421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002672957849080471, "output_power": 0.00027914387855296804, "reference_active": true, "delta_db": 0.18835998239464408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4768224147952787e-05, "output_power": 1.4781007688025333e-05, "reference_active": true, "delta_db": 0.003757675522578492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.446503663842908e-07, "output_power": 6.575787516846912e-07, "reference_active": true, "delta_db": -1.087292033619162}], "active_band_count": 3, "active_band_mae_db": 0.42646989717879485}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7611276702488967, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d46dea3b82b0b33af9d15fb06601a63dc105b3d6d4b7d3148b542876f6e10240", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9661255618962661, "standard_deviation": 0.002960085089916245, "minimum": 0.9629264301008531, "maximum": 0.971500904927175, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7337723211811178, "standard_deviation": 0.011507089355177877, "minimum": 1.7148076083130301, "maximum": 1.749045183700209, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6775693777513635, "standard_deviation": 0.292935475023509, "minimum": 0.1457167615598188, "maximum": 1.1324644955318603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1950088872834127, "standard_deviation": 0.009684179044065214, "minimum": 0.17691082441375075, "maximum": 0.20569949382262154, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9895321451568145, "standard_deviation": 0.0029913374407408632, "minimum": 0.983629093880765, "maximum": 0.9926358809895591, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.004399579056688, "standard_deviation": 0.047114572052356404, "minimum": 0.9429774425030212, "maximum": 1.073707274452048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6710029774202146, "standard_deviation": 0.21004557351392678, "minimum": 0.41163885553174956, "maximum": 1.0495721605211097, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10371799483662955, "standard_deviation": 0.01558571063879624, "minimum": 0.08618088658824724, "maximum": 0.13097144191751744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9796664470627421, "standard_deviation": 0.0072682310324240175, "minimum": 0.9621645513654679, "maximum": 0.9850812325421932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5540137044224667, "standard_deviation": 0.03395751355787916, "minimum": 1.5164777112746501, "maximum": 1.621174883147234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6246431023283638, "standard_deviation": 0.275719428810807, "minimum": 0.22257671867800202, "maximum": 1.1245464746130345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16378794021301532, "standard_deviation": 0.02722592181264507, "minimum": 0.14043644916423248, "maximum": 0.2281511194972015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9954228542072866, "standard_deviation": 0.0017984016702839117, "minimum": 0.9911384367314863, "maximum": 0.9967962790797184, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.86382584591508, "standard_deviation": 0.035985137398194714, "minimum": 0.8312879939047189, "maximum": 0.9439825471579181, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6260068432138381, "standard_deviation": 0.1674115991066217, "minimum": 0.3876152540967807, "maximum": 0.9148737268350805, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07909322845691688, "standard_deviation": 0.01554824306661625, "minimum": 0.06306915390040425, "maximum": 0.10967824148710692, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9698966633423067, "standard_deviation": 0.0038121356436455604, "minimum": 0.9636376574298823, "maximum": 0.9745257850625249, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8633838625692838, "standard_deviation": 0.04495023330495584, "minimum": 1.7816510855112588, "maximum": 1.933142861189325, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8029333800924743, "standard_deviation": 0.24916769863429392, "minimum": 0.45630376685737534, "maximum": 1.235489260589482, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17820247721181973, "standard_deviation": 0.010886894150629984, "minimum": 0.16455348778662898, "maximum": 0.19577995671339785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9903621630895855, "standard_deviation": 0.0015148892771346687, "minimum": 0.9882766134051609, "maximum": 0.9923929688942751, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.234046629298815, "standard_deviation": 0.0681128346492726, "minimum": 1.1280929374658868, "maximum": 1.3180728283038434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6859657974272972, "standard_deviation": 0.38958359489051286, "minimum": 0.09050206863399361, "maximum": 1.4151000256245727, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09852126064993921, "standard_deviation": 0.00726903485292974, "minimum": 0.0898968026248077, "maximum": 0.10876983735978174, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.40301859038946336, "standard_deviation": 0.1072815993068326, "minimum": 0.2764922007003826, "maximum": 0.5679905502772128, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.120598671520531, "standard_deviation": 0.009061484763701442, "minimum": 6.103294790114281, "maximum": 6.132769207518812, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47409993473981316, "standard_deviation": 0.059433508183885965, "minimum": 0.4067670696033654, "maximum": 0.5887349657319398, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14731108283616, "standard_deviation": 0.1178857626729633, "minimum": -0.28699140290321606, "maximum": 0.08775412687628271, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9717954213832527, "standard_deviation": 0.005171173472337758, "minimum": 0.9600998252454167, "maximum": 0.9764566317274828, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1860815640105065, "standard_deviation": 0.03230045123899632, "minimum": 1.1507470389982613, "maximum": 1.2345432052860572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6727284968507307, "standard_deviation": 0.3399230137087514, "minimum": 0.25065819342927836, "maximum": 1.266882217929938, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15997877949899497, "standard_deviation": 0.014726485612997728, "minimum": 0.14716649852594854, "maximum": 0.19357266302710283, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9913120688904662, "standard_deviation": 0.001740093205226168, "minimum": 0.9893253434711303, "maximum": 0.9937250267793952, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.627340150802125, "standard_deviation": 0.036595995670412385, "minimum": 0.594066087378139, "maximum": 0.707258539831495, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8168005281938463, "standard_deviation": 0.3742206322441933, "minimum": 0.3347480726599537, "maximum": 1.3918884498346282, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08680018874114834, "standard_deviation": 0.008489154770555849, "minimum": 0.07606789725868227, "maximum": 0.09941391353094765, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-76324-0024", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.9076402187347412, "predicted_mos": 1.9076402187347412, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "072d325b7e92c5ff219f65a406b942ea3ae2a0de2b4d684a8e7d5c74a1b70f07", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8295684000477195, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2263940903873503, "全帯域_10ms_rms_envelope_nrmse": 0.16694479222940575, "全帯域_10ms_envelope_correlation": 0.9691487318333699, "全帯域_10ms_modulation_mae_db": 0.5099150462578902, "全帯域_40ms_log_rms_mae_db": 0.6396750311663291, "全帯域_40ms_rms_envelope_nrmse": 0.10695403818740934, "全帯域_40ms_envelope_correlation": 0.9862354967124499, "全帯域_40ms_modulation_mae_db": 1.1549744084907527, "300–1000 Hz_10ms_log_rms_mae_db": 1.784932989105563, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18161541373515538, "300–1000 Hz_10ms_envelope_correlation": 0.9688266121766055, "300–1000 Hz_10ms_modulation_mae_db": 0.43989740281630124, "300–1000 Hz_40ms_log_rms_mae_db": 1.109863992271045, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10449519287120596, "300–1000 Hz_40ms_envelope_correlation": 0.9892619182550789, "300–1000 Hz_40ms_modulation_mae_db": 0.6750383320338887, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7219526274406944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2028213873501928, "1000–3000 Hz_10ms_envelope_correlation": 0.9665151070870227, "1000–3000 Hz_10ms_modulation_mae_db": 0.7731168913528549, "1000–3000 Hz_40ms_log_rms_mae_db": 1.013152132654008, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1207314236345674, "1000–3000 Hz_40ms_envelope_correlation": 0.9882031567456306, "1000–3000 Hz_40ms_modulation_mae_db": 0.567049214631182, "3000–6000 Hz_10ms_log_rms_mae_db": 1.469283678323159, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14650178408616626, "3000–6000 Hz_10ms_envelope_correlation": 0.9838659960454073, "3000–6000 Hz_10ms_modulation_mae_db": 0.15484986775285595, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7691629809852147, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06653611312741077, "3000–6000 Hz_40ms_envelope_correlation": 0.9965049713500096, "3000–6000 Hz_40ms_modulation_mae_db": 0.7247966630730236, "mr_stft_db": 6.075610736890095, "band_mae_db": 0.5440411619142153, "rms_gain_error_db": -0.15365187695014454, "output_peak": 0.4610304045493481}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.232673784826266, "全帯域_10ms_rms_envelope_nrmse": 0.159332179962096, "全帯域_10ms_envelope_correlation": 0.9720043930261063, "全帯域_10ms_modulation_mae_db": 0.2996220144591691, "全帯域_40ms_log_rms_mae_db": 0.6424594109354567, "全帯域_40ms_rms_envelope_nrmse": 0.09438726640710923, "全帯域_40ms_envelope_correlation": 0.9894038058751597, "全帯域_40ms_modulation_mae_db": 0.5762126552797949, "300–1000 Hz_10ms_log_rms_mae_db": 1.7188244186764214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18869425774028195, "300–1000 Hz_10ms_envelope_correlation": 0.9661903691239807, "300–1000 Hz_10ms_modulation_mae_db": 0.7835000080900404, "300–1000 Hz_40ms_log_rms_mae_db": 1.12087540099169, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11222219176480913, "300–1000 Hz_40ms_envelope_correlation": 0.987306546829075, "300–1000 Hz_40ms_modulation_mae_db": 0.30464553081847795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.701416672071391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2087912618232802, "1000–3000 Hz_10ms_envelope_correlation": 0.961853667850125, "1000–3000 Hz_10ms_modulation_mae_db": 0.7512327125537922, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9854407630487426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13214661284890822, "1000–3000 Hz_40ms_envelope_correlation": 0.9833459553699336, "1000–3000 Hz_40ms_modulation_mae_db": 0.5410906919589675, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5550090883028507, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15524647787218587, "3000–6000 Hz_10ms_envelope_correlation": 0.9817134469580888, "3000–6000 Hz_10ms_modulation_mae_db": 0.874274538928872, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8232066259807479, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06257382954881044, "3000–6000 Hz_40ms_envelope_correlation": 0.9968516448922674, "3000–6000 Hz_40ms_modulation_mae_db": 1.0913307534191754, "mr_stft_db": 6.088927802020641, "band_mae_db": 0.3424620081187622, "rms_gain_error_db": -0.16170458548413227, "output_peak": 0.4486662496075292}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1987590592514406, "全帯域_10ms_rms_envelope_nrmse": 0.16256712574463694, "全帯域_10ms_envelope_correlation": 0.9708710846454364, "全帯域_10ms_modulation_mae_db": 0.891346465495174, "全帯域_40ms_log_rms_mae_db": 0.7099528842300309, "全帯域_40ms_rms_envelope_nrmse": 0.09699701226073992, "全帯域_40ms_envelope_correlation": 0.989687329408466, "全帯域_40ms_modulation_mae_db": 1.3090993262828066, "300–1000 Hz_10ms_log_rms_mae_db": 1.6663060622547066, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16618173655860133, "300–1000 Hz_10ms_envelope_correlation": 0.9739314389241205, "300–1000 Hz_10ms_modulation_mae_db": 1.123636983153376, "300–1000 Hz_40ms_log_rms_mae_db": 1.04269175151988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08860083111160541, "300–1000 Hz_40ms_envelope_correlation": 0.9918957712721379, "300–1000 Hz_40ms_modulation_mae_db": 0.49120570442162315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6617911609538658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1848893081589033, "1000–3000 Hz_10ms_envelope_correlation": 0.9688228969728476, "1000–3000 Hz_10ms_modulation_mae_db": 0.13849016785172918, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9122560657838742, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08577327603886555, "1000–3000 Hz_40ms_envelope_correlation": 0.9927788545631705, "1000–3000 Hz_40ms_modulation_mae_db": 0.44284068007571226, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4925825092560991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15960474564894314, "3000–6000 Hz_10ms_envelope_correlation": 0.9809435035887213, "3000–6000 Hz_10ms_modulation_mae_db": 0.7329477866584391, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8335294012332766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07985141105797866, "3000–6000 Hz_40ms_envelope_correlation": 0.9957227790375126, "3000–6000 Hz_40ms_modulation_mae_db": 0.7336138496723166, "mr_stft_db": 6.079662218131184, "band_mae_db": 0.29529678848556756, "rms_gain_error_db": -0.2598763362031978, "output_peak": 0.41462471669677836}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1772143435293296, "全帯域_10ms_rms_envelope_nrmse": 0.15014688598913398, "全帯域_10ms_envelope_correlation": 0.9755365466430063, "全帯域_10ms_modulation_mae_db": 0.3876806200877885, "全帯域_40ms_log_rms_mae_db": 0.6460690465583526, "全帯域_40ms_rms_envelope_nrmse": 0.08373236118493052, "全帯域_40ms_envelope_correlation": 0.9923568813249299, "全帯域_40ms_modulation_mae_db": 0.6699504798517509, "300–1000 Hz_10ms_log_rms_mae_db": 1.7730971315603785, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1773387869829432, "300–1000 Hz_10ms_envelope_correlation": 0.9702675778978457, "300–1000 Hz_10ms_modulation_mae_db": 0.6049076749939181, "300–1000 Hz_40ms_log_rms_mae_db": 1.1701186712282046, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10382844469219503, "300–1000 Hz_40ms_envelope_correlation": 0.989270259871176, "300–1000 Hz_40ms_modulation_mae_db": 1.3387460986294133, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7062709998754158, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1977102269653677, "1000–3000 Hz_10ms_envelope_correlation": 0.9649057127283638, "1000–3000 Hz_10ms_modulation_mae_db": 1.0991625216076637, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9454717463524871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10358484082136324, "1000–3000 Hz_40ms_envelope_correlation": 0.9893836056454619, "1000–3000 Hz_40ms_modulation_mae_db": 1.0052356949562864, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4950495961032344, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13746100526133226, "3000–6000 Hz_10ms_envelope_correlation": 0.9857570615101274, "3000–6000 Hz_10ms_modulation_mae_db": 0.5102589008892713, "3000–6000 Hz_40ms_log_rms_mae_db": 0.851496597641115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06540883850604386, "3000–6000 Hz_40ms_envelope_correlation": 0.996653974251021, "3000–6000 Hz_40ms_modulation_mae_db": 0.36729208000559516, "mr_stft_db": 6.0711360727019965, "band_mae_db": 0.4777308075971212, "rms_gain_error_db": -0.25130264487225795, "output_peak": 0.42948009830749734}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.137822236257037, "全帯域_10ms_rms_envelope_nrmse": 0.15191951444730042, "全帯域_10ms_envelope_correlation": 0.9744267468798232, "全帯域_10ms_modulation_mae_db": 0.9634426239522883, "全帯域_40ms_log_rms_mae_db": 0.5740936005341951, "全帯域_40ms_rms_envelope_nrmse": 0.08105171156435559, "全帯域_40ms_envelope_correlation": 0.9926372270231983, "全帯域_40ms_modulation_mae_db": 0.3900054691625047, "300–1000 Hz_10ms_log_rms_mae_db": 1.7980156254975832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18822973391594477, "300–1000 Hz_10ms_envelope_correlation": 0.9664955086851333, "300–1000 Hz_10ms_modulation_mae_db": 0.9435899880453121, "300–1000 Hz_40ms_log_rms_mae_db": 1.0504819170513682, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09042166631017637, "300–1000 Hz_40ms_envelope_correlation": 0.991509296170328, "300–1000 Hz_40ms_modulation_mae_db": 0.46464726428339787, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6956978214772904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1787361729027225, "1000–3000 Hz_10ms_envelope_correlation": 0.9708722426783232, "1000–3000 Hz_10ms_modulation_mae_db": 0.5908544287083196, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9045032986512709, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08855896700043657, "1000–3000 Hz_40ms_envelope_correlation": 0.9924264221222413, "1000–3000 Hz_40ms_modulation_mae_db": 1.1538293962483608, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4656613215336658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16029776327800851, "3000–6000 Hz_10ms_envelope_correlation": 0.9817071507899853, "3000–6000 Hz_10ms_modulation_mae_db": 1.0746476008997556, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7789820311160944, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08528993423014652, "3000–6000 Hz_40ms_envelope_correlation": 0.9963032522252107, "3000–6000 Hz_40ms_modulation_mae_db": 0.6946661642023025, "mr_stft_db": 6.085187804731565, "band_mae_db": 0.2710466742458602, "rms_gain_error_db": -0.19550871422719246, "output_peak": 0.44805347568814535}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1469386934173664, "全帯域_10ms_rms_envelope_nrmse": 0.16201469837324278, "全帯域_10ms_envelope_correlation": 0.970787488741215, "全帯域_10ms_modulation_mae_db": 1.1519928929909575, "全帯域_40ms_log_rms_mae_db": 0.6014808851291541, "全帯域_40ms_rms_envelope_nrmse": 0.08082440256787024, "全帯域_40ms_envelope_correlation": 0.9920978177297034, "全帯域_40ms_modulation_mae_db": 0.9323578709416894, "300–1000 Hz_10ms_log_rms_mae_db": 1.7542368065257263, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16187265425152828, "300–1000 Hz_10ms_envelope_correlation": 0.9753060320310087, "300–1000 Hz_10ms_modulation_mae_db": 0.38335494615472077, "300–1000 Hz_40ms_log_rms_mae_db": 1.1568977538822371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09212226077318549, "300–1000 Hz_40ms_envelope_correlation": 0.9917039548947048, "300–1000 Hz_40ms_modulation_mae_db": 0.16630932507365165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.758263916087311, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20690567927915912, "1000–3000 Hz_10ms_envelope_correlation": 0.9618203421812589, "1000–3000 Hz_10ms_modulation_mae_db": 0.6100773547712589, "1000–3000 Hz_40ms_log_rms_mae_db": 0.934225380796759, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10619571096757364, "1000–3000 Hz_40ms_envelope_correlation": 0.9889367719487222, "1000–3000 Hz_40ms_modulation_mae_db": 0.5866266199077769, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5366115590700775, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15162331999012532, "3000–6000 Hz_10ms_envelope_correlation": 0.983077689923091, "3000–6000 Hz_10ms_modulation_mae_db": 0.9208787627878804, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8371255096786734, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.081201850654674, "3000–6000 Hz_40ms_envelope_correlation": 0.9957893997813259, "3000–6000 Hz_40ms_modulation_mae_db": 0.9304225913770341, "mr_stft_db": 6.09221799334268, "band_mae_db": 0.2990643132591274, "rms_gain_error_db": -0.08928429938359793, "output_peak": 0.4160110086951991}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2351954285908684, "全帯域_10ms_rms_envelope_nrmse": 0.16267049347528975, "全帯域_10ms_envelope_correlation": 0.970885436384082, "全帯域_10ms_modulation_mae_db": 0.06376295948831338, "全帯域_40ms_log_rms_mae_db": 0.6135457956425165, "全帯域_40ms_rms_envelope_nrmse": 0.08798114597678139, "全帯域_40ms_envelope_correlation": 0.9907016528368043, "全帯域_40ms_modulation_mae_db": 0.5191206620287944, "300–1000 Hz_10ms_log_rms_mae_db": 1.74195908163353, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1683650008311233, "300–1000 Hz_10ms_envelope_correlation": 0.973294593524396, "300–1000 Hz_10ms_modulation_mae_db": 0.6529674229680305, "300–1000 Hz_40ms_log_rms_mae_db": 1.063248755744913, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08935185662583053, "300–1000 Hz_40ms_envelope_correlation": 0.992313080584531, "300–1000 Hz_40ms_modulation_mae_db": 1.1311288478110113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7282586416592283, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1972179663465861, "1000–3000 Hz_10ms_envelope_correlation": 0.9646823201603515, "1000–3000 Hz_10ms_modulation_mae_db": 0.5683864236760584, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9534762752670101, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08972368423443412, "1000–3000 Hz_40ms_envelope_correlation": 0.9920002113012172, "1000–3000 Hz_40ms_modulation_mae_db": 0.5506028052342901, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5015129377222451, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16968554073098577, "3000–6000 Hz_10ms_envelope_correlation": 0.9783899182690574, "3000–6000 Hz_10ms_modulation_mae_db": 0.30644103966645925, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7996833942564685, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07525523379507944, "3000–6000 Hz_40ms_envelope_correlation": 0.9955198270522319, "3000–6000 Hz_40ms_modulation_mae_db": 0.5358497695139968, "mr_stft_db": 6.0976868386554415, "band_mae_db": 0.4810927176819996, "rms_gain_error_db": -0.08622147212663067, "output_peak": 0.5390077210695939}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1908986495157174, "全帯域_10ms_rms_envelope_nrmse": 0.19971557628035702, "全帯域_10ms_envelope_correlation": 0.9587229506426489, "全帯域_10ms_modulation_mae_db": 1.1884695114484138, "全帯域_40ms_log_rms_mae_db": 0.6297610867167072, "全帯域_40ms_rms_envelope_nrmse": 0.09578924514637344, "全帯域_40ms_envelope_correlation": 0.99020096017319, "全帯域_40ms_modulation_mae_db": 1.1090272019764964, "300–1000 Hz_10ms_log_rms_mae_db": 1.7845067815189428, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19714441756303952, "300–1000 Hz_10ms_envelope_correlation": 0.9633928038940275, "300–1000 Hz_10ms_modulation_mae_db": 0.9736241466171881, "300–1000 Hz_40ms_log_rms_mae_db": 1.1111862124605607, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10622580796514779, "300–1000 Hz_40ms_envelope_correlation": 0.9882411898275296, "300–1000 Hz_40ms_modulation_mae_db": 0.756085396060579, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6429203111452906, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19430077877243432, "1000–3000 Hz_10ms_envelope_correlation": 0.9658634254696185, "1000–3000 Hz_10ms_modulation_mae_db": 0.6760990129162289, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9488933486591729, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10603958948175284, "1000–3000 Hz_40ms_envelope_correlation": 0.9887960688170275, "1000–3000 Hz_40ms_modulation_mae_db": 0.716507192113149, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5483704852615947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22186511875494874, "3000–6000 Hz_10ms_envelope_correlation": 0.9643772364362124, "3000–6000 Hz_10ms_modulation_mae_db": 0.5840724387852133, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8641947068649237, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10495584159781786, "3000–6000 Hz_40ms_envelope_correlation": 0.9920635085975804, "3000–6000 Hz_40ms_modulation_mae_db": 0.6215546346276248, "mr_stft_db": 6.079457782112837, "band_mae_db": 0.32100592807674, "rms_gain_error_db": 0.1791873855378889, "output_peak": 0.5918307601849075}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64800, "output_samples": 64800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0804373705992685, "spectral_convergence": 0.6190669887330673, "spectral_convergence_status": "ok", "amplitude_floor": 1.7331847370499167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.778774907269372, "spectral_convergence": 0.5720445554535147, "spectral_convergence_status": "ok", "amplitude_floor": 1.0846095570828337e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.367619932801646, "spectral_convergence": 0.607883168031517, "spectral_convergence_status": "ok", "amplitude_floor": 4.5533844599041527e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024118742205032635, "output_power": 0.00016837646020729678, "delta_db": -1.5607328048589795, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 9.191122965591724e-05, "output_power": 0.00010513149651060366, "delta_db": 0.5836427003691472, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 7.435432994216677e-05, "output_power": 6.49727762672252e-05, "delta_db": -0.5857483981851275, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.4154339925686723e-05, "output_power": 2.5373930049751917e-05, "delta_db": 0.21392564415900958, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006210431709364446, "output_power": 0.0006301963764884562, "delta_db": 0.06354111141213464, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023011417890304625, "output_power": 0.0002441231755331913, "delta_db": 0.2566563125008933, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5440411619142153, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15365187695014454, "pitch": null, "output_rms_dbfs": -29.07218193601523, "output_peak": 0.4610304045493481, "reference_rms_dbfs": -28.918530059065088, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.075610736890095}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.0001629307662014554, "log_rms_mae_db": 1.2263940903873503, "rms_envelope_nrmse": 0.16694479222940575, "envelope_correlation": 0.9691487318333699, "rms_gain_error_db": -0.15320177605125815, "reference_mean_power": 0.0012840985730840112, "output_mean_power": 0.001239590347431906}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 7.3397819245367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003947037290166488, "output_power": 0.0003665875106028348, "reference_active": true, "delta_db": -0.3209356472175392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980542214464163e-05, "output_power": 5.355285244266497e-05, "reference_active": true, "delta_db": -1.1510654973610177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6036857037353872e-05, "output_power": 2.5692961082362454e-05, "reference_active": true, "delta_db": -0.05774399419511383}], "active_band_count": 3, "active_band_mae_db": 0.5099150462578902}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.000144561944924798, "log_rms_mae_db": 0.6396750311663291, "rms_envelope_nrmse": 0.10695403818740934, "envelope_correlation": 0.9862354967124499, "rms_gain_error_db": -0.15309088995473558, "reference_mean_power": 0.0012928029739092497, "output_mean_power": 0.0012480249091922615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 6.614408925582191e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037454976712341086, "output_power": 0.00034054184355870453, "reference_active": true, "delta_db": -0.4133904858577232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9110632304140427e-05, "output_power": 2.2241325029919215e-05, "reference_active": true, "delta_db": -1.1689098190107807}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1315453531682465e-06, "output_power": 1.3817589476581189e-06, "reference_active": true, "delta_db": -1.8826229206037546}], "active_band_count": 3, "active_band_mae_db": 1.1549744084907527}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 3.733123977436472e-05, "log_rms_mae_db": 1.784932989105563, "rms_envelope_nrmse": 0.18161541373515538, "envelope_correlation": 0.9688266121766055, "rms_gain_error_db": -0.24382324470577393, "reference_mean_power": 0.00013419941652038683, "output_mean_power": 0.0001268727353062099}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.2377163404532544e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0020888510118017e-05, "output_power": 1.7718684748980593e-05, "reference_active": true, "delta_db": -0.5305186595478577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.868636044401026e-06, "output_power": 6.745153834580618e-06, "reference_active": true, "delta_db": -0.6690759899660841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3701828329102724e-06, "output_power": 1.4086019258060919e-06, "reference_active": true, "delta_db": 0.1200975589349619}], "active_band_count": 3, "active_band_mae_db": 0.43989740281630124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.100732475570721e-05, "log_rms_mae_db": 1.109863992271045, "rms_envelope_nrmse": 0.10449519287120596, "envelope_correlation": 0.9892619182550789, "rms_gain_error_db": -0.2438222471032864, "reference_mean_power": 0.00013520032575819994, "output_mean_power": 0.00012781902878718576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 4.6212903051863556e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.815754516308902e-05, "output_power": 1.581520227170818e-05, "reference_active": true, "delta_db": -0.5998238226747378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8683197262456875e-06, "output_power": 3.1291472811547835e-06, "reference_active": true, "delta_db": -0.9209635762857814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2646920446881671e-07, "output_power": 1.126035356025466e-07, "reference_active": true, "delta_db": -0.5043275971411472}], "active_band_count": 3, "active_band_mae_db": 0.6750383320338887}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 4.662083951365485e-05, "log_rms_mae_db": 1.7219526274406944, "rms_envelope_nrmse": 0.2028213873501928, "envelope_correlation": 0.9665151070870227, "rms_gain_error_db": 0.3668754328729146, "reference_mean_power": 0.00017506320579060826, "output_mean_power": 0.00019049449088950812}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 9.028764248682407e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2773698500231605e-05, "output_power": 4.910087123913577e-05, "reference_active": true, "delta_db": 0.5991239436536768}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.603092418254862e-06, "output_power": 7.784162333520212e-06, "reference_active": true, "delta_db": 0.7146450848541279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4970630790605404e-06, "output_power": 4.408203455058926e-06, "reference_active": true, "delta_db": 1.0055816455507596}], "active_band_count": 3, "active_band_mae_db": 0.7731168913528549}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.913562930844258e-05, "log_rms_mae_db": 1.013152132654008, "rms_envelope_nrmse": 0.1207314236345674, "envelope_correlation": 0.9882031567456306, "rms_gain_error_db": 0.36717435323301917, "reference_mean_power": 0.00017624479575510055, "output_mean_power": 0.00019179343487601432}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 7.899096535479963e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7672515224694995e-05, "output_power": 4.459268423863172e-05, "reference_active": true, "delta_db": 0.7323899825363727}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.050253319949336e-06, "output_power": 3.7650490814051164e-06, "reference_active": true, "delta_db": 0.9143473357218092}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55667878122912e-07, "output_power": 2.5248473984645393e-07, "reference_active": true, "delta_db": -0.054410325635364076}], "active_band_count": 3, "active_band_mae_db": 0.567049214631182}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.00015276264453937597, "log_rms_mae_db": 1.469283678323159, "rms_envelope_nrmse": 0.14650178408616626, "envelope_correlation": 0.9838659960454073, "rms_gain_error_db": 0.02964282103714957, "reference_mean_power": 0.0006070254152411961, "output_mean_power": 0.0006111828465234907}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.389907119886337e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002863305566547835, "output_power": 0.00029165659580219025, "reference_active": true, "delta_db": 0.08004104849844976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.608243104449444e-05, "output_power": 3.341752790801879e-05, "reference_active": true, "delta_db": -0.3332147091761709}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0810295137794261e-05, "output_power": 1.0683367460795068e-05, "reference_active": true, "delta_db": -0.05129384558394716}], "active_band_count": 3, "active_band_mae_db": 0.15484986775285595}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.00013215266595876854, "log_rms_mae_db": 0.7691629809852147, "rms_envelope_nrmse": 0.06653611312741077, "envelope_correlation": 0.9965049713500096, "rms_gain_error_db": 0.030330917353455677, "reference_mean_power": 0.0006109513084873906, "output_mean_power": 0.000615233097487626}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 5.083168365591421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002672957849080471, "output_power": 0.0002737533231810635, "reference_active": true, "delta_db": 0.10367289672131812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4768224147952787e-05, "output_power": 1.3458825421806564e-05, "reference_active": true, "delta_db": -0.4032111552413803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.446503663842908e-07, "output_power": 5.75342400544985e-07, "reference_active": true, "delta_db": -1.6675059372563723}], "active_band_count": 3, "active_band_mae_db": 0.7247966630730236}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7602294317093002, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "27f1c812db50c3f4f7ee8c18b42a0e9d20b832974f6a2339ae7eb32bfe8e3467", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9656669643909889, "standard_deviation": 0.003132765478108811, "minimum": 0.9618203421812589, "maximum": 0.9708722426783232, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.702071518838811, "standard_deviation": 0.036681389745536884, "minimum": 1.6429203111452906, "maximum": 1.758263916087311, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6509274391797382, "standard_deviation": 0.2675876837872305, "minimum": 0.13849016785172918, "maximum": 1.0991625216076637, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19642159769983075, "standard_deviation": 0.01039534551268113, "minimum": 0.1787361729027225, "maximum": 0.2087912618232802, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9894838808141757, "standard_deviation": 0.003069298941326966, "minimum": 0.9833459553699336, "maximum": 0.9927788545631705, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9496773764016656, "standard_deviation": 0.03589789910106016, "minimum": 0.9045032986512709, "maximum": 1.013152132654008, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6954727868907156, "standard_deviation": 0.25170008029720914, "minimum": 0.44284068007571226, "maximum": 1.1538293962483608, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1040942631284877, "standard_deviation": 0.016281921853676058, "minimum": 0.08577327603886555, "maximum": 0.13214661284890822, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9799790004400863, "standard_deviation": 0.006665067762330951, "minimum": 0.9643772364362124, "maximum": 0.9857570615101274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5080101469466158, "standard_deviation": 0.03463537037984133, "minimum": 1.4656613215336658, "maximum": 1.5550090883028507, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6447963670460934, "standard_deviation": 0.3157177993157775, "minimum": 0.15484986775285595, "maximum": 1.0746476008997556, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16278571945283699, "standard_deviation": 0.02575222182126527, "minimum": 0.13746100526133226, "maximum": 0.22186511875494874, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9956761696483949, "standard_deviation": 0.00153570575130895, "minimum": 0.9920635085975804, "maximum": 0.9968516448922674, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8196726559695643, "standard_deviation": 0.03403905638528281, "minimum": 0.7691629809852147, "maximum": 0.8641947068649237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7124408132363836, "standard_deviation": 0.22378653325190762, "minimum": 0.36729208000559516, "maximum": 1.0913307534191754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07763413156474519, "standard_deviation": 0.013772787644248042, "minimum": 0.06257382954881044, "maximum": 0.10495584159781786, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9697131170321398, "standard_deviation": 0.004237660132784962, "minimum": 0.9633928038940275, "maximum": 0.9753060320310087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7527348620966066, "standard_deviation": 0.04350594818983488, "minimum": 1.6663060622547066, "maximum": 1.7980156254975832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7381848216048609, "standard_deviation": 0.26407634737216906, "minimum": 0.38335494615472077, "maximum": 1.123636983153376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17868025019732722, "standard_deviation": 0.012466673663139414, "minimum": 0.16187265425152828, "maximum": 0.19714441756303952, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9901877522130702, "standard_deviation": 0.00189996650708088, "minimum": 0.987306546829075, "maximum": 0.992313080584531, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1031705568937373, "standard_deviation": 0.04753751135364424, "minimum": 1.04269175151988, "maximum": 1.1701186712282046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6659758123915054, "standard_deviation": 0.4015137458222015, "minimum": 0.16630932507365165, "maximum": 1.3387460986294133, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09840853151426945, "standard_deviation": 0.009257209459187782, "minimum": 0.08860083111160541, "maximum": 0.11222219176480913, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3789675499224242, "standard_deviation": 0.10498807515775166, "minimum": 0.2710466742458602, "maximum": 0.5440411619142153, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.083735906073304, "standard_deviation": 0.008906735580526088, "minimum": 6.0711360727019965, "maximum": 6.0976868386554415, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4685880543498749, "standard_deviation": 0.06350385709795173, "minimum": 0.41462471669677836, "maximum": 0.5918307601849075, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1272953179636581, "standard_deviation": 0.13966839438884923, "minimum": -0.2598763362031978, "maximum": 0.1791873855378889, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.970297922349461, "standard_deviation": 0.0051216123371985054, "minimum": 0.9587229506426489, "maximum": 0.9755365466430063, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1932370357219222, "standard_deviation": 0.037649090569630624, "minimum": 1.137822236257037, "maximum": 1.2351954285908684, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6820290167724994, "standard_deviation": 0.42174936069538077, "minimum": 0.06376295948831338, "maximum": 1.1884695114484138, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16441390831268282, "standard_deviation": 0.015354390769210583, "minimum": 0.15014688598913398, "maximum": 0.19971557628035702, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9904151463854877, "standard_deviation": 0.0020938454042251885, "minimum": 0.9862354967124499, "maximum": 0.9926372270231983, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6321297176140928, "standard_deviation": 0.0397723860788549, "minimum": 0.5740936005341951, "maximum": 0.7099528842300309, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8325935092518237, "standard_deviation": 0.3388402591756027, "minimum": 0.3900054691625047, "maximum": 1.3090993262828066, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09096464791194621, "standard_deviation": 0.009175582617732818, "minimum": 0.08082440256787024, "maximum": 0.10695403818740934, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-76324-0024", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.2734498977661133, "predicted_mos": 2.2734498977661133, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "6a999d955a15732b2410bc1296c9460418a2286f9c281d4cf7ebd778269a2a39", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.810061499942094, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.20000168939772, "全帯域_10ms_rms_envelope_nrmse": 0.1718294491955126, "全帯域_10ms_envelope_correlation": 0.9674198808003104, "全帯域_10ms_modulation_mae_db": 1.4359096252536265, "全帯域_40ms_log_rms_mae_db": 0.6235817518224778, "全帯域_40ms_rms_envelope_nrmse": 0.09876820434100891, "全帯域_40ms_envelope_correlation": 0.9891713536592065, "全帯域_40ms_modulation_mae_db": 1.5634130843462557, "300–1000 Hz_10ms_log_rms_mae_db": 2.3221834209021597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2137577514791524, "300–1000 Hz_10ms_envelope_correlation": 0.9564637406152516, "300–1000 Hz_10ms_modulation_mae_db": 0.311385257829933, "300–1000 Hz_40ms_log_rms_mae_db": 1.645109687538519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12289577194618498, "300–1000 Hz_40ms_envelope_correlation": 0.9843226589314995, "300–1000 Hz_40ms_modulation_mae_db": 0.5561821276926583, "1000–3000 Hz_10ms_log_rms_mae_db": 2.208397492659734, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23059576886822022, "1000–3000 Hz_10ms_envelope_correlation": 0.9512550538232826, "1000–3000 Hz_10ms_modulation_mae_db": 0.5267202868105139, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3616586690643016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1371739975087695, "1000–3000 Hz_40ms_envelope_correlation": 0.9812337679656508, "1000–3000 Hz_40ms_modulation_mae_db": 0.5874404035026286, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0740983796649495, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20130348753220348, "3000–6000 Hz_10ms_envelope_correlation": 0.9689131127867855, "3000–6000 Hz_10ms_modulation_mae_db": 1.0313915877263777, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1982271775300795, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1177448528398156, "3000–6000 Hz_40ms_envelope_correlation": 0.9888251564867047, "3000–6000 Hz_40ms_modulation_mae_db": 0.9377445824146666, "mr_stft_db": 6.399975825473017, "band_mae_db": 0.35897566973591494, "rms_gain_error_db": -0.3031888038870255, "output_peak": 0.41279511964378746}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1413256075694007, "全帯域_10ms_rms_envelope_nrmse": 0.17039770418335232, "全帯域_10ms_envelope_correlation": 0.9679423816104167, "全帯域_10ms_modulation_mae_db": 0.7703747920608994, "全帯域_40ms_log_rms_mae_db": 0.604621366073165, "全帯域_40ms_rms_envelope_nrmse": 0.083881789031145, "全帯域_40ms_envelope_correlation": 0.9913834906744442, "全帯域_40ms_modulation_mae_db": 0.540825143552042, "300–1000 Hz_10ms_log_rms_mae_db": 2.270810553449259, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1804364826654091, "300–1000 Hz_10ms_envelope_correlation": 0.9691641180236833, "300–1000 Hz_10ms_modulation_mae_db": 0.9327256929781064, "300–1000 Hz_40ms_log_rms_mae_db": 1.5011171736067794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09474956608852327, "300–1000 Hz_40ms_envelope_correlation": 0.9912080152696282, "300–1000 Hz_40ms_modulation_mae_db": 0.5759701264531522, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2803518778052725, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2244170889703428, "1000–3000 Hz_10ms_envelope_correlation": 0.9548065372602894, "1000–3000 Hz_10ms_modulation_mae_db": 0.4714048136191744, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4341509245241975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11470621726592028, "1000–3000 Hz_40ms_envelope_correlation": 0.9870743151343627, "1000–3000 Hz_40ms_modulation_mae_db": 0.4375191029793866, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0383594858388485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18524797662910072, "3000–6000 Hz_10ms_envelope_correlation": 0.9737690921612382, "3000–6000 Hz_10ms_modulation_mae_db": 0.9582829764189026, "3000–6000 Hz_40ms_log_rms_mae_db": 1.140347340497323, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09176757366174053, "3000–6000 Hz_40ms_envelope_correlation": 0.993457817759544, "3000–6000 Hz_40ms_modulation_mae_db": 1.197660926276891, "mr_stft_db": 6.393183361742689, "band_mae_db": 0.17242031094261404, "rms_gain_error_db": -0.014621918885561035, "output_peak": 0.4034687509809015}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.131672433386449, "全帯域_10ms_rms_envelope_nrmse": 0.1630178485362005, "全帯域_10ms_envelope_correlation": 0.9719589771837045, "全帯域_10ms_modulation_mae_db": 1.0360098940742102, "全帯域_40ms_log_rms_mae_db": 0.5803999260265927, "全帯域_40ms_rms_envelope_nrmse": 0.10739611617916638, "全帯域_40ms_envelope_correlation": 0.9867850584502275, "全帯域_40ms_modulation_mae_db": 1.1592365792766646, "300–1000 Hz_10ms_log_rms_mae_db": 2.219585650177883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18252858661371385, "300–1000 Hz_10ms_envelope_correlation": 0.9684097854526457, "300–1000 Hz_10ms_modulation_mae_db": 0.5556264775496288, "300–1000 Hz_40ms_log_rms_mae_db": 1.498120510550059, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09516326557807926, "300–1000 Hz_40ms_envelope_correlation": 0.9909935848402897, "300–1000 Hz_40ms_modulation_mae_db": 0.5440492172242914, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2438901521822565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21282702804314108, "1000–3000 Hz_10ms_envelope_correlation": 0.9584349688280756, "1000–3000 Hz_10ms_modulation_mae_db": 0.6584225039408441, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3476944391058912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1053534202451664, "1000–3000 Hz_40ms_envelope_correlation": 0.9894505910472299, "1000–3000 Hz_40ms_modulation_mae_db": 0.6208229424903222, "3000–6000 Hz_10ms_log_rms_mae_db": 2.10731839278592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18288738286767173, "3000–6000 Hz_10ms_envelope_correlation": 0.9744866551974971, "3000–6000 Hz_10ms_modulation_mae_db": 0.07148512637912963, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2420216802300867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0996241647028212, "3000–6000 Hz_40ms_envelope_correlation": 0.9919780549049035, "3000–6000 Hz_40ms_modulation_mae_db": 0.6196500424851266, "mr_stft_db": 6.407063333416208, "band_mae_db": 0.3355198726321135, "rms_gain_error_db": 0.06795910924454669, "output_peak": 0.48364745119833796}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1510692359949302, "全帯域_10ms_rms_envelope_nrmse": 0.15700422284521884, "全帯域_10ms_envelope_correlation": 0.9726688456935197, "全帯域_10ms_modulation_mae_db": 0.566264669839159, "全帯域_40ms_log_rms_mae_db": 0.6131885268818135, "全帯域_40ms_rms_envelope_nrmse": 0.09383006771329293, "全帯域_40ms_envelope_correlation": 0.9896760309669284, "全帯域_40ms_modulation_mae_db": 1.0266924783599833, "300–1000 Hz_10ms_log_rms_mae_db": 2.2747067806797725, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18125332655655207, "300–1000 Hz_10ms_envelope_correlation": 0.9693361371762507, "300–1000 Hz_10ms_modulation_mae_db": 0.8628832879651744, "300–1000 Hz_40ms_log_rms_mae_db": 1.6119090143446428, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1022844844294703, "300–1000 Hz_40ms_envelope_correlation": 0.9895806517393048, "300–1000 Hz_40ms_modulation_mae_db": 0.4075985120116614, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0953424294912444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19933170337991202, "1000–3000 Hz_10ms_envelope_correlation": 0.9647066698551159, "1000–3000 Hz_10ms_modulation_mae_db": 0.1397331373263425, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3046577168272617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1014085068800915, "1000–3000 Hz_40ms_envelope_correlation": 0.9901135439108725, "1000–3000 Hz_40ms_modulation_mae_db": 0.4170319522888972, "3000–6000 Hz_10ms_log_rms_mae_db": 2.028722727133448, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20139048960717773, "3000–6000 Hz_10ms_envelope_correlation": 0.9688758773817282, "3000–6000 Hz_10ms_modulation_mae_db": 0.6436137252298199, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1620781825511648, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11250331684237104, "3000–6000 Hz_40ms_envelope_correlation": 0.9898309589717044, "3000–6000 Hz_40ms_modulation_mae_db": 0.10713765953013359, "mr_stft_db": 6.411941744348287, "band_mae_db": 0.4115996351840307, "rms_gain_error_db": -0.2099881033250338, "output_peak": 0.5815060979787483}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1306630475133348, "全帯域_10ms_rms_envelope_nrmse": 0.16871905071953885, "全帯域_10ms_envelope_correlation": 0.9685712913471562, "全帯域_10ms_modulation_mae_db": 0.5725584606806323, "全帯域_40ms_log_rms_mae_db": 0.5979188014854694, "全帯域_40ms_rms_envelope_nrmse": 0.07644004671655716, "全帯域_40ms_envelope_correlation": 0.993075692423633, "全帯域_40ms_modulation_mae_db": 0.34520951573660597, "300–1000 Hz_10ms_log_rms_mae_db": 2.2703415167973646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1918365854582425, "300–1000 Hz_10ms_envelope_correlation": 0.9651065832725101, "300–1000 Hz_10ms_modulation_mae_db": 0.8759802382672626, "300–1000 Hz_40ms_log_rms_mae_db": 1.5412271960949375, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10235672719917978, "300–1000 Hz_40ms_envelope_correlation": 0.9898528711636698, "300–1000 Hz_40ms_modulation_mae_db": 0.4067996301050587, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0984576505073695, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19204551493246141, "1000–3000 Hz_10ms_envelope_correlation": 0.9671653517109897, "1000–3000 Hz_10ms_modulation_mae_db": 0.5305696911076825, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2682202733454484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09510747671583944, "1000–3000 Hz_40ms_envelope_correlation": 0.9912491147785838, "1000–3000 Hz_40ms_modulation_mae_db": 0.20282797744415246, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0374292186847347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1647662871894142, "3000–6000 Hz_10ms_envelope_correlation": 0.9792940217010841, "3000–6000 Hz_10ms_modulation_mae_db": 0.6084290331316994, "3000–6000 Hz_40ms_log_rms_mae_db": 1.143699489108686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07407618206935615, "3000–6000 Hz_40ms_envelope_correlation": 0.9956803012880953, "3000–6000 Hz_40ms_modulation_mae_db": 0.7513797696865225, "mr_stft_db": 6.404763561187365, "band_mae_db": 0.2922878306734195, "rms_gain_error_db": -0.12093114941976386, "output_peak": 0.4420725920745708}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0709082276807853, "全帯域_10ms_rms_envelope_nrmse": 0.15206236153787156, "全帯域_10ms_envelope_correlation": 0.9750683150988277, "全帯域_10ms_modulation_mae_db": 0.05185003867452187, "全帯域_40ms_log_rms_mae_db": 0.5693110307455349, "全帯域_40ms_rms_envelope_nrmse": 0.08263452293229497, "全帯域_40ms_envelope_correlation": 0.9919574383828472, "全帯域_40ms_modulation_mae_db": 0.4628957190254795, "300–1000 Hz_10ms_log_rms_mae_db": 2.3511776379738434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2028913374932654, "300–1000 Hz_10ms_envelope_correlation": 0.9609661478267466, "300–1000 Hz_10ms_modulation_mae_db": 0.9884888529365033, "300–1000 Hz_40ms_log_rms_mae_db": 1.6651906203378863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12733448972366132, "300–1000 Hz_40ms_envelope_correlation": 0.9830683744274248, "300–1000 Hz_40ms_modulation_mae_db": 0.5542425828583634, "1000–3000 Hz_10ms_log_rms_mae_db": 2.207716585591576, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23921729305635966, "1000–3000 Hz_10ms_envelope_correlation": 0.949373287562273, "1000–3000 Hz_10ms_modulation_mae_db": 0.8414964281318092, "1000–3000 Hz_40ms_log_rms_mae_db": 1.337338578982368, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11745500677988653, "1000–3000 Hz_40ms_envelope_correlation": 0.9867652563118317, "1000–3000 Hz_40ms_modulation_mae_db": 0.6200795530507438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.045557929800704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16766972980126313, "3000–6000 Hz_10ms_envelope_correlation": 0.9786712160128419, "3000–6000 Hz_10ms_modulation_mae_db": 0.7050990445399048, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1608519278436626, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09681481594491492, "3000–6000 Hz_40ms_envelope_correlation": 0.9924658556924157, "3000–6000 Hz_40ms_modulation_mae_db": 0.9514932721167703, "mr_stft_db": 6.4067786047087205, "band_mae_db": 0.2268346531767974, "rms_gain_error_db": 0.08426463216682822, "output_peak": 0.43762771803179873}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1724150430762543, "全帯域_10ms_rms_envelope_nrmse": 0.15781672544693667, "全帯域_10ms_envelope_correlation": 0.9732803428575159, "全帯域_10ms_modulation_mae_db": 1.7017553458242631, "全帯域_40ms_log_rms_mae_db": 0.6182228880861004, "全帯域_40ms_rms_envelope_nrmse": 0.09814618820785297, "全帯域_40ms_envelope_correlation": 0.9900933115940925, "全帯域_40ms_modulation_mae_db": 2.1872907922632074, "300–1000 Hz_10ms_log_rms_mae_db": 2.2987324204409765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19742680147138927, "300–1000 Hz_10ms_envelope_correlation": 0.9634302856505771, "300–1000 Hz_10ms_modulation_mae_db": 0.31817851303253203, "300–1000 Hz_40ms_log_rms_mae_db": 1.6519340985839386, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11566879770470304, "300–1000 Hz_40ms_envelope_correlation": 0.9863287941396564, "300–1000 Hz_40ms_modulation_mae_db": 0.1228761811226297, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1817630857610983, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20795457589376903, "1000–3000 Hz_10ms_envelope_correlation": 0.961101000685405, "1000–3000 Hz_10ms_modulation_mae_db": 0.2994217677838407, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2659516711171737, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09910720533855759, "1000–3000 Hz_40ms_envelope_correlation": 0.990324992454848, "1000–3000 Hz_40ms_modulation_mae_db": 0.3439883371735279, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0386058266419953, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19168946129893455, "3000–6000 Hz_10ms_envelope_correlation": 0.9721997558539995, "3000–6000 Hz_10ms_modulation_mae_db": 0.9358933289951202, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1999145303234051, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11644177897454436, "3000–6000 Hz_40ms_envelope_correlation": 0.9898559568355828, "3000–6000 Hz_40ms_modulation_mae_db": 1.0573989939820925, "mr_stft_db": 6.409658405684314, "band_mae_db": 0.46553400530767636, "rms_gain_error_db": -0.3580691002114371, "output_peak": 0.47412134075500023}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1479939468630826, "全帯域_10ms_rms_envelope_nrmse": 0.1611185713944982, "全帯域_10ms_envelope_correlation": 0.9717770749780374, "全帯域_10ms_modulation_mae_db": 0.6221564004415938, "全帯域_40ms_log_rms_mae_db": 0.6371107090204401, "全帯域_40ms_rms_envelope_nrmse": 0.09433845590358707, "全帯域_40ms_envelope_correlation": 0.9893260721971281, "全帯域_40ms_modulation_mae_db": 0.5540240623770932, "300–1000 Hz_10ms_log_rms_mae_db": 2.2317049586797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18609940627712962, "300–1000 Hz_10ms_envelope_correlation": 0.9671451637670342, "300–1000 Hz_10ms_modulation_mae_db": 0.7246581793924287, "300–1000 Hz_40ms_log_rms_mae_db": 1.545977467790273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10067511368173118, "300–1000 Hz_40ms_envelope_correlation": 0.9898816745180147, "300–1000 Hz_40ms_modulation_mae_db": 0.8381081629506552, "1000–3000 Hz_10ms_log_rms_mae_db": 2.199667285481419, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21392671936110838, "1000–3000 Hz_10ms_envelope_correlation": 0.9610337943218306, "1000–3000 Hz_10ms_modulation_mae_db": 0.6426307572513034, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3257573935898281, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12513753159051957, "1000–3000 Hz_40ms_envelope_correlation": 0.9860384305745878, "1000–3000 Hz_40ms_modulation_mae_db": 0.46352750395360104, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0217561184249444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16802670953227836, "3000–6000 Hz_10ms_envelope_correlation": 0.9789309058133376, "3000–6000 Hz_10ms_modulation_mae_db": 0.45748756486452574, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1370720565158488, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08765800256949274, "3000–6000 Hz_40ms_envelope_correlation": 0.9948775061047656, "3000–6000 Hz_40ms_modulation_mae_db": 0.39569090925043127, "mr_stft_db": 6.422239916515395, "band_mae_db": 0.33076019782698474, "rms_gain_error_db": -0.00971709366556628, "output_peak": 0.45124926445874564}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64800, "output_samples": 64800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.444326907203737, "spectral_convergence": 0.6338295294210455, "spectral_convergence_status": "ok", "amplitude_floor": 1.7331847370499167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.183751137791801, "spectral_convergence": 0.5933952457032895, "spectral_convergence_status": "ok", "amplitude_floor": 1.0846095570828337e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.57184943142351, "spectral_convergence": 0.5983949500307847, "spectral_convergence_status": "ok", "amplitude_floor": 4.5533844599041527e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024118742205032635, "output_power": 0.00017348662421577514, "delta_db": -1.430886591558541, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 9.191122965591724e-05, "output_power": 9.817683163983938e-05, "delta_db": 0.2864043599669217, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 7.435432994216677e-05, "output_power": 7.113856715938674e-05, "delta_db": -0.19201150676966366, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.4154339925686723e-05, "output_power": 2.4016893135846156e-05, "delta_db": -0.024783482000048252, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006210431709364446, "output_power": 0.0005968276027549435, "delta_db": -0.17272889922458728, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023011417890304625, "output_power": 0.00023262013167601463, "delta_db": 0.04703917889572738, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.35897566973591494, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.3031888038870255, "pitch": null, "output_rms_dbfs": -29.221718862952116, "output_peak": 0.41279511964378746, "reference_rms_dbfs": -28.918530059065088, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.399975825473017}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.0001629307662014554, "log_rms_mae_db": 1.20000168939772, "rms_envelope_nrmse": 0.1718294491955126, "envelope_correlation": 0.9674198808003104, "rms_gain_error_db": -0.3029805818404783, "reference_mean_power": 0.0012840985730840112, "output_mean_power": 0.0011975683375406512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 7.3397819245367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003947037290166488, "output_power": 0.00034259738417505954, "reference_active": true, "delta_db": -0.614871867857841}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980542214464163e-05, "output_power": 5.593973379741269e-05, "reference_active": true, "delta_db": -0.9616876246360853}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6036857037353872e-05, "output_power": 1.3882625881482484e-05, "reference_active": true, "delta_db": -2.7311693832669537}], "active_band_count": 3, "active_band_mae_db": 1.4359096252536265}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.000144561944924798, "log_rms_mae_db": 0.6235817518224778, "rms_envelope_nrmse": 0.09876820434100891, "envelope_correlation": 0.9891713536592065, "rms_gain_error_db": -0.30318214451623743, "reference_mean_power": 0.0012928029739092497, "output_mean_power": 0.0012056302273899355}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 6.614408925582191e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037454976712341086, "output_power": 0.0003152706190860207, "reference_active": true, "delta_db": -0.748260318172822}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9110632304140427e-05, "output_power": 2.5972922081609006e-05, "reference_active": true, "delta_db": -0.4953082608835154}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1315453531682465e-06, "output_power": 9.638900789511172e-07, "reference_active": true, "delta_db": -3.44667067398243}], "active_band_count": 3, "active_band_mae_db": 1.5634130843462557}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 3.733123977436472e-05, "log_rms_mae_db": 2.3221834209021597, "rms_envelope_nrmse": 0.2137577514791524, "envelope_correlation": 0.9564637406152516, "rms_gain_error_db": -0.1284912265967893, "reference_mean_power": 0.00013419941652038683, "output_mean_power": 0.00013028712624899266}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.2377163404532544e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0020888510118017e-05, "output_power": 1.9809002665740515e-05, "reference_active": true, "delta_db": -0.04620736756563362}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.868636044401026e-06, "output_power": 6.730222609419898e-06, "reference_active": true, "delta_db": -0.6787002871070235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3701828329102724e-06, "output_power": 1.437816071760262e-06, "reference_active": true, "delta_db": 0.20924811881714192}], "active_band_count": 3, "active_band_mae_db": 0.311385257829933}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.100732475570721e-05, "log_rms_mae_db": 1.645109687538519, "rms_envelope_nrmse": 0.12289577194618498, "envelope_correlation": 0.9843226589314995, "rms_gain_error_db": -0.12848778043245418, "reference_mean_power": 0.00013520032575819994, "output_mean_power": 0.00013125896032371078}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 4.6212903051863556e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.815754516308902e-05, "output_power": 1.8093765647886313e-05, "reference_active": true, "delta_db": -0.015281720561935393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8683197262456875e-06, "output_power": 3.1180276870502716e-06, "reference_active": true, "delta_db": -0.936423951527683}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2646920446881671e-07, "output_power": 1.0722615071013684e-07, "reference_active": true, "delta_db": -0.7168407109883563}], "active_band_count": 3, "active_band_mae_db": 0.5561821276926583}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 4.662083951365485e-05, "log_rms_mae_db": 2.208397492659734, "rms_envelope_nrmse": 0.23059576886822022, "envelope_correlation": 0.9512550538232826, "rms_gain_error_db": -0.09645281838931566, "reference_mean_power": 0.00017506320579060826, "output_mean_power": 0.000171218069191562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 9.028764248682407e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2773698500231605e-05, "output_power": 4.1319890645782686e-05, "reference_active": true, "delta_db": -0.15017640410872107}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.603092418254862e-06, "output_power": 7.293813009902248e-06, "reference_active": true, "delta_db": 0.43207249403949705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4970630790605404e-06, "output_power": 4.400425394485637e-06, "reference_active": true, "delta_db": 0.9979119622833235}], "active_band_count": 3, "active_band_mae_db": 0.5267202868105139}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.913562930844258e-05, "log_rms_mae_db": 1.3616586690643016, "rms_envelope_nrmse": 0.1371739975087695, "envelope_correlation": 0.9812337679656508, "rms_gain_error_db": -0.096659102797806, "reference_mean_power": 0.00017624479575510055, "output_mean_power": 0.00017236551904827128}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 7.899096535479963e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7672515224694995e-05, "output_power": 3.68761421524142e-05, "reference_active": true, "delta_db": -0.09279136611651159}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.050253319949336e-06, "output_power": 3.6705154676864293e-06, "reference_active": true, "delta_db": 0.8039115005386381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55667878122912e-07, "output_power": 3.120599262207275e-07, "reference_active": true, "delta_db": 0.865618343852736}], "active_band_count": 3, "active_band_mae_db": 0.5874404035026286}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.00015276264453937597, "log_rms_mae_db": 2.0740983796649495, "rms_envelope_nrmse": 0.20130348753220348, "envelope_correlation": 0.9689131127867855, "rms_gain_error_db": -0.12681609949347739, "reference_mean_power": 0.0006070254152411961, "output_mean_power": 0.0005895562730722539}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.389907119886337e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002863305566547835, "output_power": 0.0002748639622872591, "reference_active": true, "delta_db": -0.1774989482496235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.608243104449444e-05, "output_power": 2.9793471474156622e-05, "reference_active": true, "delta_db": -0.8317468106930491}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0810295137794261e-05, "output_power": 6.688745173520811e-06, "reference_active": true, "delta_db": -2.0849290042364608}], "active_band_count": 3, "active_band_mae_db": 1.0313915877263777}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.00013215266595876854, "log_rms_mae_db": 1.1982271775300795, "rms_envelope_nrmse": 0.1177448528398156, "envelope_correlation": 0.9888251564867047, "rms_gain_error_db": -0.12684388864938584, "reference_mean_power": 0.0006109513084873906, "output_mean_power": 0.0005933653891238769}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 5.083168365591421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002672957849080471, "output_power": 0.000254199034667085, "reference_active": true, "delta_db": -0.21818213283517252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4768224147952787e-05, "output_power": 1.3884813396878541e-05, "reference_active": true, "delta_db": -0.2678822784937822}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.446503663842908e-07, "output_power": 4.942652082636644e-07, "reference_active": true, "delta_db": -2.327169335915045}], "active_band_count": 3, "active_band_mae_db": 0.9377445824146666}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.748427494791344, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6e8bc1c292f16dc6fc44697b294a575630fa11807da0edf1199c095ccd738ac3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9584845830059078, "standard_deviation": 0.006281793520213401, "minimum": 0.949373287562273, "maximum": 0.9671653517109897, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1894483199349963, "standard_deviation": 0.06460747549538842, "minimum": 2.0953424294912444, "maximum": 2.2803518778052725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5137999232464389, "standard_deviation": 0.21811777095622287, "minimum": 0.1397331373263425, "maximum": 0.8414964281318092, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21503946156316434, "standard_deviation": 0.01579461951564001, "minimum": 0.19204551493246141, "maximum": 0.23921729305635966, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9877812515222459, "standard_deviation": 0.0032549291595572022, "minimum": 0.9812337679656508, "maximum": 0.9912491147785838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3306787083195588, "standard_deviation": 0.05450046296894965, "minimum": 1.2659516711171737, "maximum": 1.4341509245241975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.46165472161040744, "standard_deviation": 0.14622334970299308, "minimum": 0.20282797744415246, "maximum": 0.6208229424903222, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11193117029059385, "standard_deviation": 0.014398938172784466, "minimum": 0.09510747671583944, "maximum": 0.1371739975087695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9743925796135641, "standard_deviation": 0.004286055325118062, "minimum": 0.9688758773817282, "maximum": 0.9792940217010841, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0489810098719428, "standard_deviation": 0.028161137865180326, "minimum": 2.0217561184249444, "maximum": 2.10731839278592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6764602984106849, "standard_deviation": 0.31468285266951707, "minimum": 0.07148512637912963, "maximum": 1.0313915877263777, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18287269055725547, "standard_deviation": 0.014858166663222466, "minimum": 0.1647662871894142, "maximum": 0.20139048960717773, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921214510054646, "standard_deviation": 0.002490859958239213, "minimum": 0.9888251564867047, "maximum": 0.9956803012880953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1730265480750321, "standard_deviation": 0.03704665807105775, "minimum": 1.1370720565158488, "maximum": 1.2420216802300867, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7522695194678293, "standard_deviation": 0.36380935811659176, "minimum": 0.10713765953013359, "maximum": 1.197660926276891, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09957883595063206, "standard_deviation": 0.015318310829524166, "minimum": 0.07407618206935615, "maximum": 0.1177448528398156, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9650027452230874, "standard_deviation": 0.004535342472956589, "minimum": 0.9564637406152516, "maximum": 0.9693361371762507, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.27990536738762, "standard_deviation": 0.04372737644687384, "minimum": 2.219585650177883, "maximum": 2.3511776379738434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6962408124939462, "standard_deviation": 0.27072253318314554, "minimum": 0.311385257829933, "maximum": 0.9884888529365033, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19202878475185678, "standard_deviation": 0.011923696891432942, "minimum": 0.1804364826654091, "maximum": 0.2137577514791524, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.988154578128686, "standard_deviation": 0.003143490884690198, "minimum": 0.9830683744274248, "maximum": 0.9912080152696282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5825732211058794, "standard_deviation": 0.0688933495842443, "minimum": 1.498120510550059, "maximum": 1.6651906203378863, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5007283175523087, "standard_deviation": 0.20268557207672205, "minimum": 0.1228761811226297, "maximum": 0.8381081629506552, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10764102704394164, "standard_deviation": 0.012603907758512656, "minimum": 0.09474956608852327, "maximum": 0.12733448972366132, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.32424152193494393, "standard_deviation": 0.09451377657331142, "minimum": 0.17242031094261404, "maximum": 0.46553400530767636, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.4069505941345, "standard_deviation": 0.008527407007863022, "minimum": 6.393183361742689, "maximum": 6.422239916515395, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46081104189023636, "standard_deviation": 0.05585651633598118, "minimum": 0.4034687509809015, "maximum": 0.5815060979787483, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10803655349787658, "standard_deviation": 0.16797461280028164, "minimum": -0.3580691002114371, "maximum": 0.08426463216682822, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9710858886961861, "standard_deviation": 0.0027791461812154915, "minimum": 0.9674198808003104, "maximum": 0.9750683150988277, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1432561539352446, "standard_deviation": 0.03719544115240983, "minimum": 1.0709082276807853, "maximum": 1.20000168939772, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8446099033561134, "standard_deviation": 0.5285992675185768, "minimum": 0.05185003867452187, "maximum": 1.7017553458242631, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1627457417323912, "standard_deviation": 0.007079782778733002, "minimum": 0.15206236153787156, "maximum": 0.1718294491955126, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9901835560435635, "standard_deviation": 0.0019466564148637572, "minimum": 0.9867850584502275, "maximum": 0.993075692423633, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6055443750176992, "standard_deviation": 0.022504152077643967, "minimum": 0.5693110307455349, "maximum": 0.6371107090204401, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9799484218671665, "standard_deviation": 0.6410831962899123, "minimum": 0.34520951573660597, "maximum": 2.1872907922632074, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09192942387811318, "standard_deviation": 0.010177930922556169, "minimum": 0.07644004671655716, "maximum": 0.10739611617916638, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-76324-0024", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.7294917106628418, "predicted_mos": 1.7294917106628418, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "24719b7bc661aac52ef99dcf146dee070769c2de3b46338f1a0a882fb67906fa", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.94309449987486, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1558985592913456, "全帯域_10ms_rms_envelope_nrmse": 0.18578523034942393, "全帯域_10ms_envelope_correlation": 0.9633651967377477, "全帯域_10ms_modulation_mae_db": 0.9359451619742086, "全帯域_40ms_log_rms_mae_db": 0.6329113262028343, "全帯域_40ms_rms_envelope_nrmse": 0.11695930275102405, "全帯域_40ms_envelope_correlation": 0.9844036413187142, "全帯域_40ms_modulation_mae_db": 1.2380065725080978, "300–1000 Hz_10ms_log_rms_mae_db": 1.6588392130021854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17384975246584625, "300–1000 Hz_10ms_envelope_correlation": 0.971460441569652, "300–1000 Hz_10ms_modulation_mae_db": 0.6795467124169742, "300–1000 Hz_40ms_log_rms_mae_db": 1.1409701488905277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09863874938102744, "300–1000 Hz_40ms_envelope_correlation": 0.9903121785574722, "300–1000 Hz_40ms_modulation_mae_db": 1.1041912175675945, "1000–3000 Hz_10ms_log_rms_mae_db": 1.522535903075991, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18680095133268484, "1000–3000 Hz_10ms_envelope_correlation": 0.9695748259475717, "1000–3000 Hz_10ms_modulation_mae_db": 0.891113787342196, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9011592999746089, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10674242891747268, "1000–3000 Hz_40ms_envelope_correlation": 0.9892584275969899, "1000–3000 Hz_40ms_modulation_mae_db": 0.8579946267913806, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3965491866411095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17102000806481973, "3000–6000 Hz_10ms_envelope_correlation": 0.9787500547919186, "3000–6000 Hz_10ms_modulation_mae_db": 0.36978213236942487, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7730324881604737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08928213559263433, "3000–6000 Hz_40ms_envelope_correlation": 0.9942317857224288, "3000–6000 Hz_40ms_modulation_mae_db": 0.6112333658724721, "mr_stft_db": 6.01753406077954, "band_mae_db": 0.29126126211011183, "rms_gain_error_db": 0.08487585123023822, "output_peak": 0.6107200357868366}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.244736994108379, "全帯域_10ms_rms_envelope_nrmse": 0.15975689937954948, "全帯域_10ms_envelope_correlation": 0.9736131055866089, "全帯域_10ms_modulation_mae_db": 0.6450039338870096, "全帯域_40ms_log_rms_mae_db": 0.6221080741725112, "全帯域_40ms_rms_envelope_nrmse": 0.08534747218796054, "全帯域_40ms_envelope_correlation": 0.9922767256522723, "全帯域_40ms_modulation_mae_db": 0.3002771268906317, "300–1000 Hz_10ms_log_rms_mae_db": 1.7563142474620053, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1738295884232783, "300–1000 Hz_10ms_envelope_correlation": 0.9727571105063028, "300–1000 Hz_10ms_modulation_mae_db": 0.419488587619973, "300–1000 Hz_40ms_log_rms_mae_db": 1.2112825500392082, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12257529301701485, "300–1000 Hz_40ms_envelope_correlation": 0.986583384038637, "300–1000 Hz_40ms_modulation_mae_db": 0.5930670250969432, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6087637011639533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23179762541803675, "1000–3000 Hz_10ms_envelope_correlation": 0.9543305268708328, "1000–3000 Hz_10ms_modulation_mae_db": 2.1290143483544526, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9559185729662312, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14423534352533376, "1000–3000 Hz_40ms_envelope_correlation": 0.9809696999671347, "1000–3000 Hz_40ms_modulation_mae_db": 2.1556361493627825, "3000–6000 Hz_10ms_log_rms_mae_db": 1.372719762364299, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16682496193880839, "3000–6000 Hz_10ms_envelope_correlation": 0.9813169949065288, "3000–6000 Hz_10ms_modulation_mae_db": 0.5240194797843767, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8061635053926949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09069696720019585, "3000–6000 Hz_40ms_envelope_correlation": 0.9952324728709576, "3000–6000 Hz_40ms_modulation_mae_db": 0.43485497988888633, "mr_stft_db": 6.021978729772005, "band_mae_db": 0.34243737166017146, "rms_gain_error_db": 0.14478983031651166, "output_peak": 0.5584946873698161}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2304985544176885, "全帯域_10ms_rms_envelope_nrmse": 0.15459663906228396, "全帯域_10ms_envelope_correlation": 0.9743005513050493, "全帯域_10ms_modulation_mae_db": 0.9531305137801379, "全帯域_40ms_log_rms_mae_db": 0.6986522215751512, "全帯域_40ms_rms_envelope_nrmse": 0.08180576799372741, "全帯域_40ms_envelope_correlation": 0.9921020932288298, "全帯域_40ms_modulation_mae_db": 0.6657001104739088, "300–1000 Hz_10ms_log_rms_mae_db": 1.7296055425623775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19504436699679667, "300–1000 Hz_10ms_envelope_correlation": 0.9640792377144419, "300–1000 Hz_10ms_modulation_mae_db": 0.9152210795256718, "300–1000 Hz_40ms_log_rms_mae_db": 1.1341054902009229, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1064529607001828, "300–1000 Hz_40ms_envelope_correlation": 0.9882339183183948, "300–1000 Hz_40ms_modulation_mae_db": 0.5530850136740003, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5564588259873555, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19403847551915626, "1000–3000 Hz_10ms_envelope_correlation": 0.9662909261897177, "1000–3000 Hz_10ms_modulation_mae_db": 1.0446641000322823, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9363248166100386, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10979206979125794, "1000–3000 Hz_40ms_envelope_correlation": 0.9881165085253134, "1000–3000 Hz_40ms_modulation_mae_db": 1.1103076340368443, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3288622519803805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13598868251736637, "3000–6000 Hz_10ms_envelope_correlation": 0.9859940590406067, "3000–6000 Hz_10ms_modulation_mae_db": 0.6178026902972672, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7513661510409245, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06791007422194857, "3000–6000 Hz_40ms_envelope_correlation": 0.9962847172901479, "3000–6000 Hz_40ms_modulation_mae_db": 0.37534365280152526, "mr_stft_db": 6.010769477325706, "band_mae_db": 0.2951041358673115, "rms_gain_error_db": 0.08362849824158447, "output_peak": 0.4104471463237924}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1803377005320883, "全帯域_10ms_rms_envelope_nrmse": 0.19192083398966048, "全帯域_10ms_envelope_correlation": 0.958939160156128, "全帯域_10ms_modulation_mae_db": 0.9150209528162098, "全帯域_40ms_log_rms_mae_db": 0.6951215600598841, "全帯域_40ms_rms_envelope_nrmse": 0.12134664660475014, "全帯域_40ms_envelope_correlation": 0.9827307450471235, "全帯域_40ms_modulation_mae_db": 1.236586622928805, "300–1000 Hz_10ms_log_rms_mae_db": 1.7631836749385985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16866959663714293, "300–1000 Hz_10ms_envelope_correlation": 0.9733110585831513, "300–1000 Hz_10ms_modulation_mae_db": 0.7579180933203366, "300–1000 Hz_40ms_log_rms_mae_db": 1.147425394285086, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10384102112836625, "300–1000 Hz_40ms_envelope_correlation": 0.9895648718979878, "300–1000 Hz_40ms_modulation_mae_db": 0.41366041183083874, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5235478515813814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17349602875304082, "1000–3000 Hz_10ms_envelope_correlation": 0.9729602135540852, "1000–3000 Hz_10ms_modulation_mae_db": 0.9833381201762713, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9039244019994629, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09857378147205645, "1000–3000 Hz_40ms_envelope_correlation": 0.9913085592400642, "1000–3000 Hz_40ms_modulation_mae_db": 1.226955535648767, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3511478887957193, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1539507997108626, "3000–6000 Hz_10ms_envelope_correlation": 0.9821291724066603, "3000–6000 Hz_10ms_modulation_mae_db": 0.6753308066972993, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7608583149115861, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08748970841872214, "3000–6000 Hz_40ms_envelope_correlation": 0.9943329455207904, "3000–6000 Hz_40ms_modulation_mae_db": 0.6831186880880905, "mr_stft_db": 6.024005190237175, "band_mae_db": 0.4023284849286129, "rms_gain_error_db": -0.29183165183113546, "output_peak": 0.3666540718564333}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2841474782700535, "全帯域_10ms_rms_envelope_nrmse": 0.18617904095314097, "全帯域_10ms_envelope_correlation": 0.9659649767095851, "全帯域_10ms_modulation_mae_db": 1.630948777712353, "全帯域_40ms_log_rms_mae_db": 0.7537801992152537, "全帯域_40ms_rms_envelope_nrmse": 0.10869717184444239, "全帯域_40ms_envelope_correlation": 0.988391246123479, "全帯域_40ms_modulation_mae_db": 1.6221037012824755, "300–1000 Hz_10ms_log_rms_mae_db": 1.6722646656049522, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17251252039383355, "300–1000 Hz_10ms_envelope_correlation": 0.9720614020893344, "300–1000 Hz_10ms_modulation_mae_db": 1.0563780527953714, "300–1000 Hz_40ms_log_rms_mae_db": 1.1268012734679567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09698194452438513, "300–1000 Hz_40ms_envelope_correlation": 0.9901787765345006, "300–1000 Hz_40ms_modulation_mae_db": 0.6592901822938194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5223815096190754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18425455100543792, "1000–3000 Hz_10ms_envelope_correlation": 0.9692668685892654, "1000–3000 Hz_10ms_modulation_mae_db": 0.3162280435597203, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9702016736218314, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10731907890121792, "1000–3000 Hz_40ms_envelope_correlation": 0.9889435220463509, "1000–3000 Hz_40ms_modulation_mae_db": 0.7252019018739783, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3917572095593211, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16482213300026533, "3000–6000 Hz_10ms_envelope_correlation": 0.9824216204342582, "3000–6000 Hz_10ms_modulation_mae_db": 0.5501708986423366, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7724851907869794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10940914587696443, "3000–6000 Hz_40ms_envelope_correlation": 0.9929852486393012, "3000–6000 Hz_40ms_modulation_mae_db": 0.5568354788017301, "mr_stft_db": 6.003201081705565, "band_mae_db": 0.34008425438671913, "rms_gain_error_db": 0.2681186608948511, "output_peak": 0.46168677543777475}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1437684750308286, "全帯域_10ms_rms_envelope_nrmse": 0.14548416326697924, "全帯域_10ms_envelope_correlation": 0.9766363782041463, "全帯域_10ms_modulation_mae_db": 0.45703888445126367, "全帯域_40ms_log_rms_mae_db": 0.6032146792147635, "全帯域_40ms_rms_envelope_nrmse": 0.07805760078717967, "全帯域_40ms_envelope_correlation": 0.9930694594320126, "全帯域_40ms_modulation_mae_db": 0.8843194159166242, "300–1000 Hz_10ms_log_rms_mae_db": 1.7298362783432173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19144056031921391, "300–1000 Hz_10ms_envelope_correlation": 0.9655316862733946, "300–1000 Hz_10ms_modulation_mae_db": 0.4853996752280154, "300–1000 Hz_40ms_log_rms_mae_db": 1.1563715863938773, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11965864801933006, "300–1000 Hz_40ms_envelope_correlation": 0.9857975769249948, "300–1000 Hz_40ms_modulation_mae_db": 0.5780185199616942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.52320674648391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17930234838838754, "1000–3000 Hz_10ms_envelope_correlation": 0.9711711225948054, "1000–3000 Hz_10ms_modulation_mae_db": 1.229135708778235, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9259385702548185, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10490811091150515, "1000–3000 Hz_40ms_envelope_correlation": 0.9891140502129452, "1000–3000 Hz_40ms_modulation_mae_db": 1.2851901635802447, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3625835928915855, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15247649152070222, "3000–6000 Hz_10ms_envelope_correlation": 0.982576303203225, "3000–6000 Hz_10ms_modulation_mae_db": 0.8115333024020087, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7733316899628662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07415870830152957, "3000–6000 Hz_40ms_envelope_correlation": 0.9962719370253106, "3000–6000 Hz_40ms_modulation_mae_db": 0.41727308418758313, "mr_stft_db": 6.02268182310306, "band_mae_db": 0.21925070520236736, "rms_gain_error_db": -0.2064588128459296, "output_peak": 0.4292004890448385}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1723733747198215, "全帯域_10ms_rms_envelope_nrmse": 0.17232877635995583, "全帯域_10ms_envelope_correlation": 0.9671720255884069, "全帯域_10ms_modulation_mae_db": 0.1440301660385556, "全帯域_40ms_log_rms_mae_db": 0.633050772727583, "全帯域_40ms_rms_envelope_nrmse": 0.09187249672817592, "全帯域_40ms_envelope_correlation": 0.9897102834686474, "全帯域_40ms_modulation_mae_db": 0.37294354384852085, "300–1000 Hz_10ms_log_rms_mae_db": 1.6341560803841002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1878990895436137, "300–1000 Hz_10ms_envelope_correlation": 0.9679402410722896, "300–1000 Hz_10ms_modulation_mae_db": 0.6979067015791554, "300–1000 Hz_40ms_log_rms_mae_db": 1.1363667300893938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12346680597674306, "300–1000 Hz_40ms_envelope_correlation": 0.986829204822555, "300–1000 Hz_40ms_modulation_mae_db": 0.8136513635952963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5195921211703132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19053190951616458, "1000–3000 Hz_10ms_envelope_correlation": 0.9670017790596653, "1000–3000 Hz_10ms_modulation_mae_db": 0.7730180650202075, "1000–3000 Hz_40ms_log_rms_mae_db": 0.846224296523766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09852993412909215, "1000–3000 Hz_40ms_envelope_correlation": 0.9911157926610524, "1000–3000 Hz_40ms_modulation_mae_db": 0.9553954733755249, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3951100613791094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1594828127786812, "3000–6000 Hz_10ms_envelope_correlation": 0.9809911700812569, "3000–6000 Hz_10ms_modulation_mae_db": 0.4520539608573962, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7962084147623965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09138385408176247, "3000–6000 Hz_40ms_envelope_correlation": 0.9934574519839938, "3000–6000 Hz_40ms_modulation_mae_db": 0.47376192906332065, "mr_stft_db": 6.011042368950034, "band_mae_db": 0.27981093261335965, "rms_gain_error_db": -0.07583994951375421, "output_peak": 0.4807141060453374}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1975327054190361, "全帯域_10ms_rms_envelope_nrmse": 0.17882136694977377, "全帯域_10ms_envelope_correlation": 0.9649942325672679, "全帯域_10ms_modulation_mae_db": 0.346432204621219, "全帯域_40ms_log_rms_mae_db": 0.6059033343480342, "全帯域_40ms_rms_envelope_nrmse": 0.09137293614490617, "全帯域_40ms_envelope_correlation": 0.9899158194620432, "全帯域_40ms_modulation_mae_db": 0.060099221881920704, "300–1000 Hz_10ms_log_rms_mae_db": 1.7051696250282053, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17134200230899285, "300–1000 Hz_10ms_envelope_correlation": 0.9723693348440132, "300–1000 Hz_10ms_modulation_mae_db": 0.912532374952344, "300–1000 Hz_40ms_log_rms_mae_db": 1.0857265034964265, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09823852861103341, "300–1000 Hz_40ms_envelope_correlation": 0.9899475653995258, "300–1000 Hz_40ms_modulation_mae_db": 0.6625819328630503, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5623318158832915, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20448391655645745, "1000–3000 Hz_10ms_envelope_correlation": 0.9623221265949141, "1000–3000 Hz_10ms_modulation_mae_db": 0.8766872490038201, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8979010516039302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10407317074530055, "1000–3000 Hz_40ms_envelope_correlation": 0.9892675710670271, "1000–3000 Hz_40ms_modulation_mae_db": 0.9928964154433985, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3680096349162887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15593349522097985, "3000–6000 Hz_10ms_envelope_correlation": 0.9815850455378742, "3000–6000 Hz_10ms_modulation_mae_db": 0.6251242663240106, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7535989486366308, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08484246508190224, "3000–6000 Hz_40ms_envelope_correlation": 0.9942712656369564, "3000–6000 Hz_40ms_modulation_mae_db": 0.49374928726520345, "mr_stft_db": 6.001187317012512, "band_mae_db": 0.1988006276731825, "rms_gain_error_db": -0.008061835497644526, "output_peak": 0.5014468031541431}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64800, "output_samples": 64800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.945681742141508, "spectral_convergence": 0.6185260123312283, "spectral_convergence_status": "ok", "amplitude_floor": 1.7331847370499167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.700641800079171, "spectral_convergence": 0.5784077131577435, "spectral_convergence_status": "ok", "amplitude_floor": 1.0846095570828337e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.406278640117943, "spectral_convergence": 0.6270051752859177, "spectral_convergence_status": "ok", "amplitude_floor": 4.5533844599041527e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024118742205032635, "output_power": 0.00023597275415659718, "delta_db": -0.09492794121347466, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 9.191122965591724e-05, "output_power": 0.00010247816129394064, "delta_db": 0.4726274806011037, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 7.435432994216677e-05, "output_power": 6.881445937193055e-05, "delta_db": -0.33626561982023273, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.4154339925686723e-05, "output_power": 2.8102335651940916e-05, "delta_db": 0.6574724278867494, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006210431709364446, "output_power": 0.0006440817805108143, "delta_db": 0.15819223615366443, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023011417890304625, "output_power": 0.00022863104040130744, "delta_db": -0.028081866985445958, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29126126211011183, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.08487585123023822, "pitch": null, "output_rms_dbfs": -28.833654207834847, "output_peak": 0.6107200357868366, "reference_rms_dbfs": -28.918530059065088, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.01753406077954}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.0001629307662014554, "log_rms_mae_db": 1.1558985592913456, "rms_envelope_nrmse": 0.18578523034942393, "envelope_correlation": 0.9633651967377477, "rms_gain_error_db": 0.08489700010964471, "reference_mean_power": 0.0012840985730840112, "output_mean_power": 0.0013094474172443084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 7.3397819245367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003947037290166488, "output_power": 0.0004065769105422222, "reference_active": true, "delta_db": 0.12871481996206138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980542214464163e-05, "output_power": 5.477178357264495e-05, "reference_active": true, "delta_db": -1.0533227441003756}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6036857037353872e-05, "output_power": 1.7906419980199954e-05, "reference_active": true, "delta_db": -1.625797921860189}], "active_band_count": 3, "active_band_mae_db": 0.9359451619742086}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.000144561944924798, "log_rms_mae_db": 0.6329113262028343, "rms_envelope_nrmse": 0.11695930275102405, "envelope_correlation": 0.9844036413187142, "rms_gain_error_db": 0.08495333036693908, "reference_mean_power": 0.0012928029739092497, "output_mean_power": 0.0013183407474022216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 6.614408925582191e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037454976712341086, "output_power": 0.0003729959109928656, "reference_active": true, "delta_db": -0.018054605486231366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9110632304140427e-05, "output_power": 2.512196167257525e-05, "reference_active": true, "delta_db": -0.6399809004157931}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1315453531682465e-06, "output_power": 1.0546203723847186e-06, "reference_active": true, "delta_db": -3.0559842116222686}], "active_band_count": 3, "active_band_mae_db": 1.2380065725080978}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 3.733123977436472e-05, "log_rms_mae_db": 1.6588392130021854, "rms_envelope_nrmse": 0.17384975246584625, "envelope_correlation": 0.971460441569652, "rms_gain_error_db": -0.22309851893031318, "reference_mean_power": 0.00013419941652038683, "output_mean_power": 0.00012747962453970446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.2377163404532544e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0020888510118017e-05, "output_power": 1.8324317982976314e-05, "reference_active": true, "delta_db": -0.3845552768586431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.868636044401026e-06, "output_power": 6.5258214208310075e-06, "reference_active": true, "delta_db": -0.8126427279217219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3701828329102724e-06, "output_power": 1.6631167408575158e-06, "reference_active": true, "delta_db": 0.8414421324705577}], "active_band_count": 3, "active_band_mae_db": 0.6795467124169742}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.100732475570721e-05, "log_rms_mae_db": 1.1409701488905277, "rms_envelope_nrmse": 0.09863874938102744, "envelope_correlation": 0.9903121785574722, "rms_gain_error_db": -0.223110882987187, "reference_mean_power": 0.00013520032575819994, "output_mean_power": 0.00012843004943400524}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 4.6212903051863556e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.815754516308902e-05, "output_power": 1.63820001525336e-05, "reference_active": true, "delta_db": -0.4469020741096035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8683197262456875e-06, "output_power": 3.179182845055423e-06, "reference_active": true, "delta_db": -0.8520685610453329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2646920446881671e-07, "output_power": 7.954712445241203e-08, "reference_active": true, "delta_db": -2.0136030175478474}], "active_band_count": 3, "active_band_mae_db": 1.1041912175675945}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 4.662083951365485e-05, "log_rms_mae_db": 1.522535903075991, "rms_envelope_nrmse": 0.18680095133268484, "envelope_correlation": 0.9695748259475717, "rms_gain_error_db": 0.15423529817853665, "reference_mean_power": 0.00017506320579060826, "output_mean_power": 0.00018139211595055805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 9.028764248682407e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2773698500231605e-05, "output_power": 4.117616461802114e-05, "reference_active": true, "delta_db": -0.1653091237805963}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.603092418254862e-06, "output_power": 8.53409353220564e-06, "reference_active": true, "delta_db": 1.1141002255118058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4970630790605404e-06, "output_power": 4.8205508380471076e-06, "reference_active": true, "delta_db": 1.3939320127341859}], "active_band_count": 3, "active_band_mae_db": 0.891113787342196}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.913562930844258e-05, "log_rms_mae_db": 0.9011592999746089, "rms_envelope_nrmse": 0.10674242891747268, "envelope_correlation": 0.9892584275969899, "rms_gain_error_db": 0.15422725087514963, "reference_mean_power": 0.00017624479575510055, "output_mean_power": 0.00018261608454452917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 7.899096535479963e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7672515224694995e-05, "output_power": 3.666480927375998e-05, "reference_active": true, "delta_db": -0.11775186877733909}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.050253319949336e-06, "output_power": 3.884287771112946e-06, "reference_active": true, "delta_db": 1.0497548905276797}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55667878122912e-07, "output_power": 3.534466577052401e-07, "reference_active": true, "delta_db": 1.4064771210691231}], "active_band_count": 3, "active_band_mae_db": 0.8579946267913806}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.00015276264453937597, "log_rms_mae_db": 1.3965491866411095, "rms_envelope_nrmse": 0.17102000806481973, "envelope_correlation": 0.9787500547919186, "rms_gain_error_db": 0.10632849707543003, "reference_mean_power": 0.0006070254152411961, "output_mean_power": 0.0006220706687658903}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.389907119886337e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002863305566547835, "output_power": 0.00031464718884675175, "reference_active": true, "delta_db": 0.4095615840294969}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.608243104449444e-05, "output_power": 3.4735483118016884e-05, "reference_active": true, "delta_db": -0.16522446702917537}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0810295137794261e-05, "output_power": 9.558318908029681e-06, "reference_active": true, "delta_db": -0.5345603460496026}], "active_band_count": 3, "active_band_mae_db": 0.36978213236942487}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.00013215266595876854, "log_rms_mae_db": 0.7730324881604737, "rms_envelope_nrmse": 0.08928213559263433, "envelope_correlation": 0.9942317857224288, "rms_gain_error_db": 0.10654585045838957, "reference_mean_power": 0.0006109513084873906, "output_mean_power": 0.0006261252013035177}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 5.083168365591421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002672957849080471, "output_power": 0.00029297833427209015, "reference_active": true, "delta_db": 0.39843395243413543}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4768224147952787e-05, "output_power": 1.5924641965270736e-05, "reference_active": true, "delta_db": 0.32741401515567375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.446503663842908e-07, "output_power": 6.544730409180536e-07, "reference_active": true, "delta_db": -1.107852130027607}], "active_band_count": 3, "active_band_mae_db": 0.6112333658724721}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.763269327721692, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "aae4f4456628c90d4faa5f72ae638b506a629b56e4ad696a5bc26b92a071f23a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9666147986751071, "standard_deviation": 0.005933014513057703, "minimum": 0.9543305268708328, "maximum": 0.9729602135540852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.542352309370659, "standard_deviation": 0.03170929703806, "minimum": 1.5195921211703132, "maximum": 1.6087637011639533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0303999277833982, "standard_deviation": 0.516607471340136, "minimum": 0.3162280435597203, "maximum": 2.1290143483544526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19308822581117077, "standard_deviation": 0.018233349922662314, "minimum": 0.17349602875304082, "maximum": 0.23179762541803675, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9885117664146097, "standard_deviation": 0.0032363979137090683, "minimum": 0.9809696999671347, "maximum": 0.9913085592400642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9171990854443359, "standard_deviation": 0.03890275782686168, "minimum": 0.846224296523766, "maximum": 0.9702016736218314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.163697237514115, "standard_deviation": 0.4414020178137118, "minimum": 0.7252019018739783, "maximum": 2.1556361493627825, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10927173979915457, "standard_deviation": 0.01468154188885306, "minimum": 0.09852993412909215, "maximum": 0.14423534352533376, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9819705525502911, "standard_deviation": 0.0020245179288130785, "minimum": 0.9787500547919186, "maximum": 0.9859940590406067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3708424485659767, "standard_deviation": 0.02365571833239578, "minimum": 1.3288622519803805, "maximum": 1.3965491866411095, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.578227192171765, "standard_deviation": 0.13665444825011666, "minimum": 0.36978213236942487, "maximum": 0.8115333024020087, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1575624230940607, "standard_deviation": 0.010890905219335579, "minimum": 0.13598868251736637, "maximum": 0.17102000806481973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9946334780862358, "standard_deviation": 0.0012102246246661458, "minimum": 0.9929852486393012, "maximum": 0.9962847172901479, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.773380587956819, "standard_deviation": 0.019381547417154883, "minimum": 0.7513661510409245, "maximum": 0.8061635053926949, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5057713082461014, "standard_deviation": 0.10435068087136508, "minimum": 0.37534365280152526, "maximum": 0.6831186880880905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08689663234695745, "standard_deviation": 0.012388918065692431, "minimum": 0.06791007422194857, "maximum": 0.10940914587696443, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9699388140815725, "standard_deviation": 0.0035815936945361756, "minimum": 0.9640792377144419, "maximum": 0.9733110585831513, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7061711659157053, "standard_deviation": 0.04698191525207749, "minimum": 1.6341560803841002, "maximum": 1.7631836749385985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7405489096797302, "standard_deviation": 0.2183898557469957, "minimum": 0.419488587619973, "maximum": 1.0563780527953714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17932343463608977, "standard_deviation": 0.010359632700532199, "minimum": 0.16866959663714293, "maximum": 0.19504436699679667, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9884309345617586, "standard_deviation": 0.0018178630488192014, "minimum": 0.9857975769249948, "maximum": 0.9903121785574722, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1423812096079249, "standard_deviation": 0.03490336020977675, "minimum": 1.0857265034964265, "maximum": 1.2112825500392082, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6721932083604046, "standard_deviation": 0.208068446181774, "minimum": 0.41366041183083874, "maximum": 1.1041912175675945, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10873174391976037, "standard_deviation": 0.011384883794640432, "minimum": 0.09698194452438513, "maximum": 0.12346680597674306, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.29613472180522954, "standard_deviation": 0.06657859890372043, "minimum": 0.1988006276731825, "maximum": 0.4023284849286129, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.0140500061107, "standard_deviation": 0.008876718613467812, "minimum": 6.001187317012512, "maximum": 6.024005190237175, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47742051437737154, "standard_deviation": 0.07951392729231724, "minimum": 0.3666540718564333, "maximum": 0.6107200357868366, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -9.742612565979449e-05, "standard_deviation": 0.18541725791715308, "minimum": -0.29183165183113546, "maximum": 0.2681186608948511, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9681232033568675, "standard_deviation": 0.0061308495828931026, "minimum": 0.958939160156128, "maximum": 0.9766363782041463, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2011617302236552, "standard_deviation": 0.048202559336252414, "minimum": 1.1437684750308286, "maximum": 1.2841474782700535, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7534438244101196, "standard_deviation": 0.46382912302451396, "minimum": 0.1440301660385556, "maximum": 1.630948777712353, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17185911878884597, "standard_deviation": 0.01686186673362341, "minimum": 0.14548416326697924, "maximum": 0.19192083398966048, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9890750017166403, "standard_deviation": 0.003761065414702213, "minimum": 0.9827307450471235, "maximum": 0.9930694594320126, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.655592770939502, "standard_deviation": 0.054002290517006314, "minimum": 0.6032146792147635, "maximum": 0.7537801992152537, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7975045394663731, "standard_deviation": 0.5428709220817385, "minimum": 0.060099221881920704, "maximum": 1.6221037012824755, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09693242438027078, "standard_deviation": 0.016524164040612535, "minimum": 0.07805760078717967, "maximum": 0.12134664660475014, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-76324-0024", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.9852147102355957, "predicted_mos": 1.9852147102355957, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "4dfd2a63e5c961b5b939244fc11e3092553a482d62891f5effd94550033a28c8", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8720899000763893, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1902424003080259, "全帯域_10ms_rms_envelope_nrmse": 0.15382830693553712, "全帯域_10ms_envelope_correlation": 0.9739023859454373, "全帯域_10ms_modulation_mae_db": 0.41779727577358766, "全帯域_40ms_log_rms_mae_db": 0.6103789064135362, "全帯域_40ms_rms_envelope_nrmse": 0.09208868601961598, "全帯域_40ms_envelope_correlation": 0.9898518894856271, "全帯域_40ms_modulation_mae_db": 1.0749357724998578, "300–1000 Hz_10ms_log_rms_mae_db": 1.923925980566381, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17875934887504222, "300–1000 Hz_10ms_envelope_correlation": 0.9698617262686222, "300–1000 Hz_10ms_modulation_mae_db": 0.6335702706284526, "300–1000 Hz_40ms_log_rms_mae_db": 1.24910609555258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09891819790955617, "300–1000 Hz_40ms_envelope_correlation": 0.990606830772704, "300–1000 Hz_40ms_modulation_mae_db": 0.6690313578938575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8050557640137952, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20222532628002274, "1000–3000 Hz_10ms_envelope_correlation": 0.9666506328618261, "1000–3000 Hz_10ms_modulation_mae_db": 0.8935614766213954, "1000–3000 Hz_40ms_log_rms_mae_db": 1.131635060438447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12038731603844188, "1000–3000 Hz_40ms_envelope_correlation": 0.988267164533844, "1000–3000 Hz_40ms_modulation_mae_db": 0.7338163921817732, "3000–6000 Hz_10ms_log_rms_mae_db": 1.57751139200469, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1478983156334787, "3000–6000 Hz_10ms_envelope_correlation": 0.9838099411190168, "3000–6000 Hz_10ms_modulation_mae_db": 0.2486026886639379, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8819326740449888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06856241941435685, "3000–6000 Hz_40ms_envelope_correlation": 0.9964265407944758, "3000–6000 Hz_40ms_modulation_mae_db": 0.43242780976360945, "mr_stft_db": 6.150174724518706, "band_mae_db": 0.6269311218673873, "rms_gain_error_db": -0.11907262181862005, "output_peak": 0.4546380206545812}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1758444973977136, "全帯域_10ms_rms_envelope_nrmse": 0.15362640356585405, "全帯域_10ms_envelope_correlation": 0.9738903190589507, "全帯域_10ms_modulation_mae_db": 0.5143717823153112, "全帯域_40ms_log_rms_mae_db": 0.5745214257683231, "全帯域_40ms_rms_envelope_nrmse": 0.08932520735439288, "全帯域_40ms_envelope_correlation": 0.9904370055409032, "全帯域_40ms_modulation_mae_db": 0.775123650509601, "300–1000 Hz_10ms_log_rms_mae_db": 1.9242166782752845, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1911041201333224, "300–1000 Hz_10ms_envelope_correlation": 0.9653016610902809, "300–1000 Hz_10ms_modulation_mae_db": 0.9810168540553946, "300–1000 Hz_40ms_log_rms_mae_db": 1.3146005813870902, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10942936387704345, "300–1000 Hz_40ms_envelope_correlation": 0.9880771527916912, "300–1000 Hz_40ms_modulation_mae_db": 0.4591221307177365, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7818660447394532, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20706164068307098, "1000–3000 Hz_10ms_envelope_correlation": 0.962714900385512, "1000–3000 Hz_10ms_modulation_mae_db": 0.6444711451912045, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1002187247288593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13180493776289706, "1000–3000 Hz_40ms_envelope_correlation": 0.9836360962393056, "1000–3000 Hz_40ms_modulation_mae_db": 0.44245070998307884, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6158174969460093, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15721201181006755, "3000–6000 Hz_10ms_envelope_correlation": 0.9813522992260192, "3000–6000 Hz_10ms_modulation_mae_db": 0.7688001590766952, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8788058685113933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06560065408468069, "3000–6000 Hz_40ms_envelope_correlation": 0.9965617049574168, "3000–6000 Hz_40ms_modulation_mae_db": 0.9276004561562766, "mr_stft_db": 6.151681543972008, "band_mae_db": 0.3641463852775862, "rms_gain_error_db": -0.13950558300241808, "output_peak": 0.45776265691214196}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1340948505056632, "全帯域_10ms_rms_envelope_nrmse": 0.15227451483357404, "全帯域_10ms_envelope_correlation": 0.9746731802428509, "全帯域_10ms_modulation_mae_db": 0.8954781390148523, "全帯域_40ms_log_rms_mae_db": 0.6814680533871235, "全帯域_40ms_rms_envelope_nrmse": 0.09321601445157539, "全帯域_40ms_envelope_correlation": 0.9906715566477327, "全帯域_40ms_modulation_mae_db": 1.298990300164369, "300–1000 Hz_10ms_log_rms_mae_db": 1.8327968177303002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16985565472121444, "300–1000 Hz_10ms_envelope_correlation": 0.9727850462163258, "300–1000 Hz_10ms_modulation_mae_db": 1.2456305682889726, "300–1000 Hz_40ms_log_rms_mae_db": 1.2633139329506735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09626937832319675, "300–1000 Hz_40ms_envelope_correlation": 0.9904742086939333, "300–1000 Hz_40ms_modulation_mae_db": 0.6795195529023438, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7780161976083357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1864154059865158, "1000–3000 Hz_10ms_envelope_correlation": 0.9684960504677306, "1000–3000 Hz_10ms_modulation_mae_db": 0.09224919826849519, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0074626792482002, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08852780301377926, "1000–3000 Hz_40ms_envelope_correlation": 0.9922331561568211, "1000–3000 Hz_40ms_modulation_mae_db": 0.4605686238860121, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5971660889502526, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15877331257153887, "3000–6000 Hz_10ms_envelope_correlation": 0.9809384876395862, "3000–6000 Hz_10ms_modulation_mae_db": 0.6441869979165272, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9252682847882009, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08061570684326494, "3000–6000 Hz_40ms_envelope_correlation": 0.9952728846568842, "3000–6000 Hz_40ms_modulation_mae_db": 0.6483465727990093, "mr_stft_db": 6.151958079443273, "band_mae_db": 0.3972312862246612, "rms_gain_error_db": -0.26153852948397927, "output_peak": 0.4058290360118067}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1452389973763994, "全帯域_10ms_rms_envelope_nrmse": 0.14272111560383602, "全帯域_10ms_envelope_correlation": 0.9778732863828453, "全帯域_10ms_modulation_mae_db": 0.31463459163806456, "全帯域_40ms_log_rms_mae_db": 0.6414635157356323, "全帯域_40ms_rms_envelope_nrmse": 0.0717911993586051, "全帯域_40ms_envelope_correlation": 0.9944021350711287, "全帯域_40ms_modulation_mae_db": 0.46164438751490255, "300–1000 Hz_10ms_log_rms_mae_db": 1.9023278637220271, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1746904183526545, "300–1000 Hz_10ms_envelope_correlation": 0.9712308941147324, "300–1000 Hz_10ms_modulation_mae_db": 0.7051962810651734, "300–1000 Hz_40ms_log_rms_mae_db": 1.3315102161120635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10341296884911783, "300–1000 Hz_40ms_envelope_correlation": 0.9895198695579565, "300–1000 Hz_40ms_modulation_mae_db": 1.397374147093726, "1000–3000 Hz_10ms_log_rms_mae_db": 1.790787133511817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19694198100882768, "1000–3000 Hz_10ms_envelope_correlation": 0.9654758434376659, "1000–3000 Hz_10ms_modulation_mae_db": 1.1705726357476098, "1000–3000 Hz_40ms_log_rms_mae_db": 1.034781031644466, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.104604011350082, "1000–3000 Hz_40ms_envelope_correlation": 0.9893182141885325, "1000–3000 Hz_40ms_modulation_mae_db": 1.0876022109560293, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6127052350569082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14330702701628645, "3000–6000 Hz_10ms_envelope_correlation": 0.9845343422738536, "3000–6000 Hz_10ms_modulation_mae_db": 0.5270907468366088, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9050996477978791, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07239923932761079, "3000–6000 Hz_40ms_envelope_correlation": 0.9958337947783855, "3000–6000 Hz_40ms_modulation_mae_db": 0.3991808091924933, "mr_stft_db": 6.136111617673901, "band_mae_db": 0.5097137172783764, "rms_gain_error_db": -0.1996794019949784, "output_peak": 0.4462599146354653}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1230579567532837, "全帯域_10ms_rms_envelope_nrmse": 0.14515179183592056, "全帯域_10ms_envelope_correlation": 0.977271229883706, "全帯域_10ms_modulation_mae_db": 1.0241409712957148, "全帯域_40ms_log_rms_mae_db": 0.5668972484303579, "全帯域_40ms_rms_envelope_nrmse": 0.07905440643378639, "全帯域_40ms_envelope_correlation": 0.9939573439776057, "全帯域_40ms_modulation_mae_db": 0.6560902363267006, "300–1000 Hz_10ms_log_rms_mae_db": 1.9177161448755837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18439652558161604, "300–1000 Hz_10ms_envelope_correlation": 0.9677703609159233, "300–1000 Hz_10ms_modulation_mae_db": 0.9264272929935412, "300–1000 Hz_40ms_log_rms_mae_db": 1.1831480831556727, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09016204805853163, "300–1000 Hz_40ms_envelope_correlation": 0.9917041753695964, "300–1000 Hz_40ms_modulation_mae_db": 0.5304956667736992, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7896941243055466, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17819352232466887, "1000–3000 Hz_10ms_envelope_correlation": 0.9711346712758299, "1000–3000 Hz_10ms_modulation_mae_db": 0.45055807906169837, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9996845415118382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0899446754954947, "1000–3000 Hz_40ms_envelope_correlation": 0.9921714720141829, "1000–3000 Hz_40ms_modulation_mae_db": 0.8515413450354096, "3000–6000 Hz_10ms_log_rms_mae_db": 1.601446228154097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16430110592643163, "3000–6000 Hz_10ms_envelope_correlation": 0.9806001033261285, "3000–6000 Hz_10ms_modulation_mae_db": 1.108527703288155, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8813627013658751, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09186905917082923, "3000–6000 Hz_40ms_envelope_correlation": 0.9952345085273678, "3000–6000 Hz_40ms_modulation_mae_db": 0.654739414897926, "mr_stft_db": 6.155706169083178, "band_mae_db": 0.33682609140345926, "rms_gain_error_db": -0.28813345318459904, "output_peak": 0.4453604364404575}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1358993691980175, "全帯域_10ms_rms_envelope_nrmse": 0.15967650022719573, "全帯域_10ms_envelope_correlation": 0.9716756086924935, "全帯域_10ms_modulation_mae_db": 1.2254436619817912, "全帯域_40ms_log_rms_mae_db": 0.5962778675005257, "全帯域_40ms_rms_envelope_nrmse": 0.07579013272138917, "全帯域_40ms_envelope_correlation": 0.9930478184186308, "全帯域_40ms_modulation_mae_db": 0.9519421182544622, "300–1000 Hz_10ms_log_rms_mae_db": 1.9644849650422058, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1657377349364226, "300–1000 Hz_10ms_envelope_correlation": 0.9741659489989615, "300–1000 Hz_10ms_modulation_mae_db": 0.4717192715595693, "300–1000 Hz_40ms_log_rms_mae_db": 1.3824732919932932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0965971341956461, "300–1000 Hz_40ms_envelope_correlation": 0.990992101206563, "300–1000 Hz_40ms_modulation_mae_db": 0.03855746944028001, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8022067300003488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20495235320429472, "1000–3000 Hz_10ms_envelope_correlation": 0.9628804083914855, "1000–3000 Hz_10ms_modulation_mae_db": 0.7631040070122556, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0421508507455262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10670299108281968, "1000–3000 Hz_40ms_envelope_correlation": 0.9890618832839205, "1000–3000 Hz_40ms_modulation_mae_db": 0.5728498712859028, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6464514664684176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15736806117866198, "3000–6000 Hz_10ms_envelope_correlation": 0.9814148121137144, "3000–6000 Hz_10ms_modulation_mae_db": 0.612211294375601, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9192910637175189, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08044574202095069, "3000–6000 Hz_40ms_envelope_correlation": 0.9955175876685181, "3000–6000 Hz_40ms_modulation_mae_db": 0.5731716925485738, "mr_stft_db": 6.162334820600388, "band_mae_db": 0.2686984083156769, "rms_gain_error_db": -0.08913178456225522, "output_peak": 0.46298226191446085}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1976276233130614, "全帯域_10ms_rms_envelope_nrmse": 0.15267094535190182, "全帯域_10ms_envelope_correlation": 0.9742175490368855, "全帯域_10ms_modulation_mae_db": 0.41943208520991854, "全帯域_40ms_log_rms_mae_db": 0.6016396269582803, "全帯域_40ms_rms_envelope_nrmse": 0.08148508437782782, "全帯域_40ms_envelope_correlation": 0.9919875082606676, "全帯域_40ms_modulation_mae_db": 0.7627899518101846, "300–1000 Hz_10ms_log_rms_mae_db": 1.8971998023293317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17187578661348127, "300–1000 Hz_10ms_envelope_correlation": 0.9721580624440513, "300–1000 Hz_10ms_modulation_mae_db": 0.5487645727384234, "300–1000 Hz_40ms_log_rms_mae_db": 1.234434969809416, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08978687327900794, "300–1000 Hz_40ms_envelope_correlation": 0.9923272662315771, "300–1000 Hz_40ms_modulation_mae_db": 0.9827154204421555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8086900712706842, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19871506057474955, "1000–3000 Hz_10ms_envelope_correlation": 0.9642515227199601, "1000–3000 Hz_10ms_modulation_mae_db": 0.713750704868631, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0372790470935827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09119416079151574, "1000–3000 Hz_40ms_envelope_correlation": 0.9918314659399003, "1000–3000 Hz_40ms_modulation_mae_db": 0.6932124054913474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5991402220538409, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16517976345161736, "3000–6000 Hz_10ms_envelope_correlation": 0.9797127177292059, "3000–6000 Hz_10ms_modulation_mae_db": 0.369042711323808, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9006318239756677, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07490967237723989, "3000–6000 Hz_40ms_envelope_correlation": 0.9956711809623944, "3000–6000 Hz_40ms_modulation_mae_db": 0.623917969467179, "mr_stft_db": 6.164174692447673, "band_mae_db": 0.5577163694188271, "rms_gain_error_db": -0.09727951614329655, "output_peak": 0.5374680731533127}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1588348845178151, "全帯域_10ms_rms_envelope_nrmse": 0.19088147072169925, "全帯域_10ms_envelope_correlation": 0.9610766256991564, "全帯域_10ms_modulation_mae_db": 0.4351031225999469, "全帯域_40ms_log_rms_mae_db": 0.63118001181986, "全帯域_40ms_rms_envelope_nrmse": 0.0887924702659556, "全帯域_40ms_envelope_correlation": 0.9909218679801869, "全帯域_40ms_modulation_mae_db": 0.19728556593281008, "300–1000 Hz_10ms_log_rms_mae_db": 1.993996687975477, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19869159750775386, "300–1000 Hz_10ms_envelope_correlation": 0.9626154878115943, "300–1000 Hz_10ms_modulation_mae_db": 0.8512706136412956, "300–1000 Hz_40ms_log_rms_mae_db": 1.3742631990787044, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10851081454043014, "300–1000 Hz_40ms_envelope_correlation": 0.9878720009341608, "300–1000 Hz_40ms_modulation_mae_db": 0.5919775318644168, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8001900040251617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20000598593824775, "1000–3000 Hz_10ms_envelope_correlation": 0.9641185363855703, "1000–3000 Hz_10ms_modulation_mae_db": 0.8580918338714717, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0962165726883344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11110985987963667, "1000–3000 Hz_40ms_envelope_correlation": 0.9878607734445874, "1000–3000 Hz_40ms_modulation_mae_db": 0.6347806941615847, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6874462741990763, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23039464620068045, "3000–6000 Hz_10ms_envelope_correlation": 0.961678972222614, "3000–6000 Hz_10ms_modulation_mae_db": 0.7547347162011899, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9985529768696525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11327522455552441, "3000–6000 Hz_40ms_envelope_correlation": 0.990685868537229, "3000–6000 Hz_40ms_modulation_mae_db": 0.7494927400914381, "mr_stft_db": 6.147318376248374, "band_mae_db": 0.43066911364822724, "rms_gain_error_db": 0.08110911526971963, "output_peak": 0.5883256922828451}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64800, "output_samples": 64800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.157415436104894, "spectral_convergence": 0.6189817202198741, "spectral_convergence_status": "ok", "amplitude_floor": 1.7331847370499167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.841528665804624, "spectral_convergence": 0.5801058377589344, "spectral_convergence_status": "ok", "amplitude_floor": 1.0846095570828337e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.4515800716466, "spectral_convergence": 0.611153603228024, "spectral_convergence_status": "ok", "amplitude_floor": 4.5533844599041527e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024118742205032635, "output_power": 0.00016276490394821403, "delta_db": -1.7079388930668584, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 9.191122965591724e-05, "output_power": 0.00010627720286796806, "delta_db": 0.6307153914786889, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 7.435432994216677e-05, "output_power": 6.535612469151741e-05, "delta_db": -0.5601997157012925, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.4154339925686723e-05, "output_power": 2.6190389360312216e-05, "delta_db": 0.3514678106330449, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006210431709364446, "output_power": 0.0006338514288643961, "delta_db": 0.08865683085337933, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023011417890304625, "output_power": 0.00025363208290910135, "delta_db": 0.42260808947105977, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6269311218673873, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11907262181862005, "pitch": null, "output_rms_dbfs": -29.037602680883708, "output_peak": 0.4546380206545812, "reference_rms_dbfs": -28.918530059065088, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.150174724518706}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.0001629307662014554, "log_rms_mae_db": 1.1902424003080259, "rms_envelope_nrmse": 0.15382830693553712, "envelope_correlation": 0.9739023859454373, "rms_gain_error_db": -0.118636819029666, "reference_mean_power": 0.0012840985730840112, "output_mean_power": 0.0012494954571324684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 7.3397819245367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003947037290166488, "output_power": 0.0003778806858018224, "reference_active": true, "delta_db": -0.18916534600477813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980542214464163e-05, "output_power": 5.685301937266219e-05, "reference_active": true, "delta_db": -0.8913562349045812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6036857037353872e-05, "output_power": 2.502081972582154e-05, "reference_active": true, "delta_db": -0.1728702464114036}], "active_band_count": 3, "active_band_mae_db": 0.41779727577358766}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.000144561944924798, "log_rms_mae_db": 0.6103789064135362, "rms_envelope_nrmse": 0.09208868601961598, "envelope_correlation": 0.9898518894856271, "rms_gain_error_db": -0.11856282214637157, "reference_mean_power": 0.0012928029739092497, "output_mean_power": 0.0012579867309136994}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 6.614408925582191e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037454976712341086, "output_power": 0.00035193154809281055, "reference_active": true, "delta_db": -0.27051331475698254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9110632304140427e-05, "output_power": 2.3260285164528575e-05, "reference_active": true, "delta_db": -0.9743660398099361}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1315453531682465e-06, "output_power": 1.3511444489905243e-06, "reference_active": true, "delta_db": -1.9799279629326552}], "active_band_count": 3, "active_band_mae_db": 1.0749357724998578}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 3.733123977436472e-05, "log_rms_mae_db": 1.923925980566381, "rms_envelope_nrmse": 0.17875934887504222, "envelope_correlation": 0.9698617262686222, "rms_gain_error_db": -0.24202378818650463, "reference_mean_power": 0.00013419941652038683, "output_mean_power": 0.00012692531466976432}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.2377163404532544e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0020888510118017e-05, "output_power": 1.7482275877015452e-05, "reference_active": true, "delta_db": -0.5888537790564699}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.868636044401026e-06, "output_power": 6.386429874216393e-06, "reference_active": true, "delta_db": -0.9064131013860298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3701828329102724e-06, "output_power": 1.5042600269953465e-06, "reference_active": true, "delta_db": 0.40544393144285823}], "active_band_count": 3, "active_band_mae_db": 0.6335702706284526}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.100732475570721e-05, "log_rms_mae_db": 1.24910609555258, "rms_envelope_nrmse": 0.09891819790955617, "envelope_correlation": 0.990606830772704, "rms_gain_error_db": -0.24202418192482783, "reference_mean_power": 0.00013520032575819994, "output_mean_power": 0.00012787195935328908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 4.6212903051863556e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.815754516308902e-05, "output_power": 1.565176099602594e-05, "reference_active": true, "delta_db": -0.6449392549270705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8683197262456875e-06, "output_power": 2.950462224229088e-06, "reference_active": true, "delta_db": -1.1763230382275356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2646920446881671e-07, "output_power": 1.211718137574134e-07, "reference_active": true, "delta_db": -0.18583178052696608}], "active_band_count": 3, "active_band_mae_db": 0.6690313578938575}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 4.662083951365485e-05, "log_rms_mae_db": 1.8050557640137952, "rms_envelope_nrmse": 0.20222532628002274, "envelope_correlation": 0.9666506328618261, "rms_gain_error_db": 0.3829386150705727, "reference_mean_power": 0.00017506320579060826, "output_mean_power": 0.0001912003745044842}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 9.028764248682407e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2773698500231605e-05, "output_power": 4.998347700091457e-05, "reference_active": true, "delta_db": 0.6764965978315647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.603092418254862e-06, "output_power": 7.729522013101926e-06, "reference_active": true, "delta_db": 0.6840526256534901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4970630790605404e-06, "output_power": 4.739330403591195e-06, "reference_active": true, "delta_db": 1.3201352063791318}], "active_band_count": 3, "active_band_mae_db": 0.8935614766213954}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.913562930844258e-05, "log_rms_mae_db": 1.131635060438447, "rms_envelope_nrmse": 0.12038731603844188, "envelope_correlation": 0.988267164533844, "rms_gain_error_db": 0.3830820132029627, "reference_mean_power": 0.00017624479575510055, "output_mean_power": 0.00019249723826032678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 7.899096535479963e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7672515224694995e-05, "output_power": 4.540775216202056e-05, "reference_active": true, "delta_db": 0.8110538631707762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.050253319949336e-06, "output_power": 3.830904747237566e-06, "reference_active": true, "delta_db": 0.989654451990234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55667878122912e-07, "output_power": 2.8038208001404015e-07, "reference_active": true, "delta_db": 0.4007408613843089}], "active_band_count": 3, "active_band_mae_db": 0.7338163921817732}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.00015276264453937597, "log_rms_mae_db": 1.57751139200469, "rms_envelope_nrmse": 0.1478983156334787, "envelope_correlation": 0.9838099411190168, "rms_gain_error_db": 0.11554293243329467, "reference_mean_power": 0.0006070254152411961, "output_mean_power": 0.0006233919185946291}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.389907119886337e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002863305566547835, "output_power": 0.0002991854287494846, "reference_active": true, "delta_db": 0.1907274065105175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.608243104449444e-05, "output_power": 3.60819802317068e-05, "reference_active": true, "delta_db": -5.426095843735021e-05}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0810295137794261e-05, "output_power": 1.228401083620237e-05, "reference_active": true, "delta_db": 0.5550263985228588}], "active_band_count": 3, "active_band_mae_db": 0.2486026886639379}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.00013215266595876854, "log_rms_mae_db": 0.8819326740449888, "rms_envelope_nrmse": 0.06856241941435685, "envelope_correlation": 0.9964265407944758, "rms_gain_error_db": 0.11628281460822766, "reference_mean_power": 0.0006109513084873906, "output_mean_power": 0.0006275305606398273}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 5.083168365591421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002672957849080471, "output_power": 0.00028225946415662956, "reference_active": true, "delta_db": 0.23656402479486718}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4768224147952787e-05, "output_power": 1.4855619081909742e-05, "reference_active": true, "delta_db": 0.02562479647423238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.446503663842908e-07, "output_power": 6.655298116584012e-07, "reference_active": true, "delta_db": -1.0350946080217287}], "active_band_count": 3, "active_band_mae_db": 0.43242780976360945}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.758472867062703, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "580e5e1b76d5c37b18427e84e823b016e20ad5f62cb95fbc6a8c65098dec57f6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9657153207406975, "standard_deviation": 0.0029219771055586483, "minimum": 0.962714900385512, "maximum": 0.9711346712758299, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7945632586843927, "standard_deviation": 0.011165856642717294, "minimum": 1.7780161976083357, "maximum": 1.8086900712706842, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6982948850803452, "standard_deviation": 0.32131679507402017, "minimum": 0.09224919826849519, "maximum": 1.1705726357476098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1968139095000498, "standard_deviation": 0.00977753632044072, "minimum": 0.17819352232466887, "maximum": 0.20706164068307098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9892975282251368, "standard_deviation": 0.0028916625670449645, "minimum": 0.9836360962393056, "maximum": 0.9922331561568211, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0561785635124068, "standard_deviation": 0.04752878169354959, "minimum": 0.9996845415118382, "maximum": 1.131635060438447, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6846027816226422, "standard_deviation": 0.2126610354815457, "minimum": 0.44245070998307884, "maximum": 1.0876022109560293, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10553446942683338, "standard_deviation": 0.015501030205031439, "minimum": 0.08852780301377926, "maximum": 0.13180493776289706, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792552094562674, "standard_deviation": 0.007284544176767247, "minimum": 0.961678972222614, "maximum": 0.9845343422738536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6172105504791614, "standard_deviation": 0.03456501279965821, "minimum": 1.57751139200469, "maximum": 1.6874462741990763, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6291496272103154, "standard_deviation": 0.2641638696720062, "minimum": 0.2486026886639379, "maximum": 1.108527703288155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16555428047359538, "standard_deviation": 0.027242944765269513, "minimum": 0.14330702701628645, "maximum": 0.23039464620068045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995150508860334, "standard_deviation": 0.0018688232680639028, "minimum": 0.990685868537229, "maximum": 0.9965617049574168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9113681301338971, "standard_deviation": 0.0393497495539837, "minimum": 0.8788058685113933, "maximum": 0.9985529768696525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6261096831145632, "standard_deviation": 0.16871396847971704, "minimum": 0.3991808091924933, "maximum": 0.9276004561562766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08095971472430719, "standard_deviation": 0.015414125745186578, "minimum": 0.06560065408468069, "maximum": 0.11327522455552441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9694861484825614, "standard_deviation": 0.0039779287343443825, "minimum": 0.9626154878115943, "maximum": 0.9741659489989615, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.919583117564574, "standard_deviation": 0.04770143712143584, "minimum": 1.8327968177303002, "maximum": 1.993996687975477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7954494656213529, "standard_deviation": 0.25561540829275176, "minimum": 0.4717192715595693, "maximum": 1.2456305682889726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17938889834018842, "standard_deviation": 0.011282558208645205, "minimum": 0.1657377349364226, "maximum": 0.19869159750775386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9901967006947727, "standard_deviation": 0.0016055475890371877, "minimum": 0.9878720009341608, "maximum": 0.9923272662315771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2916062962549368, "standard_deviation": 0.0705574779928107, "minimum": 1.1831480831556727, "maximum": 1.3824732919932932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.668599159641027, "standard_deviation": 0.3956329567096092, "minimum": 0.03855746944028001, "maximum": 1.397374147093726, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09913584737906625, "standard_deviation": 0.007509819581023346, "minimum": 0.08978687327900794, "maximum": 0.10942936387704345, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.43649156167927516, "standard_deviation": 0.12025033835779506, "minimum": 0.2686984083156769, "maximum": 0.6269311218673873, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.152432502998437, "standard_deviation": 0.008824187276304032, "minimum": 6.136111617673901, "maximum": 6.164174692447673, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4748282615006339, "standard_deviation": 0.058663711664851285, "minimum": 0.4058290360118067, "maximum": 0.5883256922828451, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13915397186505338, "standard_deviation": 0.11588855148620959, "minimum": -0.28813345318459904, "maximum": 0.08110911526971963, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9730725231177907, "standard_deviation": 0.005233955637738916, "minimum": 0.9610766256991564, "maximum": 0.9778732863828453, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1576050724212474, "standard_deviation": 0.027701160142459623, "minimum": 1.1230579567532837, "maximum": 1.1976276233130614, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6558002037286483, "standard_deviation": 0.3412649911491403, "minimum": 0.31463459163806456, "maximum": 1.2254436619817912, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15635388113443982, "standard_deviation": 0.014921490950602714, "minimum": 0.14272111560383602, "maximum": 0.19088147072169925, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9919096406728103, "standard_deviation": 0.0017162529874269908, "minimum": 0.9898518894856271, "maximum": 0.9944021350711287, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6129783320017048, "standard_deviation": 0.03751224073775013, "minimum": 0.5668972484303579, "maximum": 0.6814680533871235, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.772350247876611, "standard_deviation": 0.34682518974299575, "minimum": 0.19728556593281008, "maximum": 1.298990300164369, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08394290012289354, "standard_deviation": 0.008008046984018635, "minimum": 0.0717911993586051, "maximum": 0.09321601445157539, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-76324-0024", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.0508463382720947, "predicted_mos": 2.0508463382720947, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "0c5bcd2d2748b37539cca73af371b323aa1557a41e3a9f5968ce8209c72b44ba", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.691376100294292, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1624127640494024, "全帯域_10ms_rms_envelope_nrmse": 0.13109841683152823, "全帯域_10ms_envelope_correlation": 0.9810939887176288, "全帯域_10ms_modulation_mae_db": 0.6110473475394369, "全帯域_40ms_log_rms_mae_db": 0.6270319494496428, "全帯域_40ms_rms_envelope_nrmse": 0.0676722168328213, "全帯域_40ms_envelope_correlation": 0.9948031874479069, "全帯域_40ms_modulation_mae_db": 0.5716980578785659, "300–1000 Hz_10ms_log_rms_mae_db": 1.9170980684556411, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17569046803231453, "300–1000 Hz_10ms_envelope_correlation": 0.9708693917900899, "300–1000 Hz_10ms_modulation_mae_db": 0.5187614953116649, "300–1000 Hz_40ms_log_rms_mae_db": 1.293427409610041, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10095308494548068, "300–1000 Hz_40ms_envelope_correlation": 0.9900104689094117, "300–1000 Hz_40ms_modulation_mae_db": 0.4189594478851224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7749672226381787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2007800234471254, "1000–3000 Hz_10ms_envelope_correlation": 0.9634736819142173, "1000–3000 Hz_10ms_modulation_mae_db": 0.4447512661636786, "1000–3000 Hz_40ms_log_rms_mae_db": 0.976954592933921, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10303532857040322, "1000–3000 Hz_40ms_envelope_correlation": 0.9908686728938468, "1000–3000 Hz_40ms_modulation_mae_db": 0.3313826609268167, "3000–6000 Hz_10ms_log_rms_mae_db": 1.556852265270956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15042996829791172, "3000–6000 Hz_10ms_envelope_correlation": 0.9828102910241703, "3000–6000 Hz_10ms_modulation_mae_db": 0.6026563918616145, "3000–6000 Hz_40ms_log_rms_mae_db": 0.87639595702749, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07604080501434216, "3000–6000 Hz_40ms_envelope_correlation": 0.9953468046604435, "3000–6000 Hz_40ms_modulation_mae_db": 0.11308296495771963, "mr_stft_db": 6.1141193093863215, "band_mae_db": 0.3314555586137491, "rms_gain_error_db": -0.17826556925565396, "output_peak": 0.4655473929686023}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.136988451500577, "全帯域_10ms_rms_envelope_nrmse": 0.15425762864931805, "全帯域_10ms_envelope_correlation": 0.9736131704090951, "全帯域_10ms_modulation_mae_db": 0.8905340043648078, "全帯域_40ms_log_rms_mae_db": 0.5800745335791592, "全帯域_40ms_rms_envelope_nrmse": 0.0869324889643753, "全帯域_40ms_envelope_correlation": 0.9908966174079598, "全帯域_40ms_modulation_mae_db": 0.5784233684221148, "300–1000 Hz_10ms_log_rms_mae_db": 1.8875080439980103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1597088435354829, "300–1000 Hz_10ms_envelope_correlation": 0.9758871793077945, "300–1000 Hz_10ms_modulation_mae_db": 0.418372923007148, "300–1000 Hz_40ms_log_rms_mae_db": 1.2187137170824327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08408762644396314, "300–1000 Hz_40ms_envelope_correlation": 0.9927524950448413, "300–1000 Hz_40ms_modulation_mae_db": 0.5763127206841983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7460409358186144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2068053540872253, "1000–3000 Hz_10ms_envelope_correlation": 0.9611435828238423, "1000–3000 Hz_10ms_modulation_mae_db": 0.6055552694263816, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0136697149848695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10716567313061381, "1000–3000 Hz_40ms_envelope_correlation": 0.9900812054881643, "1000–3000 Hz_40ms_modulation_mae_db": 0.7147765999051291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5379209283070143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16945584960629967, "3000–6000 Hz_10ms_envelope_correlation": 0.9780696488861974, "3000–6000 Hz_10ms_modulation_mae_db": 0.6282046356942609, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8727074010193214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09621421102314086, "3000–6000 Hz_40ms_envelope_correlation": 0.992698797843321, "3000–6000 Hz_40ms_modulation_mae_db": 0.5729904576125588, "mr_stft_db": 6.102168989520595, "band_mae_db": 0.2634770370367416, "rms_gain_error_db": -0.1283672630562387, "output_peak": 0.5357188640869527}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.133033343766842, "全帯域_10ms_rms_envelope_nrmse": 0.16353172955914086, "全帯域_10ms_envelope_correlation": 0.9719299162518983, "全帯域_10ms_modulation_mae_db": 1.6378246829329475, "全帯域_40ms_log_rms_mae_db": 0.5793927886130206, "全帯域_40ms_rms_envelope_nrmse": 0.0923012403555786, "全帯域_40ms_envelope_correlation": 0.9903412962964396, "全帯域_40ms_modulation_mae_db": 1.6889483761652286, "300–1000 Hz_10ms_log_rms_mae_db": 1.8979492501546378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18106305303094522, "300–1000 Hz_10ms_envelope_correlation": 0.9691790902634988, "300–1000 Hz_10ms_modulation_mae_db": 0.9458491536168947, "300–1000 Hz_40ms_log_rms_mae_db": 1.2522385522635813, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10641425523170756, "300–1000 Hz_40ms_envelope_correlation": 0.9891456705130496, "300–1000 Hz_40ms_modulation_mae_db": 1.6532683471862162, "1000–3000 Hz_10ms_log_rms_mae_db": 1.721173272099172, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18543724049814517, "1000–3000 Hz_10ms_envelope_correlation": 0.9687773047761925, "1000–3000 Hz_10ms_modulation_mae_db": 0.3193265912991153, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9463809914744639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08362108814263451, "1000–3000 Hz_40ms_envelope_correlation": 0.9930746204376366, "1000–3000 Hz_40ms_modulation_mae_db": 0.33308794568147954, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5692318720534233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.162634995509429, "3000–6000 Hz_10ms_envelope_correlation": 0.9798248272215447, "3000–6000 Hz_10ms_modulation_mae_db": 0.25552717185341706, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8350130568322482, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07309911851036743, "3000–6000 Hz_40ms_envelope_correlation": 0.9957360123555026, "3000–6000 Hz_40ms_modulation_mae_db": 0.3763396669618599, "mr_stft_db": 6.102620151634929, "band_mae_db": 0.39868744361086833, "rms_gain_error_db": 0.1187738761816385, "output_peak": 0.38986589723371945}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1511781367145804, "全帯域_10ms_rms_envelope_nrmse": 0.1729987503779084, "全帯域_10ms_envelope_correlation": 0.9678806921044483, "全帯域_10ms_modulation_mae_db": 0.6151132941223153, "全帯域_40ms_log_rms_mae_db": 0.6072885350663665, "全帯域_40ms_rms_envelope_nrmse": 0.09241029843119578, "全帯域_40ms_envelope_correlation": 0.9901241425115747, "全帯域_40ms_modulation_mae_db": 0.9842772074439231, "300–1000 Hz_10ms_log_rms_mae_db": 1.936356342915686, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18656870151801666, "300–1000 Hz_10ms_envelope_correlation": 0.9675089955491464, "300–1000 Hz_10ms_modulation_mae_db": 0.8425571742039494, "300–1000 Hz_40ms_log_rms_mae_db": 1.303096068938001, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11711204549275189, "300–1000 Hz_40ms_envelope_correlation": 0.9872802076345369, "300–1000 Hz_40ms_modulation_mae_db": 0.3635387796129896, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6950580154614787, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18986053074382223, "1000–3000 Hz_10ms_envelope_correlation": 0.9671666190531556, "1000–3000 Hz_10ms_modulation_mae_db": 0.6581916504564136, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0416191762655085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10302138161330555, "1000–3000 Hz_40ms_envelope_correlation": 0.9894428502302037, "1000–3000 Hz_40ms_modulation_mae_db": 0.7712860754593743, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5810599863479662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18387758335735133, "3000–6000 Hz_10ms_envelope_correlation": 0.9785496853403737, "3000–6000 Hz_10ms_modulation_mae_db": 0.8806458673448124, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8703153129535443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10798535775035659, "3000–6000 Hz_40ms_envelope_correlation": 0.993973983908412, "3000–6000 Hz_40ms_modulation_mae_db": 0.5068410211089758, "mr_stft_db": 6.095871944639534, "band_mae_db": 0.545580435045692, "rms_gain_error_db": 0.0447593011674552, "output_peak": 0.3943629387839083}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.148932254466146, "全帯域_10ms_rms_envelope_nrmse": 0.14485261249981926, "全帯域_10ms_envelope_correlation": 0.9767772813752804, "全帯域_10ms_modulation_mae_db": 0.5614371581855365, "全帯域_40ms_log_rms_mae_db": 0.6045935673545205, "全帯域_40ms_rms_envelope_nrmse": 0.07999708721208432, "全帯域_40ms_envelope_correlation": 0.9924914576369387, "全帯域_40ms_modulation_mae_db": 0.9044593475093997, "300–1000 Hz_10ms_log_rms_mae_db": 1.8656613512055356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1775283780906089, "300–1000 Hz_10ms_envelope_correlation": 0.9707326819565345, "300–1000 Hz_10ms_modulation_mae_db": 0.5760220274495019, "300–1000 Hz_40ms_log_rms_mae_db": 1.1840684818476683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09797097430948366, "300–1000 Hz_40ms_envelope_correlation": 0.9916153045042655, "300–1000 Hz_40ms_modulation_mae_db": 0.30494881168929194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7226337515019354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1825932671983695, "1000–3000 Hz_10ms_envelope_correlation": 0.9702494620060852, "1000–3000 Hz_10ms_modulation_mae_db": 0.2761334410340287, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9742864443090433, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0914754527744756, "1000–3000 Hz_40ms_envelope_correlation": 0.9918364004930483, "1000–3000 Hz_40ms_modulation_mae_db": 0.7821067104149773, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5596485393971813, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15479215199208224, "3000–6000 Hz_10ms_envelope_correlation": 0.9817405175568823, "3000–6000 Hz_10ms_modulation_mae_db": 0.4414565902284471, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8663146184480968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08528608788410694, "3000–6000 Hz_40ms_envelope_correlation": 0.9941440496227237, "3000–6000 Hz_40ms_modulation_mae_db": 0.46311330655468225, "mr_stft_db": 6.095630054057668, "band_mae_db": 0.4764071300658992, "rms_gain_error_db": -0.17023241240626052, "output_peak": 0.47474238543475705}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1628522709147553, "全帯域_10ms_rms_envelope_nrmse": 0.14934952500132942, "全帯域_10ms_envelope_correlation": 0.975275287740149, "全帯域_10ms_modulation_mae_db": 0.6063014964078728, "全帯域_40ms_log_rms_mae_db": 0.5655128408374821, "全帯域_40ms_rms_envelope_nrmse": 0.06882364803610949, "全帯域_40ms_envelope_correlation": 0.9945854135564952, "全帯域_40ms_modulation_mae_db": 0.31297010209335313, "300–1000 Hz_10ms_log_rms_mae_db": 1.8542555880864295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19121255181183522, "300–1000 Hz_10ms_envelope_correlation": 0.965390331737787, "300–1000 Hz_10ms_modulation_mae_db": 1.0861009307691922, "300–1000 Hz_40ms_log_rms_mae_db": 1.233254415585249, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11484024614410893, "300–1000 Hz_40ms_envelope_correlation": 0.986397221034594, "300–1000 Hz_40ms_modulation_mae_db": 0.5665242417575688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7010171469943245, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17826258161197195, "1000–3000 Hz_10ms_envelope_correlation": 0.9710267330301768, "1000–3000 Hz_10ms_modulation_mae_db": 0.21596982368121628, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9401449377503598, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08527306809855453, "1000–3000 Hz_40ms_envelope_correlation": 0.9929537851694658, "1000–3000 Hz_40ms_modulation_mae_db": 0.5799048354736039, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5362440812182636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1587568314443207, "3000–6000 Hz_10ms_envelope_correlation": 0.9808460315987242, "3000–6000 Hz_10ms_modulation_mae_db": 0.39318033870953295, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8513738983457532, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07352068299067414, "3000–6000 Hz_40ms_envelope_correlation": 0.9960090118430849, "3000–6000 Hz_40ms_modulation_mae_db": 0.27485879986209477, "mr_stft_db": 6.1066355700464365, "band_mae_db": 0.3480890926591147, "rms_gain_error_db": -0.17073811748333267, "output_peak": 0.47466303663444137}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1949160753593766, "全帯域_10ms_rms_envelope_nrmse": 0.17891479373722072, "全帯域_10ms_envelope_correlation": 0.9644082184281815, "全帯域_10ms_modulation_mae_db": 0.4202079115285277, "全帯域_40ms_log_rms_mae_db": 0.6842116436398694, "全帯域_40ms_rms_envelope_nrmse": 0.10973108360366636, "全帯域_40ms_envelope_correlation": 0.9853473980890465, "全帯域_40ms_modulation_mae_db": 0.8098160110441183, "300–1000 Hz_10ms_log_rms_mae_db": 1.8358332037854324, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17112127198364543, "300–1000 Hz_10ms_envelope_correlation": 0.9726503471273998, "300–1000 Hz_10ms_modulation_mae_db": 0.5024669988901619, "300–1000 Hz_40ms_log_rms_mae_db": 1.1974999988457473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10606496284682387, "300–1000 Hz_40ms_envelope_correlation": 0.9892956507228727, "300–1000 Hz_40ms_modulation_mae_db": 0.6349617214457913, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6730393367524672, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18865529567990075, "1000–3000 Hz_10ms_envelope_correlation": 0.9678867998483689, "1000–3000 Hz_10ms_modulation_mae_db": 0.29652574456537034, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9656297510296229, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08972641978115328, "1000–3000 Hz_40ms_envelope_correlation": 0.9920029823554752, "1000–3000 Hz_40ms_modulation_mae_db": 0.5544395227081291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6153184402826442, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1757136200703938, "3000–6000 Hz_10ms_envelope_correlation": 0.9770538210868083, "3000–6000 Hz_10ms_modulation_mae_db": 1.0281862123293646, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8731098552147577, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10025728436217395, "3000–6000 Hz_40ms_envelope_correlation": 0.9921451619641944, "3000–6000 Hz_40ms_modulation_mae_db": 0.8670644887878124, "mr_stft_db": 6.108325592060214, "band_mae_db": 0.49549590068953836, "rms_gain_error_db": -0.13363530191837392, "output_peak": 0.517100843940648}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1219762552244805, "全帯域_10ms_rms_envelope_nrmse": 0.14411474832035998, "全帯域_10ms_envelope_correlation": 0.9769867081512091, "全帯域_10ms_modulation_mae_db": 0.2661827455798065, "全帯域_40ms_log_rms_mae_db": 0.6222935520920846, "全帯域_40ms_rms_envelope_nrmse": 0.07825932206151037, "全帯域_40ms_envelope_correlation": 0.9930371292136828, "全帯域_40ms_modulation_mae_db": 0.6552703969873873, "300–1000 Hz_10ms_log_rms_mae_db": 1.8359510129157606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18090764497609363, "300–1000 Hz_10ms_envelope_correlation": 0.9694104552146061, "300–1000 Hz_10ms_modulation_mae_db": 0.30525295046191425, "300–1000 Hz_40ms_log_rms_mae_db": 1.155790790194684, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09362247314153054, "300–1000 Hz_40ms_envelope_correlation": 0.9909421591316795, "300–1000 Hz_40ms_modulation_mae_db": 0.4277326195189189, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7210733231284612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18247040991100316, "1000–3000 Hz_10ms_envelope_correlation": 0.9697084932615895, "1000–3000 Hz_10ms_modulation_mae_db": 0.1504813335734413, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9938551621668804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08801628067922457, "1000–3000 Hz_40ms_envelope_correlation": 0.9923180609276702, "1000–3000 Hz_40ms_modulation_mae_db": 0.6708147005555051, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5459425714657722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16169130376769889, "3000–6000 Hz_10ms_envelope_correlation": 0.9815747793611289, "3000–6000 Hz_10ms_modulation_mae_db": 0.7138669363215776, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8421993948264181, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08946291208836747, "3000–6000 Hz_40ms_envelope_correlation": 0.9960461892897211, "3000–6000 Hz_40ms_modulation_mae_db": 1.146034679395943, "mr_stft_db": 6.090346320414631, "band_mae_db": 0.3207824927823104, "rms_gain_error_db": -0.1736503992919156, "output_peak": 0.4519543381170029}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64800, "output_samples": 64800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.126961392329644, "spectral_convergence": 0.5997421122258598, "spectral_convergence_status": "ok", "amplitude_floor": 1.7331847370499167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.792101240642273, "spectral_convergence": 0.5618858334990505, "spectral_convergence_status": "ok", "amplitude_floor": 1.0846095570828337e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.423295295187048, "spectral_convergence": 0.618333570497495, "spectral_convergence_status": "ok", "amplitude_floor": 4.5533844599041527e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024118742205032635, "output_power": 0.00020815864799001512, "delta_db": -0.6396019711211197, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 9.191122965591724e-05, "output_power": 0.00010304344825641492, "delta_db": 0.49651807080483473, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 7.435432994216677e-05, "output_power": 6.61220193051037e-05, "delta_db": -0.5096015620084878, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.4154339925686723e-05, "output_power": 2.5062790621768464e-05, "delta_db": 0.16034252098631877, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006210431709364446, "output_power": 0.0006006707165201912, "delta_db": -0.1448533054996761, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023011417890304625, "output_power": 0.0002281191723387766, "delta_db": -0.037815921262057184, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3314555586137491, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.17826556925565396, "pitch": null, "output_rms_dbfs": -29.096795628320745, "output_peak": 0.4655473929686023, "reference_rms_dbfs": -28.918530059065088, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.1141193093863215}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.0001629307662014554, "log_rms_mae_db": 1.1624127640494024, "rms_envelope_nrmse": 0.13109841683152823, "envelope_correlation": 0.9810939887176288, "rms_gain_error_db": -0.17796631972230242, "reference_mean_power": 0.0012840985730840112, "output_mean_power": 0.0012325420119775553}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 7.3397819245367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003947037290166488, "output_power": 0.0003846511317529167, "reference_active": true, "delta_db": -0.1120421487711379}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980542214464163e-05, "output_power": 5.300571390517416e-05, "reference_active": true, "delta_db": -1.1956646964823308}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6036857037353872e-05, "output_power": 2.938543189742876e-05, "reference_active": true, "delta_db": 0.5254351973648421}], "active_band_count": 3, "active_band_mae_db": 0.6110473475394369}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.000144561944924798, "log_rms_mae_db": 0.6270319494496428, "rms_envelope_nrmse": 0.0676722168328213, "envelope_correlation": 0.9948031874479069, "rms_gain_error_db": -0.1783133117286664, "reference_mean_power": 0.0012928029739092497, "output_mean_power": 0.001240797790035354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 6.614408925582191e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037454976712341086, "output_power": 0.0003621743737284751, "reference_active": true, "delta_db": -0.14591813573044243}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9110632304140427e-05, "output_power": 2.369987552155746e-05, "reference_active": true, "delta_db": -0.8930557375501882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1315453531682465e-06, "output_power": 1.8242424083070892e-06, "reference_active": true, "delta_db": -0.6761203003550671}], "active_band_count": 3, "active_band_mae_db": 0.5716980578785659}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 3.733123977436472e-05, "log_rms_mae_db": 1.9170980684556411, "rms_envelope_nrmse": 0.17569046803231453, "envelope_correlation": 0.9708693917900899, "rms_gain_error_db": -0.21481545882932507, "reference_mean_power": 0.00013419941652038683, "output_mean_power": 0.00012772299143167816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.2377163404532544e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0020888510118017e-05, "output_power": 1.897672740425808e-05, "reference_active": true, "delta_db": -0.23262028080654307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.868636044401026e-06, "output_power": 6.951545326455962e-06, "reference_active": true, "delta_db": -0.5381809907725581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3701828329102724e-06, "output_power": 1.6418249205788645e-06, "reference_active": true, "delta_db": 0.7854832143558936}], "active_band_count": 3, "active_band_mae_db": 0.5187614953116649}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.100732475570721e-05, "log_rms_mae_db": 1.293427409610041, "rms_envelope_nrmse": 0.10095308494548068, "envelope_correlation": 0.9900104689094117, "rms_gain_error_db": -0.2148138875294284, "reference_mean_power": 0.00013520032575819994, "output_mean_power": 0.0001286756436342104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 4.6212903051863556e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.815754516308902e-05, "output_power": 1.728208842606964e-05, "reference_active": true, "delta_db": -0.21460910121754131}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8683197262456875e-06, "output_power": 3.5347753967273808e-06, "reference_active": true, "delta_db": -0.39160538922948684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2646920446881671e-07, "output_power": 1.088725515952344e-07, "reference_active": true, "delta_db": -0.6506638532083392}], "active_band_count": 3, "active_band_mae_db": 0.4189594478851224}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 4.662083951365485e-05, "log_rms_mae_db": 1.7749672226381787, "rms_envelope_nrmse": 0.2007800234471254, "envelope_correlation": 0.9634736819142173, "rms_gain_error_db": -0.3574961860527171, "reference_mean_power": 0.00017506320579060826, "output_mean_power": 0.00016122977892240524}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 9.028764248682407e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2773698500231605e-05, "output_power": 4.0071435942647734e-05, "reference_active": true, "delta_db": -0.2834189842053551}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.603092418254862e-06, "output_power": 6.874263037846233e-06, "reference_active": true, "delta_db": 0.17478770237801297}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4970630790605404e-06, "output_power": 4.278663943094564e-06, "reference_active": true, "delta_db": 0.8760471119076678}], "active_band_count": 3, "active_band_mae_db": 0.4447512661636786}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.913562930844258e-05, "log_rms_mae_db": 0.976954592933921, "rms_envelope_nrmse": 0.10303532857040322, "envelope_correlation": 0.9908686728938468, "rms_gain_error_db": -0.3573938032689578, "reference_mean_power": 0.00017624479575510055, "output_mean_power": 0.00016232182671649186}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 7.899096535479963e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7672515224694995e-05, "output_power": 3.610659529165849e-05, "reference_active": true, "delta_db": -0.18438078948832448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.050253319949336e-06, "output_power": 3.384874855563597e-06, "reference_active": true, "delta_db": 0.4520670824051354}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55667878122912e-07, "output_power": 2.3545404509166192e-07, "reference_active": true, "delta_db": -0.35770011088699016}], "active_band_count": 3, "active_band_mae_db": 0.3313826609268167}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.00015276264453937597, "log_rms_mae_db": 1.556852265270956, "rms_envelope_nrmse": 0.15042996829791172, "envelope_correlation": 0.9828102910241703, "rms_gain_error_db": -0.03170305762361357, "reference_mean_power": 0.0006070254152411961, "output_mean_power": 0.0006026103256076493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.389907119886337e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002863305566547835, "output_power": 0.00029809136761669804, "reference_active": true, "delta_db": 0.174817019663273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.608243104449444e-05, "output_power": 3.2722222896921836e-05, "reference_active": true, "delta_db": -0.4245299169595779}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0810295137794261e-05, "output_power": 1.4279065143422522e-05, "reference_active": true, "delta_db": 1.2086222389619923}], "active_band_count": 3, "active_band_mae_db": 0.6026563918616145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.00013215266595876854, "log_rms_mae_db": 0.87639595702749, "rms_envelope_nrmse": 0.07604080501434216, "envelope_correlation": 0.9953468046604435, "rms_gain_error_db": -0.03242434587434097, "reference_mean_power": 0.0006109513084873906, "output_mean_power": 0.0006064069424779747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 5.083168365591421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002672957849080471, "output_power": 0.0002815435293591323, "reference_active": true, "delta_db": 0.22553440265196315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4768224147952787e-05, "output_power": 1.4406335266948178e-05, "reference_active": true, "delta_db": -0.10774757790094655}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.446503663842908e-07, "output_power": 8.458116568090606e-07, "reference_active": true, "delta_db": 0.005966914320249183}], "active_band_count": 3, "active_band_mae_db": 0.11308296495771963}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7630073016096627, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4abefe55945304323d3119c13a508f2fd6dcb09eedb0f198b1e13525fddb64cb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9674290845892035, "standard_deviation": 0.0034505100829252105, "minimum": 0.9611435828238423, "maximum": 0.9710267330301768, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7193753755493288, "standard_deviation": 0.03138158032932951, "minimum": 1.6730393367524672, "maximum": 1.7749672226381787, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3708668900249557, "standard_deviation": 0.182370652296564, "minimum": 0.1504813335734413, "maximum": 0.6581916504564136, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18935808789719544, "standard_deviation": 0.009763637247520184, "minimum": 0.17826258161197195, "maximum": 0.2068053540872253, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9915723222494388, "standard_deviation": 0.001321788504783536, "minimum": 0.9894428502302037, "maximum": 0.9930746204376366, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9815675963643337, "standard_deviation": 0.03395367903913066, "minimum": 0.9401449377503598, "maximum": 1.0416191762655085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5922248813906269, "standard_deviation": 0.17971542420868955, "minimum": 0.3313826609268167, "maximum": 0.7821067104149773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09391683659879563, "standard_deviation": 0.009107492861891798, "minimum": 0.08362108814263451, "maximum": 0.10716567313061381, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9800587002594787, "standard_deviation": 0.0020225724239286134, "minimum": 0.9770538210868083, "maximum": 0.9828102910241703, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5627773355429027, "standard_deviation": 0.026149689207691354, "minimum": 1.5362440812182636, "maximum": 1.6153184402826442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6179655180428784, "standard_deviation": 0.2563099708409799, "minimum": 0.25552717185341706, "maximum": 1.0281862123293646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1646690380056859, "standard_deviation": 0.011097792398055095, "minimum": 0.15042996829791172, "maximum": 0.18387758335735133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945125014359254, "standard_deviation": 0.0015167803040234853, "minimum": 0.9921451619641944, "maximum": 0.9960461892897211, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8609286868334536, "standard_deviation": 0.01584375797785642, "minimum": 0.8350130568322482, "maximum": 0.87639595702749, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5400406731552058, "standard_deviation": 0.3298682348314428, "minimum": 0.11308296495771963, "maximum": 1.146034679395943, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0877333074529412, "standard_deviation": 0.013097297831674294, "minimum": 0.07309911851036743, "maximum": 0.10798535775035659, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9702035591183571, "standard_deviation": 0.0031874753786367877, "minimum": 0.965390331737787, "maximum": 0.9758871793077945, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8788266076896416, "standard_deviation": 0.037185301098842063, "minimum": 1.8358332037854324, "maximum": 1.936356342915686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6494229567138035, "standard_deviation": 0.2756162570191755, "minimum": 0.30525295046191425, "maximum": 1.0861009307691922, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17797511412236783, "standard_deviation": 0.00966041687118136, "minimum": 0.1597088435354829, "maximum": 0.19121255181183522, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9896798971869064, "standard_deviation": 0.0021332991925489386, "minimum": 0.986397221034594, "maximum": 0.9927524950448413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2297611792959255, "standard_deviation": 0.05164591142137292, "minimum": 1.155790790194684, "maximum": 1.303096068938001, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6182808362225122, "standard_deviation": 0.43342911421040614, "minimum": 0.30494881168929194, "maximum": 1.6532683471862162, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10263320856948129, "standard_deviation": 0.010917959969668014, "minimum": 0.08408762644396314, "maximum": 0.11711204549275189, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3974968863129892, "standard_deviation": 0.09885761292506894, "minimum": 0.2634770370367416, "maximum": 0.545580435045692, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.1019647414700415, "standard_deviation": 0.007771008451321336, "minimum": 6.090346320414631, "maximum": 6.1141193093863215, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46299446215000406, "standard_deviation": 0.05165404605955704, "minimum": 0.38986589723371945, "maximum": 0.5357188640869527, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09891948575783521, "standard_deviation": 0.11478570897764838, "minimum": -0.17826556925565396, "maximum": 0.1187738761816385, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9734956578972364, "standard_deviation": 0.0053510072644533824, "minimum": 0.9644082184281815, "maximum": 0.9810939887176288, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1515361939995201, "standard_deviation": 0.022560580174410615, "minimum": 1.1219762552244805, "maximum": 1.1949160753593766, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7010810800826563, "standard_deviation": 0.41839615775857225, "minimum": 0.2661827455798065, "maximum": 1.6378246829329475, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15488977562207812, "standard_deviation": 0.016007982056830718, "minimum": 0.13109841683152823, "maximum": 0.17891479373722072, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9914533302700055, "standard_deviation": 0.0030522430397187227, "minimum": 0.9853473980890465, "maximum": 0.9948031874479069, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6087999263290181, "standard_deviation": 0.03738809814075756, "minimum": 0.5655128408374821, "maximum": 0.6842116436398694, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8132328584430113, "standard_deviation": 0.412857171007363, "minimum": 0.31297010209335313, "maximum": 1.6889483761652286, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08451592318716769, "standard_deviation": 0.013908370081536519, "minimum": 0.0676722168328213, "maximum": 0.10973108360366636, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_6930-81414-0026", "language": "en", "style": "normal", "speaker": "librispeech_6930", "source_utterance_id": "6930-81414-0026", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 49200, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_6930-81414-0026.wav", "source_file_sha256": "beecbca2d7a6a73c685c0be456e71a86fe3f7dfade1d3a900a2c5965d3f4f1d2", "decoded_pcm_sha256": "6ea614ae159242a4cb64e926932b9c1cf0c8b84fb4912932e606a31b6d48c909", "original_source_sha256": "84048bb63a1968e9856f38f430d293fd5036cc1b8c31312f3aff96d087732145", "corpus_entry_id": "librispeech_6930-81414-0026"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.508353233337402, "predicted_mos": 4.508353233337402, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "c3e7e496a47ea45c5ebb3ea20ddf9db01aa5c6faab9b6e6155057327712135a5", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.120946491999348, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_6930-81414-0026", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c3e7e496a47ea45c5ebb3ea20ddf9db01aa5c6faab9b6e6155057327712135a5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "c88868dff5818474fe738bb4157cb999f7d1139e9923ec83577117e882de4b31"}, "rows": [{"case": "librispeech_6930-81414-0026", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.8023667335510254, "predicted_mos": 2.8023667335510254, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "2d7c74d5108131969b0ebfa69b94c412e2c24f4dee66e91085998e9ccca85c5a", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.298542782998993, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_6930-81414-0026", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "2d7c74d5108131969b0ebfa69b94c412e2c24f4dee66e91085998e9ccca85c5a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.881573690688317, "全帯域_10ms_rms_envelope_nrmse": 0.09746140676566788, "全帯域_10ms_envelope_correlation": 0.9865318507999721, "全帯域_10ms_modulation_mae_db": 0.5363328585892173, "全帯域_40ms_log_rms_mae_db": 0.46035219150879886, "全帯域_40ms_rms_envelope_nrmse": 0.05553120813638943, "全帯域_40ms_envelope_correlation": 0.9951472607612849, "全帯域_40ms_modulation_mae_db": 0.5678218134636884, "300–1000 Hz_10ms_log_rms_mae_db": 1.2046443111517249, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08449218865389015, "300–1000 Hz_10ms_envelope_correlation": 0.9928074354706783, "300–1000 Hz_10ms_modulation_mae_db": 0.6306413627779787, "300–1000 Hz_40ms_log_rms_mae_db": 0.6881097439561322, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041956331125891644, "300–1000 Hz_40ms_envelope_correlation": 0.9981346108766541, "300–1000 Hz_40ms_modulation_mae_db": 0.541090970523071, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0841701597575526, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11616367610154593, "1000–3000 Hz_10ms_envelope_correlation": 0.9875774141820802, "1000–3000 Hz_10ms_modulation_mae_db": 0.71513609593143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5872752479018694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04984414969147072, "1000–3000 Hz_40ms_envelope_correlation": 0.9976161414851543, "1000–3000 Hz_40ms_modulation_mae_db": 0.7685344188087871, "3000–6000 Hz_10ms_log_rms_mae_db": 1.165039900596889, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1640340368067153, "3000–6000 Hz_10ms_envelope_correlation": 0.9717215694835294, "3000–6000 Hz_10ms_modulation_mae_db": 0.515054810182917, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6739896004596249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10285986193722212, "3000–6000 Hz_40ms_envelope_correlation": 0.9879750995919542, "3000–6000 Hz_40ms_modulation_mae_db": 0.423438832630424, "mr_stft_db": 5.745338288435271, "band_mae_db": 0.13581504649097317, "rms_gain_error_db": -0.04641423986944226, "output_peak": 0.459282036868441}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9229473048664406, "全帯域_10ms_rms_envelope_nrmse": 0.09635023293103767, "全帯域_10ms_envelope_correlation": 0.9869156680054092, "全帯域_10ms_modulation_mae_db": 0.2428830709349993, "全帯域_40ms_log_rms_mae_db": 0.4344473107529396, "全帯域_40ms_rms_envelope_nrmse": 0.04286450227564586, "全帯域_40ms_envelope_correlation": 0.9973499683595627, "全帯域_40ms_modulation_mae_db": 0.228534138859565, "300–1000 Hz_10ms_log_rms_mae_db": 1.2801069625983517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10574369101869907, "300–1000 Hz_10ms_envelope_correlation": 0.9886658957085267, "300–1000 Hz_10ms_modulation_mae_db": 0.6580920318831079, "300–1000 Hz_40ms_log_rms_mae_db": 0.7733496961950408, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049139718978368926, "300–1000 Hz_40ms_envelope_correlation": 0.997439352740377, "300–1000 Hz_40ms_modulation_mae_db": 0.5301352457716189, "1000–3000 Hz_10ms_log_rms_mae_db": 1.094871399410936, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1284286868809965, "1000–3000 Hz_10ms_envelope_correlation": 0.9849066821427052, "1000–3000 Hz_10ms_modulation_mae_db": 0.5838051270436594, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6268897659662573, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06277331250144728, "1000–3000 Hz_40ms_envelope_correlation": 0.9965098794343703, "1000–3000 Hz_40ms_modulation_mae_db": 0.8763106104852928, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1522328204082208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1511602054512687, "3000–6000 Hz_10ms_envelope_correlation": 0.9762394085157007, "3000–6000 Hz_10ms_modulation_mae_db": 0.4418988670345973, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6710980283057792, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07994045694756033, "3000–6000 Hz_40ms_envelope_correlation": 0.9936036721674744, "3000–6000 Hz_40ms_modulation_mae_db": 0.49000917820227796, "mr_stft_db": 5.73825857660363, "band_mae_db": 0.16728821304321886, "rms_gain_error_db": -0.11693830951556107, "output_peak": 0.4261299099928275}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8267295164268705, "全帯域_10ms_rms_envelope_nrmse": 0.10206652528251996, "全帯域_10ms_envelope_correlation": 0.9852492289771484, "全帯域_10ms_modulation_mae_db": 0.23989136066675165, "全帯域_40ms_log_rms_mae_db": 0.455584565251397, "全帯域_40ms_rms_envelope_nrmse": 0.04855393803939606, "全帯域_40ms_envelope_correlation": 0.9963342339465328, "全帯域_40ms_modulation_mae_db": 0.19003768259285647, "300–1000 Hz_10ms_log_rms_mae_db": 1.1094160326957274, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10740394694131825, "300–1000 Hz_10ms_envelope_correlation": 0.9882109113198928, "300–1000 Hz_10ms_modulation_mae_db": 0.15928961497199814, "300–1000 Hz_40ms_log_rms_mae_db": 0.6319886181058043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061874455852863795, "300–1000 Hz_40ms_envelope_correlation": 0.9957794922520382, "300–1000 Hz_40ms_modulation_mae_db": 0.45058872568925273, "1000–3000 Hz_10ms_log_rms_mae_db": 1.127233350157583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14045078098040448, "1000–3000 Hz_10ms_envelope_correlation": 0.9820906706579998, "1000–3000 Hz_10ms_modulation_mae_db": 1.3435245372531448, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6581581289650407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07576210149868132, "1000–3000 Hz_40ms_envelope_correlation": 0.9944877353169101, "1000–3000 Hz_40ms_modulation_mae_db": 1.596928497790816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1873567291410685, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15173168856760622, "3000–6000 Hz_10ms_envelope_correlation": 0.9758908704248443, "3000–6000 Hz_10ms_modulation_mae_db": 0.43906799842395866, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6851025876420925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07318698195528238, "3000–6000 Hz_40ms_envelope_correlation": 0.9944609090742443, "3000–6000 Hz_40ms_modulation_mae_db": 0.5208422019569637, "mr_stft_db": 5.728218063982709, "band_mae_db": 0.08594022750519235, "rms_gain_error_db": -0.0471618805854002, "output_peak": 0.4747997508162648}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.904376285300358, "全帯域_10ms_rms_envelope_nrmse": 0.09768202857220343, "全帯域_10ms_envelope_correlation": 0.9864884184573668, "全帯域_10ms_modulation_mae_db": 0.5659147711298327, "全帯域_40ms_log_rms_mae_db": 0.446874414812408, "全帯域_40ms_rms_envelope_nrmse": 0.04359803977077722, "全帯域_40ms_envelope_correlation": 0.9970019509286769, "全帯域_40ms_modulation_mae_db": 0.4706274314282599, "300–1000 Hz_10ms_log_rms_mae_db": 1.1857037039114429, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09744122664357527, "300–1000 Hz_10ms_envelope_correlation": 0.9908400128437489, "300–1000 Hz_10ms_modulation_mae_db": 0.8900014738382849, "300–1000 Hz_40ms_log_rms_mae_db": 0.6869087524281957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051330806405747366, "300–1000 Hz_40ms_envelope_correlation": 0.997533671343594, "300–1000 Hz_40ms_modulation_mae_db": 0.9945134790797692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.14667764459317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12980474331366712, "1000–3000 Hz_10ms_envelope_correlation": 0.9846025028827947, "1000–3000 Hz_10ms_modulation_mae_db": 0.5618712800811213, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6134837054682546, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057764245226190836, "1000–3000 Hz_40ms_envelope_correlation": 0.9967797102545262, "1000–3000 Hz_40ms_modulation_mae_db": 0.7846030378230694, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1944528682551567, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1547743639804683, "3000–6000 Hz_10ms_envelope_correlation": 0.9745863943110021, "3000–6000 Hz_10ms_modulation_mae_db": 0.6726769789751835, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6654451916148316, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07237130869816968, "3000–6000 Hz_40ms_envelope_correlation": 0.9942867709455453, "3000–6000 Hz_40ms_modulation_mae_db": 0.8259614619014886, "mr_stft_db": 5.722557508223379, "band_mae_db": 0.1818818825143905, "rms_gain_error_db": -0.004587898453265432, "output_peak": 0.4299379183246855}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8659353717047884, "全帯域_10ms_rms_envelope_nrmse": 0.09961178414056361, "全帯域_10ms_envelope_correlation": 0.9863605336762863, "全帯域_10ms_modulation_mae_db": 0.6703881631642946, "全帯域_40ms_log_rms_mae_db": 0.44702738598586894, "全帯域_40ms_rms_envelope_nrmse": 0.05151542127605626, "全帯域_40ms_envelope_correlation": 0.9965898636083769, "全帯域_40ms_modulation_mae_db": 1.0683217428374556, "300–1000 Hz_10ms_log_rms_mae_db": 1.2096574751102842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0986350944635823, "300–1000 Hz_10ms_envelope_correlation": 0.9900394427167144, "300–1000 Hz_10ms_modulation_mae_db": 0.9729502693721281, "300–1000 Hz_40ms_log_rms_mae_db": 0.7289595986725024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04711688668729318, "300–1000 Hz_40ms_envelope_correlation": 0.9975822639855569, "300–1000 Hz_40ms_modulation_mae_db": 1.0984800068750955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1592851258306016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13760785929239464, "1000–3000 Hz_10ms_envelope_correlation": 0.9829017857436677, "1000–3000 Hz_10ms_modulation_mae_db": 0.7151637038294579, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6355012783639679, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06412223171441661, "1000–3000 Hz_40ms_envelope_correlation": 0.9968274410613068, "1000–3000 Hz_40ms_modulation_mae_db": 0.6363624742419156, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1465301748116454, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15435399499189023, "3000–6000 Hz_10ms_envelope_correlation": 0.975338327663697, "3000–6000 Hz_10ms_modulation_mae_db": 1.0468002081774022, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6509469066509592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08310651916151747, "3000–6000 Hz_40ms_envelope_correlation": 0.9935279223982222, "3000–6000 Hz_40ms_modulation_mae_db": 1.4340560078896705, "mr_stft_db": 5.694937426216796, "band_mae_db": 0.18775823587873539, "rms_gain_error_db": -0.2051820647624234, "output_peak": 0.46078747783976476}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.855760885062232, "全帯域_10ms_rms_envelope_nrmse": 0.0894051191579652, "全帯域_10ms_envelope_correlation": 0.9890132580923089, "全帯域_10ms_modulation_mae_db": 0.42319507670037454, "全帯域_40ms_log_rms_mae_db": 0.45996694655348963, "全帯域_40ms_rms_envelope_nrmse": 0.04869352556211402, "全帯域_40ms_envelope_correlation": 0.9968828642872442, "全帯域_40ms_modulation_mae_db": 0.5645641259568607, "300–1000 Hz_10ms_log_rms_mae_db": 1.2810154631924435, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09926400733131328, "300–1000 Hz_10ms_envelope_correlation": 0.9898946759701321, "300–1000 Hz_10ms_modulation_mae_db": 0.27218593429439103, "300–1000 Hz_40ms_log_rms_mae_db": 0.831609516150564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06723564277059546, "300–1000 Hz_40ms_envelope_correlation": 0.9950043887447529, "300–1000 Hz_40ms_modulation_mae_db": 0.4483980048093004, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1361417752012346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12002931050630974, "1000–3000 Hz_10ms_envelope_correlation": 0.9868772092663259, "1000–3000 Hz_10ms_modulation_mae_db": 0.5652780456920824, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6785914529349346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06483289717539899, "1000–3000 Hz_40ms_envelope_correlation": 0.9962868939223543, "1000–3000 Hz_40ms_modulation_mae_db": 0.6745260220203121, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0749385971642462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1404809516671721, "3000–6000 Hz_10ms_envelope_correlation": 0.9800202560491916, "3000–6000 Hz_10ms_modulation_mae_db": 0.6916789795391436, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6209754659964798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08102135049653705, "3000–6000 Hz_40ms_envelope_correlation": 0.9941772191900252, "3000–6000 Hz_40ms_modulation_mae_db": 0.9832009821338309, "mr_stft_db": 5.716951507440875, "band_mae_db": 0.14311765585435315, "rms_gain_error_db": -0.18144981741871583, "output_peak": 0.4512344729730007}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8064492371113511, "全帯域_10ms_rms_envelope_nrmse": 0.09355977512412371, "全帯域_10ms_envelope_correlation": 0.9876991195760229, "全帯域_10ms_modulation_mae_db": 0.4574684306848191, "全帯域_40ms_log_rms_mae_db": 0.4576436391951363, "全帯域_40ms_rms_envelope_nrmse": 0.052103186140325955, "全帯域_40ms_envelope_correlation": 0.9957845564182298, "全帯域_40ms_modulation_mae_db": 0.404596107096068, "300–1000 Hz_10ms_log_rms_mae_db": 1.2603571612056501, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09082372200099727, "300–1000 Hz_10ms_envelope_correlation": 0.9915525318427542, "300–1000 Hz_10ms_modulation_mae_db": 0.5167017399262727, "300–1000 Hz_40ms_log_rms_mae_db": 0.8118306419279738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047691212267266685, "300–1000 Hz_40ms_envelope_correlation": 0.9975297635412885, "300–1000 Hz_40ms_modulation_mae_db": 0.6371721482342884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0987673106569982, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14506755024548967, "1000–3000 Hz_10ms_envelope_correlation": 0.9818804994640676, "1000–3000 Hz_10ms_modulation_mae_db": 0.47491218302350857, "1000–3000 Hz_40ms_log_rms_mae_db": 0.644592980896528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0931315851627533, "1000–3000 Hz_40ms_envelope_correlation": 0.9924381638200912, "1000–3000 Hz_40ms_modulation_mae_db": 0.4126480404021771, "3000–6000 Hz_10ms_log_rms_mae_db": 1.131960583092656, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13946609072366725, "3000–6000 Hz_10ms_envelope_correlation": 0.9794909978588268, "3000–6000 Hz_10ms_modulation_mae_db": 0.5560912785026194, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6809473299293183, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07519508349787038, "3000–6000 Hz_40ms_envelope_correlation": 0.9938234552353483, "3000–6000 Hz_40ms_modulation_mae_db": 0.5554064433893043, "mr_stft_db": 5.721266128151882, "band_mae_db": 0.08855024803835775, "rms_gain_error_db": -0.02464011081094427, "output_peak": 0.42882209538864374}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8585656518411413, "全帯域_10ms_rms_envelope_nrmse": 0.1043244331010428, "全帯域_10ms_envelope_correlation": 0.9846869414448534, "全帯域_10ms_modulation_mae_db": 0.2430620938796519, "全帯域_40ms_log_rms_mae_db": 0.4506811490301917, "全帯域_40ms_rms_envelope_nrmse": 0.0521890515343434, "全帯域_40ms_envelope_correlation": 0.9957332811806185, "全帯域_40ms_modulation_mae_db": 0.19632280490691636, "300–1000 Hz_10ms_log_rms_mae_db": 1.1756495356517744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09072462850128925, "300–1000 Hz_10ms_envelope_correlation": 0.9918341697073038, "300–1000 Hz_10ms_modulation_mae_db": 0.5140013793786588, "300–1000 Hz_40ms_log_rms_mae_db": 0.7206080161518642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05155481350421005, "300–1000 Hz_40ms_envelope_correlation": 0.997268113377284, "300–1000 Hz_40ms_modulation_mae_db": 0.6543779835326013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1121750585526746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12970572118074328, "1000–3000 Hz_10ms_envelope_correlation": 0.9847369513302315, "1000–3000 Hz_10ms_modulation_mae_db": 0.6266998206684036, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5996469209474664, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0606254273341967, "1000–3000 Hz_40ms_envelope_correlation": 0.9964929463382666, "1000–3000 Hz_40ms_modulation_mae_db": 0.5759363950116471, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2026788856113144, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1651270856908495, "3000–6000 Hz_10ms_envelope_correlation": 0.9712072801674656, "3000–6000 Hz_10ms_modulation_mae_db": 0.23933568222375623, "3000–6000 Hz_40ms_log_rms_mae_db": 0.735154927005545, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08670781585807905, "3000–6000 Hz_40ms_envelope_correlation": 0.9913521241542668, "3000–6000 Hz_40ms_modulation_mae_db": 0.2767965451280188, "mr_stft_db": 5.7236695181254165, "band_mae_db": 0.12731579399772597, "rms_gain_error_db": 0.020733521030579533, "output_peak": 0.428881333201476}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49200, "output_samples": 49200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.511450880142349, "spectral_convergence": 0.4371875544940863, "spectral_convergence_status": "ok", "amplitude_floor": 6.39186387874099e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3502928163020655, "spectral_convergence": 0.40633808653215414, "spectral_convergence_status": "ok", "amplitude_floor": 4.1641566181737356e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.374271168861396, "spectral_convergence": 0.4480178045806461, "spectral_convergence_status": "ok", "amplitude_floor": 3.471921787916526e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.471560290376807e-05, "output_power": 7.444976733217594e-05, "delta_db": -0.015479604130695884, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002820835097806379, "output_power": 0.0002886126092484974, "delta_db": 0.09937602511066057, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026780543872196867, "output_power": 0.0002734319870142359, "delta_db": 0.09029925838574622, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028630139823372175, "output_power": 0.0002801030694765344, "delta_db": -0.0950560101188026, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006167999147673381, "output_power": 0.000573245552383771, "delta_db": -0.31803611157142286, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00037968124815517825, "output_power": 0.0003972679007985066, "delta_db": 0.19664326962851097, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13581504649097317, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04641423986944226, "pitch": null, "output_rms_dbfs": -27.242025800294375, "output_peak": 0.459282036868441, "reference_rms_dbfs": -27.195611560424933, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.745338288435271}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 0.00010992699921133752, "log_rms_mae_db": 0.881573690688317, "rms_envelope_nrmse": 0.09746140676566788, "envelope_correlation": 0.9865318507999721, "rms_gain_error_db": -0.04652389374250499, "reference_mean_power": 0.0019104810338505282, "output_mean_power": 0.0018901241941689634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 4.701235881229352e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003075331225832077, "output_power": 0.00029881520550637805, "reference_active": true, "delta_db": -0.12489204725355721}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8607043791082874e-05, "output_power": 2.7279873375160053e-05, "reference_active": true, "delta_db": -0.20630630658061686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1458430393639735e-05, "output_power": 1.5378187546237784e-05, "reference_active": true, "delta_db": 1.2778002219334779}], "active_band_count": 3, "active_band_mae_db": 0.5363328585892173}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 9.464213695185308e-05, "log_rms_mae_db": 0.46035219150879886, "rms_envelope_nrmse": 0.05553120813638943, "envelope_correlation": 0.9951472607612849, "rms_gain_error_db": -0.046595307891397375, "reference_mean_power": 0.0019290839559002245, "output_mean_power": 0.0019084975125679736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 4.127760176364045e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002739478020838986, "output_power": 0.0002663099669448849, "reference_active": true, "delta_db": -0.1228039969379175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7486141765201905e-05, "output_power": 1.627585166751315e-05, "reference_active": true, "delta_db": -0.3115027182797466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8391125647823e-07, "output_power": 7.821007818746762e-07, "reference_active": true, "delta_db": 1.2691587251734013}], "active_band_count": 3, "active_band_mae_db": 0.5678218134636884}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 6.478025356347348e-05, "log_rms_mae_db": 1.2046443111517249, "rms_envelope_nrmse": 0.08449218865389015, "envelope_correlation": 0.9928074354706783, "rms_gain_error_db": 0.06249188102647943, "reference_mean_power": 0.0004974887922363912, "output_mean_power": 0.0005046990522747669}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.1951630543463553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001563474072286115, "output_power": 0.00015730761492200004, "reference_active": true, "delta_db": 0.02659062830219024}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.160597194093242e-06, "output_power": 8.427487805955944e-06, "reference_active": true, "delta_db": 0.13976191125053308}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0536955073753773e-06, "output_power": 3.0555768282886437e-06, "reference_active": true, "delta_db": 1.7255715487812129}], "active_band_count": 3, "active_band_mae_db": 0.6306413627779787}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 5.647638237474489e-05, "log_rms_mae_db": 0.6881097439561322, "rms_envelope_nrmse": 0.041956331125891644, "envelope_correlation": 0.9981346108766541, "rms_gain_error_db": 0.06248767756113066, "reference_mean_power": 0.00050239739349007, "output_mean_power": 0.0005096783021059495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.0265825460069485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014180308081562638, "output_power": 0.00014243803528739667, "reference_active": true, "delta_db": 0.019403081849613323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4691763208394e-06, "output_power": 4.519096438019893e-06, "reference_active": true, "delta_db": 0.048241202569682194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.310040798443831e-08, "output_power": 1.3320313517802128e-07, "reference_active": true, "delta_db": 1.5556286271499176}], "active_band_count": 3, "active_band_mae_db": 0.541090970523071}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 7.855020595588598e-05, "log_rms_mae_db": 1.0841701597575526, "rms_envelope_nrmse": 0.11616367610154593, "envelope_correlation": 0.9875774141820802, "rms_gain_error_db": -0.062115557566651644, "reference_mean_power": 0.0004940955013379756, "output_mean_power": 0.0004870789307249858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 1.3474634060369984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.364408948380063e-05, "output_power": 9.081872434869449e-05, "reference_active": true, "delta_db": -0.1330497384150961}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.996833569277459e-06, "output_power": 6.492795511925149e-06, "reference_active": true, "delta_db": -0.32469818503756276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2622375016010212e-06, "output_power": 3.336600679754529e-06, "reference_active": true, "delta_db": 1.687660364341631}], "active_band_count": 3, "active_band_mae_db": 0.71513609593143}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 6.30584220600245e-05, "log_rms_mae_db": 0.5872752479018694, "rms_envelope_nrmse": 0.04984414969147072, "envelope_correlation": 0.9976161414851543, "rms_gain_error_db": -0.062115506915248, "reference_mean_power": 0.0004989699589215838, "output_mean_power": 0.0004918841726585553}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 1.243238840477423e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.827683631948728e-05, "output_power": 8.557620518810508e-05, "reference_active": true, "delta_db": -0.13493736332239503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8441959267659476e-06, "output_power": 3.919175483239256e-06, "reference_active": true, "delta_db": 0.08389195482017296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.483409815447214e-08, "output_power": 1.5333521707903197e-07, "reference_active": true, "delta_db": 2.0867739382837938}], "active_band_count": 3, "active_band_mae_db": 0.7685344188087871}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 9.512970769085293e-05, "log_rms_mae_db": 1.165039900596889, "rms_envelope_nrmse": 0.1640340368067153, "envelope_correlation": 0.9717215694835294, "rms_gain_error_db": -0.1490512898572669, "reference_mean_power": 0.0007033343720650757, "output_mean_power": 0.0006796052309429386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.9605352394438525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001169604983627503, "output_power": 0.00010919736468171281, "reference_active": true, "delta_db": -0.29827052689836425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1203242070906313e-05, "output_power": 1.93271955651207e-05, "reference_active": true, "delta_db": -0.40233430432745737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.096712651632301e-05, "output_power": 1.332136762189874e-05, "reference_active": true, "delta_db": 0.8445595993229292}], "active_band_count": 3, "active_band_mae_db": 0.515054810182917}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 7.035164526339387e-05, "log_rms_mae_db": 0.6739896004596249, "rms_envelope_nrmse": 0.10285986193722212, "envelope_correlation": 0.9879750995919542, "rms_gain_error_db": -0.14905133350440924, "reference_mean_power": 0.0007102740547296548, "output_mean_power": 0.0006863107752416324}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.6373521771123283e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010173798550296795, "output_power": 9.501719098334256e-05, "reference_active": true, "delta_db": -0.29680946869172825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.290709427246265e-05, "output_power": 1.1688630461292627e-05, "reference_active": true, "delta_db": -0.4306485358969673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.316189457175965e-07, "output_power": 7.157164476919365e-07, "reference_active": true, "delta_db": 0.5428584933025763}], "active_band_count": 3, "active_band_mae_db": 0.423438832630424}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8520118141910049, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d1376474ae8cc91a1d8b2a53e2d9962753ce0f55f45db3fe62c906e04893165c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9844467144587341, "standard_deviation": 0.002089151230144575, "minimum": 0.9818804994640676, "maximum": 0.9875774141820802, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1199152280200937, "standard_deviation": 0.02671307712969964, "minimum": 1.0841701597575526, "maximum": 1.1592851258306016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.698298849190351, "standard_deviation": 0.2729018547886844, "minimum": 0.47491218302350857, "maximum": 1.3435245372531448, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13090729106269391, "standard_deviation": 0.009863324983069865, "minimum": 0.11616367610154593, "maximum": 0.14506755024548967, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9959298639541224, "standard_deviation": 0.001665793258843199, "minimum": 0.9924381638200912, "maximum": 0.9976161414851543, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6305174351805398, "standard_deviation": 0.030294417584564325, "minimum": 0.5872752479018694, "maximum": 0.6785914529349346, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7907311870730022, "standard_deviation": 0.3554122758218491, "minimum": 0.4126480404021771, "maximum": 1.596928497790816, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06610699378806947, "standard_deviation": 0.013116809280600786, "minimum": 0.04984414969147072, "maximum": 0.0931315851627533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9755618880592822, "standard_deviation": 0.003175676222120567, "minimum": 0.9712072801674656, "maximum": 0.9800202560491916, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1568988198851495, "standard_deviation": 0.041349998294736914, "minimum": 1.0749385971642462, "maximum": 1.2026788856113144, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5753256003824472, "standard_deviation": 0.2386163343731006, "minimum": 0.23933568222375623, "maximum": 1.0468002081774022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1526410522349547, "standard_deviation": 0.009393415831808718, "minimum": 0.13946609072366725, "maximum": 0.1651270856908495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9929008965946351, "standard_deviation": 0.0022162912235764425, "minimum": 0.9879750995919542, "maximum": 0.9944609090742443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6729575047005788, "standard_deviation": 0.032353882930089735, "minimum": 0.6209754659964798, "maximum": 0.735154927005545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6887139566539973, "standard_deviation": 0.37541910154885183, "minimum": 0.2767965451280188, "maximum": 1.4340560078896705, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0817986723190298, "standard_deviation": 0.009856722408345434, "minimum": 0.07237130869816968, "maximum": 0.10285986193722212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904806344474689, "standard_deviation": 0.001581459308264585, "minimum": 0.9882109113198928, "maximum": 0.9928074354706783, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.213318830689675, "standard_deviation": 0.05894562677864374, "minimum": 1.1094160326957274, "maximum": 1.2810154631924435, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5767329758053525, "standard_deviation": 0.2775395900308331, "minimum": 0.15928961497199814, "maximum": 0.9729502693721281, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09681606319433311, "standard_deviation": 0.00780466498494112, "minimum": 0.08449218865389015, "maximum": 0.10740394694131825, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970339571076933, "standard_deviation": 0.0010636609341666782, "minimum": 0.9950043887447529, "maximum": 0.9981346108766541, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7341705729485097, "standard_deviation": 0.06768410491414056, "minimum": 0.6319886181058043, "maximum": 0.831609516150564, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6693445705643747, "standard_deviation": 0.2460112749644531, "minimum": 0.4483980048093004, "maximum": 1.0984800068750955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05223748344902964, "standard_deviation": 0.008291919063792477, "minimum": 0.041956331125891644, "maximum": 0.06723564277059546, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1397084129153684, "standard_deviation": 0.038797213152822384, "minimum": 0.08594022750519235, "maximum": 0.18775823587873539, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.723899627147494, "standard_deviation": 0.01502530971647326, "minimum": 5.694937426216796, "maximum": 5.745338288435271, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.444984374425638, "standard_deviation": 0.018839910147629032, "minimum": 0.4261299099928275, "maximum": 0.4747997508162648, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07570510004814661, "standard_deviation": 0.08305314494582276, "minimum": -0.2051820647624234, "maximum": 0.020733521030579533, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9866181273786709, "standard_deviation": 0.0013463023542375672, "minimum": 0.9846869414448534, "maximum": 0.9890132580923089, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8652922428751874, "standard_deviation": 0.03817786327598644, "minimum": 0.8064492371113511, "maximum": 0.9229473048664406, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42239197821874264, "standard_deviation": 0.16647760774138293, "minimum": 0.23989136066675165, "maximum": 0.6703881631642946, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09755766313439054, "standard_deviation": 0.004694497209305163, "minimum": 0.0894051191579652, "maximum": 0.1043244331010428, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963529974363159, "standard_deviation": 0.0007476900185531923, "minimum": 0.9951472607612849, "maximum": 0.9973499683595627, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45157220038627877, "standard_deviation": 0.008761130007428453, "minimum": 0.4344473107529396, "maximum": 0.46035219150879886, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4613532308927088, "standard_deviation": 0.29075374641656454, "minimum": 0.19003768259285647, "maximum": 1.0683217428374556, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04938110909188102, "standard_deviation": 0.004387324098308752, "minimum": 0.04286450227564586, "maximum": 0.05553120813638943, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-81414-0026", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.7282776832580566, "predicted_mos": 2.7282776832580566, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "ef85206b744c83b372388e6dee3dbcdde51bd0df20b7664fd6f157e416e649b0", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9709498998709023, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8613686437382686, "全帯域_10ms_rms_envelope_nrmse": 0.10193537884305282, "全帯域_10ms_envelope_correlation": 0.9852873900180925, "全帯域_10ms_modulation_mae_db": 0.5240038728738124, "全帯域_40ms_log_rms_mae_db": 0.4516456189568147, "全帯域_40ms_rms_envelope_nrmse": 0.05938463837259588, "全帯域_40ms_envelope_correlation": 0.9944269807076431, "全帯域_40ms_modulation_mae_db": 0.5475521255097974, "300–1000 Hz_10ms_log_rms_mae_db": 1.1418927110041923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08780131506545463, "300–1000 Hz_10ms_envelope_correlation": 0.992289842812512, "300–1000 Hz_10ms_modulation_mae_db": 0.823038836833572, "300–1000 Hz_40ms_log_rms_mae_db": 0.6022729554706745, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04286391036090096, "300–1000 Hz_40ms_envelope_correlation": 0.9980956152275692, "300–1000 Hz_40ms_modulation_mae_db": 0.6802494346386241, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0695779904844962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12096940657977973, "1000–3000 Hz_10ms_envelope_correlation": 0.9865726172639865, "1000–3000 Hz_10ms_modulation_mae_db": 0.8492284789515997, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5636581759861315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0523497912955623, "1000–3000 Hz_40ms_envelope_correlation": 0.9973425157381257, "1000–3000 Hz_40ms_modulation_mae_db": 1.0250240127797738, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1509911527623988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17164097092717706, "3000–6000 Hz_10ms_envelope_correlation": 0.9692015782541002, "3000–6000 Hz_10ms_modulation_mae_db": 0.5040310491208354, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6557298353108564, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11137871195778272, "3000–6000 Hz_40ms_envelope_correlation": 0.9858852085159018, "3000–6000 Hz_40ms_modulation_mae_db": 0.4128082449267398, "mr_stft_db": 5.717333412003216, "band_mae_db": 0.13964861690276553, "rms_gain_error_db": -0.023193341994383873, "output_peak": 0.45724582488060345}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9177509690987238, "全帯域_10ms_rms_envelope_nrmse": 0.09832734462028463, "全帯域_10ms_envelope_correlation": 0.9863361787595969, "全帯域_10ms_modulation_mae_db": 0.2469983371258229, "全帯域_40ms_log_rms_mae_db": 0.4403648810864724, "全帯域_40ms_rms_envelope_nrmse": 0.045252271996520224, "全帯域_40ms_envelope_correlation": 0.996964249565503, "全帯域_40ms_modulation_mae_db": 0.24384738079704785, "300–1000 Hz_10ms_log_rms_mae_db": 1.195215176625002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.109196724903189, "300–1000 Hz_10ms_envelope_correlation": 0.9879367857415863, "300–1000 Hz_10ms_modulation_mae_db": 0.7177652097327619, "300–1000 Hz_40ms_log_rms_mae_db": 0.6605440822950297, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051288429651923186, "300–1000 Hz_40ms_envelope_correlation": 0.9972082770010661, "300–1000 Hz_40ms_modulation_mae_db": 0.5945395966861402, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0849828130103827, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12584061724423065, "1000–3000 Hz_10ms_envelope_correlation": 0.9854701433755733, "1000–3000 Hz_10ms_modulation_mae_db": 0.4631564133217477, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5765675843418726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06068537122112064, "1000–3000 Hz_40ms_envelope_correlation": 0.9966085245903948, "1000–3000 Hz_40ms_modulation_mae_db": 0.7586864429094603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1494462395584635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15133880612513056, "3000–6000 Hz_10ms_envelope_correlation": 0.9761028994486903, "3000–6000 Hz_10ms_modulation_mae_db": 0.4511564785176052, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6661196679541629, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08003245887737581, "3000–6000 Hz_40ms_envelope_correlation": 0.993502084994541, "3000–6000 Hz_40ms_modulation_mae_db": 0.5191994868951703, "mr_stft_db": 5.712510392934916, "band_mae_db": 0.14632115651312996, "rms_gain_error_db": -0.10479352954488919, "output_peak": 0.4281474044147247}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8167895637295718, "全帯域_10ms_rms_envelope_nrmse": 0.10363944260001673, "全帯域_10ms_envelope_correlation": 0.9847962825932229, "全帯域_10ms_modulation_mae_db": 0.2753018257710094, "全帯域_40ms_log_rms_mae_db": 0.4574707429413105, "全帯域_40ms_rms_envelope_nrmse": 0.05131780207724827, "全帯域_40ms_envelope_correlation": 0.9958799277483044, "全帯域_40ms_modulation_mae_db": 0.17129398826232345, "300–1000 Hz_10ms_log_rms_mae_db": 1.0950334239585107, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10502953882075604, "300–1000 Hz_10ms_envelope_correlation": 0.9887394013869751, "300–1000 Hz_10ms_modulation_mae_db": 0.20673483555575126, "300–1000 Hz_40ms_log_rms_mae_db": 0.5830164643128137, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0633182271210758, "300–1000 Hz_40ms_envelope_correlation": 0.9955840123367727, "300–1000 Hz_40ms_modulation_mae_db": 0.48932143109636117, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0844324847490918, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14137576946974284, "1000–3000 Hz_10ms_envelope_correlation": 0.9819018577435896, "1000–3000 Hz_10ms_modulation_mae_db": 1.4542007688768914, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5922848698539187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07931326203993735, "1000–3000 Hz_40ms_envelope_correlation": 0.9939819859967166, "1000–3000 Hz_40ms_modulation_mae_db": 1.7424535777692431, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1327096335093192, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15140562693737297, "3000–6000 Hz_10ms_envelope_correlation": 0.9759728941659954, "3000–6000 Hz_10ms_modulation_mae_db": 0.41049930772733184, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6276594827263678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07365965125539017, "3000–6000 Hz_40ms_envelope_correlation": 0.9942400513232053, "3000–6000 Hz_40ms_modulation_mae_db": 0.538059342409221, "mr_stft_db": 5.697840240680438, "band_mae_db": 0.08532747473664039, "rms_gain_error_db": -0.031420413295592274, "output_peak": 0.48226345813363825}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9084777640840546, "全帯域_10ms_rms_envelope_nrmse": 0.09645084264635108, "全帯域_10ms_envelope_correlation": 0.9868967337497617, "全帯域_10ms_modulation_mae_db": 0.479990903393175, "全帯域_40ms_log_rms_mae_db": 0.44728392291999236, "全帯域_40ms_rms_envelope_nrmse": 0.041844509139581765, "全帯域_40ms_envelope_correlation": 0.9972634078117163, "全帯域_40ms_modulation_mae_db": 0.3397907026830756, "300–1000 Hz_10ms_log_rms_mae_db": 1.1226077569945918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10270184337304286, "300–1000 Hz_10ms_envelope_correlation": 0.9897861635039488, "300–1000 Hz_10ms_modulation_mae_db": 0.9255490990512688, "300–1000 Hz_40ms_log_rms_mae_db": 0.5952736701955641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05475934379888243, "300–1000 Hz_40ms_envelope_correlation": 0.9971403084529983, "300–1000 Hz_40ms_modulation_mae_db": 1.0096585498872939, "1000–3000 Hz_10ms_log_rms_mae_db": 1.122828831866373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12698241533426827, "1000–3000 Hz_10ms_envelope_correlation": 0.9853752969602406, "1000–3000 Hz_10ms_modulation_mae_db": 0.537086087518008, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5913014077425388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05709253085068744, "1000–3000 Hz_40ms_envelope_correlation": 0.9969065576212061, "1000–3000 Hz_40ms_modulation_mae_db": 0.7135952241684903, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1507495970261437, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15388758436511088, "3000–6000 Hz_10ms_envelope_correlation": 0.9749486237568696, "3000–6000 Hz_10ms_modulation_mae_db": 0.4759994549716689, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6364341297668533, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07214865393132323, "3000–6000 Hz_40ms_envelope_correlation": 0.9942048848299465, "3000–6000 Hz_40ms_modulation_mae_db": 0.6284757483941821, "mr_stft_db": 5.693027603999518, "band_mae_db": 0.16920645963286027, "rms_gain_error_db": 0.012769497943667933, "output_peak": 0.4324669009456237}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8547314024667385, "全帯域_10ms_rms_envelope_nrmse": 0.10033564509348329, "全帯域_10ms_envelope_correlation": 0.9861432636086824, "全帯域_10ms_modulation_mae_db": 0.6809417664350481, "全帯域_40ms_log_rms_mae_db": 0.44621664865098837, "全帯域_40ms_rms_envelope_nrmse": 0.05102690559161335, "全帯域_40ms_envelope_correlation": 0.9966668993798303, "全帯域_40ms_modulation_mae_db": 1.071020627409798, "300–1000 Hz_10ms_log_rms_mae_db": 1.1220535430077982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09785599990421512, "300–1000 Hz_10ms_envelope_correlation": 0.9901902405163089, "300–1000 Hz_10ms_modulation_mae_db": 0.915933729774762, "300–1000 Hz_40ms_log_rms_mae_db": 0.585974263396906, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04628682762646144, "300–1000 Hz_40ms_envelope_correlation": 0.9976608826972849, "300–1000 Hz_40ms_modulation_mae_db": 1.0294457773882755, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1050505101700037, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13542611714529854, "1000–3000 Hz_10ms_envelope_correlation": 0.9834035394999852, "1000–3000 Hz_10ms_modulation_mae_db": 0.7295775693373491, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5495778422023919, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06122420036233018, "1000–3000 Hz_40ms_envelope_correlation": 0.997060648765439, "1000–3000 Hz_40ms_modulation_mae_db": 0.6639998500943441, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1144511243056052, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1539466486714312, "3000–6000 Hz_10ms_envelope_correlation": 0.9755818285361226, "3000–6000 Hz_10ms_modulation_mae_db": 1.052062566687123, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6121623795086358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08171066897453202, "3000–6000 Hz_40ms_envelope_correlation": 0.9939351948243444, "3000–6000 Hz_40ms_modulation_mae_db": 1.450425831247952, "mr_stft_db": 5.671238107339273, "band_mae_db": 0.1933913220597602, "rms_gain_error_db": -0.2048193357235074, "output_peak": 0.45921051266287416}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8442711838662129, "全帯域_10ms_rms_envelope_nrmse": 0.08856043537693134, "全帯域_10ms_envelope_correlation": 0.9893035426912526, "全帯域_10ms_modulation_mae_db": 0.40037420672723156, "全帯域_40ms_log_rms_mae_db": 0.4625027906021167, "全帯域_40ms_rms_envelope_nrmse": 0.04970383346543847, "全帯域_40ms_envelope_correlation": 0.9968225110706743, "全帯域_40ms_modulation_mae_db": 0.4749849649813782, "300–1000 Hz_10ms_log_rms_mae_db": 1.2412755130516733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09509464268787114, "300–1000 Hz_10ms_envelope_correlation": 0.9907374684301552, "300–1000 Hz_10ms_modulation_mae_db": 0.1788532974250737, "300–1000 Hz_40ms_log_rms_mae_db": 0.7390884139810021, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06456497305911023, "300–1000 Hz_40ms_envelope_correlation": 0.9953967334647452, "300–1000 Hz_40ms_modulation_mae_db": 0.3745062183402494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0815469741577357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12186610048799888, "1000–3000 Hz_10ms_envelope_correlation": 0.986459565439618, "1000–3000 Hz_10ms_modulation_mae_db": 0.6454511874810466, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5884845453108986, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0649597808687327, "1000–3000 Hz_40ms_envelope_correlation": 0.996254980219638, "1000–3000 Hz_40ms_modulation_mae_db": 0.760710154577578, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0477389896045726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14024125399797935, "3000–6000 Hz_10ms_envelope_correlation": 0.9802749167416247, "3000–6000 Hz_10ms_modulation_mae_db": 0.5754670954429262, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5881026637903787, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08168936269591195, "3000–6000 Hz_40ms_envelope_correlation": 0.9942871184435973, "3000–6000 Hz_40ms_modulation_mae_db": 0.796791810953882, "mr_stft_db": 5.690195243733896, "band_mae_db": 0.16442377074124237, "rms_gain_error_db": -0.19313980975100734, "output_peak": 0.4400181129633347}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8409680981358734, "全帯域_10ms_rms_envelope_nrmse": 0.0946407348216529, "全帯域_10ms_envelope_correlation": 0.9874411727239754, "全帯域_10ms_modulation_mae_db": 0.44171126726857457, "全帯域_40ms_log_rms_mae_db": 0.47794664781230417, "全帯域_40ms_rms_envelope_nrmse": 0.05201117651746835, "全帯域_40ms_envelope_correlation": 0.9958144860052551, "全帯域_40ms_modulation_mae_db": 0.4309210702066242, "300–1000 Hz_10ms_log_rms_mae_db": 1.155005793981361, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08816388419868869, "300–1000 Hz_10ms_envelope_correlation": 0.9920348539698022, "300–1000 Hz_10ms_modulation_mae_db": 0.5303663719536765, "300–1000 Hz_40ms_log_rms_mae_db": 0.6603069466126774, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046287688836560446, "300–1000 Hz_40ms_envelope_correlation": 0.9976426219513632, "300–1000 Hz_40ms_modulation_mae_db": 0.6197933922093494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.047061744229654, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1438020382422875, "1000–3000 Hz_10ms_envelope_correlation": 0.982381805199876, "1000–3000 Hz_10ms_modulation_mae_db": 0.3637798054292178, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5599824748507016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09288336933716815, "1000–3000 Hz_40ms_envelope_correlation": 0.9926428253462803, "1000–3000 Hz_40ms_modulation_mae_db": 0.2994581069331735, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1112586742605841, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14089103857141472, "3000–6000 Hz_10ms_envelope_correlation": 0.9790680648732325, "3000–6000 Hz_10ms_modulation_mae_db": 0.5831603735927577, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6577281703538197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07578980682984592, "3000–6000 Hz_40ms_envelope_correlation": 0.9937352730254205, "3000–6000 Hz_40ms_modulation_mae_db": 0.6460459001939877, "mr_stft_db": 5.69186267543704, "band_mae_db": 0.11298315502616298, "rms_gain_error_db": -0.022504220312088808, "output_peak": 0.43123915234092813}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8826154051924695, "全帯域_10ms_rms_envelope_nrmse": 0.10264721972080434, "全帯域_10ms_envelope_correlation": 0.9851416074942413, "全帯域_10ms_modulation_mae_db": 0.3487825881438298, "全帯域_40ms_log_rms_mae_db": 0.46939162702849696, "全帯域_40ms_rms_envelope_nrmse": 0.05277044854633481, "全帯域_40ms_envelope_correlation": 0.9956239509035685, "全帯域_40ms_modulation_mae_db": 0.2980257747958739, "300–1000 Hz_10ms_log_rms_mae_db": 1.0988815696984673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0921673864648636, "300–1000 Hz_10ms_envelope_correlation": 0.99155184951926, "300–1000 Hz_10ms_modulation_mae_db": 0.48995453541911244, "300–1000 Hz_40ms_log_rms_mae_db": 0.581646152638086, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0529050301138186, "300–1000 Hz_40ms_envelope_correlation": 0.9970976221106717, "300–1000 Hz_40ms_modulation_mae_db": 0.6711944293203859, "1000–3000 Hz_10ms_log_rms_mae_db": 1.102194048996169, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13185018561238723, "1000–3000 Hz_10ms_envelope_correlation": 0.9843234191292946, "1000–3000 Hz_10ms_modulation_mae_db": 0.6947452538093527, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5824235829166134, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05976514585449463, "1000–3000 Hz_40ms_envelope_correlation": 0.9966391460510137, "1000–3000 Hz_40ms_modulation_mae_db": 0.6700798316822999, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1631274144854589, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16285032240402939, "3000–6000 Hz_10ms_envelope_correlation": 0.9719428739187534, "3000–6000 Hz_10ms_modulation_mae_db": 0.3324541927689979, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6831623261886641, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08753297519170372, "3000–6000 Hz_40ms_envelope_correlation": 0.9911713039974106, "3000–6000 Hz_40ms_modulation_mae_db": 0.14933624104648052, "mr_stft_db": 5.702252229338726, "band_mae_db": 0.12566595675295802, "rms_gain_error_db": 0.014845637128255807, "output_peak": 0.4343673327540343}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49200, "output_samples": 49200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.485954887838408, "spectral_convergence": 0.43996700100391184, "spectral_convergence_status": "ok", "amplitude_floor": 6.39186387874099e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.317347444334282, "spectral_convergence": 0.4081770104282217, "spectral_convergence_status": "ok", "amplitude_floor": 4.1641566181737356e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.3486979038369595, "spectral_convergence": 0.44728642975424526, "spectral_convergence_status": "ok", "amplitude_floor": 3.471921787916526e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.471560290376807e-05, "output_power": 7.393705020856768e-05, "delta_db": -0.04549185284717028, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002820835097806379, "output_power": 0.00028944344375575926, "delta_db": 0.111860180876162, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026780543872196867, "output_power": 0.0002748737212692904, "delta_db": 0.11313829379648119, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028630139823372175, "output_power": 0.0002832747775398633, "delta_db": -0.046155618654388234, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006167999147673381, "output_power": 0.0005759516938259866, "delta_db": -0.29758245190916266, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00037968124815517825, "output_power": 0.0003997472445989818, "delta_db": 0.2236633033332288, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13964861690276553, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.023193341994383873, "pitch": null, "output_rms_dbfs": -27.218804902419315, "output_peak": 0.45724582488060345, "reference_rms_dbfs": -27.195611560424933, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.717333412003216}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 0.00010992699921133752, "log_rms_mae_db": 0.8613686437382686, "rms_envelope_nrmse": 0.10193537884305282, "envelope_correlation": 0.9852873900180925, "rms_gain_error_db": -0.02329571322703801, "reference_mean_power": 0.0019104810338505282, "output_mean_power": 0.001900260580371364}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 4.701235881229352e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003075331225832077, "output_power": 0.0002977083039294638, "reference_active": true, "delta_db": -0.14100949412703453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8607043791082874e-05, "output_power": 2.8297924185835645e-05, "reference_active": true, "delta_db": -0.04718402069808083}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1458430393639735e-05, "output_power": 1.575821208675093e-05, "reference_active": true, "delta_db": 1.383818103796322}], "active_band_count": 3, "active_band_mae_db": 0.5240038728738124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 9.464213695185308e-05, "log_rms_mae_db": 0.4516456189568147, "rms_envelope_nrmse": 0.05938463837259588, "envelope_correlation": 0.9944269807076431, "rms_gain_error_db": -0.023331789561499207, "reference_mean_power": 0.0019290839559002245, "output_mean_power": 0.0019187480439440275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 4.127760176364045e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002739478020838986, "output_power": 0.0002651883105594205, "reference_active": true, "delta_db": -0.14113443835554246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7486141765201905e-05, "output_power": 1.694558542825308e-05, "reference_active": true, "delta_db": -0.13637417779846622}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8391125647823e-07, "output_power": 7.995794496224796e-07, "reference_active": true, "delta_db": 1.3651477603753837}], "active_band_count": 3, "active_band_mae_db": 0.5475521255097974}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 6.478025356347348e-05, "log_rms_mae_db": 1.1418927110041923, "rms_envelope_nrmse": 0.08780131506545463, "envelope_correlation": 0.992289842812512, "rms_gain_error_db": 0.08262888215980259, "reference_mean_power": 0.0004974887922363912, "output_mean_power": 0.0005070446320675386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.1951630543463553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001563474072286115, "output_power": 0.00015789588101855576, "reference_active": true, "delta_db": 0.04280117288923997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.160597194093242e-06, "output_power": 8.671573993372923e-06, "reference_active": true, "delta_db": 0.2637599255127749}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0536955073753773e-06, "output_power": 3.3790264729490547e-06, "reference_active": true, "delta_db": 2.162555412098701}], "active_band_count": 3, "active_band_mae_db": 0.823038836833572}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 5.647638237474489e-05, "log_rms_mae_db": 0.6022729554706745, "rms_envelope_nrmse": 0.04286391036090096, "envelope_correlation": 0.9980956152275692, "rms_gain_error_db": 0.08262583512031607, "reference_mean_power": 0.00050239739349007, "output_mean_power": 0.0005120471592193095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.0265825460069485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014180308081562638, "output_power": 0.00014308417756079132, "reference_active": true, "delta_db": 0.03905945114554068}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4691763208394e-06, "output_power": 4.6451995918880165e-06, "reference_active": true, "delta_db": 0.1677689014529256}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.310040798443831e-08, "output_power": 1.4201806114222525e-07, "reference_active": true, "delta_db": 1.833919951317406}], "active_band_count": 3, "active_band_mae_db": 0.6802494346386241}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 7.855020595588598e-05, "log_rms_mae_db": 1.0695779904844962, "rms_envelope_nrmse": 0.12096940657977973, "envelope_correlation": 0.9865726172639865, "rms_gain_error_db": -0.02670934753665926, "reference_mean_power": 0.0004940955013379756, "output_mean_power": 0.0004910661120558372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 1.3474634060369984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.364408948380063e-05, "output_power": 9.165393373644886e-05, "reference_active": true, "delta_db": -0.09329261915339739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.996833569277459e-06, "output_power": 6.5736601505349885e-06, "reference_active": true, "delta_db": -0.27094295775432664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2622375016010212e-06, "output_power": 3.740099735625929e-06, "reference_active": true, "delta_db": 2.183449859947075}], "active_band_count": 3, "active_band_mae_db": 0.8492284789515997}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 6.30584220600245e-05, "log_rms_mae_db": 0.5636581759861315, "rms_envelope_nrmse": 0.0523497912955623, "envelope_correlation": 0.9973425157381257, "rms_gain_error_db": -0.02670775180132276, "reference_mean_power": 0.0004989699589215838, "output_mean_power": 0.0004959108656677557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 1.243238840477423e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.827683631948728e-05, "output_power": 8.637405413052273e-05, "reference_active": true, "delta_db": -0.09463455934097746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8441959267659476e-06, "output_power": 4.0192047589675286e-06, "reference_active": true, "delta_db": 0.1933461774491352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.483409815447214e-08, "output_power": 1.8016633691304106e-07, "reference_active": true, "delta_db": 2.787091301549209}], "active_band_count": 3, "active_band_mae_db": 1.0250240127797738}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 9.512970769085293e-05, "log_rms_mae_db": 1.1509911527623988, "rms_envelope_nrmse": 0.17164097092717706, "envelope_correlation": 0.9692015782541002, "rms_gain_error_db": -0.12167264672378825, "reference_mean_power": 0.0007033343720650757, "output_mean_power": 0.0006839031078603319}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.9605352394438525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001169604983627503, "output_power": 0.0001083007199214842, "reference_active": true, "delta_db": -0.33407866547281795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1203242070906313e-05, "output_power": 2.0145302643058246e-05, "reference_active": true, "delta_db": -0.2222847539899273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.096712651632301e-05, "output_power": 1.3666768131203523e-05, "reference_active": true, "delta_db": 0.9557297278997612}], "active_band_count": 3, "active_band_mae_db": 0.5040310491208354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 7.035164526339387e-05, "log_rms_mae_db": 0.6557298353108564, "rms_envelope_nrmse": 0.11137871195778272, "envelope_correlation": 0.9858852085159018, "rms_gain_error_db": -0.12167237460727756, "reference_mean_power": 0.0007102740547296548, "output_mean_power": 0.0006906511087637765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.6373521771123283e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010173798550296795, "output_power": 9.416506963009072e-05, "reference_active": true, "delta_db": -0.3359330210953132}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.290709427246265e-05, "output_power": 1.2218713388643832e-05, "reference_active": true, "delta_db": -0.23803004335107986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.316189457175965e-07, "output_power": 7.360398284566275e-07, "reference_active": true, "delta_db": 0.6644616703338262}], "active_band_count": 3, "active_band_mae_db": 0.4128082449267398}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8528278346755117, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6b53fbf1e4b71920bc1a866916c15c8d2a3bbf0e85ea7d9d3c3f56dafef2def7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9844860305765204, "standard_deviation": 0.0017853330878644825, "minimum": 0.9819018577435896, "maximum": 0.9865726172639865, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0872094247079882, "standard_deviation": 0.023234892860968234, "minimum": 1.047061744229654, "maximum": 1.122828831866373, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7171531955906516, "standard_deviation": 0.3355788062296499, "minimum": 0.3637798054292178, "maximum": 1.4542007688768914, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1310140812644992, "standard_deviation": 0.008607817536348896, "minimum": 0.12096940657977973, "maximum": 0.1438020382422875, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9959296480411017, "standard_deviation": 0.0016859250442335153, "minimum": 0.9926428253462803, "maximum": 0.9973425157381257, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.5755350604006334, "standard_deviation": 0.016044853692450765, "minimum": 0.5495778422023919, "maximum": 0.5922848698539187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8292509001142955, "standard_deviation": 0.41895165015647845, "minimum": 0.2994581069331735, "maximum": 1.7424535777692431, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06603418147875417, "standard_deviation": 0.013398450623052399, "minimum": 0.0523497912955623, "maximum": 0.09288336933716815, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9753867099619237, "standard_deviation": 0.0035586781358614484, "minimum": 0.9692015782541002, "maximum": 0.9802749167416247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1275591031890684, "standard_deviation": 0.03716048793972167, "minimum": 1.0477389896045726, "maximum": 1.1631274144854589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5481038148536558, "standard_deviation": 0.21971804278844748, "minimum": 0.3324541927689979, "maximum": 1.052062566687123, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15327528149995578, "standard_deviation": 0.010421862296055007, "minimum": 0.14024125399797935, "maximum": 0.17164097092717706, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926201399942959, "standard_deviation": 0.0029062369308883794, "minimum": 0.9858852085159018, "maximum": 0.9942871184435973, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6408873319499674, "standard_deviation": 0.0310159193054004, "minimum": 0.5881026637903787, "maximum": 0.6831623261886641, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6426428257584519, "standard_deviation": 0.3777449753010889, "minimum": 0.14933624104648052, "maximum": 1.450425831247952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0829927862142332, "standard_deviation": 0.01250751715019175, "minimum": 0.07214865393132323, "maximum": 0.11137871195778272, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904083257350685, "standard_deviation": 0.001555182469466534, "minimum": 0.9879367857415863, "maximum": 0.992289842812512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1464956860401996, "standard_deviation": 0.050048453224554125, "minimum": 1.0950334239585107, "maximum": 1.2412755130516733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5985244894682473, "standard_deviation": 0.2972191997490206, "minimum": 0.1788532974250737, "maximum": 0.9255490990512688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09725141692726014, "standard_deviation": 0.00788895203071455, "minimum": 0.08780131506545463, "maximum": 0.109196724903189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969782591553089, "standard_deviation": 0.000977927898111276, "minimum": 0.9953967334647452, "maximum": 0.9980956152275692, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6260153686128441, "standard_deviation": 0.056167142640284835, "minimum": 0.581646152638086, "maximum": 0.7390884139810021, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.683588603695835, "standard_deviation": 0.23033353764771394, "minimum": 0.3745062183402494, "maximum": 1.0294457773882755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05278430382109164, "standard_deviation": 0.007916872710764874, "minimum": 0.04286391036090096, "maximum": 0.06456497305911023, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14212098904568998, "standard_deviation": 0.03428602921038126, "minimum": 0.08532747473664039, "maximum": 0.1933913220597602, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.697032488183378, "standard_deviation": 0.014315103219092554, "minimum": 5.671238107339273, "maximum": 5.717333412003216, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4456198373869702, "standard_deviation": 0.018919804104651047, "minimum": 0.4281474044147247, "maximum": 0.48226345813363825, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06903193944369314, "standard_deviation": 0.08825889462286869, "minimum": -0.2048193357235074, "maximum": 0.014845637128255807, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9864182714548533, "standard_deviation": 0.0014746439457554595, "minimum": 0.9847962825932229, "maximum": 0.9893035426912526, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8658716287889892, "standard_deviation": 0.034697258951499416, "minimum": 0.8167895637295718, "maximum": 0.9177509690987238, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.424763095967313, "standard_deviation": 0.14097640658193836, "minimum": 0.2469983371258229, "maximum": 0.6809417664350481, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09831713046532214, "standard_deviation": 0.005020457016411263, "minimum": 0.08856043537693134, "maximum": 0.10363944260001673, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9961828016490619, "standard_deviation": 0.0009292515905061746, "minimum": 0.9944269807076431, "maximum": 0.9972634078117163, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.456602859999812, "standard_deviation": 0.012748301929721196, "minimum": 0.4403648810864724, "maximum": 0.47794664781230417, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44717957933073976, "standard_deviation": 0.2806945086845308, "minimum": 0.17129398826232345, "maximum": 1.071020627409798, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05041394821335014, "standard_deviation": 0.005214713153657364, "minimum": 0.041844509139581765, "maximum": 0.05938463837259588, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-81414-0026", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.442427396774292, "predicted_mos": 2.442427396774292, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "f8f4336efaf18ce0077cfa2c23b46ccc9ccb17b297ccef22e504db084105c3c8", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.742913099937141, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8272749865249177, "全帯域_10ms_rms_envelope_nrmse": 0.099253333783937, "全帯域_10ms_envelope_correlation": 0.9860455348513302, "全帯域_10ms_modulation_mae_db": 0.281871120305956, "全帯域_40ms_log_rms_mae_db": 0.3573932089215751, "全帯域_40ms_rms_envelope_nrmse": 0.04278972119939521, "全帯域_40ms_envelope_correlation": 0.997168289993354, "全帯域_40ms_modulation_mae_db": 0.34701739369750034, "300–1000 Hz_10ms_log_rms_mae_db": 1.5072198718888465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10827069666018281, "300–1000 Hz_10ms_envelope_correlation": 0.9882008641038297, "300–1000 Hz_10ms_modulation_mae_db": 1.0036183942131787, "300–1000 Hz_40ms_log_rms_mae_db": 0.9620621285049321, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05388502625727817, "300–1000 Hz_40ms_envelope_correlation": 0.9969266943418489, "300–1000 Hz_40ms_modulation_mae_db": 0.9804992478659127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4448773161104085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1660154868520019, "1000–3000 Hz_10ms_envelope_correlation": 0.9744120015522141, "1000–3000 Hz_10ms_modulation_mae_db": 0.5986542535251892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8191770177586526, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0879069294950723, "1000–3000 Hz_40ms_envelope_correlation": 0.9925273908692781, "1000–3000 Hz_40ms_modulation_mae_db": 0.5763979233565318, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3086141517961405, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13938268675045282, "3000–6000 Hz_10ms_envelope_correlation": 0.9794404765941503, "3000–6000 Hz_10ms_modulation_mae_db": 0.6345242546628775, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6916334845162273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06210375648029297, "3000–6000 Hz_40ms_envelope_correlation": 0.9958002184575053, "3000–6000 Hz_40ms_modulation_mae_db": 0.681968823717649, "mr_stft_db": 5.887963494765885, "band_mae_db": 0.12850138552415083, "rms_gain_error_db": -0.056721941009594205, "output_peak": 0.4406083656631514}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7371125234920672, "全帯域_10ms_rms_envelope_nrmse": 0.08953757823215559, "全帯域_10ms_envelope_correlation": 0.9886865621474762, "全帯域_10ms_modulation_mae_db": 0.42757368736773066, "全帯域_40ms_log_rms_mae_db": 0.3446719330083214, "全帯域_40ms_rms_envelope_nrmse": 0.03747977362566745, "全帯域_40ms_envelope_correlation": 0.9978966000136245, "全帯域_40ms_modulation_mae_db": 0.4731108859810485, "300–1000 Hz_10ms_log_rms_mae_db": 1.3725316782170804, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10285390033763761, "300–1000 Hz_10ms_envelope_correlation": 0.989164899996954, "300–1000 Hz_10ms_modulation_mae_db": 0.5210159440210343, "300–1000 Hz_40ms_log_rms_mae_db": 0.8753555096875956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05363880326317315, "300–1000 Hz_40ms_envelope_correlation": 0.996880882215151, "300–1000 Hz_40ms_modulation_mae_db": 0.5023618952791048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.347193834957147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13739237710701507, "1000–3000 Hz_10ms_envelope_correlation": 0.9825655614987283, "1000–3000 Hz_10ms_modulation_mae_db": 0.20755451381990486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8121122892022296, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07037074491904068, "1000–3000 Hz_40ms_envelope_correlation": 0.995442820231407, "1000–3000 Hz_40ms_modulation_mae_db": 0.26622708794381933, "3000–6000 Hz_10ms_log_rms_mae_db": 1.387430081416975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14452734825076824, "3000–6000 Hz_10ms_envelope_correlation": 0.9777088060533086, "3000–6000 Hz_10ms_modulation_mae_db": 0.4145212062071037, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7543782015147417, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06768499507650173, "3000–6000 Hz_40ms_envelope_correlation": 0.9946872270245632, "3000–6000 Hz_40ms_modulation_mae_db": 0.5255811349357192, "mr_stft_db": 5.884525368268747, "band_mae_db": 0.08175267898015437, "rms_gain_error_db": -0.06772119960037909, "output_peak": 0.4348875810166388}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8007946039216101, "全帯域_10ms_rms_envelope_nrmse": 0.09679508110473252, "全帯域_10ms_envelope_correlation": 0.9871206370287504, "全帯域_10ms_modulation_mae_db": 0.3256468460534945, "全帯域_40ms_log_rms_mae_db": 0.3903706704962961, "全帯域_40ms_rms_envelope_nrmse": 0.04676217558981377, "全帯域_40ms_envelope_correlation": 0.9973167655739579, "全帯域_40ms_modulation_mae_db": 0.2477680761426606, "300–1000 Hz_10ms_log_rms_mae_db": 1.2568753988043317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0988705795214959, "300–1000 Hz_10ms_envelope_correlation": 0.9899890374430188, "300–1000 Hz_10ms_modulation_mae_db": 0.5634633571876632, "300–1000 Hz_40ms_log_rms_mae_db": 0.7485152223559911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04868956221342297, "300–1000 Hz_40ms_envelope_correlation": 0.9974554707310758, "300–1000 Hz_40ms_modulation_mae_db": 0.5738719572926717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.334591310652017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1394007614438594, "1000–3000 Hz_10ms_envelope_correlation": 0.982213199333524, "1000–3000 Hz_10ms_modulation_mae_db": 0.5404300936896408, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7568321979662221, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06083955565679315, "1000–3000 Hz_40ms_envelope_correlation": 0.9969817049949035, "1000–3000 Hz_40ms_modulation_mae_db": 0.7739086221649272, "3000–6000 Hz_10ms_log_rms_mae_db": 1.346276516743502, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15166218460288586, "3000–6000 Hz_10ms_envelope_correlation": 0.9762754475569816, "3000–6000 Hz_10ms_modulation_mae_db": 0.2856672814589482, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7233243579547929, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07957297049911914, "3000–6000 Hz_40ms_envelope_correlation": 0.9943059386671087, "3000–6000 Hz_40ms_modulation_mae_db": 0.22442174171140486, "mr_stft_db": 5.87456778802775, "band_mae_db": 0.1654092558255079, "rms_gain_error_db": -0.20104877881809088, "output_peak": 0.4444472368750081}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8345774251059704, "全帯域_10ms_rms_envelope_nrmse": 0.09415693255443897, "全帯域_10ms_envelope_correlation": 0.9875436309497273, "全帯域_10ms_modulation_mae_db": 0.3970480518939497, "全帯域_40ms_log_rms_mae_db": 0.41628790352767897, "全帯域_40ms_rms_envelope_nrmse": 0.04519116531165295, "全帯域_40ms_envelope_correlation": 0.9970414594808671, "全帯域_40ms_modulation_mae_db": 0.2972810735220682, "300–1000 Hz_10ms_log_rms_mae_db": 1.4649985432855577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10752202323837794, "300–1000 Hz_10ms_envelope_correlation": 0.9881669172927954, "300–1000 Hz_10ms_modulation_mae_db": 0.5904703062488604, "300–1000 Hz_40ms_log_rms_mae_db": 0.9487186782081309, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0584430620165955, "300–1000 Hz_40ms_envelope_correlation": 0.9962793705358256, "300–1000 Hz_40ms_modulation_mae_db": 0.9578900725224422, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3809247971517142, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14671566121386517, "1000–3000 Hz_10ms_envelope_correlation": 0.9801013186371139, "1000–3000 Hz_10ms_modulation_mae_db": 0.4457635831875142, "1000–3000 Hz_40ms_log_rms_mae_db": 0.845784390973472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08453663338818423, "1000–3000 Hz_40ms_envelope_correlation": 0.9930311004037224, "1000–3000 Hz_40ms_modulation_mae_db": 0.6086230221228048, "3000–6000 Hz_10ms_log_rms_mae_db": 1.325727827251148, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14884399697532835, "3000–6000 Hz_10ms_envelope_correlation": 0.9769039046136315, "3000–6000 Hz_10ms_modulation_mae_db": 0.5603638309713072, "3000–6000 Hz_40ms_log_rms_mae_db": 0.730442929243106, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07533348328469161, "3000–6000 Hz_40ms_envelope_correlation": 0.9944690307467492, "3000–6000 Hz_40ms_modulation_mae_db": 0.2953093266626434, "mr_stft_db": 5.885910630776394, "band_mae_db": 0.14484546658815742, "rms_gain_error_db": -0.12081297130390169, "output_peak": 0.439412465407298}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8081733707976878, "全帯域_10ms_rms_envelope_nrmse": 0.10028743339871993, "全帯域_10ms_envelope_correlation": 0.985823853872759, "全帯域_10ms_modulation_mae_db": 0.05341720091520358, "全帯域_40ms_log_rms_mae_db": 0.33856853669318726, "全帯域_40ms_rms_envelope_nrmse": 0.04269490278150825, "全帯域_40ms_envelope_correlation": 0.9971369849945849, "全帯域_40ms_modulation_mae_db": 0.15956220566396226, "300–1000 Hz_10ms_log_rms_mae_db": 1.3964390639108193, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10662488433428899, "300–1000 Hz_10ms_envelope_correlation": 0.9883824452256527, "300–1000 Hz_10ms_modulation_mae_db": 0.3959716883906333, "300–1000 Hz_40ms_log_rms_mae_db": 0.9026611047256007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05787437660698339, "300–1000 Hz_40ms_envelope_correlation": 0.9963262668354657, "300–1000 Hz_40ms_modulation_mae_db": 0.5224098723912508, "1000–3000 Hz_10ms_log_rms_mae_db": 1.363944176822349, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15038690052306267, "1000–3000 Hz_10ms_envelope_correlation": 0.97912773325059, "1000–3000 Hz_10ms_modulation_mae_db": 0.30796655685974134, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7817468481254961, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06725719920631909, "1000–3000 Hz_40ms_envelope_correlation": 0.9955937353992235, "1000–3000 Hz_40ms_modulation_mae_db": 0.5787241705374719, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2951427072075352, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1550348363295729, "3000–6000 Hz_10ms_envelope_correlation": 0.9749127447371791, "3000–6000 Hz_10ms_modulation_mae_db": 0.08274246026902773, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6860146582607062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0719905309693218, "3000–6000 Hz_40ms_envelope_correlation": 0.9941132280264696, "3000–6000 Hz_40ms_modulation_mae_db": 0.3186252224389836, "mr_stft_db": 5.886475595353869, "band_mae_db": 0.08424867740528492, "rms_gain_error_db": 0.01348094008970741, "output_peak": 0.43094694424258007}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7997600509011749, "全帯域_10ms_rms_envelope_nrmse": 0.09329114015331742, "全帯域_10ms_envelope_correlation": 0.9877327199652907, "全帯域_10ms_modulation_mae_db": 0.3217706391493735, "全帯域_40ms_log_rms_mae_db": 0.41471938558440374, "全帯域_40ms_rms_envelope_nrmse": 0.04276685368484555, "全帯域_40ms_envelope_correlation": 0.9971929375076538, "全帯域_40ms_modulation_mae_db": 0.6654898579769788, "300–1000 Hz_10ms_log_rms_mae_db": 1.4172509029607738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10945912375433532, "300–1000 Hz_10ms_envelope_correlation": 0.9878265390579481, "300–1000 Hz_10ms_modulation_mae_db": 0.4838484669659126, "300–1000 Hz_40ms_log_rms_mae_db": 1.0021117585279729, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06500342474463362, "300–1000 Hz_40ms_envelope_correlation": 0.9954035325046455, "300–1000 Hz_40ms_modulation_mae_db": 0.5301303856394036, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3350124519733688, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14259465200545093, "1000–3000 Hz_10ms_envelope_correlation": 0.9818962307066748, "1000–3000 Hz_10ms_modulation_mae_db": 0.7275176682567404, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7611278183697824, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0673438692549682, "1000–3000 Hz_40ms_envelope_correlation": 0.9958702125493881, "1000–3000 Hz_40ms_modulation_mae_db": 0.9985079762860573, "3000–6000 Hz_10ms_log_rms_mae_db": 1.326655500273224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13584825525803457, "3000–6000 Hz_10ms_envelope_correlation": 0.9810566937393121, "3000–6000 Hz_10ms_modulation_mae_db": 0.915567007582388, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7717192742068831, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06884480645980821, "3000–6000 Hz_40ms_envelope_correlation": 0.9957422762617962, "3000–6000 Hz_40ms_modulation_mae_db": 1.271370469686967, "mr_stft_db": 5.873961364628059, "band_mae_db": 0.1084732679525508, "rms_gain_error_db": -0.047850571195187014, "output_peak": 0.4438241438582437}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8110238680897449, "全帯域_10ms_rms_envelope_nrmse": 0.09373280639336803, "全帯域_10ms_envelope_correlation": 0.9875234967348636, "全帯域_10ms_modulation_mae_db": 0.29276490092581925, "全帯域_40ms_log_rms_mae_db": 0.41736649393854397, "全帯域_40ms_rms_envelope_nrmse": 0.048281294867120315, "全帯域_40ms_envelope_correlation": 0.9964431654133191, "全帯域_40ms_modulation_mae_db": 0.44135750413807245, "300–1000 Hz_10ms_log_rms_mae_db": 1.3728453525534903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1050111118111551, "300–1000 Hz_10ms_envelope_correlation": 0.9889647464464988, "300–1000 Hz_10ms_modulation_mae_db": 0.45960826480776884, "300–1000 Hz_40ms_log_rms_mae_db": 0.8932715717853125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047791796999065786, "300–1000 Hz_40ms_envelope_correlation": 0.9976658114958431, "300–1000 Hz_40ms_modulation_mae_db": 0.517765384757511, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4283352543204628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14521535923889736, "1000–3000 Hz_10ms_envelope_correlation": 0.9806937802732366, "1000–3000 Hz_10ms_modulation_mae_db": 0.9377786988945774, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8014889202377479, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07695582853256486, "1000–3000 Hz_40ms_envelope_correlation": 0.9947767061226049, "1000–3000 Hz_40ms_modulation_mae_db": 0.9098378055382361, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3521919369530695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14412155841472238, "3000–6000 Hz_10ms_envelope_correlation": 0.9781147864282501, "3000–6000 Hz_10ms_modulation_mae_db": 0.3324989092115873, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7225208257248728, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07617441856370198, "3000–6000 Hz_40ms_envelope_correlation": 0.9937209500708287, "3000–6000 Hz_40ms_modulation_mae_db": 0.4895832087672764, "mr_stft_db": 5.900678578260735, "band_mae_db": 0.16934371955143981, "rms_gain_error_db": -0.08421826744498359, "output_peak": 0.46008632694731505}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8095037272276426, "全帯域_10ms_rms_envelope_nrmse": 0.09488516494552603, "全帯域_10ms_envelope_correlation": 0.9872371259906146, "全帯域_10ms_modulation_mae_db": 0.7000141068151216, "全帯域_40ms_log_rms_mae_db": 0.42684066505386914, "全帯域_40ms_rms_envelope_nrmse": 0.04501929597952231, "全帯域_40ms_envelope_correlation": 0.9968021719822603, "全帯域_40ms_modulation_mae_db": 0.6405272653695392, "300–1000 Hz_10ms_log_rms_mae_db": 1.2837216947572192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10117499811103645, "300–1000 Hz_10ms_envelope_correlation": 0.9895559125727078, "300–1000 Hz_10ms_modulation_mae_db": 0.3575940692958593, "300–1000 Hz_40ms_log_rms_mae_db": 0.8217027664767039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056679975881194486, "300–1000 Hz_40ms_envelope_correlation": 0.9964915006063395, "300–1000 Hz_40ms_modulation_mae_db": 0.3624588696131898, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3342796078786683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13176922314617823, "1000–3000 Hz_10ms_envelope_correlation": 0.9839762469045455, "1000–3000 Hz_10ms_modulation_mae_db": 0.29946400070757606, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7131762211731589, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05896485332061213, "1000–3000 Hz_40ms_envelope_correlation": 0.9966142781037964, "1000–3000 Hz_40ms_modulation_mae_db": 0.5136401953306606, "3000–6000 Hz_10ms_log_rms_mae_db": 1.355890643495678, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14694833449868352, "3000–6000 Hz_10ms_envelope_correlation": 0.9773701212247667, "3000–6000 Hz_10ms_modulation_mae_db": 1.0652417965111214, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7379761120505258, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07716814064925855, "3000–6000 Hz_40ms_envelope_correlation": 0.9932879177425449, "3000–6000 Hz_40ms_modulation_mae_db": 0.9166113458635081, "mr_stft_db": 5.878469796363507, "band_mae_db": 0.08038945344961407, "rms_gain_error_db": -0.019412691141324705, "output_peak": 0.43418127402615075}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49200, "output_samples": 49200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.684937973049688, "spectral_convergence": 0.44873357974734296, "spectral_convergence_status": "ok", "amplitude_floor": 6.39186387874099e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.537967967678216, "spectral_convergence": 0.41135102789087696, "spectral_convergence_status": "ok", "amplitude_floor": 4.1641566181737356e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.440984543569753, "spectral_convergence": 0.44179301701257223, "spectral_convergence_status": "ok", "amplitude_floor": 3.471921787916526e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.471560290376807e-05, "output_power": 7.117167678733824e-05, "delta_db": -0.2110410744189472, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002820835097806379, "output_power": 0.0002882304561455359, "delta_db": 0.09362170411433772, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026780543872196867, "output_power": 0.0002753102102455606, "delta_db": 0.120029254535916, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028630139823372175, "output_power": 0.00027481980493361514, "delta_db": -0.17775442009256223, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006167999147673381, "output_power": 0.0005936103097190766, "delta_db": -0.16642869943431907, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00037968124815517825, "output_power": 0.00037949480273205105, "delta_db": -0.0021331605488227236, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12850138552415083, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.056721941009594205, "pitch": null, "output_rms_dbfs": -27.252333501434528, "output_peak": 0.4406083656631514, "reference_rms_dbfs": -27.195611560424933, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.887963494765885}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 0.00010992699921133752, "log_rms_mae_db": 0.8272749865249177, "rms_envelope_nrmse": 0.099253333783937, "envelope_correlation": 0.9860455348513302, "rms_gain_error_db": -0.05670733823284902, "reference_mean_power": 0.0019104810338505282, "output_mean_power": 0.001885697376266291}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 4.701235881229352e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003075331225832077, "output_power": 0.0003080298152769599, "reference_active": true, "delta_db": 0.007008574741771766}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8607043791082874e-05, "output_power": 2.634875882458017e-05, "reference_active": true, "delta_db": -0.3571281847846322}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1458430393639735e-05, "output_power": 1.0255987214090121e-05, "reference_active": true, "delta_db": -0.481476601391464}], "active_band_count": 3, "active_band_mae_db": 0.281871120305956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 9.464213695185308e-05, "log_rms_mae_db": 0.3573932089215751, "rms_envelope_nrmse": 0.04278972119939521, "envelope_correlation": 0.997168289993354, "rms_gain_error_db": -0.05642057067553206, "reference_mean_power": 0.0019290839559002245, "output_mean_power": 0.0019041847029119925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 4.127760176364045e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002739478020838986, "output_power": 0.0002739488087452036, "reference_active": true, "delta_db": 1.5958757227369904e-05}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7486141765201905e-05, "output_power": 1.514737287028564e-05, "reference_active": true, "delta_db": -0.6235667880449465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8391125647823e-07, "output_power": 5.303955895868214e-07, "reference_active": true, "delta_db": -0.4174694342903273}], "active_band_count": 3, "active_band_mae_db": 0.34701739369750034}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 6.478025356347348e-05, "log_rms_mae_db": 1.5072198718888465, "rms_envelope_nrmse": 0.10827069666018281, "envelope_correlation": 0.9882008641038297, "rms_gain_error_db": 0.07932585664720464, "reference_mean_power": 0.0004974887922363912, "output_mean_power": 0.0005066591460183994}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.1951630543463553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001563474072286115, "output_power": 0.00015519587617349054, "reference_active": true, "delta_db": -0.032105064504018996}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.160597194093242e-06, "output_power": 7.986080764398529e-06, "reference_active": true, "delta_db": -0.09388243707739086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0536955073753773e-06, "output_power": 3.990458707390239e-06, "reference_active": true, "delta_db": 2.8848676810581257}], "active_band_count": 3, "active_band_mae_db": 1.0036183942131787}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 5.647638237474489e-05, "log_rms_mae_db": 0.9620621285049321, "rms_envelope_nrmse": 0.05388502625727817, "envelope_correlation": 0.9969266943418489, "rms_gain_error_db": 0.07932516270453685, "reference_mean_power": 0.00050239739349007, "output_mean_power": 0.0005116581471727777}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.0265825460069485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014180308081562638, "output_power": 0.0001407542389906893, "reference_active": true, "delta_db": -0.032241833931390226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4691763208394e-06, "output_power": 4.255782915303492e-06, "reference_active": true, "delta_db": -0.21248022290429897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.310040798443831e-08, "output_power": 1.7323241220100495e-07, "reference_active": true, "delta_db": 2.6967756867620487}], "active_band_count": 3, "active_band_mae_db": 0.9804992478659127}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 7.855020595588598e-05, "log_rms_mae_db": 1.4448773161104085, "rms_envelope_nrmse": 0.1660154868520019, "envelope_correlation": 0.9744120015522141, "rms_gain_error_db": -0.10555568068019168, "reference_mean_power": 0.0004940955013379756, "output_mean_power": 0.00048223122906507996}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 1.3474634060369984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.364408948380063e-05, "output_power": 9.231499157006375e-05, "reference_active": true, "delta_db": -0.06208136926099371}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.996833569277459e-06, "output_power": 5.734953796357209e-06, "reference_active": true, "delta_db": -0.8637162029690506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2622375016010212e-06, "output_power": 2.764106144271561e-06, "reference_active": true, "delta_db": 0.8701651883455231}], "active_band_count": 3, "active_band_mae_db": 0.5986542535251892}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 6.30584220600245e-05, "log_rms_mae_db": 0.8191770177586526, "rms_envelope_nrmse": 0.0879069294950723, "envelope_correlation": 0.9925273908692781, "rms_gain_error_db": -0.10556304125008847, "reference_mean_power": 0.0004989699589215838, "output_mean_power": 0.000486987815304985}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 1.243238840477423e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.827683631948728e-05, "output_power": 8.754206547401032e-05, "reference_active": true, "delta_db": -0.03629971273510046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8441959267659476e-06, "output_power": 3.154331959067635e-06, "reference_active": true, "delta_db": -0.858981180459314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.483409815447214e-08, "output_power": 1.1490946594654113e-07, "reference_active": true, "delta_db": 0.8339128768751811}], "active_band_count": 3, "active_band_mae_db": 0.5763979233565318}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 9.512970769085293e-05, "log_rms_mae_db": 1.3086141517961405, "rms_envelope_nrmse": 0.13938268675045282, "envelope_correlation": 0.9794404765941503, "rms_gain_error_db": -0.1484513517015001, "reference_mean_power": 0.0007033343720650757, "output_mean_power": 0.0006796991186824017}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.9605352394438525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001169604983627503, "output_power": 0.00011643745394590705, "reference_active": true, "delta_db": -0.019465096590874094}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1203242070906313e-05, "output_power": 1.862281221601161e-05, "reference_active": true, "delta_db": -0.5635700752146521}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.096712651632301e-05, "output_power": 8.09168735582679e-06, "reference_active": true, "delta_db": -1.3205375921831064}], "active_band_count": 3, "active_band_mae_db": 0.6345242546628775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 7.035164526339387e-05, "log_rms_mae_db": 0.6916334845162273, "rms_envelope_nrmse": 0.06210375648029297, "envelope_correlation": 0.9958002184575053, "rms_gain_error_db": -0.14845424376931962, "reference_mean_power": 0.0007102740547296548, "output_mean_power": 0.000686405139159958}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.6373521771123283e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010173798550296795, "output_power": 0.00010132269331869926, "reference_active": true, "delta_db": -0.017764084756816647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.290709427246265e-05, "output_power": 1.0330515871264058e-05, "reference_active": true, "delta_db": -0.9670647285251552}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.316189457175965e-07, "output_power": 4.947062334235031e-07, "reference_active": true, "delta_db": -1.0610776578709749}], "active_band_count": 3, "active_band_mae_db": 0.681968823717649}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8379992507330117, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9847811f2474282e44912f6c3bf33b99597cc9ae89257e0d4e1c6d3d6ad737d3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9806232590195785, "standard_deviation": 0.0029339672180700733, "minimum": 0.9744120015522141, "maximum": 0.9839762469045455, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.371144843733267, "standard_deviation": 0.04380007271655528, "minimum": 1.3342796078786683, "maximum": 1.4448773161104085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5081411711176106, "standard_deviation": 0.2448932668514221, "minimum": 0.20755451381990486, "maximum": 0.9377786988945774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14493630269129135, "standard_deviation": 0.010307371514173939, "minimum": 0.13176922314617823, "maximum": 0.1660154868520019, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951047435842906, "standard_deviation": 0.001594247775490823, "minimum": 0.9925273908692781, "maximum": 0.9969817049949035, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7864307129758452, "standard_deviation": 0.04206447475961339, "minimum": 0.7131762211731589, "maximum": 0.845784390973472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6532333504100636, "standard_deviation": 0.23370567972110912, "minimum": 0.26622708794381933, "maximum": 0.9985079762860573, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07177195172169433, "standard_deviation": 0.010526432870706228, "minimum": 0.05896485332061213, "maximum": 0.0879069294950723, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9777228726184475, "standard_deviation": 0.001890023474613838, "minimum": 0.9749127447371791, "maximum": 0.9810566937393121, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3372411706421592, "standard_deviation": 0.029297227610662707, "minimum": 1.2951427072075352, "maximum": 1.387430081416975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5363908433592952, "standard_deviation": 0.32922984892368085, "minimum": 0.08274246026902773, "maximum": 1.0652417965111214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1457961501350561, "standard_deviation": 0.006266553197838371, "minimum": 0.13584825525803457, "maximum": 0.1550348363295729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945158483746956, "standard_deviation": 0.0008889107768821911, "minimum": 0.9932879177425449, "maximum": 0.9958002184575053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.727251230433982, "standard_deviation": 0.028897269443484395, "minimum": 0.6860146582607062, "maximum": 0.7717192742068831, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.590433909223019, "standard_deviation": 0.35612299128366803, "minimum": 0.22442174171140486, "maximum": 1.271370469686967, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.072359137747837, "standard_deviation": 0.005831880864937163, "minimum": 0.06210375648029297, "maximum": 0.07957297049911914, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9887814202674257, "standard_deviation": 0.0007582386862903539, "minimum": 0.9878265390579481, "maximum": 0.9899890374430188, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3839853132972648, "standard_deviation": 0.0840719107766602, "minimum": 1.2568753988043317, "maximum": 1.5072198718888465, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5469488113913639, "standard_deviation": 0.2005869269312118, "minimum": 0.3575940692958593, "maximum": 1.0036183942131787, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10497341472106378, "standard_deviation": 0.003710655303768844, "minimum": 0.0988705795214959, "maximum": 0.10945912375433532, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966786911582743, "standard_deviation": 0.0007196139466592561, "minimum": 0.9954035325046455, "maximum": 0.9976658114958431, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8942998425340299, "standard_deviation": 0.0811787208501849, "minimum": 0.7485152223559911, "maximum": 1.0021117585279729, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6184234606701858, "standard_deviation": 0.22502564171479464, "minimum": 0.3624588696131898, "maximum": 0.9804992478659127, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05525075349779338, "standard_deviation": 0.005572641878509662, "minimum": 0.047791796999065786, "maximum": 0.06500342474463362, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12037048815960752, "standard_deviation": 0.037087062491964716, "minimum": 0.08038945344961407, "maximum": 0.16934371955143981, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.884069077055618, "standard_deviation": 0.008653860902681498, "minimum": 5.873961364628059, "maximum": 5.900678578260735, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44104929225454825, "standard_deviation": 0.00903739795410973, "minimum": 0.43094694424258007, "maximum": 0.46008632694731505, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07303818505296922, "standard_deviation": 0.0655572761770415, "minimum": -0.20104877881809088, "maximum": 0.01348094008970741, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9872141951926015, "standard_deviation": 0.0009218787247681251, "minimum": 0.985823853872759, "maximum": 0.9886865621474762, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.803527569507602, "standard_deviation": 0.029457347871645154, "minimum": 0.7371125234920672, "maximum": 0.8345774251059704, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35001331917833106, "standard_deviation": 0.18047944091770518, "minimum": 0.05341720091520358, "maximum": 0.7000141068151216, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09524243382077444, "standard_deviation": 0.0034617895068058653, "minimum": 0.08953757823215559, "maximum": 0.10028743339871993, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971247968699527, "standard_deviation": 0.0004165965512889911, "minimum": 0.9964431654133191, "maximum": 0.9978966000136245, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.38827734965298444, "standard_deviation": 0.03614986779275794, "minimum": 0.33856853669318726, "maximum": 0.42684066505386914, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4090142828114788, "standard_deviation": 0.18090528008138762, "minimum": 0.15956220566396226, "maximum": 0.6654898579769788, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.043873147879940726, "standard_deviation": 0.00328394933123584, "minimum": 0.03747977362566745, "maximum": 0.048281294867120315, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-81414-0026", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.5709784030914307, "predicted_mos": 2.5709784030914307, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "61b1421215201e90b4c46fa941f8734dd6b386b60dfaa5b412d80a7b843f3a92", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8243871997110546, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.844908456643161, "全帯域_10ms_rms_envelope_nrmse": 0.09829381966281515, "全帯域_10ms_envelope_correlation": 0.9862625002395825, "全帯域_10ms_modulation_mae_db": 0.18704729410143403, "全帯域_40ms_log_rms_mae_db": 0.4392696015455033, "全帯域_40ms_rms_envelope_nrmse": 0.044146595534013275, "全帯域_40ms_envelope_correlation": 0.9970954008975432, "全帯域_40ms_modulation_mae_db": 0.29659406083372386, "300–1000 Hz_10ms_log_rms_mae_db": 1.1469331993454783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10457875552833822, "300–1000 Hz_10ms_envelope_correlation": 0.9888319420024425, "300–1000 Hz_10ms_modulation_mae_db": 0.718358581557427, "300–1000 Hz_40ms_log_rms_mae_db": 0.6548935732297853, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052410993092919356, "300–1000 Hz_40ms_envelope_correlation": 0.9969992723241861, "300–1000 Hz_40ms_modulation_mae_db": 0.9005610533597652, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0248459659717704, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13663332101443235, "1000–3000 Hz_10ms_envelope_correlation": 0.9827887857020854, "1000–3000 Hz_10ms_modulation_mae_db": 0.6467860426172756, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6069580202010233, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06528397968687587, "1000–3000 Hz_40ms_envelope_correlation": 0.995844555345558, "1000–3000 Hz_40ms_modulation_mae_db": 0.7765481402433035, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1057133154858523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15111889712484539, "3000–6000 Hz_10ms_envelope_correlation": 0.976239951836504, "3000–6000 Hz_10ms_modulation_mae_db": 0.5561357273772202, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6285357514121162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07839613655592562, "3000–6000 Hz_40ms_envelope_correlation": 0.9940188915506583, "3000–6000 Hz_40ms_modulation_mae_db": 0.8199794234944319, "mr_stft_db": 5.670065687863963, "band_mae_db": 0.13278929188716282, "rms_gain_error_db": -0.09301465672797257, "output_peak": 0.4538809902305671}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9089191085107494, "全帯域_10ms_rms_envelope_nrmse": 0.08893649448832018, "全帯域_10ms_envelope_correlation": 0.9887528091187814, "全帯域_10ms_modulation_mae_db": 0.3961134968945989, "全帯域_40ms_log_rms_mae_db": 0.4151847408399663, "全帯域_40ms_rms_envelope_nrmse": 0.04004787440426748, "全帯域_40ms_envelope_correlation": 0.9975604327063684, "全帯域_40ms_modulation_mae_db": 0.2833382998832203, "300–1000 Hz_10ms_log_rms_mae_db": 1.1487019059621952, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10770859410207084, "300–1000 Hz_10ms_envelope_correlation": 0.9882084701664247, "300–1000 Hz_10ms_modulation_mae_db": 1.0936467039993893, "300–1000 Hz_40ms_log_rms_mae_db": 0.5855264561497581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047837033501864995, "300–1000 Hz_40ms_envelope_correlation": 0.9974978513574125, "300–1000 Hz_40ms_modulation_mae_db": 1.6051693082290797, "1000–3000 Hz_10ms_log_rms_mae_db": 1.03778540393987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11978470235328997, "1000–3000 Hz_10ms_envelope_correlation": 0.9872966963741873, "1000–3000 Hz_10ms_modulation_mae_db": 0.3222492506897929, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6240500142592751, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0632541214269838, "1000–3000 Hz_40ms_envelope_correlation": 0.9969543754986746, "1000–3000 Hz_40ms_modulation_mae_db": 0.41677381919632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0397267226488367, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13099578801015238, "3000–6000 Hz_10ms_envelope_correlation": 0.9819651554673281, "3000–6000 Hz_10ms_modulation_mae_db": 0.15011396400430707, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5787531002580159, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.058766373715657766, "3000–6000 Hz_40ms_envelope_correlation": 0.996120999979565, "3000–6000 Hz_40ms_modulation_mae_db": 0.32729519175596505, "mr_stft_db": 5.673490019460956, "band_mae_db": 0.18450714559587578, "rms_gain_error_db": -0.06828710226546517, "output_peak": 0.46163888577185214}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8541424718034618, "全帯域_10ms_rms_envelope_nrmse": 0.09938821208530346, "全帯域_10ms_envelope_correlation": 0.9859335339652482, "全帯域_10ms_modulation_mae_db": 0.361289577065745, "全帯域_40ms_log_rms_mae_db": 0.4259462858748319, "全帯域_40ms_rms_envelope_nrmse": 0.045472649041582616, "全帯域_40ms_envelope_correlation": 0.9967481173453631, "全帯域_40ms_modulation_mae_db": 0.38057687005146296, "300–1000 Hz_10ms_log_rms_mae_db": 1.185366606711899, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11838215466615376, "300–1000 Hz_10ms_envelope_correlation": 0.985744585984882, "300–1000 Hz_10ms_modulation_mae_db": 0.9777303928025921, "300–1000 Hz_40ms_log_rms_mae_db": 0.6405489262712752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05982643750831834, "300–1000 Hz_40ms_envelope_correlation": 0.9960488021628238, "300–1000 Hz_40ms_modulation_mae_db": 0.9860723168935328, "1000–3000 Hz_10ms_log_rms_mae_db": 0.995932216937696, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12678584131454487, "1000–3000 Hz_10ms_envelope_correlation": 0.9853477602001314, "1000–3000 Hz_10ms_modulation_mae_db": 0.4109168726626911, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5486291683826319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07018652402503817, "1000–3000 Hz_40ms_envelope_correlation": 0.9954575509474298, "1000–3000 Hz_40ms_modulation_mae_db": 0.46485477914615747, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0518773363104492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.147868104752362, "3000–6000 Hz_10ms_envelope_correlation": 0.976934482149204, "3000–6000 Hz_10ms_modulation_mae_db": 0.3232761337981131, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5652712263590016, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06748596998933366, "3000–6000 Hz_40ms_envelope_correlation": 0.9947543212147049, "3000–6000 Hz_40ms_modulation_mae_db": 0.2879346878712005, "mr_stft_db": 5.6550222528699505, "band_mae_db": 0.1366875975080881, "rms_gain_error_db": -0.03187499180258923, "output_peak": 0.48770924116429093}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8672346267994607, "全帯域_10ms_rms_envelope_nrmse": 0.08837538557425016, "全帯域_10ms_envelope_correlation": 0.9889305725605686, "全帯域_10ms_modulation_mae_db": 0.6737958842842833, "全帯域_40ms_log_rms_mae_db": 0.4376850092451214, "全帯域_40ms_rms_envelope_nrmse": 0.03971163201935698, "全帯域_40ms_envelope_correlation": 0.997635537577991, "全帯域_40ms_modulation_mae_db": 1.0166297339422463, "300–1000 Hz_10ms_log_rms_mae_db": 1.1110361781219849, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09832047350105479, "300–1000 Hz_10ms_envelope_correlation": 0.9903390877633393, "300–1000 Hz_10ms_modulation_mae_db": 0.6840179483544021, "300–1000 Hz_40ms_log_rms_mae_db": 0.6741610449244867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05472771168593935, "300–1000 Hz_40ms_envelope_correlation": 0.9968201375075961, "300–1000 Hz_40ms_modulation_mae_db": 0.39134776628402457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0202780422412991, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1306036280039861, "1000–3000 Hz_10ms_envelope_correlation": 0.9848510089194167, "1000–3000 Hz_10ms_modulation_mae_db": 0.4590897951618486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5526708367516853, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07235064207397501, "1000–3000 Hz_40ms_envelope_correlation": 0.995977679279118, "1000–3000 Hz_40ms_modulation_mae_db": 0.8975306826077613, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0576117032541503, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13237402349355695, "3000–6000 Hz_10ms_envelope_correlation": 0.9817010876346783, "3000–6000 Hz_10ms_modulation_mae_db": 0.734305676948683, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5932577722548737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06511103063562342, "3000–6000 Hz_40ms_envelope_correlation": 0.9955433658404057, "3000–6000 Hz_40ms_modulation_mae_db": 0.9865116582326444, "mr_stft_db": 5.6645522738322756, "band_mae_db": 0.16817842740712452, "rms_gain_error_db": -0.08432311342851329, "output_peak": 0.42863637717438663}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8606931542425993, "全帯域_10ms_rms_envelope_nrmse": 0.10442486828204539, "全帯域_10ms_envelope_correlation": 0.9845427929126136, "全帯域_10ms_modulation_mae_db": 0.5116422943118656, "全帯域_40ms_log_rms_mae_db": 0.45187503043097754, "全帯域_40ms_rms_envelope_nrmse": 0.051195175496988723, "全帯域_40ms_envelope_correlation": 0.9960317027807226, "全帯域_40ms_modulation_mae_db": 0.49176241276788085, "300–1000 Hz_10ms_log_rms_mae_db": 1.1096033489454267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09441497641336699, "300–1000 Hz_10ms_envelope_correlation": 0.9908660538506017, "300–1000 Hz_10ms_modulation_mae_db": 0.7516705098750913, "300–1000 Hz_40ms_log_rms_mae_db": 0.681753451470327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05045644899008645, "300–1000 Hz_40ms_envelope_correlation": 0.9972306896015549, "300–1000 Hz_40ms_modulation_mae_db": 0.6584703535923491, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0727445861238734, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12082775312405933, "1000–3000 Hz_10ms_envelope_correlation": 0.9867242326826188, "1000–3000 Hz_10ms_modulation_mae_db": 0.49871805672544234, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5867163662963256, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.052830354663540226, "1000–3000 Hz_40ms_envelope_correlation": 0.997625407054255, "1000–3000 Hz_40ms_modulation_mae_db": 0.3907238552257681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.140448609888099, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15713274360722024, "3000–6000 Hz_10ms_envelope_correlation": 0.974098196530819, "3000–6000 Hz_10ms_modulation_mae_db": 0.7137834118347354, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6200374652668674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07764782438372454, "3000–6000 Hz_40ms_envelope_correlation": 0.9933028874532518, "3000–6000 Hz_40ms_modulation_mae_db": 0.765489339845378, "mr_stft_db": 5.664862702366331, "band_mae_db": 0.1527597557027589, "rms_gain_error_db": -0.10312840834014206, "output_peak": 0.4469306670345537}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.891076860476179, "全帯域_10ms_rms_envelope_nrmse": 0.10278359978302345, "全帯域_10ms_envelope_correlation": 0.9850549597654409, "全帯域_10ms_modulation_mae_db": 0.3234117612982678, "全帯域_40ms_log_rms_mae_db": 0.42344754767880105, "全帯域_40ms_rms_envelope_nrmse": 0.04488276446880528, "全帯域_40ms_envelope_correlation": 0.9970153973093971, "全帯域_40ms_modulation_mae_db": 0.15900461996649493, "300–1000 Hz_10ms_log_rms_mae_db": 1.1070784007872965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10880797852219039, "300–1000 Hz_10ms_envelope_correlation": 0.9878839574009793, "300–1000 Hz_10ms_modulation_mae_db": 0.7112072349578139, "300–1000 Hz_40ms_log_rms_mae_db": 0.624578019624931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058445767464290095, "300–1000 Hz_40ms_envelope_correlation": 0.9962415327134532, "300–1000 Hz_40ms_modulation_mae_db": 0.6019648439502384, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0180943950021262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14590146075574179, "1000–3000 Hz_10ms_envelope_correlation": 0.9804736876879282, "1000–3000 Hz_10ms_modulation_mae_db": 0.5903326418815252, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6077242528307051, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07193585393151357, "1000–3000 Hz_40ms_envelope_correlation": 0.9951427547310996, "1000–3000 Hz_40ms_modulation_mae_db": 0.7293090539040709, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0860116161148654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14023707179388958, "3000–6000 Hz_10ms_envelope_correlation": 0.9794306779680857, "3000–6000 Hz_10ms_modulation_mae_db": 0.46762965044055793, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6121514105384507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06271626731210293, "3000–6000 Hz_40ms_envelope_correlation": 0.9960770295701087, "3000–6000 Hz_40ms_modulation_mae_db": 0.2638930363828476, "mr_stft_db": 5.662777215106786, "band_mae_db": 0.12217595494873623, "rms_gain_error_db": -0.10659602131818276, "output_peak": 0.44508104899251816}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8751469727899496, "全帯域_10ms_rms_envelope_nrmse": 0.09169824114870774, "全帯域_10ms_envelope_correlation": 0.9881343425833344, "全帯域_10ms_modulation_mae_db": 0.3525077678719488, "全帯域_40ms_log_rms_mae_db": 0.47554658015354295, "全帯域_40ms_rms_envelope_nrmse": 0.04578524099868971, "全帯域_40ms_envelope_correlation": 0.996759464371914, "全帯域_40ms_modulation_mae_db": 0.35520396487798317, "300–1000 Hz_10ms_log_rms_mae_db": 1.1191548322851252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1082916178318093, "300–1000 Hz_10ms_envelope_correlation": 0.9880430252562948, "300–1000 Hz_10ms_modulation_mae_db": 0.6128392751714672, "300–1000 Hz_40ms_log_rms_mae_db": 0.6650880824273674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06521009468246165, "300–1000 Hz_40ms_envelope_correlation": 0.9953079942731702, "300–1000 Hz_40ms_modulation_mae_db": 0.7938818029929592, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9758396047983217, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11659609464713075, "1000–3000 Hz_10ms_envelope_correlation": 0.987518495716945, "1000–3000 Hz_10ms_modulation_mae_db": 0.6678371322528692, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5373840164135345, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.054137115537612905, "1000–3000 Hz_40ms_envelope_correlation": 0.9971716201946733, "1000–3000 Hz_40ms_modulation_mae_db": 0.7513965113065506, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1162280089471825, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14864720939636283, "3000–6000 Hz_10ms_envelope_correlation": 0.976825214491713, "3000–6000 Hz_10ms_modulation_mae_db": 0.7790801247769966, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6326669370714724, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07578198438306857, "3000–6000 Hz_40ms_envelope_correlation": 0.9935884749407791, "3000–6000 Hz_40ms_modulation_mae_db": 0.7402884249839948, "mr_stft_db": 5.669545771066538, "band_mae_db": 0.06852397472094852, "rms_gain_error_db": -0.045485008909907636, "output_peak": 0.4683883443107289}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8979734016052343, "全帯域_10ms_rms_envelope_nrmse": 0.09260633599749567, "全帯域_10ms_envelope_correlation": 0.9879529917701598, "全帯域_10ms_modulation_mae_db": 0.2519077171199999, "全帯域_40ms_log_rms_mae_db": 0.4135930032713777, "全帯域_40ms_rms_envelope_nrmse": 0.043412108424923115, "全帯域_40ms_envelope_correlation": 0.9973863652263912, "全帯域_40ms_modulation_mae_db": 0.288053308372079, "300–1000 Hz_10ms_log_rms_mae_db": 1.1212935484201003, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10048143608006922, "300–1000 Hz_10ms_envelope_correlation": 0.9898938768103368, "300–1000 Hz_10ms_modulation_mae_db": 0.8949903171952251, "300–1000 Hz_40ms_log_rms_mae_db": 0.6689308664360633, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053598118170402345, "300–1000 Hz_40ms_envelope_correlation": 0.9969991691423369, "300–1000 Hz_40ms_modulation_mae_db": 0.8157293236100122, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0121718617333362, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12744738526572538, "1000–3000 Hz_10ms_envelope_correlation": 0.985909443304478, "1000–3000 Hz_10ms_modulation_mae_db": 0.6093611443912701, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5720562372888119, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07238633463781766, "1000–3000 Hz_40ms_envelope_correlation": 0.9963576969775657, "1000–3000 Hz_40ms_modulation_mae_db": 0.9360191556681056, "3000–6000 Hz_10ms_log_rms_mae_db": 1.038338374757782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1352238139429445, "3000–6000 Hz_10ms_envelope_correlation": 0.9809894287639519, "3000–6000 Hz_10ms_modulation_mae_db": 0.4554501415176094, "3000–6000 Hz_40ms_log_rms_mae_db": 0.5619715016115826, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06381298775078868, "3000–6000 Hz_40ms_envelope_correlation": 0.9961668834474885, "3000–6000 Hz_40ms_modulation_mae_db": 0.6971081280730228, "mr_stft_db": 5.677159188501441, "band_mae_db": 0.1913476922641859, "rms_gain_error_db": -0.13447065751115994, "output_peak": 0.43854635287232424}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49200, "output_samples": 49200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.385247016930068, "spectral_convergence": 0.4468599129880323, "spectral_convergence_status": "ok", "amplitude_floor": 6.39186387874099e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.274187858084441, "spectral_convergence": 0.41126182510886444, "spectral_convergence_status": "ok", "amplitude_floor": 4.1641566181737356e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.350762188577381, "spectral_convergence": 0.4552690505007266, "spectral_convergence_status": "ok", "amplitude_floor": 3.471921787916526e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.471560290376807e-05, "output_power": 7.777224577291142e-05, "delta_db": 0.17413334682578416, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002820835097806379, "output_power": 0.00028577629986985785, "delta_db": 0.05648510209678681, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026780543872196867, "output_power": 0.00026557343791712174, "delta_db": -0.03634756957254616, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028630139823372175, "output_power": 0.00028928728071624815, "delta_db": 0.04505870130691602, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006167999147673381, "output_power": 0.0005987238766631326, "delta_db": -0.12917727197669596, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00037968124815517825, "output_power": 0.000349837026192937, "delta_db": -0.3555337595442478, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13278929188716282, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09301465672797257, "pitch": null, "output_rms_dbfs": -27.288626217152906, "output_peak": 0.4538809902305671, "reference_rms_dbfs": -27.195611560424933, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.670065687863963}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 0.00010992699921133752, "log_rms_mae_db": 0.844908456643161, "rms_envelope_nrmse": 0.09829381966281515, "envelope_correlation": 0.9862625002395825, "rms_gain_error_db": -0.09303871581766393, "reference_mean_power": 0.0019104810338505282, "output_mean_power": 0.0018699881697728806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 4.701235881229352e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003075331225832077, "output_power": 0.00029456162862600936, "reference_active": true, "delta_db": -0.18715725560910637}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8607043791082874e-05, "output_power": 2.6537294403061754e-05, "reference_active": true, "delta_db": -0.3261633828197884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1458430393639735e-05, "output_power": 1.158529922638399e-05, "reference_active": true, "delta_db": 0.047821243875407315}], "active_band_count": 3, "active_band_mae_db": 0.18704729410143403}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 9.464213695185308e-05, "log_rms_mae_db": 0.4392696015455033, "rms_envelope_nrmse": 0.044146595534013275, "envelope_correlation": 0.9970954008975432, "rms_gain_error_db": -0.09399764633625739, "reference_mean_power": 0.0019290839559002245, "output_mean_power": 0.001887779929322862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 4.127760176364045e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002739478020838986, "output_power": 0.00026129887597565814, "reference_active": true, "delta_db": -0.20530278852808398}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7486141765201905e-05, "output_power": 1.5975082095744423e-05, "reference_active": true, "delta_db": -0.39250896347439884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8391125647823e-07, "output_power": 5.459461459614689e-07, "reference_active": true, "delta_db": -0.2919704304986887}], "active_band_count": 3, "active_band_mae_db": 0.29659406083372386}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 6.478025356347348e-05, "log_rms_mae_db": 1.1469331993454783, "rms_envelope_nrmse": 0.10457875552833822, "envelope_correlation": 0.9888319420024425, "rms_gain_error_db": -0.05893795608102115, "reference_mean_power": 0.0004974887922363912, "output_mean_power": 0.0004907829937957176}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.1951630543463553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001563474072286115, "output_power": 0.00015501880431340222, "reference_active": true, "delta_db": -0.03706300782043108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.160597194093242e-06, "output_power": 8.636640465077228e-06, "reference_active": true, "delta_db": 0.24622899046504135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0536955073753773e-06, "output_power": 3.1601989736438067e-06, "reference_active": true, "delta_db": 1.871783746386809}], "active_band_count": 3, "active_band_mae_db": 0.718358581557427}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 5.647638237474489e-05, "log_rms_mae_db": 0.6548935732297853, "rms_envelope_nrmse": 0.052410993092919356, "envelope_correlation": 0.9969992723241861, "rms_gain_error_db": -0.05893711521104193, "reference_mean_power": 0.00050239739349007, "output_mean_power": 0.0004956255265243351}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.0265825460069485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014180308081562638, "output_power": 0.0001404419391612894, "reference_active": true, "delta_db": -0.04188849052991272}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4691763208394e-06, "output_power": 4.64792641233356e-06, "reference_active": true, "delta_db": 0.17031754462442578}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.310040798443831e-08, "output_power": 1.65157878895633e-07, "reference_active": true, "delta_db": 2.489477124924957}], "active_band_count": 3, "active_band_mae_db": 0.9005610533597652}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 7.855020595588598e-05, "log_rms_mae_db": 1.0248459659717704, "rms_envelope_nrmse": 0.13663332101443235, "envelope_correlation": 0.9827887857020854, "rms_gain_error_db": -0.023494595246343692, "reference_mean_power": 0.0004940955013379756, "output_mean_power": 0.0004914297455805485}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 1.3474634060369984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.364408948380063e-05, "output_power": 9.549867455361498e-05, "reference_active": true, "delta_db": 0.08516972666233533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.996833569277459e-06, "output_power": 6.752361857559809e-06, "reference_active": true, "delta_db": -0.1544583567046997}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2622375016010212e-06, "output_power": 3.3466569869072736e-06, "reference_active": true, "delta_db": 1.7007300444847917}], "active_band_count": 3, "active_band_mae_db": 0.6467860426172756}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 6.30584220600245e-05, "log_rms_mae_db": 0.6069580202010233, "rms_envelope_nrmse": 0.06528397968687587, "envelope_correlation": 0.995844555345558, "rms_gain_error_db": -0.023494857835449995, "reference_mean_power": 0.0004989699589215838, "output_mean_power": 0.0004962778743684545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 1.243238840477423e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.827683631948728e-05, "output_power": 9.045762577455925e-05, "reference_active": true, "delta_db": 0.10598424260553671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8441959267659476e-06, "output_power": 3.681279230837838e-06, "reference_active": true, "delta_db": -0.18806753588152034}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.483409815447214e-08, "output_power": 1.5153877868404807e-07, "reference_active": true, "delta_db": 2.0355926422428534}], "active_band_count": 3, "active_band_mae_db": 0.7765481402433035}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 9.512970769085293e-05, "log_rms_mae_db": 1.1057133154858523, "rms_envelope_nrmse": 0.15111889712484539, "envelope_correlation": 0.976239951836504, "rms_gain_error_db": -0.30396844228052006, "reference_mean_power": 0.0007033343720650757, "output_mean_power": 0.0006557903138180267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.9605352394438525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001169604983627503, "output_power": 0.0001025109414908041, "reference_active": true, "delta_db": -0.5726898789877831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1203242070906313e-05, "output_power": 1.744086678972288e-05, "reference_active": true, "delta_db": -0.8483420654534328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.096712651632301e-05, "output_power": 1.0359894427001368e-05, "reference_active": true, "delta_db": -0.2473752376904444}], "active_band_count": 3, "active_band_mae_db": 0.5561357273772202}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 7.035164526339387e-05, "log_rms_mae_db": 0.6285357514121162, "rms_envelope_nrmse": 0.07839613655592562, "envelope_correlation": 0.9940188915506583, "rms_gain_error_db": -0.3039694131894482, "reference_mean_power": 0.0007102740547296548, "output_mean_power": 0.00066226073915332}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.6373521771123283e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010173798550296795, "output_power": 8.835670143351603e-05, "reference_active": true, "delta_db": -0.6124363973444038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.290709427246265e-05, "output_power": 1.04266627403322e-05, "reference_active": true, "delta_db": -0.9268315599402908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.316189457175965e-07, "output_power": 5.109614282329222e-07, "reference_active": true, "delta_db": -0.920670313198601}], "active_band_count": 3, "active_band_mae_db": 0.8199794234944319}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8451281269407177, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "53521dc50d23afe2c78fe74354df46962314b81538c797705403a81bb036536d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9851137638234738, "standard_deviation": 0.0024152571597983014, "minimum": 0.9804736876879282, "maximum": 0.987518495716945, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.0197115095935367, "standard_deviation": 0.0285745056292344, "minimum": 0.9758396047983217, "maximum": 1.0727445861238734, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5256613670478394, "standard_deviation": 0.1229248701947973, "minimum": 0.3222492506897929, "maximum": 0.6678371322528692, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1280725233098638, "standard_deviation": 0.009654711603265732, "minimum": 0.11659609464713075, "maximum": 0.14590146075574179, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963164550035468, "standard_deviation": 0.0008706363206234892, "minimum": 0.9951427547310996, "maximum": 0.997625407054255, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.5795236140529991, "standard_deviation": 0.03182385559687722, "minimum": 0.5373840164135345, "maximum": 0.6240500142592751, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6703944996622546, "standard_deviation": 0.21654225754248954, "minimum": 0.3907238552257681, "maximum": 0.9360191556681056, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06529561574791964, "standard_deviation": 0.008036773811250609, "minimum": 0.052830354663540226, "maximum": 0.07238633463781766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9785230243552855, "standard_deviation": 0.002904501055351599, "minimum": 0.974098196530819, "maximum": 0.9819651554673281, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.0794944609259023, "standard_deviation": 0.038380650935869155, "minimum": 1.038338374757782, "maximum": 1.140448609888099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5224718538372779, "standard_deviation": 0.21852873368841835, "minimum": 0.15011396400430707, "maximum": 0.7790801247769966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14294970651516675, "standard_deviation": 0.00960903918817662, "minimum": 0.13099578801015238, "maximum": 0.15713274360722024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9949466067496202, "standard_deviation": 0.001192010078827209, "minimum": 0.9933028874532518, "maximum": 0.9961668834474885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.5990806455965476, "standard_deviation": 0.02821589159733009, "minimum": 0.5619715016115826, "maximum": 0.6326669370714724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6110624863299357, "standard_deviation": 0.2772366948167703, "minimum": 0.2638930363828476, "maximum": 0.9865116582326444, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06871482184077815, "standard_deviation": 0.007530422273739308, "minimum": 0.058766373715657766, "maximum": 0.07839613655592562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9887263749044126, "standard_deviation": 0.0016426980567944862, "minimum": 0.985744585984882, "maximum": 0.9908660538506017, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1311460025724382, "standard_deviation": 0.027218477644849366, "minimum": 1.1070784007872965, "maximum": 1.185366606711899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8055576204891761, "standard_deviation": 0.16559793462284064, "minimum": 0.6128392751714672, "maximum": 1.0936467039993893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10512324833063169, "standard_deviation": 0.007453533563445388, "minimum": 0.09441497641336699, "maximum": 0.11838215466615376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966431811353167, "standard_deviation": 0.0007228713691776809, "minimum": 0.9953079942731702, "maximum": 0.9974978513574125, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6494350525667493, "standard_deviation": 0.031829674337543225, "minimum": 0.5855264561497581, "maximum": 0.681753451470327, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8441495961139951, "standard_deviation": 0.3593584777480253, "minimum": 0.39134776628402457, "maximum": 1.6051693082290797, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.055314075637035325, "standard_deviation": 0.005597435803324758, "minimum": 0.047837033501864995, "maximum": 0.06521009468246165, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1446212300043601, "standard_deviation": 0.039447878099905735, "minimum": 0.06852397472094852, "maximum": 0.1913476922641859, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6671843888835305, "standard_deviation": 0.006895076971120399, "minimum": 5.6550222528699505, "maximum": 5.677159188501441, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4538514884439028, "standard_deviation": 0.01856717308129587, "minimum": 0.42863637717438663, "maximum": 0.48770924116429093, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08339749503799158, "standard_deviation": 0.03369939373869821, "minimum": -0.13447065751115994, "maximum": -0.03187499180258923, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9869455628644661, "standard_deviation": 0.0017101784910734407, "minimum": 0.9845427929126136, "maximum": 0.9889305725605686, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8750118816088495, "standard_deviation": 0.02249537520217201, "minimum": 0.844908456643161, "maximum": 0.9089191085107494, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3822144741185179, "standard_deviation": 0.15203976925762802, "minimum": 0.18704729410143403, "maximum": 0.6737958842842833, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09581336962774516, "standard_deviation": 0.006227915427804866, "minimum": 0.08837538557425016, "maximum": 0.10442486828204539, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970290522769614, "standard_deviation": 0.0005249161585477915, "minimum": 0.9960317027807226, "maximum": 0.997635537577991, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4353184748800153, "standard_deviation": 0.020770006103639774, "minimum": 0.4135930032713777, "maximum": 0.47554658015354295, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4088954088368864, "standard_deviation": 0.2633078212240364, "minimum": 0.15900461996649493, "maximum": 1.0166297339422463, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04433175504857839, "standard_deviation": 0.00361349882447466, "minimum": 0.03971163201935698, "maximum": 0.051195175496988723, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-81414-0026", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.88059663772583, "predicted_mos": 2.88059663772583, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "2b214903b054e4f235956f1b851c34818a0331b35d91b2a4b362b5d83fa16231", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8303203000687063, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8661046981919739, "全帯域_10ms_rms_envelope_nrmse": 0.09680617405093482, "全帯域_10ms_envelope_correlation": 0.9867150499664622, "全帯域_10ms_modulation_mae_db": 0.5093592514888998, "全帯域_40ms_log_rms_mae_db": 0.43180485204566604, "全帯域_40ms_rms_envelope_nrmse": 0.054946630623589766, "全帯域_40ms_envelope_correlation": 0.995250013375243, "全帯域_40ms_modulation_mae_db": 0.5369927962090003, "300–1000 Hz_10ms_log_rms_mae_db": 1.2365096595702791, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08463097164756159, "300–1000 Hz_10ms_envelope_correlation": 0.9927863591426539, "300–1000 Hz_10ms_modulation_mae_db": 0.6304347620619174, "300–1000 Hz_40ms_log_rms_mae_db": 0.7241075457883703, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04209333830354627, "300–1000 Hz_40ms_envelope_correlation": 0.9981287336653669, "300–1000 Hz_40ms_modulation_mae_db": 0.541637221041602, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1158036995209897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11626996297565702, "1000–3000 Hz_10ms_envelope_correlation": 0.9875441258006941, "1000–3000 Hz_10ms_modulation_mae_db": 0.720155918870741, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6139876737511464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.049893310242519674, "1000–3000 Hz_40ms_envelope_correlation": 0.9976015140292217, "1000–3000 Hz_40ms_modulation_mae_db": 0.7738039236920616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1873461805386276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16414790052903203, "3000–6000 Hz_10ms_envelope_correlation": 0.9716180284310127, "3000–6000 Hz_10ms_modulation_mae_db": 0.5208897219304619, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6874561719662844, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10284941839611689, "3000–6000 Hz_40ms_envelope_correlation": 0.9879222691120257, "3000–6000 Hz_40ms_modulation_mae_db": 0.4274426421973925, "mr_stft_db": 5.759885189893162, "band_mae_db": 0.15557586950446042, "rms_gain_error_db": -0.045935031494998245, "output_peak": 0.4595306165309517}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9101150854763626, "全帯域_10ms_rms_envelope_nrmse": 0.09636429164015123, "全帯域_10ms_envelope_correlation": 0.9868994937079343, "全帯域_10ms_modulation_mae_db": 0.23814022230250578, "全帯域_40ms_log_rms_mae_db": 0.4220849372683773, "全帯域_40ms_rms_envelope_nrmse": 0.04301002112802469, "全帯域_40ms_envelope_correlation": 0.9973261813775273, "全帯域_40ms_modulation_mae_db": 0.2167239785087254, "300–1000 Hz_10ms_log_rms_mae_db": 1.313070542151657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1059527040062108, "300–1000 Hz_10ms_envelope_correlation": 0.9886261707138819, "300–1000 Hz_10ms_modulation_mae_db": 0.6601219226649131, "300–1000 Hz_40ms_log_rms_mae_db": 0.8112298487430618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049318344923082263, "300–1000 Hz_40ms_envelope_correlation": 0.9974306750933958, "300–1000 Hz_40ms_modulation_mae_db": 0.5339591196704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.115550505062111, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12852278490364152, "1000–3000 Hz_10ms_envelope_correlation": 0.9848601470783827, "1000–3000 Hz_10ms_modulation_mae_db": 0.5882823967992645, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6504835677630293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06268055116322996, "1000–3000 Hz_40ms_envelope_correlation": 0.9964969769970549, "1000–3000 Hz_40ms_modulation_mae_db": 0.8824337614164225, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1563678290609214, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15113567039875142, "3000–6000 Hz_10ms_envelope_correlation": 0.9761799908449316, "3000–6000 Hz_10ms_modulation_mae_db": 0.4433871524973247, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6771595640448257, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07969177226407709, "3000–6000 Hz_40ms_envelope_correlation": 0.993582004926694, "3000–6000 Hz_40ms_modulation_mae_db": 0.4901155857487322, "mr_stft_db": 5.754887704754093, "band_mae_db": 0.15886825883737343, "rms_gain_error_db": -0.11665723427634109, "output_peak": 0.42611892636217574}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8080049409391498, "全帯域_10ms_rms_envelope_nrmse": 0.10209862232893667, "全帯域_10ms_envelope_correlation": 0.9852236806979574, "全帯域_10ms_modulation_mae_db": 0.2387568033427823, "全帯域_40ms_log_rms_mae_db": 0.43815540432944194, "全帯域_40ms_rms_envelope_nrmse": 0.04861030220207569, "全帯域_40ms_envelope_correlation": 0.9963150542578928, "全帯域_40ms_modulation_mae_db": 0.19977946540315808, "300–1000 Hz_10ms_log_rms_mae_db": 1.1313436639300547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10761725844026095, "300–1000 Hz_10ms_envelope_correlation": 0.9881620986850002, "300–1000 Hz_10ms_modulation_mae_db": 0.16259686846954063, "300–1000 Hz_40ms_log_rms_mae_db": 0.6632218958544417, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062022449148549194, "300–1000 Hz_40ms_envelope_correlation": 0.9957618458147878, "300–1000 Hz_40ms_modulation_mae_db": 0.4560920351657581, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1545117384095143, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14057536405402796, "1000–3000 Hz_10ms_envelope_correlation": 0.9820540452734468, "1000–3000 Hz_10ms_modulation_mae_db": 1.3431841537533622, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6881068060387738, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0759392333993082, "1000–3000 Hz_40ms_envelope_correlation": 0.9944623174600585, "1000–3000 Hz_40ms_modulation_mae_db": 1.5965150228435843, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2113871396348421, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1517609992077958, "3000–6000 Hz_10ms_envelope_correlation": 0.9758162411401377, "3000–6000 Hz_10ms_modulation_mae_db": 0.4437456533711563, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7086406626733617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07301527194968112, "3000–6000 Hz_40ms_envelope_correlation": 0.994427711778863, "3000–6000 Hz_40ms_modulation_mae_db": 0.5261867982944134, "mr_stft_db": 5.746859371251258, "band_mae_db": 0.09823475981393577, "rms_gain_error_db": -0.04593615723257217, "output_peak": 0.47569945248652645}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8906050850497622, "全帯域_10ms_rms_envelope_nrmse": 0.09684793622450621, "全帯域_10ms_envelope_correlation": 0.9867295445989114, "全帯域_10ms_modulation_mae_db": 0.5726130280484496, "全帯域_40ms_log_rms_mae_db": 0.4174337480415666, "全帯域_40ms_rms_envelope_nrmse": 0.04172661886005147, "全帯域_40ms_envelope_correlation": 0.9972587745039438, "全帯域_40ms_modulation_mae_db": 0.48637337224370825, "300–1000 Hz_10ms_log_rms_mae_db": 1.2154409296398325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09764675653543205, "300–1000 Hz_10ms_envelope_correlation": 0.9908045068556635, "300–1000 Hz_10ms_modulation_mae_db": 0.8901688156448905, "300–1000 Hz_40ms_log_rms_mae_db": 0.7258183882290149, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05155490508694051, "300–1000 Hz_40ms_envelope_correlation": 0.9975179251054762, "300–1000 Hz_40ms_modulation_mae_db": 0.994650640452968, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1666553951981213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12993574877166608, "1000–3000 Hz_10ms_envelope_correlation": 0.9845611612366342, "1000–3000 Hz_10ms_modulation_mae_db": 0.5653433792919619, "1000–3000 Hz_40ms_log_rms_mae_db": 0.636193868598067, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05769423212291598, "1000–3000 Hz_40ms_envelope_correlation": 0.9967809343610927, "1000–3000 Hz_40ms_modulation_mae_db": 0.7882753989955326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2180181429176895, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15490119820383164, "3000–6000 Hz_10ms_envelope_correlation": 0.9744983532498411, "3000–6000 Hz_10ms_modulation_mae_db": 0.674396639088625, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6873005284685134, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07229663807023379, "3000–6000 Hz_40ms_envelope_correlation": 0.9942535243416634, "3000–6000 Hz_40ms_modulation_mae_db": 0.8328540498680157, "mr_stft_db": 5.74096793404937, "band_mae_db": 0.16972303835802763, "rms_gain_error_db": -0.005966267640572339, "output_peak": 0.42953873667557313}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8694581855940716, "全帯域_10ms_rms_envelope_nrmse": 0.0995937241682573, "全帯域_10ms_envelope_correlation": 0.9863724480656934, "全帯域_10ms_modulation_mae_db": 0.6616986217375022, "全帯域_40ms_log_rms_mae_db": 0.45245022649958244, "全帯域_40ms_rms_envelope_nrmse": 0.05165614939337208, "全帯域_40ms_envelope_correlation": 0.9965759903808582, "全帯域_40ms_modulation_mae_db": 1.0640642642351856, "300–1000 Hz_10ms_log_rms_mae_db": 1.2472621139067597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0987567692725966, "300–1000 Hz_10ms_envelope_correlation": 0.9900090958342058, "300–1000 Hz_10ms_modulation_mae_db": 0.9749608419741849, "300–1000 Hz_40ms_log_rms_mae_db": 0.7666815349717918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047147931879592445, "300–1000 Hz_40ms_envelope_correlation": 0.9975785457180867, "300–1000 Hz_40ms_modulation_mae_db": 1.09978991554535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.202189857238946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13766238813360565, "1000–3000 Hz_10ms_envelope_correlation": 0.9828527340154872, "1000–3000 Hz_10ms_modulation_mae_db": 0.7179533181552182, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6708395468118187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06399805321666178, "1000–3000 Hz_40ms_envelope_correlation": 0.9968074805422585, "1000–3000 Hz_40ms_modulation_mae_db": 0.639952002505671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.178349041228256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1544221816929193, "3000–6000 Hz_10ms_envelope_correlation": 0.9752526961633455, "3000–6000 Hz_10ms_modulation_mae_db": 1.0497560850561458, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6744942288018553, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08291183199945741, "3000–6000 Hz_40ms_envelope_correlation": 0.993500027160512, "3000–6000 Hz_40ms_modulation_mae_db": 1.437781442683623, "mr_stft_db": 5.715881420757468, "band_mae_db": 0.21050381015617733, "rms_gain_error_db": -0.20636981666415616, "output_peak": 0.4598959836013772}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.819097422404468, "全帯域_10ms_rms_envelope_nrmse": 0.08917320690269061, "全帯域_10ms_envelope_correlation": 0.9890780800634827, "全帯域_10ms_modulation_mae_db": 0.4222081937951437, "全帯域_40ms_log_rms_mae_db": 0.433709308753442, "全帯域_40ms_rms_envelope_nrmse": 0.0483633960605767, "全帯域_40ms_envelope_correlation": 0.9969343891994487, "全帯域_40ms_modulation_mae_db": 0.5509766393214001, "300–1000 Hz_10ms_log_rms_mae_db": 1.3144713044384764, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09942832553005383, "300–1000 Hz_10ms_envelope_correlation": 0.9898582397067472, "300–1000 Hz_10ms_modulation_mae_db": 0.2757403261892747, "300–1000 Hz_40ms_log_rms_mae_db": 0.8694917330522339, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06730098757762759, "300–1000 Hz_40ms_envelope_correlation": 0.9949943611779397, "300–1000 Hz_40ms_modulation_mae_db": 0.4545471174512626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1618257126772027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12010863087047093, "1000–3000 Hz_10ms_envelope_correlation": 0.9868381440260233, "1000–3000 Hz_10ms_modulation_mae_db": 0.5717560749952346, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7044356162826299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06481255803205402, "1000–3000 Hz_40ms_envelope_correlation": 0.9962682951841453, "1000–3000 Hz_40ms_modulation_mae_db": 0.6815107932432086, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0874735128212183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14053694371491313, "3000–6000 Hz_10ms_envelope_correlation": 0.9799461730868899, "3000–6000 Hz_10ms_modulation_mae_db": 0.6970413696993004, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6404892675969872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08084579273486178, "3000–6000 Hz_40ms_envelope_correlation": 0.9941511438051782, "3000–6000 Hz_40ms_modulation_mae_db": 0.9871306750646672, "mr_stft_db": 5.737259483458332, "band_mae_db": 0.16491060479544642, "rms_gain_error_db": -0.18230081490653613, "output_peak": 0.45160786988020823}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7713335817958703, "全帯域_10ms_rms_envelope_nrmse": 0.09313138053027548, "全帯域_10ms_envelope_correlation": 0.9878014445965948, "全帯域_10ms_modulation_mae_db": 0.46142863527974676, "全帯域_40ms_log_rms_mae_db": 0.42349107553772064, "全帯域_40ms_rms_envelope_nrmse": 0.05178734257570954, "全帯域_40ms_envelope_correlation": 0.9958336297448233, "全帯域_40ms_modulation_mae_db": 0.41156380448555024, "300–1000 Hz_10ms_log_rms_mae_db": 1.3000003495051888, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09101885816853833, "300–1000 Hz_10ms_envelope_correlation": 0.9915127418521728, "300–1000 Hz_10ms_modulation_mae_db": 0.522263532490321, "300–1000 Hz_40ms_log_rms_mae_db": 0.8541456686757889, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047778658396572846, "300–1000 Hz_40ms_envelope_correlation": 0.9975197455069946, "300–1000 Hz_40ms_modulation_mae_db": 0.6432947493160815, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1348442011447464, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14524939868780137, "1000–3000 Hz_10ms_envelope_correlation": 0.9818406378964698, "1000–3000 Hz_10ms_modulation_mae_db": 0.4793254151637297, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6819947249743321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09334933632290293, "1000–3000 Hz_40ms_envelope_correlation": 0.9924099318013432, "1000–3000 Hz_40ms_modulation_mae_db": 0.4185423232520855, "3000–6000 Hz_10ms_log_rms_mae_db": 1.160571126549819, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13961187098892638, "3000–6000 Hz_10ms_envelope_correlation": 0.9793996227007162, "3000–6000 Hz_10ms_modulation_mae_db": 0.5569972735397387, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7060247379264087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07513771445929941, "3000–6000 Hz_40ms_envelope_correlation": 0.9937846645145115, "3000–6000 Hz_40ms_modulation_mae_db": 0.555985071397007, "mr_stft_db": 5.739600588566238, "band_mae_db": 0.11351694083967541, "rms_gain_error_db": -0.025900569349496288, "output_peak": 0.42743217414184004}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8495756807644229, "全帯域_10ms_rms_envelope_nrmse": 0.10428984389888155, "全帯域_10ms_envelope_correlation": 0.9846890048940352, "全帯域_10ms_modulation_mae_db": 0.24888059651563207, "全帯域_40ms_log_rms_mae_db": 0.4494044055042211, "全帯域_40ms_rms_envelope_nrmse": 0.05213081868774356, "全帯域_40ms_envelope_correlation": 0.9957402052248928, "全帯域_40ms_modulation_mae_db": 0.21611428166739352, "300–1000 Hz_10ms_log_rms_mae_db": 1.2088585294962892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09090982687931914, "300–1000 Hz_10ms_envelope_correlation": 0.9918087412954535, "300–1000 Hz_10ms_modulation_mae_db": 0.5163667192805282, "300–1000 Hz_40ms_log_rms_mae_db": 0.7578624964171303, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05173983925290649, "300–1000 Hz_40ms_envelope_correlation": 0.9972588578933478, "300–1000 Hz_40ms_modulation_mae_db": 0.6591644607827772, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1358667564985858, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1301117101091312, "1000–3000 Hz_10ms_envelope_correlation": 0.9846376028157755, "1000–3000 Hz_10ms_modulation_mae_db": 0.6376019157172318, "1000–3000 Hz_40ms_log_rms_mae_db": 0.619782020411562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06096506323523548, "1000–3000 Hz_40ms_envelope_correlation": 0.9964498940196446, "1000–3000 Hz_40ms_modulation_mae_db": 0.5873555404024243, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2265128061672586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.165283879510206, "3000–6000 Hz_10ms_envelope_correlation": 0.9711147201575144, "3000–6000 Hz_10ms_modulation_mae_db": 0.2448113371944521, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7493480794412428, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08672697108811776, "3000–6000 Hz_40ms_envelope_correlation": 0.9913166906436326, "3000–6000 Hz_40ms_modulation_mae_db": 0.28153849906185996, "mr_stft_db": 5.742485432768782, "band_mae_db": 0.1481738839246538, "rms_gain_error_db": 0.021397380910621315, "output_peak": 0.4278927792841748}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49200, "output_samples": 49200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5186757399436335, "spectral_convergence": 0.43657805448646425, "spectral_convergence_status": "ok", "amplitude_floor": 6.39186387874099e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.368127539170457, "spectral_convergence": 0.407143836287345, "spectral_convergence_status": "ok", "amplitude_floor": 4.1641566181737356e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.392852290565396, "spectral_convergence": 0.44908212602384406, "spectral_convergence_status": "ok", "amplitude_floor": 3.471921787916526e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.471560290376807e-05, "output_power": 7.243855958164895e-05, "delta_db": -0.1344149934493162, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002820835097806379, "output_power": 0.00028877448286268717, "delta_db": 0.10181116161477519, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026780543872196867, "output_power": 0.00027357327729836905, "delta_db": 0.09254280506119934, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028630139823372175, "output_power": 0.00028041910933917504, "delta_db": -0.09015863448806766, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006167999147673381, "output_power": 0.0005741989682450433, "delta_db": -0.31081897234847183, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00037968124815517825, "output_power": 0.00039791472763850823, "delta_db": 0.20370865006493227, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15557586950446042, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.045935031494998245, "pitch": null, "output_rms_dbfs": -27.241546591919928, "output_peak": 0.4595306165309517, "reference_rms_dbfs": -27.195611560424933, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.759885189893162}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 0.00010992699921133752, "log_rms_mae_db": 0.8661046981919739, "rms_envelope_nrmse": 0.09680617405093482, "envelope_correlation": 0.9867150499664622, "rms_gain_error_db": -0.046024116741302024, "reference_mean_power": 0.0019104810338505282, "output_mean_power": 0.0018903417182215923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 4.701235881229352e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003075331225832077, "output_power": 0.00030026301759616643, "reference_active": true, "delta_db": -0.10390053057581625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8607043791082874e-05, "output_power": 2.7416741198810285e-05, "reference_active": true, "delta_db": -0.1845714825557709}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1458430393639735e-05, "output_power": 1.5243535447818748e-05, "reference_active": true, "delta_db": 1.2396057413351123}], "active_band_count": 3, "active_band_mae_db": 0.5093592514888998}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 9.464213695185308e-05, "log_rms_mae_db": 0.43180485204566604, "rms_envelope_nrmse": 0.054946630623589766, "envelope_correlation": 0.995250013375243, "rms_gain_error_db": -0.04600646813731693, "reference_mean_power": 0.0019290839559002245, "output_mean_power": 0.0019087562944410115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 4.127760176364045e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002739478020838986, "output_power": 0.00026762169680840796, "reference_active": true, "delta_db": -0.10146500446904223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7486141765201905e-05, "output_power": 1.6482468276387423e-05, "reference_active": true, "delta_db": -0.25671746482380586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8391125647823e-07, "output_power": 7.791596242346506e-07, "reference_active": true, "delta_db": 1.2527959193341527}], "active_band_count": 3, "active_band_mae_db": 0.5369927962090003}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 6.478025356347348e-05, "log_rms_mae_db": 1.2365096595702791, "rms_envelope_nrmse": 0.08463097164756159, "envelope_correlation": 0.9927863591426539, "rms_gain_error_db": 0.06454061408633373, "reference_mean_power": 0.0004974887922363912, "output_mean_power": 0.0005049371942734064}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.1951630543463553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001563474072286115, "output_power": 0.000157217621571023, "reference_active": true, "delta_db": 0.02410538309645402}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.160597194093242e-06, "output_power": 8.43077332033075e-06, "reference_active": true, "delta_db": 0.14145470835525167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0536955073753773e-06, "output_power": 3.0556982996590294e-06, "reference_active": true, "delta_db": 1.7257441947340466}], "active_band_count": 3, "active_band_mae_db": 0.6304347620619174}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 5.647638237474489e-05, "log_rms_mae_db": 0.7241075457883703, "rms_envelope_nrmse": 0.04209333830354627, "envelope_correlation": 0.9981287336653669, "rms_gain_error_db": 0.06453472993402015, "reference_mean_power": 0.00050239739349007, "output_mean_power": 0.0005099185962264657}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.0265825460069485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014180308081562638, "output_power": 0.00014235590101406323, "reference_active": true, "delta_db": 0.0168980802940298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4691763208394e-06, "output_power": 4.520793098329044e-06, "reference_active": true, "delta_db": 0.049871422088665435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.310040798443831e-08, "output_power": 1.3328025047084716e-07, "reference_active": true, "delta_db": 1.5581421607421106}], "active_band_count": 3, "active_band_mae_db": 0.541637221041602}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 7.855020595588598e-05, "log_rms_mae_db": 1.1158036995209897, "rms_envelope_nrmse": 0.11626996297565702, "envelope_correlation": 0.9875441258006941, "rms_gain_error_db": -0.05575698513305847, "reference_mean_power": 0.0004940955013379756, "output_mean_power": 0.00048779259280760495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 1.3474634060369984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.364408948380063e-05, "output_power": 9.069729682723606e-05, "reference_active": true, "delta_db": -0.13886027899939266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.996833569277459e-06, "output_power": 6.492474191578599e-06, "reference_active": true, "delta_db": -0.32491311727485084}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2622375016010212e-06, "output_power": 3.3435485483247395e-06, "reference_active": true, "delta_db": 1.6966943603379798}], "active_band_count": 3, "active_band_mae_db": 0.720155918870741}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 6.30584220600245e-05, "log_rms_mae_db": 0.6139876737511464, "rms_envelope_nrmse": 0.049893310242519674, "envelope_correlation": 0.9976015140292217, "rms_gain_error_db": -0.05575841392260875, "reference_mean_power": 0.0004989699589215838, "output_mean_power": 0.0004926047075150637}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 1.243238840477423e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.827683631948728e-05, "output_power": 8.547636320992328e-05, "reference_active": true, "delta_db": -0.1400072462849295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8441959267659476e-06, "output_power": 3.919262287833894e-06, "reference_active": true, "delta_db": 0.08398814428434337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.483409815447214e-08, "output_power": 1.5371142776999682e-07, "reference_active": true, "delta_db": 2.097416380506912}], "active_band_count": 3, "active_band_mae_db": 0.7738039236920616}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 9.512970769085293e-05, "log_rms_mae_db": 1.1873461805386276, "rms_envelope_nrmse": 0.16414790052903203, "envelope_correlation": 0.9716180284310127, "rms_gain_error_db": -0.14228898171142668, "reference_mean_power": 0.0007033343720650757, "output_mean_power": 0.0006806642542491354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.9605352394438525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001169604983627503, "output_power": 0.00010894211248957886, "reference_active": true, "delta_db": -0.3084341772902589}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1203242070906313e-05, "output_power": 1.930822555128404e-05, "reference_active": true, "delta_db": -0.4065990814438365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.096712651632301e-05, "output_power": 1.3330807102623319e-05, "reference_active": true, "delta_db": 0.8476359070572902}], "active_band_count": 3, "active_band_mae_db": 0.5208897219304619}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 7.035164526339387e-05, "log_rms_mae_db": 0.6874561719662844, "rms_envelope_nrmse": 0.10284941839611689, "envelope_correlation": 0.9879222691120257, "rms_gain_error_db": -0.14228964147892212, "reference_mean_power": 0.0007102740547296548, "output_mean_power": 0.0006873801502264587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.6373521771123283e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010173798550296795, "output_power": 9.483296187221633e-05, "reference_active": true, "delta_db": -0.3052381910581462}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.290709427246265e-05, "output_power": 1.1675843782389302e-05, "reference_active": true, "delta_db": -0.43540206434025014}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.316189457175965e-07, "output_power": 7.155235224465434e-07, "reference_active": true, "delta_db": 0.541687671193781}], "active_band_count": 3, "active_band_mae_db": 0.4274426421973925}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8505219157408774, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1d23cb8cabff969275299165c52c2962d868ade91e6aa3b062dfae0e4c96e32c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9843985747678642, "standard_deviation": 0.0020898424564951053, "minimum": 0.9818406378964698, "maximum": 0.9875441258006941, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1484059832187772, "standard_deviation": 0.029097523248487307, "minimum": 1.115550505062111, "maximum": 1.202189857238946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.702950321593343, "standard_deviation": 0.27097583686124777, "minimum": 0.4793254151637297, "maximum": 1.3431841537533622, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13105449856325022, "standard_deviation": 0.00987602299310055, "minimum": 0.11626996297565702, "maximum": 0.14524939868780137, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9959096680493524, "standard_deviation": 0.0016705808199883238, "minimum": 0.9924099318013432, "maximum": 0.9976015140292217, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.65822797807892, "standard_deviation": 0.033236327311809394, "minimum": 0.6139876737511464, "maximum": 0.7044356162826299, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7960485957938738, "standard_deviation": 0.35301011266880133, "minimum": 0.4185423232520855, "maximum": 1.5965150228435843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0661665422168535, "standard_deviation": 0.013184141125483282, "minimum": 0.049893310242519674, "maximum": 0.09334933632290293, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9754782282217986, "standard_deviation": 0.003182518190070043, "minimum": 0.9711147201575144, "maximum": 0.9799461730868899, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.178253222364829, "standard_deviation": 0.044917967286281814, "minimum": 1.0874735128212183, "maximum": 1.2265128061672586, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5788781540471506, "standard_deviation": 0.23810952641359193, "minimum": 0.2448113371944521, "maximum": 1.0497560850561458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15272508053079697, "standard_deviation": 0.0094095886496972, "minimum": 0.13961187098892638, "maximum": 0.165283879510206, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992867254535385, "standard_deviation": 0.0022237414433479187, "minimum": 0.9879222691120257, "maximum": 0.994427711778863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6913641551149349, "standard_deviation": 0.03158198011778393, "minimum": 0.6404892675969872, "maximum": 0.7493480794412428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6923793455394638, "standard_deviation": 0.37575665044119544, "minimum": 0.28153849906185996, "maximum": 1.437781442683623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08168442637023066, "standard_deviation": 0.009897394248741432, "minimum": 0.07229663807023379, "maximum": 0.10284941839611689, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904459942607223, "standard_deviation": 0.0015885816386725514, "minimum": 0.9881620986850002, "maximum": 0.9927863591426539, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2458696365798172, "standard_deviation": 0.06285329796116083, "minimum": 1.1313436639300547, "maximum": 1.3144713044384764, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5790817235969462, "standard_deviation": 0.27654429254366125, "minimum": 0.16259686846954063, "maximum": 0.9749608419741849, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09699518380999667, "standard_deviation": 0.007820625897944035, "minimum": 0.08463097164756159, "maximum": 0.10761725844026095, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970238362469244, "standard_deviation": 0.0010657453880311076, "minimum": 0.9949943611779397, "maximum": 0.9981287336653669, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7715698889664792, "standard_deviation": 0.06997684251604513, "minimum": 0.6632218958544417, "maximum": 0.8694917330522339, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6728919074282749, "standard_deviation": 0.2443772757460694, "minimum": 0.4545471174512626, "maximum": 1.09978991554535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0523695568211022, "standard_deviation": 0.008284567691343647, "minimum": 0.04209333830354627, "maximum": 0.06730098757762759, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15243839577871876, "standard_deviation": 0.034534028687530596, "minimum": 0.09823475981393577, "maximum": 0.21050381015617733, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7422283906873375, "standard_deviation": 0.013206490833300553, "minimum": 5.715881420757468, "maximum": 5.759885189893162, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4447145673703534, "standard_deviation": 0.019329211086430775, "minimum": 0.42611892636217574, "maximum": 0.47569945248652645, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07595856383175639, "standard_deviation": 0.08337335366658045, "minimum": -0.20636981666415616, "maximum": 0.021397380910621315, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.986688593323884, "standard_deviation": 0.0013752512228373012, "minimum": 0.9846890048940352, "maximum": 0.9890780800634827, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8480368350270102, "standard_deviation": 0.04594544323427855, "minimum": 0.7713335817958703, "maximum": 0.9101150854763626, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41913566906383276, "standard_deviation": 0.16327592449215605, "minimum": 0.23814022230250578, "maximum": 0.6616986217375022, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09728814746807923, "standard_deviation": 0.004809078778371206, "minimum": 0.08917320690269061, "maximum": 0.10428984389888155, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9964042797580787, "standard_deviation": 0.0007556920121073947, "minimum": 0.995250013375243, "maximum": 0.9973261813775273, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4335667447475023, "standard_deviation": 0.012678566639354783, "minimum": 0.4174337480415666, "maximum": 0.45245022649958244, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46032357525926515, "standard_deviation": 0.2849855259365253, "minimum": 0.19977946540315808, "maximum": 1.0640642642351856, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04902890994139293, "standard_deviation": 0.004618614703080647, "minimum": 0.04172661886005147, "maximum": 0.054946630623589766, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_6930-81414-0026", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.369460105895996, "predicted_mos": 2.369460105895996, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "8a49c2ee0c75c268ffa8d97a2fdb55e871391d33404892cee5b90984b99ee64d", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9759482997469604, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8997501076592342, "全帯域_10ms_rms_envelope_nrmse": 0.09398660389334425, "全帯域_10ms_envelope_correlation": 0.9874206267579076, "全帯域_10ms_modulation_mae_db": 0.6172659913155718, "全帯域_40ms_log_rms_mae_db": 0.46018549005459847, "全帯域_40ms_rms_envelope_nrmse": 0.044555603504592006, "全帯域_40ms_envelope_correlation": 0.9968878319545506, "全帯域_40ms_modulation_mae_db": 0.8641606842511435, "300–1000 Hz_10ms_log_rms_mae_db": 1.205655786700764, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10635502067640247, "300–1000 Hz_10ms_envelope_correlation": 0.9884001076478063, "300–1000 Hz_10ms_modulation_mae_db": 0.6313621088875536, "300–1000 Hz_40ms_log_rms_mae_db": 0.7154096924172432, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06002757118344445, "300–1000 Hz_40ms_envelope_correlation": 0.9960255554791104, "300–1000 Hz_40ms_modulation_mae_db": 0.672582443218828, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1524003352254972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12964916608575547, "1000–3000 Hz_10ms_envelope_correlation": 0.9846113174248636, "1000–3000 Hz_10ms_modulation_mae_db": 0.22864881746462892, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5819126330998655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06141617416464626, "1000–3000 Hz_40ms_envelope_correlation": 0.9967210306117961, "1000–3000 Hz_40ms_modulation_mae_db": 0.1916581352159169, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1708430375540586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15167090324433183, "3000–6000 Hz_10ms_envelope_correlation": 0.975903212805256, "3000–6000 Hz_10ms_modulation_mae_db": 0.6485449694645093, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6946471020561796, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08403263321991117, "3000–6000 Hz_40ms_envelope_correlation": 0.991912044301681, "3000–6000 Hz_40ms_modulation_mae_db": 0.8408286994142479, "mr_stft_db": 5.718347537717972, "band_mae_db": 0.08856893465264976, "rms_gain_error_db": -0.03558384515035499, "output_peak": 0.4233474578129446}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7940099984227799, "全帯域_10ms_rms_envelope_nrmse": 0.09253661824920223, "全帯域_10ms_envelope_correlation": 0.9878329893406272, "全帯域_10ms_modulation_mae_db": 0.4667469775713984, "全帯域_40ms_log_rms_mae_db": 0.44466594919377767, "全帯域_40ms_rms_envelope_nrmse": 0.04265586651801808, "全帯域_40ms_envelope_correlation": 0.99715149337631, "全帯域_40ms_modulation_mae_db": 0.42203619311473933, "300–1000 Hz_10ms_log_rms_mae_db": 1.2127581046467784, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11144435237869797, "300–1000 Hz_10ms_envelope_correlation": 0.9872521574217288, "300–1000 Hz_10ms_modulation_mae_db": 1.1733952160689889, "300–1000 Hz_40ms_log_rms_mae_db": 0.7090121636413008, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06092284798521089, "300–1000 Hz_40ms_envelope_correlation": 0.9959335458104563, "300–1000 Hz_40ms_modulation_mae_db": 1.2579308005844136, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0900983974105196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12014419196783754, "1000–3000 Hz_10ms_envelope_correlation": 0.9866761155139652, "1000–3000 Hz_10ms_modulation_mae_db": 0.6329776863935027, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6147892884465129, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05427887283374796, "1000–3000 Hz_40ms_envelope_correlation": 0.9972041886644077, "1000–3000 Hz_40ms_modulation_mae_db": 0.907888068724333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1219450554221828, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1428927749950993, "3000–6000 Hz_10ms_envelope_correlation": 0.9784950686190069, "3000–6000 Hz_10ms_modulation_mae_db": 0.6390554802535549, "3000–6000 Hz_40ms_log_rms_mae_db": 0.58313906135347, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06659036911735217, "3000–6000 Hz_40ms_envelope_correlation": 0.9949795980307088, "3000–6000 Hz_40ms_modulation_mae_db": 0.3875091612155001, "mr_stft_db": 5.705917598904001, "band_mae_db": 0.10390838754953517, "rms_gain_error_db": -0.04347684529938564, "output_peak": 0.4536479630113029}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8250184051461924, "全帯域_10ms_rms_envelope_nrmse": 0.07891990142468341, "全帯域_10ms_envelope_correlation": 0.9912651749576178, "全帯域_10ms_modulation_mae_db": 0.24228950707681787, "全帯域_40ms_log_rms_mae_db": 0.3966096740724869, "全帯域_40ms_rms_envelope_nrmse": 0.03617271926102287, "全帯域_40ms_envelope_correlation": 0.9981582633222296, "全帯域_40ms_modulation_mae_db": 0.43231198874021154, "300–1000 Hz_10ms_log_rms_mae_db": 1.1044011745744604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08375066022487017, "300–1000 Hz_10ms_envelope_correlation": 0.9928208244390899, "300–1000 Hz_10ms_modulation_mae_db": 0.6299599618895022, "300–1000 Hz_40ms_log_rms_mae_db": 0.6133927305703231, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04826957338445738, "300–1000 Hz_40ms_envelope_correlation": 0.9974814150081128, "300–1000 Hz_40ms_modulation_mae_db": 0.7072486516289666, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0262250388279128, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1110101710431771, "1000–3000 Hz_10ms_envelope_correlation": 0.9886326664479087, "1000–3000 Hz_10ms_modulation_mae_db": 0.30019048467225184, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5882635159214931, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.051716726928110114, "1000–3000 Hz_40ms_envelope_correlation": 0.9974573108749389, "1000–3000 Hz_40ms_modulation_mae_db": 0.38307643922962065, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1756991581115683, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12681284073560997, "3000–6000 Hz_10ms_envelope_correlation": 0.9834562829709654, "3000–6000 Hz_10ms_modulation_mae_db": 0.45045254283538433, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6614641128628608, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0631467649530795, "3000–6000 Hz_40ms_envelope_correlation": 0.996247257603276, "3000–6000 Hz_40ms_modulation_mae_db": 0.48140117553381906, "mr_stft_db": 5.714580679333518, "band_mae_db": 0.10396944679808545, "rms_gain_error_db": -0.10919667957526123, "output_peak": 0.4743658113432926}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7697848294708087, "全帯域_10ms_rms_envelope_nrmse": 0.08533282619593677, "全帯域_10ms_envelope_correlation": 0.989650842877811, "全帯域_10ms_modulation_mae_db": 0.3159389107865991, "全帯域_40ms_log_rms_mae_db": 0.4280570757284858, "全帯域_40ms_rms_envelope_nrmse": 0.04113479462988241, "全帯域_40ms_envelope_correlation": 0.9974101106184141, "全帯域_40ms_modulation_mae_db": 0.29854538745553927, "300–1000 Hz_10ms_log_rms_mae_db": 1.2252210578719909, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09192137523082407, "300–1000 Hz_10ms_envelope_correlation": 0.9914191415269564, "300–1000 Hz_10ms_modulation_mae_db": 0.7838379959292726, "300–1000 Hz_40ms_log_rms_mae_db": 0.6497155679374058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04815506321552777, "300–1000 Hz_40ms_envelope_correlation": 0.9974599929334901, "300–1000 Hz_40ms_modulation_mae_db": 0.9197452024110433, "1000–3000 Hz_10ms_log_rms_mae_db": 1.120026535963424, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1411344734288001, "1000–3000 Hz_10ms_envelope_correlation": 0.9819924500246789, "1000–3000 Hz_10ms_modulation_mae_db": 0.9201192517321667, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6238511714314452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07499186158577391, "1000–3000 Hz_40ms_envelope_correlation": 0.9954031767246068, "1000–3000 Hz_40ms_modulation_mae_db": 1.2082054302640042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1358645350985375, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1297095653514193, "3000–6000 Hz_10ms_envelope_correlation": 0.9822459502268727, "3000–6000 Hz_10ms_modulation_mae_db": 0.2393598159875857, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6608523295285512, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.059869525998938675, "3000–6000 Hz_40ms_envelope_correlation": 0.9959107479993055, "3000–6000 Hz_40ms_modulation_mae_db": 0.25719679464991696, "mr_stft_db": 5.705572454362574, "band_mae_db": 0.14956816170032533, "rms_gain_error_db": -0.06415219697561586, "output_peak": 0.4293203662747513}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8294162105371649, "全帯域_10ms_rms_envelope_nrmse": 0.08871264508431655, "全帯域_10ms_envelope_correlation": 0.9890815534734573, "全帯域_10ms_modulation_mae_db": 0.42677207723300253, "全帯域_40ms_log_rms_mae_db": 0.4523618605115588, "全帯域_40ms_rms_envelope_nrmse": 0.04891737011054326, "全帯域_40ms_envelope_correlation": 0.9967144008916964, "全帯域_40ms_modulation_mae_db": 0.21674326611894892, "300–1000 Hz_10ms_log_rms_mae_db": 1.254163314242096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11126008542979521, "300–1000 Hz_10ms_envelope_correlation": 0.9873202363168964, "300–1000 Hz_10ms_modulation_mae_db": 0.3621373754793183, "300–1000 Hz_40ms_log_rms_mae_db": 0.7950130997320541, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07741903672877969, "300–1000 Hz_40ms_envelope_correlation": 0.993398266957471, "300–1000 Hz_40ms_modulation_mae_db": 0.18247763679739495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0694267789291907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12222743041695294, "1000–3000 Hz_10ms_envelope_correlation": 0.9865503964819183, "1000–3000 Hz_10ms_modulation_mae_db": 0.5307246716930539, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6073259802278119, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05691906845748757, "1000–3000 Hz_40ms_envelope_correlation": 0.9975378045839671, "1000–3000 Hz_40ms_modulation_mae_db": 0.6547232819934201, "3000–6000 Hz_10ms_log_rms_mae_db": 1.0931749635022976, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13216099108945095, "3000–6000 Hz_10ms_envelope_correlation": 0.9819370259653813, "3000–6000 Hz_10ms_modulation_mae_db": 0.39447331132697167, "3000–6000 Hz_40ms_log_rms_mae_db": 0.606465317639359, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06933333743379286, "3000–6000 Hz_40ms_envelope_correlation": 0.9953369370087116, "3000–6000 Hz_40ms_modulation_mae_db": 0.17947349863598536, "mr_stft_db": 5.697316854803691, "band_mae_db": 0.16031801206450527, "rms_gain_error_db": -0.15972073883533217, "output_peak": 0.427472502799862}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7591649073389285, "全帯域_10ms_rms_envelope_nrmse": 0.09455429745805351, "全帯域_10ms_envelope_correlation": 0.9872975610663965, "全帯域_10ms_modulation_mae_db": 0.6201164249174386, "全帯域_40ms_log_rms_mae_db": 0.3910434649163834, "全帯域_40ms_rms_envelope_nrmse": 0.048578746417733285, "全帯域_40ms_envelope_correlation": 0.9962760058544566, "全帯域_40ms_modulation_mae_db": 0.6717804723263546, "300–1000 Hz_10ms_log_rms_mae_db": 1.195464296571039, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11246778422684799, "300–1000 Hz_10ms_envelope_correlation": 0.9874383667526377, "300–1000 Hz_10ms_modulation_mae_db": 0.37616208983131644, "300–1000 Hz_40ms_log_rms_mae_db": 0.6852146328987679, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05913726104737901, "300–1000 Hz_40ms_envelope_correlation": 0.9964354792467964, "300–1000 Hz_40ms_modulation_mae_db": 0.5007528849411336, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1068002922526663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11932314569053211, "1000–3000 Hz_10ms_envelope_correlation": 0.9868938235531778, "1000–3000 Hz_10ms_modulation_mae_db": 0.47238119079167395, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6370601330944446, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05851009241653668, "1000–3000 Hz_40ms_envelope_correlation": 0.9966674815641627, "1000–3000 Hz_40ms_modulation_mae_db": 0.5700926907498166, "3000–6000 Hz_10ms_log_rms_mae_db": 1.177109103803114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1501163467282674, "3000–6000 Hz_10ms_envelope_correlation": 0.9760118958014948, "3000–6000 Hz_10ms_modulation_mae_db": 0.8001478179506059, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6426693060739833, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08038573445850307, "3000–6000 Hz_40ms_envelope_correlation": 0.9927669785349126, "3000–6000 Hz_40ms_modulation_mae_db": 0.7883998559663928, "mr_stft_db": 5.703290757188397, "band_mae_db": 0.10317929433129529, "rms_gain_error_db": -0.017300173729689464, "output_peak": 0.4530148044419058}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8341425081272288, "全帯域_10ms_rms_envelope_nrmse": 0.09765220952532895, "全帯域_10ms_envelope_correlation": 0.986509834189721, "全帯域_10ms_modulation_mae_db": 0.5368003712248003, "全帯域_40ms_log_rms_mae_db": 0.3846059123345931, "全帯域_40ms_rms_envelope_nrmse": 0.04229213550456493, "全帯域_40ms_envelope_correlation": 0.9974756044150227, "全帯域_40ms_modulation_mae_db": 0.26175606050543376, "300–1000 Hz_10ms_log_rms_mae_db": 1.246910260203413, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11133890557820236, "300–1000 Hz_10ms_envelope_correlation": 0.9872742581523707, "300–1000 Hz_10ms_modulation_mae_db": 0.669090835879457, "300–1000 Hz_40ms_log_rms_mae_db": 0.815059628779539, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06466852522320021, "300–1000 Hz_40ms_envelope_correlation": 0.9954623470024534, "300–1000 Hz_40ms_modulation_mae_db": 0.6156828899373978, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2059445679678893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13375717254250627, "1000–3000 Hz_10ms_envelope_correlation": 0.9835893607366988, "1000–3000 Hz_10ms_modulation_mae_db": 1.0496082711266828, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6651738769588803, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06534742470819145, "1000–3000 Hz_40ms_envelope_correlation": 0.9962638687682905, "1000–3000 Hz_40ms_modulation_mae_db": 0.9055255591958655, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1540458931030886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1420834006299944, "3000–6000 Hz_10ms_envelope_correlation": 0.978543538051753, "3000–6000 Hz_10ms_modulation_mae_db": 0.49283115690015017, "3000–6000 Hz_40ms_log_rms_mae_db": 0.597321609758353, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0662417062256364, "3000–6000 Hz_40ms_envelope_correlation": 0.9952178508699561, "3000–6000 Hz_40ms_modulation_mae_db": 0.4587881301954051, "mr_stft_db": 5.711432853110921, "band_mae_db": 0.1087763938693601, "rms_gain_error_db": -0.11942772554240046, "output_peak": 0.4369676829958905}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8330577511870262, "全帯域_10ms_rms_envelope_nrmse": 0.09098311356670602, "全帯域_10ms_envelope_correlation": 0.9882451284565217, "全帯域_10ms_modulation_mae_db": 0.2446942054140001, "全帯域_40ms_log_rms_mae_db": 0.37825768103333635, "全帯域_40ms_rms_envelope_nrmse": 0.036228281088045405, "全帯域_40ms_envelope_correlation": 0.9979335636081638, "全帯域_40ms_modulation_mae_db": 0.2679382469850016, "300–1000 Hz_10ms_log_rms_mae_db": 1.1507835205867871, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0977222359964001, "300–1000 Hz_10ms_envelope_correlation": 0.9902446741542535, "300–1000 Hz_10ms_modulation_mae_db": 0.6447047309983164, "300–1000 Hz_40ms_log_rms_mae_db": 0.6278559248018015, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049761782834881356, "300–1000 Hz_40ms_envelope_correlation": 0.997282193954882, "300–1000 Hz_40ms_modulation_mae_db": 0.574372123091907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.0774539237707308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13917503356309965, "1000–3000 Hz_10ms_envelope_correlation": 0.9827459620670038, "1000–3000 Hz_10ms_modulation_mae_db": 0.5895471857068829, "1000–3000 Hz_40ms_log_rms_mae_db": 0.5711692695754409, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07269046515081354, "1000–3000 Hz_40ms_envelope_correlation": 0.9951832489972566, "1000–3000 Hz_40ms_modulation_mae_db": 0.7251919139669206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.1923534582663666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14814864618857174, "3000–6000 Hz_10ms_envelope_correlation": 0.9767521088989117, "3000–6000 Hz_10ms_modulation_mae_db": 0.34889013267099084, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6816212004658634, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06763292357624512, "3000–6000 Hz_40ms_envelope_correlation": 0.9950345623318124, "3000–6000 Hz_40ms_modulation_mae_db": 0.40728052946222976, "mr_stft_db": 5.709331654339548, "band_mae_db": 0.1315546834742199, "rms_gain_error_db": -0.014726551148821324, "output_peak": 0.43500187021580033}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49200, "output_samples": 49200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.502131835158897, "spectral_convergence": 0.4528053178046751, "spectral_convergence_status": "ok", "amplitude_floor": 6.39186387874099e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.328259666152926, "spectral_convergence": 0.4122484461947544, "spectral_convergence_status": "ok", "amplitude_floor": 4.1641566181737356e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.324651111842096, "spectral_convergence": 0.4461433937920227, "spectral_convergence_status": "ok", "amplitude_floor": 3.471921787916526e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.471560290376807e-05, "output_power": 7.574783723273716e-05, "delta_db": 0.05958932027861894, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002820835097806379, "output_power": 0.00028550796840171857, "delta_db": 0.052405350877439374, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026780543872196867, "output_power": 0.0002671224101461447, "delta_db": -0.01109068085721844, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028630139823372175, "output_power": 0.0002627911417818718, "delta_db": -0.3721274718021129, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006167999147673381, "output_power": 0.0006213476883536692, "delta_db": 0.031903819070731365, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00037968124815517825, "output_power": 0.00037930577246351083, "delta_db": -0.004296965029777459, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.08856893465264976, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03558384515035499, "pitch": null, "output_rms_dbfs": -27.231195405575285, "output_peak": 0.4233474578129446, "reference_rms_dbfs": -27.195611560424933, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.718347537717972}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 0.00010992699921133752, "log_rms_mae_db": 0.8997501076592342, "rms_envelope_nrmse": 0.09398660389334425, "envelope_correlation": 0.9874206267579076, "rms_gain_error_db": -0.03556944119259793, "reference_mean_power": 0.0019104810338505282, "output_mean_power": 0.0018948977779166421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 4.701235881229352e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003075331225832077, "output_power": 0.0002994331518353693, "reference_active": true, "delta_db": -0.11592016215116877}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8607043791082874e-05, "output_power": 2.857516960589545e-05, "reference_active": true, "delta_db": -0.0048416398573663715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1458430393639735e-05, "output_power": 1.7069812396403074e-05, "reference_active": true, "delta_db": 1.7310361719381804}], "active_band_count": 3, "active_band_mae_db": 0.6172659913155718}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 9.464213695185308e-05, "log_rms_mae_db": 0.46018549005459847, "rms_envelope_nrmse": 0.044555603504592006, "envelope_correlation": 0.9968878319545506, "rms_gain_error_db": -0.035276622059756756, "reference_mean_power": 0.0019290839559002245, "output_mean_power": 0.0019134779713436182}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 4.127760176364045e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002739478020838986, "output_power": 0.0002659192862921214, "reference_active": true, "delta_db": -0.12917983896928836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7486141765201905e-05, "output_power": 1.700584108581e-05, "reference_active": true, "delta_db": -0.12095878602115147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8391125647823e-07, "output_power": 1.001339136000965e-06, "reference_active": true, "delta_db": 2.3423434277629904}], "active_band_count": 3, "active_band_mae_db": 0.8641606842511435}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 6.478025356347348e-05, "log_rms_mae_db": 1.205655786700764, "rms_envelope_nrmse": 0.10635502067640247, "envelope_correlation": 0.9884001076478063, "rms_gain_error_db": -0.010778097405643845, "reference_mean_power": 0.0004974887922363912, "output_mean_power": 0.000496255680871124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.1951630543463553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001563474072286115, "output_power": 0.0001520095869718815, "reference_active": true, "delta_db": -0.12219704577924814}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.160597194093242e-06, "output_power": 7.493458371002829e-06, "reference_active": true, "delta_db": -0.37039642690440544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0536955073753773e-06, "output_power": 2.83586360151177e-06, "reference_active": true, "delta_db": 1.4014928539790072}], "active_band_count": 3, "active_band_mae_db": 0.6313621088875536}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 5.647638237474489e-05, "log_rms_mae_db": 0.7154096924172432, "rms_envelope_nrmse": 0.06002757118344445, "envelope_correlation": 0.9960255554791104, "rms_gain_error_db": -0.010783369461557526, "reference_mean_power": 0.00050239739349007, "output_mean_power": 0.000501151506947944}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.0265825460069485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014180308081562638, "output_power": 0.0001382862165188661, "reference_active": true, "delta_db": -0.10906771853003955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4691763208394e-06, "output_power": 4.0709535256286504e-06, "reference_active": true, "delta_db": -0.4053134460274259}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.310040798443831e-08, "output_power": 1.3160979136560137e-07, "reference_active": true, "delta_db": 1.5033661650990184}], "active_band_count": 3, "active_band_mae_db": 0.672582443218828}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 7.855020595588598e-05, "log_rms_mae_db": 1.1524003352254972, "rms_envelope_nrmse": 0.12964916608575547, "envelope_correlation": 0.9846113174248636, "rms_gain_error_db": -0.18836965591689936, "reference_mean_power": 0.0004940955013379756, "output_mean_power": 0.0004731228618061252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 1.3474634060369984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.364408948380063e-05, "output_power": 9.654927628498278e-05, "reference_active": true, "delta_db": 0.13268651444757185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.996833569277459e-06, "output_power": 7.294757809047223e-06, "reference_active": true, "delta_db": 0.18109333938997746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2622375016010212e-06, "output_power": 2.464647745043275e-06, "reference_active": true, "delta_db": 0.3721665985563374}], "active_band_count": 3, "active_band_mae_db": 0.22864881746462892}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 6.30584220600245e-05, "log_rms_mae_db": 0.5819126330998655, "rms_envelope_nrmse": 0.06141617416464626, "envelope_correlation": 0.9967210306117961, "rms_gain_error_db": -0.1883702883865641, "reference_mean_power": 0.0004989699589215838, "output_mean_power": 0.00047779034599822363}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 1.243238840477423e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.827683631948728e-05, "output_power": 9.102347481705256e-05, "reference_active": true, "delta_db": 0.1330665021129999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8441959267659476e-06, "output_power": 3.864247090942343e-06, "reference_active": true, "delta_db": 0.022593744490132087}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.483409815447214e-08, "output_power": 1.0444699135593292e-07, "reference_active": true, "delta_db": 0.4193141590446187}], "active_band_count": 3, "active_band_mae_db": 0.1916581352159169}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 9.512970769085293e-05, "log_rms_mae_db": 1.1708430375540586, "rms_envelope_nrmse": 0.15167090324433183, "envelope_correlation": 0.975903212805256, "rms_gain_error_db": -0.010222479269381664, "reference_mean_power": 0.0007033343720650757, "output_mean_power": 0.0007016808014520911}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.9605352394438525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001169604983627503, "output_power": 0.00011086017969113088, "reference_active": true, "delta_db": -0.2326363192790679}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1203242070906313e-05, "output_power": 2.067482188894354e-05, "reference_active": true, "delta_db": -0.10960494735237733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.096712651632301e-05, "output_power": 1.5864712918137874e-05, "reference_active": true, "delta_db": 1.6033936417620829}], "active_band_count": 3, "active_band_mae_db": 0.6485449694645093}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 7.035164526339387e-05, "log_rms_mae_db": 0.6946471020561796, "rms_envelope_nrmse": 0.08403263321991117, "envelope_correlation": 0.991912044301681, "rms_gain_error_db": -0.010223163957559855, "reference_mean_power": 0.0007102740547296548, "output_mean_power": 0.0007086040568966993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.6373521771123283e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010173798550296795, "output_power": 9.782171875716666e-05, "reference_active": true, "delta_db": -0.17047844749611754}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.290709427246265e-05, "output_power": 1.2094736079788933e-05, "reference_active": true, "delta_db": -0.28232086094956144}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.316189457175965e-07, "output_power": 1.0172409326517815e-06, "reference_active": true, "delta_db": 2.0696867897970646}], "active_band_count": 3, "active_band_mae_db": 0.8408286994142479}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8401069971554366, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "747d30140670843342692603d2b7885032e270886135f7bd569d2fd6fe59c142", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9852115115312768, "standard_deviation": 0.0023270167827238223, "minimum": 0.9819924500246789, "maximum": 0.9886326664479087, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1060469837934788, "standard_deviation": 0.05491377482255769, "minimum": 1.0262250388279128, "maximum": 1.2059445679678893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5905246949476055, "standard_deviation": 0.28131719476729417, "minimum": 0.22864881746462892, "maximum": 1.0496082711266828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12705259809233266, "standard_deviation": 0.010590146073272732, "minimum": 0.1110101710431771, "maximum": 0.1411344734288001, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9965547638486782, "standard_deviation": 0.0008892122173850117, "minimum": 0.9951832489972566, "maximum": 0.9975378045839671, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6111932335944867, "standard_deviation": 0.031067116770094026, "minimum": 0.5711692695754409, "maximum": 0.6651738769588803, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6932951899174872, "standard_deviation": 0.32111053853496563, "minimum": 0.1916581352159169, "maximum": 1.2082054302640042, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06198383578066344, "standard_deviation": 0.008434413835176464, "minimum": 0.051716726928110114, "maximum": 0.07499186158577391, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9791681354174553, "standard_deviation": 0.0029959449119450936, "minimum": 0.975903212805256, "maximum": 0.9834562829709654, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.1526294006076518, "standard_deviation": 0.03338227332201437, "minimum": 1.0931749635022976, "maximum": 1.1923534582663666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5017194034237191, "standard_deviation": 0.18356657754791417, "minimum": 0.2393598159875857, "maximum": 0.8001478179506059, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14044943362034312, "standard_deviation": 0.00969011237231722, "minimum": 0.12681284073560997, "maximum": 0.15167090324433183, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9946757470850455, "standard_deviation": 0.0015226126712778068, "minimum": 0.991912044301681, "maximum": 0.996247257603276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6410225049673275, "standard_deviation": 0.04104371398047589, "minimum": 0.58313906135347, "maximum": 0.6946471020561796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.47510973063418716, "standard_deviation": 0.23279927997848637, "minimum": 0.17947349863598536, "maximum": 0.8408286994142479, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06965412437293236, "standard_deviation": 0.008325626655520724, "minimum": 0.059869525998938675, "maximum": 0.08403263321991117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9890212208014675, "standard_deviation": 0.0021925155207262246, "minimum": 0.9872521574217288, "maximum": 0.9928208244390899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.199419689424666, "standard_deviation": 0.050063348199861024, "minimum": 1.1044011745744604, "maximum": 1.254163314242096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6588312893704656, "standard_deviation": 0.2536095969413172, "minimum": 0.3621373754793183, "maximum": 1.1733952160689889, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10328255246775504, "standard_deviation": 0.010892052547001967, "minimum": 0.08375066022487017, "maximum": 0.11246778422684799, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961848495490966, "standard_deviation": 0.0013601277918005726, "minimum": 0.993398266957471, "maximum": 0.9974814150081128, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7013341800973044, "standard_deviation": 0.07370148535977285, "minimum": 0.6133927305703231, "maximum": 0.815059628779539, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6788490790763856, "standard_deviation": 0.3133279098703876, "minimum": 0.18247763679739495, "maximum": 1.2579308005844136, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05854520770036009, "standard_deviation": 0.00994707320229522, "minimum": 0.04815506321552777, "maximum": 0.07741903672877969, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11873041430499703, "standard_deviation": 0.025447902469043465, "minimum": 0.08856893465264976, "maximum": 0.16031801206450527, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.708223798720078, "standard_deviation": 0.00665550154201373, "minimum": 5.697316854803691, "maximum": 5.718347537717972, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44164230736196874, "standard_deviation": 0.017310441944436795, "minimum": 0.4233474578129446, "maximum": 0.4743658113432926, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07044809453210765, "standard_deviation": 0.0531604800801243, "minimum": -0.15972073883533217, "maximum": -0.014726551148821324, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9884129638900074, "standard_deviation": 0.0015273582563732012, "minimum": 0.986509834189721, "maximum": 0.9912651749576178, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8180430897361703, "standard_deviation": 0.0442585916357126, "minimum": 0.7591649073389285, "maximum": 0.8997501076592342, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.43382805819245357, "standard_deviation": 0.15420119147377898, "minimum": 0.24228950707681787, "maximum": 0.6201164249174386, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09033477692469645, "standard_deviation": 0.005951034300332647, "minimum": 0.07891990142468341, "maximum": 0.09765220952532895, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972509092551054, "standard_deviation": 0.0006262276931396563, "minimum": 0.9962760058544566, "maximum": 0.9981582633222296, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4169733884806526, "standard_deviation": 0.03304316880501114, "minimum": 0.37825768103333635, "maximum": 0.46018549005459847, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4294090374371716, "standard_deviation": 0.22814408192993957, "minimum": 0.21674326611894892, "maximum": 0.8641606842511435, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.042566939629300284, "standard_deviation": 0.004834750443936697, "minimum": 0.03617271926102287, "maximum": 0.04891737011054326, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7021-79740-0002", "language": "en", "style": "normal", "speaker": "librispeech_7021", "source_utterance_id": "7021-79740-0002", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7021-79740-0002.wav", "source_file_sha256": "923529e19409395917f7d88da5d228de38bbceaaa4bafaba5dc795236e43dc93", "decoded_pcm_sha256": "32223cc598805a0301d5fd306db57f06a7d2c624ffce3ec523d6a45df0f1b192", "original_source_sha256": "df09e0bfaf8000a8183b367c2fbc6ce3404f4152ef14d9340515b37edb7155ba", "corpus_entry_id": "librispeech_7021-79740-0002"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.963557004928589, "predicted_mos": 3.963557004928589, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4f5ecd7381a554666ca2e944f50c4eb5316c4bb7ec21321405fca3a208837baf", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2387361790024443, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7021-79740-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "4f5ecd7381a554666ca2e944f50c4eb5316c4bb7ec21321405fca3a208837baf", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "6bdd332e5cfad3622e452101f852fa3f8dd6eb5b9be2cdc45f5708ddd2557256"}, "rows": [{"case": "librispeech_7021-79740-0002", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.5207464694976807, "predicted_mos": 2.5207464694976807, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7d5fbf48eb56399ab79d658857aea04339044be9b0cf1887f027b7a441fe9f1b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3295524820023275, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7021-79740-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "7d5fbf48eb56399ab79d658857aea04339044be9b0cf1887f027b7a441fe9f1b", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.911839150665703, "全帯域_10ms_rms_envelope_nrmse": 0.09652956266587373, "全帯域_10ms_envelope_correlation": 0.9916960791861289, "全帯域_10ms_modulation_mae_db": 0.17012003789572513, "全帯域_40ms_log_rms_mae_db": 0.5188160588084456, "全帯域_40ms_rms_envelope_nrmse": 0.046744800065905756, "全帯域_40ms_envelope_correlation": 0.998164536865333, "全帯域_40ms_modulation_mae_db": 0.35040319387619756, "300–1000 Hz_10ms_log_rms_mae_db": 1.5831758467077854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1812377001033731, "300–1000 Hz_10ms_envelope_correlation": 0.9792688159122904, "300–1000 Hz_10ms_modulation_mae_db": 0.6717968095601109, "300–1000 Hz_40ms_log_rms_mae_db": 1.0480528272154759, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12778667434507832, "300–1000 Hz_40ms_envelope_correlation": 0.9910858600712504, "300–1000 Hz_40ms_modulation_mae_db": 1.0361354131703868, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8660617782073001, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20254396066112829, "1000–3000 Hz_10ms_envelope_correlation": 0.9723726231505592, "1000–3000 Hz_10ms_modulation_mae_db": 0.2144049658985967, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5041183360763235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12641738477203543, "1000–3000 Hz_40ms_envelope_correlation": 0.9888914621524562, "1000–3000 Hz_40ms_modulation_mae_db": 0.24033798651925045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.811915194895922, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1932510185973416, "3000–6000 Hz_10ms_envelope_correlation": 0.9765040887842299, "3000–6000 Hz_10ms_modulation_mae_db": 0.5764407034994536, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4509389542694715, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1086379573018809, "3000–6000 Hz_40ms_envelope_correlation": 0.9922541355943757, "3000–6000 Hz_40ms_modulation_mae_db": 0.649684420405085, "mr_stft_db": 5.389307820306563, "band_mae_db": 0.18997943115966856, "rms_gain_error_db": -0.15663202089411576, "output_peak": 0.5314650263977498}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9338892856841418, "全帯域_10ms_rms_envelope_nrmse": 0.10681044594708025, "全帯域_10ms_envelope_correlation": 0.9896668815728772, "全帯域_10ms_modulation_mae_db": 0.16744017191486646, "全帯域_40ms_log_rms_mae_db": 0.5551850283966258, "全帯域_40ms_rms_envelope_nrmse": 0.05245065281613216, "全帯域_40ms_envelope_correlation": 0.9973496005128883, "全帯域_40ms_modulation_mae_db": 0.17929091264821673, "300–1000 Hz_10ms_log_rms_mae_db": 1.5204137923485403, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1612856389998144, "300–1000 Hz_10ms_envelope_correlation": 0.9821372032971901, "300–1000 Hz_10ms_modulation_mae_db": 0.7015397042841564, "300–1000 Hz_40ms_log_rms_mae_db": 0.9363336034378456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.085998612816738, "300–1000 Hz_40ms_envelope_correlation": 0.994725031742785, "300–1000 Hz_40ms_modulation_mae_db": 0.7303991879197698, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8906574171402184, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2465325719238476, "1000–3000 Hz_10ms_envelope_correlation": 0.9600660522469251, "1000–3000 Hz_10ms_modulation_mae_db": 2.3582321889976505, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5081502100598967, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15330487047121533, "1000–3000 Hz_40ms_envelope_correlation": 0.9862828057368698, "1000–3000 Hz_40ms_modulation_mae_db": 2.9927774884364404, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7696728999583515, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18534441915653901, "3000–6000 Hz_10ms_envelope_correlation": 0.978604949495376, "3000–6000 Hz_10ms_modulation_mae_db": 0.4517474739245803, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4299527120661804, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10984273290523067, "3000–6000 Hz_40ms_envelope_correlation": 0.992909419710716, "3000–6000 Hz_40ms_modulation_mae_db": 0.40864360810204875, "mr_stft_db": 5.383282614445956, "band_mae_db": 0.23154616290756935, "rms_gain_error_db": -0.04627722545927727, "output_peak": 0.6381534805013808}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9128764234311799, "全帯域_10ms_rms_envelope_nrmse": 0.1106759054304605, "全帯域_10ms_envelope_correlation": 0.9889090315326248, "全帯域_10ms_modulation_mae_db": 0.6862823973934565, "全帯域_40ms_log_rms_mae_db": 0.5602506311391724, "全帯域_40ms_rms_envelope_nrmse": 0.06060191500883329, "全帯域_40ms_envelope_correlation": 0.9964439986866611, "全帯域_40ms_modulation_mae_db": 0.3510647538758362, "300–1000 Hz_10ms_log_rms_mae_db": 1.5833303954206157, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19133135582743405, "300–1000 Hz_10ms_envelope_correlation": 0.9748224151248887, "300–1000 Hz_10ms_modulation_mae_db": 1.3123335710517683, "300–1000 Hz_40ms_log_rms_mae_db": 1.050523466182682, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11546316566007607, "300–1000 Hz_40ms_envelope_correlation": 0.9903771034487422, "300–1000 Hz_40ms_modulation_mae_db": 0.9441535816530484, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8769764767750319, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22018196599087964, "1000–3000 Hz_10ms_envelope_correlation": 0.9679453767474265, "1000–3000 Hz_10ms_modulation_mae_db": 1.7927666726099292, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5292035436460591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1353442928489163, "1000–3000 Hz_40ms_envelope_correlation": 0.9886215983617253, "1000–3000 Hz_40ms_modulation_mae_db": 2.290843574808816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8160214591912534, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18504453647020178, "3000–6000 Hz_10ms_envelope_correlation": 0.9783022276144636, "3000–6000 Hz_10ms_modulation_mae_db": 0.724691321552306, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3951068980647106, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10207227218372975, "3000–6000 Hz_40ms_envelope_correlation": 0.9933116114196089, "3000–6000 Hz_40ms_modulation_mae_db": 0.829140271820985, "mr_stft_db": 5.386335021637603, "band_mae_db": 0.3064525773343733, "rms_gain_error_db": -0.027752001396948337, "output_peak": 0.7234395224059734}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9160342256597362, "全帯域_10ms_rms_envelope_nrmse": 0.09704838159841647, "全帯域_10ms_envelope_correlation": 0.9914956910596837, "全帯域_10ms_modulation_mae_db": 0.31651526652089657, "全帯域_40ms_log_rms_mae_db": 0.5633460090068511, "全帯域_40ms_rms_envelope_nrmse": 0.053830249227895414, "全帯域_40ms_envelope_correlation": 0.9972776301830265, "全帯域_40ms_modulation_mae_db": 0.4487156190690615, "300–1000 Hz_10ms_log_rms_mae_db": 1.492012681626378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14899833866310638, "300–1000 Hz_10ms_envelope_correlation": 0.9850093250125189, "300–1000 Hz_10ms_modulation_mae_db": 0.6232631596082537, "300–1000 Hz_40ms_log_rms_mae_db": 0.9936486450379944, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0876975801567857, "300–1000 Hz_40ms_envelope_correlation": 0.9950606256276538, "300–1000 Hz_40ms_modulation_mae_db": 0.6301697585445277, "1000–3000 Hz_10ms_log_rms_mae_db": 1.789410708452459, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22482954941227065, "1000–3000 Hz_10ms_envelope_correlation": 0.9666277349501454, "1000–3000 Hz_10ms_modulation_mae_db": 1.9058028578382842, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4586104782129552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1310704642419028, "1000–3000 Hz_40ms_envelope_correlation": 0.9897713028702225, "1000–3000 Hz_40ms_modulation_mae_db": 2.805746597255286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8212011000526744, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18779818312461902, "3000–6000 Hz_10ms_envelope_correlation": 0.9777701088143077, "3000–6000 Hz_10ms_modulation_mae_db": 1.0870324276756083, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4626847830238934, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11580887229319585, "3000–6000 Hz_40ms_envelope_correlation": 0.9915104345987343, "3000–6000 Hz_40ms_modulation_mae_db": 1.1904077949339091, "mr_stft_db": 5.393231383776175, "band_mae_db": 0.41169423595858845, "rms_gain_error_db": -0.09444376375520173, "output_peak": 0.5240456961279236}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9018973771668239, "全帯域_10ms_rms_envelope_nrmse": 0.10933564678365103, "全帯域_10ms_envelope_correlation": 0.9891305513857149, "全帯域_10ms_modulation_mae_db": 0.4757572470168805, "全帯域_40ms_log_rms_mae_db": 0.5421906886153711, "全帯域_40ms_rms_envelope_nrmse": 0.06273920760620993, "全帯域_40ms_envelope_correlation": 0.9961762980939153, "全帯域_40ms_modulation_mae_db": 0.17942805292239436, "300–1000 Hz_10ms_log_rms_mae_db": 1.5344981472481862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18307350298727412, "300–1000 Hz_10ms_envelope_correlation": 0.9769638542382855, "300–1000 Hz_10ms_modulation_mae_db": 0.7797091714650485, "300–1000 Hz_40ms_log_rms_mae_db": 1.0255536228368936, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1128581020744489, "300–1000 Hz_40ms_envelope_correlation": 0.9913044732243778, "300–1000 Hz_40ms_modulation_mae_db": 0.7044857056417033, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8395956300305545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2197908064227271, "1000–3000 Hz_10ms_envelope_correlation": 0.9681566792425116, "1000–3000 Hz_10ms_modulation_mae_db": 0.801049920719963, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4733415616210508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14359529510797947, "1000–3000 Hz_40ms_envelope_correlation": 0.9855644526159223, "1000–3000 Hz_40ms_modulation_mae_db": 0.6931417536801096, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7570563825611263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18440878895172821, "3000–6000 Hz_10ms_envelope_correlation": 0.9795605118133589, "3000–6000 Hz_10ms_modulation_mae_db": 0.4565885943422732, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3691874724526731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10769759942425929, "3000–6000 Hz_40ms_envelope_correlation": 0.9929594398223934, "3000–6000 Hz_40ms_modulation_mae_db": 0.4919981424911657, "mr_stft_db": 5.385891158131861, "band_mae_db": 0.28748250724750807, "rms_gain_error_db": -0.03848517223270467, "output_peak": 0.6043629611708807}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8995180851464378, "全帯域_10ms_rms_envelope_nrmse": 0.09875182946816527, "全帯域_10ms_envelope_correlation": 0.9912800654915744, "全帯域_10ms_modulation_mae_db": 0.3903406438011676, "全帯域_40ms_log_rms_mae_db": 0.5645298005010232, "全帯域_40ms_rms_envelope_nrmse": 0.052575455054073364, "全帯域_40ms_envelope_correlation": 0.9975522495871249, "全帯域_40ms_modulation_mae_db": 0.21153774798684574, "300–1000 Hz_10ms_log_rms_mae_db": 1.5222779521134258, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1701164798235395, "300–1000 Hz_10ms_envelope_correlation": 0.9808433985333318, "300–1000 Hz_10ms_modulation_mae_db": 0.9813269961077987, "300–1000 Hz_40ms_log_rms_mae_db": 1.0428793689605853, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10081997391424061, "300–1000 Hz_40ms_envelope_correlation": 0.9941052201386869, "300–1000 Hz_40ms_modulation_mae_db": 0.5116112396437288, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8708595164732167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23259745320292577, "1000–3000 Hz_10ms_envelope_correlation": 0.963441473181811, "1000–3000 Hz_10ms_modulation_mae_db": 0.8530325410777825, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5050657393039648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12780529404676938, "1000–3000 Hz_40ms_envelope_correlation": 0.9890749632212639, "1000–3000 Hz_40ms_modulation_mae_db": 1.14165343120996, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8251930872902107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18605960008732247, "3000–6000 Hz_10ms_envelope_correlation": 0.9782426527891366, "3000–6000 Hz_10ms_modulation_mae_db": 0.45561880543899913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.433504504971056, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11334242662689531, "3000–6000 Hz_40ms_envelope_correlation": 0.9916925206074696, "3000–6000 Hz_40ms_modulation_mae_db": 0.42604927917367785, "mr_stft_db": 5.3991977633311885, "band_mae_db": 0.15047100958663615, "rms_gain_error_db": -0.1495463335840824, "output_peak": 0.6187384213051708}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9372761604462121, "全帯域_10ms_rms_envelope_nrmse": 0.10115233834758738, "全帯域_10ms_envelope_correlation": 0.9908169483045873, "全帯域_10ms_modulation_mae_db": 0.1624860258894352, "全帯域_40ms_log_rms_mae_db": 0.5677387353420599, "全帯域_40ms_rms_envelope_nrmse": 0.05655832032718111, "全帯域_40ms_envelope_correlation": 0.9971050383742797, "全帯域_40ms_modulation_mae_db": 0.33943152822657474, "300–1000 Hz_10ms_log_rms_mae_db": 1.4824479559049075, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18312524235502933, "300–1000 Hz_10ms_envelope_correlation": 0.9784772170064259, "300–1000 Hz_10ms_modulation_mae_db": 0.6446714460395423, "300–1000 Hz_40ms_log_rms_mae_db": 0.9929547088058696, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12449895150135278, "300–1000 Hz_40ms_envelope_correlation": 0.9913699293734409, "300–1000 Hz_40ms_modulation_mae_db": 0.5690149851403925, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8273832939066843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21506471428397492, "1000–3000 Hz_10ms_envelope_correlation": 0.969005626437207, "1000–3000 Hz_10ms_modulation_mae_db": 1.424415596030708, "1000–3000 Hz_40ms_log_rms_mae_db": 1.471075733686045, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12787695276261432, "1000–3000 Hz_40ms_envelope_correlation": 0.9883910337640062, "1000–3000 Hz_40ms_modulation_mae_db": 1.5039468558714584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8005738577440302, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1774697418704711, "3000–6000 Hz_10ms_envelope_correlation": 0.9806390926874767, "3000–6000 Hz_10ms_modulation_mae_db": 0.9877531740001544, "3000–6000 Hz_40ms_log_rms_mae_db": 1.460851118231619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1114701724226688, "3000–6000 Hz_40ms_envelope_correlation": 0.9928582645938181, "3000–6000 Hz_40ms_modulation_mae_db": 1.382732458421777, "mr_stft_db": 5.400892840096596, "band_mae_db": 0.21108279403470376, "rms_gain_error_db": -0.14320797432841786, "output_peak": 0.5851888444453393}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9823601278236965, "全帯域_10ms_rms_envelope_nrmse": 0.11445453869995037, "全帯域_10ms_envelope_correlation": 0.9881012644746338, "全帯域_10ms_modulation_mae_db": 0.6298980038510026, "全帯域_40ms_log_rms_mae_db": 0.6064980784704904, "全帯域_40ms_rms_envelope_nrmse": 0.057949075325056075, "全帯域_40ms_envelope_correlation": 0.9968467361798049, "全帯域_40ms_modulation_mae_db": 0.4828184765200962, "300–1000 Hz_10ms_log_rms_mae_db": 1.5195012939599493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18862439516945512, "300–1000 Hz_10ms_envelope_correlation": 0.9763662095022729, "300–1000 Hz_10ms_modulation_mae_db": 0.7147156458104224, "300–1000 Hz_40ms_log_rms_mae_db": 0.9796395085959391, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11510504647079069, "300–1000 Hz_40ms_envelope_correlation": 0.9922947623836094, "300–1000 Hz_40ms_modulation_mae_db": 0.6633535604540529, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8724757916152006, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26697085874621534, "1000–3000 Hz_10ms_envelope_correlation": 0.9520968468412386, "1000–3000 Hz_10ms_modulation_mae_db": 2.17822224764949, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4821885456005683, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14742351217489777, "1000–3000 Hz_40ms_envelope_correlation": 0.9871019373465183, "1000–3000 Hz_40ms_modulation_mae_db": 2.5996267174329946, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7533597639299923, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1764697040769174, "3000–6000 Hz_10ms_envelope_correlation": 0.9804921584489271, "3000–6000 Hz_10ms_modulation_mae_db": 0.5121937725795238, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3746173329153124, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10146894341557887, "3000–6000 Hz_40ms_envelope_correlation": 0.9932685393478443, "3000–6000 Hz_40ms_modulation_mae_db": 0.4555589074379, "mr_stft_db": 5.3843662418323355, "band_mae_db": 0.2577037023105218, "rms_gain_error_db": -0.10169735259443652, "output_peak": 0.6363787183650133}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.715018718804462, "spectral_convergence": 0.31123105781703253, "spectral_convergence_status": "ok", "amplitude_floor": 1.8465213976940698e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.214207892173723, "spectral_convergence": 0.33605872822811217, "spectral_convergence_status": "ok", "amplitude_floor": 1.4089022879434144e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.238696849941505, "spectral_convergence": 0.3543949801218316, "spectral_convergence_status": "ok", "amplitude_floor": 1.0260349239920231e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002262162256014784, "output_power": 0.002222043959152864, "delta_db": -0.0777110551734707, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014828754084214857, "output_power": 0.001377784111548232, "delta_db": -0.3192349081582862, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026062691889949753, "output_power": 0.00026740743268524494, "delta_db": 0.11154204585145917, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020447377727008734, "output_power": 0.0001950278623766884, "delta_db": -0.20540959184021812, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.078753769921038e-05, "output_power": 3.7422595189341946e-05, "delta_db": -0.37393586461188316, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.800369964531977e-05, "output_power": 2.7670123582876806e-05, "delta_db": -0.05204312132269388, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18997943115966856, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15663202089411576, "pitch": null, "output_rms_dbfs": -23.84328061077129, "output_peak": 0.5314650263977498, "reference_rms_dbfs": -23.686648589877173, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.389307820306563}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022610981418459577, "log_rms_mae_db": 0.911839150665703, "rms_envelope_nrmse": 0.09652956266587373, "envelope_correlation": 0.9916960791861289, "rms_gain_error_db": -0.15655008861979156, "reference_mean_power": 0.00428070632223443, "output_mean_power": 0.004129147769632986}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.20587323032393e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001500467490243704, "output_power": 0.0014298697630912936, "reference_active": true, "delta_db": -0.2093010785938129}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013253274491379155, "output_power": 0.00012654774062608956, "reference_active": true, "delta_db": -0.20068797124842916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.272532196136399e-05, "output_power": 3.197767038009117e-05, "reference_active": true, "delta_db": -0.10037106384493329}], "active_band_count": 3, "active_band_mae_db": 0.17012003789572513}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019856877416737668, "log_rms_mae_db": 0.5188160588084456, "rms_envelope_nrmse": 0.046744800065905756, "envelope_correlation": 0.998164536865333, "rms_gain_error_db": -0.15674839819579367, "reference_mean_power": 0.004296662851053013, "output_mean_power": 0.004144350111434947}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.009968768189794e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014002280466658247, "output_power": 0.0013377164158606925, "reference_active": true, "delta_db": -0.19834715781301654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3360463722840526e-05, "output_power": 5.1179729812186676e-05, "reference_active": true, "delta_db": -0.18121606278435706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498078024970164e-06, "output_power": 2.1401359527198764e-06, "reference_active": true, "delta_db": -0.6716463610312191}], "active_band_count": 3, "active_band_mae_db": 0.35040319387619756}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000212457290437723, "log_rms_mae_db": 1.5831758467077854, "rms_envelope_nrmse": 0.1812377001033731, "envelope_correlation": 0.9792688159122904, "rms_gain_error_db": -0.6064464455211112, "reference_mean_power": 0.0012391682631200216, "output_mean_power": 0.0010776696134646596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.921692896601052e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00036532110371103373, "output_power": 0.00032044458996103524, "reference_active": true, "delta_db": -0.5692181724690936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3180096390035205e-05, "output_power": 3.66201388741739e-05, "reference_active": true, "delta_db": -0.7156362052660227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1750484667690236e-05, "output_power": 9.9312102221514e-06, "reference_active": true, "delta_db": -0.7305360509452165}], "active_band_count": 3, "active_band_mae_db": 0.6717968095601109}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018128574308840577, "log_rms_mae_db": 1.0480528272154759, "rms_envelope_nrmse": 0.12778667434507832, "envelope_correlation": 0.9910858600712504, "rms_gain_error_db": -0.6070594904402985, "reference_mean_power": 0.0012444705639415084, "output_mean_power": 0.0010821281119227327}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.573537659747316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003540640123239247, "output_power": 0.0003112649385847036, "reference_active": true, "delta_db": -0.559515827981872}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0107771514146375e-05, "output_power": 1.734904231349995e-05, "reference_active": true, "delta_db": -0.6408843542056073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973740035567712e-07, "output_power": 5.138797056752525e-07, "reference_active": true, "delta_db": -1.9080060573236814}], "active_band_count": 3, "active_band_mae_db": 1.0361354131703868}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001386433646872303, "log_rms_mae_db": 1.8660617782073001, "rms_envelope_nrmse": 0.20254396066112829, "envelope_correlation": 0.9723726231505592, "rms_gain_error_db": -0.23904714437681363, "reference_mean_power": 0.00021050674147203463, "output_mean_power": 0.00019923300904676938}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9020912968770603e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.015318920526352e-05, "output_power": 8.417509000640578e-05, "reference_active": true, "delta_db": 0.21262782318045853}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8603151739641485e-05, "output_power": 3.798309911210081e-05, "reference_active": true, "delta_db": -0.07032367181331363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0682238748218467e-05, "output_power": 2.2471076975353028e-05, "reference_active": true, "delta_db": 0.360263402702018}], "active_band_count": 3, "active_band_mae_db": 0.2144049658985967}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.950462169412164e-05, "log_rms_mae_db": 1.5041183360763235, "rms_envelope_nrmse": 0.12641738477203543, "envelope_correlation": 0.9888914621524562, "rms_gain_error_db": -0.24266381033809642, "reference_mean_power": 0.0002110528457073582, "output_mean_power": 0.00019958359048081672}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7773203833688661e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04144461796749e-05, "output_power": 9.663545828947421e-05, "reference_active": true, "delta_db": 0.28898684180789935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4052463298203573e-05, "output_power": 2.390662391883518e-05, "reference_active": true, "delta_db": -0.026413110261225694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4632154850806674e-06, "output_power": 1.6064591604038328e-06, "reference_active": true, "delta_db": 0.4056140074886264}], "active_band_count": 3, "active_band_mae_db": 0.24033798651925045}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.6125957709065377e-05, "log_rms_mae_db": 1.811915194895922, "rms_envelope_nrmse": 0.1932510185973416, "envelope_correlation": 0.9765040887842299, "rms_gain_error_db": -0.0016481975284485095, "reference_mean_power": 2.4404568133397883e-05, "output_mean_power": 2.4395308076254228e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.103592290864487e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1042852972405107e-05, "output_power": 1.1699523187994041e-05, "reference_active": true, "delta_db": 0.25086872623617407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8700267025557585e-06, "output_power": 4.1628763733983614e-06, "reference_active": true, "delta_db": 0.316795520710222}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.131168166122288e-06, "output_power": 8.656879814482391e-07, "reference_active": true, "delta_db": -1.161657863551965}], "active_band_count": 3, "active_band_mae_db": 0.5764407034994536}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.642002234154929e-05, "log_rms_mae_db": 1.4509389542694715, "rms_envelope_nrmse": 0.1086379573018809, "envelope_correlation": 0.9922541355943757, "rms_gain_error_db": -0.0027360155867368185, "reference_mean_power": 2.4522755214627932e-05, "output_mean_power": 2.4507310968103082e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9657184130662428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1868339578364968e-05, "output_power": 1.2569048517691915e-05, "reference_active": true, "delta_db": 0.2491243874491236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9915566770914906e-06, "output_power": 2.073759099449458e-06, "reference_active": true, "delta_db": 0.17565634305075056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.546174499350975e-08, "output_power": 6.016508063693071e-08, "reference_active": true, "delta_db": -1.5242725307153808}], "active_band_count": 3, "active_band_mae_db": 0.649684420405085}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8719065031824998, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2cb4f8435b1149d7ddc22a0f7cc6fb42b382d9bceec6371c6a30f1c131f8ef7b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9649640515997281, "standard_deviation": 0.006380146368435288, "minimum": 0.9520968468412386, "maximum": 0.9723726231505592, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8541775765750832, "standard_deviation": 0.03323128606340362, "minimum": 1.789410708452459, "maximum": 1.8906574171402184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.4409908738528006, "standard_deviation": 0.7544766659053344, "minimum": 0.2144049658985967, "maximum": 2.3582321889976505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22856398508049616, "standard_deviation": 0.020123446961827846, "minimum": 0.20254396066112829, "maximum": 0.26697085874621534, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.987962444508623, "standard_deviation": 0.0014785580277426609, "minimum": 0.9855644526159223, "maximum": 0.9897713028702225, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.491469268525858, "standard_deviation": 0.023776741926746628, "minimum": 1.4586104782129552, "maximum": 1.5292035436460591, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.7835093006517895, "standard_deviation": 1.0344872818511408, "minimum": 0.24033798651925045, "maximum": 2.9927774884364404, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13660475830329136, "standard_deviation": 0.010242849148245173, "minimum": 0.12641738477203543, "maximum": 0.15330487047121533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9787644738059096, "standard_deviation": 0.0014011365585333824, "minimum": 0.9765040887842299, "maximum": 0.9806390926874767, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.794374218202945, "standard_deviation": 0.029686866160703894, "minimum": 1.7533597639299923, "maximum": 1.8251930872902107, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6565082841266123, "standard_deviation": 0.2533242112264514, "minimum": 0.4517474739245803, "maximum": 1.0870324276756083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18448074904189257, "standard_deviation": 0.005409195757048928, "minimum": 0.1764697040769174, "maximum": 0.1932510185973416, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.99259554571187, "standard_deviation": 0.0006943762760280574, "minimum": 0.9915104345987343, "maximum": 0.9933116114196089, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4221054719993647, "standard_deviation": 0.03771160003150201, "minimum": 1.3691874724526731, "maximum": 1.4626847830238934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7292768603483186, "standard_deviation": 0.3745337999351128, "minimum": 0.40864360810204875, "maximum": 1.382732458421777, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10879262207167992, "standard_deviation": 0.005045830878345939, "minimum": 0.10146894341557887, "maximum": 0.11580887229319585, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9792360548284005, "standard_deviation": 0.003333528720681986, "minimum": 0.9748224151248887, "maximum": 0.9850093250125189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5297072581662237, "standard_deviation": 0.03714321891443778, "minimum": 1.4824479559049075, "maximum": 1.5833303954206157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8036695629908877, "standard_deviation": 0.23434892760718778, "minimum": 0.6232631596082537, "maximum": 1.3123335710517683, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17597408174112825, "standard_deviation": 0.014652524992629694, "minimum": 0.14899833866310638, "maximum": 0.19133135582743405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9925403757513183, "standard_deviation": 0.0018254630653828066, "minimum": 0.9903771034487422, "maximum": 0.9950606256276538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0086982188841607, "standard_deviation": 0.04018881537068828, "minimum": 0.9363336034378456, "maximum": 1.050523466182682, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7236654290209512, "standard_deviation": 0.18044723116904643, "minimum": 0.5116112396437288, "maximum": 1.0361354131703868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10877851336743888, "standard_deviation": 0.015750115485443517, "minimum": 0.085998612816738, "maximum": 0.12778667434507832, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2558015525674462, "standard_deviation": 0.08096782637163505, "minimum": 0.15047100958663615, "maximum": 0.41169423595858845, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.390313105444784, "standard_deviation": 0.006767217586219628, "minimum": 5.383282614445956, "maximum": 5.400892840096596, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6077215838399289, "standard_deviation": 0.06391407751873324, "minimum": 0.5240456961279236, "maximum": 0.7234395224059734, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09475523053064808, "standard_deviation": 0.052416777251762935, "minimum": -0.15663202089411576, "maximum": -0.027752001396948337, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9901370641259781, "standard_deviation": 0.0013591895442899564, "minimum": 0.9881012644746338, "maximum": 0.9916960791861289, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9244613545029914, "standard_deviation": 0.026995290614082713, "minimum": 0.8995180851464378, "maximum": 0.9823601278236965, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3748549742854288, "standard_deviation": 0.20904423577053263, "minimum": 0.1624860258894352, "maximum": 0.6862823973934565, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10434483111764811, "standard_deviation": 0.006856347017296683, "minimum": 0.09652956266587373, "maximum": 0.11445453869995037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971145110603792, "standard_deviation": 0.0006299282997656223, "minimum": 0.9961762980939153, "maximum": 0.998164536865333, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.559819378785005, "standard_deviation": 0.02475580322839817, "minimum": 0.5188160588084456, "maximum": 0.6064980784704904, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3178362856406529, "standard_deviation": 0.11749203663641702, "minimum": 0.17929091264821673, "maximum": 0.4828184765200962, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05543120942891089, "standard_deviation": 0.005110338165494093, "minimum": 0.046744800065905756, "maximum": 0.06273920760620993, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0002", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.3839011192321777, "predicted_mos": 2.3839011192321777, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f3cd1bc586768afd2eaa5542ed8e9af9bc8e4dc5eeceae90e1dd6def34f1b3b7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.894784899894148, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9381841428165986, "全帯域_10ms_rms_envelope_nrmse": 0.0969949771046853, "全帯域_10ms_envelope_correlation": 0.9916505530085421, "全帯域_10ms_modulation_mae_db": 0.20676230975997598, "全帯域_40ms_log_rms_mae_db": 0.5549796273426548, "全帯域_40ms_rms_envelope_nrmse": 0.04962606144632943, "全帯域_40ms_envelope_correlation": 0.9979415398627747, "全帯域_40ms_modulation_mae_db": 0.39809425828195094, "300–1000 Hz_10ms_log_rms_mae_db": 1.590907531171752, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18451593856183968, "300–1000 Hz_10ms_envelope_correlation": 0.9781860955737609, "300–1000 Hz_10ms_modulation_mae_db": 0.5054598018442483, "300–1000 Hz_40ms_log_rms_mae_db": 1.0449683608802043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12574475365577098, "300–1000 Hz_40ms_envelope_correlation": 0.9912372600549045, "300–1000 Hz_40ms_modulation_mae_db": 0.9823765904623704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.86107221292373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19230520065267112, "1000–3000 Hz_10ms_envelope_correlation": 0.9752579488141314, "1000–3000 Hz_10ms_modulation_mae_db": 0.4536484391707453, "1000–3000 Hz_40ms_log_rms_mae_db": 1.458644340009563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12399130092604131, "1000–3000 Hz_40ms_envelope_correlation": 0.9894557782762385, "1000–3000 Hz_40ms_modulation_mae_db": 0.6504608622443467, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7293317731231523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19266096502560234, "3000–6000 Hz_10ms_envelope_correlation": 0.9765948264881473, "3000–6000 Hz_10ms_modulation_mae_db": 0.7641621337887597, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3938486741472118, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11336693166047426, "3000–6000 Hz_40ms_envelope_correlation": 0.9914601693088766, "3000–6000 Hz_40ms_modulation_mae_db": 0.8428170727951162, "mr_stft_db": 5.275544197573036, "band_mae_db": 0.25172556402398344, "rms_gain_error_db": -0.1693516368059185, "output_peak": 0.5337492787825456}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9640282514742697, "全帯域_10ms_rms_envelope_nrmse": 0.1062567025059177, "全帯域_10ms_envelope_correlation": 0.989782387939923, "全帯域_10ms_modulation_mae_db": 0.21576449419257662, "全帯域_40ms_log_rms_mae_db": 0.5731918478224068, "全帯域_40ms_rms_envelope_nrmse": 0.05390192097127996, "全帯域_40ms_envelope_correlation": 0.997203588859616, "全帯域_40ms_modulation_mae_db": 0.200021579174147, "300–1000 Hz_10ms_log_rms_mae_db": 1.5428309574147934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16368680458126722, "300–1000 Hz_10ms_envelope_correlation": 0.981612941795159, "300–1000 Hz_10ms_modulation_mae_db": 0.6821273209565041, "300–1000 Hz_40ms_log_rms_mae_db": 0.9792551939406009, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09088877706399248, "300–1000 Hz_40ms_envelope_correlation": 0.994148774670902, "300–1000 Hz_40ms_modulation_mae_db": 0.6558115341639161, "1000–3000 Hz_10ms_log_rms_mae_db": 1.922492585083634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2523514657316781, "1000–3000 Hz_10ms_envelope_correlation": 0.957644241286403, "1000–3000 Hz_10ms_modulation_mae_db": 1.8445088539691286, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5205261343064629, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15103688815674113, "1000–3000 Hz_40ms_envelope_correlation": 0.9863161934567765, "1000–3000 Hz_40ms_modulation_mae_db": 2.4395204664203356, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6674406686971364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17612279200925987, "3000–6000 Hz_10ms_envelope_correlation": 0.9809233802092124, "3000–6000 Hz_10ms_modulation_mae_db": 0.7610509555245053, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3375574019603416, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10453807193090062, "3000–6000 Hz_40ms_envelope_correlation": 0.993906356064908, "3000–6000 Hz_40ms_modulation_mae_db": 0.7918494138364119, "mr_stft_db": 5.272371142014592, "band_mae_db": 0.266690179564896, "rms_gain_error_db": -0.04421139783903762, "output_peak": 0.6628537926474489}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.934803120386354, "全帯域_10ms_rms_envelope_nrmse": 0.10864320198800588, "全帯域_10ms_envelope_correlation": 0.9893074907387035, "全帯域_10ms_modulation_mae_db": 0.6364597796191296, "全帯域_40ms_log_rms_mae_db": 0.5743134334646113, "全帯域_40ms_rms_envelope_nrmse": 0.062076741281872315, "全帯域_40ms_envelope_correlation": 0.9962646206604889, "全帯域_40ms_modulation_mae_db": 0.3674331970096623, "300–1000 Hz_10ms_log_rms_mae_db": 1.603251465004395, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19450534188847265, "300–1000 Hz_10ms_envelope_correlation": 0.9739392377390858, "300–1000 Hz_10ms_modulation_mae_db": 1.29783918700676, "300–1000 Hz_40ms_log_rms_mae_db": 1.0586409421268512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1197373921032185, "300–1000 Hz_40ms_envelope_correlation": 0.989654096032441, "300–1000 Hz_40ms_modulation_mae_db": 0.9461833204426675, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8784724439724911, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21193171558774415, "1000–3000 Hz_10ms_envelope_correlation": 0.9706853215880011, "1000–3000 Hz_10ms_modulation_mae_db": 1.6369385999260322, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5255282444811353, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13033233451893866, "1000–3000 Hz_40ms_envelope_correlation": 0.9899312829076494, "1000–3000 Hz_40ms_modulation_mae_db": 2.361557013952204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7165918154168207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17948064405501046, "3000–6000 Hz_10ms_envelope_correlation": 0.9799261666399702, "3000–6000 Hz_10ms_modulation_mae_db": 0.8961792472472806, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3442439642530781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10379594311859693, "3000–6000 Hz_40ms_envelope_correlation": 0.9936941456797349, "3000–6000 Hz_40ms_modulation_mae_db": 1.0212211495620342, "mr_stft_db": 5.279428202633421, "band_mae_db": 0.34539141750963087, "rms_gain_error_db": -0.023657040104504454, "output_peak": 0.7241042422487175}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9275596188693334, "全帯域_10ms_rms_envelope_nrmse": 0.09664486426826209, "全帯域_10ms_envelope_correlation": 0.9915736054847936, "全帯域_10ms_modulation_mae_db": 0.28017970284038296, "全帯域_40ms_log_rms_mae_db": 0.5567277566765785, "全帯域_40ms_rms_envelope_nrmse": 0.05333653048103856, "全帯域_40ms_envelope_correlation": 0.9973342264361728, "全帯域_40ms_modulation_mae_db": 0.34562443978966, "300–1000 Hz_10ms_log_rms_mae_db": 1.4902013578436948, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15137578103631158, "300–1000 Hz_10ms_envelope_correlation": 0.9845442198104299, "300–1000 Hz_10ms_modulation_mae_db": 1.1214809154610836, "300–1000 Hz_40ms_log_rms_mae_db": 0.9731910631577744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0862979720585934, "300–1000 Hz_40ms_envelope_correlation": 0.995300523130293, "300–1000 Hz_40ms_modulation_mae_db": 0.8604503539213648, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8004886193983543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2097225761636797, "1000–3000 Hz_10ms_envelope_correlation": 0.9725130656930395, "1000–3000 Hz_10ms_modulation_mae_db": 2.0298119334256994, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4679243083480522, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13271162795935448, "1000–3000 Hz_40ms_envelope_correlation": 0.991115579940528, "1000–3000 Hz_40ms_modulation_mae_db": 2.8877438331367897, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7377485281507967, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18688015477847889, "3000–6000 Hz_10ms_envelope_correlation": 0.9778617189437881, "3000–6000 Hz_10ms_modulation_mae_db": 0.9682511852552165, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4136672220495758, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11694333271203221, "3000–6000 Hz_40ms_envelope_correlation": 0.9911901207651873, "3000–6000 Hz_40ms_modulation_mae_db": 0.9842222894066329, "mr_stft_db": 5.2811400239529895, "band_mae_db": 0.42658966170814105, "rms_gain_error_db": -0.09445198193973708, "output_peak": 0.520646034368468}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9064731917832386, "全帯域_10ms_rms_envelope_nrmse": 0.1091854606166092, "全帯域_10ms_envelope_correlation": 0.9891595838365548, "全帯域_10ms_modulation_mae_db": 0.6177179393963775, "全帯域_40ms_log_rms_mae_db": 0.541770085565317, "全帯域_40ms_rms_envelope_nrmse": 0.06275612987355084, "全帯域_40ms_envelope_correlation": 0.996176647762251, "全帯域_40ms_modulation_mae_db": 0.2657941376545152, "300–1000 Hz_10ms_log_rms_mae_db": 1.5368241591978218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1815317138474387, "300–1000 Hz_10ms_envelope_correlation": 0.9773876428484737, "300–1000 Hz_10ms_modulation_mae_db": 0.5189550139502929, "300–1000 Hz_40ms_log_rms_mae_db": 1.0244835842862612, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11669666494863785, "300–1000 Hz_40ms_envelope_correlation": 0.9906891619643293, "300–1000 Hz_40ms_modulation_mae_db": 0.4061539135300774, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8310780710249106, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20998685016938898, "1000–3000 Hz_10ms_envelope_correlation": 0.9710412794174165, "1000–3000 Hz_10ms_modulation_mae_db": 0.8439424862437205, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4409283655856833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14064496439099458, "1000–3000 Hz_40ms_envelope_correlation": 0.9862440976211133, "1000–3000 Hz_40ms_modulation_mae_db": 1.0291774064290409, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6901083456925563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.188166213270366, "3000–6000 Hz_10ms_envelope_correlation": 0.9785903170064456, "3000–6000 Hz_10ms_modulation_mae_db": 0.33257195871080675, "3000–6000 Hz_40ms_log_rms_mae_db": 1.312080627766481, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11308740185473486, "3000–6000 Hz_40ms_envelope_correlation": 0.9921064713697032, "3000–6000 Hz_40ms_modulation_mae_db": 0.373405826276396, "mr_stft_db": 5.272329321128447, "band_mae_db": 0.3440878112460089, "rms_gain_error_db": -0.039880718865448694, "output_peak": 0.6054919920663691}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8976580893423931, "全帯域_10ms_rms_envelope_nrmse": 0.09523522761753807, "全帯域_10ms_envelope_correlation": 0.9918414976219609, "全帯域_10ms_modulation_mae_db": 0.2507492222172573, "全帯域_40ms_log_rms_mae_db": 0.5550662028590261, "全帯域_40ms_rms_envelope_nrmse": 0.05077550567888151, "全帯域_40ms_envelope_correlation": 0.9976428742906189, "全帯域_40ms_modulation_mae_db": 0.10914433326897825, "300–1000 Hz_10ms_log_rms_mae_db": 1.587608241190108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17214460042032015, "300–1000 Hz_10ms_envelope_correlation": 0.9800463918299741, "300–1000 Hz_10ms_modulation_mae_db": 1.1302271789408316, "300–1000 Hz_40ms_log_rms_mae_db": 1.060821778462092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09762203506270913, "300–1000 Hz_40ms_envelope_correlation": 0.9941474067696927, "300–1000 Hz_40ms_modulation_mae_db": 0.5083214518798231, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8928830607149814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24786126172272388, "1000–3000 Hz_10ms_envelope_correlation": 0.9585283773103457, "1000–3000 Hz_10ms_modulation_mae_db": 0.8535148146454002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5339562085936007, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13404516370864702, "1000–3000 Hz_40ms_envelope_correlation": 0.9873457149861944, "1000–3000 Hz_40ms_modulation_mae_db": 0.8186165349528204, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6937323974982603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18130751651560764, "3000–6000 Hz_10ms_envelope_correlation": 0.9791277593667688, "3000–6000 Hz_10ms_modulation_mae_db": 0.5684448364867274, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3291690406361327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11188596023984533, "3000–6000 Hz_40ms_envelope_correlation": 0.9917612041976546, "3000–6000 Hz_40ms_modulation_mae_db": 0.5911330336971599, "mr_stft_db": 5.279293849255242, "band_mae_db": 0.13535725365901363, "rms_gain_error_db": -0.12012505772939665, "output_peak": 0.54938646238832}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9651426774229728, "全帯域_10ms_rms_envelope_nrmse": 0.10002338018199834, "全帯域_10ms_envelope_correlation": 0.9910414263186376, "全帯域_10ms_modulation_mae_db": 0.25518006100664486, "全帯域_40ms_log_rms_mae_db": 0.5921658442173408, "全帯域_40ms_rms_envelope_nrmse": 0.05498289592040735, "全帯域_40ms_envelope_correlation": 0.9972992509846208, "全帯域_40ms_modulation_mae_db": 0.44297225437984644, "300–1000 Hz_10ms_log_rms_mae_db": 1.4696008668004366, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17714561061622516, "300–1000 Hz_10ms_envelope_correlation": 0.9804594233488386, "300–1000 Hz_10ms_modulation_mae_db": 0.6886763896176321, "300–1000 Hz_40ms_log_rms_mae_db": 0.9893948131098106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12419080304792289, "300–1000 Hz_40ms_envelope_correlation": 0.9918960846635501, "300–1000 Hz_40ms_modulation_mae_db": 0.552332592429191, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8832283404151249, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21248472868917198, "1000–3000 Hz_10ms_envelope_correlation": 0.9699087661422929, "1000–3000 Hz_10ms_modulation_mae_db": 1.4195498259081987, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5047561421493902, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13419499981643215, "1000–3000 Hz_40ms_envelope_correlation": 0.9873382157805521, "1000–3000 Hz_40ms_modulation_mae_db": 1.4352327580391597, "3000–6000 Hz_10ms_log_rms_mae_db": 1.722478368299069, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18092197778002925, "3000–6000 Hz_10ms_envelope_correlation": 0.9803282665880307, "3000–6000 Hz_10ms_modulation_mae_db": 1.3293370931963429, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4081110782605835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1178456064107386, "3000–6000 Hz_40ms_envelope_correlation": 0.9925628553894962, "3000–6000 Hz_40ms_modulation_mae_db": 1.7737115954187612, "mr_stft_db": 5.290059235882751, "band_mae_db": 0.2177938816647582, "rms_gain_error_db": -0.1493121888681243, "output_peak": 0.5692036908099081}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9542037220780685, "全帯域_10ms_rms_envelope_nrmse": 0.11661524374128099, "全帯域_10ms_envelope_correlation": 0.9876293368208381, "全帯域_10ms_modulation_mae_db": 0.5853867596252805, "全帯域_40ms_log_rms_mae_db": 0.5821994795163049, "全帯域_40ms_rms_envelope_nrmse": 0.059921549029587545, "全帯域_40ms_envelope_correlation": 0.9965780570803975, "全帯域_40ms_modulation_mae_db": 0.44909990260751337, "300–1000 Hz_10ms_log_rms_mae_db": 1.5227562998927584, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1934455800610443, "300–1000 Hz_10ms_envelope_correlation": 0.974802995261509, "300–1000 Hz_10ms_modulation_mae_db": 0.5612834857120588, "300–1000 Hz_40ms_log_rms_mae_db": 0.9793946678700535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11408606155556657, "300–1000 Hz_40ms_envelope_correlation": 0.9921476380285076, "300–1000 Hz_40ms_modulation_mae_db": 1.0157178105528988, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8664006990072552, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25494649152335264, "1000–3000 Hz_10ms_envelope_correlation": 0.9587385818744285, "1000–3000 Hz_10ms_modulation_mae_db": 2.8815699596595823, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4778485023151775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15857139399312714, "1000–3000 Hz_40ms_envelope_correlation": 0.9876925041055302, "1000–3000 Hz_40ms_modulation_mae_db": 3.5901454592197233, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6650460890432734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1744728818000889, "3000–6000 Hz_10ms_envelope_correlation": 0.9807082448001828, "3000–6000 Hz_10ms_modulation_mae_db": 0.6886367194114887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.31509796532919, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1059540119526683, "3000–6000 Hz_40ms_envelope_correlation": 0.9925315106731019, "3000–6000 Hz_40ms_modulation_mae_db": 0.6061256676022577, "mr_stft_db": 5.268927173466037, "band_mae_db": 0.3389110901873343, "rms_gain_error_db": -0.0850042230203841, "output_peak": 0.5755441238037785}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.605564065543419, "spectral_convergence": 0.3072566275794316, "spectral_convergence_status": "ok", "amplitude_floor": 1.8465213976940698e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.088702722609624, "spectral_convergence": 0.3310156224325588, "spectral_convergence_status": "ok", "amplitude_floor": 1.4089022879434144e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.132365804566064, "spectral_convergence": 0.3525676165981625, "spectral_convergence_status": "ok", "amplitude_floor": 1.0260349239920231e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002262162256014784, "output_power": 0.0022103042362978656, "delta_db": -0.10071695882889273, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014828754084214857, "output_power": 0.001374095228952867, "delta_db": -0.33087831338739826, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026062691889949753, "output_power": 0.0002722220265540912, "delta_db": 0.18903992883460086, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020447377727008734, "output_power": 0.00019596287250531492, "delta_db": -0.18463822938942936, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.078753769921038e-05, "output_power": 3.4789387335677714e-05, "delta_db": -0.6908070760759115, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.800369964531977e-05, "output_power": 2.791181790902352e-05, "delta_db": -0.014272877627667798, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25172556402398344, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1693516368059185, "pitch": null, "output_rms_dbfs": -23.856000226683094, "output_peak": 0.5337492787825456, "reference_rms_dbfs": -23.686648589877173, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.275544197573036}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022610981418459577, "log_rms_mae_db": 0.9381841428165986, "rms_envelope_nrmse": 0.0969949771046853, "envelope_correlation": 0.9916505530085421, "rms_gain_error_db": -0.16919338153486366, "reference_mean_power": 0.00428070632223443, "output_mean_power": 0.004117144368981999}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.20587323032393e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001500467490243704, "output_power": 0.0014190773220348137, "reference_active": true, "delta_db": -0.24220530516000524}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013253274491379155, "output_power": 0.00012750642655087998, "reference_active": true, "delta_db": -0.16791118290465423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.272532196136399e-05, "output_power": 3.1179337557270887e-05, "reference_active": true, "delta_db": -0.2101704412152684}], "active_band_count": 3, "active_band_mae_db": 0.20676230975997598}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019856877416737668, "log_rms_mae_db": 0.5549796273426548, "rms_envelope_nrmse": 0.04962606144632943, "envelope_correlation": 0.9979415398627747, "rms_gain_error_db": -0.1693419486455932, "reference_mean_power": 0.004296662851053013, "output_mean_power": 0.0041323498478120484}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.009968768189794e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014002280466658247, "output_power": 0.0013282506392267742, "reference_active": true, "delta_db": -0.229187387638755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3360463722840526e-05, "output_power": 5.1596626527048624e-05, "reference_active": true, "delta_db": -0.14598287447603153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498078024970164e-06, "output_power": 2.0686868211710415e-06, "reference_active": true, "delta_db": -0.8191125127310663}], "active_band_count": 3, "active_band_mae_db": 0.39809425828195094}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000212457290437723, "log_rms_mae_db": 1.590907531171752, "rms_envelope_nrmse": 0.18451593856183968, "envelope_correlation": 0.9781860955737609, "rms_gain_error_db": -0.5870055232379974, "reference_mean_power": 0.0012391682631200216, "output_mean_power": 0.0010825045480094902}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.921692896601052e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00036532110371103373, "output_power": 0.0003212538133711654, "reference_active": true, "delta_db": -0.5582646950685287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3180096390035205e-05, "output_power": 3.5914465233033e-05, "reference_active": true, "delta_db": -0.8001420330025842}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1750484667690236e-05, "output_power": 1.1330746216007707e-05, "reference_active": true, "delta_db": -0.15797267746163193}], "active_band_count": 3, "active_band_mae_db": 0.5054598018442483}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018128574308840577, "log_rms_mae_db": 1.0449683608802043, "rms_envelope_nrmse": 0.12574475365577098, "envelope_correlation": 0.9912372600549045, "rms_gain_error_db": -0.5875473834446464, "reference_mean_power": 0.0012444705639415084, "output_mean_power": 0.001087000866157091}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.573537659747316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003540640123239247, "output_power": 0.0003130584651948427, "reference_active": true, "delta_db": -0.5345633482909148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0107771514146375e-05, "output_power": 1.668970723113946e-05, "reference_active": true, "delta_db": -0.8091522326508395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973740035567712e-07, "output_power": 5.512145253729772e-07, "reference_active": true, "delta_db": -1.6034141904453567}], "active_band_count": 3, "active_band_mae_db": 0.9823765904623704}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001386433646872303, "log_rms_mae_db": 1.86107221292373, "rms_envelope_nrmse": 0.19230520065267112, "envelope_correlation": 0.9752579488141314, "rms_gain_error_db": -0.2650362765219243, "reference_mean_power": 0.00021050674147203463, "output_mean_power": 0.00019804431537220443}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9020912968770603e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.015318920526352e-05, "output_power": 8.253846730104222e-05, "reference_active": true, "delta_db": 0.1273559285720723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8603151739641485e-05, "output_power": 4.386858438971344e-05, "reference_active": true, "delta_db": 0.5553085630744866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0682238748218467e-05, "output_power": 2.4178292641039603e-05, "reference_active": true, "delta_db": 0.678280825865677}], "active_band_count": 3, "active_band_mae_db": 0.4536484391707453}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.950462169412164e-05, "log_rms_mae_db": 1.458644340009563, "rms_envelope_nrmse": 0.12399130092604131, "envelope_correlation": 0.9894557782762385, "rms_gain_error_db": -0.26890275277364123, "reference_mean_power": 0.0002110528457073582, "output_mean_power": 0.00019838139370159815}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7773203833688661e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04144461796749e-05, "output_power": 9.699605125941359e-05, "reference_active": true, "delta_db": 0.30516227921042577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4052463298203573e-05, "output_power": 2.7912811000394546e-05, "reference_active": true, "delta_db": 0.6464401432294733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4632154850806674e-06, "output_power": 1.8419859151762088e-06, "reference_active": true, "delta_db": 0.999780164293141}], "active_band_count": 3, "active_band_mae_db": 0.6504608622443467}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.6125957709065377e-05, "log_rms_mae_db": 1.7293317731231523, "rms_envelope_nrmse": 0.19266096502560234, "envelope_correlation": 0.9765948264881473, "rms_gain_error_db": -0.009263395438868427, "reference_mean_power": 2.4404568133397883e-05, "output_mean_power": 2.4352569260314302e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.103592290864487e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1042852972405107e-05, "output_power": 1.2003133716049186e-05, "reference_active": true, "delta_db": 0.3621335431924564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8700267025557585e-06, "output_power": 4.544606438824698e-06, "reference_active": true, "delta_db": 0.6978231786010711}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.131168166122288e-06, "output_power": 8.516756027431936e-07, "reference_active": true, "delta_db": -1.2325296795727516}], "active_band_count": 3, "active_band_mae_db": 0.7641621337887597}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.642002234154929e-05, "log_rms_mae_db": 1.3938486741472118, "rms_envelope_nrmse": 0.11336693166047426, "envelope_correlation": 0.9914601693088766, "rms_gain_error_db": -0.010293245084783006, "reference_mean_power": 2.4522755214627932e-05, "output_mean_power": 2.4464702477019173e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9657184130662428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1868339578364968e-05, "output_power": 1.2982396579737185e-05, "reference_active": true, "delta_db": 0.38964907698951035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9915566770914906e-06, "output_power": 2.2538938916772538e-06, "reference_active": true, "delta_db": 0.537407961982067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.546174499350975e-08, "output_power": 5.910610462875056e-08, "reference_active": true, "delta_db": -1.6013941794137712}], "active_band_count": 3, "active_band_mae_db": 0.8428170727951162}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8688049338159565, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "435fa70d8ed0a80680d8a9887fed6159604c5b5b2ff88472e8281ef56471e9bb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9667896977657573, "standard_deviation": 0.007212666361244224, "minimum": 0.957644241286403, "maximum": 0.9752579488141314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8670145040675603, "standard_deviation": 0.037595555785523635, "minimum": 1.8004886193983543, "maximum": 1.922492585083634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.4954356141185636, "standard_deviation": 0.7816088239284174, "minimum": 0.4536484391707453, "maximum": 2.8815699596595823, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22394878628005133, "standard_deviation": 0.023945915972220427, "minimum": 0.19230520065267112, "maximum": 0.25494649152335264, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9881799208843227, "standard_deviation": 0.0017798879910401386, "minimum": 0.9862440976211133, "maximum": 0.991115579940528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4912640307236331, "standard_deviation": 0.034553082605353634, "minimum": 1.4409283655856833, "maximum": 1.5339562085936007, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.9015567917993026, "standard_deviation": 1.0718079658198485, "minimum": 0.6504608622443467, "maximum": 3.5901454592197233, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13819108418378456, "standard_deviation": 0.01142033783552057, "minimum": 0.12399130092604131, "maximum": 0.15857139399312714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792575850053182, "standard_deviation": 0.0015106041467264082, "minimum": 0.9765948264881473, "maximum": 0.9809233802092124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7028097482401332, "standard_deviation": 0.027830419644309944, "minimum": 1.6650460890432734, "maximum": 1.7377485281507967, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.788579266202641, "standard_deviation": 0.29374992532699495, "minimum": 0.33257195871080675, "maximum": 1.3293370931963429, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18250164315430542, "standard_deviation": 0.0062412110434581294, "minimum": 0.1744728818000889, "maximum": 0.19266096502560234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9924016041810828, "standard_deviation": 0.0009877121187891094, "minimum": 0.9911901207651873, "maximum": 0.993906356064908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3567219968003243, "standard_deviation": 0.04186636350679127, "minimum": 1.312080627766481, "maximum": 1.4136672220495758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8730607560743463, "standard_deviation": 0.42314855827022274, "minimum": 0.373405826276396, "maximum": 1.7737115954187612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11092715748499889, "standard_deviation": 0.0055033050515165965, "minimum": 0.10379594311859693, "maximum": 0.1178456064107386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9788723685259039, "standard_deviation": 0.003528428861088258, "minimum": 0.9739392377390858, "maximum": 0.9845442198104299, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.54299760981447, "standard_deviation": 0.048606836764209266, "minimum": 1.4696008668004366, "maximum": 1.603251465004395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8132561616861764, "standard_deviation": 0.3179961184973846, "minimum": 0.5054598018442483, "maximum": 1.29783918700676, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17729392137661493, "standard_deviation": 0.014690456669543906, "minimum": 0.15137578103631158, "maximum": 0.19450534188847265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9924026181643275, "standard_deviation": 0.0019515985994521421, "minimum": 0.989654096032441, "maximum": 0.995300523130293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.013768800479206, "standard_deviation": 0.03766226806328575, "minimum": 0.9731910631577744, "maximum": 1.060821778462092, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7409184459227887, "standard_deviation": 0.23881096423143275, "minimum": 0.4061539135300774, "maximum": 1.0157178105528988, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10940805743705147, "standard_deviation": 0.015506081769601741, "minimum": 0.0862979720585934, "maximum": 0.12574475365577098, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2908183574454708, "standard_deviation": 0.09116214433291314, "minimum": 0.13535725365901363, "maximum": 0.42658966170814105, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.277386643238314, "standard_deviation": 0.0066351538358909775, "minimum": 5.268927173466037, "maximum": 5.290059235882751, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5926224521394444, "standard_deviation": 0.06942790826748114, "minimum": 0.520646034368468, "maximum": 0.7241042422487175, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09074928064656893, "standard_deviation": 0.05314716822347918, "minimum": -0.1693516368059185, "maximum": -0.023657040104504454, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9902482352212443, "standard_deviation": 0.0015135834134631704, "minimum": 0.9876293368208381, "maximum": 0.9918414976219609, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9360066017716535, "standard_deviation": 0.02503035650548546, "minimum": 0.8976580893423931, "maximum": 0.9651426774229728, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3810250335822032, "standard_deviation": 0.19408805950321686, "minimum": 0.20676230975997598, "maximum": 0.6364597796191296, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1036998822530372, "standard_deviation": 0.007632336747148346, "minimum": 0.09523522761753807, "maximum": 0.11661524374128099, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970551007421176, "standard_deviation": 0.0006453050581434881, "minimum": 0.996176647762251, "maximum": 0.9979415398627747, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.56630178468303, "standard_deviation": 0.01682362956636857, "minimum": 0.541770085565317, "maximum": 0.5921658442173408, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.32227301277078424, "standard_deviation": 0.12103486507310207, "minimum": 0.10914433326897825, "maximum": 0.44909990260751337, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.055922166835368436, "standard_deviation": 0.0050477024200507965, "minimum": 0.04962606144632943, "maximum": 0.06275612987355084, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0002", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.2652428150177, "predicted_mos": 2.2652428150177, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c70b3f61210caae0eccfd12b455f3274fcfcb2fdcf7065888c25fcdf1ff1b4c7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.997947699856013, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9551604939024909, "全帯域_10ms_rms_envelope_nrmse": 0.1056390063255905, "全帯域_10ms_envelope_correlation": 0.9899003420173895, "全帯域_10ms_modulation_mae_db": 0.8025872762665348, "全帯域_40ms_log_rms_mae_db": 0.5703359342734695, "全帯域_40ms_rms_envelope_nrmse": 0.04699341837151452, "全帯域_40ms_envelope_correlation": 0.9979665489247516, "全帯域_40ms_modulation_mae_db": 0.4325142234440334, "300–1000 Hz_10ms_log_rms_mae_db": 1.7768055064147965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1751271191298522, "300–1000 Hz_10ms_envelope_correlation": 0.9793254688223036, "300–1000 Hz_10ms_modulation_mae_db": 1.2331178988270646, "300–1000 Hz_40ms_log_rms_mae_db": 1.1488417108256566, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09699593273515407, "300–1000 Hz_40ms_envelope_correlation": 0.9943102431178281, "300–1000 Hz_40ms_modulation_mae_db": 1.3795985966433861, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2990029101129053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2917850681622588, "1000–3000 Hz_10ms_envelope_correlation": 0.9486240890956549, "1000–3000 Hz_10ms_modulation_mae_db": 1.3133600725747037, "1000–3000 Hz_40ms_log_rms_mae_db": 1.823005403533963, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.20033266863652482, "1000–3000 Hz_40ms_envelope_correlation": 0.9755435595606322, "1000–3000 Hz_40ms_modulation_mae_db": 1.2387550722258633, "3000–6000 Hz_10ms_log_rms_mae_db": 2.323479384494787, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24781611263668657, "3000–6000 Hz_10ms_envelope_correlation": 0.9612502449225196, "3000–6000 Hz_10ms_modulation_mae_db": 1.302705423785838, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8565098571597394, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1664234921769464, "3000–6000 Hz_40ms_envelope_correlation": 0.9820022531471533, "3000–6000 Hz_40ms_modulation_mae_db": 1.7864079231996115, "mr_stft_db": 5.770168880560395, "band_mae_db": 0.35683000377547397, "rms_gain_error_db": -0.10378776515306505, "output_peak": 0.5775682551440748}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9122801002667695, "全帯域_10ms_rms_envelope_nrmse": 0.09739286720695971, "全帯域_10ms_envelope_correlation": 0.991505120938802, "全帯域_10ms_modulation_mae_db": 0.19047003341081514, "全帯域_40ms_log_rms_mae_db": 0.5302341260733604, "全帯域_40ms_rms_envelope_nrmse": 0.054185028949895825, "全帯域_40ms_envelope_correlation": 0.9971952657896618, "全帯域_40ms_modulation_mae_db": 0.07946489248473393, "300–1000 Hz_10ms_log_rms_mae_db": 1.6185499984368836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15957993722519018, "300–1000 Hz_10ms_envelope_correlation": 0.9824398171599363, "300–1000 Hz_10ms_modulation_mae_db": 1.3719109160517762, "300–1000 Hz_40ms_log_rms_mae_db": 1.0468243547883664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08565640492364684, "300–1000 Hz_40ms_envelope_correlation": 0.9948509345280184, "300–1000 Hz_40ms_modulation_mae_db": 1.1099665342454768, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2518437795289343, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2907507267509928, "1000–3000 Hz_10ms_envelope_correlation": 0.9456421726976854, "1000–3000 Hz_10ms_modulation_mae_db": 2.2781897855497415, "1000–3000 Hz_40ms_log_rms_mae_db": 1.705814632175455, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16217050520710058, "1000–3000 Hz_40ms_envelope_correlation": 0.9824107480977534, "1000–3000 Hz_40ms_modulation_mae_db": 2.1889456109531857, "3000–6000 Hz_10ms_log_rms_mae_db": 2.367282441914832, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22878370204626536, "3000–6000 Hz_10ms_envelope_correlation": 0.9672444116122857, "3000–6000 Hz_10ms_modulation_mae_db": 0.6369326644358101, "3000–6000 Hz_40ms_log_rms_mae_db": 1.868468867168327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1453595378659774, "3000–6000 Hz_40ms_envelope_correlation": 0.9867861572544048, "3000–6000 Hz_40ms_modulation_mae_db": 0.7258899793918389, "mr_stft_db": 5.783602093504033, "band_mae_db": 0.28102315966669483, "rms_gain_error_db": 0.04501827662249194, "output_peak": 0.7190469290555576}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8770236213688958, "全帯域_10ms_rms_envelope_nrmse": 0.11203630465677197, "全帯域_10ms_envelope_correlation": 0.9885896254317575, "全帯域_10ms_modulation_mae_db": 0.5572113739660031, "全帯域_40ms_log_rms_mae_db": 0.525163647312226, "全帯域_40ms_rms_envelope_nrmse": 0.052665608507275974, "全帯域_40ms_envelope_correlation": 0.9973557979897242, "全帯域_40ms_modulation_mae_db": 0.13271054696056522, "300–1000 Hz_10ms_log_rms_mae_db": 1.6351967387941841, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2000985575918323, "300–1000 Hz_10ms_envelope_correlation": 0.9722957507917913, "300–1000 Hz_10ms_modulation_mae_db": 1.3987904767943176, "300–1000 Hz_40ms_log_rms_mae_db": 1.070743818356181, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10558463042637677, "300–1000 Hz_40ms_envelope_correlation": 0.9922095433658596, "300–1000 Hz_40ms_modulation_mae_db": 0.7713179656178166, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2380755717536545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2798897695258593, "1000–3000 Hz_10ms_envelope_correlation": 0.9465465982595128, "1000–3000 Hz_10ms_modulation_mae_db": 1.817291613029733, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7310211835180231, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1427075826140888, "1000–3000 Hz_40ms_envelope_correlation": 0.9860116098244446, "1000–3000 Hz_40ms_modulation_mae_db": 2.197279664082215, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3507997961487788, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23009012183289, "3000–6000 Hz_10ms_envelope_correlation": 0.9669456152988384, "3000–6000 Hz_10ms_modulation_mae_db": 1.5344519253920013, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8843963518527558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13790397699113568, "3000–6000 Hz_40ms_envelope_correlation": 0.9879007075904409, "3000–6000 Hz_40ms_modulation_mae_db": 1.7090691775963058, "mr_stft_db": 5.7658158622203395, "band_mae_db": 0.31946442928041024, "rms_gain_error_db": -0.07585714596522938, "output_peak": 0.6299544729761697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.932937767988155, "全帯域_10ms_rms_envelope_nrmse": 0.10339824335659135, "全帯域_10ms_envelope_correlation": 0.9902817149289395, "全帯域_10ms_modulation_mae_db": 0.43654136385078823, "全帯域_40ms_log_rms_mae_db": 0.5576032128344073, "全帯域_40ms_rms_envelope_nrmse": 0.05830706881836959, "全帯域_40ms_envelope_correlation": 0.9967019059789719, "全帯域_40ms_modulation_mae_db": 0.7426036299820781, "300–1000 Hz_10ms_log_rms_mae_db": 1.6830507606680132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16446048809253133, "300–1000 Hz_10ms_envelope_correlation": 0.9813791413893624, "300–1000 Hz_10ms_modulation_mae_db": 0.7482719838648193, "300–1000 Hz_40ms_log_rms_mae_db": 1.125970784743243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08504717589888061, "300–1000 Hz_40ms_envelope_correlation": 0.9949122432344416, "300–1000 Hz_40ms_modulation_mae_db": 0.8158717152631715, "1000–3000 Hz_10ms_log_rms_mae_db": 2.261093608772175, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.247066453140399, "1000–3000 Hz_10ms_envelope_correlation": 0.959245355360528, "1000–3000 Hz_10ms_modulation_mae_db": 1.9783522343802513, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7504496767633968, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14490790727293681, "1000–3000 Hz_40ms_envelope_correlation": 0.9871152730389081, "1000–3000 Hz_40ms_modulation_mae_db": 2.5554839663465994, "3000–6000 Hz_10ms_log_rms_mae_db": 2.331967621239408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2512466230558933, "3000–6000 Hz_10ms_envelope_correlation": 0.9615176807309374, "3000–6000 Hz_10ms_modulation_mae_db": 0.9730783697860469, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8571262894188885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.165634100023958, "3000–6000 Hz_40ms_envelope_correlation": 0.9827025745958928, "3000–6000 Hz_40ms_modulation_mae_db": 1.2944032731960462, "mr_stft_db": 5.770198080035395, "band_mae_db": 0.27275282684397767, "rms_gain_error_db": -0.041139115217841735, "output_peak": 0.5410412658832292}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9373393999579441, "全帯域_10ms_rms_envelope_nrmse": 0.09776516148492778, "全帯域_10ms_envelope_correlation": 0.9913918504627846, "全帯域_10ms_modulation_mae_db": 0.5856907183842507, "全帯域_40ms_log_rms_mae_db": 0.5356990967374795, "全帯域_40ms_rms_envelope_nrmse": 0.04447292477466345, "全帯域_40ms_envelope_correlation": 0.9982415098090818, "全帯域_40ms_modulation_mae_db": 0.3253581294576096, "300–1000 Hz_10ms_log_rms_mae_db": 1.733539008218647, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16119870626111577, "300–1000 Hz_10ms_envelope_correlation": 0.9821785435466279, "300–1000 Hz_10ms_modulation_mae_db": 1.4260950620039896, "300–1000 Hz_40ms_log_rms_mae_db": 1.1277395614697827, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09025892585742035, "300–1000 Hz_40ms_envelope_correlation": 0.9944768052025655, "300–1000 Hz_40ms_modulation_mae_db": 0.921572209881127, "1000–3000 Hz_10ms_log_rms_mae_db": 2.339869652920185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2941398264028199, "1000–3000 Hz_10ms_envelope_correlation": 0.9409816520622736, "1000–3000 Hz_10ms_modulation_mae_db": 2.4643533936532926, "1000–3000 Hz_40ms_log_rms_mae_db": 1.8278278361332658, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15644419975825294, "1000–3000 Hz_40ms_envelope_correlation": 0.9845829793307982, "1000–3000 Hz_40ms_modulation_mae_db": 2.83627660805413, "3000–6000 Hz_10ms_log_rms_mae_db": 2.382427108861981, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23571312916227194, "3000–6000 Hz_10ms_envelope_correlation": 0.9653762220789239, "3000–6000 Hz_10ms_modulation_mae_db": 0.2528292978495185, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9172519040637426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14108104655391232, "3000–6000 Hz_40ms_envelope_correlation": 0.9873123759912348, "3000–6000 Hz_40ms_modulation_mae_db": 0.23550003141323828, "mr_stft_db": 5.781834213330005, "band_mae_db": 0.3320063984434181, "rms_gain_error_db": -0.12145098845048435, "output_peak": 0.8108699125338376}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8863065265116058, "全帯域_10ms_rms_envelope_nrmse": 0.0998852151681531, "全帯域_10ms_envelope_correlation": 0.9909560126480209, "全帯域_10ms_modulation_mae_db": 0.31165164955429137, "全帯域_40ms_log_rms_mae_db": 0.5065418091807251, "全帯域_40ms_rms_envelope_nrmse": 0.044260852973426815, "全帯域_40ms_envelope_correlation": 0.9981479846144958, "全帯域_40ms_modulation_mae_db": 0.21316801352073567, "300–1000 Hz_10ms_log_rms_mae_db": 1.7352536682110145, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1759409412397143, "300–1000 Hz_10ms_envelope_correlation": 0.9788863948455141, "300–1000 Hz_10ms_modulation_mae_db": 0.3605474642152107, "300–1000 Hz_40ms_log_rms_mae_db": 1.1816026503216064, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09586026614182368, "300–1000 Hz_40ms_envelope_correlation": 0.9940745890558522, "300–1000 Hz_40ms_modulation_mae_db": 0.4831718906522813, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2450103921395734, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21077978698787614, "1000–3000 Hz_10ms_envelope_correlation": 0.9712179809648495, "1000–3000 Hz_10ms_modulation_mae_db": 0.4870718956688999, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6670663101768162, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11608609170827573, "1000–3000 Hz_40ms_envelope_correlation": 0.9909352792840453, "1000–3000 Hz_40ms_modulation_mae_db": 0.5290790408504351, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3295583123081784, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21444496101710198, "3000–6000 Hz_10ms_envelope_correlation": 0.9713445588910695, "3000–6000 Hz_10ms_modulation_mae_db": 1.4245259235883445, "3000–6000 Hz_40ms_log_rms_mae_db": 1.883467626773895, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13245341993949966, "3000–6000 Hz_40ms_envelope_correlation": 0.9892539180791551, "3000–6000 Hz_40ms_modulation_mae_db": 1.7718094471727646, "mr_stft_db": 5.764629843267237, "band_mae_db": 0.14873644358486818, "rms_gain_error_db": -0.07240529771326662, "output_peak": 0.7024640569023428}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8963646232761859, "全帯域_10ms_rms_envelope_nrmse": 0.09665092354770312, "全帯域_10ms_envelope_correlation": 0.9915220282108356, "全帯域_10ms_modulation_mae_db": 0.37009015292795694, "全帯域_40ms_log_rms_mae_db": 0.5051051734973974, "全帯域_40ms_rms_envelope_nrmse": 0.04859275171880716, "全帯域_40ms_envelope_correlation": 0.997737466985795, "全帯域_40ms_modulation_mae_db": 0.1897355345620492, "300–1000 Hz_10ms_log_rms_mae_db": 1.7215385451442236, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16001690335306168, "300–1000 Hz_10ms_envelope_correlation": 0.9823792049178375, "300–1000 Hz_10ms_modulation_mae_db": 0.7906461864246479, "300–1000 Hz_40ms_log_rms_mae_db": 1.1778519261970448, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09378416444906384, "300–1000 Hz_40ms_envelope_correlation": 0.9936413853040927, "300–1000 Hz_40ms_modulation_mae_db": 0.35942556386880115, "1000–3000 Hz_10ms_log_rms_mae_db": 2.292998198257468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2772657386043011, "1000–3000 Hz_10ms_envelope_correlation": 0.9479385175343815, "1000–3000 Hz_10ms_modulation_mae_db": 2.4026172948883713, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7463724936019458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15164701455508992, "1000–3000 Hz_40ms_envelope_correlation": 0.9839629612106321, "1000–3000 Hz_40ms_modulation_mae_db": 3.126645505892748, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3895161686407023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22579612560075787, "3000–6000 Hz_10ms_envelope_correlation": 0.9692252227070041, "3000–6000 Hz_10ms_modulation_mae_db": 1.0202024671921412, "3000–6000 Hz_40ms_log_rms_mae_db": 1.884137071853502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.146859077117479, "3000–6000 Hz_40ms_envelope_correlation": 0.9867102760467286, "3000–6000 Hz_40ms_modulation_mae_db": 1.2373860761623925, "mr_stft_db": 5.777355664118641, "band_mae_db": 0.2389055293772668, "rms_gain_error_db": -0.060832790877619045, "output_peak": 0.7396149253111691}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8889214831542245, "全帯域_10ms_rms_envelope_nrmse": 0.09413488973274392, "全帯域_10ms_envelope_correlation": 0.9919525836763871, "全帯域_10ms_modulation_mae_db": 0.08167755966874278, "全帯域_40ms_log_rms_mae_db": 0.5013211355971039, "全帯域_40ms_rms_envelope_nrmse": 0.05028342798231709, "全帯域_40ms_envelope_correlation": 0.9975509747373992, "全帯域_40ms_modulation_mae_db": 0.25519064490637106, "300–1000 Hz_10ms_log_rms_mae_db": 1.75609934669956, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1848986023604307, "300–1000 Hz_10ms_envelope_correlation": 0.9765098949416182, "300–1000 Hz_10ms_modulation_mae_db": 0.9408347988201223, "300–1000 Hz_40ms_log_rms_mae_db": 1.2159017010848032, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12569706949905413, "300–1000 Hz_40ms_envelope_correlation": 0.9890531578359446, "300–1000 Hz_40ms_modulation_mae_db": 0.731354601081109, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2689436965996648, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.30580173554509127, "1000–3000 Hz_10ms_envelope_correlation": 0.9404443656255898, "1000–3000 Hz_10ms_modulation_mae_db": 0.6169335421998444, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7846265556943905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16591174216998042, "1000–3000 Hz_40ms_envelope_correlation": 0.9819455317480036, "1000–3000 Hz_40ms_modulation_mae_db": 0.3952996127732145, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3603994394210734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21617515658948164, "3000–6000 Hz_10ms_envelope_correlation": 0.9708438503575303, "3000–6000 Hz_10ms_modulation_mae_db": 0.7361115729487554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8734024620729868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13146177123721886, "3000–6000 Hz_40ms_envelope_correlation": 0.9890938502673254, "3000–6000 Hz_40ms_modulation_mae_db": 0.7549483627179777, "mr_stft_db": 5.78425050598302, "band_mae_db": 0.11018025352313343, "rms_gain_error_db": -0.0271447128899963, "output_peak": 0.615990730038577}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.142657418788388, "spectral_convergence": 0.3281763502000893, "spectral_convergence_status": "ok", "amplitude_floor": 1.8465213976940698e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.657581446993145, "spectral_convergence": 0.35095851454643434, "spectral_convergence_status": "ok", "amplitude_floor": 1.4089022879434144e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.510267775899651, "spectral_convergence": 0.37006488499062, "spectral_convergence_status": "ok", "amplitude_floor": 1.0260349239920231e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002262162256014784, "output_power": 0.0021919437843549608, "delta_db": -0.13694340114387302, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014828754084214857, "output_power": 0.0013654188961412053, "delta_db": -0.35838754232319425, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026062691889949753, "output_power": 0.0003231492838079017, "delta_db": 0.9338392785875466, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020447377727008734, "output_power": 0.00022958986007485328, "delta_db": 0.5031508352628749, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.078753769921038e-05, "output_power": 3.9399761645449394e-05, "delta_db": -0.15033893700633408, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.800369964531977e-05, "output_power": 2.8382288731831772e-05, "delta_db": 0.058320028329021056, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.35683000377547397, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10378776515306505, "pitch": null, "output_rms_dbfs": -23.790436355030238, "output_peak": 0.5775682551440748, "reference_rms_dbfs": -23.686648589877173, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.770168880560395}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022610981418459577, "log_rms_mae_db": 0.9551604939024909, "rms_envelope_nrmse": 0.1056390063255905, "envelope_correlation": 0.9899003420173895, "rms_gain_error_db": -0.10359063288194267, "reference_mean_power": 0.00428070632223443, "output_mean_power": 0.004179808365170201}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.20587323032393e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001500467490243704, "output_power": 0.0014548001997460164, "reference_active": true, "delta_db": -0.13423238245001323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013253274491379155, "output_power": 0.00013785734575824963, "reference_active": true, "delta_db": 0.17106719705934004}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.272532196136399e-05, "output_power": 5.310435678261893e-05, "reference_active": true, "delta_db": 2.102462249290251}], "active_band_count": 3, "active_band_mae_db": 0.8025872762665348}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019856877416737668, "log_rms_mae_db": 0.5703359342734695, "rms_envelope_nrmse": 0.04699341837151452, "envelope_correlation": 0.9979665489247516, "rms_gain_error_db": -0.10360959854584706, "reference_mean_power": 0.004296662851053013, "output_mean_power": 0.004195370470988052}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.009968768189794e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014002280466658247, "output_power": 0.0013784523179885331, "reference_active": true, "delta_db": -0.06807024320453411}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3360463722840526e-05, "output_power": 5.854972482101878e-05, "reference_active": true, "delta_db": 0.403052631541039}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498078024970164e-06, "output_power": 3.021676580288395e-06, "reference_active": true, "delta_db": 0.8264197955865271}], "active_band_count": 3, "active_band_mae_db": 0.4325142234440334}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000212457290437723, "log_rms_mae_db": 1.7768055064147965, "rms_envelope_nrmse": 0.1751271191298522, "envelope_correlation": 0.9793254688223036, "rms_gain_error_db": -0.38118584288806995, "reference_mean_power": 0.0012391682631200216, "output_mean_power": 0.0011350413875405576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.921692896601052e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00036532110371103373, "output_power": 0.00028269762796553124, "reference_active": true, "delta_db": -1.1135259660705827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3180096390035205e-05, "output_power": 3.0956946431930134e-05, "reference_active": true, "delta_db": -1.4452549179758365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1750484667690236e-05, "output_power": 1.527963786605783e-05, "reference_active": true, "delta_db": 1.1405728124347743}], "active_band_count": 3, "active_band_mae_db": 1.2331178988270646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018128574308840577, "log_rms_mae_db": 1.1488417108256566, "rms_envelope_nrmse": 0.09699593273515407, "envelope_correlation": 0.9943102431178281, "rms_gain_error_db": -0.38126339200456283, "reference_mean_power": 0.0012444705639415084, "output_mean_power": 0.001139877783647724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.573537659747316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003540640123239247, "output_power": 0.0002795951922266722, "reference_active": true, "delta_db": -1.0255208732602252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0107771514146375e-05, "output_power": 1.445398052008186e-05, "reference_active": true, "delta_db": -1.4337647659393102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973740035567712e-07, "output_power": 5.41640409009742e-07, "reference_active": true, "delta_db": -1.679510150730623}], "active_band_count": 3, "active_band_mae_db": 1.3795985966433861}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001386433646872303, "log_rms_mae_db": 2.2990029101129053, "rms_envelope_nrmse": 0.2917850681622588, "envelope_correlation": 0.9486240890956549, "rms_gain_error_db": 0.4575497010454183, "reference_mean_power": 0.00021050674147203463, "output_mean_power": 0.00023389502198453507}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9020912968770603e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.015318920526352e-05, "output_power": 0.00011541477237003006, "reference_active": true, "delta_db": 1.5834059223566133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8603151739641485e-05, "output_power": 6.425063267407248e-05, "reference_active": true, "delta_db": 2.212546445821584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0682238748218467e-05, "output_power": 2.1380129534385983e-05, "reference_active": true, "delta_db": 0.14412784954591315}], "active_band_count": 3, "active_band_mae_db": 1.3133600725747037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.950462169412164e-05, "log_rms_mae_db": 1.823005403533963, "rms_envelope_nrmse": 0.20033266863652482, "envelope_correlation": 0.9755435595606322, "rms_gain_error_db": 0.4592273009616033, "reference_mean_power": 0.0002110528457073582, "output_mean_power": 0.0002345924021883906}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7773203833688661e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04144461796749e-05, "output_power": 0.00013498950793047282, "reference_active": true, "delta_db": 1.7406218779951455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4052463298203573e-05, "output_power": 3.772163756278209e-05, "reference_active": true, "delta_db": 1.9543097733544559}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4632154850806674e-06, "output_power": 1.4560454516968533e-06, "reference_active": true, "delta_db": -0.021333565327988743}], "active_band_count": 3, "active_band_mae_db": 1.2387550722258633}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.6125957709065377e-05, "log_rms_mae_db": 2.323479384494787, "rms_envelope_nrmse": 0.24781611263668657, "envelope_correlation": 0.9612502449225196, "rms_gain_error_db": -0.07943592618519127, "reference_mean_power": 2.4404568133397883e-05, "output_mean_power": 2.3962246646733648e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.103592290864487e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1042852972405107e-05, "output_power": 1.048310553246785e-05, "reference_active": true, "delta_db": -0.22591332066602404}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8700267025557585e-06, "output_power": 2.4256098590745774e-06, "reference_active": true, "delta_db": -2.0289301241235678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.131168166122288e-06, "output_power": 7.730363070856627e-07, "reference_active": true, "delta_db": -1.653272826567922}], "active_band_count": 3, "active_band_mae_db": 1.302705423785838}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.642002234154929e-05, "log_rms_mae_db": 1.8565098571597394, "rms_envelope_nrmse": 0.1664234921769464, "envelope_correlation": 0.9820022531471533, "rms_gain_error_db": -0.08022007411250211, "reference_mean_power": 2.4522755214627932e-05, "output_mean_power": 2.407394453668321e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9657184130662428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1868339578364968e-05, "output_power": 1.0791810860288037e-05, "reference_active": true, "delta_db": -0.41295638721254363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9915566770914906e-06, "output_power": 1.1696120854730096e-06, "reference_active": true, "delta_db": -2.3115082321860476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.546174499350975e-08, "output_power": 4.6590335855667495e-08, "reference_active": true, "delta_db": -2.634759150200243}], "active_band_count": 3, "active_band_mae_db": 1.7864079231996115}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8470230091281645, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "27fe32a245c2634cd6ab71b20ab7d16b48b8bcf51e7cafd9aa08630b3fe3b005", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9500800914500594, "standard_deviation": 0.010318694725166338, "minimum": 0.9404443656255898, "maximum": 0.9712179809648495, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.27460472626057, "standard_deviation": 0.03409223587576257, "minimum": 2.2380755717536545, "maximum": 2.339869652920185, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6697712289931046, "standard_deviation": 0.7828387567460549, "minimum": 0.4870718956688999, "maximum": 2.4643533936532926, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2746848881399498, "standard_deviation": 0.03111653585833265, "minimum": 0.21077978698787614, "maximum": 0.30580173554509127, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9840634927619022, "standard_deviation": 0.004478783817349028, "minimum": 0.9755435595606322, "maximum": 0.9909352792840453, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.754523011449657, "standard_deviation": 0.05558288334345458, "minimum": 1.6670663101768162, "maximum": 1.8278278361332658, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.883470635147299, "standard_deviation": 1.03932695980642, "minimum": 0.3952996127732145, "maximum": 3.126645505892748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15502596399028126, "standard_deviation": 0.02392279862528377, "minimum": 0.11608609170827573, "maximum": 0.20033266863652482, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9667184758248886, "standard_deviation": 0.003850469437489921, "minimum": 0.9612502449225196, "maximum": 0.9713445588910695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3544287841287175, "standard_deviation": 0.02481351901466914, "minimum": 2.323479384494787, "maximum": 2.3895161686407023, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.985104705622307, "standard_deviation": 0.43368334394013186, "minimum": 0.2528292978495185, "maximum": 1.5344519253920013, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2312582414926686, "standard_deviation": 0.013315152099676274, "minimum": 0.21444496101710198, "maximum": 0.2512466230558933, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9864702641215419, "standard_deviation": 0.0027177526210203497, "minimum": 0.9820022531471533, "maximum": 0.9892539180791551, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8780950537954797, "standard_deviation": 0.019488566179389155, "minimum": 1.8565098571597394, "maximum": 1.9172519040637426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.189426783856272, "standard_deviation": 0.5722784709540076, "minimum": 0.23550003141323828, "maximum": 1.7864079231996115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14589705273826592, "standard_deviation": 0.013566197251368269, "minimum": 0.13146177123721886, "maximum": 0.1664234921769464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.979424277051874, "standard_deviation": 0.0035625261783079295, "minimum": 0.9722957507917913, "maximum": 0.9824398171599363, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7075041965734152, "standard_deviation": 0.05675530942951186, "minimum": 1.6185499984368836, "maximum": 1.7768055064147965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0337768483752434, "standard_deviation": 0.38613868835411896, "minimum": 0.3605474642152107, "maximum": 1.4260950620039896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17266515690671605, "standard_deviation": 0.014387683495830627, "minimum": 0.15957993722519018, "maximum": 0.2000985575918323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9934411127055753, "standard_deviation": 0.001970938195825401, "minimum": 0.9890531578359446, "maximum": 0.9949122432344416, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1369345634733354, "standard_deviation": 0.05701237270461209, "minimum": 1.0468243547883664, "maximum": 1.2159017010848032, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8215348846566461, "standard_deviation": 0.3258738194938821, "minimum": 0.35942556386880115, "maximum": 1.3795985966433861, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09736057124142754, "standard_deviation": 0.01322771671461622, "minimum": 0.08504717589888061, "maximum": 0.12569706949905413, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2574873805619054, "standard_deviation": 0.08776722728114647, "minimum": 0.11018025352313343, "maximum": 0.35683000377547397, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.774731892877384, "standard_deviation": 0.008013694842886778, "minimum": 5.764629843267237, "maximum": 5.78425050598302, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6670688184806197, "standard_deviation": 0.09084993887992812, "minimum": 0.5410412658832292, "maximum": 0.8108699125338376, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05719994245562632, "standard_deviation": 0.05141003946061672, "minimum": -0.12145098845048435, "maximum": 0.04501827662249194, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9907624097893646, "standard_deviation": 0.0011131712590805776, "minimum": 0.9885896254317575, "maximum": 0.9919525836763871, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.910791752053284, "standard_deviation": 0.02826478801557167, "minimum": 0.8770236213688958, "maximum": 0.9551604939024909, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4169900160036729, "standard_deviation": 0.23121757626818648, "minimum": 0.08167755966874278, "maximum": 0.8025872762665348, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10086282643493018, "standard_deviation": 0.0058515440705094925, "minimum": 0.09413488973274392, "maximum": 0.11203630465677197, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976121818537351, "standard_deviation": 0.00052048203326056, "minimum": 0.9967019059789719, "maximum": 0.9982415098090818, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5290005169382712, "standard_deviation": 0.02515487573834087, "minimum": 0.5013211355971039, "maximum": 0.5703359342734695, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.29634320191477204, "standard_deviation": 0.21122820066835965, "minimum": 0.07946489248473393, "maximum": 0.7426036299820781, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0499701352620338, "standard_deviation": 0.004901993392372507, "minimum": 0.044260852973426815, "maximum": 0.05830706881836959, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0002", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.377389430999756, "predicted_mos": 2.377389430999756, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8d196cfd1327f7b7eb724672b263801b9dac1587df6b10b1ebabcb2b7728e2af", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.94409320037812, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9259208103938907, "全帯域_10ms_rms_envelope_nrmse": 0.09197513264042323, "全帯域_10ms_envelope_correlation": 0.9923265163699115, "全帯域_10ms_modulation_mae_db": 0.10838002420010577, "全帯域_40ms_log_rms_mae_db": 0.5495675247629465, "全帯域_40ms_rms_envelope_nrmse": 0.04984167317100106, "全帯域_40ms_envelope_correlation": 0.9976298466111255, "全帯域_40ms_modulation_mae_db": 0.15048663262616602, "300–1000 Hz_10ms_log_rms_mae_db": 1.4912162837624265, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16321655752416428, "300–1000 Hz_10ms_envelope_correlation": 0.9818138408702874, "300–1000 Hz_10ms_modulation_mae_db": 0.46991153517980183, "300–1000 Hz_40ms_log_rms_mae_db": 1.019607778217326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09886084787269694, "300–1000 Hz_40ms_envelope_correlation": 0.9934405350026291, "300–1000 Hz_40ms_modulation_mae_db": 0.6143476487802677, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7235115156011798, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21998901077632182, "1000–3000 Hz_10ms_envelope_correlation": 0.9673419671062348, "1000–3000 Hz_10ms_modulation_mae_db": 1.005688318039451, "1000–3000 Hz_40ms_log_rms_mae_db": 1.417772793625876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1284323521236121, "1000–3000 Hz_40ms_envelope_correlation": 0.9886998851467369, "1000–3000 Hz_40ms_modulation_mae_db": 1.1810256582294545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5921405753615074, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16902313019383636, "3000–6000 Hz_10ms_envelope_correlation": 0.9827321471222324, "3000–6000 Hz_10ms_modulation_mae_db": 0.41407709321466685, "3000–6000 Hz_40ms_log_rms_mae_db": 1.287536844124258, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09966353342395598, "3000–6000 Hz_40ms_envelope_correlation": 0.9938889741459289, "3000–6000 Hz_40ms_modulation_mae_db": 0.42951916897986325, "mr_stft_db": 5.2444549281270305, "band_mae_db": 0.22007708946203333, "rms_gain_error_db": -0.06275094371660686, "output_peak": 0.5841965272327668}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9327867564336979, "全帯域_10ms_rms_envelope_nrmse": 0.09509803288313176, "全帯域_10ms_envelope_correlation": 0.9918150998509915, "全帯域_10ms_modulation_mae_db": 0.20807278743542892, "全帯域_40ms_log_rms_mae_db": 0.530644855247175, "全帯域_40ms_rms_envelope_nrmse": 0.04391237337492157, "全帯域_40ms_envelope_correlation": 0.9981978378837282, "全帯域_40ms_modulation_mae_db": 0.268650049424271, "300–1000 Hz_10ms_log_rms_mae_db": 1.4110567908709524, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1585340760185083, "300–1000 Hz_10ms_envelope_correlation": 0.9847064436200645, "300–1000 Hz_10ms_modulation_mae_db": 0.6745091459151841, "300–1000 Hz_40ms_log_rms_mae_db": 0.9357152672853315, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11181471903881701, "300–1000 Hz_40ms_envelope_correlation": 0.9937125181225368, "300–1000 Hz_40ms_modulation_mae_db": 0.6401697308652182, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7086244259438834, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20830822970455892, "1000–3000 Hz_10ms_envelope_correlation": 0.9707555910486703, "1000–3000 Hz_10ms_modulation_mae_db": 1.6187257615036723, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3722547198978408, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10683189312129372, "1000–3000 Hz_40ms_envelope_correlation": 0.9923184432482253, "1000–3000 Hz_40ms_modulation_mae_db": 2.3434841538326556, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5848300094663303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20342888194625003, "3000–6000 Hz_10ms_envelope_correlation": 0.9766554901548314, "3000–6000 Hz_10ms_modulation_mae_db": 0.7372826106799248, "3000–6000 Hz_40ms_log_rms_mae_db": 1.301297217323299, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13113960918308526, "3000–6000 Hz_40ms_envelope_correlation": 0.9906659313395545, "3000–6000 Hz_40ms_modulation_mae_db": 0.7234588278535447, "mr_stft_db": 5.250327717730097, "band_mae_db": 0.23044458412677934, "rms_gain_error_db": -0.08376892565165452, "output_peak": 0.5813301314809621}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.916407023127772, "全帯域_10ms_rms_envelope_nrmse": 0.10686993672504748, "全帯域_10ms_envelope_correlation": 0.9896503258498097, "全帯域_10ms_modulation_mae_db": 0.6355075794028998, "全帯域_40ms_log_rms_mae_db": 0.5430430777505424, "全帯域_40ms_rms_envelope_nrmse": 0.060810537013638324, "全帯域_40ms_envelope_correlation": 0.9964122219279148, "全帯域_40ms_modulation_mae_db": 0.3352061010027762, "300–1000 Hz_10ms_log_rms_mae_db": 1.4717415502766849, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18786344121783255, "300–1000 Hz_10ms_envelope_correlation": 0.9760555453937465, "300–1000 Hz_10ms_modulation_mae_db": 1.795630203887119, "300–1000 Hz_40ms_log_rms_mae_db": 0.9937888940126595, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10333728410514005, "300–1000 Hz_40ms_envelope_correlation": 0.9933676919491646, "300–1000 Hz_40ms_modulation_mae_db": 1.4517160661104567, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6881797642353784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22554026746816402, "1000–3000 Hz_10ms_envelope_correlation": 0.9666090969152076, "1000–3000 Hz_10ms_modulation_mae_db": 0.8192163554616273, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3905319435512276, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1493808727368064, "1000–3000 Hz_40ms_envelope_correlation": 0.9844714464101408, "1000–3000 Hz_40ms_modulation_mae_db": 1.1892421281004333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5590598168354237, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18106644347190565, "3000–6000 Hz_10ms_envelope_correlation": 0.9793076593634955, "3000–6000 Hz_10ms_modulation_mae_db": 0.8882272650779216, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2453933911750965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10357183150933344, "3000–6000 Hz_40ms_envelope_correlation": 0.9933104405459834, "3000–6000 Hz_40ms_modulation_mae_db": 1.0024988091653972, "mr_stft_db": 5.25567489135959, "band_mae_db": 0.20566787713878054, "rms_gain_error_db": -0.03361328712726614, "output_peak": 0.5953566017313353}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9156086007364989, "全帯域_10ms_rms_envelope_nrmse": 0.10532111511805071, "全帯域_10ms_envelope_correlation": 0.9899695997731099, "全帯域_10ms_modulation_mae_db": 0.5501507888884638, "全帯域_40ms_log_rms_mae_db": 0.5300701387135898, "全帯域_40ms_rms_envelope_nrmse": 0.05307792648918594, "全帯域_40ms_envelope_correlation": 0.9974053130259657, "全帯域_40ms_modulation_mae_db": 0.38212884732996694, "300–1000 Hz_10ms_log_rms_mae_db": 1.5241108546390025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16993980515653467, "300–1000 Hz_10ms_envelope_correlation": 0.9813207698183268, "300–1000 Hz_10ms_modulation_mae_db": 1.3483227840566514, "300–1000 Hz_40ms_log_rms_mae_db": 1.0437924834686438, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10777018895767819, "300–1000 Hz_40ms_envelope_correlation": 0.9935641847033805, "300–1000 Hz_40ms_modulation_mae_db": 0.9399619406422947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6989498866636938, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24647009163172978, "1000–3000 Hz_10ms_envelope_correlation": 0.9591806910579125, "1000–3000 Hz_10ms_modulation_mae_db": 0.5792854600215788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4476237339529319, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14634760223042687, "1000–3000 Hz_40ms_envelope_correlation": 0.9848864478788759, "1000–3000 Hz_40ms_modulation_mae_db": 0.6577778702856297, "3000–6000 Hz_10ms_log_rms_mae_db": 1.567537529949229, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16328988577385148, "3000–6000 Hz_10ms_envelope_correlation": 0.9835685651478806, "3000–6000 Hz_10ms_modulation_mae_db": 0.8029820820839606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.265699165836094, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10708770947225503, "3000–6000 Hz_40ms_envelope_correlation": 0.993207216294817, "3000–6000 Hz_40ms_modulation_mae_db": 0.8688269178756988, "mr_stft_db": 5.2461188075064955, "band_mae_db": 0.20350869335152943, "rms_gain_error_db": -0.11680783851741114, "output_peak": 0.5610109333865472}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0034225743064202, "全帯域_10ms_rms_envelope_nrmse": 0.11833343086622271, "全帯域_10ms_envelope_correlation": 0.9877001712257977, "全帯域_10ms_modulation_mae_db": 0.9595940848522456, "全帯域_40ms_log_rms_mae_db": 0.6140887873857042, "全帯域_40ms_rms_envelope_nrmse": 0.07248930271459487, "全帯域_40ms_envelope_correlation": 0.9955715792274555, "全帯域_40ms_modulation_mae_db": 0.657253901520355, "300–1000 Hz_10ms_log_rms_mae_db": 1.4643767351098047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18132060433356065, "300–1000 Hz_10ms_envelope_correlation": 0.9798310498777322, "300–1000 Hz_10ms_modulation_mae_db": 0.7046904434452137, "300–1000 Hz_40ms_log_rms_mae_db": 1.0204662112060758, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12665962733368324, "300–1000 Hz_40ms_envelope_correlation": 0.992002392014172, "300–1000 Hz_40ms_modulation_mae_db": 0.7103100684061646, "1000–3000 Hz_10ms_log_rms_mae_db": 1.752933230294317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19827549404831304, "1000–3000 Hz_10ms_envelope_correlation": 0.9737245862535837, "1000–3000 Hz_10ms_modulation_mae_db": 0.9187489838404407, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5058849893899577, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12261170037698076, "1000–3000 Hz_40ms_envelope_correlation": 0.9899261884669086, "1000–3000 Hz_40ms_modulation_mae_db": 0.7889822396187602, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5703084307034163, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2044983711619642, "3000–6000 Hz_10ms_envelope_correlation": 0.9744835540150096, "3000–6000 Hz_10ms_modulation_mae_db": 0.4403648290662676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2719948426791703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13516727570858877, "3000–6000 Hz_40ms_envelope_correlation": 0.9884060608660491, "3000–6000 Hz_40ms_modulation_mae_db": 0.5808953583392452, "mr_stft_db": 5.243970307006779, "band_mae_db": 0.24385551471951772, "rms_gain_error_db": -0.2541360986626217, "output_peak": 0.6629068695450551}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9441318032275848, "全帯域_10ms_rms_envelope_nrmse": 0.11043183652555738, "全帯域_10ms_envelope_correlation": 0.9890991160190906, "全帯域_10ms_modulation_mae_db": 0.4842428567908294, "全帯域_40ms_log_rms_mae_db": 0.585616765811027, "全帯域_40ms_rms_envelope_nrmse": 0.06024857526360419, "全帯域_40ms_envelope_correlation": 0.9968005637314794, "全帯域_40ms_modulation_mae_db": 0.7620412359712331, "300–1000 Hz_10ms_log_rms_mae_db": 1.4383011600564968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1844424605736781, "300–1000 Hz_10ms_envelope_correlation": 0.9768665586181631, "300–1000 Hz_10ms_modulation_mae_db": 1.2935807091014497, "300–1000 Hz_40ms_log_rms_mae_db": 0.9464936625404275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10004650415242841, "300–1000 Hz_40ms_envelope_correlation": 0.9937316602325896, "300–1000 Hz_40ms_modulation_mae_db": 0.8821029718344278, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6495171452296764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2077004505862361, "1000–3000 Hz_10ms_envelope_correlation": 0.9710899193351151, "1000–3000 Hz_10ms_modulation_mae_db": 0.906806852705452, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4038710303528514, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11096930350786768, "1000–3000 Hz_40ms_envelope_correlation": 0.9919725144429336, "1000–3000 Hz_40ms_modulation_mae_db": 1.3415359877231097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5804318612703991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18820376877610093, "3000–6000 Hz_10ms_envelope_correlation": 0.9776218779371161, "3000–6000 Hz_10ms_modulation_mae_db": 0.9756677735171819, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2580645450069918, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1116516618326101, "3000–6000 Hz_40ms_envelope_correlation": 0.9921926973420179, "3000–6000 Hz_40ms_modulation_mae_db": 1.1534957180845014, "mr_stft_db": 5.242445199673951, "band_mae_db": 0.43911248283575555, "rms_gain_error_db": -0.17655202607195722, "output_peak": 0.5146071093427387}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9446488155030135, "全帯域_10ms_rms_envelope_nrmse": 0.10919766197453512, "全帯域_10ms_envelope_correlation": 0.9892403720785465, "全帯域_10ms_modulation_mae_db": 0.24637744414476734, "全帯域_40ms_log_rms_mae_db": 0.5451323594301868, "全帯域_40ms_rms_envelope_nrmse": 0.0682971841192043, "全帯域_40ms_envelope_correlation": 0.9956318410607032, "全帯域_40ms_modulation_mae_db": 0.5358445691018267, "300–1000 Hz_10ms_log_rms_mae_db": 1.4870842729826041, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18294984760983374, "300–1000 Hz_10ms_envelope_correlation": 0.9783208723215693, "300–1000 Hz_10ms_modulation_mae_db": 0.6831485745296488, "300–1000 Hz_40ms_log_rms_mae_db": 0.9433798888093305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11520136237658062, "300–1000 Hz_40ms_envelope_correlation": 0.9928251386842157, "300–1000 Hz_40ms_modulation_mae_db": 0.48791140358878754, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7329634157482585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20567801368997787, "1000–3000 Hz_10ms_envelope_correlation": 0.9718606286491619, "1000–3000 Hz_10ms_modulation_mae_db": 0.4823050708087128, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4223369048722243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12616307838769458, "1000–3000 Hz_40ms_envelope_correlation": 0.9887550441112419, "1000–3000 Hz_40ms_modulation_mae_db": 0.6224792486667489, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5305848991584514, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15576101484928667, "3000–6000 Hz_10ms_envelope_correlation": 0.9846623804248708, "3000–6000 Hz_10ms_modulation_mae_db": 0.30988126008041866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2750532979554943, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09914941197498249, "3000–6000 Hz_40ms_envelope_correlation": 0.9935675825032203, "3000–6000 Hz_40ms_modulation_mae_db": 0.30138208614817447, "mr_stft_db": 5.255141070005997, "band_mae_db": 0.11035144689751399, "rms_gain_error_db": -0.1364952208254439, "output_peak": 0.6087587745253841}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9295285465452396, "全帯域_10ms_rms_envelope_nrmse": 0.10431925224950404, "全帯域_10ms_envelope_correlation": 0.9902114373269432, "全帯域_10ms_modulation_mae_db": 1.0154645031156095, "全帯域_40ms_log_rms_mae_db": 0.5301810079516938, "全帯域_40ms_rms_envelope_nrmse": 0.056778825251907764, "全帯域_40ms_envelope_correlation": 0.9970891637477091, "全帯域_40ms_modulation_mae_db": 0.5562424392746005, "300–1000 Hz_10ms_log_rms_mae_db": 1.5243755648957036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19323358087464046, "300–1000 Hz_10ms_envelope_correlation": 0.9756833272313784, "300–1000 Hz_10ms_modulation_mae_db": 1.2804497773589036, "300–1000 Hz_40ms_log_rms_mae_db": 1.0562804676268276, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12085543565461884, "300–1000 Hz_40ms_envelope_correlation": 0.9921328336953191, "300–1000 Hz_40ms_modulation_mae_db": 0.882215835480761, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6912010821489276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2384066725280935, "1000–3000 Hz_10ms_envelope_correlation": 0.9646710381142615, "1000–3000 Hz_10ms_modulation_mae_db": 3.535194695329261, "1000–3000 Hz_40ms_log_rms_mae_db": 1.425101194803703, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17777562859796278, "1000–3000 Hz_40ms_envelope_correlation": 0.982368436291963, "1000–3000 Hz_40ms_modulation_mae_db": 4.116138072187916, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6469801040045728, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18310350237639386, "3000–6000 Hz_10ms_envelope_correlation": 0.9787126020395865, "3000–6000 Hz_10ms_modulation_mae_db": 0.869249085015324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2985490820873495, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10080418169032714, "3000–6000 Hz_40ms_envelope_correlation": 0.9934389119596138, "3000–6000 Hz_40ms_modulation_mae_db": 0.8913239287719791, "mr_stft_db": 5.25609304814771, "band_mae_db": 0.4003696251533691, "rms_gain_error_db": -0.14111488372781889, "output_peak": 0.8102479293140239}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.504710043808572, "spectral_convergence": 0.3136203530352015, "spectral_convergence_status": "ok", "amplitude_floor": 1.8465213976940698e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.037519005376107, "spectral_convergence": 0.33585254539017423, "spectral_convergence_status": "ok", "amplitude_floor": 1.4089022879434144e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.1911357351964105, "spectral_convergence": 0.3640848252802257, "spectral_convergence_status": "ok", "amplitude_floor": 1.0260349239920231e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002262162256014784, "output_power": 0.0022642880535055855, "delta_db": 0.0040792326200190625, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014828754084214857, "output_power": 0.0014057358990176656, "delta_db": -0.23200927244327682, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026062691889949753, "output_power": 0.00028610256873393825, "delta_db": 0.40502487199066606, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020447377727008734, "output_power": 0.00019456631851673707, "delta_db": -0.2156995835873462, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.078753769921038e-05, "output_power": 3.790135017565014e-05, "delta_db": -0.3187280690931482, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.800369964531977e-05, "output_power": 2.8953932771231076e-05, "delta_db": 0.14492150703774392, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22007708946203333, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06275094371660686, "pitch": null, "output_rms_dbfs": -23.74939953359378, "output_peak": 0.5841965272327668, "reference_rms_dbfs": -23.686648589877173, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.2444549281270305}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022610981418459577, "log_rms_mae_db": 0.9259208103938907, "rms_envelope_nrmse": 0.09197513264042323, "envelope_correlation": 0.9923265163699115, "rms_gain_error_db": -0.06288437603918036, "reference_mean_power": 0.00428070632223443, "output_mean_power": 0.004219169728342415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.20587323032393e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001500467490243704, "output_power": 0.001480758082994861, "reference_active": true, "delta_db": -0.05742478259810632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013253274491379155, "output_power": 0.00012786731803617803, "reference_active": true, "delta_db": -0.15563636700908245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.272532196136399e-05, "output_power": 3.189157997768241e-05, "reference_active": true, "delta_db": -0.11207892299312852}], "active_band_count": 3, "active_band_mae_db": 0.10838002420010577}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019856877416737668, "log_rms_mae_db": 0.5495675247629465, "rms_envelope_nrmse": 0.04984167317100106, "envelope_correlation": 0.9976298466111255, "rms_gain_error_db": -0.06300203678967299, "reference_mean_power": 0.004296662851053013, "output_mean_power": 0.0042347821447702464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.009968768189794e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014002280466658247, "output_power": 0.0013841309812563836, "reference_active": true, "delta_db": -0.050215827042504185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3360463722840526e-05, "output_power": 5.236090866496759e-05, "reference_active": true, "delta_db": -0.08212420471275073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498078024970164e-06, "output_power": 2.321100859477692e-06, "reference_active": true, "delta_db": -0.31911986612324317}], "active_band_count": 3, "active_band_mae_db": 0.15048663262616602}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000212457290437723, "log_rms_mae_db": 1.4912162837624265, "rms_envelope_nrmse": 0.16321655752416428, "envelope_correlation": 0.9818138408702874, "rms_gain_error_db": -0.2726820929712712, "reference_mean_power": 0.0012391682631200216, "output_mean_power": 0.001163756374064829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.921692896601052e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00036532110371103373, "output_power": 0.00037717391481985004, "reference_active": true, "delta_db": 0.13866888418902734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3180096390035205e-05, "output_power": 3.5200595233682346e-05, "reference_active": true, "delta_db": -0.8873360010939058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1750484667690236e-05, "output_power": 1.0756792921605985e-05, "reference_active": true, "delta_db": -0.38372972025647245}], "active_band_count": 3, "active_band_mae_db": 0.46991153517980183}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018128574308840577, "log_rms_mae_db": 1.019607778217326, "rms_envelope_nrmse": 0.09886084787269694, "envelope_correlation": 0.9934405350026291, "rms_gain_error_db": -0.2727854767665196, "reference_mean_power": 0.0012444705639415084, "output_mean_power": 0.0011687081720819503}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.573537659747316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003540640123239247, "output_power": 0.00036436334203256653, "reference_active": true, "delta_db": 0.12452890317857344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0107771514146375e-05, "output_power": 1.5551124935342993e-05, "reference_active": true, "delta_db": -1.116021312295146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973740035567712e-07, "output_power": 6.940852244952949e-07, "reference_active": true, "delta_db": -0.6024927308670839}], "active_band_count": 3, "active_band_mae_db": 0.6143476487802677}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001386433646872303, "log_rms_mae_db": 1.7235115156011798, "rms_envelope_nrmse": 0.21998901077632182, "envelope_correlation": 0.9673419671062348, "rms_gain_error_db": -0.29157261232633286, "reference_mean_power": 0.00021050674147203463, "output_mean_power": 0.00019683791105454794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9020912968770603e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.015318920526352e-05, "output_power": 7.362239565271336e-05, "reference_active": true, "delta_db": -0.3691086195821154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8603151739641485e-05, "output_power": 3.590646798505141e-05, "reference_active": true, "delta_db": -0.31450076986602216}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0682238748218467e-05, "output_power": 1.2085148482923436e-05, "reference_active": true, "delta_db": -2.3334555646702153}], "active_band_count": 3, "active_band_mae_db": 1.005688318039451}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.950462169412164e-05, "log_rms_mae_db": 1.417772793625876, "rms_envelope_nrmse": 0.1284323521236121, "envelope_correlation": 0.9886998851467369, "rms_gain_error_db": -0.29225484605098906, "reference_mean_power": 0.0002110528457073582, "output_mean_power": 0.0001973175560400839}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7773203833688661e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04144461796749e-05, "output_power": 8.140800717440808e-05, "reference_active": true, "delta_db": -0.45570702866165347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4052463298203573e-05, "output_power": 2.055396823470438e-05, "reference_active": true, "delta_db": -0.6826387961979802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4632154850806674e-06, "output_power": 8.410758702391964e-07, "reference_active": true, "delta_db": -2.40473114982873}], "active_band_count": 3, "active_band_mae_db": 1.1810256582294545}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.6125957709065377e-05, "log_rms_mae_db": 1.5921405753615074, "rms_envelope_nrmse": 0.16902313019383636, "envelope_correlation": 0.9827321471222324, "rms_gain_error_db": 0.19436171054703702, "reference_mean_power": 2.4404568133397883e-05, "output_mean_power": 2.552156474594649e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.103592290864487e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1042852972405107e-05, "output_power": 1.2427896745967945e-05, "reference_active": true, "delta_db": 0.5131634648410901}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8700267025557585e-06, "output_power": 4.2286778917073236e-06, "reference_active": true, "delta_db": 0.3849064358892976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.131168166122288e-06, "output_power": 1.224456362944066e-06, "reference_active": true, "delta_db": 0.34416137891361276}], "active_band_count": 3, "active_band_mae_db": 0.41407709321466685}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.642002234154929e-05, "log_rms_mae_db": 1.287536844124258, "rms_envelope_nrmse": 0.09966353342395598, "envelope_correlation": 0.9938889741459289, "rms_gain_error_db": 0.19369240835234222, "reference_mean_power": 2.4522755214627932e-05, "output_mean_power": 2.5641209311563284e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9657184130662428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1868339578364968e-05, "output_power": 1.3361518048273309e-05, "reference_active": true, "delta_db": 0.5146583873195201}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9915566770914906e-06, "output_power": 2.1901412594117447e-06, "reference_active": true, "delta_db": 0.41279456433001654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.546174499350975e-08, "output_power": 9.287144689385862e-08, "reference_active": true, "delta_db": 0.36110455529005303}], "active_band_count": 3, "active_band_mae_db": 0.42951916897986325}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8725207667070045, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7561d518f6e621dbfcb55dc8068179e8b55135c99e80414aeda23decb991c466", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9681541898100184, "standard_deviation": 0.004718593086618963, "minimum": 0.9591806910579125, "maximum": 0.9737245862535837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7057350582331643, "standard_deviation": 0.03166638344876455, "minimum": 1.6495171452296764, "maximum": 1.752933230294317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2332464372137744, "standard_deviation": 0.9905130737346259, "minimum": 0.4823050708087128, "maximum": 3.535194695329261, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21879602880417437, "standard_deviation": 0.017002602804695923, "minimum": 0.19827549404831304, "maximum": 0.24647009163172978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9879248007496282, "standard_deviation": 0.003645234140101792, "minimum": 0.982368436291963, "maximum": 0.9923184432482253, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4231721638058266, "standard_deviation": 0.04053810665842336, "minimum": 1.3722547198978408, "maximum": 1.5058849893899577, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5300831698305886, "standard_deviation": 1.1806025654531842, "minimum": 0.6224792486667489, "maximum": 4.116138072187916, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1335640538853306, "standard_deviation": 0.023282268599075076, "minimum": 0.10683189312129372, "maximum": 0.17777562859796278, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9797180345256279, "standard_deviation": 0.003599804679813677, "minimum": 0.9744835540150096, "maximum": 0.9846623804248708, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5789841533436664, "standard_deviation": 0.03333503300922607, "minimum": 1.5305848991584514, "maximum": 1.6469801040045728, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6797164998419583, "standard_deviation": 0.2535818770451333, "minimum": 0.30988126008041866, "maximum": 0.9756677735171819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18104687481869863, "standard_deviation": 0.01775079058155744, "minimum": 0.15576101484928667, "maximum": 0.2044983711619642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9923347268746481, "standard_deviation": 0.0018938835675854989, "minimum": 0.9884060608660491, "maximum": 0.9938889741459289, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2754485482734692, "standard_deviation": 0.019495662710103962, "minimum": 1.2453933911750965, "maximum": 1.301297217323299, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7439251019023005, "standard_deviation": 0.2912113387296289, "minimum": 0.30138208614817447, "maximum": 1.1534957180845014, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11102940184939228, "standard_deviation": 0.014310181924010724, "minimum": 0.09914941197498249, "maximum": 0.13516727570858877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9793248009689086, "standard_deviation": 0.003171227073045741, "minimum": 0.9756833272313784, "maximum": 0.9847064436200645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4765329015742092, "standard_deviation": 0.03927745948520167, "minimum": 1.4110567908709524, "maximum": 1.5243755648957036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0312803966842465, "standard_deviation": 0.4607309138000632, "minimum": 0.46991153517980183, "maximum": 1.795630203887119, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1776875466635941, "standard_deviation": 0.0123522135307842, "minimum": 0.1585340760185083, "maximum": 0.19323358087464046, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9930971193005009, "standard_deviation": 0.0006962039842016904, "minimum": 0.992002392014172, "maximum": 0.9937316602325896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9949405816458278, "standard_deviation": 0.04769500054228636, "minimum": 0.9357152672853315, "maximum": 1.0562804676268276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8260919582135473, "standard_deviation": 0.29708903938518644, "minimum": 0.48791140358878754, "maximum": 1.4517160661104567, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11056824618645542, "standard_deviation": 0.009968983628998387, "minimum": 0.09886084787269694, "maximum": 0.12665962733368324, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25667341421065987, "standard_deviation": 0.10886579308916779, "minimum": 0.11035144689751399, "maximum": 0.43911248283575555, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.249278246194706, "standard_deviation": 0.005743864823245017, "minimum": 5.242445199673951, "maximum": 5.25609304814771, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6148018595698517, "standard_deviation": 0.08937672256391604, "minimum": 0.5146071093427387, "maximum": 0.8102479293140239, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12565490303759755, "standard_deviation": 0.06940124645333509, "minimum": -0.2541360986626217, "maximum": -0.03361328712726614, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900015798117752, "standard_deviation": 0.0014905888945161733, "minimum": 0.9877001712257977, "maximum": 0.9923265163699115, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9390568662842648, "standard_deviation": 0.028196922125325454, "minimum": 0.9156086007364989, "maximum": 1.0034225743064202, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5259737586037938, "standard_deviation": 0.3371246239891931, "minimum": 0.10838002420010577, "maximum": 1.0154645031156095, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10519329987280907, "standard_deviation": 0.008420773567478309, "minimum": 0.09197513264042323, "maximum": 0.11833343086622271, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968422959020102, "standard_deviation": 0.0009336785303167572, "minimum": 0.9955715792274555, "maximum": 0.9981978378837282, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5535430646316082, "standard_deviation": 0.03050150673416281, "minimum": 0.5300701387135898, "maximum": 0.6140887873857042, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4559817220313994, "standard_deviation": 0.20670012959517955, "minimum": 0.15048663262616602, "maximum": 0.7620412359712331, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05818204967475726, "standard_deviation": 0.00940668122951575, "minimum": 0.04391237337492157, "maximum": 0.07248930271459487, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0002", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.531750202178955, "predicted_mos": 2.531750202178955, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3a1fb88d92649ab8e926e07c2b304e5d0cdba0aa7a63e79b641fdbbc83aad035", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.25325489975512, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9140403907448001, "全帯域_10ms_rms_envelope_nrmse": 0.09689488845114819, "全帯域_10ms_envelope_correlation": 0.9916351361332025, "全帯域_10ms_modulation_mae_db": 0.16609178050461718, "全帯域_40ms_log_rms_mae_db": 0.5217024354801426, "全帯域_40ms_rms_envelope_nrmse": 0.04702635652328298, "全帯域_40ms_envelope_correlation": 0.9981447526856433, "全帯域_40ms_modulation_mae_db": 0.3396496817052457, "300–1000 Hz_10ms_log_rms_mae_db": 1.5944543649980651, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1811911505660728, "300–1000 Hz_10ms_envelope_correlation": 0.9792325947603185, "300–1000 Hz_10ms_modulation_mae_db": 0.6639754624486219, "300–1000 Hz_40ms_log_rms_mae_db": 1.0578235589777993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12758145746618668, "300–1000 Hz_40ms_envelope_correlation": 0.9910683099473807, "300–1000 Hz_40ms_modulation_mae_db": 1.027489353901099, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8784965928109867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2030090100123161, "1000–3000 Hz_10ms_envelope_correlation": 0.9722310614193306, "1000–3000 Hz_10ms_modulation_mae_db": 0.22104210314317008, "1000–3000 Hz_40ms_log_rms_mae_db": 1.51090676942978, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12673876167460754, "1000–3000 Hz_40ms_envelope_correlation": 0.9887924611957538, "1000–3000 Hz_40ms_modulation_mae_db": 0.24804694091629562, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8339759840595642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19378140926940227, "3000–6000 Hz_10ms_envelope_correlation": 0.9764151216490775, "3000–6000 Hz_10ms_modulation_mae_db": 0.5776336829665767, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4727172153716375, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10914815662920661, "3000–6000 Hz_40ms_envelope_correlation": 0.992201757244525, "3000–6000 Hz_40ms_modulation_mae_db": 0.6511666768319007, "mr_stft_db": 5.40646583809955, "band_mae_db": 0.18661391262666627, "rms_gain_error_db": -0.15821189319927448, "output_peak": 0.5376258399908612}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9273180173602329, "全帯域_10ms_rms_envelope_nrmse": 0.10705190425955557, "全帯域_10ms_envelope_correlation": 0.9896205322360961, "全帯域_10ms_modulation_mae_db": 0.14548344367872695, "全帯域_40ms_log_rms_mae_db": 0.5524400773219291, "全帯域_40ms_rms_envelope_nrmse": 0.052290542667215716, "全帯域_40ms_envelope_correlation": 0.9973627660264894, "全帯域_40ms_modulation_mae_db": 0.15841673238317103, "300–1000 Hz_10ms_log_rms_mae_db": 1.5284714082662796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16146488521174737, "300–1000 Hz_10ms_envelope_correlation": 0.9820946673825901, "300–1000 Hz_10ms_modulation_mae_db": 0.7015073167256732, "300–1000 Hz_40ms_log_rms_mae_db": 0.9421533259042749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08592234742077988, "300–1000 Hz_40ms_envelope_correlation": 0.9947242322709876, "300–1000 Hz_40ms_modulation_mae_db": 0.7298155853787026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9012985044163784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24645190773201817, "1000–3000 Hz_10ms_envelope_correlation": 0.960042040694519, "1000–3000 Hz_10ms_modulation_mae_db": 2.3413836232663843, "1000–3000 Hz_40ms_log_rms_mae_db": 1.516825808471921, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15288648465145155, "1000–3000 Hz_40ms_envelope_correlation": 0.9862945982060142, "1000–3000 Hz_40ms_modulation_mae_db": 2.974452654304768, "3000–6000 Hz_10ms_log_rms_mae_db": 1.79216320375785, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18554127775127618, "3000–6000 Hz_10ms_envelope_correlation": 0.9785385238544233, "3000–6000 Hz_10ms_modulation_mae_db": 0.44129796853729425, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4531314150830053, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10977952559110338, "3000–6000 Hz_40ms_envelope_correlation": 0.9928809222711446, "3000–6000 Hz_40ms_modulation_mae_db": 0.3994143275920073, "mr_stft_db": 5.400115157039508, "band_mae_db": 0.22851067625104537, "rms_gain_error_db": -0.04269031061520703, "output_peak": 0.6373085666912401}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9178045131462743, "全帯域_10ms_rms_envelope_nrmse": 0.11124104389518116, "全帯域_10ms_envelope_correlation": 0.988794174040683, "全帯域_10ms_modulation_mae_db": 0.6895719069231645, "全帯域_40ms_log_rms_mae_db": 0.5593099614861025, "全帯域_40ms_rms_envelope_nrmse": 0.06080125822468624, "全帯域_40ms_envelope_correlation": 0.9964195866639232, "全帯域_40ms_modulation_mae_db": 0.3565627419222122, "300–1000 Hz_10ms_log_rms_mae_db": 1.5938791248437525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19170952365577862, "300–1000 Hz_10ms_envelope_correlation": 0.9747295555832146, "300–1000 Hz_10ms_modulation_mae_db": 1.3138380019240958, "300–1000 Hz_40ms_log_rms_mae_db": 1.0552640947141656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11577290081319033, "300–1000 Hz_40ms_envelope_correlation": 0.9903218760826744, "300–1000 Hz_40ms_modulation_mae_db": 0.9451316087576805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8885308751301269, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22050437108308554, "1000–3000 Hz_10ms_envelope_correlation": 0.9677909773579844, "1000–3000 Hz_10ms_modulation_mae_db": 1.78767598790788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5394729697390495, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1353453707161127, "1000–3000 Hz_40ms_envelope_correlation": 0.9885351554167396, "1000–3000 Hz_40ms_modulation_mae_db": 2.272906269196313, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8377216098743012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18518258221474873, "3000–6000 Hz_10ms_envelope_correlation": 0.9782753000639767, "3000–6000 Hz_10ms_modulation_mae_db": 0.7160692622160707, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4191086176125873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1019819191743129, "3000–6000 Hz_40ms_envelope_correlation": 0.9933107592893783, "3000–6000 Hz_40ms_modulation_mae_db": 0.8217031852998081, "mr_stft_db": 5.404063931472059, "band_mae_db": 0.3022856728103411, "rms_gain_error_db": -0.028051190064153984, "output_peak": 0.7229944436571886}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9104996425356358, "全帯域_10ms_rms_envelope_nrmse": 0.0975349626710659, "全帯域_10ms_envelope_correlation": 0.9914071339503653, "全帯域_10ms_modulation_mae_db": 0.29915506967612343, "全帯域_40ms_log_rms_mae_db": 0.5607977590960641, "全帯域_40ms_rms_envelope_nrmse": 0.05372398769007583, "全帯域_40ms_envelope_correlation": 0.9972848999017181, "全帯域_40ms_modulation_mae_db": 0.4367160337359269, "300–1000 Hz_10ms_log_rms_mae_db": 1.5033040975201126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14921071794358765, "300–1000 Hz_10ms_envelope_correlation": 0.9849426251995677, "300–1000 Hz_10ms_modulation_mae_db": 0.6217231351644926, "300–1000 Hz_40ms_log_rms_mae_db": 1.003621177021748, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08771231047793852, "300–1000 Hz_40ms_envelope_correlation": 0.9950287363912922, "300–1000 Hz_40ms_modulation_mae_db": 0.6162032330518256, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8001472532097438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22488581164105026, "1000–3000 Hz_10ms_envelope_correlation": 0.9665542249194833, "1000–3000 Hz_10ms_modulation_mae_db": 1.884059045715649, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4663828255409574, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1307383626568518, "1000–3000 Hz_40ms_envelope_correlation": 0.9897469176021662, "1000–3000 Hz_40ms_modulation_mae_db": 2.780043898751487, "3000–6000 Hz_10ms_log_rms_mae_db": 1.841565079724174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18774551843771511, "3000–6000 Hz_10ms_envelope_correlation": 0.9777840143827932, "3000–6000 Hz_10ms_modulation_mae_db": 1.085647036493591, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4816500621338151, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11548827984618167, "3000–6000 Hz_40ms_envelope_correlation": 0.9915441356019741, "3000–6000 Hz_40ms_modulation_mae_db": 1.1842071583330338, "mr_stft_db": 5.410910103202979, "band_mae_db": 0.40709574786705155, "rms_gain_error_db": -0.09220564493674788, "output_peak": 0.5291382615860986}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8957220532197406, "全帯域_10ms_rms_envelope_nrmse": 0.10904802100495543, "全帯域_10ms_envelope_correlation": 0.9891870366837461, "全帯域_10ms_modulation_mae_db": 0.45214027591050093, "全帯域_40ms_log_rms_mae_db": 0.5415008000684493, "全帯域_40ms_rms_envelope_nrmse": 0.06291211989847934, "全帯域_40ms_envelope_correlation": 0.9961561249361005, "全帯域_40ms_modulation_mae_db": 0.1589954798854207, "300–1000 Hz_10ms_log_rms_mae_db": 1.5447045052557657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1832651243039156, "300–1000 Hz_10ms_envelope_correlation": 0.9768983460259457, "300–1000 Hz_10ms_modulation_mae_db": 0.7739566376478256, "300–1000 Hz_40ms_log_rms_mae_db": 1.0339596013333276, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11292017726724077, "300–1000 Hz_40ms_envelope_correlation": 0.991268428685105, "300–1000 Hz_40ms_modulation_mae_db": 0.6986753635361014, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8511455697378003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22057495852873737, "1000–3000 Hz_10ms_envelope_correlation": 0.9679914874563298, "1000–3000 Hz_10ms_modulation_mae_db": 0.8077133154172245, "1000–3000 Hz_40ms_log_rms_mae_db": 1.486519584195316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14444772154206117, "1000–3000 Hz_40ms_envelope_correlation": 0.9854305459523222, "1000–3000 Hz_40ms_modulation_mae_db": 0.6997077255984897, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7795221825092447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18495526634707818, "3000–6000 Hz_10ms_envelope_correlation": 0.9795172100893609, "3000–6000 Hz_10ms_modulation_mae_db": 0.4539470722616037, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3957795956716805, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10821297650139354, "3000–6000 Hz_40ms_envelope_correlation": 0.9929524094600565, "3000–6000 Hz_40ms_modulation_mae_db": 0.4916037086612857, "mr_stft_db": 5.4041371440320995, "band_mae_db": 0.28932819807270777, "rms_gain_error_db": -0.03964866947866691, "output_peak": 0.5952066363061926}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8969263794093675, "全帯域_10ms_rms_envelope_nrmse": 0.09861288451861784, "全帯域_10ms_envelope_correlation": 0.991306578343058, "全帯域_10ms_modulation_mae_db": 0.39761903883927974, "全帯域_40ms_log_rms_mae_db": 0.564860549447654, "全帯域_40ms_rms_envelope_nrmse": 0.05255936002565914, "全帯域_40ms_envelope_correlation": 0.9975552903285427, "全帯域_40ms_modulation_mae_db": 0.2169011310259925, "300–1000 Hz_10ms_log_rms_mae_db": 1.5337235668117104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1702117671523666, "300–1000 Hz_10ms_envelope_correlation": 0.980782118566251, "300–1000 Hz_10ms_modulation_mae_db": 0.980486539346742, "300–1000 Hz_40ms_log_rms_mae_db": 1.052673848458725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10067935445169643, "300–1000 Hz_40ms_envelope_correlation": 0.9940780845880625, "300–1000 Hz_40ms_modulation_mae_db": 0.49634332090457695, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8825681086152293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.232918331418265, "1000–3000 Hz_10ms_envelope_correlation": 0.9633184869454088, "1000–3000 Hz_10ms_modulation_mae_db": 0.8256975775973756, "1000–3000 Hz_40ms_log_rms_mae_db": 1.512801706651181, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12774986070818434, "1000–3000 Hz_40ms_envelope_correlation": 0.9890276254034285, "1000–3000 Hz_40ms_modulation_mae_db": 1.1104396515271084, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8489609805953866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18630136319011797, "3000–6000 Hz_10ms_envelope_correlation": 0.978219993202246, "3000–6000 Hz_10ms_modulation_mae_db": 0.4536818611770814, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4588373880306973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11347578192067119, "3000–6000 Hz_40ms_envelope_correlation": 0.9916923958731619, "3000–6000 Hz_40ms_modulation_mae_db": 0.4248391382501529, "mr_stft_db": 5.416808007108165, "band_mae_db": 0.15096489002729432, "rms_gain_error_db": -0.1499516174177005, "output_peak": 0.6151619285615781}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9354641950240364, "全帯域_10ms_rms_envelope_nrmse": 0.10160757424443415, "全帯域_10ms_envelope_correlation": 0.9907350421744011, "全帯域_10ms_modulation_mae_db": 0.14461948208900013, "全帯域_40ms_log_rms_mae_db": 0.5676290312538318, "全帯域_40ms_rms_envelope_nrmse": 0.05674640678389044, "全帯域_40ms_envelope_correlation": 0.9970896616784287, "全帯域_40ms_modulation_mae_db": 0.32926362720315455, "300–1000 Hz_10ms_log_rms_mae_db": 1.4961643997998062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18313658319161624, "300–1000 Hz_10ms_envelope_correlation": 0.9784302328141408, "300–1000 Hz_10ms_modulation_mae_db": 0.6410375760208211, "300–1000 Hz_40ms_log_rms_mae_db": 1.0042915597714817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12434921974298113, "300–1000 Hz_40ms_envelope_correlation": 0.9913451569837742, "300–1000 Hz_40ms_modulation_mae_db": 0.5638775406819326, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8356692209108527, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21564029632325554, "1000–3000 Hz_10ms_envelope_correlation": 0.9688631218964655, "1000–3000 Hz_10ms_modulation_mae_db": 1.4329959715609604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4754285679048778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12837444092609626, "1000–3000 Hz_40ms_envelope_correlation": 0.9882959036985783, "1000–3000 Hz_40ms_modulation_mae_db": 1.5117844947693149, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8209333682775972, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1775416503397942, "3000–6000 Hz_10ms_envelope_correlation": 0.9805984299268269, "3000–6000 Hz_10ms_modulation_mae_db": 0.9789290368382391, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4788492626695342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11122176072701924, "3000–6000 Hz_40ms_envelope_correlation": 0.9928555473381903, "3000–6000 Hz_40ms_modulation_mae_db": 1.3752150977968391, "mr_stft_db": 5.417609248193894, "band_mae_db": 0.2077338037231192, "rms_gain_error_db": -0.14472146086889925, "output_peak": 0.5771539339892345}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9744296388357585, "全帯域_10ms_rms_envelope_nrmse": 0.11471154812874633, "全帯域_10ms_envelope_correlation": 0.988046265794441, "全帯域_10ms_modulation_mae_db": 0.642874554318318, "全帯域_40ms_log_rms_mae_db": 0.6022186613316257, "全帯域_40ms_rms_envelope_nrmse": 0.058157628695851976, "全帯域_40ms_envelope_correlation": 0.9968217887378572, "全帯域_40ms_modulation_mae_db": 0.501978095938272, "300–1000 Hz_10ms_log_rms_mae_db": 1.531503938385832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18863714767163078, "300–1000 Hz_10ms_envelope_correlation": 0.9763263010650274, "300–1000 Hz_10ms_modulation_mae_db": 0.7165186402758345, "300–1000 Hz_40ms_log_rms_mae_db": 0.9899282256668589, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11486165261521664, "300–1000 Hz_40ms_envelope_correlation": 0.9922889905901505, "300–1000 Hz_40ms_modulation_mae_db": 0.6498698836578668, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8844756196641168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2670062831759009, "1000–3000 Hz_10ms_envelope_correlation": 0.952039470763282, "1000–3000 Hz_10ms_modulation_mae_db": 2.1535355267737715, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4909216090106752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.146845952041247, "1000–3000 Hz_40ms_envelope_correlation": 0.9871427146800615, "1000–3000 Hz_40ms_modulation_mae_db": 2.5725094207443497, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7765360985091279, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1770744706001191, "3000–6000 Hz_10ms_envelope_correlation": 0.9803981396310849, "3000–6000 Hz_10ms_modulation_mae_db": 0.5138060996843782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4011853513722925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10188703444033143, "3000–6000 Hz_40ms_envelope_correlation": 0.9932371398089266, "3000–6000 Hz_40ms_modulation_mae_db": 0.45706248060265425, "mr_stft_db": 5.401235256702087, "band_mae_db": 0.2534609559746275, "rms_gain_error_db": -0.10094996082619141, "output_peak": 0.6365277271921296}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.735293205364362, "spectral_convergence": 0.3106172394175445, "spectral_convergence_status": "ok", "amplitude_floor": 1.8465213976940698e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.231766097002524, "spectral_convergence": 0.3334575763874893, "spectral_convergence_status": "ok", "amplitude_floor": 1.4089022879434144e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.252338211931764, "spectral_convergence": 0.3524871079721255, "spectral_convergence_status": "ok", "amplitude_floor": 1.0260349239920231e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002262162256014784, "output_power": 0.0022168727087848966, "delta_db": -0.08782994986494208, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014828754084214857, "output_power": 0.0013803493349359088, "delta_db": -0.31115652685471606, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026062691889949753, "output_power": 0.00026783462177684865, "delta_db": 0.11847445719916547, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020447377727008734, "output_power": 0.00019555418975042457, "delta_db": -0.19370494687754275, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.078753769921038e-05, "output_power": 3.7575532370752684e-05, "delta_db": -0.3562234581313364, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.800369964531977e-05, "output_power": 2.7668524338644653e-05, "delta_db": -0.05229413683229481, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18661391262666627, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15821189319927448, "pitch": null, "output_rms_dbfs": -23.84486048307645, "output_peak": 0.5376258399908612, "reference_rms_dbfs": -23.686648589877173, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.40646583809955}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022610981418459577, "log_rms_mae_db": 0.9140403907448001, "rms_envelope_nrmse": 0.09689488845114819, "envelope_correlation": 0.9916351361332025, "rms_gain_error_db": -0.1581288356917787, "reference_mean_power": 0.00428070632223443, "output_mean_power": 0.004127647014846819}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.20587323032393e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001500467490243704, "output_power": 0.0014266932002760919, "reference_active": true, "delta_db": -0.21895998888850413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013253274491379155, "output_power": 0.00012700865767135985, "reference_active": true, "delta_db": -0.18489866747166045}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.272532196136399e-05, "output_power": 3.202154331775025e-05, "reference_active": true, "delta_db": -0.09441668515368698}], "active_band_count": 3, "active_band_mae_db": 0.16609178050461718}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019856877416737668, "log_rms_mae_db": 0.5217024354801426, "rms_envelope_nrmse": 0.04702635652328298, "envelope_correlation": 0.9981447526856433, "rms_gain_error_db": -0.15832570097659968, "reference_mean_power": 0.004296662851053013, "output_mean_power": 0.0041428452090443955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.009968768189794e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014002280466658247, "output_power": 0.0013346573270990605, "reference_active": true, "delta_db": -0.20828997327899862}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3360463722840526e-05, "output_power": 5.152084980613656e-05, "reference_active": true, "delta_db": -0.15236577267000512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498078024970164e-06, "output_power": 2.1467262610976637e-06, "reference_active": true, "delta_db": -0.6582932991667333}], "active_band_count": 3, "active_band_mae_db": 0.3396496817052457}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000212457290437723, "log_rms_mae_db": 1.5944543649980651, "rms_envelope_nrmse": 0.1811911505660728, "envelope_correlation": 0.9792325947603185, "rms_gain_error_db": -0.5997103564825083, "reference_mean_power": 0.0012391682631200216, "output_mean_power": 0.001079342421065778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.921692896601052e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00036532110371103373, "output_power": 0.0003210209689754436, "reference_active": true, "delta_db": -0.5614135977940806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3180096390035205e-05, "output_power": 3.6696422460725435e-05, "reference_active": true, "delta_db": -0.706598805788742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1750484667690236e-05, "output_power": 9.94636475633362e-06, "reference_active": true, "delta_db": -0.7239139837630432}], "active_band_count": 3, "active_band_mae_db": 0.6639754624486219}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018128574308840577, "log_rms_mae_db": 1.0578235589777993, "rms_envelope_nrmse": 0.12758145746618668, "envelope_correlation": 0.9910683099473807, "rms_gain_error_db": -0.6003177752341692, "reference_mean_power": 0.0012444705639415084, "output_mean_power": 0.0010838092442518106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.573537659747316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003540640123239247, "output_power": 0.00031183872193569974, "reference_active": true, "delta_db": -0.5515174473152546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0107771514146375e-05, "output_power": 1.7383216492178005e-05, "reference_active": true, "delta_db": -0.6323380272327452}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973740035567712e-07, "output_power": 5.149923925728396e-07, "reference_active": true, "delta_db": -1.8986125871552972}], "active_band_count": 3, "active_band_mae_db": 1.027489353901099}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001386433646872303, "log_rms_mae_db": 1.8784965928109867, "rms_envelope_nrmse": 0.2030090100123161, "envelope_correlation": 0.9722310614193306, "rms_gain_error_db": -0.22593010251875742, "reference_mean_power": 0.00021050674147203463, "output_mean_power": 0.00019983566424164285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9020912968770603e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.015318920526352e-05, "output_power": 8.458816385730515e-05, "reference_active": true, "delta_db": 0.23388790773224621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8603151739641485e-05, "output_power": 3.821875273499465e-05, "reference_active": true, "delta_db": -0.04346254131915563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0682238748218467e-05, "output_power": 2.260347091959318e-05, "reference_active": true, "delta_db": 0.3857758603781084}], "active_band_count": 3, "active_band_mae_db": 0.22104210314317008}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.950462169412164e-05, "log_rms_mae_db": 1.51090676942978, "rms_envelope_nrmse": 0.12673876167460754, "envelope_correlation": 0.9887924611957538, "rms_gain_error_db": -0.22954483607305784, "reference_mean_power": 0.0002110528457073582, "output_mean_power": 0.000200187395215132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7773203833688661e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04144461796749e-05, "output_power": 9.714904300483927e-05, "reference_active": true, "delta_db": 0.3120070037521876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4052463298203573e-05, "output_power": 2.405691637806103e-05, "reference_active": true, "delta_db": 0.0008039796098416104}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4632154850806674e-06, "output_power": 1.6159996877842262e-06, "reference_active": true, "delta_db": 0.4313298393868576}], "active_band_count": 3, "active_band_mae_db": 0.24804694091629562}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.6125957709065377e-05, "log_rms_mae_db": 1.8339759840595642, "rms_envelope_nrmse": 0.19378140926940227, "envelope_correlation": 0.9764151216490775, "rms_gain_error_db": 0.010418399998661234, "reference_mean_power": 2.4404568133397883e-05, "output_mean_power": 2.446318314658874e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.103592290864487e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1042852972405107e-05, "output_power": 1.1729877453757677e-05, "reference_active": true, "delta_db": 0.26212185020685846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8700267025557585e-06, "output_power": 4.168283253350018e-06, "reference_active": true, "delta_db": 0.32243261941484547}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.131168166122288e-06, "output_power": 8.683454171745747e-07, "reference_active": true, "delta_db": -1.148346579278026}], "active_band_count": 3, "active_band_mae_db": 0.5776336829665767}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.642002234154929e-05, "log_rms_mae_db": 1.4727172153716375, "rms_envelope_nrmse": 0.10914815662920661, "envelope_correlation": 0.992201757244525, "rms_gain_error_db": 0.009326133608630294, "reference_mean_power": 2.4522755214627932e-05, "output_mean_power": 2.4575472492265454e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9657184130662428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1868339578364968e-05, "output_power": 1.2600769464208057e-05, "reference_active": true, "delta_db": 0.2600710214980557}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9915566770914906e-06, "output_power": 2.07588386251061e-06, "reference_active": true, "delta_db": 0.18010382425293306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.546174499350975e-08, "output_power": 6.031693123874747e-08, "reference_active": true, "delta_db": -1.5133251847447133}], "active_band_count": 3, "active_band_mae_db": 0.6511666768319007}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.871711677059177, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "76f1bed00e14e388850b4faf1c82e4dcdd508a3de2a825afa1dfc67d52ede6d0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9648538589316005, "standard_deviation": 0.0063423681244242506, "minimum": 0.952039470763282, "maximum": 0.9722310614193306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8652914680619044, "standard_deviation": 0.03380367415245523, "minimum": 1.8001472532097438, "maximum": 1.9012985044163784, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.4317628939228018, "standard_deviation": 0.7466315731310444, "minimum": 0.22104210314317008, "maximum": 2.3413836232663843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22887387123932862, "standard_deviation": 0.019922598514294867, "minimum": 0.2030090100123161, "maximum": 0.2670062831759009, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9879082402693831, "standard_deviation": 0.0014778241177362324, "minimum": 0.9854305459523222, "maximum": 0.9897469176021662, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4999074801179697, "standard_deviation": 0.024266392076695956, "minimum": 1.4663828255409574, "maximum": 1.5394729697390495, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.771236381976016, "standard_deviation": 1.0233963411700133, "minimum": 0.24804694091629562, "maximum": 2.974452654304768, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13664086936457653, "standard_deviation": 0.010078011300075386, "minimum": 0.12673876167460754, "maximum": 0.15288648465145155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9787183415999736, "standard_deviation": 0.0013963935817543026, "minimum": 0.9764151216490775, "maximum": 0.9805984299268269, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8164223134134057, "standard_deviation": 0.029311379457773307, "minimum": 1.7765360985091279, "maximum": 1.8489609805953866, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6526265025218544, "standard_deviation": 0.2525901178174669, "minimum": 0.44129796853729425, "maximum": 1.085647036493591, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18476544226878147, "standard_deviation": 0.005407319677438802, "minimum": 0.1770744706001191, "maximum": 0.19378140926940227, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9925843833609197, "standard_deviation": 0.0006839779894752369, "minimum": 0.9915441356019741, "maximum": 0.9933107592893783, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.445157363493156, "standard_deviation": 0.03490266894583461, "minimum": 1.3957795956716805, "maximum": 1.4816500621338151, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7256514716709602, "standard_deviation": 0.3724086223715955, "minimum": 0.3994143275920073, "maximum": 1.3752150977968391, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1088994293537775, "standard_deviation": 0.0049001816045512354, "minimum": 0.10188703444033143, "maximum": 0.11548827984618167, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.979179555174632, "standard_deviation": 0.0033378956689309944, "minimum": 0.9747295555832146, "maximum": 0.9849426251995677, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5407756757351656, "standard_deviation": 0.03664148927847062, "minimum": 1.4961643997998062, "maximum": 1.5944543649980651, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8016304136942634, "standard_deviation": 0.23588319976140548, "minimum": 0.6217231351644926, "maximum": 1.3138380019240958, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17610336246208946, "standard_deviation": 0.014636065503130466, "minimum": 0.14921071794358765, "maximum": 0.19170952365577862, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9925154769424284, "standard_deviation": 0.0018329939243095616, "minimum": 0.9903218760826744, "maximum": 0.9950287363912922, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0174644239810475, "standard_deviation": 0.04029291480102397, "minimum": 0.9421533259042749, "maximum": 1.0578235589777993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7159257362337232, "standard_deviation": 0.18351363021143396, "minimum": 0.49634332090457695, "maximum": 1.027489353901099, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1087249275319038, "standard_deviation": 0.015724592441201078, "minimum": 0.08592234742077988, "maximum": 0.12758145746618668, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25324923216910666, "standard_deviation": 0.0801436766458854, "minimum": 0.15096489002729432, "maximum": 0.40709574786705155, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.407668085731293, "standard_deviation": 0.006742259571301149, "minimum": 5.400115157039508, "maximum": 5.417609248193894, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6063896672468154, "standard_deviation": 0.062351468473920466, "minimum": 0.5291382615860986, "maximum": 0.7229944436571886, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09455384342585518, "standard_deviation": 0.05319973405408852, "minimum": -0.15821189319927448, "maximum": -0.028051190064153984, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900914874194991, "standard_deviation": 0.0013580537143389913, "minimum": 0.988046265794441, "maximum": 0.9916351361332025, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9215256037844808, "standard_deviation": 0.02532380336570785, "minimum": 0.8957220532197406, "maximum": 0.9744296388357585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36719444399246637, "standard_deviation": 0.21772696572032663, "minimum": 0.14461948208900013, "maximum": 0.6895719069231645, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10458785339671306, "standard_deviation": 0.006826692095275912, "minimum": 0.09689488845114819, "maximum": 0.11471154812874633, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971043588698378, "standard_deviation": 0.000636199682178053, "minimum": 0.9961561249361005, "maximum": 0.9981447526856433, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5588074094357249, "standard_deviation": 0.02305289909035957, "minimum": 0.5217024354801426, "maximum": 0.6022186613316257, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31231044047492446, "standard_deviation": 0.12577495143456574, "minimum": 0.15841673238317103, "maximum": 0.501978095938272, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05552720756364271, "standard_deviation": 0.005148623833097694, "minimum": 0.04702635652328298, "maximum": 0.06291211989847934, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0002", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.66607928276062, "predicted_mos": 2.66607928276062, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "00dcbd485cfbc12527bbc011bd03f0b0261011216e6dbabe7b0414238a335efa", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.030192099977285, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8663074717515549, "全帯域_10ms_rms_envelope_nrmse": 0.09508123360762046, "全帯域_10ms_envelope_correlation": 0.9918257517066925, "全帯域_10ms_modulation_mae_db": 0.2231402927014172, "全帯域_40ms_log_rms_mae_db": 0.5147271544605307, "全帯域_40ms_rms_envelope_nrmse": 0.05007429376840121, "全帯域_40ms_envelope_correlation": 0.9976564531616651, "全帯域_40ms_modulation_mae_db": 0.34542243884891244, "300–1000 Hz_10ms_log_rms_mae_db": 1.4841522436858376, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13109539426015254, "300–1000 Hz_10ms_envelope_correlation": 0.9885313942552154, "300–1000 Hz_10ms_modulation_mae_db": 0.7536271339209875, "300–1000 Hz_40ms_log_rms_mae_db": 0.930941306426492, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07303968673374335, "300–1000 Hz_40ms_envelope_correlation": 0.9967608717932219, "300–1000 Hz_40ms_modulation_mae_db": 0.8335095316714257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7863414478181014, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26812897951463316, "1000–3000 Hz_10ms_envelope_correlation": 0.952601962211501, "1000–3000 Hz_10ms_modulation_mae_db": 2.0904131345469414, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4089256436130508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14153795373005784, "1000–3000 Hz_40ms_envelope_correlation": 0.9860007356033444, "1000–3000 Hz_40ms_modulation_mae_db": 1.6724645407633876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7886262837403484, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19437671414916263, "3000–6000 Hz_10ms_envelope_correlation": 0.9761381523042892, "3000–6000 Hz_10ms_modulation_mae_db": 0.6003506315494865, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4167557544187566, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1218702724668367, "3000–6000 Hz_40ms_envelope_correlation": 0.9903408839836811, "3000–6000 Hz_40ms_modulation_mae_db": 0.4922682809556573, "mr_stft_db": 5.366622511903633, "band_mae_db": 0.25649833031039554, "rms_gain_error_db": -0.08675553137746518, "output_peak": 0.4961052867774213}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9060339649482103, "全帯域_10ms_rms_envelope_nrmse": 0.09098123512475408, "全帯域_10ms_envelope_correlation": 0.9925503169681688, "全帯域_10ms_modulation_mae_db": 0.7519515849539847, "全帯域_40ms_log_rms_mae_db": 0.5517113109413754, "全帯域_40ms_rms_envelope_nrmse": 0.04687516247515252, "全帯域_40ms_envelope_correlation": 0.9979959083403652, "全帯域_40ms_modulation_mae_db": 0.7196006982124223, "300–1000 Hz_10ms_log_rms_mae_db": 1.4937093134136246, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17282867109978015, "300–1000 Hz_10ms_envelope_correlation": 0.9806417447275522, "300–1000 Hz_10ms_modulation_mae_db": 0.8760958102816391, "300–1000 Hz_40ms_log_rms_mae_db": 0.993597832069124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11432220813179009, "300–1000 Hz_40ms_envelope_correlation": 0.992514518609845, "300–1000 Hz_40ms_modulation_mae_db": 0.6353302581869735, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8358652028213407, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24066782976361784, "1000–3000 Hz_10ms_envelope_correlation": 0.9607907325673369, "1000–3000 Hz_10ms_modulation_mae_db": 1.7562638557754322, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4786755695365972, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12959870116835145, "1000–3000 Hz_40ms_envelope_correlation": 0.9881351933529133, "1000–3000 Hz_40ms_modulation_mae_db": 1.8699347314581178, "3000–6000 Hz_10ms_log_rms_mae_db": 1.776723309887726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18720139339601147, "3000–6000 Hz_10ms_envelope_correlation": 0.9780476472648594, "3000–6000 Hz_10ms_modulation_mae_db": 0.33440687196795266, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4110246340626775, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1191440338985693, "3000–6000 Hz_40ms_envelope_correlation": 0.990676905766732, "3000–6000 Hz_40ms_modulation_mae_db": 0.30583448994336665, "mr_stft_db": 5.364771045888648, "band_mae_db": 0.14269845325616756, "rms_gain_error_db": -0.11108460893868065, "output_peak": 0.5418280807412665}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9074196370162472, "全帯域_10ms_rms_envelope_nrmse": 0.10738200080467422, "全帯域_10ms_envelope_correlation": 0.9895575728623085, "全帯域_10ms_modulation_mae_db": 0.17038308827444337, "全帯域_40ms_log_rms_mae_db": 0.5586117854097652, "全帯域_40ms_rms_envelope_nrmse": 0.05443627349568809, "全帯域_40ms_envelope_correlation": 0.9971291203369688, "全帯域_40ms_modulation_mae_db": 0.1965338733961707, "300–1000 Hz_10ms_log_rms_mae_db": 1.4442901763368354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18677565690168546, "300–1000 Hz_10ms_envelope_correlation": 0.975994407796878, "300–1000 Hz_10ms_modulation_mae_db": 1.1635994792817546, "300–1000 Hz_40ms_log_rms_mae_db": 0.937251514815537, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09980383628987223, "300–1000 Hz_40ms_envelope_correlation": 0.993229271815596, "300–1000 Hz_40ms_modulation_mae_db": 0.7582645962129752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8303679054570288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22621602372975907, "1000–3000 Hz_10ms_envelope_correlation": 0.9658762692644531, "1000–3000 Hz_10ms_modulation_mae_db": 1.4321717924728905, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4563868973390168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12929166535063905, "1000–3000 Hz_40ms_envelope_correlation": 0.9896118455282324, "1000–3000 Hz_40ms_modulation_mae_db": 1.8890488541760737, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7830192513051732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2009300607444921, "3000–6000 Hz_10ms_envelope_correlation": 0.9772807820052074, "3000–6000 Hz_10ms_modulation_mae_db": 0.626674611332756, "3000–6000 Hz_40ms_log_rms_mae_db": 1.448464166790143, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13298693575656287, "3000–6000 Hz_40ms_envelope_correlation": 0.9904188376542242, "3000–6000 Hz_40ms_modulation_mae_db": 0.5639528752256425, "mr_stft_db": 5.376921156720395, "band_mae_db": 0.4080272536903828, "rms_gain_error_db": -0.03112889884720282, "output_peak": 0.6148674731759586}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9260010472129852, "全帯域_10ms_rms_envelope_nrmse": 0.10173667053062078, "全帯域_10ms_envelope_correlation": 0.9906218124355257, "全帯域_10ms_modulation_mae_db": 0.18320299261480422, "全帯域_40ms_log_rms_mae_db": 0.5242304365968393, "全帯域_40ms_rms_envelope_nrmse": 0.04546776303184873, "全帯域_40ms_envelope_correlation": 0.9980255765562198, "全帯域_40ms_modulation_mae_db": 0.5936974432354613, "300–1000 Hz_10ms_log_rms_mae_db": 1.4834698392195678, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16188618695529094, "300–1000 Hz_10ms_envelope_correlation": 0.982621150194539, "300–1000 Hz_10ms_modulation_mae_db": 1.2456855378635376, "300–1000 Hz_40ms_log_rms_mae_db": 0.9988506876194744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.095949352904243, "300–1000 Hz_40ms_envelope_correlation": 0.9945610759098303, "300–1000 Hz_40ms_modulation_mae_db": 0.7967041028120524, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8369978069726478, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22599218173661992, "1000–3000 Hz_10ms_envelope_correlation": 0.9672244636541926, "1000–3000 Hz_10ms_modulation_mae_db": 2.4252964253384914, "1000–3000 Hz_40ms_log_rms_mae_db": 1.506833165055112, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.140841272389448, "1000–3000 Hz_40ms_envelope_correlation": 0.9891866266598098, "1000–3000 Hz_40ms_modulation_mae_db": 2.859699645195406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.797705898573488, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1867853851452391, "3000–6000 Hz_10ms_envelope_correlation": 0.9790272945053319, "3000–6000 Hz_10ms_modulation_mae_db": 0.3158366720796673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4452706118897796, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1147754299475782, "3000–6000 Hz_40ms_envelope_correlation": 0.9919466455718285, "3000–6000 Hz_40ms_modulation_mae_db": 0.38800074708968363, "mr_stft_db": 5.369289387502431, "band_mae_db": 0.27873508595747065, "rms_gain_error_db": -0.05230537503816551, "output_peak": 0.6217309886533413}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9179116078002169, "全帯域_10ms_rms_envelope_nrmse": 0.10266982643706406, "全帯域_10ms_envelope_correlation": 0.9904653689110112, "全帯域_10ms_modulation_mae_db": 0.34489978693549367, "全帯域_40ms_log_rms_mae_db": 0.5248291535760481, "全帯域_40ms_rms_envelope_nrmse": 0.050132470055398386, "全帯域_40ms_envelope_correlation": 0.9975574813128325, "全帯域_40ms_modulation_mae_db": 0.123646174036736, "300–1000 Hz_10ms_log_rms_mae_db": 1.5045878104568347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2014085759073479, "300–1000 Hz_10ms_envelope_correlation": 0.9721159231829802, "300–1000 Hz_10ms_modulation_mae_db": 1.8924365684687263, "300–1000 Hz_40ms_log_rms_mae_db": 0.9833814607570377, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10003575099465911, "300–1000 Hz_40ms_envelope_correlation": 0.9927677223556207, "300–1000 Hz_40ms_modulation_mae_db": 1.7678094497883048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8409334373059472, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21225683004370885, "1000–3000 Hz_10ms_envelope_correlation": 0.9696222533083834, "1000–3000 Hz_10ms_modulation_mae_db": 0.6085370762146893, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4542219765487248, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1294392761911798, "1000–3000 Hz_40ms_envelope_correlation": 0.9884150232926153, "1000–3000 Hz_40ms_modulation_mae_db": 0.8074325862354068, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8097553294989452, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17962902915564874, "3000–6000 Hz_10ms_envelope_correlation": 0.9800046352928433, "3000–6000 Hz_10ms_modulation_mae_db": 0.4370284421674224, "3000–6000 Hz_40ms_log_rms_mae_db": 1.429156407822605, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1026033940311576, "3000–6000 Hz_40ms_envelope_correlation": 0.9932267822146428, "3000–6000 Hz_40ms_modulation_mae_db": 0.4728686822604087, "mr_stft_db": 5.377879172374706, "band_mae_db": 0.2711451020887987, "rms_gain_error_db": -0.001006947024423529, "output_peak": 0.5705121742568887}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.916200685449208, "全帯域_10ms_rms_envelope_nrmse": 0.10728237119950673, "全帯域_10ms_envelope_correlation": 0.9896142466445933, "全帯域_10ms_modulation_mae_db": 0.08872496733982511, "全帯域_40ms_log_rms_mae_db": 0.5190248039218648, "全帯域_40ms_rms_envelope_nrmse": 0.058315977582635736, "全帯域_40ms_envelope_correlation": 0.9967172143037886, "全帯域_40ms_modulation_mae_db": 0.21377417792423933, "300–1000 Hz_10ms_log_rms_mae_db": 1.5094156634578075, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19556210284546519, "300–1000 Hz_10ms_envelope_correlation": 0.9735851430586048, "300–1000 Hz_10ms_modulation_mae_db": 0.18648177685609468, "300–1000 Hz_40ms_log_rms_mae_db": 0.9812291613122854, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11429937616780525, "300–1000 Hz_40ms_envelope_correlation": 0.9905370700616597, "300–1000 Hz_40ms_modulation_mae_db": 0.1256538579229447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.854529550471574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2655007159379801, "1000–3000 Hz_10ms_envelope_correlation": 0.9535722507444712, "1000–3000 Hz_10ms_modulation_mae_db": 1.2921298918701032, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4535576191410888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1278269274705241, "1000–3000 Hz_40ms_envelope_correlation": 0.9886249699998593, "1000–3000 Hz_40ms_modulation_mae_db": 1.4476813144474805, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8102104252229605, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2114027111603533, "3000–6000 Hz_10ms_envelope_correlation": 0.9719042236575264, "3000–6000 Hz_10ms_modulation_mae_db": 0.6969337671148281, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3936221909490774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12832157935526384, "3000–6000 Hz_40ms_envelope_correlation": 0.990054969097503, "3000–6000 Hz_40ms_modulation_mae_db": 0.777315968228358, "mr_stft_db": 5.382990615290627, "band_mae_db": 0.08356831591312165, "rms_gain_error_db": 0.0037602332949128814, "output_peak": 0.5901188741229249}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8909365783664307, "全帯域_10ms_rms_envelope_nrmse": 0.09836035668405657, "全帯域_10ms_envelope_correlation": 0.9912640221332649, "全帯域_10ms_modulation_mae_db": 0.6740287678479877, "全帯域_40ms_log_rms_mae_db": 0.5019926964592542, "全帯域_40ms_rms_envelope_nrmse": 0.04780782410017184, "全帯域_40ms_envelope_correlation": 0.9978924490229278, "全帯域_40ms_modulation_mae_db": 0.1861616694877851, "300–1000 Hz_10ms_log_rms_mae_db": 1.597317093139016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18992332546571883, "300–1000 Hz_10ms_envelope_correlation": 0.9756212593115429, "300–1000 Hz_10ms_modulation_mae_db": 0.9090288846182283, "300–1000 Hz_40ms_log_rms_mae_db": 1.017397537364259, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11166227919315164, "300–1000 Hz_40ms_envelope_correlation": 0.9922811904590993, "300–1000 Hz_40ms_modulation_mae_db": 0.7524991254490137, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7996233157364783, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1955516402781899, "1000–3000 Hz_10ms_envelope_correlation": 0.974541576413399, "1000–3000 Hz_10ms_modulation_mae_db": 0.5869299014118908, "1000–3000 Hz_40ms_log_rms_mae_db": 1.414163522248271, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11406213600503749, "1000–3000 Hz_40ms_envelope_correlation": 0.990773801270198, "1000–3000 Hz_40ms_modulation_mae_db": 0.5698931483807502, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8023066991980659, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16696213121720366, "3000–6000 Hz_10ms_envelope_correlation": 0.9827911416041095, "3000–6000 Hz_10ms_modulation_mae_db": 0.42548389865028763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4196658347246478, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1097040097679534, "3000–6000 Hz_40ms_envelope_correlation": 0.9922370425655551, "3000–6000 Hz_40ms_modulation_mae_db": 0.5830419379775935, "mr_stft_db": 5.3613224453839905, "band_mae_db": 0.2187233292276715, "rms_gain_error_db": -0.10503251203821644, "output_peak": 0.580541606878523}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9463298149360414, "全帯域_10ms_rms_envelope_nrmse": 0.10338231627957453, "全帯域_10ms_envelope_correlation": 0.9904486946444345, "全帯域_10ms_modulation_mae_db": 0.5287833312775824, "全帯域_40ms_log_rms_mae_db": 0.5664704279793142, "全帯域_40ms_rms_envelope_nrmse": 0.06031806941170204, "全帯域_40ms_envelope_correlation": 0.9967295276219477, "全帯域_40ms_modulation_mae_db": 0.4743677552305015, "300–1000 Hz_10ms_log_rms_mae_db": 1.5801798591368184, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16082201863200335, "300–1000 Hz_10ms_envelope_correlation": 0.9824794481211169, "300–1000 Hz_10ms_modulation_mae_db": 1.1354261659378908, "300–1000 Hz_40ms_log_rms_mae_db": 1.0451606560685038, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10146494706520129, "300–1000 Hz_40ms_envelope_correlation": 0.9931897021178923, "300–1000 Hz_40ms_modulation_mae_db": 1.2285625762761885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.792654045683715, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25719460908870234, "1000–3000 Hz_10ms_envelope_correlation": 0.956208155552727, "1000–3000 Hz_10ms_modulation_mae_db": 0.8207184875135779, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4309587279461158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15034048373946463, "1000–3000 Hz_40ms_envelope_correlation": 0.9841038485937017, "1000–3000 Hz_40ms_modulation_mae_db": 0.782334958811239, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7740721465655735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1780685509569042, "3000–6000 Hz_10ms_envelope_correlation": 0.9799459925863112, "3000–6000 Hz_10ms_modulation_mae_db": 0.26730128991494384, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4007608438532175, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09820313951435339, "3000–6000 Hz_40ms_envelope_correlation": 0.9937074506295177, "3000–6000 Hz_40ms_modulation_mae_db": 0.3808839349135778, "mr_stft_db": 5.3675472899996635, "band_mae_db": 0.22870681782363925, "rms_gain_error_db": -0.16110980713888629, "output_peak": 0.6696834436930577}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.696949944820346, "spectral_convergence": 0.2979142587586459, "spectral_convergence_status": "ok", "amplitude_floor": 1.8465213976940698e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.178274372074365, "spectral_convergence": 0.30840152791863656, "spectral_convergence_status": "ok", "amplitude_floor": 1.4089022879434144e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.22464321881619, "spectral_convergence": 0.341221486401597, "spectral_convergence_status": "ok", "amplitude_floor": 1.0260349239920231e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002262162256014784, "output_power": 0.0022172917352950836, "delta_db": -0.08700913726774576, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014828754084214857, "output_power": 0.0013830909025979994, "delta_db": -0.30253938303619843, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026062691889949753, "output_power": 0.00032085278963770006, "delta_db": 0.9028654971508836, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020447377727008734, "output_power": 0.0002042882301846775, "delta_db": -0.003942738287399304, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.078753769921038e-05, "output_power": 3.9847224316021493e-05, "delta_db": -0.10129413571911997, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.800369964531977e-05, "output_power": 2.8930059043592375e-05, "delta_db": 0.14133909040102619, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25649833031039554, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08675553137746518, "pitch": null, "output_rms_dbfs": -23.773404121254636, "output_peak": 0.4961052867774213, "reference_rms_dbfs": -23.686648589877173, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.366622511903633}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022610981418459577, "log_rms_mae_db": 0.8663074717515549, "rms_envelope_nrmse": 0.09508123360762046, "envelope_correlation": 0.9918257517066925, "rms_gain_error_db": -0.0867075514721763, "reference_mean_power": 0.00428070632223443, "output_mean_power": 0.004196088882528432}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.20587323032393e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001500467490243704, "output_power": 0.0014172648520306482, "reference_active": true, "delta_db": -0.2477557344119008}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013253274491379155, "output_power": 0.0001228281926318062, "reference_active": true, "delta_db": -0.3302513140344732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.272532196136399e-05, "output_power": 3.204369172666348e-05, "reference_active": true, "delta_db": -0.09141382965787757}], "active_band_count": 3, "active_band_mae_db": 0.2231402927014172}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019856877416737668, "log_rms_mae_db": 0.5147271544605307, "rms_envelope_nrmse": 0.05007429376840121, "envelope_correlation": 0.9976564531616651, "rms_gain_error_db": -0.08673330682164375, "reference_mean_power": 0.004296662851053013, "output_mean_power": 0.004211705018845554}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.009968768189794e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014002280466658247, "output_power": 0.0013230302340990325, "reference_active": true, "delta_db": -0.2462900348389918}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3360463722840526e-05, "output_power": 5.221837721084662e-05, "reference_active": true, "delta_db": -0.09396223971766524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498078024970164e-06, "output_power": 2.128161071635628e-06, "reference_active": true, "delta_db": -0.6960150419900802}], "active_band_count": 3, "active_band_mae_db": 0.34542243884891244}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000212457290437723, "log_rms_mae_db": 1.4841522436858376, "rms_envelope_nrmse": 0.13109539426015254, "envelope_correlation": 0.9885313942552154, "rms_gain_error_db": -0.2742283461490685, "reference_mean_power": 0.0012391682631200216, "output_mean_power": 0.0011633421063809918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.921692896601052e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00036532110371103373, "output_power": 0.0003249848824216898, "reference_active": true, "delta_db": -0.5081160204229723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3180096390035205e-05, "output_power": 3.515513536703137e-05, "reference_active": true, "delta_db": -0.8929483291341596}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1750484667690236e-05, "output_power": 9.639934320204213e-06, "reference_active": true, "delta_db": -0.8598170522058303}], "active_band_count": 3, "active_band_mae_db": 0.7536271339209875}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018128574308840577, "log_rms_mae_db": 0.930941306426492, "rms_envelope_nrmse": 0.07303968673374335, "envelope_correlation": 0.9967608717932219, "rms_gain_error_db": -0.2743283508142575, "reference_mean_power": 0.0012444705639415084, "output_mean_power": 0.0011682930507013562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.573537659747316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003540640123239247, "output_power": 0.00031539231808513045, "reference_active": true, "delta_db": -0.5023067541916307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0107771514146375e-05, "output_power": 1.6086435532860253e-05, "reference_active": true, "delta_db": -0.9690411881945921}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973740035567712e-07, "output_power": 6.291352327613462e-07, "reference_active": true, "delta_db": -1.029180652628054}], "active_band_count": 3, "active_band_mae_db": 0.8335095316714257}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001386433646872303, "log_rms_mae_db": 1.7863414478181014, "rms_envelope_nrmse": 0.26812897951463316, "envelope_correlation": 0.952601962211501, "rms_gain_error_db": 0.07391718068858642, "reference_mean_power": 0.00021050674147203463, "output_mean_power": 0.00021412024258028625}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9020912968770603e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.015318920526352e-05, "output_power": 9.865174200559992e-05, "reference_active": true, "delta_db": 0.9018395136271866}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8603151739641485e-05, "output_power": 4.588836915373831e-05, "reference_active": true, "delta_db": 0.7507985946558065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0682238748218467e-05, "output_power": 7.140644591135945e-06, "reference_active": true, "delta_db": -4.6186012953578315}], "active_band_count": 3, "active_band_mae_db": 2.0904131345469414}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.950462169412164e-05, "log_rms_mae_db": 1.4089256436130508, "rms_envelope_nrmse": 0.14153795373005784, "envelope_correlation": 0.9860007356033444, "rms_gain_error_db": 0.07536247059131022, "reference_mean_power": 0.0002110528457073582, "output_mean_power": 0.000214747174977669}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7773203833688661e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04144461796749e-05, "output_power": 0.00010735169208753384, "reference_active": true, "delta_db": 0.7457106778055864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4052463298203573e-05, "output_power": 2.443207703046801e-05, "reference_active": true, "delta_db": 0.0680082834807908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4632154850806674e-06, "output_power": 5.558285142239422e-07, "reference_active": true, "delta_db": -4.203674661003785}], "active_band_count": 3, "active_band_mae_db": 1.6724645407633876}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.6125957709065377e-05, "log_rms_mae_db": 1.7886262837403484, "rms_envelope_nrmse": 0.19437671414916263, "envelope_correlation": 0.9761381523042892, "rms_gain_error_db": -0.10867299870661087, "reference_mean_power": 2.4404568133397883e-05, "output_mean_power": 2.3801472551484262e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.103592290864487e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1042852972405107e-05, "output_power": 1.1468299578555555e-05, "reference_active": true, "delta_db": 0.16417739312573568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8700267025557585e-06, "output_power": 3.976605394338154e-06, "reference_active": true, "delta_db": 0.11798535715648517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.131168166122288e-06, "output_power": 7.973302847215071e-07, "reference_active": true, "delta_db": -1.518889144366239}], "active_band_count": 3, "active_band_mae_db": 0.6003506315494865}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.642002234154929e-05, "log_rms_mae_db": 1.4167557544187566, "rms_envelope_nrmse": 0.1218702724668367, "envelope_correlation": 0.9903408839836811, "rms_gain_error_db": -0.10959576547488811, "reference_mean_power": 2.4522755214627932e-05, "output_mean_power": 2.391165777874876e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9657184130662428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1868339578364968e-05, "output_power": 1.2186016422585432e-05, "reference_active": true, "delta_db": 0.11471795160653649}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9915566770914906e-06, "output_power": 2.0094113950729347e-06, "reference_active": true, "delta_db": 0.038761903771796606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.546174499350975e-08, "output_power": 6.30143204885736e-08, "reference_active": true, "delta_db": -1.3233249874886386}], "active_band_count": 3, "active_band_mae_db": 0.4922682809556573}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8801609835292206, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d25e6c7721890c54a7aa883df3c319a9087f3c362e1a701c1f58eccd08474583", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.962554707964558, "standard_deviation": 0.008014332006033116, "minimum": 0.952601962211501, "maximum": 0.974541576413399, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8221640890333541, "standard_deviation": 0.025458831143883684, "minimum": 1.7863414478181014, "maximum": 1.854529550471574, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3765575706430022, "standard_deviation": 0.6851657996695647, "minimum": 0.5869299014118908, "maximum": 2.4252964253384914, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2364386012616514, "standard_deviation": 0.026097934313176453, "minimum": 0.1955516402781899, "maximum": 0.26812897951463316, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9881065055375843, "standard_deviation": 0.002116854737990216, "minimum": 0.9841038485937017, "maximum": 0.990773801270198, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4504653901784972, "standard_deviation": 0.032636010705784056, "minimum": 1.4089256436130508, "maximum": 1.506833165055112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4873112224334828, "standard_deviation": 0.7585680639082847, "minimum": 0.5698931483807502, "maximum": 2.859699645195406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1328673020055878, "standard_deviation": 0.011073368429205248, "minimum": 0.11406213600503749, "maximum": 0.15034048373946463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9781424836525597, "standard_deviation": 0.0032248570242946494, "minimum": 0.9719042236575264, "maximum": 0.9827911416041095, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7928024179990352, "standard_deviation": 0.014279488987550777, "minimum": 1.7740721465655735, "maximum": 1.8102104252229605, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4630020230971681, "standard_deviation": 0.1598895977198483, "minimum": 0.26730128991494384, "maximum": 0.6969337671148281, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1881694969906269, "standard_deviation": 0.013983244195724005, "minimum": 0.16696213121720366, "maximum": 0.2114027111603533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9915761896854605, "standard_deviation": 0.0014055482177374712, "minimum": 0.990054969097503, "maximum": 0.9937074506295177, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.420590055563863, "standard_deviation": 0.01959225439192734, "minimum": 1.3936221909490774, "maximum": 1.448464166790143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.49552086457428607, "standard_deviation": 0.14782140841724195, "minimum": 0.30583448994336665, "maximum": 0.777315968228358, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11595109934228441, "standard_deviation": 0.012083279955888395, "minimum": 0.09820313951435339, "maximum": 0.13298693575656287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9789488088310536, "standard_deviation": 0.005553657589131641, "minimum": 0.9721159231829802, "maximum": 0.9885313942552154, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5121402498557928, "standard_deviation": 0.05140538966296207, "minimum": 1.4442901763368354, "maximum": 1.597317093139016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0202976696536072, "standard_deviation": 0.48473158489168966, "minimum": 0.18648177685609468, "maximum": 1.8924365684687263, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17503774150843054, "standard_deviation": 0.02325906772649318, "minimum": 0.13109539426015254, "maximum": 0.2014085759073479, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9932301778903456, "standard_deviation": 0.0018173044533107122, "minimum": 0.9905370700616597, "maximum": 0.9967608717932219, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9859762695540892, "standard_deviation": 0.03807163078371319, "minimum": 0.930941306426492, "maximum": 1.0451606560685038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8622916872899848, "standard_deviation": 0.47453281541006537, "minimum": 0.1256538579229447, "maximum": 1.7678094497883048, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10132217968505824, "standard_deviation": 0.013507000032410569, "minimum": 0.07303968673374335, "maximum": 0.11432220813179009, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23601283603345596, "standard_deviation": 0.09663428318730537, "minimum": 0.08356831591312165, "maximum": 0.4080272536903828, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.370917953133011, "standard_deviation": 0.007488742476672436, "minimum": 5.3613224453839905, "maximum": 5.382990615290627, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5856734910374227, "standard_deviation": 0.05268633637365534, "minimum": 0.4961052867774213, "maximum": 0.6696834436930577, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06808293088851594, "standard_deviation": 0.057931290804753434, "minimum": -0.16110980713888629, "maximum": 0.0037602332949128814, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.99079347328825, "standard_deviation": 0.0010383731952504412, "minimum": 0.9895575728623085, "maximum": 0.9925503169681688, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9096426009351117, "standard_deviation": 0.023817004449916405, "minimum": 0.8663074717515549, "maximum": 0.9463298149360414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37063935149319227, "standard_deviation": 0.2505943941892622, "minimum": 0.08872496733982511, "maximum": 0.7519515849539847, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10085950133348393, "standard_deviation": 0.005749804964168891, "minimum": 0.09098123512475408, "maximum": 0.10738200080467422, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974629663320894, "standard_deviation": 0.0005394328559730054, "minimum": 0.9967172143037886, "maximum": 0.9980255765562198, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5326997211681239, "standard_deviation": 0.023176184682586854, "minimum": 0.5019926964592542, "maximum": 0.5664704279793142, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3566505287965286, "standard_deviation": 0.2175565463400348, "minimum": 0.123646174036736, "maximum": 0.7196006982124223, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05167847924012482, "standard_deviation": 0.005449701251831096, "minimum": 0.04546776303184873, "maximum": 0.06031806941170204, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7021-79740-0011", "language": "en", "style": "normal", "speaker": "librispeech_7021", "source_utterance_id": "7021-79740-0011", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7021-79740-0011.wav", "source_file_sha256": "459de26d771395691267de58176712c0c1c093b32d84789d67d87ef93b8ef2c6", "decoded_pcm_sha256": "6cda455d70e2130c6b8f8269b688ec0381f52c4c07376a3f0f4bbbf73ca24fc7", "original_source_sha256": "af3d9a4d3df12f73bc4432b8be4762df32f7381a8c20009735ca8e982084c2a6", "corpus_entry_id": "librispeech_7021-79740-0011"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.028003215789795, "predicted_mos": 4.028003215789795, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6f9f5b456608d7f44ee223d639bfb84e148501c5efa2b76e94910a9cfb3a5904", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.397404164999898, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7021-79740-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6f9f5b456608d7f44ee223d639bfb84e148501c5efa2b76e94910a9cfb3a5904", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "15c66be27acb18c9df694af8bc9d44c5bd348f8389a0837a1e4616376dda091f"}, "rows": [{"case": "librispeech_7021-79740-0011", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.0932514667510986, "predicted_mos": 3.0932514667510986, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d92a9fdd8cde0306c5b788a1328aad1a48f808c4438613e6ce165a4bdd2ea1b5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2404902890011726, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7021-79740-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d92a9fdd8cde0306c5b788a1328aad1a48f808c4438613e6ce165a4bdd2ea1b5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6972909975645729, "全帯域_10ms_rms_envelope_nrmse": 0.12146574211681992, "全帯域_10ms_envelope_correlation": 0.9856239755872622, "全帯域_10ms_modulation_mae_db": 0.3635878266231683, "全帯域_40ms_log_rms_mae_db": 0.4174029108966334, "全帯域_40ms_rms_envelope_nrmse": 0.07080882270949616, "全帯域_40ms_envelope_correlation": 0.9949347649698107, "全帯域_40ms_modulation_mae_db": 0.5988808321757345, "300–1000 Hz_10ms_log_rms_mae_db": 1.2667171562836925, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16389977057879107, "300–1000 Hz_10ms_envelope_correlation": 0.9800146167408933, "300–1000 Hz_10ms_modulation_mae_db": 0.5188624230489012, "300–1000 Hz_40ms_log_rms_mae_db": 0.9433105068253119, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07659343035513368, "300–1000 Hz_40ms_envelope_correlation": 0.9955663881964572, "300–1000 Hz_40ms_modulation_mae_db": 0.1971634418602283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5799812327697307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2632564133246132, "1000–3000 Hz_10ms_envelope_correlation": 0.9586863251865073, "1000–3000 Hz_10ms_modulation_mae_db": 2.5189864435542018, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2985339356640626, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18411088220812497, "1000–3000 Hz_40ms_envelope_correlation": 0.982495976292515, "1000–3000 Hz_40ms_modulation_mae_db": 3.2818867400789427, "3000–6000 Hz_10ms_log_rms_mae_db": 1.599481136214782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16065528187160769, "3000–6000 Hz_10ms_envelope_correlation": 0.9844033370119832, "3000–6000 Hz_10ms_modulation_mae_db": 0.23902416329191298, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3911124857688808, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09002396071183763, "3000–6000 Hz_40ms_envelope_correlation": 0.9951675402930302, "3000–6000 Hz_40ms_modulation_mae_db": 0.5165802705519333, "mr_stft_db": 5.421988627480662, "band_mae_db": 0.38304764385297974, "rms_gain_error_db": -0.13726773426726888, "output_peak": 0.5245035119599527}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6750708782002955, "全帯域_10ms_rms_envelope_nrmse": 0.09264695284211773, "全帯域_10ms_envelope_correlation": 0.9917983976228255, "全帯域_10ms_modulation_mae_db": 0.409840040322402, "全帯域_40ms_log_rms_mae_db": 0.3947759259788984, "全帯域_40ms_rms_envelope_nrmse": 0.045810718446483115, "全帯域_40ms_envelope_correlation": 0.9981121742840161, "全帯域_40ms_modulation_mae_db": 0.4352829411191337, "300–1000 Hz_10ms_log_rms_mae_db": 1.252818371846539, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1381990776146893, "300–1000 Hz_10ms_envelope_correlation": 0.9864955913441921, "300–1000 Hz_10ms_modulation_mae_db": 0.5210899097997838, "300–1000 Hz_40ms_log_rms_mae_db": 0.9669278114470876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07523898857459196, "300–1000 Hz_40ms_envelope_correlation": 0.9966594680961219, "300–1000 Hz_40ms_modulation_mae_db": 0.6816538237261002, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6525150420098436, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2528662496323375, "1000–3000 Hz_10ms_envelope_correlation": 0.9600202298859142, "1000–3000 Hz_10ms_modulation_mae_db": 1.910191027331364, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3438804730770757, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1200640688330708, "1000–3000 Hz_40ms_envelope_correlation": 0.991945963114786, "1000–3000 Hz_40ms_modulation_mae_db": 2.013839207928368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.638319470924441, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17801423972422414, "3000–6000 Hz_10ms_envelope_correlation": 0.9811736884727786, "3000–6000 Hz_10ms_modulation_mae_db": 0.6968045101922385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4130542389507195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11302008822235173, "3000–6000 Hz_40ms_envelope_correlation": 0.9922270417035067, "3000–6000 Hz_40ms_modulation_mae_db": 0.9100534653419765, "mr_stft_db": 5.418284642786923, "band_mae_db": 0.32315196059927875, "rms_gain_error_db": -0.15428524504853391, "output_peak": 0.4668337056807469}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7210280565273351, "全帯域_10ms_rms_envelope_nrmse": 0.10386108893676214, "全帯域_10ms_envelope_correlation": 0.9895326193115586, "全帯域_10ms_modulation_mae_db": 0.5051910518530714, "全帯域_40ms_log_rms_mae_db": 0.44480741649706157, "全帯域_40ms_rms_envelope_nrmse": 0.052378156154925505, "全帯域_40ms_envelope_correlation": 0.9973238427088656, "全帯域_40ms_modulation_mae_db": 0.5287055218440796, "300–1000 Hz_10ms_log_rms_mae_db": 1.3575879945692895, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1902995313263248, "300–1000 Hz_10ms_envelope_correlation": 0.9733507951832655, "300–1000 Hz_10ms_modulation_mae_db": 0.6559966148880356, "300–1000 Hz_40ms_log_rms_mae_db": 1.0101195225968025, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1167987235217787, "300–1000 Hz_40ms_envelope_correlation": 0.9902526691377211, "300–1000 Hz_40ms_modulation_mae_db": 0.49519061951851145, "1000–3000 Hz_10ms_log_rms_mae_db": 1.608605537409589, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19280787297580515, "1000–3000 Hz_10ms_envelope_correlation": 0.977240380175687, "1000–3000 Hz_10ms_modulation_mae_db": 1.59100092474129, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3459589787524584, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09037531528755073, "1000–3000 Hz_40ms_envelope_correlation": 0.9948560472217335, "1000–3000 Hz_40ms_modulation_mae_db": 2.0435489706795082, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6303065987783385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16399541662580883, "3000–6000 Hz_10ms_envelope_correlation": 0.9838443155003836, "3000–6000 Hz_10ms_modulation_mae_db": 0.6849340958311524, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4175779295456385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09650525091449194, "3000–6000 Hz_40ms_envelope_correlation": 0.9945715547193321, "3000–6000 Hz_40ms_modulation_mae_db": 0.783897090064969, "mr_stft_db": 5.429421531272378, "band_mae_db": 0.22783607717253887, "rms_gain_error_db": -0.11390169414960535, "output_peak": 0.566242991286385}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6574447054563968, "全帯域_10ms_rms_envelope_nrmse": 0.09314973879833369, "全帯域_10ms_envelope_correlation": 0.9916795717184264, "全帯域_10ms_modulation_mae_db": 0.3965455915208614, "全帯域_40ms_log_rms_mae_db": 0.4231117892479231, "全帯域_40ms_rms_envelope_nrmse": 0.05351062445474432, "全帯域_40ms_envelope_correlation": 0.9972481650040819, "全帯域_40ms_modulation_mae_db": 0.6084132262395505, "300–1000 Hz_10ms_log_rms_mae_db": 1.2912427815800858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14341381003443268, "300–1000 Hz_10ms_envelope_correlation": 0.9853718423486731, "300–1000 Hz_10ms_modulation_mae_db": 0.6126820872774582, "300–1000 Hz_40ms_log_rms_mae_db": 0.9937917367959689, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07722773198258763, "300–1000 Hz_40ms_envelope_correlation": 0.9964153959815062, "300–1000 Hz_40ms_modulation_mae_db": 0.8705124874134528, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5967667734952002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21815762399957978, "1000–3000 Hz_10ms_envelope_correlation": 0.970449506427704, "1000–3000 Hz_10ms_modulation_mae_db": 1.9432052164184823, "1000–3000 Hz_40ms_log_rms_mae_db": 1.314516923401999, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12701702496430042, "1000–3000 Hz_40ms_envelope_correlation": 0.99040244032007, "1000–3000 Hz_40ms_modulation_mae_db": 1.499218746954078, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6344852051921328, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16747024983314343, "3000–6000 Hz_10ms_envelope_correlation": 0.9832880353926604, "3000–6000 Hz_10ms_modulation_mae_db": 0.654336667431555, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4232934249873659, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10687451776970008, "3000–6000 Hz_40ms_envelope_correlation": 0.9935084963036752, "3000–6000 Hz_40ms_modulation_mae_db": 0.8848894599523257, "mr_stft_db": 5.420124786610443, "band_mae_db": 0.28931394256298765, "rms_gain_error_db": -0.14606210008793852, "output_peak": 0.5977976861982758}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7525056817251734, "全帯域_10ms_rms_envelope_nrmse": 0.11006838053414902, "全帯域_10ms_envelope_correlation": 0.988243533297929, "全帯域_10ms_modulation_mae_db": 0.3438814880794238, "全帯域_40ms_log_rms_mae_db": 0.4746382100800232, "全帯域_40ms_rms_envelope_nrmse": 0.05518546881469832, "全帯域_40ms_envelope_correlation": 0.9969774592522495, "全帯域_40ms_modulation_mae_db": 0.07704650852460498, "300–1000 Hz_10ms_log_rms_mae_db": 1.3040875943005001, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16410209495629544, "300–1000 Hz_10ms_envelope_correlation": 0.9802567330076073, "300–1000 Hz_10ms_modulation_mae_db": 0.3884388057100761, "300–1000 Hz_40ms_log_rms_mae_db": 1.0002675894114075, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09951025617377927, "300–1000 Hz_40ms_envelope_correlation": 0.9928292574057199, "300–1000 Hz_40ms_modulation_mae_db": 0.3933831391790659, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5827312715007815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26829628541661515, "1000–3000 Hz_10ms_envelope_correlation": 0.9546712685242434, "1000–3000 Hz_10ms_modulation_mae_db": 1.6741265418587468, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3024506483496192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15463515721262153, "1000–3000 Hz_40ms_envelope_correlation": 0.9848009091316087, "1000–3000 Hz_40ms_modulation_mae_db": 1.7557184121862743, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6044268646907844, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17404965746043136, "3000–6000 Hz_10ms_envelope_correlation": 0.9816316717491195, "3000–6000 Hz_10ms_modulation_mae_db": 0.3018534286263259, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3955148395885304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10220102234860384, "3000–6000 Hz_40ms_envelope_correlation": 0.9935429962643827, "3000–6000 Hz_40ms_modulation_mae_db": 0.476020279579624, "mr_stft_db": 5.4157623872788845, "band_mae_db": 0.19891591960461055, "rms_gain_error_db": -0.12316792148619511, "output_peak": 0.5409064252652199}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.693299726029243, "全帯域_10ms_rms_envelope_nrmse": 0.10236458196642967, "全帯域_10ms_envelope_correlation": 0.989813434577867, "全帯域_10ms_modulation_mae_db": 0.6294937544656086, "全帯域_40ms_log_rms_mae_db": 0.4387221074356771, "全帯域_40ms_rms_envelope_nrmse": 0.048934899429363805, "全帯域_40ms_envelope_correlation": 0.9975881492510107, "全帯域_40ms_modulation_mae_db": 0.30686448054694987, "300–1000 Hz_10ms_log_rms_mae_db": 1.3280033702148808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15984968902886987, "300–1000 Hz_10ms_envelope_correlation": 0.9811233388727606, "300–1000 Hz_10ms_modulation_mae_db": 0.4785594972682252, "300–1000 Hz_40ms_log_rms_mae_db": 1.0506123430609278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08412083274298063, "300–1000 Hz_40ms_envelope_correlation": 0.9944755013885441, "300–1000 Hz_40ms_modulation_mae_db": 0.21921117030355167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6126718815608236, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23822417343501, "1000–3000 Hz_10ms_envelope_correlation": 0.9697085636150001, "1000–3000 Hz_10ms_modulation_mae_db": 2.45256030355702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3517691772492522, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16551904526989542, "1000–3000 Hz_40ms_envelope_correlation": 0.9901187648277749, "1000–3000 Hz_40ms_modulation_mae_db": 2.6390222461323174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6671863103577471, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18364562719660965, "3000–6000 Hz_10ms_envelope_correlation": 0.9797659359653667, "3000–6000 Hz_10ms_modulation_mae_db": 0.6716307162456571, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4686943753596458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10279938414592671, "3000–6000 Hz_40ms_envelope_correlation": 0.9934930438317882, "3000–6000 Hz_40ms_modulation_mae_db": 0.8068413900957445, "mr_stft_db": 5.4273200032842945, "band_mae_db": 0.4030074005706064, "rms_gain_error_db": -0.09659193378740932, "output_peak": 0.5954566220722719}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6683732233240455, "全帯域_10ms_rms_envelope_nrmse": 0.08431521095017246, "全帯域_10ms_envelope_correlation": 0.9930704354727874, "全帯域_10ms_modulation_mae_db": 0.3160525477186356, "全帯域_40ms_log_rms_mae_db": 0.41538255454407746, "全帯域_40ms_rms_envelope_nrmse": 0.048897150241530254, "全帯域_40ms_envelope_correlation": 0.9975045112587626, "全帯域_40ms_modulation_mae_db": 0.4585403330389239, "300–1000 Hz_10ms_log_rms_mae_db": 1.2981897606628596, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1464748336200249, "300–1000 Hz_10ms_envelope_correlation": 0.9840572140856663, "300–1000 Hz_10ms_modulation_mae_db": 0.39615364426305105, "300–1000 Hz_40ms_log_rms_mae_db": 1.0228221227955556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09260402098422307, "300–1000 Hz_40ms_envelope_correlation": 0.9932911543003057, "300–1000 Hz_40ms_modulation_mae_db": 0.6345830819209833, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6139004843535032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2405065034552823, "1000–3000 Hz_10ms_envelope_correlation": 0.9637149572696125, "1000–3000 Hz_10ms_modulation_mae_db": 2.1047900570509275, "1000–3000 Hz_40ms_log_rms_mae_db": 1.351255998657728, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14123834706613422, "1000–3000 Hz_40ms_envelope_correlation": 0.9873339562222896, "1000–3000 Hz_40ms_modulation_mae_db": 1.928528928592873, "3000–6000 Hz_10ms_log_rms_mae_db": 1.58699633281844, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16232226683766537, "3000–6000 Hz_10ms_envelope_correlation": 0.98467844934977, "3000–6000 Hz_10ms_modulation_mae_db": 0.20911339907885704, "3000–6000 Hz_40ms_log_rms_mae_db": 1.379226522314007, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09241090478558886, "3000–6000 Hz_40ms_envelope_correlation": 0.9950788022488014, "3000–6000 Hz_40ms_modulation_mae_db": 0.2531852690631877, "mr_stft_db": 5.432051424572994, "band_mae_db": 0.15369323847678212, "rms_gain_error_db": -0.03963444818787813, "output_peak": 0.481412921951673}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6991119226857888, "全帯域_10ms_rms_envelope_nrmse": 0.09751647721313234, "全帯域_10ms_envelope_correlation": 0.9907491607786042, "全帯域_10ms_modulation_mae_db": 0.5243557829096864, "全帯域_40ms_log_rms_mae_db": 0.4209493320473198, "全帯域_40ms_rms_envelope_nrmse": 0.045516708385446115, "全帯域_40ms_envelope_correlation": 0.9979253304655554, "全帯域_40ms_modulation_mae_db": 0.724830933714857, "300–1000 Hz_10ms_log_rms_mae_db": 1.3298925200202332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13837616825999033, "300–1000 Hz_10ms_envelope_correlation": 0.9858419585198374, "300–1000 Hz_10ms_modulation_mae_db": 0.15630301046316275, "300–1000 Hz_40ms_log_rms_mae_db": 0.9597270442801524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07159093624479886, "300–1000 Hz_40ms_envelope_correlation": 0.9960534930061382, "300–1000 Hz_40ms_modulation_mae_db": 0.31947261681853106, "1000–3000 Hz_10ms_log_rms_mae_db": 1.598505109148853, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20480309097504804, "1000–3000 Hz_10ms_envelope_correlation": 0.9757312046969204, "1000–3000 Hz_10ms_modulation_mae_db": 1.0983793596875842, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3027661869774747, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12265924337052905, "1000–3000 Hz_40ms_envelope_correlation": 0.9935471221076851, "1000–3000 Hz_40ms_modulation_mae_db": 1.049623084375332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6213481217203018, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15681700496469086, "3000–6000 Hz_10ms_envelope_correlation": 0.9851522348433163, "3000–6000 Hz_10ms_modulation_mae_db": 0.3247845941363329, "3000–6000 Hz_40ms_log_rms_mae_db": 1.405173699782302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0927110802657457, "3000–6000 Hz_40ms_envelope_correlation": 0.9946840603012354, "3000–6000 Hz_40ms_modulation_mae_db": 0.5493617727176904, "mr_stft_db": 5.40691337617801, "band_mae_db": 0.3499073030741337, "rms_gain_error_db": -0.09504674473474603, "output_peak": 0.5426011043822001}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.499854327413498, "spectral_convergence": 0.3149980354503007, "spectral_convergence_status": "ok", "amplitude_floor": 1.6898412267444875e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.218924412951365, "spectral_convergence": 0.31724433508229705, "spectral_convergence_status": "ok", "amplitude_floor": 1.0945626100175482e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5471871420771235, "spectral_convergence": 0.3446413556431549, "spectral_convergence_status": "ok", "amplitude_floor": 6.58943811507865e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017588563541266742, "output_power": 0.0017097007639443392, "delta_db": -0.12310266307904857, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012071674288590822, "output_power": 0.0012110272198034536, "delta_db": 0.013863957005423325, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003927549262280726, "output_power": 0.00041835306563303476, "delta_db": 0.27421314788269013, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003207071656425, "output_power": 0.00022701159042556285, "delta_db": -1.500606322586201, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.4115036625953016e-05, "output_power": 5.232359455049811e-05, "delta_db": -0.1462038501640568, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.305762815858983e-05, "output_power": 4.073995178400233e-05, "delta_db": -0.24029592240045874, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.38304764385297974, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13726773426726888, "pitch": null, "output_rms_dbfs": -24.366190528442246, "output_peak": 0.5245035119599527, "reference_rms_dbfs": -24.228922794174977, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.421988627480662}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002198976712894503, "log_rms_mae_db": 0.6972909975645729, "rms_envelope_nrmse": 0.12146574211681992, "envelope_correlation": 0.9856239755872622, "rms_gain_error_db": -0.13643651261414583, "reference_mean_power": 0.0037766562988771884, "output_mean_power": 0.0036598544290446272}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.069519300574134e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006799849644412746, "output_power": 0.0005838182497876481, "reference_active": true, "delta_db": -0.6622164323067621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022275189153607582, "output_power": 0.00020762664359866653, "reference_active": true, "delta_db": -0.3053831720812203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.238302101062001e-05, "output_power": 4.3602189393541405e-05, "reference_active": true, "delta_db": 0.12316387548152245}], "active_band_count": 3, "active_band_mae_db": 0.3635878266231683}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018113949472991762, "log_rms_mae_db": 0.4174029108966334, "rms_envelope_nrmse": 0.07080882270949616, "envelope_correlation": 0.9949347649698107, "rms_gain_error_db": -0.1376323719846041, "reference_mean_power": 0.003780325824366865, "output_mean_power": 0.0036624018598350374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9093794448028363e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006511405512347742, "output_power": 0.0005574604464042262, "reference_active": true, "delta_db": -0.674606846143492}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001166725476043267, "output_power": 0.00010410760204460224, "reference_active": true, "delta_db": -0.4948623771126134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.399065253982206e-06, "output_power": 2.9419966560329086e-06, "reference_active": true, "delta_db": -0.627173273271098}], "active_band_count": 3, "active_band_mae_db": 0.5988808321757345}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013526658282762358, "log_rms_mae_db": 1.2667171562836925, "rms_envelope_nrmse": 0.16389977057879107, "envelope_correlation": 0.9800146167408933, "rms_gain_error_db": -0.15741857249414778, "reference_mean_power": 0.001047203990993015, "output_mean_power": 0.0010099257172552784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.584657585149468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003504540347149448, "output_power": 0.0003223902167115446, "reference_active": true, "delta_db": -0.3624921011783817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.5576057457372e-05, "output_power": 9.472581356337975e-05, "reference_active": true, "delta_db": -0.038807678189902356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9623586773434205e-05, "output_power": 2.5603959941539844e-05, "reference_active": true, "delta_db": 1.1552874897784196}], "active_band_count": 3, "active_band_mae_db": 0.5188624230489012}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012031399971827572, "log_rms_mae_db": 0.9433105068253119, "rms_envelope_nrmse": 0.07659343035513368, "envelope_correlation": 0.9955663881964572, "rms_gain_error_db": -0.15771326381778727, "reference_mean_power": 0.0010510881038568295, "output_mean_power": 0.0010136027832799788}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.061243701565635e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00034327462417827815, "output_power": 0.000321908213394924, "reference_active": true, "delta_db": -0.27909642622735203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.625199841508274e-05, "output_power": 4.549110563133212e-05, "reference_active": true, "delta_db": -0.07204009889655541}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5417688576161987e-06, "output_power": 1.458760143751741e-06, "reference_active": true, "delta_db": -0.24035380045677748}], "active_band_count": 3, "active_band_mae_db": 0.1971634418602283}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013601220330575488, "log_rms_mae_db": 1.5799812327697307, "rms_envelope_nrmse": 0.2632564133246132, "envelope_correlation": 0.9586863251865073, "rms_gain_error_db": -0.84572448736287, "reference_mean_power": 0.00024923569921471797, "output_mean_power": 0.00020513407139844268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.70738105242738e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001701759171968848, "output_power": 0.0001241473216934333, "reference_active": true, "delta_db": -1.3696074530972024}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6462817859538857e-05, "output_power": 1.0363567202485479e-05, "reference_active": true, "delta_db": -4.071268196814728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.097841074264338e-06, "output_power": 5.042750740775929e-06, "reference_active": true, "delta_db": 2.116083680750675}], "active_band_count": 3, "active_band_mae_db": 2.5189864435542018}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011511226838452259, "log_rms_mae_db": 1.2985339356640626, "rms_envelope_nrmse": 0.18411088220812497, "envelope_correlation": 0.982495976292515, "rms_gain_error_db": -0.8495969296508854, "reference_mean_power": 0.00024963838926192667, "output_mean_power": 0.00020528238216657537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6223780118670214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016948627965662844, "output_power": 0.00012159008448077403, "reference_active": true, "delta_db": -1.4423638651517876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1767373976886565e-05, "output_power": 3.780607954828993e-06, "reference_active": true, "delta_db": -4.931179121065187}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8952968721807594e-07, "output_power": 4.215887072684593e-07, "reference_active": true, "delta_db": 3.4721172340198536}], "active_band_count": 3, "active_band_mae_db": 3.2818867400789427}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.3800826555528964e-05, "log_rms_mae_db": 1.599481136214782, "rms_envelope_nrmse": 0.16065528187160769, "envelope_correlation": 0.9844033370119832, "rms_gain_error_db": -0.15554519770089253, "reference_mean_power": 4.104766861392576e-05, "output_mean_power": 3.9603537105601436e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.890895346126465e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.549789437651056e-06, "output_power": 9.629758887984645e-06, "reference_active": true, "delta_db": 0.036216173262990106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.275656140671785e-06, "output_power": 3.0760918999589616e-06, "reference_active": true, "delta_db": -0.2729899965975486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.939546621487433e-07, "output_power": 5.407128162951225e-07, "reference_active": true, "delta_db": -0.40786632001520023}], "active_band_count": 3, "active_band_mae_db": 0.23902416329191298}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.772714769720092e-05, "log_rms_mae_db": 1.3911124857688808, "rms_envelope_nrmse": 0.09002396071183763, "envelope_correlation": 0.9951675402930302, "rms_gain_error_db": -0.15630496000552693, "reference_mean_power": 4.122071638798201e-05, "output_mean_power": 3.976353983359455e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.778938094011469e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.936665760227185e-06, "output_power": 9.972639895672156e-06, "reference_active": true, "delta_db": 0.015694555890503895}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7465237791634235e-06, "output_power": 1.6518634416268248e-06, "reference_active": true, "delta_db": -0.2420036134042558}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8043700755065915e-08, "output_power": 3.5680603304654044e-08, "reference_active": true, "delta_db": -1.2920426423610403}], "active_band_count": 3, "active_band_mae_db": 0.5165802705519333}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8970444907887062, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "982325c1f1b9a7ed8123fa3f7923a55fb6c79926028f7bc83eee2330f4b3de1c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9662778044726987, "standard_deviation": 0.008253014231245115, "minimum": 0.9546712685242434, "maximum": 0.977240380175687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6057096665310406, "standard_deviation": 0.022792640107465536, "minimum": 1.5799812327697307, "maximum": 1.6525150420098436, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.9116549842749522, "standard_deviation": 0.4660448799529294, "minimum": 1.0983793596875842, "maximum": 2.5189864435542018, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2348647766517864, "standard_deviation": 0.02736665205025601, "minimum": 0.19280787297580515, "maximum": 0.26829628541661515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9894376474048079, "standard_deviation": 0.004276929625670694, "minimum": 0.982495976292515, "maximum": 0.9948560472217335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3263915402662088, "standard_deviation": 0.023902747759988685, "minimum": 1.2985339356640626, "maximum": 1.3517691772492522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 2.0264232921159615, "standard_deviation": 0.6831508212877567, "minimum": 1.049623084375332, "maximum": 3.2818867400789427, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1382023855265284, "standard_deviation": 0.029559073429227384, "minimum": 0.09037531528755073, "maximum": 0.18411088220812497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9829922085356724, "standard_deviation": 0.001922829588420612, "minimum": 0.9797659359653667, "maximum": 0.9851522348433163, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.622818755087121, "standard_deviation": 0.025548946861255906, "minimum": 1.58699633281844, "maximum": 1.6671863103577471, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.472810196854254, "standard_deviation": 0.22135335864533354, "minimum": 0.20911339907885704, "maximum": 0.6968045101922385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1683712180642727, "standard_deviation": 0.009320674559097297, "minimum": 0.15681700496469086, "maximum": 0.18364562719660965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994034191958219, "standard_deviation": 0.0010115786231848405, "minimum": 0.9922270417035067, "maximum": 0.9951675402930302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4117059395371363, "standard_deviation": 0.02726084206210163, "minimum": 1.379226522314007, "maximum": 1.4686943753596458, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6476036246709314, "standard_deviation": 0.23328165854449226, "minimum": 0.2531852690631877, "maximum": 0.9100534653419765, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09956827614553082, "standard_deviation": 0.008021658245761374, "minimum": 0.09002396071183763, "maximum": 0.11302008822235173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9820640112628619, "standard_deviation": 0.004360499038816745, "minimum": 0.9733507951832655, "maximum": 0.9864955913441921, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3035674436847602, "standard_deviation": 0.03448155483867042, "minimum": 1.252818371846539, "maximum": 1.3575879945692895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.46601074908983675, "standard_deviation": 0.15602559312690265, "minimum": 0.15630301046316275, "maximum": 0.6559966148880356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1555768719274273, "standard_deviation": 0.017705968441501335, "minimum": 0.1381990776146893, "maximum": 0.1902995313263248, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9944429159390642, "standard_deviation": 0.002208736166981978, "minimum": 0.9902526691377211, "maximum": 0.9966594680961219, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9934473346516517, "standard_deviation": 0.0354750562304174, "minimum": 0.9433105068253119, "maximum": 1.0506123430609278, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4763962975925531, "standard_deviation": 0.23858688500142194, "minimum": 0.1971634418602283, "maximum": 0.8705124874134528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08671061507248423, "standard_deviation": 0.015413582548214387, "minimum": 0.07159093624479886, "maximum": 0.1167987235217787, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2911091857392397, "standard_deviation": 0.09011682785140807, "minimum": 0.15369323847678212, "maximum": 0.4030074005706064, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.421483347433074, "standard_deviation": 0.008160109618795161, "minimum": 5.40691337617801, "maximum": 5.432051424572994, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5394693710995906, "standard_deviation": 0.0480118026381535, "minimum": 0.4668337056807469, "maximum": 0.5977976861982758, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11324472771869691, "standard_deviation": 0.036770068569035246, "minimum": -0.15428524504853391, "maximum": -0.03963444818787813, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900638910459075, "standard_deviation": 0.002345188716345495, "minimum": 0.9856239755872622, "maximum": 0.9930704354727874, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6955156489391063, "standard_deviation": 0.030528297871098703, "minimum": 0.6574447054563968, "maximum": 0.7525056817251734, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4361185104366072, "standard_deviation": 0.10716582272928166, "minimum": 0.3160525477186356, "maximum": 0.6294937544656086, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10067352166973961, "standard_deviation": 0.011539958215760602, "minimum": 0.08431521095017246, "maximum": 0.12146574211681992, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972017996492941, "standard_deviation": 0.0009856106343001862, "minimum": 0.9949347649698107, "maximum": 0.9981121742840161, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42872378084095175, "standard_deviation": 0.023956294883402122, "minimum": 0.3947759259788984, "maximum": 0.4746382100800232, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46732059715047924, "standard_deviation": 0.20229343675900907, "minimum": 0.07704650852460498, "maximum": 0.724830933714857, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05263031857958595, "standard_deviation": 0.008128682848434944, "minimum": 0.045516708385446115, "maximum": 0.07080882270949616, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0011", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.0509207248687744, "predicted_mos": 3.0509207248687744, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "73729de47718d407904574c2af38ef66e5709feb93a37d45c5191443a8485a23", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9427327001467347, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7146017809397696, "全帯域_10ms_rms_envelope_nrmse": 0.11192244290421814, "全帯域_10ms_envelope_correlation": 0.9877969345099947, "全帯域_10ms_modulation_mae_db": 0.29718838805738246, "全帯域_40ms_log_rms_mae_db": 0.4180018107653724, "全帯域_40ms_rms_envelope_nrmse": 0.061393757376735476, "全帯域_40ms_envelope_correlation": 0.9961947689419316, "全帯域_40ms_modulation_mae_db": 0.573679587674259, "300–1000 Hz_10ms_log_rms_mae_db": 1.276470096543674, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16102415939831197, "300–1000 Hz_10ms_envelope_correlation": 0.9807064911653013, "300–1000 Hz_10ms_modulation_mae_db": 0.5107305796343795, "300–1000 Hz_40ms_log_rms_mae_db": 0.9447233884954, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07495519648741038, "300–1000 Hz_40ms_envelope_correlation": 0.995680903872095, "300–1000 Hz_40ms_modulation_mae_db": 0.14056493233944908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5958034441756468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23404128617365486, "1000–3000 Hz_10ms_envelope_correlation": 0.9672712708916696, "1000–3000 Hz_10ms_modulation_mae_db": 2.172673559429344, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3255892531609703, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15475595511683002, "1000–3000 Hz_40ms_envelope_correlation": 0.9876021709850934, "1000–3000 Hz_40ms_modulation_mae_db": 2.911806376157982, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5298210637970797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14539241490793517, "3000–6000 Hz_10ms_envelope_correlation": 0.9873164208055208, "3000–6000 Hz_10ms_modulation_mae_db": 0.46708385259484775, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3992496134687813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08242778199660035, "3000–6000 Hz_40ms_envelope_correlation": 0.9960671176197712, "3000–6000 Hz_40ms_modulation_mae_db": 0.7204772917025085, "mr_stft_db": 5.331702379021852, "band_mae_db": 0.4144418258081308, "rms_gain_error_db": -0.11618014655448286, "output_peak": 0.5209947313030736}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6876564887437232, "全帯域_10ms_rms_envelope_nrmse": 0.0879889470288282, "全帯域_10ms_envelope_correlation": 0.9925694423778519, "全帯域_10ms_modulation_mae_db": 0.2902133926090132, "全帯域_40ms_log_rms_mae_db": 0.4047639901204441, "全帯域_40ms_rms_envelope_nrmse": 0.04236215269723739, "全帯域_40ms_envelope_correlation": 0.9983437968543807, "全帯域_40ms_modulation_mae_db": 0.38780495380072916, "300–1000 Hz_10ms_log_rms_mae_db": 1.2417874489547314, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13337867374611204, "300–1000 Hz_10ms_envelope_correlation": 0.9872030730470132, "300–1000 Hz_10ms_modulation_mae_db": 0.46722739496927906, "300–1000 Hz_40ms_log_rms_mae_db": 0.9514254667494426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06831133356907597, "300–1000 Hz_40ms_envelope_correlation": 0.997064510587665, "300–1000 Hz_40ms_modulation_mae_db": 0.6227989420456168, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6584862143269747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24298184266672027, "1000–3000 Hz_10ms_envelope_correlation": 0.9631625903806511, "1000–3000 Hz_10ms_modulation_mae_db": 1.745238977794516, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3929204586214503, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11193280549002028, "1000–3000 Hz_40ms_envelope_correlation": 0.9927282115774623, "1000–3000 Hz_40ms_modulation_mae_db": 1.7475049464070436, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5800264674650049, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16490551842978615, "3000–6000 Hz_10ms_envelope_correlation": 0.9836941667254294, "3000–6000 Hz_10ms_modulation_mae_db": 0.9301211133660852, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4087490642324378, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10178323371926094, "3000–6000 Hz_40ms_envelope_correlation": 0.9935789996167045, "3000–6000 Hz_40ms_modulation_mae_db": 1.1177448899299198, "mr_stft_db": 5.328481062962699, "band_mae_db": 0.30395760529033566, "rms_gain_error_db": -0.1322220361622549, "output_peak": 0.4528175840040277}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7416448170188018, "全帯域_10ms_rms_envelope_nrmse": 0.11339685940848285, "全帯域_10ms_envelope_correlation": 0.9875298085284187, "全帯域_10ms_modulation_mae_db": 0.7826032057378013, "全帯域_40ms_log_rms_mae_db": 0.46663590341375616, "全帯域_40ms_rms_envelope_nrmse": 0.05610564477066517, "全帯域_40ms_envelope_correlation": 0.9969992670526289, "全帯域_40ms_modulation_mae_db": 0.5416285609995144, "300–1000 Hz_10ms_log_rms_mae_db": 1.365245178235832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19792948312780267, "300–1000 Hz_10ms_envelope_correlation": 0.9711095496441168, "300–1000 Hz_10ms_modulation_mae_db": 0.8977838501451544, "300–1000 Hz_40ms_log_rms_mae_db": 1.0322639794115096, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11995642518803969, "300–1000 Hz_40ms_envelope_correlation": 0.9897364991366558, "300–1000 Hz_40ms_modulation_mae_db": 0.5600007748968038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.66939620922731, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21040471881425596, "1000–3000 Hz_10ms_envelope_correlation": 0.9724019152304335, "1000–3000 Hz_10ms_modulation_mae_db": 1.8280064878884383, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4026649984523447, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09729265516163806, "1000–3000 Hz_40ms_envelope_correlation": 0.9940470540267948, "1000–3000 Hz_40ms_modulation_mae_db": 2.289971566453069, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5740051285467211, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15335633934182846, "3000–6000 Hz_10ms_envelope_correlation": 0.9861253888467384, "3000–6000 Hz_10ms_modulation_mae_db": 0.7383726569594365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4106232650557833, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09026512733068214, "3000–6000 Hz_40ms_envelope_correlation": 0.9956199500342412, "3000–6000 Hz_40ms_modulation_mae_db": 0.7622233551507532, "mr_stft_db": 5.344267451444009, "band_mae_db": 0.28401210140544336, "rms_gain_error_db": -0.1410239526329569, "output_peak": 0.5328315140799201}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.669469553410015, "全帯域_10ms_rms_envelope_nrmse": 0.09635826540178626, "全帯域_10ms_envelope_correlation": 0.9911098582296225, "全帯域_10ms_modulation_mae_db": 0.55572302157432, "全帯域_40ms_log_rms_mae_db": 0.4246148207188078, "全帯域_40ms_rms_envelope_nrmse": 0.058911608040201804, "全帯域_40ms_envelope_correlation": 0.9966516067186578, "全帯域_40ms_modulation_mae_db": 0.7313263645656761, "300–1000 Hz_10ms_log_rms_mae_db": 1.2841287876147642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14303318847572577, "300–1000 Hz_10ms_envelope_correlation": 0.9852900213430955, "300–1000 Hz_10ms_modulation_mae_db": 0.6633802350575174, "300–1000 Hz_40ms_log_rms_mae_db": 0.9728250723296807, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07518000055673454, "300–1000 Hz_40ms_envelope_correlation": 0.9964379764755785, "300–1000 Hz_40ms_modulation_mae_db": 0.8757005226057747, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6375343603764305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2223086799408924, "1000–3000 Hz_10ms_envelope_correlation": 0.9706867305835237, "1000–3000 Hz_10ms_modulation_mae_db": 1.6624928754264132, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3698924472830725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14992883799402834, "1000–3000 Hz_40ms_envelope_correlation": 0.9883191053960536, "1000–3000 Hz_40ms_modulation_mae_db": 1.3499038498378635, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5616651700878192, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15632946835712705, "3000–6000 Hz_10ms_envelope_correlation": 0.9859934293910676, "3000–6000 Hz_10ms_modulation_mae_db": 1.013270859254406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4318843690166516, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09999755417205418, "3000–6000 Hz_40ms_envelope_correlation": 0.9949889634825314, "3000–6000 Hz_40ms_modulation_mae_db": 1.1423924627230873, "mr_stft_db": 5.329809941607995, "band_mae_db": 0.3922000128747951, "rms_gain_error_db": -0.15748188172359398, "output_peak": 0.5063805809839494}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.734263419190138, "全帯域_10ms_rms_envelope_nrmse": 0.11201792433018307, "全帯域_10ms_envelope_correlation": 0.987806647561063, "全帯域_10ms_modulation_mae_db": 0.34765760101378107, "全帯域_40ms_log_rms_mae_db": 0.4652645818124959, "全帯域_40ms_rms_envelope_nrmse": 0.05505055741582487, "全帯域_40ms_envelope_correlation": 0.9969669297738578, "全帯域_40ms_modulation_mae_db": 0.08074193359130717, "300–1000 Hz_10ms_log_rms_mae_db": 1.3174322952957547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1682454145141177, "300–1000 Hz_10ms_envelope_correlation": 0.9792306556634082, "300–1000 Hz_10ms_modulation_mae_db": 0.44641761895654214, "300–1000 Hz_40ms_log_rms_mae_db": 1.01653798327639, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10024155574980086, "300–1000 Hz_40ms_envelope_correlation": 0.9927529269085738, "300–1000 Hz_40ms_modulation_mae_db": 0.2998013787843496, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6309398191669477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.270060652999524, "1000–3000 Hz_10ms_envelope_correlation": 0.9541069374472364, "1000–3000 Hz_10ms_modulation_mae_db": 1.2204249970705752, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3765118149859688, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1581796568183972, "1000–3000 Hz_40ms_envelope_correlation": 0.9842762453244938, "1000–3000 Hz_40ms_modulation_mae_db": 1.2686616551300434, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5452189294337864, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15892497313955173, "3000–6000 Hz_10ms_envelope_correlation": 0.9847167860767644, "3000–6000 Hz_10ms_modulation_mae_db": 0.6732927319908271, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3891426317709603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09073660465213192, "3000–6000 Hz_40ms_envelope_correlation": 0.9950267841521934, "3000–6000 Hz_40ms_modulation_mae_db": 0.7506749832502662, "mr_stft_db": 5.328395355623241, "band_mae_db": 0.2534068608420679, "rms_gain_error_db": -0.11399459730612962, "output_peak": 0.5425232480594325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7109219999935401, "全帯域_10ms_rms_envelope_nrmse": 0.10583738356651248, "全帯域_10ms_envelope_correlation": 0.9891199166655066, "全帯域_10ms_modulation_mae_db": 0.5531863986969322, "全帯域_40ms_log_rms_mae_db": 0.4501963259095705, "全帯域_40ms_rms_envelope_nrmse": 0.055063179273144996, "全帯域_40ms_envelope_correlation": 0.9969596213160012, "全帯域_40ms_modulation_mae_db": 0.3224625713307558, "300–1000 Hz_10ms_log_rms_mae_db": 1.3444442824234286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16066412155189766, "300–1000 Hz_10ms_envelope_correlation": 0.9809457281705853, "300–1000 Hz_10ms_modulation_mae_db": 0.29556023065359155, "300–1000 Hz_40ms_log_rms_mae_db": 1.0544371393891319, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08544169104054217, "300–1000 Hz_40ms_envelope_correlation": 0.9943145168031534, "300–1000 Hz_40ms_modulation_mae_db": 0.3284314042408018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.649168308756009, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2613756665389959, "1000–3000 Hz_10ms_envelope_correlation": 0.964029416646276, "1000–3000 Hz_10ms_modulation_mae_db": 2.3976811667770517, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4002629464393395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19395166696365843, "1000–3000 Hz_40ms_envelope_correlation": 0.9857255175478956, "1000–3000 Hz_40ms_modulation_mae_db": 2.680813383096434, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5744806649063952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16457005480976863, "3000–6000 Hz_10ms_envelope_correlation": 0.9836401225108157, "3000–6000 Hz_10ms_modulation_mae_db": 0.9667802798556405, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4271716270269208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0864246064217874, "3000–6000 Hz_40ms_envelope_correlation": 0.9953372522855547, "3000–6000 Hz_40ms_modulation_mae_db": 0.9875725078302161, "mr_stft_db": 5.330459727564951, "band_mae_db": 0.4649370902236623, "rms_gain_error_db": -0.11478309871364437, "output_peak": 0.5445419403326197}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6895259731714038, "全帯域_10ms_rms_envelope_nrmse": 0.08677798524512456, "全帯域_10ms_envelope_correlation": 0.9926625293729715, "全帯域_10ms_modulation_mae_db": 0.22790232313256753, "全帯域_40ms_log_rms_mae_db": 0.42744695204457506, "全帯域_40ms_rms_envelope_nrmse": 0.04884609730210909, "全帯域_40ms_envelope_correlation": 0.9975233723000164, "全帯域_40ms_modulation_mae_db": 0.39482169085766233, "300–1000 Hz_10ms_log_rms_mae_db": 1.3085773941651957, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14657424928852603, "300–1000 Hz_10ms_envelope_correlation": 0.9840783121388537, "300–1000 Hz_10ms_modulation_mae_db": 0.5416871149934799, "300–1000 Hz_40ms_log_rms_mae_db": 1.0136692365619924, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09332316602218142, "300–1000 Hz_40ms_envelope_correlation": 0.9933088140909195, "300–1000 Hz_40ms_modulation_mae_db": 0.734830968697659, "1000–3000 Hz_10ms_log_rms_mae_db": 1.641415529147795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23883668690439452, "1000–3000 Hz_10ms_envelope_correlation": 0.9643253714247713, "1000–3000 Hz_10ms_modulation_mae_db": 2.2776793619199664, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3991492521984812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13026787485789804, "1000–3000 Hz_40ms_envelope_correlation": 0.9892564274172279, "1000–3000 Hz_40ms_modulation_mae_db": 2.010231341252234, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5147529771598427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14575726596854605, "3000–6000 Hz_10ms_envelope_correlation": 0.9872517489447051, "3000–6000 Hz_10ms_modulation_mae_db": 0.36311143766853066, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3458819253151537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07837421176038126, "3000–6000 Hz_40ms_envelope_correlation": 0.9961917590277755, "3000–6000 Hz_40ms_modulation_mae_db": 0.43678373458265823, "mr_stft_db": 5.33458277945094, "band_mae_db": 0.16597996404342075, "rms_gain_error_db": -0.05633622224582899, "output_peak": 0.4848314194386004}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7129318099652571, "全帯域_10ms_rms_envelope_nrmse": 0.09641206853757578, "全帯域_10ms_envelope_correlation": 0.9909766310282595, "全帯域_10ms_modulation_mae_db": 0.6074616502841396, "全帯域_40ms_log_rms_mae_db": 0.4254683691434518, "全帯域_40ms_rms_envelope_nrmse": 0.048594745088362624, "全帯域_40ms_envelope_correlation": 0.9976444017389751, "全帯域_40ms_modulation_mae_db": 0.7786761637497225, "300–1000 Hz_10ms_log_rms_mae_db": 1.324694943535076, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13587798026512352, "300–1000 Hz_10ms_envelope_correlation": 0.9863553967313716, "300–1000 Hz_10ms_modulation_mae_db": 0.49886928337375713, "300–1000 Hz_40ms_log_rms_mae_db": 0.9820893480941476, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07359648656556256, "300–1000 Hz_40ms_envelope_correlation": 0.9958328742544461, "300–1000 Hz_40ms_modulation_mae_db": 0.7510176956694474, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6199227832401315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2315576360817716, "1000–3000 Hz_10ms_envelope_correlation": 0.9702467555109773, "1000–3000 Hz_10ms_modulation_mae_db": 1.2245987518248393, "1000–3000 Hz_40ms_log_rms_mae_db": 1.347123211876871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1498603211148427, "1000–3000 Hz_40ms_envelope_correlation": 0.9915489559202084, "1000–3000 Hz_40ms_modulation_mae_db": 1.2313001949839475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5868078852785092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14963296800937922, "3000–6000 Hz_10ms_envelope_correlation": 0.98643366284029, "3000–6000 Hz_10ms_modulation_mae_db": 0.6400402142942495, "3000–6000 Hz_40ms_log_rms_mae_db": 1.431140864161948, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08942298734121713, "3000–6000 Hz_40ms_envelope_correlation": 0.9950546545370643, "3000–6000 Hz_40ms_modulation_mae_db": 0.8516130634201636, "mr_stft_db": 5.323048169855028, "band_mae_db": 0.44207834905722354, "rms_gain_error_db": -0.1072507153549182, "output_peak": 0.5316168414586944}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.413127453184108, "spectral_convergence": 0.3121303445665241, "spectral_convergence_status": "ok", "amplitude_floor": 1.6898412267444875e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.125431934574202, "spectral_convergence": 0.3148510717885902, "spectral_convergence_status": "ok", "amplitude_floor": 1.0945626100175482e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.456547749307247, "spectral_convergence": 0.34380396281746145, "spectral_convergence_status": "ok", "amplitude_floor": 6.58943811507865e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017588563541266742, "output_power": 0.0017109132118818648, "delta_db": -0.12002392063903411, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012071674288590822, "output_power": 0.0012171333171116672, "delta_db": 0.03570641689608924, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003927549262280726, "output_power": 0.0004243355115597691, "delta_db": 0.33587737293543296, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003207071656425, "output_power": 0.00023429500827664467, "delta_db": -1.3634562766473541, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.4115036625953016e-05, "output_power": 4.9978448238951234e-05, "delta_db": -0.345351889582996, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.305762815858983e-05, "output_power": 4.0311280614337425e-05, "delta_db": -0.2862350781478786, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4144418258081308, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11618014655448286, "pitch": null, "output_rms_dbfs": -24.34510294072946, "output_peak": 0.5209947313030736, "reference_rms_dbfs": -24.228922794174977, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.331702379021852}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002198976712894503, "log_rms_mae_db": 0.7146017809397696, "rms_envelope_nrmse": 0.11192244290421814, "envelope_correlation": 0.9877969345099947, "rms_gain_error_db": -0.11493731953121386, "reference_mean_power": 0.0037766562988771884, "output_mean_power": 0.0036780169891092694}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.069519300574134e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006799849644412746, "output_power": 0.0006034028470076034, "reference_active": true, "delta_db": -0.5189195490923313}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022275189153607582, "output_power": 0.00020638813059176917, "reference_active": true, "delta_db": -0.33136683269801026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.238302101062001e-05, "output_power": 4.198208767050498e-05, "reference_active": true, "delta_db": -0.04127878238180577}], "active_band_count": 3, "active_band_mae_db": 0.29718838805738246}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018113949472991762, "log_rms_mae_db": 0.4180018107653724, "rms_envelope_nrmse": 0.061393757376735476, "envelope_correlation": 0.9961947689419316, "rms_gain_error_db": -0.11608570222245126, "reference_mean_power": 0.003780325824366865, "output_mean_power": 0.0036806172978504437}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9093794448028363e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006511405512347742, "output_power": 0.0005755136821569102, "reference_active": true, "delta_db": -0.5361908992369264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001166725476043267, "output_power": 0.00010382400767931643, "reference_active": true, "delta_db": -0.5067089207405584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.399065253982206e-06, "output_power": 2.9076732950418116e-06, "reference_active": true, "delta_db": -0.6781389430452923}], "active_band_count": 3, "active_band_mae_db": 0.573679587674259}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013526658282762358, "log_rms_mae_db": 1.276470096543674, "rms_envelope_nrmse": 0.16102415939831197, "envelope_correlation": 0.9807064911653013, "rms_gain_error_db": -0.1238097712760623, "reference_mean_power": 0.001047203990993015, "output_mean_power": 0.0010177715612044278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.584657585149468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003504540347149448, "output_power": 0.0003281936240502753, "reference_active": true, "delta_db": -0.2850092471823628}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.5576057457372e-05, "output_power": 9.29619394880822e-05, "reference_active": true, "delta_db": -0.12043935834741404}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9623586773434205e-05, "output_power": 2.5436227662871582e-05, "reference_active": true, "delta_db": 1.1267431333733617}], "active_band_count": 3, "active_band_mae_db": 0.5107305796343795}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012031399971827572, "log_rms_mae_db": 0.9447233884954, "rms_envelope_nrmse": 0.07495519648741038, "envelope_correlation": 0.995680903872095, "rms_gain_error_db": -0.12389335978767908, "reference_mean_power": 0.0010510881038568295, "output_mean_power": 0.0010215268467695451}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.061243701565635e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00034327462417827815, "output_power": 0.00032650974392450506, "reference_active": true, "delta_db": -0.2174555403375617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.625199841508274e-05, "output_power": 4.472626537570734e-05, "reference_active": true, "delta_db": -0.14567865808438313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5417688576161987e-06, "output_power": 1.5211190635314367e-06, "reference_active": true, "delta_db": -0.05856059859640246}], "active_band_count": 3, "active_band_mae_db": 0.14056493233944908}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013601220330575488, "log_rms_mae_db": 1.5958034441756468, "rms_envelope_nrmse": 0.23404128617365486, "envelope_correlation": 0.9672712708916696, "rms_gain_error_db": -0.6958290242463171, "reference_mean_power": 0.00024923569921471797, "output_mean_power": 0.00021233781598675022}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.70738105242738e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001701759171968848, "output_power": 0.0001342835616438741, "reference_active": true, "delta_db": -1.0287524834547925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6462817859538857e-05, "output_power": 1.1165850815319855e-05, "reference_active": true, "delta_db": -3.747442665777634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.097841074264338e-06, "output_power": 4.626384099711215e-06, "reference_active": true, "delta_db": 1.741825529055604}], "active_band_count": 3, "active_band_mae_db": 2.172673559429344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011511226838452259, "log_rms_mae_db": 1.3255892531609703, "rms_envelope_nrmse": 0.15475595511683002, "envelope_correlation": 0.9876021709850934, "rms_gain_error_db": -0.6980117106603211, "reference_mean_power": 0.00024963838926192667, "output_mean_power": 0.00021257402737795514}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6223780118670214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016948627965662844, "output_power": 0.00013175799081034962, "reference_active": true, "delta_db": -1.093575831084621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1767373976886565e-05, "output_power": 4.1234803022272745e-06, "reference_active": true, "delta_db": -4.554156316220938}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8952968721807594e-07, "output_power": 3.8587437938946115e-07, "reference_active": true, "delta_db": 3.087686981168388}], "active_band_count": 3, "active_band_mae_db": 2.911806376157982}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.3800826555528964e-05, "log_rms_mae_db": 1.5298210637970797, "rms_envelope_nrmse": 0.14539241490793517, "envelope_correlation": 0.9873164208055208, "rms_gain_error_db": -0.2185504845098623, "reference_mean_power": 4.104766861392576e-05, "output_mean_power": 3.9033136206360444e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.890895346126465e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.549789437651056e-06, "output_power": 9.587984970307967e-06, "reference_active": true, "delta_db": 0.017335486078785496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.275656140671785e-06, "output_power": 3.149443542141589e-06, "reference_active": true, "delta_db": -0.17064478308458408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.939546621487433e-07, "output_power": 4.4918595326071573e-07, "reference_active": true, "delta_db": -1.2132712886211736}], "active_band_count": 3, "active_band_mae_db": 0.46708385259484775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.772714769720092e-05, "log_rms_mae_db": 1.3992496134687813, "rms_envelope_nrmse": 0.08242778199660035, "envelope_correlation": 0.9960671176197712, "rms_gain_error_db": -0.21904679196951765, "reference_mean_power": 4.122071638798201e-05, "output_mean_power": 3.919321194552364e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.778938094011469e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.936665760227185e-06, "output_power": 9.942364344418607e-06, "reference_active": true, "delta_db": 0.00248992402925024}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7465237791634235e-06, "output_power": 1.6485505615971723e-06, "reference_active": true, "delta_db": -0.250722313530353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8043700755065915e-08, "output_power": 3.0960965076437776e-08, "reference_active": true, "delta_db": -1.9082196375479223}], "active_band_count": 3, "active_band_mae_db": 0.7204772917025085}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8987590506149186, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7ef728192d2871a45250a7f780d95e9e02497ad72d4f70a8e5cc2aae25e6d42d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9657788735144424, "standard_deviation": 0.005835767285825323, "minimum": 0.9541069374472364, "maximum": 0.9724019152304335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6378333335521558, "standard_deviation": 0.02298751130617767, "minimum": 1.5958034441756468, "maximum": 1.66939620922731, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.8160995222663932, "standard_deviation": 0.4491098334534085, "minimum": 1.2204249970705752, "maximum": 2.3976811667770517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23894589626502616, "standard_deviation": 0.019484649038989103, "minimum": 0.21040471881425596, "maximum": 0.270060652999524, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9891879610244038, "standard_deviation": 0.003403910554174143, "minimum": 0.9842762453244938, "maximum": 0.9940470540267948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3767642978773122, "standard_deviation": 0.028106699109123237, "minimum": 1.3255892531609703, "maximum": 1.4026649984523447, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.9362741641648271, "standard_deviation": 0.6501353687755984, "minimum": 1.2313001949839475, "maximum": 2.911806376157982, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14327122168966414, "standard_deviation": 0.02994189634667088, "minimum": 0.09729265516163806, "maximum": 0.19395166696365843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9856464657676665, "standard_deviation": 0.001465620461102841, "minimum": 0.9836401225108157, "maximum": 0.9873164208055208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5583472858343947, "standard_deviation": 0.02586508342979272, "minimum": 1.5147529771598427, "maximum": 1.5868078852785092, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7240091432480029, "standard_deviation": 0.23639534505297694, "minimum": 0.36311143766853066, "maximum": 1.013270859254406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15485862537049033, "standard_deviation": 0.0077086104623493475, "minimum": 0.14539241490793517, "maximum": 0.16490551842978615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9952331850944796, "standard_deviation": 0.0008146140927503962, "minimum": 0.9935789996167045, "maximum": 0.9961917590277755, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4054804200060795, "standard_deviation": 0.02858819714967113, "minimum": 1.3458819253151537, "maximum": 1.4318843690166516, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8461852860736966, "standard_deviation": 0.23332597405773345, "minimum": 0.43678373458265823, "maximum": 1.1423924627230873, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0899290134242644, "standard_deviation": 0.007973008690042142, "minimum": 0.07837421176038126, "maximum": 0.10178323371926094, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9818649034879683, "standard_deviation": 0.005211181580691025, "minimum": 0.9711095496441168, "maximum": 0.9872030730470132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3078475533460572, "standard_deviation": 0.03949883264023432, "minimum": 1.2417874489547314, "maximum": 1.365245178235832, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5402070384729627, "standard_deviation": 0.17711465979765711, "minimum": 0.29556023065359155, "maximum": 0.8977838501451544, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1558409087959522, "standard_deviation": 0.021140443692751025, "minimum": 0.13337867374611204, "maximum": 0.19792948312780267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9943911277661359, "standard_deviation": 0.002408474993172285, "minimum": 0.9897364991366558, "maximum": 0.997064510587665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9959964517884619, "standard_deviation": 0.03929876306784575, "minimum": 0.9447233884954, "maximum": 1.0544371393891319, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5391433274099878, "standard_deviation": 0.2575814050094354, "minimum": 0.14056493233944908, "maximum": 0.8757005226057747, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08637573189741846, "standard_deviation": 0.017380070416366925, "minimum": 0.06831133356907597, "maximum": 0.11995642518803969, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3401267261931349, "standard_deviation": 0.10454705558903149, "minimum": 0.16597996404342075, "maximum": 0.4649370902236623, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.331343358441339, "standard_deviation": 0.006169981530556402, "minimum": 5.323048169855028, "maximum": 5.344267451444009, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5145672324575398, "standard_deviation": 0.03186997407904398, "minimum": 0.4528175840040277, "maximum": 0.5445419403326197, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11740908133672623, "standard_deviation": 0.029847388953604635, "minimum": -0.15748188172359398, "maximum": -0.05633622224582899, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899464710342111, "standard_deviation": 0.0021512273076880525, "minimum": 0.9875298085284187, "maximum": 0.9926625293729715, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7076269803040811, "standard_deviation": 0.024309949542261844, "minimum": 0.669469553410015, "maximum": 0.7416448170188018, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45774199763824214, "standard_deviation": 0.1947338253216632, "minimum": 0.22790232313256753, "maximum": 0.7826032057378013, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1013389845528389, "standard_deviation": 0.010895644952909149, "minimum": 0.08677798524512456, "maximum": 0.11339685940848285, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971604705870563, "standard_deviation": 0.0006613249933206022, "minimum": 0.9961947689419316, "maximum": 0.9983437968543807, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4352990942410592, "standard_deviation": 0.022696428790567585, "minimum": 0.4047639901204441, "maximum": 0.46663590341375616, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4763927283212034, "standard_deviation": 0.2282925922207795, "minimum": 0.08074193359130717, "maximum": 0.7786761637497225, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05329096774553518, "standard_deviation": 0.006241067585599883, "minimum": 0.04236215269723739, "maximum": 0.061393757376735476, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0011", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.8594892024993896, "predicted_mos": 2.8594892024993896, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4ee1b296afb9316440ffc16db95f9883743ccdcebc85aa286c472775ae591f9f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.2225015996955335, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7148800983171492, "全帯域_10ms_rms_envelope_nrmse": 0.10560839729516701, "全帯域_10ms_envelope_correlation": 0.989151842871253, "全帯域_10ms_modulation_mae_db": 0.43001533619219573, "全帯域_40ms_log_rms_mae_db": 0.4259665842105792, "全帯域_40ms_rms_envelope_nrmse": 0.05855515157200543, "全帯域_40ms_envelope_correlation": 0.9965506800129137, "全帯域_40ms_modulation_mae_db": 0.4171549715027682, "300–1000 Hz_10ms_log_rms_mae_db": 1.432443939373493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16937501857177212, "300–1000 Hz_10ms_envelope_correlation": 0.9787840467259881, "300–1000 Hz_10ms_modulation_mae_db": 0.5777680754687115, "300–1000 Hz_40ms_log_rms_mae_db": 1.1072888059479367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09001220992359553, "300–1000 Hz_40ms_envelope_correlation": 0.9941751757913553, "300–1000 Hz_40ms_modulation_mae_db": 0.4556672664663166, "1000–3000 Hz_10ms_log_rms_mae_db": 1.909323210357812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21672437432864736, "1000–3000 Hz_10ms_envelope_correlation": 0.9706380323994294, "1000–3000 Hz_10ms_modulation_mae_db": 1.6095525358881912, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5395541000791741, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1355066889376152, "1000–3000 Hz_40ms_envelope_correlation": 0.9881841887843598, "1000–3000 Hz_40ms_modulation_mae_db": 1.3008276467246551, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1063566602769606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23447830316434967, "3000–6000 Hz_10ms_envelope_correlation": 0.9689701527881635, "3000–6000 Hz_10ms_modulation_mae_db": 0.7180130076495197, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7551049902594114, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1483418131417778, "3000–6000 Hz_40ms_envelope_correlation": 0.9877909227025565, "3000–6000 Hz_40ms_modulation_mae_db": 0.373611033616593, "mr_stft_db": 5.792030023740753, "band_mae_db": 0.24615331586922803, "rms_gain_error_db": -0.1113771177795046, "output_peak": 0.5150319456572028}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6709171813706198, "全帯域_10ms_rms_envelope_nrmse": 0.10525245417535257, "全帯域_10ms_envelope_correlation": 0.9893147901678869, "全帯域_10ms_modulation_mae_db": 0.23553872616664326, "全帯域_40ms_log_rms_mae_db": 0.4344430571491878, "全帯域_40ms_rms_envelope_nrmse": 0.058534860155048336, "全帯域_40ms_envelope_correlation": 0.9966702947041733, "全帯域_40ms_modulation_mae_db": 0.4193527645763739, "300–1000 Hz_10ms_log_rms_mae_db": 1.4243824409381272, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16753162651051287, "300–1000 Hz_10ms_envelope_correlation": 0.9802037379000161, "300–1000 Hz_10ms_modulation_mae_db": 0.6270173538968199, "300–1000 Hz_40ms_log_rms_mae_db": 1.1307668780714484, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10548695468279239, "300–1000 Hz_40ms_envelope_correlation": 0.9930732788557388, "300–1000 Hz_40ms_modulation_mae_db": 0.751680402635435, "1000–3000 Hz_10ms_log_rms_mae_db": 1.82165951162252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23193382349760952, "1000–3000 Hz_10ms_envelope_correlation": 0.9673533092093488, "1000–3000 Hz_10ms_modulation_mae_db": 0.43914017770436714, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4801173767417477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14436253212380226, "1000–3000 Hz_40ms_envelope_correlation": 0.9869678307298723, "1000–3000 Hz_40ms_modulation_mae_db": 0.4248687020359739, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1033270460606888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20116666772296465, "3000–6000 Hz_10ms_envelope_correlation": 0.9760474944617646, "3000–6000 Hz_10ms_modulation_mae_db": 0.34859476214424584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.754641362729311, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12389529745841575, "3000–6000 Hz_40ms_envelope_correlation": 0.9908621134311667, "3000–6000 Hz_40ms_modulation_mae_db": 0.5225038171564492, "mr_stft_db": 5.789399751436515, "band_mae_db": 0.14408844663944037, "rms_gain_error_db": -0.15234724126451085, "output_peak": 0.4919000839132015}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7324571913131317, "全帯域_10ms_rms_envelope_nrmse": 0.1061886576516549, "全帯域_10ms_envelope_correlation": 0.9889969987347471, "全帯域_10ms_modulation_mae_db": 0.7625351126427796, "全帯域_40ms_log_rms_mae_db": 0.42436635365721126, "全帯域_40ms_rms_envelope_nrmse": 0.04637218911196044, "全帯域_40ms_envelope_correlation": 0.9977845523622472, "全帯域_40ms_modulation_mae_db": 0.5950114308024578, "300–1000 Hz_10ms_log_rms_mae_db": 1.4440419393153758, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.167445352294796, "300–1000 Hz_10ms_envelope_correlation": 0.979146839954228, "300–1000 Hz_10ms_modulation_mae_db": 0.2924549867695397, "300–1000 Hz_40ms_log_rms_mae_db": 1.0726216481566027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10116650013693551, "300–1000 Hz_40ms_envelope_correlation": 0.9921383936991541, "300–1000 Hz_40ms_modulation_mae_db": 0.33943920974531333, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8163719930290663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26119848008853486, "1000–3000 Hz_10ms_envelope_correlation": 0.9570631050007428, "1000–3000 Hz_10ms_modulation_mae_db": 2.9568726341746263, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4583498326975763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12191298263757404, "1000–3000 Hz_40ms_envelope_correlation": 0.9908274516197184, "1000–3000 Hz_40ms_modulation_mae_db": 2.9415642587945356, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0292037361941944, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2160729425330072, "3000–6000 Hz_10ms_envelope_correlation": 0.9719625253751519, "3000–6000 Hz_10ms_modulation_mae_db": 0.7759208557091757, "3000–6000 Hz_40ms_log_rms_mae_db": 1.698353868656032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1409853528670802, "3000–6000 Hz_40ms_envelope_correlation": 0.987967034909529, "3000–6000 Hz_40ms_modulation_mae_db": 1.230946127523976, "mr_stft_db": 5.763662328208679, "band_mae_db": 0.2410248247402861, "rms_gain_error_db": -0.059589166624349585, "output_peak": 0.5135784530576605}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.703370442211774, "全帯域_10ms_rms_envelope_nrmse": 0.1029438256911764, "全帯域_10ms_envelope_correlation": 0.9896541335528208, "全帯域_10ms_modulation_mae_db": 0.38090885756574316, "全帯域_40ms_log_rms_mae_db": 0.4249185609222968, "全帯域_40ms_rms_envelope_nrmse": 0.0437215470347976, "全帯域_40ms_envelope_correlation": 0.9980255538513786, "全帯域_40ms_modulation_mae_db": 0.18219979808580644, "300–1000 Hz_10ms_log_rms_mae_db": 1.4324396917458484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15209246410470958, "300–1000 Hz_10ms_envelope_correlation": 0.9830427022231356, "300–1000 Hz_10ms_modulation_mae_db": 0.2612283494360613, "300–1000 Hz_40ms_log_rms_mae_db": 1.0421822038770907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07924400653495012, "300–1000 Hz_40ms_envelope_correlation": 0.9956431722321611, "300–1000 Hz_40ms_modulation_mae_db": 0.15106625051071143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.857457468086634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20621286188590304, "1000–3000 Hz_10ms_envelope_correlation": 0.9734459860298275, "1000–3000 Hz_10ms_modulation_mae_db": 0.9746235313584264, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4633022691860167, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10355387207654816, "1000–3000 Hz_40ms_envelope_correlation": 0.9934790835065853, "1000–3000 Hz_40ms_modulation_mae_db": 1.0765916078488207, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0566250958215893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2098508045436493, "3000–6000 Hz_10ms_envelope_correlation": 0.9747075461583286, "3000–6000 Hz_10ms_modulation_mae_db": 0.4887441455320894, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7193945725982935, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13404120103955044, "3000–6000 Hz_40ms_envelope_correlation": 0.9897338002873788, "3000–6000 Hz_40ms_modulation_mae_db": 0.9449024969039433, "mr_stft_db": 5.7852946537176, "band_mae_db": 0.2516902991354552, "rms_gain_error_db": -0.05631704776669386, "output_peak": 0.4787786141304103}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6556724188079517, "全帯域_10ms_rms_envelope_nrmse": 0.10028792742884046, "全帯域_10ms_envelope_correlation": 0.9902093467274108, "全帯域_10ms_modulation_mae_db": 0.18351239200588795, "全帯域_40ms_log_rms_mae_db": 0.39839530742654283, "全帯域_40ms_rms_envelope_nrmse": 0.049712578354781704, "全帯域_40ms_envelope_correlation": 0.997487090089855, "全帯域_40ms_modulation_mae_db": 0.23443990162371217, "300–1000 Hz_10ms_log_rms_mae_db": 1.3617312447420182, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1581237921462431, "300–1000 Hz_10ms_envelope_correlation": 0.9814846843190234, "300–1000 Hz_10ms_modulation_mae_db": 0.4220506835158937, "300–1000 Hz_40ms_log_rms_mae_db": 1.008271900253798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09151514711245146, "300–1000 Hz_40ms_envelope_correlation": 0.9934571319961321, "300–1000 Hz_40ms_modulation_mae_db": 0.25091473266536723, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8802212510572542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24101478620262243, "1000–3000 Hz_10ms_envelope_correlation": 0.9637194599286131, "1000–3000 Hz_10ms_modulation_mae_db": 1.416466364538122, "1000–3000 Hz_40ms_log_rms_mae_db": 1.568853740500104, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12965118946947268, "1000–3000 Hz_40ms_envelope_correlation": 0.9902630001134514, "1000–3000 Hz_40ms_modulation_mae_db": 1.5447940909461826, "3000–6000 Hz_10ms_log_rms_mae_db": 2.08506068801146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24035171723385362, "3000–6000 Hz_10ms_envelope_correlation": 0.9653151106595184, "3000–6000 Hz_10ms_modulation_mae_db": 0.7394723999794565, "3000–6000 Hz_40ms_log_rms_mae_db": 1.772163216305087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14999194008319083, "3000–6000 Hz_40ms_envelope_correlation": 0.9863665318972122, "3000–6000 Hz_40ms_modulation_mae_db": 0.3015051220431965, "mr_stft_db": 5.780566354516128, "band_mae_db": 0.3625541400704382, "rms_gain_error_db": -0.08708031969300478, "output_peak": 0.5018542560370293}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6755428714336527, "全帯域_10ms_rms_envelope_nrmse": 0.09133841558118119, "全帯域_10ms_envelope_correlation": 0.9922082122603548, "全帯域_10ms_modulation_mae_db": 0.4265971814369532, "全帯域_40ms_log_rms_mae_db": 0.43959680108615834, "全帯域_40ms_rms_envelope_nrmse": 0.05198696795077602, "全帯域_40ms_envelope_correlation": 0.9976680436442972, "全帯域_40ms_modulation_mae_db": 0.6223889047319869, "300–1000 Hz_10ms_log_rms_mae_db": 1.4762522829164189, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1339297230293694, "300–1000 Hz_10ms_envelope_correlation": 0.9871012679395761, "300–1000 Hz_10ms_modulation_mae_db": 0.4527434799626963, "300–1000 Hz_40ms_log_rms_mae_db": 1.1211293974402583, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07545388708849496, "300–1000 Hz_40ms_envelope_correlation": 0.9962476054503793, "300–1000 Hz_40ms_modulation_mae_db": 0.7641871225175155, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8309778245145585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20762040119425834, "1000–3000 Hz_10ms_envelope_correlation": 0.97306082067993, "1000–3000 Hz_10ms_modulation_mae_db": 2.0663253908951393, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4937593793115875, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10319249986216836, "1000–3000 Hz_40ms_envelope_correlation": 0.9934970559328724, "1000–3000 Hz_40ms_modulation_mae_db": 2.2583394425903047, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1137943410422264, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23459099400709157, "3000–6000 Hz_10ms_envelope_correlation": 0.9671944067125533, "3000–6000 Hz_10ms_modulation_mae_db": 0.8230013822892254, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7968265443833704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15726707934291043, "3000–6000 Hz_40ms_envelope_correlation": 0.98512104534418, "3000–6000 Hz_40ms_modulation_mae_db": 0.5065807315977895, "mr_stft_db": 5.77387010436965, "band_mae_db": 0.22368010772777788, "rms_gain_error_db": -0.2016280871023055, "output_peak": 0.49367294928065436}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6658690472480154, "全帯域_10ms_rms_envelope_nrmse": 0.10204503428139304, "全帯域_10ms_envelope_correlation": 0.9898568700248321, "全帯域_10ms_modulation_mae_db": 0.4721570342865216, "全帯域_40ms_log_rms_mae_db": 0.41786464359808456, "全帯域_40ms_rms_envelope_nrmse": 0.06702341891028885, "全帯域_40ms_envelope_correlation": 0.9953390060075182, "全帯域_40ms_modulation_mae_db": 0.37388692468679746, "300–1000 Hz_10ms_log_rms_mae_db": 1.377833864678414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.158487091110227, "300–1000 Hz_10ms_envelope_correlation": 0.9813646176614311, "300–1000 Hz_10ms_modulation_mae_db": 0.55380882388336, "300–1000 Hz_40ms_log_rms_mae_db": 1.0665855558483892, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10280768967245568, "300–1000 Hz_40ms_envelope_correlation": 0.9916935225047225, "300–1000 Hz_40ms_modulation_mae_db": 0.16163378001614037, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8480650722830023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21950412321031973, "1000–3000 Hz_10ms_envelope_correlation": 0.971534384578708, "1000–3000 Hz_10ms_modulation_mae_db": 1.8620295465178502, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4722390291018106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1257766982685148, "1000–3000 Hz_40ms_envelope_correlation": 0.9930830310287252, "1000–3000 Hz_40ms_modulation_mae_db": 1.9286972190370024, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1086126200196915, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21897530891775763, "3000–6000 Hz_10ms_envelope_correlation": 0.9722001588374529, "3000–6000 Hz_10ms_modulation_mae_db": 0.877414024513468, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7836026577151558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13485988999300672, "3000–6000 Hz_40ms_envelope_correlation": 0.9895553877609532, "3000–6000 Hz_40ms_modulation_mae_db": 0.4627182983664151, "mr_stft_db": 5.78469924188878, "band_mae_db": 0.3611741625675666, "rms_gain_error_db": -0.07724665249574515, "output_peak": 0.4846607926956733}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6877071790147826, "全帯域_10ms_rms_envelope_nrmse": 0.10248806668809393, "全帯域_10ms_envelope_correlation": 0.9898990747453317, "全帯域_10ms_modulation_mae_db": 0.36607571599644206, "全帯域_40ms_log_rms_mae_db": 0.4336034989347201, "全帯域_40ms_rms_envelope_nrmse": 0.06128431020754696, "全帯域_40ms_envelope_correlation": 0.9963638866905128, "全帯域_40ms_modulation_mae_db": 0.43830070925101633, "300–1000 Hz_10ms_log_rms_mae_db": 1.4746834220351792, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1556719277640839, "300–1000 Hz_10ms_envelope_correlation": 0.9828103747863591, "300–1000 Hz_10ms_modulation_mae_db": 0.4552433920206007, "300–1000 Hz_40ms_log_rms_mae_db": 1.172087715399659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.101755199284401, "300–1000 Hz_40ms_envelope_correlation": 0.9931973505803757, "300–1000 Hz_40ms_modulation_mae_db": 0.6736725405571886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8267928172051022, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23448829424355963, "1000–3000 Hz_10ms_envelope_correlation": 0.9656610438684631, "1000–3000 Hz_10ms_modulation_mae_db": 1.9142213533473107, "1000–3000 Hz_40ms_log_rms_mae_db": 1.510450689489989, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1350117089765282, "1000–3000 Hz_40ms_envelope_correlation": 0.9891361536698435, "1000–3000 Hz_40ms_modulation_mae_db": 2.6236945570140535, "3000–6000 Hz_10ms_log_rms_mae_db": 2.049693155210458, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20447167445081002, "3000–6000 Hz_10ms_envelope_correlation": 0.975810275082804, "3000–6000 Hz_10ms_modulation_mae_db": 0.4493702795961822, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7257116917055637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12333575802182292, "3000–6000 Hz_40ms_envelope_correlation": 0.9912887719638782, "3000–6000 Hz_40ms_modulation_mae_db": 0.771678802922239, "mr_stft_db": 5.763135687199405, "band_mae_db": 0.29179771581897435, "rms_gain_error_db": -0.1577816355265428, "output_peak": 0.5283377018708553}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.919224857345897, "spectral_convergence": 0.3187486265174082, "spectral_convergence_status": "ok", "amplitude_floor": 1.6898412267444875e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.669514806525417, "spectral_convergence": 0.3268009003740418, "spectral_convergence_status": "ok", "amplitude_floor": 1.0945626100175482e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.787350407350944, "spectral_convergence": 0.34683350322519274, "spectral_convergence_status": "ok", "amplitude_floor": 6.58943811507865e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017588563541266742, "output_power": 0.001722903595698511, "delta_db": -0.08969394682893794, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012071674288590822, "output_power": 0.0012119699453028835, "delta_db": 0.01724341207790196, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003927549262280726, "output_power": 0.0003621891128439914, "delta_db": -0.3518624941428114, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003207071656425, "output_power": 0.00028690960957271095, "delta_db": -0.4836356910998693, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.4115036625953016e-05, "output_power": 5.1089377913212366e-05, "delta_db": -0.24987342167271087, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.305762815858983e-05, "output_power": 4.597388401719993e-05, "delta_db": 0.2846109293931366, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24615331586922803, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1113771177795046, "pitch": null, "output_rms_dbfs": -24.34029991195448, "output_peak": 0.5150319456572028, "reference_rms_dbfs": -24.228922794174977, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.792030023740753}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002198976712894503, "log_rms_mae_db": 0.7148800983171492, "rms_envelope_nrmse": 0.10560839729516701, "envelope_correlation": 0.989151842871253, "rms_gain_error_db": -0.11214926226000267, "reference_mean_power": 0.0037766562988771884, "output_mean_power": 0.0036803789381346062}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.069519300574134e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006799849644412746, "output_power": 0.0006338705265959856, "reference_active": true, "delta_db": -0.3049875123704309}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022275189153607582, "output_power": 0.00019168375286708093, "reference_active": true, "delta_db": -0.6523609701411743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.238302101062001e-05, "output_power": 4.575743401141173e-05, "reference_active": true, "delta_db": 0.33269752606498215}], "active_band_count": 3, "active_band_mae_db": 0.43001533619219573}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018113949472991762, "log_rms_mae_db": 0.4259665842105792, "rms_envelope_nrmse": 0.05855515157200543, "envelope_correlation": 0.9965506800129137, "rms_gain_error_db": -0.11291662237419806, "reference_mean_power": 0.003780325824366865, "output_mean_power": 0.0036833040524250687}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9093794448028363e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006511405512347742, "output_power": 0.00060751424723611, "reference_active": true, "delta_db": -0.3011827552017353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001166725476043267, "output_power": 9.529329178491409e-05, "reference_active": true, "delta_db": -0.8790635165658224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.399065253982206e-06, "output_power": 3.4552650425800217e-06, "reference_active": true, "delta_db": 0.07121864274074678}], "active_band_count": 3, "active_band_mae_db": 0.4171549715027682}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013526658282762358, "log_rms_mae_db": 1.432443939373493, "rms_envelope_nrmse": 0.16937501857177212, "envelope_correlation": 0.9787840467259881, "rms_gain_error_db": -0.25945845433882925, "reference_mean_power": 0.001047203990993015, "output_mean_power": 0.0009864735456932376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.584657585149468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003504540347149448, "output_power": 0.00031456992972160044, "reference_active": true, "delta_db": -0.46913858985471407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.5576057457372e-05, "output_power": 8.745302309673952e-05, "reference_active": true, "delta_db": -0.3857428490107168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9623586773434205e-05, "output_power": 2.4022636447111936e-05, "reference_active": true, "delta_db": 0.8784227875407036}], "active_band_count": 3, "active_band_mae_db": 0.5777680754687115}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012031399971827572, "log_rms_mae_db": 1.1072888059479367, "rms_envelope_nrmse": 0.09001220992359553, "envelope_correlation": 0.9941751757913553, "rms_gain_error_db": -0.25978340424460444, "reference_mean_power": 0.0010510881038568295, "output_mean_power": 0.0009900583260148879}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.061243701565635e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00034327462417827815, "output_power": 0.0003123052669294273, "reference_active": true, "delta_db": -0.4106239181689466}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.625199841508274e-05, "output_power": 4.095300889095438e-05, "reference_active": true, "delta_db": -0.5284468642639866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5417688576161987e-06, "output_power": 1.7014232797531118e-06, "reference_active": true, "delta_db": 0.4279310169660167}], "active_band_count": 3, "active_band_mae_db": 0.4556672664663166}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013601220330575488, "log_rms_mae_db": 1.909323210357812, "rms_envelope_nrmse": 0.21672437432864736, "envelope_correlation": 0.9706380323994294, "rms_gain_error_db": -0.14955071692263064, "reference_mean_power": 0.00024923569921471797, "output_mean_power": 0.00024079927600002536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.70738105242738e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001701759171968848, "output_power": 0.00016731796350215878, "reference_active": true, "delta_db": -0.07355530040686323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6462817859538857e-05, "output_power": 1.8729653460204038e-05, "reference_active": true, "delta_db": -1.5010634556380857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.097841074264338e-06, "output_power": 6.553344803914109e-06, "reference_active": true, "delta_db": 3.2540388516196246}], "active_band_count": 3, "active_band_mae_db": 1.6095525358881912}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011511226838452259, "log_rms_mae_db": 1.5395541000791741, "rms_envelope_nrmse": 0.1355066889376152, "envelope_correlation": 0.9881841887843598, "rms_gain_error_db": -0.1491143798909356, "reference_mean_power": 0.00024963838926192667, "output_mean_power": 0.00024121256880593243}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6223780118670214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016948627965662844, "output_power": 0.00016722087055879214, "reference_active": true, "delta_db": -0.05844066666455466}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1767373976886565e-05, "output_power": 7.931913302551097e-06, "reference_active": true, "delta_db": -1.7130159728030303}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8952968721807594e-07, "output_power": 3.095849756613238e-07, "reference_active": true, "delta_db": 2.13102630070638}], "active_band_count": 3, "active_band_mae_db": 1.3008276467246551}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.3800826555528964e-05, "log_rms_mae_db": 2.1063566602769606, "rms_envelope_nrmse": 0.23447830316434967, "envelope_correlation": 0.9689701527881635, "rms_gain_error_db": 0.3346609367587149, "reference_mean_power": 4.104766861392576e-05, "output_mean_power": 4.433580406943199e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.890895346126465e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.549789437651056e-06, "output_power": 9.541927222051849e-06, "reference_active": true, "delta_db": -0.003576961783116218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.275656140671785e-06, "output_power": 2.4938059181162305e-06, "reference_active": true, "delta_db": -1.1843565453336184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.939546621487433e-07, "output_power": 7.419315631796138e-07, "reference_active": true, "delta_db": 0.9661055158318245}], "active_band_count": 3, "active_band_mae_db": 0.7180130076495197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.772714769720092e-05, "log_rms_mae_db": 1.7551049902594114, "rms_envelope_nrmse": 0.1483418131417778, "envelope_correlation": 0.9877909227025565, "rms_gain_error_db": 0.33537582956428275, "reference_mean_power": 4.122071638798201e-05, "output_mean_power": 4.4530043379780395e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.778938094011469e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.936665760227185e-06, "output_power": 9.735083933356018e-06, "reference_active": true, "delta_db": -0.0890098144988491}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7465237791634235e-06, "output_power": 1.3944218695088432e-06, "reference_active": true, "delta_db": -0.9778031755154577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8043700755065915e-08, "output_power": 4.744980611323335e-08, "reference_active": true, "delta_db": -0.05402011083547198}], "active_band_count": 3, "active_band_mae_db": 0.373611033616593}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8756361654492282, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "22c33f31953de997d03c822a6f306732c8320d2c216377dbff3a00637a2d375a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9678095177118828, "standard_deviation": 0.005583894075910709, "minimum": 0.9570631050007428, "maximum": 0.9734459860298275, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8488586435194936, "standard_deviation": 0.032328550294520066, "minimum": 1.8163719930290663, "maximum": 1.909323210357812, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6549039418030043, "standard_deviation": 0.7534974467629318, "minimum": 0.43914017770436714, "maximum": 2.9568726341746263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22733714308143188, "standard_deviation": 0.01857382461817028, "minimum": 0.20621286188590304, "maximum": 0.26119848008853486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9906797244231785, "standard_deviation": 0.002510935989800218, "minimum": 0.9869678307298723, "maximum": 0.9934970559328724, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4983283021385008, "standard_deviation": 0.03905786947860797, "minimum": 1.4583498326975763, "maximum": 1.568853740500104, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.762422190623941, "standard_deviation": 0.8388420460661343, "minimum": 0.4248687020359739, "maximum": 2.9415642587945356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12487102154402796, "standard_deviation": 0.014893284012882597, "minimum": 0.10319249986216836, "maximum": 0.14436253212380226, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9715259587594672, "standard_deviation": 0.004025568414450573, "minimum": 0.9653151106595184, "maximum": 0.9760474944617646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.081584167829659, "standard_deviation": 0.03218306873911806, "minimum": 2.0292037361941944, "maximum": 2.1137943410422264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6525663571766703, "standard_deviation": 0.19536848754986463, "minimum": 0.34859476214424584, "maximum": 0.877414024513468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21999480157168547, "standard_deviation": 0.014890129935173392, "minimum": 0.20116666772296465, "maximum": 0.24035171723385362, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9885857010371069, "standard_deviation": 0.0021610196394406906, "minimum": 0.98512104534418, "maximum": 0.9912887719638782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7507248630440282, "standard_deviation": 0.033912510907825745, "minimum": 1.698353868656032, "maximum": 1.7968265443833704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6393058037663252, "standard_deviation": 0.31805099093368355, "minimum": 0.3015051220431965, "maximum": 1.230946127523976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1390897914934694, "standard_deviation": 0.012310890920631158, "minimum": 0.12333575802182292, "maximum": 0.15726707934291043, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9817422839387197, "standard_deviation": 0.002662282196923547, "minimum": 0.9787840467259881, "maximum": 0.9871012679395761, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4279761032181093, "standard_deviation": 0.040927984136206574, "minimum": 1.3617312447420182, "maximum": 1.4762522829164189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.45528939311921035, "standard_deviation": 0.13064872562586408, "minimum": 0.2612283494360613, "maximum": 0.6270173538968199, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15783212444146424, "standard_deviation": 0.01152129636908131, "minimum": 0.1339297230293694, "maximum": 0.16937501857177212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9937032038887523, "standard_deviation": 0.0015883913086165969, "minimum": 0.9916935225047225, "maximum": 0.9962476054503793, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.090116763124398, "standard_deviation": 0.0527587930602338, "minimum": 1.008271900253798, "maximum": 1.172087715399659, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44353266313924855, "standard_deviation": 0.25741201560727667, "minimum": 0.15106625051071143, "maximum": 0.7641871225175155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09343019930450958, "standard_deviation": 0.011352813330226284, "minimum": 0.07545388708849496, "maximum": 0.10548695468279239, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2652703765711458, "standard_deviation": 0.0726222166298818, "minimum": 0.14408844663944037, "maximum": 0.3625541400704382, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7790822681346885, "standard_deviation": 0.011120612265076628, "minimum": 5.763135687199405, "maximum": 5.792030023740753, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.500976849580336, "standard_deviation": 0.01690495371186117, "minimum": 0.4787786141304103, "maximum": 0.5283377018708553, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11292090853158215, "standard_deviation": 0.05268088252153417, "minimum": -0.2016280871023055, "maximum": -0.05631704776669386, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899114086355797, "standard_deviation": 0.001014356253530395, "minimum": 0.9889969987347471, "maximum": 0.9922082122603548, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6883020537146347, "standard_deviation": 0.026501865732748823, "minimum": 0.6556724188079517, "maximum": 0.7324571913131317, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40716754453664583, "standard_deviation": 0.17454651863618448, "minimum": 0.18351239200588795, "maximum": 0.7625351126427796, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10201909734910744, "standard_deviation": 0.004764040052874236, "minimum": 0.09133841558118119, "maximum": 0.1061886576516549, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.996986138420362, "standard_deviation": 0.0009120949393600966, "minimum": 0.9953390060075182, "maximum": 0.9980255538513786, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42489435087309757, "standard_deviation": 0.012740454315991575, "minimum": 0.39839530742654283, "maximum": 0.43959680108615834, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4103419256576149, "standard_deviation": 0.15320172200856913, "minimum": 0.18219979808580644, "maximum": 0.6223889047319869, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05464887791215067, "standard_deviation": 0.007990002808153238, "minimum": 0.0437215470347976, "maximum": 0.06702341891028885, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0011", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.9737770557403564, "predicted_mos": 2.9737770557403564, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "591a76eca0b655c6fc87cde915f58308387f176874cef42c7ad90933ff268d0a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9730114000849426, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.688471162176001, "全帯域_10ms_rms_envelope_nrmse": 0.0909085508182643, "全帯域_10ms_envelope_correlation": 0.9919659237548188, "全帯域_10ms_modulation_mae_db": 0.2986885189894582, "全帯域_40ms_log_rms_mae_db": 0.41088314613607735, "全帯域_40ms_rms_envelope_nrmse": 0.05556919869817085, "全帯域_40ms_envelope_correlation": 0.9967964521993861, "全帯域_40ms_modulation_mae_db": 0.3672534295846858, "300–1000 Hz_10ms_log_rms_mae_db": 1.3012063485326386, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15154627886565628, "300–1000 Hz_10ms_envelope_correlation": 0.9829844323984701, "300–1000 Hz_10ms_modulation_mae_db": 0.2956015088383683, "300–1000 Hz_40ms_log_rms_mae_db": 0.999388141886959, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08369182116124767, "300–1000 Hz_40ms_envelope_correlation": 0.9946638496391412, "300–1000 Hz_40ms_modulation_mae_db": 0.09362024685606578, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4403879083851854, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16585744274109418, "1000–3000 Hz_10ms_envelope_correlation": 0.983184890006158, "1000–3000 Hz_10ms_modulation_mae_db": 0.43374574712515396, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2329533087336702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10197358884401926, "1000–3000 Hz_40ms_envelope_correlation": 0.9934553479702499, "1000–3000 Hz_40ms_modulation_mae_db": 0.5302439294517517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.459818036261604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15308603718930622, "3000–6000 Hz_10ms_envelope_correlation": 0.9857890021891027, "3000–6000 Hz_10ms_modulation_mae_db": 0.6884647312762988, "3000–6000 Hz_40ms_log_rms_mae_db": 1.344091765127022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08629490989002392, "3000–6000 Hz_40ms_envelope_correlation": 0.9954008354257085, "3000–6000 Hz_40ms_modulation_mae_db": 0.7542495516361445, "mr_stft_db": 5.295822989952403, "band_mae_db": 0.23283782475415957, "rms_gain_error_db": -0.039555018160055556, "output_peak": 0.521746848498798}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7306534895039453, "全帯域_10ms_rms_envelope_nrmse": 0.0968150620995078, "全帯域_10ms_envelope_correlation": 0.9910995816390772, "全帯域_10ms_modulation_mae_db": 0.45083287152628876, "全帯域_40ms_log_rms_mae_db": 0.43398135201511956, "全帯域_40ms_rms_envelope_nrmse": 0.0560560375783952, "全帯域_40ms_envelope_correlation": 0.9970953184764322, "全帯域_40ms_modulation_mae_db": 0.32511986725637926, "300–1000 Hz_10ms_log_rms_mae_db": 1.2373956772834263, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15062323370064565, "300–1000 Hz_10ms_envelope_correlation": 0.9835107881663213, "300–1000 Hz_10ms_modulation_mae_db": 0.16748949403304225, "300–1000 Hz_40ms_log_rms_mae_db": 0.9949097668110409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07889333407741018, "300–1000 Hz_40ms_envelope_correlation": 0.9958492597458711, "300–1000 Hz_40ms_modulation_mae_db": 0.24209598056517914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4657043963663552, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2190032462040614, "1000–3000 Hz_10ms_envelope_correlation": 0.9705499608453173, "1000–3000 Hz_10ms_modulation_mae_db": 2.4635034061145253, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2378185230273127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12828320457727824, "1000–3000 Hz_40ms_envelope_correlation": 0.9909107042114965, "1000–3000 Hz_40ms_modulation_mae_db": 2.373848282903736, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4557588408552542, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1534861864791469, "3000–6000 Hz_10ms_envelope_correlation": 0.9860632422054374, "3000–6000 Hz_10ms_modulation_mae_db": 0.230170236936562, "3000–6000 Hz_40ms_log_rms_mae_db": 1.339462124580525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09745206436877528, "3000–6000 Hz_40ms_envelope_correlation": 0.9942577661802428, "3000–6000 Hz_40ms_modulation_mae_db": 0.2217958741590099, "mr_stft_db": 5.302527276786193, "band_mae_db": 0.26800353290040496, "rms_gain_error_db": -0.18054898309018078, "output_peak": 0.5202128132271824}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.693316939705966, "全帯域_10ms_rms_envelope_nrmse": 0.09963843403735158, "全帯域_10ms_envelope_correlation": 0.9904134077181626, "全帯域_10ms_modulation_mae_db": 0.2520137249095265, "全帯域_40ms_log_rms_mae_db": 0.41761968923256293, "全帯域_40ms_rms_envelope_nrmse": 0.04930609368738799, "全帯域_40ms_envelope_correlation": 0.9976661026785493, "全帯域_40ms_modulation_mae_db": 0.6077789979130754, "300–1000 Hz_10ms_log_rms_mae_db": 1.2907004820750458, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15104320109190222, "300–1000 Hz_10ms_envelope_correlation": 0.9831457824832734, "300–1000 Hz_10ms_modulation_mae_db": 0.3063192647868655, "300–1000 Hz_40ms_log_rms_mae_db": 0.9864007677920997, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09143207753270252, "300–1000 Hz_40ms_envelope_correlation": 0.9936712988443794, "300–1000 Hz_40ms_modulation_mae_db": 0.33296701737372375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4996309825866647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20887265725172954, "1000–3000 Hz_10ms_envelope_correlation": 0.9738438102474967, "1000–3000 Hz_10ms_modulation_mae_db": 1.483698947000083, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3103426342234439, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11187605462765576, "1000–3000 Hz_40ms_envelope_correlation": 0.994196188108724, "1000–3000 Hz_40ms_modulation_mae_db": 1.4284625780392057, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4632546025407893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16248544700087283, "3000–6000 Hz_10ms_envelope_correlation": 0.9840737708651257, "3000–6000 Hz_10ms_modulation_mae_db": 0.3095768806913446, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3453207677483094, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10984384019521397, "3000–6000 Hz_40ms_envelope_correlation": 0.9925053912732023, "3000–6000 Hz_40ms_modulation_mae_db": 0.20681635812340185, "mr_stft_db": 5.293680774401351, "band_mae_db": 0.3159601175369812, "rms_gain_error_db": -0.1331562988488883, "output_peak": 0.4720228727298581}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7493156886113063, "全帯域_10ms_rms_envelope_nrmse": 0.10762932336137447, "全帯域_10ms_envelope_correlation": 0.988735669797337, "全帯域_10ms_modulation_mae_db": 0.47310883612201177, "全帯域_40ms_log_rms_mae_db": 0.4428235727870862, "全帯域_40ms_rms_envelope_nrmse": 0.06401028921876015, "全帯域_40ms_envelope_correlation": 0.9958524521915354, "全帯域_40ms_modulation_mae_db": 0.4790887314288081, "300–1000 Hz_10ms_log_rms_mae_db": 1.288226546606556, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1614352273561611, "300–1000 Hz_10ms_envelope_correlation": 0.9812221130226221, "300–1000 Hz_10ms_modulation_mae_db": 0.7272254682814766, "300–1000 Hz_40ms_log_rms_mae_db": 1.025361578289129, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10729922736277353, "300–1000 Hz_40ms_envelope_correlation": 0.9919848201686644, "300–1000 Hz_40ms_modulation_mae_db": 0.5635635103598134, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5176774283627128, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2678119180254031, "1000–3000 Hz_10ms_envelope_correlation": 0.9547620567843597, "1000–3000 Hz_10ms_modulation_mae_db": 2.5534669729263864, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3230862196850048, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1669302528641217, "1000–3000 Hz_40ms_envelope_correlation": 0.9823835110540377, "1000–3000 Hz_40ms_modulation_mae_db": 2.379334160884403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.47891606136372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15668745496366324, "3000–6000 Hz_10ms_envelope_correlation": 0.9854292549164989, "3000–6000 Hz_10ms_modulation_mae_db": 0.4789757068969611, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3489390341722987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09769558702530698, "3000–6000 Hz_40ms_envelope_correlation": 0.9941982659537021, "3000–6000 Hz_40ms_modulation_mae_db": 0.4331998394830026, "mr_stft_db": 5.308715483632953, "band_mae_db": 0.29984377060931405, "rms_gain_error_db": -0.11976281733003045, "output_peak": 0.5548178442103697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6951600050552099, "全帯域_10ms_rms_envelope_nrmse": 0.09729001416365499, "全帯域_10ms_envelope_correlation": 0.9911589238877005, "全帯域_10ms_modulation_mae_db": 0.5030825781260369, "全帯域_40ms_log_rms_mae_db": 0.4372564912114893, "全帯域_40ms_rms_envelope_nrmse": 0.05579172511479029, "全帯域_40ms_envelope_correlation": 0.9973189453613093, "全帯域_40ms_modulation_mae_db": 0.42274173241518254, "300–1000 Hz_10ms_log_rms_mae_db": 1.2314155446514707, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13339134185553894, "300–1000 Hz_10ms_envelope_correlation": 0.9873718060275609, "300–1000 Hz_10ms_modulation_mae_db": 0.32777802077038903, "300–1000 Hz_40ms_log_rms_mae_db": 0.9507087068961165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07908161888631811, "300–1000 Hz_40ms_envelope_correlation": 0.9960005503606226, "300–1000 Hz_40ms_modulation_mae_db": 0.5305100694599008, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5589979904431703, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2477812047005668, "1000–3000 Hz_10ms_envelope_correlation": 0.9637635073814993, "1000–3000 Hz_10ms_modulation_mae_db": 1.9899645278967786, "1000–3000 Hz_40ms_log_rms_mae_db": 1.312125907915318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16400086329320765, "1000–3000 Hz_40ms_envelope_correlation": 0.986954319210025, "1000–3000 Hz_40ms_modulation_mae_db": 1.8220903680248373, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4576238608095373, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14144365178397975, "3000–6000 Hz_10ms_envelope_correlation": 0.9880143968427657, "3000–6000 Hz_10ms_modulation_mae_db": 0.14755183905987448, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3463761704894701, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08796695192207893, "3000–6000 Hz_40ms_envelope_correlation": 0.9952231479410808, "3000–6000 Hz_40ms_modulation_mae_db": 0.2910676423228364, "mr_stft_db": 5.305056378301124, "band_mae_db": 0.32098769543822786, "rms_gain_error_db": -0.21866836293832845, "output_peak": 0.5514459469299766}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7064290756194754, "全帯域_10ms_rms_envelope_nrmse": 0.1165398287161194, "全帯域_10ms_envelope_correlation": 0.9870994906098247, "全帯域_10ms_modulation_mae_db": 0.4726669165683854, "全帯域_40ms_log_rms_mae_db": 0.43238859597104357, "全帯域_40ms_rms_envelope_nrmse": 0.0683500999103281, "全帯域_40ms_envelope_correlation": 0.9957586029966596, "全帯域_40ms_modulation_mae_db": 0.5514703569139624, "300–1000 Hz_10ms_log_rms_mae_db": 1.281215020271058, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1479310185994457, "300–1000 Hz_10ms_envelope_correlation": 0.9849949810620154, "300–1000 Hz_10ms_modulation_mae_db": 0.8085618585428974, "300–1000 Hz_40ms_log_rms_mae_db": 0.982969906035587, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09181083579099272, "300–1000 Hz_40ms_envelope_correlation": 0.9952057285737572, "300–1000 Hz_40ms_modulation_mae_db": 0.9378394365466773, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4824924740644019, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.30765941336531505, "1000–3000 Hz_10ms_envelope_correlation": 0.9485219161901329, "1000–3000 Hz_10ms_modulation_mae_db": 2.889727639626899, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2462221861333247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.22483499634171053, "1000–3000 Hz_40ms_envelope_correlation": 0.980904387843856, "1000–3000 Hz_40ms_modulation_mae_db": 2.7094609735738415, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4446180893840286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14753761537115412, "3000–6000 Hz_10ms_envelope_correlation": 0.9870416857704672, "3000–6000 Hz_10ms_modulation_mae_db": 0.3227971331518159, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3415119065654164, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0917836169023917, "3000–6000 Hz_40ms_envelope_correlation": 0.9948473467773801, "3000–6000 Hz_40ms_modulation_mae_db": 0.27176834813491274, "mr_stft_db": 5.2873886680551765, "band_mae_db": 0.46011335395476766, "rms_gain_error_db": -0.23508889465019844, "output_peak": 0.6253396292687907}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7269270048005803, "全帯域_10ms_rms_envelope_nrmse": 0.12587197776196665, "全帯域_10ms_envelope_correlation": 0.9845743454641624, "全帯域_10ms_modulation_mae_db": 0.6069763484933209, "全帯域_40ms_log_rms_mae_db": 0.44429372049810595, "全帯域_40ms_rms_envelope_nrmse": 0.05662029217543388, "全帯域_40ms_envelope_correlation": 0.9968855900329501, "全帯域_40ms_modulation_mae_db": 0.40874080501501986, "300–1000 Hz_10ms_log_rms_mae_db": 1.2774989859654056, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1859382815003285, "300–1000 Hz_10ms_envelope_correlation": 0.9743726824118437, "300–1000 Hz_10ms_modulation_mae_db": 0.7170542855127895, "300–1000 Hz_40ms_log_rms_mae_db": 0.9901625373520714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08433064756963132, "300–1000 Hz_40ms_envelope_correlation": 0.9949754571227218, "300–1000 Hz_40ms_modulation_mae_db": 0.6102066448364815, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5181175292164393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2448321419461175, "1000–3000 Hz_10ms_envelope_correlation": 0.9626911222374782, "1000–3000 Hz_10ms_modulation_mae_db": 1.7090813841101686, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3132948057957985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14376369742645256, "1000–3000 Hz_40ms_envelope_correlation": 0.9875810847643656, "1000–3000 Hz_40ms_modulation_mae_db": 1.77410923451391, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4726625181574706, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14525587294475692, "3000–6000 Hz_10ms_envelope_correlation": 0.9872495186314351, "3000–6000 Hz_10ms_modulation_mae_db": 0.31329215923155745, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3438224468126938, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08722320604917701, "3000–6000 Hz_40ms_envelope_correlation": 0.9954012299875198, "3000–6000 Hz_40ms_modulation_mae_db": 0.47587667844395165, "mr_stft_db": 5.296716283891063, "band_mae_db": 0.3085712314856139, "rms_gain_error_db": -0.1480885432393887, "output_peak": 0.5410389742383751}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6700164048107781, "全帯域_10ms_rms_envelope_nrmse": 0.09563330693344434, "全帯域_10ms_envelope_correlation": 0.991176252016952, "全帯域_10ms_modulation_mae_db": 0.521976015885196, "全帯域_40ms_log_rms_mae_db": 0.39549367372339, "全帯域_40ms_rms_envelope_nrmse": 0.04999695275646068, "全帯域_40ms_envelope_correlation": 0.9975918107601921, "全帯域_40ms_modulation_mae_db": 0.4144862090169134, "300–1000 Hz_10ms_log_rms_mae_db": 1.3258669328396995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15165645651410106, "300–1000 Hz_10ms_envelope_correlation": 0.9829303123044175, "300–1000 Hz_10ms_modulation_mae_db": 0.4835278460132218, "300–1000 Hz_40ms_log_rms_mae_db": 1.0191185789404105, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07961227136925039, "300–1000 Hz_40ms_envelope_correlation": 0.9952037516120306, "300–1000 Hz_40ms_modulation_mae_db": 0.4282291894268897, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4610990701065458, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21712485269798779, "1000–3000 Hz_10ms_envelope_correlation": 0.9708263530345018, "1000–3000 Hz_10ms_modulation_mae_db": 1.8954894788216758, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2352094994423226, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13284244366043593, "1000–3000 Hz_40ms_envelope_correlation": 0.9895638050882686, "1000–3000 Hz_40ms_modulation_mae_db": 2.185389574543481, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4622656983189273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14521058821055183, "3000–6000 Hz_10ms_envelope_correlation": 0.9876159399775588, "3000–6000 Hz_10ms_modulation_mae_db": 0.11578065482350915, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3121704616652108, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09332817885265283, "3000–6000 Hz_40ms_envelope_correlation": 0.9947890236939332, "3000–6000 Hz_40ms_modulation_mae_db": 0.13903562406693268, "mr_stft_db": 5.297464772555769, "band_mae_db": 0.30584002014688544, "rms_gain_error_db": -0.13029819928001027, "output_peak": 0.4223042714266265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.31739450261553, "spectral_convergence": 0.3155247966184585, "spectral_convergence_status": "ok", "amplitude_floor": 1.6898412267444875e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.080040160333091, "spectral_convergence": 0.34194937346457627, "spectral_convergence_status": "ok", "amplitude_floor": 1.0945626100175482e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.49003430690859, "spectral_convergence": 0.3745822267914529, "spectral_convergence_status": "ok", "amplitude_floor": 6.58943811507865e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017588563541266742, "output_power": 0.0016918842202167685, "delta_db": -0.16859732172723063, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012071674288590822, "output_power": 0.00122210359735025, "delta_db": 0.05340513452182462, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003927549262280726, "output_power": 0.0004355692930649627, "delta_db": 0.44935614105377303, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003207071656425, "output_power": 0.00029739607402929556, "delta_db": -0.327734325534862, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.4115036625953016e-05, "output_power": 5.0995323151787524e-05, "delta_db": -0.2578760851502814, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.305762815858983e-05, "output_power": 4.446884985420029e-05, "delta_db": 0.14005794053698586, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23283782475415957, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.039555018160055556, "pitch": null, "output_rms_dbfs": -24.268477812335032, "output_peak": 0.521746848498798, "reference_rms_dbfs": -24.228922794174977, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.295822989952403}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002198976712894503, "log_rms_mae_db": 0.688471162176001, "rms_envelope_nrmse": 0.0909085508182643, "envelope_correlation": 0.9919659237548188, "rms_gain_error_db": -0.03790150471149828, "reference_mean_power": 0.0037766562988771884, "output_mean_power": 0.0037438402789581725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.069519300574134e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006799849644412746, "output_power": 0.0007426108975898927, "reference_active": true, "delta_db": 0.38262008212320087}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022275189153607582, "output_power": 0.00020430652188497623, "reference_active": true, "delta_db": -0.3753917018292825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.238302101062001e-05, "output_power": 4.105693566719188e-05, "reference_active": true, "delta_db": -0.13805377301589125}], "active_band_count": 3, "active_band_mae_db": 0.2986885189894582}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018113949472991762, "log_rms_mae_db": 0.41088314613607735, "rms_envelope_nrmse": 0.05556919869817085, "envelope_correlation": 0.9967964521993861, "rms_gain_error_db": -0.035906544080107794, "reference_mean_power": 0.003780325824366865, "output_mean_power": 0.003749199743686076}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9093794448028363e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006511405512347742, "output_power": 0.0007091590652244836, "reference_active": true, "delta_db": 0.3706891603034831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001166725476043267, "output_power": 0.00010654326880970563, "reference_active": true, "delta_db": -0.39442663949680046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.399065253982206e-06, "output_power": 3.1455388481260733e-06, "reference_active": true, "delta_db": -0.33664448895377386}], "active_band_count": 3, "active_band_mae_db": 0.3672534295846858}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013526658282762358, "log_rms_mae_db": 1.3012063485326386, "rms_envelope_nrmse": 0.15154627886565628, "envelope_correlation": 0.9829844323984701, "rms_gain_error_db": -0.16316595395985925, "reference_mean_power": 0.001047203990993015, "output_mean_power": 0.0010085900822098925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.584657585149468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003504540347149448, "output_power": 0.00034852087831034384, "reference_active": true, "delta_db": -0.024022644709134486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.5576057457372e-05, "output_power": 9.041592658530751e-05, "reference_active": true, "delta_db": -0.24104174447942434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9623586773434205e-05, "output_power": 2.2643961291219355e-05, "reference_active": true, "delta_db": 0.6217401373265459}], "active_band_count": 3, "active_band_mae_db": 0.2956015088383683}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012031399971827572, "log_rms_mae_db": 0.999388141886959, "rms_envelope_nrmse": 0.08369182116124767, "envelope_correlation": 0.9946638496391412, "rms_gain_error_db": -0.16232341641766318, "reference_mean_power": 0.0010510881038568295, "output_mean_power": 0.0010125273875764924}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.061243701565635e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00034327462417827815, "output_power": 0.00034441034762754526, "reference_active": true, "delta_db": 0.014344907789033074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.625199841508274e-05, "output_power": 4.437040050985785e-05, "reference_active": true, "delta_db": -0.18037153191281396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5417688576161987e-06, "output_power": 1.572655865795049e-06, "reference_active": true, "delta_db": 0.08614430086635025}], "active_band_count": 3, "active_band_mae_db": 0.09362024685606578}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013601220330575488, "log_rms_mae_db": 1.4403879083851854, "rms_envelope_nrmse": 0.16585744274109418, "envelope_correlation": 0.983184890006158, "rms_gain_error_db": 0.025793702504137317, "reference_mean_power": 0.00024923569921471797, "output_mean_power": 0.00025072036926299443}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.70738105242738e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001701759171968848, "output_power": 0.00018457482208524315, "reference_active": true, "delta_db": 0.35274358471732964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6462817859538857e-05, "output_power": 2.8696625762156208e-05, "reference_active": true, "delta_db": 0.35194746440156655}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.097841074264338e-06, "output_power": 3.5539699632855943e-06, "reference_active": true, "delta_db": 0.5965461922565656}], "active_band_count": 3, "active_band_mae_db": 0.43374574712515396}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011511226838452259, "log_rms_mae_db": 1.2329533087336702, "rms_envelope_nrmse": 0.10197358884401926, "envelope_correlation": 0.9934553479702499, "rms_gain_error_db": 0.027544878671775724, "reference_mean_power": 0.00024963838926192667, "output_mean_power": 0.0002512267381233278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6223780118670214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016948627965662844, "output_power": 0.00018406371575725443, "reference_active": true, "delta_db": 0.35833638306060034}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1767373976886565e-05, "output_power": 1.3016159616409605e-05, "reference_active": true, "delta_db": 0.43803309778252036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8952968721807594e-07, "output_power": 2.2756919677810965e-07, "reference_active": true, "delta_db": 0.7943623075121342}], "active_band_count": 3, "active_band_mae_db": 0.5302439294517517}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.3800826555528964e-05, "log_rms_mae_db": 1.459818036261604, "rms_envelope_nrmse": 0.15308603718930622, "envelope_correlation": 0.9857890021891027, "rms_gain_error_db": -0.10388146876572771, "reference_mean_power": 4.104766861392576e-05, "output_mean_power": 4.007747470100012e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.890895346126465e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.549789437651056e-06, "output_power": 9.441296455189098e-06, "reference_active": true, "delta_db": -0.04962161350049053}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.275656140671785e-06, "output_power": 2.934780874485783e-06, "reference_active": true, "delta_db": -0.4772262559070758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.939546621487433e-07, "output_power": 4.1677212389453484e-07, "reference_active": true, "delta_db": -1.53854632442133}], "active_band_count": 3, "active_band_mae_db": 0.6884647312762988}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.772714769720092e-05, "log_rms_mae_db": 1.344091765127022, "rms_envelope_nrmse": 0.08629490989002392, "envelope_correlation": 0.9954008354257085, "rms_gain_error_db": -0.10435573910994443, "reference_mean_power": 4.122071638798201e-05, "output_mean_power": 4.024203749197766e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.778938094011469e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.936665760227185e-06, "output_power": 9.679349288988563e-06, "reference_active": true, "delta_db": -0.1139451960449323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7465237791634235e-06, "output_power": 1.54960424942847e-06, "reference_active": true, "delta_db": -0.5195370432943065}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8043700755065915e-08, "output_power": 3.3014886737806757e-08, "reference_active": true, "delta_db": -1.629266415569195}], "active_band_count": 3, "active_band_mae_db": 0.7542495516361445}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8910232834433728, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "31ee2c65d3db07033af1041180778c613037143939fc6b896a6381b0f75b2bc1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.966017952090868, "standard_deviation": 0.01101015379075237, "minimum": 0.9485219161901329, "maximum": 0.983184890006158, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4930134724414343, "standard_deviation": 0.03832330241557361, "minimum": 1.4403879083851854, "maximum": 1.5589979904431703, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.927334762952709, "standard_deviation": 0.764003593830452, "minimum": 0.43374574712515396, "maximum": 2.889727639626899, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23486785961653442, "standard_deviation": 0.04251831726572095, "minimum": 0.16585744274109418, "maximum": 0.30765941336531505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9882436685313779, "standard_deviation": 0.004807105712682826, "minimum": 0.980904387843856, "maximum": 0.994196188108724, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2763816356195243, "standard_deviation": 0.04132237918909637, "minimum": 1.2329533087336702, "maximum": 1.3230862196850048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.9003673877418956, "standard_deviation": 0.6877981700336185, "minimum": 0.5302439294517517, "maximum": 2.7094609735738415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14681313770436022, "standard_deviation": 0.03885237513271844, "minimum": 0.10197358884401926, "maximum": 0.22483499634171053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9864096014247989, "standard_deviation": 0.0013124624822647144, "minimum": 0.9840737708651257, "maximum": 0.9880143968427657, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4618647134614164, "standard_deviation": 0.010463098920717475, "minimum": 1.4446180893840286, "maximum": 1.47891606136372, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.32582616775849044, "standard_deviation": 0.18533473027303438, "minimum": 0.11578065482350915, "maximum": 0.6884647312762988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15064910674292897, "standard_deviation": 0.007006471351068784, "minimum": 0.14144365178397975, "maximum": 0.16248544700087283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9945778759040962, "standard_deviation": 0.0009584596838047419, "minimum": 0.9925053912732023, "maximum": 0.9954012299875198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3402118346451184, "standard_deviation": 0.011692033166637738, "minimum": 1.3121704616652108, "maximum": 1.3489390341722987, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3492262395462741, "standard_deviation": 0.19900362417864279, "minimum": 0.13903562406693268, "maximum": 0.7542495516361445, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09394854440070258, "standard_deviation": 0.007779413485276624, "minimum": 0.08629490989002392, "maximum": 0.10984384019521397, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9825666122345655, "standard_deviation": 0.003769592331468242, "minimum": 0.9743726824118437, "maximum": 0.9873718060275609, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2791906922781626, "standard_deviation": 0.03142730900965857, "minimum": 1.2314155446514707, "maximum": 1.3258669328396995, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4791947183473813, "standard_deviation": 0.24209722599239086, "minimum": 0.16748949403304225, "maximum": 0.8085618585428974, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15419562993547242, "standard_deviation": 0.014967764133597967, "minimum": 0.13339134185553894, "maximum": 0.1859382815003285, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946943395083986, "standard_deviation": 0.0013108945152466938, "minimum": 0.9919848201686644, "maximum": 0.9960005503606226, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9936274980004267, "standard_deviation": 0.023026428197343, "minimum": 0.9507087068961165, "maximum": 1.025361578289129, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4673790119280914, "standard_deviation": 0.2577421423461685, "minimum": 0.09362024685606578, "maximum": 0.9378394365466773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0870189792187908, "standard_deviation": 0.009679949954812073, "minimum": 0.07889333407741018, "maximum": 0.10729922736277353, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3140196933532943, "standard_deviation": 0.0659016673918807, "minimum": 0.23283782475415957, "maximum": 0.46011335395476766, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.2984215784470035, "standard_deviation": 0.006778309058498091, "minimum": 5.2873886680551765, "maximum": 5.308715483632953, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5261161500662471, "standard_deviation": 0.060107409032179336, "minimum": 0.4223042714266265, "maximum": 0.6253396292687907, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.15064588969213513, "standard_deviation": 0.06172898503919031, "minimum": -0.23508889465019844, "maximum": -0.039555018160055556, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9895279493610043, "standard_deviation": 0.002553375177988263, "minimum": 0.9845743454641624, "maximum": 0.9919659237548188, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7075362212854077, "standard_deviation": 0.02614173466954062, "minimum": 0.6700164048107781, "maximum": 0.7493156886113063, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4474182263275281, "standard_deviation": 0.11694246876474776, "minimum": 0.2520137249095265, "maximum": 0.6069763484933209, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10379081223646043, "standard_deviation": 0.011985173710523763, "minimum": 0.0909085508182643, "maximum": 0.12587197776196665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968706593371267, "standard_deviation": 0.0007253931242957913, "minimum": 0.9957586029966596, "maximum": 0.9976661026785493, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42684253019685936, "standard_deviation": 0.017207238962491805, "minimum": 0.39549367372339, "maximum": 0.44429372049810595, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44708501619300334, "standard_deviation": 0.09420132612667059, "minimum": 0.32511986725637926, "maximum": 0.6077789979130754, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.056962586142465896, "standard_deviation": 0.00644469829973985, "minimum": 0.04930609368738799, "maximum": 0.0683500999103281, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0011", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.048887252807617, "predicted_mos": 3.048887252807617, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "585c84b25d7af6c2d2cdb2be53d7a0aa52805cbc362b540c6e41bb87c895828c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9301080997101963, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6981364087499667, "全帯域_10ms_rms_envelope_nrmse": 0.12164750217850853, "全帯域_10ms_envelope_correlation": 0.9855781465128293, "全帯域_10ms_modulation_mae_db": 0.35621613062903107, "全帯域_40ms_log_rms_mae_db": 0.4186976800011373, "全帯域_40ms_rms_envelope_nrmse": 0.07090713454593864, "全帯域_40ms_envelope_correlation": 0.9949148960512011, "全帯域_40ms_modulation_mae_db": 0.6016240537143481, "300–1000 Hz_10ms_log_rms_mae_db": 1.2758890164132046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16403391398352735, "300–1000 Hz_10ms_envelope_correlation": 0.9799772627679145, "300–1000 Hz_10ms_modulation_mae_db": 0.5178995123476906, "300–1000 Hz_40ms_log_rms_mae_db": 0.9515052831305203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07662690486139737, "300–1000 Hz_40ms_envelope_correlation": 0.9955520460993749, "300–1000 Hz_40ms_modulation_mae_db": 0.19634081806606632, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5890235890458662, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2632282740762697, "1000–3000 Hz_10ms_envelope_correlation": 0.958644181385783, "1000–3000 Hz_10ms_modulation_mae_db": 2.5191200624257757, "1000–3000 Hz_40ms_log_rms_mae_db": 1.305442386013484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1839656530951573, "1000–3000 Hz_40ms_envelope_correlation": 0.9824648586184773, "1000–3000 Hz_40ms_modulation_mae_db": 3.281919570917051, "3000–6000 Hz_10ms_log_rms_mae_db": 1.611724893209612, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16066270948051867, "3000–6000 Hz_10ms_envelope_correlation": 0.9843996607216396, "3000–6000 Hz_10ms_modulation_mae_db": 0.23700912149289433, "3000–6000 Hz_40ms_log_rms_mae_db": 1.396331206566409, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08974464275148433, "3000–6000 Hz_40ms_envelope_correlation": 0.995184382753728, "3000–6000 Hz_40ms_modulation_mae_db": 0.5168550695416538, "mr_stft_db": 5.4381091526948095, "band_mae_db": 0.3847744556931874, "rms_gain_error_db": -0.13592992441920956, "output_peak": 0.5247168935688251}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6697586881078104, "全帯域_10ms_rms_envelope_nrmse": 0.09256342834019889, "全帯域_10ms_envelope_correlation": 0.9918124430961651, "全帯域_10ms_modulation_mae_db": 0.4111661044925862, "全帯域_40ms_log_rms_mae_db": 0.39370389510936804, "全帯域_40ms_rms_envelope_nrmse": 0.045592805299383296, "全帯域_40ms_envelope_correlation": 0.9981312685499291, "全帯域_40ms_modulation_mae_db": 0.4314626768457474, "300–1000 Hz_10ms_log_rms_mae_db": 1.2636967739539435, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1382188397312259, "300–1000 Hz_10ms_envelope_correlation": 0.9864716176687441, "300–1000 Hz_10ms_modulation_mae_db": 0.5180531549804817, "300–1000 Hz_40ms_log_rms_mae_db": 0.9732159408330058, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0750239790715603, "300–1000 Hz_40ms_envelope_correlation": 0.9966611941386279, "300–1000 Hz_40ms_modulation_mae_db": 0.6794349716159692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6604176016664898, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2529448529264144, "1000–3000 Hz_10ms_envelope_correlation": 0.9599728656533827, "1000–3000 Hz_10ms_modulation_mae_db": 1.9109507042619167, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3465532430507452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11995718686394197, "1000–3000 Hz_40ms_envelope_correlation": 0.9919210952959876, "1000–3000 Hz_40ms_modulation_mae_db": 2.0143014361953506, "3000–6000 Hz_10ms_log_rms_mae_db": 1.648806546051499, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1784786342843435, "3000–6000 Hz_10ms_envelope_correlation": 0.9811210555909959, "3000–6000 Hz_10ms_modulation_mae_db": 0.6898351214867374, "3000–6000 Hz_40ms_log_rms_mae_db": 1.415961838299672, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11348565530880951, "3000–6000 Hz_40ms_envelope_correlation": 0.99219448870046, "3000–6000 Hz_40ms_modulation_mae_db": 0.9054901661705399, "mr_stft_db": 5.43389391711671, "band_mae_db": 0.3201015339176193, "rms_gain_error_db": -0.15377358296581667, "output_peak": 0.4650567447956773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7263256470654721, "全帯域_10ms_rms_envelope_nrmse": 0.10394647890466037, "全帯域_10ms_envelope_correlation": 0.9895136198467892, "全帯域_10ms_modulation_mae_db": 0.506419833523489, "全帯域_40ms_log_rms_mae_db": 0.44027424208613397, "全帯域_40ms_rms_envelope_nrmse": 0.05221156743920957, "全帯域_40ms_envelope_correlation": 0.9973398649862665, "全帯域_40ms_modulation_mae_db": 0.531798559925298, "300–1000 Hz_10ms_log_rms_mae_db": 1.369476104241568, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19030386328523918, "300–1000 Hz_10ms_envelope_correlation": 0.9733324957361811, "300–1000 Hz_10ms_modulation_mae_db": 0.6558300845311368, "300–1000 Hz_40ms_log_rms_mae_db": 1.0158470248345441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11655228846954317, "300–1000 Hz_40ms_envelope_correlation": 0.9902751252959304, "300–1000 Hz_40ms_modulation_mae_db": 0.49133574107115896, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6157209050915127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1928887331804061, "1000–3000 Hz_10ms_envelope_correlation": 0.9772384851822479, "1000–3000 Hz_10ms_modulation_mae_db": 1.592916293160868, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3520807699513302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09040225664695409, "1000–3000 Hz_40ms_envelope_correlation": 0.9948585868177047, "1000–3000 Hz_40ms_modulation_mae_db": 2.044608539526086, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6379302229148682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16400518322939403, "3000–6000 Hz_10ms_envelope_correlation": 0.9838335605978317, "3000–6000 Hz_10ms_modulation_mae_db": 0.6848150340453746, "3000–6000 Hz_40ms_log_rms_mae_db": 1.424715436789939, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09621787732697204, "3000–6000 Hz_40ms_envelope_correlation": 0.9945838561284279, "3000–6000 Hz_40ms_modulation_mae_db": 0.7866513885553682, "mr_stft_db": 5.443947055980311, "band_mae_db": 0.22679700740659847, "rms_gain_error_db": -0.11331433880138277, "output_peak": 0.5661453330938029}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6605275335666498, "全帯域_10ms_rms_envelope_nrmse": 0.0931244765571047, "全帯域_10ms_envelope_correlation": 0.9916901344073773, "全帯域_10ms_modulation_mae_db": 0.40990362581322987, "全帯域_40ms_log_rms_mae_db": 0.4217000715731202, "全帯域_40ms_rms_envelope_nrmse": 0.05362711432592965, "全帯域_40ms_envelope_correlation": 0.9972428824840275, "全帯域_40ms_modulation_mae_db": 0.6246669318817535, "300–1000 Hz_10ms_log_rms_mae_db": 1.3052641259085964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14359561161578277, "300–1000 Hz_10ms_envelope_correlation": 0.9853089022029601, "300–1000 Hz_10ms_modulation_mae_db": 0.6093120422631834, "300–1000 Hz_40ms_log_rms_mae_db": 1.0045444200339406, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0772441347408324, "300–1000 Hz_40ms_envelope_correlation": 0.9963840356014001, "300–1000 Hz_40ms_modulation_mae_db": 0.868045156571382, "1000–3000 Hz_10ms_log_rms_mae_db": 1.601604182682888, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21848184344179825, "1000–3000 Hz_10ms_envelope_correlation": 0.9703350301694458, "1000–3000 Hz_10ms_modulation_mae_db": 1.9436012646534835, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3157959707363682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12722035896329584, "1000–3000 Hz_40ms_envelope_correlation": 0.9903289677375391, "1000–3000 Hz_40ms_modulation_mae_db": 1.5003549132166896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6412729334209792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16769981970133388, "3000–6000 Hz_10ms_envelope_correlation": 0.9832083626064453, "3000–6000 Hz_10ms_modulation_mae_db": 0.6483183559609954, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4240064303903202, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10680884309074365, "3000–6000 Hz_40ms_envelope_correlation": 0.9934704130453768, "3000–6000 Hz_40ms_modulation_mae_db": 0.8799485240869407, "mr_stft_db": 5.435625878036023, "band_mae_db": 0.28679292846887033, "rms_gain_error_db": -0.14783547714032144, "output_peak": 0.5899805487916077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.745806688873048, "全帯域_10ms_rms_envelope_nrmse": 0.10996412615065193, "全帯域_10ms_envelope_correlation": 0.9882644362040035, "全帯域_10ms_modulation_mae_db": 0.34341027348611836, "全帯域_40ms_log_rms_mae_db": 0.467370922372422, "全帯域_40ms_rms_envelope_nrmse": 0.05525675796106858, "全帯域_40ms_envelope_correlation": 0.9969667612632344, "全帯域_40ms_modulation_mae_db": 0.07503910043145676, "300–1000 Hz_10ms_log_rms_mae_db": 1.3125281070976813, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16411562223560933, "300–1000 Hz_10ms_envelope_correlation": 0.9802381804640017, "300–1000 Hz_10ms_modulation_mae_db": 0.38810507158324525, "300–1000 Hz_40ms_log_rms_mae_db": 1.0051376526156806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09935396776471758, "300–1000 Hz_40ms_envelope_correlation": 0.9928335446948728, "300–1000 Hz_40ms_modulation_mae_db": 0.39121194980512386, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5880400885639305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2685666800502507, "1000–3000 Hz_10ms_envelope_correlation": 0.9545786574200059, "1000–3000 Hz_10ms_modulation_mae_db": 1.673926106012915, "1000–3000 Hz_40ms_log_rms_mae_db": 1.30327706271828, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15488481129314172, "1000–3000 Hz_40ms_envelope_correlation": 0.9847265748655155, "1000–3000 Hz_40ms_modulation_mae_db": 1.7558725351957942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6123664928332926, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1744586001060361, "3000–6000 Hz_10ms_envelope_correlation": 0.9815586593888164, "3000–6000 Hz_10ms_modulation_mae_db": 0.2942920563393914, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4009596782067066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10242789975862693, "3000–6000 Hz_40ms_envelope_correlation": 0.9935120255793011, "3000–6000 Hz_40ms_modulation_mae_db": 0.47122113913965635, "mr_stft_db": 5.432030518279213, "band_mae_db": 0.19665076459796438, "rms_gain_error_db": -0.1222448086889761, "output_peak": 0.5406467583386371}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6900939542600627, "全帯域_10ms_rms_envelope_nrmse": 0.10160832382053119, "全帯域_10ms_envelope_correlation": 0.9899640149986211, "全帯域_10ms_modulation_mae_db": 0.6148972572676142, "全帯域_40ms_log_rms_mae_db": 0.43608789114878777, "全帯域_40ms_rms_envelope_nrmse": 0.048802768416973875, "全帯域_40ms_envelope_correlation": 0.9976023456353589, "全帯域_40ms_modulation_mae_db": 0.2999273012083968, "300–1000 Hz_10ms_log_rms_mae_db": 1.3407762219114991, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15998649589393182, "300–1000 Hz_10ms_envelope_correlation": 0.9810939770709439, "300–1000 Hz_10ms_modulation_mae_db": 0.48242753403943345, "300–1000 Hz_40ms_log_rms_mae_db": 1.060043089152131, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08417699342886319, "300–1000 Hz_40ms_envelope_correlation": 0.9944643303642504, "300–1000 Hz_40ms_modulation_mae_db": 0.217224894806615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.621591687357771, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23809234861209344, "1000–3000 Hz_10ms_envelope_correlation": 0.9696709519572189, "1000–3000 Hz_10ms_modulation_mae_db": 2.452461088679121, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3578082280449282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1652200995134959, "1000–3000 Hz_40ms_envelope_correlation": 0.9900939024090814, "1000–3000 Hz_40ms_modulation_mae_db": 2.6387270956309816, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6770706859020879, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1841092856101456, "3000–6000 Hz_10ms_envelope_correlation": 0.979700855003215, "3000–6000 Hz_10ms_modulation_mae_db": 0.6675685439310796, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4729172218469213, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10300659541469036, "3000–6000 Hz_40ms_envelope_correlation": 0.9934874068919395, "3000–6000 Hz_40ms_modulation_mae_db": 0.8038601631117618, "mr_stft_db": 5.4438675897306865, "band_mae_db": 0.40503976895601057, "rms_gain_error_db": -0.09687273253393941, "output_peak": 0.5931742696854466}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6701415599174818, "全帯域_10ms_rms_envelope_nrmse": 0.08463500525795555, "全帯域_10ms_envelope_correlation": 0.9930175180888624, "全帯域_10ms_modulation_mae_db": 0.31415520347288406, "全帯域_40ms_log_rms_mae_db": 0.41404524036134177, "全帯域_40ms_rms_envelope_nrmse": 0.04904062344191993, "全帯域_40ms_envelope_correlation": 0.9974894556713864, "全帯域_40ms_modulation_mae_db": 0.4604912250182244, "300–1000 Hz_10ms_log_rms_mae_db": 1.3098892751136015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14664797089970566, "300–1000 Hz_10ms_envelope_correlation": 0.9840174305242575, "300–1000 Hz_10ms_modulation_mae_db": 0.39007122381066833, "300–1000 Hz_40ms_log_rms_mae_db": 1.0332735317889934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09252539178146411, "300–1000 Hz_40ms_envelope_correlation": 0.9932969691397971, "300–1000 Hz_40ms_modulation_mae_db": 0.628809470517409, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6200551590512962, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2407093374081604, "1000–3000 Hz_10ms_envelope_correlation": 0.9636506886275743, "1000–3000 Hz_10ms_modulation_mae_db": 2.104647889268893, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3523388805303869, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14141449002201845, "1000–3000 Hz_40ms_envelope_correlation": 0.9872814030389512, "1000–3000 Hz_40ms_modulation_mae_db": 1.9281583789123868, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5993307281128673, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16283013327411777, "3000–6000 Hz_10ms_envelope_correlation": 0.9846372920635522, "3000–6000 Hz_10ms_modulation_mae_db": 0.21291458697283874, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3866677839428327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09293813430588459, "3000–6000 Hz_40ms_envelope_correlation": 0.995064228986326, "3000–6000 Hz_40ms_modulation_mae_db": 0.2524581681583811, "mr_stft_db": 5.446643484052545, "band_mae_db": 0.15179171016005327, "rms_gain_error_db": -0.03916413696640994, "output_peak": 0.48149007065334404}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.696461599035607, "全帯域_10ms_rms_envelope_nrmse": 0.09693257863450257, "全帯域_10ms_envelope_correlation": 0.9908628805769941, "全帯域_10ms_modulation_mae_db": 0.5495159042339823, "全帯域_40ms_log_rms_mae_db": 0.41459017147353305, "全帯域_40ms_rms_envelope_nrmse": 0.045252690781721804, "全帯域_40ms_envelope_correlation": 0.997954778562222, "全帯域_40ms_modulation_mae_db": 0.7429948736470227, "300–1000 Hz_10ms_log_rms_mae_db": 1.3398419112309363, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1386211152523005, "300–1000 Hz_10ms_envelope_correlation": 0.9857892921366409, "300–1000 Hz_10ms_modulation_mae_db": 0.15643552889692428, "300–1000 Hz_40ms_log_rms_mae_db": 0.9652135817909617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07181587127279965, "300–1000 Hz_40ms_envelope_correlation": 0.9960204895289984, "300–1000 Hz_40ms_modulation_mae_db": 0.3208061167117429, "1000–3000 Hz_10ms_log_rms_mae_db": 1.603988187735267, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20478915245386864, "1000–3000 Hz_10ms_envelope_correlation": 0.9756846733680138, "1000–3000 Hz_10ms_modulation_mae_db": 1.096299833636074, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3031580187363971, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12252250754467406, "1000–3000 Hz_40ms_envelope_correlation": 0.9935056964711085, "1000–3000 Hz_40ms_modulation_mae_db": 1.047117706778436, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6312772020461457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15715578732788288, "3000–6000 Hz_10ms_envelope_correlation": 0.9851057741442566, "3000–6000 Hz_10ms_modulation_mae_db": 0.3218994166795815, "3000–6000 Hz_40ms_log_rms_mae_db": 1.410846593713802, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09269766370696784, "3000–6000 Hz_40ms_envelope_correlation": 0.9946914195133554, "3000–6000 Hz_40ms_modulation_mae_db": 0.5476017152468334, "mr_stft_db": 5.421770798912216, "band_mae_db": 0.3514772887550212, "rms_gain_error_db": -0.09679399473795412, "output_peak": 0.5431183506527602}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.515855329325583, "spectral_convergence": 0.3147950294778251, "spectral_convergence_status": "ok", "amplitude_floor": 1.6898412267444875e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.237179573320262, "spectral_convergence": 0.31685717698686156, "spectral_convergence_status": "ok", "amplitude_floor": 1.0945626100175482e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.561292555438582, "spectral_convergence": 0.34454817976526986, "spectral_convergence_status": "ok", "amplitude_floor": 6.58943811507865e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017588563541266742, "output_power": 0.0017091878174282649, "delta_db": -0.1244058341992061, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012071674288590822, "output_power": 0.001211810357754823, "delta_db": 0.016671512128753452, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003927549262280726, "output_power": 0.00041894315305634984, "delta_db": 0.28033455983927485, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003207071656425, "output_power": 0.00022730533412252969, "delta_db": -1.4949903625560903, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.4115036625953016e-05, "output_power": 5.244481489890767e-05, "delta_db": -0.13615399788934276, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.305762815858983e-05, "output_power": 4.059205666710301e-05, "delta_db": -0.25609046754645703, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3847744556931874, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13592992441920956, "pitch": null, "output_rms_dbfs": -24.364852718594182, "output_peak": 0.5247168935688251, "reference_rms_dbfs": -24.228922794174977, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.4381091526948095}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002198976712894503, "log_rms_mae_db": 0.6981364087499667, "rms_envelope_nrmse": 0.12164750217850853, "envelope_correlation": 0.9855781465128293, "rms_gain_error_db": -0.1350858622297878, "reference_mean_power": 0.0037766562988771884, "output_mean_power": 0.003660992816184911}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.069519300574134e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006799849644412746, "output_power": 0.0005841501577164421, "reference_active": true, "delta_db": -0.6597481157850262}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022275189153607582, "output_power": 0.0002077403313498121, "reference_active": true, "delta_db": -0.3030058061762122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.238302101062001e-05, "output_power": 4.342915270703064e-05, "reference_active": true, "delta_db": 0.10589446992585476}], "active_band_count": 3, "active_band_mae_db": 0.35621613062903107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018113949472991762, "log_rms_mae_db": 0.4186976800011373, "rms_envelope_nrmse": 0.07090713454593864, "envelope_correlation": 0.9949148960512011, "rms_gain_error_db": -0.13632972981138242, "reference_mean_power": 0.003780325824366865, "output_mean_power": 0.0036635005416916166}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9093794448028363e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006511405512347742, "output_power": 0.0005577029336198833, "reference_active": true, "delta_db": -0.6727181388340846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001166725476043267, "output_power": 0.00010426213756347943, "reference_active": true, "delta_db": -0.4884205650205523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.399065253982206e-06, "output_power": 2.930799821100527e-06, "reference_active": true, "delta_db": -0.6437334572884073}], "active_band_count": 3, "active_band_mae_db": 0.6016240537143481}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013526658282762358, "log_rms_mae_db": 1.2758890164132046, "rms_envelope_nrmse": 0.16403391398352735, "envelope_correlation": 0.9799772627679145, "rms_gain_error_db": -0.15335542409446426, "reference_mean_power": 0.001047203990993015, "output_mean_power": 0.0010108710201284228}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.584657585149468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003504540347149448, "output_power": 0.00032273914135501705, "reference_active": true, "delta_db": -0.35779425073516047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.5576057457372e-05, "output_power": 9.472838439687806e-05, "reference_active": true, "delta_db": -0.038689813416854554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9623586773434205e-05, "output_power": 2.561532304835519e-05, "reference_active": true, "delta_db": 1.1572144728910567}], "active_band_count": 3, "active_band_mae_db": 0.5178995123476906}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012031399971827572, "log_rms_mae_db": 0.9515052831305203, "rms_envelope_nrmse": 0.07662690486139737, "envelope_correlation": 0.9955520460993749, "rms_gain_error_db": -0.15365954250143474, "reference_mean_power": 0.0010510881038568295, "output_mean_power": 0.0010145493256818025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.061243701565635e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00034327462417827815, "output_power": 0.0003222401014173751, "reference_active": true, "delta_db": -0.2746211479040088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.625199841508274e-05, "output_power": 4.548200953333902e-05, "reference_active": true, "delta_db": -0.07290857199902523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5417688576161987e-06, "output_power": 1.4583776352181181e-06, "reference_active": true, "delta_db": -0.241492734295165}], "active_band_count": 3, "active_band_mae_db": 0.19634081806606632}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013601220330575488, "log_rms_mae_db": 1.5890235890458662, "rms_envelope_nrmse": 0.2632282740762697, "envelope_correlation": 0.958644181385783, "rms_gain_error_db": -0.8387454905492451, "reference_mean_power": 0.00024923569921471797, "output_mean_power": 0.00020546398140247496}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.70738105242738e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001701759171968848, "output_power": 0.00012425806377206616, "reference_active": true, "delta_db": -1.3657351798161006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6462817859538857e-05, "output_power": 1.03604628774266e-05, "reference_active": true, "delta_db": -4.072569286658731}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.097841074264338e-06, "output_power": 5.0462028653455e-06, "reference_active": true, "delta_db": 2.119055720802496}], "active_band_count": 3, "active_band_mae_db": 2.5191200624257757}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011511226838452259, "log_rms_mae_db": 1.305442386013484, "rms_envelope_nrmse": 0.1839656530951573, "envelope_correlation": 0.9824648586184773, "rms_gain_error_db": -0.8427343024208861, "reference_mean_power": 0.00024963838926192667, "output_mean_power": 0.0002056070213618599}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6223780118670214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016948627965662844, "output_power": 0.00012169274167570569, "reference_active": true, "delta_db": -1.4386987107603368}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1767373976886565e-05, "output_power": 3.7798804788989325e-06, "reference_active": true, "delta_db": -4.932014883955078}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8952968721807594e-07, "output_power": 4.218730256440287e-07, "reference_active": true, "delta_db": 3.475045118035739}], "active_band_count": 3, "active_band_mae_db": 3.281919570917051}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.3800826555528964e-05, "log_rms_mae_db": 1.611724893209612, "rms_envelope_nrmse": 0.16066270948051867, "envelope_correlation": 0.9843996607216396, "rms_gain_error_db": -0.14470467866608983, "reference_mean_power": 4.104766861392576e-05, "output_mean_power": 3.970251583615551e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.890895346126465e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.549789437651056e-06, "output_power": 9.64273828263851e-06, "reference_active": true, "delta_db": 0.04206583589770836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.275656140671785e-06, "output_power": 3.0801999347408554e-06, "reference_active": true, "delta_db": -0.26719398448521603}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.939546621487433e-07, "output_power": 5.414726700320191e-07, "reference_active": true, "delta_db": -0.4017675440957586}], "active_band_count": 3, "active_band_mae_db": 0.23700912149289433}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.772714769720092e-05, "log_rms_mae_db": 1.396331206566409, "rms_envelope_nrmse": 0.08974464275148433, "envelope_correlation": 0.995184382753728, "rms_gain_error_db": -0.14550487597239045, "reference_mean_power": 4.122071638798201e-05, "output_mean_power": 3.9862547307209595e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.778938094011469e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.936665760227185e-06, "output_power": 9.985574838522194e-06, "reference_active": true, "delta_db": 0.021323892146124273}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7465237791634235e-06, "output_power": 1.6541650445370407e-06, "reference_active": true, "delta_db": -0.23595663825500385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8043700755065915e-08, "output_power": 3.567040049197159e-08, "reference_active": true, "delta_db": -1.2932846782238334}], "active_band_count": 3, "active_band_mae_db": 0.5168550695416538}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8967152366755056, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f0f21234c9f69b3fdf846ee5013500c1b4c799761a0ddd0d85a7a721f0bcfa6e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9662219417204589, "standard_deviation": 0.008266727660000505, "minimum": 0.9545786574200059, "maximum": 0.9772384851822479, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6125551751493776, "standard_deviation": 0.02325874506437633, "minimum": 1.5880400885639305, "maximum": 1.6604176016664898, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.911740405262381, "standard_deviation": 0.4663943337722754, "minimum": 1.096299833636074, "maximum": 2.5191200624257757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2349626527686577, "standard_deviation": 0.027377328491219176, "minimum": 0.1928887331804061, "maximum": 0.2685666800502507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893976356567957, "standard_deviation": 0.004289169308406184, "minimum": 0.9824648586184773, "maximum": 0.9948585868177047, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3295568199727401, "standard_deviation": 0.024703268454509427, "minimum": 1.3031580187363971, "maximum": 1.3578082280449282, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 2.026382522046597, "standard_deviation": 0.6835104301834596, "minimum": 1.047117706778436, "maximum": 3.281919570917051, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13819842049283493, "standard_deviation": 0.029512848104605516, "minimum": 0.09040225664695409, "maximum": 0.1839656530951573, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9829456525145941, "standard_deviation": 0.0019376319050303308, "minimum": 0.979700855003215, "maximum": 0.9851057741442566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.632472463061419, "standard_deviation": 0.024777738011324373, "minimum": 1.5993307281128673, "maximum": 1.6770706859020879, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4695815296136116, "standard_deviation": 0.21989837506029777, "minimum": 0.21291458697283874, "maximum": 0.6898351214867374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16867501912672156, "standard_deviation": 0.009423285292626496, "minimum": 0.15715578732788288, "maximum": 0.1841092856101456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940235276998644, "standard_deviation": 0.0010275539936102213, "minimum": 0.99219448870046, "maximum": 0.995184382753728, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4165507737195755, "standard_deviation": 0.026395929777021484, "minimum": 1.3866677839428327, "maximum": 1.4729172218469213, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6455107917513919, "standard_deviation": 0.23254838113592197, "minimum": 0.2524581681583811, "maximum": 0.9054901661705399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0996659139580224, "standard_deviation": 0.008149807318731855, "minimum": 0.08974464275148433, "maximum": 0.11348565530880951, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9820286448214555, "standard_deviation": 0.004350835333154972, "minimum": 0.9733324957361811, "maximum": 0.9864716176687441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.314670191983879, "standard_deviation": 0.03492239756931024, "minimum": 1.2636967739539435, "maximum": 1.369476104241568, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4647667690565954, "standard_deviation": 0.15579177975665362, "minimum": 0.15643552889692428, "maximum": 0.6558300845311368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15569042911216532, "standard_deviation": 0.017654672412507604, "minimum": 0.1382188397312259, "maximum": 0.19030386328523918, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9944359668579066, "standard_deviation": 0.002193543539946377, "minimum": 0.9902751252959304, "maximum": 0.9966611941386279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.001097565522472, "standard_deviation": 0.036410471369633025, "minimum": 0.9515052831305203, "maximum": 1.060043089152131, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4741511398956834, "standard_deviation": 0.23757456976877867, "minimum": 0.19634081806606632, "maximum": 0.868045156571382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08666494142389722, "standard_deviation": 0.015307889338297423, "minimum": 0.07181587127279965, "maximum": 0.11655228846954317, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2904281822444156, "standard_deviation": 0.09158988813585837, "minimum": 0.15179171016005327, "maximum": 0.40503976895601057, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.436986049350314, "standard_deviation": 0.008087124681552491, "minimum": 5.421770798912216, "maximum": 5.446643484052545, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5380411211975127, "standard_deviation": 0.046693618544528886, "minimum": 0.4650567447956773, "maximum": 0.5931742696854466, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11324112453175125, "standard_deviation": 0.03676437555566251, "minimum": -0.15377358296581667, "maximum": -0.03916413696640994, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900878992164552, "standard_deviation": 0.0023522786201078216, "minimum": 0.9855781465128293, "maximum": 0.9930175180888624, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6946565099470123, "standard_deviation": 0.029377180995617078, "minimum": 0.6605275335666498, "maximum": 0.745806688873048, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4382105416148669, "standard_deviation": 0.10749895081998549, "minimum": 0.31415520347288406, "maximum": 0.6148972572676142, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10055273998051423, "standard_deviation": 0.011536791753866087, "minimum": 0.08463500525795555, "maximum": 0.12164750217850853, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972052816504533, "standard_deviation": 0.0009985630179970052, "minimum": 0.9949148960512011, "maximum": 0.9981312685499291, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.42580876426573055, "standard_deviation": 0.02205425874658768, "minimum": 0.39370389510936804, "maximum": 0.467370922372422, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47100059033403097, "standard_deviation": 0.20910639194711259, "minimum": 0.07503910043145676, "maximum": 0.7429948736470227, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.052586432776518165, "standard_deviation": 0.008225272629279981, "minimum": 0.045252690781721804, "maximum": 0.07090713454593864, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-79740-0011", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.830599784851074, "predicted_mos": 2.830599784851074, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "aa7bcfe460cdf75b9206a82e84093187c7aae8b1564bb8ee2481fc5a71870834", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.000608999747783, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7473922610957785, "全帯域_10ms_rms_envelope_nrmse": 0.12386061687366662, "全帯域_10ms_envelope_correlation": 0.9851182763666029, "全帯域_10ms_modulation_mae_db": 0.8167296035478201, "全帯域_40ms_log_rms_mae_db": 0.4523255707131807, "全帯域_40ms_rms_envelope_nrmse": 0.05270848080941795, "全帯域_40ms_envelope_correlation": 0.9974158891621098, "全帯域_40ms_modulation_mae_db": 0.539443828816755, "300–1000 Hz_10ms_log_rms_mae_db": 1.3492427696465439, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1642098174005739, "300–1000 Hz_10ms_envelope_correlation": 0.9800371731709856, "300–1000 Hz_10ms_modulation_mae_db": 0.6436777717495593, "300–1000 Hz_40ms_log_rms_mae_db": 1.0011633070474895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07750058087291456, "300–1000 Hz_40ms_envelope_correlation": 0.9953751830451218, "300–1000 Hz_40ms_modulation_mae_db": 0.5253154158302051, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6371014784761688, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2876386882888918, "1000–3000 Hz_10ms_envelope_correlation": 0.9478395243174716, "1000–3000 Hz_10ms_modulation_mae_db": 3.4801546006373023, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3726385209247962, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12782073562124366, "1000–3000 Hz_40ms_envelope_correlation": 0.9906937351018583, "1000–3000 Hz_40ms_modulation_mae_db": 3.35668872194479, "3000–6000 Hz_10ms_log_rms_mae_db": 1.614995020928597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17742350219061598, "3000–6000 Hz_10ms_envelope_correlation": 0.9813860135554061, "3000–6000 Hz_10ms_modulation_mae_db": 0.371882118415441, "3000–6000 Hz_40ms_log_rms_mae_db": 1.396396522259968, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10249366499085764, "3000–6000 Hz_40ms_envelope_correlation": 0.9936697698387644, "3000–6000 Hz_40ms_modulation_mae_db": 0.6997832821125702, "mr_stft_db": 5.409916697210327, "band_mae_db": 0.36111838304189625, "rms_gain_error_db": -0.1697000972502214, "output_peak": 0.5991260904264106}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6926270214138293, "全帯域_10ms_rms_envelope_nrmse": 0.10258177034794112, "全帯域_10ms_envelope_correlation": 0.989763088358441, "全帯域_10ms_modulation_mae_db": 0.3313167577902209, "全帯域_40ms_log_rms_mae_db": 0.418757274615322, "全帯域_40ms_rms_envelope_nrmse": 0.04879552873818213, "全帯域_40ms_envelope_correlation": 0.9976244382721859, "全帯域_40ms_modulation_mae_db": 0.17606167103827794, "300–1000 Hz_10ms_log_rms_mae_db": 1.2397089628129854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13621916883475377, "300–1000 Hz_10ms_envelope_correlation": 0.986273857282472, "300–1000 Hz_10ms_modulation_mae_db": 0.5203697563755311, "300–1000 Hz_40ms_log_rms_mae_db": 0.9452240197205435, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07124438667149269, "300–1000 Hz_40ms_envelope_correlation": 0.9961740267034853, "300–1000 Hz_40ms_modulation_mae_db": 0.796403401447766, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5541131778662816, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2333183823360052, "1000–3000 Hz_10ms_envelope_correlation": 0.9660138389848606, "1000–3000 Hz_10ms_modulation_mae_db": 2.294709975865999, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3143602345926497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10837714594834229, "1000–3000 Hz_40ms_envelope_correlation": 0.9934256932103624, "1000–3000 Hz_40ms_modulation_mae_db": 2.3858745749951322, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6021262924726418, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15431086020528528, "3000–6000 Hz_10ms_envelope_correlation": 0.9857809906961302, "3000–6000 Hz_10ms_modulation_mae_db": 0.250614576557275, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4054453534756137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09198340606098812, "3000–6000 Hz_40ms_envelope_correlation": 0.9948624581371154, "3000–6000 Hz_40ms_modulation_mae_db": 0.23279740433785043, "mr_stft_db": 5.410107124674884, "band_mae_db": 0.24836967580290512, "rms_gain_error_db": -0.10254606270919572, "output_peak": 0.5605971478936178}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.699221683348064, "全帯域_10ms_rms_envelope_nrmse": 0.10399272171307114, "全帯域_10ms_envelope_correlation": 0.9894704704840045, "全帯域_10ms_modulation_mae_db": 0.5520104898194335, "全帯域_40ms_log_rms_mae_db": 0.4245941927508786, "全帯域_40ms_rms_envelope_nrmse": 0.049051748756339546, "全帯域_40ms_envelope_correlation": 0.9975440907971141, "全帯域_40ms_modulation_mae_db": 0.2744984384591579, "300–1000 Hz_10ms_log_rms_mae_db": 1.3191619523159805, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16369476310110342, "300–1000 Hz_10ms_envelope_correlation": 0.9801240483211037, "300–1000 Hz_10ms_modulation_mae_db": 0.29520907465542906, "300–1000 Hz_40ms_log_rms_mae_db": 1.024332444220432, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1058694398581857, "300–1000 Hz_40ms_envelope_correlation": 0.9913530165088363, "300–1000 Hz_40ms_modulation_mae_db": 0.3057144212530785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6302930614617808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21405202405227866, "1000–3000 Hz_10ms_envelope_correlation": 0.9714063196034303, "1000–3000 Hz_10ms_modulation_mae_db": 1.5001533745008002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3065665065518377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09695052426148783, "1000–3000 Hz_40ms_envelope_correlation": 0.9940492620228543, "1000–3000 Hz_40ms_modulation_mae_db": 1.4800575749988572, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6180011536101497, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18482631398825253, "3000–6000 Hz_10ms_envelope_correlation": 0.9793866570004892, "3000–6000 Hz_10ms_modulation_mae_db": 0.34675904351120695, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3881039679831697, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11382763174282849, "3000–6000 Hz_40ms_envelope_correlation": 0.9920066575403823, "3000–6000 Hz_40ms_modulation_mae_db": 0.5033983308159747, "mr_stft_db": 5.404525554419504, "band_mae_db": 0.1783840614271753, "rms_gain_error_db": -0.08075263857720552, "output_peak": 0.5942473018747011}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6562383452015412, "全帯域_10ms_rms_envelope_nrmse": 0.0905500596115043, "全帯域_10ms_envelope_correlation": 0.9920220893245544, "全帯域_10ms_modulation_mae_db": 0.7274609365735208, "全帯域_40ms_log_rms_mae_db": 0.42502008660170665, "全帯域_40ms_rms_envelope_nrmse": 0.05821497317637142, "全帯域_40ms_envelope_correlation": 0.9965071301513206, "全帯域_40ms_modulation_mae_db": 0.9373624368799846, "300–1000 Hz_10ms_log_rms_mae_db": 1.257582122345638, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14017603232157688, "300–1000 Hz_10ms_envelope_correlation": 0.9854340987933162, "300–1000 Hz_10ms_modulation_mae_db": 0.35039628974971504, "300–1000 Hz_40ms_log_rms_mae_db": 0.9260250467064188, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0771632795958968, "300–1000 Hz_40ms_envelope_correlation": 0.9953728554766262, "300–1000 Hz_40ms_modulation_mae_db": 0.6370624102688516, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5532321018233821, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2185058269780564, "1000–3000 Hz_10ms_envelope_correlation": 0.9711257883933496, "1000–3000 Hz_10ms_modulation_mae_db": 1.110614657978905, "1000–3000 Hz_40ms_log_rms_mae_db": 1.312618062587158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1458808433911286, "1000–3000 Hz_40ms_envelope_correlation": 0.9882668375199184, "1000–3000 Hz_40ms_modulation_mae_db": 1.5628537295045453, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6447448093562456, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17042974239640465, "3000–6000 Hz_10ms_envelope_correlation": 0.9829022586405125, "3000–6000 Hz_10ms_modulation_mae_db": 0.36729423865446503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4298162785328092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10172623465575914, "3000–6000 Hz_40ms_envelope_correlation": 0.9938311340282208, "3000–6000 Hz_40ms_modulation_mae_db": 0.5006847192340382, "mr_stft_db": 5.406646038927522, "band_mae_db": 0.2819454531546211, "rms_gain_error_db": -0.08045436038352947, "output_peak": 0.5543133933407812}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7228123153630965, "全帯域_10ms_rms_envelope_nrmse": 0.09986956718125768, "全帯域_10ms_envelope_correlation": 0.9904060565607425, "全帯域_10ms_modulation_mae_db": 0.34782479323553367, "全帯域_40ms_log_rms_mae_db": 0.4412852405491268, "全帯域_40ms_rms_envelope_nrmse": 0.06289791096814393, "全帯域_40ms_envelope_correlation": 0.9961076624164253, "全帯域_40ms_modulation_mae_db": 0.3589758189399859, "300–1000 Hz_10ms_log_rms_mae_db": 1.2887432331455035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15818136604048058, "300–1000 Hz_10ms_envelope_correlation": 0.9814493144827243, "300–1000 Hz_10ms_modulation_mae_db": 0.30157681485596005, "300–1000 Hz_40ms_log_rms_mae_db": 0.948746019438503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0922249604848062, "300–1000 Hz_40ms_envelope_correlation": 0.9933441512294963, "300–1000 Hz_40ms_modulation_mae_db": 0.35104579342347625, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5320426506038167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2410915605823893, "1000–3000 Hz_10ms_envelope_correlation": 0.9664060677307253, "1000–3000 Hz_10ms_modulation_mae_db": 2.5728877572357534, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3004912975399416, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17335310339909352, "1000–3000 Hz_40ms_envelope_correlation": 0.9852662793110083, "1000–3000 Hz_40ms_modulation_mae_db": 2.4326634342783406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6337775367724303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16780270220121796, "3000–6000 Hz_10ms_envelope_correlation": 0.98487584485306, "3000–6000 Hz_10ms_modulation_mae_db": 0.3374796508349247, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4440734313937569, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1093226546359709, "3000–6000 Hz_40ms_envelope_correlation": 0.994027493425935, "3000–6000 Hz_40ms_modulation_mae_db": 0.36320892661744936, "mr_stft_db": 5.410499061916635, "band_mae_db": 0.3958853793379378, "rms_gain_error_db": -0.1513499074011631, "output_peak": 0.5348003313267763}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6947950057313361, "全帯域_10ms_rms_envelope_nrmse": 0.09698360262360735, "全帯域_10ms_envelope_correlation": 0.9908635928650772, "全帯域_10ms_modulation_mae_db": 0.3798161112113005, "全帯域_40ms_log_rms_mae_db": 0.4222621297482961, "全帯域_40ms_rms_envelope_nrmse": 0.05021999598240039, "全帯域_40ms_envelope_correlation": 0.9974622274937607, "全帯域_40ms_modulation_mae_db": 0.5693026752778758, "300–1000 Hz_10ms_log_rms_mae_db": 1.2521304434829337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13966342896176487, "300–1000 Hz_10ms_envelope_correlation": 0.9855822020986291, "300–1000 Hz_10ms_modulation_mae_db": 0.7538674254827802, "300–1000 Hz_40ms_log_rms_mae_db": 0.9149196674735491, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07715279108464283, "300–1000 Hz_40ms_envelope_correlation": 0.9955253217686081, "300–1000 Hz_40ms_modulation_mae_db": 1.0060695220168563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.61482674445369, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22756759013878766, "1000–3000 Hz_10ms_envelope_correlation": 0.9676647417667706, "1000–3000 Hz_10ms_modulation_mae_db": 1.6149509602488308, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3622077711855638, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12438469443766777, "1000–3000 Hz_40ms_envelope_correlation": 0.9901208714362795, "1000–3000 Hz_40ms_modulation_mae_db": 1.7961714495922656, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6093804340344868, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1703266440319559, "3000–6000 Hz_10ms_envelope_correlation": 0.9827355469050388, "3000–6000 Hz_10ms_modulation_mae_db": 0.1968304091572087, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4414272405812842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10286739113125144, "3000–6000 Hz_40ms_envelope_correlation": 0.993540852436987, "3000–6000 Hz_40ms_modulation_mae_db": 0.3287982042347129, "mr_stft_db": 5.415854581270103, "band_mae_db": 0.21317689389025227, "rms_gain_error_db": -0.09858993567449359, "output_peak": 0.4901786949326572}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6803234958972414, "全帯域_10ms_rms_envelope_nrmse": 0.10428854036614926, "全帯域_10ms_envelope_correlation": 0.9894063978121619, "全帯域_10ms_modulation_mae_db": 0.48158243757346947, "全帯域_40ms_log_rms_mae_db": 0.43772121557446314, "全帯域_40ms_rms_envelope_nrmse": 0.05547182309821551, "全帯域_40ms_envelope_correlation": 0.9967868451504965, "全帯域_40ms_modulation_mae_db": 0.2564020638216876, "300–1000 Hz_10ms_log_rms_mae_db": 1.2637423984097267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14270147582887055, "300–1000 Hz_10ms_envelope_correlation": 0.9849386509901616, "300–1000 Hz_10ms_modulation_mae_db": 0.6071593497319518, "300–1000 Hz_40ms_log_rms_mae_db": 0.9146012763006768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06621436503058292, "300–1000 Hz_40ms_envelope_correlation": 0.9966730049037891, "300–1000 Hz_40ms_modulation_mae_db": 0.431543092944306, "1000–3000 Hz_10ms_log_rms_mae_db": 1.565715717629088, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22527874990688904, "1000–3000 Hz_10ms_envelope_correlation": 0.9681916151158687, "1000–3000 Hz_10ms_modulation_mae_db": 1.105717113138004, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2907277678509694, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14179020075885568, "1000–3000 Hz_40ms_envelope_correlation": 0.9872494109854562, "1000–3000 Hz_40ms_modulation_mae_db": 1.4559835312025236, "3000–6000 Hz_10ms_log_rms_mae_db": 1.604088182673025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14484205772311914, "3000–6000 Hz_10ms_envelope_correlation": 0.9874829791647332, "3000–6000 Hz_10ms_modulation_mae_db": 0.27881983152359363, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3881913649523077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09016846673988799, "3000–6000 Hz_40ms_envelope_correlation": 0.9950625348223261, "3000–6000 Hz_40ms_modulation_mae_db": 0.34725113866393115, "mr_stft_db": 5.41628553817196, "band_mae_db": 0.26568218999630505, "rms_gain_error_db": -0.028291495751921955, "output_peak": 0.5491253825623524}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6580067762317716, "全帯域_10ms_rms_envelope_nrmse": 0.09590371318059016, "全帯域_10ms_envelope_correlation": 0.9910442389598553, "全帯域_10ms_modulation_mae_db": 0.36191243851279803, "全帯域_40ms_log_rms_mae_db": 0.38819364044013493, "全帯域_40ms_rms_envelope_nrmse": 0.04396287148637638, "全帯域_40ms_envelope_correlation": 0.9979830698459886, "全帯域_40ms_modulation_mae_db": 0.3076056597339332, "300–1000 Hz_10ms_log_rms_mae_db": 1.297968347116677, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15030568409654937, "300–1000 Hz_10ms_envelope_correlation": 0.983249976454097, "300–1000 Hz_10ms_modulation_mae_db": 0.4162027419591814, "300–1000 Hz_40ms_log_rms_mae_db": 0.9715958887058129, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07299846197993355, "300–1000 Hz_40ms_envelope_correlation": 0.9960202834645051, "300–1000 Hz_40ms_modulation_mae_db": 0.5185965005320347, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5890538134409737, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22866294757183653, "1000–3000 Hz_10ms_envelope_correlation": 0.9682075744349038, "1000–3000 Hz_10ms_modulation_mae_db": 1.8947973360107413, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2906756558677177, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1055931611251138, "1000–3000 Hz_40ms_envelope_correlation": 0.9930371692611227, "1000–3000 Hz_40ms_modulation_mae_db": 2.0433932873814933, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5871991900808553, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15189687909340516, "3000–6000 Hz_10ms_envelope_correlation": 0.9860832063373254, "3000–6000 Hz_10ms_modulation_mae_db": 0.08110211556699914, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3947436846906451, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.091438078551594, "3000–6000 Hz_40ms_envelope_correlation": 0.9949129192372754, "3000–6000 Hz_40ms_modulation_mae_db": 0.2983093074056191, "mr_stft_db": 5.408186199594602, "band_mae_db": 0.15591721888952184, "rms_gain_error_db": -0.01851015341455947, "output_peak": 0.6175371278802309}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.4878520352188636, "spectral_convergence": 0.3209958947358993, "spectral_convergence_status": "ok", "amplitude_floor": 1.6898412267444875e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.204244921466151, "spectral_convergence": 0.3309381620313787, "spectral_convergence_status": "ok", "amplitude_floor": 1.0945626100175482e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.537653134945966, "spectral_convergence": 0.3541345911867455, "spectral_convergence_status": "ok", "amplitude_floor": 6.58943811507865e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017588563541266742, "output_power": 0.0016927588749845513, "delta_db": -0.16635272614069632, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012071674288590822, "output_power": 0.0011777248791431444, "delta_db": -0.10723659512752351, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003927549262280726, "output_power": 0.0004205270575718614, "delta_db": 0.29672303434430497, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003207071656425, "output_power": 0.0002440612404220948, "delta_db": -1.1860984938111834, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.4115036625953016e-05, "output_power": 5.167230136151368e-05, "delta_db": -0.2006015230004724, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.305762815858983e-05, "output_power": 4.518766448554139e-05, "delta_db": 0.20969792582719707, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.36111838304189625, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1697000972502214, "pitch": null, "output_rms_dbfs": -24.398622891425198, "output_peak": 0.5991260904264106, "reference_rms_dbfs": -24.228922794174977, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.409916697210327}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002198976712894503, "log_rms_mae_db": 0.7473922610957785, "rms_envelope_nrmse": 0.12386061687366662, "envelope_correlation": 0.9851182763666029, "rms_gain_error_db": -0.16730935965948524, "reference_mean_power": 0.0037766562988771884, "output_mean_power": 0.0036339297461006254}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.069519300574134e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006799849644412746, "output_power": 0.0006181723554819888, "reference_active": true, "delta_db": -0.4138973024700713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022275189153607582, "output_power": 0.0001990872245258279, "reference_active": true, "delta_db": -0.4877800846884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.238302101062001e-05, "output_power": 6.0540088545003e-05, "reference_active": true, "delta_db": 1.5485114234849888}], "active_band_count": 3, "active_band_mae_db": 0.8167296035478201}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018113949472991762, "log_rms_mae_db": 0.4523255707131807, "rms_envelope_nrmse": 0.05270848080941795, "envelope_correlation": 0.9974158891621098, "rms_gain_error_db": -0.1645857458784934, "reference_mean_power": 0.003780325824366865, "output_mean_power": 0.003639742488910441}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9093794448028363e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006511405512347742, "output_power": 0.0005974066821076431, "reference_active": true, "delta_db": -0.37404666887873517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001166725476043267, "output_power": 0.00010837906561165268, "reference_active": true, "delta_db": -0.3202327858150017}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.399065253982206e-06, "output_power": 4.204987462585421e-06, "reference_active": true, "delta_db": 0.9240520317565283}], "active_band_count": 3, "active_band_mae_db": 0.539443828816755}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013526658282762358, "log_rms_mae_db": 1.3492427696465439, "rms_envelope_nrmse": 0.1642098174005739, "envelope_correlation": 0.9800371731709856, "rms_gain_error_db": -0.10909146016227368, "reference_mean_power": 0.001047203990993015, "output_mean_power": 0.0010212266570741547}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.584657585149468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003504540347149448, "output_power": 0.00033843465882330383, "reference_active": true, "delta_db": -0.1515623190765676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.5576057457372e-05, "output_power": 8.703019090773476e-05, "reference_active": true, "delta_db": -0.4067917558608744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9623586773434205e-05, "output_power": 2.6918218503432648e-05, "reference_active": true, "delta_db": 1.372679240311236}], "active_band_count": 3, "active_band_mae_db": 0.6436777717495593}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012031399971827572, "log_rms_mae_db": 1.0011633070474895, "rms_envelope_nrmse": 0.07750058087291456, "envelope_correlation": 0.9953751830451218, "rms_gain_error_db": -0.11002422931579227, "reference_mean_power": 0.0010510881038568295, "output_mean_power": 0.0010247942922388326}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.061243701565635e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00034327462417827815, "output_power": 0.0003367086305091779, "reference_active": true, "delta_db": -0.08387451989150245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.625199841508274e-05, "output_power": 4.3505669682012316e-05, "reference_active": true, "delta_db": -0.26584643958218723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5417688576161987e-06, "output_power": 2.0447574436638678e-06, "reference_active": true, "delta_db": 1.2262252880169258}], "active_band_count": 3, "active_band_mae_db": 0.5253154158302051}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013601220330575488, "log_rms_mae_db": 1.6371014784761688, "rms_envelope_nrmse": 0.2876386882888918, "envelope_correlation": 0.9478395243174716, "rms_gain_error_db": -0.4511361049236787, "reference_mean_power": 0.00024923569921471797, "output_mean_power": 0.00022464493855816923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.70738105242738e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001701759171968848, "output_power": 0.0001504103230027551, "reference_active": true, "delta_db": -0.5362045610212807}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6462817859538857e-05, "output_power": 1.8765134634901105e-05, "reference_active": true, "delta_db": -1.4928440290111922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.097841074264338e-06, "output_power": 2.148823055073457e-05, "reference_active": true, "delta_db": 8.411415211879433}], "active_band_count": 3, "active_band_mae_db": 3.4801546006373023}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011511226838452259, "log_rms_mae_db": 1.3726385209247962, "rms_envelope_nrmse": 0.12782073562124366, "envelope_correlation": 0.9906937351018583, "rms_gain_error_db": -0.4533248026612039, "reference_mean_power": 0.00024963838926192667, "output_mean_power": 0.00022489452949705786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6223780118670214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016948627965662844, "output_power": 0.00015833731746931121, "reference_active": true, "delta_db": -0.2955126382022698}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1767373976886565e-05, "output_power": 9.566911157735688e-06, "reference_active": true, "delta_db": -0.8990781496671272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8952968721807594e-07, "output_power": 1.4629341692133318e-06, "reference_active": true, "delta_db": 8.875475377964971}], "active_band_count": 3, "active_band_mae_db": 3.35668872194479}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.3800826555528964e-05, "log_rms_mae_db": 1.614995020928597, "rms_envelope_nrmse": 0.17742350219061598, "envelope_correlation": 0.9813860135554061, "rms_gain_error_db": 0.11844797077316409, "reference_mean_power": 4.104766861392576e-05, "output_mean_power": 4.218259493438726e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.890895346126465e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.549789437651056e-06, "output_power": 9.589411980575728e-06, "reference_active": true, "delta_db": 0.017981812259172433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.275656140671785e-06, "output_power": 2.8920282555501683e-06, "reference_active": true, "delta_db": -0.5409577400702518}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.939546621487433e-07, "output_power": 5.224955732447589e-07, "reference_active": true, "delta_db": -0.5567068029168987}], "active_band_count": 3, "active_band_mae_db": 0.371882118415441}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.772714769720092e-05, "log_rms_mae_db": 1.396396522259968, "rms_envelope_nrmse": 0.10249366499085764, "envelope_correlation": 0.9936697698387644, "rms_gain_error_db": 0.11828805449533006, "reference_mean_power": 4.122071638798201e-05, "output_mean_power": 4.235886753304331e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.778938094011469e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.936665760227185e-06, "output_power": 9.862945096948472e-06, "reference_active": true, "delta_db": -0.03234066108391716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7465237791634235e-06, "output_power": 1.5414726238890745e-06, "reference_active": true, "delta_db": -0.5423868680771561}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8043700755065915e-08, "output_power": 3.3820048147700354e-08, "reference_active": true, "delta_db": -1.5246223171766373}], "active_band_count": 3, "active_band_mae_db": 0.6997832821125702}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8865243730546271, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a306701c6077781fd83d038e726642ff753d411feac372f60d7e835e5193a6ae", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9658569337934226, "standard_deviation": 0.007538038217496211, "minimum": 0.9478395243174716, "maximum": 0.9714063196034303, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5845473432193975, "standard_deviation": 0.03929742615440336, "minimum": 1.5320426506038167, "maximum": 1.6371014784761688, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.9467482219520418, "standard_deviation": 0.8091575124794966, "minimum": 1.105717113138004, "maximum": 3.4801546006373023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23451447123189184, "standard_deviation": 0.023020666618044, "minimum": 0.21405202405227866, "maximum": 0.2876386882888918, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9902636573561074, "standard_deviation": 0.0031684551074811513, "minimum": 0.9852662793110083, "maximum": 0.9940492620228543, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3187857271375791, "standard_deviation": 0.03140664868285297, "minimum": 1.2906756558677177, "maximum": 1.3726385209247962, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 2.0642107879872436, "standard_deviation": 0.6485739171100114, "minimum": 1.4559835312025236, "maximum": 3.35668872194479, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12801880111786665, "standard_deviation": 0.025159748751246543, "minimum": 0.09695052426148783, "maximum": 0.17335310339909352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.983829187144087, "standard_deviation": 0.002701751939642993, "minimum": 0.9793866570004892, "maximum": 0.9874829791647332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6142890774910539, "standard_deviation": 0.01825054945877685, "minimum": 1.5871991900808553, "maximum": 1.6447448093562456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.2788477480276393, "standard_deviation": 0.10082170375892503, "minimum": 0.08110211556699914, "maximum": 0.371882118415441, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16523233772878207, "standard_deviation": 0.013653868050191845, "minimum": 0.14484205772311914, "maximum": 0.18482631398825253, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9939892274333758, "standard_deviation": 0.0010019881411763322, "minimum": 0.9920066575403823, "maximum": 0.9950625348223261, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4110247304836943, "standard_deviation": 0.023684117164416787, "minimum": 1.3881039679831697, "maximum": 1.4440734313937569, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4092789141777683, "standard_deviation": 0.15000632552376084, "minimum": 0.23279740433785043, "maximum": 0.6997832821125702, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10047844106364222, "standard_deviation": 0.008684432285542026, "minimum": 0.09016846673988799, "maximum": 0.11382763174282849, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9833861651991862, "standard_deviation": 0.0025465012404166042, "minimum": 0.9800371731709856, "maximum": 0.986273857282472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2835350286594986, "standard_deviation": 0.03740608156538765, "minimum": 1.2397089628129854, "maximum": 1.3492427696465439, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4860574030700135, "standard_deviation": 0.17160968630451257, "minimum": 0.29520907465542906, "maximum": 0.7538674254827802, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14939396707320918, "standard_deviation": 0.011339502616946382, "minimum": 0.13621916883475377, "maximum": 0.1642098174005739, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949797303875585, "standard_deviation": 0.0017652618510336341, "minimum": 0.9913530165088363, "maximum": 0.9966730049037891, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9558259587016782, "standard_deviation": 0.04039222788125575, "minimum": 0.9146012763006768, "maximum": 1.024332444220432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5714688197145718, "standard_deviation": 0.2352786780096938, "minimum": 0.3057144212530785, "maximum": 1.0060695220168563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08004603319730691, "standard_deviation": 0.012855167572034972, "minimum": 0.06621436503058292, "maximum": 0.1058694398581857, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.26255990694257686, "standard_deviation": 0.08364319063328339, "minimum": 0.15591721888952184, "maximum": 0.3958853793379378, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.410252599523192, "standard_deviation": 0.004106608243411257, "minimum": 5.404525554419504, "maximum": 5.41628553817196, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5624906837796909, "standard_deviation": 0.040752933294604166, "minimum": 0.4901786949326572, "maximum": 0.6175371278802309, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09127433139528629, "standard_deviation": 0.0526843519062853, "minimum": -0.1697000972502214, "maximum": -0.01851015341455947, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9897617763414299, "standard_deviation": 0.0020755745690345796, "minimum": 0.9851182763666029, "maximum": 0.9920220893245544, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6939271130353324, "standard_deviation": 0.03075043585647957, "minimum": 0.6562383452015412, "maximum": 0.7473922610957785, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4998316960330121, "standard_deviation": 0.18525035646610008, "minimum": 0.3313167577902209, "maximum": 0.8167296035478201, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10225382398722345, "standard_deviation": 0.009891459557574834, "minimum": 0.0905500596115043, "maximum": 0.12386061687366662, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971789191611752, "standard_deviation": 0.0006400365877147051, "minimum": 0.9961076624164253, "maximum": 0.9979830698459886, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4262699188741386, "standard_deviation": 0.019131740229447896, "minimum": 0.38819364044013493, "maximum": 0.4523255707131807, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4274565741209573, "standard_deviation": 0.24714788518736716, "minimum": 0.17606167103827794, "maximum": 0.9373624368799846, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05266541662693091, "standard_deviation": 0.006013417045559633, "minimum": 0.04396287148637638, "maximum": 0.06289791096814393, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7021-85628-0025", "language": "en", "style": "normal", "speaker": "librispeech_7021", "source_utterance_id": "7021-85628-0025", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 44400, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7021-85628-0025.wav", "source_file_sha256": "a56ee995082f1c52cf9089bbaac38edef5b9dbb74fd1d76112919622b9040740", "decoded_pcm_sha256": "f8114fb1a86ed60fb5b4e5ba870142361984ad8304c63676d1f9c029942602ee", "original_source_sha256": "477a753670424f1fd835f77fdd15e54570114ec1a4f99c206e555e94137688c6", "corpus_entry_id": "librispeech_7021-85628-0025"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.993018388748169, "predicted_mos": 3.993018388748169, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "9344b5161e42054fd8435a7b6f4d9974066d7a4113abea38945b3ea07bec26b2", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2686831449973397, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7021-85628-0025", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9344b5161e42054fd8435a7b6f4d9974066d7a4113abea38945b3ea07bec26b2", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "58e5faaa9c6edaa7abbb651251a2000da6e90e3c1943e767c58f215528a210cc"}, "rows": [{"case": "librispeech_7021-85628-0025", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.9801688194274902, "predicted_mos": 2.9801688194274902, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "5dcbbeb9927cf567c0a7f2235563f0b7f3c335ec92a78568a86fbf4d9c22414f", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4205825610006286, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7021-85628-0025", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5dcbbeb9927cf567c0a7f2235563f0b7f3c335ec92a78568a86fbf4d9c22414f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.899457840026121, "全帯域_10ms_rms_envelope_nrmse": 0.055688698636609295, "全帯域_10ms_envelope_correlation": 0.9974925863021693, "全帯域_10ms_modulation_mae_db": 0.4069678712181599, "全帯域_40ms_log_rms_mae_db": 0.5695138786154408, "全帯域_40ms_rms_envelope_nrmse": 0.03316671324216737, "全帯域_40ms_envelope_correlation": 0.9991604454125889, "全帯域_40ms_modulation_mae_db": 0.3906683314724826, "300–1000 Hz_10ms_log_rms_mae_db": 1.1868829756181882, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12158037082060723, "300–1000 Hz_10ms_envelope_correlation": 0.9897513187196038, "300–1000 Hz_10ms_modulation_mae_db": 0.5155250801938457, "300–1000 Hz_40ms_log_rms_mae_db": 0.7583274118494441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07177334832818288, "300–1000 Hz_40ms_envelope_correlation": 0.9964102792721422, "300–1000 Hz_40ms_modulation_mae_db": 0.31284782583441834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4241633577062305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17201617876543493, "1000–3000 Hz_10ms_envelope_correlation": 0.9823368666584857, "1000–3000 Hz_10ms_modulation_mae_db": 1.743462898555455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0744382057680821, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11421204392370819, "1000–3000 Hz_40ms_envelope_correlation": 0.9934675274067543, "1000–3000 Hz_40ms_modulation_mae_db": 1.5700591845602059, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6989452385595563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20318536710253235, "3000–6000 Hz_10ms_envelope_correlation": 0.9697318658994543, "3000–6000 Hz_10ms_modulation_mae_db": 1.3139135862536555, "3000–6000 Hz_40ms_log_rms_mae_db": 1.203259046796022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12049928258877989, "3000–6000 Hz_40ms_envelope_correlation": 0.9900766563265382, "3000–6000 Hz_40ms_modulation_mae_db": 1.9609809294929976, "mr_stft_db": 5.557384799373479, "band_mae_db": 0.2818919341583493, "rms_gain_error_db": -0.11834622192308406, "output_peak": 0.4034554721393152}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.049937409791038, "全帯域_10ms_rms_envelope_nrmse": 0.05846796475674196, "全帯域_10ms_envelope_correlation": 0.9971212895985485, "全帯域_10ms_modulation_mae_db": 0.5904060736628257, "全帯域_40ms_log_rms_mae_db": 0.6474086884182881, "全帯域_40ms_rms_envelope_nrmse": 0.03141058575653575, "全帯域_40ms_envelope_correlation": 0.9991447487545709, "全帯域_40ms_modulation_mae_db": 0.6682254514427436, "300–1000 Hz_10ms_log_rms_mae_db": 1.2345608741263436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09504373830833476, "300–1000 Hz_10ms_envelope_correlation": 0.9939530327028816, "300–1000 Hz_10ms_modulation_mae_db": 0.3626969756296587, "300–1000 Hz_40ms_log_rms_mae_db": 0.7926148032820435, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05484906658891295, "300–1000 Hz_40ms_envelope_correlation": 0.998154755763669, "300–1000 Hz_40ms_modulation_mae_db": 0.2795266739484064, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4158040494198636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19104140438725092, "1000–3000 Hz_10ms_envelope_correlation": 0.9763111294461642, "1000–3000 Hz_10ms_modulation_mae_db": 1.313004635215531, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0845836564718747, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1259086555254741, "1000–3000 Hz_40ms_envelope_correlation": 0.9891307866395639, "1000–3000 Hz_40ms_modulation_mae_db": 1.6589457851765685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.67615477478914, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1976359336284693, "3000–6000 Hz_10ms_envelope_correlation": 0.9707079823231255, "3000–6000 Hz_10ms_modulation_mae_db": 0.9415397069099293, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1755426029981486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11346769915216594, "3000–6000 Hz_40ms_envelope_correlation": 0.9902165220209477, "3000–6000 Hz_40ms_modulation_mae_db": 1.5762570030610776, "mr_stft_db": 5.52011598703951, "band_mae_db": 0.2697167534658685, "rms_gain_error_db": -0.05980803185306692, "output_peak": 0.4040101863736414}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9198512999215767, "全帯域_10ms_rms_envelope_nrmse": 0.054751365573791726, "全帯域_10ms_envelope_correlation": 0.9975541481342255, "全帯域_10ms_modulation_mae_db": 0.5340266629066165, "全帯域_40ms_log_rms_mae_db": 0.5306795683864816, "全帯域_40ms_rms_envelope_nrmse": 0.028760562807624606, "全帯域_40ms_envelope_correlation": 0.9993794048933997, "全帯域_40ms_modulation_mae_db": 0.5947729891165485, "300–1000 Hz_10ms_log_rms_mae_db": 1.1204696387435933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12395624816225591, "300–1000 Hz_10ms_envelope_correlation": 0.990440635134569, "300–1000 Hz_10ms_modulation_mae_db": 1.20181017393095, "300–1000 Hz_40ms_log_rms_mae_db": 0.783223664946415, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07719869816476543, "300–1000 Hz_40ms_envelope_correlation": 0.997182464515038, "300–1000 Hz_40ms_modulation_mae_db": 1.1473639170195498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4103164082861321, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16974258963465497, "1000–3000 Hz_10ms_envelope_correlation": 0.9815927297537846, "1000–3000 Hz_10ms_modulation_mae_db": 2.141216568352846, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0524956297623997, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10517205600922246, "1000–3000 Hz_40ms_envelope_correlation": 0.9932590080429948, "1000–3000 Hz_40ms_modulation_mae_db": 1.8162145360907516, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7592922779550935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20675126938021934, "3000–6000 Hz_10ms_envelope_correlation": 0.9677347351981818, "3000–6000 Hz_10ms_modulation_mae_db": 1.06270048746958, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2073678808337054, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10860975936465268, "3000–6000 Hz_40ms_envelope_correlation": 0.9910492584245011, "3000–6000 Hz_40ms_modulation_mae_db": 1.5438389165063817, "mr_stft_db": 5.533872359043155, "band_mae_db": 0.3873408062458276, "rms_gain_error_db": -0.10611242393095136, "output_peak": 0.3934509479133978}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0065001710820118, "全帯域_10ms_rms_envelope_nrmse": 0.05865731744950916, "全帯域_10ms_envelope_correlation": 0.9971840907553506, "全帯域_10ms_modulation_mae_db": 0.377078376744747, "全帯域_40ms_log_rms_mae_db": 0.6222724579042244, "全帯域_40ms_rms_envelope_nrmse": 0.032607730827830125, "全帯域_40ms_envelope_correlation": 0.999176757121142, "全帯域_40ms_modulation_mae_db": 0.5850438460501288, "300–1000 Hz_10ms_log_rms_mae_db": 1.1824551001319834, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.123022430329416, "300–1000 Hz_10ms_envelope_correlation": 0.9898114109841654, "300–1000 Hz_10ms_modulation_mae_db": 0.6964256761851928, "300–1000 Hz_40ms_log_rms_mae_db": 0.8302138704351948, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0782093106737209, "300–1000 Hz_40ms_envelope_correlation": 0.9960902504799964, "300–1000 Hz_40ms_modulation_mae_db": 0.9419630748071345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.411718649191756, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16013688229852402, "1000–3000 Hz_10ms_envelope_correlation": 0.9855765755475978, "1000–3000 Hz_10ms_modulation_mae_db": 0.7938163782261934, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1145011746565903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11135343519482295, "1000–3000 Hz_40ms_envelope_correlation": 0.9945691376434658, "1000–3000 Hz_40ms_modulation_mae_db": 0.8843038660647521, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7943964482382706, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2243060752426139, "3000–6000 Hz_10ms_envelope_correlation": 0.9617267784113901, "3000–6000 Hz_10ms_modulation_mae_db": 1.1691141044301974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.306784251863298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13861250496456368, "3000–6000 Hz_40ms_envelope_correlation": 0.9842533305855001, "3000–6000 Hz_40ms_modulation_mae_db": 1.6520100791959573, "mr_stft_db": 5.526957512042858, "band_mae_db": 0.30248589952273336, "rms_gain_error_db": -0.10894514062240973, "output_peak": 0.41398462973249267}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0123303759244382, "全帯域_10ms_rms_envelope_nrmse": 0.057765674657970445, "全帯域_10ms_envelope_correlation": 0.9974137329676028, "全帯域_10ms_modulation_mae_db": 0.20335108048457826, "全帯域_40ms_log_rms_mae_db": 0.6126915474600516, "全帯域_40ms_rms_envelope_nrmse": 0.036847685291044645, "全帯域_40ms_envelope_correlation": 0.9990662977152627, "全帯域_40ms_modulation_mae_db": 0.33116871033646256, "300–1000 Hz_10ms_log_rms_mae_db": 1.1992784327252706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13861516135884633, "300–1000 Hz_10ms_envelope_correlation": 0.9875907154810302, "300–1000 Hz_10ms_modulation_mae_db": 1.3561751795542758, "300–1000 Hz_40ms_log_rms_mae_db": 0.8007045766820716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08521153780703346, "300–1000 Hz_40ms_envelope_correlation": 0.9961385162319638, "300–1000 Hz_40ms_modulation_mae_db": 1.4477028371426799, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4424127615644664, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1842082717661323, "1000–3000 Hz_10ms_envelope_correlation": 0.9787594370465937, "1000–3000 Hz_10ms_modulation_mae_db": 1.7973636386561072, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1354884288104286, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11449286135667229, "1000–3000 Hz_40ms_envelope_correlation": 0.9927322639372468, "1000–3000 Hz_40ms_modulation_mae_db": 1.6477637726394727, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7222658436032856, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2045764878123108, "3000–6000 Hz_10ms_envelope_correlation": 0.9683890459394124, "3000–6000 Hz_10ms_modulation_mae_db": 1.054953408446461, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2081483792489995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11799069626525946, "3000–6000 Hz_40ms_envelope_correlation": 0.9891501143887296, "3000–6000 Hz_40ms_modulation_mae_db": 1.5641716507362562, "mr_stft_db": 5.5219292840079435, "band_mae_db": 0.3962460655407943, "rms_gain_error_db": -0.16005121378714748, "output_peak": 0.4019032400718028}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9740315309152534, "全帯域_10ms_rms_envelope_nrmse": 0.048109444098261324, "全帯域_10ms_envelope_correlation": 0.9981939327333365, "全帯域_10ms_modulation_mae_db": 0.511706513497085, "全帯域_40ms_log_rms_mae_db": 0.6146395649424551, "全帯域_40ms_rms_envelope_nrmse": 0.02755313499093859, "全帯域_40ms_envelope_correlation": 0.9995021372704236, "全帯域_40ms_modulation_mae_db": 0.6064882314110194, "300–1000 Hz_10ms_log_rms_mae_db": 1.1401106839074968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1246119829010383, "300–1000 Hz_10ms_envelope_correlation": 0.9901723597951705, "300–1000 Hz_10ms_modulation_mae_db": 1.0693188584654458, "300–1000 Hz_40ms_log_rms_mae_db": 0.78619281905876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0826338401804348, "300–1000 Hz_40ms_envelope_correlation": 0.996312619141445, "300–1000 Hz_40ms_modulation_mae_db": 1.2455818943510244, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4367565663809394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1733564216050006, "1000–3000 Hz_10ms_envelope_correlation": 0.9814455513247987, "1000–3000 Hz_10ms_modulation_mae_db": 1.4191873517554943, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0439676353578407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10729331947241974, "1000–3000 Hz_40ms_envelope_correlation": 0.9939039158085428, "1000–3000 Hz_40ms_modulation_mae_db": 1.3931195427749186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.737173264525771, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20387912553842005, "3000–6000 Hz_10ms_envelope_correlation": 0.9695565036046098, "3000–6000 Hz_10ms_modulation_mae_db": 1.278793562095043, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1539894515740423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11943501483411685, "3000–6000 Hz_40ms_envelope_correlation": 0.9905004052631544, "3000–6000 Hz_40ms_modulation_mae_db": 1.9205077653177185, "mr_stft_db": 5.510030118855608, "band_mae_db": 0.29038575663829724, "rms_gain_error_db": -0.12849180258213722, "output_peak": 0.40810324426387345}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0194851888675696, "全帯域_10ms_rms_envelope_nrmse": 0.05598402941171665, "全帯域_10ms_envelope_correlation": 0.9973749678242533, "全帯域_10ms_modulation_mae_db": 0.3022944345034004, "全帯域_40ms_log_rms_mae_db": 0.5920052341182511, "全帯域_40ms_rms_envelope_nrmse": 0.02975706892857233, "全帯域_40ms_envelope_correlation": 0.9992506440904653, "全帯域_40ms_modulation_mae_db": 0.436224503534373, "300–1000 Hz_10ms_log_rms_mae_db": 1.2405696601600367, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11783899681356125, "300–1000 Hz_10ms_envelope_correlation": 0.9908197146319715, "300–1000 Hz_10ms_modulation_mae_db": 0.8260553744011255, "300–1000 Hz_40ms_log_rms_mae_db": 0.8296484128369428, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07723791109016322, "300–1000 Hz_40ms_envelope_correlation": 0.9963333635633655, "300–1000 Hz_40ms_modulation_mae_db": 1.1494054751605767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4712168287728582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17793075355248847, "1000–3000 Hz_10ms_envelope_correlation": 0.9798852137213732, "1000–3000 Hz_10ms_modulation_mae_db": 1.0370317576513972, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0796970257819183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11207773134241386, "1000–3000 Hz_40ms_envelope_correlation": 0.9925390056197932, "1000–3000 Hz_40ms_modulation_mae_db": 1.1806625311981318, "3000–6000 Hz_10ms_log_rms_mae_db": 1.78281758652631, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2120831269741959, "3000–6000 Hz_10ms_envelope_correlation": 0.9658714874885154, "3000–6000 Hz_10ms_modulation_mae_db": 1.2096019702023575, "3000–6000 Hz_40ms_log_rms_mae_db": 1.265552014045784, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11579143803271631, "3000–6000 Hz_40ms_envelope_correlation": 0.9891466843649056, "3000–6000 Hz_40ms_modulation_mae_db": 1.7656845222970112, "mr_stft_db": 5.517131128806749, "band_mae_db": 0.2550903186961385, "rms_gain_error_db": -0.06704538125101665, "output_peak": 0.4129964826572811}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9885537657443607, "全帯域_10ms_rms_envelope_nrmse": 0.05418061398076397, "全帯域_10ms_envelope_correlation": 0.9975133317300194, "全帯域_10ms_modulation_mae_db": 0.1693304341297626, "全帯域_40ms_log_rms_mae_db": 0.592343584229158, "全帯域_40ms_rms_envelope_nrmse": 0.028352164659518373, "全帯域_40ms_envelope_correlation": 0.9992822476820298, "全帯域_40ms_modulation_mae_db": 0.07710533847621363, "300–1000 Hz_10ms_log_rms_mae_db": 1.1827573475410684, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1119372257225289, "300–1000 Hz_10ms_envelope_correlation": 0.9916368205438408, "300–1000 Hz_10ms_modulation_mae_db": 0.8076970287283238, "300–1000 Hz_40ms_log_rms_mae_db": 0.7720236603934905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05884453037725517, "300–1000 Hz_40ms_envelope_correlation": 0.9980407995339824, "300–1000 Hz_40ms_modulation_mae_db": 1.4317508386006184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4314555179558457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16087456792182375, "1000–3000 Hz_10ms_envelope_correlation": 0.9830693722997357, "1000–3000 Hz_10ms_modulation_mae_db": 1.0590794134299752, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0637872447423677, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10771510863209764, "1000–3000 Hz_40ms_envelope_correlation": 0.9921046136796526, "1000–3000 Hz_40ms_modulation_mae_db": 0.5404761615496523, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7033659062530178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1971091104215029, "3000–6000 Hz_10ms_envelope_correlation": 0.971072079381983, "3000–6000 Hz_10ms_modulation_mae_db": 1.039557561408311, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2014247629358017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11983341693342803, "3000–6000 Hz_40ms_envelope_correlation": 0.9892604159117975, "3000–6000 Hz_40ms_modulation_mae_db": 1.5008939526519278, "mr_stft_db": 5.530368062433138, "band_mae_db": 0.173221267924776, "rms_gain_error_db": -0.031065245698977402, "output_peak": 0.4173552164203079}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44400, "output_samples": 44400, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.706681866558752, "spectral_convergence": 0.1762024768009932, "spectral_convergence_status": "ok", "amplitude_floor": 1.585777823535548e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.47054657855901, "spectral_convergence": 0.16887670418664757, "spectral_convergence_status": "ok", "amplitude_floor": 7.694479165248034e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.494925953002674, "spectral_convergence": 0.17062610383400675, "spectral_convergence_status": "ok", "amplitude_floor": 6.411814451459315e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00186841781908955, "output_power": 0.0018268591289262817, "delta_db": -0.0976894077065373, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005255584463676, "output_power": 0.0005078957376648721, "delta_db": -0.14846451942278616, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025377475865110864, "output_power": 0.0002598000907756701, "delta_db": 0.10190875064160207, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00014708318852606447, "output_power": 0.00012605327185559258, "delta_db": -0.6700891333763035, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.522695850195022e-06, "output_power": 8.31326580586995e-06, "delta_db": -0.5898824669296983, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.071083630879648e-06, "output_power": 5.955722892928364e-06, "delta_db": -0.0833173268731684, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2818919341583493, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11834622192308406, "pitch": null, "output_rms_dbfs": -25.63062143497438, "output_peak": 0.4034554721393152, "reference_rms_dbfs": -25.512275213051296, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.557384799373479}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44400, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 0.00014622417526559514, "log_rms_mae_db": 0.899457840026121, "rms_envelope_nrmse": 0.055688698636609295, "envelope_correlation": 0.9974925863021693, "rms_gain_error_db": -0.11834537444483925, "reference_mean_power": 0.0028155003489198547, "output_mean_power": 0.002739813795211109}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.019039396656961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008418483111337408, "output_power": 0.0008232034654873151, "reference_active": true, "delta_db": -0.09726654943836424}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001428627689371398, "output_power": 0.0001385317617918381, "reference_active": true, "delta_db": -0.13369705837210175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9192224262112575e-05, "output_power": 1.5280279884435492e-05, "reference_active": true, "delta_db": -0.9899400058440138}], "active_band_count": 3, "active_band_mae_db": 0.4069678712181599}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 0.00013893685565863478, "log_rms_mae_db": 0.5695138786154408, "rms_envelope_nrmse": 0.03316671324216737, "envelope_correlation": 0.9991604454125889, "rms_gain_error_db": -0.11834466699825345, "reference_mean_power": 0.0028463248833832306, "output_mean_power": 0.002769810152736431}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.8422021834325858e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007582285194051116, "output_power": 0.0007412596729059769, "reference_active": true, "delta_db": -0.09829741965075042}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661169958011628e-05, "output_power": 6.596757814511792e-05, "reference_active": true, "delta_db": -0.04219974504323849}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7101796636835213e-06, "output_power": 1.3486242196654485e-06, "reference_active": true, "delta_db": -1.0315078297234588}], "active_band_count": 3, "active_band_mae_db": 0.3906683314724826}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 8.753945313428826e-05, "log_rms_mae_db": 1.1868829756181882, "rms_envelope_nrmse": 0.12158037082060723, "envelope_correlation": 0.9897513187196038, "rms_gain_error_db": -0.1725871927145746, "reference_mean_power": 0.0005201241069420424, "output_mean_power": 0.0004998598607198908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 5.057137989146479e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002701505367125327, "output_power": 0.0002576468981954571, "reference_active": true, "delta_db": -0.20580916327896104}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.4045222297363644e-05, "output_power": 5.123243963800989e-05, "reference_active": true, "delta_db": -0.23212270468733276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.5635511670477814e-06, "output_power": 5.859511513657722e-06, "reference_active": true, "delta_db": -1.1086433726152432}], "active_band_count": 3, "active_band_mae_db": 0.5155250801938457}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 7.440703161891069e-05, "log_rms_mae_db": 0.7583274118494441, "rms_envelope_nrmse": 0.07177334832818288, "envelope_correlation": 0.9964102792721422, "rms_gain_error_db": -0.17258738056962802, "reference_mean_power": 0.0005258188868038473, "output_mean_power": 0.000505332747789611}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 4.6537380366659205e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025071266719842557, "output_power": 0.00023920773899939438, "reference_active": true, "delta_db": -0.20401051026701686}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3806665370442115e-05, "output_power": 2.2456127565667746e-05, "reference_active": true, "delta_db": -0.2536370077715295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0797322752712393e-06, "output_power": 9.665548124626214e-07, "reference_active": true, "delta_db": -0.48089595946470876}], "active_band_count": 3, "active_band_mae_db": 0.31284782583441834}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 5.3294712554339956e-05, "log_rms_mae_db": 1.4241633577062305, "rms_envelope_nrmse": 0.17201617876543493, "envelope_correlation": 0.9823368666584857, "rms_gain_error_db": -0.5683210526947059, "reference_mean_power": 0.00010733282194912887, "output_mean_power": 9.416737029209266e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 1.1016693525679884e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607161402154175e-05, "output_power": 5.806626946533129e-05, "reference_active": true, "delta_db": -0.5609099086419901}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0221078826476055e-05, "output_power": 7.84104105529766e-06, "reference_active": true, "delta_db": -1.1512300983760684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.376144897878774e-06, "output_power": 1.5017450482271955e-06, "reference_active": true, "delta_db": -3.518248688648307}], "active_band_count": 3, "active_band_mae_db": 1.743462898555455}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 4.1357257962539876e-05, "log_rms_mae_db": 1.0744382057680821, "rms_envelope_nrmse": 0.11421204392370819, "envelope_correlation": 0.9934675274067543, "rms_gain_error_db": -0.56832145853255, "reference_mean_power": 0.00010850799286328448, "output_mean_power": 9.519838576136563e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.0153781665834753e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.269855261867839e-05, "output_power": 5.390791746815787e-05, "reference_active": true, "delta_db": -0.6560496055954057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.20737764206477e-06, "output_power": 3.6928820214224057e-06, "reference_active": true, "delta_db": -0.5664606156212608}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.246816673622798e-07, "output_power": 1.0064701634532107e-07, "reference_active": true, "delta_db": -3.487667332463951}], "active_band_count": 3, "active_band_mae_db": 1.5700591845602059}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 1.102648736164134e-05, "log_rms_mae_db": 1.6989452385595563, "rms_envelope_nrmse": 0.20318536710253235, "envelope_correlation": 0.9697318658994543, "rms_gain_error_db": -0.4502510877755921, "reference_mean_power": 5.352135557413054e-06, "output_mean_power": 4.8250519742064594e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.71214230899019e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.798500894794614e-06, "output_power": 1.5882920711376425e-06, "reference_active": true, "delta_db": -0.5398029034892148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2650926706890884e-07, "output_power": 2.7889509672860227e-07, "reference_active": true, "delta_db": -1.8448759285396923}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0854303858213565e-07, "output_power": 7.583955612532673e-08, "reference_active": true, "delta_db": -1.5570619267320593}], "active_band_count": 3, "active_band_mae_db": 1.3139135862536555}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 9.954651742580568e-06, "log_rms_mae_db": 1.203259046796022, "rms_envelope_nrmse": 0.12049928258877989, "envelope_correlation": 0.9900766563265382, "rms_gain_error_db": -0.45025318315405577, "reference_mean_power": 5.410729753830906e-06, "output_mean_power": 4.877873402956944e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 2.394714459074953e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.814627629314412e-06, "output_power": 1.5455001542131842e-06, "reference_active": true, "delta_db": -0.6971846637598924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0097168111890166e-07, "output_power": 1.354352445412002e-07, "reference_active": true, "delta_db": -1.7140316913980822}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.411981302222167e-09, "output_power": 4.2316370643726225e-09, "reference_active": true, "delta_db": -3.4717264333210185}], "active_band_count": 3, "active_band_mae_db": 1.9609809294929976}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8881367057512989, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dda517451f345982c698c998c09232f129f2f227d632a76d7708b2917ac92179", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9811221094748167, "standard_deviation": 0.002820676072441444, "minimum": 0.9763111294461642, "maximum": 0.9855765755475978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4304805174097615, "standard_deviation": 0.020199254392003086, "minimum": 1.4103164082861321, "maximum": 1.4712168287728582, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.413020330230375, "standard_deviation": 0.4547021184696489, "minimum": 0.7938163782261934, "maximum": 2.141216568352846, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17366338374141374, "standard_deviation": 0.010664365883573175, "minimum": 0.16013688229852402, "maximum": 0.19104140438725092, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927132823472518, "standard_deviation": 0.0016459471778289194, "minimum": 0.9891307866395639, "maximum": 0.9945691376434658, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.081119875168938, "standard_deviation": 0.030773305206684066, "minimum": 1.0439676353578407, "maximum": 1.1354884288104286, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3364431725068067, "standard_deviation": 0.4393434562363639, "minimum": 0.5404761615496523, "maximum": 1.8162145360907516, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1122781514321039, "standard_deviation": 0.006453097725792112, "minimum": 0.10517205600922246, "maximum": 0.1259086555254741, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9680988097808341, "standard_deviation": 0.003075720420684707, "minimum": 0.9617267784113901, "maximum": 0.971072079381983, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7343014175563056, "standard_deviation": 0.04197103479271163, "minimum": 1.67615477478914, "maximum": 1.7943964482382706, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1337717984019418, "standard_deviation": 0.12960083691873134, "minimum": 0.9415397069099293, "maximum": 1.3139135862536555, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2061908120125331, "standard_deviation": 0.008749221467432395, "minimum": 0.1971091104215029, "maximum": 0.2243060752426139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9892066734107593, "standard_deviation": 0.0021176407741221896, "minimum": 0.9842533305855001, "maximum": 0.9910492584245011, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2152585487869754, "standard_deviation": 0.04882303967999063, "minimum": 1.1539894515740423, "maximum": 1.306784251863298, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.685543102407416, "standard_deviation": 0.17706793120648096, "minimum": 1.5008939526519278, "maximum": 1.9609809294929976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11927997651696035, "standard_deviation": 0.008758929673805668, "minimum": 0.10860975936465268, "maximum": 0.13861250496456368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990522000999154, "standard_deviation": 0.0018096419569305726, "minimum": 0.9875907154810302, "maximum": 0.9939530327028816, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1858855891192475, "standard_deviation": 0.04129361191785649, "minimum": 1.1204696387435933, "maximum": 1.2405696601600367, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8544630433861022, "standard_deviation": 0.33873942607747026, "minimum": 0.3626969756296587, "maximum": 1.3561751795542758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11957576930207359, "standard_deviation": 0.012456396309912026, "minimum": 0.09504373830833476, "maximum": 0.13861516135884633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968328810627003, "standard_deviation": 0.0008502985627948934, "minimum": 0.9960902504799964, "maximum": 0.998154755763669, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7941186524355452, "standard_deviation": 0.02552753628147993, "minimum": 0.7583274118494441, "maximum": 0.8302138704351948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.994517817108051, "standard_deviation": 0.46068008390737925, "minimum": 0.2795266739484064, "maximum": 1.4477028371426799, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0732447804013086, "standard_deviation": 0.010925183524553746, "minimum": 0.05484906658891295, "maximum": 0.08521153780703346, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2945473502740981, "standard_deviation": 0.07180050465195771, "minimum": 0.173221267924776, "maximum": 0.3962460655407943, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.527223656450305, "standard_deviation": 0.014342166123002488, "minimum": 5.510030118855608, "maximum": 5.557384799373479, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40690742744651404, "standard_deviation": 0.007784401911530453, "minimum": 0.3934509479133978, "maximum": 0.4173552164203079, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09748318270609885, "standard_deviation": 0.04187501610104052, "minimum": -0.16005121378714748, "maximum": -0.031065245698977402, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9974810100056882, "standard_deviation": 0.0003269158175996879, "minimum": 0.9971212895985485, "maximum": 0.9981939327333365, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9837684477840462, "standard_deviation": 0.05112717937985202, "minimum": 0.899457840026121, "maximum": 1.049937409791038, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3868951808933969, "standard_deviation": 0.1546062509737203, "minimum": 0.1693304341297626, "maximum": 0.5904060736628257, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05545063857067057, "standard_deviation": 0.0034034163231295433, "minimum": 0.048109444098261324, "maximum": 0.05865731744950916, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992453353674853, "standard_deviation": 0.00014111494026459995, "minimum": 0.9990662977152627, "maximum": 0.9995021372704236, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5976943155092939, "standard_deviation": 0.03572468480242028, "minimum": 0.5306795683864816, "maximum": 0.6474086884182881, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4612121752299965, "standard_deviation": 0.19536873019401632, "minimum": 0.07710533847621363, "maximum": 0.6682254514427436, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031056955813028975, "standard_deviation": 0.003095373595405067, "minimum": 0.02755313499093859, "maximum": 0.036847685291044645, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-85628-0025", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.0384740829467773, "predicted_mos": 3.0384740829467773, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "5d1d7d84a3db8156cf0ec14289864c031ccd5e19623820b20b2bfb6eb8111796", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.817105000372976, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8951912292139265, "全帯域_10ms_rms_envelope_nrmse": 0.05637181682833043, "全帯域_10ms_envelope_correlation": 0.9974296561732496, "全帯域_10ms_modulation_mae_db": 0.4120450776173204, "全帯域_40ms_log_rms_mae_db": 0.5780607515619969, "全帯域_40ms_rms_envelope_nrmse": 0.03301111405255345, "全帯域_40ms_envelope_correlation": 0.9991732738932101, "全帯域_40ms_modulation_mae_db": 0.3372966167324137, "300–1000 Hz_10ms_log_rms_mae_db": 1.1981758389964334, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12545774191155398, "300–1000 Hz_10ms_envelope_correlation": 0.9890418768234684, "300–1000 Hz_10ms_modulation_mae_db": 0.4768006625483034, "300–1000 Hz_40ms_log_rms_mae_db": 0.7818838269073859, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0752319895853338, "300–1000 Hz_40ms_envelope_correlation": 0.9959829477458221, "300–1000 Hz_40ms_modulation_mae_db": 0.25878381869755546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4364109066680366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17625877796108574, "1000–3000 Hz_10ms_envelope_correlation": 0.9815481939542431, "1000–3000 Hz_10ms_modulation_mae_db": 1.7614574738131787, "1000–3000 Hz_40ms_log_rms_mae_db": 1.107793354455716, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1158808168482445, "1000–3000 Hz_40ms_envelope_correlation": 0.9935028461221366, "1000–3000 Hz_40ms_modulation_mae_db": 1.5183289443225239, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6020812598861762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19927461652147213, "3000–6000 Hz_10ms_envelope_correlation": 0.9718493107907614, "3000–6000 Hz_10ms_modulation_mae_db": 1.5031458583842572, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1546700102616905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12842717311149224, "3000–6000 Hz_40ms_envelope_correlation": 0.9896438944138446, "3000–6000 Hz_40ms_modulation_mae_db": 1.9826979773615667, "mr_stft_db": 5.466854423347839, "band_mae_db": 0.35910923370582876, "rms_gain_error_db": -0.11952433444721536, "output_peak": 0.40427436728401395}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0830382803810554, "全帯域_10ms_rms_envelope_nrmse": 0.055937526423638195, "全帯域_10ms_envelope_correlation": 0.9973660417608466, "全帯域_10ms_modulation_mae_db": 0.6060249097112491, "全帯域_40ms_log_rms_mae_db": 0.6658677974082539, "全帯域_40ms_rms_envelope_nrmse": 0.030306391841458814, "全帯域_40ms_envelope_correlation": 0.9992037411594523, "全帯域_40ms_modulation_mae_db": 0.6603358571520986, "300–1000 Hz_10ms_log_rms_mae_db": 1.2302643153416117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0928330657916038, "300–1000 Hz_10ms_envelope_correlation": 0.9941668074571945, "300–1000 Hz_10ms_modulation_mae_db": 0.335534922865685, "300–1000 Hz_40ms_log_rms_mae_db": 0.7842203291624992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05271691652789745, "300–1000 Hz_40ms_envelope_correlation": 0.9982266062779267, "300–1000 Hz_40ms_modulation_mae_db": 0.3656015810207947, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4336124852018328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19456695653392633, "1000–3000 Hz_10ms_envelope_correlation": 0.9753352839386816, "1000–3000 Hz_10ms_modulation_mae_db": 1.470033724355547, "1000–3000 Hz_40ms_log_rms_mae_db": 1.137466333009407, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1293910160100796, "1000–3000 Hz_40ms_envelope_correlation": 0.9885313631009007, "1000–3000 Hz_40ms_modulation_mae_db": 1.6976236137291354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6147009261360274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.204406033694618, "3000–6000 Hz_10ms_envelope_correlation": 0.968928828322962, "3000–6000 Hz_10ms_modulation_mae_db": 0.9529948015399747, "3000–6000 Hz_40ms_log_rms_mae_db": 1.122279367700886, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12466330544015308, "3000–6000 Hz_40ms_envelope_correlation": 0.9886760802958099, "3000–6000 Hz_40ms_modulation_mae_db": 1.4507338082535286, "mr_stft_db": 5.451499549495644, "band_mae_db": 0.2990615374294506, "rms_gain_error_db": -0.05739870748707133, "output_peak": 0.4029915507463027}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9412721418722937, "全帯域_10ms_rms_envelope_nrmse": 0.05446360525159286, "全帯域_10ms_envelope_correlation": 0.9975909262653676, "全帯域_10ms_modulation_mae_db": 0.5131737557302621, "全帯域_40ms_log_rms_mae_db": 0.5194796273473753, "全帯域_40ms_rms_envelope_nrmse": 0.028417864511287656, "全帯域_40ms_envelope_correlation": 0.9994086221228494, "全帯域_40ms_modulation_mae_db": 0.6234795140009304, "300–1000 Hz_10ms_log_rms_mae_db": 1.1304388219635226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12596740753762473, "300–1000 Hz_10ms_envelope_correlation": 0.9900842441467848, "300–1000 Hz_10ms_modulation_mae_db": 1.2137853978633553, "300–1000 Hz_40ms_log_rms_mae_db": 0.7903426158526532, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07939500225233868, "300–1000 Hz_40ms_envelope_correlation": 0.9969303445756057, "300–1000 Hz_40ms_modulation_mae_db": 1.182455662743803, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4247767298539682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16056493031830785, "1000–3000 Hz_10ms_envelope_correlation": 0.9834264676460943, "1000–3000 Hz_10ms_modulation_mae_db": 2.016062331537226, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0706083748770419, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09869841413416243, "1000–3000 Hz_40ms_envelope_correlation": 0.9939053627427519, "1000–3000 Hz_40ms_modulation_mae_db": 1.7147062410488096, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7468618253536952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.216312074352137, "3000–6000 Hz_10ms_envelope_correlation": 0.9649032102016856, "3000–6000 Hz_10ms_modulation_mae_db": 1.1685729883632272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.207303875553403, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12707178196727006, "3000–6000 Hz_40ms_envelope_correlation": 0.9880622216716365, "3000–6000 Hz_40ms_modulation_mae_db": 1.5542781830779628, "mr_stft_db": 5.4583864803337905, "band_mae_db": 0.3977995503035801, "rms_gain_error_db": -0.11070106812814297, "output_peak": 0.3949109457748863}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0428832919375617, "全帯域_10ms_rms_envelope_nrmse": 0.057171716260062976, "全帯域_10ms_envelope_correlation": 0.9973514798929276, "全帯域_10ms_modulation_mae_db": 0.4462336094305703, "全帯域_40ms_log_rms_mae_db": 0.6379097518891957, "全帯域_40ms_rms_envelope_nrmse": 0.032020692223996304, "全帯域_40ms_envelope_correlation": 0.9992349294554943, "全帯域_40ms_modulation_mae_db": 0.6346128440656029, "300–1000 Hz_10ms_log_rms_mae_db": 1.2172182762601211, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12099151267137394, "300–1000 Hz_10ms_envelope_correlation": 0.9899866864970293, "300–1000 Hz_10ms_modulation_mae_db": 0.6725526360811768, "300–1000 Hz_40ms_log_rms_mae_db": 0.8278757337629702, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0741679232920428, "300–1000 Hz_40ms_envelope_correlation": 0.996327188452454, "300–1000 Hz_40ms_modulation_mae_db": 0.9136054508857295, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4077243297844133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1514459792684068, "1000–3000 Hz_10ms_envelope_correlation": 0.9876831101450158, "1000–3000 Hz_10ms_modulation_mae_db": 1.0934228871980476, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1278250673806285, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10419512863849895, "1000–3000 Hz_40ms_envelope_correlation": 0.9959437216696423, "1000–3000 Hz_40ms_modulation_mae_db": 1.1123869227772716, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7569170899706854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21895940171659076, "3000–6000 Hz_10ms_envelope_correlation": 0.9638335889792131, "3000–6000 Hz_10ms_modulation_mae_db": 1.3637099220803937, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2958486876540656, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14399907100791867, "3000–6000 Hz_40ms_envelope_correlation": 0.983747559096085, "3000–6000 Hz_40ms_modulation_mae_db": 1.7289789293380862, "mr_stft_db": 5.44431310004406, "band_mae_db": 0.32699269798356273, "rms_gain_error_db": -0.1162708151776585, "output_peak": 0.41316181658621504}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.035693843278277, "全帯域_10ms_rms_envelope_nrmse": 0.059339146677212413, "全帯域_10ms_envelope_correlation": 0.9972569526630907, "全帯域_10ms_modulation_mae_db": 0.2589737713880609, "全帯域_40ms_log_rms_mae_db": 0.6126661441645208, "全帯域_40ms_rms_envelope_nrmse": 0.038886090891265136, "全帯域_40ms_envelope_correlation": 0.9989250447369944, "全帯域_40ms_modulation_mae_db": 0.34575535070049696, "300–1000 Hz_10ms_log_rms_mae_db": 1.2332667137362652, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1431356115733186, "300–1000 Hz_10ms_envelope_correlation": 0.9864974783154877, "300–1000 Hz_10ms_modulation_mae_db": 1.3274025412098016, "300–1000 Hz_40ms_log_rms_mae_db": 0.8282363199397382, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08749292679885276, "300–1000 Hz_40ms_envelope_correlation": 0.9956325763677447, "300–1000 Hz_40ms_modulation_mae_db": 1.4133765121161217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.457499344504735, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17706513216574088, "1000–3000 Hz_10ms_envelope_correlation": 0.9806072863440393, "1000–3000 Hz_10ms_modulation_mae_db": 1.8470100474730426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1337224859532526, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1113864695531873, "1000–3000 Hz_40ms_envelope_correlation": 0.9933206783698675, "1000–3000 Hz_40ms_modulation_mae_db": 1.7741282152328655, "3000–6000 Hz_10ms_log_rms_mae_db": 1.729171837741981, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20949104550250158, "3000–6000 Hz_10ms_envelope_correlation": 0.9672766160615652, "3000–6000 Hz_10ms_modulation_mae_db": 1.307266765322254, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2272999194404883, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1310890103484318, "3000–6000 Hz_40ms_envelope_correlation": 0.9872846970891463, "3000–6000 Hz_40ms_modulation_mae_db": 1.6439853068863108, "mr_stft_db": 5.445907270106319, "band_mae_db": 0.44313911673216455, "rms_gain_error_db": -0.16131989531034482, "output_peak": 0.40283864729006524}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9889407851579123, "全帯域_10ms_rms_envelope_nrmse": 0.04917831052763973, "全帯域_10ms_envelope_correlation": 0.9980789254190752, "全帯域_10ms_modulation_mae_db": 0.6091020463308878, "全帯域_40ms_log_rms_mae_db": 0.5803572514798286, "全帯域_40ms_rms_envelope_nrmse": 0.028437952743009594, "全帯域_40ms_envelope_correlation": 0.9994281501257001, "全帯域_40ms_modulation_mae_db": 0.7961689969128374, "300–1000 Hz_10ms_log_rms_mae_db": 1.151154865550299, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12996368787241297, "300–1000 Hz_10ms_envelope_correlation": 0.989160636585254, "300–1000 Hz_10ms_modulation_mae_db": 0.9267756031781852, "300–1000 Hz_40ms_log_rms_mae_db": 0.7581537105425531, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08538724762513321, "300–1000 Hz_40ms_envelope_correlation": 0.9959404509506375, "300–1000 Hz_40ms_modulation_mae_db": 1.2340188786584363, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4425176521491645, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16975170340690748, "1000–3000 Hz_10ms_envelope_correlation": 0.9820472248722735, "1000–3000 Hz_10ms_modulation_mae_db": 1.3247685091808243, "1000–3000 Hz_40ms_log_rms_mae_db": 1.037867314096726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10201472749064171, "1000–3000 Hz_40ms_envelope_correlation": 0.9943793293478974, "1000–3000 Hz_40ms_modulation_mae_db": 1.3539904802918403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6717061278634453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20587174334168484, "3000–6000 Hz_10ms_envelope_correlation": 0.969374193974295, "3000–6000 Hz_10ms_modulation_mae_db": 1.3374904087605033, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1300212521980115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12392034251876817, "3000–6000 Hz_40ms_envelope_correlation": 0.990271591886512, "3000–6000 Hz_40ms_modulation_mae_db": 1.7924778755229653, "mr_stft_db": 5.438320437630026, "band_mae_db": 0.34464726635535214, "rms_gain_error_db": -0.11846180953544119, "output_peak": 0.4091120397456014}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0307145233519563, "全帯域_10ms_rms_envelope_nrmse": 0.05493212041133573, "全帯域_10ms_envelope_correlation": 0.9974794245491325, "全帯域_10ms_modulation_mae_db": 0.36884250492323584, "全帯域_40ms_log_rms_mae_db": 0.6015184126912653, "全帯域_40ms_rms_envelope_nrmse": 0.0288777570788617, "全帯域_40ms_envelope_correlation": 0.9993037656823587, "全帯域_40ms_modulation_mae_db": 0.49715355352531193, "300–1000 Hz_10ms_log_rms_mae_db": 1.251862298934774, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1185499981690503, "300–1000 Hz_10ms_envelope_correlation": 0.9906242599726829, "300–1000 Hz_10ms_modulation_mae_db": 0.8826345284824632, "300–1000 Hz_40ms_log_rms_mae_db": 0.8432468273559259, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07785490888394388, "300–1000 Hz_40ms_envelope_correlation": 0.9961721695004155, "300–1000 Hz_40ms_modulation_mae_db": 1.2125135385792922, "1000–3000 Hz_10ms_log_rms_mae_db": 1.488174192354302, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17390387061907406, "1000–3000 Hz_10ms_envelope_correlation": 0.9810841008251463, "1000–3000 Hz_10ms_modulation_mae_db": 1.170861183991157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.132874318808488, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11073867046065201, "1000–3000 Hz_40ms_envelope_correlation": 0.9930529932991605, "1000–3000 Hz_40ms_modulation_mae_db": 1.3659015123230749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7586035716040702, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21296637793960835, "3000–6000 Hz_10ms_envelope_correlation": 0.9657728998824461, "3000–6000 Hz_10ms_modulation_mae_db": 1.4842476996990606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2678995833008049, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1274453072603265, "3000–6000 Hz_40ms_envelope_correlation": 0.9874439755029665, "3000–6000 Hz_40ms_modulation_mae_db": 1.933539678732025, "mr_stft_db": 5.449494874347693, "band_mae_db": 0.28343990161522287, "rms_gain_error_db": -0.07073947458926919, "output_peak": 0.41233033836554484}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.029536931773013, "全帯域_10ms_rms_envelope_nrmse": 0.05350451407290256, "全帯域_10ms_envelope_correlation": 0.997574218357654, "全帯域_10ms_modulation_mae_db": 0.14524382817572976, "全帯域_40ms_log_rms_mae_db": 0.6323429582902264, "全帯域_40ms_rms_envelope_nrmse": 0.029180795131607536, "全帯域_40ms_envelope_correlation": 0.9992378446446707, "全帯域_40ms_modulation_mae_db": 0.0877154296900644, "300–1000 Hz_10ms_log_rms_mae_db": 1.2325141550764132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11585762512168461, "300–1000 Hz_10ms_envelope_correlation": 0.990926128386438, "300–1000 Hz_10ms_modulation_mae_db": 0.7470575026273295, "300–1000 Hz_40ms_log_rms_mae_db": 0.7863862702838659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05985780916456715, "300–1000 Hz_40ms_envelope_correlation": 0.9978634135200598, "300–1000 Hz_40ms_modulation_mae_db": 1.4682086355598942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4532232257473987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16245164880815738, "1000–3000 Hz_10ms_envelope_correlation": 0.9828343206342001, "1000–3000 Hz_10ms_modulation_mae_db": 1.0911252033479444, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0938056823613118, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10881231280626708, "1000–3000 Hz_40ms_envelope_correlation": 0.992124913841187, "1000–3000 Hz_40ms_modulation_mae_db": 0.582194360809365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6467475791280128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19494862857481243, "3000–6000 Hz_10ms_envelope_correlation": 0.9723908378506635, "3000–6000 Hz_10ms_modulation_mae_db": 1.350230895313981, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1226652998174405, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11719793127184117, "3000–6000 Hz_40ms_envelope_correlation": 0.9909345666497947, "3000–6000 Hz_40ms_modulation_mae_db": 1.7717095957325053, "mr_stft_db": 5.460802857249484, "band_mae_db": 0.2597658893196566, "rms_gain_error_db": -0.029339739386094882, "output_peak": 0.42817073687795026}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44400, "output_samples": 44400, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.611697811298446, "spectral_convergence": 0.17782543919640392, "spectral_convergence_status": "ok", "amplitude_floor": 1.585777823535548e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.377245945299212, "spectral_convergence": 0.16978028961260028, "spectral_convergence_status": "ok", "amplitude_floor": 7.694479165248034e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.411619513445859, "spectral_convergence": 0.1708148892063818, "spectral_convergence_status": "ok", "amplitude_floor": 6.411814451459315e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00186841781908955, "output_power": 0.0018242862888065432, "delta_db": -0.10381006478220546, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005255584463676, "output_power": 0.0005103731448063084, "delta_db": -0.12733205839238554, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025377475865110864, "output_power": 0.00026071252074642793, "delta_db": 0.11713465289771308, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00014708318852606447, "output_power": 0.0001252196298654453, "delta_db": -0.6989062037469339, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.522695850195022e-06, "output_power": 7.800419097742528e-06, "delta_db": -0.8664197650399914, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.071083630879648e-06, "output_power": 5.7432926942356874e-06, "delta_db": -0.24105265737574322, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.35910923370582876, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11952433444721536, "pitch": null, "output_rms_dbfs": -25.63179954749851, "output_peak": 0.40427436728401395, "reference_rms_dbfs": -25.512275213051296, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.466854423347839}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44400, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 0.00014622417526559514, "log_rms_mae_db": 0.8951912292139265, "rms_envelope_nrmse": 0.05637181682833043, "envelope_correlation": 0.9974296561732496, "rms_gain_error_db": -0.11952357741589036, "reference_mean_power": 0.0028155003489198547, "output_mean_power": 0.0027390706084896702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.019039396656961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008418483111337408, "output_power": 0.0008225103053016442, "reference_active": true, "delta_db": -0.10092497008060065}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001428627689371398, "output_power": 0.00013856131103039384, "reference_active": true, "delta_db": -0.13277079405469158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9192224262112575e-05, "output_power": 1.5236364819880437e-05, "reference_active": true, "delta_db": -1.0024394687166691}], "active_band_count": 3, "active_band_mae_db": 0.4120450776173204}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 0.00013893685565863478, "log_rms_mae_db": 0.5780607515619969, "rms_envelope_nrmse": 0.03301111405255345, "envelope_correlation": 0.9991732738932101, "rms_gain_error_db": -0.11952255133000882, "reference_mean_power": 0.0028463248833832306, "output_mean_power": 0.002769059032533339}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.8422021834325858e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007582285194051116, "output_power": 0.000740580389764316, "reference_active": true, "delta_db": -0.10227907662128154}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661169958011628e-05, "output_power": 6.583009572301213e-05, "reference_active": true, "delta_db": -0.05126028111848598}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7101796636835213e-06, "output_power": 1.4034814876870516e-06, "reference_active": true, "delta_db": -0.8583504924574735}], "active_band_count": 3, "active_band_mae_db": 0.3372966167324137}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 8.753945313428826e-05, "log_rms_mae_db": 1.1981758389964334, "rms_envelope_nrmse": 0.12545774191155398, "envelope_correlation": 0.9890418768234684, "rms_gain_error_db": -0.1564351425490374, "reference_mean_power": 0.0005201241069420424, "output_mean_power": 0.0005017223743520536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 5.057137989146479e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002701505367125327, "output_power": 0.00025826179664985624, "reference_active": true, "delta_db": -0.19545666738977566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.4045222297363644e-05, "output_power": 5.1957685871466105e-05, "reference_active": true, "delta_db": -0.17107507596556837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.5635511670477814e-06, "output_power": 5.920231983056113e-06, "reference_active": true, "delta_db": -1.0638702442895662}], "active_band_count": 3, "active_band_mae_db": 0.4768006625483034}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 7.440703161891069e-05, "log_rms_mae_db": 0.7818838269073859, "rms_envelope_nrmse": 0.0752319895853338, "envelope_correlation": 0.9959829477458221, "rms_gain_error_db": -0.15643531451727286, "reference_mean_power": 0.0005258188868038473, "output_mean_power": 0.0005072156556462998}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 4.6537380366659205e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025071266719842557, "output_power": 0.00023967703975211524, "reference_active": true, "delta_db": -0.1954984505169767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3806665370442115e-05, "output_power": 2.2628021097935408e-05, "reference_active": true, "delta_db": -0.2205199249419504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0797322752712393e-06, "output_power": 9.93762891161403e-07, "reference_active": true, "delta_db": -0.36033308063373914}], "active_band_count": 3, "active_band_mae_db": 0.25878381869755546}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 5.3294712554339956e-05, "log_rms_mae_db": 1.4364109066680366, "rms_envelope_nrmse": 0.17625877796108574, "envelope_correlation": 0.9815481939542431, "rms_gain_error_db": -0.6058702420218521, "reference_mean_power": 0.00010733282194912887, "output_mean_power": 9.335670684763394e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 1.1016693525679884e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607161402154175e-05, "output_power": 5.727559545186404e-05, "reference_active": true, "delta_db": -0.6204530311991345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0221078826476055e-05, "output_power": 7.745718739459114e-06, "reference_active": true, "delta_db": -1.2043501469674798}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.376144897878774e-06, "output_power": 1.5221734868180165e-06, "reference_active": true, "delta_db": -3.4595692432729215}], "active_band_count": 3, "active_band_mae_db": 1.7614574738131787}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 4.1357257962539876e-05, "log_rms_mae_db": 1.107793354455716, "rms_envelope_nrmse": 0.1158808168482445, "envelope_correlation": 0.9935028461221366, "rms_gain_error_db": -0.6058705891198166, "reference_mean_power": 0.00010850799286328448, "output_mean_power": 9.437884783794828e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.0153781665834753e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.269855261867839e-05, "output_power": 5.325791610147658e-05, "reference_active": true, "delta_db": -0.7087334618548897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.20737764206477e-06, "output_power": 3.6739665371283957e-06, "reference_active": true, "delta_db": -0.5887629839658902}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.246816673622798e-07, "output_power": 1.0612521898460353e-07, "reference_active": true, "delta_db": -3.2574903871467913}], "active_band_count": 3, "active_band_mae_db": 1.5183289443225239}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 1.102648736164134e-05, "log_rms_mae_db": 1.6020812598861762, "rms_envelope_nrmse": 0.19927461652147213, "envelope_correlation": 0.9718493107907614, "rms_gain_error_db": -0.6116150869761381, "reference_mean_power": 5.352135557413054e-06, "output_mean_power": 4.649064760882697e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.71214230899019e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.798500894794614e-06, "output_power": 1.5030239659217403e-06, "reference_active": true, "delta_db": -0.7794475269909712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2650926706890884e-07, "output_power": 2.9339565304892724e-07, "reference_active": true, "delta_db": -1.6247479678355037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0854303858213565e-07, "output_power": 6.68463605584307e-08, "reference_active": true, "delta_db": -2.105242080326297}], "active_band_count": 3, "active_band_mae_db": 1.5031458583842572}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 9.954651742580568e-06, "log_rms_mae_db": 1.1546700102616905, "rms_envelope_nrmse": 0.12842717311149224, "envelope_correlation": 0.9896438944138446, "rms_gain_error_db": -0.6116164122297775, "reference_mean_power": 5.410729753830906e-06, "output_mean_power": 4.699960433239448e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 2.394714459074953e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.814627629314412e-06, "output_power": 1.4600981304593284e-06, "reference_active": true, "delta_db": -0.9440547422646979}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0097168111890166e-07, "output_power": 1.5136963614300514e-07, "reference_active": true, "delta_db": -1.2309609839272757}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.411981302222167e-09, "output_power": 3.94796446635862e-09, "reference_active": true, "delta_db": -3.773078205892727}], "active_band_count": 3, "active_band_mae_db": 1.9826979773615667}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8922228938571731, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0cb7723e6f0d591a9deda6e0b7f49aaf951069e4b4597e9570818280aad1a8de", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9818207485449617, "standard_deviation": 0.0034262623285618845, "minimum": 0.9753352839386816, "maximum": 0.9876831101450158, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4429923582829816, "standard_deviation": 0.02409412273294775, "minimum": 1.4077243297844133, "maximum": 1.488174192354302, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.471842670112121, "standard_deviation": 0.3627619434139726, "minimum": 1.0911252033479444, "maximum": 2.016062331537226, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17075112488520083, "standard_deviation": 0.013058450611499604, "minimum": 0.1514459792684068, "maximum": 0.19456695653392633, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.993095151061693, "standard_deviation": 0.0021516866231091427, "minimum": 0.9885313631009007, "maximum": 0.9959437216696423, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1052453663678214, "standard_deviation": 0.03585530576964096, "minimum": 1.037867314096726, "maximum": 1.137466333009407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.389907536316861, "standard_deviation": 0.3956162570622606, "minimum": 0.582194360809365, "maximum": 1.7741282152328655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1101396944927167, "standard_deviation": 0.009558129499407427, "minimum": 0.09869841413416243, "maximum": 0.1293910160100796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.968041185757949, "standard_deviation": 0.003145667285535702, "minimum": 0.9638335889792131, "maximum": 0.9723908378506635, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6908487772105119, "standard_deviation": 0.06497403513998423, "minimum": 1.6020812598861762, "maximum": 1.7586035716040702, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3084574174329564, "standard_deviation": 0.17742979784163163, "minimum": 0.9529948015399747, "maximum": 1.5031458583842572, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20777874020542814, "standard_deviation": 0.008277441189423272, "minimum": 0.19494862857481243, "maximum": 0.21895940171659076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9882580733257245, "standard_deviation": 0.0022458347628090286, "minimum": 0.983747559096085, "maximum": 0.9909345666497947, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1909984994908487, "standard_deviation": 0.06857813697482637, "minimum": 1.122279367700886, "maximum": 1.2958486876540656, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.7323001693631188, "standard_deviation": 0.1801399656084522, "minimum": 1.4507338082535286, "maximum": 1.9826979773615667, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12797674036577522, "standard_deviation": 0.007665521174538919, "minimum": 0.11719793127184117, "maximum": 0.14399907100791867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9900610147730424, "standard_deviation": 0.002153979032471417, "minimum": 0.9864974783154877, "maximum": 0.9941668074571945, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2056119107324301, "standard_deviation": 0.04316401323599256, "minimum": 1.1304388219635226, "maximum": 1.251862298934774, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8228179743570375, "standard_deviation": 0.339608680982907, "minimum": 0.335534922865685, "maximum": 1.3274025412098016, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12159458133107787, "standard_deviation": 0.014332152505145699, "minimum": 0.0928330657916038, "maximum": 0.1431356115733186, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966344621738332, "standard_deviation": 0.0009525019765473407, "minimum": 0.9956325763677447, "maximum": 0.9982266062779267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.800043204225949, "standard_deviation": 0.029402438737644684, "minimum": 0.7581537105425531, "maximum": 0.8432468273559259, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0060705097827034, "standard_deviation": 0.4601782832353207, "minimum": 0.25878381869755546, "maximum": 1.4682086355598942, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07401309051626372, "standard_deviation": 0.012013151622078246, "minimum": 0.05271691652789745, "maximum": 0.08749292679885276, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3392443991806023, "standard_deviation": 0.060759905390633195, "minimum": 0.2597658893196566, "maximum": 0.44313911673216455, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.4519473740693565, "standard_deviation": 0.009477608832416258, "minimum": 5.438320437630026, "maximum": 5.466854423347839, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40847380533382244, "standard_deviation": 0.00991840506213337, "minimum": 0.3949109457748863, "maximum": 0.42817073687795026, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09796948050765478, "standard_deviation": 0.04224479125841388, "minimum": -0.16131989531034482, "maximum": -0.029339739386094882, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.997515953135168, "standard_deviation": 0.0002539238395901093, "minimum": 0.9972569526630907, "maximum": 0.9980789254190752, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0059088783707495, "standard_deviation": 0.061063225670959016, "minimum": 0.8951912292139265, "maximum": 1.0830382803810554, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41995493791341454, "standard_deviation": 0.16205605559482394, "minimum": 0.14524382817572976, "maximum": 0.6091020463308878, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.055112344556589365, "standard_deviation": 0.002991064923743634, "minimum": 0.04917831052763973, "maximum": 0.059339146677212413, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992394214775913, "standard_deviation": 0.000157158484782882, "minimum": 0.9989250447369944, "maximum": 0.9994281501257001, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6035253368540829, "standard_deviation": 0.04509639574426974, "minimum": 0.5194796273473753, "maximum": 0.6658677974082539, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49781477034746957, "standard_deviation": 0.22897402850065884, "minimum": 0.0877154296900644, "maximum": 0.7961689969128374, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031142332309255026, "standard_deviation": 0.0035585064579711297, "minimum": 0.028417864511287656, "maximum": 0.038886090891265136, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-85628-0025", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.71769118309021, "predicted_mos": 2.71769118309021, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "3659ba1daf32b5cf2868eea4f59b3aaf99b0f161534520744a6007a2089c7e86", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8577634003013372, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9586888464638618, "全帯域_10ms_rms_envelope_nrmse": 0.055648376193552916, "全帯域_10ms_envelope_correlation": 0.9975885554152432, "全帯域_10ms_modulation_mae_db": 0.3844270228605334, "全帯域_40ms_log_rms_mae_db": 0.6063643029812048, "全帯域_40ms_rms_envelope_nrmse": 0.03208113454140041, "全帯域_40ms_envelope_correlation": 0.9993300001231075, "全帯域_40ms_modulation_mae_db": 0.4445522290709764, "300–1000 Hz_10ms_log_rms_mae_db": 1.2696078776172863, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11101344886403228, "300–1000 Hz_10ms_envelope_correlation": 0.9922236746369293, "300–1000 Hz_10ms_modulation_mae_db": 1.6583140969320471, "300–1000 Hz_40ms_log_rms_mae_db": 0.7756290201342808, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06669635551452716, "300–1000 Hz_40ms_envelope_correlation": 0.9978191230275932, "300–1000 Hz_40ms_modulation_mae_db": 1.706063831756569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6314522396800224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17008581043569482, "1000–3000 Hz_10ms_envelope_correlation": 0.9818981665192481, "1000–3000 Hz_10ms_modulation_mae_db": 1.4224124931967073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1536096899371497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10344986879468032, "1000–3000 Hz_40ms_envelope_correlation": 0.9939923829288457, "1000–3000 Hz_40ms_modulation_mae_db": 1.7155663387034517, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1984764282464457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23671161554830872, "3000–6000 Hz_10ms_envelope_correlation": 0.9577025457981847, "3000–6000 Hz_10ms_modulation_mae_db": 1.3491052583999543, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5310171202225638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13769024795515264, "3000–6000 Hz_40ms_envelope_correlation": 0.985469858670688, "3000–6000 Hz_40ms_modulation_mae_db": 2.25233782811656, "mr_stft_db": 5.8500004784067, "band_mae_db": 0.29305919263430624, "rms_gain_error_db": -0.15151590739057058, "output_peak": 0.3995242843724481}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8964548874620956, "全帯域_10ms_rms_envelope_nrmse": 0.05630600633763099, "全帯域_10ms_envelope_correlation": 0.9974644377629646, "全帯域_10ms_modulation_mae_db": 0.11656491553078985, "全帯域_40ms_log_rms_mae_db": 0.49103873026652856, "全帯域_40ms_rms_envelope_nrmse": 0.029435040695378573, "全帯域_40ms_envelope_correlation": 0.9994122526735175, "全帯域_40ms_modulation_mae_db": 0.29074003994748526, "300–1000 Hz_10ms_log_rms_mae_db": 1.196188096121612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10679601345814164, "300–1000 Hz_10ms_envelope_correlation": 0.9928815795417594, "300–1000 Hz_10ms_modulation_mae_db": 0.9636928682007998, "300–1000 Hz_40ms_log_rms_mae_db": 0.7687664872345706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0710162935088904, "300–1000 Hz_40ms_envelope_correlation": 0.997349807141606, "300–1000 Hz_40ms_modulation_mae_db": 1.0305504645425847, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7233842147310068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16846340182926606, "1000–3000 Hz_10ms_envelope_correlation": 0.9814529620416395, "1000–3000 Hz_10ms_modulation_mae_db": 0.504207317899962, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2807150928835183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09550228701186814, "1000–3000 Hz_40ms_envelope_correlation": 0.9938739900523481, "1000–3000 Hz_40ms_modulation_mae_db": 0.5656287290673707, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1589978606010565, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2521612155511579, "3000–6000 Hz_10ms_envelope_correlation": 0.9518388927629073, "3000–6000 Hz_10ms_modulation_mae_db": 1.1918199389303243, "3000–6000 Hz_40ms_log_rms_mae_db": 1.562889614467665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15619236527477215, "3000–6000 Hz_40ms_envelope_correlation": 0.9806182653004762, "3000–6000 Hz_40ms_modulation_mae_db": 2.209494160387996, "mr_stft_db": 5.8233526351963105, "band_mae_db": 0.1893375566042138, "rms_gain_error_db": -0.13102983250582179, "output_peak": 0.41399161437488796}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9005811400840503, "全帯域_10ms_rms_envelope_nrmse": 0.07024279466258168, "全帯域_10ms_envelope_correlation": 0.9958114956227225, "全帯域_10ms_modulation_mae_db": 0.36878636294748035, "全帯域_40ms_log_rms_mae_db": 0.5438756122502194, "全帯域_40ms_rms_envelope_nrmse": 0.040217248787563824, "全帯域_40ms_envelope_correlation": 0.9985400951273158, "全帯域_40ms_modulation_mae_db": 0.5217653179875971, "300–1000 Hz_10ms_log_rms_mae_db": 1.2950769456949354, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13012562858381077, "300–1000 Hz_10ms_envelope_correlation": 0.988846373285795, "300–1000 Hz_10ms_modulation_mae_db": 1.0225869406663175, "300–1000 Hz_40ms_log_rms_mae_db": 0.8061378395684048, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07852984089510934, "300–1000 Hz_40ms_envelope_correlation": 0.996478028349854, "300–1000 Hz_40ms_modulation_mae_db": 1.007795226063752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.668573461353443, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1771467610090303, "1000–3000 Hz_10ms_envelope_correlation": 0.9794868665764491, "1000–3000 Hz_10ms_modulation_mae_db": 1.1635207098628566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2097487271913696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11181703543328451, "1000–3000 Hz_40ms_envelope_correlation": 0.99164010023739, "1000–3000 Hz_40ms_modulation_mae_db": 1.2148648875653445, "3000–6000 Hz_10ms_log_rms_mae_db": 2.220274873472707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25238487271233234, "3000–6000 Hz_10ms_envelope_correlation": 0.9522839131759785, "3000–6000 Hz_10ms_modulation_mae_db": 0.8017757680643034, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5590057909499444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14728485754382248, "3000–6000 Hz_40ms_envelope_correlation": 0.982622317344561, "3000–6000 Hz_40ms_modulation_mae_db": 1.2877415925595137, "mr_stft_db": 5.829081909547576, "band_mae_db": 0.16671878822048206, "rms_gain_error_db": -0.01912801149724501, "output_peak": 0.4111505432139932}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0677151204787072, "全帯域_10ms_rms_envelope_nrmse": 0.06311328452408307, "全帯域_10ms_envelope_correlation": 0.9967461249223406, "全帯域_10ms_modulation_mae_db": 0.6891920450533272, "全帯域_40ms_log_rms_mae_db": 0.6237733761928964, "全帯域_40ms_rms_envelope_nrmse": 0.03390962554723066, "全帯域_40ms_envelope_correlation": 0.9991281607139955, "全帯域_40ms_modulation_mae_db": 0.9767509523707537, "300–1000 Hz_10ms_log_rms_mae_db": 1.314791070622057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14724088324374093, "300–1000 Hz_10ms_envelope_correlation": 0.9860634838498983, "300–1000 Hz_10ms_modulation_mae_db": 1.7025137070992151, "300–1000 Hz_40ms_log_rms_mae_db": 0.8776970084168975, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09121126962221038, "300–1000 Hz_40ms_envelope_correlation": 0.9956736284202893, "300–1000 Hz_40ms_modulation_mae_db": 2.1955084691810374, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7030713652967275, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21299367813399012, "1000–3000 Hz_10ms_envelope_correlation": 0.970368393441818, "1000–3000 Hz_10ms_modulation_mae_db": 1.042072365414602, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2362712980286605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1262747420765621, "1000–3000 Hz_40ms_envelope_correlation": 0.9899180418696969, "1000–3000 Hz_40ms_modulation_mae_db": 0.8529964431755808, "3000–6000 Hz_10ms_log_rms_mae_db": 2.265987722693781, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26850279422986095, "3000–6000 Hz_10ms_envelope_correlation": 0.9449414059005775, "3000–6000 Hz_10ms_modulation_mae_db": 1.2374775892573608, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6111499832598573, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15867149736881542, "3000–6000 Hz_40ms_envelope_correlation": 0.9801150838733769, "3000–6000 Hz_40ms_modulation_mae_db": 1.8121703932704587, "mr_stft_db": 5.850511676080835, "band_mae_db": 0.2664250982413943, "rms_gain_error_db": -0.11778153540693798, "output_peak": 0.39744488553171675}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0284668758278248, "全帯域_10ms_rms_envelope_nrmse": 0.05549858670458127, "全帯域_10ms_envelope_correlation": 0.9974538104941377, "全帯域_10ms_modulation_mae_db": 0.33067301859565873, "全帯域_40ms_log_rms_mae_db": 0.6055981212136499, "全帯域_40ms_rms_envelope_nrmse": 0.0274765748875163, "全帯域_40ms_envelope_correlation": 0.999406773840476, "全帯域_40ms_modulation_mae_db": 0.7588396921421944, "300–1000 Hz_10ms_log_rms_mae_db": 1.1551282034823875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09524208680090977, "300–1000 Hz_10ms_envelope_correlation": 0.9943445919747308, "300–1000 Hz_10ms_modulation_mae_db": 0.8274960822722264, "300–1000 Hz_40ms_log_rms_mae_db": 0.7505977484104077, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05502664645627974, "300–1000 Hz_40ms_envelope_correlation": 0.9986276210639915, "300–1000 Hz_40ms_modulation_mae_db": 1.2443358156780624, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6794516866089857, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17255327079850227, "1000–3000 Hz_10ms_envelope_correlation": 0.9810248949550917, "1000–3000 Hz_10ms_modulation_mae_db": 1.5607213166855398, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2533124172178516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11189191676012307, "1000–3000 Hz_40ms_envelope_correlation": 0.9922924962282893, "1000–3000 Hz_40ms_modulation_mae_db": 1.3102272615996045, "3000–6000 Hz_10ms_log_rms_mae_db": 2.285200761001217, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2857095988440101, "3000–6000 Hz_10ms_envelope_correlation": 0.9390302882412251, "3000–6000 Hz_10ms_modulation_mae_db": 1.1705206089147875, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7558888591558741, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18094779986108087, "3000–6000 Hz_40ms_envelope_correlation": 0.9735528925039786, "3000–6000 Hz_40ms_modulation_mae_db": 1.0519225100681966, "mr_stft_db": 5.817726302818926, "band_mae_db": 0.21340743840944507, "rms_gain_error_db": -0.08899158148702505, "output_peak": 0.4036659364031571}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9713547234819253, "全帯域_10ms_rms_envelope_nrmse": 0.05217379041473575, "全帯域_10ms_envelope_correlation": 0.9977751790829835, "全帯域_10ms_modulation_mae_db": 0.522134150043643, "全帯域_40ms_log_rms_mae_db": 0.5987570551317558, "全帯域_40ms_rms_envelope_nrmse": 0.029217076799514748, "全帯域_40ms_envelope_correlation": 0.9993357221965333, "全帯域_40ms_modulation_mae_db": 0.5143633692413001, "300–1000 Hz_10ms_log_rms_mae_db": 1.2242134808243783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1236790041005876, "300–1000 Hz_10ms_envelope_correlation": 0.9902410731050901, "300–1000 Hz_10ms_modulation_mae_db": 1.2605970472054804, "300–1000 Hz_40ms_log_rms_mae_db": 0.7445868120669354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07338910871169896, "300–1000 Hz_40ms_envelope_correlation": 0.9973149537110111, "300–1000 Hz_40ms_modulation_mae_db": 1.2293853788872127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.767194644609386, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17567542773970202, "1000–3000 Hz_10ms_envelope_correlation": 0.980164389772709, "1000–3000 Hz_10ms_modulation_mae_db": 1.686631426079165, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2906512640122927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10640428080512845, "1000–3000 Hz_40ms_envelope_correlation": 0.9930001337265945, "1000–3000 Hz_40ms_modulation_mae_db": 1.4507047154954629, "3000–6000 Hz_10ms_log_rms_mae_db": 2.214529753472154, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2625701330012358, "3000–6000 Hz_10ms_envelope_correlation": 0.9477272945834053, "3000–6000 Hz_10ms_modulation_mae_db": 1.5817861359067782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6533402824486396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17066643673258675, "3000–6000 Hz_40ms_envelope_correlation": 0.9772016673455305, "3000–6000 Hz_40ms_modulation_mae_db": 2.613471233439979, "mr_stft_db": 5.8339690067475685, "band_mae_db": 0.23230987945517986, "rms_gain_error_db": -0.09846474002167918, "output_peak": 0.3990973655199175}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9876018287626167, "全帯域_10ms_rms_envelope_nrmse": 0.057494403630490534, "全帯域_10ms_envelope_correlation": 0.9973264622993059, "全帯域_10ms_modulation_mae_db": 0.602479784965915, "全帯域_40ms_log_rms_mae_db": 0.6703030842221255, "全帯域_40ms_rms_envelope_nrmse": 0.03226404749861178, "全帯域_40ms_envelope_correlation": 0.9992231636840399, "全帯域_40ms_modulation_mae_db": 0.992793578239407, "300–1000 Hz_10ms_log_rms_mae_db": 1.3806194422628522, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12014245183929323, "300–1000 Hz_10ms_envelope_correlation": 0.9908543305526277, "300–1000 Hz_10ms_modulation_mae_db": 1.4528035384734108, "300–1000 Hz_40ms_log_rms_mae_db": 0.9456420951975978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07869117081544495, "300–1000 Hz_40ms_envelope_correlation": 0.9966578602561665, "300–1000 Hz_40ms_modulation_mae_db": 1.5769767191390567, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7258845029013805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19259883719445517, "1000–3000 Hz_10ms_envelope_correlation": 0.9760005363250449, "1000–3000 Hz_10ms_modulation_mae_db": 1.504238458980364, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2891506625262237, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10945621411756469, "1000–3000 Hz_40ms_envelope_correlation": 0.9926541121501883, "1000–3000 Hz_40ms_modulation_mae_db": 1.5760530259170158, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1538737666571146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24025393120044036, "3000–6000 Hz_10ms_envelope_correlation": 0.9568571229033256, "3000–6000 Hz_10ms_modulation_mae_db": 1.1274176928479673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6002658495523372, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14440441014818825, "3000–6000 Hz_40ms_envelope_correlation": 0.9832595158125157, "3000–6000 Hz_40ms_modulation_mae_db": 1.459896933841467, "mr_stft_db": 5.836895471934003, "band_mae_db": 0.2287157758846947, "rms_gain_error_db": -0.12151602127304915, "output_peak": 0.4004861688362545}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.985665873908858, "全帯域_10ms_rms_envelope_nrmse": 0.05929850463375569, "全帯域_10ms_envelope_correlation": 0.9970260957699589, "全帯域_10ms_modulation_mae_db": 0.4044924563612589, "全帯域_40ms_log_rms_mae_db": 0.5759822504150223, "全帯域_40ms_rms_envelope_nrmse": 0.02953562674121049, "全帯域_40ms_envelope_correlation": 0.9992212402204325, "全帯域_40ms_modulation_mae_db": 0.5900741577247791, "300–1000 Hz_10ms_log_rms_mae_db": 1.209344674745112, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10120341513030909, "300–1000 Hz_10ms_envelope_correlation": 0.9931745299357109, "300–1000 Hz_10ms_modulation_mae_db": 0.8699167486504727, "300–1000 Hz_40ms_log_rms_mae_db": 0.8094300232172071, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0601644741885239, "300–1000 Hz_40ms_envelope_correlation": 0.9977828910174635, "300–1000 Hz_40ms_modulation_mae_db": 1.0539307158725606, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6395702797350897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16352869809799986, "1000–3000 Hz_10ms_envelope_correlation": 0.9828320090681002, "1000–3000 Hz_10ms_modulation_mae_db": 1.1890709715140975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1942361377161157, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09278539553851282, "1000–3000 Hz_40ms_envelope_correlation": 0.9947660363140716, "1000–3000 Hz_40ms_modulation_mae_db": 1.2550067870956971, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0453293583751098, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.252521189418829, "3000–6000 Hz_10ms_envelope_correlation": 0.9525563290803837, "3000–6000 Hz_10ms_modulation_mae_db": 0.8219665962512289, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4724477040144248, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15186457994390629, "3000–6000 Hz_40ms_envelope_correlation": 0.9815042768904608, "3000–6000 Hz_40ms_modulation_mae_db": 1.4824202654622842, "mr_stft_db": 5.829351545097945, "band_mae_db": 0.13904484692591088, "rms_gain_error_db": -0.011169237367588836, "output_peak": 0.4118408129110635}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44400, "output_samples": 44400, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.032942476007613, "spectral_convergence": 0.17977751760007604, "spectral_convergence_status": "ok", "amplitude_floor": 1.585777823535548e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.814302017718933, "spectral_convergence": 0.1717164926864614, "spectral_convergence_status": "ok", "amplitude_floor": 7.694479165248034e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.702756941493554, "spectral_convergence": 0.17040825512052843, "spectral_convergence_status": "ok", "amplitude_floor": 6.411814451459315e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00186841781908955, "output_power": 0.0018269823704193918, "delta_db": -0.09739643878367009, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005255584463676, "output_power": 0.0005020174198991079, "delta_db": -0.1990223278965765, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025377475865110864, "output_power": 0.00023855234915703044, "delta_db": -0.2686472576374011, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00014708318852606447, "output_power": 0.00013238327346279266, "delta_db": -0.45729920346644276, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.522695850195022e-06, "output_power": 8.132436360105848e-06, "delta_db": -0.6853923999679221, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.071083630879648e-06, "output_power": 6.000762833958839e-06, "delta_db": -0.050597528053824975, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29305919263430624, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15151590739057058, "pitch": null, "output_rms_dbfs": -25.663791120441864, "output_peak": 0.3995242843724481, "reference_rms_dbfs": -25.512275213051296, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8500004784067}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44400, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 0.00014622417526559514, "log_rms_mae_db": 0.9586888464638618, "rms_envelope_nrmse": 0.055648376193552916, "envelope_correlation": 0.9975885554152432, "rms_gain_error_db": -0.15151552231510465, "reference_mean_power": 0.0028155003489198547, "output_mean_power": 0.0027189676051743053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.019039396656961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008418483111337408, "output_power": 0.0008151700673682024, "reference_active": true, "delta_db": -0.13985620758855227}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001428627689371398, "output_power": 0.0001304399858156382, "reference_active": true, "delta_db": -0.39508320369759514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9192224262112575e-05, "output_power": 1.6645281492346096e-05, "reference_active": true, "delta_db": -0.6183416572954527}], "active_band_count": 3, "active_band_mae_db": 0.3844270228605334}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 0.00013893685565863478, "log_rms_mae_db": 0.6063643029812048, "rms_envelope_nrmse": 0.03208113454140041, "envelope_correlation": 0.9993300001231075, "rms_gain_error_db": -0.15151474776776, "reference_mean_power": 0.0028463248833832306, "output_mean_power": 0.00274873577441463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.8422021834325858e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007582285194051116, "output_power": 0.0007344326109955982, "reference_active": true, "delta_db": -0.13848163015255646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661169958011628e-05, "output_power": 6.0186105943371076e-05, "reference_active": true, "delta_db": -0.4405426944010161}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7101796636835213e-06, "output_power": 1.4374028496911986e-06, "reference_active": true, "delta_db": -0.7546323626593565}], "active_band_count": 3, "active_band_mae_db": 0.4445522290709764}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 8.753945313428826e-05, "log_rms_mae_db": 1.2696078776172863, "rms_envelope_nrmse": 0.11101344886403228, "envelope_correlation": 0.9922236746369293, "rms_gain_error_db": -0.3574339580406627, "reference_mean_power": 0.0005201241069420424, "output_mean_power": 0.0004790309647740594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 5.057137989146479e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002701505367125327, "output_power": 0.00024219450950589022, "reference_active": true, "delta_db": -0.4744154113398958}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.4045222297363644e-05, "output_power": 4.258775384696231e-05, "reference_active": true, "delta_db": -1.0347257228738915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.5635511670477814e-06, "output_power": 3.4052235917252615e-06, "reference_active": true, "delta_db": -3.4658011565823537}], "active_band_count": 3, "active_band_mae_db": 1.6583140969320471}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 7.440703161891069e-05, "log_rms_mae_db": 0.7756290201342808, "rms_envelope_nrmse": 0.06669635551452716, "envelope_correlation": 0.9978191230275932, "rms_gain_error_db": -0.3574340647945792, "reference_mean_power": 0.0005258188868038473, "output_mean_power": 0.00048427580857848606}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 4.6537380366659205e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025071266719842557, "output_power": 0.00022334968264706197, "reference_active": true, "delta_db": -0.50190937410646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3806665370442115e-05, "output_power": 2.0013240330332058e-05, "reference_active": true, "delta_db": -0.7538115683282167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0797322752712393e-06, "output_power": 4.436791581375963e-07, "reference_active": true, "delta_db": -3.8624705528350307}], "active_band_count": 3, "active_band_mae_db": 1.706063831756569}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 5.3294712554339956e-05, "log_rms_mae_db": 1.6314522396800224, "rms_envelope_nrmse": 0.17008581043569482, "envelope_correlation": 0.9818981665192481, "rms_gain_error_db": -0.4172919398379881, "reference_mean_power": 0.00010733282194912887, "output_mean_power": 9.749971670463438e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 1.1016693525679884e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607161402154175e-05, "output_power": 6.148872225122112e-05, "reference_active": true, "delta_db": -0.31219447635934333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0221078826476055e-05, "output_power": 8.026557110052952e-06, "reference_active": true, "delta_db": -1.0496743753828284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.376144897878774e-06, "output_power": 1.729355242051714e-06, "reference_active": true, "delta_db": -2.9053686278479507}], "active_band_count": 3, "active_band_mae_db": 1.4224124931967073}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 4.1357257962539876e-05, "log_rms_mae_db": 1.1536096899371497, "rms_envelope_nrmse": 0.10344986879468032, "envelope_correlation": 0.9939923829288457, "rms_gain_error_db": -0.4172926088541902, "reference_mean_power": 0.00010850799286328448, "output_mean_power": 9.856721124537764e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.0153781665834753e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.269855261867839e-05, "output_power": 5.760450491358436e-05, "reference_active": true, "delta_db": -0.3680106696155935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.20737764206477e-06, "output_power": 3.4362289939880875e-06, "reference_active": true, "delta_db": -0.8792939681082819}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.246816673622798e-07, "output_power": 9.154364489574014e-08, "reference_active": true, "delta_db": -3.8993943783864795}], "active_band_count": 3, "active_band_mae_db": 1.7155663387034517}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 1.102648736164134e-05, "log_rms_mae_db": 2.1984764282464457, "rms_envelope_nrmse": 0.23671161554830872, "envelope_correlation": 0.9577025457981847, "rms_gain_error_db": -0.29625590728135776, "reference_mean_power": 5.352135557413054e-06, "output_mean_power": 4.999211536448959e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.71214230899019e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.798500894794614e-06, "output_power": 1.5703526834376723e-06, "reference_active": true, "delta_db": -0.5891345723725206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2650926706890884e-07, "output_power": 2.205165536319093e-07, "reference_active": true, "delta_db": -2.8648727539037933}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0854303858213565e-07, "output_power": 9.468280821619892e-08, "reference_active": true, "delta_db": -0.5933084489235487}], "active_band_count": 3, "active_band_mae_db": 1.3491052583999543}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 9.954651742580568e-06, "log_rms_mae_db": 1.5310171202225638, "rms_envelope_nrmse": 0.13769024795515264, "envelope_correlation": 0.985469858670688, "rms_gain_error_db": -0.2962609768776258, "reference_mean_power": 5.410729753830906e-06, "output_mean_power": 5.053936085996531e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 2.394714459074953e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.814627629314412e-06, "output_power": 1.516857715834047e-06, "reference_active": true, "delta_db": -0.7784267406906797}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0097168111890166e-07, "output_power": 1.0230792112118998e-07, "reference_active": true, "delta_db": -2.9322560541067273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.411981302222167e-09, "output_power": 4.667109460620469e-09, "reference_active": true, "delta_db": -3.0463306895522733}], "active_band_count": 3, "active_band_mae_db": 2.25233782811656}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8786602796379714, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "98965219dd06ed2a129a1501f857b3a657823eb0816e6d203e9e9c1a3aab509e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9791535273375125, "standard_deviation": 0.004107598822390747, "minimum": 0.970368393441818, "maximum": 0.9828320090681002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6923227993645051, "standard_deviation": 0.04633769571914917, "minimum": 1.6314522396800224, "maximum": 1.767194644609386, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2591093824541617, "standard_deviation": 0.376359591341137, "minimum": 0.504207317899962, "maximum": 1.686631426079165, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17913073565483006, "standard_deviation": 0.016152192436936566, "minimum": 0.16352869809799986, "maximum": 0.21299367813399012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927671616884279, "standard_deviation": 0.0015307258132666096, "minimum": 0.9899180418696969, "maximum": 0.9947660363140716, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2384619111891477, "standard_deviation": 0.04969048653531149, "minimum": 1.1536096899371497, "maximum": 1.2906512640122927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.242631023577441, "standard_deviation": 0.3767858974175332, "minimum": 0.5656287290673707, "maximum": 1.7155663387034517, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10719771756721552, "standard_deviation": 0.010498852378507409, "minimum": 0.09278539553851282, "maximum": 0.1262747420765621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9503672240557486, "standard_deviation": 0.006227729038370164, "minimum": 0.9390302882412251, "maximum": 0.9577025457981847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1928338155649483, "standard_deviation": 0.07517506553388978, "minimum": 2.0453293583751098, "maximum": 2.285200761001217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.160233698571588, "standard_deviation": 0.2575829031525175, "minimum": 0.8017757680643034, "maximum": 1.5817861359067782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2563519188132719, "standard_deviation": 0.01577087728186804, "minimum": 0.23671161554830872, "maximum": 0.2857095988440101, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9805429847176984, "standard_deviation": 0.003724741451983678, "minimum": 0.9735528925039786, "maximum": 0.985469858670688, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5932506505089132, "standard_deviation": 0.08535004979365501, "minimum": 1.4724477040144248, "maximum": 1.7558888591558741, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.771181864643307, "standard_deviation": 0.5432823858173723, "minimum": 1.0519225100681966, "maximum": 2.613471233439979, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1559652743535406, "standard_deviation": 0.014179570167739358, "minimum": 0.13769024795515264, "maximum": 0.18094779986108087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9910787046103178, "standard_deviation": 0.002686142511014455, "minimum": 0.9860634838498983, "maximum": 0.9943445919747308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2556212239213276, "standard_deviation": 0.0733261938936767, "minimum": 1.1551282034823875, "maximum": 1.3806194422628522, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.219740128687496, "standard_deviation": 0.3509307344608171, "minimum": 0.8274960822722264, "maximum": 1.7025137070992151, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11693036650260316, "standard_deviation": 0.016920560063351766, "minimum": 0.09524208680090977, "maximum": 0.14724088324374093, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972129891234969, "standard_deviation": 0.0009212144721129754, "minimum": 0.9956736284202893, "maximum": 0.9986276210639915, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8098108792807878, "standard_deviation": 0.06937032861300967, "minimum": 0.7445868120669354, "maximum": 0.9456420951975978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3805683276401046, "standard_deviation": 0.41717262857453574, "minimum": 1.007795226063752, "maximum": 2.1955084691810374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0718406449640856, "standard_deviation": 0.011436037433922214, "minimum": 0.05502664645627974, "maximum": 0.09121126962221038, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21612732204695337, "standard_deviation": 0.050676458075601054, "minimum": 0.13904484692591088, "maximum": 0.29305919263430624, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.833861128228733, "standard_deviation": 0.011716145974508468, "minimum": 5.817726302818926, "maximum": 5.850511676080835, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40465020139542984, "standard_deviation": 0.006639221847597896, "minimum": 0.39744488553171675, "maximum": 0.41399161437488796, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0924496083687397, "standard_deviation": 0.05140111074089436, "minimum": -0.15151590739057058, "maximum": -0.011169237367588836, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.997149020171207, "standard_deviation": 0.0006296874767117865, "minimum": 0.9958114956227225, "maximum": 0.9977751790829835, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9745661620587425, "standard_deviation": 0.058202484965422366, "minimum": 0.8964548874620956, "maximum": 1.0677151204787072, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42734371954482586, "standard_deviation": 0.17725852356631944, "minimum": 0.11656491553078985, "maximum": 0.6891920450533272, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05872196838767649, "standard_deviation": 0.005636679247644666, "minimum": 0.05217379041473575, "maximum": 0.07024279466258168, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991996760724272, "standard_deviation": 0.00028403957293610794, "minimum": 0.9985400951273158, "maximum": 0.9994122526735175, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5894615665841753, "standard_deviation": 0.05384100158100915, "minimum": 0.49103873026652856, "maximum": 0.6703030842221255, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6362349170905616, "standard_deviation": 0.25193902947407637, "minimum": 0.29074003994748526, "maximum": 0.992793578239407, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03176704693730335, "standard_deviation": 0.003994106413711809, "minimum": 0.0274765748875163, "maximum": 0.040217248787563824, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-85628-0025", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.136471748352051, "predicted_mos": 3.136471748352051, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "0ee5fedf54e5ca3bf133a9601f0eb677c557d2420288f014600764403a7835d1", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9127177000045776, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.020881912389937, "全帯域_10ms_rms_envelope_nrmse": 0.050730593175873424, "全帯域_10ms_envelope_correlation": 0.9978664130792263, "全帯域_10ms_modulation_mae_db": 0.3384826127736234, "全帯域_40ms_log_rms_mae_db": 0.6484698728205627, "全帯域_40ms_rms_envelope_nrmse": 0.024380287373597886, "全帯域_40ms_envelope_correlation": 0.9995303481982089, "全帯域_40ms_modulation_mae_db": 0.3485626223384128, "300–1000 Hz_10ms_log_rms_mae_db": 1.1982228097118117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10689639402443143, "300–1000 Hz_10ms_envelope_correlation": 0.9925590912124655, "300–1000 Hz_10ms_modulation_mae_db": 1.0385710587691204, "300–1000 Hz_40ms_log_rms_mae_db": 0.8483163913174819, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06254269512207629, "300–1000 Hz_40ms_envelope_correlation": 0.9978585789188237, "300–1000 Hz_40ms_modulation_mae_db": 0.9686121288168956, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3680388926258789, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18736450646573286, "1000–3000 Hz_10ms_envelope_correlation": 0.9773132722837841, "1000–3000 Hz_10ms_modulation_mae_db": 0.579428598253973, "1000–3000 Hz_40ms_log_rms_mae_db": 1.03631827164454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11567878484954107, "1000–3000 Hz_40ms_envelope_correlation": 0.9916136581326531, "1000–3000 Hz_40ms_modulation_mae_db": 0.4976077943483274, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5357312505906928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19555369951757556, "3000–6000 Hz_10ms_envelope_correlation": 0.9722125185982589, "3000–6000 Hz_10ms_modulation_mae_db": 0.5379097925793429, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0735546209926867, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11495644560208836, "3000–6000 Hz_40ms_envelope_correlation": 0.9914274176116835, "3000–6000 Hz_40ms_modulation_mae_db": 0.7192844547404734, "mr_stft_db": 5.395636582808122, "band_mae_db": 0.34799135700051725, "rms_gain_error_db": -0.07733992152274705, "output_peak": 0.4120268388074869}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0758748255758286, "全帯域_10ms_rms_envelope_nrmse": 0.058188429298838516, "全帯域_10ms_envelope_correlation": 0.9974095127073355, "全帯域_10ms_modulation_mae_db": 0.3858873969665762, "全帯域_40ms_log_rms_mae_db": 0.6499342631151682, "全帯域_40ms_rms_envelope_nrmse": 0.037408617651772114, "全帯域_40ms_envelope_correlation": 0.99906610924953, "全帯域_40ms_modulation_mae_db": 0.4537256414755839, "300–1000 Hz_10ms_log_rms_mae_db": 1.1017789045319777, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11424420889146894, "300–1000 Hz_10ms_envelope_correlation": 0.9921905719969767, "300–1000 Hz_10ms_modulation_mae_db": 1.4472009257096994, "300–1000 Hz_40ms_log_rms_mae_db": 0.8177359249812903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07376298114636173, "300–1000 Hz_40ms_envelope_correlation": 0.9976390338278505, "300–1000 Hz_40ms_modulation_mae_db": 1.2369675022424405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4236685817658066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17742667423524808, "1000–3000 Hz_10ms_envelope_correlation": 0.9811940489947648, "1000–3000 Hz_10ms_modulation_mae_db": 0.882510471561937, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0567985028271867, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11257882109685766, "1000–3000 Hz_40ms_envelope_correlation": 0.9940289766380007, "1000–3000 Hz_40ms_modulation_mae_db": 1.1647824572529035, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5486269116582767, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21542359604469355, "3000–6000 Hz_10ms_envelope_correlation": 0.9666946098058099, "3000–6000 Hz_10ms_modulation_mae_db": 1.3831377937234066, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0749416127684623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14827484810906946, "3000–6000 Hz_40ms_envelope_correlation": 0.9851999908273117, "3000–6000 Hz_40ms_modulation_mae_db": 1.8663210830984642, "mr_stft_db": 5.406142140621385, "band_mae_db": 0.3696865060316288, "rms_gain_error_db": -0.17068211120308682, "output_peak": 0.4079312670612925}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9746531857617257, "全帯域_10ms_rms_envelope_nrmse": 0.05500693236691692, "全帯域_10ms_envelope_correlation": 0.997485143645812, "全帯域_10ms_modulation_mae_db": 0.5554293865605758, "全帯域_40ms_log_rms_mae_db": 0.5696569893023015, "全帯域_40ms_rms_envelope_nrmse": 0.026732246813458538, "全帯域_40ms_envelope_correlation": 0.9994292068926709, "全帯域_40ms_modulation_mae_db": 0.7202820986963815, "300–1000 Hz_10ms_log_rms_mae_db": 1.064179024685801, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12172188409647557, "300–1000 Hz_10ms_envelope_correlation": 0.9899442399277641, "300–1000 Hz_10ms_modulation_mae_db": 0.8717427116648571, "300–1000 Hz_40ms_log_rms_mae_db": 0.6548214435184201, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06893787039122408, "300–1000 Hz_40ms_envelope_correlation": 0.9969958215061219, "300–1000 Hz_40ms_modulation_mae_db": 0.8044210916094124, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4081148932386038, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16295458303845833, "1000–3000 Hz_10ms_envelope_correlation": 0.9831419458830422, "1000–3000 Hz_10ms_modulation_mae_db": 1.5029543874322744, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0827376510188618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10093164148185153, "1000–3000 Hz_40ms_envelope_correlation": 0.9939208491548753, "1000–3000 Hz_40ms_modulation_mae_db": 1.4054701701943053, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5563369693629017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21787709663454377, "3000–6000 Hz_10ms_envelope_correlation": 0.9647838365853741, "3000–6000 Hz_10ms_modulation_mae_db": 1.1616495531034832, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0750230564928251, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14167146894272484, "3000–6000 Hz_40ms_envelope_correlation": 0.9853984105288195, "3000–6000 Hz_40ms_modulation_mae_db": 1.5046961747898153, "mr_stft_db": 5.422400789123362, "band_mae_db": 0.2705324289323447, "rms_gain_error_db": -0.08165329617838384, "output_peak": 0.3999293383564907}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9209066544887394, "全帯域_10ms_rms_envelope_nrmse": 0.058055595170350825, "全帯域_10ms_envelope_correlation": 0.9971831413849711, "全帯域_10ms_modulation_mae_db": 0.3442378716139154, "全帯域_40ms_log_rms_mae_db": 0.5128050722849873, "全帯域_40ms_rms_envelope_nrmse": 0.03221799325960465, "全帯域_40ms_envelope_correlation": 0.999125561725712, "全帯域_40ms_modulation_mae_db": 0.5762345323920085, "300–1000 Hz_10ms_log_rms_mae_db": 1.1950422889056345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11819926488178288, "300–1000 Hz_10ms_envelope_correlation": 0.9918183347422094, "300–1000 Hz_10ms_modulation_mae_db": 1.1064525563996277, "300–1000 Hz_40ms_log_rms_mae_db": 0.8460137429931274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08430784313403734, "300–1000 Hz_40ms_envelope_correlation": 0.9966934023617943, "300–1000 Hz_40ms_modulation_mae_db": 1.2255477030581858, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3493668816645847, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1745802764199305, "1000–3000 Hz_10ms_envelope_correlation": 0.9834104369851411, "1000–3000 Hz_10ms_modulation_mae_db": 1.8648709051149623, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0259204492853697, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12187905446211196, "1000–3000 Hz_40ms_envelope_correlation": 0.9941841233103874, "1000–3000 Hz_40ms_modulation_mae_db": 1.7716450235353876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.511332595103073, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18516768767883524, "3000–6000 Hz_10ms_envelope_correlation": 0.9763008711431078, "3000–6000 Hz_10ms_modulation_mae_db": 1.1030696692507118, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0401761118702235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1198324375966547, "3000–6000 Hz_40ms_envelope_correlation": 0.9916185269380343, "3000–6000 Hz_40ms_modulation_mae_db": 1.3180627243953873, "mr_stft_db": 5.402939179525883, "band_mae_db": 0.4160136538718258, "rms_gain_error_db": -0.07534349405157957, "output_peak": 0.3935692592518714}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.043307589467435, "全帯域_10ms_rms_envelope_nrmse": 0.05243417098947876, "全帯域_10ms_envelope_correlation": 0.9976732651108913, "全帯域_10ms_modulation_mae_db": 0.37154957040590153, "全帯域_40ms_log_rms_mae_db": 0.6297518032181272, "全帯域_40ms_rms_envelope_nrmse": 0.028029135205937465, "全帯域_40ms_envelope_correlation": 0.9993010892322362, "全帯域_40ms_modulation_mae_db": 0.4666049998966919, "300–1000 Hz_10ms_log_rms_mae_db": 1.0790952931616347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10244474379159207, "300–1000 Hz_10ms_envelope_correlation": 0.9927555353031011, "300–1000 Hz_10ms_modulation_mae_db": 1.0404750633131374, "300–1000 Hz_40ms_log_rms_mae_db": 0.6359075183471524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05905627747635589, "300–1000 Hz_40ms_envelope_correlation": 0.9975873001124608, "300–1000 Hz_40ms_modulation_mae_db": 1.3795848676454512, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2816590580973874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.140788706796058, "1000–3000 Hz_10ms_envelope_correlation": 0.9870606246919954, "1000–3000 Hz_10ms_modulation_mae_db": 0.42040797770631827, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9836660009826272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07909375911404509, "1000–3000 Hz_40ms_envelope_correlation": 0.9956886500406331, "1000–3000 Hz_40ms_modulation_mae_db": 0.16377979269007223, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5792041259862664, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.203801408607418, "3000–6000 Hz_10ms_envelope_correlation": 0.9689821756247524, "3000–6000 Hz_10ms_modulation_mae_db": 0.8048710852000917, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0863343367039682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12932865401059854, "3000–6000 Hz_40ms_envelope_correlation": 0.9874245235201348, "3000–6000 Hz_40ms_modulation_mae_db": 1.120744785884754, "mr_stft_db": 5.397249791845873, "band_mae_db": 0.19458891968088174, "rms_gain_error_db": -0.031536943337314466, "output_peak": 0.3993664684594961}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0219110726731262, "全帯域_10ms_rms_envelope_nrmse": 0.05972774116769467, "全帯域_10ms_envelope_correlation": 0.9970348033474521, "全帯域_10ms_modulation_mae_db": 0.620752553245462, "全帯域_40ms_log_rms_mae_db": 0.6468242328330799, "全帯域_40ms_rms_envelope_nrmse": 0.0360193425825779, "全帯域_40ms_envelope_correlation": 0.9989106790987606, "全帯域_40ms_modulation_mae_db": 0.6394889322065039, "300–1000 Hz_10ms_log_rms_mae_db": 1.0555504849653043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1233224279315614, "300–1000 Hz_10ms_envelope_correlation": 0.9896275233689596, "300–1000 Hz_10ms_modulation_mae_db": 1.0401925657435134, "300–1000 Hz_40ms_log_rms_mae_db": 0.6677801387628279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06659508874572347, "300–1000 Hz_40ms_envelope_correlation": 0.9972080137540968, "300–1000 Hz_40ms_modulation_mae_db": 0.9334670115809304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4028294783196698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19261727130353323, "1000–3000 Hz_10ms_envelope_correlation": 0.9759965747690857, "1000–3000 Hz_10ms_modulation_mae_db": 2.047724717003035, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1100683245255054, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11017442842160416, "1000–3000 Hz_40ms_envelope_correlation": 0.9925794283294335, "1000–3000 Hz_40ms_modulation_mae_db": 1.8631815576541648, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5874336726991138, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2109385091732281, "3000–6000 Hz_10ms_envelope_correlation": 0.9663338328504004, "3000–6000 Hz_10ms_modulation_mae_db": 1.0647646510923379, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0951814790304712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13186525414469033, "3000–6000 Hz_40ms_envelope_correlation": 0.9862536711362515, "3000–6000 Hz_40ms_modulation_mae_db": 1.5459179959759177, "mr_stft_db": 5.418492747657623, "band_mae_db": 0.2533640569750533, "rms_gain_error_db": -0.08776609850930027, "output_peak": 0.3981706764255617}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9971259274084354, "全帯域_10ms_rms_envelope_nrmse": 0.058269314689489836, "全帯域_10ms_envelope_correlation": 0.9972559030398184, "全帯域_10ms_modulation_mae_db": 0.43513169318049966, "全帯域_40ms_log_rms_mae_db": 0.5914873303497153, "全帯域_40ms_rms_envelope_nrmse": 0.03275229481883571, "全帯域_40ms_envelope_correlation": 0.9992044008744667, "全帯域_40ms_modulation_mae_db": 0.6528346928667358, "300–1000 Hz_10ms_log_rms_mae_db": 1.133415772674431, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12192242662814455, "300–1000 Hz_10ms_envelope_correlation": 0.9908828189829832, "300–1000 Hz_10ms_modulation_mae_db": 0.9660861752619049, "300–1000 Hz_40ms_log_rms_mae_db": 0.7808820572503895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08191916256717974, "300–1000 Hz_40ms_envelope_correlation": 0.9966764659858921, "300–1000 Hz_40ms_modulation_mae_db": 1.1099182340745213, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3883179281763487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18452557365303576, "1000–3000 Hz_10ms_envelope_correlation": 0.9792006383902533, "1000–3000 Hz_10ms_modulation_mae_db": 0.46739706673267417, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1017341608453581, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13032882522705677, "1000–3000 Hz_40ms_envelope_correlation": 0.9905461082431832, "1000–3000 Hz_40ms_modulation_mae_db": 0.3095295779689376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.495871102637501, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1909462196407447, "3000–6000 Hz_10ms_envelope_correlation": 0.9734539053205575, "3000–6000 Hz_10ms_modulation_mae_db": 0.8623699778059902, "3000–6000 Hz_40ms_log_rms_mae_db": 1.077322481217908, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10859039053512691, "3000–6000 Hz_40ms_envelope_correlation": 0.99239083374395, "3000–6000 Hz_40ms_modulation_mae_db": 1.036700977045066, "mr_stft_db": 5.406530707505798, "band_mae_db": 0.33313367669686167, "rms_gain_error_db": -0.12529147563705778, "output_peak": 0.412604307101403}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9359305035187868, "全帯域_10ms_rms_envelope_nrmse": 0.049896512426115265, "全帯域_10ms_envelope_correlation": 0.998098198345926, "全帯域_10ms_modulation_mae_db": 0.3687106099718762, "全帯域_40ms_log_rms_mae_db": 0.5519184127474945, "全帯域_40ms_rms_envelope_nrmse": 0.0280012566781733, "全帯域_40ms_envelope_correlation": 0.9995403976185541, "全帯域_40ms_modulation_mae_db": 0.5539782179624522, "300–1000 Hz_10ms_log_rms_mae_db": 1.1593945625936284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11950858273683596, "300–1000 Hz_10ms_envelope_correlation": 0.9908173070117313, "300–1000 Hz_10ms_modulation_mae_db": 1.263296675617976, "300–1000 Hz_40ms_log_rms_mae_db": 0.7931863290844253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07022678167543045, "300–1000 Hz_40ms_envelope_correlation": 0.9974561823468739, "300–1000 Hz_40ms_modulation_mae_db": 1.794892178736984, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3858128540946542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18149820132832797, "1000–3000 Hz_10ms_envelope_correlation": 0.9795843080000536, "1000–3000 Hz_10ms_modulation_mae_db": 1.6616380044553898, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0342295003836544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11363912239252802, "1000–3000 Hz_40ms_envelope_correlation": 0.9930563641269281, "1000–3000 Hz_40ms_modulation_mae_db": 1.435750937304083, "3000–6000 Hz_10ms_log_rms_mae_db": 1.544996467647194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1849370190435109, "3000–6000 Hz_10ms_envelope_correlation": 0.9762448698081384, "3000–6000 Hz_10ms_modulation_mae_db": 0.992708791933086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0723226053006667, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11586972803911674, "3000–6000 Hz_40ms_envelope_correlation": 0.9922680960986447, "3000–6000 Hz_40ms_modulation_mae_db": 1.282101852029909, "mr_stft_db": 5.408777762094831, "band_mae_db": 0.33055278093636314, "rms_gain_error_db": -0.14744764042583983, "output_peak": 0.4000029345678454}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44400, "output_samples": 44400, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.4774966942776935, "spectral_convergence": 0.1768164581540036, "spectral_convergence_status": "ok", "amplitude_floor": 1.585777823535548e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.297427874642165, "spectral_convergence": 0.17195528067525187, "spectral_convergence_status": "ok", "amplitude_floor": 7.694479165248034e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.411985179504509, "spectral_convergence": 0.17345260613830293, "spectral_convergence_status": "ok", "amplitude_floor": 6.411814451459315e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00186841781908955, "output_power": 0.001867631218237765, "delta_db": -0.0018287576582924105, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005255584463676, "output_power": 0.0005017171642661967, "delta_db": -0.2016206117552852, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025377475865110864, "output_power": 0.00024403546379487228, "delta_db": -0.16995479819919085, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00014708318852606447, "output_power": 0.00013416547536099192, "delta_db": -0.39922262364649413, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.522695850195022e-06, "output_power": 7.793150329226882e-06, "delta_db": -0.8704685959378854, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.071083630879648e-06, "output_power": 5.48000481354835e-06, "delta_db": -0.4448527548059555, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.34799135700051725, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07733992152274705, "pitch": null, "output_rms_dbfs": -25.589615134574043, "output_peak": 0.4120268388074869, "reference_rms_dbfs": -25.512275213051296, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.395636582808122}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44400, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 0.00014622417526559514, "log_rms_mae_db": 1.020881912389937, "rms_envelope_nrmse": 0.050730593175873424, "envelope_correlation": 0.9978664130792263, "rms_gain_error_db": -0.07733903319434626, "reference_mean_power": 0.0028155003489198547, "output_mean_power": 0.0027658057946290115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.019039396656961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008418483111337408, "output_power": 0.0008320510199701918, "reference_active": true, "delta_db": -0.05083887661731466}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001428627689371398, "output_power": 0.00014252614712612416, "reference_active": true, "delta_db": -0.010245181545720388}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9192224262112575e-05, "output_power": 1.5405965881556637e-05, "reference_active": true, "delta_db": -0.9543637801578352}], "active_band_count": 3, "active_band_mae_db": 0.3384826127736234}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 0.00013893685565863478, "log_rms_mae_db": 0.6484698728205627, "rms_envelope_nrmse": 0.024380287373597886, "envelope_correlation": 0.9995303481982089, "rms_gain_error_db": -0.07733785529065938, "reference_mean_power": 0.0028463248833832306, "output_mean_power": 0.0027960870237662867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.8422021834325858e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007582285194051116, "output_power": 0.0007500953435589488, "reference_active": true, "delta_db": -0.04683646147098115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661169958011628e-05, "output_power": 6.738276608463906e-05, "reference_active": true, "delta_db": 0.049983202334014806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7101796636835213e-06, "output_power": 1.374532197848703e-06, "reference_active": true, "delta_db": -0.9488682032102425}], "active_band_count": 3, "active_band_mae_db": 0.3485626223384128}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 8.753945313428826e-05, "log_rms_mae_db": 1.1982228097118117, "rms_envelope_nrmse": 0.10689639402443143, "envelope_correlation": 0.9925590912124655, "rms_gain_error_db": -0.29531743057858967, "reference_mean_power": 0.0005201241069420424, "output_mean_power": 0.000485931709677312}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 5.057137989146479e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002701505367125327, "output_power": 0.0002477270987258442, "reference_active": true, "delta_db": -0.37632318286335403}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.4045222297363644e-05, "output_power": 4.878106305903046e-05, "reference_active": true, "delta_db": -0.4450604708739566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.5635511670477814e-06, "output_power": 4.4595680665840864e-06, "reference_active": true, "delta_db": -2.2943295225700506}], "active_band_count": 3, "active_band_mae_db": 1.0385710587691204}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 7.440703161891069e-05, "log_rms_mae_db": 0.8483163913174819, "rms_envelope_nrmse": 0.06254269512207629, "envelope_correlation": 0.9978585789188237, "rms_gain_error_db": -0.29531757427906097, "reference_mean_power": 0.0005258188868038473, "output_mean_power": 0.0004912521046044066}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 4.6537380366659205e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025071266719842557, "output_power": 0.0002288538278754761, "reference_active": true, "delta_db": -0.3961809617941927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3806665370442115e-05, "output_power": 2.250991185706847e-05, "reference_active": true, "delta_db": -0.2432477307048161}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0797322752712393e-06, "output_power": 6.40730376564183e-07, "reference_active": true, "delta_db": -2.266407693951678}], "active_band_count": 3, "active_band_mae_db": 0.9686121288168956}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 5.3294712554339956e-05, "log_rms_mae_db": 1.3680388926258789, "rms_envelope_nrmse": 0.18736450646573286, "envelope_correlation": 0.9773132722837841, "rms_gain_error_db": -0.3684467651746617, "reference_mean_power": 0.00010733282194912887, "output_mean_power": 9.860248750250874e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 1.1016693525679884e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607161402154175e-05, "output_power": 6.129803021151129e-05, "reference_active": true, "delta_db": -0.3256839718480721}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0221078826476055e-05, "output_power": 9.038684274834819e-06, "reference_active": true, "delta_db": -0.5339152108072736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.376144897878774e-06, "output_power": 2.757734271184927e-06, "reference_active": true, "delta_db": -0.8786866121065733}], "active_band_count": 3, "active_band_mae_db": 0.579428598253973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 4.1357257962539876e-05, "log_rms_mae_db": 1.03631827164454, "rms_envelope_nrmse": 0.11567878484954107, "envelope_correlation": 0.9916136581326531, "rms_gain_error_db": -0.36844685737626587, "reference_mean_power": 0.00010850799286328448, "output_mean_power": 9.968206918243471e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.0153781665834753e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.269855261867839e-05, "output_power": 5.7756154387075865e-05, "reference_active": true, "delta_db": -0.35659246728252203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.20737764206477e-06, "output_power": 3.8145059680110687e-06, "reference_active": true, "delta_db": -0.4257319607452793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.246816673622798e-07, "output_power": 1.9077336607127365e-07, "reference_active": true, "delta_db": -0.7104989550171809}], "active_band_count": 3, "active_band_mae_db": 0.4976077943483274}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 1.102648736164134e-05, "log_rms_mae_db": 1.5357312505906928, "rms_envelope_nrmse": 0.19555369951757556, "envelope_correlation": 0.9722125185982589, "rms_gain_error_db": -0.513319424670864, "reference_mean_power": 5.352135557413054e-06, "output_mean_power": 4.7554887885005965e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.71214230899019e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.798500894794614e-06, "output_power": 1.6238441808259272e-06, "reference_active": true, "delta_db": -0.44366304905755133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2650926706890884e-07, "output_power": 3.860773283089987e-07, "reference_active": true, "delta_db": -0.4325417261291944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0854303858213565e-07, "output_power": 9.159030703931717e-08, "reference_active": true, "delta_db": -0.7375246025512829}], "active_band_count": 3, "active_band_mae_db": 0.5379097925793429}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 9.954651742580568e-06, "log_rms_mae_db": 1.0735546209926867, "rms_envelope_nrmse": 0.11495644560208836, "envelope_correlation": 0.9914274176116835, "rms_gain_error_db": -0.5133201106943281, "reference_mean_power": 5.410729753830906e-06, "output_mean_power": 4.807550246335756e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 2.394714459074953e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.814627629314412e-06, "output_power": 1.6099293871905951e-06, "reference_active": true, "delta_db": -0.5198069116894576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0097168111890166e-07, "output_power": 1.96540268436407e-07, "reference_active": true, "delta_db": -0.09683320499805059}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.411981302222167e-09, "output_power": 6.600240048687709e-09, "reference_active": true, "delta_db": -1.541213247533912}], "active_band_count": 3, "active_band_mae_db": 0.7192844547404734}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9018596542279491, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6a58312c7a4e1a652c030f9649b638a5f52ef5b58192af7ac794d68ac43620d2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.980862731249765, "standard_deviation": 0.0036032591388648018, "minimum": 0.9759965747690857, "maximum": 0.9870606246919954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3759760709978668, "standard_deviation": 0.04463249198161435, "minimum": 1.2816590580973874, "maximum": 1.4236685817658066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1783665160325705, "standard_deviation": 0.6646391832811486, "minimum": 0.42040797770631827, "maximum": 2.047724717003035, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1752194741550406, "standard_deviation": 0.01655034164170317, "minimum": 0.140788706796058, "maximum": 0.19261727130353323, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9932022697470118, "standard_deviation": 0.0016183777022753816, "minimum": 0.9905461082431832, "maximum": 0.9956886500406331, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.053934107689138, "standard_deviation": 0.0425855531427502, "minimum": 0.9836660009826272, "maximum": 1.1100683245255054, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0764684138685228, "standard_deviation": 0.665860223521264, "minimum": 0.16377979269007223, "maximum": 1.8631815576541648, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11053805463069953, "standard_deviation": 0.015314141234887113, "minimum": 0.07909375911404509, "maximum": 0.13032882522705677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.97062582746705, "standard_deviation": 0.004551318640587768, "minimum": 0.9647838365853741, "maximum": 0.9763008711431078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5449416369606275, "standard_deviation": 0.03104201153831908, "minimum": 1.495871102637501, "maximum": 1.5874336726991138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9888101643360563, "standard_deviation": 0.2556054722978154, "minimum": 0.5379097925793429, "maximum": 1.3831377937234066, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20058065454256874, "standard_deviation": 0.01329615262635548, "minimum": 0.1849370190435109, "maximum": 0.21787709663454377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9889976838006038, "standard_deviation": 0.003215213102368014, "minimum": 0.9851999908273117, "maximum": 0.99239083374395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0743570380471514, "standard_deviation": 0.015855646541748265, "minimum": 1.0401761118702235, "maximum": 1.0951814790304712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2992287559949733, "standard_deviation": 0.3510040603830322, "minimum": 0.7192844547404734, "maximum": 1.8663210830984642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12629865337250873, "standard_deviation": 0.013898855151071908, "minimum": 0.10859039053512691, "maximum": 0.14827484810906946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991324427818274, "standard_deviation": 0.001183409439889973, "minimum": 0.9896275233689596, "maximum": 0.9927555353031011, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1233348926537778, "standard_deviation": 0.05690596273618914, "minimum": 1.0555504849653043, "maximum": 1.1982228097118117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0967522165599797, "standard_deviation": 0.18051493884449504, "minimum": 0.8717427116648571, "maximum": 1.4472009257096994, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1160324916227866, "standard_deviation": 0.007637079975263625, "minimum": 0.10244474379159207, "maximum": 0.1233224279315614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972643498517393, "standard_deviation": 0.0004442429431078204, "minimum": 0.9966764659858921, "maximum": 0.9978585789188237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7555804432818893, "standard_deviation": 0.0885427212833456, "minimum": 0.6359075183471524, "maximum": 0.8483163913174819, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1816763397206027, "standard_deviation": 0.31045087863506254, "minimum": 0.8044210916094124, "maximum": 1.794892178736984, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07091858753229863, "standard_deviation": 0.008802108860328098, "minimum": 0.05905627747635589, "maximum": 0.08430784313403734, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.31448292251568455, "standard_deviation": 0.07083841243327874, "minimum": 0.19458891968088174, "maximum": 0.4160136538718258, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.40727121264786, "standard_deviation": 0.009361622054308962, "minimum": 5.395636582808122, "maximum": 5.422400789123362, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.402950136253931, "standard_deviation": 0.006988758220714531, "minimum": 0.3935692592518714, "maximum": 0.412604307101403, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09963262260816369, "standard_deviation": 0.04500904812749328, "minimum": -0.17068211120308682, "maximum": -0.031536943337314466, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9975007975826791, "standard_deviation": 0.00035994341024805546, "minimum": 0.9970348033474521, "maximum": 0.998098198345926, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9988239589105018, "standard_deviation": 0.05284613814792609, "minimum": 0.9209066544887394, "maximum": 1.0758748255758286, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4275227118398038, "standard_deviation": 0.1048438599814622, "minimum": 0.3384826127736234, "maximum": 0.620752553245462, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.055288661160594776, "standard_deviation": 0.0038309462243483606, "minimum": 0.049896512426115265, "maximum": 0.05972774116769467, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992634741112675, "standard_deviation": 0.00022763295713556887, "minimum": 0.9989106790987606, "maximum": 0.9995403976185541, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6001059970839295, "standard_deviation": 0.05186287715959779, "minimum": 0.5128050722849873, "maximum": 0.6499342631151682, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5514639672293463, "standard_deviation": 0.12251335948096852, "minimum": 0.3485626223384128, "maximum": 0.7202820986963815, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.030692646797994695, "standard_deviation": 0.004627793497775389, "minimum": 0.024380287373597886, "maximum": 0.037408617651772114, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-85628-0025", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.0017619132995605, "predicted_mos": 3.0017619132995605, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "42013615992ffe31ac4e71850de584dc61f04fa5ebe98e19cc70546fe437fdf4", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8578059999272227, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.885563441212085, "全帯域_10ms_rms_envelope_nrmse": 0.05598126670171148, "全帯域_10ms_envelope_correlation": 0.9974661064176399, "全帯域_10ms_modulation_mae_db": 0.39580847978318895, "全帯域_40ms_log_rms_mae_db": 0.5599435874126664, "全帯域_40ms_rms_envelope_nrmse": 0.03338639899476875, "全帯域_40ms_envelope_correlation": 0.9991489391470761, "全帯域_40ms_modulation_mae_db": 0.383693927276927, "300–1000 Hz_10ms_log_rms_mae_db": 1.1993872429809873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12175826692103785, "300–1000 Hz_10ms_envelope_correlation": 0.9897111842606956, "300–1000 Hz_10ms_modulation_mae_db": 0.5115938835237012, "300–1000 Hz_40ms_log_rms_mae_db": 0.7717695030799602, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07162042731288679, "300–1000 Hz_40ms_envelope_correlation": 0.9964135588854115, "300–1000 Hz_40ms_modulation_mae_db": 0.31186658870283795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4457940588797882, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17217325075871537, "1000–3000 Hz_10ms_envelope_correlation": 0.9822552098979531, "1000–3000 Hz_10ms_modulation_mae_db": 1.7417917973158061, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0847590012793196, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11387401522859032, "1000–3000 Hz_40ms_envelope_correlation": 0.9934653695584859, "1000–3000 Hz_40ms_modulation_mae_db": 1.574531939309413, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7571572695325886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20399077810118285, "3000–6000 Hz_10ms_envelope_correlation": 0.9693320947153307, "3000–6000 Hz_10ms_modulation_mae_db": 1.2858882745676934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2760365151624946, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11991874331936317, "3000–6000 Hz_40ms_envelope_correlation": 0.9899552711545216, "3000–6000 Hz_40ms_modulation_mae_db": 1.932587392550639, "mr_stft_db": 5.583456464082359, "band_mae_db": 0.27291656080885, "rms_gain_error_db": -0.119075239144832, "output_peak": 0.4032771281173782}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.029553304082854, "全帯域_10ms_rms_envelope_nrmse": 0.058944627728168504, "全帯域_10ms_envelope_correlation": 0.9970734425903354, "全帯域_10ms_modulation_mae_db": 0.5854527762773937, "全帯域_40ms_log_rms_mae_db": 0.6284568700294182, "全帯域_40ms_rms_envelope_nrmse": 0.031765455964090646, "全帯域_40ms_envelope_correlation": 0.9991246390507023, "全帯域_40ms_modulation_mae_db": 0.6607549046687574, "300–1000 Hz_10ms_log_rms_mae_db": 1.2436974629263895, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09530325038802208, "300–1000 Hz_10ms_envelope_correlation": 0.9939031442625619, "300–1000 Hz_10ms_modulation_mae_db": 0.37106328980170816, "300–1000 Hz_40ms_log_rms_mae_db": 0.80236830599669, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054953873873347714, "300–1000 Hz_40ms_envelope_correlation": 0.9981285450600906, "300–1000 Hz_40ms_modulation_mae_db": 0.27941371728991515, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4376037925925746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19229044969626755, "1000–3000 Hz_10ms_envelope_correlation": 0.9760343037361457, "1000–3000 Hz_10ms_modulation_mae_db": 1.308289952717973, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0980136270439842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1269627932405254, "1000–3000 Hz_40ms_envelope_correlation": 0.9889625967288367, "1000–3000 Hz_40ms_modulation_mae_db": 1.6574330565331963, "3000–6000 Hz_10ms_log_rms_mae_db": 1.744055134771831, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1985434241217695, "3000–6000 Hz_10ms_envelope_correlation": 0.9703744743438143, "3000–6000 Hz_10ms_modulation_mae_db": 0.918984166114992, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2582513570979734, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11350387644514547, "3000–6000 Hz_40ms_envelope_correlation": 0.9900674948707577, "3000–6000 Hz_40ms_modulation_mae_db": 1.5603187642611316, "mr_stft_db": 5.5471890639351935, "band_mae_db": 0.2630035896261528, "rms_gain_error_db": -0.05992383409001558, "output_peak": 0.40425759049235693}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9141386010616717, "全帯域_10ms_rms_envelope_nrmse": 0.055176012150881054, "全帯域_10ms_envelope_correlation": 0.9975117896351376, "全帯域_10ms_modulation_mae_db": 0.515861925644293, "全帯域_40ms_log_rms_mae_db": 0.524444040589251, "全帯域_40ms_rms_envelope_nrmse": 0.02897466148855552, "全帯域_40ms_envelope_correlation": 0.999365365747066, "全帯域_40ms_modulation_mae_db": 0.5779909507066089, "300–1000 Hz_10ms_log_rms_mae_db": 1.132072714080808, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12415424823913285, "300–1000 Hz_10ms_envelope_correlation": 0.9903739659061445, "300–1000 Hz_10ms_modulation_mae_db": 1.193482015032583, "300–1000 Hz_40ms_log_rms_mae_db": 0.7971678285092446, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07702069246050083, "300–1000 Hz_40ms_envelope_correlation": 0.9971678064486688, "300–1000 Hz_40ms_modulation_mae_db": 1.147504767958926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.420350677154468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1701370872025253, "1000–3000 Hz_10ms_envelope_correlation": 0.9814691081624894, "1000–3000 Hz_10ms_modulation_mae_db": 2.13479873096291, "1000–3000 Hz_40ms_log_rms_mae_db": 1.06210198648883, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10520998716731307, "1000–3000 Hz_40ms_envelope_correlation": 0.9932032042430029, "1000–3000 Hz_40ms_modulation_mae_db": 1.809762480180189, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8098671534375435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2077585018472317, "3000–6000 Hz_10ms_envelope_correlation": 0.9673733096520346, "3000–6000 Hz_10ms_modulation_mae_db": 1.0381101467097213, "3000–6000 Hz_40ms_log_rms_mae_db": 1.273958486600542, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10853445204835485, "3000–6000 Hz_40ms_envelope_correlation": 0.9909313861475207, "3000–6000 Hz_40ms_modulation_mae_db": 1.5212176438513563, "mr_stft_db": 5.559314208385872, "band_mae_db": 0.3838210901844883, "rms_gain_error_db": -0.10501783138844781, "output_peak": 0.3937602848034653}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9772746183055262, "全帯域_10ms_rms_envelope_nrmse": 0.058574671847716926, "全帯域_10ms_envelope_correlation": 0.9971910029981091, "全帯域_10ms_modulation_mae_db": 0.3668133550372996, "全帯域_40ms_log_rms_mae_db": 0.6050204536701569, "全帯域_40ms_rms_envelope_nrmse": 0.03252599637896042, "全帯域_40ms_envelope_correlation": 0.9991802634157244, "全帯域_40ms_modulation_mae_db": 0.582727805262332, "300–1000 Hz_10ms_log_rms_mae_db": 1.1966239940187877, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12327665857081006, "300–1000 Hz_10ms_envelope_correlation": 0.9897488742301255, "300–1000 Hz_10ms_modulation_mae_db": 0.6958444909888032, "300–1000 Hz_40ms_log_rms_mae_db": 0.8457144510856839, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07819672181988756, "300–1000 Hz_40ms_envelope_correlation": 0.9960701808793425, "300–1000 Hz_40ms_modulation_mae_db": 0.9461610011910287, "1000–3000 Hz_10ms_log_rms_mae_db": 1.426253290418122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16069916308673093, "1000–3000 Hz_10ms_envelope_correlation": 0.9854155374877108, "1000–3000 Hz_10ms_modulation_mae_db": 0.7886105313199544, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1258973894266877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11150859055538534, "1000–3000 Hz_40ms_envelope_correlation": 0.9945013109718842, "1000–3000 Hz_40ms_modulation_mae_db": 0.8787428141206495, "3000–6000 Hz_10ms_log_rms_mae_db": 1.856639074395386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22570156287805118, "3000–6000 Hz_10ms_envelope_correlation": 0.9612769055391175, "3000–6000 Hz_10ms_modulation_mae_db": 1.1499490118493234, "3000–6000 Hz_40ms_log_rms_mae_db": 1.397766406844425, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13928417160938494, "3000–6000 Hz_40ms_envelope_correlation": 0.9840358377961161, "3000–6000 Hz_40ms_modulation_mae_db": 1.6374120693202021, "mr_stft_db": 5.554728050581471, "band_mae_db": 0.2995227065247939, "rms_gain_error_db": -0.10817060039429602, "output_peak": 0.41390248266894614}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9944294595767849, "全帯域_10ms_rms_envelope_nrmse": 0.05748784680254883, "全帯域_10ms_envelope_correlation": 0.9974424913839409, "全帯域_10ms_modulation_mae_db": 0.18850868631059034, "全帯域_40ms_log_rms_mae_db": 0.6069585027249368, "全帯域_40ms_rms_envelope_nrmse": 0.036846587609812964, "全帯域_40ms_envelope_correlation": 0.9990681590036189, "全帯域_40ms_modulation_mae_db": 0.3292985966543896, "300–1000 Hz_10ms_log_rms_mae_db": 1.211411577362962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13883006720660937, "300–1000 Hz_10ms_envelope_correlation": 0.9875225288787978, "300–1000 Hz_10ms_modulation_mae_db": 1.3516807504141297, "300–1000 Hz_40ms_log_rms_mae_db": 0.8122501174440528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08509740062654148, "300–1000 Hz_40ms_envelope_correlation": 0.9961227895867394, "300–1000 Hz_40ms_modulation_mae_db": 1.4488175011868678, "1000–3000 Hz_10ms_log_rms_mae_db": 1.461752645571248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1848734599375357, "1000–3000 Hz_10ms_envelope_correlation": 0.978546414067886, "1000–3000 Hz_10ms_modulation_mae_db": 1.7991978612885997, "1000–3000 Hz_40ms_log_rms_mae_db": 1.152378845179751, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11467086337296391, "1000–3000 Hz_40ms_envelope_correlation": 0.9926425484573096, "1000–3000 Hz_40ms_modulation_mae_db": 1.6500912343381824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7799278720779657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2058238783528345, "3000–6000 Hz_10ms_envelope_correlation": 0.9679429176626432, "3000–6000 Hz_10ms_modulation_mae_db": 1.0254858252725845, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2854587692657773, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11823672458788943, "3000–6000 Hz_40ms_envelope_correlation": 0.9889602148033728, "3000–6000 Hz_40ms_modulation_mae_db": 1.5368921237085598, "mr_stft_db": 5.543574337516964, "band_mae_db": 0.3858313078952967, "rms_gain_error_db": -0.16053714023340038, "output_peak": 0.4013502441197866}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9685774166981816, "全帯域_10ms_rms_envelope_nrmse": 0.0482153293807484, "全帯域_10ms_envelope_correlation": 0.9981815979812889, "全帯域_10ms_modulation_mae_db": 0.5078915420348946, "全帯域_40ms_log_rms_mae_db": 0.6019561559945685, "全帯域_40ms_rms_envelope_nrmse": 0.027490208166801512, "全帯域_40ms_envelope_correlation": 0.9995012726952186, "全帯域_40ms_modulation_mae_db": 0.6043338500267925, "300–1000 Hz_10ms_log_rms_mae_db": 1.1608781107911192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12474097601910175, "300–1000 Hz_10ms_envelope_correlation": 0.9901298237456824, "300–1000 Hz_10ms_modulation_mae_db": 1.0668232431799052, "300–1000 Hz_40ms_log_rms_mae_db": 0.8024864421626268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08242634957812864, "300–1000 Hz_40ms_envelope_correlation": 0.9963150748449224, "300–1000 Hz_40ms_modulation_mae_db": 1.2484555862703508, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4466242418536897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17375940421784764, "1000–3000 Hz_10ms_envelope_correlation": 0.9812954490100695, "1000–3000 Hz_10ms_modulation_mae_db": 1.4167017109357136, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0538856568567203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10705703860236497, "1000–3000 Hz_40ms_envelope_correlation": 0.9938702730559086, "1000–3000 Hz_40ms_modulation_mae_db": 1.3903304530149558, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8059009813985059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2042441534936612, "3000–6000 Hz_10ms_envelope_correlation": 0.9692927642170031, "3000–6000 Hz_10ms_modulation_mae_db": 1.259362272429619, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2349301827097543, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11864764216978284, "3000–6000 Hz_40ms_envelope_correlation": 0.9904217976316411, "3000–6000 Hz_40ms_modulation_mae_db": 1.905747571281136, "mr_stft_db": 5.536390605560282, "band_mae_db": 0.280643958701404, "rms_gain_error_db": -0.12714551328935547, "output_peak": 0.4077818540243804}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0011715546606503, "全帯域_10ms_rms_envelope_nrmse": 0.05649116320526654, "全帯域_10ms_envelope_correlation": 0.9973261829136615, "全帯域_10ms_modulation_mae_db": 0.30083510627226223, "全帯域_40ms_log_rms_mae_db": 0.5923848106026275, "全帯域_40ms_rms_envelope_nrmse": 0.030148909796901314, "全帯域_40ms_envelope_correlation": 0.999229036084128, "全帯域_40ms_modulation_mae_db": 0.4397172977453951, "300–1000 Hz_10ms_log_rms_mae_db": 1.2584787550239287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11823838294021194, "300–1000 Hz_10ms_envelope_correlation": 0.9907274306028809, "300–1000 Hz_10ms_modulation_mae_db": 0.8225593824019244, "300–1000 Hz_40ms_log_rms_mae_db": 0.8473813120448194, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07730640263961022, "300–1000 Hz_40ms_envelope_correlation": 0.9962949428844611, "300–1000 Hz_40ms_modulation_mae_db": 1.1512752198981822, "1000–3000 Hz_10ms_log_rms_mae_db": 1.482056653257018, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17873089941851797, "1000–3000 Hz_10ms_envelope_correlation": 0.9796381024185784, "1000–3000 Hz_10ms_modulation_mae_db": 1.0342303626485836, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0938776421618066, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11241848238580704, "1000–3000 Hz_40ms_envelope_correlation": 0.9924125955668185, "1000–3000 Hz_40ms_modulation_mae_db": 1.1800554365246028, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8339261086957663, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21364875560194796, "3000–6000 Hz_10ms_envelope_correlation": 0.9654025620844756, "3000–6000 Hz_10ms_modulation_mae_db": 1.1891834601408007, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3473176083298963, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11643996045304301, "3000–6000 Hz_40ms_envelope_correlation": 0.9889678483201462, "3000–6000 Hz_40ms_modulation_mae_db": 1.7570978524269654, "mr_stft_db": 5.543526092422602, "band_mae_db": 0.2504108011167859, "rms_gain_error_db": -0.06607695044662959, "output_peak": 0.41310191846712935}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9776620729130732, "全帯域_10ms_rms_envelope_nrmse": 0.05352663400137045, "全帯域_10ms_envelope_correlation": 0.9975734217482548, "全帯域_10ms_modulation_mae_db": 0.17621216183095187, "全帯域_40ms_log_rms_mae_db": 0.5775478977245837, "全帯域_40ms_rms_envelope_nrmse": 0.027888503682619636, "全帯域_40ms_envelope_correlation": 0.9993062895488193, "全帯域_40ms_modulation_mae_db": 0.06940874436023335, "300–1000 Hz_10ms_log_rms_mae_db": 1.1959051907221903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11211008949144603, "300–1000 Hz_10ms_envelope_correlation": 0.991602011183278, "300–1000 Hz_10ms_modulation_mae_db": 0.8032809530578514, "300–1000 Hz_40ms_log_rms_mae_db": 0.7874011685486915, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05876061224137344, "300–1000 Hz_40ms_envelope_correlation": 0.9980377206034734, "300–1000 Hz_40ms_modulation_mae_db": 1.4322946554225062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4464629411128822, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16134060294561442, "1000–3000 Hz_10ms_envelope_correlation": 0.9829590821333426, "1000–3000 Hz_10ms_modulation_mae_db": 1.0506879737966588, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0783409090161165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10798543461857153, "1000–3000 Hz_40ms_envelope_correlation": 0.9920410065986829, "1000–3000 Hz_40ms_modulation_mae_db": 0.537498895153525, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7623316570808667, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1982590974909205, "3000–6000 Hz_10ms_envelope_correlation": 0.9706762446996834, "3000–6000 Hz_10ms_modulation_mae_db": 1.0243501138208086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2845705677303463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12039667068317647, "3000–6000 Hz_40ms_envelope_correlation": 0.9890220249885775, "3000–6000 Hz_40ms_modulation_mae_db": 1.4838393528032505, "mr_stft_db": 5.556354432245212, "band_mae_db": 0.16314512458286565, "rms_gain_error_db": -0.03167705822844751, "output_peak": 0.41782998818896866}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44400, "output_samples": 44400, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.727404506738652, "spectral_convergence": 0.17460434660595814, "spectral_convergence_status": "ok", "amplitude_floor": 1.585777823535548e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.49957340527267, "spectral_convergence": 0.16875454801886744, "spectral_convergence_status": "ok", "amplitude_floor": 7.694479165248034e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.523391480235755, "spectral_convergence": 0.17045220958529966, "spectral_convergence_status": "ok", "amplitude_floor": 6.411814451459315e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00186841781908955, "output_power": 0.0018253567751846124, "delta_db": -0.10126238364117907, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005255584463676, "output_power": 0.0005084386319537655, "delta_db": -0.1438247862129992, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025377475865110864, "output_power": 0.000260062505824243, "delta_db": 0.10629319445381065, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00014708318852606447, "output_power": 0.00012620092817021238, "delta_db": -0.6650048707574939, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.522695850195022e-06, "output_power": 8.379380706127644e-06, "delta_db": -0.555479909517106, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.071083630879648e-06, "output_power": 5.980022160979629e-06, "delta_db": -0.0656342202705114, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27291656080885, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.119075239144832, "pitch": null, "output_rms_dbfs": -25.63135045219613, "output_peak": 0.4032771281173782, "reference_rms_dbfs": -25.512275213051296, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.583456464082359}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44400, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 0.00014622417526559514, "log_rms_mae_db": 0.885563441212085, "rms_envelope_nrmse": 0.05598126670171148, "envelope_correlation": 0.9974661064176399, "rms_gain_error_db": -0.119074400855268, "reference_mean_power": 0.0028155003489198547, "output_mean_power": 0.0027393539162435094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.019039396656961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008418483111337408, "output_power": 0.0008235163237403198, "reference_active": true, "delta_db": -0.09561632800516791}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001428627689371398, "output_power": 0.00013854855613470803, "reference_active": true, "delta_db": -0.13317059077605528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9192224262112575e-05, "output_power": 1.539080932139639e-05, "reference_active": true, "delta_db": -0.9586385205683436}], "active_band_count": 3, "active_band_mae_db": 0.39580847978318895}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 0.00013893685565863478, "log_rms_mae_db": 0.5599435874126664, "rms_envelope_nrmse": 0.03338639899476875, "envelope_correlation": 0.9991489391470761, "rms_gain_error_db": -0.11907385699178077, "reference_mean_power": 0.0028463248833832306, "output_mean_power": 0.002769345134555576}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.8422021834325858e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007582285194051116, "output_power": 0.0007415216681445658, "reference_active": true, "delta_db": -0.09676269454411313}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661169958011628e-05, "output_power": 6.600629651864086e-05, "reference_active": true, "delta_db": -0.03965148668611192}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7101796636835213e-06, "output_power": 1.3538638050632348e-06, "reference_active": true, "delta_db": -1.014667600600556}], "active_band_count": 3, "active_band_mae_db": 0.383693927276927}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 8.753945313428826e-05, "log_rms_mae_db": 1.1993872429809873, "rms_envelope_nrmse": 0.12175826692103785, "envelope_correlation": 0.9897111842606956, "rms_gain_error_db": -0.1689567364675193, "reference_mean_power": 0.0005201241069420424, "output_mean_power": 0.0005002778899932121}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 5.057137989146479e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002701505367125327, "output_power": 0.0002576431100685403, "reference_active": true, "delta_db": -0.20587301713175665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.4045222297363644e-05, "output_power": 5.121016993374301e-05, "reference_active": true, "delta_db": -0.2340109052846244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.5635511670477814e-06, "output_power": 5.878086549933685e-06, "reference_active": true, "delta_db": -1.0948977281547223}], "active_band_count": 3, "active_band_mae_db": 0.5115938835237012}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 7.440703161891069e-05, "log_rms_mae_db": 0.7717695030799602, "rms_envelope_nrmse": 0.07162042731288679, "envelope_correlation": 0.9964135588854115, "rms_gain_error_db": -0.168956952883342, "reference_mean_power": 0.0005258188868038473, "output_mean_power": 0.0005057553506737296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 4.6537380366659205e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025071266719842557, "output_power": 0.0002391948863842002, "reference_active": true, "delta_db": -0.2042438626594178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3806665370442115e-05, "output_power": 2.2465124152051008e-05, "reference_active": true, "delta_db": -0.25189744454697494}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0797322752712393e-06, "output_power": 9.668747919045022e-07, "reference_active": true, "delta_db": -0.4794584589021212}], "active_band_count": 3, "active_band_mae_db": 0.31186658870283795}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 5.3294712554339956e-05, "log_rms_mae_db": 1.4457940588797882, "rms_envelope_nrmse": 0.17217325075871537, "envelope_correlation": 0.9822552098979531, "rms_gain_error_db": -0.5603567710542081, "reference_mean_power": 0.00010733282194912887, "output_mean_power": 9.434021696158462e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 1.1016693525679884e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607161402154175e-05, "output_power": 5.811558351491445e-05, "reference_active": true, "delta_db": -0.5572231330654153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0221078826476055e-05, "output_power": 7.8383134340776e-06, "reference_active": true, "delta_db": -1.1527411183065723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.376144897878774e-06, "output_power": 1.5027265633837889e-06, "reference_active": true, "delta_db": -3.5154111405754302}], "active_band_count": 3, "active_band_mae_db": 1.7417917973158061}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 4.1357257962539876e-05, "log_rms_mae_db": 1.0847590012793196, "rms_envelope_nrmse": 0.11387401522859032, "envelope_correlation": 0.9934653695584859, "rms_gain_error_db": -0.5603572413850167, "reference_mean_power": 0.00010850799286328448, "output_mean_power": 9.537312347040382e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.0153781665834753e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.269855261867839e-05, "output_power": 5.395837042835319e-05, "reference_active": true, "delta_db": -0.6519869008831715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.20737764206477e-06, "output_power": 3.690641898699129e-06, "reference_active": true, "delta_db": -0.569095869404903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.246816673622798e-07, "output_power": 1.0030355675832083e-07, "reference_active": true, "delta_db": -3.502513047640165}], "active_band_count": 3, "active_band_mae_db": 1.574531939309413}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 1.102648736164134e-05, "log_rms_mae_db": 1.7571572695325886, "rms_envelope_nrmse": 0.20399077810118285, "envelope_correlation": 0.9693320947153307, "rms_gain_error_db": -0.4156706750239815, "reference_mean_power": 5.352135557413054e-06, "output_mean_power": 4.8636244952923e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.71214230899019e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.798500894794614e-06, "output_power": 1.5992781504369316e-06, "reference_active": true, "delta_db": -0.509866543329969}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2650926706890884e-07, "output_power": 2.7919479152101226e-07, "reference_active": true, "delta_db": -1.8402115972249637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0854303858213565e-07, "output_power": 7.67084675808199e-08, "reference_active": true, "delta_db": -1.5075866831481473}], "active_band_count": 3, "active_band_mae_db": 1.2858882745676934}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 9.954651742580568e-06, "log_rms_mae_db": 1.2760365151624946, "rms_envelope_nrmse": 0.11991874331936317, "envelope_correlation": 0.9899552711545216, "rms_gain_error_db": -0.41567335978926034, "reference_mean_power": 5.410729753830906e-06, "output_mean_power": 4.916867522825587e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 2.394714459074953e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.814627629314412e-06, "output_power": 1.5562551248754023e-06, "reference_active": true, "delta_db": -0.6670672450069832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0097168111890166e-07, "output_power": 1.3522571934694057e-07, "reference_active": true, "delta_db": -1.7207556498556609}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.411981302222167e-09, "output_power": 4.29227093439384e-09, "reference_active": true, "delta_db": -3.4099392827892725}], "active_band_count": 3, "active_band_mae_db": 1.932587392550639}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8875360857667918, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a8bd78956f16c1c4f6cb5916847e2895bd9cc80790246d719a5574709c5f2e57", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9809516508642719, "standard_deviation": 0.002871837784071555, "minimum": 0.9760343037361457, "maximum": 0.9854155374877108, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4458622876049738, "standard_deviation": 0.019519365379192723, "minimum": 1.420350677154468, "maximum": 1.482056653257018, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.409288615123275, "standard_deviation": 0.45571178153665576, "minimum": 0.7886105313199544, "maximum": 2.13479873096291, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17425053965796936, "standard_deviation": 0.010889979143539535, "minimum": 0.16069916308673093, "maximum": 0.19229044969626755, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9926373631476162, "standard_deviation": 0.0016865110144948952, "minimum": 0.9889625967288367, "maximum": 0.9945013109718842, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.093656882181652, "standard_deviation": 0.0325092565523871, "minimum": 1.0538856568567203, "maximum": 1.152378845179751, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3348057886468392, "standard_deviation": 0.44033560762108126, "minimum": 0.537498895153525, "maximum": 1.809762480180189, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1124609006464402, "standard_deviation": 0.006760195243835612, "minimum": 0.10520998716731307, "maximum": 0.1269627932405254, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9677089091142628, "standard_deviation": 0.0031147332048952867, "minimum": 0.9612769055391175, "maximum": 0.9706762446996834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7937256564238067, "standard_deviation": 0.03957511966668093, "minimum": 1.744055134771831, "maximum": 1.856639074395386, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.111414158863193, "standard_deviation": 0.12941045841721255, "minimum": 0.918984166114992, "maximum": 1.2858882745676934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20724626898594992, "standard_deviation": 0.008944231689994538, "minimum": 0.1982590974909205, "maximum": 0.22570156287805118, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9890452344640817, "standard_deviation": 0.0021525230398296794, "minimum": 0.9840358377961161, "maximum": 0.9909313861475207, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2947862367176513, "standard_deviation": 0.0524523101061258, "minimum": 1.2349301827097543, "maximum": 1.397766406844425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.666889096275405, "standard_deviation": 0.17713566217889365, "minimum": 1.4838393528032505, "maximum": 1.932587392550639, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11937028016451752, "standard_deviation": 0.008945399079948933, "minimum": 0.10853445204835485, "maximum": 0.13928417160938494, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904648703837708, "standard_deviation": 0.0018143216363905761, "minimum": 0.9875225288787978, "maximum": 0.9939031442625619, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1998068809883966, "standard_deviation": 0.040764721323069006, "minimum": 1.132072714080808, "maximum": 1.2584787550239287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8520410010500757, "standard_deviation": 0.3353550598420532, "minimum": 0.37106328980170816, "maximum": 1.3516807504141297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1198014924720465, "standard_deviation": 0.012439006844786774, "minimum": 0.09530325038802208, "maximum": 0.13883006720660937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968188273991386, "standard_deviation": 0.0008502312129821136, "minimum": 0.9960701808793425, "maximum": 0.9981285450600906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.808317391108971, "standard_deviation": 0.02647623506153674, "minimum": 0.7717695030799602, "maximum": 0.8473813120448194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9957236297400769, "standard_deviation": 0.46139731363058634, "minimum": 0.27941371728991515, "maximum": 1.4488175011868678, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07317281006903457, "standard_deviation": 0.01086940203163196, "minimum": 0.054953873873347714, "maximum": 0.08509740062654148, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.28741189243007964, "standard_deviation": 0.07249707553835912, "minimum": 0.16314512458286565, "maximum": 0.3858313078952967, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.553066656841245, "standard_deviation": 0.014490909766356532, "minimum": 5.536390605560282, "maximum": 5.583456464082359, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4069076863603015, "standard_deviation": 0.007849475199491508, "minimum": 0.3937602848034653, "maximum": 0.41782998818896866, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09720302090192805, "standard_deviation": 0.04178129850136852, "minimum": -0.16053714023340038, "maximum": -0.03167705822844751, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9974707544585459, "standard_deviation": 0.0003328152446323605, "minimum": 0.9970734425903354, "maximum": 0.9981815979812889, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9685463085638533, "standard_deviation": 0.046998228111565774, "minimum": 0.885563441212085, "maximum": 1.029553304082854, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3796730041488593, "standard_deviation": 0.15208737125889932, "minimum": 0.17621216183095187, "maximum": 0.5854527762773937, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05554969397730152, "standard_deviation": 0.003454405795414138, "minimum": 0.0482153293807484, "maximum": 0.058944627728168504, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9992404955865442, "standard_deviation": 0.00014302375116752257, "minimum": 0.9990681590036189, "maximum": 0.9995012726952186, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5870890398435261, "standard_deviation": 0.03257740183901815, "minimum": 0.524444040589251, "maximum": 0.6284568700294182, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4559907595876795, "standard_deviation": 0.19488772193963225, "minimum": 0.06940874436023335, "maximum": 0.6607549046687574, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.031128340260313844, "standard_deviation": 0.003150662141493519, "minimum": 0.027490208166801512, "maximum": 0.036846587609812964, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7021-85628-0025", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.216762065887451, "predicted_mos": 3.216762065887451, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "c3a220f3d2a4b250a858d03415cc7664c663b9749370f6545e46d716b569e14e", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.065560299903154, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8387393222398823, "全帯域_10ms_rms_envelope_nrmse": 0.05821004908634618, "全帯域_10ms_envelope_correlation": 0.9971700406419763, "全帯域_10ms_modulation_mae_db": 0.23443646940319576, "全帯域_40ms_log_rms_mae_db": 0.4758507247300194, "全帯域_40ms_rms_envelope_nrmse": 0.03221406791925102, "全帯域_40ms_envelope_correlation": 0.9991277944659482, "全帯域_40ms_modulation_mae_db": 0.4296355436307871, "300–1000 Hz_10ms_log_rms_mae_db": 1.1223729330365237, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11589102279902369, "300–1000 Hz_10ms_envelope_correlation": 0.9910965921732598, "300–1000 Hz_10ms_modulation_mae_db": 0.8720889215252047, "300–1000 Hz_40ms_log_rms_mae_db": 0.7263350688826837, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06936193429884067, "300–1000 Hz_40ms_envelope_correlation": 0.997158129814313, "300–1000 Hz_40ms_modulation_mae_db": 1.0407780882657127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.42259871327577, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18103403130954704, "1000–3000 Hz_10ms_envelope_correlation": 0.9802928332364008, "1000–3000 Hz_10ms_modulation_mae_db": 1.7603508119816142, "1000–3000 Hz_40ms_log_rms_mae_db": 1.132879079897875, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12721871371465193, "1000–3000 Hz_40ms_envelope_correlation": 0.9913893070196539, "1000–3000 Hz_40ms_modulation_mae_db": 1.737924504089328, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7217085203552884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19018600350553297, "3000–6000 Hz_10ms_envelope_correlation": 0.9729594495512256, "3000–6000 Hz_10ms_modulation_mae_db": 0.9677460710053657, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2174936144228947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10596253530830888, "3000–6000 Hz_40ms_envelope_correlation": 0.9914573382303448, "3000–6000 Hz_40ms_modulation_mae_db": 1.5284886666730808, "mr_stft_db": 5.513619589539952, "band_mae_db": 0.21792971023938768, "rms_gain_error_db": -0.07813887406594977, "output_peak": 0.41209785267548554}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9036370431030137, "全帯域_10ms_rms_envelope_nrmse": 0.05032793737190958, "全帯域_10ms_envelope_correlation": 0.997956468109549, "全帯域_10ms_modulation_mae_db": 0.5441342320288963, "全帯域_40ms_log_rms_mae_db": 0.5519869546816565, "全帯域_40ms_rms_envelope_nrmse": 0.02824771499538316, "全帯域_40ms_envelope_correlation": 0.9994065558721367, "全帯域_40ms_modulation_mae_db": 0.5493526463387453, "300–1000 Hz_10ms_log_rms_mae_db": 1.1797027469251473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09954440844676339, "300–1000 Hz_10ms_envelope_correlation": 0.9932709908518763, "300–1000 Hz_10ms_modulation_mae_db": 0.8855877359154977, "300–1000 Hz_40ms_log_rms_mae_db": 0.7734541940739763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050165809773105634, "300–1000 Hz_40ms_envelope_correlation": 0.9984502942725487, "300–1000 Hz_40ms_modulation_mae_db": 1.0865257276455635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4876689344883751, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1674678915693971, "1000–3000 Hz_10ms_envelope_correlation": 0.9838194892409098, "1000–3000 Hz_10ms_modulation_mae_db": 1.4144967578820875, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1334976925721567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10906471003797993, "1000–3000 Hz_40ms_envelope_correlation": 0.9948361250238853, "1000–3000 Hz_40ms_modulation_mae_db": 1.5026879812947407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7692105992340852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2138106234812663, "3000–6000 Hz_10ms_envelope_correlation": 0.9671682928904068, "3000–6000 Hz_10ms_modulation_mae_db": 1.782046429080385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.221390451584936, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12604853673397742, "3000–6000 Hz_40ms_envelope_correlation": 0.9903707622395944, "3000–6000 Hz_40ms_modulation_mae_db": 2.0262563634124366, "mr_stft_db": 5.51512662795836, "band_mae_db": 0.306956737993763, "rms_gain_error_db": -0.10628230409962969, "output_peak": 0.39296191212278136}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9652057329692141, "全帯域_10ms_rms_envelope_nrmse": 0.054382880781350326, "全帯域_10ms_envelope_correlation": 0.9976195491536262, "全帯域_10ms_modulation_mae_db": 0.3420985101582243, "全帯域_40ms_log_rms_mae_db": 0.5763231973405227, "全帯域_40ms_rms_envelope_nrmse": 0.029737775576235845, "全帯域_40ms_envelope_correlation": 0.9993623382899212, "全帯域_40ms_modulation_mae_db": 0.4078720695905042, "300–1000 Hz_10ms_log_rms_mae_db": 1.1876757807692562, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11152484193059903, "300–1000 Hz_10ms_envelope_correlation": 0.991397284656517, "300–1000 Hz_10ms_modulation_mae_db": 0.6516555848334938, "300–1000 Hz_40ms_log_rms_mae_db": 0.7225180579774318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06823248701154899, "300–1000 Hz_40ms_envelope_correlation": 0.9967419313477185, "300–1000 Hz_40ms_modulation_mae_db": 0.5439881474104816, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5431012191814713, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17963320012993778, "1000–3000 Hz_10ms_envelope_correlation": 0.981180539044395, "1000–3000 Hz_10ms_modulation_mae_db": 1.9372739162858792, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1527395478857523, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11620221611267721, "1000–3000 Hz_40ms_envelope_correlation": 0.9940444449248831, "1000–3000 Hz_40ms_modulation_mae_db": 2.083124068012162, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7383758160559841, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21548623848841714, "3000–6000 Hz_10ms_envelope_correlation": 0.9650946726193931, "3000–6000 Hz_10ms_modulation_mae_db": 0.778854391036831, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2175461035384125, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11944463965978563, "3000–6000 Hz_40ms_envelope_correlation": 0.9883369317242304, "3000–6000 Hz_40ms_modulation_mae_db": 1.0992352109526424, "mr_stft_db": 5.5199576065864315, "band_mae_db": 0.25099032492535633, "rms_gain_error_db": -0.12014578834374455, "output_peak": 0.4013140590917175}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0035303451217659, "全帯域_10ms_rms_envelope_nrmse": 0.06559329260475319, "全帯域_10ms_envelope_correlation": 0.9963600913728964, "全帯域_10ms_modulation_mae_db": 0.3500489498086996, "全帯域_40ms_log_rms_mae_db": 0.5754084337927534, "全帯域_40ms_rms_envelope_nrmse": 0.03853092892583465, "全帯域_40ms_envelope_correlation": 0.9986630130582288, "全帯域_40ms_modulation_mae_db": 0.4885204824255373, "300–1000 Hz_10ms_log_rms_mae_db": 1.117720088913834, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09987654746883061, "300–1000 Hz_10ms_envelope_correlation": 0.9933135458811114, "300–1000 Hz_10ms_modulation_mae_db": 0.6925994899792781, "300–1000 Hz_40ms_log_rms_mae_db": 0.7648407685470685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06328807537615529, "300–1000 Hz_40ms_envelope_correlation": 0.9974399931402882, "300–1000 Hz_40ms_modulation_mae_db": 0.7992037004117428, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4700363775719625, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23207930464575036, "1000–3000 Hz_10ms_envelope_correlation": 0.9645098634104754, "1000–3000 Hz_10ms_modulation_mae_db": 0.2514649756809764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1459873251847077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13798525633897174, "1000–3000 Hz_40ms_envelope_correlation": 0.9869515216571347, "1000–3000 Hz_40ms_modulation_mae_db": 0.07241353155272615, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7831810488981525, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2084336184246865, "3000–6000 Hz_10ms_envelope_correlation": 0.9679264993031674, "3000–6000 Hz_10ms_modulation_mae_db": 1.509700932618868, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2594737753941578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11823640432328279, "3000–6000 Hz_40ms_envelope_correlation": 0.9903995137159093, "3000–6000 Hz_40ms_modulation_mae_db": 2.2304357566582333, "mr_stft_db": 5.507314974461572, "band_mae_db": 0.26047466282901094, "rms_gain_error_db": 0.015732898226769976, "output_peak": 0.429293017523294}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9316082080293248, "全帯域_10ms_rms_envelope_nrmse": 0.05401420388498547, "全帯域_10ms_envelope_correlation": 0.9976291059167183, "全帯域_10ms_modulation_mae_db": 0.3704348587748889, "全帯域_40ms_log_rms_mae_db": 0.5536831797017584, "全帯域_40ms_rms_envelope_nrmse": 0.029145179559114526, "全帯域_40ms_envelope_correlation": 0.9993622652723851, "全帯域_40ms_modulation_mae_db": 0.31779173825570456, "300–1000 Hz_10ms_log_rms_mae_db": 1.0786804576000069, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12151562541571573, "300–1000 Hz_10ms_envelope_correlation": 0.9900382347169028, "300–1000 Hz_10ms_modulation_mae_db": 0.5519556347239499, "300–1000 Hz_40ms_log_rms_mae_db": 0.7100842774411212, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07250597467283135, "300–1000 Hz_40ms_envelope_correlation": 0.9967019266987371, "300–1000 Hz_40ms_modulation_mae_db": 0.4442905956128183, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4242093875991617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16358235532681453, "1000–3000 Hz_10ms_envelope_correlation": 0.9828441683789436, "1000–3000 Hz_10ms_modulation_mae_db": 1.6814415796478455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1193474299597042, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11702792348759293, "1000–3000 Hz_40ms_envelope_correlation": 0.9911754358655942, "1000–3000 Hz_40ms_modulation_mae_db": 1.743342151645609, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7463725165577118, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21899727144572295, "3000–6000 Hz_10ms_envelope_correlation": 0.963797210587741, "3000–6000 Hz_10ms_modulation_mae_db": 1.4356646996513869, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2134528512525815, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13136129938564323, "3000–6000 Hz_40ms_envelope_correlation": 0.986610441611216, "3000–6000 Hz_40ms_modulation_mae_db": 1.991940947529452, "mr_stft_db": 5.500256994965551, "band_mae_db": 0.2832196576181178, "rms_gain_error_db": -0.10179660478521015, "output_peak": 0.40626758578087036}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9851786731426556, "全帯域_10ms_rms_envelope_nrmse": 0.0620945993298152, "全帯域_10ms_envelope_correlation": 0.9968906933735842, "全帯域_10ms_modulation_mae_db": 0.3847972319117025, "全帯域_40ms_log_rms_mae_db": 0.5309494829115191, "全帯域_40ms_rms_envelope_nrmse": 0.032987220938491456, "全帯域_40ms_envelope_correlation": 0.9992305604311394, "全帯域_40ms_modulation_mae_db": 0.592682175968185, "300–1000 Hz_10ms_log_rms_mae_db": 1.215807974081438, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12415549967567212, "300–1000 Hz_10ms_envelope_correlation": 0.9895910177419908, "300–1000 Hz_10ms_modulation_mae_db": 0.9624507971122708, "300–1000 Hz_40ms_log_rms_mae_db": 0.7704830212201261, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07085236723741634, "300–1000 Hz_40ms_envelope_correlation": 0.996908842215336, "300–1000 Hz_40ms_modulation_mae_db": 1.0819775871097341, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4239786312604372, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15077907943175975, "1000–3000 Hz_10ms_envelope_correlation": 0.9873781766533578, "1000–3000 Hz_10ms_modulation_mae_db": 1.4052400160673422, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0685613325285184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10635086393074426, "1000–3000 Hz_40ms_envelope_correlation": 0.9950648413742574, "1000–3000 Hz_40ms_modulation_mae_db": 1.397942320079558, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7238310607366745, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20830313303268552, "3000–6000 Hz_10ms_envelope_correlation": 0.9671771544092606, "3000–6000 Hz_10ms_modulation_mae_db": 1.3072164805717634, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2371121784314048, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11250894693633783, "3000–6000 Hz_40ms_envelope_correlation": 0.9902135512599396, "3000–6000 Hz_40ms_modulation_mae_db": 1.7014617283088285, "mr_stft_db": 5.50812635455003, "band_mae_db": 0.3099453693960902, "rms_gain_error_db": -0.13462490286274154, "output_peak": 0.40308740360485407}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0208622677564294, "全帯域_10ms_rms_envelope_nrmse": 0.06729904909910212, "全帯域_10ms_envelope_correlation": 0.9964716369729653, "全帯域_10ms_modulation_mae_db": 0.8032394046227912, "全帯域_40ms_log_rms_mae_db": 0.653248607154916, "全帯域_40ms_rms_envelope_nrmse": 0.039170122981565934, "全帯域_40ms_envelope_correlation": 0.9990061098029858, "全帯域_40ms_modulation_mae_db": 0.7971468977327798, "300–1000 Hz_10ms_log_rms_mae_db": 1.1635507019810436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1468884207880169, "300–1000 Hz_10ms_envelope_correlation": 0.9868128003372771, "300–1000 Hz_10ms_modulation_mae_db": 1.1299591477329909, "300–1000 Hz_40ms_log_rms_mae_db": 0.766099981262628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09238746122967055, "300–1000 Hz_40ms_envelope_correlation": 0.9963127099793664, "300–1000 Hz_40ms_modulation_mae_db": 1.500855656957335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.459500516298389, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19793744160437418, "1000–3000 Hz_10ms_envelope_correlation": 0.9781799060062686, "1000–3000 Hz_10ms_modulation_mae_db": 1.6032450338297863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0998166943280936, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14238008282344491, "1000–3000 Hz_40ms_envelope_correlation": 0.9912945742458841, "1000–3000 Hz_40ms_modulation_mae_db": 0.9210130557709747, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6347203864974382, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18216948714060804, "3000–6000 Hz_10ms_envelope_correlation": 0.9750871630048814, "3000–6000 Hz_10ms_modulation_mae_db": 0.8692186910452447, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0629289103855788, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09353881220477267, "3000–6000 Hz_40ms_envelope_correlation": 0.9929084562427847, "3000–6000 Hz_40ms_modulation_mae_db": 1.3185181966253137, "mr_stft_db": 5.522729383795407, "band_mae_db": 0.39217174100456287, "rms_gain_error_db": -0.18784629546165083, "output_peak": 0.39368077769672954}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9464536375426689, "全帯域_10ms_rms_envelope_nrmse": 0.053615045937099905, "全帯域_10ms_envelope_correlation": 0.9975960240827602, "全帯域_10ms_modulation_mae_db": 0.4172417350771051, "全帯域_40ms_log_rms_mae_db": 0.524412407333046, "全帯域_40ms_rms_envelope_nrmse": 0.030276512466072387, "全帯域_40ms_envelope_correlation": 0.9992210581780944, "全帯域_40ms_modulation_mae_db": 0.45069140544265734, "300–1000 Hz_10ms_log_rms_mae_db": 1.1899908818878806, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11404604816448322, "300–1000 Hz_10ms_envelope_correlation": 0.991538867479184, "300–1000 Hz_10ms_modulation_mae_db": 1.003874670677214, "300–1000 Hz_40ms_log_rms_mae_db": 0.7866383789262422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06927681114081242, "300–1000 Hz_40ms_envelope_correlation": 0.9973232519973922, "300–1000 Hz_40ms_modulation_mae_db": 1.1044962717443523, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4431420530751071, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17350134460886948, "1000–3000 Hz_10ms_envelope_correlation": 0.9802890195194256, "1000–3000 Hz_10ms_modulation_mae_db": 1.1015698730753674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1010191041214596, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11367247624751832, "1000–3000 Hz_40ms_envelope_correlation": 0.9910697101725628, "1000–3000 Hz_40ms_modulation_mae_db": 1.2268579913420483, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6515166491138666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1981292217533145, "3000–6000 Hz_10ms_envelope_correlation": 0.9705405476172593, "3000–6000 Hz_10ms_modulation_mae_db": 1.2822565931002274, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1426038881765248, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10945681967076562, "3000–6000 Hz_40ms_envelope_correlation": 0.9910576174724826, "3000–6000 Hz_40ms_modulation_mae_db": 2.1491866517587366, "mr_stft_db": 5.521116258489811, "band_mae_db": 0.17198603875853732, "rms_gain_error_db": -0.0684261871225898, "output_peak": 0.40564753903076967}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44400, "output_samples": 44400, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.66245374192615, "spectral_convergence": 0.17624342114842584, "spectral_convergence_status": "ok", "amplitude_floor": 1.585777823535548e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4132433575400265, "spectral_convergence": 0.1672963945046179, "spectral_convergence_status": "ok", "amplitude_floor": 7.694479165248034e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.46516166915368, "spectral_convergence": 0.1715082949930735, "spectral_convergence_status": "ok", "amplitude_floor": 6.411814451459315e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00186841781908955, "output_power": 0.001856188343149464, "delta_db": -0.028519595214117004, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005255584463676, "output_power": 0.000513897203623491, "delta_db": -0.0974476564945858, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025377475865110864, "output_power": 0.00025026659438514274, "delta_db": -0.06045539620351148, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00014708318852606447, "output_power": 0.00012510593460302909, "delta_db": -0.7028512445649631, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.522695850195022e-06, "output_power": 8.72856505969549e-06, "delta_db": -0.3781706007695633, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.071083630879648e-06, "output_power": 6.127447406296374e-06, "delta_db": 0.04013376818958545, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21792971023938768, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07813887406594977, "pitch": null, "output_rms_dbfs": -25.590414087117246, "output_peak": 0.41209785267548554, "reference_rms_dbfs": -25.512275213051296, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.513619589539952}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44400, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 0.00014622417526559514, "log_rms_mae_db": 0.8387393222398823, "rms_envelope_nrmse": 0.05821004908634618, "envelope_correlation": 0.9971700406419763, "rms_gain_error_db": -0.07813853620663594, "reference_mean_power": 0.0028155003489198547, "output_mean_power": 0.0027652966777440727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.019039396656961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008418483111337408, "output_power": 0.0008023262555160089, "reference_active": true, "delta_db": -0.208828405853113}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001428627689371398, "output_power": 0.00013871835955415172, "reference_active": true, "delta_db": -0.12785118964166767}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9192224262112575e-05, "output_power": 1.7638527578988197e-05, "reference_active": true, "delta_db": -0.36662981271480666}], "active_band_count": 3, "active_band_mae_db": 0.23443646940319576}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 0.00013893685565863478, "log_rms_mae_db": 0.4758507247300194, "rms_envelope_nrmse": 0.03221406791925102, "envelope_correlation": 0.9991277944659482, "rms_gain_error_db": -0.07813903506904732, "reference_mean_power": 0.0028463248833832306, "output_mean_power": 0.0027955712535097698}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.8422021834325858e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007582285194051116, "output_power": 0.0007207594477208144, "reference_active": true, "delta_db": -0.22009771772001727}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661169958011628e-05, "output_power": 6.397186542111703e-05, "reference_active": true, "delta_db": -0.17561499768867211}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7101796636835213e-06, "output_power": 1.3922664126287004e-06, "reference_active": true, "delta_db": -0.8931939154836721}], "active_band_count": 3, "active_band_mae_db": 0.4296355436307871}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 8.753945313428826e-05, "log_rms_mae_db": 1.1223729330365237, "rms_envelope_nrmse": 0.11589102279902369, "envelope_correlation": 0.9910965921732598, "rms_gain_error_db": -0.2926633376324781, "reference_mean_power": 0.0005201241069420424, "output_mean_power": 0.0004862287666616799}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 5.057137989146479e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002701505367125327, "output_power": 0.00024500373809075907, "reference_active": true, "delta_db": -0.4243312412827336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.4045222297363644e-05, "output_power": 4.444980876059529e-05, "reference_active": true, "delta_db": -0.8488741069451249}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.5635511670477814e-06, "output_power": 5.551618413954796e-06, "reference_active": true, "delta_db": -1.3430614163477554}], "active_band_count": 3, "active_band_mae_db": 0.8720889215252047}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 7.440703161891069e-05, "log_rms_mae_db": 0.7263350688826837, "rms_envelope_nrmse": 0.06936193429884067, "envelope_correlation": 0.997158129814313, "rms_gain_error_db": -0.2926634586441741, "reference_mean_power": 0.0005258188868038473, "output_mean_power": 0.0004915524165900808}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 4.6537380366659205e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025071266719842557, "output_power": 0.00022716542083608265, "reference_active": true, "delta_db": -0.42834053458229826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3806665370442115e-05, "output_power": 1.9191309476676473e-05, "reference_active": true, "delta_db": -0.9359395861800284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0797322752712393e-06, "output_power": 7.202954460990645e-07, "reference_active": true, "delta_db": -1.7580541440348119}], "active_band_count": 3, "active_band_mae_db": 1.0407780882657127}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 5.3294712554339956e-05, "log_rms_mae_db": 1.42259871327577, "rms_envelope_nrmse": 0.18103403130954704, "envelope_correlation": 0.9802928332364008, "rms_gain_error_db": -0.5982234623683326, "reference_mean_power": 0.00010733282194912887, "output_mean_power": 9.3521228166868e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 1.1016693525679884e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607161402154175e-05, "output_power": 5.488380808843661e-05, "reference_active": true, "delta_db": -0.805706789700717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0221078826476055e-05, "output_power": 7.623536768845549e-06, "reference_active": true, "delta_db": -1.2734023837228678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.376144897878774e-06, "output_power": 1.6152016560566147e-06, "reference_active": true, "delta_db": -3.201943262521258}], "active_band_count": 3, "active_band_mae_db": 1.7603508119816142}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 4.1357257962539876e-05, "log_rms_mae_db": 1.132879079897875, "rms_envelope_nrmse": 0.12721871371465193, "envelope_correlation": 0.9913893070196539, "rms_gain_error_db": -0.5982236388783878, "reference_mean_power": 0.00010850799286328448, "output_mean_power": 9.45451741767235e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.0153781665834753e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.269855261867839e-05, "output_power": 5.026651780903135e-05, "reference_active": true, "delta_db": -0.9597871465019006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.20737764206477e-06, "output_power": 3.6204387295940715e-06, "reference_active": true, "delta_db": -0.652502926780371}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.246816673622798e-07, "output_power": 9.804361010542878e-08, "reference_active": true, "delta_db": -3.601483438985713}], "active_band_count": 3, "active_band_mae_db": 1.737924504089328}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 1.102648736164134e-05, "log_rms_mae_db": 1.7217085203552884, "rms_envelope_nrmse": 0.19018600350553297, "envelope_correlation": 0.9729594495512256, "rms_gain_error_db": -0.2222000766137514, "reference_mean_power": 5.352135557413054e-06, "output_mean_power": 5.085188977178402e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.71214230899019e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.798500894794614e-06, "output_power": 1.5935005793449312e-06, "reference_active": true, "delta_db": -0.5255843252073402}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2650926706890884e-07, "output_power": 2.9834925892975056e-07, "reference_active": true, "delta_db": -1.5520350839321382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0854303858213565e-07, "output_power": 8.975116504238354e-08, "reference_active": true, "delta_db": -0.8256188038766189}], "active_band_count": 3, "active_band_mae_db": 0.9677460710053657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 9.954651742580568e-06, "log_rms_mae_db": 1.2174936144228947, "rms_envelope_nrmse": 0.10596253530830888, "envelope_correlation": 0.9914573382303448, "rms_gain_error_db": -0.22220156152268117, "reference_mean_power": 5.410729753830906e-06, "output_mean_power": 5.140858933771371e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 2.394714459074953e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.814627629314412e-06, "output_power": 1.5533533394342868e-06, "reference_active": true, "delta_db": -0.6751726369559136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0097168111890166e-07, "output_power": 1.454972009042863e-07, "reference_active": true, "delta_db": -1.4028022699572171}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.411981302222167e-09, "output_power": 5.283624513457461e-09, "reference_active": true, "delta_db": -2.507491093106112}], "active_band_count": 3, "active_band_mae_db": 1.5284886666730808}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8972107728594382, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b07d66648fff3e0d9208eb7ed566296ccb9911c6a7273a129dfcbfdb238fc55b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9798117494362721, "standard_deviation": 0.006778661963310716, "minimum": 0.9645098634104754, "maximum": 0.9873781766533578, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4592794790938342, "standard_deviation": 0.04145524486193939, "minimum": 1.42259871327577, "maximum": 1.5431012191814713, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3943853705563622, "standard_deviation": 0.5274212926168003, "minimum": 0.2514649756809764, "maximum": 1.9372739162858792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18075183107830628, "standard_deviation": 0.024938137935717173, "minimum": 0.15077907943175975, "maximum": 0.23207930464575036, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919782450354819, "standard_deviation": 0.002659642209757719, "minimum": 0.9869515216571347, "maximum": 0.9950648413742574, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1192310258097833, "standard_deviation": 0.028042186318021717, "minimum": 1.0685613325285184, "maximum": 1.1527395478857523, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3356632004733933, "standard_deviation": 0.6208361919136438, "minimum": 0.07241353155272615, "maximum": 2.083124068012162, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12123778033669766, "standard_deviation": 0.013273841188274365, "minimum": 0.10635086393074426, "maximum": 0.14238008282344491, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9687188737479169, "standard_deviation": 0.003866019469095955, "minimum": 0.963797210587741, "maximum": 0.9750871630048814, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7211145746811503, "standard_deviation": 0.05266734952824438, "minimum": 1.6347203864974382, "maximum": 1.7831810488981525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.241588036013759, "standard_deviation": 0.3454556727129763, "minimum": 0.778854391036831, "maximum": 1.782046429080385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20443944965902924, "standard_deviation": 0.013047128774484053, "minimum": 0.18216948714060804, "maximum": 0.21899727144572295, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9901693265620627, "standard_deviation": 0.0019298303931373124, "minimum": 0.986610441611216, "maximum": 0.9929084562427847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1965002216483114, "standard_deviation": 0.06341662281047039, "minimum": 1.0629289103855788, "maximum": 1.2594737753941578, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.7556904402398406, "standard_deviation": 0.4115922119686646, "minimum": 1.0992352109526424, "maximum": 2.2304357566582333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11456974927785926, "standard_deviation": 0.011938539237579336, "minimum": 0.09353881220477267, "maximum": 0.13136129938564323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990882416729765, "standard_deviation": 0.002114564972984933, "minimum": 0.9868128003372771, "maximum": 0.9933135458811114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.156937695649391, "standard_deviation": 0.04617044647207631, "minimum": 1.0786804576000069, "maximum": 1.215807974081438, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8437714978124875, "standard_deviation": 0.1960647288115615, "minimum": 0.5519556347239499, "maximum": 1.1299591477329909, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11668030183613809, "standard_deviation": 0.015127720115270651, "minimum": 0.09954440844676339, "maximum": 0.1468884207880169, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971296349332126, "standard_deviation": 0.0006467641479586147, "minimum": 0.9963127099793664, "maximum": 0.9984502942725487, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7525567185414098, "standard_deviation": 0.028404573394206434, "minimum": 0.7100842774411212, "maximum": 0.7866383789262422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9502644718947175, "standard_deviation": 0.34120284340585755, "minimum": 0.4442905956128183, "maximum": 1.500855656957335, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06950886509254765, "standard_deviation": 0.011642130496863206, "minimum": 0.050165809773105634, "maximum": 0.09238746122967055, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27420928034560327, "standard_deviation": 0.06621451481725844, "minimum": 0.17198603875853732, "maximum": 0.39217174100456287, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.513530973793389, "standard_deviation": 0.007838093201912451, "minimum": 5.500256994965551, "maximum": 5.522729383795407, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40554376844081275, "standard_deviation": 0.011529312699134571, "minimum": 0.39296191212278136, "maximum": 0.429293017523294, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0976910073143433, "standard_deviation": 0.058786137927534836, "minimum": -0.18784629546165083, "maximum": 0.015732898226769976, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9972117012030095, "standard_deviation": 0.0005880554472228124, "minimum": 0.9963600913728964, "maximum": 0.997956468109549, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9494019037381194, "standard_deviation": 0.05883651360578274, "minimum": 0.8387393222398823, "maximum": 1.0208622677564294, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.430803923973188, "standard_deviation": 0.17338936867738777, "minimum": 0.23443646940319576, "maximum": 0.8032394046227912, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.05819213226192025, "standard_deviation": 0.006184120699550006, "minimum": 0.05032793737190958, "maximum": 0.06729904909910212, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991724619213549, "standard_deviation": 0.00024547962126707684, "minimum": 0.9986630130582288, "maximum": 0.9994065558721367, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.555232873455774, "standard_deviation": 0.05114822481647974, "minimum": 0.4758507247300194, "maximum": 0.653248607154916, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5042116199231126, "standard_deviation": 0.14559323281298292, "minimum": 0.31779173825570456, "maximum": 0.7971468977327798, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03253869042024363, "standard_deviation": 0.00419319752930323, "minimum": 0.02824771499538316, "maximum": 0.039170122981565934, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7127-75946-0002", "language": "en", "style": "normal", "speaker": "librispeech_7127", "source_utterance_id": "7127-75946-0002", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7127-75946-0002.wav", "source_file_sha256": "8b6d743af5caee3fd2ae188db26787caa376c129c15bd8262c20ab7ef32422b7", "decoded_pcm_sha256": "93c54b2cc025d80c3059dc9d273712d4259e0ddbef769ef56e3f3d16805ad734", "original_source_sha256": "8a4077cfaf005d616cab7fe18bc2a434b627b24de056d78cafc33f09faa1cf6c", "corpus_entry_id": "librispeech_7127-75946-0002"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.0933690071105957, "predicted_mos": 3.0933690071105957, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3d3ac76699a876710096c0ed4bc097114e9e5718acc096decce5c4f6b7632133", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.285229685003287, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7127-75946-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "3d3ac76699a876710096c0ed4bc097114e9e5718acc096decce5c4f6b7632133", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0981fb6efc50f918e211f5056d645e7f31a403d249c14e3bd928b63982ab99f0"}, "rows": [{"case": "librispeech_7127-75946-0002", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.791358709335327, "predicted_mos": 2.791358709335327, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "56a1cf0493b2852db732f2762f0268797d6966a4f55c76fa408b3f27c6a8d669", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2155863199986925, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7127-75946-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "56a1cf0493b2852db732f2762f0268797d6966a4f55c76fa408b3f27c6a8d669", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4134403434059284, "全帯域_10ms_rms_envelope_nrmse": 0.09572844428019968, "全帯域_10ms_envelope_correlation": 0.9913269981224757, "全帯域_10ms_modulation_mae_db": 0.14126981940753247, "全帯域_40ms_log_rms_mae_db": 0.7762778390040584, "全帯域_40ms_rms_envelope_nrmse": 0.05203350735416465, "全帯域_40ms_envelope_correlation": 0.9972173520391215, "全帯域_40ms_modulation_mae_db": 0.21360059139204546, "300–1000 Hz_10ms_log_rms_mae_db": 1.6186621558133623, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13863300533552722, "300–1000 Hz_10ms_envelope_correlation": 0.9854122277987598, "300–1000 Hz_10ms_modulation_mae_db": 0.1038908757042462, "300–1000 Hz_40ms_log_rms_mae_db": 1.0882510637248568, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08204533936570227, "300–1000 Hz_40ms_envelope_correlation": 0.9945204025883454, "300–1000 Hz_40ms_modulation_mae_db": 0.3966276499394965, "1000–3000 Hz_10ms_log_rms_mae_db": 1.689654081033315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17553572698251396, "1000–3000 Hz_10ms_envelope_correlation": 0.9774350797175976, "1000–3000 Hz_10ms_modulation_mae_db": 0.2707314580361293, "1000–3000 Hz_40ms_log_rms_mae_db": 1.079631288603362, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0987967252707256, "1000–3000 Hz_40ms_envelope_correlation": 0.9922636942053253, "1000–3000 Hz_40ms_modulation_mae_db": 0.49840972715530457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.974923429437559, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1626255659135105, "3000–6000 Hz_10ms_envelope_correlation": 0.9833223603718654, "3000–6000 Hz_10ms_modulation_mae_db": 0.3479074928465431, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2075028571411048, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09485777496629175, "3000–6000 Hz_40ms_envelope_correlation": 0.9941794875179649, "3000–6000 Hz_40ms_modulation_mae_db": 0.49040997823457094, "mr_stft_db": 5.978360204717693, "band_mae_db": 0.12265206074331164, "rms_gain_error_db": 0.0056594564428315616, "output_peak": 0.4953672375256083}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5213805959142515, "全帯域_10ms_rms_envelope_nrmse": 0.09365870862284863, "全帯域_10ms_envelope_correlation": 0.9917422400307069, "全帯域_10ms_modulation_mae_db": 0.3020739147853142, "全帯域_40ms_log_rms_mae_db": 0.7869245710324538, "全帯域_40ms_rms_envelope_nrmse": 0.05008805246849996, "全帯域_40ms_envelope_correlation": 0.9975752362378215, "全帯域_40ms_modulation_mae_db": 0.44708990535403026, "300–1000 Hz_10ms_log_rms_mae_db": 1.6153487719482378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1349808451628153, "300–1000 Hz_10ms_envelope_correlation": 0.9861498381027646, "300–1000 Hz_10ms_modulation_mae_db": 0.5549067182903261, "300–1000 Hz_40ms_log_rms_mae_db": 1.0538416180361672, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08642340975790837, "300–1000 Hz_40ms_envelope_correlation": 0.9941170384146321, "300–1000 Hz_40ms_modulation_mae_db": 0.6897792100218259, "1000–3000 Hz_10ms_log_rms_mae_db": 1.763165604929127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1677230078562574, "1000–3000 Hz_10ms_envelope_correlation": 0.9798968292860625, "1000–3000 Hz_10ms_modulation_mae_db": 0.3429123843748972, "1000–3000 Hz_40ms_log_rms_mae_db": 1.102311707570286, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09508767505818802, "1000–3000 Hz_40ms_envelope_correlation": 0.993974261549805, "1000–3000 Hz_40ms_modulation_mae_db": 0.18683680026043717, "3000–6000 Hz_10ms_log_rms_mae_db": 2.000938295417677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15956075416898693, "3000–6000 Hz_10ms_envelope_correlation": 0.9838611507323973, "3000–6000 Hz_10ms_modulation_mae_db": 0.5456083246081397, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1997805613330197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08604198741076069, "3000–6000 Hz_40ms_envelope_correlation": 0.9951566882393886, "3000–6000 Hz_40ms_modulation_mae_db": 0.8554858016272652, "mr_stft_db": 5.9781552111446326, "band_mae_db": 0.2196568108256405, "rms_gain_error_db": -0.11811551864520249, "output_peak": 0.5435638943314531}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.379514200081713, "全帯域_10ms_rms_envelope_nrmse": 0.08641372673214642, "全帯域_10ms_envelope_correlation": 0.992982710607254, "全帯域_10ms_modulation_mae_db": 0.10742409289044591, "全帯域_40ms_log_rms_mae_db": 0.6782326412007319, "全帯域_40ms_rms_envelope_nrmse": 0.04610435120209137, "全帯域_40ms_envelope_correlation": 0.9979574040445175, "全帯域_40ms_modulation_mae_db": 0.2696669742179388, "300–1000 Hz_10ms_log_rms_mae_db": 1.542346778219775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11410283210367826, "300–1000 Hz_10ms_envelope_correlation": 0.9900797210848252, "300–1000 Hz_10ms_modulation_mae_db": 0.07021194174500743, "300–1000 Hz_40ms_log_rms_mae_db": 0.9767603452274147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0640377123534962, "300–1000 Hz_40ms_envelope_correlation": 0.9966569807712357, "300–1000 Hz_40ms_modulation_mae_db": 0.22784598761109173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7443814842474499, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17065916712232207, "1000–3000 Hz_10ms_envelope_correlation": 0.9789648165012127, "1000–3000 Hz_10ms_modulation_mae_db": 0.4327193938668142, "1000–3000 Hz_40ms_log_rms_mae_db": 1.085911619832663, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10251743433531985, "1000–3000 Hz_40ms_envelope_correlation": 0.9924858194344711, "1000–3000 Hz_40ms_modulation_mae_db": 0.908521236400985, "3000–6000 Hz_10ms_log_rms_mae_db": 2.011735378585876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1781819752773643, "3000–6000 Hz_10ms_envelope_correlation": 0.9793265031605045, "3000–6000 Hz_10ms_modulation_mae_db": 0.19020593313079429, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1753532434092735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10237954275397373, "3000–6000 Hz_40ms_envelope_correlation": 0.9928495718038008, "3000–6000 Hz_40ms_modulation_mae_db": 0.15327113523015543, "mr_stft_db": 5.990055487916692, "band_mae_db": 0.156012213221406, "rms_gain_error_db": -0.11014428398898395, "output_peak": 0.4665098542188126}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5181777392904132, "全帯域_10ms_rms_envelope_nrmse": 0.0920961235312683, "全帯域_10ms_envelope_correlation": 0.9920222208476296, "全帯域_10ms_modulation_mae_db": 0.20313213685711728, "全帯域_40ms_log_rms_mae_db": 0.8395531026754293, "全帯域_40ms_rms_envelope_nrmse": 0.05051090401129935, "全帯域_40ms_envelope_correlation": 0.9974915891395439, "全帯域_40ms_modulation_mae_db": 0.18982125697274255, "300–1000 Hz_10ms_log_rms_mae_db": 1.6677501272469422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12952446499388154, "300–1000 Hz_10ms_envelope_correlation": 0.9872919714963346, "300–1000 Hz_10ms_modulation_mae_db": 0.20759475212585654, "300–1000 Hz_40ms_log_rms_mae_db": 1.0917764937867258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0738949727339951, "300–1000 Hz_40ms_envelope_correlation": 0.9955572132554152, "300–1000 Hz_40ms_modulation_mae_db": 0.21008936315085872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.67978926629551, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17398455870882684, "1000–3000 Hz_10ms_envelope_correlation": 0.9777923554239144, "1000–3000 Hz_10ms_modulation_mae_db": 0.2531031377152592, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0267860775085405, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10420514246894573, "1000–3000 Hz_40ms_envelope_correlation": 0.9914189735051887, "1000–3000 Hz_40ms_modulation_mae_db": 0.3623230107817123, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0134413138793947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15955355269594781, "3000–6000 Hz_10ms_envelope_correlation": 0.9836441995017022, "3000–6000 Hz_10ms_modulation_mae_db": 0.9242953970350415, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2266928319374022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08564509109484654, "3000–6000 Hz_40ms_envelope_correlation": 0.995459066333, "3000–6000 Hz_40ms_modulation_mae_db": 0.8827675755445524, "mr_stft_db": 5.965848985593219, "band_mae_db": 0.15483093186376432, "rms_gain_error_db": -0.09571126082915825, "output_peak": 0.5045302117195151}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3544679143997547, "全帯域_10ms_rms_envelope_nrmse": 0.0871098357117876, "全帯域_10ms_envelope_correlation": 0.9928168076504095, "全帯域_10ms_modulation_mae_db": 0.20024481679108722, "全帯域_40ms_log_rms_mae_db": 0.6704158981564425, "全帯域_40ms_rms_envelope_nrmse": 0.0432590228404348, "全帯域_40ms_envelope_correlation": 0.9981420650407073, "全帯域_40ms_modulation_mae_db": 0.304208479779624, "300–1000 Hz_10ms_log_rms_mae_db": 1.660970350766748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12723797000298223, "300–1000 Hz_10ms_envelope_correlation": 0.9876745407064716, "300–1000 Hz_10ms_modulation_mae_db": 0.31078889912981583, "300–1000 Hz_40ms_log_rms_mae_db": 1.0605627842612113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07085950622017859, "300–1000 Hz_40ms_envelope_correlation": 0.9958898411890295, "300–1000 Hz_40ms_modulation_mae_db": 0.13829629397484258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7398421722020827, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16632095910160147, "1000–3000 Hz_10ms_envelope_correlation": 0.9802579528030948, "1000–3000 Hz_10ms_modulation_mae_db": 1.2956887530235126, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1012850148835607, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09622418683999605, "1000–3000 Hz_40ms_envelope_correlation": 0.993768707666505, "1000–3000 Hz_40ms_modulation_mae_db": 1.4304545831802766, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0538340282293235, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1501238992206015, "3000–6000 Hz_10ms_envelope_correlation": 0.9853644273712721, "3000–6000 Hz_10ms_modulation_mae_db": 0.5088653599232543, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2245524414682913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08610394146954313, "3000–6000 Hz_40ms_envelope_correlation": 0.9950658242629017, "3000–6000 Hz_40ms_modulation_mae_db": 0.5993627830446204, "mr_stft_db": 5.989590031815394, "band_mae_db": 0.19233295971446793, "rms_gain_error_db": -0.07769120071618629, "output_peak": 0.4901647535441965}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4707967281360739, "全帯域_10ms_rms_envelope_nrmse": 0.09180911591533492, "全帯域_10ms_envelope_correlation": 0.9920002975195771, "全帯域_10ms_modulation_mae_db": 0.2015435853692943, "全帯域_40ms_log_rms_mae_db": 0.7417494937271367, "全帯域_40ms_rms_envelope_nrmse": 0.049062742762429994, "全帯域_40ms_envelope_correlation": 0.9975681189338044, "全帯域_40ms_modulation_mae_db": 0.1585975435823088, "300–1000 Hz_10ms_log_rms_mae_db": 1.6218208028556644, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12596513241567514, "300–1000 Hz_10ms_envelope_correlation": 0.9878708649067113, "300–1000 Hz_10ms_modulation_mae_db": 0.3298561556539635, "300–1000 Hz_40ms_log_rms_mae_db": 1.0044401599709807, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06929679018421475, "300–1000 Hz_40ms_envelope_correlation": 0.9961364348580023, "300–1000 Hz_40ms_modulation_mae_db": 0.6102455941413719, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6947151827983145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1621339809804912, "1000–3000 Hz_10ms_envelope_correlation": 0.9807822619027479, "1000–3000 Hz_10ms_modulation_mae_db": 0.3955142904225076, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0529259597930347, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0870946692760759, "1000–3000 Hz_40ms_envelope_correlation": 0.9943355362950179, "1000–3000 Hz_40ms_modulation_mae_db": 0.382794044758463, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9782419711840076, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16260161833887324, "3000–6000 Hz_10ms_envelope_correlation": 0.9838035249899346, "3000–6000 Hz_10ms_modulation_mae_db": 0.26409783088922634, "3000–6000 Hz_40ms_log_rms_mae_db": 1.16664205068153, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09153780096402228, "3000–6000 Hz_40ms_envelope_correlation": 0.9949456910673999, "3000–6000 Hz_40ms_modulation_mae_db": 0.4956509076401987, "mr_stft_db": 5.987629420694705, "band_mae_db": 0.15126093089575227, "rms_gain_error_db": -0.06552580422172449, "output_peak": 0.5247788999647317}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4215437711143268, "全帯域_10ms_rms_envelope_nrmse": 0.09969717471468789, "全帯域_10ms_envelope_correlation": 0.990595686790113, "全帯域_10ms_modulation_mae_db": 0.1418605253431363, "全帯域_40ms_log_rms_mae_db": 0.7149721935026547, "全帯域_40ms_rms_envelope_nrmse": 0.04985375648426135, "全帯域_40ms_envelope_correlation": 0.9975203449389777, "全帯域_40ms_modulation_mae_db": 0.3078975664997465, "300–1000 Hz_10ms_log_rms_mae_db": 1.6841926976455421, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13610364834384908, "300–1000 Hz_10ms_envelope_correlation": 0.9858197000354204, "300–1000 Hz_10ms_modulation_mae_db": 0.32576832744047884, "300–1000 Hz_40ms_log_rms_mae_db": 1.1143896403610993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.084507530002226, "300–1000 Hz_40ms_envelope_correlation": 0.9941885904367986, "300–1000 Hz_40ms_modulation_mae_db": 0.5180536864528489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7050463827970754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19251438921841088, "1000–3000 Hz_10ms_envelope_correlation": 0.9729146770768315, "1000–3000 Hz_10ms_modulation_mae_db": 0.6416158016757438, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1097188294557843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10899135240842345, "1000–3000 Hz_40ms_envelope_correlation": 0.9913816669254554, "1000–3000 Hz_40ms_modulation_mae_db": 0.7011771914070551, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0026613036592384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1768158661167247, "3000–6000 Hz_10ms_envelope_correlation": 0.9797047783739907, "3000–6000 Hz_10ms_modulation_mae_db": 0.13185231389996058, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1847186546882407, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10046614146554758, "3000–6000 Hz_40ms_envelope_correlation": 0.9931116126080743, "3000–6000 Hz_40ms_modulation_mae_db": 0.23074445941291702, "mr_stft_db": 5.988857490356587, "band_mae_db": 0.2423575162578986, "rms_gain_error_db": -0.08060725024576455, "output_peak": 0.4946484098520997}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4346136252000536, "全帯域_10ms_rms_envelope_nrmse": 0.0878376494570175, "全帯域_10ms_envelope_correlation": 0.9926874189601805, "全帯域_10ms_modulation_mae_db": 0.24405378513423603, "全帯域_40ms_log_rms_mae_db": 0.8036805559219904, "全帯域_40ms_rms_envelope_nrmse": 0.046517512630597516, "全帯域_40ms_envelope_correlation": 0.9978103105416881, "全帯域_40ms_modulation_mae_db": 0.3113632868781131, "300–1000 Hz_10ms_log_rms_mae_db": 1.607275701974514, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12027670824271254, "300–1000 Hz_10ms_envelope_correlation": 0.9889601184566351, "300–1000 Hz_10ms_modulation_mae_db": 0.40344778168771334, "300–1000 Hz_40ms_log_rms_mae_db": 0.9833519875135528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0634525834487511, "300–1000 Hz_40ms_envelope_correlation": 0.996702262462439, "300–1000 Hz_40ms_modulation_mae_db": 0.7018048868501685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6680745652284679, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16793218728200654, "1000–3000 Hz_10ms_envelope_correlation": 0.9792872042966505, "1000–3000 Hz_10ms_modulation_mae_db": 0.5306329234383619, "1000–3000 Hz_40ms_log_rms_mae_db": 1.091122860707679, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10512586453059156, "1000–3000 Hz_40ms_envelope_correlation": 0.9913833827039406, "1000–3000 Hz_40ms_modulation_mae_db": 0.5888145230349203, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0233570781977317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17094590816370006, "3000–6000 Hz_10ms_envelope_correlation": 0.9809873885359146, "3000–6000 Hz_10ms_modulation_mae_db": 0.4261129784530739, "3000–6000 Hz_40ms_log_rms_mae_db": 1.206455169629138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0862406724844487, "3000–6000 Hz_40ms_envelope_correlation": 0.9949847408651624, "3000–6000 Hz_40ms_modulation_mae_db": 0.6961242855716844, "mr_stft_db": 5.970208329602826, "band_mae_db": 0.1824346888343041, "rms_gain_error_db": -0.05996067747246232, "output_peak": 0.49804847762675314}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.202198848547445, "spectral_convergence": 0.34231717343091633, "spectral_convergence_status": "ok", "amplitude_floor": 1.3425996786840756e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.777185230289433, "spectral_convergence": 0.3283627915755123, "spectral_convergence_status": "ok", "amplitude_floor": 1.1106411258473627e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.955696535316201, "spectral_convergence": 0.35162139043285984, "spectral_convergence_status": "ok", "amplitude_floor": 1.0520318003394391e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00033395677419724207, "output_power": 0.0003502196592873623, "delta_db": 0.20650263785916945, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008500397987721807, "output_power": 0.0008307882859363759, "delta_db": -0.09948895430559654, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006268367014596879, "output_power": 0.0006263733220259437, "delta_db": -0.0032116426856213746, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003695716856827319, "output_power": 0.00037443334797534924, "delta_db": 0.05675830305652907, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013451399439061329, "output_power": 0.00013015280093472997, "delta_db": -0.14313950530820607, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011413253897640505, "output_power": 0.00012025152999959797, "delta_db": 0.22681132124474734, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12265206074331164, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.0056594564428315616, "pitch": null, "output_rms_dbfs": -26.13997332841372, "output_peak": 0.4953672375256083, "reference_rms_dbfs": -26.14563278485655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.978360204717693}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018099058448312497, "log_rms_mae_db": 1.4134403434059284, "rms_envelope_nrmse": 0.09572844428019968, "envelope_correlation": 0.9913269981224757, "rms_gain_error_db": 0.0056590277481043194, "reference_mean_power": 0.002431072908198481, "output_mean_power": 0.0024342427564928233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.757963657265947e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003340894048273399, "output_power": 0.0003486095949921278, "reference_active": true, "delta_db": 0.18476633465766618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010057920506755965, "output_power": 9.759183302755095e-05, "reference_active": true, "delta_db": -0.13094723577724965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.918423836914529e-05, "output_power": 1.9667726067101536e-05, "reference_active": true, "delta_db": 0.10809588778768159}], "active_band_count": 3, "active_band_mae_db": 0.14126981940753247}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016835992502135956, "log_rms_mae_db": 0.7762778390040584, "rms_envelope_nrmse": 0.05203350735416465, "envelope_correlation": 0.9972173520391215, "rms_gain_error_db": 0.005774794582712859, "reference_mean_power": 0.002443196853891242, "output_mean_power": 0.0024464477228326215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.921742652983223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003035779346246488, "output_power": 0.00031898275246175923, "reference_active": true, "delta_db": 0.21496999254221164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.67325523785903e-05, "output_power": 4.4196532152704215e-05, "reference_active": true, "delta_db": -0.24231307311358374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.222466914385115e-06, "output_power": 1.1718856684164305e-06, "reference_active": true, "delta_db": -0.183518708520341}], "active_band_count": 3, "active_band_mae_db": 0.21360059139204546}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015437420870825604, "log_rms_mae_db": 1.6186621558133623, "rms_envelope_nrmse": 0.13863300533552722, "envelope_correlation": 0.9854122277987598, "rms_gain_error_db": 0.03118310346018604, "reference_mean_power": 0.0008784594297426283, "output_mean_power": 0.000884789620850305}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.115238295382225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001974652277483934, "output_power": 0.00020353165661395707, "reference_active": true, "delta_db": 0.13141337024180755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.173113277871683e-05, "output_power": 5.043112972739569e-05, "reference_active": true, "delta_db": -0.11053291376943533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0925495372248406e-05, "output_power": 1.1102320791426117e-05, "reference_active": true, "delta_db": 0.06972634310149567}], "active_band_count": 3, "active_band_mae_db": 0.1038908757042462}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012941193693393203, "log_rms_mae_db": 1.0882510637248568, "rms_envelope_nrmse": 0.08204533936570227, "envelope_correlation": 0.9945204025883454, "rms_gain_error_db": 0.03118717183815387, "reference_mean_power": 0.000882875660341548, "output_mean_power": 0.0008892385078888}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7308162041776265e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018289387957124914, "output_power": 0.0001901997430140493, "reference_active": true, "delta_db": 0.17010753482698643}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5176917167074748e-05, "output_power": 2.4601620124658157e-05, "reference_active": true, "delta_db": -0.10038842815184934}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.42204663895331e-07, "output_power": 7.624415419140916e-07, "reference_active": true, "delta_db": -0.9193869868396537}], "active_band_count": 3, "active_band_mae_db": 0.3966276499394965}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.92340600003817e-05, "log_rms_mae_db": 1.689654081033315, "rms_envelope_nrmse": 0.17553572698251396, "envelope_correlation": 0.9774350797175976, "rms_gain_error_db": -0.03342449834539025, "reference_mean_power": 0.000424662888055856, "output_mean_power": 0.0004214071103689792}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0384779490480225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014012639512762435, "output_power": 0.00013815947559076512, "reference_active": true, "delta_db": -0.0613927343161194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6627820389923143e-05, "output_power": 3.27656051112726e-05, "reference_active": true, "delta_db": -0.48392883026336153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.517673552823662e-06, "output_power": 8.010022027522802e-06, "reference_active": true, "delta_db": -0.26687280952890696}], "active_band_count": 3, "active_band_mae_db": 0.2707314580361293}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.48214657797314e-05, "log_rms_mae_db": 1.079631288603362, "rms_envelope_nrmse": 0.0987967252707256, "envelope_correlation": 0.9922636942053253, "rms_gain_error_db": -0.03342950089197465, "reference_mean_power": 0.0004267980543892402, "output_mean_power": 0.0004235254190983263}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.813635389636035e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013656501467579018, "output_power": 0.00013455195847742658, "reference_active": true, "delta_db": -0.0644943217997748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7671441991318854e-05, "output_power": 1.5260537811592752e-05, "reference_active": true, "delta_db": -0.6370215013636165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.379326657430796e-07, "output_power": 4.480810543882925e-07, "reference_active": true, "delta_db": -0.7937133583025225}], "active_band_count": 3, "active_band_mae_db": 0.49840972715530457}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.51121938209953e-05, "log_rms_mae_db": 1.974923429437559, "rms_envelope_nrmse": 0.1626255659135105, "envelope_correlation": 0.9833223603718654, "rms_gain_error_db": 0.13000665906661765, "reference_mean_power": 0.00015588925116025816, "output_mean_power": 0.00016062636701087149}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103456000035232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4374604191221866e-05, "output_power": 5.8228932112659314e-05, "reference_active": true, "delta_db": 0.2974271683691706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5832401077981513e-05, "output_power": 1.5415211410060753e-05, "reference_active": true, "delta_db": -0.11597298020888352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.903828543673632e-06, "output_power": 3.3764358449559398e-06, "reference_active": true, "delta_db": -0.6303223299615751}], "active_band_count": 3, "active_band_mae_db": 0.3479074928465431}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091623719145545e-05, "log_rms_mae_db": 1.2075028571411048, "rms_envelope_nrmse": 0.09485777496629175, "envelope_correlation": 0.9941794875179649, "rms_gain_error_db": 0.1299981251870267, "reference_mean_power": 0.00015667279701186957, "output_mean_power": 0.0001614334058023303}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1374269590951097e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.419500527522151e-05, "output_power": 5.790015094431331e-05, "reference_active": true, "delta_db": 0.28720433027307385}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661491941713126e-06, "output_power": 6.809588612420164e-06, "reference_active": true, "delta_db": 0.09549368807474375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7948760162947e-07, "output_power": 2.1752508911278828e-07, "reference_active": true, "delta_db": -1.088531916355895}], "active_band_count": 3, "active_band_mae_db": 0.49040997823457094}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7728375975227398, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ab42f167b8567bbba510533a0df3d15ebd95e60e8725e9993120c29334f7ee3e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9784163971260139, "standard_deviation": 0.002503161634390917, "minimum": 0.9729146770768315, "maximum": 0.9807822619027479, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7105835924414179, "standard_deviation": 0.03430753665179865, "minimum": 1.6680745652284679, "maximum": 1.763165604929127, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5203647678191532, "standard_deviation": 0.3389982722700943, "minimum": 0.2531031377152592, "maximum": 1.2956887530235126, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1721004971565538, "standard_deviation": 0.009284600444295358, "minimum": 0.1621339809804912, "maximum": 0.19251438921841088, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9926265052857135, "standard_deviation": 0.001239031528660376, "minimum": 0.9913816669254554, "maximum": 0.9943355362950179, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0812116697943637, "standard_deviation": 0.028147200458382485, "minimum": 1.0267860775085405, "maximum": 1.1097188294557843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6324163896223943, "standard_deviation": 0.39124137960639754, "minimum": 0.18683680026043717, "maximum": 1.4304545831802766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09975538127353328, "standard_deviation": 0.006938958650626066, "minimum": 0.0870946692760759, "maximum": 0.10899135240842345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9825017916296976, "standard_deviation": 0.002201124253404451, "minimum": 0.9793265031605045, "maximum": 0.9853644273712721, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.007391599823851, "standard_deviation": 0.02517131839338833, "minimum": 1.974923429437559, "maximum": 2.0538340282293235, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.41736820384825424, "standard_deviation": 0.25165382307214573, "minimum": 0.13185231389996058, "maximum": 0.9242953970350415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16505114248696362, "standard_deviation": 0.009566681716316437, "minimum": 0.1501238992206015, "maximum": 0.1781819752773643, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944690853372116, "standard_deviation": 0.000989331253658401, "minimum": 0.9928495718038008, "maximum": 0.995459066333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.198962226286, "standard_deviation": 0.021903877692914867, "minimum": 1.16664205068153, "maximum": 1.2266928319374022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5504771157882455, "standard_deviation": 0.26556293073637616, "minimum": 0.15327113523015543, "maximum": 0.8827675755445524, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0916591190761793, "standard_deviation": 0.006873110204698858, "minimum": 0.08564509109484654, "maximum": 0.10237954275397373, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9874073728234903, "standard_deviation": 0.0016006555537314714, "minimum": 0.9854122277987598, "maximum": 0.9900797210848252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6272959233088482, "standard_deviation": 0.044516726287060375, "minimum": 1.542346778219775, "maximum": 1.6841926976455421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.28830818147217596, "standard_deviation": 0.1585907387221352, "minimum": 0.07021194174500743, "maximum": 0.5549067182903261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12835307582514016, "standard_deviation": 0.00833487250955144, "minimum": 0.11410283210367826, "maximum": 0.13863300533552722, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954710954969872, "standard_deviation": 0.0010643336941125504, "minimum": 0.9941170384146321, "maximum": 0.996702262462439, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.046671761610251, "standard_deviation": 0.05247293423736353, "minimum": 0.9767603452274147, "maximum": 1.1143896403610993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4365928340178131, "standard_deviation": 0.22573341909654665, "minimum": 0.13829629397484258, "maximum": 0.7018048868501685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07431473050830906, "standard_deviation": 0.009033061697874479, "minimum": 0.0634525834487511, "maximum": 0.08642340975790837, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17769226404456817, "standard_deviation": 0.03946094462916587, "minimum": 0.12265206074331164, "maximum": 0.2423575162578986, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.981088145230219, "standard_deviation": 0.009430137516623565, "minimum": 5.965848985593219, "maximum": 5.990055487916692, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5022014673478963, "standard_deviation": 0.023194181244617303, "minimum": 0.4665098542188126, "maximum": 0.5435638943314531, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07526206745958136, "standard_deviation": 0.03851353912855744, "minimum": -0.11811551864520249, "maximum": 0.0056594564428315616, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920217975660433, "standard_deviation": 0.0008100793440928121, "minimum": 0.990595686790113, "maximum": 0.992982710607254, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4392418646928145, "standard_deviation": 0.060624996092014403, "minimum": 1.3544679143997547, "maximum": 1.5213805959142515, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.19270033457227045, "standard_deviation": 0.06250202650051853, "minimum": 0.10742409289044591, "maximum": 0.3020739147853142, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09179384737066136, "standard_deviation": 0.004598914665135412, "minimum": 0.08641372673214642, "maximum": 0.09969717471468789, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976603026145228, "standard_deviation": 0.0002934665696386104, "minimum": 0.9972173520391215, "maximum": 0.9981420650407073, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7514757869026122, "standard_deviation": 0.06061245347372473, "minimum": 0.6704158981564425, "maximum": 0.8395531026754293, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.27528070058456866, "standard_deviation": 0.0906131949179039, "minimum": 0.1585975435823088, "maximum": 0.44708990535403026, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04842873121922237, "standard_deviation": 0.0028858584474143324, "minimum": 0.0432590228404348, "maximum": 0.05203350735416465, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0002", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5930912494659424, "predicted_mos": 2.5930912494659424, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2178556cf271a38bfd429fb0fb067a5337acd893e8cd44fdddf8e2d785f48b32", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.903061999939382, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4238164956892085, "全帯域_10ms_rms_envelope_nrmse": 0.09768180320169624, "全帯域_10ms_envelope_correlation": 0.990959890417419, "全帯域_10ms_modulation_mae_db": 0.13495965841140645, "全帯域_40ms_log_rms_mae_db": 0.7786853883578082, "全帯域_40ms_rms_envelope_nrmse": 0.05340211878804564, "全帯域_40ms_envelope_correlation": 0.9970661584852673, "全帯域_40ms_modulation_mae_db": 0.2762173277028841, "300–1000 Hz_10ms_log_rms_mae_db": 1.6532704035370995, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14062923048083295, "300–1000 Hz_10ms_envelope_correlation": 0.9850397665949019, "300–1000 Hz_10ms_modulation_mae_db": 0.16156345016027526, "300–1000 Hz_40ms_log_rms_mae_db": 1.0658573711117598, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08228671770156627, "300–1000 Hz_40ms_envelope_correlation": 0.9945326761624674, "300–1000 Hz_40ms_modulation_mae_db": 0.5579630582575513, "1000–3000 Hz_10ms_log_rms_mae_db": 1.677419504978893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17766544526498218, "1000–3000 Hz_10ms_envelope_correlation": 0.9768382970942681, "1000–3000 Hz_10ms_modulation_mae_db": 0.22708575669177752, "1000–3000 Hz_40ms_log_rms_mae_db": 1.048260612126243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09984130637846257, "1000–3000 Hz_40ms_envelope_correlation": 0.99210645195345, "1000–3000 Hz_40ms_modulation_mae_db": 0.4754809607435286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8930507758101136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15705991582176865, "3000–6000 Hz_10ms_envelope_correlation": 0.9843792596856267, "3000–6000 Hz_10ms_modulation_mae_db": 0.5135427067769714, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1186709240651367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09457064213471267, "3000–6000 Hz_40ms_envelope_correlation": 0.9942029266238543, "3000–6000 Hz_40ms_modulation_mae_db": 0.6027335645486329, "mr_stft_db": 5.933486717585886, "band_mae_db": 0.10675770442336796, "rms_gain_error_db": -0.0011405744529001374, "output_peak": 0.4929828801387704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5118972839667304, "全帯域_10ms_rms_envelope_nrmse": 0.09780721465776104, "全帯域_10ms_envelope_correlation": 0.9910181272700482, "全帯域_10ms_modulation_mae_db": 0.26538302629368793, "全帯域_40ms_log_rms_mae_db": 0.791017577939534, "全帯域_40ms_rms_envelope_nrmse": 0.05389136493731616, "全帯域_40ms_envelope_correlation": 0.9972203660941689, "全帯域_40ms_modulation_mae_db": 0.387907313997838, "300–1000 Hz_10ms_log_rms_mae_db": 1.6650828277188046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13604765468354865, "300–1000 Hz_10ms_envelope_correlation": 0.9859026967930161, "300–1000 Hz_10ms_modulation_mae_db": 0.5150063741078014, "300–1000 Hz_40ms_log_rms_mae_db": 1.0690857627122912, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0860261115136399, "300–1000 Hz_40ms_envelope_correlation": 0.9941412953524207, "300–1000 Hz_40ms_modulation_mae_db": 0.6611795019695909, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7732909724570032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1816671419984887, "1000–3000 Hz_10ms_envelope_correlation": 0.9766816557786677, "1000–3000 Hz_10ms_modulation_mae_db": 0.25364543017608326, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0883817259066133, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10986944230493678, "1000–3000 Hz_40ms_envelope_correlation": 0.9922333988561638, "1000–3000 Hz_40ms_modulation_mae_db": 0.26780385396181633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9204672647756245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1601563626461323, "3000–6000 Hz_10ms_envelope_correlation": 0.9838905767458412, "3000–6000 Hz_10ms_modulation_mae_db": 0.6389226787217995, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1374105397043706, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08770460695622187, "3000–6000 Hz_40ms_envelope_correlation": 0.9950992719142708, "3000–6000 Hz_40ms_modulation_mae_db": 0.9140602766226277, "mr_stft_db": 5.930455358734693, "band_mae_db": 0.23771034157535167, "rms_gain_error_db": -0.13672602996898325, "output_peak": 0.5392631696578325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3888123962469061, "全帯域_10ms_rms_envelope_nrmse": 0.08748074013194314, "全帯域_10ms_envelope_correlation": 0.9928010608818683, "全帯域_10ms_modulation_mae_db": 0.13311292512499495, "全帯域_40ms_log_rms_mae_db": 0.6990523923115518, "全帯域_40ms_rms_envelope_nrmse": 0.0476219998342709, "全帯域_40ms_envelope_correlation": 0.9978043419118087, "全帯域_40ms_modulation_mae_db": 0.3153794592968987, "300–1000 Hz_10ms_log_rms_mae_db": 1.5553303436783978, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11686528399579214, "300–1000 Hz_10ms_envelope_correlation": 0.9896038648698569, "300–1000 Hz_10ms_modulation_mae_db": 0.12141876875789077, "300–1000 Hz_40ms_log_rms_mae_db": 0.9515799610614384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0655032339559547, "300–1000 Hz_40ms_envelope_correlation": 0.9965068630656927, "300–1000 Hz_40ms_modulation_mae_db": 0.3610534201711939, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7115595162956103, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1699902357548137, "1000–3000 Hz_10ms_envelope_correlation": 0.979008774210889, "1000–3000 Hz_10ms_modulation_mae_db": 0.5394955566801621, "1000–3000 Hz_40ms_log_rms_mae_db": 1.035568555786666, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10007558483813798, "1000–3000 Hz_40ms_envelope_correlation": 0.9926877291205563, "1000–3000 Hz_40ms_modulation_mae_db": 1.053660114409446, "3000–6000 Hz_10ms_log_rms_mae_db": 1.940530473580136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17498044843421143, "3000–6000 Hz_10ms_envelope_correlation": 0.9800755318237531, "3000–6000 Hz_10ms_modulation_mae_db": 0.13087135842665484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1176275158495486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10131413540498786, "3000–6000 Hz_40ms_envelope_correlation": 0.9930354080077061, "3000–6000 Hz_40ms_modulation_mae_db": 0.2076348897253172, "mr_stft_db": 5.94567035749592, "band_mae_db": 0.15075945198926688, "rms_gain_error_db": -0.1084651919009025, "output_peak": 0.4632921738791468}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5347393003142022, "全帯域_10ms_rms_envelope_nrmse": 0.09449444333295066, "全帯域_10ms_envelope_correlation": 0.9915985463909031, "全帯域_10ms_modulation_mae_db": 0.2813333800894609, "全帯域_40ms_log_rms_mae_db": 0.8628116407569927, "全帯域_40ms_rms_envelope_nrmse": 0.051597358951481466, "全帯域_40ms_envelope_correlation": 0.9973740268566884, "全帯域_40ms_modulation_mae_db": 0.3330138759173475, "300–1000 Hz_10ms_log_rms_mae_db": 1.6932813145682293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12967659304457455, "300–1000 Hz_10ms_envelope_correlation": 0.9873100296615828, "300–1000 Hz_10ms_modulation_mae_db": 0.18505711129676575, "300–1000 Hz_40ms_log_rms_mae_db": 1.0476665688936735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07300537659104996, "300–1000 Hz_40ms_envelope_correlation": 0.9956941365175828, "300–1000 Hz_40ms_modulation_mae_db": 0.11347346187723772, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6637141516431173, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17699381265371622, "1000–3000 Hz_10ms_envelope_correlation": 0.9769966496747238, "1000–3000 Hz_10ms_modulation_mae_db": 0.2783775324380328, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0005400177278574, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10516773714967086, "1000–3000 Hz_40ms_envelope_correlation": 0.9912628315304068, "1000–3000 Hz_40ms_modulation_mae_db": 0.3503597889860817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9395777456383048, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16390193764079491, "3000–6000 Hz_10ms_envelope_correlation": 0.9825697193815709, "3000–6000 Hz_10ms_modulation_mae_db": 0.8654775694995408, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1781382926405117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08903906597075363, "3000–6000 Hz_40ms_envelope_correlation": 0.9948650725211016, "3000–6000 Hz_40ms_modulation_mae_db": 0.7936195244029279, "mr_stft_db": 5.923004953403989, "band_mae_db": 0.1547471950180794, "rms_gain_error_db": -0.09124947517272014, "output_peak": 0.49188586023589986}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3699714057752568, "全帯域_10ms_rms_envelope_nrmse": 0.08895244675411039, "全帯域_10ms_envelope_correlation": 0.992511714930162, "全帯域_10ms_modulation_mae_db": 0.2157439854469825, "全帯域_40ms_log_rms_mae_db": 0.6637489532009889, "全帯域_40ms_rms_envelope_nrmse": 0.043228777371793085, "全帯域_40ms_envelope_correlation": 0.9981489346664034, "全帯域_40ms_modulation_mae_db": 0.35497283641072236, "300–1000 Hz_10ms_log_rms_mae_db": 1.6775609704139949, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12760626664409166, "300–1000 Hz_10ms_envelope_correlation": 0.9876103647345191, "300–1000 Hz_10ms_modulation_mae_db": 0.262332238394942, "300–1000 Hz_40ms_log_rms_mae_db": 1.013113830253323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06878513312756891, "300–1000 Hz_40ms_envelope_correlation": 0.9961288498801426, "300–1000 Hz_40ms_modulation_mae_db": 0.21736155804596557, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7316352684951426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1752426994721374, "1000–3000 Hz_10ms_envelope_correlation": 0.9782696005298775, "1000–3000 Hz_10ms_modulation_mae_db": 1.290461514341353, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0678232907704683, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09873105143441943, "1000–3000 Hz_40ms_envelope_correlation": 0.9938824777023911, "1000–3000 Hz_40ms_modulation_mae_db": 1.4332998569824411, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0074383843772883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14372075014028876, "3000–6000 Hz_10ms_envelope_correlation": 0.9866329161563018, "3000–6000 Hz_10ms_modulation_mae_db": 0.4934908743411787, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1820755177601756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08119698250222969, "3000–6000 Hz_40ms_envelope_correlation": 0.9957043591781078, "3000–6000 Hz_40ms_modulation_mae_db": 0.6458380754501332, "mr_stft_db": 5.944014661668175, "band_mae_db": 0.2208724488923615, "rms_gain_error_db": -0.08055788944141098, "output_peak": 0.4919719205891108}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4543501595546509, "全帯域_10ms_rms_envelope_nrmse": 0.089928316265276, "全帯域_10ms_envelope_correlation": 0.9923294684834842, "全帯域_10ms_modulation_mae_db": 0.1536615089645108, "全帯域_40ms_log_rms_mae_db": 0.7287824902413712, "全帯域_40ms_rms_envelope_nrmse": 0.046795314092648345, "全帯域_40ms_envelope_correlation": 0.9977972600579287, "全帯域_40ms_modulation_mae_db": 0.18070632765973768, "300–1000 Hz_10ms_log_rms_mae_db": 1.633722615831582, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12475946507889714, "300–1000 Hz_10ms_envelope_correlation": 0.9881035576522007, "300–1000 Hz_10ms_modulation_mae_db": 0.3364192632641938, "300–1000 Hz_40ms_log_rms_mae_db": 0.9686306894215093, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06777639680397034, "300–1000 Hz_40ms_envelope_correlation": 0.9962863357123037, "300–1000 Hz_40ms_modulation_mae_db": 0.5834917892506212, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6882234473768813, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1695795602473222, "1000–3000 Hz_10ms_envelope_correlation": 0.9789750076201245, "1000–3000 Hz_10ms_modulation_mae_db": 0.4403039205921204, "1000–3000 Hz_40ms_log_rms_mae_db": 1.025973625256269, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08752559702430279, "1000–3000 Hz_40ms_envelope_correlation": 0.9943827991868268, "1000–3000 Hz_40ms_modulation_mae_db": 0.4655569917345288, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9136925040344144, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15791596900514346, "3000–6000 Hz_10ms_envelope_correlation": 0.9845400749106975, "3000–6000 Hz_10ms_modulation_mae_db": 0.19378162186312595, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0996885017199776, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08874215798087677, "3000–6000 Hz_40ms_envelope_correlation": 0.9951381156309007, "3000–6000 Hz_40ms_modulation_mae_db": 0.5540859425576276, "mr_stft_db": 5.943055027175763, "band_mae_db": 0.1355814585056735, "rms_gain_error_db": -0.06847669575215823, "output_peak": 0.52717660533708}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4459054449087267, "全帯域_10ms_rms_envelope_nrmse": 0.09853007507644541, "全帯域_10ms_envelope_correlation": 0.9908352465715121, "全帯域_10ms_modulation_mae_db": 0.24594858142710052, "全帯域_40ms_log_rms_mae_db": 0.7282549244732575, "全帯域_40ms_rms_envelope_nrmse": 0.04935283340553015, "全帯域_40ms_envelope_correlation": 0.9975930843205741, "全帯域_40ms_modulation_mae_db": 0.15802719528612905, "300–1000 Hz_10ms_log_rms_mae_db": 1.6679517763955118, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13666798429701352, "300–1000 Hz_10ms_envelope_correlation": 0.9857016888232867, "300–1000 Hz_10ms_modulation_mae_db": 0.5872409987875267, "300–1000 Hz_40ms_log_rms_mae_db": 1.0651049830109935, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08406266980151239, "300–1000 Hz_40ms_envelope_correlation": 0.9942573566497904, "300–1000 Hz_40ms_modulation_mae_db": 0.8518559342944908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7161798653477547, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18798091935825015, "1000–3000 Hz_10ms_envelope_correlation": 0.9743593146499797, "1000–3000 Hz_10ms_modulation_mae_db": 0.6949925678574708, "1000–3000 Hz_40ms_log_rms_mae_db": 1.081494346698187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10282175102207497, "1000–3000 Hz_40ms_envelope_correlation": 0.9926896108724695, "1000–3000 Hz_40ms_modulation_mae_db": 0.7659170330413, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9419671145364679, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17902828118435216, "3000–6000 Hz_10ms_envelope_correlation": 0.9792268967639178, "3000–6000 Hz_10ms_modulation_mae_db": 0.2602780269634101, "3000–6000 Hz_40ms_log_rms_mae_db": 1.171945506899768, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10163524537526672, "3000–6000 Hz_40ms_envelope_correlation": 0.993004125951912, "3000–6000 Hz_40ms_modulation_mae_db": 0.28973304158298, "mr_stft_db": 5.943036670339571, "band_mae_db": 0.25973552862516813, "rms_gain_error_db": -0.0860185185329487, "output_peak": 0.49014739609789176}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4302273404131156, "全帯域_10ms_rms_envelope_nrmse": 0.08828793515153846, "全帯域_10ms_envelope_correlation": 0.992620413872951, "全帯域_10ms_modulation_mae_db": 0.3340361415531113, "全帯域_40ms_log_rms_mae_db": 0.8001693716532605, "全帯域_40ms_rms_envelope_nrmse": 0.04784022218693453, "全帯域_40ms_envelope_correlation": 0.9977024580502135, "全帯域_40ms_modulation_mae_db": 0.4415065673588954, "300–1000 Hz_10ms_log_rms_mae_db": 1.6293247798479673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12016430526411788, "300–1000 Hz_10ms_envelope_correlation": 0.9889956402967444, "300–1000 Hz_10ms_modulation_mae_db": 0.5072330637768162, "300–1000 Hz_40ms_log_rms_mae_db": 0.9768876435433657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0630167809171158, "300–1000 Hz_40ms_envelope_correlation": 0.9967591752436519, "300–1000 Hz_40ms_modulation_mae_db": 0.8982905447328986, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6729735294908141, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16806749213922026, "1000–3000 Hz_10ms_envelope_correlation": 0.979270066179433, "1000–3000 Hz_10ms_modulation_mae_db": 0.5398141869705803, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0576705327039106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10889055937384384, "1000–3000 Hz_40ms_envelope_correlation": 0.9907659756272381, "1000–3000 Hz_40ms_modulation_mae_db": 0.6577398372358131, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9575033610587116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1635902156514252, "3000–6000 Hz_10ms_envelope_correlation": 0.9825689714119226, "3000–6000 Hz_10ms_modulation_mae_db": 0.5250092889790795, "3000–6000 Hz_40ms_log_rms_mae_db": 1.150176269265811, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08264903435049695, "3000–6000 Hz_40ms_envelope_correlation": 0.9953809901332991, "3000–6000 Hz_40ms_modulation_mae_db": 0.7916689399529785, "mr_stft_db": 5.933458944306209, "band_mae_db": 0.18649249789435451, "rms_gain_error_db": -0.07471663940110027, "output_peak": 0.5026776692275206}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.150792574486069, "spectral_convergence": 0.3416238429306199, "spectral_convergence_status": "ok", "amplitude_floor": 1.3425996786840756e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.741090443910569, "spectral_convergence": 0.32880104242062363, "spectral_convergence_status": "ok", "amplitude_floor": 1.1106411258473627e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9085771343610185, "spectral_convergence": 0.3509151558081843, "spectral_convergence_status": "ok", "amplitude_floor": 1.0520318003394391e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00033395677419724207, "output_power": 0.00034684667329845843, "delta_db": 0.1644727601515939, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008500397987721807, "output_power": 0.0008343223139244897, "delta_db": -0.08105400879725974, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006268367014596879, "output_power": 0.0006269957249349041, "delta_db": 0.0011016307553691018, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003695716856827319, "output_power": 0.00037085011633397386, "delta_db": 0.014997287426201438, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013451399439061329, "output_power": 0.00012950158499262058, "delta_db": -0.1649238539273837, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011413253897640505, "output_power": 0.00011989722931482846, "delta_db": 0.2139966854823999, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10675770442336796, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0011405744529001374, "pitch": null, "output_rms_dbfs": -26.146773359309453, "output_peak": 0.4929828801387704, "reference_rms_dbfs": -26.14563278485655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.933486717585886}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018099058448312497, "log_rms_mae_db": 1.4238164956892085, "rms_envelope_nrmse": 0.09768180320169624, "envelope_correlation": 0.990959890417419, "rms_gain_error_db": -0.0011422867781375123, "reference_mean_power": 0.002431072908198481, "output_mean_power": 0.0024304335684452163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.757963657265947e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003340894048273399, "output_power": 0.00034992787043505155, "reference_active": true, "delta_db": 0.20115830956249148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010057920506755965, "output_power": 9.652140458821717e-05, "reference_active": true, "delta_db": -0.178845656564227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.918423836914529e-05, "output_power": 1.9294434879885167e-05, "reference_active": true, "delta_db": 0.024875009107500838}], "active_band_count": 3, "active_band_mae_db": 0.13495965841140645}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016835992502135956, "log_rms_mae_db": 0.7786853883578082, "rms_envelope_nrmse": 0.05340211878804564, "envelope_correlation": 0.9970661584852673, "rms_gain_error_db": -0.0010237980744757405, "reference_mean_power": 0.002443196853891242, "output_mean_power": 0.002442620966899531}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.921742652983223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003035779346246488, "output_power": 0.00032026591127511957, "reference_active": true, "delta_db": 0.23240513389222234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.67325523785903e-05, "output_power": 4.39177285838464e-05, "reference_active": true, "delta_db": -0.26979631011939237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.222466914385115e-06, "output_power": 1.1339451818428462e-06, "reference_active": true, "delta_db": -0.3264505390970376}], "active_band_count": 3, "active_band_mae_db": 0.2762173277028841}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015437420870825604, "log_rms_mae_db": 1.6532704035370995, "rms_envelope_nrmse": 0.14062923048083295, "envelope_correlation": 0.9850397665949019, "rms_gain_error_db": 0.05154135692158569, "reference_mean_power": 0.0008784594297426283, "output_mean_power": 0.0008889469511973474}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.115238295382225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001974652277483934, "output_power": 0.0002044663484688993, "reference_active": true, "delta_db": 0.15131210601895076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.173113277871683e-05, "output_power": 4.9668751047547504e-05, "reference_active": true, "delta_db": -0.1766874892449754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0925495372248406e-05, "output_power": 1.0538336498960081e-05, "reference_active": true, "delta_db": -0.15669075521689962}], "active_band_count": 3, "active_band_mae_db": 0.16156345016027526}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012941193693393203, "log_rms_mae_db": 1.0658573711117598, "rms_envelope_nrmse": 0.08228671770156627, "envelope_correlation": 0.9945326761624674, "rms_gain_error_db": 0.051543679904074366, "reference_mean_power": 0.000882875660341548, "output_mean_power": 0.0008934163830101711}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7308162041776265e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018289387957124914, "output_power": 0.00019058097721444574, "reference_active": true, "delta_db": 0.17880377158847416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5176917167074748e-05, "output_power": 2.4407625000142747e-05, "reference_active": true, "delta_db": -0.13477028892522838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.42204663895331e-07, "output_power": 6.888326724593421e-07, "reference_active": true, "delta_db": -1.3603151142589516}], "active_band_count": 3, "active_band_mae_db": 0.5579630582575513}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.92340600003817e-05, "log_rms_mae_db": 1.677419504978893, "rms_envelope_nrmse": 0.17766544526498218, "envelope_correlation": 0.9768382970942681, "rms_gain_error_db": -0.06382418020608838, "reference_mean_power": 0.000424662888055856, "output_mean_power": 0.00041846765081351506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0384779490480225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014012639512762435, "output_power": 0.0001382559588399644, "reference_active": true, "delta_db": -0.058360910531649005}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6627820389923143e-05, "output_power": 3.316349018461687e-05, "reference_active": true, "delta_db": -0.4315084568483099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.517673552823662e-06, "output_power": 8.150461514548271e-06, "reference_active": true, "delta_db": -0.1913879026953737}], "active_band_count": 3, "active_band_mae_db": 0.22708575669177752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.48214657797314e-05, "log_rms_mae_db": 1.048260612126243, "rms_envelope_nrmse": 0.09984130637846257, "envelope_correlation": 0.99210645195345, "rms_gain_error_db": -0.0638280136901652, "reference_mean_power": 0.0004267980543892402, "output_mean_power": 0.000420571296822932}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.813635389636035e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013656501467579018, "output_power": 0.00013468098272741483, "reference_active": true, "delta_db": -0.06033179073351777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7671441991318854e-05, "output_power": 1.5321514468406195e-05, "reference_active": true, "delta_db": -0.6197029378940219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.379326657430796e-07, "output_power": 4.529884221280118e-07, "reference_active": true, "delta_db": -0.7464081536030459}], "active_band_count": 3, "active_band_mae_db": 0.4754809607435286}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.51121938209953e-05, "log_rms_mae_db": 1.8930507758101136, "rms_envelope_nrmse": 0.15705991582176865, "envelope_correlation": 0.9843792596856267, "rms_gain_error_db": 0.0899154055879567, "reference_mean_power": 0.00015588925116025816, "output_mean_power": 0.0001591503915407937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103456000035232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4374604191221866e-05, "output_power": 5.8673187839728905e-05, "reference_active": true, "delta_db": 0.330435763884216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5832401077981513e-05, "output_power": 1.4928158542018059e-05, "reference_active": true, "delta_db": -0.255405444246067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.903828543673632e-06, "output_power": 3.133372618561833e-06, "reference_active": true, "delta_db": -0.9547869122006313}], "active_band_count": 3, "active_band_mae_db": 0.5135427067769714}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091623719145545e-05, "log_rms_mae_db": 1.1186709240651367, "rms_envelope_nrmse": 0.09457064213471267, "envelope_correlation": 0.9942029266238543, "rms_gain_error_db": 0.08990792656548569, "reference_mean_power": 0.00015667279701186957, "output_mean_power": 0.00015995005340460768}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1374269590951097e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.419500527522151e-05, "output_power": 5.822958525523215e-05, "reference_active": true, "delta_db": 0.31184433925715654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661491941713126e-06, "output_power": 6.375260073318851e-06, "reference_active": true, "delta_db": -0.19073600729635684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7948760162947e-07, "output_power": 2.069190673176593e-07, "reference_active": true, "delta_db": -1.3056203470923855}], "active_band_count": 3, "active_band_mae_db": 0.6027335645486329}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7735377522108721, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a1a3b59cc9c1d96bb6404ad9c7125510e42c940e7b5783ae6584e9bc1dd3b1e1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9775499207172453, "standard_deviation": 0.0016640026505721035, "minimum": 0.9743593146499797, "maximum": 0.979270066179433, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.704374532010652, "standard_deviation": 0.03646981906778939, "minimum": 1.6637141516431173, "maximum": 1.7732909724570032, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5330220582184475, "standard_deviation": 0.3474654809829055, "minimum": 0.22708575669177752, "maximum": 1.290461514341353, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17589841336111633, "standard_deviation": 0.006763085071717183, "minimum": 0.16806749213922026, "maximum": 0.18798091935825015, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9925014093561878, "standard_deviation": 0.0012124790762986822, "minimum": 0.9907659756272381, "maximum": 0.9943827991868268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0507140883720267, "standard_deviation": 0.029452932737036332, "minimum": 1.0005400177278574, "maximum": 1.0883817259066133, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6837273046368695, "standard_deviation": 0.393015964303646, "minimum": 0.26780385396181633, "maximum": 1.4332998569824411, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10161537869073115, "standard_deviation": 0.0070456875741597815, "minimum": 0.08752559702430279, "maximum": 0.10986944230493678, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.982985493359954, "standard_deviation": 0.002432619411469682, "minimum": 0.9792268967639178, "maximum": 0.9866329161563018, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9392784529763825, "standard_deviation": 0.03408457815668468, "minimum": 1.8930507758101136, "maximum": 2.0074383843772883, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4526717656964701, "standard_deviation": 0.24590857244868702, "minimum": 0.13087135842665484, "maximum": 0.8654775694995408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1625442350655146, "standard_deviation": 0.01095995786477236, "minimum": 0.14372075014028876, "maximum": 0.17902828118435216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.994553783745144, "standard_deviation": 0.0010401923523131586, "minimum": 0.993004125951912, "maximum": 0.9957043591781078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1444666334881624, "standard_deviation": 0.031113900511629943, "minimum": 1.0996885017199776, "maximum": 1.1820755177601756, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5999217818554031, "standard_deviation": 0.24706620819804517, "minimum": 0.2076348897253172, "maximum": 0.9140602766226277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09085648383444327, "standard_deviation": 0.007724867553045442, "minimum": 0.08119698250222969, "maximum": 0.10163524537526672, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9872834511782635, "standard_deviation": 0.001627383532142815, "minimum": 0.9850397665949019, "maximum": 0.9896038648698569, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6469406289989483, "standard_deviation": 0.042717036702225784, "minimum": 1.5553303436783978, "maximum": 1.6932813145682293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3345339085682765, "standard_deviation": 0.1809313103416833, "minimum": 0.12141876875789077, "maximum": 0.5872409987875267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12905209793610856, "standard_deviation": 0.008361373340165866, "minimum": 0.11686528399579214, "maximum": 0.14062923048083295, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955383360730066, "standard_deviation": 0.0010667011580987537, "minimum": 0.9941412953524207, "maximum": 0.9967591752436519, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0197408512510444, "standard_deviation": 0.04860610522968489, "minimum": 0.9515799610614384, "maximum": 1.0690857627122912, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5305836585749437, "standard_deviation": 0.28280682644487365, "minimum": 0.11347346187723772, "maximum": 0.8982905447328986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07380780255154729, "standard_deviation": 0.00905540356960578, "minimum": 0.0630167809171158, "maximum": 0.0860261115136399, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18158207836545293, "standard_deviation": 0.05373357340866541, "minimum": 0.10675770442336796, "maximum": 0.25973552862516813, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.937022836338776, "standard_deviation": 0.00811574339162442, "minimum": 5.923004953403989, "maximum": 5.94567035749592, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.49992470939540656, "standard_deviation": 0.023638215211735977, "minimum": 0.4632921738791468, "maximum": 0.5392631696578325, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08091887682789052, "standard_deviation": 0.03884759393955044, "minimum": -0.13672602996898325, "maximum": -0.0011405744529001374, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9918343086022935, "standard_deviation": 0.0008229825547889875, "minimum": 0.9908352465715121, "maximum": 0.9928010608818683, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4449649783585996, "standard_deviation": 0.0561502235310095, "minimum": 1.3699714057752568, "maximum": 1.5347393003142022, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2205224009139069, "standard_deviation": 0.07435140728424497, "minimum": 0.13311292512499495, "maximum": 0.3340361415531113, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09289537182146518, "standard_deviation": 0.004724402027732649, "minimum": 0.08748074013194314, "maximum": 0.09853007507644541, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975883288053816, "standard_deviation": 0.0003528708154079856, "minimum": 0.9970661584852673, "maximum": 0.9981489346664034, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7565653423668456, "standard_deviation": 0.06365288504021868, "minimum": 0.6637489532009889, "maximum": 0.8628116407569927, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3059663629538066, "standard_deviation": 0.09766440652968676, "minimum": 0.15802719528612905, "maximum": 0.4415065673588954, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04921624869600254, "standard_deviation": 0.0036097561715760116, "minimum": 0.043228777371793085, "maximum": 0.05389136493731616, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0002", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.464322090148926, "predicted_mos": 2.464322090148926, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4c15ebcfba60fdc8155d9d69816d89acd59be34760aecc52d26a1dd261efebc0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.84831009991467, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4346520000117153, "全帯域_10ms_rms_envelope_nrmse": 0.09668805193454395, "全帯域_10ms_envelope_correlation": 0.9913095669485932, "全帯域_10ms_modulation_mae_db": 0.10219224460831165, "全帯域_40ms_log_rms_mae_db": 0.7225013625186534, "全帯域_40ms_rms_envelope_nrmse": 0.05596733119576833, "全帯域_40ms_envelope_correlation": 0.9970916416677889, "全帯域_40ms_modulation_mae_db": 0.12564371764293333, "300–1000 Hz_10ms_log_rms_mae_db": 1.8664152367370144, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1317165142616717, "300–1000 Hz_10ms_envelope_correlation": 0.9867824998943647, "300–1000 Hz_10ms_modulation_mae_db": 0.14502317112387916, "300–1000 Hz_40ms_log_rms_mae_db": 1.2366234437748562, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0762147962349473, "300–1000 Hz_40ms_envelope_correlation": 0.995324661985446, "300–1000 Hz_40ms_modulation_mae_db": 0.14988164755147357, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0271702499719675, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19182382816762306, "1000–3000 Hz_10ms_envelope_correlation": 0.9737805678751207, "1000–3000 Hz_10ms_modulation_mae_db": 0.36570806886314994, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2905851223154505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12920634421382649, "1000–3000 Hz_40ms_envelope_correlation": 0.9883790060375651, "1000–3000 Hz_40ms_modulation_mae_db": 0.40062548080853233, "3000–6000 Hz_10ms_log_rms_mae_db": 2.673395743823498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18898914577166856, "3000–6000 Hz_10ms_envelope_correlation": 0.9785487605163302, "3000–6000 Hz_10ms_modulation_mae_db": 1.0379697234174863, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6419294160572733, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11563161651408366, "3000–6000 Hz_40ms_envelope_correlation": 0.9935962903716408, "3000–6000 Hz_40ms_modulation_mae_db": 1.2676505110946878, "mr_stft_db": 6.331604214851772, "band_mae_db": 0.29158725970652477, "rms_gain_error_db": -0.16610159553763745, "output_peak": 0.469351753294972}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2751044311096926, "全帯域_10ms_rms_envelope_nrmse": 0.08953090272365195, "全帯域_10ms_envelope_correlation": 0.9924991884123024, "全帯域_10ms_modulation_mae_db": 0.1825282587812448, "全帯域_40ms_log_rms_mae_db": 0.635872713324416, "全帯域_40ms_rms_envelope_nrmse": 0.04878345501317159, "全帯域_40ms_envelope_correlation": 0.9977353344508278, "全帯域_40ms_modulation_mae_db": 0.13514553296842582, "300–1000 Hz_10ms_log_rms_mae_db": 1.8382773947940736, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12072057416198613, "300–1000 Hz_10ms_envelope_correlation": 0.9889526723979422, "300–1000 Hz_10ms_modulation_mae_db": 0.6099929150879456, "300–1000 Hz_40ms_log_rms_mae_db": 1.2132201506979425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0710557889195791, "300–1000 Hz_40ms_envelope_correlation": 0.996066682470067, "300–1000 Hz_40ms_modulation_mae_db": 0.8596031267900542, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0353802310565783, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18648815958778578, "1000–3000 Hz_10ms_envelope_correlation": 0.9743987475997791, "1000–3000 Hz_10ms_modulation_mae_db": 0.49077667261390623, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3126217394899282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10584234670429093, "1000–3000 Hz_40ms_envelope_correlation": 0.9913588072546612, "1000–3000 Hz_40ms_modulation_mae_db": 0.49145965198056274, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6652295849576824, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1805776376590473, "3000–6000 Hz_10ms_envelope_correlation": 0.9790242855149197, "3000–6000 Hz_10ms_modulation_mae_db": 0.8402449895941256, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6361626976686154, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10068825214485191, "3000–6000 Hz_40ms_envelope_correlation": 0.9934946309436061, "3000–6000 Hz_40ms_modulation_mae_db": 0.9749413778415438, "mr_stft_db": 6.307711756119327, "band_mae_db": 0.14337989822882982, "rms_gain_error_db": -0.12505953229398106, "output_peak": 0.5423913869664874}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.371952400316347, "全帯域_10ms_rms_envelope_nrmse": 0.08894558868757782, "全帯域_10ms_envelope_correlation": 0.9925174192642345, "全帯域_10ms_modulation_mae_db": 0.4222602088454102, "全帯域_40ms_log_rms_mae_db": 0.6796323136898602, "全帯域_40ms_rms_envelope_nrmse": 0.047026101209353785, "全帯域_40ms_envelope_correlation": 0.9978036687397628, "全帯域_40ms_modulation_mae_db": 0.852592860281815, "300–1000 Hz_10ms_log_rms_mae_db": 1.8214024437871903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12282896147124416, "300–1000 Hz_10ms_envelope_correlation": 0.9885336170090112, "300–1000 Hz_10ms_modulation_mae_db": 0.41178309454963985, "300–1000 Hz_40ms_log_rms_mae_db": 1.1961921221653982, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06834645331434781, "300–1000 Hz_40ms_envelope_correlation": 0.9961867979758994, "300–1000 Hz_40ms_modulation_mae_db": 1.0040828226466252, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0878535213280704, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18733398260252285, "1000–3000 Hz_10ms_envelope_correlation": 0.9749228000938477, "1000–3000 Hz_10ms_modulation_mae_db": 0.4124358391682527, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3545835821683876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11265647463498717, "1000–3000 Hz_40ms_envelope_correlation": 0.9914304655389122, "1000–3000 Hz_40ms_modulation_mae_db": 0.7238996417481581, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7009744731546066, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17475897619941605, "3000–6000 Hz_10ms_envelope_correlation": 0.9804331010503835, "3000–6000 Hz_10ms_modulation_mae_db": 0.7500489781278441, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6917314588413683, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10072540816475367, "3000–6000 Hz_40ms_envelope_correlation": 0.9931655167061935, "3000–6000 Hz_40ms_modulation_mae_db": 1.0365128282118306, "mr_stft_db": 6.334229223032987, "band_mae_db": 0.15479953513748543, "rms_gain_error_db": -0.08573120620775511, "output_peak": 0.44282158072373146}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4486834946508458, "全帯域_10ms_rms_envelope_nrmse": 0.0882081904382598, "全帯域_10ms_envelope_correlation": 0.9927059699897378, "全帯域_10ms_modulation_mae_db": 0.5490451250217815, "全帯域_40ms_log_rms_mae_db": 0.7635252374416126, "全帯域_40ms_rms_envelope_nrmse": 0.0498795726662435, "全帯域_40ms_envelope_correlation": 0.9975977694224814, "全帯域_40ms_modulation_mae_db": 0.5968730908009365, "300–1000 Hz_10ms_log_rms_mae_db": 1.8724018821406172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13018238855887784, "300–1000 Hz_10ms_envelope_correlation": 0.9872557761177727, "300–1000 Hz_10ms_modulation_mae_db": 0.5263331906113197, "300–1000 Hz_40ms_log_rms_mae_db": 1.2733827506251092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08482538235477663, "300–1000 Hz_40ms_envelope_correlation": 0.9944880206466139, "300–1000 Hz_40ms_modulation_mae_db": 0.9075045228998159, "1000–3000 Hz_10ms_log_rms_mae_db": 2.082484249094619, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1856688978645355, "1000–3000 Hz_10ms_envelope_correlation": 0.9746045338305162, "1000–3000 Hz_10ms_modulation_mae_db": 0.8143899269769156, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4029500435541036, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10914367120076522, "1000–3000 Hz_40ms_envelope_correlation": 0.990644107268005, "1000–3000 Hz_40ms_modulation_mae_db": 0.8505547573167184, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6696579719851385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18810892962394446, "3000–6000 Hz_10ms_envelope_correlation": 0.9771058005159893, "3000–6000 Hz_10ms_modulation_mae_db": 1.2162849059963616, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6959912786566442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10289980696451237, "3000–6000 Hz_40ms_envelope_correlation": 0.9931115215152869, "3000–6000 Hz_40ms_modulation_mae_db": 1.2769978979778898, "mr_stft_db": 6.321546800629998, "band_mae_db": 0.21741959993379856, "rms_gain_error_db": -0.11611067479701316, "output_peak": 0.4922079032726841}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4254323636856476, "全帯域_10ms_rms_envelope_nrmse": 0.087445871536611, "全帯域_10ms_envelope_correlation": 0.9928599207921691, "全帯域_10ms_modulation_mae_db": 0.41675134240444295, "全帯域_40ms_log_rms_mae_db": 0.727841031616858, "全帯域_40ms_rms_envelope_nrmse": 0.047013831925296325, "全帯域_40ms_envelope_correlation": 0.9979315211843635, "全帯域_40ms_modulation_mae_db": 0.6424255473008783, "300–1000 Hz_10ms_log_rms_mae_db": 1.8298206878191106, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13010954052919696, "300–1000 Hz_10ms_envelope_correlation": 0.9871762175716383, "300–1000 Hz_10ms_modulation_mae_db": 0.4778585572285465, "300–1000 Hz_40ms_log_rms_mae_db": 1.1891371373464545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07648423406025341, "300–1000 Hz_40ms_envelope_correlation": 0.9954902297331227, "300–1000 Hz_40ms_modulation_mae_db": 0.6372766787494478, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0853413143394652, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17054767039073587, "1000–3000 Hz_10ms_envelope_correlation": 0.9794947376725694, "1000–3000 Hz_10ms_modulation_mae_db": 0.8647198683266785, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3574488932438413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09808205585116257, "1000–3000 Hz_40ms_envelope_correlation": 0.9938838113785503, "1000–3000 Hz_40ms_modulation_mae_db": 1.1733253676814555, "3000–6000 Hz_10ms_log_rms_mae_db": 2.683115959301111, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1904948713553668, "3000–6000 Hz_10ms_envelope_correlation": 0.9763833537295386, "3000–6000 Hz_10ms_modulation_mae_db": 0.951324546427133, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6536911401723489, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10499776847867061, "3000–6000 Hz_40ms_envelope_correlation": 0.9926395701583142, "3000–6000 Hz_40ms_modulation_mae_db": 1.0930096961911429, "mr_stft_db": 6.322956702772454, "band_mae_db": 0.2128353625655368, "rms_gain_error_db": -0.1325577327405408, "output_peak": 0.4924140519903504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3289947125434183, "全帯域_10ms_rms_envelope_nrmse": 0.09054827960849852, "全帯域_10ms_envelope_correlation": 0.9923069075041255, "全帯域_10ms_modulation_mae_db": 0.4656268153698906, "全帯域_40ms_log_rms_mae_db": 0.6749167359744991, "全帯域_40ms_rms_envelope_nrmse": 0.04411074501087936, "全帯域_40ms_envelope_correlation": 0.9981949417668134, "全帯域_40ms_modulation_mae_db": 0.5608373452147886, "300–1000 Hz_10ms_log_rms_mae_db": 1.9088806930939797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12826803103608792, "300–1000 Hz_10ms_envelope_correlation": 0.987440691998014, "300–1000 Hz_10ms_modulation_mae_db": 0.4863947201369352, "300–1000 Hz_40ms_log_rms_mae_db": 1.2896140203737736, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07653706373528708, "300–1000 Hz_40ms_envelope_correlation": 0.995297628710476, "300–1000 Hz_40ms_modulation_mae_db": 0.6057244105092487, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0777622355458423, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19323200897149323, "1000–3000 Hz_10ms_envelope_correlation": 0.9727017153522587, "1000–3000 Hz_10ms_modulation_mae_db": 0.8620416286382054, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2873152265862504, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09860149794640198, "1000–3000 Hz_40ms_envelope_correlation": 0.9931381728024405, "1000–3000 Hz_40ms_modulation_mae_db": 1.0694699851423641, "3000–6000 Hz_10ms_log_rms_mae_db": 2.613312035952703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17341077995459905, "3000–6000 Hz_10ms_envelope_correlation": 0.9806471115232204, "3000–6000 Hz_10ms_modulation_mae_db": 0.5861425121478887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.621625104289589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.110375860827836, "3000–6000 Hz_40ms_envelope_correlation": 0.9919327286027373, "3000–6000 Hz_40ms_modulation_mae_db": 1.2590485977247068, "mr_stft_db": 6.311621728763129, "band_mae_db": 0.23333797646157695, "rms_gain_error_db": -0.12240992107854859, "output_peak": 0.5036152949676714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2965242834517496, "全帯域_10ms_rms_envelope_nrmse": 0.08476542838524101, "全帯域_10ms_envelope_correlation": 0.9932275313718654, "全帯域_10ms_modulation_mae_db": 0.14425874747083484, "全帯域_40ms_log_rms_mae_db": 0.7285245032636067, "全帯域_40ms_rms_envelope_nrmse": 0.04313696008874538, "全帯域_40ms_envelope_correlation": 0.9981866139648435, "全帯域_40ms_modulation_mae_db": 0.08183768584512036, "300–1000 Hz_10ms_log_rms_mae_db": 1.9514509813362968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12571381770246481, "300–1000 Hz_10ms_envelope_correlation": 0.9879942276529369, "300–1000 Hz_10ms_modulation_mae_db": 0.18796413114891744, "300–1000 Hz_40ms_log_rms_mae_db": 1.3235814371097645, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07342178713516968, "300–1000 Hz_40ms_envelope_correlation": 0.9957306126448251, "300–1000 Hz_40ms_modulation_mae_db": 0.5243713906590479, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0858960506806175, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16568915375850307, "1000–3000 Hz_10ms_envelope_correlation": 0.9801678162735625, "1000–3000 Hz_10ms_modulation_mae_db": 0.39422338120011063, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3427281800188753, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09000615368796572, "1000–3000 Hz_40ms_envelope_correlation": 0.9943421440031953, "1000–3000 Hz_40ms_modulation_mae_db": 0.577508209516085, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6526697836536073, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17144528817666732, "3000–6000 Hz_10ms_envelope_correlation": 0.981108581389607, "3000–6000 Hz_10ms_modulation_mae_db": 0.8870458355881209, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6414006375108496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09904181870663294, "3000–6000 Hz_40ms_envelope_correlation": 0.9933839224955132, "3000–6000 Hz_40ms_modulation_mae_db": 0.957816989982249, "mr_stft_db": 6.315261485691205, "band_mae_db": 0.13085153817495282, "rms_gain_error_db": -0.09369012743748732, "output_peak": 0.5222058236541995}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3565465603472233, "全帯域_10ms_rms_envelope_nrmse": 0.09309722530255492, "全帯域_10ms_envelope_correlation": 0.9919244771218171, "全帯域_10ms_modulation_mae_db": 0.24437744581210583, "全帯域_40ms_log_rms_mae_db": 0.7431144515226948, "全帯域_40ms_rms_envelope_nrmse": 0.054368915339588035, "全帯域_40ms_envelope_correlation": 0.9972270996387156, "全帯域_40ms_modulation_mae_db": 0.22360613682504016, "300–1000 Hz_10ms_log_rms_mae_db": 1.9720356513160997, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13189915088571805, "300–1000 Hz_10ms_envelope_correlation": 0.9867040842997948, "300–1000 Hz_10ms_modulation_mae_db": 0.29581425170852277, "300–1000 Hz_40ms_log_rms_mae_db": 1.315284811591941, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07826251189509803, "300–1000 Hz_40ms_envelope_correlation": 0.9950694580856188, "300–1000 Hz_40ms_modulation_mae_db": 0.39604779405951335, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0569374308526167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1928346030898534, "1000–3000 Hz_10ms_envelope_correlation": 0.9742675789224068, "1000–3000 Hz_10ms_modulation_mae_db": 0.7283545512248004, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3512851745694068, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12120537046853734, "1000–3000 Hz_40ms_envelope_correlation": 0.9911754217939653, "1000–3000 Hz_40ms_modulation_mae_db": 0.7194162479600082, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6307139627604856, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17524981986745075, "3000–6000 Hz_10ms_envelope_correlation": 0.9802534182343577, "3000–6000 Hz_10ms_modulation_mae_db": 0.7107472569260705, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5970446809831755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10301122173005152, "3000–6000 Hz_40ms_envelope_correlation": 0.9930649898343439, "3000–6000 Hz_40ms_modulation_mae_db": 0.9476102005408992, "mr_stft_db": 6.316470249725591, "band_mae_db": 0.2775773333889194, "rms_gain_error_db": -0.15175089153000165, "output_peak": 0.5201532497110252}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.62180062024938, "spectral_convergence": 0.3425436007976942, "spectral_convergence_status": "ok", "amplitude_floor": 1.3425996786840756e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.241473321939361, "spectral_convergence": 0.3338507084283393, "spectral_convergence_status": "ok", "amplitude_floor": 1.1106411258473627e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.1315387023665755, "spectral_convergence": 0.3542300251785522, "spectral_convergence_status": "ok", "amplitude_floor": 1.0520318003394391e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00033395677419724207, "output_power": 0.00034992591219115915, "delta_db": 0.20285845997279037, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008500397987721807, "output_power": 0.0008184924529182404, "delta_db": -0.16424580548528828, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006268367014596879, "output_power": 0.0006093832035615671, "delta_db": -0.122639370149889, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003695716856827319, "output_power": 0.0003320565520109433, "delta_db": -0.464866368519104, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013451399439061329, "output_power": 0.0001287402357343379, "delta_db": -0.19053169225032335, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011413253897640505, "output_power": 9.930504040340169e-05, "delta_db": -0.6043818618617536, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29158725970652477, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16610159553763745, "pitch": null, "output_rms_dbfs": -26.31173438039419, "output_peak": 0.469351753294972, "reference_rms_dbfs": -26.14563278485655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.331604214851772}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018099058448312497, "log_rms_mae_db": 1.4346520000117153, "rms_envelope_nrmse": 0.09668805193454395, "envelope_correlation": 0.9913095669485932, "rms_gain_error_db": -0.16610520326614814, "reference_mean_power": 0.002431072908198481, "output_mean_power": 0.0023398470159311765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.757963657265947e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003340894048273399, "output_power": 0.00032514443624585804, "reference_active": true, "delta_db": -0.11786375904452825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010057920506755965, "output_power": 0.00010103120251878763, "reference_active": true, "delta_db": 0.019473233563635367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.918423836914529e-05, "output_power": 1.845102884881772e-05, "reference_active": true, "delta_db": -0.16923974121677132}], "active_band_count": 3, "active_band_mae_db": 0.10219224460831165}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016835992502135956, "log_rms_mae_db": 0.7225013625186534, "rms_envelope_nrmse": 0.05596733119576833, "envelope_correlation": 0.9970916416677889, "rms_gain_error_db": -0.16605398496348067, "reference_mean_power": 0.002443196853891242, "output_mean_power": 0.0023515437438034305}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.921742652983223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003035779346246488, "output_power": 0.00029774476579419245, "reference_active": true, "delta_db": -0.08426066403834324}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.67325523785903e-05, "output_power": 4.810654298691756e-05, "reference_active": true, "delta_db": 0.1258464751722428}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.222466914385115e-06, "output_power": 1.2703386606060946e-06, "reference_active": true, "delta_db": 0.1668240137182139}], "active_band_count": 3, "active_band_mae_db": 0.12564371764293333}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015437420870825604, "log_rms_mae_db": 1.8664152367370144, "rms_envelope_nrmse": 0.1317165142616717, "envelope_correlation": 0.9867824998943647, "rms_gain_error_db": -0.058124710216638405, "reference_mean_power": 0.0008784594297426283, "output_mean_power": 0.0008667807108698395}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.115238295382225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001974652277483934, "output_power": 0.00019555652248478647, "reference_active": true, "delta_db": -0.04218324711781283}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.173113277871683e-05, "output_power": 4.9985043535140455e-05, "reference_active": true, "delta_db": -0.14911913746006505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0925495372248406e-05, "output_power": 1.1556274528053474e-05, "reference_active": true, "delta_db": 0.2437671287937596}], "active_band_count": 3, "active_band_mae_db": 0.14502317112387916}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012941193693393203, "log_rms_mae_db": 1.2366234437748562, "rms_envelope_nrmse": 0.0762147962349473, "envelope_correlation": 0.995324661985446, "rms_gain_error_db": -0.05812404273088268, "reference_mean_power": 0.000882875660341548, "output_mean_power": 0.0008711383635790212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7308162041776265e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018289387957124914, "output_power": 0.00018372082455632685, "reference_active": true, "delta_db": 0.019592135770529765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5176917167074748e-05, "output_power": 2.4807095470409645e-05, "reference_active": true, "delta_db": -0.0642663301786306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.42204663895331e-07, "output_power": 8.660971350103018e-07, "reference_active": true, "delta_db": -0.3657864767052604}], "active_band_count": 3, "active_band_mae_db": 0.14988164755147357}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.92340600003817e-05, "log_rms_mae_db": 2.0271702499719675, "rms_envelope_nrmse": 0.19182382816762306, "envelope_correlation": 0.9737805678751207, "rms_gain_error_db": -0.4499681825852866, "reference_mean_power": 0.000424662888055856, "output_mean_power": 0.00038286660804228855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0384779490480225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014012639512762435, "output_power": 0.00012357017935448188, "reference_active": true, "delta_db": -0.546062725228733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6627820389923143e-05, "output_power": 3.419826771293867e-05, "reference_active": true, "delta_db": -0.29806968132161027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.517673552823662e-06, "output_power": 8.035664782675055e-06, "reference_active": true, "delta_db": -0.2529918000391065}], "active_band_count": 3, "active_band_mae_db": 0.36570806886314994}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.48214657797314e-05, "log_rms_mae_db": 1.2905851223154505, "rms_envelope_nrmse": 0.12920634421382649, "envelope_correlation": 0.9883790060375651, "rms_gain_error_db": -0.44997912939364726, "reference_mean_power": 0.0004267980543892402, "output_mean_power": 0.0003847906565826074}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.813635389636035e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013656501467579018, "output_power": 0.00012091336218911412, "reference_active": true, "delta_db": -0.5286515813163203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7671441991318854e-05, "output_power": 1.6887334850710944e-05, "reference_active": true, "delta_db": -0.1971087454475589}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.379326657430796e-07, "output_power": 4.820768825385158e-07, "reference_active": true, "delta_db": -0.47611611566171774}], "active_band_count": 3, "active_band_mae_db": 0.40062548080853233}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.51121938209953e-05, "log_rms_mae_db": 2.673395743823498, "rms_envelope_nrmse": 0.18898914577166856, "envelope_correlation": 0.9785487605163302, "rms_gain_error_db": -0.5346023223146567, "reference_mean_power": 0.00015588925116025816, "output_mean_power": 0.00013783386634569204}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103456000035232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4374604191221866e-05, "output_power": 4.559460440364083e-05, "reference_active": true, "delta_db": -0.7648265666202989}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5832401077981513e-05, "output_power": 1.1606924019779716e-05, "reference_active": true, "delta_db": -1.348296416935821}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.903828543673632e-06, "output_power": 3.1003599386531553e-06, "reference_active": true, "delta_db": -1.000786186696339}], "active_band_count": 3, "active_band_mae_db": 1.0379697234174863}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091623719145545e-05, "log_rms_mae_db": 1.6419294160572733, "rms_envelope_nrmse": 0.11563161651408366, "envelope_correlation": 0.9935962903716408, "rms_gain_error_db": -0.5346318754066716, "reference_mean_power": 0.00015667279701186957, "output_mean_power": 0.00013852571780030706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1374269590951097e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.419500527522151e-05, "output_power": 4.538244951745319e-05, "reference_active": true, "delta_db": -0.7707132967323406}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661491941713126e-06, "output_power": 5.161946619521751e-06, "reference_active": true, "delta_db": -1.1075799774594384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7948760162947e-07, "output_power": 1.7943068742094858e-07, "reference_active": true, "delta_db": -1.924658259092285}], "active_band_count": 3, "active_band_mae_db": 1.2676505110946878}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7651786302962744, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7606c51ea713e516156cc6cfbef8e84bb4197c7d7888272ae94628e42be3dc3b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9755423122025076, "standard_deviation": 0.0027356272984513288, "minimum": 0.9727017153522587, "maximum": 0.9801678162735625, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.067353160358722, "standard_deviation": 0.024408423235892655, "minimum": 2.0271702499719675, "maximum": 2.0878535213280704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6165812421265024, "standard_deviation": 0.22146513115620303, "minimum": 0.36570806886314994, "maximum": 0.8647198683266785, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18420228805413158, "standard_deviation": 0.010423250008331024, "minimum": 0.16568915375850307, "maximum": 0.19323200897149323, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9917939920096619, "standard_deviation": 0.0019403893595574075, "minimum": 0.9883790060375651, "maximum": 0.9943421440031953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3374397452432807, "standard_deviation": 0.03878382626537222, "minimum": 1.2873152265862504, "maximum": 1.4029500435541036, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7507824177692355, "standard_deviation": 0.27035624083693377, "minimum": 0.40062548080853233, "maximum": 1.1733253676814555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10809298933849218, "standard_deviation": 0.01288489612834845, "minimum": 0.09000615368796572, "maximum": 0.12920634421382649, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9791880515592932, "standard_deviation": 0.0017378170573425795, "minimum": 0.9763833537295386, "maximum": 0.981108581389607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.661133689448604, "standard_deviation": 0.028269655565697348, "minimum": 2.613312035952703, "maximum": 2.7009744731546066, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8724760935281288, "standard_deviation": 0.198635479507335, "minimum": 0.5861425121478887, "maximum": 1.2162849059963616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18037943107602003, "standard_deviation": 0.0077691439573167155, "minimum": 0.17144528817666732, "maximum": 0.1904948713553668, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9930486463284545, "standard_deviation": 0.0005400689309484088, "minimum": 0.9919327286027373, "maximum": 0.9935962903716408, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.647447051772483, "standard_deviation": 0.03329125012245642, "minimum": 1.5970446809831755, "maximum": 1.6959912786566442, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1016985124456187, "standard_deviation": 0.14534005949739232, "minimum": 0.9476102005408992, "maximum": 1.2769978979778898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10467146919142409, "standard_deviation": 0.005628968104207593, "minimum": 0.09904181870663294, "maximum": 0.11563161651408366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9876049733676844, "standard_deviation": 0.0008151260143323211, "minimum": 0.9867040842997948, "maximum": 0.9889526723979422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8825856213780479, "standard_deviation": 0.056430873914841036, "minimum": 1.8214024437871903, "maximum": 1.9720356513160997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3926455039494633, "standard_deviation": 0.1665259728984973, "minimum": 0.14502317112387916, "maximum": 0.6099929150879456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12767987232590594, "standard_deviation": 0.004182226564576994, "minimum": 0.12072057416198613, "maximum": 0.13189915088571805, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954567615315086, "standard_deviation": 0.0005493252640778152, "minimum": 0.9944880206466139, "maximum": 0.9961867979758994, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.254629484210655, "standard_deviation": 0.053120034550903, "minimum": 1.1891371373464545, "maximum": 1.3235814371097645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6355615492331533, "standard_deviation": 0.2844703831067393, "minimum": 0.14988164755147357, "maximum": 1.0040828226466252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07564350220618238, "standard_deviation": 0.004960239733902508, "minimum": 0.06834645331434781, "maximum": 0.08482538235477663, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20772356294970307, "standard_deviation": 0.06040255291306803, "minimum": 0.13085153817495282, "maximum": 0.29158725970652477, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.320175270198308, "standard_deviation": 0.009289983378631054, "minimum": 6.307711756119327, "maximum": 6.334229223032987, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4981451305726402, "standard_deviation": 0.031642597499527275, "minimum": 0.44282158072373146, "maximum": 0.5423913869664874, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12417646020287065, "standard_deviation": 0.026898548490960076, "minimum": -0.16610159553763745, "maximum": -0.08573120620775511, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924188726756056, "standard_deviation": 0.0005897410573230098, "minimum": 0.9913095669485932, "maximum": 0.9932275313718654, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.36723628076458, "standard_deviation": 0.06508533480839, "minimum": 1.2751044311096926, "maximum": 1.4486834946508458, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31588002353925276, "standard_deviation": 0.1674572329444123, "minimum": 0.10219224460831165, "maximum": 0.5490451250217815, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08990369232711737, "standard_deviation": 0.0036452705487229676, "minimum": 0.08476542838524101, "maximum": 0.09668805193454395, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977210738544495, "standard_deviation": 0.0004049675714474083, "minimum": 0.9970916416677889, "maximum": 0.9981949417668134, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7094910436690252, "standard_deviation": 0.04213087731274902, "minimum": 0.635872713324416, "maximum": 0.7635252374416126, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4023702396099923, "standard_deviation": 0.2942191787141334, "minimum": 0.08183768584512036, "maximum": 0.852592860281815, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048785864056130786, "standard_deviation": 0.004534689365772505, "minimum": 0.04313696008874538, "maximum": 0.05596733119576833, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0002", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.0706076622009277, "predicted_mos": 2.0706076622009277, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c7e42699e20ed8a1931ad8c828c94f76c4e0602664bc23e1cb0f341fcebcf039", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9480642997659743, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3124638772356925, "全帯域_10ms_rms_envelope_nrmse": 0.09417033340474808, "全帯域_10ms_envelope_correlation": 0.9916308587738816, "全帯域_10ms_modulation_mae_db": 0.44721925966581627, "全帯域_40ms_log_rms_mae_db": 0.7038741353482635, "全帯域_40ms_rms_envelope_nrmse": 0.052873107667123116, "全帯域_40ms_envelope_correlation": 0.9971580490133847, "全帯域_40ms_modulation_mae_db": 0.5141965425090687, "300–1000 Hz_10ms_log_rms_mae_db": 1.5695752938278025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12258679007379274, "300–1000 Hz_10ms_envelope_correlation": 0.9885640951208298, "300–1000 Hz_10ms_modulation_mae_db": 0.42606827918174855, "300–1000 Hz_40ms_log_rms_mae_db": 0.9874363779244344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07025754983817108, "300–1000 Hz_40ms_envelope_correlation": 0.9959779490297239, "300–1000 Hz_40ms_modulation_mae_db": 0.8076643426454829, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5974699374283876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17229486076104183, "1000–3000 Hz_10ms_envelope_correlation": 0.978284064832768, "1000–3000 Hz_10ms_modulation_mae_db": 0.5975167901728787, "1000–3000 Hz_40ms_log_rms_mae_db": 1.046450138449957, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10261630706115842, "1000–3000 Hz_40ms_envelope_correlation": 0.9917184394439397, "1000–3000 Hz_40ms_modulation_mae_db": 0.6140022745685869, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7277479023339726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18744794861673175, "3000–6000 Hz_10ms_envelope_correlation": 0.9776316746145596, "3000–6000 Hz_10ms_modulation_mae_db": 0.618574980571099, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0465701033749635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12521642631610475, "3000–6000 Hz_40ms_envelope_correlation": 0.9895600712167568, "3000–6000 Hz_40ms_modulation_mae_db": 0.8958512838878634, "mr_stft_db": 5.851881679949035, "band_mae_db": 0.10394613369299278, "rms_gain_error_db": -0.024945508282830854, "output_peak": 0.5090521650491484}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4748353798454439, "全帯域_10ms_rms_envelope_nrmse": 0.09261677676750943, "全帯域_10ms_envelope_correlation": 0.991879287995491, "全帯域_10ms_modulation_mae_db": 0.17272619071335363, "全帯域_40ms_log_rms_mae_db": 0.7160177037124044, "全帯域_40ms_rms_envelope_nrmse": 0.04756038933642064, "全帯域_40ms_envelope_correlation": 0.997747404677586, "全帯域_40ms_modulation_mae_db": 0.24026474930001315, "300–1000 Hz_10ms_log_rms_mae_db": 1.5807553306021342, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12812750559033795, "300–1000 Hz_10ms_envelope_correlation": 0.9874386983706908, "300–1000 Hz_10ms_modulation_mae_db": 0.26256147493118515, "300–1000 Hz_40ms_log_rms_mae_db": 1.0122144214000763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07266596748825549, "300–1000 Hz_40ms_envelope_correlation": 0.9956916853290345, "300–1000 Hz_40ms_modulation_mae_db": 0.37269974716357585, "1000–3000 Hz_10ms_log_rms_mae_db": 1.530509724715744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1486672770885206, "1000–3000 Hz_10ms_envelope_correlation": 0.9844051874195091, "1000–3000 Hz_10ms_modulation_mae_db": 0.38254464672317495, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9206401190901546, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08259541109405305, "1000–3000 Hz_40ms_envelope_correlation": 0.9956652156186621, "1000–3000 Hz_40ms_modulation_mae_db": 0.6083153673537885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7090943934039797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1665309789105103, "3000–6000 Hz_10ms_envelope_correlation": 0.9819259205745441, "3000–6000 Hz_10ms_modulation_mae_db": 0.34744791198480945, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0395633058376814, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08854802407871001, "3000–6000 Hz_40ms_envelope_correlation": 0.9947282601088961, "3000–6000 Hz_40ms_modulation_mae_db": 0.20664538606062088, "mr_stft_db": 5.858130656583562, "band_mae_db": 0.20334979767806735, "rms_gain_error_db": -0.08371587701768671, "output_peak": 0.47553699070029476}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.430373553977012, "全帯域_10ms_rms_envelope_nrmse": 0.08638208981520892, "全帯域_10ms_envelope_correlation": 0.9929762732839974, "全帯域_10ms_modulation_mae_db": 0.20378688056429087, "全帯域_40ms_log_rms_mae_db": 0.755833325754791, "全帯域_40ms_rms_envelope_nrmse": 0.048234063392844424, "全帯域_40ms_envelope_correlation": 0.997715550809315, "全帯域_40ms_modulation_mae_db": 0.3080305307601226, "300–1000 Hz_10ms_log_rms_mae_db": 1.5574943583922798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1286683917746768, "300–1000 Hz_10ms_envelope_correlation": 0.9873853975336231, "300–1000 Hz_10ms_modulation_mae_db": 0.1347663931924635, "300–1000 Hz_40ms_log_rms_mae_db": 1.0564726087397198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08130580215593335, "300–1000 Hz_40ms_envelope_correlation": 0.9945840415922871, "300–1000 Hz_40ms_modulation_mae_db": 0.5999841668477502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5290851892007462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16012777238503306, "1000–3000 Hz_10ms_envelope_correlation": 0.9813600800140683, "1000–3000 Hz_10ms_modulation_mae_db": 0.5377812144597881, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0034774605303693, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09892470929002786, "1000–3000 Hz_40ms_envelope_correlation": 0.9927031755577159, "1000–3000 Hz_40ms_modulation_mae_db": 0.43601686220648733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7366372654817992, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15415801712902244, "3000–6000 Hz_10ms_envelope_correlation": 0.9849887037410249, "3000–6000 Hz_10ms_modulation_mae_db": 0.33311662931397484, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0371338077088232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0923160876454379, "3000–6000 Hz_40ms_envelope_correlation": 0.9949693737282957, "3000–6000 Hz_40ms_modulation_mae_db": 0.4148649903474942, "mr_stft_db": 5.855922827341376, "band_mae_db": 0.1805600599180023, "rms_gain_error_db": -0.09663428382391086, "output_peak": 0.6096671068045736}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4245738678041517, "全帯域_10ms_rms_envelope_nrmse": 0.09433171301927518, "全帯域_10ms_envelope_correlation": 0.9916949635658044, "全帯域_10ms_modulation_mae_db": 0.4565731063015151, "全帯域_40ms_log_rms_mae_db": 0.7512115070135404, "全帯域_40ms_rms_envelope_nrmse": 0.05189962882514932, "全帯域_40ms_envelope_correlation": 0.9974769926941676, "全帯域_40ms_modulation_mae_db": 0.5521112443938178, "300–1000 Hz_10ms_log_rms_mae_db": 1.5454855342995835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12990384623716472, "300–1000 Hz_10ms_envelope_correlation": 0.9870957883806823, "300–1000 Hz_10ms_modulation_mae_db": 0.3771102561490443, "300–1000 Hz_40ms_log_rms_mae_db": 1.0150060530560372, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07435842263979267, "300–1000 Hz_40ms_envelope_correlation": 0.9955494747316033, "300–1000 Hz_40ms_modulation_mae_db": 0.7489020651755188, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5199301245263999, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15326605194186288, "1000–3000 Hz_10ms_envelope_correlation": 0.9840683271312444, "1000–3000 Hz_10ms_modulation_mae_db": 0.7372205546904489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9634403083739753, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0940199011635135, "1000–3000 Hz_40ms_envelope_correlation": 0.9949723412193842, "1000–3000 Hz_40ms_modulation_mae_db": 0.781971096516087, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7179571209953348, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17218242984670432, "3000–6000 Hz_10ms_envelope_correlation": 0.9825034259142116, "3000–6000 Hz_10ms_modulation_mae_db": 0.9540079757910753, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0487223319818229, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1117653817965417, "3000–6000 Hz_40ms_envelope_correlation": 0.9940814995984826, "3000–6000 Hz_40ms_modulation_mae_db": 1.4750510443651133, "mr_stft_db": 5.852576102833184, "band_mae_db": 0.30471633900326384, "rms_gain_error_db": -0.14688833471011317, "output_peak": 0.5374615702655371}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4118018362273819, "全帯域_10ms_rms_envelope_nrmse": 0.08368697417044219, "全帯域_10ms_envelope_correlation": 0.9934319272345402, "全帯域_10ms_modulation_mae_db": 0.4332738016268776, "全帯域_40ms_log_rms_mae_db": 0.7779486652251528, "全帯域_40ms_rms_envelope_nrmse": 0.04493458414143368, "全帯域_40ms_envelope_correlation": 0.9980601730625193, "全帯域_40ms_modulation_mae_db": 0.7107673314597195, "300–1000 Hz_10ms_log_rms_mae_db": 1.532378147046417, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12065211714097672, "300–1000 Hz_10ms_envelope_correlation": 0.988904958831493, "300–1000 Hz_10ms_modulation_mae_db": 0.5911153446832552, "300–1000 Hz_40ms_log_rms_mae_db": 1.0217247442490045, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07453885019101969, "300–1000 Hz_40ms_envelope_correlation": 0.9955492479762725, "300–1000 Hz_40ms_modulation_mae_db": 0.8347546917114173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5399282945600659, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15646002250402316, "1000–3000 Hz_10ms_envelope_correlation": 0.9824756962363196, "1000–3000 Hz_10ms_modulation_mae_db": 0.5242829080621326, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9922756712141181, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09286601160523408, "1000–3000 Hz_40ms_envelope_correlation": 0.9940168074736171, "1000–3000 Hz_40ms_modulation_mae_db": 0.6795846214465194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6863651154890928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14659357818649937, "3000–6000 Hz_10ms_envelope_correlation": 0.9860694242154003, "3000–6000 Hz_10ms_modulation_mae_db": 0.6669953474656705, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0175626133794136, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0824917945482417, "3000–6000 Hz_40ms_envelope_correlation": 0.9955275121130935, "3000–6000 Hz_40ms_modulation_mae_db": 0.7774716104598941, "mr_stft_db": 5.8400801668982725, "band_mae_db": 0.21767237641053686, "rms_gain_error_db": -0.1060048817950973, "output_peak": 0.4519914470762072}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.375432639080462, "全帯域_10ms_rms_envelope_nrmse": 0.08843704012832634, "全帯域_10ms_envelope_correlation": 0.9926457090199087, "全帯域_10ms_modulation_mae_db": 0.4782752521506599, "全帯域_40ms_log_rms_mae_db": 0.7133575617883988, "全帯域_40ms_rms_envelope_nrmse": 0.04737630597530888, "全帯域_40ms_envelope_correlation": 0.9978384503040066, "全帯域_40ms_modulation_mae_db": 0.36284052414478357, "300–1000 Hz_10ms_log_rms_mae_db": 1.5109044940028784, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12465575103567882, "300–1000 Hz_10ms_envelope_correlation": 0.9882360496732502, "300–1000 Hz_10ms_modulation_mae_db": 0.6893176561920908, "300–1000 Hz_40ms_log_rms_mae_db": 0.9511678839628648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0713388450020937, "300–1000 Hz_40ms_envelope_correlation": 0.9960776617806538, "300–1000 Hz_40ms_modulation_mae_db": 0.796431587574046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5048702210908005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1562775930126748, "1000–3000 Hz_10ms_envelope_correlation": 0.982414898346872, "1000–3000 Hz_10ms_modulation_mae_db": 0.8207863428492171, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9538978523378749, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0943644600428935, "1000–3000 Hz_40ms_envelope_correlation": 0.9936390117858869, "1000–3000 Hz_40ms_modulation_mae_db": 0.9540724321974133, "3000–6000 Hz_10ms_log_rms_mae_db": 1.70125825463695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15859632810346114, "3000–6000 Hz_10ms_envelope_correlation": 0.9836231211228637, "3000–6000 Hz_10ms_modulation_mae_db": 0.4323323119619109, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0206792491037395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09319981773244428, "3000–6000 Hz_40ms_envelope_correlation": 0.994110339163645, "3000–6000 Hz_40ms_modulation_mae_db": 0.5330222750574408, "mr_stft_db": 5.831479079767351, "band_mae_db": 0.15177043192396691, "rms_gain_error_db": -0.11228174806834709, "output_peak": 0.5111721151193706}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4020532076750634, "全帯域_10ms_rms_envelope_nrmse": 0.09171824714437414, "全帯域_10ms_envelope_correlation": 0.9920370504007558, "全帯域_10ms_modulation_mae_db": 0.2417599731260385, "全帯域_40ms_log_rms_mae_db": 0.6592315352666847, "全帯域_40ms_rms_envelope_nrmse": 0.04352024745196806, "全帯域_40ms_envelope_correlation": 0.9980584884123157, "全帯域_40ms_modulation_mae_db": 0.20825530954330407, "300–1000 Hz_10ms_log_rms_mae_db": 1.5545114836236624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1302405623784612, "300–1000 Hz_10ms_envelope_correlation": 0.9870921268906396, "300–1000 Hz_10ms_modulation_mae_db": 0.4651567669411502, "300–1000 Hz_40ms_log_rms_mae_db": 0.963803593309218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07336548641537058, "300–1000 Hz_40ms_envelope_correlation": 0.9955946444590866, "300–1000 Hz_40ms_modulation_mae_db": 0.2000149780995146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5912526609790212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15837151859830426, "1000–3000 Hz_10ms_envelope_correlation": 0.9816045828947673, "1000–3000 Hz_10ms_modulation_mae_db": 0.3675657020694429, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9708386087685553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07974192225353634, "1000–3000 Hz_40ms_envelope_correlation": 0.9949815391664475, "1000–3000 Hz_40ms_modulation_mae_db": 0.718649154739946, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7614292725614726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14919109100878367, "3000–6000 Hz_10ms_envelope_correlation": 0.9858695084140899, "3000–6000 Hz_10ms_modulation_mae_db": 0.4195586939346498, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0593741485518042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08195507581343467, "3000–6000 Hz_40ms_envelope_correlation": 0.9955883159152796, "3000–6000 Hz_40ms_modulation_mae_db": 0.48974132375056495, "mr_stft_db": 5.858272961656176, "band_mae_db": 0.04710898396132233, "rms_gain_error_db": -0.01353915904345287, "output_peak": 0.51546038950686}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4318597666609534, "全帯域_10ms_rms_envelope_nrmse": 0.09593342128476813, "全帯域_10ms_envelope_correlation": 0.9912499026712821, "全帯域_10ms_modulation_mae_db": 0.2126643902206314, "全帯域_40ms_log_rms_mae_db": 0.7858870136424153, "全帯域_40ms_rms_envelope_nrmse": 0.0540156572209427, "全帯域_40ms_envelope_correlation": 0.9970131682737949, "全帯域_40ms_modulation_mae_db": 0.27744310395768607, "300–1000 Hz_10ms_log_rms_mae_db": 1.5665858407405941, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13425762861699883, "300–1000 Hz_10ms_envelope_correlation": 0.9863260036258715, "300–1000 Hz_10ms_modulation_mae_db": 0.2111184824484367, "300–1000 Hz_40ms_log_rms_mae_db": 1.0045021430622776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0762736322598241, "300–1000 Hz_40ms_envelope_correlation": 0.9952752888288203, "300–1000 Hz_40ms_modulation_mae_db": 0.38667480194620074, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5249242874868998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17687991376850756, "1000–3000 Hz_10ms_envelope_correlation": 0.9780308414816538, "1000–3000 Hz_10ms_modulation_mae_db": 0.3851211596384589, "1000–3000 Hz_40ms_log_rms_mae_db": 1.020428800541123, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10983142848107022, "1000–3000 Hz_40ms_envelope_correlation": 0.9922701190358322, "1000–3000 Hz_40ms_modulation_mae_db": 0.5252865756906536, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7984858218536302, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1553844055252528, "3000–6000 Hz_10ms_envelope_correlation": 0.984342385974383, "3000–6000 Hz_10ms_modulation_mae_db": 0.42313895260384343, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0845225085695287, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09157085190836588, "3000–6000 Hz_40ms_envelope_correlation": 0.9943071229723629, "3000–6000 Hz_40ms_modulation_mae_db": 0.4037407870702457, "mr_stft_db": 5.851780174828272, "band_mae_db": 0.19653577253921647, "rms_gain_error_db": -0.0419182275970175, "output_peak": 0.46539798957164086}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.983686598977633, "spectral_convergence": 0.3408620859769362, "spectral_convergence_status": "ok", "amplitude_floor": 1.3425996786840756e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.654928566986144, "spectral_convergence": 0.3347998637417288, "spectral_convergence_status": "ok", "amplitude_floor": 1.1106411258473627e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.917029873883329, "spectral_convergence": 0.3624325821909682, "spectral_convergence_status": "ok", "amplitude_floor": 1.0520318003394391e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00033395677419724207, "output_power": 0.0003413728023970656, "delta_db": 0.09538659963401142, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008500397987721807, "output_power": 0.0008342074590268222, "delta_db": -0.08165191060475485, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006268367014596879, "output_power": 0.0006237013801067583, "delta_db": -0.021777115559174505, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003695716856827319, "output_power": 0.0003728708245469141, "delta_db": 0.03859712273243559, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013451399439061329, "output_power": 0.0001260695228021301, "delta_db": -0.28157360301792134, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011413253897640505, "output_power": 0.00011691723166233617, "delta_db": 0.10469045060965894, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.10394613369299278, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.024945508282830854, "pitch": null, "output_rms_dbfs": -26.170578293139386, "output_peak": 0.5090521650491484, "reference_rms_dbfs": -26.14563278485655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.851881679949035}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018099058448312497, "log_rms_mae_db": 1.3124638772356925, "rms_envelope_nrmse": 0.09417033340474808, "envelope_correlation": 0.9916308587738816, "rms_gain_error_db": -0.024949926645299724, "reference_mean_power": 0.002431072908198481, "output_mean_power": 0.0024171465985853365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.757963657265947e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003340894048273399, "output_power": 0.00032664941698950904, "reference_active": true, "delta_db": -0.09780815432715272}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010057920506755965, "output_power": 9.382852565034788e-05, "reference_active": true, "delta_db": -0.3017330662356347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.918423836914529e-05, "output_power": 1.544304390528065e-05, "reference_active": true, "delta_db": -0.9421165584346614}], "active_band_count": 3, "active_band_mae_db": 0.44721925966581627}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016835992502135956, "log_rms_mae_db": 0.7038741353482635, "rms_envelope_nrmse": 0.052873107667123116, "envelope_correlation": 0.9971580490133847, "rms_gain_error_db": -0.024963253974081313, "reference_mean_power": 0.002443196853891242, "output_mean_power": 0.0024291936381588045}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.921742652983223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003035779346246488, "output_power": 0.00029671057981894947, "reference_active": true, "delta_db": -0.09937169697808156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.67325523785903e-05, "output_power": 4.349125545469987e-05, "reference_active": true, "delta_db": -0.3121755685943135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.222466914385115e-06, "output_power": 9.421778233236573e-07, "reference_active": true, "delta_db": -1.131042361954811}], "active_band_count": 3, "active_band_mae_db": 0.5141965425090687}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015437420870825604, "log_rms_mae_db": 1.5695752938278025, "rms_envelope_nrmse": 0.12258679007379274, "envelope_correlation": 0.9885640951208298, "rms_gain_error_db": -0.01689988028260986, "reference_mean_power": 0.0008784594297426283, "output_mean_power": 0.000875047686768212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.115238295382225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001974652277483934, "output_power": 0.00020426340641994107, "reference_active": true, "delta_db": 0.14699939728183117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.173113277871683e-05, "output_power": 5.077220767341479e-05, "reference_active": true, "delta_db": -0.08125940669009496}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0925495372248406e-05, "output_power": 8.579194898434618e-06, "reference_active": true, "delta_db": -1.0499460335733195}], "active_band_count": 3, "active_band_mae_db": 0.42606827918174855}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012941193693393203, "log_rms_mae_db": 0.9874363779244344, "rms_envelope_nrmse": 0.07025754983817108, "envelope_correlation": 0.9959779490297239, "rms_gain_error_db": -0.01689921385733225, "reference_mean_power": 0.000882875660341548, "output_mean_power": 0.0008794469006512062}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7308162041776265e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018289387957124914, "output_power": 0.00018934740433306852, "reference_active": true, "delta_db": 0.15060183648586803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5176917167074748e-05, "output_power": 2.4416496163180008e-05, "reference_active": true, "delta_db": -0.13319209474344076}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.42204663895331e-07, "output_power": 5.757385901621771e-07, "reference_active": true, "delta_db": -2.1391990967071397}], "active_band_count": 3, "active_band_mae_db": 0.8076643426454829}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.92340600003817e-05, "log_rms_mae_db": 1.5974699374283876, "rms_envelope_nrmse": 0.17229486076104183, "envelope_correlation": 0.978284064832768, "rms_gain_error_db": -0.07843526254915996, "reference_mean_power": 0.000424662888055856, "output_mean_power": 0.00041706215479828573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0384779490480225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014012639512762435, "output_power": 0.00012287261444922807, "reference_active": true, "delta_db": -0.5706485038199007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6627820389923143e-05, "output_power": 3.0336762540101618e-05, "reference_active": true, "delta_db": -0.8184184373795306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.517673552823662e-06, "output_power": 7.761982306271337e-06, "reference_active": true, "delta_db": -0.4034834293192046}], "active_band_count": 3, "active_band_mae_db": 0.5975167901728787}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.48214657797314e-05, "log_rms_mae_db": 1.046450138449957, "rms_envelope_nrmse": 0.10261630706115842, "envelope_correlation": 0.9917184394439397, "rms_gain_error_db": -0.0784368855785887, "reference_mean_power": 0.0004267980543892402, "output_mean_power": 0.000419158948690361}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.813635389636035e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013656501467579018, "output_power": 0.00011908304592075002, "reference_active": true, "delta_db": -0.5948952116546381}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7671441991318854e-05, "output_power": 1.4861312986094308e-05, "reference_active": true, "delta_db": -0.7521480870473706}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.379326657430796e-07, "output_power": 4.799893097466227e-07, "reference_active": true, "delta_db": -0.494963525003752}], "active_band_count": 3, "active_band_mae_db": 0.6140022745685869}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.51121938209953e-05, "log_rms_mae_db": 1.7277479023339726, "rms_envelope_nrmse": 0.18744794861673175, "envelope_correlation": 0.9776316746145596, "rms_gain_error_db": 0.09180871849474902, "reference_mean_power": 0.00015588925116025816, "output_mean_power": 0.00015921978850375163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103456000035232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4374604191221866e-05, "output_power": 5.0507951551196605e-05, "reference_active": true, "delta_db": -0.3203635340957311}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5832401077981513e-05, "output_power": 1.3529372241176196e-05, "reference_active": true, "delta_db": -0.6826913727253812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.903828543673632e-06, "output_power": 3.207921495770126e-06, "reference_active": true, "delta_db": -0.8526700348921846}], "active_band_count": 3, "active_band_mae_db": 0.618574980571099}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091623719145545e-05, "log_rms_mae_db": 1.0465701033749635, "rms_envelope_nrmse": 0.12521642631610475, "envelope_correlation": 0.9895600712167568, "rms_gain_error_db": 0.09180801286721779, "reference_mean_power": 0.00015667279701186957, "output_mean_power": 0.00016002004862944642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1374269590951097e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.419500527522151e-05, "output_power": 5.0081149271587655e-05, "reference_active": true, "delta_db": -0.3428497630917561}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661491941713126e-06, "output_power": 5.716932813647517e-06, "reference_active": true, "delta_db": -0.6640841852872702}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7948760162947e-07, "output_power": 1.8980190041974202e-07, "reference_active": true, "delta_db": -1.680619903284564}], "active_band_count": 3, "active_band_mae_db": 0.8958512838878634}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.782540524414501, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a664101c1740869a8572118257de232f9bce581dd9c1fcfc8b5f96691e1eb93a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9815804597946503, "standard_deviation": 0.002365872388401096, "minimum": 0.9780308414816538, "maximum": 0.9844051874195091, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5422463049985082, "standard_deviation": 0.03372714848227923, "minimum": 1.5048702210908005, "maximum": 1.5974699374283876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5441024148331928, "standard_deviation": 0.16889372853680806, "minimum": 0.3675657020694429, "maximum": 0.8207863428492171, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.160293126257496, "standard_deviation": 0.00954867532241592, "minimum": 0.1486672770885206, "maximum": 0.17687991376850756, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937458311626857, "standard_deviation": 0.0014230425449462739, "minimum": 0.9917184394439397, "maximum": 0.9956652156186621, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9839311199132659, "standard_deviation": 0.03998470152087769, "minimum": 0.9206401190901546, "maximum": 1.046450138449957, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6647372980899353, "standard_deviation": 0.1595093070588782, "minimum": 0.43601686220648733, "maximum": 0.9540724321974133, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09437001887393587, "standard_deviation": 0.009876443192730552, "minimum": 0.07974192225353634, "maximum": 0.10983142848107022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9833692705713846, "standard_deviation": 0.0027497773935841927, "minimum": 0.9776316746145596, "maximum": 0.9860694242154003, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.729871893344529, "standard_deviation": 0.035934995497476294, "minimum": 1.6863651154890928, "maximum": 1.7984858218536302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5243966004533792, "standard_deviation": 0.21091254130805331, "minimum": 0.33311662931397484, "maximum": 0.9540079757910753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1612605971658707, "standard_deviation": 0.013552594661879781, "minimum": 0.14659357818649937, "maximum": 0.18744794861673175, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9941090618521016, "standard_deviation": 0.001929361184028016, "minimum": 0.9895600712167568, "maximum": 0.9955883159152796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.044266008563472, "standard_deviation": 0.021435616543098997, "minimum": 1.0175626133794136, "maximum": 1.0845225085695287, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6495485876249046, "standard_deviation": 0.39807778829732243, "minimum": 0.20664538606062088, "maximum": 1.4750510443651133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09588293247991012, "standard_deviation": 0.015015849865539926, "minimum": 0.08195507581343467, "maximum": 0.12521642631610475, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.987630389803385, "standard_deviation": 0.0008649419028627114, "minimum": 0.9863260036258715, "maximum": 0.988904958831493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.552211310316919, "standard_deviation": 0.022363685611370206, "minimum": 1.5109044940028784, "maximum": 1.5807553306021342, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3946518317149218, "standard_deviation": 0.18911987875529832, "minimum": 0.1347663931924635, "maximum": 0.6893176561920908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12738657410601095, "standard_deviation": 0.004467878469794907, "minimum": 0.12065211714097672, "maximum": 0.13425762861699883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955374992159353, "standard_deviation": 0.00046150804738782426, "minimum": 0.9945840415922871, "maximum": 0.9960776617806538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.001540978212954, "standard_deviation": 0.03357159974057699, "minimum": 0.9511678839628648, "maximum": 1.0564726087397198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5933907976454383, "standard_deviation": 0.2437220456382946, "minimum": 0.2000149780995146, "maximum": 0.8347546917114173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07426306949880758, "standard_deviation": 0.00341591359856719, "minimum": 0.07025754983817108, "maximum": 0.08130580215593335, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1757074868909211, "standard_deviation": 0.07735451555611073, "minimum": 0.04710898396132233, "maximum": 0.30471633900326384, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.850015456232153, "standard_deviation": 0.009444583226730493, "minimum": 5.831479079767351, "maximum": 5.858272961656176, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.509467471761704, "standard_deviation": 0.049655268724499615, "minimum": 0.4519914470762072, "maximum": 0.6096671068045736, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07824100254230704, "standard_deviation": 0.046837760785736045, "minimum": -0.14688833471011317, "maximum": -0.01353915904345287, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9921932466182076, "standard_deviation": 0.0007492485586670306, "minimum": 0.9912499026712821, "maximum": 0.9934319272345402, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.40792426606327, "standard_deviation": 0.04788241794597968, "minimum": 1.3124638772356925, "maximum": 1.4748353798454439, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33078485679614794, "standard_deviation": 0.13343053668853916, "minimum": 0.17272619071335363, "maximum": 0.4782752521506599, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09090957446683155, "standard_deviation": 0.004308274555772685, "minimum": 0.08368697417044219, "maximum": 0.09593342128476813, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976335346558862, "standard_deviation": 0.0003890897636536825, "minimum": 0.9970131682737949, "maximum": 0.9980601730625193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7329201809689563, "standard_deviation": 0.042484810140231644, "minimum": 0.6592315352666847, "maximum": 0.7858870136424153, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3967386670085644, "standard_deviation": 0.17720476624989348, "minimum": 0.20825530954330407, "maximum": 0.7107673314597195, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04880174800139885, "standard_deviation": 0.003781572064793759, "minimum": 0.04352024745196806, "maximum": 0.0540156572209427, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0002", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.93658447265625, "predicted_mos": 2.93658447265625, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "47bb6c3861b5fffbdf3ddec31a4b0a5a47881cef13a871cc819d3e7da381600b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.869748499710113, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3838317740755839, "全帯域_10ms_rms_envelope_nrmse": 0.09534115364976303, "全帯域_10ms_envelope_correlation": 0.9913944586854666, "全帯域_10ms_modulation_mae_db": 0.14060846042435846, "全帯域_40ms_log_rms_mae_db": 0.7427490014897609, "全帯域_40ms_rms_envelope_nrmse": 0.05178240159127826, "全帯域_40ms_envelope_correlation": 0.9972442316723727, "全帯域_40ms_modulation_mae_db": 0.20348134469492288, "300–1000 Hz_10ms_log_rms_mae_db": 1.655577633409515, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13891218980446304, "300–1000 Hz_10ms_envelope_correlation": 0.9853587358306028, "300–1000 Hz_10ms_modulation_mae_db": 0.10551901342344892, "300–1000 Hz_40ms_log_rms_mae_db": 1.141764603635641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08226031442697844, "300–1000 Hz_40ms_envelope_correlation": 0.9944994332594529, "300–1000 Hz_40ms_modulation_mae_db": 0.3973957873428123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.761154567267277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17578751025399347, "1000–3000 Hz_10ms_envelope_correlation": 0.9773765630095298, "1000–3000 Hz_10ms_modulation_mae_db": 0.27184549613133996, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1681534768931754, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09887205669760576, "1000–3000 Hz_40ms_envelope_correlation": 0.9922497058417545, "1000–3000 Hz_40ms_modulation_mae_db": 0.5024858899591909, "3000–6000 Hz_10ms_log_rms_mae_db": 2.061823100332808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16284455783771895, "3000–6000 Hz_10ms_envelope_correlation": 0.9833040027315411, "3000–6000 Hz_10ms_modulation_mae_db": 0.35033387768171514, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3176821509464542, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09494507916847582, "3000–6000 Hz_40ms_envelope_correlation": 0.9941821711979419, "3000–6000 Hz_40ms_modulation_mae_db": 0.4925089323825042, "mr_stft_db": 6.011346325918552, "band_mae_db": 0.11814990246791794, "rms_gain_error_db": 0.005400078237815243, "output_peak": 0.49564794536866247}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4563118857298896, "全帯域_10ms_rms_envelope_nrmse": 0.09311007947690289, "全帯域_10ms_envelope_correlation": 0.991845617814633, "全帯域_10ms_modulation_mae_db": 0.2982750047471268, "全帯域_40ms_log_rms_mae_db": 0.7374852774822924, "全帯域_40ms_rms_envelope_nrmse": 0.04931647459175331, "全帯域_40ms_envelope_correlation": 0.9976610008362949, "全帯域_40ms_modulation_mae_db": 0.4424575461004084, "300–1000 Hz_10ms_log_rms_mae_db": 1.65728017351208, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13526738618080672, "300–1000 Hz_10ms_envelope_correlation": 0.9860881697301076, "300–1000 Hz_10ms_modulation_mae_db": 0.5561750351063731, "300–1000 Hz_40ms_log_rms_mae_db": 1.1086531889946012, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08662337866717869, "300–1000 Hz_40ms_envelope_correlation": 0.9940869869780579, "300–1000 Hz_40ms_modulation_mae_db": 0.6923107141221588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.838483875923862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16789876198811018, "1000–3000 Hz_10ms_envelope_correlation": 0.9798351238030479, "1000–3000 Hz_10ms_modulation_mae_db": 0.34502428990834405, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1922813529500678, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09496730538478843, "1000–3000 Hz_40ms_envelope_correlation": 0.9939657378924444, "1000–3000 Hz_40ms_modulation_mae_db": 0.1860614222494551, "3000–6000 Hz_10ms_log_rms_mae_db": 2.083156820607428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15975682257913584, "3000–6000 Hz_10ms_envelope_correlation": 0.9838488805775374, "3000–6000 Hz_10ms_modulation_mae_db": 0.5486764436462553, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3047885920759068, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0861234076497821, "3000–6000 Hz_40ms_envelope_correlation": 0.9951632587762228, "3000–6000 Hz_40ms_modulation_mae_db": 0.8585642659115825, "mr_stft_db": 6.0094453320444385, "band_mae_db": 0.21126262042605445, "rms_gain_error_db": -0.1202303733694716, "output_peak": 0.5434042064454803}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3347102259274681, "全帯域_10ms_rms_envelope_nrmse": 0.08641932918981055, "全帯域_10ms_envelope_correlation": 0.9929814483168073, "全帯域_10ms_modulation_mae_db": 0.10770527227544396, "全帯域_40ms_log_rms_mae_db": 0.6524490565636386, "全帯域_40ms_rms_envelope_nrmse": 0.04608799585195759, "全帯域_40ms_envelope_correlation": 0.9979582791587508, "全帯域_40ms_modulation_mae_db": 0.2774565802721929, "300–1000 Hz_10ms_log_rms_mae_db": 1.5840078907755397, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11440373161581852, "300–1000 Hz_10ms_envelope_correlation": 0.990035266337289, "300–1000 Hz_10ms_modulation_mae_db": 0.06866675418244177, "300–1000 Hz_40ms_log_rms_mae_db": 1.042407419608427, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06423506658597897, "300–1000 Hz_40ms_envelope_correlation": 0.996643164154936, "300–1000 Hz_40ms_modulation_mae_db": 0.22345157491203338, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8141523237778598, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1708869910230349, "1000–3000 Hz_10ms_envelope_correlation": 0.9788940744294239, "1000–3000 Hz_10ms_modulation_mae_db": 0.42976398486559764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1658266337275647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10245322049878061, "1000–3000 Hz_40ms_envelope_correlation": 0.9924721955967833, "1000–3000 Hz_40ms_modulation_mae_db": 0.9088088734903216, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1008617149468423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17831957098489443, "3000–6000 Hz_10ms_envelope_correlation": 0.9793009565960341, "3000–6000 Hz_10ms_modulation_mae_db": 0.1900042952762395, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2851209993212631, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10230035512057625, "3000–6000 Hz_40ms_envelope_correlation": 0.9928527123084324, "3000–6000 Hz_40ms_modulation_mae_db": 0.15330964390328247, "mr_stft_db": 6.021234584031327, "band_mae_db": 0.15016453540299296, "rms_gain_error_db": -0.10979496361222348, "output_peak": 0.46628335374441066}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.468219859252963, "全帯域_10ms_rms_envelope_nrmse": 0.09210044113405864, "全帯域_10ms_envelope_correlation": 0.9920141241654453, "全帯域_10ms_modulation_mae_db": 0.21568294265067955, "全帯域_40ms_log_rms_mae_db": 0.8194386986545814, "全帯域_40ms_rms_envelope_nrmse": 0.05052437180586205, "全帯域_40ms_envelope_correlation": 0.997486125400179, "全帯域_40ms_modulation_mae_db": 0.19696745211426225, "300–1000 Hz_10ms_log_rms_mae_db": 1.714369718861493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1298985935131973, "300–1000 Hz_10ms_envelope_correlation": 0.9872207700893171, "300–1000 Hz_10ms_modulation_mae_db": 0.21167234263161552, "300–1000 Hz_40ms_log_rms_mae_db": 1.1627806877594073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07417122463729867, "300–1000 Hz_40ms_envelope_correlation": 0.9955283906603182, "300–1000 Hz_40ms_modulation_mae_db": 0.20741624262868533, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7429323485548132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1742168707415613, "1000–3000 Hz_10ms_envelope_correlation": 0.9777309583314021, "1000–3000 Hz_10ms_modulation_mae_db": 0.2525998149645287, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1124247348475458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10423213829721298, "1000–3000 Hz_40ms_envelope_correlation": 0.9914016504506107, "1000–3000 Hz_40ms_modulation_mae_db": 0.36469332230204904, "3000–6000 Hz_10ms_log_rms_mae_db": 2.101774716606695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1597420412491825, "3000–6000 Hz_10ms_envelope_correlation": 0.9836013192000123, "3000–6000 Hz_10ms_modulation_mae_db": 0.9230983462039375, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3359798348084972, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0855499803960341, "3000–6000 Hz_40ms_envelope_correlation": 0.9954507823980384, "3000–6000 Hz_40ms_modulation_mae_db": 0.8819795763544785, "mr_stft_db": 5.996522513714459, "band_mae_db": 0.1566098931036498, "rms_gain_error_db": -0.09551425391454949, "output_peak": 0.5054296302273346}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3192293676535087, "全帯域_10ms_rms_envelope_nrmse": 0.08717455686082794, "全帯域_10ms_envelope_correlation": 0.992804255561741, "全帯域_10ms_modulation_mae_db": 0.21209725521752013, "全帯域_40ms_log_rms_mae_db": 0.6576072496685069, "全帯域_40ms_rms_envelope_nrmse": 0.043320859585057776, "全帯域_40ms_envelope_correlation": 0.9981340690901754, "全帯域_40ms_modulation_mae_db": 0.3126742662729181, "300–1000 Hz_10ms_log_rms_mae_db": 1.7083817617736483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12748828597068518, "300–1000 Hz_10ms_envelope_correlation": 0.987629008247331, "300–1000 Hz_10ms_modulation_mae_db": 0.31097983470558815, "300–1000 Hz_40ms_log_rms_mae_db": 1.1270064384749745, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07104938726605026, "300–1000 Hz_40ms_envelope_correlation": 0.9958718150914717, "300–1000 Hz_40ms_modulation_mae_db": 0.1335617339162616, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8115509011199484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16655240847556838, "1000–3000 Hz_10ms_envelope_correlation": 0.9801872050115499, "1000–3000 Hz_10ms_modulation_mae_db": 1.2983761821584359, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1980584209182035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09623965828333543, "1000–3000 Hz_40ms_envelope_correlation": 0.9937401238873212, "1000–3000 Hz_40ms_modulation_mae_db": 1.4373814268204101, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1392458125528, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15033785043871087, "3000–6000 Hz_10ms_envelope_correlation": 0.9853204323349267, "3000–6000 Hz_10ms_modulation_mae_db": 0.507344080635801, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3300807545779905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08611586128422252, "3000–6000 Hz_40ms_envelope_correlation": 0.9950489015603716, "3000–6000 Hz_40ms_modulation_mae_db": 0.5975611518222758, "mr_stft_db": 6.02068291387094, "band_mae_db": 0.1857342239922645, "rms_gain_error_db": -0.07634586484010397, "output_peak": 0.49122573778278605}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4514808141736575, "全帯域_10ms_rms_envelope_nrmse": 0.091734134543474, "全帯域_10ms_envelope_correlation": 0.9920126139680904, "全帯域_10ms_modulation_mae_db": 0.20395695110080156, "全帯域_40ms_log_rms_mae_db": 0.7240096108498267, "全帯域_40ms_rms_envelope_nrmse": 0.04896544704499559, "全帯域_40ms_envelope_correlation": 0.9975774016823218, "全帯域_40ms_modulation_mae_db": 0.15445330061435739, "300–1000 Hz_10ms_log_rms_mae_db": 1.6695925871362616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12622141071851875, "300–1000 Hz_10ms_envelope_correlation": 0.9878231532919163, "300–1000 Hz_10ms_modulation_mae_db": 0.33177658500138824, "300–1000 Hz_40ms_log_rms_mae_db": 1.065360235892252, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06947550065886518, "300–1000 Hz_40ms_envelope_correlation": 0.9961184923691379, "300–1000 Hz_40ms_modulation_mae_db": 0.6151436776500384, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7621836648263636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1623496678012708, "1000–3000 Hz_10ms_envelope_correlation": 0.9807282347291749, "1000–3000 Hz_10ms_modulation_mae_db": 0.39584122406951733, "1000–3000 Hz_40ms_log_rms_mae_db": 1.146987255776035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08720519533153533, "1000–3000 Hz_40ms_envelope_correlation": 0.9943095615681015, "1000–3000 Hz_40ms_modulation_mae_db": 0.38618687179713324, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0746273548421343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1630142216086865, "3000–6000 Hz_10ms_envelope_correlation": 0.9837666062324874, "3000–6000 Hz_10ms_modulation_mae_db": 0.26557975400275285, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2785904156076617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09189756550529457, "3000–6000 Hz_40ms_envelope_correlation": 0.994937072005641, "3000–6000 Hz_40ms_modulation_mae_db": 0.49624259397231096, "mr_stft_db": 6.021228716423864, "band_mae_db": 0.1460944359084703, "rms_gain_error_db": -0.06481732368198632, "output_peak": 0.5240570707831407}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3586550027697102, "全帯域_10ms_rms_envelope_nrmse": 0.09954274202308942, "全帯域_10ms_envelope_correlation": 0.9906251494257612, "全帯域_10ms_modulation_mae_db": 0.1510361886782362, "全帯域_40ms_log_rms_mae_db": 0.6808321738055855, "全帯域_40ms_rms_envelope_nrmse": 0.04993519221196345, "全帯域_40ms_envelope_correlation": 0.997513968043861, "全帯域_40ms_modulation_mae_db": 0.306346061520116, "300–1000 Hz_10ms_log_rms_mae_db": 1.74057480419894, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13638637856620706, "300–1000 Hz_10ms_envelope_correlation": 0.9857575407440556, "300–1000 Hz_10ms_modulation_mae_db": 0.3246633961663845, "300–1000 Hz_40ms_log_rms_mae_db": 1.1903303502015483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08463016124927061, "300–1000 Hz_40ms_envelope_correlation": 0.9941670074738812, "300–1000 Hz_40ms_modulation_mae_db": 0.5186981183971987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7868612564384743, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19265833734837837, "1000–3000 Hz_10ms_envelope_correlation": 0.9728584609251377, "1000–3000 Hz_10ms_modulation_mae_db": 0.6414604577513329, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2057741986048511, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10883089917857174, "1000–3000 Hz_40ms_envelope_correlation": 0.9913891273012191, "1000–3000 Hz_40ms_modulation_mae_db": 0.7057792245770308, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0937229820313332, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1769207559329606, "3000–6000 Hz_10ms_envelope_correlation": 0.9796932539210125, "3000–6000 Hz_10ms_modulation_mae_db": 0.131733626627355, "3000–6000 Hz_40ms_log_rms_mae_db": 1.28532828091133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10041322403561224, "3000–6000 Hz_40ms_envelope_correlation": 0.9931175629772494, "3000–6000 Hz_40ms_modulation_mae_db": 0.22917090442800325, "mr_stft_db": 6.0191781730377985, "band_mae_db": 0.23601284316175275, "rms_gain_error_db": -0.08110282475906268, "output_peak": 0.4942573603440653}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.390469033221302, "全帯域_10ms_rms_envelope_nrmse": 0.08780227270691895, "全帯域_10ms_envelope_correlation": 0.9926918978495495, "全帯域_10ms_modulation_mae_db": 0.23668474356835878, "全帯域_40ms_log_rms_mae_db": 0.7753152267210291, "全帯域_40ms_rms_envelope_nrmse": 0.0465849384300296, "全帯域_40ms_envelope_correlation": 0.9978020428828019, "全帯域_40ms_modulation_mae_db": 0.31045666292821794, "300–1000 Hz_10ms_log_rms_mae_db": 1.656104516742359, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12056772935712742, "300–1000 Hz_10ms_envelope_correlation": 0.9889107646425986, "300–1000 Hz_10ms_modulation_mae_db": 0.4053029770257855, "300–1000 Hz_40ms_log_rms_mae_db": 1.0466748595396334, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06367074891905232, "300–1000 Hz_40ms_envelope_correlation": 0.9966832066826737, "300–1000 Hz_40ms_modulation_mae_db": 0.705410860258547, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7379728626520676, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1681967855533778, "1000–3000 Hz_10ms_envelope_correlation": 0.9792136819239556, "1000–3000 Hz_10ms_modulation_mae_db": 0.5300170698967168, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1696660228263687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10519877580192866, "1000–3000 Hz_40ms_envelope_correlation": 0.991351863509416, "1000–3000 Hz_40ms_modulation_mae_db": 0.5909633088169417, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0964352345924313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.171271486803979, "3000–6000 Hz_10ms_envelope_correlation": 0.9809327450387626, "3000–6000 Hz_10ms_modulation_mae_db": 0.42578553746772, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3018472013431197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08645455085082815, "3000–6000 Hz_40ms_envelope_correlation": 0.9949625671415693, "3000–6000 Hz_40ms_modulation_mae_db": 0.6954970283313576, "mr_stft_db": 6.0021953876754255, "band_mae_db": 0.17868114384032283, "rms_gain_error_db": -0.05875555931410801, "output_peak": 0.49798416398402107}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.232750546231528, "spectral_convergence": 0.3414379501133128, "spectral_convergence_status": "ok", "amplitude_floor": 1.3425996786840756e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.812875392504542, "spectral_convergence": 0.3283425867146186, "spectral_convergence_status": "ok", "amplitude_floor": 1.1106411258473627e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.988413039019585, "spectral_convergence": 0.3521334322578276, "spectral_convergence_status": "ok", "amplitude_floor": 1.0520318003394391e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00033395677419724207, "output_power": 0.0003482247049692358, "delta_db": 0.1816932170237706, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008500397987721807, "output_power": 0.0008313781635616135, "delta_db": -0.09640646387906929, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006268367014596879, "output_power": 0.0006266894727564931, "delta_db": -0.0010201719754167507, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003695716856827319, "output_power": 0.00037489813315533726, "delta_db": 0.062145869705165, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013451399439061329, "output_power": 0.00013043512169384565, "delta_db": -0.13372921533425838, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011413253897640505, "output_power": 0.00012044809242177482, "delta_db": 0.23390447688982757, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11814990246791794, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.005400078237815243, "pitch": null, "output_rms_dbfs": -26.140232706618733, "output_peak": 0.49564794536866247, "reference_rms_dbfs": -26.14563278485655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.011346325918552}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018099058448312497, "log_rms_mae_db": 1.3838317740755839, "rms_envelope_nrmse": 0.09534115364976303, "envelope_correlation": 0.9913944586854666, "rms_gain_error_db": 0.005397999083880922, "reference_mean_power": 0.002431072908198481, "output_mean_power": 0.0024340964529898537}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.757963657265947e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003340894048273399, "output_power": 0.00034844940451437677, "reference_active": true, "delta_db": 0.18277023843583484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010057920506755965, "output_power": 9.776168670802729e-05, "reference_active": true, "delta_db": -0.12339512872023466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.918423836914529e-05, "output_power": 1.9702011281856416e-05, "reference_active": true, "delta_db": 0.11566001411700588}], "active_band_count": 3, "active_band_mae_db": 0.14060846042435846}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016835992502135956, "log_rms_mae_db": 0.7427490014897609, "rms_envelope_nrmse": 0.05178240159127826, "envelope_correlation": 0.9972442316723727, "rms_gain_error_db": 0.005508448019862429, "reference_mean_power": 0.002443196853891242, "output_mean_power": 0.002446297690311189}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.921742652983223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003035779346246488, "output_power": 0.0003188050859601179, "reference_active": true, "delta_db": 0.21255039221026925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.67325523785903e-05, "output_power": 4.4344756612250365e-05, "reference_active": true, "delta_db": -0.2277722581113344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.222466914385115e-06, "output_power": 1.1755063393472448e-06, "reference_active": true, "delta_db": -0.17012138376316502}], "active_band_count": 3, "active_band_mae_db": 0.20348134469492288}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015437420870825604, "log_rms_mae_db": 1.655577633409515, "rms_envelope_nrmse": 0.13891218980446304, "envelope_correlation": 0.9853587358306028, "rms_gain_error_db": 0.03388839619233376, "reference_mean_power": 0.0008784594297426283, "output_mean_power": 0.0008853409427520747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.115238295382225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001974652277483934, "output_power": 0.00020330410313025026, "reference_active": true, "delta_db": 0.12655513296790136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.173113277871683e-05, "output_power": 5.03586624843829e-05, "reference_active": true, "delta_db": -0.11677801627272487}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0925495372248406e-05, "output_power": 1.1111265537661913e-05, "reference_active": true, "delta_db": 0.07322389102972056}], "active_band_count": 3, "active_band_mae_db": 0.10551901342344892}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012941193693393203, "log_rms_mae_db": 1.141764603635641, "rms_envelope_nrmse": 0.08226031442697844, "envelope_correlation": 0.9944994332594529, "rms_gain_error_db": 0.033891718496033514, "reference_mean_power": 0.000882875660341548, "output_mean_power": 0.0008897924490823837}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7308162041776265e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018289387957124914, "output_power": 0.00019000657562798375, "reference_active": true, "delta_db": 0.16569458675343932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5176917167074748e-05, "output_power": 2.4566619182169467e-05, "reference_active": true, "delta_db": -0.10657157359947343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.42204663895331e-07, "output_power": 7.623477616451205e-07, "reference_active": true, "delta_db": -0.9199212016755242}], "active_band_count": 3, "active_band_mae_db": 0.3973957873428123}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.92340600003817e-05, "log_rms_mae_db": 1.761154567267277, "rms_envelope_nrmse": 0.17578751025399347, "envelope_correlation": 0.9773765630095298, "rms_gain_error_db": -0.027458215713381806, "reference_mean_power": 0.000424662888055856, "output_mean_power": 0.00042198643196548374}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0384779490480225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014012639512762435, "output_power": 0.00013814959490620582, "reference_active": true, "delta_db": -0.061703337712210815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6627820389923143e-05, "output_power": 3.272655328768607e-05, "reference_active": true, "delta_db": -0.4891080742878544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.517673552823662e-06, "output_power": 8.013984234036504e-06, "reference_active": true, "delta_db": -0.2647250763939547}], "active_band_count": 3, "active_band_mae_db": 0.27184549613133996}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.48214657797314e-05, "log_rms_mae_db": 1.1681534768931754, "rms_envelope_nrmse": 0.09887205669760576, "envelope_correlation": 0.9922497058417545, "rms_gain_error_db": -0.027464785284822185, "reference_mean_power": 0.0004267980543892402, "output_mean_power": 0.0004241074997735666}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.813635389636035e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013656501467579018, "output_power": 0.0001345432215024607, "reference_active": true, "delta_db": -0.06477633502779476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7671441991318854e-05, "output_power": 1.5243532800542446e-05, "reference_active": true, "delta_db": -0.6418635982252359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.379326657430796e-07, "output_power": 4.4734866308237344e-07, "reference_active": true, "delta_db": -0.8008177366245418}], "active_band_count": 3, "active_band_mae_db": 0.5024858899591909}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.51121938209953e-05, "log_rms_mae_db": 2.061823100332808, "rms_envelope_nrmse": 0.16284455783771895, "envelope_correlation": 0.9833040027315411, "rms_gain_error_db": 0.13887408431082793, "reference_mean_power": 0.00015588925116025816, "output_mean_power": 0.00016095466899545646}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103456000035232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4374604191221866e-05, "output_power": 5.825822091411622e-05, "reference_active": true, "delta_db": 0.29961109419993054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5832401077981513e-05, "output_power": 1.540960487553514e-05, "reference_active": true, "delta_db": -0.11755280278518958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.903828543673632e-06, "output_power": 3.373703887485381e-06, "reference_active": true, "delta_db": -0.6338377360600255}], "active_band_count": 3, "active_band_mae_db": 0.35033387768171514}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091623719145545e-05, "log_rms_mae_db": 1.3176821509464542, "rms_envelope_nrmse": 0.09494507916847582, "envelope_correlation": 0.9941821711979419, "rms_gain_error_db": 0.13886251160586258, "reference_mean_power": 0.00015667279701186957, "output_mean_power": 0.00016176324409393895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1374269590951097e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.419500527522151e-05, "output_power": 5.7913848846480986e-05, "reference_active": true, "delta_db": 0.28823165397548667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661491941713126e-06, "output_power": 6.810016503952872e-06, "reference_active": true, "delta_db": 0.09576657547490262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7948760162947e-07, "output_power": 2.1727496573704783e-07, "reference_active": true, "delta_db": -1.0935285676971231}], "active_band_count": 3, "active_band_mae_db": 0.4925089323825042}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7709495031864387, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "398c7a981c5a07260c9e2f3d2242e7b7ccac4a3b19f30e1024b1ad0e55680b65", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9783530377704026, "standard_deviation": 0.0025004742004736706, "minimum": 0.9728584609251377, "maximum": 0.9807282347291749, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7819114750700833, "standard_deviation": 0.03667017951488862, "minimum": 1.7379728626520676, "maximum": 1.838483875923862, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5206160649682267, "standard_deviation": 0.33974305103519825, "minimum": 0.2525998149645287, "maximum": 1.2983761821584359, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1723184166481619, "standard_deviation": 0.00926239775973299, "minimum": 0.1623496678012708, "maximum": 0.19265833734837837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9926099957559563, "standard_deviation": 0.001235548628887228, "minimum": 0.991351863509416, "maximum": 0.9943095615681015, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1698965120679765, "standard_deviation": 0.030305296590305786, "minimum": 1.1124247348475458, "maximum": 1.2057741986048511, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6352950425015667, "standard_deviation": 0.39275855804746035, "minimum": 0.1860614222494551, "maximum": 1.4373814268204101, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09974990618421986, "standard_deviation": 0.006896048844947468, "minimum": 0.08720519533153533, "maximum": 0.10883089917857174, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9824710245790393, "standard_deviation": 0.002197375280221794, "minimum": 0.9793009565960341, "maximum": 0.9853204323349267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0939559670640593, "standard_deviation": 0.022956266779351926, "minimum": 2.061823100332808, "maximum": 2.1392458125528, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.41781949519272205, "standard_deviation": 0.25127802028268126, "minimum": 0.131733626627355, "maximum": 0.9230983462039375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1652759134294086, "standard_deviation": 0.009538963932861525, "minimum": 0.15033785043871087, "maximum": 0.17831957098489443, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944643785456834, "standard_deviation": 0.0009831251553974942, "minimum": 0.9928527123084324, "maximum": 0.9954507823980384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.304927278699028, "standard_deviation": 0.021511429655739808, "minimum": 1.2785904156076617, "maximum": 1.3359798348084972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5506042621382243, "standard_deviation": 0.2660130236909539, "minimum": 0.15330964390328247, "maximum": 0.8819795763544785, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09172500300135322, "standard_deviation": 0.006829252593055731, "minimum": 0.0855499803960341, "maximum": 0.10230035512057625, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9873529261141523, "standard_deviation": 0.0016061608257999596, "minimum": 0.9853587358306028, "maximum": 0.990035266337289, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6732361358012295, "standard_deviation": 0.048250160488339015, "minimum": 1.5840078907755397, "maximum": 1.74057480419894, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.2893444922803782, "standard_deviation": 0.15887296716720203, "minimum": 0.06866675418244177, "maximum": 0.5561750351063731, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.128643213215853, "standard_deviation": 0.008332718618699183, "minimum": 0.11440373161581852, "maximum": 0.13891218980446304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954498120837412, "standard_deviation": 0.001067898182895733, "minimum": 0.9940869869780579, "maximum": 0.9966832066826737, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1106222230133107, "standard_deviation": 0.05491191731642549, "minimum": 1.042407419608427, "maximum": 1.1903303502015483, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.43667358865346695, "standard_deviation": 0.22915738734455263, "minimum": 0.1335617339162616, "maximum": 0.705410860258547, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07451447280133414, "standard_deviation": 0.009021555406201472, "minimum": 0.06367074891905232, "maximum": 0.08662337866717869, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17283869978792818, "standard_deviation": 0.03808151194654797, "minimum": 0.11814990246791794, "maximum": 0.23601284316175275, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.012729243339601, "standard_deviation": 0.009530013989270886, "minimum": 5.996522513714459, "maximum": 6.021234584031327, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5022861835849877, "standard_deviation": 0.023053098304054224, "minimum": 0.46628335374441066, "maximum": 0.5434042064454803, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0751451356567113, "standard_deviation": 0.03881814638204837, "minimum": -0.1202303733694716, "maximum": 0.005400078237815243, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9920461957234368, "standard_deviation": 0.0007885615231860534, "minimum": 0.9906251494257612, "maximum": 0.9929814483168073, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3953634953505105, "standard_deviation": 0.057509730601949374, "minimum": 1.3192293676535087, "maximum": 1.468219859252963, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.19575585233281567, "standard_deviation": 0.06060495848104201, "minimum": 0.10770527227544396, "maximum": 0.2982750047471268, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09165308869810568, "standard_deviation": 0.004479417009942485, "minimum": 0.08641932918981055, "maximum": 0.09954274202308942, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976721398458448, "standard_deviation": 0.00028383939804898367, "minimum": 0.9972442316723727, "maximum": 0.9981340690901754, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7237357869044028, "standard_deviation": 0.05814466331659584, "minimum": 0.6524490565636386, "maximum": 0.8194386986545814, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.27553665181467446, "standard_deviation": 0.09058568536255948, "minimum": 0.15445330061435739, "maximum": 0.4424575461004084, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048314710139112205, "standard_deviation": 0.0027735833298925855, "minimum": 0.043320859585057776, "maximum": 0.05178240159127826, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0002", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.0164971351623535, "predicted_mos": 2.0164971351623535, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a2ae690286fb2d9eaf777e5d3d5d62620819be8d3bcd4148f5698002b895d3cc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.753609500359744, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4074628271950242, "全帯域_10ms_rms_envelope_nrmse": 0.09295022994795543, "全帯域_10ms_envelope_correlation": 0.991872016525611, "全帯域_10ms_modulation_mae_db": 0.22296929804257892, "全帯域_40ms_log_rms_mae_db": 0.7507576282523883, "全帯域_40ms_rms_envelope_nrmse": 0.052355027622318646, "全帯域_40ms_envelope_correlation": 0.9973011054700812, "全帯域_40ms_modulation_mae_db": 0.46427727762233895, "300–1000 Hz_10ms_log_rms_mae_db": 1.5564504193526416, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11432046904513049, "300–1000 Hz_10ms_envelope_correlation": 0.9900225884696908, "300–1000 Hz_10ms_modulation_mae_db": 0.0880005977575803, "300–1000 Hz_40ms_log_rms_mae_db": 0.9903006646494326, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05979709830402301, "300–1000 Hz_40ms_envelope_correlation": 0.9971249081658377, "300–1000 Hz_40ms_modulation_mae_db": 0.3622572477786073, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6969336383226392, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18853748584998273, "1000–3000 Hz_10ms_envelope_correlation": 0.9738168995728411, "1000–3000 Hz_10ms_modulation_mae_db": 0.6314468778856743, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0961424180364645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11308811269723722, "1000–3000 Hz_40ms_envelope_correlation": 0.9899960528083854, "1000–3000 Hz_40ms_modulation_mae_db": 0.6926144782633191, "3000–6000 Hz_10ms_log_rms_mae_db": 2.024986862470253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16163176325489464, "3000–6000 Hz_10ms_envelope_correlation": 0.9830461833491705, "3000–6000 Hz_10ms_modulation_mae_db": 0.8253889442201338, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2590152955965659, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08620509418947132, "3000–6000 Hz_40ms_envelope_correlation": 0.994951331325958, "3000–6000 Hz_40ms_modulation_mae_db": 1.1113779365448673, "mr_stft_db": 5.963903084207451, "band_mae_db": 0.14932239024517827, "rms_gain_error_db": -0.10078709118625007, "output_peak": 0.5620758468027767}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4048986834021866, "全帯域_10ms_rms_envelope_nrmse": 0.09564905579563315, "全帯域_10ms_envelope_correlation": 0.9913476817159677, "全帯域_10ms_modulation_mae_db": 0.07911893195806304, "全帯域_40ms_log_rms_mae_db": 0.7733155174245426, "全帯域_40ms_rms_envelope_nrmse": 0.04979126712505177, "全帯域_40ms_envelope_correlation": 0.9975200603742467, "全帯域_40ms_modulation_mae_db": 0.42457256850126174, "300–1000 Hz_10ms_log_rms_mae_db": 1.6426213903528848, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13018233676375088, "300–1000 Hz_10ms_envelope_correlation": 0.9870353990546148, "300–1000 Hz_10ms_modulation_mae_db": 0.5097909527020872, "300–1000 Hz_40ms_log_rms_mae_db": 1.0379670234626468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0750655937681588, "300–1000 Hz_40ms_envelope_correlation": 0.9953970203159483, "300–1000 Hz_40ms_modulation_mae_db": 0.7597259241490862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6604649617748148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18388194395291155, "1000–3000 Hz_10ms_envelope_correlation": 0.9751216719842514, "1000–3000 Hz_10ms_modulation_mae_db": 0.40991252563248076, "1000–3000 Hz_40ms_log_rms_mae_db": 1.025760304960553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10125395612410629, "1000–3000 Hz_40ms_envelope_correlation": 0.9919996994691415, "1000–3000 Hz_40ms_modulation_mae_db": 0.5339904600454302, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0178890690983478, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1718535848830291, "3000–6000 Hz_10ms_envelope_correlation": 0.9815138160955422, "3000–6000 Hz_10ms_modulation_mae_db": 0.40513749749844585, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2311759829124618, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09634769745548447, "3000–6000 Hz_40ms_envelope_correlation": 0.9940802674757115, "3000–6000 Hz_40ms_modulation_mae_db": 0.48268907507113745, "mr_stft_db": 5.957042869200841, "band_mae_db": 0.1485789091496711, "rms_gain_error_db": -0.07923874533059698, "output_peak": 0.5210595603634758}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.357274939596584, "全帯域_10ms_rms_envelope_nrmse": 0.09294211903242408, "全帯域_10ms_envelope_correlation": 0.9919479111106182, "全帯域_10ms_modulation_mae_db": 0.41725510856352027, "全帯域_40ms_log_rms_mae_db": 0.6906400321348887, "全帯域_40ms_rms_envelope_nrmse": 0.054350786073251646, "全帯域_40ms_envelope_correlation": 0.9972118255830447, "全帯域_40ms_modulation_mae_db": 0.6926603296981497, "300–1000 Hz_10ms_log_rms_mae_db": 1.6508025017506653, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12667111081041207, "300–1000 Hz_10ms_envelope_correlation": 0.9878444949228611, "300–1000 Hz_10ms_modulation_mae_db": 0.23324463939114426, "300–1000 Hz_40ms_log_rms_mae_db": 1.0244338654551635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07424051256626793, "300–1000 Hz_40ms_envelope_correlation": 0.9957627329755582, "300–1000 Hz_40ms_modulation_mae_db": 0.6545603752237634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.698560564395744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17686677129404602, "1000–3000 Hz_10ms_envelope_correlation": 0.9769758859260174, "1000–3000 Hz_10ms_modulation_mae_db": 0.8229774746436226, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0077919328702225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10310539616153795, "1000–3000 Hz_40ms_envelope_correlation": 0.9916703425026219, "1000–3000 Hz_40ms_modulation_mae_db": 1.0563037433735225, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0160230272506254, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18015513598973615, "3000–6000 Hz_10ms_envelope_correlation": 0.9804010139727241, "3000–6000 Hz_10ms_modulation_mae_db": 0.8127793385158973, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2128963832107666, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10832919333764977, "3000–6000 Hz_40ms_envelope_correlation": 0.994420366558762, "3000–6000 Hz_40ms_modulation_mae_db": 1.0484458283862341, "mr_stft_db": 5.968154866139421, "band_mae_db": 0.20948231075002086, "rms_gain_error_db": -0.14907412774584014, "output_peak": 0.4762336481109517}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2728022264120546, "全帯域_10ms_rms_envelope_nrmse": 0.09393690518983426, "全帯域_10ms_envelope_correlation": 0.9916246710468968, "全帯域_10ms_modulation_mae_db": 0.24499065451109406, "全帯域_40ms_log_rms_mae_db": 0.6356942486413417, "全帯域_40ms_rms_envelope_nrmse": 0.048732965480843216, "全帯域_40ms_envelope_correlation": 0.9976032991187481, "全帯域_40ms_modulation_mae_db": 0.18969006657884882, "300–1000 Hz_10ms_log_rms_mae_db": 1.6091607184386794, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13517501000294263, "300–1000 Hz_10ms_envelope_correlation": 0.9860166466544246, "300–1000 Hz_10ms_modulation_mae_db": 0.32011131021226974, "300–1000 Hz_40ms_log_rms_mae_db": 0.9881093519964473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07030637637314459, "300–1000 Hz_40ms_envelope_correlation": 0.9960446972150763, "300–1000 Hz_40ms_modulation_mae_db": 0.10831426741411476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6927856607214533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18007487482176007, "1000–3000 Hz_10ms_envelope_correlation": 0.9761610454733277, "1000–3000 Hz_10ms_modulation_mae_db": 0.2295675775455109, "1000–3000 Hz_40ms_log_rms_mae_db": 1.087866572783452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10275273413012621, "1000–3000 Hz_40ms_envelope_correlation": 0.991885137160999, "1000–3000 Hz_40ms_modulation_mae_db": 0.44418248705481583, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9771412127743322, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15546087759334465, "3000–6000 Hz_10ms_envelope_correlation": 0.9844394402086111, "3000–6000 Hz_10ms_modulation_mae_db": 0.6005080610079051, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1772055701973212, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08435471395153062, "3000–6000 Hz_40ms_envelope_correlation": 0.9951808472715868, "3000–6000 Hz_40ms_modulation_mae_db": 0.6558068369907813, "mr_stft_db": 5.96830660890127, "band_mae_db": 0.12206505344486453, "rms_gain_error_db": -0.06872248714458698, "output_peak": 0.4655604760316303}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.366729961813685, "全帯域_10ms_rms_envelope_nrmse": 0.09597684569193217, "全帯域_10ms_envelope_correlation": 0.9913123010590492, "全帯域_10ms_modulation_mae_db": 0.11458249633070339, "全帯域_40ms_log_rms_mae_db": 0.6673872482330185, "全帯域_40ms_rms_envelope_nrmse": 0.048962932840472824, "全帯域_40ms_envelope_correlation": 0.99766759486835, "全帯域_40ms_modulation_mae_db": 0.05735502118287824, "300–1000 Hz_10ms_log_rms_mae_db": 1.5228108044735105, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13128473419611972, "300–1000 Hz_10ms_envelope_correlation": 0.9868076164602271, "300–1000 Hz_10ms_modulation_mae_db": 0.2263682045660007, "300–1000 Hz_40ms_log_rms_mae_db": 0.9513824117544946, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07142756487137811, "300–1000 Hz_40ms_envelope_correlation": 0.995850558257626, "300–1000 Hz_40ms_modulation_mae_db": 0.38809776264918544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.675544884408506, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15766068966562197, "1000–3000 Hz_10ms_envelope_correlation": 0.9817925593738706, "1000–3000 Hz_10ms_modulation_mae_db": 0.6052137153707597, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0635907620198122, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09031980708014378, "1000–3000 Hz_40ms_envelope_correlation": 0.9937411973526882, "1000–3000 Hz_40ms_modulation_mae_db": 0.5463585706547621, "3000–6000 Hz_10ms_log_rms_mae_db": 2.002252754059755, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1636731042752581, "3000–6000 Hz_10ms_envelope_correlation": 0.9826493536175643, "3000–6000 Hz_10ms_modulation_mae_db": 0.23486961566192563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2080200289511493, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09068042042455249, "3000–6000 Hz_40ms_envelope_correlation": 0.9943919345775982, "3000–6000 Hz_40ms_modulation_mae_db": 0.49235743048816305, "mr_stft_db": 5.949200504755656, "band_mae_db": 0.1447097075951719, "rms_gain_error_db": -0.10983060589559605, "output_peak": 0.5178114464255077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3800582719000352, "全帯域_10ms_rms_envelope_nrmse": 0.09591918790393172, "全帯域_10ms_envelope_correlation": 0.9912792235169533, "全帯域_10ms_modulation_mae_db": 0.24807192407941128, "全帯域_40ms_log_rms_mae_db": 0.7220331151763544, "全帯域_40ms_rms_envelope_nrmse": 0.05112793838931139, "全帯域_40ms_envelope_correlation": 0.9973101689519195, "全帯域_40ms_modulation_mae_db": 0.5049916822044027, "300–1000 Hz_10ms_log_rms_mae_db": 1.6221280030058791, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11798367009902683, "300–1000 Hz_10ms_envelope_correlation": 0.9893786627800621, "300–1000 Hz_10ms_modulation_mae_db": 0.36501565204292313, "300–1000 Hz_40ms_log_rms_mae_db": 1.0462092989612592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06665286185052538, "300–1000 Hz_40ms_envelope_correlation": 0.9963634225907613, "300–1000 Hz_40ms_modulation_mae_db": 0.8725985317247137, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6446030011206993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.173367007129331, "1000–3000 Hz_10ms_envelope_correlation": 0.9783550773506602, "1000–3000 Hz_10ms_modulation_mae_db": 0.5652173567590628, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9867606381353203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10600039788800442, "1000–3000 Hz_40ms_envelope_correlation": 0.991278509520959, "1000–3000 Hz_40ms_modulation_mae_db": 0.7043722544212851, "3000–6000 Hz_10ms_log_rms_mae_db": 2.025722533678287, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1797466395276813, "3000–6000 Hz_10ms_envelope_correlation": 0.9789264143210422, "3000–6000 Hz_10ms_modulation_mae_db": 0.3575928146325336, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2240406046848442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09416360199679029, "3000–6000 Hz_40ms_envelope_correlation": 0.9941223180747335, "3000–6000 Hz_40ms_modulation_mae_db": 0.512243150479247, "mr_stft_db": 5.949068118125015, "band_mae_db": 0.14920799549643363, "rms_gain_error_db": -0.004729927595861074, "output_peak": 0.5105415124331573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4413693590579921, "全帯域_10ms_rms_envelope_nrmse": 0.08633628550722558, "全帯域_10ms_envelope_correlation": 0.9930253594181608, "全帯域_10ms_modulation_mae_db": 0.2898691961979905, "全帯域_40ms_log_rms_mae_db": 0.7591446717674726, "全帯域_40ms_rms_envelope_nrmse": 0.04370862751368468, "全帯域_40ms_envelope_correlation": 0.9982071207318522, "全帯域_40ms_modulation_mae_db": 0.31358538489278226, "300–1000 Hz_10ms_log_rms_mae_db": 1.6048292516769036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1273537333484736, "300–1000 Hz_10ms_envelope_correlation": 0.9876518801474957, "300–1000 Hz_10ms_modulation_mae_db": 0.6448352324451966, "300–1000 Hz_40ms_log_rms_mae_db": 0.9993868743243276, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07135599934583345, "300–1000 Hz_40ms_envelope_correlation": 0.9959997452709025, "300–1000 Hz_40ms_modulation_mae_db": 0.9660791989323846, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6812620681701869, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16216770933139515, "1000–3000 Hz_10ms_envelope_correlation": 0.9807190598693151, "1000–3000 Hz_10ms_modulation_mae_db": 0.5320819184132383, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0133805093462755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08309895745666471, "1000–3000 Hz_40ms_envelope_correlation": 0.9947754365975591, "1000–3000 Hz_40ms_modulation_mae_db": 0.65826583591334, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0097523906001036, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13892316950531924, "3000–6000 Hz_10ms_envelope_correlation": 0.9875755798824395, "3000–6000 Hz_10ms_modulation_mae_db": 0.45184670206398986, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2005143069168782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07335238255766457, "3000–6000 Hz_40ms_envelope_correlation": 0.9965382199554554, "3000–6000 Hz_40ms_modulation_mae_db": 0.6048181663595857, "mr_stft_db": 5.974977832779817, "band_mae_db": 0.1340948300825567, "rms_gain_error_db": -0.12025750705483139, "output_peak": 0.5129966031469321}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3574941991381084, "全帯域_10ms_rms_envelope_nrmse": 0.09349978526324018, "全帯域_10ms_envelope_correlation": 0.9917812968317001, "全帯域_10ms_modulation_mae_db": 0.17654849503020795, "全帯域_40ms_log_rms_mae_db": 0.6904333070088807, "全帯域_40ms_rms_envelope_nrmse": 0.04756426210044806, "全帯域_40ms_envelope_correlation": 0.997717837215347, "全帯域_40ms_modulation_mae_db": 0.21480231707728895, "300–1000 Hz_10ms_log_rms_mae_db": 1.6608253017705628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13611499874724378, "300–1000 Hz_10ms_envelope_correlation": 0.986138564696534, "300–1000 Hz_10ms_modulation_mae_db": 0.16708212276951542, "300–1000 Hz_40ms_log_rms_mae_db": 1.0236753926656832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07156797732402315, "300–1000 Hz_40ms_envelope_correlation": 0.9959481560703699, "300–1000 Hz_40ms_modulation_mae_db": 0.48116970309569473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.669402882270523, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15685933282349013, "1000–3000 Hz_10ms_envelope_correlation": 0.9820149993238105, "1000–3000 Hz_10ms_modulation_mae_db": 0.3435371703498793, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0019522461644044, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08513173965776777, "1000–3000 Hz_40ms_envelope_correlation": 0.9945613167626589, "1000–3000 Hz_40ms_modulation_mae_db": 0.5685023810741718, "3000–6000 Hz_10ms_log_rms_mae_db": 2.072779707011084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16881211635085466, "3000–6000 Hz_10ms_envelope_correlation": 0.9814432630108313, "3000–6000 Hz_10ms_modulation_mae_db": 0.93572617503272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2385790149359501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08770148876873747, "3000–6000 Hz_40ms_envelope_correlation": 0.9947652283888446, "3000–6000 Hz_40ms_modulation_mae_db": 0.8572808178894359, "mr_stft_db": 5.965889508349043, "band_mae_db": 0.15251013323371546, "rms_gain_error_db": 0.01158446892085285, "output_peak": 0.5469148393040125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.200657155891104, "spectral_convergence": 0.3379184481441776, "spectral_convergence_status": "ok", "amplitude_floor": 1.3425996786840756e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.772000098971314, "spectral_convergence": 0.32808326325734344, "spectral_convergence_status": "ok", "amplitude_floor": 1.1106411258473627e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.919051997759932, "spectral_convergence": 0.34424600441424275, "spectral_convergence_status": "ok", "amplitude_floor": 1.0520318003394391e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00033395677419724207, "output_power": 0.0003505893259724902, "delta_db": 0.21108432045060954, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008500397987721807, "output_power": 0.0007980797811703932, "delta_db": -0.27392951430360657, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006268367014596879, "output_power": 0.000623584356788951, "delta_db": -0.022592046409796205, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003695716856827319, "output_power": 0.0003602974149523883, "delta_db": -0.1103754466648178, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013451399439061329, "output_power": 0.00013002949554209446, "delta_db": -0.14725591571312777, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011413253897640505, "output_power": 0.00011074898861436553, "delta_db": -0.1306970979291118, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14932239024517827, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10078709118625007, "pitch": null, "output_rms_dbfs": -26.2464198760428, "output_peak": 0.5620758468027767, "reference_rms_dbfs": -26.14563278485655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.963903084207451}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018099058448312497, "log_rms_mae_db": 1.4074628271950242, "rms_envelope_nrmse": 0.09295022994795543, "envelope_correlation": 0.991872016525611, "rms_gain_error_db": -0.10078198708416153, "reference_mean_power": 0.002431072908198481, "output_mean_power": 0.0023753071983741722}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.757963657265947e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003340894048273399, "output_power": 0.00032130919138449244, "reference_active": true, "delta_db": -0.16939553645889588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010057920506755965, "output_power": 9.130215981503284e-05, "reference_active": true, "delta_db": -0.4202714764296822}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.918423836914529e-05, "output_power": 1.9537484664323424e-05, "reference_active": true, "delta_db": 0.0792408812391586}], "active_band_count": 3, "active_band_mae_db": 0.22296929804257892}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016835992502135956, "log_rms_mae_db": 0.7507576282523883, "rms_envelope_nrmse": 0.052355027622318646, "envelope_correlation": 0.9973011054700812, "rms_gain_error_db": -0.10070359898713116, "reference_mean_power": 0.002443196853891242, "output_mean_power": 0.0023871961235966176}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.921742652983223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003035779346246488, "output_power": 0.0002908886202858965, "reference_active": true, "delta_db": -0.1854347019004829}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.67325523785903e-05, "output_power": 4.196508877840848e-05, "reference_active": true, "delta_db": -0.4673135519777358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.222466914385115e-06, "output_power": 1.4495909595272687e-06, "reference_active": true, "delta_db": 0.7400835789887982}], "active_band_count": 3, "active_band_mae_db": 0.46427727762233895}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015437420870825604, "log_rms_mae_db": 1.5564504193526416, "rms_envelope_nrmse": 0.11432046904513049, "envelope_correlation": 0.9900225884696908, "rms_gain_error_db": -0.08792890657642191, "reference_mean_power": 0.0008784594297426283, "output_mean_power": 0.0008608526454842463}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.115238295382225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001974652277483934, "output_power": 0.00019949246214468955, "reference_active": true, "delta_db": 0.04435859983760836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.173113277871683e-05, "output_power": 4.945600807270728e-05, "reference_active": true, "delta_db": -0.19532929835879426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0925495372248406e-05, "output_power": 1.0864500091200565e-05, "reference_active": true, "delta_db": -0.024313895076338278}], "active_band_count": 3, "active_band_mae_db": 0.0880005977575803}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012941193693393203, "log_rms_mae_db": 0.9903006646494326, "rms_envelope_nrmse": 0.05979709830402301, "envelope_correlation": 0.9971249081658377, "rms_gain_error_db": -0.08792754259660845, "reference_mean_power": 0.000882875660341548, "output_mean_power": 0.0008651806341940567}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7308162041776265e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018289387957124914, "output_power": 0.00018530862162114577, "reference_active": true, "delta_db": 0.05696453334379393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5176917167074748e-05, "output_power": 2.4076983616989387e-05, "reference_active": true, "delta_db": -0.1940047381077811}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.42204663895331e-07, "output_power": 7.772576451274059e-07, "reference_active": true, "delta_db": -0.835802471884247}], "active_band_count": 3, "active_band_mae_db": 0.3622572477786073}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.92340600003817e-05, "log_rms_mae_db": 1.6969336383226392, "rms_envelope_nrmse": 0.18853748584998273, "envelope_correlation": 0.9738168995728411, "rms_gain_error_db": -0.17566253019440992, "reference_mean_power": 0.000424662888055856, "output_mean_power": 0.00040782895446657923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0384779490480225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014012639512762435, "output_power": 0.00011945256880567927, "reference_active": true, "delta_db": -0.6932445594117558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6627820389923143e-05, "output_power": 2.7785308770631875e-05, "reference_active": true, "delta_db": -1.1999584854556107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.517673552823662e-06, "output_power": 8.515442729864504e-06, "reference_active": true, "delta_db": -0.001137588789656449}], "active_band_count": 3, "active_band_mae_db": 0.6314468778856743}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.48214657797314e-05, "log_rms_mae_db": 1.0961424180364645, "rms_envelope_nrmse": 0.11308811269723722, "envelope_correlation": 0.9899960528083854, "rms_gain_error_db": -0.1756648612340599, "reference_mean_power": 0.0004267980543892402, "output_mean_power": 0.000409879261315337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.813635389636035e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013656501467579018, "output_power": 0.00011598415420210988, "reference_active": true, "delta_db": -0.7094079591226545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7671441991318854e-05, "output_power": 1.3331897934835848e-05, "reference_active": true, "delta_db": -1.2238000925509998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.379326657430796e-07, "output_power": 5.561493778842485e-07, "reference_active": true, "delta_db": 0.14463538311630342}], "active_band_count": 3, "active_band_mae_db": 0.6926144782633191}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.51121938209953e-05, "log_rms_mae_db": 2.024986862470253, "rms_envelope_nrmse": 0.16163176325489464, "envelope_correlation": 0.9830461833491705, "rms_gain_error_db": -0.04522367766709418, "reference_mean_power": 0.00015588925116025816, "output_mean_power": 0.00015427437763420207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103456000035232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4374604191221866e-05, "output_power": 5.199399540054293e-05, "reference_active": true, "delta_db": -0.19442917193052428}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5832401077981513e-05, "output_power": 1.3389298918508877e-05, "reference_active": true, "delta_db": -0.727889458327182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.903828543673632e-06, "output_power": 2.7296434020550713e-06, "reference_active": true, "delta_db": -1.5538482024026947}], "active_band_count": 3, "active_band_mae_db": 0.8253889442201338}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091623719145545e-05, "log_rms_mae_db": 1.2590152955965659, "rms_envelope_nrmse": 0.08620509418947132, "envelope_correlation": 0.994951331325958, "rms_gain_error_db": -0.045231471370467013, "reference_mean_power": 0.00015667279701186957, "output_mean_power": 0.00015504952840308673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1374269590951097e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.419500527522151e-05, "output_power": 5.142284269938503e-05, "reference_active": true, "delta_db": -0.22803181750529314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661491941713126e-06, "output_power": 5.512715993792479e-06, "reference_active": true, "delta_db": -0.8220588796481759}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7948760162947e-07, "output_power": 1.651802943288499e-07, "reference_active": true, "delta_db": -2.284043112481133}], "active_band_count": 3, "active_band_mae_db": 1.1113779365448673}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7805813716225983, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e738c9e31bd36f3a83bce8d7b4e65d6e809dbf9cb089b648e79a67a9ee559bf9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9781196498592617, "standard_deviation": 0.0031195748038903176, "minimum": 0.9738168995728411, "maximum": 0.9820149993238105, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6774447076480707, "standard_deviation": 0.01893894192578136, "minimum": 1.6446030011206993, "maximum": 1.698560564395744, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5174943270750285, "standard_deviation": 0.18563668942436798, "minimum": 0.2295675775455109, "maximum": 0.8229774746436226, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17242697685856734, "standard_deviation": 0.01216257664401726, "minimum": 0.15685933282349013, "maximum": 0.18853748584998273, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9924884615218765, "standard_deviation": 0.0016921931376484064, "minimum": 0.9899960528083854, "maximum": 0.9947754365975591, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.035405673039563, "standard_deviation": 0.041510116208227206, "minimum": 0.9867606381353203, "maximum": 1.0961424180364645, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.650573776350081, "standard_deviation": 0.1863649368230851, "minimum": 0.44418248705481583, "maximum": 1.0563037433735225, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09809388764944854, "standard_deviation": 0.010673380308978643, "minimum": 0.08309895745666471, "maximum": 0.11308811269723722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9824993830572407, "standard_deviation": 0.0026486942303192746, "minimum": 0.9789264143210422, "maximum": 0.9875755798824395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0183184446178486, "standard_deviation": 0.026986289783619898, "minimum": 1.9771412127743322, "maximum": 2.072779707011084, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.577981143579194, "standard_deviation": 0.25548585120301637, "minimum": 0.23486961566192563, "maximum": 0.93572617503272, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16503204892251472, "standard_deviation": 0.013589590448751125, "minimum": 0.13892316950531924, "maximum": 0.18015513598973615, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9948063142035813, "standard_deviation": 0.0007998349094531933, "minimum": 0.9940802674757115, "maximum": 0.9965382199554554, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.218930898425742, "standard_deviation": 0.02508412951943192, "minimum": 1.1772055701973212, "maximum": 1.2590152955965659, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7206274052761814, "standard_deviation": 0.25314715237951857, "minimum": 0.48268907507113745, "maximum": 1.1113779365448673, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09014182408523513, "standard_deviation": 0.010160055662349836, "minimum": 0.07335238255766457, "maximum": 0.10832919333764977, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9876119816482388, "standard_deviation": 0.001448316455223822, "minimum": 0.9860166466544246, "maximum": 0.9900225884696908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.608703548852716, "standard_deviation": 0.04772361973736111, "minimum": 1.5228108044735105, "maximum": 1.6608253017705628, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3193060889858397, "standard_deviation": 0.1840272141024736, "minimum": 0.0880005977575803, "maximum": 0.6448352324451966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1273857578766375, "standard_deviation": 0.007741426017622058, "minimum": 0.11432046904513049, "maximum": 0.13611499874724378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960614051077601, "standard_deviation": 0.0005090927976823044, "minimum": 0.9953970203159483, "maximum": 0.9971249081658377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0076831104086819, "standard_deviation": 0.03130860475454539, "minimum": 0.9513824117544946, "maximum": 1.0462092989612592, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5741003763709438, "standard_deviation": 0.2900114515633667, "minimum": 0.10831426741411476, "maximum": 0.9660791989323846, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0700517480504193, "standard_deviation": 0.004860223334535035, "minimum": 0.05979709830402301, "maximum": 0.0750655937681588, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15124641624970156, "standard_deviation": 0.025624894205944337, "minimum": 0.12206505344486453, "maximum": 0.20948231075002086, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.962067924057314, "standard_deviation": 0.00941447764323341, "minimum": 5.949068118125015, "maximum": 5.974977832779817, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5141492415773055, "standard_deviation": 0.03217033412276054, "minimum": 0.4655604760316303, "maximum": 0.5620758468027767, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07763200287908872, "standard_deviation": 0.05586983785729087, "minimum": -0.14907412774584014, "maximum": 0.01158446892085285, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9917738076531196, "standard_deviation": 0.0005691876103798829, "minimum": 0.9912792235169533, "maximum": 0.9930253594181608, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3735113085644588, "standard_deviation": 0.049991319283018, "minimum": 1.2728022264120546, "maximum": 1.4413693590579921, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.22417576308919618, "standard_deviation": 0.10550661970746154, "minimum": 0.07911893195806304, "maximum": 0.41725510856352027, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09340130179152206, "standard_deviation": 0.003131656673099348, "minimum": 0.08633628550722558, "maximum": 0.09597684569193217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975673765391988, "standard_deviation": 0.0003182444072925913, "minimum": 0.9972118255830447, "maximum": 0.9982071207318522, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7111757210798609, "standard_deviation": 0.04827574102337592, "minimum": 0.6356942486413417, "maximum": 0.7733155174245426, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35774183096974393, "standard_deviation": 0.2037239714215485, "minimum": 0.05735502118287824, "maximum": 0.6926603296981497, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04957422589317278, "standard_deviation": 0.0032190529197904166, "minimum": 0.04370862751368468, "maximum": 0.054350786073251646, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7127-75946-0029", "language": "en", "style": "normal", "speaker": "librispeech_7127", "source_utterance_id": "7127-75946-0029", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7127-75946-0029.wav", "source_file_sha256": "53a1d3b899a3e84ad12a95320898a370d3ab0b60bb5cfb7e5ccc8cdfb4d71394", "decoded_pcm_sha256": "92db6e9b9692ce0f9a8c0ad4504635e4a6e649b6d8760fca3400f3ea75d63e54", "original_source_sha256": "e26b6b1a8cceb2a05b3aae15719c785d4d24a2a09d91dd6292da8a4b4e32401a", "corpus_entry_id": "librispeech_7127-75946-0029"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 2.536989450454712, "predicted_mos": 2.536989450454712, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5a57ed2ebacc92b40830bf9a348d6aabd69d4af04f5e2c4fdc7112113f875372", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.137082934001228, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7127-75946-0029", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "5a57ed2ebacc92b40830bf9a348d6aabd69d4af04f5e2c4fdc7112113f875372", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "fb56543512bf946cb8e8680e6e9940fdf458e66eefce85babf56618639407acb"}, "rows": [{"case": "librispeech_7127-75946-0029", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.7738618850708008, "predicted_mos": 1.7738618850708008, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cd1ba5e2588a64a4a5d6054633f6ffbdf4e2c4a3c398bfc01e2b626567ad5f62", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.428462218002096, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7127-75946-0029", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "cd1ba5e2588a64a4a5d6054633f6ffbdf4e2c4a3c398bfc01e2b626567ad5f62", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3368349920904776, "全帯域_10ms_rms_envelope_nrmse": 0.09645992858033271, "全帯域_10ms_envelope_correlation": 0.9898604958130571, "全帯域_10ms_modulation_mae_db": 0.25721474790706883, "全帯域_40ms_log_rms_mae_db": 0.7151288870307932, "全帯域_40ms_rms_envelope_nrmse": 0.05287913137691423, "全帯域_40ms_envelope_correlation": 0.9969127147315747, "全帯域_40ms_modulation_mae_db": 0.5978366236231266, "300–1000 Hz_10ms_log_rms_mae_db": 1.535438454309173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13610314672615778, "300–1000 Hz_10ms_envelope_correlation": 0.9847000022697425, "300–1000 Hz_10ms_modulation_mae_db": 0.3989536883720733, "300–1000 Hz_40ms_log_rms_mae_db": 0.9556694919099502, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07697935707934347, "300–1000 Hz_40ms_envelope_correlation": 0.9948471720556119, "300–1000 Hz_40ms_modulation_mae_db": 0.2706019087777752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5333571272381616, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.142013372523658, "1000–3000 Hz_10ms_envelope_correlation": 0.9853137044940611, "1000–3000 Hz_10ms_modulation_mae_db": 0.5947455725246412, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9095325041136335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09120171069232914, "1000–3000 Hz_40ms_envelope_correlation": 0.9945148561063877, "1000–3000 Hz_40ms_modulation_mae_db": 0.7272769349952032, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8148549306539066, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18689685644389614, "3000–6000 Hz_10ms_envelope_correlation": 0.9755646154108505, "3000–6000 Hz_10ms_modulation_mae_db": 0.36573320842426, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0657585336238982, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10116978791527247, "3000–6000 Hz_40ms_envelope_correlation": 0.9928836745467785, "3000–6000 Hz_40ms_modulation_mae_db": 0.6283674268573043, "mr_stft_db": 5.777675751018631, "band_mae_db": 0.29340767378757515, "rms_gain_error_db": -0.15528778496019363, "output_peak": 0.4417781320422417}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3380982187857233, "全帯域_10ms_rms_envelope_nrmse": 0.10070584562439365, "全帯域_10ms_envelope_correlation": 0.9889052897313079, "全帯域_10ms_modulation_mae_db": 0.30439995902400757, "全帯域_40ms_log_rms_mae_db": 0.7566582471467119, "全帯域_40ms_rms_envelope_nrmse": 0.050879057882202317, "全帯域_40ms_envelope_correlation": 0.9971428584960057, "全帯域_40ms_modulation_mae_db": 0.5035233188879441, "300–1000 Hz_10ms_log_rms_mae_db": 1.5973547416833942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13304241438343728, "300–1000 Hz_10ms_envelope_correlation": 0.9853805889665219, "300–1000 Hz_10ms_modulation_mae_db": 0.4492211260352376, "300–1000 Hz_40ms_log_rms_mae_db": 1.0967719168594592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07330122818460069, "300–1000 Hz_40ms_envelope_correlation": 0.9953043160744354, "300–1000 Hz_40ms_modulation_mae_db": 0.3864086154190589, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5262840275838634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14765371244749684, "1000–3000 Hz_10ms_envelope_correlation": 0.9837888922207181, "1000–3000 Hz_10ms_modulation_mae_db": 0.7366405995076196, "1000–3000 Hz_40ms_log_rms_mae_db": 0.850127245292307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07924822623563876, "1000–3000 Hz_40ms_envelope_correlation": 0.9960004086264009, "1000–3000 Hz_40ms_modulation_mae_db": 0.9157916677338064, "3000–6000 Hz_10ms_log_rms_mae_db": 1.862513309934812, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21103911967373892, "3000–6000 Hz_10ms_envelope_correlation": 0.9685941898206878, "3000–6000 Hz_10ms_modulation_mae_db": 0.6972389421009778, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0919074252041023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10807302340345323, "3000–6000 Hz_40ms_envelope_correlation": 0.9915990199487618, "3000–6000 Hz_40ms_modulation_mae_db": 0.31407418250025143, "mr_stft_db": 5.780184951229562, "band_mae_db": 0.17202947768886853, "rms_gain_error_db": -0.1488015969430136, "output_peak": 0.4750389921963021}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2376159905608874, "全帯域_10ms_rms_envelope_nrmse": 0.09219256640874972, "全帯域_10ms_envelope_correlation": 0.9906790664301091, "全帯域_10ms_modulation_mae_db": 0.4826302007003311, "全帯域_40ms_log_rms_mae_db": 0.6917953764834958, "全帯域_40ms_rms_envelope_nrmse": 0.05250796223342249, "全帯域_40ms_envelope_correlation": 0.9968328268228128, "全帯域_40ms_modulation_mae_db": 0.602349959850044, "300–1000 Hz_10ms_log_rms_mae_db": 1.61915622497834, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1271053008863116, "300–1000 Hz_10ms_envelope_correlation": 0.9868160858338711, "300–1000 Hz_10ms_modulation_mae_db": 0.3068752415644556, "300–1000 Hz_40ms_log_rms_mae_db": 1.0559909382281616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07189859215675076, "300–1000 Hz_40ms_envelope_correlation": 0.9957778432201502, "300–1000 Hz_40ms_modulation_mae_db": 0.4058247932501751, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4991764157578331, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1427720034705334, "1000–3000 Hz_10ms_envelope_correlation": 0.9843402340648676, "1000–3000 Hz_10ms_modulation_mae_db": 0.34039885926097124, "1000–3000 Hz_40ms_log_rms_mae_db": 0.882189555882346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0697867357512427, "1000–3000 Hz_40ms_envelope_correlation": 0.9963322967209832, "1000–3000 Hz_40ms_modulation_mae_db": 0.2017023177756394, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7976644402396633, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1968163134207714, "3000–6000 Hz_10ms_envelope_correlation": 0.9743015617395845, "3000–6000 Hz_10ms_modulation_mae_db": 0.3720487095987614, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0592491199322336, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09316222018810087, "3000–6000 Hz_40ms_envelope_correlation": 0.9943122017526813, "3000–6000 Hz_40ms_modulation_mae_db": 0.37351629832008376, "mr_stft_db": 5.764414993962419, "band_mae_db": 0.23779993772006958, "rms_gain_error_db": -0.12566318648430325, "output_peak": 0.4773873202502974}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3061382468212066, "全帯域_10ms_rms_envelope_nrmse": 0.09335330354896723, "全帯域_10ms_envelope_correlation": 0.9904856268056453, "全帯域_10ms_modulation_mae_db": 0.3766184629128991, "全帯域_40ms_log_rms_mae_db": 0.7151739639701321, "全帯域_40ms_rms_envelope_nrmse": 0.049261927286383095, "全帯域_40ms_envelope_correlation": 0.9973009172594324, "全帯域_40ms_modulation_mae_db": 0.5551766894307568, "300–1000 Hz_10ms_log_rms_mae_db": 1.6015814884644657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14241178279987643, "300–1000 Hz_10ms_envelope_correlation": 0.9833740489703322, "300–1000 Hz_10ms_modulation_mae_db": 0.3643664495817786, "300–1000 Hz_40ms_log_rms_mae_db": 1.0669806002759687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08109394034706105, "300–1000 Hz_40ms_envelope_correlation": 0.9946331074686167, "300–1000 Hz_40ms_modulation_mae_db": 0.4568478542961317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.529828034534112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13862224949434623, "1000–3000 Hz_10ms_envelope_correlation": 0.9854224853686976, "1000–3000 Hz_10ms_modulation_mae_db": 0.42947203148033153, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8885410803316363, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07949851586049794, "1000–3000 Hz_40ms_envelope_correlation": 0.9953592253204306, "1000–3000 Hz_40ms_modulation_mae_db": 0.3971671009519702, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7998782778672888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17883979077583032, "3000–6000 Hz_10ms_envelope_correlation": 0.9775245456713024, "3000–6000 Hz_10ms_modulation_mae_db": 0.40908560699697166, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9944243987178474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08990386366466573, "3000–6000 Hz_40ms_envelope_correlation": 0.994143659141141, "3000–6000 Hz_40ms_modulation_mae_db": 0.4605358753078493, "mr_stft_db": 5.762218203026078, "band_mae_db": 0.253717171586418, "rms_gain_error_db": -0.14025025888944487, "output_peak": 0.45181580280320505}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.234896789593921, "全帯域_10ms_rms_envelope_nrmse": 0.09953784753932808, "全帯域_10ms_envelope_correlation": 0.9890418021416704, "全帯域_10ms_modulation_mae_db": 0.08034098026362732, "全帯域_40ms_log_rms_mae_db": 0.6261666425434372, "全帯域_40ms_rms_envelope_nrmse": 0.04994348936905008, "全帯域_40ms_envelope_correlation": 0.9970323008062485, "全帯域_40ms_modulation_mae_db": 0.5331784291817298, "300–1000 Hz_10ms_log_rms_mae_db": 1.673388900818447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1487934263869097, "300–1000 Hz_10ms_envelope_correlation": 0.9816946822718626, "300–1000 Hz_10ms_modulation_mae_db": 0.2993329916433425, "300–1000 Hz_40ms_log_rms_mae_db": 1.056311984169734, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08261091371248669, "300–1000 Hz_40ms_envelope_correlation": 0.9940591876110245, "300–1000 Hz_40ms_modulation_mae_db": 0.39262689600871914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5608039270613943, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15102179652511502, "1000–3000 Hz_10ms_envelope_correlation": 0.9823465101916461, "1000–3000 Hz_10ms_modulation_mae_db": 0.28158580753126555, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9362228184077688, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07993669829584958, "1000–3000 Hz_40ms_envelope_correlation": 0.9948944622284757, "1000–3000 Hz_40ms_modulation_mae_db": 0.3455977158827166, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8321164110126769, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1923670720731857, "3000–6000 Hz_10ms_envelope_correlation": 0.9740631679033374, "3000–6000 Hz_10ms_modulation_mae_db": 0.7433400795638477, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0843542629662508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09453602133507832, "3000–6000 Hz_40ms_envelope_correlation": 0.9933649516175505, "3000–6000 Hz_40ms_modulation_mae_db": 0.3256424202398038, "mr_stft_db": 5.764395908811924, "band_mae_db": 0.21476688873934277, "rms_gain_error_db": -0.07985545810073405, "output_peak": 0.4466828744902491}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.176099142060794, "全帯域_10ms_rms_envelope_nrmse": 0.09143756475184608, "全帯域_10ms_envelope_correlation": 0.9908486556869585, "全帯域_10ms_modulation_mae_db": 0.15804402797244732, "全帯域_40ms_log_rms_mae_db": 0.5996488579114003, "全帯域_40ms_rms_envelope_nrmse": 0.04674900460530332, "全帯域_40ms_envelope_correlation": 0.9975358064137171, "全帯域_40ms_modulation_mae_db": 0.4415997782307073, "300–1000 Hz_10ms_log_rms_mae_db": 1.5841266425408276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13599629584610876, "300–1000 Hz_10ms_envelope_correlation": 0.9848552274396586, "300–1000 Hz_10ms_modulation_mae_db": 0.44643596462530627, "300–1000 Hz_40ms_log_rms_mae_db": 1.07955258120655, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08808455384776323, "300–1000 Hz_40ms_envelope_correlation": 0.9935049693468804, "300–1000 Hz_40ms_modulation_mae_db": 0.35833240024999274, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6125915096506829, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15235373152343468, "1000–3000 Hz_10ms_envelope_correlation": 0.9820267899337465, "1000–3000 Hz_10ms_modulation_mae_db": 0.31204049971270903, "1000–3000 Hz_40ms_log_rms_mae_db": 0.94750687465869, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08862175217081597, "1000–3000 Hz_40ms_envelope_correlation": 0.9936755643282853, "1000–3000 Hz_40ms_modulation_mae_db": 0.44335608496145795, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8273622527153388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19415046060747876, "3000–6000 Hz_10ms_envelope_correlation": 0.9736430780809603, "3000–6000 Hz_10ms_modulation_mae_db": 0.7740708255115646, "3000–6000 Hz_40ms_log_rms_mae_db": 1.046759769036195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09861726570675995, "3000–6000 Hz_40ms_envelope_correlation": 0.9934318364728465, "3000–6000 Hz_40ms_modulation_mae_db": 1.168081674326175, "mr_stft_db": 5.78813748104137, "band_mae_db": 0.1616006253863033, "rms_gain_error_db": -0.12347267606417468, "output_peak": 0.4520298688485294}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2640920126929982, "全帯域_10ms_rms_envelope_nrmse": 0.0919405154451718, "全帯域_10ms_envelope_correlation": 0.9907724678832002, "全帯域_10ms_modulation_mae_db": 0.09457609411876618, "全帯域_40ms_log_rms_mae_db": 0.7553372982450447, "全帯域_40ms_rms_envelope_nrmse": 0.049429430425444, "全帯域_40ms_envelope_correlation": 0.997251860317252, "全帯域_40ms_modulation_mae_db": 0.11998869351997839, "300–1000 Hz_10ms_log_rms_mae_db": 1.6672787618807219, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13802360916364403, "300–1000 Hz_10ms_envelope_correlation": 0.9845190153765915, "300–1000 Hz_10ms_modulation_mae_db": 0.2426062210298269, "300–1000 Hz_40ms_log_rms_mae_db": 1.0947788175574078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08332295238479251, "300–1000 Hz_40ms_envelope_correlation": 0.9944284038546731, "300–1000 Hz_40ms_modulation_mae_db": 0.159675670925026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5567098487815343, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1454627624225485, "1000–3000 Hz_10ms_envelope_correlation": 0.9840020934838626, "1000–3000 Hz_10ms_modulation_mae_db": 0.5573532210030337, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9056123804018364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0761995053948375, "1000–3000 Hz_40ms_envelope_correlation": 0.9959809104742487, "1000–3000 Hz_40ms_modulation_mae_db": 0.7203623943336138, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8338614328786647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1950312116884566, "3000–6000 Hz_10ms_envelope_correlation": 0.9732230315654701, "3000–6000 Hz_10ms_modulation_mae_db": 0.7765662738900249, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1103025716267663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1051222120548451, "3000–6000 Hz_40ms_envelope_correlation": 0.9919860516184267, "3000–6000 Hz_40ms_modulation_mae_db": 0.5675193067200688, "mr_stft_db": 5.775500699075419, "band_mae_db": 0.27596180434458795, "rms_gain_error_db": -0.13247110633384876, "output_peak": 0.4445429484464023}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.268406428781566, "全帯域_10ms_rms_envelope_nrmse": 0.09751837645764663, "全帯域_10ms_envelope_correlation": 0.989675543632366, "全帯域_10ms_modulation_mae_db": 0.16628231730681392, "全帯域_40ms_log_rms_mae_db": 0.6779391974776844, "全帯域_40ms_rms_envelope_nrmse": 0.05014844695370034, "全帯域_40ms_envelope_correlation": 0.9973601513055433, "全帯域_40ms_modulation_mae_db": 0.40893304042543743, "300–1000 Hz_10ms_log_rms_mae_db": 1.6428994687146887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14086826692167223, "300–1000 Hz_10ms_envelope_correlation": 0.9838114548226172, "300–1000 Hz_10ms_modulation_mae_db": 0.3770374263753296, "300–1000 Hz_40ms_log_rms_mae_db": 1.0663285960629654, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07811482693597463, "300–1000 Hz_40ms_envelope_correlation": 0.9951602214490692, "300–1000 Hz_40ms_modulation_mae_db": 0.48333361434442895, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4942637491418556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15546383954281218, "1000–3000 Hz_10ms_envelope_correlation": 0.9820214926852123, "1000–3000 Hz_10ms_modulation_mae_db": 0.35066361014773034, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8544729017689614, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09233714163837022, "1000–3000 Hz_40ms_envelope_correlation": 0.9942575916327734, "1000–3000 Hz_40ms_modulation_mae_db": 0.38050891307841633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8608785153689176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19421676348459938, "3000–6000 Hz_10ms_envelope_correlation": 0.9734589802226893, "3000–6000 Hz_10ms_modulation_mae_db": 0.6114056578684722, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1017824567140062, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10470333729909065, "3000–6000 Hz_40ms_envelope_correlation": 0.9920222601460323, "3000–6000 Hz_40ms_modulation_mae_db": 0.3480875975342248, "mr_stft_db": 5.783551424816864, "band_mae_db": 0.24952332616479211, "rms_gain_error_db": -0.17006370165665985, "output_peak": 0.4416561014456898}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.982918525017961, "spectral_convergence": 0.3633001323198922, "spectral_convergence_status": "ok", "amplitude_floor": 1.016885378852613e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.516073380782424, "spectral_convergence": 0.3466246460461061, "spectral_convergence_status": "ok", "amplitude_floor": 6.009137305261044e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.834035347255508, "spectral_convergence": 0.3817182617266673, "spectral_convergence_status": "ok", "amplitude_floor": 5.349766097366943e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003378185892142299, "output_power": 0.0003504825745219664, "delta_db": 0.15982886450746123, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005186258824964237, "output_power": 0.0005050597436658888, "delta_db": -0.1151143285576541, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004467018619566432, "output_power": 0.0004207683414014543, "delta_db": -0.2597470664551126, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002054215054829419, "output_power": 0.00018457828830834234, "delta_db": -0.46465293502758, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00019049740954433386, "output_power": 0.0001733107183727815, "delta_db": -0.41063651936462775, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.3523016958893975e-05, "output_power": 4.7180835286903344e-05, "delta_db": 0.3504663288130152, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29340767378757515, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15528778496019363, "pitch": null, "output_rms_dbfs": -27.743339932666345, "output_peak": 0.4417781320422417, "reference_rms_dbfs": -27.58805214770615, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.777675751018631}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001220285373493014, "log_rms_mae_db": 1.3368349920904776, "rms_envelope_nrmse": 0.09645992858033271, "envelope_correlation": 0.9898604958130571, "rms_gain_error_db": -0.1552053357753301, "reference_mean_power": 0.0017439355594579833, "output_mean_power": 0.001682712417531892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.14297331662316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004136999026359357, "output_power": 0.00040177697101146975, "reference_active": true, "delta_db": -0.12700378380979852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.498679838909322e-05, "output_power": 7.418152993752812e-05, "reference_active": true, "delta_db": -0.5905568278893244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.720897303077822e-05, "output_power": 1.6995994848242813e-05, "reference_active": true, "delta_db": -0.05408363202208354}], "active_band_count": 3, "active_band_mae_db": 0.25721474790706883}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011225274665232883, "log_rms_mae_db": 0.7151288870307932, "rms_envelope_nrmse": 0.05287913137691423, "envelope_correlation": 0.9969127147315747, "rms_gain_error_db": -0.1556681508488518, "reference_mean_power": 0.0017512823148994152, "output_mean_power": 0.0016896211878615567}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.129560410778096e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037542654743209525, "output_power": 0.00036496644760462116, "reference_active": true, "delta_db": -0.12272039140589593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.98989275491616e-05, "output_power": 3.534542816605722e-05, "reference_active": true, "delta_db": -0.5262797539131874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1683754663794263e-06, "output_power": 8.977004781379841e-07, "reference_active": true, "delta_db": -1.1445097255502965}], "active_band_count": 3, "active_band_mae_db": 0.5978366236231266}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.832439224759049e-05, "log_rms_mae_db": 1.535438454309173, "rms_envelope_nrmse": 0.13610314672615778, "envelope_correlation": 0.9847000022697425, "rms_gain_error_db": -0.07274597946832478, "reference_mean_power": 0.0005656706566344367, "output_mean_power": 0.0005562743730938219}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3997327154946766e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018576385507894965, "output_power": 0.00018048968828050544, "reference_active": true, "delta_db": -0.1250882033836296}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.516179020835253e-05, "output_power": 3.2534261645218286e-05, "reference_active": true, "delta_db": -0.337300233930353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.50699704718427e-06, "output_power": 8.890240461059145e-06, "reference_active": true, "delta_db": 0.7344726278022373}], "active_band_count": 3, "active_band_mae_db": 0.3989536883720733}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.746258281389908e-05, "log_rms_mae_db": 0.9556694919099502, "rms_envelope_nrmse": 0.07697935707934347, "envelope_correlation": 0.9948471720556119, "rms_gain_error_db": -0.07377799418145159, "reference_mean_power": 0.0005675319915431874, "output_mean_power": 0.0005579721828250547}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.076532458167812e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017606095149491155, "output_power": 0.00017273306517473055, "reference_active": true, "delta_db": -0.08287564893437711}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7577844172032785e-05, "output_power": 1.6236454090984424e-05, "reference_active": true, "delta_db": -0.34474421232015107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.080515782562606e-07, "output_power": 5.550426951644897e-07, "reference_active": true, "delta_db": 0.3841858650787974}], "active_band_count": 3, "active_band_mae_db": 0.2706019087777752}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.679296131568703e-05, "log_rms_mae_db": 1.5333571272381616, "rms_envelope_nrmse": 0.142013372523658, "envelope_correlation": 0.9853137044940611, "rms_gain_error_db": -0.3942301073921477, "reference_mean_power": 0.00025821561430829635, "output_mean_power": 0.00023580851850613983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9030353181622983e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.270493557085603e-05, "output_power": 8.387925757530443e-05, "reference_active": true, "delta_db": -0.43448278589987455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6852701179097796e-05, "output_power": 1.3581145262465909e-05, "reference_active": true, "delta_db": -0.9373312582484484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.066518069212071e-06, "output_power": 3.6981152281344622e-06, "reference_active": true, "delta_db": -0.4124226734256006}], "active_band_count": 3, "active_band_mae_db": 0.5947455725246412}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.121002038281381e-05, "log_rms_mae_db": 0.9095325041136335, "rms_envelope_nrmse": 0.09120171069232914, "envelope_correlation": 0.9945148561063877, "rms_gain_error_db": -0.39468705626161565, "reference_mean_power": 0.0002594441947231775, "output_mean_power": 0.00023690555912838745}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7574651639406604e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.903046567125758e-05, "output_power": 8.008260788897697e-05, "reference_active": true, "delta_db": -0.4600043745160196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.408774632836393e-06, "output_power": 6.657276732804538e-06, "reference_active": true, "delta_db": -1.0143610264005527}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6645360300144737e-07, "output_power": 2.2639928802807824e-07, "reference_active": true, "delta_db": -0.7074654040690374}], "active_band_count": 3, "active_band_mae_db": 0.7272769349952032}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.318455452770038e-05, "log_rms_mae_db": 1.8148549306539066, "rms_envelope_nrmse": 0.18689685644389614, "envelope_correlation": 0.9755646154108505, "rms_gain_error_db": -0.28274134837465603, "reference_mean_power": 0.0001492769552865986, "output_mean_power": 0.00013986808655171948}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0711725750134131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.027413944353342e-05, "output_power": 5.963533595870418e-05, "reference_active": true, "delta_db": -0.04627348291938775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2776930907508557e-05, "output_power": 1.0662014820648397e-05, "reference_active": true, "delta_db": -0.7858726443335029}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.16053005381717e-06, "output_power": 2.0326142319433045e-06, "reference_active": true, "delta_db": -0.2650534980198894}], "active_band_count": 3, "active_band_mae_db": 0.36573320842426}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.04724555086843e-05, "log_rms_mae_db": 1.0657585336238982, "rms_envelope_nrmse": 0.10116978791527247, "envelope_correlation": 0.9928836745467785, "rms_gain_error_db": -0.2829380192901695, "reference_mean_power": 0.00014998968981604926, "output_mean_power": 0.00014052953365417816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.934560164313186e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.89522015257842e-05, "output_power": 5.793448135605076e-05, "reference_active": true, "delta_db": -0.07562904730397715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6946349724579084e-06, "output_power": 4.903747357805555e-06, "reference_active": true, "delta_db": -0.6493780378628287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8905823102195032e-07, "output_power": 1.4473916752655867e-07, "reference_active": true, "delta_db": -1.160095195405107}], "active_band_count": 3, "active_band_mae_db": 0.6283674268573043}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8323886632994423, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "807cb3508dc3507e7677bad500d02ef50cfe97624cc3fe600686dd0152f54013", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9836577753053515, "standard_deviation": 0.001389488600244397, "minimum": 0.9820214926852123, "maximum": 0.9854224853686976, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5391268299686796, "standard_deviation": 0.03792764635430353, "minimum": 1.4942637491418556, "maximum": 1.6125915096506829, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4503625251460378, "standard_deviation": 0.16227642490145627, "minimum": 0.28158580753126555, "maximum": 0.7366405995076196, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14692043349374312, "standard_deviation": 0.005758754488343153, "minimum": 0.13862224949434623, "maximum": 0.15546383954281218, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951269144297482, "standard_deviation": 0.0009479304399577219, "minimum": 0.9936755643282853, "maximum": 0.9963322967209832, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8967756701071474, "standard_deviation": 0.03508503920218688, "minimum": 0.850127245292307, "maximum": 0.94750687465869, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.516470391214103, "standard_deviation": 0.24254888795352147, "minimum": 0.2017023177756394, "maximum": 0.9157916677338064, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08210378575494773, "standard_deviation": 0.007893811896602182, "minimum": 0.0697867357512427, "maximum": 0.09233714163837022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9737966463018602, "standard_deviation": 0.0025321112231870197, "minimum": 0.9685941898206878, "maximum": 0.9775245456713024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8286411963339084, "standard_deviation": 0.02450788362962604, "minimum": 1.7976644402396633, "maximum": 1.862513309934812, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.59368616299436, "standard_deviation": 0.18306138324773175, "minimum": 0.36573320842426, "maximum": 0.7765662738900249, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19366969852099467, "standard_deviation": 0.009114168687631412, "minimum": 0.17883979077583032, "maximum": 0.21103911967373892, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9929679569055273, "standard_deviation": 0.0010213178651434299, "minimum": 0.9915990199487618, "maximum": 0.9943122017526813, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0693173172276627, "standard_deviation": 0.037229068589718514, "minimum": 0.9944243987178474, "maximum": 1.1103025716267663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5232280977257202, "standard_deviation": 0.285020229262221, "minimum": 0.31407418250025143, "maximum": 1.168081674326175, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09941096644590829, "standard_deviation": 0.006465060856354443, "minimum": 0.08990386366466573, "maximum": 0.10807302340345323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9843938882438997, "standard_deviation": 0.0015037016062371647, "minimum": 0.9816946822718626, "maximum": 0.9868160858338711, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6151530854237572, "standard_deviation": 0.04584510428517313, "minimum": 1.535438454309173, "maximum": 1.673388900818447, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3606036386534188, "standard_deviation": 0.07323097031664841, "minimum": 0.2426062210298269, "maximum": 0.4492211260352376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13779303038926474, "standard_deviation": 0.006488800333910588, "minimum": 0.1271053008863116, "maximum": 0.1487934263869097, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947144026350576, "standard_deviation": 0.0007246403729416923, "minimum": 0.9935049693468804, "maximum": 0.9957778432201502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0590481157837746, "standard_deviation": 0.04462559106816892, "minimum": 0.9556694919099502, "maximum": 1.0967719168594592, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3642064691589134, "standard_deviation": 0.10449580282557129, "minimum": 0.159675670925026, "maximum": 0.48333361434442895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07942579558109664, "standard_deviation": 0.005411131749344426, "minimum": 0.07189859215675076, "maximum": 0.08808455384776323, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23235086317724468, "standard_deviation": 0.046849935503263905, "minimum": 0.1616006253863033, "maximum": 0.29340767378757515, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.774509926622784, "standard_deviation": 0.009753800710817202, "minimum": 5.762218203026078, "maximum": 5.78813748104137, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4538665050653646, "standard_deviation": 0.014356867673709124, "minimum": 0.4416561014456898, "maximum": 0.4773873202502974, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13448322117904657, "standard_deviation": 0.027056399545200192, "minimum": -0.17006370165665985, "maximum": -0.07985545810073405, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900336185155394, "standard_deviation": 0.0007788373331455449, "minimum": 0.9889052897313079, "maximum": 0.9908486556869585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2702727276734467, "standard_deviation": 0.05547899958271835, "minimum": 1.176099142060794, "maximum": 1.3380982187857233, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24001334877574515, "standard_deviation": 0.14180868028056728, "minimum": 0.08034098026362732, "maximum": 0.4826302007003311, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09539324354455449, "standard_deviation": 0.003645543372171308, "minimum": 0.09143756475184608, "maximum": 0.10070584562439365, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971711795190733, "standard_deviation": 0.00023705947159622433, "minimum": 0.9968328268228128, "maximum": 0.9975358064137171, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6922310588510874, "standard_deviation": 0.056472958922612095, "minimum": 0.5996488579114003, "maximum": 0.7566582471467119, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4703233166437156, "standard_deviation": 0.15733128560009357, "minimum": 0.11998869351997839, "maximum": 0.602349959850044, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.050224806266552485, "standard_deviation": 0.0019424658087090006, "minimum": 0.04674900460530332, "maximum": 0.05287913137691423, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0029", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.9462153911590576, "predicted_mos": 1.9462153911590576, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7bbf71d41175829607ef503bcfade470f0d6ecbf97c0ddc716d91a43f2661747", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.832720100414008, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3432342718523933, "全帯域_10ms_rms_envelope_nrmse": 0.09686744997274395, "全帯域_10ms_envelope_correlation": 0.9897577335792188, "全帯域_10ms_modulation_mae_db": 0.2369636766812749, "全帯域_40ms_log_rms_mae_db": 0.73656199214943, "全帯域_40ms_rms_envelope_nrmse": 0.05422009389830893, "全帯域_40ms_envelope_correlation": 0.9967115211203353, "全帯域_40ms_modulation_mae_db": 0.5750572958041502, "300–1000 Hz_10ms_log_rms_mae_db": 1.5200289372174416, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13789691649184316, "300–1000 Hz_10ms_envelope_correlation": 0.984309162763195, "300–1000 Hz_10ms_modulation_mae_db": 0.34163896363949203, "300–1000 Hz_40ms_log_rms_mae_db": 0.876012559163852, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07736254630573126, "300–1000 Hz_40ms_envelope_correlation": 0.9947613914994323, "300–1000 Hz_40ms_modulation_mae_db": 0.1971627691034695, "1000–3000 Hz_10ms_log_rms_mae_db": 1.552103013087923, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1493463487676144, "1000–3000 Hz_10ms_envelope_correlation": 0.9839731084059625, "1000–3000 Hz_10ms_modulation_mae_db": 0.5215826440699942, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9093622328965767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09434535559115828, "1000–3000 Hz_40ms_envelope_correlation": 0.994498130775114, "1000–3000 Hz_40ms_modulation_mae_db": 0.7215211645983975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7703564069927258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1870676581373201, "3000–6000 Hz_10ms_envelope_correlation": 0.9754577805655407, "3000–6000 Hz_10ms_modulation_mae_db": 0.3698168368117141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0245526985165951, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09920559099854889, "3000–6000 Hz_40ms_envelope_correlation": 0.9929881104189024, "3000–6000 Hz_40ms_modulation_mae_db": 0.7180220316260969, "mr_stft_db": 5.737658603692605, "band_mae_db": 0.31114615694402253, "rms_gain_error_db": -0.14900788776096868, "output_peak": 0.4423006137314879}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3407616361017338, "全帯域_10ms_rms_envelope_nrmse": 0.09991664790760534, "全帯域_10ms_envelope_correlation": 0.9890850293331569, "全帯域_10ms_modulation_mae_db": 0.3284180541007858, "全帯域_40ms_log_rms_mae_db": 0.747927017053004, "全帯域_40ms_rms_envelope_nrmse": 0.050125021575988554, "全帯域_40ms_envelope_correlation": 0.9972347501725872, "全帯域_40ms_modulation_mae_db": 0.5650052104855036, "300–1000 Hz_10ms_log_rms_mae_db": 1.5618501685186532, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1325315770276076, "300–1000 Hz_10ms_envelope_correlation": 0.9855093860955529, "300–1000 Hz_10ms_modulation_mae_db": 0.4455574975683398, "300–1000 Hz_40ms_log_rms_mae_db": 0.9951882582529038, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07211755970260661, "300–1000 Hz_40ms_envelope_correlation": 0.9954194382227466, "300–1000 Hz_40ms_modulation_mae_db": 0.3523038471090374, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5117373333623838, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14666015810401972, "1000–3000 Hz_10ms_envelope_correlation": 0.9841301574797857, "1000–3000 Hz_10ms_modulation_mae_db": 0.8707746788447236, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8387355291948932, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07984788501760019, "1000–3000 Hz_40ms_envelope_correlation": 0.9960575196027714, "1000–3000 Hz_40ms_modulation_mae_db": 1.1131598405406926, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8168262003653204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20883984589502833, "3000–6000 Hz_10ms_envelope_correlation": 0.969339173482992, "3000–6000 Hz_10ms_modulation_mae_db": 0.6473637974838334, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0514806632832365, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10865314779057354, "3000–6000 Hz_40ms_envelope_correlation": 0.991647660448744, "3000–6000 Hz_40ms_modulation_mae_db": 0.34264941054691195, "mr_stft_db": 5.739804161931346, "band_mae_db": 0.19466134459873896, "rms_gain_error_db": -0.14750856463266204, "output_peak": 0.4755206628643635}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2109582271200237, "全帯域_10ms_rms_envelope_nrmse": 0.09062961258304363, "全帯域_10ms_envelope_correlation": 0.9910119190657416, "全帯域_10ms_modulation_mae_db": 0.4898758682703325, "全帯域_40ms_log_rms_mae_db": 0.6835876189434346, "全帯域_40ms_rms_envelope_nrmse": 0.05319172580784165, "全帯域_40ms_envelope_correlation": 0.9967604631243022, "全帯域_40ms_modulation_mae_db": 0.6183965867355002, "300–1000 Hz_10ms_log_rms_mae_db": 1.5781490760381884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12352375678402477, "300–1000 Hz_10ms_envelope_correlation": 0.9875232794411718, "300–1000 Hz_10ms_modulation_mae_db": 0.4309314881829715, "300–1000 Hz_40ms_log_rms_mae_db": 0.978900311943929, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07167197641035729, "300–1000 Hz_40ms_envelope_correlation": 0.9957263755414059, "300–1000 Hz_40ms_modulation_mae_db": 0.49135070509084083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4807426139295667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1463743921833571, "1000–3000 Hz_10ms_envelope_correlation": 0.9836513956627725, "1000–3000 Hz_10ms_modulation_mae_db": 0.20900623695239254, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8329065866569727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07170742669516447, "1000–3000 Hz_40ms_envelope_correlation": 0.9963161271462391, "1000–3000 Hz_40ms_modulation_mae_db": 0.34694644204691877, "3000–6000 Hz_10ms_log_rms_mae_db": 1.762767055802301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19924289417386762, "3000–6000 Hz_10ms_envelope_correlation": 0.973717415531552, "3000–6000 Hz_10ms_modulation_mae_db": 0.3310404816485121, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0223267338014026, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0942947702202787, "3000–6000 Hz_40ms_envelope_correlation": 0.9942250780560304, "3000–6000 Hz_40ms_modulation_mae_db": 0.31053859253223776, "mr_stft_db": 5.726147269810163, "band_mae_db": 0.26922753996312654, "rms_gain_error_db": -0.13038741305960724, "output_peak": 0.49069180171090654}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3263203786068396, "全帯域_10ms_rms_envelope_nrmse": 0.09406330535890553, "全帯域_10ms_envelope_correlation": 0.9903214281890527, "全帯域_10ms_modulation_mae_db": 0.26388780983957977, "全帯域_40ms_log_rms_mae_db": 0.7287893504831426, "全帯域_40ms_rms_envelope_nrmse": 0.0495328519311599, "全帯域_40ms_envelope_correlation": 0.9972380962277289, "全帯域_40ms_modulation_mae_db": 0.4037967559260726, "300–1000 Hz_10ms_log_rms_mae_db": 1.5976735420329171, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14106700439279127, "300–1000 Hz_10ms_envelope_correlation": 0.9836029114292724, "300–1000 Hz_10ms_modulation_mae_db": 0.3611796957542693, "300–1000 Hz_40ms_log_rms_mae_db": 1.0193346933115606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07883980900692883, "300–1000 Hz_40ms_envelope_correlation": 0.9947810089791761, "300–1000 Hz_40ms_modulation_mae_db": 0.354187874732498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5647625734460626, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14847731747412027, "1000–3000 Hz_10ms_envelope_correlation": 0.9833165476725256, "1000–3000 Hz_10ms_modulation_mae_db": 0.40785636533553715, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8817180479549752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08735430442595228, "1000–3000 Hz_40ms_envelope_correlation": 0.9944643129668839, "1000–3000 Hz_40ms_modulation_mae_db": 0.3753596854482091, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7483937016193893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17856061828432046, "3000–6000 Hz_10ms_envelope_correlation": 0.9776116734655942, "3000–6000 Hz_10ms_modulation_mae_db": 0.373275256889501, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9772189914718462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08946237333306448, "3000–6000 Hz_40ms_envelope_correlation": 0.9941852176178276, "3000–6000 Hz_40ms_modulation_mae_db": 0.46646241013182294, "mr_stft_db": 5.729415054095058, "band_mae_db": 0.25205032197978644, "rms_gain_error_db": -0.132648824907507, "output_peak": 0.45213650546167256}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.235958210891517, "全帯域_10ms_rms_envelope_nrmse": 0.09790390183158364, "全帯域_10ms_envelope_correlation": 0.9893904984394478, "全帯域_10ms_modulation_mae_db": 0.1372214947473449, "全帯域_40ms_log_rms_mae_db": 0.6400424488282646, "全帯域_40ms_rms_envelope_nrmse": 0.04899193256972179, "全帯域_40ms_envelope_correlation": 0.9971369764819704, "全帯域_40ms_modulation_mae_db": 0.5960580219158965, "300–1000 Hz_10ms_log_rms_mae_db": 1.6467428399422874, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1477556832262063, "300–1000 Hz_10ms_envelope_correlation": 0.981978630771942, "300–1000 Hz_10ms_modulation_mae_db": 0.37448695408940075, "300–1000 Hz_40ms_log_rms_mae_db": 0.9925411380622033, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0805276388948227, "300–1000 Hz_40ms_envelope_correlation": 0.994326316367554, "300–1000 Hz_40ms_modulation_mae_db": 0.33630133252588923, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5908783796409032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15530324948697297, "1000–3000 Hz_10ms_envelope_correlation": 0.9813748658145375, "1000–3000 Hz_10ms_modulation_mae_db": 0.2720346091091142, "1000–3000 Hz_40ms_log_rms_mae_db": 0.910117317228362, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08525681995911093, "1000–3000 Hz_40ms_envelope_correlation": 0.9942693242273347, "1000–3000 Hz_40ms_modulation_mae_db": 0.36710868579626227, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7831894725086348, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18682215507205563, "3000–6000 Hz_10ms_envelope_correlation": 0.975548124575037, "3000–6000 Hz_10ms_modulation_mae_db": 0.7216428265612489, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0604989310313548, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09467764390237504, "3000–6000 Hz_40ms_envelope_correlation": 0.9934252538676432, "3000–6000 Hz_40ms_modulation_mae_db": 0.2664325060416491, "mr_stft_db": 5.724359442693498, "band_mae_db": 0.2334328584664113, "rms_gain_error_db": -0.07675610770599907, "output_peak": 0.44604158251413023}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1784789215178708, "全帯域_10ms_rms_envelope_nrmse": 0.09268045126359441, "全帯域_10ms_envelope_correlation": 0.990589245008347, "全帯域_10ms_modulation_mae_db": 0.21384842441526228, "全帯域_40ms_log_rms_mae_db": 0.6062304541888681, "全帯域_40ms_rms_envelope_nrmse": 0.047976463938294626, "全帯域_40ms_envelope_correlation": 0.9973959491341676, "全帯域_40ms_modulation_mae_db": 0.3734394457807529, "300–1000 Hz_10ms_log_rms_mae_db": 1.591938603922735, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1395480625124507, "300–1000 Hz_10ms_envelope_correlation": 0.9840096857970029, "300–1000 Hz_10ms_modulation_mae_db": 0.4533016711361781, "300–1000 Hz_40ms_log_rms_mae_db": 1.0348122699903286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09037807243264977, "300–1000 Hz_40ms_envelope_correlation": 0.9931007757443872, "300–1000 Hz_40ms_modulation_mae_db": 0.3584398635515164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5817272075590565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15125581004282607, "1000–3000 Hz_10ms_envelope_correlation": 0.9823423709698712, "1000–3000 Hz_10ms_modulation_mae_db": 0.4136705898788436, "1000–3000 Hz_40ms_log_rms_mae_db": 0.897566195719245, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08462626875523586, "1000–3000 Hz_40ms_envelope_correlation": 0.9943551855865241, "1000–3000 Hz_40ms_modulation_mae_db": 0.5843186552525034, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7711110346809122, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19707066480203062, "3000–6000 Hz_10ms_envelope_correlation": 0.9728757811591148, "3000–6000 Hz_10ms_modulation_mae_db": 0.8128795401071466, "3000–6000 Hz_40ms_log_rms_mae_db": 0.989347234210437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10083133425324622, "3000–6000 Hz_40ms_envelope_correlation": 0.9931834144173649, "3000–6000 Hz_40ms_modulation_mae_db": 1.226460438325625, "mr_stft_db": 5.75026613836426, "band_mae_db": 0.16801256308867954, "rms_gain_error_db": -0.12490075148509311, "output_peak": 0.4479123869711039}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2640868641966645, "全帯域_10ms_rms_envelope_nrmse": 0.09287374054490304, "全帯域_10ms_envelope_correlation": 0.9905450792616305, "全帯域_10ms_modulation_mae_db": 0.11979660838716709, "全帯域_40ms_log_rms_mae_db": 0.7409033843307287, "全帯域_40ms_rms_envelope_nrmse": 0.05024086694964813, "全帯域_40ms_envelope_correlation": 0.9971140480212217, "全帯域_40ms_modulation_mae_db": 0.18184749778829423, "300–1000 Hz_10ms_log_rms_mae_db": 1.6239918603813837, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1373739294872525, "300–1000 Hz_10ms_envelope_correlation": 0.9846162476895052, "300–1000 Hz_10ms_modulation_mae_db": 0.22915791303694458, "300–1000 Hz_40ms_log_rms_mae_db": 1.008305740173988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08144107729337173, "300–1000 Hz_40ms_envelope_correlation": 0.9946340625885938, "300–1000 Hz_40ms_modulation_mae_db": 0.14073687088189563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.550804395478232, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14381946546933982, "1000–3000 Hz_10ms_envelope_correlation": 0.9844766091791085, "1000–3000 Hz_10ms_modulation_mae_db": 0.5365268500242352, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8655374913639566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07452266912124993, "1000–3000 Hz_40ms_envelope_correlation": 0.9963318674012196, "1000–3000 Hz_40ms_modulation_mae_db": 0.699271768525863, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7856245941320492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1948066470418628, "3000–6000 Hz_10ms_envelope_correlation": 0.9733905511654238, "3000–6000 Hz_10ms_modulation_mae_db": 0.7584539197743969, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0733896929434987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10769481236194456, "3000–6000 Hz_40ms_envelope_correlation": 0.9918137867682814, "3000–6000 Hz_40ms_modulation_mae_db": 0.519784487725274, "mr_stft_db": 5.739850023884315, "band_mae_db": 0.29453940825985525, "rms_gain_error_db": -0.12391974889650699, "output_peak": 0.4433158754814195}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2940855424055009, "全帯域_10ms_rms_envelope_nrmse": 0.09786966120694328, "全帯域_10ms_envelope_correlation": 0.989672237968105, "全帯域_10ms_modulation_mae_db": 0.25766094100720377, "全帯域_40ms_log_rms_mae_db": 0.6930539315857888, "全帯域_40ms_rms_envelope_nrmse": 0.050467181473026014, "全帯域_40ms_envelope_correlation": 0.9974197225412001, "全帯域_40ms_modulation_mae_db": 0.3610814504923432, "300–1000 Hz_10ms_log_rms_mae_db": 1.6141569705280543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14326696515485896, "300–1000 Hz_10ms_envelope_correlation": 0.983217656020524, "300–1000 Hz_10ms_modulation_mae_db": 0.5612851790506643, "300–1000 Hz_40ms_log_rms_mae_db": 0.97679824802062, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07663382283335234, "300–1000 Hz_40ms_envelope_correlation": 0.9953349641678606, "300–1000 Hz_40ms_modulation_mae_db": 0.3557459893232863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4883784999093659, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1596935925892531, "1000–3000 Hz_10ms_envelope_correlation": 0.9811076371781691, "1000–3000 Hz_10ms_modulation_mae_db": 0.3977998832178728, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7930833766683417, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08897931824010995, "1000–3000 Hz_40ms_envelope_correlation": 0.995019009902253, "1000–3000 Hz_40ms_modulation_mae_db": 0.4431881710981112, "3000–6000 Hz_10ms_log_rms_mae_db": 1.799257960714231, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19004259925503755, "3000–6000 Hz_10ms_envelope_correlation": 0.9746681099297468, "3000–6000 Hz_10ms_modulation_mae_db": 0.5913991844720167, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0517931054099259, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10411356829440312, "3000–6000 Hz_40ms_envelope_correlation": 0.9922707742978057, "3000–6000 Hz_40ms_modulation_mae_db": 0.4603692123785919, "mr_stft_db": 5.737952529523323, "band_mae_db": 0.2670338517453601, "rms_gain_error_db": -0.1883810800569078, "output_peak": 0.4404781378687279}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.939895745929034, "spectral_convergence": 0.36170538709252903, "spectral_convergence_status": "ok", "amplitude_floor": 1.016885378852613e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.480172392844494, "spectral_convergence": 0.3470831729589168, "spectral_convergence_status": "ok", "amplitude_floor": 6.009137305261044e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.792907672304287, "spectral_convergence": 0.3826638484372865, "spectral_convergence_status": "ok", "amplitude_floor": 5.349766097366943e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003378185892142299, "output_power": 0.00034928036761968766, "delta_db": 0.14490631556950492, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005186258824964237, "output_power": 0.0005068838395683769, "delta_db": -0.09945741534526265, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004467018619566432, "output_power": 0.0004241630192485963, "delta_db": -0.22484960949714664, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002054215054829419, "output_power": 0.0001831204122488162, "delta_db": -0.4990915036028953, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00019049740954433386, "output_power": 0.00017202812919693325, "delta_db": -0.44289607931336966, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.3523016958893975e-05, "output_power": 4.833776814920789e-05, "delta_db": 0.4556760183359563, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.31114615694402253, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14900788776096868, "pitch": null, "output_rms_dbfs": -27.73706003546712, "output_peak": 0.4423006137314879, "reference_rms_dbfs": -27.58805214770615, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.737658603692605}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001220285373493014, "log_rms_mae_db": 1.3432342718523933, "rms_envelope_nrmse": 0.09686744997274395, "envelope_correlation": 0.9897577335792188, "rms_gain_error_db": -0.14893376595292876, "reference_mean_power": 0.0017439355594579833, "output_mean_power": 0.0016851441481678068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.14297331662316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004136999026359357, "output_power": 0.00040421632847495627, "reference_active": true, "delta_db": -0.10071565657303723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.498679838909322e-05, "output_power": 7.409671348687705e-05, "reference_active": true, "delta_db": -0.5955252331872454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.720897303077822e-05, "output_power": 1.7267122417523796e-05, "reference_active": true, "delta_db": 0.014650140283542102}], "active_band_count": 3, "active_band_mae_db": 0.2369636766812749}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011225274665232883, "log_rms_mae_db": 0.73656199214943, "rms_envelope_nrmse": 0.05422009389830893, "envelope_correlation": 0.9967115211203353, "rms_gain_error_db": -0.1494291152577394, "reference_mean_power": 0.0017512823148994152, "output_mean_power": 0.001692050226862109}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.129560410778096e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037542654743209525, "output_power": 0.00036692087600432926, "reference_active": true, "delta_db": -0.09952557822929935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.98989275491616e-05, "output_power": 3.4982892288108416e-05, "reference_active": true, "delta_db": -0.5710550947458619}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1683754663794263e-06, "output_power": 9.164806661191808e-07, "reference_active": true, "delta_db": -1.0545912144372895}], "active_band_count": 3, "active_band_mae_db": 0.5750572958041502}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.832439224759049e-05, "log_rms_mae_db": 1.5200289372174416, "rms_envelope_nrmse": 0.13789691649184316, "envelope_correlation": 0.984309162763195, "rms_gain_error_db": -0.03905321325338117, "reference_mean_power": 0.0005656706566344367, "output_mean_power": 0.000560606759032676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3997327154946766e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018576385507894965, "output_power": 0.00018052472488958295, "reference_active": true, "delta_db": -0.12424523393533908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.516179020835253e-05, "output_power": 3.311719962572926e-05, "reference_active": true, "delta_db": -0.26017372463236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.50699704718427e-06, "output_power": 8.699935608937385e-06, "reference_active": true, "delta_db": 0.640497932350777}], "active_band_count": 3, "active_band_mae_db": 0.34163896363949203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.746258281389908e-05, "log_rms_mae_db": 0.876012559163852, "rms_envelope_nrmse": 0.07736254630573126, "envelope_correlation": 0.9947613914994323, "rms_gain_error_db": -0.040171055532019995, "reference_mean_power": 0.0005675319915431874, "output_mean_power": 0.0005623066789648741}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.076532458167812e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017606095149491155, "output_power": 0.00017288747356596264, "reference_active": true, "delta_db": -0.0789951667745239}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7577844172032785e-05, "output_power": 1.6592203314146544e-05, "reference_active": true, "delta_db": -0.25061549376880593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.080515782562606e-07, "output_power": 5.396293586295714e-07, "reference_active": true, "delta_db": 0.26187764676707864}], "active_band_count": 3, "active_band_mae_db": 0.1971627691034695}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.679296131568703e-05, "log_rms_mae_db": 1.552103013087923, "rms_envelope_nrmse": 0.1493463487676144, "envelope_correlation": 0.9839731084059625, "rms_gain_error_db": -0.45226113707634846, "reference_mean_power": 0.00025821561430829635, "output_mean_power": 0.00023267857041852413}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9030353181622983e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.270493557085603e-05, "output_power": 8.304348587499274e-05, "reference_active": true, "delta_db": -0.47797285361348446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6852701179097796e-05, "output_power": 1.3889810506675157e-05, "reference_active": true, "delta_db": -0.8397319936556773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.066518069212071e-06, "output_power": 3.841655331308043e-06, "reference_active": true, "delta_db": -0.24704308494082106}], "active_band_count": 3, "active_band_mae_db": 0.5215826440699942}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.121002038281381e-05, "log_rms_mae_db": 0.9093622328965767, "rms_envelope_nrmse": 0.09434535559115828, "envelope_correlation": 0.994498130775114, "rms_gain_error_db": -0.45272549950306834, "reference_mean_power": 0.0002594441947231775, "output_mean_power": 0.00023376065069604488}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7574651639406604e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.903046567125758e-05, "output_power": 7.941510724948243e-05, "reference_active": true, "delta_db": -0.4963551819980617}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.408774632836393e-06, "output_power": 6.733993691477097e-06, "reference_active": true, "delta_db": -0.964600077262554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6645360300144737e-07, "output_power": 2.2660045299311212e-07, "reference_active": true, "delta_db": -0.7036082345345769}], "active_band_count": 3, "active_band_mae_db": 0.7215211645983975}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.318455452770038e-05, "log_rms_mae_db": 1.7703564069927258, "rms_envelope_nrmse": 0.1870676581373201, "envelope_correlation": 0.9754577805655407, "rms_gain_error_db": -0.2257548179909362, "reference_mean_power": 0.0001492769552865986, "output_mean_power": 0.00014171547821860558}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0711725750134131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.027413944353342e-05, "output_power": 6.242332644725777e-05, "reference_active": true, "delta_db": 0.15215889351410342}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2776930907508557e-05, "output_power": 1.0695552729705853e-05, "reference_active": true, "delta_db": -0.7722331315229509}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.16053005381717e-06, "output_power": 2.070400974272267e-06, "reference_active": true, "delta_db": -0.18505848539808817}], "active_band_count": 3, "active_band_mae_db": 0.3698168368117141}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.04724555086843e-05, "log_rms_mae_db": 1.0245526985165951, "rms_envelope_nrmse": 0.09920559099854889, "envelope_correlation": 0.9929881104189024, "rms_gain_error_db": -0.22588137971422867, "reference_mean_power": 0.00014998968981604926, "output_mean_power": 0.00014238796036469652}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.934560164313186e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.89522015257842e-05, "output_power": 6.0513953738575984e-05, "reference_active": true, "delta_db": 0.11355500779868423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6946349724579084e-06, "output_power": 4.796511265300876e-06, "reference_active": true, "delta_db": -0.7454042211053412}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8905823102195032e-07, "output_power": 1.403087977593454e-07, "reference_active": true, "delta_db": -1.2951068659742653}], "active_band_count": 3, "active_band_mae_db": 0.7180220316260969}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8303197671479006, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6aaf0ec7db348451f0e1f2536123bf441fddb7ec17d72aa82fbe15df812ce0ae", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9830465865453416, "standard_deviation": 0.0012854293872353838, "minimum": 0.9811076371781691, "maximum": 0.9844766091791085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5401417520516867, "standard_deviation": 0.041715031460942666, "minimum": 1.4807426139295667, "maximum": 1.5908783796409032, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4536564821790891, "standard_deviation": 0.20182233365228616, "minimum": 0.20900623695239254, "maximum": 0.8707746788447236, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15011629176468794, "standard_deviation": 0.005193450079503159, "minimum": 0.14381946546933982, "maximum": 0.1596935925892531, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951639347010425, "standard_deviation": 0.0009179386740645758, "minimum": 0.9942693242273347, "maximum": 0.9963318674012196, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8661283472104154, "standard_deviation": 0.04179122281080838, "minimum": 0.7930833766683417, "maximum": 0.910117317228362, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5813593016633698, "standard_deviation": 0.26152003117553363, "minimum": 0.34694644204691877, "maximum": 1.1131598405406926, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08333000597569773, "standard_deviation": 0.007555690774088255, "minimum": 0.07170742669516447, "maximum": 0.09434535559115828, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9740760762343752, "standard_deviation": 0.0024320632897524832, "minimum": 0.969339173482992, "maximum": 0.9776116734655942, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7796908033519454, "standard_deviation": 0.02150092061581468, "minimum": 1.7483937016193893, "maximum": 1.8168262003653204, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5757339804685462, "standard_deviation": 0.19256038997686314, "minimum": 0.3310404816485121, "maximum": 0.8128795401071466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19280663533269038, "standard_deviation": 0.009258278329450783, "minimum": 0.17856061828432046, "maximum": 0.20883984589502833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992967411986575, "standard_deviation": 0.0009913467388458688, "minimum": 0.991647660448744, "maximum": 0.9942250780560304, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0313260063335372, "standard_deviation": 0.034355653681340574, "minimum": 0.9772189914718462, "maximum": 1.0733896929434987, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5388398861635262, "standard_deviation": 0.3120223167476875, "minimum": 0.2664325060416491, "maximum": 1.226460438325625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09986665514430432, "standard_deviation": 0.006804485921830215, "minimum": 0.08946237333306448, "maximum": 0.10865314779057354, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9843458700010208, "standard_deviation": 0.0016503613275443841, "minimum": 0.981978630771942, "maximum": 0.9875232794411718, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5918164998227076, "standard_deviation": 0.039322488621557114, "minimum": 1.5200289372174416, "maximum": 1.6467428399422874, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3996924203072825, "standard_deviation": 0.09756786913104562, "minimum": 0.22915791303694458, "maximum": 0.5612851790506643, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13787048688462938, "standard_deviation": 0.007313890496989343, "minimum": 0.12352375678402477, "maximum": 0.1477556832262063, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947605416388945, "standard_deviation": 0.0008156270219480335, "minimum": 0.9931007757443872, "maximum": 0.9957263755414059, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9852366523649231, "standard_deviation": 0.04832368460428036, "minimum": 0.876012559163852, "maximum": 1.0348122699903286, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.32327865653980414, "standard_deviation": 0.10805607261475098, "minimum": 0.14073687088189563, "maximum": 0.49135070509084083, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07862156285997757, "standard_deviation": 0.0059279119208907735, "minimum": 0.07167197641035729, "maximum": 0.09037807243264977, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24876300563074757, "standard_deviation": 0.04845203983122037, "minimum": 0.16801256308867954, "maximum": 0.31114615694402253, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.735681652999321, "standard_deviation": 0.008573297366611794, "minimum": 5.724359442693498, "maximum": 5.75026613836426, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4547996958254765, "standard_deviation": 0.018291473814152064, "minimum": 0.4404781378687279, "maximum": 0.49069180171090654, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1341887973131565, "standard_deviation": 0.031261300072540556, "minimum": -0.1883810800569078, "maximum": -0.07675610770599907, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900466463555876, "standard_deviation": 0.0006685800239978391, "minimum": 0.9890850293331569, "maximum": 0.9910119190657416, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2742355065865678, "standard_deviation": 0.06214602311221574, "minimum": 1.1784789215178708, "maximum": 1.3432342718523933, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2559591096811189, "standard_deviation": 0.11637752196964121, "minimum": 0.11979660838716709, "maximum": 0.4898758682703325, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09535059633366536, "standard_deviation": 0.003234281188867392, "minimum": 0.09062961258304363, "maximum": 0.09991664790760534, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971264408529392, "standard_deviation": 0.00026436159162386693, "minimum": 0.9967115211203353, "maximum": 0.9974197225412001, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6971370246953326, "standard_deviation": 0.051789917036067674, "minimum": 0.6062304541888681, "maximum": 0.747927017053004, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4593352831160642, "standard_deviation": 0.15384000635642497, "minimum": 0.18184749778829423, "maximum": 0.6183965867355002, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.050593267267998704, "standard_deviation": 0.002097672485624505, "minimum": 0.047976463938294626, "maximum": 0.05422009389830893, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0029", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.754737377166748, "predicted_mos": 1.754737377166748, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8b24c5cf1c3b84be3f02e97ef3ef815cf890da018d0c74bee36ba679b1761420", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7771557001397014, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2467671439841002, "全帯域_10ms_rms_envelope_nrmse": 0.10101156832194705, "全帯域_10ms_envelope_correlation": 0.9887135496112271, "全帯域_10ms_modulation_mae_db": 0.16576302349440789, "全帯域_40ms_log_rms_mae_db": 0.647251121073083, "全帯域_40ms_rms_envelope_nrmse": 0.04791781998197991, "全帯域_40ms_envelope_correlation": 0.9972582396673728, "全帯域_40ms_modulation_mae_db": 0.1145639013245517, "300–1000 Hz_10ms_log_rms_mae_db": 1.8677067114946986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15549241734317049, "300–1000 Hz_10ms_envelope_correlation": 0.9800511300340643, "300–1000 Hz_10ms_modulation_mae_db": 0.3708288077944835, "300–1000 Hz_40ms_log_rms_mae_db": 1.2799661659175772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09208467860812898, "300–1000 Hz_40ms_envelope_correlation": 0.9928735473307598, "300–1000 Hz_40ms_modulation_mae_db": 0.39951460430930136, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8233490391570526, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1576629798084164, "1000–3000 Hz_10ms_envelope_correlation": 0.9807343682517436, "1000–3000 Hz_10ms_modulation_mae_db": 0.12637824682152696, "1000–3000 Hz_40ms_log_rms_mae_db": 1.074549733037375, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0835318989007945, "1000–3000 Hz_40ms_envelope_correlation": 0.9942153008170661, "1000–3000 Hz_40ms_modulation_mae_db": 0.10163177259373053, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3689575089432653, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20748211070261505, "3000–6000 Hz_10ms_envelope_correlation": 0.9697146400273498, "3000–6000 Hz_10ms_modulation_mae_db": 0.8207537918158576, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4446967263553463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10571908555485077, "3000–6000 Hz_40ms_envelope_correlation": 0.992238515880285, "3000–6000 Hz_40ms_modulation_mae_db": 0.3944841042210818, "mr_stft_db": 6.0682673739703965, "band_mae_db": 0.1635423927278177, "rms_gain_error_db": -0.07399741556961655, "output_peak": 0.4515470368129577}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.228454500945988, "全帯域_10ms_rms_envelope_nrmse": 0.09279307234152522, "全帯域_10ms_envelope_correlation": 0.9904630575117835, "全帯域_10ms_modulation_mae_db": 0.048159705310684864, "全帯域_40ms_log_rms_mae_db": 0.6530826870833205, "全帯域_40ms_rms_envelope_nrmse": 0.04475619311598627, "全帯域_40ms_envelope_correlation": 0.9975942623767653, "全帯域_40ms_modulation_mae_db": 0.18226257949632277, "300–1000 Hz_10ms_log_rms_mae_db": 1.7124920442576546, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1398755204945219, "300–1000 Hz_10ms_envelope_correlation": 0.9838299627589707, "300–1000 Hz_10ms_modulation_mae_db": 0.25597881888571944, "300–1000 Hz_40ms_log_rms_mae_db": 1.1584092434623043, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07473219763795876, "300–1000 Hz_40ms_envelope_correlation": 0.9951022663945026, "300–1000 Hz_40ms_modulation_mae_db": 0.21620804914100036, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8797907469108108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14632850955274035, "1000–3000 Hz_10ms_envelope_correlation": 0.9833939388995941, "1000–3000 Hz_10ms_modulation_mae_db": 0.1953742836818773, "1000–3000 Hz_40ms_log_rms_mae_db": 1.158416084503631, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08028295559472881, "1000–3000 Hz_40ms_envelope_correlation": 0.9946958312733063, "1000–3000 Hz_40ms_modulation_mae_db": 0.14477684322991463, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3328465086823136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23441105876925317, "3000–6000 Hz_10ms_envelope_correlation": 0.9617832005061467, "3000–6000 Hz_10ms_modulation_mae_db": 0.3005557896325782, "3000–6000 Hz_40ms_log_rms_mae_db": 1.413167801552862, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11676016835823819, "3000–6000 Hz_40ms_envelope_correlation": 0.9898542776701446, "3000–6000 Hz_40ms_modulation_mae_db": 0.8355365585385076, "mr_stft_db": 6.048875653379933, "band_mae_db": 0.18157054832632316, "rms_gain_error_db": -0.06112054581047158, "output_peak": 0.4547124529764821}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1633723142404553, "全帯域_10ms_rms_envelope_nrmse": 0.09601598163141199, "全帯域_10ms_envelope_correlation": 0.9899551855217269, "全帯域_10ms_modulation_mae_db": 0.3166470885910997, "全帯域_40ms_log_rms_mae_db": 0.6046266367467851, "全帯域_40ms_rms_envelope_nrmse": 0.052446796021243955, "全帯域_40ms_envelope_correlation": 0.9969765474959965, "全帯域_40ms_modulation_mae_db": 0.3545725962262938, "300–1000 Hz_10ms_log_rms_mae_db": 1.7721857587159568, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13949039821083767, "300–1000 Hz_10ms_envelope_correlation": 0.9840384930489303, "300–1000 Hz_10ms_modulation_mae_db": 0.5174471120202426, "300–1000 Hz_40ms_log_rms_mae_db": 1.1530638345527575, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07666938030057763, "300–1000 Hz_40ms_envelope_correlation": 0.995174450047012, "300–1000 Hz_40ms_modulation_mae_db": 0.7026218908385383, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8987027153202458, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15173975567108575, "1000–3000 Hz_10ms_envelope_correlation": 0.9823561876150935, "1000–3000 Hz_10ms_modulation_mae_db": 0.7724166295390154, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1088386356548057, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07786637931917972, "1000–3000 Hz_40ms_envelope_correlation": 0.9954941341551025, "1000–3000 Hz_40ms_modulation_mae_db": 0.8319746769565611, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3942233943543707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20178293343694959, "3000–6000 Hz_10ms_envelope_correlation": 0.9713033138490482, "3000–6000 Hz_10ms_modulation_mae_db": 0.27885594689861726, "3000–6000 Hz_40ms_log_rms_mae_db": 1.473089098786907, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09466149849357224, "3000–6000 Hz_40ms_envelope_correlation": 0.9933843326795168, "3000–6000 Hz_40ms_modulation_mae_db": 0.242501300190498, "mr_stft_db": 6.0607522090817225, "band_mae_db": 0.31105072761057184, "rms_gain_error_db": -0.15640187222796045, "output_peak": 0.4435292382755928}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2794442544774545, "全帯域_10ms_rms_envelope_nrmse": 0.09533348191687542, "全帯域_10ms_envelope_correlation": 0.9899865077557707, "全帯域_10ms_modulation_mae_db": 0.15210170128469336, "全帯域_40ms_log_rms_mae_db": 0.6589540569864506, "全帯域_40ms_rms_envelope_nrmse": 0.04893930971751884, "全帯域_40ms_envelope_correlation": 0.9972507247605665, "全帯域_40ms_modulation_mae_db": 0.21944054907404506, "300–1000 Hz_10ms_log_rms_mae_db": 1.8788566240714237, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13829512336924274, "300–1000 Hz_10ms_envelope_correlation": 0.9842589591628091, "300–1000 Hz_10ms_modulation_mae_db": 0.5460339883727973, "300–1000 Hz_40ms_log_rms_mae_db": 1.2296079448631525, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07892448360782234, "300–1000 Hz_40ms_envelope_correlation": 0.9948012185591847, "300–1000 Hz_40ms_modulation_mae_db": 0.430635688684539, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8538681594043445, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15949823464639656, "1000–3000 Hz_10ms_envelope_correlation": 0.9804813016956709, "1000–3000 Hz_10ms_modulation_mae_db": 0.6936643943379627, "1000–3000 Hz_40ms_log_rms_mae_db": 1.103678547703262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08989090258461561, "1000–3000 Hz_40ms_envelope_correlation": 0.9938806516737503, "1000–3000 Hz_40ms_modulation_mae_db": 0.7070257221080817, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3781812467710597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2103312165505824, "3000–6000 Hz_10ms_envelope_correlation": 0.9687714970682724, "3000–6000 Hz_10ms_modulation_mae_db": 0.49270969510155566, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4807640451084745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1176018961243052, "3000–6000 Hz_40ms_envelope_correlation": 0.9899280360709096, "3000–6000 Hz_40ms_modulation_mae_db": 0.9407717575117349, "mr_stft_db": 6.076794930090988, "band_mae_db": 0.2813771686442325, "rms_gain_error_db": -0.114396700867846, "output_peak": 0.43851195179403646}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2406462444720046, "全帯域_10ms_rms_envelope_nrmse": 0.09084439467416959, "全帯域_10ms_envelope_correlation": 0.9910939030062068, "全帯域_10ms_modulation_mae_db": 0.560766926328509, "全帯域_40ms_log_rms_mae_db": 0.658354913461199, "全帯域_40ms_rms_envelope_nrmse": 0.049243501369326696, "全帯域_40ms_envelope_correlation": 0.9974449277717462, "全帯域_40ms_modulation_mae_db": 0.7662454029377462, "300–1000 Hz_10ms_log_rms_mae_db": 1.8266768601241392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13278139955004217, "300–1000 Hz_10ms_envelope_correlation": 0.9854576790723881, "300–1000 Hz_10ms_modulation_mae_db": 0.5215828554539418, "300–1000 Hz_40ms_log_rms_mae_db": 1.1926262034608561, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07608804806928322, "300–1000 Hz_40ms_envelope_correlation": 0.9950033733177123, "300–1000 Hz_40ms_modulation_mae_db": 0.7930319866545993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.85995406016363, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14453935325600997, "1000–3000 Hz_10ms_envelope_correlation": 0.9841222019788767, "1000–3000 Hz_10ms_modulation_mae_db": 0.795701384200615, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0998285149860845, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07850680358652205, "1000–3000 Hz_40ms_envelope_correlation": 0.9955204863554189, "1000–3000 Hz_40ms_modulation_mae_db": 0.9185005361114887, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3572970256183696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20487308169452217, "3000–6000 Hz_10ms_envelope_correlation": 0.9708571504851163, "3000–6000 Hz_10ms_modulation_mae_db": 1.4232039662886755, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4238498854824273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1140593999999171, "3000–6000 Hz_40ms_envelope_correlation": 0.9914495678845912, "3000–6000 Hz_40ms_modulation_mae_db": 1.712156077295389, "mr_stft_db": 6.056792255861525, "band_mae_db": 0.17607035946922814, "rms_gain_error_db": -0.16879304117204497, "output_peak": 0.4431857511690637}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.205386937783791, "全帯域_10ms_rms_envelope_nrmse": 0.08830170952001372, "全帯域_10ms_envelope_correlation": 0.9913673586453644, "全帯域_10ms_modulation_mae_db": 0.2870488843832551, "全帯域_40ms_log_rms_mae_db": 0.6126356763455477, "全帯域_40ms_rms_envelope_nrmse": 0.0417342445945912, "全帯域_40ms_envelope_correlation": 0.9978997248062136, "全帯域_40ms_modulation_mae_db": 0.1402923850473283, "300–1000 Hz_10ms_log_rms_mae_db": 1.796907609041935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15022640199512832, "300–1000 Hz_10ms_envelope_correlation": 0.9813493930154407, "300–1000 Hz_10ms_modulation_mae_db": 0.4108826913877206, "300–1000 Hz_40ms_log_rms_mae_db": 1.2034294706637794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09250864672732559, "300–1000 Hz_40ms_envelope_correlation": 0.9926149494686637, "300–1000 Hz_40ms_modulation_mae_db": 0.4166346049502166, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8477790774188925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15417837686900285, "1000–3000 Hz_10ms_envelope_correlation": 0.9818157348291255, "1000–3000 Hz_10ms_modulation_mae_db": 0.47315098038057074, "1000–3000 Hz_40ms_log_rms_mae_db": 1.109185932124553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08466363043127267, "1000–3000 Hz_40ms_envelope_correlation": 0.9941577516035299, "1000–3000 Hz_40ms_modulation_mae_db": 0.3713555652420785, "3000–6000 Hz_10ms_log_rms_mae_db": 2.357575851187089, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19344691518534085, "3000–6000 Hz_10ms_envelope_correlation": 0.9738978305174392, "3000–6000 Hz_10ms_modulation_mae_db": 0.23107485275203055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3926795937924115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09010331493278685, "3000–6000 Hz_40ms_envelope_correlation": 0.9939346099857995, "3000–6000 Hz_40ms_modulation_mae_db": 0.4915315014140935, "mr_stft_db": 6.051378862618854, "band_mae_db": 0.15101323977284883, "rms_gain_error_db": -0.04627007037606495, "output_peak": 0.44656735172030987}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3098821835624086, "全帯域_10ms_rms_envelope_nrmse": 0.09867181957888693, "全帯域_10ms_envelope_correlation": 0.989232604459796, "全帯域_10ms_modulation_mae_db": 0.23197210723233927, "全帯域_40ms_log_rms_mae_db": 0.7169508787493275, "全帯域_40ms_rms_envelope_nrmse": 0.04852886373531266, "全帯域_40ms_envelope_correlation": 0.9972532137508423, "全帯域_40ms_modulation_mae_db": 0.3176204267998814, "300–1000 Hz_10ms_log_rms_mae_db": 1.8887571782433776, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13662353902694918, "300–1000 Hz_10ms_envelope_correlation": 0.9846002214874016, "300–1000 Hz_10ms_modulation_mae_db": 0.6833203766657103, "300–1000 Hz_40ms_log_rms_mae_db": 1.2730454969028868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07295632532807007, "300–1000 Hz_40ms_envelope_correlation": 0.9953624103937515, "300–1000 Hz_40ms_modulation_mae_db": 0.4264568454414784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8624027588256262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15775133567817584, "1000–3000 Hz_10ms_envelope_correlation": 0.9809238992999277, "1000–3000 Hz_10ms_modulation_mae_db": 0.5742866663882942, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1372165231232003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08070511088672522, "1000–3000 Hz_40ms_envelope_correlation": 0.9952245580597316, "1000–3000 Hz_40ms_modulation_mae_db": 0.5474025530989964, "3000–6000 Hz_10ms_log_rms_mae_db": 2.348934727011592, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2064550092987911, "3000–6000 Hz_10ms_envelope_correlation": 0.9700591513775345, "3000–6000 Hz_10ms_modulation_mae_db": 0.5324143144216149, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4212542536636301, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09919677634784205, "3000–6000 Hz_40ms_envelope_correlation": 0.9932645392458522, "3000–6000 Hz_40ms_modulation_mae_db": 0.687266542445815, "mr_stft_db": 6.049865870767236, "band_mae_db": 0.1828892167029611, "rms_gain_error_db": -0.10168674613725615, "output_peak": 0.4466598471465441}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.238760249614594, "全帯域_10ms_rms_envelope_nrmse": 0.09750333853145236, "全帯域_10ms_envelope_correlation": 0.9894666183207107, "全帯域_10ms_modulation_mae_db": 0.15863150269378176, "全帯域_40ms_log_rms_mae_db": 0.6684296651558032, "全帯域_40ms_rms_envelope_nrmse": 0.04765804501956918, "全帯域_40ms_envelope_correlation": 0.997282978910046, "全帯域_40ms_modulation_mae_db": 0.15518458030196894, "300–1000 Hz_10ms_log_rms_mae_db": 1.8027204878409742, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14700186868164028, "300–1000 Hz_10ms_envelope_correlation": 0.9821976435370202, "300–1000 Hz_10ms_modulation_mae_db": 0.5365038943794613, "300–1000 Hz_40ms_log_rms_mae_db": 1.209305445761516, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0851097605367122, "300–1000 Hz_40ms_envelope_correlation": 0.9938872086954245, "300–1000 Hz_40ms_modulation_mae_db": 0.4150201147258284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8742528478028972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.155759493685992, "1000–3000 Hz_10ms_envelope_correlation": 0.981273783120704, "1000–3000 Hz_10ms_modulation_mae_db": 0.12568902799467493, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1614446804095588, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08714174283833354, "1000–3000 Hz_40ms_envelope_correlation": 0.9937156705566961, "1000–3000 Hz_40ms_modulation_mae_db": 0.11727797008905673, "3000–6000 Hz_10ms_log_rms_mae_db": 2.295986543938852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21516028481167607, "3000–6000 Hz_10ms_envelope_correlation": 0.967296861320341, "3000–6000 Hz_10ms_modulation_mae_db": 1.0181034965030602, "3000–6000 Hz_40ms_log_rms_mae_db": 1.401380925276304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10756925319781037, "3000–6000 Hz_40ms_envelope_correlation": 0.9915596756602141, "3000–6000 Hz_40ms_modulation_mae_db": 1.6222911235765018, "mr_stft_db": 6.05835493217629, "band_mae_db": 0.17364554472767668, "rms_gain_error_db": -0.07205075265971828, "output_peak": 0.49297234977537235}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.324551520784919, "spectral_convergence": 0.3707902428765554, "spectral_convergence_status": "ok", "amplitude_floor": 1.016885378852613e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.881923647496567, "spectral_convergence": 0.3490142147076327, "spectral_convergence_status": "ok", "amplitude_floor": 6.009137305261044e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.998326953629704, "spectral_convergence": 0.3842121120354688, "spectral_convergence_status": "ok", "amplitude_floor": 5.349766097366943e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003378185892142299, "output_power": 0.00036041261107209923, "delta_db": 0.28116434913886174, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005186258824964237, "output_power": 0.0005126722973340739, "delta_db": -0.05014336217639416, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004467018619566432, "output_power": 0.00041397603631652215, "delta_db": -0.33042560717030167, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002054215054829419, "output_power": 0.0002022643123664676, "delta_db": -0.06726645373155113, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00019049740954433386, "output_power": 0.00018047595346269973, "delta_db": -0.23469729462520333, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.3523016958893975e-05, "output_power": 4.3347420973199696e-05, "delta_db": -0.0175572895245941, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1635423927278177, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07399741556961655, "pitch": null, "output_rms_dbfs": -27.66204956327577, "output_peak": 0.4515470368129577, "reference_rms_dbfs": -27.58805214770615, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0682673739703965}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001220285373493014, "log_rms_mae_db": 1.2467671439841002, "rms_envelope_nrmse": 0.10101156832194705, "envelope_correlation": 0.9887135496112271, "rms_gain_error_db": -0.0739183391881294, "reference_mean_power": 0.0017439355594579833, "output_mean_power": 0.00171450438130725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.14297331662316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004136999026359357, "output_power": 0.00041102217585119215, "reference_active": true, "delta_db": -0.028201647160257092}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.498679838909322e-05, "output_power": 8.799408963048054e-05, "reference_active": true, "delta_db": 0.1510203365771841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.720897303077822e-05, "output_power": 1.851661780971581e-05, "reference_active": true, "delta_db": 0.3180670867457825}], "active_band_count": 3, "active_band_mae_db": 0.16576302349440789}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011225274665232883, "log_rms_mae_db": 0.647251121073083, "rms_envelope_nrmse": 0.04791781998197991, "envelope_correlation": 0.9972582396673728, "rms_gain_error_db": -0.07329898807299372, "reference_mean_power": 0.0017512823148994152, "output_mean_power": 0.0017219727052126333}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.129560410778096e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037542654743209525, "output_power": 0.0003733945898543552, "reference_active": true, "delta_db": -0.023569583856651224}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.98989275491616e-05, "output_power": 4.169879831048954e-05, "reference_active": true, "delta_db": 0.19162317154775763}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1683754663794263e-06, "output_power": 1.2034618397777939e-06, "reference_active": true, "delta_db": 0.1284989485692463}], "active_band_count": 3, "active_band_mae_db": 0.1145639013245517}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.832439224759049e-05, "log_rms_mae_db": 1.8677067114946986, "rms_envelope_nrmse": 0.15549241734317049, "envelope_correlation": 0.9800511300340643, "rms_gain_error_db": -0.17986422493019855, "reference_mean_power": 0.0005656706566344367, "output_mean_power": 0.0005427217522482139}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3997327154946766e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018576385507894965, "output_power": 0.00017181504453113499, "reference_active": true, "delta_db": -0.3390002620737789}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.516179020835253e-05, "output_power": 3.0506121901494608e-05, "reference_active": true, "delta_db": -0.6168397704387047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.50699704718427e-06, "output_power": 7.782710646833188e-06, "reference_active": true, "delta_db": 0.15664639087096693}], "active_band_count": 3, "active_band_mae_db": 0.3708288077944835}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.746258281389908e-05, "log_rms_mae_db": 1.2799661659175772, "rms_envelope_nrmse": 0.09208467860812898, "envelope_correlation": 0.9928735473307598, "rms_gain_error_db": -0.17890058374974277, "reference_mean_power": 0.0005675319915431874, "output_mean_power": 0.0005446284062943993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.076532458167812e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017606095149491155, "output_power": 0.00016278164629441604, "reference_active": true, "delta_db": -0.3405760819509951}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7577844172032785e-05, "output_power": 1.5701837522469513e-05, "reference_active": true, "delta_db": -0.49015131015226426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.080515782562606e-07, "output_power": 4.6679496477582513e-07, "reference_active": true, "delta_db": -0.3678164208246445}], "active_band_count": 3, "active_band_mae_db": 0.39951460430930136}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.679296131568703e-05, "log_rms_mae_db": 1.8233490391570526, "rms_envelope_nrmse": 0.1576629798084164, "envelope_correlation": 0.9807343682517436, "rms_gain_error_db": -0.04442237852690648, "reference_mean_power": 0.00025821561430829635, "output_mean_power": 0.0002555878841104046}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9030353181622983e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.270493557085603e-05, "output_power": 9.386553272909891e-05, "reference_active": true, "delta_db": 0.05403292899503816}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6852701179097796e-05, "output_power": 1.7279966126716926e-05, "reference_active": true, "delta_db": 0.10873366580612788}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.066518069212071e-06, "output_power": 3.868885632684962e-06, "reference_active": true, "delta_db": -0.2163681456634148}], "active_band_count": 3, "active_band_mae_db": 0.12637824682152696}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.121002038281381e-05, "log_rms_mae_db": 1.074549733037375, "rms_envelope_nrmse": 0.0835318989007945, "envelope_correlation": 0.9942153008170661, "rms_gain_error_db": -0.04427880202654264, "reference_mean_power": 0.0002594441947231775, "output_mean_power": 0.0002568124518865439}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7574651639406604e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.903046567125758e-05, "output_power": 9.017649128587171e-05, "reference_active": true, "delta_db": 0.055546882102409184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.408774632836393e-06, "output_power": 8.459977010743465e-06, "reference_active": true, "delta_db": 0.026364699590381725}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6645360300144737e-07, "output_power": 2.5311811265435753e-07, "reference_active": true, "delta_db": -0.22298373608840072}], "active_band_count": 3, "active_band_mae_db": 0.10163177259373053}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.318455452770038e-05, "log_rms_mae_db": 2.3689575089432653, "rms_envelope_nrmse": 0.20748211070261505, "envelope_correlation": 0.9697146400273498, "rms_gain_error_db": -0.2935776753329108, "reference_mean_power": 0.0001492769552865986, "output_mean_power": 0.00013951952882145928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0711725750134131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.027413944353342e-05, "output_power": 5.48594444146783e-05, "reference_active": true, "delta_db": -0.40879613506036405}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2776930907508557e-05, "output_power": 1.0497715336211387e-05, "reference_active": true, "delta_db": -0.8533175436785351}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.16053005381717e-06, "output_power": 2.8482301644700375e-06, "reference_active": true, "delta_db": 1.2001476967086737}], "active_band_count": 3, "active_band_mae_db": 0.8207537918158576}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.04724555086843e-05, "log_rms_mae_db": 1.4446967263553463, "rms_envelope_nrmse": 0.10571908555485077, "envelope_correlation": 0.992238515880285, "rms_gain_error_db": -0.29366291781407045, "reference_mean_power": 0.00014998968981604926, "output_mean_power": 0.0001401829242427315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.934560164313186e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.89522015257842e-05, "output_power": 5.380698710119324e-05, "reference_active": true, "delta_db": -0.39661353538936284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6946349724579084e-06, "output_power": 5.014482440480463e-06, "reference_active": true, "delta_db": -0.5523977606577708}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8905823102195032e-07, "output_power": 1.7912305526809834e-07, "reference_active": true, "delta_db": -0.23444101661611172}], "active_band_count": 3, "active_band_mae_db": 0.3944841042210818}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8073950815267376, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "391cfec9098855a6d8228bdc871ed7c1a62f517963379af07447233a737a533b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.981887676961342, "standard_deviation": 0.0013147705057441502, "minimum": 0.9804813016956709, "maximum": 0.9841222019788767, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8625124256254373, "standard_deviation": 0.0226092204823335, "minimum": 1.8233490391570526, "maximum": 1.8987027153202458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46958270166806715, "standard_deviation": 0.2854801604092413, "minimum": 0.12568902799467493, "maximum": 0.795701384200615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15343225489597745, "standard_deviation": 0.005499789501305866, "minimum": 0.14453935325600997, "maximum": 0.15949823464639656, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946130480618252, "standard_deviation": 0.0007258239741066826, "minimum": 0.9937156705566961, "maximum": 0.9955204863554189, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1191448314428087, "standard_deviation": 0.03038923856277971, "minimum": 1.074549733037375, "maximum": 1.1614446804095588, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4674932049287385, "standard_deviation": 0.33164294078337336, "minimum": 0.10163177259373053, "maximum": 0.9185005361114887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08282367801777152, "standard_deviation": 0.004253270354139534, "minimum": 0.07786637931917972, "maximum": 0.08989090258461561, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9692104556439061, "standard_deviation": 0.0035668368471716864, "minimum": 0.9617832005061467, "maximum": 0.9738978305174392, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3542503508133636, "standard_deviation": 0.029971951596483484, "minimum": 2.295986543938852, "maximum": 2.3942233943543707, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6372089816767488, "standard_deviation": 0.4200538418924744, "minimum": 0.23107485275203055, "maximum": 1.4232039662886755, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2092428263062163, "standard_deviation": 0.011980748573297253, "minimum": 0.19344691518534085, "maximum": 0.23441105876925317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9919516943846641, "standard_deviation": 0.0015441314947971503, "minimum": 0.9898542776701446, "maximum": 0.9939346099857995, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4313602912522954, "standard_deviation": 0.03216996910816392, "minimum": 1.3926795937924115, "maximum": 1.4807640451084745, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8658173706492027, "standard_deviation": 0.5448099425910645, "minimum": 0.242501300190498, "maximum": 1.712156077295389, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10570892412616534, "standard_deviation": 0.010314711248253066, "minimum": 0.09010331493278685, "maximum": 0.1176018961243052, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9832229352646281, "standard_deviation": 0.0018369109057099408, "minimum": 0.9800511300340643, "maximum": 0.9854576790723881, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.81828790922377, "standard_deviation": 0.05995797907023201, "minimum": 1.7124920442576546, "maximum": 1.8887571782433776, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4803223181200096, "standard_deviation": 0.1303314364927043, "minimum": 0.25597881888571944, "maximum": 0.6833203766657103, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1424733335839416, "standard_deviation": 0.007665990760791762, "minimum": 0.13278139955004217, "maximum": 0.15549241734317049, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9943524280258764, "standard_deviation": 0.001089268021199392, "minimum": 0.9926149494686637, "maximum": 0.9953624103937515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2124317256981039, "standard_deviation": 0.04698936110474855, "minimum": 1.1530638345527575, "maximum": 1.2799661659175772, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4750154730931877, "standard_deviation": 0.18398607189165386, "minimum": 0.21620804914100036, "maximum": 0.7930319866545993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08113419010198485, "standard_deviation": 0.007775791010415394, "minimum": 0.07295632532807007, "maximum": 0.09250864672732559, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2026448997477075, "standard_deviation": 0.05919340229911539, "minimum": 0.15101323977284883, "maximum": 0.31105072761057184, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.058885260993368, "standard_deviation": 0.009671869124895954, "minimum": 6.048875653379933, "maximum": 6.076794930090988, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45221074745879486, "standard_deviation": 0.01721852755386983, "minimum": 0.43851195179403646, "maximum": 0.49297234977537235, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09933964310262236, "standard_deviation": 0.044683150167897964, "minimum": -0.16879304117204497, "maximum": -0.04627007037606495, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900348481040733, "standard_deviation": 0.0009103796304068313, "minimum": 0.9887135496112271, "maximum": 0.9913673586453644, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2390892286350996, "standard_deviation": 0.044213195274697144, "minimum": 1.1633723142404553, "maximum": 1.3098821835624086, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24013636741484637, "standard_deviation": 0.1547831956776255, "minimum": 0.048159705310684864, "maximum": 0.560766926328509, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09505942081453528, "standard_deviation": 0.004209000210079142, "minimum": 0.08830170952001372, "maximum": 0.10101156832194705, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973700774424437, "standard_deviation": 0.0002775529891846133, "minimum": 0.9969765474959965, "maximum": 0.9978997248062136, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6525357044501896, "standard_deviation": 0.034625564734176634, "minimum": 0.6046266367467851, "maximum": 0.7169508787493275, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2812728026510173, "standard_deviation": 0.21350284714200513, "minimum": 0.1145639013245517, "maximum": 0.7662454029377462, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04765309669444109, "standard_deviation": 0.003194307784256881, "minimum": 0.0417342445945912, "maximum": 0.052446796021243955, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0029", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.7918293476104736, "predicted_mos": 1.7918293476104736, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8623ca6bdff21b72998f436099581c160fcfb44a2166bc5542ad9b7ca59f0bd3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.98584290035069, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2956459572400296, "全帯域_10ms_rms_envelope_nrmse": 0.09889124324979935, "全帯域_10ms_envelope_correlation": 0.9892315737392123, "全帯域_10ms_modulation_mae_db": 0.2951676493892415, "全帯域_40ms_log_rms_mae_db": 0.6931109756401654, "全帯域_40ms_rms_envelope_nrmse": 0.05178451188809336, "全帯域_40ms_envelope_correlation": 0.9969071348627998, "全帯域_40ms_modulation_mae_db": 0.5198993037569639, "300–1000 Hz_10ms_log_rms_mae_db": 1.4524774881716616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14709705336271717, "300–1000 Hz_10ms_envelope_correlation": 0.9821012566272944, "300–1000 Hz_10ms_modulation_mae_db": 0.3832992035181599, "300–1000 Hz_40ms_log_rms_mae_db": 0.8667298307630517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08074904957744374, "300–1000 Hz_40ms_envelope_correlation": 0.9943189520652532, "300–1000 Hz_40ms_modulation_mae_db": 0.22674957005512672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5185304879100885, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14857711570373688, "1000–3000 Hz_10ms_envelope_correlation": 0.982994417135228, "1000–3000 Hz_10ms_modulation_mae_db": 0.5098632792174461, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8803582048028841, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08483410041012691, "1000–3000 Hz_40ms_envelope_correlation": 0.9942871630529675, "1000–3000 Hz_40ms_modulation_mae_db": 0.7973097758303954, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6215291938762226, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17192452656424073, "3000–6000 Hz_10ms_envelope_correlation": 0.9793935635188527, "3000–6000 Hz_10ms_modulation_mae_db": 0.44466510787756236, "3000–6000 Hz_40ms_log_rms_mae_db": 0.989417217214499, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0941312596572895, "3000–6000 Hz_40ms_envelope_correlation": 0.9938344786970902, "3000–6000 Hz_40ms_modulation_mae_db": 0.7685115869124872, "mr_stft_db": 5.664052043044183, "band_mae_db": 0.17205109161611745, "rms_gain_error_db": -0.12015399204078875, "output_peak": 0.4361680499464913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3451148313779258, "全帯域_10ms_rms_envelope_nrmse": 0.09313077482732927, "全帯域_10ms_envelope_correlation": 0.9905110278447624, "全帯域_10ms_modulation_mae_db": 0.29046333616029824, "全帯域_40ms_log_rms_mae_db": 0.6768609397822547, "全帯域_40ms_rms_envelope_nrmse": 0.046971524333968874, "全帯域_40ms_envelope_correlation": 0.9975257930847127, "全帯域_40ms_modulation_mae_db": 0.40071613754541624, "300–1000 Hz_10ms_log_rms_mae_db": 1.5044278421308335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13237238313253427, "300–1000 Hz_10ms_envelope_correlation": 0.9855372010557456, "300–1000 Hz_10ms_modulation_mae_db": 0.25305984915832996, "300–1000 Hz_40ms_log_rms_mae_db": 0.9220020041751349, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07638531064880628, "300–1000 Hz_40ms_envelope_correlation": 0.9949191998192837, "300–1000 Hz_40ms_modulation_mae_db": 0.15813496893095558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3958656720759044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1346329939094287, "1000–3000 Hz_10ms_envelope_correlation": 0.9863689195988379, "1000–3000 Hz_10ms_modulation_mae_db": 0.3711933114372345, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7889347665866584, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06502581604508308, "1000–3000 Hz_40ms_envelope_correlation": 0.9972463208745248, "1000–3000 Hz_40ms_modulation_mae_db": 0.5630126375566472, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6332579777372869, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17574781399341174, "3000–6000 Hz_10ms_envelope_correlation": 0.9788153179772505, "3000–6000 Hz_10ms_modulation_mae_db": 0.6793958602173907, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0068902954877852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0980541047623222, "3000–6000 Hz_40ms_envelope_correlation": 0.9938926525441948, "3000–6000 Hz_40ms_modulation_mae_db": 0.6941681688011586, "mr_stft_db": 5.665875659404939, "band_mae_db": 0.2524501672067964, "rms_gain_error_db": -0.12684867627112112, "output_peak": 0.4510984742842414}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3054999397448575, "全帯域_10ms_rms_envelope_nrmse": 0.09415397607151109, "全帯域_10ms_envelope_correlation": 0.9902307425719561, "全帯域_10ms_modulation_mae_db": 0.14711123148030442, "全帯域_40ms_log_rms_mae_db": 0.6880165934648675, "全帯域_40ms_rms_envelope_nrmse": 0.04661993435567714, "全帯域_40ms_envelope_correlation": 0.9974838800381824, "全帯域_40ms_modulation_mae_db": 0.32669048506406373, "300–1000 Hz_10ms_log_rms_mae_db": 1.4103173598548135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13405625701917293, "300–1000 Hz_10ms_envelope_correlation": 0.9851691554129555, "300–1000 Hz_10ms_modulation_mae_db": 0.17865743534597536, "300–1000 Hz_40ms_log_rms_mae_db": 0.9233002147207079, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08039757588171753, "300–1000 Hz_40ms_envelope_correlation": 0.9944055652985501, "300–1000 Hz_40ms_modulation_mae_db": 0.3377021190381173, "1000–3000 Hz_10ms_log_rms_mae_db": 1.404453474623382, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13116270015036258, "1000–3000 Hz_10ms_envelope_correlation": 0.9869835903181097, "1000–3000 Hz_10ms_modulation_mae_db": 0.626358736950433, "1000–3000 Hz_40ms_log_rms_mae_db": 0.835978119632748, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07745343456212732, "1000–3000 Hz_40ms_envelope_correlation": 0.9955665817393421, "1000–3000 Hz_40ms_modulation_mae_db": 0.8473871516879545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.649292060818321, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19371697320241482, "3000–6000 Hz_10ms_envelope_correlation": 0.9737256486423457, "3000–6000 Hz_10ms_modulation_mae_db": 0.62073718054095, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9820635300660008, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09332974994033721, "3000–6000 Hz_40ms_envelope_correlation": 0.993596225464631, "3000–6000 Hz_40ms_modulation_mae_db": 0.4906020067527401, "mr_stft_db": 5.652597619830097, "band_mae_db": 0.1753489395613004, "rms_gain_error_db": -0.10234577516452792, "output_peak": 0.45525095866987586}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3728627331342442, "全帯域_10ms_rms_envelope_nrmse": 0.09963851697663713, "全帯域_10ms_envelope_correlation": 0.9892144324176686, "全帯域_10ms_modulation_mae_db": 0.40914046113738367, "全帯域_40ms_log_rms_mae_db": 0.7439947482426079, "全帯域_40ms_rms_envelope_nrmse": 0.051514156479484784, "全帯域_40ms_envelope_correlation": 0.9971440882704493, "全帯域_40ms_modulation_mae_db": 0.41046551220954725, "300–1000 Hz_10ms_log_rms_mae_db": 1.583776256603372, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1571768062146705, "300–1000 Hz_10ms_envelope_correlation": 0.9802874367866256, "300–1000 Hz_10ms_modulation_mae_db": 0.8311210127950647, "300–1000 Hz_40ms_log_rms_mae_db": 1.0298733236203073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09028000236786422, "300–1000 Hz_40ms_envelope_correlation": 0.9941627459259083, "300–1000 Hz_40ms_modulation_mae_db": 0.6290806694462175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.429851896798454, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14042701122353604, "1000–3000 Hz_10ms_envelope_correlation": 0.9848617533063352, "1000–3000 Hz_10ms_modulation_mae_db": 0.3042656981834439, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8252372454375201, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0716903698923676, "1000–3000 Hz_40ms_envelope_correlation": 0.9961063864336204, "1000–3000 Hz_40ms_modulation_mae_db": 0.35153795305990804, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6890562569748824, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20357657367248758, "3000–6000 Hz_10ms_envelope_correlation": 0.9711914011689128, "3000–6000 Hz_10ms_modulation_mae_db": 1.199527938722275, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0482997592634613, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12096685107404265, "3000–6000 Hz_40ms_envelope_correlation": 0.9900872796420406, "3000–6000 Hz_40ms_modulation_mae_db": 0.8648413904338415, "mr_stft_db": 5.696897954149268, "band_mae_db": 0.30673280697790967, "rms_gain_error_db": -0.16661893742990552, "output_peak": 0.46489583228577996}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3352093773526625, "全帯域_10ms_rms_envelope_nrmse": 0.09541886390701185, "全帯域_10ms_envelope_correlation": 0.989928723303355, "全帯域_10ms_modulation_mae_db": 0.3502378715684807, "全帯域_40ms_log_rms_mae_db": 0.7667781585555985, "全帯域_40ms_rms_envelope_nrmse": 0.04835933571734523, "全帯域_40ms_envelope_correlation": 0.997241577682586, "全帯域_40ms_modulation_mae_db": 0.49274788373718076, "300–1000 Hz_10ms_log_rms_mae_db": 1.491766534161283, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14233930145876633, "300–1000 Hz_10ms_envelope_correlation": 0.983257734794584, "300–1000 Hz_10ms_modulation_mae_db": 0.23904722150083982, "300–1000 Hz_40ms_log_rms_mae_db": 0.8922823518477537, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.078137369066309, "300–1000 Hz_40ms_envelope_correlation": 0.9946376923820557, "300–1000 Hz_40ms_modulation_mae_db": 0.23533387905689343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4374909045647115, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14383526210971953, "1000–3000 Hz_10ms_envelope_correlation": 0.9840293902404855, "1000–3000 Hz_10ms_modulation_mae_db": 0.32150032912059623, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8469583404880113, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07379063098237784, "1000–3000 Hz_40ms_envelope_correlation": 0.9957242413692731, "1000–3000 Hz_40ms_modulation_mae_db": 0.38285908567133164, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5740221626107782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18972053015221854, "3000–6000 Hz_10ms_envelope_correlation": 0.9747193163128192, "3000–6000 Hz_10ms_modulation_mae_db": 0.4757996697057731, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9240429107503414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09614214911303771, "3000–6000 Hz_40ms_envelope_correlation": 0.9931561329639537, "3000–6000 Hz_40ms_modulation_mae_db": 0.3154359991033233, "mr_stft_db": 5.663431036088995, "band_mae_db": 0.2396238070451695, "rms_gain_error_db": -0.09011998114617623, "output_peak": 0.45018247697738095}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2778642169981302, "全帯域_10ms_rms_envelope_nrmse": 0.10025948874542502, "全帯域_10ms_envelope_correlation": 0.9890733231692806, "全帯域_10ms_modulation_mae_db": 0.2714314993341173, "全帯域_40ms_log_rms_mae_db": 0.7104416159145424, "全帯域_40ms_rms_envelope_nrmse": 0.05608516182085213, "全帯域_40ms_envelope_correlation": 0.9965417947427513, "全帯域_40ms_modulation_mae_db": 0.28332034551876956, "300–1000 Hz_10ms_log_rms_mae_db": 1.5136202032078538, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13938673897768933, "300–1000 Hz_10ms_envelope_correlation": 0.984089376179102, "300–1000 Hz_10ms_modulation_mae_db": 0.39636128339684035, "300–1000 Hz_40ms_log_rms_mae_db": 0.9668735478412092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08485815154528588, "300–1000 Hz_40ms_envelope_correlation": 0.993664631634859, "300–1000 Hz_40ms_modulation_mae_db": 0.3134397753151334, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4355347481768748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14383953438425615, "1000–3000 Hz_10ms_envelope_correlation": 0.9846709632306568, "1000–3000 Hz_10ms_modulation_mae_db": 0.6102604299908564, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8732352940208223, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08737212383214003, "1000–3000 Hz_40ms_envelope_correlation": 0.9948147358583439, "1000–3000 Hz_40ms_modulation_mae_db": 0.4187089636924581, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6324414536932463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17973389687903196, "3000–6000 Hz_10ms_envelope_correlation": 0.9785314516186221, "3000–6000 Hz_10ms_modulation_mae_db": 0.7518573120448061, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0257142451957157, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09972832738252724, "3000–6000 Hz_40ms_envelope_correlation": 0.9947197742187206, "3000–6000 Hz_40ms_modulation_mae_db": 1.0666362389914497, "mr_stft_db": 5.671426175564427, "band_mae_db": 0.2696165602512292, "rms_gain_error_db": -0.1703189213248152, "output_peak": 0.4749919954125647}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2308492609054125, "全帯域_10ms_rms_envelope_nrmse": 0.10059589229215828, "全帯域_10ms_envelope_correlation": 0.9890931669380413, "全帯域_10ms_modulation_mae_db": 0.6173628749077572, "全帯域_40ms_log_rms_mae_db": 0.6245502192779326, "全帯域_40ms_rms_envelope_nrmse": 0.0548108353095856, "全帯域_40ms_envelope_correlation": 0.9968787795335599, "全帯域_40ms_modulation_mae_db": 0.7008767965796835, "300–1000 Hz_10ms_log_rms_mae_db": 1.4658390575818616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14686968737063605, "300–1000 Hz_10ms_envelope_correlation": 0.9823627645191196, "300–1000 Hz_10ms_modulation_mae_db": 0.4001525345509713, "300–1000 Hz_40ms_log_rms_mae_db": 0.9085880418380413, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08225407154559558, "300–1000 Hz_40ms_envelope_correlation": 0.9945813418035666, "300–1000 Hz_40ms_modulation_mae_db": 0.356238483637809, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4405739845869934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13669590835644885, "1000–3000 Hz_10ms_envelope_correlation": 0.9857048509818296, "1000–3000 Hz_10ms_modulation_mae_db": 0.3347401152639551, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8735491558502629, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07470683127951622, "1000–3000 Hz_40ms_envelope_correlation": 0.9957497998075586, "1000–3000 Hz_40ms_modulation_mae_db": 0.2615693620776895, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6819497333824638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1994987744274741, "3000–6000 Hz_10ms_envelope_correlation": 0.9725790889182849, "3000–6000 Hz_10ms_modulation_mae_db": 0.6411673199300839, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0181530138322803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10537670430747285, "3000–6000 Hz_40ms_envelope_correlation": 0.9932579324932542, "3000–6000 Hz_40ms_modulation_mae_db": 0.7095571218120965, "mr_stft_db": 5.668495372584132, "band_mae_db": 0.20809983182705757, "rms_gain_error_db": -0.19697812600696182, "output_peak": 0.44007087678134776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3262799356362844, "全帯域_10ms_rms_envelope_nrmse": 0.09821812364661758, "全帯域_10ms_envelope_correlation": 0.9893789742568262, "全帯域_10ms_modulation_mae_db": 0.11824488622339473, "全帯域_40ms_log_rms_mae_db": 0.6944257252321844, "全帯域_40ms_rms_envelope_nrmse": 0.0518380637621315, "全帯域_40ms_envelope_correlation": 0.9968773644830575, "全帯域_40ms_modulation_mae_db": 0.1831738790350278, "300–1000 Hz_10ms_log_rms_mae_db": 1.5272616631860223, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14475855569026166, "300–1000 Hz_10ms_envelope_correlation": 0.9826691175418136, "300–1000 Hz_10ms_modulation_mae_db": 0.3288431590006298, "300–1000 Hz_40ms_log_rms_mae_db": 0.9890059483648199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0795305194617371, "300–1000 Hz_40ms_envelope_correlation": 0.9944606466234471, "300–1000 Hz_40ms_modulation_mae_db": 0.4361335702031903, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4043183659827112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1535777084721465, "1000–3000 Hz_10ms_envelope_correlation": 0.9819039299395066, "1000–3000 Hz_10ms_modulation_mae_db": 0.5401647512382769, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8273861013625979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09602676458956903, "1000–3000 Hz_40ms_envelope_correlation": 0.9927721516651878, "1000–3000 Hz_40ms_modulation_mae_db": 0.372604812269775, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6220605842799067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19188511954046425, "3000–6000 Hz_10ms_envelope_correlation": 0.9742542906433528, "3000–6000 Hz_10ms_modulation_mae_db": 0.6504099301684393, "3000–6000 Hz_40ms_log_rms_mae_db": 1.013643314314125, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09438667429236305, "3000–6000 Hz_40ms_envelope_correlation": 0.993423676439214, "3000–6000 Hz_40ms_modulation_mae_db": 0.09327752449606315, "mr_stft_db": 5.658044670538612, "band_mae_db": 0.2521810327464183, "rms_gain_error_db": -0.10758001686760416, "output_peak": 0.44072305465733913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.823735095009048, "spectral_convergence": 0.36743609793101945, "spectral_convergence_status": "ok", "amplitude_floor": 1.016885378852613e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.373302844728376, "spectral_convergence": 0.3479565371267535, "spectral_convergence_status": "ok", "amplitude_floor": 6.009137305261044e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.795118189395126, "spectral_convergence": 0.3851677413486691, "spectral_convergence_status": "ok", "amplitude_floor": 5.349766097366943e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003378185892142299, "output_power": 0.00032477945614308595, "delta_db": -0.17094993675043715, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005186258824964237, "output_power": 0.0005333844685160467, "delta_db": 0.12186178594287073, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004467018619566432, "output_power": 0.00041624946625760885, "delta_db": -0.30664073081587245, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002054215054829419, "output_power": 0.00019249574701428316, "delta_db": -0.28224769048137177, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00019049740954433386, "output_power": 0.00018449064820654355, "delta_db": -0.1391471759752867, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.3523016958893975e-05, "output_power": 4.363800778519496e-05, "delta_db": 0.011459229730865932, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17205109161611745, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12015399204078875, "pitch": null, "output_rms_dbfs": -27.708206139746938, "output_peak": 0.4361680499464913, "reference_rms_dbfs": -27.58805214770615, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.664052043044183}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001220285373493014, "log_rms_mae_db": 1.2956459572400296, "rms_envelope_nrmse": 0.09889124324979935, "envelope_correlation": 0.9892315737392123, "rms_gain_error_db": -0.12012912217016669, "reference_mean_power": 0.0017439355594579833, "output_mean_power": 0.0016963580379265198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.14297331662316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004136999026359357, "output_power": 0.0004026689760959424, "reference_active": true, "delta_db": -0.117372483007544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.498679838909322e-05, "output_power": 7.541932843535589e-05, "reference_active": true, "delta_db": -0.5186880791607904}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.720897303077822e-05, "output_power": 1.6248404245343906e-05, "reference_active": true, "delta_db": -0.24944238599939006}], "active_band_count": 3, "active_band_mae_db": 0.2951676493892415}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011225274665232883, "log_rms_mae_db": 0.6931109756401654, "rms_envelope_nrmse": 0.05178451188809336, "envelope_correlation": 0.9969071348627998, "rms_gain_error_db": -0.11980568852460773, "reference_mean_power": 0.0017512823148994152, "output_mean_power": 0.0017036312318111661}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.129560410778096e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037542654743209525, "output_power": 0.0003648017558565272, "reference_active": true, "delta_db": -0.12468059516065598}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.98989275491616e-05, "output_power": 3.623064710375865e-05, "reference_active": true, "delta_db": -0.41885131464158376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1683754663794263e-06, "output_power": 9.246254198399679e-07, "reference_active": true, "delta_db": -1.016166001468652}], "active_band_count": 3, "active_band_mae_db": 0.5198993037569639}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.832439224759049e-05, "log_rms_mae_db": 1.4524774881716616, "rms_envelope_nrmse": 0.14709705336271717, "envelope_correlation": 0.9821012566272944, "rms_gain_error_db": -0.07575695197624986, "reference_mean_power": 0.0005656706566344367, "output_mean_power": 0.0005558888405961562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3997327154946766e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018576385507894965, "output_power": 0.00017810954507105613, "reference_active": true, "delta_db": -0.18274020849856148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.516179020835253e-05, "output_power": 3.2600781793801354e-05, "reference_active": true, "delta_db": -0.3284296346389737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.50699704718427e-06, "output_power": 8.696390276574565e-06, "reference_active": true, "delta_db": 0.6387277674169445}], "active_band_count": 3, "active_band_mae_db": 0.3832992035181599}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.746258281389908e-05, "log_rms_mae_db": 0.8667298307630517, "rms_envelope_nrmse": 0.08074904957744374, "envelope_correlation": 0.9943189520652532, "rms_gain_error_db": -0.07490111046882017, "reference_mean_power": 0.0005675319915431874, "output_mean_power": 0.0005578279059232801}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.076532458167812e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017606095149491155, "output_power": 0.00016929497593750855, "reference_active": true, "delta_db": -0.17018974594828581}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7577844172032785e-05, "output_power": 1.580488374477076e-05, "reference_active": true, "delta_db": -0.46174304455177007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.080515782562606e-07, "output_power": 5.024307516561355e-07, "reference_active": true, "delta_db": -0.04831591966532424}], "active_band_count": 3, "active_band_mae_db": 0.22674957005512672}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.679296131568703e-05, "log_rms_mae_db": 1.5185304879100885, "rms_envelope_nrmse": 0.14857711570373688, "envelope_correlation": 0.982994417135228, "rms_gain_error_db": -0.1501329667001289, "reference_mean_power": 0.00025821561430829635, "output_mean_power": 0.00024944178396671304}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9030353181622983e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.270493557085603e-05, "output_power": 8.629791274335205e-05, "reference_active": true, "delta_db": -0.3110256469379706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6852701179097796e-05, "output_power": 1.273558351985571e-05, "reference_active": true, "delta_db": -1.2165067194114512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.066518069212071e-06, "output_power": 4.064592011429163e-06, "reference_active": true, "delta_db": -0.002057471302916643}], "active_band_count": 3, "active_band_mae_db": 0.5098632792174461}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.121002038281381e-05, "log_rms_mae_db": 0.8803582048028841, "rms_envelope_nrmse": 0.08483410041012691, "envelope_correlation": 0.9942871630529675, "rms_gain_error_db": -0.15023143547864287, "reference_mean_power": 0.0002594441947231775, "output_mean_power": 0.0002506229363119146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7574651639406604e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.903046567125758e-05, "output_power": 8.221859210930744e-05, "reference_active": true, "delta_db": -0.34568609189254335}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.408774632836393e-06, "output_power": 6.219371464647128e-06, "reference_active": true, "delta_db": -1.3098621621185775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6645360300144737e-07, "output_power": 2.248969107211699e-07, "reference_active": true, "delta_db": -0.7363810734800655}], "active_band_count": 3, "active_band_mae_db": 0.7973097758303954}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.318455452770038e-05, "log_rms_mae_db": 1.6215291938762226, "rms_envelope_nrmse": 0.17192452656424073, "envelope_correlation": 0.9793935635188527, "rms_gain_error_db": -0.2547871016142784, "reference_mean_power": 0.0001492769552865986, "output_mean_power": 0.00014077127958465383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0711725750134131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.027413944353342e-05, "output_power": 5.899787174175404e-05, "reference_active": true, "delta_db": -0.09294673095142836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2776930907508557e-05, "output_power": 9.85002461311404e-06, "reference_active": true, "delta_db": -1.1298923060188588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.16053005381717e-06, "output_power": 2.2165418757115253e-06, "reference_active": true, "delta_db": 0.11115628666239984}], "active_band_count": 3, "active_band_mae_db": 0.44466510787756236}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.04724555086843e-05, "log_rms_mae_db": 0.989417217214499, "rms_envelope_nrmse": 0.0941312596572895, "envelope_correlation": 0.9938344786970902, "rms_gain_error_db": -0.25473679090158013, "reference_mean_power": 0.00014998968981604926, "output_mean_power": 0.00014144504165475216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.934560164313186e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.89522015257842e-05, "output_power": 5.704181439336793e-05, "reference_active": true, "delta_db": -0.14306696977646088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6946349724579084e-06, "output_power": 4.532434540528793e-06, "reference_active": true, "delta_db": -0.9913435038581618}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8905823102195032e-07, "output_power": 1.4437206263719316e-07, "reference_active": true, "delta_db": -1.171124287102839}], "active_band_count": 3, "active_band_mae_db": 0.7685115869124872}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8212849409627758, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b7bf0ca3a67884ae51aee2925ad0dfe52d788c9eca6dbb5882279ed24e1b6213", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9846897268438737, "standard_deviation": 0.001699273120634269, "minimum": 0.9819039299395066, "maximum": 0.9869835903181097, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4333274418398898, "standard_deviation": 0.038591216870172265, "minimum": 1.3958656720759044, "maximum": 1.5185304879100885, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4522933314252803, "standard_deviation": 0.13400587218774743, "minimum": 0.3042656981834439, "maximum": 0.626358736950433, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1415935292887044, "standard_deviation": 0.007416772876576012, "minimum": 0.13116270015036258, "maximum": 0.1535777084721465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9952834226001023, "standard_deviation": 0.0013384921766000147, "minimum": 0.9927721516651878, "maximum": 0.9972463208745248, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8439546535226881, "standard_deviation": 0.031129000017559458, "minimum": 0.7889347665866584, "maximum": 0.8803582048028841, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.49937371773076994, "standard_deviation": 0.2166261754894827, "minimum": 0.2615693620776895, "maximum": 0.8473871516879545, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07886250894916351, "standard_deviation": 0.0099323271947604, "minimum": 0.06502581604508308, "maximum": 0.09602676458956903, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.975401259850055, "standard_deviation": 0.003109161557618033, "minimum": 0.9711914011689128, "maximum": 0.9793935635188527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6379511779216385, "standard_deviation": 0.03652340213868357, "minimum": 1.5740221626107782, "maximum": 1.6890562569748824, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.68294503990091, "standard_deviation": 0.23237274511282505, "minimum": 0.44466510787756236, "maximum": 1.199527938722275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18822552605396797, "standard_deviation": 0.011351828638569046, "minimum": 0.17192452656424073, "maximum": 0.20357657367248758, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932460190578873, "standard_deviation": 0.0013667707873556013, "minimum": 0.9900872796420406, "maximum": 0.9947197742187206, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.001028035765526, "standard_deviation": 0.037324779305919195, "minimum": 0.9240429107503414, "maximum": 1.0482997592634613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.625378754662895, "standard_deviation": 0.31207062220723336, "minimum": 0.09327752449606315, "maximum": 1.0666362389914497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10026447756617404, "standard_deviation": 0.009236150314534654, "minimum": 0.09332974994033721, "maximum": 0.12096685107404265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.983184255364655, "standard_deviation": 0.0017240103938737966, "minimum": 0.9802874367866256, "maximum": 0.9855372010557456, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4936858006122127, "standard_deviation": 0.052350395387685954, "minimum": 1.4103173598548135, "maximum": 1.583776256603372, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.37631771240835143, "standard_deviation": 0.20114803845565096, "minimum": 0.17865743534597536, "maximum": 0.8311210127950647, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14300709790330604, "standard_deviation": 0.007948613829666034, "minimum": 0.13237238313253427, "maximum": 0.1571768062146705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9943938469441154, "standard_deviation": 0.0003714800458228999, "minimum": 0.993664631634859, "maximum": 0.9949191998192837, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9373319078963782, "standard_deviation": 0.053956395842688104, "minimum": 0.8667298307630517, "maximum": 1.0298733236203073, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3366016294604304, "standard_deviation": 0.1466516113659914, "minimum": 0.15813496893095558, "maximum": 0.6290806694462175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08157400626184491, "standard_deviation": 0.0043427424660013566, "minimum": 0.07638531064880628, "maximum": 0.09028000236786422, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23451302965399984, "standard_deviation": 0.04661010567753978, "minimum": 0.17205109161611745, "maximum": 0.30673280697790967, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.667602566400582, "standard_deviation": 0.013218648716182838, "minimum": 5.652597619830097, "maximum": 5.696897954149268, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45167271487687766, "standard_deviation": 0.013249724575740824, "minimum": 0.4361680499464913, "maximum": 0.4749919954125647, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1351205532814876, "standard_deviation": 0.0381867090642321, "minimum": -0.19697812600696182, "maximum": -0.09011998114617623, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9895827455301378, "standard_deviation": 0.0005607400834581411, "minimum": 0.9890733231692806, "maximum": 0.9905110278447624, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3111657815486932, "standard_deviation": 0.04407507940500322, "minimum": 1.2308492609054125, "maximum": 1.3728627331342442, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31239497627512225, "standard_deviation": 0.1564179157601159, "minimum": 0.11824488622339473, "maximum": 0.6173628749077572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0975383599645612, "standard_deviation": 0.002898994941948231, "minimum": 0.09313077482732927, "maximum": 0.10059589229215828, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970750515872624, "standard_deviation": 0.0003365725145089855, "minimum": 0.9965417947427513, "maximum": 0.9975257930847127, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6997723720137692, "standard_deviation": 0.0430177569759367, "minimum": 0.6245502192779326, "maximum": 0.7667781585555985, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4147362929308316, "standard_deviation": 0.15940945766258072, "minimum": 0.1831738790350278, "maximum": 0.7008767965796835, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05099794045839232, "standard_deviation": 0.0034713144550066843, "minimum": 0.04661993435567714, "maximum": 0.05608516182085213, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0029", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.7998406887054443, "predicted_mos": 1.7998406887054443, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "bfece6ccfde3e42024b7068b6af7e158c24d4db9a7398a8be720ad4bd22b068a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.810366000048816, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.28575061553603, "全帯域_10ms_rms_envelope_nrmse": 0.09634029793985581, "全帯域_10ms_envelope_correlation": 0.9898893520170052, "全帯域_10ms_modulation_mae_db": 0.269445445330953, "全帯域_40ms_log_rms_mae_db": 0.681701041234958, "全帯域_40ms_rms_envelope_nrmse": 0.052708303140534055, "全帯域_40ms_envelope_correlation": 0.9969393641667018, "全帯域_40ms_modulation_mae_db": 0.5879087423537214, "300–1000 Hz_10ms_log_rms_mae_db": 1.6174118204433305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13638603672841085, "300–1000 Hz_10ms_envelope_correlation": 0.9846412509739252, "300–1000 Hz_10ms_modulation_mae_db": 0.40312735056165083, "300–1000 Hz_40ms_log_rms_mae_db": 1.0697286201697125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07721273249308527, "300–1000 Hz_40ms_envelope_correlation": 0.9948228535610477, "300–1000 Hz_40ms_modulation_mae_db": 0.2710606643478517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5803618866762958, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14200017845024016, "1000–3000 Hz_10ms_envelope_correlation": 0.9852713982976798, "1000–3000 Hz_10ms_modulation_mae_db": 0.5938537657930837, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9875997204299539, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09094795225681189, "1000–3000 Hz_40ms_envelope_correlation": 0.994492822376693, "1000–3000 Hz_40ms_modulation_mae_db": 0.727906496088433, "3000–6000 Hz_10ms_log_rms_mae_db": 1.895510911245056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18692674560125996, "3000–6000 Hz_10ms_envelope_correlation": 0.9755299263983275, "3000–6000 Hz_10ms_modulation_mae_db": 0.3596808002623237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1727331128693539, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10097550321260537, "3000–6000 Hz_40ms_envelope_correlation": 0.9928629557425152, "3000–6000 Hz_40ms_modulation_mae_db": 0.6241948386227124, "mr_stft_db": 5.807372449133901, "band_mae_db": 0.2875904151279404, "rms_gain_error_db": -0.15581104386034772, "output_peak": 0.4417002118673188}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2892139705083256, "全帯域_10ms_rms_envelope_nrmse": 0.10061451138193943, "全帯域_10ms_envelope_correlation": 0.9889238398005796, "全帯域_10ms_modulation_mae_db": 0.3038212009872684, "全帯域_40ms_log_rms_mae_db": 0.7055239776518886, "全帯域_40ms_rms_envelope_nrmse": 0.050802726062303634, "全帯域_40ms_envelope_correlation": 0.9971514495093061, "全帯域_40ms_modulation_mae_db": 0.5111598131022831, "300–1000 Hz_10ms_log_rms_mae_db": 1.6714105021115135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13337638464990023, "300–1000 Hz_10ms_envelope_correlation": 0.9853096373559198, "300–1000 Hz_10ms_modulation_mae_db": 0.45234723688640904, "300–1000 Hz_40ms_log_rms_mae_db": 1.1889216454353224, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07348001169503218, "300–1000 Hz_40ms_envelope_correlation": 0.9952843435299574, "300–1000 Hz_40ms_modulation_mae_db": 0.38747189391006603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5748331852673194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1476711238820184, "1000–3000 Hz_10ms_envelope_correlation": 0.983751443918083, "1000–3000 Hz_10ms_modulation_mae_db": 0.7349972274264028, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9242928941014448, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07897992960679903, "1000–3000 Hz_40ms_envelope_correlation": 0.9959909811916907, "1000–3000 Hz_40ms_modulation_mae_db": 0.9176999244705474, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9390307480387705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2111225874619472, "3000–6000 Hz_10ms_envelope_correlation": 0.9685506280909659, "3000–6000 Hz_10ms_modulation_mae_db": 0.6949993040066316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1898553422546385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10793216308383215, "3000–6000 Hz_40ms_envelope_correlation": 0.9915810722510199, "3000–6000 Hz_40ms_modulation_mae_db": 0.3100820273201962, "mr_stft_db": 5.807378716870037, "band_mae_db": 0.16438283270903573, "rms_gain_error_db": -0.14873113782482753, "output_peak": 0.4746586103180451}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2127210473780556, "全帯域_10ms_rms_envelope_nrmse": 0.0921287914359416, "全帯域_10ms_envelope_correlation": 0.9906965189463148, "全帯域_10ms_modulation_mae_db": 0.48332367136475535, "全帯域_40ms_log_rms_mae_db": 0.6844762310597963, "全帯域_40ms_rms_envelope_nrmse": 0.0527574359939911, "全帯域_40ms_envelope_correlation": 0.9968055054433101, "全帯域_40ms_modulation_mae_db": 0.5980520435081655, "300–1000 Hz_10ms_log_rms_mae_db": 1.6956400469199335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12739140005602537, "300–1000 Hz_10ms_envelope_correlation": 0.986749324031606, "300–1000 Hz_10ms_modulation_mae_db": 0.3062937660428447, "300–1000 Hz_40ms_log_rms_mae_db": 1.1660015182717545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07196995329684215, "300–1000 Hz_40ms_envelope_correlation": 0.9957619191614017, "300–1000 Hz_40ms_modulation_mae_db": 0.408331030074177, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5483415422414972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14304172467137458, "1000–3000 Hz_10ms_envelope_correlation": 0.9842551318957015, "1000–3000 Hz_10ms_modulation_mae_db": 0.3442743522234235, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9654563132593924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06984196276428999, "1000–3000 Hz_40ms_envelope_correlation": 0.9962928225030752, "1000–3000 Hz_40ms_modulation_mae_db": 0.20401707552275142, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8699644178615709, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1971054161653978, "3000–6000 Hz_10ms_envelope_correlation": 0.9742753575954097, "3000–6000 Hz_10ms_modulation_mae_db": 0.3763813358481561, "3000–6000 Hz_40ms_log_rms_mae_db": 1.143501269999767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09346143061067877, "3000–6000 Hz_40ms_envelope_correlation": 0.9943042499473336, "3000–6000 Hz_40ms_modulation_mae_db": 0.3778030950384618, "mr_stft_db": 5.794351324616877, "band_mae_db": 0.23083989961208137, "rms_gain_error_db": -0.12699485597877935, "output_peak": 0.4766889094418523}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2618169832197665, "全帯域_10ms_rms_envelope_nrmse": 0.09332717041829915, "全帯域_10ms_envelope_correlation": 0.9904832111605653, "全帯域_10ms_modulation_mae_db": 0.37738263671897365, "全帯域_40ms_log_rms_mae_db": 0.6842060682322828, "全帯域_40ms_rms_envelope_nrmse": 0.049239717429959806, "全帯域_40ms_envelope_correlation": 0.9972965271667945, "全帯域_40ms_modulation_mae_db": 0.5657750486538297, "300–1000 Hz_10ms_log_rms_mae_db": 1.6627133271791745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14274180865639835, "300–1000 Hz_10ms_envelope_correlation": 0.9832922626249739, "300–1000 Hz_10ms_modulation_mae_db": 0.369352754803006, "300–1000 Hz_40ms_log_rms_mae_db": 1.159948340761377, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08128560924715265, "300–1000 Hz_40ms_envelope_correlation": 0.9946061020215847, "300–1000 Hz_40ms_modulation_mae_db": 0.4606015215510548, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5807565691494594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1388710676727646, "1000–3000 Hz_10ms_envelope_correlation": 0.9853546029653996, "1000–3000 Hz_10ms_modulation_mae_db": 0.4340510538732773, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9702673181442034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07961078588703709, "1000–3000 Hz_40ms_envelope_correlation": 0.9953226499241986, "1000–3000 Hz_40ms_modulation_mae_db": 0.40459006959295146, "3000–6000 Hz_10ms_log_rms_mae_db": 1.875998127243128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1788448005368341, "3000–6000 Hz_10ms_envelope_correlation": 0.977515257632434, "3000–6000 Hz_10ms_modulation_mae_db": 0.41023117013639515, "3000–6000 Hz_40ms_log_rms_mae_db": 1.098160497295103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08969750532121318, "3000–6000 Hz_40ms_envelope_correlation": 0.9941441723301864, "3000–6000 Hz_40ms_modulation_mae_db": 0.46246893149152224, "mr_stft_db": 5.790396397326194, "band_mae_db": 0.24926704974248629, "rms_gain_error_db": -0.13863016732919206, "output_peak": 0.45256420267126335}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2073743708822569, "全帯域_10ms_rms_envelope_nrmse": 0.0995786534314999, "全帯域_10ms_envelope_correlation": 0.9890249150500652, "全帯域_10ms_modulation_mae_db": 0.08406062386269025, "全帯域_40ms_log_rms_mae_db": 0.6102244788608157, "全帯域_40ms_rms_envelope_nrmse": 0.05014371016891287, "全帯域_40ms_envelope_correlation": 0.9970031925563866, "全帯域_40ms_modulation_mae_db": 0.531810090821201, "300–1000 Hz_10ms_log_rms_mae_db": 1.7468008069675507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14911572280952173, "300–1000 Hz_10ms_envelope_correlation": 0.9816208191736655, "300–1000 Hz_10ms_modulation_mae_db": 0.3035497899710816, "300–1000 Hz_40ms_log_rms_mae_db": 1.1547934737648364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08284555729881027, "300–1000 Hz_40ms_envelope_correlation": 0.9940365053532204, "300–1000 Hz_40ms_modulation_mae_db": 0.39650066312040155, "1000–3000 Hz_10ms_log_rms_mae_db": 1.609819922177558, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15126071830191018, "1000–3000 Hz_10ms_envelope_correlation": 0.9822733465113354, "1000–3000 Hz_10ms_modulation_mae_db": 0.2784652412126974, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0148009195984589, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07994182032567065, "1000–3000 Hz_40ms_envelope_correlation": 0.9948674169647845, "1000–3000 Hz_40ms_modulation_mae_db": 0.34508358086246127, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9071740273425049, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19251367452824084, "3000–6000 Hz_10ms_envelope_correlation": 0.974022904144891, "3000–6000 Hz_10ms_modulation_mae_db": 0.7456802122129501, "3000–6000 Hz_40ms_log_rms_mae_db": 1.185986154017168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0945174016340503, "3000–6000 Hz_40ms_envelope_correlation": 0.993346811499048, "3000–6000 Hz_40ms_modulation_mae_db": 0.3273049786572896, "mr_stft_db": 5.793544399979974, "band_mae_db": 0.2089254226242487, "rms_gain_error_db": -0.07985967836676891, "output_peak": 0.44688528084834667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1506590842233995, "全帯域_10ms_rms_envelope_nrmse": 0.09144011441021455, "全帯域_10ms_envelope_correlation": 0.9908468711212077, "全帯域_10ms_modulation_mae_db": 0.1537381881163251, "全帯域_40ms_log_rms_mae_db": 0.6020199787870048, "全帯域_40ms_rms_envelope_nrmse": 0.04678061765652065, "全帯域_40ms_envelope_correlation": 0.9975322396330563, "全帯域_40ms_modulation_mae_db": 0.43399829643858406, "300–1000 Hz_10ms_log_rms_mae_db": 1.6602786618587588, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1363272993344435, "300–1000 Hz_10ms_envelope_correlation": 0.9847757613335092, "300–1000 Hz_10ms_modulation_mae_db": 0.45176898648898406, "300–1000 Hz_40ms_log_rms_mae_db": 1.1868772580047473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08830331818197534, "300–1000 Hz_40ms_envelope_correlation": 0.9934709020357959, "300–1000 Hz_40ms_modulation_mae_db": 0.35988194227496945, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6772546507238881, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15256200164746161, "1000–3000 Hz_10ms_envelope_correlation": 0.9819639856605971, "1000–3000 Hz_10ms_modulation_mae_db": 0.3095024534973057, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0378166302514995, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08865779362604045, "1000–3000 Hz_40ms_envelope_correlation": 0.9936455553953342, "1000–3000 Hz_40ms_modulation_mae_db": 0.44342500683969616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9006312617471766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19416116558014865, "3000–6000 Hz_10ms_envelope_correlation": 0.9736074488766485, "3000–6000 Hz_10ms_modulation_mae_db": 0.7688145311415434, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1505845289851522, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09823826328570187, "3000–6000 Hz_40ms_envelope_correlation": 0.9934326458245373, "3000–6000 Hz_40ms_modulation_mae_db": 1.1669955997902732, "mr_stft_db": 5.817662184126282, "band_mae_db": 0.15748669074878804, "rms_gain_error_db": -0.12351715895137103, "output_peak": 0.45141544727036803}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.211737651493397, "全帯域_10ms_rms_envelope_nrmse": 0.09189657590339247, "全帯域_10ms_envelope_correlation": 0.9907786431788516, "全帯域_10ms_modulation_mae_db": 0.0992648312596656, "全帯域_40ms_log_rms_mae_db": 0.7017422901493694, "全帯域_40ms_rms_envelope_nrmse": 0.04937161737329041, "全帯域_40ms_envelope_correlation": 0.9972601745973191, "全帯域_40ms_modulation_mae_db": 0.12829984148235565, "300–1000 Hz_10ms_log_rms_mae_db": 1.7448372546202575, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13823415778271633, "300–1000 Hz_10ms_envelope_correlation": 0.9844626137825089, "300–1000 Hz_10ms_modulation_mae_db": 0.24528050962937775, "300–1000 Hz_40ms_log_rms_mae_db": 1.2014785632350267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08344291547130639, "300–1000 Hz_40ms_envelope_correlation": 0.9944061358835081, "300–1000 Hz_40ms_modulation_mae_db": 0.16271879745240903, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6130995822137941, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1456313459914745, "1000–3000 Hz_10ms_envelope_correlation": 0.9839449389979277, "1000–3000 Hz_10ms_modulation_mae_db": 0.5573043538202074, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9789203126131759, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07613704268388446, "1000–3000 Hz_40ms_envelope_correlation": 0.9959608347343245, "1000–3000 Hz_40ms_modulation_mae_db": 0.7233000041586143, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9070999315617867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1951461279252926, "3000–6000 Hz_10ms_envelope_correlation": 0.9731780023086786, "3000–6000 Hz_10ms_modulation_mae_db": 0.7766423132979151, "3000–6000 Hz_40ms_log_rms_mae_db": 1.200804488633512, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10488232229349562, "3000–6000 Hz_40ms_envelope_correlation": 0.9919889399034594, "3000–6000 Hz_40ms_modulation_mae_db": 0.5643132827578988, "mr_stft_db": 5.805218304488592, "band_mae_db": 0.2663193511450374, "rms_gain_error_db": -0.1333901311223045, "output_peak": 0.44524039824952016}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2459513701042384, "全帯域_10ms_rms_envelope_nrmse": 0.09734199375094103, "全帯域_10ms_envelope_correlation": 0.9897139018754678, "全帯域_10ms_modulation_mae_db": 0.16694277490701104, "全帯域_40ms_log_rms_mae_db": 0.6767903167128645, "全帯域_40ms_rms_envelope_nrmse": 0.049816715369112544, "全帯域_40ms_envelope_correlation": 0.9974002550973995, "全帯域_40ms_modulation_mae_db": 0.413051909523636, "300–1000 Hz_10ms_log_rms_mae_db": 1.7021273669473493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14115636325585312, "300–1000 Hz_10ms_envelope_correlation": 0.9837379212888591, "300–1000 Hz_10ms_modulation_mae_db": 0.38201652237293837, "300–1000 Hz_40ms_log_rms_mae_db": 1.1516777930115343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07827519296157649, "300–1000 Hz_40ms_envelope_correlation": 0.9951343918440159, "300–1000 Hz_40ms_modulation_mae_db": 0.48667681713104044, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5504789736053008, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15565690012546501, "1000–3000 Hz_10ms_envelope_correlation": 0.9819390957970113, "1000–3000 Hz_10ms_modulation_mae_db": 0.35523715716040133, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9175297686456071, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09231771074846178, "1000–3000 Hz_40ms_envelope_correlation": 0.9942168224140155, "1000–3000 Hz_40ms_modulation_mae_db": 0.3886126752114316, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9294740409315034, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1941651232530399, "3000–6000 Hz_10ms_envelope_correlation": 0.973458694182765, "3000–6000 Hz_10ms_modulation_mae_db": 0.6173790663448177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1929594597354503, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10426020031420617, "3000–6000 Hz_40ms_envelope_correlation": 0.9920571862303295, "3000–6000 Hz_40ms_modulation_mae_db": 0.34155498591555683, "mr_stft_db": 5.811679856203193, "band_mae_db": 0.24065503251495832, "rms_gain_error_db": -0.16988968131478882, "output_peak": 0.4417496193284896}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0103394101427705, "spectral_convergence": 0.3628879908659735, "spectral_convergence_status": "ok", "amplitude_floor": 1.016885378852613e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.546588970332748, "spectral_convergence": 0.34649700077674944, "spectral_convergence_status": "ok", "amplitude_floor": 6.009137305261044e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.865188966926184, "spectral_convergence": 0.38233293282540115, "spectral_convergence_status": "ok", "amplitude_floor": 5.349766097366943e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003378185892142299, "output_power": 0.0003487641584913587, "delta_db": 0.1384830342909462, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005186258824964237, "output_power": 0.0005054794389001199, "delta_db": -0.11150692098886007, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004467018619566432, "output_power": 0.000421076264430183, "delta_db": -0.2565700127027862, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002054215054829419, "output_power": 0.00018483919130014926, "delta_db": -0.45851847985165195, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00019049740954433386, "output_power": 0.00017373067482784137, "delta_db": -0.4001256775070935, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.3523016958893975e-05, "output_power": 4.7288204993607415e-05, "delta_db": 0.36033836542630454, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2875904151279404, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15581104386034772, "pitch": null, "output_rms_dbfs": -27.743863191566497, "output_peak": 0.4417002118673188, "reference_rms_dbfs": -27.58805214770615, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.807372449133901}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001220285373493014, "log_rms_mae_db": 1.28575061553603, "rms_envelope_nrmse": 0.09634029793985581, "envelope_correlation": 0.9898893520170052, "rms_gain_error_db": -0.1557360108024469, "reference_mean_power": 0.0017439355594579833, "output_mean_power": 0.0016825068153564259}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.14297331662316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004136999026359357, "output_power": 0.00040242472948789036, "reference_active": true, "delta_db": -0.12000757893784868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.498679838909322e-05, "output_power": 7.412802831469264e-05, "reference_active": true, "delta_db": -0.5936902012001293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.720897303077822e-05, "output_power": 1.683802299928733e-05, "reference_active": true, "delta_db": -0.09463855585488118}], "active_band_count": 3, "active_band_mae_db": 0.269445445330953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011225274665232883, "log_rms_mae_db": 0.681701041234958, "rms_envelope_nrmse": 0.052708303140534055, "envelope_correlation": 0.9969393641667018, "rms_gain_error_db": -0.15623822462705697, "reference_mean_power": 0.0017512823148994152, "output_mean_power": 0.0016893994154099782}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.129560410778096e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037542654743209525, "output_power": 0.00036586600291273, "reference_active": true, "delta_db": -0.11202923768126878}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.98989275491616e-05, "output_power": 3.535161151990721e-05, "reference_active": true, "delta_db": -0.5255200625667099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1683754663794263e-06, "output_power": 9.014979352547273e-07, "reference_active": true, "delta_db": -1.1261769268131854}], "active_band_count": 3, "active_band_mae_db": 0.5879087423537214}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.832439224759049e-05, "log_rms_mae_db": 1.6174118204433305, "rms_envelope_nrmse": 0.13638603672841085, "envelope_correlation": 0.9846412509739252, "rms_gain_error_db": -0.06890675947601049, "reference_mean_power": 0.0005656706566344367, "output_mean_power": 0.0005567663443338889}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3997327154946766e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018576385507894965, "output_power": 0.00018033106156912075, "reference_active": true, "delta_db": -0.1289067592678885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.516179020835253e-05, "output_power": 3.251140726878476e-05, "reference_active": true, "delta_db": -0.34035209891893137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.50699704718427e-06, "output_power": 8.90181499956378e-06, "reference_active": true, "delta_db": 0.7401231934981325}], "active_band_count": 3, "active_band_mae_db": 0.40312735056165083}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.746258281389908e-05, "log_rms_mae_db": 1.0697286201697125, "rms_envelope_nrmse": 0.07721273249308527, "envelope_correlation": 0.9948228535610477, "rms_gain_error_db": -0.06997428346092195, "reference_mean_power": 0.0005675319915431874, "output_mean_power": 0.0005584610894439479}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.076532458167812e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017606095149491155, "output_power": 0.00017259443366235034, "reference_active": true, "delta_db": -0.08636259418384166}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7577844172032785e-05, "output_power": 1.622758132114563e-05, "reference_active": true, "delta_db": -0.34711815938324164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.080515782562606e-07, "output_power": 5.54469841019311e-07, "reference_active": true, "delta_db": 0.3797012394764717}], "active_band_count": 3, "active_band_mae_db": 0.2710606643478517}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.679296131568703e-05, "log_rms_mae_db": 1.5803618866762958, "rms_envelope_nrmse": 0.14200017845024016, "envelope_correlation": 0.9852713982976798, "rms_gain_error_db": -0.38713894788795933, "reference_mean_power": 0.00025821561430829635, "output_mean_power": 0.00023619386111998168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9030353181622983e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.270493557085603e-05, "output_power": 8.373113018919382e-05, "reference_active": true, "delta_db": -0.4421590312701975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6852701179097796e-05, "output_power": 1.359689689238045e-05, "reference_active": true, "delta_db": -0.9322971600011605}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.066518069212071e-06, "output_power": 3.7026460295737385e-06, "reference_active": true, "delta_db": -0.4071051061078931}], "active_band_count": 3, "active_band_mae_db": 0.5938537657930837}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.121002038281381e-05, "log_rms_mae_db": 0.9875997204299539, "rms_envelope_nrmse": 0.09094795225681189, "envelope_correlation": 0.994492822376693, "rms_gain_error_db": -0.38761588769041666, "reference_mean_power": 0.0002594441947231775, "output_mean_power": 0.0002372916021778083}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7574651639406604e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.903046567125758e-05, "output_power": 7.993799158633392e-05, "reference_active": true, "delta_db": -0.4678541238027701}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.408774632836393e-06, "output_power": 6.662395549908563e-06, "reference_active": true, "delta_db": -1.0110229951270993}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6645360300144737e-07, "output_power": 2.265360690818526e-07, "reference_active": true, "delta_db": -0.7048423693354301}], "active_band_count": 3, "active_band_mae_db": 0.727906496088433}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.318455452770038e-05, "log_rms_mae_db": 1.895510911245056, "rms_envelope_nrmse": 0.18692674560125996, "envelope_correlation": 0.9755299263983275, "rms_gain_error_db": -0.27185492338985773, "reference_mean_power": 0.0001492769552865986, "output_mean_power": 0.00014021913256203366}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0711725750134131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.027413944353342e-05, "output_power": 5.966810796834622e-05, "reference_active": true, "delta_db": -0.04388751603730741}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2776930907508557e-05, "output_power": 1.0671916658866718e-05, "reference_active": true, "delta_db": -0.7818412129849593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.16053005381717e-06, "output_power": 2.03811621747366e-06, "reference_active": true, "delta_db": -0.25331367176470415}], "active_band_count": 3, "active_band_mae_db": 0.3596808002623237}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.04724555086843e-05, "log_rms_mae_db": 1.1727331128693539, "rms_envelope_nrmse": 0.10097550321260537, "envelope_correlation": 0.9928629557425152, "rms_gain_error_db": -0.2720618738549394, "reference_mean_power": 0.00014998968981604926, "output_mean_power": 0.00014088190632713293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.934560164313186e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.89522015257842e-05, "output_power": 5.7967685317297786e-05, "reference_active": true, "delta_db": -0.0731406904500358}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6946349724579084e-06, "output_power": 4.910647406632566e-06, "reference_active": true, "delta_db": -0.643271387995787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8905823102195032e-07, "output_power": 1.4486996200989646e-07, "reference_active": true, "delta_db": -1.1561724374223143}], "active_band_count": 3, "active_band_mae_db": 0.6241948386227124}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8309637521027219, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4ca76058bf8b3b7ebdf7f454d3222f3d390801252ea346b4b473808e98e868bf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9835942430054669, "standard_deviation": 0.001395820794596304, "minimum": 0.9819390957970113, "maximum": 0.9853546029653996, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5918682890068891, "standard_deviation": 0.041777600282567884, "minimum": 1.5483415422414972, "maximum": 1.6772546507238881, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.45096070062584986, "standard_deviation": 0.16168760808851101, "minimum": 0.2784652412126974, "maximum": 0.7349972274264028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14708688259283864, "standard_deviation": 0.005769877874727084, "minimum": 0.1388710676727646, "maximum": 0.15565690012546501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950987381880145, "standard_deviation": 0.0009505968794048776, "minimum": 0.9936455553953342, "maximum": 0.9962928225030752, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9745854846304669, "standard_deviation": 0.040866171433958424, "minimum": 0.9175297686456071, "maximum": 1.0378166302514995, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5193293540933608, "standard_deviation": 0.24189856421275172, "minimum": 0.20401707552275142, "maximum": 0.9176999244705474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08205437473737442, "standard_deviation": 0.007856580303835703, "minimum": 0.06984196276428999, "maximum": 0.09231771074846178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.973767277403765, "standard_deviation": 0.002539944367704176, "minimum": 0.9685506280909659, "maximum": 0.977515257632434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.903110433246437, "standard_deviation": 0.02365034647109927, "minimum": 1.8699644178615709, "maximum": 1.9390307480387705, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5937260916563416, "standard_deviation": 0.18272008835183912, "minimum": 0.3596808002623237, "maximum": 0.7766423132979151, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19374820513152013, "standard_deviation": 0.009146477986679566, "minimum": 0.1788448005368341, "maximum": 0.2111225874619472, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9929647542160537, "standard_deviation": 0.0010177705437298965, "minimum": 0.9915810722510199, "maximum": 0.9943042499473336, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.166823106723768, "standard_deviation": 0.034418913201307605, "minimum": 1.098160497295103, "maximum": 1.200804488633512, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5218397174492388, "standard_deviation": 0.28484727593929476, "minimum": 0.3100820273201962, "maximum": 1.1669955997902732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09924559871947292, "standard_deviation": 0.006362401977047574, "minimum": 0.08969750532121318, "maximum": 0.10793216308383215, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.984323698820621, "standard_deviation": 0.0015067838995760179, "minimum": 0.9816208191736655, "maximum": 0.986749324031606, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6876524733809837, "standard_deviation": 0.04409532850638942, "minimum": 1.6174118204433305, "maximum": 1.7468008069675507, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.36421711459453654, "standard_deviation": 0.07413405183765615, "minimum": 0.24528050962937775, "maximum": 0.45234723688640904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13809114665915867, "standard_deviation": 0.006495251636056745, "minimum": 0.12739140005602537, "maximum": 0.14911572280952173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946903941738165, "standard_deviation": 0.0007288210307475274, "minimum": 0.9934709020357959, "maximum": 0.9957619191614017, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.159928401581789, "standard_deviation": 0.0406388074930811, "minimum": 1.0697286201697125, "maximum": 1.2014785632350267, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3666554162327462, "standard_deviation": 0.10492603496119889, "minimum": 0.16271879745240903, "maximum": 0.48667681713104044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07960191133072259, "standard_deviation": 0.00543818073004875, "minimum": 0.07196995329684215, "maximum": 0.08830331818197534, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22568333677807204, "standard_deviation": 0.046260238032375264, "minimum": 0.15748669074878804, "maximum": 0.2875904151279404, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.803450454093131, "standard_deviation": 0.009672991719082464, "minimum": 5.790396397326194, "maximum": 5.817662184126282, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4538628349994005, "standard_deviation": 0.014034967407745909, "minimum": 0.4417002118673188, "maximum": 0.4766889094418523, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1346029818435475, "standard_deviation": 0.026965150523388166, "minimum": -0.16988968131478882, "maximum": -0.07985967836676891, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9900446566437571, "standard_deviation": 0.0007772810104949489, "minimum": 0.9889238398005796, "maximum": 0.9908468711212077, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2331531366681836, "standard_deviation": 0.04666082707699062, "minimum": 1.1506590842233995, "maximum": 1.2892139705083256, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2422474215684553, "standard_deviation": 0.14136179808917435, "minimum": 0.08406062386269025, "maximum": 0.48332367136475535, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09533351358401049, "standard_deviation": 0.003629728680897074, "minimum": 0.09144011441021455, "maximum": 0.10061451138193943, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971735885212842, "standard_deviation": 0.00024567415617943313, "minimum": 0.9968055054433101, "maximum": 0.9975322396330563, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6683355478361225, "standard_deviation": 0.039706342391959026, "minimum": 0.6020199787870048, "maximum": 0.7055239776518886, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47125697323547205, "standard_deviation": 0.15412093628155907, "minimum": 0.12829984148235565, "maximum": 0.5980520435081655, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05020260539932814, "standard_deviation": 0.00195103999974823, "minimum": 0.04678061765652065, "maximum": 0.0527574359939911, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75946-0029", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.88688063621521, "predicted_mos": 1.88688063621521, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f4e7ef6c840acd8089dc606054720f08bd651568c554f8fa7b34e1a626b818b8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8633117000572383, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2571978987959964, "全帯域_10ms_rms_envelope_nrmse": 0.09618106813024908, "全帯域_10ms_envelope_correlation": 0.9899588723105477, "全帯域_10ms_modulation_mae_db": 0.5566478966907765, "全帯域_40ms_log_rms_mae_db": 0.6449476347616254, "全帯域_40ms_rms_envelope_nrmse": 0.05169149546178845, "全帯域_40ms_envelope_correlation": 0.9970994746206104, "全帯域_40ms_modulation_mae_db": 0.2952892396472046, "300–1000 Hz_10ms_log_rms_mae_db": 1.5862103122505709, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14076669604232056, "300–1000 Hz_10ms_envelope_correlation": 0.9840362426327038, "300–1000 Hz_10ms_modulation_mae_db": 0.7209555838603277, "300–1000 Hz_40ms_log_rms_mae_db": 1.0489561451017813, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08591506380102004, "300–1000 Hz_40ms_envelope_correlation": 0.9942910878804206, "300–1000 Hz_40ms_modulation_mae_db": 0.5775488238706848, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5335720348878221, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.153265258618596, "1000–3000 Hz_10ms_envelope_correlation": 0.9822786364226009, "1000–3000 Hz_10ms_modulation_mae_db": 0.7396999352534538, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8654696502812911, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07458309380976175, "1000–3000 Hz_40ms_envelope_correlation": 0.9964094051596143, "1000–3000 Hz_40ms_modulation_mae_db": 0.5687682099810473, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7621124307052838, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17510779837148216, "3000–6000 Hz_10ms_envelope_correlation": 0.9784784379466629, "3000–6000 Hz_10ms_modulation_mae_db": 0.420530475314507, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0298604057058807, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09432894327260859, "3000–6000 Hz_40ms_envelope_correlation": 0.9934185375128337, "3000–6000 Hz_40ms_modulation_mae_db": 0.4989605421180557, "mr_stft_db": 5.7540001155023965, "band_mae_db": 0.22954873187903016, "rms_gain_error_db": -0.16213620047333746, "output_peak": 0.4570570751927278}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2720455889607418, "全帯域_10ms_rms_envelope_nrmse": 0.09145217515847145, "全帯域_10ms_envelope_correlation": 0.9908273860318333, "全帯域_10ms_modulation_mae_db": 0.28810675015206233, "全帯域_40ms_log_rms_mae_db": 0.6894853953086403, "全帯域_40ms_rms_envelope_nrmse": 0.044314933122754877, "全帯域_40ms_envelope_correlation": 0.9978118982244446, "全帯域_40ms_modulation_mae_db": 0.27645873263536846, "300–1000 Hz_10ms_log_rms_mae_db": 1.5534844690974599, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13186282679828865, "300–1000 Hz_10ms_envelope_correlation": 0.985649089819033, "300–1000 Hz_10ms_modulation_mae_db": 0.38653773133149, "300–1000 Hz_40ms_log_rms_mae_db": 0.9819047467672007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07006528555615905, "300–1000 Hz_40ms_envelope_correlation": 0.9958182090960278, "300–1000 Hz_40ms_modulation_mae_db": 0.2879108346367994, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5108664374787262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1486507064857885, "1000–3000 Hz_10ms_envelope_correlation": 0.9829110929064944, "1000–3000 Hz_10ms_modulation_mae_db": 0.3060726136832849, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8941731783388711, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07555759634000785, "1000–3000 Hz_40ms_envelope_correlation": 0.9955035962807408, "1000–3000 Hz_40ms_modulation_mae_db": 0.4770716873332603, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7885472047566024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17212322953411321, "3000–6000 Hz_10ms_envelope_correlation": 0.9792094480279075, "3000–6000 Hz_10ms_modulation_mae_db": 0.8339262541606018, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0585656833159502, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08183181536326889, "3000–6000 Hz_40ms_envelope_correlation": 0.9951981943649842, "3000–6000 Hz_40ms_modulation_mae_db": 1.1224304033212886, "mr_stft_db": 5.757853358855545, "band_mae_db": 0.15086414970841278, "rms_gain_error_db": -0.12349082411404226, "output_peak": 0.49401545434789496}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2506000878670338, "全帯域_10ms_rms_envelope_nrmse": 0.09231189021585492, "全帯域_10ms_envelope_correlation": 0.9908775768223855, "全帯域_10ms_modulation_mae_db": 0.3245987288739338, "全帯域_40ms_log_rms_mae_db": 0.5886084828377127, "全帯域_40ms_rms_envelope_nrmse": 0.05289894185539894, "全帯域_40ms_envelope_correlation": 0.9970903789930123, "全帯域_40ms_modulation_mae_db": 0.27393512316458335, "300–1000 Hz_10ms_log_rms_mae_db": 1.532299396869887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15436014341160317, "300–1000 Hz_10ms_envelope_correlation": 0.9805566490768326, "300–1000 Hz_10ms_modulation_mae_db": 0.4202964933190035, "300–1000 Hz_40ms_log_rms_mae_db": 0.966552009869646, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10282557105070414, "300–1000 Hz_40ms_envelope_correlation": 0.9913172960473913, "300–1000 Hz_40ms_modulation_mae_db": 0.3885058723597645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.534108476185219, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15324495496574167, "1000–3000 Hz_10ms_envelope_correlation": 0.982681570300351, "1000–3000 Hz_10ms_modulation_mae_db": 0.6426159028156143, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8666055407404246, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08669288697598869, "1000–3000 Hz_40ms_envelope_correlation": 0.9952879375813282, "1000–3000 Hz_40ms_modulation_mae_db": 0.6773596986705842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.839468617512253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19685044025602527, "3000–6000 Hz_10ms_envelope_correlation": 0.9727041086646118, "3000–6000 Hz_10ms_modulation_mae_db": 1.017332181472365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0815421905549953, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0924888893643631, "3000–6000 Hz_40ms_envelope_correlation": 0.9938289060903979, "3000–6000 Hz_40ms_modulation_mae_db": 0.8811032680303207, "mr_stft_db": 5.758660552393607, "band_mae_db": 0.2313262350476519, "rms_gain_error_db": -0.1897398013818147, "output_peak": 0.45245112070022797}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1785519181489932, "全帯域_10ms_rms_envelope_nrmse": 0.09547506424799858, "全帯域_10ms_envelope_correlation": 0.9901329265892576, "全帯域_10ms_modulation_mae_db": 0.3495854798675702, "全帯域_40ms_log_rms_mae_db": 0.6031106468458362, "全帯域_40ms_rms_envelope_nrmse": 0.04822108626220261, "全帯域_40ms_envelope_correlation": 0.9975940533763978, "全帯域_40ms_modulation_mae_db": 0.5069748809682547, "300–1000 Hz_10ms_log_rms_mae_db": 1.6401791669045238, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14789668579196005, "300–1000 Hz_10ms_envelope_correlation": 0.9819728450208073, "300–1000 Hz_10ms_modulation_mae_db": 0.5157149384788605, "300–1000 Hz_40ms_log_rms_mae_db": 1.0734393005280147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08428971594779484, "300–1000 Hz_40ms_envelope_correlation": 0.9940678213629972, "300–1000 Hz_40ms_modulation_mae_db": 0.5041051709843956, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5371530127405797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14553754260824073, "1000–3000 Hz_10ms_envelope_correlation": 0.9839277270677247, "1000–3000 Hz_10ms_modulation_mae_db": 0.5162233336065031, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9396335049903718, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07518830982548982, "1000–3000 Hz_40ms_envelope_correlation": 0.9960318037512996, "1000–3000 Hz_40ms_modulation_mae_db": 0.4933823586090254, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8389190550301195, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17679917193014996, "3000–6000 Hz_10ms_envelope_correlation": 0.9781173044262834, "3000–6000 Hz_10ms_modulation_mae_db": 0.6135646774103684, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0638444081198057, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0838392267543113, "3000–6000 Hz_40ms_envelope_correlation": 0.9951572764760305, "3000–6000 Hz_40ms_modulation_mae_db": 0.6041875337288664, "mr_stft_db": 5.761448397705402, "band_mae_db": 0.2008825475859307, "rms_gain_error_db": -0.17281889924211427, "output_peak": 0.4486230777590959}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2427257300954697, "全帯域_10ms_rms_envelope_nrmse": 0.09015210302128153, "全帯域_10ms_envelope_correlation": 0.9911391926976895, "全帯域_10ms_modulation_mae_db": 0.3259756048391416, "全帯域_40ms_log_rms_mae_db": 0.7077507534770004, "全帯域_40ms_rms_envelope_nrmse": 0.04709471672500105, "全帯域_40ms_envelope_correlation": 0.9975844556457829, "全帯域_40ms_modulation_mae_db": 0.43584963071084104, "300–1000 Hz_10ms_log_rms_mae_db": 1.6129634181115524, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15836735452316802, "300–1000 Hz_10ms_envelope_correlation": 0.979643060625391, "300–1000 Hz_10ms_modulation_mae_db": 0.3917217692440412, "300–1000 Hz_40ms_log_rms_mae_db": 1.0719746979075921, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10529026895236118, "300–1000 Hz_40ms_envelope_correlation": 0.9911771522789006, "300–1000 Hz_40ms_modulation_mae_db": 0.6167763131513178, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4858487485896703, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13963894479025646, "1000–3000 Hz_10ms_envelope_correlation": 0.9849405883349454, "1000–3000 Hz_10ms_modulation_mae_db": 0.24203749523571813, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8493392093973618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0765412249668569, "1000–3000 Hz_40ms_envelope_correlation": 0.9951776295406978, "1000–3000 Hz_40ms_modulation_mae_db": 0.21863564544518702, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7867696068606658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2000446176659504, "3000–6000 Hz_10ms_envelope_correlation": 0.9720576873072315, "3000–6000 Hz_10ms_modulation_mae_db": 0.7332209129533003, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0455179973721638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09892488114732137, "3000–6000 Hz_40ms_envelope_correlation": 0.993618029040162, "3000–6000 Hz_40ms_modulation_mae_db": 0.8041874438191329, "mr_stft_db": 5.753851814893703, "band_mae_db": 0.23940245507510496, "rms_gain_error_db": -0.14116536354876005, "output_peak": 0.44334282557980303}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.162388175034996, "全帯域_10ms_rms_envelope_nrmse": 0.08814513948660348, "全帯域_10ms_envelope_correlation": 0.9914179700832585, "全帯域_10ms_modulation_mae_db": 0.269299984946829, "全帯域_40ms_log_rms_mae_db": 0.6004441894850766, "全帯域_40ms_rms_envelope_nrmse": 0.04534086393351668, "全帯域_40ms_envelope_correlation": 0.9975731298939468, "全帯域_40ms_modulation_mae_db": 0.4123370322900021, "300–1000 Hz_10ms_log_rms_mae_db": 1.6340143208340883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.126066848241356, "300–1000 Hz_10ms_envelope_correlation": 0.9869602221281759, "300–1000 Hz_10ms_modulation_mae_db": 0.7570326793005244, "300–1000 Hz_40ms_log_rms_mae_db": 1.060317394959729, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07264594227474193, "300–1000 Hz_40ms_envelope_correlation": 0.995507751199055, "300–1000 Hz_40ms_modulation_mae_db": 0.9108810983219552, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5183132453055599, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15274295507604388, "1000–3000 Hz_10ms_envelope_correlation": 0.9819926448514433, "1000–3000 Hz_10ms_modulation_mae_db": 0.4770402274637748, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8720050148497821, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06857862633574087, "1000–3000 Hz_40ms_envelope_correlation": 0.9964734948808603, "1000–3000 Hz_40ms_modulation_mae_db": 0.4842483337107466, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7774038111625128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17848589600008608, "3000–6000 Hz_10ms_envelope_correlation": 0.9776827881114986, "3000–6000 Hz_10ms_modulation_mae_db": 0.35153434104490966, "3000–6000 Hz_40ms_log_rms_mae_db": 1.002557991840239, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08500295800838278, "3000–6000 Hz_40ms_envelope_correlation": 0.9946218768441623, "3000–6000 Hz_40ms_modulation_mae_db": 0.7245112888648518, "mr_stft_db": 5.758650893677642, "band_mae_db": 0.22193855804313412, "rms_gain_error_db": -0.08051882224673124, "output_peak": 0.4395494031368296}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2532625126308257, "全帯域_10ms_rms_envelope_nrmse": 0.09339987358650446, "全帯域_10ms_envelope_correlation": 0.9904958263992029, "全帯域_10ms_modulation_mae_db": 0.2974772877815588, "全帯域_40ms_log_rms_mae_db": 0.6316933878524024, "全帯域_40ms_rms_envelope_nrmse": 0.046410874757724895, "全帯域_40ms_envelope_correlation": 0.9977297510894176, "全帯域_40ms_modulation_mae_db": 0.20521247950534938, "300–1000 Hz_10ms_log_rms_mae_db": 1.5761837905493143, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14229789373802754, "300–1000 Hz_10ms_envelope_correlation": 0.9838828259697505, "300–1000 Hz_10ms_modulation_mae_db": 0.5802128474074911, "300–1000 Hz_40ms_log_rms_mae_db": 1.0128594147955587, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08914541443889792, "300–1000 Hz_40ms_envelope_correlation": 0.994104229326251, "300–1000 Hz_40ms_modulation_mae_db": 0.3506299173810388, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5224047860892331, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1408529670411615, "1000–3000 Hz_10ms_envelope_correlation": 0.9846389393101064, "1000–3000 Hz_10ms_modulation_mae_db": 0.5009895726376347, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8558724388510192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0683712109074746, "1000–3000 Hz_40ms_envelope_correlation": 0.996156619719534, "1000–3000 Hz_40ms_modulation_mae_db": 0.46182004476802385, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8998588817364253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2107727342609945, "3000–6000 Hz_10ms_envelope_correlation": 0.9689251211403582, "3000–6000 Hz_10ms_modulation_mae_db": 0.789917679123516, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0952703186727435, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09557481594358018, "3000–6000 Hz_40ms_envelope_correlation": 0.9931724098033001, "3000–6000 Hz_40ms_modulation_mae_db": 0.35074334282935293, "mr_stft_db": 5.765743427835685, "band_mae_db": 0.22530664377215118, "rms_gain_error_db": -0.14997637774972641, "output_peak": 0.44211030595194223}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2264354488282494, "全帯域_10ms_rms_envelope_nrmse": 0.0929780057713586, "全帯域_10ms_envelope_correlation": 0.9905102754116838, "全帯域_10ms_modulation_mae_db": 0.19434964019985487, "全帯域_40ms_log_rms_mae_db": 0.6316830972431406, "全帯域_40ms_rms_envelope_nrmse": 0.04289580810234563, "全帯域_40ms_envelope_correlation": 0.9979402665369047, "全帯域_40ms_modulation_mae_db": 0.3379224865522765, "300–1000 Hz_10ms_log_rms_mae_db": 1.5818578388775844, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13842703813037152, "300–1000 Hz_10ms_envelope_correlation": 0.9842052776813153, "300–1000 Hz_10ms_modulation_mae_db": 0.19105844022422971, "300–1000 Hz_40ms_log_rms_mae_db": 0.958325328557097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06948981205626412, "300–1000 Hz_40ms_envelope_correlation": 0.9959486331860142, "300–1000 Hz_40ms_modulation_mae_db": 0.29052719731093907, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6066958603199133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1601435623883094, "1000–3000 Hz_10ms_envelope_correlation": 0.9801721584250664, "1000–3000 Hz_10ms_modulation_mae_db": 0.5371452269911726, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9000656433099165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07811787568107062, "1000–3000 Hz_40ms_envelope_correlation": 0.995296615419001, "1000–3000 Hz_40ms_modulation_mae_db": 0.4953293521468504, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8050360199985152, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1774613967599374, "3000–6000 Hz_10ms_envelope_correlation": 0.9781841288262001, "3000–6000 Hz_10ms_modulation_mae_db": 0.4110426147047514, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0745914355352268, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09224075331872437, "3000–6000 Hz_40ms_envelope_correlation": 0.9937428371630775, "3000–6000 Hz_40ms_modulation_mae_db": 0.08581196787973448, "mr_stft_db": 5.762224667223535, "band_mae_db": 0.22292615063068852, "rms_gain_error_db": -0.11211722180861529, "output_peak": 0.4807124190036757}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.976919063265204, "spectral_convergence": 0.3614649962759089, "spectral_convergence_status": "ok", "amplitude_floor": 1.016885378852613e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.484543600137932, "spectral_convergence": 0.3381310120403975, "spectral_convergence_status": "ok", "amplitude_floor": 6.009137305261044e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.800537683104053, "spectral_convergence": 0.37422666449703157, "spectral_convergence_status": "ok", "amplitude_floor": 5.349766097366943e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003378185892142299, "output_power": 0.00035312035200878535, "delta_db": 0.19239204839279325, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005186258824964237, "output_power": 0.0005090957032525831, "delta_db": -0.08054755210632156, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004467018619566432, "output_power": 0.00040134753610648443, "delta_db": -0.4649716155098938, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002054215054829419, "output_power": 0.00019054989542607932, "delta_db": -0.3263719296380406, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00019049740954433386, "output_power": 0.00018230023359926616, "delta_db": -0.1910184919185105, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.3523016958893975e-05, "output_power": 4.231749164104826e-05, "delta_db": -0.12199075370862143, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22954873187903016, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16213620047333746, "pitch": null, "output_rms_dbfs": -27.75018834817949, "output_peak": 0.4570570751927278, "reference_rms_dbfs": -27.58805214770615, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7540001155023965}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001220285373493014, "log_rms_mae_db": 1.2571978987959964, "rms_envelope_nrmse": 0.09618106813024908, "envelope_correlation": 0.9899588723105477, "rms_gain_error_db": -0.16224644153680529, "reference_mean_power": 0.0017439355594579833, "output_mean_power": 0.0016799864891070657}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.14297331662316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004136999026359357, "output_power": 0.000395981071595856, "reference_active": true, "delta_db": -0.19010992097263912}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.498679838909322e-05, "output_power": 7.501452354505882e-05, "reference_active": true, "delta_db": -0.5420611367706544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.720897303077822e-05, "output_power": 2.135660509596812e-05, "reference_active": true, "delta_db": 0.937772632329036}], "active_band_count": 3, "active_band_mae_db": 0.5566478966907765}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011225274665232883, "log_rms_mae_db": 0.6449476347616254, "rms_envelope_nrmse": 0.05169149546178845, "envelope_correlation": 0.9970994746206104, "rms_gain_error_db": -0.16225849189962677, "reference_mean_power": 0.0017512823148994152, "output_mean_power": 0.0016870591623484236}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.129560410778096e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037542654743209525, "output_power": 0.0003596618750642799, "reference_active": true, "delta_db": -0.18630575409969757}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.98989275491616e-05, "output_power": 3.5056558679494856e-05, "reference_active": true, "delta_db": -0.5619194094609937}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1683754663794263e-06, "output_power": 1.1319263593324303e-06, "reference_active": true, "delta_db": -0.13764255538092252}], "active_band_count": 3, "active_band_mae_db": 0.2952892396472046}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.832439224759049e-05, "log_rms_mae_db": 1.5862103122505709, "rms_envelope_nrmse": 0.14076669604232056, "envelope_correlation": 0.9840362426327038, "rms_gain_error_db": -0.2679787998351687, "reference_mean_power": 0.0005656706566344367, "output_mean_power": 0.0005318213534353914}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3997327154946766e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018576385507894965, "output_power": 0.00017146818405929093, "reference_active": true, "delta_db": -0.34777666753908376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.516179020835253e-05, "output_power": 2.7736509378770954e-05, "reference_active": true, "delta_db": -1.0301917389387423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.50699704718427e-06, "output_power": 8.99406619424566e-06, "reference_active": true, "delta_db": 0.7848983451031567}], "active_band_count": 3, "active_band_mae_db": 0.7209555838603277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.746258281389908e-05, "log_rms_mae_db": 1.0489561451017813, "rms_envelope_nrmse": 0.08591506380102004, "envelope_correlation": 0.9942910878804206, "rms_gain_error_db": -0.26924904885412654, "reference_mean_power": 0.0005675319915431874, "output_mean_power": 0.0005334152683468791}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.076532458167812e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017606095149491155, "output_power": 0.0001622951819146949, "reference_active": true, "delta_db": -0.3535741756400493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7577844172032785e-05, "output_power": 1.3903689992286076e-05, "reference_active": true, "delta_db": -1.0183553428545324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.080515782562606e-07, "output_power": 5.520513803716574e-07, "reference_active": true, "delta_db": 0.360716953117473}], "active_band_count": 3, "active_band_mae_db": 0.5775488238706848}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.679296131568703e-05, "log_rms_mae_db": 1.5335720348878221, "rms_envelope_nrmse": 0.153265258618596, "envelope_correlation": 0.9822786364226009, "rms_gain_error_db": -0.3283455263804247, "reference_mean_power": 0.00025821561430829635, "output_mean_power": 0.00023941312084811917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9030353181622983e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.270493557085603e-05, "output_power": 8.360180931782276e-05, "reference_active": true, "delta_db": -0.4488717983056052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6852701179097796e-05, "output_power": 1.3245274217885016e-05, "reference_active": true, "delta_db": -1.046085662746881}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.066518069212071e-06, "output_power": 4.804375469567865e-06, "reference_active": true, "delta_db": 0.7241423447078752}], "active_band_count": 3, "active_band_mae_db": 0.7396999352534538}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.121002038281381e-05, "log_rms_mae_db": 0.8654696502812911, "rms_envelope_nrmse": 0.07458309380976175, "envelope_correlation": 0.9964094051596143, "rms_gain_error_db": -0.3282456295855497, "reference_mean_power": 0.0002594441947231775, "output_mean_power": 0.0002405577729564771}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7574651639406604e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.903046567125758e-05, "output_power": 7.981378393602063e-05, "reference_active": true, "delta_db": -0.47460743941814043}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.408774632836393e-06, "output_power": 6.8502095607191505e-06, "reference_active": true, "delta_db": -0.8902885534795444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6645360300144737e-07, "output_power": 2.882454501746745e-07, "reference_active": true, "delta_db": 0.341408637045457}], "active_band_count": 3, "active_band_mae_db": 0.5687682099810473}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.318455452770038e-05, "log_rms_mae_db": 1.7621124307052838, "rms_envelope_nrmse": 0.17510779837148216, "envelope_correlation": 0.9784784379466629, "rms_gain_error_db": -0.11883378311045302, "reference_mean_power": 0.0001492769552865986, "output_mean_power": 0.00014524774216664928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0711725750134131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.027413944353342e-05, "output_power": 5.718964358125984e-05, "reference_active": true, "delta_db": -0.22813628574681913}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2776930907508557e-05, "output_power": 1.1074802711983778e-05, "reference_active": true, "delta_db": -0.6209054792623452}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.16053005381717e-06, "output_power": 2.3758295407447675e-06, "reference_active": true, "delta_db": 0.41254966093435674}], "active_band_count": 3, "active_band_mae_db": 0.420530475314507}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.04724555086843e-05, "log_rms_mae_db": 1.0298604057058807, "rms_envelope_nrmse": 0.09432894327260859, "envelope_correlation": 0.9934185375128337, "rms_gain_error_db": -0.11854195929066495, "reference_mean_power": 0.00014998968981604926, "output_mean_power": 0.00014595104574110569}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.934560164313186e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.89522015257842e-05, "output_power": 5.572939070217993e-05, "reference_active": true, "delta_db": -0.24415733260700462}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6946349724579084e-06, "output_power": 5.194726942968098e-06, "reference_active": true, "delta_db": -0.399031668006124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8905823102195032e-07, "output_power": 1.553196228811886e-07, "reference_active": true, "delta_db": -0.8536926257410384}], "active_band_count": 3, "active_band_mae_db": 0.4989605421180557}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.824441156214327, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b457fe808a4682a06ef9624bacd84d8b01d4888fae36439b6c2b9a8a144259dc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9829429197023416, "standard_deviation": 0.0015550717530996665, "minimum": 0.9801721584250664, "maximum": 0.9849405883349454, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5311203251995904, "standard_deviation": 0.034770086257241574, "minimum": 1.4858487485896703, "maximum": 1.6066958603199133, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4952280384608945, "standard_deviation": 0.1621008335394651, "minimum": 0.24203749523571813, "maximum": 0.7396999352534538, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14925961149676725, "standard_deviation": 0.006969071489037067, "minimum": 0.13963894479025646, "maximum": 0.1601435623883094, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957921377916344, "standard_deviation": 0.0005340131355299751, "minimum": 0.9951776295406978, "maximum": 0.9964734948808603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8803955225948799, "standard_deviation": 0.029604773058342422, "minimum": 0.8493392093973618, "maximum": 0.9396335049903718, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.48457691633309063, "standard_deviation": 0.12858882253984577, "minimum": 0.21863564544518702, "maximum": 0.6773596986705842, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07545385310529888, "standard_deviation": 0.005767734606959832, "minimum": 0.0683712109074746, "maximum": 0.08669288697598869, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9756698780563442, "standard_deviation": 0.0038566302806222847, "minimum": 0.9689251211403582, "maximum": 0.9792094480279075, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8122644534702972, "standard_deviation": 0.04491031268617481, "minimum": 1.7621124307052838, "maximum": 1.8998588817364253, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6463836420230399, "standard_deviation": 0.237645001898986, "minimum": 0.35153434104490966, "maximum": 1.017332181472365, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18595566059734236, "standard_deviation": 0.014411729004529564, "minimum": 0.17212322953411321, "maximum": 0.2107727342609945, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940947584118685, "standard_deviation": 0.0007886849775790217, "minimum": 0.9931724098033001, "maximum": 0.9951981943649842, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0564688038896257, "standard_deviation": 0.029894483868381602, "minimum": 1.002557991840239, "maximum": 1.0952703186727435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6339919738239504, "standard_deviation": 0.3244003773143323, "minimum": 0.08581196787973448, "maximum": 1.1224304033212886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09052903539657006, "standard_deviation": 0.006188620203514125, "minimum": 0.08183181536326889, "maximum": 0.09892488114732137, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9833632766192512, "standard_deviation": 0.002485535434774607, "minimum": 0.979643060625391, "maximum": 0.9869602221281759, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5896490891868726, "standard_deviation": 0.03763704568854974, "minimum": 1.532299396869887, "maximum": 1.6401791669045238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.495441310395746, "standard_deviation": 0.18814788000088511, "minimum": 0.19105844022422971, "maximum": 0.7570326793005244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14250568583463696, "standard_deviation": 0.010852202137309854, "minimum": 0.126066848241356, "maximum": 0.15836735452316802, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9940290225471322, "standard_deviation": 0.0018759834076537672, "minimum": 0.9911771522789006, "maximum": 0.9959486331860142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0217911298108275, "standard_deviation": 0.04804477464304478, "minimum": 0.958325328557097, "maximum": 1.0734393005280147, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4908606535021119, "standard_deviation": 0.21092499234137457, "minimum": 0.2879108346367994, "maximum": 0.9108810983219552, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0849583842597429, "standard_deviation": 0.01396011060331351, "minimum": 0.06948981205626412, "maximum": 0.10529026895236118, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21527443396776302, "standard_deviation": 0.028291828614493298, "minimum": 0.15086414970841278, "maximum": 0.23940245507510496, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.759054153510939, "standard_deviation": 0.004051420013376183, "minimum": 5.753851814893703, "maximum": 5.765743427835685, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4572327102090247, "standard_deviation": 0.019768671329067674, "minimum": 0.4395494031368296, "maximum": 0.49401545434789496, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1414954388206427, "standard_deviation": 0.035278782701823455, "minimum": -0.1897398013818147, "maximum": -0.08051882224673124, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9906700032932323, "standard_deviation": 0.0004924368501494383, "minimum": 0.9899588723105477, "maximum": 0.9914179700832585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2304009200452883, "standard_deviation": 0.0393966889502605, "minimum": 1.162388175034996, "maximum": 1.2720455889607418, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3257551716689659, "standard_deviation": 0.10457558044368376, "minimum": 0.19434964019985487, "maximum": 0.5566478966907765, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09251191495229028, "standard_deviation": 0.0026470778666004445, "minimum": 0.08814513948660348, "maximum": 0.09618106813024908, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975529260475646, "standard_deviation": 0.00030961452553080734, "minimum": 0.9970903789930123, "maximum": 0.9979402665369047, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6372154484764293, "standard_deviation": 0.04260476742688588, "minimum": 0.5886084828377127, "maximum": 0.7077507534770004, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.34299745068423504, "standard_deviation": 0.10057096682839413, "minimum": 0.20521247950534938, "maximum": 0.5069748809682547, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04735859002759164, "standard_deviation": 0.0034730132952365865, "minimum": 0.04289580810234563, "maximum": 0.05289894185539894, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7127-75947-0011", "language": "en", "style": "normal", "speaker": "librispeech_7127", "source_utterance_id": "7127-75947-0011", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 57920, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7127-75947-0011.wav", "source_file_sha256": "10f8b8ed3cb1e91e5e1d1be369eab7ab08c6ef4433791d8a772fd744c3a082cb", "decoded_pcm_sha256": "aebe57a65af629b724d8ca28eb599d83587cc213e6e303682e713453ab9de80a", "original_source_sha256": "e1bbd2d0c4d8800ab59fe4f4a1f027601dc96c787ad8bf50fd97e051270ee72b", "corpus_entry_id": "librispeech_7127-75947-0011"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.551081895828247, "predicted_mos": 3.551081895828247, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "58c9722bc309337ffb7846c1a829a7be5b0377c076fa205290110ce9beca5085", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5533459059988672, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7127-75947-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "58c9722bc309337ffb7846c1a829a7be5b0377c076fa205290110ce9beca5085", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0a958123a4ddc0a331da3b5e89c6e5b2f141b1c7b4279ba5e6848c7973c3be80"}, "rows": [{"case": "librispeech_7127-75947-0011", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.0063230991363525, "predicted_mos": 2.0063230991363525, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "f4b3d64b6fc8c764f82816d47f1873745d0456906ae75d962bc6612d9107a030", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2617003970008227, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7127-75947-0011", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f4b3d64b6fc8c764f82816d47f1873745d0456906ae75d962bc6612d9107a030", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1425516379704386, "全帯域_10ms_rms_envelope_nrmse": 0.10504957589362922, "全帯域_10ms_envelope_correlation": 0.9903931163103559, "全帯域_10ms_modulation_mae_db": 0.778179357180696, "全帯域_40ms_log_rms_mae_db": 0.6356380353389879, "全帯域_40ms_rms_envelope_nrmse": 0.06550022743329359, "全帯域_40ms_envelope_correlation": 0.9966258975216492, "全帯域_40ms_modulation_mae_db": 0.8914916447164095, "300–1000 Hz_10ms_log_rms_mae_db": 1.3808445562215055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1650047236866238, "300–1000 Hz_10ms_envelope_correlation": 0.9786976590120443, "300–1000 Hz_10ms_modulation_mae_db": 1.1090685138237915, "300–1000 Hz_40ms_log_rms_mae_db": 0.8542266937630744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09820913712836714, "300–1000 Hz_40ms_envelope_correlation": 0.9927804131453972, "300–1000 Hz_40ms_modulation_mae_db": 0.3603785483643868, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6181363123130417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1601299614401602, "1000–3000 Hz_10ms_envelope_correlation": 0.9811447029761511, "1000–3000 Hz_10ms_modulation_mae_db": 1.070363836953318, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9555809880844722, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08741959268710972, "1000–3000 Hz_40ms_envelope_correlation": 0.9956547845088468, "1000–3000 Hz_40ms_modulation_mae_db": 1.1046104392437655, "3000–6000 Hz_10ms_log_rms_mae_db": 1.871146713358335, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1988203683703296, "3000–6000 Hz_10ms_envelope_correlation": 0.9707287941279131, "3000–6000 Hz_10ms_modulation_mae_db": 0.7589311083556187, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1216638359602886, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11722938518291978, "3000–6000 Hz_40ms_envelope_correlation": 0.9896069648740043, "3000–6000 Hz_40ms_modulation_mae_db": 0.9479106496344402, "mr_stft_db": 5.893675823889201, "band_mae_db": 0.4717744821835114, "rms_gain_error_db": -0.26810514224247306, "output_peak": 0.49012875191297084}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2273672360253263, "全帯域_10ms_rms_envelope_nrmse": 0.12802519672526097, "全帯域_10ms_envelope_correlation": 0.984991435079685, "全帯域_10ms_modulation_mae_db": 2.1174370291886624, "全帯域_40ms_log_rms_mae_db": 0.7442334868035211, "全帯域_40ms_rms_envelope_nrmse": 0.06396898546081577, "全帯域_40ms_envelope_correlation": 0.9960795040915529, "全帯域_40ms_modulation_mae_db": 2.4381302380132595, "300–1000 Hz_10ms_log_rms_mae_db": 1.369936930414512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14407969533801718, "300–1000 Hz_10ms_envelope_correlation": 0.9835988187823868, "300–1000 Hz_10ms_modulation_mae_db": 0.9854758171894508, "300–1000 Hz_40ms_log_rms_mae_db": 0.8823816722715904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08084149049896552, "300–1000 Hz_40ms_envelope_correlation": 0.9947437896696567, "300–1000 Hz_40ms_modulation_mae_db": 0.596970054522114, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6977476575464143, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1811704197971343, "1000–3000 Hz_10ms_envelope_correlation": 0.9746360376556736, "1000–3000 Hz_10ms_modulation_mae_db": 1.1055101636930076, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0603407396610531, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09567403164483969, "1000–3000 Hz_40ms_envelope_correlation": 0.9930845512471231, "1000–3000 Hz_40ms_modulation_mae_db": 1.1415911301634332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.859058622683157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20174358482455512, "3000–6000 Hz_10ms_envelope_correlation": 0.9697237778778552, "3000–6000 Hz_10ms_modulation_mae_db": 0.9437021235797838, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0605546344908565, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10676173760109049, "3000–6000 Hz_40ms_envelope_correlation": 0.9911351860091289, "3000–6000 Hz_40ms_modulation_mae_db": 1.028491796674612, "mr_stft_db": 5.8748772898845685, "band_mae_db": 0.350498134286137, "rms_gain_error_db": -0.0720043920660684, "output_peak": 0.46106291347488165}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.151531234661461, "全帯域_10ms_rms_envelope_nrmse": 0.1062246845834087, "全帯域_10ms_envelope_correlation": 0.9896731418791884, "全帯域_10ms_modulation_mae_db": 0.9826663520308664, "全帯域_40ms_log_rms_mae_db": 0.7336742777182046, "全帯域_40ms_rms_envelope_nrmse": 0.05855539562097802, "全帯域_40ms_envelope_correlation": 0.9967000196108191, "全帯域_40ms_modulation_mae_db": 0.8100428296661426, "300–1000 Hz_10ms_log_rms_mae_db": 1.514038992139336, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18634280901795094, "300–1000 Hz_10ms_envelope_correlation": 0.9727311508231469, "300–1000 Hz_10ms_modulation_mae_db": 2.417285403257901, "300–1000 Hz_40ms_log_rms_mae_db": 0.9717065990061555, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12060263798536522, "300–1000 Hz_40ms_envelope_correlation": 0.9880355518862901, "300–1000 Hz_40ms_modulation_mae_db": 1.602783063128421, "1000–3000 Hz_10ms_log_rms_mae_db": 1.526159743380276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1573933383889069, "1000–3000 Hz_10ms_envelope_correlation": 0.9808829712984115, "1000–3000 Hz_10ms_modulation_mae_db": 0.4471091482146705, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9443854561572502, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08550057600831365, "1000–3000 Hz_40ms_envelope_correlation": 0.9942545064806932, "1000–3000 Hz_40ms_modulation_mae_db": 0.5222227039246489, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9345970352407398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19413503152340647, "3000–6000 Hz_10ms_envelope_correlation": 0.9719405190797583, "3000–6000 Hz_10ms_modulation_mae_db": 0.8006925764018477, "3000–6000 Hz_40ms_log_rms_mae_db": 1.122821413157327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10720812024295318, "3000–6000 Hz_40ms_envelope_correlation": 0.9912415602723375, "3000–6000 Hz_40ms_modulation_mae_db": 0.8439921398621183, "mr_stft_db": 5.856679448698412, "band_mae_db": 0.32416716803305506, "rms_gain_error_db": -0.02629137083390276, "output_peak": 0.5011734486866285}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1256226947117023, "全帯域_10ms_rms_envelope_nrmse": 0.11592455742316385, "全帯域_10ms_envelope_correlation": 0.9876655706312847, "全帯域_10ms_modulation_mae_db": 0.4957950466164471, "全帯域_40ms_log_rms_mae_db": 0.577715967312868, "全帯域_40ms_rms_envelope_nrmse": 0.06142686631844267, "全帯域_40ms_envelope_correlation": 0.9964208136885083, "全帯域_40ms_modulation_mae_db": 0.8520947159504687, "300–1000 Hz_10ms_log_rms_mae_db": 1.4960054967287926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15267985232679135, "300–1000 Hz_10ms_envelope_correlation": 0.9815543124296431, "300–1000 Hz_10ms_modulation_mae_db": 1.126003870220166, "300–1000 Hz_40ms_log_rms_mae_db": 0.9105324722893051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07946398199779314, "300–1000 Hz_40ms_envelope_correlation": 0.994933536065298, "300–1000 Hz_40ms_modulation_mae_db": 0.6271830574236293, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6676931247432751, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19194790933861028, "1000–3000 Hz_10ms_envelope_correlation": 0.9726311236079042, "1000–3000 Hz_10ms_modulation_mae_db": 0.7480303329245045, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0409309958938295, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13304587728286693, "1000–3000 Hz_40ms_envelope_correlation": 0.9864126449017614, "1000–3000 Hz_40ms_modulation_mae_db": 0.5607089721812811, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9159339681863201, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21013385632907874, "3000–6000 Hz_10ms_envelope_correlation": 0.9675146620522375, "3000–6000 Hz_10ms_modulation_mae_db": 0.8520733300696519, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1300170965177232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12647095872740907, "3000–6000 Hz_40ms_envelope_correlation": 0.9886701907790674, "3000–6000 Hz_40ms_modulation_mae_db": 0.6609316649499807, "mr_stft_db": 5.86577991242081, "band_mae_db": 0.33890381126256197, "rms_gain_error_db": -0.08628510576852731, "output_peak": 0.5174065015043324}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1559302022787328, "全帯域_10ms_rms_envelope_nrmse": 0.10232327246989524, "全帯域_10ms_envelope_correlation": 0.9904811271647341, "全帯域_10ms_modulation_mae_db": 1.20550009894024, "全帯域_40ms_log_rms_mae_db": 0.6402795586082001, "全帯域_40ms_rms_envelope_nrmse": 0.05724966044406118, "全帯域_40ms_envelope_correlation": 0.9968612602773343, "全帯域_40ms_modulation_mae_db": 1.1798873182671974, "300–1000 Hz_10ms_log_rms_mae_db": 1.5104601877055361, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1587756282045244, "300–1000 Hz_10ms_envelope_correlation": 0.980266918518895, "300–1000 Hz_10ms_modulation_mae_db": 1.9414925275513728, "300–1000 Hz_40ms_log_rms_mae_db": 0.8782368794141437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08717987021302635, "300–1000 Hz_40ms_envelope_correlation": 0.9937851928465751, "300–1000 Hz_40ms_modulation_mae_db": 1.3437875747282997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6221163329768775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17537523124363955, "1000–3000 Hz_10ms_envelope_correlation": 0.9769214007273196, "1000–3000 Hz_10ms_modulation_mae_db": 0.958999828929436, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0152663008932732, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1068994665773686, "1000–3000 Hz_40ms_envelope_correlation": 0.9911302811667128, "1000–3000 Hz_40ms_modulation_mae_db": 0.7943806694298846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8662983438763427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1897879616449309, "3000–6000 Hz_10ms_envelope_correlation": 0.9742116264931528, "3000–6000 Hz_10ms_modulation_mae_db": 0.8278086384663014, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1800696796975882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12277357764621201, "3000–6000 Hz_40ms_envelope_correlation": 0.9898911311619585, "3000–6000 Hz_40ms_modulation_mae_db": 0.8666301329551503, "mr_stft_db": 5.8647387595468885, "band_mae_db": 0.3832318299488235, "rms_gain_error_db": 0.02947932644143884, "output_peak": 0.48554911318587407}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1545373655075162, "全帯域_10ms_rms_envelope_nrmse": 0.11721942158187929, "全帯域_10ms_envelope_correlation": 0.9873873347641882, "全帯域_10ms_modulation_mae_db": 0.3822467443642439, "全帯域_40ms_log_rms_mae_db": 0.6836563014988609, "全帯域_40ms_rms_envelope_nrmse": 0.06443519433598437, "全帯域_40ms_envelope_correlation": 0.9961258851430325, "全帯域_40ms_modulation_mae_db": 0.4074636958247699, "300–1000 Hz_10ms_log_rms_mae_db": 1.5869866687244716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17250515432370744, "300–1000 Hz_10ms_envelope_correlation": 0.9768990414661534, "300–1000 Hz_10ms_modulation_mae_db": 1.889338430557709, "300–1000 Hz_40ms_log_rms_mae_db": 1.057685887189995, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11207065355197922, "300–1000 Hz_40ms_envelope_correlation": 0.9897608081429273, "300–1000 Hz_40ms_modulation_mae_db": 1.3770444715063732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6213150494119837, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1772919832091837, "1000–3000 Hz_10ms_envelope_correlation": 0.975745271056744, "1000–3000 Hz_10ms_modulation_mae_db": 0.8614101357481214, "1000–3000 Hz_40ms_log_rms_mae_db": 1.028510363359371, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0951847471699463, "1000–3000 Hz_40ms_envelope_correlation": 0.993329521326934, "1000–3000 Hz_40ms_modulation_mae_db": 0.534356629479522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8390267639600746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19043497453704955, "3000–6000 Hz_10ms_envelope_correlation": 0.9731089069446991, "3000–6000 Hz_10ms_modulation_mae_db": 0.6506355607458479, "3000–6000 Hz_40ms_log_rms_mae_db": 1.088930433488991, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09961422262396208, "3000–6000 Hz_40ms_envelope_correlation": 0.9926993978494599, "3000–6000 Hz_40ms_modulation_mae_db": 0.7969025946269506, "mr_stft_db": 5.864238161488179, "band_mae_db": 0.3028063090811693, "rms_gain_error_db": -0.12589654573056638, "output_peak": 0.4751830093268758}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2179835314294767, "全帯域_10ms_rms_envelope_nrmse": 0.12407261425115934, "全帯域_10ms_envelope_correlation": 0.9858735471171883, "全帯域_10ms_modulation_mae_db": 0.827933059545007, "全帯域_40ms_log_rms_mae_db": 0.6785318734551349, "全帯域_40ms_rms_envelope_nrmse": 0.06975968998273523, "全帯域_40ms_envelope_correlation": 0.9955233948922295, "全帯域_40ms_modulation_mae_db": 1.113569270783245, "300–1000 Hz_10ms_log_rms_mae_db": 1.5383744314140442, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15566222249845874, "300–1000 Hz_10ms_envelope_correlation": 0.9808187217316671, "300–1000 Hz_10ms_modulation_mae_db": 1.3516084212120874, "300–1000 Hz_40ms_log_rms_mae_db": 1.022775037063001, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0889468901261836, "300–1000 Hz_40ms_envelope_correlation": 0.9934979059774554, "300–1000 Hz_40ms_modulation_mae_db": 0.8465745375494835, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6147076305091177, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16413907417851242, "1000–3000 Hz_10ms_envelope_correlation": 0.9790666900167007, "1000–3000 Hz_10ms_modulation_mae_db": 0.6347987744494173, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9780564186056594, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08485438527259308, "1000–3000 Hz_40ms_envelope_correlation": 0.9943860660185742, "1000–3000 Hz_40ms_modulation_mae_db": 0.7086519804041748, "3000–6000 Hz_10ms_log_rms_mae_db": 1.897292622857706, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2112792488541249, "3000–6000 Hz_10ms_envelope_correlation": 0.9685624215555851, "3000–6000 Hz_10ms_modulation_mae_db": 0.8537098439861222, "3000–6000 Hz_40ms_log_rms_mae_db": 1.115924934526816, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12474559715557529, "3000–6000 Hz_40ms_envelope_correlation": 0.988714793120292, "3000–6000 Hz_40ms_modulation_mae_db": 0.6400594107615033, "mr_stft_db": 5.875498365301662, "band_mae_db": 0.30275163317041975, "rms_gain_error_db": -0.16327571814631942, "output_peak": 0.38951463248324275}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1691321016576217, "全帯域_10ms_rms_envelope_nrmse": 0.12282995187555949, "全帯域_10ms_envelope_correlation": 0.9867801877285396, "全帯域_10ms_modulation_mae_db": 0.4620260472027778, "全帯域_40ms_log_rms_mae_db": 0.7837664945549193, "全帯域_40ms_rms_envelope_nrmse": 0.07343426326619011, "全帯域_40ms_envelope_correlation": 0.9958233530452004, "全帯域_40ms_modulation_mae_db": 0.6541694287162968, "300–1000 Hz_10ms_log_rms_mae_db": 1.466835122094508, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15781987547608542, "300–1000 Hz_10ms_envelope_correlation": 0.9806692316958034, "300–1000 Hz_10ms_modulation_mae_db": 1.641244303391021, "300–1000 Hz_40ms_log_rms_mae_db": 0.9509098744347084, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09863747186794687, "300–1000 Hz_40ms_envelope_correlation": 0.9927650969993592, "300–1000 Hz_40ms_modulation_mae_db": 0.8965724273658257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.628914509772897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19114912425028543, "1000–3000 Hz_10ms_envelope_correlation": 0.9750975679368431, "1000–3000 Hz_10ms_modulation_mae_db": 1.1051540966869544, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9897546412910166, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1257273501003278, "1000–3000 Hz_40ms_envelope_correlation": 0.9924316059114252, "1000–3000 Hz_40ms_modulation_mae_db": 0.9210689782162472, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8796271651922627, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19561637255622658, "3000–6000 Hz_10ms_envelope_correlation": 0.971450629766315, "3000–6000 Hz_10ms_modulation_mae_db": 0.9402704682466304, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1221618248592047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11185939059609343, "3000–6000 Hz_40ms_envelope_correlation": 0.9902222630260872, "3000–6000 Hz_40ms_modulation_mae_db": 0.587947466322257, "mr_stft_db": 5.878975851923783, "band_mae_db": 0.5009776138481082, "rms_gain_error_db": -0.3123063774655326, "output_peak": 0.45897434549706345}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 57920, "output_samples": 57920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.161591197319182, "spectral_convergence": 0.402623695250189, "spectral_convergence_status": "ok", "amplitude_floor": 1.9579592332350965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.575583397715234, "spectral_convergence": 0.38526651522276567, "spectral_convergence_status": "ok", "amplitude_floor": 1.206925486771671e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.943852876633188, "spectral_convergence": 0.4150991731174834, "spectral_convergence_status": "ok", "amplitude_floor": 6.395923898397892e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042562388767747, "output_power": 0.0004003586566039608, "delta_db": -0.26576770962895924, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001322032073723835, "output_power": 0.0012978795203060136, "delta_db": -0.08007612061781921, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006256947892797981, "output_power": 0.0005467382545208097, "delta_db": -0.5858307629398811, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002997481206139804, "output_power": 0.0002742804858738867, "delta_db": -0.38561558678756613, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003961082609243847, "output_power": 0.0003634118435947291, "delta_db": -0.37414822942010567, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.769256628656105e-06, "output_power": 1.0099510275067949e-05, "delta_db": 1.139208483706737, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4717744821835114, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.26810514224247306, "pitch": null, "output_rms_dbfs": -25.386863505086655, "output_peak": 0.49012875191297084, "reference_rms_dbfs": -25.11875836284418, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.893675823889201}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 57920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.00020370994921487513, "log_rms_mae_db": 1.1425516379704386, "rms_envelope_nrmse": 0.10504957589362922, "envelope_correlation": 0.9903931163103559, "rms_gain_error_db": -0.2681053251697937, "reference_mean_power": 0.003081231895334411, "output_mean_power": 0.002896768920573158}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 1.3803128421607645e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003568424263155193, "output_power": 0.0003388262348704594, "reference_active": true, "delta_db": -0.22499453838605532}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014048200587458577, "output_power": 9.93225454390697e-05, "reference_active": true, "delta_db": -1.5057285855910119}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7025084230927585e-05, "output_power": 1.956459057857752e-05, "reference_active": true, "delta_db": 0.6038149475650207}], "active_band_count": 3, "active_band_mae_db": 0.778179357180696}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.00013932647455664127, "log_rms_mae_db": 0.6356380353389879, "rms_envelope_nrmse": 0.06550022743329359, "envelope_correlation": 0.9966258975216492, "rms_gain_error_db": -0.26810698862923926, "reference_mean_power": 0.0031070091895842884, "output_mean_power": 0.0029210018962254696}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 1.2870957120852427e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031478645736461515, "output_power": 0.0002977634571088019, "reference_active": true, "delta_db": -0.24144641935771025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.928842911180188e-05, "output_power": 4.903569250356234e-05, "reference_active": true, "delta_db": -1.501484022634723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4070650648317675e-06, "output_power": 1.1354283431849943e-06, "reference_active": true, "delta_db": -0.9315444921567955}], "active_band_count": 3, "active_band_mae_db": 0.8914916447164095}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.0001235612100218032, "log_rms_mae_db": 1.3808445562215055, "rms_envelope_nrmse": 0.1650047236866238, "envelope_correlation": 0.9786976590120443, "rms_gain_error_db": -0.29212436417039167, "reference_mean_power": 0.0011039001556015875, "output_mean_power": 0.0010320894961495723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 4.707425731708226e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018855725502559516, "output_power": 0.00014946829088729156, "reference_active": true, "delta_db": -1.0089417864220234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8855557263471545e-05, "output_power": 5.941563014062834e-05, "reference_active": true, "delta_db": 0.0411323135522727}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.265121532076037e-06, "output_power": 7.205175353787902e-06, "reference_active": true, "delta_db": 2.2771314414970782}], "active_band_count": 3, "active_band_mae_db": 1.1090685138237915}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.0001123986305082509, "log_rms_mae_db": 0.8542266937630744, "rms_envelope_nrmse": 0.09820913712836714, "envelope_correlation": 0.9927804131453972, "rms_gain_error_db": -0.29212495489577595, "reference_mean_power": 0.0011131375427710934, "output_mean_power": 0.0010407258333708894}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 4.401493945076042e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017713633289459933, "output_power": 0.0001423286988778817, "reference_active": true, "delta_db": -0.9501517040304244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7898955892530673e-05, "output_power": 2.8137964422589175e-05, "reference_active": true, "delta_db": 0.037047259055620146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0644420897956173e-07, "output_power": 3.9774732899857516e-07, "reference_active": true, "delta_db": -0.09393668200711593}], "active_band_count": 3, "active_band_mae_db": 0.3603785483643868}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 8.851421864498589e-05, "log_rms_mae_db": 1.6181363123130417, "rms_envelope_nrmse": 0.1601299614401602, "envelope_correlation": 0.9811447029761511, "rms_gain_error_db": -0.4530788885245241, "reference_mean_power": 0.000506505585091631, "output_mean_power": 0.00045632719312458606}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 2.2793822474038483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010569673324300439, "output_power": 8.864271461219374e-05, "reference_active": true, "delta_db": -0.7641851731540288}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5368696588833574e-05, "output_power": 1.6455754112672933e-05, "reference_active": true, "delta_db": 0.296807525908598}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.905327294990813e-06, "output_power": 6.407185797439468e-06, "reference_active": true, "delta_db": 2.150098811797327}], "active_band_count": 3, "active_band_mae_db": 1.070363836953318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 8.08613786301264e-05, "log_rms_mae_db": 0.9555809880844722, "rms_envelope_nrmse": 0.08741959268710972, "envelope_correlation": 0.9956547845088468, "rms_gain_error_db": -0.45308355936228867, "reference_mean_power": 0.0005107405079369301, "output_mean_power": 0.00046014207661431784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 2.1269882857538688e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.701418164823122e-05, "output_power": 8.393145224313569e-05, "reference_active": true, "delta_db": -0.6291048693213309}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.796784971798785e-06, "output_power": 7.366295854238649e-06, "reference_active": true, "delta_db": 0.34945627065846296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1410692673718103e-07, "output_power": 3.665706887229157e-07, "reference_active": true, "delta_db": 2.3352701777515024}], "active_band_count": 3, "active_band_mae_db": 1.1046104392437655}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 5.208111658419103e-05, "log_rms_mae_db": 1.871146713358335, "rms_envelope_nrmse": 0.1988203683703296, "envelope_correlation": 0.9707287941279131, "rms_gain_error_db": -0.2709651945807448, "reference_mean_power": 0.0001411400109581116, "output_mean_power": 0.00013260308480047935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 8.582618674096986e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.483177751159981e-05, "output_power": 3.033268217560976e-05, "reference_active": true, "delta_db": -0.6006482302917702}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4244491686748372e-05, "output_power": 1.4821597820961572e-05, "reference_active": true, "delta_db": 0.1724806855976206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3188700900305924e-06, "output_power": 3.2782559624722884e-06, "reference_active": true, "delta_db": 1.5036644091774654}], "active_band_count": 3, "active_band_mae_db": 0.7589311083556187}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 4.6774501347655295e-05, "log_rms_mae_db": 1.1216638359602886, "rms_envelope_nrmse": 0.11722938518291978, "envelope_correlation": 0.9896069648740043, "rms_gain_error_db": -0.270979469250174, "reference_mean_power": 0.0001423195946220122, "output_mean_power": 0.00013371088124953642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 7.830084455082624e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.343144196598961e-05, "output_power": 3.0075817448740817e-05, "reference_active": true, "delta_db": -0.4593766887622742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.21519729141624e-06, "output_power": 8.0635497632945e-06, "reference_active": true, "delta_db": 0.48278059609666213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.774546333615459e-07, "output_power": 2.749443564514483e-07, "reference_active": true, "delta_db": 1.901574664044384}], "active_band_count": 3, "active_band_mae_db": 0.9479106496344402}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8039334609921682, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b2f3896669f8f2196667233403183bf2f8bb8c61d6e24e5043faf63ae39f7081", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9770157206594685, "standard_deviation": 0.0030798986206252064, "minimum": 0.9726311236079042, "maximum": 0.9811447029761511, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6245987950817353, "standard_deviation": 0.04938665818372988, "minimum": 1.526159743380276, "maximum": 1.6977476575464143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8664220396999287, "standard_deviation": 0.24137504403456247, "minimum": 0.4471091482146705, "maximum": 1.1055101636930076, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17482463023080408, "standard_deviation": 0.013311220637932795, "minimum": 0.1573933383889069, "maximum": 0.19194790933861028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9925854951952588, "standard_deviation": 0.0028409312659557007, "minimum": 0.9864126449017614, "maximum": 0.9956547845088468, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0016032379932407, "standard_deviation": 0.04139393212765432, "minimum": 0.9443854561572502, "maximum": 1.0603407396610531, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7859489378803697, "standard_deviation": 0.24954619345824744, "minimum": 0.5222227039246489, "maximum": 1.1415911301634332, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10178825334292071, "standard_deviation": 0.018577008213806246, "minimum": 0.08485438527259308, "maximum": 0.13304587728286693, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9709051672371896, "standard_deviation": 0.0022541832650898315, "minimum": 0.9675146620522375, "maximum": 0.9742116264931528, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8828726544193672, "standard_deviation": 0.031370382781220695, "minimum": 1.8390267639600746, "maximum": 1.9345970352407398, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8284779562314755, "standard_deviation": 0.09577109712380222, "minimum": 0.6506355607458479, "maximum": 0.9437021235797838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19899392482996275, "standard_deviation": 0.008243093246740687, "minimum": 0.1897879616449309, "maximum": 0.2112792488541249, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.990272685886542, "standard_deviation": 0.0013710208067473323, "minimum": 0.9886701907790674, "maximum": 0.9926993978494599, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1177679815873494, "standard_deviation": 0.03422599514028212, "minimum": 1.0605546344908565, "maximum": 1.1800696796975882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7966082319733765, "standard_deviation": 0.1559605153996675, "minimum": 0.587947466322257, "maximum": 1.028491796674612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11458287372202691, "standard_deviation": 0.009752173395127677, "minimum": 0.09961422262396208, "maximum": 0.12647095872740907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9794044818074674, "standard_deviation": 0.0033352634182108832, "minimum": 0.9727311508231469, "maximum": 0.9835988187823868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.482935298180338, "standard_deviation": 0.07487746275331941, "minimum": 1.369936930414512, "maximum": 1.5869866687244716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.5576896609004374, "standard_deviation": 0.5015253135886094, "minimum": 0.9854758171894508, "maximum": 2.417285403257901, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16160874510901993, "standard_deviation": 0.013020721120929887, "minimum": 0.14407969533801718, "maximum": 0.18634280901795094, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9925377868416199, "standard_deviation": 0.0024253477227381554, "minimum": 0.9880355518862901, "maximum": 0.994933536065298, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9410568894289968, "standard_deviation": 0.07287663006752294, "minimum": 0.8542266937630744, "maximum": 1.057685887189995, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9564117168235666, "standard_deviation": 0.43968790719082274, "minimum": 0.3603785483643868, "maximum": 1.602783063128421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09574401667120339, "standard_deviation": 0.014665943915018576, "minimum": 0.07946398199779314, "maximum": 0.12060263798536522, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3718888727267233, "standard_deviation": 0.07572978117944448, "minimum": 0.30275163317041975, "maximum": 0.5009776138481082, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.871807951644188, "standard_deviation": 0.011496003524908958, "minimum": 5.856679448698412, "maximum": 5.893675823889201, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4723740895089837, "standard_deviation": 0.03877997641849167, "minimum": 0.38951463248324275, "maximum": 0.5174065015043324, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12808566572649388, "standard_deviation": 0.11641493385475511, "minimum": -0.3123063774655326, "maximum": 0.02947932644143884, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9879056825843955, "standard_deviation": 0.002075697598580905, "minimum": 0.984991435079685, "maximum": 0.9904811271647341, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1680820005302845, "standard_deviation": 0.03598739441742529, "minimum": 1.1256226947117023, "maximum": 1.2273672360253263, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9064729668836176, "standard_deviation": 0.5638245593598866, "minimum": 0.3822467443642439, "maximum": 2.1174370291886624, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11520865935049451, "standard_deviation": 0.009676370893469902, "minimum": 0.10232327246989524, "maximum": 0.12802519672526097, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9962700160337907, "standard_deviation": 0.00046258340808642427, "minimum": 0.9955233948922295, "maximum": 0.9968612602773343, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.684686999411337, "standard_deviation": 0.0672116648589249, "minimum": 0.577715967312868, "maximum": 0.7837664945549193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0433561427422235, "standard_deviation": 0.6139881089701547, "minimum": 0.4074636958247699, "maximum": 2.4381302380132595, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06429128535781262, "standard_deviation": 0.00542182666986768, "minimum": 0.05724966044406118, "maximum": 0.07343426326619011, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75947-0011", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.9151132106781006, "predicted_mos": 1.9151132106781006, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "dfc7f3552b2f146abd6d9ae8de8cc61fd95245bd20e5a9d2cb65ef7a59e1062e", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.85176909994334, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1331961529731363, "全帯域_10ms_rms_envelope_nrmse": 0.10221983518479871, "全帯域_10ms_envelope_correlation": 0.9908806000181627, "全帯域_10ms_modulation_mae_db": 0.6719752646837679, "全帯域_40ms_log_rms_mae_db": 0.63033145366312, "全帯域_40ms_rms_envelope_nrmse": 0.06449369478892188, "全帯域_40ms_envelope_correlation": 0.9966735493253049, "全帯域_40ms_modulation_mae_db": 0.9985585114913248, "300–1000 Hz_10ms_log_rms_mae_db": 1.3760710719707898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15967390000389298, "300–1000 Hz_10ms_envelope_correlation": 0.9799749957192961, "300–1000 Hz_10ms_modulation_mae_db": 0.8514306014591378, "300–1000 Hz_40ms_log_rms_mae_db": 0.8454171764491258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0948489650082179, "300–1000 Hz_40ms_envelope_correlation": 0.9931170663556862, "300–1000 Hz_40ms_modulation_mae_db": 0.5074216121429189, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6111628724355291, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1586578213311444, "1000–3000 Hz_10ms_envelope_correlation": 0.9817966599238034, "1000–3000 Hz_10ms_modulation_mae_db": 0.8205209387443881, "1000–3000 Hz_40ms_log_rms_mae_db": 0.922088843911502, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08794305400886286, "1000–3000 Hz_40ms_envelope_correlation": 0.9959161988215336, "1000–3000 Hz_40ms_modulation_mae_db": 0.7722408607989276, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8482872782545139, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1962377466359519, "3000–6000 Hz_10ms_envelope_correlation": 0.9716600384820681, "3000–6000 Hz_10ms_modulation_mae_db": 0.6562823720707346, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0675146388792507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11873100153515781, "3000–6000 Hz_40ms_envelope_correlation": 0.989676400858614, "3000–6000 Hz_40ms_modulation_mae_db": 0.8559573410920773, "mr_stft_db": 5.8486441868628525, "band_mae_db": 0.4247659534653459, "rms_gain_error_db": -0.2547552171223373, "output_peak": 0.49065665211551496}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2385814724496462, "全帯域_10ms_rms_envelope_nrmse": 0.13860271631997542, "全帯域_10ms_envelope_correlation": 0.9825299084334119, "全帯域_10ms_modulation_mae_db": 2.4779357154187, "全帯域_40ms_log_rms_mae_db": 0.7581251977967163, "全帯域_40ms_rms_envelope_nrmse": 0.07785030038704042, "全帯域_40ms_envelope_correlation": 0.9941683049347476, "全帯域_40ms_modulation_mae_db": 2.889917727102995, "300–1000 Hz_10ms_log_rms_mae_db": 1.3463132767445756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14417666812381347, "300–1000 Hz_10ms_envelope_correlation": 0.9835756231007642, "300–1000 Hz_10ms_modulation_mae_db": 0.8030081936269947, "300–1000 Hz_40ms_log_rms_mae_db": 0.833706849233125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08025070185469572, "300–1000 Hz_40ms_envelope_correlation": 0.994781892610095, "300–1000 Hz_40ms_modulation_mae_db": 0.4076436310569243, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6742117483785481, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17349026843742837, "1000–3000 Hz_10ms_envelope_correlation": 0.9769392715297058, "1000–3000 Hz_10ms_modulation_mae_db": 1.1807263762512157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0017281770608704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09245248301170025, "1000–3000 Hz_40ms_envelope_correlation": 0.993851491255118, "1000–3000 Hz_40ms_modulation_mae_db": 1.1200399606430003, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8620089547906828, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19529628586032682, "3000–6000 Hz_10ms_envelope_correlation": 0.9717277288075428, "3000–6000 Hz_10ms_modulation_mae_db": 1.060510477224112, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0612816957731193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10486392114768088, "3000–6000 Hz_40ms_envelope_correlation": 0.9919125641358082, "3000–6000 Hz_40ms_modulation_mae_db": 1.0089407623574933, "mr_stft_db": 5.827129086186975, "band_mae_db": 0.33331756418299346, "rms_gain_error_db": -0.030619719732988507, "output_peak": 0.47800500258867396}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1808103910517984, "全帯域_10ms_rms_envelope_nrmse": 0.10878384522995968, "全帯域_10ms_envelope_correlation": 0.9891889745559409, "全帯域_10ms_modulation_mae_db": 0.8466065016032069, "全帯域_40ms_log_rms_mae_db": 0.7486283046263443, "全帯域_40ms_rms_envelope_nrmse": 0.05881317616524801, "全帯域_40ms_envelope_correlation": 0.9966788047756665, "全帯域_40ms_modulation_mae_db": 0.6248817234616476, "300–1000 Hz_10ms_log_rms_mae_db": 1.4950243593651709, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18454122740531279, "300–1000 Hz_10ms_envelope_correlation": 0.9733458458518831, "300–1000 Hz_10ms_modulation_mae_db": 2.444645537686036, "300–1000 Hz_40ms_log_rms_mae_db": 0.9456329172546017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11678531175814946, "300–1000 Hz_40ms_envelope_correlation": 0.9888074804170693, "300–1000 Hz_40ms_modulation_mae_db": 1.680243249082509, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5492169263376872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15804671718222316, "1000–3000 Hz_10ms_envelope_correlation": 0.9807522821002995, "1000–3000 Hz_10ms_modulation_mae_db": 0.346552950157837, "1000–3000 Hz_40ms_log_rms_mae_db": 0.94556768642035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08405987497344188, "1000–3000 Hz_40ms_envelope_correlation": 0.9945365271122695, "1000–3000 Hz_40ms_modulation_mae_db": 0.47736955255830016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8984333351536997, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20226335214353783, "3000–6000 Hz_10ms_envelope_correlation": 0.9697960927426433, "3000–6000 Hz_10ms_modulation_mae_db": 0.7944673098590592, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0491492433586191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10998226329638291, "3000–6000 Hz_40ms_envelope_correlation": 0.9914556016883119, "3000–6000 Hz_40ms_modulation_mae_db": 0.6755582846554754, "mr_stft_db": 5.814090276336052, "band_mae_db": 0.2654368946103806, "rms_gain_error_db": -0.025151878744506882, "output_peak": 0.4989066404586845}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.129640868140872, "全帯域_10ms_rms_envelope_nrmse": 0.11476562403542666, "全帯域_10ms_envelope_correlation": 0.9879283965043729, "全帯域_10ms_modulation_mae_db": 0.36204701917401244, "全帯域_40ms_log_rms_mae_db": 0.572774198008384, "全帯域_40ms_rms_envelope_nrmse": 0.06125892835722138, "全帯域_40ms_envelope_correlation": 0.9964055122561518, "全帯域_40ms_modulation_mae_db": 0.5035749509771307, "300–1000 Hz_10ms_log_rms_mae_db": 1.4767549307859424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1516990213303567, "300–1000 Hz_10ms_envelope_correlation": 0.9817779798179992, "300–1000 Hz_10ms_modulation_mae_db": 1.0305790258164433, "300–1000 Hz_40ms_log_rms_mae_db": 0.8865069640644683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0780955448191478, "300–1000 Hz_40ms_envelope_correlation": 0.9950525787537811, "300–1000 Hz_40ms_modulation_mae_db": 0.532020457565217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6670703108406257, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1919142988431011, "1000–3000 Hz_10ms_envelope_correlation": 0.9724052828219618, "1000–3000 Hz_10ms_modulation_mae_db": 0.8003420114560745, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0008311138291783, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13318007291239484, "1000–3000 Hz_40ms_envelope_correlation": 0.9862150870911605, "1000–3000 Hz_40ms_modulation_mae_db": 0.6258714832439478, "3000–6000 Hz_10ms_log_rms_mae_db": 1.837113586075189, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2141355812304126, "3000–6000 Hz_10ms_envelope_correlation": 0.966198169597284, "3000–6000 Hz_10ms_modulation_mae_db": 0.8567062807721596, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0654258206571998, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12652466287806483, "3000–6000 Hz_40ms_envelope_correlation": 0.9886676945649634, "3000–6000 Hz_40ms_modulation_mae_db": 0.6061127636812297, "mr_stft_db": 5.821881053277425, "band_mae_db": 0.2944870030083305, "rms_gain_error_db": -0.05465953247354569, "output_peak": 0.5160500643593692}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1667596331879493, "全帯域_10ms_rms_envelope_nrmse": 0.1041296659770543, "全帯域_10ms_envelope_correlation": 0.9901552387605196, "全帯域_10ms_modulation_mae_db": 0.8097869353296844, "全帯域_40ms_log_rms_mae_db": 0.6603421988377061, "全帯域_40ms_rms_envelope_nrmse": 0.06160269701343944, "全帯域_40ms_envelope_correlation": 0.9963801819640944, "全帯域_40ms_modulation_mae_db": 0.6567235205353358, "300–1000 Hz_10ms_log_rms_mae_db": 1.516668825257144, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16006435444573922, "300–1000 Hz_10ms_envelope_correlation": 0.9800960260444719, "300–1000 Hz_10ms_modulation_mae_db": 1.8545906319268877, "300–1000 Hz_40ms_log_rms_mae_db": 0.9031643300198108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09032204378595914, "300–1000 Hz_40ms_envelope_correlation": 0.993416213515798, "300–1000 Hz_40ms_modulation_mae_db": 1.1950628239428507, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6414053909932953, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1769669880191389, "1000–3000 Hz_10ms_envelope_correlation": 0.9763744368396667, "1000–3000 Hz_10ms_modulation_mae_db": 0.8537027032272286, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0372103515514943, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10741161616345209, "1000–3000 Hz_40ms_envelope_correlation": 0.9909547209925099, "1000–3000 Hz_40ms_modulation_mae_db": 0.7008191173969568, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8502404062032016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18366753635527783, "3000–6000 Hz_10ms_envelope_correlation": 0.9770941506633217, "3000–6000 Hz_10ms_modulation_mae_db": 1.0971025635482299, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1006231718812982, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11779820360854033, "3000–6000 Hz_40ms_envelope_correlation": 0.9923384417020769, "3000–6000 Hz_40ms_modulation_mae_db": 1.1564983937586246, "mr_stft_db": 5.821261598973965, "band_mae_db": 0.39226516098223313, "rms_gain_error_db": 0.04648559758590945, "output_peak": 0.48645244326131937}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1530646778715132, "全帯域_10ms_rms_envelope_nrmse": 0.12250032046918867, "全帯域_10ms_envelope_correlation": 0.9862590936383406, "全帯域_10ms_modulation_mae_db": 0.5830221661216863, "全帯域_40ms_log_rms_mae_db": 0.6882041178065678, "全帯域_40ms_rms_envelope_nrmse": 0.07465107825056647, "全帯域_40ms_envelope_correlation": 0.9948261622619915, "全帯域_40ms_modulation_mae_db": 0.9647611195206579, "300–1000 Hz_10ms_log_rms_mae_db": 1.5419272964273416, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1698571098365718, "300–1000 Hz_10ms_envelope_correlation": 0.9777748278914653, "300–1000 Hz_10ms_modulation_mae_db": 1.8998672520654418, "300–1000 Hz_40ms_log_rms_mae_db": 1.0007130475366657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10579253176742669, "300–1000 Hz_40ms_envelope_correlation": 0.9909944497707396, "300–1000 Hz_40ms_modulation_mae_db": 1.3495380807942745, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6040615167779975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1737500301296759, "1000–3000 Hz_10ms_envelope_correlation": 0.976884257400136, "1000–3000 Hz_10ms_modulation_mae_db": 0.7244254399233911, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9769520755761846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09434153149309046, "1000–3000 Hz_40ms_envelope_correlation": 0.9936469383702724, "1000–3000 Hz_40ms_modulation_mae_db": 0.3977816558304402, "3000–6000 Hz_10ms_log_rms_mae_db": 1.852468633067329, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19766980811694157, "3000–6000 Hz_10ms_envelope_correlation": 0.9712522106159623, "3000–6000 Hz_10ms_modulation_mae_db": 0.6934557364654088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0731582602434846, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10822715506832921, "3000–6000 Hz_40ms_envelope_correlation": 0.9918148959863711, "3000–6000 Hz_40ms_modulation_mae_db": 0.7840810153125743, "mr_stft_db": 5.8153886945515225, "band_mae_db": 0.3192798956713731, "rms_gain_error_db": -0.15798477022955199, "output_peak": 0.4750105275617528}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2446164218536382, "全帯域_10ms_rms_envelope_nrmse": 0.12980672674308918, "全帯域_10ms_envelope_correlation": 0.9845507974677885, "全帯域_10ms_modulation_mae_db": 1.0262282106235814, "全帯域_40ms_log_rms_mae_db": 0.6784214487188155, "全帯域_40ms_rms_envelope_nrmse": 0.07451998906069246, "全帯域_40ms_envelope_correlation": 0.9949194937577494, "全帯域_40ms_modulation_mae_db": 1.079658397214868, "300–1000 Hz_10ms_log_rms_mae_db": 1.5573707580639828, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16139557258992152, "300–1000 Hz_10ms_envelope_correlation": 0.9794049101108724, "300–1000 Hz_10ms_modulation_mae_db": 1.451270414755997, "300–1000 Hz_40ms_log_rms_mae_db": 1.0117244995427237, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09381927051158695, "300–1000 Hz_40ms_envelope_correlation": 0.9927428904351088, "300–1000 Hz_40ms_modulation_mae_db": 0.8139970866225035, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6692886911373102, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1681616199525969, "1000–3000 Hz_10ms_envelope_correlation": 0.9781318944302079, "1000–3000 Hz_10ms_modulation_mae_db": 0.5760056659421354, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9793816100494708, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08705327540222954, "1000–3000 Hz_40ms_envelope_correlation": 0.9942431134118473, "1000–3000 Hz_40ms_modulation_mae_db": 0.5852507739824151, "3000–6000 Hz_10ms_log_rms_mae_db": 1.862095461710708, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21018455829792276, "3000–6000 Hz_10ms_envelope_correlation": 0.9685286546716604, "3000–6000 Hz_10ms_modulation_mae_db": 0.8859347842796512, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1048580312916962, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1237647927963518, "3000–6000 Hz_40ms_envelope_correlation": 0.9886298564373072, "3000–6000 Hz_40ms_modulation_mae_db": 0.6168895967248705, "mr_stft_db": 5.831632751038415, "band_mae_db": 0.27488065795035593, "rms_gain_error_db": -0.18346207950196172, "output_peak": 0.3886350351335566}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1794348611946321, "全帯域_10ms_rms_envelope_nrmse": 0.11907334172353713, "全帯域_10ms_envelope_correlation": 0.9877985927639087, "全帯域_10ms_modulation_mae_db": 0.5697650827653701, "全帯域_40ms_log_rms_mae_db": 0.794042985021918, "全帯域_40ms_rms_envelope_nrmse": 0.07190173660370985, "全帯域_40ms_envelope_correlation": 0.9962353917984232, "全帯域_40ms_modulation_mae_db": 0.6249170217272078, "300–1000 Hz_10ms_log_rms_mae_db": 1.4951223042426338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.156532088936398, "300–1000 Hz_10ms_envelope_correlation": 0.9809546561731148, "300–1000 Hz_10ms_modulation_mae_db": 1.4671820769707657, "300–1000 Hz_40ms_log_rms_mae_db": 0.962552274695083, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09803775886937881, "300–1000 Hz_40ms_envelope_correlation": 0.9928204903382597, "300–1000 Hz_40ms_modulation_mae_db": 0.7251715287715741, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6376590359352077, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1931609924447684, "1000–3000 Hz_10ms_envelope_correlation": 0.9752291235344953, "1000–3000 Hz_10ms_modulation_mae_db": 1.0846183726001504, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9870993592693682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1295878109522985, "1000–3000 Hz_40ms_envelope_correlation": 0.9925002586644338, "1000–3000 Hz_40ms_modulation_mae_db": 0.981691237204808, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8081727086370654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19896222945630107, "3000–6000 Hz_10ms_envelope_correlation": 0.9704851289815135, "3000–6000 Hz_10ms_modulation_mae_db": 1.0104946622017537, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0728860121135437, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11502762810014328, "3000–6000 Hz_40ms_envelope_correlation": 0.9898214509840291, "3000–6000 Hz_40ms_modulation_mae_db": 0.8299955966906557, "mr_stft_db": 5.833796685876778, "band_mae_db": 0.5096953011156284, "rms_gain_error_db": -0.33629901004046614, "output_peak": 0.45774064171945744}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 57920, "output_samples": 57920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.110782632038451, "spectral_convergence": 0.39493891149184707, "spectral_convergence_status": "ok", "amplitude_floor": 1.9579592332350965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.532466741992767, "spectral_convergence": 0.37942193105549127, "spectral_convergence_status": "ok", "amplitude_floor": 1.206925486771671e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.902683186557341, "spectral_convergence": 0.41278439308552006, "spectral_convergence_status": "ok", "amplitude_floor": 6.395923898397892e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042562388767747, "output_power": 0.0003971665868513867, "delta_db": -0.30053289589886667, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001322032073723835, "output_power": 0.0013086300916827485, "delta_db": -0.044250890737942895, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006256947892797981, "output_power": 0.0005546074055918328, "delta_db": -0.5237687388248564, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002997481206139804, "output_power": 0.00027459963665786444, "delta_db": -0.38056510547817424, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003961082609243847, "output_power": 0.000357220011105226, "delta_db": -0.4487812016769247, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.769256628656105e-06, "output_power": 9.450377474996408e-06, "delta_db": 0.8506968881753106, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4247659534653459, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2547552171223373, "pitch": null, "output_rms_dbfs": -25.373513579966517, "output_peak": 0.49065665211551496, "reference_rms_dbfs": -25.11875836284418, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8486441868628525}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 57920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.00020370994921487513, "log_rms_mae_db": 1.1331961529731363, "rms_envelope_nrmse": 0.10221983518479871, "envelope_correlation": 0.9908806000181627, "rms_gain_error_db": -0.25475537104029783, "reference_mean_power": 0.003081231895334411, "output_mean_power": 0.0029056871159633887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 1.3803128421607645e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003568424263155193, "output_power": 0.00033950431129440027, "reference_active": true, "delta_db": -0.21631189984119548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014048200587458577, "output_power": 0.00010056420208028868, "reference_active": true, "delta_db": -1.4517728761304005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7025084230927585e-05, "output_power": 1.8444775392095564e-05, "reference_active": true, "delta_db": 0.3478410180797076}], "active_band_count": 3, "active_band_mae_db": 0.6719752646837679}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.00013932647455664127, "log_rms_mae_db": 0.63033145366312, "rms_envelope_nrmse": 0.06449369478892188, "envelope_correlation": 0.9966735493253049, "rms_gain_error_db": -0.2547564629781359, "reference_mean_power": 0.0031070091895842884, "output_mean_power": 0.002929995082529953}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 1.2870957120852427e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031478645736461515, "output_power": 0.0002979120418782848, "reference_active": true, "delta_db": -0.23927981868572173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.928842911180188e-05, "output_power": 4.921000225784714e-05, "reference_active": true, "delta_db": -1.486073302797114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4070650648317675e-06, "output_power": 1.0502240008708093e-06, "reference_active": true, "delta_db": -1.2703224129911386}], "active_band_count": 3, "active_band_mae_db": 0.9985585114913248}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.0001235612100218032, "log_rms_mae_db": 1.3760710719707898, "rms_envelope_nrmse": 0.15967390000389298, "envelope_correlation": 0.9799749957192961, "rms_gain_error_db": -0.2500127607360977, "reference_mean_power": 0.0011039001556015875, "output_mean_power": 0.0010421458861209203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 4.707425731708226e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018855725502559516, "output_power": 0.00015511590453917137, "reference_active": true, "delta_db": -0.8478691747510009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8855557263471545e-05, "output_power": 5.8551281320168884e-05, "reference_active": true, "delta_db": -0.02251072605356723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.265121532076037e-06, "output_power": 6.28525064083532e-06, "reference_active": true, "delta_db": 1.6839119035728456}], "active_band_count": 3, "active_band_mae_db": 0.8514306014591378}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.0001123986305082509, "log_rms_mae_db": 0.8454171764491258, "rms_envelope_nrmse": 0.0948489650082179, "envelope_correlation": 0.9931170663556862, "rms_gain_error_db": -0.2500132392208877, "reference_mean_power": 0.0011131375427710934, "output_mean_power": 0.001050866400543753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 4.401493945076042e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017713633289459933, "output_power": 0.00014700433137372013, "reference_active": true, "delta_db": -0.8097751849741739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7898955892530673e-05, "output_power": 2.7440120494279574e-05, "reference_active": true, "delta_db": -0.07201936176749786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0644420897956173e-07, "output_power": 3.507146528978262e-07, "reference_active": true, "delta_db": -0.640470289687085}], "active_band_count": 3, "active_band_mae_db": 0.5074216121429189}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 8.851421864498589e-05, "log_rms_mae_db": 1.6111628724355291, "rms_envelope_nrmse": 0.1586578213311444, "envelope_correlation": 0.9817966599238034, "rms_gain_error_db": -0.4893258621939891, "reference_mean_power": 0.000506505585091631, "output_mean_power": 0.00045253445635050517}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 2.2793822474038483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010569673324300439, "output_power": 8.826757428693937e-05, "reference_active": true, "delta_db": -0.7826037307639457}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5368696588833574e-05, "output_power": 1.5844940083900213e-05, "reference_active": true, "delta_db": 0.13253564514751434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.905327294990813e-06, "output_power": 5.575704907898108e-06, "reference_active": true, "delta_db": 1.5464234403217045}], "active_band_count": 3, "active_band_mae_db": 0.8205209387443881}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 8.08613786301264e-05, "log_rms_mae_db": 0.922088843911502, "rms_envelope_nrmse": 0.08794305400886286, "envelope_correlation": 0.9959161988215336, "rms_gain_error_db": -0.4893309926834628, "reference_mean_power": 0.0005107405079369301, "output_mean_power": 0.0004563175843631825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 2.1269882857538688e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.701418164823122e-05, "output_power": 8.358380952413757e-05, "reference_active": true, "delta_db": -0.6471306336236122}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.796784971798785e-06, "output_power": 6.884111763385199e-06, "reference_active": true, "delta_db": 0.05544382000257288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1410692673718103e-07, "output_power": 3.1048850812034386e-07, "reference_active": true, "delta_db": 1.6141481287705974}], "active_band_count": 3, "active_band_mae_db": 0.7722408607989276}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 5.208111658419103e-05, "log_rms_mae_db": 1.8482872782545139, "rms_envelope_nrmse": 0.1962377466359519, "envelope_correlation": 0.9716600384820681, "rms_gain_error_db": -0.3455447784682456, "reference_mean_power": 0.0001411400109581116, "output_mean_power": 0.00013034538793972673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 8.582618674096986e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.483177751159981e-05, "output_power": 2.9969783841817714e-05, "reference_active": true, "delta_db": -0.652920270528089}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4244491686748372e-05, "output_power": 1.4976796772788274e-05, "reference_active": true, "delta_db": 0.2177198058671476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3188700900305924e-06, "output_power": 2.986050325436866e-06, "reference_active": true, "delta_db": 1.098207039816967}], "active_band_count": 3, "active_band_mae_db": 0.6562823720707346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 4.6774501347655295e-05, "log_rms_mae_db": 1.0675146388792507, "rms_envelope_nrmse": 0.11873100153515781, "envelope_correlation": 0.989676400858614, "rms_gain_error_db": -0.3455569512293521, "reference_mean_power": 0.0001423195946220122, "output_mean_power": 0.00013143438668558418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 7.830084455082624e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.343144196598961e-05, "output_power": 2.9749497928021297e-05, "reference_active": true, "delta_db": -0.5067546831081366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.21519729141624e-06, "output_power": 8.102526548293448e-06, "reference_active": true, "delta_db": 0.5037225180832643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.774546333615459e-07, "output_power": 2.539959697269638e-07, "reference_active": true, "delta_db": 1.557394822084831}], "active_band_count": 3, "active_band_mae_db": 0.8559573410920773}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8099092041007993, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4c046f6702f2867e2abdcee795e939f1fb4b7bcee4277e8605d05d0696ed5b8f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9773141510725346, "standard_deviation": 0.002981731715950113, "minimum": 0.9724052828219618, "maximum": 0.9817966599238034, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.631759561604525, "standard_deviation": 0.042409585317651835, "minimum": 1.5492169263376872, "maximum": 1.6742117483785481, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7983618072878027, "standard_deviation": 0.26477568167793364, "minimum": 0.346552950157837, "maximum": 1.1807263762512157, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17426859204250966, "standard_deviation": 0.013204924896647196, "minimum": 0.15804671718222316, "maximum": 0.1931609924447684, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927330419648931, "standard_deviation": 0.003010739035093886, "minimum": 0.9862150870911605, "maximum": 0.9959161988215336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9813574022085523, "standard_deviation": 0.03538299096976544, "minimum": 0.922088843911502, "maximum": 1.0372103515514943, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7076330802073495, "standard_deviation": 0.24497798408906146, "minimum": 0.3977816558304402, "maximum": 1.1200399606430003, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10200371486468379, "standard_deviation": 0.019476092862576323, "minimum": 0.08405987497344188, "maximum": 0.13318007291239484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9708427718202496, "standard_deviation": 0.0031361797668841124, "minimum": 0.966198169597284, "maximum": 0.9770941506633217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8523525454865486, "standard_deviation": 0.025401616112485904, "minimum": 1.8081727086370654, "maximum": 1.8984333351536997, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8818692733026386, "standard_deviation": 0.16451680691181586, "minimum": 0.6562823720707346, "maximum": 1.0971025635482299, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19980213726208407, "standard_deviation": 0.00939517397215231, "minimum": 0.18366753635527783, "maximum": 0.2141355812304126, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905396132946852, "standard_deviation": 0.0015117776731655391, "minimum": 0.9886298564373072, "maximum": 0.9923384417020769, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0743621092747766, "standard_deviation": 0.01911171560797178, "minimum": 1.0491492433586191, "maximum": 1.1048580312916962, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8167542192841251, "standard_deviation": 0.19249991118574813, "minimum": 0.6061127636812297, "maximum": 1.1564983937586246, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11561495355383139, "standard_deviation": 0.007579619176514443, "minimum": 0.10486392114768088, "maximum": 0.12652466287806483, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9796131080887334, "standard_deviation": 0.0030531790206201055, "minimum": 0.9733458458518831, "maximum": 0.9835756231007642, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4756566028571978, "standard_deviation": 0.07573753003652166, "minimum": 1.3463132767445756, "maximum": 1.5573707580639828, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.475321716788463, "standard_deviation": 0.5736299041473731, "minimum": 0.8030081936269947, "maximum": 2.444645537686036, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1609924928340008, "standard_deviation": 0.012100311164381131, "minimum": 0.14417666812381347, "maximum": 0.18454122740531279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9927166327745671, "standard_deviation": 0.002022394775066554, "minimum": 0.9888074804170693, "maximum": 0.9950525787537811, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9236772573494505, "standard_deviation": 0.06729664955868335, "minimum": 0.833706849233125, "maximum": 1.0117244995427237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9013873087473465, "standard_deviation": 0.45791666594557023, "minimum": 0.4076436310569243, "maximum": 1.680243249082509, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09474401604682031, "standard_deviation": 0.01268511453746047, "minimum": 0.0780955448191478, "maximum": 0.11678531175814946, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.35176605387333015, "standard_deviation": 0.08447441278824716, "minimum": 0.2654368946103806, "maximum": 0.5096953011156284, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.826728041637998, "standard_deviation": 0.011314777692819427, "minimum": 5.814090276336052, "maximum": 5.8486441868628525, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47393212589979106, "standard_deviation": 0.03852573234554428, "minimum": 0.3886350351335566, "maximum": 0.5160500643593692, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12455582628243109, "standard_deviation": 0.13050467735009832, "minimum": -0.33629901004046614, "maximum": 0.04648559758590945, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9874114502678057, "standard_deviation": 0.0028405429945323173, "minimum": 0.9825299084334119, "maximum": 0.9908806000181627, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1782630598403983, "standard_deviation": 0.04344351416941965, "minimum": 1.129640868140872, "maximum": 1.2446164218536382, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9184208619650012, "standard_deviation": 0.6617086171249503, "minimum": 0.36204701917401244, "maximum": 2.4779357154187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11748525946037872, "standard_deviation": 0.012648853502374687, "minimum": 0.10221983518479871, "maximum": 0.13860271631997542, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9957859251342662, "standard_deviation": 0.000986530980849935, "minimum": 0.9941683049347476, "maximum": 0.9966788047756665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6913587380599465, "standard_deviation": 0.07290658538204667, "minimum": 0.572774198008384, "maximum": 0.794042985021918, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0428741215038961, "standard_deviation": 0.7755505272595359, "minimum": 0.5035749509771307, "maximum": 2.889917727102995, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06813645007835499, "standard_deviation": 0.007386373141460588, "minimum": 0.05881317616524801, "maximum": 0.07785030038704042, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75947-0011", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.1988871097564697, "predicted_mos": 2.1988871097564697, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "a3c8736a21165dbcbde3a718328684075f651676031d0042b0d9b58551f08ff3", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.849523199722171, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.082432756893315, "全帯域_10ms_rms_envelope_nrmse": 0.1256181559938269, "全帯域_10ms_envelope_correlation": 0.9859124761077388, "全帯域_10ms_modulation_mae_db": 0.765056040633617, "全帯域_40ms_log_rms_mae_db": 0.6346497718621422, "全帯域_40ms_rms_envelope_nrmse": 0.0695078544934022, "全帯域_40ms_envelope_correlation": 0.9961603740741006, "全帯域_40ms_modulation_mae_db": 0.7755787571994314, "300–1000 Hz_10ms_log_rms_mae_db": 1.5128465908392754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15511460727505097, "300–1000 Hz_10ms_envelope_correlation": 0.9813753425356774, "300–1000 Hz_10ms_modulation_mae_db": 1.1553083190086524, "300–1000 Hz_40ms_log_rms_mae_db": 0.9893923921295646, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09852823108761598, "300–1000 Hz_40ms_envelope_correlation": 0.9928880732803802, "300–1000 Hz_40ms_modulation_mae_db": 0.41245067870943536, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8891785816082167, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18625773123116252, "1000–3000 Hz_10ms_envelope_correlation": 0.9729061965038992, "1000–3000 Hz_10ms_modulation_mae_db": 0.30180851983779194, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1470917928518318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0864766221203981, "1000–3000 Hz_40ms_envelope_correlation": 0.9945500645946034, "1000–3000 Hz_40ms_modulation_mae_db": 0.5395792276851606, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2643381029428684, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21176257991892453, "3000–6000 Hz_10ms_envelope_correlation": 0.966685084242522, "3000–6000 Hz_10ms_modulation_mae_db": 0.6978453145301948, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3386582910457636, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11602866993886533, "3000–6000 Hz_40ms_envelope_correlation": 0.9893033372556315, "3000–6000 Hz_40ms_modulation_mae_db": 0.51839991916129, "mr_stft_db": 6.125948863829869, "band_mae_db": 0.513981420891532, "rms_gain_error_db": -0.2770829727360738, "output_peak": 0.42984869503408707}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1473087965833897, "全帯域_10ms_rms_envelope_nrmse": 0.12249741306616824, "全帯域_10ms_envelope_correlation": 0.9863104154614627, "全帯域_10ms_modulation_mae_db": 1.7297933166547985, "全帯域_40ms_log_rms_mae_db": 0.6366905746392959, "全帯域_40ms_rms_envelope_nrmse": 0.08392289929286699, "全帯域_40ms_envelope_correlation": 0.9932288484977969, "全帯域_40ms_modulation_mae_db": 1.7243833880748796, "300–1000 Hz_10ms_log_rms_mae_db": 1.5731401486384018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1776102431298201, "300–1000 Hz_10ms_envelope_correlation": 0.9750485062964233, "300–1000 Hz_10ms_modulation_mae_db": 2.2787764737951544, "300–1000 Hz_40ms_log_rms_mae_db": 0.9719159621402711, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10811318745430194, "300–1000 Hz_40ms_envelope_correlation": 0.9905322239567202, "300–1000 Hz_40ms_modulation_mae_db": 1.2729593623894109, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9065572836853848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17256447612378162, "1000–3000 Hz_10ms_envelope_correlation": 0.97785876281017, "1000–3000 Hz_10ms_modulation_mae_db": 0.6666378978773372, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2298950277857825, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10768900817172625, "1000–3000 Hz_40ms_envelope_correlation": 0.99249727709039, "1000–3000 Hz_40ms_modulation_mae_db": 0.5504870300495784, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3663213773497977, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22394004690223512, "3000–6000 Hz_10ms_envelope_correlation": 0.9629526124141665, "3000–6000 Hz_10ms_modulation_mae_db": 0.6834088710880285, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4380057390743928, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1219081524119751, "3000–6000 Hz_40ms_envelope_correlation": 0.9882624515344998, "3000–6000 Hz_40ms_modulation_mae_db": 0.16144949108453074, "mr_stft_db": 6.119647855930367, "band_mae_db": 0.566513175125633, "rms_gain_error_db": -0.058582147060499, "output_peak": 0.5925933193479833}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1936238467249276, "全帯域_10ms_rms_envelope_nrmse": 0.11455111898261966, "全帯域_10ms_envelope_correlation": 0.9879835239397597, "全帯域_10ms_modulation_mae_db": 0.823161044548324, "全帯域_40ms_log_rms_mae_db": 0.6551306979470356, "全帯域_40ms_rms_envelope_nrmse": 0.0591155923018976, "全帯域_40ms_envelope_correlation": 0.9968263345558676, "全帯域_40ms_modulation_mae_db": 1.264275359551852, "300–1000 Hz_10ms_log_rms_mae_db": 1.630875300332235, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15526213528893962, "300–1000 Hz_10ms_envelope_correlation": 0.9809367245661251, "300–1000 Hz_10ms_modulation_mae_db": 1.2425243984009486, "300–1000 Hz_40ms_log_rms_mae_db": 1.0383157876901157, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09143450414749488, "300–1000 Hz_40ms_envelope_correlation": 0.9930990871110933, "300–1000 Hz_40ms_modulation_mae_db": 0.8886582156634457, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9323235102362468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16742709812549225, "1000–3000 Hz_10ms_envelope_correlation": 0.9781850602219084, "1000–3000 Hz_10ms_modulation_mae_db": 0.5237410057003978, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1713864842502941, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09110323191528213, "1000–3000 Hz_40ms_envelope_correlation": 0.9935050102924997, "1000–3000 Hz_40ms_modulation_mae_db": 0.5372463259241634, "3000–6000 Hz_10ms_log_rms_mae_db": 2.293558756512615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2249302297247509, "3000–6000 Hz_10ms_envelope_correlation": 0.9621479403338854, "3000–6000 Hz_10ms_modulation_mae_db": 1.082842414764975, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4047059512052364, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12327060303573799, "3000–6000 Hz_40ms_envelope_correlation": 0.9881218251210858, "3000–6000 Hz_40ms_modulation_mae_db": 1.3984964900685342, "mr_stft_db": 6.127122116471036, "band_mae_db": 0.3205199961212036, "rms_gain_error_db": -0.14635079160246717, "output_peak": 0.4485624666676407}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2060088911393647, "全帯域_10ms_rms_envelope_nrmse": 0.11579532208947618, "全帯域_10ms_envelope_correlation": 0.9876472714189775, "全帯域_10ms_modulation_mae_db": 0.8264273255152513, "全帯域_40ms_log_rms_mae_db": 0.706860033057191, "全帯域_40ms_rms_envelope_nrmse": 0.06171565207282902, "全帯域_40ms_envelope_correlation": 0.9964262802401309, "全帯域_40ms_modulation_mae_db": 1.4091749944563239, "300–1000 Hz_10ms_log_rms_mae_db": 1.5717378598971892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15437660854455285, "300–1000 Hz_10ms_envelope_correlation": 0.9811601924311335, "300–1000 Hz_10ms_modulation_mae_db": 1.339917959727094, "300–1000 Hz_40ms_log_rms_mae_db": 0.9494705019842429, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08165573034690773, "300–1000 Hz_40ms_envelope_correlation": 0.9945118906715266, "300–1000 Hz_40ms_modulation_mae_db": 0.7378558705624775, "1000–3000 Hz_10ms_log_rms_mae_db": 1.992464812028144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19591830079374198, "1000–3000 Hz_10ms_envelope_correlation": 0.9703825537615681, "1000–3000 Hz_10ms_modulation_mae_db": 0.6469708902982984, "1000–3000 Hz_40ms_log_rms_mae_db": 1.27168745530864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11523779958008282, "1000–3000 Hz_40ms_envelope_correlation": 0.989320310114476, "1000–3000 Hz_40ms_modulation_mae_db": 0.28609871474625986, "3000–6000 Hz_10ms_log_rms_mae_db": 2.359996450360551, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22874690124447847, "3000–6000 Hz_10ms_envelope_correlation": 0.9607587919445317, "3000–6000 Hz_10ms_modulation_mae_db": 0.6021967557566458, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3999161876999675, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12703261222547124, "3000–6000 Hz_40ms_envelope_correlation": 0.9873922477342161, "3000–6000 Hz_40ms_modulation_mae_db": 1.019032926796033, "mr_stft_db": 6.131023688093976, "band_mae_db": 0.39572824158409414, "rms_gain_error_db": -0.11685213384531228, "output_peak": 0.4315206447831632}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.127798478601474, "全帯域_10ms_rms_envelope_nrmse": 0.13103939858892524, "全帯域_10ms_envelope_correlation": 0.984427492784972, "全帯域_10ms_modulation_mae_db": 1.5347970326902083, "全帯域_40ms_log_rms_mae_db": 0.677874727022762, "全帯域_40ms_rms_envelope_nrmse": 0.07168086860044383, "全帯域_40ms_envelope_correlation": 0.9950866517970564, "全帯域_40ms_modulation_mae_db": 1.6169647277822579, "300–1000 Hz_10ms_log_rms_mae_db": 1.4921933288148648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1625543331922658, "300–1000 Hz_10ms_envelope_correlation": 0.9790785604254726, "300–1000 Hz_10ms_modulation_mae_db": 1.083741739275675, "300–1000 Hz_40ms_log_rms_mae_db": 0.911545906294677, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08724518556494563, "300–1000 Hz_40ms_envelope_correlation": 0.9937289887646953, "300–1000 Hz_40ms_modulation_mae_db": 0.5783494739804209, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8879695397341336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15764227182363177, "1000–3000 Hz_10ms_envelope_correlation": 0.9808004822616252, "1000–3000 Hz_10ms_modulation_mae_db": 0.7224870618656096, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1437030087926305, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09038385453120859, "1000–3000 Hz_40ms_envelope_correlation": 0.9934039794014674, "1000–3000 Hz_40ms_modulation_mae_db": 0.8075487848749955, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2814331127597365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21719722220267795, "3000–6000 Hz_10ms_envelope_correlation": 0.9653576977621665, "3000–6000 Hz_10ms_modulation_mae_db": 0.5358599045793516, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4079490164831745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12113616027503123, "3000–6000 Hz_40ms_envelope_correlation": 0.9884770273824031, "3000–6000 Hz_40ms_modulation_mae_db": 0.47724086582371905, "mr_stft_db": 6.133038515302295, "band_mae_db": 0.3697220215225516, "rms_gain_error_db": 0.03143540553846187, "output_peak": 0.42274244409585765}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0925464019708266, "全帯域_10ms_rms_envelope_nrmse": 0.10452315558297559, "全帯域_10ms_envelope_correlation": 0.9900004843711269, "全帯域_10ms_modulation_mae_db": 0.7172943057811801, "全帯域_40ms_log_rms_mae_db": 0.619224360347337, "全帯域_40ms_rms_envelope_nrmse": 0.0675546032989667, "全帯域_40ms_envelope_correlation": 0.9957261045834381, "全帯域_40ms_modulation_mae_db": 1.0091578719837149, "300–1000 Hz_10ms_log_rms_mae_db": 1.6150696861937504, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14905824916614072, "300–1000 Hz_10ms_envelope_correlation": 0.9824467146236654, "300–1000 Hz_10ms_modulation_mae_db": 1.5185010742072205, "300–1000 Hz_40ms_log_rms_mae_db": 0.9910313693967039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0950813108961901, "300–1000 Hz_40ms_envelope_correlation": 0.9927039751729543, "300–1000 Hz_40ms_modulation_mae_db": 1.1314538850787088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9242109660763984, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1954079476452698, "1000–3000 Hz_10ms_envelope_correlation": 0.9717354434600887, "1000–3000 Hz_10ms_modulation_mae_db": 0.9341602558179437, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1562368323853587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10941132057326335, "1000–3000 Hz_40ms_envelope_correlation": 0.9910257853908929, "1000–3000 Hz_40ms_modulation_mae_db": 1.1233667452178138, "3000–6000 Hz_10ms_log_rms_mae_db": 2.310358245152179, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21685870696375298, "3000–6000 Hz_10ms_envelope_correlation": 0.9648769651401348, "3000–6000 Hz_10ms_modulation_mae_db": 0.839877962556567, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4030269919606209, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12781584583960232, "3000–6000 Hz_40ms_envelope_correlation": 0.9870418616618114, "3000–6000 Hz_40ms_modulation_mae_db": 0.7460525722581854, "mr_stft_db": 6.129866244712834, "band_mae_db": 0.43656618547416065, "rms_gain_error_db": -0.13028190982462995, "output_peak": 0.4966063564249457}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1589672139580947, "全帯域_10ms_rms_envelope_nrmse": 0.13231152334038226, "全帯域_10ms_envelope_correlation": 0.9840064704583433, "全帯域_10ms_modulation_mae_db": 0.527798035179046, "全帯域_40ms_log_rms_mae_db": 0.6344139859842914, "全帯域_40ms_rms_envelope_nrmse": 0.06699949966326603, "全帯域_40ms_envelope_correlation": 0.9956926221762145, "全帯域_40ms_modulation_mae_db": 0.14136460884708066, "300–1000 Hz_10ms_log_rms_mae_db": 1.6353568125569478, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17517825425367126, "300–1000 Hz_10ms_envelope_correlation": 0.9758116820681743, "300–1000 Hz_10ms_modulation_mae_db": 1.5640521825948184, "300–1000 Hz_40ms_log_rms_mae_db": 1.103280683621932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12295681340055417, "300–1000 Hz_40ms_envelope_correlation": 0.9874854359321941, "300–1000 Hz_40ms_modulation_mae_db": 0.8757002595128153, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8818565118066397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16169593732602175, "1000–3000 Hz_10ms_envelope_correlation": 0.9795858219995638, "1000–3000 Hz_10ms_modulation_mae_db": 0.48221123516411984, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1010614991216625, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07606645035970323, "1000–3000 Hz_40ms_envelope_correlation": 0.995537486438447, "1000–3000 Hz_40ms_modulation_mae_db": 0.48505839588294636, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4078320514082114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2177548655429155, "3000–6000 Hz_10ms_envelope_correlation": 0.9645102266000578, "3000–6000 Hz_10ms_modulation_mae_db": 1.1293066143194768, "3000–6000 Hz_40ms_log_rms_mae_db": 1.445038129780007, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12659808898149516, "3000–6000 Hz_40ms_envelope_correlation": 0.9874287344935265, "3000–6000 Hz_40ms_modulation_mae_db": 1.07178025098326, "mr_stft_db": 6.131011032193752, "band_mae_db": 0.363713660675355, "rms_gain_error_db": -0.014296136382986463, "output_peak": 0.4470554765346788}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.177925436772241, "全帯域_10ms_rms_envelope_nrmse": 0.13079992278332442, "全帯域_10ms_envelope_correlation": 0.9842295287214852, "全帯域_10ms_modulation_mae_db": 0.5633572600342795, "全帯域_40ms_log_rms_mae_db": 0.693258019242351, "全帯域_40ms_rms_envelope_nrmse": 0.057020969796474315, "全帯域_40ms_envelope_correlation": 0.9968850709025209, "全帯域_40ms_modulation_mae_db": 0.7135179405272183, "300–1000 Hz_10ms_log_rms_mae_db": 1.4874517322826786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16393942333121805, "300–1000 Hz_10ms_envelope_correlation": 0.9787151026216577, "300–1000 Hz_10ms_modulation_mae_db": 1.5060866385120726, "300–1000 Hz_40ms_log_rms_mae_db": 0.8940578582651643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0853018099250505, "300–1000 Hz_40ms_envelope_correlation": 0.9941318347873458, "300–1000 Hz_40ms_modulation_mae_db": 1.3197486864221946, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9192285082378355, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17964632637801758, "1000–3000 Hz_10ms_envelope_correlation": 0.9749261869576314, "1000–3000 Hz_10ms_modulation_mae_db": 0.9269452533989632, "1000–3000 Hz_40ms_log_rms_mae_db": 1.151904373011126, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08739124440786636, "1000–3000 Hz_40ms_envelope_correlation": 0.9937637439418827, "1000–3000 Hz_40ms_modulation_mae_db": 0.9375839691813096, "3000–6000 Hz_10ms_log_rms_mae_db": 2.399243599309468, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21857937107050415, "3000–6000 Hz_10ms_envelope_correlation": 0.9642299950644317, "3000–6000 Hz_10ms_modulation_mae_db": 0.6583769144458688, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3561371582096025, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11690374205220061, "3000–6000 Hz_40ms_envelope_correlation": 0.9894729302493062, "3000–6000 Hz_40ms_modulation_mae_db": 0.9064784655291757, "mr_stft_db": 6.131326717036408, "band_mae_db": 0.27029337115316615, "rms_gain_error_db": -0.0649337141638647, "output_peak": 0.43548078566074133}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 57920, "output_samples": 57920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.418200036469725, "spectral_convergence": 0.4184668453045852, "spectral_convergence_status": "ok", "amplitude_floor": 1.9579592332350965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.900624837642527, "spectral_convergence": 0.40442759445884435, "spectral_convergence_status": "ok", "amplitude_floor": 1.206925486771671e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.059021717377354, "spectral_convergence": 0.43354970841776486, "spectral_convergence_status": "ok", "amplitude_floor": 6.395923898397892e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042562388767747, "output_power": 0.00039692623745678517, "delta_db": -0.30316186862841343, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001322032073723835, "output_power": 0.0012755888243112154, "delta_db": -0.1553128601744097, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006256947892797981, "output_power": 0.0005427870962319566, "delta_db": -0.617330236785803, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002997481206139804, "output_power": 0.0002801897118993599, "delta_db": -0.29304284132501884, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003961082609243847, "output_power": 0.00038023322467287287, "delta_db": -0.17763837087209267, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.769256628656105e-06, "output_power": 1.1069288375059494e-05, "delta_db": 1.5374023475634546, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.513981420891532, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2770829727360738, "pitch": null, "output_rms_dbfs": -25.39584133558025, "output_peak": 0.42984869503408707, "reference_rms_dbfs": -25.11875836284418, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.125948863829869}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 57920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.00020370994921487513, "log_rms_mae_db": 1.082432756893315, "rms_envelope_nrmse": 0.1256181559938269, "envelope_correlation": 0.9859124761077388, "rms_gain_error_db": -0.27708294548254975, "reference_mean_power": 0.003081231895334411, "output_mean_power": 0.0028907869816938905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 1.3803128421607645e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003568424263155193, "output_power": 0.00034508539021214604, "reference_active": true, "delta_db": -0.1454991057926786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014048200587458577, "output_power": 9.499009726551648e-05, "reference_active": true, "delta_db": -1.6994236731514625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7025084230927585e-05, "output_power": 1.888486147585776e-05, "reference_active": true, "delta_db": 0.45024534295671}], "active_band_count": 3, "active_band_mae_db": 0.765056040633617}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.00013932647455664127, "log_rms_mae_db": 0.6346497718621422, "rms_envelope_nrmse": 0.0695078544934022, "envelope_correlation": 0.9961603740741006, "rms_gain_error_db": -0.27708166051264327, "reference_mean_power": 0.0031070091895842884, "output_mean_power": 0.0029149718942989285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 1.2870957120852427e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031478645736461515, "output_power": 0.00030185911092826427, "reference_active": true, "delta_db": -0.1821175144518565}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.928842911180188e-05, "output_power": 4.694683435538355e-05, "reference_active": true, "delta_db": -1.690544023086204}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4070650648317675e-06, "output_power": 1.2673795841781028e-06, "reference_active": true, "delta_db": -0.45407473406023335}], "active_band_count": 3, "active_band_mae_db": 0.7755787571994314}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.0001235612100218032, "log_rms_mae_db": 1.5128465908392754, "rms_envelope_nrmse": 0.15511460727505097, "envelope_correlation": 0.9813753425356774, "rms_gain_error_db": -0.3058255700965256, "reference_mean_power": 0.0011039001556015875, "output_mean_power": 0.0010288385710740195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 4.707425731708226e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018855725502559516, "output_power": 0.00016542057203286305, "reference_active": true, "delta_db": -0.5685372889550391}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8855557263471545e-05, "output_power": 5.639082839515058e-05, "reference_active": true, "delta_db": -0.18579001570903503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.265121532076037e-06, "output_power": 7.963265138281925e-06, "reference_active": true, "delta_db": 2.7115976523618834}], "active_band_count": 3, "active_band_mae_db": 1.1553083190086524}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.0001123986305082509, "log_rms_mae_db": 0.9893923921295646, "rms_envelope_nrmse": 0.09852823108761598, "envelope_correlation": 0.9928880732803802, "rms_gain_error_db": -0.3058257622574194, "reference_mean_power": 0.0011131375427710934, "output_mean_power": 0.0010374478003550285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 4.401493945076042e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017713633289459933, "output_power": 0.0001567096443584192, "reference_active": true, "delta_db": -0.5321192458727346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7898955892530673e-05, "output_power": 2.7256816441353558e-05, "reference_active": true, "delta_db": -0.10112820814305924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0644420897956173e-07, "output_power": 4.6710164667798215e-07, "reference_active": true, "delta_db": 0.6041045821125124}], "active_band_count": 3, "active_band_mae_db": 0.41245067870943536}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 8.851421864498589e-05, "log_rms_mae_db": 1.8891785816082167, "rms_envelope_nrmse": 0.18625773123116252, "envelope_correlation": 0.9729061965038992, "rms_gain_error_db": -0.26992743933673174, "reference_mean_power": 0.000506505585091631, "output_mean_power": 0.00047598305776232116}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 2.2793822474038483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010569673324300439, "output_power": 9.483844862116683e-05, "reference_active": true, "delta_db": -0.47077123701849466}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5368696588833574e-05, "output_power": 1.426359657063324e-05, "reference_active": true, "delta_db": -0.32407989949505633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.905327294990813e-06, "output_power": 4.006036289238911e-06, "reference_active": true, "delta_db": 0.11057442299982481}], "active_band_count": 3, "active_band_mae_db": 0.30180851983779194}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 8.08613786301264e-05, "log_rms_mae_db": 1.1470917928518318, "rms_envelope_nrmse": 0.0864766221203981, "envelope_correlation": 0.9945500645946034, "rms_gain_error_db": -0.2699225777957778, "reference_mean_power": 0.0005107405079369301, "output_mean_power": 0.0004799633172459803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 2.1269882857538688e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.701418164823122e-05, "output_power": 8.817277539719083e-05, "reference_active": true, "delta_db": -0.415007133799669}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.796784971798785e-06, "output_power": 6.66817231882592e-06, "reference_active": true, "delta_db": -0.08296716139165652}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1410692673718103e-07, "output_power": 1.6540718816370406e-07, "reference_active": true, "delta_db": -1.1207633878641563}], "active_band_count": 3, "active_band_mae_db": 0.5395792276851606}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 5.208111658419103e-05, "log_rms_mae_db": 2.2643381029428684, "rms_envelope_nrmse": 0.21176257991892453, "envelope_correlation": 0.966685084242522, "rms_gain_error_db": -0.051940501306181996, "reference_mean_power": 0.0001411400109581116, "output_mean_power": 0.0001394620666753967}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 8.582618674096986e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.483177751159981e-05, "output_power": 3.14518872076387e-05, "reference_active": true, "delta_db": -0.44328928108998833}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4244491686748372e-05, "output_power": 1.0547127402061209e-05, "reference_active": true, "delta_db": -1.3051276407565424}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3188700900305924e-06, "output_power": 2.51066241418376e-06, "reference_active": true, "delta_db": 0.3451190217440534}], "active_band_count": 3, "active_band_mae_db": 0.6978453145301948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 4.6774501347655295e-05, "log_rms_mae_db": 1.3386582910457636, "rms_envelope_nrmse": 0.11602866993886533, "envelope_correlation": 0.9893033372556315, "rms_gain_error_db": -0.051915127981779996, "reference_mean_power": 0.0001423195946220122, "output_mean_power": 0.00014062844845695967}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 7.830084455082624e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.343144196598961e-05, "output_power": 2.9938418401237536e-05, "reference_active": true, "delta_db": -0.4792625546944961}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.21519729141624e-06, "output_power": 5.642533170163592e-06, "reference_active": true, "delta_db": -1.06774090085308}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.774546333615459e-07, "output_power": 1.7778985408589513e-07, "reference_active": true, "delta_db": 0.008196301936293938}], "active_band_count": 3, "active_band_mae_db": 0.51839991916129}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7796250899994189, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "71465c0035c4b94c7fe905f72f7a7cd6e5a7e5a4ddfa1129c20642cba9e8cd93", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9757975634970568, "standard_deviation": 0.0038595194308359178, "minimum": 0.9703825537615681, "maximum": 0.9808004822616252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.916723714176625, "standard_deviation": 0.03575115474913873, "minimum": 1.8818565118066397, "maximum": 1.992464812028144, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6506202649950577, "standard_deviation": 0.2165689352248376, "minimum": 0.30180851983779194, "maximum": 0.9341602558179437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1770700111808899, "standard_deviation": 0.014685631250560147, "minimum": 0.15764227182363177, "maximum": 0.19591830079374198, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929504571580824, "standard_deviation": 0.0019825601265842613, "minimum": 0.989320310114476, "maximum": 0.995537486438447, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1716208091884157, "standard_deviation": 0.053976675099192024, "minimum": 1.1010614991216625, "maximum": 1.27168745530864, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6583711491952784, "standard_deviation": 0.27398656709918245, "minimum": 0.28609871474625986, "maximum": 1.1233667452178138, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09546994145744135, "standard_deviation": 0.013638249236822022, "minimum": 0.07606645035970323, "maximum": 0.11523779958008282, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9639399141877371, "standard_deviation": 0.001895069179094205, "minimum": 0.9607587919445317, "maximum": 0.966685084242522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3353852119744287, "standard_deviation": 0.05506297656351886, "minimum": 2.2643381029428684, "maximum": 2.4078320514082114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7787143440051385, "standard_deviation": 0.22019166307740362, "minimum": 0.5358599045793516, "maximum": 1.1293066143194768, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21997124044627994, "standard_deviation": 0.0054637766819873, "minimum": 0.21176257991892453, "maximum": 0.22874690124447847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9881875519290599, "standard_deviation": 0.0008873571655439059, "minimum": 0.9870418616618114, "maximum": 0.9894729302493062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3991796831823455, "standard_deviation": 0.036348988335703575, "minimum": 1.3386582910457636, "maximum": 1.445038129780007, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.787366372713091, "standard_deviation": 0.39356029006321985, "minimum": 0.16144949108453074, "maximum": 1.3984964900685342, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12258673434504738, "standard_deviation": 0.004496622991348535, "minimum": 0.11602866993886533, "maximum": 0.12781584583960232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9793216031960411, "standard_deviation": 0.0026986507289532683, "minimum": 0.9750485062964233, "maximum": 0.9824467146236654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5648339324444178, "standard_deviation": 0.06087759753691391, "minimum": 1.4874517322826786, "maximum": 1.6353568125569478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4611135981902046, "standard_deviation": 0.3747152894025162, "minimum": 1.083741739275675, "maximum": 2.2787764737951544, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16163673177270743, "standard_deviation": 0.010273207033155674, "minimum": 0.14905824916614072, "maximum": 0.1776102431298201, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9923851887096138, "standard_deviation": 0.0023202418786273872, "minimum": 0.9874854359321941, "maximum": 0.9945118906715266, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9811263076903339, "standard_deviation": 0.06747278205213515, "minimum": 0.8940578582651643, "maximum": 1.103280683621932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9021470540398635, "standard_deviation": 0.3244764798183148, "minimum": 0.41245067870943536, "maximum": 1.3197486864221946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09628959660288261, "standard_deviation": 0.013615453313422123, "minimum": 0.08165573034690773, "maximum": 0.12295681340055417, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.404629759068462, "standard_deviation": 0.09803541143918991, "minimum": 0.27029337115316615, "maximum": 0.566513175125633, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.128623129196317, "standard_deviation": 0.0043057334908195525, "minimum": 6.119647855930367, "maximum": 6.133038515302295, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46305127356863723, "standard_deviation": 0.05713943042899945, "minimum": 0.42274244409585765, "maximum": 0.5925933193479833, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09711805000967144, "standard_deviation": 0.09426923412637789, "minimum": -0.2770829727360738, "maximum": 0.03143540553846187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9863147079079833, "standard_deviation": 0.0021228331044513987, "minimum": 0.9840064704583433, "maximum": 0.9900004843711269, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1483264778304543, "standard_deviation": 0.04510981809348422, "minimum": 1.082432756893315, "maximum": 1.2060088911393647, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.935960545129588, "standard_deviation": 0.4465662609596753, "minimum": 0.527798035179046, "maximum": 1.7297933166547985, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12214200130346231, "standard_deviation": 0.009842168650361225, "minimum": 0.10452315558297559, "maximum": 0.13231152334038226, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9957540358533907, "standard_deviation": 0.001187591350464904, "minimum": 0.9932288484977969, "maximum": 0.9968850709025209, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6572627712628007, "standard_deviation": 0.03182203957944965, "minimum": 0.619224360347337, "maximum": 0.706860033057191, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0818022060528447, "standard_deviation": 0.5289997392464235, "minimum": 0.14136460884708066, "maximum": 1.7243833880748796, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06718974244001834, "standard_deviation": 0.008487580965997958, "minimum": 0.057020969796474315, "maximum": 0.08392289929286699, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75947-0011", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.9203927516937256, "predicted_mos": 1.9203927516937256, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "bec4d67de65a06a3107ef8116a253c7f94f714b6746fbaf9a54fd7cc865e2371", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7084511001594365, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1175478003218062, "全帯域_10ms_rms_envelope_nrmse": 0.1210627025987107, "全帯域_10ms_envelope_correlation": 0.9864991866805016, "全帯域_10ms_modulation_mae_db": 0.47402659955732745, "全帯域_40ms_log_rms_mae_db": 0.6575885307942271, "全帯域_40ms_rms_envelope_nrmse": 0.06404904886085619, "全帯域_40ms_envelope_correlation": 0.9961727444470715, "全帯域_40ms_modulation_mae_db": 0.7341801609113717, "300–1000 Hz_10ms_log_rms_mae_db": 1.4948065078430781, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16718008997238729, "300–1000 Hz_10ms_envelope_correlation": 0.9781728099692432, "300–1000 Hz_10ms_modulation_mae_db": 1.8025663874658935, "300–1000 Hz_40ms_log_rms_mae_db": 0.9352457774901828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10382444829718322, "300–1000 Hz_40ms_envelope_correlation": 0.9911849532685385, "300–1000 Hz_40ms_modulation_mae_db": 1.087884172783599, "1000–3000 Hz_10ms_log_rms_mae_db": 1.468624386303414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1559142382918217, "1000–3000 Hz_10ms_envelope_correlation": 0.9811646868444166, "1000–3000 Hz_10ms_modulation_mae_db": 1.0974534609414535, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9160881426054237, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08259529721983541, "1000–3000 Hz_40ms_envelope_correlation": 0.9945203682109499, "1000–3000 Hz_40ms_modulation_mae_db": 0.8436601639192377, "3000–6000 Hz_10ms_log_rms_mae_db": 1.719301010673098, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18134649163297856, "3000–6000 Hz_10ms_envelope_correlation": 0.9767151882312004, "3000–6000 Hz_10ms_modulation_mae_db": 0.3805661604211581, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0469908453068888, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11300580489859038, "3000–6000 Hz_40ms_envelope_correlation": 0.9918813626826722, "3000–6000 Hz_40ms_modulation_mae_db": 0.7681843278133207, "mr_stft_db": 5.763814202163874, "band_mae_db": 0.21872396383404125, "rms_gain_error_db": -0.1307724619884443, "output_peak": 0.4269936877032939}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.231967910200071, "全帯域_10ms_rms_envelope_nrmse": 0.13240660259828926, "全帯域_10ms_envelope_correlation": 0.9838330720865847, "全帯域_10ms_modulation_mae_db": 1.1225768344614857, "全帯域_40ms_log_rms_mae_db": 0.6956944947872327, "全帯域_40ms_rms_envelope_nrmse": 0.07012637322741914, "全帯域_40ms_envelope_correlation": 0.995430076363619, "全帯域_40ms_modulation_mae_db": 0.8864866388934992, "300–1000 Hz_10ms_log_rms_mae_db": 1.3568248117809414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1373008590478723, "300–1000 Hz_10ms_envelope_correlation": 0.9851217138754463, "300–1000 Hz_10ms_modulation_mae_db": 1.5872257820223787, "300–1000 Hz_40ms_log_rms_mae_db": 0.822794008072732, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06904247757447307, "300–1000 Hz_40ms_envelope_correlation": 0.9962231942341763, "300–1000 Hz_40ms_modulation_mae_db": 0.9297681142839593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5944095590435732, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15875607118758292, "1000–3000 Hz_10ms_envelope_correlation": 0.9807003377932215, "1000–3000 Hz_10ms_modulation_mae_db": 1.2236645525818168, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9251505769418512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07886499477342207, "1000–3000 Hz_40ms_envelope_correlation": 0.9956254322191866, "1000–3000 Hz_40ms_modulation_mae_db": 1.2093759761350207, "3000–6000 Hz_10ms_log_rms_mae_db": 1.68186072030785, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.212702284612152, "3000–6000 Hz_10ms_envelope_correlation": 0.96695443114014, "3000–6000 Hz_10ms_modulation_mae_db": 1.392301444695087, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9983529840429254, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13103567310699607, "3000–6000 Hz_40ms_envelope_correlation": 0.9866061606944082, "3000–6000 Hz_40ms_modulation_mae_db": 0.8649496601776715, "mr_stft_db": 5.774606978593533, "band_mae_db": 0.32449952518122643, "rms_gain_error_db": -0.1511521764377977, "output_peak": 0.40792702108863527}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2308733460612609, "全帯域_10ms_rms_envelope_nrmse": 0.11701675378383618, "全帯域_10ms_envelope_correlation": 0.9876880682867398, "全帯域_10ms_modulation_mae_db": 0.7966525438591807, "全帯域_40ms_log_rms_mae_db": 0.7389493639367027, "全帯域_40ms_rms_envelope_nrmse": 0.06497497613451111, "全帯域_40ms_envelope_correlation": 0.9964526494247753, "全帯域_40ms_modulation_mae_db": 0.6234012969781553, "300–1000 Hz_10ms_log_rms_mae_db": 1.499167679274522, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17554127564988314, "300–1000 Hz_10ms_envelope_correlation": 0.9756102993329924, "300–1000 Hz_10ms_modulation_mae_db": 1.167264834737693, "300–1000 Hz_40ms_log_rms_mae_db": 0.961635287450738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09739398871789116, "300–1000 Hz_40ms_envelope_correlation": 0.9921740769318144, "300–1000 Hz_40ms_modulation_mae_db": 0.4338344406932073, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5282377393831437, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1726920359513806, "1000–3000 Hz_10ms_envelope_correlation": 0.9775677391050976, "1000–3000 Hz_10ms_modulation_mae_db": 0.7492525518390268, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9646321781204663, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10281738862725222, "1000–3000 Hz_40ms_envelope_correlation": 0.9928534482071895, "1000–3000 Hz_40ms_modulation_mae_db": 0.6153185146547124, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6975613649692505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20029459780260458, "3000–6000 Hz_10ms_envelope_correlation": 0.9711100597069089, "3000–6000 Hz_10ms_modulation_mae_db": 0.6539595815587297, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0224312555277997, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12474593047231554, "3000–6000 Hz_40ms_envelope_correlation": 0.9896593700650308, "3000–6000 Hz_40ms_modulation_mae_db": 0.720605377760933, "mr_stft_db": 5.780066858630923, "band_mae_db": 0.4866452657109537, "rms_gain_error_db": -0.22705257217821587, "output_peak": 0.45276284957685203}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2269880722028976, "全帯域_10ms_rms_envelope_nrmse": 0.1079526303244138, "全帯域_10ms_envelope_correlation": 0.9892776144002815, "全帯域_10ms_modulation_mae_db": 0.9240959132613957, "全帯域_40ms_log_rms_mae_db": 0.7454351718794107, "全帯域_40ms_rms_envelope_nrmse": 0.06007293108599665, "全帯域_40ms_envelope_correlation": 0.9965055391373595, "全帯域_40ms_modulation_mae_db": 0.9310745177803453, "300–1000 Hz_10ms_log_rms_mae_db": 1.5017160611122968, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19492767397758656, "300–1000 Hz_10ms_envelope_correlation": 0.970822260301362, "300–1000 Hz_10ms_modulation_mae_db": 1.677860737827795, "300–1000 Hz_40ms_log_rms_mae_db": 0.9672922369359069, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12288654810081222, "300–1000 Hz_40ms_envelope_correlation": 0.9878196032348913, "300–1000 Hz_40ms_modulation_mae_db": 1.1162691628321317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3907281910680407, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15486105183539609, "1000–3000 Hz_10ms_envelope_correlation": 0.9814732931799665, "1000–3000 Hz_10ms_modulation_mae_db": 0.5072293114249127, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8916629957054053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09035207298045, "1000–3000 Hz_40ms_envelope_correlation": 0.9936733342533253, "1000–3000 Hz_40ms_modulation_mae_db": 0.49080184752177186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7066303426593836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2199150318048552, "3000–6000 Hz_10ms_envelope_correlation": 0.967641952403447, "3000–6000 Hz_10ms_modulation_mae_db": 0.6341334156568467, "3000–6000 Hz_40ms_log_rms_mae_db": 1.083653665368596, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15924249599520762, "3000–6000 Hz_40ms_envelope_correlation": 0.982869207685263, "3000–6000 Hz_40ms_modulation_mae_db": 0.8040722450357191, "mr_stft_db": 5.77852907948872, "band_mae_db": 0.3263103028270502, "rms_gain_error_db": -0.039324040399334306, "output_peak": 0.5384451853440365}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1645842543599698, "全帯域_10ms_rms_envelope_nrmse": 0.11660983984104364, "全帯域_10ms_envelope_correlation": 0.9875292551650687, "全帯域_10ms_modulation_mae_db": 0.8037385282888557, "全帯域_40ms_log_rms_mae_db": 0.6847488012885067, "全帯域_40ms_rms_envelope_nrmse": 0.05905483889987746, "全帯域_40ms_envelope_correlation": 0.9968030706305143, "全帯域_40ms_modulation_mae_db": 0.49611863318434307, "300–1000 Hz_10ms_log_rms_mae_db": 1.4099956059835743, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1628440268907751, "300–1000 Hz_10ms_envelope_correlation": 0.9791093311630434, "300–1000 Hz_10ms_modulation_mae_db": 1.5754915163144094, "300–1000 Hz_40ms_log_rms_mae_db": 0.8842588547215604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10548232240509933, "300–1000 Hz_40ms_envelope_correlation": 0.9908382670448912, "300–1000 Hz_40ms_modulation_mae_db": 1.4716973871488896, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4863472185176243, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13970253836456648, "1000–3000 Hz_10ms_envelope_correlation": 0.985390268211867, "1000–3000 Hz_10ms_modulation_mae_db": 0.5225461712098846, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9172897284070005, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07807574212689843, "1000–3000 Hz_40ms_envelope_correlation": 0.9959947466908508, "1000–3000 Hz_40ms_modulation_mae_db": 0.48467444976545365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7337997483960714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20285007059081825, "3000–6000 Hz_10ms_envelope_correlation": 0.9693253132106735, "3000–6000 Hz_10ms_modulation_mae_db": 0.7049763462954312, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0207959332806367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1089973073091551, "3000–6000 Hz_40ms_envelope_correlation": 0.9907013575692891, "3000–6000 Hz_40ms_modulation_mae_db": 0.8029771422147078, "mr_stft_db": 5.771440262252237, "band_mae_db": 0.3815394220325672, "rms_gain_error_db": -0.13824800703689055, "output_peak": 0.45821582126078886}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.215758578825931, "全帯域_10ms_rms_envelope_nrmse": 0.13024858473128859, "全帯域_10ms_envelope_correlation": 0.9850169212548807, "全帯域_10ms_modulation_mae_db": 1.0288350260039725, "全帯域_40ms_log_rms_mae_db": 0.7361740905976821, "全帯域_40ms_rms_envelope_nrmse": 0.08570868186978285, "全帯域_40ms_envelope_correlation": 0.9939168389673128, "全帯域_40ms_modulation_mae_db": 1.3350756925792406, "300–1000 Hz_10ms_log_rms_mae_db": 1.3015095156766192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14575512808099814, "300–1000 Hz_10ms_envelope_correlation": 0.9835193291711268, "300–1000 Hz_10ms_modulation_mae_db": 1.0086141237418123, "300–1000 Hz_40ms_log_rms_mae_db": 0.8335561635141534, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08905827217753176, "300–1000 Hz_40ms_envelope_correlation": 0.9941422357996622, "300–1000 Hz_40ms_modulation_mae_db": 0.3886422983912274, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5343995768003051, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19472974215580435, "1000–3000 Hz_10ms_envelope_correlation": 0.9707893547273031, "1000–3000 Hz_10ms_modulation_mae_db": 1.1128183789980508, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9125410268160586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10728916367707331, "1000–3000 Hz_40ms_envelope_correlation": 0.990768830983155, "1000–3000 Hz_40ms_modulation_mae_db": 1.0139575769328522, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6402799857694128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18284235392316334, "3000–6000 Hz_10ms_envelope_correlation": 0.9751949749006291, "3000–6000 Hz_10ms_modulation_mae_db": 0.35516812286826016, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8986759075987223, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08383807334543575, "3000–6000 Hz_40ms_envelope_correlation": 0.9946043671691658, "3000–6000 Hz_40ms_modulation_mae_db": 0.4059912551249103, "mr_stft_db": 5.752652778398942, "band_mae_db": 0.4200177488047821, "rms_gain_error_db": -0.32036610076361044, "output_peak": 0.3997044982171329}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.100331904373152, "全帯域_10ms_rms_envelope_nrmse": 0.12438354551088829, "全帯域_10ms_envelope_correlation": 0.9860148327638281, "全帯域_10ms_modulation_mae_db": 1.0837765492216718, "全帯域_40ms_log_rms_mae_db": 0.5818078463514632, "全帯域_40ms_rms_envelope_nrmse": 0.07484142588159623, "全帯域_40ms_envelope_correlation": 0.9951127112433232, "全帯域_40ms_modulation_mae_db": 0.9913058304814596, "300–1000 Hz_10ms_log_rms_mae_db": 1.3238031825427585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14129583065559073, "300–1000 Hz_10ms_envelope_correlation": 0.9845755013965604, "300–1000 Hz_10ms_modulation_mae_db": 1.2465046224203382, "300–1000 Hz_40ms_log_rms_mae_db": 0.831587892593166, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0756578597933517, "300–1000 Hz_40ms_envelope_correlation": 0.9954228493854284, "300–1000 Hz_40ms_modulation_mae_db": 0.8683467720038299, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4672478166682386, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15854922918133849, "1000–3000 Hz_10ms_envelope_correlation": 0.9807861304101757, "1000–3000 Hz_10ms_modulation_mae_db": 0.9886478262147896, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9034306404015903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08672822970830239, "1000–3000 Hz_40ms_envelope_correlation": 0.9946671125779091, "1000–3000 Hz_40ms_modulation_mae_db": 1.0075902296827968, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7001782483629575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1816648077774805, "3000–6000 Hz_10ms_envelope_correlation": 0.9776966009277361, "3000–6000 Hz_10ms_modulation_mae_db": 1.3142541429154293, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0760098818247459, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11600716771719559, "3000–6000 Hz_40ms_envelope_correlation": 0.992715797459244, "3000–6000 Hz_40ms_modulation_mae_db": 1.2867775254797404, "mr_stft_db": 5.773727977589643, "band_mae_db": 0.42322160086086935, "rms_gain_error_db": -0.233467910207388, "output_peak": 0.499440780944342}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1575448356877989, "全帯域_10ms_rms_envelope_nrmse": 0.10646557468951397, "全帯域_10ms_envelope_correlation": 0.9896878731579389, "全帯域_10ms_modulation_mae_db": 1.1657830920848877, "全帯域_40ms_log_rms_mae_db": 0.626383868079687, "全帯域_40ms_rms_envelope_nrmse": 0.05072116075910514, "全帯域_40ms_envelope_correlation": 0.9975339425179031, "全帯域_40ms_modulation_mae_db": 0.9974825124978501, "300–1000 Hz_10ms_log_rms_mae_db": 1.4858011874671548, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1878536104173776, "300–1000 Hz_10ms_envelope_correlation": 0.9732982204175644, "300–1000 Hz_10ms_modulation_mae_db": 1.669919180811392, "300–1000 Hz_40ms_log_rms_mae_db": 1.0176544442697024, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12981912900238407, "300–1000 Hz_40ms_envelope_correlation": 0.9868531505434569, "300–1000 Hz_40ms_modulation_mae_db": 1.3545654928718094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3983924364637597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14600808403380552, "1000–3000 Hz_10ms_envelope_correlation": 0.9841510912121939, "1000–3000 Hz_10ms_modulation_mae_db": 0.8948594846089694, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8376781686126141, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07824078610470672, "1000–3000 Hz_40ms_envelope_correlation": 0.9963018116106299, "1000–3000 Hz_40ms_modulation_mae_db": 0.7097775735576454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6960442082786795, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.189342106502504, "3000–6000 Hz_10ms_envelope_correlation": 0.9738455626105461, "3000–6000 Hz_10ms_modulation_mae_db": 0.24686915980263321, "3000–6000 Hz_40ms_log_rms_mae_db": 1.005627741071378, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12361266978530468, "3000–6000 Hz_40ms_envelope_correlation": 0.9889383685031621, "3000–6000 Hz_40ms_modulation_mae_db": 0.20973134036739016, "mr_stft_db": 5.780812426074665, "band_mae_db": 0.28298351439646935, "rms_gain_error_db": 0.008554357990544142, "output_peak": 0.42168315119220506}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 57920, "output_samples": 57920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.944260080362357, "spectral_convergence": 0.4089483640086001, "spectral_convergence_status": "ok", "amplitude_floor": 1.9579592332350965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4521782669468415, "spectral_convergence": 0.40671388323978297, "spectral_convergence_status": "ok", "amplitude_floor": 1.206925486771671e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.895004259182424, "spectral_convergence": 0.45133068168069984, "spectral_convergence_status": "ok", "amplitude_floor": 6.395923898397892e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042562388767747, "output_power": 0.000424026788932713, "delta_db": -0.016326991972462866, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001322032073723835, "output_power": 0.0012877396161416874, "delta_db": -0.11413935088347663, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006256947892797981, "output_power": 0.0005960470041221343, "delta_db": -0.21082028785756762, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002997481206139804, "output_power": 0.0002804354749125596, "delta_db": -0.28923518052645486, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003961082609243847, "output_power": 0.00038853997729051314, "delta_db": -0.0837818934913768, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.769256628656105e-06, "output_power": 8.91627541250227e-06, "delta_db": 0.5980400782729088, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21872396383404125, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1307724619884443, "pitch": null, "output_rms_dbfs": -25.249530824832625, "output_peak": 0.4269936877032939, "reference_rms_dbfs": -25.11875836284418, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.763814202163874}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 57920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.00020370994921487513, "log_rms_mae_db": 1.1175478003218062, "rms_envelope_nrmse": 0.1210627025987107, "envelope_correlation": 0.9864991866805016, "rms_gain_error_db": -0.13077255117355077, "reference_mean_power": 0.003081231895334411, "output_mean_power": 0.0029898343673932697}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 1.3803128421607645e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003568424263155193, "output_power": 0.000340403471751648, "reference_active": true, "delta_db": -0.20482502875898118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014048200587458577, "output_power": 0.00010746403931382537, "reference_active": true, "delta_db": -1.1635753906637516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7025084230927585e-05, "output_power": 1.681594705536854e-05, "reference_active": true, "delta_db": -0.05367937924924952}], "active_band_count": 3, "active_band_mae_db": 0.47402659955732745}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.00013932647455664127, "log_rms_mae_db": 0.6575885307942271, "rms_envelope_nrmse": 0.06404904886085619, "envelope_correlation": 0.9961727444470715, "rms_gain_error_db": -0.1307737995507405, "reference_mean_power": 0.0031070091895842884, "output_mean_power": 0.003014846171963493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 1.2870957120852427e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031478645736461515, "output_power": 0.0002979758088727877, "reference_active": true, "delta_db": -0.23835032653678837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.928842911180188e-05, "output_power": 5.5002474844788516e-05, "reference_active": true, "delta_db": -1.0027848415935359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4070650648317675e-06, "output_power": 1.1276482464057022e-06, "reference_active": true, "delta_db": -0.961405314603791}], "active_band_count": 3, "active_band_mae_db": 0.7341801609113717}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.0001235612100218032, "log_rms_mae_db": 1.4948065078430781, "rms_envelope_nrmse": 0.16718008997238729, "envelope_correlation": 0.9781728099692432, "rms_gain_error_db": 0.04994610616142418, "reference_mean_power": 0.0011039001556015875, "output_mean_power": 0.0011166688595560038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 4.707425731708226e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018855725502559516, "output_power": 0.00015984796764266653, "reference_active": true, "delta_db": -0.7173612831403902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8855557263471545e-05, "output_power": 5.369159809517887e-05, "reference_active": true, "delta_db": -0.39881145453128436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.265121532076037e-06, "output_power": 1.1457346648394745e-05, "reference_active": true, "delta_db": 4.291526424726006}], "active_band_count": 3, "active_band_mae_db": 1.8025663874658935}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.0001123986305082509, "log_rms_mae_db": 0.9352457774901828, "rms_envelope_nrmse": 0.10382444829718322, "envelope_correlation": 0.9911849532685385, "rms_gain_error_db": 0.04994633237559596, "reference_mean_power": 0.0011131375427710934, "output_mean_power": 0.001126013153321052}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 4.401493945076042e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017713633289459933, "output_power": 0.00015087791650178974, "reference_active": true, "delta_db": -0.6968197140926409}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7898955892530673e-05, "output_power": 2.4342082423825074e-05, "reference_active": true, "delta_db": -0.5923022163833024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0644420897956173e-07, "output_power": 6.404039454483034e-07, "reference_active": true, "delta_db": 1.974530587874854}], "active_band_count": 3, "active_band_mae_db": 1.087884172783599}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 8.851421864498589e-05, "log_rms_mae_db": 1.468624386303414, "rms_envelope_nrmse": 0.1559142382918217, "envelope_correlation": 0.9811646868444166, "rms_gain_error_db": -0.08062683529660718, "reference_mean_power": 0.000506505585091631, "output_mean_power": 0.0004971890497484447}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 2.2793822474038483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010569673324300439, "output_power": 8.898926651562952e-05, "reference_active": true, "delta_db": -0.7472393766423104}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5368696588833574e-05, "output_power": 1.3583113844884556e-05, "reference_active": true, "delta_db": -0.5363769608905026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.905327294990813e-06, "output_power": 6.2020011341740485e-06, "reference_active": true, "delta_db": 2.0087440452915475}], "active_band_count": 3, "active_band_mae_db": 1.0974534609414535}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 8.08613786301264e-05, "log_rms_mae_db": 0.9160881426054237, "rms_envelope_nrmse": 0.08259529721983541, "envelope_correlation": 0.9945203682109499, "rms_gain_error_db": -0.08062311887154978, "reference_mean_power": 0.0005107405079369301, "output_mean_power": 0.0005013465055177151}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 2.1269882857538688e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.701418164823122e-05, "output_power": 8.36219492314364e-05, "reference_active": true, "delta_db": -0.6451493783329052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.796784971798785e-06, "output_power": 6.076872901842621e-06, "reference_active": true, "delta_db": -0.4862337732529358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1410692673718103e-07, "output_power": 2.9552243172628143e-07, "reference_active": true, "delta_db": 1.3995973401718722}], "active_band_count": 3, "active_band_mae_db": 0.8436601639192377}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 5.208111658419103e-05, "log_rms_mae_db": 1.719301010673098, "rms_envelope_nrmse": 0.18134649163297856, "envelope_correlation": 0.9767151882312004, "rms_gain_error_db": -0.49633748517905335, "reference_mean_power": 0.0001411400109581116, "output_mean_power": 0.00012589729477996433}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 8.582618674096986e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.483177751159981e-05, "output_power": 3.27471882736095e-05, "reference_active": true, "delta_db": -0.2680162094057798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4244491686748372e-05, "output_power": 1.1858909509398385e-05, "reference_active": true, "delta_db": -0.7960220094959026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3188700900305924e-06, "output_power": 2.277772748062574e-06, "reference_active": true, "delta_db": -0.07766026236179202}], "active_band_count": 3, "active_band_mae_db": 0.3805661604211581}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 4.6774501347655295e-05, "log_rms_mae_db": 1.0469908453068888, "rms_envelope_nrmse": 0.11300580489859038, "envelope_correlation": 0.9918813626826722, "rms_gain_error_db": -0.49639166489884734, "reference_mean_power": 0.0001423195946220122, "output_mean_power": 0.000126947903064689}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 7.830084455082624e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.343144196598961e-05, "output_power": 3.134896320458688e-05, "reference_active": true, "delta_db": -0.279319268941817}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.21519729141624e-06, "output_power": 6.281213817583858e-06, "reference_active": true, "delta_db": -0.6020463349359149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.774546333615459e-07, "output_power": 1.27869982386843e-07, "reference_active": true, "delta_db": -1.42318737956223}], "active_band_count": 3, "active_band_mae_db": 0.7681843278133207}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8007530201591319, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b834f1b678da47663d566c72654d2aeb52d27edd63a52b86ece09b171be0f697", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9802528626855302, "standard_deviation": 0.004490664523405105, "minimum": 0.9707893547273031, "maximum": 0.985390268211867, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4835483655310124, "standard_deviation": 0.06890632743056418, "minimum": 1.3907281910680407, "maximum": 1.5944095590435732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.887058967227363, "standard_deviation": 0.271253274748245, "minimum": 0.5072293114249127, "maximum": 1.2236645525818168, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.160151623875212, "standard_deviation": 0.016994006680859177, "minimum": 0.13970253836456648, "maximum": 0.19472974215580435, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9943006355941495, "standard_deviation": 0.0018428973257401673, "minimum": 0.990768830983155, "maximum": 0.9963018116106299, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9085591822013013, "standard_deviation": 0.03566534765926834, "minimum": 0.8376781686126141, "maximum": 0.9646321781204663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7968945415211863, "standard_deviation": 0.26607473019729333, "minimum": 0.48467444976545365, "maximum": 1.2093759761350207, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08812045940224257, "standard_deviation": 0.011370718579811385, "minimum": 0.07807574212689843, "maximum": 0.10728916367707331, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9723105103914101, "standard_deviation": 0.004139067901367523, "minimum": 0.96695443114014, "maximum": 0.9776966009277361, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6969569536770879, "standard_deviation": 0.02777967646093775, "minimum": 1.6402799857694128, "maximum": 1.7337997483960714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7102785467766969, "standard_deviation": 0.4290357469294092, "minimum": 0.24686915980263321, "maximum": 1.392301444695087, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19636971808081954, "standard_deviation": 0.014894115821979883, "minimum": 0.18134649163297856, "maximum": 0.2199150318048552, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9897469989785294, "standard_deviation": 0.003697577183058524, "minimum": 0.982869207685263, "maximum": 0.9946043671691658, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0190672767527116, "standard_deviation": 0.057708227889210026, "minimum": 0.8986759075987223, "maximum": 1.083653665368596, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7329111092467991, "standard_deviation": 0.3198581459568128, "minimum": 0.20973134036739016, "maximum": 1.2867775254797404, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12006064032877509, "standard_deviation": 0.02135051372571202, "minimum": 0.08383807334543575, "maximum": 0.15924249599520762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9787786832034173, "standard_deviation": 0.005347443554831294, "minimum": 0.970822260301362, "maximum": 0.9851217138754463, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4217030689601182, "standard_deviation": 0.0847118285654907, "minimum": 1.3015095156766192, "maximum": 1.5017160611122968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.466930898167714, "standard_deviation": 0.2861180692980115, "minimum": 1.0086141237418123, "maximum": 1.8025663874658935, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16408731183655884, "standard_deviation": 0.02148522360993972, "minimum": 0.1373008590478723, "maximum": 0.19492767397758656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9918322913053574, "standard_deviation": 0.0033790411970575076, "minimum": 0.9868531505434569, "maximum": 0.9962231942341763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9067530831310178, "standard_deviation": 0.07401805043851539, "minimum": 0.822794008072732, "maximum": 1.0176544442697024, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9563759801260817, "standard_deviation": 0.39111554540916454, "minimum": 0.3886422983912274, "maximum": 1.4716973871488896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09914563075859083, "standard_deviation": 0.02114317828138208, "minimum": 0.06904247757447307, "maximum": 0.12981912900238407, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.35799266795599494, "standard_deviation": 0.0864833368074605, "minimum": 0.21872396383404125, "maximum": 0.4866452657109537, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.771956320399067, "standard_deviation": 0.009537628466691863, "minimum": 5.752652778398942, "maximum": 5.780812426074665, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4506466244159108, "standard_deviation": 0.04776375955617775, "minimum": 0.3997044982171329, "maximum": 0.5384451853440365, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1539786138776421, "standard_deviation": 0.10670369641101735, "minimum": -0.32036610076361044, "maximum": 0.008554357990544142, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.986943352974478, "standard_deviation": 0.0020115193108277323, "minimum": 0.9838330720865847, "maximum": 0.9896878731579389, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1806995877541109, "standard_deviation": 0.053131323634664884, "minimum": 1.100331904373152, "maximum": 1.231967910200071, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9249356358423472, "standard_deviation": 0.22914749283714325, "minimum": 0.47402659955732745, "maximum": 1.1657830920848877, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11951827925974805, "standard_deviation": 0.009457864260481806, "minimum": 0.10646557468951397, "maximum": 0.13240660259828926, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9959909465914849, "standard_deviation": 0.001130057047514635, "minimum": 0.9939168389673128, "maximum": 0.9975339425179031, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.683347770964364, "standard_deviation": 0.05868826603451676, "minimum": 0.5818078463514632, "maximum": 0.7454351718794107, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8743906604132832, "standard_deviation": 0.2591621675423849, "minimum": 0.49611863318434307, "maximum": 1.3350756925792406, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0661936795898931, "standard_deviation": 0.010723772229737045, "minimum": 0.05072116075910514, "maximum": 0.08570868186978285, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75947-0011", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.9943931102752686, "predicted_mos": 1.9943931102752686, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "9fc12b527353b6027e40b991ea47f152a14fd635d07d5cf5c3a58759f217a339", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9905037004500628, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.114995272152549, "全帯域_10ms_rms_envelope_nrmse": 0.1060988775617219, "全帯域_10ms_envelope_correlation": 0.9902079269884586, "全帯域_10ms_modulation_mae_db": 0.7391304508277342, "全帯域_40ms_log_rms_mae_db": 0.610440659673779, "全帯域_40ms_rms_envelope_nrmse": 0.06680534479579985, "全帯域_40ms_envelope_correlation": 0.9964840612217317, "全帯域_40ms_modulation_mae_db": 1.0108918288128752, "300–1000 Hz_10ms_log_rms_mae_db": 1.3946965117266994, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1653808061947589, "300–1000 Hz_10ms_envelope_correlation": 0.97858722311798, "300–1000 Hz_10ms_modulation_mae_db": 1.1089898424566178, "300–1000 Hz_40ms_log_rms_mae_db": 0.8748730507310157, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0986422476567982, "300–1000 Hz_40ms_envelope_correlation": 0.9926985708193817, "300–1000 Hz_40ms_modulation_mae_db": 0.3612017633238587, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6471252632908961, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16017549681863574, "1000–3000 Hz_10ms_envelope_correlation": 0.9811178077548075, "1000–3000 Hz_10ms_modulation_mae_db": 1.0727677371099285, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9915992846770609, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08744347410705239, "1000–3000 Hz_40ms_envelope_correlation": 0.9956363682308518, "1000–3000 Hz_40ms_modulation_mae_db": 1.10743747547225, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9146319899565276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19928212085560337, "3000–6000 Hz_10ms_envelope_correlation": 0.9705701267594835, "3000–6000 Hz_10ms_modulation_mae_db": 0.7617141881997272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1916047572478627, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11786867917552982, "3000–6000 Hz_40ms_envelope_correlation": 0.9894501788173858, "3000–6000 Hz_40ms_modulation_mae_db": 0.9510966072058927, "mr_stft_db": 5.907316022587865, "band_mae_db": 0.48305034861926827, "rms_gain_error_db": -0.2729629708833871, "output_peak": 0.48895101683767817}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1953417829721678, "全帯域_10ms_rms_envelope_nrmse": 0.1283575085026532, "全帯域_10ms_envelope_correlation": 0.9849210785727031, "全帯域_10ms_modulation_mae_db": 2.1278414474724876, "全帯域_40ms_log_rms_mae_db": 0.7136268520041863, "全帯域_40ms_rms_envelope_nrmse": 0.06558809256881082, "全帯域_40ms_envelope_correlation": 0.9958666372651732, "全帯域_40ms_modulation_mae_db": 2.4357300301881657, "300–1000 Hz_10ms_log_rms_mae_db": 1.3805821983265467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14429337477734736, "300–1000 Hz_10ms_envelope_correlation": 0.9835462792780908, "300–1000 Hz_10ms_modulation_mae_db": 0.9883147151076083, "300–1000 Hz_40ms_log_rms_mae_db": 0.8951728393556051, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08119407801505418, "300–1000 Hz_40ms_envelope_correlation": 0.9946934511661981, "300–1000 Hz_40ms_modulation_mae_db": 0.5961666035257732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7242949449785245, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18128937509875107, "1000–3000 Hz_10ms_envelope_correlation": 0.9745880706746348, "1000–3000 Hz_10ms_modulation_mae_db": 1.1060533885399055, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0871660365069507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09581915304129893, "1000–3000 Hz_40ms_envelope_correlation": 0.9930474732292467, "1000–3000 Hz_40ms_modulation_mae_db": 1.1468040899452625, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9012168329721626, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2021338433417002, "3000–6000 Hz_10ms_envelope_correlation": 0.9696022858184915, "3000–6000 Hz_10ms_modulation_mae_db": 0.9369495710244761, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1189262501839898, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10734146125853476, "3000–6000 Hz_40ms_envelope_correlation": 0.9910170391721373, "3000–6000 Hz_40ms_modulation_mae_db": 1.0190551533252532, "mr_stft_db": 5.889832460595536, "band_mae_db": 0.3586346897809117, "rms_gain_error_db": -0.0631522169082553, "output_peak": 0.46131048253657847}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1272696992887088, "全帯域_10ms_rms_envelope_nrmse": 0.10511278322336205, "全帯域_10ms_envelope_correlation": 0.9898782298211687, "全帯域_10ms_modulation_mae_db": 1.046227538024419, "全帯域_40ms_log_rms_mae_db": 0.7040946442063604, "全帯域_40ms_rms_envelope_nrmse": 0.05558472834504527, "全帯域_40ms_envelope_correlation": 0.9970298361452912, "全帯域_40ms_modulation_mae_db": 0.8809312914498829, "300–1000 Hz_10ms_log_rms_mae_db": 1.5229413897950326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18717066851336506, "300–1000 Hz_10ms_envelope_correlation": 0.9724960342732255, "300–1000 Hz_10ms_modulation_mae_db": 2.4167875811355337, "300–1000 Hz_40ms_log_rms_mae_db": 0.9724730280565814, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12179654540804104, "300–1000 Hz_40ms_envelope_correlation": 0.9878036337218644, "300–1000 Hz_40ms_modulation_mae_db": 1.5950209346692128, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5516083107571683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15743380196761775, "1000–3000 Hz_10ms_envelope_correlation": 0.9808643832081609, "1000–3000 Hz_10ms_modulation_mae_db": 0.44780334836317165, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9748663667031089, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0855435816877023, "1000–3000 Hz_40ms_envelope_correlation": 0.9942388677886309, "1000–3000 Hz_40ms_modulation_mae_db": 0.5239375451173187, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9907331339774945, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19439590970678333, "3000–6000 Hz_10ms_envelope_correlation": 0.9718542855097526, "3000–6000 Hz_10ms_modulation_mae_db": 0.7904416785010683, "3000–6000 Hz_40ms_log_rms_mae_db": 1.201635490140058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10753061582485914, "3000–6000 Hz_40ms_envelope_correlation": 0.9911617508350692, "3000–6000 Hz_40ms_modulation_mae_db": 0.8324165865162202, "mr_stft_db": 5.872637005157863, "band_mae_db": 0.3129640222641155, "rms_gain_error_db": -0.03500475590008646, "output_peak": 0.5011832241237107}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.117652062346049, "全帯域_10ms_rms_envelope_nrmse": 0.11696256360212487, "全帯域_10ms_envelope_correlation": 0.9874408130998896, "全帯域_10ms_modulation_mae_db": 0.45840043404946423, "全帯域_40ms_log_rms_mae_db": 0.5660331323059691, "全帯域_40ms_rms_envelope_nrmse": 0.06221579957492078, "全帯域_40ms_envelope_correlation": 0.9963136698834223, "全帯域_40ms_modulation_mae_db": 0.7940800832446427, "300–1000 Hz_10ms_log_rms_mae_db": 1.5129038507615997, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15283587696704187, "300–1000 Hz_10ms_envelope_correlation": 0.9815130123408691, "300–1000 Hz_10ms_modulation_mae_db": 1.124662568518656, "300–1000 Hz_40ms_log_rms_mae_db": 0.9202872942716026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07968329795149533, "300–1000 Hz_40ms_envelope_correlation": 0.994901288298202, "300–1000 Hz_40ms_modulation_mae_db": 0.6244821440985278, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6904938565663237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1920670092074813, "1000–3000 Hz_10ms_envelope_correlation": 0.9725922307271794, "1000–3000 Hz_10ms_modulation_mae_db": 0.7505757239583758, "1000–3000 Hz_40ms_log_rms_mae_db": 1.071156309819977, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13317946310928702, "1000–3000 Hz_40ms_envelope_correlation": 0.9863793081568883, "1000–3000 Hz_40ms_modulation_mae_db": 0.5642386581378196, "3000–6000 Hz_10ms_log_rms_mae_db": 1.96722645683293, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21056219446474772, "3000–6000 Hz_10ms_envelope_correlation": 0.9673434494471692, "3000–6000 Hz_10ms_modulation_mae_db": 0.8436551902654172, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1954234640547916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.127104882395121, "3000–6000 Hz_40ms_envelope_correlation": 0.988487491123473, "3000–6000 Hz_40ms_modulation_mae_db": 0.6499492636080316, "mr_stft_db": 5.882187834458814, "band_mae_db": 0.34739179383185165, "rms_gain_error_db": -0.07858977709040085, "output_peak": 0.517972335147211}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1390925412566986, "全帯域_10ms_rms_envelope_nrmse": 0.10296719441942855, "全帯域_10ms_envelope_correlation": 0.9903568801036461, "全帯域_10ms_modulation_mae_db": 1.1361687747624727, "全帯域_40ms_log_rms_mae_db": 0.6222766994130617, "全帯域_40ms_rms_envelope_nrmse": 0.057335147537721946, "全帯域_40ms_envelope_correlation": 0.9968515387919566, "全帯域_40ms_modulation_mae_db": 1.0690102654973592, "300–1000 Hz_10ms_log_rms_mae_db": 1.5199191838442117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15905331140872372, "300–1000 Hz_10ms_envelope_correlation": 0.9802011247459467, "300–1000 Hz_10ms_modulation_mae_db": 1.9420272574899184, "300–1000 Hz_40ms_log_rms_mae_db": 0.889128428607628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08767207233542038, "300–1000 Hz_40ms_envelope_correlation": 0.9937191958975148, "300–1000 Hz_40ms_modulation_mae_db": 1.343430631430164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6520958384232753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17560279405673576, "1000–3000 Hz_10ms_envelope_correlation": 0.976862714533842, "1000–3000 Hz_10ms_modulation_mae_db": 0.9634673628579323, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0424891513655328, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10720421037994474, "1000–3000 Hz_40ms_envelope_correlation": 0.991079892371037, "1000–3000 Hz_40ms_modulation_mae_db": 0.7996945034250231, "3000–6000 Hz_10ms_log_rms_mae_db": 1.919560254423038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19033784736575, "3000–6000 Hz_10ms_envelope_correlation": 0.9740011579178433, "3000–6000 Hz_10ms_modulation_mae_db": 0.8153974485576887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2462600540688202, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12350812825658605, "3000–6000 Hz_40ms_envelope_correlation": 0.9896756191902797, "3000–6000 Hz_40ms_modulation_mae_db": 0.8504508897602784, "mr_stft_db": 5.879480700226488, "band_mae_db": 0.38436013294700294, "rms_gain_error_db": 0.030065592356028945, "output_peak": 0.4862435739507034}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1349828126103374, "全帯域_10ms_rms_envelope_nrmse": 0.11682484177394273, "全帯域_10ms_envelope_correlation": 0.9874754108194067, "全帯域_10ms_modulation_mae_db": 0.34580554947569936, "全帯域_40ms_log_rms_mae_db": 0.6739323735760809, "全帯域_40ms_rms_envelope_nrmse": 0.0650508200154244, "全帯域_40ms_envelope_correlation": 0.9960553436423841, "全帯域_40ms_modulation_mae_db": 0.5214921905315841, "300–1000 Hz_10ms_log_rms_mae_db": 1.596154206402817, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17312882725542778, "300–1000 Hz_10ms_envelope_correlation": 0.9767354245409002, "300–1000 Hz_10ms_modulation_mae_db": 1.8877958072569998, "300–1000 Hz_40ms_log_rms_mae_db": 1.0616656619460358, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11300433090744323, "300–1000 Hz_40ms_envelope_correlation": 0.9895934439788768, "300–1000 Hz_40ms_modulation_mae_db": 1.3727201096870256, "1000–3000 Hz_10ms_log_rms_mae_db": 1.654122577125373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17738162219487025, "1000–3000 Hz_10ms_envelope_correlation": 0.9757070342931571, "1000–3000 Hz_10ms_modulation_mae_db": 0.8615904130162276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0692105641927334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09529470848415002, "1000–3000 Hz_40ms_envelope_correlation": 0.993299216262355, "1000–3000 Hz_40ms_modulation_mae_db": 0.5329468337892042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8894656527538551, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19045580209641266, "3000–6000 Hz_10ms_envelope_correlation": 0.9730892947734386, "3000–6000 Hz_10ms_modulation_mae_db": 0.6436854541297542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1634769624984524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09956473704767929, "3000–6000 Hz_40ms_envelope_correlation": 0.9926827618404437, "3000–6000 Hz_40ms_modulation_mae_db": 0.788757753691328, "mr_stft_db": 5.880500143651748, "band_mae_db": 0.3116135270404435, "rms_gain_error_db": -0.12888054911824226, "output_peak": 0.4754074637067898}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1946058711586964, "全帯域_10ms_rms_envelope_nrmse": 0.12483988874048077, "全帯域_10ms_envelope_correlation": 0.9856795418109324, "全帯域_10ms_modulation_mae_db": 0.9011495395494956, "全帯域_40ms_log_rms_mae_db": 0.6710456890756686, "全帯域_40ms_rms_envelope_nrmse": 0.06819346758470078, "全帯域_40ms_envelope_correlation": 0.9957143147001645, "全帯域_40ms_modulation_mae_db": 1.0115269165422924, "300–1000 Hz_10ms_log_rms_mae_db": 1.5482867145717938, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1559561819218875, "300–1000 Hz_10ms_envelope_correlation": 0.9807473196848414, "300–1000 Hz_10ms_modulation_mae_db": 1.3533449215250084, "300–1000 Hz_40ms_log_rms_mae_db": 1.0303609516925798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08937211740965884, "300–1000 Hz_40ms_envelope_correlation": 0.9934366555464925, "300–1000 Hz_40ms_modulation_mae_db": 0.843906892172118, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6383190702101802, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16425733737197817, "1000–3000 Hz_10ms_envelope_correlation": 0.9790275508009498, "1000–3000 Hz_10ms_modulation_mae_db": 0.6373742064530753, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0112217407817252, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08501038209173518, "1000–3000 Hz_40ms_envelope_correlation": 0.9943554906396872, "1000–3000 Hz_40ms_modulation_mae_db": 0.7112385169427803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9450059439510874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21139346051048505, "3000–6000 Hz_10ms_envelope_correlation": 0.9685481206243203, "3000–6000 Hz_10ms_modulation_mae_db": 0.8548531654352213, "3000–6000 Hz_40ms_log_rms_mae_db": 1.176104288308472, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1247973921111512, "3000–6000 Hz_40ms_envelope_correlation": 0.9887161045943729, "3000–6000 Hz_40ms_modulation_mae_db": 0.6416994991641127, "mr_stft_db": 5.891432651037026, "band_mae_db": 0.3003128730898494, "rms_gain_error_db": -0.1568328192147308, "output_peak": 0.3891365457300192}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1385594550662985, "全帯域_10ms_rms_envelope_nrmse": 0.12378553384873495, "全帯域_10ms_envelope_correlation": 0.9865890063858936, "全帯域_10ms_modulation_mae_db": 0.49317511299129607, "全帯域_40ms_log_rms_mae_db": 0.765987337953097, "全帯域_40ms_rms_envelope_nrmse": 0.07517953088130792, "全帯域_40ms_envelope_correlation": 0.9956057274873843, "全帯域_40ms_modulation_mae_db": 0.6747003771373036, "300–1000 Hz_10ms_log_rms_mae_db": 1.4822394649941502, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1582006908539525, "300–1000 Hz_10ms_envelope_correlation": 0.9805589381055377, "300–1000 Hz_10ms_modulation_mae_db": 1.6503696157380414, "300–1000 Hz_40ms_log_rms_mae_db": 0.9670382002206656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0990468151155914, "300–1000 Hz_40ms_envelope_correlation": 0.9926832380859832, "300–1000 Hz_40ms_modulation_mae_db": 0.9004546444460991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6592154799768455, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19116346418521102, "1000–3000 Hz_10ms_envelope_correlation": 0.9750713683412152, "1000–3000 Hz_10ms_modulation_mae_db": 1.1054797422746578, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0178689039020954, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12570539601911812, "1000–3000 Hz_40ms_envelope_correlation": 0.992414284479696, "1000–3000 Hz_40ms_modulation_mae_db": 0.9195627049579684, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9292808993293347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1956022264486144, "3000–6000 Hz_10ms_envelope_correlation": 0.9714542923481184, "3000–6000 Hz_10ms_modulation_mae_db": 0.9365597171056694, "3000–6000 Hz_40ms_log_rms_mae_db": 1.179730445873864, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11175857987806297, "3000–6000 Hz_40ms_envelope_correlation": 0.990230679294649, "3000–6000 Hz_40ms_modulation_mae_db": 0.5893529475768583, "mr_stft_db": 5.894696537607221, "band_mae_db": 0.5121064700179473, "rms_gain_error_db": -0.31817954597736386, "output_peak": 0.45911007421184985}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 57920, "output_samples": 57920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.172343533635325, "spectral_convergence": 0.40130124726676036, "spectral_convergence_status": "ok", "amplitude_floor": 1.9579592332350965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.590234501787052, "spectral_convergence": 0.3837924210215586, "spectral_convergence_status": "ok", "amplitude_floor": 1.206925486771671e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.959370032341218, "spectral_convergence": 0.41413604313065794, "spectral_convergence_status": "ok", "amplitude_floor": 6.395923898397892e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042562388767747, "output_power": 0.00039504395168052473, "delta_db": -0.3238057735742655, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001322032073723835, "output_power": 0.001299353169746352, "delta_db": -0.0751478144916175, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006256947892797981, "output_power": 0.0005468933909464567, "delta_db": -0.5845986315484404, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002997481206139804, "output_power": 0.00027439025140270606, "delta_db": -0.38387791204507726, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003961082609243847, "output_power": 0.0003637050895126796, "delta_db": -0.3706452140146729, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.769256628656105e-06, "output_power": 1.0148506672563277e-05, "delta_db": 1.1602267460415359, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.48305034861926827, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2729629708833871, "pitch": null, "output_rms_dbfs": -25.391721333727567, "output_peak": 0.48895101683767817, "reference_rms_dbfs": -25.11875836284418, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.907316022587865}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 57920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.00020370994921487513, "log_rms_mae_db": 1.114995272152549, "rms_envelope_nrmse": 0.1060988775617219, "envelope_correlation": 0.9902079269884586, "rms_gain_error_db": -0.27296325672708655, "reference_mean_power": 0.003081231895334411, "output_mean_power": 0.002893530464140216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 1.3803128421607645e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003568424263155193, "output_power": 0.000337594820458767, "reference_active": true, "delta_db": -0.24080708799619266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014048200587458577, "output_power": 9.86691435587228e-05, "reference_active": true, "delta_db": -1.5343934111416475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7025084230927585e-05, "output_power": 1.884986979089257e-05, "reference_active": true, "delta_db": 0.44219085334536223}], "active_band_count": 3, "active_band_mae_db": 0.7391304508277342}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.00013932647455664127, "log_rms_mae_db": 0.610440659673779, "rms_envelope_nrmse": 0.06680534479579985, "envelope_correlation": 0.9964840612217317, "rms_gain_error_db": -0.27296527579763497, "reference_mean_power": 0.0031070091895842884, "output_mean_power": 0.0029177361095119882}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 1.2870957120852427e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031478645736461515, "output_power": 0.0002964588743221451, "reference_active": true, "delta_db": -0.2605158479454869}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.928842911180188e-05, "output_power": 4.842428193880203e-05, "reference_active": true, "delta_db": -1.5559752575241155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4070650648317675e-06, "output_power": 1.0633977626526992e-06, "reference_active": true, "delta_db": -1.2161843809690234}], "active_band_count": 3, "active_band_mae_db": 1.0108918288128752}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.0001235612100218032, "log_rms_mae_db": 1.3946965117266994, "rms_envelope_nrmse": 0.1653808061947589, "envelope_correlation": 0.97858722311798, "rms_gain_error_db": -0.2897072181832289, "reference_mean_power": 0.0011039001556015875, "output_mean_power": 0.0010326640844659102}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 4.707425731708226e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018855725502559516, "output_power": 0.00014930976650953708, "reference_active": true, "delta_db": -1.0135503088527582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8855557263471545e-05, "output_power": 5.9335629754553085e-05, "reference_active": true, "delta_db": 0.035280799806755817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.265121532076037e-06, "output_power": 7.2068461759854225e-06, "reference_active": true, "delta_db": 2.2781384187103395}], "active_band_count": 3, "active_band_mae_db": 1.1089898424566178}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.0001123986305082509, "log_rms_mae_db": 0.8748730507310157, "rms_envelope_nrmse": 0.0986422476567982, "envelope_correlation": 0.9926985708193817, "rms_gain_error_db": -0.2897079128746043, "reference_mean_power": 0.0011131375427710934, "output_mean_power": 0.001041305204809909}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 4.401493945076042e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017713633289459933, "output_power": 0.0001421290601007037, "reference_active": true, "delta_db": -0.9562476554186939}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7898955892530673e-05, "output_power": 2.812449046207564e-05, "reference_active": true, "delta_db": 0.03496712699445244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0644420897956173e-07, "output_power": 3.9788896014276244e-07, "reference_active": true, "delta_db": -0.09239050755842991}], "active_band_count": 3, "active_band_mae_db": 0.3612017633238587}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 8.851421864498589e-05, "log_rms_mae_db": 1.6471252632908961, "rms_envelope_nrmse": 0.16017549681863574, "envelope_correlation": 0.9811178077548075, "rms_gain_error_db": -0.4512885312454395, "reference_mean_power": 0.000506505585091631, "output_mean_power": 0.00045651535050829763}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 2.2793822474038483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010569673324300439, "output_power": 8.858043158264855e-05, "reference_active": true, "delta_db": -0.7672377289563691}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5368696588833574e-05, "output_power": 1.646862910879462e-05, "reference_active": true, "delta_db": 0.3002041210809838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.905327294990813e-06, "output_power": 6.4083108923822366e-06, "reference_active": true, "delta_db": 2.1508613612924323}], "active_band_count": 3, "active_band_mae_db": 1.0727677371099285}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 8.08613786301264e-05, "log_rms_mae_db": 0.9915992846770609, "rms_envelope_nrmse": 0.08744347410705239, "envelope_correlation": 0.9956363682308518, "rms_gain_error_db": -0.4512935541669767, "reference_mean_power": 0.0005107405079369301, "output_mean_power": 0.0004603317696696442}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 2.1269882857538688e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.701418164823122e-05, "output_power": 8.38613235432717e-05, "reference_active": true, "delta_db": -0.632735122054755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.796784971798785e-06, "output_power": 7.376244600456839e-06, "reference_active": true, "delta_db": 0.3553177926309785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1410692673718103e-07, "output_power": 3.6648539235146955e-07, "reference_active": true, "delta_db": 2.3342595117310165}], "active_band_count": 3, "active_band_mae_db": 1.10743747547225}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 5.208111658419103e-05, "log_rms_mae_db": 1.9146319899565276, "rms_envelope_nrmse": 0.19928212085560337, "envelope_correlation": 0.9705701267594835, "rms_gain_error_db": -0.2642512113372013, "reference_mean_power": 0.0001411400109581116, "output_mean_power": 0.00013280824131472647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 8.582618674096986e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.483177751159981e-05, "output_power": 3.0411869189724892e-05, "reference_active": true, "delta_db": -0.5893252385249073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4244491686748372e-05, "output_power": 1.485940476246589e-05, "reference_active": true, "delta_db": 0.18354456722154916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3188700900305924e-06, "output_power": 3.2847603878708226e-06, "reference_active": true, "delta_db": 1.5122727588527252}], "active_band_count": 3, "active_band_mae_db": 0.7617141881997272}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 4.6774501347655295e-05, "log_rms_mae_db": 1.1916047572478627, "rms_envelope_nrmse": 0.11786867917552982, "envelope_correlation": 0.9894501788173858, "rms_gain_error_db": -0.2642656890831058, "reference_mean_power": 0.0001423195946220122, "output_mean_power": 0.00013391774542614358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 7.830084455082624e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.343144196598961e-05, "output_power": 3.0158428885386405e-05, "reference_active": true, "delta_db": -0.44746395945747447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.21519729141624e-06, "output_power": 8.080585511037082e-06, "reference_active": true, "delta_db": 0.4919461956906762}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.774546333615459e-07, "output_power": 2.757244695992499e-07, "reference_active": true, "delta_db": 1.9138796664695277}], "active_band_count": 3, "active_band_mae_db": 0.9510966072058927}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8038787165827951, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "484575bc1cb25c7954015c37d796f085d7105cf694b24130e742fe986d182569", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9769788950417434, "standard_deviation": 0.003085756631176134, "minimum": 0.9725922307271794, "maximum": 0.9811178077548075, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6521594176660734, "standard_deviation": 0.049358936935382805, "minimum": 1.5516083107571683, "maximum": 1.7242949449785245, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8681389903216592, "standard_deviation": 0.24133342639741473, "minimum": 0.44780334836317165, "maximum": 1.1060533885399055, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17492136261266011, "standard_deviation": 0.013319291268997332, "minimum": 0.15743380196761775, "maximum": 0.1920670092074813, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9925563626447991, "standard_deviation": 0.0028461325497523607, "minimum": 0.9863793081568883, "maximum": 0.9956363682308518, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.033197294743648, "standard_deviation": 0.04068622482622048, "minimum": 0.9748663667031089, "maximum": 1.0871660365069507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7882325409734534, "standard_deviation": 0.25041916600179454, "minimum": 0.5239375451173187, "maximum": 1.1468040899452625, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10190004611503609, "standard_deviation": 0.018576581530449294, "minimum": 0.08501038209173518, "maximum": 0.13317946310928702, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9708078766498272, "standard_deviation": 0.002253044482118214, "minimum": 0.9673434494471692, "maximum": 0.9740011579178433, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9321401455245537, "standard_deviation": 0.03397487382295775, "minimum": 1.8894656527538551, "maximum": 1.9907331339774945, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8229070516523778, "standard_deviation": 0.09587864813673515, "minimum": 0.6436854541297542, "maximum": 0.9369495710244761, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1992704255987621, "standard_deviation": 0.0082581254281262, "minimum": 0.19033784736575, "maximum": 0.21139346051048505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9901777031084764, "standard_deviation": 0.0014002768761212179, "minimum": 0.988487491123473, "maximum": 0.9926827618404437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1841452140470388, "standard_deviation": 0.03607896765815078, "minimum": 1.1189262501839898, "maximum": 1.2462600540688202, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7903473376059968, "standard_deviation": 0.15387328831096056, "minimum": 0.5893529475768583, "maximum": 1.0190551533252532, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11493430949344052, "standard_deviation": 0.009901653376405157, "minimum": 0.09956473704767929, "maximum": 0.127104882395121, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.979298169510924, "standard_deviation": 0.0033974574008556716, "minimum": 0.9724960342732255, "maximum": 0.9835462792780908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4947154400528564, "standard_deviation": 0.07378374660391766, "minimum": 1.3805821983265467, "maximum": 1.596154206402817, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.5590365386535479, "standard_deviation": 0.5011555711087281, "minimum": 0.9883147151076083, "maximum": 2.4167875811355337, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16200246723656309, "standard_deviation": 0.013234124395778184, "minimum": 0.14429337477734736, "maximum": 0.18717066851336506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9924411846893142, "standard_deviation": 0.00249257275353062, "minimum": 0.9878036337218644, "maximum": 0.994901288298202, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9513749318602143, "standard_deviation": 0.06860402477456388, "minimum": 0.8748730507310157, "maximum": 1.0616656619460358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9546729654190974, "standard_deviation": 0.43767422106044684, "minimum": 0.3612017633238587, "maximum": 1.5950209346692128, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09630143809993782, "standard_deviation": 0.014970828579915234, "minimum": 0.07968329795149533, "maximum": 0.12179654540804104, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3763042321989238, "standard_deviation": 0.08019759233442135, "minimum": 0.3003128730898494, "maximum": 0.5121064700179473, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.88726041941532, "standard_deviation": 0.010871632198782191, "minimum": 5.872637005157863, "maximum": 5.907316022587865, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47241433953056755, "standard_deviation": 0.038936464647608006, "minimum": 0.3891365457300192, "maximum": 0.517972335147211, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1279421303420547, "standard_deviation": 0.11849897810704021, "minimum": -0.31817954597736386, "maximum": 0.030065592356028945, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9878186109502624, "standard_deviation": 0.0021090765383021065, "minimum": 0.9849210785727031, "maximum": 0.9903568801036461, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1453124371064383, "standard_deviation": 0.031922979435217984, "minimum": 1.114995272152549, "maximum": 1.1953417829721678, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9059873558941336, "standard_deviation": 0.5704627824791956, "minimum": 0.34580554947569936, "maximum": 2.1278414474724876, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11561864895905613, "standard_deviation": 0.00984397848096579, "minimum": 0.10296719441942855, "maximum": 0.1283575085026532, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9962401411421885, "standard_deviation": 0.0005230348471963839, "minimum": 0.9956057274873843, "maximum": 0.9970298361452912, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6659296735260254, "standard_deviation": 0.06413965857022634, "minimum": 0.5660331323059691, "maximum": 0.765987337953097, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0497953729255132, "standard_deviation": 0.5901413804672889, "minimum": 0.5214921905315841, "maximum": 2.4357300301881657, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06449411641296647, "standard_deviation": 0.006216688594148038, "minimum": 0.05558472834504527, "maximum": 0.07517953088130792, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7127-75947-0011", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.001338243484497, "predicted_mos": 2.001338243484497, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "cc891ad891a0d66c23c42db2ac18b7ddc55fa180ece2f4955650277b8211682f", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.782246300019324, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1394813154872487, "全帯域_10ms_rms_envelope_nrmse": 0.11230788295526206, "全帯域_10ms_envelope_correlation": 0.9885023431828546, "全帯域_10ms_modulation_mae_db": 1.4050730501968662, "全帯域_40ms_log_rms_mae_db": 0.6695661134509083, "全帯域_40ms_rms_envelope_nrmse": 0.0630040013294494, "全帯域_40ms_envelope_correlation": 0.9961814920143609, "全帯域_40ms_modulation_mae_db": 1.2209947871225506, "300–1000 Hz_10ms_log_rms_mae_db": 1.4530733866819276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15955341443012908, "300–1000 Hz_10ms_envelope_correlation": 0.9798320672486919, "300–1000 Hz_10ms_modulation_mae_db": 1.6468992240704126, "300–1000 Hz_40ms_log_rms_mae_db": 0.9189190419968694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10420173485661645, "300–1000 Hz_40ms_envelope_correlation": 0.9911554743094553, "300–1000 Hz_40ms_modulation_mae_db": 0.8850583033571585, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6017687453497436, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1764555979344039, "1000–3000 Hz_10ms_envelope_correlation": 0.9759869545675798, "1000–3000 Hz_10ms_modulation_mae_db": 0.7221382441149359, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9909400378191344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10322821748544699, "1000–3000 Hz_40ms_envelope_correlation": 0.9914177792913462, "1000–3000 Hz_40ms_modulation_mae_db": 0.5708040959498321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8004278674735483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22840373382636317, "3000–6000 Hz_10ms_envelope_correlation": 0.9613370756628189, "3000–6000 Hz_10ms_modulation_mae_db": 0.9139955347066548, "3000–6000 Hz_40ms_log_rms_mae_db": 1.062504848581175, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12941178162276393, "3000–6000 Hz_40ms_envelope_correlation": 0.9866951315790637, "3000–6000 Hz_40ms_modulation_mae_db": 0.8400803592184941, "mr_stft_db": 5.865566765134214, "band_mae_db": 0.3149803221709182, "rms_gain_error_db": 0.021243447869255805, "output_peak": 0.47198819295493327}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1744488911293172, "全帯域_10ms_rms_envelope_nrmse": 0.11186426086743034, "全帯域_10ms_envelope_correlation": 0.9885815946220646, "全帯域_10ms_modulation_mae_db": 0.2916665248431177, "全帯域_40ms_log_rms_mae_db": 0.6850781891895196, "全帯域_40ms_rms_envelope_nrmse": 0.054303818243771375, "全帯域_40ms_envelope_correlation": 0.9974108013695326, "全帯域_40ms_modulation_mae_db": 0.5542642141035867, "300–1000 Hz_10ms_log_rms_mae_db": 1.5446477363054754, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16763227826981328, "300–1000 Hz_10ms_envelope_correlation": 0.9777197970655516, "300–1000 Hz_10ms_modulation_mae_db": 1.8374952154569886, "300–1000 Hz_40ms_log_rms_mae_db": 1.0010138585658677, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1000708056438667, "300–1000 Hz_40ms_envelope_correlation": 0.9917743306319425, "300–1000 Hz_40ms_modulation_mae_db": 0.8465870336532134, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5668522454274505, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16940825382155525, "1000–3000 Hz_10ms_envelope_correlation": 0.977748999315418, "1000–3000 Hz_10ms_modulation_mae_db": 1.2181728469627158, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9540161352987183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09951832215328708, "1000–3000 Hz_40ms_envelope_correlation": 0.9924312187276539, "1000–3000 Hz_40ms_modulation_mae_db": 0.838672272680376, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8963702628369459, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19667954195142745, "3000–6000 Hz_10ms_envelope_correlation": 0.9716561706144694, "3000–6000 Hz_10ms_modulation_mae_db": 0.8495782153063698, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1211426880417505, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11213184432626526, "3000–6000 Hz_40ms_envelope_correlation": 0.9901577142933267, "3000–6000 Hz_40ms_modulation_mae_db": 0.2543687423213504, "mr_stft_db": 5.8330643050808915, "band_mae_db": 0.3094524082854402, "rms_gain_error_db": -0.15972824803625219, "output_peak": 0.4203907621290476}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.100353620386209, "全帯域_10ms_rms_envelope_nrmse": 0.09197444625965713, "全帯域_10ms_envelope_correlation": 0.9922511659748074, "全帯域_10ms_modulation_mae_db": 1.0041089622106592, "全帯域_40ms_log_rms_mae_db": 0.6419487211338866, "全帯域_40ms_rms_envelope_nrmse": 0.044988706373553275, "全帯域_40ms_envelope_correlation": 0.9980390997945643, "全帯域_40ms_modulation_mae_db": 0.8748266894209936, "300–1000 Hz_10ms_log_rms_mae_db": 1.4406794772766252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13334541798798125, "300–1000 Hz_10ms_envelope_correlation": 0.9861534092937888, "300–1000 Hz_10ms_modulation_mae_db": 1.3373010094067421, "300–1000 Hz_40ms_log_rms_mae_db": 0.9278005243139681, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08270731809832488, "300–1000 Hz_40ms_envelope_correlation": 0.9944637166248683, "300–1000 Hz_40ms_modulation_mae_db": 0.8863836507503682, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5907529204499509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18166798122630945, "1000–3000 Hz_10ms_envelope_correlation": 0.974818846253293, "1000–3000 Hz_10ms_modulation_mae_db": 0.963817780576096, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9322248850445543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09306997406013783, "1000–3000 Hz_40ms_envelope_correlation": 0.9930934758623419, "1000–3000 Hz_40ms_modulation_mae_db": 0.6568126908525704, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8113310216868022, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2124713875739469, "3000–6000 Hz_10ms_envelope_correlation": 0.9663498397083227, "3000–6000 Hz_10ms_modulation_mae_db": 0.7298059089210004, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0418688920044492, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11160965597032861, "3000–6000 Hz_40ms_envelope_correlation": 0.9902546030234995, "3000–6000 Hz_40ms_modulation_mae_db": 0.5938341315546544, "mr_stft_db": 5.864409003538273, "band_mae_db": 0.28586239171126887, "rms_gain_error_db": 0.007889751202630137, "output_peak": 0.47191787214144926}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.099858947251375, "全帯域_10ms_rms_envelope_nrmse": 0.13958659495201206, "全帯域_10ms_envelope_correlation": 0.982543333244471, "全帯域_10ms_modulation_mae_db": 1.906352449927078, "全帯域_40ms_log_rms_mae_db": 0.6163600438962433, "全帯域_40ms_rms_envelope_nrmse": 0.08917558482090925, "全帯域_40ms_envelope_correlation": 0.9925077276773769, "全帯域_40ms_modulation_mae_db": 1.7377378344726855, "300–1000 Hz_10ms_log_rms_mae_db": 1.3934111499666015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14379025378515867, "300–1000 Hz_10ms_envelope_correlation": 0.983648036568848, "300–1000 Hz_10ms_modulation_mae_db": 1.251615277635522, "300–1000 Hz_40ms_log_rms_mae_db": 0.8859264274520376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07957327412173927, "300–1000 Hz_40ms_envelope_correlation": 0.9949899554436246, "300–1000 Hz_40ms_modulation_mae_db": 0.6238837188181968, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6125481568858204, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1762156072241423, "1000–3000 Hz_10ms_envelope_correlation": 0.9759495906325397, "1000–3000 Hz_10ms_modulation_mae_db": 0.9997987130870722, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9523451688776388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08662610004893931, "1000–3000 Hz_40ms_envelope_correlation": 0.9943544274477748, "1000–3000 Hz_40ms_modulation_mae_db": 0.8693974723908382, "3000–6000 Hz_10ms_log_rms_mae_db": 1.909883366103169, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22528497775070977, "3000–6000 Hz_10ms_envelope_correlation": 0.9648606332354404, "3000–6000 Hz_10ms_modulation_mae_db": 0.5866480840594491, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1015051973880345, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13329415476690765, "3000–6000 Hz_40ms_envelope_correlation": 0.9874795826373955, "3000–6000 Hz_40ms_modulation_mae_db": 0.1269804526760764, "mr_stft_db": 5.853257103420895, "band_mae_db": 0.5361372362440636, "rms_gain_error_db": 0.08284968715561002, "output_peak": 0.4873949477852467}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.127582599813946, "全帯域_10ms_rms_envelope_nrmse": 0.12111851091495991, "全帯域_10ms_envelope_correlation": 0.9865875328258816, "全帯域_10ms_modulation_mae_db": 0.5747761932031829, "全帯域_40ms_log_rms_mae_db": 0.6184324860395891, "全帯域_40ms_rms_envelope_nrmse": 0.05478244021292247, "全帯域_40ms_envelope_correlation": 0.9971142063003693, "全帯域_40ms_modulation_mae_db": 0.26269957371943037, "300–1000 Hz_10ms_log_rms_mae_db": 1.4137983239470542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16354543270313515, "300–1000 Hz_10ms_envelope_correlation": 0.9788504404148725, "300–1000 Hz_10ms_modulation_mae_db": 1.4774800456597015, "300–1000 Hz_40ms_log_rms_mae_db": 0.8915027499332081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09063169073646714, "300–1000 Hz_40ms_envelope_correlation": 0.9932404480965645, "300–1000 Hz_40ms_modulation_mae_db": 0.8041353907895514, "1000–3000 Hz_10ms_log_rms_mae_db": 1.625349707392565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17854849846634757, "1000–3000 Hz_10ms_envelope_correlation": 0.9753701413027157, "1000–3000 Hz_10ms_modulation_mae_db": 1.265894971029849, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9936668745223618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08847442832487488, "1000–3000 Hz_40ms_envelope_correlation": 0.9937263506831596, "1000–3000 Hz_40ms_modulation_mae_db": 0.9082976346766943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9055060678435032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19245850038610868, "3000–6000 Hz_10ms_envelope_correlation": 0.9726735245696267, "3000–6000 Hz_10ms_modulation_mae_db": 0.4712170022597959, "3000–6000 Hz_40ms_log_rms_mae_db": 1.078413875094878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10725339642544768, "3000–6000 Hz_40ms_envelope_correlation": 0.9910692448452793, "3000–6000 Hz_40ms_modulation_mae_db": 0.3241726049351464, "mr_stft_db": 5.844958052797142, "band_mae_db": 0.3011622480037473, "rms_gain_error_db": -0.0006593460367109685, "output_peak": 0.4829236431261705}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1421267434051328, "全帯域_10ms_rms_envelope_nrmse": 0.11981329738675511, "全帯域_10ms_envelope_correlation": 0.9869046519392582, "全帯域_10ms_modulation_mae_db": 0.9312036346871319, "全帯域_40ms_log_rms_mae_db": 0.6829977612496331, "全帯域_40ms_rms_envelope_nrmse": 0.07779652547574221, "全帯域_40ms_envelope_correlation": 0.9944376096412926, "全帯域_40ms_modulation_mae_db": 1.5788736650649469, "300–1000 Hz_10ms_log_rms_mae_db": 1.3818941085392922, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1411998855810886, "300–1000 Hz_10ms_envelope_correlation": 0.9842262237204485, "300–1000 Hz_10ms_modulation_mae_db": 1.228086132763704, "300–1000 Hz_40ms_log_rms_mae_db": 0.7891651672775483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07581837092263696, "300–1000 Hz_40ms_envelope_correlation": 0.9952947552856815, "300–1000 Hz_40ms_modulation_mae_db": 0.712649003856665, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6366554959871809, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1535529980140979, "1000–3000 Hz_10ms_envelope_correlation": 0.9818432180345483, "1000–3000 Hz_10ms_modulation_mae_db": 1.1527072458233196, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9292920150179599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07391619635320726, "1000–3000 Hz_40ms_envelope_correlation": 0.9955969928535878, "1000–3000 Hz_40ms_modulation_mae_db": 1.082671166528775, "3000–6000 Hz_10ms_log_rms_mae_db": 1.940631506652804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19499256861910366, "3000–6000 Hz_10ms_envelope_correlation": 0.9733488119703687, "3000–6000 Hz_10ms_modulation_mae_db": 0.8821464124564035, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1535148840945115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10808658231044724, "3000–6000 Hz_40ms_envelope_correlation": 0.9938378707013208, "3000–6000 Hz_40ms_modulation_mae_db": 0.4727591891882102, "mr_stft_db": 5.861364182570359, "band_mae_db": 0.38293342996207097, "rms_gain_error_db": -0.18697398724595676, "output_peak": 0.4051627201950289}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2002726444007703, "全帯域_10ms_rms_envelope_nrmse": 0.11550284560697681, "全帯域_10ms_envelope_correlation": 0.9878896127112806, "全帯域_10ms_modulation_mae_db": 0.7430515752976211, "全帯域_40ms_log_rms_mae_db": 0.7747918137047937, "全帯域_40ms_rms_envelope_nrmse": 0.05917751734235571, "全帯域_40ms_envelope_correlation": 0.9966596937885909, "全帯域_40ms_modulation_mae_db": 0.8101611006311819, "300–1000 Hz_10ms_log_rms_mae_db": 1.4173223326090554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15288412822466352, "300–1000 Hz_10ms_envelope_correlation": 0.9820240820710617, "300–1000 Hz_10ms_modulation_mae_db": 1.1672582565893561, "300–1000 Hz_40ms_log_rms_mae_db": 0.8832306144022498, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09213303651020524, "300–1000 Hz_40ms_envelope_correlation": 0.9932574454706745, "300–1000 Hz_40ms_modulation_mae_db": 0.767201871480129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5433036291497566, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15893985116884363, "1000–3000 Hz_10ms_envelope_correlation": 0.9820299177476403, "1000–3000 Hz_10ms_modulation_mae_db": 0.5154264220983605, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9659464302729053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09390230056404004, "1000–3000 Hz_40ms_envelope_correlation": 0.9954256654223457, "1000–3000 Hz_40ms_modulation_mae_db": 0.3810902559614107, "3000–6000 Hz_10ms_log_rms_mae_db": 1.91520547488204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21481277148140657, "3000–6000 Hz_10ms_envelope_correlation": 0.9659624182752741, "3000–6000 Hz_10ms_modulation_mae_db": 0.23770307352040346, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0472627022277514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09755393149687475, "3000–6000 Hz_40ms_envelope_correlation": 0.9924689083722766, "3000–6000 Hz_40ms_modulation_mae_db": 0.08689318452729905, "mr_stft_db": 5.831557071777169, "band_mae_db": 0.45736939898921447, "rms_gain_error_db": 0.04265245045329955, "output_peak": 0.47687104271144587}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2093057384727366, "全帯域_10ms_rms_envelope_nrmse": 0.17244046453069733, "全帯域_10ms_envelope_correlation": 0.9733865524904224, "全帯域_10ms_modulation_mae_db": 1.7531783122310527, "全帯域_40ms_log_rms_mae_db": 0.7053148241764368, "全帯域_40ms_rms_envelope_nrmse": 0.0890546428291473, "全帯域_40ms_envelope_correlation": 0.9926010213832883, "全帯域_40ms_modulation_mae_db": 1.9790284509630027, "300–1000 Hz_10ms_log_rms_mae_db": 1.459093103117145, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17542172328437688, "300–1000 Hz_10ms_envelope_correlation": 0.976164948729221, "300–1000 Hz_10ms_modulation_mae_db": 1.2219570596634106, "300–1000 Hz_40ms_log_rms_mae_db": 0.9043737177433938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10565434133788357, "300–1000 Hz_40ms_envelope_correlation": 0.9909371027360393, "300–1000 Hz_40ms_modulation_mae_db": 0.39080142177255217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5749214653824901, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16406493370402592, "1000–3000 Hz_10ms_envelope_correlation": 0.9791845825691161, "1000–3000 Hz_10ms_modulation_mae_db": 0.8991709334490375, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9719792865602704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08989066248442588, "1000–3000 Hz_40ms_envelope_correlation": 0.9936736922485455, "1000–3000 Hz_40ms_modulation_mae_db": 0.5636545942293947, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8980331633796705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2392969934701358, "3000–6000 Hz_10ms_envelope_correlation": 0.9575538353596915, "3000–6000 Hz_10ms_modulation_mae_db": 0.8465122412474889, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1125298008075597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1209802335723937, "3000–6000 Hz_40ms_envelope_correlation": 0.9883878102434871, "3000–6000 Hz_40ms_modulation_mae_db": 0.48571811475898247, "mr_stft_db": 5.86725987932376, "band_mae_db": 0.3339085637668981, "rms_gain_error_db": 0.10024763024562092, "output_peak": 0.49281823466465974}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 57920, "output_samples": 57920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.143905667217532, "spectral_convergence": 0.4198455828263042, "spectral_convergence_status": "ok", "amplitude_floor": 1.9579592332350965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.542154610064972, "spectral_convergence": 0.39963775317982203, "spectral_convergence_status": "ok", "amplitude_floor": 1.206925486771671e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.910640018120137, "spectral_convergence": 0.42078784712383266, "spectral_convergence_status": "ok", "amplitude_floor": 6.395923898397892e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042562388767747, "output_power": 0.0004646737822422267, "delta_db": 0.3812217519274364, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001322032073723835, "output_power": 0.0013588449950541553, "delta_db": 0.11927927443539962, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006256947892797981, "output_power": 0.0005714012508278844, "delta_db": -0.39421351492346246, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002997481206139804, "output_power": 0.00029989366399528944, "delta_db": 0.0021082149457777705, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003961082609243847, "output_power": 0.00038769272378649106, "delta_db": -0.09326249556296669, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.769256628656105e-06, "output_power": 9.557826334168463e-06, "delta_db": 0.8997966812304664, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3149803221709182, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.021243447869255805, "pitch": null, "output_rms_dbfs": -25.097514914974923, "output_peak": 0.47198819295493327, "reference_rms_dbfs": -25.11875836284418, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.865566765134214}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 57920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.00020370994921487513, "log_rms_mae_db": 1.1394813154872487, "rms_envelope_nrmse": 0.11230788295526206, "envelope_correlation": 0.9885023431828546, "rms_gain_error_db": 0.021243649981954955, "reference_mean_power": 0.003081231895334411, "output_mean_power": 0.003096340759828729}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 1.3803128421607645e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003568424263155193, "output_power": 0.00036362631649109026, "reference_active": true, "delta_db": 0.08178822948996071}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014048200587458577, "output_power": 0.0001583465371987713, "reference_active": true, "delta_db": 0.5198787075589272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7025084230927585e-05, "output_power": 3.91241117400838e-05, "reference_active": true, "delta_db": 3.6135522135417113}], "active_band_count": 3, "active_band_mae_db": 1.4050730501968662}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.00013932647455664127, "log_rms_mae_db": 0.6695661134509083, "rms_envelope_nrmse": 0.0630040013294494, "envelope_correlation": 0.9961814920143609, "rms_gain_error_db": 0.021243360440922204, "reference_mean_power": 0.0031070091895842884, "output_mean_power": 0.003122244245250383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 1.2870957120852427e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031478645736461515, "output_power": 0.0003301148717834663, "reference_active": true, "delta_db": 0.2064904986002027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.928842911180188e-05, "output_power": 9.000228515805408e-05, "reference_active": true, "delta_db": 1.1359282107633335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4070650648317675e-06, "output_power": 2.4008816126036198e-06, "reference_active": true, "delta_db": 2.320565652004116}], "active_band_count": 3, "active_band_mae_db": 1.2209947871225506}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.0001235612100218032, "log_rms_mae_db": 1.4530733866819276, "rms_envelope_nrmse": 0.15955341443012908, "envelope_correlation": 0.9798320672486919, "rms_gain_error_db": -0.15538190483287007, "reference_mean_power": 0.0011039001556015875, "output_mean_power": 0.0010651029905413777}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 4.707425731708226e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018855725502559516, "output_power": 0.00015181725508541807, "reference_active": true, "delta_db": -0.9412211225139421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8855557263471545e-05, "output_power": 4.8573466797796025e-05, "reference_active": true, "delta_db": -0.8338837494378027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.265121532076037e-06, "output_power": 8.840782183093315e-06, "reference_active": true, "delta_db": 3.1655928002594935}], "active_band_count": 3, "active_band_mae_db": 1.6468992240704126}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.0001123986305082509, "log_rms_mae_db": 0.9189190419968694, "rms_envelope_nrmse": 0.10420173485661645, "envelope_correlation": 0.9911554743094553, "rms_gain_error_db": -0.15538201388756218, "reference_mean_power": 0.0011131375427710934, "output_mean_power": 0.001074015697798117}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 4.401493945076042e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017713633289459933, "output_power": 0.00014135083026833284, "reference_active": true, "delta_db": -0.9800928580171859}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7898955892530673e-05, "output_power": 2.2846290228849025e-05, "reference_active": true, "delta_db": -0.8677226070468614}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0644420897956173e-07, "output_power": 4.894821788979435e-07, "reference_active": true, "delta_db": 0.8073594450074283}], "active_band_count": 3, "active_band_mae_db": 0.8850583033571585}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 8.851421864498589e-05, "log_rms_mae_db": 1.6017687453497436, "rms_envelope_nrmse": 0.1764555979344039, "envelope_correlation": 0.9759869545675798, "rms_gain_error_db": -0.04233065387843799, "reference_mean_power": 0.000506505585091631, "output_mean_power": 0.0005015926605842743}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 2.2793822474038483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010569673324300439, "output_power": 9.70751353004371e-05, "reference_active": true, "delta_db": -0.3695356029910893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5368696588833574e-05, "output_power": 1.697616557364288e-05, "reference_active": true, "delta_db": 0.43202565610145727}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.905327294990813e-06, "output_power": 5.347401383952891e-06, "reference_active": true, "delta_db": 1.364853473252261}], "active_band_count": 3, "active_band_mae_db": 0.7221382441149359}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 8.08613786301264e-05, "log_rms_mae_db": 0.9909400378191344, "rms_envelope_nrmse": 0.10322821748544699, "envelope_correlation": 0.9914177792913462, "rms_gain_error_db": -0.042323306660966856, "reference_mean_power": 0.0005107405079369301, "output_mean_power": 0.0005057873618499598}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 2.1269882857538688e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.701418164823122e-05, "output_power": 9.153083731483307e-05, "reference_active": true, "delta_db": -0.25267789327591395}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.796784971798785e-06, "output_power": 8.020141913628507e-06, "reference_active": true, "delta_db": 0.7187852256959364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1410692673718103e-07, "output_power": 2.5393679538265966e-07, "reference_active": true, "delta_db": 0.7409491688776457}], "active_band_count": 3, "active_band_mae_db": 0.5708040959498321}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 5.208111658419103e-05, "log_rms_mae_db": 1.8004278674735483, "rms_envelope_nrmse": 0.22840373382636317, "envelope_correlation": 0.9613370756628189, "rms_gain_error_db": -0.04516737764586588, "reference_mean_power": 0.0001411400109581116, "output_mean_power": 0.00013967973713306154}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 8.582618674096986e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.483177751159981e-05, "output_power": 3.203468837178381e-05, "reference_active": true, "delta_db": -0.3635513404197996}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4244491686748372e-05, "output_power": 1.78657265407538e-05, "reference_active": true, "delta_db": 0.9837372613872672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3188700900305924e-06, "output_power": 3.197026374524382e-06, "reference_active": true, "delta_db": 1.3946980023128974}], "active_band_count": 3, "active_band_mae_db": 0.9139955347066548}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 4.6774501347655295e-05, "log_rms_mae_db": 1.062504848581175, "rms_envelope_nrmse": 0.12941178162276393, "envelope_correlation": 0.9866951315790637, "rms_gain_error_db": -0.04521015994603897, "reference_mean_power": 0.0001423195946220122, "output_mean_power": 0.00014084572901964702}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 7.830084455082624e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.343144196598961e-05, "output_power": 3.204188220126297e-05, "reference_active": true, "delta_db": -0.1843708955202228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.21519729141624e-06, "output_power": 9.528878263720761e-06, "reference_active": true, "delta_db": 1.2079356776622565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.774546333615459e-07, "output_power": 2.300810198564197e-07, "reference_active": true, "delta_db": 1.1279345044730031}], "active_band_count": 3, "active_band_mae_db": 0.8400803592184941}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8070441213267848, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ef8c557894a6e9b903ca4c4d154b9bf13461066e08eb0a3ae1391117c771486c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9778665313028564, "standard_deviation": 0.002869213789540534, "minimum": 0.974818846253293, "maximum": 0.9820299177476403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5940190457531198, "standard_deviation": 0.031369541009960096, "minimum": 1.5433036291497566, "maximum": 1.6366554959871809, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9671408946426733, "standard_deviation": 0.25536407209623, "minimum": 0.5154264220983605, "maximum": 1.265894971029849, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16985671519496576, "standard_deviation": 0.010126124822657095, "minimum": 0.1535529980140979, "maximum": 0.18166798122630945, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937149503170944, "standard_deviation": 0.001423807044055265, "minimum": 0.9914177792913462, "maximum": 0.9955969928535878, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9613013541766929, "standard_deviation": 0.024097717636479592, "minimum": 0.9292920150179599, "maximum": 0.9936668745223618, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7339250229087364, "standard_deviation": 0.22909561059102546, "minimum": 0.3810902559614107, "maximum": 1.082671166528775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0910782751842949, "standard_deviation": 0.00889098042279883, "minimum": 0.07391619635320726, "maximum": 0.10322821748544699, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9667177886745015, "standard_deviation": 0.005617752299178801, "minimum": 0.9575538353596915, "maximum": 0.9733488119703687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8846735913573103, "standard_deviation": 0.05060318146684563, "minimum": 1.8004278674735483, "maximum": 1.940631506652804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6897008090596958, "standard_deviation": 0.23957088257489656, "minimum": 0.23770307352040346, "maximum": 0.9139955347066548, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21305005938240024, "standard_deviation": 0.01730339036119923, "minimum": 0.19245850038610868, "maximum": 0.2392969934701358, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900438582119562, "standard_deviation": 0.002447043917469811, "minimum": 0.9866951315790637, "maximum": 0.9938378707013208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0898428610300137, "standard_deviation": 0.03905504044612331, "minimum": 1.0418688920044492, "maximum": 1.1535148840945115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.39810084739752666, "standard_deviation": 0.25167690224387995, "minimum": 0.08689318452729905, "maximum": 0.8400803592184941, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1150401975614286, "standard_deviation": 0.012012000439573301, "minimum": 0.09755393149687475, "maximum": 0.13329415476690765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9810773756390605, "standard_deviation": 0.003486734413450453, "minimum": 0.976164948729221, "maximum": 0.9861534092937888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4379899523053972, "standard_deviation": 0.05099351232052696, "minimum": 1.3818941085392922, "maximum": 1.5446477363054754, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3960115276557297, "standard_deviation": 0.23892878774913243, "minimum": 1.1672582565893561, "maximum": 1.8374952154569886, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1546715667832933, "standard_deviation": 0.014437133153703704, "minimum": 0.13334541798798125, "maximum": 0.17542172328437688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9931391535748564, "standard_deviation": 0.001711205093726571, "minimum": 0.9909371027360393, "maximum": 0.9952947552856815, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9002415127106429, "standard_deviation": 0.058784233606906953, "minimum": 0.7891651672775483, "maximum": 1.0010138585658677, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7395875493097293, "standard_deviation": 0.16684189450906783, "minimum": 0.39080142177255217, "maximum": 0.8863836507503682, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09134882152846752, "standard_deviation": 0.011348892639118145, "minimum": 0.07581837092263696, "maximum": 0.10565434133788357, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.36522574989170276, "standard_deviation": 0.08866748466121384, "minimum": 0.28586239171126887, "maximum": 0.5361372362440636, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.852679545455338, "standard_deviation": 0.014553061763332554, "minimum": 5.831557071777169, "maximum": 5.86725987932376, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46368342696349774, "standard_deviation": 0.0324936818941691, "minimum": 0.4051627201950289, "maximum": 0.49281823466465974, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.011559826799062938, "standard_deviation": 0.10599879036527642, "minimum": -0.18697398724595676, "maximum": 0.10024763024562092, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9858308483738801, "standard_deviation": 0.005701583230598131, "minimum": 0.9733865524904224, "maximum": 0.9922511659748074, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.149178812543342, "standard_deviation": 0.04190281576196881, "minimum": 1.099858947251375, "maximum": 1.2093057384727366, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0761763378245888, "standard_deviation": 0.5681722259324502, "minimum": 0.2916665248431177, "maximum": 1.906352449927078, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.12307603793421884, "standard_deviation": 0.023875396324342126, "minimum": 0.09197444625965713, "maximum": 0.17244046453069733, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9956189564961719, "standard_deviation": 0.002167500582730521, "minimum": 0.9925077276773769, "maximum": 0.9980390997945643, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6743112441051262, "standard_deviation": 0.051823293021919896, "minimum": 0.6163600438962433, "maximum": 0.7747918137047937, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1273232894372973, "standard_deviation": 0.6036788350014003, "minimum": 0.26269957371943037, "maximum": 1.9790284509630027, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.06653540457848137, "standard_deviation": 0.01675504226590239, "minimum": 0.044988706373553275, "maximum": 0.08917558482090925, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7176-88083-0002", "language": "en", "style": "normal", "speaker": "librispeech_7176", "source_utterance_id": "7176-88083-0002", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7176-88083-0002.wav", "source_file_sha256": "a4a4a8b65fa5cee2a1f4c98fda46d5c91168bb11e01acbcfd0a6badf80bf0d8a", "decoded_pcm_sha256": "22dedd61331f17693761f502975e31bf01741aced72867d6ade55e38b2e09d33", "original_source_sha256": "3c042266b649c153544bdb4b6c17ac6faf27e9a2b589a3d673225c7ea91180cf", "corpus_entry_id": "librispeech_7176-88083-0002"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.166062831878662, "predicted_mos": 4.166062831878662, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "825749ad1963b2c54ac9985616177c60eaf18eac859b4852d8b4ba9d6141470f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.624037371999293, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7176-88083-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "825749ad1963b2c54ac9985616177c60eaf18eac859b4852d8b4ba9d6141470f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "6ce789c9d4e40824a7597a188ba764b0912cc3d4dceaf7c8ac7aeb28469d4407"}, "rows": [{"case": "librispeech_7176-88083-0002", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.9633100032806396, "predicted_mos": 2.9633100032806396, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1e4ce4975c6d2bae4645f3ff73662ef25f1017102efe69a4291490cb6036e183", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2743088569986867, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7176-88083-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1e4ce4975c6d2bae4645f3ff73662ef25f1017102efe69a4291490cb6036e183", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9916239927482712, "全帯域_10ms_rms_envelope_nrmse": 0.07536986515436682, "全帯域_10ms_envelope_correlation": 0.9940743766623579, "全帯域_10ms_modulation_mae_db": 0.11344900299175953, "全帯域_40ms_log_rms_mae_db": 0.5647600886072333, "全帯域_40ms_rms_envelope_nrmse": 0.046280983246517235, "全帯域_40ms_envelope_correlation": 0.9975700177055397, "全帯域_40ms_modulation_mae_db": 0.2921725257539143, "300–1000 Hz_10ms_log_rms_mae_db": 1.6788067391280934, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1247573348291501, "300–1000 Hz_10ms_envelope_correlation": 0.9870895327929452, "300–1000 Hz_10ms_modulation_mae_db": 0.5928910085755784, "300–1000 Hz_40ms_log_rms_mae_db": 1.0216071197765597, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07199222847047462, "300–1000 Hz_40ms_envelope_correlation": 0.995535504679353, "300–1000 Hz_40ms_modulation_mae_db": 0.7743104545633273, "1000–3000 Hz_10ms_log_rms_mae_db": 1.990776603740161, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19796972856137965, "1000–3000 Hz_10ms_envelope_correlation": 0.9682446922460031, "1000–3000 Hz_10ms_modulation_mae_db": 1.0217236078573455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2122002131510108, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10743774883353348, "1000–3000 Hz_40ms_envelope_correlation": 0.9909089146860234, "1000–3000 Hz_40ms_modulation_mae_db": 1.0892693435527747, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3750784619818748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21228832948479165, "3000–6000 Hz_10ms_envelope_correlation": 0.9682258607698793, "3000–6000 Hz_10ms_modulation_mae_db": 0.6754303697951552, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6983914987071371, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15208886739127658, "3000–6000 Hz_40ms_envelope_correlation": 0.9821887933315253, "3000–6000 Hz_40ms_modulation_mae_db": 0.8205342344243762, "mr_stft_db": 5.793660698582452, "band_mae_db": 0.15243830840781716, "rms_gain_error_db": -0.09251933505921982, "output_peak": 0.6546100276042567}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0295683202824177, "全帯域_10ms_rms_envelope_nrmse": 0.07444602592243078, "全帯域_10ms_envelope_correlation": 0.9941847356558723, "全帯域_10ms_modulation_mae_db": 0.3434158525509979, "全帯域_40ms_log_rms_mae_db": 0.5535837166727049, "全帯域_40ms_rms_envelope_nrmse": 0.040991677395955294, "全帯域_40ms_envelope_correlation": 0.9980830879968697, "全帯域_40ms_modulation_mae_db": 0.6872521841321086, "300–1000 Hz_10ms_log_rms_mae_db": 1.7434049766954176, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13832341015554797, "300–1000 Hz_10ms_envelope_correlation": 0.9841813229129986, "300–1000 Hz_10ms_modulation_mae_db": 0.6087282259756327, "300–1000 Hz_40ms_log_rms_mae_db": 1.0672098787434903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08616123197778014, "300–1000 Hz_40ms_envelope_correlation": 0.9937351370928967, "300–1000 Hz_40ms_modulation_mae_db": 1.1700786624430963, "1000–3000 Hz_10ms_log_rms_mae_db": 2.03205241899531, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20134811248223689, "1000–3000 Hz_10ms_envelope_correlation": 0.9667449208250245, "1000–3000 Hz_10ms_modulation_mae_db": 0.9857113960792191, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2463068725713933, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11022457509547845, "1000–3000 Hz_40ms_envelope_correlation": 0.9889738170528393, "1000–3000 Hz_40ms_modulation_mae_db": 0.9642087018009065, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3127791930223753, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2324772031193097, "3000–6000 Hz_10ms_envelope_correlation": 0.961944472655798, "3000–6000 Hz_10ms_modulation_mae_db": 0.5723564746486826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.642932113107352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16258073988230243, "3000–6000 Hz_40ms_envelope_correlation": 0.9796734646845218, "3000–6000 Hz_40ms_modulation_mae_db": 0.5212062297303649, "mr_stft_db": 5.797629177630154, "band_mae_db": 0.1311954521403715, "rms_gain_error_db": -0.06623478563702008, "output_peak": 0.6200520832466695}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.050348116746462, "全帯域_10ms_rms_envelope_nrmse": 0.08914293541234519, "全帯域_10ms_envelope_correlation": 0.9916576491118282, "全帯域_10ms_modulation_mae_db": 0.5002471680553815, "全帯域_40ms_log_rms_mae_db": 0.5927525440293279, "全帯域_40ms_rms_envelope_nrmse": 0.05054867516645408, "全帯域_40ms_envelope_correlation": 0.9970855656487009, "全帯域_40ms_modulation_mae_db": 0.6440669348263366, "300–1000 Hz_10ms_log_rms_mae_db": 1.7047171403151269, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13636779241510003, "300–1000 Hz_10ms_envelope_correlation": 0.9847107107296686, "300–1000 Hz_10ms_modulation_mae_db": 0.6511540168602621, "300–1000 Hz_40ms_log_rms_mae_db": 1.0788220762235876, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09390363044047553, "300–1000 Hz_40ms_envelope_correlation": 0.992439883055666, "300–1000 Hz_40ms_modulation_mae_db": 0.9499030151810698, "1000–3000 Hz_10ms_log_rms_mae_db": 2.057738756836606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22061801006294202, "1000–3000 Hz_10ms_envelope_correlation": 0.9603312107578975, "1000–3000 Hz_10ms_modulation_mae_db": 1.4259975643996103, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2761678782275878, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15694299863265265, "1000–3000 Hz_40ms_envelope_correlation": 0.9779407183915442, "1000–3000 Hz_40ms_modulation_mae_db": 1.2329951042196525, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4286829466761146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23559698558535352, "3000–6000 Hz_10ms_envelope_correlation": 0.9602335533194132, "3000–6000 Hz_10ms_modulation_mae_db": 0.9829381346559769, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7485621947693704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16203061018667808, "3000–6000 Hz_40ms_envelope_correlation": 0.9793521193348421, "3000–6000 Hz_40ms_modulation_mae_db": 1.2744633980598872, "mr_stft_db": 5.7976221431178665, "band_mae_db": 0.13346248639601963, "rms_gain_error_db": -0.09907790555106223, "output_peak": 0.6275242007057816}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.050420862018648, "全帯域_10ms_rms_envelope_nrmse": 0.07279951423194767, "全帯域_10ms_envelope_correlation": 0.9944469715656237, "全帯域_10ms_modulation_mae_db": 0.7311412006867802, "全帯域_40ms_log_rms_mae_db": 0.5572846854664314, "全帯域_40ms_rms_envelope_nrmse": 0.03694479899760593, "全帯域_40ms_envelope_correlation": 0.9984507497719457, "全帯域_40ms_modulation_mae_db": 0.944297249461898, "300–1000 Hz_10ms_log_rms_mae_db": 1.7631922190620912, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1142471958039095, "300–1000 Hz_10ms_envelope_correlation": 0.9893091923488533, "300–1000 Hz_10ms_modulation_mae_db": 0.45497134108180887, "300–1000 Hz_40ms_log_rms_mae_db": 1.0584365896826036, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06470497349323909, "300–1000 Hz_40ms_envelope_correlation": 0.9965824488579705, "300–1000 Hz_40ms_modulation_mae_db": 0.5983826779985131, "1000–3000 Hz_10ms_log_rms_mae_db": 2.022615998044689, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19667629527563715, "1000–3000 Hz_10ms_envelope_correlation": 0.9682653245071902, "1000–3000 Hz_10ms_modulation_mae_db": 1.1858404746170923, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1584374023054493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10512184911787165, "1000–3000 Hz_40ms_envelope_correlation": 0.9903517631892332, "1000–3000 Hz_40ms_modulation_mae_db": 1.019590208420049, "3000–6000 Hz_10ms_log_rms_mae_db": 2.373794240214399, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24370873586316283, "3000–6000 Hz_10ms_envelope_correlation": 0.9577384588960905, "3000–6000 Hz_10ms_modulation_mae_db": 0.5462375043934364, "3000–6000 Hz_40ms_log_rms_mae_db": 1.769582935871235, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1786655044807194, "3000–6000 Hz_40ms_envelope_correlation": 0.975064921567436, "3000–6000 Hz_40ms_modulation_mae_db": 0.5862284148255703, "mr_stft_db": 5.7907743569175585, "band_mae_db": 0.11635633150550852, "rms_gain_error_db": -0.06702724937657766, "output_peak": 0.6268727283224683}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0579148301855055, "全帯域_10ms_rms_envelope_nrmse": 0.07702075879173972, "全帯域_10ms_envelope_correlation": 0.9937955331262951, "全帯域_10ms_modulation_mae_db": 0.4396190926846309, "全帯域_40ms_log_rms_mae_db": 0.5877637826555477, "全帯域_40ms_rms_envelope_nrmse": 0.04202814616844212, "全帯域_40ms_envelope_correlation": 0.9980143329500479, "全帯域_40ms_modulation_mae_db": 0.63610505300994, "300–1000 Hz_10ms_log_rms_mae_db": 1.835055620370222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13144312288694587, "300–1000 Hz_10ms_envelope_correlation": 0.9855460157396195, "300–1000 Hz_10ms_modulation_mae_db": 0.38809555409316676, "300–1000 Hz_40ms_log_rms_mae_db": 1.085800979432717, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07600334458372464, "300–1000 Hz_40ms_envelope_correlation": 0.9948952754613083, "300–1000 Hz_40ms_modulation_mae_db": 0.7676508533314008, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9852775783223238, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17892976631738494, "1000–3000 Hz_10ms_envelope_correlation": 0.9747682014383859, "1000–3000 Hz_10ms_modulation_mae_db": 1.0261648853980065, "1000–3000 Hz_40ms_log_rms_mae_db": 1.215281213570066, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10236934558605768, "1000–3000 Hz_40ms_envelope_correlation": 0.9924828271248672, "1000–3000 Hz_40ms_modulation_mae_db": 1.0906003591866786, "3000–6000 Hz_10ms_log_rms_mae_db": 2.363400338440286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23938827477917657, "3000–6000 Hz_10ms_envelope_correlation": 0.9601517330067691, "3000–6000 Hz_10ms_modulation_mae_db": 0.6342672472415135, "3000–6000 Hz_40ms_log_rms_mae_db": 1.718158259241121, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17033269812414692, "3000–6000 Hz_40ms_envelope_correlation": 0.9780999844162152, "3000–6000 Hz_40ms_modulation_mae_db": 1.0091828539514234, "mr_stft_db": 5.8092349157681396, "band_mae_db": 0.2108918012307345, "rms_gain_error_db": -0.08878228835201901, "output_peak": 0.6051738021148055}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0037084239441718, "全帯域_10ms_rms_envelope_nrmse": 0.07654814862057842, "全帯域_10ms_envelope_correlation": 0.9938254883708595, "全帯域_10ms_modulation_mae_db": 0.13519439497924532, "全帯域_40ms_log_rms_mae_db": 0.5416136632269436, "全帯域_40ms_rms_envelope_nrmse": 0.04408933312193431, "全帯域_40ms_envelope_correlation": 0.9977267497569817, "全帯域_40ms_modulation_mae_db": 0.22483272616437044, "300–1000 Hz_10ms_log_rms_mae_db": 1.7900541564747054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1336620519550536, "300–1000 Hz_10ms_envelope_correlation": 0.9849963637328752, "300–1000 Hz_10ms_modulation_mae_db": 0.4039071796939678, "300–1000 Hz_40ms_log_rms_mae_db": 1.0919899823413632, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07315554348771405, "300–1000 Hz_40ms_envelope_correlation": 0.9950628453556161, "300–1000 Hz_40ms_modulation_mae_db": 0.5318460321179074, "1000–3000 Hz_10ms_log_rms_mae_db": 2.020294933648681, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2114785677517045, "1000–3000 Hz_10ms_envelope_correlation": 0.9635856923694498, "1000–3000 Hz_10ms_modulation_mae_db": 0.7956015517853153, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2429873619825513, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12304129543763798, "1000–3000 Hz_40ms_envelope_correlation": 0.9862035488980073, "1000–3000 Hz_40ms_modulation_mae_db": 0.8069207487377111, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4066919953668187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23749056598690443, "3000–6000 Hz_10ms_envelope_correlation": 0.9585679800515524, "3000–6000 Hz_10ms_modulation_mae_db": 1.1146151514242344, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7474648201601817, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17577367451471004, "3000–6000 Hz_40ms_envelope_correlation": 0.975506377999531, "3000–6000 Hz_40ms_modulation_mae_db": 1.3849152722255162, "mr_stft_db": 5.804699694272114, "band_mae_db": 0.059342345823167005, "rms_gain_error_db": -0.0375412774262993, "output_peak": 0.6365135489250192}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0208394366797584, "全帯域_10ms_rms_envelope_nrmse": 0.07480501374195547, "全帯域_10ms_envelope_correlation": 0.9940963995072977, "全帯域_10ms_modulation_mae_db": 0.22941977659576251, "全帯域_40ms_log_rms_mae_db": 0.5318505612936432, "全帯域_40ms_rms_envelope_nrmse": 0.038832036354334964, "全帯域_40ms_envelope_correlation": 0.998213240294825, "全帯域_40ms_modulation_mae_db": 0.4943323329274816, "300–1000 Hz_10ms_log_rms_mae_db": 1.79157264292815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12632599540265801, "300–1000 Hz_10ms_envelope_correlation": 0.9865889790019494, "300–1000 Hz_10ms_modulation_mae_db": 0.5837592229814411, "300–1000 Hz_40ms_log_rms_mae_db": 1.0152275885702937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06180842008977785, "300–1000 Hz_40ms_envelope_correlation": 0.9965446024320643, "300–1000 Hz_40ms_modulation_mae_db": 0.924240469072258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9848885191996044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1936355196539254, "1000–3000 Hz_10ms_envelope_correlation": 0.9692569198000447, "1000–3000 Hz_10ms_modulation_mae_db": 0.7885287258509975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1962148433059399, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10721473465414229, "1000–3000 Hz_40ms_envelope_correlation": 0.9899897021348718, "1000–3000 Hz_40ms_modulation_mae_db": 1.0035316873111093, "3000–6000 Hz_10ms_log_rms_mae_db": 2.334552032198062, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22184413077914517, "3000–6000 Hz_10ms_envelope_correlation": 0.9640623695754312, "3000–6000 Hz_10ms_modulation_mae_db": 1.1702727381485607, "3000–6000 Hz_40ms_log_rms_mae_db": 1.70493708757528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15392122049125664, "3000–6000 Hz_40ms_envelope_correlation": 0.9810525651482742, "3000–6000 Hz_40ms_modulation_mae_db": 1.4661563876352537, "mr_stft_db": 5.782527990277956, "band_mae_db": 0.11660398071723298, "rms_gain_error_db": -0.005360713133256161, "output_peak": 0.6440497871062396}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.038159757116177, "全帯域_10ms_rms_envelope_nrmse": 0.0793655972683145, "全帯域_10ms_envelope_correlation": 0.993355798980665, "全帯域_10ms_modulation_mae_db": 0.08629126593106413, "全帯域_40ms_log_rms_mae_db": 0.5636947150699907, "全帯域_40ms_rms_envelope_nrmse": 0.04503992791199488, "全帯域_40ms_envelope_correlation": 0.9976284963670936, "全帯域_40ms_modulation_mae_db": 0.07553139478114887, "300–1000 Hz_10ms_log_rms_mae_db": 1.7106349646110899, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1252539536603416, "300–1000 Hz_10ms_envelope_correlation": 0.9871270452782925, "300–1000 Hz_10ms_modulation_mae_db": 0.3335737443425922, "300–1000 Hz_40ms_log_rms_mae_db": 0.9764825278295636, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07129276854898701, "300–1000 Hz_40ms_envelope_correlation": 0.9959106469822616, "300–1000 Hz_40ms_modulation_mae_db": 0.3771860456932361, "1000–3000 Hz_10ms_log_rms_mae_db": 2.033367284443966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17580676000092782, "1000–3000 Hz_10ms_envelope_correlation": 0.9747610209919125, "1000–3000 Hz_10ms_modulation_mae_db": 0.9305980098010633, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2625034560279171, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10245132637256292, "1000–3000 Hz_40ms_envelope_correlation": 0.9908335131838374, "1000–3000 Hz_40ms_modulation_mae_db": 0.8097160915796308, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2987013303138424, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21604147915493208, "3000–6000 Hz_10ms_envelope_correlation": 0.9660796936573113, "3000–6000 Hz_10ms_modulation_mae_db": 0.6943188685688041, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6310683992515183, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15381485291605237, "3000–6000 Hz_40ms_envelope_correlation": 0.9811248372378274, "3000–6000 Hz_40ms_modulation_mae_db": 0.6705354775918363, "mr_stft_db": 5.7615537840409266, "band_mae_db": 0.16886704206632788, "rms_gain_error_db": -0.05368217359387188, "output_peak": 0.617461933514075}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.266604310828496, "spectral_convergence": 0.2321593795754191, "spectral_convergence_status": "ok", "amplitude_floor": 1.9341972091070217e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.548284789967771, "spectral_convergence": 0.22243813787394856, "spectral_convergence_status": "ok", "amplitude_floor": 1.0469896576426271e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.566092994951086, "spectral_convergence": 0.2468552700271514, "spectral_convergence_status": "ok", "amplitude_floor": 8.978224937278799e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002340108897553237, "output_power": 0.0023316359341233133, "delta_db": -0.01575328032788712, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003418531859998669, "output_power": 0.0032984451468959907, "delta_db": -0.1553036562384305, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009669299209010629, "output_power": 0.0009659130680778934, "delta_db": -0.004569575627928017, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016643839195516676, "output_power": 0.00015262520471136308, "delta_db": -0.3762725174544351, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.655994855693747e-05, "output_power": 7.155351831026947e-05, "delta_db": -0.29370639927834363, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.076442519269658e-05, "output_power": 4.1417488944801e-05, "delta_db": 0.06902442151987863, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15243830840781716, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09251933505921982, "pitch": null, "output_rms_dbfs": -21.63575213012575, "output_peak": 0.6546100276042567, "reference_rms_dbfs": -21.54323279506653, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.793660698582452}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019906122792766575, "log_rms_mae_db": 0.9916239927482712, "rms_envelope_nrmse": 0.07536986515436682, "envelope_correlation": 0.9940743766623579, "rms_gain_error_db": -0.09208788832678033, "reference_mean_power": 0.007012794178240996, "output_mean_power": 0.00686566017981849}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.192818590388021e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015356197181510699, "output_power": 0.0015290402873784433, "reference_active": true, "delta_db": -0.018647516288668845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003622112164404606, "output_power": 0.0003662962932245749, "reference_active": true, "delta_db": 0.04870628997733283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.688762087391039e-05, "output_power": 4.403103320637732e-05, "reference_active": true, "delta_db": -0.2729932027092769}], "active_band_count": 3, "active_band_mae_db": 0.11344900299175953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018229956701751592, "log_rms_mae_db": 0.5647600886072333, "rms_envelope_nrmse": 0.046280983246517235, "envelope_correlation": 0.9975700177055397, "rms_gain_error_db": -0.08911776963923354, "reference_mean_power": 0.0070179714231847195, "output_mean_power": 0.006875429266946632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8443307374696506e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014276197130393435, "output_power": 0.001423135467637682, "reference_active": true, "delta_db": -0.013662938528623932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018503527061359352, "output_power": 0.00019102939907956928, "reference_active": true, "delta_db": 0.13845689920219445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.327921114523828e-06, "output_power": 3.6630222973479004e-06, "reference_active": true, "delta_db": -0.7243977395309245}], "active_band_count": 3, "active_band_mae_db": 0.2921725257539143}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015227518148493586, "log_rms_mae_db": 1.6788067391280934, "rms_envelope_nrmse": 0.1247573348291501, "envelope_correlation": 0.9870895327929452, "rms_gain_error_db": -0.17514496554566714, "reference_mean_power": 0.0024461505543329222, "output_mean_power": 0.002349463414745233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3707367109300066e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00065188242251555, "output_power": 0.0006258527148739383, "reference_active": true, "delta_db": -0.1769713030805922}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000159373128948768, "output_power": 0.00014496311490733136, "reference_active": true, "delta_db": -0.4115758678876117}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8816561344170712e-05, "output_power": 1.4306311441176632e-05, "reference_active": true, "delta_db": -1.1901258547585314}], "active_band_count": 3, "active_band_mae_db": 0.5928910085755784}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013579343349222632, "log_rms_mae_db": 1.0216071197765597, "rms_envelope_nrmse": 0.07199222847047462, "envelope_correlation": 0.995535504679353, "rms_gain_error_db": -0.17169847550764591, "reference_mean_power": 0.0024424650949128645, "output_mean_power": 0.0023477860532397626}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.250982515304928e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006212562593255464, "output_power": 0.0005953759072003867, "reference_active": true, "delta_db": -0.18479521124587425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.001783099426803e-05, "output_power": 7.184239753250818e-05, "reference_active": true, "delta_db": -0.4680595776114105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.744470580402989e-06, "output_power": 1.1875611790096263e-06, "reference_active": true, "delta_db": -1.670076574832697}], "active_band_count": 3, "active_band_mae_db": 0.7743104545633273}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.560305347439803e-05, "log_rms_mae_db": 1.990776603740161, "rms_envelope_nrmse": 0.19796972856137965, "envelope_correlation": 0.9682446922460031, "rms_gain_error_db": -0.386127806851646, "reference_mean_power": 0.0001982579107701744, "output_mean_power": 0.00018139183933048246}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.44612649270932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.480334346914858e-05, "output_power": 3.051889906053579e-05, "reference_active": true, "delta_db": -0.5705210470646515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0911626587448635e-05, "output_power": 8.561421161640364e-06, "reference_active": true, "delta_db": -1.0534363353857377}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7887330599139967e-06, "output_power": 2.001177452891532e-06, "reference_active": true, "delta_db": -1.4412134411216475}], "active_band_count": 3, "active_band_mae_db": 1.0217236078573455}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.513956828007693e-05, "log_rms_mae_db": 1.2122002131510108, "rms_envelope_nrmse": 0.10743774883353348, "envelope_correlation": 0.9909089146860234, "rms_gain_error_db": -0.39785383619664466, "reference_mean_power": 0.00019869265760544437, "output_mean_power": 0.0001812994284908505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.64113687643322e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.340875190885911e-05, "output_power": 2.8916869271127806e-05, "reference_active": true, "delta_db": -0.6270897986268388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.757829899772919e-06, "output_power": 4.46613665256933e-06, "reference_active": true, "delta_db": -1.1032682336217894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2467549734271334e-07, "output_power": 1.5769238144886127e-07, "reference_active": true, "delta_db": -1.5374499984096959}], "active_band_count": 3, "active_band_mae_db": 1.0892693435527747}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.7519381942341847e-05, "log_rms_mae_db": 2.3750784619818748, "rms_envelope_nrmse": 0.21228832948479165, "envelope_correlation": 0.9682258607698793, "rms_gain_error_db": 0.15834387041335332, "reference_mean_power": 3.309162956541551e-05, "output_mean_power": 3.432041582778854e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.646064632792655e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.7250471980521725e-06, "output_power": 6.983327855092649e-06, "reference_active": true, "delta_db": 0.1636709521823742}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3903630336351795e-06, "output_power": 2.209082011800528e-06, "reference_active": true, "delta_db": -0.3425202465120184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2031554280127943e-06, "output_power": 8.478357860810506e-07, "reference_active": true, "delta_db": -1.5200999106910729}], "active_band_count": 3, "active_band_mae_db": 0.6754303697951552}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2421508904297867e-05, "log_rms_mae_db": 1.6983914987071371, "rms_envelope_nrmse": 0.15208886739127658, "envelope_correlation": 0.9821887933315253, "rms_gain_error_db": 0.15674491413389074, "reference_mean_power": 3.3247421347363265e-05, "output_mean_power": 3.4469299589526416e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.522637378578784e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.9685546500066154e-06, "output_power": 7.115032444096477e-06, "reference_active": true, "delta_db": 0.09034174489628359}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.482924821402799e-06, "output_power": 1.261206552184029e-06, "reference_active": true, "delta_db": -0.7033291620704112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.259963132949643e-08, "output_power": 6.30690193079201e-08, "reference_active": true, "delta_db": -1.667931796306434}], "active_band_count": 3, "active_band_mae_db": 0.8205342344243762}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8959118046931626, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bfc9efa04f90c245a5fbbd54c19e8e310878056c7a37a233f072e00054780d78", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9682447478669884, "standard_deviation": 0.004965919943862794, "minimum": 0.9603312107578975, "maximum": 0.9747682014383859, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.015876511653918, "standard_deviation": 0.02648863256795282, "minimum": 1.9848885191996044, "maximum": 2.057738756836606, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0200207769735812, "standard_deviation": 0.20890192402515512, "minimum": 0.7885287258509975, "maximum": 1.4259975643996103, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19705784501326729, "standard_deviation": 0.014997001083553218, "minimum": 0.17580676000092782, "maximum": 0.22061801006294202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.988460600582653, "standard_deviation": 0.004625140999353131, "minimum": 0.9779407183915442, "maximum": 0.9924828271248672, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2262624051427395, "standard_deviation": 0.03837203904455728, "minimum": 1.1584374023054493, "maximum": 1.2761678782275878, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.002104030601064, "standard_deviation": 0.14421886789330302, "minimum": 0.8069207487377111, "maximum": 1.2329951042196525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11435048421624214, "standard_deviation": 0.018422369121721117, "minimum": 0.10236934558605768, "maximum": 0.15694299863265265, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9621255152415307, "standard_deviation": 0.0037035037439212084, "minimum": 0.9577384588960905, "maximum": 0.9682258607698793, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3617100672767215, "standard_deviation": 0.04462259023861089, "minimum": 2.2987013303138424, "maximum": 2.4286829466761146, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7988045611095455, "standard_deviation": 0.25068760898710757, "minimum": 0.5462375043934364, "maximum": 1.1702727381485607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.229854463094097, "standard_deviation": 0.011616077746430205, "minimum": 0.21228832948479165, "maximum": 0.24370873586316283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9790078829650216, "standard_deviation": 0.0026178963874553636, "minimum": 0.975064921567436, "maximum": 0.9821887933315253, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7076371635853995, "standard_deviation": 0.049787210810492394, "minimum": 1.6310683992515183, "maximum": 1.769582935871235, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9666527835555285, "standard_deviation": 0.37286247237442205, "minimum": 0.5212062297303649, "maximum": 1.4661563876352537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16365102099839282, "standard_deviation": 0.01032273566373908, "minimum": 0.15208886739127658, "maximum": 0.1786655044807194, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9861936453171503, "standard_deviation": 0.001674950557759565, "minimum": 0.9841813229129986, "maximum": 0.9893091923488533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.752179807448112, "standard_deviation": 0.052665004544307885, "minimum": 1.6788067391280934, "maximum": 1.835055620370222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5021350367005563, "standard_deviation": 0.12056945693306544, "minimum": 0.3335737443425922, "maximum": 0.6511540168602621, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12879760713858834, "standard_deviation": 0.007795738946700467, "minimum": 0.1142471958039095, "maximum": 0.13832341015554797, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9950882929896421, "standard_deviation": 0.0014198899727565163, "minimum": 0.992439883055666, "maximum": 0.9965824488579705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0494470928250224, "standard_deviation": 0.04081418888895577, "minimum": 0.9764825278295636, "maximum": 1.0919899823413632, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.761699776300101, "standard_deviation": 0.2552768028298069, "minimum": 0.3771860456932361, "maximum": 1.1700786624430963, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07487776763652162, "standard_deviation": 0.010616733040506901, "minimum": 0.06180842008977785, "maximum": 0.09390363044047553, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1361447185358974, "standard_deviation": 0.0441217028824118, "minimum": 0.059342345823167005, "maximum": 0.2108918012307345, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7922128450758965, "standard_deviation": 0.014843033099294661, "minimum": 5.7615537840409266, "maximum": 5.8092349157681396, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6290322639424144, "standard_deviation": 0.015704402758717925, "minimum": 0.6051738021148055, "maximum": 0.6546100276042567, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06377821601616576, "standard_deviation": 0.031434344849449465, "minimum": -0.09907790555106223, "maximum": -0.005360713133256161, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936796191225998, "standard_deviation": 0.0008784278156283897, "minimum": 0.9916576491118282, "maximum": 0.9944469715656237, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0303229674651764, "standard_deviation": 0.02367672017926734, "minimum": 0.9916239927482712, "maximum": 1.0579148301855055, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32234721930945276, "standard_deviation": 0.22550896647865398, "minimum": 0.08629126593106413, "maximum": 0.7311412006867802, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07743723239295983, "standard_deviation": 0.005118095665683174, "minimum": 0.07279951423194767, "maximum": 0.08914293541234519, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978465300615005, "standard_deviation": 0.00043112597656771567, "minimum": 0.9970855656487009, "maximum": 0.9984507497719457, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5616629696277278, "standard_deviation": 0.020828319632654826, "minimum": 0.5318505612936432, "maximum": 0.5927525440293279, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4998238001321498, "standard_deviation": 0.28557162024956234, "minimum": 0.07553139478114887, "maximum": 0.944297249461898, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04309444729540485, "standard_deviation": 0.004343125565093516, "minimum": 0.03694479899760593, "maximum": 0.05054867516645408, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0002", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.057685375213623, "predicted_mos": 3.057685375213623, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4da9cbd9b0966378d51dad9b2c4f285c9351d78c148e46167907b28ddd2f7c5c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9374345000833273, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9868415013617499, "全帯域_10ms_rms_envelope_nrmse": 0.07452040117058752, "全帯域_10ms_envelope_correlation": 0.9941870961477455, "全帯域_10ms_modulation_mae_db": 0.14437408437676127, "全帯域_40ms_log_rms_mae_db": 0.5590138583147582, "全帯域_40ms_rms_envelope_nrmse": 0.04405857647190959, "全帯域_40ms_envelope_correlation": 0.997782496397107, "全帯域_40ms_modulation_mae_db": 0.29394196923358534, "300–1000 Hz_10ms_log_rms_mae_db": 1.6912754831152235, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13186719761524898, "300–1000 Hz_10ms_envelope_correlation": 0.9854205634875727, "300–1000 Hz_10ms_modulation_mae_db": 0.6143248764580415, "300–1000 Hz_40ms_log_rms_mae_db": 1.0389476366498898, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07426974495380124, "300–1000 Hz_40ms_envelope_correlation": 0.9950666321182199, "300–1000 Hz_40ms_modulation_mae_db": 0.7032031121539003, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0249144693428844, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19070222389114466, "1000–3000 Hz_10ms_envelope_correlation": 0.9718792165948719, "1000–3000 Hz_10ms_modulation_mae_db": 1.2502085964830159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2324559272521536, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10892377227689382, "1000–3000 Hz_40ms_envelope_correlation": 0.9924156628290999, "1000–3000 Hz_40ms_modulation_mae_db": 1.3471107711268868, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3614720348380445, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20807322030518388, "3000–6000 Hz_10ms_envelope_correlation": 0.9689043862646919, "3000–6000 Hz_10ms_modulation_mae_db": 0.7069073034518943, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6960167203646526, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15090706056232783, "3000–6000 Hz_40ms_envelope_correlation": 0.9820305481337975, "3000–6000 Hz_40ms_modulation_mae_db": 0.8625289563727069, "mr_stft_db": 5.7442385814911106, "band_mae_db": 0.20325582928133798, "rms_gain_error_db": -0.08224511663749398, "output_peak": 0.6536229176509786}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0353309043497756, "全帯域_10ms_rms_envelope_nrmse": 0.07607109787105976, "全帯域_10ms_envelope_correlation": 0.9939093210196502, "全帯域_10ms_modulation_mae_db": 0.3072893923817686, "全帯域_40ms_log_rms_mae_db": 0.5592282252960294, "全帯域_40ms_rms_envelope_nrmse": 0.040007860498634196, "全帯域_40ms_envelope_correlation": 0.9981582482475663, "全帯域_40ms_modulation_mae_db": 0.7171649128587422, "300–1000 Hz_10ms_log_rms_mae_db": 1.784980966609672, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13735322503224032, "300–1000 Hz_10ms_envelope_correlation": 0.9842917993769476, "300–1000 Hz_10ms_modulation_mae_db": 0.5529645352376237, "300–1000 Hz_40ms_log_rms_mae_db": 1.0804477986554102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08264075872186111, "300–1000 Hz_40ms_envelope_correlation": 0.9941201176320447, "300–1000 Hz_40ms_modulation_mae_db": 1.1834167309589265, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9951813238006308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18436456375498617, "1000–3000 Hz_10ms_envelope_correlation": 0.9722265884293193, "1000–3000 Hz_10ms_modulation_mae_db": 1.173563804009411, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2071928982701647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10601612393626811, "1000–3000 Hz_40ms_envelope_correlation": 0.9902138086512577, "1000–3000 Hz_40ms_modulation_mae_db": 1.2368437716430372, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3327846750280847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22926701237284133, "3000–6000 Hz_10ms_envelope_correlation": 0.9624415134308388, "3000–6000 Hz_10ms_modulation_mae_db": 0.743240548809729, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6716651803512712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16013281943325694, "3000–6000 Hz_40ms_envelope_correlation": 0.979892573664664, "3000–6000 Hz_40ms_modulation_mae_db": 0.7130931122221411, "mr_stft_db": 5.743883059815964, "band_mae_db": 0.15524366941367096, "rms_gain_error_db": -0.05193441666104837, "output_peak": 0.6236023969282881}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0770593379053612, "全帯域_10ms_rms_envelope_nrmse": 0.08870616888174716, "全帯域_10ms_envelope_correlation": 0.9917305361187052, "全帯域_10ms_modulation_mae_db": 0.565881029973938, "全帯域_40ms_log_rms_mae_db": 0.6080971243019855, "全帯域_40ms_rms_envelope_nrmse": 0.05028733134866132, "全帯域_40ms_envelope_correlation": 0.9971015545540338, "全帯域_40ms_modulation_mae_db": 0.728335763322971, "300–1000 Hz_10ms_log_rms_mae_db": 1.756961577437648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13798019859155036, "300–1000 Hz_10ms_envelope_correlation": 0.9842592673071224, "300–1000 Hz_10ms_modulation_mae_db": 0.6628404129009207, "300–1000 Hz_40ms_log_rms_mae_db": 1.0827455270481592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09279798492347087, "300–1000 Hz_40ms_envelope_correlation": 0.9925300846279393, "300–1000 Hz_40ms_modulation_mae_db": 0.874512295404826, "1000–3000 Hz_10ms_log_rms_mae_db": 2.068976885200675, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20219625199139404, "1000–3000 Hz_10ms_envelope_correlation": 0.9666937720107244, "1000–3000 Hz_10ms_modulation_mae_db": 1.449939449684086, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3161955050363496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14255967787992327, "1000–3000 Hz_40ms_envelope_correlation": 0.9821783867088848, "1000–3000 Hz_40ms_modulation_mae_db": 1.365207098143011, "3000–6000 Hz_10ms_log_rms_mae_db": 2.427648557361343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.236894085185821, "3000–6000 Hz_10ms_envelope_correlation": 0.9596414016026097, "3000–6000 Hz_10ms_modulation_mae_db": 0.793307378708067, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7664809255701865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1627171966022266, "3000–6000 Hz_40ms_envelope_correlation": 0.9790865492596791, "3000–6000 Hz_40ms_modulation_mae_db": 0.9942596640975507, "mr_stft_db": 5.751021165829616, "band_mae_db": 0.18653431438404053, "rms_gain_error_db": -0.09213447044566297, "output_peak": 0.6269858214058616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.055671207004195, "全帯域_10ms_rms_envelope_nrmse": 0.0734187818161316, "全帯域_10ms_envelope_correlation": 0.9943663592711499, "全帯域_10ms_modulation_mae_db": 0.7639763712484681, "全帯域_40ms_log_rms_mae_db": 0.5696513175118515, "全帯域_40ms_rms_envelope_nrmse": 0.03789317997730135, "全帯域_40ms_envelope_correlation": 0.998391494897586, "全帯域_40ms_modulation_mae_db": 0.9978121581600191, "300–1000 Hz_10ms_log_rms_mae_db": 1.791632153846399, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11692847328167429, "300–1000 Hz_10ms_envelope_correlation": 0.9887625229801029, "300–1000 Hz_10ms_modulation_mae_db": 0.4906623564294228, "300–1000 Hz_40ms_log_rms_mae_db": 1.0619548583993554, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06573785073316991, "300–1000 Hz_40ms_envelope_correlation": 0.9964592170452937, "300–1000 Hz_40ms_modulation_mae_db": 0.6484261447154938, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0367071872943323, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18148306266719386, "1000–3000 Hz_10ms_envelope_correlation": 0.9745393288160861, "1000–3000 Hz_10ms_modulation_mae_db": 1.3499467502736626, "1000–3000 Hz_40ms_log_rms_mae_db": 1.208386543502835, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10394142648494374, "1000–3000 Hz_40ms_envelope_correlation": 0.9929458749656137, "1000–3000 Hz_40ms_modulation_mae_db": 1.3105027933386963, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3745944714744693, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2420197678574666, "3000–6000 Hz_10ms_envelope_correlation": 0.9578890715693312, "3000–6000 Hz_10ms_modulation_mae_db": 0.7331866217768893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7483309984410227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.176497968842266, "3000–6000 Hz_40ms_envelope_correlation": 0.975370153603489, "3000–6000 Hz_40ms_modulation_mae_db": 0.8417436131210633, "mr_stft_db": 5.7366056899301965, "band_mae_db": 0.2150034303196108, "rms_gain_error_db": -0.07790189130778563, "output_peak": 0.628615583306393}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.056176911550763, "全帯域_10ms_rms_envelope_nrmse": 0.0806412234055841, "全帯域_10ms_envelope_correlation": 0.9931900758115506, "全帯域_10ms_modulation_mae_db": 0.43319819667500575, "全帯域_40ms_log_rms_mae_db": 0.5851761785838296, "全帯域_40ms_rms_envelope_nrmse": 0.04613490358331686, "全帯域_40ms_envelope_correlation": 0.9975847904403585, "全帯域_40ms_modulation_mae_db": 0.6145965546574836, "300–1000 Hz_10ms_log_rms_mae_db": 1.86940149580895, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13885129811546798, "300–1000 Hz_10ms_envelope_correlation": 0.9838678699131292, "300–1000 Hz_10ms_modulation_mae_db": 0.4737711496284886, "300–1000 Hz_40ms_log_rms_mae_db": 1.0962325762797538, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08334147302663028, "300–1000 Hz_40ms_envelope_correlation": 0.9938512251070867, "300–1000 Hz_40ms_modulation_mae_db": 0.9787154722449626, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0477216351269747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17998447878338478, "1000–3000 Hz_10ms_envelope_correlation": 0.9748864301683874, "1000–3000 Hz_10ms_modulation_mae_db": 1.1286258703113496, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2503781641448364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1100917084707293, "1000–3000 Hz_40ms_envelope_correlation": 0.991473855741288, "1000–3000 Hz_40ms_modulation_mae_db": 1.175297847188398, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3249339611042705, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23743733675420048, "3000–6000 Hz_10ms_envelope_correlation": 0.9605359950137299, "3000–6000 Hz_10ms_modulation_mae_db": 0.6332186932322572, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7135435870310411, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17174092959257536, "3000–6000 Hz_40ms_envelope_correlation": 0.9775565353556326, "3000–6000 Hz_40ms_modulation_mae_db": 0.9000407644586943, "mr_stft_db": 5.753079261492094, "band_mae_db": 0.2455624763399936, "rms_gain_error_db": -0.09346321730477931, "output_peak": 0.6024858382823397}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0261202395382183, "全帯域_10ms_rms_envelope_nrmse": 0.07990469592619598, "全帯域_10ms_envelope_correlation": 0.9932666466351694, "全帯域_10ms_modulation_mae_db": 0.1737123839221654, "全帯域_40ms_log_rms_mae_db": 0.553130132510895, "全帯域_40ms_rms_envelope_nrmse": 0.04553039299605693, "全帯域_40ms_envelope_correlation": 0.9975688545636122, "全帯域_40ms_modulation_mae_db": 0.2165611571198225, "300–1000 Hz_10ms_log_rms_mae_db": 1.8166609502330935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1377393166159409, "300–1000 Hz_10ms_envelope_correlation": 0.9841031403883541, "300–1000 Hz_10ms_modulation_mae_db": 0.3588535172488417, "300–1000 Hz_40ms_log_rms_mae_db": 1.099110476284307, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07270002651075118, "300–1000 Hz_40ms_envelope_correlation": 0.9951315081295032, "300–1000 Hz_40ms_modulation_mae_db": 0.6007885393267668, "1000–3000 Hz_10ms_log_rms_mae_db": 2.026355885283547, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20189818541727259, "1000–3000 Hz_10ms_envelope_correlation": 0.9665121920180603, "1000–3000 Hz_10ms_modulation_mae_db": 1.0162739227448134, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2346298008784868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12281398449051494, "1000–3000 Hz_40ms_envelope_correlation": 0.9863984308870106, "1000–3000 Hz_40ms_modulation_mae_db": 1.094793693104443, "3000–6000 Hz_10ms_log_rms_mae_db": 2.42882265963488, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23970855988016854, "3000–6000 Hz_10ms_envelope_correlation": 0.9579075796349965, "3000–6000 Hz_10ms_modulation_mae_db": 1.0436874153456286, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8132931564875316, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17760570180614485, "3000–6000 Hz_40ms_envelope_correlation": 0.975330724232864, "3000–6000 Hz_40ms_modulation_mae_db": 1.3540718544008612, "mr_stft_db": 5.748512815236308, "band_mae_db": 0.09242251985667543, "rms_gain_error_db": -0.03126225915328623, "output_peak": 0.635442885290828}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0370626451790517, "全帯域_10ms_rms_envelope_nrmse": 0.07260846672365892, "全帯域_10ms_envelope_correlation": 0.9944346282285654, "全帯域_10ms_modulation_mae_db": 0.2265533747461975, "全帯域_40ms_log_rms_mae_db": 0.5522069556262518, "全帯域_40ms_rms_envelope_nrmse": 0.03972830936853214, "全帯域_40ms_envelope_correlation": 0.9981337913337737, "全帯域_40ms_modulation_mae_db": 0.4681249728062677, "300–1000 Hz_10ms_log_rms_mae_db": 1.7736494941352967, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1248509134348803, "300–1000 Hz_10ms_envelope_correlation": 0.9869272555265329, "300–1000 Hz_10ms_modulation_mae_db": 0.6927885920366861, "300–1000 Hz_40ms_log_rms_mae_db": 1.013474557010023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.065456914782462, "300–1000 Hz_40ms_envelope_correlation": 0.9961875763128935, "300–1000 Hz_40ms_modulation_mae_db": 0.9638003450063811, "1000–3000 Hz_10ms_log_rms_mae_db": 2.066334551940671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18866538892762855, "1000–3000 Hz_10ms_envelope_correlation": 0.9716311893238817, "1000–3000 Hz_10ms_modulation_mae_db": 0.8819652986591592, "1000–3000 Hz_40ms_log_rms_mae_db": 1.258364243198056, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10756819506810798, "1000–3000 Hz_40ms_envelope_correlation": 0.9913032150807574, "1000–3000 Hz_40ms_modulation_mae_db": 1.1308905685658168, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3342967371213734, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22021664585414438, "3000–6000 Hz_10ms_envelope_correlation": 0.9645130124899164, "3000–6000 Hz_10ms_modulation_mae_db": 1.1368193332778398, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7121935059895095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15707157289568668, "3000–6000 Hz_40ms_envelope_correlation": 0.9803381336431043, "3000–6000 Hz_40ms_modulation_mae_db": 1.4059193152203713, "mr_stft_db": 5.729861110264757, "band_mae_db": 0.19751219924685617, "rms_gain_error_db": -0.02386034787546754, "output_peak": 0.6387409304394731}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.02751892240179, "全帯域_10ms_rms_envelope_nrmse": 0.07932331462857195, "全帯域_10ms_envelope_correlation": 0.9933771357455771, "全帯域_10ms_modulation_mae_db": 0.17043624191915152, "全帯域_40ms_log_rms_mae_db": 0.5624155021318628, "全帯域_40ms_rms_envelope_nrmse": 0.0441222792633962, "全帯域_40ms_envelope_correlation": 0.9977515863040531, "全帯域_40ms_modulation_mae_db": 0.07621609141382203, "300–1000 Hz_10ms_log_rms_mae_db": 1.7199581433868854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12447398988976092, "300–1000 Hz_10ms_envelope_correlation": 0.9873164427111976, "300–1000 Hz_10ms_modulation_mae_db": 0.3344073933051324, "300–1000 Hz_40ms_log_rms_mae_db": 0.966454025676199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07007637193684978, "300–1000 Hz_40ms_envelope_correlation": 0.9961056660295434, "300–1000 Hz_40ms_modulation_mae_db": 0.4117450751276049, "1000–3000 Hz_10ms_log_rms_mae_db": 2.04752049917173, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17324001222831864, "1000–3000 Hz_10ms_envelope_correlation": 0.9765348255317, "1000–3000 Hz_10ms_modulation_mae_db": 1.0269400275318383, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2694682414496645, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10835514940306162, "1000–3000 Hz_40ms_envelope_correlation": 0.9912400518527476, "1000–3000 Hz_40ms_modulation_mae_db": 0.9356261298017697, "3000–6000 Hz_10ms_log_rms_mae_db": 2.296721961257412, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21636474601270922, "3000–6000 Hz_10ms_envelope_correlation": 0.9657730952539053, "3000–6000 Hz_10ms_modulation_mae_db": 0.8450841647914707, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6626408137279909, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15595639819699336, "3000–6000 Hz_40ms_envelope_correlation": 0.9805446431249406, "3000–6000 Hz_40ms_modulation_mae_db": 0.7363482820471551, "mr_stft_db": 5.711719066488894, "band_mae_db": 0.22400680854601432, "rms_gain_error_db": -0.06814569567438258, "output_peak": 0.6095330176846702}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.198809512993465, "spectral_convergence": 0.2340668973360759, "spectral_convergence_status": "ok", "amplitude_floor": 1.9341972091070217e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.514613673852529, "spectral_convergence": 0.22288408963836334, "spectral_convergence_status": "ok", "amplitude_floor": 1.0469896576426271e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.51929255762734, "spectral_convergence": 0.24771318300174744, "spectral_convergence_status": "ok", "amplitude_floor": 8.978224937278799e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002340108897553237, "output_power": 0.0023385466188534376, "delta_db": -0.0029003590512521665, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003418531859998669, "output_power": 0.003304672461523496, "delta_db": -0.14711210387790463, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009669299209010629, "output_power": 0.000978480503357427, "delta_db": 0.051571772290059444, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016643839195516676, "output_power": 0.0001462892589289148, "delta_db": -0.5604107133422417, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.655994855693747e-05, "output_power": 6.935196519005972e-05, "delta_db": -0.42942860995277754, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.076442519269658e-05, "output_power": 4.0501413571054746e-05, "delta_db": -0.028111417173792407, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20325582928133798, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08224511663749398, "pitch": null, "output_rms_dbfs": -21.625477911704024, "output_peak": 0.6536229176509786, "reference_rms_dbfs": -21.54323279506653, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7442385814911106}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019906122792766575, "log_rms_mae_db": 0.9868415013617499, "rms_envelope_nrmse": 0.07452040117058752, "envelope_correlation": 0.9941870961477455, "rms_gain_error_db": -0.08180753118613165, "reference_mean_power": 0.007012794178240996, "output_mean_power": 0.006881931407203824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.192818590388021e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015356197181510699, "output_power": 0.0015304889876666232, "reference_active": true, "delta_db": -0.014534709830233045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003622112164404606, "output_power": 0.00036870882017459136, "reference_active": true, "delta_db": 0.07721632032820903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.688762087391039e-05, "output_power": 4.334321024044911e-05, "reference_active": true, "delta_db": -0.34137122297184175}], "active_band_count": 3, "active_band_mae_db": 0.14437408437676127}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018229956701751592, "log_rms_mae_db": 0.5590138583147582, "rms_envelope_nrmse": 0.04405857647190959, "envelope_correlation": 0.997782496397107, "rms_gain_error_db": -0.07983037081399605, "reference_mean_power": 0.0070179714231847195, "output_mean_power": 0.0068901481229169965}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8443307374696506e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014276197130393435, "output_power": 0.0014274760479663702, "reference_active": true, "delta_db": -0.0004370637884678694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018503527061359352, "output_power": 0.0001914001993346026, "reference_active": true, "delta_db": 0.14687866131375238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.327921114523828e-06, "output_power": 3.6545029599846035e-06, "reference_active": true, "delta_db": -0.7345101825985358}], "active_band_count": 3, "active_band_mae_db": 0.29394196923358534}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015227518148493586, "log_rms_mae_db": 1.6912754831152235, "rms_envelope_nrmse": 0.13186719761524898, "envelope_correlation": 0.9854205634875727, "rms_gain_error_db": -0.12785344420052835, "reference_mean_power": 0.0024461505543329222, "output_mean_power": 0.0023751871708025465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3707367109300066e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00065188242251555, "output_power": 0.0006323917514161535, "reference_active": true, "delta_db": -0.13183074207796178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000159373128948768, "output_power": 0.00014381592851402266, "reference_active": true, "delta_db": -0.446081096192913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8816561344170712e-05, "output_power": 1.4061575451947896e-05, "reference_active": true, "delta_db": -1.2650627911032497}], "active_band_count": 3, "active_band_mae_db": 0.6143248764580415}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013579343349222632, "log_rms_mae_db": 1.0389476366498898, "rms_envelope_nrmse": 0.07426974495380124, "envelope_correlation": 0.9950666321182199, "rms_gain_error_db": -0.12902074615725398, "reference_mean_power": 0.0024424650949128645, "output_mean_power": 0.0023709712692238038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.250982515304928e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006212562593255464, "output_power": 0.0006023178307791013, "reference_active": true, "delta_db": -0.1344505735514205}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.001783099426803e-05, "output_power": 7.051977871097882e-05, "reference_active": true, "delta_db": -0.5487583416021998}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.744470580402989e-06, "output_power": 1.2560982413568673e-06, "reference_active": true, "delta_db": -1.4264004213080808}], "active_band_count": 3, "active_band_mae_db": 0.7032031121539003}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.560305347439803e-05, "log_rms_mae_db": 2.0249144693428844, "rms_envelope_nrmse": 0.19070222389114466, "envelope_correlation": 0.9718792165948719, "rms_gain_error_db": -0.5668816094283124, "reference_mean_power": 0.0001982579107701744, "output_mean_power": 0.00017399724197748287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.44612649270932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.480334346914858e-05, "output_power": 2.981631476926075e-05, "reference_active": true, "delta_db": -0.6716700292888876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0911626587448635e-05, "output_power": 8.224209842835382e-06, "reference_active": true, "delta_db": -1.2279531241697563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7887330599139967e-06, "output_power": 1.8209863132175492e-06, "reference_active": true, "delta_db": -1.8510026359904037}], "active_band_count": 3, "active_band_mae_db": 1.2502085964830159}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.513956828007693e-05, "log_rms_mae_db": 1.2324559272521536, "rms_envelope_nrmse": 0.10892377227689382, "envelope_correlation": 0.9924156628290999, "rms_gain_error_db": -0.5715288510063322, "reference_mean_power": 0.00019869265760544437, "output_mean_power": 0.00017419229200178678}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.64113687643322e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.340875190885911e-05, "output_power": 2.8389204278911e-05, "reference_active": true, "delta_db": -0.7070703162194079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.757829899772919e-06, "output_power": 4.223224237588375e-06, "reference_active": true, "delta_db": -1.3461468915378292}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2467549734271334e-07, "output_power": 1.4214912841250515e-07, "reference_active": true, "delta_db": -1.9881151056234239}], "active_band_count": 3, "active_band_mae_db": 1.3471107711268868}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.7519381942341847e-05, "log_rms_mae_db": 2.3614720348380445, "rms_envelope_nrmse": 0.20807322030518388, "envelope_correlation": 0.9689043862646919, "rms_gain_error_db": 0.0441172992027756, "reference_mean_power": 3.309162956541551e-05, "output_mean_power": 3.342950023524223e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.646064632792655e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.7250471980521725e-06, "output_power": 7.007709448971579e-06, "reference_active": true, "delta_db": 0.17880750306191687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3903630336351795e-06, "output_power": 2.276616677655156e-06, "reference_active": true, "delta_db": -0.21173950924824414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2031554280127943e-06, "output_power": 8.078007367071885e-07, "reference_active": true, "delta_db": -1.7301748980455218}], "active_band_count": 3, "active_band_mae_db": 0.7069073034518943}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2421508904297867e-05, "log_rms_mae_db": 1.6960167203646526, "rms_envelope_nrmse": 0.15090706056232783, "envelope_correlation": 0.9820305481337975, "rms_gain_error_db": 0.04335658360673514, "reference_mean_power": 3.3247421347363265e-05, "output_mean_power": 3.358100007067316e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.522637378578784e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.9685546500066154e-06, "output_power": 7.170451172794226e-06, "reference_active": true, "delta_db": 0.12403772530795401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.482924821402799e-06, "output_power": 1.2991046155886727e-06, "reference_active": true, "delta_db": -0.5747500874367149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.259963132949643e-08, "output_power": 5.994174088344477e-08, "reference_active": true, "delta_db": -1.8887990563734518}], "active_band_count": 3, "active_band_mae_db": 0.8625289563727069}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8934649326962746, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9f72d4fcc1ee0d097b38ab6d5c51593db8fa1da55a8a03091773167fee595ffe", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9718629428616289, "standard_deviation": 0.0036580259990015777, "minimum": 0.9665121920180603, "maximum": 0.9765348255317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0392140546451802, "standard_deviation": 0.02413562817333398, "minimum": 1.9951813238006308, "maximum": 2.068976885200675, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.159682964962167, "standard_deviation": 0.18712684342497204, "minimum": 0.8819652986591592, "maximum": 1.449939449684086, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1878167709576654, "standard_deviation": 0.010281481203106091, "minimum": 0.17324001222831864, "maximum": 0.20219625199139404, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9897711608395825, "standard_deviation": 0.003658603654543724, "minimum": 0.9821783867088848, "maximum": 0.9929458749656137, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2471339154665682, "standard_deviation": 0.035603793116345844, "minimum": 1.2071928982701647, "maximum": 1.3161955050363496, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1995340841140072, "standard_deviation": 0.14582268028714812, "minimum": 0.9356261298017697, "maximum": 1.365207098143011, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11378375475130535, "standard_deviation": 0.012940301877801555, "minimum": 0.10394142648494374, "maximum": 0.14255967787992327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9622007569075024, "standard_deviation": 0.00395417447931185, "minimum": 0.9578890715693312, "maximum": 0.9689043862646919, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3601593822274847, "standard_deviation": 0.04805193167188634, "minimum": 2.296721961257412, "maximum": 2.42882265963488, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.829431432424222, "standard_deviation": 0.1740880057789695, "minimum": 0.6332186932322572, "maximum": 1.1368193332778398, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22874767177781694, "standard_deviation": 0.01248989216079524, "minimum": 0.20807322030518388, "maximum": 0.2420197678574666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9787687326272714, "standard_deviation": 0.0024628673027963536, "minimum": 0.975330724232864, "maximum": 0.9820305481337975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7230206109954007, "standard_deviation": 0.050565230283184354, "minimum": 1.6626408137279909, "maximum": 1.8132931564875316, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.976000695242568, "standard_deviation": 0.264932767266719, "minimum": 0.7130931122221411, "maximum": 1.4059193152203713, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1640787059914347, "standard_deviation": 0.010013670270679925, "minimum": 0.15090706056232783, "maximum": 0.17760570180614485, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9856186077113699, "standard_deviation": 0.0018318054426693855, "minimum": 0.9838678699131292, "maximum": 0.9887625229801029, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7755650330716461, "standard_deviation": 0.055275389232958196, "minimum": 1.6912754831152235, "maximum": 1.86940149580895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5225766041556447, "standard_deviation": 0.13288051635424483, "minimum": 0.3344073933051324, "maximum": 0.6927885920366861, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13125557657209552, "standard_deviation": 0.008235683036941036, "minimum": 0.11692847328167429, "maximum": 0.13885129811546798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949315033753156, "standard_deviation": 0.0013596589835498404, "minimum": 0.9925300846279393, "maximum": 0.9964592170452937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0549209320003872, "standard_deviation": 0.046189075064558834, "minimum": 0.966454025676199, "maximum": 1.099110476284307, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7955759643673577, "standard_deviation": 0.24898222309266965, "minimum": 0.4117450751276049, "maximum": 1.1834167309589265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07587764069862454, "standard_deviation": 0.009602758206198258, "minimum": 0.065456914782462, "maximum": 0.09279798492347087, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.189942655923525, "standard_deviation": 0.04757298948040129, "minimum": 0.09242251985667543, "maximum": 0.2455624763399936, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.739865093818618, "standard_deviation": 0.01369060656076651, "minimum": 5.711719066488894, "maximum": 5.753079261492094, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.627378673873604, "standard_deviation": 0.01619946192077287, "minimum": 0.6024858382823397, "maximum": 0.6536229176509786, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06511842688248834, "standard_deviation": 0.026783486055731248, "minimum": -0.09346321730477931, "maximum": -0.02386034787546754, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9935577248722641, "standard_deviation": 0.0008891167918120563, "minimum": 0.9917305361187052, "maximum": 0.9944346282285654, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0377227086613632, "standard_deviation": 0.026843101824815658, "minimum": 0.9868415013617499, "maximum": 1.0770593379053612, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34817763440543203, "standard_deviation": 0.2228202914964173, "minimum": 0.14437408437676127, "maximum": 0.7639763712484681, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07814926880294212, "standard_deviation": 0.005246541099084521, "minimum": 0.07260846672365892, "maximum": 0.08870616888174716, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978091020922614, "standard_deviation": 0.00041043210000004206, "minimum": 0.9971015545540338, "maximum": 0.998391494897586, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5686149117846829, "standard_deviation": 0.019130816596536825, "minimum": 0.5522069556262518, "maximum": 0.6080971243019855, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5140941974465891, "standard_deviation": 0.3075396077532102, "minimum": 0.07621609141382203, "maximum": 0.9978121581600191, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04347035418847607, "standard_deviation": 0.004065225470732864, "minimum": 0.03789317997730135, "maximum": 0.05028733134866132, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0002", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.9018921852111816, "predicted_mos": 2.9018921852111816, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "98d05a4b1b9cd3a9a9e1a0c84b8946495feb2a4b76e4577c16eef8d2f0a2e128", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.857490499969572, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9996166187685508, "全帯域_10ms_rms_envelope_nrmse": 0.07755862375614675, "全帯域_10ms_envelope_correlation": 0.9936474491592896, "全帯域_10ms_modulation_mae_db": 0.362654540113935, "全帯域_40ms_log_rms_mae_db": 0.5493172071676128, "全帯域_40ms_rms_envelope_nrmse": 0.03820701509480585, "全帯域_40ms_envelope_correlation": 0.9982801468208392, "全帯域_40ms_modulation_mae_db": 0.49299865662493253, "300–1000 Hz_10ms_log_rms_mae_db": 1.9146785852239232, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14549015303932386, "300–1000 Hz_10ms_envelope_correlation": 0.9822128450352914, "300–1000 Hz_10ms_modulation_mae_db": 0.1034842299077338, "300–1000 Hz_40ms_log_rms_mae_db": 1.2025088767802405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08080948230365374, "300–1000 Hz_40ms_envelope_correlation": 0.9939752196428367, "300–1000 Hz_40ms_modulation_mae_db": 0.46520947478330643, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3306848855030458, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21018416432897052, "1000–3000 Hz_10ms_envelope_correlation": 0.9636179243486868, "1000–3000 Hz_10ms_modulation_mae_db": 1.1767330997916972, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3810506090842882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1068054512043725, "1000–3000 Hz_40ms_envelope_correlation": 0.9897072168581127, "1000–3000 Hz_40ms_modulation_mae_db": 1.477678888150314, "3000–6000 Hz_10ms_log_rms_mae_db": 2.974531946629342, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2764164074856819, "3000–6000 Hz_10ms_envelope_correlation": 0.9442779952736617, "3000–6000 Hz_10ms_modulation_mae_db": 1.031333650254229, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1466502598089674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19227166604577278, "3000–6000 Hz_40ms_envelope_correlation": 0.9706269338034832, "3000–6000 Hz_40ms_modulation_mae_db": 1.3932393476828109, "mr_stft_db": 6.197799914557379, "band_mae_db": 0.15710949377746938, "rms_gain_error_db": -0.028008796398925475, "output_peak": 0.7147043853914787}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0033930010942036, "全帯域_10ms_rms_envelope_nrmse": 0.07636768227753218, "全帯域_10ms_envelope_correlation": 0.9938704068667354, "全帯域_10ms_modulation_mae_db": 0.3009459148113793, "全帯域_40ms_log_rms_mae_db": 0.5631839620525779, "全帯域_40ms_rms_envelope_nrmse": 0.040036551791684215, "全帯域_40ms_envelope_correlation": 0.998159343939038, "全帯域_40ms_modulation_mae_db": 0.3376837016701033, "300–1000 Hz_10ms_log_rms_mae_db": 1.82972212139738, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1180409062979755, "300–1000 Hz_10ms_envelope_correlation": 0.9884214717533794, "300–1000 Hz_10ms_modulation_mae_db": 0.7210029189121854, "300–1000 Hz_40ms_log_rms_mae_db": 1.086367400756678, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06301118467119832, "300–1000 Hz_40ms_envelope_correlation": 0.9966227364528062, "300–1000 Hz_40ms_modulation_mae_db": 1.1547909628294375, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3455023468439906, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2269126028621175, "1000–3000 Hz_10ms_envelope_correlation": 0.9581537490780299, "1000–3000 Hz_10ms_modulation_mae_db": 0.5460804241050198, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4143733636437292, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12780456681189922, "1000–3000 Hz_40ms_envelope_correlation": 0.9851140666918364, "1000–3000 Hz_40ms_modulation_mae_db": 0.5657179798961026, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9852817616115987, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27021940560257623, "3000–6000 Hz_10ms_envelope_correlation": 0.9514597896533761, "3000–6000 Hz_10ms_modulation_mae_db": 0.6371076176463535, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0943548895914788, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1903929081535639, "3000–6000 Hz_40ms_envelope_correlation": 0.974529785034667, "3000–6000 Hz_40ms_modulation_mae_db": 1.1140303791056365, "mr_stft_db": 6.158233070083907, "band_mae_db": 0.12578876692975352, "rms_gain_error_db": -0.059010505568287575, "output_peak": 0.6107505191181749}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0371182236146408, "全帯域_10ms_rms_envelope_nrmse": 0.079551608179296, "全帯域_10ms_envelope_correlation": 0.9933249537502895, "全帯域_10ms_modulation_mae_db": 0.2866744677461473, "全帯域_40ms_log_rms_mae_db": 0.5535814295094803, "全帯域_40ms_rms_envelope_nrmse": 0.04566023536998094, "全帯域_40ms_envelope_correlation": 0.9975292990666518, "全帯域_40ms_modulation_mae_db": 0.47906959632175145, "300–1000 Hz_10ms_log_rms_mae_db": 1.9347939951021227, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13025393269379926, "300–1000 Hz_10ms_envelope_correlation": 0.9857870229720799, "300–1000 Hz_10ms_modulation_mae_db": 0.41346481543424174, "300–1000 Hz_40ms_log_rms_mae_db": 1.1714490477425912, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07598721673552664, "300–1000 Hz_40ms_envelope_correlation": 0.9946821169396979, "300–1000 Hz_40ms_modulation_mae_db": 0.7071404634976449, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3475841859392013, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23616425606829297, "1000–3000 Hz_10ms_envelope_correlation": 0.954708323444607, "1000–3000 Hz_10ms_modulation_mae_db": 0.7606452695147577, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3638001431633624, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1338081988814574, "1000–3000 Hz_40ms_envelope_correlation": 0.9838190395371145, "1000–3000 Hz_40ms_modulation_mae_db": 0.9467699776611219, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9750179182243164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29668457229137374, "3000–6000 Hz_10ms_envelope_correlation": 0.9361117849552362, "3000–6000 Hz_10ms_modulation_mae_db": 1.0689512375880827, "3000–6000 Hz_40ms_log_rms_mae_db": 2.125217736169015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20382599963481593, "3000–6000 Hz_40ms_envelope_correlation": 0.9670928417926995, "3000–6000 Hz_40ms_modulation_mae_db": 1.3011420378583154, "mr_stft_db": 6.182720079445027, "band_mae_db": 0.054555905812644405, "rms_gain_error_db": -0.004573097346737943, "output_peak": 0.6574198459213545}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.015112285523769, "全帯域_10ms_rms_envelope_nrmse": 0.07372971008454024, "全帯域_10ms_envelope_correlation": 0.9942893790928702, "全帯域_10ms_modulation_mae_db": 0.41832118360243364, "全帯域_40ms_log_rms_mae_db": 0.5349435362625875, "全帯域_40ms_rms_envelope_nrmse": 0.0361056461930929, "全帯域_40ms_envelope_correlation": 0.9985153807438188, "全帯域_40ms_modulation_mae_db": 0.7917630908624465, "300–1000 Hz_10ms_log_rms_mae_db": 1.9446576108812368, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13072770505445794, "300–1000 Hz_10ms_envelope_correlation": 0.9859721450908621, "300–1000 Hz_10ms_modulation_mae_db": 0.8420899641552978, "300–1000 Hz_40ms_log_rms_mae_db": 1.1930926380383542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07734170458395322, "300–1000 Hz_40ms_envelope_correlation": 0.9950822676618887, "300–1000 Hz_40ms_modulation_mae_db": 1.128259363016152, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2963849356611816, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21561811055603344, "1000–3000 Hz_10ms_envelope_correlation": 0.9616508713657482, "1000–3000 Hz_10ms_modulation_mae_db": 0.5845005399082482, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3506913621601158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10642277167341094, "1000–3000 Hz_40ms_envelope_correlation": 0.9899765398669323, "1000–3000 Hz_40ms_modulation_mae_db": 0.7870102978478926, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9703458112814247, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2663056412805198, "3000–6000 Hz_10ms_envelope_correlation": 0.9493432634025507, "3000–6000 Hz_10ms_modulation_mae_db": 0.40277605150230245, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1214086099179306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17962758933069073, "3000–6000 Hz_40ms_envelope_correlation": 0.9748631365831479, "3000–6000 Hz_40ms_modulation_mae_db": 0.7327817390942557, "mr_stft_db": 6.1715621689112234, "band_mae_db": 0.14318814361224463, "rms_gain_error_db": -0.06576202313450105, "output_peak": 0.6509742351833708}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9987107174125657, "全帯域_10ms_rms_envelope_nrmse": 0.07720031541950333, "全帯域_10ms_envelope_correlation": 0.9937146405664373, "全帯域_10ms_modulation_mae_db": 0.1158724068671606, "全帯域_40ms_log_rms_mae_db": 0.5330459165815584, "全帯域_40ms_rms_envelope_nrmse": 0.03755606458931157, "全帯域_40ms_envelope_correlation": 0.9983660049863932, "全帯域_40ms_modulation_mae_db": 0.07586745443927849, "300–1000 Hz_10ms_log_rms_mae_db": 1.971275156426885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12833748824463, "300–1000 Hz_10ms_envelope_correlation": 0.9862146674270437, "300–1000 Hz_10ms_modulation_mae_db": 0.3104278767950565, "300–1000 Hz_40ms_log_rms_mae_db": 1.1909058923985918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07211414441566913, "300–1000 Hz_40ms_envelope_correlation": 0.9953975255779712, "300–1000 Hz_40ms_modulation_mae_db": 0.13529230304228756, "1000–3000 Hz_10ms_log_rms_mae_db": 2.35639324374185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21532365758730732, "1000–3000 Hz_10ms_envelope_correlation": 0.9621216225698883, "1000–3000 Hz_10ms_modulation_mae_db": 0.5224323718194922, "1000–3000 Hz_40ms_log_rms_mae_db": 1.435561836844469, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11610829893234602, "1000–3000 Hz_40ms_envelope_correlation": 0.9876907222194561, "1000–3000 Hz_40ms_modulation_mae_db": 0.7683404949070214, "3000–6000 Hz_10ms_log_rms_mae_db": 3.0634160244491606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2774625418675964, "3000–6000 Hz_10ms_envelope_correlation": 0.9461253172301703, "3000–6000 Hz_10ms_modulation_mae_db": 1.2146967945250482, "3000–6000 Hz_40ms_log_rms_mae_db": 2.138227660140457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1871511243658066, "3000–6000 Hz_40ms_envelope_correlation": 0.9734806517455349, "3000–6000 Hz_40ms_modulation_mae_db": 1.5285141517338259, "mr_stft_db": 6.181619172503549, "band_mae_db": 0.11273942525846192, "rms_gain_error_db": -0.04962071248545355, "output_peak": 0.6177925807680554}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9657732308475512, "全帯域_10ms_rms_envelope_nrmse": 0.07043801677786325, "全帯域_10ms_envelope_correlation": 0.9948411666350694, "全帯域_10ms_modulation_mae_db": 0.38926666660406584, "全帯域_40ms_log_rms_mae_db": 0.5172001202572922, "全帯域_40ms_rms_envelope_nrmse": 0.033056827552328305, "全帯域_40ms_envelope_correlation": 0.99883581947211, "全帯域_40ms_modulation_mae_db": 0.4454887155360941, "300–1000 Hz_10ms_log_rms_mae_db": 1.7987578455230155, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13341894398719142, "300–1000 Hz_10ms_envelope_correlation": 0.9855006691503718, "300–1000 Hz_10ms_modulation_mae_db": 0.7991939855953119, "300–1000 Hz_40ms_log_rms_mae_db": 1.0692619841471391, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0829281187510038, "300–1000 Hz_40ms_envelope_correlation": 0.9944203117390401, "300–1000 Hz_40ms_modulation_mae_db": 0.9576987242799366, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4201426325054216, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22628048908721607, "1000–3000 Hz_10ms_envelope_correlation": 0.957816698966332, "1000–3000 Hz_10ms_modulation_mae_db": 0.5043342730385134, "1000–3000 Hz_40ms_log_rms_mae_db": 1.462458135674651, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12575988187181494, "1000–3000 Hz_40ms_envelope_correlation": 0.9855630024011183, "1000–3000 Hz_40ms_modulation_mae_db": 0.7258637559128163, "3000–6000 Hz_10ms_log_rms_mae_db": 2.970059075876258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2713890533111892, "3000–6000 Hz_10ms_envelope_correlation": 0.9464884504417743, "3000–6000 Hz_10ms_modulation_mae_db": 1.2765303890406356, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1116721345635243, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18852493415692992, "3000–6000 Hz_40ms_envelope_correlation": 0.9718282118226231, "3000–6000 Hz_40ms_modulation_mae_db": 1.5004785133725127, "mr_stft_db": 6.169152881664942, "band_mae_db": 0.10875866323687293, "rms_gain_error_db": -0.09292972409628197, "output_peak": 0.6333432486512092}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0142356494213574, "全帯域_10ms_rms_envelope_nrmse": 0.07887491956076, "全帯域_10ms_envelope_correlation": 0.9934553867830215, "全帯域_10ms_modulation_mae_db": 0.18135704753486784, "全帯域_40ms_log_rms_mae_db": 0.5413004721161436, "全帯域_40ms_rms_envelope_nrmse": 0.04642053338437675, "全帯域_40ms_envelope_correlation": 0.9974946068454296, "全帯域_40ms_modulation_mae_db": 0.18434801714506002, "300–1000 Hz_10ms_log_rms_mae_db": 2.004244616908089, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11364410630760842, "300–1000 Hz_10ms_envelope_correlation": 0.9893062942612101, "300–1000 Hz_10ms_modulation_mae_db": 0.7152391134391047, "300–1000 Hz_40ms_log_rms_mae_db": 1.1880336511527245, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0641898455045705, "300–1000 Hz_40ms_envelope_correlation": 0.9964548869117587, "300–1000 Hz_40ms_modulation_mae_db": 0.7929009872786653, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3939784028199855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2597388876461305, "1000–3000 Hz_10ms_envelope_correlation": 0.9459399652130566, "1000–3000 Hz_10ms_modulation_mae_db": 0.7729152035426896, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4325931107008236, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15400395851013832, "1000–3000 Hz_40ms_envelope_correlation": 0.978969576480154, "1000–3000 Hz_40ms_modulation_mae_db": 0.7573197474285701, "3000–6000 Hz_10ms_log_rms_mae_db": 2.998892368081694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2783912296564774, "3000–6000 Hz_10ms_envelope_correlation": 0.9462492656898074, "3000–6000 Hz_10ms_modulation_mae_db": 0.6177772984355031, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0966289364250046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1853063273365706, "3000–6000 Hz_40ms_envelope_correlation": 0.9744320374284814, "3000–6000 Hz_40ms_modulation_mae_db": 0.68744834445867, "mr_stft_db": 6.191454634224239, "band_mae_db": 0.09609785183264148, "rms_gain_error_db": -0.06175277868769098, "output_peak": 0.6111499028497933}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.970565804013979, "全帯域_10ms_rms_envelope_nrmse": 0.06566581449943701, "全帯域_10ms_envelope_correlation": 0.9955816551191455, "全帯域_10ms_modulation_mae_db": 0.6752729237910847, "全帯域_40ms_log_rms_mae_db": 0.5301103077765718, "全帯域_40ms_rms_envelope_nrmse": 0.03898766239742235, "全帯域_40ms_envelope_correlation": 0.9983677798684013, "全帯域_40ms_modulation_mae_db": 0.767057113196801, "300–1000 Hz_10ms_log_rms_mae_db": 1.9075640409351442, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1167333328582253, "300–1000 Hz_10ms_envelope_correlation": 0.9886625722745467, "300–1000 Hz_10ms_modulation_mae_db": 0.6926133946466475, "300–1000 Hz_40ms_log_rms_mae_db": 1.164921792739344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07596930185800876, "300–1000 Hz_40ms_envelope_correlation": 0.9949028696485115, "300–1000 Hz_40ms_modulation_mae_db": 1.2351056761289436, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3740487825733307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22395775388474634, "1000–3000 Hz_10ms_envelope_correlation": 0.9590417125736651, "1000–3000 Hz_10ms_modulation_mae_db": 0.7692885500682399, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4191423787993827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13183387340181327, "1000–3000 Hz_40ms_envelope_correlation": 0.9841180432806741, "1000–3000 Hz_40ms_modulation_mae_db": 0.9246574770383695, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9562121914135524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28953714445836565, "3000–6000 Hz_10ms_envelope_correlation": 0.9402993462571516, "3000–6000 Hz_10ms_modulation_mae_db": 0.9196400186248184, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1522541668784307, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2018431525165143, "3000–6000 Hz_40ms_envelope_correlation": 0.9683474085903971, "3000–6000 Hz_40ms_modulation_mae_db": 1.2190905004995702, "mr_stft_db": 6.183681064238978, "band_mae_db": 0.11414496644793708, "rms_gain_error_db": -0.10879668327525503, "output_peak": 0.6347259710277215}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.740311276639905, "spectral_convergence": 0.24453794410086757, "spectral_convergence_status": "ok", "amplitude_floor": 1.9341972091070217e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.037792365730293, "spectral_convergence": 0.2394584113585776, "spectral_convergence_status": "ok", "amplitude_floor": 1.0469896576426271e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.815296101301941, "spectral_convergence": 0.2612652982613241, "spectral_convergence_status": "ok", "amplitude_floor": 8.978224937278799e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002340108897553237, "output_power": 0.002354241063693842, "delta_db": 0.02614862702692957, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003418531859998669, "output_power": 0.0033210274234815977, "delta_db": -0.12567169975594147, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009669299209010629, "output_power": 0.001013905072590228, "delta_db": 0.2060229649743096, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016643839195516676, "output_power": 0.00016032939675919425, "delta_db": -0.16240352669074165, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.655994855693747e-05, "output_power": 7.232664116404528e-05, "delta_db": -0.24703335943319846, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.076442519269658e-05, "output_power": 4.244426343051399e-05, "delta_db": 0.1753767847836954, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15710949377746938, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.028008796398925475, "pitch": null, "output_rms_dbfs": -21.57124159146546, "output_peak": 0.7147043853914787, "reference_rms_dbfs": -21.54323279506653, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.197799914557379}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019906122792766575, "log_rms_mae_db": 0.9996166187685508, "rms_envelope_nrmse": 0.07755862375614675, "envelope_correlation": 0.9936474491592896, "rms_gain_error_db": -0.027365592206190253, "reference_mean_power": 0.007012794178240996, "output_mean_power": 0.006968744365017582}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.192818590388021e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015356197181510699, "output_power": 0.0015313729161570888, "reference_active": true, "delta_db": -0.012027181479647145}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003622112164404606, "output_power": 0.00036362052498467855, "reference_active": true, "delta_db": 0.016864947019264727}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.688762087391039e-05, "output_power": 3.67410063451767e-05, "reference_active": true, "delta_db": -1.0590714918428932}], "active_band_count": 3, "active_band_mae_db": 0.362654540113935}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018229956701751592, "log_rms_mae_db": 0.5493172071676128, "rms_envelope_nrmse": 0.03820701509480585, "envelope_correlation": 0.9982801468208392, "rms_gain_error_db": -0.02777392603445533, "reference_mean_power": 0.0070179714231847195, "output_mean_power": 0.006973233419337882}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8443307374696506e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014276197130393435, "output_power": 0.0014207239132894886, "reference_active": true, "delta_db": -0.021028457402668503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018503527061359352, "output_power": 0.00018664395458593253, "reference_active": true, "delta_db": 0.03759408046116242}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.327921114523828e-06, "output_power": 3.1206279434635215e-06, "reference_active": true, "delta_db": -1.4203734320109667}], "active_band_count": 3, "active_band_mae_db": 0.49299865662493253}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015227518148493586, "log_rms_mae_db": 1.9146785852239232, "rms_envelope_nrmse": 0.14549015303932386, "envelope_correlation": 0.9822128450352914, "rms_gain_error_db": -0.023663703877545125, "reference_mean_power": 0.0024461505543329222, "output_mean_power": 0.0024328582907088584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3707367109300066e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00065188242251555, "output_power": 0.0006616957751514883, "reference_active": true, "delta_db": 0.0648909093686721}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000159373128948768, "output_power": 0.0001512823049415938, "reference_active": true, "delta_db": -0.22626966350852346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8816561344170712e-05, "output_power": 1.8733160281516976e-05, "reference_active": true, "delta_db": -0.01929211684600586}], "active_band_count": 3, "active_band_mae_db": 0.1034842299077338}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013579343349222632, "log_rms_mae_db": 1.2025088767802405, "rms_envelope_nrmse": 0.08080948230365374, "envelope_correlation": 0.9939752196428367, "rms_gain_error_db": -0.029384125617617017, "reference_mean_power": 0.0024424650949128645, "output_mean_power": 0.0024259952902351356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.250982515304928e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006212562593255464, "output_power": 0.0006335594856325227, "reference_active": true, "delta_db": 0.08516620146525987}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.001783099426803e-05, "output_power": 7.53794140180607e-05, "reference_active": true, "delta_db": -0.25934017963195455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.744470580402989e-06, "output_power": 1.3694666021509721e-06, "reference_active": true, "delta_db": -1.0511220432527049}], "active_band_count": 3, "active_band_mae_db": 0.46520947478330643}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.560305347439803e-05, "log_rms_mae_db": 2.3306848855030458, "rms_envelope_nrmse": 0.21018416432897052, "envelope_correlation": 0.9636179243486868, "rms_gain_error_db": -0.1414971062766049, "reference_mean_power": 0.0001982579107701744, "output_mean_power": 0.00019190258048626987}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.44612649270932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.480334346914858e-05, "output_power": 2.976960977914425e-05, "reference_active": true, "delta_db": -0.6784782558265012}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0911626587448635e-05, "output_power": 8.866465927206731e-06, "reference_active": true, "delta_db": -0.9013894588047768}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7887330599139967e-06, "output_power": 1.779810590777578e-06, "reference_active": true, "delta_db": -1.9503315847438136}], "active_band_count": 3, "active_band_mae_db": 1.1767330997916972}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.513956828007693e-05, "log_rms_mae_db": 1.3810506090842882, "rms_envelope_nrmse": 0.1068054512043725, "envelope_correlation": 0.9897072168581127, "rms_gain_error_db": -0.1514403019920921, "reference_mean_power": 0.00019869265760544437, "output_mean_power": 0.0001918835693647686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.64113687643322e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.340875190885911e-05, "output_power": 2.8146042930081685e-05, "reference_active": true, "delta_db": -0.7444290573304623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.757829899772919e-06, "output_power": 4.297924221207659e-06, "reference_active": true, "delta_db": -1.2700007665605817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2467549734271334e-07, "output_power": 1.2873452285240277e-07, "reference_active": true, "delta_db": -2.418606840559898}], "active_band_count": 3, "active_band_mae_db": 1.477678888150314}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.7519381942341847e-05, "log_rms_mae_db": 2.974531946629342, "rms_envelope_nrmse": 0.2764164074856819, "envelope_correlation": 0.9442779952736617, "rms_gain_error_db": -0.05795208118064582, "reference_mean_power": 3.309162956541551e-05, "output_mean_power": 3.265298929777356e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.646064632792655e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.7250471980521725e-06, "output_power": 6.68506230432785e-06, "reference_active": true, "delta_db": -0.025898774597037315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3903630336351795e-06, "output_power": 2.037362592889818e-06, "reference_active": true, "delta_db": -0.6939553592465935}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2031554280127943e-06, "output_power": 6.964774392194126e-07, "reference_active": true, "delta_db": -2.3741468169190565}], "active_band_count": 3, "active_band_mae_db": 1.031333650254229}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2421508904297867e-05, "log_rms_mae_db": 2.1466502598089674, "rms_envelope_nrmse": 0.19227166604577278, "envelope_correlation": 0.9706269338034832, "rms_gain_error_db": -0.060606856546682464, "reference_mean_power": 3.3247421347363265e-05, "output_mean_power": 3.27866678964957e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.522637378578784e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.9685546500066154e-06, "output_power": 6.665916430810545e-06, "reference_active": true, "delta_db": -0.1928284559022233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.482924821402799e-06, "output_power": 1.0471099891982991e-06, "reference_active": true, "delta_db": -1.511268318383813}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.259963132949643e-08, "output_power": 5.236572621128827e-08, "reference_active": true, "delta_db": -2.475621268762396}], "active_band_count": 3, "active_band_mae_db": 1.3932393476828109}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8701931467826463, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4dec8bac289f7621e0e41600b0d550e06119bb1a4c5b6bd7481d1b67948a285c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9578813584450018, "standard_deviation": 0.00559575194128191, "minimum": 0.9459399652130566, "maximum": 0.9636179243486868, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.3580899269485007, "standard_deviation": 0.03821638539293086, "minimum": 2.2963849356611816, "maximum": 2.4201426325054216, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7046162164735823, "standard_deviation": 0.2228856375567959, "minimum": 0.5043342730385134, "maximum": 1.1767330997916972, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22677249025260185, "standard_deviation": 0.01564718857685305, "minimum": 0.21018416432897052, "maximum": 0.2597388876461305, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9856197759169248, "standard_deviation": 0.0035848760523177277, "minimum": 0.978969576480154, "maximum": 0.9899765398669323, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4074588675088529, "standard_deviation": 0.03865002540374929, "minimum": 1.3506913621601158, "maximum": 1.462458135674651, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.869169827355276, "standard_deviation": 0.2729287652885705, "minimum": 0.5657179798961026, "maximum": 1.477678888150314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12531837516090658, "standard_deviation": 0.015713348656627754, "minimum": 0.10642277167341094, "maximum": 0.15400395851013832, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9450444016129661, "standard_deviation": 0.004887847010564217, "minimum": 0.9361117849552362, "maximum": 0.9514597896533761, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.986719637195918, "standard_deviation": 0.033368996380468564, "minimum": 2.9562121914135524, "maximum": 3.0634160244491606, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8961016322021216, "standard_deviation": 0.3123322680883157, "minimum": 0.40277605150230245, "maximum": 1.2765303890406356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.27830074949422257, "standard_deviation": 0.010180129947231022, "minimum": 0.2663056412805198, "maximum": 0.29668457229137374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9719001258501293, "standard_deviation": 0.002975162610894712, "minimum": 0.9670928417926995, "maximum": 0.9748631365831479, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.123301799186851, "standard_deviation": 0.021705959668753636, "minimum": 2.0943548895914788, "maximum": 2.1522541668784307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1845906267256996, "standard_deviation": 0.32351590329054897, "minimum": 0.68744834445867, "maximum": 1.5285141517338259, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19111796269258308, "standard_deviation": 0.008164767724809709, "minimum": 0.17962758933069073, "maximum": 0.20382599963481593, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.986509710995598, "standard_deviation": 0.0022831697001495337, "minimum": 0.9822128450352914, "maximum": 0.9893062942612101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9132117465497247, "standard_deviation": 0.06888628315781956, "minimum": 1.7987578455230155, "maximum": 2.004244616908089, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5746895373606975, "standard_deviation": 0.2658772914007315, "minimum": 0.1034842299077338, "maximum": 0.8420899641552978, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12708082106040147, "standard_deviation": 0.010510938473903744, "minimum": 0.11364410630760842, "maximum": 0.14549015303932386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9951922418218139, "standard_deviation": 0.0009347120210220419, "minimum": 0.9939752196428367, "maximum": 0.9966227364528062, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.158317660469458, "standard_deviation": 0.05131276209846531, "minimum": 1.0692619841471391, "maximum": 1.2025088767802405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8220497443570467, "standard_deviation": 0.3791949087236216, "minimum": 0.13529230304228756, "maximum": 1.2351056761289436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07404387485294801, "standard_deviation": 0.007227020411345304, "minimum": 0.06301118467119832, "maximum": 0.0829281187510038, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11404790211350317, "standard_deviation": 0.031012674293582235, "minimum": 0.054555905812644405, "maximum": 0.15710949377746938, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.179527873203655, "standard_deviation": 0.012719441626957487, "minimum": 6.158233070083907, "maximum": 6.197799914557379, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6413575861138948, "standard_deviation": 0.0343158505078612, "minimum": 0.6107505191181749, "maximum": 0.7147043853914787, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0588067901241417, "standard_deviation": 0.03315473145960617, "minimum": -0.10879668327525503, "maximum": -0.004573097346737943, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940906297466073, "standard_deviation": 0.0007747840208412747, "minimum": 0.9933249537502895, "maximum": 0.9955816551191455, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0005656913370773, "standard_deviation": 0.02347461445336251, "minimum": 0.9657732308475512, "maximum": 1.0371182236146408, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3412956438838843, "standard_deviation": 0.1695535072056768, "minimum": 0.1158724068671606, "maximum": 0.6752729237910847, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07492333631938485, "standard_deviation": 0.004758390554606607, "minimum": 0.06566581449943701, "maximum": 0.079551608179296, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981935477178352, "standard_deviation": 0.00046515769546706334, "minimum": 0.9974946068454296, "maximum": 0.99883581947211, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5403353689654781, "standard_deviation": 0.014643968785992586, "minimum": 0.5172001202572922, "maximum": 0.5631839620525779, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4467845432245584, "standard_deviation": 0.2516005553334644, "minimum": 0.07586745443927849, "maximum": 0.7917630908624465, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03950381704662537, "standard_deviation": 0.004549401012788718, "minimum": 0.033056827552328305, "maximum": 0.04642053338437675, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0002", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.186124563217163, "predicted_mos": 3.186124563217163, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "693dcfa4de8ec35ca3e58e401ed47758b8d0095f668c6a92e13e447863f1dbec", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.087868300266564, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.984796844826022, "全帯域_10ms_rms_envelope_nrmse": 0.07013794772063599, "全帯域_10ms_envelope_correlation": 0.9948340892144749, "全帯域_10ms_modulation_mae_db": 0.18539099640135948, "全帯域_40ms_log_rms_mae_db": 0.5319767879602244, "全帯域_40ms_rms_envelope_nrmse": 0.037103200019884784, "全帯域_40ms_envelope_correlation": 0.9984177175068987, "全帯域_40ms_modulation_mae_db": 0.29913196355, "300–1000 Hz_10ms_log_rms_mae_db": 1.6631803811825314, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12603902900641528, "300–1000 Hz_10ms_envelope_correlation": 0.9868150502747745, "300–1000 Hz_10ms_modulation_mae_db": 0.6262156126096753, "300–1000 Hz_40ms_log_rms_mae_db": 1.0416502323049834, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07217172574257907, "300–1000 Hz_40ms_envelope_correlation": 0.9955499724867959, "300–1000 Hz_40ms_modulation_mae_db": 0.8555941146984413, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8912674302203576, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17438186219726795, "1000–3000 Hz_10ms_envelope_correlation": 0.9751338168956561, "1000–3000 Hz_10ms_modulation_mae_db": 0.7959742350299269, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1512132750793227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1004805127335995, "1000–3000 Hz_40ms_envelope_correlation": 0.9908557328889557, "1000–3000 Hz_40ms_modulation_mae_db": 0.8826493771484797, "3000–6000 Hz_10ms_log_rms_mae_db": 2.174221458558142, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22043089272470126, "3000–6000 Hz_10ms_envelope_correlation": 0.9644703854456425, "3000–6000 Hz_10ms_modulation_mae_db": 0.8211522353103186, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5888912274068352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15580610868839992, "3000–6000 Hz_40ms_envelope_correlation": 0.9805723428699097, "3000–6000 Hz_40ms_modulation_mae_db": 1.0522865964451809, "mr_stft_db": 5.658981646245188, "band_mae_db": 0.09436440406229807, "rms_gain_error_db": -0.05673415854181935, "output_peak": 0.6490371573824004}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0311879841354537, "全帯域_10ms_rms_envelope_nrmse": 0.07726734425230859, "全帯域_10ms_envelope_correlation": 0.993697288639423, "全帯域_10ms_modulation_mae_db": 0.13971930317603, "全帯域_40ms_log_rms_mae_db": 0.535166276960632, "全帯域_40ms_rms_envelope_nrmse": 0.038950018108387434, "全帯域_40ms_envelope_correlation": 0.9982191298167024, "全帯域_40ms_modulation_mae_db": 0.19928487336445513, "300–1000 Hz_10ms_log_rms_mae_db": 1.606646316522918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12640386785651403, "300–1000 Hz_10ms_envelope_correlation": 0.986635395407222, "300–1000 Hz_10ms_modulation_mae_db": 0.6253760734824777, "300–1000 Hz_40ms_log_rms_mae_db": 1.0071368256015925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0738500398238377, "300–1000 Hz_40ms_envelope_correlation": 0.9951914082388222, "300–1000 Hz_40ms_modulation_mae_db": 0.49564906418464494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.912495541929079, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2116582377300727, "1000–3000 Hz_10ms_envelope_correlation": 0.9632045050395525, "1000–3000 Hz_10ms_modulation_mae_db": 0.9670658024895659, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1973412299654977, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11236543966124672, "1000–3000 Hz_40ms_envelope_correlation": 0.9891991283988474, "1000–3000 Hz_40ms_modulation_mae_db": 0.9007114462871909, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1835230678712536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21878100942316048, "3000–6000 Hz_10ms_envelope_correlation": 0.9661466409109581, "3000–6000 Hz_10ms_modulation_mae_db": 0.5081251324544839, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6087540977000372, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15729019517876844, "3000–6000 Hz_40ms_envelope_correlation": 0.9808412734411505, "3000–6000 Hz_40ms_modulation_mae_db": 0.6019799459038907, "mr_stft_db": 5.661037255869658, "band_mae_db": 0.10353986785958845, "rms_gain_error_db": -0.021675630183648, "output_peak": 0.6163839061196398}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0192046942878672, "全帯域_10ms_rms_envelope_nrmse": 0.07946572688493918, "全帯域_10ms_envelope_correlation": 0.9933311925221657, "全帯域_10ms_modulation_mae_db": 0.7065762633168243, "全帯域_40ms_log_rms_mae_db": 0.553498663105241, "全帯域_40ms_rms_envelope_nrmse": 0.04798556495971033, "全帯域_40ms_envelope_correlation": 0.9972814578806525, "全帯域_40ms_modulation_mae_db": 0.8023690762989343, "300–1000 Hz_10ms_log_rms_mae_db": 1.6666606001209547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12345355437043283, "300–1000 Hz_10ms_envelope_correlation": 0.987190319482442, "300–1000 Hz_10ms_modulation_mae_db": 0.6334548445791935, "300–1000 Hz_40ms_log_rms_mae_db": 1.0610347192547318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07622865844563584, "300–1000 Hz_40ms_envelope_correlation": 0.9946703104961337, "300–1000 Hz_40ms_modulation_mae_db": 0.7496871000496196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8915551717626449, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1865699417662259, "1000–3000 Hz_10ms_envelope_correlation": 0.9715105817184517, "1000–3000 Hz_10ms_modulation_mae_db": 0.7077374828972713, "1000–3000 Hz_40ms_log_rms_mae_db": 1.17499096551313, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10811365330880138, "1000–3000 Hz_40ms_envelope_correlation": 0.9896430630020334, "1000–3000 Hz_40ms_modulation_mae_db": 0.9739345928709963, "3000–6000 Hz_10ms_log_rms_mae_db": 2.149004560068884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22025825658528358, "3000–6000 Hz_10ms_envelope_correlation": 0.964579678126793, "3000–6000 Hz_10ms_modulation_mae_db": 0.630970124632189, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5695819835251879, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15673261470063066, "3000–6000 Hz_40ms_envelope_correlation": 0.980337570001432, "3000–6000 Hz_40ms_modulation_mae_db": 0.8155563238626128, "mr_stft_db": 5.658691472127561, "band_mae_db": 0.11482422194791493, "rms_gain_error_db": -0.04146362143971949, "output_peak": 0.6320913196322335}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0584372352325042, "全帯域_10ms_rms_envelope_nrmse": 0.07234785277571602, "全帯域_10ms_envelope_correlation": 0.9944954796738755, "全帯域_10ms_modulation_mae_db": 0.3775641377672616, "全帯域_40ms_log_rms_mae_db": 0.5571049149816033, "全帯域_40ms_rms_envelope_nrmse": 0.035339583170481366, "全帯域_40ms_envelope_correlation": 0.9985658855241921, "全帯域_40ms_modulation_mae_db": 0.4378024960321049, "300–1000 Hz_10ms_log_rms_mae_db": 1.5948434166836105, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12825610771332674, "300–1000 Hz_10ms_envelope_correlation": 0.986241405789433, "300–1000 Hz_10ms_modulation_mae_db": 0.46390325120842224, "300–1000 Hz_40ms_log_rms_mae_db": 0.9762651258381134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06762536884149514, "300–1000 Hz_40ms_envelope_correlation": 0.9960522023929355, "300–1000 Hz_40ms_modulation_mae_db": 0.5173238964578897, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9137778805943875, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20183839732879383, "1000–3000 Hz_10ms_envelope_correlation": 0.9669984491611084, "1000–3000 Hz_10ms_modulation_mae_db": 1.1503773092374123, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2153726069504733, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1280656064862899, "1000–3000 Hz_40ms_envelope_correlation": 0.9854801909509883, "1000–3000 Hz_40ms_modulation_mae_db": 1.345386579237812, "3000–6000 Hz_10ms_log_rms_mae_db": 2.087012411325569, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21451050672225708, "3000–6000 Hz_10ms_envelope_correlation": 0.9667780413296239, "3000–6000 Hz_10ms_modulation_mae_db": 0.9092756167534263, "3000–6000 Hz_40ms_log_rms_mae_db": 1.526716703812122, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1529839107470086, "3000–6000 Hz_40ms_envelope_correlation": 0.981403202528056, "3000–6000 Hz_40ms_modulation_mae_db": 1.2062212924201532, "mr_stft_db": 5.659412373107396, "band_mae_db": 0.10429415697463684, "rms_gain_error_db": -0.0454465564124638, "output_peak": 0.624882072334514}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.042608586961441, "全帯域_10ms_rms_envelope_nrmse": 0.08366001913346037, "全帯域_10ms_envelope_correlation": 0.9926112565730772, "全帯域_10ms_modulation_mae_db": 0.46220437826601596, "全帯域_40ms_log_rms_mae_db": 0.5804544141310205, "全帯域_40ms_rms_envelope_nrmse": 0.04338279501618718, "全帯域_40ms_envelope_correlation": 0.9977715174023587, "全帯域_40ms_modulation_mae_db": 0.8450900721167415, "300–1000 Hz_10ms_log_rms_mae_db": 1.6528222165853188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12528540581138378, "300–1000 Hz_10ms_envelope_correlation": 0.9868212732283529, "300–1000 Hz_10ms_modulation_mae_db": 0.5323504336554193, "300–1000 Hz_40ms_log_rms_mae_db": 0.9952254064667752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07191321571853564, "300–1000 Hz_40ms_envelope_correlation": 0.9952337124304788, "300–1000 Hz_40ms_modulation_mae_db": 0.925475815145481, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8824302238425215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1732292210745298, "1000–3000 Hz_10ms_envelope_correlation": 0.9759115623721283, "1000–3000 Hz_10ms_modulation_mae_db": 0.8360892250952995, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1893753547555628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10367523189024601, "1000–3000 Hz_40ms_envelope_correlation": 0.9912846075546049, "1000–3000 Hz_40ms_modulation_mae_db": 0.8528332192457165, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1870294167397093, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23083029131678015, "3000–6000 Hz_10ms_envelope_correlation": 0.9615458379508657, "3000–6000 Hz_10ms_modulation_mae_db": 0.28607316313322456, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6621832297211836, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15794795146437163, "3000–6000 Hz_40ms_envelope_correlation": 0.9801941509265117, "3000–6000 Hz_40ms_modulation_mae_db": 0.6580767915942007, "mr_stft_db": 5.663340759447024, "band_mae_db": 0.1772148347785575, "rms_gain_error_db": -0.016734241934115344, "output_peak": 0.6587065526170026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0404565818394849, "全帯域_10ms_rms_envelope_nrmse": 0.07517580110258645, "全帯域_10ms_envelope_correlation": 0.9940641672741677, "全帯域_10ms_modulation_mae_db": 0.3200557638151506, "全帯域_40ms_log_rms_mae_db": 0.5864390220104779, "全帯域_40ms_rms_envelope_nrmse": 0.04535043262460241, "全帯域_40ms_envelope_correlation": 0.9976182820004148, "全帯域_40ms_modulation_mae_db": 0.42889682164264814, "300–1000 Hz_10ms_log_rms_mae_db": 1.677624680557215, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10695207955192178, "300–1000 Hz_10ms_envelope_correlation": 0.9904676874035547, "300–1000 Hz_10ms_modulation_mae_db": 0.5169279600414938, "300–1000 Hz_40ms_log_rms_mae_db": 1.0215239861439236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05437604527764937, "300–1000 Hz_40ms_envelope_correlation": 0.9974229465520614, "300–1000 Hz_40ms_modulation_mae_db": 0.7572699246652994, "1000–3000 Hz_10ms_log_rms_mae_db": 1.936702335763263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1909400344135247, "1000–3000 Hz_10ms_envelope_correlation": 0.9703721732727908, "1000–3000 Hz_10ms_modulation_mae_db": 1.1157764737487088, "1000–3000 Hz_40ms_log_rms_mae_db": 1.197134417656055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11568197017133657, "1000–3000 Hz_40ms_envelope_correlation": 0.9886661008511745, "1000–3000 Hz_40ms_modulation_mae_db": 1.2935302875153827, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0917691891483585, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20671900672067822, "3000–6000 Hz_10ms_envelope_correlation": 0.9690599844897608, "3000–6000 Hz_10ms_modulation_mae_db": 0.8565499999933298, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5387385068098816, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14719315873702202, "3000–6000 Hz_40ms_envelope_correlation": 0.9827415624263149, "3000–6000 Hz_40ms_modulation_mae_db": 1.2565288509247217, "mr_stft_db": 5.66154047280066, "band_mae_db": 0.16540638333132504, "rms_gain_error_db": -0.06281236569393703, "output_peak": 0.6378753939110161}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0744575933117433, "全帯域_10ms_rms_envelope_nrmse": 0.08721186721954405, "全帯域_10ms_envelope_correlation": 0.991959155776693, "全帯域_10ms_modulation_mae_db": 0.4441696871177799, "全帯域_40ms_log_rms_mae_db": 0.590377489269557, "全帯域_40ms_rms_envelope_nrmse": 0.046248638450689265, "全帯域_40ms_envelope_correlation": 0.997478091623913, "全帯域_40ms_modulation_mae_db": 0.7704357896604807, "300–1000 Hz_10ms_log_rms_mae_db": 1.6856947157634137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13664215359575863, "300–1000 Hz_10ms_envelope_correlation": 0.9842876582031888, "300–1000 Hz_10ms_modulation_mae_db": 0.6339853605327678, "300–1000 Hz_40ms_log_rms_mae_db": 1.0598683751732523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07523818291685845, "300–1000 Hz_40ms_envelope_correlation": 0.994815151212983, "300–1000 Hz_40ms_modulation_mae_db": 1.1130744869268197, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8349943122173429, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17042128930389466, "1000–3000 Hz_10ms_envelope_correlation": 0.9765265352764215, "1000–3000 Hz_10ms_modulation_mae_db": 0.30286490896315604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1270600627632406, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09343869050809668, "1000–3000 Hz_40ms_envelope_correlation": 0.9929044170607833, "1000–3000 Hz_40ms_modulation_mae_db": 0.38272951192774757, "3000–6000 Hz_10ms_log_rms_mae_db": 2.164529601083685, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22770626929479465, "3000–6000 Hz_10ms_envelope_correlation": 0.9626083797571791, "3000–6000 Hz_10ms_modulation_mae_db": 0.7885482790522126, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6029987113252686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16049935995718853, "3000–6000 Hz_40ms_envelope_correlation": 0.9795364990207799, "3000–6000 Hz_40ms_modulation_mae_db": 1.0053843015198185, "mr_stft_db": 5.6638687158969425, "band_mae_db": 0.11177398834871234, "rms_gain_error_db": -0.0365176527081166, "output_peak": 0.7212654603751186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0293924320156997, "全帯域_10ms_rms_envelope_nrmse": 0.07342291261068912, "全帯域_10ms_envelope_correlation": 0.9943218760771243, "全帯域_10ms_modulation_mae_db": 0.27205929241220445, "全帯域_40ms_log_rms_mae_db": 0.5421346554739456, "全帯域_40ms_rms_envelope_nrmse": 0.03785952023955754, "全帯域_40ms_envelope_correlation": 0.9983443782605766, "全帯域_40ms_modulation_mae_db": 0.41205926385909325, "300–1000 Hz_10ms_log_rms_mae_db": 1.7030036484338722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12918131008981526, "300–1000 Hz_10ms_envelope_correlation": 0.9860802995465318, "300–1000 Hz_10ms_modulation_mae_db": 0.8349113098157884, "300–1000 Hz_40ms_log_rms_mae_db": 1.0107410725231318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07322335220077039, "300–1000 Hz_40ms_envelope_correlation": 0.995377640017486, "300–1000 Hz_40ms_modulation_mae_db": 0.9891437380315645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8501808936837956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19454937750724477, "1000–3000 Hz_10ms_envelope_correlation": 0.9694136849240526, "1000–3000 Hz_10ms_modulation_mae_db": 0.801345362786036, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1397419862791187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1328935968284991, "1000–3000 Hz_40ms_envelope_correlation": 0.9842680774367423, "1000–3000 Hz_40ms_modulation_mae_db": 1.2453647034208235, "3000–6000 Hz_10ms_log_rms_mae_db": 2.126164740832373, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21604175539220702, "3000–6000 Hz_10ms_envelope_correlation": 0.9666607563306542, "3000–6000 Hz_10ms_modulation_mae_db": 0.9914505690350307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5751915555162275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15396769362845913, "3000–6000 Hz_40ms_envelope_correlation": 0.9814088278155743, "3000–6000 Hz_40ms_modulation_mae_db": 1.107997122315359, "mr_stft_db": 5.6615137365172385, "band_mae_db": 0.054712210186966655, "rms_gain_error_db": -0.05582577976330851, "output_peak": 0.6317413577059073}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.04286025122928, "spectral_convergence": 0.2300189723186392, "spectral_convergence_status": "ok", "amplitude_floor": 1.9341972091070217e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.396710972520469, "spectral_convergence": 0.2231534409264312, "spectral_convergence_status": "ok", "amplitude_floor": 1.0469896576426271e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.537373714985813, "spectral_convergence": 0.24258862133553358, "spectral_convergence_status": "ok", "amplitude_floor": 8.978224937278799e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002340108897553237, "output_power": 0.002382714952929884, "delta_db": 0.0783602239727215, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003418531859998669, "output_power": 0.0033189315166893105, "delta_db": -0.12841340582041025, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009669299209010629, "output_power": 0.0009383707432938106, "delta_db": -0.1302054054747568, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016643839195516676, "output_power": 0.00016197355010290544, "delta_db": -0.11809410132021249, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.655994855693747e-05, "output_power": 7.66105973754733e-05, "delta_db": 0.0028721583608452617, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.076442519269658e-05, "output_power": 3.9760991889606134e-05, "delta_db": -0.10824112942484218, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09436440406229807, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05673415854181935, "pitch": null, "output_rms_dbfs": -21.59996695360835, "output_peak": 0.6490371573824004, "reference_rms_dbfs": -21.54323279506653, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.658981646245188}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019906122792766575, "log_rms_mae_db": 0.984796844826022, "rms_envelope_nrmse": 0.07013794772063599, "envelope_correlation": 0.9948340892144749, "rms_gain_error_db": -0.05595084370531987, "reference_mean_power": 0.007012794178240996, "output_mean_power": 0.006923026729468655}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.192818590388021e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015356197181510699, "output_power": 0.001508030673861643, "reference_active": true, "delta_db": -0.0787350472438055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003622112164404606, "output_power": 0.0003467341175267193, "reference_active": true, "delta_db": -0.18965317853418515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.688762087391039e-05, "output_power": 4.38813237623218e-05, "reference_active": true, "delta_db": -0.2877847634260878}], "active_band_count": 3, "active_band_mae_db": 0.18539099640135948}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018229956701751592, "log_rms_mae_db": 0.5319767879602244, "rms_envelope_nrmse": 0.037103200019884784, "envelope_correlation": 0.9984177175068987, "rms_gain_error_db": -0.05482446852169813, "reference_mean_power": 0.0070179714231847195, "output_mean_power": 0.006929934800264365}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8443307374696506e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014276197130393435, "output_power": 0.0013999443639043654, "reference_active": true, "delta_db": -0.08501759762182813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018503527061359352, "output_power": 0.00018023122188220902, "reference_active": true, "delta_db": -0.1142449258379951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.327921114523828e-06, "output_power": 3.685241893335376e-06, "reference_active": true, "delta_db": -0.6981333671901767}], "active_band_count": 3, "active_band_mae_db": 0.29913196355}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015227518148493586, "log_rms_mae_db": 1.6631803811825314, "rms_envelope_nrmse": 0.12603902900641528, "envelope_correlation": 0.9868150502747745, "rms_gain_error_db": -0.18629499441747524, "reference_mean_power": 0.0024461505543329222, "output_mean_power": 0.0023434391647613726}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3707367109300066e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00065188242251555, "output_power": 0.0006229695906996778, "reference_active": true, "delta_db": -0.19702423036322195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000159373128948768, "output_power": 0.0001313023934546231, "reference_active": true, "delta_db": -0.8414245643340872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8816561344170712e-05, "output_power": 1.550673825651384e-05, "reference_active": true, "delta_db": -0.8401980431317169}], "active_band_count": 3, "active_band_mae_db": 0.6262156126096753}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013579343349222632, "log_rms_mae_db": 1.0416502323049834, "rms_envelope_nrmse": 0.07217172574257907, "envelope_correlation": 0.9955499724867959, "rms_gain_error_db": -0.18177122547336005, "reference_mean_power": 0.0024424650949128645, "output_mean_power": 0.0023423470574752536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.250982515304928e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006212562593255464, "output_power": 0.0005893660604546663, "reference_active": true, "delta_db": -0.2288565467579049}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.001783099426803e-05, "output_power": 6.644243140834863e-05, "reference_active": true, "delta_db": -0.8074125825212533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.744470580402989e-06, "output_power": 1.226344049744623e-06, "reference_active": true, "delta_db": -1.5305132148161658}], "active_band_count": 3, "active_band_mae_db": 0.8555941146984413}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.560305347439803e-05, "log_rms_mae_db": 1.8912674302203576, "rms_envelope_nrmse": 0.17438186219726795, "envelope_correlation": 0.9751338168956561, "rms_gain_error_db": -0.10410235245893015, "reference_mean_power": 0.0001982579107701744, "output_mean_power": 0.0001935620843241682}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.44612649270932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.480334346914858e-05, "output_power": 3.134468741946535e-05, "reference_active": true, "delta_db": -0.45457024267518026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0911626587448635e-05, "output_power": 1.0599332701618156e-05, "reference_active": true, "delta_db": -0.1261097094157176}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7887330599139967e-06, "output_power": 1.8394274782202543e-06, "reference_active": true, "delta_db": -1.8072427529988833}], "active_band_count": 3, "active_band_mae_db": 0.7959742350299269}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.513956828007693e-05, "log_rms_mae_db": 1.1512132750793227, "rms_envelope_nrmse": 0.1004805127335995, "envelope_correlation": 0.9908557328889557, "rms_gain_error_db": -0.10464050344841491, "reference_mean_power": 0.00019869265760544437, "output_mean_power": 0.00019396249786005434}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.64113687643322e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.340875190885911e-05, "output_power": 2.9963056593012866e-05, "reference_active": true, "delta_db": -0.47274136916690745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.757829899772919e-06, "output_power": 5.272207032324964e-06, "reference_active": true, "delta_db": -0.3826637455345155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2467549734271334e-07, "output_power": 1.486967272672077e-07, "reference_active": true, "delta_db": -1.7925430167440157}], "active_band_count": 3, "active_band_mae_db": 0.8826493771484797}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.7519381942341847e-05, "log_rms_mae_db": 2.174221458558142, "rms_envelope_nrmse": 0.22043089272470126, "envelope_correlation": 0.9644703854456425, "rms_gain_error_db": -0.12535434548504482, "reference_mean_power": 3.309162956541551e-05, "output_mean_power": 3.215012897877261e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.646064632792655e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.7250471980521725e-06, "output_power": 6.493740033045791e-06, "reference_active": true, "delta_db": -0.15200438349339515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3903630336351795e-06, "output_power": 2.399521193137099e-06, "reference_active": true, "delta_db": 0.016607261446773623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2031554280127943e-06, "output_power": 7.093118716160598e-07, "reference_active": true, "delta_db": -2.294845060990787}], "active_band_count": 3, "active_band_mae_db": 0.8211522353103186}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2421508904297867e-05, "log_rms_mae_db": 1.5888912274068352, "rms_envelope_nrmse": 0.15580610868839992, "envelope_correlation": 0.9805723428699097, "rms_gain_error_db": -0.1264396531638953, "reference_mean_power": 3.3247421347363265e-05, "output_mean_power": 3.229341710238416e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.522637378578784e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.9685546500066154e-06, "output_power": 6.662332617122417e-06, "reference_active": true, "delta_db": -0.1951639923118992}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.482924821402799e-06, "output_power": 1.337518591922088e-06, "reference_active": true, "delta_db": -0.4481930696014046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.259963132949643e-08, "output_power": 5.1910950965956464e-08, "reference_active": true, "delta_db": -2.513502727422239}], "active_band_count": 3, "active_band_mae_db": 1.0522865964451809}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8969858160424163, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b9333e00820c75d38e23084ee5207943b480ce6ef96bd5ff3a40bbcc34237b0d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9711339135825202, "standard_deviation": 0.004653177452763912, "minimum": 0.9632045050395525, "maximum": 0.9765265352764215, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.889175473751674, "standard_deviation": 0.033635907249817965, "minimum": 1.8349943122173429, "maximum": 1.936702335763263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8346538500309221, "standard_deviation": 0.26666164428844435, "minimum": 0.30286490896315604, "maximum": 1.1503773092374123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18794854516519427, "standard_deviation": 0.014723237920161572, "minimum": 0.17042128930389466, "maximum": 0.2116582377300727, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9890376647680162, "standard_deviation": 0.0029091264622076556, "minimum": 0.9842680774367423, "maximum": 0.9929044170607833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1740287373703002, "standard_deviation": 0.03145356642686172, "minimum": 1.1270600627632406, "maximum": 1.2153726069504733, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9846424647067686, "standard_deviation": 0.314133418698476, "minimum": 0.38272951192774757, "maximum": 1.345386579237812, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11183933769851448, "standard_deviation": 0.013468249983425115, "minimum": 0.09343869050809668, "maximum": 0.1328935968284991, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9652312130426847, "standard_deviation": 0.0024318826078168414, "minimum": 0.9615458379508657, "maximum": 0.9690599844897608, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1454068057034967, "standard_deviation": 0.039733679155918834, "minimum": 2.087012411325569, "maximum": 2.1870294167397093, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7240181400455269, "standard_deviation": 0.2342367022167179, "minimum": 0.28607316313322456, "maximum": 0.9914505690350307, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2194097485224828, "standard_deviation": 0.007538661726004652, "minimum": 0.20671900672067822, "maximum": 0.23083029131678015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9808794286287161, "standard_deviation": 0.0009772102121716583, "minimum": 0.9795364990207799, "maximum": 0.9827415624263149, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.584132001977093, "standard_deviation": 0.04263728278867135, "minimum": 1.526716703812122, "maximum": 1.6621832297211836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9630039031232422, "standard_deviation": 0.24528751473864174, "minimum": 0.6019799459038907, "maximum": 1.2565288509247217, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1553026241377311, "standard_deviation": 0.004020919022995613, "minimum": 0.14719315873702202, "maximum": 0.16049935995718853, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9868173861669374, "standard_deviation": 0.0017226015719564028, "minimum": 0.9842876582031888, "maximum": 0.9904676874035547, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6563094969812293, "standard_deviation": 0.037624456226733945, "minimum": 1.5948434166836105, "maximum": 1.7030036484338722, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6083906057406547, "standard_deviation": 0.11226581520733514, "minimum": 0.46390325120842224, "maximum": 0.8349113098157884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12527668849944604, "standard_deviation": 0.008402922016831588, "minimum": 0.10695207955192178, "maximum": 0.13664215359575863, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955391679784621, "standard_deviation": 0.0008727123216046721, "minimum": 0.9946703104961337, "maximum": 0.9974229465520614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.021680717913313, "standard_deviation": 0.030471299870491103, "minimum": 0.9762651258381134, "maximum": 1.0610347192547318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.80040226751997, "standard_deviation": 0.21696513811744972, "minimum": 0.49564906418464494, "maximum": 1.1130744869268197, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0705783236209202, "standard_deviation": 0.0070396378566121584, "minimum": 0.05437604527764937, "maximum": 0.07622865844563584, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11576625843624999, "standard_deviation": 0.03914008742241593, "minimum": 0.054712210186966655, "maximum": 0.1772148347785575, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.661048304001459, "standard_deviation": 0.001933577568671256, "minimum": 5.658691472127561, "maximum": 5.6638687158969425, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.646497902509729, "standard_deviation": 0.032992285795186524, "minimum": 0.6163839061196398, "maximum": 0.7212654603751186, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04215125083464102, "standard_deviation": 0.016625626902148083, "minimum": -0.06281236569393703, "maximum": -0.016734241934115344, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936643132188752, "standard_deviation": 0.0009841626399204758, "minimum": 0.991959155776693, "maximum": 0.9948340892144749, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.035067744076277, "standard_deviation": 0.02675192539867219, "minimum": 0.984796844826022, "maximum": 1.0744575933117433, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3634674777840783, "standard_deviation": 0.17976122606126335, "minimum": 0.13971930317603, "maximum": 0.7065762633168243, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07733618396248497, "standard_deviation": 0.005839993919079316, "minimum": 0.07013794772063599, "maximum": 0.08721186721954405, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979620575019637, "standard_deviation": 0.00048350145496400246, "minimum": 0.9972814578806525, "maximum": 0.9985658855241921, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5596440279865877, "standard_deviation": 0.023333044363186485, "minimum": 0.5319767879602244, "maximum": 0.590377489269557, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5243837945655573, "standard_deviation": 0.24678750171104188, "minimum": 0.19928487336445513, "maximum": 0.8450900721167415, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04152746907368754, "standard_deviation": 0.004781137314844594, "minimum": 0.035339583170481366, "maximum": 0.04798556495971033, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0002", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.9683306217193604, "predicted_mos": 2.9683306217193604, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "347606a6cc46ec9271c3db04af6f5f0ed7cb5ec7d69e9ddbf97567e13f85ea28", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7903981003910303, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9902616675516199, "全帯域_10ms_rms_envelope_nrmse": 0.07575872112930519, "全帯域_10ms_envelope_correlation": 0.9940121003100458, "全帯域_10ms_modulation_mae_db": 0.11008197984174341, "全帯域_40ms_log_rms_mae_db": 0.5597813783920297, "全帯域_40ms_rms_envelope_nrmse": 0.04640505327174857, "全帯域_40ms_envelope_correlation": 0.9975567586828754, "全帯域_40ms_modulation_mae_db": 0.2906942442903628, "300–1000 Hz_10ms_log_rms_mae_db": 1.6849852727050074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1247297065364729, "300–1000 Hz_10ms_envelope_correlation": 0.9870917148925846, "300–1000 Hz_10ms_modulation_mae_db": 0.592803276849176, "300–1000 Hz_40ms_log_rms_mae_db": 1.0272821112791755, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07178509780084241, "300–1000 Hz_40ms_envelope_correlation": 0.9955569884951523, "300–1000 Hz_40ms_modulation_mae_db": 0.7741990797133526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9975684098101498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19805269707805725, "1000–3000 Hz_10ms_envelope_correlation": 0.968189929878265, "1000–3000 Hz_10ms_modulation_mae_db": 1.0161132846268228, "1000–3000 Hz_40ms_log_rms_mae_db": 1.216805621296895, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1072718123720191, "1000–3000 Hz_40ms_envelope_correlation": 0.9908960748024864, "1000–3000 Hz_40ms_modulation_mae_db": 1.0826764033360146, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3850254392679457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21257192914569928, "3000–6000 Hz_10ms_envelope_correlation": 0.9682225926716438, "3000–6000 Hz_10ms_modulation_mae_db": 0.673567040817324, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7065748017240665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15224055051657637, "3000–6000 Hz_40ms_envelope_correlation": 0.9822196869845571, "3000–6000 Hz_40ms_modulation_mae_db": 0.8190778435535809, "mr_stft_db": 5.7941297445871776, "band_mae_db": 0.15184303889609727, "rms_gain_error_db": -0.09301484720912424, "output_peak": 0.6547071696621664}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.029160351502369, "全帯域_10ms_rms_envelope_nrmse": 0.0746726199630301, "全帯域_10ms_envelope_correlation": 0.9941508513943795, "全帯域_10ms_modulation_mae_db": 0.33988701708174696, "全帯域_40ms_log_rms_mae_db": 0.5517739584810263, "全帯域_40ms_rms_envelope_nrmse": 0.0410802280520459, "全帯域_40ms_envelope_correlation": 0.9980775375326011, "全帯域_40ms_modulation_mae_db": 0.6829272560171437, "300–1000 Hz_10ms_log_rms_mae_db": 1.7505216942861592, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13829691720856968, "300–1000 Hz_10ms_envelope_correlation": 0.9841812621840811, "300–1000 Hz_10ms_modulation_mae_db": 0.6078080904861279, "300–1000 Hz_40ms_log_rms_mae_db": 1.0729782956452616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0859079993188878, "300–1000 Hz_40ms_envelope_correlation": 0.9937673911716783, "300–1000 Hz_40ms_modulation_mae_db": 1.1702645070378292, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0386684074542027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20178937791839446, "1000–3000 Hz_10ms_envelope_correlation": 0.9666055471293381, "1000–3000 Hz_10ms_modulation_mae_db": 0.9795609363917546, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2508713398005613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11058351348357724, "1000–3000 Hz_40ms_envelope_correlation": 0.988888612149235, "1000–3000 Hz_40ms_modulation_mae_db": 0.9558174975542183, "3000–6000 Hz_10ms_log_rms_mae_db": 2.32220044967057, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23298736248973573, "3000–6000 Hz_10ms_envelope_correlation": 0.9618731504157215, "3000–6000 Hz_10ms_modulation_mae_db": 0.5815651433560828, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6499118752458604, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16297928613400564, "3000–6000 Hz_40ms_envelope_correlation": 0.9796483746189083, "3000–6000 Hz_40ms_modulation_mae_db": 0.5259680045494542, "mr_stft_db": 5.80031063991105, "band_mae_db": 0.12911442420198996, "rms_gain_error_db": -0.06796727285687525, "output_peak": 0.6199619114020759}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0509200312417495, "全帯域_10ms_rms_envelope_nrmse": 0.08890108653919815, "全帯域_10ms_envelope_correlation": 0.9917020581470852, "全帯域_10ms_modulation_mae_db": 0.4980192158625609, "全帯域_40ms_log_rms_mae_db": 0.589652869021777, "全帯域_40ms_rms_envelope_nrmse": 0.0502827821408666, "全帯域_40ms_envelope_correlation": 0.9971149225853497, "全帯域_40ms_modulation_mae_db": 0.6418105525300989, "300–1000 Hz_10ms_log_rms_mae_db": 1.7100115014684913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13626207848564154, "300–1000 Hz_10ms_envelope_correlation": 0.9847289804076986, "300–1000 Hz_10ms_modulation_mae_db": 0.6472610464484495, "300–1000 Hz_40ms_log_rms_mae_db": 1.082682157748573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09361438205525077, "300–1000 Hz_40ms_envelope_correlation": 0.9924824533959412, "300–1000 Hz_40ms_modulation_mae_db": 0.9451963750963372, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0640592766318937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22094953846962861, "1000–3000 Hz_10ms_envelope_correlation": 0.9602185962808769, "1000–3000 Hz_10ms_modulation_mae_db": 1.4227668364421564, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2811992218487849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15713692284769318, "1000–3000 Hz_40ms_envelope_correlation": 0.9778746202022782, "1000–3000 Hz_40ms_modulation_mae_db": 1.2271482352621723, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4391327044359383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23578505636012334, "3000–6000 Hz_10ms_envelope_correlation": 0.9602381470332073, "3000–6000 Hz_10ms_modulation_mae_db": 0.9743567350244016, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7563022507243415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16202891225686913, "3000–6000 Hz_40ms_envelope_correlation": 0.9793989808579933, "3000–6000 Hz_40ms_modulation_mae_db": 1.2662965395904637, "mr_stft_db": 5.8000113608696955, "band_mae_db": 0.13240981149293504, "rms_gain_error_db": -0.09840472684596298, "output_peak": 0.6276492587519327}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.049856471666231, "全帯域_10ms_rms_envelope_nrmse": 0.07300911840658461, "全帯域_10ms_envelope_correlation": 0.9944165166338643, "全帯域_10ms_modulation_mae_db": 0.7378541017045994, "全帯域_40ms_log_rms_mae_db": 0.5555783687748702, "全帯域_40ms_rms_envelope_nrmse": 0.036931249751762124, "全帯域_40ms_envelope_correlation": 0.998454957868454, "全帯域_40ms_modulation_mae_db": 0.9519947296301282, "300–1000 Hz_10ms_log_rms_mae_db": 1.768132488144803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11428175750629287, "300–1000 Hz_10ms_envelope_correlation": 0.9892960744027235, "300–1000 Hz_10ms_modulation_mae_db": 0.45230690655055605, "300–1000 Hz_40ms_log_rms_mae_db": 1.0630960728016339, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06452724495504847, "300–1000 Hz_40ms_envelope_correlation": 0.9965951758670212, "300–1000 Hz_40ms_modulation_mae_db": 0.5962164845458635, "1000–3000 Hz_10ms_log_rms_mae_db": 2.029320553787313, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19683171741992803, "1000–3000 Hz_10ms_envelope_correlation": 0.9682041913229685, "1000–3000 Hz_10ms_modulation_mae_db": 1.1815019263533773, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1606684029518897, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10508165977168624, "1000–3000 Hz_40ms_envelope_correlation": 0.9903315183685427, "1000–3000 Hz_40ms_modulation_mae_db": 1.0140096000463252, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3847131427554853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2442716771015206, "3000–6000 Hz_10ms_envelope_correlation": 0.9576318369359628, "3000–6000 Hz_10ms_modulation_mae_db": 0.546513099768388, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7764397401784413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17916616523426768, "3000–6000 Hz_40ms_envelope_correlation": 0.9749919078341552, "3000–6000 Hz_40ms_modulation_mae_db": 0.5936879753323651, "mr_stft_db": 5.7913273059897605, "band_mae_db": 0.1114022447581799, "rms_gain_error_db": -0.06851560783823497, "output_peak": 0.6268860559692566}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.057423561965794, "全帯域_10ms_rms_envelope_nrmse": 0.07664128875911512, "全帯域_10ms_envelope_correlation": 0.9938633902035385, "全帯域_10ms_modulation_mae_db": 0.442038272844224, "全帯域_40ms_log_rms_mae_db": 0.5902206399635646, "全帯域_40ms_rms_envelope_nrmse": 0.04181953230290343, "全帯域_40ms_envelope_correlation": 0.9980424559847819, "全帯域_40ms_modulation_mae_db": 0.6382912966587946, "300–1000 Hz_10ms_log_rms_mae_db": 1.8423309613320937, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1317582619519428, "300–1000 Hz_10ms_envelope_correlation": 0.9854694789838568, "300–1000 Hz_10ms_modulation_mae_db": 0.3884864376220249, "300–1000 Hz_40ms_log_rms_mae_db": 1.0904812459643558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07628113452387643, "300–1000 Hz_40ms_envelope_correlation": 0.9948463042915057, "300–1000 Hz_40ms_modulation_mae_db": 0.7670464720192184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9933948553766219, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1790800674063796, "1000–3000 Hz_10ms_envelope_correlation": 0.9746880081058855, "1000–3000 Hz_10ms_modulation_mae_db": 1.0197281714531132, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2206749355588922, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10229334357964147, "1000–3000 Hz_40ms_envelope_correlation": 0.9924470823077463, "1000–3000 Hz_40ms_modulation_mae_db": 1.0837073757414786, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3733820013151545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2398086776184378, "3000–6000 Hz_10ms_envelope_correlation": 0.9601150823786628, "3000–6000 Hz_10ms_modulation_mae_db": 0.630304533276179, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7249068062894637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17061313451313784, "3000–6000 Hz_40ms_envelope_correlation": 0.9781121779758245, "3000–6000 Hz_40ms_modulation_mae_db": 1.004471811398312, "mr_stft_db": 5.809805950333822, "band_mae_db": 0.20711115030120877, "rms_gain_error_db": -0.09152489730359842, "output_peak": 0.60534477797273}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.002176715322572, "全帯域_10ms_rms_envelope_nrmse": 0.07676033127787929, "全帯域_10ms_envelope_correlation": 0.9937911582434291, "全帯域_10ms_modulation_mae_db": 0.13791161607556768, "全帯域_40ms_log_rms_mae_db": 0.5408821126198351, "全帯域_40ms_rms_envelope_nrmse": 0.04423360738438579, "全帯域_40ms_envelope_correlation": 0.9977117979297083, "全帯域_40ms_modulation_mae_db": 0.22621668102498602, "300–1000 Hz_10ms_log_rms_mae_db": 1.7951585919594455, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1337886235606484, "300–1000 Hz_10ms_envelope_correlation": 0.9849705510610924, "300–1000 Hz_10ms_modulation_mae_db": 0.40529181215893145, "300–1000 Hz_40ms_log_rms_mae_db": 1.0953542995105523, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07318769987749604, "300–1000 Hz_40ms_envelope_correlation": 0.9950584553674802, "300–1000 Hz_40ms_modulation_mae_db": 0.5310716224361737, "1000–3000 Hz_10ms_log_rms_mae_db": 2.028246354248927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21182327162361528, "1000–3000 Hz_10ms_envelope_correlation": 0.9634888842467828, "1000–3000 Hz_10ms_modulation_mae_db": 0.7898023680729042, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2504231342635979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12325485435393306, "1000–3000 Hz_40ms_envelope_correlation": 0.9861596960283984, "1000–3000 Hz_40ms_modulation_mae_db": 0.8003174974336376, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4171590040183544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23769657653403123, "3000–6000 Hz_10ms_envelope_correlation": 0.9584913008020517, "3000–6000 Hz_10ms_modulation_mae_db": 1.1036830777959785, "3000–6000 Hz_40ms_log_rms_mae_db": 1.753075605251075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17586653226012208, "3000–6000 Hz_40ms_envelope_correlation": 0.9754470582002601, "3000–6000 Hz_40ms_modulation_mae_db": 1.3740730640325118, "mr_stft_db": 5.80636321234575, "band_mae_db": 0.058116310592599314, "rms_gain_error_db": -0.037428951294485355, "output_peak": 0.636032475699119}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.019096830828543, "全帯域_10ms_rms_envelope_nrmse": 0.07521634097688167, "全帯域_10ms_envelope_correlation": 0.9940312267705362, "全帯域_10ms_modulation_mae_db": 0.227875069741722, "全帯域_40ms_log_rms_mae_db": 0.5308885276704451, "全帯域_40ms_rms_envelope_nrmse": 0.039183308963849015, "全帯域_40ms_envelope_correlation": 0.9981807671852745, "全帯域_40ms_modulation_mae_db": 0.49383284714163955, "300–1000 Hz_10ms_log_rms_mae_db": 1.7992028994682778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12647171039357877, "300–1000 Hz_10ms_envelope_correlation": 0.9865584941915548, "300–1000 Hz_10ms_modulation_mae_db": 0.5829526177225682, "300–1000 Hz_40ms_log_rms_mae_db": 1.0198129994128504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06184310669094209, "300–1000 Hz_40ms_envelope_correlation": 0.9965383362619387, "300–1000 Hz_40ms_modulation_mae_db": 0.9232668076728258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9912383834426017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1938270553051076, "1000–3000 Hz_10ms_envelope_correlation": 0.9691834382436219, "1000–3000 Hz_10ms_modulation_mae_db": 0.7810470174234668, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2012478945342, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10718328052662096, "1000–3000 Hz_40ms_envelope_correlation": 0.9899679253482488, "1000–3000 Hz_40ms_modulation_mae_db": 0.994612569806955, "3000–6000 Hz_10ms_log_rms_mae_db": 2.342404023979405, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22203461653740705, "3000–6000 Hz_10ms_envelope_correlation": 0.9640294600327747, "3000–6000 Hz_10ms_modulation_mae_db": 1.1668199971116666, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7095119562267411, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15386725367323356, "3000–6000 Hz_40ms_envelope_correlation": 0.9810700245649987, "3000–6000 Hz_40ms_modulation_mae_db": 1.4557755532419145, "mr_stft_db": 5.784210539742027, "band_mae_db": 0.11180293339397852, "rms_gain_error_db": -0.005208883510696871, "output_peak": 0.6440418893975685}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0328186300112243, "全帯域_10ms_rms_envelope_nrmse": 0.07958538762114027, "全帯域_10ms_envelope_correlation": 0.9933179260595404, "全帯域_10ms_modulation_mae_db": 0.08789053637373383, "全帯域_40ms_log_rms_mae_db": 0.5583251457206974, "全帯域_40ms_rms_envelope_nrmse": 0.04526582774191747, "全帯域_40ms_envelope_correlation": 0.9976027518758257, "全帯域_40ms_modulation_mae_db": 0.07492886071440437, "300–1000 Hz_10ms_log_rms_mae_db": 1.7162045156136374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12518195767395762, "300–1000 Hz_10ms_envelope_correlation": 0.9871348682717216, "300–1000 Hz_10ms_modulation_mae_db": 0.33617702853675735, "300–1000 Hz_40ms_log_rms_mae_db": 0.9806278962282654, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07105638167837584, "300–1000 Hz_40ms_envelope_correlation": 0.9959321329327036, "300–1000 Hz_40ms_modulation_mae_db": 0.374463147378059, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0415716488821456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1759542516103578, "1000–3000 Hz_10ms_envelope_correlation": 0.9747088164621736, "1000–3000 Hz_10ms_modulation_mae_db": 0.9254496224416782, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2676862657574803, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10244471895638266, "1000–3000 Hz_40ms_envelope_correlation": 0.990812418232362, "1000–3000 Hz_40ms_modulation_mae_db": 0.803498979771533, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3069922638127065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21637118506766562, "3000–6000 Hz_10ms_envelope_correlation": 0.9660164585216967, "3000–6000 Hz_10ms_modulation_mae_db": 0.6924436321240236, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6359259677117084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15398898570635297, "3000–6000 Hz_40ms_envelope_correlation": 0.9811006766517474, "3000–6000 Hz_40ms_modulation_mae_db": 0.659119654150455, "mr_stft_db": 5.763617275697111, "band_mae_db": 0.16762441983337983, "rms_gain_error_db": -0.05295869274950735, "output_peak": 0.6175010653222428}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2685353071993, "spectral_convergence": 0.23193179952197146, "spectral_convergence_status": "ok", "amplitude_floor": 1.9341972091070217e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.550084904468893, "spectral_convergence": 0.2212177628381431, "spectral_convergence_status": "ok", "amplitude_floor": 1.0469896576426271e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.563769022093339, "spectral_convergence": 0.24572858504240702, "spectral_convergence_status": "ok", "amplitude_floor": 8.978224937278799e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002340108897553237, "output_power": 0.002328480275391046, "delta_db": -0.021635045168318347, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003418531859998669, "output_power": 0.0032998345792126455, "delta_db": -0.15347462557568814, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009669299209010629, "output_power": 0.000966499337115711, "delta_db": -0.0019343884891735504, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016643839195516676, "output_power": 0.0001527874869275682, "delta_db": -0.37165723593069855, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.655994855693747e-05, "output_power": 7.170452456639419e-05, "delta_db": -0.28455072372394213, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.076442519269658e-05, "output_power": 4.1501323258178236e-05, "delta_db": 0.07780621448876275, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15184303889609727, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09301484720912424, "pitch": null, "output_rms_dbfs": -21.636247642275656, "output_peak": 0.6547071696621664, "reference_rms_dbfs": -21.54323279506653, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7941297445871776}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019906122792766575, "log_rms_mae_db": 0.9902616675516199, "rms_envelope_nrmse": 0.07575872112930519, "envelope_correlation": 0.9940121003100458, "rms_gain_error_db": -0.09259337711270946, "reference_mean_power": 0.007012794178240996, "output_mean_power": 0.006864861110889596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.192818590388021e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015356197181510699, "output_power": 0.0015290578349359402, "reference_active": true, "delta_db": -0.018597676112931977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003622112164404606, "output_power": 0.0003663513621806527, "reference_active": true, "delta_db": 0.049359158886892504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.688762087391039e-05, "output_power": 4.4139690770304576e-05, "reference_active": true, "delta_db": -0.26228910452540577}], "active_band_count": 3, "active_band_mae_db": 0.11008197984174341}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018229956701751592, "log_rms_mae_db": 0.5597813783920297, "rms_envelope_nrmse": 0.04640505327174857, "envelope_correlation": 0.9975567586828754, "rms_gain_error_db": -0.08958291482920083, "reference_mean_power": 0.0070179714231847195, "output_mean_power": 0.0068746929228922565}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8443307374696506e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014276197130393435, "output_power": 0.001423117263113174, "reference_active": true, "delta_db": -0.013718493149294984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018503527061359352, "output_power": 0.00019112136299422448, "reference_active": true, "delta_db": 0.14054714344105027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.327921114523828e-06, "output_power": 3.6685768958430464e-06, "reference_active": true, "delta_db": -0.717817096280743}], "active_band_count": 3, "active_band_mae_db": 0.2906942442903628}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015227518148493586, "log_rms_mae_db": 1.6849852727050074, "rms_envelope_nrmse": 0.1247297065364729, "envelope_correlation": 0.9870917148925846, "rms_gain_error_db": -0.17275069929448889, "reference_mean_power": 0.0024461505543329222, "output_mean_power": 0.002350759031447707}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3707367109300066e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00065188242251555, "output_power": 0.0006259191233733508, "reference_active": true, "delta_db": -0.17651050273356195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000159373128948768, "output_power": 0.00014496373343703112, "reference_active": true, "delta_db": -0.41155733741780276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8816561344170712e-05, "output_power": 1.430559947569396e-05, "reference_active": true, "delta_db": -1.1903419903961632}], "active_band_count": 3, "active_band_mae_db": 0.592803276849176}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013579343349222632, "log_rms_mae_db": 1.0272821112791755, "rms_envelope_nrmse": 0.07178509780084241, "envelope_correlation": 0.9955569884951523, "rms_gain_error_db": -0.1693516016644459, "reference_mean_power": 0.0024424650949128645, "output_mean_power": 0.002349055110741583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.250982515304928e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006212562593255464, "output_power": 0.0005954096059202474, "reference_active": true, "delta_db": -0.18454940429029146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.001783099426803e-05, "output_power": 7.184932259225301e-05, "reference_active": true, "delta_db": -0.46764097099528174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.744470580402989e-06, "output_power": 1.1874708662099895e-06, "reference_active": true, "delta_db": -1.6704068638544842}], "active_band_count": 3, "active_band_mae_db": 0.7741990797133526}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.560305347439803e-05, "log_rms_mae_db": 1.9975684098101498, "rms_envelope_nrmse": 0.19805269707805725, "envelope_correlation": 0.968189929878265, "rms_gain_error_db": -0.38022206512105705, "reference_mean_power": 0.0001982579107701744, "output_mean_power": 0.0001816386723206643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.44612649270932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.480334346914858e-05, "output_power": 3.054398226890704e-05, "reference_active": true, "delta_db": -0.5669530858549582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0911626587448635e-05, "output_power": 8.582775016331583e-06, "reference_active": true, "delta_db": -1.0426176700377026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7887330599139967e-06, "output_power": 2.002304094235207e-06, "reference_active": true, "delta_db": -1.4387690979878078}], "active_band_count": 3, "active_band_mae_db": 1.0161132846268228}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.513956828007693e-05, "log_rms_mae_db": 1.216805621296895, "rms_envelope_nrmse": 0.1072718123720191, "envelope_correlation": 0.9908960748024864, "rms_gain_error_db": -0.3920320550870537, "reference_mean_power": 0.00019869265760544437, "output_mean_power": 0.00018154262599741347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.64113687643322e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.340875190885911e-05, "output_power": 2.893888992821407e-05, "reference_active": true, "delta_db": -0.6237838355024748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.757829899772919e-06, "output_power": 4.4756582450176965e-06, "reference_active": true, "delta_db": -1.0940191361652716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2467549734271334e-07, "output_power": 1.5795489453129125e-07, "reference_active": true, "delta_db": -1.5302262383402971}], "active_band_count": 3, "active_band_mae_db": 1.0826764033360146}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.7519381942341847e-05, "log_rms_mae_db": 2.3850254392679457, "rms_envelope_nrmse": 0.21257192914569928, "envelope_correlation": 0.9682225926716438, "rms_gain_error_db": 0.17037038425554984, "reference_mean_power": 3.309162956541551e-05, "output_mean_power": 3.441558788341776e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.646064632792655e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.7250471980521725e-06, "output_power": 7.003267404424354e-06, "reference_active": true, "delta_db": 0.17605372576280248}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3903630336351795e-06, "output_power": 2.216281363151222e-06, "reference_active": true, "delta_db": -0.3283896946982854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2031554280127943e-06, "output_power": 8.485861993556952e-07, "reference_active": true, "delta_db": -1.516257701990884}], "active_band_count": 3, "active_band_mae_db": 0.673567040817324}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2421508904297867e-05, "log_rms_mae_db": 1.7065748017240665, "rms_envelope_nrmse": 0.15224055051657637, "envelope_correlation": 0.9822196869845571, "rms_gain_error_db": 0.16875728948506669, "reference_mean_power": 3.3247421347363265e-05, "output_mean_power": 3.456477198044832e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.522637378578784e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.9685546500066154e-06, "output_power": 7.134857491668324e-06, "reference_active": true, "delta_db": 0.10242592850433108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.482924821402799e-06, "output_power": 1.2646881427129592e-06, "reference_active": true, "delta_db": -0.6913568774333989}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.259963132949643e-08, "output_power": 6.313412781259752e-08, "reference_active": true, "delta_db": -1.6634507247230128}], "active_band_count": 3, "active_band_mae_db": 0.8190778435535809}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8957919683375861, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "86effc9465af462822c4380a17a856f744b5968a4e9bb5f5a90308a4896a1c4e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.968160926458739, "standard_deviation": 0.0049835972903322315, "minimum": 0.9602185962808769, "maximum": 0.9747088164621736, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.023008486204232, "standard_deviation": 0.026392025426902985, "minimum": 1.9912383834426017, "maximum": 2.0640592766318937, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0144962704006593, "standard_deviation": 0.21001408737565364, "minimum": 0.7810470174234668, "maximum": 1.4227668364421564, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1972884971039336, "standard_deviation": 0.015075209775327253, "minimum": 0.1759542516103578, "maximum": 0.22094953846962861, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9884222434299121, "standard_deviation": 0.00463920474868056, "minimum": 0.9778746202022782, "maximum": 0.9924470823077463, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2311971020015375, "standard_deviation": 0.03923822868329353, "minimum": 1.1606684029518897, "maximum": 1.2811992218487849, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9952235198690418, "standard_deviation": 0.1443800810891666, "minimum": 0.8003174974336376, "maximum": 1.2271482352621723, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11440626323644423, "standard_deviation": 0.018511161550204092, "minimum": 0.10229334357964147, "maximum": 0.15713692284769318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9620772535989651, "standard_deviation": 0.0037222978970075492, "minimum": 0.9576318369359628, "maximum": 0.9682225926716438, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.371376128656945, "standard_deviation": 0.045447207510461295, "minimum": 2.3069922638127065, "maximum": 2.4391327044359383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7961566574092556, "standard_deviation": 0.24651341164960927, "minimum": 0.546513099768388, "maximum": 1.1668199971116666, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23019088510682756, "standard_deviation": 0.011675110968380685, "minimum": 0.21257192914569928, "maximum": 0.2442716771015206, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9789986109610556, "standard_deviation": 0.00264899480931199, "minimum": 0.9749919078341552, "maximum": 0.9822196869845571, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.714081125418962, "standard_deviation": 0.05015992798326483, "minimum": 1.6359259677117084, "maximum": 1.7764397401784413, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9623088057311322, "standard_deviation": 0.36760982386986674, "minimum": 0.5259680045494542, "maximum": 1.4557755532419145, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16384385253682066, "standard_deviation": 0.010422984894598324, "minimum": 0.15224055051657637, "maximum": 0.17916616523426768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9861789280494142, "standard_deviation": 0.0016760575923303943, "minimum": 0.9841812621840811, "maximum": 0.9892960744027235, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7583184906222393, "standard_deviation": 0.05308865995496952, "minimum": 1.6849852727050074, "maximum": 1.8423309613320937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.501635902046824, "standard_deviation": 0.11910264980407444, "minimum": 0.33617702853675735, "maximum": 0.6472610464484495, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12884637666463808, "standard_deviation": 0.007795166460130503, "minimum": 0.11428175750629287, "maximum": 0.13829691720856968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9950971547229276, "standard_deviation": 0.0014091063528780424, "minimum": 0.9924824533959412, "maximum": 0.9965951758670212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0540393848238334, "standard_deviation": 0.04066007097819929, "minimum": 0.9806278962282654, "maximum": 1.0953542995105523, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7602155619874573, "standard_deviation": 0.2556206774518238, "minimum": 0.374463147378059, "maximum": 1.1702645070378292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07477538086258997, "standard_deviation": 0.01054698289411416, "minimum": 0.06184310669094209, "maximum": 0.09361438205525077, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13367804168379607, "standard_deviation": 0.04404784227504761, "minimum": 0.058116310592599314, "maximum": 0.20711115030120877, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.793722003684549, "standard_deviation": 0.01465860576385478, "minimum": 5.763617275697111, "maximum": 5.809805950333822, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6290155755221365, "standard_deviation": 0.015658617513518634, "minimum": 0.60534477797273, "maximum": 0.6547071696621664, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06437798495106067, "standard_deviation": 0.03185341430396662, "minimum": -0.09840472684596298, "maximum": -0.005208883510696871, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936606534703024, "standard_deviation": 0.0008522368865721318, "minimum": 0.9917020581470852, "maximum": 0.9944165166338643, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.028964282511263, "standard_deviation": 0.024071904957818933, "minimum": 0.9902616675516199, "maximum": 1.057423561965794, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32269472619073725, "standard_deviation": 0.227127149175549, "minimum": 0.08789053637373383, "maximum": 0.7378541017045994, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0775681118341418, "standard_deviation": 0.0049568169919490795, "minimum": 0.07300911840658461, "maximum": 0.08890108653919815, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978427437056088, "standard_deviation": 0.00042600018117895243, "minimum": 0.9971149225853497, "maximum": 0.998454957868454, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5596378750805306, "standard_deviation": 0.021025744758251738, "minimum": 0.5308885276704451, "maximum": 0.5902206399635646, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5000870585009447, "standard_deviation": 0.2869758536045611, "minimum": 0.07492886071440437, "maximum": 0.9519947296301282, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04315019870118486, "standard_deviation": 0.004269233407063124, "minimum": 0.036931249751762124, "maximum": 0.0502827821408666, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0002", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.89178204536438, "predicted_mos": 2.89178204536438, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "88e6f774766acf3fd7dbfa5364cf8a37f938206c545cc667d584a38810912828", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7876400002278388, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0419614710725116, "全帯域_10ms_rms_envelope_nrmse": 0.0793860472181451, "全帯域_10ms_envelope_correlation": 0.9934868750542311, "全帯域_10ms_modulation_mae_db": 0.4388966597430617, "全帯域_40ms_log_rms_mae_db": 0.5561920000521707, "全帯域_40ms_rms_envelope_nrmse": 0.042104614046314895, "全帯域_40ms_envelope_correlation": 0.9981288309209572, "全帯域_40ms_modulation_mae_db": 0.519193389947799, "300–1000 Hz_10ms_log_rms_mae_db": 1.6418734340013657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11549813860882774, "300–1000 Hz_10ms_envelope_correlation": 0.9889363122452952, "300–1000 Hz_10ms_modulation_mae_db": 0.4355543243641298, "300–1000 Hz_40ms_log_rms_mae_db": 0.9558321603191158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059729388320832716, "300–1000 Hz_40ms_envelope_correlation": 0.9970250485503686, "300–1000 Hz_40ms_modulation_mae_db": 0.7076630084766696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.980592941064255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1854593943941509, "1000–3000 Hz_10ms_envelope_correlation": 0.9719127935458339, "1000–3000 Hz_10ms_modulation_mae_db": 1.132999415880622, "1000–3000 Hz_40ms_log_rms_mae_db": 1.196743481232035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10075117636730142, "1000–3000 Hz_40ms_envelope_correlation": 0.9913582083820092, "1000–3000 Hz_40ms_modulation_mae_db": 1.141561156673269, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3549512537716444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22689790708014737, "3000–6000 Hz_10ms_envelope_correlation": 0.9625876625660849, "3000–6000 Hz_10ms_modulation_mae_db": 0.9452287276635811, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6835884419453562, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1581974032331479, "3000–6000 Hz_40ms_envelope_correlation": 0.9800433544514824, "3000–6000 Hz_40ms_modulation_mae_db": 1.1388107952428865, "mr_stft_db": 5.778518202135834, "band_mae_db": 0.15886126549067583, "rms_gain_error_db": -0.1274285893127981, "output_peak": 0.6267310956187829}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0041894368971047, "全帯域_10ms_rms_envelope_nrmse": 0.07043560187131451, "全帯域_10ms_envelope_correlation": 0.9947815721692077, "全帯域_10ms_modulation_mae_db": 0.7374723302459968, "全帯域_40ms_log_rms_mae_db": 0.531619958001679, "全帯域_40ms_rms_envelope_nrmse": 0.03490694007623789, "全帯域_40ms_envelope_correlation": 0.998591099288719, "全帯域_40ms_modulation_mae_db": 0.7410981494488987, "300–1000 Hz_10ms_log_rms_mae_db": 1.7147024464628566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11756058897745778, "300–1000 Hz_10ms_envelope_correlation": 0.9884107610035896, "300–1000 Hz_10ms_modulation_mae_db": 0.7216700747587961, "300–1000 Hz_40ms_log_rms_mae_db": 1.0361765296709857, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06193221939744499, "300–1000 Hz_40ms_envelope_correlation": 0.9965311849269575, "300–1000 Hz_40ms_modulation_mae_db": 0.918983066172706, "1000–3000 Hz_10ms_log_rms_mae_db": 2.029534201872327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.176784593263863, "1000–3000 Hz_10ms_envelope_correlation": 0.9744095344634893, "1000–3000 Hz_10ms_modulation_mae_db": 0.7272616589108184, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2122733777325632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10116702988588877, "1000–3000 Hz_40ms_envelope_correlation": 0.9909355769134996, "1000–3000 Hz_40ms_modulation_mae_db": 0.6692259942983765, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3450945252681503, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2280174592105605, "3000–6000 Hz_10ms_envelope_correlation": 0.962630528686467, "3000–6000 Hz_10ms_modulation_mae_db": 0.44561396183477425, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6789986001199426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15813375532176202, "3000–6000 Hz_40ms_envelope_correlation": 0.98025577905723, "3000–6000 Hz_40ms_modulation_mae_db": 0.6511194782267586, "mr_stft_db": 5.79302278395797, "band_mae_db": 0.09747295072631301, "rms_gain_error_db": -0.05449567472128536, "output_peak": 0.635016419524921}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0238550932216453, "全帯域_10ms_rms_envelope_nrmse": 0.07867498917768612, "全帯域_10ms_envelope_correlation": 0.9934970215475422, "全帯域_10ms_modulation_mae_db": 0.08308240781515487, "全帯域_40ms_log_rms_mae_db": 0.5587232116956969, "全帯域_40ms_rms_envelope_nrmse": 0.04757647948018286, "全帯域_40ms_envelope_correlation": 0.9973371475225606, "全帯域_40ms_modulation_mae_db": 0.224669840662199, "300–1000 Hz_10ms_log_rms_mae_db": 1.692263583419897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11718100272388776, "300–1000 Hz_10ms_envelope_correlation": 0.9884992720753425, "300–1000 Hz_10ms_modulation_mae_db": 0.8279142206073856, "300–1000 Hz_40ms_log_rms_mae_db": 0.9866731906562228, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06646975455564795, "300–1000 Hz_40ms_envelope_correlation": 0.9959955736946802, "300–1000 Hz_40ms_modulation_mae_db": 1.0749861242295176, "1000–3000 Hz_10ms_log_rms_mae_db": 1.982579372058272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1880526348478538, "1000–3000 Hz_10ms_envelope_correlation": 0.9714957539691246, "1000–3000 Hz_10ms_modulation_mae_db": 0.8525459820509423, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1857871172630008, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09614047568928348, "1000–3000 Hz_40ms_envelope_correlation": 0.9930012824972978, "1000–3000 Hz_40ms_modulation_mae_db": 0.850557158493674, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3504529056842163, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23317833345910252, "3000–6000 Hz_10ms_envelope_correlation": 0.9600983737886996, "3000–6000 Hz_10ms_modulation_mae_db": 1.45804355260524, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6870551810629355, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16518084137113778, "3000–6000 Hz_40ms_envelope_correlation": 0.978199766336781, "3000–6000 Hz_40ms_modulation_mae_db": 1.7239895189654575, "mr_stft_db": 5.782054722976323, "band_mae_db": 0.21527588073476242, "rms_gain_error_db": 0.021921956564892472, "output_peak": 0.6532410613403411}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0353187654691103, "全帯域_10ms_rms_envelope_nrmse": 0.09100607433256602, "全帯域_10ms_envelope_correlation": 0.9914335129502833, "全帯域_10ms_modulation_mae_db": 0.7255140584210578, "全帯域_40ms_log_rms_mae_db": 0.5703914645364041, "全帯域_40ms_rms_envelope_nrmse": 0.049969462567997516, "全帯域_40ms_envelope_correlation": 0.9973608764986374, "全帯域_40ms_modulation_mae_db": 1.2207452250702773, "300–1000 Hz_10ms_log_rms_mae_db": 1.7586540178131749, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15224058426991566, "300–1000 Hz_10ms_envelope_correlation": 0.9806976880444553, "300–1000 Hz_10ms_modulation_mae_db": 0.4350397639125398, "300–1000 Hz_40ms_log_rms_mae_db": 1.0335065042405138, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0820899015438031, "300–1000 Hz_40ms_envelope_correlation": 0.9944006774899586, "300–1000 Hz_40ms_modulation_mae_db": 1.010280492557305, "1000–3000 Hz_10ms_log_rms_mae_db": 2.000003867713502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18115352247247654, "1000–3000 Hz_10ms_envelope_correlation": 0.9737938869241706, "1000–3000 Hz_10ms_modulation_mae_db": 1.4056793776867682, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2144884576503403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10793760123771197, "1000–3000 Hz_40ms_envelope_correlation": 0.9908726754374354, "1000–3000 Hz_40ms_modulation_mae_db": 1.4010858626000768, "3000–6000 Hz_10ms_log_rms_mae_db": 2.334761767695059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2193517717830814, "3000–6000 Hz_10ms_envelope_correlation": 0.964936014559233, "3000–6000 Hz_10ms_modulation_mae_db": 0.6471381951631866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6847706567331369, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15808059317155265, "3000–6000 Hz_40ms_envelope_correlation": 0.9800606988859647, "3000–6000 Hz_40ms_modulation_mae_db": 1.0962951769407747, "mr_stft_db": 5.797377849319507, "band_mae_db": 0.2019883572665779, "rms_gain_error_db": -0.15132816522040837, "output_peak": 0.6316981087296154}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9813712046162574, "全帯域_10ms_rms_envelope_nrmse": 0.07484340855833053, "全帯域_10ms_envelope_correlation": 0.9941509497570372, "全帯域_10ms_modulation_mae_db": 0.21160665354393596, "全帯域_40ms_log_rms_mae_db": 0.5423906842284175, "全帯域_40ms_rms_envelope_nrmse": 0.04161028545846135, "全帯域_40ms_envelope_correlation": 0.9979944322215111, "全帯域_40ms_modulation_mae_db": 0.375624307923264, "300–1000 Hz_10ms_log_rms_mae_db": 1.6856695511470812, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12054717714256302, "300–1000 Hz_10ms_envelope_correlation": 0.9878088240953364, "300–1000 Hz_10ms_modulation_mae_db": 0.4419415895551782, "300–1000 Hz_40ms_log_rms_mae_db": 0.9726113642205695, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06692960035100057, "300–1000 Hz_40ms_envelope_correlation": 0.9958720362460383, "300–1000 Hz_40ms_modulation_mae_db": 0.5233432007393821, "1000–3000 Hz_10ms_log_rms_mae_db": 2.069601612724531, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19395241322071793, "1000–3000 Hz_10ms_envelope_correlation": 0.9693509051443666, "1000–3000 Hz_10ms_modulation_mae_db": 0.6494412562937445, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2219410237641557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10844554443161458, "1000–3000 Hz_40ms_envelope_correlation": 0.9892829819446421, "1000–3000 Hz_40ms_modulation_mae_db": 0.9313812006444814, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3323944059537816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2153770342025863, "3000–6000 Hz_10ms_envelope_correlation": 0.9671391363696858, "3000–6000 Hz_10ms_modulation_mae_db": 0.48018219533527234, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6650700694801173, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15599358950609382, "3000–6000 Hz_40ms_envelope_correlation": 0.9811200984582272, "3000–6000 Hz_40ms_modulation_mae_db": 0.5357709892241737, "mr_stft_db": 5.793475613248769, "band_mae_db": 0.10482209695737212, "rms_gain_error_db": 0.043696970384171814, "output_peak": 0.6617175561522753}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0068528661958558, "全帯域_10ms_rms_envelope_nrmse": 0.07978922723059943, "全帯域_10ms_envelope_correlation": 0.9932855398224684, "全帯域_10ms_modulation_mae_db": 0.4974025597620351, "全帯域_40ms_log_rms_mae_db": 0.5415132048629704, "全帯域_40ms_rms_envelope_nrmse": 0.041982326886413936, "全帯域_40ms_envelope_correlation": 0.9979449035813737, "全帯域_40ms_modulation_mae_db": 0.8028045046869084, "300–1000 Hz_10ms_log_rms_mae_db": 1.6879597961905903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13101885681435388, "300–1000 Hz_10ms_envelope_correlation": 0.9856337088821873, "300–1000 Hz_10ms_modulation_mae_db": 0.4182591913758314, "300–1000 Hz_40ms_log_rms_mae_db": 0.9818662456986793, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07200979352700779, "300–1000 Hz_40ms_envelope_correlation": 0.9954237830460164, "300–1000 Hz_40ms_modulation_mae_db": 0.6822272625283136, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0435473259856383, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1963146384068595, "1000–3000 Hz_10ms_envelope_correlation": 0.9683102169168286, "1000–3000 Hz_10ms_modulation_mae_db": 1.0288471330349285, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2266499488625826, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09967173160816074, "1000–3000 Hz_40ms_envelope_correlation": 0.9911939491396076, "1000–3000 Hz_40ms_modulation_mae_db": 1.0426729242863728, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3788994012135, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2301584975119725, "3000–6000 Hz_10ms_envelope_correlation": 0.9624225663405838, "3000–6000 Hz_10ms_modulation_mae_db": 0.3876765107990292, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7243932148161523, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1617786860835782, "3000–6000 Hz_40ms_envelope_correlation": 0.979671285329913, "3000–6000 Hz_40ms_modulation_mae_db": 0.8017831675306368, "mr_stft_db": 5.775596647001922, "band_mae_db": 0.17847089059740093, "rms_gain_error_db": -0.046971555479620106, "output_peak": 0.6548613605345697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9853883124483087, "全帯域_10ms_rms_envelope_nrmse": 0.07978081905684159, "全帯域_10ms_envelope_correlation": 0.9932957738937546, "全帯域_10ms_modulation_mae_db": 0.2605632278728086, "全帯域_40ms_log_rms_mae_db": 0.5228912960899336, "全帯域_40ms_rms_envelope_nrmse": 0.03616632853416432, "全帯域_40ms_envelope_correlation": 0.9984557883466773, "全帯域_40ms_modulation_mae_db": 0.31089306342121653, "300–1000 Hz_10ms_log_rms_mae_db": 1.6882307302788468, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12494058128989464, "300–1000 Hz_10ms_envelope_correlation": 0.9868967032919507, "300–1000 Hz_10ms_modulation_mae_db": 0.608203687792389, "300–1000 Hz_40ms_log_rms_mae_db": 0.944759872822912, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06591022754553857, "300–1000 Hz_40ms_envelope_correlation": 0.9959940132247852, "300–1000 Hz_40ms_modulation_mae_db": 0.8926425249689024, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9738431221223118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17757427526912758, "1000–3000 Hz_10ms_envelope_correlation": 0.9743277784535914, "1000–3000 Hz_10ms_modulation_mae_db": 1.0497561949960315, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1822079899351348, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10047345916835168, "1000–3000 Hz_40ms_envelope_correlation": 0.9914155139741869, "1000–3000 Hz_40ms_modulation_mae_db": 1.0259004630275739, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3270556484785354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22565372611246268, "3000–6000 Hz_10ms_envelope_correlation": 0.9651888341149018, "3000–6000 Hz_10ms_modulation_mae_db": 0.7131221674034488, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6914178715955634, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1611939191898806, "3000–6000 Hz_40ms_envelope_correlation": 0.9808891546078299, "3000–6000 Hz_40ms_modulation_mae_db": 0.9579759899936846, "mr_stft_db": 5.788866161248953, "band_mae_db": 0.12179842088187949, "rms_gain_error_db": 0.010000093451745717, "output_peak": 0.6726202654303536}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.985596682994136, "全帯域_10ms_rms_envelope_nrmse": 0.07136646009311447, "全帯域_10ms_envelope_correlation": 0.9946349595370995, "全帯域_10ms_modulation_mae_db": 0.3676430449633345, "全帯域_40ms_log_rms_mae_db": 0.5215480610191326, "全帯域_40ms_rms_envelope_nrmse": 0.03518668244084968, "全帯域_40ms_envelope_correlation": 0.9985609330996578, "全帯域_40ms_modulation_mae_db": 0.5946539132034155, "300–1000 Hz_10ms_log_rms_mae_db": 1.6706284352308491, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11687008229053258, "300–1000 Hz_10ms_envelope_correlation": 0.9886073251951105, "300–1000 Hz_10ms_modulation_mae_db": 0.2674436368299497, "300–1000 Hz_40ms_log_rms_mae_db": 0.9772542749303696, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06449771070986893, "300–1000 Hz_40ms_envelope_correlation": 0.9963577698850931, "300–1000 Hz_40ms_modulation_mae_db": 0.46767238775592146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.99353733490078, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1884565541264197, "1000–3000 Hz_10ms_envelope_correlation": 0.9712058719911085, "1000–3000 Hz_10ms_modulation_mae_db": 1.3575789439897328, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2238867416388852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11259914375323682, "1000–3000 Hz_40ms_envelope_correlation": 0.9895051140732342, "1000–3000 Hz_40ms_modulation_mae_db": 1.214127637293867, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2958865393992047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2447383230000683, "3000–6000 Hz_10ms_envelope_correlation": 0.9579083207609662, "3000–6000 Hz_10ms_modulation_mae_db": 0.5037957741235628, "3000–6000 Hz_40ms_log_rms_mae_db": 1.68856053630712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.170503812605429, "3000–6000 Hz_40ms_envelope_correlation": 0.9777058403810768, "3000–6000 Hz_40ms_modulation_mae_db": 0.5822018857432466, "mr_stft_db": 5.78600870810153, "band_mae_db": 0.1045419667456587, "rms_gain_error_db": -0.0387785681399309, "output_peak": 0.6430804711665424}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.28161329583759, "spectral_convergence": 0.2343311827978693, "spectral_convergence_status": "ok", "amplitude_floor": 1.9341972091070217e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.530826724011775, "spectral_convergence": 0.21959151381521022, "spectral_convergence_status": "ok", "amplitude_floor": 1.0469896576426271e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.523114586558138, "spectral_convergence": 0.2524080718104137, "spectral_convergence_status": "ok", "amplitude_floor": 8.978224937278799e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002340108897553237, "output_power": 0.002337611744201151, "delta_db": -0.004636873842438033, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003418531859998669, "output_power": 0.0032319305188865197, "delta_db": -0.24377616192045803, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009669299209010629, "output_power": 0.0009679879324656929, "delta_db": 0.0047494381149381796, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016643839195516676, "output_power": 0.00015877357706651292, "delta_db": -0.20475281935655984, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.655994855693747e-05, "output_power": 7.129759715563416e-05, "delta_db": -0.30926739280745236, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.076442519269658e-05, "output_power": 3.905555665595148e-05, "delta_db": -0.18598490690220848, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15886126549067583, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1274285893127981, "pitch": null, "output_rms_dbfs": -21.67066138437933, "output_peak": 0.6267310956187829, "reference_rms_dbfs": -21.54323279506653, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.778518202135834}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019906122792766575, "log_rms_mae_db": 1.0419614710725116, "rms_envelope_nrmse": 0.0793860472181451, "envelope_correlation": 0.9934868750542311, "rms_gain_error_db": -0.12710912974849015, "reference_mean_power": 0.007012794178240996, "output_mean_power": 0.00681051854420442}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.192818590388021e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015356197181510699, "output_power": 0.001485891650199444, "reference_active": true, "delta_db": -0.14296537806728796}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003622112164404606, "output_power": 0.00032415279111815837, "reference_active": true, "delta_db": -0.4821212935011587}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.688762087391039e-05, "output_power": 3.9985070490750096e-05, "reference_active": true, "delta_db": -0.6916033076607385}], "active_band_count": 3, "active_band_mae_db": 0.4388966597430617}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018229956701751592, "log_rms_mae_db": 0.5561920000521707, "rms_envelope_nrmse": 0.042104614046314895, "envelope_correlation": 0.9981288309209572, "rms_gain_error_db": -0.12802122541890387, "reference_mean_power": 0.0070179714231847195, "output_mean_power": 0.006814115222026544}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8443307374696506e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014276197130393435, "output_power": 0.001381423805225113, "reference_active": true, "delta_db": -0.14285600541321045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018503527061359352, "output_power": 0.0001657995666572316, "reference_active": true, "delta_db": -0.47671128487585285}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.327921114523828e-06, "output_power": 3.4872095734672223e-06, "reference_active": true, "delta_db": -0.9380128795543337}], "active_band_count": 3, "active_band_mae_db": 0.519193389947799}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015227518148493586, "log_rms_mae_db": 1.6418734340013657, "rms_envelope_nrmse": 0.11549813860882774, "envelope_correlation": 0.9889363122452952, "rms_gain_error_db": -0.1645791326586224, "reference_mean_power": 0.0024461505543329222, "output_mean_power": 0.0023551863194143387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3707367109300066e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00065188242251555, "output_power": 0.0006280214080062709, "reference_active": true, "delta_db": -0.1619482256869359}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000159373128948768, "output_power": 0.00013064367093645952, "reference_active": true, "delta_db": -0.8632672408753173}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8816561344170712e-05, "output_power": 1.7635814008519627e-05, "reference_active": true, "delta_db": -0.2814475065301362}], "active_band_count": 3, "active_band_mae_db": 0.4355543243641298}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013579343349222632, "log_rms_mae_db": 0.9558321603191158, "rms_envelope_nrmse": 0.059729388320832716, "envelope_correlation": 0.9970250485503686, "rms_gain_error_db": -0.17093201124167529, "reference_mean_power": 0.0024424650949128645, "output_mean_power": 0.0023482004386372454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.250982515304928e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006212562593255464, "output_power": 0.0005960545324415949, "reference_active": true, "delta_db": -0.1798478266633885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.001783099426803e-05, "output_power": 6.373562080811048e-05, "reference_active": true, "delta_db": -0.9880455463425932}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.744470580402989e-06, "output_power": 1.40008397384033e-06, "reference_active": true, "delta_db": -0.9550956524240268}], "active_band_count": 3, "active_band_mae_db": 0.7076630084766696}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.560305347439803e-05, "log_rms_mae_db": 1.980592941064255, "rms_envelope_nrmse": 0.1854593943941509, "envelope_correlation": 0.9719127935458339, "rms_gain_error_db": -0.2690732530413124, "reference_mean_power": 0.0001982579107701744, "output_mean_power": 0.00018634734092710727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.44612649270932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.480334346914858e-05, "output_power": 3.140788975665985e-05, "reference_active": true, "delta_db": -0.4458220966058881}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0911626587448635e-05, "output_power": 8.674241644610395e-06, "reference_active": true, "delta_db": -0.9965797890915764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7887330599139967e-06, "output_power": 1.7772450323795828e-06, "reference_active": true, "delta_db": -1.9565963619444018}], "active_band_count": 3, "active_band_mae_db": 1.132999415880622}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.513956828007693e-05, "log_rms_mae_db": 1.196743481232035, "rms_envelope_nrmse": 0.10075117636730142, "envelope_correlation": 0.9913582083820092, "rms_gain_error_db": -0.2725796108063719, "reference_mean_power": 0.00019869265760544437, "output_mean_power": 0.00018660524977566581}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.64113687643322e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.340875190885911e-05, "output_power": 2.9713179174679608e-05, "reference_active": true, "delta_db": -0.5091112965817003}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.757829899772919e-06, "output_power": 4.417292657703926e-06, "reference_active": true, "delta_db": -1.1510265721450779}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2467549734271334e-07, "output_power": 1.496584185144267e-07, "reference_active": true, "delta_db": -1.7645456012930292}], "active_band_count": 3, "active_band_mae_db": 1.141561156673269}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.7519381942341847e-05, "log_rms_mae_db": 2.3549512537716444, "rms_envelope_nrmse": 0.22689790708014737, "envelope_correlation": 0.9625876625660849, "rms_gain_error_db": -0.03396204144152415, "reference_mean_power": 3.309162956541551e-05, "output_mean_power": 3.283386059616695e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.646064632792655e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.7250471980521725e-06, "output_power": 6.260731348290195e-06, "reference_active": true, "delta_db": -0.3107026832157794}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3903630336351795e-06, "output_power": 2.2860268543935574e-06, "reference_active": true, "delta_db": -0.1938253610263695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2031554280127943e-06, "output_power": 7.034057608434683e-07, "reference_active": true, "delta_db": -2.3311581387485942}], "active_band_count": 3, "active_band_mae_db": 0.9452287276635811}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2421508904297867e-05, "log_rms_mae_db": 1.6835884419453562, "rms_envelope_nrmse": 0.1581974032331479, "envelope_correlation": 0.9800433544514824, "rms_gain_error_db": -0.03565022914864858, "reference_mean_power": 3.3247421347363265e-05, "output_mean_power": 3.297561806968271e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.522637378578784e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.9685546500066154e-06, "output_power": 6.419162807419978e-06, "reference_active": true, "delta_db": -0.35664319577769044}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.482924821402799e-06, "output_power": 1.271106973426427e-06, "reference_active": true, "delta_db": -0.6693703321062637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.259963132949643e-08, "output_power": 5.340321112113333e-08, "reference_active": true, "delta_db": -2.3904188578447054}], "active_band_count": 3, "active_band_mae_db": 1.1388107952428865}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8968717249052971, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2a5ffba116ae8ef936d38a5773eb2efd3a1f28b0029675aea26a0b0b1f0f7c9f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9718508426760641, "standard_deviation": 0.0022611364706712246, "minimum": 0.9683102169168286, "maximum": 0.9744095344634893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0091549723052022, "standard_deviation": 0.03453481796098796, "minimum": 1.9738431221223118, "maximum": 2.069601612724531, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0255137453554486, "standard_deviation": 0.2744469380399026, "minimum": 0.6494412562937445, "maximum": 1.4056793776867682, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18596850325018363, "standard_deviation": 0.007167085784819295, "minimum": 0.176784593263863, "maximum": 0.1963146384068595, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9909456627952391, "standard_deviation": 0.0011663918461058424, "minimum": 0.9892829819446421, "maximum": 0.9930012824972978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2079972672598371, "standard_deviation": 0.017485799273778874, "minimum": 1.1822079899351348, "maximum": 1.2266499488625826, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0345640496647115, "standard_deviation": 0.22556204242423508, "minimum": 0.6692259942983765, "maximum": 1.4010858626000768, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10339827026769369, "standard_deviation": 0.005578521402758471, "minimum": 0.09614047568928348, "maximum": 0.11259914375323682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9628639296483277, "standard_deviation": 0.002940069700422907, "minimum": 0.9579083207609662, "maximum": 0.9671391363696858, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3399370559330115, "standard_deviation": 0.02411498323326598, "minimum": 2.2958865393992047, "maximum": 2.3788994012135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6976001356160119, "standard_deviation": 0.3559217063249425, "minimum": 0.3876765107990292, "maximum": 1.45804355260524, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2279216315449977, "standard_deviation": 0.008883754068583336, "minimum": 0.2153770342025863, "maximum": 0.2447383230000683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9797432471885632, "standard_deviation": 0.0012071744013105144, "minimum": 0.9777058403810768, "maximum": 0.9811200984582272, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6879818215075404, "standard_deviation": 0.016793066540625665, "minimum": 1.6650700694801173, "maximum": 1.7243932148161523, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9359933752334524, "standard_deviation": 0.3914742510642442, "minimum": 0.5357709892241737, "maximum": 1.7239895189654575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16113282506032273, "standard_deviation": 0.004750363817075621, "minimum": 0.15599358950609382, "maximum": 0.170503812605429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9869363243541585, "standard_deviation": 0.0027461448324875687, "minimum": 0.9806976880444553, "maximum": 0.9889363122452952, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6924977493180828, "standard_deviation": 0.03379828446706638, "minimum": 1.6418734340013657, "maximum": 1.7586540178131749, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5195033111495249, "standard_deviation": 0.18432821706910504, "minimum": 0.2674436368299497, "maximum": 0.8279142206073856, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12448212651467913, "standard_deviation": 0.01235674583694754, "minimum": 0.11549813860882774, "maximum": 0.15224058426991566, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959500108829873, "standard_deviation": 0.0007880517186190073, "minimum": 0.9944006774899586, "maximum": 0.9970250485503686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.986085017819921, "standard_deviation": 0.03307702110582141, "minimum": 0.944759872822912, "maximum": 1.0361765296709857, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7847247584285897, "standard_deviation": 0.22363406385975473, "minimum": 0.46767238775592146, "maximum": 1.0749861242295176, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06744607449389309, "standard_deviation": 0.006943554736103159, "minimum": 0.059729388320832716, "maximum": 0.0820899015438031, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14790397867508004, "standard_deviation": 0.04702521021774744, "minimum": 0.09747295072631301, "maximum": 0.21527588073476242, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.786865085998851, "standard_deviation": 0.007715105523137897, "minimum": 5.775596647001922, "maximum": 5.797377849319507, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6473707923121752, "standard_deviation": 0.015931109139757184, "minimum": 0.6267310956187829, "maximum": 0.6726202654303536, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.042922941559154104, "standard_deviation": 0.06925017779288038, "minimum": -0.15132816522040837, "maximum": 0.043696970384171814, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993570775591453, "standard_deviation": 0.0010469325138169805, "minimum": 0.9914335129502833, "maximum": 0.9947815721692077, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0080667291143661, "standard_deviation": 0.023571080648450354, "minimum": 0.9813712046162574, "maximum": 1.0419614710725116, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41527261779592317, "standard_deviation": 0.23464245172167253, "minimum": 0.08308240781515487, "maximum": 0.7374723302459968, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07816032844232472, "standard_deviation": 0.0064241703858037, "minimum": 0.07043560187131451, "maximum": 0.09100607433256602, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980467514350118, "standard_deviation": 0.0004959576643102133, "minimum": 0.9973371475225606, "maximum": 0.998591099288719, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5431587350608006, "standard_deviation": 0.017598586983258297, "minimum": 0.5215480610191326, "maximum": 0.5703914645364041, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5987102992954972, "standard_deviation": 0.3224256220049794, "minimum": 0.224669840662199, "maximum": 1.2207452250702773, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041187889936327805, "standard_deviation": 0.005609442636537091, "minimum": 0.03490694007623789, "maximum": 0.049969462567997516, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7176-88083-0004", "language": "en", "style": "normal", "speaker": "librispeech_7176", "source_utterance_id": "7176-88083-0004", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7176-88083-0004.wav", "source_file_sha256": "07b85712a5f328803081a96f277c0f1be4d86857192a5cbd233009b1e1993069", "decoded_pcm_sha256": "43bbb6d389351da5d47cefcacfede3e848cb81497c672b951f8b533d7fbedfed", "original_source_sha256": "a0879f1d42b65a0bc10e3b725393bac049736d153e7a1c1b5d35e48bda4e3778", "corpus_entry_id": "librispeech_7176-88083-0004"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.5081281661987305, "predicted_mos": 4.5081281661987305, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "44210103bf3e3cc157184e7ee954d25869da0ee1188ff6ed4d1c8f815fd7acf6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.3673611320009513, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7176-88083-0004", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "44210103bf3e3cc157184e7ee954d25869da0ee1188ff6ed4d1c8f815fd7acf6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "45dd31e9af00a32cf192a87f3fc4d575cf62b88c59a697486955eb85310466db"}, "rows": [{"case": "librispeech_7176-88083-0004", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.031992197036743, "predicted_mos": 3.031992197036743, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f042e8256103dbdda2d74f293eeee5aef04a539062c37ef2fa4bd5cb89ebf8c0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.159854470999562, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7176-88083-0004", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f042e8256103dbdda2d74f293eeee5aef04a539062c37ef2fa4bd5cb89ebf8c0", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9783308235871506, "全帯域_10ms_rms_envelope_nrmse": 0.06502984414144199, "全帯域_10ms_envelope_correlation": 0.9953061134361834, "全帯域_10ms_modulation_mae_db": 0.14322919273413445, "全帯域_40ms_log_rms_mae_db": 0.5011227107009143, "全帯域_40ms_rms_envelope_nrmse": 0.03404353890137764, "全帯域_40ms_envelope_correlation": 0.9985761674364564, "全帯域_40ms_modulation_mae_db": 0.16849876255412954, "300–1000 Hz_10ms_log_rms_mae_db": 1.6270299724063833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10941739343932572, "300–1000 Hz_10ms_envelope_correlation": 0.9896489916749044, "300–1000 Hz_10ms_modulation_mae_db": 0.4742084335531381, "300–1000 Hz_40ms_log_rms_mae_db": 0.9128009875108323, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062008437235577744, "300–1000 Hz_40ms_envelope_correlation": 0.9965464602596453, "300–1000 Hz_40ms_modulation_mae_db": 0.31036018705972296, "1000–3000 Hz_10ms_log_rms_mae_db": 2.016831169465673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18609395139636892, "1000–3000 Hz_10ms_envelope_correlation": 0.971827017451505, "1000–3000 Hz_10ms_modulation_mae_db": 0.8536250571198121, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2399527674930344, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11231236494867138, "1000–3000 Hz_40ms_envelope_correlation": 0.9896192257114809, "1000–3000 Hz_40ms_modulation_mae_db": 1.3355996532152952, "3000–6000 Hz_10ms_log_rms_mae_db": 2.199959904266082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19051529505439743, "3000–6000 Hz_10ms_envelope_correlation": 0.975064844364947, "3000–6000 Hz_10ms_modulation_mae_db": 0.9230067338343435, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3764802374821827, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12178361582632534, "3000–6000 Hz_40ms_envelope_correlation": 0.9890250720569139, "3000–6000 Hz_40ms_modulation_mae_db": 1.0660741365731383, "mr_stft_db": 5.774167438718635, "band_mae_db": 0.18994829501954735, "rms_gain_error_db": -0.04636331636925427, "output_peak": 0.588625573448276}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0001859419369175, "全帯域_10ms_rms_envelope_nrmse": 0.06862237586657748, "全帯域_10ms_envelope_correlation": 0.9947505711319223, "全帯域_10ms_modulation_mae_db": 0.29383951625556093, "全帯域_40ms_log_rms_mae_db": 0.5083189419949408, "全帯域_40ms_rms_envelope_nrmse": 0.03594408258346302, "全帯域_40ms_envelope_correlation": 0.998393213981814, "全帯域_40ms_modulation_mae_db": 0.520859549102901, "300–1000 Hz_10ms_log_rms_mae_db": 1.6152414129680521, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11427616993694448, "300–1000 Hz_10ms_envelope_correlation": 0.9887649795429075, "300–1000 Hz_10ms_modulation_mae_db": 0.9695300422672428, "300–1000 Hz_40ms_log_rms_mae_db": 1.0137569132121, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07104256484556383, "300–1000 Hz_40ms_envelope_correlation": 0.9954715039483458, "300–1000 Hz_40ms_modulation_mae_db": 1.203511075129329, "1000–3000 Hz_10ms_log_rms_mae_db": 1.984181156462679, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20007874984193866, "1000–3000 Hz_10ms_envelope_correlation": 0.9675779257176064, "1000–3000 Hz_10ms_modulation_mae_db": 1.0316288401159053, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2258409283548075, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1151503476860594, "1000–3000 Hz_40ms_envelope_correlation": 0.9897581652134197, "1000–3000 Hz_40ms_modulation_mae_db": 0.8524919516301873, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1604472914643256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19828902165641465, "3000–6000 Hz_10ms_envelope_correlation": 0.9729612545318845, "3000–6000 Hz_10ms_modulation_mae_db": 0.6814964499077888, "3000–6000 Hz_40ms_log_rms_mae_db": 1.366633808488942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12112359450744918, "3000–6000 Hz_40ms_envelope_correlation": 0.9891373170330425, "3000–6000 Hz_40ms_modulation_mae_db": 0.6859300503832454, "mr_stft_db": 5.758848924794468, "band_mae_db": 0.18300593073371438, "rms_gain_error_db": -0.04346828374689085, "output_peak": 0.6228387027719172}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0260640288028071, "全帯域_10ms_rms_envelope_nrmse": 0.06984837407392223, "全帯域_10ms_envelope_correlation": 0.9946047248113592, "全帯域_10ms_modulation_mae_db": 0.17978883429138617, "全帯域_40ms_log_rms_mae_db": 0.5129365351001662, "全帯域_40ms_rms_envelope_nrmse": 0.036584965885092254, "全帯域_40ms_envelope_correlation": 0.9984021155268069, "全帯域_40ms_modulation_mae_db": 0.34462952492862603, "300–1000 Hz_10ms_log_rms_mae_db": 1.6715516585178973, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1163957765684877, "300–1000 Hz_10ms_envelope_correlation": 0.9882530206986159, "300–1000 Hz_10ms_modulation_mae_db": 0.5802751279131861, "300–1000 Hz_40ms_log_rms_mae_db": 0.9743327719698005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06748650421638319, "300–1000 Hz_40ms_envelope_correlation": 0.9958709045760037, "300–1000 Hz_40ms_modulation_mae_db": 0.7548576691915064, "1000–3000 Hz_10ms_log_rms_mae_db": 2.012069286255003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21124586289070546, "1000–3000 Hz_10ms_envelope_correlation": 0.9631555745374927, "1000–3000 Hz_10ms_modulation_mae_db": 0.747800155670566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2309310002583553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14314875498247245, "1000–3000 Hz_40ms_envelope_correlation": 0.9813946835079063, "1000–3000 Hz_40ms_modulation_mae_db": 0.30318279301806356, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2527905346399084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21258602904658935, "3000–6000 Hz_10ms_envelope_correlation": 0.9692022533321138, "3000–6000 Hz_10ms_modulation_mae_db": 0.45879825753634024, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4420246755863755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1337314834967558, "3000–6000 Hz_40ms_envelope_correlation": 0.9867441415868329, "3000–6000 Hz_40ms_modulation_mae_db": 0.23864816730443328, "mr_stft_db": 5.7812638708403155, "band_mae_db": 0.13701268946689996, "rms_gain_error_db": -0.07858159878740563, "output_peak": 0.5906156292984599}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0286574044093253, "全帯域_10ms_rms_envelope_nrmse": 0.06615823810030698, "全帯域_10ms_envelope_correlation": 0.9951270583273837, "全帯域_10ms_modulation_mae_db": 0.10912086082182436, "全帯域_40ms_log_rms_mae_db": 0.48247066654315424, "全帯域_40ms_rms_envelope_nrmse": 0.033115686937177924, "全帯域_40ms_envelope_correlation": 0.9986484607032884, "全帯域_40ms_modulation_mae_db": 0.39410759718755733, "300–1000 Hz_10ms_log_rms_mae_db": 1.6415633310596844, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1189134754291228, "300–1000 Hz_10ms_envelope_correlation": 0.9876378400856505, "300–1000 Hz_10ms_modulation_mae_db": 0.1114383529031448, "300–1000 Hz_40ms_log_rms_mae_db": 1.002092708675228, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07656315687179376, "300–1000 Hz_40ms_envelope_correlation": 0.9944427227962106, "300–1000 Hz_40ms_modulation_mae_db": 0.14515389158148223, "1000–3000 Hz_10ms_log_rms_mae_db": 2.007440926634883, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21363050755326482, "1000–3000 Hz_10ms_envelope_correlation": 0.9620510367072391, "1000–3000 Hz_10ms_modulation_mae_db": 0.47801453699279706, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2985169155829532, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13025019800999055, "1000–3000 Hz_40ms_envelope_correlation": 0.9847258771835594, "1000–3000 Hz_40ms_modulation_mae_db": 0.4287295588426893, "3000–6000 Hz_10ms_log_rms_mae_db": 2.235365546669539, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20655131526435905, "3000–6000 Hz_10ms_envelope_correlation": 0.9708017406257319, "3000–6000 Hz_10ms_modulation_mae_db": 0.5863800519941104, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4163428148010238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12364914357862061, "3000–6000 Hz_40ms_envelope_correlation": 0.9886392268854076, "3000–6000 Hz_40ms_modulation_mae_db": 0.6729799407819588, "mr_stft_db": 5.767571562153587, "band_mae_db": 0.18852172458516483, "rms_gain_error_db": -0.048361877497486015, "output_peak": 0.6263942612092884}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0097628650489006, "全帯域_10ms_rms_envelope_nrmse": 0.07096170517896422, "全帯域_10ms_envelope_correlation": 0.9944596459046138, "全帯域_10ms_modulation_mae_db": 0.23783506093086904, "全帯域_40ms_log_rms_mae_db": 0.5181553886264678, "全帯域_40ms_rms_envelope_nrmse": 0.03856041677848678, "全帯域_40ms_envelope_correlation": 0.9982477513274807, "全帯域_40ms_modulation_mae_db": 0.5199391732349004, "300–1000 Hz_10ms_log_rms_mae_db": 1.698474795787088, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1206081709430958, "300–1000 Hz_10ms_envelope_correlation": 0.9873972776061758, "300–1000 Hz_10ms_modulation_mae_db": 0.6002545042743632, "300–1000 Hz_40ms_log_rms_mae_db": 1.005239131731457, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07479721055718402, "300–1000 Hz_40ms_envelope_correlation": 0.9948661219287837, "300–1000 Hz_40ms_modulation_mae_db": 0.6338817661612489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9867268443272235, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17837256054600378, "1000–3000 Hz_10ms_envelope_correlation": 0.974119106273275, "1000–3000 Hz_10ms_modulation_mae_db": 0.6639107552084219, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2259784533717897, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10279809514880572, "1000–3000 Hz_40ms_envelope_correlation": 0.9913738740735824, "1000–3000 Hz_40ms_modulation_mae_db": 1.0364884830479015, "3000–6000 Hz_10ms_log_rms_mae_db": 2.170713903191145, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18362043830214617, "3000–6000 Hz_10ms_envelope_correlation": 0.9768597399850926, "3000–6000 Hz_10ms_modulation_mae_db": 0.5250508550661573, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3972112895911266, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11761184259274889, "3000–6000 Hz_40ms_envelope_correlation": 0.9897074835438211, "3000–6000 Hz_40ms_modulation_mae_db": 0.4079827075391533, "mr_stft_db": 5.769118657208252, "band_mae_db": 0.20925599666770775, "rms_gain_error_db": -0.0893663383009603, "output_peak": 0.5798614843308623}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0202961484059043, "全帯域_10ms_rms_envelope_nrmse": 0.06714523262720061, "全帯域_10ms_envelope_correlation": 0.9949767278647867, "全帯域_10ms_modulation_mae_db": 0.15643533786458982, "全帯域_40ms_log_rms_mae_db": 0.5066417784119547, "全帯域_40ms_rms_envelope_nrmse": 0.034131992867606625, "全帯域_40ms_envelope_correlation": 0.9985544556353343, "全帯域_40ms_modulation_mae_db": 0.21866292804130474, "300–1000 Hz_10ms_log_rms_mae_db": 1.6784001466737437, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12519850079288322, "300–1000 Hz_10ms_envelope_correlation": 0.986266216034277, "300–1000 Hz_10ms_modulation_mae_db": 0.4249856444970937, "300–1000 Hz_40ms_log_rms_mae_db": 1.0152596300569727, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0717819282332044, "300–1000 Hz_40ms_envelope_correlation": 0.995171489390362, "300–1000 Hz_40ms_modulation_mae_db": 0.26528615956534224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9177945169877246, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16862441166100595, "1000–3000 Hz_10ms_envelope_correlation": 0.9766811874068586, "1000–3000 Hz_10ms_modulation_mae_db": 0.33138106782240295, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1583109528882232, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09087004964387904, "1000–3000 Hz_40ms_envelope_correlation": 0.9930103069085014, "1000–3000 Hz_40ms_modulation_mae_db": 0.11013734470164947, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2580005902897358, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21864523450734988, "3000–6000 Hz_10ms_envelope_correlation": 0.9671677232658356, "3000–6000 Hz_10ms_modulation_mae_db": 0.36707845534550526, "3000–6000 Hz_40ms_log_rms_mae_db": 1.438270698316383, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14076110032568687, "3000–6000 Hz_40ms_envelope_correlation": 0.9852224193523779, "3000–6000 Hz_40ms_modulation_mae_db": 0.21498214459146955, "mr_stft_db": 5.75933573450537, "band_mae_db": 0.15737949621864109, "rms_gain_error_db": -0.04146967427253889, "output_peak": 0.6182350949540695}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0218372175604096, "全帯域_10ms_rms_envelope_nrmse": 0.07099592787225006, "全帯域_10ms_envelope_correlation": 0.9944013409650517, "全帯域_10ms_modulation_mae_db": 0.20796023053650092, "全帯域_40ms_log_rms_mae_db": 0.5276327755711798, "全帯域_40ms_rms_envelope_nrmse": 0.03664722684360775, "全帯域_40ms_envelope_correlation": 0.9983498987969806, "全帯域_40ms_modulation_mae_db": 0.4116794553864263, "300–1000 Hz_10ms_log_rms_mae_db": 1.7416917160106735, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12208962221868738, "300–1000 Hz_10ms_envelope_correlation": 0.9870146728290874, "300–1000 Hz_10ms_modulation_mae_db": 0.3129882428450145, "300–1000 Hz_40ms_log_rms_mae_db": 1.0406056616471093, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07166071471661878, "300–1000 Hz_40ms_envelope_correlation": 0.9952294932987714, "300–1000 Hz_40ms_modulation_mae_db": 0.5951633147838676, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9566321517244603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17811035138248982, "1000–3000 Hz_10ms_envelope_correlation": 0.9742359006393231, "1000–3000 Hz_10ms_modulation_mae_db": 0.6331118317280378, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1973235540761242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10814155507082264, "1000–3000 Hz_40ms_envelope_correlation": 0.9903246983999257, "1000–3000 Hz_40ms_modulation_mae_db": 0.40482927561210064, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2534130912397545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22285826348127616, "3000–6000 Hz_10ms_envelope_correlation": 0.9657327348905488, "3000–6000 Hz_10ms_modulation_mae_db": 0.7887359631184396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4175522948198507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1384261293417016, "3000–6000 Hz_40ms_envelope_correlation": 0.9859322980925784, "3000–6000 Hz_40ms_modulation_mae_db": 0.7019011058052, "mr_stft_db": 5.771919621800218, "band_mae_db": 0.1812429063894108, "rms_gain_error_db": -0.054373678097377545, "output_peak": 0.604011379046174}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0599024824808285, "全帯域_10ms_rms_envelope_nrmse": 0.07091354825213361, "全帯域_10ms_envelope_correlation": 0.9943905334035613, "全帯域_10ms_modulation_mae_db": 0.16130740236750482, "全帯域_40ms_log_rms_mae_db": 0.5391000590783698, "全帯域_40ms_rms_envelope_nrmse": 0.035665951424451926, "全帯域_40ms_envelope_correlation": 0.9983979224094077, "全帯域_40ms_modulation_mae_db": 0.07014318026385548, "300–1000 Hz_10ms_log_rms_mae_db": 1.672388514825989, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11182119159006905, "300–1000 Hz_10ms_envelope_correlation": 0.9889991650122972, "300–1000 Hz_10ms_modulation_mae_db": 0.15785724125784673, "300–1000 Hz_40ms_log_rms_mae_db": 0.9522792028444352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06439877175282341, "300–1000 Hz_40ms_envelope_correlation": 0.9959839868496477, "300–1000 Hz_40ms_modulation_mae_db": 0.18780042221469406, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0113685021777696, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2014049228421212, "1000–3000 Hz_10ms_envelope_correlation": 0.9666502932653838, "1000–3000 Hz_10ms_modulation_mae_db": 0.6422315695125971, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2475522023181025, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12048041754291952, "1000–3000 Hz_40ms_envelope_correlation": 0.987708743821092, "1000–3000 Hz_40ms_modulation_mae_db": 0.6992523065248933, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2258144996908977, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21249951575678172, "3000–6000 Hz_10ms_envelope_correlation": 0.9695298542949742, "3000–6000 Hz_10ms_modulation_mae_db": 0.30396076902505514, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4026324958789584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13030081705940774, "3000–6000 Hz_40ms_envelope_correlation": 0.9876233840631359, "3000–6000 Hz_40ms_modulation_mae_db": 0.30721849901750814, "mr_stft_db": 5.755895948608942, "band_mae_db": 0.19220678321999096, "rms_gain_error_db": -0.007387873049336672, "output_peak": 0.5951096289156746}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.287141005375266, "spectral_convergence": 0.2227452528199766, "spectral_convergence_status": "ok", "amplitude_floor": 2.016382236479878e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5478315987032465, "spectral_convergence": 0.2124403608990162, "spectral_convergence_status": "ok", "amplitude_floor": 1.164576113083389e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.487529712077393, "spectral_convergence": 0.22347993081643275, "spectral_convergence_status": "ok", "amplitude_floor": 9.482830334434223e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0029188205763026483, "output_power": 0.0029807285902641797, "delta_db": 0.0911503406072773, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003712420548895462, "output_power": 0.003680639971970273, "delta_db": -0.037338296279713856, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009931195946472759, "output_power": 0.0009078893911539231, "delta_db": -0.389686094581452, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003469926396566426, "output_power": 0.0003152634032185979, "delta_db": -0.41646703329642815, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013263724937728663, "output_power": 0.0001306366625935787, "delta_db": -0.06600430084515352, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.438064081917431e-05, "output_power": 6.647520198584789e-05, "delta_db": 0.13904370450725945, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18994829501954735, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04636331636925427, "pitch": null, "output_rms_dbfs": -20.92500863572831, "output_peak": 0.588625573448276, "reference_rms_dbfs": -20.87864531935906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.774167438718635}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023582673937264833, "log_rms_mae_db": 0.9783308235871506, "rms_envelope_nrmse": 0.06502984414144199, "envelope_correlation": 0.9953061134361834, "rms_gain_error_db": -0.04636193847668112, "reference_mean_power": 0.008175179133340451, "output_mean_power": 0.008088371379709568}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4118743138857815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017924334098852407, "output_power": 0.0017357387641495664, "reference_active": true, "delta_db": -0.1395866765941599}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00033647515979494424, "output_power": 0.00032596555009050624, "reference_active": true, "delta_db": -0.13781304287636173}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.450579873055231e-05, "output_power": 5.6450981079906906e-05, "reference_active": true, "delta_db": 0.15228785873188172}], "active_band_count": 3, "active_band_mae_db": 0.14322919273413445}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002175646104262597, "log_rms_mae_db": 0.5011227107009143, "rms_envelope_nrmse": 0.03404353890137764, "envelope_correlation": 0.9985761674364564, "rms_gain_error_db": -0.04634859696890891, "reference_mean_power": 0.008216199965234223, "output_mean_power": 0.008128981605994814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.000498094679175e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016116003385924513, "output_power": 0.001575424941494624, "reference_active": true, "delta_db": -0.09859633196882671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017199808112094875, "output_power": 0.00016492228687447707, "reference_active": true, "delta_db": -0.18244253510728572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5420165509687798e-06, "output_power": 3.3636038751474566e-06, "reference_active": true, "delta_db": -0.22445742058627624}], "active_band_count": 3, "active_band_mae_db": 0.16849876255412954}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019245421130728438, "log_rms_mae_db": 1.6270299724063833, "rms_envelope_nrmse": 0.10941739343932572, "envelope_correlation": 0.9896489916749044, "rms_gain_error_db": -0.1678807732501522, "reference_mean_power": 0.0024199693169718968, "output_mean_power": 0.0023282080270196825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3235427499409267e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006913978889533925, "output_power": 0.0006250502090968322, "reference_active": true, "delta_db": -0.4381314437949765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018527392767596858, "output_power": 0.0001675460444925476, "reference_active": true, "delta_db": -0.4368012908710995}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604035138941515e-05, "output_power": 2.9045420691836734e-05, "reference_active": true, "delta_db": 0.5476925659933384}], "active_band_count": 3, "active_band_mae_db": 0.4742084335531381}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017723694424605393, "log_rms_mae_db": 0.9128009875108323, "rms_envelope_nrmse": 0.062008437235577744, "envelope_correlation": 0.9965464602596453, "rms_gain_error_db": -0.16788571810385267, "reference_mean_power": 0.0024321306730885443, "output_mean_power": 0.0023399055801399623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1677626980235007e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006540469587465989, "output_power": 0.000595105460913663, "reference_active": true, "delta_db": -0.410149950414928}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.291953991407944e-05, "output_power": 8.460457445282537e-05, "reference_active": true, "delta_db": -0.4071320537021608}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.049791938534597e-06, "output_power": 2.104212695348649e-06, "reference_active": true, "delta_db": 0.11379855706208006}], "active_band_count": 3, "active_band_mae_db": 0.31036018705972296}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.001235221787968e-05, "log_rms_mae_db": 2.016831169465673, "rms_envelope_nrmse": 0.18609395139636892, "envelope_correlation": 0.971827017451505, "rms_gain_error_db": -0.37103866510237704, "reference_mean_power": 0.0003849600012580844, "output_mean_power": 0.00035343677892640216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5531826233320727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012212996791699553, "output_power": 0.00011185039081810427, "reference_active": true, "delta_db": -0.3818473710176013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.773842605512037e-05, "output_power": 1.9665554239717633e-05, "reference_active": true, "delta_db": -1.4937562376603402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.9827801273791894e-06, "output_power": 5.9634998270634445e-06, "reference_active": true, "delta_db": -0.6852715626814945}], "active_band_count": 3, "active_band_mae_db": 0.8536250571198121}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.393121399816863e-05, "log_rms_mae_db": 1.2399527674930344, "rms_envelope_nrmse": 0.11231236494867138, "envelope_correlation": 0.9896192257114809, "rms_gain_error_db": -0.37104112834096953, "reference_mean_power": 0.0003868956457514561, "output_mean_power": 0.0003552137178147671}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2949024312838256e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011569002203634676, "output_power": 0.00010501299522968904, "reference_active": true, "delta_db": -0.4205285799047542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4800082503354946e-05, "output_power": 1.0140622643305405e-05, "reference_active": true, "delta_db": -1.641995144990909}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.803144704877755e-07, "output_power": 3.7088210279607163e-07, "reference_active": true, "delta_db": -1.9442752347502221}], "active_band_count": 3, "active_band_mae_db": 1.3355996532152952}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.314726805023413e-05, "log_rms_mae_db": 2.199959904266082, "rms_envelope_nrmse": 0.19051529505439743, "envelope_correlation": 0.975064844364947, "rms_gain_error_db": -0.13803235862124458, "reference_mean_power": 8.262304385157997e-05, "output_mean_power": 8.003831821928204e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.2225580529948354e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.126031645648209e-05, "output_power": 1.9985474907323803e-05, "reference_active": true, "delta_db": -0.26855251945583947}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.562486047803945e-06, "output_power": 8.902720969847038e-06, "reference_active": true, "delta_db": -0.3104805262298005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.527384735880018e-06, "output_power": 2.1303654770268615e-06, "reference_active": true, "delta_db": -2.1899871558173905}], "active_band_count": 3, "active_band_mae_db": 0.9230067338343435}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6400870566820594e-05, "log_rms_mae_db": 1.3764802374821827, "rms_envelope_nrmse": 0.12178361582632534, "envelope_correlation": 0.9890250720569139, "rms_gain_error_db": -0.13803478461100765, "reference_mean_power": 8.30385039555115e-05, "output_mean_power": 8.044073640461026e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.767621782464082e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.197465758003415e-05, "output_power": 2.0521894743233252e-05, "reference_active": true, "delta_db": -0.2970466062350518}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.865612603723312e-06, "output_power": 4.805583625097372e-06, "reference_active": true, "delta_db": -0.8656723737798099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.828054895066371e-07, "output_power": 1.7698542060262644e-07, "reference_active": true, "delta_db": -2.0355034297045536}], "active_band_count": 3, "active_band_mae_db": 1.0660741365731383}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9085402427246426, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "175d3e8ad94e36883237ff2b7f1c54487eec05efdf5f17c8e38701643ae5277e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9695372552498355, "standard_deviation": 0.005454909235840775, "minimum": 0.9620510367072391, "maximum": 0.9766811874068586, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.986630569254427, "standard_deviation": 0.03432732546124863, "minimum": 1.9177945169877246, "maximum": 2.016831169465673, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6727129767713175, "standard_deviation": 0.21520012702864416, "minimum": 0.33138106782240295, "maximum": 1.0316288401159053, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19219516476423731, "standard_deviation": 0.016697412711921916, "minimum": 0.16862441166100595, "maximum": 0.21363050755326482, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9884894468524335, "standard_deviation": 0.003781947941462813, "minimum": 0.9813946835079063, "maximum": 0.9930103069085014, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2280508467929239, "standard_deviation": 0.04020077929577735, "minimum": 1.1583109528882232, "maximum": 1.2985169155829532, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6463389208240975, "standard_deviation": 0.41119076873603894, "minimum": 0.11013734470164947, "maximum": 1.3355996532152952, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1153939728792026, "standard_deviation": 0.016209077903818512, "minimum": 0.09087004964387904, "maximum": 0.14314875498247245, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.970915018161391, "standard_deviation": 0.003826807908794256, "minimum": 0.9657327348905488, "maximum": 0.9768597399850926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2195631701814236, "standard_deviation": 0.038324424194599103, "minimum": 2.1604472914643256, "maximum": 2.2580005902897358, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5793134419784676, "standard_deviation": 0.21071487078418347, "minimum": 0.30396076902505514, "maximum": 0.9230067338343435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2056956391336643, "standard_deviation": 0.013780759917393595, "minimum": 0.18362043830214617, "maximum": 0.22285826348127616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9877539178267638, "standard_deviation": 0.0016428758617961618, "minimum": 0.9852224193523779, "maximum": 0.9897074835438211, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4071435393706055, "standard_deviation": 0.026936003585367546, "minimum": 1.366633808488942, "maximum": 1.4420246755863755, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5369645939995134, "standard_deviation": 0.2951844038859048, "minimum": 0.21498214459146955, "maximum": 1.0660741365731383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.128423465841087, "standard_deviation": 0.008627456841370895, "minimum": 0.11761184259274889, "maximum": 0.14076110032568687, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9879977704354894, "standard_deviation": 0.0011236412896186096, "minimum": 0.986266216034277, "maximum": 0.9896489916749044, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6682926935311888, "standard_deviation": 0.04081218717415375, "minimum": 1.6152414129680521, "maximum": 1.7416917160106735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.45394219868887875, "standard_deviation": 0.2749780590174799, "minimum": 0.1114383529031448, "maximum": 0.9695300422672428, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11734003761482702, "standard_deviation": 0.005358241525040633, "minimum": 0.10941739343932572, "maximum": 0.12519850079288322, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954478353809713, "standard_deviation": 0.0006704651888436366, "minimum": 0.9944427227962106, "maximum": 0.9965464602596453, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.989545875955992, "standard_deviation": 0.0410020832886786, "minimum": 0.9128009875108323, "maximum": 1.0406056616471093, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5120018107108992, "standard_deviation": 0.35859776173638924, "minimum": 0.14515389158148223, "maximum": 1.203511075129329, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06996741105364365, "standard_deviation": 0.004995089565087953, "minimum": 0.062008437235577744, "maximum": 0.07656315687179376, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17982172778763464, "standard_deviation": 0.02250850476568768, "minimum": 0.13701268946689996, "maximum": 0.20925599666770775, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.767265219828723, "standard_deviation": 0.008715884324172855, "minimum": 5.755895948608942, "maximum": 5.7812638708403155, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6032114692468402, "standard_deviation": 0.017449046133077774, "minimum": 0.5798614843308623, "maximum": 0.6263942612092884, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05117158001515627, "standard_deviation": 0.024847813049397302, "minimum": -0.0893663383009603, "maximum": -0.007387873049336672, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947520894806078, "standard_deviation": 0.0003503051317535775, "minimum": 0.9943905334035613, "maximum": 0.9953061134361834, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0181296140290303, "standard_deviation": 0.023643554298973312, "minimum": 0.9783308235871506, "maximum": 1.0599024824808285, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.18618955447529634, "standard_deviation": 0.05862753532184096, "minimum": 0.10912086082182436, "maximum": 0.29383951625556093, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06870940576409965, "standard_deviation": 0.0023591272875660824, "minimum": 0.06502984414144199, "maximum": 0.07099592787225006, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984462482271961, "standard_deviation": 0.00013376828184240286, "minimum": 0.9982477513274807, "maximum": 0.9986484607032884, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5120473570033934, "standard_deviation": 0.017113497276733433, "minimum": 0.48247066654315424, "maximum": 0.5391000590783698, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33106502133746263, "standard_deviation": 0.1645316657974624, "minimum": 0.07014318026385548, "maximum": 0.520859549102901, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03558673277765799, "standard_deviation": 0.0017607494794881966, "minimum": 0.033115686937177924, "maximum": 0.03856041677848678, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0004", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.856560230255127, "predicted_mos": 2.856560230255127, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8917ede9534ae553e9cd3f0da43c1dfa3be1080681efa880096b1349b4c1b3bc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.063583299983293, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9881335449907396, "全帯域_10ms_rms_envelope_nrmse": 0.06708179082557958, "全帯域_10ms_envelope_correlation": 0.9949904676424628, "全帯域_10ms_modulation_mae_db": 0.22848666066540554, "全帯域_40ms_log_rms_mae_db": 0.50846248719716, "全帯域_40ms_rms_envelope_nrmse": 0.03460214594022085, "全帯域_40ms_envelope_correlation": 0.9985156265422227, "全帯域_40ms_modulation_mae_db": 0.09946322298641996, "300–1000 Hz_10ms_log_rms_mae_db": 1.648409852025631, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11195673063138456, "300–1000 Hz_10ms_envelope_correlation": 0.9891150216454441, "300–1000 Hz_10ms_modulation_mae_db": 0.5767805535479098, "300–1000 Hz_40ms_log_rms_mae_db": 0.9058412378156193, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061827965533251546, "300–1000 Hz_40ms_envelope_correlation": 0.9965349586773836, "300–1000 Hz_40ms_modulation_mae_db": 0.38975988125079414, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0550235925936415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18082806731225398, "1000–3000 Hz_10ms_envelope_correlation": 0.973961587600433, "1000–3000 Hz_10ms_modulation_mae_db": 0.8455793949234831, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2443002014403959, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11252652726730535, "1000–3000 Hz_40ms_envelope_correlation": 0.9902736410340252, "1000–3000 Hz_40ms_modulation_mae_db": 1.4098856333950407, "3000–6000 Hz_10ms_log_rms_mae_db": 2.193174287418817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18753302870830274, "3000–6000 Hz_10ms_envelope_correlation": 0.9758555522420513, "3000–6000 Hz_10ms_modulation_mae_db": 0.6425707047895667, "3000–6000 Hz_40ms_log_rms_mae_db": 1.406423876959401, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11512460110034921, "3000–6000 Hz_40ms_envelope_correlation": 0.9902007281155333, "3000–6000 Hz_40ms_modulation_mae_db": 0.7507365358613397, "mr_stft_db": 5.728252568765913, "band_mae_db": 0.17350895355844517, "rms_gain_error_db": -0.03754684053277464, "output_peak": 0.5822143968231266}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9874284352016987, "全帯域_10ms_rms_envelope_nrmse": 0.0669619044383639, "全帯域_10ms_envelope_correlation": 0.9950023652098016, "全帯域_10ms_modulation_mae_db": 0.3288430496643273, "全帯域_40ms_log_rms_mae_db": 0.5003261049392421, "全帯域_40ms_rms_envelope_nrmse": 0.03511311927020469, "全帯域_40ms_envelope_correlation": 0.9984626927718708, "全帯域_40ms_modulation_mae_db": 0.5324626137027954, "300–1000 Hz_10ms_log_rms_mae_db": 1.6373860870265948, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11918536952312636, "300–1000 Hz_10ms_envelope_correlation": 0.9876701777907333, "300–1000 Hz_10ms_modulation_mae_db": 1.0370889691124099, "300–1000 Hz_40ms_log_rms_mae_db": 0.9902118999544716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07264118105902459, "300–1000 Hz_40ms_envelope_correlation": 0.9951501284681906, "300–1000 Hz_40ms_modulation_mae_db": 1.2684768787691894, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0351843273612986, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19914129798194533, "1000–3000 Hz_10ms_envelope_correlation": 0.968688040189415, "1000–3000 Hz_10ms_modulation_mae_db": 0.8422165369727962, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2555784624907846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12031024270775631, "1000–3000 Hz_40ms_envelope_correlation": 0.9897632234734859, "1000–3000 Hz_40ms_modulation_mae_db": 0.6620774162858416, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1863301387559515, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1999180724604805, "3000–6000 Hz_10ms_envelope_correlation": 0.9724766204757318, "3000–6000 Hz_10ms_modulation_mae_db": 0.7951734601198178, "3000–6000 Hz_40ms_log_rms_mae_db": 1.433138628404697, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12412054022699352, "3000–6000 Hz_40ms_envelope_correlation": 0.9887488378793107, "3000–6000 Hz_40ms_modulation_mae_db": 0.7839061965218219, "mr_stft_db": 5.723970753054668, "band_mae_db": 0.22157321816614794, "rms_gain_error_db": -0.03836815623641071, "output_peak": 0.6263504917562552}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.033014987535261, "全帯域_10ms_rms_envelope_nrmse": 0.06979296800438234, "全帯域_10ms_envelope_correlation": 0.9946386472731221, "全帯域_10ms_modulation_mae_db": 0.12723604636682215, "全帯域_40ms_log_rms_mae_db": 0.5103361350193404, "全帯域_40ms_rms_envelope_nrmse": 0.03680201627464905, "全帯域_40ms_envelope_correlation": 0.9984170209017876, "全帯域_40ms_modulation_mae_db": 0.27315376141999237, "300–1000 Hz_10ms_log_rms_mae_db": 1.6918511844772917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11758196160106271, "300–1000 Hz_10ms_envelope_correlation": 0.9880173760896248, "300–1000 Hz_10ms_modulation_mae_db": 0.6458251492688923, "300–1000 Hz_40ms_log_rms_mae_db": 0.9361434511511415, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06775606172891721, "300–1000 Hz_40ms_envelope_correlation": 0.9958455575445242, "300–1000 Hz_40ms_modulation_mae_db": 0.8721310923213451, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0333308921151736, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19056521437190801, "1000–3000 Hz_10ms_envelope_correlation": 0.9703624187048852, "1000–3000 Hz_10ms_modulation_mae_db": 0.5114856289054951, "1000–3000 Hz_40ms_log_rms_mae_db": 1.259302209650287, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12801314196142194, "1000–3000 Hz_40ms_envelope_correlation": 0.9858986078232279, "1000–3000 Hz_40ms_modulation_mae_db": 0.363258073579734, "3000–6000 Hz_10ms_log_rms_mae_db": 2.234845860538677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20444795183408648, "3000–6000 Hz_10ms_envelope_correlation": 0.9713154332257984, "3000–6000 Hz_10ms_modulation_mae_db": 0.48005564248434124, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4702417936222323, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.131728153178907, "3000–6000 Hz_40ms_envelope_correlation": 0.9871390042600932, "3000–6000 Hz_40ms_modulation_mae_db": 0.31093630036395664, "mr_stft_db": 5.731082532733255, "band_mae_db": 0.16708951932311913, "rms_gain_error_db": -0.09141496201232463, "output_peak": 0.555150124091766}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0121279468304785, "全帯域_10ms_rms_envelope_nrmse": 0.06437666938003157, "全帯域_10ms_envelope_correlation": 0.9954049072596197, "全帯域_10ms_modulation_mae_db": 0.14605540511322654, "全帯域_40ms_log_rms_mae_db": 0.48332973740154705, "全帯域_40ms_rms_envelope_nrmse": 0.031248577526800554, "全帯域_40ms_envelope_correlation": 0.9988275677702002, "全帯域_40ms_modulation_mae_db": 0.35064252066718965, "300–1000 Hz_10ms_log_rms_mae_db": 1.6301070605268424, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1170696080636868, "300–1000 Hz_10ms_envelope_correlation": 0.988037528987994, "300–1000 Hz_10ms_modulation_mae_db": 0.2260874141134531, "300–1000 Hz_40ms_log_rms_mae_db": 0.947156229699898, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07512082103436654, "300–1000 Hz_40ms_envelope_correlation": 0.99467438384762, "300–1000 Hz_40ms_modulation_mae_db": 0.22231434888919854, "1000–3000 Hz_10ms_log_rms_mae_db": 2.054063563626413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20742370461372692, "1000–3000 Hz_10ms_envelope_correlation": 0.9650046299178402, "1000–3000 Hz_10ms_modulation_mae_db": 0.5471310101912339, "1000–3000 Hz_40ms_log_rms_mae_db": 1.304764087932626, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12927141072889278, "1000–3000 Hz_40ms_envelope_correlation": 0.9864502612633267, "1000–3000 Hz_40ms_modulation_mae_db": 0.8599551663366419, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2509718715845928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19798711022094365, "3000–6000 Hz_10ms_envelope_correlation": 0.9729990719497342, "3000–6000 Hz_10ms_modulation_mae_db": 0.7377527983786828, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4624438273640674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11965067747398632, "3000–6000 Hz_40ms_envelope_correlation": 0.9894864267826595, "3000–6000 Hz_40ms_modulation_mae_db": 0.807605577550901, "mr_stft_db": 5.729288745079717, "band_mae_db": 0.19147950019218152, "rms_gain_error_db": -0.06402780499893507, "output_peak": 0.5971519214730726}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0294611258875175, "全帯域_10ms_rms_envelope_nrmse": 0.07219078523710452, "全帯域_10ms_envelope_correlation": 0.9942717489807478, "全帯域_10ms_modulation_mae_db": 0.24258637257911808, "全帯域_40ms_log_rms_mae_db": 0.5367563476731472, "全帯域_40ms_rms_envelope_nrmse": 0.04037769024102233, "全帯域_40ms_envelope_correlation": 0.9980787449497932, "全帯域_40ms_modulation_mae_db": 0.5091120866892856, "300–1000 Hz_10ms_log_rms_mae_db": 1.7135905134177554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12549631616933712, "300–1000 Hz_10ms_envelope_correlation": 0.9862693525929106, "300–1000 Hz_10ms_modulation_mae_db": 0.5745405158138751, "300–1000 Hz_40ms_log_rms_mae_db": 0.9996029076624571, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07784899283199348, "300–1000 Hz_40ms_envelope_correlation": 0.9943341169985646, "300–1000 Hz_40ms_modulation_mae_db": 0.6075257197822231, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0545630724850654, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1785327204116751, "1000–3000 Hz_10ms_envelope_correlation": 0.9746776181140995, "1000–3000 Hz_10ms_modulation_mae_db": 0.7402834286373811, "1000–3000 Hz_40ms_log_rms_mae_db": 1.226894103485758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10303452182223785, "1000–3000 Hz_40ms_envelope_correlation": 0.992254612259175, "1000–3000 Hz_40ms_modulation_mae_db": 1.1491173348264374, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2207417685379927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18962964369006513, "3000–6000 Hz_10ms_envelope_correlation": 0.9753641029490298, "3000–6000 Hz_10ms_modulation_mae_db": 0.5882706932612413, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4800687261467242, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12726118312214227, "3000–6000 Hz_40ms_envelope_correlation": 0.988020374168204, "3000–6000 Hz_40ms_modulation_mae_db": 0.4910623000114674, "mr_stft_db": 5.732940872234053, "band_mae_db": 0.22805771144185463, "rms_gain_error_db": -0.09269180782547562, "output_peak": 0.5808227596676391}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0307724954152087, "全帯域_10ms_rms_envelope_nrmse": 0.06966283375755779, "全帯域_10ms_envelope_correlation": 0.9945849067100676, "全帯域_10ms_modulation_mae_db": 0.1493236069921988, "全帯域_40ms_log_rms_mae_db": 0.5135040812540096, "全帯域_40ms_rms_envelope_nrmse": 0.034775116646291744, "全帯域_40ms_envelope_correlation": 0.998486088462336, "全帯域_40ms_modulation_mae_db": 0.23268651780245483, "300–1000 Hz_10ms_log_rms_mae_db": 1.6958188145708326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13084067060497112, "300–1000 Hz_10ms_envelope_correlation": 0.9849159996997362, "300–1000 Hz_10ms_modulation_mae_db": 0.4742390799846888, "300–1000 Hz_40ms_log_rms_mae_db": 0.9877015690787591, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07310020164464041, "300–1000 Hz_40ms_envelope_correlation": 0.9948735915156784, "300–1000 Hz_40ms_modulation_mae_db": 0.3090257434157459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9491780271900112, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16762219957224436, "1000–3000 Hz_10ms_envelope_correlation": 0.9771864985868696, "1000–3000 Hz_10ms_modulation_mae_db": 0.42863422259721634, "1000–3000 Hz_40ms_log_rms_mae_db": 1.171245758149151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08818210711936705, "1000–3000 Hz_40ms_envelope_correlation": 0.993837964368652, "1000–3000 Hz_40ms_modulation_mae_db": 0.22196160665433937, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2438613084836954, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20523589931286346, "3000–6000 Hz_10ms_envelope_correlation": 0.9709669610498703, "3000–6000 Hz_10ms_modulation_mae_db": 0.5590717631690268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.453519492855117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13475270856037347, "3000–6000 Hz_40ms_envelope_correlation": 0.9865780563124332, "3000–6000 Hz_40ms_modulation_mae_db": 0.4622777300907301, "mr_stft_db": 5.711669349966084, "band_mae_db": 0.17712465895075666, "rms_gain_error_db": -0.02832354056363333, "output_peak": 0.6173087351854879}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0081919632506018, "全帯域_10ms_rms_envelope_nrmse": 0.06779088724034009, "全帯域_10ms_envelope_correlation": 0.9949198038876693, "全帯域_10ms_modulation_mae_db": 0.19400499745731523, "全帯域_40ms_log_rms_mae_db": 0.5116448653029932, "全帯域_40ms_rms_envelope_nrmse": 0.03448526880391625, "全帯域_40ms_envelope_correlation": 0.9985714312503067, "全帯域_40ms_modulation_mae_db": 0.4704267055862276, "300–1000 Hz_10ms_log_rms_mae_db": 1.6768487236648508, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11589148866886247, "300–1000 Hz_10ms_envelope_correlation": 0.9883943783999427, "300–1000 Hz_10ms_modulation_mae_db": 0.5844069704935233, "300–1000 Hz_40ms_log_rms_mae_db": 0.9632866274538673, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06829332744463727, "300–1000 Hz_40ms_envelope_correlation": 0.9957655959949175, "300–1000 Hz_40ms_modulation_mae_db": 0.7794068786656257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.988154392072927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16927090341928855, "1000–3000 Hz_10ms_envelope_correlation": 0.9779914226439425, "1000–3000 Hz_10ms_modulation_mae_db": 0.38274113126852427, "1000–3000 Hz_40ms_log_rms_mae_db": 1.19537153767027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10545685581932628, "1000–3000 Hz_40ms_envelope_correlation": 0.9924719265297812, "1000–3000 Hz_40ms_modulation_mae_db": 0.5575575587818931, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2714503434745428, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22003908694736954, "3000–6000 Hz_10ms_envelope_correlation": 0.9666635780031746, "3000–6000 Hz_10ms_modulation_mae_db": 0.9457489278854361, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4424801083309347, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13684042189812712, "3000–6000 Hz_40ms_envelope_correlation": 0.9866518370409194, "3000–6000 Hz_40ms_modulation_mae_db": 0.8765066540120926, "mr_stft_db": 5.729596581232168, "band_mae_db": 0.24258076125899064, "rms_gain_error_db": -0.06803827760014547, "output_peak": 0.6019103966879992}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0303327264869377, "全帯域_10ms_rms_envelope_nrmse": 0.07328116814392825, "全帯域_10ms_envelope_correlation": 0.9940115873391722, "全帯域_10ms_modulation_mae_db": 0.17591345291080676, "全帯域_40ms_log_rms_mae_db": 0.5381949865301179, "全帯域_40ms_rms_envelope_nrmse": 0.03669103332618828, "全帯域_40ms_envelope_correlation": 0.9983059625143583, "全帯域_40ms_modulation_mae_db": 0.05119741489917958, "300–1000 Hz_10ms_log_rms_mae_db": 1.7054693823945928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1165571675025951, "300–1000 Hz_10ms_envelope_correlation": 0.988022217335843, "300–1000 Hz_10ms_modulation_mae_db": 0.16605388324454182, "300–1000 Hz_40ms_log_rms_mae_db": 0.9518492962631733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06508784842240145, "300–1000 Hz_40ms_envelope_correlation": 0.9958441993515611, "300–1000 Hz_40ms_modulation_mae_db": 0.2635429090771144, "1000–3000 Hz_10ms_log_rms_mae_db": 2.032279904715206, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19554070872169427, "1000–3000 Hz_10ms_envelope_correlation": 0.9699105064964815, "1000–3000 Hz_10ms_modulation_mae_db": 0.8317208987920655, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2420916622939968, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12074838969841589, "1000–3000 Hz_40ms_envelope_correlation": 0.9895603689682058, "1000–3000 Hz_40ms_modulation_mae_db": 1.1400336007370335, "3000–6000 Hz_10ms_log_rms_mae_db": 2.24221510034998, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2066504792280634, "3000–6000 Hz_10ms_envelope_correlation": 0.9711318776644089, "3000–6000 Hz_10ms_modulation_mae_db": 0.3995146085212357, "3000–6000 Hz_40ms_log_rms_mae_db": 1.46186932664321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13056562926679782, "3000–6000 Hz_40ms_envelope_correlation": 0.9876532038553534, "3000–6000 Hz_40ms_modulation_mae_db": 0.4495597337372554, "mr_stft_db": 5.712406539807393, "band_mae_db": 0.1918438502975193, "rms_gain_error_db": -0.001023821493656366, "output_peak": 0.5973498503828856}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2260157436525345, "spectral_convergence": 0.2232997898006059, "spectral_convergence_status": "ok", "amplitude_floor": 2.016382236479878e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.518567748075958, "spectral_convergence": 0.21171182646334316, "spectral_convergence_status": "ok", "amplitude_floor": 1.164576113083389e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.4401742145692475, "spectral_convergence": 0.22321542753241644, "spectral_convergence_status": "ok", "amplitude_floor": 9.482830334434223e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0029188205763026483, "output_power": 0.002981119270743382, "delta_db": 0.09171952781400657, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003712420548895462, "output_power": 0.003699906371115591, "delta_db": -0.014664338412784424, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009931195946472759, "output_power": 0.0009127136162918374, "delta_db": -0.3666702119106046, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003469926396566426, "output_power": 0.00030909121740892664, "delta_db": -0.5023359762244349, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013263724937728663, "output_power": 0.00013103640545626048, "delta_db": -0.05273535759588294, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.438064081917431e-05, "output_power": 6.418927407608009e-05, "delta_db": -0.01292830939295747, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17350895355844517, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03754684053277464, "pitch": null, "output_rms_dbfs": -20.916192159891832, "output_peak": 0.5822143968231266, "reference_rms_dbfs": -20.87864531935906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.728252568765913}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023582673937264833, "log_rms_mae_db": 0.9881335449907396, "rms_envelope_nrmse": 0.06708179082557958, "envelope_correlation": 0.9949904676424628, "rms_gain_error_db": -0.03754545606707692, "reference_mean_power": 0.008175179133340451, "output_mean_power": 0.008104808018699923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4118743138857815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017924334098852407, "output_power": 0.0017304350386534158, "reference_active": true, "delta_db": -0.15287729988777118}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00033647515979494424, "output_power": 0.00033023521421282414, "reference_active": true, "delta_db": -0.08129626211496462}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.450579873055231e-05, "output_power": 6.047436964657724e-05, "reference_active": true, "delta_db": 0.45128641999348074}], "active_band_count": 3, "active_band_mae_db": 0.22848666066540554}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002175646104262597, "log_rms_mae_db": 0.50846248719716, "rms_envelope_nrmse": 0.03460214594022085, "envelope_correlation": 0.9985156265422227, "rms_gain_error_db": -0.03753259195023606, "reference_mean_power": 0.008216199965234223, "output_mean_power": 0.00814549987494848}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.000498094679175e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016116003385924513, "output_power": 0.0015716214537324282, "reference_active": true, "delta_db": -0.10909401356567859}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017199808112094875, "output_power": 0.0001664294870973972, "reference_active": true, "delta_db": -0.14293326998285863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5420165509687798e-06, "output_power": 3.5800313088990155e-06, "reference_active": true, "delta_db": 0.04636238541072268}], "active_band_count": 3, "active_band_mae_db": 0.09946322298641996}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019245421130728438, "log_rms_mae_db": 1.648409852025631, "rms_envelope_nrmse": 0.11195673063138456, "envelope_correlation": 0.9891150216454441, "rms_gain_error_db": -0.1594758581298248, "reference_mean_power": 0.0024199693169718968, "output_mean_power": 0.002332718178373632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3235427499409267e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006913978889533925, "output_power": 0.0006213492454193889, "reference_active": true, "delta_db": -0.4639227392076244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018527392767596858, "output_power": 0.00016677511822651937, "reference_active": true, "delta_db": -0.45683051151660947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604035138941515e-05, "output_power": 3.0850858621189784e-05, "reference_active": true, "delta_db": 0.8095884099194955}], "active_band_count": 3, "active_band_mae_db": 0.5767805535479098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017723694424605393, "log_rms_mae_db": 0.9058412378156193, "rms_envelope_nrmse": 0.061827965533251546, "envelope_correlation": 0.9965349586773836, "rms_gain_error_db": -0.15947305872092196, "reference_mean_power": 0.0024321306730885443, "output_mean_power": 0.002344442572291262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1677626980235007e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006540469587465989, "output_power": 0.0005923289793130154, "reference_active": true, "delta_db": -0.43045949837684494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.291953991407944e-05, "output_power": 8.442855394258848e-05, "reference_active": true, "delta_db": -0.4161769982825715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.049791938534597e-06, "output_power": 2.2078730354195246e-06, "reference_active": true, "delta_db": 0.32264314709296604}], "active_band_count": 3, "active_band_mae_db": 0.38975988125079414}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.001235221787968e-05, "log_rms_mae_db": 2.0550235925936415, "rms_envelope_nrmse": 0.18082806731225398, "envelope_correlation": 0.973961587600433, "rms_gain_error_db": -0.4524852528865974, "reference_mean_power": 0.0003849600012580844, "output_mean_power": 0.0003468702727309381}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5531826233320727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012212996791699553, "output_power": 0.00010704891404371914, "reference_active": true, "delta_db": -0.57239977053606}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.773842605512037e-05, "output_power": 2.1111797548345733e-05, "reference_active": true, "delta_db": -1.1855660196142146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.9827801273791894e-06, "output_power": 5.83648169528445e-06, "reference_active": true, "delta_db": -0.7787723946201747}], "active_band_count": 3, "active_band_mae_db": 0.8455793949234831}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.393121399816863e-05, "log_rms_mae_db": 1.2443002014403959, "rms_envelope_nrmse": 0.11252652726730535, "envelope_correlation": 0.9902736410340252, "rms_gain_error_db": -0.4524868089825062, "reference_mean_power": 0.0003868956457514561, "output_mean_power": 0.0003486142706677867}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2949024312838256e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011569002203634676, "output_power": 0.00010055839378612436, "reference_active": true, "delta_db": -0.6087757602621591}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4800082503354946e-05, "output_power": 1.0810751668107121e-05, "reference_active": true, "delta_db": -1.3640824502534545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.803144704877755e-07, "output_power": 3.4513065139290825e-07, "reference_active": true, "delta_db": -2.2567986896695085}], "active_band_count": 3, "active_band_mae_db": 1.4098856333950407}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.314726805023413e-05, "log_rms_mae_db": 2.193174287418817, "rms_envelope_nrmse": 0.18753302870830274, "envelope_correlation": 0.9758555522420513, "rms_gain_error_db": -0.10099689354971644, "reference_mean_power": 8.262304385157997e-05, "output_mean_power": 8.072378203651743e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.2225580529948354e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.126031645648209e-05, "output_power": 2.0821975811516242e-05, "reference_active": true, "delta_db": -0.09047787010653856}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.562486047803945e-06, "output_power": 9.434740642713866e-06, "reference_active": true, "delta_db": -0.05840848420845284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.527384735880018e-06, "output_power": 2.3419102922403185e-06, "reference_active": true, "delta_db": -1.7788257600537085}], "active_band_count": 3, "active_band_mae_db": 0.6425707047895667}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6400870566820594e-05, "log_rms_mae_db": 1.406423876959401, "rms_envelope_nrmse": 0.11512460110034921, "envelope_correlation": 0.9902007281155333, "rms_gain_error_db": -0.10099723743324424, "reference_mean_power": 8.30385039555115e-05, "output_mean_power": 8.112968550532223e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.767621782464082e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.197465758003415e-05, "output_power": 2.163410711960244e-05, "reference_active": true, "delta_db": -0.0678314065508539}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.865612603723312e-06, "output_power": 5.120960841874273e-06, "reference_active": true, "delta_db": -0.5896192151743538}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.828054895066371e-07, "output_power": 1.958898513452467e-07, "reference_active": true, "delta_db": -1.5947589858588116}], "active_band_count": 3, "active_band_mae_db": 0.7507365358613397}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9076527153443976, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fb5b0c463ae87bd8e02ae96b131c37009d1b5dd248dacb3468d112ad77cf7619", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9722228402817458, "standard_deviation": 0.004478026529819759, "minimum": 0.9650046299178402, "maximum": 0.9779914226439425, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.025222221519967, "standard_deviation": 0.03770634414720979, "minimum": 1.9491780271900112, "maximum": 2.0550235925936415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6412240315360245, "standard_deviation": 0.19491914395229587, "minimum": 0.38274113126852427, "maximum": 0.8455793949234831, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18611560205059208, "standard_deviation": 0.014360380017222786, "minimum": 0.16762219957224436, "maximum": 0.20742370461372692, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.990063825714985, "standard_deviation": 0.0028192541768855767, "minimum": 0.9858986078232279, "maximum": 0.993837964368652, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2374435028891586, "standard_deviation": 0.04083213403034902, "minimum": 1.171245758149151, "maximum": 1.304764087932626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7954807988246202, "standard_deviation": 0.41671681997347265, "minimum": 0.22196160665433937, "maximum": 1.4098856333950407, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11344289964059043, "standard_deviation": 0.014012356344103373, "minimum": 0.08818210711936705, "maximum": 0.12927141072889278, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.972096649694975, "standard_deviation": 0.0028809693688995022, "minimum": 0.9666635780031746, "maximum": 0.9758555522420513, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.230448834893031, "standard_deviation": 0.02895827145990688, "minimum": 2.1863301387559515, "maximum": 2.2714503434745428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6435198248261685, "standard_deviation": 0.17714603988001681, "minimum": 0.3995146085212357, "maximum": 0.9457489278854361, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20143015905027184, "standard_deviation": 0.01030712692457236, "minimum": 0.18753302870830274, "maximum": 0.22003908694736954, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9880598085518133, "standard_deviation": 0.0013248328216309334, "minimum": 0.9865780563124332, "maximum": 0.9902007281155333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4512732225407978, "standard_deviation": 0.023448239173764672, "minimum": 1.406423876959401, "maximum": 1.4800687261467242, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6165738785186956, "standard_deviation": 0.21075591589081089, "minimum": 0.31093630036395664, "maximum": 0.8765066540120926, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12750548935345957, "standard_deviation": 0.0074962898217322636, "minimum": 0.11512460110034921, "maximum": 0.13684042189812712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9875552565677785, "standard_deviation": 0.0013327532608946886, "minimum": 0.9849159996997362, "maximum": 0.9891150216454441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6749352022630488, "standard_deviation": 0.03223928187343133, "minimum": 1.6301070605268424, "maximum": 1.7135905134177554, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5356278169474118, "standard_deviation": 0.26869064031494244, "minimum": 0.16605388324454182, "maximum": 1.0370889691124099, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11932241409562827, "standard_deviation": 0.00600691888989869, "minimum": 0.11195673063138456, "maximum": 0.13084067060497112, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9953778165498051, "standard_deviation": 0.000738324011789611, "minimum": 0.9943341169985646, "maximum": 0.9965349586773836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9602241523849233, "standard_deviation": 0.03160612269016902, "minimum": 0.9058412378156193, "maximum": 0.9996029076624571, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5890229315214045, "standard_deviation": 0.3658479223046568, "minimum": 0.22231434888919854, "maximum": 1.2684768787691894, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07020954996240406, "standard_deviation": 0.005384165478451152, "minimum": 0.061827965533251546, "maximum": 0.07784899283199348, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19915727164862687, "standard_deviation": 0.028049336170437672, "minimum": 0.16708951932311913, "maximum": 0.24258076125899064, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7249009928591565, "standard_deviation": 0.008344700227097669, "minimum": 5.711669349966084, "maximum": 5.732940872234053, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.594782334508529, "standard_deviation": 0.022330435732359318, "minimum": 0.555150124091766, "maximum": 0.6263504917562552, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05267940140791948, "standard_deviation": 0.03198533301230868, "minimum": -0.09269180782547562, "maximum": -0.001023821493656366, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994728054287833, "standard_deviation": 0.00044545792973560726, "minimum": 0.9940115873391722, "maximum": 0.9954049072596197, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0149329031998056, "standard_deviation": 0.01910156193730445, "minimum": 0.9874284352016987, "maximum": 1.033014987535261, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.19905619896865256, "standard_deviation": 0.06612467143780189, "minimum": 0.12723604636682215, "maximum": 0.3288430496643273, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06889237587841102, "standard_deviation": 0.002931491428850353, "minimum": 0.06437666938003157, "maximum": 0.07328116814392825, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984581418953595, "standard_deviation": 0.00021454526764205973, "minimum": 0.9980787449497932, "maximum": 0.9988275677702002, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5128193431646947, "standard_deviation": 0.018006419853502174, "minimum": 0.48332973740154705, "maximum": 0.5381949865301179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31489310546919314, "standard_deviation": 0.18321407167602405, "minimum": 0.05119741489917958, "maximum": 0.5324626137027954, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.035511871003661724, "standard_deviation": 0.002604668567324178, "minimum": 0.031248577526800554, "maximum": 0.04037769024102233, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0004", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.7500109672546387, "predicted_mos": 2.7500109672546387, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6b257c7e20700335d3029c1c2734c930d2848326acb4931fd49393adae8b65b4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.880379800219089, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0076237647272792, "全帯域_10ms_rms_envelope_nrmse": 0.06898166459223498, "全帯域_10ms_envelope_correlation": 0.9946984193073387, "全帯域_10ms_modulation_mae_db": 0.18479192320508506, "全帯域_40ms_log_rms_mae_db": 0.5368439136470413, "全帯域_40ms_rms_envelope_nrmse": 0.035285717747860774, "全帯域_40ms_envelope_correlation": 0.998457879138836, "全帯域_40ms_modulation_mae_db": 0.5118040169124837, "300–1000 Hz_10ms_log_rms_mae_db": 1.9116425487676776, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11992429079885994, "300–1000 Hz_10ms_envelope_correlation": 0.9876144804128992, "300–1000 Hz_10ms_modulation_mae_db": 0.4608055503239498, "300–1000 Hz_40ms_log_rms_mae_db": 1.1668816007236342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07470376243271674, "300–1000 Hz_40ms_envelope_correlation": 0.9949891671937907, "300–1000 Hz_40ms_modulation_mae_db": 0.7423298979368123, "1000–3000 Hz_10ms_log_rms_mae_db": 2.250282017131702, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2116307166527656, "1000–3000 Hz_10ms_envelope_correlation": 0.9628198433486107, "1000–3000 Hz_10ms_modulation_mae_db": 0.6925333040030882, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3934734256651597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12742777793670057, "1000–3000 Hz_40ms_envelope_correlation": 0.9851998340032836, "1000–3000 Hz_40ms_modulation_mae_db": 0.32472521328414555, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7790851076062593, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25127140879321525, "3000–6000 Hz_10ms_envelope_correlation": 0.957000284212706, "3000–6000 Hz_10ms_modulation_mae_db": 0.5356382451432306, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7638953156804027, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15226279336535264, "3000–6000 Hz_40ms_envelope_correlation": 0.9828827125150721, "3000–6000 Hz_40ms_modulation_mae_db": 0.7112881052449266, "mr_stft_db": 6.131917137100781, "band_mae_db": 0.16406913802343973, "rms_gain_error_db": -0.045755532057624594, "output_peak": 0.5939186801255382}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9976325631822168, "全帯域_10ms_rms_envelope_nrmse": 0.0673530819497006, "全帯域_10ms_envelope_correlation": 0.9949958281385727, "全帯域_10ms_modulation_mae_db": 0.2260646214281997, "全帯域_40ms_log_rms_mae_db": 0.5035483293914335, "全帯域_40ms_rms_envelope_nrmse": 0.03200135652942504, "全帯域_40ms_envelope_correlation": 0.9987993084562817, "全帯域_40ms_modulation_mae_db": 0.3354623401405645, "300–1000 Hz_10ms_log_rms_mae_db": 1.896287072798552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12454712597406334, "300–1000 Hz_10ms_envelope_correlation": 0.9869340686941421, "300–1000 Hz_10ms_modulation_mae_db": 0.8084925351283075, "300–1000 Hz_40ms_log_rms_mae_db": 1.1493575476978442, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07751201427651518, "300–1000 Hz_40ms_envelope_correlation": 0.9950064621730311, "300–1000 Hz_40ms_modulation_mae_db": 0.7358429330872581, "1000–3000 Hz_10ms_log_rms_mae_db": 2.283875503394969, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2349327265955874, "1000–3000 Hz_10ms_envelope_correlation": 0.9549116593295881, "1000–3000 Hz_10ms_modulation_mae_db": 1.0217236521638065, "1000–3000 Hz_40ms_log_rms_mae_db": 1.369911893760292, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15151290048263621, "1000–3000 Hz_40ms_envelope_correlation": 0.9790494389682604, "1000–3000 Hz_40ms_modulation_mae_db": 0.8207733426281401, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8052249455034284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2627402042246428, "3000–6000 Hz_10ms_envelope_correlation": 0.9528008777055514, "3000–6000 Hz_10ms_modulation_mae_db": 0.8064027703942468, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7831443282594932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1576809712841352, "3000–6000 Hz_40ms_envelope_correlation": 0.9816413864803523, "3000–6000 Hz_40ms_modulation_mae_db": 0.9208078762813274, "mr_stft_db": 6.1090853151927185, "band_mae_db": 0.1531760781079925, "rms_gain_error_db": -0.07337721298171294, "output_peak": 0.5894331359891488}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.033006965668178, "全帯域_10ms_rms_envelope_nrmse": 0.06967084821974569, "全帯域_10ms_envelope_correlation": 0.9947064003981001, "全帯域_10ms_modulation_mae_db": 0.46634379693774, "全帯域_40ms_log_rms_mae_db": 0.5164632429658734, "全帯域_40ms_rms_envelope_nrmse": 0.03623462980663272, "全帯域_40ms_envelope_correlation": 0.9985186764770636, "全帯域_40ms_modulation_mae_db": 0.8021406174382189, "300–1000 Hz_10ms_log_rms_mae_db": 1.922954088622317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1279286233981639, "300–1000 Hz_10ms_envelope_correlation": 0.9860725078277768, "300–1000 Hz_10ms_modulation_mae_db": 0.5971527914403305, "300–1000 Hz_40ms_log_rms_mae_db": 1.150619306147952, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0775963734206598, "300–1000 Hz_40ms_envelope_correlation": 0.9948814259884079, "300–1000 Hz_40ms_modulation_mae_db": 0.9429769732813692, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1820158650482817, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1855778364266299, "1000–3000 Hz_10ms_envelope_correlation": 0.9720423483594405, "1000–3000 Hz_10ms_modulation_mae_db": 0.5808004105685208, "1000–3000 Hz_40ms_log_rms_mae_db": 1.269458081992802, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09706024687694924, "1000–3000 Hz_40ms_envelope_correlation": 0.992804478741937, "1000–3000 Hz_40ms_modulation_mae_db": 0.6886822575335433, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8026204094973215, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2541037909898695, "3000–6000 Hz_10ms_envelope_correlation": 0.9569536746114641, "3000–6000 Hz_10ms_modulation_mae_db": 1.0347498190787239, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7590786717953657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.162756593922555, "3000–6000 Hz_40ms_envelope_correlation": 0.9809803811181625, "3000–6000 Hz_40ms_modulation_mae_db": 1.1497084338590107, "mr_stft_db": 6.1081583279215295, "band_mae_db": 0.23686439356447028, "rms_gain_error_db": -0.1039755447177535, "output_peak": 0.5817057603727848}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9885397434037817, "全帯域_10ms_rms_envelope_nrmse": 0.06499430153281262, "全帯域_10ms_envelope_correlation": 0.9952961732457207, "全帯域_10ms_modulation_mae_db": 0.10175036116653559, "全帯域_40ms_log_rms_mae_db": 0.49470877536915964, "全帯域_40ms_rms_envelope_nrmse": 0.033391498139252566, "全帯域_40ms_envelope_correlation": 0.9985952600314821, "全帯域_40ms_modulation_mae_db": 0.31944302342267755, "300–1000 Hz_10ms_log_rms_mae_db": 1.8459343642817507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11067533837586946, "300–1000 Hz_10ms_envelope_correlation": 0.989227428061618, "300–1000 Hz_10ms_modulation_mae_db": 0.4605858149669711, "300–1000 Hz_40ms_log_rms_mae_db": 1.061480498994816, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060178324801167776, "300–1000 Hz_40ms_envelope_correlation": 0.9964947228286618, "300–1000 Hz_40ms_modulation_mae_db": 0.6763169431897214, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2568869018805775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20932583430637508, "1000–3000 Hz_10ms_envelope_correlation": 0.9637967492704612, "1000–3000 Hz_10ms_modulation_mae_db": 0.855844448184385, "1000–3000 Hz_40ms_log_rms_mae_db": 1.35705601278194, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11537339663307299, "1000–3000 Hz_40ms_envelope_correlation": 0.9887854844205379, "1000–3000 Hz_40ms_modulation_mae_db": 0.4860068713862975, "3000–6000 Hz_10ms_log_rms_mae_db": 2.742133306533236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24845222539255477, "3000–6000 Hz_10ms_envelope_correlation": 0.9576471953839586, "3000–6000 Hz_10ms_modulation_mae_db": 1.162830811866211, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7272191483333177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15545184955368263, "3000–6000 Hz_40ms_envelope_correlation": 0.9820875265571825, "3000–6000 Hz_40ms_modulation_mae_db": 1.2674354645039845, "mr_stft_db": 6.101448462701768, "band_mae_db": 0.1676857353130298, "rms_gain_error_db": 0.006502227466424832, "output_peak": 0.5887661495524877}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9803245633959544, "全帯域_10ms_rms_envelope_nrmse": 0.0654424457178482, "全帯域_10ms_envelope_correlation": 0.9952271441021966, "全帯域_10ms_modulation_mae_db": 0.4020772176594553, "全帯域_40ms_log_rms_mae_db": 0.48288104129747034, "全帯域_40ms_rms_envelope_nrmse": 0.03189393633594616, "全帯域_40ms_envelope_correlation": 0.9987167426643595, "全帯域_40ms_modulation_mae_db": 0.6346487850161967, "300–1000 Hz_10ms_log_rms_mae_db": 1.8908188498682499, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1125835557461838, "300–1000 Hz_10ms_envelope_correlation": 0.989059853462887, "300–1000 Hz_10ms_modulation_mae_db": 0.9225323880161177, "300–1000 Hz_40ms_log_rms_mae_db": 1.1155497057408519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06650297975998805, "300–1000 Hz_40ms_envelope_correlation": 0.9959980161511176, "300–1000 Hz_40ms_modulation_mae_db": 1.011692221978847, "1000–3000 Hz_10ms_log_rms_mae_db": 2.240351111141725, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20351273420299817, "1000–3000 Hz_10ms_envelope_correlation": 0.9659693332238319, "1000–3000 Hz_10ms_modulation_mae_db": 0.3470228157116532, "1000–3000 Hz_40ms_log_rms_mae_db": 1.391861096897544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12420325788516849, "1000–3000 Hz_40ms_envelope_correlation": 0.9858937399011338, "1000–3000 Hz_40ms_modulation_mae_db": 0.49088661446047516, "3000–6000 Hz_10ms_log_rms_mae_db": 2.767943688578147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24282395202390114, "3000–6000 Hz_10ms_envelope_correlation": 0.9596546689764865, "3000–6000 Hz_10ms_modulation_mae_db": 0.9143534526256891, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7225069676702263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14767050265153817, "3000–6000 Hz_40ms_envelope_correlation": 0.9839690268569526, "3000–6000 Hz_40ms_modulation_mae_db": 1.0358466050034487, "mr_stft_db": 6.122110164309281, "band_mae_db": 0.10280538373603093, "rms_gain_error_db": -0.009298307714807762, "output_peak": 0.6221954718017702}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9976830084534695, "全帯域_10ms_rms_envelope_nrmse": 0.0684144784367255, "全帯域_10ms_envelope_correlation": 0.9947814942591686, "全帯域_10ms_modulation_mae_db": 0.17165652086821984, "全帯域_40ms_log_rms_mae_db": 0.5373095869033985, "全帯域_40ms_rms_envelope_nrmse": 0.03620972912801989, "全帯域_40ms_envelope_correlation": 0.9983550835801148, "全帯域_40ms_modulation_mae_db": 0.18393039877585626, "300–1000 Hz_10ms_log_rms_mae_db": 1.8791328269587348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10951689975773571, "300–1000 Hz_10ms_envelope_correlation": 0.9898240967057798, "300–1000 Hz_10ms_modulation_mae_db": 0.6154320502403365, "300–1000 Hz_40ms_log_rms_mae_db": 1.1773380194555332, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06593948301359935, "300–1000 Hz_40ms_envelope_correlation": 0.9962664825942047, "300–1000 Hz_40ms_modulation_mae_db": 0.732269898865578, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2777099142263, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21923901786042038, "1000–3000 Hz_10ms_envelope_correlation": 0.9600140196327978, "1000–3000 Hz_10ms_modulation_mae_db": 0.4892112546104867, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3860573317350438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12540902204279905, "1000–3000 Hz_40ms_envelope_correlation": 0.9855593191933067, "1000–3000 Hz_40ms_modulation_mae_db": 0.5223582977681366, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7765425007763134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26215886513090475, "3000–6000 Hz_10ms_envelope_correlation": 0.9535233488398264, "3000–6000 Hz_10ms_modulation_mae_db": 0.9578322968165199, "3000–6000 Hz_40ms_log_rms_mae_db": 1.714286528483553, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15303289385804453, "3000–6000 Hz_40ms_envelope_correlation": 0.9829229656767303, "3000–6000 Hz_40ms_modulation_mae_db": 0.9441952447723088, "mr_stft_db": 6.124307340833649, "band_mae_db": 0.14880145390370447, "rms_gain_error_db": -0.026610027604673618, "output_peak": 0.5995993538583679}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.014879601606265, "全帯域_10ms_rms_envelope_nrmse": 0.06630353622741761, "全帯域_10ms_envelope_correlation": 0.9951068442170898, "全帯域_10ms_modulation_mae_db": 0.19062003949773118, "全帯域_40ms_log_rms_mae_db": 0.5140545312170645, "全帯域_40ms_rms_envelope_nrmse": 0.03402153827663216, "全帯域_40ms_envelope_correlation": 0.9985702147305349, "全帯域_40ms_modulation_mae_db": 0.4352570643117408, "300–1000 Hz_10ms_log_rms_mae_db": 1.9436588536690482, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11695108872517868, "300–1000 Hz_10ms_envelope_correlation": 0.9879649418807752, "300–1000 Hz_10ms_modulation_mae_db": 0.3921733552830769, "300–1000 Hz_40ms_log_rms_mae_db": 1.1215227052016565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06560497193031284, "300–1000 Hz_40ms_envelope_correlation": 0.9957884515068722, "300–1000 Hz_40ms_modulation_mae_db": 0.4334226898929174, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2526403809308904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21372709768592962, "1000–3000 Hz_10ms_envelope_correlation": 0.9620354162462394, "1000–3000 Hz_10ms_modulation_mae_db": 0.43082804140307746, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4034037407383078, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12078400698037872, "1000–3000 Hz_40ms_envelope_correlation": 0.9867088375454556, "1000–3000 Hz_40ms_modulation_mae_db": 0.29563203134051713, "3000–6000 Hz_10ms_log_rms_mae_db": 2.75180755378769, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25416459422472715, "3000–6000 Hz_10ms_envelope_correlation": 0.9556787038622673, "3000–6000 Hz_10ms_modulation_mae_db": 1.0698895587272015, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7749628638076325, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15972434463361607, "3000–6000 Hz_40ms_envelope_correlation": 0.9814070902785705, "3000–6000 Hz_40ms_modulation_mae_db": 1.2284638247188016, "mr_stft_db": 6.114939942301464, "band_mae_db": 0.13729995693395589, "rms_gain_error_db": -0.04776462013484928, "output_peak": 0.580140825502904}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.01875073441677, "全帯域_10ms_rms_envelope_nrmse": 0.07107156003127965, "全帯域_10ms_envelope_correlation": 0.9943622840297119, "全帯域_10ms_modulation_mae_db": 0.23497653842099822, "全帯域_40ms_log_rms_mae_db": 0.538276104574294, "全帯域_40ms_rms_envelope_nrmse": 0.03922365846063748, "全帯域_40ms_envelope_correlation": 0.998061675975992, "全帯域_40ms_modulation_mae_db": 0.08746895587647134, "300–1000 Hz_10ms_log_rms_mae_db": 1.8366171993630869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11893033004493168, "300–1000 Hz_10ms_envelope_correlation": 0.987792734851196, "300–1000 Hz_10ms_modulation_mae_db": 0.4649328331377931, "300–1000 Hz_40ms_log_rms_mae_db": 1.0841179398760692, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07741598591074542, "300–1000 Hz_40ms_envelope_correlation": 0.9945275400151095, "300–1000 Hz_40ms_modulation_mae_db": 0.508135878186471, "1000–3000 Hz_10ms_log_rms_mae_db": 2.333598450247887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21830739915390154, "1000–3000 Hz_10ms_envelope_correlation": 0.9603804215217817, "1000–3000 Hz_10ms_modulation_mae_db": 0.7298369284099347, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4536442873221156, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13145826725808668, "1000–3000 Hz_40ms_envelope_correlation": 0.9847127840125369, "1000–3000 Hz_40ms_modulation_mae_db": 0.5598499881626603, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7324782178639544, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2366638620742432, "3000–6000 Hz_10ms_envelope_correlation": 0.9618647432455146, "3000–6000 Hz_10ms_modulation_mae_db": 1.6891884699234694, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7131436682573875, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15217694785584254, "3000–6000 Hz_40ms_envelope_correlation": 0.9840129552311804, "3000–6000 Hz_40ms_modulation_mae_db": 1.7395080888417898, "mr_stft_db": 6.111747569265845, "band_mae_db": 0.16073527772032487, "rms_gain_error_db": -0.02013858095775207, "output_peak": 0.5813142730554728}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.716379476492258, "spectral_convergence": 0.22454011938807283, "spectral_convergence_status": "ok", "amplitude_floor": 2.016382236479878e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.997667487843802, "spectral_convergence": 0.21833353129064167, "spectral_convergence_status": "ok", "amplitude_floor": 1.164576113083389e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.681704446966282, "spectral_convergence": 0.22538852130481096, "spectral_convergence_status": "ok", "amplitude_floor": 9.482830334434223e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0029188205763026483, "output_power": 0.003020330989675689, "delta_db": 0.14847139528109243, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003712420548895462, "output_power": 0.003615746448711172, "delta_db": -0.11459199579668614, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009931195946472759, "output_power": 0.0009145904630014198, "delta_db": -0.35774882367291816, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003469926396566426, "output_power": 0.0003326906889372616, "delta_db": -0.18279616338747665, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013263724937728663, "output_power": 0.00013240547273629143, "delta_db": -0.007595706916577633, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.438064081917431e-05, "output_power": 6.700024171146274e-05, "delta_db": 0.17321074308588752, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.16406913802343973, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.045755532057624594, "pitch": null, "output_rms_dbfs": -20.92440085141668, "output_peak": 0.5939186801255382, "reference_rms_dbfs": -20.87864531935906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.131917137100781}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023582673937264833, "log_rms_mae_db": 1.0076237647272792, "rms_envelope_nrmse": 0.06898166459223498, "envelope_correlation": 0.9946984193073387, "rms_gain_error_db": -0.04575450830256549, "reference_mean_power": 0.008175179133340451, "output_mean_power": 0.00808950274670718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4118743138857815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017924334098852407, "output_power": 0.0017722455054594688, "reference_active": true, "delta_db": -0.04919146681685432}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00033647515979494424, "output_power": 0.0003194444486271404, "reference_active": true, "delta_db": -0.2255766277547503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.450579873055231e-05, "output_power": 5.1107182664998243e-05, "reference_active": true, "delta_db": -0.2796076750436505}], "active_band_count": 3, "active_band_mae_db": 0.18479192320508506}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002175646104262597, "log_rms_mae_db": 0.5368439136470413, "rms_envelope_nrmse": 0.035285717747860774, "envelope_correlation": 0.998457879138836, "rms_gain_error_db": -0.045757400332277515, "reference_mean_power": 0.008216199965234223, "output_mean_power": 0.008130088263782037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.000498094679175e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016116003385924513, "output_power": 0.0015931339728457218, "reference_active": true, "delta_db": -0.05005051107658125}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017199808112094875, "output_power": 0.0001613594559487103, "reference_active": true, "delta_db": -0.27729180864196135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5420165509687798e-06, "output_power": 2.6819063780321552e-06, "reference_active": true, "delta_db": -1.2080697310189086}], "active_band_count": 3, "active_band_mae_db": 0.5118040169124837}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019245421130728438, "log_rms_mae_db": 1.9116425487676776, "rms_envelope_nrmse": 0.11992429079885994, "envelope_correlation": 0.9876144804128992, "rms_gain_error_db": -0.21635540181569826, "reference_mean_power": 0.0024199693169718968, "output_mean_power": 0.0023023657674784087}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3235427499409267e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006913978889533925, "output_power": 0.0006490475439218961, "reference_active": true, "delta_db": -0.27451538343975745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018527392767596858, "output_power": 0.0001617391359676863, "reference_active": true, "delta_db": -0.5899918962580345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604035138941515e-05, "output_power": 2.272571089079616e-05, "reference_active": true, "delta_db": -0.5179093712740575}], "active_band_count": 3, "active_band_mae_db": 0.4608055503239498}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017723694424605393, "log_rms_mae_db": 1.1668816007236342, "rms_envelope_nrmse": 0.07470376243271674, "envelope_correlation": 0.9949891671937907, "rms_gain_error_db": -0.2163657997589297, "reference_mean_power": 0.0024321306730885443, "output_mean_power": 0.002313930576607517}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1677626980235007e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006540469587465989, "output_power": 0.0006083882996887211, "reference_active": true, "delta_db": -0.314280773141448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.291953991407944e-05, "output_power": 8.23872824660947e-05, "reference_active": true, "delta_db": -0.5224687280224131}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.049791938534597e-06, "output_power": 1.4882838241747012e-06, "reference_active": true, "delta_db": -1.3902401926465757}], "active_band_count": 3, "active_band_mae_db": 0.7423298979368123}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.001235221787968e-05, "log_rms_mae_db": 2.250282017131702, "rms_envelope_nrmse": 0.2116307166527656, "envelope_correlation": 0.9628198433486107, "rms_gain_error_db": -0.16366465422143248, "reference_mean_power": 0.0003849600012580844, "output_mean_power": 0.00037072266773455946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5531826233320727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012212996791699553, "output_power": 0.00012147224933459688, "reference_active": true, "delta_db": -0.02345169500548425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.773842605512037e-05, "output_power": 2.547343764576527e-05, "reference_active": true, "delta_db": -0.3699425747478434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.9827801273791894e-06, "output_power": 1.0290795110403876e-05, "reference_active": true, "delta_db": 1.684205642255937}], "active_band_count": 3, "active_band_mae_db": 0.6925333040030882}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.393121399816863e-05, "log_rms_mae_db": 1.3934734256651597, "rms_envelope_nrmse": 0.12742777793670057, "envelope_correlation": 0.9851998340032836, "rms_gain_error_db": -0.16366702430289348, "reference_mean_power": 0.0003868956457514561, "output_mean_power": 0.00037258652115634025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2949024312838256e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011569002203634676, "output_power": 0.00011823614488411353, "reference_active": true, "delta_db": 0.0945435721624892}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4800082503354946e-05, "output_power": 1.2365691147799795e-05, "reference_active": true, "delta_db": -0.7804574126085878}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.803144704877755e-07, "output_power": 5.67212684802551e-07, "reference_active": true, "delta_db": -0.09917465508135972}], "active_band_count": 3, "active_band_mae_db": 0.32472521328414555}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.314726805023413e-05, "log_rms_mae_db": 2.7790851076062593, "rms_envelope_nrmse": 0.25127140879321525, "envelope_correlation": 0.957000284212706, "rms_gain_error_db": 0.02207847765742968, "reference_mean_power": 8.262304385157997e-05, "output_mean_power": 8.304414884916153e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.2225580529948354e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.126031645648209e-05, "output_power": 2.1466799901448614e-05, "reference_active": true, "delta_db": 0.041975834784726815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.562486047803945e-06, "output_power": 9.043284581937166e-06, "reference_active": true, "delta_db": -0.24244616690102058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.527384735880018e-06, "output_power": 2.6013785678444364e-06, "reference_active": true, "delta_db": -1.3224927337439443}], "active_band_count": 3, "active_band_mae_db": 0.5356382451432306}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6400870566820594e-05, "log_rms_mae_db": 1.7638953156804027, "rms_envelope_nrmse": 0.15226279336535264, "envelope_correlation": 0.9828827125150721, "rms_gain_error_db": 0.022076775772499453, "reference_mean_power": 8.30385039555115e-05, "output_mean_power": 8.346169372282702e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.767621782464082e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.197465758003415e-05, "output_power": 2.216971021268433e-05, "reference_active": true, "delta_db": 0.03837899943177168}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.865612603723312e-06, "output_power": 4.802484956985935e-06, "reference_active": true, "delta_db": -0.8684736329448685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.828054895066371e-07, "output_power": 2.131997381281996e-07, "reference_active": true, "delta_db": -1.2270116833581395}], "active_band_count": 3, "active_band_mae_db": 0.7112881052449266}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8860256628903732, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "23e5da8cf99766ecffdbe44dc7dd3550cdb03e255f9000aeb8bc81e69f698248", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9627462238665939, "standard_deviation": 0.004969377621782023, "minimum": 0.9549116593295881, "maximum": 0.9720423483594405, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2596700180002918, "standard_deviation": 0.04297082099601415, "minimum": 2.1820158650482817, "maximum": 2.333598450247887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6434751068818692, "standard_deviation": 0.2264212127644232, "minimum": 0.3470228157116532, "maximum": 1.0217236521638065, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21203167036057596, "standard_deviation": 0.014134675680019304, "minimum": 0.1855778364266299, "maximum": 0.2349327265955874, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9860892395983065, "standard_deviation": 0.003878725407062601, "minimum": 0.9790494389682604, "maximum": 0.992804478741937, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3781082338616506, "standard_deviation": 0.052302974598477235, "minimum": 1.269458081992802, "maximum": 1.4536442873221156, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5236143270704895, "standard_deviation": 0.1735691437289847, "minimum": 0.29563203134051713, "maximum": 0.8207733426281401, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.124153609511974, "standard_deviation": 0.01529091322113936, "minimum": 0.09706024687694924, "maximum": 0.15151290048263621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9568904371047218, "standard_deviation": 0.0029881557809745825, "minimum": 0.9528008777055514, "maximum": 0.9618647432455146, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.7697294662682936, "standard_deviation": 0.026590213412899497, "minimum": 2.7324782178639544, "maximum": 2.8052249455034284, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0213606780719116, "standard_deviation": 0.33083480235083357, "minimum": 0.5356382451432306, "maximum": 1.6891884699234694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25154736285675733, "standard_deviation": 0.008932616173356122, "minimum": 0.2366638620742432, "maximum": 0.2627402042246428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9824880055892754, "standard_deviation": 0.001145321917700964, "minimum": 0.9809803811181625, "maximum": 0.9840129552311804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7447796865359222, "standard_deviation": 0.028505275276177425, "minimum": 1.7131436682573875, "maximum": 1.7831443282594932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1246567054031997, "standard_deviation": 0.30769024830713676, "minimum": 0.7112881052449266, "maximum": 1.7395080888417898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15509461214059583, "standard_deviation": 0.0048234147123195886, "minimum": 0.14767050265153817, "maximum": 0.162756593922555, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9880612639871342, "standard_deviation": 0.0012519368126901376, "minimum": 0.9860725078277768, "maximum": 0.9898240967057798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.890880725541177, "standard_deviation": 0.036600616575247144, "minimum": 1.8366171993630869, "maximum": 1.9436588536690482, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5902634148171104, "standard_deviation": 0.18793258221049863, "minimum": 0.3921733552830769, "maximum": 0.9225323880161177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11763215660262331, "standard_deviation": 0.006557221764691576, "minimum": 0.10951689975773571, "maximum": 0.1279286233981639, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954940335563994, "standard_deviation": 0.0007310218500404485, "minimum": 0.9945275400151095, "maximum": 0.9964947228286618, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1283584154797945, "standard_deviation": 0.040461999084504124, "minimum": 1.061480498994816, "maximum": 1.1773380194555332, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7228734295523718, "standard_deviation": 0.19440692731583994, "minimum": 0.4334226898929174, "maximum": 1.011692221978847, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07068173694321314, "standard_deviation": 0.006887233301148572, "minimum": 0.060178324801167776, "maximum": 0.0775963734206598, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15892967716286854, "standard_deviation": 0.03767080860096178, "minimum": 0.10280538373603093, "maximum": 0.23686439356447028, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.115464282453379, "standard_deviation": 0.00998330135036209, "minimum": 6.101448462701768, "maximum": 6.131917137100781, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5921342062823093, "standard_deviation": 0.013891912695448828, "minimum": 0.580140825502904, "maximum": 0.6221954718017702, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04005219983784361, "standard_deviation": 0.03581524164875959, "minimum": -0.1039755447177535, "maximum": 0.006502227466424832, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9948968234622374, "standard_deviation": 0.00031558584980904833, "minimum": 0.9943622840297119, "maximum": 0.9952961732457207, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0048051181067392, "standard_deviation": 0.01717678819941423, "minimum": 0.9803245633959544, "maximum": 1.033006965668178, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24728512739799563, "standard_deviation": 0.12338412158144667, "minimum": 0.10175036116653559, "maximum": 0.46634379693774, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06777898958847062, "standard_deviation": 0.002132828678626391, "minimum": 0.06499430153281262, "maximum": 0.07107156003127965, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985093551318331, "standard_deviation": 0.00022838534639011267, "minimum": 0.998061675975992, "maximum": 0.9987993084562817, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5155106906707169, "standard_deviation": 0.021017346209268314, "minimum": 0.48288104129747034, "maximum": 0.538276104574294, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4137694002367762, "standard_deviation": 0.23422206140793858, "minimum": 0.08746895587647134, "maximum": 0.8021406174382189, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03478275805305085, "standard_deviation": 0.0024702599043984427, "minimum": 0.03189393633594616, "maximum": 0.03922365846063748, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0004", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.032092332839966, "predicted_mos": 3.032092332839966, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c8148faefa1600c6c411d1fe932f0f2190eaefc234815997040476bf4b1e4f74", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8278802996501327, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.040808169479281, "全帯域_10ms_rms_envelope_nrmse": 0.07268249667908712, "全帯域_10ms_envelope_correlation": 0.9941227294658773, "全帯域_10ms_modulation_mae_db": 0.3457607477561066, "全帯域_40ms_log_rms_mae_db": 0.5162385964184124, "全帯域_40ms_rms_envelope_nrmse": 0.037434184832489226, "全帯域_40ms_envelope_correlation": 0.9982852536790309, "全帯域_40ms_modulation_mae_db": 0.590451520526846, "300–1000 Hz_10ms_log_rms_mae_db": 1.6452957472986174, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12204585717043047, "300–1000 Hz_10ms_envelope_correlation": 0.9869014985905397, "300–1000 Hz_10ms_modulation_mae_db": 0.45744964186743203, "300–1000 Hz_40ms_log_rms_mae_db": 0.9776799394732884, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06598781191775611, "300–1000 Hz_40ms_envelope_correlation": 0.9958518331773104, "300–1000 Hz_40ms_modulation_mae_db": 0.7178171910565568, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8608057931996946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18418114905025335, "1000–3000 Hz_10ms_envelope_correlation": 0.9727819222894489, "1000–3000 Hz_10ms_modulation_mae_db": 0.8769891327311807, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2051618092869427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1052912940885069, "1000–3000 Hz_40ms_envelope_correlation": 0.9916487527347966, "1000–3000 Hz_40ms_modulation_mae_db": 1.3540611887608012, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0084692344226913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18872430806588314, "3000–6000 Hz_10ms_envelope_correlation": 0.9762272701647731, "3000–6000 Hz_10ms_modulation_mae_db": 0.7891781594932418, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3270571734532612, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12589242457419003, "3000–6000 Hz_40ms_envelope_correlation": 0.9895325398550585, "3000–6000 Hz_40ms_modulation_mae_db": 0.9331896581830156, "mr_stft_db": 5.641650098452284, "band_mae_db": 0.2367206242276998, "rms_gain_error_db": -0.0628885522607816, "output_peak": 0.6295245139213238}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0530638302422757, "全帯域_10ms_rms_envelope_nrmse": 0.07235763827405528, "全帯域_10ms_envelope_correlation": 0.9942050121089889, "全帯域_10ms_modulation_mae_db": 0.2229150273594437, "全帯域_40ms_log_rms_mae_db": 0.5338530790329613, "全帯域_40ms_rms_envelope_nrmse": 0.038810968257152204, "全帯域_40ms_envelope_correlation": 0.998184778116787, "全帯域_40ms_modulation_mae_db": 0.19296970345327016, "300–1000 Hz_10ms_log_rms_mae_db": 1.6405447069825905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1273768517712352, "300–1000 Hz_10ms_envelope_correlation": 0.9862005667160412, "300–1000 Hz_10ms_modulation_mae_db": 0.3452713029828065, "300–1000 Hz_40ms_log_rms_mae_db": 0.9714135803642753, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07819363459808192, "300–1000 Hz_40ms_envelope_correlation": 0.9948208503033881, "300–1000 Hz_40ms_modulation_mae_db": 0.3350427649395096, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8983129087208748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17157786624353283, "1000–3000 Hz_10ms_envelope_correlation": 0.9776554106703085, "1000–3000 Hz_10ms_modulation_mae_db": 0.7935290234601838, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1532510634582072, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09957212652872925, "1000–3000 Hz_40ms_envelope_correlation": 0.9941796460649573, "1000–3000 Hz_40ms_modulation_mae_db": 1.0013781463956575, "3000–6000 Hz_10ms_log_rms_mae_db": 1.997907045937677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1971664270836002, "3000–6000 Hz_10ms_envelope_correlation": 0.9732892704123004, "3000–6000 Hz_10ms_modulation_mae_db": 0.3304263380608323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3397176396971744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12092174486254016, "3000–6000 Hz_40ms_envelope_correlation": 0.9891199821792312, "3000–6000 Hz_40ms_modulation_mae_db": 0.3422823210618225, "mr_stft_db": 5.637799726247969, "band_mae_db": 0.2659443549061494, "rms_gain_error_db": -0.07043491732839366, "output_peak": 0.5713472758035723}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0205154563738053, "全帯域_10ms_rms_envelope_nrmse": 0.0684359902033071, "全帯域_10ms_envelope_correlation": 0.9947843426084557, "全帯域_10ms_modulation_mae_db": 0.21420395594752697, "全帯域_40ms_log_rms_mae_db": 0.5129276449230593, "全帯域_40ms_rms_envelope_nrmse": 0.03486785175963494, "全帯域_40ms_envelope_correlation": 0.9984973300522918, "全帯域_40ms_modulation_mae_db": 0.41619137784961063, "300–1000 Hz_10ms_log_rms_mae_db": 1.5518553972662488, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12150291630213964, "300–1000 Hz_10ms_envelope_correlation": 0.9870694143484892, "300–1000 Hz_10ms_modulation_mae_db": 0.24459114309143928, "300–1000 Hz_40ms_log_rms_mae_db": 0.9684598946013142, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06450104405223801, "300–1000 Hz_40ms_envelope_correlation": 0.9961454833209465, "300–1000 Hz_40ms_modulation_mae_db": 0.18027100474120916, "1000–3000 Hz_10ms_log_rms_mae_db": 1.839198337210631, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1918740252184955, "1000–3000 Hz_10ms_envelope_correlation": 0.9697194243876681, "1000–3000 Hz_10ms_modulation_mae_db": 0.39036044919533214, "1000–3000 Hz_40ms_log_rms_mae_db": 1.13986236583439, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12357503194033577, "1000–3000 Hz_40ms_envelope_correlation": 0.9863525493989007, "1000–3000 Hz_40ms_modulation_mae_db": 0.22719717743916332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.991148456227327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18879987299950363, "3000–6000 Hz_10ms_envelope_correlation": 0.9756186823245199, "3000–6000 Hz_10ms_modulation_mae_db": 0.6744103654029076, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2883807050772722, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.111804795421064, "3000–6000 Hz_40ms_envelope_correlation": 0.9912145740170794, "3000–6000 Hz_40ms_modulation_mae_db": 0.7681119798427561, "mr_stft_db": 5.622880344204621, "band_mae_db": 0.14448817075714995, "rms_gain_error_db": -0.04617834949512296, "output_peak": 0.6075941341159682}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0284535015055365, "全帯域_10ms_rms_envelope_nrmse": 0.0644560640287867, "全帯域_10ms_envelope_correlation": 0.9953675689658804, "全帯域_10ms_modulation_mae_db": 0.14639030938103348, "全帯域_40ms_log_rms_mae_db": 0.5112844192395961, "全帯域_40ms_rms_envelope_nrmse": 0.029913799870148843, "全帯域_40ms_envelope_correlation": 0.9988809363916349, "全帯域_40ms_modulation_mae_db": 0.28820643952830244, "300–1000 Hz_10ms_log_rms_mae_db": 1.60424602093589, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11621233932108095, "300–1000 Hz_10ms_envelope_correlation": 0.9880972074905499, "300–1000 Hz_10ms_modulation_mae_db": 0.4533728334292995, "300–1000 Hz_40ms_log_rms_mae_db": 0.9777704213438982, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06642273682651505, "300–1000 Hz_40ms_envelope_correlation": 0.9956944052394635, "300–1000 Hz_40ms_modulation_mae_db": 0.665361619317963, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8102479455030445, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1720598166525494, "1000–3000 Hz_10ms_envelope_correlation": 0.9758866937680027, "1000–3000 Hz_10ms_modulation_mae_db": 0.6783305896269871, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1989968211574393, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10569783393928614, "1000–3000 Hz_40ms_envelope_correlation": 0.9905879479959137, "1000–3000 Hz_40ms_modulation_mae_db": 1.0774030891251871, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9711069334894216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1893994537375124, "3000–6000 Hz_10ms_envelope_correlation": 0.9755479809762915, "3000–6000 Hz_10ms_modulation_mae_db": 0.17173870429678043, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2983615918238682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11736283767216096, "3000–6000 Hz_40ms_envelope_correlation": 0.9897923162428727, "3000–6000 Hz_40ms_modulation_mae_db": 0.12066072020494391, "mr_stft_db": 5.641892117118459, "band_mae_db": 0.16620368650389006, "rms_gain_error_db": -0.026619416341482145, "output_peak": 0.6247689398417979}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.027887298765097, "全帯域_10ms_rms_envelope_nrmse": 0.06988982227761219, "全帯域_10ms_envelope_correlation": 0.9945876068925725, "全帯域_10ms_modulation_mae_db": 0.2857578283534602, "全帯域_40ms_log_rms_mae_db": 0.5411268772531507, "全帯域_40ms_rms_envelope_nrmse": 0.036630072440998696, "全帯域_40ms_envelope_correlation": 0.998381615142101, "全帯域_40ms_modulation_mae_db": 0.23982308998923588, "300–1000 Hz_10ms_log_rms_mae_db": 1.5292265632854567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11733202992987708, "300–1000 Hz_10ms_envelope_correlation": 0.9880135667869306, "300–1000 Hz_10ms_modulation_mae_db": 0.27526985704679746, "300–1000 Hz_40ms_log_rms_mae_db": 0.8934717928606277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06868184670993265, "300–1000 Hz_40ms_envelope_correlation": 0.9956637951675902, "300–1000 Hz_40ms_modulation_mae_db": 0.3557294490513181, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8403240746112342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17948499903632412, "1000–3000 Hz_10ms_envelope_correlation": 0.9739901416452861, "1000–3000 Hz_10ms_modulation_mae_db": 0.3677009859303782, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1941836084686923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1043722120638476, "1000–3000 Hz_40ms_envelope_correlation": 0.9913873046231537, "1000–3000 Hz_40ms_modulation_mae_db": 0.7967960293831675, "3000–6000 Hz_10ms_log_rms_mae_db": 2.024861514392913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20805334174478618, "3000–6000 Hz_10ms_envelope_correlation": 0.9703640369331084, "3000–6000 Hz_10ms_modulation_mae_db": 0.22602418098481328, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3361438125859832, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13324279719531762, "3000–6000 Hz_40ms_envelope_correlation": 0.9868630889737081, "3000–6000 Hz_40ms_modulation_mae_db": 0.37514134783601644, "mr_stft_db": 5.636389648533798, "band_mae_db": 0.1611623610705985, "rms_gain_error_db": -0.06724758191764979, "output_peak": 0.5934483598713867}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0598645206441974, "全帯域_10ms_rms_envelope_nrmse": 0.07030415194623398, "全帯域_10ms_envelope_correlation": 0.9945316253037426, "全帯域_10ms_modulation_mae_db": 0.2954255487779524, "全帯域_40ms_log_rms_mae_db": 0.543053098137875, "全帯域_40ms_rms_envelope_nrmse": 0.03450372908010201, "全帯域_40ms_envelope_correlation": 0.9985865306245467, "全帯域_40ms_modulation_mae_db": 0.34485323605303436, "300–1000 Hz_10ms_log_rms_mae_db": 1.6341168885853297, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12618081441574827, "300–1000 Hz_10ms_envelope_correlation": 0.9861911434578547, "300–1000 Hz_10ms_modulation_mae_db": 0.7324291655371801, "300–1000 Hz_40ms_log_rms_mae_db": 0.9337116307905811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0746203321338002, "300–1000 Hz_40ms_envelope_correlation": 0.9949773322822968, "300–1000 Hz_40ms_modulation_mae_db": 0.8137503728840111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8242119683946627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17648918819664686, "1000–3000 Hz_10ms_envelope_correlation": 0.9743655050787787, "1000–3000 Hz_10ms_modulation_mae_db": 0.6681484087970677, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1483379761702792, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09527412445290233, "1000–3000 Hz_40ms_envelope_correlation": 0.9922206441416518, "1000–3000 Hz_40ms_modulation_mae_db": 0.43125414824049985, "3000–6000 Hz_10ms_log_rms_mae_db": 1.989268600800338, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20869505528824892, "3000–6000 Hz_10ms_envelope_correlation": 0.9701505027265498, "3000–6000 Hz_10ms_modulation_mae_db": 0.6218053769787463, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2718419770475615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12901856995531774, "3000–6000 Hz_40ms_envelope_correlation": 0.9883088709291629, "3000–6000 Hz_40ms_modulation_mae_db": 0.6571005697581679, "mr_stft_db": 5.643340556097637, "band_mae_db": 0.19525272453883047, "rms_gain_error_db": -0.07689257534066933, "output_peak": 0.5928545379151695}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.031510986423613, "全帯域_10ms_rms_envelope_nrmse": 0.06549262015669448, "全帯域_10ms_envelope_correlation": 0.995215002941581, "全帯域_10ms_modulation_mae_db": 0.29103005066412607, "全帯域_40ms_log_rms_mae_db": 0.5298138740601528, "全帯域_40ms_rms_envelope_nrmse": 0.03441721001814604, "全帯域_40ms_envelope_correlation": 0.9985065450881923, "全帯域_40ms_modulation_mae_db": 0.34624841692886915, "300–1000 Hz_10ms_log_rms_mae_db": 1.5569591789271227, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09789313756886699, "300–1000 Hz_10ms_envelope_correlation": 0.9916192285511557, "300–1000 Hz_10ms_modulation_mae_db": 0.3374939543122943, "300–1000 Hz_40ms_log_rms_mae_db": 0.9569547847325206, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05241483709183001, "300–1000 Hz_40ms_envelope_correlation": 0.997410384055378, "300–1000 Hz_40ms_modulation_mae_db": 0.42111358247708264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.824574762296981, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18248467873295757, "1000–3000 Hz_10ms_envelope_correlation": 0.9735355534695, "1000–3000 Hz_10ms_modulation_mae_db": 0.4100011960363638, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1738007352592028, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11060797012029357, "1000–3000 Hz_40ms_envelope_correlation": 0.9908713619300246, "1000–3000 Hz_40ms_modulation_mae_db": 0.5771922472997723, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0285399775351087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21432630736959807, "3000–6000 Hz_10ms_envelope_correlation": 0.9702868827975469, "3000–6000 Hz_10ms_modulation_mae_db": 0.5098661564105891, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3556426981061191, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1307707735262884, "3000–6000 Hz_40ms_envelope_correlation": 0.9884370445170365, "3000–6000 Hz_40ms_modulation_mae_db": 0.5572624596769957, "mr_stft_db": 5.638768254699303, "band_mae_db": 0.22185443560706877, "rms_gain_error_db": -0.010262663271636805, "output_peak": 0.5909919042404562}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.015971635507483, "全帯域_10ms_rms_envelope_nrmse": 0.06577348921491277, "全帯域_10ms_envelope_correlation": 0.9951885429291544, "全帯域_10ms_modulation_mae_db": 0.25375644773890643, "全帯域_40ms_log_rms_mae_db": 0.49536100727471793, "全帯域_40ms_rms_envelope_nrmse": 0.03237419764639029, "全帯域_40ms_envelope_correlation": 0.99870218789703, "全帯域_40ms_modulation_mae_db": 0.32765934232836, "300–1000 Hz_10ms_log_rms_mae_db": 1.68569813607351, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11599034564919866, "300–1000 Hz_10ms_envelope_correlation": 0.9884575168810891, "300–1000 Hz_10ms_modulation_mae_db": 0.7014310289940692, "300–1000 Hz_40ms_log_rms_mae_db": 1.0083829958838553, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06768355393716821, "300–1000 Hz_40ms_envelope_correlation": 0.9960018666519141, "300–1000 Hz_40ms_modulation_mae_db": 0.8606720361534655, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8996743127749558, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17497982285093766, "1000–3000 Hz_10ms_envelope_correlation": 0.9759773818870989, "1000–3000 Hz_10ms_modulation_mae_db": 0.6668877847328593, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2324973917967723, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10985984533142898, "1000–3000 Hz_40ms_envelope_correlation": 0.9911463578085581, "1000–3000 Hz_40ms_modulation_mae_db": 0.8177047477877736, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9794715526218902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20703472063508266, "3000–6000 Hz_10ms_envelope_correlation": 0.97048664818638, "3000–6000 Hz_10ms_modulation_mae_db": 0.8161086306631632, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3096017110965705, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12904957669055866, "3000–6000 Hz_40ms_envelope_correlation": 0.9879558655752146, "3000–6000 Hz_40ms_modulation_mae_db": 0.9880627459357291, "mr_stft_db": 5.656488789623247, "band_mae_db": 0.23387733592281898, "rms_gain_error_db": -0.03676373965210041, "output_peak": 0.5952598804256574}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.055937140574072, "spectral_convergence": 0.22854104353632473, "spectral_convergence_status": "ok", "amplitude_floor": 2.016382236479878e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.414139371236208, "spectral_convergence": 0.214437317309872, "spectral_convergence_status": "ok", "amplitude_floor": 1.164576113083389e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.454873783546572, "spectral_convergence": 0.2274259379494993, "spectral_convergence_status": "ok", "amplitude_floor": 9.482830334434223e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0029188205763026483, "output_power": 0.0030148549092760455, "delta_db": 0.14059017275535873, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003712420548895462, "output_power": 0.003568935259530476, "delta_db": -0.17118497975066566, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009931195946472759, "output_power": 0.0009738821427139563, "delta_db": -0.08495148177858974, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003469926396566426, "output_power": 0.0003051776357004694, "delta_db": -0.5576755862116987, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013263724937728663, "output_power": 0.0001283077365152748, "delta_db": -0.14412663300458, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.438064081917431e-05, "output_power": 5.978273861823063e-05, "delta_db": -0.32179489186530597, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2367206242276998, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0628885522607816, "pitch": null, "output_rms_dbfs": -20.94153387161984, "output_peak": 0.6295245139213238, "reference_rms_dbfs": -20.87864531935906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.641650098452284}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023582673937264833, "log_rms_mae_db": 1.040808169479281, "rms_envelope_nrmse": 0.07268249667908712, "envelope_correlation": 0.9941227294658773, "rms_gain_error_db": -0.06288670948248265, "reference_mean_power": 0.008175179133340451, "output_mean_power": 0.00805765385291147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4118743138857815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017924334098852407, "output_power": 0.0017740572174893005, "reference_active": true, "delta_db": -0.04475407572379923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00033647515979494424, "output_power": 0.0003220208355351632, "reference_active": true, "delta_db": -0.19069035528855954}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.450579873055231e-05, "output_power": 4.53167678577481e-05, "reference_active": true, "delta_db": -0.801837812255961}], "active_band_count": 3, "active_band_mae_db": 0.3457607477561066}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002175646104262597, "log_rms_mae_db": 0.5162385964184124, "rms_envelope_nrmse": 0.037434184832489226, "envelope_correlation": 0.9982852536790309, "rms_gain_error_db": -0.06286103029203689, "reference_mean_power": 0.008216199965234223, "output_mean_power": 0.008098132857739715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.000498094679175e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016116003385924513, "output_power": 0.0016112416851332338, "reference_active": true, "delta_db": -0.0009666078433257176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017199808112094875, "output_power": 0.00016011835031525342, "reference_active": true, "delta_db": -0.3108249481276313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5420165509687798e-06, "output_power": 2.531012474472847e-06, "reference_active": true, "delta_db": -1.4595630056095812}], "active_band_count": 3, "active_band_mae_db": 0.590451520526846}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019245421130728438, "log_rms_mae_db": 1.6452957472986174, "rms_envelope_nrmse": 0.12204585717043047, "envelope_correlation": 0.9869014985905397, "rms_gain_error_db": -0.12338018717241961, "reference_mean_power": 0.0024199693169718968, "output_mean_power": 0.0023521869752949278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3235427499409267e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006913978889533925, "output_power": 0.0006660037514834896, "reference_active": true, "delta_db": -0.1625137369357246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018527392767596858, "output_power": 0.00016151111292822608, "reference_active": true, "delta_db": -0.5961189860003157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604035138941515e-05, "output_power": 2.2229863413616557e-05, "reference_active": true, "delta_db": -0.6137162026662558}], "active_band_count": 3, "active_band_mae_db": 0.45744964186743203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017723694424605393, "log_rms_mae_db": 0.9776799394732884, "rms_envelope_nrmse": 0.06598781191775611, "envelope_correlation": 0.9958518331773104, "rms_gain_error_db": -0.12338607544339741, "reference_mean_power": 0.0024321306730885443, "output_mean_power": 0.002364004491669219}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1677626980235007e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006540469587465989, "output_power": 0.0006318642564445304, "reference_active": true, "delta_db": -0.14985141864531046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.291953991407944e-05, "output_power": 8.103328135613625e-05, "reference_active": true, "delta_db": -0.594436251944623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.049791938534597e-06, "output_power": 1.4818129658582846e-06, "reference_active": true, "delta_db": -1.4091639025797371}], "active_band_count": 3, "active_band_mae_db": 0.7178171910565568}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.001235221787968e-05, "log_rms_mae_db": 1.8608057931996946, "rms_envelope_nrmse": 0.18418114905025335, "envelope_correlation": 0.9727819222894489, "rms_gain_error_db": -0.4383930071573128, "reference_mean_power": 0.0003849600012580844, "output_mean_power": 0.00034799764612364337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5531826233320727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012212996791699553, "output_power": 0.0001051243234560633, "reference_active": true, "delta_db": -0.6511902916656469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.773842605512037e-05, "output_power": 2.2090002804207265e-05, "reference_active": true, "delta_db": -0.9888604352218311}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.9827801273791894e-06, "output_power": 5.558232382784512e-06, "reference_active": true, "delta_db": -0.9909166713060642}], "active_band_count": 3, "active_band_mae_db": 0.8769891327311807}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.393121399816863e-05, "log_rms_mae_db": 1.2051618092869427, "rms_envelope_nrmse": 0.1052912940885069, "envelope_correlation": 0.9916487527347966, "rms_gain_error_db": -0.43839480399878583, "reference_mean_power": 0.0003868956457514561, "output_mean_power": 0.00034974729289227784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2949024312838256e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011569002203634676, "output_power": 9.982014706940215e-05, "reference_active": true, "delta_db": -0.6407769840116833}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4800082503354946e-05, "output_power": 1.128394974260183e-05, "reference_active": true, "delta_db": -1.1780299319436596}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.803144704877755e-07, "output_power": 3.461989409969244e-07, "reference_active": true, "delta_db": -2.2433766503270602}], "active_band_count": 3, "active_band_mae_db": 1.3540611887608012}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.314726805023413e-05, "log_rms_mae_db": 2.0084692344226913, "rms_envelope_nrmse": 0.18872430806588314, "envelope_correlation": 0.9762272701647731, "rms_gain_error_db": -0.4324998746170431, "reference_mean_power": 8.262304385157997e-05, "output_mean_power": 7.479132082309308e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.2225580529948354e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.126031645648209e-05, "output_power": 1.9179892834620625e-05, "reference_active": true, "delta_db": -0.4472354836831436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.562486047803945e-06, "output_power": 8.181253062465553e-06, "reference_active": true, "delta_db": -0.6775098803936699}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.527384735880018e-06, "output_power": 2.649560983018143e-06, "reference_active": true, "delta_db": -1.2427891144029115}], "active_band_count": 3, "active_band_mae_db": 0.7891781594932418}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6400870566820594e-05, "log_rms_mae_db": 1.3270571734532612, "rms_envelope_nrmse": 0.12589242457419003, "envelope_correlation": 0.9895325398550585, "rms_gain_error_db": -0.43250132832111565, "reference_mean_power": 8.30385039555115e-05, "output_mean_power": 7.516737488292701e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.767621782464082e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.197465758003415e-05, "output_power": 1.9853550672566765e-05, "reference_active": true, "delta_db": -0.44083927746453455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.865612603723312e-06, "output_power": 4.556712510109158e-06, "reference_active": true, "delta_db": -1.0966174732210439}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.828054895066371e-07, "output_power": 2.114835601899378e-07, "reference_active": true, "delta_db": -1.2621122238634683}], "active_band_count": 3, "active_band_mae_db": 0.9331896581830156}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.908890489199964, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "57dc86dcd5dc078854792a047ecd4cddf5424637a0e9ffec7370c31492c177c3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9742390041495115, "standard_deviation": 0.002406275676335842, "minimum": 0.9697194243876681, "maximum": 0.9776554106703085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8496687628390098, "standard_deviation": 0.03384745716374682, "minimum": 1.8102479455030445, "maximum": 1.8996743127749558, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6064934463137941, "standard_deviation": 0.19387187416975366, "minimum": 0.3677009859303782, "maximum": 0.8769891327311807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17914144324771217, "standard_deviation": 0.006871506445593124, "minimum": 0.17157786624353283, "maximum": 0.1918740252184955, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9910493205872445, "standard_deviation": 0.0022025342009105034, "minimum": 0.9863525493989007, "maximum": 0.9941796460649573, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.180761471428991, "standard_deviation": 0.03232415718554189, "minimum": 1.13986236583439, "maximum": 1.2324973917967723, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7853733468040027, "standard_deviation": 0.36583007679835267, "minimum": 0.22719717743916332, "maximum": 1.3540611887608012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10678130480816632, "standard_deviation": 0.00844930487645154, "minimum": 0.09527412445290233, "maximum": 0.12357503194033577, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9727464093151836, "standard_deviation": 0.0027276380853208356, "minimum": 0.9701505027265498, "maximum": 0.9762272701647731, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9988466644284206, "standard_deviation": 0.020521178400389403, "minimum": 1.9711069334894216, "maximum": 2.0285399775351087, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5174447390363843, "standard_deviation": 0.25015436488460624, "minimum": 0.17173870429678043, "maximum": 0.8161086306631632, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20027493586552692, "standard_deviation": 0.010471531127645256, "minimum": 0.18872430806588314, "maximum": 0.21432630736959807, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9889030352861705, "standard_deviation": 0.001320501202479136, "minimum": 0.9868630889737081, "maximum": 0.9912145740170794, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3158434136109762, "standard_deviation": 0.02860248752621519, "minimum": 1.2718419770475615, "maximum": 1.3556426981061191, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5927264753124308, "standard_deviation": 0.3026736527877591, "minimum": 0.12066072020494391, "maximum": 0.9880627459357291, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1247579399871797, "standard_deviation": 0.007398774252183999, "minimum": 0.111804795421064, "maximum": 0.13324279719531762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9878187678528313, "standard_deviation": 0.001758149463881768, "minimum": 0.9861911434578547, "maximum": 0.9916192285511557, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6059928299193456, "standard_deviation": 0.054920360011459134, "minimum": 1.5292265632854567, "maximum": 1.68569813607351, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4434136159076648, "standard_deviation": 0.18479825179336068, "minimum": 0.24459114309143928, "maximum": 0.7324291655371801, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11806678651607216, "standard_deviation": 0.009222808025680176, "minimum": 0.09789313756886699, "maximum": 0.1273768517712352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958207437747859, "standard_deviation": 0.0007944367486019596, "minimum": 0.9948208503033881, "maximum": 0.997410384055378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9609806300062951, "standard_deviation": 0.03441300213962814, "minimum": 0.8934717928606277, "maximum": 1.0083829958838553, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5437197525776395, "standard_deviation": 0.25199083102259445, "minimum": 0.18027100474120916, "maximum": 0.8606720361534655, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06731322465841527, "standard_deviation": 0.007617718158200377, "minimum": 0.05241483709183001, "maximum": 0.07819363459808192, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20318796169177572, "standard_deviation": 0.04308745334173743, "minimum": 0.14448817075714995, "maximum": 0.2659443549061494, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.639901191872164, "standard_deviation": 0.009271615511378095, "minimum": 5.622880344204621, "maximum": 5.656488789623247, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6007236932669164, "standard_deviation": 0.019113394944163625, "minimum": 0.5713472758035723, "maximum": 0.6295245139213238, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04966097445097959, "standard_deviation": 0.02366124629167996, "minimum": -0.07689257534066933, "maximum": -0.010262663271636805, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947503039020316, "standard_deviation": 0.0004714198755652388, "minimum": 0.9941227294658773, "maximum": 0.9953675689658804, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.034759424867661, "standard_deviation": 0.015377716753627455, "minimum": 1.015971635507483, "maximum": 1.0598645206441974, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.25690498949731944, "standard_deviation": 0.06161897782976757, "minimum": 0.14639030938103348, "maximum": 0.3457607477561066, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06867403409758621, "standard_deviation": 0.003165768441400233, "minimum": 0.0644560640287867, "maximum": 0.07268249667908712, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985031471239518, "standard_deviation": 0.00022454606700581472, "minimum": 0.998184778116787, "maximum": 0.9988809363916349, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5229573245424907, "standard_deviation": 0.016661554228371685, "minimum": 0.49536100727471793, "maximum": 0.543053098137875, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3433003908321911, "standard_deviation": 0.12129935361387141, "minimum": 0.19296970345327016, "maximum": 0.590451520526846, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03486900173813278, "standard_deviation": 0.002836905974804881, "minimum": 0.029913799870148843, "maximum": 0.038810968257152204, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0004", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.062346935272217, "predicted_mos": 3.062346935272217, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a18132775a21bf6d656c36b9b89d5f564477233479c8cb4f52fda127018032a5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8212736998684704, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9793757752517841, "全帯域_10ms_rms_envelope_nrmse": 0.06500225338486938, "全帯域_10ms_envelope_correlation": 0.9953105225297812, "全帯域_10ms_modulation_mae_db": 0.151396746146276, "全帯域_40ms_log_rms_mae_db": 0.5016829215437346, "全帯域_40ms_rms_envelope_nrmse": 0.034039980121859915, "全帯域_40ms_envelope_correlation": 0.9985768783446667, "全帯域_40ms_modulation_mae_db": 0.16361975137997806, "300–1000 Hz_10ms_log_rms_mae_db": 1.632003612858974, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10953542872743129, "300–1000 Hz_10ms_envelope_correlation": 0.9896230287551361, "300–1000 Hz_10ms_modulation_mae_db": 0.47609254762280956, "300–1000 Hz_40ms_log_rms_mae_db": 0.9152460179466254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062071687324640495, "300–1000 Hz_40ms_envelope_correlation": 0.9965335464277756, "300–1000 Hz_40ms_modulation_mae_db": 0.31137573617194336, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0244224321199304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18623670530524097, "1000–3000 Hz_10ms_envelope_correlation": 0.9717630363443269, "1000–3000 Hz_10ms_modulation_mae_db": 0.8489821189450585, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2453156843110198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11230571090266488, "1000–3000 Hz_40ms_envelope_correlation": 0.9895901230640236, "1000–3000 Hz_40ms_modulation_mae_db": 1.332123137497203, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2079705726116936, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1905960904799329, "3000–6000 Hz_10ms_envelope_correlation": 0.9750494952131868, "3000–6000 Hz_10ms_modulation_mae_db": 0.9207834526677239, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3802969777321885, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12173024088755098, "3000–6000 Hz_40ms_envelope_correlation": 0.9890287030242705, "3000–6000 Hz_40ms_modulation_mae_db": 1.0640465382007045, "mr_stft_db": 5.775618890133316, "band_mae_db": 0.18647894759444458, "rms_gain_error_db": -0.04676430112593434, "output_peak": 0.5889633010791506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0012871457625014, "全帯域_10ms_rms_envelope_nrmse": 0.06902949879125392, "全帯域_10ms_envelope_correlation": 0.9946881029412937, "全帯域_10ms_modulation_mae_db": 0.30547517516974737, "全帯域_40ms_log_rms_mae_db": 0.5068472757704983, "全帯域_40ms_rms_envelope_nrmse": 0.036391593079656046, "全帯域_40ms_envelope_correlation": 0.9983529624680577, "全帯域_40ms_modulation_mae_db": 0.528435804797889, "300–1000 Hz_10ms_log_rms_mae_db": 1.620377120890823, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11433195670118267, "300–1000 Hz_10ms_envelope_correlation": 0.9887493749166797, "300–1000 Hz_10ms_modulation_mae_db": 0.9684289426393047, "300–1000 Hz_40ms_log_rms_mae_db": 1.0173528968301666, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07099055040741788, "300–1000 Hz_40ms_envelope_correlation": 0.9954721050595883, "300–1000 Hz_40ms_modulation_mae_db": 1.2028964626986154, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9895343018103366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20016431485280278, "1000–3000 Hz_10ms_envelope_correlation": 0.9675253997647537, "1000–3000 Hz_10ms_modulation_mae_db": 1.0304194067589847, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2319727973781545, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11512276777243467, "1000–3000 Hz_40ms_envelope_correlation": 0.9897299817212589, "1000–3000 Hz_40ms_modulation_mae_db": 0.8505971037579249, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1676968850054137, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19830553256693748, "3000–6000 Hz_10ms_envelope_correlation": 0.972964853051617, "3000–6000 Hz_10ms_modulation_mae_db": 0.6789047232334394, "3000–6000 Hz_40ms_log_rms_mae_db": 1.370797909834583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12101121182880692, "3000–6000 Hz_40ms_envelope_correlation": 0.9891537682910382, "3000–6000 Hz_40ms_modulation_mae_db": 0.6835013956225597, "mr_stft_db": 5.7604453312082535, "band_mae_db": 0.1805286007899861, "rms_gain_error_db": -0.04410903896990441, "output_peak": 0.6226656934087819}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.029259567256702, "全帯域_10ms_rms_envelope_nrmse": 0.06996869179423848, "全帯域_10ms_envelope_correlation": 0.9945850025718729, "全帯域_10ms_modulation_mae_db": 0.1754989312360062, "全帯域_40ms_log_rms_mae_db": 0.516749358816109, "全帯域_40ms_rms_envelope_nrmse": 0.03666260305366978, "全帯域_40ms_envelope_correlation": 0.998393846841706, "全帯域_40ms_modulation_mae_db": 0.3364141411658237, "300–1000 Hz_10ms_log_rms_mae_db": 1.6754800829464755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1164791747135964, "300–1000 Hz_10ms_envelope_correlation": 0.9882316492092885, "300–1000 Hz_10ms_modulation_mae_db": 0.5791539753106986, "300–1000 Hz_40ms_log_rms_mae_db": 0.9756650897932522, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06747654089610883, "300–1000 Hz_40ms_envelope_correlation": 0.9958656918136561, "300–1000 Hz_40ms_modulation_mae_db": 0.7540850646035892, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0160943921220196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2114989400552804, "1000–3000 Hz_10ms_envelope_correlation": 0.9630624004360168, "1000–3000 Hz_10ms_modulation_mae_db": 0.753081195031159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2337841068169226, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14327463357058223, "1000–3000 Hz_40ms_envelope_correlation": 0.9813461753445057, "1000–3000 Hz_40ms_modulation_mae_db": 0.3073615931290515, "3000–6000 Hz_10ms_log_rms_mae_db": 2.259815204566916, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2126721128275525, "3000–6000 Hz_10ms_envelope_correlation": 0.9691984295449431, "3000–6000 Hz_10ms_modulation_mae_db": 0.4611553006638112, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4456991347069614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13367282904170016, "3000–6000 Hz_40ms_envelope_correlation": 0.9867656066549045, "3000–6000 Hz_40ms_modulation_mae_db": 0.2402649855383848, "mr_stft_db": 5.782962942701812, "band_mae_db": 0.13703839160541392, "rms_gain_error_db": -0.07817336113998424, "output_peak": 0.5907765789317241}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.028154950837362, "全帯域_10ms_rms_envelope_nrmse": 0.06631529850319355, "全帯域_10ms_envelope_correlation": 0.9951034321876941, "全帯域_10ms_modulation_mae_db": 0.11518180268795973, "全帯域_40ms_log_rms_mae_db": 0.48301639425899273, "全帯域_40ms_rms_envelope_nrmse": 0.03325715232507481, "全帯域_40ms_envelope_correlation": 0.9986361352015971, "全帯域_40ms_modulation_mae_db": 0.4012568530370384, "300–1000 Hz_10ms_log_rms_mae_db": 1.6482814068705098, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11914639363596818, "300–1000 Hz_10ms_envelope_correlation": 0.987584481701098, "300–1000 Hz_10ms_modulation_mae_db": 0.11309444036034867, "300–1000 Hz_40ms_log_rms_mae_db": 1.0064339162826799, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07672954637070253, "300–1000 Hz_40ms_envelope_correlation": 0.994411205302114, "300–1000 Hz_40ms_modulation_mae_db": 0.1492839635927635, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0131396255217098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21383068946162817, "1000–3000 Hz_10ms_envelope_correlation": 0.9619714551315545, "1000–3000 Hz_10ms_modulation_mae_db": 0.47418931610093756, "1000–3000 Hz_40ms_log_rms_mae_db": 1.303486585578815, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1303757130775205, "1000–3000 Hz_40ms_envelope_correlation": 0.9846767161039847, "1000–3000 Hz_40ms_modulation_mae_db": 0.42765747541690025, "3000–6000 Hz_10ms_log_rms_mae_db": 2.242722722683455, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20664926240746742, "3000–6000 Hz_10ms_envelope_correlation": 0.9707922867989802, "3000–6000 Hz_10ms_modulation_mae_db": 0.5856149787933584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4204272619007408, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12360337014564426, "3000–6000 Hz_40ms_envelope_correlation": 0.9886538268129154, "3000–6000 Hz_40ms_modulation_mae_db": 0.6723328458015546, "mr_stft_db": 5.768463157869846, "band_mae_db": 0.1885826513740505, "rms_gain_error_db": -0.04804867277288126, "output_peak": 0.6260688198638393}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0083826989011109, "全帯域_10ms_rms_envelope_nrmse": 0.07083648337394656, "全帯域_10ms_envelope_correlation": 0.9944794125219365, "全帯域_10ms_modulation_mae_db": 0.24090763331638906, "全帯域_40ms_log_rms_mae_db": 0.5158682391845152, "全帯域_40ms_rms_envelope_nrmse": 0.03844343214399066, "全帯域_40ms_envelope_correlation": 0.9982586683645323, "全帯域_40ms_modulation_mae_db": 0.5184941592902822, "300–1000 Hz_10ms_log_rms_mae_db": 1.7047344765245402, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12066343175956737, "300–1000 Hz_10ms_envelope_correlation": 0.9873828028997974, "300–1000 Hz_10ms_modulation_mae_db": 0.5984551274236871, "300–1000 Hz_40ms_log_rms_mae_db": 1.0094072255781994, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07477094727878257, "300–1000 Hz_40ms_envelope_correlation": 0.9948651029103689, "300–1000 Hz_40ms_modulation_mae_db": 0.6328759583929777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9924379274808985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17849577158659152, "1000–3000 Hz_10ms_envelope_correlation": 0.974062547744289, "1000–3000 Hz_10ms_modulation_mae_db": 0.6621833133947873, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2306920142453108, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10279181276591566, "1000–3000 Hz_40ms_envelope_correlation": 0.9913452843129575, "1000–3000 Hz_40ms_modulation_mae_db": 1.0369726280040508, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1777211169274167, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18370880215408814, "3000–6000 Hz_10ms_envelope_correlation": 0.9768456370374436, "3000–6000 Hz_10ms_modulation_mae_db": 0.5258065886330345, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4013081972053794, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11760795340044135, "3000–6000 Hz_40ms_envelope_correlation": 0.9897060432112125, "3000–6000 Hz_40ms_modulation_mae_db": 0.4098637367572771, "mr_stft_db": 5.771130125827626, "band_mae_db": 0.20789498837170436, "rms_gain_error_db": -0.08898322439220108, "output_peak": 0.5783485987519573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0209548996143083, "全帯域_10ms_rms_envelope_nrmse": 0.06722797101143228, "全帯域_10ms_envelope_correlation": 0.994964100209595, "全帯域_10ms_modulation_mae_db": 0.15656464067435152, "全帯域_40ms_log_rms_mae_db": 0.508800968186498, "全帯域_40ms_rms_envelope_nrmse": 0.03415755526335722, "全帯域_40ms_envelope_correlation": 0.9985518996420372, "全帯域_40ms_modulation_mae_db": 0.21812848967302054, "300–1000 Hz_10ms_log_rms_mae_db": 1.6832292394698805, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12549680517110084, "300–1000 Hz_10ms_envelope_correlation": 0.9861955949776589, "300–1000 Hz_10ms_modulation_mae_db": 0.42828483136498274, "300–1000 Hz_40ms_log_rms_mae_db": 1.021930705575107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0719912346177384, "300–1000 Hz_40ms_envelope_correlation": 0.9951351763000855, "300–1000 Hz_40ms_modulation_mae_db": 0.26008929376064627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9244179331419784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16878261529648636, "1000–3000 Hz_10ms_envelope_correlation": 0.9766235989126076, "1000–3000 Hz_10ms_modulation_mae_db": 0.3308228164506033, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1632617106496783, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09090875827895777, "1000–3000 Hz_40ms_envelope_correlation": 0.9929814078948934, "1000–3000 Hz_40ms_modulation_mae_db": 0.1100199190260821, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2649192223080568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2187515325791439, "3000–6000 Hz_10ms_envelope_correlation": 0.967153130163833, "3000–6000 Hz_10ms_modulation_mae_db": 0.36838093251315057, "3000–6000 Hz_40ms_log_rms_mae_db": 1.441381098752905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14073862802730824, "3000–6000 Hz_40ms_envelope_correlation": 0.9852306783442927, "3000–6000 Hz_40ms_modulation_mae_db": 0.21350459167559618, "mr_stft_db": 5.760145808562459, "band_mae_db": 0.15596550569542686, "rms_gain_error_db": -0.04112784338406409, "output_peak": 0.6182814189039152}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0239131226131832, "全帯域_10ms_rms_envelope_nrmse": 0.07105517757901877, "全帯域_10ms_envelope_correlation": 0.9943916303312721, "全帯域_10ms_modulation_mae_db": 0.20396809844066277, "全帯域_40ms_log_rms_mae_db": 0.5283823155259955, "全帯域_40ms_rms_envelope_nrmse": 0.03667108046940683, "全帯域_40ms_envelope_correlation": 0.9983468704471629, "全帯域_40ms_modulation_mae_db": 0.41238995875321965, "300–1000 Hz_10ms_log_rms_mae_db": 1.7469062470921315, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12223163591710696, "300–1000 Hz_10ms_envelope_correlation": 0.9869811148032618, "300–1000 Hz_10ms_modulation_mae_db": 0.31529158593922174, "300–1000 Hz_40ms_log_rms_mae_db": 1.0459417633858454, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07173394077488561, "300–1000 Hz_40ms_envelope_correlation": 0.9952134214955295, "300–1000 Hz_40ms_modulation_mae_db": 0.5904008151004078, "1000–3000 Hz_10ms_log_rms_mae_db": 1.962158308781878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1781768364148203, "1000–3000 Hz_10ms_envelope_correlation": 0.9741956755689397, "1000–3000 Hz_10ms_modulation_mae_db": 0.6328034418262224, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2009403944695458, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10808190850125407, "1000–3000 Hz_40ms_envelope_correlation": 0.9903074328010965, "1000–3000 Hz_40ms_modulation_mae_db": 0.4027006091084318, "3000–6000 Hz_10ms_log_rms_mae_db": 2.260808416929483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2229522115500732, "3000–6000 Hz_10ms_envelope_correlation": 0.9657107417998578, "3000–6000 Hz_10ms_modulation_mae_db": 0.7888754923385287, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4217969020317445, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13841962214850861, "3000–6000 Hz_40ms_envelope_correlation": 0.9859257235569568, "3000–6000 Hz_40ms_modulation_mae_db": 0.6989986850386002, "mr_stft_db": 5.772274379058733, "band_mae_db": 0.1786651695512135, "rms_gain_error_db": -0.05382767493537401, "output_peak": 0.6039420488189843}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0591438212638777, "全帯域_10ms_rms_envelope_nrmse": 0.0709596968787572, "全帯域_10ms_envelope_correlation": 0.9943834144082008, "全帯域_10ms_modulation_mae_db": 0.15597740511115643, "全帯域_40ms_log_rms_mae_db": 0.5400343427136125, "全帯域_40ms_rms_envelope_nrmse": 0.0357739309606383, "全帯域_40ms_envelope_correlation": 0.9983884013590453, "全帯域_40ms_modulation_mae_db": 0.06912166527107726, "300–1000 Hz_10ms_log_rms_mae_db": 1.6766317233613908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11197934638286032, "300–1000 Hz_10ms_envelope_correlation": 0.9889657146018284, "300–1000 Hz_10ms_modulation_mae_db": 0.1529534156423, "300–1000 Hz_40ms_log_rms_mae_db": 0.9544399396715743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06445513501225715, "300–1000 Hz_40ms_envelope_correlation": 0.995972592344014, "300–1000 Hz_40ms_modulation_mae_db": 0.18322716905417505, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0165447175458637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.201568901838518, "1000–3000 Hz_10ms_envelope_correlation": 0.9665767023925794, "1000–3000 Hz_10ms_modulation_mae_db": 0.6452308726102232, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2496160891715242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12049995306165731, "1000–3000 Hz_40ms_envelope_correlation": 0.9876740606314274, "1000–3000 Hz_40ms_modulation_mae_db": 0.6940908534473796, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2328596455880176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21248617770146153, "3000–6000 Hz_10ms_envelope_correlation": 0.9695568384696192, "3000–6000 Hz_10ms_modulation_mae_db": 0.30470198717292307, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4058104506622966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13010755890815498, "3000–6000 Hz_40ms_envelope_correlation": 0.9876714019295851, "3000–6000 Hz_40ms_modulation_mae_db": 0.3055077117818205, "mr_stft_db": 5.757352414145067, "band_mae_db": 0.19121084458087212, "rms_gain_error_db": -0.007210620759964143, "output_peak": 0.5947846462160018}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.291826743189389, "spectral_convergence": 0.22249741225105793, "spectral_convergence_status": "ok", "amplitude_floor": 2.016382236479878e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.549341600279898, "spectral_convergence": 0.2115858897272592, "spectral_convergence_status": "ok", "amplitude_floor": 1.164576113083389e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.485688326930662, "spectral_convergence": 0.22265879068341507, "spectral_convergence_status": "ok", "amplitude_floor": 9.482830334434223e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0029188205763026483, "output_power": 0.0029782049526686704, "delta_db": 0.08747182350457114, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003712420548895462, "output_power": 0.003681625783624045, "delta_db": -0.03617525061360835, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009931195946472759, "output_power": 0.0009082751290423498, "delta_db": -0.38784128551200037, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003469926396566426, "output_power": 0.00031559532500197835, "delta_db": -0.4118970145014478, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013263724937728663, "output_power": 0.0001307827004422478, "delta_db": -0.06115206360556631, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.438064081917431e-05, "output_power": 6.640318642708663e-05, "delta_db": 0.13433624782947345, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18647894759444458, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04676430112593434, "pitch": null, "output_rms_dbfs": -20.92540962048499, "output_peak": 0.5889633010791506, "reference_rms_dbfs": -20.87864531935906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.775618890133316}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023582673937264833, "log_rms_mae_db": 0.9793757752517841, "rms_envelope_nrmse": 0.06500225338486938, "envelope_correlation": 0.9953105225297812, "rms_gain_error_db": -0.04676292670656092, "reference_mean_power": 0.008175179133340451, "output_mean_power": 0.008087624607155116}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4118743138857815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017924334098852407, "output_power": 0.001735894302196127, "reference_active": true, "delta_db": -0.1391975265077223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00033647515979494424, "output_power": 0.00032566284284059083, "reference_active": true, "delta_db": -0.14184798362335854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.450579873055231e-05, "output_power": 5.672273734740408e-05, "reference_active": true, "delta_db": 0.17314472830774721}], "active_band_count": 3, "active_band_mae_db": 0.151396746146276}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002175646104262597, "log_rms_mae_db": 0.5016829215437346, "rms_envelope_nrmse": 0.034039980121859915, "envelope_correlation": 0.9985768783446667, "rms_gain_error_db": -0.046749529042424305, "reference_mean_power": 0.008216199965234223, "output_mean_power": 0.008128231189134538}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.000498094679175e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016116003385924513, "output_power": 0.0015760826283394303, "reference_active": true, "delta_db": -0.09678367713404058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017199808112094875, "output_power": 0.00016470233429280182, "reference_active": true, "delta_db": -0.18823847391437717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5420165509687798e-06, "output_power": 3.3780562415423987e-06, "reference_active": true, "delta_db": -0.2058371030915164}], "active_band_count": 3, "active_band_mae_db": 0.16361975137997806}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019245421130728438, "log_rms_mae_db": 1.632003612858974, "rms_envelope_nrmse": 0.10953542872743129, "envelope_correlation": 0.9896230287551361, "rms_gain_error_db": -0.16613066958167746, "reference_mean_power": 0.0024199693169718968, "output_mean_power": 0.0023291464286512476}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3235427499409267e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006913978889533925, "output_power": 0.0006250200338406174, "reference_active": true, "delta_db": -0.43834111116912133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018527392767596858, "output_power": 0.0001676276848375356, "reference_active": true, "delta_db": -0.4346856147946344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604035138941515e-05, "output_power": 2.9096014618072375e-05, "reference_active": true, "delta_db": 0.5552509169046731}], "active_band_count": 3, "active_band_mae_db": 0.47609254762280956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017723694424605393, "log_rms_mae_db": 0.9152460179466254, "rms_envelope_nrmse": 0.062071687324640495, "envelope_correlation": 0.9965335464277756, "rms_gain_error_db": -0.16613580139569928, "reference_mean_power": 0.0024321306730885443, "output_mean_power": 0.002340848595785993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1677626980235007e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006540469587465989, "output_power": 0.0005950929677000876, "reference_active": true, "delta_db": -0.41024112401399854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.291953991407944e-05, "output_power": 8.466289311258967e-05, "reference_active": true, "delta_db": -0.40413945583393723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.049791938534597e-06, "output_power": 2.107096587073536e-06, "reference_active": true, "delta_db": 0.11974662866789429}], "active_band_count": 3, "active_band_mae_db": 0.31137573617194336}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.001235221787968e-05, "log_rms_mae_db": 2.0244224321199304, "rms_envelope_nrmse": 0.18623670530524097, "envelope_correlation": 0.9717630363443269, "rms_gain_error_db": -0.36669665998409373, "reference_mean_power": 0.0003849600012580844, "output_mean_power": 0.00035379031593157163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5531826233320727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012212996791699553, "output_power": 0.00011192104186681694, "reference_active": true, "delta_db": -0.37910498685819477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.773842605512037e-05, "output_power": 1.9693541602540373e-05, "reference_active": true, "delta_db": -1.487579896956651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.9827801273791894e-06, "output_power": 5.970383384275802e-06, "reference_active": true, "delta_db": -0.6802614730203296}], "active_band_count": 3, "active_band_mae_db": 0.8489821189450585}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.393121399816863e-05, "log_rms_mae_db": 1.2453156843110198, "rms_envelope_nrmse": 0.11230571090266488, "envelope_correlation": 0.9895901230640236, "rms_gain_error_db": -0.3666992096370748, "reference_mean_power": 0.0003868956457514561, "output_mean_power": 0.00035556902518773495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2949024312838256e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011569002203634676, "output_power": 0.00010508239954140553, "reference_active": true, "delta_db": -0.4176592251787203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4800082503354946e-05, "output_power": 1.0151751852044034e-05, "reference_active": true, "delta_db": -1.6372314300014668}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.803144704877755e-07, "output_power": 3.711209954067785e-07, "reference_active": true, "delta_db": -1.9414787573114216}], "active_band_count": 3, "active_band_mae_db": 1.332123137497203}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.314726805023413e-05, "log_rms_mae_db": 2.2079705726116936, "rms_envelope_nrmse": 0.1905960904799329, "envelope_correlation": 0.9750494952131868, "rms_gain_error_db": -0.1317478574192738, "reference_mean_power": 8.262304385157997e-05, "output_mean_power": 8.015422229816233e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.2225580529948354e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.126031645648209e-05, "output_power": 1.999453400860767e-05, "reference_active": true, "delta_db": -0.26658437694031734}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.562486047803945e-06, "output_power": 8.912612489562078e-06, "reference_active": true, "delta_db": -0.305657902077298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.527384735880018e-06, "output_power": 2.1303061608330807e-06, "reference_active": true, "delta_db": -2.1901080789855563}], "active_band_count": 3, "active_band_mae_db": 0.9207834526677239}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6400870566820594e-05, "log_rms_mae_db": 1.3802969777321885, "rms_envelope_nrmse": 0.12173024088755098, "envelope_correlation": 0.9890287030242705, "rms_gain_error_db": -0.13175034493827342, "reference_mean_power": 8.30385039555115e-05, "output_mean_power": 8.055722208692752e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.767621782464082e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.197465758003415e-05, "output_power": 2.0531998025584912e-05, "reference_active": true, "delta_db": -0.2949090258171719}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.865612603723312e-06, "output_power": 4.810174891470217e-06, "reference_active": true, "delta_db": -0.8615250945011808}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.828054895066371e-07, "output_power": 1.7697718617781085e-07, "reference_active": true, "delta_db": -2.035705494283761}], "active_band_count": 3, "active_band_mae_db": 1.0640465382007045}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9084358271243488, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "71d94f27bfef7868c8f9691d04879963d73ae621f611cbfdc7ce445c76e6a2da", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9694726020368835, "standard_deviation": 0.005468005535019409, "minimum": 0.9619714551315545, "maximum": 0.9766235989126076, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9923437048155768, "standard_deviation": 0.03410963299736468, "minimum": 1.9244179331419784, "maximum": 2.0244224321199304, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.672214060139747, "standard_deviation": 0.21522121380788573, "minimum": 0.3308228164506033, "maximum": 1.0304194067589847, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19234434685142107, "standard_deviation": 0.016732199675140554, "minimum": 0.16878261529648636, "maximum": 0.21383068946162817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9884563977342684, "standard_deviation": 0.0037911244041243884, "minimum": 0.9813461753445057, "maximum": 0.9929814078948934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2323836728276214, "standard_deviation": 0.040162766068204236, "minimum": 1.1632617106496783, "maximum": 1.303486585578815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6451904149233779, "standard_deviation": 0.4099844399039151, "minimum": 0.1100199190260821, "maximum": 1.332123137497203, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11542015724137339, "standard_deviation": 0.01625364443470479, "minimum": 0.09090875827895777, "maximum": 0.14327463357058223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9709089265099351, "standard_deviation": 0.0038267918725411643, "minimum": 0.9657107417998578, "maximum": 0.9768456370374436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2268142233275565, "standard_deviation": 0.038259199359454196, "minimum": 2.1676968850054137, "maximum": 2.2649192223080568, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5792779320019963, "standard_deviation": 0.2094906912713444, "minimum": 0.30470198717292307, "maximum": 0.9207834526677239, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20576521528333214, "standard_deviation": 0.013783685802263764, "minimum": 0.18370880215408814, "maximum": 0.2229522115500732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.987766968978147, "standard_deviation": 0.001643016619402177, "minimum": 0.9852306783442927, "maximum": 0.9897060432112125, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4109397416033498, "standard_deviation": 0.026758722253456105, "minimum": 1.370797909834583, "maximum": 1.4456991347069614, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5360025613020623, "standard_deviation": 0.2942895957906472, "minimum": 0.21350459167559618, "maximum": 1.0640465382007045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12836142679851442, "standard_deviation": 0.008634622623994602, "minimum": 0.11760795340044135, "maximum": 0.14073862802730824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9879642202330936, "standard_deviation": 0.001135114963739663, "minimum": 0.9861955949776589, "maximum": 0.9896230287551361, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.673455488751841, "standard_deviation": 0.04079987425801666, "minimum": 1.620377120890823, "maximum": 1.7469062470921315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4539693582879191, "standard_deviation": 0.2747474040238865, "minimum": 0.11309444036034867, "maximum": 0.9684289426393047, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11748302162610176, "standard_deviation": 0.0053989876339651825, "minimum": 0.10953542872743129, "maximum": 0.12549680517110084, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954336052066415, "standard_deviation": 0.0006755522612112681, "minimum": 0.994411205302114, "maximum": 0.9965335464277756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9933021943829312, "standard_deviation": 0.04228284909876887, "minimum": 0.9152460179466254, "maximum": 1.0459417633858454, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5105293079218898, "standard_deviation": 0.35857796102054706, "minimum": 0.1492839635927635, "maximum": 1.2028964626986154, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07002744783531668, "standard_deviation": 0.005013794922952221, "minimum": 0.062071687324640495, "maximum": 0.07672954637070253, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.178295637445389, "standard_deviation": 0.022108922467514815, "minimum": 0.13703839160541392, "maximum": 0.20789498837170436, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.768549131188388, "standard_deviation": 0.008782610481923683, "minimum": 5.757352414145067, "maximum": 5.782962942701812, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6029788882467944, "standard_deviation": 0.017626847063090083, "minimum": 0.5783485987519573, "maximum": 0.6260688198638393, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05103059218503845, "standard_deviation": 0.024721913879641864, "minimum": -0.08898322439220108, "maximum": -0.007210620759964143, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947382022127058, "standard_deviation": 0.00034865292016718157, "minimum": 0.9943834144082008, "maximum": 0.9953105225297812, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0188089976876036, "standard_deviation": 0.02337525632190527, "minimum": 0.9793757752517841, "maximum": 1.0591438212638777, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.18812130409781863, "standard_deviation": 0.06055303468470186, "minimum": 0.11518180268795973, "maximum": 0.30547517516974737, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06879938391458876, "standard_deviation": 0.0023416655929324667, "minimum": 0.06500225338486938, "maximum": 0.07105517757901877, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984382078336007, "standard_deviation": 0.000132892477975671, "minimum": 0.9982586683645323, "maximum": 0.9986361352015971, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5126727269999944, "standard_deviation": 0.017202490289615152, "minimum": 0.48301639425899273, "maximum": 0.5400343427136125, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3309826029210411, "standard_deviation": 0.16693069808583216, "minimum": 0.06912166527107726, "maximum": 0.528435804797889, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.035674665927206695, "standard_deviation": 0.0017311004189242607, "minimum": 0.03325715232507481, "maximum": 0.03844343214399066, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0004", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.0066463947296143, "predicted_mos": 3.0066463947296143, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a24fea48bcd7e25c3b21632bd5bdd9902d032173544a666a856ccd93f05e2684", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.651271200273186, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0058837201619695, "全帯域_10ms_rms_envelope_nrmse": 0.06681319610548346, "全帯域_10ms_envelope_correlation": 0.9950182686886759, "全帯域_10ms_modulation_mae_db": 0.5121984374558525, "全帯域_40ms_log_rms_mae_db": 0.5116137071541471, "全帯域_40ms_rms_envelope_nrmse": 0.03326013021318172, "全帯域_40ms_envelope_correlation": 0.9986103582462428, "全帯域_40ms_modulation_mae_db": 0.6387242264044469, "300–1000 Hz_10ms_log_rms_mae_db": 1.6127412816423992, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1265350337663396, "300–1000 Hz_10ms_envelope_correlation": 0.9860140718806885, "300–1000 Hz_10ms_modulation_mae_db": 1.0848402550341427, "300–1000 Hz_40ms_log_rms_mae_db": 0.9451688041476286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07900230006715965, "300–1000 Hz_40ms_envelope_correlation": 0.9941470416978466, "300–1000 Hz_40ms_modulation_mae_db": 1.1361490603834705, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8997665850133794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1784848134712627, "1000–3000 Hz_10ms_envelope_correlation": 0.9737409271219271, "1000–3000 Hz_10ms_modulation_mae_db": 0.5560932312780458, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1584164067282394, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09567660509519495, "1000–3000 Hz_40ms_envelope_correlation": 0.9916267768345506, "1000–3000 Hz_40ms_modulation_mae_db": 0.4602682529008854, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1852143931168677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2005448877101278, "3000–6000 Hz_10ms_envelope_correlation": 0.9737469317653216, "3000–6000 Hz_10ms_modulation_mae_db": 0.676061226329073, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3714881758713178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11944070222787828, "3000–6000 Hz_40ms_envelope_correlation": 0.9902220762744715, "3000–6000 Hz_40ms_modulation_mae_db": 0.5466296405886513, "mr_stft_db": 5.743725209393294, "band_mae_db": 0.1255157193275867, "rms_gain_error_db": -0.024337697091652975, "output_peak": 0.5783774518923506}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9966416708032514, "全帯域_10ms_rms_envelope_nrmse": 0.06758236878783448, "全帯域_10ms_envelope_correlation": 0.9949177796954753, "全帯域_10ms_modulation_mae_db": 0.19808391985332985, "全帯域_40ms_log_rms_mae_db": 0.4940001897199386, "全帯域_40ms_rms_envelope_nrmse": 0.03273940574653788, "全帯域_40ms_envelope_correlation": 0.9986530495547818, "全帯域_40ms_modulation_mae_db": 0.186814541473598, "300–1000 Hz_10ms_log_rms_mae_db": 1.6649766262534373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12658862455931208, "300–1000 Hz_10ms_envelope_correlation": 0.9858650464344424, "300–1000 Hz_10ms_modulation_mae_db": 0.3333888561942269, "300–1000 Hz_40ms_log_rms_mae_db": 1.0095802827983442, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06772895193667153, "300–1000 Hz_40ms_envelope_correlation": 0.9954902070201871, "300–1000 Hz_40ms_modulation_mae_db": 0.5510906387296922, "1000–3000 Hz_10ms_log_rms_mae_db": 1.953586988121975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18196567753664616, "1000–3000 Hz_10ms_envelope_correlation": 0.9726793038444981, "1000–3000 Hz_10ms_modulation_mae_db": 0.39039977016096095, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1877638003967241, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10722905359707059, "1000–3000 Hz_40ms_envelope_correlation": 0.9895749471917892, "1000–3000 Hz_40ms_modulation_mae_db": 0.7622837706419571, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1922083520438833, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2034282673037175, "3000–6000 Hz_10ms_envelope_correlation": 0.9716416847782348, "3000–6000 Hz_10ms_modulation_mae_db": 0.7188169703228081, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3827144127587385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11944232502518584, "3000–6000 Hz_40ms_envelope_correlation": 0.9893849916449156, "3000–6000 Hz_40ms_modulation_mae_db": 0.7286954169469841, "mr_stft_db": 5.7526956673360905, "band_mae_db": 0.11330889287843661, "rms_gain_error_db": 0.012924359854907547, "output_peak": 0.6115365909799293}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0057375752550985, "全帯域_10ms_rms_envelope_nrmse": 0.06859701873880059, "全帯域_10ms_envelope_correlation": 0.9947905285401643, "全帯域_10ms_modulation_mae_db": 0.312465124543674, "全帯域_40ms_log_rms_mae_db": 0.502666604976441, "全帯域_40ms_rms_envelope_nrmse": 0.03591059428033917, "全帯域_40ms_envelope_correlation": 0.9984108579380698, "全帯域_40ms_modulation_mae_db": 0.1275006366623377, "300–1000 Hz_10ms_log_rms_mae_db": 1.624504958397257, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11643053140964445, "300–1000 Hz_10ms_envelope_correlation": 0.9882205385938092, "300–1000 Hz_10ms_modulation_mae_db": 0.705469951906419, "300–1000 Hz_40ms_log_rms_mae_db": 0.9501048259222487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06288901127209438, "300–1000 Hz_40ms_envelope_correlation": 0.9964113951823246, "300–1000 Hz_40ms_modulation_mae_db": 0.46579128075443404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8973088102713964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16892243482877325, "1000–3000 Hz_10ms_envelope_correlation": 0.9764188333021089, "1000–3000 Hz_10ms_modulation_mae_db": 0.43378870490521737, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1138057344366468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09209147031888217, "1000–3000 Hz_40ms_envelope_correlation": 0.9923220251784755, "1000–3000 Hz_40ms_modulation_mae_db": 0.4589741479618044, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1541912485966117, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2019045037563912, "3000–6000 Hz_10ms_envelope_correlation": 0.9730738124536581, "3000–6000 Hz_10ms_modulation_mae_db": 0.7671051404916787, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3579001986297197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12172461885861646, "3000–6000 Hz_40ms_envelope_correlation": 0.9895817017005887, "3000–6000 Hz_40ms_modulation_mae_db": 0.518787705021084, "mr_stft_db": 5.74588267256246, "band_mae_db": 0.16295469464359133, "rms_gain_error_db": 0.015779940924858304, "output_peak": 0.5959766633325696}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9838799443562928, "全帯域_10ms_rms_envelope_nrmse": 0.06544380903549388, "全帯域_10ms_envelope_correlation": 0.9952822715527951, "全帯域_10ms_modulation_mae_db": 0.1492670219031944, "全帯域_40ms_log_rms_mae_db": 0.499195588492431, "全帯域_40ms_rms_envelope_nrmse": 0.03491519048297971, "全帯域_40ms_envelope_correlation": 0.9985613473270503, "全帯域_40ms_modulation_mae_db": 0.4079639840257033, "300–1000 Hz_10ms_log_rms_mae_db": 1.643208465322469, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11775401258283737, "300–1000 Hz_10ms_envelope_correlation": 0.9882843414890975, "300–1000 Hz_10ms_modulation_mae_db": 0.6315591425324258, "300–1000 Hz_40ms_log_rms_mae_db": 0.9932384315610706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06789019414300676, "300–1000 Hz_40ms_envelope_correlation": 0.9962621191209199, "300–1000 Hz_40ms_modulation_mae_db": 1.038019319462424, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9048953631257446, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1780986034096162, "1000–3000 Hz_10ms_envelope_correlation": 0.9743415698192572, "1000–3000 Hz_10ms_modulation_mae_db": 0.44166155659259615, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1571404664591298, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09605496421324002, "1000–3000 Hz_40ms_envelope_correlation": 0.9929028394853783, "1000–3000 Hz_40ms_modulation_mae_db": 0.7845693204996357, "3000–6000 Hz_10ms_log_rms_mae_db": 2.195200396418967, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19727347329736625, "3000–6000 Hz_10ms_envelope_correlation": 0.9736824534979898, "3000–6000 Hz_10ms_modulation_mae_db": 0.5002545970709283, "3000–6000 Hz_40ms_log_rms_mae_db": 1.412029077504719, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12410842533140816, "3000–6000 Hz_40ms_envelope_correlation": 0.988692307966169, "3000–6000 Hz_40ms_modulation_mae_db": 0.3457798718207123, "mr_stft_db": 5.740929851352476, "band_mae_db": 0.21052759349719197, "rms_gain_error_db": -0.07861149731944442, "output_peak": 0.5780162671581485}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9734845852919628, "全帯域_10ms_rms_envelope_nrmse": 0.066968861763818, "全帯域_10ms_envelope_correlation": 0.994995355302792, "全帯域_10ms_modulation_mae_db": 0.2357404239725929, "全帯域_40ms_log_rms_mae_db": 0.49704828362367015, "全帯域_40ms_rms_envelope_nrmse": 0.032384057494661625, "全帯域_40ms_envelope_correlation": 0.998677698384818, "全帯域_40ms_modulation_mae_db": 0.3564558117532628, "300–1000 Hz_10ms_log_rms_mae_db": 1.6854469288714602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12342945574404705, "300–1000 Hz_10ms_envelope_correlation": 0.9865649161895238, "300–1000 Hz_10ms_modulation_mae_db": 0.21084263910443146, "300–1000 Hz_40ms_log_rms_mae_db": 0.9774649659896401, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06748319734972523, "300–1000 Hz_40ms_envelope_correlation": 0.9955611128763782, "300–1000 Hz_40ms_modulation_mae_db": 0.36154837467151424, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9738503502116476, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1716605395815983, "1000–3000 Hz_10ms_envelope_correlation": 0.97587575025497, "1000–3000 Hz_10ms_modulation_mae_db": 0.2820841180380747, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1627787205412552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08723416247765428, "1000–3000 Hz_40ms_envelope_correlation": 0.9937675112588317, "1000–3000 Hz_40ms_modulation_mae_db": 0.6172794438213484, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2207973703760104, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1891022194380271, "3000–6000 Hz_10ms_envelope_correlation": 0.9754672921148484, "3000–6000 Hz_10ms_modulation_mae_db": 0.7850269450720652, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3381098032902925, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11529027112285582, "3000–6000 Hz_40ms_envelope_correlation": 0.9902264588952383, "3000–6000 Hz_40ms_modulation_mae_db": 0.8734153052694165, "mr_stft_db": 5.7488698778544345, "band_mae_db": 0.15075408571088675, "rms_gain_error_db": -0.012432034953384082, "output_peak": 0.5813299739705482}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0188230993348883, "全帯域_10ms_rms_envelope_nrmse": 0.06538902718735175, "全帯域_10ms_envelope_correlation": 0.9952459097246191, "全帯域_10ms_modulation_mae_db": 0.12853850525445534, "全帯域_40ms_log_rms_mae_db": 0.5282562846745859, "全帯域_40ms_rms_envelope_nrmse": 0.03467422837456361, "全帯域_40ms_envelope_correlation": 0.9985183946692564, "全帯域_40ms_modulation_mae_db": 0.06994061981596773, "300–1000 Hz_10ms_log_rms_mae_db": 1.63842478259924, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11391372475831416, "300–1000 Hz_10ms_envelope_correlation": 0.9886206677106779, "300–1000 Hz_10ms_modulation_mae_db": 0.31019370550095027, "300–1000 Hz_40ms_log_rms_mae_db": 0.9570945095908656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06740674641869913, "300–1000 Hz_40ms_envelope_correlation": 0.9956603968707456, "300–1000 Hz_40ms_modulation_mae_db": 0.28387013635616415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.976955675938786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18610642438299033, "1000–3000 Hz_10ms_envelope_correlation": 0.9719722370304356, "1000–3000 Hz_10ms_modulation_mae_db": 0.6946628751422682, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2321502723619042, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11443188227296193, "1000–3000 Hz_40ms_envelope_correlation": 0.9894132969088368, "1000–3000 Hz_40ms_modulation_mae_db": 0.4633619814317895, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2084602559081703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2069880146041762, "3000–6000 Hz_10ms_envelope_correlation": 0.9704926587936565, "3000–6000 Hz_10ms_modulation_mae_db": 0.664699820302001, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3942716646629922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13416394098385562, "3000–6000 Hz_40ms_envelope_correlation": 0.9867033132547559, "3000–6000 Hz_40ms_modulation_mae_db": 0.7266899940864889, "mr_stft_db": 5.755633661558284, "band_mae_db": 0.15503949234911474, "rms_gain_error_db": -0.050822469027495994, "output_peak": 0.5808428249212055}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0135468715606262, "全帯域_10ms_rms_envelope_nrmse": 0.07095082636355907, "全帯域_10ms_envelope_correlation": 0.9943812616709368, "全帯域_10ms_modulation_mae_db": 0.15261941514855962, "全帯域_40ms_log_rms_mae_db": 0.5225748021814102, "全帯域_40ms_rms_envelope_nrmse": 0.035004572036763944, "全帯域_40ms_envelope_correlation": 0.9984714079963742, "全帯域_40ms_modulation_mae_db": 0.21870272055516182, "300–1000 Hz_10ms_log_rms_mae_db": 1.679512822885601, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11729704201278668, "300–1000 Hz_10ms_envelope_correlation": 0.987955725800607, "300–1000 Hz_10ms_modulation_mae_db": 0.3037944052843479, "300–1000 Hz_40ms_log_rms_mae_db": 1.007113921892981, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07104382794088893, "300–1000 Hz_40ms_envelope_correlation": 0.9952153574445667, "300–1000 Hz_40ms_modulation_mae_db": 0.23615667910988147, "1000–3000 Hz_10ms_log_rms_mae_db": 1.962433236320074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19254788892283284, "1000–3000 Hz_10ms_envelope_correlation": 0.9695992929060484, "1000–3000 Hz_10ms_modulation_mae_db": 0.501564969454049, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1880004477492943, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11674897986439114, "1000–3000 Hz_40ms_envelope_correlation": 0.9884124719369028, "1000–3000 Hz_40ms_modulation_mae_db": 0.5757465834928134, "3000–6000 Hz_10ms_log_rms_mae_db": 2.177156542534851, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19841208419878467, "3000–6000 Hz_10ms_envelope_correlation": 0.9733054795919203, "3000–6000 Hz_10ms_modulation_mae_db": 0.5048948375916961, "3000–6000 Hz_40ms_log_rms_mae_db": 1.357179344083129, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12257246574755658, "3000–6000 Hz_40ms_envelope_correlation": 0.9889409742617371, "3000–6000 Hz_40ms_modulation_mae_db": 0.5015270961770675, "mr_stft_db": 5.758984875402195, "band_mae_db": 0.18674320853947066, "rms_gain_error_db": -0.037793059086650406, "output_peak": 0.6350024189241382}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.996107096791931, "全帯域_10ms_rms_envelope_nrmse": 0.07156682023222287, "全帯域_10ms_envelope_correlation": 0.994280092245985, "全帯域_10ms_modulation_mae_db": 0.3003041977917136, "全帯域_40ms_log_rms_mae_db": 0.5140323609561223, "全帯域_40ms_rms_envelope_nrmse": 0.039354597623343215, "全帯域_40ms_envelope_correlation": 0.9980538222537867, "全帯域_40ms_modulation_mae_db": 0.10198564671529443, "300–1000 Hz_10ms_log_rms_mae_db": 1.6724511595796667, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11704734478429826, "300–1000 Hz_10ms_envelope_correlation": 0.9879325079510767, "300–1000 Hz_10ms_modulation_mae_db": 0.5301785374993785, "300–1000 Hz_40ms_log_rms_mae_db": 0.9696790573186859, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.073620915941146, "300–1000 Hz_40ms_envelope_correlation": 0.9947106766373618, "300–1000 Hz_40ms_modulation_mae_db": 0.3725791884587946, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9863519724225394, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17705537987141548, "1000–3000 Hz_10ms_envelope_correlation": 0.9746941355605627, "1000–3000 Hz_10ms_modulation_mae_db": 0.547839276337675, "1000–3000 Hz_40ms_log_rms_mae_db": 1.221759547059811, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0994244412982826, "1000–3000 Hz_40ms_envelope_correlation": 0.9923124376300481, "1000–3000 Hz_40ms_modulation_mae_db": 0.47809649262634585, "3000–6000 Hz_10ms_log_rms_mae_db": 2.142240466228026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20783538829860634, "3000–6000 Hz_10ms_envelope_correlation": 0.9703554378185906, "3000–6000 Hz_10ms_modulation_mae_db": 0.749149497522136, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3461422554533167, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1283769829616207, "3000–6000 Hz_40ms_envelope_correlation": 0.9882544221148509, "3000–6000 Hz_40ms_modulation_mae_db": 0.768353853328891, "mr_stft_db": 5.748397794155679, "band_mae_db": 0.19442986933516068, "rms_gain_error_db": -0.029994015020402255, "output_peak": 0.6214001716502682}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.260626371138841, "spectral_convergence": 0.2264954372263858, "spectral_convergence_status": "ok", "amplitude_floor": 2.016382236479878e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.515485171461843, "spectral_convergence": 0.21680081675230597, "spectral_convergence_status": "ok", "amplitude_floor": 1.164576113083389e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.455064085579199, "spectral_convergence": 0.22892307509877177, "spectral_convergence_status": "ok", "amplitude_floor": 9.482830334434223e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0029188205763026483, "output_power": 0.002980492346114973, "delta_db": 0.09080611739686084, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003712420548895462, "output_power": 0.0036862269273496215, "delta_db": -0.03075100648546712, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009931195946472759, "output_power": 0.0009120591958815191, "delta_db": -0.36978524281565556, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003469926396566426, "output_power": 0.0003411923707646161, "delta_db": -0.07320951156778206, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013263724937728663, "output_power": 0.0001384200710710054, "delta_db": 0.1853356096218865, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.438064081917431e-05, "output_power": 6.433311973615323e-05, "delta_db": -0.0032068280778681625, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1255157193275867, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.024337697091652975, "pitch": null, "output_rms_dbfs": -20.902983016450708, "output_peak": 0.5783774518923506, "reference_rms_dbfs": -20.87864531935906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.743725209393294}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023582673937264833, "log_rms_mae_db": 1.0058837201619695, "rms_envelope_nrmse": 0.06681319610548346, "envelope_correlation": 0.9950182686886759, "rms_gain_error_db": -0.024336112221119182, "reference_mean_power": 0.008175179133340451, "output_mean_power": 0.008129496836835075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4118743138857815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017924334098852407, "output_power": 0.001768521470043678, "reference_active": true, "delta_db": -0.05832693740298402}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00033647515979494424, "output_power": 0.00030666283836962185, "reference_active": true, "delta_db": -0.4029185700499443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.450579873055231e-05, "output_power": 4.255079980542341e-05, "reference_active": true, "delta_db": -1.075349804914629}], "active_band_count": 3, "active_band_mae_db": 0.5121984374558525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002175646104262597, "log_rms_mae_db": 0.5116137071541471, "rms_envelope_nrmse": 0.03326013021318172, "envelope_correlation": 0.9986103582462428, "rms_gain_error_db": -0.02435442747864209, "reference_mean_power": 0.008216199965234223, "output_mean_power": 0.008170253991326983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.000498094679175e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016116003385924513, "output_power": 0.0015997100977141962, "reference_active": true, "delta_db": -0.03216063785973303}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017199808112094875, "output_power": 0.00015347051031252846, "reference_active": true, "delta_db": -0.49498664558162475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5420165509687798e-06, "output_power": 2.572456541553067e-06, "reference_active": true, "delta_db": -1.3890253957719831}], "active_band_count": 3, "active_band_mae_db": 0.6387242264044469}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019245421130728438, "log_rms_mae_db": 1.6127412816423992, "rms_envelope_nrmse": 0.1265350337663396, "envelope_correlation": 0.9860140718806885, "rms_gain_error_db": -0.1802164651716297, "reference_mean_power": 0.0024199693169718968, "output_mean_power": 0.0023216043724627704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3235427499409267e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006913978889533925, "output_power": 0.0006387345477313415, "reference_active": true, "delta_db": -0.344076423601097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018527392767596858, "output_power": 0.0001518772361012379, "reference_active": true, "delta_db": -0.8632162320340147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604035138941515e-05, "output_power": 1.598032485527865e-05, "reference_active": true, "delta_db": -2.0472281094673166}], "active_band_count": 3, "active_band_mae_db": 1.0848402550341427}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017723694424605393, "log_rms_mae_db": 0.9451688041476286, "rms_envelope_nrmse": 0.07900230006715965, "envelope_correlation": 0.9941470416978466, "rms_gain_error_db": -0.18023614137385702, "reference_mean_power": 0.0024321306730885443, "output_mean_power": 0.0023332608325416612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1677626980235007e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006540469587465989, "output_power": 0.0006013927006903586, "reference_active": true, "delta_db": -0.36450777960348874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.291953991407944e-05, "output_power": 7.731350736059756e-05, "reference_active": true, "delta_db": -0.7985167507470549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.049791938534597e-06, "output_power": 1.2222710226422615e-06, "reference_active": true, "delta_db": -2.245422650799868}], "active_band_count": 3, "active_band_mae_db": 1.1361490603834705}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.001235221787968e-05, "log_rms_mae_db": 1.8997665850133794, "rms_envelope_nrmse": 0.1784848134712627, "envelope_correlation": 0.9737409271219271, "rms_gain_error_db": -0.07078695944288468, "reference_mean_power": 0.0003849600012580844, "output_mean_power": 0.00037873628172333825}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5531826233320727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012212996791699553, "output_power": 0.00011667076864972276, "reference_active": true, "delta_db": -0.19860183918741653}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.773842605512037e-05, "output_power": 2.7396335213354924e-05, "reference_active": true, "delta_db": -0.05389343057356218}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.9827801273791894e-06, "output_power": 9.674016185595903e-06, "reference_active": true, "delta_db": 1.4157844240731587}], "active_band_count": 3, "active_band_mae_db": 0.5560932312780458}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.393121399816863e-05, "log_rms_mae_db": 1.1584164067282394, "rms_envelope_nrmse": 0.09567660509519495, "envelope_correlation": 0.9916267768345506, "rms_gain_error_db": -0.07078807744439064, "reference_mean_power": 0.0003868956457514561, "output_mean_power": 0.0003806405343061462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2949024312838256e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011569002203634676, "output_power": 0.000113594891222515, "reference_active": true, "delta_db": -0.07937103793002656}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4800082503354946e-05, "output_power": 1.3129707035713982e-05, "reference_active": true, "delta_db": -0.5200910063761137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.803144704877755e-07, "output_power": 4.84762625145044e-07, "reference_active": true, "delta_db": -0.7813427143965158}], "active_band_count": 3, "active_band_mae_db": 0.4602682529008854}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.314726805023413e-05, "log_rms_mae_db": 2.1852143931168677, "rms_envelope_nrmse": 0.2005448877101278, "envelope_correlation": 0.9737469317653216, "rms_gain_error_db": 0.2239113945363225, "reference_mean_power": 8.262304385157997e-05, "output_mean_power": 8.699460687529394e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.2225580529948354e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.126031645648209e-05, "output_power": 2.387979321899128e-05, "reference_active": true, "delta_db": 0.5046083717000867}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.562486047803945e-06, "output_power": 1.0617895701944377e-05, "reference_active": true, "delta_db": 0.4546764051166664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.527384735880018e-06, "output_power": 2.7578011034323014e-06, "reference_active": true, "delta_db": -1.0688989021704658}], "active_band_count": 3, "active_band_mae_db": 0.676061226329073}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6400870566820594e-05, "log_rms_mae_db": 1.3714881758713178, "rms_envelope_nrmse": 0.11944070222787828, "envelope_correlation": 0.9902220762744715, "rms_gain_error_db": 0.22391065307302063, "reference_mean_power": 8.30385039555115e-05, "output_mean_power": 8.743203393444325e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.767621782464082e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.197465758003415e-05, "output_power": 2.5159380689356673e-05, "reference_active": true, "delta_db": 0.5877783012311613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.865612603723312e-06, "output_power": 5.745608245967097e-06, "reference_active": true, "delta_db": -0.08977365400451773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.828054895066371e-07, "output_power": 2.265969910328644e-07, "reference_active": true, "delta_db": -0.9623369665302749}], "active_band_count": 3, "active_band_mae_db": 0.5466296405886513}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9050264620363357, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "025318a1f9d640caacdd338c6a278f07eec4b491436ad14af9fa1a8276af1812", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9736652562299759, "standard_deviation": 0.002215550986824083, "minimum": 0.9695992929060484, "maximum": 0.9764188333021089, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9443936226781928, "standard_deviation": 0.037544448722468635, "minimum": 1.8973088102713964, "maximum": 1.9863519724225394, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.48101181273861093, "standard_deviation": 0.12405849574715247, "minimum": 0.2820841180380747, "maximum": 0.6946628751422682, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1793552202506419, "standard_deviation": 0.007579265258838623, "minimum": 0.16892243482877325, "maximum": 0.19254788892283284, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9912915383031017, "standard_deviation": 0.0019163490587983715, "minimum": 0.9884124719369028, "maximum": 0.9937675112588317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1777269244666257, "standard_deviation": 0.03819708756979515, "minimum": 1.1138057344366468, "maximum": 1.2321502723619042, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5750724991720725, "standard_deviation": 0.13588928613976226, "minimum": 0.4589741479618044, "maximum": 0.7845693204996357, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10111144489220972, "standard_deviation": 0.010632040916865163, "minimum": 0.08723416247765428, "maximum": 0.11674897986439114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9727207188517775, "standard_deviation": 0.0017611462135447432, "minimum": 0.9703554378185906, "maximum": 0.9754672921148484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1844336281529233, "standard_deviation": 0.026252619525628754, "minimum": 2.142240466228026, "maximum": 2.2207973703760104, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6707511293377983, "standard_deviation": 0.1117662607080617, "minimum": 0.5002545970709283, "maximum": 0.7850269450720652, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20068610482589966, "standard_deviation": 0.005986339672841691, "minimum": 0.1891022194380271, "maximum": 0.20783538829860634, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9890007807640909, "standard_deviation": 0.001159622919917196, "minimum": 0.9867033132547559, "maximum": 0.9902264588952383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.369979366531778, "standard_deviation": 0.02511793794735877, "minimum": 1.3381098032902925, "maximum": 1.412029077504719, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6262348604049119, "standard_deviation": 0.17483636811339498, "minimum": 0.3457798718207123, "maximum": 0.8734153052694165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12313996653237218, "standard_deviation": 0.005869927713637815, "minimum": 0.11529027112285582, "maximum": 0.13416394098385562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9874322270062403, "standard_deviation": 0.0011021864327629828, "minimum": 0.9858650464344424, "maximum": 0.9886206677106779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6526583781939412, "standard_deviation": 0.02677319935429412, "minimum": 1.6127412816423992, "maximum": 1.6854469288714602, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5137834366320403, "standard_deviation": 0.2895437370318838, "minimum": 0.21084263910443146, "maximum": 1.0848402550341427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11987447120219745, "standard_deviation": 0.004907781187565037, "minimum": 0.11391372475831416, "maximum": 0.12658862455931208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954322883562913, "standard_deviation": 0.0007501328038687486, "minimum": 0.9941470416978466, "maximum": 0.9964113951823246, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9761805999026831, "standard_deviation": 0.02510906156297158, "minimum": 0.9451688041476286, "maximum": 1.0095802827983442, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5556505847407969, "standard_deviation": 0.3432794230508392, "minimum": 0.23615667910988147, "maximum": 1.1361490603834705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06963314313367396, "standard_deviation": 0.004890219164021696, "minimum": 0.06288901127209438, "maximum": 0.07900230006715965, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16240919453517993, "standard_deviation": 0.03358415431831905, "minimum": 0.11330889287843661, "maximum": 0.21052759349719197, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.749389951201865, "standard_deviation": 0.00608502486378119, "minimum": 5.740929851352476, "maximum": 5.758984875402195, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5978102953536447, "standard_deviation": 0.022233966037647497, "minimum": 0.5780162671581485, "maximum": 0.6350024189241382, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.025660808964908036, "standard_deviation": 0.03159058357263469, "minimum": -0.07861149731944442, "maximum": 0.015779940924858304, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9948639334276804, "standard_deviation": 0.00036725507069534664, "minimum": 0.994280092245985, "maximum": 0.9952822715527951, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9992630704445027, "standard_deviation": 0.015073044485921012, "minimum": 0.9734845852919628, "maximum": 1.0188230993348883, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24865213074042153, "standard_deviation": 0.12674356082040322, "minimum": 0.12853850525445534, "maximum": 0.5121984374558525, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06791399102682051, "standard_deviation": 0.0023214616846950913, "minimum": 0.06538902718735175, "maximum": 0.07156682023222287, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984946170462975, "standard_deviation": 0.00019960255676178002, "minimum": 0.9980538222537867, "maximum": 0.998677698384818, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5086734777223433, "standard_deviation": 0.012481180985463464, "minimum": 0.4940001897199386, "maximum": 0.5282562846745859, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2635110234257216, "standard_deviation": 0.19281347290117737, "minimum": 0.06994061981596773, "maximum": 0.6387242264044469, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.034780347031546366, "standard_deviation": 0.0022200274176862496, "minimum": 0.032384057494661625, "maximum": 0.039354597623343215, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7176-88083-0026", "language": "en", "style": "normal", "speaker": "librispeech_7176", "source_utterance_id": "7176-88083-0026", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 88480, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7176-88083-0026.wav", "source_file_sha256": "fedfe73b1b0e35d6139855c46f0a7fd9e7435a0a279e2f34c63411fbe48ef233", "decoded_pcm_sha256": "bc9284b0dea002361e46bae8e07d7c2c20a9984ef0ce6f6e2c9fda86e70a579e", "original_source_sha256": "c637736d1dbd70e42d9ad3e319b273baf390cbc3f3ec8b31239ea7dca8e9d5d6", "corpus_entry_id": "librispeech_7176-88083-0026"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.9027793407440186, "predicted_mos": 3.9027793407440186, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "803ff98e7c871bad36ff22d335d9a777e29a57ddf3910411406ff69435a764b2", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4535644259995024, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7176-88083-0026", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "803ff98e7c871bad36ff22d335d9a777e29a57ddf3910411406ff69435a764b2", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "d8ac2a315ecbb8b2cd771de10cee21a68a53a6c060b81e6b96a589b954912c21"}, "rows": [{"case": "librispeech_7176-88083-0026", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.384065628051758, "predicted_mos": 2.384065628051758, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "e364f8bfb37cc2ae10fa974d6effdccb5e9a415cf89cef3c685a68b4ef56a956", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.34784454200053, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7176-88083-0026", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "e364f8bfb37cc2ae10fa974d6effdccb5e9a415cf89cef3c685a68b4ef56a956", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0748890306043695, "全帯域_10ms_rms_envelope_nrmse": 0.07961774537409796, "全帯域_10ms_envelope_correlation": 0.993620381960512, "全帯域_10ms_modulation_mae_db": 0.18637700873603222, "全帯域_40ms_log_rms_mae_db": 0.562523606011493, "全帯域_40ms_rms_envelope_nrmse": 0.04120907813523354, "全帯域_40ms_envelope_correlation": 0.9981456471748053, "全帯域_40ms_modulation_mae_db": 0.09790806640651366, "300–1000 Hz_10ms_log_rms_mae_db": 1.8684406057637186, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11713586501345055, "300–1000 Hz_10ms_envelope_correlation": 0.9894649730185794, "300–1000 Hz_10ms_modulation_mae_db": 0.11168388796286423, "300–1000 Hz_40ms_log_rms_mae_db": 1.086377976567984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07561992946344913, "300–1000 Hz_40ms_envelope_correlation": 0.9953345109651792, "300–1000 Hz_40ms_modulation_mae_db": 0.28649556764794515, "1000–3000 Hz_10ms_log_rms_mae_db": 2.366847434737324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.194682383914381, "1000–3000 Hz_10ms_envelope_correlation": 0.9726028379000361, "1000–3000 Hz_10ms_modulation_mae_db": 0.9544036060221693, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5328997925558834, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12466526952886017, "1000–3000 Hz_40ms_envelope_correlation": 0.9892668074867077, "1000–3000 Hz_40ms_modulation_mae_db": 1.199585332096751, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6995523630049867, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24791193543874543, "3000–6000 Hz_10ms_envelope_correlation": 0.9552469342260619, "3000–6000 Hz_10ms_modulation_mae_db": 0.9667077002776406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8268753444727488, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16241750038724495, "3000–6000 Hz_40ms_envelope_correlation": 0.9788020038138682, "3000–6000 Hz_40ms_modulation_mae_db": 0.8353770137652825, "mr_stft_db": 6.061082787785023, "band_mae_db": 0.24126030296841794, "rms_gain_error_db": -0.08871174486387216, "output_peak": 0.6087917723465202}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1006620568168408, "全帯域_10ms_rms_envelope_nrmse": 0.08278204027538505, "全帯域_10ms_envelope_correlation": 0.9930731373711118, "全帯域_10ms_modulation_mae_db": 0.26571591165262226, "全帯域_40ms_log_rms_mae_db": 0.5904149318495587, "全帯域_40ms_rms_envelope_nrmse": 0.04483562127688225, "全帯域_40ms_envelope_correlation": 0.9977487391528344, "全帯域_40ms_modulation_mae_db": 0.34646254581174124, "300–1000 Hz_10ms_log_rms_mae_db": 1.9324631040025209, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15122689620403007, "300–1000 Hz_10ms_envelope_correlation": 0.9823541693219965, "300–1000 Hz_10ms_modulation_mae_db": 0.3419510411550948, "300–1000 Hz_40ms_log_rms_mae_db": 1.1814265299725861, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0909146067286172, "300–1000 Hz_40ms_envelope_correlation": 0.9933608491474182, "300–1000 Hz_40ms_modulation_mae_db": 0.29721638810292955, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3395411547560454, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20821555454234733, "1000–3000 Hz_10ms_envelope_correlation": 0.9681259018923771, "1000–3000 Hz_10ms_modulation_mae_db": 1.4465848251125013, "1000–3000 Hz_40ms_log_rms_mae_db": 1.481521920843018, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12676109784769035, "1000–3000 Hz_40ms_envelope_correlation": 0.9886148444595697, "1000–3000 Hz_40ms_modulation_mae_db": 1.4638806708093302, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6355682714949125, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2434408695499092, "3000–6000 Hz_10ms_envelope_correlation": 0.956906220333095, "3000–6000 Hz_10ms_modulation_mae_db": 1.0277585771528657, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7675988623927812, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1596026911682912, "3000–6000 Hz_40ms_envelope_correlation": 0.9798335237404617, "3000–6000 Hz_40ms_modulation_mae_db": 1.4245331786508952, "mr_stft_db": 6.066613067703899, "band_mae_db": 0.1908699171072504, "rms_gain_error_db": -0.07589459891172279, "output_peak": 0.6289358815875784}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0664429252953278, "全帯域_10ms_rms_envelope_nrmse": 0.08094196178267903, "全帯域_10ms_envelope_correlation": 0.9933929164568063, "全帯域_10ms_modulation_mae_db": 0.28177419214631044, "全帯域_40ms_log_rms_mae_db": 0.5668201886732626, "全帯域_40ms_rms_envelope_nrmse": 0.04186342636462954, "全帯域_40ms_envelope_correlation": 0.9980658338658549, "全帯域_40ms_modulation_mae_db": 0.4615888973540891, "300–1000 Hz_10ms_log_rms_mae_db": 1.9247840522784552, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1410503229606568, "300–1000 Hz_10ms_envelope_correlation": 0.9848302453794636, "300–1000 Hz_10ms_modulation_mae_db": 0.9868841793509743, "300–1000 Hz_40ms_log_rms_mae_db": 1.1278228604948377, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0850717778556445, "300–1000 Hz_40ms_envelope_correlation": 0.9943955120506187, "300–1000 Hz_40ms_modulation_mae_db": 1.233528031722207, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3198271784685107, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21828653112589924, "1000–3000 Hz_10ms_envelope_correlation": 0.9640623164432003, "1000–3000 Hz_10ms_modulation_mae_db": 1.1090217919718475, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4688492071390595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12606097097563482, "1000–3000 Hz_40ms_envelope_correlation": 0.987950473989321, "1000–3000 Hz_40ms_modulation_mae_db": 1.1440491289494956, "3000–6000 Hz_10ms_log_rms_mae_db": 2.672157805189313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24626661389426727, "3000–6000 Hz_10ms_envelope_correlation": 0.9562024510819758, "3000–6000 Hz_10ms_modulation_mae_db": 0.9329346247415033, "3000–6000 Hz_40ms_log_rms_mae_db": 1.817265404387342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15974625688853183, "3000–6000 Hz_40ms_envelope_correlation": 0.9794102688466149, "3000–6000 Hz_40ms_modulation_mae_db": 0.6794330661746913, "mr_stft_db": 6.055411757102694, "band_mae_db": 0.21863833533690002, "rms_gain_error_db": -0.08341279321382179, "output_peak": 0.6426161913455176}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1538115486800127, "全帯域_10ms_rms_envelope_nrmse": 0.09332085952203456, "全帯域_10ms_envelope_correlation": 0.9912508937025539, "全帯域_10ms_modulation_mae_db": 0.3708952141327355, "全帯域_40ms_log_rms_mae_db": 0.659852360224386, "全帯域_40ms_rms_envelope_nrmse": 0.057628536822175336, "全帯域_40ms_envelope_correlation": 0.9963427144431238, "全帯域_40ms_modulation_mae_db": 0.33602593133605957, "300–1000 Hz_10ms_log_rms_mae_db": 1.8733058831613796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13393871231529164, "300–1000 Hz_10ms_envelope_correlation": 0.9861668771527631, "300–1000 Hz_10ms_modulation_mae_db": 0.5344796406105736, "300–1000 Hz_40ms_log_rms_mae_db": 1.1248193958621517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08131996557897139, "300–1000 Hz_40ms_envelope_correlation": 0.9946200671539003, "300–1000 Hz_40ms_modulation_mae_db": 0.3494582038644312, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3882850736481442, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2008157360152626, "1000–3000 Hz_10ms_envelope_correlation": 0.9697997083480645, "1000–3000 Hz_10ms_modulation_mae_db": 1.2149122918071928, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5035774842416996, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12078269768186353, "1000–3000 Hz_40ms_envelope_correlation": 0.9888649097874097, "1000–3000 Hz_40ms_modulation_mae_db": 1.5456412724246649, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6837680040923066, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23550744254453218, "3000–6000 Hz_10ms_envelope_correlation": 0.9600699684488131, "3000–6000 Hz_10ms_modulation_mae_db": 0.9338822441124149, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8503731612395755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16472203888564502, "3000–6000 Hz_40ms_envelope_correlation": 0.9791046161237397, "3000–6000 Hz_40ms_modulation_mae_db": 1.0984158560518344, "mr_stft_db": 6.066164635272817, "band_mae_db": 0.18344377255817923, "rms_gain_error_db": -0.12268624602310224, "output_peak": 0.6165592686668327}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1422426369099576, "全帯域_10ms_rms_envelope_nrmse": 0.09417454648521402, "全帯域_10ms_envelope_correlation": 0.9909999060236898, "全帯域_10ms_modulation_mae_db": 0.34800692273731687, "全帯域_40ms_log_rms_mae_db": 0.6316704965402065, "全帯域_40ms_rms_envelope_nrmse": 0.05861456934989227, "全帯域_40ms_envelope_correlation": 0.9960567742647851, "全帯域_40ms_modulation_mae_db": 0.278811284733629, "300–1000 Hz_10ms_log_rms_mae_db": 1.8668566106141706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14262378637923873, "300–1000 Hz_10ms_envelope_correlation": 0.984323811346243, "300–1000 Hz_10ms_modulation_mae_db": 0.7839508278794737, "300–1000 Hz_40ms_log_rms_mae_db": 1.106688600407331, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09285545362576404, "300–1000 Hz_40ms_envelope_correlation": 0.9929401743449551, "300–1000 Hz_40ms_modulation_mae_db": 0.8738863881098116, "1000–3000 Hz_10ms_log_rms_mae_db": 2.375260163194035, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20535219086733061, "1000–3000 Hz_10ms_envelope_correlation": 0.9679607567317053, "1000–3000 Hz_10ms_modulation_mae_db": 1.2135165956225216, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4812959114978765, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11389453480283758, "1000–3000 Hz_40ms_envelope_correlation": 0.9896635872505449, "1000–3000 Hz_40ms_modulation_mae_db": 1.296460147363888, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6252534741545537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24034646347659358, "3000–6000 Hz_10ms_envelope_correlation": 0.9592219280193623, "3000–6000 Hz_10ms_modulation_mae_db": 0.21044877790777292, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7936358758147413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16610669975832953, "3000–6000 Hz_40ms_envelope_correlation": 0.9783032582598579, "3000–6000 Hz_40ms_modulation_mae_db": 0.18171222486463093, "mr_stft_db": 6.062910649945881, "band_mae_db": 0.2142821132202871, "rms_gain_error_db": -0.037897299566816485, "output_peak": 0.6535605218327193}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1355897634761898, "全帯域_10ms_rms_envelope_nrmse": 0.09436153087408528, "全帯域_10ms_envelope_correlation": 0.991070848239698, "全帯域_10ms_modulation_mae_db": 0.43641244173931276, "全帯域_40ms_log_rms_mae_db": 0.646959731703025, "全帯域_40ms_rms_envelope_nrmse": 0.05694554134774356, "全帯域_40ms_envelope_correlation": 0.9964572945520295, "全帯域_40ms_modulation_mae_db": 0.34493274327588247, "300–1000 Hz_10ms_log_rms_mae_db": 1.9628792245223174, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1496793079411562, "300–1000 Hz_10ms_envelope_correlation": 0.9828885640342352, "300–1000 Hz_10ms_modulation_mae_db": 0.5916776984905144, "300–1000 Hz_40ms_log_rms_mae_db": 1.2217473576292315, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09771196787993672, "300–1000 Hz_40ms_envelope_correlation": 0.9924549420558688, "300–1000 Hz_40ms_modulation_mae_db": 0.9950931270621606, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3463769916707684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20833211071860738, "1000–3000 Hz_10ms_envelope_correlation": 0.9679902691372682, "1000–3000 Hz_10ms_modulation_mae_db": 1.4187851944729915, "1000–3000 Hz_40ms_log_rms_mae_db": 1.523597533771619, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12424505925068215, "1000–3000 Hz_40ms_envelope_correlation": 0.9891798388695614, "1000–3000 Hz_40ms_modulation_mae_db": 1.682521276388148, "3000–6000 Hz_10ms_log_rms_mae_db": 2.593338793596412, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24477597485984767, "3000–6000 Hz_10ms_envelope_correlation": 0.9564568747120694, "3000–6000 Hz_10ms_modulation_mae_db": 1.3112798581005551, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7670014071257862, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16455280294844, "3000–6000 Hz_40ms_envelope_correlation": 0.9787866176329038, "3000–6000 Hz_40ms_modulation_mae_db": 1.3162883903199807, "mr_stft_db": 6.060142145661757, "band_mae_db": 0.27412454659099533, "rms_gain_error_db": -0.1259118985322717, "output_peak": 0.611282347113761}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0980996362010806, "全帯域_10ms_rms_envelope_nrmse": 0.08322770089312063, "全帯域_10ms_envelope_correlation": 0.99306085259231, "全帯域_10ms_modulation_mae_db": 0.5270905056564329, "全帯域_40ms_log_rms_mae_db": 0.5963953851115337, "全帯域_40ms_rms_envelope_nrmse": 0.044726431360914615, "全帯域_40ms_envelope_correlation": 0.9978344787430962, "全帯域_40ms_modulation_mae_db": 0.8267630556504365, "300–1000 Hz_10ms_log_rms_mae_db": 1.9376925161485201, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12638583990474944, "300–1000 Hz_10ms_envelope_correlation": 0.9877500572138228, "300–1000 Hz_10ms_modulation_mae_db": 0.3010431026701748, "300–1000 Hz_40ms_log_rms_mae_db": 1.1467143300241562, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07562567383476461, "300–1000 Hz_40ms_envelope_correlation": 0.9954321214279519, "300–1000 Hz_40ms_modulation_mae_db": 0.5214246095403579, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3715655633413997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21754415849419326, "1000–3000 Hz_10ms_envelope_correlation": 0.9638695267548828, "1000–3000 Hz_10ms_modulation_mae_db": 0.7044781043164732, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5219440710388017, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1260587359606579, "1000–3000 Hz_40ms_envelope_correlation": 0.9870785163708052, "1000–3000 Hz_40ms_modulation_mae_db": 0.8775638319886108, "3000–6000 Hz_10ms_log_rms_mae_db": 2.668606316646633, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2336790516028975, "3000–6000 Hz_10ms_envelope_correlation": 0.960577914835929, "3000–6000 Hz_10ms_modulation_mae_db": 0.9126739456292858, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7904128443851253, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16321740892532044, "3000–6000 Hz_40ms_envelope_correlation": 0.979253277418447, "3000–6000 Hz_40ms_modulation_mae_db": 0.9634006317879501, "mr_stft_db": 6.0529850819753435, "band_mae_db": 0.15351088480293293, "rms_gain_error_db": -0.10009562139691196, "output_peak": 0.6299639416376974}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.114727533861749, "全帯域_10ms_rms_envelope_nrmse": 0.08227498158434754, "全帯域_10ms_envelope_correlation": 0.9933317654759204, "全帯域_10ms_modulation_mae_db": 0.29787857989788463, "全帯域_40ms_log_rms_mae_db": 0.6070579910968581, "全帯域_40ms_rms_envelope_nrmse": 0.05158357622448586, "全帯域_40ms_envelope_correlation": 0.9972201595379401, "全帯域_40ms_modulation_mae_db": 0.3957896015375286, "300–1000 Hz_10ms_log_rms_mae_db": 1.9320125221862916, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14019046630109006, "300–1000 Hz_10ms_envelope_correlation": 0.9854830407876676, "300–1000 Hz_10ms_modulation_mae_db": 0.5997147344067599, "300–1000 Hz_40ms_log_rms_mae_db": 1.1165560235677412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10090808859746758, "300–1000 Hz_40ms_envelope_correlation": 0.9923861117432919, "300–1000 Hz_40ms_modulation_mae_db": 0.7742559660694369, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4418349092697773, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20511610011987916, "1000–3000 Hz_10ms_envelope_correlation": 0.9679238225531027, "1000–3000 Hz_10ms_modulation_mae_db": 1.3498332095341796, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5246995466127924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11961543755060287, "1000–3000 Hz_40ms_envelope_correlation": 0.9882144832995532, "1000–3000 Hz_40ms_modulation_mae_db": 1.632174728974995, "3000–6000 Hz_10ms_log_rms_mae_db": 2.654328201377958, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2420007663552947, "3000–6000 Hz_10ms_envelope_correlation": 0.9575955262425112, "3000–6000 Hz_10ms_modulation_mae_db": 0.632821033780545, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7687552737136996, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15930615842028845, "3000–6000 Hz_40ms_envelope_correlation": 0.9795028429565147, "3000–6000 Hz_40ms_modulation_mae_db": 0.7614215204366731, "mr_stft_db": 6.065833406220253, "band_mae_db": 0.1909219512165895, "rms_gain_error_db": -0.14808965940144023, "output_peak": 0.6307839895277987}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88480, "output_samples": 88480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.652066059165366, "spectral_convergence": 0.28271224389848043, "spectral_convergence_status": "ok", "amplitude_floor": 2.1848847242256272e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.853901910605009, "spectral_convergence": 0.2527656975944428, "spectral_convergence_status": "ok", "amplitude_floor": 1.2036494394056804e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.677280393584694, "spectral_convergence": 0.2914238930211682, "spectral_convergence_status": "ok", "amplitude_floor": 7.153950066191215e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002200971739696214, "output_power": 0.002216873959534386, "delta_db": 0.0312653570248213, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002201298666996261, "output_power": 0.0021747048389422394, "delta_db": -0.052786497761638684, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011542423166681864, "output_power": 0.0010766860182495261, "delta_db": -0.30207918927887556, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019781669780676847, "output_power": 0.00017115662718787032, "delta_db": -0.6286922813360409, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.802033486419172e-05, "output_power": 5.4455936862143405e-05, "delta_db": -0.27534996940496026, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6630669693501258e-05, "output_power": 1.7244419937923417e-05, "delta_db": 0.15738852300417072, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24126030296841794, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08871174486387216, "pitch": null, "output_rms_dbfs": -22.43278577566983, "output_peak": 0.6087917723465202, "reference_rms_dbfs": -22.34407403080596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.061082787785023}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00024519594682554, "log_rms_mae_db": 1.0748890306043695, "rms_envelope_nrmse": 0.07961774537409796, "envelope_correlation": 0.993620381960512, "rms_gain_error_db": -0.08870570389127912, "reference_mean_power": 0.00583423429533283, "output_mean_power": 0.005716277391358636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.472320009859619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016537460501756134, "output_power": 0.0016603681462395237, "reference_active": true, "delta_db": 0.01735573028716897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003778384775464376, "output_power": 0.0003650143773872345, "reference_active": true, "delta_db": -0.14996211612589688}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8517105494706513e-05, "output_power": 5.3097760120958354e-05, "reference_active": true, "delta_db": 0.3918131797950307}], "active_band_count": 3, "active_band_mae_db": 0.18637700873603222}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00020002070942310007, "log_rms_mae_db": 0.562523606011493, "rms_envelope_nrmse": 0.04120907813523354, "envelope_correlation": 0.9981456471748053, "rms_gain_error_db": -0.08869292144452445, "reference_mean_power": 0.005865946925483087, "output_mean_power": 0.0057473657696308095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.084203911038529e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015941324207372327, "output_power": 0.0015958066619550502, "reference_active": true, "delta_db": 0.004558794197568643}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018248750335661006, "output_power": 0.00017988113378533365, "reference_active": true, "delta_db": -0.06247513315122837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.552526527817181e-06, "output_power": 4.796467568518349e-06, "reference_active": true, "delta_db": 0.22669027187074395}], "active_band_count": 3, "active_band_mae_db": 0.09790806640651366}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00018690652054173922, "log_rms_mae_db": 1.8684406057637186, "rms_envelope_nrmse": 0.11713586501345055, "envelope_correlation": 0.9894649730185794, "rms_gain_error_db": -0.1757316071429379, "reference_mean_power": 0.0020929267743018004, "output_mean_power": 0.002009929725363824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 8.903808595053404e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005237826202169963, "output_power": 0.0005121465388768634, "reference_active": true, "delta_db": -0.09756841807501518}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020282121471265822, "output_power": 0.00020171327713001052, "reference_active": true, "delta_db": -0.023788942166092078}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.57818052122161e-05, "output_power": 2.454391806781538e-05, "reference_active": true, "delta_db": -0.21369430364748543}], "active_band_count": 3, "active_band_mae_db": 0.11168388796286423}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00016950242057071555, "log_rms_mae_db": 1.086377976567984, "rms_envelope_nrmse": 0.07561992946344913, "envelope_correlation": 0.9953345109651792, "rms_gain_error_db": -0.17572603546751583, "reference_mean_power": 0.002104315784196243, "output_mean_power": 0.002020869685606348}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 7.681995008528509e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000538193996495516, "output_power": 0.0005305091998746756, "reference_active": true, "delta_db": -0.06245929278941463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010305287620923108, "output_power": 0.00010437279880754349, "reference_active": true, "delta_db": 0.055272119839650864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.715218896542221e-06, "output_power": 2.2889131721630582e-06, "reference_active": true, "delta_db": -0.7417552903147699}], "active_band_count": 3, "active_band_mae_db": 0.28649556764794515}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 7.550458878286076e-05, "log_rms_mae_db": 2.366847434737324, "rms_envelope_nrmse": 0.194682383914381, "envelope_correlation": 0.9726028379000361, "rms_gain_error_db": -0.5358635959544804, "reference_mean_power": 0.00021315330376123098, "output_mean_power": 0.00018841076319576399}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 1.6207032137826829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010202948507054384, "output_power": 8.999503631237687e-05, "reference_active": true, "delta_db": -0.5450713829777045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.692530062160877e-05, "output_power": 3.161953002700687e-05, "reference_active": true, "delta_db": -0.6736862917797672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.752197568216375e-06, "output_power": 4.623808370614041e-06, "reference_active": true, "delta_db": -1.6444531433090361}], "active_band_count": 3, "active_band_mae_db": 0.9544036060221693}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 5.938009261210461e-05, "log_rms_mae_db": 1.5328997925558834, "rms_envelope_nrmse": 0.12466526952886017, "envelope_correlation": 0.9892668074867077, "rms_gain_error_db": -0.535592420337957, "reference_mean_power": 0.00021425147153668813, "output_mean_power": 0.00018939328261613651}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 1.4064708995422735e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010396272320640846, "output_power": 9.184997572668801e-05, "reference_active": true, "delta_db": -0.5379860104591033}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97231105181685e-05, "output_power": 1.7069878081089527e-05, "reference_active": true, "delta_db": -0.6274498904415362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.307963255185344e-07, "output_power": 3.6021114454574813e-07, "reference_active": true, "delta_db": -2.4333200953896132}], "active_band_count": 3, "active_band_mae_db": 1.199585332096751}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 4.4291934004628345e-05, "log_rms_mae_db": 2.6995523630049867, "rms_envelope_nrmse": 0.24791193543874543, "envelope_correlation": 0.9552469342260619, "rms_gain_error_db": -0.17017944495522397, "reference_mean_power": 2.6417005173666327e-05, "output_mean_power": 2.5401866938610563e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.694560310630941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.502841409588563e-05, "output_power": 1.355480770034765e-05, "reference_active": true, "delta_db": -0.4481979249058043}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.7026465908103935e-06, "output_power": 5.687534753113962e-06, "reference_active": true, "delta_db": -0.7132225764730977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.011540321330325e-07, "output_power": 6.03849582421338e-07, "reference_active": true, "delta_db": -1.7387025994540197}], "active_band_count": 3, "active_band_mae_db": 0.9667077002776406}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 3.2524867537670264e-05, "log_rms_mae_db": 1.8268753444727488, "rms_envelope_nrmse": 0.16241750038724495, "envelope_correlation": 0.9788020038138682, "rms_gain_error_db": -0.1702380214739213, "reference_mean_power": 2.65459352580292e-05, "output_mean_power": 2.5525498284511915e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.4345030346700597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6238306479668564e-05, "output_power": 1.4150759344944708e-05, "reference_active": true, "delta_db": -0.597609887699392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6930369104619132e-06, "output_power": 3.1745530707759113e-06, "reference_active": true, "delta_db": -0.657010564620769}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.897644209500044e-08, "output_power": 5.920338054599572e-08, "reference_active": true, "delta_db": -1.2515105889756866}], "active_band_count": 3, "active_band_mae_db": 0.8353770137652825}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.885396680789366, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "da1bfe2dc5d9b3f0db9972ad4a32a5363bba77f123c42764a33618290f562eb6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9677918924700796, "standard_deviation": 0.0028452478414733445, "minimum": 0.9638695267548828, "maximum": 0.9726028379000361, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.3686923086357505, "standard_deviation": 0.036886072524870764, "minimum": 2.3198271784685107, "maximum": 2.4418349092697773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1764419523574845, "standard_deviation": 0.2510614102686927, "minimum": 0.7044781043164732, "maximum": 1.4465848251125013, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20729309572473756, "standard_deviation": 0.007904456744226029, "minimum": 0.194682383914381, "maximum": 0.21828653112589924, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.988604182689184, "standard_deviation": 0.000833901362068673, "minimum": 0.9870785163708052, "maximum": 0.9896635872505449, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.5047981834625939, "standard_deviation": 0.024556412941931026, "minimum": 1.4688492071390595, "maximum": 1.5328997925558834, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3552345486244854, "standard_deviation": 0.2757640994610408, "minimum": 0.8775638319886108, "maximum": 1.682521276388148, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12276047544985366, "standard_deviation": 0.004408057137166911, "minimum": 0.11389453480283758, "maximum": 0.12676109784769035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9577847272374773, "standard_deviation": 0.0019495246808761291, "minimum": 0.9552469342260619, "maximum": 0.960577914835929, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.6540716536946345, "standard_deviation": 0.03453805806971319, "minimum": 2.593338793596412, "maximum": 2.6995523630049867, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8660633452128228, "standard_deviation": 0.3229260272296491, "minimum": 0.21044877790777292, "maximum": 1.3112798581005551, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24174113971526096, "standard_deviation": 0.005021762855273344, "minimum": 0.2336790516028975, "maximum": 0.24791193543874543, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9791245510990509, "standard_deviation": 0.0004836036071503983, "minimum": 0.9783032582598579, "maximum": 0.9798335237404617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.797739771691475, "standard_deviation": 0.03108595007959489, "minimum": 1.7670014071257862, "maximum": 1.8503731612395755, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9075727352564923, "standard_deviation": 0.3926699862051701, "minimum": 0.18171222486463093, "maximum": 1.4245331786508952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16245894467276142, "standard_deviation": 0.002641586670404312, "minimum": 0.15930615842028845, "maximum": 0.16610669975832953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9854077172818464, "standard_deviation": 0.0023817308206324863, "minimum": 0.9823541693219965, "maximum": 0.9894649730185794, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.912304314834672, "standard_deviation": 0.0371547204953476, "minimum": 1.8668566106141706, "maximum": 1.9628792245223174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5314231390658037, "standard_deviation": 0.27894555662745085, "minimum": 0.11168388796286423, "maximum": 0.9868841793509743, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13777889962745793, "standard_deviation": 0.011545827004674021, "minimum": 0.11713586501345055, "maximum": 0.15122689620403007, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.993865536111148, "standard_deviation": 0.001239339896084208, "minimum": 0.9923861117432919, "maximum": 0.9954321214279519, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1390191343157525, "standard_deviation": 0.043670024946704813, "minimum": 1.086377976567984, "maximum": 1.2217473576292315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6664197852649101, "standard_deviation": 0.3558592752900364, "minimum": 0.28649556764794515, "maximum": 1.233528031722207, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08750343294557689, "standard_deviation": 0.009640882015411235, "minimum": 0.07561992946344913, "maximum": 0.10090808859746758, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20838147797519407, "standard_deviation": 0.03739241674151657, "minimum": 0.15351088480293293, "maximum": 0.27412454659099533, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.061392941458458, "standard_deviation": 0.005073281539334905, "minimum": 6.0529850819753435, "maximum": 6.066613067703899, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6278117392573032, "standard_deviation": 0.015392482343177973, "minimum": 0.6087917723465202, "maximum": 0.6535605218327193, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09783748273874492, "standard_deviation": 0.034437688609657076, "minimum": -0.14808965940144023, "maximum": -0.037897299566816485, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924750877278252, "standard_deviation": 0.0011486320860230367, "minimum": 0.9909999060236898, "maximum": 0.993620381960512, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.110808141480691, "standard_deviation": 0.031599762731421926, "minimum": 1.0664429252953278, "maximum": 1.1538115486800127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.33926884708733096, "standard_deviation": 0.1066336670089405, "minimum": 0.18637700873603222, "maximum": 0.5270905056564329, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0863376708488705, "standard_deviation": 0.006410424703593362, "minimum": 0.07961774537409796, "maximum": 0.09436153087408528, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972339552168086, "standard_deviation": 0.0008394124253552617, "minimum": 0.9960567742647851, "maximum": 0.9981456471748053, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6077118364012905, "standard_deviation": 0.03580667219554921, "minimum": 0.562523606011493, "maximum": 0.659852360224386, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38603526576323505, "standard_deviation": 0.20712835884261246, "minimum": 0.09790806640651366, "maximum": 0.8267630556504365, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04967584761024462, "standard_deviation": 0.007370583527706207, "minimum": 0.04120907813523354, "maximum": 0.05861456934989227, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0026", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.4372780323028564, "predicted_mos": 2.4372780323028564, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "04f4049ee39824196683f488a54a3afcc071d7a16de1605ab386f5f0096fc3c7", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7791637000627816, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0932422808860038, "全帯域_10ms_rms_envelope_nrmse": 0.08022475201720725, "全帯域_10ms_envelope_correlation": 0.9935448181230118, "全帯域_10ms_modulation_mae_db": 0.34123414933276824, "全帯域_40ms_log_rms_mae_db": 0.5754163100231986, "全帯域_40ms_rms_envelope_nrmse": 0.04111401749161085, "全帯域_40ms_envelope_correlation": 0.9981915989197109, "全帯域_40ms_modulation_mae_db": 0.22338959673965209, "300–1000 Hz_10ms_log_rms_mae_db": 1.8511952392052813, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11597865734406529, "300–1000 Hz_10ms_envelope_correlation": 0.9897284715186085, "300–1000 Hz_10ms_modulation_mae_db": 0.31428066203647825, "300–1000 Hz_40ms_log_rms_mae_db": 1.082587858267384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07080158850800314, "300–1000 Hz_40ms_envelope_correlation": 0.996025707959351, "300–1000 Hz_40ms_modulation_mae_db": 0.5188272298804659, "1000–3000 Hz_10ms_log_rms_mae_db": 2.463948482384671, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20238581130659847, "1000–3000 Hz_10ms_envelope_correlation": 0.971148938379933, "1000–3000 Hz_10ms_modulation_mae_db": 0.9624495738100952, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5595810971277497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1308986508029206, "1000–3000 Hz_40ms_envelope_correlation": 0.9893091238286092, "1000–3000 Hz_40ms_modulation_mae_db": 1.2283238242457823, "3000–6000 Hz_10ms_log_rms_mae_db": 2.728303382877563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24561727991774687, "3000–6000 Hz_10ms_envelope_correlation": 0.9558770249848593, "3000–6000 Hz_10ms_modulation_mae_db": 0.7475014806065406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.873116723503196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16006151742876695, "3000–6000 Hz_40ms_envelope_correlation": 0.9793766577001882, "3000–6000 Hz_40ms_modulation_mae_db": 0.6770239091246036, "mr_stft_db": 6.025229135831354, "band_mae_db": 0.2560295346024502, "rms_gain_error_db": -0.10306360957753544, "output_peak": 0.6112303525685826}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0869152323892524, "全帯域_10ms_rms_envelope_nrmse": 0.08602697438668418, "全帯域_10ms_envelope_correlation": 0.992515345465721, "全帯域_10ms_modulation_mae_db": 0.34328720554876235, "全帯域_40ms_log_rms_mae_db": 0.5795336997806435, "全帯域_40ms_rms_envelope_nrmse": 0.04666164333281457, "全帯域_40ms_envelope_correlation": 0.9975596756018394, "全帯域_40ms_modulation_mae_db": 0.21304108787554943, "300–1000 Hz_10ms_log_rms_mae_db": 1.9489904017104516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15490962993192675, "300–1000 Hz_10ms_envelope_correlation": 0.981430231408168, "300–1000 Hz_10ms_modulation_mae_db": 0.38973915387575514, "300–1000 Hz_40ms_log_rms_mae_db": 1.1933883434903823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09167887091244462, "300–1000 Hz_40ms_envelope_correlation": 0.9932062743606983, "300–1000 Hz_40ms_modulation_mae_db": 0.3048360750874889, "1000–3000 Hz_10ms_log_rms_mae_db": 2.440821743214296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20869826787105766, "1000–3000 Hz_10ms_envelope_correlation": 0.9686441037496394, "1000–3000 Hz_10ms_modulation_mae_db": 1.452441027288368, "1000–3000 Hz_40ms_log_rms_mae_db": 1.506135414580557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1294167369536398, "1000–3000 Hz_40ms_envelope_correlation": 0.9890715474821961, "1000–3000 Hz_40ms_modulation_mae_db": 1.5455110913846195, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6758618712161453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23662798711016222, "3000–6000 Hz_10ms_envelope_correlation": 0.9591988517479071, "3000–6000 Hz_10ms_modulation_mae_db": 0.9936166871635294, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8196617098466688, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15569313216138436, "3000–6000 Hz_40ms_envelope_correlation": 0.9809462152403257, "3000–6000 Hz_40ms_modulation_mae_db": 1.3618087625595277, "mr_stft_db": 6.0152036376891616, "band_mae_db": 0.24641072048695525, "rms_gain_error_db": -0.0787334739098599, "output_peak": 0.6380742556824591}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1028526230558602, "全帯域_10ms_rms_envelope_nrmse": 0.07925239764275296, "全帯域_10ms_envelope_correlation": 0.9936446587527773, "全帯域_10ms_modulation_mae_db": 0.20551328937267774, "全帯域_40ms_log_rms_mae_db": 0.5695135130064922, "全帯域_40ms_rms_envelope_nrmse": 0.03993611275248167, "全帯域_40ms_envelope_correlation": 0.9982079124627489, "全帯域_40ms_modulation_mae_db": 0.3751704935723841, "300–1000 Hz_10ms_log_rms_mae_db": 1.942271121655714, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1410445461607229, "300–1000 Hz_10ms_envelope_correlation": 0.9848175960489021, "300–1000 Hz_10ms_modulation_mae_db": 1.2136388049020808, "300–1000 Hz_40ms_log_rms_mae_db": 1.101164678175904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08349186104284559, "300–1000 Hz_40ms_envelope_correlation": 0.9946091639584451, "300–1000 Hz_40ms_modulation_mae_db": 1.518726434168854, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4219178218075013, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2144950536275167, "1000–3000 Hz_10ms_envelope_correlation": 0.9655165993108241, "1000–3000 Hz_10ms_modulation_mae_db": 1.256594490756635, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5319649851395247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13287572350574567, "1000–3000 Hz_40ms_envelope_correlation": 0.9865891278335603, "1000–3000 Hz_40ms_modulation_mae_db": 1.2805635936472524, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6808273288897007, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23724072477194508, "3000–6000 Hz_10ms_envelope_correlation": 0.9589484159727906, "3000–6000 Hz_10ms_modulation_mae_db": 1.026566285597141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.822655730077711, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15804229859123972, "3000–6000 Hz_40ms_envelope_correlation": 0.9798112334714073, "3000–6000 Hz_40ms_modulation_mae_db": 0.8031605149352542, "mr_stft_db": 6.016362306134588, "band_mae_db": 0.2681001606505686, "rms_gain_error_db": -0.06185932764611832, "output_peak": 0.6431851105634804}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.155672856438737, "全帯域_10ms_rms_envelope_nrmse": 0.09432943689887681, "全帯域_10ms_envelope_correlation": 0.9909992343929573, "全帯域_10ms_modulation_mae_db": 0.40866516382210377, "全帯域_40ms_log_rms_mae_db": 0.6601864912331578, "全帯域_40ms_rms_envelope_nrmse": 0.05500955746589413, "全帯域_40ms_envelope_correlation": 0.9966018903006368, "全帯域_40ms_modulation_mae_db": 0.2889158331482765, "300–1000 Hz_10ms_log_rms_mae_db": 1.9474454695308414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14264565478962446, "300–1000 Hz_10ms_envelope_correlation": 0.9841762238479003, "300–1000 Hz_10ms_modulation_mae_db": 0.7486756627666615, "300–1000 Hz_40ms_log_rms_mae_db": 1.1585256687528158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08365683357957597, "300–1000 Hz_40ms_envelope_correlation": 0.9941508658277569, "300–1000 Hz_40ms_modulation_mae_db": 0.5423916032336694, "1000–3000 Hz_10ms_log_rms_mae_db": 2.513224800307669, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19661309534048213, "1000–3000 Hz_10ms_envelope_correlation": 0.9712650627615187, "1000–3000 Hz_10ms_modulation_mae_db": 1.101257103216648, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5648254187923096, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12010637279008611, "1000–3000 Hz_40ms_envelope_correlation": 0.989309177372732, "1000–3000 Hz_40ms_modulation_mae_db": 1.4375049424022708, "3000–6000 Hz_10ms_log_rms_mae_db": 2.709723487488766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23424508487277704, "3000–6000 Hz_10ms_envelope_correlation": 0.9602101572993289, "3000–6000 Hz_10ms_modulation_mae_db": 0.7821838820388045, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8629230780532553, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15939913065488404, "3000–6000 Hz_40ms_envelope_correlation": 0.9803134411731996, "3000–6000 Hz_40ms_modulation_mae_db": 0.890425039927038, "mr_stft_db": 6.024210628550549, "band_mae_db": 0.21892507107004258, "rms_gain_error_db": -0.09264565464648099, "output_peak": 0.6237715152774846}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1165725571762626, "全帯域_10ms_rms_envelope_nrmse": 0.09150668217113743, "全帯域_10ms_envelope_correlation": 0.9915054649220086, "全帯域_10ms_modulation_mae_db": 0.2894015440720447, "全帯域_40ms_log_rms_mae_db": 0.6265325536217365, "全帯域_40ms_rms_envelope_nrmse": 0.05732605582670971, "全帯域_40ms_envelope_correlation": 0.9962303076359478, "全帯域_40ms_modulation_mae_db": 0.3645624521228361, "300–1000 Hz_10ms_log_rms_mae_db": 1.8690926778989254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15081392958862788, "300–1000 Hz_10ms_envelope_correlation": 0.9823825946925949, "300–1000 Hz_10ms_modulation_mae_db": 0.640342712634762, "300–1000 Hz_40ms_log_rms_mae_db": 1.1084035246883766, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09522916550526257, "300–1000 Hz_40ms_envelope_correlation": 0.9925329260807779, "300–1000 Hz_40ms_modulation_mae_db": 0.7038733428903722, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4705035742703125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20853110707036454, "1000–3000 Hz_10ms_envelope_correlation": 0.9669109346950018, "1000–3000 Hz_10ms_modulation_mae_db": 1.1939005108032463, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5302373874000463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1196252414912713, "1000–3000 Hz_40ms_envelope_correlation": 0.9885545612231201, "1000–3000 Hz_40ms_modulation_mae_db": 1.2653372057795336, "3000–6000 Hz_10ms_log_rms_mae_db": 2.635305547926155, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2423534906192841, "3000–6000 Hz_10ms_envelope_correlation": 0.9583857470320352, "3000–6000 Hz_10ms_modulation_mae_db": 0.3239479507432581, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8333762371903877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16821729776297792, "3000–6000 Hz_40ms_envelope_correlation": 0.9777920543396575, "3000–6000 Hz_40ms_modulation_mae_db": 0.09673222401270433, "mr_stft_db": 6.011510002685486, "band_mae_db": 0.1939183331026394, "rms_gain_error_db": -0.03564257351270334, "output_peak": 0.6482204258809761}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1433126862168734, "全帯域_10ms_rms_envelope_nrmse": 0.09081058229362024, "全帯域_10ms_envelope_correlation": 0.9917479971222634, "全帯域_10ms_modulation_mae_db": 0.3943151197267105, "全帯域_40ms_log_rms_mae_db": 0.6388318398272814, "全帯域_40ms_rms_envelope_nrmse": 0.053289956001564116, "全帯域_40ms_envelope_correlation": 0.9969288189621509, "全帯域_40ms_modulation_mae_db": 0.34400732734678513, "300–1000 Hz_10ms_log_rms_mae_db": 1.9476158978051865, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1468274803685448, "300–1000 Hz_10ms_envelope_correlation": 0.9836184225307486, "300–1000 Hz_10ms_modulation_mae_db": 0.6157612565950066, "300–1000 Hz_40ms_log_rms_mae_db": 1.18296514643926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09211937725251253, "300–1000 Hz_40ms_envelope_correlation": 0.9934716603049012, "300–1000 Hz_40ms_modulation_mae_db": 1.0687526870040776, "1000–3000 Hz_10ms_log_rms_mae_db": 2.444900852095512, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20808676804698098, "1000–3000 Hz_10ms_envelope_correlation": 0.9686164634326712, "1000–3000 Hz_10ms_modulation_mae_db": 1.5192382840550016, "1000–3000 Hz_40ms_log_rms_mae_db": 1.578621954609153, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12831565165425685, "1000–3000 Hz_40ms_envelope_correlation": 0.989095305206992, "1000–3000 Hz_40ms_modulation_mae_db": 1.7413302397491073, "3000–6000 Hz_10ms_log_rms_mae_db": 2.655580689416205, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2470681449733168, "3000–6000 Hz_10ms_envelope_correlation": 0.9556992407805738, "3000–6000 Hz_10ms_modulation_mae_db": 1.4177496506755898, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8581434025635244, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1646353204489153, "3000–6000 Hz_40ms_envelope_correlation": 0.9792849726633046, "3000–6000 Hz_40ms_modulation_mae_db": 1.35783900641513, "mr_stft_db": 6.011973883471742, "band_mae_db": 0.32051128653944955, "rms_gain_error_db": -0.1251629825475911, "output_peak": 0.6125287542850323}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0990781909185405, "全帯域_10ms_rms_envelope_nrmse": 0.0852640998616164, "全帯域_10ms_envelope_correlation": 0.9927117210907284, "全帯域_10ms_modulation_mae_db": 0.5534171496060257, "全帯域_40ms_log_rms_mae_db": 0.6071751263041812, "全帯域_40ms_rms_envelope_nrmse": 0.04588522972601859, "全帯域_40ms_envelope_correlation": 0.9977190368740099, "全帯域_40ms_modulation_mae_db": 0.8278387972777786, "300–1000 Hz_10ms_log_rms_mae_db": 1.9465976690490108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12516527803778896, "300–1000 Hz_10ms_envelope_correlation": 0.9879502055812671, "300–1000 Hz_10ms_modulation_mae_db": 0.3443520058886018, "300–1000 Hz_40ms_log_rms_mae_db": 1.1358263786193386, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06987372207679375, "300–1000 Hz_40ms_envelope_correlation": 0.9960988693825238, "300–1000 Hz_40ms_modulation_mae_db": 0.5710181288767492, "1000–3000 Hz_10ms_log_rms_mae_db": 2.432735227388792, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2195908607840281, "1000–3000 Hz_10ms_envelope_correlation": 0.9635118384355907, "1000–3000 Hz_10ms_modulation_mae_db": 0.5986813947663355, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5340922815174065, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.129410097348707, "1000–3000 Hz_40ms_envelope_correlation": 0.9870663734280226, "1000–3000 Hz_40ms_modulation_mae_db": 0.8053528747213723, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7405874101857965, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23388675435941256, "3000–6000 Hz_10ms_envelope_correlation": 0.9605591629951187, "3000–6000 Hz_10ms_modulation_mae_db": 1.1105640846662255, "3000–6000 Hz_40ms_log_rms_mae_db": 1.848926001147863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16109868721008988, "3000–6000 Hz_40ms_envelope_correlation": 0.9802286247803768, "3000–6000 Hz_40ms_modulation_mae_db": 1.1654749718902948, "mr_stft_db": 6.00992080911652, "band_mae_db": 0.20829298557376552, "rms_gain_error_db": -0.1024301403703795, "output_peak": 0.6315748607857306}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1138884713924104, "全帯域_10ms_rms_envelope_nrmse": 0.08347187139415393, "全帯域_10ms_envelope_correlation": 0.9930932110010599, "全帯域_10ms_modulation_mae_db": 0.19540634992378905, "全帯域_40ms_log_rms_mae_db": 0.6054900970544066, "全帯域_40ms_rms_envelope_nrmse": 0.0505573980352442, "全帯域_40ms_envelope_correlation": 0.997305125213631, "全帯域_40ms_modulation_mae_db": 0.30193826116937444, "300–1000 Hz_10ms_log_rms_mae_db": 1.908359034111428, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13490691425124257, "300–1000 Hz_10ms_envelope_correlation": 0.9864480029008931, "300–1000 Hz_10ms_modulation_mae_db": 0.5814347855686562, "300–1000 Hz_40ms_log_rms_mae_db": 1.0894761464868703, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09100132770176482, "300–1000 Hz_40ms_envelope_correlation": 0.9938237053075426, "300–1000 Hz_40ms_modulation_mae_db": 0.837540434328916, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5025873829988607, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20512872707844332, "1000–3000 Hz_10ms_envelope_correlation": 0.9679463790344873, "1000–3000 Hz_10ms_modulation_mae_db": 1.213039939942932, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5497197718576967, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12366899887302626, "1000–3000 Hz_40ms_envelope_correlation": 0.9874900962636729, "1000–3000 Hz_40ms_modulation_mae_db": 1.4936939579806585, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6865357901541964, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2410596727499859, "3000–6000 Hz_10ms_envelope_correlation": 0.9578028325958495, "3000–6000 Hz_10ms_modulation_mae_db": 0.3663930824976771, "3000–6000 Hz_40ms_log_rms_mae_db": 1.784873553689471, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16063425763611297, "3000–6000 Hz_40ms_envelope_correlation": 0.9791569951821275, "3000–6000 Hz_40ms_modulation_mae_db": 0.48231012573747667, "mr_stft_db": 6.025033059064037, "band_mae_db": 0.20448448400697575, "rms_gain_error_db": -0.13931616421129478, "output_peak": 0.6295006957098437}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88480, "output_samples": 88480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.59809541188328, "spectral_convergence": 0.280308038845891, "spectral_convergence_status": "ok", "amplitude_floor": 2.1848847242256272e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.838758561695551, "spectral_convergence": 0.25330517664140645, "spectral_convergence_status": "ok", "amplitude_floor": 1.2036494394056804e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.638833433915231, "spectral_convergence": 0.29109696003392194, "spectral_convergence_status": "ok", "amplitude_floor": 7.153950066191215e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002200971739696214, "output_power": 0.0022189253174146454, "delta_db": 0.03528219104138861, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002201298666996261, "output_power": 0.002152164107078175, "delta_db": -0.09803586466654633, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011542423166681864, "output_power": 0.0010845200192602603, "delta_db": -0.27059419199064255, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019781669780676847, "output_power": 0.00016703758281186307, "delta_db": -0.7344875109841531, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.802033486419172e-05, "output_power": 5.3029380777365466e-05, "delta_db": -0.3906367509986309, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6630669693501258e-05, "output_power": 1.6603347906416936e-05, "delta_db": -0.00714069793333977, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2560295346024502, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10306360957753544, "pitch": null, "output_rms_dbfs": -22.4471376403835, "output_peak": 0.6112303525685826, "reference_rms_dbfs": -22.34407403080596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.025229135831354}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00024519594682554, "log_rms_mae_db": 1.0932422808860038, "rms_envelope_nrmse": 0.08022475201720725, "envelope_correlation": 0.9935448181230118, "rms_gain_error_db": -0.10305742285118968, "reference_mean_power": 0.00583423429533283, "output_mean_power": 0.005697418527858979}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.472320009859619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016537460501756134, "output_power": 0.0016400134023314488, "reference_active": true, "delta_db": -0.03621422769017802}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003778384775464376, "output_power": 0.00036269354511743424, "reference_active": true, "delta_db": -0.17766356156366003}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8517105494706513e-05, "output_power": 5.84624954422417e-05, "reference_active": true, "delta_db": 0.8098246587444666}], "active_band_count": 3, "active_band_mae_db": 0.34123414933276824}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00020002070942310007, "log_rms_mae_db": 0.5754163100231986, "rms_envelope_nrmse": 0.04111401749161085, "envelope_correlation": 0.9981915989197109, "rms_gain_error_db": -0.10304444268749274, "reference_mean_power": 0.005865946925483087, "output_mean_power": 0.0057284046016616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.084203911038529e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015941324207372327, "output_power": 0.0015768280393776309, "reference_active": true, "delta_db": -0.04740060328687943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018248750335661006, "output_power": 0.00017908899719981322, "reference_active": true, "delta_db": -0.08164224911512796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.552526527817181e-06, "output_power": 5.15661940891347e-06, "reference_active": true, "delta_db": 0.5411259378169488}], "active_band_count": 3, "active_band_mae_db": 0.22338959673965209}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00018690652054173922, "log_rms_mae_db": 1.8511952392052813, "rms_envelope_nrmse": 0.11597865734406529, "envelope_correlation": 0.9897284715186085, "rms_gain_error_db": -0.19511613840447578, "reference_mean_power": 0.0020929267743018004, "output_mean_power": 0.0020009784895274503}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 8.903808595053404e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005237826202169963, "output_power": 0.0005099472945329381, "reference_active": true, "delta_db": -0.11625791850486564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020282121471265822, "output_power": 0.00019308170532895722, "reference_active": true, "delta_db": -0.21372253489699483}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.57818052122161e-05, "output_power": 2.238861198098375e-05, "reference_active": true, "delta_db": -0.6128615327075744}], "active_band_count": 3, "active_band_mae_db": 0.31428066203647825}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00016950242057071555, "log_rms_mae_db": 1.082587858267384, "rms_envelope_nrmse": 0.07080158850800314, "envelope_correlation": 0.996025707959351, "rms_gain_error_db": -0.19511015846218363, "reference_mean_power": 0.002104315784196243, "output_mean_power": 0.0020118699177115846}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 7.681995008528509e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000538193996495516, "output_power": 0.0005259709167064875, "reference_active": true, "delta_db": -0.09977118213972903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010305287620923108, "output_power": 0.00010025764411823223, "reference_active": true, "delta_db": -0.11942622232499256}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.715218896542221e-06, "output_power": 1.9956134344129872e-06, "reference_active": true, "delta_db": -1.3372842851766762}], "active_band_count": 3, "active_band_mae_db": 0.5188272298804659}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 7.550458878286076e-05, "log_rms_mae_db": 2.463948482384671, "rms_envelope_nrmse": 0.20238581130659847, "envelope_correlation": 0.971148938379933, "rms_gain_error_db": -0.6662487515472222, "reference_mean_power": 0.00021315330376123098, "output_mean_power": 0.00018283830781897357}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 1.6207032137826829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010202948507054384, "output_power": 8.80373516186541e-05, "reference_active": true, "delta_db": -0.6405872540518055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.692530062160877e-05, "output_power": 3.0484898573177415e-05, "reference_active": true, "delta_db": -0.8323928532774261}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.752197568216375e-06, "output_power": 4.8753778487314155e-06, "reference_active": true, "delta_db": -1.414368614101054}], "active_band_count": 3, "active_band_mae_db": 0.9624495738100952}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 5.938009261210461e-05, "log_rms_mae_db": 1.5595810971277497, "rms_envelope_nrmse": 0.1308986508029206, "envelope_correlation": 0.9893091238286092, "rms_gain_error_db": -0.6660153832343658, "reference_mean_power": 0.00021425147153668813, "output_mean_power": 0.00018379016816403676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 1.4064708995422735e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010396272320640846, "output_power": 8.994369732068107e-05, "reference_active": true, "delta_db": -0.6290691074935427}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97231105181685e-05, "output_power": 1.651431504350217e-05, "reference_active": true, "delta_db": -0.7711484291766881}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.307963255185344e-07, "output_power": 3.727466605026122e-07, "reference_active": true, "delta_db": -2.284753936067116}], "active_band_count": 3, "active_band_mae_db": 1.2283238242457823}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 4.4291934004628345e-05, "log_rms_mae_db": 2.728303382877563, "rms_envelope_nrmse": 0.24561727991774687, "envelope_correlation": 0.9558770249848593, "rms_gain_error_db": -0.23396168513681564, "reference_mean_power": 2.6417005173666327e-05, "output_mean_power": 2.5031530978130826e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.694560310630941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.502841409588563e-05, "output_power": 1.3826478672679094e-05, "reference_active": true, "delta_db": -0.3620156511076331}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.7026465908103935e-06, "output_power": 6.018259547127164e-06, "reference_active": true, "delta_db": -0.46775407577394057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.011540321330325e-07, "output_power": 6.509169169701049e-07, "reference_active": true, "delta_db": -1.4127347149380483}], "active_band_count": 3, "active_band_mae_db": 0.7475014806065406}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 3.2524867537670264e-05, "log_rms_mae_db": 1.873116723503196, "rms_envelope_nrmse": 0.16006151742876695, "envelope_correlation": 0.9793766577001882, "rms_gain_error_db": -0.2339971005482191, "reference_mean_power": 2.65459352580292e-05, "output_mean_power": 2.5153494036496704e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.4345030346700597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6238306479668564e-05, "output_power": 1.4568261785752284e-05, "reference_active": true, "delta_db": -0.4713299701131053}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6930369104619132e-06, "output_power": 3.3616050526336635e-06, "reference_active": true, "delta_db": -0.4083696045121423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.897644209500044e-08, "output_power": 6.058433559477756e-08, "reference_active": true, "delta_db": -1.151372152748563}], "active_band_count": 3, "active_band_mae_db": 0.6770239091246036}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8845365377587068, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b8d3f26dabb1354db59a01d0f9c16afade22bf82dbcf21a7e557b16e0e4009cb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9679450399749583, "standard_deviation": 0.0026416172369079357, "minimum": 0.9635118384355907, "maximum": 0.9712650627615187, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.461329985558452, "standard_deviation": 0.032836693703402545, "minimum": 2.4219178218075013, "maximum": 2.513224800307669, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1622002905799076, "standard_deviation": 0.28918200394545235, "minimum": 0.5986813947663355, "maximum": 1.5192382840550016, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.207941211390684, "standard_deviation": 0.007037737345080724, "minimum": 0.19661309534048213, "maximum": 0.2195908607840281, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9883106640798631, "standard_deviation": 0.0010976155680954152, "minimum": 0.9865891278335603, "maximum": 0.989309177372732, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.5443972888780555, "standard_deviation": 0.023215580611199403, "minimum": 1.506135414580557, "maximum": 1.578621954609153, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3497022162388244, "standard_deviation": 0.27854291502464057, "minimum": 0.8053528747213723, "maximum": 1.7413302397491073, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1267896841774567, "standard_deviation": 0.0050107683546153515, "minimum": 0.1196252414912713, "maximum": 0.13287572350574567, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9583351791760579, "standard_deviation": 0.0018071398045543458, "minimum": 0.9556992407805738, "maximum": 0.9605591629951187, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.689090688519316, "standard_deviation": 0.035618682393862025, "minimum": 2.635305547926155, "maximum": 2.7405874101857965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8460653879985958, "standard_deviation": 0.37179530586548787, "minimum": 0.3239479507432581, "maximum": 1.4177496506755898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23976239242182884, "standard_deviation": 0.005031295823964845, "minimum": 0.23388675435941256, "maximum": 0.2470681449733168, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9796137743188233, "standard_deviation": 0.0009542094263342226, "minimum": 0.9777920543396575, "maximum": 0.9809462152403257, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8379595545090097, "standard_deviation": 0.0287725145526243, "minimum": 1.784873553689471, "maximum": 1.873116723503196, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8543468193252537, "standard_deviation": 0.4400615217147705, "minimum": 0.09673222401270433, "maximum": 1.3618087625595277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1609727052367964, "standard_deviation": 0.0038876519621513336, "minimum": 0.15569313216138436, "maximum": 0.16821729776297792, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9850689685661354, "standard_deviation": 0.0028115960888094086, "minimum": 0.981430231408168, "maximum": 0.9897284715186085, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.920195938870855, "standard_deviation": 0.0396784024760206, "minimum": 1.8511952392052813, "maximum": 1.9489904017104516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6060281305335002, "standard_deviation": 0.29075959828657005, "minimum": 0.31428066203647825, "maximum": 1.2136388049020808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13903651130906797, "standard_deviation": 0.013145070284326123, "minimum": 0.11597865734406529, "maximum": 0.15490962993192675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942398966477496, "standard_deviation": 0.001283672875656147, "minimum": 0.9925329260807779, "maximum": 0.9960988693825238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1315422181150414, "standard_deviation": 0.04282856059336261, "minimum": 1.082587858267384, "maximum": 1.1933883434903823, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7582457419338241, "standard_deviation": 0.3833197453837589, "minimum": 0.3048360750874889, "maximum": 1.518726434168854, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08473159332240038, "standard_deviation": 0.009782243570756598, "minimum": 0.06987372207679375, "maximum": 0.09522916550526257, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23958407200410586, "standard_deviation": 0.042092932780242745, "minimum": 0.1939183331026394, "maximum": 0.32051128653944955, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.01743043281793, "standard_deviation": 0.006457665784793917, "minimum": 6.00992080911652, "maximum": 6.025229135831354, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6297607463441988, "standard_deviation": 0.013467603089988381, "minimum": 0.6112303525685826, "maximum": 0.6482204258809761, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09235674080274542, "standard_deviation": 0.03344465196364891, "minimum": -0.13931616421129478, "maximum": -0.03564257351270334, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.992470306358816, "standard_deviation": 0.000970912768186043, "minimum": 0.9909992343929573, "maximum": 0.9936446587527773, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1139418623092425, "standard_deviation": 0.024242394755826478, "minimum": 1.0869152323892524, "maximum": 1.155672856438737, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3414049964256103, "standard_deviation": 0.11635085864061809, "minimum": 0.19540634992378905, "maximum": 0.5534171496060257, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08636084958325616, "standard_deviation": 0.005446104871888866, "minimum": 0.07925239764275296, "maximum": 0.09432943689887681, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973430457463344, "standard_deviation": 0.0007188139079786816, "minimum": 0.9962303076359478, "maximum": 0.9982079124627489, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6078349538563873, "standard_deviation": 0.032464236690911336, "minimum": 0.5695135130064922, "maximum": 0.6601864912331578, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3673579811565796, "standard_deviation": 0.19553231190511194, "minimum": 0.21304108787554943, "maximum": 0.8278387972777786, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04872249632904223, "standard_deviation": 0.006382605931332241, "minimum": 0.03993611275248167, "maximum": 0.05732605582670971, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0026", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.518401622772217, "predicted_mos": 2.518401622772217, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "f5e22f438448ac810f8eae6bd6d0f68a46da6ed95fc32f99497a7222697789f3", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8224193998612463, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0898832854397935, "全帯域_10ms_rms_envelope_nrmse": 0.10806445710170029, "全帯域_10ms_envelope_correlation": 0.9882435650575362, "全帯域_10ms_modulation_mae_db": 0.7273591689737401, "全帯域_40ms_log_rms_mae_db": 0.6149290811690236, "全帯域_40ms_rms_envelope_nrmse": 0.0657587774795448, "全帯域_40ms_envelope_correlation": 0.9952460387326169, "全帯域_40ms_modulation_mae_db": 0.9472308706839924, "300–1000 Hz_10ms_log_rms_mae_db": 2.065310804392717, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16364555583534127, "300–1000 Hz_10ms_envelope_correlation": 0.9794454477632434, "300–1000 Hz_10ms_modulation_mae_db": 1.0616777108691307, "300–1000 Hz_40ms_log_rms_mae_db": 1.2162320323492393, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11254215480889794, "300–1000 Hz_40ms_envelope_correlation": 0.9897723516719327, "300–1000 Hz_40ms_modulation_mae_db": 1.2817838746170456, "1000–3000 Hz_10ms_log_rms_mae_db": 2.692235041098009, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24235059852481913, "1000–3000 Hz_10ms_envelope_correlation": 0.954776318268896, "1000–3000 Hz_10ms_modulation_mae_db": 0.6345919063606388, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7067903785700507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14109030307803133, "1000–3000 Hz_40ms_envelope_correlation": 0.9832410673470523, "1000–3000 Hz_40ms_modulation_mae_db": 1.3437709215980471, "3000–6000 Hz_10ms_log_rms_mae_db": 3.2556383583434902, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3135969951411178, "3000–6000 Hz_10ms_envelope_correlation": 0.9286155626009418, "3000–6000 Hz_10ms_modulation_mae_db": 0.6537135960276376, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1834305853348184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19822101057236444, "3000–6000 Hz_40ms_envelope_correlation": 0.9685195312901266, "3000–6000 Hz_40ms_modulation_mae_db": 0.5775740329385061, "mr_stft_db": 6.451740625898126, "band_mae_db": 0.22022015984094256, "rms_gain_error_db": -0.1448641817420316, "output_peak": 0.6203227951755136}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.080983934476903, "全帯域_10ms_rms_envelope_nrmse": 0.0830984842461279, "全帯域_10ms_envelope_correlation": 0.9932748064308635, "全帯域_10ms_modulation_mae_db": 0.3660837517196509, "全帯域_40ms_log_rms_mae_db": 0.5888239557130572, "全帯域_40ms_rms_envelope_nrmse": 0.04792849693727147, "全帯域_40ms_envelope_correlation": 0.9977566876518786, "全帯域_40ms_modulation_mae_db": 0.7630394134673848, "300–1000 Hz_10ms_log_rms_mae_db": 2.0245523648264845, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.127772440204854, "300–1000 Hz_10ms_envelope_correlation": 0.9878836809715763, "300–1000 Hz_10ms_modulation_mae_db": 0.5833603981822311, "300–1000 Hz_40ms_log_rms_mae_db": 1.1423055536013584, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0760491993846842, "300–1000 Hz_40ms_envelope_correlation": 0.995939841601349, "300–1000 Hz_40ms_modulation_mae_db": 0.78602508423177, "1000–3000 Hz_10ms_log_rms_mae_db": 2.7523372827788966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2341382513730471, "1000–3000 Hz_10ms_envelope_correlation": 0.9579538390334839, "1000–3000 Hz_10ms_modulation_mae_db": 1.0268375544311878, "1000–3000 Hz_40ms_log_rms_mae_db": 1.69778782920019, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13421849376600403, "1000–3000 Hz_40ms_envelope_correlation": 0.9847062427937421, "1000–3000 Hz_40ms_modulation_mae_db": 1.610601633091049, "3000–6000 Hz_10ms_log_rms_mae_db": 3.253363452157792, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3071305533256711, "3000–6000 Hz_10ms_envelope_correlation": 0.9315787634707781, "3000–6000 Hz_10ms_modulation_mae_db": 0.5948081696348378, "3000–6000 Hz_40ms_log_rms_mae_db": 2.183409277965599, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18746668382331622, "3000–6000 Hz_40ms_envelope_correlation": 0.9720447773954751, "3000–6000 Hz_40ms_modulation_mae_db": 1.0446894830236586, "mr_stft_db": 6.452474486087227, "band_mae_db": 0.18132668454411127, "rms_gain_error_db": -0.1722439542427583, "output_peak": 0.6243692387893357}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1253717258925224, "全帯域_10ms_rms_envelope_nrmse": 0.09334635280854, "全帯域_10ms_envelope_correlation": 0.9911566448181358, "全帯域_10ms_modulation_mae_db": 0.4595640705490836, "全帯域_40ms_log_rms_mae_db": 0.5931098039397223, "全帯域_40ms_rms_envelope_nrmse": 0.04653337118525703, "全帯域_40ms_envelope_correlation": 0.997542313207715, "全帯域_40ms_modulation_mae_db": 0.30451038307997025, "300–1000 Hz_10ms_log_rms_mae_db": 2.0091240574372695, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1439535098721494, "300–1000 Hz_10ms_envelope_correlation": 0.9842798280060593, "300–1000 Hz_10ms_modulation_mae_db": 0.9069085529362938, "300–1000 Hz_40ms_log_rms_mae_db": 1.119449640349607, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08988667702316042, "300–1000 Hz_40ms_envelope_correlation": 0.9938013625851712, "300–1000 Hz_40ms_modulation_mae_db": 1.0933411156760888, "1000–3000 Hz_10ms_log_rms_mae_db": 2.774924414644321, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22921765198084798, "1000–3000 Hz_10ms_envelope_correlation": 0.9605656258378579, "1000–3000 Hz_10ms_modulation_mae_db": 1.498387674123001, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7039512522809113, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13649798515466124, "1000–3000 Hz_40ms_envelope_correlation": 0.9862072964817182, "1000–3000 Hz_40ms_modulation_mae_db": 1.8225788970882428, "3000–6000 Hz_10ms_log_rms_mae_db": 3.2143915359509716, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31180564485949264, "3000–6000 Hz_10ms_envelope_correlation": 0.9284731185495587, "3000–6000 Hz_10ms_modulation_mae_db": 0.8537028997742997, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1805413591159803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19777637104110038, "3000–6000 Hz_40ms_envelope_correlation": 0.9696281972120562, "3000–6000 Hz_40ms_modulation_mae_db": 0.944541549244963, "mr_stft_db": 6.444560822768508, "band_mae_db": 0.24559221346152157, "rms_gain_error_db": -0.061982603686398824, "output_peak": 0.8001100923221948}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.029594956935037, "全帯域_10ms_rms_envelope_nrmse": 0.08359158889779465, "全帯域_10ms_envelope_correlation": 0.9929253718119703, "全帯域_10ms_modulation_mae_db": 0.44936556996123284, "全帯域_40ms_log_rms_mae_db": 0.5744389411480498, "全帯域_40ms_rms_envelope_nrmse": 0.04076875764407341, "全帯域_40ms_envelope_correlation": 0.9981289453967713, "全帯域_40ms_modulation_mae_db": 0.692636017505016, "300–1000 Hz_10ms_log_rms_mae_db": 2.051183412730399, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13355234116231438, "300–1000 Hz_10ms_envelope_correlation": 0.9861411727459891, "300–1000 Hz_10ms_modulation_mae_db": 0.886524707051041, "300–1000 Hz_40ms_log_rms_mae_db": 1.2010948176928136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07255417249774265, "300–1000 Hz_40ms_envelope_correlation": 0.9955706085490154, "300–1000 Hz_40ms_modulation_mae_db": 1.291592389766381, "1000–3000 Hz_10ms_log_rms_mae_db": 2.7171369053018544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23944257515130524, "1000–3000 Hz_10ms_envelope_correlation": 0.955992164480018, "1000–3000 Hz_10ms_modulation_mae_db": 0.6788045684056577, "1000–3000 Hz_40ms_log_rms_mae_db": 1.713181948854742, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13889824533750897, "1000–3000 Hz_40ms_envelope_correlation": 0.9834452353657253, "1000–3000 Hz_40ms_modulation_mae_db": 1.0429376992459491, "3000–6000 Hz_10ms_log_rms_mae_db": 3.199812018398279, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30391354501720286, "3000–6000 Hz_10ms_envelope_correlation": 0.9322750321408133, "3000–6000 Hz_10ms_modulation_mae_db": 0.9954851617176627, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1277210628914625, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19001573907842073, "3000–6000 Hz_40ms_envelope_correlation": 0.9718927927705953, "3000–6000 Hz_40ms_modulation_mae_db": 0.9474138359124891, "mr_stft_db": 6.4474498037073635, "band_mae_db": 0.128069745398228, "rms_gain_error_db": -0.060109891220674144, "output_peak": 0.6333864510781345}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.109859674336452, "全帯域_10ms_rms_envelope_nrmse": 0.08295844558337938, "全帯域_10ms_envelope_correlation": 0.993254728263666, "全帯域_10ms_modulation_mae_db": 0.7157582352605347, "全帯域_40ms_log_rms_mae_db": 0.5868505527896241, "全帯域_40ms_rms_envelope_nrmse": 0.055526008674726894, "全帯域_40ms_envelope_correlation": 0.9967568508612189, "全帯域_40ms_modulation_mae_db": 0.9716728716391874, "300–1000 Hz_10ms_log_rms_mae_db": 1.9672711107588465, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13773767551826196, "300–1000 Hz_10ms_envelope_correlation": 0.9857015806309621, "300–1000 Hz_10ms_modulation_mae_db": 1.0037816227363046, "300–1000 Hz_40ms_log_rms_mae_db": 1.1241490177941096, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.086137260351354, "300–1000 Hz_40ms_envelope_correlation": 0.9943752932465098, "300–1000 Hz_40ms_modulation_mae_db": 1.289158691357147, "1000–3000 Hz_10ms_log_rms_mae_db": 2.7753853008096336, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24605807132289975, "1000–3000 Hz_10ms_envelope_correlation": 0.9533455945433138, "1000–3000 Hz_10ms_modulation_mae_db": 1.1303119249910114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.7474667621269797, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13912039899249007, "1000–3000 Hz_40ms_envelope_correlation": 0.9837800990985606, "1000–3000 Hz_40ms_modulation_mae_db": 1.5627649549225013, "3000–6000 Hz_10ms_log_rms_mae_db": 3.217984939954238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30029032657119864, "3000–6000 Hz_10ms_envelope_correlation": 0.9361740447111149, "3000–6000 Hz_10ms_modulation_mae_db": 0.2011148518672823, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1456624577895966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1860236764975243, "3000–6000 Hz_40ms_envelope_correlation": 0.9730345224082925, "3000–6000 Hz_40ms_modulation_mae_db": 0.542308664123175, "mr_stft_db": 6.436256963926063, "band_mae_db": 0.23072812847935434, "rms_gain_error_db": -0.14389173925298457, "output_peak": 0.6066758927414695}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.077387078132473, "全帯域_10ms_rms_envelope_nrmse": 0.09594840686255576, "全帯域_10ms_envelope_correlation": 0.9906767657678052, "全帯域_10ms_modulation_mae_db": 0.4852767079078693, "全帯域_40ms_log_rms_mae_db": 0.5890142555650545, "全帯域_40ms_rms_envelope_nrmse": 0.05195213158497961, "全帯域_40ms_envelope_correlation": 0.9969702940252153, "全帯域_40ms_modulation_mae_db": 0.48076818375903363, "300–1000 Hz_10ms_log_rms_mae_db": 2.008357492663554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1641439509055768, "300–1000 Hz_10ms_envelope_correlation": 0.978946976590469, "300–1000 Hz_10ms_modulation_mae_db": 0.4500740985122347, "300–1000 Hz_40ms_log_rms_mae_db": 1.1784808388659274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09683249693912672, "300–1000 Hz_40ms_envelope_correlation": 0.9920646466199058, "300–1000 Hz_40ms_modulation_mae_db": 0.47135237508057287, "1000–3000 Hz_10ms_log_rms_mae_db": 2.710302182625007, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2274013187992207, "1000–3000 Hz_10ms_envelope_correlation": 0.9605231243616236, "1000–3000 Hz_10ms_modulation_mae_db": 1.269486784722801, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6708742519182225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1253485666920153, "1000–3000 Hz_40ms_envelope_correlation": 0.9866165762047465, "1000–3000 Hz_40ms_modulation_mae_db": 1.4283752701853738, "3000–6000 Hz_10ms_log_rms_mae_db": 3.252421995109382, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30896668990822757, "3000–6000 Hz_10ms_envelope_correlation": 0.9313655370152275, "3000–6000 Hz_10ms_modulation_mae_db": 0.5760101249473267, "3000–6000 Hz_40ms_log_rms_mae_db": 2.208309529170623, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19294286712569167, "3000–6000 Hz_40ms_envelope_correlation": 0.9704523509657054, "3000–6000 Hz_40ms_modulation_mae_db": 0.6838657806381674, "mr_stft_db": 6.443247656430782, "band_mae_db": 0.1236330455717544, "rms_gain_error_db": -0.08319931667718122, "output_peak": 0.6649603728587641}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0592487631151077, "全帯域_10ms_rms_envelope_nrmse": 0.08762820307411134, "全帯域_10ms_envelope_correlation": 0.992206551220119, "全帯域_10ms_modulation_mae_db": 0.176849256470448, "全帯域_40ms_log_rms_mae_db": 0.5654099754281312, "全帯域_40ms_rms_envelope_nrmse": 0.047992109723379854, "全帯域_40ms_envelope_correlation": 0.9973619041356886, "全帯域_40ms_modulation_mae_db": 0.30378728986227527, "300–1000 Hz_10ms_log_rms_mae_db": 1.9879361527044597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13808905993344742, "300–1000 Hz_10ms_envelope_correlation": 0.9851442688777499, "300–1000 Hz_10ms_modulation_mae_db": 0.28580572468469867, "300–1000 Hz_40ms_log_rms_mae_db": 1.139551768746049, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07666485444994282, "300–1000 Hz_40ms_envelope_correlation": 0.9950354515446438, "300–1000 Hz_40ms_modulation_mae_db": 0.29561670159127695, "1000–3000 Hz_10ms_log_rms_mae_db": 2.679021093568453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23767690346385145, "1000–3000 Hz_10ms_envelope_correlation": 0.9575637289997235, "1000–3000 Hz_10ms_modulation_mae_db": 1.5171771011394235, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6761810067532603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14280739247334234, "1000–3000 Hz_40ms_envelope_correlation": 0.9848761212635314, "1000–3000 Hz_40ms_modulation_mae_db": 1.6840016555794382, "3000–6000 Hz_10ms_log_rms_mae_db": 3.2263167740883496, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31533526182254434, "3000–6000 Hz_10ms_envelope_correlation": 0.9271064528943316, "3000–6000 Hz_10ms_modulation_mae_db": 1.6224990871526181, "3000–6000 Hz_40ms_log_rms_mae_db": 2.190108555260547, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21276923120844343, "3000–6000 Hz_40ms_envelope_correlation": 0.9646875056314914, "3000–6000 Hz_40ms_modulation_mae_db": 1.6072737130872579, "mr_stft_db": 6.455867338127991, "band_mae_db": 0.22341925780464356, "rms_gain_error_db": -0.04644828982795935, "output_peak": 0.7676272271958682}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0748069422703244, "全帯域_10ms_rms_envelope_nrmse": 0.09749598086552047, "全帯域_10ms_envelope_correlation": 0.9903587976576586, "全帯域_10ms_modulation_mae_db": 0.5374574111644396, "全帯域_40ms_log_rms_mae_db": 0.5654885073128124, "全帯域_40ms_rms_envelope_nrmse": 0.054607637502586424, "全帯域_40ms_envelope_correlation": 0.9965579667788748, "全帯域_40ms_modulation_mae_db": 0.5871375716152318, "300–1000 Hz_10ms_log_rms_mae_db": 1.989280705026036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13403786634379664, "300–1000 Hz_10ms_envelope_correlation": 0.9860308262409141, "300–1000 Hz_10ms_modulation_mae_db": 0.3998316867473046, "300–1000 Hz_40ms_log_rms_mae_db": 1.149199305226395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08218888629383891, "300–1000 Hz_40ms_envelope_correlation": 0.9943119146614982, "300–1000 Hz_40ms_modulation_mae_db": 0.31505063410073225, "1000–3000 Hz_10ms_log_rms_mae_db": 2.7398998300360353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2417315152419321, "1000–3000 Hz_10ms_envelope_correlation": 0.9554810049326581, "1000–3000 Hz_10ms_modulation_mae_db": 1.6821785062017203, "1000–3000 Hz_40ms_log_rms_mae_db": 1.781081890670807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14636852133393394, "1000–3000 Hz_40ms_envelope_correlation": 0.9829805077821222, "1000–3000 Hz_40ms_modulation_mae_db": 2.097604830152385, "3000–6000 Hz_10ms_log_rms_mae_db": 3.2274425663340534, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2902520249521043, "3000–6000 Hz_10ms_envelope_correlation": 0.9388612978915862, "3000–6000 Hz_10ms_modulation_mae_db": 0.6389470387165824, "3000–6000 Hz_40ms_log_rms_mae_db": 2.177982432019378, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17914724594815995, "3000–6000 Hz_40ms_envelope_correlation": 0.9744104367462689, "3000–6000 Hz_40ms_modulation_mae_db": 0.36905546445814913, "mr_stft_db": 6.453876586910829, "band_mae_db": 0.16787516165842234, "rms_gain_error_db": -0.012041679495166088, "output_peak": 0.784438628230611}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88480, "output_samples": 88480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.132001089614744, "spectral_convergence": 0.29146056733806286, "spectral_convergence_status": "ok", "amplitude_floor": 2.1848847242256272e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.324387571690565, "spectral_convergence": 0.2706586514286883, "spectral_convergence_status": "ok", "amplitude_floor": 1.2036494394056804e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.898833216389072, "spectral_convergence": 0.27340414949460845, "spectral_convergence_status": "ok", "amplitude_floor": 7.153950066191215e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002200971739696214, "output_power": 0.002177556091973466, "delta_db": -0.04645115405810765, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002201298666996261, "output_power": 0.0021047734804105762, "delta_db": -0.1947360748262696, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011542423166681864, "output_power": 0.001098095272773026, "delta_db": -0.21656970367489303, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019781669780676847, "output_power": 0.000185900314015454, "delta_db": -0.26982824458118515, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.802033486419172e-05, "output_power": 5.376195638232234e-05, "delta_db": -0.3310516681703606, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6630669693501258e-05, "output_power": 1.7667624052095714e-05, "delta_db": 0.2626841137348394, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22022015984094256, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1448641817420316, "pitch": null, "output_rms_dbfs": -22.488938212547993, "output_peak": 0.6203227951755136, "reference_rms_dbfs": -22.34407403080596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.451740625898126}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00024519594682554, "log_rms_mae_db": 1.0898832854397935, "rms_envelope_nrmse": 0.10806445710170029, "envelope_correlation": 0.9882435650575362, "rms_gain_error_db": -0.1448646672206358, "reference_mean_power": 0.00583423429533283, "output_mean_power": 0.00564283561984156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.472320009859619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016537460501756134, "output_power": 0.0016052251462349478, "reference_active": true, "delta_db": -0.12932865481528902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003778384775464376, "output_power": 0.0002883316740177087, "reference_active": true, "delta_db": -1.1741382919855177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8517105494706513e-05, "output_power": 3.963089015309314e-05, "reference_active": true, "delta_db": -0.8786105601204136}], "active_band_count": 3, "active_band_mae_db": 0.7273591689737401}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00020002070942310007, "log_rms_mae_db": 0.6149290811690236, "rms_envelope_nrmse": 0.0657587774795448, "envelope_correlation": 0.9952460387326169, "rms_gain_error_db": -0.144852804400729, "reference_mean_power": 0.005865946925483087, "output_mean_power": 0.005673523378473411}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.084203911038529e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015941324207372327, "output_power": 0.0015354671925723504, "reference_active": true, "delta_db": -0.1628385279166309}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018248750335661006, "output_power": 0.00013661392810603376, "reference_active": true, "delta_db": -1.2573815063466107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.552526527817181e-06, "output_power": 3.2817479857197232e-06, "reference_active": true, "delta_db": -1.4214725777887356}], "active_band_count": 3, "active_band_mae_db": 0.9472308706839924}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00018690652054173922, "log_rms_mae_db": 2.065310804392717, "rms_envelope_nrmse": 0.16364555583534127, "envelope_correlation": 0.9794454477632434, "rms_gain_error_db": -0.3022001305945544, "reference_mean_power": 0.0020929267743018004, "output_mean_power": 0.0019522436585367483}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 8.903808595053404e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005237826202169963, "output_power": 0.0004878240099023832, "reference_active": true, "delta_db": -0.30887912107099014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020282121471265822, "output_power": 0.00014731356331307497, "reference_active": true, "delta_db": -1.388706447842282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.57818052122161e-05, "output_power": 1.830495082474103e-05, "reference_active": true, "delta_db": -1.4874475636941198}], "active_band_count": 3, "active_band_mae_db": 1.0616777108691307}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00016950242057071555, "log_rms_mae_db": 1.2162320323492393, "rms_envelope_nrmse": 0.11254215480889794, "envelope_correlation": 0.9897723516719327, "rms_gain_error_db": -0.3021965676628623, "reference_mean_power": 0.002104315784196243, "output_mean_power": 0.0019628687281351863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 7.681995008528509e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000538193996495516, "output_power": 0.0004930490661944103, "reference_active": true, "delta_db": -0.38048708329610725}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010305287620923108, "output_power": 7.241621636975116e-05, "reference_active": true, "delta_db": -1.5322428756884823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.715218896542221e-06, "output_power": 1.7399736882581982e-06, "reference_active": true, "delta_db": -1.9326216648665469}], "active_band_count": 3, "active_band_mae_db": 1.2817838746170456}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 7.550458878286076e-05, "log_rms_mae_db": 2.692235041098009, "rms_envelope_nrmse": 0.24235059852481913, "envelope_correlation": 0.954776318268896, "rms_gain_error_db": -0.27156998691994494, "reference_mean_power": 0.00021315330376123098, "output_mean_power": 0.00020023273048078632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 1.6207032137826829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010202948507054384, "output_power": 9.528973567643575e-05, "reference_active": true, "delta_db": -0.29679572567451973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.692530062160877e-05, "output_power": 3.442510223007414e-05, "reference_active": true, "delta_db": -0.30448801111504764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.752197568216375e-06, "output_power": 5.0026018912988675e-06, "reference_active": true, "delta_db": -1.302491982292349}], "active_band_count": 3, "active_band_mae_db": 0.6345919063606388}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 5.938009261210461e-05, "log_rms_mae_db": 1.7067903785700507, "rms_envelope_nrmse": 0.14109030307803133, "envelope_correlation": 0.9832410673470523, "rms_gain_error_db": -0.2713666213358545, "reference_mean_power": 0.00021425147153668813, "output_mean_power": 0.00020127375610054873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 1.4064708995422735e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010396272320640846, "output_power": 9.721986881314745e-05, "reference_active": true, "delta_db": -0.2912261620692374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97231105181685e-05, "output_power": 1.8194344160102217e-05, "reference_active": true, "delta_db": -0.35039002821441745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.307963255185344e-07, "output_power": 2.890144103680486e-07, "reference_active": true, "delta_db": -3.389696574510487}], "active_band_count": 3, "active_band_mae_db": 1.3437709215980471}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 4.4291934004628345e-05, "log_rms_mae_db": 3.2556383583434902, "rms_envelope_nrmse": 0.3135969951411178, "envelope_correlation": 0.9286155626009418, "rms_gain_error_db": -0.042637725574532326, "reference_mean_power": 2.6417005173666327e-05, "output_mean_power": 2.6158919939532733e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.694560310630941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.502841409588563e-05, "output_power": 1.2916563774779957e-05, "reference_active": true, "delta_db": -0.6576616059631031}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.7026465908103935e-06, "output_power": 5.499367661665168e-06, "reference_active": true, "delta_db": -0.859335655099033}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.011540321330325e-07, "output_power": 9.981903866521697e-07, "reference_active": true, "delta_db": 0.44414352702077664}], "active_band_count": 3, "active_band_mae_db": 0.6537135960276376}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 3.2524867537670264e-05, "log_rms_mae_db": 2.1834305853348184, "rms_envelope_nrmse": 0.19822101057236444, "envelope_correlation": 0.9685195312901266, "rms_gain_error_db": -0.04227490151082674, "reference_mean_power": 2.65459352580292e-05, "output_mean_power": 2.6288786581422514e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.4345030346700597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6238306479668564e-05, "output_power": 1.3674523386364865e-05, "reference_active": true, "delta_db": -0.746285356515799}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6930369104619132e-06, "output_power": 3.1521151340441943e-06, "reference_active": true, "delta_db": -0.6878157614455795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.897644209500044e-08, "output_power": 8.459791564680621e-08, "reference_active": true, "delta_db": 0.29862098085413963}], "active_band_count": 3, "active_band_mae_db": 0.5775740329385061}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8563447103879012, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4351182c9af1c0122b4b8474b5dac7a7d195ead48c016ed370402b09afdd4b90", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9570251750571969, "standard_deviation": 0.002617827344383208, "minimum": 0.9533455945433138, "maximum": 0.9605656258378579, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.7301552563577767, "standard_deviation": 0.036357944302816726, "minimum": 2.679021093568453, "maximum": 2.7753853008096336, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1797220025469302, "standard_deviation": 0.38726958418166235, "minimum": 0.6345919063606388, "maximum": 1.6821785062017203, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23725211073224045, "standard_deviation": 0.00654511723061106, "minimum": 0.2274013187992207, "maximum": 0.24605807132289975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9844816432921498, "standard_deviation": 0.001368033908232229, "minimum": 0.9829805077821222, "maximum": 0.9866165762047465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.7121644150468955, "standard_deviation": 0.03644881990189779, "minimum": 1.6708742519182225, "maximum": 1.781081890670807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5740794827328732, "standard_deviation": 0.3175258962368493, "minimum": 1.0429376992459491, "maximum": 2.097604830152385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1380437383534984, "standard_deviation": 0.006339835304342291, "minimum": 0.1253485666920153, "maximum": 0.14636852133393394, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9318062261592941, "standard_deviation": 0.004012398825014074, "minimum": 0.9271064528943316, "maximum": 0.9388612978915862, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 3.2309214550420693, "standard_deviation": 0.0207658243135145, "minimum": 3.199812018398279, "maximum": 3.2556383583434902, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7670351162297809, "standard_deviation": 0.4152772000416375, "minimum": 0.2011148518672823, "maximum": 1.6224990871526181, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30641138019969494, "standard_deviation": 0.008213619998482283, "minimum": 0.2902520249521043, "maximum": 0.31533526182254434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9705837643025015, "standard_deviation": 0.0030344463897855863, "minimum": 0.9646875056314914, "maximum": 0.9744104367462689, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.174645657443501, "standard_deviation": 0.025674251331702258, "minimum": 2.1277210628914625, "maximum": 2.208309529170623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8395903154282958, "standard_deviation": 0.3885368579398381, "minimum": 0.36905546445814913, "maximum": 1.6072737130872579, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19304535316187765, "standard_deviation": 0.010148703427208725, "minimum": 0.17914724594815995, "maximum": 0.21276923120844343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9841967227283703, "standard_deviation": 0.0032523020080050553, "minimum": 0.978946976590469, "maximum": 0.9878836809715763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0128770125674706, "standard_deviation": 0.03305026021387326, "minimum": 1.9672711107588465, "maximum": 2.065310804392717, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6972455627149048, "standard_deviation": 0.3019843013904768, "minimum": 0.28580572468469867, "maximum": 1.0616777108691307, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14286654997196774, "standard_deviation": 0.013770249563212806, "minimum": 0.127772440204854, "maximum": 0.1641439509055768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9938589338100032, "standard_deviation": 0.0020367675681033535, "minimum": 0.9897723516719327, "maximum": 0.995939841601349, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1588078718281873, "standard_deviation": 0.03577409427257371, "minimum": 1.119449640349607, "maximum": 1.2162320323492393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8529901083026268, "standard_deviation": 0.4431096867119674, "minimum": 0.29561670159127695, "maximum": 1.291592389766381, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08660696271859344, "standard_deviation": 0.01317728066904919, "minimum": 0.07255417249774265, "maximum": 0.11254215480889794, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19010804959487226, "standard_deviation": 0.04718345661964786, "minimum": 0.1236330455717544, "maximum": 0.24559221346152157, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.448184285482111, "standard_deviation": 0.0065768557762112415, "minimum": 6.436256963926063, "maximum": 6.455867338127991, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6877363372989864, "standard_deviation": 0.08191038079457451, "minimum": 0.6066758927414695, "maximum": 0.8001100923221948, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09059770701814426, "standard_deviation": 0.056536336691569294, "minimum": -0.1722439542427583, "maximum": -0.012041679495166088, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9915121538784692, "standard_deviation": 0.0017509607788662143, "minimum": 0.9882435650575362, "maximum": 0.9932748064308635, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0808920450748265, "standard_deviation": 0.029392789622781083, "minimum": 1.029594956935037, "maximum": 1.1253717258925224, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4897142715008749, "standard_deviation": 0.17951368654044061, "minimum": 0.176849256470448, "maximum": 0.7273591689737401, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09151648992996622, "standard_deviation": 0.008901088127633844, "minimum": 0.08295844558337938, "maximum": 0.10806445710170029, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970401250987475, "standard_deviation": 0.0008937137114613429, "minimum": 0.9952460387326169, "maximum": 0.9981289453967713, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5847581341331844, "standard_deviation": 0.01634981708560691, "minimum": 0.5654099754281312, "maximum": 0.6149290811690236, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6313478252015114, "standard_deviation": 0.26061727423706427, "minimum": 0.30378728986227527, "maximum": 0.9716728716391874, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05138341134147744, "standard_deviation": 0.007503927201904277, "minimum": 0.04076875764407341, "maximum": 0.0657587774795448, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0026", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.728555917739868, "predicted_mos": 2.728555917739868, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "672c74cbf4d582abaaa054695e7232af83799e4211573f3622c5b9b5968e506b", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.798051900230348, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1512005473087983, "全帯域_10ms_rms_envelope_nrmse": 0.09093038614063244, "全帯域_10ms_envelope_correlation": 0.9916558342071772, "全帯域_10ms_modulation_mae_db": 0.4369468435015351, "全帯域_40ms_log_rms_mae_db": 0.6331578000282363, "全帯域_40ms_rms_envelope_nrmse": 0.054888176804903197, "全帯域_40ms_envelope_correlation": 0.9966318165089134, "全帯域_40ms_modulation_mae_db": 0.729806580340345, "300–1000 Hz_10ms_log_rms_mae_db": 1.853648654446914, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.153281106731306, "300–1000 Hz_10ms_envelope_correlation": 0.982124150032818, "300–1000 Hz_10ms_modulation_mae_db": 1.087885165342775, "300–1000 Hz_40ms_log_rms_mae_db": 1.1059313055147362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10639392734076764, "300–1000 Hz_40ms_envelope_correlation": 0.9910169622665127, "300–1000 Hz_40ms_modulation_mae_db": 1.3798002243711733, "1000–3000 Hz_10ms_log_rms_mae_db": 2.307633800217155, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20741044053893407, "1000–3000 Hz_10ms_envelope_correlation": 0.9698011632714165, "1000–3000 Hz_10ms_modulation_mae_db": 1.3128976709823992, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4958614574957718, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12978875287038047, "1000–3000 Hz_40ms_envelope_correlation": 0.9900943706645847, "1000–3000 Hz_40ms_modulation_mae_db": 1.4992353279329702, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4186521739234337, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24124401252118333, "3000–6000 Hz_10ms_envelope_correlation": 0.9588025001809181, "3000–6000 Hz_10ms_modulation_mae_db": 0.4117221907635922, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6797217569733753, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15793217830653772, "3000–6000 Hz_40ms_envelope_correlation": 0.9804020866301104, "3000–6000 Hz_40ms_modulation_mae_db": 0.3063153262098168, "mr_stft_db": 5.9236360282273814, "band_mae_db": 0.20649192947681538, "rms_gain_error_db": -0.09852292289475098, "output_peak": 0.6294882549422605}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.106880165666878, "全帯域_10ms_rms_envelope_nrmse": 0.08053908928385324, "全帯域_10ms_envelope_correlation": 0.9934334658649712, "全帯域_10ms_modulation_mae_db": 0.395531338075325, "全帯域_40ms_log_rms_mae_db": 0.5729344707823755, "全帯域_40ms_rms_envelope_nrmse": 0.041462388560226306, "全帯域_40ms_envelope_correlation": 0.9980424212343526, "全帯域_40ms_modulation_mae_db": 0.5758957147533167, "300–1000 Hz_10ms_log_rms_mae_db": 1.8546906774656564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12823406428000975, "300–1000 Hz_10ms_envelope_correlation": 0.9872199816119408, "300–1000 Hz_10ms_modulation_mae_db": 1.0429511760618508, "300–1000 Hz_40ms_log_rms_mae_db": 1.148175806468647, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07663629233062616, "300–1000 Hz_40ms_envelope_correlation": 0.9950343306876871, "300–1000 Hz_40ms_modulation_mae_db": 1.2333785505047652, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2189760933486893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21067671397062387, "1000–3000 Hz_10ms_envelope_correlation": 0.9669284072734412, "1000–3000 Hz_10ms_modulation_mae_db": 1.4246811695524688, "1000–3000 Hz_40ms_log_rms_mae_db": 1.42173568116263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11756229308806213, "1000–3000 Hz_40ms_envelope_correlation": 0.9903170020059129, "1000–3000 Hz_40ms_modulation_mae_db": 1.5774821978168438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3453292630592966, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22205355766370333, "3000–6000 Hz_10ms_envelope_correlation": 0.9644574834694456, "3000–6000 Hz_10ms_modulation_mae_db": 0.46359572255741077, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7269662879055379, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1565102135855508, "3000–6000 Hz_40ms_envelope_correlation": 0.9802452935849475, "3000–6000 Hz_40ms_modulation_mae_db": 0.6727034331762459, "mr_stft_db": 5.8944282129229935, "band_mae_db": 0.18794217292669632, "rms_gain_error_db": -0.0377925537913648, "output_peak": 0.6733328392633451}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0671867511889026, "全帯域_10ms_rms_envelope_nrmse": 0.07523471963458543, "全帯域_10ms_envelope_correlation": 0.9942834647313109, "全帯域_10ms_modulation_mae_db": 0.38269929247119466, "全帯域_40ms_log_rms_mae_db": 0.5683037819493675, "全帯域_40ms_rms_envelope_nrmse": 0.04317521301744386, "全帯域_40ms_envelope_correlation": 0.9978932014806378, "全帯域_40ms_modulation_mae_db": 0.5584066903375607, "300–1000 Hz_10ms_log_rms_mae_db": 1.7954406609935847, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13101157793562776, "300–1000 Hz_10ms_envelope_correlation": 0.9868311700552878, "300–1000 Hz_10ms_modulation_mae_db": 0.6965288809859569, "300–1000 Hz_40ms_log_rms_mae_db": 1.0611039622911842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0788293764132992, "300–1000 Hz_40ms_envelope_correlation": 0.9950499465994719, "300–1000 Hz_40ms_modulation_mae_db": 0.6515697745711999, "1000–3000 Hz_10ms_log_rms_mae_db": 2.230485119807568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1975759892845277, "1000–3000 Hz_10ms_envelope_correlation": 0.9704036721606968, "1000–3000 Hz_10ms_modulation_mae_db": 0.9561050100683506, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4419861197339365, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1160718519831964, "1000–3000 Hz_40ms_envelope_correlation": 0.9891888512243336, "1000–3000 Hz_40ms_modulation_mae_db": 1.287066876995554, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4107414548737247, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2537471310757032, "3000–6000 Hz_10ms_envelope_correlation": 0.955138960079155, "3000–6000 Hz_10ms_modulation_mae_db": 0.31269348193488783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6971229942462323, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16744749073184212, "3000–6000 Hz_40ms_envelope_correlation": 0.9784388508731758, "3000–6000 Hz_40ms_modulation_mae_db": 0.6344704053226723, "mr_stft_db": 5.881716090082335, "band_mae_db": 0.12823443772307547, "rms_gain_error_db": -0.06123056211210086, "output_peak": 0.7005473009474108}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1179782058085068, "全帯域_10ms_rms_envelope_nrmse": 0.08393329567126906, "全帯域_10ms_envelope_correlation": 0.9928508676810183, "全帯域_10ms_modulation_mae_db": 0.39776542625666816, "全帯域_40ms_log_rms_mae_db": 0.6058021000916222, "全帯域_40ms_rms_envelope_nrmse": 0.04499685405041599, "全帯域_40ms_envelope_correlation": 0.9976782176743769, "全帯域_40ms_modulation_mae_db": 0.16989750412222668, "300–1000 Hz_10ms_log_rms_mae_db": 1.77690379135315, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1284670330842204, "300–1000 Hz_10ms_envelope_correlation": 0.9871503970934588, "300–1000 Hz_10ms_modulation_mae_db": 0.2209957094630501, "300–1000 Hz_40ms_log_rms_mae_db": 1.000268476459319, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07452830010488756, "300–1000 Hz_40ms_envelope_correlation": 0.9952728268987968, "300–1000 Hz_40ms_modulation_mae_db": 0.06025494963002539, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1762164944277127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19193723380691832, "1000–3000 Hz_10ms_envelope_correlation": 0.9720774065593034, "1000–3000 Hz_10ms_modulation_mae_db": 0.16576572028940809, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4195589993412947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11375769792262995, "1000–3000 Hz_40ms_envelope_correlation": 0.9896118467938154, "1000–3000 Hz_40ms_modulation_mae_db": 0.2843929217030116, "3000–6000 Hz_10ms_log_rms_mae_db": 2.41832709947143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22671609521925898, "3000–6000 Hz_10ms_envelope_correlation": 0.9644030144404693, "3000–6000 Hz_10ms_modulation_mae_db": 0.2886548832674637, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7036127785551995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16297589537422844, "3000–6000 Hz_40ms_envelope_correlation": 0.9798649796109296, "3000–6000 Hz_40ms_modulation_mae_db": 0.3553652584751192, "mr_stft_db": 5.907268313017781, "band_mae_db": 0.1350483820906067, "rms_gain_error_db": -0.038743580786076884, "output_peak": 0.7260476311229502}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0914817101874759, "全帯域_10ms_rms_envelope_nrmse": 0.0885387130905133, "全帯域_10ms_envelope_correlation": 0.9920506422792474, "全帯域_10ms_modulation_mae_db": 0.20827546451653378, "全帯域_40ms_log_rms_mae_db": 0.5735665270386288, "全帯域_40ms_rms_envelope_nrmse": 0.042098317651946916, "全帯域_40ms_envelope_correlation": 0.9979609059009876, "全帯域_40ms_modulation_mae_db": 0.359505526256144, "300–1000 Hz_10ms_log_rms_mae_db": 1.8122647448019509, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.138971734100424, "300–1000 Hz_10ms_envelope_correlation": 0.9849460482264245, "300–1000 Hz_10ms_modulation_mae_db": 0.13181229594585567, "300–1000 Hz_40ms_log_rms_mae_db": 1.085248197944701, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08049034007379038, "300–1000 Hz_40ms_envelope_correlation": 0.9944533911041263, "300–1000 Hz_40ms_modulation_mae_db": 0.12350441589801409, "1000–3000 Hz_10ms_log_rms_mae_db": 2.220714961462074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25457270004984306, "1000–3000 Hz_10ms_envelope_correlation": 0.9513754797164803, "1000–3000 Hz_10ms_modulation_mae_db": 1.288977880219159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.528277882391426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18435855905094983, "1000–3000 Hz_40ms_envelope_correlation": 0.9714070019353391, "1000–3000 Hz_40ms_modulation_mae_db": 1.213585538477477, "3000–6000 Hz_10ms_log_rms_mae_db": 2.409704549128694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2323123814489496, "3000–6000 Hz_10ms_envelope_correlation": 0.9608775647368586, "3000–6000 Hz_10ms_modulation_mae_db": 0.4977201366039003, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6687288842217496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15791149441279984, "3000–6000 Hz_40ms_envelope_correlation": 0.9805994747584542, "3000–6000 Hz_40ms_modulation_mae_db": 0.698595588247322, "mr_stft_db": 5.881573880619874, "band_mae_db": 0.1287983885134906, "rms_gain_error_db": -0.010232633553095848, "output_peak": 0.7110578876535041}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1135952994020406, "全帯域_10ms_rms_envelope_nrmse": 0.08988175885364025, "全帯域_10ms_envelope_correlation": 0.9918476994181704, "全帯域_10ms_modulation_mae_db": 0.13036113494241272, "全帯域_40ms_log_rms_mae_db": 0.6136824214275638, "全帯域_40ms_rms_envelope_nrmse": 0.05907362113754875, "全帯域_40ms_envelope_correlation": 0.9960209612381611, "全帯域_40ms_modulation_mae_db": 0.27767229777374985, "300–1000 Hz_10ms_log_rms_mae_db": 1.8063258581146617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14082337097829073, "300–1000 Hz_10ms_envelope_correlation": 0.9845502667034385, "300–1000 Hz_10ms_modulation_mae_db": 0.36905830173843507, "300–1000 Hz_40ms_log_rms_mae_db": 1.0693991811574686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08312331124862879, "300–1000 Hz_40ms_envelope_correlation": 0.9941181367198532, "300–1000 Hz_40ms_modulation_mae_db": 0.36689602212907757, "1000–3000 Hz_10ms_log_rms_mae_db": 2.168772295629887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2104586223900619, "1000–3000 Hz_10ms_envelope_correlation": 0.9664970080637317, "1000–3000 Hz_10ms_modulation_mae_db": 0.6279839578148565, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4318103020679807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13238906834699218, "1000–3000 Hz_40ms_envelope_correlation": 0.9861025871472174, "1000–3000 Hz_40ms_modulation_mae_db": 0.7711717433045339, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4249797220262024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24087269338675396, "3000–6000 Hz_10ms_envelope_correlation": 0.9578145339802375, "3000–6000 Hz_10ms_modulation_mae_db": 0.8936758240652226, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7066630299546552, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1613956180626061, "3000–6000 Hz_40ms_envelope_correlation": 0.9797936362628694, "3000–6000 Hz_40ms_modulation_mae_db": 1.0857537106301633, "mr_stft_db": 5.910546308963077, "band_mae_db": 0.1655898700297553, "rms_gain_error_db": 0.010962074064209315, "output_peak": 0.6552223794219834}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0422090019394448, "全帯域_10ms_rms_envelope_nrmse": 0.09444484062921435, "全帯域_10ms_envelope_correlation": 0.9912628516020003, "全帯域_10ms_modulation_mae_db": 0.4336861005162869, "全帯域_40ms_log_rms_mae_db": 0.5563744710153287, "全帯域_40ms_rms_envelope_nrmse": 0.0641924334423881, "全帯域_40ms_envelope_correlation": 0.995714545623076, "全帯域_40ms_modulation_mae_db": 0.7145568337278982, "300–1000 Hz_10ms_log_rms_mae_db": 1.8010396386073704, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13592710585745196, "300–1000 Hz_10ms_envelope_correlation": 0.9862132460068364, "300–1000 Hz_10ms_modulation_mae_db": 0.8117642922641113, "300–1000 Hz_40ms_log_rms_mae_db": 1.0633271015502974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08320726403622175, "300–1000 Hz_40ms_envelope_correlation": 0.9950328458616667, "300–1000 Hz_40ms_modulation_mae_db": 1.2681277472129064, "1000–3000 Hz_10ms_log_rms_mae_db": 2.215877889407036, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20512023621450176, "1000–3000 Hz_10ms_envelope_correlation": 0.9690990697460861, "1000–3000 Hz_10ms_modulation_mae_db": 0.6122751429874645, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4193071420244652, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12388853440873922, "1000–3000 Hz_40ms_envelope_correlation": 0.9893815078319654, "1000–3000 Hz_40ms_modulation_mae_db": 0.9717839575811252, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3554072497226635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24127196174242507, "3000–6000 Hz_10ms_envelope_correlation": 0.9579420922738948, "3000–6000 Hz_10ms_modulation_mae_db": 1.2935490518939623, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6901318114681607, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16110318185733782, "3000–6000 Hz_40ms_envelope_correlation": 0.9804131694752101, "3000–6000 Hz_40ms_modulation_mae_db": 1.352252208597138, "mr_stft_db": 5.888290500512859, "band_mae_db": 0.26048806642477434, "rms_gain_error_db": -0.1884008026960587, "output_peak": 0.6140145003789026}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0922005921777647, "全帯域_10ms_rms_envelope_nrmse": 0.09363331469369053, "全帯域_10ms_envelope_correlation": 0.9910942409372114, "全帯域_10ms_modulation_mae_db": 0.21003406805536792, "全帯域_40ms_log_rms_mae_db": 0.6180690946776414, "全帯域_40ms_rms_envelope_nrmse": 0.05766866229662391, "全帯域_40ms_envelope_correlation": 0.9961718159955429, "全帯域_40ms_modulation_mae_db": 0.2569031495167466, "300–1000 Hz_10ms_log_rms_mae_db": 1.9068352810189708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15202357037452932, "300–1000 Hz_10ms_envelope_correlation": 0.9820706827512056, "300–1000 Hz_10ms_modulation_mae_db": 0.1575726688373291, "300–1000 Hz_40ms_log_rms_mae_db": 1.1388327145958361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09913541880747706, "300–1000 Hz_40ms_envelope_correlation": 0.9918356052118591, "300–1000 Hz_40ms_modulation_mae_db": 0.7033977363160782, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1926601636724588, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20798411455101512, "1000–3000 Hz_10ms_envelope_correlation": 0.9684813902932811, "1000–3000 Hz_10ms_modulation_mae_db": 1.550435606015298, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4373591217637594, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13032961978704813, "1000–3000 Hz_40ms_envelope_correlation": 0.9883120354383185, "1000–3000 Hz_40ms_modulation_mae_db": 1.7704327602555274, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4222156083846773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23382230071268062, "3000–6000 Hz_10ms_envelope_correlation": 0.9626146339705391, "3000–6000 Hz_10ms_modulation_mae_db": 0.35242373161707863, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6929189049813391, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15690664200612375, "3000–6000 Hz_40ms_envelope_correlation": 0.9816599206184977, "3000–6000 Hz_40ms_modulation_mae_db": 0.5159616414534437, "mr_stft_db": 5.89653069772012, "band_mae_db": 0.25855458939910464, "rms_gain_error_db": -0.04301658962457876, "output_peak": 0.7011876894976001}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88480, "output_samples": 88480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.420218808310974, "spectral_convergence": 0.28776620737686026, "spectral_convergence_status": "ok", "amplitude_floor": 2.1848847242256272e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.686648334763412, "spectral_convergence": 0.267704842622441, "spectral_convergence_status": "ok", "amplitude_floor": 1.2036494394056804e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.66404094160776, "spectral_convergence": 0.2815841439322636, "spectral_convergence_status": "ok", "amplitude_floor": 7.153950066191215e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002200971739696214, "output_power": 0.0022192243437586835, "delta_db": 0.035867414619998685, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002201298666996261, "output_power": 0.002115151056173691, "delta_db": -0.17337582093661258, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011542423166681864, "output_power": 0.0011252034066703172, "delta_db": -0.11065953940388606, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019781669780676847, "output_power": 0.00016317037645093745, "delta_db": -0.8362163232540544, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.802033486419172e-05, "output_power": 5.874267489583431e-05, "delta_db": 0.053734875889393185, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6630669693501258e-05, "output_power": 1.6742468799232984e-05, "delta_db": 0.0290976027569473, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20649192947681538, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09852292289475098, "pitch": null, "output_rms_dbfs": -22.442596953700715, "output_peak": 0.6294882549422605, "reference_rms_dbfs": -22.34407403080596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9236360282273814}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00024519594682554, "log_rms_mae_db": 1.1512005473087983, "rms_envelope_nrmse": 0.09093038614063244, "envelope_correlation": 0.9916558342071772, "rms_gain_error_db": -0.09852288370307094, "reference_mean_power": 0.00583423429533283, "output_mean_power": 0.005703370401665572}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.472320009859619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016537460501756134, "output_power": 0.0016173109510722028, "reference_active": true, "delta_db": -0.09675292687364231}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003778384775464376, "output_power": 0.0003268607527186163, "reference_active": true, "delta_db": -0.6294340613958256}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8517105494706513e-05, "output_power": 4.240621676404703e-05, "reference_active": true, "delta_db": -0.5846535422351375}], "active_band_count": 3, "active_band_mae_db": 0.4369468435015351}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00020002070942310007, "log_rms_mae_db": 0.6331578000282363, "rms_envelope_nrmse": 0.054888176804903197, "envelope_correlation": 0.9966318165089134, "rms_gain_error_db": -0.09851000942590711, "reference_mean_power": 0.005865946925483087, "output_mean_power": 0.0057343887056149345}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.084203911038529e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015941324207372327, "output_power": 0.0015537641486753902, "reference_active": true, "delta_db": -0.11139297973258205}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018248750335661006, "output_power": 0.00015822273817084473, "reference_active": true, "delta_db": -0.619642335096487}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.552526527817181e-06, "output_power": 3.253973730346563e-06, "reference_active": true, "delta_db": -1.4583844261919656}], "active_band_count": 3, "active_band_mae_db": 0.729806580340345}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00018690652054173922, "log_rms_mae_db": 1.853648654446914, "rms_envelope_nrmse": 0.153281106731306, "envelope_correlation": 0.982124150032818, "rms_gain_error_db": -0.30965395017583125, "reference_mean_power": 0.0020929267743018004, "output_mean_power": 0.0019488958859538964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 8.903808595053404e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005237826202169963, "output_power": 0.0004828965795369003, "reference_active": true, "delta_db": -0.35296954669618935}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020282121471265822, "output_power": 0.00014841743622157183, "reference_active": true, "delta_db": -1.356284541435746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.57818052122161e-05, "output_power": 1.80249132026434e-05, "reference_active": true, "delta_db": -1.5544014078963895}], "active_band_count": 3, "active_band_mae_db": 1.087885165342775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00016950242057071555, "log_rms_mae_db": 1.1059313055147362, "rms_envelope_nrmse": 0.10639392734076764, "envelope_correlation": 0.9910169622665127, "rms_gain_error_db": -0.3096433281603865, "reference_mean_power": 0.002104315784196243, "output_mean_power": 0.001959505920334337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 7.681995008528509e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000538193996495516, "output_power": 0.0004858132326508012, "reference_active": true, "delta_db": -0.4446950893392364}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010305287620923108, "output_power": 7.542183173263368e-05, "reference_active": true, "delta_db": -1.355630417654947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.715218896542221e-06, "output_power": 1.5845185609014385e-06, "reference_active": true, "delta_db": -2.3390751661193363}], "active_band_count": 3, "active_band_mae_db": 1.3798002243711733}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 7.550458878286076e-05, "log_rms_mae_db": 2.307633800217155, "rms_envelope_nrmse": 0.20741044053893407, "envelope_correlation": 0.9698011632714165, "rms_gain_error_db": -0.6798032800837217, "reference_mean_power": 0.00021315330376123098, "output_mean_power": 0.00018226855071614293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 1.6207032137826829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010202948507054384, "output_power": 8.380012811323171e-05, "reference_active": true, "delta_db": -0.8548101225241816}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.692530062160877e-05, "output_power": 2.7556010018629598e-05, "reference_active": true, "delta_db": -1.2710770576153725}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.752197568216375e-06, "output_power": 4.447997739977506e-06, "reference_active": true, "delta_db": -1.8128058328076433}], "active_band_count": 3, "active_band_mae_db": 1.3128976709823992}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 5.938009261210461e-05, "log_rms_mae_db": 1.4958614574957718, "rms_envelope_nrmse": 0.12978875287038047, "envelope_correlation": 0.9900943706645847, "rms_gain_error_db": -0.67975999262401, "reference_mean_power": 0.00021425147153668813, "output_mean_power": 0.00018320942605601917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 1.4064708995422735e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010396272320640846, "output_power": 8.404639703752691e-05, "reference_active": true, "delta_db": -0.9235854639542056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97231105181685e-05, "output_power": 1.4641542658337404e-05, "reference_active": true, "delta_db": -1.2938857113693738}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.307963255185344e-07, "output_power": 3.7313473046555574e-07, "reference_active": true, "delta_db": -2.2802348084753317}], "active_band_count": 3, "active_band_mae_db": 1.4992353279329702}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 4.4291934004628345e-05, "log_rms_mae_db": 2.4186521739234337, "rms_envelope_nrmse": 0.24124401252118333, "envelope_correlation": 0.9588025001809181, "rms_gain_error_db": 0.09563645779178878, "reference_mean_power": 2.6417005173666327e-05, "output_mean_power": 2.7005189368182884e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.694560310630941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.502841409588563e-05, "output_power": 1.5353666824305273e-05, "reference_active": true, "delta_db": 0.09298958958998185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.7026465908103935e-06, "output_power": 7.738561251208042e-06, "reference_active": true, "delta_db": 0.6241390337752785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.011540321330325e-07, "output_power": 1.0153197289248164e-06, "reference_active": true, "delta_db": 0.5180379489255161}], "active_band_count": 3, "active_band_mae_db": 0.4117221907635922}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 3.2524867537670264e-05, "log_rms_mae_db": 1.6797217569733753, "rms_envelope_nrmse": 0.15793217830653772, "envelope_correlation": 0.9804020866301104, "rms_gain_error_db": 0.09601898111591792, "reference_mean_power": 2.65459352580292e-05, "output_mean_power": 2.7139380438617308e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.4345030346700597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6238306479668564e-05, "output_power": 1.654291148435491e-05, "reference_active": true, "delta_db": 0.08071212021361857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6930369104619132e-06, "output_power": 4.44201742356285e-06, "reference_active": true, "delta_db": 0.8019660943497124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.897644209500044e-08, "output_power": 7.831965908594614e-08, "reference_active": true, "delta_db": -0.03626776406611938}], "active_band_count": 3, "active_band_mae_db": 0.3063153262098168}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8765563222904734, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c04c26bb170f37df8060189c11ed5ce1a953a8f80fe18b648a6454444ef2cf85", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9668329496355546, "standard_deviation": 0.006502446595831987, "minimum": 0.9513754797164803, "maximum": 0.9720774065593034, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.216417102246573, "standard_deviation": 0.04308498990118679, "minimum": 2.168772295629887, "maximum": 2.307633800217155, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9923902697411755, "standard_deviation": 0.48563411564101167, "minimum": 0.16576572028940809, "maximum": 1.550435606015298, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2107170063508032, "standard_deviation": 0.018892486612874276, "minimum": 0.19193723380691832, "maximum": 0.25457270004984306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9868019003801859, "standard_deviation": 0.006360777724424142, "minimum": 0.9714070019353391, "maximum": 0.9903170020059129, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.449487088247658, "standard_deviation": 0.04044582353347573, "minimum": 1.4193071420244652, "maximum": 1.528277882391426, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1718939155083805, "standard_deviation": 0.48302414532250604, "minimum": 0.2843929217030116, "maximum": 1.7704327602555274, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1310182971822498, "standard_deviation": 0.022685019222435647, "minimum": 0.11375769792262995, "maximum": 0.18435855905094983, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9602563478914397, "standard_deviation": 0.003389297707290696, "minimum": 0.955138960079155, "maximum": 0.9644574834694456, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.4006696400737653, "standard_deviation": 0.03158815007114048, "minimum": 2.3453292630592966, "maximum": 2.4249797220262024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5642543778379397, "standard_deviation": 0.3509831981138348, "minimum": 0.2886548832674637, "maximum": 1.2935490518939623, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23650501672133228, "standard_deviation": 0.00993544708654236, "minimum": 0.22205355766370333, "maximum": 0.2537471310757032, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9801771764767744, "standard_deviation": 0.000906652329348122, "minimum": 0.9784388508731758, "maximum": 0.9816599206184977, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.695733306038281, "standard_deviation": 0.017663042277592717, "minimum": 1.6687288842217496, "maximum": 1.7269662879055379, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7026771965139902, "standard_deviation": 0.35594011884152105, "minimum": 0.3063153262098168, "maximum": 1.352252208597138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16027283929212832, "standard_deviation": 0.0037278012721422807, "minimum": 0.1565102135855508, "maximum": 0.16744749073184212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9851382428101763, "standard_deviation": 0.0021139458864072704, "minimum": 0.9820706827512056, "maximum": 0.9872199816119408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8258936633502822, "standard_deviation": 0.042535556391681224, "minimum": 1.77690379135315, "maximum": 1.9068352810189708, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5648210613299205, "standard_deviation": 0.39468235823246633, "minimum": 0.13181229594585567, "maximum": 1.087885165342775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1385924454177325, "standard_deviation": 0.009824651204439996, "minimum": 0.12823406428000975, "maximum": 0.153281106731306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9939767556687467, "standard_deviation": 0.0016325879117100497, "minimum": 0.9910169622665127, "maximum": 0.9952728268987968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0840358432477737, "standard_deviation": 0.04748449344789679, "minimum": 1.000268476459319, "maximum": 1.148175806468647, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.723366177579155, "standard_deviation": 0.5237172783767089, "minimum": 0.06025494963002539, "maximum": 1.3798002243711733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08529302879446232, "standard_deviation": 0.011348888306827652, "minimum": 0.07452830010488756, "maximum": 0.10639392734076764, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18389347957303986, "standard_deviation": 0.05454010003169428, "minimum": 0.12823443772307547, "maximum": 0.26048806642477434, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.897998754008302, "standard_deviation": 0.014854972414978358, "minimum": 5.881573880619874, "maximum": 5.9236360282273814, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6763623104034946, "standard_deviation": 0.04038056155064754, "minimum": 0.6140145003789026, "maximum": 0.7260476311229502, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05837219642422719, "standard_deviation": 0.061756372742357145, "minimum": -0.1884008026960587, "maximum": 0.010962074064209315, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9923098833401385, "standard_deviation": 0.0011169337709574473, "minimum": 0.9910942409372114, "maximum": 0.9942834647313109, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0978415342099765, "standard_deviation": 0.03314364786328165, "minimum": 1.0422090019394448, "maximum": 1.1512005473087983, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3244124585419155, "standard_deviation": 0.1211060682206158, "minimum": 0.13036113494241272, "maximum": 0.4369468435015351, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08714201474967484, "standard_deviation": 0.006707726274182875, "minimum": 0.07523471963458543, "maximum": 0.09444484062921435, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997014235707006, "standard_deviation": 0.0009782277500725743, "minimum": 0.995714545623076, "maximum": 0.9980424212343526, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5927363333763455, "standard_deviation": 0.02819277127906389, "minimum": 0.5563744710153287, "maximum": 0.6331578000282363, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.45533053710349847, "standard_deviation": 0.21688915989735183, "minimum": 0.16989750412222668, "maximum": 0.729806580340345, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05094445837018713, "standard_deviation": 0.008994117032277307, "minimum": 0.041462388560226306, "maximum": 0.0641924334423881, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0026", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.410280466079712, "predicted_mos": 2.410280466079712, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "213bbcc0607f050c499f661ab0facee94566c2fba9fee65c45ef679c78b80708", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.72780690016225, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.070686313792307, "全帯域_10ms_rms_envelope_nrmse": 0.0796209427433675, "全帯域_10ms_envelope_correlation": 0.9936226188784953, "全帯域_10ms_modulation_mae_db": 0.17726964842564108, "全帯域_40ms_log_rms_mae_db": 0.5573862082609475, "全帯域_40ms_rms_envelope_nrmse": 0.04095810547669491, "全帯域_40ms_envelope_correlation": 0.9981731086657406, "全帯域_40ms_modulation_mae_db": 0.08822876871399678, "300–1000 Hz_10ms_log_rms_mae_db": 1.8770849056225745, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1176158441721202, "300–1000 Hz_10ms_envelope_correlation": 0.9893653981577238, "300–1000 Hz_10ms_modulation_mae_db": 0.09999012659315054, "300–1000 Hz_40ms_log_rms_mae_db": 1.0925292895352723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0760622239184011, "300–1000 Hz_40ms_envelope_correlation": 0.9952600019550496, "300–1000 Hz_40ms_modulation_mae_db": 0.2857074678283055, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3764792293530075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.194970002219374, "1000–3000 Hz_10ms_envelope_correlation": 0.972461034792148, "1000–3000 Hz_10ms_modulation_mae_db": 0.9442188939263277, "1000–3000 Hz_40ms_log_rms_mae_db": 1.537758409625263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12481377151558687, "1000–3000 Hz_40ms_envelope_correlation": 0.9891540303756016, "1000–3000 Hz_40ms_modulation_mae_db": 1.1917129235088946, "3000–6000 Hz_10ms_log_rms_mae_db": 2.708483647691708, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2482869105271393, "3000–6000 Hz_10ms_envelope_correlation": 0.9551462136184682, "3000–6000 Hz_10ms_modulation_mae_db": 0.9587598248925008, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8367093997288386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16267458614452918, "3000–6000 Hz_40ms_envelope_correlation": 0.9787275151396084, "3000–6000 Hz_40ms_modulation_mae_db": 0.8299268780719271, "mr_stft_db": 6.064021079048229, "band_mae_db": 0.23603423342913768, "rms_gain_error_db": -0.09017399952427668, "output_peak": 0.6088429249160301}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1034579643719158, "全帯域_10ms_rms_envelope_nrmse": 0.0826178281314555, "全帯域_10ms_envelope_correlation": 0.9931010545236497, "全帯域_10ms_modulation_mae_db": 0.26071933924846385, "全帯域_40ms_log_rms_mae_db": 0.5891699779166526, "全帯域_40ms_rms_envelope_nrmse": 0.04468613454065005, "全帯域_40ms_envelope_correlation": 0.9977643515990875, "全帯域_40ms_modulation_mae_db": 0.34617806036293547, "300–1000 Hz_10ms_log_rms_mae_db": 1.9373079792633556, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15135834466695727, "300–1000 Hz_10ms_envelope_correlation": 0.9823121833190053, "300–1000 Hz_10ms_modulation_mae_db": 0.3433282922029419, "300–1000 Hz_40ms_log_rms_mae_db": 1.1843105918319004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09082870353988157, "300–1000 Hz_40ms_envelope_correlation": 0.9933571527368941, "300–1000 Hz_40ms_modulation_mae_db": 0.2976759042349479, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3472955760046545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20839690158335747, "1000–3000 Hz_10ms_envelope_correlation": 0.9680311049427347, "1000–3000 Hz_10ms_modulation_mae_db": 1.4389738181886305, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4873043645393087, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12670638006864218, "1000–3000 Hz_40ms_envelope_correlation": 0.988565539207165, "1000–3000 Hz_40ms_modulation_mae_db": 1.4570726479586218, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6453296126411265, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24367558550263488, "3000–6000 Hz_10ms_envelope_correlation": 0.9568343894575904, "3000–6000 Hz_10ms_modulation_mae_db": 1.0197021762611713, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7760879158456289, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15961260219205453, "3000–6000 Hz_40ms_envelope_correlation": 0.9798019642749101, "3000–6000 Hz_40ms_modulation_mae_db": 1.4198374487042642, "mr_stft_db": 6.0689740746462055, "band_mae_db": 0.18369496335641186, "rms_gain_error_db": -0.07593969834485957, "output_peak": 0.6289008113741715}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0642559468925077, "全帯域_10ms_rms_envelope_nrmse": 0.0801983972246207, "全帯域_10ms_envelope_correlation": 0.9935140214568075, "全帯域_10ms_modulation_mae_db": 0.2783114619321963, "全帯域_40ms_log_rms_mae_db": 0.5667800000402977, "全帯域_40ms_rms_envelope_nrmse": 0.04155088628228896, "全帯域_40ms_envelope_correlation": 0.9980933156902282, "全帯域_40ms_modulation_mae_db": 0.45997149967691753, "300–1000 Hz_10ms_log_rms_mae_db": 1.9295902940567344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14113091976288206, "300–1000 Hz_10ms_envelope_correlation": 0.984800837695889, "300–1000 Hz_10ms_modulation_mae_db": 0.977884679200122, "300–1000 Hz_40ms_log_rms_mae_db": 1.1279533093409504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08483874804235134, "300–1000 Hz_40ms_envelope_correlation": 0.9944126934527738, "300–1000 Hz_40ms_modulation_mae_db": 1.2259131244193018, "1000–3000 Hz_10ms_log_rms_mae_db": 2.324441418185821, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21863868619925172, "1000–3000 Hz_10ms_envelope_correlation": 0.9639089519027265, "1000–3000 Hz_10ms_modulation_mae_db": 1.0996569885982306, "1000–3000 Hz_40ms_log_rms_mae_db": 1.474677520625219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12625610412958893, "1000–3000 Hz_40ms_envelope_correlation": 0.9878439953129526, "1000–3000 Hz_40ms_modulation_mae_db": 1.136071376071584, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6825639228988107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24681990572462523, "3000–6000 Hz_10ms_envelope_correlation": 0.9560733913719164, "3000–6000 Hz_10ms_modulation_mae_db": 0.9238808151432564, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8274023942067152, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16025615281258238, "3000–6000 Hz_40ms_envelope_correlation": 0.9793072682191599, "3000–6000 Hz_40ms_modulation_mae_db": 0.6756836620361447, "mr_stft_db": 6.0578915233162585, "band_mae_db": 0.21172917881433342, "rms_gain_error_db": -0.08214286666353793, "output_peak": 0.6425290340108445}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1503365269937325, "全帯域_10ms_rms_envelope_nrmse": 0.09316055863810983, "全帯域_10ms_envelope_correlation": 0.9912818881763753, "全帯域_10ms_modulation_mae_db": 0.3608003911475248, "全帯域_40ms_log_rms_mae_db": 0.6571491658120325, "全帯域_40ms_rms_envelope_nrmse": 0.05749572741849621, "全帯域_40ms_envelope_correlation": 0.9963609973355416, "全帯域_40ms_modulation_mae_db": 0.3276394921186659, "300–1000 Hz_10ms_log_rms_mae_db": 1.8805088846368196, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13422388139537414, "300–1000 Hz_10ms_envelope_correlation": 0.9860988890206432, "300–1000 Hz_10ms_modulation_mae_db": 0.539397511342525, "300–1000 Hz_40ms_log_rms_mae_db": 1.1294006588396728, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08156614636606908, "300–1000 Hz_40ms_envelope_correlation": 0.9945722310251176, "300–1000 Hz_40ms_modulation_mae_db": 0.35230446341541505, "1000–3000 Hz_10ms_log_rms_mae_db": 2.394817863934509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20099078826011846, "1000–3000 Hz_10ms_envelope_correlation": 0.9697154729621134, "1000–3000 Hz_10ms_modulation_mae_db": 1.2058149396202662, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5077599576047558, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12069268216437927, "1000–3000 Hz_40ms_envelope_correlation": 0.9888294220526578, "1000–3000 Hz_40ms_modulation_mae_db": 1.5380243289908329, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6917484374247356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23579745396658414, "3000–6000 Hz_10ms_envelope_correlation": 0.9599487311023568, "3000–6000 Hz_10ms_modulation_mae_db": 0.9256782373869781, "3000–6000 Hz_40ms_log_rms_mae_db": 1.857021692204639, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16484160254060043, "3000–6000 Hz_40ms_envelope_correlation": 0.97901109000398, "3000–6000 Hz_40ms_modulation_mae_db": 1.091738092216317, "mr_stft_db": 6.069171078210364, "band_mae_db": 0.182491925191258, "rms_gain_error_db": -0.12278820688267562, "output_peak": 0.6165776955375686}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.141137388683101, "全帯域_10ms_rms_envelope_nrmse": 0.09165298275849579, "全帯域_10ms_envelope_correlation": 0.9914795703877012, "全帯域_10ms_modulation_mae_db": 0.277526790556995, "全帯域_40ms_log_rms_mae_db": 0.6255170515795511, "全帯域_40ms_rms_envelope_nrmse": 0.05594610183572717, "全帯域_40ms_envelope_correlation": 0.9964148599955424, "全帯域_40ms_modulation_mae_db": 0.40117328442746863, "300–1000 Hz_10ms_log_rms_mae_db": 1.8770056802131971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14265028437167246, "300–1000 Hz_10ms_envelope_correlation": 0.9843108303989893, "300–1000 Hz_10ms_modulation_mae_db": 0.7768573724870779, "300–1000 Hz_40ms_log_rms_mae_db": 1.112473113544792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09265229545250808, "300–1000 Hz_40ms_envelope_correlation": 0.9929597400845446, "300–1000 Hz_40ms_modulation_mae_db": 0.8689476360995384, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3806854934098154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20572769260386026, "1000–3000 Hz_10ms_envelope_correlation": 0.9678224731464875, "1000–3000 Hz_10ms_modulation_mae_db": 1.2073559754667604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4868902265789483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11418122128302063, "1000–3000 Hz_40ms_envelope_correlation": 0.9895600304611681, "1000–3000 Hz_40ms_modulation_mae_db": 1.2895119099215553, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6344406755233765, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24098997067794223, "3000–6000 Hz_10ms_envelope_correlation": 0.9591054035209323, "3000–6000 Hz_10ms_modulation_mae_db": 0.20358850066994796, "3000–6000 Hz_40ms_log_rms_mae_db": 1.804359308781919, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1667199736441279, "3000–6000 Hz_40ms_envelope_correlation": 0.9782170688184377, "3000–6000 Hz_40ms_modulation_mae_db": 0.17760591239185564, "mr_stft_db": 6.066972910652299, "band_mae_db": 0.2090963072228009, "rms_gain_error_db": -0.04517342751661506, "output_peak": 0.6535210472289148}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.131988998270632, "全帯域_10ms_rms_envelope_nrmse": 0.09447881620189837, "全帯域_10ms_envelope_correlation": 0.9910487968112017, "全帯域_10ms_modulation_mae_db": 0.4503663101995934, "全帯域_40ms_log_rms_mae_db": 0.6429288553616881, "全帯域_40ms_rms_envelope_nrmse": 0.056608911107486844, "全帯域_40ms_envelope_correlation": 0.9965034207993311, "全帯域_40ms_modulation_mae_db": 0.3471872116811596, "300–1000 Hz_10ms_log_rms_mae_db": 1.9716557409734017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14976620181484132, "300–1000 Hz_10ms_envelope_correlation": 0.9828552849677481, "300–1000 Hz_10ms_modulation_mae_db": 0.5808978949619487, "300–1000 Hz_40ms_log_rms_mae_db": 1.2265730506693393, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09760396080347705, "300–1000 Hz_40ms_envelope_correlation": 0.992454571131437, "300–1000 Hz_40ms_modulation_mae_db": 0.9852742278200081, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3548743002725216, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20859369954074086, "1000–3000 Hz_10ms_envelope_correlation": 0.9678595495293422, "1000–3000 Hz_10ms_modulation_mae_db": 1.4098861680906376, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5292475363703775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12435529986635621, "1000–3000 Hz_40ms_envelope_correlation": 0.9890824933800728, "1000–3000 Hz_40ms_modulation_mae_db": 1.6758168744746802, "3000–6000 Hz_10ms_log_rms_mae_db": 2.60088013258626, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24514696892928178, "3000–6000 Hz_10ms_envelope_correlation": 0.9563087556273079, "3000–6000 Hz_10ms_modulation_mae_db": 1.2973170804736047, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7727480358433325, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16464470417331667, "3000–6000 Hz_40ms_envelope_correlation": 0.9787015122810194, "3000–6000 Hz_40ms_modulation_mae_db": 1.3074390169336734, "mr_stft_db": 6.064317052715016, "band_mae_db": 0.2715214167184396, "rms_gain_error_db": -0.12652964943814268, "output_peak": 0.611066060678741}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0952549753498564, "全帯域_10ms_rms_envelope_nrmse": 0.08273064617346688, "全帯域_10ms_envelope_correlation": 0.9931476519852991, "全帯域_10ms_modulation_mae_db": 0.5265185952868393, "全帯域_40ms_log_rms_mae_db": 0.597008422564925, "全帯域_40ms_rms_envelope_nrmse": 0.044779024455402716, "全帯域_40ms_envelope_correlation": 0.9978310368810507, "全帯域_40ms_modulation_mae_db": 0.834130818037211, "300–1000 Hz_10ms_log_rms_mae_db": 1.9446077251863154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12663010481491213, "300–1000 Hz_10ms_envelope_correlation": 0.9876919342802082, "300–1000 Hz_10ms_modulation_mae_db": 0.28913698689876605, "300–1000 Hz_40ms_log_rms_mae_db": 1.1528222014406149, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07569993660184508, "300–1000 Hz_40ms_envelope_correlation": 0.9954077061065784, "300–1000 Hz_40ms_modulation_mae_db": 0.5121603674671019, "1000–3000 Hz_10ms_log_rms_mae_db": 2.379186242073785, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21804705247382053, "1000–3000 Hz_10ms_envelope_correlation": 0.9636820221757116, "1000–3000 Hz_10ms_modulation_mae_db": 0.6947551406955407, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5274590763070606, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12658021111990306, "1000–3000 Hz_40ms_envelope_correlation": 0.9869165278891354, "1000–3000 Hz_40ms_modulation_mae_db": 0.8702585348999219, "3000–6000 Hz_10ms_log_rms_mae_db": 2.677341608771496, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23401391151048512, "3000–6000 Hz_10ms_envelope_correlation": 0.9604479170639029, "3000–6000 Hz_10ms_modulation_mae_db": 0.9054528231991531, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7991798353000439, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1633305260466014, "3000–6000 Hz_40ms_envelope_correlation": 0.9791697975223345, "3000–6000 Hz_40ms_modulation_mae_db": 0.9576697024097501, "mr_stft_db": 6.056109469956264, "band_mae_db": 0.14462498115703223, "rms_gain_error_db": -0.10066245063229376, "output_peak": 0.6300019216509476}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1145006538473563, "全帯域_10ms_rms_envelope_nrmse": 0.08207240752512224, "全帯域_10ms_envelope_correlation": 0.9933647626934267, "全帯域_10ms_modulation_mae_db": 0.3035849415327458, "全帯域_40ms_log_rms_mae_db": 0.6083400868172697, "全帯域_40ms_rms_envelope_nrmse": 0.051666489843514414, "全帯域_40ms_envelope_correlation": 0.9972080880147562, "全帯域_40ms_modulation_mae_db": 0.3961213398626155, "300–1000 Hz_10ms_log_rms_mae_db": 1.9382463971884156, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14020474308573908, "300–1000 Hz_10ms_envelope_correlation": 0.98546112600625, "300–1000 Hz_10ms_modulation_mae_db": 0.5885123253649137, "300–1000 Hz_40ms_log_rms_mae_db": 1.1196131920983643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10077457334931582, "300–1000 Hz_40ms_envelope_correlation": 0.9923851028608318, "300–1000 Hz_40ms_modulation_mae_db": 0.7633468927180048, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4492078291077575, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20540752223555753, "1000–3000 Hz_10ms_envelope_correlation": 0.9678224889993475, "1000–3000 Hz_10ms_modulation_mae_db": 1.3413234842398847, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5315066298368043, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1198862948562701, "1000–3000 Hz_40ms_envelope_correlation": 0.9881249240412739, "1000–3000 Hz_40ms_modulation_mae_db": 1.6250127867484674, "3000–6000 Hz_10ms_log_rms_mae_db": 2.661741246666472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24237764415283117, "3000–6000 Hz_10ms_envelope_correlation": 0.957517144022829, "3000–6000 Hz_10ms_modulation_mae_db": 0.6243135034499647, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7781875504204243, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1594879247251777, "3000–6000 Hz_40ms_envelope_correlation": 0.9794749920989694, "3000–6000 Hz_40ms_modulation_mae_db": 0.7549874853303136, "mr_stft_db": 6.069574128798408, "band_mae_db": 0.18674807512858774, "rms_gain_error_db": -0.14727191385473376, "output_peak": 0.630391686841694}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88480, "output_samples": 88480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.658291421188946, "spectral_convergence": 0.2821392019963134, "spectral_convergence_status": "ok", "amplitude_floor": 2.1848847242256272e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.858046362125464, "spectral_convergence": 0.25042692022002006, "spectral_convergence_status": "ok", "amplitude_floor": 1.2036494394056804e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.675725453830277, "spectral_convergence": 0.289394881208201, "spectral_convergence_status": "ok", "amplitude_floor": 7.153950066191215e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002200971739696214, "output_power": 0.002211242690830804, "delta_db": 0.02021944095904135, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002201298666996261, "output_power": 0.0021770804230613495, "delta_db": -0.04804498082276489, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011542423166681864, "output_power": 0.0010775614236301079, "delta_db": -0.29854956923847786, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019781669780676847, "output_power": 0.00017145660053140166, "delta_db": -0.6210873910070751, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.802033486419172e-05, "output_power": 5.45891187740117e-05, "delta_db": -0.2647414740749917, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6630669693501258e-05, "output_power": 1.7268952400889813e-05, "delta_db": 0.16356254447247506, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23603423342913768, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09017399952427668, "pitch": null, "output_rms_dbfs": -22.434248030330238, "output_peak": 0.6088429249160301, "reference_rms_dbfs": -22.34407403080596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.064021079048229}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00024519594682554, "log_rms_mae_db": 1.070686313792307, "rms_envelope_nrmse": 0.0796209427433675, "envelope_correlation": 0.9936226188784953, "rms_gain_error_db": -0.09016800685790081, "reference_mean_power": 0.00583423429533283, "output_mean_power": 0.0057143530007353035}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.472320009859619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016537460501756134, "output_power": 0.001661952202887439, "reference_active": true, "delta_db": 0.021497095604472327}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003778384775464376, "output_power": 0.00036229623591452285, "reference_active": true, "delta_db": -0.18242360691110973}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8517105494706513e-05, "output_power": 5.232192421960262e-05, "reference_active": true, "delta_db": 0.3278882427613412}], "active_band_count": 3, "active_band_mae_db": 0.17726964842564108}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00020002070942310007, "log_rms_mae_db": 0.5573862082609475, "rms_envelope_nrmse": 0.04095810547669491, "envelope_correlation": 0.9981731086657406, "rms_gain_error_db": -0.09015534736531107, "reference_mean_power": 0.005865946925483087, "output_mean_power": 0.005745430750412886}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.084203911038529e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015941324207372327, "output_power": 0.0015979610292306968, "reference_active": true, "delta_db": 0.01041789261790312}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018248750335661006, "output_power": 0.00017817278908738217, "reference_active": true, "delta_db": -0.10391751140698291}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.552526527817181e-06, "output_power": 4.7128929517463035e-06, "reference_active": true, "delta_db": 0.15035090211710433}], "active_band_count": 3, "active_band_mae_db": 0.08822876871399678}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00018690652054173922, "log_rms_mae_db": 1.8770849056225745, "rms_envelope_nrmse": 0.1176158441721202, "envelope_correlation": 0.9893653981577238, "rms_gain_error_db": -0.17099888091662963, "reference_mean_power": 0.0020929267743018004, "output_mean_power": 0.0020121212411442608}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 8.903808595053404e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005237826202169963, "output_power": 0.0005128645116355864, "reference_active": true, "delta_db": -0.09148435374986248}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020282121471265822, "output_power": 0.00020225313354679716, "reference_active": true, "delta_db": -0.012181204680554572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.57818052122161e-05, "output_power": 2.4642390802366307e-05, "reference_active": true, "delta_db": -0.19630482134903457}], "active_band_count": 3, "active_band_mae_db": 0.09999012659315054}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00016950242057071555, "log_rms_mae_db": 1.0925292895352723, "rms_envelope_nrmse": 0.0760622239184011, "envelope_correlation": 0.9952600019550496, "rms_gain_error_db": -0.17099330932235476, "reference_mean_power": 0.002104315784196243, "output_mean_power": 0.0020230731296742416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 7.681995008528509e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000538193996495516, "output_power": 0.0005312520892901086, "reference_active": true, "delta_db": -0.056381978995300584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010305287620923108, "output_power": 0.00010470575517880951, "reference_active": true, "delta_db": 0.0691043598505457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.715218896542221e-06, "output_power": 2.2942526445900537e-06, "reference_active": true, "delta_db": -0.7316360646390702}], "active_band_count": 3, "active_band_mae_db": 0.2857074678283055}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 7.550458878286076e-05, "log_rms_mae_db": 2.3764792293530075, "rms_envelope_nrmse": 0.194970002219374, "envelope_correlation": 0.972461034792148, "rms_gain_error_db": -0.5274462409444235, "reference_mean_power": 0.00021315330376123098, "output_mean_power": 0.00018877628894631965}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 1.6207032137826829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010202948507054384, "output_power": 9.016404424571844e-05, "reference_active": true, "delta_db": -0.536923113775604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.692530062160877e-05, "output_power": 3.165807817705076e-05, "reference_active": true, "delta_db": -0.6683949252449829}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.752197568216375e-06, "output_power": 4.642065636960789e-06, "reference_active": true, "delta_db": -1.6273386427583962}], "active_band_count": 3, "active_band_mae_db": 0.9442188939263277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 5.938009261210461e-05, "log_rms_mae_db": 1.537758409625263, "rms_envelope_nrmse": 0.12481377151558687, "envelope_correlation": 0.9891540303756016, "rms_gain_error_db": -0.5271740173565486, "reference_mean_power": 0.00021425147153668813, "output_mean_power": 0.00018976076029070843}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 1.4064708995422735e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010396272320640846, "output_power": 9.201382785905792e-05, "reference_active": true, "delta_db": -0.5302454879216355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97231105181685e-05, "output_power": 1.7094922518861526e-05, "reference_active": true, "delta_db": -0.6210827155223363}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.307963255185344e-07, "output_power": 3.610007449694417e-07, "reference_active": true, "delta_db": -2.423810567082712}], "active_band_count": 3, "active_band_mae_db": 1.1917129235088946}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 4.4291934004628345e-05, "log_rms_mae_db": 2.708483647691708, "rms_envelope_nrmse": 0.2482869105271393, "envelope_correlation": 0.9551462136184682, "rms_gain_error_db": -0.1559339806279899, "reference_mean_power": 2.6417005173666327e-05, "output_mean_power": 2.548532540556058e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.694560310630941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.502841409588563e-05, "output_power": 1.3595823989593785e-05, "reference_active": true, "delta_db": -0.43507619584828705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.7026465908103935e-06, "output_power": 5.688025972854804e-06, "reference_active": true, "delta_db": -0.7128475021745712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.011540321330325e-07, "output_power": 6.052899353308954e-07, "reference_active": true, "delta_db": -1.7283557766546442}], "active_band_count": 3, "active_band_mae_db": 0.9587598248925008}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 3.2524867537670264e-05, "log_rms_mae_db": 1.8367093997288386, "rms_envelope_nrmse": 0.16267458614452918, "envelope_correlation": 0.9787275151396084, "rms_gain_error_db": -0.15598672108195055, "reference_mean_power": 2.65459352580292e-05, "output_mean_power": 2.5609397359306698e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.4345030346700597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6238306479668564e-05, "output_power": 1.4189707309705744e-05, "reference_active": true, "delta_db": -0.5856729658186914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6930369104619132e-06, "output_power": 3.1760434470762574e-06, "reference_active": true, "delta_db": -0.6549721349606161}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.897644209500044e-08, "output_power": 5.923576635293234e-08, "reference_active": true, "delta_db": -1.2491355334364738}], "active_band_count": 3, "active_band_mae_db": 0.8299268780719271}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8851855943566838, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b853feeeea1842190a628c904e6fa991def194b78dc9b4848a644ed45c72878a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9676628873063264, "standard_deviation": 0.002863560370565236, "minimum": 0.9636820221757116, "maximum": 0.972461034792148, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.375873494042734, "standard_deviation": 0.03717570706035066, "minimum": 2.324441418185821, "maximum": 2.4492078291077575, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1677481761032849, "standard_deviation": 0.251755142177216, "minimum": 0.6947551406955407, "maximum": 1.4389738181886305, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2075965431395101, "standard_deviation": 0.007965343784191264, "minimum": 0.194970002219374, "maximum": 0.21863868619925172, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9885096203400033, "standard_deviation": 0.0008519893332405423, "minimum": 0.9869165278891354, "maximum": 0.9895600304611681, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.510325465185967, "standard_deviation": 0.024522155721594847, "minimum": 1.474677520625219, "maximum": 1.537758409625263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3479351728218196, "standard_deviation": 0.2759925459160187, "minimum": 0.8702585348999219, "maximum": 1.6758168744746802, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1229339956254684, "standard_deviation": 0.004392182415116079, "minimum": 0.11418122128302063, "maximum": 0.12670638006864218, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.957672743223163, "standard_deviation": 0.0019443329630307203, "minimum": 0.9551462136184682, "maximum": 0.9604479170639029, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.662816160525498, "standard_deviation": 0.034788613282555884, "minimum": 2.60088013258626, "maximum": 2.708483647691708, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8573366201845722, "standard_deviation": 0.32145479529079674, "minimum": 0.20358850066994796, "maximum": 1.2973170804736047, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24213854387394046, "standard_deviation": 0.005052950215264243, "minimum": 0.23401391151048512, "maximum": 0.2482869105271393, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9790514010448025, "standard_deviation": 0.0004996169605569783, "minimum": 0.9782170688184377, "maximum": 0.9798019642749101, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8064620165414427, "standard_deviation": 0.031226819688924344, "minimum": 1.7727480358433325, "maximum": 1.857021692204639, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9018610247617806, "standard_deviation": 0.3917741729234544, "minimum": 0.17760591239185564, "maximum": 1.4198374487042642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16269600903487377, "standard_deviation": 0.002692541212610877, "minimum": 0.1594879247251777, "maximum": 0.1667199736441279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9853620604808071, "standard_deviation": 0.0023607954054932928, "minimum": 0.9823121833190053, "maximum": 0.9893653981577238, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9195009508926018, "standard_deviation": 0.036343793462821655, "minimum": 1.8770056802131971, "maximum": 1.9716557409734017, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5245006486314308, "standard_deviation": 0.2790723703475166, "minimum": 0.09999012659315054, "maximum": 0.977884679200122, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13794754051056232, "standard_deviation": 0.011415546753873214, "minimum": 0.1176158441721202, "maximum": 0.15135834466695727, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9938511499191534, "standard_deviation": 0.0012178078268535062, "minimum": 0.9923851028608318, "maximum": 0.9954077061065784, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1432094259126133, "standard_deviation": 0.043465282924905375, "minimum": 1.0925292895352723, "maximum": 1.2265730506693393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6614162605003279, "standard_deviation": 0.352195438110825, "minimum": 0.2857074678283055, "maximum": 1.2259131244193018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08750332350923115, "standard_deviation": 0.00947393226301585, "minimum": 0.07569993660184508, "maximum": 0.10077457334931582, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20324263512725016, "standard_deviation": 0.038478883144904476, "minimum": 0.14462498115703223, "maximum": 0.2715214167184396, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.06462891466788, "standard_deviation": 0.005181026198585449, "minimum": 6.056109469956264, "maximum": 6.069574128798408, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6277288977798641, "standard_deviation": 0.015383613234111894, "minimum": 0.6088429249160301, "maximum": 0.6535210472289148, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09883527660714189, "standard_deviation": 0.03261818811266043, "minimum": -0.14727191385473376, "maximum": -0.04517342751661506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9925700456141195, "standard_deviation": 0.001096022641245153, "minimum": 0.9910487968112017, "maximum": 0.9936226188784953, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.108952346025176, "standard_deviation": 0.03159986885854563, "minimum": 1.0642559468925077, "maximum": 1.1503365269937325, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32938718479124995, "standard_deviation": 0.11231494933981638, "minimum": 0.17726964842564108, "maximum": 0.5265185952868393, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08581657242456711, "standard_deviation": 0.006173164073478991, "minimum": 0.0796209427433675, "maximum": 0.09447881620189837, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972936473726598, "standard_deviation": 0.0007743341216345391, "minimum": 0.9963609973355416, "maximum": 0.9981731086657406, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6055349710441705, "standard_deviation": 0.035112566059559154, "minimum": 0.5573862082609475, "maximum": 0.6571491658120325, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4000788093601213, "standard_deviation": 0.20709068214238394, "minimum": 0.08822876871399678, "maximum": 0.834130818037211, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.049211422620032655, "standard_deviation": 0.006986325906139868, "minimum": 0.04095810547669491, "maximum": 0.05749572741849621, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7176-88083-0026", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.3130977153778076, "predicted_mos": 2.3130977153778076, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "2dcce639af17c19500d764a6c8073143e9bbcaa162fa8a3a7515babcf6059ef3", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.783261700067669, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.075801817496445, "全帯域_10ms_rms_envelope_nrmse": 0.08791507853344162, "全帯域_10ms_envelope_correlation": 0.9921825801253262, "全帯域_10ms_modulation_mae_db": 0.11877510640914413, "全帯域_40ms_log_rms_mae_db": 0.5813110501121278, "全帯域_40ms_rms_envelope_nrmse": 0.04493155945115816, "全帯域_40ms_envelope_correlation": 0.9977524931258598, "全帯域_40ms_modulation_mae_db": 0.3207263639289396, "300–1000 Hz_10ms_log_rms_mae_db": 1.839583074000309, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11943140098853115, "300–1000 Hz_10ms_envelope_correlation": 0.9891412572843068, "300–1000 Hz_10ms_modulation_mae_db": 0.46409168033087056, "300–1000 Hz_40ms_log_rms_mae_db": 1.0286201637116925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07100369234709039, "300–1000 Hz_40ms_envelope_correlation": 0.996054080799831, "300–1000 Hz_40ms_modulation_mae_db": 0.5752376312024415, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3149895423075266, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21659004285080058, "1000–3000 Hz_10ms_envelope_correlation": 0.9644659361830685, "1000–3000 Hz_10ms_modulation_mae_db": 1.1238441054281894, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4446037711975268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12408938814456782, "1000–3000 Hz_40ms_envelope_correlation": 0.9881073748513837, "1000–3000 Hz_40ms_modulation_mae_db": 1.2853113406687766, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6019821877187463, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24322320983592044, "3000–6000 Hz_10ms_envelope_correlation": 0.9572397892059822, "3000–6000 Hz_10ms_modulation_mae_db": 0.38833204923022074, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7395371838655047, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1573157262502848, "3000–6000 Hz_40ms_envelope_correlation": 0.980054109050295, "3000–6000 Hz_40ms_modulation_mae_db": 0.5856977441187455, "mr_stft_db": 6.044946624481703, "band_mae_db": 0.19954869498079633, "rms_gain_error_db": -0.08080004203342739, "output_peak": 0.6309706515142317}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0634734222796263, "全帯域_10ms_rms_envelope_nrmse": 0.08308578432598404, "全帯域_10ms_envelope_correlation": 0.9930387210744307, "全帯域_10ms_modulation_mae_db": 0.2355698968868473, "全帯域_40ms_log_rms_mae_db": 0.5705905663192413, "全帯域_40ms_rms_envelope_nrmse": 0.04489802118841922, "全帯域_40ms_envelope_correlation": 0.9977701278709152, "全帯域_40ms_modulation_mae_db": 0.20822461910544585, "300–1000 Hz_10ms_log_rms_mae_db": 1.838483076281888, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12919378004379778, "300–1000 Hz_10ms_envelope_correlation": 0.987164694053957, "300–1000 Hz_10ms_modulation_mae_db": 0.2686502405279883, "300–1000 Hz_40ms_log_rms_mae_db": 1.0869533696208786, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08261139119337912, "300–1000 Hz_40ms_envelope_correlation": 0.994424517084346, "300–1000 Hz_40ms_modulation_mae_db": 0.3213053800044558, "1000–3000 Hz_10ms_log_rms_mae_db": 2.464984838238334, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22210087618085161, "1000–3000 Hz_10ms_envelope_correlation": 0.9621504159686183, "1000–3000 Hz_10ms_modulation_mae_db": 0.7134828088374269, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5917616265650452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13130436904193327, "1000–3000 Hz_40ms_envelope_correlation": 0.9855662719015668, "1000–3000 Hz_40ms_modulation_mae_db": 0.9793571802983037, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6430420481389985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2428671982907614, "3000–6000 Hz_10ms_envelope_correlation": 0.9573295405116377, "3000–6000 Hz_10ms_modulation_mae_db": 1.335705981149536, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7780382145988638, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15830552792772506, "3000–6000 Hz_40ms_envelope_correlation": 0.9807395734112128, "3000–6000 Hz_40ms_modulation_mae_db": 0.9422809004360172, "mr_stft_db": 6.056192085804163, "band_mae_db": 0.20005324214304562, "rms_gain_error_db": -0.0883722386043482, "output_peak": 0.6541725721784297}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0508083631401226, "全帯域_10ms_rms_envelope_nrmse": 0.08113167791443339, "全帯域_10ms_envelope_correlation": 0.9934228829961913, "全帯域_10ms_modulation_mae_db": 0.3232423087072513, "全帯域_40ms_log_rms_mae_db": 0.5730150674960856, "全帯域_40ms_rms_envelope_nrmse": 0.045527827122661645, "全帯域_40ms_envelope_correlation": 0.9977654100461973, "全帯域_40ms_modulation_mae_db": 0.6110358947627197, "300–1000 Hz_10ms_log_rms_mae_db": 1.802641453626251, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1229737240649187, "300–1000 Hz_10ms_envelope_correlation": 0.9887321018808375, "300–1000 Hz_10ms_modulation_mae_db": 0.5861749793266698, "300–1000 Hz_40ms_log_rms_mae_db": 0.997702998363922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06966651473854928, "300–1000 Hz_40ms_envelope_correlation": 0.9965900564947293, "300–1000 Hz_40ms_modulation_mae_db": 0.573850473071656, "1000–3000 Hz_10ms_log_rms_mae_db": 2.4047648194318803, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22065716398208363, "1000–3000 Hz_10ms_envelope_correlation": 0.9651136857091551, "1000–3000 Hz_10ms_modulation_mae_db": 1.4151965945810039, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6027441700792098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13890748291038194, "1000–3000 Hz_40ms_envelope_correlation": 0.987778487096493, "1000–3000 Hz_40ms_modulation_mae_db": 1.5538043052507924, "3000–6000 Hz_10ms_log_rms_mae_db": 2.629029082456886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25521542095441135, "3000–6000 Hz_10ms_envelope_correlation": 0.9524334624466239, "3000–6000 Hz_10ms_modulation_mae_db": 1.1227623443546981, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7777072932870457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1663542464398411, "3000–6000 Hz_40ms_envelope_correlation": 0.9778757031938271, "3000–6000 Hz_40ms_modulation_mae_db": 0.9698626204377021, "mr_stft_db": 6.0454215826446225, "band_mae_db": 0.261798233200317, "rms_gain_error_db": -0.10271160823323841, "output_peak": 0.610586976391011}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1071530671875038, "全帯域_10ms_rms_envelope_nrmse": 0.10022458877551174, "全帯域_10ms_envelope_correlation": 0.9898047915408972, "全帯域_10ms_modulation_mae_db": 0.3248130056153014, "全帯域_40ms_log_rms_mae_db": 0.6021840437159693, "全帯域_40ms_rms_envelope_nrmse": 0.05454389978451491, "全帯域_40ms_envelope_correlation": 0.9966240660402979, "全帯域_40ms_modulation_mae_db": 0.1950568110292811, "300–1000 Hz_10ms_log_rms_mae_db": 1.8909456789289443, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12949323147615527, "300–1000 Hz_10ms_envelope_correlation": 0.9870833906087599, "300–1000 Hz_10ms_modulation_mae_db": 0.5801044966364318, "300–1000 Hz_40ms_log_rms_mae_db": 1.1695307805841373, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07711905510903362, "300–1000 Hz_40ms_envelope_correlation": 0.9951786006469818, "300–1000 Hz_40ms_modulation_mae_db": 0.8147455904684459, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3860674049195207, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21140124891790546, "1000–3000 Hz_10ms_envelope_correlation": 0.9657447491025485, "1000–3000 Hz_10ms_modulation_mae_db": 0.48210915962663087, "1000–3000 Hz_40ms_log_rms_mae_db": 1.540311907656782, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12592954947639104, "1000–3000 Hz_40ms_envelope_correlation": 0.9866358331256827, "1000–3000 Hz_40ms_modulation_mae_db": 0.7395634595671313, "3000–6000 Hz_10ms_log_rms_mae_db": 2.636798726175459, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24150912036326933, "3000–6000 Hz_10ms_envelope_correlation": 0.9577068433003498, "3000–6000 Hz_10ms_modulation_mae_db": 1.3773524366240812, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7478206077561116, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15958014734750106, "3000–6000 Hz_40ms_envelope_correlation": 0.9799127287217319, "3000–6000 Hz_40ms_modulation_mae_db": 1.1333940541286867, "mr_stft_db": 6.04048546937851, "band_mae_db": 0.1884173219624544, "rms_gain_error_db": -0.07533478415528852, "output_peak": 0.725020108112007}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.056933316550978, "全帯域_10ms_rms_envelope_nrmse": 0.09314411627449325, "全帯域_10ms_envelope_correlation": 0.9912965403877664, "全帯域_10ms_modulation_mae_db": 0.6252846396255851, "全帯域_40ms_log_rms_mae_db": 0.5788713017035029, "全帯域_40ms_rms_envelope_nrmse": 0.05951582231435309, "全帯域_40ms_envelope_correlation": 0.9960893190347988, "全帯域_40ms_modulation_mae_db": 0.8870184997042071, "300–1000 Hz_10ms_log_rms_mae_db": 1.855829840254203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14329787193904753, "300–1000 Hz_10ms_envelope_correlation": 0.9844586805864384, "300–1000 Hz_10ms_modulation_mae_db": 0.7014573276501684, "300–1000 Hz_40ms_log_rms_mae_db": 1.1011973034254174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09191340770726741, "300–1000 Hz_40ms_envelope_correlation": 0.9935042890610988, "300–1000 Hz_40ms_modulation_mae_db": 0.7921980025703919, "1000–3000 Hz_10ms_log_rms_mae_db": 2.356971790927903, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2128176924286212, "1000–3000 Hz_10ms_envelope_correlation": 0.9655051993674126, "1000–3000 Hz_10ms_modulation_mae_db": 1.2109862134147356, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5265336309621265, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1281163094282966, "1000–3000 Hz_40ms_envelope_correlation": 0.9866149931487465, "1000–3000 Hz_40ms_modulation_mae_db": 1.6315906990248088, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6515466178963365, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2399590553049842, "3000–6000 Hz_10ms_envelope_correlation": 0.958023971341158, "3000–6000 Hz_10ms_modulation_mae_db": 0.5611931649233819, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8076458723475415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1587476663546314, "3000–6000 Hz_40ms_envelope_correlation": 0.9799192764243053, "3000–6000 Hz_40ms_modulation_mae_db": 0.6515751820857685, "mr_stft_db": 6.048635860647, "band_mae_db": 0.22916861424191712, "rms_gain_error_db": -0.12225506508542777, "output_peak": 0.6471352737753906}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0455273750269403, "全帯域_10ms_rms_envelope_nrmse": 0.08308186839183615, "全帯域_10ms_envelope_correlation": 0.9929959740980545, "全帯域_10ms_modulation_mae_db": 0.2118745789436717, "全帯域_40ms_log_rms_mae_db": 0.551117719456081, "全帯域_40ms_rms_envelope_nrmse": 0.04766315840123902, "全帯域_40ms_envelope_correlation": 0.9973958788041432, "全帯域_40ms_modulation_mae_db": 0.33878830685155353, "300–1000 Hz_10ms_log_rms_mae_db": 1.9204017414818984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12691048581520287, "300–1000 Hz_10ms_envelope_correlation": 0.9874611984489793, "300–1000 Hz_10ms_modulation_mae_db": 0.165775579912841, "300–1000 Hz_40ms_log_rms_mae_db": 1.0936718176948153, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06771041735946903, "300–1000 Hz_40ms_envelope_correlation": 0.9960652249124915, "300–1000 Hz_40ms_modulation_mae_db": 0.21693118113841303, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2871091289217462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20059941813197435, "1000–3000 Hz_10ms_envelope_correlation": 0.9692876459385111, "1000–3000 Hz_10ms_modulation_mae_db": 0.6517153963964566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4463333344072775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10911684000712464, "1000–3000 Hz_40ms_envelope_correlation": 0.9898404531283612, "1000–3000 Hz_40ms_modulation_mae_db": 0.8600088258537744, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6334192427324994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2436875602030167, "3000–6000 Hz_10ms_envelope_correlation": 0.9569329105532934, "3000–6000 Hz_10ms_modulation_mae_db": 0.9057616014883519, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7807054100783453, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15459283303598775, "3000–6000 Hz_40ms_envelope_correlation": 0.9808993127738449, "3000–6000 Hz_40ms_modulation_mae_db": 1.0160943121465271, "mr_stft_db": 6.029404050469079, "band_mae_db": 0.07702787008501771, "rms_gain_error_db": -0.04325668978164253, "output_peak": 0.6462386785836811}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0984464565453027, "全帯域_10ms_rms_envelope_nrmse": 0.07321018748615406, "全帯域_10ms_envelope_correlation": 0.9945663053843743, "全帯域_10ms_modulation_mae_db": 0.21492696222550542, "全帯域_40ms_log_rms_mae_db": 0.5808571411100866, "全帯域_40ms_rms_envelope_nrmse": 0.03781768002154236, "全帯域_40ms_envelope_correlation": 0.9983613907523121, "全帯域_40ms_modulation_mae_db": 0.08704209751234082, "300–1000 Hz_10ms_log_rms_mae_db": 1.8793781079118315, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11574186282421194, "300–1000 Hz_10ms_envelope_correlation": 0.9896985949910734, "300–1000 Hz_10ms_modulation_mae_db": 0.45022279698456874, "300–1000 Hz_40ms_log_rms_mae_db": 1.0617098850003723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06459702053023804, "300–1000 Hz_40ms_envelope_correlation": 0.9966222108299378, "300–1000 Hz_40ms_modulation_mae_db": 0.6747614639669056, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3362136622473413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21350018795137538, "1000–3000 Hz_10ms_envelope_correlation": 0.9655386681522652, "1000–3000 Hz_10ms_modulation_mae_db": 1.2897657820148354, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5173517627859618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11978282884411032, "1000–3000 Hz_40ms_envelope_correlation": 0.9890492679735008, "1000–3000 Hz_40ms_modulation_mae_db": 1.4404482818159838, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7157418262385677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.261931876907854, "3000–6000 Hz_10ms_envelope_correlation": 0.9501104958855072, "3000–6000 Hz_10ms_modulation_mae_db": 0.3027606843658316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8579207586019135, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16847224567075528, "3000–6000 Hz_40ms_envelope_correlation": 0.97713897477574, "3000–6000 Hz_40ms_modulation_mae_db": 0.46610718302008, "mr_stft_db": 6.037459367551544, "band_mae_db": 0.17514726132645994, "rms_gain_error_db": -0.03400458961574482, "output_peak": 0.6210126636344139}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1099703014888225, "全帯域_10ms_rms_envelope_nrmse": 0.08895242052841701, "全帯域_10ms_envelope_correlation": 0.9921244725257752, "全帯域_10ms_modulation_mae_db": 0.3896133915177198, "全帯域_40ms_log_rms_mae_db": 0.601929100343519, "全帯域_40ms_rms_envelope_nrmse": 0.05451920082635861, "全帯域_40ms_envelope_correlation": 0.9968207145576952, "全帯域_40ms_modulation_mae_db": 0.3290561879004616, "300–1000 Hz_10ms_log_rms_mae_db": 1.8062781402361479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1229856083581156, "300–1000 Hz_10ms_envelope_correlation": 0.9885692157485436, "300–1000 Hz_10ms_modulation_mae_db": 0.6357322177862545, "300–1000 Hz_40ms_log_rms_mae_db": 1.000735729839235, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07087354915407791, "300–1000 Hz_40ms_envelope_correlation": 0.9962595868248005, "300–1000 Hz_40ms_modulation_mae_db": 0.7643026651314365, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3721012544881424, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21853213184148926, "1000–3000 Hz_10ms_envelope_correlation": 0.9652082737097503, "1000–3000 Hz_10ms_modulation_mae_db": 1.8121320504984555, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4659509055488789, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13566090772229983, "1000–3000 Hz_40ms_envelope_correlation": 0.9876407351231621, "1000–3000 Hz_40ms_modulation_mae_db": 1.9677447489099882, "3000–6000 Hz_10ms_log_rms_mae_db": 2.643426983586308, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24382956742735432, "3000–6000 Hz_10ms_envelope_correlation": 0.9574184321968605, "3000–6000 Hz_10ms_modulation_mae_db": 0.5924863963276547, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7897558412758479, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15799939205544686, "3000–6000 Hz_40ms_envelope_correlation": 0.9800148957088102, "3000–6000 Hz_40ms_modulation_mae_db": 0.6944473953442837, "mr_stft_db": 6.053548267721205, "band_mae_db": 0.24778066988696942, "rms_gain_error_db": -0.1370263117211295, "output_peak": 0.6871236402225231}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88480, "output_samples": 88480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.65199829177599, "spectral_convergence": 0.2909329555098987, "spectral_convergence_status": "ok", "amplitude_floor": 2.1848847242256272e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.825631647230173, "spectral_convergence": 0.2687230196052209, "spectral_convergence_status": "ok", "amplitude_floor": 1.2036494394056804e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.657209934438946, "spectral_convergence": 0.2863345009382881, "spectral_convergence_status": "ok", "amplitude_floor": 7.153950066191215e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002200971739696214, "output_power": 0.0022617918456292515, "delta_db": 0.11838167704987404, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002201298666996261, "output_power": 0.002059998991218137, "delta_db": -0.28811962903964106, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011542423166681864, "output_power": 0.0011503689741987758, "delta_db": -0.014598321647966686, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019781669780676847, "output_power": 0.00017691109448146332, "delta_db": -0.48507878503600504, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.802033486419172e-05, "output_power": 5.546431828657598e-05, "delta_db": -0.19566551570545387, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6630669693501258e-05, "output_power": 1.7000220802936495e-05, "delta_db": 0.09544824140583738, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19954869498079633, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08080004203342739, "pitch": null, "output_rms_dbfs": -22.424874072839387, "output_peak": 0.6309706515142317, "reference_rms_dbfs": -22.34407403080596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.044946624481703}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00024519594682554, "log_rms_mae_db": 1.075801817496445, "rms_envelope_nrmse": 0.08791507853344162, "envelope_correlation": 0.9921825801253262, "rms_gain_error_db": -0.08080014753338333, "reference_mean_power": 0.00583423429533283, "output_mean_power": 0.005726692331086446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.472320009859619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016537460501756134, "output_power": 0.0016513340385231983, "reference_active": true, "delta_db": -0.006338869308975298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003778384775464376, "output_power": 0.00035102829466871724, "reference_active": true, "delta_db": -0.3196405841177945}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8517105494706513e-05, "output_power": 4.8857300772995814e-05, "reference_active": true, "delta_db": 0.03034586580066259}], "active_band_count": 3, "active_band_mae_db": 0.11877510640914413}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00020002070942310007, "log_rms_mae_db": 0.5813110501121278, "rms_envelope_nrmse": 0.04493155945115816, "envelope_correlation": 0.9977524931258598, "rms_gain_error_db": -0.08079015852360998, "reference_mean_power": 0.005865946925483087, "output_mean_power": 0.005757833648231279}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.084203911038529e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015941324207372327, "output_power": 0.0015947275700946585, "reference_active": true, "delta_db": 0.0016210814245995837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018248750335661006, "output_power": 0.00017286466866296414, "reference_active": true, "delta_db": -0.23526891482624682}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.552526527817181e-06, "output_power": 3.852330823215471e-06, "reference_active": true, "delta_db": -0.7252890955359724}], "active_band_count": 3, "active_band_mae_db": 0.3207263639289396}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00018690652054173922, "log_rms_mae_db": 1.839583074000309, "rms_envelope_nrmse": 0.11943140098853115, "envelope_correlation": 0.9891412572843068, "rms_gain_error_db": -0.20082321117845936, "reference_mean_power": 0.0020929267743018004, "output_mean_power": 0.0019983507265060973}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 8.903808595053404e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005237826202169963, "output_power": 0.0005132500795700151, "reference_active": true, "delta_db": -0.08822058505117597}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020282121471265822, "output_power": 0.000186227811003829, "reference_active": true, "delta_db": -0.37068840986510787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.57818052122161e-05, "output_power": 2.079585201380991e-05, "reference_active": true, "delta_db": -0.9333660460763278}], "active_band_count": 3, "active_band_mae_db": 0.46409168033087056}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00016950242057071555, "log_rms_mae_db": 1.0286201637116925, "rms_envelope_nrmse": 0.07100369234709039, "envelope_correlation": 0.996054080799831, "rms_gain_error_db": -0.20081902692003245, "reference_mean_power": 0.002104315784196243, "output_mean_power": 0.0020092270208835455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 7.681995008528509e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000538193996495516, "output_power": 0.0005298352152501403, "reference_active": true, "delta_db": -0.06798028839364174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010305287620923108, "output_power": 9.522234484744756e-05, "reference_active": true, "delta_db": -0.34321245698898684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.715218896542221e-06, "output_power": 2.0061011765642757e-06, "reference_active": true, "delta_db": -1.314520148224696}], "active_band_count": 3, "active_band_mae_db": 0.5752376312024415}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 7.550458878286076e-05, "log_rms_mae_db": 2.3149895423075266, "rms_envelope_nrmse": 0.21659004285080058, "envelope_correlation": 0.9644659361830685, "rms_gain_error_db": -0.4123059004808086, "reference_mean_power": 0.00021315330376123098, "output_mean_power": 0.00019384806861517934}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 1.6207032137826829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010202948507054384, "output_power": 9.35326793130383e-05, "reference_active": true, "delta_db": -0.3776231961904004}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.692530062160877e-05, "output_power": 3.2750909860632566e-05, "reference_active": true, "delta_db": -0.5210067000829942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.752197568216375e-06, "output_power": 3.8208052265155144e-06, "reference_active": true, "delta_db": -2.4729024200111733}], "active_band_count": 3, "active_band_mae_db": 1.1238441054281894}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 5.938009261210461e-05, "log_rms_mae_db": 1.4446037711975268, "rms_envelope_nrmse": 0.12408938814456782, "envelope_correlation": 0.9881073748513837, "rms_gain_error_db": -0.41214275569886427, "reference_mean_power": 0.00021425147153668813, "output_mean_power": 0.00019485409529117055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 1.4064708995422735e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010396272320640846, "output_power": 9.553933213197041e-05, "reference_active": true, "delta_db": -0.3669544589530952}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97231105181685e-05, "output_power": 1.7191787252680877e-05, "reference_active": true, "delta_db": -0.5965438014125729}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.307963255185344e-07, "output_power": 3.24075036540382e-07, "reference_active": true, "delta_db": -2.892435761640662}], "active_band_count": 3, "active_band_mae_db": 1.2853113406687766}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 4.4291934004628345e-05, "log_rms_mae_db": 2.6019821877187463, "rms_envelope_nrmse": 0.24322320983592044, "envelope_correlation": 0.9572397892059822, "rms_gain_error_db": -0.051390218079872464, "reference_mean_power": 2.6417005173666327e-05, "output_mean_power": 2.6106254017263638e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.694560310630941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.502841409588563e-05, "output_power": 1.4274821853039936e-05, "reference_active": true, "delta_db": -0.22342456149749138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.7026465908103935e-06, "output_power": 6.612141238940506e-06, "reference_active": true, "delta_db": -0.05904199268395842}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.011540321330325e-07, "output_power": 7.35438008203007e-07, "reference_active": true, "delta_db": -0.8825295935092125}], "active_band_count": 3, "active_band_mae_db": 0.38833204923022074}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 3.2524867537670264e-05, "log_rms_mae_db": 1.7395371838655047, "rms_envelope_nrmse": 0.1573157262502848, "envelope_correlation": 0.980054109050295, "rms_gain_error_db": -0.0514653161061912, "reference_mean_power": 2.65459352580292e-05, "output_mean_power": 2.623321383077228e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.4345030346700597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6238306479668564e-05, "output_power": 1.5027140641715801e-05, "reference_active": true, "delta_db": -0.336643828954156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6930369104619132e-06, "output_power": 3.7735080542318314e-06, "reference_active": true, "delta_db": 0.093616330165816}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.897644209500044e-08, "output_power": 5.8185431244059543e-08, "reference_active": true, "delta_db": -1.3268330732362645}], "active_band_count": 3, "active_band_mae_db": 0.5856977441187455}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8889874891855623, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "079749c9804855f19908d1a686d33abc0ba34d3d68bb8e812be96caa552e6fa5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9653768217664163, "standard_deviation": 0.0019570257039270016, "minimum": 0.9621504159686183, "maximum": 0.9692876459385111, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.3654003051852994, "standard_deviation": 0.055462122482012814, "minimum": 2.2871091289217462, "maximum": 2.464984838238334, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0874040138497167, "standard_deviation": 0.4450125083492156, "minimum": 0.48210915962663087, "maximum": 1.8121320504984555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21452484528563767, "standard_deviation": 0.0067924709627041275, "minimum": 0.20059941813197435, "maximum": 0.22210087618085161, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9876541770436121, "standard_deviation": 0.0013847835897742551, "minimum": 0.9855662719015668, "maximum": 0.9898404531283612, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.516948888650351, "standard_deviation": 0.061431498575166, "minimum": 1.4446037711975268, "maximum": 1.6027441700792098, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.307228605173695, "standard_deviation": 0.4226225415632398, "minimum": 0.7395634595671313, "maximum": 1.9677447489099882, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12661345944688818, "standard_deviation": 0.009387608847770318, "minimum": 0.10911684000712464, "maximum": 0.13890748291038194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9558994306801766, "standard_deviation": 0.0029404163120265382, "minimum": 0.9501104958855072, "maximum": 0.958023971341158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.6443733393679754, "standard_deviation": 0.032420028476814515, "minimum": 2.6019821877187463, "maximum": 2.7157418262385677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8232943323079696, "standard_deviation": 0.42226342711430054, "minimum": 0.3027606843658316, "maximum": 1.3773524366240812, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24652787616094646, "standard_deviation": 0.007753005292836945, "minimum": 0.2399590553049842, "maximum": 0.261931876907854, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9795693217574709, "standard_deviation": 0.001340961638142385, "minimum": 0.97713897477574, "maximum": 0.9808993127738449, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7848913977263967, "standard_deviation": 0.03668799848206836, "minimum": 1.7395371838655047, "maximum": 1.8579207586019135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8074324239647263, "standard_deviation": 0.23817302628323964, "minimum": 0.46610718302008, "maximum": 1.1333940541286867, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16017097313527168, "standard_deviation": 0.0047356188873075005, "minimum": 0.15459283303598775, "maximum": 0.16847224567075528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.987788641700362, "standard_deviation": 0.001650084557064828, "minimum": 0.9844586805864384, "maximum": 0.9896985949910734, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8541926390901842, "standard_deviation": 0.04103130011386256, "minimum": 1.802641453626251, "maximum": 1.9204017414818984, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48152616489447414, "standard_deviation": 0.18482298022558027, "minimum": 0.165775579912841, "maximum": 0.7014573276501684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1262534956887476, "standard_deviation": 0.008354254253599494, "minimum": 0.11574186282421194, "maximum": 0.14329787193904753, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955873208317771, "standard_deviation": 0.0011233153077081731, "minimum": 0.9935042890610988, "maximum": 0.9966222108299378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0675152560300587, "standard_deviation": 0.057920438685739255, "minimum": 0.997702998363922, "maximum": 1.1695307805841373, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5916665484442682, "standard_deviation": 0.2206465161514221, "minimum": 0.21693118113841303, "maximum": 0.8147455904684459, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0744368810173881, "standard_deviation": 0.009013912581697942, "minimum": 0.06459702053023804, "maximum": 0.09191340770726741, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1973677384783722, "standard_deviation": 0.05703577823167075, "minimum": 0.07702787008501771, "maximum": 0.261798233200317, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.044511663587228, "standard_deviation": 0.008703888423388354, "minimum": 6.029404050469079, "maximum": 6.056192085804163, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6527825705514609, "standard_deviation": 0.03730142174733499, "minimum": 0.610586976391011, "maximum": 0.725020108112007, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08547016615378089, "standard_deviation": 0.03560111047230559, "minimum": -0.1370263117211295, "maximum": -0.03400458961574482, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924290335166019, "standard_deviation": 0.0014427895436586635, "minimum": 0.9898047915408972, "maximum": 0.9945663053843743, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0760142649644675, "standard_deviation": 0.02594507673038859, "minimum": 1.0455273750269403, "maximum": 1.1099703014888225, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3055124862413783, "standard_deviation": 0.15434222581078724, "minimum": 0.11877510640914413, "maximum": 0.6252846396255851, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08634321527878391, "standard_deviation": 0.008186288809138872, "minimum": 0.07321018748615406, "maximum": 0.10022458877551174, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973224250290273, "standard_deviation": 0.0007489465388521292, "minimum": 0.9960893190347988, "maximum": 0.9983613907523121, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5799844987820767, "standard_deviation": 0.016682992502992, "minimum": 0.551117719456081, "maximum": 0.6021840437159693, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37211859759936866, "standard_deviation": 0.258371423816646, "minimum": 0.08704209751234082, "maximum": 0.8870184997042071, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04867714613878088, "standard_deviation": 0.007002151642019641, "minimum": 0.03781768002154236, "maximum": 0.05951582231435309, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7729-102255-0028", "language": "en", "style": "normal", "speaker": "librispeech_7729", "source_utterance_id": "7729-102255-0028", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7729-102255-0028.wav", "source_file_sha256": "0d3626b297b28d736c5fd61a7e940661a04e979fa4e6347d27c97d0bd60c26a8", "decoded_pcm_sha256": "7a4ab1cb22b23d73175ff25f3f2557bc2724e5cd7a3952f1cc5626d2a9bfbd3f", "original_source_sha256": "4654ddc035bcd2a1c3def2c04709f8f93eff6f9d9da968a1f044f65edf3a5b15", "corpus_entry_id": "librispeech_7729-102255-0028"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.1194844245910645, "predicted_mos": 4.1194844245910645, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2dd107c270a40dd1c3df5531d50e44b8fed6e67e55658ed3cf75430f90f39a37", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.419066908998502, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7729-102255-0028", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "2dd107c270a40dd1c3df5531d50e44b8fed6e67e55658ed3cf75430f90f39a37", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "bdefd8b41a55fe29ad275ad0cdc0930d498f0e9c563afc84f3f746edbb5622c1"}, "rows": [{"case": "librispeech_7729-102255-0028", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.6804444789886475, "predicted_mos": 2.6804444789886475, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0cea792a0ff047d051df404eea5a1df2d280320af27f5db066c37d8548b1515a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.671989290000056, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7729-102255-0028", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "0cea792a0ff047d051df404eea5a1df2d280320af27f5db066c37d8548b1515a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8289138257108376, "全帯域_10ms_rms_envelope_nrmse": 0.07234304411445358, "全帯域_10ms_envelope_correlation": 0.9945707173893922, "全帯域_10ms_modulation_mae_db": 0.4598495251577503, "全帯域_40ms_log_rms_mae_db": 0.4653608451904468, "全帯域_40ms_rms_envelope_nrmse": 0.03996374716539286, "全帯域_40ms_envelope_correlation": 0.9982776986068054, "全帯域_40ms_modulation_mae_db": 0.43132172240126687, "300–1000 Hz_10ms_log_rms_mae_db": 1.5009618874186526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10109367830209409, "300–1000 Hz_10ms_envelope_correlation": 0.991006387939488, "300–1000 Hz_10ms_modulation_mae_db": 0.487984929463311, "300–1000 Hz_40ms_log_rms_mae_db": 1.0176619654148231, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05698446033684822, "300–1000 Hz_40ms_envelope_correlation": 0.997105995262388, "300–1000 Hz_40ms_modulation_mae_db": 0.41676683827971955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7863823675519024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17177325806657717, "1000–3000 Hz_10ms_envelope_correlation": 0.9779367386429185, "1000–3000 Hz_10ms_modulation_mae_db": 0.641729486755357, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2133661033137064, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0935775416264806, "1000–3000 Hz_40ms_envelope_correlation": 0.9937322047079662, "1000–3000 Hz_40ms_modulation_mae_db": 0.385946674461471, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9074377297390983, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1945171596099572, "3000–6000 Hz_10ms_envelope_correlation": 0.9733215841310912, "3000–6000 Hz_10ms_modulation_mae_db": 0.8999600618270863, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4662310384250246, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12934064736756615, "3000–6000 Hz_40ms_envelope_correlation": 0.989069728692841, "3000–6000 Hz_40ms_modulation_mae_db": 0.798588506961296, "mr_stft_db": 5.868014600905933, "band_mae_db": 0.19484830857246457, "rms_gain_error_db": -0.08556196092638113, "output_peak": 0.732358975637002}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.885012548642195, "全帯域_10ms_rms_envelope_nrmse": 0.08784686991568964, "全帯域_10ms_envelope_correlation": 0.991905873633453, "全帯域_10ms_modulation_mae_db": 0.6866238230835319, "全帯域_40ms_log_rms_mae_db": 0.5172892989557791, "全帯域_40ms_rms_envelope_nrmse": 0.05270748466161683, "全帯域_40ms_envelope_correlation": 0.9968702670648474, "全帯域_40ms_modulation_mae_db": 0.736092718020795, "300–1000 Hz_10ms_log_rms_mae_db": 1.5970120590202306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12385029797705557, "300–1000 Hz_10ms_envelope_correlation": 0.9862746029998866, "300–1000 Hz_10ms_modulation_mae_db": 0.6311814679733703, "300–1000 Hz_40ms_log_rms_mae_db": 1.0833026827175998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07365158756864239, "300–1000 Hz_40ms_envelope_correlation": 0.9947780592791423, "300–1000 Hz_40ms_modulation_mae_db": 0.7871487953774009, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7652632809983535, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17627739831811512, "1000–3000 Hz_10ms_envelope_correlation": 0.9768889679671687, "1000–3000 Hz_10ms_modulation_mae_db": 0.6248342833496227, "1000–3000 Hz_40ms_log_rms_mae_db": 1.276148286798015, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10063193069021358, "1000–3000 Hz_40ms_envelope_correlation": 0.9929080925058434, "1000–3000 Hz_40ms_modulation_mae_db": 0.7753396379692395, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9142504675823677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20298835270133275, "3000–6000 Hz_10ms_envelope_correlation": 0.9701033889435816, "3000–6000 Hz_10ms_modulation_mae_db": 0.7885877480120064, "3000–6000 Hz_40ms_log_rms_mae_db": 1.478281272546841, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1411257471847126, "3000–6000 Hz_40ms_envelope_correlation": 0.9855670212820058, "3000–6000 Hz_40ms_modulation_mae_db": 0.7665655851027259, "mr_stft_db": 5.882273124607718, "band_mae_db": 0.2605224380079303, "rms_gain_error_db": -0.06573966549089634, "output_peak": 0.6669469868144029}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8891773892022926, "全帯域_10ms_rms_envelope_nrmse": 0.08232084797847453, "全帯域_10ms_envelope_correlation": 0.9930785498035078, "全帯域_10ms_modulation_mae_db": 0.2658540820512771, "全帯域_40ms_log_rms_mae_db": 0.5303426608382106, "全帯域_40ms_rms_envelope_nrmse": 0.04664200022533671, "全帯域_40ms_envelope_correlation": 0.9978389602407458, "全帯域_40ms_modulation_mae_db": 0.2009660141810258, "300–1000 Hz_10ms_log_rms_mae_db": 1.4797609469256858, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11549125435192178, "300–1000 Hz_10ms_envelope_correlation": 0.9883607080478916, "300–1000 Hz_10ms_modulation_mae_db": 0.41019248282860854, "300–1000 Hz_40ms_log_rms_mae_db": 1.0152669053370267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06728112711169254, "300–1000 Hz_40ms_envelope_correlation": 0.9961340059497954, "300–1000 Hz_40ms_modulation_mae_db": 0.5273651885162046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7859101400664108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1788087400226358, "1000–3000 Hz_10ms_envelope_correlation": 0.9776666139433671, "1000–3000 Hz_10ms_modulation_mae_db": 1.4670754444210514, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2491929187523183, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10643872557115919, "1000–3000 Hz_40ms_envelope_correlation": 0.9939124117881006, "1000–3000 Hz_40ms_modulation_mae_db": 1.0726503240693541, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9059143651945523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20077488838540036, "3000–6000 Hz_10ms_envelope_correlation": 0.9704108310953403, "3000–6000 Hz_10ms_modulation_mae_db": 0.5216018005208353, "3000–6000 Hz_40ms_log_rms_mae_db": 1.501365223870333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12999576327723597, "3000–6000 Hz_40ms_envelope_correlation": 0.9872215571274092, "3000–6000 Hz_40ms_modulation_mae_db": 0.6363983353199966, "mr_stft_db": 5.876642017233196, "band_mae_db": 0.30285482720083107, "rms_gain_error_db": -0.15265644763391154, "output_peak": 0.7705850813186453}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.905152759787651, "全帯域_10ms_rms_envelope_nrmse": 0.08242802983805997, "全帯域_10ms_envelope_correlation": 0.9931255089232938, "全帯域_10ms_modulation_mae_db": 0.4907520485495584, "全帯域_40ms_log_rms_mae_db": 0.5117957523504421, "全帯域_40ms_rms_envelope_nrmse": 0.04928368055284277, "全帯域_40ms_envelope_correlation": 0.9976091458290854, "全帯域_40ms_modulation_mae_db": 0.3498542317656244, "300–1000 Hz_10ms_log_rms_mae_db": 1.6315835428518939, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10784123707054168, "300–1000 Hz_10ms_envelope_correlation": 0.9898839369465231, "300–1000 Hz_10ms_modulation_mae_db": 0.4919874454019137, "300–1000 Hz_40ms_log_rms_mae_db": 1.1336715400109767, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06519442730878296, "300–1000 Hz_40ms_envelope_correlation": 0.996340620513455, "300–1000 Hz_40ms_modulation_mae_db": 0.3688057715624722, "1000–3000 Hz_10ms_log_rms_mae_db": 1.785206365804266, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17120665356615644, "1000–3000 Hz_10ms_envelope_correlation": 0.9796241827531295, "1000–3000 Hz_10ms_modulation_mae_db": 0.9029231013233204, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2808261852142222, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10480767623054556, "1000–3000 Hz_40ms_envelope_correlation": 0.9940079972589764, "1000–3000 Hz_40ms_modulation_mae_db": 1.2663958936219, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9030940286544984, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20051770920367007, "3000–6000 Hz_10ms_envelope_correlation": 0.9709337332676583, "3000–6000 Hz_10ms_modulation_mae_db": 0.6167319616943162, "3000–6000 Hz_40ms_log_rms_mae_db": 1.459592287539612, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11666438482735816, "3000–6000 Hz_40ms_envelope_correlation": 0.9908329831931931, "3000–6000 Hz_40ms_modulation_mae_db": 0.6463430514355085, "mr_stft_db": 5.870178370432893, "band_mae_db": 0.23635200804176623, "rms_gain_error_db": -0.16971095328298058, "output_peak": 0.6955126731982096}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8480767941297012, "全帯域_10ms_rms_envelope_nrmse": 0.08096421131901882, "全帯域_10ms_envelope_correlation": 0.9932047296760877, "全帯域_10ms_modulation_mae_db": 0.3360585263976274, "全帯域_40ms_log_rms_mae_db": 0.4362927044003156, "全帯域_40ms_rms_envelope_nrmse": 0.04093354330451298, "全帯域_40ms_envelope_correlation": 0.9982556768366319, "全帯域_40ms_modulation_mae_db": 0.37884509786076537, "300–1000 Hz_10ms_log_rms_mae_db": 1.5574674945202713, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11370953430864521, "300–1000 Hz_10ms_envelope_correlation": 0.9886550610660395, "300–1000 Hz_10ms_modulation_mae_db": 0.5648654168373064, "300–1000 Hz_40ms_log_rms_mae_db": 1.0002445497988128, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06188266741056775, "300–1000 Hz_40ms_envelope_correlation": 0.996703314097884, "300–1000 Hz_40ms_modulation_mae_db": 0.7301477334595302, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7412136461679104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17397382230089697, "1000–3000 Hz_10ms_envelope_correlation": 0.9786429883657766, "1000–3000 Hz_10ms_modulation_mae_db": 0.6332459681702426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1932538033803455, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1058095651710545, "1000–3000 Hz_40ms_envelope_correlation": 0.9936510308986309, "1000–3000 Hz_40ms_modulation_mae_db": 0.7867717998366889, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8937681558642374, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18071694849783074, "3000–6000 Hz_10ms_envelope_correlation": 0.9761274222125592, "3000–6000 Hz_10ms_modulation_mae_db": 0.4066456362196813, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4908849793448475, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11297960659943662, "3000–6000 Hz_40ms_envelope_correlation": 0.9904424919213952, "3000–6000 Hz_40ms_modulation_mae_db": 0.39642336137574813, "mr_stft_db": 5.861002419218717, "band_mae_db": 0.254477824441239, "rms_gain_error_db": -0.11341604456734532, "output_peak": 0.6870004406649706}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8733591648855453, "全帯域_10ms_rms_envelope_nrmse": 0.07728894608094479, "全帯域_10ms_envelope_correlation": 0.9939980121209121, "全帯域_10ms_modulation_mae_db": 0.4320086584169342, "全帯域_40ms_log_rms_mae_db": 0.4603907699357897, "全帯域_40ms_rms_envelope_nrmse": 0.04243672799826802, "全帯域_40ms_envelope_correlation": 0.998322046906428, "全帯域_40ms_modulation_mae_db": 0.2808362060430831, "300–1000 Hz_10ms_log_rms_mae_db": 1.564801475742827, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11122373673464087, "300–1000 Hz_10ms_envelope_correlation": 0.9894229720068687, "300–1000 Hz_10ms_modulation_mae_db": 0.5248486823299607, "300–1000 Hz_40ms_log_rms_mae_db": 1.0129002594113312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06478615645663995, "300–1000 Hz_40ms_envelope_correlation": 0.9966667383946227, "300–1000 Hz_40ms_modulation_mae_db": 0.48690245524985715, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7158108667700784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17236409196504604, "1000–3000 Hz_10ms_envelope_correlation": 0.9783600233882285, "1000–3000 Hz_10ms_modulation_mae_db": 1.4765183781495492, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1768820564976843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0999472669469087, "1000–3000 Hz_40ms_envelope_correlation": 0.9935717846431119, "1000–3000 Hz_40ms_modulation_mae_db": 1.0333495418450391, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8751188911340344, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19635000337886524, "3000–6000 Hz_10ms_envelope_correlation": 0.9733368622874832, "3000–6000 Hz_10ms_modulation_mae_db": 0.6243015307657919, "3000–6000 Hz_40ms_log_rms_mae_db": 1.417697882968764, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12885745858864434, "3000–6000 Hz_40ms_envelope_correlation": 0.989958977286654, "3000–6000 Hz_40ms_modulation_mae_db": 0.4296520157974355, "mr_stft_db": 5.8590329547751665, "band_mae_db": 0.23769097030159747, "rms_gain_error_db": -0.16648728069852123, "output_peak": 0.7006573873525156}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8816241844020237, "全帯域_10ms_rms_envelope_nrmse": 0.07776410106613203, "全帯域_10ms_envelope_correlation": 0.9937942284165271, "全帯域_10ms_modulation_mae_db": 0.4333092168480855, "全帯域_40ms_log_rms_mae_db": 0.4605837978378284, "全帯域_40ms_rms_envelope_nrmse": 0.042504814335032955, "全帯域_40ms_envelope_correlation": 0.9981642775192728, "全帯域_40ms_modulation_mae_db": 0.6657165878264577, "300–1000 Hz_10ms_log_rms_mae_db": 1.5360406640417783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10616437690580101, "300–1000 Hz_10ms_envelope_correlation": 0.9900744843927415, "300–1000 Hz_10ms_modulation_mae_db": 0.6450017987604473, "300–1000 Hz_40ms_log_rms_mae_db": 1.0293879861525184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06269290825287917, "300–1000 Hz_40ms_envelope_correlation": 0.9964638254056246, "300–1000 Hz_40ms_modulation_mae_db": 0.9058858396056468, "1000–3000 Hz_10ms_log_rms_mae_db": 1.751438074626612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16678644149445737, "1000–3000 Hz_10ms_envelope_correlation": 0.9798329096405544, "1000–3000 Hz_10ms_modulation_mae_db": 0.5969486881356495, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2527109427549963, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10402389807939261, "1000–3000 Hz_40ms_envelope_correlation": 0.9928907098572098, "1000–3000 Hz_40ms_modulation_mae_db": 0.3246222509823482, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8901161168556477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19911037532959525, "3000–6000 Hz_10ms_envelope_correlation": 0.9708948777417858, "3000–6000 Hz_10ms_modulation_mae_db": 0.8294637088553035, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4704696592899307, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1305452261099273, "3000–6000 Hz_40ms_envelope_correlation": 0.9870652142912175, "3000–6000 Hz_40ms_modulation_mae_db": 0.867866775670285, "mr_stft_db": 5.872517317587523, "band_mae_db": 0.2502491581816059, "rms_gain_error_db": -0.13071573328917907, "output_peak": 0.7148335556515647}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8740872685220815, "全帯域_10ms_rms_envelope_nrmse": 0.08439768618765027, "全帯域_10ms_envelope_correlation": 0.9928033063372608, "全帯域_10ms_modulation_mae_db": 0.3200294306875813, "全帯域_40ms_log_rms_mae_db": 0.49720578361096396, "全帯域_40ms_rms_envelope_nrmse": 0.05236895038508897, "全帯域_40ms_envelope_correlation": 0.9972739837103247, "全帯域_40ms_modulation_mae_db": 0.3049349807813612, "300–1000 Hz_10ms_log_rms_mae_db": 1.5424648296358507, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11896088192276107, "300–1000 Hz_10ms_envelope_correlation": 0.9878527794312478, "300–1000 Hz_10ms_modulation_mae_db": 0.28249940115059596, "300–1000 Hz_40ms_log_rms_mae_db": 1.0406162995026018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0753608442397128, "300–1000 Hz_40ms_envelope_correlation": 0.9952817770385077, "300–1000 Hz_40ms_modulation_mae_db": 0.31591630129818077, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7741524265316055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17364258677006256, "1000–3000 Hz_10ms_envelope_correlation": 0.9772351765053269, "1000–3000 Hz_10ms_modulation_mae_db": 0.672700357475032, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2378088354085983, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11415973383576054, "1000–3000 Hz_40ms_envelope_correlation": 0.9897862657637543, "1000–3000 Hz_40ms_modulation_mae_db": 0.516462204150056, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8792683285837295, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18984521063148993, "3000–6000 Hz_10ms_envelope_correlation": 0.9737811648048182, "3000–6000 Hz_10ms_modulation_mae_db": 0.5978285831701579, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4462718314518388, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12112316800884795, "3000–6000 Hz_40ms_envelope_correlation": 0.9893164264487754, "3000–6000 Hz_40ms_modulation_mae_db": 0.4006866674702427, "mr_stft_db": 5.867505469233919, "band_mae_db": 0.22282177137770487, "rms_gain_error_db": -0.1758984350370547, "output_peak": 0.7168436841485855}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.831401749845571, "spectral_convergence": 0.2441740088042335, "spectral_convergence_status": "ok", "amplitude_floor": 1.7851227833840954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.678837421009062, "spectral_convergence": 0.24579782703155553, "spectral_convergence_status": "ok", "amplitude_floor": 1.2048994709482952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.093804631863166, "spectral_convergence": 0.26371629967906607, "spectral_convergence_status": "ok", "amplitude_floor": 9.468117719284587e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005960090393190935, "output_power": 0.0005959652331942916, "delta_db": -0.00031921424350028684, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030564977634421338, "output_power": 0.0030341648195548678, "delta_db": -0.03184913835334149, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008630746056010172, "output_power": 0.0008189859206461384, "delta_db": -0.22771902693857066, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003160541845765363, "output_power": 0.00029760619727962375, "delta_db": -0.2611957417560195, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001143536135409032, "output_power": 0.0001011806825508324, "delta_db": -0.5315228796613143, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.420642909107181e-05, "output_power": 3.513630791227747e-05, "delta_db": 0.1164838504820412, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19484830857246457, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08556196092638113, "pitch": null, "output_rms_dbfs": -23.113097927345223, "output_peak": 0.732358975637002, "reference_rms_dbfs": -23.02753596641884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.868014600905933}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023539858364162597, "log_rms_mae_db": 0.8289138257108376, "rms_envelope_nrmse": 0.07234304411445358, "envelope_correlation": 0.9945707173893922, "rms_gain_error_db": -0.08558361261993448, "reference_mean_power": 0.004983368761309487, "output_mean_power": 0.004886126023967444}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0209735395978665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009552928696150586, "output_power": 0.0009241385313918479, "reference_active": true, "delta_db": -0.14399457882945305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014624552237472608, "output_power": 0.00014958261217747887, "reference_active": true, "delta_db": 0.09798534999501755}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.510798757145359e-05, "output_power": 3.262637095126887e-05, "reference_active": true, "delta_db": 1.1375686466487802}], "active_band_count": 3, "active_band_mae_db": 0.4598495251577503}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002083747898533542, "log_rms_mae_db": 0.4653608451904468, "rms_envelope_nrmse": 0.03996374716539286, "envelope_correlation": 0.9982776986068054, "rms_gain_error_db": -0.08564672694531954, "reference_mean_power": 0.004999205734592124, "output_mean_power": 0.004901582729889482}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8665599913970187e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008900977663526511, "output_power": 0.0008692844429145922, "reference_active": true, "delta_db": -0.10275803826410412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.407197015766335e-05, "output_power": 7.656431883288195e-05, "reference_active": true, "delta_db": 0.14372526982523612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8378799869222006e-06, "output_power": 2.3391891160880795e-06, "reference_active": true, "delta_db": 1.0474818591144603}], "active_band_count": 3, "active_band_mae_db": 0.43132172240126687}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019850808969518194, "log_rms_mae_db": 1.5009618874186526, "rms_envelope_nrmse": 0.10109367830209409, "envelope_correlation": 0.991006387939488, "rms_gain_error_db": -0.15686679204423834, "reference_mean_power": 0.003121339390070273, "output_mean_power": 0.0030106087104380184}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3811041144776226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007780170156378303, "output_power": 0.0007233914399555234, "reference_active": true, "delta_db": -0.3161572998313241}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001204320773242115, "output_power": 0.00011948089092936503, "reference_active": true, "delta_db": -0.03443725197894915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.313510568503667e-05, "output_power": 2.9895616339372457e-05, "reference_active": true, "delta_db": 1.1133602365796598}], "active_band_count": 3, "active_band_mae_db": 0.487984929463311}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016710142967257498, "log_rms_mae_db": 1.0176619654148231, "rms_envelope_nrmse": 0.05698446033684822, "envelope_correlation": 0.997105995262388, "rms_gain_error_db": -0.15727944059370852, "reference_mean_power": 0.003133268891618282, "output_mean_power": 0.003021827873312669}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2683362614683906e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007382464262949856, "output_power": 0.0006932925541527815, "reference_active": true, "delta_db": -0.2728481733546407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8108189032543396e-05, "output_power": 5.88061739915802e-05, "reference_active": true, "delta_db": 0.05185583870942052}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.916170233531451e-06, "output_power": 2.3713395227204287e-06, "reference_active": true, "delta_db": 0.9255965027750973}], "active_band_count": 3, "active_band_mae_db": 0.41676683827971955}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012483338724103395, "log_rms_mae_db": 1.7863823675519024, "rms_envelope_nrmse": 0.17177325806657717, "envelope_correlation": 0.9779367386429185, "rms_gain_error_db": -0.2858747019568601, "reference_mean_power": 0.00037365748369267646, "output_mean_power": 0.00034985349322828483}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.54909855359075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.931499876625785e-05, "output_power": 8.507887844503962e-05, "reference_active": true, "delta_db": -0.21102640396429811}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4977670974468328e-05, "output_power": 1.2102255912161374e-05, "reference_active": true, "delta_db": -0.9257795368071404}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3989606403687574e-06, "output_power": 4.075532750943967e-06, "reference_active": true, "delta_db": 0.7883825194946326}], "active_band_count": 3, "active_band_mae_db": 0.641729486755357}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010596065945345535, "log_rms_mae_db": 1.2133661033137064, "rms_envelope_nrmse": 0.0935775416264806, "envelope_correlation": 0.9937322047079662, "rms_gain_error_db": -0.28545440820648993, "reference_mean_power": 0.00037540546219302315, "output_mean_power": 0.0003515241336293242}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4355698091451528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.577569295489126e-05, "output_power": 8.180052753777729e-05, "reference_active": true, "delta_db": -0.2060813079288791}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.83374089374301e-06, "output_power": 6.585512738820078e-06, "reference_active": true, "delta_db": -0.7537960869112266}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527528900320414e-07, "output_power": 2.414903979932628e-07, "reference_active": true, "delta_db": -0.19796262854430743}], "active_band_count": 3, "active_band_mae_db": 0.385946674461471}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.3292043050207156e-05, "log_rms_mae_db": 1.9074377297390983, "rms_envelope_nrmse": 0.1945171596099572, "envelope_correlation": 0.9733215841310912, "rms_gain_error_db": -0.4617867891710302, "reference_mean_power": 5.712588023845408e-05, "output_mean_power": 5.1363454641834504e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.279319333258023e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4391078150208969e-05, "output_power": 1.1959237703909694e-05, "reference_active": true, "delta_db": -0.8038983352602671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.344791688657969e-06, "output_power": 3.3849937026398474e-06, "reference_active": true, "delta_db": -1.0841109393346036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.199993664659886e-07, "output_power": 6.801840218803261e-07, "reference_active": true, "delta_db": -0.8118709108863884}], "active_band_count": 3, "active_band_mae_db": 0.8999600618270863}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.870517765122232e-05, "log_rms_mae_db": 1.4662310384250246, "rms_envelope_nrmse": 0.12934064736756615, "envelope_correlation": 0.989069728692841, "rms_gain_error_db": -0.4617938412306819, "reference_mean_power": 5.74115094616415e-05, "output_mean_power": 5.1620187934233514e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.057417179151539e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3970684782124775e-05, "output_power": 1.1453731723132242e-05, "reference_active": true, "delta_db": -0.8627068729572086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.241188995835654e-06, "output_power": 1.8349617575087463e-06, "reference_active": true, "delta_db": -0.8685146389060594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.11366674713116e-08, "output_power": 4.3881152068541975e-08, "reference_active": true, "delta_db": -0.6645440090206202}], "active_band_count": 3, "active_band_mae_db": 0.798588506961296}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8417242102011504, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ab9ce921b11afd340542634a79c8bcc6d90968b650910e17f6f19c465444f3d8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9782734501508088, "standard_deviation": 0.001060786351373049, "minimum": 0.9768889679671687, "maximum": 0.9798329096405544, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7631721460646423, "standard_deviation": 0.02543656514136286, "minimum": 1.7158108667700784, "maximum": 1.7863823675519024, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8769969634724781, "standard_deviation": 0.37924854181063566, "minimum": 0.5969486881356495, "maximum": 1.4765183781495492, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17310412406299344, "standard_deviation": 0.003575531708445652, "minimum": 0.16678644149445737, "maximum": 0.1788087400226358, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930575621779492, "standard_deviation": 0.0013865387644870334, "minimum": 0.9897862657637543, "maximum": 0.9940079972589764, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2350236415149858, "standard_deviation": 0.03761589339939349, "minimum": 1.1768820564976843, "maximum": 1.2808261852142222, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7701922908670121, "standard_deviation": 0.34180368265859185, "minimum": 0.3246222509823482, "maximum": 1.2663958936219, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10367454226893941, "standard_deviation": 0.005959911371465129, "minimum": 0.0935775416264806, "maximum": 0.11415973383576054, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9723637330605397, "standard_deviation": 0.002109987979526526, "minimum": 0.9701033889435816, "maximum": 0.9761274222125592, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8961210104510209, "standard_deviation": 0.013979447695224042, "minimum": 1.8751188911340344, "maximum": 1.9142504675823677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6606401288831474, "standard_deviation": 0.1661258537076547, "minimum": 0.4066456362196813, "maximum": 0.8999600618270863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19560258096726768, "standard_deviation": 0.007311748295788431, "minimum": 0.18071694849783074, "maximum": 0.20298835270133275, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9886843000304364, "standard_deviation": 0.0018654567214088681, "minimum": 0.9855670212820058, "maximum": 0.9908329831931931, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.466349271929649, "standard_deviation": 0.02619487447295233, "minimum": 1.417697882968764, "maximum": 1.501365223870333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6178155373916547, "standard_deviation": 0.1890030827546254, "minimum": 0.39642336137574813, "maximum": 0.867866775670285, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12632900024546614, "standard_deviation": 0.008978866797804737, "minimum": 0.11297960659943662, "maximum": 0.1411257471847126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9889413666038358, "standard_deviation": 0.0014824900397937167, "minimum": 0.9862746029998866, "maximum": 0.991006387939488, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.551261612519649, "standard_deviation": 0.048870126577775855, "minimum": 1.4797609469256858, "maximum": 1.6315835428518939, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5048202030931892, "standard_deviation": 0.1185951500014577, "minimum": 0.28249940115059596, "maximum": 0.6450017987604473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11229187469668266, "standard_deviation": 0.0073115712667144745, "minimum": 0.10109367830209409, "maximum": 0.12385029797705557, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961842919926774, "standard_deviation": 0.0007790841970293025, "minimum": 0.9947780592791423, "maximum": 0.997105995262388, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0416315235432112, "standard_deviation": 0.044986958380619474, "minimum": 1.0002445497988128, "maximum": 1.1336715400109767, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5673673654186264, "standard_deviation": 0.21474427364461812, "minimum": 0.31591630129818077, "maximum": 0.9058858396056468, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06597927233572072, "standard_deviation": 0.0060839560577787485, "minimum": 0.05698446033684822, "maximum": 0.0753608442397128, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24497716326564245, "standard_deviation": 0.03126106984660078, "minimum": 0.19484830857246457, "maximum": 0.30285482720083107, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.869645784249384, "standard_deviation": 0.00766725848016321, "minimum": 5.8590329547751665, "maximum": 5.882273124607718, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.710592348098237, "standard_deviation": 0.031436050405402295, "minimum": 0.6669469868144029, "maximum": 0.7705850813186453, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13252331511578372, "standard_deviation": 0.0411564874735619, "minimum": -0.1758984350370547, "maximum": -0.06573966549089634, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933101157875542, "standard_deviation": 0.000812776843348587, "minimum": 0.991905873633453, "maximum": 0.9945707173893922, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.873175491910291, "standard_deviation": 0.02414245665131169, "minimum": 0.8289138257108376, "maximum": 0.905152759787651, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42806066389904324, "standard_deviation": 0.1299917022156823, "minimum": 0.2658540820512771, "maximum": 0.6866238230835319, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08066921706255295, "standard_deviation": 0.00478867112484444, "minimum": 0.07234304411445358, "maximum": 0.08784686991568964, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978265070892677, "standard_deviation": 0.0005364549106760823, "minimum": 0.9968702670648474, "maximum": 0.998322046906428, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48490770163997204, "standard_deviation": 0.03364805925899179, "minimum": 0.4362927044003156, "maximum": 0.5303426608382106, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4185709448600474, "standard_deviation": 0.188072802636006, "minimum": 0.2009660141810258, "maximum": 0.736092718020795, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04585511857851152, "standard_deviation": 0.005121373123170417, "minimum": 0.03996374716539286, "maximum": 0.05270748466161683, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0028", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.8368942737579346, "predicted_mos": 2.8368942737579346, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "85a6c9d1df24f694003823972a25196c1749b613de7807114289087a076e0321", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.868520500138402, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8535218963506587, "全帯域_10ms_rms_envelope_nrmse": 0.07468418944974632, "全帯域_10ms_envelope_correlation": 0.9941974007733908, "全帯域_10ms_modulation_mae_db": 0.4751627146997766, "全帯域_40ms_log_rms_mae_db": 0.48285122815969644, "全帯域_40ms_rms_envelope_nrmse": 0.041933209720120025, "全帯域_40ms_envelope_correlation": 0.9980785361564642, "全帯域_40ms_modulation_mae_db": 0.4102965192696559, "300–1000 Hz_10ms_log_rms_mae_db": 1.4964655374493636, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10397494515758399, "300–1000 Hz_10ms_envelope_correlation": 0.990435680756421, "300–1000 Hz_10ms_modulation_mae_db": 0.46748246523384424, "300–1000 Hz_40ms_log_rms_mae_db": 1.0101164068035722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05860890758819586, "300–1000 Hz_40ms_envelope_correlation": 0.9968757220329545, "300–1000 Hz_40ms_modulation_mae_db": 0.32928970680327885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7528944174952648, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1697201796329029, "1000–3000 Hz_10ms_envelope_correlation": 0.9787021414206398, "1000–3000 Hz_10ms_modulation_mae_db": 0.6589040034959277, "1000–3000 Hz_40ms_log_rms_mae_db": 1.194683434971599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0935239544228181, "1000–3000 Hz_40ms_envelope_correlation": 0.9940080326125741, "1000–3000 Hz_40ms_modulation_mae_db": 0.27596121564920084, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8230635771394637, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18777263893719565, "3000–6000 Hz_10ms_envelope_correlation": 0.9754991367311776, "3000–6000 Hz_10ms_modulation_mae_db": 0.6614742212698103, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4381770310271231, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12703403393930165, "3000–6000 Hz_40ms_envelope_correlation": 0.9898206643142489, "3000–6000 Hz_40ms_modulation_mae_db": 0.5988000531523466, "mr_stft_db": 5.792216069421297, "band_mae_db": 0.20427140213368555, "rms_gain_error_db": -0.0802477093752226, "output_peak": 0.7421917753509304}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8954679261399457, "全帯域_10ms_rms_envelope_nrmse": 0.08817227176724948, "全帯域_10ms_envelope_correlation": 0.9918379899821794, "全帯域_10ms_modulation_mae_db": 0.6363704819806503, "全帯域_40ms_log_rms_mae_db": 0.5153644739354444, "全帯域_40ms_rms_envelope_nrmse": 0.0509722674908903, "全帯域_40ms_envelope_correlation": 0.9970693610663546, "全帯域_40ms_modulation_mae_db": 0.6220685959971556, "300–1000 Hz_10ms_log_rms_mae_db": 1.6055382194926235, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12477425730713088, "300–1000 Hz_10ms_envelope_correlation": 0.9860593070659746, "300–1000 Hz_10ms_modulation_mae_db": 0.5502373554608547, "300–1000 Hz_40ms_log_rms_mae_db": 1.0855905071961012, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07280771989543824, "300–1000 Hz_40ms_envelope_correlation": 0.9948850862882994, "300–1000 Hz_40ms_modulation_mae_db": 0.6311672660815787, "1000–3000 Hz_10ms_log_rms_mae_db": 1.726197536235867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17049457962039374, "1000–3000 Hz_10ms_envelope_correlation": 0.978563203143631, "1000–3000 Hz_10ms_modulation_mae_db": 0.7833005145493601, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2348236512887827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09709461598125158, "1000–3000 Hz_40ms_envelope_correlation": 0.9935457760830763, "1000–3000 Hz_40ms_modulation_mae_db": 0.6711187824782018, "3000–6000 Hz_10ms_log_rms_mae_db": 1.830558053565912, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.193709696666887, "3000–6000 Hz_10ms_envelope_correlation": 0.9731113531215633, "3000–6000 Hz_10ms_modulation_mae_db": 0.7629223380548863, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4509881332845014, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1371300357707988, "3000–6000 Hz_40ms_envelope_correlation": 0.9866272927805222, "3000–6000 Hz_40ms_modulation_mae_db": 0.7921311456549596, "mr_stft_db": 5.804814294873743, "band_mae_db": 0.2507153413349395, "rms_gain_error_db": -0.06112623041999634, "output_peak": 0.666949239650658}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9031287730018374, "全帯域_10ms_rms_envelope_nrmse": 0.08431592213169643, "全帯域_10ms_envelope_correlation": 0.9927111939685199, "全帯域_10ms_modulation_mae_db": 0.35857420501460496, "全帯域_40ms_log_rms_mae_db": 0.5458925014792118, "全帯域_40ms_rms_envelope_nrmse": 0.047886375244959534, "全帯域_40ms_envelope_correlation": 0.9976889160602733, "全帯域_40ms_modulation_mae_db": 0.27946882346036933, "300–1000 Hz_10ms_log_rms_mae_db": 1.493573186804838, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11616806394782102, "300–1000 Hz_10ms_envelope_correlation": 0.9881504933728409, "300–1000 Hz_10ms_modulation_mae_db": 0.46208815669576264, "300–1000 Hz_40ms_log_rms_mae_db": 1.0168612226746199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06555302164454097, "300–1000 Hz_40ms_envelope_correlation": 0.9962729609762812, "300–1000 Hz_40ms_modulation_mae_db": 0.268487933630372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7595424139368934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17684679269316167, "1000–3000 Hz_10ms_envelope_correlation": 0.9787346026987818, "1000–3000 Hz_10ms_modulation_mae_db": 1.4082981212717691, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2349669269793906, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10674893582201879, "1000–3000 Hz_40ms_envelope_correlation": 0.9944562105429338, "1000–3000 Hz_40ms_modulation_mae_db": 1.0001220377266042, "3000–6000 Hz_10ms_log_rms_mae_db": 1.825209973229804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20143650626868273, "3000–6000 Hz_10ms_envelope_correlation": 0.9703425751473186, "3000–6000 Hz_10ms_modulation_mae_db": 0.43763766119592534, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4738538102331473, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1345213582964332, "3000–6000 Hz_40ms_envelope_correlation": 0.9863726854897895, "3000–6000 Hz_40ms_modulation_mae_db": 0.645458920265044, "mr_stft_db": 5.806158414409343, "band_mae_db": 0.3010374197483166, "rms_gain_error_db": -0.14944540412506224, "output_peak": 0.7655911618854836}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9151289736438284, "全帯域_10ms_rms_envelope_nrmse": 0.08419048976271798, "全帯域_10ms_envelope_correlation": 0.9928507188564333, "全帯域_10ms_modulation_mae_db": 0.5221028469864891, "全帯域_40ms_log_rms_mae_db": 0.5176497317443487, "全帯域_40ms_rms_envelope_nrmse": 0.050674928814339826, "全帯域_40ms_envelope_correlation": 0.9974902411813027, "全帯域_40ms_modulation_mae_db": 0.4141796782300358, "300–1000 Hz_10ms_log_rms_mae_db": 1.6207469767341611, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10976017603077062, "300–1000 Hz_10ms_envelope_correlation": 0.9895190540005804, "300–1000 Hz_10ms_modulation_mae_db": 0.5135530428755369, "300–1000 Hz_40ms_log_rms_mae_db": 1.1147693255816327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0661361657905172, "300–1000 Hz_40ms_envelope_correlation": 0.9962362907438853, "300–1000 Hz_40ms_modulation_mae_db": 0.3493469060208943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6753811700320933, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16641922358945083, "1000–3000 Hz_10ms_envelope_correlation": 0.9815544939712223, "1000–3000 Hz_10ms_modulation_mae_db": 0.9610176324412135, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1877290449244793, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1046139507286144, "1000–3000 Hz_40ms_envelope_correlation": 0.9947859066369433, "1000–3000 Hz_40ms_modulation_mae_db": 1.3123687671930535, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7916952801928585, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18969148287163112, "3000–6000 Hz_10ms_envelope_correlation": 0.9746053711720337, "3000–6000 Hz_10ms_modulation_mae_db": 0.5954071118594403, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4517215129499474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11775991696039535, "3000–6000 Hz_40ms_envelope_correlation": 0.9911643284720326, "3000–6000 Hz_40ms_modulation_mae_db": 0.7690843821059556, "mr_stft_db": 5.792853201476899, "band_mae_db": 0.28055067411967144, "rms_gain_error_db": -0.1781621167126191, "output_peak": 0.6985782158043242}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8545351069320717, "全帯域_10ms_rms_envelope_nrmse": 0.08108349293029334, "全帯域_10ms_envelope_correlation": 0.9931926065452155, "全帯域_10ms_modulation_mae_db": 0.2950794251383509, "全帯域_40ms_log_rms_mae_db": 0.4509564329192311, "全帯域_40ms_rms_envelope_nrmse": 0.04223460171346055, "全帯域_40ms_envelope_correlation": 0.9981430780688973, "全帯域_40ms_modulation_mae_db": 0.3427079616722315, "300–1000 Hz_10ms_log_rms_mae_db": 1.5860479775407332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11574831866103147, "300–1000 Hz_10ms_envelope_correlation": 0.988219175164901, "300–1000 Hz_10ms_modulation_mae_db": 0.6192636700121894, "300–1000 Hz_40ms_log_rms_mae_db": 1.0407870188646877, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06327123806514849, "300–1000 Hz_40ms_envelope_correlation": 0.996519835125107, "300–1000 Hz_40ms_modulation_mae_db": 0.8167497631823096, "1000–3000 Hz_10ms_log_rms_mae_db": 1.675346659808849, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17575034307412696, "1000–3000 Hz_10ms_envelope_correlation": 0.9785085375176669, "1000–3000 Hz_10ms_modulation_mae_db": 0.7439751764576433, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1440904563256318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10437211498362246, "1000–3000 Hz_40ms_envelope_correlation": 0.994355091208875, "1000–3000 Hz_40ms_modulation_mae_db": 0.5131609525638938, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7974197847197324, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1726703859361716, "3000–6000 Hz_10ms_envelope_correlation": 0.9783612568748838, "3000–6000 Hz_10ms_modulation_mae_db": 0.35555753752051505, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4343421019825395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.112627499462723, "3000–6000 Hz_40ms_envelope_correlation": 0.9906513979444238, "3000–6000 Hz_40ms_modulation_mae_db": 0.3351108121036142, "mr_stft_db": 5.785926207768952, "band_mae_db": 0.2461827092798763, "rms_gain_error_db": -0.11740393775435967, "output_peak": 0.6849347597068002}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8525918531938139, "全帯域_10ms_rms_envelope_nrmse": 0.07735750755161008, "全帯域_10ms_envelope_correlation": 0.9940086629557414, "全帯域_10ms_modulation_mae_db": 0.5076392241869203, "全帯域_40ms_log_rms_mae_db": 0.4491143435618551, "全帯域_40ms_rms_envelope_nrmse": 0.04328594533087476, "全帯域_40ms_envelope_correlation": 0.9982629810308937, "全帯域_40ms_modulation_mae_db": 0.40180113593042366, "300–1000 Hz_10ms_log_rms_mae_db": 1.5425002424518612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1117282025614316, "300–1000 Hz_10ms_envelope_correlation": 0.9893474291824745, "300–1000 Hz_10ms_modulation_mae_db": 0.5646048090107582, "300–1000 Hz_40ms_log_rms_mae_db": 0.9913830950593391, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06619748829130559, "300–1000 Hz_40ms_envelope_correlation": 0.9965164712057657, "300–1000 Hz_40ms_modulation_mae_db": 0.4884568796268627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6778747386980053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16866719122984503, "1000–3000 Hz_10ms_envelope_correlation": 0.9793734011542563, "1000–3000 Hz_10ms_modulation_mae_db": 1.4004710005062548, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1699733767086058, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09874895836618527, "1000–3000 Hz_40ms_envelope_correlation": 0.9936975983007963, "1000–3000 Hz_40ms_modulation_mae_db": 1.0127324215047258, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7793435350586977, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18924759067779762, "3000–6000 Hz_10ms_envelope_correlation": 0.975718699739106, "3000–6000 Hz_10ms_modulation_mae_db": 0.5017573989657809, "3000–6000 Hz_40ms_log_rms_mae_db": 1.435459633623283, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12911679320129515, "3000–6000 Hz_40ms_envelope_correlation": 0.990186994798442, "3000–6000 Hz_40ms_modulation_mae_db": 0.3921990503917327, "mr_stft_db": 5.783488654715074, "band_mae_db": 0.266044353828312, "rms_gain_error_db": -0.17142476648799904, "output_peak": 0.7024929874316242}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8953854701023757, "全帯域_10ms_rms_envelope_nrmse": 0.08014604088602205, "全帯域_10ms_envelope_correlation": 0.9933629791043532, "全帯域_10ms_modulation_mae_db": 0.2963567449870967, "全帯域_40ms_log_rms_mae_db": 0.4765293871192322, "全帯域_40ms_rms_envelope_nrmse": 0.04368394241864112, "全帯域_40ms_envelope_correlation": 0.9980045343635918, "全帯域_40ms_modulation_mae_db": 0.4367808974251904, "300–1000 Hz_10ms_log_rms_mae_db": 1.511867886094711, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10879445939196906, "300–1000 Hz_10ms_envelope_correlation": 0.9895151873000269, "300–1000 Hz_10ms_modulation_mae_db": 0.401871452694823, "300–1000 Hz_40ms_log_rms_mae_db": 1.0212213243159514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0637292438291203, "300–1000 Hz_40ms_envelope_correlation": 0.9962622812915828, "300–1000 Hz_40ms_modulation_mae_db": 0.6609164678070988, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6827665174282835, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1617657021359559, "1000–3000 Hz_10ms_envelope_correlation": 0.981347689519322, "1000–3000 Hz_10ms_modulation_mae_db": 0.5723196981424583, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2033782322602415, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09966053447158642, "1000–3000 Hz_40ms_envelope_correlation": 0.9938677128193297, "1000–3000 Hz_40ms_modulation_mae_db": 0.42983439321147837, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7829741304224165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18833352760160446, "3000–6000 Hz_10ms_envelope_correlation": 0.9741523014931134, "3000–6000 Hz_10ms_modulation_mae_db": 0.70796083986724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4269259822030007, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1295832213585691, "3000–6000 Hz_40ms_envelope_correlation": 0.9874400846480659, "3000–6000 Hz_40ms_modulation_mae_db": 0.7649948547742852, "mr_stft_db": 5.793546149863386, "band_mae_db": 0.24030752886053589, "rms_gain_error_db": -0.11853600393102967, "output_peak": 0.7125817432048939}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9016287891973283, "全帯域_10ms_rms_envelope_nrmse": 0.08650566388523405, "全帯域_10ms_envelope_correlation": 0.9923907575046169, "全帯域_10ms_modulation_mae_db": 0.3409411100575573, "全帯域_40ms_log_rms_mae_db": 0.5151995272366924, "全帯域_40ms_rms_envelope_nrmse": 0.05315384566100627, "全帯域_40ms_envelope_correlation": 0.9971470573325618, "全帯域_40ms_modulation_mae_db": 0.34177399552805926, "300–1000 Hz_10ms_log_rms_mae_db": 1.5606927547559777, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12099993469784749, "300–1000 Hz_10ms_envelope_correlation": 0.9873887880998881, "300–1000 Hz_10ms_modulation_mae_db": 0.332339457407343, "300–1000 Hz_40ms_log_rms_mae_db": 1.0496262082769372, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07569963940490913, "300–1000 Hz_40ms_envelope_correlation": 0.9952191708288786, "300–1000 Hz_40ms_modulation_mae_db": 0.36975582971218107, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7416863900844655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1736372366810108, "1000–3000 Hz_10ms_envelope_correlation": 0.9773650017508547, "1000–3000 Hz_10ms_modulation_mae_db": 0.6438587982894327, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2130802272052061, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11200577331274145, "1000–3000 Hz_40ms_envelope_correlation": 0.9903592656966355, "1000–3000 Hz_40ms_modulation_mae_db": 0.49736562831187753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8275120370528548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1857133423442104, "3000–6000 Hz_10ms_envelope_correlation": 0.9751322925886058, "3000–6000 Hz_10ms_modulation_mae_db": 0.6221426083303639, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4623098395828182, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12225705996322672, "3000–6000 Hz_40ms_envelope_correlation": 0.989300586012646, "3000–6000 Hz_40ms_modulation_mae_db": 0.47133908414798825, "mr_stft_db": 5.7974354771400955, "band_mae_db": 0.2245518229101869, "rms_gain_error_db": -0.170118499039162, "output_peak": 0.7057018958329827}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.750881586155913, "spectral_convergence": 0.24643175036431333, "spectral_convergence_status": "ok", "amplitude_floor": 1.7851227833840954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5976347212973385, "spectral_convergence": 0.24617790434089035, "spectral_convergence_status": "ok", "amplitude_floor": 1.2048994709482952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.028131900810639, "spectral_convergence": 0.2639177382817331, "spectral_convergence_status": "ok", "amplitude_floor": 9.468117719284587e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005960090393190935, "output_power": 0.0005940573727055427, "delta_db": -0.014244563027471862, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030564977634421338, "output_power": 0.0030535788134336823, "delta_db": -0.004149486384042831, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008630746056010172, "output_power": 0.0008135383616735953, "delta_db": -0.25670301950042285, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003160541845765363, "output_power": 0.00029351723947764055, "delta_db": -0.3212793054881322, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001143536135409032, "output_power": 9.98138674956044e-05, "delta_db": -0.5905900918260824, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.420642909107181e-05, "output_power": 3.451230246653042e-05, "delta_db": 0.038661946575961205, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20427140213368555, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0802477093752226, "pitch": null, "output_rms_dbfs": -23.107783675794064, "output_peak": 0.7421917753509304, "reference_rms_dbfs": -23.02753596641884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.792216069421297}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023539858364162597, "log_rms_mae_db": 0.8535218963506587, "rms_envelope_nrmse": 0.07468418944974632, "envelope_correlation": 0.9941974007733908, "rms_gain_error_db": -0.08011911169324712, "reference_mean_power": 0.004983368761309487, "output_mean_power": 0.004892277850920436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0209735395978665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009552928696150586, "output_power": 0.0009267975831533185, "reference_active": true, "delta_db": -0.13151643555409048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014624552237472608, "output_power": 0.00015051000047608415, "reference_active": true, "delta_db": 0.12482779035083114}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.510798757145359e-05, "output_power": 3.2864444569632575e-05, "reference_active": true, "delta_db": 1.1691439181944083}], "active_band_count": 3, "active_band_mae_db": 0.4751627146997766}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002083747898533542, "log_rms_mae_db": 0.48285122815969644, "rms_envelope_nrmse": 0.041933209720120025, "envelope_correlation": 0.9980785361564642, "rms_gain_error_db": -0.08007351528325923, "reference_mean_power": 0.004999205734592124, "output_mean_power": 0.00490787686778482}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8665599913970187e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008900977663526511, "output_power": 0.0008723214465071445, "reference_active": true, "delta_db": -0.0876116122364904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.407197015766335e-05, "output_power": 7.713930371460004e-05, "reference_active": true, "delta_db": 0.17621818339659157}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8378799869222006e-06, "output_power": 2.2962709143961985e-06, "reference_active": true, "delta_db": 0.9670597621758856}], "active_band_count": 3, "active_band_mae_db": 0.4102965192696559}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019850808969518194, "log_rms_mae_db": 1.4964655374493636, "rms_envelope_nrmse": 0.10397494515758399, "envelope_correlation": 0.990435680756421, "rms_gain_error_db": -0.14479184023176472, "reference_mean_power": 0.003121339390070273, "output_mean_power": 0.003018990935091999}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3811041144776226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007780170156378303, "output_power": 0.0007272673867211956, "reference_active": true, "delta_db": -0.2929498265491314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001204320773242115, "output_power": 0.00011843174691000302, "reference_active": true, "delta_db": -0.07274042208193747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.313510568503667e-05, "output_power": 2.9372925412846565e-05, "reference_active": true, "delta_db": 1.0367571470704637}], "active_band_count": 3, "active_band_mae_db": 0.46748246523384424}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016710142967257498, "log_rms_mae_db": 1.0101164068035722, "rms_envelope_nrmse": 0.05860890758819586, "envelope_correlation": 0.9968757220329545, "rms_gain_error_db": -0.14501193340801932, "reference_mean_power": 0.003133268891618282, "output_mean_power": 0.0030303756910310723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2683362614683906e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007382464262949856, "output_power": 0.0006970074422480848, "reference_active": true, "delta_db": -0.2496393803682792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8108189032543396e-05, "output_power": 5.8219358794794635e-05, "reference_active": true, "delta_db": 0.008300773101461968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.916170233531451e-06, "output_power": 2.266871936926659e-06, "reference_active": true, "delta_db": 0.7299289669400953}], "active_band_count": 3, "active_band_mae_db": 0.32928970680327885}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012483338724103395, "log_rms_mae_db": 1.7528944174952648, "rms_envelope_nrmse": 0.1697201796329029, "envelope_correlation": 0.9787021414206398, "rms_gain_error_db": -0.33908123168341986, "reference_mean_power": 0.00037365748369267646, "output_mean_power": 0.0003455934968796582}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.54909855359075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.931499876625785e-05, "output_power": 8.368039377945529e-05, "reference_active": true, "delta_db": -0.2830068135278467}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4977670974468328e-05, "output_power": 1.2828259465969479e-05, "reference_active": true, "delta_db": -0.6727655044908666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3989606403687574e-06, "output_power": 4.2997193045255884e-06, "reference_active": true, "delta_db": 1.02093969246907}], "active_band_count": 3, "active_band_mae_db": 0.6589040034959277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010596065945345535, "log_rms_mae_db": 1.194683434971599, "rms_envelope_nrmse": 0.0935239544228181, "envelope_correlation": 0.9940080326125741, "rms_gain_error_db": -0.33865834565690833, "reference_mean_power": 0.00037540546219302315, "output_mean_power": 0.00034724400197027564}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4355698091451528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.577569295489126e-05, "output_power": 8.072817305437556e-05, "reference_active": true, "delta_db": -0.26339111121446424}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.83374089374301e-06, "output_power": 6.9731474349426715e-06, "reference_active": true, "delta_db": -0.5054035517160711}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527528900320414e-07, "output_power": 2.493372881662043e-07, "reference_active": true, "delta_db": -0.059088984017067234}], "active_band_count": 3, "active_band_mae_db": 0.27596121564920084}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.3292043050207156e-05, "log_rms_mae_db": 1.8230635771394637, "rms_envelope_nrmse": 0.18777263893719565, "envelope_correlation": 0.9754991367311776, "rms_gain_error_db": -0.5234273703393817, "reference_mean_power": 5.712588023845408e-05, "output_mean_power": 5.0639588512533825e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.279319333258023e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4391078150208969e-05, "output_power": 1.2508532746704185e-05, "reference_active": true, "delta_db": -0.6088696175705075}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.344791688657969e-06, "output_power": 3.4236423876443204e-06, "reference_active": true, "delta_db": -1.0348056432637511}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.199993664659886e-07, "output_power": 7.581214359873388e-07, "reference_active": true, "delta_db": -0.3407474029751722}], "active_band_count": 3, "active_band_mae_db": 0.6614742212698103}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.870517765122232e-05, "log_rms_mae_db": 1.4381770310271231, "rms_envelope_nrmse": 0.12703403393930165, "envelope_correlation": 0.9898206643142489, "rms_gain_error_db": -0.5234257290890832, "reference_mean_power": 5.74115094616415e-05, "output_mean_power": 5.089280553029351e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.057417179151539e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3970684782124775e-05, "output_power": 1.1936415201428915e-05, "reference_active": true, "delta_db": -0.683437767815934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.241188995835654e-06, "output_power": 1.8112062186031663e-06, "reference_active": true, "delta_db": -0.9251058080986009}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.11366674713116e-08, "output_power": 4.897187854094767e-08, "reference_active": true, "delta_db": -0.18785658354250495}], "active_band_count": 3, "active_band_mae_db": 0.5988000531523466}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.842757526311582, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "154ad695294f8e48772e2efdc64daaf9d5123ee7c00d4293e61d34a62c947e74", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9792686338970469, "standard_deviation": 0.0014571940838366817, "minimum": 0.9773650017508547, "maximum": 0.9815544939712223, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7114612304649652, "standard_deviation": 0.03725721432216022, "minimum": 1.675346659808849, "maximum": 1.7595424139368934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8965181181442574, "standard_deviation": 0.3341378792248486, "minimum": 0.5723196981424583, "maximum": 1.4082981212717691, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17041265608210598, "standard_deviation": 0.00499338325991635, "minimum": 0.1617657021359559, "maximum": 0.17684679269316167, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9936344492376454, "standard_deviation": 0.0013863170649679088, "minimum": 0.9903592656966355, "maximum": 0.9947859066369433, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1978406688329921, "standard_deviation": 0.031137696864146327, "minimum": 1.1440904563256318, "maximum": 1.2349669269793906, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7140830248298795, "standard_deviation": 0.3567483402710363, "minimum": 0.27596121564920084, "maximum": 1.3123687671930535, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10209610476110481, "standard_deviation": 0.005940204200376634, "minimum": 0.0935239544228181, "maximum": 0.11200577331274145, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9746153733584753, "standard_deviation": 0.002300019646892297, "minimum": 0.9703425751473186, "maximum": 0.9783612568748838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8072220464227176, "standard_deviation": 0.02149122806338481, "minimum": 1.7793435350586977, "maximum": 1.830558053565912, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5806074646329953, "standard_deviation": 0.1389996878450923, "minimum": 0.35555753752051505, "maximum": 0.7629223380548863, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18857189641302258, "standard_deviation": 0.008065182787560663, "minimum": 0.1726703859361716, "maximum": 0.20143650626868273, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9889455043075214, "standard_deviation": 0.0018718337140370878, "minimum": 0.9863726854897895, "maximum": 0.9911643284720326, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.446722255610795, "standard_deviation": 0.01588940562807715, "minimum": 1.4269259822030007, "maximum": 1.4738538102331473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5961397878244908, "standard_deviation": 0.17911572424193553, "minimum": 0.3351108121036142, "maximum": 0.7921311456549596, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12625373986909286, "standard_deviation": 0.008283646649494934, "minimum": 0.112627499462723, "maximum": 0.1371300357707988, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9885793893678885, "standard_deviation": 0.0014079591360771338, "minimum": 0.9860593070659746, "maximum": 0.990435680756421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5521790976655336, "standard_deviation": 0.04931771762114065, "minimum": 1.493573186804838, "maximum": 1.6207469767341611, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.488930051173889, "standard_deviation": 0.09285833935898528, "minimum": 0.332339457407343, "maximum": 0.6192636700121894, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11399354471944825, "standard_deviation": 0.006798918410398758, "minimum": 0.10397494515758399, "maximum": 0.12477425730713088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960984773115943, "standard_deviation": 0.0006843283631870677, "minimum": 0.9948850862882994, "maximum": 0.9968757220329545, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0412943885966053, "standard_deviation": 0.04121979484422705, "minimum": 0.9913830950593391, "maximum": 1.1147693255816327, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.489271344108072, "standard_deviation": 0.1945843646069131, "minimum": 0.268487933630372, "maximum": 0.8167497631823096, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06650042806364698, "standard_deviation": 0.005423660907902895, "minimum": 0.05860890758819586, "maximum": 0.07569963940490913, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2517076565269405, "standard_deviation": 0.03075245928715903, "minimum": 0.20427140213368555, "maximum": 0.3010374197483166, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.794554808708599, "standard_deviation": 0.008057773991781255, "minimum": 5.783488654715074, "maximum": 5.806158414409343, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7098777223584622, "standard_deviation": 0.031213902105090167, "minimum": 0.666949239650658, "maximum": 0.7655911618854836, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13080808348068135, "standard_deviation": 0.04395934002626028, "minimum": -0.1781621167126191, "maximum": -0.06112623041999634, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930690387113064, "standard_deviation": 0.0007936378641535962, "minimum": 0.9918379899821794, "maximum": 0.9941974007733908, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8839235985702325, "standard_deviation": 0.025885349244523792, "minimum": 0.8525918531938139, "maximum": 0.9151289736438284, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.42902834413143076, "standard_deviation": 0.12434335438893444, "minimum": 0.2950794251383509, "maximum": 0.6363704819806503, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08205694729557121, "standard_deviation": 0.004596345012632206, "minimum": 0.07468418944974632, "maximum": 0.08817227176724948, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977355881575425, "standard_deviation": 0.0004608590045973291, "minimum": 0.9970693610663546, "maximum": 0.9982629810308937, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.494194703269464, "standard_deviation": 0.03475359631326704, "minimum": 0.4491143435618551, "maximum": 0.5458925014792118, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40613470093914017, "standard_deviation": 0.10133181332604496, "minimum": 0.27946882346036933, "maximum": 0.6220685959971556, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04672813954928655, "standard_deviation": 0.0044804344177847765, "minimum": 0.041933209720120025, "maximum": 0.05315384566100627, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0028", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.7170963287353516, "predicted_mos": 2.7170963287353516, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d2b543ebe86e70ca7474e253969b7e28792acbb056cb75b87e3adddd4575178c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.775569799821824, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8802663807172079, "全帯域_10ms_rms_envelope_nrmse": 0.0875510952017504, "全帯域_10ms_envelope_correlation": 0.9923011111816834, "全帯域_10ms_modulation_mae_db": 0.49363829954187294, "全帯域_40ms_log_rms_mae_db": 0.47220734986349494, "全帯域_40ms_rms_envelope_nrmse": 0.05100930172474126, "全帯域_40ms_envelope_correlation": 0.9975461937799094, "全帯域_40ms_modulation_mae_db": 0.5003824636272668, "300–1000 Hz_10ms_log_rms_mae_db": 1.805724169616414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1276876468440612, "300–1000 Hz_10ms_envelope_correlation": 0.9859212336725739, "300–1000 Hz_10ms_modulation_mae_db": 0.8214243198265171, "300–1000 Hz_40ms_log_rms_mae_db": 1.1827954071682072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08363903931104577, "300–1000 Hz_40ms_envelope_correlation": 0.9940427185569701, "300–1000 Hz_40ms_modulation_mae_db": 0.7043708837910033, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0426660193514734, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19117948313607486, "1000–3000 Hz_10ms_envelope_correlation": 0.9742891334861361, "1000–3000 Hz_10ms_modulation_mae_db": 1.0140448158218476, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3835838875170945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11684686161170309, "1000–3000 Hz_40ms_envelope_correlation": 0.9924774614232444, "1000–3000 Hz_40ms_modulation_mae_db": 1.2374492062425662, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3508844200973518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22728134243868375, "3000–6000 Hz_10ms_envelope_correlation": 0.9620369892453413, "3000–6000 Hz_10ms_modulation_mae_db": 0.7928098538019507, "3000–6000 Hz_40ms_log_rms_mae_db": 1.821993025825247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13996967260859827, "3000–6000 Hz_40ms_envelope_correlation": 0.9855554696040555, "3000–6000 Hz_40ms_modulation_mae_db": 0.4954466034633676, "mr_stft_db": 6.1664561240360065, "band_mae_db": 0.3447831613097731, "rms_gain_error_db": -0.19605919454830212, "output_peak": 0.6876760884185386}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8525730707880487, "全帯域_10ms_rms_envelope_nrmse": 0.08230578332508394, "全帯域_10ms_envelope_correlation": 0.9930182143735153, "全帯域_10ms_modulation_mae_db": 0.40489566643302827, "全帯域_40ms_log_rms_mae_db": 0.4530684525079295, "全帯域_40ms_rms_envelope_nrmse": 0.05347033452079015, "全帯域_40ms_envelope_correlation": 0.996943773086847, "全帯域_40ms_modulation_mae_db": 0.5132894106064154, "300–1000 Hz_10ms_log_rms_mae_db": 1.6550991852158536, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11240721216412243, "300–1000 Hz_10ms_envelope_correlation": 0.9889955017179262, "300–1000 Hz_10ms_modulation_mae_db": 0.8108821519894418, "300–1000 Hz_40ms_log_rms_mae_db": 1.0604236131456806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07603333943975704, "300–1000 Hz_40ms_envelope_correlation": 0.9948576683598028, "300–1000 Hz_40ms_modulation_mae_db": 1.0890919258177163, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0746849944297825, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18022352287315593, "1000–3000 Hz_10ms_envelope_correlation": 0.9760206694033622, "1000–3000 Hz_10ms_modulation_mae_db": 1.0282102920887954, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4117763338408755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10126392883101414, "1000–3000 Hz_40ms_envelope_correlation": 0.9931572212272018, "1000–3000 Hz_40ms_modulation_mae_db": 1.472000617975507, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3349102537649764, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22700856381795217, "3000–6000 Hz_10ms_envelope_correlation": 0.962285794591843, "3000–6000 Hz_10ms_modulation_mae_db": 0.8637833545394317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.807318621276965, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14625590863961685, "3000–6000 Hz_40ms_envelope_correlation": 0.9835938376431508, "3000–6000 Hz_40ms_modulation_mae_db": 0.5710449408453501, "mr_stft_db": 6.143946651522821, "band_mae_db": 0.25257760209821994, "rms_gain_error_db": -0.12997875584418392, "output_peak": 0.7187835801710012}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8506270932776633, "全帯域_10ms_rms_envelope_nrmse": 0.08013063431073751, "全帯域_10ms_envelope_correlation": 0.9933911641101312, "全帯域_10ms_modulation_mae_db": 0.2942182832057078, "全帯域_40ms_log_rms_mae_db": 0.46020543328875896, "全帯域_40ms_rms_envelope_nrmse": 0.045009959634207175, "全帯域_40ms_envelope_correlation": 0.9979143969626179, "全帯域_40ms_modulation_mae_db": 0.5036931962213576, "300–1000 Hz_10ms_log_rms_mae_db": 1.657639946790977, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11045636457530589, "300–1000 Hz_10ms_envelope_correlation": 0.9892558586768895, "300–1000 Hz_10ms_modulation_mae_db": 0.5046638437317765, "300–1000 Hz_40ms_log_rms_mae_db": 1.0354250765097184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06381768087771676, "300–1000 Hz_40ms_envelope_correlation": 0.9963789966356734, "300–1000 Hz_40ms_modulation_mae_db": 0.9688824132749199, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0010580350729947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1894069612240167, "1000–3000 Hz_10ms_envelope_correlation": 0.9743505918622694, "1000–3000 Hz_10ms_modulation_mae_db": 0.6065505913334319, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3983862265421052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12194591530648753, "1000–3000 Hz_40ms_envelope_correlation": 0.9907946378555651, "1000–3000 Hz_40ms_modulation_mae_db": 0.20787143408443465, "3000–6000 Hz_10ms_log_rms_mae_db": 2.275008955285363, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23352764494398717, "3000–6000 Hz_10ms_envelope_correlation": 0.9601123372237625, "3000–6000 Hz_10ms_modulation_mae_db": 0.7123448158437536, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7479320610048006, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15161030678969584, "3000–6000 Hz_40ms_envelope_correlation": 0.9833436285495782, "3000–6000 Hz_40ms_modulation_mae_db": 0.6455106539067249, "mr_stft_db": 6.155711033898392, "band_mae_db": 0.289048425348527, "rms_gain_error_db": -0.13131080968604264, "output_peak": 0.6951173686225478}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9029409367309431, "全帯域_10ms_rms_envelope_nrmse": 0.08435413168325626, "全帯域_10ms_envelope_correlation": 0.9928288711909775, "全帯域_10ms_modulation_mae_db": 0.6872491560459016, "全帯域_40ms_log_rms_mae_db": 0.5024353874957248, "全帯域_40ms_rms_envelope_nrmse": 0.043722024790446894, "全帯域_40ms_envelope_correlation": 0.9982756390328638, "全帯域_40ms_modulation_mae_db": 0.24785224711461096, "300–1000 Hz_10ms_log_rms_mae_db": 1.7394158570514113, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11878642615083673, "300–1000 Hz_10ms_envelope_correlation": 0.9878860873891135, "300–1000 Hz_10ms_modulation_mae_db": 0.8817482344778386, "300–1000 Hz_40ms_log_rms_mae_db": 1.1389900621068996, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0656899153554238, "300–1000 Hz_40ms_envelope_correlation": 0.996669417633011, "300–1000 Hz_40ms_modulation_mae_db": 0.4308659248449223, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0771202899458645, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18907802386990025, "1000–3000 Hz_10ms_envelope_correlation": 0.9735612426044095, "1000–3000 Hz_10ms_modulation_mae_db": 0.8071447733233116, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4488259183207666, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10235255590352847, "1000–3000 Hz_40ms_envelope_correlation": 0.9931978252607786, "1000–3000 Hz_40ms_modulation_mae_db": 0.7463541750742294, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3199242601591714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22989078496691231, "3000–6000 Hz_10ms_envelope_correlation": 0.9614175378713253, "3000–6000 Hz_10ms_modulation_mae_db": 1.1079873831590017, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7728437012155405, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14372626276392142, "3000–6000 Hz_40ms_envelope_correlation": 0.9853751185426952, "3000–6000 Hz_40ms_modulation_mae_db": 1.0657283801313466, "mr_stft_db": 6.159276725225497, "band_mae_db": 0.2176503021917624, "rms_gain_error_db": -0.18171771539287984, "output_peak": 0.7156757567471298}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8597679047662931, "全帯域_10ms_rms_envelope_nrmse": 0.0830131529531078, "全帯域_10ms_envelope_correlation": 0.9928058901389359, "全帯域_10ms_modulation_mae_db": 0.31002626369100933, "全帯域_40ms_log_rms_mae_db": 0.45340038410672323, "全帯域_40ms_rms_envelope_nrmse": 0.04422474670066748, "全帯域_40ms_envelope_correlation": 0.9978758213747757, "全帯域_40ms_modulation_mae_db": 0.20065238769086446, "300–1000 Hz_10ms_log_rms_mae_db": 1.663990126956659, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11405116126252168, "300–1000 Hz_10ms_envelope_correlation": 0.9883966482702893, "300–1000 Hz_10ms_modulation_mae_db": 0.5455931627729974, "300–1000 Hz_40ms_log_rms_mae_db": 1.0449205788184441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06192505841630544, "300–1000 Hz_40ms_envelope_correlation": 0.9964318449613777, "300–1000 Hz_40ms_modulation_mae_db": 0.14579737873147017, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0323692202959385, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18054857476093877, "1000–3000 Hz_10ms_envelope_correlation": 0.9767338797558702, "1000–3000 Hz_10ms_modulation_mae_db": 1.0811346528041301, "1000–3000 Hz_40ms_log_rms_mae_db": 1.358738763114615, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10399705275990828, "1000–3000 Hz_40ms_envelope_correlation": 0.9939049166653843, "1000–3000 Hz_40ms_modulation_mae_db": 0.8931672398519074, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2512079281011155, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21927794838305043, "3000–6000 Hz_10ms_envelope_correlation": 0.9648476063149609, "3000–6000 Hz_10ms_modulation_mae_db": 0.2802953737982134, "3000–6000 Hz_40ms_log_rms_mae_db": 1.695866100337361, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13371692369789867, "3000–6000 Hz_40ms_envelope_correlation": 0.9866465507404996, "3000–6000 Hz_40ms_modulation_mae_db": 0.22216110820414392, "mr_stft_db": 6.144425268755665, "band_mae_db": 0.2505064310541577, "rms_gain_error_db": -0.09285096735560787, "output_peak": 0.6695070266096274}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8185728852332994, "全帯域_10ms_rms_envelope_nrmse": 0.07252948589847129, "全帯域_10ms_envelope_correlation": 0.9945753058724287, "全帯域_10ms_modulation_mae_db": 0.25003034742807384, "全帯域_40ms_log_rms_mae_db": 0.42507725268893254, "全帯域_40ms_rms_envelope_nrmse": 0.03841775578642931, "全帯域_40ms_envelope_correlation": 0.9984693359149582, "全帯域_40ms_modulation_mae_db": 0.15587744808000834, "300–1000 Hz_10ms_log_rms_mae_db": 1.6535872012218162, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10188312725647522, "300–1000 Hz_10ms_envelope_correlation": 0.9909269415503551, "300–1000 Hz_10ms_modulation_mae_db": 0.29413387930238033, "300–1000 Hz_40ms_log_rms_mae_db": 1.0544114577432648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059774214536776456, "300–1000 Hz_40ms_envelope_correlation": 0.9968773628427317, "300–1000 Hz_40ms_modulation_mae_db": 0.4482644621307305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.975529763487399, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18500966815618441, "1000–3000 Hz_10ms_envelope_correlation": 0.974327819462297, "1000–3000 Hz_10ms_modulation_mae_db": 0.9476113187532705, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4388924801578098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11523972522661853, "1000–3000 Hz_40ms_envelope_correlation": 0.9901732562726386, "1000–3000 Hz_40ms_modulation_mae_db": 0.6493835709937422, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3196002914694143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21887700036499735, "3000–6000 Hz_10ms_envelope_correlation": 0.9649457477816735, "3000–6000 Hz_10ms_modulation_mae_db": 0.9812369862683159, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7841893627436565, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13201129850297186, "3000–6000 Hz_40ms_envelope_correlation": 0.9869280184935116, "3000–6000 Hz_40ms_modulation_mae_db": 0.6118331433015659, "mr_stft_db": 6.137893460399653, "band_mae_db": 0.20544915312476322, "rms_gain_error_db": -0.10839185719875745, "output_peak": 0.6758332631779664}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8477356340586513, "全帯域_10ms_rms_envelope_nrmse": 0.07685399168018375, "全帯域_10ms_envelope_correlation": 0.9938645099971825, "全帯域_10ms_modulation_mae_db": 0.18213299831971316, "全帯域_40ms_log_rms_mae_db": 0.4472649767460297, "全帯域_40ms_rms_envelope_nrmse": 0.03917763366392371, "全帯域_40ms_envelope_correlation": 0.9983857782393145, "全帯域_40ms_modulation_mae_db": 0.3306010950937759, "300–1000 Hz_10ms_log_rms_mae_db": 1.6996320952514614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11119367947339855, "300–1000 Hz_10ms_envelope_correlation": 0.9891698568442623, "300–1000 Hz_10ms_modulation_mae_db": 0.46259917468667533, "300–1000 Hz_40ms_log_rms_mae_db": 1.1383333502335113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05879203806022795, "300–1000 Hz_40ms_envelope_correlation": 0.9970866583360013, "300–1000 Hz_40ms_modulation_mae_db": 0.6421656017038012, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0009556480325785, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17919392231742862, "1000–3000 Hz_10ms_envelope_correlation": 0.9761462821078208, "1000–3000 Hz_10ms_modulation_mae_db": 0.9960738479239467, "1000–3000 Hz_40ms_log_rms_mae_db": 1.354335160421885, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10639722764308499, "1000–3000 Hz_40ms_envelope_correlation": 0.9920464029149554, "1000–3000 Hz_40ms_modulation_mae_db": 0.5135702072680576, "3000–6000 Hz_10ms_log_rms_mae_db": 2.330693232977736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23578126851490344, "3000–6000 Hz_10ms_envelope_correlation": 0.959176890114709, "3000–6000 Hz_10ms_modulation_mae_db": 0.5110375028727638, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7664287253747493, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14159312420500386, "3000–6000 Hz_40ms_envelope_correlation": 0.9848435331698234, "3000–6000 Hz_40ms_modulation_mae_db": 0.32734503149044136, "mr_stft_db": 6.160983965556543, "band_mae_db": 0.2358920404614087, "rms_gain_error_db": -0.10148560533180465, "output_peak": 0.740393916492142}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8986279260673028, "全帯域_10ms_rms_envelope_nrmse": 0.07843517766555219, "全帯域_10ms_envelope_correlation": 0.9937766487199122, "全帯域_10ms_modulation_mae_db": 0.6066359874749119, "全帯域_40ms_log_rms_mae_db": 0.48064257026818474, "全帯域_40ms_rms_envelope_nrmse": 0.04421437764797157, "全帯域_40ms_envelope_correlation": 0.9981095750119544, "全帯域_40ms_modulation_mae_db": 0.7419038110168829, "300–1000 Hz_10ms_log_rms_mae_db": 1.6512508536316342, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1088515891923024, "300–1000 Hz_10ms_envelope_correlation": 0.9896679957993885, "300–1000 Hz_10ms_modulation_mae_db": 0.9103792849984197, "300–1000 Hz_40ms_log_rms_mae_db": 1.0608868107420357, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06791143730292094, "300–1000 Hz_40ms_envelope_correlation": 0.9959640129268271, "300–1000 Hz_40ms_modulation_mae_db": 0.9482034996957429, "1000–3000 Hz_10ms_log_rms_mae_db": 2.068004818659226, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17221488935732557, "1000–3000 Hz_10ms_envelope_correlation": 0.9784825629534705, "1000–3000 Hz_10ms_modulation_mae_db": 0.33643415696654655, "1000–3000 Hz_40ms_log_rms_mae_db": 1.437194170082023, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10712376762831236, "1000–3000 Hz_40ms_envelope_correlation": 0.9924616320586185, "1000–3000 Hz_40ms_modulation_mae_db": 0.6546963023297329, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2546327047065073, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22545437645252714, "3000–6000 Hz_10ms_envelope_correlation": 0.962652085739883, "3000–6000 Hz_10ms_modulation_mae_db": 1.113377241661938, "3000–6000 Hz_40ms_log_rms_mae_db": 1.745598293450301, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13833901102390547, "3000–6000 Hz_40ms_envelope_correlation": 0.9855400219684224, "3000–6000 Hz_40ms_modulation_mae_db": 0.8367038985868506, "mr_stft_db": 6.1577210745327235, "band_mae_db": 0.2382540897072388, "rms_gain_error_db": -0.1583992209674253, "output_peak": 0.7153155576612354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.153570288726446, "spectral_convergence": 0.25603327295062106, "spectral_convergence_status": "ok", "amplitude_floor": 1.7851227833840954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.019561532170997, "spectral_convergence": 0.2546847378488782, "spectral_convergence_status": "ok", "amplitude_floor": 1.2048994709482952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.3262365512105765, "spectral_convergence": 0.27063077918510875, "spectral_convergence_status": "ok", "amplitude_floor": 9.468117719284587e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005960090393190935, "output_power": 0.0005972384295384697, "delta_db": 0.008948983194731325, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030564977634421338, "output_power": 0.0029577180999823807, "delta_db": -0.14267303295696215, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008630746056010172, "output_power": 0.0007938332813119241, "delta_db": -0.36319035805705396, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003160541845765363, "output_power": 0.0002728159894627364, "delta_db": -0.6389172450668106, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001143536135409032, "output_power": 0.00010133925314549587, "delta_db": -0.524721934550309, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.420642909107181e-05, "output_power": 3.742247273871073e-05, "delta_db": 0.3902474140327711, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3447831613097731, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.19605919454830212, "pitch": null, "output_rms_dbfs": -23.223595160967143, "output_peak": 0.6876760884185386, "reference_rms_dbfs": -23.02753596641884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.1664561240360065}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023539858364162597, "log_rms_mae_db": 0.8802663807172079, "rms_envelope_nrmse": 0.0875510952017504, "envelope_correlation": 0.9923011111816834, "rms_gain_error_db": -0.19586860359205355, "reference_mean_power": 0.004983368761309487, "output_mean_power": 0.0047636096360238155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0209735395978665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009552928696150586, "output_power": 0.0008868567345593926, "reference_active": true, "delta_db": -0.3228306785576871}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014624552237472608, "output_power": 0.00012884134816653738, "reference_active": true, "delta_db": -0.550273175560409}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.510798757145359e-05, "output_power": 2.887972207382303e-05, "reference_active": true, "delta_db": 0.6078110445075228}], "active_band_count": 3, "active_band_mae_db": 0.49363829954187294}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002083747898533542, "log_rms_mae_db": 0.47220734986349494, "rms_envelope_nrmse": 0.05100930172474126, "envelope_correlation": 0.9975461937799094, "rms_gain_error_db": -0.1962819916190904, "reference_mean_power": 0.004999205734592124, "output_mean_power": 0.004778293373609376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8665599913970187e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008900977663526511, "output_power": 0.0008257805426087704, "reference_active": true, "delta_db": -0.325730655731871}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.407197015766335e-05, "output_power": 6.580752883471375e-05, "reference_active": true, "delta_db": -0.5137831338200557}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8378799869222006e-06, "output_power": 2.140328866313193e-06, "reference_active": true, "delta_db": 0.6616336013298736}], "active_band_count": 3, "active_band_mae_db": 0.5003824636272668}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019850808969518194, "log_rms_mae_db": 1.805724169616414, "rms_envelope_nrmse": 0.1276876468440612, "envelope_correlation": 0.9859212336725739, "rms_gain_error_db": -0.28862840292720404, "reference_mean_power": 0.003121339390070273, "output_mean_power": 0.00292064083309926}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3811041144776226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007780170156378303, "output_power": 0.000672120518070679, "reference_active": true, "delta_db": -0.6354194186953914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001204320773242115, "output_power": 9.74899883480561e-05, "reference_active": true, "delta_db": -0.9178215904820103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.313510568503667e-05, "output_power": 2.8534790135801878e-05, "reference_active": true, "delta_db": 0.9110319503021493}], "active_band_count": 3, "active_band_mae_db": 0.8214243198265171}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016710142967257498, "log_rms_mae_db": 1.1827954071682072, "rms_envelope_nrmse": 0.08363903931104577, "envelope_correlation": 0.9940427185569701, "rms_gain_error_db": -0.289132475208169, "reference_mean_power": 0.003133268891618282, "output_mean_power": 0.0029314630156610697}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2683362614683906e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007382464262949856, "output_power": 0.0006375588190989615, "reference_active": true, "delta_db": -0.6368109576054161}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8108189032543396e-05, "output_power": 4.7277679268330286e-05, "reference_active": true, "delta_db": -0.8958119059887798}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.916170233531451e-06, "output_power": 2.1901964372884426e-06, "reference_active": true, "delta_db": 0.580489787778814}], "active_band_count": 3, "active_band_mae_db": 0.7043708837910033}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012483338724103395, "log_rms_mae_db": 2.0426660193514734, "rms_envelope_nrmse": 0.19117948313607486, "envelope_correlation": 0.9742891334861361, "rms_gain_error_db": -0.5969920986134271, "reference_mean_power": 0.00037365748369267646, "output_mean_power": 0.0003256675409941384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.54909855359075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.931499876625785e-05, "output_power": 7.771276776741516e-05, "reference_active": true, "delta_db": -0.6043201977879794}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4977670974468328e-05, "output_power": 9.921382394648133e-06, "reference_active": true, "delta_db": -1.7887209714763517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3989606403687574e-06, "output_power": 2.9270949315675144e-06, "reference_active": true, "delta_db": -0.6490932782012114}], "active_band_count": 3, "active_band_mae_db": 1.0140448158218476}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010596065945345535, "log_rms_mae_db": 1.3835838875170945, "rms_envelope_nrmse": 0.11684686161170309, "envelope_correlation": 0.9924774614232444, "rms_gain_error_db": -0.5969466526210832, "reference_mean_power": 0.00037540546219302315, "output_mean_power": 0.0003271944452649797}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4355698091451528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.577569295489126e-05, "output_power": 7.380228471819083e-05, "reference_active": true, "delta_db": -0.6529442866294682}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.83374089374301e-06, "output_power": 5.377775468800386e-06, "reference_active": true, "delta_db": -1.6336653712845455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527528900320414e-07, "output_power": 1.8202135123554193e-07, "reference_active": true, "delta_db": -1.4257379608136846}], "active_band_count": 3, "active_band_mae_db": 1.2374492062425662}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.3292043050207156e-05, "log_rms_mae_db": 2.3508844200973518, "rms_envelope_nrmse": 0.22728134243868375, "envelope_correlation": 0.9620369892453413, "rms_gain_error_db": -0.2521660157347059, "reference_mean_power": 5.712588023845408e-05, "output_mean_power": 5.3903417869087425e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.279319333258023e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4391078150208969e-05, "output_power": 1.2110663554632652e-05, "reference_active": true, "delta_db": -0.7492539241423062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.344791688657969e-06, "output_power": 3.5074105406190436e-06, "reference_active": true, "delta_db": -0.9298235619956254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.199993664659886e-07, "output_power": 9.632715293365214e-07, "reference_active": true, "delta_db": 0.6993520752679204}], "active_band_count": 3, "active_band_mae_db": 0.7928098538019507}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.870517765122232e-05, "log_rms_mae_db": 1.821993025825247, "rms_envelope_nrmse": 0.13996967260859827, "envelope_correlation": 0.9855554696040555, "rms_gain_error_db": -0.2521529797823199, "reference_mean_power": 5.74115094616415e-05, "output_mean_power": 5.417309739830513e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.057417179151539e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3970684782124775e-05, "output_power": 1.1633261221098277e-05, "reference_active": true, "delta_db": -0.7951621371587924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.241188995835654e-06, "output_power": 1.938792109121427e-06, "reference_active": true, "delta_db": -0.6294723795215927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.11366674713116e-08, "output_power": 5.186841194676304e-08, "reference_active": true, "delta_db": 0.061705293709717765}], "active_band_count": 3, "active_band_mae_db": 0.4954466034633676}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8281109306862897, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "69531b3f3ea41f4615157a91813992ed7b4b8d0bd9733b0deaf3f70a5e9f0fed", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9754890227044544, "standard_deviation": 0.0016486391981229726, "minimum": 0.9735612426044095, "maximum": 0.9784825629534705, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.034048598659407, "standard_deviation": 0.038442103250134106, "minimum": 1.975529763487399, "maximum": 2.0771202899458645, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.85215055612691, "standard_deviation": 0.2585054398870494, "minimum": 0.33643415696654655, "maximum": 1.0811346528041301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18335688071187814, "standard_deviation": 0.006463507184399108, "minimum": 0.17221488935732557, "maximum": 0.19117948313607486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9922766692097983, "standard_deviation": 0.001254899219421862, "minimum": 0.9901732562726386, "maximum": 0.9939049166653843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4039666174996468, "standard_deviation": 0.03655264591016558, "minimum": 1.354335160421885, "maximum": 1.4488259183207666, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7968115942275222, "standard_deviation": 0.4019381719058792, "minimum": 0.20787143408443465, "maximum": 1.472000617975507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10939587936383216, "standard_deviation": 0.007617740858426293, "minimum": 0.10126392883101414, "maximum": 0.12194591530648753, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9621843736104373, "standard_deviation": 0.002031715938038277, "minimum": 0.959176890114709, "maximum": 0.9649457477816735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3046077558202045, "standard_deviation": 0.038590766687222666, "minimum": 2.2512079281011155, "maximum": 2.3508844200973518, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7953590639931711, "standard_deviation": 0.2905983314625938, "minimum": 0.2802953737982134, "maximum": 1.113377241661938, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22713736623537673, "standard_deviation": 0.006047524764160259, "minimum": 0.21887700036499735, "maximum": 0.23578126851490344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.985228272338967, "standard_deviation": 0.0012816141340112673, "minimum": 0.9833436285495782, "maximum": 0.9869280184935116, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7677712364035776, "standard_deviation": 0.03938213169605046, "minimum": 1.695866100337361, "maximum": 1.821993025825247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5969717199912239, "standard_deviation": 0.2683127990428249, "minimum": 0.22216110820414392, "maximum": 1.0657283801313466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14090281352895154, "standard_deviation": 0.006435330172326679, "minimum": 0.13201129850297186, "maximum": 0.15161030678969584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9887775154900997, "standard_deviation": 0.0014609829972255393, "minimum": 0.9859212336725739, "maximum": 0.9909269415503551, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6907924294670285, "standard_deviation": 0.05566597541964276, "minimum": 1.6512508536316342, "maximum": 1.805724169616414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6539280064732559, "standard_deviation": 0.2300415173040495, "minimum": 0.29413387930238033, "maximum": 0.9103792849984197, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11316465086487801, "standard_deviation": 0.007568508084900457, "minimum": 0.10188312725647522, "maximum": 0.1276876468440612, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960385850315494, "standard_deviation": 0.001059251571393369, "minimum": 0.9940427185569701, "maximum": 0.9970866583360013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0895232945584703, "standard_deviation": 0.05522004011619936, "minimum": 1.0354250765097184, "maximum": 1.1827954071682072, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6722052612487883, "standard_deviation": 0.32191908372325284, "minimum": 0.14579737873147017, "maximum": 1.0890919258177163, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06719784041252178, "standard_deviation": 0.008587013164159447, "minimum": 0.05879203806022795, "maximum": 0.08363903931104577, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25427015066198133, "standard_deviation": 0.044300815919910176, "minimum": 0.20544915312476322, "maximum": 0.3447831613097731, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.153301787990912, "standard_deviation": 0.009977221016650228, "minimum": 6.137893460399653, "maximum": 6.1664561240360065, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7022878197375235, "standard_deviation": 0.024238356725064435, "minimum": 0.6695070266096274, "maximum": 0.740393916492142, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1375242657906255, "standard_deviation": 0.037898153468504, "minimum": -0.19605919454830212, "maximum": -0.09285096735560787, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933202144480958, "standard_deviation": 0.0007290504030379381, "minimum": 0.9923011111816834, "maximum": 0.9945753058724287, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8638889789549262, "standard_deviation": 0.028373676644914036, "minimum": 0.8185728852332994, "maximum": 0.9029409367309431, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4036033752675274, "standard_deviation": 0.17854199389988765, "minimum": 0.18213299831971316, "maximum": 0.6872491560459016, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0806466815897679, "standard_deviation": 0.004708644570585868, "minimum": 0.07252948589847129, "maximum": 0.0875510952017504, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997940064175405, "standard_deviation": 0.000503225279650531, "minimum": 0.996943773086847, "maximum": 0.9984693359149582, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4617877258707223, "standard_deviation": 0.023371998141884655, "minimum": 0.42507725268893254, "maximum": 0.5024353874957248, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3992815074313978, "standard_deviation": 0.19923350702016096, "minimum": 0.15587744808000834, "maximum": 0.7419038110168829, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04490576680864719, "standard_deviation": 0.00518052208601032, "minimum": 0.03841775578642931, "maximum": 0.05347033452079015, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0028", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.778362989425659, "predicted_mos": 2.778362989425659, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6cbd0d388b604e1f43347dff6f1dc052b8d0fb9780b463bf1149da2bb1de9bcd", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7106572999618948, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.930399393811061, "全帯域_10ms_rms_envelope_nrmse": 0.092833126990436, "全帯域_10ms_envelope_correlation": 0.9909613551099107, "全帯域_10ms_modulation_mae_db": 0.23898313586163145, "全帯域_40ms_log_rms_mae_db": 0.5595683658994187, "全帯域_40ms_rms_envelope_nrmse": 0.05137291840931202, "全帯域_40ms_envelope_correlation": 0.9969944971379767, "全帯域_40ms_modulation_mae_db": 0.11355430739465351, "300–1000 Hz_10ms_log_rms_mae_db": 1.5367319502023704, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11560378025726809, "300–1000 Hz_10ms_envelope_correlation": 0.9880292281493446, "300–1000 Hz_10ms_modulation_mae_db": 0.454541971696641, "300–1000 Hz_40ms_log_rms_mae_db": 1.005359770136125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06398131497385935, "300–1000 Hz_40ms_envelope_correlation": 0.9960177253095029, "300–1000 Hz_40ms_modulation_mae_db": 0.18592331443352442, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6622494212144998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18221544200593223, "1000–3000 Hz_10ms_envelope_correlation": 0.975014893810155, "1000–3000 Hz_10ms_modulation_mae_db": 0.8039917805675273, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1767664676321492, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11311953132079071, "1000–3000 Hz_40ms_envelope_correlation": 0.9901338702061719, "1000–3000 Hz_40ms_modulation_mae_db": 0.6509606671987186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7365919322250472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19461247555900138, "3000–6000 Hz_10ms_envelope_correlation": 0.9722166064613218, "3000–6000 Hz_10ms_modulation_mae_db": 0.5475093569045649, "3000–6000 Hz_40ms_log_rms_mae_db": 1.369121560014756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13347705143544614, "3000–6000 Hz_40ms_envelope_correlation": 0.9863603844290297, "3000–6000 Hz_40ms_modulation_mae_db": 0.7095478073658056, "mr_stft_db": 5.769748425438926, "band_mae_db": 0.18815585034974372, "rms_gain_error_db": 0.011108265467309239, "output_peak": 0.7069389801548915}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8789687469346757, "全帯域_10ms_rms_envelope_nrmse": 0.07868226475020114, "全帯域_10ms_envelope_correlation": 0.993600841462585, "全帯域_10ms_modulation_mae_db": 0.5404576710572303, "全帯域_40ms_log_rms_mae_db": 0.47540407017785613, "全帯域_40ms_rms_envelope_nrmse": 0.0454231454584228, "全帯域_40ms_envelope_correlation": 0.9978268575757845, "全帯域_40ms_modulation_mae_db": 0.432813419392246, "300–1000 Hz_10ms_log_rms_mae_db": 1.5066344996069077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11160276035514315, "300–1000 Hz_10ms_envelope_correlation": 0.9891675367372053, "300–1000 Hz_10ms_modulation_mae_db": 0.5800235694678163, "300–1000 Hz_40ms_log_rms_mae_db": 1.0433688595640684, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07074813333828722, "300–1000 Hz_40ms_envelope_correlation": 0.9956309394319872, "300–1000 Hz_40ms_modulation_mae_db": 0.3712146005560017, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6634792481878644, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16913061000328541, "1000–3000 Hz_10ms_envelope_correlation": 0.9802532332826139, "1000–3000 Hz_10ms_modulation_mae_db": 0.570441642390461, "1000–3000 Hz_40ms_log_rms_mae_db": 1.136746295233105, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10894771889927997, "1000–3000 Hz_40ms_envelope_correlation": 0.9933835961438531, "1000–3000 Hz_40ms_modulation_mae_db": 0.7855004636554445, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7461612626600092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18603596871322992, "3000–6000 Hz_10ms_envelope_correlation": 0.974703701649474, "3000–6000 Hz_10ms_modulation_mae_db": 0.3017863184239804, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3860589935840857, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11777136040803306, "3000–6000 Hz_40ms_envelope_correlation": 0.9893023087914811, "3000–6000 Hz_40ms_modulation_mae_db": 0.4162442982543266, "mr_stft_db": 5.78081838784211, "band_mae_db": 0.2805808463360784, "rms_gain_error_db": -0.11752299448156324, "output_peak": 0.733605941446147}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8854418867990377, "全帯域_10ms_rms_envelope_nrmse": 0.08454767416965, "全帯域_10ms_envelope_correlation": 0.9925038998960657, "全帯域_10ms_modulation_mae_db": 0.9494072044743636, "全帯域_40ms_log_rms_mae_db": 0.46744369049035694, "全帯域_40ms_rms_envelope_nrmse": 0.04486394577857019, "全帯域_40ms_envelope_correlation": 0.9977876789373206, "全帯域_40ms_modulation_mae_db": 0.8927714539750156, "300–1000 Hz_10ms_log_rms_mae_db": 1.48223343589135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10970822306973863, "300–1000 Hz_10ms_envelope_correlation": 0.9892527976973751, "300–1000 Hz_10ms_modulation_mae_db": 0.984065903837426, "300–1000 Hz_40ms_log_rms_mae_db": 0.9586797241936179, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0574161565135267, "300–1000 Hz_40ms_envelope_correlation": 0.9969307250681615, "300–1000 Hz_40ms_modulation_mae_db": 0.9538865998059466, "1000–3000 Hz_10ms_log_rms_mae_db": 1.648888143817497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15826436419391213, "1000–3000 Hz_10ms_envelope_correlation": 0.9837700023676822, "1000–3000 Hz_10ms_modulation_mae_db": 1.3374165938523008, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1860323342218921, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10391495787351038, "1000–3000 Hz_40ms_envelope_correlation": 0.9950701217912116, "1000–3000 Hz_40ms_modulation_mae_db": 0.9640794969703048, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7115806778490639, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1705911050302967, "3000–6000 Hz_10ms_envelope_correlation": 0.9787632551771585, "3000–6000 Hz_10ms_modulation_mae_db": 0.20193356637926177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3287750531691178, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11553763211077307, "3000–6000 Hz_40ms_envelope_correlation": 0.98987881991289, "3000–6000 Hz_40ms_modulation_mae_db": 0.42509874806073916, "mr_stft_db": 5.758848821760396, "band_mae_db": 0.2319134613322161, "rms_gain_error_db": -0.07596354624458945, "output_peak": 0.7021492798595016}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8811943789933905, "全帯域_10ms_rms_envelope_nrmse": 0.08472072481583935, "全帯域_10ms_envelope_correlation": 0.9925531061960693, "全帯域_10ms_modulation_mae_db": 0.19411434795506866, "全帯域_40ms_log_rms_mae_db": 0.500248590776652, "全帯域_40ms_rms_envelope_nrmse": 0.045492859437236056, "全帯域_40ms_envelope_correlation": 0.9978213917448271, "全帯域_40ms_modulation_mae_db": 0.17374215356445477, "300–1000 Hz_10ms_log_rms_mae_db": 1.4757254271329918, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11138681005117339, "300–1000 Hz_10ms_envelope_correlation": 0.9890869445897645, "300–1000 Hz_10ms_modulation_mae_db": 0.27203309678642623, "300–1000 Hz_40ms_log_rms_mae_db": 1.0239702200106744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06546866322887074, "300–1000 Hz_40ms_envelope_correlation": 0.9961926820820443, "300–1000 Hz_40ms_modulation_mae_db": 0.23623577861410214, "1000–3000 Hz_10ms_log_rms_mae_db": 1.643098368778596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15692860522647553, "1000–3000 Hz_10ms_envelope_correlation": 0.9826766182023566, "1000–3000 Hz_10ms_modulation_mae_db": 0.7024164369863937, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1360365356877573, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09101491247784764, "1000–3000 Hz_40ms_envelope_correlation": 0.9953494702491981, "1000–3000 Hz_40ms_modulation_mae_db": 1.0713062325724414, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6799043182679583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1693088093484832, "3000–6000 Hz_10ms_envelope_correlation": 0.9791287463236367, "3000–6000 Hz_10ms_modulation_mae_db": 0.7592939028621598, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3037138108945174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10602296786726842, "3000–6000 Hz_40ms_envelope_correlation": 0.9916854789708717, "3000–6000 Hz_40ms_modulation_mae_db": 0.578284396897151, "mr_stft_db": 5.771974400160912, "band_mae_db": 0.25891070110177067, "rms_gain_error_db": -0.11870517730186847, "output_peak": 0.7318720552139033}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8674259165061103, "全帯域_10ms_rms_envelope_nrmse": 0.07881864591280395, "全帯域_10ms_envelope_correlation": 0.9936775703256338, "全帯域_10ms_modulation_mae_db": 0.36199257865772766, "全帯域_40ms_log_rms_mae_db": 0.5172148534410348, "全帯域_40ms_rms_envelope_nrmse": 0.045553350596614424, "全帯域_40ms_envelope_correlation": 0.9979408879247839, "全帯域_40ms_modulation_mae_db": 0.44026262033739166, "300–1000 Hz_10ms_log_rms_mae_db": 1.5055248055598682, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10761895035978522, "300–1000 Hz_10ms_envelope_correlation": 0.989952070076542, "300–1000 Hz_10ms_modulation_mae_db": 0.5589207327377134, "300–1000 Hz_40ms_log_rms_mae_db": 1.0144572104914649, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061523987741531456, "300–1000 Hz_40ms_envelope_correlation": 0.9968454315932561, "300–1000 Hz_40ms_modulation_mae_db": 0.7788074996265743, "1000–3000 Hz_10ms_log_rms_mae_db": 1.617329785756415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16222905337556293, "1000–3000 Hz_10ms_envelope_correlation": 0.9817665112154194, "1000–3000 Hz_10ms_modulation_mae_db": 0.7395135706889303, "1000–3000 Hz_40ms_log_rms_mae_db": 1.160091049993401, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09360286843950671, "1000–3000 Hz_40ms_envelope_correlation": 0.995523717155456, "1000–3000 Hz_40ms_modulation_mae_db": 0.42701149584532033, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7421478052755783, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1849550087110351, "3000–6000 Hz_10ms_envelope_correlation": 0.974952271124544, "3000–6000 Hz_10ms_modulation_mae_db": 0.4763376380637904, "3000–6000 Hz_40ms_log_rms_mae_db": 1.368986625455308, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1066463436422821, "3000–6000 Hz_40ms_envelope_correlation": 0.9915039193450227, "3000–6000 Hz_40ms_modulation_mae_db": 0.33868858053426815, "mr_stft_db": 5.767285239370853, "band_mae_db": 0.24744568097053646, "rms_gain_error_db": -0.15167050918793598, "output_peak": 0.7229417852661769}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8846740333416265, "全帯域_10ms_rms_envelope_nrmse": 0.08459401035121503, "全帯域_10ms_envelope_correlation": 0.9925760967829473, "全帯域_10ms_modulation_mae_db": 0.4884015132417572, "全帯域_40ms_log_rms_mae_db": 0.46935316481608386, "全帯域_40ms_rms_envelope_nrmse": 0.04394305438687711, "全帯域_40ms_envelope_correlation": 0.9979540672400481, "全帯域_40ms_modulation_mae_db": 0.12422159921467472, "300–1000 Hz_10ms_log_rms_mae_db": 1.5674283345432662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11378067850209476, "300–1000 Hz_10ms_envelope_correlation": 0.9885326374215336, "300–1000 Hz_10ms_modulation_mae_db": 0.5304480737223444, "300–1000 Hz_40ms_log_rms_mae_db": 1.0509072824755659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06589376549327211, "300–1000 Hz_40ms_envelope_correlation": 0.9960362732710945, "300–1000 Hz_40ms_modulation_mae_db": 0.3157415151989747, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5988821643162952, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16889992306869073, "1000–3000 Hz_10ms_envelope_correlation": 0.9787480579732829, "1000–3000 Hz_10ms_modulation_mae_db": 1.8652559376129074, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1086703143933252, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09618617376284568, "1000–3000 Hz_40ms_envelope_correlation": 0.993336202587239, "1000–3000 Hz_40ms_modulation_mae_db": 1.3877734450406525, "3000–6000 Hz_10ms_log_rms_mae_db": 1.807660685233112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1913267818312753, "3000–6000 Hz_10ms_envelope_correlation": 0.9745307898436055, "3000–6000 Hz_10ms_modulation_mae_db": 1.1133648812180545, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4001519372569444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12373039528123399, "3000–6000 Hz_40ms_envelope_correlation": 0.9905704112179501, "3000–6000 Hz_40ms_modulation_mae_db": 1.0548809118478317, "mr_stft_db": 5.781358084925408, "band_mae_db": 0.20276506050854204, "rms_gain_error_db": -0.11070797258863842, "output_peak": 0.769710662470845}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9160237805754397, "全帯域_10ms_rms_envelope_nrmse": 0.08214631654478978, "全帯域_10ms_envelope_correlation": 0.9930657044193687, "全帯域_10ms_modulation_mae_db": 0.45091037916381155, "全帯域_40ms_log_rms_mae_db": 0.49855482146033253, "全帯域_40ms_rms_envelope_nrmse": 0.04520336271112225, "全帯域_40ms_envelope_correlation": 0.9979064377683285, "全帯域_40ms_modulation_mae_db": 0.13848290962796622, "300–1000 Hz_10ms_log_rms_mae_db": 1.5831600431674484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11042475207253628, "300–1000 Hz_10ms_envelope_correlation": 0.9892604197279027, "300–1000 Hz_10ms_modulation_mae_db": 0.5405623644298746, "300–1000 Hz_40ms_log_rms_mae_db": 1.1197421701060581, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06617795500338962, "300–1000 Hz_40ms_envelope_correlation": 0.9960438324589235, "300–1000 Hz_40ms_modulation_mae_db": 0.36935950234985954, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6496092154646642, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15847991385440532, "1000–3000 Hz_10ms_envelope_correlation": 0.9829195508086442, "1000–3000 Hz_10ms_modulation_mae_db": 0.7198621651972914, "1000–3000 Hz_40ms_log_rms_mae_db": 1.162278572877262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09944308568482396, "1000–3000 Hz_40ms_envelope_correlation": 0.9948451304298921, "1000–3000 Hz_40ms_modulation_mae_db": 0.9092986971779334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8356243866016968, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2287609752904519, "3000–6000 Hz_10ms_envelope_correlation": 0.9618216601844454, "3000–6000 Hz_10ms_modulation_mae_db": 0.6887900774664284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.435145739223066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15474946819530508, "3000–6000 Hz_40ms_envelope_correlation": 0.9814466920844621, "3000–6000 Hz_40ms_modulation_mae_db": 0.6390416235106843, "mr_stft_db": 5.765195166064135, "band_mae_db": 0.19730576599913627, "rms_gain_error_db": -0.13499349873693473, "output_peak": 0.7506509375585669}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8893363255759088, "全帯域_10ms_rms_envelope_nrmse": 0.08433627031209838, "全帯域_10ms_envelope_correlation": 0.9925475852298725, "全帯域_10ms_modulation_mae_db": 0.5762755046823784, "全帯域_40ms_log_rms_mae_db": 0.5076848995872223, "全帯域_40ms_rms_envelope_nrmse": 0.048236963353945624, "全帯域_40ms_envelope_correlation": 0.9973833296195794, "全帯域_40ms_modulation_mae_db": 0.2878076712268289, "300–1000 Hz_10ms_log_rms_mae_db": 1.5025637141037125, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.120191610249401, "300–1000 Hz_10ms_envelope_correlation": 0.9871197601812318, "300–1000 Hz_10ms_modulation_mae_db": 0.8045307689210527, "300–1000 Hz_40ms_log_rms_mae_db": 1.013662244450318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07240644407597562, "300–1000 Hz_40ms_envelope_correlation": 0.9949615547872073, "300–1000 Hz_40ms_modulation_mae_db": 0.3837132892748462, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6788585261159927, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15598256349500536, "1000–3000 Hz_10ms_envelope_correlation": 0.9829648561149833, "1000–3000 Hz_10ms_modulation_mae_db": 0.32252325104820656, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2002248510067473, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09914442802623641, "1000–3000 Hz_40ms_envelope_correlation": 0.9942011665899693, "1000–3000 Hz_40ms_modulation_mae_db": 0.4802113397954069, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7551360747312281, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19492310327927023, "3000–6000 Hz_10ms_envelope_correlation": 0.9721444739373201, "3000–6000 Hz_10ms_modulation_mae_db": 0.8830554906667946, "3000–6000 Hz_40ms_log_rms_mae_db": 1.366038889661173, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11577446926577915, "3000–6000 Hz_40ms_envelope_correlation": 0.9897479777642676, "3000–6000 Hz_40ms_modulation_mae_db": 0.7988010574676538, "mr_stft_db": 5.777933247785332, "band_mae_db": 0.20086321055432133, "rms_gain_error_db": -0.056924217702658896, "output_peak": 0.7193593735289778}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.667770551879372, "spectral_convergence": 0.265267092194358, "spectral_convergence_status": "ok", "amplitude_floor": 1.7851227833840954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.564310106606256, "spectral_convergence": 0.2542050984728654, "spectral_convergence_status": "ok", "amplitude_floor": 1.2048994709482952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.077164617831149, "spectral_convergence": 0.28669495399400036, "spectral_convergence_status": "ok", "amplitude_floor": 9.468117719284587e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005960090393190935, "output_power": 0.0005992418866804719, "delta_db": 0.02349316219873504, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030564977634421338, "output_power": 0.0031186789486888165, "delta_db": 0.08746586429207039, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008630746056010172, "output_power": 0.0008324536355847056, "delta_db": -0.15688284127524205, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003160541845765363, "output_power": 0.0003034390592669863, "delta_db": -0.1769006150609999, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001143536135409032, "output_power": 0.00010305129724897687, "delta_db": -0.45196429524839254, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.420642909107181e-05, "output_power": 3.6085320894615775e-05, "delta_db": 0.23222832402302263, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18815585034974372, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.011108265467309239, "pitch": null, "output_rms_dbfs": -23.01642770095153, "output_peak": 0.7069389801548915, "reference_rms_dbfs": -23.02753596641884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.769748425438926}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023539858364162597, "log_rms_mae_db": 0.930399393811061, "rms_envelope_nrmse": 0.092833126990436, "envelope_correlation": 0.9909613551099107, "rms_gain_error_db": 0.011132714108440965, "reference_mean_power": 0.004983368761309487, "output_mean_power": 0.004996159526465157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0209735395978665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009552928696150586, "output_power": 0.0009566153351105194, "reference_active": true, "delta_db": 0.006008024416273117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014624552237472608, "output_power": 0.00015135690840614225, "reference_active": true, "delta_db": 0.1491967022730061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.510798757145359e-05, "output_power": 2.857500881826753e-05, "reference_active": true, "delta_db": 0.5617446808956151}], "active_band_count": 3, "active_band_mae_db": 0.23898313586163145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002083747898533542, "log_rms_mae_db": 0.5595683658994187, "rms_envelope_nrmse": 0.05137291840931202, "envelope_correlation": 0.9969944971379767, "rms_gain_error_db": 0.011293481754952265, "reference_mean_power": 0.004999205734592124, "output_mean_power": 0.005012222687976218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8665599913970187e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008900977663526511, "output_power": 0.0008953731081319137, "reference_active": true, "delta_db": 0.025663353284990763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.407197015766335e-05, "output_power": 7.783736727961583e-05, "reference_active": true, "delta_db": 0.2153424209575389}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8378799869222006e-06, "output_power": 1.8805412468264966e-06, "reference_active": true, "delta_db": 0.09965714794143092}], "active_band_count": 3, "active_band_mae_db": 0.11355430739465351}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019850808969518194, "log_rms_mae_db": 1.5367319502023704, "rms_envelope_nrmse": 0.11560378025726809, "envelope_correlation": 0.9880292281493446, "rms_gain_error_db": -0.028807427840608257, "reference_mean_power": 0.003121339390070273, "output_mean_power": 0.003100703576807306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3811041144776226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007780170156378303, "output_power": 0.0007514496400666337, "reference_active": true, "delta_db": -0.15089214574457932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001204320773242115, "output_power": 0.00011143265164861758, "reference_active": true, "delta_db": -0.33729712429733555}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.313510568503667e-05, "output_power": 2.830187133604641e-05, "reference_active": true, "delta_db": 0.8754366450480082}], "active_band_count": 3, "active_band_mae_db": 0.454541971696641}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016710142967257498, "log_rms_mae_db": 1.005359770136125, "rms_envelope_nrmse": 0.06398131497385935, "envelope_correlation": 0.9960177253095029, "rms_gain_error_db": -0.029143241439690624, "reference_mean_power": 0.003133268891618282, "output_mean_power": 0.003112313544333786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2683362614683906e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007382464262949856, "output_power": 0.0007160959898276239, "reference_active": true, "delta_db": -0.13230111645465453}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8108189032543396e-05, "output_power": 5.467922151211287e-05, "reference_active": true, "delta_db": -0.26415017937818386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.916170233531451e-06, "output_power": 1.8462998085251614e-06, "reference_active": true, "delta_db": -0.16131864746773486}], "active_band_count": 3, "active_band_mae_db": 0.18592331443352442}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012483338724103395, "log_rms_mae_db": 1.6622494212144998, "rms_envelope_nrmse": 0.18221544200593223, "envelope_correlation": 0.975014893810155, "rms_gain_error_db": -0.22249699994712058, "reference_mean_power": 0.00037365748369267646, "output_mean_power": 0.0003549964293518892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.54909855359075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.931499876625785e-05, "output_power": 8.394349736829296e-05, "reference_active": true, "delta_db": -0.26937337143849577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4977670974468328e-05, "output_power": 1.7575955430013656e-05, "reference_active": true, "delta_db": 0.6947465671973053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3989606403687574e-06, "output_power": 4.743857947671317e-06, "reference_active": true, "delta_db": 1.4478554030667805}], "active_band_count": 3, "active_band_mae_db": 0.8039917805675273}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010596065945345535, "log_rms_mae_db": 1.1767664676321492, "rms_envelope_nrmse": 0.11311953132079071, "envelope_correlation": 0.9901338702061719, "rms_gain_error_db": -0.22226864252607165, "reference_mean_power": 0.00037540546219302315, "output_mean_power": 0.0003566758649669329}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4355698091451528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.577569295489126e-05, "output_power": 8.181200394346624e-05, "reference_active": true, "delta_db": -0.20547204656902585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.83374089374301e-06, "output_power": 9.51456037992468e-06, "reference_active": true, "delta_db": 0.8441952365174905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527528900320414e-07, "output_power": 3.11184245756442e-07, "reference_active": true, "delta_db": 0.9032147185096395}], "active_band_count": 3, "active_band_mae_db": 0.6509606671987186}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.3292043050207156e-05, "log_rms_mae_db": 1.7365919322250472, "rms_envelope_nrmse": 0.19461247555900138, "envelope_correlation": 0.9722166064613218, "rms_gain_error_db": -0.1826500654351524, "reference_mean_power": 5.712588023845408e-05, "output_mean_power": 5.4773172780945955e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.279319333258023e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4391078150208969e-05, "output_power": 1.246615155161386e-05, "reference_active": true, "delta_db": -0.6236092928450173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.344791688657969e-06, "output_power": 4.0555967512740215e-06, "reference_active": true, "delta_db": -0.29914192512210813}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.199993664659886e-07, "output_power": 9.678124444762768e-07, "reference_active": true, "delta_db": 0.7197768527465693}], "active_band_count": 3, "active_band_mae_db": 0.5475093569045649}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.870517765122232e-05, "log_rms_mae_db": 1.369121560014756, "rms_envelope_nrmse": 0.13347705143544614, "envelope_correlation": 0.9863603844290297, "rms_gain_error_db": -0.1826841104980376, "reference_mean_power": 5.74115094616415e-05, "output_mean_power": 5.5046606953027315e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.057417179151539e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3970684782124775e-05, "output_power": 1.2209905115433965e-05, "reference_active": true, "delta_db": -0.5850540485400869}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.241188995835654e-06, "output_power": 2.1499342681648717e-06, "reference_active": true, "delta_db": -0.18053299378485094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.11366674713116e-08, "output_power": 6.999033303559259e-08, "reference_active": true, "delta_db": 1.363056379772479}], "active_band_count": 3, "active_band_mae_db": 0.7095478073658056}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8508305417420676, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d08526e0d4fc01d063de309534ee810de5cabdad23e3deb291a707260ddb7fed", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9810142154718922, "standard_deviation": 0.002929300718719594, "minimum": 0.975014893810155, "maximum": 0.9837700023676822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6452993592064782, "standard_deviation": 0.025959260016907293, "minimum": 1.5988821643162952, "maximum": 1.6788585261159927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8826776722930023, "standard_deviation": 0.48832107121079565, "minimum": 0.32252325104820656, "maximum": 1.8652559376129074, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1640163094029087, "standard_deviation": 0.008967573631379679, "minimum": 0.15598256349500536, "maximum": 0.18221544200593223, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939804093941239, "standard_deviation": 0.0017650223786057643, "minimum": 0.9901338702061719, "maximum": 0.995523717155456, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1583558026307048, "standard_deviation": 0.030024103737473648, "minimum": 1.1086703143933252, "maximum": 1.2002248510067473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8345177297820279, "standard_deviation": 0.3189781368251351, "minimum": 0.42701149584532033, "maximum": 1.3877734450406525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10067170956060517, "standard_deviation": 0.007570088113900994, "minimum": 0.09101491247784764, "maximum": 0.11311953132079071, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9735326880876882, "standard_deviation": 0.005396623013623953, "minimum": 0.9618216601844454, "maximum": 0.9791287463236367, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7518508928554617, "standard_deviation": 0.04971473062964183, "minimum": 1.6799043182679583, "maximum": 1.8356243866016968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6215089039981294, "standard_deviation": 0.30205415943584557, "minimum": 0.20193356637926177, "maximum": 1.1133648812180545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19006427847038046, "standard_deviation": 0.01849960784940415, "minimum": 0.1693088093484832, "maximum": 0.2287609752904519, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9888119990644968, "standard_deviation": 0.0034039486016517687, "minimum": 0.9814466920844621, "maximum": 0.9916854789708717, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3697490761573712, "standard_deviation": 0.040569760588100356, "minimum": 1.3037138108945174, "maximum": 1.435145739223066, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6200734279923075, "standard_deviation": 0.23599387582948628, "minimum": 0.33868858053426815, "maximum": 1.0548809118478317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12171371102576511, "standard_deviation": 0.016017803029846702, "minimum": 0.10602296786726842, "maximum": 0.15474946819530508, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9888001743226125, "standard_deviation": 0.0008824054455406273, "minimum": 0.9871197601812318, "maximum": 0.989952070076542, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5200002762759894, "standard_deviation": 0.03890893987476976, "minimum": 1.4757254271329918, "maximum": 1.5831600431674484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5906408101999119, "standard_deviation": 0.2164043480564958, "minimum": 0.27203309678642623, "maximum": 0.984065903837426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11253969561464255, "standard_deviation": 0.00393503313521618, "minimum": 0.10761895035978522, "maximum": 0.120191610249401, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960823955002721, "standard_deviation": 0.0006304129431777369, "minimum": 0.9949615547872073, "maximum": 0.9969307250681615, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0287684351784865, "standard_deviation": 0.046130165873002886, "minimum": 0.9586797241936179, "maximum": 1.1197421701060581, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44936026248247873, "standard_deviation": 0.2706008497381552, "minimum": 0.18592331443352442, "maximum": 0.9538865998059466, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0654520525460891, "standard_deviation": 0.004770485765768192, "minimum": 0.0574161565135267, "maximum": 0.07240644407597562, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2259925721440431, "standard_deviation": 0.03377304118050667, "minimum": 0.18815585034974372, "maximum": 0.2805808463360784, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.771645221668509, "standard_deviation": 0.007982952109712385, "minimum": 5.758848821760396, "maximum": 5.781358084925408, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7296536269373763, "standard_deviation": 0.022326985982807832, "minimum": 0.7021492798595016, "maximum": 0.769710662470845, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09442245634711, "standard_deviation": 0.05236606723451132, "minimum": -0.15167050918793598, "maximum": 0.011108265467309239, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9926857699278067, "standard_deviation": 0.0008469086819182621, "minimum": 0.9909613551099107, "maximum": 0.9936775703256338, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8916830578171563, "standard_deviation": 0.020856948956591865, "minimum": 0.8674259165061103, "maximum": 0.930399393811061, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4750677918867461, "standard_deviation": 0.23535228633755884, "minimum": 0.19411434795506866, "maximum": 0.9494072044743636, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0838348792308792, "standard_deviation": 0.004434478733650714, "minimum": 0.07868226475020114, "maximum": 0.092833126990436, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977018934935811, "standard_deviation": 0.0003384198235657178, "minimum": 0.9969944971379767, "maximum": 0.9979540672400481, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49943405708111965, "standard_deviation": 0.03049001860883318, "minimum": 0.46744369049035694, "maximum": 0.5595683658994187, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.32545701684165396, "standard_deviation": 0.26493773814503563, "minimum": 0.11355430739465351, "maximum": 0.8927714539750156, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04626120001651256, "standard_deviation": 0.0023985762893998573, "minimum": 0.04394305438687711, "maximum": 0.05137291840931202, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0028", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.724931240081787, "predicted_mos": 2.724931240081787, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "713e18f86a60698c4ada61a7c82caa31fcce9a27f583dad3ddaecd25d628dec8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.097137100063264, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.832702367238592, "全帯域_10ms_rms_envelope_nrmse": 0.07236631472236497, "全帯域_10ms_envelope_correlation": 0.9945672619672976, "全帯域_10ms_modulation_mae_db": 0.46149456462078137, "全帯域_40ms_log_rms_mae_db": 0.46155320167268865, "全帯域_40ms_rms_envelope_nrmse": 0.03996357516872767, "全帯域_40ms_envelope_correlation": 0.99827789074844, "全帯域_40ms_modulation_mae_db": 0.42983576554373193, "300–1000 Hz_10ms_log_rms_mae_db": 1.503535097929047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10111875855724442, "300–1000 Hz_10ms_envelope_correlation": 0.9909997534726843, "300–1000 Hz_10ms_modulation_mae_db": 0.4886278155993909, "300–1000 Hz_40ms_log_rms_mae_db": 1.0196121553399988, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05696601777719017, "300–1000 Hz_40ms_envelope_correlation": 0.997105709107865, "300–1000 Hz_40ms_modulation_mae_db": 0.41771233762975574, "1000–3000 Hz_10ms_log_rms_mae_db": 1.796065577341627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17182887101058392, "1000–3000 Hz_10ms_envelope_correlation": 0.9779177409598574, "1000–3000 Hz_10ms_modulation_mae_db": 0.6430988106364263, "1000–3000 Hz_40ms_log_rms_mae_db": 1.226472430665496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09359826439090384, "1000–3000 Hz_40ms_envelope_correlation": 0.9937238079220914, "1000–3000 Hz_40ms_modulation_mae_db": 0.3849736517290418, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9152653596188547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1945829972569677, "3000–6000 Hz_10ms_envelope_correlation": 0.9732877266673845, "3000–6000 Hz_10ms_modulation_mae_db": 0.8990641444108621, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4794213728744177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12930728350793502, "3000–6000 Hz_40ms_envelope_correlation": 0.9890564316555318, "3000–6000 Hz_40ms_modulation_mae_db": 0.7983057304032393, "mr_stft_db": 5.869873397407339, "band_mae_db": 0.19465915900803346, "rms_gain_error_db": -0.08563522301591411, "output_peak": 0.7320522824075959}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8821782178592095, "全帯域_10ms_rms_envelope_nrmse": 0.08785768411153154, "全帯域_10ms_envelope_correlation": 0.9919038912383024, "全帯域_10ms_modulation_mae_db": 0.6900866796822333, "全帯域_40ms_log_rms_mae_db": 0.5125105028412902, "全帯域_40ms_rms_envelope_nrmse": 0.052742984864431706, "全帯域_40ms_envelope_correlation": 0.9968656687294398, "全帯域_40ms_modulation_mae_db": 0.7399459378342387, "300–1000 Hz_10ms_log_rms_mae_db": 1.6010352946231925, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12392847923961142, "300–1000 Hz_10ms_envelope_correlation": 0.9862559842502392, "300–1000 Hz_10ms_modulation_mae_db": 0.6328831593149938, "300–1000 Hz_40ms_log_rms_mae_db": 1.0878023758420519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07371747501672561, "300–1000 Hz_40ms_envelope_correlation": 0.9947670746803375, "300–1000 Hz_40ms_modulation_mae_db": 0.788754068711472, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7745073758820626, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17634029171103405, "1000–3000 Hz_10ms_envelope_correlation": 0.976866163342058, "1000–3000 Hz_10ms_modulation_mae_db": 0.6261706469279993, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2896914066285896, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10066815169595052, "1000–3000 Hz_40ms_envelope_correlation": 0.992895902163374, "1000–3000 Hz_40ms_modulation_mae_db": 0.772964201116601, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9224180195905554, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2030474327952631, "3000–6000 Hz_10ms_envelope_correlation": 0.9700741094150087, "3000–6000 Hz_10ms_modulation_mae_db": 0.7882472619979944, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4882654806698314, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14111399159503168, "3000–6000 Hz_40ms_envelope_correlation": 0.9855538650847286, "3000–6000 Hz_40ms_modulation_mae_db": 0.7660359255113729, "mr_stft_db": 5.884514096473459, "band_mae_db": 0.261338321818491, "rms_gain_error_db": -0.06554880657071555, "output_peak": 0.6671881753249833}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8870584451100225, "全帯域_10ms_rms_envelope_nrmse": 0.08236607026470391, "全帯域_10ms_envelope_correlation": 0.9930714711746533, "全帯域_10ms_modulation_mae_db": 0.26690734238740665, "全帯域_40ms_log_rms_mae_db": 0.5285968802692279, "全帯域_40ms_rms_envelope_nrmse": 0.04669359769341373, "全帯域_40ms_envelope_correlation": 0.9978345943709596, "全帯域_40ms_modulation_mae_db": 0.20015141081647128, "300–1000 Hz_10ms_log_rms_mae_db": 1.4821628243819818, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11551931436755399, "300–1000 Hz_10ms_envelope_correlation": 0.9883526702809502, "300–1000 Hz_10ms_modulation_mae_db": 0.4107529617417794, "300–1000 Hz_40ms_log_rms_mae_db": 1.0169368166533324, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0672629931140808, "300–1000 Hz_40ms_envelope_correlation": 0.9961339428261909, "300–1000 Hz_40ms_modulation_mae_db": 0.5267832047837252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.79314600893367, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.178813930657738, "1000–3000 Hz_10ms_envelope_correlation": 0.977654246027363, "1000–3000 Hz_10ms_modulation_mae_db": 1.4678278675691085, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2581735513411565, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10639318772757876, "1000–3000 Hz_40ms_envelope_correlation": 0.9939082962841435, "1000–3000 Hz_40ms_modulation_mae_db": 1.0731749944604994, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9120694625327244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20085897049470777, "3000–6000 Hz_10ms_envelope_correlation": 0.9703831002252413, "3000–6000 Hz_10ms_modulation_mae_db": 0.5215290530267507, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5107982374289737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13000270344436757, "3000–6000 Hz_40ms_envelope_correlation": 0.9872130671914839, "3000–6000 Hz_40ms_modulation_mae_db": 0.6364225087915331, "mr_stft_db": 5.87746379685941, "band_mae_db": 0.30331580255133755, "rms_gain_error_db": -0.1529129868400272, "output_peak": 0.7706740667728899}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8999663086593753, "全帯域_10ms_rms_envelope_nrmse": 0.08255915839132213, "全帯域_10ms_envelope_correlation": 0.9931016808162437, "全帯域_10ms_modulation_mae_db": 0.49042916331160513, "全帯域_40ms_log_rms_mae_db": 0.5062701251848738, "全帯域_40ms_rms_envelope_nrmse": 0.04933302408383125, "全帯域_40ms_envelope_correlation": 0.9976027063137813, "全帯域_40ms_modulation_mae_db": 0.3502466261861243, "300–1000 Hz_10ms_log_rms_mae_db": 1.6344673988109084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10787104183086113, "300–1000 Hz_10ms_envelope_correlation": 0.9898753804674892, "300–1000 Hz_10ms_modulation_mae_db": 0.4926023154394021, "300–1000 Hz_40ms_log_rms_mae_db": 1.136766674828244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06517290182584522, "300–1000 Hz_40ms_envelope_correlation": 0.9963402461416868, "300–1000 Hz_40ms_modulation_mae_db": 0.3677324023619473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.791364389725002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17124258412544008, "1000–3000 Hz_10ms_envelope_correlation": 0.9796038050923187, "1000–3000 Hz_10ms_modulation_mae_db": 0.9011228788631778, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2900944971307884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10479641723245064, "1000–3000 Hz_40ms_envelope_correlation": 0.9939976959937105, "1000–3000 Hz_40ms_modulation_mae_db": 1.2648181331173045, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9083871701806328, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20061800743949357, "3000–6000 Hz_10ms_envelope_correlation": 0.9708912242253761, "3000–6000 Hz_10ms_modulation_mae_db": 0.6146487041291931, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4709478817430413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11666223396630679, "3000–6000 Hz_40ms_envelope_correlation": 0.9908150467512884, "3000–6000 Hz_40ms_modulation_mae_db": 0.6447840884598625, "mr_stft_db": 5.872109764973718, "band_mae_db": 0.23539206095422516, "rms_gain_error_db": -0.16952337887449168, "output_peak": 0.6957205225494536}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8479839778198931, "全帯域_10ms_rms_envelope_nrmse": 0.08109763773107462, "全帯域_10ms_envelope_correlation": 0.9931813825294695, "全帯域_10ms_modulation_mae_db": 0.3351372516639766, "全帯域_40ms_log_rms_mae_db": 0.4384253305057612, "全帯域_40ms_rms_envelope_nrmse": 0.0410007711328321, "全帯域_40ms_envelope_correlation": 0.998248746449189, "全帯域_40ms_modulation_mae_db": 0.3803868536592645, "300–1000 Hz_10ms_log_rms_mae_db": 1.560608790806329, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11375389654864906, "300–1000 Hz_10ms_envelope_correlation": 0.9886435304896395, "300–1000 Hz_10ms_modulation_mae_db": 0.5641358883005667, "300–1000 Hz_40ms_log_rms_mae_db": 1.0031896637366362, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06187990226262642, "300–1000 Hz_40ms_envelope_correlation": 0.9967007936817103, "300–1000 Hz_40ms_modulation_mae_db": 0.7296912190086732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7501895778568064, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1740109709388816, "1000–3000 Hz_10ms_envelope_correlation": 0.9786245951980023, "1000–3000 Hz_10ms_modulation_mae_db": 0.631567378647925, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2025416673778806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10581245160992413, "1000–3000 Hz_40ms_envelope_correlation": 0.9936414147024272, "1000–3000 Hz_40ms_modulation_mae_db": 0.7857907357411152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9031514880222773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18079057578033697, "3000–6000 Hz_10ms_envelope_correlation": 0.9761059246147744, "3000–6000 Hz_10ms_modulation_mae_db": 0.4064546778417004, "3000–6000 Hz_40ms_log_rms_mae_db": 1.503304173166996, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.112976837686651, "3000–6000 Hz_40ms_envelope_correlation": 0.9904369718595073, "3000–6000 Hz_40ms_modulation_mae_db": 0.3952712015811419, "mr_stft_db": 5.862588131460477, "band_mae_db": 0.253319533457148, "rms_gain_error_db": -0.11324076684139617, "output_peak": 0.6868023635237672}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8672626557629021, "全帯域_10ms_rms_envelope_nrmse": 0.0773024173808138, "全帯域_10ms_envelope_correlation": 0.9939958616957028, "全帯域_10ms_modulation_mae_db": 0.4335480676578283, "全帯域_40ms_log_rms_mae_db": 0.45726840941767827, "全帯域_40ms_rms_envelope_nrmse": 0.04244001089339652, "全帯域_40ms_envelope_correlation": 0.998321840077741, "全帯域_40ms_modulation_mae_db": 0.28260514620404914, "300–1000 Hz_10ms_log_rms_mae_db": 1.5683679810920779, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.111242525527137, "300–1000 Hz_10ms_envelope_correlation": 0.9894158745510793, "300–1000 Hz_10ms_modulation_mae_db": 0.525175105269781, "300–1000 Hz_40ms_log_rms_mae_db": 1.0176014227309567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06475908387913488, "300–1000 Hz_40ms_envelope_correlation": 0.9966666354875865, "300–1000 Hz_40ms_modulation_mae_db": 0.4863770709404102, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7218907321831338, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1724122138792666, "1000–3000 Hz_10ms_envelope_correlation": 0.9783385212627798, "1000–3000 Hz_10ms_modulation_mae_db": 1.4777342606535069, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1843962861533428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09996302539412327, "1000–3000 Hz_40ms_envelope_correlation": 0.993559369020876, "1000–3000 Hz_40ms_modulation_mae_db": 1.034371763693801, "3000–6000 Hz_10ms_log_rms_mae_db": 1.882328922328677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.196412855281422, "3000–6000 Hz_10ms_envelope_correlation": 0.9732956536075176, "3000–6000 Hz_10ms_modulation_mae_db": 0.6223285660770592, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4276030620763456, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12880199658275987, "3000–6000 Hz_40ms_envelope_correlation": 0.9899417583955301, "3000–6000 Hz_40ms_modulation_mae_db": 0.4287571913238671, "mr_stft_db": 5.861381523847254, "band_mae_db": 0.2364920089096311, "rms_gain_error_db": -0.16651194926869564, "output_peak": 0.7009099964283662}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8774385313519563, "全帯域_10ms_rms_envelope_nrmse": 0.07778950161235905, "全帯域_10ms_envelope_correlation": 0.9937897928630948, "全帯域_10ms_modulation_mae_db": 0.42387601633677696, "全帯域_40ms_log_rms_mae_db": 0.4578976237637284, "全帯域_40ms_rms_envelope_nrmse": 0.042523201055994254, "全帯域_40ms_envelope_correlation": 0.9981620392842686, "全帯域_40ms_modulation_mae_db": 0.660041497788674, "300–1000 Hz_10ms_log_rms_mae_db": 1.5386335105773592, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1061852595638154, "300–1000 Hz_10ms_envelope_correlation": 0.9900681799146609, "300–1000 Hz_10ms_modulation_mae_db": 0.6447596490271245, "300–1000 Hz_40ms_log_rms_mae_db": 1.032652891686803, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06266969170227266, "300–1000 Hz_40ms_envelope_correlation": 0.9964639142658831, "300–1000 Hz_40ms_modulation_mae_db": 0.9058239668686525, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7606465906953295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16686582178917683, "1000–3000 Hz_10ms_envelope_correlation": 0.9798047516812615, "1000–3000 Hz_10ms_modulation_mae_db": 0.5979599350963495, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2629558855191438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10408416602265452, "1000–3000 Hz_40ms_envelope_correlation": 0.9928713655416808, "1000–3000 Hz_40ms_modulation_mae_db": 0.32189791746747265, "3000–6000 Hz_10ms_log_rms_mae_db": 1.895135468778246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19922795162665263, "3000–6000 Hz_10ms_envelope_correlation": 0.9708594993210496, "3000–6000 Hz_10ms_modulation_mae_db": 0.8308991233182953, "3000–6000 Hz_40ms_log_rms_mae_db": 1.479728242477752, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13057998702851673, "3000–6000 Hz_40ms_envelope_correlation": 0.9870531256463577, "3000–6000 Hz_40ms_modulation_mae_db": 0.8688573014034683, "mr_stft_db": 5.874631000499288, "band_mae_db": 0.24921403933427366, "rms_gain_error_db": -0.13058872356577717, "output_peak": 0.7148320124201968}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8711241288974689, "全帯域_10ms_rms_envelope_nrmse": 0.08443745980948103, "全帯域_10ms_envelope_correlation": 0.9927949573359521, "全帯域_10ms_modulation_mae_db": 0.31845444944276036, "全帯域_40ms_log_rms_mae_db": 0.4980206867919754, "全帯域_40ms_rms_envelope_nrmse": 0.052387289905687334, "全帯域_40ms_envelope_correlation": 0.9972703469933994, "全帯域_40ms_modulation_mae_db": 0.3038335175846547, "300–1000 Hz_10ms_log_rms_mae_db": 1.5459220706651635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1189632344386843, "300–1000 Hz_10ms_envelope_correlation": 0.987849217374799, "300–1000 Hz_10ms_modulation_mae_db": 0.2832326207477424, "300–1000 Hz_40ms_log_rms_mae_db": 1.0429273371474925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07530817365779909, "300–1000 Hz_40ms_envelope_correlation": 0.9952862994296191, "300–1000 Hz_40ms_modulation_mae_db": 0.31533756484197223, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7830505550029447, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17370906005906733, "1000–3000 Hz_10ms_envelope_correlation": 0.977215302902401, "1000–3000 Hz_10ms_modulation_mae_db": 0.6745494205747757, "1000–3000 Hz_40ms_log_rms_mae_db": 1.247237714584435, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11420994343961982, "1000–3000 Hz_40ms_envelope_correlation": 0.9897737601480815, "1000–3000 Hz_40ms_modulation_mae_db": 0.516109224579213, "3000–6000 Hz_10ms_log_rms_mae_db": 1.883357439994661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18988365507149974, "3000–6000 Hz_10ms_envelope_correlation": 0.9737624446574344, "3000–6000 Hz_10ms_modulation_mae_db": 0.5970012101712672, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4518826890501133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12106806193599283, "3000–6000 Hz_40ms_envelope_correlation": 0.9893140634838286, "3000–6000 Hz_40ms_modulation_mae_db": 0.3992063408071828, "mr_stft_db": 5.868966778743084, "band_mae_db": 0.2230490628160922, "rms_gain_error_db": -0.1756154442334447, "output_peak": 0.7169349935777957}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8331459297773955, "spectral_convergence": 0.2440358739642662, "spectral_convergence_status": "ok", "amplitude_floor": 1.7851227833840954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.680648559417003, "spectral_convergence": 0.24543981422405914, "spectral_convergence_status": "ok", "amplitude_floor": 1.2048994709482952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.095825703027617, "spectral_convergence": 0.26343397057927265, "spectral_convergence_status": "ok", "amplitude_floor": 9.468117719284587e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005960090393190935, "output_power": 0.000595216424596988, "delta_db": -0.005779397121916293, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030564977634421338, "output_power": 0.003034532402361471, "delta_db": -0.031323031418333966, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008630746056010172, "output_power": 0.0008191662754674622, "delta_db": -0.2267627408964326, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003160541845765363, "output_power": 0.00029770317397166245, "delta_db": -0.25978079871523846, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001143536135409032, "output_power": 0.0001012188613021041, "delta_db": -0.5298844549016256, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.420642909107181e-05, "output_power": 3.5119651073569814e-05, "delta_db": 0.11442453099465384, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19465915900803346, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08563522301591411, "pitch": null, "output_rms_dbfs": -23.113171189434752, "output_peak": 0.7320522824075959, "reference_rms_dbfs": -23.02753596641884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.869873397407339}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023539858364162597, "log_rms_mae_db": 0.832702367238592, "rms_envelope_nrmse": 0.07236631472236497, "envelope_correlation": 0.9945672619672976, "rms_gain_error_db": -0.08565677197169502, "reference_mean_power": 0.004983368761309487, "output_mean_power": 0.004886043715115596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0209735395978665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009552928696150586, "output_power": 0.0009241314910243009, "reference_active": true, "delta_db": -0.14402766482597057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014624552237472608, "output_power": 0.00014952189353304563, "reference_active": true, "delta_db": 0.09622210189378345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.510798757145359e-05, "output_power": 3.267648231196389e-05, "reference_active": true, "delta_db": 1.14423392714259}], "active_band_count": 3, "active_band_mae_db": 0.46149456462078137}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002083747898533542, "log_rms_mae_db": 0.46155320167268865, "rms_envelope_nrmse": 0.03996357516872767, "envelope_correlation": 0.99827789074844, "rms_gain_error_db": -0.08572504616139552, "reference_mean_power": 0.004999205734592124, "output_mean_power": 0.004901494337180963}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8665599913970187e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008900977663526511, "output_power": 0.0008692634877687745, "reference_active": true, "delta_db": -0.1028627314282779}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.407197015766335e-05, "output_power": 7.653290440371173e-05, "reference_active": true, "delta_db": 0.14194298884798315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8378799869222006e-06, "output_power": 2.337692084434685e-06, "reference_active": true, "delta_db": 1.0447015763549348}], "active_band_count": 3, "active_band_mae_db": 0.42983576554373193}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019850808969518194, "log_rms_mae_db": 1.503535097929047, "rms_envelope_nrmse": 0.10111875855724442, "envelope_correlation": 0.9909997534726843, "rms_gain_error_db": -0.15626095585416735, "reference_mean_power": 0.003121339390070273, "output_mean_power": 0.0030110287164503893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3811041144776226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007780170156378303, "output_power": 0.0007234942501148437, "reference_active": true, "delta_db": -0.3155401137570133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001204320773242115, "output_power": 0.00011949148762420162, "reference_active": true, "delta_db": -0.03405209565880773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.313510568503667e-05, "output_power": 2.9915799338237528e-05, "reference_active": true, "delta_db": 1.1162912373823517}], "active_band_count": 3, "active_band_mae_db": 0.4886278155993909}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016710142967257498, "log_rms_mae_db": 1.0196121553399988, "rms_envelope_nrmse": 0.05696601777719017, "envelope_correlation": 0.997105709107865, "rms_gain_error_db": -0.15667764115720878, "reference_mean_power": 0.003133268891618282, "output_mean_power": 0.0030222466353254512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2683362614683906e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007382464262949856, "output_power": 0.0006934063809108311, "reference_active": true, "delta_db": -0.2721351947581342}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8108189032543396e-05, "output_power": 5.8810799495407216e-05, "reference_active": true, "delta_db": 0.05219742729643742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.916170233531451e-06, "output_power": 2.37309174435269e-06, "reference_active": true, "delta_db": 0.9288043908346955}], "active_band_count": 3, "active_band_mae_db": 0.41771233762975574}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012483338724103395, "log_rms_mae_db": 1.796065577341627, "rms_envelope_nrmse": 0.17182887101058392, "envelope_correlation": 0.9779177409598574, "rms_gain_error_db": -0.2845106232760809, "reference_mean_power": 0.00037365748369267646, "output_mean_power": 0.00034996339622398023}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.54909855359075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.931499876625785e-05, "output_power": 8.510685951172958e-05, "reference_active": true, "delta_db": -0.20959831449962174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4977670974468328e-05, "output_power": 1.2099030640362176e-05, "reference_active": true, "delta_db": -0.9269370932455413}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3989606403687574e-06, "output_power": 4.079643725956076e-06, "reference_active": true, "delta_db": 0.7927610241641159}], "active_band_count": 3, "active_band_mae_db": 0.6430988106364263}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010596065945345535, "log_rms_mae_db": 1.226472430665496, "rms_envelope_nrmse": 0.09359826439090384, "envelope_correlation": 0.9937238079220914, "rms_gain_error_db": -0.2840948922761263, "reference_mean_power": 0.00037540546219302315, "output_mean_power": 0.00035163419200875343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4355698091451528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.577569295489126e-05, "output_power": 8.182756329757703e-05, "reference_active": true, "delta_db": -0.20464616542468067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.83374089374301e-06, "output_power": 6.582934704058555e-06, "reference_active": true, "delta_db": -0.7554965550799231}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527528900320414e-07, "output_power": 2.4166753184382106e-07, "reference_active": true, "delta_db": -0.19477823468252162}], "active_band_count": 3, "active_band_mae_db": 0.3849736517290418}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.3292043050207156e-05, "log_rms_mae_db": 1.9152653596188547, "rms_envelope_nrmse": 0.1945829972569677, "envelope_correlation": 0.9732877266673845, "rms_gain_error_db": -0.45843649306622947, "reference_mean_power": 5.712588023845408e-05, "output_mean_power": 5.140309345409918e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.279319333258023e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4391078150208969e-05, "output_power": 1.19620584828228e-05, "reference_active": true, "delta_db": -0.8028741025324531}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.344791688657969e-06, "output_power": 3.3861393686001717e-06, "reference_active": true, "delta_db": -1.0826412993077825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.199993664659886e-07, "output_power": 6.802143876075305e-07, "reference_active": true, "delta_db": -0.811677031392351}], "active_band_count": 3, "active_band_mae_db": 0.8990641444108621}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.870517765122232e-05, "log_rms_mae_db": 1.4794213728744177, "rms_envelope_nrmse": 0.12930728350793502, "envelope_correlation": 0.9890564316555318, "rms_gain_error_db": -0.4584446827864992, "reference_mean_power": 5.74115094616415e-05, "output_mean_power": 5.166001134309519e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.057417179151539e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3970684782124775e-05, "output_power": 1.145645842120699e-05, "reference_active": true, "delta_db": -0.8616731059515136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.241188995835654e-06, "output_power": 1.8354397016654161e-06, "reference_active": true, "delta_db": -0.8673835990817698}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.11366674713116e-08, "output_power": 4.3867852387825164e-08, "reference_active": true, "delta_db": -0.6658604861764345}], "active_band_count": 3, "active_band_mae_db": 0.7983057304032393}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.841683053706987, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8752f64b0574900421301e2e6e4b7ec6cc5ed35733d5332f9fa1714f5b17a499", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9782531408082552, "standard_deviation": 0.0010589105265457048, "minimum": 0.976866163342058, "maximum": 0.9798047516812615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7713576009525722, "standard_deviation": 0.025751334837730854, "minimum": 1.7218907321831338, "maximum": 1.796065577341627, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8775038998711586, "standard_deviation": 0.37933230819544494, "minimum": 0.5979599350963495, "maximum": 1.4777342606535069, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17315296802139857, "standard_deviation": 0.003560310211238064, "minimum": 0.16686582178917683, "maximum": 0.178813930657738, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930464514720481, "standard_deviation": 0.0013880792853674497, "minimum": 0.9897737601480815, "maximum": 0.9939976959937105, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.245195429925104, "standard_deviation": 0.038452725275617126, "minimum": 1.1843962861533428, "maximum": 1.2900944971307884, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7692625777381311, "standard_deviation": 0.34234643400545967, "minimum": 0.32189791746747265, "maximum": 1.2648181331173045, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1036907009391507, "standard_deviation": 0.005960893883163974, "minimum": 0.09359826439090384, "maximum": 0.11420994343961982, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9723324603417233, "standard_deviation": 0.0021135985653043922, "minimum": 0.9700741094150087, "maximum": 0.9761059246147744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9027641663808286, "standard_deviation": 0.014703390086425674, "minimum": 1.882328922328677, "maximum": 1.9224180195905554, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6600215926216403, "standard_deviation": 0.166351473869031, "minimum": 0.4064546778417004, "maximum": 0.8990641444108621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19567780571829294, "standard_deviation": 0.007320282011364263, "minimum": 0.18079057578033697, "maximum": 0.2030474327952631, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9886730412585321, "standard_deviation": 0.0018651729306974477, "minimum": 0.9855538650847286, "maximum": 0.9908150467512884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4764938924359339, "standard_deviation": 0.026938988075418963, "minimum": 1.4276030620763456, "maximum": 1.5107982374289737, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6172050360352084, "standard_deviation": 0.18962275383077418, "minimum": 0.3952712015811419, "maximum": 0.8688573014034683, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12631413696844518, "standard_deviation": 0.008980543691453674, "minimum": 0.112976837686651, "maximum": 0.14111399159503168, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9889325738501927, "standard_deviation": 0.001485305354062827, "minimum": 0.9862559842502392, "maximum": 0.9909997534726843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5543416211107575, "standard_deviation": 0.049201002675209105, "minimum": 1.4821628243819818, "maximum": 1.6344673988109084, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5052711894300975, "standard_deviation": 0.11848772538358428, "minimum": 0.2832326207477424, "maximum": 0.6447596490271245, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11232281375919459, "standard_deviation": 0.007321572748384749, "minimum": 0.10111875855724442, "maximum": 0.12392847923961142, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961830769526099, "standard_deviation": 0.0007808756839025872, "minimum": 0.9947670746803375, "maximum": 0.997105709107865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0446861672456895, "standard_deviation": 0.04526114108249372, "minimum": 1.0031896637366362, "maximum": 1.136766674828244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.567276479393326, "standard_deviation": 0.21510468511603748, "minimum": 0.31533756484197223, "maximum": 0.9058239668686525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06596702990445935, "standard_deviation": 0.006090870575826981, "minimum": 0.05696601777719017, "maximum": 0.07530817365779909, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.244597498606154, "standard_deviation": 0.03147355491443035, "minimum": 0.19465915900803346, "maximum": 0.30331580255133755, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.8714410612830035, "standard_deviation": 0.007615292637862672, "minimum": 5.861381523847254, "maximum": 5.884514096473459, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7106393016256312, "standard_deviation": 0.03138091475683674, "minimum": 0.6671881753249833, "maximum": 0.7706740667728899, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1324471599013078, "standard_deviation": 0.04115561611693162, "minimum": -0.1756154442334447, "maximum": -0.06554880657071555, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933007874525895, "standard_deviation": 0.0008141500629810088, "minimum": 0.9919038912383024, "maximum": 0.9945672619672976, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8707143290874274, "standard_deviation": 0.021623744146094057, "minimum": 0.832702367238592, "maximum": 0.8999663086593753, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4274916918879211, "standard_deviation": 0.13111009400433166, "minimum": 0.26690734238740665, "maximum": 0.6900866796822333, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08072203050295639, "standard_deviation": 0.004796612190835895, "minimum": 0.07236631472236497, "maximum": 0.08785768411153154, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978229791209023, "standard_deviation": 0.0005375270003699054, "minimum": 0.9968656687294398, "maximum": 0.998321840077741, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.482567845055903, "standard_deviation": 0.03263364024286532, "minimum": 0.4384253305057612, "maximum": 0.5285968802692279, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.41838084445215107, "standard_deviation": 0.18792142238460466, "minimum": 0.20015141081647128, "maximum": 0.7399459378342387, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04588555684978932, "standard_deviation": 0.005126165846406038, "minimum": 0.03996357516872767, "maximum": 0.052742984864431706, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0028", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6660878658294678, "predicted_mos": 2.6660878658294678, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3d015b587d79ceee0ba42f995fe69c3c93fc19127edaa120156cebf5ecce6e32", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8549648001790047, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8240120839364242, "全帯域_10ms_rms_envelope_nrmse": 0.08020586955049584, "全帯域_10ms_envelope_correlation": 0.993360783813788, "全帯域_10ms_modulation_mae_db": 0.08911794548345435, "全帯域_40ms_log_rms_mae_db": 0.44258785959399277, "全帯域_40ms_rms_envelope_nrmse": 0.04220144307288695, "全帯域_40ms_envelope_correlation": 0.9981859116509135, "全帯域_40ms_modulation_mae_db": 0.07580232320512637, "300–1000 Hz_10ms_log_rms_mae_db": 1.5834234903435243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1112794927709207, "300–1000 Hz_10ms_envelope_correlation": 0.989167194240986, "300–1000 Hz_10ms_modulation_mae_db": 0.21517904001608576, "300–1000 Hz_40ms_log_rms_mae_db": 1.0603852354665342, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06041428066738489, "300–1000 Hz_40ms_envelope_correlation": 0.9969213168042865, "300–1000 Hz_40ms_modulation_mae_db": 0.28202450468495227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7311624044089415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17131486277167296, "1000–3000 Hz_10ms_envelope_correlation": 0.9787865761525695, "1000–3000 Hz_10ms_modulation_mae_db": 0.7316647998989749, "1000–3000 Hz_40ms_log_rms_mae_db": 1.18867016831262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09873836796473542, "1000–3000 Hz_40ms_envelope_correlation": 0.9939761754663891, "1000–3000 Hz_40ms_modulation_mae_db": 0.5000357152106333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.829246428876345, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18533534241079305, "3000–6000 Hz_10ms_envelope_correlation": 0.9748597406218399, "3000–6000 Hz_10ms_modulation_mae_db": 0.44587159429068207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3820081767594523, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11644179648755655, "3000–6000 Hz_40ms_envelope_correlation": 0.989777888826616, "3000–6000 Hz_40ms_modulation_mae_db": 0.36927295921511866, "mr_stft_db": 5.852032126181626, "band_mae_db": 0.28392703369653877, "rms_gain_error_db": -0.12488591061287876, "output_peak": 0.7013184253552694}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9014500669612779, "全帯域_10ms_rms_envelope_nrmse": 0.07863323767722838, "全帯域_10ms_envelope_correlation": 0.9936203947751435, "全帯域_10ms_modulation_mae_db": 0.4515794216406758, "全帯域_40ms_log_rms_mae_db": 0.5139709645181584, "全帯域_40ms_rms_envelope_nrmse": 0.04097503468227428, "全帯域_40ms_envelope_correlation": 0.9982714626642573, "全帯域_40ms_modulation_mae_db": 0.21749072931661362, "300–1000 Hz_10ms_log_rms_mae_db": 1.5384338571821585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1020626345895769, "300–1000 Hz_10ms_envelope_correlation": 0.9909023445941614, "300–1000 Hz_10ms_modulation_mae_db": 0.4689796884117879, "300–1000 Hz_40ms_log_rms_mae_db": 1.0124891932333653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05563857621746073, "300–1000 Hz_40ms_envelope_correlation": 0.9973628210520697, "300–1000 Hz_40ms_modulation_mae_db": 0.28572253935355846, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7872183259344443, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16952439074511563, "1000–3000 Hz_10ms_envelope_correlation": 0.9794665562092175, "1000–3000 Hz_10ms_modulation_mae_db": 0.7071526405772256, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2454936284537617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10490641711414865, "1000–3000 Hz_40ms_envelope_correlation": 0.9932334352013698, "1000–3000 Hz_40ms_modulation_mae_db": 0.4850956429254632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8690861375600805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1914682810799704, "3000–6000 Hz_10ms_envelope_correlation": 0.973147134004274, "3000–6000 Hz_10ms_modulation_mae_db": 0.47629656493388445, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4375971973584605, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1164604060654615, "3000–6000 Hz_40ms_envelope_correlation": 0.9895643923235171, "3000–6000 Hz_40ms_modulation_mae_db": 0.3916149938900746, "mr_stft_db": 5.868535857472712, "band_mae_db": 0.2446462685614571, "rms_gain_error_db": -0.11951988034897705, "output_peak": 0.7009556307587759}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8623456190610347, "全帯域_10ms_rms_envelope_nrmse": 0.08668984589435834, "全帯域_10ms_envelope_correlation": 0.9922403997282321, "全帯域_10ms_modulation_mae_db": 0.5156863737094065, "全帯域_40ms_log_rms_mae_db": 0.46148239629429066, "全帯域_40ms_rms_envelope_nrmse": 0.049487073861019505, "全帯域_40ms_envelope_correlation": 0.9974352898946105, "全帯域_40ms_modulation_mae_db": 0.31003968260185805, "300–1000 Hz_10ms_log_rms_mae_db": 1.4980645553788052, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11117895291956828, "300–1000 Hz_10ms_envelope_correlation": 0.9891381732758373, "300–1000 Hz_10ms_modulation_mae_db": 0.6366610449740681, "300–1000 Hz_40ms_log_rms_mae_db": 0.9371087651377417, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06092428178096754, "300–1000 Hz_40ms_envelope_correlation": 0.9967513769410381, "300–1000 Hz_40ms_modulation_mae_db": 0.3323612510672748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.729901734419129, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18271926816262432, "1000–3000 Hz_10ms_envelope_correlation": 0.9771225833261167, "1000–3000 Hz_10ms_modulation_mae_db": 1.174634288029509, "1000–3000 Hz_40ms_log_rms_mae_db": 1.211557319653389, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11562690697413096, "1000–3000 Hz_40ms_envelope_correlation": 0.9930605498790787, "1000–3000 Hz_40ms_modulation_mae_db": 1.304821350530753, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8691538659633125, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1990224177067798, "3000–6000 Hz_10ms_envelope_correlation": 0.9709161814183063, "3000–6000 Hz_10ms_modulation_mae_db": 0.24335437659398163, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4369140919384995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1159693557831444, "3000–6000 Hz_40ms_envelope_correlation": 0.989841602480393, "3000–6000 Hz_40ms_modulation_mae_db": 0.3213553266387073, "mr_stft_db": 5.84603385396401, "band_mae_db": 0.3434894254215802, "rms_gain_error_db": -0.13476024845752113, "output_peak": 0.6817296330622383}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8651453906317641, "全帯域_10ms_rms_envelope_nrmse": 0.0828854964501173, "全帯域_10ms_envelope_correlation": 0.9927892045226294, "全帯域_10ms_modulation_mae_db": 0.2943238688831683, "全帯域_40ms_log_rms_mae_db": 0.4549972679718379, "全帯域_40ms_rms_envelope_nrmse": 0.04948188119997126, "全帯域_40ms_envelope_correlation": 0.9972324211670374, "全帯域_40ms_modulation_mae_db": 0.18956845926152263, "300–1000 Hz_10ms_log_rms_mae_db": 1.5192569423879299, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11967257364618684, "300–1000 Hz_10ms_envelope_correlation": 0.9872016594738564, "300–1000 Hz_10ms_modulation_mae_db": 0.5586579741444366, "300–1000 Hz_40ms_log_rms_mae_db": 1.0321760475645438, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07519912394564604, "300–1000 Hz_40ms_envelope_correlation": 0.9945371507344298, "300–1000 Hz_40ms_modulation_mae_db": 0.3751143860493931, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6999925138793963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16838721516242197, "1000–3000 Hz_10ms_envelope_correlation": 0.9798042067007049, "1000–3000 Hz_10ms_modulation_mae_db": 0.72213706848959, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1807896700511382, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09603649559969192, "1000–3000 Hz_40ms_envelope_correlation": 0.9947058482662607, "1000–3000 Hz_40ms_modulation_mae_db": 1.089184224773821, "3000–6000 Hz_10ms_log_rms_mae_db": 1.92968125565986, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19439830865095664, "3000–6000 Hz_10ms_envelope_correlation": 0.9723373967153663, "3000–6000 Hz_10ms_modulation_mae_db": 0.3790794682383665, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4928396737992338, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11964598072073532, "3000–6000 Hz_40ms_envelope_correlation": 0.9893916128597627, "3000–6000 Hz_40ms_modulation_mae_db": 0.3290039910354272, "mr_stft_db": 5.841678220272484, "band_mae_db": 0.23755898233597064, "rms_gain_error_db": -0.05195493940920782, "output_peak": 0.7555409354041872}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8610000723705342, "全帯域_10ms_rms_envelope_nrmse": 0.0891623163457951, "全帯域_10ms_envelope_correlation": 0.9917065385571975, "全帯域_10ms_modulation_mae_db": 0.7778379697344543, "全帯域_40ms_log_rms_mae_db": 0.44821173422237254, "全帯域_40ms_rms_envelope_nrmse": 0.04712180137648422, "全帯域_40ms_envelope_correlation": 0.9975675571520243, "全帯域_40ms_modulation_mae_db": 0.5483307593851213, "300–1000 Hz_10ms_log_rms_mae_db": 1.4649425917534231, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.124923008328383, "300–1000 Hz_10ms_envelope_correlation": 0.9860690262701693, "300–1000 Hz_10ms_modulation_mae_db": 1.0242184246024462, "300–1000 Hz_40ms_log_rms_mae_db": 0.9406456318372479, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06912335946034609, "300–1000 Hz_40ms_envelope_correlation": 0.9954807741363564, "300–1000 Hz_40ms_modulation_mae_db": 0.5596287113817042, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7023473763817183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16757466599459744, "1000–3000 Hz_10ms_envelope_correlation": 0.9796580542768628, "1000–3000 Hz_10ms_modulation_mae_db": 0.8220525869459241, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1924858847752595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0983498399949575, "1000–3000 Hz_40ms_envelope_correlation": 0.9938376783598187, "1000–3000 Hz_40ms_modulation_mae_db": 0.49113471627487854, "3000–6000 Hz_10ms_log_rms_mae_db": 1.875022263101586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19348451542366077, "3000–6000 Hz_10ms_envelope_correlation": 0.9734114302412926, "3000–6000 Hz_10ms_modulation_mae_db": 0.5852170778553935, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4395419585725633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1201295388121177, "3000–6000 Hz_40ms_envelope_correlation": 0.9906712721561898, "3000–6000 Hz_40ms_modulation_mae_db": 0.6186631334297655, "mr_stft_db": 5.8463597281605715, "band_mae_db": 0.28288541945764273, "rms_gain_error_db": -0.09133324552698069, "output_peak": 0.7251917512408528}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8361559453224712, "全帯域_10ms_rms_envelope_nrmse": 0.07367341406008651, "全帯域_10ms_envelope_correlation": 0.994366712446182, "全帯域_10ms_modulation_mae_db": 0.21195323660466606, "全帯域_40ms_log_rms_mae_db": 0.44483246343558824, "全帯域_40ms_rms_envelope_nrmse": 0.03887535390871137, "全帯域_40ms_envelope_correlation": 0.9984036922653005, "全帯域_40ms_modulation_mae_db": 0.255343816313911, "300–1000 Hz_10ms_log_rms_mae_db": 1.4794683291001933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10032940262841351, "300–1000 Hz_10ms_envelope_correlation": 0.9912014521530386, "300–1000 Hz_10ms_modulation_mae_db": 0.3740471594887711, "300–1000 Hz_40ms_log_rms_mae_db": 0.9796935525247035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06108950446731482, "300–1000 Hz_40ms_envelope_correlation": 0.9967146652640946, "300–1000 Hz_40ms_modulation_mae_db": 0.26584513495054685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7199144494393124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17850377184855507, "1000–3000 Hz_10ms_envelope_correlation": 0.9763926139402396, "1000–3000 Hz_10ms_modulation_mae_db": 0.5239867760893443, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1563501836100603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09714444001729754, "1000–3000 Hz_40ms_envelope_correlation": 0.9936709014369619, "1000–3000 Hz_40ms_modulation_mae_db": 0.4112607070728725, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8827826135619385, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19997295469332757, "3000–6000 Hz_10ms_envelope_correlation": 0.9710353346593866, "3000–6000 Hz_10ms_modulation_mae_db": 0.6354867841571333, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4536177578112994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12693769953523065, "3000–6000 Hz_40ms_envelope_correlation": 0.9887253449387075, "3000–6000 Hz_40ms_modulation_mae_db": 0.9086131601785093, "mr_stft_db": 5.847367921955331, "band_mae_db": 0.20350618868993717, "rms_gain_error_db": -0.09750438804379478, "output_peak": 0.7193929698584592}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.879639872501324, "全帯域_10ms_rms_envelope_nrmse": 0.07936160592886594, "全帯域_10ms_envelope_correlation": 0.9933916275172859, "全帯域_10ms_modulation_mae_db": 0.3526452768839785, "全帯域_40ms_log_rms_mae_db": 0.5082307363944304, "全帯域_40ms_rms_envelope_nrmse": 0.04381403784534649, "全帯域_40ms_envelope_correlation": 0.9978447989245391, "全帯域_40ms_modulation_mae_db": 0.23292257264458197, "300–1000 Hz_10ms_log_rms_mae_db": 1.4066771678402923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10422227416518758, "300–1000 Hz_10ms_envelope_correlation": 0.9903058014094517, "300–1000 Hz_10ms_modulation_mae_db": 0.5819532216953429, "300–1000 Hz_40ms_log_rms_mae_db": 0.9228913045133943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05902863152304499, "300–1000 Hz_40ms_envelope_correlation": 0.996700736500782, "300–1000 Hz_40ms_modulation_mae_db": 0.4120304866570856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7232647526088727, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17406147314664067, "1000–3000 Hz_10ms_envelope_correlation": 0.9782454948035111, "1000–3000 Hz_10ms_modulation_mae_db": 0.32311805130908855, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2330310604917383, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1110843015297041, "1000–3000 Hz_40ms_envelope_correlation": 0.9921086225697259, "1000–3000 Hz_40ms_modulation_mae_db": 0.15372363893248261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.895830701689707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1864865054939841, "3000–6000 Hz_10ms_envelope_correlation": 0.9745136721467815, "3000–6000 Hz_10ms_modulation_mae_db": 0.2627212058859661, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4250246345121236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11292513780514112, "3000–6000 Hz_40ms_envelope_correlation": 0.990303956372351, "3000–6000 Hz_40ms_modulation_mae_db": 0.341827215039524, "mr_stft_db": 5.8479539557626055, "band_mae_db": 0.1955715029630979, "rms_gain_error_db": -0.056905738199739465, "output_peak": 0.7387334546926021}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8332818270508724, "全帯域_10ms_rms_envelope_nrmse": 0.06963015247297605, "全帯域_10ms_envelope_correlation": 0.9949739243120769, "全帯域_10ms_modulation_mae_db": 0.5072530947358359, "全帯域_40ms_log_rms_mae_db": 0.4540230341022828, "全帯域_40ms_rms_envelope_nrmse": 0.04144679414359545, "全帯域_40ms_envelope_correlation": 0.998143082120027, "全帯域_40ms_modulation_mae_db": 0.5600562771861206, "300–1000 Hz_10ms_log_rms_mae_db": 1.4615467784414307, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09752820364332389, "300–1000 Hz_10ms_envelope_correlation": 0.9915841085624548, "300–1000 Hz_10ms_modulation_mae_db": 1.0231195451803448, "300–1000 Hz_40ms_log_rms_mae_db": 0.9671377821117761, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06252973441799917, "300–1000 Hz_40ms_envelope_correlation": 0.996371248760814, "300–1000 Hz_40ms_modulation_mae_db": 1.2025403215637405, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7572034508917098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16183983532588303, "1000–3000 Hz_10ms_envelope_correlation": 0.9809513789715274, "1000–3000 Hz_10ms_modulation_mae_db": 0.793570645805668, "1000–3000 Hz_40ms_log_rms_mae_db": 1.220103291242991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09382200459737089, "1000–3000 Hz_40ms_envelope_correlation": 0.9942961249047224, "1000–3000 Hz_40ms_modulation_mae_db": 0.3235754000572329, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8487378094640712, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2072024487764845, "3000–6000 Hz_10ms_envelope_correlation": 0.969076271413588, "3000–6000 Hz_10ms_modulation_mae_db": 0.7994847441953383, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4386592187310852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1394760740753584, "3000–6000 Hz_40ms_envelope_correlation": 0.9863809211857365, "3000–6000 Hz_40ms_modulation_mae_db": 0.7872872822110137, "mr_stft_db": 5.851610991833454, "band_mae_db": 0.23253915539838754, "rms_gain_error_db": -0.08631111300224767, "output_peak": 0.6863762685034106}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.810685637647587, "spectral_convergence": 0.26128497028461184, "spectral_convergence_status": "ok", "amplitude_floor": 1.7851227833840954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.653638569113901, "spectral_convergence": 0.24802628475890426, "spectral_convergence_status": "ok", "amplitude_floor": 1.2048994709482952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.09177217178339, "spectral_convergence": 0.27264504223940794, "spectral_convergence_status": "ok", "amplitude_floor": 9.468117719284587e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005960090393190935, "output_power": 0.0006023373575676869, "delta_db": 0.045869529061579956, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030564977634421338, "output_power": 0.0029705642114145276, "delta_db": -0.12385137856196757, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008630746056010172, "output_power": 0.0008410513761836129, "delta_db": -0.11225812697537238, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003160541845765363, "output_power": 0.0002920708085846253, "delta_db": -0.342733917989396, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001143536135409032, "output_power": 9.61348522306039e-05, "delta_db": -0.7536902965493366, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.420642909107181e-05, "output_power": 3.686579761577698e-05, "delta_db": 0.32515895304157993, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.28392703369653877, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12488591061287876, "pitch": null, "output_rms_dbfs": -23.15242187703172, "output_peak": 0.7013184253552694, "reference_rms_dbfs": -23.02753596641884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.852032126181626}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023539858364162597, "log_rms_mae_db": 0.8240120839364242, "rms_envelope_nrmse": 0.08020586955049584, "envelope_correlation": 0.993360783813788, "rms_gain_error_db": -0.124493594670769, "reference_mean_power": 0.004983368761309487, "output_mean_power": 0.004842545009362741}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0209735395978665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009552928696150586, "output_power": 0.0009331569477147001, "reference_active": true, "delta_db": -0.1018184224702034}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014624552237472608, "output_power": 0.0001441276540086295, "reference_active": true, "delta_db": -0.06335260484504279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.510798757145359e-05, "output_power": 2.5705744459755932e-05, "reference_active": true, "delta_db": 0.10218280913511686}], "active_band_count": 3, "active_band_mae_db": 0.08911794548345435}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002083747898533542, "log_rms_mae_db": 0.44258785959399277, "rms_envelope_nrmse": 0.04220144307288695, "envelope_correlation": 0.9981859116509135, "rms_gain_error_db": -0.12413609449114235, "reference_mean_power": 0.004999205734592124, "output_mean_power": 0.004858334358916974}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8665599913970187e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008900977663526511, "output_power": 0.0008732270239712433, "reference_active": true, "delta_db": -0.08310543738624275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.407197015766335e-05, "output_power": 7.340686756235276e-05, "reference_active": true, "delta_db": -0.03917203983824059}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8378799869222006e-06, "output_power": 1.8829123236407784e-06, "reference_active": true, "delta_db": 0.10512949239089575}], "active_band_count": 3, "active_band_mae_db": 0.07580232320512637}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019850808969518194, "log_rms_mae_db": 1.5834234903435243, "rms_envelope_nrmse": 0.1112794927709207, "envelope_correlation": 0.989167194240986, "rms_gain_error_db": -0.20178960062869258, "reference_mean_power": 0.003121339390070273, "output_mean_power": 0.0029796279053909227}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3811041144776226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007780170156378303, "output_power": 0.000735527344181225, "reference_active": true, "delta_db": -0.2439027251707128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001204320773242115, "output_power": 0.00010979964691787328, "reference_active": true, "delta_db": -0.40141233974012863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.313510568503667e-05, "output_power": 2.313392281535865e-05, "reference_active": true, "delta_db": -0.00022205513741575168}], "active_band_count": 3, "active_band_mae_db": 0.21517904001608576}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016710142967257498, "log_rms_mae_db": 1.0603852354665342, "rms_envelope_nrmse": 0.06041428066738489, "envelope_correlation": 0.9969213168042865, "rms_gain_error_db": -0.20158343580107313, "reference_mean_power": 0.003133268891618282, "output_mean_power": 0.0029911577878366886}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2683362614683906e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007382464262949856, "output_power": 0.0007002493363386143, "reference_active": true, "delta_db": -0.2294864741531043}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8108189032543396e-05, "output_power": 5.3964859596739086e-05, "reference_active": true, "delta_db": -0.3212628919879728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.916170233531451e-06, "output_power": 1.7902005021277173e-06, "reference_active": true, "delta_db": -0.2953241479137796}], "active_band_count": 3, "active_band_mae_db": 0.28202450468495227}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012483338724103395, "log_rms_mae_db": 1.7311624044089415, "rms_envelope_nrmse": 0.17131486277167296, "envelope_correlation": 0.9787865761525695, "rms_gain_error_db": -0.4430882195448531, "reference_mean_power": 0.00037365748369267646, "output_mean_power": 0.0003374153708635912}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.54909855359075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.931499876625785e-05, "output_power": 7.393483995335913e-05, "reference_active": true, "delta_db": -0.8207525946955818}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4977670974468328e-05, "output_power": 1.3062007278599118e-05, "reference_active": true, "delta_db": -0.5943436445985544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3989606403687574e-06, "output_power": 4.067578579114189e-06, "reference_active": true, "delta_db": 0.7798981604027884}], "active_band_count": 3, "active_band_mae_db": 0.7316647998989749}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010596065945345535, "log_rms_mae_db": 1.18867016831262, "rms_envelope_nrmse": 0.09873836796473542, "envelope_correlation": 0.9939761754663891, "rms_gain_error_db": -0.44271071341853535, "reference_mean_power": 0.00037540546219302315, "output_mean_power": 0.00033902327588516214}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4355698091451528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.577569295489126e-05, "output_power": 7.141442202398719e-05, "reference_active": true, "delta_db": -0.7957830958809348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.83374089374301e-06, "output_power": 7.072601874949582e-06, "reference_active": true, "delta_db": -0.44389991041858806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527528900320414e-07, "output_power": 2.68372828880482e-07, "reference_active": true, "delta_db": 0.2604241393323772}], "active_band_count": 3, "active_band_mae_db": 0.5000357152106333}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.3292043050207156e-05, "log_rms_mae_db": 1.829246428876345, "rms_envelope_nrmse": 0.18533534241079305, "envelope_correlation": 0.9748597406218399, "rms_gain_error_db": -0.1863716653505897, "reference_mean_power": 5.712588023845408e-05, "output_mean_power": 5.472625610834957e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.279319333258023e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4391078150208969e-05, "output_power": 1.3571082569345156e-05, "reference_active": true, "delta_db": -0.2547883876857675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.344791688657969e-06, "output_power": 3.615337691417592e-06, "reference_active": true, "delta_db": -0.7982009010703632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.199993664659886e-07, "output_power": 8.755401495237422e-07, "reference_active": true, "delta_db": 0.2846254941159154}], "active_band_count": 3, "active_band_mae_db": 0.44587159429068207}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.870517765122232e-05, "log_rms_mae_db": 1.3820081767594523, "rms_envelope_nrmse": 0.11644179648755655, "envelope_correlation": 0.989777888826616, "rms_gain_error_db": -0.18637637428126308, "reference_mean_power": 5.74115094616415e-05, "output_mean_power": 5.4999827583593695e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.057417179151539e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3970684782124775e-05, "output_power": 1.3145232195516084e-05, "reference_active": true, "delta_db": -0.2644943203379134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.241188995835654e-06, "output_power": 1.948113587079216e-06, "reference_active": true, "delta_db": -0.6086420609209321}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.11366674713116e-08, "output_power": 5.397599683567081e-08, "reference_active": true, "delta_db": 0.23468249638651048}], "active_band_count": 3, "active_band_mae_db": 0.36927295921511866}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8534421397402135, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "031dc126afe347246618611e6b82e10f31ce90fe4046cb7fe89b713400c3c176", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9788034330475937, "standard_deviation": 0.001499248057413035, "minimum": 0.9763926139402396, "maximum": 0.9809513789715274, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7313756259954407, "standard_deviation": 0.028815633980884336, "minimum": 1.6999925138793963, "maximum": 1.7872183259344443, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7247896071431655, "standard_deviation": 0.24453405780706547, "minimum": 0.32311805130908855, "maximum": 1.174634288029509, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1717406853946889, "standard_deviation": 0.006583271798061868, "minimum": 0.16183983532588303, "maximum": 0.18271926816262432, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.993611167010541, "standard_deviation": 0.0008070317285373303, "minimum": 0.9921086225697259, "maximum": 0.9947058482662607, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2035601508238698, "standard_deviation": 0.02941381602885172, "minimum": 1.1563501836100603, "maximum": 1.2454936284537617, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5948539244722671, "standard_deviation": 0.39317492931434683, "minimum": 0.15372363893248261, "maximum": 1.304821350530753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10196359672400462, "standard_deviation": 0.0078069122732633365, "minimum": 0.09382200459737089, "maximum": 0.11562690697413096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9724121451526044, "standard_deviation": 0.001971695215106936, "minimum": 0.969076271413588, "maximum": 0.9748597406218399, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8749426344846125, "standard_deviation": 0.030087358690870796, "minimum": 1.829246428876345, "maximum": 1.92968125565986, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.47843897701884325, "standard_deviation": 0.18968862592567906, "minimum": 0.24335437659398163, "maximum": 0.7994847441953383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1946713467794946, "standard_deviation": 0.007269925670173694, "minimum": 0.18533534241079305, "maximum": 0.2072024487764845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9893321238929091, "standard_deviation": 0.0013272590943903106, "minimum": 0.9863809211857365, "maximum": 0.9906712721561898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4382753386853397, "standard_deviation": 0.030608812273431752, "minimum": 1.3820081767594523, "maximum": 1.4928396737992338, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5084547577047676, "standard_deviation": 0.23238417269992726, "minimum": 0.3213553266387073, "maximum": 0.9086131601785093, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1209982486605932, "standard_deviation": 0.008544911170761355, "minimum": 0.11292513780514112, "maximum": 0.1394760740753584, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9894462199974945, "standard_deviation": 0.001968568154505137, "minimum": 0.9860690262701693, "maximum": 0.9915841085624548, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4939767140534697, "standard_deviation": 0.05393031050704705, "minimum": 1.4066771678402923, "maximum": 1.5834234903435243, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6103520123141604, "standard_deviation": 0.28730799028431875, "minimum": 0.21517904001608576, "maximum": 1.0242184246024462, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1088995678364451, "standard_deviation": 0.009673559885741425, "minimum": 0.09752820364332389, "maximum": 0.124923008328383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9963550112742339, "standard_deviation": 0.0009111638036502488, "minimum": 0.9945371507344298, "maximum": 0.9973628210520697, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9815659390486633, "standard_deviation": 0.049331841871915263, "minimum": 0.9228913045133943, "maximum": 1.0603852354665342, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.46440841696353197, "standard_deviation": 0.31320437186744965, "minimum": 0.26584513495054685, "maximum": 1.2025403215637405, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06299343656002053, "standard_deviation": 0.006223708502489077, "minimum": 0.05563857621746073, "maximum": 0.07519912394564604, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2530154970655765, "standard_deviation": 0.04856877237257725, "minimum": 0.1955715029630979, "maximum": 0.3434894254215802, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.85019658195035, "standard_deviation": 0.008100632525393264, "minimum": 5.841678220272484, "maximum": 5.868535857472712, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7136548836094745, "standard_deviation": 0.025707710630131057, "minimum": 0.6817296330622383, "maximum": 0.7555409354041872, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09539693295016843, "standard_deviation": 0.030409756247504055, "minimum": -0.13476024845752113, "maximum": -0.05195493940920782, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993306198209067, "standard_deviation": 0.001067514999175295, "minimum": 0.9917065385571975, "maximum": 0.9949739243120769, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8578788597294629, "standard_deviation": 0.025852177815603377, "minimum": 0.8240120839364242, "maximum": 0.9014500669612779, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.40004964845945495, "standard_deviation": 0.21237691000255635, "minimum": 0.08911794548345435, "maximum": 0.7778379697344543, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08003024229749042, "standard_deviation": 0.006401989063670884, "minimum": 0.06963015247297605, "maximum": 0.0891623163457951, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978855269798387, "standard_deviation": 0.00043194434792623326, "minimum": 0.9972324211670374, "maximum": 0.9984036922653005, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4660420570666192, "standard_deviation": 0.028491016810455832, "minimum": 0.44258785959399277, "maximum": 0.5139709645181584, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2986943274893569, "standard_deviation": 0.1711914757401469, "minimum": 0.07580232320512637, "maximum": 0.5600562771861206, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04417542751128619, "standard_deviation": 0.004050733764393514, "minimum": 0.03887535390871137, "maximum": 0.049487073861019505, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7729-102255-0038", "language": "en", "style": "normal", "speaker": "librispeech_7729", "source_utterance_id": "7729-102255-0038", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7729-102255-0038.wav", "source_file_sha256": "81c19f33a420f81dd0cf46240ae9e3b6c2fcd454e82e980097221c2a20cb93fb", "decoded_pcm_sha256": "4488be03955742e6e44fcdb33f0ae71643965b6765824a166c40ca31fbd9f66a", "original_source_sha256": "7bad471a79798adfc54e63acefeaf2a9579bc1f06779fd3aa49cc060df7c47f7", "corpus_entry_id": "librispeech_7729-102255-0038"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.430001258850098, "predicted_mos": 4.430001258850098, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "681e09ee06b146f3f30633f5c64fa9efb0061deaf08f51d8cfda52461d584029", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.175717279998935, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7729-102255-0038", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "681e09ee06b146f3f30633f5c64fa9efb0061deaf08f51d8cfda52461d584029", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "9e25b2715b44c695b1208e22e4cec275bbd491d99983858945470dba564f7b1d"}, "rows": [{"case": "librispeech_7729-102255-0038", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.520419120788574, "predicted_mos": 3.520419120788574, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c9a2bebecde2b9e2d517569f92e865c3d5f23d1b172938565ded73032072eade", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0409031179988233, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7729-102255-0038", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c9a2bebecde2b9e2d517569f92e865c3d5f23d1b172938565ded73032072eade", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8954954358291931, "全帯域_10ms_rms_envelope_nrmse": 0.07742873178447383, "全帯域_10ms_envelope_correlation": 0.9940153017003215, "全帯域_10ms_modulation_mae_db": 0.2811113956674324, "全帯域_40ms_log_rms_mae_db": 0.5262895240794215, "全帯域_40ms_rms_envelope_nrmse": 0.043160368676388006, "全帯域_40ms_envelope_correlation": 0.9980396916730307, "全帯域_40ms_modulation_mae_db": 0.2727481370918798, "300–1000 Hz_10ms_log_rms_mae_db": 1.443254738728556, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09926406989776053, "300–1000 Hz_10ms_envelope_correlation": 0.9919123742531167, "300–1000 Hz_10ms_modulation_mae_db": 0.48876014243212973, "300–1000 Hz_40ms_log_rms_mae_db": 0.9760436132411416, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05703400267910744, "300–1000 Hz_40ms_envelope_correlation": 0.9971971470088099, "300–1000 Hz_40ms_modulation_mae_db": 0.27070575519791634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8270786682073783, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19459728580538796, "1000–3000 Hz_10ms_envelope_correlation": 0.9756835840940548, "1000–3000 Hz_10ms_modulation_mae_db": 1.1270754893959065, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3137650708508386, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12479447072068275, "1000–3000 Hz_40ms_envelope_correlation": 0.9926161247072947, "1000–3000 Hz_40ms_modulation_mae_db": 1.4654722995484892, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9304687669600047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22534494316683884, "3000–6000 Hz_10ms_envelope_correlation": 0.9659202594533381, "3000–6000 Hz_10ms_modulation_mae_db": 1.2332401529186237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.426050321499835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1351432419265408, "3000–6000 Hz_40ms_envelope_correlation": 0.9871394034480774, "3000–6000 Hz_40ms_modulation_mae_db": 0.7969470584734196, "mr_stft_db": 5.777514283752712, "band_mae_db": 0.23756275408516989, "rms_gain_error_db": -0.07170162222058957, "output_peak": 0.5532742873628058}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.867862963910355, "全帯域_10ms_rms_envelope_nrmse": 0.07352925208275032, "全帯域_10ms_envelope_correlation": 0.9947120387388355, "全帯域_10ms_modulation_mae_db": 0.3216989443143383, "全帯域_40ms_log_rms_mae_db": 0.5296721230230933, "全帯域_40ms_rms_envelope_nrmse": 0.03972623633361613, "全帯域_40ms_envelope_correlation": 0.9984871570754821, "全帯域_40ms_modulation_mae_db": 0.4268307116659505, "300–1000 Hz_10ms_log_rms_mae_db": 1.447479491423445, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09901866243065424, "300–1000 Hz_10ms_envelope_correlation": 0.9921553614488778, "300–1000 Hz_10ms_modulation_mae_db": 0.3244104360640393, "300–1000 Hz_40ms_log_rms_mae_db": 0.9593274569508368, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05995294839821521, "300–1000 Hz_40ms_envelope_correlation": 0.9971822467807815, "300–1000 Hz_40ms_modulation_mae_db": 0.6276250886366607, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8115912378887116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17948515103568075, "1000–3000 Hz_10ms_envelope_correlation": 0.9786665034809484, "1000–3000 Hz_10ms_modulation_mae_db": 0.7774174700166866, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2838161725246013, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10688064552586841, "1000–3000 Hz_40ms_envelope_correlation": 0.9941216146425791, "1000–3000 Hz_40ms_modulation_mae_db": 0.4928857365740775, "3000–6000 Hz_10ms_log_rms_mae_db": 1.918200400521027, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2132677026309289, "3000–6000 Hz_10ms_envelope_correlation": 0.9694551683565116, "3000–6000 Hz_10ms_modulation_mae_db": 0.9068743648367352, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3863065426345298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12372877161833412, "3000–6000 Hz_40ms_envelope_correlation": 0.9893998688353, "3000–6000 Hz_40ms_modulation_mae_db": 0.43159808458825455, "mr_stft_db": 5.794795144302238, "band_mae_db": 0.25745770922438055, "rms_gain_error_db": -0.12355307341891607, "output_peak": 0.5755199588900837}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8943544427876161, "全帯域_10ms_rms_envelope_nrmse": 0.08017272327216284, "全帯域_10ms_envelope_correlation": 0.9937170407833795, "全帯域_10ms_modulation_mae_db": 0.25641377139551996, "全帯域_40ms_log_rms_mae_db": 0.5336472970824518, "全帯域_40ms_rms_envelope_nrmse": 0.04840347053058611, "全帯域_40ms_envelope_correlation": 0.9977076738837694, "全帯域_40ms_modulation_mae_db": 0.5202909438144596, "300–1000 Hz_10ms_log_rms_mae_db": 1.4358635506513788, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1029698702142228, "300–1000 Hz_10ms_envelope_correlation": 0.991461489979667, "300–1000 Hz_10ms_modulation_mae_db": 0.2584237251313766, "300–1000 Hz_40ms_log_rms_mae_db": 0.9958375654672817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06459662941698256, "300–1000 Hz_40ms_envelope_correlation": 0.9966310701886343, "300–1000 Hz_40ms_modulation_mae_db": 0.41830269248128166, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8071358048036086, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18032986485875918, "1000–3000 Hz_10ms_envelope_correlation": 0.9793674680939124, "1000–3000 Hz_10ms_modulation_mae_db": 1.3113718912287762, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3112288449287062, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1246142218091813, "1000–3000 Hz_40ms_envelope_correlation": 0.9920540115239825, "1000–3000 Hz_40ms_modulation_mae_db": 2.055499709885011, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9166287927507102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2199625061016357, "3000–6000 Hz_10ms_envelope_correlation": 0.9676975940532564, "3000–6000 Hz_10ms_modulation_mae_db": 1.0931262896564762, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4202425673569705, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1378875682291895, "3000–6000 Hz_40ms_envelope_correlation": 0.9874384400884495, "3000–6000 Hz_40ms_modulation_mae_db": 0.6987576162041912, "mr_stft_db": 5.792212751967768, "band_mae_db": 0.25371087516376506, "rms_gain_error_db": -0.1408380316904104, "output_peak": 0.6059750426636918}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8789491374747759, "全帯域_10ms_rms_envelope_nrmse": 0.07106643505059013, "全帯域_10ms_envelope_correlation": 0.9950676548606873, "全帯域_10ms_modulation_mae_db": 0.3667240379026031, "全帯域_40ms_log_rms_mae_db": 0.5007714069016497, "全帯域_40ms_rms_envelope_nrmse": 0.040425833793592506, "全帯域_40ms_envelope_correlation": 0.9984195160852035, "全帯域_40ms_modulation_mae_db": 0.6614125535649283, "300–1000 Hz_10ms_log_rms_mae_db": 1.515157647520286, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09257053326597389, "300–1000 Hz_10ms_envelope_correlation": 0.9932745344674231, "300–1000 Hz_10ms_modulation_mae_db": 0.44182058513488065, "300–1000 Hz_40ms_log_rms_mae_db": 1.0067084296095343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05958235360512092, "300–1000 Hz_40ms_envelope_correlation": 0.9973166510344209, "300–1000 Hz_40ms_modulation_mae_db": 0.4724895872497275, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8570780238147162, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17587765580725537, "1000–3000 Hz_10ms_envelope_correlation": 0.9805883862763028, "1000–3000 Hz_10ms_modulation_mae_db": 0.6525115033976788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3186322425572254, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11689815449341039, "1000–3000 Hz_40ms_envelope_correlation": 0.9935674787366457, "1000–3000 Hz_40ms_modulation_mae_db": 0.8951988394820446, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9637177425240635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20001395259974553, "3000–6000 Hz_10ms_envelope_correlation": 0.9733654240910216, "3000–6000 Hz_10ms_modulation_mae_db": 0.5752663601652409, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4219298203634783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13420724289495273, "3000–6000 Hz_40ms_envelope_correlation": 0.987859786600403, "3000–6000 Hz_40ms_modulation_mae_db": 0.6942614409647335, "mr_stft_db": 5.774576493745298, "band_mae_db": 0.29342134229402195, "rms_gain_error_db": -0.1220635148114154, "output_peak": 0.531329525190005}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8803712665017224, "全帯域_10ms_rms_envelope_nrmse": 0.08007796061280846, "全帯域_10ms_envelope_correlation": 0.9939312620887044, "全帯域_10ms_modulation_mae_db": 0.4063782867689228, "全帯域_40ms_log_rms_mae_db": 0.5067489702938899, "全帯域_40ms_rms_envelope_nrmse": 0.04700218902350171, "全帯域_40ms_envelope_correlation": 0.9980981512443762, "全帯域_40ms_modulation_mae_db": 0.5032878802516239, "300–1000 Hz_10ms_log_rms_mae_db": 1.4744482568717574, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11468630972276415, "300–1000 Hz_10ms_envelope_correlation": 0.9897943367367862, "300–1000 Hz_10ms_modulation_mae_db": 0.5546995140482689, "300–1000 Hz_40ms_log_rms_mae_db": 1.0247434523821182, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07515432809684518, "300–1000 Hz_40ms_envelope_correlation": 0.99590492329251, "300–1000 Hz_40ms_modulation_mae_db": 0.5698451248925238, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8540843526177384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1937786182177767, "1000–3000 Hz_10ms_envelope_correlation": 0.9764085121425324, "1000–3000 Hz_10ms_modulation_mae_db": 1.0350133073623116, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3224088910278646, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12672838258301378, "1000–3000 Hz_40ms_envelope_correlation": 0.9928421491004847, "1000–3000 Hz_40ms_modulation_mae_db": 1.6304368885765388, "3000–6000 Hz_10ms_log_rms_mae_db": 1.936541072220316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18421492266379424, "3000–6000 Hz_10ms_envelope_correlation": 0.9773767088401359, "3000–6000 Hz_10ms_modulation_mae_db": 1.10560013624803, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3746096292490317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11321909668347466, "3000–6000 Hz_40ms_envelope_correlation": 0.9913393401321237, "3000–6000 Hz_40ms_modulation_mae_db": 0.5808726768729847, "mr_stft_db": 5.78719436961519, "band_mae_db": 0.2754054430762969, "rms_gain_error_db": -0.19451853022532262, "output_peak": 0.5000314075070275}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8858310292624134, "全帯域_10ms_rms_envelope_nrmse": 0.07814645866275721, "全帯域_10ms_envelope_correlation": 0.9941065434967602, "全帯域_10ms_modulation_mae_db": 0.0849673387475608, "全帯域_40ms_log_rms_mae_db": 0.5089497156144759, "全帯域_40ms_rms_envelope_nrmse": 0.04454420308113068, "全帯域_40ms_envelope_correlation": 0.9981760769608549, "全帯域_40ms_modulation_mae_db": 0.3328141630932502, "300–1000 Hz_10ms_log_rms_mae_db": 1.4753718372321707, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10423509573841228, "300–1000 Hz_10ms_envelope_correlation": 0.9913715036514791, "300–1000 Hz_10ms_modulation_mae_db": 0.19933985658892964, "300–1000 Hz_40ms_log_rms_mae_db": 0.9897362266255665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0646517076792016, "300–1000 Hz_40ms_envelope_correlation": 0.9967900486898977, "300–1000 Hz_40ms_modulation_mae_db": 0.4858769790092022, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8490852267972844, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17909346692233924, "1000–3000 Hz_10ms_envelope_correlation": 0.9799817282693313, "1000–3000 Hz_10ms_modulation_mae_db": 0.5588329603066579, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2918741864996588, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11738198491095221, "1000–3000 Hz_40ms_envelope_correlation": 0.9938232289328921, "1000–3000 Hz_40ms_modulation_mae_db": 0.9578435136687843, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9032368974967562, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20348211870707572, "3000–6000 Hz_10ms_envelope_correlation": 0.9723284224246104, "3000–6000 Hz_10ms_modulation_mae_db": 0.5234693460014807, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4121219591332592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1175359849101301, "3000–6000 Hz_40ms_envelope_correlation": 0.9907763659036305, "3000–6000 Hz_40ms_modulation_mae_db": 0.7718634098703049, "mr_stft_db": 5.783354184495113, "band_mae_db": 0.25599199936443334, "rms_gain_error_db": -0.1593565548012254, "output_peak": 0.5240934675955937}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8801152620469676, "全帯域_10ms_rms_envelope_nrmse": 0.07466265016039965, "全帯域_10ms_envelope_correlation": 0.9944989558877247, "全帯域_10ms_modulation_mae_db": 0.19394584284959904, "全帯域_40ms_log_rms_mae_db": 0.5335166570015949, "全帯域_40ms_rms_envelope_nrmse": 0.04340058350555131, "全帯域_40ms_envelope_correlation": 0.9980833762139495, "全帯域_40ms_modulation_mae_db": 0.34379963420606047, "300–1000 Hz_10ms_log_rms_mae_db": 1.515792321245599, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10318039183342077, "300–1000 Hz_10ms_envelope_correlation": 0.991416325178062, "300–1000 Hz_10ms_modulation_mae_db": 0.29383726355392387, "300–1000 Hz_40ms_log_rms_mae_db": 1.084229889944227, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06510643990919542, "300–1000 Hz_40ms_envelope_correlation": 0.9965519312883928, "300–1000 Hz_40ms_modulation_mae_db": 0.3792926547597057, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8508027510614797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17771392685204046, "1000–3000 Hz_10ms_envelope_correlation": 0.9790380747674972, "1000–3000 Hz_10ms_modulation_mae_db": 1.0604238656781708, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3496299701991268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11141587689853111, "1000–3000 Hz_40ms_envelope_correlation": 0.993185844264852, "1000–3000 Hz_40ms_modulation_mae_db": 1.4042708452136914, "3000–6000 Hz_10ms_log_rms_mae_db": 1.969425331571645, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21885925414808707, "3000–6000 Hz_10ms_envelope_correlation": 0.9678039625481025, "3000–6000 Hz_10ms_modulation_mae_db": 0.5815853003345003, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4278703289916448, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15517594634008888, "3000–6000 Hz_40ms_envelope_correlation": 0.9830371659510709, "3000–6000 Hz_40ms_modulation_mae_db": 0.7598637790238761, "mr_stft_db": 5.77657263313665, "band_mae_db": 0.1968832918234211, "rms_gain_error_db": -0.09997924226513756, "output_peak": 0.5478805747871103}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.860163690935969, "全帯域_10ms_rms_envelope_nrmse": 0.06746441047727225, "全帯域_10ms_envelope_correlation": 0.9955879400912913, "全帯域_10ms_modulation_mae_db": 0.24880704564658132, "全帯域_40ms_log_rms_mae_db": 0.5043550607168803, "全帯域_40ms_rms_envelope_nrmse": 0.038495415209240706, "全帯域_40ms_envelope_correlation": 0.9985984482275599, "全帯域_40ms_modulation_mae_db": 0.3203681208727318, "300–1000 Hz_10ms_log_rms_mae_db": 1.4045205204897209, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0888537252823454, "300–1000 Hz_10ms_envelope_correlation": 0.9936594758366556, "300–1000 Hz_10ms_modulation_mae_db": 0.37273078060613035, "300–1000 Hz_40ms_log_rms_mae_db": 0.9810357973980285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05186261051665318, "300–1000 Hz_40ms_envelope_correlation": 0.997869184655986, "300–1000 Hz_40ms_modulation_mae_db": 0.5364309133537989, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8444695788132672, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18184425838041893, "1000–3000 Hz_10ms_envelope_correlation": 0.980237851441463, "1000–3000 Hz_10ms_modulation_mae_db": 0.9300782459350009, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3427656776133141, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1246471890699459, "1000–3000 Hz_40ms_envelope_correlation": 0.9936585565226206, "1000–3000 Hz_40ms_modulation_mae_db": 1.0571075038828113, "3000–6000 Hz_10ms_log_rms_mae_db": 1.949177026153072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21799937891969515, "3000–6000 Hz_10ms_envelope_correlation": 0.9690298533437781, "3000–6000 Hz_10ms_modulation_mae_db": 1.3351068727807303, "3000–6000 Hz_40ms_log_rms_mae_db": 1.447897964589873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1452278121230906, "3000–6000 Hz_40ms_envelope_correlation": 0.987059946868349, "3000–6000 Hz_40ms_modulation_mae_db": 1.8399549721679147, "mr_stft_db": 5.782637877222136, "band_mae_db": 0.306290637009218, "rms_gain_error_db": -0.12557907590370906, "output_peak": 0.5461346135950694}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.790758351985625, "spectral_convergence": 0.22322276057560483, "spectral_convergence_status": "ok", "amplitude_floor": 1.6152428194526273e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5717952876199615, "spectral_convergence": 0.21869408802998583, "spectral_convergence_status": "ok", "amplitude_floor": 1.0855275118818902e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.969989211652551, "spectral_convergence": 0.22539520473883337, "spectral_convergence_status": "ok", "amplitude_floor": 7.682277294566263e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006930713321042831, "output_power": 0.0006938583728270318, "delta_db": 0.004928980587804499, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002559270540127681, "output_power": 0.002524217537201354, "delta_db": -0.059894178158398106, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007234666478039212, "output_power": 0.0007256427026694927, "delta_db": 0.013043174836670649, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002266743638240336, "output_power": 0.00019155020540999326, "delta_db": -0.7311978349109766, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.48754354732022e-05, "output_power": 5.919930609048113e-05, "delta_db": -0.3976366991247863, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.432671244156477e-05, "output_power": 4.66157942333647e-05, "delta_db": 0.21867565689238316, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23756275408516989, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07170162222058957, "pitch": null, "output_rms_dbfs": -23.725231340596842, "output_peak": 0.5532742873628058, "reference_rms_dbfs": -23.65352971837625, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.777514283752712}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019088238002709922, "log_rms_mae_db": 0.8954954358291931, "rms_envelope_nrmse": 0.07742873178447383, "envelope_correlation": 0.9940153017003215, "rms_gain_error_db": -0.07170160288489347, "reference_mean_power": 0.004315280923443156, "output_mean_power": 0.00424462094854805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8761677888507434e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007125355736933842, "output_power": 0.0006985487252296143, "reference_active": true, "delta_db": -0.0860984713803187}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011942531791187452, "output_power": 0.00011090340718205121, "reference_active": true, "delta_db": -0.3215151727608679}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5392584995669358e-05, "output_power": 2.8072362874769632e-05, "reference_active": true, "delta_db": 0.4357205428611105}], "active_band_count": 3, "active_band_mae_db": 0.2811113956674324}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018007320980187037, "log_rms_mae_db": 0.5262895240794215, "rms_envelope_nrmse": 0.043160368676388006, "envelope_correlation": 0.9980396916730307, "rms_gain_error_db": -0.07170165400106189, "reference_mean_power": 0.004336982504671315, "output_mean_power": 0.004265967130004361}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7431428852385876e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006439047386291532, "output_power": 0.000628655125603781, "reference_active": true, "delta_db": -0.10409160337264363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.115354721039444e-05, "output_power": 4.742207023456157e-05, "reference_active": true, "delta_db": -0.3289524556651225}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9654159892814645e-06, "output_power": 1.7985994224872172e-06, "reference_active": true, "delta_db": -0.3852003522378733}], "active_band_count": 3, "active_band_mae_db": 0.2727481370918798}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017279669295308916, "log_rms_mae_db": 1.443254738728556, "rms_envelope_nrmse": 0.09926406989776053, "envelope_correlation": 0.9919123742531167, "rms_gain_error_db": -0.09339908313581641, "reference_mean_power": 0.0026069520032741206, "output_mean_power": 0.002551485632276885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2087875823984064e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005529275902426358, "output_power": 0.0005415971183594929, "reference_active": true, "delta_db": -0.08991916150182894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648535673845476e-05, "output_power": 8.44204862850918e-05, "reference_active": true, "delta_db": -0.5801355743582672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.705863634087677e-05, "output_power": 2.0491175837394054e-05, "reference_active": true, "delta_db": 0.796225691436293}], "active_band_count": 3, "active_band_mae_db": 0.48876014243212973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014230358058689408, "log_rms_mae_db": 0.9760436132411416, "rms_envelope_nrmse": 0.05703400267910744, "envelope_correlation": 0.9971971470088099, "rms_gain_error_db": -0.09339909735567226, "reference_mean_power": 0.0026200630509491097, "output_mean_power": 0.002564317716579293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1215946456987494e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005050917134712818, "output_power": 0.0004957499007345646, "reference_active": true, "delta_db": -0.08107607623317033}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.798176008445663e-05, "output_power": 3.237275746377132e-05, "reference_active": true, "delta_db": -0.6939539264238392}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3776930382412419e-06, "output_power": 1.389508443403488e-06, "reference_active": true, "delta_db": 0.037087262936739446}], "active_band_count": 3, "active_band_mae_db": 0.27070575519791634}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.058549471038879e-05, "log_rms_mae_db": 1.8270786682073783, "rms_envelope_nrmse": 0.19459728580538796, "envelope_correlation": 0.9756835840940548, "rms_gain_error_db": -0.7078214931950342, "reference_mean_power": 0.0002448429483315042, "output_mean_power": 0.00020802017202895627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7588735706755028e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.377351846027054e-05, "output_power": 8.042129922128388e-05, "reference_active": true, "delta_db": -0.6670912647030007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.692404819271496e-06, "output_power": 8.480136247314399e-06, "reference_active": true, "delta_db": -0.5802871490274226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.8112011913022686e-06, "output_power": 2.3317236095019386e-06, "reference_active": true, "delta_db": -2.1338480544572964}], "active_band_count": 3, "active_band_mae_db": 1.1270754893959065}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.399057463688353e-05, "log_rms_mae_db": 1.3137650708508386, "rms_envelope_nrmse": 0.12479447072068275, "envelope_correlation": 0.9926161247072947, "rms_gain_error_db": -0.7078266458401349, "reference_mean_power": 0.0002460738466011548, "output_mean_power": 0.0002090657032141054}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6710797473991227e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.068314304092372e-05, "output_power": 7.722355075400873e-05, "reference_active": true, "delta_db": -0.6977679749319298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.419207768284792e-06, "output_power": 3.932825494639112e-06, "reference_active": true, "delta_db": -0.5063974378156985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.686031483587781e-07, "output_power": 1.2879112898191801e-07, "reference_active": true, "delta_db": -3.192251485897839}], "active_band_count": 3, "active_band_mae_db": 1.4654722995484892}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.748604525835054e-05, "log_rms_mae_db": 1.9304687669600047, "rms_envelope_nrmse": 0.22534494316683884, "envelope_correlation": 0.9659202594533381, "rms_gain_error_db": -0.13638932644921095, "reference_mean_power": 6.169396290931914e-05, "output_mean_power": 5.978658335984603e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.220333375078007e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9572001812371347e-05, "output_power": 1.70295886417672e-05, "reference_active": true, "delta_db": -0.6043108981482693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.501529531199527e-06, "output_power": 7.112239339937008e-06, "reference_active": true, "delta_db": -0.7749070472169002}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.287934963899872e-07, "output_power": 1.584781588486281e-06, "reference_active": true, "delta_db": 2.3205025133907013}], "active_band_count": 3, "active_band_mae_db": 1.2332401529186237}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.636845752844561e-05, "log_rms_mae_db": 1.426050321499835, "rms_envelope_nrmse": 0.1351432419265408, "envelope_correlation": 0.9871394034480774, "rms_gain_error_db": -0.13638911590031752, "reference_mean_power": 6.200418008997283e-05, "output_mean_power": 6.008721253286268e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.99689742660592e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0058245625049685e-05, "output_power": 1.772253059071491e-05, "reference_active": true, "delta_db": -0.5376721054402374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.381270388794365e-06, "output_power": 3.831901029158667e-06, "reference_active": true, "delta_db": -0.5818577282294686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.024894795639882e-07, "output_power": 1.373442309783833e-07, "reference_active": true, "delta_db": 1.271311341750553}], "active_band_count": 3, "active_band_mae_db": 0.7969470584734196}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8643183546051689, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "60250edfe0bb447291fd478d371232416afdcf4dc23a21dfd323a7f961f41897", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9787465135707553, "standard_deviation": 0.0017915324957375443, "minimum": 0.9756835840940548, "maximum": 0.9805883862763028, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8376657055005232, "standard_deviation": 0.019711230507049295, "minimum": 1.8071358048036086, "maximum": 1.8570780238147162, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9315905916651487, "standard_deviation": 0.2536441487108512, "minimum": 0.5588329603066579, "maximum": 1.3113718912287762, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18284002848495734, "standard_deviation": 0.007223194802356179, "minimum": 0.17587765580725537, "maximum": 0.19459728580538796, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.993233626053919, "standard_deviation": 0.0006924608768974821, "minimum": 0.9920540115239825, "maximum": 0.9941216146425791, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.316765132025167, "standard_deviation": 0.02247636224916307, "minimum": 1.2838161725246013, "maximum": 1.3496299701991268, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.244839417103931, "standard_deviation": 0.4906621707149962, "minimum": 0.4928857365740775, "maximum": 2.055499709885011, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11917011575144823, "standard_deviation": 0.007249787264813875, "minimum": 0.10688064552586841, "maximum": 0.12672838258301378, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9703721741388444, "standard_deviation": 0.003742827760457108, "minimum": 0.9659202594533381, "maximum": 0.9773767088401359, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9359245037746995, "standard_deviation": 0.023468800132876562, "minimum": 1.9032368974967562, "maximum": 1.969425331571645, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9192836028677271, "standard_deviation": 0.32196356480183125, "minimum": 0.5234693460014807, "maximum": 1.3351068727807303, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21039309736722514, "standard_deviation": 0.013594699935773403, "minimum": 0.18421492266379424, "maximum": 0.22534494316683884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9880062897284255, "standard_deviation": 0.0026024102527089656, "minimum": 0.9830371659510709, "maximum": 0.9913393401321237, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4146286417273277, "standard_deviation": 0.023626300051920908, "minimum": 1.3746096292490317, "maximum": 1.447897964589873, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8217648797707099, "standard_deviation": 0.4285017064075045, "minimum": 0.43159808458825455, "maximum": 1.8399549721679147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13276570809072516, "standard_deviation": 0.014077173330236349, "minimum": 0.11321909668347466, "maximum": 0.15517594634008888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918806751940084, "standard_deviation": 0.0012063423920718054, "minimum": 0.9897943367367862, "maximum": 0.9936594758366556, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4639860455203642, "standard_deviation": 0.03886017787125692, "minimum": 1.4045205204897209, "maximum": 1.515792321245599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3667527879449599, "standard_deviation": 0.12115662877238696, "minimum": 0.19933985658892964, "maximum": 0.5546995140482689, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10059733229819426, "standard_deviation": 0.00784351721458084, "minimum": 0.0888537252823454, "maximum": 0.11468630972276415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969304003674291, "standard_deviation": 0.0005939246139144022, "minimum": 0.99590492329251, "maximum": 0.997869184655986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0022078039523419, "standard_deviation": 0.03858829716631083, "minimum": 0.9593274569508368, "maximum": 1.084229889944227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.47007109944760206, "standard_deviation": 0.11345524601865803, "minimum": 0.27070575519791634, "maximum": 0.6276250886366607, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.062242627537665185, "standard_deviation": 0.006901890412228945, "minimum": 0.05186261051665318, "maximum": 0.07515432809684518, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2595905065050883, "standard_deviation": 0.033901168649113855, "minimum": 0.1968832918234211, "maximum": 0.306290637009218, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.783607217279638, "standard_deviation": 0.007380083492842011, "minimum": 5.774576493745298, "maximum": 5.794795144302238, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5480298596989234, "standard_deviation": 0.03229367806356647, "minimum": 0.5000314075070275, "maximum": 0.6059750426636918, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12969870566709077, "standard_deviation": 0.036965672749954395, "minimum": -0.19451853022532262, "maximum": -0.07170162222058957, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994454592205963, "standard_deviation": 0.0006392563387307878, "minimum": 0.9937170407833795, "maximum": 0.9955879400912913, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8803929035936265, "standard_deviation": 0.012076936190195793, "minimum": 0.860163690935969, "maximum": 0.8954954358291931, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27000583291156965, "standard_deviation": 0.10099919926617541, "minimum": 0.0849673387475608, "maximum": 0.4063782867689228, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07531857776290184, "standard_deviation": 0.0045049925150006135, "minimum": 0.06746441047727225, "maximum": 0.08017272327216284, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982012614205283, "standard_deviation": 0.0002883512518166032, "minimum": 0.9977076738837694, "maximum": 0.9985984482275599, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5179938443391822, "standard_deviation": 0.014051237961622989, "minimum": 0.5007714069016497, "maximum": 0.5336472970824518, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4226940180701106, "standard_deviation": 0.13108407909991696, "minimum": 0.2727481370918798, "maximum": 0.6614125535649283, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.043144787519200894, "standard_deviation": 0.0034898464170210865, "minimum": 0.038495415209240706, "maximum": 0.04840347053058611, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0038", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.4715521335601807, "predicted_mos": 3.4715521335601807, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b4ab7c10dcbed0824545f76681354dc10ae7358d25616196f6b5f7fdb9189f6b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.968704499769956, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9140181198369538, "全帯域_10ms_rms_envelope_nrmse": 0.08263067984613716, "全帯域_10ms_envelope_correlation": 0.9931828914358801, "全帯域_10ms_modulation_mae_db": 0.2953997772573616, "全帯域_40ms_log_rms_mae_db": 0.5477040108995779, "全帯域_40ms_rms_envelope_nrmse": 0.047598647073269204, "全帯域_40ms_envelope_correlation": 0.9976021623118526, "全帯域_40ms_modulation_mae_db": 0.2952827579967725, "300–1000 Hz_10ms_log_rms_mae_db": 1.4738423340118345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11003608971408486, "300–1000 Hz_10ms_envelope_correlation": 0.9900362090734847, "300–1000 Hz_10ms_modulation_mae_db": 0.5331232007974588, "300–1000 Hz_40ms_log_rms_mae_db": 1.0048070557559512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0656556353350186, "300–1000 Hz_40ms_envelope_correlation": 0.9962309090089669, "300–1000 Hz_40ms_modulation_mae_db": 0.30165716984783675, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7841451417888508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19363410438334924, "1000–3000 Hz_10ms_envelope_correlation": 0.9768328201548843, "1000–3000 Hz_10ms_modulation_mae_db": 1.1994928616136376, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2788001401660083, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12788866788378783, "1000–3000 Hz_40ms_envelope_correlation": 0.9929600359650094, "1000–3000 Hz_40ms_modulation_mae_db": 1.4956056979447636, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8668667519116773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21805173214693824, "3000–6000 Hz_10ms_envelope_correlation": 0.9681535059745364, "3000–6000 Hz_10ms_modulation_mae_db": 1.0469662722391477, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3769792478927687, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1395332839905279, "3000–6000 Hz_40ms_envelope_correlation": 0.9866537097547061, "3000–6000 Hz_40ms_modulation_mae_db": 0.7189672201682523, "mr_stft_db": 5.715794316879106, "band_mae_db": 0.2608236463548351, "rms_gain_error_db": -0.07246006802900345, "output_peak": 0.5482518255072668}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8595878473347554, "全帯域_10ms_rms_envelope_nrmse": 0.07274018973884579, "全帯域_10ms_envelope_correlation": 0.9948556309000283, "全帯域_10ms_modulation_mae_db": 0.3649647387808484, "全帯域_40ms_log_rms_mae_db": 0.5357926767270328, "全帯域_40ms_rms_envelope_nrmse": 0.03896897070172852, "全帯域_40ms_envelope_correlation": 0.9985862726195357, "全帯域_40ms_modulation_mae_db": 0.5531896001792931, "300–1000 Hz_10ms_log_rms_mae_db": 1.4467637590504734, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09982511693474783, "300–1000 Hz_10ms_envelope_correlation": 0.9920571440816943, "300–1000 Hz_10ms_modulation_mae_db": 0.34534059549861734, "300–1000 Hz_40ms_log_rms_mae_db": 0.9360107455679969, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05931255117510527, "300–1000 Hz_40ms_envelope_correlation": 0.9972979622300862, "300–1000 Hz_40ms_modulation_mae_db": 0.706564315343689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7573718448436824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17183416710161323, "1000–3000 Hz_10ms_envelope_correlation": 0.9809663227370927, "1000–3000 Hz_10ms_modulation_mae_db": 0.5995176314761763, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2067685122976404, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10576345466011965, "1000–3000 Hz_40ms_envelope_correlation": 0.9945610429014691, "1000–3000 Hz_40ms_modulation_mae_db": 0.38921783537323557, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8946592670166897, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20332466685800427, "3000–6000 Hz_10ms_envelope_correlation": 0.9725552655322415, "3000–6000 Hz_10ms_modulation_mae_db": 0.9070569583787716, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3534156424971227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12246195668946704, "3000–6000 Hz_40ms_envelope_correlation": 0.9901752556709045, "3000–6000 Hz_40ms_modulation_mae_db": 0.3883357205160391, "mr_stft_db": 5.734023932874983, "band_mae_db": 0.28796396064036683, "rms_gain_error_db": -0.13336178697126025, "output_peak": 0.579889908785155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9052608412448565, "全帯域_10ms_rms_envelope_nrmse": 0.07877053106153743, "全帯域_10ms_envelope_correlation": 0.993935521755733, "全帯域_10ms_modulation_mae_db": 0.23900943503039054, "全帯域_40ms_log_rms_mae_db": 0.5433391622036902, "全帯域_40ms_rms_envelope_nrmse": 0.04822113962649116, "全帯域_40ms_envelope_correlation": 0.9977163419299222, "全帯域_40ms_modulation_mae_db": 0.4943475879522094, "300–1000 Hz_10ms_log_rms_mae_db": 1.4704563095425385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10092342009772315, "300–1000 Hz_10ms_envelope_correlation": 0.9917871803504279, "300–1000 Hz_10ms_modulation_mae_db": 0.30974836085288365, "300–1000 Hz_40ms_log_rms_mae_db": 0.9894860709269951, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06330144165611773, "300–1000 Hz_40ms_envelope_correlation": 0.9967472598862991, "300–1000 Hz_40ms_modulation_mae_db": 0.46789606880581586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8011188644016225, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17640841179681, "1000–3000 Hz_10ms_envelope_correlation": 0.9809523662835996, "1000–3000 Hz_10ms_modulation_mae_db": 1.365218904961262, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2568932675774787, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12241285524939867, "1000–3000 Hz_40ms_envelope_correlation": 0.9930524726836526, "1000–3000 Hz_40ms_modulation_mae_db": 2.0056328454429657, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8378104406944542, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19794615648716474, "3000–6000 Hz_10ms_envelope_correlation": 0.9746494572383092, "3000–6000 Hz_10ms_modulation_mae_db": 0.7583236573128639, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3314964887240555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12869227063336547, "3000–6000 Hz_40ms_envelope_correlation": 0.9899899225585518, "3000–6000 Hz_40ms_modulation_mae_db": 0.4094141286573363, "mr_stft_db": 5.727928362910493, "band_mae_db": 0.29357003933628784, "rms_gain_error_db": -0.13785787368834643, "output_peak": 0.5772346310046798}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8837981545291486, "全帯域_10ms_rms_envelope_nrmse": 0.07149551906919184, "全帯域_10ms_envelope_correlation": 0.9949801941751308, "全帯域_10ms_modulation_mae_db": 0.36611197261115525, "全帯域_40ms_log_rms_mae_db": 0.5020355379764684, "全帯域_40ms_rms_envelope_nrmse": 0.041103302764862945, "全帯域_40ms_envelope_correlation": 0.9983266138533603, "全帯域_40ms_modulation_mae_db": 0.7036804885140592, "300–1000 Hz_10ms_log_rms_mae_db": 1.4837962662883317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09353070055145432, "300–1000 Hz_10ms_envelope_correlation": 0.9930316530078962, "300–1000 Hz_10ms_modulation_mae_db": 0.4433175909845011, "300–1000 Hz_40ms_log_rms_mae_db": 0.9441183997111515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05909634286152146, "300–1000 Hz_40ms_envelope_correlation": 0.9972561466254338, "300–1000 Hz_40ms_modulation_mae_db": 0.5054833291446212, "1000–3000 Hz_10ms_log_rms_mae_db": 1.801458497887766, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17785513750759893, "1000–3000 Hz_10ms_envelope_correlation": 0.9809504698584044, "1000–3000 Hz_10ms_modulation_mae_db": 0.567029698618391, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2559630487754851, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12133353491219448, "1000–3000 Hz_40ms_envelope_correlation": 0.9937205778892092, "1000–3000 Hz_40ms_modulation_mae_db": 0.8011367425644379, "3000–6000 Hz_10ms_log_rms_mae_db": 1.891084396342335, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19991070581589304, "3000–6000 Hz_10ms_envelope_correlation": 0.973627405419758, "3000–6000 Hz_10ms_modulation_mae_db": 0.43434316508844933, "3000–6000 Hz_40ms_log_rms_mae_db": 1.356387611170619, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13470602071165028, "3000–6000 Hz_40ms_envelope_correlation": 0.9880922376898106, "3000–6000 Hz_40ms_modulation_mae_db": 0.7575496219440024, "mr_stft_db": 5.702050864705736, "band_mae_db": 0.2901032077254729, "rms_gain_error_db": -0.11170684432761742, "output_peak": 0.53093433387621}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8983249251435619, "全帯域_10ms_rms_envelope_nrmse": 0.08065809365875026, "全帯域_10ms_envelope_correlation": 0.99380210343351, "全帯域_10ms_modulation_mae_db": 0.3367858532144441, "全帯域_40ms_log_rms_mae_db": 0.5147580473227681, "全帯域_40ms_rms_envelope_nrmse": 0.0478809478012805, "全帯域_40ms_envelope_correlation": 0.9979626959535672, "全帯域_40ms_modulation_mae_db": 0.4419727017076999, "300–1000 Hz_10ms_log_rms_mae_db": 1.4658957307526472, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.115820826328923, "300–1000 Hz_10ms_envelope_correlation": 0.9894597926945198, "300–1000 Hz_10ms_modulation_mae_db": 0.535616205210276, "300–1000 Hz_40ms_log_rms_mae_db": 0.9953948721740943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07459447152602425, "300–1000 Hz_40ms_envelope_correlation": 0.9958437071552476, "300–1000 Hz_40ms_modulation_mae_db": 0.6146227860063015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7996049160507162, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18814162039239415, "1000–3000 Hz_10ms_envelope_correlation": 0.9786564144181232, "1000–3000 Hz_10ms_modulation_mae_db": 1.0758419069974114, "1000–3000 Hz_40ms_log_rms_mae_db": 1.25115221719215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1254140655426347, "1000–3000 Hz_40ms_envelope_correlation": 0.9937138229579144, "1000–3000 Hz_40ms_modulation_mae_db": 1.7643548593697698, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8771272477568162, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17563043297817996, "3000–6000 Hz_10ms_envelope_correlation": 0.9796794974143156, "3000–6000 Hz_10ms_modulation_mae_db": 0.9187326011782222, "3000–6000 Hz_40ms_log_rms_mae_db": 1.353192366529684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11067115061758663, "3000–6000 Hz_40ms_envelope_correlation": 0.9920291809959156, "3000–6000 Hz_40ms_modulation_mae_db": 0.39445395476923056, "mr_stft_db": 5.72555340266491, "band_mae_db": 0.2864328321216351, "rms_gain_error_db": -0.1864658513270363, "output_peak": 0.5072179261932854}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8982694867743647, "全帯域_10ms_rms_envelope_nrmse": 0.07626366395576031, "全帯域_10ms_envelope_correlation": 0.9943592697284491, "全帯域_10ms_modulation_mae_db": 0.11902459883457332, "全帯域_40ms_log_rms_mae_db": 0.5101742736682349, "全帯域_40ms_rms_envelope_nrmse": 0.04197800250021776, "全帯域_40ms_envelope_correlation": 0.9983680099704402, "全帯域_40ms_modulation_mae_db": 0.337601284541946, "300–1000 Hz_10ms_log_rms_mae_db": 1.4907511990443079, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10368017736846283, "300–1000 Hz_10ms_envelope_correlation": 0.9913755318118268, "300–1000 Hz_10ms_modulation_mae_db": 0.22367657454394366, "300–1000 Hz_40ms_log_rms_mae_db": 0.9717781584217203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061648739761534094, "300–1000 Hz_40ms_envelope_correlation": 0.9970158478060899, "300–1000 Hz_40ms_modulation_mae_db": 0.4842908120048777, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8314744125947637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18270032222804672, "1000–3000 Hz_10ms_envelope_correlation": 0.9797473028559245, "1000–3000 Hz_10ms_modulation_mae_db": 0.4509525183935457, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2346004540487385, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1208092313202013, "1000–3000 Hz_40ms_envelope_correlation": 0.9940364342881823, "1000–3000 Hz_40ms_modulation_mae_db": 0.9055037254850112, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8929094948529575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20554352017117639, "3000–6000 Hz_10ms_envelope_correlation": 0.9719128041576154, "3000–6000 Hz_10ms_modulation_mae_db": 0.23788730199916186, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4004943501952696, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12392124125431248, "3000–6000 Hz_40ms_envelope_correlation": 0.9898376459805587, "3000–6000 Hz_40ms_modulation_mae_db": 0.5629510999038, "mr_stft_db": 5.723915256460413, "band_mae_db": 0.2707396511348195, "rms_gain_error_db": -0.14649015721891978, "output_peak": 0.5089160528931875}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8978625105657926, "全帯域_10ms_rms_envelope_nrmse": 0.0738967479340095, "全帯域_10ms_envelope_correlation": 0.994610957203119, "全帯域_10ms_modulation_mae_db": 0.19451345214419477, "全帯域_40ms_log_rms_mae_db": 0.5495046977576918, "全帯域_40ms_rms_envelope_nrmse": 0.043453990054545456, "全帯域_40ms_envelope_correlation": 0.9980732972491851, "全帯域_40ms_modulation_mae_db": 0.49316948047500686, "300–1000 Hz_10ms_log_rms_mae_db": 1.5005697303574976, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10103460768756972, "300–1000 Hz_10ms_envelope_correlation": 0.9917659476981198, "300–1000 Hz_10ms_modulation_mae_db": 0.34311041314195734, "300–1000 Hz_40ms_log_rms_mae_db": 1.0064497823830134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06291391045585552, "300–1000 Hz_40ms_envelope_correlation": 0.9967786791493848, "300–1000 Hz_40ms_modulation_mae_db": 0.47329281932346784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8332627780764847, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17753763691264887, "1000–3000 Hz_10ms_envelope_correlation": 0.979762715633401, "1000–3000 Hz_10ms_modulation_mae_db": 1.0779634943290324, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2767019659001662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11221645331924643, "1000–3000 Hz_40ms_envelope_correlation": 0.9938076013865215, "1000–3000 Hz_40ms_modulation_mae_db": 1.4219977031194386, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8986844690496736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2215079106101165, "3000–6000 Hz_10ms_envelope_correlation": 0.9671229033432103, "3000–6000 Hz_10ms_modulation_mae_db": 0.4167236884076466, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3873780487551461, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16328710578880362, "3000–6000 Hz_40ms_envelope_correlation": 0.981381863756468, "3000–6000 Hz_40ms_modulation_mae_db": 0.8095787830966358, "mr_stft_db": 5.714300896014675, "band_mae_db": 0.24603012438983782, "rms_gain_error_db": -0.09735328540695756, "output_peak": 0.5288363035878926}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8563836040964287, "全帯域_10ms_rms_envelope_nrmse": 0.0700410938150047, "全帯域_10ms_envelope_correlation": 0.9952191853388436, "全帯域_10ms_modulation_mae_db": 0.2782991686213731, "全帯域_40ms_log_rms_mae_db": 0.5009974484072778, "全帯域_40ms_rms_envelope_nrmse": 0.039693753267890444, "全帯域_40ms_envelope_correlation": 0.9984817430444485, "全帯域_40ms_modulation_mae_db": 0.400979059460788, "300–1000 Hz_10ms_log_rms_mae_db": 1.4113190503560542, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09176750765781778, "300–1000 Hz_10ms_envelope_correlation": 0.9931686608098982, "300–1000 Hz_10ms_modulation_mae_db": 0.3854552980342764, "300–1000 Hz_40ms_log_rms_mae_db": 0.9687495144562883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05124613620075485, "300–1000 Hz_40ms_envelope_correlation": 0.997861037247035, "300–1000 Hz_40ms_modulation_mae_db": 0.6108429491972295, "1000–3000 Hz_10ms_log_rms_mae_db": 1.798715352744987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18575397544103317, "1000–3000 Hz_10ms_envelope_correlation": 0.9805076437676467, "1000–3000 Hz_10ms_modulation_mae_db": 0.9786552485755712, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2617692600071582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12975600832588743, "1000–3000 Hz_40ms_envelope_correlation": 0.9941705179994514, "1000–3000 Hz_40ms_modulation_mae_db": 1.1056084483972155, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8822834315542065, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21687636890990108, "3000–6000 Hz_10ms_envelope_correlation": 0.9700136026184283, "3000–6000 Hz_10ms_modulation_mae_db": 1.4629089266829893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4142224395261342, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1534264456816974, "3000–6000 Hz_40ms_envelope_correlation": 0.9860161251287621, "3000–6000 Hz_40ms_modulation_mae_db": 1.9357900457391397, "mr_stft_db": 5.717158562270545, "band_mae_db": 0.35533942859983797, "rms_gain_error_db": -0.12131984044356309, "output_peak": 0.5425303896230208}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.740675882206421, "spectral_convergence": 0.2255968672507007, "spectral_convergence_status": "ok", "amplitude_floor": 1.6152428194526273e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.511837952171307, "spectral_convergence": 0.21972505321442137, "spectral_convergence_status": "ok", "amplitude_floor": 1.0855275118818902e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.894869116259591, "spectral_convergence": 0.22441474175272466, "spectral_convergence_status": "ok", "amplitude_floor": 7.682277294566263e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006930713321042831, "output_power": 0.0006918822643691445, "delta_db": -0.007457371628630463, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002559270540127681, "output_power": 0.002534268817877594, "delta_db": -0.04263517547539064, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007234666478039212, "output_power": 0.0007232393065361603, "delta_db": -0.0013649360965712592, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002266743638240336, "output_power": 0.0001889360027233827, "delta_db": -0.7908768268222066, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.48754354732022e-05, "output_power": 5.645812059727169e-05, "delta_db": -0.6035386898090357, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.432671244156477e-05, "output_power": 4.555881403483577e-05, "delta_db": 0.1190688782971761, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2608236463548351, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07246006802900345, "pitch": null, "output_rms_dbfs": -23.725989786405254, "output_peak": 0.5482518255072668, "reference_rms_dbfs": -23.65352971837625, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.715794316879106}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019088238002709922, "log_rms_mae_db": 0.9140181198369538, "rms_envelope_nrmse": 0.08263067984613716, "envelope_correlation": 0.9931828914358801, "rms_gain_error_db": -0.07246005267913866, "reference_mean_power": 0.004315280923443156, "output_mean_power": 0.0042438797347117505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8761677888507434e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007125355736933842, "output_power": 0.0006995771222362405, "reference_active": true, "delta_db": -0.07970952969040199}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011942531791187452, "output_power": 0.00010936313440222911, "reference_active": true, "delta_db": -0.3822545721934718}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5392584995669358e-05, "output_power": 2.799822103596067e-05, "reference_active": true, "delta_db": 0.424235229888211}], "active_band_count": 3, "active_band_mae_db": 0.2953997772573616}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018007320980187037, "log_rms_mae_db": 0.5477040108995779, "rms_envelope_nrmse": 0.047598647073269204, "envelope_correlation": 0.9976021623118526, "rms_gain_error_db": -0.07246008940917524, "reference_mean_power": 0.004336982504671315, "output_mean_power": 0.004265222202735372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7431428852385876e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006439047386291532, "output_power": 0.0006296456720933803, "reference_active": true, "delta_db": -0.09725398610097871}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.115354721039444e-05, "output_power": 4.682996234081167e-05, "reference_active": true, "delta_db": -0.38351946432511175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9654159892814645e-06, "output_power": 1.7903873574784548e-06, "reference_active": true, "delta_db": -0.40507482356422714}], "active_band_count": 3, "active_band_mae_db": 0.2952827579967725}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017279669295308916, "log_rms_mae_db": 1.4738423340118345, "rms_envelope_nrmse": 0.11003608971408486, "envelope_correlation": 0.9900362090734847, "rms_gain_error_db": -0.07735245440866834, "reference_mean_power": 0.0026069520032741206, "output_mean_power": 0.002560930485184618}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2087875823984064e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005529275902426358, "output_power": 0.0005466127117403046, "reference_active": true, "delta_db": -0.04988533910881175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648535673845476e-05, "output_power": 8.21366516790906e-05, "reference_active": true, "delta_db": -0.6992441215985534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.705863634087677e-05, "output_power": 2.0747621909006653e-05, "reference_active": true, "delta_db": 0.8502401416850109}], "active_band_count": 3, "active_band_mae_db": 0.5331232007974588}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014230358058689408, "log_rms_mae_db": 1.0048070557559512, "rms_envelope_nrmse": 0.0656556353350186, "envelope_correlation": 0.9962309090089669, "rms_gain_error_db": -0.07735246692849049, "reference_mean_power": 0.0026200630509491097, "output_mean_power": 0.0025738100711142553}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1215946456987494e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005050917134712818, "output_power": 0.0004996735742783544, "reference_active": true, "delta_db": -0.04683861575645967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.798176008445663e-05, "output_power": 3.162929406649657e-05, "reference_active": true, "delta_db": -0.7948558716934937}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3776930382412419e-06, "output_power": 1.3577654759507767e-06, "reference_active": true, "delta_db": -0.06327702209355696}], "active_band_count": 3, "active_band_mae_db": 0.30165716984783675}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.058549471038879e-05, "log_rms_mae_db": 1.7841451417888508, "rms_envelope_nrmse": 0.19363410438334924, "envelope_correlation": 0.9768328201548843, "rms_gain_error_db": -0.7904985659148054, "reference_mean_power": 0.0002448429483315042, "output_mean_power": 0.0002040975276647124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7588735706755028e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.377351846027054e-05, "output_power": 8.02307993280431e-05, "reference_active": true, "delta_db": -0.6773909236507214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.692404819271496e-06, "output_power": 8.566399052532404e-06, "reference_active": true, "delta_db": -0.5363324339687339}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.8112011913022686e-06, "output_power": 2.2008292674583636e-06, "reference_active": true, "delta_db": -2.3847552272214574}], "active_band_count": 3, "active_band_mae_db": 1.1994928616136376}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.399057463688353e-05, "log_rms_mae_db": 1.2788001401660083, "rms_envelope_nrmse": 0.12788866788378783, "envelope_correlation": 0.9929600359650094, "rms_gain_error_db": -0.7905033264862548, "reference_mean_power": 0.0002460738466011548, "output_mean_power": 0.00020512336174637977}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6710797473991227e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.068314304092372e-05, "output_power": 7.717845096784016e-05, "reference_active": true, "delta_db": -0.7003050649445762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.419207768284792e-06, "output_power": 3.898152198871189e-06, "reference_active": true, "delta_db": -0.5448562868043354}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.686031483587781e-07, "output_power": 1.273343350193363e-07, "reference_active": true, "delta_db": -3.241655742085379}], "active_band_count": 3, "active_band_mae_db": 1.4956056979447636}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.748604525835054e-05, "log_rms_mae_db": 1.8668667519116773, "rms_envelope_nrmse": 0.21805173214693824, "envelope_correlation": 0.9681535059745364, "rms_gain_error_db": -0.2885878411691284, "reference_mean_power": 6.169396290931914e-05, "output_mean_power": 5.772765072912419e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.220333375078007e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9572001812371347e-05, "output_power": 1.577223848129642e-05, "reference_active": true, "delta_db": -0.9374191218119253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.501529531199527e-06, "output_power": 6.836636972360838e-06, "reference_active": true, "delta_db": -0.9465454840343743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.287934963899872e-07, "output_power": 1.2405451832390424e-06, "reference_active": true, "delta_db": 1.2569342108711432}], "active_band_count": 3, "active_band_mae_db": 1.0469662722391477}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.636845752844561e-05, "log_rms_mae_db": 1.3769792478927687, "rms_envelope_nrmse": 0.1395332839905279, "envelope_correlation": 0.9866537097547061, "rms_gain_error_db": -0.2885876127516645, "reference_mean_power": 6.200418008997283e-05, "output_mean_power": 5.801792706195953e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.99689742660592e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0058245625049685e-05, "output_power": 1.621108968772222e-05, "reference_active": true, "delta_db": -0.9248073667816341}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.381270388794365e-06, "output_power": 3.691782873743495e-06, "reference_active": true, "delta_db": -0.7436390558425331}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.024894795639882e-07, "output_power": 1.1468980427802412e-07, "reference_active": true, "delta_db": 0.4884552378805902}], "active_band_count": 3, "active_band_mae_db": 0.7189672201682523}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8644827063712937, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cc6928ff2cdb69bc58002fa81ee5e7d79823df5fdc28c77e055427841e3868aa", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9797970069636346, "standard_deviation": 0.001445615886485421, "minimum": 0.9768328201548843, "maximum": 0.9809663227370927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8008939760486091, "standard_deviation": 0.024390803690291517, "minimum": 1.7573718448436824, "maximum": 1.8332627780764847, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9143340331206284, "standard_deviation": 0.3329653421429433, "minimum": 0.4509525183935457, "maximum": 1.365218904961262, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1817331719704368, "standard_deviation": 0.007159286453464012, "minimum": 0.17183416710161323, "maximum": 0.19363410438334924, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937528132589262, "standard_deviation": 0.0005388727857113768, "minimum": 0.9929600359650094, "maximum": 0.9945610429014691, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2528311082456032, "standard_deviation": 0.023321808873610676, "minimum": 1.2067685122976404, "maximum": 1.2788001401660083, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2361322322121047, "standard_deviation": 0.535304923716733, "minimum": 0.38921783537323557, "maximum": 2.0056328454429657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12069928390168382, "standard_deviation": 0.008056308660443228, "minimum": 0.10576345466011965, "maximum": 0.12975600832588743, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9722143052123018, "standard_deviation": 0.003984230063164111, "minimum": 0.9671229033432103, "maximum": 0.9796794974143156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8801781873973513, "standard_deviation": 0.02006836982360118, "minimum": 1.8378104406944542, "maximum": 1.8986844690496736, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7728678214109066, "standard_deviation": 0.3997183332295308, "minimum": 0.23788730199916186, "maximum": 1.4629089266829893, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20484893674717175, "standard_deviation": 0.014760253812851204, "minimum": 0.17563043297817996, "maximum": 0.2215079106101165, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9880219926919597, "standard_deviation": 0.003334996121249103, "minimum": 0.981381863756468, "maximum": 0.9920291809959156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.37169577441135, "standard_deviation": 0.027873658305187784, "minimum": 1.3314964887240555, "maximum": 1.4142224395261342, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7471300718493046, "standard_deviation": 0.5097340861950196, "minimum": 0.3883357205160391, "maximum": 1.9357900457391397, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13458743442092636, "standard_deviation": 0.01719880164641372, "minimum": 0.11067115061758663, "maximum": 0.16328710578880362, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9915852649409834, "standard_deviation": 0.0013014531037665415, "minimum": 0.9894597926945198, "maximum": 0.9931686608098982, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4679242974254607, "standard_deviation": 0.02812264677056349, "minimum": 1.4113190503560542, "maximum": 1.5005697303574976, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3899235298829893, "standard_deviation": 0.1087911538398489, "minimum": 0.22367657454394366, "maximum": 0.535616205210276, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10207730579259794, "standard_deviation": 0.007942860409956436, "minimum": 0.09176750765781778, "maximum": 0.115820826328923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968789436385679, "standard_deviation": 0.0006347289991099133, "minimum": 0.9958437071552476, "maximum": 0.997861037247035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9770993249246513, "standard_deviation": 0.026703058026627702, "minimum": 0.9360107455679969, "maximum": 1.0064497823830134, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5205812812092299, "standard_deviation": 0.12309421792086153, "minimum": 0.30165716984783675, "maximum": 0.706564315343689, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06222115362149147, "standard_deviation": 0.006606757042883014, "minimum": 0.05124613620075485, "maximum": 0.07459447152602425, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.28637536128788665, "standard_deviation": 0.03240223420572699, "minimum": 0.24603012438983782, "maximum": 0.35533942859983797, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.720090699347607, "standard_deviation": 0.009899827689624404, "minimum": 5.702050864705736, "maximum": 5.734023932874983, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5404764214338372, "standard_deviation": 0.027511018043130307, "minimum": 0.5072179261932854, "maximum": 0.579889908785155, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12587696342658805, "standard_deviation": 0.03419274454444971, "minimum": -0.1864658513270363, "maximum": -0.07246006802900345, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943682192463368, "standard_deviation": 0.0006874325716446187, "minimum": 0.9931828914358801, "maximum": 0.9952191853388436, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8891881861907327, "standard_deviation": 0.02103542208410195, "minimum": 0.8563836040964287, "maximum": 0.9140181198369538, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27426362456179265, "standard_deviation": 0.08679776642220498, "minimum": 0.11902459883457332, "maximum": 0.36611197261115525, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07581206488490463, "standard_deviation": 0.004535187813242637, "minimum": 0.0700410938150047, "maximum": 0.08263067984613716, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981396421165389, "standard_deviation": 0.0003600399447281953, "minimum": 0.9976021623118526, "maximum": 0.9985862726195357, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5255382318703428, "standard_deviation": 0.020683431156777298, "minimum": 0.5009974484072778, "maximum": 0.5495046977576918, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46502787010347185, "standard_deviation": 0.12876550544065352, "minimum": 0.2952827579967725, "maximum": 0.7036804885140592, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04361234422378575, "standard_deviation": 0.0038033874501830693, "minimum": 0.03896897070172852, "maximum": 0.04822113962649116, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0038", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.5889651775360107, "predicted_mos": 3.5889651775360107, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "26175ddd3c5f46fede33638cfc2bb99f2850537860c425a349b620b81f62a764", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.846362499985844, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8495531878254714, "全帯域_10ms_rms_envelope_nrmse": 0.07518365721806461, "全帯域_10ms_envelope_correlation": 0.99461942769451, "全帯域_10ms_modulation_mae_db": 0.2292477509717847, "全帯域_40ms_log_rms_mae_db": 0.5151071170933661, "全帯域_40ms_rms_envelope_nrmse": 0.046648926745794704, "全帯域_40ms_envelope_correlation": 0.9980313211447602, "全帯域_40ms_modulation_mae_db": 0.6114105618896027, "300–1000 Hz_10ms_log_rms_mae_db": 1.6557810820370824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.105217427059708, "300–1000 Hz_10ms_envelope_correlation": 0.9914038708854003, "300–1000 Hz_10ms_modulation_mae_db": 0.3940639307254698, "300–1000 Hz_40ms_log_rms_mae_db": 1.202604345409968, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06778543613858723, "300–1000 Hz_40ms_envelope_correlation": 0.9966755565391135, "300–1000 Hz_40ms_modulation_mae_db": 1.08057909446264, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1291837515990597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19986696637810056, "1000–3000 Hz_10ms_envelope_correlation": 0.9736306012630893, "1000–3000 Hz_10ms_modulation_mae_db": 0.6439104350045302, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5802148681308814, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1394662154833344, "1000–3000 Hz_40ms_envelope_correlation": 0.9889507066345523, "1000–3000 Hz_40ms_modulation_mae_db": 0.5750200175357972, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3886835368259884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25232105717293013, "3000–6000 Hz_10ms_envelope_correlation": 0.9570630476785807, "3000–6000 Hz_10ms_modulation_mae_db": 0.8017453626944927, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7880866864716265, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16215915137523962, "3000–6000 Hz_40ms_envelope_correlation": 0.982016543794031, "3000–6000 Hz_40ms_modulation_mae_db": 1.2066886291879524, "mr_stft_db": 6.084683129763079, "band_mae_db": 0.24949988080571908, "rms_gain_error_db": -0.17252654175019808, "output_peak": 0.5250883133177172}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8785877220719843, "全帯域_10ms_rms_envelope_nrmse": 0.0745385883217944, "全帯域_10ms_envelope_correlation": 0.9946049363738056, "全帯域_10ms_modulation_mae_db": 0.18016549757144498, "全帯域_40ms_log_rms_mae_db": 0.4834548914119135, "全帯域_40ms_rms_envelope_nrmse": 0.03919552351009178, "全帯域_40ms_envelope_correlation": 0.9985943440779431, "全帯域_40ms_modulation_mae_db": 0.25354932724690665, "300–1000 Hz_10ms_log_rms_mae_db": 1.625267720040387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10112352108644428, "300–1000 Hz_10ms_envelope_correlation": 0.9920132353412668, "300–1000 Hz_10ms_modulation_mae_db": 0.36629959754731206, "300–1000 Hz_40ms_log_rms_mae_db": 1.1974336695934842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06130890777481775, "300–1000 Hz_40ms_envelope_correlation": 0.9972988106810026, "300–1000 Hz_40ms_modulation_mae_db": 0.35747531218596357, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0933204296009302, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16737894367202588, "1000–3000 Hz_10ms_envelope_correlation": 0.9815337753741818, "1000–3000 Hz_10ms_modulation_mae_db": 0.7224340869973621, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5683113681838026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10930197572289795, "1000–3000 Hz_40ms_envelope_correlation": 0.9934831351322012, "1000–3000 Hz_40ms_modulation_mae_db": 0.5691601017908808, "3000–6000 Hz_10ms_log_rms_mae_db": 2.418584753410095, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25903131173160354, "3000–6000 Hz_10ms_envelope_correlation": 0.9549567406793099, "3000–6000 Hz_10ms_modulation_mae_db": 1.2093792820932074, "3000–6000 Hz_40ms_log_rms_mae_db": 1.819963024739827, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15031090119776777, "3000–6000 Hz_40ms_envelope_correlation": 0.984166107451749, "3000–6000 Hz_40ms_modulation_mae_db": 1.171988975978646, "mr_stft_db": 6.075927962663973, "band_mae_db": 0.1852383472510192, "rms_gain_error_db": -0.14054818918023043, "output_peak": 0.5490074002755548}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8126684839221914, "全帯域_10ms_rms_envelope_nrmse": 0.0755209917604366, "全帯域_10ms_envelope_correlation": 0.9942778226755093, "全帯域_10ms_modulation_mae_db": 0.25164365315826814, "全帯域_40ms_log_rms_mae_db": 0.44437779522292703, "全帯域_40ms_rms_envelope_nrmse": 0.03636457827145746, "全帯域_40ms_envelope_correlation": 0.9985906573274679, "全帯域_40ms_modulation_mae_db": 0.237708922851409, "300–1000 Hz_10ms_log_rms_mae_db": 1.5126724284760198, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09917848751059812, "300–1000 Hz_10ms_envelope_correlation": 0.9918998823623417, "300–1000 Hz_10ms_modulation_mae_db": 0.2485090383177059, "300–1000 Hz_40ms_log_rms_mae_db": 1.057313076202538, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05056894101035122, "300–1000 Hz_40ms_envelope_correlation": 0.9977742062166537, "300–1000 Hz_40ms_modulation_mae_db": 0.18470946803287977, "1000–3000 Hz_10ms_log_rms_mae_db": 2.147286728623163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19253714655860282, "1000–3000 Hz_10ms_envelope_correlation": 0.9758694782321807, "1000–3000 Hz_10ms_modulation_mae_db": 1.5725046036460029, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6772608966487357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13777129983888883, "1000–3000 Hz_40ms_envelope_correlation": 0.9893545754339051, "1000–3000 Hz_40ms_modulation_mae_db": 1.9494291446495202, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4271158093224448, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25676794951223486, "3000–6000 Hz_10ms_envelope_correlation": 0.9554870441399532, "3000–6000 Hz_10ms_modulation_mae_db": 1.0308951057145943, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8521141823753835, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15674647659291166, "3000–6000 Hz_40ms_envelope_correlation": 0.9826969693108095, "3000–6000 Hz_40ms_modulation_mae_db": 0.7271776249960561, "mr_stft_db": 6.0721589234432765, "band_mae_db": 0.24339034433179785, "rms_gain_error_db": -0.03238358611799954, "output_peak": 0.5639586132458183}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8829306978356773, "全帯域_10ms_rms_envelope_nrmse": 0.08044964590993191, "全帯域_10ms_envelope_correlation": 0.99356498958559, "全帯域_10ms_modulation_mae_db": 0.3356862385651474, "全帯域_40ms_log_rms_mae_db": 0.5102192722197513, "全帯域_40ms_rms_envelope_nrmse": 0.043749457131029165, "全帯域_40ms_envelope_correlation": 0.9980291545398412, "全帯域_40ms_modulation_mae_db": 0.47835440764766385, "300–1000 Hz_10ms_log_rms_mae_db": 1.638965798602643, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1019812342969995, "300–1000 Hz_10ms_envelope_correlation": 0.9915191578363596, "300–1000 Hz_10ms_modulation_mae_db": 0.38518414029111697, "300–1000 Hz_40ms_log_rms_mae_db": 1.143320582514327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06070342175215316, "300–1000 Hz_40ms_envelope_correlation": 0.9969201845037791, "300–1000 Hz_40ms_modulation_mae_db": 0.7750783851874247, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1448184527090364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18991128323579476, "1000–3000 Hz_10ms_envelope_correlation": 0.9754912020167145, "1000–3000 Hz_10ms_modulation_mae_db": 1.2084830962290654, "1000–3000 Hz_40ms_log_rms_mae_db": 1.624100209743767, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1268752159136215, "1000–3000 Hz_40ms_envelope_correlation": 0.990153865436979, "1000–3000 Hz_40ms_modulation_mae_db": 1.6668292985847073, "3000–6000 Hz_10ms_log_rms_mae_db": 2.370127528966673, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25563090526129895, "3000–6000 Hz_10ms_envelope_correlation": 0.9564923115260511, "3000–6000 Hz_10ms_modulation_mae_db": 1.3937108706250445, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7944873789414317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1452127858579641, "3000–6000 Hz_40ms_envelope_correlation": 0.985238047779239, "3000–6000 Hz_40ms_modulation_mae_db": 0.6097431899509114, "mr_stft_db": 6.076943377268776, "band_mae_db": 0.1601722331592997, "rms_gain_error_db": -0.09292932108151855, "output_peak": 0.5782420379391933}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8962998049788641, "全帯域_10ms_rms_envelope_nrmse": 0.07650530348931722, "全帯域_10ms_envelope_correlation": 0.9943002532021099, "全帯域_10ms_modulation_mae_db": 0.2725101400854654, "全帯域_40ms_log_rms_mae_db": 0.5139319164376979, "全帯域_40ms_rms_envelope_nrmse": 0.04721684293808724, "全帯域_40ms_envelope_correlation": 0.9978319775678168, "全帯域_40ms_modulation_mae_db": 0.1657513931245466, "300–1000 Hz_10ms_log_rms_mae_db": 1.7082794928488378, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10701026827700617, "300–1000 Hz_10ms_envelope_correlation": 0.9910501933487468, "300–1000 Hz_10ms_modulation_mae_db": 0.40940232763917755, "300–1000 Hz_40ms_log_rms_mae_db": 1.1913436147385301, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07076069931786659, "300–1000 Hz_40ms_envelope_correlation": 0.9962568956221933, "300–1000 Hz_40ms_modulation_mae_db": 0.2600564595807773, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1218647554081236, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19131701991122002, "1000–3000 Hz_10ms_envelope_correlation": 0.9767005794138544, "1000–3000 Hz_10ms_modulation_mae_db": 1.4172513556844237, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5840518275772235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13448932286632545, "1000–3000 Hz_40ms_envelope_correlation": 0.9907088983277512, "1000–3000 Hz_40ms_modulation_mae_db": 1.934355151461317, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3914804394368336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2385549319840065, "3000–6000 Hz_10ms_envelope_correlation": 0.9642023888640359, "3000–6000 Hz_10ms_modulation_mae_db": 0.4734164381118713, "3000–6000 Hz_40ms_log_rms_mae_db": 1.759895564375208, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13943836509907254, "3000–6000 Hz_40ms_envelope_correlation": 0.9876010042727698, "3000–6000 Hz_40ms_modulation_mae_db": 0.4411766848093676, "mr_stft_db": 6.085066704504167, "band_mae_db": 0.22054047471073682, "rms_gain_error_db": -0.13964539188661165, "output_peak": 0.5097303167076662}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8318873247973546, "全帯域_10ms_rms_envelope_nrmse": 0.06987009453650522, "全帯域_10ms_envelope_correlation": 0.9952517426069285, "全帯域_10ms_modulation_mae_db": 0.24199048333059903, "全帯域_40ms_log_rms_mae_db": 0.47679256066974124, "全帯域_40ms_rms_envelope_nrmse": 0.037446266348994395, "全帯域_40ms_envelope_correlation": 0.9986887641529568, "全帯域_40ms_modulation_mae_db": 0.4929283655911075, "300–1000 Hz_10ms_log_rms_mae_db": 1.5915219574680646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09988879830873512, "300–1000 Hz_10ms_envelope_correlation": 0.9919942284411185, "300–1000 Hz_10ms_modulation_mae_db": 0.6244968794301843, "300–1000 Hz_40ms_log_rms_mae_db": 1.0992355529846356, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06171267432424615, "300–1000 Hz_40ms_envelope_correlation": 0.9969703514409358, "300–1000 Hz_40ms_modulation_mae_db": 0.9647808243137703, "1000–3000 Hz_10ms_log_rms_mae_db": 2.201406269732584, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22190793564756514, "1000–3000 Hz_10ms_envelope_correlation": 0.9703565455751022, "1000–3000 Hz_10ms_modulation_mae_db": 1.7473687723416766, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6308370425378316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16315346656104454, "1000–3000 Hz_40ms_envelope_correlation": 0.9882929177516775, "1000–3000 Hz_40ms_modulation_mae_db": 2.3559822471604277, "3000–6000 Hz_10ms_log_rms_mae_db": 2.36111368795773, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2436038871855229, "3000–6000 Hz_10ms_envelope_correlation": 0.960436560952777, "3000–6000 Hz_10ms_modulation_mae_db": 1.2970245832412104, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8164274480436444, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15315238131649705, "3000–6000 Hz_40ms_envelope_correlation": 0.9849557312075096, "3000–6000 Hz_40ms_modulation_mae_db": 1.616336549189109, "mr_stft_db": 6.079037966429666, "band_mae_db": 0.32323511543594624, "rms_gain_error_db": -0.1263783683758314, "output_peak": 0.5093904819909244}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8856770534848275, "全帯域_10ms_rms_envelope_nrmse": 0.07672724164184362, "全帯域_10ms_envelope_correlation": 0.9942944213532693, "全帯域_10ms_modulation_mae_db": 0.32291409099923274, "全帯域_40ms_log_rms_mae_db": 0.4981152149538618, "全帯域_40ms_rms_envelope_nrmse": 0.040765932631398195, "全帯域_40ms_envelope_correlation": 0.998482790116762, "全帯域_40ms_modulation_mae_db": 0.3446390230090299, "300–1000 Hz_10ms_log_rms_mae_db": 1.6455458553036815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10242718124813184, "300–1000 Hz_10ms_envelope_correlation": 0.9915142032627732, "300–1000 Hz_10ms_modulation_mae_db": 0.552305626204007, "300–1000 Hz_40ms_log_rms_mae_db": 1.1452275941502945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0610524630341918, "300–1000 Hz_40ms_envelope_correlation": 0.9969786104467586, "300–1000 Hz_40ms_modulation_mae_db": 0.5209268016820675, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1446766718456995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21445303242415983, "1000–3000 Hz_10ms_envelope_correlation": 0.97331530090842, "1000–3000 Hz_10ms_modulation_mae_db": 1.022500393905388, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5776040948549988, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15986715199971796, "1000–3000 Hz_40ms_envelope_correlation": 0.9895321915664761, "1000–3000 Hz_40ms_modulation_mae_db": 1.209676356192295, "3000–6000 Hz_10ms_log_rms_mae_db": 2.443203135285612, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24639472236205207, "3000–6000 Hz_10ms_envelope_correlation": 0.9590713403225989, "3000–6000 Hz_10ms_modulation_mae_db": 1.1603630855699425, "3000–6000 Hz_40ms_log_rms_mae_db": 1.80570478310948, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14718601408057871, "3000–6000 Hz_40ms_envelope_correlation": 0.9850874261044572, "3000–6000 Hz_40ms_modulation_mae_db": 1.5385221033443208, "mr_stft_db": 6.094284336856956, "band_mae_db": 0.3004490478562853, "rms_gain_error_db": -0.1476462424620423, "output_peak": 0.5566764868477939}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9064670817353085, "全帯域_10ms_rms_envelope_nrmse": 0.0645782548083869, "全帯域_10ms_envelope_correlation": 0.9958812624543104, "全帯域_10ms_modulation_mae_db": 0.2451858074487582, "全帯域_40ms_log_rms_mae_db": 0.4788114465539617, "全帯域_40ms_rms_envelope_nrmse": 0.032231848732491186, "全帯域_40ms_envelope_correlation": 0.9989688951827165, "全帯域_40ms_modulation_mae_db": 0.4016920601000897, "300–1000 Hz_10ms_log_rms_mae_db": 1.568969079845576, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08770777305834555, "300–1000 Hz_10ms_envelope_correlation": 0.9937886799468115, "300–1000 Hz_10ms_modulation_mae_db": 0.4190159021344435, "300–1000 Hz_40ms_log_rms_mae_db": 1.1027921450463498, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05181728012559665, "300–1000 Hz_40ms_envelope_correlation": 0.9978230371068098, "300–1000 Hz_40ms_modulation_mae_db": 0.5326619232743641, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1235010193498067, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2040623592215796, "1000–3000 Hz_10ms_envelope_correlation": 0.9741182410076441, "1000–3000 Hz_10ms_modulation_mae_db": 1.5526865056198442, "1000–3000 Hz_40ms_log_rms_mae_db": 1.619947943980966, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13732717106890863, "1000–3000 Hz_40ms_envelope_correlation": 0.9917799404999873, "1000–3000 Hz_40ms_modulation_mae_db": 2.231914848344482, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3938124084114745, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22870773842735093, "3000–6000 Hz_10ms_envelope_correlation": 0.9648823292248022, "3000–6000 Hz_10ms_modulation_mae_db": 1.0200121368853423, "3000–6000 Hz_40ms_log_rms_mae_db": 1.811818307692947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14019079488333805, "3000–6000 Hz_40ms_envelope_correlation": 0.9866580121752917, "3000–6000 Hz_40ms_modulation_mae_db": 0.8145260912680831, "mr_stft_db": 6.078885757582675, "band_mae_db": 0.297366639748113, "rms_gain_error_db": -0.08538340455391663, "output_peak": 0.5269454326374193}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.128785256010527, "spectral_convergence": 0.23322730801230876, "spectral_convergence_status": "ok", "amplitude_floor": 1.6152428194526273e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.911588003277493, "spectral_convergence": 0.2303765294475851, "spectral_convergence_status": "ok", "amplitude_floor": 1.0855275118818902e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.213676130001214, "spectral_convergence": 0.23915443768987196, "spectral_convergence_status": "ok", "amplitude_floor": 7.682277294566263e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006930713321042831, "output_power": 0.0007110828549571327, "delta_db": 0.11142272162476007, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002559270540127681, "output_power": 0.00242061455971413, "delta_db": -0.24190556331636828, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007234666478039212, "output_power": 0.0007155353103012132, "delta_db": -0.04787444482645682, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002266743638240336, "output_power": 0.0001912750694239296, "delta_db": -0.7374403726402633, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.48754354732022e-05, "output_power": 6.0418116234802304e-05, "delta_db": -0.3091310568531078, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.432671244156477e-05, "output_power": 4.483199187901814e-05, "delta_db": 0.04922512557335827, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24949988080571908, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.17252654175019808, "pitch": null, "output_rms_dbfs": -23.82605626012645, "output_peak": 0.5250883133177172, "reference_rms_dbfs": -23.65352971837625, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.084683129763079}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019088238002709922, "log_rms_mae_db": 0.8495531878254714, "rms_envelope_nrmse": 0.07518365721806461, "envelope_correlation": 0.99461942769451, "rms_gain_error_db": -0.17252653427261733, "reference_mean_power": 0.004315280923443156, "output_mean_power": 0.0041472137521828935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8761677888507434e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007125355736933842, "output_power": 0.000685832411625058, "reference_active": true, "delta_db": -0.16588546084897096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011942531791187452, "output_power": 0.0001143825303313376, "reference_active": true, "delta_db": -0.18736706423566823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5392584995669358e-05, "output_power": 2.35102821965431e-05, "reference_active": true, "delta_db": -0.3344907278307149}], "active_band_count": 3, "active_band_mae_db": 0.2292477509717847}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018007320980187037, "log_rms_mae_db": 0.5151071170933661, "rms_envelope_nrmse": 0.046648926745794704, "envelope_correlation": 0.9980313211447602, "rms_gain_error_db": -0.1725263908268479, "reference_mean_power": 0.004336982504671315, "output_mean_power": 0.004168070259973186}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7431428852385876e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006439047386291532, "output_power": 0.0006215856242612805, "reference_active": true, "delta_db": -0.1532065941611777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.115354721039444e-05, "output_power": 4.8612796884737825e-05, "reference_active": true, "delta_db": -0.2212514651163925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9654159892814645e-06, "output_power": 1.404355702912117e-06, "reference_active": true, "delta_db": -1.4597736263912382}], "active_band_count": 3, "active_band_mae_db": 0.6114105618896027}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017279669295308916, "log_rms_mae_db": 1.6557810820370824, "rms_envelope_nrmse": 0.105217427059708, "envelope_correlation": 0.9914038708854003, "rms_gain_error_db": -0.2699064206724145, "reference_mean_power": 0.0026069520032741206, "output_mean_power": 0.0024498663547707906}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2087875823984064e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005529275902426358, "output_power": 0.000508563972489137, "reference_active": true, "delta_db": -0.36322670319274936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648535673845476e-05, "output_power": 8.32873220351346e-05, "reference_active": true, "delta_db": -0.6388250862152169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.705863634087677e-05, "output_power": 1.636553889665912e-05, "reference_active": true, "delta_db": -0.1801400027684429}], "active_band_count": 3, "active_band_mae_db": 0.3940639307254698}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014230358058689408, "log_rms_mae_db": 1.202604345409968, "rms_envelope_nrmse": 0.06778543613858723, "envelope_correlation": 0.9966755565391135, "rms_gain_error_db": -0.26990653667399367, "reference_mean_power": 0.0026200630509491097, "output_mean_power": 0.0024621873116146063}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1215946456987494e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005050917134712818, "output_power": 0.00046448744760604286, "reference_active": true, "delta_db": -0.3639626150610051}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.798176008445663e-05, "output_power": 3.310542711061761e-05, "reference_active": true, "delta_db": -0.5967589086826025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3776930382412419e-06, "output_power": 8.147997514911758e-07, "reference_active": true, "delta_db": -2.281015759644313}], "active_band_count": 3, "active_band_mae_db": 1.08057909446264}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.058549471038879e-05, "log_rms_mae_db": 2.1291837515990597, "rms_envelope_nrmse": 0.19986696637810056, "envelope_correlation": 0.9736306012630893, "rms_gain_error_db": -0.6587804834230369, "reference_mean_power": 0.0002448429483315042, "output_mean_power": 0.00021038247122234094}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7588735706755028e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.377351846027054e-05, "output_power": 8.176352759544618e-05, "reference_active": true, "delta_db": -0.5952059106142151}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.692404819271496e-06, "output_power": 7.960724734974431e-06, "reference_active": true, "delta_db": -0.8547893767550423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.8112011913022686e-06, "output_power": 4.258292379584577e-06, "reference_active": true, "delta_db": 0.48173601764433294}], "active_band_count": 3, "active_band_mae_db": 0.6439104350045302}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.399057463688353e-05, "log_rms_mae_db": 1.5802148681308814, "rms_envelope_nrmse": 0.1394662154833344, "envelope_correlation": 0.9889507066345523, "rms_gain_error_db": -0.6587818804958943, "reference_mean_power": 0.0002460738466011548, "output_mean_power": 0.0002114400584140381}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6710797473991227e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.068314304092372e-05, "output_power": 7.878573578904719e-05, "reference_active": true, "delta_db": -0.6107896888992985}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.419207768284792e-06, "output_power": 3.6761240020081458e-06, "reference_active": true, "delta_db": -0.7995426785814902}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.686031483587781e-07, "output_power": 2.887911414788282e-07, "reference_active": true, "delta_db": 0.3147276851266029}], "active_band_count": 3, "active_band_mae_db": 0.5750200175357972}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.748604525835054e-05, "log_rms_mae_db": 2.3886835368259884, "rms_envelope_nrmse": 0.25232105717293013, "envelope_correlation": 0.9570630476785807, "rms_gain_error_db": -0.31901689176032805, "reference_mean_power": 6.169396290931914e-05, "output_mean_power": 5.732459285429972e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.220333375078007e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9572001812371347e-05, "output_power": 1.726433342827387e-05, "reference_active": true, "delta_db": -0.5448543229909721}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.501529531199527e-06, "output_power": 5.683286117953962e-06, "reference_active": true, "delta_db": -1.748975471286356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.287934963899872e-07, "output_power": 9.052708503301227e-07, "reference_active": true, "delta_db": -0.11140629380615026}], "active_band_count": 3, "active_band_mae_db": 0.8017453626944927}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.636845752844561e-05, "log_rms_mae_db": 1.7880866864716265, "rms_envelope_nrmse": 0.16215915137523962, "envelope_correlation": 0.982016543794031, "rms_gain_error_db": -0.31901789491936106, "reference_mean_power": 6.200418008997283e-05, "output_mean_power": 5.761282612269173e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.99689742660592e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0058245625049685e-05, "output_power": 1.7198556258367852e-05, "reference_active": true, "delta_db": -0.6680095383640304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.381270388794365e-06, "output_power": 2.9281187707415177e-06, "reference_active": true, "delta_db": -1.7501136770563883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.024894795639882e-07, "output_power": 7.771145146060795e-08, "reference_active": true, "delta_db": -1.201942672143438}], "active_band_count": 3, "active_band_mae_db": 1.2066886291879524}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8429552739656521, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bb5395b2afb29a042e8501019a21488abb078fbd99771e9973292c44ddefb6af", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9751269654738983, "standard_deviation": 0.0032404394057107093, "minimum": 0.9703565455751022, "maximum": 0.9815337753741818, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.13825725985855, "standard_deviation": 0.031033840879111926, "minimum": 2.0933204296009302, "maximum": 2.201406269732584, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2358924061785366, "standard_deviation": 0.40857503866475525, "minimum": 0.6439104350045302, "maximum": 1.7473687723416766, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19767933588113107, "standard_deviation": 0.016730161992092832, "minimum": 0.16737894367202588, "maximum": 0.22190793564756514, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9902820288479413, "standard_deviation": 0.0016835706844251918, "minimum": 0.9882929177516775, "maximum": 0.9934831351322012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.6077910314572756, "standard_deviation": 0.03696494507969353, "minimum": 1.5683113681838026, "maximum": 1.6772608966487357, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5615458957149282, "standard_deviation": 0.703351511914307, "minimum": 0.5691601017908808, "maximum": 2.3559822471604277, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1385314774318424, "standard_deviation": 0.01720019324320121, "minimum": 0.10930197572289795, "maximum": 0.16315346656104454, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9590739704235136, "standard_deviation": 0.003825119258881298, "minimum": 0.9549567406793099, "maximum": 0.9648823292248022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3992651624521066, "standard_deviation": 0.02824963340790749, "minimum": 2.36111368795773, "maximum": 2.443203135285612, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0483183581169633, "standard_deviation": 0.2956978882663212, "minimum": 0.4734164381118713, "maximum": 1.3937108706250445, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24762656295462498, "standard_deviation": 0.010403685996109111, "minimum": 0.22870773842735093, "maximum": 0.25903131173160354, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9848024802619821, "standard_deviation": 0.0018581374349591084, "minimum": 0.982016543794031, "maximum": 0.9876010042727698, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8060621719686936, "standard_deviation": 0.02684125479506543, "minimum": 1.759895564375208, "maximum": 1.8521141823753835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0157699810905558, "standard_deviation": 0.43319480128709564, "minimum": 0.4411766848093676, "maximum": 1.616336549189109, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1492996088004212, "standard_deviation": 0.007910880989788106, "minimum": 0.13943836509907254, "maximum": 0.16215915137523962, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918979314281023, "standard_deviation": 0.0008324640090002511, "minimum": 0.9910501933487468, "maximum": 0.9937886799468115, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6183754268277863, "standard_deviation": 0.05979350565213939, "minimum": 1.5126724284760198, "maximum": 1.7082794928488378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.42490968028617715, "standard_deviation": 0.11552482651229737, "minimum": 0.2485090383177059, "maximum": 0.6244968794301843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10056683635574608, "standard_deviation": 0.005814512242895877, "minimum": 0.08770777305834555, "maximum": 0.10701026827700617, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970872065696559, "standard_deviation": 0.0005305533527702154, "minimum": 0.9962568956221933, "maximum": 0.9978230371068098, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.142408822580016, "standard_deviation": 0.053089309952285654, "minimum": 1.057313076202538, "maximum": 1.202604345409968, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5845335335899859, "standard_deviation": 0.3274126082306161, "minimum": 0.18470946803287977, "maximum": 1.08057909446264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06071372793472632, "standard_deviation": 0.006912526485321809, "minimum": 0.05056894101035122, "maximum": 0.07076069931786659, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24748651041236464, "standard_deviation": 0.05766460506880487, "minimum": 0.1601722331592997, "maximum": 0.32323511543594624, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.080873519814071, "standard_deviation": 0.006919074466793312, "minimum": 6.0721589234432765, "maximum": 6.094284336856956, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.539879885370261, "standard_deviation": 0.025750362315400337, "minimum": 0.5093904819909244, "maximum": 0.5782420379391933, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11718013067604358, "standard_deviation": 0.04458756778921038, "minimum": -0.17252654175019808, "maximum": -0.03238358611799954, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9945993569932541, "standard_deviation": 0.000699496924989321, "minimum": 0.99356498958559, "maximum": 0.9958812624543104, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.86800891958146, "standard_deviation": 0.033023985612632284, "minimum": 0.8126684839221914, "maximum": 0.9064670817353085, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2599179577663376, "standard_deviation": 0.050392629044890544, "minimum": 0.18016549757144498, "maximum": 0.3356862385651474, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07417172221078505, "standard_deviation": 0.00485366365435624, "minimum": 0.0645782548083869, "maximum": 0.08044964590993191, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984022380137831, "standard_deviation": 0.0003935298814162001, "minimum": 0.9978319775678168, "maximum": 0.9989688951827165, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4901012768204025, "standard_deviation": 0.024206084763482436, "minimum": 0.44437779522292703, "maximum": 0.5151071170933661, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3732542576825445, "standard_deviation": 0.15093666887926674, "minimum": 0.1657513931245466, "maximum": 0.6114105618896027, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.040452422038668015, "standard_deviation": 0.005210262511497328, "minimum": 0.032231848732491186, "maximum": 0.04721684293808724, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0038", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.709907054901123, "predicted_mos": 3.709907054901123, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cfe371865404302dddc6a41be0689a6e2f2a7fb34ce0a3c02f538034589ca79f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7878757999278605, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9321144161808209, "全帯域_10ms_rms_envelope_nrmse": 0.07681459836189466, "全帯域_10ms_envelope_correlation": 0.9941970829578448, "全帯域_10ms_modulation_mae_db": 0.31534102595609226, "全帯域_40ms_log_rms_mae_db": 0.5337855241830363, "全帯域_40ms_rms_envelope_nrmse": 0.04269871086347837, "全帯域_40ms_envelope_correlation": 0.9981981796604472, "全帯域_40ms_modulation_mae_db": 0.39728017206352595, "300–1000 Hz_10ms_log_rms_mae_db": 1.3470100696700054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09561237877643985, "300–1000 Hz_10ms_envelope_correlation": 0.9927525776156374, "300–1000 Hz_10ms_modulation_mae_db": 0.3185285031662303, "300–1000 Hz_40ms_log_rms_mae_db": 0.9378257017180063, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058695042373497965, "300–1000 Hz_40ms_envelope_correlation": 0.9973629256263834, "300–1000 Hz_40ms_modulation_mae_db": 0.5986892232598495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7447026369338487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18294225781696002, "1000–3000 Hz_10ms_envelope_correlation": 0.9787762119918632, "1000–3000 Hz_10ms_modulation_mae_db": 1.0803038024138152, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2587117938865697, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12018680743773263, "1000–3000 Hz_40ms_envelope_correlation": 0.9930631663172804, "1000–3000 Hz_40ms_modulation_mae_db": 1.4319796811817038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8239113888805403, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24068732069711682, "3000–6000 Hz_10ms_envelope_correlation": 0.962174050487859, "3000–6000 Hz_10ms_modulation_mae_db": 0.7121464895575174, "3000–6000 Hz_40ms_log_rms_mae_db": 1.333917647524462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17540838624106436, "3000–6000 Hz_40ms_envelope_correlation": 0.9805665291553916, "3000–6000 Hz_40ms_modulation_mae_db": 0.9085872941287997, "mr_stft_db": 5.679372298864375, "band_mae_db": 0.2968732886804258, "rms_gain_error_db": -0.11791216468917079, "output_peak": 0.5246377007878213}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9713608438817006, "全帯域_10ms_rms_envelope_nrmse": 0.07740459145345657, "全帯域_10ms_envelope_correlation": 0.9941371563623234, "全帯域_10ms_modulation_mae_db": 0.2164220210636173, "全帯域_40ms_log_rms_mae_db": 0.546928530125438, "全帯域_40ms_rms_envelope_nrmse": 0.04083015514346377, "全帯域_40ms_envelope_correlation": 0.998421865012214, "全帯域_40ms_modulation_mae_db": 0.5470520035023028, "300–1000 Hz_10ms_log_rms_mae_db": 1.406965965704325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09877312686761557, "300–1000 Hz_10ms_envelope_correlation": 0.9922222445394565, "300–1000 Hz_10ms_modulation_mae_db": 0.3284920953605413, "300–1000 Hz_40ms_log_rms_mae_db": 0.957976187037358, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05295655332738326, "300–1000 Hz_40ms_envelope_correlation": 0.9979355237149147, "300–1000 Hz_40ms_modulation_mae_db": 0.486235907056672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.672347810138815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1610872503654781, "1000–3000 Hz_10ms_envelope_correlation": 0.9857806394272043, "1000–3000 Hz_10ms_modulation_mae_db": 1.3611336859751086, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2176896404996544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11541480739854093, "1000–3000 Hz_40ms_envelope_correlation": 0.9953880583289402, "1000–3000 Hz_40ms_modulation_mae_db": 1.6221431821890562, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7941172614160121, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2029805412620984, "3000–6000 Hz_10ms_envelope_correlation": 0.9727544962872758, "3000–6000 Hz_10ms_modulation_mae_db": 0.36891271052689983, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3778206087436298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13972098074714173, "3000–6000 Hz_40ms_envelope_correlation": 0.9870497866888509, "3000–6000 Hz_40ms_modulation_mae_db": 0.6344591732252753, "mr_stft_db": 5.689586640857418, "band_mae_db": 0.30271592983199874, "rms_gain_error_db": -0.13242507916860188, "output_peak": 0.5471823179271622}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.917900307958827, "全帯域_10ms_rms_envelope_nrmse": 0.07406678434705502, "全帯域_10ms_envelope_correlation": 0.9947384898132272, "全帯域_10ms_modulation_mae_db": 0.48255321729760553, "全帯域_40ms_log_rms_mae_db": 0.5492965479936357, "全帯域_40ms_rms_envelope_nrmse": 0.043237839219539595, "全帯域_40ms_envelope_correlation": 0.9983031934058934, "全帯域_40ms_modulation_mae_db": 0.6189237941139721, "300–1000 Hz_10ms_log_rms_mae_db": 1.3434444055772359, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10184347418146111, "300–1000 Hz_10ms_envelope_correlation": 0.9918616649213414, "300–1000 Hz_10ms_modulation_mae_db": 0.8003694495736283, "300–1000 Hz_40ms_log_rms_mae_db": 0.9854107587882074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0644430458623325, "300–1000 Hz_40ms_envelope_correlation": 0.9969039831104911, "300–1000 Hz_40ms_modulation_mae_db": 0.9907739663734837, "1000–3000 Hz_10ms_log_rms_mae_db": 1.721456250538914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16922423304157902, "1000–3000 Hz_10ms_envelope_correlation": 0.9821234991668673, "1000–3000 Hz_10ms_modulation_mae_db": 1.0121527449234746, "1000–3000 Hz_40ms_log_rms_mae_db": 1.211659897812395, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11083202979541576, "1000–3000 Hz_40ms_envelope_correlation": 0.9943687231891082, "1000–3000 Hz_40ms_modulation_mae_db": 1.3412443581979872, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7063448711486673, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18592619434737886, "3000–6000 Hz_10ms_envelope_correlation": 0.9774102916439533, "3000–6000 Hz_10ms_modulation_mae_db": 0.2838469353968283, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2550073122058318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12153746852114827, "3000–6000 Hz_40ms_envelope_correlation": 0.9898480256391281, "3000–6000 Hz_40ms_modulation_mae_db": 0.3053003111310109, "mr_stft_db": 5.693257605769929, "band_mae_db": 0.272857254559609, "rms_gain_error_db": -0.1589294693552898, "output_peak": 0.5396550018464363}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8728047434376083, "全帯域_10ms_rms_envelope_nrmse": 0.07279094428966704, "全帯域_10ms_envelope_correlation": 0.9948996459112395, "全帯域_10ms_modulation_mae_db": 0.32406142817375183, "全帯域_40ms_log_rms_mae_db": 0.5187116628899087, "全帯域_40ms_rms_envelope_nrmse": 0.041152408103176184, "全帯域_40ms_envelope_correlation": 0.9984617724751998, "全帯域_40ms_modulation_mae_db": 0.3087189534698626, "300–1000 Hz_10ms_log_rms_mae_db": 1.3793785370249108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09505787251847136, "300–1000 Hz_10ms_envelope_correlation": 0.9931122457637573, "300–1000 Hz_10ms_modulation_mae_db": 0.5733949466880137, "300–1000 Hz_40ms_log_rms_mae_db": 0.9908622834007208, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06178678326770733, "300–1000 Hz_40ms_envelope_correlation": 0.9973512370142529, "300–1000 Hz_40ms_modulation_mae_db": 0.5255726268663578, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7460176906975065, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16584465174943022, "1000–3000 Hz_10ms_envelope_correlation": 0.98328558884287, "1000–3000 Hz_10ms_modulation_mae_db": 1.5292853610999844, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2375314145467218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11246952888246003, "1000–3000 Hz_40ms_envelope_correlation": 0.9944182096297205, "1000–3000 Hz_40ms_modulation_mae_db": 1.7565221300085379, "3000–6000 Hz_10ms_log_rms_mae_db": 1.756149179594662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19409227976749246, "3000–6000 Hz_10ms_envelope_correlation": 0.9748040138925455, "3000–6000 Hz_10ms_modulation_mae_db": 0.5249718758012429, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2907959431273894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1263637280377592, "3000–6000 Hz_40ms_envelope_correlation": 0.9888602283540177, "3000–6000 Hz_40ms_modulation_mae_db": 0.9296919245104621, "mr_stft_db": 5.691686055998237, "band_mae_db": 0.22996523970105165, "rms_gain_error_db": -0.15046335940820102, "output_peak": 0.5430588599689489}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8887542918184296, "全帯域_10ms_rms_envelope_nrmse": 0.07268483440250112, "全帯域_10ms_envelope_correlation": 0.9947517736246688, "全帯域_10ms_modulation_mae_db": 0.045398963984828226, "全帯域_40ms_log_rms_mae_db": 0.5366959695664146, "全帯域_40ms_rms_envelope_nrmse": 0.03965935891849168, "全帯域_40ms_envelope_correlation": 0.9983792996359925, "全帯域_40ms_modulation_mae_db": 0.0690336257612222, "300–1000 Hz_10ms_log_rms_mae_db": 1.3798054663103, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09589661651503419, "300–1000 Hz_10ms_envelope_correlation": 0.992563538842035, "300–1000 Hz_10ms_modulation_mae_db": 0.24209885480326254, "300–1000 Hz_40ms_log_rms_mae_db": 0.9491584601602232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059337512008383914, "300–1000 Hz_40ms_envelope_correlation": 0.9971066812496067, "300–1000 Hz_40ms_modulation_mae_db": 0.252245037748195, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7049106195266415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.189248987581598, "1000–3000 Hz_10ms_envelope_correlation": 0.9768990067575393, "1000–3000 Hz_10ms_modulation_mae_db": 0.6407285367667539, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2055491350640273, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12279622312280539, "1000–3000 Hz_40ms_envelope_correlation": 0.9924874102971425, "1000–3000 Hz_40ms_modulation_mae_db": 0.8099454397677052, "3000–6000 Hz_10ms_log_rms_mae_db": 1.802742641023744, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20395837764032082, "3000–6000 Hz_10ms_envelope_correlation": 0.9724037081310036, "3000–6000 Hz_10ms_modulation_mae_db": 0.6117813650437856, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3264626742582917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12553586763270805, "3000–6000 Hz_40ms_envelope_correlation": 0.989750252023947, "3000–6000 Hz_40ms_modulation_mae_db": 0.5962557906315282, "mr_stft_db": 5.695960966379686, "band_mae_db": 0.3068406908252333, "rms_gain_error_db": -0.08289121229802629, "output_peak": 0.5477006363081867}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8781685779811615, "全帯域_10ms_rms_envelope_nrmse": 0.06731275495204103, "全帯域_10ms_envelope_correlation": 0.9956846230567649, "全帯域_10ms_modulation_mae_db": 0.2243295123848422, "全帯域_40ms_log_rms_mae_db": 0.49328014085827426, "全帯域_40ms_rms_envelope_nrmse": 0.03921227770724023, "全帯域_40ms_envelope_correlation": 0.9986323963485086, "全帯域_40ms_modulation_mae_db": 0.24051581351865714, "300–1000 Hz_10ms_log_rms_mae_db": 1.3438696435683153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0867810212379528, "300–1000 Hz_10ms_envelope_correlation": 0.994195862580654, "300–1000 Hz_10ms_modulation_mae_db": 0.40927899215076974, "300–1000 Hz_40ms_log_rms_mae_db": 0.9384125893829467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05214362044695642, "300–1000 Hz_40ms_envelope_correlation": 0.9981261877776754, "300–1000 Hz_40ms_modulation_mae_db": 0.36226499382185623, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7520667082089252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1746879939061218, "1000–3000 Hz_10ms_envelope_correlation": 0.9792135892658609, "1000–3000 Hz_10ms_modulation_mae_db": 0.932425921283634, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2272492100098278, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10910732165658013, "1000–3000 Hz_40ms_envelope_correlation": 0.9927144682495197, "1000–3000 Hz_40ms_modulation_mae_db": 0.9798002853719744, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7477089572973135, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22782387493500134, "3000–6000 Hz_10ms_envelope_correlation": 0.9660392339014794, "3000–6000 Hz_10ms_modulation_mae_db": 1.1059946959266733, "3000–6000 Hz_40ms_log_rms_mae_db": 1.310675588160524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13833748796699144, "3000–6000 Hz_40ms_envelope_correlation": 0.9887215454158645, "3000–6000 Hz_40ms_modulation_mae_db": 0.7043161545188577, "mr_stft_db": 5.687978974159421, "band_mae_db": 0.2744847024154569, "rms_gain_error_db": -0.15082094549674327, "output_peak": 0.5315655205645536}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.855971275549386, "全帯域_10ms_rms_envelope_nrmse": 0.07391139785193862, "全帯域_10ms_envelope_correlation": 0.9946992877509456, "全帯域_10ms_modulation_mae_db": 0.1614077029284899, "全帯域_40ms_log_rms_mae_db": 0.4821666431934709, "全帯域_40ms_rms_envelope_nrmse": 0.037780703656639235, "全帯域_40ms_envelope_correlation": 0.9986921252807508, "全帯域_40ms_modulation_mae_db": 0.3518375052218903, "300–1000 Hz_10ms_log_rms_mae_db": 1.3839812965593812, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09638535423366344, "300–1000 Hz_10ms_envelope_correlation": 0.992567774933898, "300–1000 Hz_10ms_modulation_mae_db": 0.3821493163868614, "300–1000 Hz_40ms_log_rms_mae_db": 0.9978311564143714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05660939941530964, "300–1000 Hz_40ms_envelope_correlation": 0.9974906283870117, "300–1000 Hz_40ms_modulation_mae_db": 0.12015136012776345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.796569938852452, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1934805527938661, "1000–3000 Hz_10ms_envelope_correlation": 0.9766597423765387, "1000–3000 Hz_10ms_modulation_mae_db": 1.016357514507747, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2952149856718698, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13006713669013228, "1000–3000 Hz_40ms_envelope_correlation": 0.9922757825254087, "1000–3000 Hz_40ms_modulation_mae_db": 1.3947171442905368, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8423838354203386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23339302956728233, "3000–6000 Hz_10ms_envelope_correlation": 0.9654775231853023, "3000–6000 Hz_10ms_modulation_mae_db": 1.1501303324728218, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3971887151848608, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14871386485720017, "3000–6000 Hz_40ms_envelope_correlation": 0.9883313884619481, "3000–6000 Hz_40ms_modulation_mae_db": 1.3155311854536882, "mr_stft_db": 5.707019262469885, "band_mae_db": 0.35046019593299554, "rms_gain_error_db": -0.13271908706512284, "output_peak": 0.5427621117138627}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8906015747179757, "全帯域_10ms_rms_envelope_nrmse": 0.07165353080483813, "全帯域_10ms_envelope_correlation": 0.9949487261894154, "全帯域_10ms_modulation_mae_db": 0.15448762120436274, "全帯域_40ms_log_rms_mae_db": 0.5326285925866359, "全帯域_40ms_rms_envelope_nrmse": 0.03875401677762949, "全帯域_40ms_envelope_correlation": 0.9985185710199006, "全帯域_40ms_modulation_mae_db": 0.12357527168585077, "300–1000 Hz_10ms_log_rms_mae_db": 1.4079171187434443, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09540272601393453, "300–1000 Hz_10ms_envelope_correlation": 0.992601633428791, "300–1000 Hz_10ms_modulation_mae_db": 0.3338371300349521, "300–1000 Hz_40ms_log_rms_mae_db": 0.936579532468259, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05170553330119042, "300–1000 Hz_40ms_envelope_correlation": 0.9978257969102878, "300–1000 Hz_40ms_modulation_mae_db": 0.36849633815769045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6668705249721039, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17118152284320204, "1000–3000 Hz_10ms_envelope_correlation": 0.9840343030494493, "1000–3000 Hz_10ms_modulation_mae_db": 0.9704218946623083, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1776917691639162, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11800593346307779, "1000–3000 Hz_40ms_envelope_correlation": 0.9959050838541338, "1000–3000 Hz_40ms_modulation_mae_db": 0.9959006546762771, "3000–6000 Hz_10ms_log_rms_mae_db": 1.742717055785909, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.198980370302219, "3000–6000 Hz_10ms_envelope_correlation": 0.9736089534931222, "3000–6000 Hz_10ms_modulation_mae_db": 0.22362805264022312, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3398131347185633, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1344362724339822, "3000–6000 Hz_40ms_envelope_correlation": 0.9873592250056381, "3000–6000 Hz_40ms_modulation_mae_db": 0.4554282416609427, "mr_stft_db": 5.695367639553512, "band_mae_db": 0.23893334495642463, "rms_gain_error_db": -0.1072447927272497, "output_peak": 0.5009755440833752}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6462800748794875, "spectral_convergence": 0.22385309931721517, "spectral_convergence_status": "ok", "amplitude_floor": 1.6152428194526273e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.456138779499231, "spectral_convergence": 0.2191499521550117, "spectral_convergence_status": "ok", "amplitude_floor": 1.0855275118818902e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.935698042214405, "spectral_convergence": 0.23317301863868672, "spectral_convergence_status": "ok", "amplitude_floor": 7.682277294566263e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006930713321042831, "output_power": 0.0006945956370373707, "delta_db": 0.009541157960239762, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002559270540127681, "output_power": 0.0024816331971121847, "delta_db": -0.13378607403254808, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007234666478039212, "output_power": 0.0007300981263577558, "delta_db": 0.039627192067535386, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002266743638240336, "output_power": 0.00019074177914343225, "delta_db": -0.7495657633077313, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.48754354732022e-05, "output_power": 5.912120664187562e-05, "delta_db": -0.4033699678052052, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.432671244156477e-05, "output_power": 4.0006500574851694e-05, "delta_db": -0.44534957690929494, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2968732886804258, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11791216468917079, "pitch": null, "output_rms_dbfs": -23.77144188306542, "output_peak": 0.5246377007878213, "reference_rms_dbfs": -23.65352971837625, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.679372298864375}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019088238002709922, "log_rms_mae_db": 0.9321144161808209, "rms_envelope_nrmse": 0.07681459836189466, "envelope_correlation": 0.9941970829578448, "rms_gain_error_db": -0.11791219623078449, "reference_mean_power": 0.004315280923443156, "output_mean_power": 0.004199695991991854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8761677888507434e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007125355736933842, "output_power": 0.0006914660802558906, "reference_active": true, "delta_db": -0.13035670859910792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011942531791187452, "output_power": 0.00011114426718375944, "reference_active": true, "delta_db": -0.31209339331837666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5392584995669358e-05, "output_power": 2.851439834720063e-05, "reference_active": true, "delta_db": 0.5035729759507922}], "active_band_count": 3, "active_band_mae_db": 0.31534102595609226}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018007320980187037, "log_rms_mae_db": 0.5337855241830363, "rms_envelope_nrmse": 0.04269871086347837, "envelope_correlation": 0.9981981796604472, "rms_gain_error_db": -0.11791239964793579, "reference_mean_power": 0.004336982504671315, "output_mean_power": 0.00422081609800658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7431428852385876e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006439047386291532, "output_power": 0.0006229666826114283, "reference_active": true, "delta_db": -0.14356800744151993}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.115354721039444e-05, "output_power": 4.8724081996160746e-05, "reference_active": true, "delta_db": -0.21132089585574715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9654159892814645e-06, "output_power": 1.620911574919697e-06, "reference_active": true, "delta_db": -0.8369516128933108}], "active_band_count": 3, "active_band_mae_db": 0.39728017206352595}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017279669295308916, "log_rms_mae_db": 1.3470100696700054, "rms_envelope_nrmse": 0.09561237877643985, "envelope_correlation": 0.9927525776156374, "rms_gain_error_db": -0.20286652856032783, "reference_mean_power": 0.0026069520032741206, "output_mean_power": 0.002487977122298361}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2087875823984064e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005529275902426358, "output_power": 0.0005286886358754188, "reference_active": true, "delta_db": -0.1946828573458312}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648535673845476e-05, "output_power": 8.344942826044156e-05, "reference_active": true, "delta_db": -0.6303804138571291}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.705863634087677e-05, "output_power": 1.7579095875173335e-05, "reference_active": true, "delta_db": 0.13052223829573065}], "active_band_count": 3, "active_band_mae_db": 0.3185285031662303}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014230358058689408, "log_rms_mae_db": 0.9378257017180063, "rms_envelope_nrmse": 0.058695042373497965, "envelope_correlation": 0.9973629256263834, "rms_gain_error_db": -0.20286655325598782, "reference_mean_power": 0.0026200630509491097, "output_mean_power": 0.0025004897997685134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1215946456987494e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005050917134712818, "output_power": 0.00048312515292712197, "reference_active": true, "delta_db": -0.19310594807810696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.798176008445663e-05, "output_power": 3.265766961830608e-05, "reference_active": true, "delta_db": -0.6558989500668673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3776930382412419e-06, "output_power": 1.107761294426189e-06, "reference_active": true, "delta_db": -0.9470627716345743}], "active_band_count": 3, "active_band_mae_db": 0.5986892232598495}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.058549471038879e-05, "log_rms_mae_db": 1.7447026369338487, "rms_envelope_nrmse": 0.18294225781696002, "envelope_correlation": 0.9787762119918632, "rms_gain_error_db": -0.6817412542929477, "reference_mean_power": 0.0002448429483315042, "output_mean_power": 0.00020927313249956349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7588735706755028e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.377351846027054e-05, "output_power": 7.815036432200846e-05, "reference_active": true, "delta_db": -0.7914920444578029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.692404819271496e-06, "output_power": 7.64500621598161e-06, "reference_active": true, "delta_db": -1.0305370200603416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.8112011913022686e-06, "output_power": 2.748993046326104e-06, "reference_active": true, "delta_db": -1.418882342723301}], "active_band_count": 3, "active_band_mae_db": 1.0803038024138152}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.399057463688353e-05, "log_rms_mae_db": 1.2587117938865697, "rms_envelope_nrmse": 0.12018680743773263, "envelope_correlation": 0.9930631663172804, "rms_gain_error_db": -0.681744632752894, "reference_mean_power": 0.0002460738466011548, "output_mean_power": 0.0002103250471175774}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6710797473991227e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.068314304092372e-05, "output_power": 7.52046421103856e-05, "reference_active": true, "delta_db": -0.8128191534011809}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.419207768284792e-06, "output_power": 3.1245989553344254e-06, "reference_active": true, "delta_db": -1.5055013709368046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.686031483587781e-07, "output_power": 1.7035278849618924e-07, "reference_active": true, "delta_db": -1.9776185192071263}], "active_band_count": 3, "active_band_mae_db": 1.4319796811817038}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.748604525835054e-05, "log_rms_mae_db": 1.8239113888805403, "rms_envelope_nrmse": 0.24068732069711682, "envelope_correlation": 0.962174050487859, "rms_gain_error_db": -0.6434427513540774, "reference_mean_power": 6.169396290931914e-05, "output_mean_power": 5.319837823081112e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.220333375078007e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9572001812371347e-05, "output_power": 1.6015785594061274e-05, "reference_active": true, "delta_db": -0.8708700110069338}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.501529531199527e-06, "output_power": 6.926941265121303e-06, "reference_active": true, "delta_db": -0.8895556255268375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.287934963899872e-07, "output_power": 1.0127926686401737e-06, "reference_active": true, "delta_db": 0.37601383213878076}], "active_band_count": 3, "active_band_mae_db": 0.7121464895575174}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.636845752844561e-05, "log_rms_mae_db": 1.333917647524462, "rms_envelope_nrmse": 0.17540838624106436, "envelope_correlation": 0.9805665291553916, "rms_gain_error_db": -0.6434434970884262, "reference_mean_power": 6.200418008997283e-05, "output_mean_power": 5.346586768581165e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.99689742660592e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0058245625049685e-05, "output_power": 1.6349686478668722e-05, "reference_active": true, "delta_db": -0.8878351614890329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.381270388794365e-06, "output_power": 3.5313439311678605e-06, "reference_active": true, "delta_db": -0.9366003912659703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.024894795639882e-07, "output_power": 8.328113874108961e-08, "reference_active": true, "delta_db": -0.901326329631396}], "active_band_count": 3, "active_band_mae_db": 0.9085872941287997}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8621912195621663, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "22e421ddbfe102c27c36f2c0f3980f0de2d83de98d03df8b3d7d11e96a3ddde4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9808465726097741, "standard_deviation": 0.003425781111800242, "minimum": 0.9766597423765387, "maximum": 0.9857806394272043, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.725617772483651, "standard_deviation": 0.04350562731175437, "minimum": 1.6668705249721039, "maximum": 1.796569938852452, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0678511827041033, "standard_deviation": 0.27126985287758826, "minimum": 0.6407285367667539, "maximum": 1.5292853610999844, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17596218126227942, "standard_deviation": 0.011499442217443846, "minimum": 0.1610872503654781, "maximum": 0.1934805527938661, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938276127989067, "standard_deviation": 0.0013846232320454423, "minimum": 0.9922757825254087, "maximum": 0.9959050838541338, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2289122308318727, "standard_deviation": 0.03576527583625607, "minimum": 1.1776917691639162, "maximum": 1.2952149856718698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2915316094604723, "standard_deviation": 0.33270832093613173, "minimum": 0.8099454397677052, "maximum": 1.7565221300085379, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11735997355584311, "standard_deviation": 0.006947286128573974, "minimum": 0.10910732165658013, "maximum": 0.13006713669013228, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9705840338778176, "standard_deviation": 0.005332350934881763, "minimum": 0.962174050487859, "maximum": 0.9774102916439533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7770093988208981, "standard_deviation": 0.0461234325799574, "minimum": 1.7063448711486673, "maximum": 1.8423838354203386, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.622676557170749, "standard_deviation": 0.3521190948139531, "minimum": 0.22362805264022312, "maximum": 1.1501303324728218, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21098024856486375, "standard_deviation": 0.020137948631494106, "minimum": 0.18592619434737886, "maximum": 0.24068732069711682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9875608725930982, "standard_deviation": 0.002997054352024401, "minimum": 0.9805665291553916, "maximum": 0.9898480256391281, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3289602029904442, "standard_deviation": 0.045435884316842874, "minimum": 1.2550073122058318, "maximum": 1.3971887151848608, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7311962594075706, "standard_deviation": 0.31578480611850007, "minimum": 0.3053003111310109, "maximum": 1.3155311854536882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13875675705474944, "standard_deviation": 0.01726739703266237, "minimum": 0.12153746852114827, "maximum": 0.17540838624106436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9927346928281964, "standard_deviation": 0.0006946246858386862, "minimum": 0.9918616649213414, "maximum": 0.994195862580654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3740465628947398, "standard_deviation": 0.026643964588390263, "minimum": 1.3434444055772359, "maximum": 1.4079171187434443, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4235186610205325, "standard_deviation": 0.18023893395741736, "minimum": 0.24209885480326254, "maximum": 0.8003694495736283, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09571907129307161, "standard_deviation": 0.004271975148579435, "minimum": 0.0867810212379528, "maximum": 0.10184347418146111, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975128704738279, "standard_deviation": 0.00042046808645592533, "minimum": 0.9969039831104911, "maximum": 0.9981261877776754, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9617570836712617, "standard_deviation": 0.025725345357069943, "minimum": 0.936579532468259, "maximum": 0.9978311564143714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.46305368167648353, "standard_deviation": 0.2625298541917805, "minimum": 0.12015136012776345, "maximum": 0.9907739663734837, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.057209686250345185, "standard_deviation": 0.0046967380779261375, "minimum": 0.05170553330119042, "maximum": 0.0644430458623325, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2841413308628995, "standard_deviation": 0.03895045016776839, "minimum": 0.22996523970105165, "maximum": 0.35046019593299554, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.692528680506558, "standard_deviation": 0.00786951070887527, "minimum": 5.679372298864375, "maximum": 5.707019262469885, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5346922116500434, "standard_deviation": 0.015732851445257733, "minimum": 0.5009755440833752, "maximum": 0.5477006363081867, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1291757637760507, "standard_deviation": 0.025579477451163696, "minimum": -0.1589294693552898, "maximum": -0.08289121229802629, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9947570982083037, "standard_deviation": 0.000480922494639554, "minimum": 0.9941371563623234, "maximum": 0.9956846230567649, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9009595039407388, "standard_deviation": 0.03742855957413076, "minimum": 0.855971275549386, "maximum": 0.9713608438817006, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24050018662419875, "standard_deviation": 0.13295155949450116, "minimum": 0.045398963984828226, "maximum": 0.48255321729760553, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07332992955792403, "standard_deviation": 0.003147566009507751, "minimum": 0.06731275495204103, "maximum": 0.07740459145345657, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984509253548634, "standard_deviation": 0.00016365581640909142, "minimum": 0.9981981796604472, "maximum": 0.9986921252807508, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5241867014246018, "standard_deviation": 0.024548659283454385, "minimum": 0.4821666431934709, "maximum": 0.5492965479936357, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3321171424171605, "standard_deviation": 0.19076203964821267, "minimum": 0.0690336257612222, "maximum": 0.6189237941139721, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.040415683798707316, "standard_deviation": 0.0019129556106295528, "minimum": 0.037780703656639235, "maximum": 0.043237839219539595, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0038", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.4814248085021973, "predicted_mos": 3.4814248085021973, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b0e18d8e6b7bad0d11e4dca03d2a5211202dfb8a4bdd6f204ed31dac725cf6f9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.296516099944711, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8849456620787739, "全帯域_10ms_rms_envelope_nrmse": 0.07744855360401826, "全帯域_10ms_envelope_correlation": 0.9940124369203946, "全帯域_10ms_modulation_mae_db": 0.2808856853632222, "全帯域_40ms_log_rms_mae_db": 0.5107686789945202, "全帯域_40ms_rms_envelope_nrmse": 0.043208943509797974, "全帯域_40ms_envelope_correlation": 0.9980350160705834, "全帯域_40ms_modulation_mae_db": 0.2728206524844476, "300–1000 Hz_10ms_log_rms_mae_db": 1.4543993540006057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09933661723674762, "300–1000 Hz_10ms_envelope_correlation": 0.9918990198686499, "300–1000 Hz_10ms_modulation_mae_db": 0.4901867193794995, "300–1000 Hz_40ms_log_rms_mae_db": 0.9902401731352998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057063753428927506, "300–1000 Hz_40ms_envelope_correlation": 0.9971924295425334, "300–1000 Hz_40ms_modulation_mae_db": 0.2714523881467036, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8410496877845257, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19464143215326082, "1000–3000 Hz_10ms_envelope_correlation": 0.975648269920408, "1000–3000 Hz_10ms_modulation_mae_db": 1.126510957416656, "1000–3000 Hz_40ms_log_rms_mae_db": 1.332785235196777, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12476732239539724, "1000–3000 Hz_40ms_envelope_correlation": 0.9925971655892889, "1000–3000 Hz_40ms_modulation_mae_db": 1.4645821800517451, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9508989926187112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22545742576309163, "3000–6000 Hz_10ms_envelope_correlation": 0.9658909846015498, "3000–6000 Hz_10ms_modulation_mae_db": 1.2331124614055269, "3000–6000 Hz_40ms_log_rms_mae_db": 1.453132901644808, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13514628119474353, "3000–6000 Hz_40ms_envelope_correlation": 0.9871335498650775, "3000–6000 Hz_40ms_modulation_mae_db": 0.7966564594183035, "mr_stft_db": 5.7833627239439815, "band_mae_db": 0.23657943702686976, "rms_gain_error_db": -0.07166509429280052, "output_peak": 0.5531156538892668}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8577670914879175, "全帯域_10ms_rms_envelope_nrmse": 0.07354104570996178, "全帯域_10ms_envelope_correlation": 0.9947100936820812, "全帯域_10ms_modulation_mae_db": 0.3203568152229949, "全帯域_40ms_log_rms_mae_db": 0.5192965412233217, "全帯域_40ms_rms_envelope_nrmse": 0.039744004319761495, "全帯域_40ms_envelope_correlation": 0.9984854549516531, "全帯域_40ms_modulation_mae_db": 0.4241081494718199, "300–1000 Hz_10ms_log_rms_mae_db": 1.4584247696336374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0990570074328651, "300–1000 Hz_10ms_envelope_correlation": 0.9921470108375733, "300–1000 Hz_10ms_modulation_mae_db": 0.32414852421419843, "300–1000 Hz_40ms_log_rms_mae_db": 0.9751556730379717, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059913571048490204, "300–1000 Hz_40ms_envelope_correlation": 0.9971839657432062, "300–1000 Hz_40ms_modulation_mae_db": 0.628339993983932, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8289214466502748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17953809373036125, "1000–3000 Hz_10ms_envelope_correlation": 0.9786352799937879, "1000–3000 Hz_10ms_modulation_mae_db": 0.779275154837244, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3091448101676253, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10684705225181686, "1000–3000 Hz_40ms_envelope_correlation": 0.9941090871635343, "1000–3000 Hz_40ms_modulation_mae_db": 0.4925893253592843, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9280610122488893, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2133110623381997, "3000–6000 Hz_10ms_envelope_correlation": 0.9694402552857071, "3000–6000 Hz_10ms_modulation_mae_db": 0.9066453787000528, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4003505967107586, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12361712586322934, "3000–6000 Hz_40ms_envelope_correlation": 0.9894084547295091, "3000–6000 Hz_40ms_modulation_mae_db": 0.43152524178907076, "mr_stft_db": 5.801093012578728, "band_mae_db": 0.2559053374401736, "rms_gain_error_db": -0.12352876989411651, "output_peak": 0.5756144552015973}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8830807386284473, "全帯域_10ms_rms_envelope_nrmse": 0.08019462228892377, "全帯域_10ms_envelope_correlation": 0.9937128935052303, "全帯域_10ms_modulation_mae_db": 0.25616710197803033, "全帯域_40ms_log_rms_mae_db": 0.5318844554264224, "全帯域_40ms_rms_envelope_nrmse": 0.04843955113729759, "全帯域_40ms_envelope_correlation": 0.9977031720643268, "全帯域_40ms_modulation_mae_db": 0.5189818647836605, "300–1000 Hz_10ms_log_rms_mae_db": 1.4438310586518956, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10301643837543827, "300–1000 Hz_10ms_envelope_correlation": 0.9914515324276657, "300–1000 Hz_10ms_modulation_mae_db": 0.25801677159067915, "300–1000 Hz_40ms_log_rms_mae_db": 1.0025944706352419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06459837644373913, "300–1000 Hz_40ms_envelope_correlation": 0.9966284488978504, "300–1000 Hz_40ms_modulation_mae_db": 0.41866052192607656, "1000–3000 Hz_10ms_log_rms_mae_db": 1.817045703376861, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18034893596806761, "1000–3000 Hz_10ms_envelope_correlation": 0.9793440870748792, "1000–3000 Hz_10ms_modulation_mae_db": 1.3114691874289648, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3272156772378298, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12456976459879364, "1000–3000 Hz_40ms_envelope_correlation": 0.9920432953508144, "1000–3000 Hz_40ms_modulation_mae_db": 2.0560592741383785, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9363623901956561, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22005190880462222, "3000–6000 Hz_10ms_envelope_correlation": 0.9676574206133354, "3000–6000 Hz_10ms_modulation_mae_db": 1.0930666239615359, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4443985044500274, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13779978696512774, "3000–6000 Hz_40ms_envelope_correlation": 0.9874333626599698, "3000–6000 Hz_40ms_modulation_mae_db": 0.698627878696751, "mr_stft_db": 5.796830651402677, "band_mae_db": 0.25158032007817493, "rms_gain_error_db": -0.14065288464099285, "output_peak": 0.6060859567983243}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8667029417028764, "全帯域_10ms_rms_envelope_nrmse": 0.07116230541047282, "全帯域_10ms_envelope_correlation": 0.9950541749230657, "全帯域_10ms_modulation_mae_db": 0.3645340072486612, "全帯域_40ms_log_rms_mae_db": 0.48910740053784574, "全帯域_40ms_rms_envelope_nrmse": 0.04044567280998405, "全帯域_40ms_envelope_correlation": 0.9984182397990714, "全帯域_40ms_modulation_mae_db": 0.6593067512893137, "300–1000 Hz_10ms_log_rms_mae_db": 1.5298963287535159, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09262420736634489, "300–1000 Hz_10ms_envelope_correlation": 0.9932645421025957, "300–1000 Hz_10ms_modulation_mae_db": 0.44127014821120136, "300–1000 Hz_40ms_log_rms_mae_db": 1.0257559728266616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05956743151185436, "300–1000 Hz_40ms_envelope_correlation": 0.9973161520811539, "300–1000 Hz_40ms_modulation_mae_db": 0.4728101186155175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.871779906665313, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17590304136296703, "1000–3000 Hz_10ms_envelope_correlation": 0.980561516722074, "1000–3000 Hz_10ms_modulation_mae_db": 0.6516764511643299, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3390767041915832, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11683593719600248, "1000–3000 Hz_40ms_envelope_correlation": 0.9935572969107432, "1000–3000 Hz_40ms_modulation_mae_db": 0.8950338441857659, "3000–6000 Hz_10ms_log_rms_mae_db": 1.98355637060559, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20008526010427138, "3000–6000 Hz_10ms_envelope_correlation": 0.9733330411424322, "3000–6000 Hz_10ms_modulation_mae_db": 0.5752745060115442, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4509865779680369, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1341853681762104, "3000–6000 Hz_40ms_envelope_correlation": 0.9878423453307207, "3000–6000 Hz_40ms_modulation_mae_db": 0.6910835756414621, "mr_stft_db": 5.780044851416663, "band_mae_db": 0.2937538413388197, "rms_gain_error_db": -0.12221314707552654, "output_peak": 0.5308898720609736}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8736411285424779, "全帯域_10ms_rms_envelope_nrmse": 0.0801901245183251, "全帯域_10ms_envelope_correlation": 0.9939141958926418, "全帯域_10ms_modulation_mae_db": 0.40448365933891806, "全帯域_40ms_log_rms_mae_db": 0.5066425275159899, "全帯域_40ms_rms_envelope_nrmse": 0.0471694119450764, "全帯域_40ms_envelope_correlation": 0.99808228371453, "全帯域_40ms_modulation_mae_db": 0.5000889845283298, "300–1000 Hz_10ms_log_rms_mae_db": 1.4866232077077917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11471522077633824, "300–1000 Hz_10ms_envelope_correlation": 0.9897861875850129, "300–1000 Hz_10ms_modulation_mae_db": 0.5544334236317728, "300–1000 Hz_40ms_log_rms_mae_db": 1.0452803410241103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07510865165613802, "300–1000 Hz_40ms_envelope_correlation": 0.9959081712566326, "300–1000 Hz_40ms_modulation_mae_db": 0.5702307837166475, "1000–3000 Hz_10ms_log_rms_mae_db": 1.868843080195578, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1938324547201729, "1000–3000 Hz_10ms_envelope_correlation": 0.9763711332462466, "1000–3000 Hz_10ms_modulation_mae_db": 1.034109050839284, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3458624218857582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12670192269807803, "1000–3000 Hz_40ms_envelope_correlation": 0.9928243191606035, "1000–3000 Hz_40ms_modulation_mae_db": 1.6292376947008667, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9533464298614067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18426020705037124, "3000–6000 Hz_10ms_envelope_correlation": 0.9773590688440881, "3000–6000 Hz_10ms_modulation_mae_db": 1.1049417821087253, "3000–6000 Hz_40ms_log_rms_mae_db": 1.401369546433878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11314506274416322, "3000–6000 Hz_40ms_envelope_correlation": 0.991338374086899, "3000–6000 Hz_40ms_modulation_mae_db": 0.5805081806163691, "mr_stft_db": 5.79234319894693, "band_mae_db": 0.27526376766455624, "rms_gain_error_db": -0.19483065690325374, "output_peak": 0.5002489107715833}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8775109779263252, "全帯域_10ms_rms_envelope_nrmse": 0.07809409365524804, "全帯域_10ms_envelope_correlation": 0.9941140188539409, "全帯域_10ms_modulation_mae_db": 0.08727480571532402, "全帯域_40ms_log_rms_mae_db": 0.5043800090776428, "全帯域_40ms_rms_envelope_nrmse": 0.044481879621342715, "全帯域_40ms_envelope_correlation": 0.9981811924251923, "全帯域_40ms_modulation_mae_db": 0.33728667424755027, "300–1000 Hz_10ms_log_rms_mae_db": 1.4874357648443897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10427194580332, "300–1000 Hz_10ms_envelope_correlation": 0.9913627891491266, "300–1000 Hz_10ms_modulation_mae_db": 0.20042530769230393, "300–1000 Hz_40ms_log_rms_mae_db": 1.007594751493478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06463208997498977, "300–1000 Hz_40ms_envelope_correlation": 0.9967898487800231, "300–1000 Hz_40ms_modulation_mae_db": 0.4861018189799206, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8652553832287897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17913729542282816, "1000–3000 Hz_10ms_envelope_correlation": 0.9799500279458657, "1000–3000 Hz_10ms_modulation_mae_db": 0.5606499966779509, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3138937688333163, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11734132237502583, "1000–3000 Hz_40ms_envelope_correlation": 0.9938098791545001, "1000–3000 Hz_40ms_modulation_mae_db": 0.9571242211169751, "3000–6000 Hz_10ms_log_rms_mae_db": 1.921311471127489, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20353443330412394, "3000–6000 Hz_10ms_envelope_correlation": 0.9723040485484614, "3000–6000 Hz_10ms_modulation_mae_db": 0.521677561733713, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4403208424611624, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11746324929044204, "3000–6000 Hz_40ms_envelope_correlation": 0.9907686472884272, "3000–6000 Hz_40ms_modulation_mae_db": 0.7698292990728426, "mr_stft_db": 5.790300299419293, "band_mae_db": 0.2555433045533031, "rms_gain_error_db": -0.15912149732463693, "output_peak": 0.5244879532301122}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8695759653299667, "全帯域_10ms_rms_envelope_nrmse": 0.07488519651834367, "全帯域_10ms_envelope_correlation": 0.9944659154441023, "全帯域_10ms_modulation_mae_db": 0.19480699280408662, "全帯域_40ms_log_rms_mae_db": 0.5244668850254949, "全帯域_40ms_rms_envelope_nrmse": 0.043538559933976546, "全帯域_40ms_envelope_correlation": 0.9980710805322424, "全帯域_40ms_modulation_mae_db": 0.33936362723055474, "300–1000 Hz_10ms_log_rms_mae_db": 1.5295024424574597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10322649667972204, "300–1000 Hz_10ms_envelope_correlation": 0.9914061152109114, "300–1000 Hz_10ms_modulation_mae_db": 0.29364816583032344, "300–1000 Hz_40ms_log_rms_mae_db": 1.1001839864352352, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06509272890257888, "300–1000 Hz_40ms_envelope_correlation": 0.9965510368860858, "300–1000 Hz_40ms_modulation_mae_db": 0.3796732642158725, "1000–3000 Hz_10ms_log_rms_mae_db": 1.863606866890891, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17776233816041145, "1000–3000 Hz_10ms_envelope_correlation": 0.9790077497002557, "1000–3000 Hz_10ms_modulation_mae_db": 1.0596538739578767, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3717297470499046, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11139992015537067, "1000–3000 Hz_40ms_envelope_correlation": 0.9931695900303069, "1000–3000 Hz_40ms_modulation_mae_db": 1.4030412154431362, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9842209577511123, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21893635137522333, "3000–6000 Hz_10ms_envelope_correlation": 0.9677814171947915, "3000–6000 Hz_10ms_modulation_mae_db": 0.581281531354143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4521238538641261, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15518744586616007, "3000–6000 Hz_40ms_envelope_correlation": 0.9830268387898218, "3000–6000 Hz_40ms_modulation_mae_db": 0.758332209218945, "mr_stft_db": 5.781787926997509, "band_mae_db": 0.1943977130444063, "rms_gain_error_db": -0.10034875645317223, "output_peak": 0.5473702728454599}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8499052058834894, "全帯域_10ms_rms_envelope_nrmse": 0.06757833318310551, "全帯域_10ms_envelope_correlation": 0.9955717320065193, "全帯域_10ms_modulation_mae_db": 0.2479654663189518, "全帯域_40ms_log_rms_mae_db": 0.4992302455391599, "全帯域_40ms_rms_envelope_nrmse": 0.038528785486448856, "全帯域_40ms_envelope_correlation": 0.9985951439019193, "全帯域_40ms_modulation_mae_db": 0.3193510653022233, "300–1000 Hz_10ms_log_rms_mae_db": 1.4156907397529916, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0888969135978427, "300–1000 Hz_10ms_envelope_correlation": 0.9936510008514243, "300–1000 Hz_10ms_modulation_mae_db": 0.3725203167410456, "300–1000 Hz_40ms_log_rms_mae_db": 0.9934764131870202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051831156338916794, "300–1000 Hz_40ms_envelope_correlation": 0.9978695856220374, "300–1000 Hz_40ms_modulation_mae_db": 0.5366426873665304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8597348183859603, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18188688745113588, "1000–3000 Hz_10ms_envelope_correlation": 0.9802041862388687, "1000–3000 Hz_10ms_modulation_mae_db": 0.9286239992871724, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3632176196476613, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12459669207790379, "1000–3000 Hz_40ms_envelope_correlation": 0.9936457557170144, "1000–3000 Hz_40ms_modulation_mae_db": 1.0557450099414956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9636189514815747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21801985359130935, "3000–6000 Hz_10ms_envelope_correlation": 0.9689976608782951, "3000–6000 Hz_10ms_modulation_mae_db": 1.3339186602076518, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4722727883388647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14515082271944646, "3000–6000 Hz_40ms_envelope_correlation": 0.9870404937822529, "3000–6000 Hz_40ms_modulation_mae_db": 1.8385113574135403, "mr_stft_db": 5.7883458262082, "band_mae_db": 0.3040115581515949, "rms_gain_error_db": -0.12545083177248853, "output_peak": 0.5463248215653651}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.79667735506413, "spectral_convergence": 0.22308577506136787, "spectral_convergence_status": "ok", "amplitude_floor": 1.6152428194526273e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.576366471767568, "spectral_convergence": 0.21845135965214252, "spectral_convergence_status": "ok", "amplitude_floor": 1.0855275118818902e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.977044345000248, "spectral_convergence": 0.22523113951275436, "spectral_convergence_status": "ok", "amplitude_floor": 7.682277294566263e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006930713321042831, "output_power": 0.000693101319578818, "delta_db": 0.0001879043667889845, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002559270540127681, "output_power": 0.0025246795063790066, "delta_db": -0.059099427687188275, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007234666478039212, "output_power": 0.0007257758369128022, "delta_db": 0.013839905287701886, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002266743638240336, "output_power": 0.00019165605243009017, "delta_db": -0.728798668562173, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.48754354732022e-05, "output_power": 5.924320703579346e-05, "delta_db": -0.3944172572758504, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.432671244156477e-05, "output_power": 4.666366743460426e-05, "delta_db": 0.22313345898151582, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23657943702686976, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07166509429280052, "pitch": null, "output_rms_dbfs": -23.725194812669056, "output_peak": 0.5531156538892668, "reference_rms_dbfs": -23.65352971837625, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7833627239439815}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019088238002709922, "log_rms_mae_db": 0.8849456620787739, "rms_envelope_nrmse": 0.07744855360401826, "envelope_correlation": 0.9940124369203946, "rms_gain_error_db": -0.07166508162936376, "reference_mean_power": 0.004315280923443156, "output_mean_power": 0.004244656643115804}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8761677888507434e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007125355736933842, "output_power": 0.0006984941615485208, "reference_active": true, "delta_db": -0.086437712299085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011942531791187452, "output_power": 0.00011092829019805154, "reference_active": true, "delta_db": -0.3205408704663806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5392584995669358e-05, "output_power": 2.8072090942789078e-05, "reference_active": true, "delta_db": 0.4356784733242011}], "active_band_count": 3, "active_band_mae_db": 0.2808856853632222}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018007320980187037, "log_rms_mae_db": 0.5107686789945202, "rms_envelope_nrmse": 0.043208943509797974, "envelope_correlation": 0.9980350160705834, "rms_gain_error_db": -0.07166518145525629, "reference_mean_power": 0.004336982504671315, "output_mean_power": 0.004266002956233193}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7431428852385876e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006439047386291532, "output_power": 0.000628602694808395, "reference_active": true, "delta_db": -0.10445382669329345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.115354721039444e-05, "output_power": 4.7433339121607035e-05, "reference_active": true, "delta_db": -0.3279205660728939}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9654159892814645e-06, "output_power": 1.7982320273484188e-06, "reference_active": true, "delta_db": -0.38608756468715544}], "active_band_count": 3, "active_band_mae_db": 0.2728206524844476}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017279669295308916, "log_rms_mae_db": 1.4543993540006057, "rms_envelope_nrmse": 0.09933661723674762, "envelope_correlation": 0.9918990198686499, "rms_gain_error_db": -0.09268112603586812, "reference_mean_power": 0.0026069520032741206, "output_mean_power": 0.002551907467857856}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2087875823984064e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005529275902426358, "output_power": 0.0005415473749119603, "reference_active": true, "delta_db": -0.09031806127787258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648535673845476e-05, "output_power": 8.440057230359399e-05, "reference_active": true, "delta_db": -0.5811601541576874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.705863634087677e-05, "output_power": 2.050465682778602e-05, "reference_active": true, "delta_db": 0.7990819427029384}], "active_band_count": 3, "active_band_mae_db": 0.4901867193794995}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014230358058689408, "log_rms_mae_db": 0.9902401731352998, "rms_envelope_nrmse": 0.057063753428927506, "envelope_correlation": 0.9971924295425334, "rms_gain_error_db": -0.09268115114972712, "reference_mean_power": 0.0026200630509491097, "output_mean_power": 0.002564741667247531}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1215946456987494e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005050917134712818, "output_power": 0.0004957072658012772, "reference_active": true, "delta_db": -0.08144958941960778}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.798176008445663e-05, "output_power": 3.236497105138137e-05, "reference_active": true, "delta_db": -0.6949986328195213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3776930382412419e-06, "output_power": 1.38977136138581e-06, "reference_active": true, "delta_db": 0.03790894220098164}], "active_band_count": 3, "active_band_mae_db": 0.2714523881467036}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.058549471038879e-05, "log_rms_mae_db": 1.8410496877845257, "rms_envelope_nrmse": 0.19464143215326082, "envelope_correlation": 0.975648269920408, "rms_gain_error_db": -0.7053475964466306, "reference_mean_power": 0.0002448429483315042, "output_mean_power": 0.00020813870151748843}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7588735706755028e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.377351846027054e-05, "output_power": 8.041360489562794e-05, "reference_active": true, "delta_db": -0.6675067967910431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.692404819271496e-06, "output_power": 8.478253463211822e-06, "reference_active": true, "delta_db": -0.5812514891135766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.8112011913022686e-06, "output_power": 2.333374336114181e-06, "reference_active": true, "delta_db": -2.130774586345349}], "active_band_count": 3, "active_band_mae_db": 1.126510957416656}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.399057463688353e-05, "log_rms_mae_db": 1.332785235196777, "rms_envelope_nrmse": 0.12476732239539724, "envelope_correlation": 0.9925971655892889, "rms_gain_error_db": -0.705353028999155, "reference_mean_power": 0.0002460738466011548, "output_mean_power": 0.00020918481496208404}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6710797473991227e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.068314304092372e-05, "output_power": 7.721453608694329e-05, "reference_active": true, "delta_db": -0.6982749768311687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.419207768284792e-06, "output_power": 3.93266371315206e-06, "reference_active": true, "delta_db": -0.5065760937269959}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.686031483587781e-07, "output_power": 1.2889069095900213e-07, "reference_active": true, "delta_db": -3.1888954695970706}], "active_band_count": 3, "active_band_mae_db": 1.4645821800517451}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.748604525835054e-05, "log_rms_mae_db": 1.9508989926187112, "rms_envelope_nrmse": 0.22545742576309163, "envelope_correlation": 0.9658909846015498, "rms_gain_error_db": -0.13176606055559453, "reference_mean_power": 6.169396290931914e-05, "output_mean_power": 5.9850262835575974e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.220333375078007e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9572001812371347e-05, "output_power": 1.7031496740022474e-05, "reference_active": true, "delta_db": -0.6038243155659625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.501529531199527e-06, "output_power": 7.1165670333138e-06, "reference_active": true, "delta_db": -0.7722652326309645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.287934963899872e-07, "output_power": 1.5857836993470118e-06, "reference_active": true, "delta_db": 2.323247836019654}], "active_band_count": 3, "active_band_mae_db": 1.2331124614055269}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.636845752844561e-05, "log_rms_mae_db": 1.453132901644808, "rms_envelope_nrmse": 0.13514628119474353, "envelope_correlation": 0.9871335498650775, "rms_gain_error_db": -0.13176596210580127, "reference_mean_power": 6.200418008997283e-05, "output_mean_power": 6.015121066006673e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.99689742660592e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0058245625049685e-05, "output_power": 1.7724240481149722e-05, "reference_active": true, "delta_db": -0.5372531132676818}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.381270388794365e-06, "output_power": 3.834399202464619e-06, "reference_active": true, "delta_db": -0.5790273070687111}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.024894795639882e-07, "output_power": 1.3741944290995408e-07, "reference_active": true, "delta_db": 1.2736889579185178}], "active_band_count": 3, "active_band_mae_db": 0.7966564594183035}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8642460730425844, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c5f30a05369b26edba6d6dfc7a22166cbba9ed3466739b5d8d527f09d43c7f32", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9787152813552982, "standard_deviation": 0.001794393889204983, "minimum": 0.975648269920408, "maximum": 0.980561516722074, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.852029611647274, "standard_deviation": 0.02042236455158735, "minimum": 1.817045703376861, "maximum": 1.871779906665313, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9314960839511848, "standard_deviation": 0.25308652044641994, "minimum": 0.5606499966779509, "maximum": 1.3114691874289648, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18288130987115064, "standard_deviation": 0.007228165981865986, "minimum": 0.17590304136296703, "maximum": 0.19464143215326082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9932195486346007, "standard_deviation": 0.0006933490086207599, "minimum": 0.9920432953508144, "maximum": 0.9941090871635343, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.337865748026307, "standard_deviation": 0.022018449110255977, "minimum": 1.3091448101676253, "maximum": 1.3717297470499046, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2441765956172057, "standard_deviation": 0.4907620512558508, "minimum": 0.4925893253592843, "maximum": 2.0560592741383785, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11913249171854856, "standard_deviation": 0.0072474247664221165, "minimum": 0.10684705225181686, "maximum": 0.12670192269807803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9703454871385826, "standard_deviation": 0.0037460569664382972, "minimum": 0.9658909846015498, "maximum": 0.9773590688440881, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9526720719863038, "standard_deviation": 0.023687857352342027, "minimum": 1.921311471127489, "maximum": 1.9842209577511123, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9187398131853617, "standard_deviation": 0.3220283623782011, "minimum": 0.521677561733713, "maximum": 1.3339186602076518, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21045706279140158, "standard_deviation": 0.013607160683148364, "minimum": 0.18426020705037124, "maximum": 0.22545742576309163, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9879990083165847, "standard_deviation": 0.002606133959934526, "minimum": 0.9830268387898218, "maximum": 0.991338374086899, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.439369451483958, "standard_deviation": 0.025529688248597797, "minimum": 1.4003505967107586, "maximum": 1.4722727883388647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8206342752334106, "standard_deviation": 0.4282600526542237, "minimum": 0.43152524178907076, "maximum": 1.8385113574135403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13271189285244034, "standard_deviation": 0.014101460206541615, "minimum": 0.11314506274416322, "maximum": 0.15518744586616007, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918710247541199, "standard_deviation": 0.0012061817974423619, "minimum": 0.9897861875850129, "maximum": 0.9936510008514243, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.475725458225286, "standard_deviation": 0.04045787256099578, "minimum": 1.4156907397529916, "maximum": 1.5298963287535159, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.36683117216137806, "standard_deviation": 0.1211219985093733, "minimum": 0.20042530769230393, "maximum": 0.5544334236317728, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10064310590857735, "standard_deviation": 0.00783761898395425, "minimum": 0.0888969135978427, "maximum": 0.11471522077633824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969299548511903, "standard_deviation": 0.0005932507145593501, "minimum": 0.9959081712566326, "maximum": 0.9978695856220374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0175352227218772, "standard_deviation": 0.03983914404196043, "minimum": 0.9751556730379717, "maximum": 1.1001839864352352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4704889471189001, "standard_deviation": 0.11341458689534893, "minimum": 0.2714523881467036, "maximum": 0.628339993983932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06222596991320434, "standard_deviation": 0.006894253981255929, "minimum": 0.051831156338916794, "maximum": 0.07510865165613802, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2583794099122373, "standard_deviation": 0.034326194477325926, "minimum": 0.1943977130444063, "maximum": 0.3040115581515949, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.789263561364248, "standard_deviation": 0.00740933391531857, "minimum": 5.780044851416663, "maximum": 5.801093012578728, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5480172370453352, "standard_deviation": 0.03227481586541084, "minimum": 0.5002489107715833, "maximum": 0.6060859567983243, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12972645479462347, "standard_deviation": 0.036973591566200496, "minimum": -0.19483065690325374, "maximum": -0.07166509429280052, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994444432653497, "standard_deviation": 0.0006353537709011858, "minimum": 0.9937128935052303, "maximum": 0.9955717320065193, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8703912139475343, "standard_deviation": 0.012115055319461174, "minimum": 0.8499052058834894, "maximum": 0.8849456620787739, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.26955931674877365, "standard_deviation": 0.09956666196706566, "minimum": 0.08727480571532402, "maximum": 0.40448365933891806, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07538678436104987, "standard_deviation": 0.00447603574553546, "minimum": 0.06757833318310551, "maximum": 0.08019462228892377, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981964479324399, "standard_deviation": 0.000290329782342995, "minimum": 0.9977031720643268, "maximum": 0.9985951439019193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5107220929175497, "standard_deviation": 0.013971453524924847, "minimum": 0.48910740053784574, "maximum": 0.5318844554264224, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4214134711672375, "standard_deviation": 0.1301734870739694, "minimum": 0.2728206524844476, "maximum": 0.6593067512893137, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04319460109546071, "standard_deviation": 0.003511558094283272, "minimum": 0.038528785486448856, "maximum": 0.04843955113729759, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0038", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.380887031555176, "predicted_mos": 3.380887031555176, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f94f8cb1448515e0e856b40c638bd7c4c3b87766b3c4ccb296eb97f1a063554e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8368490003049374, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8733712850617911, "全帯域_10ms_rms_envelope_nrmse": 0.07550546639844266, "全帯域_10ms_envelope_correlation": 0.9943590467730457, "全帯域_10ms_modulation_mae_db": 0.35134031239161495, "全帯域_40ms_log_rms_mae_db": 0.5316296356414119, "全帯域_40ms_rms_envelope_nrmse": 0.038799102776971926, "全帯域_40ms_envelope_correlation": 0.9984938801409372, "全帯域_40ms_modulation_mae_db": 0.5071562755525997, "300–1000 Hz_10ms_log_rms_mae_db": 1.432205657078764, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09339187012282175, "300–1000 Hz_10ms_envelope_correlation": 0.99297584908476, "300–1000 Hz_10ms_modulation_mae_db": 0.4028021279870355, "300–1000 Hz_40ms_log_rms_mae_db": 0.9375062202495784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05285656952960772, "300–1000 Hz_40ms_envelope_correlation": 0.9977889598157369, "300–1000 Hz_40ms_modulation_mae_db": 0.5065980730106951, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8532459046329905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1988641099981384, "1000–3000 Hz_10ms_envelope_correlation": 0.9750460360686639, "1000–3000 Hz_10ms_modulation_mae_db": 0.843020657202619, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3031375941548324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13487147808881375, "1000–3000 Hz_40ms_envelope_correlation": 0.991347247929168, "1000–3000 Hz_40ms_modulation_mae_db": 1.2354103913753478, "3000–6000 Hz_10ms_log_rms_mae_db": 1.898218820550516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19874124598821222, "3000–6000 Hz_10ms_envelope_correlation": 0.9735606299346552, "3000–6000 Hz_10ms_modulation_mae_db": 0.8264538105446354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4125681730848423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12097666785879084, "3000–6000 Hz_40ms_envelope_correlation": 0.9900113856849599, "3000–6000 Hz_40ms_modulation_mae_db": 0.6963017365901837, "mr_stft_db": 5.78138425671177, "band_mae_db": 0.23836627103054567, "rms_gain_error_db": -0.09967594611040477, "output_peak": 0.5736970316085873}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9366629305101316, "全帯域_10ms_rms_envelope_nrmse": 0.07187279621663371, "全帯域_10ms_envelope_correlation": 0.9948742347147408, "全帯域_10ms_modulation_mae_db": 0.1932303116036335, "全帯域_40ms_log_rms_mae_db": 0.5480010134502126, "全帯域_40ms_rms_envelope_nrmse": 0.03912123984717639, "全帯域_40ms_envelope_correlation": 0.9984353265829223, "全帯域_40ms_modulation_mae_db": 0.1496189333216823, "300–1000 Hz_10ms_log_rms_mae_db": 1.4246132835807648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09141481523014057, "300–1000 Hz_10ms_envelope_correlation": 0.9934276362925648, "300–1000 Hz_10ms_modulation_mae_db": 0.5877954187900191, "300–1000 Hz_40ms_log_rms_mae_db": 0.9640231290112933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05092321681288939, "300–1000 Hz_40ms_envelope_correlation": 0.998157980928066, "300–1000 Hz_40ms_modulation_mae_db": 0.4219684620195558, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8444120901451657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16509579344778355, "1000–3000 Hz_10ms_envelope_correlation": 0.9821576955024871, "1000–3000 Hz_10ms_modulation_mae_db": 0.8299988527088208, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3240636846318121, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11212679451113275, "1000–3000 Hz_40ms_envelope_correlation": 0.992926999635475, "1000–3000 Hz_40ms_modulation_mae_db": 1.0786040593369692, "3000–6000 Hz_10ms_log_rms_mae_db": 1.878717874217545, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21569366869251655, "3000–6000 Hz_10ms_envelope_correlation": 0.9687393718524516, "3000–6000 Hz_10ms_modulation_mae_db": 0.4666184786565206, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3636677369388037, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12298986173636012, "3000–6000 Hz_40ms_envelope_correlation": 0.9894480227559789, "3000–6000 Hz_40ms_modulation_mae_db": 0.8016509420679926, "mr_stft_db": 5.766199975079071, "band_mae_db": 0.21801310924657388, "rms_gain_error_db": -0.08693469776225546, "output_peak": 0.525682733989181}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8634189842208015, "全帯域_10ms_rms_envelope_nrmse": 0.07958865921189488, "全帯域_10ms_envelope_correlation": 0.9936613676137791, "全帯域_10ms_modulation_mae_db": 0.16577442175718296, "全帯域_40ms_log_rms_mae_db": 0.5138723169865109, "全帯域_40ms_rms_envelope_nrmse": 0.044666533566215, "全帯域_40ms_envelope_correlation": 0.9978964269778949, "全帯域_40ms_modulation_mae_db": 0.5932139978299842, "300–1000 Hz_10ms_log_rms_mae_db": 1.4236498393422086, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09685881305499214, "300–1000 Hz_10ms_envelope_correlation": 0.9923086046497159, "300–1000 Hz_10ms_modulation_mae_db": 0.25619081211420985, "300–1000 Hz_40ms_log_rms_mae_db": 0.988228965462998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05618647798760253, "300–1000 Hz_40ms_envelope_correlation": 0.9973049675150523, "300–1000 Hz_40ms_modulation_mae_db": 0.5152811376533624, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8437638326648484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2127438328501899, "1000–3000 Hz_10ms_envelope_correlation": 0.9744178815054932, "1000–3000 Hz_10ms_modulation_mae_db": 0.7556195087876848, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3540970421982383, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15508875143019177, "1000–3000 Hz_40ms_envelope_correlation": 0.9912187914221265, "1000–3000 Hz_40ms_modulation_mae_db": 0.5692535579589509, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8780049010288606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20480751983608098, "3000–6000 Hz_10ms_envelope_correlation": 0.972275928393144, "3000–6000 Hz_10ms_modulation_mae_db": 0.45734226814487516, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4050113318727508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12221283932467027, "3000–6000 Hz_40ms_envelope_correlation": 0.9895975109192368, "3000–6000 Hz_40ms_modulation_mae_db": 0.5176892458553729, "mr_stft_db": 5.780081541374142, "band_mae_db": 0.2613021061756106, "rms_gain_error_db": -0.06550667319150036, "output_peak": 0.5185728282837275}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8572068230060407, "全帯域_10ms_rms_envelope_nrmse": 0.07281181592458505, "全帯域_10ms_envelope_correlation": 0.9948452377137311, "全帯域_10ms_modulation_mae_db": 0.4717777286295545, "全帯域_40ms_log_rms_mae_db": 0.518237248973931, "全帯域_40ms_rms_envelope_nrmse": 0.04309898432754226, "全帯域_40ms_envelope_correlation": 0.9982153011661067, "全帯域_40ms_modulation_mae_db": 0.6556620830013441, "300–1000 Hz_10ms_log_rms_mae_db": 1.430970138917856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09481467310530871, "300–1000 Hz_10ms_envelope_correlation": 0.9928954921127437, "300–1000 Hz_10ms_modulation_mae_db": 0.7771040477072728, "300–1000 Hz_40ms_log_rms_mae_db": 0.9368622236527456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05689298444014989, "300–1000 Hz_40ms_envelope_correlation": 0.9975671234237451, "300–1000 Hz_40ms_modulation_mae_db": 0.9319108647468429, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8156124500462836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1828491551158467, "1000–3000 Hz_10ms_envelope_correlation": 0.9803840480288268, "1000–3000 Hz_10ms_modulation_mae_db": 1.0458090498718893, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3360483149536095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.125268426314843, "1000–3000 Hz_40ms_envelope_correlation": 0.9940532145190402, "1000–3000 Hz_40ms_modulation_mae_db": 1.4257733169391698, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8796529604958079, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22960642374567608, "3000–6000 Hz_10ms_envelope_correlation": 0.9648044783344089, "3000–6000 Hz_10ms_modulation_mae_db": 0.9387956912025195, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3920287389929198, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15191098426150712, "3000–6000 Hz_40ms_envelope_correlation": 0.9846480907008833, "3000–6000 Hz_40ms_modulation_mae_db": 0.4088629387740968, "mr_stft_db": 5.7821026220676375, "band_mae_db": 0.27878756684359896, "rms_gain_error_db": -0.13405737869685522, "output_peak": 0.5140489651758792}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8767199648730167, "全帯域_10ms_rms_envelope_nrmse": 0.0767224434740533, "全帯域_10ms_envelope_correlation": 0.9941719585205379, "全帯域_10ms_modulation_mae_db": 0.1585884482744004, "全帯域_40ms_log_rms_mae_db": 0.5045850817961499, "全帯域_40ms_rms_envelope_nrmse": 0.04326086973825101, "全帯域_40ms_envelope_correlation": 0.9980912611774065, "全帯域_40ms_modulation_mae_db": 0.3540225668214824, "300–1000 Hz_10ms_log_rms_mae_db": 1.4449969835021526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09207407349289179, "300–1000 Hz_10ms_envelope_correlation": 0.9930934484504073, "300–1000 Hz_10ms_modulation_mae_db": 0.36349631869921745, "300–1000 Hz_40ms_log_rms_mae_db": 1.028994741196409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055453115857498646, "300–1000 Hz_40ms_envelope_correlation": 0.9974198498060033, "300–1000 Hz_40ms_modulation_mae_db": 0.39695435646359845, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8321986168716333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19048563819479317, "1000–3000 Hz_10ms_envelope_correlation": 0.9763152229483355, "1000–3000 Hz_10ms_modulation_mae_db": 0.6719994306687341, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3334521987076648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12200207492557642, "1000–3000 Hz_40ms_envelope_correlation": 0.9924738524673472, "1000–3000 Hz_40ms_modulation_mae_db": 1.0017391667206257, "3000–6000 Hz_10ms_log_rms_mae_db": 1.901537035691857, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20049659124660654, "3000–6000 Hz_10ms_envelope_correlation": 0.9732918282877355, "3000–6000 Hz_10ms_modulation_mae_db": 0.6842301253223088, "3000–6000 Hz_40ms_log_rms_mae_db": 1.398510672254578, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1293167020268967, "3000–6000 Hz_40ms_envelope_correlation": 0.9889185545009203, "3000–6000 Hz_40ms_modulation_mae_db": 0.9272661631347047, "mr_stft_db": 5.784455626856307, "band_mae_db": 0.2157569416601728, "rms_gain_error_db": -0.09854068523634571, "output_peak": 0.5638534459183646}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8785308213214027, "全帯域_10ms_rms_envelope_nrmse": 0.08040716454875896, "全帯域_10ms_envelope_correlation": 0.9939292844045993, "全帯域_10ms_modulation_mae_db": 0.5971880223523183, "全帯域_40ms_log_rms_mae_db": 0.5520800414751676, "全帯域_40ms_rms_envelope_nrmse": 0.05029566964593681, "全帯域_40ms_envelope_correlation": 0.9978030571659032, "全帯域_40ms_modulation_mae_db": 0.9060097461732953, "300–1000 Hz_10ms_log_rms_mae_db": 1.4385814582130567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10909425057988387, "300–1000 Hz_10ms_envelope_correlation": 0.9911385846778785, "300–1000 Hz_10ms_modulation_mae_db": 0.8800818319303515, "300–1000 Hz_40ms_log_rms_mae_db": 1.0385795500346147, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07561861650041644, "300–1000 Hz_40ms_envelope_correlation": 0.9961572882564591, "300–1000 Hz_40ms_modulation_mae_db": 1.149414581486724, "1000–3000 Hz_10ms_log_rms_mae_db": 1.788217337039972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1679788773345128, "1000–3000 Hz_10ms_envelope_correlation": 0.9803474911029256, "1000–3000 Hz_10ms_modulation_mae_db": 1.2733886530896508, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2860020061754676, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10538618642983262, "1000–3000 Hz_40ms_envelope_correlation": 0.9926369349017358, "1000–3000 Hz_40ms_modulation_mae_db": 1.4441444587087329, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9121897525178515, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21372782384050987, "3000–6000 Hz_10ms_envelope_correlation": 0.9694101490961246, "3000–6000 Hz_10ms_modulation_mae_db": 0.7978063831980071, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4089071617395459, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13545555115504132, "3000–6000 Hz_40ms_envelope_correlation": 0.9875864197302241, "3000–6000 Hz_40ms_modulation_mae_db": 0.9400719773314953, "mr_stft_db": 5.7669516439138055, "band_mae_db": 0.22642814373527176, "rms_gain_error_db": -0.20585630638568614, "output_peak": 0.5632657354010622}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8967154176540786, "全帯域_10ms_rms_envelope_nrmse": 0.0733493885053513, "全帯域_10ms_envelope_correlation": 0.9946423421298062, "全帯域_10ms_modulation_mae_db": 0.15287978581578418, "全帯域_40ms_log_rms_mae_db": 0.5282827996421291, "全帯域_40ms_rms_envelope_nrmse": 0.039237654380758116, "全帯域_40ms_envelope_correlation": 0.9984100520910804, "全帯域_40ms_modulation_mae_db": 0.2984805086672479, "300–1000 Hz_10ms_log_rms_mae_db": 1.4616299243015278, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09272837809290964, "300–1000 Hz_10ms_envelope_correlation": 0.9930887443084437, "300–1000 Hz_10ms_modulation_mae_db": 0.36598636255904177, "300–1000 Hz_40ms_log_rms_mae_db": 1.0127055167999501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056475928243556565, "300–1000 Hz_40ms_envelope_correlation": 0.9974439231384196, "300–1000 Hz_40ms_modulation_mae_db": 0.41489602743124615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8920024183437876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1900478098027006, "1000–3000 Hz_10ms_envelope_correlation": 0.9774916662833391, "1000–3000 Hz_10ms_modulation_mae_db": 0.9986770863540482, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3719437638801975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1286768546681289, "1000–3000 Hz_40ms_envelope_correlation": 0.9923961005364725, "1000–3000 Hz_40ms_modulation_mae_db": 1.1262620610388339, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9302648123471662, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2143602467938806, "3000–6000 Hz_10ms_envelope_correlation": 0.9693864976776307, "3000–6000 Hz_10ms_modulation_mae_db": 0.6964008401454116, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4391213297199539, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.131608381263388, "3000–6000 Hz_40ms_envelope_correlation": 0.9886597019116896, "3000–6000 Hz_40ms_modulation_mae_db": 0.6574900662027175, "mr_stft_db": 5.78407189806603, "band_mae_db": 0.279624534641977, "rms_gain_error_db": -0.07822933757896156, "output_peak": 0.537329008735319}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8779078676897832, "全帯域_10ms_rms_envelope_nrmse": 0.07767596779191459, "全帯域_10ms_envelope_correlation": 0.9943266493777452, "全帯域_10ms_modulation_mae_db": 0.3619364624442753, "全帯域_40ms_log_rms_mae_db": 0.5037344302966186, "全帯域_40ms_rms_envelope_nrmse": 0.04656980267246933, "全帯域_40ms_envelope_correlation": 0.9981510979339007, "全帯域_40ms_modulation_mae_db": 0.5718504917598467, "300–1000 Hz_10ms_log_rms_mae_db": 1.50060255514868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10411895565581636, "300–1000 Hz_10ms_envelope_correlation": 0.9916768090483398, "300–1000 Hz_10ms_modulation_mae_db": 0.31310390218749223, "300–1000 Hz_40ms_log_rms_mae_db": 1.0348025153273388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0667830961665976, "300–1000 Hz_40ms_envelope_correlation": 0.9968833290117171, "300–1000 Hz_40ms_modulation_mae_db": 0.3919452995471171, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8163600939677227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17955300818574677, "1000–3000 Hz_10ms_envelope_correlation": 0.9802031326935251, "1000–3000 Hz_10ms_modulation_mae_db": 1.540823964781004, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3068797315205667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12672315157111422, "1000–3000 Hz_40ms_envelope_correlation": 0.9924424465723938, "1000–3000 Hz_40ms_modulation_mae_db": 2.1491435294911345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9228876663922236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2044989241152657, "3000–6000 Hz_10ms_envelope_correlation": 0.9720277454314391, "3000–6000 Hz_10ms_modulation_mae_db": 0.9869248121965409, "3000–6000 Hz_40ms_log_rms_mae_db": 1.400012428920152, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12207168466121984, "3000–6000 Hz_40ms_envelope_correlation": 0.9899787076835519, "3000–6000 Hz_40ms_modulation_mae_db": 0.6975602285854113, "mr_stft_db": 5.758997805034666, "band_mae_db": 0.30110710992582523, "rms_gain_error_db": -0.19597249995405575, "output_peak": 0.5003380669020228}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.798099688969426, "spectral_convergence": 0.22827507493236393, "spectral_convergence_status": "ok", "amplitude_floor": 1.6152428194526273e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.562387961317118, "spectral_convergence": 0.2219782495669947, "spectral_convergence_status": "ok", "amplitude_floor": 1.0855275118818902e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.983665119848767, "spectral_convergence": 0.23371319630912749, "spectral_convergence_status": "ok", "amplitude_floor": 7.682277294566263e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006930713321042831, "output_power": 0.0007094659878672826, "delta_db": 0.10153644757130284, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002559270540127681, "output_power": 0.0024833915094613834, "delta_db": -0.13071005557233115, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007234666478039212, "output_power": 0.0007285905300176561, "delta_db": 0.03065007413755473, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002266743638240336, "output_power": 0.0001880527438807483, "delta_db": -0.8112273092595748, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.48754354732022e-05, "output_power": 6.072206399509109e-05, "delta_db": -0.2873376086622321, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.432671244156477e-05, "output_power": 4.36306727268809e-05, "delta_db": -0.06873613098027852, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23836627103054567, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09967594611040477, "pitch": null, "output_rms_dbfs": -23.753205664486657, "output_peak": 0.5736970316085873, "reference_rms_dbfs": -23.65352971837625, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.78138425671177}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019088238002709922, "log_rms_mae_db": 0.8733712850617911, "rms_envelope_nrmse": 0.07550546639844266, "envelope_correlation": 0.9943590467730457, "rms_gain_error_db": -0.09967589251041009, "reference_mean_power": 0.004315280923443156, "output_mean_power": 0.004217367861394049}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8761677888507434e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007125355736933842, "output_power": 0.0006822198637517579, "reference_active": true, "delta_db": -0.18882191264875015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011942531791187452, "output_power": 0.00010609273749355428, "reference_active": true, "delta_db": -0.5141075047053494}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5392584995669358e-05, "output_power": 2.7530624743384107e-05, "reference_active": true, "delta_db": 0.35109151982074516}], "active_band_count": 3, "active_band_mae_db": 0.35134031239161495}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018007320980187037, "log_rms_mae_db": 0.5316296356414119, "rms_envelope_nrmse": 0.038799102776971926, "envelope_correlation": 0.9984938801409372, "rms_gain_error_db": -0.09967567943831462, "reference_mean_power": 0.004336982504671315, "output_mean_power": 0.00423857724502551}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7431428852385876e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006439047386291532, "output_power": 0.0006142965163165506, "reference_active": true, "delta_db": -0.20443568730527373}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.115354721039444e-05, "output_power": 4.7119899717928876e-05, "reference_active": true, "delta_db": -0.3567139746803532}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9654159892814645e-06, "output_power": 1.5755150945781175e-06, "reference_active": true, "delta_db": -0.960319164672172}], "active_band_count": 3, "active_band_mae_db": 0.5071562755525997}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017279669295308916, "log_rms_mae_db": 1.432205657078764, "rms_envelope_nrmse": 0.09339187012282175, "envelope_correlation": 0.99297584908476, "rms_gain_error_db": -0.16176291182078276, "reference_mean_power": 0.0026069520032741206, "output_mean_power": 0.002511636259920507}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2087875823984064e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005529275902426358, "output_power": 0.0005332014451630015, "reference_active": true, "delta_db": -0.15776943277214653}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648535673845476e-05, "output_power": 8.006200193832803e-05, "reference_active": true, "delta_db": -0.8103496154330634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.705863634087677e-05, "output_power": 1.8029058358483603e-05, "reference_active": true, "delta_db": 0.24028733575589672}], "active_band_count": 3, "active_band_mae_db": 0.4028021279870355}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014230358058689408, "log_rms_mae_db": 0.9375062202495784, "rms_envelope_nrmse": 0.05285656952960772, "envelope_correlation": 0.9977889598157369, "rms_gain_error_db": -0.16176293804421385, "reference_mean_power": 0.0026200630509491097, "output_mean_power": 0.002524267924396741}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1215946456987494e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005050917134712818, "output_power": 0.00048655655945610256, "reference_active": true, "delta_db": -0.16236911717453095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.798176008445663e-05, "output_power": 3.287044994705719e-05, "reference_active": true, "delta_db": -0.6276943722341057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3776930382412419e-06, "output_power": 1.1646069858356474e-06, "reference_active": true, "delta_db": -0.7297307296234488}], "active_band_count": 3, "active_band_mae_db": 0.5065980730106951}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.058549471038879e-05, "log_rms_mae_db": 1.8532459046329905, "rms_envelope_nrmse": 0.1988641099981384, "envelope_correlation": 0.9750460360686639, "rms_gain_error_db": -0.7693684051376153, "reference_mean_power": 0.0002448429483315042, "output_mean_power": 0.00020509296326958327}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7588735706755028e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.377351846027054e-05, "output_power": 7.200078319683527e-05, "reference_active": true, "delta_db": -1.1474299085792818}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.692404819271496e-06, "output_power": 7.4107044783226835e-06, "reference_active": true, "delta_db": -1.1657204991240662}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.8112011913022686e-06, "output_power": 3.6263583592613744e-06, "reference_active": true, "delta_db": -0.21591156390450905}], "active_band_count": 3, "active_band_mae_db": 0.843020657202619}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.399057463688353e-05, "log_rms_mae_db": 1.3031375941548324, "rms_envelope_nrmse": 0.13487147808881375, "envelope_correlation": 0.991347247929168, "rms_gain_error_db": -0.7693685096815804, "reference_mean_power": 0.0002460738466011548, "output_mean_power": 0.00020612402158861349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6710797473991227e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.068314304092372e-05, "output_power": 6.971687234588022e-05, "reference_active": true, "delta_db": -1.1418866872822537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.419207768284792e-06, "output_power": 3.529481026911471e-06, "reference_active": true, "delta_db": -0.976335686785482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.686031483587781e-07, "output_power": 1.8634177104115395e-07, "reference_active": true, "delta_db": -1.588008800058308}], "active_band_count": 3, "active_band_mae_db": 1.2354103913753478}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.748604525835054e-05, "log_rms_mae_db": 1.898218820550516, "rms_envelope_nrmse": 0.19874124598821222, "envelope_correlation": 0.9735606299346552, "rms_gain_error_db": -0.24776668561427867, "reference_mean_power": 6.169396290931914e-05, "output_mean_power": 5.827281517371157e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.220333375078007e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9572001812371347e-05, "output_power": 1.666418936871439e-05, "reference_active": true, "delta_db": -0.6985105509510261}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.501529531199527e-06, "output_power": 6.963487215611746e-06, "reference_active": true, "delta_db": -0.8667028513612195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.287934963899872e-07, "output_power": 1.1463941696441932e-06, "reference_active": true, "delta_db": 0.9141480293216608}], "active_band_count": 3, "active_band_mae_db": 0.8264538105446354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.636845752844561e-05, "log_rms_mae_db": 1.4125681730848423, "rms_envelope_nrmse": 0.12097666785879084, "envelope_correlation": 0.9900113856849599, "rms_gain_error_db": -0.24776729778007348, "reference_mean_power": 6.200418008997283e-05, "output_mean_power": 5.856582146284956e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.99689742660592e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0058245625049685e-05, "output_power": 1.677471006202707e-05, "reference_active": true, "delta_db": -0.7763792274137469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.381270388794365e-06, "output_power": 3.8065271145394403e-06, "reference_active": true, "delta_db": -0.6107112866584751}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.024894795639882e-07, "output_power": 8.719625208152274e-08, "reference_active": true, "delta_db": -0.7018146956983294}], "active_band_count": 3, "active_band_mae_db": 0.6963017365901837}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8600890544744458, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "68d560f6c3c2ba68d8317943a8208dd596379532bf01b815e67d308b00232e5c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9782953967666996, "standard_deviation": 0.002860841627867275, "minimum": 0.9744178815054932, "maximum": 0.9821576955024871, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8357265929640505, "standard_deviation": 0.030852519992642245, "minimum": 1.788217337039972, "maximum": 1.8920024183437876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9949171504330565, "standard_deviation": 0.29031376004610654, "minimum": 0.6719994306687341, "maximum": 1.540823964781004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18595227811621398, "standard_deviation": 0.015722793542180816, "minimum": 0.16509579344778355, "maximum": 0.2127438328501899, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9924369484979698, "standard_deviation": 0.0008917779359025886, "minimum": 0.9912187914221265, "maximum": 0.9940532145190402, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3269530420277986, "standard_deviation": 0.028135828457293052, "minimum": 1.2860020061754676, "maximum": 1.3719437638801975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2537913176962205, "standard_deviation": 0.4546399024169646, "minimum": 0.5692535579589509, "maximum": 2.1491435294911345, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12626796474245416, "standard_deviation": 0.014956538880229698, "minimum": 0.10538618642983262, "maximum": 0.15508875143019177, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9704370786259486, "standard_deviation": 0.0029419249452216907, "minimum": 0.9648044783344089, "maximum": 0.9735606299346552, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9001842279052286, "standard_deviation": 0.020509137470453673, "minimum": 1.8780049010288606, "maximum": 1.9302648123471662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7318215511763524, "standard_deviation": 0.19654746720938474, "minimum": 0.45734226814487516, "maximum": 0.9869248121965409, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21024155553234355, "standard_deviation": 0.010166958748368926, "minimum": 0.19874124598821222, "maximum": 0.22960642374567608, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9886060492359305, "standard_deviation": 0.001787124412930547, "minimum": 0.9846480907008833, "maximum": 0.9900113856849599, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4024784466904432, "standard_deviation": 0.02114419677211196, "minimum": 1.3636677369388037, "maximum": 1.4391213297199539, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7058616623177469, "standard_deviation": 0.18484249208154313, "minimum": 0.4088629387740968, "maximum": 0.9400719773314953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12956783403598426, "standard_deviation": 0.010457222614459798, "minimum": 0.12097666785879084, "maximum": 0.15191098426150712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925756460781068, "standard_deviation": 0.000798579537791754, "minimum": 0.9911385846778785, "maximum": 0.9934276362925648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4446562300106263, "standard_deviation": 0.025747206688726567, "minimum": 1.4236498393422086, "maximum": 1.50060255514868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.49332010274683, "standard_deviation": 0.22960006451873155, "minimum": 0.25619081211420985, "maximum": 0.8800818319303515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0968119786668456, "standard_deviation": 0.0064163584880417195, "minimum": 0.09141481523014057, "maximum": 0.10909425057988387, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973404277368999, "standard_deviation": 0.0006033241305866802, "minimum": 0.9961572882564591, "maximum": 0.998157980928066, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.992712857716866, "standard_deviation": 0.04242917829702623, "minimum": 0.9368622236527456, "maximum": 1.0385795500346147, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5911211002948927, "standard_deviation": 0.2872939589952456, "minimum": 0.3919452995471171, "maximum": 1.149414581486724, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05889875069228985, "standard_deviation": 0.008202402954359178, "minimum": 0.05092321681288939, "maximum": 0.07561861650041644, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25242322290744695, "standard_deviation": 0.03226178534264933, "minimum": 0.2157569416601728, "maximum": 0.30110710992582523, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.775530671137929, "standard_deviation": 0.00989061657514262, "minimum": 5.758997805034666, "maximum": 5.784455626856307, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5370984770017679, "standard_deviation": 0.026981972058391923, "minimum": 0.5003380669020228, "maximum": 0.5736970316085873, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12059669061450812, "standard_deviation": 0.05347774189621794, "minimum": -0.20585630638568614, "maximum": -0.06550667319150036, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9943512651559981, "standard_deviation": 0.0004291312794105888, "minimum": 0.9936613676137791, "maximum": 0.9948742347147408, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8825667617921308, "standard_deviation": 0.024764836101887077, "minimum": 0.8572068230060407, "maximum": 0.9366629305101316, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3065894366585955, "standard_deviation": 0.16690919384653302, "minimum": 0.15287978581578418, "maximum": 0.5971880223523183, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0759917127589543, "standard_deviation": 0.003164545804277117, "minimum": 0.07187279621663371, "maximum": 0.08040716454875896, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981870504045189, "standard_deviation": 0.0002531781826619998, "minimum": 0.9978030571659032, "maximum": 0.9984938801409372, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5250528210327664, "standard_deviation": 0.018344462997001248, "minimum": 0.5037344302966186, "maximum": 0.5520800414751676, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5045018253909354, "standard_deviation": 0.23528142107624997, "minimum": 0.1496189333216823, "maximum": 0.9060097461732953, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04313123211941511, "standard_deviation": 0.004062639169566023, "minimum": 0.038799102776971926, "maximum": 0.05029566964593681, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_7729-102255-0042", "language": "en", "style": "normal", "speaker": "librispeech_7729", "source_utterance_id": "7729-102255-0042", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_7729-102255-0042.wav", "source_file_sha256": "4f29a2ccd1d3d8cb85f33e7a8f2e804212c0c63122cb7d938ed4e59fc57e2376", "decoded_pcm_sha256": "3e552debe1c0d524847e6550ec6f2d9550fc271ad1a9ac47e3543e154db1f552", "original_source_sha256": "258a5f83b294976c312b375ef48ed79028fb41459458f6ece40b90fab7bce317", "corpus_entry_id": "librispeech_7729-102255-0042"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.228043079376221, "predicted_mos": 4.228043079376221, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "771bae28a9ddc1de4e6b039e3c3bf8346c4942f076859b624472c4dc44a30465", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4838812979978684, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_7729-102255-0042", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "771bae28a9ddc1de4e6b039e3c3bf8346c4942f076859b624472c4dc44a30465", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "8e0eb42509778a5dd3e14ba6fb8dbb8055bbeca1cb9858080ff50cf3162ecbc6"}, "rows": [{"case": "librispeech_7729-102255-0042", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.77156138420105, "predicted_mos": 2.77156138420105, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4e33d50b4dde2b88e9c04c59c7297cb62b623ad84749fc3d698751d660af6bf5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.0218310810014373, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_7729-102255-0042", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "4e33d50b4dde2b88e9c04c59c7297cb62b623ad84749fc3d698751d660af6bf5", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0316491564463814, "全帯域_10ms_rms_envelope_nrmse": 0.1078450159022606, "全帯域_10ms_envelope_correlation": 0.9888792622664058, "全帯域_10ms_modulation_mae_db": 0.3158714884255038, "全帯域_40ms_log_rms_mae_db": 0.5871278600476428, "全帯域_40ms_rms_envelope_nrmse": 0.06656698401215559, "全帯域_40ms_envelope_correlation": 0.9955820720404198, "全帯域_40ms_modulation_mae_db": 0.2874379024329135, "300–1000 Hz_10ms_log_rms_mae_db": 1.3738829458161927, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16631006594779965, "300–1000 Hz_10ms_envelope_correlation": 0.9776526400787292, "300–1000 Hz_10ms_modulation_mae_db": 0.2957473936356065, "300–1000 Hz_40ms_log_rms_mae_db": 0.8177021889903923, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1145018835452634, "300–1000 Hz_40ms_envelope_correlation": 0.9892708986311325, "300–1000 Hz_40ms_modulation_mae_db": 0.18218116634792825, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9546458182152646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1972907987414528, "1000–3000 Hz_10ms_envelope_correlation": 0.9718529172589004, "1000–3000 Hz_10ms_modulation_mae_db": 0.7788624121727853, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2867375401925225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12046671790203554, "1000–3000 Hz_40ms_envelope_correlation": 0.9906942887720972, "1000–3000 Hz_40ms_modulation_mae_db": 0.6752295878379675, "3000–6000 Hz_10ms_log_rms_mae_db": 2.052384543549282, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2503815699950419, "3000–6000 Hz_10ms_envelope_correlation": 0.9575863012164471, "3000–6000 Hz_10ms_modulation_mae_db": 1.7460698318901793, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7614977393940179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18766778795412004, "3000–6000 Hz_40ms_envelope_correlation": 0.9757788383612684, "3000–6000 Hz_40ms_modulation_mae_db": 3.0003215175343065, "mr_stft_db": 6.174367429925327, "band_mae_db": 0.31389087968923707, "rms_gain_error_db": -0.1313595315088841, "output_peak": 0.6267188387334389}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0334100447481565, "全帯域_10ms_rms_envelope_nrmse": 0.09529080264614005, "全帯域_10ms_envelope_correlation": 0.9912417108471627, "全帯域_10ms_modulation_mae_db": 0.13251593952360788, "全帯域_40ms_log_rms_mae_db": 0.5268210520742622, "全帯域_40ms_rms_envelope_nrmse": 0.052931558626915655, "全帯域_40ms_envelope_correlation": 0.9971107144769712, "全帯域_40ms_modulation_mae_db": 0.17631381310322325, "300–1000 Hz_10ms_log_rms_mae_db": 1.264501152338722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1209943935809065, "300–1000 Hz_10ms_envelope_correlation": 0.9879955156667202, "300–1000 Hz_10ms_modulation_mae_db": 0.20837148319845303, "300–1000 Hz_40ms_log_rms_mae_db": 0.7466090467509798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0650740773836972, "300–1000 Hz_40ms_envelope_correlation": 0.9963919863650758, "300–1000 Hz_40ms_modulation_mae_db": 0.49246142278724186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8918866669539187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20085925654927728, "1000–3000 Hz_10ms_envelope_correlation": 0.971025090827275, "1000–3000 Hz_10ms_modulation_mae_db": 0.9050327862822475, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2507965185071848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13448825020795319, "1000–3000 Hz_40ms_envelope_correlation": 0.9880217776928442, "1000–3000 Hz_40ms_modulation_mae_db": 0.6908047867638566, "3000–6000 Hz_10ms_log_rms_mae_db": 2.127518453428907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26676052615895124, "3000–6000 Hz_10ms_envelope_correlation": 0.9517072767131347, "3000–6000 Hz_10ms_modulation_mae_db": 1.8849214249595685, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8639717076319475, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19662661081803762, "3000–6000 Hz_40ms_envelope_correlation": 0.9731802666152417, "3000–6000 Hz_40ms_modulation_mae_db": 2.939470354360919, "mr_stft_db": 6.181285211483111, "band_mae_db": 0.24520277773945356, "rms_gain_error_db": -0.045227762906419534, "output_peak": 0.7107929744604389}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9912268252856818, "全帯域_10ms_rms_envelope_nrmse": 0.0991837452432422, "全帯域_10ms_envelope_correlation": 0.9907415423084593, "全帯域_10ms_modulation_mae_db": 0.2562408086624955, "全帯域_40ms_log_rms_mae_db": 0.5501522948284677, "全帯域_40ms_rms_envelope_nrmse": 0.05543981162253139, "全帯域_40ms_envelope_correlation": 0.9971867102957785, "全帯域_40ms_modulation_mae_db": 0.11562715174897116, "300–1000 Hz_10ms_log_rms_mae_db": 1.2853826472862595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13444511906044757, "300–1000 Hz_10ms_envelope_correlation": 0.9858401991764418, "300–1000 Hz_10ms_modulation_mae_db": 0.36562967182931416, "300–1000 Hz_40ms_log_rms_mae_db": 0.7662827103351489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07737756077930628, "300–1000 Hz_40ms_envelope_correlation": 0.9958523917277827, "300–1000 Hz_40ms_modulation_mae_db": 0.5328153176231997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9497695342830337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20930115942461794, "1000–3000 Hz_10ms_envelope_correlation": 0.9686964882079511, "1000–3000 Hz_10ms_modulation_mae_db": 0.6718018036975258, "1000–3000 Hz_40ms_log_rms_mae_db": 1.234572086504249, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13313644638819297, "1000–3000 Hz_40ms_envelope_correlation": 0.9890252883604649, "1000–3000 Hz_40ms_modulation_mae_db": 0.6004290194259669, "3000–6000 Hz_10ms_log_rms_mae_db": 2.15268050242441, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26072635039556696, "3000–6000 Hz_10ms_envelope_correlation": 0.954154801873616, "3000–6000 Hz_10ms_modulation_mae_db": 1.7462442851313316, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8546615611614246, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1984764913400022, "3000–6000 Hz_40ms_envelope_correlation": 0.9725032952258139, "3000–6000 Hz_40ms_modulation_mae_db": 2.57916494299784, "mr_stft_db": 6.186131971589821, "band_mae_db": 0.350115340012691, "rms_gain_error_db": -0.1774230826045985, "output_peak": 0.6024480491812773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.99254827372966, "全帯域_10ms_rms_envelope_nrmse": 0.10325048456205867, "全帯域_10ms_envelope_correlation": 0.9897380621103349, "全帯域_10ms_modulation_mae_db": 0.6867109895508204, "全帯域_40ms_log_rms_mae_db": 0.5617362264716781, "全帯域_40ms_rms_envelope_nrmse": 0.05474839274441555, "全帯域_40ms_envelope_correlation": 0.9969048896655238, "全帯域_40ms_modulation_mae_db": 0.5795639734021226, "300–1000 Hz_10ms_log_rms_mae_db": 1.3377895605146832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1404227923825348, "300–1000 Hz_10ms_envelope_correlation": 0.9837627221689678, "300–1000 Hz_10ms_modulation_mae_db": 0.6109967464240598, "300–1000 Hz_40ms_log_rms_mae_db": 0.8027935778900007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07847728920180579, "300–1000 Hz_40ms_envelope_correlation": 0.9946441227715036, "300–1000 Hz_40ms_modulation_mae_db": 0.5330186250372752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.91158317821349, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19973778536950212, "1000–3000 Hz_10ms_envelope_correlation": 0.9702151227730068, "1000–3000 Hz_10ms_modulation_mae_db": 0.8279009779505516, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2112781954428988, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12309675138219385, "1000–3000 Hz_40ms_envelope_correlation": 0.98877150430562, "1000–3000 Hz_40ms_modulation_mae_db": 0.7023414220997367, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1476686473098834, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2765512608279537, "3000–6000 Hz_10ms_envelope_correlation": 0.9480584434766158, "3000–6000 Hz_10ms_modulation_mae_db": 1.990019140219012, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8544579933629384, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20332974332860837, "3000–6000 Hz_40ms_envelope_correlation": 0.9720014596635954, "3000–6000 Hz_40ms_modulation_mae_db": 2.974212706934408, "mr_stft_db": 6.176979730674831, "band_mae_db": 0.25654898697333456, "rms_gain_error_db": -0.031113938663374485, "output_peak": 0.6691609735161648}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0275845369692702, "全帯域_10ms_rms_envelope_nrmse": 0.10265838415941758, "全帯域_10ms_envelope_correlation": 0.9898582803878608, "全帯域_10ms_modulation_mae_db": 0.6519165640819639, "全帯域_40ms_log_rms_mae_db": 0.5610172153231308, "全帯域_40ms_rms_envelope_nrmse": 0.055567411595561056, "全帯域_40ms_envelope_correlation": 0.9968010037575076, "全帯域_40ms_modulation_mae_db": 0.488507170964215, "300–1000 Hz_10ms_log_rms_mae_db": 1.3395428047420963, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1422840452744487, "300–1000 Hz_10ms_envelope_correlation": 0.9833592461576782, "300–1000 Hz_10ms_modulation_mae_db": 0.556886740606213, "300–1000 Hz_40ms_log_rms_mae_db": 0.836768052925472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07790468597868712, "300–1000 Hz_40ms_envelope_correlation": 0.9946857956019546, "300–1000 Hz_40ms_modulation_mae_db": 0.2837432370200706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8864070860178181, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20121136534442635, "1000–3000 Hz_10ms_envelope_correlation": 0.9703326823381159, "1000–3000 Hz_10ms_modulation_mae_db": 1.104129366721214, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1828287268033053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11946026642528071, "1000–3000 Hz_40ms_envelope_correlation": 0.9906181901351659, "1000–3000 Hz_40ms_modulation_mae_db": 1.0971475931958317, "3000–6000 Hz_10ms_log_rms_mae_db": 2.107695892117394, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2411730989484071, "3000–6000 Hz_10ms_envelope_correlation": 0.9607477126479141, "3000–6000 Hz_10ms_modulation_mae_db": 1.7936107929003293, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7991447225741957, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18544179256757853, "3000–6000 Hz_40ms_envelope_correlation": 0.9763866467347152, "3000–6000 Hz_40ms_modulation_mae_db": 2.6096299047354288, "mr_stft_db": 6.169326756268066, "band_mae_db": 0.23203300995933232, "rms_gain_error_db": -0.015950538592426775, "output_peak": 0.6048455785509151}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9570812358171202, "全帯域_10ms_rms_envelope_nrmse": 0.09699504076243311, "全帯域_10ms_envelope_correlation": 0.9912148803138722, "全帯域_10ms_modulation_mae_db": 0.3626841104527256, "全帯域_40ms_log_rms_mae_db": 0.5605270998582708, "全帯域_40ms_rms_envelope_nrmse": 0.057718161503060375, "全帯域_40ms_envelope_correlation": 0.9969868467941478, "全帯域_40ms_modulation_mae_db": 0.5520338907471142, "300–1000 Hz_10ms_log_rms_mae_db": 1.3355713606910573, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1366174360165828, "300–1000 Hz_10ms_envelope_correlation": 0.985485447670683, "300–1000 Hz_10ms_modulation_mae_db": 0.651535190187538, "300–1000 Hz_40ms_log_rms_mae_db": 0.8723462640611565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08128549476992102, "300–1000 Hz_40ms_envelope_correlation": 0.9954776204363621, "300–1000 Hz_40ms_modulation_mae_db": 0.8560426955690396, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9237056543621107, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1912535433864789, "1000–3000 Hz_10ms_envelope_correlation": 0.9735570650420085, "1000–3000 Hz_10ms_modulation_mae_db": 0.9136557201590428, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2218059014583877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11844426275504466, "1000–3000 Hz_40ms_envelope_correlation": 0.9909082854676267, "1000–3000 Hz_40ms_modulation_mae_db": 0.8675210087752485, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1808437411291988, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26722749018234676, "3000–6000 Hz_10ms_envelope_correlation": 0.9516654762130219, "3000–6000 Hz_10ms_modulation_mae_db": 1.908727614086482, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9047845008136302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2073460282640427, "3000–6000 Hz_40ms_envelope_correlation": 0.9706881956394865, "3000–6000 Hz_40ms_modulation_mae_db": 2.7115830102049885, "mr_stft_db": 6.165755719575962, "band_mae_db": 0.30078005088965865, "rms_gain_error_db": -0.19300634643695364, "output_peak": 0.6681898894140221}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9871747014089857, "全帯域_10ms_rms_envelope_nrmse": 0.09345656844142777, "全帯域_10ms_envelope_correlation": 0.9916149376747686, "全帯域_10ms_modulation_mae_db": 0.545108940338041, "全帯域_40ms_log_rms_mae_db": 0.5393481687853826, "全帯域_40ms_rms_envelope_nrmse": 0.047800148125985265, "全帯域_40ms_envelope_correlation": 0.9977319928847271, "全帯域_40ms_modulation_mae_db": 0.7365379615697369, "300–1000 Hz_10ms_log_rms_mae_db": 1.2981181191651086, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12590628609315305, "300–1000 Hz_10ms_envelope_correlation": 0.9870654486734874, "300–1000 Hz_10ms_modulation_mae_db": 0.8798621272461299, "300–1000 Hz_40ms_log_rms_mae_db": 0.7926662838572329, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07067315218497881, "300–1000 Hz_40ms_envelope_correlation": 0.9958515209766413, "300–1000 Hz_40ms_modulation_mae_db": 1.1242973024001683, "1000–3000 Hz_10ms_log_rms_mae_db": 1.922302146320812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2014501016354518, "1000–3000 Hz_10ms_envelope_correlation": 0.9697178751959233, "1000–3000 Hz_10ms_modulation_mae_db": 0.6488383143235096, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2250775836339425, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12155721151009459, "1000–3000 Hz_40ms_envelope_correlation": 0.9891389045649871, "1000–3000 Hz_40ms_modulation_mae_db": 0.37277363054937235, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1275165289714835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24636003196947906, "3000–6000 Hz_10ms_envelope_correlation": 0.9589267504576237, "3000–6000 Hz_10ms_modulation_mae_db": 1.778812861932978, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8282892939098678, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19174090732547858, "3000–6000 Hz_40ms_envelope_correlation": 0.9744394293670624, "3000–6000 Hz_40ms_modulation_mae_db": 2.879963672061475, "mr_stft_db": 6.180208455179071, "band_mae_db": 0.3397415400945012, "rms_gain_error_db": -0.09418189678501242, "output_peak": 0.7014949713752271}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9480521623889507, "全帯域_10ms_rms_envelope_nrmse": 0.08431922593047443, "全帯域_10ms_envelope_correlation": 0.9932272525249451, "全帯域_10ms_modulation_mae_db": 0.36344973665029706, "全帯域_40ms_log_rms_mae_db": 0.525266545578855, "全帯域_40ms_rms_envelope_nrmse": 0.04361358346612015, "全帯域_40ms_envelope_correlation": 0.9981704439722542, "全帯域_40ms_modulation_mae_db": 0.5446600490462278, "300–1000 Hz_10ms_log_rms_mae_db": 1.2911396594108275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12716044473056176, "300–1000 Hz_10ms_envelope_correlation": 0.9870829541023517, "300–1000 Hz_10ms_modulation_mae_db": 0.5544200728412371, "300–1000 Hz_40ms_log_rms_mae_db": 0.763118198768823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07509196061811205, "300–1000 Hz_40ms_envelope_correlation": 0.9956795029024782, "300–1000 Hz_40ms_modulation_mae_db": 0.7928623210759532, "1000–3000 Hz_10ms_log_rms_mae_db": 1.934024057948758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19149094012213713, "1000–3000 Hz_10ms_envelope_correlation": 0.9736442254788695, "1000–3000 Hz_10ms_modulation_mae_db": 0.7269658868743117, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2556481854573462, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1330387964718501, "1000–3000 Hz_40ms_envelope_correlation": 0.9878821760920856, "1000–3000 Hz_40ms_modulation_mae_db": 0.8314487970045064, "3000–6000 Hz_10ms_log_rms_mae_db": 2.135258665045449, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25135377037175877, "3000–6000 Hz_10ms_envelope_correlation": 0.9573442286664577, "3000–6000 Hz_10ms_modulation_mae_db": 1.761137027480428, "3000–6000 Hz_40ms_log_rms_mae_db": 1.822913919694887, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19120064744818804, "3000–6000 Hz_40ms_envelope_correlation": 0.9746517517369164, "3000–6000 Hz_40ms_modulation_mae_db": 3.0348662434133113, "mr_stft_db": 6.163430956239568, "band_mae_db": 0.21141923721421751, "rms_gain_error_db": -0.1092795768547063, "output_peak": 0.6329053837670663}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.107642311213053, "spectral_convergence": 0.3157295093083662, "spectral_convergence_status": "ok", "amplitude_floor": 2.172633843844252e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.9498654584247115, "spectral_convergence": 0.308316978026789, "spectral_convergence_status": "ok", "amplitude_floor": 1.076658885105748e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.465594520138217, "spectral_convergence": 0.33901452880923627, "spectral_convergence_status": "ok", "amplitude_floor": 9.276051433709775e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006687421878497317, "output_power": 0.0006628249680797721, "delta_db": -0.038598621162446875, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0031496712549154435, "output_power": 0.0030755848291104644, "delta_db": -0.10337516924446294, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007488394009410867, "output_power": 0.0007201388703629092, "delta_db": -0.16972434052525961, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002473939509459554, "output_power": 0.00022205188180291752, "delta_db": -0.46934618444152204, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011342358064728732, "output_power": 9.538755383275278e-05, "delta_db": -0.7521164166543666, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.620279276522665e-05, "output_power": 5.008257871057917e-05, "delta_db": 0.3501845461073643, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.31389087968923707, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1313595315088841, "pitch": null, "output_rms_dbfs": -23.164063217695276, "output_peak": 0.6267188387334389, "reference_rms_dbfs": -23.03270368618639, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.174367429925327}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023158079660483383, "log_rms_mae_db": 1.0316491564463814, "rms_envelope_nrmse": 0.1078450159022606, "envelope_correlation": 0.9888792622664058, "rms_gain_error_db": -0.1309267704680247, "reference_mean_power": 0.004973550834679421, "output_mean_power": 0.00482585072673356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9210188012408756e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006213961937331268, "output_power": 0.0006028846400720695, "reference_active": true, "delta_db": -0.13134369407195695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026574930428793645, "output_power": 0.0003050257238962953, "reference_active": true, "delta_db": 0.5986433023241695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2739241814293584e-05, "output_power": 3.4421624651284793e-05, "reference_active": true, "delta_db": 0.21762746888038487}], "active_band_count": 3, "active_band_mae_db": 0.3158714884255038}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000211279420280134, "log_rms_mae_db": 0.5871278600476428, "rms_envelope_nrmse": 0.06656698401215559, "envelope_correlation": 0.9955820720404198, "rms_gain_error_db": -0.13082350150014818, "reference_mean_power": 0.0049805093532051875, "output_mean_power": 0.004832717511029132}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7395423077013874e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005644300449164411, "output_power": 0.0005577590906017772, "reference_active": true, "delta_db": -0.05163466092577161}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013986675109572895, "output_power": 0.00016438435805452307, "reference_active": true, "delta_db": 0.7014600304660014}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.497453327177472e-06, "output_power": 2.561057223934781e-06, "reference_active": true, "delta_db": 0.1092190159069675}], "active_band_count": 3, "active_band_mae_db": 0.2874379024329135}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020019856693346264, "log_rms_mae_db": 1.3738829458161927, "rms_envelope_nrmse": 0.16631006594779965, "envelope_correlation": 0.9776526400787292, "rms_gain_error_db": -0.3250084315167402, "reference_mean_power": 0.0028106313477950454, "output_mean_power": 0.0026079727100948224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3651110338311905e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046335294017358356, "output_power": 0.00042527591166434755, "reference_active": true, "delta_db": -0.37241138902869425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024122957535154213, "output_power": 0.0002349596026156313, "reference_active": true, "delta_db": -0.11437353327878258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.948878321904196e-05, "output_power": 3.233721236142055e-05, "reference_active": true, "delta_db": 0.40045725859934267}], "active_band_count": 3, "active_band_mae_db": 0.2957473936356065}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018155803316515472, "log_rms_mae_db": 0.8177021889903923, "rms_envelope_nrmse": 0.1145018835452634, "envelope_correlation": 0.9892708986311325, "rms_gain_error_db": -0.32613022270471365, "reference_mean_power": 0.002819303754025505, "output_mean_power": 0.0026153441626505193}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2479023732657914e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004464610508434935, "output_power": 0.00041525088175187994, "reference_active": true, "delta_db": -0.31473013815373424}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001291081626943261, "output_power": 0.000132255928719741, "reference_active": true, "delta_db": 0.10461448701919182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4462923050468237e-06, "output_power": 2.3756828461547646e-06, "reference_active": true, "delta_db": -0.12719887387085865}], "active_band_count": 3, "active_band_mae_db": 0.18218116634792825}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.07685175719321e-05, "log_rms_mae_db": 1.9546458182152646, "rms_envelope_nrmse": 0.1972907987414528, "envelope_correlation": 0.9718529172589004, "rms_gain_error_db": -0.5182570711213158, "reference_mean_power": 0.0002950078039052426, "output_mean_power": 0.00026182300165882103}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.385626430084388e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.894665895613538e-05, "output_power": 8.848844785567533e-05, "reference_active": true, "delta_db": -0.48514556989907026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.27901607459427e-05, "output_power": 1.9203029381823017e-05, "reference_active": true, "delta_db": -0.7437764198900207}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.304247232293777e-06, "output_power": 5.554739971426715e-06, "reference_active": true, "delta_db": 1.1076652467292645}], "active_band_count": 3, "active_band_mae_db": 0.7788624121727853}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.03385602019198e-05, "log_rms_mae_db": 1.2867375401925225, "rms_envelope_nrmse": 0.12046671790203554, "envelope_correlation": 0.9906942887720972, "rms_gain_error_db": -0.5146450178665123, "reference_mean_power": 0.0002923716436469955, "output_mean_power": 0.00025969928113074743}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2489747353423383e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.236418785147827e-05, "output_power": 8.410802994798965e-05, "reference_active": true, "delta_db": -0.40666155213137883}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2735133445304299e-05, "output_power": 1.1035968766432714e-05, "reference_active": true, "delta_db": -0.6219303739968953}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6865666311394477e-07, "output_power": 3.379926840687231e-07, "reference_active": true, "delta_db": 0.9970968373856282}], "active_band_count": 3, "active_band_mae_db": 0.6752295878379675}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.524569706134418e-05, "log_rms_mae_db": 2.052384543549282, "rms_envelope_nrmse": 0.2503815699950419, "envelope_correlation": 0.9575863012164471, "rms_gain_error_db": -0.2838406360876601, "reference_mean_power": 6.736853190004857e-05, "output_mean_power": 6.310634387835234e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.630234817861519e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0381998821210514e-05, "output_power": 1.9920182482670494e-05, "reference_active": true, "delta_db": -0.09953459592125763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.955680541944287e-06, "output_power": 4.397848055722571e-06, "reference_active": true, "delta_db": 0.4601900952927391}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.594012590914479e-07, "output_power": 1.936414172320496e-06, "reference_active": true, "delta_db": 4.6784848044565415}], "active_band_count": 3, "active_band_mae_db": 1.7460698318901793}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.8717196486707034e-05, "log_rms_mae_db": 1.7614977393940179, "rms_envelope_nrmse": 0.18766778795412004, "envelope_correlation": 0.9757788383612684, "rms_gain_error_db": -0.28469392809873106, "reference_mean_power": 6.764325505404543e-05, "output_mean_power": 6.335123786010445e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.430902640057432e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9875215171126693e-05, "output_power": 2.0856999152208315e-05, "reference_active": true, "delta_db": 0.20939984416373253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8203988319997446e-06, "output_power": 2.299697396080936e-06, "reference_active": true, "delta_db": 1.0150414528715657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.685144573008789e-08, "output_power": 2.208548068571059e-07, "reference_active": true, "delta_db": 7.776523255567621}], "active_band_count": 3, "active_band_mae_db": 3.0003215175343065}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.794908164951432, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bafedb6438a33290e3262dd8700ba9654a072dd083a060a9ca504bf79ddd27c4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9711301833902564, "standard_deviation": 0.001778216658990171, "minimum": 0.9686964882079511, "maximum": 0.9736442254788695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9217905177894008, "standard_deviation": 0.024687482272383075, "minimum": 1.8864070860178181, "maximum": 1.9546458182152646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8221484085226486, "standard_deviation": 0.1504106954984122, "minimum": 0.6488383143235096, "maximum": 1.104129366721214, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19907436882166804, "standard_deviation": 0.0058588013870839025, "minimum": 0.1912535433864789, "maximum": 0.20930115942461794, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893825519238615, "standard_deviation": 0.0012093996633618974, "minimum": 0.9878821760920856, "maximum": 0.9909082854676267, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2335930922499796, "standard_deviation": 0.03138266801163986, "minimum": 1.1828287268033053, "maximum": 1.2867375401925225, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7297119807065608, "standard_deviation": 0.2116941714479005, "minimum": 0.37277363054937235, "maximum": 1.0971475931958317, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12546108788033072, "standard_deviation": 0.00685401048038182, "minimum": 0.11844426275504466, "maximum": 0.13448825020795319, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9550238739081038, "standard_deviation": 0.004334183306235287, "minimum": 0.9480584434766158, "maximum": 0.9607477126479141, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1289458717470007, "standard_deviation": 0.03772909899589433, "minimum": 2.052384543549282, "maximum": 2.1808437411291988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.8261928723250387, "standard_deviation": 0.09059349628176136, "minimum": 1.7460698318901793, "maximum": 1.990019140219012, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25756676235618825, "standard_deviation": 0.012146842025507427, "minimum": 0.2411730989484071, "maximum": 0.2765512608279537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9737037354180125, "standard_deviation": 0.0019512027436740028, "minimum": 0.9706881956394865, "maximum": 0.9763866467347152, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8362151798178636, "standard_deviation": 0.04365760241577655, "minimum": 1.7614977393940179, "maximum": 1.9047845008136302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.8411515440303345, "standard_deviation": 0.1815669737050269, "minimum": 2.57916494299784, "maximum": 3.0348662434133113, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.195228751130757, "standard_deviation": 0.00761858351710705, "minimum": 0.18544179256757853, "maximum": 0.2073460282640427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9847805217118825, "standard_deviation": 0.0033023962270980396, "minimum": 0.9776526400787292, "maximum": 0.9879955156667202, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3157410312456186, "standard_deviation": 0.03642867287830395, "minimum": 1.264501152338722, "maximum": 1.3738829458161927, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5154311782460689, "standard_deviation": 0.2166935976697511, "minimum": 0.20837148319845303, "maximum": 0.8798621272461299, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13676757288580438, "standard_deviation": 0.014045970311396093, "minimum": 0.1209943935809065, "maximum": 0.16631006594779965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947317299266163, "standard_deviation": 0.0022850869314380655, "minimum": 0.9892708986311325, "maximum": 0.9963919863650758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7997857904474008, "standard_deviation": 0.0419322179098224, "minimum": 0.7466090467509798, "maximum": 0.8723462640611565, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5996777609826096, "standard_deviation": 0.31002219207596754, "minimum": 0.18218116634792825, "maximum": 1.1242973024001683, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08004826305772145, "standard_deviation": 0.014829806441122724, "minimum": 0.0650740773836972, "maximum": 0.1145018835452634, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.28121647782155323, "standard_deviation": 0.05184968116666078, "minimum": 0.21141923721421751, "maximum": 0.350115340012691, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.174685778866969, "standard_deviation": 0.007978251400150724, "minimum": 6.163430956239568, "maximum": 6.186131971589821, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6520695823748188, "standard_deviation": 0.0416543868572155, "minimum": 0.6024480491812773, "maximum": 0.7107929744604389, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09969283429404696, "standard_deviation": 0.0660700276168865, "minimum": -0.19300634643695364, "maximum": -0.015950538592426775, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9908144910542263, "standard_deviation": 0.0013431535976454797, "minimum": 0.9888792622664058, "maximum": 0.9932272525249451, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9960908670992759, "standard_deviation": 0.03293747229381426, "minimum": 0.9480521623889507, "maximum": 1.0334100447481565, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4143123222106819, "standard_deviation": 0.19533979272171262, "minimum": 0.13251593952360788, "maximum": 0.6867109895508204, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0978749084559318, "standard_deviation": 0.00720738293809875, "minimum": 0.08431922593047443, "maximum": 0.1078450159022606, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970593342359163, "standard_deviation": 0.0007544198223666005, "minimum": 0.9955820720404198, "maximum": 0.9981704439722542, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5514995578709612, "standard_deviation": 0.020677519105242237, "minimum": 0.525266545578855, "maximum": 0.5871278600476428, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43508523912681557, "standard_deviation": 0.21748603697779947, "minimum": 0.11562715174897116, "maximum": 0.7365379615697369, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05429825646209313, "standard_deviation": 0.006799048922439045, "minimum": 0.04361358346612015, "maximum": 0.06656698401215559, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0042", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.794076681137085, "predicted_mos": 2.794076681137085, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d68651149df712c83c8c8ad07d089e9174e14a853f3fec1908c7ad042552ff54", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6720695002004504, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0256961121283632, "全帯域_10ms_rms_envelope_nrmse": 0.10693707066581161, "全帯域_10ms_envelope_correlation": 0.9890646226630039, "全帯域_10ms_modulation_mae_db": 0.23004703660734727, "全帯域_40ms_log_rms_mae_db": 0.5897507703058551, "全帯域_40ms_rms_envelope_nrmse": 0.06549332415726364, "全帯域_40ms_envelope_correlation": 0.9957333339066095, "全帯域_40ms_modulation_mae_db": 0.3604231162681359, "300–1000 Hz_10ms_log_rms_mae_db": 1.360204990121563, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16746527693557953, "300–1000 Hz_10ms_envelope_correlation": 0.9773189267447414, "300–1000 Hz_10ms_modulation_mae_db": 0.4987383069469913, "300–1000 Hz_40ms_log_rms_mae_db": 0.8227525785822739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11226369171916073, "300–1000 Hz_40ms_envelope_correlation": 0.9897374793075049, "300–1000 Hz_40ms_modulation_mae_db": 0.23621431616309332, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8980697989410735, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19628606494031375, "1000–3000 Hz_10ms_envelope_correlation": 0.9729901232238469, "1000–3000 Hz_10ms_modulation_mae_db": 0.8804759792087268, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2123655413890686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12267757692882461, "1000–3000 Hz_40ms_envelope_correlation": 0.9912969180527688, "1000–3000 Hz_40ms_modulation_mae_db": 0.7945004340083729, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9853289056199848, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2477728711506788, "3000–6000 Hz_10ms_envelope_correlation": 0.9584317434694072, "3000–6000 Hz_10ms_modulation_mae_db": 1.757441571835345, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6859864649191338, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18597602863177629, "3000–6000 Hz_40ms_envelope_correlation": 0.976082288771298, "3000–6000 Hz_40ms_modulation_mae_db": 3.0326718500682186, "mr_stft_db": 6.090454657357412, "band_mae_db": 0.3409395377531003, "rms_gain_error_db": -0.1333031874294607, "output_peak": 0.6270973554460322}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.037422188195346, "全帯域_10ms_rms_envelope_nrmse": 0.09483862544733246, "全帯域_10ms_envelope_correlation": 0.9913333272596127, "全帯域_10ms_modulation_mae_db": 0.1482320117960724, "全帯域_40ms_log_rms_mae_db": 0.5278422382854729, "全帯域_40ms_rms_envelope_nrmse": 0.05268093832735471, "全帯域_40ms_envelope_correlation": 0.9971639159218865, "全帯域_40ms_modulation_mae_db": 0.11508701977795771, "300–1000 Hz_10ms_log_rms_mae_db": 1.277437112376503, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11970895527461073, "300–1000 Hz_10ms_envelope_correlation": 0.9882839026419611, "300–1000 Hz_10ms_modulation_mae_db": 0.22597789876543697, "300–1000 Hz_40ms_log_rms_mae_db": 0.7662242413136761, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06763594516723091, "300–1000 Hz_40ms_envelope_correlation": 0.996141374802352, "300–1000 Hz_40ms_modulation_mae_db": 0.4472792592206267, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8187234455919135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19311633409319087, "1000–3000 Hz_10ms_envelope_correlation": 0.974507524281297, "1000–3000 Hz_10ms_modulation_mae_db": 0.7075063416596902, "1000–3000 Hz_40ms_log_rms_mae_db": 1.198178435867914, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13151406498390153, "1000–3000 Hz_40ms_envelope_correlation": 0.9899540166838839, "1000–3000 Hz_40ms_modulation_mae_db": 1.0123859964725133, "3000–6000 Hz_10ms_log_rms_mae_db": 2.017449599905955, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2562471265385037, "3000–6000 Hz_10ms_envelope_correlation": 0.9554638755390495, "3000–6000 Hz_10ms_modulation_mae_db": 1.8095130146327776, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8008748894940731, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19171004959846125, "3000–6000 Hz_40ms_envelope_correlation": 0.9744651947423509, "3000–6000 Hz_40ms_modulation_mae_db": 2.930547621276698, "mr_stft_db": 6.094050343736932, "band_mae_db": 0.28276212004578666, "rms_gain_error_db": -0.0666305150908281, "output_peak": 0.7046970767349703}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.008419578386354, "全帯域_10ms_rms_envelope_nrmse": 0.10059926665952633, "全帯域_10ms_envelope_correlation": 0.9904602210301023, "全帯域_10ms_modulation_mae_db": 0.16830839410621953, "全帯域_40ms_log_rms_mae_db": 0.5529686864428407, "全帯域_40ms_rms_envelope_nrmse": 0.05606902937853265, "全帯域_40ms_envelope_correlation": 0.997107450887745, "全帯域_40ms_modulation_mae_db": 0.15911448282630447, "300–1000 Hz_10ms_log_rms_mae_db": 1.3239428902199095, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13201978133737027, "300–1000 Hz_10ms_envelope_correlation": 0.9863874440519604, "300–1000 Hz_10ms_modulation_mae_db": 0.36138458485704267, "300–1000 Hz_40ms_log_rms_mae_db": 0.7668909661372066, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07651966832507359, "300–1000 Hz_40ms_envelope_correlation": 0.9959688943985611, "300–1000 Hz_40ms_modulation_mae_db": 0.45787249360325494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8774265865804003, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20393835382203568, "1000–3000 Hz_10ms_envelope_correlation": 0.9719645721160569, "1000–3000 Hz_10ms_modulation_mae_db": 0.7863903006173262, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1954580917285058, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1349013202416662, "1000–3000 Hz_40ms_envelope_correlation": 0.9905068925760868, "1000–3000 Hz_40ms_modulation_mae_db": 0.9579769649855453, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0637264208507737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.254829621191626, "3000–6000 Hz_10ms_envelope_correlation": 0.9560212136539065, "3000–6000 Hz_10ms_modulation_mae_db": 1.642783381550475, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7869291333228499, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1976226583756932, "3000–6000 Hz_40ms_envelope_correlation": 0.9725095236439514, "3000–6000 Hz_40ms_modulation_mae_db": 2.557577449593287, "mr_stft_db": 6.099873354657658, "band_mae_db": 0.3975850646580766, "rms_gain_error_db": -0.17606479421947205, "output_peak": 0.5969030943400021}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0187086882918857, "全帯域_10ms_rms_envelope_nrmse": 0.10141013968446688, "全帯域_10ms_envelope_correlation": 0.9900999791113397, "全帯域_10ms_modulation_mae_db": 0.39124010551416294, "全帯域_40ms_log_rms_mae_db": 0.5883976340639941, "全帯域_40ms_rms_envelope_nrmse": 0.053461260341698257, "全帯域_40ms_envelope_correlation": 0.9970696054244091, "全帯域_40ms_modulation_mae_db": 0.3100105434808481, "300–1000 Hz_10ms_log_rms_mae_db": 1.372121469407154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14236681509191806, "300–1000 Hz_10ms_envelope_correlation": 0.9833206242125684, "300–1000 Hz_10ms_modulation_mae_db": 0.4764561550375999, "300–1000 Hz_40ms_log_rms_mae_db": 0.8392089105054821, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07988053836042634, "300–1000 Hz_40ms_envelope_correlation": 0.9944747194048045, "300–1000 Hz_40ms_modulation_mae_db": 0.3534680804378309, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8876574582304055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19856661279627597, "1000–3000 Hz_10ms_envelope_correlation": 0.9713087909408618, "1000–3000 Hz_10ms_modulation_mae_db": 0.9634757813061534, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1990690270653022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1252275842382528, "1000–3000 Hz_40ms_envelope_correlation": 0.9894197259670442, "1000–3000 Hz_40ms_modulation_mae_db": 0.9036297580413258, "3000–6000 Hz_10ms_log_rms_mae_db": 2.059561891403961, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2716272634000156, "3000–6000 Hz_10ms_envelope_correlation": 0.9501293556704814, "3000–6000 Hz_10ms_modulation_mae_db": 1.9819374564651515, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7859872752682653, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20387076857125033, "3000–6000 Hz_40ms_envelope_correlation": 0.9720433967175657, "3000–6000 Hz_40ms_modulation_mae_db": 2.977290257988107, "mr_stft_db": 6.098629951659359, "band_mae_db": 0.2953693839072632, "rms_gain_error_db": -0.05178972818777809, "output_peak": 0.6697932705766729}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0580883592678856, "全帯域_10ms_rms_envelope_nrmse": 0.10432143160433927, "全帯域_10ms_envelope_correlation": 0.9895345103619104, "全帯域_10ms_modulation_mae_db": 0.635592212416314, "全帯域_40ms_log_rms_mae_db": 0.5651549044284009, "全帯域_40ms_rms_envelope_nrmse": 0.056142748263024636, "全帯域_40ms_envelope_correlation": 0.9967340444201305, "全帯域_40ms_modulation_mae_db": 0.4941401783120563, "300–1000 Hz_10ms_log_rms_mae_db": 1.3542440241957965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14471543664089567, "300–1000 Hz_10ms_envelope_correlation": 0.9828064194924541, "300–1000 Hz_10ms_modulation_mae_db": 0.5292655908246543, "300–1000 Hz_40ms_log_rms_mae_db": 0.8016964767407996, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08019947403933417, "300–1000 Hz_40ms_envelope_correlation": 0.9943357357759559, "300–1000 Hz_40ms_modulation_mae_db": 0.2203296098294176, "1000–3000 Hz_10ms_log_rms_mae_db": 1.833878258635497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19734458938353966, "1000–3000 Hz_10ms_envelope_correlation": 0.9729359757305903, "1000–3000 Hz_10ms_modulation_mae_db": 1.016673894120126, "1000–3000 Hz_40ms_log_rms_mae_db": 1.158609897050682, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12732888927652383, "1000–3000 Hz_40ms_envelope_correlation": 0.9908842172893758, "1000–3000 Hz_40ms_modulation_mae_db": 0.9570157808294341, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0525608888345768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23544408656033816, "3000–6000 Hz_10ms_envelope_correlation": 0.9626247786588528, "3000–6000 Hz_10ms_modulation_mae_db": 1.7184953592448462, "3000–6000 Hz_40ms_log_rms_mae_db": 1.747198300700023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1844829499908088, "3000–6000 Hz_40ms_envelope_correlation": 0.9765905363988514, "3000–6000 Hz_40ms_modulation_mae_db": 2.5501902531943417, "mr_stft_db": 6.085495600814473, "band_mae_db": 0.29201223231411205, "rms_gain_error_db": -0.010061168433065008, "output_peak": 0.6072852216062125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9515459536527917, "全帯域_10ms_rms_envelope_nrmse": 0.0945408952723439, "全帯域_10ms_envelope_correlation": 0.9916890576964024, "全帯域_10ms_modulation_mae_db": 0.3215978247798588, "全帯域_40ms_log_rms_mae_db": 0.5544155069263318, "全帯域_40ms_rms_envelope_nrmse": 0.05684542050717879, "全帯域_40ms_envelope_correlation": 0.9971079136644786, "全帯域_40ms_modulation_mae_db": 0.5470169527023035, "300–1000 Hz_10ms_log_rms_mae_db": 1.3412536371140487, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13399023178958214, "300–1000 Hz_10ms_envelope_correlation": 0.9860540579432244, "300–1000 Hz_10ms_modulation_mae_db": 0.569677656381064, "300–1000 Hz_40ms_log_rms_mae_db": 0.863206722686408, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07864637562366639, "300–1000 Hz_40ms_envelope_correlation": 0.9958170338378363, "300–1000 Hz_40ms_modulation_mae_db": 0.7788203161057075, "1000–3000 Hz_10ms_log_rms_mae_db": 1.83964379043937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1835324527128037, "1000–3000 Hz_10ms_envelope_correlation": 0.9767161995756132, "1000–3000 Hz_10ms_modulation_mae_db": 1.000113601914442, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1770960137754387, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11400002625560257, "1000–3000 Hz_40ms_envelope_correlation": 0.9928908112730503, "1000–3000 Hz_40ms_modulation_mae_db": 1.024356986228721, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0886358273114216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26109033140610566, "3000–6000 Hz_10ms_envelope_correlation": 0.9542612632946015, "3000–6000 Hz_10ms_modulation_mae_db": 1.8290064158868855, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8443107885027785, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2091080705173616, "3000–6000 Hz_40ms_envelope_correlation": 0.9704825314973162, "3000–6000 Hz_40ms_modulation_mae_db": 2.727798574735154, "mr_stft_db": 6.078706642778122, "band_mae_db": 0.32898880479362747, "rms_gain_error_db": -0.19567294702758736, "output_peak": 0.6730794559365978}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9989012099084926, "全帯域_10ms_rms_envelope_nrmse": 0.0966544252063788, "全帯域_10ms_envelope_correlation": 0.9909950019361963, "全帯域_10ms_modulation_mae_db": 0.39702607565019904, "全帯域_40ms_log_rms_mae_db": 0.5563348693578513, "全帯域_40ms_rms_envelope_nrmse": 0.0487740624916993, "全帯域_40ms_envelope_correlation": 0.9975747969367561, "全帯域_40ms_modulation_mae_db": 0.5263735744237783, "300–1000 Hz_10ms_log_rms_mae_db": 1.3177269262120006, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12987363523477172, "300–1000 Hz_10ms_envelope_correlation": 0.9861415765688006, "300–1000 Hz_10ms_modulation_mae_db": 0.5881192818540493, "300–1000 Hz_40ms_log_rms_mae_db": 0.7987737820105254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06921047995160907, "300–1000 Hz_40ms_envelope_correlation": 0.9958703914825366, "300–1000 Hz_40ms_modulation_mae_db": 0.7661631503568627, "1000–3000 Hz_10ms_log_rms_mae_db": 1.880219938043493, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19639401365461762, "1000–3000 Hz_10ms_envelope_correlation": 0.9717671948761911, "1000–3000 Hz_10ms_modulation_mae_db": 0.7772026872726804, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1916297247212533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1200507063918242, "1000–3000 Hz_40ms_envelope_correlation": 0.9901481737466653, "1000–3000 Hz_40ms_modulation_mae_db": 0.5403210193907991, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0333892331471564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23912505065329362, "3000–6000 Hz_10ms_envelope_correlation": 0.9613359449418918, "3000–6000 Hz_10ms_modulation_mae_db": 1.7316137336203263, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7496623840288095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18775315292331088, "3000–6000 Hz_40ms_envelope_correlation": 0.9752892865871245, "3000–6000 Hz_40ms_modulation_mae_db": 2.9428834629596516, "mr_stft_db": 6.097507238937603, "band_mae_db": 0.35089770230864475, "rms_gain_error_db": -0.06373764550162656, "output_peak": 0.7095896692062693}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9745552961787823, "全帯域_10ms_rms_envelope_nrmse": 0.0851413042696163, "全帯域_10ms_envelope_correlation": 0.9931409359207624, "全帯域_10ms_modulation_mae_db": 0.5672719334610292, "全帯域_40ms_log_rms_mae_db": 0.536758979938412, "全帯域_40ms_rms_envelope_nrmse": 0.04584874765014946, "全帯域_40ms_envelope_correlation": 0.9980274084857476, "全帯域_40ms_modulation_mae_db": 0.3770090740042078, "300–1000 Hz_10ms_log_rms_mae_db": 1.3207344901194815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12805770206105097, "300–1000 Hz_10ms_envelope_correlation": 0.987011617346097, "300–1000 Hz_10ms_modulation_mae_db": 0.46024509963932414, "300–1000 Hz_40ms_log_rms_mae_db": 0.7987511375747771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08143394731207562, "300–1000 Hz_40ms_envelope_correlation": 0.9949349354511549, "300–1000 Hz_40ms_modulation_mae_db": 0.689703776475314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.853792683420718, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1813611715831303, "1000–3000 Hz_10ms_envelope_correlation": 0.9779289382316673, "1000–3000 Hz_10ms_modulation_mae_db": 0.8160413395747333, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1845876829417863, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1240858909121921, "1000–3000 Hz_40ms_envelope_correlation": 0.9913856756516581, "1000–3000 Hz_40ms_modulation_mae_db": 0.9390374968713422, "3000–6000 Hz_10ms_log_rms_mae_db": 2.048292532927016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24509663746764923, "3000–6000 Hz_10ms_envelope_correlation": 0.9593314608101124, "3000–6000 Hz_10ms_modulation_mae_db": 1.7129101609800428, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7587821259124998, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1900337173636547, "3000–6000 Hz_40ms_envelope_correlation": 0.9747910004679075, "3000–6000 Hz_40ms_modulation_mae_db": 2.9884593824641024, "mr_stft_db": 6.084226309351901, "band_mae_db": 0.2664716144275642, "rms_gain_error_db": -0.13115000029616405, "output_peak": 0.6343123717709148}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.023824289564632, "spectral_convergence": 0.3143262034538573, "spectral_convergence_status": "ok", "amplitude_floor": 2.172633843844252e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.858583438354741, "spectral_convergence": 0.3056221602876089, "spectral_convergence_status": "ok", "amplitude_floor": 1.076658885105748e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.388956244152865, "spectral_convergence": 0.335129693904583, "spectral_convergence_status": "ok", "amplitude_floor": 9.276051433709775e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006687421878497317, "output_power": 0.0006650722739513689, "delta_db": -0.023898786866719795, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0031496712549154435, "output_power": 0.0030775762459102894, "delta_db": -0.10056405688453944, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007488394009410867, "output_power": 0.000720919601274573, "delta_db": -0.16501853341701536, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002473939509459554, "output_power": 0.0002160454930567205, "delta_db": -0.5884386555757248, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011342358064728732, "output_power": 9.412693487035565e-05, "delta_db": -0.8098943668126446, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.620279276522665e-05, "output_power": 5.01707404122101e-05, "delta_db": 0.3578228269619577, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3409395377531003, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1333031874294607, "pitch": null, "output_rms_dbfs": -23.166006873615853, "output_peak": 0.6270973554460322, "reference_rms_dbfs": -23.03270368618639, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.090454657357412}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023158079660483383, "log_rms_mae_db": 1.0256961121283632, "rms_envelope_nrmse": 0.10693707066581161, "envelope_correlation": 0.9890646226630039, "rms_gain_error_db": -0.132732914148008, "reference_mean_power": 0.004973550834679421, "output_mean_power": 0.004823844169440579}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9210188012408756e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006213961937331268, "output_power": 0.0005929382112620393, "reference_active": true, "delta_db": -0.2035914983778833}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026574930428793645, "output_power": 0.00029272758857531664, "reference_active": true, "delta_db": 0.4199151888822633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2739241814293584e-05, "output_power": 3.224075292400729e-05, "reference_active": true, "delta_db": -0.06663442256189524}], "active_band_count": 3, "active_band_mae_db": 0.23004703660734727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000211279420280134, "log_rms_mae_db": 0.5897507703058551, "rms_envelope_nrmse": 0.06549332415726364, "envelope_correlation": 0.9957333339066095, "rms_gain_error_db": -0.1325026250543013, "reference_mean_power": 0.0049805093532051875, "output_mean_power": 0.004830849386603315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7395423077013874e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005644300449164411, "output_power": 0.0005459965486676997, "reference_active": true, "delta_db": -0.14420225959778243}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013986675109572895, "output_power": 0.0001587633300685479, "reference_active": true, "delta_db": 0.5503571271934494}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.497453327177472e-06, "output_power": 2.2846852853603623e-06, "reference_active": true, "delta_db": -0.3867099620131758}], "active_band_count": 3, "active_band_mae_db": 0.3604231162681359}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020019856693346264, "log_rms_mae_db": 1.360204990121563, "rms_envelope_nrmse": 0.16746527693557953, "envelope_correlation": 0.9773189267447414, "rms_gain_error_db": -0.3256838782586239, "reference_mean_power": 0.0028106313477950454, "output_mean_power": 0.0026075671305247146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3651110338311905e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046335294017358356, "output_power": 0.0004186802099128686, "reference_active": true, "delta_db": -0.4402948991389438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024122957535154213, "output_power": 0.00021815921829123735, "reference_active": true, "delta_db": -0.4365698355509814}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.948878321904196e-05, "output_power": 3.400884411345503e-05, "reference_active": true, "delta_db": 0.6193501861510486}], "active_band_count": 3, "active_band_mae_db": 0.4987383069469913}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018155803316515472, "log_rms_mae_db": 0.8227525785822739, "rms_envelope_nrmse": 0.11226369171916073, "envelope_correlation": 0.9897374793075049, "rms_gain_error_db": -0.32678085481907226, "reference_mean_power": 0.002819303754025505, "output_mean_power": 0.002614952377924676}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2479023732657914e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004464610508434935, "output_power": 0.0004070885123436312, "reference_active": true, "delta_db": -0.40094729960242526}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001291081626943261, "output_power": 0.0001247717911765599, "reference_active": true, "delta_db": -0.1483729113732419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4462923050468237e-06, "output_power": 2.3581752418998933e-06, "reference_active": true, "delta_db": -0.15932273751361276}], "active_band_count": 3, "active_band_mae_db": 0.23621431616309332}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.07685175719321e-05, "log_rms_mae_db": 1.8980697989410735, "rms_envelope_nrmse": 0.19628606494031375, "envelope_correlation": 0.9729901232238469, "rms_gain_error_db": -0.6257593156325002, "reference_mean_power": 0.0002950078039052426, "output_mean_power": 0.00025542157176759717}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.385626430084388e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.894665895613538e-05, "output_power": 8.733561746061422e-05, "reference_active": true, "delta_db": -0.5420973930880204}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.27901607459427e-05, "output_power": 1.7527785703012353e-05, "reference_active": true, "delta_db": -1.140203335581019}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.304247232293777e-06, "output_power": 5.367968142897395e-06, "reference_active": true, "delta_db": 0.9591272089571412}], "active_band_count": 3, "active_band_mae_db": 0.8804759792087268}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.03385602019198e-05, "log_rms_mae_db": 1.2123655413890686, "rms_envelope_nrmse": 0.12267757692882461, "envelope_correlation": 0.9912969180527688, "rms_gain_error_db": -0.6238569051346942, "reference_mean_power": 0.0002923716436469955, "output_mean_power": 0.00025325006105864783}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2489747353423383e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.236418785147827e-05, "output_power": 8.205976849067382e-05, "reference_active": true, "delta_db": -0.5137332849291102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2735133445304299e-05, "output_power": 9.98426258613502e-06, "reference_active": true, "delta_db": -1.0568750565062963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6865666311394477e-07, "output_power": 3.2395665967301103e-07, "reference_active": true, "delta_db": 0.8128929605897117}], "active_band_count": 3, "active_band_mae_db": 0.7945004340083729}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.524569706134418e-05, "log_rms_mae_db": 1.9853289056199848, "rms_envelope_nrmse": 0.2477728711506788, "envelope_correlation": 0.9584317434694072, "rms_gain_error_db": -0.30978105033885217, "reference_mean_power": 6.736853190004857e-05, "output_mean_power": 6.27305330923648e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.630234817861519e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0381998821210514e-05, "output_power": 2.0431169431489266e-05, "reference_active": true, "delta_db": 0.010464531895952521}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.955680541944287e-06, "output_power": 4.346339029023274e-06, "reference_active": true, "delta_db": 0.4090238871624072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.594012590914479e-07, "output_power": 2.0157348376156414e-06, "reference_active": true, "delta_db": 4.852836296447675}], "active_band_count": 3, "active_band_mae_db": 1.757441571835345}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.8717196486707034e-05, "log_rms_mae_db": 1.6859864649191338, "rms_envelope_nrmse": 0.18597602863177629, "envelope_correlation": 0.976082288771298, "rms_gain_error_db": -0.3105220819213616, "reference_mean_power": 6.764325505404543e-05, "output_mean_power": 6.297559651196573e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.430902640057432e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9875215171126693e-05, "output_power": 2.138957162866828e-05, "reference_active": true, "delta_db": 0.3189024792393984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8203988319997446e-06, "output_power": 2.2289283566184093e-06, "reference_active": true, "delta_db": 0.879295612006983}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.685144573008789e-08, "output_power": 2.2721462208420393e-07, "reference_active": true, "delta_db": 7.899817458958275}], "active_band_count": 3, "active_band_mae_db": 3.0326718500682186}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7973453993682132, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2aa67d7f0c3f498097da7a47ba07424464dd674f4714b8f2d47e725643e7d2f0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9737649148720156, "standard_deviation": 0.0024246574680107597, "minimum": 0.9713087909408618, "maximum": 0.9779289382316673, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8611764949853589, "standard_deviation": 0.02867230779593067, "minimum": 1.8187234455919135, "maximum": 1.8980697989410735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8684849907092347, "standard_deviation": 0.11476033133788008, "minimum": 0.7075063416596902, "maximum": 1.016673894120126, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19381744912323845, "standard_deviation": 0.007668044064000426, "minimum": 0.1813611715831303, "maximum": 0.20393835382203568, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9908108039050667, "standard_deviation": 0.001075654462589939, "minimum": 0.9894197259670442, "maximum": 0.9928908112730503, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.189624301817494, "standard_deviation": 0.016308690414986594, "minimum": 1.158609897050682, "maximum": 1.2123655413890686, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8911530546035067, "standard_deviation": 0.15861790623051233, "minimum": 0.5403210193907991, "maximum": 1.024356986228721, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12497325740359848, "standard_deviation": 0.0065209898197166214, "minimum": 0.11400002625560257, "maximum": 0.1349013202416662, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9571999545047879, "standard_deviation": 0.004064276974087338, "minimum": 0.9501293556704814, "maximum": 0.9626247786588528, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0436181625001058, "standard_deviation": 0.03156403675770071, "minimum": 1.9853289056199848, "maximum": 2.0886358273114216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7729626367769813, "standard_deviation": 0.10255142854311666, "minimum": 1.642783381550475, "maximum": 1.9819374564651515, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25140412354602637, "standard_deviation": 0.011931084890468978, "minimum": 0.23544408656033816, "maximum": 0.2716272634000156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9740317198532957, "standard_deviation": 0.0021369372208349335, "minimum": 0.9704825314973162, "maximum": 0.9765905363988514, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.769966420268554, "standard_deviation": 0.04649912823167494, "minimum": 1.6859864649191338, "maximum": 1.8443107885027785, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.838427356534945, "standard_deviation": 0.19752637587049396, "minimum": 2.5501902531943417, "maximum": 3.0326718500682186, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19381967449653964, "standard_deviation": 0.008892260751768438, "minimum": 0.1844829499908088, "maximum": 0.2091080705173616, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9846655711252259, "standard_deviation": 0.003482734979649337, "minimum": 0.9773189267447414, "maximum": 0.9882839026419611, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.333458192470807, "standard_deviation": 0.030098238337630135, "minimum": 1.277437112376503, "maximum": 1.372121469407154, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4637330717882703, "standard_deviation": 0.11915946079108172, "minimum": 0.22597789876543697, "maximum": 0.5881192818540493, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13727472929572238, "standard_deviation": 0.014537887612147995, "minimum": 0.11970895527461073, "maximum": 0.16746527693557953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946600705575883, "standard_deviation": 0.0021097639106575587, "minimum": 0.9897374793075049, "maximum": 0.996141374802352, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8071881019438936, "standard_deviation": 0.03355104663236684, "minimum": 0.7662242413136761, "maximum": 0.863206722686408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.49373137527401345, "standard_deviation": 0.2261734384215884, "minimum": 0.2203296098294176, "maximum": 0.7788203161057075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08072376506232211, "standard_deviation": 0.01374301892701004, "minimum": 0.06763594516723091, "maximum": 0.11226369171916073, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3193783075260219, "standard_deviation": 0.04331316405180198, "minimum": 0.2664716144275642, "maximum": 0.3975850646580766, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.091118012411682, "standard_deviation": 0.007709875402430839, "minimum": 6.078706642778122, "maximum": 6.099873354657658, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6528446894522091, "standard_deviation": 0.04278995568420731, "minimum": 0.5969030943400021, "maximum": 0.7095896692062693, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10355124827324774, "standard_deviation": 0.06520232710387391, "minimum": -0.19567294702758736, "maximum": -0.010061168433065008, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9907897069974163, "standard_deviation": 0.0012989280330451486, "minimum": 0.9890646226630039, "maximum": 0.9931409359207624, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0091671732512375, "standard_deviation": 0.034184396031710126, "minimum": 0.9515459536527917, "maximum": 1.0580883592678856, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3574144492914004, "standard_deviation": 0.1777478021524937, "minimum": 0.1482320117960724, "maximum": 0.635592212416314, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09805539485122694, "standard_deviation": 0.006838010808535878, "minimum": 0.0851413042696163, "maximum": 0.10693707066581161, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970648087059704, "standard_deviation": 0.0006635874501048227, "minimum": 0.9957333339066095, "maximum": 0.9980274084857476, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5589529487186449, "standard_deviation": 0.02196641017114316, "minimum": 0.5278422382854729, "maximum": 0.5897507703058551, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36114686772444904, "standard_deviation": 0.16202485000671704, "minimum": 0.11508701977795771, "maximum": 0.5470169527023035, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05441444138961268, "standard_deviation": 0.005897633656675048, "minimum": 0.04584874765014946, "maximum": 0.06549332415726364, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0042", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.6133832931518555, "predicted_mos": 2.6133832931518555, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4582a9b5b8a4b25249fc0731b09d900272635ada56dc22249e106df2f37f17a4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7508739000186324, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0065421839817645, "全帯域_10ms_rms_envelope_nrmse": 0.13250661501084107, "全帯域_10ms_envelope_correlation": 0.9831068528296316, "全帯域_10ms_modulation_mae_db": 0.22140585531191062, "全帯域_40ms_log_rms_mae_db": 0.5636133815039958, "全帯域_40ms_rms_envelope_nrmse": 0.07919662427450964, "全帯域_40ms_envelope_correlation": 0.9936398552100812, "全帯域_40ms_modulation_mae_db": 0.3102331671205088, "300–1000 Hz_10ms_log_rms_mae_db": 1.4520950444131522, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19035547160575395, "300–1000 Hz_10ms_envelope_correlation": 0.9702631156744828, "300–1000 Hz_10ms_modulation_mae_db": 0.5693913060625184, "300–1000 Hz_40ms_log_rms_mae_db": 0.9487944867862632, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12427989155509958, "300–1000 Hz_40ms_envelope_correlation": 0.9868495771349072, "300–1000 Hz_40ms_modulation_mae_db": 0.7807085850009701, "1000–3000 Hz_10ms_log_rms_mae_db": 2.293892842993446, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24899088820772164, "1000–3000 Hz_10ms_envelope_correlation": 0.9530277979343982, "1000–3000 Hz_10ms_modulation_mae_db": 1.4377203999144499, "1000–3000 Hz_40ms_log_rms_mae_db": 1.515591806261901, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1550985252132413, "1000–3000 Hz_40ms_envelope_correlation": 0.9818756150967435, "1000–3000 Hz_40ms_modulation_mae_db": 1.322926209843851, "3000–6000 Hz_10ms_log_rms_mae_db": 2.643263776354078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2979587184596587, "3000–6000 Hz_10ms_envelope_correlation": 0.9399332344999994, "3000–6000 Hz_10ms_modulation_mae_db": 1.9002578590012897, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2091705332608758, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22286584795258502, "3000–6000 Hz_40ms_envelope_correlation": 0.9673816066970579, "3000–6000 Hz_40ms_modulation_mae_db": 2.8003934511789024, "mr_stft_db": 6.49042574267403, "band_mae_db": 0.344507867110496, "rms_gain_error_db": -0.1252930581321595, "output_peak": 0.7601350909989353}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0034153611184469, "全帯域_10ms_rms_envelope_nrmse": 0.09543018783136782, "全帯域_10ms_envelope_correlation": 0.9913631371645448, "全帯域_10ms_modulation_mae_db": 0.11354995610052238, "全帯域_40ms_log_rms_mae_db": 0.5608125599655239, "全帯域_40ms_rms_envelope_nrmse": 0.05622802750112866, "全帯域_40ms_envelope_correlation": 0.9969454612994743, "全帯域_40ms_modulation_mae_db": 0.17994343406966704, "300–1000 Hz_10ms_log_rms_mae_db": 1.4309925160119783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14334242623340324, "300–1000 Hz_10ms_envelope_correlation": 0.9833546323568483, "300–1000 Hz_10ms_modulation_mae_db": 0.29340592935651844, "300–1000 Hz_40ms_log_rms_mae_db": 0.8979515373050663, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09022300026427353, "300–1000 Hz_40ms_envelope_correlation": 0.9933175519190577, "300–1000 Hz_40ms_modulation_mae_db": 0.3044502840955787, "1000–3000 Hz_10ms_log_rms_mae_db": 2.15409283405091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21336888655898864, "1000–3000 Hz_10ms_envelope_correlation": 0.9665558846192954, "1000–3000 Hz_10ms_modulation_mae_db": 1.0356216535607414, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4641570110963846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13696546994313857, "1000–3000 Hz_40ms_envelope_correlation": 0.9871214075245452, "1000–3000 Hz_40ms_modulation_mae_db": 0.9718778861974521, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5944806089493286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29245809667563016, "3000–6000 Hz_10ms_envelope_correlation": 0.9423138060657861, "3000–6000 Hz_10ms_modulation_mae_db": 1.8425624001067762, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2108212795559528, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20946996152650055, "3000–6000 Hz_40ms_envelope_correlation": 0.9695405163376043, "3000–6000 Hz_40ms_modulation_mae_db": 2.5710596532645074, "mr_stft_db": 6.467596914005207, "band_mae_db": 0.2980710517847694, "rms_gain_error_db": -0.1382601566160838, "output_peak": 0.6844306286202212}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9635942450195776, "全帯域_10ms_rms_envelope_nrmse": 0.10076447561643453, "全帯域_10ms_envelope_correlation": 0.9902874951627264, "全帯域_10ms_modulation_mae_db": 0.16600096934617117, "全帯域_40ms_log_rms_mae_db": 0.5496339165273443, "全帯域_40ms_rms_envelope_nrmse": 0.052009476574412516, "全帯域_40ms_envelope_correlation": 0.9973888056119333, "全帯域_40ms_modulation_mae_db": 0.10568725182341994, "300–1000 Hz_10ms_log_rms_mae_db": 1.3991068298283653, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14550101455533274, "300–1000 Hz_10ms_envelope_correlation": 0.9829279413758613, "300–1000 Hz_10ms_modulation_mae_db": 0.3553425683207321, "300–1000 Hz_40ms_log_rms_mae_db": 0.8923060694958617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0814316923515075, "300–1000 Hz_40ms_envelope_correlation": 0.9949159055873666, "300–1000 Hz_40ms_modulation_mae_db": 0.5707682807033105, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2736100693876726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22785028028850854, "1000–3000 Hz_10ms_envelope_correlation": 0.9614095997011337, "1000–3000 Hz_10ms_modulation_mae_db": 1.1300169486031157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4674369487922403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14172773037650027, "1000–3000 Hz_40ms_envelope_correlation": 0.9859384739071311, "1000–3000 Hz_40ms_modulation_mae_db": 1.0983738512123804, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5968998923781483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28390100226465054, "3000–6000 Hz_10ms_envelope_correlation": 0.945341719623327, "3000–6000 Hz_10ms_modulation_mae_db": 2.0997048980158435, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2535534056210387, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20834442623340263, "3000–6000 Hz_40ms_envelope_correlation": 0.9703837031370078, "3000–6000 Hz_40ms_modulation_mae_db": 2.9623900470260147, "mr_stft_db": 6.475291802861503, "band_mae_db": 0.2960719385932951, "rms_gain_error_db": -0.1276843565305657, "output_peak": 0.6083074192274571}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0393951703033697, "全帯域_10ms_rms_envelope_nrmse": 0.09519337989636405, "全帯域_10ms_envelope_correlation": 0.9912744163874012, "全帯域_10ms_modulation_mae_db": 0.06984994378410789, "全帯域_40ms_log_rms_mae_db": 0.5978867721923565, "全帯域_40ms_rms_envelope_nrmse": 0.05724813336867804, "全帯域_40ms_envelope_correlation": 0.996615614273289, "全帯域_40ms_modulation_mae_db": 0.14793701879889762, "300–1000 Hz_10ms_log_rms_mae_db": 1.5053723654044688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13122555913261483, "300–1000 Hz_10ms_envelope_correlation": 0.9860060546907651, "300–1000 Hz_10ms_modulation_mae_db": 0.4318445839657304, "300–1000 Hz_40ms_log_rms_mae_db": 0.9654287432305798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07226478984247309, "300–1000 Hz_40ms_envelope_correlation": 0.9957790623353577, "300–1000 Hz_40ms_modulation_mae_db": 0.5431093742385893, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2651784595850666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21816259349886621, "1000–3000 Hz_10ms_envelope_correlation": 0.9641520295518197, "1000–3000 Hz_10ms_modulation_mae_db": 0.6310447174341006, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5189778265299143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1305390545876835, "1000–3000 Hz_40ms_envelope_correlation": 0.9870300226897576, "1000–3000 Hz_40ms_modulation_mae_db": 0.790190520845219, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5912675454108296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2818417049062349, "3000–6000 Hz_10ms_envelope_correlation": 0.9465197386484692, "3000–6000 Hz_10ms_modulation_mae_db": 1.6720702938527225, "3000–6000 Hz_40ms_log_rms_mae_db": 2.228455884264804, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20502292251507642, "3000–6000 Hz_40ms_envelope_correlation": 0.9710028923245713, "3000–6000 Hz_40ms_modulation_mae_db": 2.9512804755861057, "mr_stft_db": 6.4842655687374195, "band_mae_db": 0.18122270013703146, "rms_gain_error_db": -0.03538205888860538, "output_peak": 0.6321939512222063}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.978465867914496, "全帯域_10ms_rms_envelope_nrmse": 0.1017182401061465, "全帯域_10ms_envelope_correlation": 0.9900696875555074, "全帯域_10ms_modulation_mae_db": 0.16164397231375108, "全帯域_40ms_log_rms_mae_db": 0.5518772996779521, "全帯域_40ms_rms_envelope_nrmse": 0.05125319250191569, "全帯域_40ms_envelope_correlation": 0.9973881145340835, "全帯域_40ms_modulation_mae_db": 0.2336483471941356, "300–1000 Hz_10ms_log_rms_mae_db": 1.4929726714977771, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15106501799701597, "300–1000 Hz_10ms_envelope_correlation": 0.9812500477703034, "300–1000 Hz_10ms_modulation_mae_db": 0.4732663699288224, "300–1000 Hz_40ms_log_rms_mae_db": 0.928774338191624, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08251824926195511, "300–1000 Hz_40ms_envelope_correlation": 0.9942221565108346, "300–1000 Hz_40ms_modulation_mae_db": 0.7554046706307576, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2751510495484615, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21988242181540762, "1000–3000 Hz_10ms_envelope_correlation": 0.9636849346506273, "1000–3000 Hz_10ms_modulation_mae_db": 0.8921734893393009, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4583143059353028, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13507742296587447, "1000–3000 Hz_40ms_envelope_correlation": 0.9864655664310665, "1000–3000 Hz_40ms_modulation_mae_db": 0.8099065314595726, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6352532122197267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2939344595869388, "3000–6000 Hz_10ms_envelope_correlation": 0.9418675250711626, "3000–6000 Hz_10ms_modulation_mae_db": 1.963014711483616, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2920737437998584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21393128626555982, "3000–6000 Hz_40ms_envelope_correlation": 0.9685097857084748, "3000–6000 Hz_40ms_modulation_mae_db": 2.955912945742039, "mr_stft_db": 6.479885169172692, "band_mae_db": 0.29926506978849665, "rms_gain_error_db": -0.10040686217349543, "output_peak": 0.6191938269583478}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9563813712473627, "全帯域_10ms_rms_envelope_nrmse": 0.11545236381139844, "全帯域_10ms_envelope_correlation": 0.9871726045482176, "全帯域_10ms_modulation_mae_db": 0.4242753810898794, "全帯域_40ms_log_rms_mae_db": 0.5450910235379401, "全帯域_40ms_rms_envelope_nrmse": 0.059143467957856584, "全帯域_40ms_envelope_correlation": 0.996430877508466, "全帯域_40ms_modulation_mae_db": 0.16936830792339605, "300–1000 Hz_10ms_log_rms_mae_db": 1.4323266518395028, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15429484834700555, "300–1000 Hz_10ms_envelope_correlation": 0.9804672560333191, "300–1000 Hz_10ms_modulation_mae_db": 0.40917232016530997, "300–1000 Hz_40ms_log_rms_mae_db": 0.9468532306576389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08586333307620038, "300–1000 Hz_40ms_envelope_correlation": 0.9936387509389513, "300–1000 Hz_40ms_modulation_mae_db": 0.1727198210473411, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2749334674125934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22360706231116476, "1000–3000 Hz_10ms_envelope_correlation": 0.962242877398368, "1000–3000 Hz_10ms_modulation_mae_db": 0.833596718335934, "1000–3000 Hz_40ms_log_rms_mae_db": 1.509941575053824, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1442289530802225, "1000–3000 Hz_40ms_envelope_correlation": 0.9840174350802146, "1000–3000 Hz_40ms_modulation_mae_db": 0.9788365328558544, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5850667081481866, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.287980404027737, "3000–6000 Hz_10ms_envelope_correlation": 0.9453777273968451, "3000–6000 Hz_10ms_modulation_mae_db": 2.0254929134008033, "3000–6000 Hz_40ms_log_rms_mae_db": 2.225822259895052, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2031178140701165, "3000–6000 Hz_40ms_envelope_correlation": 0.9720419753220431, "3000–6000 Hz_40ms_modulation_mae_db": 3.2309777845777625, "mr_stft_db": 6.488922281008816, "band_mae_db": 0.2547836411698042, "rms_gain_error_db": -0.07785807343827547, "output_peak": 0.6704169425611071}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9638639170475994, "全帯域_10ms_rms_envelope_nrmse": 0.11011988255046655, "全帯域_10ms_envelope_correlation": 0.9885459165028452, "全帯域_10ms_modulation_mae_db": 0.3663421245809217, "全帯域_40ms_log_rms_mae_db": 0.5263302011588159, "全帯域_40ms_rms_envelope_nrmse": 0.06527737668269919, "全帯域_40ms_envelope_correlation": 0.9957165675238352, "全帯域_40ms_modulation_mae_db": 0.31736785637819925, "300–1000 Hz_10ms_log_rms_mae_db": 1.4241575872679562, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.159201382170419, "300–1000 Hz_10ms_envelope_correlation": 0.979332186653253, "300–1000 Hz_10ms_modulation_mae_db": 0.2409716754659971, "300–1000 Hz_40ms_log_rms_mae_db": 0.8728818197497638, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09027707968241518, "300–1000 Hz_40ms_envelope_correlation": 0.9928772194721, "300–1000 Hz_40ms_modulation_mae_db": 0.27815108089118756, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2065391128503373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2191297293221618, "1000–3000 Hz_10ms_envelope_correlation": 0.9642136807692071, "1000–3000 Hz_10ms_modulation_mae_db": 0.7791458913194713, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4588457923448643, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13987362644685356, "1000–3000 Hz_40ms_envelope_correlation": 0.9857434946267687, "1000–3000 Hz_40ms_modulation_mae_db": 0.8955173123400031, "3000–6000 Hz_10ms_log_rms_mae_db": 2.577943996459666, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29277740318302914, "3000–6000 Hz_10ms_envelope_correlation": 0.9426595541398065, "3000–6000 Hz_10ms_modulation_mae_db": 1.8418095267927332, "3000–6000 Hz_40ms_log_rms_mae_db": 2.20220742651842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20645674649788603, "3000–6000 Hz_40ms_envelope_correlation": 0.9706427771398468, "3000–6000 Hz_40ms_modulation_mae_db": 3.0531139353531564, "mr_stft_db": 6.4724754610996, "band_mae_db": 0.2626802130896044, "rms_gain_error_db": 0.06147568284344564, "output_peak": 0.6928453284179655}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0073449193318005, "全帯域_10ms_rms_envelope_nrmse": 0.09486134906811387, "全帯域_10ms_envelope_correlation": 0.9913651893953918, "全帯域_10ms_modulation_mae_db": 0.12376308064292489, "全帯域_40ms_log_rms_mae_db": 0.5440535147145463, "全帯域_40ms_rms_envelope_nrmse": 0.04817955384314348, "全帯域_40ms_envelope_correlation": 0.9976978008514101, "全帯域_40ms_modulation_mae_db": 0.09354614266581933, "300–1000 Hz_10ms_log_rms_mae_db": 1.47594238976163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13899253203102957, "300–1000 Hz_10ms_envelope_correlation": 0.9844954041172486, "300–1000 Hz_10ms_modulation_mae_db": 0.3062256523035008, "300–1000 Hz_40ms_log_rms_mae_db": 0.8904839283054999, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07529770526407219, "300–1000 Hz_40ms_envelope_correlation": 0.9957490922800118, "300–1000 Hz_40ms_modulation_mae_db": 0.2508292820906092, "1000–3000 Hz_10ms_log_rms_mae_db": 2.202253411826104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21547232950021783, "1000–3000 Hz_10ms_envelope_correlation": 0.9656639689748031, "1000–3000 Hz_10ms_modulation_mae_db": 0.6388150211047999, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4113744388855969, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1413629640773087, "1000–3000 Hz_40ms_envelope_correlation": 0.9857882842596501, "1000–3000 Hz_40ms_modulation_mae_db": 0.8733328758249423, "3000–6000 Hz_10ms_log_rms_mae_db": 2.579773391772042, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2747664579055827, "3000–6000 Hz_10ms_envelope_correlation": 0.9491044685459251, "3000–6000 Hz_10ms_modulation_mae_db": 1.7653635939559116, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2357606732118174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20201457809397777, "3000–6000 Hz_40ms_envelope_correlation": 0.9720374010216043, "3000–6000 Hz_40ms_modulation_mae_db": 2.6234697890371605, "mr_stft_db": 6.475302154739038, "band_mae_db": 0.2802850053213927, "rms_gain_error_db": -0.09645145544124932, "output_peak": 0.7077655151868443}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.457705275728085, "spectral_convergence": 0.3307522555762788, "spectral_convergence_status": "ok", "amplitude_floor": 2.172633843844252e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.301553952395869, "spectral_convergence": 0.31301464504323556, "spectral_convergence_status": "ok", "amplitude_floor": 1.076658885105748e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.712017999898135, "spectral_convergence": 0.3370915688520199, "spectral_convergence_status": "ok", "amplitude_floor": 9.276051433709775e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006687421878497317, "output_power": 0.0006709876073524112, "delta_db": 0.014557774503056946, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0031496712549154435, "output_power": 0.0029563466870560356, "delta_db": -0.2750986504730689, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007488394009410867, "output_power": 0.00083100672227255, "delta_db": 0.4521584976798567, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002473939509459554, "output_power": 0.00022997452837961456, "delta_db": -0.31709339544251536, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011342358064728732, "output_power": 9.545706868569815e-05, "delta_db": -0.7489525949043729, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.620279276522665e-05, "output_power": 4.904410986075845e-05, "delta_db": 0.25918628966010543, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.344507867110496, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1252930581321595, "pitch": null, "output_rms_dbfs": -23.15799674431855, "output_peak": 0.7601350909989353, "reference_rms_dbfs": -23.03270368618639, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.49042574267403}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023158079660483383, "log_rms_mae_db": 1.0065421839817645, "rms_envelope_nrmse": 0.13250661501084107, "envelope_correlation": 0.9831068528296316, "rms_gain_error_db": -0.12570364719467175, "reference_mean_power": 0.004973550834679421, "output_mean_power": 0.0048316581172516875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9210188012408756e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006213961937331268, "output_power": 0.000572715714634431, "reference_active": true, "delta_db": -0.35429488924876584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026574930428793645, "output_power": 0.0002501936872324078, "reference_active": true, "delta_db": -0.2619578866117679}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2739241814293584e-05, "output_power": 3.237964915113427e-05, "reference_active": true, "delta_db": -0.047964790075198216}], "active_band_count": 3, "active_band_mae_db": 0.22140585531191062}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000211279420280134, "log_rms_mae_db": 0.5636133815039958, "rms_envelope_nrmse": 0.07919662427450964, "envelope_correlation": 0.9936398552100812, "rms_gain_error_db": -0.12578517462192063, "reference_mean_power": 0.0049805093532051875, "output_mean_power": 0.004838327285214958}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7395423077013874e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005644300449164411, "output_power": 0.0005201525986332159, "reference_active": true, "delta_db": -0.35479350904249096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013986675109572895, "output_power": 0.0001331973015409705, "reference_active": true, "delta_db": -0.2121906045621495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.497453327177472e-06, "output_power": 2.296814081104405e-06, "reference_active": true, "delta_db": -0.363715387756886}], "active_band_count": 3, "active_band_mae_db": 0.3102331671205088}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020019856693346264, "log_rms_mae_db": 1.4520950444131522, "rms_envelope_nrmse": 0.19035547160575395, "envelope_correlation": 0.9702631156744828, "rms_gain_error_db": -0.2605368491263786, "reference_mean_power": 0.0028106313477950454, "output_mean_power": 0.002646977202274331}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3651110338311905e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046335294017358356, "output_power": 0.00040172714334525255, "reference_active": true, "delta_db": -0.6198074661776034}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024122957535154213, "output_power": 0.000198991979546512, "reference_active": true, "delta_db": -0.8359497998123169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.948878321904196e-05, "output_power": 2.782371957744574e-05, "reference_active": true, "delta_db": -0.25241665219763504}], "active_band_count": 3, "active_band_mae_db": 0.5693913060625184}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018155803316515472, "log_rms_mae_db": 0.9487944867862632, "rms_envelope_nrmse": 0.12427989155509958, "envelope_correlation": 0.9868495771349072, "rms_gain_error_db": -0.26111502108431517, "reference_mean_power": 0.002819303754025505, "output_mean_power": 0.002654791188548718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2479023732657914e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004464610508434935, "output_power": 0.00038427174912026186, "reference_active": true, "delta_db": -0.6514511994088275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001291081626943261, "output_power": 0.00010982647398103749, "reference_active": true, "delta_db": -0.7024666019126835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4462923050468237e-06, "output_power": 1.9484423200428697e-06, "reference_active": true, "delta_db": -0.9882079536813994}], "active_band_count": 3, "active_band_mae_db": 0.7807085850009701}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.07685175719321e-05, "log_rms_mae_db": 2.293892842993446, "rms_envelope_nrmse": 0.24899088820772164, "envelope_correlation": 0.9530277979343982, "rms_gain_error_db": -0.41163109148823396, "reference_mean_power": 0.0002950078039052426, "output_mean_power": 0.0002683307196849896}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.385626430084388e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.894665895613538e-05, "output_power": 9.186507602852404e-05, "reference_active": true, "delta_db": -0.32250695555550246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.27901607459427e-05, "output_power": 2.1068014868439503e-05, "reference_active": true, "delta_db": -0.34123772219939075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.304247232293777e-06, "output_power": 9.97329949770728e-06, "reference_active": true, "delta_db": 3.6494165219884565}], "active_band_count": 3, "active_band_mae_db": 1.4377203999144499}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.03385602019198e-05, "log_rms_mae_db": 1.515591806261901, "rms_envelope_nrmse": 0.1550985252132413, "envelope_correlation": 0.9818756150967435, "rms_gain_error_db": -0.4058587341148716, "reference_mean_power": 0.0002923716436469955, "output_mean_power": 0.0002662866388121065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2489747353423383e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.236418785147827e-05, "output_power": 8.987919053070686e-05, "reference_active": true, "delta_db": -0.11844463456738065}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2735133445304299e-05, "output_power": 1.172365205445486e-05, "reference_active": true, "delta_db": -0.35940579526750893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6865666311394477e-07, "output_power": 6.00192323134128e-07, "reference_active": true, "delta_db": 3.4909281996966635}], "active_band_count": 3, "active_band_mae_db": 1.322926209843851}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.524569706134418e-05, "log_rms_mae_db": 2.643263776354078, "rms_envelope_nrmse": 0.2979587184596587, "envelope_correlation": 0.9399332344999994, "rms_gain_error_db": -0.5490993283520601, "reference_mean_power": 6.736853190004857e-05, "output_mean_power": 5.936727985068305e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.630234817861519e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0381998821210514e-05, "output_power": 1.742212453867262e-05, "reference_active": true, "delta_db": -0.6814565825216174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.955680541944287e-06, "output_power": 3.5994704152041976e-06, "reference_active": true, "delta_db": -0.40982602657844874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.594012590914479e-07, "output_power": 1.9058945530294213e-06, "reference_active": true, "delta_db": 4.609490967903803}], "active_band_count": 3, "active_band_mae_db": 1.9002578590012897}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.8717196486707034e-05, "log_rms_mae_db": 2.2091705332608758, "rms_envelope_nrmse": 0.22286584795258502, "envelope_correlation": 0.9673816066970579, "rms_gain_error_db": -0.5488301776986615, "reference_mean_power": 6.764325505404543e-05, "output_mean_power": 5.961306893753091e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.430902640057432e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9875215171126693e-05, "output_power": 1.799950894649646e-05, "reference_active": true, "delta_db": -0.43051182008333383}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8203988319997446e-06, "output_power": 1.959453020421276e-06, "reference_active": true, "delta_db": 0.31968307170217286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.685144573008789e-08, "output_power": 2.145621323647008e-07, "reference_active": true, "delta_db": 7.650985461751201}], "active_band_count": 3, "active_band_mae_db": 2.8003934511789024}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.782151013496921, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f2c3a43b26632eb385cbe007cecc90b1fef2986bda1ad66ebd7feee866223618", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9626188466999566, "standard_deviation": 0.0042155422253261025, "minimum": 0.9530277979343982, "maximum": 0.9665558846192954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.243206405956824, "standard_deviation": 0.04923152631451772, "minimum": 2.15409283405091, "maximum": 2.293892842993446, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9222668549514892, "standard_deviation": 0.2715323338542581, "minimum": 0.6310447174341006, "maximum": 1.4377203999144499, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22330802393787963, "standard_deviation": 0.011314869755979595, "minimum": 0.21336888655898864, "maximum": 0.24899088820772164, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9854975374519847, "standard_deviation": 0.0017554414705090152, "minimum": 0.9818756150967435, "maximum": 0.9871214075245452, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4755799631125037, "standard_deviation": 0.03696005519841604, "minimum": 1.4113744388855969, "maximum": 1.5189778265299143, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9676202150724094, "standard_deviation": 0.17481689610620998, "minimum": 0.790190520845219, "maximum": 1.322926209843851, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14060921833635287, "standard_deviation": 0.0072770544565130685, "minimum": 0.1305390545876835, "maximum": 0.1550985252132413, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9441397217489151, "standard_deviation": 0.002968234805537077, "minimum": 0.9399332344999994, "maximum": 0.9491044685459251, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.600493641461501, "standard_deviation": 0.024921333795949234, "minimum": 2.577943996459666, "maximum": 2.643263776354078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.888784524576212, "standard_deviation": 0.13905767800168786, "minimum": 1.6720702938527225, "maximum": 2.0997048980158435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2882022808761827, "standard_deviation": 0.007621568566812966, "minimum": 0.2747664579055827, "maximum": 0.2979587184596587, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9701925822110262, "standard_deviation": 0.0016422107833104127, "minimum": 0.9673816066970579, "maximum": 0.9720419753220431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.2322331507659774, "standard_deviation": 0.029265184561202428, "minimum": 2.20220742651842, "maximum": 2.2920737437998584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.893574760220706, "standard_deviation": 0.2193824247202747, "minimum": 2.5710596532645074, "maximum": 3.2309777845777625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20890294789438807, "standard_deviation": 0.006791129990165453, "minimum": 0.20201457809397777, "maximum": 0.22286584795258502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9810120798340102, "standard_deviation": 0.0048542638696887665, "minimum": 0.9702631156744828, "maximum": 0.9860060546907651, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.451620757003104, "standard_deviation": 0.03684079273686346, "minimum": 1.3991068298283653, "maximum": 1.5053723654044688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3849525506961412, "standard_deviation": 0.10742997524194388, "minimum": 0.2409716754659971, "maximum": 0.5693913060625184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15174728150907185, "standard_deviation": 0.01791485300011842, "minimum": 0.13122555913261483, "maximum": 0.19035547160575395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9934186645223233, "standard_deviation": 0.0028634254009410583, "minimum": 0.9868495771349072, "maximum": 0.9957790623353577, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9179342692152872, "standard_deviation": 0.033808811470204585, "minimum": 0.8728818197497638, "maximum": 0.9654287432305798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.457017672337293, "standard_deviation": 0.23687566833454976, "minimum": 0.1727198210473411, "maximum": 0.7807085850009701, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08776946766224956, "standard_deviation": 0.0160918565223767, "minimum": 0.07226478984247309, "maximum": 0.12427989155509958, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27711093587436125, "standard_deviation": 0.04742419574015815, "minimum": 0.18122270013703146, "maximum": 0.344507867110496, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.479270636787288, "standard_deviation": 0.008079496131374108, "minimum": 6.467596914005207, "maximum": 6.49042574267403, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6719110878991357, "standard_deviation": 0.05080525238700507, "minimum": 0.6083074192274571, "maximum": 0.7601350909989353, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07998254229712362, "standard_deviation": 0.06591911726588573, "minimum": -0.1382601566160838, "maximum": 0.06147568284344564, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9891481624432832, "standard_deviation": 0.002859883013408161, "minimum": 0.9831068528296316, "maximum": 0.9913651893953918, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9898753794955522, "standard_deviation": 0.028878632063328632, "minimum": 0.9563813712473627, "maximum": 1.0393951703033697, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.20585391039627365, "standard_deviation": 0.12592801090921868, "minimum": 0.06984994378410789, "maximum": 0.4242753810898794, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1057558117363916, "standard_deviation": 0.013129910919115443, "minimum": 0.09486134906811387, "maximum": 0.13250661501084107, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9964778871015716, "standard_deviation": 0.0013104150319236667, "minimum": 0.9936398552100812, "maximum": 0.9976978008514101, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5549123336598094, "standard_deviation": 0.020790474362858555, "minimum": 0.5263302011588159, "maximum": 0.5978867721923565, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.19471644074675545, "standard_deviation": 0.08545164609628524, "minimum": 0.09354614266581933, "maximum": 0.31736785637819925, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05856698158804298, "standard_deviation": 0.009874534354550027, "minimum": 0.04817955384314348, "maximum": 0.07919662427450964, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0042", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8291780948638916, "predicted_mos": 2.8291780948638916, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "92e52792a07f0c00e0f46e112ebc0a39f57a84f8556ee9cc61ecbf44ec782ae1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.149081599898636, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0508719311761217, "全帯域_10ms_rms_envelope_nrmse": 0.09973914727090898, "全帯域_10ms_envelope_correlation": 0.9905010561464481, "全帯域_10ms_modulation_mae_db": 0.8688126291025021, "全帯域_40ms_log_rms_mae_db": 0.5847705168650821, "全帯域_40ms_rms_envelope_nrmse": 0.05842875103808736, "全帯域_40ms_envelope_correlation": 0.9966269194836912, "全帯域_40ms_modulation_mae_db": 0.9476111698289605, "300–1000 Hz_10ms_log_rms_mae_db": 1.2424280151540024, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13072501001608286, "300–1000 Hz_10ms_envelope_correlation": 0.9862305124350965, "300–1000 Hz_10ms_modulation_mae_db": 1.0594267338475318, "300–1000 Hz_40ms_log_rms_mae_db": 0.7939137853285635, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07974545566245522, "300–1000 Hz_40ms_envelope_correlation": 0.9949170669364052, "300–1000 Hz_40ms_modulation_mae_db": 1.441131799676213, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7870093745944802, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19694205570474577, "1000–3000 Hz_10ms_envelope_correlation": 0.9725973768720287, "1000–3000 Hz_10ms_modulation_mae_db": 1.1176469092332464, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1851957259212007, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12291079777641656, "1000–3000 Hz_40ms_envelope_correlation": 0.9911006482624046, "1000–3000 Hz_40ms_modulation_mae_db": 0.9167681141270632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9569211346443884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24951304137306285, "3000–6000 Hz_10ms_envelope_correlation": 0.9578256971918127, "3000–6000 Hz_10ms_modulation_mae_db": 1.4454448694203539, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7132808863433686, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18905523420738465, "3000–6000 Hz_40ms_envelope_correlation": 0.9751867099481355, "3000–6000 Hz_40ms_modulation_mae_db": 2.7882653700922133, "mr_stft_db": 6.069685399380693, "band_mae_db": 0.3403523006884741, "rms_gain_error_db": -0.12029663154227892, "output_peak": 0.5765836453726296}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0417087094410524, "全帯域_10ms_rms_envelope_nrmse": 0.0948538838055876, "全帯域_10ms_envelope_correlation": 0.991402429264411, "全帯域_10ms_modulation_mae_db": 0.454205461341281, "全帯域_40ms_log_rms_mae_db": 0.5840268166948677, "全帯域_40ms_rms_envelope_nrmse": 0.04983638301613642, "全帯域_40ms_envelope_correlation": 0.9974927266164789, "全帯域_40ms_modulation_mae_db": 0.31588344898227827, "300–1000 Hz_10ms_log_rms_mae_db": 1.3211988422477172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12931183515350736, "300–1000 Hz_10ms_envelope_correlation": 0.9863762168228711, "300–1000 Hz_10ms_modulation_mae_db": 0.531956681950347, "300–1000 Hz_40ms_log_rms_mae_db": 0.757584660907916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07244510983842813, "300–1000 Hz_40ms_envelope_correlation": 0.9955567865298819, "300–1000 Hz_40ms_modulation_mae_db": 0.23964960672217125, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7959778054022444, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1941584824434347, "1000–3000 Hz_10ms_envelope_correlation": 0.9726240862553427, "1000–3000 Hz_10ms_modulation_mae_db": 0.6970236487583409, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1670174631489945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12006073796857908, "1000–3000 Hz_40ms_envelope_correlation": 0.9904283305245166, "1000–3000 Hz_40ms_modulation_mae_db": 0.9993862772665109, "3000–6000 Hz_10ms_log_rms_mae_db": 1.940543865268535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2453985506616536, "3000–6000 Hz_10ms_envelope_correlation": 0.9592336707512958, "3000–6000 Hz_10ms_modulation_mae_db": 1.72888690803813, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7504199936444294, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18871186775989338, "3000–6000 Hz_40ms_envelope_correlation": 0.9751594549676202, "3000–6000 Hz_40ms_modulation_mae_db": 2.6456134864736125, "mr_stft_db": 6.0849591855788185, "band_mae_db": 0.19903393096849706, "rms_gain_error_db": -0.04708190800849222, "output_peak": 0.6834409867519687}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9969625463353906, "全帯域_10ms_rms_envelope_nrmse": 0.09885427430286721, "全帯域_10ms_envelope_correlation": 0.9907137871214622, "全帯域_10ms_modulation_mae_db": 0.3479532379327068, "全帯域_40ms_log_rms_mae_db": 0.5631475430498324, "全帯域_40ms_rms_envelope_nrmse": 0.05278682936638244, "全帯域_40ms_envelope_correlation": 0.9973410876115143, "全帯域_40ms_modulation_mae_db": 0.3499552028950596, "300–1000 Hz_10ms_log_rms_mae_db": 1.2439208844269276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14798105848289006, "300–1000 Hz_10ms_envelope_correlation": 0.9822780128424438, "300–1000 Hz_10ms_modulation_mae_db": 0.37751549738444456, "300–1000 Hz_40ms_log_rms_mae_db": 0.7464837216010584, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08601676420097286, "300–1000 Hz_40ms_envelope_correlation": 0.9940489799785475, "300–1000 Hz_40ms_modulation_mae_db": 0.361222896184987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7185944713744483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17733246949697193, "1000–3000 Hz_10ms_envelope_correlation": 0.978861531896646, "1000–3000 Hz_10ms_modulation_mae_db": 0.9908159020891144, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1522032614203543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11718390209337387, "1000–3000 Hz_40ms_envelope_correlation": 0.9926357714966949, "1000–3000 Hz_40ms_modulation_mae_db": 1.2295942240409357, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9554609061370267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24358174849608522, "3000–6000 Hz_10ms_envelope_correlation": 0.9602714109817497, "3000–6000 Hz_10ms_modulation_mae_db": 1.6685480436421463, "3000–6000 Hz_40ms_log_rms_mae_db": 1.739078328307379, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19409907205414356, "3000–6000 Hz_40ms_envelope_correlation": 0.9747516946180645, "3000–6000 Hz_40ms_modulation_mae_db": 2.837252290057559, "mr_stft_db": 6.0539898378545125, "band_mae_db": 0.28767058978197085, "rms_gain_error_db": -0.14067555600983067, "output_peak": 0.7385470205704271}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.006641362704958, "全帯域_10ms_rms_envelope_nrmse": 0.09169821953932705, "全帯域_10ms_envelope_correlation": 0.9919592558682212, "全帯域_10ms_modulation_mae_db": 0.0971646816695514, "全帯域_40ms_log_rms_mae_db": 0.5383921315323542, "全帯域_40ms_rms_envelope_nrmse": 0.04522246869393475, "全帯域_40ms_envelope_correlation": 0.9980064249470759, "全帯域_40ms_modulation_mae_db": 0.08633438729783893, "300–1000 Hz_10ms_log_rms_mae_db": 1.246551956713359, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13067581597048974, "300–1000 Hz_10ms_envelope_correlation": 0.9861718590834773, "300–1000 Hz_10ms_modulation_mae_db": 0.35412714140648555, "300–1000 Hz_40ms_log_rms_mae_db": 0.7291280490315943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06992680634821552, "300–1000 Hz_40ms_envelope_correlation": 0.9961186322762738, "300–1000 Hz_40ms_modulation_mae_db": 0.4592491112923051, "1000–3000 Hz_10ms_log_rms_mae_db": 1.805923318892035, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1794944792961077, "1000–3000 Hz_10ms_envelope_correlation": 0.9779003745266707, "1000–3000 Hz_10ms_modulation_mae_db": 1.0118852396879898, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2182112679281807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11642394714375152, "1000–3000 Hz_40ms_envelope_correlation": 0.9924085185154375, "1000–3000 Hz_40ms_modulation_mae_db": 0.7543558965888253, "3000–6000 Hz_10ms_log_rms_mae_db": 1.952810462641708, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2521333364333477, "3000–6000 Hz_10ms_envelope_correlation": 0.9575042261513739, "3000–6000 Hz_10ms_modulation_mae_db": 2.324604176997338, "3000–6000 Hz_40ms_log_rms_mae_db": 1.755592241440196, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1939627174510262, "3000–6000 Hz_40ms_envelope_correlation": 0.973650161784026, "3000–6000 Hz_40ms_modulation_mae_db": 3.49615233772233, "mr_stft_db": 6.07535385989224, "band_mae_db": 0.35748806054070154, "rms_gain_error_db": -0.10501000398589665, "output_peak": 0.6093267922303374}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.020638149241593, "全帯域_10ms_rms_envelope_nrmse": 0.10127857527068469, "全帯域_10ms_envelope_correlation": 0.990143783607856, "全帯域_10ms_modulation_mae_db": 0.3491622176417024, "全帯域_40ms_log_rms_mae_db": 0.5821143059417403, "全帯域_40ms_rms_envelope_nrmse": 0.052418279833927356, "全帯域_40ms_envelope_correlation": 0.9971850640310902, "全帯域_40ms_modulation_mae_db": 0.21661231722039886, "300–1000 Hz_10ms_log_rms_mae_db": 1.236628338730083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13701348205995528, "300–1000 Hz_10ms_envelope_correlation": 0.9846723648816151, "300–1000 Hz_10ms_modulation_mae_db": 0.2970286589357564, "300–1000 Hz_40ms_log_rms_mae_db": 0.75074116128798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07772964752885503, "300–1000 Hz_40ms_envelope_correlation": 0.9949742066790511, "300–1000 Hz_40ms_modulation_mae_db": 0.42969170336635937, "1000–3000 Hz_10ms_log_rms_mae_db": 1.820646258220194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18796272217925603, "1000–3000 Hz_10ms_envelope_correlation": 0.9735422456172548, "1000–3000 Hz_10ms_modulation_mae_db": 1.3055524906521414, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1551691767569594, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11052562180974321, "1000–3000 Hz_40ms_envelope_correlation": 0.9907948685966795, "1000–3000 Hz_40ms_modulation_mae_db": 1.2941667243796386, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9565463325639239, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2587966652099625, "3000–6000 Hz_10ms_envelope_correlation": 0.9545475726278302, "3000–6000 Hz_10ms_modulation_mae_db": 1.6659449689664951, "3000–6000 Hz_40ms_log_rms_mae_db": 1.716363647663884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19272920105305447, "3000–6000 Hz_40ms_envelope_correlation": 0.9742199635720551, "3000–6000 Hz_40ms_modulation_mae_db": 2.656316693827694, "mr_stft_db": 6.065554244004564, "band_mae_db": 0.19013800108346265, "rms_gain_error_db": -0.05736320891069895, "output_peak": 0.7838732116752469}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0333567867597313, "全帯域_10ms_rms_envelope_nrmse": 0.10088326974204663, "全帯域_10ms_envelope_correlation": 0.9903486170788672, "全帯域_10ms_modulation_mae_db": 0.21063155484216034, "全帯域_40ms_log_rms_mae_db": 0.5760860264639789, "全帯域_40ms_rms_envelope_nrmse": 0.06467655745449943, "全帯域_40ms_envelope_correlation": 0.9957645998656585, "全帯域_40ms_modulation_mae_db": 0.23893079606547318, "300–1000 Hz_10ms_log_rms_mae_db": 1.2956337583789561, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14016870966453665, "300–1000 Hz_10ms_envelope_correlation": 0.9841455439964589, "300–1000 Hz_10ms_modulation_mae_db": 0.32907159655655543, "300–1000 Hz_40ms_log_rms_mae_db": 0.8046889873804609, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09194103235722985, "300–1000 Hz_40ms_envelope_correlation": 0.992738293686714, "300–1000 Hz_40ms_modulation_mae_db": 0.3604995516047096, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7884929656196145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19615439168595922, "1000–3000 Hz_10ms_envelope_correlation": 0.9733486587185898, "1000–3000 Hz_10ms_modulation_mae_db": 0.8232823245486566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2055787036995609, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1238179868403995, "1000–3000 Hz_40ms_envelope_correlation": 0.9914472580012055, "1000–3000 Hz_40ms_modulation_mae_db": 0.7260395667597089, "3000–6000 Hz_10ms_log_rms_mae_db": 1.974192036068934, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23941405025211882, "3000–6000 Hz_10ms_envelope_correlation": 0.961564547697968, "3000–6000 Hz_10ms_modulation_mae_db": 1.7899027470618074, "3000–6000 Hz_40ms_log_rms_mae_db": 1.710256922103697, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19178601815129573, "3000–6000 Hz_40ms_envelope_correlation": 0.9742277447857726, "3000–6000 Hz_40ms_modulation_mae_db": 2.743124375517541, "mr_stft_db": 6.06279305260523, "band_mae_db": 0.39216419844943445, "rms_gain_error_db": 0.037254381758983275, "output_peak": 0.7005758078733568}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.002833174248306, "全帯域_10ms_rms_envelope_nrmse": 0.09780850691041558, "全帯域_10ms_envelope_correlation": 0.9910231850110262, "全帯域_10ms_modulation_mae_db": 0.4234749566579496, "全帯域_40ms_log_rms_mae_db": 0.5557064511366403, "全帯域_40ms_rms_envelope_nrmse": 0.05717751388712238, "全帯域_40ms_envelope_correlation": 0.9969858413258849, "全帯域_40ms_modulation_mae_db": 0.5308647918558899, "300–1000 Hz_10ms_log_rms_mae_db": 1.2471505715440132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.140545480305893, "300–1000 Hz_10ms_envelope_correlation": 0.9845310083223312, "300–1000 Hz_10ms_modulation_mae_db": 0.570781938026089, "300–1000 Hz_40ms_log_rms_mae_db": 0.8257352266134306, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08854348363342614, "300–1000 Hz_40ms_envelope_correlation": 0.9942904051498969, "300–1000 Hz_40ms_modulation_mae_db": 0.550489119282945, "1000–3000 Hz_10ms_log_rms_mae_db": 1.771319609007774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18167367781446359, "1000–3000 Hz_10ms_envelope_correlation": 0.97614078519874, "1000–3000 Hz_10ms_modulation_mae_db": 0.583576113658618, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2076903040654945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11717257379707385, "1000–3000 Hz_40ms_envelope_correlation": 0.990656944782573, "1000–3000 Hz_40ms_modulation_mae_db": 0.5955420958510848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9621427609952198, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23937371544811833, "3000–6000 Hz_10ms_envelope_correlation": 0.9613484067853467, "3000–6000 Hz_10ms_modulation_mae_db": 1.7643432592107056, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7451776873680012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18595015027702033, "3000–6000 Hz_40ms_envelope_correlation": 0.9757131127771136, "3000–6000 Hz_40ms_modulation_mae_db": 2.5281451561068664, "mr_stft_db": 6.062319356893549, "band_mae_db": 0.2543677125096568, "rms_gain_error_db": -0.17931870707896064, "output_peak": 0.6657824536772299}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0235794881966267, "全帯域_10ms_rms_envelope_nrmse": 0.11111138057824843, "全帯域_10ms_envelope_correlation": 0.9881896512047574, "全帯域_10ms_modulation_mae_db": 0.200435515627395, "全帯域_40ms_log_rms_mae_db": 0.6075750612687911, "全帯域_40ms_rms_envelope_nrmse": 0.06630650964295988, "全帯域_40ms_envelope_correlation": 0.9954773392076448, "全帯域_40ms_modulation_mae_db": 0.28328442057614517, "300–1000 Hz_10ms_log_rms_mae_db": 1.3497289820956329, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16046266965783948, "300–1000 Hz_10ms_envelope_correlation": 0.978956856618797, "300–1000 Hz_10ms_modulation_mae_db": 0.4983258057494389, "300–1000 Hz_40ms_log_rms_mae_db": 0.8184817984938674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09588437918718992, "300–1000 Hz_40ms_envelope_correlation": 0.991997185194993, "300–1000 Hz_40ms_modulation_mae_db": 0.6103139745965609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7846445452058628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18971602283442998, "1000–3000 Hz_10ms_envelope_correlation": 0.9740316557565166, "1000–3000 Hz_10ms_modulation_mae_db": 0.46580146259152094, "1000–3000 Hz_40ms_log_rms_mae_db": 1.199404322014121, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12214533048572405, "1000–3000 Hz_40ms_envelope_correlation": 0.9900983396397679, "1000–3000 Hz_40ms_modulation_mae_db": 0.5667792073124321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9868329794867703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24844481523755474, "3000–6000 Hz_10ms_envelope_correlation": 0.9583515752130841, "3000–6000 Hz_10ms_modulation_mae_db": 1.9131506531005826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7888997003078475, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19161029901122312, "3000–6000 Hz_40ms_envelope_correlation": 0.9741685197090216, "3000–6000 Hz_40ms_modulation_mae_db": 2.8849479831961755, "mr_stft_db": 6.086794995279971, "band_mae_db": 0.27947045475972593, "rms_gain_error_db": 0.012389531507874852, "output_peak": 0.723765890198643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.933018602386531, "spectral_convergence": 0.2999097513137156, "spectral_convergence_status": "ok", "amplitude_floor": 2.172633843844252e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.832299472325052, "spectral_convergence": 0.29136405625982637, "spectral_convergence_status": "ok", "amplitude_floor": 1.076658885105748e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.443738123430496, "spectral_convergence": 0.3261172722627203, "spectral_convergence_status": "ok", "amplitude_floor": 9.276051433709775e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006687421878497317, "output_power": 0.0006769027111387536, "delta_db": 0.05267531849038004, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0031496712549154435, "output_power": 0.0030568213117209324, "delta_db": -0.12995174394037623, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007488394009410867, "output_power": 0.0007389909789706872, "delta_db": -0.05749550277212897, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002473939509459554, "output_power": 0.00021393671539442877, "delta_db": -0.6310375272070289, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011342358064728732, "output_power": 9.894830843538779e-05, "delta_db": -0.5929497917393646, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.620279276522665e-05, "output_power": 5.2779906876816586e-05, "delta_db": 0.5780039199815659, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3403523006884741, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12029663154227892, "pitch": null, "output_rms_dbfs": -23.153000317728672, "output_peak": 0.5765836453726296, "reference_rms_dbfs": -23.03270368618639, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.069685399380693}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023158079660483383, "log_rms_mae_db": 1.0508719311761217, "rms_envelope_nrmse": 0.09973914727090898, "envelope_correlation": 0.9905010561464481, "rms_gain_error_db": -0.12012444016981841, "reference_mean_power": 0.004973550834679421, "output_mean_power": 0.004837869143323596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9210188012408756e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006213961937331268, "output_power": 0.000583306088994602, "reference_active": true, "delta_db": -0.27472078771727027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026574930428793645, "output_power": 0.00024218823855304988, "reference_active": true, "delta_db": -0.4031908767644876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2739241814293584e-05, "output_power": 2.0999840462746718e-05, "reference_active": true, "delta_db": -1.9285262228257485}], "active_band_count": 3, "active_band_mae_db": 0.8688126291025021}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000211279420280134, "log_rms_mae_db": 0.5847705168650821, "rms_envelope_nrmse": 0.05842875103808736, "envelope_correlation": 0.9966269194836912, "rms_gain_error_db": -0.1200262972277541, "reference_mean_power": 0.0049805093532051875, "output_mean_power": 0.004844747310514378}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7395423077013874e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005644300449164411, "output_power": 0.0005302924109470906, "reference_active": true, "delta_db": -0.2709471148427598}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013986675109572895, "output_power": 0.00012759196392947458, "reference_active": true, "delta_db": -0.3989116469772657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.497453327177472e-06, "output_power": 1.514258102057464e-06, "reference_active": true, "delta_db": -2.1729747476668564}], "active_band_count": 3, "active_band_mae_db": 0.9476111698289605}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020019856693346264, "log_rms_mae_db": 1.2424280151540024, "rms_envelope_nrmse": 0.13072501001608286, "envelope_correlation": 0.9862305124350965, "rms_gain_error_db": -0.23912524763329857, "reference_mean_power": 0.0028106313477950454, "output_mean_power": 0.0026600595612328664}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3651110338311905e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046335294017358356, "output_power": 0.00042230701721968186, "reference_active": true, "delta_db": -0.4028362525807108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024122957535154213, "output_power": 0.00019182164673150746, "reference_active": true, "delta_db": -0.9953293736651043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.948878321904196e-05, "output_power": 1.9572459150026856e-05, "reference_active": true, "delta_db": -1.7801145752967804}], "active_band_count": 3, "active_band_mae_db": 1.0594267338475318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018155803316515472, "log_rms_mae_db": 0.7939137853285635, "rms_envelope_nrmse": 0.07974545566245522, "envelope_correlation": 0.9949170669364052, "rms_gain_error_db": -0.2396947146058071, "reference_mean_power": 0.002819303754025505, "output_mean_power": 0.0026679175147328697}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2479023732657914e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004464610508434935, "output_power": 0.00040171833692033204, "reference_active": true, "delta_db": -0.4586192104714223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001291081626943261, "output_power": 0.00010347261063082249, "reference_active": true, "delta_db": -0.9612829428789511}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4462923050468237e-06, "output_power": 1.2536001188969193e-06, "reference_active": true, "delta_db": -2.9034932456782654}], "active_band_count": 3, "active_band_mae_db": 1.441131799676213}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.07685175719321e-05, "log_rms_mae_db": 1.7870093745944802, "rms_envelope_nrmse": 0.19694205570474577, "envelope_correlation": 0.9725973768720287, "rms_gain_error_db": -0.6022372236968112, "reference_mean_power": 0.0002950078039052426, "output_mean_power": 0.0002568087294781331}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.385626430084388e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.894665895613538e-05, "output_power": 7.593813095303907e-05, "reference_active": true, "delta_db": -1.1494123060468164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.27901607459427e-05, "output_power": 1.6707979949476597e-05, "reference_active": true, "delta_db": -1.3482344298699944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.304247232293777e-06, "output_power": 5.241150227020936e-06, "reference_active": true, "delta_db": 0.8552939917829281}], "active_band_count": 3, "active_band_mae_db": 1.1176469092332464}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.03385602019198e-05, "log_rms_mae_db": 1.1851957259212007, "rms_envelope_nrmse": 0.12291079777641656, "envelope_correlation": 0.9911006482624046, "rms_gain_error_db": -0.6006129115299589, "reference_mean_power": 0.0002923716436469955, "output_mean_power": 0.000254609121316168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2489747353423383e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.236418785147827e-05, "output_power": 7.200212842325577e-05, "reference_active": true, "delta_db": -1.081582812725074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2735133445304299e-05, "output_power": 9.407386480708402e-06, "reference_active": true, "delta_db": -1.3153451370987757}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6865666311394477e-07, "output_power": 2.914306704990898e-07, "reference_active": true, "delta_db": 0.35337639255733994}], "active_band_count": 3, "active_band_mae_db": 0.9167681141270632}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.524569706134418e-05, "log_rms_mae_db": 1.9569211346443884, "rms_envelope_nrmse": 0.24951304137306285, "envelope_correlation": 0.9578256971918127, "rms_gain_error_db": -0.2798120913881705, "reference_mean_power": 6.736853190004857e-05, "output_mean_power": 6.316490890402222e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.630234817861519e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0381998821210514e-05, "output_power": 2.0046621128434298e-05, "reference_active": true, "delta_db": -0.07205589640813871}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.955680541944287e-06, "output_power": 4.120890870496212e-06, "reference_active": true, "delta_db": 0.17769902666916204}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.594012590914479e-07, "output_power": 1.6896927497883187e-06, "reference_active": true, "delta_db": 4.086579685183761}], "active_band_count": 3, "active_band_mae_db": 1.4454448694203539}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.8717196486707034e-05, "log_rms_mae_db": 1.7132808863433686, "rms_envelope_nrmse": 0.18905523420738465, "envelope_correlation": 0.9751867099481355, "rms_gain_error_db": -0.27942227392512337, "reference_mean_power": 6.764325505404543e-05, "output_mean_power": 6.342818302162577e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.430902640057432e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9875215171126693e-05, "output_power": 2.0775969487171685e-05, "reference_active": true, "delta_db": 0.19249459720173434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8203988319997446e-06, "output_power": 2.152882400573171e-06, "reference_active": true, "delta_db": 0.7285375930985456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.685144573008789e-08, "output_power": 2.0456480793481012e-07, "reference_active": true, "delta_db": 7.44376391997636}], "active_band_count": 3, "active_band_mae_db": 2.7882653700922133}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7999532048121203, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1cc17cdf0f135630eefe92c4277b6d943afb13bcd06400bdc6d33444d722a470", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9748808393552237, "standard_deviation": 0.0024412709874289114, "minimum": 0.9725973768720287, "maximum": 0.978861531896646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7840760435395815, "standard_deviation": 0.030320515729247145, "minimum": 1.7185944713744483, "maximum": 1.820646258220194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8744480114024535, "standard_deviation": 0.28363030473790224, "minimum": 0.46580146259152094, "maximum": 1.3055524906521414, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18792928768192113, "standard_deviation": 0.007685328638965056, "minimum": 0.17733246949697193, "maximum": 0.19694205570474577, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99119633497741, "standard_deviation": 0.0009150352352705274, "minimum": 0.9900983396397679, "maximum": 0.9926357714966949, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1863087781193582, "standard_deviation": 0.02542219662968979, "minimum": 1.1522032614203543, "maximum": 1.2182112679281807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.885329013290775, "standard_deviation": 0.2745494510104801, "minimum": 0.5667792073124321, "maximum": 1.2941667243796386, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1187801122393827, "standard_deviation": 0.004378969749835529, "minimum": 0.11052562180974321, "maximum": 0.1238179868403995, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9588308884250576, "standard_deviation": 0.002311310949321949, "minimum": 0.9545475726278302, "maximum": 0.961564547697968, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9606813097258133, "standard_deviation": 0.014114540966733405, "minimum": 1.940543865268535, "maximum": 1.9868329794867703, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7876032033046947, "standard_deviation": 0.25471925907296056, "minimum": 1.4454448694203539, "maximum": 2.324604176997338, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24708199038898798, "standard_deviation": 0.006588057101486551, "minimum": 0.23937371544811833, "maximum": 0.2587966652099625, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9746346702702262, "standard_deviation": 0.0006846274298343507, "minimum": 0.973650161784026, "maximum": 0.9757131127771136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7398836758973504, "standard_deviation": 0.026532092413286038, "minimum": 1.710256922103697, "maximum": 1.7888997003078475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.822477211624249, "standard_deviation": 0.2954289502013799, "minimum": 2.5281451561068664, "maximum": 3.49615233772233, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19098806999563017, "standard_deviation": 0.0028505607155566923, "minimum": 0.18595015027702033, "maximum": 0.19409907205414356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9841702968753864, "standard_deviation": 0.002514538246342889, "minimum": 0.978956856618797, "maximum": 0.9863762168228711, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2729051686613366, "standard_deviation": 0.04341413705005995, "minimum": 1.236628338730083, "maximum": 1.3497289820956329, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5022792567320811, "standard_deviation": 0.24658990081975335, "minimum": 0.2970286589357564, "maximum": 1.0594267338475318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13961050766389932, "standard_deviation": 0.010540076753035118, "minimum": 0.12931183515350736, "maximum": 0.16046266965783948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9943301945539704, "standard_deviation": 0.0013897560961138248, "minimum": 0.991997185194993, "maximum": 0.9961186322762738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7783446738306088, "standard_deviation": 0.03669099117410186, "minimum": 0.7291280490315943, "maximum": 0.8257352266134306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5565309703407814, "standard_deviation": 0.37568738183482975, "minimum": 0.23964960672217125, "maximum": 1.441131799676213, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08277908484459659, "standard_deviation": 0.009304935005826723, "minimum": 0.06992680634821552, "maximum": 0.09588437918718992, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.28758565609774045, "standard_deviation": 0.07281491744978426, "minimum": 0.19013800108346265, "maximum": 0.39216419844943445, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.070181241436197, "standard_deviation": 0.011472772169924976, "minimum": 6.0539898378545125, "maximum": 6.086794995279971, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6852369760437299, "standard_deviation": 0.06786586676820693, "minimum": 0.5765836453726296, "maximum": 0.7838732116752469, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07501276278366249, "standard_deviation": 0.07509661193675696, "minimum": -0.17931870707896064, "maximum": 0.037254381758983275, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9905352206628812, "standard_deviation": 0.0011174479452551998, "minimum": 0.9881896512047574, "maximum": 0.9919592558682212, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0220740185129724, "standard_deviation": 0.019218417017723293, "minimum": 0.9969625463353906, "maximum": 1.0508719311761217, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3689800318519061, "standard_deviation": 0.23551331368904102, "minimum": 0.0971646816695514, "maximum": 0.8688126291025021, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09952840717751077, "standard_deviation": 0.005678077445976597, "minimum": 0.09169821953932705, "maximum": 0.11111138057824843, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968600003861299, "standard_deviation": 0.0008646683866541657, "minimum": 0.9954773392076448, "maximum": 0.9980064249470759, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5739773566191608, "standard_deviation": 0.021160492972923862, "minimum": 0.5383921315323542, "maximum": 0.6075750612687911, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3711845668402556, "standard_deviation": 0.26498625386431224, "minimum": 0.08633438729783893, "maximum": 0.9476111698289605, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05585666161663125, "standard_deviation": 0.0072301584051965365, "minimum": 0.04522246869393475, "maximum": 0.06630650964295988, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0042", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.7242965698242188, "predicted_mos": 2.7242965698242188, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1a12274df4bcdae7e3acb4c2cc171e7437d40bcdffd2dd83cc63e4b1930dcb8e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8371562999673188, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.028160484431816, "全帯域_10ms_rms_envelope_nrmse": 0.10787437544210587, "全帯域_10ms_envelope_correlation": 0.9888755151064336, "全帯域_10ms_modulation_mae_db": 0.31800331086120964, "全帯域_40ms_log_rms_mae_db": 0.5790922814181473, "全帯域_40ms_rms_envelope_nrmse": 0.06663441227019418, "全帯域_40ms_envelope_correlation": 0.9955754418356879, "全帯域_40ms_modulation_mae_db": 0.2910224535391433, "300–1000 Hz_10ms_log_rms_mae_db": 1.378843211884009, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16631461039324305, "300–1000 Hz_10ms_envelope_correlation": 0.9776435361315802, "300–1000 Hz_10ms_modulation_mae_db": 0.2963690478526876, "300–1000 Hz_40ms_log_rms_mae_db": 0.8234993136713396, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11443366390797359, "300–1000 Hz_40ms_envelope_correlation": 0.9892754766163724, "300–1000 Hz_40ms_modulation_mae_db": 0.1808753525471054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9635798220596181, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1973748714921216, "1000–3000 Hz_10ms_envelope_correlation": 0.9718093654093161, "1000–3000 Hz_10ms_modulation_mae_db": 0.7793607502157828, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3065677253561858, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12044309131511152, "1000–3000 Hz_40ms_envelope_correlation": 0.9906765422883783, "1000–3000 Hz_40ms_modulation_mae_db": 0.6760385735982443, "3000–6000 Hz_10ms_log_rms_mae_db": 2.063474770073016, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2504184556664843, "3000–6000 Hz_10ms_envelope_correlation": 0.9575756626115172, "3000–6000 Hz_10ms_modulation_mae_db": 1.7441932125929238, "3000–6000 Hz_40ms_log_rms_mae_db": 1.778834503808462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18764000050611496, "3000–6000 Hz_40ms_envelope_correlation": 0.9757785599934886, "3000–6000 Hz_40ms_modulation_mae_db": 3.0019170200880017, "mr_stft_db": 6.178003523353609, "band_mae_db": 0.31514230933363035, "rms_gain_error_db": -0.13209457235786862, "output_peak": 0.626623729675691}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.028009609242622, "全帯域_10ms_rms_envelope_nrmse": 0.09507568232522927, "全帯域_10ms_envelope_correlation": 0.9912814125989289, "全帯域_10ms_modulation_mae_db": 0.13918206406004116, "全帯域_40ms_log_rms_mae_db": 0.5218776319565911, "全帯域_40ms_rms_envelope_nrmse": 0.05277953513824997, "全帯域_40ms_envelope_correlation": 0.9971279145496935, "全帯域_40ms_modulation_mae_db": 0.1738134406642657, "300–1000 Hz_10ms_log_rms_mae_db": 1.2696665099161026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1210310104179908, "300–1000 Hz_10ms_envelope_correlation": 0.9879862319166415, "300–1000 Hz_10ms_modulation_mae_db": 0.20748060981219388, "300–1000 Hz_40ms_log_rms_mae_db": 0.751196071409582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06507790572332188, "300–1000 Hz_40ms_envelope_correlation": 0.9963882907709559, "300–1000 Hz_40ms_modulation_mae_db": 0.49289534739534835, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9044050494686569, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20095670812575195, "1000–3000 Hz_10ms_envelope_correlation": 0.9709767317072058, "1000–3000 Hz_10ms_modulation_mae_db": 0.9053722338519425, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2668211833890919, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13454526202113457, "1000–3000 Hz_40ms_envelope_correlation": 0.9879869591578501, "1000–3000 Hz_40ms_modulation_mae_db": 0.6912131723483766, "3000–6000 Hz_10ms_log_rms_mae_db": 2.136388904556287, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2668192584916012, "3000–6000 Hz_10ms_envelope_correlation": 0.95169468524464, "3000–6000 Hz_10ms_modulation_mae_db": 1.8839648802361477, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8772611981668053, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19662933294245635, "3000–6000 Hz_40ms_envelope_correlation": 0.9731768363871346, "3000–6000 Hz_40ms_modulation_mae_db": 2.942119750992136, "mr_stft_db": 6.184909047957138, "band_mae_db": 0.24036969545679499, "rms_gain_error_db": -0.04573989003692373, "output_peak": 0.7110246458697678}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9833710711773347, "全帯域_10ms_rms_envelope_nrmse": 0.09916379991699321, "全帯域_10ms_envelope_correlation": 0.990743608517384, "全帯域_10ms_modulation_mae_db": 0.25060689672790387, "全帯域_40ms_log_rms_mae_db": 0.5473965847284784, "全帯域_40ms_rms_envelope_nrmse": 0.05545261558712449, "全帯域_40ms_envelope_correlation": 0.9971829802760941, "全帯域_40ms_modulation_mae_db": 0.11082571222835862, "300–1000 Hz_10ms_log_rms_mae_db": 1.2889962157207726, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1344387014832718, "300–1000 Hz_10ms_envelope_correlation": 0.9858331395910545, "300–1000 Hz_10ms_modulation_mae_db": 0.36297872682436155, "300–1000 Hz_40ms_log_rms_mae_db": 0.7708708597245572, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07729736363306126, "300–1000 Hz_40ms_envelope_correlation": 0.9958534671667537, "300–1000 Hz_40ms_modulation_mae_db": 0.5313981936438673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9580455111715187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20932179215246943, "1000–3000 Hz_10ms_envelope_correlation": 0.9686710108701242, "1000–3000 Hz_10ms_modulation_mae_db": 0.6727134020421115, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2471805627719752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13305978605259655, "1000–3000 Hz_40ms_envelope_correlation": 0.9890187779202027, "1000–3000 Hz_40ms_modulation_mae_db": 0.6014036201541427, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1658216923300144, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26081318291178424, "3000–6000 Hz_10ms_envelope_correlation": 0.9541431337723256, "3000–6000 Hz_10ms_modulation_mae_db": 1.7440544686615385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8689431743649219, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19846200305115794, "3000–6000 Hz_40ms_envelope_correlation": 0.9725171521705087, "3000–6000 Hz_40ms_modulation_mae_db": 2.580952221226788, "mr_stft_db": 6.189809004947485, "band_mae_db": 0.3475855643983954, "rms_gain_error_db": -0.1769226184338277, "output_peak": 0.6022457201713901}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9925454578749519, "全帯域_10ms_rms_envelope_nrmse": 0.10334679220785488, "全帯域_10ms_envelope_correlation": 0.9897193610960332, "全帯域_10ms_modulation_mae_db": 0.6877229165804208, "全帯域_40ms_log_rms_mae_db": 0.5614146624932664, "全帯域_40ms_rms_envelope_nrmse": 0.0547630322406473, "全帯域_40ms_envelope_correlation": 0.9969031844993841, "全帯域_40ms_modulation_mae_db": 0.5896811268057345, "300–1000 Hz_10ms_log_rms_mae_db": 1.3408442880476907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1405304897426381, "300–1000 Hz_10ms_envelope_correlation": 0.9837360232415604, "300–1000 Hz_10ms_modulation_mae_db": 0.6119856164060734, "300–1000 Hz_40ms_log_rms_mae_db": 0.806406715381832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0785288800863426, "300–1000 Hz_40ms_envelope_correlation": 0.9946338062778751, "300–1000 Hz_40ms_modulation_mae_db": 0.5335777598537126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9193285710630248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1998629132978326, "1000–3000 Hz_10ms_envelope_correlation": 0.9701669089661387, "1000–3000 Hz_10ms_modulation_mae_db": 0.8283122780824025, "1000–3000 Hz_40ms_log_rms_mae_db": 1.221795612782341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1231580357845926, "1000–3000 Hz_40ms_envelope_correlation": 0.9887447251323837, "1000–3000 Hz_40ms_modulation_mae_db": 0.7022623669650702, "3000–6000 Hz_10ms_log_rms_mae_db": 2.16265577770991, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27657207278268975, "3000–6000 Hz_10ms_envelope_correlation": 0.9480437539596684, "3000–6000 Hz_10ms_modulation_mae_db": 1.987359757814537, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8742785796690893, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20327009292472234, "3000–6000 Hz_40ms_envelope_correlation": 0.971999959748562, "3000–6000 Hz_40ms_modulation_mae_db": 2.9754939025981613, "mr_stft_db": 6.182340761976328, "band_mae_db": 0.25358624727985674, "rms_gain_error_db": -0.030907930719974355, "output_peak": 0.6691667483521928}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0175640755145965, "全帯域_10ms_rms_envelope_nrmse": 0.10266627418483669, "全帯域_10ms_envelope_correlation": 0.9898566660649173, "全帯域_10ms_modulation_mae_db": 0.6545575957642907, "全帯域_40ms_log_rms_mae_db": 0.5600973598217797, "全帯域_40ms_rms_envelope_nrmse": 0.0555670947168586, "全帯域_40ms_envelope_correlation": 0.9968010365003077, "全帯域_40ms_modulation_mae_db": 0.49284601800006184, "300–1000 Hz_10ms_log_rms_mae_db": 1.3466405061882143, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1424265244840746, "300–1000 Hz_10ms_envelope_correlation": 0.9833243904991443, "300–1000 Hz_10ms_modulation_mae_db": 0.5594322273618186, "300–1000 Hz_40ms_log_rms_mae_db": 0.8443825169364423, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07800555858480358, "300–1000 Hz_40ms_envelope_correlation": 0.9946691161287483, "300–1000 Hz_40ms_modulation_mae_db": 0.28590743588472006, "1000–3000 Hz_10ms_log_rms_mae_db": 1.897151315405612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2012979396536147, "1000–3000 Hz_10ms_envelope_correlation": 0.9702909913962864, "1000–3000 Hz_10ms_modulation_mae_db": 1.1043221919316897, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1949831901111436, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11947052736651712, "1000–3000 Hz_40ms_envelope_correlation": 0.9905962577309756, "1000–3000 Hz_40ms_modulation_mae_db": 1.0974556637307575, "3000–6000 Hz_10ms_log_rms_mae_db": 2.118589138027281, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2412029834201904, "3000–6000 Hz_10ms_envelope_correlation": 0.9607389138135712, "3000–6000 Hz_10ms_modulation_mae_db": 1.7916263987773846, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8141655516081385, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18538616388320164, "3000–6000 Hz_40ms_envelope_correlation": 0.9763938148324475, "3000–6000 Hz_40ms_modulation_mae_db": 2.607164788999143, "mr_stft_db": 6.171221952862585, "band_mae_db": 0.22906421628791543, "rms_gain_error_db": -0.015974282145014913, "output_peak": 0.6046121111390758}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9442267210463259, "全帯域_10ms_rms_envelope_nrmse": 0.09706311623439287, "全帯域_10ms_envelope_correlation": 0.9912009663259578, "全帯域_10ms_modulation_mae_db": 0.36026003365516096, "全帯域_40ms_log_rms_mae_db": 0.5514091095828569, "全帯域_40ms_rms_envelope_nrmse": 0.05785513974025451, "全帯域_40ms_envelope_correlation": 0.9969688678384465, "全帯域_40ms_modulation_mae_db": 0.5485033975562837, "300–1000 Hz_10ms_log_rms_mae_db": 1.3403755719916672, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13666477048801157, "300–1000 Hz_10ms_envelope_correlation": 0.9854653229387368, "300–1000 Hz_10ms_modulation_mae_db": 0.6504000177148784, "300–1000 Hz_40ms_log_rms_mae_db": 0.8765324613190123, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08125523282322536, "300–1000 Hz_40ms_envelope_correlation": 0.9954711304228818, "300–1000 Hz_40ms_modulation_mae_db": 0.8559852803554585, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9329128313526782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19134561742183798, "1000–3000 Hz_10ms_envelope_correlation": 0.9735164184411962, "1000–3000 Hz_10ms_modulation_mae_db": 0.9141744289393742, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2368466400734115, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11846114378131367, "1000–3000 Hz_40ms_envelope_correlation": 0.9908874889918476, "1000–3000 Hz_40ms_modulation_mae_db": 0.867870576307054, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1946428818788406, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2672681961746732, "3000–6000 Hz_10ms_envelope_correlation": 0.9516395324453079, "3000–6000 Hz_10ms_modulation_mae_db": 1.9063177770333033, "3000–6000 Hz_40ms_log_rms_mae_db": 1.915489696240276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20728791166224872, "3000–6000 Hz_40ms_envelope_correlation": 0.9706842646016514, "3000–6000 Hz_40ms_modulation_mae_db": 2.712265046507678, "mr_stft_db": 6.17100122645871, "band_mae_db": 0.2975295265280918, "rms_gain_error_db": -0.1927802631820161, "output_peak": 0.6681909882533831}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.979694136112105, "全帯域_10ms_rms_envelope_nrmse": 0.0936103427108844, "全帯域_10ms_envelope_correlation": 0.9915867244921241, "全帯域_10ms_modulation_mae_db": 0.5551381112022195, "全帯域_40ms_log_rms_mae_db": 0.5317603030229229, "全帯域_40ms_rms_envelope_nrmse": 0.04796685950557548, "全帯域_40ms_envelope_correlation": 0.9977149597567516, "全帯域_40ms_modulation_mae_db": 0.7459410155523734, "300–1000 Hz_10ms_log_rms_mae_db": 1.3051691699400447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12594114821304334, "300–1000 Hz_10ms_envelope_correlation": 0.9870547506930554, "300–1000 Hz_10ms_modulation_mae_db": 0.8783464087803239, "300–1000 Hz_40ms_log_rms_mae_db": 0.8014050639340913, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07065326049102852, "300–1000 Hz_40ms_envelope_correlation": 0.9958491753844187, "300–1000 Hz_40ms_modulation_mae_db": 1.1236647462029927, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9335922306025606, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2015648423477769, "1000–3000 Hz_10ms_envelope_correlation": 0.9696709866971457, "1000–3000 Hz_10ms_modulation_mae_db": 0.6493511109328715, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2438815933700325, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1215723410569153, "1000–3000 Hz_40ms_envelope_correlation": 0.9891189455777111, "1000–3000 Hz_40ms_modulation_mae_db": 0.37320125235876245, "3000–6000 Hz_10ms_log_rms_mae_db": 2.140466944290174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24635290321419628, "3000–6000 Hz_10ms_envelope_correlation": 0.9589305343389245, "3000–6000 Hz_10ms_modulation_mae_db": 1.7770201897804585, "3000–6000 Hz_40ms_log_rms_mae_db": 1.847567698136238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19166798985116354, "3000–6000 Hz_40ms_envelope_correlation": 0.9744530492346479, "3000–6000 Hz_40ms_modulation_mae_db": 2.881403665137991, "mr_stft_db": 6.183730829046794, "band_mae_db": 0.3395430849223384, "rms_gain_error_db": -0.09405844103549912, "output_peak": 0.7016332755008112}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9371146983913793, "全帯域_10ms_rms_envelope_nrmse": 0.08440339989283303, "全帯域_10ms_envelope_correlation": 0.9932124017146275, "全帯域_10ms_modulation_mae_db": 0.37279704090543664, "全帯域_40ms_log_rms_mae_db": 0.5167160378053449, "全帯域_40ms_rms_envelope_nrmse": 0.04355667392144948, "全帯域_40ms_envelope_correlation": 0.9981743754896427, "全帯域_40ms_modulation_mae_db": 0.5379426892165109, "300–1000 Hz_10ms_log_rms_mae_db": 1.2964754684720765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12719777955365855, "300–1000 Hz_10ms_envelope_correlation": 0.9870692115825581, "300–1000 Hz_10ms_modulation_mae_db": 0.5522828365239864, "300–1000 Hz_40ms_log_rms_mae_db": 0.7682155555541499, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07506199273544704, "300–1000 Hz_40ms_envelope_correlation": 0.9956764239631839, "300–1000 Hz_40ms_modulation_mae_db": 0.7913239374842437, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9481229290190596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1915878065832385, "1000–3000 Hz_10ms_envelope_correlation": 0.973598050089013, "1000–3000 Hz_10ms_modulation_mae_db": 0.7257494774668829, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2781403626137517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1330828091259809, "1000–3000 Hz_40ms_envelope_correlation": 0.9878512979594988, "1000–3000 Hz_40ms_modulation_mae_db": 0.8300557317302432, "3000–6000 Hz_10ms_log_rms_mae_db": 2.150167328844835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25142188058310216, "3000–6000 Hz_10ms_envelope_correlation": 0.9573339073145021, "3000–6000 Hz_10ms_modulation_mae_db": 1.7635559191610006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8419072812011161, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1911918850991143, "3000–6000 Hz_40ms_envelope_correlation": 0.9746583259748868, "3000–6000 Hz_40ms_modulation_mae_db": 3.036927658807055, "mr_stft_db": 6.165530884836818, "band_mae_db": 0.20990536211259644, "rms_gain_error_db": -0.10869907527816124, "output_peak": 0.6326809512329212}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.110541743331336, "spectral_convergence": 0.3155609006831057, "spectral_convergence_status": "ok", "amplitude_floor": 2.172633843844252e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.954020298628188, "spectral_convergence": 0.3075333652136106, "spectral_convergence_status": "ok", "amplitude_floor": 1.076658885105748e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.469448528101303, "spectral_convergence": 0.338366586934079, "spectral_convergence_status": "ok", "amplitude_floor": 9.276051433709775e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006687421878497317, "output_power": 0.0006605123769877924, "delta_db": -0.05377761654537222, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0031496712549154435, "output_power": 0.0030765871538702095, "delta_db": -0.1019600458789657, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007488394009410867, "output_power": 0.0007203863678858077, "delta_db": -0.16823201249541025, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002473939509459554, "output_power": 0.0002221714099125647, "delta_db": -0.4670090535133165, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011342358064728732, "output_power": 9.548886026554308e-05, "delta_db": -0.7475064359942514, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.620279276522665e-05, "output_power": 5.010777247954302e-05, "delta_db": 0.3523686915744661, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.31514230933363035, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13209457235786862, "pitch": null, "output_rms_dbfs": -23.16479825854426, "output_peak": 0.626623729675691, "reference_rms_dbfs": -23.03270368618639, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.178003523353609}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023158079660483383, "log_rms_mae_db": 1.028160484431816, "rms_envelope_nrmse": 0.10787437544210587, "envelope_correlation": 0.9888755151064336, "rms_gain_error_db": -0.13166134070048235, "reference_mean_power": 0.004973550834679421, "output_mean_power": 0.004825034546317167}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9210188012408756e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006213961937331268, "output_power": 0.0006022748474091461, "reference_active": true, "delta_db": -0.13573862460130723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026574930428793645, "output_power": 0.00030478735296643304, "reference_active": true, "delta_db": 0.5952480591361895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2739241814293584e-05, "output_power": 3.446441749045078e-05, "reference_active": true, "delta_db": 0.22302324884613223}], "active_band_count": 3, "active_band_mae_db": 0.31800331086120964}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000211279420280134, "log_rms_mae_db": 0.5790922814181473, "rms_envelope_nrmse": 0.06663441227019418, "envelope_correlation": 0.9955754418356879, "rms_gain_error_db": -0.1315598593298291, "reference_mean_power": 0.0049805093532051875, "output_mean_power": 0.004831898180400407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7395423077013874e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005644300449164411, "output_power": 0.0005572089957579224, "reference_active": true, "delta_db": -0.05592004231949194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013986675109572895, "output_power": 0.00016425958383402376, "reference_active": true, "delta_db": 0.698162312106192}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.497453327177472e-06, "output_power": 2.566822759427764e-06, "reference_active": true, "delta_db": 0.11898500619174594}], "active_band_count": 3, "active_band_mae_db": 0.2910224535391433}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020019856693346264, "log_rms_mae_db": 1.378843211884009, "rms_envelope_nrmse": 0.16631461039324305, "envelope_correlation": 0.9776435361315802, "rms_gain_error_db": -0.323691670432657, "reference_mean_power": 0.0028106313477950454, "output_mean_power": 0.002608763555423738}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3651110338311905e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046335294017358356, "output_power": 0.0004253838332032301, "reference_active": true, "delta_db": -0.3713094271843861}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024122957535154213, "output_power": 0.00023503531070171287, "reference_active": true, "delta_db": -0.11297438603810334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.948878321904196e-05, "output_power": 3.236973812166155e-05, "reference_active": true, "delta_db": 0.40482333033557333}], "active_band_count": 3, "active_band_mae_db": 0.2963690478526876}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018155803316515472, "log_rms_mae_db": 0.8234993136713396, "rms_envelope_nrmse": 0.11443366390797359, "envelope_correlation": 0.9892754766163724, "rms_gain_error_db": -0.3248115312687462, "reference_mean_power": 0.002819303754025505, "output_mean_power": 0.0026161384061299236}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2479023732657914e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004464610508434935, "output_power": 0.0004154075682611241, "reference_active": true, "delta_db": -0.3130917249850138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001291081626943261, "output_power": 0.00013229865452728365, "reference_active": true, "delta_db": 0.10601726616474431}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4462923050468237e-06, "output_power": 2.377697726765857e-06, "reference_active": true, "delta_db": -0.12351706649155805}], "active_band_count": 3, "active_band_mae_db": 0.1808753525471054}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.07685175719321e-05, "log_rms_mae_db": 1.9635798220596181, "rms_envelope_nrmse": 0.1973748714921216, "envelope_correlation": 0.9718093654093161, "rms_gain_error_db": -0.5154697979878792, "reference_mean_power": 0.0002950078039052426, "output_mean_power": 0.00026199109185573715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.385626430084388e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.894665895613538e-05, "output_power": 8.851258053437006e-05, "reference_active": true, "delta_db": -0.48396131810675697}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.27901607459427e-05, "output_power": 1.920569230583193e-05, "reference_active": true, "delta_db": -0.743174216456094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.304247232293777e-06, "output_power": 5.558938642556592e-06, "reference_active": true, "delta_db": 1.1109467160844975}], "active_band_count": 3, "active_band_mae_db": 0.7793607502157828}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.03385602019198e-05, "log_rms_mae_db": 1.3065677253561858, "rms_envelope_nrmse": 0.12044309131511152, "envelope_correlation": 0.9906765422883783, "rms_gain_error_db": -0.5118298590785326, "reference_mean_power": 0.0002923716436469955, "output_mean_power": 0.00025986767648187437}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2489747353423383e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.236418785147827e-05, "output_power": 8.41356052491217e-05, "reference_active": true, "delta_db": -0.40523792606532544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2735133445304299e-05, "output_power": 1.1038078280487229e-05, "reference_active": true, "delta_db": -0.6211003039021683}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6865666311394477e-07, "output_power": 3.383571555310527e-07, "reference_active": true, "delta_db": 1.001777490827239}], "active_band_count": 3, "active_band_mae_db": 0.6760385735982443}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.524569706134418e-05, "log_rms_mae_db": 2.063474770073016, "rms_envelope_nrmse": 0.2504184556664843, "envelope_correlation": 0.9575756626115172, "rms_gain_error_db": -0.2783042585024832, "reference_mean_power": 6.736853190004857e-05, "output_mean_power": 6.318684302162474e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.630234817861519e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0381998821210514e-05, "output_power": 1.9945394942781247e-05, "reference_active": true, "delta_db": -0.09404131852842604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.955680541944287e-06, "output_power": 4.399549420572041e-06, "reference_active": true, "delta_db": 0.4618698952040924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.594012590914479e-07, "output_power": 1.9356044615153494e-06, "reference_active": true, "delta_db": 4.676668424046253}], "active_band_count": 3, "active_band_mae_db": 1.7441932125929238}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.8717196486707034e-05, "log_rms_mae_db": 1.778834503808462, "rms_envelope_nrmse": 0.18764000050611496, "envelope_correlation": 0.9757785599934886, "rms_gain_error_db": -0.2791548648210409, "reference_mean_power": 6.764325505404543e-05, "output_mean_power": 6.343208861943855e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.430902640057432e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9875215171126693e-05, "output_power": 2.088018963552409e-05, "reference_active": true, "delta_db": 0.21422599574014967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8203988319997446e-06, "output_power": 2.3008664045620673e-06, "reference_active": true, "delta_db": 1.0172485473303334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.685144573008789e-08, "output_power": 2.2074058144969815e-07, "reference_active": true, "delta_db": 7.774276517193522}], "active_band_count": 3, "active_band_mae_db": 3.0019170200880017}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7948908241288073, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6ee39de654df69dda3147e4c7140b0c8bda845c891ce4d8b2a9e0e19dd26d085", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9710875579470533, "standard_deviation": 0.0017753755844008138, "minimum": 0.9686710108701242, "maximum": 0.973598050089013, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.932142282517841, "standard_deviation": 0.024128484596896777, "minimum": 1.897151315405612, "maximum": 1.9635798220596181, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8224194841828822, "standard_deviation": 0.15041262667344762, "minimum": 0.6493511109328715, "maximum": 1.1043221919316897, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19916406138433046, "standard_deviation": 0.00584230335781541, "minimum": 0.19134561742183798, "maximum": 0.20932179215246943, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.989360124344856, "standard_deviation": 0.0012135758953957321, "minimum": 0.9878512979594988, "maximum": 0.9908874889918476, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2495271088084916, "standard_deviation": 0.034462573061189376, "minimum": 1.1949831901111436, "maximum": 1.3065677253561858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7299376196490814, "standard_deviation": 0.21148153718585475, "minimum": 0.37320125235876245, "maximum": 1.0974556637307575, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12547412456302026, "standard_deviation": 0.00685403355528867, "minimum": 0.11846114378131367, "maximum": 0.13454526202113457, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.955012515437557, "standard_deviation": 0.0043392818139245766, "minimum": 0.9480437539596684, "maximum": 0.9607389138135712, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1415259297137945, "standard_deviation": 0.03887420987791543, "minimum": 2.063474770073016, "maximum": 2.1946428818788406, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.8247615755071616, "standard_deviation": 0.09001811294140691, "minimum": 1.7440544686615385, "maximum": 1.987359757814537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2576086166555902, "standard_deviation": 0.012152808716510448, "minimum": 0.2412029834201904, "maximum": 0.27657207278268975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9737077453679159, "standard_deviation": 0.001953738829567543, "minimum": 0.9706842646016514, "maximum": 0.9763938148324475, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8523059603993808, "standard_deviation": 0.04201678342833382, "minimum": 1.778834503808462, "maximum": 1.915489696240276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.8422805067946193, "standard_deviation": 0.1824795599869252, "minimum": 2.580952221226788, "maximum": 3.036927658807055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19519192249002246, "standard_deviation": 0.007615137135899608, "minimum": 0.18538616388320164, "maximum": 0.20728791166224872, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9847640758242915, "standard_deviation": 0.003303881881722054, "minimum": 0.9776435361315802, "maximum": 0.9879862319166415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3208763677700723, "standard_deviation": 0.036417806156526185, "minimum": 1.2696665099161026, "maximum": 1.378843211884009, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5149094364095405, "standard_deviation": 0.21666375876992033, "minimum": 0.20748060981219388, "maximum": 0.8783464087803239, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13681812934699147, "standard_deviation": 0.014046129505229397, "minimum": 0.1210310104179908, "maximum": 0.16631461039324305, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9947271108413988, "standard_deviation": 0.002282679239130621, "minimum": 0.9892754766163724, "maximum": 0.9963882907709559, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8053135697413758, "standard_deviation": 0.04215104197226083, "minimum": 0.751196071409582, "maximum": 0.8765324613190123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5994535066709311, "standard_deviation": 0.3096691384953753, "minimum": 0.1808753525471054, "maximum": 1.1236647462029927, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08003923224815049, "standard_deviation": 0.014808786449266007, "minimum": 0.06507790572332188, "maximum": 0.11443366390797359, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2790907507899525, "standard_deviation": 0.052677996902459674, "minimum": 0.20990536211259644, "maximum": 0.3475855643983954, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.178318403929934, "standard_deviation": 0.008353482666081626, "minimum": 6.165530884836818, "maximum": 6.189809004947485, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6520222712744042, "standard_deviation": 0.041820181989284985, "minimum": 0.6022457201713901, "maximum": 0.7110246458697678, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09964713414866072, "standard_deviation": 0.06594749295405167, "minimum": -0.1927802631820161, "maximum": -0.015974282145014913, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9908095819895508, "standard_deviation": 0.0013413690493957328, "minimum": 0.9888755151064336, "maximum": 0.9932124017146275, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9888357817238914, "standard_deviation": 0.03525127975038079, "minimum": 0.9371146983913793, "maximum": 1.028160484431816, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41728349621958544, "standard_deviation": 0.19590418352318975, "minimum": 0.13918206406004116, "maximum": 0.6877229165804208, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09790047286439127, "standard_deviation": 0.007198281419279994, "minimum": 0.08440339989283303, "maximum": 0.10787437544210587, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997056095093251, "standard_deviation": 0.000755389282080672, "minimum": 0.9955754418356879, "maximum": 0.9981743754896427, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5462204963536734, "standard_deviation": 0.021395842323985177, "minimum": 0.5167160378053449, "maximum": 0.5790922814181473, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4363219816953415, "standard_deviation": 0.22086191125128682, "minimum": 0.11082571222835862, "maximum": 0.7459410155523734, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05432192039004426, "standard_deviation": 0.006821823052453564, "minimum": 0.04355667392144948, "maximum": 0.06663441227019418, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_7729-102255-0042", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.062997817993164, "predicted_mos": 3.062997817993164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "dc4c9df85acc05ba4e2143abccf2fb30d7a9c6866e6a536b8905cbb76ecd6407", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.07674239994958, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9785010240651106, "全帯域_10ms_rms_envelope_nrmse": 0.09522426293966696, "全帯域_10ms_envelope_correlation": 0.9913195541252893, "全帯域_10ms_modulation_mae_db": 0.15910969704430955, "全帯域_40ms_log_rms_mae_db": 0.5363403944814025, "全帯域_40ms_rms_envelope_nrmse": 0.05011351140513811, "全帯域_40ms_envelope_correlation": 0.9975350426117718, "全帯域_40ms_modulation_mae_db": 0.1887451302828552, "300–1000 Hz_10ms_log_rms_mae_db": 1.2878357176848398, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1330459045628777, "300–1000 Hz_10ms_envelope_correlation": 0.9856036551016981, "300–1000 Hz_10ms_modulation_mae_db": 0.516819696441959, "300–1000 Hz_40ms_log_rms_mae_db": 0.7622168107311805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07131835549595529, "300–1000 Hz_40ms_envelope_correlation": 0.9959077922810786, "300–1000 Hz_40ms_modulation_mae_db": 0.3248377560187512, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9482855821603855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2285825489763228, "1000–3000 Hz_10ms_envelope_correlation": 0.9618400138231241, "1000–3000 Hz_10ms_modulation_mae_db": 1.5434936300101203, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2517347269819254, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14440443171981804, "1000–3000 Hz_40ms_envelope_correlation": 0.9864402503616432, "1000–3000 Hz_40ms_modulation_mae_db": 1.585210210957135, "3000–6000 Hz_10ms_log_rms_mae_db": 2.142739495952024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25501002718480903, "3000–6000 Hz_10ms_envelope_correlation": 0.9560091360866152, "3000–6000 Hz_10ms_modulation_mae_db": 1.7122267149964745, "3000–6000 Hz_40ms_log_rms_mae_db": 1.864089881388722, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1897328551630573, "3000–6000 Hz_40ms_envelope_correlation": 0.9749904000418549, "3000–6000 Hz_40ms_modulation_mae_db": 3.0298508956859718, "mr_stft_db": 6.171213490049499, "band_mae_db": 0.3263757737887418, "rms_gain_error_db": -0.11062391426697801, "output_peak": 0.619797978122592}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.056758714044845, "全帯域_10ms_rms_envelope_nrmse": 0.1127655172533642, "全帯域_10ms_envelope_correlation": 0.9881366205399382, "全帯域_10ms_modulation_mae_db": 0.4208894547398521, "全帯域_40ms_log_rms_mae_db": 0.6163098380458639, "全帯域_40ms_rms_envelope_nrmse": 0.06251174537906272, "全帯域_40ms_envelope_correlation": 0.9966136410886683, "全帯域_40ms_modulation_mae_db": 0.6014329172695859, "300–1000 Hz_10ms_log_rms_mae_db": 1.3997340069880173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16042300932208373, "300–1000 Hz_10ms_envelope_correlation": 0.9798440841061968, "300–1000 Hz_10ms_modulation_mae_db": 0.4446452750095997, "300–1000 Hz_40ms_log_rms_mae_db": 0.894865458179431, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09560251471578755, "300–1000 Hz_40ms_envelope_correlation": 0.9937251165559773, "300–1000 Hz_40ms_modulation_mae_db": 0.6391627100397848, "1000–3000 Hz_10ms_log_rms_mae_db": 1.890569452093403, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19960397708908098, "1000–3000 Hz_10ms_envelope_correlation": 0.9703130171827838, "1000–3000 Hz_10ms_modulation_mae_db": 0.6545742616508825, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2612507354363123, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12445499322265523, "1000–3000 Hz_40ms_envelope_correlation": 0.9886611124826483, "1000–3000 Hz_40ms_modulation_mae_db": 0.6105308128998214, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1013718814449014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24133694555867452, "3000–6000 Hz_10ms_envelope_correlation": 0.9607185066095281, "3000–6000 Hz_10ms_modulation_mae_db": 1.6239094325772687, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8474994463555237, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18538460736918913, "3000–6000 Hz_40ms_envelope_correlation": 0.9765172890967502, "3000–6000 Hz_40ms_modulation_mae_db": 2.496138218880809, "mr_stft_db": 6.169858482483648, "band_mae_db": 0.2781777562872305, "rms_gain_error_db": -0.23643460296274033, "output_peak": 0.5778297146794439}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0102383339714733, "全帯域_10ms_rms_envelope_nrmse": 0.10360601707091603, "全帯域_10ms_envelope_correlation": 0.9899070392066708, "全帯域_10ms_modulation_mae_db": 0.7744179947061295, "全帯域_40ms_log_rms_mae_db": 0.5698340547202919, "全帯域_40ms_rms_envelope_nrmse": 0.06039888592931776, "全帯域_40ms_envelope_correlation": 0.9966614189143033, "全帯域_40ms_modulation_mae_db": 0.5170897913143248, "300–1000 Hz_10ms_log_rms_mae_db": 1.2863782800085106, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15008185649382672, "300–1000 Hz_10ms_envelope_correlation": 0.9817522956175891, "300–1000 Hz_10ms_modulation_mae_db": 0.8798756081827438, "300–1000 Hz_40ms_log_rms_mae_db": 0.809314598398617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09333916000412956, "300–1000 Hz_40ms_envelope_correlation": 0.9929456472232518, "300–1000 Hz_40ms_modulation_mae_db": 0.5702066300163575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8822899202193508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20469529330171754, "1000–3000 Hz_10ms_envelope_correlation": 0.9709849115058707, "1000–3000 Hz_10ms_modulation_mae_db": 1.1787196762674577, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2233866849095567, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13377949000856962, "1000–3000 Hz_40ms_envelope_correlation": 0.989916324686412, "1000–3000 Hz_40ms_modulation_mae_db": 1.4746425857740857, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1018402343143032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2464259530938424, "3000–6000 Hz_10ms_envelope_correlation": 0.9590709115769164, "3000–6000 Hz_10ms_modulation_mae_db": 1.6778899616693461, "3000–6000 Hz_40ms_log_rms_mae_db": 1.809976775683017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1920929621941201, "3000–6000 Hz_40ms_envelope_correlation": 0.9741914063829944, "3000–6000 Hz_40ms_modulation_mae_db": 2.8021332229674116, "mr_stft_db": 6.153118838335604, "band_mae_db": 0.32211294592858514, "rms_gain_error_db": -0.19310582649001046, "output_peak": 0.6840243419356656}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9481077923940144, "全帯域_10ms_rms_envelope_nrmse": 0.11184973537420553, "全帯域_10ms_envelope_correlation": 0.9879940765154213, "全帯域_10ms_modulation_mae_db": 0.3685364831832742, "全帯域_40ms_log_rms_mae_db": 0.5106813616618107, "全帯域_40ms_rms_envelope_nrmse": 0.062284241907792616, "全帯域_40ms_envelope_correlation": 0.9960972749816318, "全帯域_40ms_modulation_mae_db": 0.21444567390756517, "300–1000 Hz_10ms_log_rms_mae_db": 1.316513128703841, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16140045254814622, "300–1000 Hz_10ms_envelope_correlation": 0.9786807228040737, "300–1000 Hz_10ms_modulation_mae_db": 0.4503100462365483, "300–1000 Hz_40ms_log_rms_mae_db": 0.8513154959287815, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09023283068754236, "300–1000 Hz_40ms_envelope_correlation": 0.9931849513944608, "300–1000 Hz_40ms_modulation_mae_db": 0.21361856483468553, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8671353166217415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19954308522790862, "1000–3000 Hz_10ms_envelope_correlation": 0.9716035546894051, "1000–3000 Hz_10ms_modulation_mae_db": 0.9155536208538688, "1000–3000 Hz_40ms_log_rms_mae_db": 1.199129376698497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12007179706007619, "1000–3000 Hz_40ms_envelope_correlation": 0.9915252270418179, "1000–3000 Hz_40ms_modulation_mae_db": 1.0074048821700006, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0927110687956505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2600809235724582, "3000–6000 Hz_10ms_envelope_correlation": 0.9541081055049463, "3000–6000 Hz_10ms_modulation_mae_db": 2.2048384271247143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8189922361041813, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20121936546065755, "3000–6000 Hz_40ms_envelope_correlation": 0.9717212653803414, "3000–6000 Hz_40ms_modulation_mae_db": 3.364889649445249, "mr_stft_db": 6.160295124282842, "band_mae_db": 0.34004247529482434, "rms_gain_error_db": -0.10060306833766637, "output_peak": 0.7543511113280158}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9597058425776964, "全帯域_10ms_rms_envelope_nrmse": 0.10041536906858714, "全帯域_10ms_envelope_correlation": 0.990295345970092, "全帯域_10ms_modulation_mae_db": 0.6922183218470129, "全帯域_40ms_log_rms_mae_db": 0.5173076020885482, "全帯域_40ms_rms_envelope_nrmse": 0.05324691153836078, "全帯域_40ms_envelope_correlation": 0.9971105747929484, "全帯域_40ms_modulation_mae_db": 0.5530401694503154, "300–1000 Hz_10ms_log_rms_mae_db": 1.3125791649432947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13871481993917656, "300–1000 Hz_10ms_envelope_correlation": 0.9842619007179634, "300–1000 Hz_10ms_modulation_mae_db": 0.5207947374914028, "300–1000 Hz_40ms_log_rms_mae_db": 0.7998120265277409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07266473468980199, "300–1000 Hz_40ms_envelope_correlation": 0.9956374029242735, "300–1000 Hz_40ms_modulation_mae_db": 0.40318110908141763, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8108810186905242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19181520272132038, "1000–3000 Hz_10ms_envelope_correlation": 0.9731066416694883, "1000–3000 Hz_10ms_modulation_mae_db": 0.8728853774925817, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1770555982033208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12607148707338772, "1000–3000 Hz_40ms_envelope_correlation": 0.9888891715220065, "1000–3000 Hz_40ms_modulation_mae_db": 0.7832877756264779, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1297423529903807, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.248990121513252, "3000–6000 Hz_10ms_envelope_correlation": 0.9581131066391709, "3000–6000 Hz_10ms_modulation_mae_db": 1.902920648459208, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8394613665561812, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1904278716167218, "3000–6000 Hz_40ms_envelope_correlation": 0.9750883816971531, "3000–6000 Hz_40ms_modulation_mae_db": 2.7177624749456624, "mr_stft_db": 6.152826912919976, "band_mae_db": 0.2600000672572604, "rms_gain_error_db": -0.07242701665630033, "output_peak": 0.6149059936387153}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9754611107967501, "全帯域_10ms_rms_envelope_nrmse": 0.09937858794997459, "全帯域_10ms_envelope_correlation": 0.9904963332622031, "全帯域_10ms_modulation_mae_db": 0.2492597764426103, "全帯域_40ms_log_rms_mae_db": 0.49968774875731825, "全帯域_40ms_rms_envelope_nrmse": 0.04822515745426744, "全帯域_40ms_envelope_correlation": 0.9976565946456383, "全帯域_40ms_modulation_mae_db": 0.2683465560560963, "300–1000 Hz_10ms_log_rms_mae_db": 1.3181812048024724, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14436065801285133, "300–1000 Hz_10ms_envelope_correlation": 0.9829253836874168, "300–1000 Hz_10ms_modulation_mae_db": 0.43496582748500195, "300–1000 Hz_40ms_log_rms_mae_db": 0.7585749456643986, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07584174072306842, "300–1000 Hz_40ms_envelope_correlation": 0.9951862909517829, "300–1000 Hz_40ms_modulation_mae_db": 0.5786436688604186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9460451265559657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19620367308700673, "1000–3000 Hz_10ms_envelope_correlation": 0.9726631150596674, "1000–3000 Hz_10ms_modulation_mae_db": 0.7568953215973112, "1000–3000 Hz_40ms_log_rms_mae_db": 1.237484479995269, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11742951787695696, "1000–3000 Hz_40ms_envelope_correlation": 0.9919272202163205, "1000–3000 Hz_40ms_modulation_mae_db": 0.7556624161805251, "3000–6000 Hz_10ms_log_rms_mae_db": 2.063232882542397, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2537019377466651, "3000–6000 Hz_10ms_envelope_correlation": 0.9567116168290288, "3000–6000 Hz_10ms_modulation_mae_db": 2.3900581060232526, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7979676476320259, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19418467084196217, "3000–6000 Hz_40ms_envelope_correlation": 0.9737809110551104, "3000–6000 Hz_40ms_modulation_mae_db": 3.5721414284747053, "mr_stft_db": 6.152867857098291, "band_mae_db": 0.41345835067897013, "rms_gain_error_db": -0.08130962439849737, "output_peak": 0.6879420040409165}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9837827892713765, "全帯域_10ms_rms_envelope_nrmse": 0.09967008190399436, "全帯域_10ms_envelope_correlation": 0.9905143291297539, "全帯域_10ms_modulation_mae_db": 0.37394282584685823, "全帯域_40ms_log_rms_mae_db": 0.5715508628076318, "全帯域_40ms_rms_envelope_nrmse": 0.054812384383526265, "全帯域_40ms_envelope_correlation": 0.997042374514839, "全帯域_40ms_modulation_mae_db": 0.665794187559854, "300–1000 Hz_10ms_log_rms_mae_db": 1.3196625221533747, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13217812599926917, "300–1000 Hz_10ms_envelope_correlation": 0.9858365463089068, "300–1000 Hz_10ms_modulation_mae_db": 0.506421213659192, "300–1000 Hz_40ms_log_rms_mae_db": 0.7946501701370742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07454188542572725, "300–1000 Hz_40ms_envelope_correlation": 0.9955317877157854, "300–1000 Hz_40ms_modulation_mae_db": 0.6317465325066477, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9488220522850686, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1854587420984208, "1000–3000 Hz_10ms_envelope_correlation": 0.9756690869734711, "1000–3000 Hz_10ms_modulation_mae_db": 1.4062148718267042, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2797702296773512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11555067946347526, "1000–3000 Hz_40ms_envelope_correlation": 0.9919545363323566, "1000–3000 Hz_40ms_modulation_mae_db": 1.6940637351470962, "3000–6000 Hz_10ms_log_rms_mae_db": 2.123331375358892, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25340848485583534, "3000–6000 Hz_10ms_envelope_correlation": 0.9566330826180673, "3000–6000 Hz_10ms_modulation_mae_db": 2.0860638448294435, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8109965126618517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19089634077720027, "3000–6000 Hz_40ms_envelope_correlation": 0.9753124205932507, "3000–6000 Hz_40ms_modulation_mae_db": 3.164490824732253, "mr_stft_db": 6.158354535856744, "band_mae_db": 0.26712317231312604, "rms_gain_error_db": -0.11232902380691252, "output_peak": 0.6803966416268472}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9569072888174163, "全帯域_10ms_rms_envelope_nrmse": 0.09982888483563729, "全帯域_10ms_envelope_correlation": 0.9904810124899853, "全帯域_10ms_modulation_mae_db": 0.2022955305737293, "全帯域_40ms_log_rms_mae_db": 0.5210604606567444, "全帯域_40ms_rms_envelope_nrmse": 0.04805409722813907, "全帯域_40ms_envelope_correlation": 0.9978171642256318, "全帯域_40ms_modulation_mae_db": 0.4758434123802677, "300–1000 Hz_10ms_log_rms_mae_db": 1.2965380678893046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13612970940174116, "300–1000 Hz_10ms_envelope_correlation": 0.9852293494139376, "300–1000 Hz_10ms_modulation_mae_db": 0.5608666916392467, "300–1000 Hz_40ms_log_rms_mae_db": 0.7933892172585937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07192461773514021, "300–1000 Hz_40ms_envelope_correlation": 0.9963211835454525, "300–1000 Hz_40ms_modulation_mae_db": 0.8856664144760354, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8690034701475398, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2010465905851976, "1000–3000 Hz_10ms_envelope_correlation": 0.970675974198967, "1000–3000 Hz_10ms_modulation_mae_db": 1.0044359927303592, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2262655529852702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12319219844337212, "1000–3000 Hz_40ms_envelope_correlation": 0.9901871046260485, "1000–3000 Hz_40ms_modulation_mae_db": 0.8227267753115779, "3000–6000 Hz_10ms_log_rms_mae_db": 2.092153135478816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25253135564717005, "3000–6000 Hz_10ms_envelope_correlation": 0.9569468001215344, "3000–6000 Hz_10ms_modulation_mae_db": 1.5725879917118661, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8490698177967948, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1938306184057506, "3000–6000 Hz_40ms_envelope_correlation": 0.9737805278024008, "3000–6000 Hz_40ms_modulation_mae_db": 2.8949770801747157, "mr_stft_db": 6.159175577043061, "band_mae_db": 0.36811055083150035, "rms_gain_error_db": -0.13208033489763693, "output_peak": 0.7167941064257032}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.106238908959763, "spectral_convergence": 0.30651056983538694, "spectral_convergence_status": "ok", "amplitude_floor": 2.172633843844252e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.926269665443899, "spectral_convergence": 0.2936849529263512, "spectral_convergence_status": "ok", "amplitude_floor": 1.076658885105748e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.481131895744836, "spectral_convergence": 0.3215828445638378, "spectral_convergence_status": "ok", "amplitude_floor": 9.276051433709775e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006687421878497317, "output_power": 0.0006689603185787371, "delta_db": 0.0014163533719189136, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0031496712549154435, "output_power": 0.0030343114247759735, "delta_db": -0.1620507462589529, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007488394009410867, "output_power": 0.0007812009284641469, "delta_db": 0.18374063605073487, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002473939509459554, "output_power": 0.0002179446645896242, "delta_db": -0.5504283473958754, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011342358064728732, "output_power": 9.65447795944591e-05, "delta_db": -0.699745579095466, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.620279276522665e-05, "output_power": 5.020598889476955e-05, "delta_db": 0.36087298055950284, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3263757737887418, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11062391426697801, "pitch": null, "output_rms_dbfs": -23.143327600453368, "output_peak": 0.619797978122592, "reference_rms_dbfs": -23.03270368618639, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.171213490049499}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023158079660483383, "log_rms_mae_db": 0.9785010240651106, "rms_envelope_nrmse": 0.09522426293966696, "envelope_correlation": 0.9913195541252893, "rms_gain_error_db": -0.1105902801501785, "reference_mean_power": 0.004973550834679421, "output_mean_power": 0.004848501487777019}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9210188012408756e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006213961937331268, "output_power": 0.0006077762936940651, "reference_active": true, "delta_db": -0.09624832298990138}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026574930428793645, "output_power": 0.00025275818139119596, "reference_active": true, "delta_db": -0.2176691431505046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2739241814293584e-05, "output_power": 3.399458785016852e-05, "reference_active": true, "delta_db": 0.1634116249925227}], "active_band_count": 3, "active_band_mae_db": 0.15910969704430955}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000211279420280134, "log_rms_mae_db": 0.5363403944814025, "rms_envelope_nrmse": 0.05011351140513811, "envelope_correlation": 0.9975350426117718, "rms_gain_error_db": -0.11091470586243632, "reference_mean_power": 0.0049805093532051875, "output_mean_power": 0.004854922364276251}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7395423077013874e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005644300449164411, "output_power": 0.0005551007463428404, "reference_active": true, "delta_db": -0.07238312178763544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013986675109572895, "output_power": 0.00013215841194384956, "reference_active": true, "delta_db": -0.24619675567627325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.497453327177472e-06, "output_power": 2.3590211220895504e-06, "reference_active": true, "delta_db": -0.2476555133846568}], "active_band_count": 3, "active_band_mae_db": 0.1887451302828552}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020019856693346264, "log_rms_mae_db": 1.2878357176848398, "rms_envelope_nrmse": 0.1330459045628777, "envelope_correlation": 0.9856036551016981, "rms_gain_error_db": -0.2058383315793598, "reference_mean_power": 0.0028106313477950454, "output_mean_power": 0.002680526176197134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3651110338311905e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046335294017358356, "output_power": 0.0004455981127979615, "reference_active": true, "delta_db": -0.169685803102588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024122957535154213, "output_power": 0.00020235283975371678, "reference_active": true, "delta_db": -0.7632124880879279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.948878321904196e-05, "output_power": 3.399483461401552e-05, "reference_active": true, "delta_db": 0.617560798135361}], "active_band_count": 3, "active_band_mae_db": 0.516819696441959}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018155803316515472, "log_rms_mae_db": 0.7622168107311805, "rms_envelope_nrmse": 0.07131835549595529, "envelope_correlation": 0.9959077922810786, "rms_gain_error_db": -0.20644641899450744, "reference_mean_power": 0.002819303754025505, "output_mean_power": 0.002688420681750054}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2479023732657914e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004464610508434935, "output_power": 0.0004302941920740622, "reference_active": true, "delta_db": -0.1601809292723265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001291081626943261, "output_power": 0.0001072100822386624, "reference_active": true, "delta_db": -0.8071807168291423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4462923050468237e-06, "output_power": 2.450323986557995e-06, "reference_active": true, "delta_db": 0.007151621954784809}], "active_band_count": 3, "active_band_mae_db": 0.3248377560187512}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.07685175719321e-05, "log_rms_mae_db": 1.9482855821603855, "rms_envelope_nrmse": 0.2285825489763228, "envelope_correlation": 0.9618400138231241, "rms_gain_error_db": -0.613190766916985, "reference_mean_power": 0.0002950078039052426, "output_mean_power": 0.0002561618363538715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.385626430084388e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.894665895613538e-05, "output_power": 7.849201071856757e-05, "reference_active": true, "delta_db": -1.0057567986914733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.27901607459427e-05, "output_power": 1.733423177619658e-05, "reference_active": true, "delta_db": -1.1884278917961137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.304247232293777e-06, "output_power": 7.542699899449319e-06, "reference_active": true, "delta_db": 2.436296199542774}], "active_band_count": 3, "active_band_mae_db": 1.5434936300101203}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.03385602019198e-05, "log_rms_mae_db": 1.2517347269819254, "rms_envelope_nrmse": 0.14440443171981804, "envelope_correlation": 0.9864402503616432, "rms_gain_error_db": -0.6193746323880523, "reference_mean_power": 0.0002923716436469955, "output_mean_power": 0.0002535115706729737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2489747353423383e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.236418785147827e-05, "output_power": 7.601107007972505e-05, "reference_active": true, "delta_db": -0.8462676933209058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2735133445304299e-05, "output_power": 1.0447642295659264e-05, "reference_active": true, "delta_db": -0.8598520518960655}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6865666311394477e-07, "output_power": 5.421865064891862e-07, "reference_active": true, "delta_db": 3.0495108876544337}], "active_band_count": 3, "active_band_mae_db": 1.585210210957135}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.524569706134418e-05, "log_rms_mae_db": 2.142739495952024, "rms_envelope_nrmse": 0.25501002718480903, "envelope_correlation": 0.9560091360866152, "rms_gain_error_db": -0.19661905185262551, "reference_mean_power": 6.736853190004857e-05, "output_mean_power": 6.43865534119418e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.630234817861519e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0381998821210514e-05, "output_power": 2.138029511492658e-05, "reference_active": true, "delta_db": 0.2076692339965275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.955680541944287e-06, "output_power": 4.247702871177577e-06, "reference_active": true, "delta_db": 0.3093291901579061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.594012590914479e-07, "output_power": 1.910372000169452e-06, "reference_active": true, "delta_db": 4.61968172083499}], "active_band_count": 3, "active_band_mae_db": 1.7122267149964745}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.8717196486707034e-05, "log_rms_mae_db": 1.864089881388722, "rms_envelope_nrmse": 0.1897328551630573, "envelope_correlation": 0.9749904000418549, "rms_gain_error_db": -0.195940907002929, "reference_mean_power": 6.764325505404543e-05, "output_mean_power": 6.465921196944034e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.430902640057432e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9875215171126693e-05, "output_power": 2.209673464561032e-05, "reference_active": true, "delta_db": 0.46016261303506645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8203988319997446e-06, "output_power": 2.2461506061538474e-06, "reference_active": true, "delta_db": 0.9127232450282905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.685144573008789e-08, "output_power": 2.1783176648825296e-07, "reference_active": true, "delta_db": 7.716666828994558}], "active_band_count": 3, "active_band_mae_db": 3.0298508956859718}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8004682366614253, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1537268a0e224c855996acdcf6e888937deb8b07ac90986a529e4b942aeb5db4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9708570393878473, "standard_deviation": 0.004030022206520497, "minimum": 0.9618400138231241, "maximum": 0.9756690869734711, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8953789923467474, "standard_deviation": 0.0493530451880541, "minimum": 1.8108810186905242, "maximum": 1.9488220522850686, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0415965940536607, "standard_deviation": 0.3117284026561526, "minimum": 0.6545742616508825, "maximum": 1.5434936300101203, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20086863913587194, "standard_deviation": 0.012690174807856489, "minimum": 0.1854587420984208, "maximum": 0.2285825489763228, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9899376184086567, "standard_deviation": 0.0019112435271644794, "minimum": 0.9864402503616432, "maximum": 0.9919545363323566, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2320096731109378, "standard_deviation": 0.0333028888338219, "minimum": 1.1770555982033208, "maximum": 1.2797702296773512, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0916911492583399, "standard_deviation": 0.4262887733681641, "minimum": 0.6105308128998214, "maximum": 1.6940637351470962, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1256193243585389, "standard_deviation": 0.0094577649102986, "minimum": 0.11555067946347526, "maximum": 0.14440443171981804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9572889082482259, "standard_deviation": 0.002009005342325095, "minimum": 0.9541081055049463, "maximum": 0.9607185066095281, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1058903033596708, "standard_deviation": 0.02521295305000006, "minimum": 2.063232882542397, "maximum": 2.142739495952024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.8963118909239467, "standard_deviation": 0.3013349303070267, "minimum": 1.5725879917118661, "maximum": 2.3900581060232526, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2514357186465883, "standard_deviation": 0.0057421640994223, "minimum": 0.24133694555867452, "maximum": 0.2600809235724582, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.974422825256232, "standard_deviation": 0.0014202537738487882, "minimum": 0.9717212653803414, "maximum": 0.9765172890967502, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8297567105222874, "standard_deviation": 0.023392175226747938, "minimum": 1.7979676476320259, "maximum": 1.864089881388722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.005297974413347, "standard_deviation": 0.35293257545962, "minimum": 2.496138218880809, "maximum": 3.5721414284747053, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19222116147858237, "standard_deviation": 0.004556486889355285, "minimum": 0.18538460736918913, "maximum": 0.20121936546065755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9830167422197229, "standard_deviation": 0.00271407598354581, "minimum": 0.9786807228040737, "maximum": 0.9858365463089068, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3171777616467069, "standard_deviation": 0.036015599750718025, "minimum": 1.2863782800085106, "maximum": 1.3997340069880173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5393373870182118, "standard_deviation": 0.14449420535244034, "minimum": 0.43496582748500195, "maximum": 0.8798756081827438, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14454181703499658, "standard_deviation": 0.011691786037355332, "minimum": 0.13217812599926917, "maximum": 0.16140045254814622, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9948050215740079, "standard_deviation": 0.0013163455713471098, "minimum": 0.9929456472232518, "maximum": 0.9963211835454525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8080173403532271, "standard_deviation": 0.04538917119234398, "minimum": 0.7585749456643986, "maximum": 0.894865458179431, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5308829232292623, "standard_deviation": 0.21078621047441287, "minimum": 0.21361856483468553, "maximum": 0.8856664144760354, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08068322993464408, "standard_deviation": 0.010445908672179387, "minimum": 0.07131835549595529, "maximum": 0.09560251471578755, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.32192513654752986, "standard_deviation": 0.05293028563188099, "minimum": 0.2600000672572604, "maximum": 0.41345835067897013, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.159713852258708, "standard_deviation": 0.0073233710232806695, "minimum": 6.152826912919976, "maximum": 6.171213490049499, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6670052364747374, "standard_deviation": 0.05844495747076045, "minimum": 0.5778297146794439, "maximum": 0.7543511113280158, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1298641764770928, "standard_deviation": 0.056760306842891184, "minimum": -0.23643460296274033, "maximum": -0.07242701665630033, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9898930389049192, "standard_deviation": 0.0011942561850311402, "minimum": 0.9879940765154213, "maximum": 0.9913195541252893, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9836828619923352, "standard_deviation": 0.035248396566375025, "minimum": 0.9481077923940144, "maximum": 1.056758714044845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.405083760547972, "standard_deviation": 0.22274801080861434, "minimum": 0.15910969704430955, "maximum": 0.7744179947061295, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10284230704954325, "standard_deviation": 0.006270712178821583, "minimum": 0.09522426293966696, "maximum": 0.1127655172533642, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970667607219291, "standard_deviation": 0.0005906816580253727, "minimum": 0.9960972749816318, "maximum": 0.9978171642256318, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5428465404024514, "standard_deviation": 0.0396693489985113, "minimum": 0.49968774875731825, "maximum": 0.6163098380458639, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4355922297776081, "standard_deviation": 0.18530380801191443, "minimum": 0.1887451302828552, "maximum": 0.665794187559854, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0549558669032006, "standard_deviation": 0.006091554254448636, "minimum": 0.04805409722813907, "maximum": 0.06251174537906272, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8224-274381-0010", "language": "en", "style": "normal", "speaker": "librispeech_8224", "source_utterance_id": "8224-274381-0010", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8224-274381-0010.wav", "source_file_sha256": "2b81359935c5b54b25f15e7d5e70937dcdaa397d62ccafeb83b7e1895f9d65e6", "decoded_pcm_sha256": "34d9175c2a6517466fd56565f6c4abc5a8a75b16d60e1e96d0615d4383e19b19", "original_source_sha256": "e9f711f02a8f7cfe4527d6d16f4ad04dc95369de860a6c4f56a645dc7da5a02b", "corpus_entry_id": "librispeech_8224-274381-0010"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.5045268535614014, "predicted_mos": 3.5045268535614014, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ecf849f4bd61cc84bae983ed3f28234ecf5f50711e044834178faf84607794b2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1117712650011526, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8224-274381-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "ecf849f4bd61cc84bae983ed3f28234ecf5f50711e044834178faf84607794b2", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "47f2be273a089762a811311e58170df296cedada30080030c701877e87b9cb90"}, "rows": [{"case": "librispeech_8224-274381-0010", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.9274075031280518, "predicted_mos": 2.9274075031280518, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b744ecac4067538c7eb3a8fd9feb67450013e955c55a518e9176b3aa5de1657e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1565256670000963, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8224-274381-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "b744ecac4067538c7eb3a8fd9feb67450013e955c55a518e9176b3aa5de1657e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2661602095078144, "全帯域_10ms_rms_envelope_nrmse": 0.07269893286329901, "全帯域_10ms_envelope_correlation": 0.9950087755901975, "全帯域_10ms_modulation_mae_db": 0.580084341142535, "全帯域_40ms_log_rms_mae_db": 0.713902130089294, "全帯域_40ms_rms_envelope_nrmse": 0.03873117095704167, "全帯域_40ms_envelope_correlation": 0.9985087796833786, "全帯域_40ms_modulation_mae_db": 0.7476230757557346, "300–1000 Hz_10ms_log_rms_mae_db": 1.693859663589309, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10405331681962583, "300–1000 Hz_10ms_envelope_correlation": 0.9918722319297164, "300–1000 Hz_10ms_modulation_mae_db": 0.78110590863544, "300–1000 Hz_40ms_log_rms_mae_db": 0.9924459722705626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060021196234967066, "300–1000 Hz_40ms_envelope_correlation": 0.9971776880205796, "300–1000 Hz_40ms_modulation_mae_db": 1.1390036152451415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7644245979057314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15136797043411143, "1000–3000 Hz_10ms_envelope_correlation": 0.9819963286252583, "1000–3000 Hz_10ms_modulation_mae_db": 0.5414260267519674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.049537008756208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08244374275843253, "1000–3000 Hz_40ms_envelope_correlation": 0.9945738944417671, "1000–3000 Hz_40ms_modulation_mae_db": 0.7868958416436787, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0638996764030355, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15540499862553817, "3000–6000 Hz_10ms_envelope_correlation": 0.9848410314486483, "3000–6000 Hz_10ms_modulation_mae_db": 0.39846291106175674, "3000–6000 Hz_40ms_log_rms_mae_db": 1.526858168603665, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10213341887501325, "3000–6000 Hz_40ms_envelope_correlation": 0.9932340400330073, "3000–6000 Hz_40ms_modulation_mae_db": 0.2896174074484356, "mr_stft_db": 5.743213247572879, "band_mae_db": 0.092516272414083, "rms_gain_error_db": -0.051409819261471555, "output_peak": 0.4289313477158413}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2328328818149372, "全帯域_10ms_rms_envelope_nrmse": 0.06414794062821681, "全帯域_10ms_envelope_correlation": 0.9961288699870392, "全帯域_10ms_modulation_mae_db": 0.0453773096473698, "全帯域_40ms_log_rms_mae_db": 0.6534466780984381, "全帯域_40ms_rms_envelope_nrmse": 0.03285374019777643, "全帯域_40ms_envelope_correlation": 0.9989429858227985, "全帯域_40ms_modulation_mae_db": 0.03336085666153788, "300–1000 Hz_10ms_log_rms_mae_db": 1.7292574882263743, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10087104013579336, "300–1000 Hz_10ms_envelope_correlation": 0.9925604271231769, "300–1000 Hz_10ms_modulation_mae_db": 0.5143094416011739, "300–1000 Hz_40ms_log_rms_mae_db": 1.0188270480788018, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054031994780206434, "300–1000 Hz_40ms_envelope_correlation": 0.9980094165146095, "300–1000 Hz_40ms_modulation_mae_db": 0.6803560213647376, "1000–3000 Hz_10ms_log_rms_mae_db": 1.743984312618773, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14549179512316252, "1000–3000 Hz_10ms_envelope_correlation": 0.9834581592367532, "1000–3000 Hz_10ms_modulation_mae_db": 0.5252655607905566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0489609601404088, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0761993440541753, "1000–3000 Hz_40ms_envelope_correlation": 0.9955312708351804, "1000–3000 Hz_40ms_modulation_mae_db": 0.7679012053205438, "3000–6000 Hz_10ms_log_rms_mae_db": 2.074260748882491, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1816243175635596, "3000–6000 Hz_10ms_envelope_correlation": 0.9800253239807719, "3000–6000 Hz_10ms_modulation_mae_db": 0.8278012516720322, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5202365901014372, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12140230107682322, "3000–6000 Hz_40ms_envelope_correlation": 0.9917946169866638, "3000–6000 Hz_40ms_modulation_mae_db": 0.902729008465644, "mr_stft_db": 5.730559147769687, "band_mae_db": 0.18205651522038382, "rms_gain_error_db": -0.05455019508072958, "output_peak": 0.45239323885396143}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3145720567081265, "全帯域_10ms_rms_envelope_nrmse": 0.06810178937199883, "全帯域_10ms_envelope_correlation": 0.9956159799685601, "全帯域_10ms_modulation_mae_db": 0.25486293298348345, "全帯域_40ms_log_rms_mae_db": 0.7112162795281928, "全帯域_40ms_rms_envelope_nrmse": 0.03702083308678836, "全帯域_40ms_envelope_correlation": 0.9986260698477748, "全帯域_40ms_modulation_mae_db": 0.14089276843372198, "300–1000 Hz_10ms_log_rms_mae_db": 1.706193997510373, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09362851531314753, "300–1000 Hz_10ms_envelope_correlation": 0.9935656852144519, "300–1000 Hz_10ms_modulation_mae_db": 0.3859184372884658, "300–1000 Hz_40ms_log_rms_mae_db": 1.0264744594251687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053790900826532685, "300–1000 Hz_40ms_envelope_correlation": 0.9979429334606117, "300–1000 Hz_40ms_modulation_mae_db": 0.4962027610578688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.749764128222726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14102760957085275, "1000–3000 Hz_10ms_envelope_correlation": 0.9846062973798054, "1000–3000 Hz_10ms_modulation_mae_db": 0.6169827705011549, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9845422783984578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07596076904548732, "1000–3000 Hz_40ms_envelope_correlation": 0.995729782725054, "1000–3000 Hz_40ms_modulation_mae_db": 0.8728274720318657, "3000–6000 Hz_10ms_log_rms_mae_db": 2.101091043828072, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1933765683277996, "3000–6000 Hz_10ms_envelope_correlation": 0.9792845759009617, "3000–6000 Hz_10ms_modulation_mae_db": 0.6058398320528863, "3000–6000 Hz_40ms_log_rms_mae_db": 1.52207619519223, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12592311204931425, "3000–6000 Hz_40ms_envelope_correlation": 0.9917443952932383, "3000–6000 Hz_40ms_modulation_mae_db": 0.5880748974195945, "mr_stft_db": 5.736223260652888, "band_mae_db": 0.15066032991807535, "rms_gain_error_db": -0.03690828507960258, "output_peak": 0.4460119494218061}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2529180450258854, "全帯域_10ms_rms_envelope_nrmse": 0.07193908211549019, "全帯域_10ms_envelope_correlation": 0.9951586772025993, "全帯域_10ms_modulation_mae_db": 0.24661766121127393, "全帯域_40ms_log_rms_mae_db": 0.7209084493223752, "全帯域_40ms_rms_envelope_nrmse": 0.036852254268988, "全帯域_40ms_envelope_correlation": 0.9987193123074556, "全帯域_40ms_modulation_mae_db": 0.442582740941085, "300–1000 Hz_10ms_log_rms_mae_db": 1.8073372672827548, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10679208886964282, "300–1000 Hz_10ms_envelope_correlation": 0.9917008239504835, "300–1000 Hz_10ms_modulation_mae_db": 1.0434157433758064, "300–1000 Hz_40ms_log_rms_mae_db": 1.131848025591888, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05827980794026606, "300–1000 Hz_40ms_envelope_correlation": 0.9977345058135331, "300–1000 Hz_40ms_modulation_mae_db": 1.1812505164816542, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7896283175593337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15486883910950994, "1000–3000 Hz_10ms_envelope_correlation": 0.981899317992805, "1000–3000 Hz_10ms_modulation_mae_db": 0.6028202846195898, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0758784483752504, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08810963877186151, "1000–3000 Hz_40ms_envelope_correlation": 0.9949050323945294, "1000–3000 Hz_40ms_modulation_mae_db": 0.8069681336611056, "3000–6000 Hz_10ms_log_rms_mae_db": 2.110172861485771, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16346843319054416, "3000–6000 Hz_10ms_envelope_correlation": 0.9833131713798868, "3000–6000 Hz_10ms_modulation_mae_db": 0.4576250846974507, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5837768840983624, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1101370454334914, "3000–6000 Hz_40ms_envelope_correlation": 0.9924091522472117, "3000–6000 Hz_40ms_modulation_mae_db": 0.45177167204769964, "mr_stft_db": 5.7318043892649095, "band_mae_db": 0.24115381943344905, "rms_gain_error_db": -0.08923190819463278, "output_peak": 0.4321700476423836}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3005546522166362, "全帯域_10ms_rms_envelope_nrmse": 0.0716622285259059, "全帯域_10ms_envelope_correlation": 0.9951686619736996, "全帯域_10ms_modulation_mae_db": 0.5170640420892217, "全帯域_40ms_log_rms_mae_db": 0.6797020392194218, "全帯域_40ms_rms_envelope_nrmse": 0.03701681760060017, "全帯域_40ms_envelope_correlation": 0.9986685910605625, "全帯域_40ms_modulation_mae_db": 0.5130737488103972, "300–1000 Hz_10ms_log_rms_mae_db": 1.7825999469805425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10207701771995342, "300–1000 Hz_10ms_envelope_correlation": 0.9923588104784252, "300–1000 Hz_10ms_modulation_mae_db": 1.053433663133489, "300–1000 Hz_40ms_log_rms_mae_db": 1.0500067808277884, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05660523879590571, "300–1000 Hz_40ms_envelope_correlation": 0.9977611759243925, "300–1000 Hz_40ms_modulation_mae_db": 1.3546753298841814, "1000–3000 Hz_10ms_log_rms_mae_db": 1.792282233514416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15326190215657362, "1000–3000 Hz_10ms_envelope_correlation": 0.982239468618248, "1000–3000 Hz_10ms_modulation_mae_db": 0.4515625749662366, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0968654255385155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09398799733527466, "1000–3000 Hz_40ms_envelope_correlation": 0.9938606731371267, "1000–3000 Hz_40ms_modulation_mae_db": 0.7085918129041113, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0568778051601178, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17469406363940249, "3000–6000 Hz_10ms_envelope_correlation": 0.9821646515153859, "3000–6000 Hz_10ms_modulation_mae_db": 0.65423806414425, "3000–6000 Hz_40ms_log_rms_mae_db": 1.530498434627443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12048323311022313, "3000–6000 Hz_40ms_envelope_correlation": 0.9914993693978134, "3000–6000 Hz_40ms_modulation_mae_db": 0.5116515113574392, "mr_stft_db": 5.732883342377389, "band_mae_db": 0.19950480807476925, "rms_gain_error_db": -0.06953825159000208, "output_peak": 0.41482774813210843}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2523000133986688, "全帯域_10ms_rms_envelope_nrmse": 0.06250465053692053, "全帯域_10ms_envelope_correlation": 0.9963201543944001, "全帯域_10ms_modulation_mae_db": 0.39611868226035424, "全帯域_40ms_log_rms_mae_db": 0.6905030971188842, "全帯域_40ms_rms_envelope_nrmse": 0.0316119165130842, "全帯域_40ms_envelope_correlation": 0.999018665160302, "全帯域_40ms_modulation_mae_db": 0.48194762580871736, "300–1000 Hz_10ms_log_rms_mae_db": 1.7783794568422613, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09042268896144041, "300–1000 Hz_10ms_envelope_correlation": 0.9941294258357157, "300–1000 Hz_10ms_modulation_mae_db": 0.8009454817270036, "300–1000 Hz_40ms_log_rms_mae_db": 1.0826252168528505, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051751860425883665, "300–1000 Hz_40ms_envelope_correlation": 0.9982527870707032, "300–1000 Hz_40ms_modulation_mae_db": 0.9607252552747604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8482306431836513, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15637312029482223, "1000–3000 Hz_10ms_envelope_correlation": 0.9807634843240165, "1000–3000 Hz_10ms_modulation_mae_db": 0.32057695854076873, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1624745355000168, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08927681700652919, "1000–3000 Hz_40ms_envelope_correlation": 0.993574776348088, "1000–3000 Hz_40ms_modulation_mae_db": 0.334324229711602, "3000–6000 Hz_10ms_log_rms_mae_db": 2.077487462251604, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16490105322420728, "3000–6000 Hz_10ms_envelope_correlation": 0.9841487979578133, "3000–6000 Hz_10ms_modulation_mae_db": 0.6824259808082562, "3000–6000 Hz_40ms_log_rms_mae_db": 1.539140204209179, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11200522018151704, "3000–6000 Hz_40ms_envelope_correlation": 0.9927261667802336, "3000–6000 Hz_40ms_modulation_mae_db": 0.6574843809964324, "mr_stft_db": 5.75028951023744, "band_mae_db": 0.13797712656961955, "rms_gain_error_db": -0.05068031648721989, "output_peak": 0.43444108360660966}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2914279168529414, "全帯域_10ms_rms_envelope_nrmse": 0.07703566002954877, "全帯域_10ms_envelope_correlation": 0.9944167451142001, "全帯域_10ms_modulation_mae_db": 0.15272736338965334, "全帯域_40ms_log_rms_mae_db": 0.6993534914004396, "全帯域_40ms_rms_envelope_nrmse": 0.04058257229949807, "全帯域_40ms_envelope_correlation": 0.998413073863298, "全帯域_40ms_modulation_mae_db": 0.16200810284333522, "300–1000 Hz_10ms_log_rms_mae_db": 1.7388092933981254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11486803945697949, "300–1000 Hz_10ms_envelope_correlation": 0.9903810879192111, "300–1000 Hz_10ms_modulation_mae_db": 0.638926887701925, "300–1000 Hz_40ms_log_rms_mae_db": 1.0671562136259305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06565873942179759, "300–1000 Hz_40ms_envelope_correlation": 0.9970672810728524, "300–1000 Hz_40ms_modulation_mae_db": 1.0409366956254889, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7812963474402652, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1520661417302432, "1000–3000 Hz_10ms_envelope_correlation": 0.9820333043787769, "1000–3000 Hz_10ms_modulation_mae_db": 0.3063945763228931, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0699704635061023, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08498105515762827, "1000–3000 Hz_40ms_envelope_correlation": 0.9945739673433833, "1000–3000 Hz_40ms_modulation_mae_db": 0.572987414059658, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0483223646972584, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16758161678326614, "3000–6000 Hz_10ms_envelope_correlation": 0.9825161005404796, "3000–6000 Hz_10ms_modulation_mae_db": 0.7616067201850288, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4796303653875926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11142381166682648, "3000–6000 Hz_40ms_envelope_correlation": 0.9919594654165252, "3000–6000 Hz_40ms_modulation_mae_db": 0.8897115883288955, "mr_stft_db": 5.733774804270585, "band_mae_db": 0.16697465916798315, "rms_gain_error_db": -0.07414337159374172, "output_peak": 0.44826203843804796}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2542205729731999, "全帯域_10ms_rms_envelope_nrmse": 0.06968270655731046, "全帯域_10ms_envelope_correlation": 0.9954773221085269, "全帯域_10ms_modulation_mae_db": 0.6934796404050472, "全帯域_40ms_log_rms_mae_db": 0.7058712968485238, "全帯域_40ms_rms_envelope_nrmse": 0.0387286659696452, "全帯域_40ms_envelope_correlation": 0.9985909395220897, "全帯域_40ms_modulation_mae_db": 1.0047696960965065, "300–1000 Hz_10ms_log_rms_mae_db": 1.6494475911052866, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09503323907154505, "300–1000 Hz_10ms_envelope_correlation": 0.9934555442730661, "300–1000 Hz_10ms_modulation_mae_db": 1.054239125229201, "300–1000 Hz_40ms_log_rms_mae_db": 1.0018075057971685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055498403951317596, "300–1000 Hz_40ms_envelope_correlation": 0.9979129712124117, "300–1000 Hz_40ms_modulation_mae_db": 1.3177815527521068, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7832011126090286, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14806950403560004, "1000–3000 Hz_10ms_envelope_correlation": 0.984131388182339, "1000–3000 Hz_10ms_modulation_mae_db": 0.73170981244765, "1000–3000 Hz_40ms_log_rms_mae_db": 1.037206075353933, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08553661774298937, "1000–3000 Hz_40ms_envelope_correlation": 0.9959944674649726, "1000–3000 Hz_40ms_modulation_mae_db": 0.8492032722207256, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0754469409336256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16615337675194566, "3000–6000 Hz_10ms_envelope_correlation": 0.9832769824543631, "3000–6000 Hz_10ms_modulation_mae_db": 0.43982584959702337, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5185894087023095, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11583153365546012, "3000–6000 Hz_40ms_envelope_correlation": 0.9916457051952776, "3000–6000 Hz_40ms_modulation_mae_db": 0.36456321714937395, "mr_stft_db": 5.742124387109495, "band_mae_db": 0.19122208516718778, "rms_gain_error_db": -0.09389888410571107, "output_peak": 0.4374934212501065}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.959459076534725, "spectral_convergence": 0.21684047756354904, "spectral_convergence_status": "ok", "amplitude_floor": 1.642409573690628e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5717608162626915, "spectral_convergence": 0.2065054461205444, "spectral_convergence_status": "ok", "amplitude_floor": 1.497573454010957e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.698419849921222, "spectral_convergence": 0.20185206160862876, "spectral_convergence_status": "ok", "amplitude_floor": 1.2731241516974033e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012546591027172457, "output_power": 0.0012627065234896585, "delta_db": 0.027766823412602767, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0029841932734943235, "output_power": 0.002922069674122261, "delta_db": -0.09136380033372127, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004877853564370709, "output_power": 0.0004930933223616283, "delta_db": 0.04700362773025027, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013442175178747268, "output_power": 0.00012605097876178213, "delta_db": -0.2792332859839188, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010124396480278944, "output_power": 0.00010131408425280945, "delta_db": 0.0030067915420961496, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001060495609474016, "output_power": 0.00010347526428446559, "delta_db": -0.10672330548190881, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.092516272414083, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.051409819261471555, "pitch": null, "output_rms_dbfs": -23.002741261021114, "output_peak": 0.4289313477158413, "reference_rms_dbfs": -22.951331441759642, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.743213247572879}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024147266762860616, "log_rms_mae_db": 1.2661602095078144, "rms_envelope_nrmse": 0.07269893286329901, "envelope_correlation": 0.9950087755901975, "rms_gain_error_db": -0.05140733786118677, "reference_mean_power": 0.005072567967533502, "output_mean_power": 0.005012878068187313}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4164286663529534e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010865583816620668, "output_power": 0.0010873067211544555, "reference_active": true, "delta_db": 0.0029900634087327146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020269542962956985, "output_power": 0.00018657519347915408, "reference_active": true, "delta_db": -0.3599005568691229}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.69603101954121e-05, "output_power": 1.9633089411075467e-05, "reference_active": true, "delta_db": -1.3773624031497493}], "active_band_count": 3, "active_band_mae_db": 0.580084341142535}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000225157536036158, "log_rms_mae_db": 0.713902130089294, "rms_envelope_nrmse": 0.03873117095704167, "envelope_correlation": 0.9985087796833786, "rms_gain_error_db": -0.05139605274872856, "reference_mean_power": 0.005098028148503288, "output_mean_power": 0.00503805174555428}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2603245746946083e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010232290392639372, "output_power": 0.0010198134772230205, "reference_active": true, "delta_db": -0.014521098968406703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010544090587602268, "output_power": 9.730642293306951e-05, "reference_active": true, "delta_db": -0.34867620611222977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0049541105469593e-06, "output_power": 1.3005805194486637e-06, "reference_active": true, "delta_db": -1.8796719221865674}], "active_band_count": 3, "active_band_mae_db": 0.7476230757557346}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020992712290116374, "log_rms_mae_db": 1.693859663589309, "rms_envelope_nrmse": 0.10405331681962583, "envelope_correlation": 0.9918722319297164, "rms_gain_error_db": -0.07072146915481821, "reference_mean_power": 0.0023388664575994473, "output_mean_power": 0.0023010882704802547}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4073975659228857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007410848450397675, "output_power": 0.0007461012260368775, "reference_active": true, "delta_db": 0.02929821480920547}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011531333502346333, "output_power": 0.00010349285389693606, "reference_active": true, "delta_db": -0.4696916960241021}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6543285086796136e-05, "output_power": 1.7358887560969963e-05, "reference_active": true, "delta_db": -1.8443278150730127}], "active_band_count": 3, "active_band_mae_db": 0.78110590863544}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002036893389343094, "log_rms_mae_db": 0.9924459722705626, "rms_envelope_nrmse": 0.060021196234967066, "envelope_correlation": 0.9971776880205796, "rms_gain_error_db": -0.07072232887173727, "reference_mean_power": 0.0023506239289763434, "output_mean_power": 0.00231265537327028}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3312664615669961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215176046332657, "output_power": 0.0007208135862349322, "reference_active": true, "delta_db": -0.004239682699403254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5162124576630795e-05, "output_power": 4.8994620110369547e-05, "reference_active": true, "delta_db": -0.514925901087646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1332065410684048e-06, "output_power": 1.0945821977157439e-06, "reference_active": true, "delta_db": -2.8978452619483748}], "active_band_count": 3, "active_band_mae_db": 1.1390036152451415}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.160365671559225e-05, "log_rms_mae_db": 1.7644245979057314, "rms_envelope_nrmse": 0.15136797043411143, "envelope_correlation": 0.9819963286252583, "rms_gain_error_db": -0.19166797458647908, "reference_mean_power": 0.00021265544148097967, "output_mean_power": 0.0002034743653537126}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.181263558889223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.569393841260949e-05, "output_power": 5.492595307114431e-05, "reference_active": true, "delta_db": -0.06030328728624361}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4402699100691371e-05, "output_power": 1.2670710800800438e-05, "reference_active": true, "delta_db": -0.5564290896649768}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.927958039351209e-06, "output_power": 2.3217223290974327e-06, "reference_active": true, "delta_db": -1.0075457033046817}], "active_band_count": 3, "active_band_mae_db": 0.5414260267519674}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.199537921297776e-05, "log_rms_mae_db": 1.049537008756208, "rms_envelope_nrmse": 0.08244374275843253, "envelope_correlation": 0.9945738944417671, "rms_gain_error_db": -0.19166520215302493, "reference_mean_power": 0.00021371687213348275, "output_mean_power": 0.00020449010088939795}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0829999638899264e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5369039282181144e-05, "output_power": 5.4248459311906034e-05, "reference_active": true, "delta_db": -0.0887957958085887}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.495407348960159e-06, "output_power": 5.750316794783564e-06, "reference_active": true, "delta_db": -0.5291462096447455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.109761653644897e-07, "output_power": 1.4124036867563154e-07, "reference_active": true, "delta_db": -1.7427455194777017}], "active_band_count": 3, "active_band_mae_db": 0.7868958416436787}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3802525725767476e-05, "log_rms_mae_db": 2.0638996764030355, "rms_envelope_nrmse": 0.15540499862553817, "envelope_correlation": 0.9848410314486483, "rms_gain_error_db": -0.06285260844113283, "reference_mean_power": 6.940550281225073e-05, "output_mean_power": 6.840827574179955e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.994304097313729e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.68316376202847e-05, "output_power": 3.469896915515909e-05, "reference_active": true, "delta_db": -0.25904456163496287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3496621932801275e-06, "output_power": 4.874740174592695e-06, "reference_active": true, "delta_db": -0.40374886576095753}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1665818120524947e-06, "output_power": 1.3187872256522718e-06, "reference_active": true, "delta_db": 0.5325953057893497}], "active_band_count": 3, "active_band_mae_db": 0.39846291106175674}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.808436973199796e-05, "log_rms_mae_db": 1.526858168603665, "rms_envelope_nrmse": 0.10213341887501325, "envelope_correlation": 0.9932340400330073, "rms_gain_error_db": -0.06285666726808203, "reference_mean_power": 6.97538862423984e-05, "output_mean_power": 6.875158930076856e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.7257017907786166e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.6073966740876416e-05, "output_power": 3.378400299535994e-05, "reference_active": true, "delta_db": -0.28482793614890284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3003381054168396e-06, "output_power": 2.173538065027244e-06, "reference_active": true, "delta_db": -0.24624423273467397}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.80849015690717e-08, "output_power": 8.440051031407228e-08, "reference_active": true, "delta_db": 0.33778005346173}], "active_band_count": 3, "active_band_mae_db": 0.2896174074484356}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8494633424499834, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "425a3795ab43c656505abd4996fd937c23ce67f7c17c17fec7b1b9214c718d90", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826409685922504, "standard_deviation": 0.001296846716837347, "minimum": 0.9807634843240165, "maximum": 0.9846062973798054, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7816014616317406, "standard_deviation": 0.03239117927154813, "minimum": 1.743984312618773, "maximum": 1.8482306431836513, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5120923206176021, "standard_deviation": 0.1469465205181045, "minimum": 0.3063945763228931, "maximum": 0.73170981244765, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15031586030685945, "standard_deviation": 0.005138912072597906, "minimum": 0.14102760957085275, "maximum": 0.15637312029482223, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9948429830862626, "standard_deviation": 0.0008707943295944178, "minimum": 0.993574776348088, "maximum": 0.9959944674649726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0656793994461116, "standard_deviation": 0.05126729116491869, "minimum": 0.9845422783984578, "maximum": 1.1624745355000168, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7124624226941614, "standard_deviation": 0.17905182151112606, "minimum": 0.334324229711602, "maximum": 0.8728274720318657, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08456199773404727, "standard_deviation": 0.006247197716209332, "minimum": 0.07596076904548732, "maximum": 0.09398799733527466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9824463293972888, "standard_deviation": 0.0019278618607261398, "minimum": 0.9792845759009617, "maximum": 0.9848410314486483, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0759448629552466, "standard_deviation": 0.02097955364187491, "minimum": 2.0483223646972584, "maximum": 2.110172861485771, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6034782117773356, "standard_deviation": 0.15776422070840504, "minimum": 0.39846291106175674, "maximum": 0.8278012516720322, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1709005535132829, "standard_deviation": 0.011949027110044098, "minimum": 0.15540499862553817, "maximum": 0.1933765683277996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9921266139187463, "standard_deviation": 0.0006064704316261537, "minimum": 0.9914993693978134, "maximum": 0.9932340400330073, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5276007813652774, "standard_deviation": 0.028697881150473142, "minimum": 1.4796303653875926, "maximum": 1.5837768840983624, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5819504604016893, "standard_deviation": 0.22613616542823245, "minimum": 0.2896174074484356, "maximum": 0.902729008465644, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11491745950608362, "standard_deviation": 0.007577803037571546, "minimum": 0.10213341887501325, "maximum": 0.12592311204931425, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925030045905309, "standard_deviation": 0.0012100682153259065, "minimum": 0.9903810879192111, "maximum": 0.9941294258357157, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7357355881168783, "standard_deviation": 0.0524607317526492, "minimum": 1.6494475911052866, "maximum": 1.8073372672827548, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.784036836086563, "standard_deviation": 0.2578659290567713, "minimum": 0.3859184372884658, "maximum": 1.054239125229201, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10096824329351597, "standard_deviation": 0.00791185924181341, "minimum": 0.09042268896144041, "maximum": 0.11486803945697949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977323448862117, "standard_deviation": 0.00040977209429632854, "minimum": 0.9970672810728524, "maximum": 0.9982527870707032, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.04639890280877, "standard_deviation": 0.04646701313572385, "minimum": 0.9924459722705626, "maximum": 1.131848025591888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0213664684607424, "standard_deviation": 0.30107035284694056, "minimum": 0.4962027610578688, "maximum": 1.3546753298841814, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.056954767797109604, "standard_deviation": 0.004388034125278579, "minimum": 0.051751860425883665, "maximum": 0.06565873942179759, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17025820199569386, "standard_deviation": 0.044626979061052355, "minimum": 0.092516272414083, "maximum": 0.24115381943344905, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.737609011156909, "standard_deviation": 0.0069172532135076984, "minimum": 5.730559147769687, "maximum": 5.75028951023744, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4368163593826081, "standard_deviation": 0.012139443465453925, "minimum": 0.41482774813210843, "maximum": 0.45239323885396143, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06504512892413891, "standard_deviation": 0.020052077820291638, "minimum": -0.09389888410571107, "maximum": -0.03690828507960258, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954118982924028, "standard_deviation": 0.000616560708650636, "minimum": 0.9944167451142001, "maximum": 0.9963201543944001, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2706232935622763, "standard_deviation": 0.028351151064905898, "minimum": 1.2328328818149372, "maximum": 1.3145720567081265, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3607914966411173, "standard_deviation": 0.22407689667138764, "minimum": 0.0453773096473698, "maximum": 0.6934796404050472, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06972162382858631, "standard_deviation": 0.004733144358277197, "minimum": 0.06250465053692053, "maximum": 0.07703566002954877, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986860521584575, "standard_deviation": 0.00020585862539937585, "minimum": 0.998413073863298, "maximum": 0.999018665160302, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6968629327031962, "standard_deviation": 0.021979887322763306, "minimum": 0.6534466780984381, "maximum": 0.7209084493223752, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4407823269188794, "standard_deviation": 0.32793682764152254, "minimum": 0.03336085666153788, "maximum": 1.0047696960965065, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.036674746361677765, "standard_deviation": 0.0030288895519184943, "minimum": 0.0316119165130842, "maximum": 0.04058257229949807, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0010", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.0176100730895996, "predicted_mos": 3.0176100730895996, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5457e4dd714a721829e5df3f814ceab04e82cee8316af1cbc5688afb29e9142a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7941952999681234, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2651242610698832, "全帯域_10ms_rms_envelope_nrmse": 0.0724718547879536, "全帯域_10ms_envelope_correlation": 0.9950379893888484, "全帯域_10ms_modulation_mae_db": 0.6147054156227506, "全帯域_40ms_log_rms_mae_db": 0.6867089687993062, "全帯域_40ms_rms_envelope_nrmse": 0.038071613859476286, "全帯域_40ms_envelope_correlation": 0.9985575807750797, "全帯域_40ms_modulation_mae_db": 0.7813441820513606, "300–1000 Hz_10ms_log_rms_mae_db": 1.6571127012995261, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10636982252857839, "300–1000 Hz_10ms_envelope_correlation": 0.9915011330152527, "300–1000 Hz_10ms_modulation_mae_db": 0.837256787074077, "300–1000 Hz_40ms_log_rms_mae_db": 0.9091855470254739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06123212645678917, "300–1000 Hz_40ms_envelope_correlation": 0.9970481065122296, "300–1000 Hz_40ms_modulation_mae_db": 1.232190218930788, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7280120203388987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14976717368854375, "1000–3000 Hz_10ms_envelope_correlation": 0.9825347606565122, "1000–3000 Hz_10ms_modulation_mae_db": 0.5683688413168729, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9652614644302766, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08512215545266678, "1000–3000 Hz_40ms_envelope_correlation": 0.9944310694343426, "1000–3000 Hz_40ms_modulation_mae_db": 0.7241096015496095, "3000–6000 Hz_10ms_log_rms_mae_db": 2.019302547481736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1572258256117282, "3000–6000 Hz_10ms_envelope_correlation": 0.9844467390838522, "3000–6000 Hz_10ms_modulation_mae_db": 0.4947632597114202, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4421639372576884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10218805308527949, "3000–6000 Hz_40ms_envelope_correlation": 0.9932018308987239, "3000–6000 Hz_40ms_modulation_mae_db": 0.36543106195441877, "mr_stft_db": 5.685001311719566, "band_mae_db": 0.09170358227646731, "rms_gain_error_db": -0.04853210826903252, "output_peak": 0.43395842994948863}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.276586977591687, "全帯域_10ms_rms_envelope_nrmse": 0.06398357298436658, "全帯域_10ms_envelope_correlation": 0.9961592092488187, "全帯域_10ms_modulation_mae_db": 0.08704765733570885, "全帯域_40ms_log_rms_mae_db": 0.6630302499875182, "全帯域_40ms_rms_envelope_nrmse": 0.0322154785023167, "全帯域_40ms_envelope_correlation": 0.9989979165664673, "全帯域_40ms_modulation_mae_db": 0.06936592533450657, "300–1000 Hz_10ms_log_rms_mae_db": 1.711471591963572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10086165463469546, "300–1000 Hz_10ms_envelope_correlation": 0.992547775485717, "300–1000 Hz_10ms_modulation_mae_db": 0.583986540024883, "300–1000 Hz_40ms_log_rms_mae_db": 0.9635607912361907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05388531261243788, "300–1000 Hz_40ms_envelope_correlation": 0.9980039453606077, "300–1000 Hz_40ms_modulation_mae_db": 0.7466443849468054, "1000–3000 Hz_10ms_log_rms_mae_db": 1.729207687191454, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1441878011033689, "1000–3000 Hz_10ms_envelope_correlation": 0.9838981027017166, "1000–3000 Hz_10ms_modulation_mae_db": 0.3655297465370821, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9973441786670602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0773150346302951, "1000–3000 Hz_40ms_envelope_correlation": 0.9955768060028343, "1000–3000 Hz_40ms_modulation_mae_db": 0.5723360615767749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.991005593153989, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1857729080100544, "3000–6000 Hz_10ms_envelope_correlation": 0.9792174060316332, "3000–6000 Hz_10ms_modulation_mae_db": 0.9200668894429195, "3000–6000 Hz_40ms_log_rms_mae_db": 1.460635723261673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12335089129829324, "3000–6000 Hz_40ms_envelope_correlation": 0.991729267095946, "3000–6000 Hz_40ms_modulation_mae_db": 0.9564130568211847, "mr_stft_db": 5.667482989853867, "band_mae_db": 0.18621321913248212, "rms_gain_error_db": -0.06259574399121674, "output_peak": 0.4520485151705755}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.372820418615889, "全帯域_10ms_rms_envelope_nrmse": 0.06716538174873757, "全帯域_10ms_envelope_correlation": 0.9957483175966534, "全帯域_10ms_modulation_mae_db": 0.2743930125553798, "全帯域_40ms_log_rms_mae_db": 0.6978210458884095, "全帯域_40ms_rms_envelope_nrmse": 0.03532853222094472, "全帯域_40ms_envelope_correlation": 0.9987700142952423, "全帯域_40ms_modulation_mae_db": 0.14127861766794622, "300–1000 Hz_10ms_log_rms_mae_db": 1.6502499434846425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09254281827964933, "300–1000 Hz_10ms_envelope_correlation": 0.993745681712868, "300–1000 Hz_10ms_modulation_mae_db": 0.3693518120145183, "300–1000 Hz_40ms_log_rms_mae_db": 0.9452165906707073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05201396030160845, "300–1000 Hz_40ms_envelope_correlation": 0.9981261952520906, "300–1000 Hz_40ms_modulation_mae_db": 0.47712061347098417, "1000–3000 Hz_10ms_log_rms_mae_db": 1.749993881059595, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13868519860534642, "1000–3000 Hz_10ms_envelope_correlation": 0.9853519123407634, "1000–3000 Hz_10ms_modulation_mae_db": 0.6134474784266486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9623777300533358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07541479192647044, "1000–3000 Hz_40ms_envelope_correlation": 0.9960982261128856, "1000–3000 Hz_40ms_modulation_mae_db": 0.8469998047965707, "3000–6000 Hz_10ms_log_rms_mae_db": 1.975789515831442, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1882657405423797, "3000–6000 Hz_10ms_envelope_correlation": 0.9800115354777014, "3000–6000 Hz_10ms_modulation_mae_db": 0.6630429186517018, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3658452209930791, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11773750937622154, "3000–6000 Hz_40ms_envelope_correlation": 0.9925796635065418, "3000–6000 Hz_40ms_modulation_mae_db": 0.5845946647136476, "mr_stft_db": 5.678395404579661, "band_mae_db": 0.169876962870093, "rms_gain_error_db": -0.05502566455718307, "output_peak": 0.4460351511343287}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.252558823544265, "全帯域_10ms_rms_envelope_nrmse": 0.07654035799017318, "全帯域_10ms_envelope_correlation": 0.9945015592470217, "全帯域_10ms_modulation_mae_db": 0.24297460369324086, "全帯域_40ms_log_rms_mae_db": 0.722686974088264, "全帯域_40ms_rms_envelope_nrmse": 0.03980448119870332, "全帯域_40ms_envelope_correlation": 0.9984846319765894, "全帯域_40ms_modulation_mae_db": 0.4557028623314352, "300–1000 Hz_10ms_log_rms_mae_db": 1.763318964390312, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1134793171818633, "300–1000 Hz_10ms_envelope_correlation": 0.9905441144066568, "300–1000 Hz_10ms_modulation_mae_db": 1.0438988857744687, "300–1000 Hz_40ms_log_rms_mae_db": 1.0539269210167184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061289870257719065, "300–1000 Hz_40ms_envelope_correlation": 0.9974116325542276, "300–1000 Hz_40ms_modulation_mae_db": 1.248035442587061, "1000–3000 Hz_10ms_log_rms_mae_db": 1.756723787783305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15172659145606543, "1000–3000 Hz_10ms_envelope_correlation": 0.9829855279186649, "1000–3000 Hz_10ms_modulation_mae_db": 0.49878531292205036, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9771310109464114, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08816254698950225, "1000–3000 Hz_40ms_envelope_correlation": 0.9952771244046011, "1000–3000 Hz_40ms_modulation_mae_db": 0.8569704514434234, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0420167317232716, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16564938311871538, "3000–6000 Hz_10ms_envelope_correlation": 0.9828376473917579, "3000–6000 Hz_10ms_modulation_mae_db": 0.5169886186744982, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4920849733252353, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11334284247713833, "3000–6000 Hz_40ms_envelope_correlation": 0.9919197667785178, "3000–6000 Hz_40ms_modulation_mae_db": 0.5101151868508023, "mr_stft_db": 5.675767066393429, "band_mae_db": 0.2523371033987159, "rms_gain_error_db": -0.08686752560480507, "output_peak": 0.43510987649297}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3420682635904353, "全帯域_10ms_rms_envelope_nrmse": 0.0723009067566752, "全帯域_10ms_envelope_correlation": 0.9950747092070801, "全帯域_10ms_modulation_mae_db": 0.6018869092146032, "全帯域_40ms_log_rms_mae_db": 0.7277409294817561, "全帯域_40ms_rms_envelope_nrmse": 0.037906872454373665, "全帯域_40ms_envelope_correlation": 0.99859088749789, "全帯域_40ms_modulation_mae_db": 0.6180376051178367, "300–1000 Hz_10ms_log_rms_mae_db": 1.747661373146162, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1041628647127165, "300–1000 Hz_10ms_envelope_correlation": 0.9919738812076757, "300–1000 Hz_10ms_modulation_mae_db": 1.0678780732374733, "300–1000 Hz_40ms_log_rms_mae_db": 0.9760794925812533, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05741921251938834, "300–1000 Hz_40ms_envelope_correlation": 0.9976131834539526, "300–1000 Hz_40ms_modulation_mae_db": 1.4094836014162315, "1000–3000 Hz_10ms_log_rms_mae_db": 1.770642899303404, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15073417735231207, "1000–3000 Hz_10ms_envelope_correlation": 0.9831901539310826, "1000–3000 Hz_10ms_modulation_mae_db": 0.34455879737789, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9955087837789144, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09226177950417033, "1000–3000 Hz_40ms_envelope_correlation": 0.9945731516444792, "1000–3000 Hz_40ms_modulation_mae_db": 0.5312743335602161, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9802584138730805, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17520621712672102, "3000–6000 Hz_10ms_envelope_correlation": 0.9817978816624691, "3000–6000 Hz_10ms_modulation_mae_db": 0.672307664439141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4541057356447227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12284090760432201, "3000–6000 Hz_40ms_envelope_correlation": 0.9909037384290641, "3000–6000 Hz_40ms_modulation_mae_db": 0.49803419665636567, "mr_stft_db": 5.669773697935732, "band_mae_db": 0.2192278710784675, "rms_gain_error_db": -0.06422811370895976, "output_peak": 0.41649359254950186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2327164163342035, "全帯域_10ms_rms_envelope_nrmse": 0.06511561675774764, "全帯域_10ms_envelope_correlation": 0.9960041793960618, "全帯域_10ms_modulation_mae_db": 0.3938742607304995, "全帯域_40ms_log_rms_mae_db": 0.7066321712745092, "全帯域_40ms_rms_envelope_nrmse": 0.03266041999179305, "全帯域_40ms_envelope_correlation": 0.9989517766953085, "全帯域_40ms_modulation_mae_db": 0.49510368991374243, "300–1000 Hz_10ms_log_rms_mae_db": 1.742764228876263, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09442445093394747, "300–1000 Hz_10ms_envelope_correlation": 0.9935050674016122, "300–1000 Hz_10ms_modulation_mae_db": 0.8527761302430571, "300–1000 Hz_40ms_log_rms_mae_db": 1.0331480344767305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05200980615912332, "300–1000 Hz_40ms_envelope_correlation": 0.9981617490291232, "300–1000 Hz_40ms_modulation_mae_db": 1.0471322730407358, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7932521398133947, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15345606897559094, "1000–3000 Hz_10ms_envelope_correlation": 0.9817545758142812, "1000–3000 Hz_10ms_modulation_mae_db": 0.2336145107737159, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0169018231355529, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08911350562199748, "1000–3000 Hz_40ms_envelope_correlation": 0.9939924606613552, "1000–3000 Hz_40ms_modulation_mae_db": 0.38436918943408527, "3000–6000 Hz_10ms_log_rms_mae_db": 2.005627930486806, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1611319291582668, "3000–6000 Hz_10ms_envelope_correlation": 0.9846124843747995, "3000–6000 Hz_10ms_modulation_mae_db": 0.6612383286876676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.438545492586793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10862954451274925, "3000–6000 Hz_40ms_envelope_correlation": 0.9929450475808778, "3000–6000 Hz_40ms_modulation_mae_db": 0.6104442172731396, "mr_stft_db": 5.687537668824064, "band_mae_db": 0.1544263134489466, "rms_gain_error_db": -0.05261859327508665, "output_peak": 0.43476485542664217}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2948462928784223, "全帯域_10ms_rms_envelope_nrmse": 0.07672396771180649, "全帯域_10ms_envelope_correlation": 0.9944510524272293, "全帯域_10ms_modulation_mae_db": 0.20743513067674968, "全帯域_40ms_log_rms_mae_db": 0.6922853620424521, "全帯域_40ms_rms_envelope_nrmse": 0.04126542983162298, "全帯域_40ms_envelope_correlation": 0.9983392634618203, "全帯域_40ms_modulation_mae_db": 0.20936022704658744, "300–1000 Hz_10ms_log_rms_mae_db": 1.6610879713215732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11426281737403067, "300–1000 Hz_10ms_envelope_correlation": 0.9904112040676745, "300–1000 Hz_10ms_modulation_mae_db": 0.5475198473589143, "300–1000 Hz_40ms_log_rms_mae_db": 0.993381431273434, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06529274913460219, "300–1000 Hz_40ms_envelope_correlation": 0.9970046846770886, "300–1000 Hz_40ms_modulation_mae_db": 0.9448717471301705, "1000–3000 Hz_10ms_log_rms_mae_db": 1.75531479361862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1456993493152399, "1000–3000 Hz_10ms_envelope_correlation": 0.9837207357157305, "1000–3000 Hz_10ms_modulation_mae_db": 0.3758947536063222, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9938053057971358, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08354018350918245, "1000–3000 Hz_40ms_envelope_correlation": 0.9949818939652576, "1000–3000 Hz_40ms_modulation_mae_db": 0.6057539927106995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9745132972129946, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16599741555934633, "3000–6000 Hz_10ms_envelope_correlation": 0.982791991110925, "3000–6000 Hz_10ms_modulation_mae_db": 0.7116856646700667, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4371105757144111, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1130038838205727, "3000–6000 Hz_40ms_envelope_correlation": 0.9916835623425597, "3000–6000 Hz_40ms_modulation_mae_db": 0.8201038893817065, "mr_stft_db": 5.676172637252716, "band_mae_db": 0.1695685327677979, "rms_gain_error_db": -0.06687605796100186, "output_peak": 0.4494607933161201}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2677024145718017, "全帯域_10ms_rms_envelope_nrmse": 0.06868099058619938, "全帯域_10ms_envelope_correlation": 0.9956024755193367, "全帯域_10ms_modulation_mae_db": 0.5936944671990857, "全帯域_40ms_log_rms_mae_db": 0.7168765479715931, "全帯域_40ms_rms_envelope_nrmse": 0.0381895491883923, "全帯域_40ms_envelope_correlation": 0.9986236829721986, "全帯域_40ms_modulation_mae_db": 0.9085833339022679, "300–1000 Hz_10ms_log_rms_mae_db": 1.5796951127733616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09302461945330001, "300–1000 Hz_10ms_envelope_correlation": 0.9937199900456501, "300–1000 Hz_10ms_modulation_mae_db": 1.0150351657649328, "300–1000 Hz_40ms_log_rms_mae_db": 0.9031749860944851, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05365282220785193, "300–1000 Hz_40ms_envelope_correlation": 0.9980464517469263, "300–1000 Hz_40ms_modulation_mae_db": 1.2876546965239313, "1000–3000 Hz_10ms_log_rms_mae_db": 1.764527729602863, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14808775806711846, "1000–3000 Hz_10ms_envelope_correlation": 0.9845283379785589, "1000–3000 Hz_10ms_modulation_mae_db": 0.6933399138781886, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9965890354634562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08994885228645225, "1000–3000 Hz_40ms_envelope_correlation": 0.9957993407583516, "1000–3000 Hz_40ms_modulation_mae_db": 0.8616344367140737, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9984820770256044, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16015279068463414, "3000–6000 Hz_10ms_envelope_correlation": 0.9842513161619533, "3000–6000 Hz_10ms_modulation_mae_db": 0.4159598451301743, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4119663773630668, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11087144008263133, "3000–6000 Hz_40ms_envelope_correlation": 0.9921782539491136, "3000–6000 Hz_40ms_modulation_mae_db": 0.3578140027696001, "mr_stft_db": 5.687874948799489, "band_mae_db": 0.2118765038671605, "rms_gain_error_db": -0.0910739555674079, "output_peak": 0.43800848545283677}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.91213523305926, "spectral_convergence": 0.21445866629152957, "spectral_convergence_status": "ok", "amplitude_floor": 1.642409573690628e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.510631021568783, "spectral_convergence": 0.20475118572197012, "spectral_convergence_status": "ok", "amplitude_floor": 1.497573454010957e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.632237680530653, "spectral_convergence": 0.1999652519542752, "spectral_convergence_status": "ok", "amplitude_floor": 1.2731241516974033e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012546591027172457, "output_power": 0.0012570170023833527, "delta_db": 0.008154101658377988, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0029841932734943235, "output_power": 0.0029355079367379784, "delta_db": -0.07143688198250371, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004877853564370709, "output_power": 0.0004904929928747302, "delta_db": 0.024040490467724523, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013442175178747268, "output_power": 0.00012395397642235575, "delta_db": -0.35209087417688933, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010124396480278944, "output_power": 9.999713534836252e-05, "delta_db": -0.05381585375655873, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001060495609474016, "output_power": 0.00010506076182879993, "delta_db": -0.040683291616749676, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.09170358227646731, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.04853210826903252, "pitch": null, "output_rms_dbfs": -22.999863550028678, "output_peak": 0.43395842994948863, "reference_rms_dbfs": -22.951331441759642, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.685001311719566}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024147266762860616, "log_rms_mae_db": 1.2651242610698832, "rms_envelope_nrmse": 0.0724718547879536, "envelope_correlation": 0.9950379893888484, "rms_gain_error_db": -0.04852986958944501, "reference_mean_power": 0.005072567967533502, "output_mean_power": 0.005016200509013469}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4164286663529534e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010865583816620668, "output_power": 0.0010911613905891834, "reference_active": true, "delta_db": 0.0183592432756257}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020269542962956985, "output_power": 0.00018676137814688202, "reference_active": true, "delta_db": -0.35556886334333515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.69603101954121e-05, "output_power": 1.921790626104966e-05, "reference_active": true, "delta_db": -1.4701881402492911}], "active_band_count": 3, "active_band_mae_db": 0.6147054156227506}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000225157536036158, "log_rms_mae_db": 0.6867089687993062, "rms_envelope_nrmse": 0.038071613859476286, "envelope_correlation": 0.9985575807750797, "rms_gain_error_db": -0.04851773481873675, "reference_mean_power": 0.005098028148503288, "output_mean_power": 0.005041391857321002}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2603245746946083e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010232290392639372, "output_power": 0.0010224667842374466, "reference_active": true, "delta_db": -0.003236484753262097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010544090587602268, "output_power": 9.771318050600383e-05, "reference_active": true, "delta_db": -0.33055978884499315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0049541105469593e-06, "output_power": 1.2620623461213453e-06, "reference_active": true, "delta_db": -2.0102362725558267}], "active_band_count": 3, "active_band_mae_db": 0.7813441820513606}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020992712290116374, "log_rms_mae_db": 1.6571127012995261, "rms_envelope_nrmse": 0.10636982252857839, "envelope_correlation": 0.9915011330152527, "rms_gain_error_db": -0.05674398541372223, "reference_mean_power": 0.0023388664575994473, "output_mean_power": 0.002308506103019864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4073975659228857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007410848450397675, "output_power": 0.0007479281638360231, "reference_active": true, "delta_db": 0.039919550718662405}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011531333502346333, "output_power": 0.00010181102573897149, "reference_active": true, "delta_db": -0.5408471977145899}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6543285086796136e-05, "output_power": 1.7015875931986167e-05, "reference_active": true, "delta_db": -1.9310036127889787}], "active_band_count": 3, "active_band_mae_db": 0.837256787074077}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002036893389343094, "log_rms_mae_db": 0.9091855470254739, "rms_envelope_nrmse": 0.06123212645678917, "envelope_correlation": 0.9970481065122296, "rms_gain_error_db": -0.05674283275881396, "reference_mean_power": 0.0023506239289763434, "output_mean_power": 0.0023201115688055453}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3312664615669961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215176046332657, "output_power": 0.0007220688896761049, "reference_active": true, "delta_db": 0.0033170172295653765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5162124576630795e-05, "output_power": 4.832888661652924e-05, "reference_active": true, "delta_db": -0.5743419460180863}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1332065410684048e-06, "output_power": 1.0402596119378338e-06, "reference_active": true, "delta_db": -3.1189116935447125}], "active_band_count": 3, "active_band_mae_db": 1.232190218930788}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.160365671559225e-05, "log_rms_mae_db": 1.7280120203388987, "rms_envelope_nrmse": 0.14976717368854375, "envelope_correlation": 0.9825347606565122, "rms_gain_error_db": -0.2557380623979217, "reference_mean_power": 0.00021265544148097967, "output_mean_power": 0.00020049460632447057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.181263558889223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.569393841260949e-05, "output_power": 5.429274031841607e-05, "reference_active": true, "delta_db": -0.11066167871085775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4402699100691371e-05, "output_power": 1.312857305537882e-05, "reference_active": true, "delta_db": -0.4022636237320225}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.927958039351209e-06, "output_power": 2.2250856167021543e-06, "reference_active": true, "delta_db": -1.1921812215077385}], "active_band_count": 3, "active_band_mae_db": 0.5683688413168729}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.199537921297776e-05, "log_rms_mae_db": 0.9652614644302766, "rms_envelope_nrmse": 0.08512215545266678, "envelope_correlation": 0.9944310694343426, "rms_gain_error_db": -0.25572275383412885, "reference_mean_power": 0.00021371687213348275, "output_mean_power": 0.00020149604865522073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0829999638899264e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5369039282181144e-05, "output_power": 5.384147332640698e-05, "reference_active": true, "delta_db": -0.12150052578386311}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.495407348960159e-06, "output_power": 5.981772709959702e-06, "reference_active": true, "delta_db": -0.357764852625451}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.109761653644897e-07, "output_power": 1.4286539682924177e-07, "reference_active": true, "delta_db": -1.6930634262395141}], "active_band_count": 3, "active_band_mae_db": 0.7241096015496095}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3802525725767476e-05, "log_rms_mae_db": 2.019302547481736, "rms_envelope_nrmse": 0.1572258256117282, "envelope_correlation": 0.9844467390838522, "rms_gain_error_db": -0.08909479937947784, "reference_mean_power": 6.940550281225073e-05, "output_mean_power": 6.79961659107286e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.994304097313729e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.68316376202847e-05, "output_power": 3.521026241758106e-05, "reference_active": true, "delta_db": -0.1955176720364885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3496621932801275e-06, "output_power": 4.68042256680579e-06, "reference_active": true, "delta_db": -0.5804129453112746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1665818120524947e-06, "output_power": 1.373254773503062e-06, "reference_active": true, "delta_db": 0.7083591617864976}], "active_band_count": 3, "active_band_mae_db": 0.4947632597114202}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.808436973199796e-05, "log_rms_mae_db": 1.4421639372576884, "rms_envelope_nrmse": 0.10218805308527949, "envelope_correlation": 0.9932018308987239, "rms_gain_error_db": -0.08909312276151587, "reference_mean_power": 6.97538862423984e-05, "output_mean_power": 6.833750150542298e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.7257017907786166e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.6073966740876416e-05, "output_power": 3.422388156157175e-05, "reference_active": true, "delta_db": -0.22864636268041577}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3003381054168396e-06, "output_power": 2.0614495257738733e-06, "reference_active": true, "delta_db": -0.47618967993064093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.80849015690717e-08, "output_power": 8.545014057185813e-08, "reference_active": true, "delta_db": 0.39145714325219955}], "active_band_count": 3, "active_band_mae_db": 0.36543106195441877}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8491427640715692, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "518750e874dc73bccb3f801c8a0bd00542537cb1e2e48342eb7f060d75195235", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9834955133821638, "standard_deviation": 0.0011359694401460597, "minimum": 0.9817545758142812, "maximum": 0.9853519123407634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7559593673389418, "standard_deviation": 0.021424509519868552, "minimum": 1.7280120203388987, "maximum": 1.7932521398133947, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46169241935484634, "standard_deviation": 0.1566589645878548, "minimum": 0.2336145107737159, "maximum": 0.6933399138781886, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14779301482044824, "standard_deviation": 0.004788544944085951, "minimum": 0.13868519860534642, "maximum": 0.15345606897559094, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950912591230134, "standard_deviation": 0.0007279392025057299, "minimum": 0.9939924606613552, "maximum": 0.9960982261128856, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.988114916534018, "standard_deviation": 0.018431552870762473, "minimum": 0.9623777300533358, "maximum": 1.0169018231355529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6729309839731816, "standard_deviation": 0.17742818522425474, "minimum": 0.38436918943408527, "maximum": 0.8616344367140737, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08510985624009212, "standard_deviation": 0.006060782104449747, "minimum": 0.07541479192647044, "maximum": 0.09226177950417033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9824958751618864, "standard_deviation": 0.002034680355696025, "minimum": 0.9792174060316332, "maximum": 0.9846124843747995, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9983745133486155, "standard_deviation": 0.023473946837252905, "minimum": 1.9745132972129946, "maximum": 2.0420167317232716, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6320066486759487, "standard_deviation": 0.1565295395693762, "minimum": 0.4159598451301743, "maximum": 0.9200668894429195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16992527622648074, "standard_deviation": 0.011851583527685783, "minimum": 0.1572258256117282, "maximum": 0.1882657405423797, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992142641322668, "standard_deviation": 0.0007487303742341329, "minimum": 0.9909037384290641, "maximum": 0.9932018308987239, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4378072545183338, "standard_deviation": 0.03703115767954322, "minimum": 1.3658452209930791, "maximum": 1.4920849733252353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5878687845526083, "standard_deviation": 0.2093520659948399, "minimum": 0.3578140027696001, "maximum": 0.9564130568211847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11399563403215099, "standard_deviation": 0.00716285569399175, "minimum": 0.10218805308527949, "maximum": 0.12335089129829324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9922436059178884, "standard_deviation": 0.0013622319581871871, "minimum": 0.9904112040676745, "maximum": 0.993745681712868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6891702359069267, "standard_deviation": 0.06279763927807687, "minimum": 1.5796951127733616, "maximum": 1.763318964390312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7897129051865406, "standard_deviation": 0.2609537859608679, "minimum": 0.3693518120145183, "maximum": 1.0678780732374733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10239104563734763, "standard_deviation": 0.008729113258485096, "minimum": 0.09254281827964933, "maximum": 0.11426281737403067, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976769935732808, "standard_deviation": 0.0004784130317656833, "minimum": 0.9970046846770886, "maximum": 0.9981617490291232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9722092242968742, "standard_deviation": 0.05397839445246293, "minimum": 0.9031749860944851, "maximum": 1.0539269210167184, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0491416222558385, "standard_deviation": 0.3138218845292645, "minimum": 0.47712061347098417, "maximum": 1.4094836014162315, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.057099482456190044, "standard_deviation": 0.005012094480716558, "minimum": 0.05200980615912332, "maximum": 0.06529274913460219, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18190376110501635, "standard_deviation": 0.04848971283162915, "minimum": 0.09170358227646731, "maximum": 0.2523371033987159, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6785007156698155, "standard_deviation": 0.0077667378395509115, "minimum": 5.667482989853867, "maximum": 5.687874948799489, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.43823496243655796, "standard_deviation": 0.01127900358511226, "minimum": 0.41649359254950186, "maximum": 0.4520485151705755, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0659772203668367, "standard_deviation": 0.015511627273926845, "minimum": -0.0910739555674079, "maximum": -0.04853210826903252, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9953224365038813, "standard_deviation": 0.0006546740964316035, "minimum": 0.9944510524272293, "maximum": 0.9961592092488187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2880529835245733, "standard_deviation": 0.04712086269494612, "minimum": 1.2327164163342035, "maximum": 1.372820418615889, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3770014321285023, "standard_deviation": 0.20550802069114205, "minimum": 0.08704765733570885, "maximum": 0.6147054156227506, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07037283116545745, "standard_deviation": 0.004900578361740499, "minimum": 0.06398357298436658, "maximum": 0.07672396771180649, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986644692800746, "standard_deviation": 0.00022724298760474178, "minimum": 0.9983392634618203, "maximum": 0.9989979165664673, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.701722781191726, "standard_deviation": 0.02137085844578472, "minimum": 0.6630302499875182, "maximum": 0.7277409294817561, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.45984705542071036, "standard_deviation": 0.30418840927421015, "minimum": 0.06936592533450657, "maximum": 0.9085833339022679, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.036930297155952876, "standard_deviation": 0.0032496888835791393, "minimum": 0.0322154785023167, "maximum": 0.04126542983162298, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0010", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.529853582382202, "predicted_mos": 2.529853582382202, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "67a043748f1bff2b03a6e5eed5fce86fd09bee24b20d49247aa408d35ae6e556", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.192888000048697, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2285891404902243, "全帯域_10ms_rms_envelope_nrmse": 0.06769034867702226, "全帯域_10ms_envelope_correlation": 0.9956776312227112, "全帯域_10ms_modulation_mae_db": 0.21318080766043476, "全帯域_40ms_log_rms_mae_db": 0.6439774887238734, "全帯域_40ms_rms_envelope_nrmse": 0.037857801431896924, "全帯域_40ms_envelope_correlation": 0.9985758878125524, "全帯域_40ms_modulation_mae_db": 0.15468256145455414, "300–1000 Hz_10ms_log_rms_mae_db": 1.9197331545735508, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09283347311566933, "300–1000 Hz_10ms_envelope_correlation": 0.9937010592530409, "300–1000 Hz_10ms_modulation_mae_db": 0.9017053370267263, "300–1000 Hz_40ms_log_rms_mae_db": 1.2079985021708421, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055078772955287264, "300–1000 Hz_40ms_envelope_correlation": 0.9978453975617317, "300–1000 Hz_40ms_modulation_mae_db": 0.8663432835632728, "1000–3000 Hz_10ms_log_rms_mae_db": 2.213794422287487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16924786089686128, "1000–3000 Hz_10ms_envelope_correlation": 0.977978727472348, "1000–3000 Hz_10ms_modulation_mae_db": 1.2032513954046704, "1000–3000 Hz_40ms_log_rms_mae_db": 1.327471496169193, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09715787346182597, "1000–3000 Hz_40ms_envelope_correlation": 0.9932138561732522, "1000–3000 Hz_40ms_modulation_mae_db": 1.7831110491242448, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7973842510533293, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18049453723457023, "3000–6000 Hz_10ms_envelope_correlation": 0.9807659092132145, "3000–6000 Hz_10ms_modulation_mae_db": 0.5347793716764357, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0424251464747014, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11507217231826528, "3000–6000 Hz_40ms_envelope_correlation": 0.992199915022075, "3000–6000 Hz_40ms_modulation_mae_db": 0.5865361701210997, "mr_stft_db": 6.13303460986221, "band_mae_db": 0.19863710256410327, "rms_gain_error_db": -0.05089807643004243, "output_peak": 0.43017433626863066}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2083594308053947, "全帯域_10ms_rms_envelope_nrmse": 0.06901354327217495, "全帯域_10ms_envelope_correlation": 0.9955343997948181, "全帯域_10ms_modulation_mae_db": 0.1514730381845811, "全帯域_40ms_log_rms_mae_db": 0.630617840657911, "全帯域_40ms_rms_envelope_nrmse": 0.03858878417304284, "全帯域_40ms_envelope_correlation": 0.9985500051792646, "全帯域_40ms_modulation_mae_db": 0.22952091514159664, "300–1000 Hz_10ms_log_rms_mae_db": 1.9496794563416546, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09716434223347774, "300–1000 Hz_10ms_envelope_correlation": 0.9931055660463414, "300–1000 Hz_10ms_modulation_mae_db": 0.632116136568211, "300–1000 Hz_40ms_log_rms_mae_db": 1.2087337763325605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056673977615237875, "300–1000 Hz_40ms_envelope_correlation": 0.9977537308945895, "300–1000 Hz_40ms_modulation_mae_db": 0.8834804009707273, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2382377122433463, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1602108885925225, "1000–3000 Hz_10ms_envelope_correlation": 0.9798440712255029, "1000–3000 Hz_10ms_modulation_mae_db": 0.8883123504756453, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3884139886695388, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09154522647469462, "1000–3000 Hz_40ms_envelope_correlation": 0.9931133020912274, "1000–3000 Hz_40ms_modulation_mae_db": 1.2489386816201746, "3000–6000 Hz_10ms_log_rms_mae_db": 2.779338801623123, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1762759487401115, "3000–6000 Hz_10ms_envelope_correlation": 0.9806824275781927, "3000–6000 Hz_10ms_modulation_mae_db": 0.4788963210730038, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0268663245215164, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11734552686670188, "3000–6000 Hz_40ms_envelope_correlation": 0.9914600459111808, "3000–6000 Hz_40ms_modulation_mae_db": 0.4284062614877808, "mr_stft_db": 6.108239975964527, "band_mae_db": 0.116938394668695, "rms_gain_error_db": -0.06744605625220813, "output_peak": 0.5064005611359167}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.204808073049828, "全帯域_10ms_rms_envelope_nrmse": 0.06947096507870615, "全帯域_10ms_envelope_correlation": 0.9955051418710064, "全帯域_10ms_modulation_mae_db": 0.2794373147219768, "全帯域_40ms_log_rms_mae_db": 0.6272100266018255, "全帯域_40ms_rms_envelope_nrmse": 0.039061735486202856, "全帯域_40ms_envelope_correlation": 0.9985670268034933, "全帯域_40ms_modulation_mae_db": 0.3276375798453754, "300–1000 Hz_10ms_log_rms_mae_db": 1.964610721294681, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10865670404467884, "300–1000 Hz_10ms_envelope_correlation": 0.9914524208951487, "300–1000 Hz_10ms_modulation_mae_db": 1.072632878285402, "300–1000 Hz_40ms_log_rms_mae_db": 1.253570968047403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061408035032025465, "300–1000 Hz_40ms_envelope_correlation": 0.9975041917233427, "300–1000 Hz_40ms_modulation_mae_db": 1.1980742296057316, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2428668622714794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1570960130791812, "1000–3000 Hz_10ms_envelope_correlation": 0.9810204281038625, "1000–3000 Hz_10ms_modulation_mae_db": 0.41422493577486835, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3687727691846943, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09070542176822957, "1000–3000 Hz_40ms_envelope_correlation": 0.9940489119494081, "1000–3000 Hz_40ms_modulation_mae_db": 0.6968872470148574, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8202982655231548, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2055154624475502, "3000–6000 Hz_10ms_envelope_correlation": 0.9747048269257764, "3000–6000 Hz_10ms_modulation_mae_db": 0.3999657698167158, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0344058315700555, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13404160457061245, "3000–6000 Hz_40ms_envelope_correlation": 0.9890834090837154, "3000–6000 Hz_40ms_modulation_mae_db": 0.17628304357531296, "mr_stft_db": 6.125440948285188, "band_mae_db": 0.23461496670042878, "rms_gain_error_db": -0.09298202966796013, "output_peak": 0.48477800720613845}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2650228522358413, "全帯域_10ms_rms_envelope_nrmse": 0.06444524416972673, "全帯域_10ms_envelope_correlation": 0.9961586195086136, "全帯域_10ms_modulation_mae_db": 0.19491260875044994, "全帯域_40ms_log_rms_mae_db": 0.6389657899022618, "全帯域_40ms_rms_envelope_nrmse": 0.03409760385064124, "全帯域_40ms_envelope_correlation": 0.9989491385649667, "全帯域_40ms_modulation_mae_db": 0.21874632947087014, "300–1000 Hz_10ms_log_rms_mae_db": 2.0098543960066158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09622054742767526, "300–1000 Hz_10ms_envelope_correlation": 0.9933226148046445, "300–1000 Hz_10ms_modulation_mae_db": 0.6945299015409728, "300–1000 Hz_40ms_log_rms_mae_db": 1.2071519325628577, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053212008979811196, "300–1000 Hz_40ms_envelope_correlation": 0.9981485484951278, "300–1000 Hz_40ms_modulation_mae_db": 0.8793994736804779, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2382876087340895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16449593838152912, "1000–3000 Hz_10ms_envelope_correlation": 0.9795275080342253, "1000–3000 Hz_10ms_modulation_mae_db": 0.41605932161795467, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3933781155189562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09727124593822124, "1000–3000 Hz_40ms_envelope_correlation": 0.993627278182367, "1000–3000 Hz_40ms_modulation_mae_db": 0.427622231820944, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8503734524956097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1886000326022349, "3000–6000 Hz_10ms_envelope_correlation": 0.9777762496608591, "3000–6000 Hz_10ms_modulation_mae_db": 0.8287802420921212, "3000–6000 Hz_40ms_log_rms_mae_db": 2.050088733054289, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11998355265712643, "3000–6000 Hz_40ms_envelope_correlation": 0.9911268437979676, "3000–6000 Hz_40ms_modulation_mae_db": 0.5389869698021882, "mr_stft_db": 6.135745309579547, "band_mae_db": 0.23696762794578521, "rms_gain_error_db": -0.10114496376062976, "output_peak": 0.4392610000282033}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1882178769365552, "全帯域_10ms_rms_envelope_nrmse": 0.05998851842525492, "全帯域_10ms_envelope_correlation": 0.9966000699425863, "全帯域_10ms_modulation_mae_db": 0.3894702230491438, "全帯域_40ms_log_rms_mae_db": 0.6066065243131885, "全帯域_40ms_rms_envelope_nrmse": 0.03287188223851168, "全帯域_40ms_envelope_correlation": 0.9989080389883228, "全帯域_40ms_modulation_mae_db": 0.3378965019569977, "300–1000 Hz_10ms_log_rms_mae_db": 1.9894519704323474, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0879224264870772, "300–1000 Hz_10ms_envelope_correlation": 0.9942030445744985, "300–1000 Hz_10ms_modulation_mae_db": 0.8109632368686143, "300–1000 Hz_40ms_log_rms_mae_db": 1.188400483938418, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04938397813850911, "300–1000 Hz_40ms_envelope_correlation": 0.9980933684536555, "300–1000 Hz_40ms_modulation_mae_db": 0.9528649142450876, "1000–3000 Hz_10ms_log_rms_mae_db": 2.264713835826252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16385914999133894, "1000–3000 Hz_10ms_envelope_correlation": 0.9788502049943206, "1000–3000 Hz_10ms_modulation_mae_db": 0.726637038515365, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3375600472423865, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.087186793728287, "1000–3000 Hz_40ms_envelope_correlation": 0.9936978406435424, "1000–3000 Hz_40ms_modulation_mae_db": 0.30260261812078126, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8471522129289655, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19761779550920028, "3000–6000 Hz_10ms_envelope_correlation": 0.9761176103944379, "3000–6000 Hz_10ms_modulation_mae_db": 0.479021270862773, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0532043783506926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1370113563150304, "3000–6000 Hz_40ms_envelope_correlation": 0.988162849616734, "3000–6000 Hz_40ms_modulation_mae_db": 0.3317767516592163, "mr_stft_db": 6.133141719973291, "band_mae_db": 0.0761508080495336, "rms_gain_error_db": -0.006282191951097282, "output_peak": 0.45922315838199246}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.192110609894052, "全帯域_10ms_rms_envelope_nrmse": 0.06426083011275527, "全帯域_10ms_envelope_correlation": 0.9961142687176133, "全帯域_10ms_modulation_mae_db": 0.19987482461141926, "全帯域_40ms_log_rms_mae_db": 0.7200286133404322, "全帯域_40ms_rms_envelope_nrmse": 0.03633152305428735, "全帯域_40ms_envelope_correlation": 0.9987091768428642, "全帯域_40ms_modulation_mae_db": 0.131637747070389, "300–1000 Hz_10ms_log_rms_mae_db": 1.9941408619793604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09515761108440986, "300–1000 Hz_10ms_envelope_correlation": 0.9933261023344098, "300–1000 Hz_10ms_modulation_mae_db": 0.35537210118469664, "300–1000 Hz_40ms_log_rms_mae_db": 1.1589152832682261, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05429274870167572, "300–1000 Hz_40ms_envelope_correlation": 0.9978639909739926, "300–1000 Hz_40ms_modulation_mae_db": 0.6869873351726863, "1000–3000 Hz_10ms_log_rms_mae_db": 2.287158488280954, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1580144349287641, "1000–3000 Hz_10ms_envelope_correlation": 0.9803413327024365, "1000–3000 Hz_10ms_modulation_mae_db": 0.31957407809929045, "1000–3000 Hz_40ms_log_rms_mae_db": 1.399586542269974, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08755072212620728, "1000–3000 Hz_40ms_envelope_correlation": 0.99377609746836, "1000–3000 Hz_40ms_modulation_mae_db": 0.7410340935336911, "3000–6000 Hz_10ms_log_rms_mae_db": 2.805945607287918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18558407453852668, "3000–6000 Hz_10ms_envelope_correlation": 0.978548923539529, "3000–6000 Hz_10ms_modulation_mae_db": 0.2950022281277584, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1049257101946854, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11393098380320768, "3000–6000 Hz_40ms_envelope_correlation": 0.9916933938163974, "3000–6000 Hz_40ms_modulation_mae_db": 0.5060666354688067, "mr_stft_db": 6.128635176138199, "band_mae_db": 0.10320042318224387, "rms_gain_error_db": -0.05145604345964225, "output_peak": 0.48270242115492173}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2174368487110658, "全帯域_10ms_rms_envelope_nrmse": 0.06527410817541184, "全帯域_10ms_envelope_correlation": 0.9960646167778995, "全帯域_10ms_modulation_mae_db": 0.35797040865353874, "全帯域_40ms_log_rms_mae_db": 0.6566830420964214, "全帯域_40ms_rms_envelope_nrmse": 0.037891042017443734, "全帯域_40ms_envelope_correlation": 0.9986795807845187, "全帯域_40ms_modulation_mae_db": 0.4465137645290316, "300–1000 Hz_10ms_log_rms_mae_db": 1.9882419106772384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09375728023585796, "300–1000 Hz_10ms_envelope_correlation": 0.9936561539037273, "300–1000 Hz_10ms_modulation_mae_db": 0.6598112135386455, "300–1000 Hz_40ms_log_rms_mae_db": 1.2571643893656423, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058737706911427226, "300–1000 Hz_40ms_envelope_correlation": 0.9976166604254946, "300–1000 Hz_40ms_modulation_mae_db": 0.7541924473019078, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2803340169255772, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16088738470898864, "1000–3000 Hz_10ms_envelope_correlation": 0.9799594356774893, "1000–3000 Hz_10ms_modulation_mae_db": 1.0507147434075312, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3979160509084165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08648004980453976, "1000–3000 Hz_40ms_envelope_correlation": 0.9945824713486013, "1000–3000 Hz_40ms_modulation_mae_db": 1.343116599586341, "3000–6000 Hz_10ms_log_rms_mae_db": 2.818262126754268, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18681904267126562, "3000–6000 Hz_10ms_envelope_correlation": 0.9801958922014244, "3000–6000 Hz_10ms_modulation_mae_db": 0.24525919242505503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.963186896315503, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11941483630236725, "3000–6000 Hz_40ms_envelope_correlation": 0.9922729037914797, "3000–6000 Hz_40ms_modulation_mae_db": 0.18310183962018597, "mr_stft_db": 6.11957401438011, "band_mae_db": 0.17464511464539953, "rms_gain_error_db": -0.10169229225251396, "output_peak": 0.4602357735248011}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1998377561776794, "全帯域_10ms_rms_envelope_nrmse": 0.07150541665895592, "全帯域_10ms_envelope_correlation": 0.9951725678945964, "全帯域_10ms_modulation_mae_db": 0.1675652878851626, "全帯域_40ms_log_rms_mae_db": 0.6305120769798505, "全帯域_40ms_rms_envelope_nrmse": 0.03896019993128783, "全帯域_40ms_envelope_correlation": 0.9984931327920296, "全帯域_40ms_modulation_mae_db": 0.2707316078119004, "300–1000 Hz_10ms_log_rms_mae_db": 1.8882523979987742, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09983766641927036, "300–1000 Hz_10ms_envelope_correlation": 0.9926085866079476, "300–1000 Hz_10ms_modulation_mae_db": 0.5504791485205064, "300–1000 Hz_40ms_log_rms_mae_db": 1.1783763792174708, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05856399473168723, "300–1000 Hz_40ms_envelope_correlation": 0.9974516251726533, "300–1000 Hz_40ms_modulation_mae_db": 0.690849430219644, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2358144194466627, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16168502759188544, "1000–3000 Hz_10ms_envelope_correlation": 0.9802688021045843, "1000–3000 Hz_10ms_modulation_mae_db": 0.7388251103128161, "1000–3000 Hz_40ms_log_rms_mae_db": 1.333120759419806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09101259580381281, "1000–3000 Hz_40ms_envelope_correlation": 0.994597975168485, "1000–3000 Hz_40ms_modulation_mae_db": 0.7312991694941605, "3000–6000 Hz_10ms_log_rms_mae_db": 2.77544546499597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17195844843874356, "3000–6000 Hz_10ms_envelope_correlation": 0.9816866465254628, "3000–6000 Hz_10ms_modulation_mae_db": 0.32370851204739814, "3000–6000 Hz_40ms_log_rms_mae_db": 2.009145150346161, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10798877954440184, "3000–6000 Hz_40ms_envelope_correlation": 0.9925981472162339, "3000–6000 Hz_40ms_modulation_mae_db": 0.4339543219273132, "mr_stft_db": 6.114418905095252, "band_mae_db": 0.19290866279646593, "rms_gain_error_db": -0.05070675707209849, "output_peak": 0.4722486681221231}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.414755496833706, "spectral_convergence": 0.22303402993172877, "spectral_convergence_status": "ok", "amplitude_floor": 1.642409573690628e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.07296033236049, "spectral_convergence": 0.21512485074921775, "spectral_convergence_status": "ok", "amplitude_floor": 1.497573454010957e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.911388000392431, "spectral_convergence": 0.21101987487707913, "spectral_convergence_status": "ok", "amplitude_floor": 1.2731241516974033e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012546591027172457, "output_power": 0.0012945659507535188, "delta_db": 0.1359843834484214, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0029841932734943235, "output_power": 0.0029043841977032803, "delta_db": -0.11772881952525223, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004877853564370709, "output_power": 0.00048106820947606054, "delta_db": -0.06022100149378627, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013442175178747268, "output_power": 0.0001283316328810524, "delta_db": -0.20135830609372019, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010124396480278944, "output_power": 9.035395670714449e-05, "delta_db": -0.4942196858859437, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001060495609474016, "output_power": 0.0001105961265412168, "delta_db": 0.18231041893749578, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19863710256410327, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.05089807643004243, "pitch": null, "output_rms_dbfs": -23.002229518189683, "output_peak": 0.43017433626863066, "reference_rms_dbfs": -22.951331441759642, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.13303460986221}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024147266762860616, "log_rms_mae_db": 1.2285891404902243, "rms_envelope_nrmse": 0.06769034867702226, "envelope_correlation": 0.9956776312227112, "rms_gain_error_db": -0.05089598838648803, "reference_mean_power": 0.005072567967533502, "output_mean_power": 0.005013468332071915}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4164286663529534e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010865583816620668, "output_power": 0.0010493701109746724, "reference_active": true, "delta_db": -0.1512437617308846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020269542962956985, "output_power": 0.00019554365056301647, "reference_active": true, "delta_db": -0.15600237668321876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.69603101954121e-05, "output_power": 2.4974411098382903e-05, "reference_active": true, "delta_db": -0.3322962845672009}], "active_band_count": 3, "active_band_mae_db": 0.21318080766043476}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000225157536036158, "log_rms_mae_db": 0.6439774887238734, "rms_envelope_nrmse": 0.037857801431896924, "envelope_correlation": 0.9985758878125524, "rms_gain_error_db": -0.05089435133045952, "reference_mean_power": 0.005098028148503288, "output_mean_power": 0.005038633780052148}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2603245746946083e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010232290392639372, "output_power": 0.0009881065581746208, "reference_active": true, "delta_db": -0.15169075167389912}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010544090587602268, "output_power": 0.00010057492851142648, "reference_active": true, "delta_db": -0.2051939587934248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0049541105469593e-06, "output_power": 1.9560868771896324e-06, "reference_active": true, "delta_db": -0.10716297389633844}], "active_band_count": 3, "active_band_mae_db": 0.15468256145455414}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020992712290116374, "log_rms_mae_db": 1.9197331545735508, "rms_envelope_nrmse": 0.09283347311566933, "envelope_correlation": 0.9937010592530409, "rms_gain_error_db": -0.1682079693355174, "reference_mean_power": 0.0023388664575994473, "output_mean_power": 0.0022500109343140343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4073975659228857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007410848450397675, "output_power": 0.0006878651202254484, "reference_active": true, "delta_db": -0.3236464405450983}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011531333502346333, "output_power": 0.00010743990720786321, "reference_active": true, "delta_db": -0.30713908150365365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6543285086796136e-05, "output_power": 1.6463479328593217e-05, "reference_active": true, "delta_db": -2.0743304890314267}], "active_band_count": 3, "active_band_mae_db": 0.9017053370267263}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002036893389343094, "log_rms_mae_db": 1.2079985021708421, "rms_envelope_nrmse": 0.055078772955287264, "envelope_correlation": 0.9978453975617317, "rms_gain_error_db": -0.16821416373833567, "reference_mean_power": 0.0023506239289763434, "output_mean_power": 0.0022613185040144396}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3312664615669961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215176046332657, "output_power": 0.0006687226482561008, "reference_active": true, "delta_db": -0.3300089999875502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5162124576630795e-05, "output_power": 5.000780813494349e-05, "reference_active": true, "delta_db": -0.42603165096748097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1332065410684048e-06, "output_power": 1.3955133218931325e-06, "reference_active": true, "delta_db": -1.8429891997347871}], "active_band_count": 3, "active_band_mae_db": 0.8663432835632728}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.160365671559225e-05, "log_rms_mae_db": 2.213794422287487, "rms_envelope_nrmse": 0.16924786089686128, "envelope_correlation": 0.977978727472348, "rms_gain_error_db": -0.3309858178502597, "reference_mean_power": 0.00021265544148097967, "output_mean_power": 0.00019705067107863325}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.181263558889223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.569393841260949e-05, "output_power": 4.8324238869498434e-05, "reference_active": true, "delta_db": -0.6164290782980686}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4402699100691371e-05, "output_power": 1.0060846800317823e-05, "reference_active": true, "delta_db": -1.558093516484444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.927958039351209e-06, "output_power": 2.103980398292736e-06, "reference_active": true, "delta_db": -1.4352315914314984}], "active_band_count": 3, "active_band_mae_db": 1.2032513954046704}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.199537921297776e-05, "log_rms_mae_db": 1.327471496169193, "rms_envelope_nrmse": 0.09715787346182597, "envelope_correlation": 0.9932138561732522, "rms_gain_error_db": -0.33100868474931844, "reference_mean_power": 0.00021371687213348275, "output_mean_power": 0.00019803317067389382}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0829999638899264e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5369039282181144e-05, "output_power": 4.721290206567406e-05, "reference_active": true, "delta_db": -0.6920629174965269}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.495407348960159e-06, "output_power": 4.370256231239656e-06, "reference_active": true, "delta_db": -1.7209949162659932}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.109761653644897e-07, "output_power": 1.0730151728600348e-07, "reference_active": true, "delta_db": -2.9362753136102135}], "active_band_count": 3, "active_band_mae_db": 1.7831110491242448}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3802525725767476e-05, "log_rms_mae_db": 2.7973842510533293, "rms_envelope_nrmse": 0.18049453723457023, "envelope_correlation": 0.9807659092132145, "rms_gain_error_db": 0.2578392504417618, "reference_mean_power": 6.940550281225073e-05, "output_mean_power": 7.365086127911988e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.994304097313729e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.68316376202847e-05, "output_power": 3.839441996556312e-05, "reference_active": true, "delta_db": 0.18047081948329477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3496621932801275e-06, "output_power": 5.246514097847962e-06, "reference_active": true, "delta_db": -0.08455514987347615}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1665818120524947e-06, "output_power": 1.58798507283822e-06, "reference_active": true, "delta_db": 1.3393121456725363}], "active_band_count": 3, "active_band_mae_db": 0.5347793716764357}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.808436973199796e-05, "log_rms_mae_db": 2.0424251464747014, "rms_envelope_nrmse": 0.11507217231826528, "envelope_correlation": 0.992199915022075, "rms_gain_error_db": 0.257844803731173, "reference_mean_power": 6.97538862423984e-05, "output_mean_power": 7.402064908910961e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.7257017907786166e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.6073966740876416e-05, "output_power": 3.7350442912350554e-05, "reference_active": true, "delta_db": 0.15101855634284508}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3003381054168396e-06, "output_power": 2.3825108942900067e-06, "reference_active": true, "delta_db": 0.15243221536389}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.80849015690717e-08, "output_power": 1.0918999372808319e-07, "reference_active": true, "delta_db": 1.456157738656564}], "active_band_count": 3, "active_band_mae_db": 0.5865361701210997}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8220793660200969, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "db344abba3463beb2bbe25beabac799511caa647f0eaf99ed5070df19d196032", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9797238137893463, "standard_deviation": 0.0009469186575048701, "minimum": 0.977978727472348, "maximum": 0.9810204281038625, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.250150920751981, "standard_deviation": 0.02494481999397896, "minimum": 2.213794422287487, "maximum": 2.287158488280954, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7196998717010177, "standard_deviation": 0.320234324750785, "minimum": 0.31957407809929045, "maximum": 1.2032513954046704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1619370872713839, "standard_deviation": 0.0039034385440863107, "minimum": 0.1570960130791812, "maximum": 0.16924786089686128, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938322166281555, "standard_deviation": 0.0005553170195464352, "minimum": 0.9931133020912274, "maximum": 0.994597975168485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3682774711728707, "standard_deviation": 0.031027988414886982, "minimum": 1.327471496169193, "maximum": 1.399586542269974, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9093264612893994, "standard_deviation": 0.503176920633409, "minimum": 0.30260261812078126, "maximum": 1.7831110491242448, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09111374113822727, "standard_deviation": 0.004215560484776071, "minimum": 0.08648004980453976, "maximum": 0.09727124593822124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9788098107548621, "standard_deviation": 0.0024717133000984612, "minimum": 0.9747048269257764, "maximum": 0.9816866465254628, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.8117750228327925, "standard_deviation": 0.027951831403756223, "minimum": 2.77544546499597, "maximum": 2.8503734524956097, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.44817661351515764, "standard_deviation": 0.1838899977782498, "minimum": 0.24525919242505503, "maximum": 0.8287802420921212, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18660816777277536, "standard_deviation": 0.010970766092210909, "minimum": 0.17195844843874356, "maximum": 0.2055154624475502, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991074688531973, "standard_deviation": 0.0016036029631783671, "minimum": 0.988162849616734, "maximum": 0.9925981472162339, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.0355310213534503, "standard_deviation": 0.040344332152172, "minimum": 1.963186896315503, "maximum": 2.1049257101946854, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.398138999207738, "standard_deviation": 0.15539987909437356, "minimum": 0.17628304357531296, "maximum": 0.5865361701210997, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12059860154721415, "standard_deviation": 0.009974690019315648, "minimum": 0.10798877954440184, "maximum": 0.1370113563150304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931719435524697, "standard_deviation": 0.0008368814860682942, "minimum": 0.9914524208951487, "maximum": 0.9942030445744985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9629956086630278, "standard_deviation": 0.04164682044906398, "minimum": 1.8882523979987742, "maximum": 2.0098543960066158, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7097012441917219, "standard_deviation": 0.2198955547226802, "minimum": 0.35537210118469664, "maximum": 1.072632878285402, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09644375638101457, "standard_deviation": 0.006044623519064729, "minimum": 0.0879224264870772, "maximum": 0.10865670404467884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977846892125735, "standard_deviation": 0.0002553578193409426, "minimum": 0.9974516251726533, "maximum": 0.9981485484951278, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2075389643629277, "standard_deviation": 0.03415179744837985, "minimum": 1.1589152832682261, "maximum": 1.2571643893656423, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8640239393449419, "standard_deviation": 0.1662601154943057, "minimum": 0.6869873351726863, "maximum": 1.1980742296057316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05591890288320764, "standard_deviation": 0.003761185002772309, "minimum": 0.04938397813850911, "maximum": 0.061408035032025465, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1667578875690819, "standard_deviation": 0.06097718846051594, "minimum": 0.0761508080495336, "maximum": 0.23696762794578521, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.124778832409791, "standard_deviation": 0.009867849330340803, "minimum": 6.108239975964527, "maximum": 6.135745309579547, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4668779907278409, "standard_deviation": 0.02499792239704271, "minimum": 0.43017433626863066, "maximum": 0.5064005611359167, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06532605135577405, "standard_deviation": 0.03267871784125761, "minimum": -0.10169229225251396, "maximum": -0.006282191951097282, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9958534144662307, "standard_deviation": 0.0004599397226276549, "minimum": 0.9951725678945964, "maximum": 0.9966000699425863, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2130478235375801, "standard_deviation": 0.024731337091432156, "minimum": 1.1882178769365552, "maximum": 1.2650228522358413, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24423556418958836, "standard_deviation": 0.08873405592919614, "minimum": 0.1514730381845811, "maximum": 0.3894702230491438, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.066456121821251, "standard_deviation": 0.003678865570407136, "minimum": 0.05998851842525492, "maximum": 0.07150541665895592, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986789984710016, "standard_deviation": 0.00016932685395954354, "minimum": 0.9984931327920296, "maximum": 0.9989491385649667, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6443251753269705, "standard_deviation": 0.033820722633392794, "minimum": 0.6066065243131885, "maximum": 0.7200286133404322, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.26467087591008936, "standard_deviation": 0.10390255951764675, "minimum": 0.131637747070389, "maximum": 0.4465137645290316, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03695757152291431, "standard_deviation": 0.0023313046926483547, "minimum": 0.03287188223851168, "maximum": 0.039061735486202856, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0010", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.6096808910369873, "predicted_mos": 2.6096808910369873, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8afea4491d735ccaab05b76370de46238e914c1959d4617eb1afc95b507b8a18", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.249000400304794, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3429981483173572, "全帯域_10ms_rms_envelope_nrmse": 0.07119793012027222, "全帯域_10ms_envelope_correlation": 0.9952540133197157, "全帯域_10ms_modulation_mae_db": 0.31143355037047105, "全帯域_40ms_log_rms_mae_db": 0.7408136022652391, "全帯域_40ms_rms_envelope_nrmse": 0.03942978365949307, "全帯域_40ms_envelope_correlation": 0.998508217409831, "全帯域_40ms_modulation_mae_db": 0.33795995571791915, "300–1000 Hz_10ms_log_rms_mae_db": 1.5998986689003853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1040727775106037, "300–1000 Hz_10ms_envelope_correlation": 0.9921160279011948, "300–1000 Hz_10ms_modulation_mae_db": 0.8461382806074478, "300–1000 Hz_40ms_log_rms_mae_db": 0.9262964067032674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06383571270879587, "300–1000 Hz_40ms_envelope_correlation": 0.997144250074381, "300–1000 Hz_40ms_modulation_mae_db": 0.7367173122176706, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6413297271973255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14315363969409425, "1000–3000 Hz_10ms_envelope_correlation": 0.9847829272233334, "1000–3000 Hz_10ms_modulation_mae_db": 0.4173697803853658, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0229115849321828, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0872770529649183, "1000–3000 Hz_40ms_envelope_correlation": 0.9950178445992314, "1000–3000 Hz_40ms_modulation_mae_db": 0.4675736090968214, "3000–6000 Hz_10ms_log_rms_mae_db": 1.735378636863876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16577474747737567, "3000–6000 Hz_10ms_envelope_correlation": 0.9826964259231038, "3000–6000 Hz_10ms_modulation_mae_db": 0.3307152992758341, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2606510831953939, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10478059266570341, "3000–6000 Hz_40ms_envelope_correlation": 0.9928672220511733, "3000–6000 Hz_40ms_modulation_mae_db": 0.2998510990852291, "mr_stft_db": 5.583330506959747, "band_mae_db": 0.29637945519674086, "rms_gain_error_db": -0.08503488302791737, "output_peak": 0.42507390906352605}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3527387327716205, "全帯域_10ms_rms_envelope_nrmse": 0.06319529055493091, "全帯域_10ms_envelope_correlation": 0.9962411191047086, "全帯域_10ms_modulation_mae_db": 0.2898055167223482, "全帯域_40ms_log_rms_mae_db": 0.7778322116516853, "全帯域_40ms_rms_envelope_nrmse": 0.033959938597571335, "全帯域_40ms_envelope_correlation": 0.9988705210078846, "全帯域_40ms_modulation_mae_db": 0.41048248848679575, "300–1000 Hz_10ms_log_rms_mae_db": 1.5900077462603535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0859064593972572, "300–1000 Hz_10ms_envelope_correlation": 0.9946443604228101, "300–1000 Hz_10ms_modulation_mae_db": 0.996559137653659, "300–1000 Hz_40ms_log_rms_mae_db": 0.9192912695490459, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04676167061692443, "300–1000 Hz_40ms_envelope_correlation": 0.9985322409905085, "300–1000 Hz_40ms_modulation_mae_db": 1.2829373870874388, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5305211843047095, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14119098521024628, "1000–3000 Hz_10ms_envelope_correlation": 0.9854460509309537, "1000–3000 Hz_10ms_modulation_mae_db": 0.5474445679848985, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8949754290047667, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0823768242075893, "1000–3000 Hz_40ms_envelope_correlation": 0.9960375639840481, "1000–3000 Hz_40ms_modulation_mae_db": 0.5559503668730718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8002508831307624, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15533047285939822, "3000–6000 Hz_10ms_envelope_correlation": 0.9850487313612251, "3000–6000 Hz_10ms_modulation_mae_db": 0.53426383767284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3296697036243308, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10193812482597191, "3000–6000 Hz_40ms_envelope_correlation": 0.9933055636592147, "3000–6000 Hz_40ms_modulation_mae_db": 0.5962082007461317, "mr_stft_db": 5.5978547004785755, "band_mae_db": 0.22602421413713678, "rms_gain_error_db": -0.05543887222460658, "output_peak": 0.45841959714767894}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3589702174029854, "全帯域_10ms_rms_envelope_nrmse": 0.0678096908112109, "全帯域_10ms_envelope_correlation": 0.9957255874716832, "全帯域_10ms_modulation_mae_db": 0.33044752255272203, "全帯域_40ms_log_rms_mae_db": 0.7790751956618713, "全帯域_40ms_rms_envelope_nrmse": 0.03544595797181782, "全帯域_40ms_envelope_correlation": 0.9988380825320826, "全帯域_40ms_modulation_mae_db": 0.17131020762487456, "300–1000 Hz_10ms_log_rms_mae_db": 1.63299371892632, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09311051929082696, "300–1000 Hz_10ms_envelope_correlation": 0.9936043604413861, "300–1000 Hz_10ms_modulation_mae_db": 0.5015819476485367, "300–1000 Hz_40ms_log_rms_mae_db": 0.9409920773383932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04927401731442667, "300–1000 Hz_40ms_envelope_correlation": 0.9982685513970193, "300–1000 Hz_40ms_modulation_mae_db": 0.5725808714362196, "1000–3000 Hz_10ms_log_rms_mae_db": 1.611142934884372, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15548158119996489, "1000–3000 Hz_10ms_envelope_correlation": 0.9827220213215657, "1000–3000 Hz_10ms_modulation_mae_db": 0.30277530697171007, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9463495269886468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1035224708676519, "1000–3000 Hz_40ms_envelope_correlation": 0.9935393241975925, "1000–3000 Hz_40ms_modulation_mae_db": 0.35004833871735225, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7755410732308778, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16791896540494206, "3000–6000 Hz_10ms_envelope_correlation": 0.9822725498814392, "3000–6000 Hz_10ms_modulation_mae_db": 0.3817951261302082, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3210670954933514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11516557406539933, "3000–6000 Hz_40ms_envelope_correlation": 0.9913308286497136, "3000–6000 Hz_40ms_modulation_mae_db": 0.47337612856009503, "mr_stft_db": 5.578226647585827, "band_mae_db": 0.22245935232317293, "rms_gain_error_db": -0.09553748909760142, "output_peak": 0.44878335760710325}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2475795984982372, "全帯域_10ms_rms_envelope_nrmse": 0.06643006442763462, "全帯域_10ms_envelope_correlation": 0.9959308616442893, "全帯域_10ms_modulation_mae_db": 0.5512890652795653, "全帯域_40ms_log_rms_mae_db": 0.7065835014627819, "全帯域_40ms_rms_envelope_nrmse": 0.03946027004035343, "全帯域_40ms_envelope_correlation": 0.998568760782062, "全帯域_40ms_modulation_mae_db": 0.5744862461529787, "300–1000 Hz_10ms_log_rms_mae_db": 1.6006494203831256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10346096514997594, "300–1000 Hz_10ms_envelope_correlation": 0.9925221377524442, "300–1000 Hz_10ms_modulation_mae_db": 0.9086608349422658, "300–1000 Hz_40ms_log_rms_mae_db": 0.9793017111815114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0652154247145773, "300–1000 Hz_40ms_envelope_correlation": 0.9973815042537191, "300–1000 Hz_40ms_modulation_mae_db": 1.0930312574046062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5586368331352345, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13228272926483448, "1000–3000 Hz_10ms_envelope_correlation": 0.9867451542126655, "1000–3000 Hz_10ms_modulation_mae_db": 0.33661234478039054, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9021635649374514, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0746257043206738, "1000–3000 Hz_40ms_envelope_correlation": 0.9961624712382051, "1000–3000 Hz_40ms_modulation_mae_db": 0.49706532318014934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.835513250662157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14864705437861192, "3000–6000 Hz_10ms_envelope_correlation": 0.9864468431930513, "3000–6000 Hz_10ms_modulation_mae_db": 0.5357799983552344, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3731812371567431, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10375062008946603, "3000–6000 Hz_40ms_envelope_correlation": 0.9935467920457585, "3000–6000 Hz_40ms_modulation_mae_db": 0.46446156114013615, "mr_stft_db": 5.591697946141811, "band_mae_db": 0.21179511993385428, "rms_gain_error_db": -0.1096044712378853, "output_peak": 0.42864115085588067}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.319345901535991, "全帯域_10ms_rms_envelope_nrmse": 0.0670774816651221, "全帯域_10ms_envelope_correlation": 0.9958106683482868, "全帯域_10ms_modulation_mae_db": 0.26804955177463713, "全帯域_40ms_log_rms_mae_db": 0.7460371007211359, "全帯域_40ms_rms_envelope_nrmse": 0.03738238706408294, "全帯域_40ms_envelope_correlation": 0.9986916690589867, "全帯域_40ms_modulation_mae_db": 0.3859894821868104, "300–1000 Hz_10ms_log_rms_mae_db": 1.5914959459212394, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09269097174502211, "300–1000 Hz_10ms_envelope_correlation": 0.9938682863674122, "300–1000 Hz_10ms_modulation_mae_db": 0.2706799760276373, "300–1000 Hz_40ms_log_rms_mae_db": 0.9210134757402411, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054641656431729475, "300–1000 Hz_40ms_envelope_correlation": 0.9980790897958002, "300–1000 Hz_40ms_modulation_mae_db": 0.22196966243666447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5729294138247125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15101153060625588, "1000–3000 Hz_10ms_envelope_correlation": 0.9827324717938118, "1000–3000 Hz_10ms_modulation_mae_db": 0.4500528497335574, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8838008634140648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0873781125622913, "1000–3000 Hz_40ms_envelope_correlation": 0.9948091826545814, "1000–3000 Hz_40ms_modulation_mae_db": 0.41120345601002617, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7498556355985762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17617486933233098, "3000–6000 Hz_10ms_envelope_correlation": 0.9805961077251614, "3000–6000 Hz_10ms_modulation_mae_db": 0.5001225911367123, "3000–6000 Hz_40ms_log_rms_mae_db": 1.331925384635243, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11249491781054279, "3000–6000 Hz_40ms_envelope_correlation": 0.9917560997714243, "3000–6000 Hz_40ms_modulation_mae_db": 0.5410743234646959, "mr_stft_db": 5.586842047834996, "band_mae_db": 0.18641483480757817, "rms_gain_error_db": -0.09206064131289024, "output_peak": 0.46061182551083274}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3206551527931727, "全帯域_10ms_rms_envelope_nrmse": 0.07153756648662991, "全帯域_10ms_envelope_correlation": 0.9951774258164051, "全帯域_10ms_modulation_mae_db": 0.5938581941758162, "全帯域_40ms_log_rms_mae_db": 0.7462918949649895, "全帯域_40ms_rms_envelope_nrmse": 0.03634596716950987, "全帯域_40ms_envelope_correlation": 0.9987046035265502, "全帯域_40ms_modulation_mae_db": 0.6453821543558715, "300–1000 Hz_10ms_log_rms_mae_db": 1.6384445204430347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10247227289386782, "300–1000 Hz_10ms_envelope_correlation": 0.9923759143880205, "300–1000 Hz_10ms_modulation_mae_db": 0.42999028454537913, "300–1000 Hz_40ms_log_rms_mae_db": 1.0632588581325888, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05919175098431443, "300–1000 Hz_40ms_envelope_correlation": 0.9976173469869037, "300–1000 Hz_40ms_modulation_mae_db": 0.37343860039531357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5819414106483867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14791425299977742, "1000–3000 Hz_10ms_envelope_correlation": 0.9829650325610091, "1000–3000 Hz_10ms_modulation_mae_db": 0.565064940983197, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8842640142519903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0825264425876443, "1000–3000 Hz_40ms_envelope_correlation": 0.9947910366800674, "1000–3000 Hz_40ms_modulation_mae_db": 0.49950666783893577, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7393981286939522, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18110198592826376, "3000–6000 Hz_10ms_envelope_correlation": 0.9795967468581833, "3000–6000 Hz_10ms_modulation_mae_db": 0.1549173456211356, "3000–6000 Hz_40ms_log_rms_mae_db": 1.297733888870424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11627914249050131, "3000–6000 Hz_40ms_envelope_correlation": 0.9912331546223532, "3000–6000 Hz_40ms_modulation_mae_db": 0.30952023925371724, "mr_stft_db": 5.573149036853355, "band_mae_db": 0.15383188423331548, "rms_gain_error_db": -0.061484957125678384, "output_peak": 0.4708103904130509}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2540909357741794, "全帯域_10ms_rms_envelope_nrmse": 0.05994083514185518, "全帯域_10ms_envelope_correlation": 0.996610388696766, "全帯域_10ms_modulation_mae_db": 0.13452645849042313, "全帯域_40ms_log_rms_mae_db": 0.6969900652628896, "全帯域_40ms_rms_envelope_nrmse": 0.031734588411744154, "全帯域_40ms_envelope_correlation": 0.9989980047845055, "全帯域_40ms_modulation_mae_db": 0.17112725226788453, "300–1000 Hz_10ms_log_rms_mae_db": 1.6262907393619357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0911960687550196, "300–1000 Hz_10ms_envelope_correlation": 0.9938512495280725, "300–1000 Hz_10ms_modulation_mae_db": 0.7174992139250667, "300–1000 Hz_40ms_log_rms_mae_db": 0.940646943686707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04792552218093169, "300–1000 Hz_40ms_envelope_correlation": 0.9983456674045122, "300–1000 Hz_40ms_modulation_mae_db": 0.8875016941746111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5823329351628237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14663039984464427, "1000–3000 Hz_10ms_envelope_correlation": 0.9832905262159393, "1000–3000 Hz_10ms_modulation_mae_db": 0.7983656455046436, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9712279558073444, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0855362394644216, "1000–3000 Hz_40ms_envelope_correlation": 0.9943361834671198, "1000–3000 Hz_40ms_modulation_mae_db": 0.46162434465761765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7128983057779943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19216230347791566, "3000–6000 Hz_10ms_envelope_correlation": 0.9774462382293037, "3000–6000 Hz_10ms_modulation_mae_db": 0.1718681300034998, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2729142440339116, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1149472344966928, "3000–6000 Hz_40ms_envelope_correlation": 0.9916838960673044, "3000–6000 Hz_40ms_modulation_mae_db": 0.21116631350566176, "mr_stft_db": 5.598888218652715, "band_mae_db": 0.18604861820073806, "rms_gain_error_db": -0.04021294198316861, "output_peak": 0.4481122786202492}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3040844299182703, "全帯域_10ms_rms_envelope_nrmse": 0.06638493481453317, "全帯域_10ms_envelope_correlation": 0.995832157047984, "全帯域_10ms_modulation_mae_db": 0.3190678910066063, "全帯域_40ms_log_rms_mae_db": 0.6939264780490423, "全帯域_40ms_rms_envelope_nrmse": 0.03532891686053533, "全帯域_40ms_envelope_correlation": 0.9987421311463199, "全帯域_40ms_modulation_mae_db": 0.46225145029306924, "300–1000 Hz_10ms_log_rms_mae_db": 1.683058514910273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09645370253377794, "300–1000 Hz_10ms_envelope_correlation": 0.9931103390753275, "300–1000 Hz_10ms_modulation_mae_db": 0.7914298181666813, "300–1000 Hz_40ms_log_rms_mae_db": 0.9458044646336555, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05421281202566308, "300–1000 Hz_40ms_envelope_correlation": 0.9978435231085807, "300–1000 Hz_40ms_modulation_mae_db": 0.9110702317105458, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5699725707115328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15199292612965803, "1000–3000 Hz_10ms_envelope_correlation": 0.9822068824508755, "1000–3000 Hz_10ms_modulation_mae_db": 0.1960119724490934, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9435969669256725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09058407979881873, "1000–3000 Hz_40ms_envelope_correlation": 0.9938568752955022, "1000–3000 Hz_40ms_modulation_mae_db": 0.242753051272716, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7484564805058258, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16435127578019332, "3000–6000 Hz_10ms_envelope_correlation": 0.9845795167997305, "3000–6000 Hz_10ms_modulation_mae_db": 0.3617473949478991, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3170911051265994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11215207559486023, "3000–6000 Hz_40ms_envelope_correlation": 0.9929629418579167, "3000–6000 Hz_40ms_modulation_mae_db": 0.43595953875648275, "mr_stft_db": 5.600262545854055, "band_mae_db": 0.18873540398742852, "rms_gain_error_db": -0.024001123991595992, "output_peak": 0.43478415758551014}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.73073450963741, "spectral_convergence": 0.21794341902882072, "spectral_convergence_status": "ok", "amplitude_floor": 1.642409573690628e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.38734730195899, "spectral_convergence": 0.2083823432348835, "spectral_convergence_status": "ok", "amplitude_floor": 1.497573454010957e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6319097092828425, "spectral_convergence": 0.20450394611939549, "spectral_convergence_status": "ok", "amplitude_floor": 1.2731241516974033e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012546591027172457, "output_power": 0.0013313949857223532, "delta_db": 0.2578117516895753, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0029841932734943235, "output_power": 0.002863557174643364, "delta_db": -0.17921088326180984, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004877853564370709, "output_power": 0.00046217995837213643, "delta_db": -0.23417649304978394, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013442175178747268, "output_power": 0.00011922626977734746, "delta_db": -0.5209759440689388, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010124396480278944, "output_power": 9.763422825179242e-05, "delta_db": -0.15767046436176446, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001060495609474016, "output_power": 9.608724806216376e-05, "delta_db": -0.42843119474857283, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29637945519674086, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08503488302791737, "pitch": null, "output_rms_dbfs": -23.03636632478756, "output_peak": 0.42507390906352605, "reference_rms_dbfs": -22.951331441759642, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.583330506959747}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024147266762860616, "log_rms_mae_db": 1.3429981483173572, "rms_envelope_nrmse": 0.07119793012027222, "envelope_correlation": 0.9952540133197157, "rms_gain_error_db": -0.085034469580191, "reference_mean_power": 0.005072567967533502, "output_mean_power": 0.004974213569543587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4164286663529534e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010865583816620668, "output_power": 0.0010710496278035163, "reference_active": true, "delta_db": -0.06243471653352646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020269542962956985, "output_power": 0.0001870938215138125, "reference_active": true, "delta_db": -0.34784510516160716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.69603101954121e-05, "output_power": 2.3895867516293587e-05, "reference_active": true, "delta_db": -0.5240208294162796}], "active_band_count": 3, "active_band_mae_db": 0.31143355037047105}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000225157536036158, "log_rms_mae_db": 0.7408136022652391, "rms_envelope_nrmse": 0.03942978365949307, "envelope_correlation": 0.998508217409831, "rms_gain_error_db": -0.0850432460902234, "reference_mean_power": 0.005098028148503288, "output_mean_power": 0.004999169988467135}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2603245746946083e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010232290392639372, "output_power": 0.0010069974994057233, "reference_active": true, "delta_db": -0.06944464825896567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010544090587602268, "output_power": 9.788548706557512e-05, "reference_active": true, "delta_db": -0.3229082226760044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0049541105469593e-06, "output_power": 1.737607719466231e-06, "reference_active": true, "delta_db": -0.6215269962187874}], "active_band_count": 3, "active_band_mae_db": 0.33795995571791915}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020992712290116374, "log_rms_mae_db": 1.5998986689003853, "rms_envelope_nrmse": 0.1040727775106037, "envelope_correlation": 0.9921160279011948, "rms_gain_error_db": -0.20554267028421294, "reference_mean_power": 0.0023388664575994473, "output_mean_power": 0.0022307513195245674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4073975659228857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007410848450397675, "output_power": 0.0007111559052635599, "reference_active": true, "delta_db": -0.17903111503115413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011531333502346333, "output_power": 0.00010518809558424757, "reference_active": true, "delta_db": -0.3991294038563352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6543285086796136e-05, "output_power": 1.690165526320499e-05, "reference_active": true, "delta_db": -1.960254322934854}], "active_band_count": 3, "active_band_mae_db": 0.8461382806074478}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002036893389343094, "log_rms_mae_db": 0.9262964067032674, "rms_envelope_nrmse": 0.06383571270879587, "envelope_correlation": 0.997144250074381, "rms_gain_error_db": -0.2055434554419226, "reference_mean_power": 0.0023506239289763434, "output_mean_power": 0.0022419648912555615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3312664615669961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215176046332657, "output_power": 0.0006873816230619048, "reference_active": true, "delta_db": -0.21049014879755618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5162124576630795e-05, "output_power": 5.0774455492784385e-05, "reference_active": true, "delta_db": -0.35995710025120464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1332065410684048e-06, "output_power": 1.4623874048129786e-06, "reference_active": true, "delta_db": -1.6397046876042507}], "active_band_count": 3, "active_band_mae_db": 0.7367173122176706}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.160365671559225e-05, "log_rms_mae_db": 1.6413297271973255, "rms_envelope_nrmse": 0.14315363969409425, "envelope_correlation": 0.9847829272233334, "rms_gain_error_db": -0.3948035209166756, "reference_mean_power": 0.00021265544148097967, "output_mean_power": 0.00019417626714759607}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.181263558889223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.569393841260949e-05, "output_power": 5.088085221732577e-05, "reference_active": true, "delta_db": -0.39253553414358133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4402699100691371e-05, "output_power": 1.320344608931227e-05, "reference_active": true, "delta_db": -0.3775659103238136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.927958039351209e-06, "output_power": 2.620378689778084e-06, "reference_active": true, "delta_db": -0.48200789668870264}], "active_band_count": 3, "active_band_mae_db": 0.4173697803853658}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.199537921297776e-05, "log_rms_mae_db": 1.0229115849321828, "rms_envelope_nrmse": 0.0872770529649183, "envelope_correlation": 0.9950178445992314, "rms_gain_error_db": -0.3948286130113674, "reference_mean_power": 0.00021371687213348275, "output_mean_power": 0.0001951443349064877}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0829999638899264e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5369039282181144e-05, "output_power": 5.036003338538552e-05, "reference_active": true, "delta_db": -0.4118097871374984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.495407348960159e-06, "output_power": 6.104854545272428e-06, "reference_active": true, "delta_db": -0.26931071467821077}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.109761653644897e-07, "output_power": 1.786789377505568e-07, "reference_active": true, "delta_db": -0.7216003254747549}], "active_band_count": 3, "active_band_mae_db": 0.4675736090968214}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3802525725767476e-05, "log_rms_mae_db": 1.735378636863876, "rms_envelope_nrmse": 0.16577474747737567, "envelope_correlation": 0.9826964259231038, "rms_gain_error_db": -0.11450359213256539, "reference_mean_power": 6.940550281225073e-05, "output_mean_power": 6.75995096401253e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.994304097313729e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.68316376202847e-05, "output_power": 3.5878962163516095e-05, "reference_active": true, "delta_db": -0.11381156895209422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3496621932801275e-06, "output_power": 4.6983031588315104e-06, "reference_active": true, "delta_db": -0.5638532290222235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1665818120524947e-06, "output_power": 1.0850933231361618e-06, "reference_active": true, "delta_db": -0.3144810998531846}], "active_band_count": 3, "active_band_mae_db": 0.3307152992758341}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.808436973199796e-05, "log_rms_mae_db": 1.2606510831953939, "rms_envelope_nrmse": 0.10478059266570341, "envelope_correlation": 0.9928672220511733, "rms_gain_error_db": -0.11450595734625157, "reference_mean_power": 6.97538862423984e-05, "output_mean_power": 6.793879082207056e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.7257017907786166e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.6073966740876416e-05, "output_power": 3.4822823501914636e-05, "reference_active": true, "delta_db": -0.15329918891229413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3003381054168396e-06, "output_power": 1.9914598037227735e-06, "reference_active": true, "delta_db": -0.6262012874555151}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.80849015690717e-08, "output_power": 7.595594740350544e-08, "reference_active": true, "delta_db": -0.12005282088787818}], "active_band_count": 3, "active_band_mae_db": 0.2998510990852291}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8479633984077969, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "72f23f742c8f1989a0c42ca8efca36ea3881353d540751da7132c3d1e2660258", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9838613833387693, "standard_deviation": 0.001608718736143414, "minimum": 0.9822068824508755, "maximum": 0.9867451542126655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5811008762336372, "standard_deviation": 0.03335055084859625, "minimum": 1.5305211843047095, "maximum": 1.6413297271973255, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4517121760991071, "standard_deviation": 0.18676841191433205, "minimum": 0.1960119724490934, "maximum": 0.7983656455046436, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14620725561868442, "standard_deviation": 0.007309885876624434, "minimum": 0.13228272926483448, "maximum": 0.15548158119996489, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9948188102645434, "standard_deviation": 0.0009355362309528431, "minimum": 0.9935393241975925, "maximum": 0.9961624712382051, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.931161238282765, "standard_deviation": 0.049282613576070014, "minimum": 0.8838008634140648, "maximum": 1.0229115849321828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4357156447058363, "standard_deviation": 0.09939096962332247, "minimum": 0.242753051272716, "maximum": 0.5559503668730718, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08672836584675116, "standard_deviation": 0.008299047939546457, "minimum": 0.0746257043206738, "maximum": 0.1035224708676519, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823353949963998, "standard_deviation": 0.0030205426130136288, "minimum": 0.9774462382293037, "maximum": 0.9864468431930513, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.762161549308003, "standard_deviation": 0.03966270361480051, "minimum": 1.7128983057779943, "maximum": 1.835513250662157, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3714012153929204, "standard_deviation": 0.15054619700738883, "minimum": 0.1549173456211356, "maximum": 0.5357799983552344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16893270932987897, "standard_deviation": 0.013983014112485353, "minimum": 0.14864705437861192, "maximum": 0.19216230347791566, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9923358123406073, "standard_deviation": 0.0009311655469699553, "minimum": 0.9912331546223532, "maximum": 0.9935467920457585, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3130292177669998, "standard_deviation": 0.035697078553278354, "minimum": 1.2606510831953939, "maximum": 1.3731812371567431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4164521755640187, "standard_deviation": 0.13144320787951128, "minimum": 0.21116631350566176, "maximum": 0.5962082007461317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11018853525489222, "standard_deviation": 0.005762543937535324, "minimum": 0.10193812482597191, "maximum": 0.11627914249050131, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9932615844845835, "standard_deviation": 0.0008792630631244307, "minimum": 0.9921160279011948, "maximum": 0.9946443604228101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6203549093883334, "standard_deviation": 0.03165552064630574, "minimum": 1.5900077462603535, "maximum": 1.683058514910273, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6828174366895843, "standard_deviation": 0.25515749378113955, "minimum": 0.2706799760276373, "maximum": 0.996559137653659, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09617046715954392, "standard_deviation": 0.00662199270562543, "minimum": 0.0859064593972572, "maximum": 0.1040727775106037, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979015217514282, "standard_deviation": 0.0004912536471650214, "minimum": 0.997144250074381, "maximum": 0.9985322409905085, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9545756508706762, "standard_deviation": 0.047883782635482944, "minimum": 0.9192912695490459, "maximum": 1.0632588581325888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7599058771078837, "standard_deviation": 0.35817524210138124, "minimum": 0.22196966243666447, "maximum": 1.2829373870874388, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05513232087217036, "standard_deviation": 0.007084275213983855, "minimum": 0.04676167061692443, "maximum": 0.0652154247145773, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20896111035249562, "standard_deviation": 0.04243478741500439, "minimum": 0.15383188423331548, "maximum": 0.29637945519674086, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.588781456295135, "standard_deviation": 0.010099804907427679, "minimum": 5.573149036853355, "maximum": 5.600262545854055, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.446904583350479, "standard_deviation": 0.016274414579372634, "minimum": 0.42507390906352605, "maximum": 0.4708103904130509, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07042192250016799, "standard_deviation": 0.02980812631781101, "minimum": -0.1096044712378853, "maximum": -0.024001123991595992, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9958227776812298, "standard_deviation": 0.000471574153405501, "minimum": 0.9951774258164051, "maximum": 0.996610388696766, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.312557889626477, "standard_deviation": 0.04229018626664111, "minimum": 1.2475795984982372, "maximum": 1.3589702174029854, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3498097187965737, "standard_deviation": 0.1509864618635598, "minimum": 0.13452645849042313, "maximum": 0.5938581941758162, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06669672425277362, "standard_deviation": 0.00383868791381164, "minimum": 0.05994083514185518, "maximum": 0.07153756648662991, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987402487810279, "standard_deviation": 0.00016029493403131362, "minimum": 0.998508217409831, "maximum": 0.9989980047845055, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7359437562549543, "standard_deviation": 0.03379033042047688, "minimum": 0.6939264780490423, "maximum": 0.7790751956618713, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.39487365463577545, "standard_deviation": 0.17036086051251378, "minimum": 0.17112725226788453, "maximum": 0.6453821543558715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.036135976221888494, "standard_deviation": 0.002638521160721762, "minimum": 0.031734588411744154, "maximum": 0.03946027004035343, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0010", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.9525399208068848, "predicted_mos": 2.9525399208068848, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1ae12bcfbc645f8ed1c399f46f538a19faf30e8e4d9640c9a4542e037a9f43ef", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.174788900185376, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2305636811357958, "全帯域_10ms_rms_envelope_nrmse": 0.07286801385188169, "全帯域_10ms_envelope_correlation": 0.9949853277655379, "全帯域_10ms_modulation_mae_db": 0.5884723255183258, "全帯域_40ms_log_rms_mae_db": 0.6879817023397672, "全帯域_40ms_rms_envelope_nrmse": 0.03879007370776727, "全帯域_40ms_envelope_correlation": 0.9985036512612364, "全帯域_40ms_modulation_mae_db": 0.7594375715008096, "300–1000 Hz_10ms_log_rms_mae_db": 1.7337331343758022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10435047431169185, "300–1000 Hz_10ms_envelope_correlation": 0.9918208536016552, "300–1000 Hz_10ms_modulation_mae_db": 0.7777943014396693, "300–1000 Hz_40ms_log_rms_mae_db": 1.054473809941377, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060174398165052545, "300–1000 Hz_40ms_envelope_correlation": 0.9971563186639884, "300–1000 Hz_40ms_modulation_mae_db": 1.1394618703148993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8233878289085101, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1520639673132731, "1000–3000 Hz_10ms_envelope_correlation": 0.9818149074772348, "1000–3000 Hz_10ms_modulation_mae_db": 0.5349284138733026, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1282603540586116, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08260887595185698, "1000–3000 Hz_40ms_envelope_correlation": 0.99451916622058, "1000–3000 Hz_40ms_modulation_mae_db": 0.7843681457998987, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1390925208713387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15636746738092855, "3000–6000 Hz_10ms_envelope_correlation": 0.9846902498550602, "3000–6000 Hz_10ms_modulation_mae_db": 0.3911693492181019, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6066510023137803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10275668089445647, "3000–6000 Hz_40ms_envelope_correlation": 0.9931591802624897, "3000–6000 Hz_40ms_modulation_mae_db": 0.280651116707249, "mr_stft_db": 5.775196352641436, "band_mae_db": 0.08507925105707669, "rms_gain_error_db": -0.051203434757911226, "output_peak": 0.42881423954006015}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2125636087148781, "全帯域_10ms_rms_envelope_nrmse": 0.06425737198197665, "全帯域_10ms_envelope_correlation": 0.9961149742459001, "全帯域_10ms_modulation_mae_db": 0.05071634150951748, "全帯域_40ms_log_rms_mae_db": 0.6236652547098412, "全帯域_40ms_rms_envelope_nrmse": 0.03290119539175529, "全帯域_40ms_envelope_correlation": 0.9989400465009338, "全帯域_40ms_modulation_mae_db": 0.03941459776385379, "300–1000 Hz_10ms_log_rms_mae_db": 1.7686579865836152, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.101254117922334, "300–1000 Hz_10ms_envelope_correlation": 0.9924902002440716, "300–1000 Hz_10ms_modulation_mae_db": 0.5129937731368288, "300–1000 Hz_40ms_log_rms_mae_db": 1.0714396605464198, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054113066486540275, "300–1000 Hz_40ms_envelope_correlation": 0.9979892212066197, "300–1000 Hz_40ms_modulation_mae_db": 0.684297441670592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.800914649217314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1457465596273689, "1000–3000 Hz_10ms_envelope_correlation": 0.9833722444980878, "1000–3000 Hz_10ms_modulation_mae_db": 0.5181374373810157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1172199496670299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07589658594726255, "1000–3000 Hz_40ms_envelope_correlation": 0.9955279809149189, "1000–3000 Hz_40ms_modulation_mae_db": 0.7623362910536348, "3000–6000 Hz_10ms_log_rms_mae_db": 2.142444713656583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1821522117810315, "3000–6000 Hz_10ms_envelope_correlation": 0.9797449481469339, "3000–6000 Hz_10ms_modulation_mae_db": 0.8228307644949346, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6048105137572937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12103547871855248, "3000–6000 Hz_40ms_envelope_correlation": 0.9916502830188723, "3000–6000 Hz_40ms_modulation_mae_db": 0.8961152704155556, "mr_stft_db": 5.762990407025555, "band_mae_db": 0.1668542867154635, "rms_gain_error_db": -0.05529280010010908, "output_peak": 0.453191350100597}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2733455212888973, "全帯域_10ms_rms_envelope_nrmse": 0.06783567310710274, "全帯域_10ms_envelope_correlation": 0.9956505905571162, "全帯域_10ms_modulation_mae_db": 0.2645657682394982, "全帯域_40ms_log_rms_mae_db": 0.6772768844865845, "全帯域_40ms_rms_envelope_nrmse": 0.036642720562528056, "全帯域_40ms_envelope_correlation": 0.9986546726788634, "全帯域_40ms_modulation_mae_db": 0.15014504164241116, "300–1000 Hz_10ms_log_rms_mae_db": 1.7357263319847078, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09387945686208814, "300–1000 Hz_10ms_envelope_correlation": 0.9935223026189101, "300–1000 Hz_10ms_modulation_mae_db": 0.38557622641477735, "300–1000 Hz_40ms_log_rms_mae_db": 1.0660013684107599, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053778379304276265, "300–1000 Hz_40ms_envelope_correlation": 0.9979329811699579, "300–1000 Hz_40ms_modulation_mae_db": 0.49562440749119396, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8043179188979523, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14124915940607277, "1000–3000 Hz_10ms_envelope_correlation": 0.9845201911075845, "1000–3000 Hz_10ms_modulation_mae_db": 0.6105957339120592, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0452338921773399, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0756916435899609, "1000–3000 Hz_40ms_envelope_correlation": 0.9957099741940979, "1000–3000 Hz_40ms_modulation_mae_db": 0.8689792910775523, "3000–6000 Hz_10ms_log_rms_mae_db": 2.176299611525295, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19539339566265082, "3000–6000 Hz_10ms_envelope_correlation": 0.9791097806704395, "3000–6000 Hz_10ms_modulation_mae_db": 0.6172245288835506, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6109509511906537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1285414050746472, "3000–6000 Hz_40ms_envelope_correlation": 0.9915940375428786, "3000–6000 Hz_40ms_modulation_mae_db": 0.5985520563372723, "mr_stft_db": 5.768456595520489, "band_mae_db": 0.15042900161110262, "rms_gain_error_db": -0.037485254590043926, "output_peak": 0.4457084665756507}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.225817863307519, "全帯域_10ms_rms_envelope_nrmse": 0.07197988059983576, "全帯域_10ms_envelope_correlation": 0.9951513710567241, "全帯域_10ms_modulation_mae_db": 0.24204195342812695, "全帯域_40ms_log_rms_mae_db": 0.6927497433498713, "全帯域_40ms_rms_envelope_nrmse": 0.03672941267070424, "全帯域_40ms_envelope_correlation": 0.9987284695404537, "全帯域_40ms_modulation_mae_db": 0.43768602165729015, "300–1000 Hz_10ms_log_rms_mae_db": 1.8466433978798167, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10706343862486496, "300–1000 Hz_10ms_envelope_correlation": 0.9916500802209147, "300–1000 Hz_10ms_modulation_mae_db": 1.043591461392043, "300–1000 Hz_40ms_log_rms_mae_db": 1.1880528787184859, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058308321024071894, "300–1000 Hz_40ms_envelope_correlation": 0.9977233080107273, "300–1000 Hz_40ms_modulation_mae_db": 1.184472779983979, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8426058815402233, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1552974564951949, "1000–3000 Hz_10ms_envelope_correlation": 0.9817409359165064, "1000–3000 Hz_10ms_modulation_mae_db": 0.603766260228214, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1463834346672257, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08812284127231976, "1000–3000 Hz_40ms_envelope_correlation": 0.9948349581867202, "1000–3000 Hz_40ms_modulation_mae_db": 0.8066048867006049, "3000–6000 Hz_10ms_log_rms_mae_db": 2.177246142284872, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.162961699003679, "3000–6000 Hz_10ms_envelope_correlation": 0.9833946811706634, "3000–6000 Hz_10ms_modulation_mae_db": 0.45362384500918673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6700518202119277, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10878986172363532, "3000–6000 Hz_40ms_envelope_correlation": 0.9925535346414431, "3000–6000 Hz_40ms_modulation_mae_db": 0.4474435642094671, "mr_stft_db": 5.762039473213616, "band_mae_db": 0.23089161370752534, "rms_gain_error_db": -0.08840481004725806, "output_peak": 0.43236690032950825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.272958204280109, "全帯域_10ms_rms_envelope_nrmse": 0.07144846683318465, "全帯域_10ms_envelope_correlation": 0.9951980648761932, "全帯域_10ms_modulation_mae_db": 0.5300042177212126, "全帯域_40ms_log_rms_mae_db": 0.6465675724654278, "全帯域_40ms_rms_envelope_nrmse": 0.03683747334078786, "全帯域_40ms_envelope_correlation": 0.9986823100858103, "全帯域_40ms_modulation_mae_db": 0.5225184025612738, "300–1000 Hz_10ms_log_rms_mae_db": 1.8195085622102896, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10236789137673827, "300–1000 Hz_10ms_envelope_correlation": 0.9923098211919794, "300–1000 Hz_10ms_modulation_mae_db": 1.051809917191245, "300–1000 Hz_40ms_log_rms_mae_db": 1.1064887488904862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056628608212642935, "300–1000 Hz_40ms_envelope_correlation": 0.997753571305767, "300–1000 Hz_40ms_modulation_mae_db": 1.3558322665209586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8561513861180055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1534291600836288, "1000–3000 Hz_10ms_envelope_correlation": 0.9821602249869218, "1000–3000 Hz_10ms_modulation_mae_db": 0.4467188577918479, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1688686277666855, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09347408616004582, "1000–3000 Hz_40ms_envelope_correlation": 0.9938929393627728, "1000–3000 Hz_40ms_modulation_mae_db": 0.7040930269419788, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1361756417463487, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17513873387285558, "3000–6000 Hz_10ms_envelope_correlation": 0.9822588788925426, "3000–6000 Hz_10ms_modulation_mae_db": 0.6632611345309165, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6208577884560156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1206991319132098, "3000–6000 Hz_40ms_envelope_correlation": 0.991632915862091, "3000–6000 Hz_40ms_modulation_mae_db": 0.5222620414918588, "mr_stft_db": 5.763780812996777, "band_mae_db": 0.1863156830362115, "rms_gain_error_db": -0.06959980355207938, "output_peak": 0.4150380811226993}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2229844883624401, "全帯域_10ms_rms_envelope_nrmse": 0.06245524288622883, "全帯域_10ms_envelope_correlation": 0.9963258405918938, "全帯域_10ms_modulation_mae_db": 0.39368985866045164, "全帯域_40ms_log_rms_mae_db": 0.6536018926586838, "全帯域_40ms_rms_envelope_nrmse": 0.03137197593573428, "全帯域_40ms_envelope_correlation": 0.9990337520111472, "全帯域_40ms_modulation_mae_db": 0.4889972318807801, "300–1000 Hz_10ms_log_rms_mae_db": 1.8239772523549331, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09069861050095518, "300–1000 Hz_10ms_envelope_correlation": 0.9940827208131654, "300–1000 Hz_10ms_modulation_mae_db": 0.7952645290933965, "300–1000 Hz_40ms_log_rms_mae_db": 1.1384727159017398, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051775226331150025, "300–1000 Hz_40ms_envelope_correlation": 0.998237925241372, "300–1000 Hz_40ms_modulation_mae_db": 0.9581695747331836, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9107015849324438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15665027427252048, "1000–3000 Hz_10ms_envelope_correlation": 0.9806658487847307, "1000–3000 Hz_10ms_modulation_mae_db": 0.32533817467170234, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2413105713929806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08893582372406915, "1000–3000 Hz_40ms_envelope_correlation": 0.993581558899557, "1000–3000 Hz_40ms_modulation_mae_db": 0.3249688197399093, "3000–6000 Hz_10ms_log_rms_mae_db": 2.153134583658536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1665279108979692, "3000–6000 Hz_10ms_envelope_correlation": 0.984030368178344, "3000–6000 Hz_10ms_modulation_mae_db": 0.6822041831128788, "3000–6000 Hz_40ms_log_rms_mae_db": 1.623359008144845, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11401073554802447, "3000–6000 Hz_40ms_envelope_correlation": 0.9926253236039002, "3000–6000 Hz_40ms_modulation_mae_db": 0.6557823457624358, "mr_stft_db": 5.783081515402409, "band_mae_db": 0.1364368682902741, "rms_gain_error_db": -0.04976202997152927, "output_peak": 0.43448705029262963}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2610098951979574, "全帯域_10ms_rms_envelope_nrmse": 0.07699942595875807, "全帯域_10ms_envelope_correlation": 0.9944223345594007, "全帯域_10ms_modulation_mae_db": 0.13604170641170563, "全帯域_40ms_log_rms_mae_db": 0.6676722027185942, "全帯域_40ms_rms_envelope_nrmse": 0.040621405357506066, "全帯域_40ms_envelope_correlation": 0.9984107701831202, "全帯域_40ms_modulation_mae_db": 0.15311004235855938, "300–1000 Hz_10ms_log_rms_mae_db": 1.7808206390730241, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11524897070737986, "300–1000 Hz_10ms_envelope_correlation": 0.9903075840512798, "300–1000 Hz_10ms_modulation_mae_db": 0.637547902226892, "300–1000 Hz_40ms_log_rms_mae_db": 1.130945745554102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06587695974328565, "300–1000 Hz_40ms_envelope_correlation": 0.9970374399127865, "300–1000 Hz_40ms_modulation_mae_db": 1.0426471802345885, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8331772742431511, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15261743738029132, "1000–3000 Hz_10ms_envelope_correlation": 0.9818679277459977, "1000–3000 Hz_10ms_modulation_mae_db": 0.2954117347277166, "1000–3000 Hz_40ms_log_rms_mae_db": 1.132303149837951, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08506715203421669, "1000–3000 Hz_40ms_envelope_correlation": 0.9945156578787658, "1000–3000 Hz_40ms_modulation_mae_db": 0.5615217730303831, "3000–6000 Hz_10ms_log_rms_mae_db": 2.122402785441978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16819416630865638, "3000–6000 Hz_10ms_envelope_correlation": 0.9824727125287491, "3000–6000 Hz_10ms_modulation_mae_db": 0.7603666382030339, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5685727953291855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11205689951775542, "3000–6000 Hz_40ms_envelope_correlation": 0.991925745604672, "3000–6000 Hz_40ms_modulation_mae_db": 0.8984413803216952, "mr_stft_db": 5.7638917982895235, "band_mae_db": 0.15294974438315115, "rms_gain_error_db": -0.07379467205236696, "output_peak": 0.4490595842825751}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2296265068177143, "全帯域_10ms_rms_envelope_nrmse": 0.06959464353918905, "全帯域_10ms_envelope_correlation": 0.9954887858549134, "全帯域_10ms_modulation_mae_db": 0.6859189835245155, "全帯域_40ms_log_rms_mae_db": 0.6745738825794545, "全帯域_40ms_rms_envelope_nrmse": 0.038614202779738004, "全帯域_40ms_envelope_correlation": 0.9986005426797889, "全帯域_40ms_modulation_mae_db": 1.010268928797046, "300–1000 Hz_10ms_log_rms_mae_db": 1.685041517339902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09522101354313665, "300–1000 Hz_10ms_envelope_correlation": 0.9934177114821972, "300–1000 Hz_10ms_modulation_mae_db": 1.0532442050831607, "300–1000 Hz_40ms_log_rms_mae_db": 1.047026438284755, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05543693763041588, "300–1000 Hz_40ms_envelope_correlation": 0.9979042083011421, "300–1000 Hz_40ms_modulation_mae_db": 1.3197401497302763, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8377821792381739, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14831814712933916, "1000–3000 Hz_10ms_envelope_correlation": 0.9840143908521303, "1000–3000 Hz_10ms_modulation_mae_db": 0.7250634644982549, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0985457090620099, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08520212522534817, "1000–3000 Hz_40ms_envelope_correlation": 0.9959746033557594, "1000–3000 Hz_40ms_modulation_mae_db": 0.8436345165559738, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1487944888056654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16796549661839963, "3000–6000 Hz_10ms_envelope_correlation": 0.9830598005681742, "3000–6000 Hz_10ms_modulation_mae_db": 0.4286649311900223, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6092944829286722, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11772616903247742, "3000–6000 Hz_40ms_envelope_correlation": 0.9914925260046304, "3000–6000 Hz_40ms_modulation_mae_db": 0.3512364845814115, "mr_stft_db": 5.772232406445387, "band_mae_db": 0.1911055711347879, "rms_gain_error_db": -0.09339075955828333, "output_peak": 0.4374796243490766}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.992023528168301, "spectral_convergence": 0.21671429975007575, "spectral_convergence_status": "ok", "amplitude_floor": 1.642409573690628e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.604492240127559, "spectral_convergence": 0.2076792737518895, "spectral_convergence_status": "ok", "amplitude_floor": 1.497573454010957e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.72907328962845, "spectral_convergence": 0.20310532390369915, "spectral_convergence_status": "ok", "amplitude_floor": 1.2731241516974033e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012546591027172457, "output_power": 0.0012590117574724593, "delta_db": 0.015040440346391935, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0029841932734943235, "output_power": 0.002923751340582295, "delta_db": -0.0888651316871662, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004877853564370709, "output_power": 0.0004936144578314487, "delta_db": 0.0515911314361038, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013442175178747268, "output_power": 0.00012644309851649347, "delta_db": -0.2657442105603295, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010124396480278944, "output_power": 0.00010173280263892155, "delta_db": 0.020918648760569897, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001060495609474016, "output_power": 0.00010439441873488813, "delta_db": -0.06831594355189873, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.08507925105707669, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.051203434757911226, "pitch": null, "output_rms_dbfs": -23.002534876517554, "output_peak": 0.42881423954006015, "reference_rms_dbfs": -22.951331441759642, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.775196352641436}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024147266762860616, "log_rms_mae_db": 1.2305636811357958, "rms_envelope_nrmse": 0.07286801385188169, "envelope_correlation": 0.9949853277655379, "rms_gain_error_db": -0.051202292144496835, "reference_mean_power": 0.005072567967533502, "output_mean_power": 0.005013114749398805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4164286663529534e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010865583816620668, "output_power": 0.0010876030906114221, "reference_active": true, "delta_db": 0.004173667619496613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020269542962956985, "output_power": 0.00018642564940778372, "reference_active": true, "delta_db": -0.3633829172828206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.69603101954121e-05, "output_power": 1.954064278155301e-05, "reference_active": true, "delta_db": -1.3978603916526602}], "active_band_count": 3, "active_band_mae_db": 0.5884723255183258}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000225157536036158, "log_rms_mae_db": 0.6879817023397672, "rms_envelope_nrmse": 0.03879007370776727, "envelope_correlation": 0.9985036512612364, "rms_gain_error_db": -0.051192558065739274, "reference_mean_power": 0.005098028148503288, "output_mean_power": 0.005038287815963855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2603245746946083e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010232290392639372, "output_power": 0.0010201925774826468, "reference_active": true, "delta_db": -0.012906974810693705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010544090587602268, "output_power": 9.731104395510753e-05, "reference_active": true, "delta_db": -0.348469967237135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0049541105469593e-06, "output_power": 1.2894688631566288e-06, "reference_active": true, "delta_db": -1.9169357724546003}], "active_band_count": 3, "active_band_mae_db": 0.7594375715008096}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020992712290116374, "log_rms_mae_db": 1.7337331343758022, "rms_envelope_nrmse": 0.10435047431169185, "envelope_correlation": 0.9918208536016552, "rms_gain_error_db": -0.067403841987374, "reference_mean_power": 0.0023388664575994473, "output_mean_power": 0.0023028467707448905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4073975659228857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007410848450397675, "output_power": 0.0007460433508134472, "reference_active": true, "delta_db": 0.028961318632022486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011531333502346333, "output_power": 0.00010356778079929731, "reference_active": true, "delta_db": -0.46654862233367006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6543285086796136e-05, "output_power": 1.7384706989847787e-05, "reference_active": true, "delta_db": -1.8378729633533153}], "active_band_count": 3, "active_band_mae_db": 0.7777943014396693}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002036893389343094, "log_rms_mae_db": 1.054473809941377, "rms_envelope_nrmse": 0.060174398165052545, "envelope_correlation": 0.9971563186639884, "rms_gain_error_db": -0.06740525290008537, "reference_mean_power": 0.0023506239289763434, "output_mean_power": 0.0023144224194173646}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3312664615669961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215176046332657, "output_power": 0.0007207191868169557, "reference_active": true, "delta_db": -0.004808482065565082}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5162124576630795e-05, "output_power": 4.905053258339272e-05, "reference_active": true, "delta_db": -0.5099725748968691}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1332065410684048e-06, "output_power": 1.0931316058524065e-06, "reference_active": true, "delta_db": -2.9036045539822637}], "active_band_count": 3, "active_band_mae_db": 1.1394618703148993}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.160365671559225e-05, "log_rms_mae_db": 1.8233878289085101, "rms_envelope_nrmse": 0.1520639673132731, "envelope_correlation": 0.9818149074772348, "rms_gain_error_db": -0.1766906281909747, "reference_mean_power": 0.00021265544148097967, "output_mean_power": 0.00020417729093394764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.181263558889223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.569393841260949e-05, "output_power": 5.4873844075428043e-05, "reference_active": true, "delta_db": -0.06442545355325477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4402699100691371e-05, "output_power": 1.2698427333972236e-05, "reference_active": true, "delta_db": -0.546939494756931}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.927958039351209e-06, "output_power": 2.329286015095422e-06, "reference_active": true, "delta_db": -0.9934202933097218}], "active_band_count": 3, "active_band_mae_db": 0.5349284138733026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.199537921297776e-05, "log_rms_mae_db": 1.1282603540586116, "rms_envelope_nrmse": 0.08260887595185698, "envelope_correlation": 0.99451916622058, "rms_gain_error_db": -0.1766956070712221, "reference_mean_power": 0.00021371687213348275, "output_mean_power": 0.00020519616920911047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0829999638899264e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5369039282181144e-05, "output_power": 5.4180022353797625e-05, "reference_active": true, "delta_db": -0.09427808178270998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.495407348960159e-06, "output_power": 5.770129083500209e-06, "reference_active": true, "delta_db": -0.5142086347678281}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.109761653644897e-07, "output_power": 1.4117949444525268e-07, "reference_active": true, "delta_db": -1.7446177208491578}], "active_band_count": 3, "active_band_mae_db": 0.7843681457998987}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3802525725767476e-05, "log_rms_mae_db": 2.1390925208713387, "rms_envelope_nrmse": 0.15636746738092855, "envelope_correlation": 0.9846902498550602, "rms_gain_error_db": -0.03081772894280242, "reference_mean_power": 6.940550281225073e-05, "output_mean_power": 6.891474157884233e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.994304097313729e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.68316376202847e-05, "output_power": 3.482329645290637e-05, "reference_active": true, "delta_db": -0.24351148990646437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3496621932801275e-06, "output_power": 4.891695876913687e-06, "reference_active": true, "delta_db": -0.38866910623683715}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1665818120524947e-06, "output_power": 1.3214415138467419e-06, "reference_active": true, "delta_db": 0.5413274515110044}], "active_band_count": 3, "active_band_mae_db": 0.3911693492181019}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.808436973199796e-05, "log_rms_mae_db": 1.6066510023137803, "rms_envelope_nrmse": 0.10275668089445647, "envelope_correlation": 0.9931591802624897, "rms_gain_error_db": -0.030824217865394885, "reference_mean_power": 6.97538862423984e-05, "output_mean_power": 6.926055813090919e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.7257017907786166e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.6073966740876416e-05, "output_power": 3.390575067472301e-05, "reference_active": true, "delta_db": -0.26920536419922264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3003381054168396e-06, "output_power": 2.182638167489817e-06, "reference_active": true, "delta_db": -0.2280992825660072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.80849015690717e-08, "output_power": 8.453410082759764e-08, "reference_active": true, "delta_db": 0.3446487033565172}], "active_band_count": 3, "active_band_mae_db": 0.280651116707249}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8488053622642499, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f57a7399db225bcbe6b338dfdcec77a899af81de0c19cc6d295f0e91d457d166", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9825195839211492, "standard_deviation": 0.0013124076626429488, "minimum": 0.9806658487847307, "maximum": 0.9845201911075845, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8386298378869719, "standard_deviation": 0.03459905314972552, "minimum": 1.800914649217314, "maximum": 1.9107015849324438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5074950096355142, "standard_deviation": 0.14636254750123423, "minimum": 0.2954117347277166, "maximum": 0.7250634644982549, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15067152021346117, "standard_deviation": 0.005196830787166582, "minimum": 0.14124915940607277, "maximum": 0.15665027427252048, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9948196048766464, "standard_deviation": 0.0008621340071101521, "minimum": 0.993581558899557, "maximum": 0.9959746033557594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1347657110787293, "standard_deviation": 0.056484966964178214, "minimum": 1.0452338921773399, "maximum": 1.2413105713929806, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.707063343862492, "standard_deviation": 0.18165725419243334, "minimum": 0.3249688197399093, "maximum": 0.8689792910775523, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.084374891738135, "standard_deviation": 0.006201467386519297, "minimum": 0.0756916435899609, "maximum": 0.09347408616004582, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9823451775013634, "standard_deviation": 0.001971369140747153, "minimum": 0.9791097806704395, "maximum": 0.9846902498550602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.149448810998827, "standard_deviation": 0.019167102837864294, "minimum": 2.122402785441978, "maximum": 2.177246142284872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6024181718303281, "standard_deviation": 0.16070040284978931, "minimum": 0.3911693492181019, "maximum": 0.8228307644949346, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17183763519077133, "standard_deviation": 0.01222837113378997, "minimum": 0.15636746738092855, "maximum": 0.19539339566265082, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920791933176222, "standard_deviation": 0.0006182762046156437, "minimum": 0.9914925260046304, "maximum": 0.9931591802624897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6143185452915467, "standard_deviation": 0.028053874227895927, "minimum": 1.5685727953291855, "maximum": 1.6700518202119277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5813105324783682, "standard_deviation": 0.2299858877934953, "minimum": 0.280651116707249, "maximum": 0.8984413803216952, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11570204530284484, "standard_deviation": 0.00805224884997869, "minimum": 0.10275668089445647, "maximum": 0.1285414050746472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9924501592780217, "standard_deviation": 0.0012188392938001825, "minimum": 0.9903075840512798, "maximum": 0.9940827208131654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7742636027252614, "standard_deviation": 0.05466812383331408, "minimum": 1.685041517339902, "maximum": 1.8466433978798167, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7822277894972516, "standard_deviation": 0.25784178158401594, "minimum": 0.38557622641477735, "maximum": 1.0532442050831607, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1012604967311486, "standard_deviation": 0.00795194864986455, "minimum": 0.09069861050095518, "maximum": 0.11524897070737986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977168717265451, "standard_deviation": 0.00041485021370584916, "minimum": 0.9970374399127865, "maximum": 0.998237925241372, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1003626707810157, "standard_deviation": 0.049400090038185004, "minimum": 1.047026438284755, "maximum": 1.1880528787184859, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.022530708834959, "standard_deviation": 0.3014016144764775, "minimum": 0.49562440749119396, "maximum": 1.3558322665209586, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05701148711217943, "standard_deviation": 0.004459026911880926, "minimum": 0.051775226331150025, "maximum": 0.06587695974328565, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1625077524919491, "standard_deviation": 0.043081340345565385, "minimum": 0.08507925105707669, "maximum": 0.23089161370752534, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7689586701919, "standard_deviation": 0.007419200142386694, "minimum": 5.762039473213616, "maximum": 5.783081515402409, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4370181620740996, "standard_deviation": 0.012310107726961256, "minimum": 0.4150380811226993, "maximum": 0.453191350100597, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06486669557869766, "standard_deviation": 0.01973399727178567, "minimum": -0.09339075955828333, "maximum": -0.037485254590043926, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954171611884599, "standard_deviation": 0.0006172598896470633, "minimum": 0.9944223345594007, "maximum": 0.9963258405918938, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.241108721138164, "standard_deviation": 0.024112291749353226, "minimum": 1.2125636087148781, "maximum": 1.2733455212888973, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36143139437666927, "standard_deviation": 0.22591981158590727, "minimum": 0.05071634150951748, "maximum": 0.6859189835245155, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06967983984476968, "standard_deviation": 0.0047383721071695035, "minimum": 0.06245524288622883, "maximum": 0.07699942595875807, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986942768676692, "standard_deviation": 0.00020840454484179547, "minimum": 0.9984107701831202, "maximum": 0.9990337520111472, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.665511141913528, "standard_deviation": 0.023064725577217223, "minimum": 0.6236652547098412, "maximum": 0.6927497433498713, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.445197229770253, "standard_deviation": 0.33015942097570355, "minimum": 0.03941459776385379, "maximum": 1.010268928797046, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03656355746831513, "standard_deviation": 0.003072989928106587, "minimum": 0.03137197593573428, "maximum": 0.040621405357506066, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0010", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.0714945793151855, "predicted_mos": 3.0714945793151855, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f43c84ddae9e2580d83bb26f9eceafb4e7b8e93e064169ce2a77aba72657e74c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.807443799916655, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2443431916223828, "全帯域_10ms_rms_envelope_nrmse": 0.06791680631018246, "全帯域_10ms_envelope_correlation": 0.9957364097512539, "全帯域_10ms_modulation_mae_db": 0.31704285514434727, "全帯域_40ms_log_rms_mae_db": 0.66400701591665, "全帯域_40ms_rms_envelope_nrmse": 0.0383344136941019, "全帯域_40ms_envelope_correlation": 0.9986508652024862, "全帯域_40ms_modulation_mae_db": 0.5207751295860864, "300–1000 Hz_10ms_log_rms_mae_db": 1.846818893282088, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09833149681087738, "300–1000 Hz_10ms_envelope_correlation": 0.9929917773928646, "300–1000 Hz_10ms_modulation_mae_db": 0.8197537262540849, "300–1000 Hz_40ms_log_rms_mae_db": 1.1093199837797254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05854561329767458, "300–1000 Hz_40ms_envelope_correlation": 0.9976586409004697, "300–1000 Hz_40ms_modulation_mae_db": 0.9636315306186481, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7661181873364702, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.152608015052407, "1000–3000 Hz_10ms_envelope_correlation": 0.9823343603436796, "1000–3000 Hz_10ms_modulation_mae_db": 0.5337038963576032, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0462820066757497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08437906120173623, "1000–3000 Hz_40ms_envelope_correlation": 0.9952771473235219, "1000–3000 Hz_40ms_modulation_mae_db": 0.8795415241425227, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0822886281733206, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1725431550796262, "3000–6000 Hz_10ms_envelope_correlation": 0.9812708103577101, "3000–6000 Hz_10ms_modulation_mae_db": 0.612810692285295, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5117699249867824, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10517040256183016, "3000–6000 Hz_40ms_envelope_correlation": 0.9927946123815644, "3000–6000 Hz_40ms_modulation_mae_db": 0.4084744015722883, "mr_stft_db": 5.727103763289672, "band_mae_db": 0.22405929839119998, "rms_gain_error_db": -0.10693338969832336, "output_peak": 0.44591588989885167}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2439827129651446, "全帯域_10ms_rms_envelope_nrmse": 0.06959872920982682, "全帯域_10ms_envelope_correlation": 0.9956659562163424, "全帯域_10ms_modulation_mae_db": 0.3095430749548909, "全帯域_40ms_log_rms_mae_db": 0.6748486063467123, "全帯域_40ms_rms_envelope_nrmse": 0.04057341057014721, "全帯域_40ms_envelope_correlation": 0.998661003009754, "全帯域_40ms_modulation_mae_db": 0.3956301726146725, "300–1000 Hz_10ms_log_rms_mae_db": 1.7647934583242366, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10167726592724033, "300–1000 Hz_10ms_envelope_correlation": 0.9928836803540525, "300–1000 Hz_10ms_modulation_mae_db": 0.5166969020747666, "300–1000 Hz_40ms_log_rms_mae_db": 1.0561328717645122, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0617563847603116, "300–1000 Hz_40ms_envelope_correlation": 0.9978283473155936, "300–1000 Hz_40ms_modulation_mae_db": 0.601604030308189, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7864186019328574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1449423602574755, "1000–3000 Hz_10ms_envelope_correlation": 0.9847114801590382, "1000–3000 Hz_10ms_modulation_mae_db": 0.7214030462941962, "1000–3000 Hz_40ms_log_rms_mae_db": 1.090570561566992, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09349136427179074, "1000–3000 Hz_40ms_envelope_correlation": 0.9944785901749696, "1000–3000 Hz_40ms_modulation_mae_db": 0.7900465618234934, "3000–6000 Hz_10ms_log_rms_mae_db": 2.09063361840356, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16458610800630535, "3000–6000 Hz_10ms_envelope_correlation": 0.9833712447650501, "3000–6000 Hz_10ms_modulation_mae_db": 0.6039615888835375, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5377506152597087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11602395135787612, "3000–6000 Hz_40ms_envelope_correlation": 0.991952478301892, "3000–6000 Hz_40ms_modulation_mae_db": 0.8740815432951249, "mr_stft_db": 5.708756414362326, "band_mae_db": 0.24931694447326183, "rms_gain_error_db": -0.15647531315293106, "output_peak": 0.4467445099481545}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2227652832740905, "全帯域_10ms_rms_envelope_nrmse": 0.06526895337966059, "全帯域_10ms_envelope_correlation": 0.9959923938900881, "全帯域_10ms_modulation_mae_db": 0.11163830959769973, "全帯域_40ms_log_rms_mae_db": 0.6463883899420266, "全帯域_40ms_rms_envelope_nrmse": 0.03533632395112308, "全帯域_40ms_envelope_correlation": 0.99877652032373, "全帯域_40ms_modulation_mae_db": 0.15935131343677705, "300–1000 Hz_10ms_log_rms_mae_db": 1.6785722594714017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09867064853167835, "300–1000 Hz_10ms_envelope_correlation": 0.99296183819341, "300–1000 Hz_10ms_modulation_mae_db": 0.48706410576528253, "300–1000 Hz_40ms_log_rms_mae_db": 0.9661012212775478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05811386324466805, "300–1000 Hz_40ms_envelope_correlation": 0.9977209520050723, "300–1000 Hz_40ms_modulation_mae_db": 0.4255862586260298, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7275859523993067, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13652860966552985, "1000–3000 Hz_10ms_envelope_correlation": 0.9860023619159637, "1000–3000 Hz_10ms_modulation_mae_db": 0.5103849301757962, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0051813695634864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07943922173328966, "1000–3000 Hz_40ms_envelope_correlation": 0.9957744859873725, "1000–3000 Hz_40ms_modulation_mae_db": 1.0103199430463503, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0204431049635008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16662452494045363, "3000–6000 Hz_10ms_envelope_correlation": 0.9826123497668731, "3000–6000 Hz_10ms_modulation_mae_db": 0.37637041956600675, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4646770475395074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.106617520055273, "3000–6000 Hz_40ms_envelope_correlation": 0.9925910266341386, "3000–6000 Hz_40ms_modulation_mae_db": 0.38904670042257355, "mr_stft_db": 5.705776817364349, "band_mae_db": 0.17573528084326914, "rms_gain_error_db": -0.06000977274620788, "output_peak": 0.46357505028828194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.236089330397946, "全帯域_10ms_rms_envelope_nrmse": 0.06729463679270903, "全帯域_10ms_envelope_correlation": 0.9957336518863329, "全帯域_10ms_modulation_mae_db": 0.14248031098141897, "全帯域_40ms_log_rms_mae_db": 0.636079344615044, "全帯域_40ms_rms_envelope_nrmse": 0.03632166831130006, "全帯域_40ms_envelope_correlation": 0.9987005505769464, "全帯域_40ms_modulation_mae_db": 0.14090076574232752, "300–1000 Hz_10ms_log_rms_mae_db": 1.7392361444467614, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09506638473132771, "300–1000 Hz_10ms_envelope_correlation": 0.9934003026261367, "300–1000 Hz_10ms_modulation_mae_db": 0.49665108978916633, "300–1000 Hz_40ms_log_rms_mae_db": 1.0155355687098893, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05337090414116327, "300–1000 Hz_40ms_envelope_correlation": 0.9980283821079623, "300–1000 Hz_40ms_modulation_mae_db": 0.6710929718224073, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7846452605360699, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15685865250527442, "1000–3000 Hz_10ms_envelope_correlation": 0.981719717412791, "1000–3000 Hz_10ms_modulation_mae_db": 0.6206439088697975, "1000–3000 Hz_40ms_log_rms_mae_db": 1.102452620440518, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09871190286757471, "1000–3000 Hz_40ms_envelope_correlation": 0.9935685042373201, "1000–3000 Hz_40ms_modulation_mae_db": 1.0066639542542983, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0665739732994695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16025026124068278, "3000–6000 Hz_10ms_envelope_correlation": 0.9846975415833529, "3000–6000 Hz_10ms_modulation_mae_db": 0.19130329166736781, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5268918857398084, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10847042271426569, "3000–6000 Hz_40ms_envelope_correlation": 0.99290279464136, "3000–6000 Hz_40ms_modulation_mae_db": 0.13708651502454672, "mr_stft_db": 5.71603236906704, "band_mae_db": 0.2103144520839051, "rms_gain_error_db": -0.05731830377764587, "output_peak": 0.43576640147611606}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2254163538111327, "全帯域_10ms_rms_envelope_nrmse": 0.07290511555304789, "全帯域_10ms_envelope_correlation": 0.9950372665813586, "全帯域_10ms_modulation_mae_db": 0.3100070455567771, "全帯域_40ms_log_rms_mae_db": 0.6665114299114355, "全帯域_40ms_rms_envelope_nrmse": 0.040689001539870435, "全帯域_40ms_envelope_correlation": 0.9984334669585961, "全帯域_40ms_modulation_mae_db": 0.4068836773935754, "300–1000 Hz_10ms_log_rms_mae_db": 1.7461286050046605, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10182017441779229, "300–1000 Hz_10ms_envelope_correlation": 0.9925211629801152, "300–1000 Hz_10ms_modulation_mae_db": 1.2341993622634326, "300–1000 Hz_40ms_log_rms_mae_db": 1.029608115814994, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056953332413934604, "300–1000 Hz_40ms_envelope_correlation": 0.9978800957160378, "300–1000 Hz_40ms_modulation_mae_db": 1.3990972680437492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8094529105088302, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15675172180378769, "1000–3000 Hz_10ms_envelope_correlation": 0.9807393181182938, "1000–3000 Hz_10ms_modulation_mae_db": 0.9304336261370132, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1032501231973675, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09307418143263915, "1000–3000 Hz_40ms_envelope_correlation": 0.9931294695141081, "1000–3000 Hz_40ms_modulation_mae_db": 0.5758369118598334, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0578114238872147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1938843420146426, "3000–6000 Hz_10ms_envelope_correlation": 0.9798241271127731, "3000–6000 Hz_10ms_modulation_mae_db": 0.724514892678819, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4961988489545377, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12956327704770149, "3000–6000 Hz_40ms_envelope_correlation": 0.9917609786863106, "3000–6000 Hz_40ms_modulation_mae_db": 0.6461991622381921, "mr_stft_db": 5.726721374140008, "band_mae_db": 0.16894108023401325, "rms_gain_error_db": -0.09561141681589654, "output_peak": 0.44901610916083573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2803015681718135, "全帯域_10ms_rms_envelope_nrmse": 0.07208588495836475, "全帯域_10ms_envelope_correlation": 0.9951008609880588, "全帯域_10ms_modulation_mae_db": 0.20688577984247392, "全帯域_40ms_log_rms_mae_db": 0.7028483044398985, "全帯域_40ms_rms_envelope_nrmse": 0.03839118468445669, "全帯域_40ms_envelope_correlation": 0.9985398824305565, "全帯域_40ms_modulation_mae_db": 0.07575941912364755, "300–1000 Hz_10ms_log_rms_mae_db": 1.8227204954148932, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1051137905594386, "300–1000 Hz_10ms_envelope_correlation": 0.991920367969969, "300–1000 Hz_10ms_modulation_mae_db": 0.5442189354084458, "300–1000 Hz_40ms_log_rms_mae_db": 1.0983675386458636, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060977282277808194, "300–1000 Hz_40ms_envelope_correlation": 0.9974056879471441, "300–1000 Hz_40ms_modulation_mae_db": 0.7331186371094148, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7879166943110896, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15183130716016865, "1000–3000 Hz_10ms_envelope_correlation": 0.9818648665091447, "1000–3000 Hz_10ms_modulation_mae_db": 0.47161124062589455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0794255507364536, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08438024337966342, "1000–3000 Hz_40ms_envelope_correlation": 0.99426595883745, "1000–3000 Hz_40ms_modulation_mae_db": 0.4693239395257991, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0855207758579635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1728124852344777, "3000–6000 Hz_10ms_envelope_correlation": 0.9812524539840237, "3000–6000 Hz_10ms_modulation_mae_db": 0.5784008732429374, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4779839215750106, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10266604078564648, "3000–6000 Hz_40ms_envelope_correlation": 0.9931738921308384, "3000–6000 Hz_40ms_modulation_mae_db": 0.5425223324798066, "mr_stft_db": 5.720992837593282, "band_mae_db": 0.13898308233656206, "rms_gain_error_db": -0.051249125601552585, "output_peak": 0.4668429524383776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2369512618866811, "全帯域_10ms_rms_envelope_nrmse": 0.06483498798796565, "全帯域_10ms_envelope_correlation": 0.9960519956679199, "全帯域_10ms_modulation_mae_db": 0.17831435472117385, "全帯域_40ms_log_rms_mae_db": 0.6828772100359443, "全帯域_40ms_rms_envelope_nrmse": 0.0347259312816787, "全帯域_40ms_envelope_correlation": 0.9988283210974175, "全帯域_40ms_modulation_mae_db": 0.39383175592524844, "300–1000 Hz_10ms_log_rms_mae_db": 1.749334994752135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08979621802609368, "300–1000 Hz_10ms_envelope_correlation": 0.9940271194386642, "300–1000 Hz_10ms_modulation_mae_db": 0.7765680988976621, "300–1000 Hz_40ms_log_rms_mae_db": 1.0009548501322916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04757742024324632, "300–1000 Hz_40ms_envelope_correlation": 0.9983462344734854, "300–1000 Hz_40ms_modulation_mae_db": 1.056098276093982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7512672188058016, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13638328338299396, "1000–3000 Hz_10ms_envelope_correlation": 0.9855942696286285, "1000–3000 Hz_10ms_modulation_mae_db": 0.5800660140198873, "1000–3000 Hz_40ms_log_rms_mae_db": 1.040594576702203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07572687994047345, "1000–3000 Hz_40ms_envelope_correlation": 0.9956813304169789, "1000–3000 Hz_40ms_modulation_mae_db": 0.9406848592990696, "3000–6000 Hz_10ms_log_rms_mae_db": 2.09268672423089, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19040513851439705, "3000–6000 Hz_10ms_envelope_correlation": 0.9787982441014587, "3000–6000 Hz_10ms_modulation_mae_db": 1.2393556714135834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5726733984468597, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12826437607174204, "3000–6000 Hz_40ms_envelope_correlation": 0.9903606530208693, "3000–6000 Hz_40ms_modulation_mae_db": 1.2360512477332037, "mr_stft_db": 5.721835169110939, "band_mae_db": 0.12902565855237305, "rms_gain_error_db": -0.06452572258305102, "output_peak": 0.4588841518895329}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2062085053538556, "全帯域_10ms_rms_envelope_nrmse": 0.06056326870170034, "全帯域_10ms_envelope_correlation": 0.9965433544552412, "全帯域_10ms_modulation_mae_db": 0.2906449707781304, "全帯域_40ms_log_rms_mae_db": 0.6582192656933793, "全帯域_40ms_rms_envelope_nrmse": 0.0341103334978251, "全帯域_40ms_envelope_correlation": 0.998837158467364, "全帯域_40ms_modulation_mae_db": 0.20470586155858275, "300–1000 Hz_10ms_log_rms_mae_db": 1.624782382929662, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08770041475717351, "300–1000 Hz_10ms_envelope_correlation": 0.9942600066269132, "300–1000 Hz_10ms_modulation_mae_db": 0.7286277965754077, "300–1000 Hz_40ms_log_rms_mae_db": 0.9257419726584354, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05053642834018606, "300–1000 Hz_40ms_envelope_correlation": 0.9980466717704695, "300–1000 Hz_40ms_modulation_mae_db": 0.7513089391426475, "1000–3000 Hz_10ms_log_rms_mae_db": 1.764384431922396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1580525783521887, "1000–3000 Hz_10ms_envelope_correlation": 0.9816386371884036, "1000–3000 Hz_10ms_modulation_mae_db": 0.8347105330000718, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0473175296962465, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09693316999958247, "1000–3000 Hz_40ms_envelope_correlation": 0.9941642656645573, "1000–3000 Hz_40ms_modulation_mae_db": 1.0043066766145021, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0447702618833845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1740781674192931, "3000–6000 Hz_10ms_envelope_correlation": 0.9809250273077088, "3000–6000 Hz_10ms_modulation_mae_db": 0.29183191576197415, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4950971135457123, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10674358793715337, "3000–6000 Hz_40ms_envelope_correlation": 0.9927274743857696, "3000–6000 Hz_40ms_modulation_mae_db": 0.32697554552617253, "mr_stft_db": 5.737951098461459, "band_mae_db": 0.25990955209029754, "rms_gain_error_db": -0.030459649326532903, "output_peak": 0.4318654044922249}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.956542144565322, "spectral_convergence": 0.22224320578116308, "spectral_convergence_status": "ok", "amplitude_floor": 1.642409573690628e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5540755819830245, "spectral_convergence": 0.2096313897617198, "spectral_convergence_status": "ok", "amplitude_floor": 1.497573454010957e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6706935633206665, "spectral_convergence": 0.20389242108274186, "spectral_convergence_status": "ok", "amplitude_floor": 1.2731241516974033e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012546591027172457, "output_power": 0.0012704444566355028, "delta_db": 0.054299407892539404, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0029841932734943235, "output_power": 0.002890706695683091, "delta_db": -0.13822918713841562, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004877853564370709, "output_power": 0.00046727504693467036, "delta_db": -0.18656168652412067, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013442175178747268, "output_power": 0.00012183716120796122, "delta_db": -0.42689779403982875, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010124396480278944, "output_power": 9.553550996837355e-05, "delta_db": -0.2520431796280697, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001060495609474016, "output_power": 9.928334970606928e-05, "delta_db": -0.2863245351242257, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22405929839119998, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10693338969832336, "pitch": null, "output_rms_dbfs": -23.058264831457965, "output_peak": 0.44591588989885167, "reference_rms_dbfs": -22.951331441759642, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.727103763289672}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024147266762860616, "log_rms_mae_db": 1.2443431916223828, "rms_envelope_nrmse": 0.06791680631018246, "envelope_correlation": 0.9957364097512539, "rms_gain_error_db": -0.10692963798953405, "reference_mean_power": 0.005072567967533502, "output_mean_power": 0.004949198938124628}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4164286663529534e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010865583816620668, "output_power": 0.0010556271063680413, "reference_active": true, "delta_db": -0.12542532783106772}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020269542962956985, "output_power": 0.00018535823686009054, "reference_active": true, "delta_db": -0.38832066581041685}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.69603101954121e-05, "output_power": 2.437735756368453e-05, "reference_active": true, "delta_db": -0.43738257179155726}], "active_band_count": 3, "active_band_mae_db": 0.31704285514434727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000225157536036158, "log_rms_mae_db": 0.66400701591665, "rms_envelope_nrmse": 0.0383344136941019, "envelope_correlation": 0.9986508652024862, "rms_gain_error_db": -0.10691952556564499, "reference_mean_power": 0.005098028148503288, "output_mean_power": 0.004974051488455284}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2603245746946083e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010232290392639372, "output_power": 0.0009937004178822396, "reference_active": true, "delta_db": -0.12717384478382798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010544090587602268, "output_power": 9.402887771759762e-05, "reference_active": true, "delta_db": -0.4974787581231239}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0049541105469593e-06, "output_power": 1.6156123277228818e-06, "reference_active": true, "delta_db": -0.9376727858513072}], "active_band_count": 3, "active_band_mae_db": 0.5207751295860864}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020992712290116374, "log_rms_mae_db": 1.846818893282088, "rms_envelope_nrmse": 0.09833149681087738, "envelope_correlation": 0.9929917773928646, "rms_gain_error_db": -0.20143383670125764, "reference_mean_power": 0.0023388664575994473, "output_mean_power": 0.002232862818411557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4073975659228857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007410848450397675, "output_power": 0.0007041012542768819, "reference_active": true, "delta_db": -0.22232814201660267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011531333502346333, "output_power": 9.889943312805022e-05, "reference_active": true, "delta_db": -0.6668573041162702}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6543285086796136e-05, "output_power": 1.84904337225459e-05, "reference_active": true, "delta_db": -1.5700757326293822}], "active_band_count": 3, "active_band_mae_db": 0.8197537262540849}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002036893389343094, "log_rms_mae_db": 1.1093199837797254, "rms_envelope_nrmse": 0.05854561329767458, "envelope_correlation": 0.9976586409004697, "rms_gain_error_db": -0.20143509713836452, "reference_mean_power": 0.0023506239289763434, "output_mean_power": 0.002244086758668079}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3312664615669961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215176046332657, "output_power": 0.0006852546103209464, "reference_active": true, "delta_db": -0.22394965996653693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5162124576630795e-05, "output_power": 4.649081435931591e-05, "reference_active": true, "delta_db": -0.7427383113947651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1332065410684048e-06, "output_power": 1.3696583347422675e-06, "reference_active": true, "delta_db": -1.9242066204946426}], "active_band_count": 3, "active_band_mae_db": 0.9636315306186481}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.160365671559225e-05, "log_rms_mae_db": 1.7661181873364702, "rms_envelope_nrmse": 0.152608015052407, "envelope_correlation": 0.9823343603436796, "rms_gain_error_db": -0.36232392114551437, "reference_mean_power": 0.00021265544148097967, "output_mean_power": 0.00019563389781772724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.181263558889223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.569393841260949e-05, "output_power": 4.993760647585736e-05, "reference_active": true, "delta_db": -0.47380207580709055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4402699100691371e-05, "output_power": 1.1725455044987853e-05, "reference_active": true, "delta_db": -0.8931418161489324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.927958039351209e-06, "output_power": 2.7742658046285924e-06, "reference_active": true, "delta_db": -0.2341677971167867}], "active_band_count": 3, "active_band_mae_db": 0.5337038963576032}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.199537921297776e-05, "log_rms_mae_db": 1.0462820066757497, "rms_envelope_nrmse": 0.08437906120173623, "envelope_correlation": 0.9952771473235219, "rms_gain_error_db": -0.3623320540830435, "reference_mean_power": 0.00021371687213348275, "output_mean_power": 0.00019661000036781578}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0829999638899264e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5369039282181144e-05, "output_power": 4.9299039071876665e-05, "reference_active": true, "delta_db": -0.504285338895721}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.495407348960159e-06, "output_power": 5.167245110274866e-06, "reference_active": true, "delta_db": -0.9934732937733}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.109761653644897e-07, "output_power": 1.6223583952789176e-07, "reference_active": true, "delta_db": -1.1408659397585472}], "active_band_count": 3, "active_band_mae_db": 0.8795415241425227}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3802525725767476e-05, "log_rms_mae_db": 2.0822886281733206, "rms_envelope_nrmse": 0.1725431550796262, "envelope_correlation": 0.9812708103577101, "rms_gain_error_db": -0.06699351538774644, "reference_mean_power": 6.940550281225073e-05, "output_mean_power": 6.834308096916733e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.994304097313729e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.68316376202847e-05, "output_power": 3.584229247137054e-05, "reference_active": true, "delta_db": -0.11825249693659629}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3496621932801275e-06, "output_power": 4.449688181858436e-06, "reference_active": true, "delta_db": -0.7999678092304014}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1665818120524947e-06, "output_power": 9.438305359980965e-07, "reference_active": true, "delta_db": -0.920211770688887}], "active_band_count": 3, "active_band_mae_db": 0.612810692285295}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.808436973199796e-05, "log_rms_mae_db": 1.5117699249867824, "rms_envelope_nrmse": 0.10517040256183016, "envelope_correlation": 0.9927946123815644, "rms_gain_error_db": -0.06699831025164492, "reference_mean_power": 6.97538862423984e-05, "output_mean_power": 6.868605570102477e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.7257017907786166e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.6073966740876416e-05, "output_power": 3.450290888962059e-05, "reference_active": true, "delta_db": -0.1933818918986035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3003381054168396e-06, "output_power": 1.9260058276688646e-06, "reference_active": true, "delta_db": -0.7713407676741026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.80849015690717e-08, "output_power": 7.353549372498728e-08, "reference_active": true, "delta_db": -0.26070054514415875}], "active_band_count": 3, "active_band_mae_db": 0.4084744015722883}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8504199091903063, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f947d251cecb51f5f685e01ffa171a090343e53e42a9585c15784f1adb8da704", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9830756264094929, "standard_deviation": 0.0020340859813630214, "minimum": 0.9807393181182938, "maximum": 0.9860023619159637, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7722236572191028, "standard_deviation": 0.025406716255919927, "minimum": 1.7275859523993067, "maximum": 1.8094529105088302, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6503696494350324, "standard_deviation": 0.16407774505700537, "minimum": 0.47161124062589455, "maximum": 0.9304336261370132, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1492445660224782, "standard_deviation": 0.008908534534221873, "minimum": 0.13638328338299396, "maximum": 0.1580525783521887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945424690195348, "standard_deviation": 0.0009653551875275332, "minimum": 0.9931294695141081, "maximum": 0.9957744859873725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0643842923223772, "standard_deviation": 0.03497267164024957, "minimum": 1.0051813695634864, "maximum": 1.1032501231973675, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8345905463207337, "standard_deviation": 0.20871290639395831, "minimum": 0.4693239395257991, "maximum": 1.0103199430463503, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08826700310334372, "standard_deviation": 0.008451616067810003, "minimum": 0.07572687994047345, "maximum": 0.09871190286757471, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9815939748723688, "standard_deviation": 0.0019065652110255883, "minimum": 0.9787982441014587, "maximum": 0.9846975415833529, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.067591063837413, "standard_deviation": 0.025463248924635338, "minimum": 2.0204431049635008, "maximum": 2.09268672423089, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.57731866818744, "standard_deviation": 0.32355639333046415, "minimum": 0.19130329166736781, "maximum": 1.2393556714135834, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17439802280623481, "standard_deviation": 0.011945478876699818, "minimum": 0.16025026124068278, "maximum": 0.1938843420146426, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922829887728428, "standard_deviation": 0.0009109796966682997, "minimum": 0.9903606530208693, "maximum": 0.9931738921308384, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.510380344505991, "standard_deviation": 0.03479969161446096, "minimum": 1.4646770475395074, "maximum": 1.5726733984468597, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5700546810364886, "standard_deviation": 0.3480477690365034, "minimum": 0.13708651502454672, "maximum": 1.2360512477332037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11293994731643603, "standard_deviation": 0.01058940087510009, "minimum": 0.10266604078564648, "maximum": 0.12956327704770149, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9931207819477657, "standard_deviation": 0.000764952977925074, "minimum": 0.991920367969969, "maximum": 0.9942600066269132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7465484042032298, "standard_deviation": 0.07141265505091587, "minimum": 1.624782382929662, "maximum": 1.846818893282088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7004725021285311, "standard_deviation": 0.2538139341456955, "minimum": 0.48706410576528253, "maximum": 1.2341993622634326, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09727204922020273, "standard_deviation": 0.0060596085173641664, "minimum": 0.08770041475717351, "maximum": 0.1051137905594386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978643765295293, "standard_deviation": 0.00028464379212126816, "minimum": 0.9974056879471441, "maximum": 0.9983462344734854, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0252202653479072, "standard_deviation": 0.06264608585559772, "minimum": 0.9257419726584354, "maximum": 1.1093199837797254, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8251922389706334, "standard_deviation": 0.3045393520704134, "minimum": 0.4255862586260298, "maximum": 1.3990972680437492, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05597890358987409, "standard_deviation": 0.00503510195043947, "minimum": 0.04757742024324632, "maximum": 0.0617563847603116, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19453566862561023, "standard_deviation": 0.048962866897286614, "minimum": 0.12902565855237305, "maximum": 0.25990955209029754, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.720646230423634, "standard_deviation": 0.010442729246066563, "minimum": 5.705776817364349, "maximum": 5.737951098461459, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.44982630869904694, "standard_deviation": 0.012571844595716919, "minimum": 0.4318654044922249, "maximum": 0.4668429524383776, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07782283671276764, "standard_deviation": 0.04002509557752413, "minimum": -0.15647531315293106, "maximum": -0.030459649326532903, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9957327361795745, "standard_deviation": 0.0004948114869328964, "minimum": 0.9950372665813586, "maximum": 0.9965433544552412, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.237007275935381, "standard_deviation": 0.021591870105429036, "minimum": 1.2062085053538556, "maximum": 1.2803015681718135, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.23331958769711403, "standard_deviation": 0.08346872874377344, "minimum": 0.11163830959769973, "maximum": 0.31704285514434727, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0675585478616822, "standard_deviation": 0.004053866887144708, "minimum": 0.06056326870170034, "maximum": 0.07290511555304789, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986784710083563, "standard_deviation": 0.00014042737025680716, "minimum": 0.9984334669585961, "maximum": 0.998837158467364, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6664724458626363, "standard_deviation": 0.020920885642340573, "minimum": 0.636079344615044, "maximum": 0.7028483044398985, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2872297619226147, "standard_deviation": 0.16092150044275058, "minimum": 0.07575941912364755, "maximum": 0.5207751295860864, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0373102834413129, "standard_deviation": 0.0025655229650835168, "minimum": 0.0341103334978251, "maximum": 0.040689001539870435, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8224-274381-0013", "language": "en", "style": "normal", "speaker": "librispeech_8224", "source_utterance_id": "8224-274381-0013", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8224-274381-0013.wav", "source_file_sha256": "2f72b476f11f44ff273c1b5ecea3140f46a345aaa117757881a2ac22dbda2fd9", "decoded_pcm_sha256": "31d48338892252220a9e3042986f534b91479b75a07438f2a0515b00bc8fe055", "original_source_sha256": "cd0ba8a197cc480b016a6386531b3e1ee14c9ff6ddc85f19abde2abc4f00070a", "corpus_entry_id": "librispeech_8224-274381-0013"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.8276984691619873, "predicted_mos": 3.8276984691619873, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1150f8c0f52c5a33501feffd6a3d20ec73f26560f9162de27a11499813385b21", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1027693549985997, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8224-274381-0013", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1150f8c0f52c5a33501feffd6a3d20ec73f26560f9162de27a11499813385b21", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "8f62ebb1c8f2288d8c4b16ccb1481b74a43e4039ba77e57303a5b3e57b940d29"}, "rows": [{"case": "librispeech_8224-274381-0013", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.0984814167022705, "predicted_mos": 3.0984814167022705, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8e052cde21cfd0f7fcaedb1d669b94883a2bd09667f6c6fbe7d2d7cfe9b5026e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.2356441002339125, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.30456649346836, "全帯域_10ms_rms_envelope_nrmse": 0.09168348245725327, "全帯域_10ms_envelope_correlation": 0.991552907832811, "全帯域_10ms_modulation_mae_db": 0.7061601657024658, "全帯域_40ms_log_rms_mae_db": 0.7109739457665045, "全帯域_40ms_rms_envelope_nrmse": 0.052710767835024465, "全帯域_40ms_envelope_correlation": 0.9970845299910067, "全帯域_40ms_modulation_mae_db": 0.809352861944649, "300–1000 Hz_10ms_log_rms_mae_db": 1.7267814624900795, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09377360022219568, "300–1000 Hz_10ms_envelope_correlation": 0.9935180678356789, "300–1000 Hz_10ms_modulation_mae_db": 0.7148367069166998, "300–1000 Hz_40ms_log_rms_mae_db": 1.1030177096750682, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05654529259712844, "300–1000 Hz_40ms_envelope_correlation": 0.9977419137448323, "300–1000 Hz_40ms_modulation_mae_db": 0.8125339226766967, "1000–3000 Hz_10ms_log_rms_mae_db": 1.806579503604515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17267612474667596, "1000–3000 Hz_10ms_envelope_correlation": 0.9802613011155167, "1000–3000 Hz_10ms_modulation_mae_db": 0.9397777384206755, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1496450435019179, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10963328949859477, "1000–3000 Hz_40ms_envelope_correlation": 0.9943415512571427, "1000–3000 Hz_40ms_modulation_mae_db": 1.2178091959262611, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8316847906663323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11858669382455524, "3000–6000 Hz_10ms_envelope_correlation": 0.9926801015449022, "3000–6000 Hz_10ms_modulation_mae_db": 0.2288722828112826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2526278126624668, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06707951774843382, "3000–6000 Hz_40ms_envelope_correlation": 0.9983986384540495, "3000–6000 Hz_40ms_modulation_mae_db": 0.5624247251612141, "mr_stft_db": 5.702051749034509, "band_mae_db": 0.23415302239410762, "rms_gain_error_db": -0.07187058759131246, "output_peak": 0.49877663814281276}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1992164260653246, "全帯域_10ms_rms_envelope_nrmse": 0.07960809638973988, "全帯域_10ms_envelope_correlation": 0.9936581427456692, "全帯域_10ms_modulation_mae_db": 0.3647915661815288, "全帯域_40ms_log_rms_mae_db": 0.6735996803765263, "全帯域_40ms_rms_envelope_nrmse": 0.043112362148956085, "全帯域_40ms_envelope_correlation": 0.9980926267648721, "全帯域_40ms_modulation_mae_db": 0.38195088718012066, "300–1000 Hz_10ms_log_rms_mae_db": 1.690543044741907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08205861196669118, "300–1000 Hz_10ms_envelope_correlation": 0.9953091914685702, "300–1000 Hz_10ms_modulation_mae_db": 0.5127024537593257, "300–1000 Hz_40ms_log_rms_mae_db": 1.013372477821548, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05004705025828628, "300–1000 Hz_40ms_envelope_correlation": 0.9985268330477134, "300–1000 Hz_40ms_modulation_mae_db": 0.49884178286588243, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8435373651983131, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19701391825837547, "1000–3000 Hz_10ms_envelope_correlation": 0.9716795013486672, "1000–3000 Hz_10ms_modulation_mae_db": 0.5768268511486879, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2259985495380559, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1336841183178799, "1000–3000 Hz_40ms_envelope_correlation": 0.9875446400240067, "1000–3000 Hz_40ms_modulation_mae_db": 0.5468429154214944, "3000–6000 Hz_10ms_log_rms_mae_db": 1.896156130021364, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15461369806715117, "3000–6000 Hz_10ms_envelope_correlation": 0.987660054926696, "3000–6000 Hz_10ms_modulation_mae_db": 0.2784790020978834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2935626183673046, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07557988172487112, "3000–6000 Hz_40ms_envelope_correlation": 0.9975637867397779, "3000–6000 Hz_40ms_modulation_mae_db": 0.46526112014709486, "mr_stft_db": 5.719376509881871, "band_mae_db": 0.2076672162134017, "rms_gain_error_db": -0.08052973340853219, "output_peak": 0.5521219483840654}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2532188897898484, "全帯域_10ms_rms_envelope_nrmse": 0.09531822891938702, "全帯域_10ms_envelope_correlation": 0.9909678580933912, "全帯域_10ms_modulation_mae_db": 0.3546189724508155, "全帯域_40ms_log_rms_mae_db": 0.6824252625425384, "全帯域_40ms_rms_envelope_nrmse": 0.04782489496779766, "全帯域_40ms_envelope_correlation": 0.9977529179455915, "全帯域_40ms_modulation_mae_db": 0.3197878760071619, "300–1000 Hz_10ms_log_rms_mae_db": 1.6565496594625868, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10678725587149596, "300–1000 Hz_10ms_envelope_correlation": 0.9915757389977009, "300–1000 Hz_10ms_modulation_mae_db": 0.3410265377834998, "300–1000 Hz_40ms_log_rms_mae_db": 0.9756084850882275, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05967433381807648, "300–1000 Hz_40ms_envelope_correlation": 0.9975557101771874, "300–1000 Hz_40ms_modulation_mae_db": 0.662415630652505, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8046569575216889, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18565600954949082, "1000–3000 Hz_10ms_envelope_correlation": 0.9739600602138444, "1000–3000 Hz_10ms_modulation_mae_db": 0.6077355791533339, "1000–3000 Hz_40ms_log_rms_mae_db": 1.169422994641661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11880187937602217, "1000–3000 Hz_40ms_envelope_correlation": 0.9888168309918874, "1000–3000 Hz_40ms_modulation_mae_db": 0.5339018520674786, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8778396149476204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12688226381617887, "3000–6000 Hz_10ms_envelope_correlation": 0.9909236021630017, "3000–6000 Hz_10ms_modulation_mae_db": 0.2486813590833621, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2819766259105514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06067891641516562, "3000–6000 Hz_40ms_envelope_correlation": 0.9978954645086544, "3000–6000 Hz_40ms_modulation_mae_db": 0.38670034082197, "mr_stft_db": 5.716164512346363, "band_mae_db": 0.15752212759376163, "rms_gain_error_db": -0.1271233001895714, "output_peak": 0.585915841415835}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.262186798601819, "全帯域_10ms_rms_envelope_nrmse": 0.08558176013512568, "全帯域_10ms_envelope_correlation": 0.9928538368130008, "全帯域_10ms_modulation_mae_db": 0.6929246642658814, "全帯域_40ms_log_rms_mae_db": 0.714714669353293, "全帯域_40ms_rms_envelope_nrmse": 0.05457945551213201, "全帯域_40ms_envelope_correlation": 0.9971290000652687, "全帯域_40ms_modulation_mae_db": 0.6228208006682094, "300–1000 Hz_10ms_log_rms_mae_db": 1.7224645693236798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08610851434323208, "300–1000 Hz_10ms_envelope_correlation": 0.9946389161150188, "300–1000 Hz_10ms_modulation_mae_db": 0.6908453897550948, "300–1000 Hz_40ms_log_rms_mae_db": 1.0315899285944103, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05233667179837855, "300–1000 Hz_40ms_envelope_correlation": 0.9981755627744818, "300–1000 Hz_40ms_modulation_mae_db": 0.7646331817078637, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8061548034775796, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18789686396775554, "1000–3000 Hz_10ms_envelope_correlation": 0.9749721875874056, "1000–3000 Hz_10ms_modulation_mae_db": 1.54669776329301, "1000–3000 Hz_40ms_log_rms_mae_db": 1.162302164807498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1159319376533927, "1000–3000 Hz_40ms_envelope_correlation": 0.9919181167523528, "1000–3000 Hz_40ms_modulation_mae_db": 1.997231508925281, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8790538944409496, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13373302484681318, "3000–6000 Hz_10ms_envelope_correlation": 0.9902420991299772, "3000–6000 Hz_10ms_modulation_mae_db": 0.902501559168495, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2448140375137917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06956594860361381, "3000–6000 Hz_40ms_envelope_correlation": 0.9974061554294993, "3000–6000 Hz_40ms_modulation_mae_db": 1.181606791046349, "mr_stft_db": 5.709489664854524, "band_mae_db": 0.23607698080829767, "rms_gain_error_db": -0.1550896166099623, "output_peak": 0.5965407853393542}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.204704119205838, "全帯域_10ms_rms_envelope_nrmse": 0.09904417373427038, "全帯域_10ms_envelope_correlation": 0.9901692862863867, "全帯域_10ms_modulation_mae_db": 0.5021016576315839, "全帯域_40ms_log_rms_mae_db": 0.6849658942161254, "全帯域_40ms_rms_envelope_nrmse": 0.049996688033963586, "全帯域_40ms_envelope_correlation": 0.9974801663526075, "全帯域_40ms_modulation_mae_db": 0.5941980437629392, "300–1000 Hz_10ms_log_rms_mae_db": 1.7500967124019398, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09523917230723757, "300–1000 Hz_10ms_envelope_correlation": 0.9933681593526101, "300–1000 Hz_10ms_modulation_mae_db": 1.0570189628928646, "300–1000 Hz_40ms_log_rms_mae_db": 1.0418693779186694, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055122321194582505, "300–1000 Hz_40ms_envelope_correlation": 0.9979529499250344, "300–1000 Hz_40ms_modulation_mae_db": 1.10089532213589, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8051574913861812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17345228151655495, "1000–3000 Hz_10ms_envelope_correlation": 0.9779058742246556, "1000–3000 Hz_10ms_modulation_mae_db": 0.8170703289804736, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1411088392036326, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10492547517791659, "1000–3000 Hz_40ms_envelope_correlation": 0.9923391304241195, "1000–3000 Hz_40ms_modulation_mae_db": 0.8821460546161809, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8280482606033142, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1371205433128787, "3000–6000 Hz_10ms_envelope_correlation": 0.9900697572626266, "3000–6000 Hz_10ms_modulation_mae_db": 0.43545705938770823, "3000–6000 Hz_40ms_log_rms_mae_db": 1.249271659704976, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06517451867958513, "3000–6000 Hz_40ms_envelope_correlation": 0.9981041325340615, "3000–6000 Hz_40ms_modulation_mae_db": 0.4355262817126048, "mr_stft_db": 5.690021392174863, "band_mae_db": 0.1637802536694022, "rms_gain_error_db": -0.11025453931613298, "output_peak": 0.4995144636336776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2460506921561443, "全帯域_10ms_rms_envelope_nrmse": 0.08134153346661231, "全帯域_10ms_envelope_correlation": 0.9933388263513997, "全帯域_10ms_modulation_mae_db": 0.16003130730554008, "全帯域_40ms_log_rms_mae_db": 0.6555795524829288, "全帯域_40ms_rms_envelope_nrmse": 0.03956909411622581, "全帯域_40ms_envelope_correlation": 0.9983517986354901, "全帯域_40ms_modulation_mae_db": 0.13789914342766815, "300–1000 Hz_10ms_log_rms_mae_db": 1.7279466975340345, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09011052055996402, "300–1000 Hz_10ms_envelope_correlation": 0.9940109096841739, "300–1000 Hz_10ms_modulation_mae_db": 0.41067248739351037, "300–1000 Hz_40ms_log_rms_mae_db": 1.0056136727206497, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052943478931482844, "300–1000 Hz_40ms_envelope_correlation": 0.9980281373767944, "300–1000 Hz_40ms_modulation_mae_db": 0.32088466364821494, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8459101328639451, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1863291896061163, "1000–3000 Hz_10ms_envelope_correlation": 0.9739383662552242, "1000–3000 Hz_10ms_modulation_mae_db": 0.5754727738064452, "1000–3000 Hz_40ms_log_rms_mae_db": 1.14569800535293, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09757803705063298, "1000–3000 Hz_40ms_envelope_correlation": 0.9930843032776745, "1000–3000 Hz_40ms_modulation_mae_db": 1.0790949221030879, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8416466413841879, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14778785241251108, "3000–6000 Hz_10ms_envelope_correlation": 0.989924987693655, "3000–6000 Hz_10ms_modulation_mae_db": 0.681556985434382, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2784727267619214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08771064584787731, "3000–6000 Hz_40ms_envelope_correlation": 0.9974551716651522, "3000–6000 Hz_40ms_modulation_mae_db": 0.9796514438562789, "mr_stft_db": 5.696699367239514, "band_mae_db": 0.20798400415986829, "rms_gain_error_db": -0.03193275827251136, "output_peak": 0.5970421272076613}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2134344938937187, "全帯域_10ms_rms_envelope_nrmse": 0.084895010906357, "全帯域_10ms_envelope_correlation": 0.9928009061734436, "全帯域_10ms_modulation_mae_db": 0.1978981760237558, "全帯域_40ms_log_rms_mae_db": 0.6309190734467373, "全帯域_40ms_rms_envelope_nrmse": 0.04852586534550721, "全帯域_40ms_envelope_correlation": 0.9975627866344903, "全帯域_40ms_modulation_mae_db": 0.5661445760244769, "300–1000 Hz_10ms_log_rms_mae_db": 1.5968023357365115, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08541683743204914, "300–1000 Hz_10ms_envelope_correlation": 0.9947081151190262, "300–1000 Hz_10ms_modulation_mae_db": 0.5547841416348169, "300–1000 Hz_40ms_log_rms_mae_db": 0.9664415788710365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05181107226955855, "300–1000 Hz_40ms_envelope_correlation": 0.9981947640906226, "300–1000 Hz_40ms_modulation_mae_db": 0.6017212704518633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7893078202648975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1749076092841505, "1000–3000 Hz_10ms_envelope_correlation": 0.9771433428325268, "1000–3000 Hz_10ms_modulation_mae_db": 1.5222248931142366, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0418298542100561, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09882417337652971, "1000–3000 Hz_40ms_envelope_correlation": 0.9928116466735012, "1000–3000 Hz_40ms_modulation_mae_db": 1.9711868454890518, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8788848243365357, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12476192210746649, "3000–6000 Hz_10ms_envelope_correlation": 0.9914388697940619, "3000–6000 Hz_10ms_modulation_mae_db": 0.39886647506243084, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2149903491606622, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06799721825551626, "3000–6000 Hz_40ms_envelope_correlation": 0.9978619515823646, "3000–6000 Hz_40ms_modulation_mae_db": 0.695931456399039, "mr_stft_db": 5.7163632371308175, "band_mae_db": 0.13467880855744813, "rms_gain_error_db": -0.08202204619692299, "output_peak": 0.5082090335906888}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2327842092411374, "全帯域_10ms_rms_envelope_nrmse": 0.08594364401495425, "全帯域_10ms_envelope_correlation": 0.9926279322719013, "全帯域_10ms_modulation_mae_db": 0.650744360759883, "全帯域_40ms_log_rms_mae_db": 0.6497176270853425, "全帯域_40ms_rms_envelope_nrmse": 0.04350939396918587, "全帯域_40ms_envelope_correlation": 0.9980845517911003, "全帯域_40ms_modulation_mae_db": 0.7744454187336927, "300–1000 Hz_10ms_log_rms_mae_db": 1.7376181627655183, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09466738030173942, "300–1000 Hz_10ms_envelope_correlation": 0.9935251054825331, "300–1000 Hz_10ms_modulation_mae_db": 0.9887619144700223, "300–1000 Hz_40ms_log_rms_mae_db": 1.045699619319457, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05564719364753509, "300–1000 Hz_40ms_envelope_correlation": 0.9979935112113305, "300–1000 Hz_40ms_modulation_mae_db": 0.9310706815887957, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8143162272824491, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18518288419664822, "1000–3000 Hz_10ms_envelope_correlation": 0.9743566474092024, "1000–3000 Hz_10ms_modulation_mae_db": 1.392551373839155, "1000–3000 Hz_40ms_log_rms_mae_db": 1.184329052099322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10306796164672725, "1000–3000 Hz_40ms_envelope_correlation": 0.9922880590018408, "1000–3000 Hz_40ms_modulation_mae_db": 2.0430064602279714, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9199258299375885, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14147515804400881, "3000–6000 Hz_10ms_envelope_correlation": 0.9885855021639852, "3000–6000 Hz_10ms_modulation_mae_db": 0.7542059842556855, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3532888831118526, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06941012421289268, "3000–6000 Hz_40ms_envelope_correlation": 0.9972274290319335, "3000–6000 Hz_40ms_modulation_mae_db": 0.5442959543189629, "mr_stft_db": 5.707361120862778, "band_mae_db": 0.14680848381677658, "rms_gain_error_db": -0.09255056252848401, "output_peak": 0.5262785731534934}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.862899294401115, "spectral_convergence": 0.2531686065195638, "spectral_convergence_status": "ok", "amplitude_floor": 2.0096490980240517e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.584952166409194, "spectral_convergence": 0.2347060113146063, "spectral_convergence_status": "ok", "amplitude_floor": 1.509105853911261e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.658303786293217, "spectral_convergence": 0.24203565671538407, "spectral_convergence_status": "ok", "amplitude_floor": 1.2442618586634314e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017656001839806658, "output_power": 0.0018603631056541667, "delta_db": 0.22705352675976848, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005544968065872631, "output_power": 0.0053349242808805, "delta_db": -0.16770789520301546, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005978914392629673, "output_power": 0.0006000730153777241, "delta_db": 0.015817623171694287, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.426968976514186e-05, "output_power": 7.25673305708916e-05, "delta_db": -0.6493024747534142, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.40147911812626e-05, "output_power": 6.134412572388305e-05, "delta_db": -0.1850735186448893, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00024101533193812626, "output_power": 0.00023229954601704456, "delta_db": -0.15996309583186413, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23415302239410762, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07187058759131246, "pitch": null, "output_rms_dbfs": -20.882262154519523, "output_peak": 0.49877663814281276, "reference_rms_dbfs": -20.810391566928214, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.702051749034509}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026421160622171636, "log_rms_mae_db": 1.30456649346836, "rms_envelope_nrmse": 0.09168348245725327, "envelope_correlation": 0.991552907832811, "rms_gain_error_db": -0.07185170044659796, "reference_mean_power": 0.008303652784602253, "output_mean_power": 0.00816740348188393}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.437173268209281e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017543654513432086, "output_power": 0.0017689499106573468, "reference_active": true, "delta_db": 0.03595469453135097}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013368683738807852, "output_power": 0.00013317896674910237, "reference_active": true, "delta_db": -0.016530082077805444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.235741942938941e-05, "output_power": 4.496592146639694e-05, "reference_active": true, "delta_db": -2.065995720498241}], "active_band_count": 3, "active_band_mae_db": 0.7061601657024658}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022872947944797367, "log_rms_mae_db": 0.7109739457665045, "rms_envelope_nrmse": 0.052710767835024465, "envelope_correlation": 0.9970845299910067, "rms_gain_error_db": -0.07206775740077336, "reference_mean_power": 0.008338532782066458, "output_mean_power": 0.008201303139496288}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.164744003117204e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015799393891106446, "output_power": 0.0015924786515649867, "reference_active": true, "delta_db": 0.034331924613819814}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.374442374979845e-05, "output_power": 5.976219201343947e-05, "reference_active": true, "delta_db": -0.2801568089159651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.293312028286094e-06, "output_power": 2.0243047543772307e-06, "reference_active": true, "delta_db": -2.113569852304162}], "active_band_count": 3, "active_band_mae_db": 0.809352861944649}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024243770094671305, "log_rms_mae_db": 1.7267814624900795, "rms_envelope_nrmse": 0.09377360022219568, "envelope_correlation": 0.9935180678356789, "rms_gain_error_db": -0.18524780788978967, "reference_mean_power": 0.003959461759548422, "output_mean_power": 0.003794122704170932}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.015177817247118e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001136879031892713, "output_power": 0.0010739319062959191, "reference_active": true, "delta_db": -0.24737511247410612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001232393146611943, "output_power": 0.00010727108062809483, "reference_active": true, "delta_db": -0.60266618879505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0364107060899577e-05, "output_power": 2.25379095737914e-05, "reference_active": true, "delta_db": -1.2944688194809428}], "active_band_count": 3, "active_band_mae_db": 0.7148367069166998}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020477863430336455, "log_rms_mae_db": 1.1030177096750682, "rms_envelope_nrmse": 0.05654529259712844, "envelope_correlation": 0.9977419137448323, "rms_gain_error_db": -0.1852483701193219, "reference_mean_power": 0.003979363820398493, "output_mean_power": 0.0038131932018629425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8942412803353574e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010558685473009428, "output_power": 0.0009934803457124313, "reference_active": true, "delta_db": -0.26450573340594546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.091449836890923e-05, "output_power": 4.5664537443497856e-05, "reference_active": true, "delta_db": -0.4726240642909355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8755256559270567e-06, "output_power": 1.2678731720439301e-06, "reference_active": true, "delta_db": -1.7004719703332092}], "active_band_count": 3, "active_band_mae_db": 0.8125339226766967}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5081112731014556e-05, "log_rms_mae_db": 1.806579503604515, "rms_envelope_nrmse": 0.17267612474667596, "envelope_correlation": 0.9802613011155167, "rms_gain_error_db": -0.6331959936514577, "reference_mean_power": 0.00010911551831198975, "output_mean_power": 9.431199259476957e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.836309277153812e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5256828161206645e-05, "output_power": 2.0531423054222612e-05, "reference_active": true, "delta_db": -0.8995975764378193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.88342602252987e-06, "output_power": 8.619856183239706e-06, "reference_active": true, "delta_db": -0.5940749595917183}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.436572483663522e-06, "output_power": 4.006444350401246e-06, "reference_active": true, "delta_db": -1.3256606792324888}], "active_band_count": 3, "active_band_mae_db": 0.9397777384206755}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.4434083199144116e-05, "log_rms_mae_db": 1.1496450435019179, "rms_envelope_nrmse": 0.10963328949859477, "envelope_correlation": 0.9943415512571427, "rms_gain_error_db": -0.6341070289933295, "reference_mean_power": 0.0001096329653054558, "output_mean_power": 9.4739362486808e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.313909407445067e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3642314296732694e-05, "output_power": 1.9381262671033016e-05, "reference_active": true, "delta_db": -0.8630791895575528}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.77067085832153e-06, "output_power": 5.55689820096058e-06, "reference_active": true, "delta_db": -0.8579926116943581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.651384635522461e-07, "output_power": 2.980894977692518e-07, "reference_active": true, "delta_db": -1.9323557865268723}], "active_band_count": 3, "active_band_mae_db": 1.2178091959262611}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010577651466639849, "log_rms_mae_db": 1.8316847906663323, "rms_envelope_nrmse": 0.11858669382455524, "envelope_correlation": 0.9926801015449022, "rms_gain_error_db": -0.34469509303108553, "reference_mean_power": 0.00018020158938811219, "output_mean_power": 0.00016645203452478457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.705326923064497e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.613613486254781e-05, "output_power": 1.6172168130139656e-05, "reference_active": true, "delta_db": 0.009687327602536823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.942687703713047e-06, "output_power": 9.026158548751425e-06, "reference_active": true, "delta_db": 0.04034893241968106}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.043858273628482e-06, "output_power": 9.015368414679005e-07, "reference_active": true, "delta_db": -0.63658058841163}], "active_band_count": 3, "active_band_mae_db": 0.2288722828112826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.602886913996206e-05, "log_rms_mae_db": 1.2526278126624668, "rms_envelope_nrmse": 0.06707951774843382, "envelope_correlation": 0.9983986384540495, "rms_gain_error_db": -0.3625947079729761, "reference_mean_power": 0.0001775816262598491, "output_mean_power": 0.00016335730427316465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.444034789789241e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7728340440048457e-05, "output_power": 1.7568795498967885e-05, "reference_active": true, "delta_db": -0.039260952806017275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.955564320473026e-06, "output_power": 5.058962872145526e-06, "reference_active": true, "delta_db": 0.08968374711149693}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.260796671427193e-08, "output_power": 6.468676037265985e-08, "reference_active": true, "delta_db": -1.558329475566128}], "active_band_count": 3, "active_band_mae_db": 0.5624247251612141}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.851647279590751, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4a7bedb578bc38bb4795600cdc0ae2943e9241e25daf90c1e37997941742ee94", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9755271601233804, "standard_deviation": 0.002729339504524305, "minimum": 0.9716795013486672, "maximum": 0.9802613011155167, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.814452537699946, "standard_deviation": 0.01993192419793909, "minimum": 1.7893078202648975, "maximum": 1.8459101328639451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9972946627195023, "standard_deviation": 0.42693183707571153, "minimum": 0.5754727738064452, "maximum": 1.54669776329301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18288936014072096, "standard_deviation": 0.008505479257540478, "minimum": 0.17267612474667596, "maximum": 0.19701391825837547, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9916430348003158, "standard_deviation": 0.0022830256468543988, "minimum": 0.9875446400240067, "maximum": 0.9943415512571427, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1525418129193843, "standard_deviation": 0.052412701785204505, "minimum": 1.0418298542100561, "maximum": 1.2259985495380559, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2839024693471008, "standard_deviation": 0.6404147520268727, "minimum": 0.5339018520674786, "maximum": 2.0430064602279714, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11030585901221202, "standard_deviation": 0.012106625849465636, "minimum": 0.09757803705063298, "maximum": 0.1336841183178799, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9901906218348633, "standard_deviation": 0.0015748419441798708, "minimum": 0.987660054926696, "maximum": 0.9926801015449022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8691549982922366, "standard_deviation": 0.03256928599071635, "minimum": 1.8280482606033142, "maximum": 1.9199258299375885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4910775884126537, "standard_deviation": 0.25609453694097867, "minimum": 0.2288722828112826, "maximum": 0.902501559168495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13562014455394544, "standard_deviation": 0.012149613680038814, "minimum": 0.11858669382455524, "maximum": 0.15461369806715117, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9977390912431866, "standard_deviation": 0.0003949306265776896, "minimum": 0.9972274290319335, "maximum": 0.9983986384540495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2711255891491908, "standard_deviation": 0.0415574945449306, "minimum": 1.2149903491606622, "maximum": 1.3532888831118526, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6564247641829393, "standard_deviation": 0.2831840004501448, "minimum": 0.38670034082197, "maximum": 1.181606791046349, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07039959643599447, "standard_deviation": 0.008165862848687484, "minimum": 0.06067891641516562, "maximum": 0.08771064584787731, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.993831775506914, "standard_deviation": 0.0011421404548670475, "minimum": 0.9915757389977009, "maximum": 0.9953091914685702, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7011003305570322, "standard_deviation": 0.05147144348461993, "minimum": 1.5968023357365115, "maximum": 1.7500967124019398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6588310743257293, "standard_deviation": 0.25810608196195006, "minimum": 0.3410265377834998, "maximum": 1.0570189628928646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09177023662557562, "standard_deviation": 0.007740822957584193, "minimum": 0.08205861196669118, "maximum": 0.10678725587149596, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980211727934997, "standard_deviation": 0.0002951444172248699, "minimum": 0.9975557101771874, "maximum": 0.9985268330477134, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0229016062511334, "standard_deviation": 0.043404498895099855, "minimum": 0.9664415788710365, "maximum": 1.1030177096750682, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7116245569659639, "standard_deviation": 0.2464419284283376, "minimum": 0.32088466364821494, "maximum": 1.10089532213589, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05426592681437859, "standard_deviation": 0.003079513125700878, "minimum": 0.05004705025828628, "maximum": 0.05967433381807648, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18608386215163297, "standard_deviation": 0.04009572827732341, "minimum": 0.13467880855744813, "maximum": 0.23607698080829767, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.707190944190655, "standard_deviation": 0.010349154564698378, "minimum": 5.690021392174863, "maximum": 5.719376509881871, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5455499263584485, "standard_deviation": 0.043089350135439956, "minimum": 0.49877663814281276, "maximum": 0.5970421272076613, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09392164301417871, "standard_deviation": 0.03734459019990621, "minimum": -0.1550896166099623, "maximum": -0.03193275827251136, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9922462120710004, "standard_deviation": 0.001221149239475981, "minimum": 0.9901692862863867, "maximum": 0.9936581427456692, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.239520265302774, "standard_deviation": 0.0349233604292183, "minimum": 1.1992164260653246, "maximum": 1.30456649346836, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45365885879018175, "standard_deviation": 0.21762726855478162, "minimum": 0.16003130730554008, "maximum": 0.7061601657024658, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08792699125296247, "standard_deviation": 0.006798448487056487, "minimum": 0.07960809638973988, "maximum": 0.09904417373427038, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9976922972725534, "standard_deviation": 0.00046309541878190467, "minimum": 0.9970845299910067, "maximum": 0.9983517986354901, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6753619631587495, "standard_deviation": 0.029243378521300076, "minimum": 0.6309190734467373, "maximum": 0.714714669353293, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5258249509686148, "standard_deviation": 0.23038676015468165, "minimum": 0.13789914342766815, "maximum": 0.809352861944649, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.047478565241099085, "standard_deviation": 0.005109292574965539, "minimum": 0.03956909411622581, "maximum": 0.05457945551213201, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0013", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.9976415634155273, "predicted_mos": 2.9976415634155273, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ef902298b4c7d7ea8b92438ceab244ec05878b1459ce4273b59b9ec080e7c988", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.213849799707532, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3269145938128766, "全帯域_10ms_rms_envelope_nrmse": 0.09639229548139193, "全帯域_10ms_envelope_correlation": 0.9906631638317308, "全帯域_10ms_modulation_mae_db": 0.7245104198057734, "全帯域_40ms_log_rms_mae_db": 0.721316379258789, "全帯域_40ms_rms_envelope_nrmse": 0.05531970817131418, "全帯域_40ms_envelope_correlation": 0.9967885790136818, "全帯域_40ms_modulation_mae_db": 0.4944220809304248, "300–1000 Hz_10ms_log_rms_mae_db": 1.6786916548871404, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09679125519475872, "300–1000 Hz_10ms_envelope_correlation": 0.9930303771560552, "300–1000 Hz_10ms_modulation_mae_db": 0.6856365009142319, "300–1000 Hz_40ms_log_rms_mae_db": 0.968251961866833, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055956082612905966, "300–1000 Hz_40ms_envelope_correlation": 0.9977449784139364, "300–1000 Hz_40ms_modulation_mae_db": 0.7966946867544973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6973813211536832, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17923782717640066, "1000–3000 Hz_10ms_envelope_correlation": 0.979985823702934, "1000–3000 Hz_10ms_modulation_mae_db": 1.1163982298635649, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0620841037560935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11788197457891249, "1000–3000 Hz_40ms_envelope_correlation": 0.9946677325309013, "1000–3000 Hz_40ms_modulation_mae_db": 1.3984679554214587, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7398688400055078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11755021922371069, "3000–6000 Hz_10ms_envelope_correlation": 0.9929558315395445, "3000–6000 Hz_10ms_modulation_mae_db": 0.2533976442388788, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1941406349157286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07070492136948349, "3000–6000 Hz_40ms_envelope_correlation": 0.9982700336034125, "3000–6000 Hz_40ms_modulation_mae_db": 0.47025174119928964, "mr_stft_db": 5.615798536483435, "band_mae_db": 0.2851097282998037, "rms_gain_error_db": -0.07625113860121244, "output_peak": 0.5087428176105644}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2288580233875606, "全帯域_10ms_rms_envelope_nrmse": 0.07531145277684541, "全帯域_10ms_envelope_correlation": 0.9943474581641779, "全帯域_10ms_modulation_mae_db": 0.3400909685626367, "全帯域_40ms_log_rms_mae_db": 0.6856038058671802, "全帯域_40ms_rms_envelope_nrmse": 0.03896002435988908, "全帯域_40ms_envelope_correlation": 0.9984765814542427, "全帯域_40ms_modulation_mae_db": 0.3319471878196685, "300–1000 Hz_10ms_log_rms_mae_db": 1.608445001654161, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07958205558311675, "300–1000 Hz_10ms_envelope_correlation": 0.9955334339520987, "300–1000 Hz_10ms_modulation_mae_db": 0.46835959609025646, "300–1000 Hz_40ms_log_rms_mae_db": 0.8856522871466971, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04724788173135827, "300–1000 Hz_40ms_envelope_correlation": 0.9986520473594656, "300–1000 Hz_40ms_modulation_mae_db": 0.49353519453635136, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7801386431705726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21056742680299106, "1000–3000 Hz_10ms_envelope_correlation": 0.967416009950971, "1000–3000 Hz_10ms_modulation_mae_db": 0.8376395238918458, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1183280956897281, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14744064454115166, "1000–3000 Hz_40ms_envelope_correlation": 0.9844094687325052, "1000–3000 Hz_40ms_modulation_mae_db": 0.6091967427401567, "3000–6000 Hz_10ms_log_rms_mae_db": 1.808757753718726, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1546751626024691, "3000–6000 Hz_10ms_envelope_correlation": 0.9878228735759934, "3000–6000 Hz_10ms_modulation_mae_db": 0.2886104425748561, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2323427068541903, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07637941369389156, "3000–6000 Hz_40ms_envelope_correlation": 0.9976455352152385, "3000–6000 Hz_40ms_modulation_mae_db": 0.2972768391248008, "mr_stft_db": 5.628944462012806, "band_mae_db": 0.20757611288806266, "rms_gain_error_db": -0.08804858812598787, "output_peak": 0.5536706977130847}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.232541627100057, "全帯域_10ms_rms_envelope_nrmse": 0.09513495207810257, "全帯域_10ms_envelope_correlation": 0.9910168136661371, "全帯域_10ms_modulation_mae_db": 0.4639576202485876, "全帯域_40ms_log_rms_mae_db": 0.6602472734415245, "全帯域_40ms_rms_envelope_nrmse": 0.049842959822233614, "全帯域_40ms_envelope_correlation": 0.9975577281644635, "全帯域_40ms_modulation_mae_db": 0.36558730710558285, "300–1000 Hz_10ms_log_rms_mae_db": 1.5805870557824222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11063045167260363, "300–1000 Hz_10ms_envelope_correlation": 0.9908915966130845, "300–1000 Hz_10ms_modulation_mae_db": 0.3224375383175176, "300–1000 Hz_40ms_log_rms_mae_db": 0.8406647807709656, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06087334198861743, "300–1000 Hz_40ms_envelope_correlation": 0.9973929483311048, "300–1000 Hz_40ms_modulation_mae_db": 0.7695443337270668, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7492463741917066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17075888183007448, "1000–3000 Hz_10ms_envelope_correlation": 0.9785519396239051, "1000–3000 Hz_10ms_modulation_mae_db": 0.4726615925633138, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0921712191534254, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10878503007923532, "1000–3000 Hz_40ms_envelope_correlation": 0.9914692704092303, "1000–3000 Hz_40ms_modulation_mae_db": 0.4904544559764121, "3000–6000 Hz_10ms_log_rms_mae_db": 1.80988917762478, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1270780614838969, "3000–6000 Hz_10ms_envelope_correlation": 0.9909028985411906, "3000–6000 Hz_10ms_modulation_mae_db": 0.48279209811518126, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2583826408246002, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06015027334402129, "3000–6000 Hz_40ms_envelope_correlation": 0.9979405526541736, "3000–6000 Hz_40ms_modulation_mae_db": 0.2321922964157904, "mr_stft_db": 5.628850221195468, "band_mae_db": 0.21554547624303133, "rms_gain_error_db": -0.13223743498442822, "output_peak": 0.575556604419594}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2901617174877713, "全帯域_10ms_rms_envelope_nrmse": 0.08528647145927165, "全帯域_10ms_envelope_correlation": 0.9928735552463881, "全帯域_10ms_modulation_mae_db": 0.402321205408573, "全帯域_40ms_log_rms_mae_db": 0.7272292138265404, "全帯域_40ms_rms_envelope_nrmse": 0.05232018185528801, "全帯域_40ms_envelope_correlation": 0.9973448331785981, "全帯域_40ms_modulation_mae_db": 0.24179766855982976, "300–1000 Hz_10ms_log_rms_mae_db": 1.6725744974033205, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09033476949357273, "300–1000 Hz_10ms_envelope_correlation": 0.9940096898532801, "300–1000 Hz_10ms_modulation_mae_db": 0.7225331296243945, "300–1000 Hz_40ms_log_rms_mae_db": 0.9388549878476197, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055189420197444626, "300–1000 Hz_40ms_envelope_correlation": 0.9978659117266822, "300–1000 Hz_40ms_modulation_mae_db": 0.8036164403194319, "1000–3000 Hz_10ms_log_rms_mae_db": 1.733270708490539, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18574932126863433, "1000–3000 Hz_10ms_envelope_correlation": 0.9763046936004984, "1000–3000 Hz_10ms_modulation_mae_db": 1.1623185048591715, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0972707482566482, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11748676457257935, "1000–3000 Hz_40ms_envelope_correlation": 0.9925422506750373, "1000–3000 Hz_40ms_modulation_mae_db": 1.5310084442155916, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7707472858422635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1291844564667623, "3000–6000 Hz_10ms_envelope_correlation": 0.9910221378023468, "3000–6000 Hz_10ms_modulation_mae_db": 1.0382240480068643, "3000–6000 Hz_40ms_log_rms_mae_db": 1.184887909720329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0681422049032463, "3000–6000 Hz_40ms_envelope_correlation": 0.9976096047691051, "3000–6000 Hz_40ms_modulation_mae_db": 1.3267954454765714, "mr_stft_db": 5.621752606524274, "band_mae_db": 0.26130449997136596, "rms_gain_error_db": -0.14361847446377912, "output_peak": 0.6494988011446321}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2304942268049703, "全帯域_10ms_rms_envelope_nrmse": 0.095232564196293, "全帯域_10ms_envelope_correlation": 0.9909429132442097, "全帯域_10ms_modulation_mae_db": 0.4968513568587183, "全帯域_40ms_log_rms_mae_db": 0.6913838511342065, "全帯域_40ms_rms_envelope_nrmse": 0.050143146956532665, "全帯域_40ms_envelope_correlation": 0.997488596281804, "全帯域_40ms_modulation_mae_db": 0.574904236328356, "300–1000 Hz_10ms_log_rms_mae_db": 1.7181632039596928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09500805300485797, "300–1000 Hz_10ms_envelope_correlation": 0.9933884993697341, "300–1000 Hz_10ms_modulation_mae_db": 1.077195513216895, "300–1000 Hz_40ms_log_rms_mae_db": 0.9071329974172186, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054784110546777696, "300–1000 Hz_40ms_envelope_correlation": 0.9979671891421974, "300–1000 Hz_40ms_modulation_mae_db": 1.0947080839572614, "1000–3000 Hz_10ms_log_rms_mae_db": 1.74175216860917, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16988748830265815, "1000–3000 Hz_10ms_envelope_correlation": 0.9793961332014045, "1000–3000 Hz_10ms_modulation_mae_db": 0.9511219237640735, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0830076347154796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10090823224183312, "1000–3000 Hz_40ms_envelope_correlation": 0.9937779177047215, "1000–3000 Hz_40ms_modulation_mae_db": 1.019922090483539, "3000–6000 Hz_10ms_log_rms_mae_db": 1.700587836196746, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13681322603550095, "3000–6000 Hz_10ms_envelope_correlation": 0.9899979409887594, "3000–6000 Hz_10ms_modulation_mae_db": 0.35978527828282186, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1319563940885486, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06634722826271869, "3000–6000 Hz_40ms_envelope_correlation": 0.9979211400219996, "3000–6000 Hz_40ms_modulation_mae_db": 0.3608918407295783, "mr_stft_db": 5.603768632248254, "band_mae_db": 0.2014213191951869, "rms_gain_error_db": -0.11927881792918962, "output_peak": 0.5018908781862825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2428888072737923, "全帯域_10ms_rms_envelope_nrmse": 0.08050491974893788, "全帯域_10ms_envelope_correlation": 0.9934796020658823, "全帯域_10ms_modulation_mae_db": 0.2800614395701198, "全帯域_40ms_log_rms_mae_db": 0.6612460121214788, "全帯域_40ms_rms_envelope_nrmse": 0.03882227363149827, "全帯域_40ms_envelope_correlation": 0.9984221906106431, "全帯域_40ms_modulation_mae_db": 0.2892153774825576, "300–1000 Hz_10ms_log_rms_mae_db": 1.6801186723757522, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09442500233843158, "300–1000 Hz_10ms_envelope_correlation": 0.9933752533985075, "300–1000 Hz_10ms_modulation_mae_db": 0.5014700974917771, "300–1000 Hz_40ms_log_rms_mae_db": 0.8865826937264515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05425125308559881, "300–1000 Hz_40ms_envelope_correlation": 0.997891347194392, "300–1000 Hz_40ms_modulation_mae_db": 0.382538165980641, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7503975543608612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18399697366006285, "1000–3000 Hz_10ms_envelope_correlation": 0.9756717271015103, "1000–3000 Hz_10ms_modulation_mae_db": 1.1332485546131028, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0559063457222797, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10327644612060397, "1000–3000 Hz_40ms_envelope_correlation": 0.9937201563011802, "1000–3000 Hz_40ms_modulation_mae_db": 1.7339996747107762, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7954324521582843, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14471615791188774, "3000–6000 Hz_10ms_envelope_correlation": 0.9905141264041767, "3000–6000 Hz_10ms_modulation_mae_db": 0.6904938080943718, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2701958408932217, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0868966192673574, "3000–6000 Hz_40ms_envelope_correlation": 0.9976239745135337, "3000–6000 Hz_40ms_modulation_mae_db": 0.9303700558465194, "mr_stft_db": 5.60950934979456, "band_mae_db": 0.24398730718750786, "rms_gain_error_db": -0.04620920033597809, "output_peak": 0.5815412386421437}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2408695559465033, "全帯域_10ms_rms_envelope_nrmse": 0.08512612421255497, "全帯域_10ms_envelope_correlation": 0.9927346072515404, "全帯域_10ms_modulation_mae_db": 0.44848803642097096, "全帯域_40ms_log_rms_mae_db": 0.6391200585948159, "全帯域_40ms_rms_envelope_nrmse": 0.049087384369137614, "全帯域_40ms_envelope_correlation": 0.9974516376644607, "全帯域_40ms_modulation_mae_db": 0.9223841971379426, "300–1000 Hz_10ms_log_rms_mae_db": 1.5675547737971796, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08500162955986101, "300–1000 Hz_10ms_envelope_correlation": 0.9946490059948268, "300–1000 Hz_10ms_modulation_mae_db": 0.5470467302720325, "300–1000 Hz_40ms_log_rms_mae_db": 0.8720978011576914, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05005895620670488, "300–1000 Hz_40ms_envelope_correlation": 0.998202582296637, "300–1000 Hz_40ms_modulation_mae_db": 0.5732875885493446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.735478364962312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1651331584381908, "1000–3000 Hz_10ms_envelope_correlation": 0.9797543393776965, "1000–3000 Hz_10ms_modulation_mae_db": 0.8294265432352379, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9981500304076509, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08426608282953586, "1000–3000 Hz_40ms_envelope_correlation": 0.9950702868890914, "1000–3000 Hz_40ms_modulation_mae_db": 1.1519340346845408, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7945557694909862, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12029798617964149, "3000–6000 Hz_10ms_envelope_correlation": 0.992073988716424, "3000–6000 Hz_10ms_modulation_mae_db": 0.48883962036181017, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1864369960370567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06648986170976655, "3000–6000 Hz_40ms_envelope_correlation": 0.9979842190263316, "3000–6000 Hz_40ms_modulation_mae_db": 0.7362632375593284, "mr_stft_db": 5.629396152332653, "band_mae_db": 0.13314184453498662, "rms_gain_error_db": -0.04856058076569602, "output_peak": 0.5230997507466153}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3004880289107805, "全帯域_10ms_rms_envelope_nrmse": 0.0905383963066147, "全帯域_10ms_envelope_correlation": 0.9917932497253107, "全帯域_10ms_modulation_mae_db": 0.31813908725139084, "全帯域_40ms_log_rms_mae_db": 0.6871196771380645, "全帯域_40ms_rms_envelope_nrmse": 0.04246395725211952, "全帯域_40ms_envelope_correlation": 0.9981538000692368, "全帯域_40ms_modulation_mae_db": 0.4991201808992242, "300–1000 Hz_10ms_log_rms_mae_db": 1.690444539487943, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09817663393717604, "300–1000 Hz_10ms_envelope_correlation": 0.9929450503957457, "300–1000 Hz_10ms_modulation_mae_db": 1.0802844090395374, "300–1000 Hz_40ms_log_rms_mae_db": 0.9466454136069169, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056514587047159635, "300–1000 Hz_40ms_envelope_correlation": 0.9978499114241023, "300–1000 Hz_40ms_modulation_mae_db": 1.0164368093047949, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7527429586337797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1810510974291262, "1000–3000 Hz_10ms_envelope_correlation": 0.976160584559962, "1000–3000 Hz_10ms_modulation_mae_db": 1.7555571560142182, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1122025578500463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10438434596133606, "1000–3000 Hz_40ms_envelope_correlation": 0.9929895505044615, "1000–3000 Hz_40ms_modulation_mae_db": 2.4708816619971135, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8474888361333932, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13918804689078623, "3000–6000 Hz_10ms_envelope_correlation": 0.9889669123327427, "3000–6000 Hz_10ms_modulation_mae_db": 0.6458721956784711, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2979918215372817, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06514853635233261, "3000–6000 Hz_40ms_envelope_correlation": 0.9975933620977403, "3000–6000 Hz_40ms_modulation_mae_db": 0.7798970505971838, "mr_stft_db": 5.624522356355665, "band_mae_db": 0.19128779462580062, "rms_gain_error_db": -0.08150051441616164, "output_peak": 0.5274768344920712}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.771020119435291, "spectral_convergence": 0.25553794581132905, "spectral_convergence_status": "ok", "amplitude_floor": 2.0096490980240517e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.498304221857376, "spectral_convergence": 0.23787516116460647, "spectral_convergence_status": "ok", "amplitude_floor": 1.509105853911261e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.578071268157639, "spectral_convergence": 0.24194497590077207, "spectral_convergence_status": "ok", "amplitude_floor": 1.2442618586634314e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017656001839806658, "output_power": 0.001853881059407062, "delta_db": 0.21189501994418988, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005544968065872631, "output_power": 0.005337667304669504, "delta_db": -0.165475484992289, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005978914392629673, "output_power": 0.0006016554583575321, "delta_db": 0.02725525931498433, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.426968976514186e-05, "output_power": 7.002159736127813e-05, "delta_db": -0.8043938124456571, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.40147911812626e-05, "output_power": 5.963892882502337e-05, "delta_db": -0.3075049865698382, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00024101533193812626, "output_power": 0.00023047896233406313, "delta_db": -0.1941338065318637, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2851097282998037, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07625113860121244, "pitch": null, "output_rms_dbfs": -20.88664270552943, "output_peak": 0.5087428176105644, "reference_rms_dbfs": -20.810391566928214, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.615798536483435}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026421160622171636, "log_rms_mae_db": 1.3269145938128766, "rms_envelope_nrmse": 0.09639229548139193, "envelope_correlation": 0.9906631638317308, "rms_gain_error_db": -0.07624659239353254, "reference_mean_power": 0.008303652784602253, "output_mean_power": 0.008159142566466429}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.437173268209281e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017543654513432086, "output_power": 0.0017510760865658169, "reference_active": true, "delta_db": -0.008150490462492464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013368683738807852, "output_power": 0.00014765280349061024, "reference_active": true, "delta_db": 0.4315304798608511}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.235741942938941e-05, "output_power": 4.8539806507025224e-05, "reference_active": true, "delta_db": -1.7338502890939766}], "active_band_count": 3, "active_band_mae_db": 0.7245104198057734}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022872947944797367, "log_rms_mae_db": 0.721316379258789, "rms_envelope_nrmse": 0.05531970817131418, "envelope_correlation": 0.9967885790136818, "rms_gain_error_db": -0.07645732584222109, "reference_mean_power": 0.008338532782066458, "output_mean_power": 0.008193017979141555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.164744003117204e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015799393891106446, "output_power": 0.0015799538558674375, "reference_active": true, "delta_db": 3.976610701199864e-05}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.374442374979845e-05, "output_power": 6.608666453770738e-05, "reference_active": true, "delta_db": 0.1567163368268805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.293312028286094e-06, "output_power": 2.426508775913194e-06, "reference_active": true, "delta_db": -1.326510139857382}], "active_band_count": 3, "active_band_mae_db": 0.4944220809304248}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024243770094671305, "log_rms_mae_db": 1.6786916548871404, "rms_envelope_nrmse": 0.09679125519475872, "envelope_correlation": 0.9930303771560552, "rms_gain_error_db": -0.17070229923172933, "reference_mean_power": 0.003959461759548422, "output_mean_power": 0.0038068513866324396}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.015177817247118e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001136879031892713, "output_power": 0.0010748895357558296, "reference_active": true, "delta_db": -0.24350421646067974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001232393146611943, "output_power": 0.00010569991707297282, "reference_active": true, "delta_db": -0.6667462794128789}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0364107060899577e-05, "output_power": 2.331817800927309e-05, "reference_active": true, "delta_db": -1.146659006869137}], "active_band_count": 3, "active_band_mae_db": 0.6856365009142319}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020477863430336455, "log_rms_mae_db": 0.968251961866833, "rms_envelope_nrmse": 0.055956082612905966, "envelope_correlation": 0.9977449784139364, "rms_gain_error_db": -0.1707027667637589, "reference_mean_power": 0.003979363820398493, "output_mean_power": 0.0038259859462576387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8942412803353574e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010558685473009428, "output_power": 0.000993973048178225, "reference_active": true, "delta_db": -0.2623524455134526}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.091449836890923e-05, "output_power": 4.5346311626367244e-05, "reference_active": true, "delta_db": -0.5029950103972299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8755256559270567e-06, "output_power": 1.2901771589155525e-06, "reference_active": true, "delta_db": -1.6247366043528095}], "active_band_count": 3, "active_band_mae_db": 0.7966946867544973}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5081112731014556e-05, "log_rms_mae_db": 1.6973813211536832, "rms_envelope_nrmse": 0.17923782717640066, "envelope_correlation": 0.979985823702934, "rms_gain_error_db": -0.7886016360984136, "reference_mean_power": 0.00010911551831198975, "output_mean_power": 9.099684948784515e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.836309277153812e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5256828161206645e-05, "output_power": 2.0379287556627113e-05, "reference_active": true, "delta_db": -0.9318981210574585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.88342602252987e-06, "output_power": 8.545575740426953e-06, "reference_active": true, "delta_db": -0.6316618818834564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.436572483663522e-06, "output_power": 3.6038085724984567e-06, "reference_active": true, "delta_db": -1.7856346866497794}], "active_band_count": 3, "active_band_mae_db": 1.1163982298635649}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.4434083199144116e-05, "log_rms_mae_db": 1.0620841037560935, "rms_envelope_nrmse": 0.11788197457891249, "envelope_correlation": 0.9946677325309013, "rms_gain_error_db": -0.7894165916482938, "reference_mean_power": 0.0001096329653054558, "output_mean_power": 9.141121926659077e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.313909407445067e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3642314296732694e-05, "output_power": 1.920403473834329e-05, "reference_active": true, "delta_db": -0.9029750357234541}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.77067085832153e-06, "output_power": 5.42893877519208e-06, "reference_active": true, "delta_db": -0.9591675808226654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.651384635522461e-07, "output_power": 2.718041785383328e-07, "reference_active": true, "delta_db": -2.3332612497182574}], "active_band_count": 3, "active_band_mae_db": 1.3984679554214587}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010577651466639849, "log_rms_mae_db": 1.7398688400055078, "rms_envelope_nrmse": 0.11755021922371069, "envelope_correlation": 0.9929558315395445, "rms_gain_error_db": -0.3875394388126894, "reference_mean_power": 0.00018020158938811219, "output_mean_power": 0.00016481801268793716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.705326923064497e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.613613486254781e-05, "output_power": 1.6609584583721833e-05, "reference_active": true, "delta_db": 0.12559255640697808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.942687703713047e-06, "output_power": 9.112809492703415e-06, "reference_active": true, "delta_db": 0.08184226653448015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.043858273628482e-06, "output_power": 9.19106265777711e-07, "reference_active": true, "delta_db": -0.5527581097751781}], "active_band_count": 3, "active_band_mae_db": 0.2533976442388788}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.602886913996206e-05, "log_rms_mae_db": 1.1941406349157286, "rms_envelope_nrmse": 0.07070492136948349, "envelope_correlation": 0.9982700336034125, "rms_gain_error_db": -0.40585169054615283, "reference_mean_power": 0.0001775816262598491, "output_mean_power": 0.00016173829470857027}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.444034789789241e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7728340440048457e-05, "output_power": 1.8014577058619426e-05, "reference_active": true, "delta_db": 0.06955987377435148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.955564320473026e-06, "output_power": 5.097331154815612e-06, "reference_active": true, "delta_db": 0.12249731605761789}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.260796671427193e-08, "output_power": 6.994852252683223e-08, "reference_active": true, "delta_db": -1.2186980337658995}], "active_band_count": 3, "active_band_mae_db": 0.47025174119928964}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8500639521621247, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6dd0d6d454826f1b8ec56a62372b06280ed4aec13209c30c76b5cc78b243b8cf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9766551563898602, "standard_deviation": 0.004114528370940657, "minimum": 0.967416009950971, "maximum": 0.979985823702934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7425510116965781, "standard_deviation": 0.023299869363098132, "minimum": 1.6973813211536832, "maximum": 1.7801386431705726, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.032296503600566, "standard_deviation": 0.36957586032339135, "minimum": 0.4726615925633138, "maximum": 1.7555571560142182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1807977718635173, "standard_deviation": 0.014084826711128106, "minimum": 0.1651331584381908, "maximum": 0.21056742680299106, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9923308292183911, "standard_deviation": 0.0034008168881245124, "minimum": 0.9844094687325052, "maximum": 0.9950702868890914, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.077390091943919, "standard_deviation": 0.03874443538911499, "minimum": 0.9981500304076509, "maximum": 1.1183280956897281, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3007331325286984, "standard_deviation": 0.6386723301730487, "minimum": 0.4904544559764121, "maximum": 2.4708816619971135, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11055369011564847, "standard_deviation": 0.01828597855343036, "minimum": 0.08426608282953586, "maximum": 0.14744064454115166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9905320887376472, "standard_deviation": 0.0016346643667820178, "minimum": 0.9878228735759934, "maximum": 0.9929558315395445, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7834159938963359, "standard_deviation": 0.04572309156068853, "minimum": 1.700587836196746, "maximum": 1.8474888361333932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5310018919191569, "standard_deviation": 0.25777672421502873, "minimum": 0.2533976442388788, "maximum": 1.0382240480068643, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13368791459933194, "standard_deviation": 0.012568625590179924, "minimum": 0.11755021922371069, "maximum": 0.1546751626024691, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9978235527376919, "standard_deviation": 0.00024455623020293364, "minimum": 0.9975933620977403, "maximum": 0.9982700336034125, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2195418681088697, "standard_deviation": 0.05478002052818565, "minimum": 1.1319563940885486, "maximum": 1.2979918215372817, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6417423133686327, "standard_deviation": 0.37326711762948794, "minimum": 0.2321922964157904, "maximum": 1.3267954454765714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07003238236285222, "standard_deviation": 0.008249244263796963, "minimum": 0.06015027334402129, "maximum": 0.0868966192673574, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9934778633416665, "standard_deviation": 0.0013664975521340144, "minimum": 0.9908915966130845, "maximum": 0.9955334339520987, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6495724249184516, "standard_deviation": 0.05588593490290513, "minimum": 1.5675547737971796, "maximum": 1.7181632039596928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6756204393708303, "standard_deviation": 0.27822687281117464, "minimum": 0.3224375383175176, "maximum": 1.0802844090395374, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09374373134804731, "standard_deviation": 0.009294855833557126, "minimum": 0.07958205558311675, "maximum": 0.11063045167260363, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979458644860648, "standard_deviation": 0.0003644361757219482, "minimum": 0.9973929483311048, "maximum": 0.9986520473594656, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9057353654425493, "standard_deviation": 0.042789586885630296, "minimum": 0.8406647807709656, "maximum": 0.968251961866833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7412951628911736, "standard_deviation": 0.24688713808795393, "minimum": 0.382538165980641, "maximum": 1.0947080839572614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054359454177070915, "standard_deviation": 0.0041330408041004175, "minimum": 0.04724788173135827, "maximum": 0.06087334198861743, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2174217603682182, "standard_deviation": 0.04683656054840202, "minimum": 0.13314184453498662, "maximum": 0.2851097282998037, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.62031778961839, "standard_deviation": 0.009713732883883757, "minimum": 5.603768632248254, "maximum": 5.629396152332653, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5526847028693734, "standard_deviation": 0.04891884427188656, "minimum": 0.5018908781862825, "maximum": 0.6494988011446321, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09196309370280413, "standard_deviation": 0.03661635282039851, "minimum": -0.14361847446377912, "maximum": -0.04620920033597809, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9922314203994221, "standard_deviation": 0.001335667295588113, "minimum": 0.9906631638317308, "maximum": 0.9943474581641779, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2616520725905391, "standard_deviation": 0.03828069302431948, "minimum": 1.2288580233875606, "maximum": 1.3269145938128766, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4343025167658463, "standard_deviation": 0.13958945460389388, "minimum": 0.2800614395701198, "maximum": 0.7245104198057734, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0879408970325015, "standard_deviation": 0.007679857864827827, "minimum": 0.07531145277684541, "maximum": 0.09639229548139193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977104933046413, "standard_deviation": 0.0005871407719704293, "minimum": 0.9967885790136818, "maximum": 0.9984765814542427, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6841582839228251, "standard_deviation": 0.03029391173815192, "minimum": 0.6391200585948159, "maximum": 0.7272292138265404, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4649222795329483, "standard_deviation": 0.21765453978826346, "minimum": 0.24179766855982976, "maximum": 0.9223841971379426, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04711995455225162, "standard_deviation": 0.006231462036430253, "minimum": 0.03882227363149827, "maximum": 0.05531970817131418, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0013", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.4383456707000732, "predicted_mos": 2.4383456707000732, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "de5807f001bda9c07b9c6a1c59d54aaed6153b3df58424ecc6b00f2dfa784fb7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8410153998993337, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2091614625908362, "全帯域_10ms_rms_envelope_nrmse": 0.09263806457453595, "全帯域_10ms_envelope_correlation": 0.9914084135146282, "全帯域_10ms_modulation_mae_db": 0.22513126692975524, "全帯域_40ms_log_rms_mae_db": 0.652280267845908, "全帯域_40ms_rms_envelope_nrmse": 0.04673534353864879, "全帯域_40ms_envelope_correlation": 0.9977739400976, "全帯域_40ms_modulation_mae_db": 0.3304371513447841, "300–1000 Hz_10ms_log_rms_mae_db": 1.9934982101704624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08596175119690376, "300–1000 Hz_10ms_envelope_correlation": 0.9945767890842785, "300–1000 Hz_10ms_modulation_mae_db": 0.7101589081398979, "300–1000 Hz_40ms_log_rms_mae_db": 1.2860942738797725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054928713464352415, "300–1000 Hz_40ms_envelope_correlation": 0.9978355402908659, "300–1000 Hz_40ms_modulation_mae_db": 0.7392639008782108, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2181187214947573, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20553115197829427, "1000–3000 Hz_10ms_envelope_correlation": 0.9682500894378269, "1000–3000 Hz_10ms_modulation_mae_db": 1.5170963838244889, "1000–3000 Hz_40ms_log_rms_mae_db": 1.428897688090704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11679962683952391, "1000–3000 Hz_40ms_envelope_correlation": 0.9897861524398393, "1000–3000 Hz_40ms_modulation_mae_db": 2.353891501411415, "3000–6000 Hz_10ms_log_rms_mae_db": 2.573325500178391, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14409776953087752, "3000–6000 Hz_10ms_envelope_correlation": 0.9882638027738571, "3000–6000 Hz_10ms_modulation_mae_db": 0.7815613105573765, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7600680262945863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0829916554452066, "3000–6000 Hz_40ms_envelope_correlation": 0.9961610991413019, "3000–6000 Hz_40ms_modulation_mae_db": 1.0042282987902074, "mr_stft_db": 6.12086531474553, "band_mae_db": 0.1363760686595615, "rms_gain_error_db": -0.09525701573824544, "output_peak": 0.48536580763254666}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2223363782689596, "全帯域_10ms_rms_envelope_nrmse": 0.07806788193923136, "全帯域_10ms_envelope_correlation": 0.9940123720992464, "全帯域_10ms_modulation_mae_db": 0.18411418669517735, "全帯域_40ms_log_rms_mae_db": 0.5899116473432024, "全帯域_40ms_rms_envelope_nrmse": 0.0413202964821814, "全帯域_40ms_envelope_correlation": 0.9983824338483422, "全帯域_40ms_modulation_mae_db": 0.46204951528350763, "300–1000 Hz_10ms_log_rms_mae_db": 1.9401073872762211, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08747117691763563, "300–1000 Hz_10ms_envelope_correlation": 0.9946278066846791, "300–1000 Hz_10ms_modulation_mae_db": 0.5847335272162643, "300–1000 Hz_40ms_log_rms_mae_db": 1.1373199061264547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051888149376293886, "300–1000 Hz_40ms_envelope_correlation": 0.9984182034580316, "300–1000 Hz_40ms_modulation_mae_db": 0.7331947391023901, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1940629812901005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19745450385833943, "1000–3000 Hz_10ms_envelope_correlation": 0.9707390912376392, "1000–3000 Hz_10ms_modulation_mae_db": 1.1970391318216151, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4246351150769854, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10840367073515386, "1000–3000 Hz_40ms_envelope_correlation": 0.9913221057215612, "1000–3000 Hz_40ms_modulation_mae_db": 1.6816067328419868, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5358870300932272, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1441096158267246, "3000–6000 Hz_10ms_envelope_correlation": 0.9882660782883429, "3000–6000 Hz_10ms_modulation_mae_db": 0.45173396605488003, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7683969115103695, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07900141164993764, "3000–6000 Hz_40ms_envelope_correlation": 0.9965459065711536, "3000–6000 Hz_40ms_modulation_mae_db": 0.6974334480391643, "mr_stft_db": 6.105452992317118, "band_mae_db": 0.18448530881463457, "rms_gain_error_db": -0.1266949379309108, "output_peak": 0.544463226644489}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1680750186902766, "全帯域_10ms_rms_envelope_nrmse": 0.12872181858999654, "全帯域_10ms_envelope_correlation": 0.9837377980050892, "全帯域_10ms_modulation_mae_db": 2.3079524500759883, "全帯域_40ms_log_rms_mae_db": 0.6202858463118234, "全帯域_40ms_rms_envelope_nrmse": 0.09950311262975038, "全帯域_40ms_envelope_correlation": 0.9896430730158063, "全帯域_40ms_modulation_mae_db": 2.944328611302613, "300–1000 Hz_10ms_log_rms_mae_db": 1.9838978915529923, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09741951884348055, "300–1000 Hz_10ms_envelope_correlation": 0.9927985289249472, "300–1000 Hz_10ms_modulation_mae_db": 0.7832719222684114, "300–1000 Hz_40ms_log_rms_mae_db": 1.2433865947061704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05767570009118415, "300–1000 Hz_40ms_envelope_correlation": 0.9973967257947791, "300–1000 Hz_40ms_modulation_mae_db": 0.7998635036453475, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2101503628993764, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1897176969144584, "1000–3000 Hz_10ms_envelope_correlation": 0.9729060146812324, "1000–3000 Hz_10ms_modulation_mae_db": 0.6480400757644554, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4055671607808855, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0967656334104256, "1000–3000 Hz_40ms_envelope_correlation": 0.9928750697599074, "1000–3000 Hz_40ms_modulation_mae_db": 1.1035948977643628, "3000–6000 Hz_10ms_log_rms_mae_db": 2.574730265313993, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15761202998378138, "3000–6000 Hz_10ms_envelope_correlation": 0.9859832840078414, "3000–6000 Hz_10ms_modulation_mae_db": 0.8636325909681682, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8236626313632598, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08988360053177578, "3000–6000 Hz_40ms_envelope_correlation": 0.9956598752886252, "3000–6000 Hz_40ms_modulation_mae_db": 0.4224392696395916, "mr_stft_db": 6.11797589948748, "band_mae_db": 0.18811313164560786, "rms_gain_error_db": 0.06464503416607585, "output_peak": 0.5639805801849697}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.305591670467864, "全帯域_10ms_rms_envelope_nrmse": 0.08539322634235579, "全帯域_10ms_envelope_correlation": 0.9926576247288519, "全帯域_10ms_modulation_mae_db": 0.36376797135277733, "全帯域_40ms_log_rms_mae_db": 0.6982213656511963, "全帯域_40ms_rms_envelope_nrmse": 0.044857901942282424, "全帯域_40ms_envelope_correlation": 0.9978578568837977, "全帯域_40ms_modulation_mae_db": 0.2616874838930119, "300–1000 Hz_10ms_log_rms_mae_db": 2.0163846012091122, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09268358029915234, "300–1000 Hz_10ms_envelope_correlation": 0.9936317663791305, "300–1000 Hz_10ms_modulation_mae_db": 0.776609411971877, "300–1000 Hz_40ms_log_rms_mae_db": 1.2564615512930057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0486511433669569, "300–1000 Hz_40ms_envelope_correlation": 0.9983567806710265, "300–1000 Hz_40ms_modulation_mae_db": 0.7982652860201208, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1920938956267175, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20562479984351748, "1000–3000 Hz_10ms_envelope_correlation": 0.9683913835827128, "1000–3000 Hz_10ms_modulation_mae_db": 1.0770085195866896, "1000–3000 Hz_40ms_log_rms_mae_db": 1.418860313088012, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11439689723297093, "1000–3000 Hz_40ms_envelope_correlation": 0.9906898223485009, "1000–3000 Hz_40ms_modulation_mae_db": 1.6508013412242395, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5270480135505973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16817325924672835, "3000–6000 Hz_10ms_envelope_correlation": 0.9841619837544738, "3000–6000 Hz_10ms_modulation_mae_db": 0.5229023081708442, "3000–6000 Hz_40ms_log_rms_mae_db": 1.723329383141979, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08804250003585988, "3000–6000 Hz_40ms_envelope_correlation": 0.9956744833936824, "3000–6000 Hz_40ms_modulation_mae_db": 0.5865891278885261, "mr_stft_db": 6.114455361117926, "band_mae_db": 0.16524322435603297, "rms_gain_error_db": -0.024674651454884813, "output_peak": 0.5285112464351422}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1669481286792858, "全帯域_10ms_rms_envelope_nrmse": 0.0834545443351176, "全帯域_10ms_envelope_correlation": 0.9931227942883558, "全帯域_10ms_modulation_mae_db": 0.8680434815703618, "全帯域_40ms_log_rms_mae_db": 0.6602600010166928, "全帯域_40ms_rms_envelope_nrmse": 0.04756087924195902, "全帯域_40ms_envelope_correlation": 0.997789357288483, "全帯域_40ms_modulation_mae_db": 1.0297493924736763, "300–1000 Hz_10ms_log_rms_mae_db": 2.0496314836040543, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10137220246447709, "300–1000 Hz_10ms_envelope_correlation": 0.9928872161931638, "300–1000 Hz_10ms_modulation_mae_db": 0.5239808719953153, "300–1000 Hz_40ms_log_rms_mae_db": 1.2885565773940937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06446401850545393, "300–1000 Hz_40ms_envelope_correlation": 0.997569124688812, "300–1000 Hz_40ms_modulation_mae_db": 0.6019049390965441, "1000–3000 Hz_10ms_log_rms_mae_db": 2.232381475607837, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19891575753197008, "1000–3000 Hz_10ms_envelope_correlation": 0.9701399009501492, "1000–3000 Hz_10ms_modulation_mae_db": 1.2546710052820491, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4005554161666829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10944941930395904, "1000–3000 Hz_40ms_envelope_correlation": 0.990730627377496, "1000–3000 Hz_40ms_modulation_mae_db": 1.8286146999825672, "3000–6000 Hz_10ms_log_rms_mae_db": 2.564351887016661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1397716265634091, "3000–6000 Hz_10ms_envelope_correlation": 0.9893750139298069, "3000–6000 Hz_10ms_modulation_mae_db": 0.4934713352734623, "3000–6000 Hz_40ms_log_rms_mae_db": 1.768081368696164, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08093957243485515, "3000–6000 Hz_40ms_envelope_correlation": 0.9965428454937209, "3000–6000 Hz_40ms_modulation_mae_db": 0.7606263474492178, "mr_stft_db": 6.1105125911357305, "band_mae_db": 0.1634809613837658, "rms_gain_error_db": -0.12954098772524478, "output_peak": 0.48098788748800586}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.111785194945054, "全帯域_10ms_rms_envelope_nrmse": 0.08204413690794851, "全帯域_10ms_envelope_correlation": 0.9932277653527871, "全帯域_10ms_modulation_mae_db": 0.3192494643327429, "全帯域_40ms_log_rms_mae_db": 0.6181115064432916, "全帯域_40ms_rms_envelope_nrmse": 0.048098186134681836, "全帯域_40ms_envelope_correlation": 0.9975635243219262, "全帯域_40ms_modulation_mae_db": 0.75268293937491, "300–1000 Hz_10ms_log_rms_mae_db": 2.0082380000027102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08927528276322666, "300–1000 Hz_10ms_envelope_correlation": 0.9941926686559629, "300–1000 Hz_10ms_modulation_mae_db": 0.681731343100826, "300–1000 Hz_40ms_log_rms_mae_db": 1.2642120153756407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05129470996679136, "300–1000 Hz_40ms_envelope_correlation": 0.9982489017567194, "300–1000 Hz_40ms_modulation_mae_db": 0.746486144006302, "1000–3000 Hz_10ms_log_rms_mae_db": 2.205496709863881, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19152384277204754, "1000–3000 Hz_10ms_envelope_correlation": 0.9731135006260018, "1000–3000 Hz_10ms_modulation_mae_db": 1.8870516183279697, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4106313375365578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11341170161008275, "1000–3000 Hz_40ms_envelope_correlation": 0.9912095892156823, "1000–3000 Hz_40ms_modulation_mae_db": 2.4464914078225717, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6338371672937857, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15083740330312037, "3000–6000 Hz_10ms_envelope_correlation": 0.9877185860751224, "3000–6000 Hz_10ms_modulation_mae_db": 0.35741069840350254, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8670636408102768, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0790048113809979, "3000–6000 Hz_40ms_envelope_correlation": 0.9967864539940783, "3000–6000 Hz_40ms_modulation_mae_db": 0.3543394846065342, "mr_stft_db": 6.103101428886977, "band_mae_db": 0.19286875806999107, "rms_gain_error_db": -0.04887539312965686, "output_peak": 0.5307262785098701}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.179466720327061, "全帯域_10ms_rms_envelope_nrmse": 0.10040522962542314, "全帯域_10ms_envelope_correlation": 0.9898561549362151, "全帯域_10ms_modulation_mae_db": 0.46292437796408614, "全帯域_40ms_log_rms_mae_db": 0.6472115006101405, "全帯域_40ms_rms_envelope_nrmse": 0.05529070736065425, "全帯域_40ms_envelope_correlation": 0.9967579961385414, "全帯域_40ms_modulation_mae_db": 0.7098874456792886, "300–1000 Hz_10ms_log_rms_mae_db": 2.0439505128724234, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10906128573058124, "300–1000 Hz_10ms_envelope_correlation": 0.9911828173523806, "300–1000 Hz_10ms_modulation_mae_db": 0.5107474835403385, "300–1000 Hz_40ms_log_rms_mae_db": 1.2931669050372114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06634358615853674, "300–1000 Hz_40ms_envelope_correlation": 0.9968549436757147, "300–1000 Hz_40ms_modulation_mae_db": 0.5757020818947073, "1000–3000 Hz_10ms_log_rms_mae_db": 2.303250083682783, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2038363513060873, "1000–3000 Hz_10ms_envelope_correlation": 0.9686936987355527, "1000–3000 Hz_10ms_modulation_mae_db": 1.2020719323104878, "1000–3000 Hz_40ms_log_rms_mae_db": 1.461410722499916, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11388428723282366, "1000–3000 Hz_40ms_envelope_correlation": 0.9901520915543772, "1000–3000 Hz_40ms_modulation_mae_db": 1.7539362424391358, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5703229419618063, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1650687899092896, "3000–6000 Hz_10ms_envelope_correlation": 0.9846462672205878, "3000–6000 Hz_10ms_modulation_mae_db": 0.7237774585631701, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7148997429700485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07585538078333356, "3000–6000 Hz_40ms_envelope_correlation": 0.9967826565720839, "3000–6000 Hz_40ms_modulation_mae_db": 1.1783253975714505, "mr_stft_db": 6.127818239474075, "band_mae_db": 0.16063868773548926, "rms_gain_error_db": -0.053530261074541965, "output_peak": 0.5745377767270425}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2433241686240109, "全帯域_10ms_rms_envelope_nrmse": 0.07891642658938855, "全帯域_10ms_envelope_correlation": 0.9939053978656222, "全帯域_10ms_modulation_mae_db": 0.5866605809545692, "全帯域_40ms_log_rms_mae_db": 0.6977508109141789, "全帯域_40ms_rms_envelope_nrmse": 0.04639601007294375, "全帯域_40ms_envelope_correlation": 0.9979483685700306, "全帯域_40ms_modulation_mae_db": 0.4457773344767845, "300–1000 Hz_10ms_log_rms_mae_db": 2.0048234489622123, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08628173273523876, "300–1000 Hz_10ms_envelope_correlation": 0.9946346206569233, "300–1000 Hz_10ms_modulation_mae_db": 0.6298645685495957, "300–1000 Hz_40ms_log_rms_mae_db": 1.2443876035471364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0516966881433936, "300–1000 Hz_40ms_envelope_correlation": 0.9982463806633054, "300–1000 Hz_40ms_modulation_mae_db": 0.7616870560954693, "1000–3000 Hz_10ms_log_rms_mae_db": 2.254965681015991, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18482718040062918, "1000–3000 Hz_10ms_envelope_correlation": 0.9746135857472085, "1000–3000 Hz_10ms_modulation_mae_db": 1.3964657986274343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4629875934979926, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1066309053869517, "1000–3000 Hz_40ms_envelope_correlation": 0.9917544558789361, "1000–3000 Hz_40ms_modulation_mae_db": 1.876423917234763, "3000–6000 Hz_10ms_log_rms_mae_db": 2.615143769267267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17049323609234138, "3000–6000 Hz_10ms_envelope_correlation": 0.9840601521042501, "3000–6000 Hz_10ms_modulation_mae_db": 0.741902917472887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8394727342691262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10014079970542598, "3000–6000 Hz_40ms_envelope_correlation": 0.9954261501316822, "3000–6000 Hz_40ms_modulation_mae_db": 1.0070452608359222, "mr_stft_db": 6.131436599556051, "band_mae_db": 0.18238036521064116, "rms_gain_error_db": -0.14103249207991966, "output_peak": 0.5192784696457745}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.357376236206689, "spectral_convergence": 0.2603084398130343, "spectral_convergence_status": "ok", "amplitude_floor": 2.0096490980240517e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.0877654713568985, "spectral_convergence": 0.2532754278514082, "spectral_convergence_status": "ok", "amplitude_floor": 1.509105853911261e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.917454236673003, "spectral_convergence": 0.2434818441892256, "spectral_convergence_status": "ok", "amplitude_floor": 1.2442618586634314e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017656001839806658, "output_power": 0.001796406664128952, "delta_db": 0.07512292087352589, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005544968065872631, "output_power": 0.005357334117864037, "delta_db": -0.14950316720221243, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005978914392629673, "output_power": 0.0005874914690498939, "delta_db": -0.07620770534350502, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.426968976514186e-05, "output_power": 7.837933247692364e-05, "delta_db": -0.314698346312356, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.40147911812626e-05, "output_power": 6.248917862461649e-05, "delta_db": -0.10475516927720406, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00024101533193812626, "output_power": 0.0002356393197582143, "delta_db": -0.09796910294856556, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1363760686595615, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09525701573824544, "pitch": null, "output_rms_dbfs": -20.90564858266646, "output_peak": 0.48536580763254666, "reference_rms_dbfs": -20.810391566928214, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.12086531474553}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026421160622171636, "log_rms_mae_db": 1.2091614625908362, "rms_envelope_nrmse": 0.09263806457453595, "envelope_correlation": 0.9914084135146282, "rms_gain_error_db": -0.0951722580603421, "reference_mean_power": 0.008303652784602253, "output_mean_power": 0.008123664051388414}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.437173268209281e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017543654513432086, "output_power": 0.0016289657261384521, "reference_active": true, "delta_db": -0.3220811947175301}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013368683738807852, "output_power": 0.000126388022277443, "reference_active": true, "delta_db": -0.24382731312654704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.235741942938941e-05, "output_power": 7.055609566372765e-05, "reference_active": true, "delta_db": -0.10948529294518858}], "active_band_count": 3, "active_band_mae_db": 0.22513126692975524}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022872947944797367, "log_rms_mae_db": 0.652280267845908, "rms_envelope_nrmse": 0.04673534353864879, "envelope_correlation": 0.9977739400976, "rms_gain_error_db": -0.09538863880462312, "reference_mean_power": 0.008338532782066458, "output_mean_power": 0.008157381555768239}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.164744003117204e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015799393891106446, "output_power": 0.0014768205018893359, "reference_active": true, "delta_db": -0.29312713728366396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.374442374979845e-05, "output_power": 5.91032544867315e-05, "reference_active": true, "delta_db": -0.3283080373363531}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.293312028286094e-06, "output_power": 3.586084413522387e-06, "reference_active": true, "delta_db": 0.3698762794143351}], "active_band_count": 3, "active_band_mae_db": 0.3304371513447841}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024243770094671305, "log_rms_mae_db": 1.9934982101704624, "rms_envelope_nrmse": 0.08596175119690376, "envelope_correlation": 0.9945767890842785, "rms_gain_error_db": -0.16191922048475318, "reference_mean_power": 0.003959461759548422, "output_mean_power": 0.003814558071774686}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.015177817247118e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001136879031892713, "output_power": 0.0010223893599119163, "reference_active": true, "delta_db": -0.4609793551670666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001232393146611943, "output_power": 0.00010659606502561884, "reference_active": true, "delta_db": -0.6300810143793827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0364107060899577e-05, "output_power": 2.3901154462694846e-05, "reference_active": true, "delta_db": -1.0394163548732442}], "active_band_count": 3, "active_band_mae_db": 0.7101589081398979}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020477863430336455, "log_rms_mae_db": 1.2860942738797725, "rms_envelope_nrmse": 0.054928713464352415, "envelope_correlation": 0.9978355402908659, "rms_gain_error_db": -0.1619237255291796, "reference_mean_power": 0.003979363820398493, "output_mean_power": 0.0038337278037714113}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8942412803353574e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010558685473009428, "output_power": 0.0009448464391808946, "reference_active": true, "delta_db": -0.48248622413542486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.091449836890923e-05, "output_power": 4.5058240783359574e-05, "reference_active": true, "delta_db": -0.5306723778775297}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8755256559270567e-06, "output_power": 1.4212147276775306e-06, "reference_active": true, "delta_db": -1.204633100621678}], "active_band_count": 3, "active_band_mae_db": 0.7392639008782108}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5081112731014556e-05, "log_rms_mae_db": 2.2181187214947573, "rms_envelope_nrmse": 0.20553115197829427, "envelope_correlation": 0.9682500894378269, "rms_gain_error_db": -0.2594898562783118, "reference_mean_power": 0.00010911551831198975, "output_mean_power": 0.00010278684465797067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.836309277153812e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5256828161206645e-05, "output_power": 2.304651327971471e-05, "reference_active": true, "delta_db": -0.3977357942571421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.88342602252987e-06, "output_power": 7.737357397474692e-06, "reference_active": true, "delta_db": -1.0631485803538279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.436572483663522e-06, "output_power": 2.6686075168414475e-06, "reference_active": true, "delta_db": -3.0904047768624965}], "active_band_count": 3, "active_band_mae_db": 1.5170963838244889}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.4434083199144116e-05, "log_rms_mae_db": 1.428897688090704, "rms_envelope_nrmse": 0.11679962683952391, "envelope_correlation": 0.9897861524398393, "rms_gain_error_db": -0.25974798021303763, "reference_mean_power": 0.0001096329653054558, "output_mean_power": 0.00010326814190262851}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.313909407445067e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3642314296732694e-05, "output_power": 2.130331992485956e-05, "reference_active": true, "delta_db": -0.4524269698851576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.77067085832153e-06, "output_power": 4.529345365988477e-06, "reference_active": true, "delta_db": -1.745962647872519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.651384635522461e-07, "output_power": 1.5179370636926585e-07, "reference_active": true, "delta_db": -4.863284886476568}], "active_band_count": 3, "active_band_mae_db": 2.353891501411415}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010577651466639849, "log_rms_mae_db": 2.573325500178391, "rms_envelope_nrmse": 0.14409776953087752, "envelope_correlation": 0.9882638027738571, "rms_gain_error_db": -0.08555711056214688, "reference_mean_power": 0.00018020158938811219, "output_mean_power": 0.00017668631214759454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.705326923064497e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.613613486254781e-05, "output_power": 1.4810074271685992e-05, "reference_active": true, "delta_db": -0.3724227847625056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.942687703713047e-06, "output_power": 7.645516953070173e-06, "reference_active": true, "delta_db": -0.6806120895308834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.043858273628482e-06, "output_power": 7.753122252461439e-07, "reference_active": true, "delta_db": -1.2916490573787405}], "active_band_count": 3, "active_band_mae_db": 0.7815613105573765}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.602886913996206e-05, "log_rms_mae_db": 1.7600680262945863, "rms_envelope_nrmse": 0.0829916554452066, "envelope_correlation": 0.9961610991413019, "rms_gain_error_db": -0.10046391628360599, "reference_mean_power": 0.0001775816262598491, "output_mean_power": 0.00017352083833641958}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.444034789789241e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7728340440048457e-05, "output_power": 1.5815525049855187e-05, "reference_active": true, "delta_db": -0.4958446858661059}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.955564320473026e-06, "output_power": 4.278681647892985e-06, "reference_active": true, "delta_db": -0.637831430811179}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.260796671427193e-08, "output_power": 6.008242749458406e-08, "reference_active": true, "delta_db": -1.8790087796933375}], "active_band_count": 3, "active_band_mae_db": 1.0042282987902074}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8262802617322141, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "117a5ed94b1b9608062aacffbac2c6254561c4302ac16f45aff5111da26a7e62", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9708559081247904, "standard_deviation": 0.0024344385113355027, "minimum": 0.9682500894378269, "maximum": 0.9746135857472085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2263149889351803, "standard_deviation": 0.03732066284009666, "minimum": 2.1920938956267175, "maximum": 2.303250083682783, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2724305581931488, "standard_deviation": 0.35697298967636615, "minimum": 0.6480400757644554, "maximum": 1.8870516183279697, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19717891057566794, "standard_deviation": 0.007825980792365945, "minimum": 0.18482718040062918, "maximum": 0.20562479984351748, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9910649892870376, "standard_deviation": 0.0009680708039517626, "minimum": 0.9897861524398393, "maximum": 0.9928750697599074, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4266931683422168, "standard_deviation": 0.02384628054009676, "minimum": 1.4005554161666829, "maximum": 1.4629875934979926, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.8369200925901303, "standard_deviation": 0.4215415159488495, "minimum": 1.1035948977643628, "maximum": 2.4464914078225717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10996776771898643, "standard_deviation": 0.0063430206641569686, "minimum": 0.0967656334104256, "maximum": 0.11679962683952391, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9865593960192853, "standard_deviation": 0.0021069600687988977, "minimum": 0.9840601521042501, "maximum": 0.9893750139298069, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.574330821834466, "standard_deviation": 0.035930693958998505, "minimum": 2.5270480135505973, "maximum": 2.6338371672937857, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6170490731830364, "standard_deviation": 0.1827267198934229, "minimum": 0.35741069840350254, "maximum": 0.8636325909681682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15502046630703403, "standard_deviation": 0.012002765569049195, "minimum": 0.1397716265634091, "maximum": 0.17049323609234138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9961974338232911, "standard_deviation": 0.0005462623620631663, "minimum": 0.9954261501316822, "maximum": 0.9967864539940783, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7831218048819761, "standard_deviation": 0.05486231469423446, "minimum": 1.7148997429700485, "maximum": 1.8670636408102768, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7513783293525769, "standard_deviation": 0.29461216447083693, "minimum": 0.3543394846065342, "maximum": 1.1783253975714505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08448246649592406, "standard_deviation": 0.007894639937239193, "minimum": 0.07585538078333356, "maximum": 0.10014079970542598, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9935665267414332, "standard_deviation": 0.0012199199948970342, "minimum": 0.9911828173523806, "maximum": 0.9946346206569233, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0050664419562736, "standard_deviation": 0.03471895892272046, "minimum": 1.9401073872762211, "maximum": 2.0496314836040543, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6501372545978157, "standard_deviation": 0.10579927231094972, "minimum": 0.5107474835403385, "maximum": 0.7832719222684114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.093690816368837, "standard_deviation": 0.008306508532830983, "minimum": 0.08596175119690376, "maximum": 0.10906128573058124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9978658251249068, "standard_deviation": 0.0005565984960475623, "minimum": 0.9968549436757147, "maximum": 0.9984182034580316, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2516981784199357, "standard_deviation": 0.050207412917678725, "minimum": 1.1373199061264547, "maximum": 1.2931669050372114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7195459563423865, "standard_deviation": 0.08473053340000486, "minimum": 0.5757020818947073, "maximum": 0.7998635036453475, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05586783863412038, "standard_deviation": 0.006480922918185781, "minimum": 0.0486511433669569, "maximum": 0.06634358615853674, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17169831323446555, "standard_deviation": 0.018801279630642893, "minimum": 0.1363760686595615, "maximum": 0.19286875806999107, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.116452303340111, "standard_deviation": 0.010099563661907836, "minimum": 6.103101428886977, "maximum": 6.131436599556051, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5284814091584801, "standard_deviation": 0.033468797296424095, "minimum": 0.48098788748800586, "maximum": 0.5745377767270425, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06937008812091605, "standard_deviation": 0.06901446683909243, "minimum": -0.14103249207991966, "maximum": 0.06464503416607585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9914910400988495, "standard_deviation": 0.0034229985629910895, "minimum": 0.9837377980050892, "maximum": 0.9940123720992464, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2008360928241686, "standard_deviation": 0.05845503311445543, "minimum": 1.111785194945054, "maximum": 1.305591670467864, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6647304724844323, "standard_deviation": 0.6992353256873447, "minimum": 0.18411418669517735, "maximum": 2.3079524500759883, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09120516611299967, "standard_deviation": 0.01688435527849093, "minimum": 0.07806788193923136, "maximum": 0.12872181858999654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9967145687705659, "standard_deviation": 0.0028937678160766596, "minimum": 0.9896430730158063, "maximum": 0.9983824338483422, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6480041182670543, "standard_deviation": 0.03817671643438508, "minimum": 0.5899116473432024, "maximum": 0.6982213656511963, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.867074984228572, "standard_deviation": 0.8762950078855893, "minimum": 0.2616874838930119, "maximum": 2.944328611302613, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05372030467538773, "standard_deviation": 0.01890886325145713, "minimum": 0.0413202964821814, "maximum": 0.09950311262975038, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0013", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.2595174312591553, "predicted_mos": 3.2595174312591553, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1be00f8cef78f01f89a16a1f2494130c88a00c388f7ff77b443a862361b602d6", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.758177500218153, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.241441697657637, "全帯域_10ms_rms_envelope_nrmse": 0.08585380906036166, "全帯域_10ms_envelope_correlation": 0.9926764647378464, "全帯域_10ms_modulation_mae_db": 0.11905780179554883, "全帯域_40ms_log_rms_mae_db": 0.6922407974287353, "全帯域_40ms_rms_envelope_nrmse": 0.045304313092125674, "全帯域_40ms_envelope_correlation": 0.9979485153089184, "全帯域_40ms_modulation_mae_db": 0.23500468485929318, "300–1000 Hz_10ms_log_rms_mae_db": 1.5747166431409318, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08826710413048808, "300–1000 Hz_10ms_envelope_correlation": 0.9943293644632113, "300–1000 Hz_10ms_modulation_mae_db": 0.31097989555740274, "300–1000 Hz_40ms_log_rms_mae_db": 0.9369105953284946, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05014543836223448, "300–1000 Hz_40ms_envelope_correlation": 0.9983437864152298, "300–1000 Hz_40ms_modulation_mae_db": 0.4433561400012809, "1000–3000 Hz_10ms_log_rms_mae_db": 1.589952009069364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16262235807244535, "1000–3000 Hz_10ms_envelope_correlation": 0.9827352689869927, "1000–3000 Hz_10ms_modulation_mae_db": 1.5633504750986489, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0459744670414, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10804302215570073, "1000–3000 Hz_40ms_envelope_correlation": 0.9942487233451112, "1000–3000 Hz_40ms_modulation_mae_db": 1.9931799698299286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.630518357547886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14783044101487502, "3000–6000 Hz_10ms_envelope_correlation": 0.9876153248373811, "3000–6000 Hz_10ms_modulation_mae_db": 0.20587405003617443, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1297944000912066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08476503841557635, "3000–6000 Hz_40ms_envelope_correlation": 0.9960835783750739, "3000–6000 Hz_40ms_modulation_mae_db": 0.5665155956552964, "mr_stft_db": 5.543681051886051, "band_mae_db": 0.24053769286727852, "rms_gain_error_db": -0.10636439033612062, "output_peak": 0.5220250830152966}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2969714067522917, "全帯域_10ms_rms_envelope_nrmse": 0.08203165942843706, "全帯域_10ms_envelope_correlation": 0.9932668335987951, "全帯域_10ms_modulation_mae_db": 0.2602250978524188, "全帯域_40ms_log_rms_mae_db": 0.703789720827162, "全帯域_40ms_rms_envelope_nrmse": 0.04391431580447964, "全帯域_40ms_envelope_correlation": 0.9979897144447243, "全帯域_40ms_modulation_mae_db": 0.5600211156799461, "300–1000 Hz_10ms_log_rms_mae_db": 1.5446004689229471, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0842351973133449, "300–1000 Hz_10ms_envelope_correlation": 0.99468824027882, "300–1000 Hz_10ms_modulation_mae_db": 0.5340235720236538, "300–1000 Hz_40ms_log_rms_mae_db": 0.9156076477353158, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048357592162290956, "300–1000 Hz_40ms_envelope_correlation": 0.9982617007839234, "300–1000 Hz_40ms_modulation_mae_db": 0.4359713593972377, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6228282427590024, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1797306380349767, "1000–3000 Hz_10ms_envelope_correlation": 0.97576843951017, "1000–3000 Hz_10ms_modulation_mae_db": 0.6325117190664248, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0797617580746934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11143450233156944, "1000–3000 Hz_40ms_envelope_correlation": 0.990508454244232, "1000–3000 Hz_40ms_modulation_mae_db": 0.5444592549953737, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6254040564685703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14543213425850632, "3000–6000 Hz_10ms_envelope_correlation": 0.9881062706158581, "3000–6000 Hz_10ms_modulation_mae_db": 0.6829846737160908, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1473335838091712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09713438011215902, "3000–6000 Hz_40ms_envelope_correlation": 0.9946812186787147, "3000–6000 Hz_40ms_modulation_mae_db": 1.0003806730789513, "mr_stft_db": 5.569970007042074, "band_mae_db": 0.16936577085505292, "rms_gain_error_db": -0.06546348046669245, "output_peak": 0.527937825644082}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2429359727720157, "全帯域_10ms_rms_envelope_nrmse": 0.08810712574227693, "全帯域_10ms_envelope_correlation": 0.9922319855891405, "全帯域_10ms_modulation_mae_db": 0.5669085591494714, "全帯域_40ms_log_rms_mae_db": 0.6604744111254076, "全帯域_40ms_rms_envelope_nrmse": 0.04364443852225309, "全帯域_40ms_envelope_correlation": 0.9980473839163164, "全帯域_40ms_modulation_mae_db": 0.3475275015382491, "300–1000 Hz_10ms_log_rms_mae_db": 1.4881414544624902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08163653264585266, "300–1000 Hz_10ms_envelope_correlation": 0.995032899349282, "300–1000 Hz_10ms_modulation_mae_db": 0.468990065545195, "300–1000 Hz_40ms_log_rms_mae_db": 0.853922419458804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04769484586495526, "300–1000 Hz_40ms_envelope_correlation": 0.9983284454041003, "300–1000 Hz_40ms_modulation_mae_db": 0.39449367275155867, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5888571501976805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.184519554202917, "1000–3000 Hz_10ms_envelope_correlation": 0.9745806413116758, "1000–3000 Hz_10ms_modulation_mae_db": 0.8742299070613968, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9909377709275432, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12298877549016302, "1000–3000 Hz_40ms_envelope_correlation": 0.9884957249526661, "1000–3000 Hz_40ms_modulation_mae_db": 0.4903561347496456, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6085429756008114, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12520162860732628, "3000–6000 Hz_10ms_envelope_correlation": 0.9911015937270015, "3000–6000 Hz_10ms_modulation_mae_db": 0.8807870859380018, "3000–6000 Hz_40ms_log_rms_mae_db": 1.104484008857171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06411924895347612, "3000–6000 Hz_40ms_envelope_correlation": 0.9976464594616251, "3000–6000 Hz_40ms_modulation_mae_db": 1.0516183101132144, "mr_stft_db": 5.54801731249193, "band_mae_db": 0.148007712585857, "rms_gain_error_db": -0.0841931193607459, "output_peak": 0.6464568415122693}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2481649875786134, "全帯域_10ms_rms_envelope_nrmse": 0.10713673445083963, "全帯域_10ms_envelope_correlation": 0.9884944413336328, "全帯域_10ms_modulation_mae_db": 1.3928372483609934, "全帯域_40ms_log_rms_mae_db": 0.7019509261687237, "全帯域_40ms_rms_envelope_nrmse": 0.07285058626583363, "全帯域_40ms_envelope_correlation": 0.994372679767706, "全帯域_40ms_modulation_mae_db": 1.577766073687152, "300–1000 Hz_10ms_log_rms_mae_db": 1.5170961623105534, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09975615941122895, "300–1000 Hz_10ms_envelope_correlation": 0.9929111154059053, "300–1000 Hz_10ms_modulation_mae_db": 0.680089133653408, "300–1000 Hz_40ms_log_rms_mae_db": 0.8878500061575169, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06161369942191467, "300–1000 Hz_40ms_envelope_correlation": 0.9976060567942368, "300–1000 Hz_40ms_modulation_mae_db": 0.6475891795583144, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5602122221593209, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15861941234324, "1000–3000 Hz_10ms_envelope_correlation": 0.9823150169589046, "1000–3000 Hz_10ms_modulation_mae_db": 0.6540229526231122, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9626728903478685, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09893307421831107, "1000–3000 Hz_40ms_envelope_correlation": 0.9940805315451616, "1000–3000 Hz_40ms_modulation_mae_db": 0.8935900774321149, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6854622801505437, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1552590913097463, "3000–6000 Hz_10ms_envelope_correlation": 0.9862524350835297, "3000–6000 Hz_10ms_modulation_mae_db": 1.144875388545193, "3000–6000 Hz_40ms_log_rms_mae_db": 1.165403867732378, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08963429826661719, "3000–6000 Hz_40ms_envelope_correlation": 0.9953786413532827, "3000–6000 Hz_40ms_modulation_mae_db": 1.227875365909536, "mr_stft_db": 5.561858374025743, "band_mae_db": 0.28259422006941354, "rms_gain_error_db": -0.07850300325680826, "output_peak": 0.5219993875767621}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2424212044310414, "全帯域_10ms_rms_envelope_nrmse": 0.08788342755395584, "全帯域_10ms_envelope_correlation": 0.9922887046300698, "全帯域_10ms_modulation_mae_db": 0.8828672400789745, "全帯域_40ms_log_rms_mae_db": 0.6672811429062562, "全帯域_40ms_rms_envelope_nrmse": 0.04974250357099951, "全帯域_40ms_envelope_correlation": 0.9974754868580287, "全帯域_40ms_modulation_mae_db": 0.9312327825486668, "300–1000 Hz_10ms_log_rms_mae_db": 1.576530948214658, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09118346520143343, "300–1000 Hz_10ms_envelope_correlation": 0.9938329006747869, "300–1000 Hz_10ms_modulation_mae_db": 0.2529634048422036, "300–1000 Hz_40ms_log_rms_mae_db": 0.9082401025546822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052676340021891435, "300–1000 Hz_40ms_envelope_correlation": 0.9980172225939049, "300–1000 Hz_40ms_modulation_mae_db": 0.34549231176828393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6482160717404426, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1823807219731093, "1000–3000 Hz_10ms_envelope_correlation": 0.976263023374331, "1000–3000 Hz_10ms_modulation_mae_db": 0.8136967484667522, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0572472439446443, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11651204326385044, "1000–3000 Hz_40ms_envelope_correlation": 0.9914081480152579, "1000–3000 Hz_40ms_modulation_mae_db": 0.8599291748998699, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6381384509268109, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1393491992836777, "3000–6000 Hz_10ms_envelope_correlation": 0.9896209047364851, "3000–6000 Hz_10ms_modulation_mae_db": 0.6638267191181976, "3000–6000 Hz_40ms_log_rms_mae_db": 1.168299332606751, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07618349573860002, "3000–6000 Hz_40ms_envelope_correlation": 0.9971444696455531, "3000–6000 Hz_40ms_modulation_mae_db": 0.5989798718698832, "mr_stft_db": 5.559869415865266, "band_mae_db": 0.19054123555692204, "rms_gain_error_db": -0.10159212588551719, "output_peak": 0.5422635845249025}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.192116138853017, "全帯域_10ms_rms_envelope_nrmse": 0.07789661207582288, "全帯域_10ms_envelope_correlation": 0.9939487468347955, "全帯域_10ms_modulation_mae_db": 0.7803390466629424, "全帯域_40ms_log_rms_mae_db": 0.6357636786901046, "全帯域_40ms_rms_envelope_nrmse": 0.0416463780618911, "全帯域_40ms_envelope_correlation": 0.9982497888735774, "全帯域_40ms_modulation_mae_db": 0.6924615640096778, "300–1000 Hz_10ms_log_rms_mae_db": 1.520525081594742, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08207514200355664, "300–1000 Hz_10ms_envelope_correlation": 0.9950477697996043, "300–1000 Hz_10ms_modulation_mae_db": 0.8195840790528944, "300–1000 Hz_40ms_log_rms_mae_db": 0.8366183218804836, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04505907816590692, "300–1000 Hz_40ms_envelope_correlation": 0.9986137798652318, "300–1000 Hz_40ms_modulation_mae_db": 0.9122583218700032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.624906516780361, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18019609207363074, "1000–3000 Hz_10ms_envelope_correlation": 0.978084092890436, "1000–3000 Hz_10ms_modulation_mae_db": 2.3691638597639817, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0250090843039947, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12356664304618152, "1000–3000 Hz_40ms_envelope_correlation": 0.9914189534644933, "1000–3000 Hz_40ms_modulation_mae_db": 3.1593074357065727, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6417532811759745, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13545557474974196, "3000–6000 Hz_10ms_envelope_correlation": 0.989547520210969, "3000–6000 Hz_10ms_modulation_mae_db": 0.45515913931005914, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1793410695185333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06609644936668216, "3000–6000 Hz_40ms_envelope_correlation": 0.997572882208921, "3000–6000 Hz_40ms_modulation_mae_db": 0.4920013341796006, "mr_stft_db": 5.554973734071062, "band_mae_db": 0.25908599350826894, "rms_gain_error_db": -0.09168355733348452, "output_peak": 0.49226287272100155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.234039924547863, "全帯域_10ms_rms_envelope_nrmse": 0.07515072304399778, "全帯域_10ms_envelope_correlation": 0.9943553055934367, "全帯域_10ms_modulation_mae_db": 0.5314272386948249, "全帯域_40ms_log_rms_mae_db": 0.6390536299194992, "全帯域_40ms_rms_envelope_nrmse": 0.041121342448989275, "全帯域_40ms_envelope_correlation": 0.9982132296356825, "全帯域_40ms_modulation_mae_db": 0.4823049219780093, "300–1000 Hz_10ms_log_rms_mae_db": 1.51953551409762, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08287989979704952, "300–1000 Hz_10ms_envelope_correlation": 0.9948154573846265, "300–1000 Hz_10ms_modulation_mae_db": 0.34284488558757187, "300–1000 Hz_40ms_log_rms_mae_db": 0.8727844131206439, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0478723152704252, "300–1000 Hz_40ms_envelope_correlation": 0.9982378367369139, "300–1000 Hz_40ms_modulation_mae_db": 0.14585957594837046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6187903818717053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18859752426368376, "1000–3000 Hz_10ms_envelope_correlation": 0.9751724807629127, "1000–3000 Hz_10ms_modulation_mae_db": 1.372242015325558, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0860339223633946, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11744374298293322, "1000–3000 Hz_40ms_envelope_correlation": 0.9922292229226723, "1000–3000 Hz_40ms_modulation_mae_db": 2.0158624858971215, "3000–6000 Hz_10ms_log_rms_mae_db": 1.593736044953767, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1205433825591268, "3000–6000 Hz_10ms_envelope_correlation": 0.9917355876485374, "3000–6000 Hz_10ms_modulation_mae_db": 0.5956861023588093, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1082701638483161, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05639319698730992, "3000–6000 Hz_40ms_envelope_correlation": 0.9982103791982048, "3000–6000 Hz_40ms_modulation_mae_db": 0.6815596615213528, "mr_stft_db": 5.559197253651559, "band_mae_db": 0.2871854332496765, "rms_gain_error_db": -0.020408645941487374, "output_peak": 0.5237353632522493}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.21482599017718, "全帯域_10ms_rms_envelope_nrmse": 0.07989103765155552, "全帯域_10ms_envelope_correlation": 0.9937316929555408, "全帯域_10ms_modulation_mae_db": 0.4956262391771518, "全帯域_40ms_log_rms_mae_db": 0.646601371560518, "全帯域_40ms_rms_envelope_nrmse": 0.042209691958666785, "全帯域_40ms_envelope_correlation": 0.9983151993741132, "全帯域_40ms_modulation_mae_db": 0.6945769516201304, "300–1000 Hz_10ms_log_rms_mae_db": 1.5797168665640924, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0897612331164903, "300–1000 Hz_10ms_envelope_correlation": 0.9941644324535955, "300–1000 Hz_10ms_modulation_mae_db": 0.5298921065850843, "300–1000 Hz_40ms_log_rms_mae_db": 0.8800734678922796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05617425280378772, "300–1000 Hz_40ms_envelope_correlation": 0.9978645471307396, "300–1000 Hz_40ms_modulation_mae_db": 0.5802764114062915, "1000–3000 Hz_10ms_log_rms_mae_db": 1.617936020624093, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16891878616325084, "1000–3000 Hz_10ms_envelope_correlation": 0.9806935990184603, "1000–3000 Hz_10ms_modulation_mae_db": 0.5827432058567695, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0655268493804775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10815994373880078, "1000–3000 Hz_40ms_envelope_correlation": 0.9938117480273986, "1000–3000 Hz_40ms_modulation_mae_db": 0.980211814238697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.665373602880492, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16413404931330344, "3000–6000 Hz_10ms_envelope_correlation": 0.9853793092450485, "3000–6000 Hz_10ms_modulation_mae_db": 0.5354211911647797, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1478637232558484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10763311816144609, "3000–6000 Hz_40ms_envelope_correlation": 0.9945975459008475, "3000–6000 Hz_40ms_modulation_mae_db": 0.6186146704632686, "mr_stft_db": 5.556227986150273, "band_mae_db": 0.28783616866845857, "rms_gain_error_db": -0.1308392736457479, "output_peak": 0.504058991257542}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.650219623221467, "spectral_convergence": 0.25161723069776837, "spectral_convergence_status": "ok", "amplitude_floor": 2.0096490980240517e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.410045892076361, "spectral_convergence": 0.22655267144411634, "spectral_convergence_status": "ok", "amplitude_floor": 1.509105853911261e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.570777640360327, "spectral_convergence": 0.2291670534779407, "spectral_convergence_status": "ok", "amplitude_floor": 1.2442618586634314e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017656001839806658, "output_power": 0.0017858112280158935, "delta_db": 0.049431837892342526, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005544968065872631, "output_power": 0.0053580822024538914, "delta_db": -0.1488967717058799, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005978914392629673, "output_power": 0.0005894299829979447, "delta_db": -0.06190111469379336, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.426968976514186e-05, "output_power": 7.245152995591147e-05, "delta_db": -0.6562383420734879, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.40147911812626e-05, "output_power": 5.879631942503526e-05, "delta_db": -0.36930192499812614, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00024101533193812626, "output_power": 0.00023243367777246753, "delta_db": -0.1574561658400414, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24053769286727852, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10636439033612062, "pitch": null, "output_rms_dbfs": -20.916755957264336, "output_peak": 0.5220250830152966, "reference_rms_dbfs": -20.810391566928214, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.543681051886051}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026421160622171636, "log_rms_mae_db": 1.241441697657637, "rms_envelope_nrmse": 0.08585380906036166, "envelope_correlation": 0.9926764647378464, "rms_gain_error_db": -0.1065381292305567, "reference_mean_power": 0.008303652784602253, "output_mean_power": 0.008102431499048405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.437173268209281e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017543654513432086, "output_power": 0.0016997118188915308, "reference_active": true, "delta_db": -0.13744771932157848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013368683738807852, "output_power": 0.00014043515997585305, "reference_active": true, "delta_db": 0.21387203956937773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.235741942938941e-05, "output_power": 7.245501227138346e-05, "reference_active": true, "delta_db": 0.005853646495690253}], "active_band_count": 3, "active_band_mae_db": 0.11905780179554883}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022872947944797367, "log_rms_mae_db": 0.6922407974287353, "rms_envelope_nrmse": 0.045304313092125674, "envelope_correlation": 0.9979485153089184, "rms_gain_error_db": -0.1069076092391873, "reference_mean_power": 0.008338532782066458, "output_mean_power": 0.008135774066415734}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.164744003117204e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015799393891106446, "output_power": 0.0015383259281196231, "reference_active": true, "delta_db": -0.11592066489459625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.374442374979845e-05, "output_power": 6.217529943981522e-05, "reference_active": true, "delta_db": -0.10824313915113917}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.293312028286094e-06, "output_power": 2.948138316671424e-06, "reference_active": true, "delta_db": -0.48085025053214414}], "active_band_count": 3, "active_band_mae_db": 0.23500468485929318}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024243770094671305, "log_rms_mae_db": 1.5747166431409318, "rms_envelope_nrmse": 0.08826710413048808, "envelope_correlation": 0.9943293644632113, "rms_gain_error_db": -0.19558261455622994, "reference_mean_power": 0.003959461759548422, "output_mean_power": 0.003785104651248439}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.015177817247118e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001136879031892713, "output_power": 0.0010602297317765426, "reference_active": true, "delta_db": -0.30314277736403195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001232393146611943, "output_power": 0.00011045801381546434, "reference_active": true, "delta_db": -0.47552044766973994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0364107060899577e-05, "output_power": 2.9304402303032707e-05, "reference_active": true, "delta_db": -0.15427646163843634}], "active_band_count": 3, "active_band_mae_db": 0.31097989555740274}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020477863430336455, "log_rms_mae_db": 0.9369105953284946, "rms_envelope_nrmse": 0.05014543836223448, "envelope_correlation": 0.9983437864152298, "rms_gain_error_db": -0.19558322327387304, "reference_mean_power": 0.003979363820398493, "output_mean_power": 0.003804129780545863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8942412803353574e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010558685473009428, "output_power": 0.0009835200350206826, "reference_active": true, "delta_db": -0.30826641866526144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.091449836890923e-05, "output_power": 4.686403820525429e-05, "reference_active": true, "delta_db": -0.36001760882262646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8755256559270567e-06, "output_power": 1.6104400783909557e-06, "reference_active": true, "delta_db": -0.6617843925159548}], "active_band_count": 3, "active_band_mae_db": 0.4433561400012809}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5081112731014556e-05, "log_rms_mae_db": 1.589952009069364, "rms_envelope_nrmse": 0.16262235807244535, "envelope_correlation": 0.9827352689869927, "rms_gain_error_db": -0.6349329359846589, "reference_mean_power": 0.00010911551831198975, "output_mean_power": 9.42742804558402e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.836309277153812e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5256828161206645e-05, "output_power": 2.0559650572944834e-05, "reference_active": true, "delta_db": -0.8936308024087612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.88342602252987e-06, "output_power": 7.96908148124894e-06, "reference_active": true, "delta_db": -0.934992485531926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.436572483663522e-06, "output_power": 2.81308185179784e-06, "reference_active": true, "delta_db": -2.861428137355259}], "active_band_count": 3, "active_band_mae_db": 1.5633504750986489}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.4434083199144116e-05, "log_rms_mae_db": 1.0459744670414, "rms_envelope_nrmse": 0.10804302215570073, "envelope_correlation": 0.9942487233451112, "rms_gain_error_db": -0.6355072097160935, "reference_mean_power": 0.0001096329653054558, "output_mean_power": 9.470882310556041e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.313909407445067e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3642314296732694e-05, "output_power": 1.9064169701640228e-05, "reference_active": true, "delta_db": -0.9347209119721425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.77067085832153e-06, "output_power": 4.961803175495907e-06, "reference_active": true, "delta_db": -1.3499216930247646}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.651384635522461e-07, "output_power": 1.986518086638255e-07, "reference_active": true, "delta_db": -3.694897304492879}], "active_band_count": 3, "active_band_mae_db": 1.9931799698299286}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010577651466639849, "log_rms_mae_db": 1.630518357547886, "rms_envelope_nrmse": 0.14783044101487502, "envelope_correlation": 0.9876153248373811, "rms_gain_error_db": -0.2017965949117269, "reference_mean_power": 0.00018020158938811219, "output_mean_power": 0.0001720200056800566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.705326923064497e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.613613486254781e-05, "output_power": 1.6691492450693494e-05, "reference_active": true, "delta_db": 0.14695655394223828}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.942687703713047e-06, "output_power": 8.409283775034023e-06, "reference_active": true, "delta_db": -0.2670905640915155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.043858273628482e-06, "output_power": 9.960566391475324e-07, "reference_active": true, "delta_db": -0.20357503207476946}], "active_band_count": 3, "active_band_mae_db": 0.20587405003617443}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.602886913996206e-05, "log_rms_mae_db": 1.1297944000912066, "rms_envelope_nrmse": 0.08476503841557635, "envelope_correlation": 0.9960835783750739, "rms_gain_error_db": -0.20741146636919372, "reference_mean_power": 0.0001775816262598491, "output_mean_power": 0.00016929997041740069}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.444034789789241e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7728340440048457e-05, "output_power": 1.796996460515032e-05, "reference_active": true, "delta_db": 0.05879138742072971}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.955564320473026e-06, "output_power": 4.724223361870984e-06, "reference_active": true, "delta_db": -0.20762694486809669}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.260796671427193e-08, "output_power": 6.657872952107811e-08, "reference_active": true, "delta_db": -1.4331284546770628}], "active_band_count": 3, "active_band_mae_db": 0.5665155956552964}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8564990644375797, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "78dc22830f89b0c5b6f73c17b04a831a6fa42546a6a73978d087eef0aacfa41e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9782015703517354, "standard_deviation": 0.0032877146874190997, "minimum": 0.9745806413116758, "maximum": 0.9827352689869927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6089623269002462, "standard_deviation": 0.027533455409762535, "minimum": 1.5602122221593209, "maximum": 1.6482160717404426, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1077451104078304, "standard_deviation": 0.6230161066618264, "minimum": 0.5827432058567695, "maximum": 2.3691638597639817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17569813589090671, "standard_deviation": 0.010915124682503287, "minimum": 0.15861941234324, "maximum": 0.18859752426368376, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.992025188314624, "standard_deviation": 0.001996705244859886, "minimum": 0.9884957249526661, "maximum": 0.9942487233451112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.039145498298002, "standard_deviation": 0.04357320852739659, "minimum": 0.9626728903478685, "maximum": 1.0860339223633946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3671120434686657, "standard_deviation": 0.933611294509642, "minimum": 0.4903561347496456, "maximum": 3.1593074357065727, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11338521840343878, "standard_deviation": 0.00837215361229367, "minimum": 0.09893307421831107, "maximum": 0.12356664304618152, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9886698682631013, "standard_deviation": 0.002238881524684401, "minimum": 0.9853793092450485, "maximum": 0.9917355876485374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.636116131213107, "standard_deviation": 0.02934850577155349, "minimum": 1.593736044953767, "maximum": 1.6854622801505437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6455767937734131, "standard_deviation": 0.28045272109710667, "minimum": 0.20587405003617443, "maximum": 1.144875388545193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14165068763703798, "standard_deviation": 0.014652864352303252, "minimum": 0.1205433825591268, "maximum": 0.16413404931330344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9964143968527779, "standard_deviation": 0.0014195426700605038, "minimum": 0.9945975459008475, "maximum": 0.9982103791982048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.143848768714922, "standard_deviation": 0.027680230957813648, "minimum": 1.104484008857171, "maximum": 1.1793410695185333, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7796931853488879, "standard_deviation": 0.2725432001910645, "minimum": 0.4920013341796006, "maximum": 1.227875365909536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08024490325023337, "standard_deviation": 0.01768605383507741, "minimum": 0.05639319698730992, "maximum": 0.10763311816144609, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943527724762289, "standard_deviation": 0.0007227795071339652, "minimum": 0.9929111154059053, "maximum": 0.9950477697996043, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5401078924135043, "standard_deviation": 0.034122088201350625, "minimum": 1.4881414544624902, "maximum": 1.5797168665640924, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4924208928559267, "standard_deviation": 0.19219510305348417, "minimum": 0.2529634048422036, "maximum": 0.8195840790528944, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08747434170243057, "standard_deviation": 0.006156196793646202, "minimum": 0.08163653264585266, "maximum": 0.09975615941122895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998159171965535, "standard_deviation": 0.00031600055367661403, "minimum": 0.9976060567942368, "maximum": 0.9986137798652318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8865008717660277, "standard_deviation": 0.033040745520826226, "minimum": 0.8366183218804836, "maximum": 0.9369105953284946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4881621215876676, "standard_deviation": 0.22828257978473676, "minimum": 0.14585957594837046, "maximum": 0.9122583218700032, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05119919525917583, "standard_deviation": 0.005419934053254261, "minimum": 0.04505907816590692, "maximum": 0.06161369942191467, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.233144278420116, "standard_deviation": 0.056335136577896405, "minimum": 0.148007712585857, "maximum": 0.28783616866845857, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.556724391897994, "standard_deviation": 0.008168436517509719, "minimum": 5.543681051886051, "maximum": 5.569970007042074, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5350924936880131, "standard_deviation": 0.04746917428526865, "minimum": 0.49226287272100155, "maximum": 0.6464568415122693, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08488094952832553, "standard_deviation": 0.03267614249035689, "minimum": -0.1308392736457479, "maximum": -0.020408645941487374, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9926242719091571, "standard_deviation": 0.0018406014922646124, "minimum": 0.9884944413336328, "maximum": 0.9943553055934367, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2391146653462075, "standard_deviation": 0.02996382523107638, "minimum": 1.192116138853017, "maximum": 1.2969714067522917, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6286610589715408, "standard_deviation": 0.3960517337024256, "minimum": 0.11905780179554883, "maximum": 1.3928372483609934, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08549389112590591, "standard_deviation": 0.009922268446895312, "minimum": 0.07515072304399778, "maximum": 0.10713673445083963, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975764997723834, "standard_deviation": 0.0013207119436828207, "minimum": 0.994372679767706, "maximum": 0.9983151993741132, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6683944598283008, "standard_deviation": 0.027805616778644616, "minimum": 0.6357636786901046, "maximum": 0.703789720827162, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6901119494901407, "standard_deviation": 0.41930021111615173, "minimum": 0.23500468485929318, "maximum": 1.577766073687152, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.047554196215654834, "standard_deviation": 0.010577203222191446, "minimum": 0.041121342448989275, "maximum": 0.07285058626583363, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0013", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.1025068759918213, "predicted_mos": 3.1025068759918213, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "dc2ac341a8cb7ed83cf2ddbd85d7facd0cef662a110117eedd297c21b6e499a5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.765127500053495, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2852213000675545, "全帯域_10ms_rms_envelope_nrmse": 0.09327104876977821, "全帯域_10ms_envelope_correlation": 0.9912517715582765, "全帯域_10ms_modulation_mae_db": 0.7241166672509856, "全帯域_40ms_log_rms_mae_db": 0.7126312157651943, "全帯域_40ms_rms_envelope_nrmse": 0.05539071864182751, "全帯域_40ms_envelope_correlation": 0.9967662741589521, "全帯域_40ms_modulation_mae_db": 0.8171046322718163, "300–1000 Hz_10ms_log_rms_mae_db": 1.7816469706864884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09414606133368954, "300–1000 Hz_10ms_envelope_correlation": 0.9934564250422636, "300–1000 Hz_10ms_modulation_mae_db": 0.7156545283020094, "300–1000 Hz_40ms_log_rms_mae_db": 1.185956727229955, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056747303972893036, "300–1000 Hz_40ms_envelope_correlation": 0.9977136980491162, "300–1000 Hz_40ms_modulation_mae_db": 0.8108826586902725, "1000–3000 Hz_10ms_log_rms_mae_db": 1.882763875433745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17360042267821824, "1000–3000 Hz_10ms_envelope_correlation": 0.979816270794173, "1000–3000 Hz_10ms_modulation_mae_db": 0.9316618160064198, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2424948573043724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11009011676804194, "1000–3000 Hz_40ms_envelope_correlation": 0.9940211922856679, "1000–3000 Hz_40ms_modulation_mae_db": 1.2111090386605208, "3000–6000 Hz_10ms_log_rms_mae_db": 1.897404680635696, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11781674323072323, "3000–6000 Hz_10ms_envelope_correlation": 0.9926093734203875, "3000–6000 Hz_10ms_modulation_mae_db": 0.24726250465726574, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3352080957124515, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06526596841991587, "3000–6000 Hz_40ms_envelope_correlation": 0.9983111808295714, "3000–6000 Hz_40ms_modulation_mae_db": 0.5583314615578471, "mr_stft_db": 5.738206486568075, "band_mae_db": 0.21412287639556782, "rms_gain_error_db": -0.06760085189731199, "output_peak": 0.4983293947893907}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1822302688143658, "全帯域_10ms_rms_envelope_nrmse": 0.07968187355870142, "全帯域_10ms_envelope_correlation": 0.9936374505468208, "全帯域_10ms_modulation_mae_db": 0.3021219529046952, "全帯域_40ms_log_rms_mae_db": 0.6449746066610441, "全帯域_40ms_rms_envelope_nrmse": 0.04410363727401473, "全帯域_40ms_envelope_correlation": 0.9979861732763456, "全帯域_40ms_modulation_mae_db": 0.3157528140211847, "300–1000 Hz_10ms_log_rms_mae_db": 1.754343232711383, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08245681643325359, "300–1000 Hz_10ms_envelope_correlation": 0.9952526522513861, "300–1000 Hz_10ms_modulation_mae_db": 0.5114313934509906, "300–1000 Hz_40ms_log_rms_mae_db": 1.0956024982095618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050191372223876325, "300–1000 Hz_40ms_envelope_correlation": 0.9985064921912523, "300–1000 Hz_40ms_modulation_mae_db": 0.4962473810209785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9179030610443695, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1980495651435182, "1000–3000 Hz_10ms_envelope_correlation": 0.9712219782882837, "1000–3000 Hz_10ms_modulation_mae_db": 0.5743343272010367, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3180429955605804, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13452450052129816, "1000–3000 Hz_40ms_envelope_correlation": 0.9871657976387072, "1000–3000 Hz_40ms_modulation_mae_db": 0.5215676303723799, "3000–6000 Hz_10ms_log_rms_mae_db": 1.96255544776404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15757507092014153, "3000–6000 Hz_10ms_envelope_correlation": 0.987500552717184, "3000–6000 Hz_10ms_modulation_mae_db": 0.26096088897037406, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3764660647290479, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08051027588931664, "3000–6000 Hz_40ms_envelope_correlation": 0.9974278623676627, "3000–6000 Hz_40ms_modulation_mae_db": 0.4380441734800789, "mr_stft_db": 5.757025198272566, "band_mae_db": 0.20538857405016875, "rms_gain_error_db": -0.07379594714174928, "output_peak": 0.5519299127115338}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1993806056657719, "全帯域_10ms_rms_envelope_nrmse": 0.09161510332097066, "全帯域_10ms_envelope_correlation": 0.9916767466297736, "全帯域_10ms_modulation_mae_db": 0.36378427922895556, "全帯域_40ms_log_rms_mae_db": 0.6446964463669915, "全帯域_40ms_rms_envelope_nrmse": 0.047906058760676315, "全帯域_40ms_envelope_correlation": 0.9977524963683134, "全帯域_40ms_modulation_mae_db": 0.3408034795195014, "300–1000 Hz_10ms_log_rms_mae_db": 1.7082781194385193, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10709639795642636, "300–1000 Hz_10ms_envelope_correlation": 0.9915158942950071, "300–1000 Hz_10ms_modulation_mae_db": 0.3387478953429202, "300–1000 Hz_40ms_log_rms_mae_db": 1.045517761538717, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05978183243163483, "300–1000 Hz_40ms_envelope_correlation": 0.9975357978885746, "300–1000 Hz_40ms_modulation_mae_db": 0.6721748428836453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8741296128085168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18673396600484268, "1000–3000 Hz_10ms_envelope_correlation": 0.9736128922866151, "1000–3000 Hz_10ms_modulation_mae_db": 0.6073578638344875, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2505539406659956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11970501918372958, "1000–3000 Hz_40ms_envelope_correlation": 0.9885602108822912, "1000–3000 Hz_40ms_modulation_mae_db": 0.51184539151348, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9272099500438213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12827130976599865, "3000–6000 Hz_10ms_envelope_correlation": 0.9908304324804543, "3000–6000 Hz_10ms_modulation_mae_db": 0.2462499527458819, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3618884574699954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06257471638699197, "3000–6000 Hz_40ms_envelope_correlation": 0.997824935582371, "3000–6000 Hz_40ms_modulation_mae_db": 0.38941198189478055, "mr_stft_db": 5.752031736745803, "band_mae_db": 0.15718556872852463, "rms_gain_error_db": -0.12920386663796127, "output_peak": 0.5863921529423993}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2235746060640178, "全帯域_10ms_rms_envelope_nrmse": 0.0834686850927127, "全帯域_10ms_envelope_correlation": 0.9931689059253579, "全帯域_10ms_modulation_mae_db": 0.4733130024801677, "全帯域_40ms_log_rms_mae_db": 0.6675336114002994, "全帯域_40ms_rms_envelope_nrmse": 0.05042328042326629, "全帯域_40ms_envelope_correlation": 0.9975414989971508, "全帯域_40ms_modulation_mae_db": 0.3143137008269491, "300–1000 Hz_10ms_log_rms_mae_db": 1.7735352372701603, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08652013854535436, "300–1000 Hz_10ms_envelope_correlation": 0.9945818112022722, "300–1000 Hz_10ms_modulation_mae_db": 0.6951868216004206, "300–1000 Hz_40ms_log_rms_mae_db": 1.1051571516829606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05250756529639126, "300–1000 Hz_40ms_envelope_correlation": 0.9981573074953893, "300–1000 Hz_40ms_modulation_mae_db": 0.771275407216841, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8697508629472215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18853483301546292, "1000–3000 Hz_10ms_envelope_correlation": 0.9746222335557116, "1000–3000 Hz_10ms_modulation_mae_db": 1.5387845842414138, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2398560801979308, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11595819807907878, "1000–3000 Hz_40ms_envelope_correlation": 0.9916806687016008, "1000–3000 Hz_40ms_modulation_mae_db": 1.9893672523065302, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9408127119988412, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13557502649983175, "3000–6000 Hz_10ms_envelope_correlation": 0.9901745649802762, "3000–6000 Hz_10ms_modulation_mae_db": 0.8517945970669455, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3312216281601872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07136745816663906, "3000–6000 Hz_40ms_envelope_correlation": 0.9974130835788017, "3000–6000 Hz_40ms_modulation_mae_db": 1.1386274707406032, "mr_stft_db": 5.7445505258116, "band_mae_db": 0.22562904112391546, "rms_gain_error_db": -0.14231924208860453, "output_peak": 0.5965918991530452}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1795802213202802, "全帯域_10ms_rms_envelope_nrmse": 0.09384811534738359, "全帯域_10ms_envelope_correlation": 0.9911968568817142, "全帯域_10ms_modulation_mae_db": 0.6320588608529417, "全帯域_40ms_log_rms_mae_db": 0.6763221539790533, "全帯域_40ms_rms_envelope_nrmse": 0.04955943990056354, "全帯域_40ms_envelope_correlation": 0.9975372173279368, "全帯域_40ms_modulation_mae_db": 0.7397821413288481, "300–1000 Hz_10ms_log_rms_mae_db": 1.8145246012908955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09561861748621346, "300–1000 Hz_10ms_envelope_correlation": 0.9933068759904664, "300–1000 Hz_10ms_modulation_mae_db": 1.0585050954573598, "300–1000 Hz_40ms_log_rms_mae_db": 1.1325641963781066, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05532970037662702, "300–1000 Hz_40ms_envelope_correlation": 0.9979274806911571, "300–1000 Hz_40ms_modulation_mae_db": 1.102001367206748, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8672447843373292, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17424211180630098, "1000–3000 Hz_10ms_envelope_correlation": 0.9776346946402773, "1000–3000 Hz_10ms_modulation_mae_db": 0.7976090888695802, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2108513489720056, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10536214174884763, "1000–3000 Hz_40ms_envelope_correlation": 0.9921611992938929, "1000–3000 Hz_40ms_modulation_mae_db": 0.8605329748757763, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8869068177550112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14025428715716184, "3000–6000 Hz_10ms_envelope_correlation": 0.9898658638119223, "3000–6000 Hz_10ms_modulation_mae_db": 0.4326151364694492, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3182344781938273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07004150520087478, "3000–6000 Hz_40ms_envelope_correlation": 0.9979765838651444, "3000–6000 Hz_40ms_modulation_mae_db": 0.4359513056485949, "mr_stft_db": 5.724920172964052, "band_mae_db": 0.15405229929900757, "rms_gain_error_db": -0.11311965809587665, "output_peak": 0.49913904078392035}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.212980459311976, "全帯域_10ms_rms_envelope_nrmse": 0.07946501030812618, "全帯域_10ms_envelope_correlation": 0.9936439225063404, "全帯域_10ms_modulation_mae_db": 0.12959527645352875, "全帯域_40ms_log_rms_mae_db": 0.6444461385817832, "全帯域_40ms_rms_envelope_nrmse": 0.0392831813482276, "全帯域_40ms_envelope_correlation": 0.9983755842884681, "全帯域_40ms_modulation_mae_db": 0.14257964004795096, "300–1000 Hz_10ms_log_rms_mae_db": 1.7831051324663412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0904553689668806, "300–1000 Hz_10ms_envelope_correlation": 0.9939592206073552, "300–1000 Hz_10ms_modulation_mae_db": 0.41327928418037807, "300–1000 Hz_40ms_log_rms_mae_db": 1.0805932779182235, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05313717505071951, "300–1000 Hz_40ms_envelope_correlation": 0.9980065209342805, "300–1000 Hz_40ms_modulation_mae_db": 0.32559287026452094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9146633414859981, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1873893828704558, "1000–3000 Hz_10ms_envelope_correlation": 0.9735812846188263, "1000–3000 Hz_10ms_modulation_mae_db": 0.5671874826895206, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2256577193836191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09818941884914448, "1000–3000 Hz_40ms_envelope_correlation": 0.9928885515200191, "1000–3000 Hz_40ms_modulation_mae_db": 1.0753461244241151, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8951075046376882, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14938628760742978, "3000–6000 Hz_10ms_envelope_correlation": 0.9900452462064596, "3000–6000 Hz_10ms_modulation_mae_db": 0.6470743426849851, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3502415495430535, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09044921334921441, "3000–6000 Hz_40ms_envelope_correlation": 0.9975289839664987, "3000–6000 Hz_40ms_modulation_mae_db": 0.9548630764659226, "mr_stft_db": 5.734015742238932, "band_mae_db": 0.21466258842512528, "rms_gain_error_db": -0.03092641922719641, "output_peak": 0.5976495207780461}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1915238328654814, "全帯域_10ms_rms_envelope_nrmse": 0.0853196511049971, "全帯域_10ms_envelope_correlation": 0.9927186364107939, "全帯域_10ms_modulation_mae_db": 0.267432726327642, "全帯域_40ms_log_rms_mae_db": 0.6143949492046429, "全帯域_40ms_rms_envelope_nrmse": 0.047777143795970776, "全帯域_40ms_envelope_correlation": 0.9976275009598013, "全帯域_40ms_modulation_mae_db": 0.7329495833051959, "300–1000 Hz_10ms_log_rms_mae_db": 1.6546915782190188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0858355704605019, "300–1000 Hz_10ms_envelope_correlation": 0.9946464276981122, "300–1000 Hz_10ms_modulation_mae_db": 0.5517216663501747, "300–1000 Hz_40ms_log_rms_mae_db": 1.047934309639118, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05203222898621487, "300–1000 Hz_40ms_envelope_correlation": 0.9981684302734305, "300–1000 Hz_40ms_modulation_mae_db": 0.5964969047114712, "1000–3000 Hz_10ms_log_rms_mae_db": 1.847093453339658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17529593499644092, "1000–3000 Hz_10ms_envelope_correlation": 0.9769885670335424, "1000–3000 Hz_10ms_modulation_mae_db": 1.5159361301616965, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1151617650284702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09833558591142139, "1000–3000 Hz_40ms_envelope_correlation": 0.9927866260878886, "1000–3000 Hz_40ms_modulation_mae_db": 1.9616184839268769, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9302646827026493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12363957893884539, "3000–6000 Hz_10ms_envelope_correlation": 0.9914933809430589, "3000–6000 Hz_10ms_modulation_mae_db": 0.41344698766253557, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2817691037896815, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06540112899245748, "3000–6000 Hz_40ms_envelope_correlation": 0.9979064099448347, "3000–6000 Hz_40ms_modulation_mae_db": 0.7170033444003311, "mr_stft_db": 5.752451813416684, "band_mae_db": 0.13655135233167331, "rms_gain_error_db": -0.07610279875104452, "output_peak": 0.5077628598603674}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2258468895414116, "全帯域_10ms_rms_envelope_nrmse": 0.08448914797981157, "全帯域_10ms_envelope_correlation": 0.9928703553276846, "全帯域_10ms_modulation_mae_db": 0.5560662519306623, "全帯域_40ms_log_rms_mae_db": 0.6374391095549139, "全帯域_40ms_rms_envelope_nrmse": 0.04072404428580632, "全帯域_40ms_envelope_correlation": 0.9983254486622407, "全帯域_40ms_modulation_mae_db": 0.7248617766987607, "300–1000 Hz_10ms_log_rms_mae_db": 1.7930063631234612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09504140506221921, "300–1000 Hz_10ms_envelope_correlation": 0.9934651755725963, "300–1000 Hz_10ms_modulation_mae_db": 0.9870147762873449, "300–1000 Hz_40ms_log_rms_mae_db": 1.1189871360989925, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05583734011392692, "300–1000 Hz_40ms_envelope_correlation": 0.9979704374377527, "300–1000 Hz_40ms_modulation_mae_db": 0.9281158364602136, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8812743948364963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18578811084918534, "1000–3000 Hz_10ms_envelope_correlation": 0.9741161640070138, "1000–3000 Hz_10ms_modulation_mae_db": 1.3890513564689464, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2553749265834242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1029802239456399, "1000–3000 Hz_40ms_envelope_correlation": 0.9921753544112277, "1000–3000 Hz_40ms_modulation_mae_db": 2.036763393315403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9800694448218323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14409029481836652, "3000–6000 Hz_10ms_envelope_correlation": 0.9881992189069873, "3000–6000 Hz_10ms_modulation_mae_db": 0.7442618867604024, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4303684339349525, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07199298341962206, "3000–6000 Hz_40ms_envelope_correlation": 0.9970114247676463, "3000–6000 Hz_40ms_modulation_mae_db": 0.5017513383892577, "mr_stft_db": 5.743829326559535, "band_mae_db": 0.13083165537665395, "rms_gain_error_db": -0.0879022645455147, "output_peak": 0.5255850222380754}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.897452634623242, "spectral_convergence": 0.2560876462750986, "spectral_convergence_status": "ok", "amplitude_floor": 2.0096490980240517e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.622938701905733, "spectral_convergence": 0.23874729659394286, "spectral_convergence_status": "ok", "amplitude_floor": 1.509105853911261e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.69422812317525, "spectral_convergence": 0.2474732429886755, "spectral_convergence_status": "ok", "amplitude_floor": 1.2442618586634314e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017656001839806658, "output_power": 0.0018601819240623542, "delta_db": 0.22663054483115172, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005544968065872631, "output_power": 0.005338634234627716, "delta_db": -0.1646888224109806, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005978914392629673, "output_power": 0.0006007387946894678, "delta_db": 0.020633437081245917, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.426968976514186e-05, "output_power": 7.297671505345873e-05, "delta_db": -0.6248708479036348, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.40147911812626e-05, "output_power": 6.199012090608164e-05, "delta_db": -0.13957849739310968, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00024101533193812626, "output_power": 0.00023507755214277528, "delta_db": -0.10833510875328399, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21412287639556782, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06760085189731199, "pitch": null, "output_rms_dbfs": -20.877992418825528, "output_peak": 0.4983293947893907, "reference_rms_dbfs": -20.810391566928214, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.738206486568075}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026421160622171636, "log_rms_mae_db": 1.2852213000675545, "rms_envelope_nrmse": 0.09327104876977821, "envelope_correlation": 0.9912517715582765, "rms_gain_error_db": -0.06759155802535627, "reference_mean_power": 0.008303652784602253, "output_mean_power": 0.00817541909675684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.437173268209281e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017543654513432086, "output_power": 0.0017769634666372838, "reference_active": true, "delta_db": 0.05558432840641726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013368683738807852, "output_power": 0.00013516291967261438, "reference_active": true, "delta_db": 0.04768915095650163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.235741942938941e-05, "output_power": 4.493403481398626e-05, "reference_active": true, "delta_db": -2.069076522390038}], "active_band_count": 3, "active_band_mae_db": 0.7241166672509856}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022872947944797367, "log_rms_mae_db": 0.7126312157651943, "rms_envelope_nrmse": 0.05539071864182751, "envelope_correlation": 0.9967662741589521, "rms_gain_error_db": -0.06783077581850842, "reference_mean_power": 0.008338532782066458, "output_mean_power": 0.008209308243816875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.164744003117204e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015799393891106446, "output_power": 0.0016020086468033828, "reference_active": true, "delta_db": 0.060244293243955555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.374442374979845e-05, "output_power": 6.06703817543707e-05, "reference_active": true, "delta_db": -0.21465471798896216}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.293312028286094e-06, "output_power": 1.9952227669818782e-06, "reference_active": true, "delta_db": -2.176414885582531}], "active_band_count": 3, "active_band_mae_db": 0.8171046322718163}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024243770094671305, "log_rms_mae_db": 1.7816469706864884, "rms_envelope_nrmse": 0.09414606133368954, "envelope_correlation": 0.9934564250422636, "rms_gain_error_db": -0.18223102763992224, "reference_mean_power": 0.003959461759548422, "output_mean_power": 0.0037967591665875864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.015177817247118e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001136879031892713, "output_power": 0.0010725631943484676, "reference_active": true, "delta_db": -0.2529136681081585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001232393146611943, "output_power": 0.00010719707315037982, "reference_active": true, "delta_db": -0.6056634674026841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0364107060899577e-05, "output_power": 2.2569496423667717e-05, "reference_active": true, "delta_db": -1.2883864493951855}], "active_band_count": 3, "active_band_mae_db": 0.7156545283020094}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020477863430336455, "log_rms_mae_db": 1.185956727229955, "rms_envelope_nrmse": 0.056747303972893036, "envelope_correlation": 0.9977136980491162, "rms_gain_error_db": -0.18223181379702896, "reference_mean_power": 0.003979363820398493, "output_mean_power": 0.0038158427192497095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8942412803353574e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010558685473009428, "output_power": 0.000992170862706236, "reference_active": true, "delta_db": -0.2702338423887226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.091449836890923e-05, "output_power": 4.566285678809907e-05, "reference_active": true, "delta_db": -0.4727839066628541}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8755256559270567e-06, "output_power": 1.2710422492651861e-06, "reference_active": true, "delta_db": -1.6896302270192411}], "active_band_count": 3, "active_band_mae_db": 0.8108826586902725}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5081112731014556e-05, "log_rms_mae_db": 1.882763875433745, "rms_envelope_nrmse": 0.17360042267821824, "envelope_correlation": 0.979816270794173, "rms_gain_error_db": -0.604169755367653, "reference_mean_power": 0.00010911551831198975, "output_mean_power": 9.494444155906028e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.836309277153812e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5256828161206645e-05, "output_power": 2.0598622526562924e-05, "reference_active": true, "delta_db": -0.8854063030850332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.88342602252987e-06, "output_power": 8.58578651123219e-06, "reference_active": true, "delta_db": -0.6112743053581173}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.436572483663522e-06, "output_power": 4.031760249899729e-06, "reference_active": true, "delta_db": -1.2983048395761088}], "active_band_count": 3, "active_band_mae_db": 0.9316618160064198}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.4434083199144116e-05, "log_rms_mae_db": 1.2424948573043724, "rms_envelope_nrmse": 0.11009011676804194, "envelope_correlation": 0.9940211922856679, "rms_gain_error_db": -0.6050839522036782, "reference_mean_power": 0.0001096329653054558, "output_mean_power": 9.537460793169159e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.313909407445067e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3642314296732694e-05, "output_power": 1.944115908538765e-05, "reference_active": true, "delta_db": -0.8496783239223284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.77067085832153e-06, "output_power": 5.529730018705948e-06, "reference_active": true, "delta_db": -0.8792777400868322}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.651384635522461e-07, "output_power": 3.0001650588589946e-07, "reference_active": true, "delta_db": -1.9043710519724018}], "active_band_count": 3, "active_band_mae_db": 1.2111090386605208}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010577651466639849, "log_rms_mae_db": 1.897404680635696, "rms_envelope_nrmse": 0.11781674323072323, "envelope_correlation": 0.9926093734203875, "rms_gain_error_db": -0.29361162097616184, "reference_mean_power": 0.00018020158938811219, "output_mean_power": 0.00016842147056705633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.705326923064497e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.613613486254781e-05, "output_power": 1.6361393950109826e-05, "reference_active": true, "delta_db": 0.06020786757089541}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.942687703713047e-06, "output_power": 9.0983433114484e-06, "reference_active": true, "delta_db": 0.07494255614021987}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.043858273628482e-06, "output_power": 9.077741860630441e-07, "reference_active": true, "delta_db": -0.6066370902606819}], "active_band_count": 3, "active_band_mae_db": 0.24726250465726574}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.602886913996206e-05, "log_rms_mae_db": 1.3352080957124515, "rms_envelope_nrmse": 0.06526596841991587, "envelope_correlation": 0.9983111808295714, "rms_gain_error_db": -0.31057773012366396, "reference_mean_power": 0.0001775816262598491, "output_mean_power": 0.0001653256564641366}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.444034789789241e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7728340440048457e-05, "output_power": 1.7761529924255483e-05, "reference_active": true, "delta_db": 0.008122889096165107}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.955564320473026e-06, "output_power": 5.097983282080196e-06, "reference_active": true, "delta_db": 0.12305289533797621}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.260796671427193e-08, "output_power": 6.490325665681829e-08, "reference_active": true, "delta_db": -1.5438186002393999}], "active_band_count": 3, "active_band_mae_db": 0.5583314615578471}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8501369785973285, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c2a588b965c5381be827082ba50234b9ac4472f139f0584fcb95586f45ead082", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9751992606530554, "standard_deviation": 0.0027492592077241296, "minimum": 0.9712219782882837, "maximum": 0.979816270794173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8818529232791668, "standard_deviation": 0.02391959258942319, "minimum": 1.847093453339658, "maximum": 1.9179030610443695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9902403311841377, "standard_deviation": 0.42684822378551973, "minimum": 0.5671874826895206, "maximum": 1.5387845842414138, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18370429092055313, "standard_deviation": 0.008606837855935372, "minimum": 0.17360042267821824, "maximum": 0.1980495651435182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9914299501026619, "standard_deviation": 0.0023377106041760614, "minimum": 0.9871657976387072, "maximum": 0.9940211922856679, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2322492042120499, "standard_deviation": 0.056825622127246254, "minimum": 1.1151617650284702, "maximum": 1.3180429955605804, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2710187861743854, "standard_deviation": 0.6467185424975185, "minimum": 0.51184539151348, "maximum": 2.036763393315403, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11064315062590023, "standard_deviation": 0.01238762628534903, "minimum": 0.09818941884914448, "maximum": 0.13452450052129816, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9900898291833413, "standard_deviation": 0.0016563575385375063, "minimum": 0.987500552717184, "maximum": 0.9926093734203875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9275414050449475, "standard_deviation": 0.03333848394057159, "minimum": 1.8869068177550112, "maximum": 1.9800694448218323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.48045828712723, "standard_deviation": 0.23887722794134592, "minimum": 0.2462499527458819, "maximum": 0.8517945970669455, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13707607486731233, "standard_deviation": 0.013428587865215186, "minimum": 0.11781674323072323, "maximum": 0.15757507092014153, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9976750581128164, "standard_deviation": 0.0004076132764705719, "minimum": 0.9970114247676463, "maximum": 0.9983111808295714, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3481747264416497, "standard_deviation": 0.04393210636682483, "minimum": 1.2817691037896815, "maximum": 1.4303684339349525, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.641748019072177, "standard_deviation": 0.27377571762833725, "minimum": 0.38941198189478055, "maximum": 1.1386274707406032, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07220040622812904, "standard_deviation": 0.009215213190541246, "minimum": 0.06257471638699197, "maximum": 0.09044921334921441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937730603324324, "standard_deviation": 0.0011431706354954687, "minimum": 0.9915158942950071, "maximum": 0.9952526522513861, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7578914044007834, "standard_deviation": 0.05211174740947668, "minimum": 1.6546915782190188, "maximum": 1.8145246012908955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6589426826214497, "standard_deviation": 0.25855157687762254, "minimum": 0.3387478953429202, "maximum": 1.0585050954573598, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09214629703056737, "standard_deviation": 0.007710544134293297, "minimum": 0.08245681643325359, "maximum": 0.10709639795642636, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979982706201191, "standard_deviation": 0.00029598570043455396, "minimum": 0.9975357978885746, "maximum": 0.9985064921912523, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1015391323369543, "standard_deviation": 0.04612444544028653, "minimum": 1.045517761538717, "maximum": 1.185956727229955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7128484085568364, "standard_deviation": 0.24578718656973692, "minimum": 0.32559287026452094, "maximum": 1.102001367206748, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05444556480653547, "standard_deviation": 0.0030678647701532368, "minimum": 0.050191372223876325, "maximum": 0.05978183243163483, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17980299446632958, "standard_deviation": 0.03889773564782952, "minimum": 0.13083165537665395, "maximum": 0.22562904112391546, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.743378875322156, "standard_deviation": 0.010691342138385443, "minimum": 5.724920172964052, "maximum": 5.757025198272566, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5454224754070973, "standard_deviation": 0.0434881283738672, "minimum": 0.4983293947893907, "maximum": 0.5976495207780461, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09012138104815742, "standard_deviation": 0.0363714188857985, "minimum": -0.14231924208860453, "maximum": -0.03092641922719641, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9925205807233453, "standard_deviation": 0.001011742933505501, "minimum": 0.9911968568817142, "maximum": 0.9936439225063404, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2125422729563575, "standard_deviation": 0.034229220888156386, "minimum": 1.1795802213202802, "maximum": 1.2852213000675545, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4310611271786974, "standard_deviation": 0.20089106441801657, "minimum": 0.12959527645352875, "maximum": 0.7241166672509856, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08639482943531018, "standard_deviation": 0.005812809050050406, "minimum": 0.07946501030812618, "maximum": 0.09384811534738359, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977390242549011, "standard_deviation": 0.0005139189564009793, "minimum": 0.9967662741589521, "maximum": 0.9983755842884681, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6553047789392403, "standard_deviation": 0.029779068193284335, "minimum": 0.6143949492046429, "maximum": 0.7126312157651943, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.516018471002526, "standard_deviation": 0.26250052839883015, "minimum": 0.14257964004795096, "maximum": 0.8171046322718163, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.046895938053794134, "standard_deviation": 0.00530953094386198, "minimum": 0.0392831813482276, "maximum": 0.05539071864182751, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274381-0013", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.2909626960754395, "predicted_mos": 3.2909626960754395, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5df39ce0e7877fc17f5acb051fff735e785a3138d44e0ca810dc39350dcdde12", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6776442998088896, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2391840054346233, "全帯域_10ms_rms_envelope_nrmse": 0.08014184518133768, "全帯域_10ms_envelope_correlation": 0.9935630042152978, "全帯域_10ms_modulation_mae_db": 0.3187948517343314, "全帯域_40ms_log_rms_mae_db": 0.6690880490015112, "全帯域_40ms_rms_envelope_nrmse": 0.0435349774172753, "全帯域_40ms_envelope_correlation": 0.9980355815751857, "全帯域_40ms_modulation_mae_db": 0.2423991327329349, "300–1000 Hz_10ms_log_rms_mae_db": 1.6938275795116917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09065109463760146, "300–1000 Hz_10ms_envelope_correlation": 0.9940317047814774, "300–1000 Hz_10ms_modulation_mae_db": 0.5036597077169304, "300–1000 Hz_40ms_log_rms_mae_db": 1.0794462828884872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05580780600179621, "300–1000 Hz_40ms_envelope_correlation": 0.9978949383075527, "300–1000 Hz_40ms_modulation_mae_db": 0.47601019132780825, "1000–3000 Hz_10ms_log_rms_mae_db": 1.821543740834228, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19249888125314338, "1000–3000 Hz_10ms_envelope_correlation": 0.9727778712483113, "1000–3000 Hz_10ms_modulation_mae_db": 0.9356311737709623, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1700407003066902, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11466628580496724, "1000–3000 Hz_40ms_envelope_correlation": 0.9911350014798851, "1000–3000 Hz_40ms_modulation_mae_db": 1.6256209392652516, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9076376787429887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1268790758870263, "3000–6000 Hz_10ms_envelope_correlation": 0.9916153971700402, "3000–6000 Hz_10ms_modulation_mae_db": 0.6657328762486117, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2433910967009112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06365810698607166, "3000–6000 Hz_40ms_envelope_correlation": 0.9982771078181684, "3000–6000 Hz_40ms_modulation_mae_db": 0.8822711522825237, "mr_stft_db": 5.686422899161415, "band_mae_db": 0.1772659928438485, "rms_gain_error_db": -0.07201553403370993, "output_peak": 0.5094233708559266}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2246035542963474, "全帯域_10ms_rms_envelope_nrmse": 0.0879054144864402, "全帯域_10ms_envelope_correlation": 0.9922651147456303, "全帯域_10ms_modulation_mae_db": 0.5989772850143228, "全帯域_40ms_log_rms_mae_db": 0.6912585352226728, "全帯域_40ms_rms_envelope_nrmse": 0.05189845077136591, "全帯域_40ms_envelope_correlation": 0.9972117878018952, "全帯域_40ms_modulation_mae_db": 0.9095675699998833, "300–1000 Hz_10ms_log_rms_mae_db": 1.758837339712508, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09612422474468928, "300–1000 Hz_10ms_envelope_correlation": 0.993095322358287, "300–1000 Hz_10ms_modulation_mae_db": 0.3569303165124278, "300–1000 Hz_40ms_log_rms_mae_db": 1.035880532917179, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053352871041951154, "300–1000 Hz_40ms_envelope_correlation": 0.9979251164164102, "300–1000 Hz_40ms_modulation_mae_db": 0.4069828505730375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.78892105495922, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.190150120819698, "1000–3000 Hz_10ms_envelope_correlation": 0.972836868498691, "1000–3000 Hz_10ms_modulation_mae_db": 1.6701034597440156, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1385684107978598, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1208124409337409, "1000–3000 Hz_40ms_envelope_correlation": 0.9888560875280821, "1000–3000 Hz_40ms_modulation_mae_db": 2.459849511581062, "3000–6000 Hz_10ms_log_rms_mae_db": 1.892791937486024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.11578453690166814, "3000–6000 Hz_10ms_envelope_correlation": 0.9923697605080809, "3000–6000 Hz_10ms_modulation_mae_db": 1.3087113157845032, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2941443025217183, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05541720997537701, "3000–6000 Hz_40ms_envelope_correlation": 0.9982560071577553, "3000–6000 Hz_40ms_modulation_mae_db": 1.6600854021896343, "mr_stft_db": 5.6960912527060055, "band_mae_db": 0.17740032966796493, "rms_gain_error_db": -0.09043819581720627, "output_peak": 0.6202480118659591}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2127151402791028, "全帯域_10ms_rms_envelope_nrmse": 0.0942770918362871, "全帯域_10ms_envelope_correlation": 0.9910649010515802, "全帯域_10ms_modulation_mae_db": 0.7417277183980192, "全帯域_40ms_log_rms_mae_db": 0.6074649479536962, "全帯域_40ms_rms_envelope_nrmse": 0.06335518173258452, "全帯域_40ms_envelope_correlation": 0.9957539840296182, "全帯域_40ms_modulation_mae_db": 1.099333861663366, "300–1000 Hz_10ms_log_rms_mae_db": 1.688894314533981, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08605815013638739, "300–1000 Hz_10ms_envelope_correlation": 0.9947024678606429, "300–1000 Hz_10ms_modulation_mae_db": 0.4698164641529714, "300–1000 Hz_40ms_log_rms_mae_db": 0.9988803234879201, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05222976130868458, "300–1000 Hz_40ms_envelope_correlation": 0.998252199438052, "300–1000 Hz_40ms_modulation_mae_db": 0.7301034558188194, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7824401523123146, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18437052249508298, "1000–3000 Hz_10ms_envelope_correlation": 0.9754459350332958, "1000–3000 Hz_10ms_modulation_mae_db": 0.8200102497572157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1098975811205798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10662896791223575, "1000–3000 Hz_40ms_envelope_correlation": 0.9929970782470412, "1000–3000 Hz_40ms_modulation_mae_db": 1.415754327741884, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8270942772078134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14212086586559605, "3000–6000 Hz_10ms_envelope_correlation": 0.9884986043191751, "3000–6000 Hz_10ms_modulation_mae_db": 0.2522044927705272, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2628822823297088, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0768483599074583, "3000–6000 Hz_40ms_envelope_correlation": 0.9966640940914984, "3000–6000 Hz_40ms_modulation_mae_db": 0.4971383382351469, "mr_stft_db": 5.6827846085455995, "band_mae_db": 0.26248889448355706, "rms_gain_error_db": -0.07250103686831079, "output_peak": 0.5312526158967193}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2139146472978186, "全帯域_10ms_rms_envelope_nrmse": 0.07722586347708285, "全帯域_10ms_envelope_correlation": 0.9940898363698022, "全帯域_10ms_modulation_mae_db": 0.3382842997565188, "全帯域_40ms_log_rms_mae_db": 0.6001582515560017, "全帯域_40ms_rms_envelope_nrmse": 0.04419364916787153, "全帯域_40ms_envelope_correlation": 0.9980090755151542, "全帯域_40ms_modulation_mae_db": 0.392571239928922, "300–1000 Hz_10ms_log_rms_mae_db": 1.6944122632944563, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08872640716524696, "300–1000 Hz_10ms_envelope_correlation": 0.99404509838317, "300–1000 Hz_10ms_modulation_mae_db": 0.5231479596657749, "300–1000 Hz_40ms_log_rms_mae_db": 0.9593696382947741, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055271002357853434, "300–1000 Hz_40ms_envelope_correlation": 0.99762079402986, "300–1000 Hz_40ms_modulation_mae_db": 0.6285840881224257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7800868389934448, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19707230886163415, "1000–3000 Hz_10ms_envelope_correlation": 0.9729669050347366, "1000–3000 Hz_10ms_modulation_mae_db": 1.343830093755009, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1799566726029498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11857991896351847, "1000–3000 Hz_40ms_envelope_correlation": 0.9925499338154934, "1000–3000 Hz_40ms_modulation_mae_db": 1.5118930031778357, "3000–6000 Hz_10ms_log_rms_mae_db": 1.924623552724017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1580699973668531, "3000–6000 Hz_10ms_envelope_correlation": 0.9862047514824792, "3000–6000 Hz_10ms_modulation_mae_db": 0.349655209059847, "3000–6000 Hz_40ms_log_rms_mae_db": 1.301346153721033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07837385631718337, "3000–6000 Hz_40ms_envelope_correlation": 0.9966666584126915, "3000–6000 Hz_40ms_modulation_mae_db": 0.14960278626550025, "mr_stft_db": 5.705283549146748, "band_mae_db": 0.20532524478765843, "rms_gain_error_db": -0.07601405139603544, "output_peak": 0.5422616253653542}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2889048502748255, "全帯域_10ms_rms_envelope_nrmse": 0.11235435149976052, "全帯域_10ms_envelope_correlation": 0.9874032543371833, "全帯域_10ms_modulation_mae_db": 1.23175849110168, "全帯域_40ms_log_rms_mae_db": 0.7004835024329894, "全帯域_40ms_rms_envelope_nrmse": 0.07070336802787787, "全帯域_40ms_envelope_correlation": 0.9946986464191249, "全帯域_40ms_modulation_mae_db": 1.8264735753326464, "300–1000 Hz_10ms_log_rms_mae_db": 1.6667899476594061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09443357269071814, "300–1000 Hz_10ms_envelope_correlation": 0.993338949686873, "300–1000 Hz_10ms_modulation_mae_db": 0.5089323973825323, "300–1000 Hz_40ms_log_rms_mae_db": 1.0665318284476721, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055694647266457356, "300–1000 Hz_40ms_envelope_correlation": 0.9977114627302789, "300–1000 Hz_40ms_modulation_mae_db": 0.4830187221230687, "1000–3000 Hz_10ms_log_rms_mae_db": 1.775339597815423, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17966525634784855, "1000–3000 Hz_10ms_envelope_correlation": 0.9767626103550796, "1000–3000 Hz_10ms_modulation_mae_db": 1.2059973812212572, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0886162195242801, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10502573232526204, "1000–3000 Hz_40ms_envelope_correlation": 0.9932408285682398, "1000–3000 Hz_40ms_modulation_mae_db": 1.728212816491866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9033876330834796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15080577822688468, "3000–6000 Hz_10ms_envelope_correlation": 0.987047002686784, "3000–6000 Hz_10ms_modulation_mae_db": 0.7434849313811195, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2927384065023306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0856135910106047, "3000–6000 Hz_40ms_envelope_correlation": 0.995762552836255, "3000–6000 Hz_40ms_modulation_mae_db": 0.9083623221795184, "mr_stft_db": 5.7016684488609615, "band_mae_db": 0.24940416406351798, "rms_gain_error_db": -0.03785680845911221, "output_peak": 0.5316174618088627}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.177708923817533, "全帯域_10ms_rms_envelope_nrmse": 0.08615749893211488, "全帯域_10ms_envelope_correlation": 0.9925661504536569, "全帯域_10ms_modulation_mae_db": 0.7877832572054166, "全帯域_40ms_log_rms_mae_db": 0.6649885866560958, "全帯域_40ms_rms_envelope_nrmse": 0.054200701573189056, "全帯域_40ms_envelope_correlation": 0.9968742739632516, "全帯域_40ms_modulation_mae_db": 0.5847553469810539, "300–1000 Hz_10ms_log_rms_mae_db": 1.6814122253391892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09035946009407886, "300–1000 Hz_10ms_envelope_correlation": 0.9938528239166591, "300–1000 Hz_10ms_modulation_mae_db": 0.3207727551344754, "300–1000 Hz_40ms_log_rms_mae_db": 1.039365757429795, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053768328767619514, "300–1000 Hz_40ms_envelope_correlation": 0.9977961163890696, "300–1000 Hz_40ms_modulation_mae_db": 0.45168959751118454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8111788618866504, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1902539768507835, "1000–3000 Hz_10ms_envelope_correlation": 0.9737443617998907, "1000–3000 Hz_10ms_modulation_mae_db": 1.020865255779016, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1522176241102904, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10597480736151609, "1000–3000 Hz_40ms_envelope_correlation": 0.9932187600908635, "1000–3000 Hz_40ms_modulation_mae_db": 1.3055303405302126, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8180442828386583, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16684290922362113, "3000–6000 Hz_10ms_envelope_correlation": 0.9843541053032134, "3000–6000 Hz_10ms_modulation_mae_db": 1.2541711021229907, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2294195666093115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09190684777083696, "3000–6000 Hz_40ms_envelope_correlation": 0.995176637825826, "3000–6000 Hz_40ms_modulation_mae_db": 1.5589767674966157, "mr_stft_db": 5.693474751306543, "band_mae_db": 0.253082383768402, "rms_gain_error_db": -0.026911285952148582, "output_peak": 0.5331747726074887}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1943899318274103, "全帯域_10ms_rms_envelope_nrmse": 0.0847640445926897, "全帯域_10ms_envelope_correlation": 0.9928213385547531, "全帯域_10ms_modulation_mae_db": 0.8506970926118704, "全帯域_40ms_log_rms_mae_db": 0.6481817951937983, "全帯域_40ms_rms_envelope_nrmse": 0.04388776852537963, "全帯域_40ms_envelope_correlation": 0.998075786448781, "全帯域_40ms_modulation_mae_db": 0.889766440184801, "300–1000 Hz_10ms_log_rms_mae_db": 1.6129351191124879, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09514251399237197, "300–1000 Hz_10ms_envelope_correlation": 0.9933868126173108, "300–1000 Hz_10ms_modulation_mae_db": 0.5345247055720687, "300–1000 Hz_40ms_log_rms_mae_db": 0.9533472739384445, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0541966027485638, "300–1000 Hz_40ms_envelope_correlation": 0.998039262120449, "300–1000 Hz_40ms_modulation_mae_db": 0.4013068308313681, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8107424825302159, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18271192376111622, "1000–3000 Hz_10ms_envelope_correlation": 0.9763616493887135, "1000–3000 Hz_10ms_modulation_mae_db": 1.7580214872748805, "1000–3000 Hz_40ms_log_rms_mae_db": 1.142847967178735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11240882419939013, "1000–3000 Hz_40ms_envelope_correlation": 0.9924651180367874, "1000–3000 Hz_40ms_modulation_mae_db": 2.1891368336178654, "3000–6000 Hz_10ms_log_rms_mae_db": 1.879479173242816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12082684156012372, "3000–6000 Hz_10ms_envelope_correlation": 0.9919361813326021, "3000–6000 Hz_10ms_modulation_mae_db": 0.19918970439417868, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2970546118159094, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07104697522833772, "3000–6000 Hz_40ms_envelope_correlation": 0.9974763167631189, "3000–6000 Hz_40ms_modulation_mae_db": 0.4139791795359393, "mr_stft_db": 5.682523483564882, "band_mae_db": 0.2493699361843549, "rms_gain_error_db": -0.09058316893742382, "output_peak": 0.504243674577223}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1876509292444477, "全帯域_10ms_rms_envelope_nrmse": 0.08879413461964604, "全帯域_10ms_envelope_correlation": 0.9922212250742426, "全帯域_10ms_modulation_mae_db": 0.4051984938526805, "全帯域_40ms_log_rms_mae_db": 0.6992827994158198, "全帯域_40ms_rms_envelope_nrmse": 0.05420236071667562, "全帯域_40ms_envelope_correlation": 0.9971076390516163, "全帯域_40ms_modulation_mae_db": 0.3548463896620224, "300–1000 Hz_10ms_log_rms_mae_db": 1.6789882030225953, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09370349697119108, "300–1000 Hz_10ms_envelope_correlation": 0.9936752858561796, "300–1000 Hz_10ms_modulation_mae_db": 0.8441209836352993, "300–1000 Hz_40ms_log_rms_mae_db": 1.0037678369084917, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055907113834809605, "300–1000 Hz_40ms_envelope_correlation": 0.9979766681810939, "300–1000 Hz_40ms_modulation_mae_db": 0.9317053907115751, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8693250936195205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19447120299684087, "1000–3000 Hz_10ms_envelope_correlation": 0.9724966501709388, "1000–3000 Hz_10ms_modulation_mae_db": 1.7289667923673868, "1000–3000 Hz_40ms_log_rms_mae_db": 1.176099164435648, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10817168296284799, "1000–3000 Hz_40ms_envelope_correlation": 0.9928259514273389, "1000–3000 Hz_40ms_modulation_mae_db": 2.529743727668184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.877729808495343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1431853891707861, "3000–6000 Hz_10ms_envelope_correlation": 0.988384354231214, "3000–6000 Hz_10ms_modulation_mae_db": 0.4630988273064456, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2817414269312868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06049396427276886, "3000–6000 Hz_40ms_envelope_correlation": 0.9978921376946982, "3000–6000 Hz_40ms_modulation_mae_db": 0.7712539211856431, "mr_stft_db": 5.678263686415815, "band_mae_db": 0.17742968129949765, "rms_gain_error_db": -0.14416688866371935, "output_peak": 0.5632189781580959}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.882157378906604, "spectral_convergence": 0.2523553915260505, "spectral_convergence_status": "ok", "amplitude_floor": 2.0096490980240517e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.560165379344506, "spectral_convergence": 0.2446010072405294, "spectral_convergence_status": "ok", "amplitude_floor": 1.509105853911261e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.616945939233138, "spectral_convergence": 0.24670100294332709, "spectral_convergence_status": "ok", "amplitude_floor": 1.2442618586634314e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017656001839806658, "output_power": 0.0018193866810606231, "delta_db": 0.13032645696839895, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005544968065872631, "output_power": 0.005366532833086956, "delta_db": -0.14205258569090637, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005978914392629673, "output_power": 0.0005835228042910801, "delta_db": -0.10564502038682103, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.426968976514186e-05, "output_power": 7.576241079999924e-05, "delta_db": -0.4621760939438497, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.40147911812626e-05, "output_power": 6.368081479333922e-05, "delta_db": -0.022717215987803036, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00024101533193812626, "output_power": 0.0002524134710535788, "delta_db": 0.20067858408531197, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1772659928438485, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07201553403370993, "pitch": null, "output_rms_dbfs": -20.882407100961924, "output_peak": 0.5094233708559266, "reference_rms_dbfs": -20.810391566928214, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.686422899161415}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026421160622171636, "log_rms_mae_db": 1.2391840054346233, "rms_envelope_nrmse": 0.08014184518133768, "envelope_correlation": 0.9935630042152978, "rms_gain_error_db": -0.0719282288508015, "reference_mean_power": 0.008303652784602253, "output_mean_power": 0.00816725956275209}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.437173268209281e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017543654513432086, "output_power": 0.0017124212867515014, "reference_active": true, "delta_db": -0.10509448387943786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013368683738807852, "output_power": 0.00012144258484088136, "reference_active": true, "delta_db": -0.4171764698567948}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.235741942938941e-05, "output_power": 6.547442747687809e-05, "reference_active": true, "delta_db": -0.4341136014667616}], "active_band_count": 3, "active_band_mae_db": 0.3187948517343314}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022872947944797367, "log_rms_mae_db": 0.6690880490015112, "rms_envelope_nrmse": 0.0435349774172753, "envelope_correlation": 0.9980355815751857, "rms_gain_error_db": -0.07165630476709532, "reference_mean_power": 0.008338532782066458, "output_mean_power": 0.00820208017161859}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.164744003117204e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015799393891106446, "output_power": 0.0015368691329187736, "reference_active": true, "delta_db": -0.1200353838100586}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.374442374979845e-05, "output_power": 5.81598303602393e-05, "reference_active": true, "delta_db": -0.398190683404068}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.293312028286094e-06, "output_power": 3.138598387216729e-06, "reference_active": true, "delta_db": -0.20897133098467816}], "active_band_count": 3, "active_band_mae_db": 0.2423991327329349}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024243770094671305, "log_rms_mae_db": 1.6938275795116917, "rms_envelope_nrmse": 0.09065109463760146, "envelope_correlation": 0.9940317047814774, "rms_gain_error_db": -0.2057675161410087, "reference_mean_power": 0.003959461759548422, "output_mean_power": 0.0037762383747376917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.015177817247118e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001136879031892713, "output_power": 0.0010795173958811789, "reference_active": true, "delta_db": -0.22484611452734393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001232393146611943, "output_power": 0.00011042744099377194, "reference_active": true, "delta_db": -0.47672266425330573}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0364107060899577e-05, "output_power": 2.5201098951616698e-05, "reference_active": true, "delta_db": -0.8094103443701413}], "active_band_count": 3, "active_band_mae_db": 0.5036597077169304}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020477863430336455, "log_rms_mae_db": 1.0794462828884872, "rms_envelope_nrmse": 0.05580780600179621, "envelope_correlation": 0.9978949383075527, "rms_gain_error_db": -0.205769050771146, "reference_mean_power": 0.003979363820398493, "output_mean_power": 0.0037952181301964525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8942412803353574e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010558685473009428, "output_power": 0.0009974421346695174, "reference_active": true, "delta_db": -0.24722143057099227}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.091449836890923e-05, "output_power": 4.856142540464545e-05, "reference_active": true, "delta_db": -0.20550043341033747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8755256559270567e-06, "output_power": 1.4982770885022524e-06, "reference_active": true, "delta_db": -0.975308710002095}], "active_band_count": 3, "active_band_mae_db": 0.47601019132780825}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5081112731014556e-05, "log_rms_mae_db": 1.821543740834228, "rms_envelope_nrmse": 0.19249888125314338, "envelope_correlation": 0.9727778712483113, "rms_gain_error_db": -0.4424685440935303, "reference_mean_power": 0.00010911551831198975, "output_mean_power": 9.854615081962659e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.836309277153812e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5256828161206645e-05, "output_power": 2.3391843818509295e-05, "reference_active": true, "delta_db": -0.33314353846686934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.88342602252987e-06, "output_power": 8.240551065062516e-06, "reference_active": true, "delta_db": -0.7895126095670567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.436572483663522e-06, "output_power": 3.688938758148928e-06, "reference_active": true, "delta_db": -1.6842373732789606}], "active_band_count": 3, "active_band_mae_db": 0.9356311737709623}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.4434083199144116e-05, "log_rms_mae_db": 1.1700407003066902, "rms_envelope_nrmse": 0.11466628580496724, "envelope_correlation": 0.9911350014798851, "rms_gain_error_db": -0.4431650018483736, "reference_mean_power": 0.0001096329653054558, "output_mean_power": 9.899759876330826e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.313909407445067e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3642314296732694e-05, "output_power": 2.196388676640301e-05, "reference_active": true, "delta_db": -0.319807904071437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.77067085832153e-06, "output_power": 5.00010843965102e-06, "reference_active": true, "delta_db": -1.3165227883117985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.651384635522461e-07, "output_power": 2.2056116215412277e-07, "reference_active": true, "delta_db": -3.2405321254125194}], "active_band_count": 3, "active_band_mae_db": 1.6256209392652516}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010577651466639849, "log_rms_mae_db": 1.9076376787429887, "rms_envelope_nrmse": 0.1268790758870263, "envelope_correlation": 0.9916153971700402, "rms_gain_error_db": 0.25341402515289724, "reference_mean_power": 0.00018020158938811219, "output_mean_power": 0.0001910293155714644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.705326923064497e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.613613486254781e-05, "output_power": 1.737796488332999e-05, "reference_active": true, "delta_db": 0.321994003182805}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.942687703713047e-06, "output_power": 8.317001571825506e-06, "reference_active": true, "delta_db": -0.31501281037031414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.043858273628482e-06, "output_power": 7.631718736494748e-07, "reference_active": true, "delta_db": -1.360191815192716}], "active_band_count": 3, "active_band_mae_db": 0.6657328762486117}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.602886913996206e-05, "log_rms_mae_db": 1.2433910967009112, "rms_envelope_nrmse": 0.06365810698607166, "envelope_correlation": 0.9982771078181684, "rms_gain_error_db": 0.26530260087005864, "reference_mean_power": 0.0001775816262598491, "output_mean_power": 0.00018876796281855799}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.444034789789241e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7728340440048457e-05, "output_power": 1.8564859102171788e-05, "reference_active": true, "delta_db": 0.20023574548538875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.955564320473026e-06, "output_power": 4.6717181292799585e-06, "reference_active": true, "delta_db": -0.25616485916180226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.260796671427193e-08, "output_power": 5.592513791730411e-08, "reference_active": true, "delta_db": -2.1904128522003803}], "active_band_count": 3, "active_band_mae_db": 0.8822711522825237}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8451471971732959, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "038472750637b8ad6424233665176ff4f1270f24d3b75943c48bf28f4cb9e078", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.974174106441207, "standard_deviation": 0.0017442476023016428, "minimum": 0.9724966501709388, "maximum": 0.9767626103550796, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8049472278688772, "standard_deviation": 0.031053937962834753, "minimum": 1.775339597815423, "maximum": 1.8693250936195205, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3104282367087179, "standard_deviation": 0.3744961099282855, "minimum": 0.8200102497572157, "maximum": 1.7580214872748805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18889927417326846, "standard_deviation": 0.006074088723240983, "minimum": 0.17966525634784855, "maximum": 0.19707230886163415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9921610948992163, "standard_deviation": 0.0014949892505064725, "minimum": 0.9888560875280821, "maximum": 0.9932408285682398, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1447805425096291, "standard_deviation": 0.03234894469870895, "minimum": 1.0886162195242801, "maximum": 1.1799566726029498, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.84571768750927, "standard_deviation": 0.4800812408355349, "minimum": 1.3055303405302126, "maximum": 2.529743727668184, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11153358255793482, "standard_deviation": 0.006034870777516936, "minimum": 0.10502573232526204, "maximum": 0.1208124409337409, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9888012696291987, "standard_deviation": 0.0029360606900524925, "minimum": 0.9843541053032134, "maximum": 0.9923697605080809, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8788485429776425, "standard_deviation": 0.03798131568252923, "minimum": 1.8180442828386583, "maximum": 1.924623552724017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6545310573835279, "standard_deviation": 0.43000934543343133, "minimum": 0.19918970439417868, "maximum": 1.3087113157845032, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1405644242753199, "standard_deviation": 0.018134753174138922, "minimum": 0.11578453690166814, "maximum": 0.16684290922362113, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9970214390750014, "standard_deviation": 0.0011531040782946393, "minimum": 0.995176637825826, "maximum": 0.9982771078181684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2753397308915262, "standard_deviation": 0.02707065453944914, "minimum": 1.2294195666093115, "maximum": 1.301346153721033, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8552087336713152, "standard_deviation": 0.5311039266761992, "minimum": 0.14960278626550025, "maximum": 1.6600854021896343, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07291986393357983, "standard_deviation": 0.012631305258294437, "minimum": 0.05541720997537701, "maximum": 0.09190684777083696, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9937660581825749, "standard_deviation": 0.000509509281275732, "minimum": 0.993095322358287, "maximum": 0.9947024678606429, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6845121240232896, "standard_deviation": 0.04001017369122306, "minimum": 1.6129351191124879, "maximum": 1.758837339712508, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.50773816122156, "standard_deviation": 0.15732748749573655, "minimum": 0.3207727551344754, "maximum": 0.8441209836352993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09189986505403565, "standard_deviation": 0.00350838775668271, "minimum": 0.08605815013638739, "maximum": 0.09612422474468928, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9979020697015958, "standard_deviation": 0.0001979295179711786, "minimum": 0.99762079402986, "maximum": 0.998252199438052, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0170736842890955, "standard_deviation": 0.04642342298565196, "minimum": 0.9533472739384445, "maximum": 1.0794462828884872, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5636751408774109, "standard_deviation": 0.18708482620632932, "minimum": 0.4013068308313681, "maximum": 0.9317053907115751, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054528516665966954, "standard_deviation": 0.0013523909121058483, "minimum": 0.05222976130868458, "maximum": 0.055907113834809605, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21897082838735016, "standard_deviation": 0.03833495698098971, "minimum": 0.1772659928438485, "maximum": 0.26248889448355706, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.690814084963496, "standard_deviation": 0.009798774221378833, "minimum": 5.678263686415815, "maximum": 5.705283549146748, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5419300638919536, "standard_deviation": 0.03656889895039948, "minimum": 0.504243674577223, "maximum": 0.6202480118659591, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0763108712659583, "standard_deviation": 0.03577669974902882, "minimum": -0.14416688866371935, "maximum": -0.026911285952148582, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9919993531002682, "standard_deviation": 0.0020677532780855956, "minimum": 0.9874032543371833, "maximum": 0.9940898363698022, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2173839978090135, "standard_deviation": 0.03517978969776332, "minimum": 1.177708923817533, "maximum": 1.2889048502748255, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.659152686209355, "standard_deviation": 0.3104938266683527, "minimum": 0.3187948517343314, "maximum": 1.23175849110168, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08895253057816987, "standard_deviation": 0.01080751214692525, "minimum": 0.07722586347708285, "maximum": 0.11235435149976052, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969708468505784, "standard_deviation": 0.001204883586907475, "minimum": 0.9946986464191249, "maximum": 0.998075786448781, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6601133084290731, "standard_deviation": 0.039183015197394755, "minimum": 0.6001582515560017, "maximum": 0.7004835024329894, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7874641945607037, "standard_deviation": 0.5187159387220037, "minimum": 0.2423991327329349, "maximum": 1.8264735753326464, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.053247057241527423, "standard_deviation": 0.009815605113299795, "minimum": 0.0435349774172753, "maximum": 0.07070336802787787, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8224-274384-0010", "language": "en", "style": "normal", "speaker": "librispeech_8224", "source_utterance_id": "8224-274384-0010", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8224-274384-0010.wav", "source_file_sha256": "a7b2c6f9a59ea20b14f4241064d11f7d00665d69e97440a8f4d1765440d9529b", "decoded_pcm_sha256": "bfff7d3cf9d00e04b5e5f471ec3e7e23254f82646da0069c2f8d619c084a00f4", "original_source_sha256": "677a2c48f5ddd5929e5c68e682b66998c48322833dac3baa41a086f87ef15077", "corpus_entry_id": "librispeech_8224-274384-0010"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.9012131690979004, "predicted_mos": 3.9012131690979004, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0b685c0750383611c038234e972881dd66164719a4c1d40086e12c28397b39de", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.2363766380003653, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8224-274384-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "0b685c0750383611c038234e972881dd66164719a4c1d40086e12c28397b39de", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "742ea1f463a5d6275ca4dea75eb76e3d478fb9b550110c351f3d7dd4d220e2be"}, "rows": [{"case": "librispeech_8224-274384-0010", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.846132278442383, "predicted_mos": 2.846132278442383, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "14ed4df693979a6c22629c5f8526f195ef829b137ad14edc5c63432527e32a32", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.207306112999504, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8224-274384-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "14ed4df693979a6c22629c5f8526f195ef829b137ad14edc5c63432527e32a32", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.116571313616395, "全帯域_10ms_rms_envelope_nrmse": 0.07657823221590346, "全帯域_10ms_envelope_correlation": 0.9941166922425843, "全帯域_10ms_modulation_mae_db": 0.2709406258653671, "全帯域_40ms_log_rms_mae_db": 0.6451411869652007, "全帯域_40ms_rms_envelope_nrmse": 0.043965853067405615, "全帯域_40ms_envelope_correlation": 0.9980717167087878, "全帯域_40ms_modulation_mae_db": 0.4202928965782095, "300–1000 Hz_10ms_log_rms_mae_db": 1.5733945255673434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0918335640905146, "300–1000 Hz_10ms_envelope_correlation": 0.9939815534461041, "300–1000 Hz_10ms_modulation_mae_db": 0.7955726260559116, "300–1000 Hz_40ms_log_rms_mae_db": 0.9605126583583109, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05760897881988081, "300–1000 Hz_40ms_envelope_correlation": 0.9977899000235828, "300–1000 Hz_40ms_modulation_mae_db": 0.6864493250276468, "1000–3000 Hz_10ms_log_rms_mae_db": 1.723218545970502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1649475545355467, "1000–3000 Hz_10ms_envelope_correlation": 0.9794901806781386, "1000–3000 Hz_10ms_modulation_mae_db": 0.7656971160104016, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0557840703530912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08404492552363735, "1000–3000 Hz_40ms_envelope_correlation": 0.9944949831981408, "1000–3000 Hz_40ms_modulation_mae_db": 0.16400868192603013, "3000–6000 Hz_10ms_log_rms_mae_db": 1.86356351373685, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2120814381953484, "3000–6000 Hz_10ms_envelope_correlation": 0.9737080833838745, "3000–6000 Hz_10ms_modulation_mae_db": 0.3754980184176699, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3440824140237992, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15289679313814705, "3000–6000 Hz_40ms_envelope_correlation": 0.9860505318224124, "3000–6000 Hz_40ms_modulation_mae_db": 0.09032097597007749, "mr_stft_db": 5.66199940154337, "band_mae_db": 0.1820255748703502, "rms_gain_error_db": -0.13082074219255316, "output_peak": 0.42138015935753836}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0862676010230843, "全帯域_10ms_rms_envelope_nrmse": 0.08336964662916521, "全帯域_10ms_envelope_correlation": 0.9928700078952443, "全帯域_10ms_modulation_mae_db": 0.20111426398400734, "全帯域_40ms_log_rms_mae_db": 0.6241109366415077, "全帯域_40ms_rms_envelope_nrmse": 0.04980937771322087, "全帯域_40ms_envelope_correlation": 0.9972913775044844, "全帯域_40ms_modulation_mae_db": 0.15800036975371892, "300–1000 Hz_10ms_log_rms_mae_db": 1.593701066453137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10483237288408011, "300–1000 Hz_10ms_envelope_correlation": 0.9919182498239371, "300–1000 Hz_10ms_modulation_mae_db": 0.5371755067233653, "300–1000 Hz_40ms_log_rms_mae_db": 0.9311773067412316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06337704309958299, "300–1000 Hz_40ms_envelope_correlation": 0.9971168158570667, "300–1000 Hz_40ms_modulation_mae_db": 0.483515388249822, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7445092154885864, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14471031083102126, "1000–3000 Hz_10ms_envelope_correlation": 0.9843630130392224, "1000–3000 Hz_10ms_modulation_mae_db": 0.3935427330434565, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0696968477988806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07594674813313286, "1000–3000 Hz_40ms_envelope_correlation": 0.9957715637594867, "1000–3000 Hz_40ms_modulation_mae_db": 0.5940335889608098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.868234439368148, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21232202242048648, "3000–6000 Hz_10ms_envelope_correlation": 0.9722910728215695, "3000–6000 Hz_10ms_modulation_mae_db": 0.33272464963340903, "3000–6000 Hz_40ms_log_rms_mae_db": 1.351785986236204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15420225358225462, "3000–6000 Hz_40ms_envelope_correlation": 0.9848121485699194, "3000–6000 Hz_40ms_modulation_mae_db": 0.2547085274669594, "mr_stft_db": 5.674492404930748, "band_mae_db": 0.16987099001066672, "rms_gain_error_db": -0.0735584088255163, "output_peak": 0.3933747179910375}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0992524452085692, "全帯域_10ms_rms_envelope_nrmse": 0.08188503778572022, "全帯域_10ms_envelope_correlation": 0.9931171262238434, "全帯域_10ms_modulation_mae_db": 0.382802455324932, "全帯域_40ms_log_rms_mae_db": 0.5648192352705685, "全帯域_40ms_rms_envelope_nrmse": 0.03854421045197329, "全帯域_40ms_envelope_correlation": 0.9983787618794099, "全帯域_40ms_modulation_mae_db": 0.45888910170942454, "300–1000 Hz_10ms_log_rms_mae_db": 1.5753243010754594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10182945174513545, "300–1000 Hz_10ms_envelope_correlation": 0.9921962196922309, "300–1000 Hz_10ms_modulation_mae_db": 0.539885071212645, "300–1000 Hz_40ms_log_rms_mae_db": 0.8788557773204061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05545622844927114, "300–1000 Hz_40ms_envelope_correlation": 0.9976190784617164, "300–1000 Hz_40ms_modulation_mae_db": 0.4903341521838307, "1000–3000 Hz_10ms_log_rms_mae_db": 1.751826933734148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14676049418925344, "1000–3000 Hz_10ms_envelope_correlation": 0.984213162124364, "1000–3000 Hz_10ms_modulation_mae_db": 0.6100399296363673, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0182049065784018, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07036483398687109, "1000–3000 Hz_40ms_envelope_correlation": 0.9969415060798615, "1000–3000 Hz_40ms_modulation_mae_db": 0.3507263603422121, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8640054484319997, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2058175304954366, "3000–6000 Hz_10ms_envelope_correlation": 0.973934401464047, "3000–6000 Hz_10ms_modulation_mae_db": 0.3955805978441444, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3494499703014837, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15366380192933332, "3000–6000 Hz_40ms_envelope_correlation": 0.9848825445436722, "3000–6000 Hz_40ms_modulation_mae_db": 0.33202558515624186, "mr_stft_db": 5.677696120226229, "band_mae_db": 0.14489719009695332, "rms_gain_error_db": -0.052183103703647205, "output_peak": 0.38585256957674163}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0169573978223228, "全帯域_10ms_rms_envelope_nrmse": 0.0722623836870178, "全帯域_10ms_envelope_correlation": 0.9946547181245585, "全帯域_10ms_modulation_mae_db": 0.20221333112453546, "全帯域_40ms_log_rms_mae_db": 0.5650649058061947, "全帯域_40ms_rms_envelope_nrmse": 0.035419193262658334, "全帯域_40ms_envelope_correlation": 0.9986575264242256, "全帯域_40ms_modulation_mae_db": 0.22461850869792244, "300–1000 Hz_10ms_log_rms_mae_db": 1.6481457870235263, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10298698903050418, "300–1000 Hz_10ms_envelope_correlation": 0.9920806822626511, "300–1000 Hz_10ms_modulation_mae_db": 0.311764611525026, "300–1000 Hz_40ms_log_rms_mae_db": 1.001510145682393, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054380644804201184, "300–1000 Hz_40ms_envelope_correlation": 0.9978132733754227, "300–1000 Hz_40ms_modulation_mae_db": 0.4523983520372292, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7769803419739794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1763641261945867, "1000–3000 Hz_10ms_envelope_correlation": 0.9765638740522418, "1000–3000 Hz_10ms_modulation_mae_db": 0.4983105031374457, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1346032516175568, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09927887100358122, "1000–3000 Hz_40ms_envelope_correlation": 0.9922851346519574, "1000–3000 Hz_40ms_modulation_mae_db": 0.42716627614903574, "3000–6000 Hz_10ms_log_rms_mae_db": 1.898934259555617, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20293513984723216, "3000–6000 Hz_10ms_envelope_correlation": 0.9747405218516286, "3000–6000 Hz_10ms_modulation_mae_db": 0.35963675785873916, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4228659731476847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15564590276504361, "3000–6000 Hz_40ms_envelope_correlation": 0.9845278168480479, "3000–6000 Hz_40ms_modulation_mae_db": 0.347892975871194, "mr_stft_db": 5.675061282290962, "band_mae_db": 0.17232728377054327, "rms_gain_error_db": -0.0667944967220103, "output_peak": 0.3959254290154375}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1112363386064663, "全帯域_10ms_rms_envelope_nrmse": 0.08928439606411509, "全帯域_10ms_envelope_correlation": 0.9918402494458799, "全帯域_10ms_modulation_mae_db": 0.17293960019028845, "全帯域_40ms_log_rms_mae_db": 0.6290318261415343, "全帯域_40ms_rms_envelope_nrmse": 0.05250586714453926, "全帯域_40ms_envelope_correlation": 0.997031308140489, "全帯域_40ms_modulation_mae_db": 0.4519159934510772, "300–1000 Hz_10ms_log_rms_mae_db": 1.5597020607891467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11588634867633511, "300–1000 Hz_10ms_envelope_correlation": 0.9899628197898005, "300–1000 Hz_10ms_modulation_mae_db": 0.2937197183408637, "300–1000 Hz_40ms_log_rms_mae_db": 0.9526757673612838, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06801616244086789, "300–1000 Hz_40ms_envelope_correlation": 0.9965290883489221, "300–1000 Hz_40ms_modulation_mae_db": 0.580650533555914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7900344792188396, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1647782048003269, "1000–3000 Hz_10ms_envelope_correlation": 0.9798846599216327, "1000–3000 Hz_10ms_modulation_mae_db": 0.6860408894466343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.137387368083321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09050980477871233, "1000–3000 Hz_40ms_envelope_correlation": 0.9943178162462941, "1000–3000 Hz_40ms_modulation_mae_db": 1.1967350022315315, "3000–6000 Hz_10ms_log_rms_mae_db": 1.871771510264445, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21601731549437636, "3000–6000 Hz_10ms_envelope_correlation": 0.9709847512492038, "3000–6000 Hz_10ms_modulation_mae_db": 0.6695095978915573, "3000–6000 Hz_40ms_log_rms_mae_db": 1.324996368799488, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17668879509361043, "3000–6000 Hz_40ms_envelope_correlation": 0.9800597842523101, "3000–6000 Hz_40ms_modulation_mae_db": 0.52395437555421, "mr_stft_db": 5.652952466106086, "band_mae_db": 0.18120808930530385, "rms_gain_error_db": -0.0967847670023268, "output_peak": 0.4152406373754022}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.033983374723199, "全帯域_10ms_rms_envelope_nrmse": 0.07471250304313315, "全帯域_10ms_envelope_correlation": 0.9942953103778266, "全帯域_10ms_modulation_mae_db": 0.09781401758579054, "全帯域_40ms_log_rms_mae_db": 0.5412973228322163, "全帯域_40ms_rms_envelope_nrmse": 0.04033480614855458, "全帯域_40ms_envelope_correlation": 0.9982112896962008, "全帯域_40ms_modulation_mae_db": 0.1504658303193705, "300–1000 Hz_10ms_log_rms_mae_db": 1.5982269501235618, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09828435821167064, "300–1000 Hz_10ms_envelope_correlation": 0.9927084628815481, "300–1000 Hz_10ms_modulation_mae_db": 0.20462205074481574, "300–1000 Hz_40ms_log_rms_mae_db": 0.9410974616801255, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055506483352914596, "300–1000 Hz_40ms_envelope_correlation": 0.9975597612815925, "300–1000 Hz_40ms_modulation_mae_db": 0.3838068527571439, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7930304291608412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16888348751677357, "1000–3000 Hz_10ms_envelope_correlation": 0.9785384767296174, "1000–3000 Hz_10ms_modulation_mae_db": 0.4860130100240261, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1381278874792318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10338839534201727, "1000–3000 Hz_40ms_envelope_correlation": 0.9918178850654609, "1000–3000 Hz_40ms_modulation_mae_db": 0.8465301938125934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.891411873058264, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1972985247349229, "3000–6000 Hz_10ms_envelope_correlation": 0.9760116561310194, "3000–6000 Hz_10ms_modulation_mae_db": 0.4546320295185897, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4053885811416722, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15148784422173397, "3000–6000 Hz_40ms_envelope_correlation": 0.9852965272641642, "3000–6000 Hz_40ms_modulation_mae_db": 0.3804335513932526, "mr_stft_db": 5.666851649922883, "band_mae_db": 0.13924661622346027, "rms_gain_error_db": -0.024641689137686402, "output_peak": 0.3964698538742363}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0607366090313168, "全帯域_10ms_rms_envelope_nrmse": 0.07467942650317336, "全帯域_10ms_envelope_correlation": 0.9942683487403826, "全帯域_10ms_modulation_mae_db": 0.6535770307042635, "全帯域_40ms_log_rms_mae_db": 0.5473869944846882, "全帯域_40ms_rms_envelope_nrmse": 0.0376671134841084, "全帯域_40ms_envelope_correlation": 0.9984236767850484, "全帯域_40ms_modulation_mae_db": 0.607843024797184, "300–1000 Hz_10ms_log_rms_mae_db": 1.6049947752350537, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10427610084274666, "300–1000 Hz_10ms_envelope_correlation": 0.9917768798512405, "300–1000 Hz_10ms_modulation_mae_db": 0.987786740958312, "300–1000 Hz_40ms_log_rms_mae_db": 0.9220259901074513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05536666517696934, "300–1000 Hz_40ms_envelope_correlation": 0.9975641505636876, "300–1000 Hz_40ms_modulation_mae_db": 1.1647517110411865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8406279060508397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1758664900162081, "1000–3000 Hz_10ms_envelope_correlation": 0.9769165402563166, "1000–3000 Hz_10ms_modulation_mae_db": 0.8507943735771665, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1383610049456392, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0982565891258275, "1000–3000 Hz_40ms_envelope_correlation": 0.9930549927272241, "1000–3000 Hz_40ms_modulation_mae_db": 0.4573399143038644, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8479310302348808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20711668860883575, "3000–6000 Hz_10ms_envelope_correlation": 0.9746360450437614, "3000–6000 Hz_10ms_modulation_mae_db": 0.3758064900795887, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3463282278462865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15304902618431093, "3000–6000 Hz_40ms_envelope_correlation": 0.9857826228414935, "3000–6000 Hz_40ms_modulation_mae_db": 0.14627251547469222, "mr_stft_db": 5.666682205514153, "band_mae_db": 0.14617798623600473, "rms_gain_error_db": -0.017721785534348153, "output_peak": 0.47538403172078014}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0955700778451378, "全帯域_10ms_rms_envelope_nrmse": 0.08300217884251626, "全帯域_10ms_envelope_correlation": 0.9930141292003846, "全帯域_10ms_modulation_mae_db": 0.257674201274734, "全帯域_40ms_log_rms_mae_db": 0.6201018641318179, "全帯域_40ms_rms_envelope_nrmse": 0.04626739663202106, "全帯域_40ms_envelope_correlation": 0.9978022873907763, "全帯域_40ms_modulation_mae_db": 0.1767594435746079, "300–1000 Hz_10ms_log_rms_mae_db": 1.5191114200729907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10677113348629097, "300–1000 Hz_10ms_envelope_correlation": 0.9915910781335902, "300–1000 Hz_10ms_modulation_mae_db": 0.6442668711172446, "300–1000 Hz_40ms_log_rms_mae_db": 0.8995514966866315, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058691585928685716, "300–1000 Hz_40ms_envelope_correlation": 0.9975787667660421, "300–1000 Hz_40ms_modulation_mae_db": 0.5557681440442661, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7340661703031248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16397868932148735, "1000–3000 Hz_10ms_envelope_correlation": 0.9803088102529189, "1000–3000 Hz_10ms_modulation_mae_db": 0.30261738198505383, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1261630010345869, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09331199245533618, "1000–3000 Hz_40ms_envelope_correlation": 0.9942590599482826, "1000–3000 Hz_40ms_modulation_mae_db": 0.8339029942854075, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8629109286045753, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2098682872664105, "3000–6000 Hz_10ms_envelope_correlation": 0.9727244253848054, "3000–6000 Hz_10ms_modulation_mae_db": 1.0060634855212136, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3456897154726495, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16306223841349327, "3000–6000 Hz_40ms_envelope_correlation": 0.9832570653397298, "3000–6000 Hz_40ms_modulation_mae_db": 1.0193080111879385, "mr_stft_db": 5.657853745933473, "band_mae_db": 0.18297467533653988, "rms_gain_error_db": -0.11705453226859967, "output_peak": 0.38013514159999773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8376189272831, "spectral_convergence": 0.25085905269234815, "spectral_convergence_status": "ok", "amplitude_floor": 1.597965137573199e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5760267567218476, "spectral_convergence": 0.2302573734222072, "spectral_convergence_status": "ok", "amplitude_floor": 1.057274363712373e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.572352520625163, "spectral_convergence": 0.23761579429727236, "spectral_convergence_status": "ok", "amplitude_floor": 9.008040625007043e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007434014661781762, "output_power": 0.0007510394580461735, "delta_db": 0.04439341228991748, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002448126060340392, "output_power": 0.0023255875612578558, "delta_db": -0.22301081113432852, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00039049712732524606, "output_power": 0.0003868048297827328, "delta_db": -0.04125955156779111, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012090645510899839, "output_power": 0.00011481585829086762, "delta_db": -0.2244761147775246, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.957546973001399e-05, "output_power": 6.0647367520810216e-05, "delta_db": 0.07744479205903812, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.142800821939945e-05, "output_power": 9.097680661335727e-05, "delta_db": 0.48156876739350146, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1820255748703502, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13082074219255316, "pitch": null, "output_rms_dbfs": -24.28306085646557, "output_peak": 0.42138015935753836, "reference_rms_dbfs": -24.152240114273017, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.66199940154337}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019565573818943504, "log_rms_mae_db": 1.116571313616395, "rms_envelope_nrmse": 0.07657823221590346, "envelope_correlation": 0.9941166922425843, "rms_gain_error_db": -0.13079475624197734, "reference_mean_power": 0.0038470996951946897, "output_mean_power": 0.0037329654062804624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.743468430119212e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007829242029289011, "output_power": 0.0007370527830515603, "reference_active": true, "delta_db": -0.26221128442999764}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011656588263872551, "output_power": 0.00011094753771600996, "reference_active": true, "delta_db": -0.2145378823449294}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.15404555717803e-05, "output_power": 2.9191787633083428e-05, "reference_active": true, "delta_db": -0.3360727108211743}], "active_band_count": 3, "active_band_mae_db": 0.2709406258653671}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018472557440126405, "log_rms_mae_db": 0.6451411869652007, "rms_envelope_nrmse": 0.043965853067405615, "envelope_correlation": 0.9980717167087878, "rms_gain_error_db": -0.13057556580665466, "reference_mean_power": 0.0038659686805529734, "output_mean_power": 0.0037514639271088008}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.614770154461247e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007106100759473788, "output_power": 0.000669288720083353, "reference_active": true, "delta_db": -0.2601785554029267}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.7305685337334634e-05, "output_power": 5.274482364747909e-05, "reference_active": true, "delta_db": -0.36017866163704737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9649709160477995e-06, "output_power": 2.277237222505313e-06, "reference_active": true, "delta_db": 0.6405214726946544}], "active_band_count": 3, "active_band_mae_db": 0.4202928965782095}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016965929826439602, "log_rms_mae_db": 1.5733945255673434, "rms_envelope_nrmse": 0.0918335640905146, "envelope_correlation": 0.9939815534461041, "rms_gain_error_db": -0.2220313240899181, "reference_mean_power": 0.0021633278035333714, "output_mean_power": 0.0020555081332970238}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3458386409803657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007177587670080827, "output_power": 0.000653908423730492, "reference_active": true, "delta_db": -0.40461573821883595}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.048912255781678e-05, "output_power": 7.93203054401011e-05, "reference_active": true, "delta_db": -0.5721199914609131}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.109541579796594e-05, "output_power": 2.2474919951350465e-05, "reference_active": true, "delta_db": -1.4099821484879858}], "active_band_count": 3, "active_band_mae_db": 0.7955726260559116}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015157137475553504, "log_rms_mae_db": 0.9605126583583109, "rms_envelope_nrmse": 0.05760897881988081, "envelope_correlation": 0.9977899000235828, "rms_gain_error_db": -0.22204951119893998, "reference_mean_power": 0.002174161455071234, "output_mean_power": 0.00206579318756844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2731603714689836e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006867133707128568, "output_power": 0.0006256336844117322, "reference_active": true, "delta_db": -0.4045538014490801}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.067232594302272e-05, "output_power": 3.415599202216126e-05, "reference_active": true, "delta_db": -0.758321058243919}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.08557261679534e-06, "output_power": 1.6965943033859764e-06, "reference_active": true, "delta_db": -0.8964731153899413}], "active_band_count": 3, "active_band_mae_db": 0.6864493250276468}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.465903254113391e-05, "log_rms_mae_db": 1.723218545970502, "rms_envelope_nrmse": 0.1649475545355467, "envelope_correlation": 0.9794901806781386, "rms_gain_error_db": -0.11426306020269805, "reference_mean_power": 0.00014077271984820836, "output_mean_power": 0.00013711728191280925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.622290255773331e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.707421130487826e-05, "output_power": 3.6503543592719583e-05, "reference_active": true, "delta_db": -0.06736895098187906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.08772682349557e-06, "output_power": 4.866742796145665e-06, "reference_active": true, "delta_db": -0.19285389892161625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8011808387082025e-06, "output_power": 2.879016593758536e-06, "reference_active": true, "delta_db": 2.0368684981277094}], "active_band_count": 3, "active_band_mae_db": 0.7656971160104016}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.107059424790359e-05, "log_rms_mae_db": 1.0557840703530912, "rms_envelope_nrmse": 0.08404492552363735, "envelope_correlation": 0.9944949831981408, "rms_gain_error_db": -0.11453435927693442, "reference_mean_power": 0.00014139647469838974, "output_mean_power": 0.0001377162364997252}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.069193023932203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.382591725297068e-05, "output_power": 3.381842138894166e-05, "reference_active": true, "delta_db": -0.000962508456313183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1244897853414024e-06, "output_power": 2.0770378797821177e-06, "reference_active": true, "delta_db": -0.09810230307937728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.400225448356044e-07, "output_power": 1.279092177770057e-07, "reference_active": true, "delta_db": -0.3929612342423999}], "active_band_count": 3, "active_band_mae_db": 0.16400868192603013}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.428054041295646e-05, "log_rms_mae_db": 1.86356351373685, "rms_envelope_nrmse": 0.2120814381953484, "envelope_correlation": 0.9737080833838745, "rms_gain_error_db": 0.2701684114897999, "reference_mean_power": 8.952225188868405e-05, "output_mean_power": 9.526817357768198e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.314971416941356e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.877588546016404e-05, "output_power": 5.985858146033324e-05, "reference_active": true, "delta_db": 0.07927240132315538}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.04956306951581e-06, "output_power": 8.825177024454063e-06, "reference_active": true, "delta_db": -0.10904185458692944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5294615410314917e-06, "output_power": 3.1393957104797034e-06, "reference_active": true, "delta_db": 0.9381797993429247}], "active_band_count": 3, "active_band_mae_db": 0.3754980184176699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5452516433588795e-05, "log_rms_mae_db": 1.3440824140237992, "rms_envelope_nrmse": 0.15289679313814705, "envelope_correlation": 0.9860505318224124, "rms_gain_error_db": 0.2701760545911256, "reference_mean_power": 8.996899751083782e-05, "output_mean_power": 9.574376174622675e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.951283845645369e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.884212028117595e-05, "output_power": 6.032834065621686e-05, "reference_active": true, "delta_db": 0.10833066678324182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.436071463262977e-06, "output_power": 4.402026914076425e-06, "reference_active": true, "delta_db": -0.033458403264473524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7711408251558852e-07, "output_power": 1.8246118014997526e-07, "reference_active": true, "delta_db": 0.1291738578625171}], "active_band_count": 3, "active_band_mae_db": 0.09032097597007749}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8963737107407982, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2d749b7b5878d06c8f3d33495515fe628850c9cf3ca5ec8502e3d66ea01d831f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9800348396318066, "standard_deviation": 0.002942210687406355, "minimum": 0.9765638740522418, "maximum": 0.9843630130392224, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7692867527376077, "standard_deviation": 0.03859038097978104, "minimum": 1.723218545970502, "maximum": 1.8406279060508397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5741319921075689, "standard_deviation": 0.18780642332311484, "minimum": 0.30261738198505383, "maximum": 0.8507943735771665, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1632861696756505, "standard_deviation": 0.01185781121467384, "minimum": 0.14471031083102126, "maximum": 0.1763641261945867, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941178677095885, "standard_deviation": 0.0017191488292918808, "minimum": 0.9918178850654609, "maximum": 0.9969415060798615, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1022910422363386, "standard_deviation": 0.04739918809038349, "minimum": 1.0182049065784018, "maximum": 1.1383610049456392, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6088053765014356, "standard_deviation": 0.33240356365598533, "minimum": 0.16400868192603013, "maximum": 1.1967350022315315, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08938777004363947, "standard_deviation": 0.011712294981432525, "minimum": 0.07036483398687109, "maximum": 0.10338839534201727, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9736288696662387, "standard_deviation": 0.0015868455101135783, "minimum": 0.9709847512492038, "maximum": 0.9760116561310194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8710953754068473, "standard_deviation": 0.016505296648832372, "minimum": 1.8479310302348808, "maximum": 1.898934259555617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.49618145334561403, "standard_deviation": 0.23185168380052235, "minimum": 0.33272464963340903, "maximum": 1.0060634855212136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20793211838288114, "standard_deviation": 0.005961624786302042, "minimum": 0.1972985247349229, "maximum": 0.21601731549437636, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9843336301852186, "standard_deviation": 0.0019260339685698404, "minimum": 0.9800597842523101, "maximum": 0.9860505318224124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3613234046211584, "standard_deviation": 0.033906049114509146, "minimum": 1.324996368799488, "maximum": 1.4228659731476847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.38686456475932074, "standard_deviation": 0.2894905964264455, "minimum": 0.09032097597007749, "maximum": 1.0193080111879385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1575870819159909, "standard_deviation": 0.00849591169988144, "minimum": 0.15148784422173397, "maximum": 0.17668879509361043, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920269932351378, "standard_deviation": 0.0011238608470805192, "minimum": 0.9899628197898005, "maximum": 0.9939815534461041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5840751107925275, "standard_deviation": 0.03749399140766135, "minimum": 1.5191114200729907, "maximum": 1.6481457870235263, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.539349149584773, "standard_deviation": 0.26777662696295207, "minimum": 0.20462205074481574, "maximum": 0.987786740958312, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10333753987090971, "standard_deviation": 0.006893588214626135, "minimum": 0.0918335640905146, "maximum": 0.11588634867633511, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997446354334754, "standard_deviation": 0.0004270713361113529, "minimum": 0.9965290883489221, "maximum": 0.9978132733754227, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9359258254922291, "standard_deviation": 0.03781337221352798, "minimum": 0.8788557773204061, "maximum": 1.001510145682393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5997093073621299, "standard_deviation": 0.24578421144947782, "minimum": 0.3838068527571439, "maximum": 1.1647517110411865, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05855047400904671, "standard_deviation": 0.004783389435099151, "minimum": 0.054380644804201184, "maximum": 0.06801616244086789, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1648410507312278, "standard_deviation": 0.01841778165755721, "minimum": 0.13924661622346027, "maximum": 0.18297467533653988, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.666698659558488, "standard_deviation": 0.00879055581841338, "minimum": 5.652952466106086, "maximum": 5.677696120226229, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4079703175638964, "standard_deviation": 0.030546577064719253, "minimum": 0.38013514159999773, "maximum": 0.47538403172078014, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07244494067333601, "standard_deviation": 0.04090696926710349, "minimum": -0.13082074219255316, "maximum": -0.017721785534348153, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9935220727813381, "standard_deviation": 0.0009611740371701468, "minimum": 0.9918402494458799, "maximum": 0.9946547181245585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0775718947345614, "standard_deviation": 0.036608642195396696, "minimum": 1.0169573978223228, "maximum": 1.116571313616395, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2798844407567398, "standard_deviation": 0.17222739785558577, "minimum": 0.09781401758579054, "maximum": 0.6535770307042635, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07947172559634308, "standard_deviation": 0.005803123673032756, "minimum": 0.0722623836870178, "maximum": 0.08928439606411509, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979834930661777, "standard_deviation": 0.000571027945881545, "minimum": 0.997031308140489, "maximum": 0.9986575264242256, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5921192840342161, "standard_deviation": 0.04147534567715133, "minimum": 0.5412973228322163, "maximum": 0.6451411869652007, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3310981461101894, "standard_deviation": 0.17452798954032184, "minimum": 0.1504658303193705, "maximum": 0.607843024797184, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04306422723806018, "standard_deviation": 0.006108197243367995, "minimum": 0.035419193262658334, "maximum": 0.05250586714453926, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274384-0010", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.880906343460083, "predicted_mos": 2.880906343460083, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "98e26795bc35a2c465284f63f87103875f32cc9c15a5e5bacc68634c9a65d391", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7678175000473857, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1189297217115433, "全帯域_10ms_rms_envelope_nrmse": 0.0770102803198733, "全帯域_10ms_envelope_correlation": 0.9940359498646059, "全帯域_10ms_modulation_mae_db": 0.32414071764280494, "全帯域_40ms_log_rms_mae_db": 0.6500534387283545, "全帯域_40ms_rms_envelope_nrmse": 0.04449323640064907, "全帯域_40ms_envelope_correlation": 0.998004988840657, "全帯域_40ms_modulation_mae_db": 0.4082569485108393, "300–1000 Hz_10ms_log_rms_mae_db": 1.560964584558203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09549336668796492, "300–1000 Hz_10ms_envelope_correlation": 0.993391613956891, "300–1000 Hz_10ms_modulation_mae_db": 0.8860161578188067, "300–1000 Hz_40ms_log_rms_mae_db": 0.9505359255999434, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060135586794007906, "300–1000 Hz_40ms_envelope_correlation": 0.9974736183406736, "300–1000 Hz_40ms_modulation_mae_db": 0.8800062936624208, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7301565808410728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16005849474008357, "1000–3000 Hz_10ms_envelope_correlation": 0.9808392079898464, "1000–3000 Hz_10ms_modulation_mae_db": 0.5954492296149607, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0820971766152292, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0888500934972116, "1000–3000 Hz_40ms_envelope_correlation": 0.9941177955920399, "1000–3000 Hz_40ms_modulation_mae_db": 0.30410869639856924, "3000–6000 Hz_10ms_log_rms_mae_db": 1.858726911288704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2167306703506722, "3000–6000 Hz_10ms_envelope_correlation": 0.9727070066269281, "3000–6000 Hz_10ms_modulation_mae_db": 0.46744569741174696, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3732428166575894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1528895441366929, "3000–6000 Hz_40ms_envelope_correlation": 0.9861675779897836, "3000–6000 Hz_40ms_modulation_mae_db": 0.21680389510629752, "mr_stft_db": 5.61885994146483, "band_mae_db": 0.18794494349071691, "rms_gain_error_db": -0.12679180684046346, "output_peak": 0.42834844743997025}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1114455548389695, "全帯域_10ms_rms_envelope_nrmse": 0.086879853488725, "全帯域_10ms_envelope_correlation": 0.9922609921282995, "全帯域_10ms_modulation_mae_db": 0.13205625531123347, "全帯域_40ms_log_rms_mae_db": 0.6496889875634886, "全帯域_40ms_rms_envelope_nrmse": 0.05220277360464472, "全帯域_40ms_envelope_correlation": 0.9970332320018994, "全帯域_40ms_modulation_mae_db": 0.2076527810863086, "300–1000 Hz_10ms_log_rms_mae_db": 1.560647816244061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10936073524803457, "300–1000 Hz_10ms_envelope_correlation": 0.9911853176880117, "300–1000 Hz_10ms_modulation_mae_db": 0.7775558267421778, "300–1000 Hz_40ms_log_rms_mae_db": 0.9271644075310973, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06516034409357493, "300–1000 Hz_40ms_envelope_correlation": 0.9969480832731419, "300–1000 Hz_40ms_modulation_mae_db": 0.7794879049353794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7568160767054992, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14325292916092167, "1000–3000 Hz_10ms_envelope_correlation": 0.9851211390655372, "1000–3000 Hz_10ms_modulation_mae_db": 0.2869581028091739, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0981150494121361, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07863417795390493, "1000–3000 Hz_40ms_envelope_correlation": 0.9960495769426603, "1000–3000 Hz_40ms_modulation_mae_db": 0.7490203724695732, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8678580363702941, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21134129217607595, "3000–6000 Hz_10ms_envelope_correlation": 0.9726844079813703, "3000–6000 Hz_10ms_modulation_mae_db": 0.20264426623068074, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3957126151130894, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15350740566783194, "3000–6000 Hz_40ms_envelope_correlation": 0.9849848749908541, "3000–6000 Hz_40ms_modulation_mae_db": 0.12789549388311294, "mr_stft_db": 5.628568463850225, "band_mae_db": 0.19311599895374928, "rms_gain_error_db": -0.08239922900118844, "output_peak": 0.3963211568466354}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0760602698446526, "全帯域_10ms_rms_envelope_nrmse": 0.08129001634248162, "全帯域_10ms_envelope_correlation": 0.9932168474864836, "全帯域_10ms_modulation_mae_db": 0.36471234219508414, "全帯域_40ms_log_rms_mae_db": 0.5450965817763845, "全帯域_40ms_rms_envelope_nrmse": 0.03611345078802756, "全帯域_40ms_envelope_correlation": 0.998578474691857, "全帯域_40ms_modulation_mae_db": 0.4532204193954071, "300–1000 Hz_10ms_log_rms_mae_db": 1.6237432842229762, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09947685874675018, "300–1000 Hz_10ms_envelope_correlation": 0.9925415813220971, "300–1000 Hz_10ms_modulation_mae_db": 0.5515956288019558, "300–1000 Hz_40ms_log_rms_mae_db": 0.9112019731694883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05100592449156464, "300–1000 Hz_40ms_envelope_correlation": 0.9979780941106496, "300–1000 Hz_40ms_modulation_mae_db": 0.5016989759399674, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7866267704616614, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14885436371281927, "1000–3000 Hz_10ms_envelope_correlation": 0.9840731695452679, "1000–3000 Hz_10ms_modulation_mae_db": 0.36996446692401924, "1000–3000 Hz_40ms_log_rms_mae_db": 1.066104648419687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07965552514475406, "1000–3000 Hz_40ms_envelope_correlation": 0.996242193076321, "1000–3000 Hz_40ms_modulation_mae_db": 0.32105688957158973, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8523223810355174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20480571438463027, "3000–6000 Hz_10ms_envelope_correlation": 0.9742151410438673, "3000–6000 Hz_10ms_modulation_mae_db": 0.3985487886090624, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3892864817420318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15206541441594718, "3000–6000 Hz_40ms_envelope_correlation": 0.9851632768469659, "3000–6000 Hz_40ms_modulation_mae_db": 0.2997706409156575, "mr_stft_db": 5.62452047647935, "band_mae_db": 0.15550464249954674, "rms_gain_error_db": -0.05021280343417607, "output_peak": 0.3893560145132104}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.034123327481177, "全帯域_10ms_rms_envelope_nrmse": 0.0724412332331893, "全帯域_10ms_envelope_correlation": 0.9946215946867547, "全帯域_10ms_modulation_mae_db": 0.19458294721143599, "全帯域_40ms_log_rms_mae_db": 0.560527037583182, "全帯域_40ms_rms_envelope_nrmse": 0.03579364428747994, "全帯域_40ms_envelope_correlation": 0.9986187703278986, "全帯域_40ms_modulation_mae_db": 0.18816528282096714, "300–1000 Hz_10ms_log_rms_mae_db": 1.6240715299449116, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10287199603050456, "300–1000 Hz_10ms_envelope_correlation": 0.9920568413530875, "300–1000 Hz_10ms_modulation_mae_db": 0.3324666422650527, "300–1000 Hz_40ms_log_rms_mae_db": 0.9648374459467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05495037857922602, "300–1000 Hz_40ms_envelope_correlation": 0.9976997767597217, "300–1000 Hz_40ms_modulation_mae_db": 0.47399412182725326, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7626404141399081, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15935577807761364, "1000–3000 Hz_10ms_envelope_correlation": 0.9814785885595887, "1000–3000 Hz_10ms_modulation_mae_db": 0.4828442108183888, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1157562126640403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09336486750434496, "1000–3000 Hz_40ms_envelope_correlation": 0.9941453733608582, "1000–3000 Hz_40ms_modulation_mae_db": 0.914316302291672, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8961505536785797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20326898935394302, "3000–6000 Hz_10ms_envelope_correlation": 0.9745931355428583, "3000–6000 Hz_10ms_modulation_mae_db": 0.3179416629654273, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4572874751855363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15586125268545403, "3000–6000 Hz_40ms_envelope_correlation": 0.9844127407696404, "3000–6000 Hz_40ms_modulation_mae_db": 0.32344577129957397, "mr_stft_db": 5.618107584206999, "band_mae_db": 0.21897744891224458, "rms_gain_error_db": -0.06142487092531472, "output_peak": 0.4011202730027089}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.130255687059357, "全帯域_10ms_rms_envelope_nrmse": 0.08936673254501798, "全帯域_10ms_envelope_correlation": 0.9918226622641111, "全帯域_10ms_modulation_mae_db": 0.22615726291172575, "全帯域_40ms_log_rms_mae_db": 0.6563269196465025, "全帯域_40ms_rms_envelope_nrmse": 0.0540713925323107, "全帯域_40ms_envelope_correlation": 0.9968416117251494, "全帯域_40ms_modulation_mae_db": 0.48477313960490404, "300–1000 Hz_10ms_log_rms_mae_db": 1.5438760979100925, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11599611054309787, "300–1000 Hz_10ms_envelope_correlation": 0.9899185319369044, "300–1000 Hz_10ms_modulation_mae_db": 0.39320323623209613, "300–1000 Hz_40ms_log_rms_mae_db": 0.9293422407225344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06508265955990143, "300–1000 Hz_40ms_envelope_correlation": 0.9968151056767763, "300–1000 Hz_40ms_modulation_mae_db": 0.6570663569000204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7986117687662637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15936627565558936, "1000–3000 Hz_10ms_envelope_correlation": 0.9816630818850898, "1000–3000 Hz_10ms_modulation_mae_db": 0.6445119555550893, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1212199668756935, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09041642278224055, "1000–3000 Hz_40ms_envelope_correlation": 0.9948762879302555, "1000–3000 Hz_40ms_modulation_mae_db": 1.173384042697942, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8830399377976335, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2156759903312286, "3000–6000 Hz_10ms_envelope_correlation": 0.9710248000055873, "3000–6000 Hz_10ms_modulation_mae_db": 0.6769452431922902, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3833868073243356, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17668406180700674, "3000–6000 Hz_40ms_envelope_correlation": 0.9799956759212461, "3000–6000 Hz_40ms_modulation_mae_db": 0.5665299630043217, "mr_stft_db": 5.6047162994363795, "band_mae_db": 0.2074634157181211, "rms_gain_error_db": -0.09568648869507235, "output_peak": 0.4101465223657267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0496343735117488, "全帯域_10ms_rms_envelope_nrmse": 0.07548704568435369, "全帯域_10ms_envelope_correlation": 0.9941769087218976, "全帯域_10ms_modulation_mae_db": 0.08483851648573426, "全帯域_40ms_log_rms_mae_db": 0.552434772640294, "全帯域_40ms_rms_envelope_nrmse": 0.04003372161341298, "全帯域_40ms_envelope_correlation": 0.9982347135034765, "全帯域_40ms_modulation_mae_db": 0.1557708541121825, "300–1000 Hz_10ms_log_rms_mae_db": 1.602239850309591, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09859580646387622, "300–1000 Hz_10ms_envelope_correlation": 0.9926516817457838, "300–1000 Hz_10ms_modulation_mae_db": 0.20632558895485734, "300–1000 Hz_40ms_log_rms_mae_db": 0.9047776163787883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05344516486280185, "300–1000 Hz_40ms_envelope_correlation": 0.9977201201874081, "300–1000 Hz_40ms_modulation_mae_db": 0.3793907391453983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8184117723261164, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16069292275970112, "1000–3000 Hz_10ms_envelope_correlation": 0.9809982218064911, "1000–3000 Hz_10ms_modulation_mae_db": 0.5627362568822449, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1496836939485808, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10152089176795734, "1000–3000 Hz_40ms_envelope_correlation": 0.9926817722056441, "1000–3000 Hz_40ms_modulation_mae_db": 1.0786418584845103, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9226242575865722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19952686685186416, "3000–6000 Hz_10ms_envelope_correlation": 0.975597947742014, "3000–6000 Hz_10ms_modulation_mae_db": 0.4270950452681935, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4646540224718687, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15265217623701618, "3000–6000 Hz_40ms_envelope_correlation": 0.9851018245193842, "3000–6000 Hz_40ms_modulation_mae_db": 0.35989352948114556, "mr_stft_db": 5.625181974559108, "band_mae_db": 0.161781548192543, "rms_gain_error_db": -0.01556672073117777, "output_peak": 0.40314408491043163}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0769264691880336, "全帯域_10ms_rms_envelope_nrmse": 0.07650593286391379, "全帯域_10ms_envelope_correlation": 0.9939834835255825, "全帯域_10ms_modulation_mae_db": 0.6229491858244925, "全帯域_40ms_log_rms_mae_db": 0.575581681311563, "全帯域_40ms_rms_envelope_nrmse": 0.03875234423561339, "全帯域_40ms_envelope_correlation": 0.9983314584708833, "全帯域_40ms_modulation_mae_db": 0.5822912047804074, "300–1000 Hz_10ms_log_rms_mae_db": 1.5980472719550427, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10579937867529735, "300–1000 Hz_10ms_envelope_correlation": 0.9915345904474726, "300–1000 Hz_10ms_modulation_mae_db": 0.9668365512679755, "300–1000 Hz_40ms_log_rms_mae_db": 0.8958096585690857, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055862298977412475, "300–1000 Hz_40ms_envelope_correlation": 0.9975234413817031, "300–1000 Hz_40ms_modulation_mae_db": 1.1167530016353704, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8292823061712926, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15903283690641085, "1000–3000 Hz_10ms_envelope_correlation": 0.9818822351333504, "1000–3000 Hz_10ms_modulation_mae_db": 0.6330559833367756, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1181653210255618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08933567036239934, "1000–3000 Hz_40ms_envelope_correlation": 0.9952344340831578, "1000–3000 Hz_40ms_modulation_mae_db": 0.33586084068634064, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8170190801893231, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20459529378731647, "3000–6000 Hz_10ms_envelope_correlation": 0.9751858562903394, "3000–6000 Hz_10ms_modulation_mae_db": 0.33011857593858834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3548278406155012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15179964308451852, "3000–6000 Hz_40ms_envelope_correlation": 0.9859246243677371, "3000–6000 Hz_40ms_modulation_mae_db": 0.17808407162928916, "mr_stft_db": 5.6178243757977535, "band_mae_db": 0.16589909751626197, "rms_gain_error_db": -0.02004979435198894, "output_peak": 0.4652550588910969}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1261547907411198, "全帯域_10ms_rms_envelope_nrmse": 0.0847365103927071, "全帯域_10ms_envelope_correlation": 0.9927158918750607, "全帯域_10ms_modulation_mae_db": 0.347797641947959, "全帯域_40ms_log_rms_mae_db": 0.6425183833861208, "全帯域_40ms_rms_envelope_nrmse": 0.04590669075634959, "全帯域_40ms_envelope_correlation": 0.9978519590395687, "全帯域_40ms_modulation_mae_db": 0.25326217838825754, "300–1000 Hz_10ms_log_rms_mae_db": 1.5225450134820186, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10887489568846306, "300–1000 Hz_10ms_envelope_correlation": 0.9912653577906114, "300–1000 Hz_10ms_modulation_mae_db": 0.5668559851242151, "300–1000 Hz_40ms_log_rms_mae_db": 0.8737696394898792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05803157120232866, "300–1000 Hz_40ms_envelope_correlation": 0.9976748173162147, "300–1000 Hz_40ms_modulation_mae_db": 0.5126330264452534, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7435708921139967, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1618052471019563, "1000–3000 Hz_10ms_envelope_correlation": 0.9815454591186418, "1000–3000 Hz_10ms_modulation_mae_db": 0.19817188527733964, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1238633266992726, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09429633413834124, "1000–3000 Hz_40ms_envelope_correlation": 0.9950018060079866, "1000–3000 Hz_40ms_modulation_mae_db": 0.759466320249652, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8516584362796782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2133605525972933, "3000–6000 Hz_10ms_envelope_correlation": 0.9716286057079208, "3000–6000 Hz_10ms_modulation_mae_db": 0.8244853936567785, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3596005021460593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16222973565792578, "3000–6000 Hz_40ms_envelope_correlation": 0.9831277631338918, "3000–6000 Hz_40ms_modulation_mae_db": 0.8039414833304058, "mr_stft_db": 5.612847659318366, "band_mae_db": 0.20839143582634553, "rms_gain_error_db": -0.11790460667586461, "output_peak": 0.3878384584796404}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.805801035883426, "spectral_convergence": 0.25028886745575524, "spectral_convergence_status": "ok", "amplitude_floor": 1.597965137573199e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.541654307418871, "spectral_convergence": 0.23138775541117396, "spectral_convergence_status": "ok", "amplitude_floor": 1.057274363712373e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.509124481092194, "spectral_convergence": 0.23806382496645592, "spectral_convergence_status": "ok", "amplitude_floor": 9.008040625007043e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007434014661781762, "output_power": 0.0007508541878776713, "delta_db": 0.04332194078501766, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002448126060340392, "output_power": 0.0023289767889181877, "delta_db": -0.2166861671066636, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00039049712732524606, "output_power": 0.00039067271922060874, "delta_db": 0.0019524202795207543, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012090645510899839, "output_power": 0.00011193006665767193, "delta_db": -0.3350272575512967, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.957546973001399e-05, "output_power": 6.0921740194026584e-05, "delta_db": 0.09704824930628549, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.142800821939945e-05, "output_power": 8.997817378606065e-05, "delta_db": 0.43363362591551735, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18794494349071691, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12679180684046346, "pitch": null, "output_rms_dbfs": -24.27903192111348, "output_peak": 0.42834844743997025, "reference_rms_dbfs": -24.152240114273017, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.61885994146483}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019565573818943504, "log_rms_mae_db": 1.1189297217115433, "rms_envelope_nrmse": 0.0770102803198733, "envelope_correlation": 0.9940359498646059, "rms_gain_error_db": -0.12676549461243203, "reference_mean_power": 0.0038470996951946897, "output_mean_power": 0.0037364303533181666}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.743468430119212e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007829242029289011, "output_power": 0.0007331136194402277, "reference_active": true, "delta_db": -0.2854843113578343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011656588263872551, "output_power": 0.00010953218296903094, "reference_active": true, "delta_db": -0.27029713297634883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.15404555717803e-05, "output_power": 2.8655229504740176e-05, "reference_active": true, "delta_db": -0.41664070859423175}], "active_band_count": 3, "active_band_mae_db": 0.32414071764280494}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018472557440126405, "log_rms_mae_db": 0.6500534387283545, "rms_envelope_nrmse": 0.04449323640064907, "envelope_correlation": 0.998004988840657, "rms_gain_error_db": -0.1265705350609783, "reference_mean_power": 0.0038659686805529734, "output_mean_power": 0.0037549250943515458}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.614770154461247e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007106100759473788, "output_power": 0.0006656439168585626, "reference_active": true, "delta_db": -0.28389393725440054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.7305685337334634e-05, "output_power": 5.156729823040422e-05, "reference_active": true, "delta_db": -0.45823332788650417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9649709160477995e-06, "output_power": 2.1959400761825484e-06, "reference_active": true, "delta_db": 0.48264358039161315}], "active_band_count": 3, "active_band_mae_db": 0.4082569485108393}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016965929826439602, "log_rms_mae_db": 1.560964584558203, "rms_envelope_nrmse": 0.09549336668796492, "envelope_correlation": 0.993391613956891, "rms_gain_error_db": -0.20571269299637507, "reference_mean_power": 0.0021633278035333714, "output_mean_power": 0.0020632462415310034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3458386409803657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007177587670080827, "output_power": 0.000653605789996318, "reference_active": true, "delta_db": -0.4066261509815703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.048912255781678e-05, "output_power": 7.694336229969575e-05, "reference_active": true, "delta_db": -0.7042521668056592}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.109541579796594e-05, "output_power": 2.1776062426214422e-05, "reference_active": true, "delta_db": -1.5471701556691908}], "active_band_count": 3, "active_band_mae_db": 0.8860161578188067}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015157137475553504, "log_rms_mae_db": 0.9505359255999434, "rms_envelope_nrmse": 0.060135586794007906, "envelope_correlation": 0.9974736183406736, "rms_gain_error_db": -0.20572170403578138, "reference_mean_power": 0.002174161455071234, "output_mean_power": 0.002073574395813684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2731603714689836e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006867133707128568, "output_power": 0.0006248432287807199, "reference_active": true, "delta_db": -0.4100443556662803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.067232594302272e-05, "output_power": 3.250000267795438e-05, "reference_active": true, "delta_db": -0.9741561250634774}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.08557261679534e-06, "output_power": 1.561864939209359e-06, "reference_active": true, "delta_db": -1.2558184002575048}], "active_band_count": 3, "active_band_mae_db": 0.8800062936624208}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.465903254113391e-05, "log_rms_mae_db": 1.7301565808410728, "rms_envelope_nrmse": 0.16005849474008357, "envelope_correlation": 0.9808392079898464, "rms_gain_error_db": -0.2220079897927492, "reference_mean_power": 0.00014077271984820836, "output_mean_power": 0.00013375736438616653}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.622290255773331e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.707421130487826e-05, "output_power": 3.591262937543793e-05, "reference_active": true, "delta_db": -0.13824717320491023}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.08772682349557e-06, "output_power": 4.993227040960871e-06, "reference_active": true, "delta_db": -0.08142470810459039}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8011808387082025e-06, "output_power": 2.583597757800432e-06, "reference_active": true, "delta_db": 1.5666758075353815}], "active_band_count": 3, "active_band_mae_db": 0.5954492296149607}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.107059424790359e-05, "log_rms_mae_db": 1.0820971766152292, "rms_envelope_nrmse": 0.0888500934972116, "envelope_correlation": 0.9941177955920399, "rms_gain_error_db": -0.2219838380790182, "reference_mean_power": 0.00014139647469838974, "output_mean_power": 0.0001343507817894755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.069193023932203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.382591725297068e-05, "output_power": 3.316597137696926e-05, "reference_active": true, "delta_db": -0.08556860541373416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1244897853414024e-06, "output_power": 2.1108636033356853e-06, "reference_active": true, "delta_db": -0.027944757044906617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.400225448356044e-07, "output_power": 1.1649752364535568e-07, "reference_active": true, "delta_db": -0.798812726737067}], "active_band_count": 3, "active_band_mae_db": 0.30410869639856924}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.428054041295646e-05, "log_rms_mae_db": 1.858726911288704, "rms_envelope_nrmse": 0.2167306703506722, "envelope_correlation": 0.9727070066269281, "rms_gain_error_db": 0.2910910579782769, "reference_mean_power": 8.952225188868405e-05, "output_mean_power": 9.572824652540319e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.314971416941356e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.877588546016404e-05, "output_power": 6.172568652912498e-05, "reference_active": true, "delta_db": 0.2126674846357029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.04956306951581e-06, "output_power": 9.048920644121982e-06, "reference_active": true, "delta_db": -0.0003083151142449465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5294615410314917e-06, "output_power": 3.326321399961529e-06, "reference_active": true, "delta_db": 1.189361292485293}], "active_band_count": 3, "active_band_mae_db": 0.46744569741174696}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5452516433588795e-05, "log_rms_mae_db": 1.3732428166575894, "rms_envelope_nrmse": 0.1528895441366929, "envelope_correlation": 0.9861675779897836, "rms_gain_error_db": 0.29110779210431226, "reference_mean_power": 8.996899751083782e-05, "output_mean_power": 9.620633281085953e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.951283845645369e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.884212028117595e-05, "output_power": 6.225397896863524e-05, "reference_active": true, "delta_db": 0.24478801184874083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.436071463262977e-06, "output_power": 4.4962978690905904e-06, "reference_active": true, "delta_db": 0.05856540658823427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7711408251558852e-07, "output_power": 1.91848726392019e-07, "reference_active": true, "delta_db": 0.34705826688191743}], "active_band_count": 3, "active_band_mae_db": 0.21680389510629752}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8951964813746158, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "31ead5853df8e32e71ba38e4605d1529faaf552a9d5ac31520132bad7798ef51", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9822001378879767, "standard_deviation": 0.0015434634966776475, "minimum": 0.9808392079898464, "maximum": 0.9851211390655372, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7782645726907265, "standard_deviation": 0.0356985877471173, "minimum": 1.7301565808410728, "maximum": 1.8292823061712926, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.471711511402249, "standard_deviation": 0.1685749966314682, "minimum": 0.19817188527733964, "maximum": 0.6445119555550893, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15655235601438697, "standard_deviation": 0.006709205947131704, "minimum": 0.14325292916092167, "maximum": 0.1618052471019563, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9947936548998655, "standard_deviation": 0.0011499781170662092, "minimum": 0.9926817722056441, "maximum": 0.996242193076321, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.109375674457525, "standard_deviation": 0.026284663468495048, "minimum": 1.066104648419687, "maximum": 1.1496836939485808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7044819153562312, "standard_deviation": 0.34889565744721895, "minimum": 0.30410869639856924, "maximum": 1.173384042697942, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08950924789389425, "standard_deviation": 0.007547110743497591, "minimum": 0.07863417795390493, "maximum": 0.10152089176795734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9734546126176107, "standard_deviation": 0.001684218704456874, "minimum": 0.9710248000055873, "maximum": 0.975597947742014, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.868674949278288, "standard_deviation": 0.032064037309084664, "minimum": 1.8170190801893231, "maximum": 1.9226242575865722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.455653084159096, "standard_deviation": 0.20282287497530307, "minimum": 0.20264426623068074, "maximum": 0.8244853936567785, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.208663171229128, "standard_deviation": 0.006408685549702034, "minimum": 0.19952686685186416, "maximum": 0.2167306703506722, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9843597948174378, "standard_deviation": 0.001995349137958525, "minimum": 0.9799956759212461, "maximum": 0.9861675779897836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3972498201570014, "standard_deviation": 0.041746238677129816, "minimum": 1.3548278406155012, "maximum": 1.4646540224718687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.35954560608122554, "standard_deviation": 0.2242994950406822, "minimum": 0.12789549388311294, "maximum": 0.8039414833304058, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15721115421154916, "standard_deviation": 0.008578577653448993, "minimum": 0.15179964308451852, "maximum": 0.17668406180700674, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918181895301075, "standard_deviation": 0.0010787493545450343, "minimum": 0.9899185319369044, "maximum": 0.993391613956891, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.579516931078362, "standard_deviation": 0.03782872797905479, "minimum": 1.5225450134820186, "maximum": 1.6240715299449116, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5851069521508921, "standard_deviation": 0.27226195271847614, "minimum": 0.20632558895485734, "maximum": 0.9668365512679755, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1045586435104986, "standard_deviation": 0.006765977697389787, "minimum": 0.09549336668796492, "maximum": 0.11599611054309787, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974791321307861, "standard_deviation": 0.0003998129522103946, "minimum": 0.9968151056767763, "maximum": 0.9979780941106496, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9196798634259395, "standard_deviation": 0.02954940342523546, "minimum": 0.8737696394898792, "maximum": 0.9648374459467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6626288025613829, "standard_deviation": 0.24824318080800972, "minimum": 0.3793907391453983, "maximum": 1.1167530016353704, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05795924107010224, "standard_deviation": 0.005203249028515718, "minimum": 0.05100592449156464, "maximum": 0.06516034409357493, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18738481638869114, "standard_deviation": 0.023932022128025143, "minimum": 0.15550464249954674, "maximum": 0.21897744891224458, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.618828346889126, "standard_deviation": 0.007597798218634891, "minimum": 5.6047162994363795, "maximum": 5.628568463850225, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41019125205617757, "standard_deviation": 0.025689875268398766, "minimum": 0.3878384584796404, "maximum": 0.4652550588910969, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07125454008190579, "standard_deviation": 0.04182305789626645, "minimum": -0.12679180684046346, "maximum": -0.01556672073117777, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933542913190995, "standard_deviation": 0.0010078926928695615, "minimum": 0.9918226622641111, "maximum": 0.9946215946867547, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.090441274297075, "standard_deviation": 0.036527468507630736, "minimum": 1.034123327481177, "maximum": 1.130255687059357, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2871543586913088, "standard_deviation": 0.16949346337315063, "minimum": 0.08483851648573426, "maximum": 0.6229491858244925, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08046470060878272, "standard_deviation": 0.006048468552584856, "minimum": 0.0724412332331893, "maximum": 0.08936673254501798, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979369010751737, "standard_deviation": 0.0006707200620960339, "minimum": 0.9968416117251494, "maximum": 0.9986187703278986, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6040284753294862, "standard_deviation": 0.04965056947501852, "minimum": 0.5450965817763845, "maximum": 0.6563269196465025, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.34167410108740925, "standard_deviation": 0.1599790785093964, "minimum": 0.1557708541121825, "maximum": 0.5822912047804074, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04342090677731099, "standard_deviation": 0.0069971993404134855, "minimum": 0.03579364428747994, "maximum": 0.0540713925323107, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274384-0010", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.55198073387146, "predicted_mos": 2.55198073387146, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "8296a31093ec84fbe0c3cd314d984385cf94404266eef946c1a80c9527633a9a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7762361001223326, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0585487679943921, "全帯域_10ms_rms_envelope_nrmse": 0.09040475766058834, "全帯域_10ms_envelope_correlation": 0.991648559527631, "全帯域_10ms_modulation_mae_db": 0.22567597446226748, "全帯域_40ms_log_rms_mae_db": 0.5874791657382639, "全帯域_40ms_rms_envelope_nrmse": 0.04736143131143207, "全帯域_40ms_envelope_correlation": 0.9976541330440042, "全帯域_40ms_modulation_mae_db": 0.3426407276176086, "300–1000 Hz_10ms_log_rms_mae_db": 1.7388904195362909, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12069685019693294, "300–1000 Hz_10ms_envelope_correlation": 0.9892177297739279, "300–1000 Hz_10ms_modulation_mae_db": 0.6155328194870019, "300–1000 Hz_40ms_log_rms_mae_db": 1.0224622157929695, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06208303946318618, "300–1000 Hz_40ms_envelope_correlation": 0.9973584074983767, "300–1000 Hz_40ms_modulation_mae_db": 0.9385434056643281, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1668598409760587, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20507387468437974, "1000–3000 Hz_10ms_envelope_correlation": 0.9691585581720359, "1000–3000 Hz_10ms_modulation_mae_db": 0.42042747675328124, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2767962761737077, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11108673713844938, "1000–3000 Hz_40ms_envelope_correlation": 0.9905861583105633, "1000–3000 Hz_40ms_modulation_mae_db": 0.2980394046875516, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4222518438739735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23488805540163263, "3000–6000 Hz_10ms_envelope_correlation": 0.9667776757673495, "3000–6000 Hz_10ms_modulation_mae_db": 0.49727234351255767, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7201129788708098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16965133212744163, "3000–6000 Hz_40ms_envelope_correlation": 0.9822257888315218, "3000–6000 Hz_40ms_modulation_mae_db": 0.5188091472381585, "mr_stft_db": 6.040489638295893, "band_mae_db": 0.1667937108169321, "rms_gain_error_db": -0.10664188757540086, "output_peak": 0.3682466977427322}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0199703717891593, "全帯域_10ms_rms_envelope_nrmse": 0.07969875919538134, "全帯域_10ms_envelope_correlation": 0.9935295188793855, "全帯域_10ms_modulation_mae_db": 0.1459930565101528, "全帯域_40ms_log_rms_mae_db": 0.5527977081341284, "全帯域_40ms_rms_envelope_nrmse": 0.046610858531098844, "全帯域_40ms_envelope_correlation": 0.9976901669153235, "全帯域_40ms_modulation_mae_db": 0.16437320331684344, "300–1000 Hz_10ms_log_rms_mae_db": 1.7676996537905456, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10602090457288237, "300–1000 Hz_10ms_envelope_correlation": 0.991662502125668, "300–1000 Hz_10ms_modulation_mae_db": 0.3953683182634213, "300–1000 Hz_40ms_log_rms_mae_db": 1.0385896115658169, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06621171545029303, "300–1000 Hz_40ms_envelope_correlation": 0.9967223766132599, "300–1000 Hz_40ms_modulation_mae_db": 0.4826092673358093, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0893062824346864, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17011388066498376, "1000–3000 Hz_10ms_envelope_correlation": 0.9781751034949328, "1000–3000 Hz_10ms_modulation_mae_db": 0.916516574027197, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2596236230228754, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09578466223176547, "1000–3000 Hz_40ms_envelope_correlation": 0.9928351116519228, "1000–3000 Hz_40ms_modulation_mae_db": 1.3284069656857185, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4973217002422623, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24226827882835322, "3000–6000 Hz_10ms_envelope_correlation": 0.9638330941955837, "3000–6000 Hz_10ms_modulation_mae_db": 1.1261770426137037, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7614386717953479, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16907987553833187, "3000–6000 Hz_40ms_envelope_correlation": 0.981837459253872, "3000–6000 Hz_40ms_modulation_mae_db": 1.0674489293437952, "mr_stft_db": 6.029361076651057, "band_mae_db": 0.17957845167169537, "rms_gain_error_db": -0.09690871617256569, "output_peak": 0.41596484001442047}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0223868659247393, "全帯域_10ms_rms_envelope_nrmse": 0.07580296302520675, "全帯域_10ms_envelope_correlation": 0.9942933345135088, "全帯域_10ms_modulation_mae_db": 0.6310178332796914, "全帯域_40ms_log_rms_mae_db": 0.5552088641540899, "全帯域_40ms_rms_envelope_nrmse": 0.03925084050017952, "全帯域_40ms_envelope_correlation": 0.9985883394718627, "全帯域_40ms_modulation_mae_db": 0.6351194498753079, "300–1000 Hz_10ms_log_rms_mae_db": 1.839762254201678, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10799543793396488, "300–1000 Hz_10ms_envelope_correlation": 0.9920236116629152, "300–1000 Hz_10ms_modulation_mae_db": 0.9636903953500008, "300–1000 Hz_40ms_log_rms_mae_db": 1.0771355301825023, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06645567061916567, "300–1000 Hz_40ms_envelope_correlation": 0.9975520290641993, "300–1000 Hz_40ms_modulation_mae_db": 1.0989374284579123, "1000–3000 Hz_10ms_log_rms_mae_db": 2.129693932067484, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20151025451777455, "1000–3000 Hz_10ms_envelope_correlation": 0.9694295341155363, "1000–3000 Hz_10ms_modulation_mae_db": 0.9193314448393006, "1000–3000 Hz_40ms_log_rms_mae_db": 1.266563159511423, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11573520904999396, "1000–3000 Hz_40ms_envelope_correlation": 0.9893795164820408, "1000–3000 Hz_40ms_modulation_mae_db": 0.07452075498750238, "3000–6000 Hz_10ms_log_rms_mae_db": 2.46600595174082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22887184070077146, "3000–6000 Hz_10ms_envelope_correlation": 0.9690815231888529, "3000–6000 Hz_10ms_modulation_mae_db": 0.6543274234943702, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7801514265126184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1716629106584428, "3000–6000 Hz_40ms_envelope_correlation": 0.9821096980277595, "3000–6000 Hz_40ms_modulation_mae_db": 0.8049632439589943, "mr_stft_db": 6.040477106189172, "band_mae_db": 0.20866709223000388, "rms_gain_error_db": -0.1448369138988244, "output_peak": 0.40618263733331694}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0974226527554158, "全帯域_10ms_rms_envelope_nrmse": 0.08758419994176679, "全帯域_10ms_envelope_correlation": 0.992111543285102, "全帯域_10ms_modulation_mae_db": 0.34926611141642977, "全帯域_40ms_log_rms_mae_db": 0.589073680066679, "全帯域_40ms_rms_envelope_nrmse": 0.047368960385514976, "全帯域_40ms_envelope_correlation": 0.9975111503794215, "全帯域_40ms_modulation_mae_db": 0.23972248650792757, "300–1000 Hz_10ms_log_rms_mae_db": 1.764148411429743, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10921067502413076, "300–1000 Hz_10ms_envelope_correlation": 0.9909883354595733, "300–1000 Hz_10ms_modulation_mae_db": 0.9137381264724261, "300–1000 Hz_40ms_log_rms_mae_db": 1.0462347389259503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06297599871735915, "300–1000 Hz_40ms_envelope_correlation": 0.996868422488568, "300–1000 Hz_40ms_modulation_mae_db": 0.8472103454898118, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0483218769800478, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17126731588700125, "1000–3000 Hz_10ms_envelope_correlation": 0.9785420598606781, "1000–3000 Hz_10ms_modulation_mae_db": 0.6239731508775316, "1000–3000 Hz_40ms_log_rms_mae_db": 1.177797200429261, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09737996633855366, "1000–3000 Hz_40ms_envelope_correlation": 0.993857413579968, "1000–3000 Hz_40ms_modulation_mae_db": 0.39910768357792675, "3000–6000 Hz_10ms_log_rms_mae_db": 2.468742484503069, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23623740931154041, "3000–6000 Hz_10ms_envelope_correlation": 0.9654109981781901, "3000–6000 Hz_10ms_modulation_mae_db": 0.6001974014377655, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7669436189600913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17269803992913513, "3000–6000 Hz_40ms_envelope_correlation": 0.9810954351400936, "3000–6000 Hz_40ms_modulation_mae_db": 0.5892146953356142, "mr_stft_db": 6.029235879935039, "band_mae_db": 0.15426868193779517, "rms_gain_error_db": -0.02926589957380253, "output_peak": 0.44567882253146307}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0550898219249296, "全帯域_10ms_rms_envelope_nrmse": 0.07618606422808821, "全帯域_10ms_envelope_correlation": 0.9941214874227535, "全帯域_10ms_modulation_mae_db": 0.16051327076416214, "全帯域_40ms_log_rms_mae_db": 0.5748690673519117, "全帯域_40ms_rms_envelope_nrmse": 0.042419172468521865, "全帯域_40ms_envelope_correlation": 0.9981410832269574, "全帯域_40ms_modulation_mae_db": 0.1471297411607688, "300–1000 Hz_10ms_log_rms_mae_db": 1.7978080526857618, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09817160011791493, "300–1000 Hz_10ms_envelope_correlation": 0.9928378830710332, "300–1000 Hz_10ms_modulation_mae_db": 0.24044529349024577, "300–1000 Hz_40ms_log_rms_mae_db": 1.0096825324890577, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054059250599404175, "300–1000 Hz_40ms_envelope_correlation": 0.9978479147883091, "300–1000 Hz_40ms_modulation_mae_db": 0.3590692732523195, "1000–3000 Hz_10ms_log_rms_mae_db": 2.136400326997843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19078854903928316, "1000–3000 Hz_10ms_envelope_correlation": 0.9724736107441008, "1000–3000 Hz_10ms_modulation_mae_db": 0.651177045348262, "1000–3000 Hz_40ms_log_rms_mae_db": 1.293975239194464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10528262114464791, "1000–3000 Hz_40ms_envelope_correlation": 0.9915136748431415, "1000–3000 Hz_40ms_modulation_mae_db": 0.5244199209482064, "3000–6000 Hz_10ms_log_rms_mae_db": 2.439396791644994, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2584219149542586, "3000–6000 Hz_10ms_envelope_correlation": 0.9585045243287309, "3000–6000 Hz_10ms_modulation_mae_db": 0.8117319330722795, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7770424061813843, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17744173364693908, "3000–6000 Hz_40ms_envelope_correlation": 0.9800755492013187, "3000–6000 Hz_40ms_modulation_mae_db": 0.9036272194343047, "mr_stft_db": 6.041243733086021, "band_mae_db": 0.1103293389919157, "rms_gain_error_db": -0.10798273333827701, "output_peak": 0.39355839921315}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.029243352379358, "全帯域_10ms_rms_envelope_nrmse": 0.08019835793768894, "全帯域_10ms_envelope_correlation": 0.993633839238861, "全帯域_10ms_modulation_mae_db": 0.663576386110401, "全帯域_40ms_log_rms_mae_db": 0.5515854218848227, "全帯域_40ms_rms_envelope_nrmse": 0.048317189666770746, "全帯域_40ms_envelope_correlation": 0.9977528159640889, "全帯域_40ms_modulation_mae_db": 0.8319557571449149, "300–1000 Hz_10ms_log_rms_mae_db": 1.7658280107287527, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11995250376539035, "300–1000 Hz_10ms_envelope_correlation": 0.9897238007306002, "300–1000 Hz_10ms_modulation_mae_db": 0.9922469563324237, "300–1000 Hz_40ms_log_rms_mae_db": 1.0555488369442605, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07508339001247173, "300–1000 Hz_40ms_envelope_correlation": 0.9963391858993775, "300–1000 Hz_40ms_modulation_mae_db": 1.1750439933729073, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1927559819270037, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18664650691043255, "1000–3000 Hz_10ms_envelope_correlation": 0.9736849484065598, "1000–3000 Hz_10ms_modulation_mae_db": 0.6271516824900779, "1000–3000 Hz_40ms_log_rms_mae_db": 1.302223840546397, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09369730355099451, "1000–3000 Hz_40ms_envelope_correlation": 0.9931227414272321, "1000–3000 Hz_40ms_modulation_mae_db": 0.7200533025475369, "3000–6000 Hz_10ms_log_rms_mae_db": 2.455113063424153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2203163028136995, "3000–6000 Hz_10ms_envelope_correlation": 0.969989464016951, "3000–6000 Hz_10ms_modulation_mae_db": 1.4013853921112542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.805349472880327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16906354292596265, "3000–6000 Hz_40ms_envelope_correlation": 0.981913574640297, "3000–6000 Hz_40ms_modulation_mae_db": 1.3025011331393845, "mr_stft_db": 6.033696267177636, "band_mae_db": 0.18641154644321964, "rms_gain_error_db": -0.16421885940123815, "output_peak": 0.4190776586507645}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0907541893433026, "全帯域_10ms_rms_envelope_nrmse": 0.091940933319649, "全帯域_10ms_envelope_correlation": 0.991312749695923, "全帯域_10ms_modulation_mae_db": 0.7754914739296357, "全帯域_40ms_log_rms_mae_db": 0.6322175461516795, "全帯域_40ms_rms_envelope_nrmse": 0.05762333652968171, "全帯域_40ms_envelope_correlation": 0.9963272872891853, "全帯域_40ms_modulation_mae_db": 0.9095542941820028, "300–1000 Hz_10ms_log_rms_mae_db": 1.8431065489774845, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12708555099170898, "300–1000 Hz_10ms_envelope_correlation": 0.9877660683519358, "300–1000 Hz_10ms_modulation_mae_db": 1.3020953806424018, "300–1000 Hz_40ms_log_rms_mae_db": 1.149990508705685, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08316204891384854, "300–1000 Hz_40ms_envelope_correlation": 0.9944761424924188, "300–1000 Hz_40ms_modulation_mae_db": 1.6697777917633603, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1459879305307332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1795842683671167, "1000–3000 Hz_10ms_envelope_correlation": 0.9759048760494303, "1000–3000 Hz_10ms_modulation_mae_db": 1.0456449650116895, "1000–3000 Hz_40ms_log_rms_mae_db": 1.261371744828134, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09755502227127372, "1000–3000 Hz_40ms_envelope_correlation": 0.9932564657806232, "1000–3000 Hz_40ms_modulation_mae_db": 0.4116718610254102, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4578017030803587, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23283536117870945, "3000–6000 Hz_10ms_envelope_correlation": 0.9664697932377675, "3000–6000 Hz_10ms_modulation_mae_db": 1.312337330865358, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7736098135580276, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17650089214355275, "3000–6000 Hz_40ms_envelope_correlation": 0.9805052166138288, "3000–6000 Hz_40ms_modulation_mae_db": 1.2124774235013198, "mr_stft_db": 6.033151319152608, "band_mae_db": 0.18816990616257723, "rms_gain_error_db": -0.04865566665800283, "output_peak": 0.49623571658766896}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0346167642236819, "全帯域_10ms_rms_envelope_nrmse": 0.08232498457483357, "全帯域_10ms_envelope_correlation": 0.9930741402132502, "全帯域_10ms_modulation_mae_db": 0.6486451889152235, "全帯域_40ms_log_rms_mae_db": 0.5791171270853054, "全帯域_40ms_rms_envelope_nrmse": 0.04373472526251436, "全帯域_40ms_envelope_correlation": 0.9979702374851372, "全帯域_40ms_modulation_mae_db": 0.7343530167513276, "300–1000 Hz_10ms_log_rms_mae_db": 1.7728318006009665, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10345462784057684, "300–1000 Hz_10ms_envelope_correlation": 0.9923810648503575, "300–1000 Hz_10ms_modulation_mae_db": 0.97062405560375, "300–1000 Hz_40ms_log_rms_mae_db": 1.082077375205573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0641419181604598, "300–1000 Hz_40ms_envelope_correlation": 0.9973165167453637, "300–1000 Hz_40ms_modulation_mae_db": 1.132629478282764, "1000–3000 Hz_10ms_log_rms_mae_db": 2.128871507351276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17663827621601352, "1000–3000 Hz_10ms_envelope_correlation": 0.9769094763157753, "1000–3000 Hz_10ms_modulation_mae_db": 0.6487107314468799, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2732197355290904, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08783636978924107, "1000–3000 Hz_40ms_envelope_correlation": 0.9940467622677055, "1000–3000 Hz_40ms_modulation_mae_db": 0.25819338836737366, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4977028811942774, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23260805401499363, "3000–6000 Hz_10ms_envelope_correlation": 0.9666956045451806, "3000–6000 Hz_10ms_modulation_mae_db": 0.7385666027283175, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8045120784587518, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1673545140248118, "3000–6000 Hz_40ms_envelope_correlation": 0.9822472772830668, "3000–6000 Hz_40ms_modulation_mae_db": 1.0099451903669225, "mr_stft_db": 6.046747508995213, "band_mae_db": 0.15483781303652452, "rms_gain_error_db": -0.08826404779919124, "output_peak": 0.3795131919135572}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.268407967876561, "spectral_convergence": 0.26478212297085724, "spectral_convergence_status": "ok", "amplitude_floor": 1.597965137573199e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.073533558408076, "spectral_convergence": 0.24898706855885624, "spectral_convergence_status": "ok", "amplitude_floor": 1.057274363712373e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.779527388603044, "spectral_convergence": 0.24499333047221036, "spectral_convergence_status": "ok", "amplitude_floor": 9.008040625007043e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007434014661781762, "output_power": 0.0007767027618630768, "delta_db": 0.19031436156495796, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002448126060340392, "output_power": 0.002311203126959561, "delta_db": -0.24995660576526527, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00039049712732524606, "output_power": 0.00039027990419461506, "delta_db": -0.00241653649122332, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012090645510899839, "output_power": 0.00012330958827022647, "delta_db": 0.0854735949872753, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.957546973001399e-05, "output_power": 6.209570827708648e-05, "delta_db": 0.17994109824818155, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.142800821939945e-05, "output_power": 8.710434485150514e-05, "delta_db": 0.2926600678446893, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1667937108169321, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10664188757540086, "pitch": null, "output_rms_dbfs": -24.258882001848416, "output_peak": 0.3682466977427322, "reference_rms_dbfs": -24.152240114273017, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.040489638295893}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019565573818943504, "log_rms_mae_db": 1.0585487679943921, "rms_envelope_nrmse": 0.09040475766058834, "envelope_correlation": 0.991648559527631, "rms_gain_error_db": -0.10661657100343196, "reference_mean_power": 0.0038470996951946897, "output_mean_power": 0.0037538056514920507}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.743468430119212e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007829242029289011, "output_power": 0.0007427645170748905, "reference_active": true, "delta_db": -0.22868570141877456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011656588263872551, "output_power": 0.0001075527258957995, "reference_active": true, "delta_db": -0.34950034569661775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.15404555717803e-05, "output_power": 3.226652156886052e-05, "reference_active": true, "delta_db": 0.09884187627141001}], "active_band_count": 3, "active_band_mae_db": 0.22567597446226748}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018472557440126405, "log_rms_mae_db": 0.5874791657382639, "rms_envelope_nrmse": 0.04736143131143207, "envelope_correlation": 0.9976541330440042, "rms_gain_error_db": -0.10661803450558235, "reference_mean_power": 0.0038659686805529734, "output_mean_power": 0.0037722157835777485}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.614770154461247e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007106100759473788, "output_power": 0.0006735673482143713, "reference_active": true, "delta_db": -0.23250335000452996}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.7305685337334634e-05, "output_power": 5.335756726705122e-05, "reference_active": true, "delta_db": -0.3100169020548085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9649709160477995e-06, "output_power": 1.7571787135893931e-06, "reference_active": true, "delta_db": -0.48540193079348737}], "active_band_count": 3, "active_band_mae_db": 0.3426407276176086}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016965929826439602, "log_rms_mae_db": 1.7388904195362909, "rms_envelope_nrmse": 0.12069685019693294, "envelope_correlation": 0.9892177297739279, "rms_gain_error_db": -0.22303895300016469, "reference_mean_power": 0.0021633278035333714, "output_mean_power": 0.002055031279629388}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3458386409803657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007177587670080827, "output_power": 0.0006782259244158891, "reference_active": true, "delta_db": -0.24604119683972317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.048912255781678e-05, "output_power": 7.484188520508199e-05, "reference_active": true, "delta_db": -0.8245165854406766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.109541579796594e-05, "output_power": 2.600712235069058e-05, "reference_active": true, "delta_db": -0.7760406761806059}], "active_band_count": 3, "active_band_mae_db": 0.6155328194870019}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015157137475553504, "log_rms_mae_db": 1.0224622157929695, "rms_envelope_nrmse": 0.06208303946318618, "envelope_correlation": 0.9973584074983767, "rms_gain_error_db": -0.22307058873375685, "reference_mean_power": 0.002174161455071234, "output_mean_power": 0.0020653075523237034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2731603714689836e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006867133707128568, "output_power": 0.0006420022457482458, "reference_active": true, "delta_db": -0.2923895619518491}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.067232594302272e-05, "output_power": 3.243581213152124e-05, "reference_active": true, "delta_db": -0.9827423295389344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.08557261679534e-06, "output_power": 1.462768109254346e-06, "reference_active": true, "delta_db": -1.5404983255022011}], "active_band_count": 3, "active_band_mae_db": 0.9385434056643281}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.465903254113391e-05, "log_rms_mae_db": 2.1668598409760587, "rms_envelope_nrmse": 0.20507387468437974, "envelope_correlation": 0.9691585581720359, "rms_gain_error_db": 0.13444713264692418, "reference_mean_power": 0.00014077271984820836, "output_mean_power": 0.00014519886277277065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.622290255773331e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.707421130487826e-05, "output_power": 3.709710686653939e-05, "reference_active": true, "delta_db": 0.002681202559282471}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.08772682349557e-06, "output_power": 5.045817734170357e-06, "reference_active": true, "delta_db": -0.03592225780699703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8011808387082025e-06, "output_power": 2.3868504852020756e-06, "reference_active": true, "delta_db": 1.2226789698935643}], "active_band_count": 3, "active_band_mae_db": 0.42042747675328124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.107059424790359e-05, "log_rms_mae_db": 1.2767962761737077, "rms_envelope_nrmse": 0.11108673713844938, "envelope_correlation": 0.9905861583105633, "rms_gain_error_db": 0.13509935550230184, "reference_mean_power": 0.00014139647469838974, "output_mean_power": 0.00014586413378774385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.069193023932203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.382591725297068e-05, "output_power": 3.433224975054091e-05, "reference_active": true, "delta_db": 0.06452681011093575}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1244897853414024e-06, "output_power": 2.336890345180083e-06, "reference_active": true, "delta_db": 0.4138368698059842}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.400225448356044e-07, "output_power": 1.272396638370538e-07, "reference_active": true, "delta_db": -0.4157545341457348}], "active_band_count": 3, "active_band_mae_db": 0.2980394046875516}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.428054041295646e-05, "log_rms_mae_db": 2.4222518438739735, "rms_envelope_nrmse": 0.23488805540163263, "envelope_correlation": 0.9667776757673495, "rms_gain_error_db": 0.12870766219671181, "reference_mean_power": 8.952225188868405e-05, "output_mean_power": 9.221504114002538e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.314971416941356e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.877588546016404e-05, "output_power": 5.6419224089658454e-05, "reference_active": true, "delta_db": -0.17772071315983948}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.04956306951581e-06, "output_power": 8.551790505672374e-06, "reference_active": true, "delta_db": -0.24570557767712325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5294615410314917e-06, "output_power": 1.9778302121732753e-06, "reference_active": true, "delta_db": -1.0683907397007102}], "active_band_count": 3, "active_band_mae_db": 0.49727234351255767}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5452516433588795e-05, "log_rms_mae_db": 1.7201129788708098, "rms_envelope_nrmse": 0.16965133212744163, "envelope_correlation": 0.9822257888315218, "rms_gain_error_db": 0.12870483500554536, "reference_mean_power": 8.996899751083782e-05, "output_mean_power": 9.267516434030864e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.951283845645369e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.884212028117595e-05, "output_power": 5.631283691538781e-05, "reference_active": true, "delta_db": -0.19080906759163857}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.436071463262977e-06, "output_power": 4.260952721187538e-06, "reference_active": true, "delta_db": -0.17491818467737716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7711408251558852e-07, "output_power": 1.3464278147713392e-07, "reference_active": true, "delta_db": -1.1907001894454599}], "active_band_count": 3, "active_band_mae_db": 0.5188091472381585}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8741307249510352, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3354d82b282d507599d447d441601c29943f30901c1cfd84d10d3c88045ba20e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9742847708948812, "standard_deviation": 0.0037083782652687306, "minimum": 0.9691585581720359, "maximum": 0.9785420598606781, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1297747099081414, "standard_deviation": 0.0445962759661118, "minimum": 2.0483218769800478, "maximum": 2.1927559819270037, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7316166338492776, "standard_deviation": 0.20726838500212422, "minimum": 0.42042747675328124, "maximum": 1.0456449650116895, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18520286578587314, "standard_deviation": 0.013206436382193083, "minimum": 0.17011388066498376, "maximum": 0.20507387468437974, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9923247305428997, "standard_deviation": 0.0016662482225464096, "minimum": 0.9893795164820408, "maximum": 0.9940467622677055, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2639463524044192, "standard_deviation": 0.0379319973763466, "minimum": 1.177797200429261, "maximum": 1.302223840546397, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5018016602284033, "standard_deviation": 0.38424771558639215, "minimum": 0.07452075498750238, "maximum": 1.3284069656857185, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10054473643936496, "standard_deviation": 0.00936813882635307, "minimum": 0.08783636978924107, "maximum": 0.11573520904999396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9658453346823257, "standard_deviation": 0.003540664744701405, "minimum": 0.9585045243287309, "maximum": 0.969989464016951, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.463042052462989, "standard_deviation": 0.025997298017018906, "minimum": 2.4222518438739735, "maximum": 2.4977028811942774, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8927494337294508, "standard_deviation": 0.341932124928856, "minimum": 0.49727234351255767, "maximum": 1.4013853921112542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23580590215049485, "standard_deviation": 0.011092688909603863, "minimum": 0.2203163028136995, "maximum": 0.2584219149542586, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9815012498739697, "standard_deviation": 0.00083872023886579, "minimum": 0.9800755492013187, "maximum": 0.9822472772830668, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.77364505840217, "standard_deviation": 0.02691629237957601, "minimum": 1.7201129788708098, "maximum": 1.805349472880327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9261233727898117, "standard_deviation": 0.2790944693869411, "minimum": 0.5188091472381585, "maximum": 1.3025011331393845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1716816051243272, "standard_deviation": 0.003662689673386004, "minimum": 0.1673545140248118, "maximum": 0.17744173364693908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9908251245032514, "standard_deviation": 0.0017643620616271153, "minimum": 0.9877660683519358, "maximum": 0.9928378830710332, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7862593939939027, "standard_deviation": 0.03761163864647145, "minimum": 1.7388904195362909, "maximum": 1.8431065489774845, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7992176682052089, "standard_deviation": 0.3522891771906781, "minimum": 0.24044529349024577, "maximum": 1.3020953806424018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11157351880543775, "standard_deviation": 0.009919754752624969, "minimum": 0.09817160011791493, "maximum": 0.12708555099170898, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968101244487342, "standard_deviation": 0.0010599751979395824, "minimum": 0.9944761424924188, "maximum": 0.9978479147883091, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.060215168726477, "standard_deviation": 0.043895315312750774, "minimum": 1.0096825324890577, "maximum": 1.149990508705685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9629776229524015, "standard_deviation": 0.41425726827755993, "minimum": 0.3590692732523195, "maximum": 1.6697777917633603, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06677162899202353, "standard_deviation": 0.008807903867298829, "minimum": 0.054059250599404175, "maximum": 0.08316204891384854, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16863206766133296, "standard_deviation": 0.02980805152604838, "minimum": 0.1103293389919157, "maximum": 0.20866709223000388, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.03680031618533, "standard_deviation": 0.006339600721295547, "minimum": 6.029235879935039, "maximum": 6.046747508995213, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41555724549838413, "standard_deviation": 0.04059939968981391, "minimum": 0.3682466977427322, "maximum": 0.49623571658766896, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09834684055216283, "standard_deviation": 0.04473290272399379, "minimum": -0.16421885940123815, "maximum": -0.02926589957380253, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9929656465970519, "standard_deviation": 0.0011383701646058332, "minimum": 0.991312749695923, "maximum": 0.9942933345135088, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0510040982918722, "standard_deviation": 0.030077234070066444, "minimum": 1.0199703717891593, "maximum": 1.0974226527554158, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.45002241192349546, "standard_deviation": 0.25648121851797956, "minimum": 0.1459930565101528, "maximum": 0.7754914739296357, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08301762748540037, "standard_deviation": 0.006247436328589927, "minimum": 0.07580296302520675, "maximum": 0.091940933319649, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977044017219976, "standard_deviation": 0.0006526395949622497, "minimum": 0.9963272872891853, "maximum": 0.9985883394718627, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5777935725708601, "standard_deviation": 0.026766328590141057, "minimum": 0.5515854218848227, "maximum": 0.6322175461516795, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5006060845695877, "standard_deviation": 0.3118548423616751, "minimum": 0.1471297411607688, "maximum": 0.9095542941820028, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.046585814331964265, "standard_deviation": 0.005416001923930304, "minimum": 0.03925084050017952, "maximum": 0.05762333652968171, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274384-0010", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.557443857192993, "predicted_mos": 2.557443857192993, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "03933f7b325bf32d716cd42a841734fd7a6b206a65a81675bcc829373c1a4df4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.801390599925071, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0208353025477241, "全帯域_10ms_rms_envelope_nrmse": 0.07570008907488653, "全帯域_10ms_envelope_correlation": 0.9942111766736953, "全帯域_10ms_modulation_mae_db": 0.756759473154747, "全帯域_40ms_log_rms_mae_db": 0.5352645704018082, "全帯域_40ms_rms_envelope_nrmse": 0.03883007358783609, "全帯域_40ms_envelope_correlation": 0.9984888026302722, "全帯域_40ms_modulation_mae_db": 0.9231185562621343, "300–1000 Hz_10ms_log_rms_mae_db": 1.419711835700657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09575480857346473, "300–1000 Hz_10ms_envelope_correlation": 0.9932066648958274, "300–1000 Hz_10ms_modulation_mae_db": 1.235425506930415, "300–1000 Hz_40ms_log_rms_mae_db": 0.8400750920434321, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05340203081820351, "300–1000 Hz_40ms_envelope_correlation": 0.9979297858753886, "300–1000 Hz_40ms_modulation_mae_db": 1.4374188968879942, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5778435590499293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1716758302382554, "1000–3000 Hz_10ms_envelope_correlation": 0.9781561270900464, "1000–3000 Hz_10ms_modulation_mae_db": 0.3417034424563426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0588952083483285, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10361975827932031, "1000–3000 Hz_40ms_envelope_correlation": 0.992285530296585, "1000–3000 Hz_40ms_modulation_mae_db": 0.36324096379769544, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6184498281922912, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2117318637896161, "3000–6000 Hz_10ms_envelope_correlation": 0.9721853585814764, "3000–6000 Hz_10ms_modulation_mae_db": 0.6422437629074875, "3000–6000 Hz_40ms_log_rms_mae_db": 1.250065110012462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1566541152518179, "3000–6000 Hz_40ms_envelope_correlation": 0.9841556952919334, "3000–6000 Hz_40ms_modulation_mae_db": 0.5171962108380536, "mr_stft_db": 5.53288337113291, "band_mae_db": 0.14812984084001626, "rms_gain_error_db": -0.11107112528906989, "output_peak": 0.41788857758081704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1628365324393273, "全帯域_10ms_rms_envelope_nrmse": 0.08018936685818338, "全帯域_10ms_envelope_correlation": 0.993420391251644, "全帯域_10ms_modulation_mae_db": 0.39576079705064804, "全帯域_40ms_log_rms_mae_db": 0.5918018567545114, "全帯域_40ms_rms_envelope_nrmse": 0.041891219692061776, "全帯域_40ms_envelope_correlation": 0.9981325581106862, "全帯域_40ms_modulation_mae_db": 0.3357491660311913, "300–1000 Hz_10ms_log_rms_mae_db": 1.5481546818587841, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10152366294292256, "300–1000 Hz_10ms_envelope_correlation": 0.9924638956587049, "300–1000 Hz_10ms_modulation_mae_db": 1.1031695315383319, "300–1000 Hz_40ms_log_rms_mae_db": 0.9792732089984889, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0604130470911277, "300–1000 Hz_40ms_envelope_correlation": 0.9974470362970318, "300–1000 Hz_40ms_modulation_mae_db": 1.095688401672439, "1000–3000 Hz_10ms_log_rms_mae_db": 1.531003854568917, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14214112717039398, "1000–3000 Hz_10ms_envelope_correlation": 0.9849018308725592, "1000–3000 Hz_10ms_modulation_mae_db": 0.5721592432122269, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9800995461435827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06871846171017136, "1000–3000 Hz_40ms_envelope_correlation": 0.9965560331316404, "1000–3000 Hz_40ms_modulation_mae_db": 0.5599916681858655, "3000–6000 Hz_10ms_log_rms_mae_db": 1.669037588382157, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21287540598335195, "3000–6000 Hz_10ms_envelope_correlation": 0.9723676577308051, "3000–6000 Hz_10ms_modulation_mae_db": 0.18550906031774153, "3000–6000 Hz_40ms_log_rms_mae_db": 1.245142666852689, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15823094591062264, "3000–6000 Hz_40ms_envelope_correlation": 0.9841263179552288, "3000–6000 Hz_40ms_modulation_mae_db": 0.37398951563387445, "mr_stft_db": 5.507924049943683, "band_mae_db": 0.1457821694098935, "rms_gain_error_db": -0.08207294719000467, "output_peak": 0.4419570302331025}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0778252511925486, "全帯域_10ms_rms_envelope_nrmse": 0.08600600835317686, "全帯域_10ms_envelope_correlation": 0.992470601932952, "全帯域_10ms_modulation_mae_db": 0.17795806080825596, "全帯域_40ms_log_rms_mae_db": 0.6042515326780445, "全帯域_40ms_rms_envelope_nrmse": 0.04660211436108946, "全帯域_40ms_envelope_correlation": 0.9977170713844006, "全帯域_40ms_modulation_mae_db": 0.2984706311962763, "300–1000 Hz_10ms_log_rms_mae_db": 1.4838443070417036, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11009167192989344, "300–1000 Hz_10ms_envelope_correlation": 0.9910083552979572, "300–1000 Hz_10ms_modulation_mae_db": 0.22505378445039506, "300–1000 Hz_40ms_log_rms_mae_db": 0.9390338601287496, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06273615552928265, "300–1000 Hz_40ms_envelope_correlation": 0.9971334379956758, "300–1000 Hz_40ms_modulation_mae_db": 0.33439735415974364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.587315041026312, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1694158922288149, "1000–3000 Hz_10ms_envelope_correlation": 0.9785341349898339, "1000–3000 Hz_10ms_modulation_mae_db": 0.5105186305794315, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0511842766326225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09738140012996964, "1000–3000 Hz_40ms_envelope_correlation": 0.9929408769795408, "1000–3000 Hz_40ms_modulation_mae_db": 0.5316014518642084, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6986929275857447, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21722026307483264, "3000–6000 Hz_10ms_envelope_correlation": 0.9713775609335202, "3000–6000 Hz_10ms_modulation_mae_db": 0.5383891618392793, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3145158284958431, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1551823200180622, "3000–6000 Hz_40ms_envelope_correlation": 0.9848270775318617, "3000–6000 Hz_40ms_modulation_mae_db": 0.4261704231030665, "mr_stft_db": 5.522838054707068, "band_mae_db": 0.1554640964153633, "rms_gain_error_db": -0.10394382276139306, "output_peak": 0.3895793049737333}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.086839438585662, "全帯域_10ms_rms_envelope_nrmse": 0.07571609431366395, "全帯域_10ms_envelope_correlation": 0.9941362786503164, "全帯域_10ms_modulation_mae_db": 0.4334983927776787, "全帯域_40ms_log_rms_mae_db": 0.5710599574481587, "全帯域_40ms_rms_envelope_nrmse": 0.039567090376430636, "全帯域_40ms_envelope_correlation": 0.9983259726739738, "全帯域_40ms_modulation_mae_db": 0.424252088279405, "300–1000 Hz_10ms_log_rms_mae_db": 1.4518615976403935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08889058755450266, "300–1000 Hz_10ms_envelope_correlation": 0.9941537925614793, "300–1000 Hz_10ms_modulation_mae_db": 0.728509724004535, "300–1000 Hz_40ms_log_rms_mae_db": 0.8508476204326703, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04988918150731788, "300–1000 Hz_40ms_envelope_correlation": 0.9981920848698749, "300–1000 Hz_40ms_modulation_mae_db": 0.8850202527052575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.599803718840708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16265537620558065, "1000–3000 Hz_10ms_envelope_correlation": 0.9804237875437313, "1000–3000 Hz_10ms_modulation_mae_db": 0.3404148939505222, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9954516894377953, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08808718251019917, "1000–3000 Hz_40ms_envelope_correlation": 0.9945851066580733, "1000–3000 Hz_40ms_modulation_mae_db": 0.6834448580302528, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6884258466831856, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2105110594486138, "3000–6000 Hz_10ms_envelope_correlation": 0.9746648483852954, "3000–6000 Hz_10ms_modulation_mae_db": 0.4053780980141963, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2861742090783395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15221147549972946, "3000–6000 Hz_40ms_envelope_correlation": 0.986603612623239, "3000–6000 Hz_40ms_modulation_mae_db": 0.4034823000578592, "mr_stft_db": 5.525989700592305, "band_mae_db": 0.2250234122172846, "rms_gain_error_db": -0.07547148497113568, "output_peak": 0.40735470318810985}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1410569826762897, "全帯域_10ms_rms_envelope_nrmse": 0.0861235773610987, "全帯域_10ms_envelope_correlation": 0.9924105923034092, "全帯域_10ms_modulation_mae_db": 0.6904719362570928, "全帯域_40ms_log_rms_mae_db": 0.7372158312083033, "全帯域_40ms_rms_envelope_nrmse": 0.051844468402639565, "全帯域_40ms_envelope_correlation": 0.9970669005897175, "全帯域_40ms_modulation_mae_db": 0.892251341424641, "300–1000 Hz_10ms_log_rms_mae_db": 1.441526232221328, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10981771527958202, "300–1000 Hz_10ms_envelope_correlation": 0.9909035164830227, "300–1000 Hz_10ms_modulation_mae_db": 0.919987202464512, "300–1000 Hz_40ms_log_rms_mae_db": 0.8778916619483552, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.071078376117147, "300–1000 Hz_40ms_envelope_correlation": 0.9959947643712277, "300–1000 Hz_40ms_modulation_mae_db": 1.29944499750357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5468611670880217, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17878684673027156, "1000–3000 Hz_10ms_envelope_correlation": 0.9759301976262088, "1000–3000 Hz_10ms_modulation_mae_db": 0.46810344670454573, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0141078093105211, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10744085528366948, "1000–3000 Hz_40ms_envelope_correlation": 0.991024246434306, "1000–3000 Hz_40ms_modulation_mae_db": 0.2741974573437035, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6156280963493386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21171710398596436, "3000–6000 Hz_10ms_envelope_correlation": 0.9720957794714976, "3000–6000 Hz_10ms_modulation_mae_db": 0.7076516177056416, "3000–6000 Hz_40ms_log_rms_mae_db": 1.216643271264091, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15662991132628745, "3000–6000 Hz_40ms_envelope_correlation": 0.9842350001882479, "3000–6000 Hz_40ms_modulation_mae_db": 0.6034151913603883, "mr_stft_db": 5.514050769851967, "band_mae_db": 0.15428196071493494, "rms_gain_error_db": -0.07017563331846134, "output_peak": 0.4475108783416478}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0867059422659635, "全帯域_10ms_rms_envelope_nrmse": 0.08406983090803598, "全帯域_10ms_envelope_correlation": 0.9930210202898226, "全帯域_10ms_modulation_mae_db": 0.6222384016119957, "全帯域_40ms_log_rms_mae_db": 0.6474912858511908, "全帯域_40ms_rms_envelope_nrmse": 0.056706401739270995, "全帯域_40ms_envelope_correlation": 0.9968099887430978, "全帯域_40ms_modulation_mae_db": 0.7299022426612076, "300–1000 Hz_10ms_log_rms_mae_db": 1.4979872519447026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11708529832427324, "300–1000 Hz_10ms_envelope_correlation": 0.9900480726582164, "300–1000 Hz_10ms_modulation_mae_db": 1.2129370914665012, "300–1000 Hz_40ms_log_rms_mae_db": 0.8822269012175324, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07872424352892006, "300–1000 Hz_40ms_envelope_correlation": 0.9956173131892844, "300–1000 Hz_40ms_modulation_mae_db": 1.2207329941042508, "1000–3000 Hz_10ms_log_rms_mae_db": 1.593908456874679, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15452017803465978, "1000–3000 Hz_10ms_envelope_correlation": 0.9829848399524814, "1000–3000 Hz_10ms_modulation_mae_db": 0.33745751215920805, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0351388287587502, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09055117002853884, "1000–3000 Hz_40ms_envelope_correlation": 0.9950792537785893, "1000–3000 Hz_40ms_modulation_mae_db": 0.7182604280822179, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6729053969458014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21196380216492502, "3000–6000 Hz_10ms_envelope_correlation": 0.9720164386342077, "3000–6000 Hz_10ms_modulation_mae_db": 0.7343409311637097, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2477819963048098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16234141247255165, "3000–6000 Hz_40ms_envelope_correlation": 0.9831611368911631, "3000–6000 Hz_40ms_modulation_mae_db": 0.8684149334527621, "mr_stft_db": 5.53225419442271, "band_mae_db": 0.17808670074999733, "rms_gain_error_db": -0.17808047464461063, "output_peak": 0.554919750444099}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.147037989705363, "全帯域_10ms_rms_envelope_nrmse": 0.08231165954842722, "全帯域_10ms_envelope_correlation": 0.9932099241960428, "全帯域_10ms_modulation_mae_db": 0.2534080683532534, "全帯域_40ms_log_rms_mae_db": 0.6303905556593018, "全帯域_40ms_rms_envelope_nrmse": 0.04765869602024492, "全帯域_40ms_envelope_correlation": 0.9977425907930713, "全帯域_40ms_modulation_mae_db": 0.2530775171776787, "300–1000 Hz_10ms_log_rms_mae_db": 1.56054570983344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10691854946106143, "300–1000 Hz_10ms_envelope_correlation": 0.9917286454846693, "300–1000 Hz_10ms_modulation_mae_db": 0.2899336293397426, "300–1000 Hz_40ms_log_rms_mae_db": 1.0124773807902303, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06437347780972018, "300–1000 Hz_40ms_envelope_correlation": 0.9972052146370022, "300–1000 Hz_40ms_modulation_mae_db": 0.3647018034481328, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5878912246654457, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15918808855555724, "1000–3000 Hz_10ms_envelope_correlation": 0.9813942388221325, "1000–3000 Hz_10ms_modulation_mae_db": 0.41732153450803366, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0357152693269172, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08568221783926944, "1000–3000 Hz_40ms_envelope_correlation": 0.9951455153130594, "1000–3000 Hz_40ms_modulation_mae_db": 0.274648688960615, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6071272442214506, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2016332334290017, "3000–6000 Hz_10ms_envelope_correlation": 0.9747511591559122, "3000–6000 Hz_10ms_modulation_mae_db": 1.1280046538550563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1851025645121467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15304737844134617, "3000–6000 Hz_40ms_envelope_correlation": 0.9851435132500507, "3000–6000 Hz_40ms_modulation_mae_db": 1.1356782170623114, "mr_stft_db": 5.5334471448264315, "band_mae_db": 0.21145457655831643, "rms_gain_error_db": -0.14585391776387768, "output_peak": 0.46803303311373756}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0902454060632023, "全帯域_10ms_rms_envelope_nrmse": 0.07472180392612846, "全帯域_10ms_envelope_correlation": 0.9943150487878121, "全帯域_10ms_modulation_mae_db": 0.1152326113315093, "全帯域_40ms_log_rms_mae_db": 0.6083850611552034, "全帯域_40ms_rms_envelope_nrmse": 0.041178111411021505, "全帯域_40ms_envelope_correlation": 0.9982124074967426, "全帯域_40ms_modulation_mae_db": 0.1450731274552217, "300–1000 Hz_10ms_log_rms_mae_db": 1.4735093061004736, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09319634444874247, "300–1000 Hz_10ms_envelope_correlation": 0.9935935422103344, "300–1000 Hz_10ms_modulation_mae_db": 0.35318527111635994, "300–1000 Hz_40ms_log_rms_mae_db": 0.8946225091795642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05413358167867532, "300–1000 Hz_40ms_envelope_correlation": 0.9978800360950261, "300–1000 Hz_40ms_modulation_mae_db": 0.5238209812931934, "1000–3000 Hz_10ms_log_rms_mae_db": 1.535239613841353, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16690379438305844, "1000–3000 Hz_10ms_envelope_correlation": 0.9799505753959216, "1000–3000 Hz_10ms_modulation_mae_db": 0.6942993317749794, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9905682192054792, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09559120061215531, "1000–3000 Hz_40ms_envelope_correlation": 0.994410948677419, "1000–3000 Hz_40ms_modulation_mae_db": 0.40092941015026734, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6610595458603803, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21004584040430896, "3000–6000 Hz_10ms_envelope_correlation": 0.973142287824181, "3000–6000 Hz_10ms_modulation_mae_db": 0.5566782998029006, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2145472773948551, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15502794934675737, "3000–6000 Hz_40ms_envelope_correlation": 0.9847684541281656, "3000–6000 Hz_40ms_modulation_mae_db": 0.5152495149479154, "mr_stft_db": 5.523374903205198, "band_mae_db": 0.215144906358542, "rms_gain_error_db": -0.09073608023039012, "output_peak": 0.40075255588276737}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.637352907544144, "spectral_convergence": 0.24791208523005945, "spectral_convergence_status": "ok", "amplitude_floor": 1.597965137573199e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.43791402053962, "spectral_convergence": 0.23196901269148415, "spectral_convergence_status": "ok", "amplitude_floor": 1.057274363712373e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.523383185314965, "spectral_convergence": 0.23869975850975406, "spectral_convergence_status": "ok", "amplitude_floor": 9.008040625007043e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007434014661781762, "output_power": 0.000750034423872551, "delta_db": 0.038577830975643206, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002448126060340392, "output_power": 0.0023535170344973293, "delta_db": -0.17116430965366208, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00039049712732524606, "output_power": 0.0003912707099661679, "delta_db": 0.008594950210982609, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012090645510899839, "output_power": 0.00011115061442301896, "delta_db": -0.3653762039807262, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.957546973001399e-05, "output_power": 5.7130639054882136e-05, "delta_db": -0.18198393237902655, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.142800821939945e-05, "output_power": 8.376874202509298e-05, "delta_db": 0.12308181784005702, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14812984084001626, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11107112528906989, "pitch": null, "output_rms_dbfs": -24.263311239562086, "output_peak": 0.41788857758081704, "reference_rms_dbfs": -24.152240114273017, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.53288337113291}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019565573818943504, "log_rms_mae_db": 1.0208353025477241, "rms_envelope_nrmse": 0.07570008907488653, "envelope_correlation": 0.9942111766736953, "rms_gain_error_db": -0.1110536808663761, "reference_mean_power": 0.0038470996951946897, "output_mean_power": 0.0037499724131996753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.743468430119212e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007829242029289011, "output_power": 0.0007724194513214871, "reference_active": true, "delta_db": -0.05866517043557239}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011656588263872551, "output_power": 9.391578474703822e-05, "reference_active": true, "delta_db": -0.938328647537215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.15404555717803e-05, "output_power": 2.3525534634286674e-05, "reference_active": true, "delta_db": -1.2732846014914534}], "active_band_count": 3, "active_band_mae_db": 0.756759473154747}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018472557440126405, "log_rms_mae_db": 0.5352645704018082, "rms_envelope_nrmse": 0.03883007358783609, "envelope_correlation": 0.9984888026302722, "rms_gain_error_db": -0.11126759011354438, "reference_mean_power": 0.0038659686805529734, "output_mean_power": 0.0037681794113836303}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.614770154461247e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007106100759473788, "output_power": 0.0007013400141415899, "reference_active": true, "delta_db": -0.0570274341107197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.7305685337334634e-05, "output_power": 4.535949263474499e-05, "reference_active": true, "delta_db": -1.0152952262260366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9649709160477995e-06, "output_power": 1.329391280275123e-06, "reference_active": true, "delta_db": -1.6970330084496466}], "active_band_count": 3, "active_band_mae_db": 0.9231185562621343}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016965929826439602, "log_rms_mae_db": 1.419711835700657, "rms_envelope_nrmse": 0.09575480857346473, "envelope_correlation": 0.9932066648958274, "rms_gain_error_db": -0.15986623076842627, "reference_mean_power": 0.0021633278035333714, "output_mean_power": 0.002085142349189148}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3458386409803657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007177587670080827, "output_power": 0.0006890470202197975, "reference_active": true, "delta_db": -0.17729646902368668}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.048912255781678e-05, "output_power": 7.126878189996644e-05, "reference_active": true, "delta_db": -1.0369704097054955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.109541579796594e-05, "output_power": 1.751843895829872e-05, "reference_active": true, "delta_db": -2.492009642062063}], "active_band_count": 3, "active_band_mae_db": 1.235425506930415}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015157137475553504, "log_rms_mae_db": 0.8400750920434321, "rms_envelope_nrmse": 0.05340203081820351, "envelope_correlation": 0.9979297858753886, "rms_gain_error_db": -0.15986452458028214, "reference_mean_power": 0.002174161455071234, "output_mean_power": 0.0020955852818814527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2731603714689836e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006867133707128568, "output_power": 0.0006543637303082389, "reference_active": true, "delta_db": -0.20956283936553174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.067232594302272e-05, "output_power": 2.96548664342219e-05, "reference_active": true, "delta_db": -1.3720303696286726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.08557261679534e-06, "output_power": 1.112138736638442e-06, "reference_active": true, "delta_db": -2.7306634816697786}], "active_band_count": 3, "active_band_mae_db": 1.4374188968879942}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.465903254113391e-05, "log_rms_mae_db": 1.5778435590499293, "rms_envelope_nrmse": 0.1716758302382554, "envelope_correlation": 0.9781561270900464, "rms_gain_error_db": -0.32227945161578875, "reference_mean_power": 0.00014077271984820836, "output_mean_power": 0.0001307045049323963}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.622290255773331e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.707421130487826e-05, "output_power": 3.604209542250774e-05, "reference_active": true, "delta_db": -0.12261888731553468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.08772682349557e-06, "output_power": 4.9464656162295305e-06, "reference_active": true, "delta_db": -0.12228790026314418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8011808387082025e-06, "output_power": 2.155647139744601e-06, "reference_active": true, "delta_db": 0.780203539790349}], "active_band_count": 3, "active_band_mae_db": 0.3417034424563426}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.107059424790359e-05, "log_rms_mae_db": 1.0588952083483285, "rms_envelope_nrmse": 0.10361975827932031, "envelope_correlation": 0.992285530296585, "rms_gain_error_db": -0.32271371120706166, "reference_mean_power": 0.00014139647469838974, "output_mean_power": 0.00013127052151727547}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.069193023932203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.382591725297068e-05, "output_power": 3.358721560437652e-05, "reference_active": true, "delta_db": -0.03075580014357171}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1244897853414024e-06, "output_power": 2.234544646664333e-06, "reference_active": true, "delta_db": 0.2193438905681529}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.400225448356044e-07, "output_power": 1.1540792866186682e-07, "reference_active": true, "delta_db": -0.8396232006813618}], "active_band_count": 3, "active_band_mae_db": 0.36324096379769544}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.428054041295646e-05, "log_rms_mae_db": 1.6184498281922912, "rms_envelope_nrmse": 0.2117318637896161, "envelope_correlation": 0.9721853585814764, "rms_gain_error_db": -0.061194369464324556, "reference_mean_power": 8.952225188868405e-05, "output_mean_power": 8.826968183232307e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.314971416941356e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.877588546016404e-05, "output_power": 5.3122357098075526e-05, "reference_active": true, "delta_db": -0.43921843586595644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.04956306951581e-06, "output_power": 8.487233245427767e-06, "reference_active": true, "delta_db": -0.2786147354520847}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5294615410314917e-06, "output_power": 1.9148654748553207e-06, "reference_active": true, "delta_db": -1.2088981174044213}], "active_band_count": 3, "active_band_mae_db": 0.6422437629074875}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5452516433588795e-05, "log_rms_mae_db": 1.250065110012462, "rms_envelope_nrmse": 0.1566541152518179, "envelope_correlation": 0.9841556952919334, "rms_gain_error_db": -0.061218176064421904, "reference_mean_power": 8.996899751083782e-05, "output_mean_power": 8.870969043716626e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.951283845645369e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.884212028117595e-05, "output_power": 5.3147118733632584e-05, "reference_active": true, "delta_db": -0.44208588363160245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.436071463262977e-06, "output_power": 4.312304100416789e-06, "reference_active": true, "delta_db": -0.12289154460328158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7711408251558852e-07, "output_power": 1.4112110660013023e-07, "reference_active": true, "delta_db": -0.9866112042792765}], "active_band_count": 3, "active_band_mae_db": 0.5171962108380536}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.896862191218306, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cdb5db60a41d1a74521718b845e6c5b25931226f7df92c14b0175a023e729f0f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9802844665366144, "standard_deviation": 0.0028417224461365735, "minimum": 0.9759301976262088, "maximum": 0.9849018308725592, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5699833294944208, "standard_deviation": 0.02778923512253628, "minimum": 1.531003854568917, "maximum": 1.599803718840708, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4602472544181613, "standard_deviation": 0.12829985378723532, "minimum": 0.33745751215920805, "maximum": 0.6942993317749794, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16316089169332398, "standard_deviation": 0.011353666149843593, "minimum": 0.14214112717039398, "maximum": 0.17878684673027156, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9940034389086516, "standard_deviation": 0.0017903587289895765, "minimum": 0.991024246434306, "maximum": 0.9965560331316404, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0201451058954996, "standard_deviation": 0.02941548455991281, "minimum": 0.9800995461435827, "maximum": 1.0588952083483285, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.47578936580185327, "standard_deviation": 0.1738168417629748, "minimum": 0.2741974573437035, "maximum": 0.7182604280822179, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09213403079916169, "standard_deviation": 0.012038657465975091, "minimum": 0.06871846171017136, "maximum": 0.10744085528366948, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9728251363396119, "standard_deviation": 0.0012589519222376002, "minimum": 0.9713775609335202, "maximum": 0.9747511591559122, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6539158092775437, "standard_deviation": 0.03535370907356914, "minimum": 1.6071272442214506, "maximum": 1.6986929275857447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6122744482007516, "standard_deviation": 0.2736936762270602, "minimum": 0.18550906031774153, "maximum": 1.1280046538550563, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21096232153507682, "standard_deviation": 0.004359915459985251, "minimum": 0.2016332334290017, "maximum": 0.21722026307483264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9846276009824863, "standard_deviation": 0.0010006462157441632, "minimum": 0.9831611368911631, "maximum": 0.986603612623239, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2449966154894045, "standard_deviation": 0.04119701127587073, "minimum": 1.1851025645121467, "maximum": 1.3145158284958431, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6054495383070289, "standard_deviation": 0.26531048979846883, "minimum": 0.37398951563387445, "maximum": 1.1356782170623114, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15616568853339685, "standard_deviation": 0.0031731021393474313, "minimum": 0.15221147549972946, "maximum": 0.16234141247255165, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921383106562764, "standard_deviation": 0.0014523718908125414, "minimum": 0.9900480726582164, "maximum": 0.9941537925614793, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4846426152926853, "standard_deviation": 0.04963776399832103, "minimum": 1.419711835700657, "maximum": 1.56054570983344, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7585252176638491, "standard_deviation": 0.4222634521931968, "minimum": 0.22505378445039506, "maximum": 1.235425506930415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10290982981430533, "standard_deviation": 0.009712423689689087, "minimum": 0.08889058755450266, "maximum": 0.11708529832427324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971749591663139, "standard_deviation": 0.0009256061830931133, "minimum": 0.9956173131892844, "maximum": 0.9981920848698749, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9095560293423779, "standard_deviation": 0.06160824804075144, "minimum": 0.8400750920434321, "maximum": 1.0124773807902303, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8951532102218227, "standard_deviation": 0.4371231616256626, "minimum": 0.33439735415974364, "maximum": 1.4374188968879942, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.061843761760049284, "standard_deviation": 0.009659142439526387, "minimum": 0.04988918150731788, "maximum": 0.07872424352892006, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17917095790804355, "standard_deviation": 0.033165859790190544, "minimum": 0.1457821694098935, "maximum": 0.2250234122172846, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.524095273585285, "standard_deviation": 0.009253146640401742, "minimum": 5.507924049943683, "maximum": 5.5334471448264315, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4409994792197518, "standard_deviation": 0.05302545256869375, "minimum": 0.3895793049737333, "maximum": 0.554919750444099, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10717568577111788, "standard_deviation": 0.03746175658932213, "minimum": -0.17808047464461063, "maximum": -0.07017563331846134, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933993792607119, "standard_deviation": 0.0007616057886018246, "minimum": 0.9924105923034092, "maximum": 0.9943150487878121, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1016728556845101, "standard_deviation": 0.04633441146384195, "minimum": 1.0208353025477241, "maximum": 1.1628365324393273, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4306659676681476, "standard_deviation": 0.2409926963916483, "minimum": 0.1152326113315093, "maximum": 0.756759473154747, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08060480379295014, "standard_deviation": 0.0047402173547880875, "minimum": 0.07472180392612846, "maximum": 0.0861235773610987, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978120365527452, "standard_deviation": 0.0006045473082337741, "minimum": 0.9968099887430978, "maximum": 0.9984888026302722, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6157325813945652, "standard_deviation": 0.06001294591538514, "minimum": 0.5352645704018082, "maximum": 0.7372158312083033, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5002368338109695, "standard_deviation": 0.30379010291855607, "minimum": 0.1450731274552217, "maximum": 0.9231185562621343, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04553477194882437, "standard_deviation": 0.006359357810693481, "minimum": 0.03883007358783609, "maximum": 0.056706401739270995, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274384-0010", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.816729784011841, "predicted_mos": 2.816729784011841, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "17c5c1dbad9e26b6209b4c9711c8931396c4eec5868910672a1e4a7fbdcde149", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7366273999214172, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0923800722549886, "全帯域_10ms_rms_envelope_nrmse": 0.07650110134622684, "全帯域_10ms_envelope_correlation": 0.9941274609590961, "全帯域_10ms_modulation_mae_db": 0.2718033427238487, "全帯域_40ms_log_rms_mae_db": 0.6353917374411323, "全帯域_40ms_rms_envelope_nrmse": 0.043960868752885136, "全帯域_40ms_envelope_correlation": 0.9980690751421075, "全帯域_40ms_modulation_mae_db": 0.41799366941070976, "300–1000 Hz_10ms_log_rms_mae_db": 1.6018453625136506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09217693656286025, "300–1000 Hz_10ms_envelope_correlation": 0.993926765049457, "300–1000 Hz_10ms_modulation_mae_db": 0.7959029133654466, "300–1000 Hz_40ms_log_rms_mae_db": 0.9808505873315507, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05776171384334925, "300–1000 Hz_40ms_envelope_correlation": 0.9977666033500124, "300–1000 Hz_40ms_modulation_mae_db": 0.6876149750496204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7557557368978602, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1655333801617483, "1000–3000 Hz_10ms_envelope_correlation": 0.9793469343087413, "1000–3000 Hz_10ms_modulation_mae_db": 0.770452135039774, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0806323420689727, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08395984349312088, "1000–3000 Hz_40ms_envelope_correlation": 0.9944831449326794, "1000–3000 Hz_40ms_modulation_mae_db": 0.16291102083536357, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8938815592008198, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2128397105963751, "3000–6000 Hz_10ms_envelope_correlation": 0.9738175367038501, "3000–6000 Hz_10ms_modulation_mae_db": 0.3813188242096814, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3721657923482524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1532539466276058, "3000–6000 Hz_40ms_envelope_correlation": 0.9862459167732353, "3000–6000 Hz_40ms_modulation_mae_db": 0.09502680066071793, "mr_stft_db": 5.689079362236111, "band_mae_db": 0.18593338718973654, "rms_gain_error_db": -0.13033083421521555, "output_peak": 0.4213873610883026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0500910201353215, "全帯域_10ms_rms_envelope_nrmse": 0.08318082992990466, "全帯域_10ms_envelope_correlation": 0.9929041904168389, "全帯域_10ms_modulation_mae_db": 0.2027617235191951, "全帯域_40ms_log_rms_mae_db": 0.5760826671926431, "全帯域_40ms_rms_envelope_nrmse": 0.04956571215799036, "全帯域_40ms_envelope_correlation": 0.997319563181517, "全帯域_40ms_modulation_mae_db": 0.16303378238256605, "300–1000 Hz_10ms_log_rms_mae_db": 1.616921959408889, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10513229114444944, "300–1000 Hz_10ms_envelope_correlation": 0.9918657640232871, "300–1000 Hz_10ms_modulation_mae_db": 0.5384256717803041, "300–1000 Hz_40ms_log_rms_mae_db": 0.9512035338332933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06353026824126227, "300–1000 Hz_40ms_envelope_correlation": 0.9970947753929001, "300–1000 Hz_40ms_modulation_mae_db": 0.48261624535059267, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7836625653926013, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1450371029750051, "1000–3000 Hz_10ms_envelope_correlation": 0.9842548685109782, "1000–3000 Hz_10ms_modulation_mae_db": 0.39590808419727613, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0954353207767669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0758098383684213, "1000–3000 Hz_40ms_envelope_correlation": 0.9957275662480053, "1000–3000 Hz_40ms_modulation_mae_db": 0.5750368058644953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8988752195760188, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2128445631472356, "3000–6000 Hz_10ms_envelope_correlation": 0.9723142855116322, "3000–6000 Hz_10ms_modulation_mae_db": 0.3015410708675013, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3746078763670952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15433194125132352, "3000–6000 Hz_40ms_envelope_correlation": 0.9848996885621383, "3000–6000 Hz_40ms_modulation_mae_db": 0.22419392814886574, "mr_stft_db": 5.7015098955337535, "band_mae_db": 0.16445862570860678, "rms_gain_error_db": -0.07443649317646894, "output_peak": 0.39544438655481684}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0574287230908959, "全帯域_10ms_rms_envelope_nrmse": 0.08150053835525897, "全帯域_10ms_envelope_correlation": 0.9931779014915028, "全帯域_10ms_modulation_mae_db": 0.3927739120473799, "全帯域_40ms_log_rms_mae_db": 0.5296188093600518, "全帯域_40ms_rms_envelope_nrmse": 0.03819372365414579, "全帯域_40ms_envelope_correlation": 0.9984057984795969, "全帯域_40ms_modulation_mae_db": 0.46485804557014987, "300–1000 Hz_10ms_log_rms_mae_db": 1.5932901419838756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10214216262091501, "300–1000 Hz_10ms_envelope_correlation": 0.9921424186085988, "300–1000 Hz_10ms_modulation_mae_db": 0.5383902418712997, "300–1000 Hz_40ms_log_rms_mae_db": 0.8877196186161849, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055586692690130286, "300–1000 Hz_40ms_envelope_correlation": 0.9976000104571533, "300–1000 Hz_40ms_modulation_mae_db": 0.4889521260378451, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7832100968754951, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14741029029888253, "1000–3000 Hz_10ms_envelope_correlation": 0.9840041012890705, "1000–3000 Hz_10ms_modulation_mae_db": 0.6245539031405164, "1000–3000 Hz_40ms_log_rms_mae_db": 1.039906667779025, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07032893211890182, "1000–3000 Hz_40ms_envelope_correlation": 0.9968620771112063, "1000–3000 Hz_40ms_modulation_mae_db": 0.3384299386960226, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8956454567311194, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2064949311788305, "3000–6000 Hz_10ms_envelope_correlation": 0.9739161276765754, "3000–6000 Hz_10ms_modulation_mae_db": 0.4262263822149361, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3761066640329036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1539326563881219, "3000–6000 Hz_40ms_envelope_correlation": 0.9849392352149667, "3000–6000 Hz_40ms_modulation_mae_db": 0.3640360732327555, "mr_stft_db": 5.6992926068593865, "band_mae_db": 0.13785701493362065, "rms_gain_error_db": -0.05115135782122303, "output_peak": 0.3849022634075823}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9954662202401671, "全帯域_10ms_rms_envelope_nrmse": 0.07213915210642247, "全帯域_10ms_envelope_correlation": 0.9946708244587465, "全帯域_10ms_modulation_mae_db": 0.2143232406551351, "全帯域_40ms_log_rms_mae_db": 0.5441204738060302, "全帯域_40ms_rms_envelope_nrmse": 0.03532039678236768, "全帯域_40ms_envelope_correlation": 0.9986622783152268, "全帯域_40ms_modulation_mae_db": 0.23458641975395036, "300–1000 Hz_10ms_log_rms_mae_db": 1.6734365443154122, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1032764877439793, "300–1000 Hz_10ms_envelope_correlation": 0.9920288462672323, "300–1000 Hz_10ms_modulation_mae_db": 0.3112667893894825, "300–1000 Hz_40ms_log_rms_mae_db": 1.0182457711047503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05447504959792102, "300–1000 Hz_40ms_envelope_correlation": 0.9977953057367064, "300–1000 Hz_40ms_modulation_mae_db": 0.451918612417116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8079480280056865, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1771410280908254, "1000–3000 Hz_10ms_envelope_correlation": 0.976346545085969, "1000–3000 Hz_10ms_modulation_mae_db": 0.5170030368388485, "1000–3000 Hz_40ms_log_rms_mae_db": 1.160336829808895, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09961952616001916, "1000–3000 Hz_40ms_envelope_correlation": 0.9921930079605991, "1000–3000 Hz_40ms_modulation_mae_db": 0.4177093487239907, "3000–6000 Hz_10ms_log_rms_mae_db": 1.927891386204417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20365444881946565, "3000–6000 Hz_10ms_envelope_correlation": 0.9747225505273253, "3000–6000 Hz_10ms_modulation_mae_db": 0.351230108266789, "3000–6000 Hz_40ms_log_rms_mae_db": 1.446416602471589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15586532583844645, "3000–6000 Hz_40ms_envelope_correlation": 0.984607050664807, "3000–6000 Hz_40ms_modulation_mae_db": 0.33787055028646695, "mr_stft_db": 5.70067050366794, "band_mae_db": 0.16785317467899155, "rms_gain_error_db": -0.06497933397370746, "output_peak": 0.39654651832184196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0872041485450885, "全帯域_10ms_rms_envelope_nrmse": 0.08880961326547926, "全帯域_10ms_envelope_correlation": 0.9919329206349894, "全帯域_10ms_modulation_mae_db": 0.17467132069744792, "全帯域_40ms_log_rms_mae_db": 0.6064764815994883, "全帯域_40ms_rms_envelope_nrmse": 0.05180286071399013, "全帯域_40ms_envelope_correlation": 0.997121076273848, "全帯域_40ms_modulation_mae_db": 0.4487594247177196, "300–1000 Hz_10ms_log_rms_mae_db": 1.586868036156407, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11614410310158887, "300–1000 Hz_10ms_envelope_correlation": 0.989911578399995, "300–1000 Hz_10ms_modulation_mae_db": 0.29335263235728276, "300–1000 Hz_40ms_log_rms_mae_db": 0.9753077915153202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06813900457837958, "300–1000 Hz_40ms_envelope_correlation": 0.9965083903769957, "300–1000 Hz_40ms_modulation_mae_db": 0.5812286502538863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8296103640877952, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1656871771039049, "1000–3000 Hz_10ms_envelope_correlation": 0.9795848008410983, "1000–3000 Hz_10ms_modulation_mae_db": 0.6731780568242448, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1738055308996098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0907702355453649, "1000–3000 Hz_40ms_envelope_correlation": 0.9941814558977904, "1000–3000 Hz_40ms_modulation_mae_db": 1.1819350961704707, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8999614748010842, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21687972453739693, "3000–6000 Hz_10ms_envelope_correlation": 0.9707537315421738, "3000–6000 Hz_10ms_modulation_mae_db": 0.6312919330621366, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3501999852549782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17709883190054404, "3000–6000 Hz_40ms_envelope_correlation": 0.9799266194370702, "3000–6000 Hz_40ms_modulation_mae_db": 0.48538404291450227, "mr_stft_db": 5.677227527273334, "band_mae_db": 0.17379135808255053, "rms_gain_error_db": -0.09954321510059094, "output_peak": 0.41509667361704716}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0293621736976428, "全帯域_10ms_rms_envelope_nrmse": 0.0747328934969793, "全帯域_10ms_envelope_correlation": 0.9942884128581453, "全帯域_10ms_modulation_mae_db": 0.09494506901783133, "全帯域_40ms_log_rms_mae_db": 0.5495362158645729, "全帯域_40ms_rms_envelope_nrmse": 0.040406595463526804, "全帯域_40ms_envelope_correlation": 0.9982032427076892, "全帯域_40ms_modulation_mae_db": 0.14725094623403215, "300–1000 Hz_10ms_log_rms_mae_db": 1.6234879755096854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09855255599163133, "300–1000 Hz_10ms_envelope_correlation": 0.9926650690521496, "300–1000 Hz_10ms_modulation_mae_db": 0.20411703469583187, "300–1000 Hz_40ms_log_rms_mae_db": 0.9627033906547863, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05563304161132437, "300–1000 Hz_40ms_envelope_correlation": 0.9975440868150551, "300–1000 Hz_40ms_modulation_mae_db": 0.38339542553796324, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8180836771796391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16948907143732414, "1000–3000 Hz_10ms_envelope_correlation": 0.9783611360892797, "1000–3000 Hz_10ms_modulation_mae_db": 0.49957661874415465, "1000–3000 Hz_40ms_log_rms_mae_db": 1.166309429262494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10359147655556687, "1000–3000 Hz_40ms_envelope_correlation": 0.9917478341301387, "1000–3000 Hz_40ms_modulation_mae_db": 0.8466138526872952, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9166321699406546, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19795000478511793, "3000–6000 Hz_10ms_envelope_correlation": 0.9759845102367702, "3000–6000 Hz_10ms_modulation_mae_db": 0.4421863446568186, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4308685201315035, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15179142787950234, "3000–6000 Hz_40ms_envelope_correlation": 0.9853340585684297, "3000–6000 Hz_40ms_modulation_mae_db": 0.3697950809297476, "mr_stft_db": 5.695064248140732, "band_mae_db": 0.13632497458922693, "rms_gain_error_db": -0.024558808257810413, "output_peak": 0.3973062074516761}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0621320166782549, "全帯域_10ms_rms_envelope_nrmse": 0.07526630942619963, "全帯域_10ms_envelope_correlation": 0.9941775980610399, "全帯域_10ms_modulation_mae_db": 0.6554595321059291, "全帯域_40ms_log_rms_mae_db": 0.5575141337895032, "全帯域_40ms_rms_envelope_nrmse": 0.0382668136985333, "全帯域_40ms_envelope_correlation": 0.9983716006480425, "全帯域_40ms_modulation_mae_db": 0.6172421346674191, "300–1000 Hz_10ms_log_rms_mae_db": 1.6336733231094505, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10460529443233833, "300–1000 Hz_10ms_envelope_correlation": 0.9917242263502583, "300–1000 Hz_10ms_modulation_mae_db": 0.9850065435659149, "300–1000 Hz_40ms_log_rms_mae_db": 0.9478488391240693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05560350028694183, "300–1000 Hz_40ms_envelope_correlation": 0.9975399307068059, "300–1000 Hz_40ms_modulation_mae_db": 1.1635617826238038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8754395898930298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17658626549166065, "1000–3000 Hz_10ms_envelope_correlation": 0.9766555177517902, "1000–3000 Hz_10ms_modulation_mae_db": 0.8599345993771853, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1742373474396992, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09850117169256412, "1000–3000 Hz_40ms_envelope_correlation": 0.9929192351571505, "1000–3000 Hz_40ms_modulation_mae_db": 0.46607812458300685, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8755594703797407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2082083181032362, "3000–6000 Hz_10ms_envelope_correlation": 0.9746455072593674, "3000–6000 Hz_10ms_modulation_mae_db": 0.37945190248238053, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3706588173024137, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15407230923570994, "3000–6000 Hz_40ms_envelope_correlation": 0.9858304166033423, "3000–6000 Hz_40ms_modulation_mae_db": 0.13997679857419723, "mr_stft_db": 5.693191360398477, "band_mae_db": 0.1531268948398498, "rms_gain_error_db": -0.015501994510771253, "output_peak": 0.4729876849864608}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0602326310918473, "全帯域_10ms_rms_envelope_nrmse": 0.08297664841362182, "全帯域_10ms_envelope_correlation": 0.9930178208408293, "全帯域_10ms_modulation_mae_db": 0.25618464313780476, "全帯域_40ms_log_rms_mae_db": 0.6005181781216822, "全帯域_40ms_rms_envelope_nrmse": 0.04609508225889189, "全帯域_40ms_envelope_correlation": 0.9978201489888211, "全帯域_40ms_modulation_mae_db": 0.1796998175944445, "300–1000 Hz_10ms_log_rms_mae_db": 1.550381942126069, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10704449905254448, "300–1000 Hz_10ms_envelope_correlation": 0.9915444585264633, "300–1000 Hz_10ms_modulation_mae_db": 0.6445983678769209, "300–1000 Hz_40ms_log_rms_mae_db": 0.924813586406002, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05884344714748098, "300–1000 Hz_40ms_envelope_correlation": 0.9975612566700213, "300–1000 Hz_40ms_modulation_mae_db": 0.5568850952772079, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7751964770435191, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1647227262524767, "1000–3000 Hz_10ms_envelope_correlation": 0.98006028276517, "1000–3000 Hz_10ms_modulation_mae_db": 0.2915689346057989, "1000–3000 Hz_40ms_log_rms_mae_db": 1.166100253076022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09352442218101988, "1000–3000 Hz_40ms_envelope_correlation": 0.9941448987977507, "1000–3000 Hz_40ms_modulation_mae_db": 0.8235946157121664, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8949568215852215, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21044043417640576, "3000–6000 Hz_10ms_envelope_correlation": 0.9725648015915732, "3000–6000 Hz_10ms_modulation_mae_db": 0.9759833844183676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.376064364451842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16310081000874097, "3000–6000 Hz_40ms_envelope_correlation": 0.9831941299529826, "3000–6000 Hz_40ms_modulation_mae_db": 0.9877329759968264, "mr_stft_db": 5.68395699228108, "band_mae_db": 0.16493777172804372, "rms_gain_error_db": -0.11624981657627388, "output_peak": 0.3779629744827314}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.865341007859445, "spectral_convergence": 0.2511212384008162, "spectral_convergence_status": "ok", "amplitude_floor": 1.597965137573199e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6058916789374935, "spectral_convergence": 0.23130916788970907, "spectral_convergence_status": "ok", "amplitude_floor": 1.057274363712373e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5960053999113955, "spectral_convergence": 0.23815003832618004, "spectral_convergence_status": "ok", "amplitude_floor": 9.008040625007043e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007434014661781762, "output_power": 0.0007480138296383804, "delta_db": 0.026862140827775858, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002448126060340392, "output_power": 0.0023268593013818606, "delta_db": -0.22063653476607412, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00039049712732524606, "output_power": 0.0003872502005489783, "delta_db": -0.036261920134713604, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012090645510899839, "output_power": 0.00011531615783683984, "delta_db": -0.20559324241014065, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.957546973001399e-05, "output_power": 6.101756141736472e-05, "delta_db": 0.10387371682343713, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.142800821939945e-05, "output_power": 9.183560441072356e-05, "delta_db": 0.5223727681762778, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18593338718973654, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13033083421521555, "pitch": null, "output_rms_dbfs": -24.282570948488235, "output_peak": 0.4213873610883026, "reference_rms_dbfs": -24.152240114273017, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.689079362236111}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019565573818943504, "log_rms_mae_db": 1.0923800722549886, "rms_envelope_nrmse": 0.07650110134622684, "envelope_correlation": 0.9941274609590961, "rms_gain_error_db": -0.13030527720860102, "reference_mean_power": 0.0038470996951946897, "output_mean_power": 0.0037333861602499518}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.743468430119212e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007829242029289011, "output_power": 0.0007367206585559737, "reference_active": true, "delta_db": -0.26416870656976144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011656588263872551, "output_power": 0.00011114438861857611, "reference_active": true, "delta_db": -0.206839152855674}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.15404555717803e-05, "output_power": 2.9135853525673622e-05, "reference_active": true, "delta_db": -0.34440216874611074}], "active_band_count": 3, "active_band_mae_db": 0.2718033427238487}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018472557440126405, "log_rms_mae_db": 0.6353917374411323, "rms_envelope_nrmse": 0.043960868752885136, "envelope_correlation": 0.9980690751421075, "rms_gain_error_db": -0.13008744785032242, "reference_mean_power": 0.0038659686805529734, "output_mean_power": 0.0037518855902637363}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.614770154461247e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007106100759473788, "output_power": 0.000669273109544057, "reference_active": true, "delta_db": -0.26027985174839235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.7305685337334634e-05, "output_power": 5.277943885774101e-05, "reference_active": true, "delta_db": -0.35732942204247753}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9649709160477995e-06, "output_power": 2.2750623330056413e-06, "reference_active": true, "delta_db": 0.6363717344412595}], "active_band_count": 3, "active_band_mae_db": 0.41799366941070976}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016965929826439602, "log_rms_mae_db": 1.6018453625136506, "rms_envelope_nrmse": 0.09217693656286025, "envelope_correlation": 0.993926765049457, "rms_gain_error_db": -0.2194315232444916, "reference_mean_power": 0.0021633278035333714, "output_mean_power": 0.002056738982830616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3458386409803657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007177587670080827, "output_power": 0.0006537849220293764, "reference_active": true, "delta_db": -0.4054360545052737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.048912255781678e-05, "output_power": 7.923215018972942e-05, "reference_active": true, "delta_db": -0.5769493512561016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.109541579796594e-05, "output_power": 2.249904244440125e-05, "reference_active": true, "delta_db": -1.4053233343349645}], "active_band_count": 3, "active_band_mae_db": 0.7959029133654466}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015157137475553504, "log_rms_mae_db": 0.9808505873315507, "rms_envelope_nrmse": 0.05776171384334925, "envelope_correlation": 0.9977666033500124, "rms_gain_error_db": -0.21945764729570166, "reference_mean_power": 0.002174161455071234, "output_mean_power": 0.0020670264182544545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2731603714689836e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006867133707128568, "output_power": 0.0006255121318617322, "reference_active": true, "delta_db": -0.4053976615524467}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.067232594302272e-05, "output_power": 3.4122868782942014e-05, "reference_active": true, "delta_db": -0.7625347316926339}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.08557261679534e-06, "output_power": 1.6972040631054533e-06, "reference_active": true, "delta_db": -0.8949125319037807}], "active_band_count": 3, "active_band_mae_db": 0.6876149750496204}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.465903254113391e-05, "log_rms_mae_db": 1.7557557368978602, "rms_envelope_nrmse": 0.1655333801617483, "envelope_correlation": 0.9793469343087413, "rms_gain_error_db": -0.09392776788109376, "reference_mean_power": 0.00014077271984820836, "output_mean_power": 0.00013776082179322332}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.622290255773331e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.707421130487826e-05, "output_power": 3.656779439814706e-05, "reference_active": true, "delta_db": -0.059731543972926474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.08772682349557e-06, "output_power": 4.87555834322804e-06, "reference_active": true, "delta_db": -0.18499426813625236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8011808387082025e-06, "output_power": 2.8988141829491124e-06, "reference_active": true, "delta_db": 2.066630593010143}], "active_band_count": 3, "active_band_mae_db": 0.770452135039774}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.107059424790359e-05, "log_rms_mae_db": 1.0806323420689727, "rms_envelope_nrmse": 0.08395984349312088, "envelope_correlation": 0.9944831449326794, "rms_gain_error_db": -0.09433676100644502, "reference_mean_power": 0.00014139647469838974, "output_mean_power": 0.00013835820073807416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.069193023932203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.382591725297068e-05, "output_power": 3.389116411672047e-05, "reference_active": true, "delta_db": 0.008369043994155554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1244897853414024e-06, "output_power": 2.0790820222623845e-06, "reference_active": true, "delta_db": -0.09383024215759324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.400225448356044e-07, "output_power": 1.2809866061194565e-07, "reference_active": true, "delta_db": -0.38653377635434194}], "active_band_count": 3, "active_band_mae_db": 0.16291102083536357}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.428054041295646e-05, "log_rms_mae_db": 1.8938815592008198, "rms_envelope_nrmse": 0.2128397105963751, "envelope_correlation": 0.9738175367038501, "rms_gain_error_db": 0.30856743727483366, "reference_mean_power": 8.952225188868405e-05, "output_mean_power": 9.611424124752708e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.314971416941356e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.877588546016404e-05, "output_power": 6.027378814420977e-05, "reference_active": true, "delta_db": 0.10929306773883218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.04956306951581e-06, "output_power": 8.888156792279923e-06, "reference_active": true, "delta_db": -0.0781590392132345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5294615410314917e-06, "output_power": 3.152670021994038e-06, "reference_active": true, "delta_db": 0.9565043656769774}], "active_band_count": 3, "active_band_mae_db": 0.3813188242096814}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5452516433588795e-05, "log_rms_mae_db": 1.3721657923482524, "rms_envelope_nrmse": 0.1532539466276058, "envelope_correlation": 0.9862459167732353, "rms_gain_error_db": 0.3085637509432118, "reference_mean_power": 8.996899751083782e-05, "output_mean_power": 9.659380108535176e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.951283845645369e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.884212028117595e-05, "output_power": 6.0742987030274986e-05, "reference_active": true, "delta_db": 0.13807831054131067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.436071463262977e-06, "output_power": 4.43573626958768e-06, "reference_active": true, "delta_db": -0.00032816933820713787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7711408251558852e-07, "output_power": 1.8319789785687753e-07, "reference_active": true, "delta_db": 0.14667392210263594}], "active_band_count": 3, "active_band_mae_db": 0.09502680066071793}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8958272150770648, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6245f95bfbeb68d504124c6b6d843f5617d1716bdde77475c2340fc1441671ef", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9798267733302621, "standard_deviation": 0.0029693280079535083, "minimum": 0.976346545085969, "maximum": 0.9842548685109782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8036133169219535, "standard_deviation": 0.03774960666912293, "minimum": 1.7557557368978602, "maximum": 1.8754395898930298, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5790219210959748, "standard_deviation": 0.19004277476776937, "minimum": 0.2915689346057989, "maximum": 0.8599345993771853, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16395088022647847, "standard_deviation": 0.011962688348250547, "minimum": 0.1450371029750051, "maximum": 0.1771410280908254, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994032402529415, "standard_deviation": 0.001730418861440807, "minimum": 0.9917478341301387, "maximum": 0.9968620771112063, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1320954651389354, "standard_deviation": 0.05227859030666246, "minimum": 1.039906667779025, "maximum": 1.1742373474396992, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6015386004091015, "standard_deviation": 0.32965268440218365, "minimum": 0.16291102083536357, "maximum": 1.1819350961704707, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08951318076437237, "standard_deviation": 0.01186499893896183, "minimum": 0.07032893211890182, "maximum": 0.10359147655556687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9735898813811584, "standard_deviation": 0.0016480063992297491, "minimum": 0.9707537315421738, "maximum": 0.9759845102367702, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9004254448023845, "standard_deviation": 0.015736888382889024, "minimum": 1.8755594703797407, "maximum": 1.927891386204417, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.48615374377232645, "standard_deviation": 0.22077281839788976, "minimum": 0.3015410708675013, "maximum": 0.9759833844183676, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20866401691800796, "standard_deviation": 0.005980507829665344, "minimum": 0.19795000478511793, "maximum": 0.21687972453739693, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9843721394721214, "standard_deviation": 0.0020131122179727215, "minimum": 0.9799266194370702, "maximum": 0.9862459167732353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.387136077795072, "standard_deviation": 0.03313991603958827, "minimum": 1.3501999852549782, "maximum": 1.446416602471589, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.37550203134300997, "standard_deviation": 0.27916036975517305, "minimum": 0.09502680066071793, "maximum": 0.9877329759968264, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15793090614124938, "standard_deviation": 0.008466727896939804, "minimum": 0.15179142787950234, "maximum": 0.17709883190054404, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9919761407846802, "standard_deviation": 0.0011234128287572456, "minimum": 0.989911578399995, "maximum": 0.993926765049457, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6099881606404298, "standard_deviation": 0.036350628426762545, "minimum": 1.550381942126069, "maximum": 1.6734365443154122, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5388825243628104, "standard_deviation": 0.26737458882540965, "minimum": 0.20411703469583187, "maximum": 0.9850065435659149, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10363429133128838, "standard_deviation": 0.006873980355974849, "minimum": 0.09217693656286025, "maximum": 0.11614410310158887, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974262949382063, "standard_deviation": 0.00042755326179507613, "minimum": 0.9965083903769957, "maximum": 0.9977953057367064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9560866398232446, "standard_deviation": 0.03897619109338197, "minimum": 0.8877196186161849, "maximum": 1.0182457711047503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5995216140685045, "standard_deviation": 0.24566059909566146, "minimum": 0.38339542553796324, "maximum": 1.1635617826238038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.058696589749598696, "standard_deviation": 0.004778891028070227, "minimum": 0.05447504959792102, "maximum": 0.06813900457837958, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1605354002188283, "standard_deviation": 0.01717323655957954, "minimum": 0.13632497458922693, "maximum": 0.18593338718973654, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.692499062048852, "standard_deviation": 0.008611125388016621, "minimum": 5.677227527273334, "maximum": 5.7015098955337535, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40770425873880745, "standard_deviation": 0.029979028189499147, "minimum": 0.3779629744827314, "maximum": 0.4729876849864608, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07209398170400769, "standard_deviation": 0.041492983837544795, "minimum": -0.13033083421521555, "maximum": -0.015501994510771253, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9935371412151486, "standard_deviation": 0.0009243690020027168, "minimum": 0.9919329206349894, "maximum": 0.9946708244587465, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0542871257167756, "standard_deviation": 0.031030223530711165, "minimum": 0.9954662202401671, "maximum": 1.0923800722549886, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2828653479880715, "standard_deviation": 0.1731535802130217, "minimum": 0.09494506901783133, "maximum": 0.6554595321059291, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07938838579251162, "standard_deviation": 0.005606525525948443, "minimum": 0.07213915210642247, "maximum": 0.08880961326547926, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979965979671062, "standard_deviation": 0.0005418494735512101, "minimum": 0.997121076273848, "maximum": 0.9986622783152268, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.574907337146888, "standard_deviation": 0.03636593609841328, "minimum": 0.5296188093600518, "maximum": 0.6353917374411323, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3341780300413739, "standard_deviation": 0.17540589920160687, "minimum": 0.14725094623403215, "maximum": 0.6172421346674191, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04295150668529139, "standard_deviation": 0.005883248866493487, "minimum": 0.03532039678236768, "maximum": 0.05180286071399013, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8224-274384-0010", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.918491840362549, "predicted_mos": 2.918491840362549, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1b1b43bd2c94a558226812d04b88b83b854374992f12d87b9bdcac1dde826862", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.586174599826336, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0615904673516061, "全帯域_10ms_rms_envelope_nrmse": 0.07703963162674593, "全帯域_10ms_envelope_correlation": 0.9939465166024871, "全帯域_10ms_modulation_mae_db": 0.16986702281653962, "全帯域_40ms_log_rms_mae_db": 0.5498656363245296, "全帯域_40ms_rms_envelope_nrmse": 0.03849630315842125, "全帯域_40ms_envelope_correlation": 0.9984476915826405, "全帯域_40ms_modulation_mae_db": 0.121470555793893, "300–1000 Hz_10ms_log_rms_mae_db": 1.5632434271781803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09489176763360542, "300–1000 Hz_10ms_envelope_correlation": 0.99334617332738, "300–1000 Hz_10ms_modulation_mae_db": 0.4818568510954369, "300–1000 Hz_40ms_log_rms_mae_db": 0.9470113868969282, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05177646247342592, "300–1000 Hz_40ms_envelope_correlation": 0.998083661582514, "300–1000 Hz_40ms_modulation_mae_db": 0.6069331011696152, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7163186340025687, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16362047504949043, "1000–3000 Hz_10ms_envelope_correlation": 0.9804984182350381, "1000–3000 Hz_10ms_modulation_mae_db": 0.7127189221369213, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0502286789239037, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08688039277219808, "1000–3000 Hz_40ms_envelope_correlation": 0.995315027353779, "1000–3000 Hz_40ms_modulation_mae_db": 0.9156476642524355, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8412070137345622, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2064476655367473, "3000–6000 Hz_10ms_envelope_correlation": 0.9736813087998811, "3000–6000 Hz_10ms_modulation_mae_db": 1.1341238004858143, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3398097334517882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16211608221826462, "3000–6000 Hz_40ms_envelope_correlation": 0.983536824212942, "3000–6000 Hz_40ms_modulation_mae_db": 1.103290412645081, "mr_stft_db": 5.644983004948322, "band_mae_db": 0.23793596097326766, "rms_gain_error_db": -0.08890561563107943, "output_peak": 0.3774462619266241}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.053945034932445, "全帯域_10ms_rms_envelope_nrmse": 0.07267395492274642, "全帯域_10ms_envelope_correlation": 0.9945950281355156, "全帯域_10ms_modulation_mae_db": 0.21155034505372936, "全帯域_40ms_log_rms_mae_db": 0.5879942308265205, "全帯域_40ms_rms_envelope_nrmse": 0.04111710890842241, "全帯域_40ms_envelope_correlation": 0.9981709666623642, "全帯域_40ms_modulation_mae_db": 0.3415916954888298, "300–1000 Hz_10ms_log_rms_mae_db": 1.604801725666875, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09792570408947955, "300–1000 Hz_10ms_envelope_correlation": 0.9928687887101972, "300–1000 Hz_10ms_modulation_mae_db": 0.4197310699852919, "300–1000 Hz_40ms_log_rms_mae_db": 0.9323540704766486, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05603524266624337, "300–1000 Hz_40ms_envelope_correlation": 0.997663587022803, "300–1000 Hz_40ms_modulation_mae_db": 0.496378491720806, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7037501730970006, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16574484560253844, "1000–3000 Hz_10ms_envelope_correlation": 0.9793987678690228, "1000–3000 Hz_10ms_modulation_mae_db": 0.3422389964384343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0412226494713681, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0954959479428471, "1000–3000 Hz_40ms_envelope_correlation": 0.993176054710218, "1000–3000 Hz_40ms_modulation_mae_db": 0.6409560686661798, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9075866739872833, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20189989649914, "3000–6000 Hz_10ms_envelope_correlation": 0.9752001387040621, "3000–6000 Hz_10ms_modulation_mae_db": 0.270484363352572, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3807678869909166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14867322891563303, "3000–6000 Hz_40ms_envelope_correlation": 0.986041416321454, "3000–6000 Hz_40ms_modulation_mae_db": 0.2880039041746269, "mr_stft_db": 5.648081087450141, "band_mae_db": 0.16025909211114628, "rms_gain_error_db": -0.06880380616008858, "output_peak": 0.40694801910211886}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.086936484955781, "全帯域_10ms_rms_envelope_nrmse": 0.07833022303831044, "全帯域_10ms_envelope_correlation": 0.9937235370122132, "全帯域_10ms_modulation_mae_db": 0.4263912652258464, "全帯域_40ms_log_rms_mae_db": 0.5720611073298275, "全帯域_40ms_rms_envelope_nrmse": 0.04468799812439363, "全帯域_40ms_envelope_correlation": 0.9978402573451849, "全帯域_40ms_modulation_mae_db": 0.6763243418905627, "300–1000 Hz_10ms_log_rms_mae_db": 1.5981606205635372, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09838073588214562, "300–1000 Hz_10ms_envelope_correlation": 0.9927710953341705, "300–1000 Hz_10ms_modulation_mae_db": 0.5967897513736478, "300–1000 Hz_40ms_log_rms_mae_db": 0.941294414494636, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05834570979205974, "300–1000 Hz_40ms_envelope_correlation": 0.997420819693951, "300–1000 Hz_40ms_modulation_mae_db": 0.9449236681051243, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7233587994064483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16411569380475005, "1000–3000 Hz_10ms_envelope_correlation": 0.9814383459612628, "1000–3000 Hz_10ms_modulation_mae_db": 0.772136938383032, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0838130703359006, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11020125220053986, "1000–3000 Hz_40ms_envelope_correlation": 0.9929425344151938, "1000–3000 Hz_40ms_modulation_mae_db": 1.050837466518572, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8183676484179847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2070365376894342, "3000–6000 Hz_10ms_envelope_correlation": 0.9739333745011974, "3000–6000 Hz_10ms_modulation_mae_db": 0.6896744319183893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3066351322306262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14613003642932432, "3000–6000 Hz_40ms_envelope_correlation": 0.9865102507950527, "3000–6000 Hz_40ms_modulation_mae_db": 0.5406442803966712, "mr_stft_db": 5.641142905701663, "band_mae_db": 0.19459952972885533, "rms_gain_error_db": -0.07539114030688093, "output_peak": 0.4451389970805937}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0836269864111345, "全帯域_10ms_rms_envelope_nrmse": 0.08574541740872009, "全帯域_10ms_envelope_correlation": 0.9924614731827028, "全帯域_10ms_modulation_mae_db": 0.24247123582923477, "全帯域_40ms_log_rms_mae_db": 0.6009555001265696, "全帯域_40ms_rms_envelope_nrmse": 0.047406046088586846, "全帯域_40ms_envelope_correlation": 0.9975342106510413, "全帯域_40ms_modulation_mae_db": 0.39807824311835344, "300–1000 Hz_10ms_log_rms_mae_db": 1.5667905577844476, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10643378305756215, "300–1000 Hz_10ms_envelope_correlation": 0.9914485626051871, "300–1000 Hz_10ms_modulation_mae_db": 0.6331448321655787, "300–1000 Hz_40ms_log_rms_mae_db": 0.9980739245097427, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0630479089577264, "300–1000 Hz_40ms_envelope_correlation": 0.996862703218198, "300–1000 Hz_40ms_modulation_mae_db": 0.8652404306567304, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7846614868590152, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16719010975598123, "1000–3000 Hz_10ms_envelope_correlation": 0.9795179251658157, "1000–3000 Hz_10ms_modulation_mae_db": 0.46582230683085596, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0986201717982569, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09775676629381731, "1000–3000 Hz_40ms_envelope_correlation": 0.9935554479948101, "1000–3000 Hz_40ms_modulation_mae_db": 0.8930375808829342, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8211270343829289, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22044968538157367, "3000–6000 Hz_10ms_envelope_correlation": 0.9697544095734032, "3000–6000 Hz_10ms_modulation_mae_db": 0.4962029824942973, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2992650882621588, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16093055872647233, "3000–6000 Hz_40ms_envelope_correlation": 0.9835568311955007, "3000–6000 Hz_40ms_modulation_mae_db": 0.6923264107413177, "mr_stft_db": 5.6432965937762605, "band_mae_db": 0.1278163766121985, "rms_gain_error_db": -0.051096856496246845, "output_peak": 0.39300945747900995}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0365083307289833, "全帯域_10ms_rms_envelope_nrmse": 0.07873006207986359, "全帯域_10ms_envelope_correlation": 0.9937869366089249, "全帯域_10ms_modulation_mae_db": 0.8369828072620855, "全帯域_40ms_log_rms_mae_db": 0.5297680894923247, "全帯域_40ms_rms_envelope_nrmse": 0.04583247304974675, "全帯域_40ms_envelope_correlation": 0.9979017197846259, "全帯域_40ms_modulation_mae_db": 1.2074576058958486, "300–1000 Hz_10ms_log_rms_mae_db": 1.5132416211492319, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11380408258152329, "300–1000 Hz_10ms_envelope_correlation": 0.9906637203405282, "300–1000 Hz_10ms_modulation_mae_db": 1.1175908708989999, "300–1000 Hz_40ms_log_rms_mae_db": 0.9048972683019751, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06950373712413416, "300–1000 Hz_40ms_envelope_correlation": 0.9967810599413346, "300–1000 Hz_40ms_modulation_mae_db": 1.538476995864268, "1000–3000 Hz_10ms_log_rms_mae_db": 1.799251830693963, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17349834187778598, "1000–3000 Hz_10ms_envelope_correlation": 0.9773845165776861, "1000–3000 Hz_10ms_modulation_mae_db": 0.6568635738986175, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0836759548724162, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09441409329668694, "1000–3000 Hz_40ms_envelope_correlation": 0.9929764725914487, "1000–3000 Hz_40ms_modulation_mae_db": 0.7593809935881327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8148947086518519, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19820741005508433, "3000–6000 Hz_10ms_envelope_correlation": 0.9757729325510305, "3000–6000 Hz_10ms_modulation_mae_db": 1.193175913086372, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3203911694171762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1579053742214881, "3000–6000 Hz_40ms_envelope_correlation": 0.9843406405313344, "3000–6000 Hz_40ms_modulation_mae_db": 1.276089186722038, "mr_stft_db": 5.649240552892465, "band_mae_db": 0.1471628375893587, "rms_gain_error_db": -0.13763999256425682, "output_peak": 0.39500509653401006}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0825054961748284, "全帯域_10ms_rms_envelope_nrmse": 0.08599295896605345, "全帯域_10ms_envelope_correlation": 0.9924127313137304, "全帯域_10ms_modulation_mae_db": 0.3649048563984784, "全帯域_40ms_log_rms_mae_db": 0.5608587173802004, "全帯域_40ms_rms_envelope_nrmse": 0.04407689869710424, "全帯域_40ms_envelope_correlation": 0.9978781502997035, "全帯域_40ms_modulation_mae_db": 0.1834939015842887, "300–1000 Hz_10ms_log_rms_mae_db": 1.5487228188818332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11200997976040925, "300–1000 Hz_10ms_envelope_correlation": 0.9905182051535779, "300–1000 Hz_10ms_modulation_mae_db": 0.21171527271391957, "300–1000 Hz_40ms_log_rms_mae_db": 0.8734054181773501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06365085858283698, "300–1000 Hz_40ms_envelope_correlation": 0.9968054992358313, "300–1000 Hz_40ms_modulation_mae_db": 0.13483490154423441, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7559906312911306, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15720761300939762, "1000–3000 Hz_10ms_envelope_correlation": 0.9815210956866988, "1000–3000 Hz_10ms_modulation_mae_db": 0.48992233947082003, "1000–3000 Hz_40ms_log_rms_mae_db": 1.121174586012243, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09408056353486337, "1000–3000 Hz_40ms_envelope_correlation": 0.993022084474473, "1000–3000 Hz_40ms_modulation_mae_db": 0.704937593019708, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8733462331974602, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2040559284177324, "3000–6000 Hz_10ms_envelope_correlation": 0.9743293805199779, "3000–6000 Hz_10ms_modulation_mae_db": 1.2075981003784166, "3000–6000 Hz_40ms_log_rms_mae_db": 1.371679764491917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15223516032058898, "3000–6000 Hz_40ms_envelope_correlation": 0.9851626150238462, "3000–6000 Hz_40ms_modulation_mae_db": 1.2396020917813322, "mr_stft_db": 5.662148215556773, "band_mae_db": 0.11724288803538079, "rms_gain_error_db": -0.05381599107923774, "output_peak": 0.44275272833844315}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.089215397874364, "全帯域_10ms_rms_envelope_nrmse": 0.08026648490755088, "全帯域_10ms_envelope_correlation": 0.9934712940821896, "全帯域_10ms_modulation_mae_db": 0.2690461145406943, "全帯域_40ms_log_rms_mae_db": 0.5771528151461802, "全帯域_40ms_rms_envelope_nrmse": 0.0435461717720307, "全帯域_40ms_envelope_correlation": 0.9980443013977754, "全帯域_40ms_modulation_mae_db": 0.4376280561013656, "300–1000 Hz_10ms_log_rms_mae_db": 1.6080785593735911, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09898546052472763, "300–1000 Hz_10ms_envelope_correlation": 0.9927740542543363, "300–1000 Hz_10ms_modulation_mae_db": 0.3933666301371915, "300–1000 Hz_40ms_log_rms_mae_db": 0.9106389376577897, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05581645116648368, "300–1000 Hz_40ms_envelope_correlation": 0.9977817619700308, "300–1000 Hz_40ms_modulation_mae_db": 0.6268889255937603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7627851168704076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16804383893984606, "1000–3000 Hz_10ms_envelope_correlation": 0.9792955400248143, "1000–3000 Hz_10ms_modulation_mae_db": 0.49246755869082454, "1000–3000 Hz_40ms_log_rms_mae_db": 1.076845817961731, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10444303520764438, "1000–3000 Hz_40ms_envelope_correlation": 0.9924676783369072, "1000–3000 Hz_40ms_modulation_mae_db": 0.7844704029792059, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8023355042954703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19837215990638224, "3000–6000 Hz_10ms_envelope_correlation": 0.9756299506140822, "3000–6000 Hz_10ms_modulation_mae_db": 0.9309109664062744, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2951974101063166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15319935630609863, "3000–6000 Hz_40ms_envelope_correlation": 0.9851199298758593, "3000–6000 Hz_40ms_modulation_mae_db": 0.8107910143007183, "mr_stft_db": 5.655315358727286, "band_mae_db": 0.17123977306958407, "rms_gain_error_db": -0.11022768476109554, "output_peak": 0.38760425398155024}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0816475236632042, "全帯域_10ms_rms_envelope_nrmse": 0.086457259295996, "全帯域_10ms_envelope_correlation": 0.9923775514044703, "全帯域_10ms_modulation_mae_db": 0.302809024406719, "全帯域_40ms_log_rms_mae_db": 0.6081180882199111, "全帯域_40ms_rms_envelope_nrmse": 0.04450830402801168, "全帯域_40ms_envelope_correlation": 0.9979534413913962, "全帯域_40ms_modulation_mae_db": 0.4710207756747833, "300–1000 Hz_10ms_log_rms_mae_db": 1.5742870848917703, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11166509979530359, "300–1000 Hz_10ms_envelope_correlation": 0.9907900310445165, "300–1000 Hz_10ms_modulation_mae_db": 0.259918942270019, "300–1000 Hz_40ms_log_rms_mae_db": 0.9689651139777823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06755391279180288, "300–1000 Hz_40ms_envelope_correlation": 0.9966976076192253, "300–1000 Hz_40ms_modulation_mae_db": 0.2878847531855343, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7374522357408018, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1509151193308412, "1000–3000 Hz_10ms_envelope_correlation": 0.9829069376587038, "1000–3000 Hz_10ms_modulation_mae_db": 0.7076143778782028, "1000–3000 Hz_40ms_log_rms_mae_db": 1.088735566960167, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08371433833681918, "1000–3000 Hz_40ms_envelope_correlation": 0.9944745785273378, "1000–3000 Hz_40ms_modulation_mae_db": 0.4384691414631017, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8635238383249835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2229385431465627, "3000–6000 Hz_10ms_envelope_correlation": 0.9708540566225395, "3000–6000 Hz_10ms_modulation_mae_db": 0.3636907035257824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3728215119278468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1532596839287523, "3000–6000 Hz_40ms_envelope_correlation": 0.9859930968198144, "3000–6000 Hz_40ms_modulation_mae_db": 0.5681076487378928, "mr_stft_db": 5.645610916322146, "band_mae_db": 0.1629627809891491, "rms_gain_error_db": -0.10751073419087731, "output_peak": 0.4319770306547064}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.813446175516483, "spectral_convergence": 0.25051061056982477, "spectral_convergence_status": "ok", "amplitude_floor": 1.597965137573199e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.57561390783795, "spectral_convergence": 0.23130091953682494, "spectral_convergence_status": "ok", "amplitude_floor": 1.057274363712373e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.545888931490531, "spectral_convergence": 0.24115327584794233, "spectral_convergence_status": "ok", "amplitude_floor": 9.008040625007043e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007434014661781762, "output_power": 0.0007772831328491594, "delta_db": 0.19355830257225984, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002448126060340392, "output_power": 0.0023454637165693055, "delta_db": -0.18605058030174693, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00039049712732524606, "output_power": 0.00039921165630103, "delta_db": 0.09585370103276458, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012090645510899839, "output_power": 0.0001115815210008334, "delta_db": -0.3485721103923983, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.957546973001399e-05, "output_power": 5.3302713140868975e-05, "delta_db": -0.4831815983416748, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.142800821939945e-05, "output_power": 7.92015834698193e-05, "delta_db": -0.12039947319876161, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23793596097326766, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08890561563107943, "pitch": null, "output_rms_dbfs": -24.241145729904098, "output_peak": 0.3774462619266241, "reference_rms_dbfs": -24.152240114273017, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.644983004948322}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019565573818943504, "log_rms_mae_db": 1.0615904673516061, "rms_envelope_nrmse": 0.07703963162674593, "envelope_correlation": 0.9939465166024871, "rms_gain_error_db": -0.0888712277255976, "reference_mean_power": 0.0038470996951946897, "output_mean_power": 0.0037691751411432117}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.743468430119212e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007829242029289011, "output_power": 0.0007565420411475321, "reference_active": true, "delta_db": -0.14886652000387468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011656588263872551, "output_power": 0.00010947084017360428, "reference_active": true, "delta_db": -0.2727300526568848}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.15404555717803e-05, "output_power": 3.218610409936585e-05, "reference_active": true, "delta_db": 0.08800449578885938}], "active_band_count": 3, "active_band_mae_db": 0.16986702281653962}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018472557440126405, "log_rms_mae_db": 0.5498656363245296, "rms_envelope_nrmse": 0.03849630315842125, "envelope_correlation": 0.9984476915826405, "rms_gain_error_db": -0.0887332400323462, "reference_mean_power": 0.0038659686805529734, "output_mean_power": 0.003787782274292429}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.614770154461247e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007106100759473788, "output_power": 0.0006916624700828367, "reference_active": true, "delta_db": -0.11737149881224067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.7305685337334634e-05, "output_power": 5.422743904135638e-05, "reference_active": true, "delta_db": -0.2397861594620142}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9649709160477995e-06, "output_power": 1.9616915698899047e-06, "reference_active": true, "delta_db": -0.0072540091074241265}], "active_band_count": 3, "active_band_mae_db": 0.121470555793893}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016965929826439602, "log_rms_mae_db": 1.5632434271781803, "rms_envelope_nrmse": 0.09489176763360542, "envelope_correlation": 0.99334617332738, "rms_gain_error_db": -0.16331723346688387, "reference_mean_power": 0.0021633278035333714, "output_mean_power": 0.002083486105800213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3458386409803657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007177587670080827, "output_power": 0.0006799631627444519, "reference_active": true, "delta_db": -0.23493120577247006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.048912255781678e-05, "output_power": 7.85779039956147e-05, "reference_active": true, "delta_db": -0.6129593686475611}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.109541579796594e-05, "output_power": 2.709744668968191e-05, "reference_active": true, "delta_db": -0.5976799788662797}], "active_band_count": 3, "active_band_mae_db": 0.4818568510954369}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015157137475553504, "log_rms_mae_db": 0.9470113868969282, "rms_envelope_nrmse": 0.05177646247342592, "envelope_correlation": 0.998083661582514, "rms_gain_error_db": -0.16333337178785587, "reference_mean_power": 0.002174161455071234, "output_mean_power": 0.002093912140008138}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2731603714689836e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006867133707128568, "output_power": 0.0006485287134393052, "reference_active": true, "delta_db": -0.24846294324826687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.067232594302272e-05, "output_power": 3.488506648321514e-05, "reference_active": true, "delta_db": -0.6665945428537075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.08557261679534e-06, "output_power": 1.6929772971247305e-06, "reference_active": true, "delta_db": -0.9057418174068712}], "active_band_count": 3, "active_band_mae_db": 0.6069331011696152}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.465903254113391e-05, "log_rms_mae_db": 1.7163186340025687, "rms_envelope_nrmse": 0.16362047504949043, "envelope_correlation": 0.9804984182350381, "rms_gain_error_db": -0.3801405281376049, "reference_mean_power": 0.00014077271984820836, "output_mean_power": 0.00012897467697087634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.622290255773331e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.707421130487826e-05, "output_power": 3.220316921840306e-05, "reference_active": true, "delta_db": -0.6117330674288016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.08772682349557e-06, "output_power": 3.7103973553874823e-06, "reference_active": true, "delta_db": -1.371033628438352}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8011808387082025e-06, "output_power": 1.737874101184769e-06, "reference_active": true, "delta_db": -0.1553900705436106}], "active_band_count": 3, "active_band_mae_db": 0.7127189221369213}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.107059424790359e-05, "log_rms_mae_db": 1.0502286789239037, "rms_envelope_nrmse": 0.08688039277219808, "envelope_correlation": 0.995315027353779, "rms_gain_error_db": -0.3797072248140127, "reference_mean_power": 0.00014139647469838974, "output_mean_power": 0.0001295590811483196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.069193023932203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.382591725297068e-05, "output_power": 2.980509511151993e-05, "reference_active": true, "delta_db": -0.5495907008412182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1244897853414024e-06, "output_power": 1.6812021090383964e-06, "reference_active": true, "delta_db": -1.0163472117133852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.400225448356044e-07, "output_power": 1.0668354990309516e-07, "reference_active": true, "delta_db": -1.1810050802027028}], "active_band_count": 3, "active_band_mae_db": 0.9156476642524355}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.428054041295646e-05, "log_rms_mae_db": 1.8412070137345622, "rms_envelope_nrmse": 0.2064476655367473, "envelope_correlation": 0.9736813087998811, "rms_gain_error_db": -0.3093256004661618, "reference_mean_power": 8.952225188868405e-05, "output_mean_power": 8.336781863243253e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.314971416941356e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.877588546016404e-05, "output_power": 5.01189500356159e-05, "reference_active": true, "delta_db": -0.6919721631599675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.04956306951581e-06, "output_power": 7.66745001616829e-06, "reference_active": true, "delta_db": -0.7197665761656877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5294615410314917e-06, "output_power": 1.599428441096105e-06, "reference_active": true, "delta_db": -1.990632662131788}], "active_band_count": 3, "active_band_mae_db": 1.1341238004858143}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5452516433588795e-05, "log_rms_mae_db": 1.3398097334517882, "rms_envelope_nrmse": 0.16211608221826462, "envelope_correlation": 0.983536824212942, "rms_gain_error_db": -0.3093426868196007, "reference_mean_power": 8.996899751083782e-05, "output_mean_power": 8.378352197038783e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.951283845645369e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.884212028117595e-05, "output_power": 4.989925454805716e-05, "reference_active": true, "delta_db": -0.7159425575483441}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.436071463262977e-06, "output_power": 3.823796024763141e-06, "reference_active": true, "delta_db": -0.6450381646753249}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7711408251558852e-07, "output_power": 1.1307433496694546e-07, "reference_active": true, "delta_db": -1.9488905157115743}], "active_band_count": 3, "active_band_mae_db": 1.103290412645081}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8930547006671253, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "392402d99f69c4eca89b5682c4562be9bb276f37258d33376bf40e0138992905", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9802451933973803, "standard_deviation": 0.0017126092738501638, "minimum": 0.9773845165776861, "maximum": 0.9829069376587038, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.747946113495167, "standard_deviation": 0.03366524625405566, "minimum": 1.7037501730970006, "maximum": 1.799251830693963, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5799731267159636, "standard_deviation": 0.1521592688509067, "minimum": 0.3422389964384343, "maximum": 0.772136938383032, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16379200467132887, "standard_deviation": 0.006938035633658209, "minimum": 0.1509151193308412, "maximum": 0.17349834187778598, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934912348005209, "standard_deviation": 0.0009434335538278048, "minimum": 0.9924676783369072, "maximum": 0.995315027353779, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0805395620419984, "standard_deviation": 0.02547843326140744, "minimum": 1.0412226494713681, "maximum": 1.121174586012243, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7734671139212838, "standard_deviation": 0.18755773321401795, "minimum": 0.4384691414631017, "maximum": 1.050837466518572, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09587329869817704, "standard_deviation": 0.008589664329181085, "minimum": 0.08371433833681918, "maximum": 0.11020125220053986, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9736444439857718, "standard_deviation": 0.0022157067069792434, "minimum": 0.9697544095734032, "maximum": 0.9757729325510305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8427985818740655, "standard_deviation": 0.035946663849324476, "minimum": 1.8023355042954703, "maximum": 1.9075866739872833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7857326577059898, "standard_deviation": 0.382302875302643, "minimum": 0.270484363352572, "maximum": 1.2075981003784166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2074259783290821, "standard_deviation": 0.009412862160643297, "minimum": 0.19820741005508433, "maximum": 0.2229385431465627, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9850327005969755, "standard_deviation": 0.0011365545296370232, "minimum": 0.983536824212942, "maximum": 0.9865102507950527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3358209621098434, "standard_deviation": 0.0353907308418112, "minimum": 1.2951974101063166, "maximum": 1.3807678869909166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8148568686874598, "standard_deviation": 0.35956713592393, "minimum": 0.2880039041746269, "maximum": 1.276089186722038, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15430618513332778, "standard_deviation": 0.00564446583643249, "minimum": 0.14613003642932432, "maximum": 0.16211608221826462, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918975788462367, "standard_deviation": 0.0011604353296610311, "minimum": 0.9905182051535779, "maximum": 0.99334617332738, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5721658019361833, "standard_deviation": 0.03199373746370017, "minimum": 1.5132416211492319, "maximum": 1.6080785593735911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5142642775800106, "standard_deviation": 0.284297750861777, "minimum": 0.21171527271391957, "maximum": 1.1175908708989999, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10426207666559456, "standard_deviation": 0.007568601450972247, "minimum": 0.09489176763360542, "maximum": 0.11380408258152329, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972620875354861, "standard_deviation": 0.0005410900373439531, "minimum": 0.9966976076192253, "maximum": 0.998083661582514, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9345800668116065, "standard_deviation": 0.038941275394798414, "minimum": 0.8734054181773501, "maximum": 0.9980739245097427, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6876951584800091, "standard_deviation": 0.43651924575125284, "minimum": 0.13483490154423441, "maximum": 1.538476995864268, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06071628544433914, "standard_deviation": 0.006204620900082076, "minimum": 0.05177646247342592, "maximum": 0.06950373712413416, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16490240488861754, "standard_deviation": 0.038252543437372694, "minimum": 0.11724288803538079, "maximum": 0.23793596097326766, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.648727329421882, "standard_deviation": 0.006917387109753723, "minimum": 5.641142905701663, "maximum": 5.662148215556773, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4099852306371321, "standard_deviation": 0.026388942531600852, "minimum": 0.3774462619266241, "maximum": 0.4451389970805937, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0866739776487204, "standard_deviation": 0.03019812365221498, "minimum": -0.13763999256425682, "maximum": -0.051096856496246845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9933468835427792, "standard_deviation": 0.0008336335214996187, "minimum": 0.9923775514044703, "maximum": 0.9945950281355156, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0719969652615433, "standard_deviation": 0.019094105784762068, "minimum": 1.0365083307289833, "maximum": 1.089215397874364, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35300283394166593, "standard_deviation": 0.21216471024677674, "minimum": 0.16986702281653962, "maximum": 0.8369828072620855, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08065449903074835, "standard_deviation": 0.004987144232108953, "minimum": 0.07267395492274642, "maximum": 0.086457259295996, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979713423893415, "standard_deviation": 0.0002657069774514941, "minimum": 0.9975342106510413, "maximum": 0.9984476915826405, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.573346773105758, "standard_deviation": 0.02620902259714315, "minimum": 0.5297680894923247, "maximum": 0.6081180882199111, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4796331469434906, "standard_deviation": 0.3405751456440106, "minimum": 0.121470555793893, "maximum": 1.2074576058958486, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04370891297833969, "standard_deviation": 0.0027729968844007738, "minimum": 0.03849630315842125, "maximum": 0.047406046088586846, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8230-279154-0018", "language": "en", "style": "normal", "speaker": "librispeech_8230", "source_utterance_id": "8230-279154-0018", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8230-279154-0018.wav", "source_file_sha256": "c9f2ef709fe2ff35c99ad4c1d5a09545ffe87d2a1626fce0e2b46479784f3535", "decoded_pcm_sha256": "2becab040601a06405a880e563913b87f3eb167e942e97f7225c55643949f142", "original_source_sha256": "6fb6d05dd68dfe641a0f58b800e9d2e86662dab952355bc7573505bc35efb2fc", "corpus_entry_id": "librispeech_8230-279154-0018"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.5863499641418457, "predicted_mos": 3.5863499641418457, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6ed97aea1b2ea4e52b1a3c908f5d690233bad4edd48c4b417c1ed84cd12b1074", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.302755952001462, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8230-279154-0018", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6ed97aea1b2ea4e52b1a3c908f5d690233bad4edd48c4b417c1ed84cd12b1074", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "812b9900ccbe6b84e9444fed1c7770731b7dfca39437dccad215d79db129df01"}, "rows": [{"case": "librispeech_8230-279154-0018", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.3600456714630127, "predicted_mos": 2.3600456714630127, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f84634eee8cb89ffc245b63dd1b8214e91b8848bd34e4929d9981a389ab5a341", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6515870559996984, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8230-279154-0018", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f84634eee8cb89ffc245b63dd1b8214e91b8848bd34e4929d9981a389ab5a341", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1579235172860352, "全帯域_10ms_rms_envelope_nrmse": 0.08057073948070388, "全帯域_10ms_envelope_correlation": 0.9935626723854157, "全帯域_10ms_modulation_mae_db": 0.23387462077981014, "全帯域_40ms_log_rms_mae_db": 0.5910190097163232, "全帯域_40ms_rms_envelope_nrmse": 0.03794522304533308, "全帯域_40ms_envelope_correlation": 0.9985702633762943, "全帯域_40ms_modulation_mae_db": 0.43607235029623226, "300–1000 Hz_10ms_log_rms_mae_db": 1.7994148652963853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10609129814832328, "300–1000 Hz_10ms_envelope_correlation": 0.9919403430907818, "300–1000 Hz_10ms_modulation_mae_db": 0.9582104483540747, "300–1000 Hz_40ms_log_rms_mae_db": 1.0693306245688254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059038634261299676, "300–1000 Hz_40ms_envelope_correlation": 0.9977556468171285, "300–1000 Hz_40ms_modulation_mae_db": 1.0040171491520455, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0883768013072688, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2515315020449861, "1000–3000 Hz_10ms_envelope_correlation": 0.9499318638911245, "1000–3000 Hz_10ms_modulation_mae_db": 0.6564370135440446, "1000–3000 Hz_40ms_log_rms_mae_db": 1.569220239362758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17115155242034089, "1000–3000 Hz_40ms_envelope_correlation": 0.9747870088799617, "1000–3000 Hz_40ms_modulation_mae_db": 0.8831642068417206, "3000–6000 Hz_10ms_log_rms_mae_db": 2.459831265113782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.46171367990428674, "3000–6000 Hz_10ms_envelope_correlation": 0.8844858868008607, "3000–6000 Hz_10ms_modulation_mae_db": 1.5699505841116739, "3000–6000 Hz_40ms_log_rms_mae_db": 2.362900282853703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39690264639253375, "3000–6000 Hz_40ms_envelope_correlation": 0.9115413103080438, "3000–6000 Hz_40ms_modulation_mae_db": 1.8372661906222263, "mr_stft_db": 6.029002326271076, "band_mae_db": 0.44231752660653695, "rms_gain_error_db": -0.10425970040012604, "output_peak": 0.515433296889439}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2444456904381642, "全帯域_10ms_rms_envelope_nrmse": 0.0786019926480825, "全帯域_10ms_envelope_correlation": 0.993912760723713, "全帯域_10ms_modulation_mae_db": 0.2327719204533536, "全帯域_40ms_log_rms_mae_db": 0.6663526597906774, "全帯域_40ms_rms_envelope_nrmse": 0.04428709911404532, "全帯域_40ms_envelope_correlation": 0.9980338325285966, "全帯域_40ms_modulation_mae_db": 0.1503065449251298, "300–1000 Hz_10ms_log_rms_mae_db": 1.8179768921618014, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10765935863235339, "300–1000 Hz_10ms_envelope_correlation": 0.9916910382608097, "300–1000 Hz_10ms_modulation_mae_db": 1.0455847257173232, "300–1000 Hz_40ms_log_rms_mae_db": 1.1668747753464088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06267392799270856, "300–1000 Hz_40ms_envelope_correlation": 0.9974091855523071, "300–1000 Hz_40ms_modulation_mae_db": 1.1626676934948355, "1000–3000 Hz_10ms_log_rms_mae_db": 2.121002005351316, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26295679340390526, "1000–3000 Hz_10ms_envelope_correlation": 0.9447996885926427, "1000–3000 Hz_10ms_modulation_mae_db": 0.5025327097031804, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6497093175546735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17843175450289298, "1000–3000 Hz_40ms_envelope_correlation": 0.9723957627439067, "1000–3000 Hz_40ms_modulation_mae_db": 0.8445527408407697, "3000–6000 Hz_10ms_log_rms_mae_db": 2.480190707671189, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.44655954898150074, "3000–6000 Hz_10ms_envelope_correlation": 0.8918167111789985, "3000–6000 Hz_10ms_modulation_mae_db": 0.9513364016726787, "3000–6000 Hz_40ms_log_rms_mae_db": 2.350715803330918, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38715486330814897, "3000–6000 Hz_40ms_envelope_correlation": 0.9157713032577295, "3000–6000 Hz_40ms_modulation_mae_db": 0.9412104709449688, "mr_stft_db": 6.0408843938620755, "band_mae_db": 0.4237447054126351, "rms_gain_error_db": -0.11832535222019823, "output_peak": 0.5241897512958076}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2448913428312276, "全帯域_10ms_rms_envelope_nrmse": 0.08568184405129378, "全帯域_10ms_envelope_correlation": 0.992719375410487, "全帯域_10ms_modulation_mae_db": 0.1830563596599203, "全帯域_40ms_log_rms_mae_db": 0.667479256296838, "全帯域_40ms_rms_envelope_nrmse": 0.05007837279377751, "全帯域_40ms_envelope_correlation": 0.9974072938012027, "全帯域_40ms_modulation_mae_db": 0.23027759738618792, "300–1000 Hz_10ms_log_rms_mae_db": 1.8297935701022146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11887428422430786, "300–1000 Hz_10ms_envelope_correlation": 0.9895244952002811, "300–1000 Hz_10ms_modulation_mae_db": 0.756618889880583, "300–1000 Hz_40ms_log_rms_mae_db": 1.1109974717044164, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07423624879912485, "300–1000 Hz_40ms_envelope_correlation": 0.9958351047486159, "300–1000 Hz_40ms_modulation_mae_db": 1.1595497268930273, "1000–3000 Hz_10ms_log_rms_mae_db": 2.092633917202453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24992460941813874, "1000–3000 Hz_10ms_envelope_correlation": 0.9496729421441625, "1000–3000 Hz_10ms_modulation_mae_db": 0.15935070791210876, "1000–3000 Hz_40ms_log_rms_mae_db": 1.573739828737956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16501238711659896, "1000–3000 Hz_40ms_envelope_correlation": 0.9763658714193777, "1000–3000 Hz_40ms_modulation_mae_db": 0.40324206864442597, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4851622434875638, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.45112255796833745, "3000–6000 Hz_10ms_envelope_correlation": 0.8899000922942315, "3000–6000 Hz_10ms_modulation_mae_db": 0.7186874128245964, "3000–6000 Hz_40ms_log_rms_mae_db": 2.328731090071372, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38696343164445357, "3000–6000 Hz_40ms_envelope_correlation": 0.9160633808967144, "3000–6000 Hz_40ms_modulation_mae_db": 0.8321467493250552, "mr_stft_db": 6.038450753425033, "band_mae_db": 0.4227741061722054, "rms_gain_error_db": -0.10786008437996791, "output_peak": 0.5069306189789967}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.223556807224831, "全帯域_10ms_rms_envelope_nrmse": 0.08751725852228116, "全帯域_10ms_envelope_correlation": 0.9924418299485797, "全帯域_10ms_modulation_mae_db": 0.17667106602870844, "全帯域_40ms_log_rms_mae_db": 0.6507917410453918, "全帯域_40ms_rms_envelope_nrmse": 0.052217608881611116, "全帯域_40ms_envelope_correlation": 0.997235299553518, "全帯域_40ms_modulation_mae_db": 0.16319914760091309, "300–1000 Hz_10ms_log_rms_mae_db": 1.792559164750395, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12174138511378614, "300–1000 Hz_10ms_envelope_correlation": 0.9892618744567051, "300–1000 Hz_10ms_modulation_mae_db": 0.31665954813107117, "300–1000 Hz_40ms_log_rms_mae_db": 1.0727272081803414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0776097867469174, "300–1000 Hz_40ms_envelope_correlation": 0.9957809821842246, "300–1000 Hz_40ms_modulation_mae_db": 0.32553032267267784, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0351129265678884, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25524246586352367, "1000–3000 Hz_10ms_envelope_correlation": 0.950111151652385, "1000–3000 Hz_10ms_modulation_mae_db": 1.0624233280362174, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5616705682482526, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1674877788657846, "1000–3000 Hz_40ms_envelope_correlation": 0.976889119671006, "1000–3000 Hz_40ms_modulation_mae_db": 1.2379210494550172, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4635128061387466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.46243022236374937, "3000–6000 Hz_10ms_envelope_correlation": 0.8917828807478059, "3000–6000 Hz_10ms_modulation_mae_db": 1.2940594661387619, "3000–6000 Hz_40ms_log_rms_mae_db": 2.32701694764634, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3971797794415383, "3000–6000 Hz_40ms_envelope_correlation": 0.91845567345123, "3000–6000 Hz_40ms_modulation_mae_db": 1.4127387869485741, "mr_stft_db": 6.021635587183003, "band_mae_db": 0.4781451342775544, "rms_gain_error_db": -0.13304564593265572, "output_peak": 0.5234098004558468}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1646382318728183, "全帯域_10ms_rms_envelope_nrmse": 0.08013425718608298, "全帯域_10ms_envelope_correlation": 0.993562291111358, "全帯域_10ms_modulation_mae_db": 0.16677945336824443, "全帯域_40ms_log_rms_mae_db": 0.5676373774817676, "全帯域_40ms_rms_envelope_nrmse": 0.041039248819459175, "全帯域_40ms_envelope_correlation": 0.9981864694320267, "全帯域_40ms_modulation_mae_db": 0.1530631018046132, "300–1000 Hz_10ms_log_rms_mae_db": 1.825004442528608, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10053866958261995, "300–1000 Hz_10ms_envelope_correlation": 0.9926925150725198, "300–1000 Hz_10ms_modulation_mae_db": 0.751238302510071, "300–1000 Hz_40ms_log_rms_mae_db": 1.1293272173923472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05315112871872688, "300–1000 Hz_40ms_envelope_correlation": 0.9981256832572565, "300–1000 Hz_40ms_modulation_mae_db": 0.789296738277225, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1408626025260626, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2584332865253078, "1000–3000 Hz_10ms_envelope_correlation": 0.9476700886444612, "1000–3000 Hz_10ms_modulation_mae_db": 0.6256203042537097, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6515384811289924, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17940655185207136, "1000–3000 Hz_40ms_envelope_correlation": 0.9725685679408252, "1000–3000 Hz_40ms_modulation_mae_db": 0.8253357640472716, "3000–6000 Hz_10ms_log_rms_mae_db": 2.496141340608973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4463848118194862, "3000–6000 Hz_10ms_envelope_correlation": 0.8875045732623472, "3000–6000 Hz_10ms_modulation_mae_db": 0.7610968909926094, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3222257181544834, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38819399527908044, "3000–6000 Hz_40ms_envelope_correlation": 0.9114750581036191, "3000–6000 Hz_40ms_modulation_mae_db": 0.9338476043503169, "mr_stft_db": 6.03488675027085, "band_mae_db": 0.38147513797872407, "rms_gain_error_db": -0.046652967739672146, "output_peak": 0.5275609654587501}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2332440849874657, "全帯域_10ms_rms_envelope_nrmse": 0.08366959556314955, "全帯域_10ms_envelope_correlation": 0.9930123821982111, "全帯域_10ms_modulation_mae_db": 0.2793771532481877, "全帯域_40ms_log_rms_mae_db": 0.6987405378857796, "全帯域_40ms_rms_envelope_nrmse": 0.05194686985161074, "全帯域_40ms_envelope_correlation": 0.9971281774989865, "全帯域_40ms_modulation_mae_db": 0.413816986856441, "300–1000 Hz_10ms_log_rms_mae_db": 1.8218341741916124, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10550431658726925, "300–1000 Hz_10ms_envelope_correlation": 0.9918023737495529, "300–1000 Hz_10ms_modulation_mae_db": 0.7821538957941169, "300–1000 Hz_40ms_log_rms_mae_db": 1.1088648226510054, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06360560878152889, "300–1000 Hz_40ms_envelope_correlation": 0.9969982098742832, "300–1000 Hz_40ms_modulation_mae_db": 0.7769793993550326, "1000–3000 Hz_10ms_log_rms_mae_db": 2.087010709802537, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23723975675688277, "1000–3000 Hz_10ms_envelope_correlation": 0.9562866646750507, "1000–3000 Hz_10ms_modulation_mae_db": 0.6121366396939659, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5612824775022038, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1600182185427859, "1000–3000 Hz_40ms_envelope_correlation": 0.9786098155635009, "1000–3000 Hz_40ms_modulation_mae_db": 0.8734711820428217, "3000–6000 Hz_10ms_log_rms_mae_db": 2.460779464123759, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.437622430429868, "3000–6000 Hz_10ms_envelope_correlation": 0.90568604848987, "3000–6000 Hz_10ms_modulation_mae_db": 0.9113664151831579, "3000–6000 Hz_40ms_log_rms_mae_db": 2.283054818600398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3805167634384577, "3000–6000 Hz_40ms_envelope_correlation": 0.9274623572037316, "3000–6000 Hz_40ms_modulation_mae_db": 1.1136804372445899, "mr_stft_db": 6.029355019004378, "band_mae_db": 0.529039936008477, "rms_gain_error_db": -0.08137680428706054, "output_peak": 0.5605651109948359}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2183336563969607, "全帯域_10ms_rms_envelope_nrmse": 0.09089364608580533, "全帯域_10ms_envelope_correlation": 0.9918060735609947, "全帯域_10ms_modulation_mae_db": 0.15391227101294924, "全帯域_40ms_log_rms_mae_db": 0.655590407140333, "全帯域_40ms_rms_envelope_nrmse": 0.04877740442472815, "全帯域_40ms_envelope_correlation": 0.9975900843331017, "全帯域_40ms_modulation_mae_db": 0.16620298669969746, "300–1000 Hz_10ms_log_rms_mae_db": 1.8466908135925515, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11651697935939527, "300–1000 Hz_10ms_envelope_correlation": 0.9901898710167242, "300–1000 Hz_10ms_modulation_mae_db": 0.6425185790730251, "300–1000 Hz_40ms_log_rms_mae_db": 1.1219698884871439, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06268752121945237, "300–1000 Hz_40ms_envelope_correlation": 0.9974392125581443, "300–1000 Hz_40ms_modulation_mae_db": 0.830375273939, "1000–3000 Hz_10ms_log_rms_mae_db": 2.050436084268127, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2413246852472176, "1000–3000 Hz_10ms_envelope_correlation": 0.9531650985200495, "1000–3000 Hz_10ms_modulation_mae_db": 0.6080948549138313, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5514400058440756, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1599771643113355, "1000–3000 Hz_40ms_envelope_correlation": 0.977850027993361, "1000–3000 Hz_40ms_modulation_mae_db": 0.861057885282456, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5542885639258497, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4659811647169401, "3000–6000 Hz_10ms_envelope_correlation": 0.8804124120285629, "3000–6000 Hz_10ms_modulation_mae_db": 1.39402666285132, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3678308326732593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.40217188347833915, "3000–6000 Hz_40ms_envelope_correlation": 0.9075177466379292, "3000–6000 Hz_40ms_modulation_mae_db": 1.2801676256074523, "mr_stft_db": 6.0379065130294185, "band_mae_db": 0.41338719525973655, "rms_gain_error_db": -0.12433908569124105, "output_peak": 0.5363590605531383}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2536030493451642, "全帯域_10ms_rms_envelope_nrmse": 0.07988708512430283, "全帯域_10ms_envelope_correlation": 0.9936608276504628, "全帯域_10ms_modulation_mae_db": 0.2523554086378231, "全帯域_40ms_log_rms_mae_db": 0.6501820661171072, "全帯域_40ms_rms_envelope_nrmse": 0.04132270419460769, "全帯域_40ms_envelope_correlation": 0.9982681493447054, "全帯域_40ms_modulation_mae_db": 0.26224346061220843, "300–1000 Hz_10ms_log_rms_mae_db": 1.7925804466400037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10739993537963714, "300–1000 Hz_10ms_envelope_correlation": 0.9915536720691857, "300–1000 Hz_10ms_modulation_mae_db": 0.5739236137971347, "300–1000 Hz_40ms_log_rms_mae_db": 1.1230315357392537, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05572572854914505, "300–1000 Hz_40ms_envelope_correlation": 0.9978469023348212, "300–1000 Hz_40ms_modulation_mae_db": 0.523441866786939, "1000–3000 Hz_10ms_log_rms_mae_db": 2.093227803270142, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25065233976128914, "1000–3000 Hz_10ms_envelope_correlation": 0.9497206501764437, "1000–3000 Hz_10ms_modulation_mae_db": 0.49958809078099264, "1000–3000 Hz_40ms_log_rms_mae_db": 1.586932234971095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1707181888858863, "1000–3000 Hz_40ms_envelope_correlation": 0.9746068289349514, "1000–3000 Hz_40ms_modulation_mae_db": 0.7187607355694917, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4961567960787776, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4469693617490467, "3000–6000 Hz_10ms_envelope_correlation": 0.8991637951646422, "3000–6000 Hz_10ms_modulation_mae_db": 1.1226553032318904, "3000–6000 Hz_40ms_log_rms_mae_db": 2.313616215322246, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3887178787954402, "3000–6000 Hz_40ms_envelope_correlation": 0.9219932762064705, "3000–6000 Hz_40ms_modulation_mae_db": 1.2277765199367587, "mr_stft_db": 6.037389336393463, "band_mae_db": 0.5297191281775735, "rms_gain_error_db": -0.09243786362088167, "output_peak": 0.5249265053616856}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.128605677753013, "spectral_convergence": 0.23862456940831944, "spectral_convergence_status": "ok", "amplitude_floor": 1.9267471569597447e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.796163144375008, "spectral_convergence": 0.22196750100842522, "spectral_convergence_status": "ok", "amplitude_floor": 1.271916808390449e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.162238156685208, "spectral_convergence": 0.23262560201367663, "spectral_convergence_status": "ok", "amplitude_floor": 1.1644398478092437e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001373055391651405, "output_power": 0.001391811801637012, "delta_db": 0.05892456851577552, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0033897235254818038, "output_power": 0.003208998252297502, "delta_db": -0.23794796911448926, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002841140646523882, "output_power": 0.0003107161166787259, "delta_db": 0.3887104709162753, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.671358771236327e-05, "output_power": 8.586275085659692e-05, "delta_db": -0.042823569961190924, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6620686952686656e-05, "output_power": 2.9270831925935675e-05, "delta_db": 0.41215807635838325, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.533501052724457e-05, "output_power": 3.5896750753571117e-05, "delta_db": 1.5133405047731074, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.44231752660653695, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10425970040012604, "pitch": null, "output_rms_dbfs": -22.956301165074834, "output_peak": 0.515433296889439, "reference_rms_dbfs": -22.85204146467471, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.029002326271076}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023735521522065317, "log_rms_mae_db": 1.1579235172860352, "rms_envelope_nrmse": 0.08057073948070388, "envelope_correlation": 0.9935626723854157, "rms_gain_error_db": -0.10422652435916274, "reference_mean_power": 0.0051897925018650195, "output_mean_power": 0.005066725099315885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6580095170174236e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00099786392963893, "output_power": 0.0009482023118209848, "reference_active": true, "delta_db": -0.2217031437563779}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00025981544310873085, "output_power": 0.00023641562757243618, "reference_active": true, "delta_db": -0.4098878049047323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.167033379867013e-05, "output_power": 4.100376048415325e-05, "reference_active": true, "delta_db": -0.07003291367832024}], "active_band_count": 3, "active_band_mae_db": 0.23387462077981014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022053657198928817, "log_rms_mae_db": 0.5910190097163232, "rms_envelope_nrmse": 0.03794522304533308, "envelope_correlation": 0.9985702633762943, "rms_gain_error_db": -0.10418640980547791, "reference_mean_power": 0.0052155458817555326, "output_mean_power": 0.005091914812474715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3916979728964525e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008985624617991317, "output_power": 0.0008517430988486422, "reference_active": true, "delta_db": -0.23239648186905848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001435836821660598, "output_power": 0.00013145368012168143, "reference_active": true, "delta_db": -0.38332337571172925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3776224720397365e-06, "output_power": 2.0271840359668352e-06, "reference_active": true, "delta_db": -0.6924971933079092}], "active_band_count": 3, "active_band_mae_db": 0.43607235029623226}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018677846007938983, "log_rms_mae_db": 1.7994148652963853, "rms_envelope_nrmse": 0.10609129814832328, "envelope_correlation": 0.9919403430907818, "rms_gain_error_db": -0.23068335045372038, "reference_mean_power": 0.0023293656178310037, "output_mean_power": 0.002208865773413974}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5191576623571145e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006490500325829311, "output_power": 0.0006121376280263238, "reference_active": true, "delta_db": -0.25429099707146596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013228128531435308, "output_power": 0.00010540591787982805, "reference_active": true, "delta_db": -0.9863341162844336}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.716163060824064e-05, "output_power": 1.1780342756025572e-05, "reference_active": true, "delta_db": -1.6340062317063244}], "active_band_count": 3, "active_band_mae_db": 0.9582104483540747}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017927232752442356, "log_rms_mae_db": 1.0693306245688254, "rms_envelope_nrmse": 0.059038634261299676, "envelope_correlation": 0.9977556468171285, "rms_gain_error_db": -0.23075792988735283, "reference_mean_power": 0.002341010543733861, "output_mean_power": 0.0022198701773390975}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.424681480871225e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006148042140063018, "output_power": 0.0005718523366876788, "reference_active": true, "delta_db": -0.31452935539432036}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.176867672883116e-05, "output_power": 4.9257341415026946e-05, "reference_active": true, "delta_db": -0.9829732974440057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.094779276972058e-06, "output_power": 7.376862500864847e-07, "reference_active": true, "delta_db": -1.7145487946178104}], "active_band_count": 3, "active_band_mae_db": 1.0040171491520455}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.042588377582574e-05, "log_rms_mae_db": 2.0883768013072688, "rms_envelope_nrmse": 0.2515315020449861, "envelope_correlation": 0.9499318638911245, "rms_gain_error_db": 0.06757214332108788, "reference_mean_power": 9.196537764000522e-05, "output_mean_power": 9.340746220860931e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3593984658429913e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0478584079151807e-05, "output_power": 2.087502784276923e-05, "reference_active": true, "delta_db": 0.08327137858503561}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.015542002555929e-06, "output_power": 6.793538535198732e-06, "reference_active": true, "delta_db": 0.5282127967565722}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6245635384567205e-06, "output_power": 2.2208508270178193e-06, "reference_active": true, "delta_db": 1.3578268652905259}], "active_band_count": 3, "active_band_mae_db": 0.6564370135440446}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6202941099492215e-05, "log_rms_mae_db": 1.569220239362758, "rms_envelope_nrmse": 0.17115155242034089, "envelope_correlation": 0.9747870088799617, "rms_gain_error_db": 0.06749303500674905, "reference_mean_power": 9.242233001846982e-05, "output_mean_power": 9.386987004620922e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9023164132494484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.934413521065944e-05, "output_power": 2.0564112838910983e-05, "reference_active": true, "delta_db": 0.26560659114944535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.243887754162173e-06, "output_power": 3.965324385072424e-06, "reference_active": true, "delta_db": 0.8721290252455707}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3065359453687705e-07, "output_power": 1.850533974881216e-07, "reference_active": true, "delta_db": 1.5117570041301458}], "active_band_count": 3, "active_band_mae_db": 0.8831642068417206}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.535256840646245e-05, "log_rms_mae_db": 2.459831265113782, "rms_envelope_nrmse": 0.46171367990428674, "envelope_correlation": 0.8844858868008607, "rms_gain_error_db": 0.757113639148226, "reference_mean_power": 2.9383590737924492e-05, "output_mean_power": 3.4979712070023386e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.113292932890967e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.65042891211228e-06, "output_power": 9.618384323686963e-06, "reference_active": true, "delta_db": -0.01444489661860714}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.01875786446688e-06, "output_power": 5.138703231832673e-06, "reference_active": true, "delta_db": 1.0676169706498269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.253460098772285e-07, "output_power": 1.6723386031625636e-06, "reference_active": true, "delta_db": 3.627789885066588}], "active_band_count": 3, "active_band_mae_db": 1.5699505841116739}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9034899521347454e-05, "log_rms_mae_db": 2.362900282853703, "rms_envelope_nrmse": 0.39690264639253375, "envelope_correlation": 0.9115413103080438, "rms_gain_error_db": 0.7570788450933339, "reference_mean_power": 2.9530847534057242e-05, "output_mean_power": 3.5154732357811306e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9758537203809177e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.950101722919118e-06, "output_power": 1.051859125244271e-05, "reference_active": true, "delta_db": 0.24130058263527307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0870738801319347e-06, "output_power": 3.090145085926904e-06, "reference_active": true, "delta_db": 1.704410476504613}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.171675766600786e-08, "output_power": 1.4028519923782228e-07, "reference_active": true, "delta_db": 3.5660875127267926}], "active_band_count": 3, "active_band_mae_db": 1.8372661906222263}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8620333338037293, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a11e59936757f46735cda0eb5977ceed19a9d676114d7e8a3e0bf986ffec576e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.95016976853704, "standard_deviation": 0.003422262338009462, "minimum": 0.9447996885926427, "maximum": 0.9562866646750507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0885828562869744, "standard_deviation": 0.03409835233292474, "minimum": 2.0351129265678884, "maximum": 2.1408626025260626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5907729561047563, "standard_deviation": 0.2481252021503499, "minimum": 0.15935070791210876, "maximum": 1.0624233280362174, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2509131798776564, "standard_deviation": 0.00845722606033386, "minimum": 0.23723975675688277, "maximum": 0.26295679340390526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9755091253933614, "standard_deviation": 0.0023117092856579442, "minimum": 0.9723957627439067, "maximum": 0.9786098155635009, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.588191644168751, "standard_deviation": 0.039903016017386835, "minimum": 1.5514400058440756, "maximum": 1.6515384811289924, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8309382040904968, "standard_deviation": 0.22891150905898036, "minimum": 0.40324206864442597, "maximum": 1.2379210494550172, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16902544956221205, "standard_deviation": 0.007417183862514727, "minimum": 0.1599771643113355, "maximum": 0.17940655185207136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8913440499959149, "standard_deviation": 0.008023350055737751, "minimum": 0.8804124120285629, "maximum": 0.90568604848987, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.48700789839358, "standard_deviation": 0.030988961403637445, "minimum": 2.459831265113782, "maximum": 2.5542885639258497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.090397392125836, "standard_deviation": 0.3076821155800839, "minimum": 0.7186874128245964, "maximum": 1.5699505841116739, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.4523479722416519, "standard_deviation": 0.009939603206343451, "minimum": 0.437622430429868, "maximum": 0.4659811647169401, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9162850132581835, "standard_deviation": 0.006381283103895916, "minimum": 0.9075177466379292, "maximum": 0.9274623572037316, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.3320114635815896, "standard_deviation": 0.027912116629839372, "minimum": 2.283054818600398, "maximum": 2.3678308326732593, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1973542981224927, "standard_deviation": 0.32493616373845546, "minimum": 0.8321467493250552, "maximum": 1.8372661906222263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.390975155222249, "standard_deviation": 0.007090193423620769, "minimum": 0.3805167634384577, "maximum": 0.40217188347833915, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.99108202286457, "standard_deviation": 0.0012522793193927681, "minimum": 0.9892618744567051, "maximum": 0.9926925150725198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8157317961579464, "standard_deviation": 0.019359950394773645, "minimum": 1.792559164750395, "maximum": 1.8466908135925515, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.728363500407175, "standard_deviation": 0.22627263230888833, "minimum": 0.31665954813107117, "maximum": 1.0455847257173232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11054077837846155, "standard_deviation": 0.007502279326402609, "minimum": 0.10053866958261995, "maximum": 0.12174138511378614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971488659158476, "standard_deviation": 0.0008929444918993557, "minimum": 0.9957809821842246, "maximum": 0.9981256832572565, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1128904430087179, "standard_deviation": 0.031396282664555505, "minimum": 1.0693306245688254, "maximum": 1.1668747753464088, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8214822713213479, "standard_deviation": 0.29323515580373066, "minimum": 0.32553032267267784, "maximum": 1.1626676934948355, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06359107313361295, "standard_deviation": 0.008477555690359783, "minimum": 0.05315112871872688, "maximum": 0.0776097867469174, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.45257535873668037, "standard_deviation": 0.05459893255910231, "minimum": 0.38147513797872407, "maximum": 0.5297191281775735, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.033688834929912, "standard_deviation": 0.006474075356278177, "minimum": 6.021635587183003, "maximum": 6.0408843938620755, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5274218887485624, "standard_deviation": 0.015922825206175852, "minimum": 0.5069306189789967, "maximum": 0.5605651109948359, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1010371880339754, "standard_deviation": 0.027623642615303583, "minimum": -0.13304564593265572, "maximum": -0.046652967739672146, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930847766236528, "standard_deviation": 0.0007231891522354877, "minimum": 0.9918060735609947, "maximum": 0.993912760723713, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2175795475478333, "standard_deviation": 0.03666199935267038, "minimum": 1.1579235172860352, "maximum": 1.2536030493451642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2098497816486246, "standard_deviation": 0.045585357887780585, "minimum": 0.15391227101294924, "maximum": 0.2793771532481877, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08336955233271276, "standard_deviation": 0.004349217723466379, "minimum": 0.0786019926480825, "maximum": 0.09089364608580533, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997802446233554, "standard_deviation": 0.0005325105889716657, "minimum": 0.9971281774989865, "maximum": 0.9985702633762943, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6434741319342772, "standard_deviation": 0.04293208879996775, "minimum": 0.5676373774817676, "maximum": 0.6987405378857796, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2468977720226779, "standard_deviation": 0.11699370209209188, "minimum": 0.1503065449251298, "maximum": 0.43607235029623226, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0459518163906466, "standard_deviation": 0.005511702582420726, "minimum": 0.03794522304533308, "maximum": 0.052217608881611116, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0018", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.1463446617126465, "predicted_mos": 2.1463446617126465, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "54f06f17b4af5c1e9634191c73155fd6f0e4a61eeba1ae59b23e9d07a124bfe2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6865511001087725, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1474291903909468, "全帯域_10ms_rms_envelope_nrmse": 0.07856922630308283, "全帯域_10ms_envelope_correlation": 0.9938962048056798, "全帯域_10ms_modulation_mae_db": 0.322427943648979, "全帯域_40ms_log_rms_mae_db": 0.5896512680107934, "全帯域_40ms_rms_envelope_nrmse": 0.03854756911385426, "全帯域_40ms_envelope_correlation": 0.9985314020361179, "全帯域_40ms_modulation_mae_db": 0.5326258421663342, "300–1000 Hz_10ms_log_rms_mae_db": 1.7687774502218758, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10202866847790132, "300–1000 Hz_10ms_envelope_correlation": 0.9925427684258212, "300–1000 Hz_10ms_modulation_mae_db": 0.9765669029387465, "300–1000 Hz_40ms_log_rms_mae_db": 0.9859534454309448, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05472839788360211, "300–1000 Hz_40ms_envelope_correlation": 0.9980950253180784, "300–1000 Hz_40ms_modulation_mae_db": 1.0270561691456663, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0785020630718294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24226430795314916, "1000–3000 Hz_10ms_envelope_correlation": 0.9525411095753689, "1000–3000 Hz_10ms_modulation_mae_db": 0.5819691536105035, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4876460889662735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16107114047718707, "1000–3000 Hz_40ms_envelope_correlation": 0.9773614383184378, "1000–3000 Hz_40ms_modulation_mae_db": 0.6125489869366809, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3818203999336043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4522103658550504, "3000–6000 Hz_10ms_envelope_correlation": 0.8880600021959394, "3000–6000 Hz_10ms_modulation_mae_db": 1.4324492003202813, "3000–6000 Hz_40ms_log_rms_mae_db": 2.18409066579015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3887651900891888, "3000–6000 Hz_40ms_envelope_correlation": 0.9139542818678924, "3000–6000 Hz_40ms_modulation_mae_db": 1.7021038504327362, "mr_stft_db": 5.938998347404511, "band_mae_db": 0.43026540798542645, "rms_gain_error_db": -0.10898167667914957, "output_peak": 0.5149108794420896}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.233123917091392, "全帯域_10ms_rms_envelope_nrmse": 0.0800543103913296, "全帯域_10ms_envelope_correlation": 0.9936594133808108, "全帯域_10ms_modulation_mae_db": 0.27208068059831275, "全帯域_40ms_log_rms_mae_db": 0.6539057223127758, "全帯域_40ms_rms_envelope_nrmse": 0.044472570713934886, "全帯域_40ms_envelope_correlation": 0.9979889232728991, "全帯域_40ms_modulation_mae_db": 0.2161435411895666, "300–1000 Hz_10ms_log_rms_mae_db": 1.789817464273233, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1043112568853933, "300–1000 Hz_10ms_envelope_correlation": 0.992152261889096, "300–1000 Hz_10ms_modulation_mae_db": 1.0698547528296798, "300–1000 Hz_40ms_log_rms_mae_db": 1.1167905266247309, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0597101843967881, "300–1000 Hz_40ms_envelope_correlation": 0.9975875729238642, "300–1000 Hz_40ms_modulation_mae_db": 1.232872290287601, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0666099794398667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24699771173441804, "1000–3000 Hz_10ms_envelope_correlation": 0.9504619739443171, "1000–3000 Hz_10ms_modulation_mae_db": 0.2009634140420361, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5209847598379411, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1660262423632409, "1000–3000 Hz_40ms_envelope_correlation": 0.9760798676252772, "1000–3000 Hz_40ms_modulation_mae_db": 0.44355344794243284, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4208045344795357, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4471497520694136, "3000–6000 Hz_10ms_envelope_correlation": 0.8921830027061372, "3000–6000 Hz_10ms_modulation_mae_db": 0.9019944249588727, "3000–6000 Hz_40ms_log_rms_mae_db": 2.212292379825442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3866407582137563, "3000–6000 Hz_40ms_envelope_correlation": 0.9163387045358633, "3000–6000 Hz_40ms_modulation_mae_db": 0.8376187087164296, "mr_stft_db": 5.952678392891002, "band_mae_db": 0.4305657604606792, "rms_gain_error_db": -0.10843320809897877, "output_peak": 0.5236662126878118}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2586377413076104, "全帯域_10ms_rms_envelope_nrmse": 0.08544656206758403, "全帯域_10ms_envelope_correlation": 0.9927738610600736, "全帯域_10ms_modulation_mae_db": 0.1448624447333525, "全帯域_40ms_log_rms_mae_db": 0.7029733282944195, "全帯域_40ms_rms_envelope_nrmse": 0.0522222138244197, "全帯域_40ms_envelope_correlation": 0.9971811174401554, "全帯域_40ms_modulation_mae_db": 0.19032683721392799, "300–1000 Hz_10ms_log_rms_mae_db": 1.8124880029902892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.117909634512078, "300–1000 Hz_10ms_envelope_correlation": 0.9897473124659324, "300–1000 Hz_10ms_modulation_mae_db": 0.8721829183188446, "300–1000 Hz_40ms_log_rms_mae_db": 1.0516371049080107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07251773996843862, "300–1000 Hz_40ms_envelope_correlation": 0.9961070260861254, "300–1000 Hz_40ms_modulation_mae_db": 1.2857967342327552, "1000–3000 Hz_10ms_log_rms_mae_db": 2.077077623697724, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24377357233608896, "1000–3000 Hz_10ms_envelope_correlation": 0.9517425209834516, "1000–3000 Hz_10ms_modulation_mae_db": 0.20116651643038788, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5251286550534107, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16213556658030465, "1000–3000 Hz_40ms_envelope_correlation": 0.9772156497330615, "1000–3000 Hz_40ms_modulation_mae_db": 0.25380591000429503, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4017067695901284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.45043254417692385, "3000–6000 Hz_10ms_envelope_correlation": 0.8910635740680289, "3000–6000 Hz_10ms_modulation_mae_db": 0.6416380208574982, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1939579862666947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3859705780751293, "3000–6000 Hz_40ms_envelope_correlation": 0.9170099707300661, "3000–6000 Hz_40ms_modulation_mae_db": 0.8437510387109981, "mr_stft_db": 5.9529838524132055, "band_mae_db": 0.4328324831001691, "rms_gain_error_db": -0.11418803615345625, "output_peak": 0.5049211549829942}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2605665674866466, "全帯域_10ms_rms_envelope_nrmse": 0.08923183350157335, "全帯域_10ms_envelope_correlation": 0.9921307083241935, "全帯域_10ms_modulation_mae_db": 0.19595377892941113, "全帯域_40ms_log_rms_mae_db": 0.6932361084277676, "全帯域_40ms_rms_envelope_nrmse": 0.05296489664338152, "全帯域_40ms_envelope_correlation": 0.9971442437802194, "全帯域_40ms_modulation_mae_db": 0.16465566634318748, "300–1000 Hz_10ms_log_rms_mae_db": 1.80324368716397, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12032315879249818, "300–1000 Hz_10ms_envelope_correlation": 0.9895278921706019, "300–1000 Hz_10ms_modulation_mae_db": 0.4264606315163935, "300–1000 Hz_40ms_log_rms_mae_db": 1.051871026764562, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07400920677427544, "300–1000 Hz_40ms_envelope_correlation": 0.9962248472480146, "300–1000 Hz_40ms_modulation_mae_db": 0.46020364423027776, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9813645411403717, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24349539428714698, "1000–3000 Hz_10ms_envelope_correlation": 0.9528143709376099, "1000–3000 Hz_10ms_modulation_mae_db": 0.871253643226054, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4488709415446095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15529371763891875, "1000–3000 Hz_40ms_envelope_correlation": 0.9790469066435805, "1000–3000 Hz_40ms_modulation_mae_db": 0.9841343027618619, "3000–6000 Hz_10ms_log_rms_mae_db": 2.388716175153092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4576928102682468, "3000–6000 Hz_10ms_envelope_correlation": 0.8940212100087006, "3000–6000 Hz_10ms_modulation_mae_db": 1.2758015889044059, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1976225414912247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39191776345134427, "3000–6000 Hz_40ms_envelope_correlation": 0.9204990220851164, "3000–6000 Hz_40ms_modulation_mae_db": 1.4342417452564042, "mr_stft_db": 5.932185506010479, "band_mae_db": 0.43883473678051227, "rms_gain_error_db": -0.13201839876388083, "output_peak": 0.5209388468086479}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.212014690052011, "全帯域_10ms_rms_envelope_nrmse": 0.0844066414169855, "全帯域_10ms_envelope_correlation": 0.9928610561235823, "全帯域_10ms_modulation_mae_db": 0.18497934651175627, "全帯域_40ms_log_rms_mae_db": 0.6087770051433423, "全帯域_40ms_rms_envelope_nrmse": 0.04418853826808932, "全帯域_40ms_envelope_correlation": 0.9979120365428013, "全帯域_40ms_modulation_mae_db": 0.18446393688084387, "300–1000 Hz_10ms_log_rms_mae_db": 1.8508794655913965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10250056212271622, "300–1000 Hz_10ms_envelope_correlation": 0.9924234381551669, "300–1000 Hz_10ms_modulation_mae_db": 0.8014554285982484, "300–1000 Hz_40ms_log_rms_mae_db": 1.0892957791802047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05361303796467008, "300–1000 Hz_40ms_envelope_correlation": 0.998129262629057, "300–1000 Hz_40ms_modulation_mae_db": 0.8244802117289596, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1173551350749507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24778376989477408, "1000–3000 Hz_10ms_envelope_correlation": 0.9506524702841842, "1000–3000 Hz_10ms_modulation_mae_db": 0.4537915092075528, "1000–3000 Hz_40ms_log_rms_mae_db": 1.569743660993047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16997098117663798, "1000–3000 Hz_40ms_envelope_correlation": 0.97471463480294, "1000–3000 Hz_40ms_modulation_mae_db": 0.5732348269607428, "3000–6000 Hz_10ms_log_rms_mae_db": 2.399805167508823, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.446155552767503, "3000–6000 Hz_10ms_envelope_correlation": 0.8880942400162868, "3000–6000 Hz_10ms_modulation_mae_db": 0.7434603414504938, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1710145165644383, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38648947229670144, "3000–6000 Hz_40ms_envelope_correlation": 0.9124497274304272, "3000–6000 Hz_40ms_modulation_mae_db": 0.9008195025335843, "mr_stft_db": 5.957060933234168, "band_mae_db": 0.3451195160457894, "rms_gain_error_db": -0.058572228634378755, "output_peak": 0.5283259917626311}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2305994892421288, "全帯域_10ms_rms_envelope_nrmse": 0.082297222698609, "全帯域_10ms_envelope_correlation": 0.9932385520439978, "全帯域_10ms_modulation_mae_db": 0.2878504151315227, "全帯域_40ms_log_rms_mae_db": 0.6901341113081957, "全帯域_40ms_rms_envelope_nrmse": 0.049045114838178104, "全帯域_40ms_envelope_correlation": 0.9974426548384564, "全帯域_40ms_modulation_mae_db": 0.49696247253569154, "300–1000 Hz_10ms_log_rms_mae_db": 1.8227514309708797, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10146580560983028, "300–1000 Hz_10ms_envelope_correlation": 0.9924039911425223, "300–1000 Hz_10ms_modulation_mae_db": 0.7998773547387161, "300–1000 Hz_40ms_log_rms_mae_db": 1.0631398488872907, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058500951444037656, "300–1000 Hz_40ms_envelope_correlation": 0.9974485089820478, "300–1000 Hz_40ms_modulation_mae_db": 0.8042558215607644, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0398875913598644, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23344370698996264, "1000–3000 Hz_10ms_envelope_correlation": 0.9564772221327308, "1000–3000 Hz_10ms_modulation_mae_db": 0.47299951268849316, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4518964939231755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15676103564495877, "1000–3000 Hz_40ms_envelope_correlation": 0.9786310475627894, "1000–3000 Hz_40ms_modulation_mae_db": 0.6716444055481046, "3000–6000 Hz_10ms_log_rms_mae_db": 2.438094394338693, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4406248239829509, "3000–6000 Hz_10ms_envelope_correlation": 0.904347162606489, "3000–6000 Hz_10ms_modulation_mae_db": 0.7240295244637879, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1939024045783957, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3821611771303471, "3000–6000 Hz_40ms_envelope_correlation": 0.9265970918976302, "3000–6000 Hz_40ms_modulation_mae_db": 0.9303266597037881, "mr_stft_db": 5.946778876265029, "band_mae_db": 0.49451276002222566, "rms_gain_error_db": -0.07438492533159258, "output_peak": 0.5623162143178225}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2370532263922156, "全帯域_10ms_rms_envelope_nrmse": 0.09754361276436936, "全帯域_10ms_envelope_correlation": 0.9905380411359356, "全帯域_10ms_modulation_mae_db": 0.1501329625563439, "全帯域_40ms_log_rms_mae_db": 0.6629990561665084, "全帯域_40ms_rms_envelope_nrmse": 0.05202780495245378, "全帯域_40ms_envelope_correlation": 0.9972397520393621, "全帯域_40ms_modulation_mae_db": 0.1577966624078497, "300–1000 Hz_10ms_log_rms_mae_db": 1.8432336541592347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12914898649717013, "300–1000 Hz_10ms_envelope_correlation": 0.9877812643378497, "300–1000 Hz_10ms_modulation_mae_db": 0.6247998368845399, "300–1000 Hz_40ms_log_rms_mae_db": 1.0830901193649034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06585305591316845, "300–1000 Hz_40ms_envelope_correlation": 0.9970851699739272, "300–1000 Hz_40ms_modulation_mae_db": 0.8302000321627164, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0410022255983096, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24053126854546167, "1000–3000 Hz_10ms_envelope_correlation": 0.953031633868614, "1000–3000 Hz_10ms_modulation_mae_db": 0.564380101895002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.503668904901297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16195231858329007, "1000–3000 Hz_40ms_envelope_correlation": 0.977560306176324, "1000–3000 Hz_40ms_modulation_mae_db": 0.7293631627857243, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4629252738349363, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4645046202302071, "3000–6000 Hz_10ms_envelope_correlation": 0.8807624854042606, "3000–6000 Hz_10ms_modulation_mae_db": 1.4910674450278918, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2312050173843505, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39802464716559005, "3000–6000 Hz_40ms_envelope_correlation": 0.9088996513189334, "3000–6000 Hz_40ms_modulation_mae_db": 1.3391206857057625, "mr_stft_db": 5.95428955619505, "band_mae_db": 0.39139963412480716, "rms_gain_error_db": -0.12789680689142424, "output_peak": 0.5355421552084542}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2943184455414551, "全帯域_10ms_rms_envelope_nrmse": 0.08063557664172032, "全帯域_10ms_envelope_correlation": 0.9935229051206017, "全帯域_10ms_modulation_mae_db": 0.1859222558096134, "全帯域_40ms_log_rms_mae_db": 0.687425644358396, "全帯域_40ms_rms_envelope_nrmse": 0.04152216061988724, "全帯域_40ms_envelope_correlation": 0.9982262355414725, "全帯域_40ms_modulation_mae_db": 0.21257617159029954, "300–1000 Hz_10ms_log_rms_mae_db": 1.7957418030144017, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10613720318169986, "300–1000 Hz_10ms_envelope_correlation": 0.9917702779045868, "300–1000 Hz_10ms_modulation_mae_db": 0.6434738814363095, "300–1000 Hz_40ms_log_rms_mae_db": 1.0463069383211943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056192761716079606, "300–1000 Hz_40ms_envelope_correlation": 0.9978130635814154, "300–1000 Hz_40ms_modulation_mae_db": 0.6567384175251023, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0556454451168635, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24690126188336578, "1000–3000 Hz_10ms_envelope_correlation": 0.9506732651440395, "1000–3000 Hz_10ms_modulation_mae_db": 0.4879769038624288, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5072746858261572, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16491250367573443, "1000–3000 Hz_40ms_envelope_correlation": 0.9762409348349237, "1000–3000 Hz_40ms_modulation_mae_db": 0.6337594734314314, "3000–6000 Hz_10ms_log_rms_mae_db": 2.451338253873731, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.44763758143461807, "3000–6000 Hz_10ms_envelope_correlation": 0.8989198817008645, "3000–6000 Hz_10ms_modulation_mae_db": 1.176509949296833, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2177655448507725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3873821869836414, "3000–6000 Hz_40ms_envelope_correlation": 0.9224404246773943, "3000–6000 Hz_40ms_modulation_mae_db": 1.3410775222085016, "mr_stft_db": 5.951873080605331, "band_mae_db": 0.5201530624018124, "rms_gain_error_db": -0.08249677236261634, "output_peak": 0.5208607425317743}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.042253612636893, "spectral_convergence": 0.23440140631249917, "spectral_convergence_status": "ok", "amplitude_floor": 1.9267471569597447e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.722918905890967, "spectral_convergence": 0.2185964917058451, "spectral_convergence_status": "ok", "amplitude_floor": 1.271916808390449e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0518225236856775, "spectral_convergence": 0.23007613746802252, "spectral_convergence_status": "ok", "amplitude_floor": 1.1644398478092437e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001373055391651405, "output_power": 0.0013883538593523117, "delta_db": 0.04812114015554794, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0033897235254818038, "output_power": 0.0032158323827159076, "delta_db": -0.22870873335142583, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002841140646523882, "output_power": 0.00030767461574310605, "delta_db": 0.3459893350646934, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.671358771236327e-05, "output_power": 8.161768347153845e-05, "delta_db": -0.26302891045572574, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6620686952686656e-05, "output_power": 2.805882915232886e-05, "delta_db": 0.22850286345381188, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.533501052724457e-05, "output_power": 3.5517732793041395e-05, "delta_db": 1.467241465431354, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.43026540798542645, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10898167667914957, "pitch": null, "output_rms_dbfs": -22.961023141353856, "output_peak": 0.5149108794420896, "reference_rms_dbfs": -22.85204146467471, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.938998347404511}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023735521522065317, "log_rms_mae_db": 1.1474291903909468, "rms_envelope_nrmse": 0.07856922630308283, "envelope_correlation": 0.9938962048056798, "rms_gain_error_db": -0.10894832921708035, "reference_mean_power": 0.0051897925018650195, "output_mean_power": 0.005061219368219885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6580095170174236e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00099786392963893, "output_power": 0.0009469582079382538, "reference_active": true, "delta_db": -0.227405114889328}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00025981544310873085, "output_power": 0.00023790683762012522, "reference_active": true, "delta_db": -0.382580372670261}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.167033379867013e-05, "output_power": 3.8379308453377416e-05, "reference_active": true, "delta_db": -0.3572983433873479}], "active_band_count": 3, "active_band_mae_db": 0.322427943648979}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022053657198928817, "log_rms_mae_db": 0.5896512680107934, "rms_envelope_nrmse": 0.03854756911385426, "envelope_correlation": 0.9985314020361179, "rms_gain_error_db": -0.10893792068199752, "reference_mean_power": 0.0052155458817555326, "output_mean_power": 0.005086346918049649}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3916979728964525e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008985624617991317, "output_power": 0.0008493150762292454, "reference_active": true, "delta_db": -0.24479438285922506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001435836821660598, "output_power": 0.00013281843508011358, "reference_active": true, "delta_db": -0.33846727606738797}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3776224720397365e-06, "output_power": 1.8822673546777815e-06, "reference_active": true, "delta_db": -1.0146158675723895}], "active_band_count": 3, "active_band_mae_db": 0.5326258421663342}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018677846007938983, "log_rms_mae_db": 1.7687774502218758, "rms_envelope_nrmse": 0.10202866847790132, "envelope_correlation": 0.9925427684258212, "rms_gain_error_db": -0.21957199880409892, "reference_mean_power": 0.0023293656178310037, "output_mean_power": 0.0022145243551412726}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5191576623571145e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006490500325829311, "output_power": 0.0006118488671328257, "reference_active": true, "delta_db": -0.25634015799698645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013228128531435308, "output_power": 0.0001065114082052338, "reference_active": true, "delta_db": -0.9410227949186873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.716163060824064e-05, "output_power": 1.1516613028069516e-05, "reference_active": true, "delta_db": -1.7323377559005657}], "active_band_count": 3, "active_band_mae_db": 0.9765669029387465}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017927232752442356, "log_rms_mae_db": 0.9859534454309448, "rms_envelope_nrmse": 0.05472839788360211, "envelope_correlation": 0.9980950253180784, "rms_gain_error_db": -0.21963331347758486, "reference_mean_power": 0.002341010543733861, "output_mean_power": 0.002225563747279112}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.424681480871225e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006148042140063018, "output_power": 0.0005718231811657232, "reference_active": true, "delta_db": -0.31475078327499795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.176867672883116e-05, "output_power": 4.950298500142575e-05, "reference_active": true, "delta_db": -0.9613691011122427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.094779276972058e-06, "output_power": 7.224731402955694e-07, "reference_active": true, "delta_db": -1.8050486230497584}], "active_band_count": 3, "active_band_mae_db": 1.0270561691456663}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.042588377582574e-05, "log_rms_mae_db": 2.0785020630718294, "rms_envelope_nrmse": 0.24226430795314916, "envelope_correlation": 0.9525411095753689, "rms_gain_error_db": -0.1654086736001535, "reference_mean_power": 9.196537764000522e-05, "output_mean_power": 8.852857857265748e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3593984658429913e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0478584079151807e-05, "output_power": 1.97399763623013e-05, "reference_active": true, "delta_db": -0.15953297265774632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.015542002555929e-06, "output_power": 5.831115559851113e-06, "reference_active": true, "delta_db": -0.1352311526677244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6245635384567205e-06, "output_power": 2.2690864128828182e-06, "reference_active": true, "delta_db": 1.4511433355060397}], "active_band_count": 3, "active_band_mae_db": 0.5819691536105035}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6202941099492215e-05, "log_rms_mae_db": 1.4876460889662735, "rms_envelope_nrmse": 0.16107114047718707, "envelope_correlation": 0.9773614383184378, "rms_gain_error_db": -0.1655144017948827, "reference_mean_power": 9.242233001846982e-05, "output_mean_power": 8.896628848353508e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9023164132494484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.934413521065944e-05, "output_power": 1.9388358540571918e-05, "reference_active": true, "delta_db": 0.009917232090197348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.243887754162173e-06, "output_power": 3.472666003150265e-06, "reference_active": true, "delta_db": 0.2959719716729464}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3065359453687705e-07, "output_power": 1.8590759739119004e-07, "reference_active": true, "delta_db": 1.531757757046899}], "active_band_count": 3, "active_band_mae_db": 0.6125489869366809}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.535256840646245e-05, "log_rms_mae_db": 2.3818203999336043, "rms_envelope_nrmse": 0.4522103658550504, "envelope_correlation": 0.8880600021959394, "rms_gain_error_db": 0.6930944384285695, "reference_mean_power": 2.9383590737924492e-05, "output_mean_power": 3.446785920472784e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.113292932890967e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.65042891211228e-06, "output_power": 9.934370458250554e-06, "reference_active": true, "delta_db": 0.12593735109214627}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.01875786446688e-06, "output_power": 4.9618676313408646e-06, "reference_active": true, "delta_db": 0.9155333417937834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.253460098772285e-07, "output_power": 1.535086728878897e-06, "reference_active": true, "delta_db": 3.2558769080749146}], "active_band_count": 3, "active_band_mae_db": 1.4324492003202813}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9034899521347454e-05, "log_rms_mae_db": 2.18409066579015, "rms_envelope_nrmse": 0.3887651900891888, "envelope_correlation": 0.9139542818678924, "rms_gain_error_db": 0.6930811918537729, "reference_mean_power": 2.9530847534057242e-05, "output_mean_power": 3.46404903150123e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9758537203809177e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.950101722919118e-06, "output_power": 1.0836323325396088e-05, "reference_active": true, "delta_db": 0.37054433986619884}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0870738801319347e-06, "output_power": 2.9128636520803046e-06, "reference_active": true, "delta_db": 1.4478233337292754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.171675766600786e-08, "output_power": 1.3158230496520485e-07, "reference_active": true, "delta_db": 3.287943877702734}], "active_band_count": 3, "active_band_mae_db": 1.7021038504327362}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8641894685987068, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b33fb6836aa0d197305d70db69f80ee758204257d77829c19e41a7afade46b87", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9522993208587894, "standard_deviation": 0.001976899576687952, "minimum": 0.9504619739443171, "maximum": 0.9564772221327308, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0571805755624726, "standard_deviation": 0.03940921508495328, "minimum": 1.9813645411403717, "maximum": 2.1173551350749507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47931259437030727, "standard_deviation": 0.21613566958860553, "minimum": 0.2009634140420361, "maximum": 0.871253643226054, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24314887420304593, "standard_deviation": 0.004670746852310943, "minimum": 0.23344370698996264, "maximum": 0.24778376989477408, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9771063482121667, "standard_deviation": 0.0014089746268716528, "minimum": 0.97471463480294, "maximum": 0.9790469066435805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.501901773880739, "standard_deviation": 0.03973875887433158, "minimum": 1.4488709415446095, "maximum": 1.569743660993047, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6127555645464092, "standard_deviation": 0.2119517131941037, "minimum": 0.25380591000429503, "maximum": 0.9841343027618619, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1622654382675341, "standard_deviation": 0.004795393456288836, "minimum": 0.15529371763891875, "maximum": 0.16997098117663798, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8921814448383384, "standard_deviation": 0.007192553299884812, "minimum": 0.8807624854042606, "maximum": 0.904347162606489, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.418151371089068, "standard_deviation": 0.030010030079306554, "minimum": 2.3818203999336043, "maximum": 2.4629252738349363, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.048368811910008, "standard_deviation": 0.3373565705711318, "minimum": 0.6416380208574982, "maximum": 1.4910674450278918, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.4508010063481142, "standard_deviation": 0.007424687533436448, "minimum": 0.4406248239829509, "maximum": 0.4645046202302071, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9172736093179154, "standard_deviation": 0.005724397559096756, "minimum": 0.9088996513189334, "maximum": 0.9265970918976302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.2002313820939334, "standard_deviation": 0.019319294826017495, "minimum": 2.1710145165644383, "maximum": 2.2312050173843505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1661324641585256, "standard_deviation": 0.3289648326090926, "minimum": 0.8376187087164296, "maximum": 1.7021038504327362, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3884189716757124, "standard_deviation": 0.004747103928545337, "minimum": 0.3821611771303471, "maximum": 0.39802464716559005, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9910436508114471, "standard_deviation": 0.0017880508490850414, "minimum": 0.9877812643378497, "maximum": 0.9925427684258212, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.81086661979816, "standard_deviation": 0.027500557631989574, "minimum": 1.7687774502218758, "maximum": 1.8508794655913965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7768339634076848, "standard_deviation": 0.2068967398050452, "minimum": 0.4264606315163935, "maximum": 1.0698547528296798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11047815950991091, "standard_deviation": 0.01051450483438642, "minimum": 0.10146580560983028, "maximum": 0.12914898649717013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973113095928163, "standard_deviation": 0.0007848564920209141, "minimum": 0.9961070260861254, "maximum": 0.998129262629057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0610105986852303, "standard_deviation": 0.038590633170793295, "minimum": 0.9859534454309448, "maximum": 1.1167905266247309, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8902004151092304, "standard_deviation": 0.27963173503832434, "minimum": 0.46020364423027776, "maximum": 1.2857967342327552, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06189066700763251, "standard_deviation": 0.00796440522338907, "minimum": 0.05361303796467008, "maximum": 0.07400920677427544, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4354604201151777, "standard_deviation": 0.05455284754415076, "minimum": 0.3451195160457894, "maximum": 0.5201530624018124, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.948356068127347, "standard_deviation": 0.008577301966564385, "minimum": 5.932185506010479, "maximum": 5.957060933234168, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5264352747177782, "standard_deviation": 0.017065314770898732, "minimum": 0.5049211549829942, "maximum": 0.5623162143178225, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10087150661443466, "standard_deviation": 0.026265617549852568, "minimum": -0.13201839876388083, "maximum": -0.058572228634378755, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9928275927493593, "standard_deviation": 0.0010835065765303746, "minimum": 0.9905380411359356, "maximum": 0.9938962048056798, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2342179084380507, "standard_deviation": 0.04301203805009067, "minimum": 1.1474291903909468, "maximum": 1.2943184455414551, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.21802622848991143, "standard_deviation": 0.0668360555749194, "minimum": 0.1448624447333525, "maximum": 0.322427943648979, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08477312322315675, "standard_deviation": 0.0061824432978538745, "minimum": 0.07856922630308283, "maximum": 0.09754361276436936, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977082956864355, "standard_deviation": 0.0005280959870468915, "minimum": 0.9971442437802194, "maximum": 0.9985314020361179, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6611377805027749, "standard_deviation": 0.04178190650465059, "minimum": 0.5896512680107934, "maximum": 0.7029733282944195, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.26944389129096263, "standard_deviation": 0.15307845124334282, "minimum": 0.1577966624078497, "maximum": 0.5326258421663342, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04687385862177485, "standard_deviation": 0.005450138417409722, "minimum": 0.03854756911385426, "maximum": 0.05296489664338152, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0018", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.9064891338348389, "predicted_mos": 1.9064891338348389, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "33a5533d1890c048de0993e84a8bd6ee9bee780b83fb59c5e076cc10ea955345", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7271074997261167, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1516424797446962, "全帯域_10ms_rms_envelope_nrmse": 0.0783160140571424, "全帯域_10ms_envelope_correlation": 0.9939161210011187, "全帯域_10ms_modulation_mae_db": 0.5771479018440521, "全帯域_40ms_log_rms_mae_db": 0.6048001596735433, "全帯域_40ms_rms_envelope_nrmse": 0.04128440534014033, "全帯域_40ms_envelope_correlation": 0.9982602609004644, "全帯域_40ms_modulation_mae_db": 0.6217028787351689, "300–1000 Hz_10ms_log_rms_mae_db": 2.062239904016516, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11284372718051933, "300–1000 Hz_10ms_envelope_correlation": 0.9906045655645636, "300–1000 Hz_10ms_modulation_mae_db": 0.9229615210053841, "300–1000 Hz_40ms_log_rms_mae_db": 1.2925110798016388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061676196379263144, "300–1000 Hz_40ms_envelope_correlation": 0.9972266673273277, "300–1000 Hz_40ms_modulation_mae_db": 0.8685210519060468, "1000–3000 Hz_10ms_log_rms_mae_db": 2.497211499822893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.27710308620350693, "1000–3000 Hz_10ms_envelope_correlation": 0.9383238155184833, "1000–3000 Hz_10ms_modulation_mae_db": 0.6913069176774463, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9139390566835186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19234650561696645, "1000–3000 Hz_40ms_envelope_correlation": 0.9682678816009959, "1000–3000 Hz_40ms_modulation_mae_db": 0.8842211488468815, "3000–6000 Hz_10ms_log_rms_mae_db": 3.1078201918166166, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.456958558439239, "3000–6000 Hz_10ms_envelope_correlation": 0.8901572656635753, "3000–6000 Hz_10ms_modulation_mae_db": 0.8222761783730167, "3000–6000 Hz_40ms_log_rms_mae_db": 2.851385720984908, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3900992288314771, "3000–6000 Hz_40ms_envelope_correlation": 0.9180077505073201, "3000–6000 Hz_40ms_modulation_mae_db": 0.7244866523001963, "mr_stft_db": 6.4332476548616455, "band_mae_db": 0.586493049392215, "rms_gain_error_db": -0.09390337888525115, "output_peak": 0.5008929379516684}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0929122433549991, "全帯域_10ms_rms_envelope_nrmse": 0.07863657736488702, "全帯域_10ms_envelope_correlation": 0.9938915340738003, "全帯域_10ms_modulation_mae_db": 0.3803791050640954, "全帯域_40ms_log_rms_mae_db": 0.5442416011471392, "全帯域_40ms_rms_envelope_nrmse": 0.040190355887553465, "全帯域_40ms_envelope_correlation": 0.9983960739523888, "全帯域_40ms_modulation_mae_db": 0.46550473509941465, "300–1000 Hz_10ms_log_rms_mae_db": 1.9872492306793597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09858803558062328, "300–1000 Hz_10ms_envelope_correlation": 0.9930903340480376, "300–1000 Hz_10ms_modulation_mae_db": 0.6615200275908384, "300–1000 Hz_40ms_log_rms_mae_db": 1.2278046652884074, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05507192386444831, "300–1000 Hz_40ms_envelope_correlation": 0.9980789274985999, "300–1000 Hz_40ms_modulation_mae_db": 0.9158509766883407, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5190529229424783, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2931342377806653, "1000–3000 Hz_10ms_envelope_correlation": 0.9361378177537004, "1000–3000 Hz_10ms_modulation_mae_db": 0.676671197774394, "1000–3000 Hz_40ms_log_rms_mae_db": 1.938126302513745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19790627349429019, "1000–3000 Hz_40ms_envelope_correlation": 0.9692109696489253, "1000–3000 Hz_40ms_modulation_mae_db": 0.8322614928686596, "3000–6000 Hz_10ms_log_rms_mae_db": 3.14604154795229, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4585055610308701, "3000–6000 Hz_10ms_envelope_correlation": 0.8929185163258131, "3000–6000 Hz_10ms_modulation_mae_db": 0.9000009831510885, "3000–6000 Hz_40ms_log_rms_mae_db": 2.852430295075606, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3947131783817095, "3000–6000 Hz_40ms_envelope_correlation": 0.9192011516618875, "3000–6000 Hz_40ms_modulation_mae_db": 1.0057332280261055, "mr_stft_db": 6.444569729116856, "band_mae_db": 0.5897704480866215, "rms_gain_error_db": -0.11142747012337238, "output_peak": 0.5173021551705511}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1458378854342994, "全帯域_10ms_rms_envelope_nrmse": 0.07682673082918667, "全帯域_10ms_envelope_correlation": 0.9940911583638362, "全帯域_10ms_modulation_mae_db": 0.2914669773298529, "全帯域_40ms_log_rms_mae_db": 0.582354069790539, "全帯域_40ms_rms_envelope_nrmse": 0.0419169725707172, "全帯域_40ms_envelope_correlation": 0.9981159151319309, "全帯域_40ms_modulation_mae_db": 0.3713710419127478, "300–1000 Hz_10ms_log_rms_mae_db": 2.0051543715933478, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09305856377638476, "300–1000 Hz_10ms_envelope_correlation": 0.9936985754075234, "300–1000 Hz_10ms_modulation_mae_db": 0.6163188904741913, "300–1000 Hz_40ms_log_rms_mae_db": 1.1936185474152687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05367440555334748, "300–1000 Hz_40ms_envelope_correlation": 0.9979362119989076, "300–1000 Hz_40ms_modulation_mae_db": 0.9030772475237049, "1000–3000 Hz_10ms_log_rms_mae_db": 2.547804828652974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.27955786725132864, "1000–3000 Hz_10ms_envelope_correlation": 0.940570071034674, "1000–3000 Hz_10ms_modulation_mae_db": 0.7206469420070091, "1000–3000 Hz_40ms_log_rms_mae_db": 1.889674834798658, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18270308674892996, "1000–3000 Hz_40ms_envelope_correlation": 0.9730633135800435, "1000–3000 Hz_40ms_modulation_mae_db": 0.8881283542936632, "3000–6000 Hz_10ms_log_rms_mae_db": 3.1427425179914446, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.45594685665793194, "3000–6000 Hz_10ms_envelope_correlation": 0.8809068118711669, "3000–6000 Hz_10ms_modulation_mae_db": 0.7751037456574656, "3000–6000 Hz_40ms_log_rms_mae_db": 2.862948966341654, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39511583018384705, "3000–6000 Hz_40ms_envelope_correlation": 0.9070393781605539, "3000–6000 Hz_40ms_modulation_mae_db": 0.8519458640491763, "mr_stft_db": 6.426436834538177, "band_mae_db": 0.4257382211167746, "rms_gain_error_db": -0.05155812406138351, "output_peak": 0.5124322562731806}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2142646557695538, "全帯域_10ms_rms_envelope_nrmse": 0.08379436219329281, "全帯域_10ms_envelope_correlation": 0.9930190750873337, "全帯域_10ms_modulation_mae_db": 0.2413510306774318, "全帯域_40ms_log_rms_mae_db": 0.6426124963227424, "全帯域_40ms_rms_envelope_nrmse": 0.04576668103197362, "全帯域_40ms_envelope_correlation": 0.9978383829513137, "全帯域_40ms_modulation_mae_db": 0.2487780011419354, "300–1000 Hz_10ms_log_rms_mae_db": 2.022025515120053, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11797062303955788, "300–1000 Hz_10ms_envelope_correlation": 0.9898455323034726, "300–1000 Hz_10ms_modulation_mae_db": 0.36743661523882615, "300–1000 Hz_40ms_log_rms_mae_db": 1.282124317154688, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06934969920498893, "300–1000 Hz_40ms_envelope_correlation": 0.9966041745287474, "300–1000 Hz_40ms_modulation_mae_db": 0.5331639627491239, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5458034570716803, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.29558712453887226, "1000–3000 Hz_10ms_envelope_correlation": 0.9312947431053988, "1000–3000 Hz_10ms_modulation_mae_db": 1.0358034050989602, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9818747439308873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.20329944198219244, "1000–3000 Hz_40ms_envelope_correlation": 0.9648716943064815, "1000–3000 Hz_40ms_modulation_mae_db": 1.2525052594983304, "3000–6000 Hz_10ms_log_rms_mae_db": 3.148651564563704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.45893331657499414, "3000–6000 Hz_10ms_envelope_correlation": 0.8883870815389786, "3000–6000 Hz_10ms_modulation_mae_db": 0.7489320931812545, "3000–6000 Hz_40ms_log_rms_mae_db": 2.8554141708215317, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3958268332245186, "3000–6000 Hz_40ms_envelope_correlation": 0.9147199394735032, "3000–6000 Hz_40ms_modulation_mae_db": 0.8027920920768818, "mr_stft_db": 6.428860442276142, "band_mae_db": 0.4561995843707896, "rms_gain_error_db": -0.1013665112584493, "output_peak": 0.5072988933916015}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1792350955714732, "全帯域_10ms_rms_envelope_nrmse": 0.0776997670606425, "全帯域_10ms_envelope_correlation": 0.9940704452455653, "全帯域_10ms_modulation_mae_db": 0.2301863427419357, "全帯域_40ms_log_rms_mae_db": 0.5902730372667198, "全帯域_40ms_rms_envelope_nrmse": 0.041957209279292786, "全帯域_40ms_envelope_correlation": 0.9982736229354324, "全帯域_40ms_modulation_mae_db": 0.376872594172278, "300–1000 Hz_10ms_log_rms_mae_db": 2.0364703911368975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09606673598680444, "300–1000 Hz_10ms_envelope_correlation": 0.9937169382505794, "300–1000 Hz_10ms_modulation_mae_db": 0.736674525250948, "300–1000 Hz_40ms_log_rms_mae_db": 1.1798973835150686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061485704585802534, "300–1000 Hz_40ms_envelope_correlation": 0.9977575977107537, "300–1000 Hz_40ms_modulation_mae_db": 1.0358518431179713, "1000–3000 Hz_10ms_log_rms_mae_db": 2.478712505060483, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2708479383628799, "1000–3000 Hz_10ms_envelope_correlation": 0.9446856097768653, "1000–3000 Hz_10ms_modulation_mae_db": 0.9905303540076204, "1000–3000 Hz_40ms_log_rms_mae_db": 1.868056510078261, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17800227588918557, "1000–3000 Hz_40ms_envelope_correlation": 0.9748530691507163, "1000–3000 Hz_40ms_modulation_mae_db": 1.1265052566161924, "3000–6000 Hz_10ms_log_rms_mae_db": 3.1946885677280012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.46318028938853095, "3000–6000 Hz_10ms_envelope_correlation": 0.8881417086186618, "3000–6000 Hz_10ms_modulation_mae_db": 0.7637450837778467, "3000–6000 Hz_40ms_log_rms_mae_db": 2.970169905763035, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3979287365508882, "3000–6000 Hz_40ms_envelope_correlation": 0.9154573898013345, "3000–6000 Hz_40ms_modulation_mae_db": 0.9599973030595691, "mr_stft_db": 6.449850329131482, "band_mae_db": 0.5613919480680428, "rms_gain_error_db": -0.12394276491126309, "output_peak": 0.5164170537450894}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1717228088419254, "全帯域_10ms_rms_envelope_nrmse": 0.0856743714803763, "全帯域_10ms_envelope_correlation": 0.9926541494255192, "全帯域_10ms_modulation_mae_db": 0.4334024934628758, "全帯域_40ms_log_rms_mae_db": 0.6101086385060084, "全帯域_40ms_rms_envelope_nrmse": 0.05202385207522682, "全帯域_40ms_envelope_correlation": 0.997075231237916, "全帯域_40ms_modulation_mae_db": 0.8390994093207865, "300–1000 Hz_10ms_log_rms_mae_db": 2.147819678677362, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11273431193443323, "300–1000 Hz_10ms_envelope_correlation": 0.9907012617630617, "300–1000 Hz_10ms_modulation_mae_db": 0.6097993490782688, "300–1000 Hz_40ms_log_rms_mae_db": 1.3567657023312991, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0635400498753288, "300–1000 Hz_40ms_envelope_correlation": 0.997145666799759, "300–1000 Hz_40ms_modulation_mae_db": 0.6229987199901927, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5107112347886997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26957587832016383, "1000–3000 Hz_10ms_envelope_correlation": 0.9416007491310456, "1000–3000 Hz_10ms_modulation_mae_db": 0.5917766359047386, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9179245641968328, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18778120486732267, "1000–3000 Hz_40ms_envelope_correlation": 0.9695689008117091, "1000–3000 Hz_40ms_modulation_mae_db": 0.6456928434431174, "3000–6000 Hz_10ms_log_rms_mae_db": 3.1298036891881154, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.46632463359228327, "3000–6000 Hz_10ms_envelope_correlation": 0.8881293507497847, "3000–6000 Hz_10ms_modulation_mae_db": 0.9762322488002019, "3000–6000 Hz_40ms_log_rms_mae_db": 2.859550481826943, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.40792314628150084, "3000–6000 Hz_40ms_envelope_correlation": 0.9122009488532462, "3000–6000 Hz_40ms_modulation_mae_db": 0.8559330013670904, "mr_stft_db": 6.444780502574287, "band_mae_db": 0.4910757147418133, "rms_gain_error_db": -0.04935012543719146, "output_peak": 0.560385710461174}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1996988575225802, "全帯域_10ms_rms_envelope_nrmse": 0.07941142665241216, "全帯域_10ms_envelope_correlation": 0.9937405753090263, "全帯域_10ms_modulation_mae_db": 0.30565192560602467, "全帯域_40ms_log_rms_mae_db": 0.6475103293992197, "全帯域_40ms_rms_envelope_nrmse": 0.04369040736389898, "全帯域_40ms_envelope_correlation": 0.9980342113591638, "全帯域_40ms_modulation_mae_db": 0.2464104768525206, "300–1000 Hz_10ms_log_rms_mae_db": 2.028964576679746, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11314054317328631, "300–1000 Hz_10ms_envelope_correlation": 0.9906559877919655, "300–1000 Hz_10ms_modulation_mae_db": 0.5399070473481569, "300–1000 Hz_40ms_log_rms_mae_db": 1.254364463103789, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05972787537078726, "300–1000 Hz_40ms_envelope_correlation": 0.9975267048992136, "300–1000 Hz_40ms_modulation_mae_db": 0.6806804680082176, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5508879166535117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.31453818735606676, "1000–3000 Hz_10ms_envelope_correlation": 0.9252838247242821, "1000–3000 Hz_10ms_modulation_mae_db": 0.6108597653484692, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9606113770720537, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.2063752955269265, "1000–3000 Hz_40ms_envelope_correlation": 0.9656935993885831, "1000–3000 Hz_40ms_modulation_mae_db": 0.748933214582618, "3000–6000 Hz_10ms_log_rms_mae_db": 3.1496494563238207, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4625968996533703, "3000–6000 Hz_10ms_envelope_correlation": 0.8824251972087671, "3000–6000 Hz_10ms_modulation_mae_db": 0.7528615443423102, "3000–6000 Hz_40ms_log_rms_mae_db": 2.900365319338044, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39555918946294405, "3000–6000 Hz_40ms_envelope_correlation": 0.9112696236100997, "3000–6000 Hz_40ms_modulation_mae_db": 0.6843114902456439, "mr_stft_db": 6.439361536178261, "band_mae_db": 0.4939017688502744, "rms_gain_error_db": -0.09850102692832875, "output_peak": 0.530966818497906}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1455143907212026, "全帯域_10ms_rms_envelope_nrmse": 0.07822794710324932, "全帯域_10ms_envelope_correlation": 0.9940055449091826, "全帯域_10ms_modulation_mae_db": 0.5362067046189707, "全帯域_40ms_log_rms_mae_db": 0.5939213207971829, "全帯域_40ms_rms_envelope_nrmse": 0.042368664916930934, "全帯域_40ms_envelope_correlation": 0.9982574968871326, "全帯域_40ms_modulation_mae_db": 0.6093100003308173, "300–1000 Hz_10ms_log_rms_mae_db": 1.9849343745513177, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11177129149302634, "300–1000 Hz_10ms_envelope_correlation": 0.9911817422691872, "300–1000 Hz_10ms_modulation_mae_db": 0.8895172450334518, "300–1000 Hz_40ms_log_rms_mae_db": 1.1719069332225538, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06458858794642248, "300–1000 Hz_40ms_envelope_correlation": 0.9974303925924433, "300–1000 Hz_40ms_modulation_mae_db": 1.129822979947503, "1000–3000 Hz_10ms_log_rms_mae_db": 2.5514426340327567, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2870001160283561, "1000–3000 Hz_10ms_envelope_correlation": 0.9337952269064295, "1000–3000 Hz_10ms_modulation_mae_db": 0.4782583780843013, "1000–3000 Hz_40ms_log_rms_mae_db": 1.9635632107812624, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19851711695933638, "1000–3000 Hz_40ms_envelope_correlation": 0.965834970863099, "1000–3000 Hz_40ms_modulation_mae_db": 0.4695809437766658, "3000–6000 Hz_10ms_log_rms_mae_db": 3.1362340070749837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.47019569206076367, "3000–6000 Hz_10ms_envelope_correlation": 0.8773572038541354, "3000–6000 Hz_10ms_modulation_mae_db": 0.9332274186221463, "3000–6000 Hz_40ms_log_rms_mae_db": 2.8946811001508754, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.4086962875706836, "3000–6000 Hz_40ms_envelope_correlation": 0.9039800906181079, "3000–6000 Hz_40ms_modulation_mae_db": 0.9351929222328291, "mr_stft_db": 6.4215706395270695, "band_mae_db": 0.4967742643033802, "rms_gain_error_db": -0.13030485628644822, "output_peak": 0.5179302955121021}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.590550790831061, "spectral_convergence": 0.24427216092713336, "spectral_convergence_status": "ok", "amplitude_floor": 1.9267471569597447e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.2708314159765175, "spectral_convergence": 0.228605489072728, "spectral_convergence_status": "ok", "amplitude_floor": 1.271916808390449e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.438360757777356, "spectral_convergence": 0.23515016485962348, "spectral_convergence_status": "ok", "amplitude_floor": 1.1644398478092437e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001373055391651405, "output_power": 0.001361778604540782, "delta_db": -0.03581551332817969, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0033897235254818038, "output_power": 0.003252936423719429, "delta_db": -0.1788870208466464, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002841140646523882, "output_power": 0.00030837189714107637, "delta_db": 0.35582059249274184, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.671358771236327e-05, "output_power": 8.10010926461009e-05, "delta_db": -0.29596277980814667, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6620686952686656e-05, "output_power": 3.680112202904912e-05, "delta_db": 1.4064180172761933, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.533501052724457e-05, "output_power": 3.375409950614492e-05, "delta_db": 1.2460543726013826, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.586493049392215, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09390337888525115, "pitch": null, "output_rms_dbfs": -22.945944843559957, "output_peak": 0.5008929379516684, "reference_rms_dbfs": -22.85204146467471, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.4332476548616455}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023735521522065317, "log_rms_mae_db": 1.1516424797446962, "rms_envelope_nrmse": 0.0783160140571424, "envelope_correlation": 0.9939161210011187, "rms_gain_error_db": -0.09391917315977691, "reference_mean_power": 0.0051897925018650195, "output_mean_power": 0.0050787645195551675}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6580095170174236e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00099786392963893, "output_power": 0.0009759725607940078, "reference_active": true, "delta_db": -0.09633716451527605}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00025981544310873085, "output_power": 0.0002357708687494174, "reference_active": true, "delta_db": -0.42174817682169236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.167033379867013e-05, "output_power": 3.151310111391643e-05, "reference_active": true, "delta_db": -1.2133583641951877}], "active_band_count": 3, "active_band_mae_db": 0.5771479018440521}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022053657198928817, "log_rms_mae_db": 0.6048001596735433, "rms_envelope_nrmse": 0.04128440534014033, "envelope_correlation": 0.9982602609004644, "rms_gain_error_db": -0.09384472381661882, "reference_mean_power": 0.0052155458817555326, "output_mean_power": 0.00510405443972379}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3916979728964525e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008985624617991317, "output_power": 0.000881249614770535, "reference_active": true, "delta_db": -0.08449331467544523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001435836821660598, "output_power": 0.000128961500471603, "reference_active": true, "delta_db": -0.46645008986531983}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3776224720397365e-06, "output_power": 1.756816386649197e-06, "reference_active": true, "delta_db": -1.3141652316647416}], "active_band_count": 3, "active_band_mae_db": 0.6217028787351689}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018677846007938983, "log_rms_mae_db": 2.062239904016516, "rms_envelope_nrmse": 0.11284372718051933, "envelope_correlation": 0.9906045655645636, "rms_gain_error_db": -0.16116662455885455, "reference_mean_power": 0.0023293656178310037, "output_mean_power": 0.002244507176848043}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5191576623571145e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006490500325829311, "output_power": 0.0006148852479353002, "reference_active": true, "delta_db": -0.23484102294807865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013228128531435308, "output_power": 0.00011280847340770503, "reference_active": true, "delta_db": -0.6915668396556593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.716163060824064e-05, "output_power": 1.1228219132054143e-05, "reference_active": true, "delta_db": -1.8424767004124143}], "active_band_count": 3, "active_band_mae_db": 0.9229615210053841}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017927232752442356, "log_rms_mae_db": 1.2925110798016388, "rms_envelope_nrmse": 0.061676196379263144, "envelope_correlation": 0.9972266673273277, "rms_gain_error_db": -0.16110598506318766, "reference_mean_power": 0.002341010543733861, "output_mean_power": 0.0022557593762489695}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.424681480871225e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006148042140063018, "output_power": 0.000578514250176377, "reference_active": true, "delta_db": -0.2642277446854594}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.176867672883116e-05, "output_power": 5.334086116606514e-05, "reference_active": true, "delta_db": -0.6370827462637005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.094779276972058e-06, "output_power": 7.394372103751804e-07, "reference_active": true, "delta_db": -1.7042526647689802}], "active_band_count": 3, "active_band_mae_db": 0.8685210519060468}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.042588377582574e-05, "log_rms_mae_db": 2.497211499822893, "rms_envelope_nrmse": 0.27710308620350693, "envelope_correlation": 0.9383238155184833, "rms_gain_error_db": -0.008495047381102447, "reference_mean_power": 9.196537764000522e-05, "output_mean_power": 9.178566394665436e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3593984658429913e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0478584079151807e-05, "output_power": 2.0118312044038456e-05, "reference_active": true, "delta_db": -0.07708385585564025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.015542002555929e-06, "output_power": 7.126013851777292e-06, "reference_active": true, "delta_db": 0.7357189867955521}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6245635384567205e-06, "output_power": 2.171943346310423e-06, "reference_active": true, "delta_db": 1.2611179103811465}], "active_band_count": 3, "active_band_mae_db": 0.6913069176774463}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6202941099492215e-05, "log_rms_mae_db": 1.9139390566835186, "rms_envelope_nrmse": 0.19234650561696645, "envelope_correlation": 0.9682678816009959, "rms_gain_error_db": -0.008487136566645036, "reference_mean_power": 9.242233001846982e-05, "output_mean_power": 9.2241891395297e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9023164132494484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.934413521065944e-05, "output_power": 1.9838878614500486e-05, "reference_active": true, "delta_db": 0.10967801035129995}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.243887754162173e-06, "output_power": 4.202720433210794e-06, "reference_active": true, "delta_db": 1.1246468327881716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3065359453687705e-07, "output_power": 1.8111533473387886e-07, "reference_active": true, "delta_db": 1.418338603401173}], "active_band_count": 3, "active_band_mae_db": 0.8842211488468815}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.535256840646245e-05, "log_rms_mae_db": 3.1078201918166166, "rms_envelope_nrmse": 0.456958558439239, "envelope_correlation": 0.8901572656635753, "rms_gain_error_db": 0.8886270792033215, "reference_mean_power": 2.9383590737924492e-05, "output_mean_power": 3.605517221976644e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.113292932890967e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.65042891211228e-06, "output_power": 9.074113984003142e-06, "reference_active": true, "delta_db": -0.2674238560552023}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.01875786446688e-06, "output_power": 3.5680347178602314e-06, "reference_active": true, "delta_db": -0.5166276863988885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.253460098772285e-07, "output_power": 1.068619083548176e-06, "reference_active": true, "delta_db": 1.6827769926649594}], "active_band_count": 3, "active_band_mae_db": 0.8222761783730167}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9034899521347454e-05, "log_rms_mae_db": 2.851385720984908, "rms_envelope_nrmse": 0.3900992288314771, "envelope_correlation": 0.9180077505073201, "rms_gain_error_db": 0.8885818174898545, "reference_mean_power": 2.9530847534057242e-05, "output_mean_power": 3.6235486213525585e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9758537203809177e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.950101722919118e-06, "output_power": 9.158162577819867e-06, "reference_active": true, "delta_db": -0.36019171736942923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0870738801319347e-06, "output_power": 2.050596319316983e-06, "reference_active": true, "delta_db": -0.07657649377967758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.171675766600786e-08, "output_power": 9.20602768487832e-08, "reference_active": true, "delta_db": 1.736691745751482}], "active_band_count": 3, "active_band_mae_db": 0.7244866523001963}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.845941441514028, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "eca1eb7db57dbec08a1fcbc4a79d88bed8bd31cc905020c4afb84b258c89ec9c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9364614822438599, "standard_deviation": 0.006247811297578902, "minimum": 0.9252838247242821, "maximum": 0.9446856097768653, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.525203374878185, "standard_deviation": 0.0279698300092037, "minimum": 2.478712505060483, "maximum": 2.5514426340327567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7244816994878674, "standard_deviation": 0.1935571928636357, "minimum": 0.4782583780843013, "maximum": 1.0358034050989602, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.28591805448023, "standard_deviation": 0.01501512498873939, "minimum": 0.26957587832016383, "maximum": 0.31453818735606676, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9689205499188193, "standard_deviation": 0.003578012082265701, "minimum": 0.9648716943064815, "maximum": 0.9748530691507163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.9292213250069024, "standard_deviation": 0.03906515736365093, "minimum": 1.868056510078261, "maximum": 1.9818747439308873, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.855978564240766, "standard_deviation": 0.25008392184418043, "minimum": 0.4695809437766658, "maximum": 1.2525052594983304, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19336640013564377, "standard_deviation": 0.009978793018879162, "minimum": 0.17800227588918557, "maximum": 0.2063752955269265, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8860528919788604, "standard_deviation": 0.005255789456591365, "minimum": 0.8773572038541354, "maximum": 0.8929185163258131, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 3.144453942829872, "standard_deviation": 0.02448850862059791, "minimum": 3.1078201918166166, "maximum": 3.1946885677280012, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8340474119881662, "standard_deviation": 0.09007470929214874, "minimum": 0.7489320931812545, "maximum": 0.9762322488002019, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.46158022592474796, "standard_deviation": 0.004926369894667191, "minimum": 0.45594685665793194, "maximum": 0.47019569206076367, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9127345340857567, "standard_deviation": 0.005243846590444181, "minimum": 0.9039800906181079, "maximum": 0.9192011516618875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.8808682450378247, "standard_deviation": 0.04079059971369849, "minimum": 2.851385720984908, "maximum": 2.970169905763035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8525490691696865, "standard_deviation": 0.11277788102495502, "minimum": 0.6843114902456439, "maximum": 1.0057332280261055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3982328038109461, "standard_deviation": 0.006597347132760125, "minimum": 0.3900992288314771, "maximum": 0.4086962875706836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916868671747988, "standard_deviation": 0.0015578854824205232, "minimum": 0.9898455323034726, "maximum": 0.9937169382505794, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.034357255306825, "standard_deviation": 0.0525783460339849, "minimum": 1.9849343745513177, "maximum": 2.147819678677362, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6680169026275081, "standard_deviation": 0.18190965216029548, "minimum": 0.36743661523882615, "maximum": 0.9229615210053841, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10702172902057944, "standard_deviation": 0.009505748769799148, "minimum": 0.09305856377638476, "maximum": 0.11797062303955788, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974632929194691, "standard_deviation": 0.00047649608657361, "minimum": 0.9966041745287474, "maximum": 0.9980789274985999, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2448741364790892, "standard_deviation": 0.06406618962517274, "minimum": 1.1719069332225538, "maximum": 1.3567657023312991, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8362459062413876, "standard_deviation": 0.20681556351221844, "minimum": 0.5331639627491239, "maximum": 1.129822979947503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06113930534754862, "standard_deviation": 0.005071065230716686, "minimum": 0.05367440555334748, "maximum": 0.06934969920498893, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.5126681248662389, "standard_deviation": 0.06045709931479784, "minimum": 0.4257382211167746, "maximum": 0.5897704480866215, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.43608470852549, "standard_deviation": 0.010084081418666364, "minimum": 6.4215706395270695, "maximum": 6.449850329131482, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5204532651254091, "standard_deviation": 0.018352506036726557, "minimum": 0.5008929379516684, "maximum": 0.560385710461174, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09504428223646098, "standard_deviation": 0.030187657059349903, "minimum": -0.13030485628644822, "maximum": -0.04935012543719146, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936735754269228, "standard_deviation": 0.000537256028042356, "minimum": 0.9926541494255192, "maximum": 0.9940911583638362, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1626035521200913, "standard_deviation": 0.037711027253623505, "minimum": 1.0929122433549991, "maximum": 1.2142646557695538, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3744740601681549, "standard_deviation": 0.1314256512113996, "minimum": 0.2301863427419357, "maximum": 0.5771479018440521, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07982339959264866, "standard_deviation": 0.0031597514214610884, "minimum": 0.07682673082918667, "maximum": 0.0856743714803763, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980313994194678, "standard_deviation": 0.0004232225267540619, "minimum": 0.997075231237916, "maximum": 0.9983960739523888, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6019777066128869, "standard_deviation": 0.03320699969390872, "minimum": 0.5442416011471392, "maximum": 0.6475103293992197, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47238114219570865, "standard_deviation": 0.20562173835905342, "minimum": 0.2464104768525206, "maximum": 0.8390994093207865, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04364981855821677, "standard_deviation": 0.0037735789989646797, "minimum": 0.040190355887553465, "maximum": 0.05202385207522682, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0018", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.1827166080474854, "predicted_mos": 2.1827166080474854, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "25c4d0439df75a87c02ceaebd442612d7256c021af77bff4c8eb59ac1ea575a1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8299186998046935, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2153520649306009, "全帯域_10ms_rms_envelope_nrmse": 0.07103960358159865, "全帯域_10ms_envelope_correlation": 0.9949592215596017, "全帯域_10ms_modulation_mae_db": 0.36894507758481465, "全帯域_40ms_log_rms_mae_db": 0.6396592317885603, "全帯域_40ms_rms_envelope_nrmse": 0.0357607649866036, "全帯域_40ms_envelope_correlation": 0.9986477270421188, "全帯域_40ms_modulation_mae_db": 0.4240899372442562, "300–1000 Hz_10ms_log_rms_mae_db": 1.6666537206001708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09607536343101557, "300–1000 Hz_10ms_envelope_correlation": 0.993190380451079, "300–1000 Hz_10ms_modulation_mae_db": 0.4630045402027149, "300–1000 Hz_40ms_log_rms_mae_db": 0.9916233307451402, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05170373967588548, "300–1000 Hz_40ms_envelope_correlation": 0.998026529979996, "300–1000 Hz_40ms_modulation_mae_db": 0.3931139933052606, "1000–3000 Hz_10ms_log_rms_mae_db": 1.938364400732502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2490448429339933, "1000–3000 Hz_10ms_envelope_correlation": 0.9500516693478994, "1000–3000 Hz_10ms_modulation_mae_db": 0.5987153942401701, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4193762319878862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16273464687629904, "1000–3000 Hz_40ms_envelope_correlation": 0.9768698283033771, "1000–3000 Hz_40ms_modulation_mae_db": 0.778991646620368, "3000–6000 Hz_10ms_log_rms_mae_db": 2.25516018169224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4493416913891345, "3000–6000 Hz_10ms_envelope_correlation": 0.8876135227913381, "3000–6000 Hz_10ms_modulation_mae_db": 0.7548693830409762, "3000–6000 Hz_40ms_log_rms_mae_db": 2.145581185213099, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38923160495810377, "3000–6000 Hz_40ms_envelope_correlation": 0.912112112642686, "3000–6000 Hz_40ms_modulation_mae_db": 0.9664565206141336, "mr_stft_db": 5.874206491385993, "band_mae_db": 0.3843947588693271, "rms_gain_error_db": -0.060087037113472776, "output_peak": 0.5175699120272388}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2796428997755716, "全帯域_10ms_rms_envelope_nrmse": 0.07970961076291827, "全帯域_10ms_envelope_correlation": 0.9936475231233465, "全帯域_10ms_modulation_mae_db": 0.15985274717508444, "全帯域_40ms_log_rms_mae_db": 0.6477824430580764, "全帯域_40ms_rms_envelope_nrmse": 0.04281760819014026, "全帯域_40ms_envelope_correlation": 0.9980603648506017, "全帯域_40ms_modulation_mae_db": 0.24274742914485406, "300–1000 Hz_10ms_log_rms_mae_db": 1.6812819148288047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09766607909689758, "300–1000 Hz_10ms_envelope_correlation": 0.9931577987786168, "300–1000 Hz_10ms_modulation_mae_db": 0.8441092799752007, "300–1000 Hz_40ms_log_rms_mae_db": 1.0577698027315594, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058452301488796446, "300–1000 Hz_40ms_envelope_correlation": 0.997678747978617, "300–1000 Hz_40ms_modulation_mae_db": 1.1472565565610404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.910431280719949, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23851092736730972, "1000–3000 Hz_10ms_envelope_correlation": 0.9540923119346534, "1000–3000 Hz_10ms_modulation_mae_db": 0.3817914556709801, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4955299462507563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16736250736165706, "1000–3000 Hz_40ms_envelope_correlation": 0.9755537838523037, "1000–3000 Hz_40ms_modulation_mae_db": 0.5868064553513966, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2024957571344017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4474472576659488, "3000–6000 Hz_10ms_envelope_correlation": 0.8951157550685036, "3000–6000 Hz_10ms_modulation_mae_db": 0.711231644956125, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1444731775530514, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38852683575003955, "3000–6000 Hz_40ms_envelope_correlation": 0.9182567397036839, "3000–6000 Hz_40ms_modulation_mae_db": 1.038012632778415, "mr_stft_db": 5.876855491159983, "band_mae_db": 0.4458840960166716, "rms_gain_error_db": -0.06333449499404575, "output_peak": 0.5223437877727821}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.223725334667443, "全帯域_10ms_rms_envelope_nrmse": 0.08997120982909178, "全帯域_10ms_envelope_correlation": 0.9919172881296454, "全帯域_10ms_modulation_mae_db": 0.11581624962770039, "全帯域_40ms_log_rms_mae_db": 0.6414287653595485, "全帯域_40ms_rms_envelope_nrmse": 0.04446742194227799, "全帯域_40ms_envelope_correlation": 0.9979407808649569, "全帯域_40ms_modulation_mae_db": 0.06767090622102105, "300–1000 Hz_10ms_log_rms_mae_db": 1.6561558442900188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11805963714634105, "300–1000 Hz_10ms_envelope_correlation": 0.9897350385355614, "300–1000 Hz_10ms_modulation_mae_db": 0.6448297074456849, "300–1000 Hz_40ms_log_rms_mae_db": 0.9562435838706657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06554081097460149, "300–1000 Hz_40ms_envelope_correlation": 0.9969106438704536, "300–1000 Hz_40ms_modulation_mae_db": 0.5445379057133549, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8870956369204153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25400309806158716, "1000–3000 Hz_10ms_envelope_correlation": 0.9511798260598636, "1000–3000 Hz_10ms_modulation_mae_db": 0.6354805263482733, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4619726015058003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17263444012184279, "1000–3000 Hz_40ms_envelope_correlation": 0.9756936916717588, "1000–3000 Hz_40ms_modulation_mae_db": 0.7728369760762983, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2111352632989236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4404821396506364, "3000–6000 Hz_10ms_envelope_correlation": 0.8997161210277935, "3000–6000 Hz_10ms_modulation_mae_db": 1.3801514583237415, "3000–6000 Hz_40ms_log_rms_mae_db": 2.0830296973746036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3805670632777509, "3000–6000 Hz_40ms_envelope_correlation": 0.9230678565027436, "3000–6000 Hz_40ms_modulation_mae_db": 1.5185999702528565, "mr_stft_db": 5.856711427692134, "band_mae_db": 0.4992993630210996, "rms_gain_error_db": -0.0914395473824483, "output_peak": 0.5225154919418693}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.226003951065667, "全帯域_10ms_rms_envelope_nrmse": 0.07792131228858043, "全帯域_10ms_envelope_correlation": 0.9939553101573301, "全帯域_10ms_modulation_mae_db": 0.32928703975764084, "全帯域_40ms_log_rms_mae_db": 0.6156751753987244, "全帯域_40ms_rms_envelope_nrmse": 0.039129068752008014, "全帯域_40ms_envelope_correlation": 0.9984133359537134, "全帯域_40ms_modulation_mae_db": 0.3724596716676129, "300–1000 Hz_10ms_log_rms_mae_db": 1.740306643867535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12060216468185732, "300–1000 Hz_10ms_envelope_correlation": 0.9892148277841292, "300–1000 Hz_10ms_modulation_mae_db": 0.5782472628430321, "300–1000 Hz_40ms_log_rms_mae_db": 1.03085229934211, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06596753273986598, "300–1000 Hz_40ms_envelope_correlation": 0.9967745306000391, "300–1000 Hz_40ms_modulation_mae_db": 0.564901445622438, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9348552752043295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24477374577952796, "1000–3000 Hz_10ms_envelope_correlation": 0.951665559839958, "1000–3000 Hz_10ms_modulation_mae_db": 0.5119671761879857, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4974604646077267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16386022719279816, "1000–3000 Hz_40ms_envelope_correlation": 0.9765232234093437, "1000–3000 Hz_40ms_modulation_mae_db": 0.42397463548989883, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2256839346806894, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4390413984024471, "3000–6000 Hz_10ms_envelope_correlation": 0.9005980606712862, "3000–6000 Hz_10ms_modulation_mae_db": 0.7929180399389999, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1309758739260944, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38107649795155607, "3000–6000 Hz_40ms_envelope_correlation": 0.9229718750852695, "3000–6000 Hz_40ms_modulation_mae_db": 0.9767573680039275, "mr_stft_db": 5.86671180728429, "band_mae_db": 0.5318146720266249, "rms_gain_error_db": -0.08213630891958605, "output_peak": 0.5445863919171128}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2567769517040168, "全帯域_10ms_rms_envelope_nrmse": 0.08585511932747068, "全帯域_10ms_envelope_correlation": 0.9926557854966004, "全帯域_10ms_modulation_mae_db": 0.23933514528911592, "全帯域_40ms_log_rms_mae_db": 0.6966593695844967, "全帯域_40ms_rms_envelope_nrmse": 0.046909722486631666, "全帯域_40ms_envelope_correlation": 0.9977055870679288, "全帯域_40ms_modulation_mae_db": 0.16663985088145347, "300–1000 Hz_10ms_log_rms_mae_db": 1.62581624380536, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10548524205457413, "300–1000 Hz_10ms_envelope_correlation": 0.991889601535138, "300–1000 Hz_10ms_modulation_mae_db": 0.8337209936785831, "300–1000 Hz_40ms_log_rms_mae_db": 0.9577801602433835, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05491511663542318, "300–1000 Hz_40ms_envelope_correlation": 0.9979453938728429, "300–1000 Hz_40ms_modulation_mae_db": 0.7771676652566423, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8763585682323485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24918466742447665, "1000–3000 Hz_10ms_envelope_correlation": 0.9517461338715723, "1000–3000 Hz_10ms_modulation_mae_db": 0.5398794913631407, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4621809872999367, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17594483047718282, "1000–3000 Hz_40ms_envelope_correlation": 0.9738649940747997, "1000–3000 Hz_40ms_modulation_mae_db": 0.6291821297026211, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1826775646220176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4328430843007298, "3000–6000 Hz_10ms_envelope_correlation": 0.8985375152022584, "3000–6000 Hz_10ms_modulation_mae_db": 0.6717332440781125, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1009913672820244, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3788197310540476, "3000–6000 Hz_40ms_envelope_correlation": 0.9193908489412221, "3000–6000 Hz_40ms_modulation_mae_db": 0.9280754801180113, "mr_stft_db": 5.870651133963432, "band_mae_db": 0.47001643839595086, "rms_gain_error_db": -0.0961822102907972, "output_peak": 0.5213748485425894}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2020022519866533, "全帯域_10ms_rms_envelope_nrmse": 0.08049268455249156, "全帯域_10ms_envelope_correlation": 0.9936085720222985, "全帯域_10ms_modulation_mae_db": 0.1919692443612883, "全帯域_40ms_log_rms_mae_db": 0.6773596103575105, "全帯域_40ms_rms_envelope_nrmse": 0.04407438230339608, "全帯域_40ms_envelope_correlation": 0.9980523162391043, "全帯域_40ms_modulation_mae_db": 0.41397585753311433, "300–1000 Hz_10ms_log_rms_mae_db": 1.633478057350431, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10540757363971726, "300–1000 Hz_10ms_envelope_correlation": 0.9922184279434405, "300–1000 Hz_10ms_modulation_mae_db": 0.6610472252996985, "300–1000 Hz_40ms_log_rms_mae_db": 0.978502263174437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05827298670447197, "300–1000 Hz_40ms_envelope_correlation": 0.99803820290356, "300–1000 Hz_40ms_modulation_mae_db": 0.8307267003950037, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9036813199286522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23951863943647608, "1000–3000 Hz_10ms_envelope_correlation": 0.9552519190133761, "1000–3000 Hz_10ms_modulation_mae_db": 0.43777369194673504, "1000–3000 Hz_40ms_log_rms_mae_db": 1.483312385091871, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16483584922584577, "1000–3000 Hz_40ms_envelope_correlation": 0.9769825495273514, "1000–3000 Hz_40ms_modulation_mae_db": 0.6797146275402115, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2292070090304246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.44890056100157016, "3000–6000 Hz_10ms_envelope_correlation": 0.889139747599996, "3000–6000 Hz_10ms_modulation_mae_db": 1.140159563985001, "3000–6000 Hz_40ms_log_rms_mae_db": 2.141163733209798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3911677328821133, "3000–6000 Hz_40ms_envelope_correlation": 0.9123043041223865, "3000–6000 Hz_40ms_modulation_mae_db": 1.0110430248285718, "mr_stft_db": 5.861865968465428, "band_mae_db": 0.42622346242427384, "rms_gain_error_db": -0.10726078154326392, "output_peak": 0.526645669448557}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2077650452975721, "全帯域_10ms_rms_envelope_nrmse": 0.07892401093387111, "全帯域_10ms_envelope_correlation": 0.9939171697335154, "全帯域_10ms_modulation_mae_db": 0.17021906232386075, "全帯域_40ms_log_rms_mae_db": 0.6090011476319732, "全帯域_40ms_rms_envelope_nrmse": 0.04623887343272624, "全帯域_40ms_envelope_correlation": 0.997911456430628, "全帯域_40ms_modulation_mae_db": 0.1622728008484078, "300–1000 Hz_10ms_log_rms_mae_db": 1.6812292971163316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10597870851548015, "300–1000 Hz_10ms_envelope_correlation": 0.9922171273821324, "300–1000 Hz_10ms_modulation_mae_db": 0.8406145585760788, "300–1000 Hz_40ms_log_rms_mae_db": 0.9933589089960361, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06748469509584439, "300–1000 Hz_40ms_envelope_correlation": 0.997190364541668, "300–1000 Hz_40ms_modulation_mae_db": 0.9149723570496873, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8996262614321262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24634284160285866, "1000–3000 Hz_10ms_envelope_correlation": 0.9506734063477236, "1000–3000 Hz_10ms_modulation_mae_db": 0.7267109084023465, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5185360019188985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17030529315759346, "1000–3000 Hz_40ms_envelope_correlation": 0.9750826730534459, "1000–3000 Hz_40ms_modulation_mae_db": 0.6889538765695988, "3000–6000 Hz_10ms_log_rms_mae_db": 2.225770229438588, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.45201668216149776, "3000–6000 Hz_10ms_envelope_correlation": 0.8922778114759802, "3000–6000 Hz_10ms_modulation_mae_db": 1.616928531137323, "3000–6000 Hz_40ms_log_rms_mae_db": 2.136402087828082, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39866558366767785, "3000–6000 Hz_40ms_envelope_correlation": 0.913175098498267, "3000–6000 Hz_40ms_modulation_mae_db": 1.7724921249121135, "mr_stft_db": 5.880646977101859, "band_mae_db": 0.5061177091130952, "rms_gain_error_db": -0.13914301231099274, "output_peak": 0.5345884211087624}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.222201406976013, "全帯域_10ms_rms_envelope_nrmse": 0.0751947583325026, "全帯域_10ms_envelope_correlation": 0.9944672504255474, "全帯域_10ms_modulation_mae_db": 0.4869243773018854, "全帯域_40ms_log_rms_mae_db": 0.5876335191218363, "全帯域_40ms_rms_envelope_nrmse": 0.037808085570639756, "全帯域_40ms_envelope_correlation": 0.9986502383265217, "全帯域_40ms_modulation_mae_db": 0.4301053298589194, "300–1000 Hz_10ms_log_rms_mae_db": 1.6728005368159855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09717439821546923, "300–1000 Hz_10ms_envelope_correlation": 0.9934324212625842, "300–1000 Hz_10ms_modulation_mae_db": 0.40000500222721014, "300–1000 Hz_40ms_log_rms_mae_db": 1.0592892307935446, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05693250353207904, "300–1000 Hz_40ms_envelope_correlation": 0.9980828453520548, "300–1000 Hz_40ms_modulation_mae_db": 0.43740977875391157, "1000–3000 Hz_10ms_log_rms_mae_db": 2.028930853001935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2574877723867649, "1000–3000 Hz_10ms_envelope_correlation": 0.9460804249036879, "1000–3000 Hz_10ms_modulation_mae_db": 0.3326750142950972, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5710856071222155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1749547586457617, "1000–3000 Hz_40ms_envelope_correlation": 0.9738685896834807, "1000–3000 Hz_40ms_modulation_mae_db": 0.5177943644977296, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2266534412067798, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4448442964727086, "3000–6000 Hz_10ms_envelope_correlation": 0.8941722234437194, "3000–6000 Hz_10ms_modulation_mae_db": 1.1055536634080734, "3000–6000 Hz_40ms_log_rms_mae_db": 2.1367571870933646, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3892486414725043, "3000–6000 Hz_40ms_envelope_correlation": 0.9159271967832759, "3000–6000 Hz_40ms_modulation_mae_db": 1.2701277548377166, "mr_stft_db": 5.887669468455052, "band_mae_db": 0.5089700590585321, "rms_gain_error_db": -0.12480286405085331, "output_peak": 0.5305694339010413}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8986223086351615, "spectral_convergence": 0.23374481221719326, "spectral_convergence_status": "ok", "amplitude_floor": 1.9267471569597447e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.631548140711266, "spectral_convergence": 0.22312012952639168, "spectral_convergence_status": "ok", "amplitude_floor": 1.271916808390449e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0924490248115495, "spectral_convergence": 0.23185342050243612, "spectral_convergence_status": "ok", "amplitude_floor": 1.1644398478092437e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001373055391651405, "output_power": 0.0013779836549274616, "delta_db": 0.015560083295653422, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0033897235254818038, "output_power": 0.003291491062006924, "delta_db": -0.12771597369388676, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002841140646523882, "output_power": 0.0002980761955305064, "delta_db": 0.20834561137844707, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.671358771236327e-05, "output_power": 8.297892895496305e-05, "delta_db": -0.19119330315308544, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6620686952686656e-05, "output_power": 2.9712006207562682e-05, "delta_db": 0.47712718782459906, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.533501052724457e-05, "output_power": 3.4069340634430475e-05, "delta_db": 1.2864263938702913, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3843947588693271, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.060087037113472776, "pitch": null, "output_rms_dbfs": -22.91212850178818, "output_peak": 0.5175699120272388, "reference_rms_dbfs": -22.85204146467471, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.874206491385993}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023735521522065317, "log_rms_mae_db": 1.2153520649306009, "rms_envelope_nrmse": 0.07103960358159865, "envelope_correlation": 0.9949592215596017, "rms_gain_error_db": -0.0600495443983199, "reference_mean_power": 0.0051897925018650195, "output_mean_power": 0.005118527486678711}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6580095170174236e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00099786392963893, "output_power": 0.0009852812710662204, "reference_active": true, "delta_db": -0.05511096730455728}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00025981544310873085, "output_power": 0.00022119522052846247, "reference_active": true, "delta_db": -0.6988922270705276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.167033379867013e-05, "output_power": 3.841879821809848e-05, "reference_active": true, "delta_db": -0.35283203837935906}], "active_band_count": 3, "active_band_mae_db": 0.36894507758481465}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022053657198928817, "log_rms_mae_db": 0.6396592317885603, "rms_envelope_nrmse": 0.0357607649866036, "envelope_correlation": 0.9986477270421188, "rms_gain_error_db": -0.06001628898650266, "reference_mean_power": 0.0052155458817555326, "output_mean_power": 0.005143966616146509}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3916979728964525e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008985624617991317, "output_power": 0.0008951120102230767, "reference_active": true, "delta_db": -0.016708872778930673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001435836821660598, "output_power": 0.00011907358342182991, "reference_active": true, "delta_db": -0.8128966294882712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3776224720397365e-06, "output_power": 2.147219615294771e-06, "reference_active": true, "delta_db": -0.4426643094655667}], "active_band_count": 3, "active_band_mae_db": 0.4240899372442562}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018677846007938983, "log_rms_mae_db": 1.6666537206001708, "rms_envelope_nrmse": 0.09607536343101557, "envelope_correlation": 0.993190380451079, "rms_gain_error_db": -0.12035292376164725, "reference_mean_power": 0.0023293656178310037, "output_mean_power": 0.002265699811277258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5191576623571145e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006490500325829311, "output_power": 0.0006007063165102272, "reference_active": true, "delta_db": -0.33615977418331144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013228128531435308, "output_power": 0.00011211875648684437, "reference_active": true, "delta_db": -0.7182013375589064}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.716163060824064e-05, "output_power": 1.588888050888656e-05, "reference_active": true, "delta_db": -0.3346525088659268}], "active_band_count": 3, "active_band_mae_db": 0.4630045402027149}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017927232752442356, "log_rms_mae_db": 0.9916233307451402, "rms_envelope_nrmse": 0.05170373967588548, "envelope_correlation": 0.998026529979996, "rms_gain_error_db": -0.1203845461669076, "reference_mean_power": 0.002341010543733861, "output_mean_power": 0.002277009880421642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.424681480871225e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006148042140063018, "output_power": 0.0005666660383477626, "reference_active": true, "delta_db": -0.35409650476462434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.176867672883116e-05, "output_power": 5.399562143964087e-05, "reference_active": true, "delta_db": -0.584097536754521}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.094779276972058e-06, "output_power": 1.0356470542769926e-06, "reference_active": true, "delta_db": -0.24114793839663645}], "active_band_count": 3, "active_band_mae_db": 0.3931139933052606}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.042588377582574e-05, "log_rms_mae_db": 1.938364400732502, "rms_envelope_nrmse": 0.2490448429339933, "envelope_correlation": 0.9500516693478994, "rms_gain_error_db": -0.0859666705269023, "reference_mean_power": 9.196537764000522e-05, "output_mean_power": 9.016286254273875e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3593984658429913e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0478584079151807e-05, "output_power": 2.0626577479257517e-05, "reference_active": true, "delta_db": 0.03127246910273746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.015542002555929e-06, "output_power": 6.160946673999413e-06, "reference_active": true, "delta_db": 0.1037268624697125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6245635384567205e-06, "output_power": 2.3815043924415026e-06, "reference_active": true, "delta_db": 1.66114685114806}], "active_band_count": 3, "active_band_mae_db": 0.5987153942401701}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6202941099492215e-05, "log_rms_mae_db": 1.4193762319878862, "rms_envelope_nrmse": 0.16273464687629904, "envelope_correlation": 0.9768698283033771, "rms_gain_error_db": -0.08595406260132298, "reference_mean_power": 9.242233001846982e-05, "output_mean_power": 9.061112173705898e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9023164132494484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.934413521065944e-05, "output_power": 2.0508351226361445e-05, "reference_active": true, "delta_db": 0.2538142742956964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.243887754162173e-06, "output_power": 3.5007536177907465e-06, "reference_active": true, "delta_db": 0.330957280681374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3065359453687705e-07, "output_power": 1.9558777266340688e-07, "reference_active": true, "delta_db": 1.7522033848840337}], "active_band_count": 3, "active_band_mae_db": 0.778991646620368}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.535256840646245e-05, "log_rms_mae_db": 2.25516018169224, "rms_envelope_nrmse": 0.4493416913891345, "envelope_correlation": 0.8876135227913381, "rms_gain_error_db": 0.6212272053995398, "reference_mean_power": 2.9383590737924492e-05, "output_mean_power": 3.3902176999988615e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.113292932890967e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.65042891211228e-06, "output_power": 9.88550132502748e-06, "reference_active": true, "delta_db": 0.10452082717514771}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.01875786446688e-06, "output_power": 4.209594210858367e-06, "reference_active": true, "delta_db": 0.20148393331908754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.253460098772285e-07, "output_power": 1.138690153436816e-06, "reference_active": true, "delta_db": 1.9586033886286935}], "active_band_count": 3, "active_band_mae_db": 0.7548693830409762}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9034899521347454e-05, "log_rms_mae_db": 2.145581185213099, "rms_envelope_nrmse": 0.38923160495810377, "envelope_correlation": 0.912112112642686, "rms_gain_error_db": 0.6211897617273278, "reference_mean_power": 2.9530847534057242e-05, "output_mean_power": 3.4071785075566744e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9758537203809177e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.950101722919118e-06, "output_power": 1.0403337289879591e-05, "reference_active": true, "delta_db": 0.19345158421324934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0870738801319347e-06, "output_power": 2.3637650380089533e-06, "reference_active": true, "delta_db": 0.5406648190776279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.171675766600786e-08, "output_power": 1.016081204565367e-07, "reference_active": true, "delta_db": 2.1652531585515233}], "active_band_count": 3, "active_band_mae_db": 0.9664565206141336}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8665051509091103, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9c3761dc89ad59464e0adcde072f9bb7861eccfc60a686479af5f90bc95770eb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9513426564148417, "standard_deviation": 0.002750120456655477, "minimum": 0.9460804249036879, "maximum": 0.9552519190133761, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9224179495215323, "standard_deviation": 0.04801385204864961, "minimum": 1.8763585682323485, "maximum": 2.028930853001935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5206242073068411, "standard_deviation": 0.1329184959701131, "minimum": 0.3326750142950972, "maximum": 0.7267109084023465, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2473583168741243, "standard_deviation": 0.006550757357266335, "minimum": 0.23851092736730972, "maximum": 0.2574877723867649, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9755549166969826, "standard_deviation": 0.001233262631132531, "minimum": 0.9738649940747997, "maximum": 0.9769825495273514, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4886817782231363, "standard_deviation": 0.04480655180362468, "minimum": 1.4193762319878862, "maximum": 1.5710856071222155, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6347818389810154, "standard_deviation": 0.12263068542236591, "minimum": 0.42397463548989883, "maximum": 0.778991646620368, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16907906913237258, "standard_deviation": 0.0051303884394275955, "minimum": 0.16273464687629904, "maximum": 0.17594483047718282, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8946463446601094, "standard_deviation": 0.004812261796822391, "minimum": 0.8876135227913381, "maximum": 0.9005980606712862, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2198479226380083, "standard_deviation": 0.021409742676997388, "minimum": 2.1826775646220176, "maximum": 2.25516018169224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.021693191108544, "standard_deviation": 0.347697662217401, "minimum": 0.6717332440781125, "maximum": 1.616928531137323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.44436463888058414, "standard_deviation": 0.006434160250571525, "minimum": 0.4328430843007298, "maximum": 0.45201668216149776, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9171507540349418, "standard_deviation": 0.0044931327682952256, "minimum": 0.912112112642686, "maximum": 0.9230678565027436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.1274217886850146, "standard_deviation": 0.022862965963101015, "minimum": 2.0830296973746036, "maximum": 2.145581185213099, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1851956095432183, "standard_deviation": 0.3099965750820642, "minimum": 0.9280754801180113, "maximum": 1.7724921249121135, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3871629613767241, "standard_deviation": 0.006643992091558058, "minimum": 0.3788197310540476, "maximum": 0.39866558366767785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918819529590852, "standard_deviation": 0.001589794733725651, "minimum": 0.9892148277841292, "maximum": 0.9934324212625842, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6697152823343298, "standard_deviation": 0.03523900481127187, "minimum": 1.62581624380536, "maximum": 1.740306643867535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6581973212810254, "standard_deviation": 0.1732010337811721, "minimum": 0.40000500222721014, "maximum": 0.8441092799752007, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10580614584766904, "standard_deviation": 0.009295296196043272, "minimum": 0.09607536343101557, "maximum": 0.12060216468185732, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997580907387404, "standard_deviation": 0.0005417015524583403, "minimum": 0.9967745306000391, "maximum": 0.9980828453520548, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0031774474871096, "standard_deviation": 0.04139914108472231, "minimum": 0.9562435838706657, "maximum": 1.0592892307935446, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7012608003321673, "standard_deviation": 0.26044902738316406, "minimum": 0.3931139933052606, "maximum": 1.1472565565610404, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059908710855870996, "standard_deviation": 0.005753725236187098, "minimum": 0.05170373967588548, "maximum": 0.06748469509584439, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4715900698656969, "standard_deviation": 0.04972589701079341, "minimum": 0.3843947588693271, "maximum": 0.5318146720266249, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.871914845688521, "standard_deviation": 0.010105955128329229, "minimum": 5.856711427692134, "maximum": 5.887669468455052, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5275242445824941, "standard_deviation": 0.008782231069956374, "minimum": 0.5175699120272388, "maximum": 0.5445863919171128, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09554828207568251, "standard_deviation": 0.02773833416751622, "minimum": -0.13914301231099274, "maximum": -0.060087037113472776, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9936410150809856, "standard_deviation": 0.000966465045956774, "minimum": 0.9919172881296454, "maximum": 0.9949592215596017, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2291837383004423, "standard_deviation": 0.026145415787474743, "minimum": 1.2020022519866533, "maximum": 1.2796428997755716, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.25779361792767386, "standard_deviation": 0.12652482233605727, "minimum": 0.11581624962770039, "maximum": 0.4869243773018854, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07988853870106563, "standard_deviation": 0.005889371411191053, "minimum": 0.07103960358159865, "maximum": 0.08997120982909178, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981727258469466, "standard_deviation": 0.00035434845957850285, "minimum": 0.9977055870679288, "maximum": 0.9986502383265217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6393999077875907, "standard_deviation": 0.03577856887545521, "minimum": 0.5876335191218363, "maximum": 0.6966593695844967, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.28499522292495494, "standard_deviation": 0.1428154844874736, "minimum": 0.06767090622102105, "maximum": 0.4301053298589194, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04215074095805295, "standard_deviation": 0.004100414659929623, "minimum": 0.0357607649866036, "maximum": 0.046909722486631666, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0018", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.3561151027679443, "predicted_mos": 2.3561151027679443, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "96d5d3ff8eff424c508f90a93a37b5e2cb87e1f0ca015050c2d4d01cc9a9a0f8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6495279003866017, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1392546639266334, "全帯域_10ms_rms_envelope_nrmse": 0.08100796376709914, "全帯域_10ms_envelope_correlation": 0.9934864907421436, "全帯域_10ms_modulation_mae_db": 0.21002231300639237, "全帯域_40ms_log_rms_mae_db": 0.5828826265317184, "全帯域_40ms_rms_envelope_nrmse": 0.03871107933042411, "全帯域_40ms_envelope_correlation": 0.9984998114949742, "全帯域_40ms_modulation_mae_db": 0.4073192222921163, "300–1000 Hz_10ms_log_rms_mae_db": 1.8475706155462954, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10630201524056031, "300–1000 Hz_10ms_envelope_correlation": 0.9919044287133347, "300–1000 Hz_10ms_modulation_mae_db": 0.9606903098089775, "300–1000 Hz_40ms_log_rms_mae_db": 1.115608639692254, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05912942359100015, "300–1000 Hz_40ms_envelope_correlation": 0.9977432169730213, "300–1000 Hz_40ms_modulation_mae_db": 1.0015446681545936, "1000–3000 Hz_10ms_log_rms_mae_db": 2.130526553811726, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2526766837804498, "1000–3000 Hz_10ms_envelope_correlation": 0.9496109796046064, "1000–3000 Hz_10ms_modulation_mae_db": 0.7111458133893472, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6225262032257335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17243835024816645, "1000–3000 Hz_40ms_envelope_correlation": 0.9745216108139189, "1000–3000 Hz_40ms_modulation_mae_db": 0.9290662340797174, "3000–6000 Hz_10ms_log_rms_mae_db": 2.511425916636438, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.46212429783276904, "3000–6000 Hz_10ms_envelope_correlation": 0.8847438884694593, "3000–6000 Hz_10ms_modulation_mae_db": 1.582217995652158, "3000–6000 Hz_40ms_log_rms_mae_db": 2.4224791713174474, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3972892739078222, "3000–6000 Hz_40ms_envelope_correlation": 0.9118322211528976, "3000–6000 Hz_40ms_modulation_mae_db": 1.8528298536548855, "mr_stft_db": 6.057198224031388, "band_mae_db": 0.44731444382035884, "rms_gain_error_db": -0.10201806503973633, "output_peak": 0.5151427906979105}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.223674327831196, "全帯域_10ms_rms_envelope_nrmse": 0.07881035189538818, "全帯域_10ms_envelope_correlation": 0.9938763073349611, "全帯域_10ms_modulation_mae_db": 0.19100982577290457, "全帯域_40ms_log_rms_mae_db": 0.6500064304116031, "全帯域_40ms_rms_envelope_nrmse": 0.04467036304659834, "全帯域_40ms_envelope_correlation": 0.997992216706926, "全帯域_40ms_modulation_mae_db": 0.10349786599335009, "300–1000 Hz_10ms_log_rms_mae_db": 1.8621284247536218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1080190466936714, "300–1000 Hz_10ms_envelope_correlation": 0.991629713398021, "300–1000 Hz_10ms_modulation_mae_db": 1.0392354974958815, "300–1000 Hz_40ms_log_rms_mae_db": 1.2053728584360943, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06291336176285164, "300–1000 Hz_40ms_envelope_correlation": 0.9973814184789255, "300–1000 Hz_40ms_modulation_mae_db": 1.1530787218625846, "1000–3000 Hz_10ms_log_rms_mae_db": 2.162070016921624, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2641712235514821, "1000–3000 Hz_10ms_envelope_correlation": 0.9444245366261795, "1000–3000 Hz_10ms_modulation_mae_db": 0.5417928370465598, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6970887890445259, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17989190289537035, "1000–3000 Hz_40ms_envelope_correlation": 0.9720484852934338, "1000–3000 Hz_40ms_modulation_mae_db": 0.87844384573472, "3000–6000 Hz_10ms_log_rms_mae_db": 2.535557403933141, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.44698728618809275, "3000–6000 Hz_10ms_envelope_correlation": 0.8920892562489919, "3000–6000 Hz_10ms_modulation_mae_db": 0.9377781501081371, "3000–6000 Hz_40ms_log_rms_mae_db": 2.4096051779471033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3873865203069193, "3000–6000 Hz_40ms_envelope_correlation": 0.9161560440016603, "3000–6000 Hz_40ms_modulation_mae_db": 0.953209817808118, "mr_stft_db": 6.067056348671774, "band_mae_db": 0.4299277896212188, "rms_gain_error_db": -0.11715476640495202, "output_peak": 0.5237719217273195}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2164428796926536, "全帯域_10ms_rms_envelope_nrmse": 0.08608962835516658, "全帯域_10ms_envelope_correlation": 0.9926429526546509, "全帯域_10ms_modulation_mae_db": 0.11517838733673212, "全帯域_40ms_log_rms_mae_db": 0.6611512465807123, "全帯域_40ms_rms_envelope_nrmse": 0.050805806292048894, "全帯域_40ms_envelope_correlation": 0.9973187851338312, "全帯域_40ms_modulation_mae_db": 0.15934878504969988, "300–1000 Hz_10ms_log_rms_mae_db": 1.8683774543014564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11910594477316529, "300–1000 Hz_10ms_envelope_correlation": 0.989480194042925, "300–1000 Hz_10ms_modulation_mae_db": 0.7501842062548855, "300–1000 Hz_40ms_log_rms_mae_db": 1.1441977940797154, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07433445870560895, "300–1000 Hz_40ms_envelope_correlation": 0.995819800630136, "300–1000 Hz_40ms_modulation_mae_db": 1.1546992061834975, "1000–3000 Hz_10ms_log_rms_mae_db": 2.132590845444768, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25070951553601395, "1000–3000 Hz_10ms_envelope_correlation": 0.9494555798639633, "1000–3000 Hz_10ms_modulation_mae_db": 0.20179979845029686, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6124080577072972, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1658517453058625, "1000–3000 Hz_40ms_envelope_correlation": 0.9761965283453148, "1000–3000 Hz_40ms_modulation_mae_db": 0.43807989891187255, "3000–6000 Hz_10ms_log_rms_mae_db": 2.536742457845606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.45128752265884947, "3000–6000 Hz_10ms_envelope_correlation": 0.8902486059595167, "3000–6000 Hz_10ms_modulation_mae_db": 0.7038725449885107, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3792942957663255, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38696593006164653, "3000–6000 Hz_40ms_envelope_correlation": 0.9165001617737124, "3000–6000 Hz_40ms_modulation_mae_db": 0.8370468828141889, "mr_stft_db": 6.063482376665152, "band_mae_db": 0.4263499145617284, "rms_gain_error_db": -0.10590713031982601, "output_peak": 0.5069588348928146}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2076967305318456, "全帯域_10ms_rms_envelope_nrmse": 0.08631601715405327, "全帯域_10ms_envelope_correlation": 0.9926553673514924, "全帯域_10ms_modulation_mae_db": 0.18098479755580668, "全帯域_40ms_log_rms_mae_db": 0.630740830423434, "全帯域_40ms_rms_envelope_nrmse": 0.05073531337773499, "全帯域_40ms_envelope_correlation": 0.9974065209194672, "全帯域_40ms_modulation_mae_db": 0.21780294870541453, "300–1000 Hz_10ms_log_rms_mae_db": 1.8378415444845777, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12191795643087798, "300–1000 Hz_10ms_envelope_correlation": 0.9892249460979402, "300–1000 Hz_10ms_modulation_mae_db": 0.3145062168094616, "300–1000 Hz_40ms_log_rms_mae_db": 1.111797066207772, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07767015653172667, "300–1000 Hz_40ms_envelope_correlation": 0.9957686875359425, "300–1000 Hz_40ms_modulation_mae_db": 0.3309361619021801, "1000–3000 Hz_10ms_log_rms_mae_db": 2.081634047613517, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2567199577518617, "1000–3000 Hz_10ms_envelope_correlation": 0.9497393447489186, "1000–3000 Hz_10ms_modulation_mae_db": 1.0977862958723303, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6141680340173576, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16925383827090681, "1000–3000 Hz_40ms_envelope_correlation": 0.9765832104957297, "1000–3000 Hz_40ms_modulation_mae_db": 1.287158576110029, "3000–6000 Hz_10ms_log_rms_mae_db": 2.515453840997642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.46286389706156417, "3000–6000 Hz_10ms_envelope_correlation": 0.892099569361363, "3000–6000 Hz_10ms_modulation_mae_db": 1.3149629695399256, "3000–6000 Hz_40ms_log_rms_mae_db": 2.384467879483368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3975614143335088, "3000–6000 Hz_40ms_envelope_correlation": 0.9188206354549785, "3000–6000 Hz_40ms_modulation_mae_db": 1.4268298274211701, "mr_stft_db": 6.048693307812751, "band_mae_db": 0.4893668015300492, "rms_gain_error_db": -0.13337694960252636, "output_peak": 0.5230934362057523}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1418997254899284, "全帯域_10ms_rms_envelope_nrmse": 0.07844396070108335, "全帯域_10ms_envelope_correlation": 0.9938347928039971, "全帯域_10ms_modulation_mae_db": 0.22494902452118556, "全帯域_40ms_log_rms_mae_db": 0.5621162020170665, "全帯域_40ms_rms_envelope_nrmse": 0.039628248216484106, "全帯域_40ms_envelope_correlation": 0.9983143932520642, "全帯域_40ms_modulation_mae_db": 0.262401764542597, "300–1000 Hz_10ms_log_rms_mae_db": 1.8742422782910015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10085805993712753, "300–1000 Hz_10ms_envelope_correlation": 0.992643690707669, "300–1000 Hz_10ms_modulation_mae_db": 0.7503978123384535, "300–1000 Hz_40ms_log_rms_mae_db": 1.1673925473066842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053340389127221687, "300–1000 Hz_40ms_envelope_correlation": 0.9981075617467379, "300–1000 Hz_40ms_modulation_mae_db": 0.7882286669966696, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1855876256643096, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2597866584064141, "1000–3000 Hz_10ms_envelope_correlation": 0.9472812338655666, "1000–3000 Hz_10ms_modulation_mae_db": 0.6531877735697968, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6965480068594583, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18094614783625382, "1000–3000 Hz_40ms_envelope_correlation": 0.9722361192775049, "1000–3000 Hz_40ms_modulation_mae_db": 0.8590747743342071, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5444440813911133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4468179770522331, "3000–6000 Hz_10ms_envelope_correlation": 0.8877411261829815, "3000–6000 Hz_10ms_modulation_mae_db": 0.7467995997544237, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3791205567700975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38848254574817564, "3000–6000 Hz_40ms_envelope_correlation": 0.9117960010068568, "3000–6000 Hz_40ms_modulation_mae_db": 0.9403358396525766, "mr_stft_db": 6.062642537518393, "band_mae_db": 0.3898241962497968, "rms_gain_error_db": -0.04964996614060968, "output_peak": 0.5270322444658224}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2138529983572641, "全帯域_10ms_rms_envelope_nrmse": 0.08442919743166745, "全帯域_10ms_envelope_correlation": 0.9928832015540284, "全帯域_10ms_modulation_mae_db": 0.26410045341253036, "全帯域_40ms_log_rms_mae_db": 0.684788010092693, "全帯域_40ms_rms_envelope_nrmse": 0.05268002939431812, "全帯域_40ms_envelope_correlation": 0.9970439939282121, "全帯域_40ms_modulation_mae_db": 0.3467539336688239, "300–1000 Hz_10ms_log_rms_mae_db": 1.866999766133172, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.105804650787508, "300–1000 Hz_10ms_envelope_correlation": 0.9917537142787555, "300–1000 Hz_10ms_modulation_mae_db": 0.7754045737189438, "300–1000 Hz_40ms_log_rms_mae_db": 1.1537120987758724, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06376225236141424, "300–1000 Hz_40ms_envelope_correlation": 0.9969789595975684, "300–1000 Hz_40ms_modulation_mae_db": 0.7646454809365628, "1000–3000 Hz_10ms_log_rms_mae_db": 2.135200224883966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23849079289913924, "1000–3000 Hz_10ms_envelope_correlation": 0.9560044569278501, "1000–3000 Hz_10ms_modulation_mae_db": 0.6356009058810798, "1000–3000 Hz_40ms_log_rms_mae_db": 1.613208728570452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16148922402588267, "1000–3000 Hz_40ms_envelope_correlation": 0.9783817056626315, "1000–3000 Hz_40ms_modulation_mae_db": 0.9048120641705787, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5138952622681843, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4381843388417028, "3000–6000 Hz_10ms_envelope_correlation": 0.9059417815257318, "3000–6000 Hz_10ms_modulation_mae_db": 0.9384423818544677, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3407414032294698, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38097920508419064, "3000–6000 Hz_40ms_envelope_correlation": 0.9277952122569888, "3000–6000 Hz_40ms_modulation_mae_db": 1.1359663655549164, "mr_stft_db": 6.057857669503432, "band_mae_db": 0.5423447856036927, "rms_gain_error_db": -0.0815354435073777, "output_peak": 0.5597365847937914}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.19884689003933, "全帯域_10ms_rms_envelope_nrmse": 0.09150393996621738, "全帯域_10ms_envelope_correlation": 0.9916866180631381, "全帯域_10ms_modulation_mae_db": 0.16534456721543112, "全帯域_40ms_log_rms_mae_db": 0.6506242851889487, "全帯域_40ms_rms_envelope_nrmse": 0.04960725379861941, "全帯域_40ms_envelope_correlation": 0.9974919267843719, "全帯域_40ms_modulation_mae_db": 0.2039720117315458, "300–1000 Hz_10ms_log_rms_mae_db": 1.8880119474136885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11671940679231171, "300–1000 Hz_10ms_envelope_correlation": 0.990150552113141, "300–1000 Hz_10ms_modulation_mae_db": 0.6352157224678006, "300–1000 Hz_40ms_log_rms_mae_db": 1.1568279440933624, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06278627669011158, "300–1000 Hz_40ms_envelope_correlation": 0.997424088307488, "300–1000 Hz_40ms_modulation_mae_db": 0.8217401357353307, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0959000992423746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2423085111277282, "1000–3000 Hz_10ms_envelope_correlation": 0.9528809620526976, "1000–3000 Hz_10ms_modulation_mae_db": 0.6269155842908157, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6059068357149098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16113234114475367, "1000–3000 Hz_40ms_envelope_correlation": 0.9776032445132861, "1000–3000 Hz_40ms_modulation_mae_db": 0.8888418409823445, "3000–6000 Hz_10ms_log_rms_mae_db": 2.603711388705472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4664058599454582, "3000–6000 Hz_10ms_envelope_correlation": 0.8806871192034978, "3000–6000 Hz_10ms_modulation_mae_db": 1.3755481051269702, "3000–6000 Hz_40ms_log_rms_mae_db": 2.4227101953200125, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.40247147068065, "3000–6000 Hz_40ms_envelope_correlation": 0.9078645102503339, "3000–6000 Hz_40ms_modulation_mae_db": 1.2598677756229304, "mr_stft_db": 6.064941115604967, "band_mae_db": 0.4182831999913997, "rms_gain_error_db": -0.12174998559658731, "output_peak": 0.5360069570140916}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.238632602130307, "全帯域_10ms_rms_envelope_nrmse": 0.08026684189481946, "全帯域_10ms_envelope_correlation": 0.9935974558359372, "全帯域_10ms_modulation_mae_db": 0.2263174616037412, "全帯域_40ms_log_rms_mae_db": 0.6459183639145268, "全帯域_40ms_rms_envelope_nrmse": 0.04156860895656622, "全帯域_40ms_envelope_correlation": 0.9982442625171867, "全帯域_40ms_modulation_mae_db": 0.23771210233340864, "300–1000 Hz_10ms_log_rms_mae_db": 1.8333498905924983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10768033748707273, "300–1000 Hz_10ms_envelope_correlation": 0.9915065926492413, "300–1000 Hz_10ms_modulation_mae_db": 0.5693002277313033, "300–1000 Hz_40ms_log_rms_mae_db": 1.1664232520882691, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05588469022888236, "300–1000 Hz_40ms_envelope_correlation": 0.9978302115464982, "300–1000 Hz_40ms_modulation_mae_db": 0.5150898040834287, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1355726400869828, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.251248201599642, "1000–3000 Hz_10ms_envelope_correlation": 0.9495649196495163, "1000–3000 Hz_10ms_modulation_mae_db": 0.5264119830769659, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6309924075112323, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17140551431170375, "1000–3000 Hz_40ms_envelope_correlation": 0.9744658709964684, "1000–3000 Hz_40ms_modulation_mae_db": 0.7392424058040771, "3000–6000 Hz_10ms_log_rms_mae_db": 2.545705472041722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4474169418571016, "3000–6000 Hz_10ms_envelope_correlation": 0.8994589263866982, "3000–6000 Hz_10ms_modulation_mae_db": 1.1522851924308413, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3671430634233595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3890831583590929, "3000–6000 Hz_40ms_envelope_correlation": 0.9223513528007972, "3000–6000 Hz_40ms_modulation_mae_db": 1.2530123565594178, "mr_stft_db": 6.0628557778777115, "band_mae_db": 0.533785422015153, "rms_gain_error_db": -0.09128549554450228, "output_peak": 0.5252271969689667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.1590027006623735, "spectral_convergence": 0.23812004774488343, "spectral_convergence_status": "ok", "amplitude_floor": 1.9267471569597447e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.825229627626069, "spectral_convergence": 0.22252489523810054, "spectral_convergence_status": "ok", "amplitude_floor": 1.271916808390449e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.18736234380572, "spectral_convergence": 0.23307432826958027, "spectral_convergence_status": "ok", "amplitude_floor": 1.1644398478092437e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001373055391651405, "output_power": 0.0013917327385738493, "delta_db": 0.058677856802941936, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0033897235254818038, "output_power": 0.0032105626409920157, "delta_db": -0.23583129664320493, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002841140646523882, "output_power": 0.0003110911752994814, "delta_db": 0.39394958313768014, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.671358771236327e-05, "output_power": 8.626741630209187e-05, "delta_db": -0.02240364047781882, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6620686952686656e-05, "output_power": 2.9470838913872564e-05, "delta_db": 0.44173240260021923, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.533501052724457e-05, "output_power": 3.60454355320171e-05, "delta_db": 1.5312918832602875, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.44731444382035884, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10201806503973633, "pitch": null, "output_rms_dbfs": -22.954059529714446, "output_peak": 0.5151427906979105, "reference_rms_dbfs": -22.85204146467471, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.057198224031388}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023735521522065317, "log_rms_mae_db": 1.1392546639266334, "rms_envelope_nrmse": 0.08100796376709914, "envelope_correlation": 0.9934864907421436, "rms_gain_error_db": -0.1019850235160229, "reference_mean_power": 0.0051897925018650195, "output_mean_power": 0.005069340835962609}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6580095170174236e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00099786392963893, "output_power": 0.0009464995195009977, "reference_active": true, "delta_db": -0.2295092639096425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00025981544310873085, "output_power": 0.00023724227701846852, "reference_active": true, "delta_db": -0.39472877751803526}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.167033379867013e-05, "output_power": 4.1726299322205544e-05, "reference_active": true, "delta_db": 0.005828897591499241}], "active_band_count": 3, "active_band_mae_db": 0.21002231300639237}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022053657198928817, "log_rms_mae_db": 0.5828826265317184, "rms_envelope_nrmse": 0.03871107933042411, "envelope_correlation": 0.9984998114949742, "rms_gain_error_db": -0.10194651485200322, "reference_mean_power": 0.0052155458817555326, "output_mean_power": 0.005094541669700997}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3916979728964525e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008985624617991317, "output_power": 0.0008502159017320517, "reference_active": true, "delta_db": -0.24019048229210002}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001435836821660598, "output_power": 0.00013204094905586793, "reference_active": true, "delta_db": -0.36396449309488016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3776224720397365e-06, "output_power": 2.0623512187840647e-06, "reference_active": true, "delta_db": -0.6178026914893687}], "active_band_count": 3, "active_band_mae_db": 0.4073192222921163}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018677846007938983, "log_rms_mae_db": 1.8475706155462954, "rms_envelope_nrmse": 0.10630201524056031, "envelope_correlation": 0.9919044287133347, "rms_gain_error_db": -0.2286245629298061, "reference_mean_power": 0.0023293656178310037, "output_mean_power": 0.002209913141859857}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5191576623571145e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006490500325829311, "output_power": 0.0006115840021455619, "reference_active": true, "delta_db": -0.2582205947108216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013228128531435308, "output_power": 0.00010520470443118273, "reference_active": true, "delta_db": -0.9946324552817531}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.716163060824064e-05, "output_power": 1.179333844019647e-05, "reference_active": true, "delta_db": -1.6292178794343575}], "active_band_count": 3, "active_band_mae_db": 0.9606903098089775}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017927232752442356, "log_rms_mae_db": 1.115608639692254, "rms_envelope_nrmse": 0.05912942359100015, "envelope_correlation": 0.9977432169730213, "rms_gain_error_db": -0.22870178222829957, "reference_mean_power": 0.002341010543733861, "output_mean_power": 0.0022209214137056863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.424681480871225e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006148042140063018, "output_power": 0.0005711621688876732, "reference_active": true, "delta_db": -0.3197740150569006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.176867672883116e-05, "output_power": 4.922079552541181e-05, "reference_active": true, "delta_db": -0.9861966888317443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.094779276972058e-06, "output_power": 7.403894777313829e-07, "reference_active": true, "delta_db": -1.6986633005751361}], "active_band_count": 3, "active_band_mae_db": 1.0015446681545936}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.042588377582574e-05, "log_rms_mae_db": 2.130526553811726, "rms_envelope_nrmse": 0.2526766837804498, "envelope_correlation": 0.9496109796046064, "rms_gain_error_db": 0.09019127364435847, "reference_mean_power": 9.196537764000522e-05, "output_mean_power": 9.389522044507942e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3593984658429913e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0478584079151807e-05, "output_power": 2.0967997233177774e-05, "reference_active": true, "delta_db": 0.10257025079055086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.015542002555929e-06, "output_power": 6.909464543933187e-06, "reference_active": true, "delta_db": 0.6016962904835041}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6245635384567205e-06, "output_power": 2.2576353240011156e-06, "reference_active": true, "delta_db": 1.4291708988939869}], "active_band_count": 3, "active_band_mae_db": 0.7111458133893472}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6202941099492215e-05, "log_rms_mae_db": 1.6225262032257335, "rms_envelope_nrmse": 0.17243835024816645, "envelope_correlation": 0.9745216108139189, "rms_gain_error_db": 0.09010948487318046, "reference_mean_power": 9.242233001846982e-05, "output_mean_power": 9.435998466070972e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9023164132494484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.934413521065944e-05, "output_power": 2.0649983782422423e-05, "reference_active": true, "delta_db": 0.28370395780065716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.243887754162173e-06, "output_power": 4.035742178354718e-06, "reference_active": true, "delta_db": 0.9485759550592101}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3065359453687705e-07, "output_power": 1.8690169563379695e-07, "reference_active": true, "delta_db": 1.5549187893792846}], "active_band_count": 3, "active_band_mae_db": 0.9290662340797174}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.535256840646245e-05, "log_rms_mae_db": 2.511425916636438, "rms_envelope_nrmse": 0.46212429783276904, "envelope_correlation": 0.8847438884694593, "rms_gain_error_db": 0.7770250256459637, "reference_mean_power": 2.9383590737924492e-05, "output_mean_power": 3.51404540713429e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.113292932890967e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.65042891211228e-06, "output_power": 9.662359757227064e-06, "reference_active": true, "delta_db": 0.005365875374769027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.01875786446688e-06, "output_power": 5.176970992393933e-06, "reference_active": true, "delta_db": 1.099838913923981}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.253460098772285e-07, "output_power": 1.6776066774690367e-06, "reference_active": true, "delta_db": 3.6414491976577237}], "active_band_count": 3, "active_band_mae_db": 1.582217995652158}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9034899521347454e-05, "log_rms_mae_db": 2.4224791713174474, "rms_envelope_nrmse": 0.3972892739078222, "envelope_correlation": 0.9118322211528976, "rms_gain_error_db": 0.7769886744836667, "reference_mean_power": 2.9530847534057242e-05, "output_mean_power": 3.531626596631214e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9758537203809177e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.950101722919118e-06, "output_power": 1.056019226116827e-05, "reference_active": true, "delta_db": 0.2584430443689937}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0870738801319347e-06, "output_power": 3.1092862961436846e-06, "reference_active": true, "delta_db": 1.7312288986402922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.171675766600786e-08, "output_power": 1.4037341444148556e-07, "reference_active": true, "delta_db": 3.568817617955371}], "active_band_count": 3, "active_band_mae_db": 1.8528298536548855}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8614364547558149, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c12b2ac20c987eb905818cdc0d564a1e752026e449739eab94c049f98d0afe61", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9498702516674122, "standard_deviation": 0.003451732808303242, "minimum": 0.9444245366261795, "maximum": 0.9560044569278501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1323852567086585, "standard_deviation": 0.03303701277824334, "minimum": 2.081634047613517, "maximum": 2.1855876256643096, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.624330123947149, "standard_deviation": 0.24715421205178312, "minimum": 0.20179979845029686, "maximum": 1.0977862958723303, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2520139430815914, "standard_deviation": 0.008536972219721741, "minimum": 0.23849079289913924, "maximum": 0.2641712235514821, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.975254596924786, "standard_deviation": 0.0023451201523803887, "minimum": 0.9720484852934338, "maximum": 0.9783817056626315, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.6366058828313708, "standard_deviation": 0.037909363301468005, "minimum": 1.6059068357149098, "maximum": 1.6970887890445259, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8655899550159434, "standard_deviation": 0.23383895120065626, "minimum": 0.43807989891187255, "maximum": 1.287158576110029, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1703011330048625, "standard_deviation": 0.0074994945034677065, "minimum": 0.16113234114475367, "maximum": 0.18094614783625382, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8916262841672801, "standard_deviation": 0.008024491450143298, "minimum": 0.8806871192034978, "maximum": 0.9059417815257318, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.5383669779774145, "standard_deviation": 0.02981108846573728, "minimum": 2.511425916636438, "maximum": 2.603711388705472, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0939883674319293, "standard_deviation": 0.31423351901149754, "minimum": 0.7038725449885107, "maximum": 1.582217995652158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.45276101517972145, "standard_deviation": 0.0099123769287542, "minimum": 0.4381843388417028, "maximum": 0.4664058599454582, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9166395173372781, "standard_deviation": 0.006388108736551315, "minimum": 0.9078645102503339, "maximum": 0.9277952122569888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.388195217907148, "standard_deviation": 0.028524641461299998, "minimum": 2.3407414032294698, "maximum": 2.4227101953200125, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2073873398860255, "standard_deviation": 0.3268382318130228, "minimum": 0.8370468828141889, "maximum": 1.8528298536548855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.39127743981025076, "standard_deviation": 0.007104778325937636, "minimum": 0.38097920508419064, "maximum": 0.40247147068065, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9910367290001285, "standard_deviation": 0.001248562990524841, "minimum": 0.9892249460979402, "maximum": 0.992643690707669, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8598152401895391, "standard_deviation": 0.018771796237622644, "minimum": 1.8333498905924983, "maximum": 1.8880119474136885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7243668208282135, "standard_deviation": 0.22643805790675087, "minimum": 0.3145062168094616, "maximum": 1.0392354974958815, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11080092726778687, "standard_deviation": 0.007455875842191493, "minimum": 0.10085805993712753, "maximum": 0.12191795643087798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971317431020397, "standard_deviation": 0.0008915668653877694, "minimum": 0.9957686875359425, "maximum": 0.9981075617467379, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.152666525085003, "standard_deviation": 0.030042328280701556, "minimum": 1.111797066207772, "maximum": 1.2053728584360943, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8162453557318561, "standard_deviation": 0.29082172129698247, "minimum": 0.3309361619021801, "maximum": 1.1546992061834975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06372762612485215, "standard_deviation": 0.008443080160565565, "minimum": 0.053340389127221687, "maximum": 0.07767015653172667, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.45964956917417465, "standard_deviation": 0.05599604972504528, "minimum": 0.3898241962497968, "maximum": 0.5423447856036927, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.060590919710696, "standard_deviation": 0.005842692957585382, "minimum": 6.048693307812751, "maximum": 6.067056348671774, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5271212458458087, "standard_deviation": 0.015692395636860475, "minimum": 0.5069588348928146, "maximum": 0.5597365847937914, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1003347252695147, "standard_deviation": 0.026380315821019965, "minimum": -0.13337694960252636, "maximum": -0.04964996614060968, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930828982925436, "standard_deviation": 0.0007548237005174956, "minimum": 0.9916866180631381, "maximum": 0.9938763073349611, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.197537602249895, "standard_deviation": 0.03701915615409012, "minimum": 1.1392546639266334, "maximum": 1.238632602130307, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.19723835380309052, "standard_deviation": 0.04520030045836983, "minimum": 0.11517838733673212, "maximum": 0.26410045341253036, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08335848764568685, "standard_deviation": 0.004529621131503043, "minimum": 0.07844396070108335, "maximum": 0.09150393996621738, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977889888421292, "standard_deviation": 0.0005399114807197379, "minimum": 0.9970439939282121, "maximum": 0.9984998114949742, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6335284993950878, "standard_deviation": 0.04102323766135228, "minimum": 0.5621162020170665, "maximum": 0.684788010092693, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24235107928961952, "standard_deviation": 0.0976752297361938, "minimum": 0.10349786599335009, "maximum": 0.4073192222921163, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04605083780159927, "standard_deviation": 0.005583983601693513, "minimum": 0.03871107933042411, "maximum": 0.05268002939431812, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0018", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.217864990234375, "predicted_mos": 2.217864990234375, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4af309bce478199da2f41dffc3ec37d915062116a1990a5b5af2a8771cb3964f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.697970800101757, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1692120359602658, "全帯域_10ms_rms_envelope_nrmse": 0.0805294043267785, "全帯域_10ms_envelope_correlation": 0.9935453460759083, "全帯域_10ms_modulation_mae_db": 0.6993356380919038, "全帯域_40ms_log_rms_mae_db": 0.6407757097999854, "全帯域_40ms_rms_envelope_nrmse": 0.04394881051985692, "全帯域_40ms_envelope_correlation": 0.9979904043429098, "全帯域_40ms_modulation_mae_db": 0.6588420968201624, "300–1000 Hz_10ms_log_rms_mae_db": 1.835501851291907, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11163398351032965, "300–1000 Hz_10ms_envelope_correlation": 0.9908304796049202, "300–1000 Hz_10ms_modulation_mae_db": 0.6965253288902932, "300–1000 Hz_40ms_log_rms_mae_db": 1.1619126385929461, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06045481301628088, "300–1000 Hz_40ms_envelope_correlation": 0.9973623468464398, "300–1000 Hz_40ms_modulation_mae_db": 0.5106549097910716, "1000–3000 Hz_10ms_log_rms_mae_db": 2.105141544069969, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23890315586779656, "1000–3000 Hz_10ms_envelope_correlation": 0.9551303488437358, "1000–3000 Hz_10ms_modulation_mae_db": 0.5389390334607481, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5917929680806773, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16317032036812856, "1000–3000 Hz_40ms_envelope_correlation": 0.9772505198559491, "1000–3000 Hz_40ms_modulation_mae_db": 0.6595101419366697, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4971850428211404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4556021308723383, "3000–6000 Hz_10ms_envelope_correlation": 0.8959867276181447, "3000–6000 Hz_10ms_modulation_mae_db": 1.10306895751693, "3000–6000 Hz_40ms_log_rms_mae_db": 2.373553638815485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39466702500594647, "3000–6000 Hz_40ms_envelope_correlation": 0.9202634957618216, "3000–6000 Hz_40ms_modulation_mae_db": 0.9697011443809357, "mr_stft_db": 6.017808929709433, "band_mae_db": 0.4539454732035515, "rms_gain_error_db": -0.0909677898653409, "output_peak": 0.5203839110350614}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1449875527616709, "全帯域_10ms_rms_envelope_nrmse": 0.06702679120098197, "全帯域_10ms_envelope_correlation": 0.9955602620468076, "全帯域_10ms_modulation_mae_db": 0.43423149339505857, "全帯域_40ms_log_rms_mae_db": 0.5984603359700512, "全帯域_40ms_rms_envelope_nrmse": 0.03210205850726643, "全帯域_40ms_envelope_correlation": 0.9989816180837239, "全帯域_40ms_modulation_mae_db": 0.5150008086226682, "300–1000 Hz_10ms_log_rms_mae_db": 1.7222858132011951, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10056435747736478, "300–1000 Hz_10ms_envelope_correlation": 0.9926852674732574, "300–1000 Hz_10ms_modulation_mae_db": 0.71645797157612, "300–1000 Hz_40ms_log_rms_mae_db": 0.9912012774163104, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05712627992161877, "300–1000 Hz_40ms_envelope_correlation": 0.9977731751234405, "300–1000 Hz_40ms_modulation_mae_db": 0.7584180736315776, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0299853320970946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26675318322747943, "1000–3000 Hz_10ms_envelope_correlation": 0.9461287913023182, "1000–3000 Hz_10ms_modulation_mae_db": 0.7889668382108451, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5463364409535476, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17678659576454223, "1000–3000 Hz_40ms_envelope_correlation": 0.9746520372063296, "1000–3000 Hz_40ms_modulation_mae_db": 0.998978091401006, "3000–6000 Hz_10ms_log_rms_mae_db": 2.418130216750802, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4484806758850071, "3000–6000 Hz_10ms_envelope_correlation": 0.8915432923899378, "3000–6000 Hz_10ms_modulation_mae_db": 0.8843593950111597, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3561080416230737, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3899776364095361, "3000–6000 Hz_40ms_envelope_correlation": 0.9151488945854115, "3000–6000 Hz_40ms_modulation_mae_db": 0.9224196771592442, "mr_stft_db": 6.025179637213108, "band_mae_db": 0.4485545853315598, "rms_gain_error_db": -0.08616208190434588, "output_peak": 0.5485121706129746}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1850389592377537, "全帯域_10ms_rms_envelope_nrmse": 0.07969036015951825, "全帯域_10ms_envelope_correlation": 0.9937183417269722, "全帯域_10ms_modulation_mae_db": 0.23766332894015677, "全帯域_40ms_log_rms_mae_db": 0.5871662583655956, "全帯域_40ms_rms_envelope_nrmse": 0.04089975444218822, "全帯域_40ms_envelope_correlation": 0.9983276104189247, "全帯域_40ms_modulation_mae_db": 0.3602478421121121, "300–1000 Hz_10ms_log_rms_mae_db": 1.7901281843174532, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11733815886392066, "300–1000 Hz_10ms_envelope_correlation": 0.9899935895850196, "300–1000 Hz_10ms_modulation_mae_db": 0.765230085625395, "300–1000 Hz_40ms_log_rms_mae_db": 1.0415928748203156, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0659552108091529, "300–1000 Hz_40ms_envelope_correlation": 0.9970118935238615, "300–1000 Hz_40ms_modulation_mae_db": 0.9132543579714101, "1000–3000 Hz_10ms_log_rms_mae_db": 2.114565553959797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2575681273162662, "1000–3000 Hz_10ms_envelope_correlation": 0.9478037856134504, "1000–3000 Hz_10ms_modulation_mae_db": 0.41360842840269013, "1000–3000 Hz_40ms_log_rms_mae_db": 1.61275282689926, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1738431703798876, "1000–3000 Hz_40ms_envelope_correlation": 0.9741427102200809, "1000–3000 Hz_40ms_modulation_mae_db": 0.5268252760382243, "3000–6000 Hz_10ms_log_rms_mae_db": 2.485541924318464, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4577596902839777, "3000–6000 Hz_10ms_envelope_correlation": 0.8896234807949694, "3000–6000 Hz_10ms_modulation_mae_db": 0.7432422033404443, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3222502885059098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39061412817706453, "3000–6000 Hz_40ms_envelope_correlation": 0.9173888522248284, "3000–6000 Hz_40ms_modulation_mae_db": 0.8891445320986535, "mr_stft_db": 6.02146821622285, "band_mae_db": 0.44297116800400316, "rms_gain_error_db": -0.10955999578026633, "output_peak": 0.5196358152893299}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.136500959196615, "全帯域_10ms_rms_envelope_nrmse": 0.07779350326449752, "全帯域_10ms_envelope_correlation": 0.993980698342975, "全帯域_10ms_modulation_mae_db": 0.39877626563680835, "全帯域_40ms_log_rms_mae_db": 0.579829421460383, "全帯域_40ms_rms_envelope_nrmse": 0.04374037983726588, "全帯域_40ms_envelope_correlation": 0.998008628354512, "全帯域_40ms_modulation_mae_db": 0.3572599259187772, "300–1000 Hz_10ms_log_rms_mae_db": 1.732042515857547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09595322120067575, "300–1000 Hz_10ms_envelope_correlation": 0.9934347048879969, "300–1000 Hz_10ms_modulation_mae_db": 0.588690454164716, "300–1000 Hz_40ms_log_rms_mae_db": 1.0575873159388207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05408523125937788, "300–1000 Hz_40ms_envelope_correlation": 0.998098444044094, "300–1000 Hz_40ms_modulation_mae_db": 0.5185006658702386, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1165664520761274, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25778045517108694, "1000–3000 Hz_10ms_envelope_correlation": 0.9462978623070787, "1000–3000 Hz_10ms_modulation_mae_db": 0.5208870854344338, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5987404583320888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16668707468994257, "1000–3000 Hz_40ms_envelope_correlation": 0.9757613177240603, "1000–3000 Hz_40ms_modulation_mae_db": 0.6154652296051242, "3000–6000 Hz_10ms_log_rms_mae_db": 2.4937308702888354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.455749974461961, "3000–6000 Hz_10ms_envelope_correlation": 0.8900639956135119, "3000–6000 Hz_10ms_modulation_mae_db": 1.253651096252811, "3000–6000 Hz_40ms_log_rms_mae_db": 2.344286715464853, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3933835348373646, "3000–6000 Hz_40ms_envelope_correlation": 0.9155346895804236, "3000–6000 Hz_40ms_modulation_mae_db": 1.3502266144106223, "mr_stft_db": 6.019173695129344, "band_mae_db": 0.4833968841151786, "rms_gain_error_db": -0.08614235558245151, "output_peak": 0.5381247013970274}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1529322206663875, "全帯域_10ms_rms_envelope_nrmse": 0.07698008440522161, "全帯域_10ms_envelope_correlation": 0.9941564690466094, "全帯域_10ms_modulation_mae_db": 0.39267530349732943, "全帯域_40ms_log_rms_mae_db": 0.5831648695284064, "全帯域_40ms_rms_envelope_nrmse": 0.04273893206374301, "全帯域_40ms_envelope_correlation": 0.9981661968316935, "全帯域_40ms_modulation_mae_db": 0.419752976929179, "300–1000 Hz_10ms_log_rms_mae_db": 1.786173377138368, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1210392516641378, "300–1000 Hz_10ms_envelope_correlation": 0.9894629622398666, "300–1000 Hz_10ms_modulation_mae_db": 0.56276992783622, "300–1000 Hz_40ms_log_rms_mae_db": 1.0664322026663509, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0768562489387895, "300–1000 Hz_40ms_envelope_correlation": 0.9959530999990622, "300–1000 Hz_40ms_modulation_mae_db": 0.41566096558163085, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1683301522982523, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2570740952260767, "1000–3000 Hz_10ms_envelope_correlation": 0.9473170936140539, "1000–3000 Hz_10ms_modulation_mae_db": 0.3671878379959827, "1000–3000 Hz_40ms_log_rms_mae_db": 1.6675885161118755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1750428009728452, "1000–3000 Hz_40ms_envelope_correlation": 0.9733471888470798, "1000–3000 Hz_40ms_modulation_mae_db": 0.5252439113844283, "3000–6000 Hz_10ms_log_rms_mae_db": 2.501997407416048, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4459943366457394, "3000–6000 Hz_10ms_envelope_correlation": 0.8927340604688068, "3000–6000 Hz_10ms_modulation_mae_db": 0.693527978567754, "3000–6000 Hz_40ms_log_rms_mae_db": 2.363246464273413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.38367603458744937, "3000–6000 Hz_40ms_envelope_correlation": 0.918045151642947, "3000–6000 Hz_40ms_modulation_mae_db": 0.8859776702648817, "mr_stft_db": 6.019797587174065, "band_mae_db": 0.43263876169365106, "rms_gain_error_db": -0.11299272786332512, "output_peak": 0.5273804007481036}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2841284570769178, "全帯域_10ms_rms_envelope_nrmse": 0.08339351248307629, "全帯域_10ms_envelope_correlation": 0.9930776797541458, "全帯域_10ms_modulation_mae_db": 0.5689200471257336, "全帯域_40ms_log_rms_mae_db": 0.6916716516060977, "全帯域_40ms_rms_envelope_nrmse": 0.04740364199317782, "全帯域_40ms_envelope_correlation": 0.9976474409851644, "全帯域_40ms_modulation_mae_db": 0.784961521552422, "300–1000 Hz_10ms_log_rms_mae_db": 1.7738678046427565, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11284546649435438, "300–1000 Hz_10ms_envelope_correlation": 0.9905594189013738, "300–1000 Hz_10ms_modulation_mae_db": 0.42524173740510524, "300–1000 Hz_40ms_log_rms_mae_db": 1.0499008679976618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06672474011151985, "300–1000 Hz_40ms_envelope_correlation": 0.9966434272681347, "300–1000 Hz_40ms_modulation_mae_db": 0.4737842795404972, "1000–3000 Hz_10ms_log_rms_mae_db": 2.141578201950332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2686701352618905, "1000–3000 Hz_10ms_envelope_correlation": 0.9417297160491822, "1000–3000 Hz_10ms_modulation_mae_db": 0.5743442366156412, "1000–3000 Hz_40ms_log_rms_mae_db": 1.646732760865287, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18289628516799486, "1000–3000 Hz_40ms_envelope_correlation": 0.9708000907345116, "1000–3000 Hz_40ms_modulation_mae_db": 0.6560845733856236, "3000–6000 Hz_10ms_log_rms_mae_db": 2.480665360832129, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.46084826093635206, "3000–6000 Hz_10ms_envelope_correlation": 0.8877836137104251, "3000–6000 Hz_10ms_modulation_mae_db": 0.8709814123030665, "3000–6000 Hz_40ms_log_rms_mae_db": 2.321206686637964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.39273018924996655, "3000–6000 Hz_40ms_envelope_correlation": 0.916101897315376, "3000–6000 Hz_40ms_modulation_mae_db": 0.9604659760189457, "mr_stft_db": 6.014516111496735, "band_mae_db": 0.5285035553041106, "rms_gain_error_db": -0.08992625805590444, "output_peak": 0.5219963772583412}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2318392418260335, "全帯域_10ms_rms_envelope_nrmse": 0.08039727315210415, "全帯域_10ms_envelope_correlation": 0.9935923963065932, "全帯域_10ms_modulation_mae_db": 0.41650453734857845, "全帯域_40ms_log_rms_mae_db": 0.6592309550623628, "全帯域_40ms_rms_envelope_nrmse": 0.04053817892822884, "全帯域_40ms_envelope_correlation": 0.9983452010251801, "全帯域_40ms_modulation_mae_db": 0.2277369344412562, "300–1000 Hz_10ms_log_rms_mae_db": 1.7783188245020425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10365141420234938, "300–1000 Hz_10ms_envelope_correlation": 0.9923657889603761, "300–1000 Hz_10ms_modulation_mae_db": 0.6588089939451827, "300–1000 Hz_40ms_log_rms_mae_db": 1.0996059519778434, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06179929851647072, "300–1000 Hz_40ms_envelope_correlation": 0.9975207941065655, "300–1000 Hz_40ms_modulation_mae_db": 0.7071527883172918, "1000–3000 Hz_10ms_log_rms_mae_db": 2.091169235331205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.255711005589986, "1000–3000 Hz_10ms_envelope_correlation": 0.947016604607769, "1000–3000 Hz_10ms_modulation_mae_db": 0.4859753833072847, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5773545529668127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1727778350294726, "1000–3000 Hz_40ms_envelope_correlation": 0.9739752781139107, "1000–3000 Hz_40ms_modulation_mae_db": 0.5815467779544311, "3000–6000 Hz_10ms_log_rms_mae_db": 2.542385572061149, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4559801702509718, "3000–6000 Hz_10ms_envelope_correlation": 0.8950525913648628, "3000–6000 Hz_10ms_modulation_mae_db": 0.7773589222182397, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3648371778085986, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.396659163728143, "3000–6000 Hz_40ms_envelope_correlation": 0.9186591241740406, "3000–6000 Hz_40ms_modulation_mae_db": 0.8843793758955997, "mr_stft_db": 6.019864497617789, "band_mae_db": 0.5441652653292716, "rms_gain_error_db": -0.10458700652149014, "output_peak": 0.5154742747482525}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.141034378376244, "全帯域_10ms_rms_envelope_nrmse": 0.0774551749041004, "全帯域_10ms_envelope_correlation": 0.9941455576540738, "全帯域_10ms_modulation_mae_db": 0.4568480314439916, "全帯域_40ms_log_rms_mae_db": 0.5838476052461204, "全帯域_40ms_rms_envelope_nrmse": 0.0425660154473533, "全帯域_40ms_envelope_correlation": 0.9982655145047074, "全帯域_40ms_modulation_mae_db": 0.530011663456866, "300–1000 Hz_10ms_log_rms_mae_db": 1.7951964804373206, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1101487097690707, "300–1000 Hz_10ms_envelope_correlation": 0.9915301189026514, "300–1000 Hz_10ms_modulation_mae_db": 0.28937125797651936, "300–1000 Hz_40ms_log_rms_mae_db": 1.1092035011732273, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06828060836981852, "300–1000 Hz_40ms_envelope_correlation": 0.9971130684786034, "300–1000 Hz_40ms_modulation_mae_db": 0.3362400184378065, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0469963287430124, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2603077261111491, "1000–3000 Hz_10ms_envelope_correlation": 0.9468028155932818, "1000–3000 Hz_10ms_modulation_mae_db": 0.7393539919686646, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5026529641255972, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16668409374933993, "1000–3000 Hz_40ms_envelope_correlation": 0.9765557941750882, "1000–3000 Hz_40ms_modulation_mae_db": 0.9624636502322629, "3000–6000 Hz_10ms_log_rms_mae_db": 2.467224190808008, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4375311812736422, "3000–6000 Hz_10ms_envelope_correlation": 0.8974534013456619, "3000–6000 Hz_10ms_modulation_mae_db": 0.7269374322358969, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3422377193437716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3784053160302986, "3000–6000 Hz_40ms_envelope_correlation": 0.9208012550863658, "3000–6000 Hz_40ms_modulation_mae_db": 0.8981097325028301, "mr_stft_db": 6.0258428384895195, "band_mae_db": 0.4562929305841128, "rms_gain_error_db": -0.137289042160478, "output_peak": 0.5238311581197054}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.152298141057871, "spectral_convergence": 0.24237129603539917, "spectral_convergence_status": "ok", "amplitude_floor": 1.9267471569597447e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7727337769904645, "spectral_convergence": 0.22248204228213242, "spectral_convergence_status": "ok", "amplitude_floor": 1.271916808390449e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.128394871079962, "spectral_convergence": 0.22966138094839844, "spectral_convergence_status": "ok", "amplitude_floor": 1.1644398478092437e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001373055391651405, "output_power": 0.0013731833151690662, "delta_db": 0.0004046005012462063, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0033897235254818038, "output_power": 0.003254554403489129, "delta_db": -0.17672741826264138, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002841140646523882, "output_power": 0.00029560119831044564, "delta_db": 0.1721345690271855, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.671358771236327e-05, "output_power": 8.647604022406563e-05, "delta_db": -0.011913603520248499, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6620686952686656e-05, "output_power": 3.243037850920847e-05, "delta_db": 0.8573275919646455, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.533501052724457e-05, "output_power": 3.5829239880497815e-05, "delta_db": 1.5051650559453422, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4539454732035515, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0909677898653409, "pitch": null, "output_rms_dbfs": -22.943009254540048, "output_peak": 0.5203839110350614, "reference_rms_dbfs": -22.85204146467471, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.017808929709433}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023735521522065317, "log_rms_mae_db": 1.1692120359602658, "rms_envelope_nrmse": 0.0805294043267785, "envelope_correlation": 0.9935453460759083, "rms_gain_error_db": -0.0909245117255867, "reference_mean_power": 0.0051897925018650195, "output_mean_power": 0.005082267770412892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6580095170174236e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00099786392963893, "output_power": 0.0009593207180187244, "reference_active": true, "delta_db": -0.17107500383189492}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00025981544310873085, "output_power": 0.0002316784915787678, "reference_active": true, "delta_db": -0.49779244526259037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.167033379867013e-05, "output_power": 2.9985619990526507e-05, "reference_active": true, "delta_db": -1.4291394651812261}], "active_band_count": 3, "active_band_mae_db": 0.6993356380919038}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022053657198928817, "log_rms_mae_db": 0.6407757097999854, "rms_envelope_nrmse": 0.04394881051985692, "envelope_correlation": 0.9979904043429098, "rms_gain_error_db": -0.09092942315653943, "reference_mean_power": 0.0052155458817555326, "output_mean_power": 0.005107481802777063}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3916979728964525e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008985624617991317, "output_power": 0.000867719323999088, "reference_active": true, "delta_db": -0.1516900237711771}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001435836821660598, "output_power": 0.00012658275010399398, "reference_active": true, "delta_db": -0.547305596982616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3776224720397365e-06, "output_power": 1.7716985504770474e-06, "reference_active": true, "delta_db": -1.2775306697066937}], "active_band_count": 3, "active_band_mae_db": 0.6588420968201624}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018677846007938983, "log_rms_mae_db": 1.835501851291907, "rms_envelope_nrmse": 0.11163398351032965, "envelope_correlation": 0.9908304796049202, "rms_gain_error_db": -0.16292826320975945, "reference_mean_power": 0.0023293656178310037, "output_mean_power": 0.002243596916890588}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5191576623571145e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006490500325829311, "output_power": 0.000603243534498229, "reference_active": true, "delta_db": -0.3178550017394626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013228128531435308, "output_power": 0.00011076997213851041, "reference_active": true, "delta_db": -0.7707635961949777}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.716163060824064e-05, "output_power": 1.362896295216978e-05, "reference_active": true, "delta_db": -1.0009573887364394}], "active_band_count": 3, "active_band_mae_db": 0.6965253288902932}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017927232752442356, "log_rms_mae_db": 1.1619126385929461, "rms_envelope_nrmse": 0.06045481301628088, "envelope_correlation": 0.9973623468464398, "rms_gain_error_db": -0.1629174792583539, "reference_mean_power": 0.002341010543733861, "output_mean_power": 0.0022548186682539947}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.424681480871225e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006148042140063018, "output_power": 0.000568014232831968, "reference_active": true, "delta_db": -0.3437761754978199}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.176867672883116e-05, "output_power": 5.195515305935141e-05, "reference_active": true, "delta_db": -0.7513966895376635}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.094779276972058e-06, "output_power": 9.900277814780187e-07, "reference_active": true, "delta_db": -0.43679186433773143}], "active_band_count": 3, "active_band_mae_db": 0.5106549097910716}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.042588377582574e-05, "log_rms_mae_db": 2.105141544069969, "rms_envelope_nrmse": 0.23890315586779656, "envelope_correlation": 0.9551303488437358, "rms_gain_error_db": 0.10174338509210301, "reference_mean_power": 9.196537764000522e-05, "output_mean_power": 9.414531156898547e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3593984658429913e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0478584079151807e-05, "output_power": 1.98898422492342e-05, "reference_active": true, "delta_db": -0.12668586900471185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.015542002555929e-06, "output_power": 6.4783100625981016e-06, "reference_active": true, "delta_db": 0.3218696668938454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6245635384567205e-06, "output_power": 2.1259980321234183e-06, "reference_active": true, "delta_db": 1.1682615644836871}], "active_band_count": 3, "active_band_mae_db": 0.5389390334607481}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6202941099492215e-05, "log_rms_mae_db": 1.5917929680806773, "rms_envelope_nrmse": 0.16317032036812856, "envelope_correlation": 0.9772505198559491, "rms_gain_error_db": 0.10180752783882004, "reference_mean_power": 9.242233001846982e-05, "output_mean_power": 9.461449287035899e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9023164132494484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.934413521065944e-05, "output_power": 1.954113967761698e-05, "reference_active": true, "delta_db": 0.04400569890975438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.243887754162173e-06, "output_power": 3.7956861794340234e-06, "reference_active": true, "delta_db": 0.6822448031624563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3065359453687705e-07, "output_power": 1.7432085802908508e-07, "reference_active": true, "delta_db": 1.2522799237377984}], "active_band_count": 3, "active_band_mae_db": 0.6595101419366697}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.535256840646245e-05, "log_rms_mae_db": 2.4971850428211404, "rms_envelope_nrmse": 0.4556021308723383, "envelope_correlation": 0.8959867276181447, "rms_gain_error_db": 1.0476922899512155, "reference_mean_power": 2.9383590737924492e-05, "output_mean_power": 3.740021470765364e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.113292932890967e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.65042891211228e-06, "output_power": 1.0730237560774245e-05, "reference_active": true, "delta_db": 0.4606272114467579}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.01875786446688e-06, "output_power": 3.806841752289479e-06, "reference_active": true, "delta_db": -0.23527016281122787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.253460098772285e-07, "output_power": 1.323963999507107e-06, "reference_active": true, "delta_db": 2.613309498292804}], "active_band_count": 3, "active_band_mae_db": 1.10306895751693}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9034899521347454e-05, "log_rms_mae_db": 2.373553638815485, "rms_envelope_nrmse": 0.39466702500594647, "envelope_correlation": 0.9202634957618216, "rms_gain_error_db": 1.047677742311222, "reference_mean_power": 2.9530847534057242e-05, "output_mean_power": 3.758752116104976e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9758537203809177e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.950101722919118e-06, "output_power": 1.0987199841033319e-05, "reference_active": true, "delta_db": 0.43059503073478655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0870738801319347e-06, "output_power": 2.1832057604371595e-06, "reference_active": true, "delta_db": 0.19556845671970952}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.171675766600786e-08, "output_power": 1.0439918217803766e-07, "reference_active": true, "delta_db": 2.2829399456883106}], "active_band_count": 3, "active_band_mae_db": 0.9697011443809357}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8599654168541827, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f2e153af093bf480eb74b81457254e73479d333b4d91e9a49545911b499eafe4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9472783772413588, "standard_deviation": 0.0036896823933288176, "minimum": 0.9417297160491822, "maximum": 0.9551303488437358, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1017916000657237, "standard_deviation": 0.045786057467685395, "minimum": 2.0299853320970946, "maximum": 2.1683301522982523, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5536578544245363, "standard_deviation": 0.14664714326803424, "minimum": 0.3671878379959827, "maximum": 0.7889668382108451, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2578459854714664, "standard_deviation": 0.008997037638982939, "minimum": 0.23890315586779656, "maximum": 0.2686701352618905, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9745606171096263, "standard_deviation": 0.002028285080741806, "minimum": 0.9708000907345116, "maximum": 0.9772505198559491, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.5929939360418932, "standard_deviation": 0.05272583351271586, "minimum": 1.5026529641255972, "maximum": 1.6675885161118755, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6907647064922212, "standard_deviation": 0.18626678398334498, "minimum": 0.5252439113844283, "maximum": 0.998978091401006, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1722360220152692, "standard_deviation": 0.006424340961220657, "minimum": 0.16317032036812856, "maximum": 0.18289628516799486, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8925301454132901, "standard_deviation": 0.0033937918232596442, "minimum": 0.8877836137104251, "maximum": 0.8974534013456619, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.485857573162072, "standard_deviation": 0.035066194003600275, "minimum": 2.418130216750802, "maximum": 2.542385572061149, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8816409246807877, "standard_deviation": 0.1988867408997285, "minimum": 0.693527978567754, "maximum": 1.253651096252811, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.45224330257624873, "standard_deviation": 0.007668785944305674, "minimum": 0.4375311812736422, "maximum": 0.46084826093635206, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9177429200464018, "standard_deviation": 0.0021064885678183567, "minimum": 0.9151488945854115, "maximum": 0.9208012550863658, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.3484658415591335, "standard_deviation": 0.019496738194419334, "minimum": 2.321206686637964, "maximum": 2.373553638815485, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9700530903414641, "standard_deviation": 0.15721849683235492, "minimum": 0.8843793758955997, "maximum": 1.3502266144106223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.39001412850322115, "standard_deviation": 0.006092537787305793, "minimum": 0.3784053160302986, "maximum": 0.396659163728143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913577913194327, "standard_deviation": 0.0013881333162338602, "minimum": 0.9894629622398666, "maximum": 0.9934347048879969, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7766893564235737, "standard_deviation": 0.0359278525944226, "minimum": 1.7222858132011951, "maximum": 1.835501851291907, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5878869696774439, "standard_deviation": 0.16080597137698016, "minimum": 0.28937125797651936, "maximum": 0.765230085625395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10914682039777537, "standard_deviation": 0.00851324104563228, "minimum": 0.09595322120067575, "maximum": 0.1210392516641378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971845311737753, "standard_deviation": 0.0006730056736280783, "minimum": 0.9959530999990622, "maximum": 0.998098444044094, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0721795788229345, "standard_deviation": 0.05128399635025066, "minimum": 0.9912012774163104, "maximum": 1.1619126385929461, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5792082573926906, "standard_deviation": 0.19469875213604607, "minimum": 0.3362400184378065, "maximum": 0.9132543579714101, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06391030386787863, "standard_deviation": 0.007148941422852139, "minimum": 0.05408523125937788, "maximum": 0.0768562489387895, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.4738085779456799, "standard_deviation": 0.041435913030355265, "minimum": 0.43263876169365106, "maximum": 0.5441652653292716, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.020456439131605, "standard_deviation": 0.003723905446582027, "minimum": 6.014516111496735, "maximum": 6.0258428384895195, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5269173511510995, "standard_deviation": 0.011035957563171708, "minimum": 0.5154742747482525, "maximum": 0.5485121706129746, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10220340721670029, "standard_deviation": 0.017725933973058036, "minimum": -0.137289042160478, "maximum": -0.08614235558245151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939720938692607, "standard_deviation": 0.0007345636960676537, "minimum": 0.9930776797541458, "maximum": 0.9955602620468076, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.180709225637736, "standard_deviation": 0.0521492531046477, "minimum": 1.136500959196615, "maximum": 1.2841284570769178, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4506193306849451, "standard_deviation": 0.1356052624814054, "minimum": 0.23766332894015677, "maximum": 0.6993356380919038, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07790826298703483, "standard_deviation": 0.0048670288414705545, "minimum": 0.06702679120098197, "maximum": 0.08339351248307629, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998216576818352, "standard_deviation": 0.00038496863950954506, "minimum": 0.9976474409851644, "maximum": 0.9989816180837239, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6155183508798754, "standard_deviation": 0.04270555625576802, "minimum": 0.579829421460383, "maximum": 0.6916716516060977, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.48172672123168037, "standard_deviation": 0.1794278441989304, "minimum": 0.2277369344412562, "maximum": 0.784961521552422, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04174222146738506, "standard_deviation": 0.004434888209963133, "minimum": 0.03210205850726643, "maximum": 0.04740364199317782, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8230-279154-0032", "language": "en", "style": "normal", "speaker": "librispeech_8230", "source_utterance_id": "8230-279154-0032", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 62080, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8230-279154-0032.wav", "source_file_sha256": "86bc8f6514c0d803372889673e4d7c121f066def12776a66668bf308e40cd609", "decoded_pcm_sha256": "d59c4c75b338357fc4c79c9980c56076c2ae54441e0d223325fd8514435d4576", "original_source_sha256": "d60da130333b0e1adba450d171b7f753f423dce0f66a00448dfb036d43a7231d", "corpus_entry_id": "librispeech_8230-279154-0032"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.941014528274536, "predicted_mos": 3.941014528274536, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "6bea6e571970db55e3f2e4de99b31d4d51feef0f6cb5f219b681b9d92d4dccb6", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9859986789997492, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8230-279154-0032", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "6bea6e571970db55e3f2e4de99b31d4d51feef0f6cb5f219b681b9d92d4dccb6", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "13541ca079afe56ea25803a0e8910f0e6fb5f0e24063e867981a992039b868f4"}, "rows": [{"case": "librispeech_8230-279154-0032", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.3638715744018555, "predicted_mos": 2.3638715744018555, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "f018d8a2b6d91b16d379908a1d8ec7094a1638f7a0d11f9b9a98d242af1aba11", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.686378741000226, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8230-279154-0032", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f018d8a2b6d91b16d379908a1d8ec7094a1638f7a0d11f9b9a98d242af1aba11", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2172694544449767, "全帯域_10ms_rms_envelope_nrmse": 0.11306326195771114, "全帯域_10ms_envelope_correlation": 0.9901767231487186, "全帯域_10ms_modulation_mae_db": 0.44170241302426033, "全帯域_40ms_log_rms_mae_db": 0.6905609090277174, "全帯域_40ms_rms_envelope_nrmse": 0.06310256315342735, "全帯域_40ms_envelope_correlation": 0.9974104970003697, "全帯域_40ms_modulation_mae_db": 0.7433750478212703, "300–1000 Hz_10ms_log_rms_mae_db": 1.633612045038869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16901078924135127, "300–1000 Hz_10ms_envelope_correlation": 0.9823250660126626, "300–1000 Hz_10ms_modulation_mae_db": 0.5564237711269033, "300–1000 Hz_40ms_log_rms_mae_db": 1.0321269365960812, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10864700235441747, "300–1000 Hz_40ms_envelope_correlation": 0.9937617943106819, "300–1000 Hz_40ms_modulation_mae_db": 1.3061492151183247, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7917396640191718, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21443174302219165, "1000–3000 Hz_10ms_envelope_correlation": 0.9693711466817648, "1000–3000 Hz_10ms_modulation_mae_db": 1.136739727267354, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1980249988944685, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14385121702907536, "1000–3000 Hz_40ms_envelope_correlation": 0.985724320635071, "1000–3000 Hz_40ms_modulation_mae_db": 0.8829874710174529, "3000–6000 Hz_10ms_log_rms_mae_db": 1.907234520683046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.188372008025285, "3000–6000 Hz_10ms_envelope_correlation": 0.9760654494349473, "3000–6000 Hz_10ms_modulation_mae_db": 0.35971044365757426, "3000–6000 Hz_40ms_log_rms_mae_db": 1.247589565858859, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09976834414222084, "3000–6000 Hz_40ms_envelope_correlation": 0.9928936618301641, "3000–6000 Hz_40ms_modulation_mae_db": 0.3133174740467401, "mr_stft_db": 5.662838783822866, "band_mae_db": 0.18038624264925263, "rms_gain_error_db": -0.28476594375425535, "output_peak": 0.8106844098526276}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.304522701694853, "全帯域_10ms_rms_envelope_nrmse": 0.09825953730385947, "全帯域_10ms_envelope_correlation": 0.992197178517917, "全帯域_10ms_modulation_mae_db": 0.15957078866172603, "全帯域_40ms_log_rms_mae_db": 0.6626616776171058, "全帯域_40ms_rms_envelope_nrmse": 0.0422916904840918, "全帯域_40ms_envelope_correlation": 0.9984787466593191, "全帯域_40ms_modulation_mae_db": 0.3088314084859545, "300–1000 Hz_10ms_log_rms_mae_db": 1.6264791324168295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14222368252059026, "300–1000 Hz_10ms_envelope_correlation": 0.9866201922997476, "300–1000 Hz_10ms_modulation_mae_db": 0.36877886884434224, "300–1000 Hz_40ms_log_rms_mae_db": 1.0005406349289403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06688910361095031, "300–1000 Hz_40ms_envelope_correlation": 0.9969735877736912, "300–1000 Hz_40ms_modulation_mae_db": 0.558006291286817, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8349236719973823, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19576287865830433, "1000–3000 Hz_10ms_envelope_correlation": 0.9745026526560692, "1000–3000 Hz_10ms_modulation_mae_db": 0.4830784262601484, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2409041479467189, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12992304282408101, "1000–3000 Hz_40ms_envelope_correlation": 0.9884326379188486, "1000–3000 Hz_40ms_modulation_mae_db": 0.3982651137288085, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9202641015778423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19830028744648207, "3000–6000 Hz_10ms_envelope_correlation": 0.9731268717506258, "3000–6000 Hz_10ms_modulation_mae_db": 0.4240812323528869, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2311366423905787, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10921964193564292, "3000–6000 Hz_40ms_envelope_correlation": 0.9912937996670271, "3000–6000 Hz_40ms_modulation_mae_db": 0.5592377686666804, "mr_stft_db": 5.655506074214571, "band_mae_db": 0.39569400327371923, "rms_gain_error_db": -0.02484587634488905, "output_peak": 0.7806487420863446}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.344652507905394, "全帯域_10ms_rms_envelope_nrmse": 0.09872024934205731, "全帯域_10ms_envelope_correlation": 0.9922500760469861, "全帯域_10ms_modulation_mae_db": 0.35514937379349537, "全帯域_40ms_log_rms_mae_db": 0.7390756740043906, "全帯域_40ms_rms_envelope_nrmse": 0.05236133969572788, "全帯域_40ms_envelope_correlation": 0.9979219350927583, "全帯域_40ms_modulation_mae_db": 0.1803566648455448, "300–1000 Hz_10ms_log_rms_mae_db": 1.6496837668166096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12766837947139537, "300–1000 Hz_10ms_envelope_correlation": 0.989508817619407, "300–1000 Hz_10ms_modulation_mae_db": 0.4098818764633814, "300–1000 Hz_40ms_log_rms_mae_db": 1.0655723772187482, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07130916832275236, "300–1000 Hz_40ms_envelope_correlation": 0.9969669330337626, "300–1000 Hz_40ms_modulation_mae_db": 0.5139458559668875, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8679099464460174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21108514690091737, "1000–3000 Hz_10ms_envelope_correlation": 0.9690109859728995, "1000–3000 Hz_10ms_modulation_mae_db": 1.253799310279957, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2206684808063775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12455702257615084, "1000–3000 Hz_40ms_envelope_correlation": 0.9887282136725128, "1000–3000 Hz_40ms_modulation_mae_db": 0.7849635324515966, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9242852899547405, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18527723710639613, "3000–6000 Hz_10ms_envelope_correlation": 0.9763111482892671, "3000–6000 Hz_10ms_modulation_mae_db": 0.7080588201681591, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3268417557910588, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11757634214735438, "3000–6000 Hz_40ms_envelope_correlation": 0.9900566529246462, "3000–6000 Hz_40ms_modulation_mae_db": 0.4847497241878081, "mr_stft_db": 5.672932573707307, "band_mae_db": 0.14362063394993665, "rms_gain_error_db": -0.1635036983234613, "output_peak": 0.8762287888041601}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2796865862924176, "全帯域_10ms_rms_envelope_nrmse": 0.15121257035303767, "全帯域_10ms_envelope_correlation": 0.9818953158292123, "全帯域_10ms_modulation_mae_db": 0.7760223206756519, "全帯域_40ms_log_rms_mae_db": 0.7080073694396004, "全帯域_40ms_rms_envelope_nrmse": 0.09207358200603555, "全帯域_40ms_envelope_correlation": 0.992938391629283, "全帯域_40ms_modulation_mae_db": 0.6593547397506625, "300–1000 Hz_10ms_log_rms_mae_db": 1.7056050294985634, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2076325529777095, "300–1000 Hz_10ms_envelope_correlation": 0.9720230694242121, "300–1000 Hz_10ms_modulation_mae_db": 1.044633837964839, "300–1000 Hz_40ms_log_rms_mae_db": 1.103859455781992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13623002143995133, "300–1000 Hz_40ms_envelope_correlation": 0.9874699010437753, "300–1000 Hz_40ms_modulation_mae_db": 0.6176157552137443, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8098718003172196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19049920918707394, "1000–3000 Hz_10ms_envelope_correlation": 0.9756684047694725, "1000–3000 Hz_10ms_modulation_mae_db": 0.6986697617064551, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1958302641847531, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11496834160710355, "1000–3000 Hz_40ms_envelope_correlation": 0.9908530004718854, "1000–3000 Hz_40ms_modulation_mae_db": 0.645234612575308, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8480653380968568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19059951606338987, "3000–6000 Hz_10ms_envelope_correlation": 0.9750544522743365, "3000–6000 Hz_10ms_modulation_mae_db": 0.434884199585152, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2498885799723156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11699245032937546, "3000–6000 Hz_40ms_envelope_correlation": 0.9899758868285685, "3000–6000 Hz_40ms_modulation_mae_db": 0.7557757038620098, "mr_stft_db": 5.660540333270994, "band_mae_db": 0.31530873957631184, "rms_gain_error_db": 0.05479955072059796, "output_peak": 0.954797367633953}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3494294063151717, "全帯域_10ms_rms_envelope_nrmse": 0.09853488756566896, "全帯域_10ms_envelope_correlation": 0.9923011364667111, "全帯域_10ms_modulation_mae_db": 0.2762486227584859, "全帯域_40ms_log_rms_mae_db": 0.732946825518419, "全帯域_40ms_rms_envelope_nrmse": 0.051818166516584704, "全帯域_40ms_envelope_correlation": 0.9980008069332069, "全帯域_40ms_modulation_mae_db": 0.18549807825984632, "300–1000 Hz_10ms_log_rms_mae_db": 1.6957266459642832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13256557500039326, "300–1000 Hz_10ms_envelope_correlation": 0.9893737450202185, "300–1000 Hz_10ms_modulation_mae_db": 0.5031225844347469, "300–1000 Hz_40ms_log_rms_mae_db": 1.1488839757800535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08270365605204905, "300–1000 Hz_40ms_envelope_correlation": 0.9966392988893097, "300–1000 Hz_40ms_modulation_mae_db": 0.5144064870664593, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8730120843873432, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19377101732886975, "1000–3000 Hz_10ms_envelope_correlation": 0.9739034882950734, "1000–3000 Hz_10ms_modulation_mae_db": 1.0626514540333105, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2013334339132549, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11151162708107354, "1000–3000 Hz_40ms_envelope_correlation": 0.9910336041858715, "1000–3000 Hz_40ms_modulation_mae_db": 0.7223928145791864, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9382466327955186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2111178537083777, "3000–6000 Hz_10ms_envelope_correlation": 0.9714234988388426, "3000–6000 Hz_10ms_modulation_mae_db": 0.6911739579686221, "3000–6000 Hz_40ms_log_rms_mae_db": 1.257393237276469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12419680352755454, "3000–6000 Hz_40ms_envelope_correlation": 0.9900239029972503, "3000–6000 Hz_40ms_modulation_mae_db": 0.7318268108900675, "mr_stft_db": 5.642666390374597, "band_mae_db": 0.372965999597088, "rms_gain_error_db": -0.17078677174291637, "output_peak": 0.7844586899394669}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2601885122359435, "全帯域_10ms_rms_envelope_nrmse": 0.09881068953505961, "全帯域_10ms_envelope_correlation": 0.9927076001072261, "全帯域_10ms_modulation_mae_db": 0.3215156722852277, "全帯域_40ms_log_rms_mae_db": 0.6250767797077634, "全帯域_40ms_rms_envelope_nrmse": 0.05752532389117129, "全帯域_40ms_envelope_correlation": 0.9980243699633793, "全帯域_40ms_modulation_mae_db": 0.5879073732410738, "300–1000 Hz_10ms_log_rms_mae_db": 1.5696093625667356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13993879914055957, "300–1000 Hz_10ms_envelope_correlation": 0.9889105796732967, "300–1000 Hz_10ms_modulation_mae_db": 0.4913660275486174, "300–1000 Hz_40ms_log_rms_mae_db": 1.0104306951775686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.091605247362234, "300–1000 Hz_40ms_envelope_correlation": 0.9966287898537554, "300–1000 Hz_40ms_modulation_mae_db": 0.8837150248634883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.837997959204074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20488645506644365, "1000–3000 Hz_10ms_envelope_correlation": 0.9708134119277838, "1000–3000 Hz_10ms_modulation_mae_db": 0.884934847234116, "1000–3000 Hz_40ms_log_rms_mae_db": 1.206412683612812, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12953332381841617, "1000–3000 Hz_40ms_envelope_correlation": 0.9877705763261947, "1000–3000 Hz_40ms_modulation_mae_db": 0.7732429526127887, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9146263008803572, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20244871262627606, "3000–6000 Hz_10ms_envelope_correlation": 0.9717078125053141, "3000–6000 Hz_10ms_modulation_mae_db": 0.5893580951132079, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2380198434301755, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13460754121304583, "3000–6000 Hz_40ms_envelope_correlation": 0.9870928818714348, "3000–6000 Hz_40ms_modulation_mae_db": 0.36288006287448676, "mr_stft_db": 5.667835093322599, "band_mae_db": 0.17882764415173893, "rms_gain_error_db": -0.2810065948294067, "output_peak": 0.7550670467643676}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2990136070276785, "全帯域_10ms_rms_envelope_nrmse": 0.0947455053300624, "全帯域_10ms_envelope_correlation": 0.9927933251277002, "全帯域_10ms_modulation_mae_db": 0.3596112444288102, "全帯域_40ms_log_rms_mae_db": 0.6970268896116919, "全帯域_40ms_rms_envelope_nrmse": 0.04653128770424206, "全帯域_40ms_envelope_correlation": 0.9982876072979451, "全帯域_40ms_modulation_mae_db": 0.34356160406021985, "300–1000 Hz_10ms_log_rms_mae_db": 1.7549373984370302, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1291223920346556, "300–1000 Hz_10ms_envelope_correlation": 0.9893159328493354, "300–1000 Hz_10ms_modulation_mae_db": 0.3548498193426193, "300–1000 Hz_40ms_log_rms_mae_db": 1.127687365521908, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.070300153326898, "300–1000 Hz_40ms_envelope_correlation": 0.9971581064634966, "300–1000 Hz_40ms_modulation_mae_db": 0.3929144974990886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.849443738229682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2606401995347373, "1000–3000 Hz_10ms_envelope_correlation": 0.955026542011491, "1000–3000 Hz_10ms_modulation_mae_db": 1.0819017314748762, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2603756551191414, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1846802995037323, "1000–3000 Hz_40ms_envelope_correlation": 0.9765370795643481, "1000–3000 Hz_40ms_modulation_mae_db": 1.147587316258822, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0017649647832747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2137328363593783, "3000–6000 Hz_10ms_envelope_correlation": 0.9691822941557898, "3000–6000 Hz_10ms_modulation_mae_db": 0.20829443599500141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3656603576553539, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11659996671826801, "3000–6000 Hz_40ms_envelope_correlation": 0.9902616521664985, "3000–6000 Hz_40ms_modulation_mae_db": 0.2251986782240575, "mr_stft_db": 5.666884052255509, "band_mae_db": 0.16626078363521424, "rms_gain_error_db": -0.12130713047471228, "output_peak": 0.7498976520573577}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2493710178666773, "全帯域_10ms_rms_envelope_nrmse": 0.09365790776599284, "全帯域_10ms_envelope_correlation": 0.9929973968148545, "全帯域_10ms_modulation_mae_db": 0.1594584871784239, "全帯域_40ms_log_rms_mae_db": 0.6339014288198381, "全帯域_40ms_rms_envelope_nrmse": 0.05704758940316207, "全帯域_40ms_envelope_correlation": 0.9974045771179262, "全帯域_40ms_modulation_mae_db": 0.2811818070704086, "300–1000 Hz_10ms_log_rms_mae_db": 1.5919179900680196, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13738535740242042, "300–1000 Hz_10ms_envelope_correlation": 0.9882319180017217, "300–1000 Hz_10ms_modulation_mae_db": 0.32581522176514716, "300–1000 Hz_40ms_log_rms_mae_db": 1.0178920841140835, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09139684070892869, "300–1000 Hz_40ms_envelope_correlation": 0.9952748713934639, "300–1000 Hz_40ms_modulation_mae_db": 0.35772300346984215, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8540253832254843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19852902826012847, "1000–3000 Hz_10ms_envelope_correlation": 0.9733588323790735, "1000–3000 Hz_10ms_modulation_mae_db": 0.49027617692765135, "1000–3000 Hz_40ms_log_rms_mae_db": 1.231005278269322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12719175217501213, "1000–3000 Hz_40ms_envelope_correlation": 0.9900172041010123, "1000–3000 Hz_40ms_modulation_mae_db": 0.36775791948643716, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9087087258641033, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20340746582635041, "3000–6000 Hz_10ms_envelope_correlation": 0.9714891328531263, "3000–6000 Hz_10ms_modulation_mae_db": 0.8517704782254331, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2818497687836725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1364571250731813, "3000–6000 Hz_40ms_envelope_correlation": 0.9863645200029154, "3000–6000 Hz_40ms_modulation_mae_db": 0.6644298153694471, "mr_stft_db": 5.655435035302776, "band_mae_db": 0.23130091570282993, "rms_gain_error_db": -0.13914202131542372, "output_peak": 0.7765169145055404}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 62080, "output_samples": 62080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.071787785202774, "spectral_convergence": 0.3229031444659713, "spectral_convergence_status": "ok", "amplitude_floor": 2.580715109285093e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.418742748690809, "spectral_convergence": 0.28921014259411215, "spectral_convergence_status": "ok", "amplitude_floor": 1.9335078654426385e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.497985817575015, "spectral_convergence": 0.3080159450730803, "spectral_convergence_status": "ok", "amplitude_floor": 1.5577782358148932e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007926248856121452, "output_power": 0.0007840741826999784, "delta_db": -0.047105494919735015, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003328913174136213, "output_power": 0.0030584654626994163, "delta_db": -0.36798887234824607, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00045950159700284816, "output_power": 0.00042603986223848896, "delta_db": -0.3283678953771306, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012582167228939335, "output_power": 0.0001311384110624928, "delta_db": 0.17974464163732354, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.488237987418478e-05, "output_power": 3.546296140049607e-05, "delta_db": 0.07168892038133948, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.708544414240077e-05, "output_power": 6.844952874841134e-05, "delta_db": 0.08742163123174113, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18038624264925263, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.28476594375425535, "pitch": null, "output_rms_dbfs": -23.46437244866538, "output_peak": 0.8106844098526276, "reference_rms_dbfs": -23.179606504911124, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.662838783822866}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 62080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.00032928880743725145, "log_rms_mae_db": 1.2172694544449767, "rms_envelope_nrmse": 0.11306326195771114, "envelope_correlation": 0.9901767231487186, "rms_gain_error_db": -0.2847702231069677, "reference_mean_power": 0.004815032927514685, "output_mean_power": 0.004509436119173556}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.902918899290217e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008077922828075312, "output_power": 0.0007710103458612352, "reference_active": true, "delta_db": -0.20239494145186493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023728667544934464, "output_power": 0.00021137344410360697, "reference_active": true, "delta_db": -0.5022292778841335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1018322593991373e-05, "output_power": 1.8220057844700633e-05, "reference_active": true, "delta_db": -0.6204830197367825}], "active_band_count": 3, "active_band_mae_db": 0.44170241302426033}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0003041508818572066, "log_rms_mae_db": 0.6905609090277174, "rms_envelope_nrmse": 0.06310256315342735, "envelope_correlation": 0.9974104970003697, "rms_gain_error_db": -0.2847743416199401, "reference_mean_power": 0.004852582666148238, "output_mean_power": 0.004544598370113806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.7242336092637014e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007426996521093883, "output_power": 0.0007086803091712293, "reference_active": true, "delta_db": -0.20362854518936696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010890208813537604, "output_power": 9.604731641904869e-05, "reference_active": true, "delta_db": -0.545509720905926}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4730800018618504e-06, "output_power": 1.047436360767676e-06, "reference_active": true, "delta_db": -1.480986877368518}], "active_band_count": 3, "active_band_mae_db": 0.7433750478212703}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.0002965570632263201, "log_rms_mae_db": 1.633612045038869, "rms_envelope_nrmse": 0.16901078924135127, "envelope_correlation": 0.9823250660126626, "rms_gain_error_db": -0.5092074939924947, "reference_mean_power": 0.002818219366063788, "output_mean_power": 0.002506421143812317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.2520173458640518e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044507178368220307, "output_power": 0.000416664883745406, "reference_active": true, "delta_db": -0.2864316215497623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017793142616898364, "output_power": 0.0001487158906505139, "reference_active": true, "delta_db": -0.7789528334477119}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6359305994808858e-05, "output_power": 1.423561355443102e-05, "reference_active": true, "delta_db": -0.6038868583832357}], "active_band_count": 3, "active_band_mae_db": 0.5564237711269033}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0002772405007925501, "log_rms_mae_db": 1.0321269365960812, "rms_envelope_nrmse": 0.10864700235441747, "envelope_correlation": 0.9937617943106819, "rms_gain_error_db": -0.5092077415355731, "reference_mean_power": 0.0028402077631144488, "output_mean_power": 0.002525976675290335}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.1541839932451843e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134314303572302, "output_power": 0.0003872843933029858, "reference_active": true, "delta_db": -0.283734937383489}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.103778868784845e-05, "output_power": 6.73700863914478e-05, "reference_active": true, "delta_db": -0.8022047747129877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3787553243482739e-06, "output_power": 7.181851306838737e-07, "reference_active": true, "delta_db": -2.8325079332584977}], "active_band_count": 3, "active_band_mae_db": 1.3061492151183247}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 6.247605344635176e-05, "log_rms_mae_db": 1.7917396640191718, "rms_envelope_nrmse": 0.21443174302219165, "envelope_correlation": 0.9693711466817648, "rms_gain_error_db": 0.19784219221404606, "reference_mean_power": 0.00014490237542137544, "output_mean_power": 0.00015165604440958355}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 6.946266094979904e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.0294766814481675e-05, "output_power": 2.8781929656164416e-05, "reference_active": true, "delta_db": -0.2224770656170421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.606273878647975e-06, "output_power": 7.774623796994447e-06, "reference_active": true, "delta_db": -0.441357792659968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.077953085521739e-06, "output_power": 2.028795590718343e-06, "reference_active": true, "delta_db": 2.746384323525052}], "active_band_count": 3, "active_band_mae_db": 1.136739727267354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 5.585752960561645e-05, "log_rms_mae_db": 1.1980249988944685, "rms_envelope_nrmse": 0.14385121702907536, "envelope_correlation": 0.985724320635071, "rms_gain_error_db": 0.19783951081659512, "reference_mean_power": 0.00014603285132237434, "output_mean_power": 0.0001528391156271271}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 6.45401857377675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9246286179300554e-05, "output_power": 2.7823678309212704e-05, "reference_active": true, "delta_db": -0.21656181838446933}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6077467608233e-06, "output_power": 3.513314587373602e-06, "reference_active": true, "delta_db": -0.11519006531910805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.605305892606176e-08, "output_power": 1.467193403451964e-07, "reference_active": true, "delta_db": 2.317210529348781}], "active_band_count": 3, "active_band_mae_db": 0.8829874710174529}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 3.703669227124254e-05, "log_rms_mae_db": 1.907234520683046, "rms_envelope_nrmse": 0.188372008025285, "envelope_correlation": 0.9760654494349473, "rms_gain_error_db": 0.08985536076731174, "reference_mean_power": 5.2354216813681856e-05, "output_mean_power": 5.3448706976167225e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 3.452458222743894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.744222722260824e-05, "output_power": 1.725524276402719e-05, "reference_active": true, "delta_db": -0.046808660527452486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.482818619951907e-06, "output_power": 7.123579879991515e-06, "reference_active": true, "delta_db": -0.2136691964445061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0100258421135412e-06, "output_power": 1.2195450085014866e-06, "reference_active": true, "delta_db": 0.8186534740007643}], "active_band_count": 3, "active_band_mae_db": 0.35971044365757426}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 2.970153096248883e-05, "log_rms_mae_db": 1.247589565858859, "rms_envelope_nrmse": 0.09976834414222084, "envelope_correlation": 0.9928936618301641, "rms_gain_error_db": 0.08985422346664435, "reference_mean_power": 5.276267962923645e-05, "output_mean_power": 5.38656947979758e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 3.130411452597279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7491368881094015e-05, "output_power": 1.7528217478271986e-05, "reference_active": true, "delta_db": 0.009139541605199029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.132236890482271e-06, "output_power": 3.914133951019267e-06, "reference_active": true, "delta_db": -0.23549526372166502}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.718357196867276e-08, "output_power": 7.884875713027119e-08, "reference_active": true, "delta_db": 0.6953176168133561}], "active_band_count": 3, "active_band_mae_db": 0.3133174740467401}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8264269761428882, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c3b2725caea98a670f4122d556b119faca419e5e8fce42beb1938016ecd8add5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9702069330867035, "standard_deviation": 0.006598926182264759, "minimum": 0.955026542011491, "maximum": 0.9756684047694725, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.839865530978297, "standard_deviation": 0.027828709368187474, "minimum": 1.7917396640191718, "maximum": 1.8730120843873432, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8865064293979835, "standard_deviation": 0.29803279008839195, "minimum": 0.4830784262601484, "maximum": 1.253799310279957, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2087007097448333, "standard_deviation": 0.022601612891257818, "minimum": 0.19049920918707394, "maximum": 0.2606401995347373, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9873870796094681, "standard_deviation": 0.004717043181349795, "minimum": 0.9765370795643481, "maximum": 0.9910336041858715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2193193678433558, "standard_deviation": 0.02325377549079447, "minimum": 1.1958302641847531, "maximum": 1.2603756551191414, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7153039665888, "standard_deviation": 0.2533935077796266, "minimum": 0.36775791948643716, "maximum": 1.147587316258822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1332770783268306, "standard_deviation": 0.02299498107818139, "minimum": 0.11151162708107354, "maximum": 0.1846802995037323, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9730450825127812, "standard_deviation": 0.0025521418522396165, "minimum": 0.9691822941557898, "maximum": 0.9763111482892671, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9203994843294674, "standard_deviation": 0.04230759538586737, "minimum": 1.8480653380968568, "maximum": 2.0017649647832747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5334164578832546, "standard_deviation": 0.21294098395820205, "minimum": 0.20829443599500141, "maximum": 0.8517704782254331, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19915698964524192, "standard_deviation": 0.010470006251633503, "minimum": 0.18527723710639613, "maximum": 0.2137328363593783, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9897453697860632, "standard_deviation": 0.0021135406593136823, "minimum": 0.9863645200029154, "maximum": 0.9928936618301641, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2747974688948105, "standard_deviation": 0.04771477579656272, "minimum": 1.2311366423905787, "maximum": 1.3656603576553539, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5121770047651621, "standard_deviation": 0.1992577307365457, "minimum": 0.2251986782240575, "maximum": 0.7557757038620098, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11942727688583042, "standard_deviation": 0.012254573543656699, "minimum": 0.09976834414222084, "maximum": 0.1364571250731813, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9857886651125751, "standard_deviation": 0.0060595776179206435, "minimum": 0.9720230694242121, "maximum": 0.989508817619407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6534464213508675, "standard_deviation": 0.06182666003103694, "minimum": 1.5696093625667356, "maximum": 1.7549373984370302, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5068590009363245, "standard_deviation": 0.23184562648227366, "minimum": 0.32581522176514716, "maximum": 1.044633837964839, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1481934409736344, "standard_deviation": 0.027297763056380365, "minimum": 0.12766837947139537, "maximum": 0.2076325529777095, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.995109160345242, "standard_deviation": 0.003294526305826892, "minimum": 0.9874699010437753, "maximum": 0.9971581064634966, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.063374190639922, "standard_deviation": 0.05718151408370758, "minimum": 1.0005406349289403, "maximum": 1.1488839757800535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6430595163105814, "standard_deviation": 0.3122946991327375, "minimum": 0.35772300346984215, "maximum": 1.3061492151183247, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08988514914727265, "standard_deviation": 0.023342571014902103, "minimum": 0.06688910361095031, "maximum": 0.13623002143995133, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24804562031701144, "standard_deviation": 0.0994068949850293, "minimum": 0.14362063394993665, "maximum": 0.39569400327371923, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.660579792033902, "standard_deviation": 0.00943534391105474, "minimum": 5.642666390374597, "maximum": 5.672932573707307, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8110374514554772, "standard_deviation": 0.07031488520242646, "minimum": 0.7498976520573577, "maximum": 0.954797367633953, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.14131981075805833, "standard_deviation": 0.11581009631314208, "minimum": -0.28476594375425535, "maximum": 0.05479955072059796, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9909148440074157, "standard_deviation": 0.0037478648414208684, "minimum": 0.9818953158292123, "maximum": 0.9929973968148545, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.288016724222889, "standard_deviation": 0.045849093494632454, "minimum": 1.2172694544449767, "maximum": 1.3494294063151717, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35615986535076016, "standard_deviation": 0.19586426277209504, "minimum": 0.1594584871784239, "maximum": 0.7760223206756519, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10587557614418117, "standard_deviation": 0.019247474227780476, "minimum": 0.09365790776599284, "maximum": 0.15121257035303767, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973083664617735, "standard_deviation": 0.0018052966072731615, "minimum": 0.992938391629283, "maximum": 0.9984787466593191, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6861571942183158, "standard_deviation": 0.04245691865855877, "minimum": 0.6250767797077634, "maximum": 0.7390756740043906, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4112583404418726, "standard_deviation": 0.2201738015664859, "minimum": 0.1803566648455448, "maximum": 0.7433750478212703, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05784394285680534, "standard_deviation": 0.01528899986904011, "minimum": 0.0422916904840918, "maximum": 0.09207358200603555, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0032", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.52359938621521, "predicted_mos": 2.52359938621521, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "c20d82c312e868d7a2bba0596112df0c51bd299348baf57a22cc35175917fb45", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.731799599714577, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.223300955276722, "全帯域_10ms_rms_envelope_nrmse": 0.11118841481055637, "全帯域_10ms_envelope_correlation": 0.9905564360050045, "全帯域_10ms_modulation_mae_db": 0.5181084610441292, "全帯域_40ms_log_rms_mae_db": 0.6986073010342088, "全帯域_40ms_rms_envelope_nrmse": 0.06274868384550905, "全帯域_40ms_envelope_correlation": 0.9974839639559694, "全帯域_40ms_modulation_mae_db": 0.821845494717316, "300–1000 Hz_10ms_log_rms_mae_db": 1.6134142370815578, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1597134128941903, "300–1000 Hz_10ms_envelope_correlation": 0.9843740423235934, "300–1000 Hz_10ms_modulation_mae_db": 0.7014423529644719, "300–1000 Hz_40ms_log_rms_mae_db": 0.9845650781869665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10436896366206953, "300–1000 Hz_40ms_envelope_correlation": 0.9943326246753735, "300–1000 Hz_40ms_modulation_mae_db": 1.3388623846911631, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7359823877447942, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2079015079168785, "1000–3000 Hz_10ms_envelope_correlation": 0.9703832668380801, "1000–3000 Hz_10ms_modulation_mae_db": 1.018062527472276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.095188018117724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13320265343816604, "1000–3000 Hz_40ms_envelope_correlation": 0.9872320319762137, "1000–3000 Hz_40ms_modulation_mae_db": 0.7422200608724773, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8431547871369882, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18503822390567196, "3000–6000 Hz_10ms_envelope_correlation": 0.9768245406764415, "3000–6000 Hz_10ms_modulation_mae_db": 0.21033748965198987, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2205826641380793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09909056670155633, "3000–6000 Hz_40ms_envelope_correlation": 0.9929646265989224, "3000–6000 Hz_40ms_modulation_mae_db": 0.17870088698305722, "mr_stft_db": 5.593917652317707, "band_mae_db": 0.1495865441806486, "rms_gain_error_db": -0.2900344820334043, "output_peak": 0.7975482547727808}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.348076739739724, "全帯域_10ms_rms_envelope_nrmse": 0.10893635682930952, "全帯域_10ms_envelope_correlation": 0.990476480921081, "全帯域_10ms_modulation_mae_db": 0.29540670433449345, "全帯域_40ms_log_rms_mae_db": 0.6683142394067336, "全帯域_40ms_rms_envelope_nrmse": 0.052162240588132895, "全帯域_40ms_envelope_correlation": 0.9976987452034942, "全帯域_40ms_modulation_mae_db": 0.4129096724138946, "300–1000 Hz_10ms_log_rms_mae_db": 1.6368733941964413, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16022100500971592, "300–1000 Hz_10ms_envelope_correlation": 0.983159297650699, "300–1000 Hz_10ms_modulation_mae_db": 0.5224520787790355, "300–1000 Hz_40ms_log_rms_mae_db": 1.020228334773399, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08309327594275061, "300–1000 Hz_40ms_envelope_correlation": 0.9952772336017076, "300–1000 Hz_40ms_modulation_mae_db": 0.36664103624559813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8075578702374122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1676488275836672, "1000–3000 Hz_10ms_envelope_correlation": 0.980549593718012, "1000–3000 Hz_10ms_modulation_mae_db": 0.22386367064217302, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1286671955565593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10002964144877388, "1000–3000 Hz_40ms_envelope_correlation": 0.992742811794048, "1000–3000 Hz_40ms_modulation_mae_db": 0.17583539184856925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8803279100561427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1999655312497041, "3000–6000 Hz_10ms_envelope_correlation": 0.9729858076055158, "3000–6000 Hz_10ms_modulation_mae_db": 0.48681414624259384, "3000–6000 Hz_40ms_log_rms_mae_db": 1.197499919653351, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11204568115471698, "3000–6000 Hz_40ms_envelope_correlation": 0.9909589207352602, "3000–6000 Hz_40ms_modulation_mae_db": 0.6102360483801215, "mr_stft_db": 5.588823143190766, "band_mae_db": 0.3670512988869894, "rms_gain_error_db": 0.0314925124963491, "output_peak": 0.7757133207166038}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3136654384152697, "全帯域_10ms_rms_envelope_nrmse": 0.09431731145718797, "全帯域_10ms_envelope_correlation": 0.9929362990408954, "全帯域_10ms_modulation_mae_db": 0.3246488273306844, "全帯域_40ms_log_rms_mae_db": 0.7263336817820928, "全帯域_40ms_rms_envelope_nrmse": 0.051150391106787166, "全帯域_40ms_envelope_correlation": 0.9980121368472588, "全帯域_40ms_modulation_mae_db": 0.2257141773214919, "300–1000 Hz_10ms_log_rms_mae_db": 1.6488061515832282, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12341834210324298, "300–1000 Hz_10ms_envelope_correlation": 0.9901829478480697, "300–1000 Hz_10ms_modulation_mae_db": 0.4280352451078504, "300–1000 Hz_40ms_log_rms_mae_db": 1.0438246440632526, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0690254166782563, "300–1000 Hz_40ms_envelope_correlation": 0.9971343671386905, "300–1000 Hz_40ms_modulation_mae_db": 0.6170288746167197, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8681678109931865, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18442053456370455, "1000–3000 Hz_10ms_envelope_correlation": 0.9764024137220669, "1000–3000 Hz_10ms_modulation_mae_db": 0.9644515392846779, "1000–3000 Hz_40ms_log_rms_mae_db": 1.149066852432814, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10039844425411809, "1000–3000 Hz_40ms_envelope_correlation": 0.9932408754945785, "1000–3000 Hz_40ms_modulation_mae_db": 0.5270759895352409, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8404161023779186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19118755289363762, "3000–6000 Hz_10ms_envelope_correlation": 0.9747645465575699, "3000–6000 Hz_10ms_modulation_mae_db": 0.34568211098420304, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2438814063700396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12232289846006736, "3000–6000 Hz_40ms_envelope_correlation": 0.9891599019609394, "3000–6000 Hz_40ms_modulation_mae_db": 0.615272290505737, "mr_stft_db": 5.611705411965741, "band_mae_db": 0.1903854143006534, "rms_gain_error_db": -0.15699051506785183, "output_peak": 0.8805874956410726}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2784417574259488, "全帯域_10ms_rms_envelope_nrmse": 0.15017454130528404, "全帯域_10ms_envelope_correlation": 0.9821752823694183, "全帯域_10ms_modulation_mae_db": 0.8208449997650328, "全帯域_40ms_log_rms_mae_db": 0.7031422501958973, "全帯域_40ms_rms_envelope_nrmse": 0.09046746006349099, "全帯域_40ms_envelope_correlation": 0.9932107649465907, "全帯域_40ms_modulation_mae_db": 0.624108900954889, "300–1000 Hz_10ms_log_rms_mae_db": 1.6947478489297951, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2063507387665112, "300–1000 Hz_10ms_envelope_correlation": 0.972489951917708, "300–1000 Hz_10ms_modulation_mae_db": 1.210895372561844, "300–1000 Hz_40ms_log_rms_mae_db": 1.0787536962777922, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13257427955681786, "300–1000 Hz_40ms_envelope_correlation": 0.988210625985943, "300–1000 Hz_40ms_modulation_mae_db": 0.7276449226068858, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8037605965998298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1755615038810444, "1000–3000 Hz_10ms_envelope_correlation": 0.9787198659028847, "1000–3000 Hz_10ms_modulation_mae_db": 0.3276803732583917, "1000–3000 Hz_40ms_log_rms_mae_db": 1.140303661165217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09833165303998093, "1000–3000 Hz_40ms_envelope_correlation": 0.9929850217902823, "1000–3000 Hz_40ms_modulation_mae_db": 0.4606965975085142, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8411782035179967, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1862427390287368, "3000–6000 Hz_10ms_envelope_correlation": 0.9761198296685898, "3000–6000 Hz_10ms_modulation_mae_db": 0.6197492011933222, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2547495524861296, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11437521807753234, "3000–6000 Hz_40ms_envelope_correlation": 0.9904378099057289, "3000–6000 Hz_40ms_modulation_mae_db": 0.5727105430612923, "mr_stft_db": 5.589898206747257, "band_mae_db": 0.2502229113768126, "rms_gain_error_db": 0.06523725962188126, "output_peak": 0.9501961688381586}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3762342139739814, "全帯域_10ms_rms_envelope_nrmse": 0.09885722328028883, "全帯域_10ms_envelope_correlation": 0.9922909681649791, "全帯域_10ms_modulation_mae_db": 0.18556880476849347, "全帯域_40ms_log_rms_mae_db": 0.7626339532022886, "全帯域_40ms_rms_envelope_nrmse": 0.053701556891225355, "全帯域_40ms_envelope_correlation": 0.9978817092456775, "全帯域_40ms_modulation_mae_db": 0.13033176865388552, "300–1000 Hz_10ms_log_rms_mae_db": 1.6898572451511034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13149595521981744, "300–1000 Hz_10ms_envelope_correlation": 0.9895338958494224, "300–1000 Hz_10ms_modulation_mae_db": 0.5242983185571694, "300–1000 Hz_40ms_log_rms_mae_db": 1.1052159292407366, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08013533222540685, "300–1000 Hz_40ms_envelope_correlation": 0.996899437701851, "300–1000 Hz_40ms_modulation_mae_db": 0.5464490888949013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.815988624766118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18356082611077587, "1000–3000 Hz_10ms_envelope_correlation": 0.9765024413231436, "1000–3000 Hz_10ms_modulation_mae_db": 0.8120556537588725, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0851711446448702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10862764889620352, "1000–3000 Hz_40ms_envelope_correlation": 0.9916238295359872, "1000–3000 Hz_40ms_modulation_mae_db": 0.7098460282163298, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8732426570233087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21750027874272337, "3000–6000 Hz_10ms_envelope_correlation": 0.9701807945653653, "3000–6000 Hz_10ms_modulation_mae_db": 0.7677729197863741, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1897419206440305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12650509921415018, "3000–6000 Hz_40ms_envelope_correlation": 0.9900604288780083, "3000–6000 Hz_40ms_modulation_mae_db": 0.7921535305940317, "mr_stft_db": 5.5763911632452015, "band_mae_db": 0.3756313494834897, "rms_gain_error_db": -0.18559853489798664, "output_peak": 0.7882357393072683}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2632927916475303, "全帯域_10ms_rms_envelope_nrmse": 0.09914487383087596, "全帯域_10ms_envelope_correlation": 0.9925612880074558, "全帯域_10ms_modulation_mae_db": 0.3282063028537044, "全帯域_40ms_log_rms_mae_db": 0.6158379993105094, "全帯域_40ms_rms_envelope_nrmse": 0.057225880350452264, "全帯域_40ms_envelope_correlation": 0.9979401544868656, "全帯域_40ms_modulation_mae_db": 0.5211911579048957, "300–1000 Hz_10ms_log_rms_mae_db": 1.5693743118197456, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14387270025662782, "300–1000 Hz_10ms_envelope_correlation": 0.9879955135605054, "300–1000 Hz_10ms_modulation_mae_db": 0.4393636904116596, "300–1000 Hz_40ms_log_rms_mae_db": 0.9707073438293, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0967374812453203, "300–1000 Hz_40ms_envelope_correlation": 0.9957641527207682, "300–1000 Hz_40ms_modulation_mae_db": 0.8348793522310342, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8111569964336156, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1920117498137957, "1000–3000 Hz_10ms_envelope_correlation": 0.9742998865994834, "1000–3000 Hz_10ms_modulation_mae_db": 0.631646155397605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1522254494124398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11809158514968363, "1000–3000 Hz_40ms_envelope_correlation": 0.9901391307748242, "1000–3000 Hz_40ms_modulation_mae_db": 0.6142828346289099, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8705756688026725, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.191375873310538, "3000–6000 Hz_10ms_envelope_correlation": 0.9751443658832358, "3000–6000 Hz_10ms_modulation_mae_db": 0.48450321662015616, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1971383978250176, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12513678009982154, "3000–6000 Hz_40ms_envelope_correlation": 0.9894886245338512, "3000–6000 Hz_40ms_modulation_mae_db": 0.32896410330944015, "mr_stft_db": 5.592868796312907, "band_mae_db": 0.2608582500036271, "rms_gain_error_db": -0.2639468557271566, "output_peak": 0.7545524317051575}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3113033931485505, "全帯域_10ms_rms_envelope_nrmse": 0.09254724491123285, "全帯域_10ms_envelope_correlation": 0.993124265957122, "全帯域_10ms_modulation_mae_db": 0.4480982873081684, "全帯域_40ms_log_rms_mae_db": 0.6796763068270385, "全帯域_40ms_rms_envelope_nrmse": 0.04566979812779633, "全帯域_40ms_envelope_correlation": 0.9983531871830765, "全帯域_40ms_modulation_mae_db": 0.3549967730037449, "300–1000 Hz_10ms_log_rms_mae_db": 1.7200616587709525, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12659722238099433, "300–1000 Hz_10ms_envelope_correlation": 0.9896295792434677, "300–1000 Hz_10ms_modulation_mae_db": 0.40903648722891967, "300–1000 Hz_40ms_log_rms_mae_db": 1.0452024209140351, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06509780742180243, "300–1000 Hz_40ms_envelope_correlation": 0.9974872582352694, "300–1000 Hz_40ms_modulation_mae_db": 0.4885671488246411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7735168104884425, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21948605148112138, "1000–3000 Hz_10ms_envelope_correlation": 0.9666307250173322, "1000–3000 Hz_10ms_modulation_mae_db": 0.6516274187342117, "1000–3000 Hz_40ms_log_rms_mae_db": 1.124771318351905, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14875324649146043, "1000–3000 Hz_40ms_envelope_correlation": 0.9839067561831745, "1000–3000 Hz_40ms_modulation_mae_db": 0.8913327847601401, "3000–6000 Hz_10ms_log_rms_mae_db": 1.910949530936084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21199898715823765, "3000–6000 Hz_10ms_envelope_correlation": 0.9698918278046708, "3000–6000 Hz_10ms_modulation_mae_db": 0.12128190964411206, "3000–6000 Hz_40ms_log_rms_mae_db": 1.251861281145049, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1162194690091388, "3000–6000 Hz_40ms_envelope_correlation": 0.9904395491246942, "3000–6000 Hz_40ms_modulation_mae_db": 0.2026496851588053, "mr_stft_db": 5.598225261031378, "band_mae_db": 0.12911144491935328, "rms_gain_error_db": -0.1178914456739162, "output_peak": 0.7470587386047202}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3053115237445234, "全帯域_10ms_rms_envelope_nrmse": 0.09796402907517597, "全帯域_10ms_envelope_correlation": 0.9923962963897642, "全帯域_10ms_modulation_mae_db": 0.16724919322594534, "全帯域_40ms_log_rms_mae_db": 0.672540108834736, "全帯域_40ms_rms_envelope_nrmse": 0.06202740342151791, "全帯域_40ms_envelope_correlation": 0.9970033334827296, "全帯域_40ms_modulation_mae_db": 0.32666696153542557, "300–1000 Hz_10ms_log_rms_mae_db": 1.5956039170526926, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14405685086137546, "300–1000 Hz_10ms_envelope_correlation": 0.9871357497333341, "300–1000 Hz_10ms_modulation_mae_db": 0.3530743659737581, "300–1000 Hz_40ms_log_rms_mae_db": 1.019019982336387, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10061303447350872, "300–1000 Hz_40ms_envelope_correlation": 0.9942524780339012, "300–1000 Hz_40ms_modulation_mae_db": 0.41974319728042636, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8149604238414858, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19312452927150903, "1000–3000 Hz_10ms_envelope_correlation": 0.975574441554265, "1000–3000 Hz_10ms_modulation_mae_db": 0.36057944064720554, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1342175458592159, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1283481813601877, "1000–3000 Hz_40ms_envelope_correlation": 0.9906544996148455, "1000–3000 Hz_40ms_modulation_mae_db": 0.23673051429622283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8958919946277883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19954851118500855, "3000–6000 Hz_10ms_envelope_correlation": 0.9725985764513099, "3000–6000 Hz_10ms_modulation_mae_db": 0.8229819474880836, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2599135512951674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12633550541564117, "3000–6000 Hz_40ms_envelope_correlation": 0.9884415642883365, "3000–6000 Hz_40ms_modulation_mae_db": 0.5624894239246699, "mr_stft_db": 5.58672144729417, "band_mae_db": 0.2426698452064041, "rms_gain_error_db": -0.169024028496788, "output_peak": 0.7792171354675879}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 62080, "output_samples": 62080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.993080574367489, "spectral_convergence": 0.31865792299974544, "spectral_convergence_status": "ok", "amplitude_floor": 2.580715109285093e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.362917438531914, "spectral_convergence": 0.28698998840462264, "spectral_convergence_status": "ok", "amplitude_floor": 1.9335078654426385e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.425754944053719, "spectral_convergence": 0.3115596565468216, "spectral_convergence_status": "ok", "amplitude_floor": 1.5577782358148932e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007926248856121452, "output_power": 0.0007774221316979601, "delta_db": -0.08410803528184008, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003328913174136213, "output_power": 0.0030650733200962466, "delta_db": -0.3586160008099951, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00045950159700284816, "output_power": 0.0004268511772392631, "delta_db": -0.32010541593038616, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012582167228939335, "output_power": 0.00012660650549292096, "delta_db": 0.027005687780469988, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.488237987418478e-05, "output_power": 3.4399520240050285e-05, "delta_db": -0.060537216720201244, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.708544414240077e-05, "output_power": 6.781768960172551e-05, "delta_db": 0.04714690856099893, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1495865441806486, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2900344820334043, "pitch": null, "output_rms_dbfs": -23.46964098694453, "output_peak": 0.7975482547727808, "reference_rms_dbfs": -23.179606504911124, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.593917652317707}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 62080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.00032928880743725145, "log_rms_mae_db": 1.223300955276722, "rms_envelope_nrmse": 0.11118841481055637, "envelope_correlation": 0.9905564360050045, "rms_gain_error_db": -0.2900407560308563, "reference_mean_power": 0.004815032927514685, "output_mean_power": 0.004503966854276475}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.902918899290217e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008077922828075312, "output_power": 0.0007705883562937926, "reference_active": true, "delta_db": -0.20477257377449593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023728667544934464, "output_power": 0.00020730449936012034, "reference_active": true, "delta_db": -0.5866462344030864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1018322593991373e-05, "output_power": 1.7632236237483334e-05, "reference_active": true, "delta_db": -0.7629065749548053}], "active_band_count": 3, "active_band_mae_db": 0.5181084610441292}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0003041508818572066, "log_rms_mae_db": 0.6986073010342088, "rms_envelope_nrmse": 0.06274868384550905, "envelope_correlation": 0.9974839639559694, "rms_gain_error_db": -0.29004489217623547, "reference_mean_power": 0.004852582666148238, "output_mean_power": 0.004539086440288577}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.7242336092637014e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007426996521093883, "output_power": 0.0007074569022731023, "reference_active": true, "delta_db": -0.2111323236772848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010890208813537604, "output_power": 9.421662715703218e-05, "reference_active": true, "delta_db": -0.6290865422877612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4730800018618504e-06, "output_power": 1.013198578635952e-06, "reference_active": true, "delta_db": -1.625317618186902}], "active_band_count": 3, "active_band_mae_db": 0.821845494717316}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.0002965570632263201, "log_rms_mae_db": 1.6134142370815578, "rms_envelope_nrmse": 0.1597134128941903, "envelope_correlation": 0.9843740423235934, "rms_gain_error_db": -0.4970473917440993, "reference_mean_power": 0.002818219366063788, "output_mean_power": 0.002513448874460477}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.2520173458640518e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044507178368220307, "output_power": 0.0004154102886121679, "reference_active": true, "delta_db": -0.2995281343263271}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017793142616898364, "output_power": 0.00014556065818740788, "reference_active": true, "delta_db": -0.872086490706091}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6359305994808858e-05, "output_power": 1.3197559614273127e-05, "reference_active": true, "delta_db": -0.9327124338609973}], "active_band_count": 3, "active_band_mae_db": 0.7014423529644719}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0002772405007925501, "log_rms_mae_db": 0.9845650781869665, "rms_envelope_nrmse": 0.10436896366206953, "envelope_correlation": 0.9943326246753735, "rms_gain_error_db": -0.49704744187316274, "reference_mean_power": 0.0028402077631144488, "output_mean_power": 0.002533059352652492}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.1541839932451843e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134314303572302, "output_power": 0.00038537044026906524, "reference_active": true, "delta_db": -0.3052509081711593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.103778868784845e-05, "output_power": 6.614810779645909e-05, "reference_active": true, "delta_db": -0.8817015627294044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3787553243482739e-06, "output_power": 7.186604321967806e-07, "reference_active": true, "delta_db": -2.8296346831729258}], "active_band_count": 3, "active_band_mae_db": 1.3388623846911631}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 6.247605344635176e-05, "log_rms_mae_db": 1.7359823877447942, "rms_envelope_nrmse": 0.2079015079168785, "envelope_correlation": 0.9703832668380801, "rms_gain_error_db": 0.05150430601616586, "reference_mean_power": 0.00014490237542137544, "output_mean_power": 0.00014663104700811494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 6.946266094979904e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.0294766814481675e-05, "output_power": 2.8407437723933703e-05, "reference_active": true, "delta_db": -0.2793555071378973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.606273878647975e-06, "output_power": 8.01009591359292e-06, "reference_active": true, "delta_db": -0.3117744618432312}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.077953085521739e-06, "output_power": 1.9006651920948637e-06, "reference_active": true, "delta_db": 2.4630576134356996}], "active_band_count": 3, "active_band_mae_db": 1.018062527472276}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 5.585752960561645e-05, "log_rms_mae_db": 1.095188018117724, "rms_envelope_nrmse": 0.13320265343816604, "envelope_correlation": 0.9872320319762137, "rms_gain_error_db": 0.051503778929132354, "reference_mean_power": 0.00014603285132237434, "output_mean_power": 0.00014777499144442307}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 6.45401857377675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9246286179300554e-05, "output_power": 2.751169785209598e-05, "reference_active": true, "delta_db": -0.2655333212580855}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6077467608233e-06, "output_power": 3.588314826272479e-06, "reference_active": true, "delta_db": -0.023455059017237524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.605305892606176e-08, "output_power": 1.344415444727619e-07, "reference_active": true, "delta_db": 1.9376718023421087}], "active_band_count": 3, "active_band_mae_db": 0.7422200608724773}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 3.703669227124254e-05, "log_rms_mae_db": 1.8431547871369882, "rms_envelope_nrmse": 0.18503822390567196, "envelope_correlation": 0.9768245406764415, "rms_gain_error_db": 0.042037880267968605, "reference_mean_power": 5.2354216813681856e-05, "output_mean_power": 5.28634442026278e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 3.452458222743894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.744222722260824e-05, "output_power": 1.7219379975818153e-05, "reference_active": true, "delta_db": -0.05584430205459733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.482818619951907e-06, "output_power": 7.316561309791363e-06, "reference_active": true, "delta_db": -0.09758202095692212}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0100258421135412e-06, "output_power": 1.1274339286425573e-06, "reference_active": true, "delta_db": 0.4775861459444502}], "active_band_count": 3, "active_band_mae_db": 0.21033748965198987}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 2.970153096248883e-05, "log_rms_mae_db": 1.2205826641380793, "rms_envelope_nrmse": 0.09909056670155633, "envelope_correlation": 0.9929646265989224, "rms_gain_error_db": 0.042037962832213185, "reference_mean_power": 5.276267962923645e-05, "output_mean_power": 5.3275880976572896e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 3.130411452597279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7491368881094015e-05, "output_power": 1.7554420136799735e-05, "reference_active": true, "delta_db": 0.01562689426702054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.132236890482271e-06, "output_power": 3.9939281041900745e-06, "reference_active": true, "delta_db": -0.1478496779431509}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.718357196867276e-08, "output_power": 7.320246392731111e-08, "reference_active": true, "delta_db": 0.37262608873900027}], "active_band_count": 3, "active_band_mae_db": 0.17870088698305722}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8260858560618694, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a92b6cb694cf43a97a4153583c17f92520e7a1a20aa454afb8eab3ba1032d260", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9748828293344085, "standard_deviation": 0.004486945921631238, "minimum": 0.9666307250173322, "maximum": 0.980549593718012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8038864401381107, "standard_deviation": 0.037797999923936565, "minimum": 1.7359823877447942, "maximum": 1.8681678109931865, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6237458473994266, "standard_deviation": 0.29873147776024234, "minimum": 0.22386367064217302, "maximum": 1.018062527472276, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1904644413278121, "standard_deviation": 0.016829454900108926, "minimum": 0.1676488275836672, "maximum": 0.21948605148112138, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9903156196454943, "standard_deviation": 0.0032516706955190086, "minimum": 0.9839067561831745, "maximum": 0.9932408754945785, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1262013981925931, "standard_deviation": 0.024250624200262742, "minimum": 1.0851711446448702, "maximum": 1.1522254494124398, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5447525252083005, "standard_deviation": 0.24772014797908135, "minimum": 0.17583539184856925, "maximum": 0.8913327847601401, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11697288175982178, "standard_deviation": 0.018450589228227798, "minimum": 0.09833165303998093, "maximum": 0.14875324649146043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9735637861515873, "standard_deviation": 0.002599437727993413, "minimum": 0.9698918278046708, "maximum": 0.9768245406764415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8694671068098625, "standard_deviation": 0.026427983054674517, "minimum": 1.8404161023779186, "maximum": 1.910949530936084, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.48239036770135435, "standard_deviation": 0.2505532571649197, "minimum": 0.12128190964411206, "maximum": 0.8229819474880836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19785721218428226, "standard_deviation": 0.011825275113939165, "minimum": 0.18503822390567196, "maximum": 0.21750027874272337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9902439282532176, "standard_deviation": 0.0013636639694838448, "minimum": 0.9884415642883365, "maximum": 0.9929646265989224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2269210866946079, "standard_deviation": 0.029149054350114712, "minimum": 1.1897419206440305, "maximum": 1.2599135512951674, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4828970639896444, "standard_deviation": 0.21991837716075727, "minimum": 0.17870088698305722, "maximum": 0.7921535305940317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11775390226657809, "standard_deviation": 0.009413393549962033, "minimum": 0.09909056670155633, "maximum": 0.12650509921415018, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.98556262226585, "standard_deviation": 0.0058570550012755195, "minimum": 0.972489951917708, "maximum": 0.9901829478480697, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6460923455731895, "standard_deviation": 0.052575298975442296, "minimum": 1.5693743118197456, "maximum": 1.7200616587709525, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5735747389480885, "standard_deviation": 0.2783771874147235, "minimum": 0.3530743659737581, "maximum": 1.210895372561844, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14946577843655945, "standard_deviation": 0.02683675780248163, "minimum": 0.12341834210324298, "maximum": 0.2063507387665112, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.994919772261688, "standard_deviation": 0.0029775088950794175, "minimum": 0.988210625985943, "maximum": 0.9974872582352694, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0334396787027338, "standard_deviation": 0.04497291185642569, "minimum": 0.9707073438293, "maximum": 1.1052159292407366, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6674770006739212, "standard_deviation": 0.31247351402887913, "minimum": 0.36664103624559813, "maximum": 1.3388623846911631, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09145569890074158, "standard_deviation": 0.02192578692082652, "minimum": 0.06509780742180243, "maximum": 0.13257427955681786, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24568963229474727, "standard_deviation": 0.0908358759562979, "minimum": 0.12911144491935328, "maximum": 0.3756313494834897, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.592318885263141, "standard_deviation": 0.010118474533352604, "minimum": 5.5763911632452015, "maximum": 5.611705411965741, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8091386606316686, "standard_deviation": 0.0701218840890271, "minimum": 0.7470587386047202, "maximum": 0.9501961688381586, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13584451122235913, "standard_deviation": 0.12703549848027698, "minimum": -0.2900344820334043, "maximum": 0.06523725962188126, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9908146646069651, "standard_deviation": 0.003632930562930071, "minimum": 0.9821752823694183, "maximum": 0.993124265957122, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3024533516715313, "standard_deviation": 0.04797695939042855, "minimum": 1.223300955276722, "maximum": 1.3762342139739814, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38601644757883147, "standard_deviation": 0.21170559643929737, "minimum": 0.16724919322594534, "maximum": 0.8208449997650328, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10664124943748893, "standard_deviation": 0.01876049423013544, "minimum": 0.09254724491123285, "maximum": 0.15017454130528404, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971979994189578, "standard_deviation": 0.0016597756912144195, "minimum": 0.9932107649465907, "maximum": 0.9983531871830765, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6908857300741881, "standard_deviation": 0.04345450300163525, "minimum": 0.6158379993105094, "maximum": 0.7626339532022886, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42722061331319294, "standard_deviation": 0.2228401768526643, "minimum": 0.13033176865388552, "maximum": 0.821845494717316, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05939417679936399, "standard_deviation": 0.013781262836168238, "minimum": 0.04566979812779633, "maximum": 0.09046746006349099, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0032", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.2086575031280518, "predicted_mos": 2.2086575031280518, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "149573e75a5d0b5540eaed4e78a2d7dc55bd9d09b2faf790186fe74c688e3fd1", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.850784699898213, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2009425789612789, "全帯域_10ms_rms_envelope_nrmse": 0.11028940402411004, "全帯域_10ms_envelope_correlation": 0.9902201978755037, "全帯域_10ms_modulation_mae_db": 0.5329178342694311, "全帯域_40ms_log_rms_mae_db": 0.5958220962575268, "全帯域_40ms_rms_envelope_nrmse": 0.06002763196206556, "全帯域_40ms_envelope_correlation": 0.9969394556492129, "全帯域_40ms_modulation_mae_db": 0.2576463850849107, "300–1000 Hz_10ms_log_rms_mae_db": 2.0272067814386325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13394025949422217, "300–1000 Hz_10ms_envelope_correlation": 0.9881868788697529, "300–1000 Hz_10ms_modulation_mae_db": 0.8115767484084467, "300–1000 Hz_40ms_log_rms_mae_db": 1.3024318082151805, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07260348043035553, "300–1000 Hz_40ms_envelope_correlation": 0.9965249402524317, "300–1000 Hz_40ms_modulation_mae_db": 0.32119421830883815, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2394615303248404, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23599032459558733, "1000–3000 Hz_10ms_envelope_correlation": 0.9611372167956372, "1000–3000 Hz_10ms_modulation_mae_db": 1.3489352570802255, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4826263911361364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15461467122955228, "1000–3000 Hz_40ms_envelope_correlation": 0.9826368173202202, "1000–3000 Hz_40ms_modulation_mae_db": 1.116665334101003, "3000–6000 Hz_10ms_log_rms_mae_db": 2.476412354352291, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20223854862658772, "3000–6000 Hz_10ms_envelope_correlation": 0.972361237131909, "3000–6000 Hz_10ms_modulation_mae_db": 1.1022822489936177, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6444654211421135, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13929483685309307, "3000–6000 Hz_40ms_envelope_correlation": 0.9869822470697209, "3000–6000 Hz_40ms_modulation_mae_db": 1.215466673549049, "mr_stft_db": 5.987062092454626, "band_mae_db": 0.3492433333269694, "rms_gain_error_db": 0.033132790708365646, "output_peak": 0.7872102090423433}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1875899578623992, "全帯域_10ms_rms_envelope_nrmse": 0.08530237637893484, "全帯域_10ms_envelope_correlation": 0.9943390106415217, "全帯域_10ms_modulation_mae_db": 0.9150391232384815, "全帯域_40ms_log_rms_mae_db": 0.6736023502496409, "全帯域_40ms_rms_envelope_nrmse": 0.050503002047982706, "全帯域_40ms_envelope_correlation": 0.9981377211639869, "全帯域_40ms_modulation_mae_db": 0.6546990101788807, "300–1000 Hz_10ms_log_rms_mae_db": 2.0047236974835454, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12582644728097866, "300–1000 Hz_10ms_envelope_correlation": 0.9902692019983207, "300–1000 Hz_10ms_modulation_mae_db": 1.0050532897618125, "300–1000 Hz_40ms_log_rms_mae_db": 1.3630474672699395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07973346712095933, "300–1000 Hz_40ms_envelope_correlation": 0.9966238035632963, "300–1000 Hz_40ms_modulation_mae_db": 0.8070312301198931, "1000–3000 Hz_10ms_log_rms_mae_db": 2.148554566096926, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18696463248198422, "1000–3000 Hz_10ms_envelope_correlation": 0.9758471994346409, "1000–3000 Hz_10ms_modulation_mae_db": 0.9476526964366457, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4063713366779018, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0990012293760191, "1000–3000 Hz_40ms_envelope_correlation": 0.9929256042361614, "1000–3000 Hz_40ms_modulation_mae_db": 0.6346562717324128, "3000–6000 Hz_10ms_log_rms_mae_db": 2.538170281505252, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1906163028342259, "3000–6000 Hz_10ms_envelope_correlation": 0.9751849901887439, "3000–6000 Hz_10ms_modulation_mae_db": 0.2299064258119826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.662523913675362, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11274909239777325, "3000–6000 Hz_40ms_envelope_correlation": 0.9910557786124369, "3000–6000 Hz_40ms_modulation_mae_db": 0.2372517638572751, "mr_stft_db": 5.996008150211032, "band_mae_db": 0.30797839210033917, "rms_gain_error_db": -0.18008688345318852, "output_peak": 0.7959607926172407}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1866144812903403, "全帯域_10ms_rms_envelope_nrmse": 0.09737143818218497, "全帯域_10ms_envelope_correlation": 0.9924290274496301, "全帯域_10ms_modulation_mae_db": 0.2029400589801835, "全帯域_40ms_log_rms_mae_db": 0.5986009489962364, "全帯域_40ms_rms_envelope_nrmse": 0.05300136392762838, "全帯域_40ms_envelope_correlation": 0.9976651107265897, "全帯域_40ms_modulation_mae_db": 0.39639045773651543, "300–1000 Hz_10ms_log_rms_mae_db": 1.8889461473553744, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1121935136114414, "300–1000 Hz_10ms_envelope_correlation": 0.9917359704381167, "300–1000 Hz_10ms_modulation_mae_db": 0.5849084088659026, "300–1000 Hz_40ms_log_rms_mae_db": 1.2057516356783557, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06672774743608864, "300–1000 Hz_40ms_envelope_correlation": 0.9970506370112453, "300–1000 Hz_40ms_modulation_mae_db": 1.0988473128953242, "1000–3000 Hz_10ms_log_rms_mae_db": 2.265227632186249, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2811744470105949, "1000–3000 Hz_10ms_envelope_correlation": 0.9517023346709502, "1000–3000 Hz_10ms_modulation_mae_db": 1.5009901059217006, "1000–3000 Hz_40ms_log_rms_mae_db": 1.479455230733301, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17839674921704482, "1000–3000 Hz_40ms_envelope_correlation": 0.9812293843041107, "1000–3000 Hz_40ms_modulation_mae_db": 1.2130491404396935, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5140759428339026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2211261981598361, "3000–6000 Hz_10ms_envelope_correlation": 0.9662855529879899, "3000–6000 Hz_10ms_modulation_mae_db": 0.6080503333045099, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6790203868008386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1355467666819511, "3000–6000 Hz_40ms_envelope_correlation": 0.9874127579708228, "3000–6000 Hz_40ms_modulation_mae_db": 0.6475142237474153, "mr_stft_db": 5.988978803302302, "band_mae_db": 0.32394386896607735, "rms_gain_error_db": 0.01744066365832572, "output_peak": 0.7873170215315549}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3066157023920653, "全帯域_10ms_rms_envelope_nrmse": 0.12313845282963132, "全帯域_10ms_envelope_correlation": 0.988120143999979, "全帯域_10ms_modulation_mae_db": 0.3472258202762298, "全帯域_40ms_log_rms_mae_db": 0.6709945783279203, "全帯域_40ms_rms_envelope_nrmse": 0.05619078433620537, "全帯域_40ms_envelope_correlation": 0.9980664822525002, "全帯域_40ms_modulation_mae_db": 0.23172998167317096, "300–1000 Hz_10ms_log_rms_mae_db": 1.893130220292096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1729899858611219, "300–1000 Hz_10ms_envelope_correlation": 0.980928725365704, "300–1000 Hz_10ms_modulation_mae_db": 0.6707241970824652, "300–1000 Hz_40ms_log_rms_mae_db": 1.2100700175576202, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07796341151777608, "300–1000 Hz_40ms_envelope_correlation": 0.9973672203931588, "300–1000 Hz_40ms_modulation_mae_db": 0.4975363145521223, "1000–3000 Hz_10ms_log_rms_mae_db": 2.190336602013105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23877347836452692, "1000–3000 Hz_10ms_envelope_correlation": 0.9619683883033646, "1000–3000 Hz_10ms_modulation_mae_db": 1.214488482168881, "1000–3000 Hz_40ms_log_rms_mae_db": 1.410870946151311, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14412151776864, "1000–3000 Hz_40ms_envelope_correlation": 0.9857133527604569, "1000–3000 Hz_40ms_modulation_mae_db": 1.2030034456339245, "3000–6000 Hz_10ms_log_rms_mae_db": 2.5905554834196947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22734404382128573, "3000–6000 Hz_10ms_envelope_correlation": 0.9645925335991706, "3000–6000 Hz_10ms_modulation_mae_db": 1.26132905811776, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7543933954005857, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13944258599276413, "3000–6000 Hz_40ms_envelope_correlation": 0.9870971761897602, "3000–6000 Hz_40ms_modulation_mae_db": 1.3525343438005348, "mr_stft_db": 5.988251632446333, "band_mae_db": 0.36260409395181914, "rms_gain_error_db": -0.27134844707345945, "output_peak": 0.775552833826749}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2575707970800873, "全帯域_10ms_rms_envelope_nrmse": 0.11206722930716183, "全帯域_10ms_envelope_correlation": 0.9898185562661874, "全帯域_10ms_modulation_mae_db": 0.24265520175719005, "全帯域_40ms_log_rms_mae_db": 0.7061378434064581, "全帯域_40ms_rms_envelope_nrmse": 0.055210855662254296, "全帯域_40ms_envelope_correlation": 0.9974517093464416, "全帯域_40ms_modulation_mae_db": 0.1618496596518846, "300–1000 Hz_10ms_log_rms_mae_db": 1.886641107374947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14102491331266329, "300–1000 Hz_10ms_envelope_correlation": 0.9869677301068239, "300–1000 Hz_10ms_modulation_mae_db": 0.3532009466457886, "300–1000 Hz_40ms_log_rms_mae_db": 1.2483459672201054, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0751594426732991, "300–1000 Hz_40ms_envelope_correlation": 0.9964080898970773, "300–1000 Hz_40ms_modulation_mae_db": 0.489311421388386, "1000–3000 Hz_10ms_log_rms_mae_db": 2.260255904245978, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24818558486925107, "1000–3000 Hz_10ms_envelope_correlation": 0.9587786674019171, "1000–3000 Hz_10ms_modulation_mae_db": 1.3672627405274802, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4952706767638104, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15639146081377717, "1000–3000 Hz_40ms_envelope_correlation": 0.983066182686076, "1000–3000 Hz_40ms_modulation_mae_db": 1.153912598557167, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6244983304873513, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20257649174574294, "3000–6000 Hz_10ms_envelope_correlation": 0.9726916063836862, "3000–6000 Hz_10ms_modulation_mae_db": 0.7501622336074956, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8137425493503054, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13617099910086297, "3000–6000 Hz_40ms_envelope_correlation": 0.9870792457142153, "3000–6000 Hz_40ms_modulation_mae_db": 0.5851963271032479, "mr_stft_db": 6.001678796914727, "band_mae_db": 0.32112728705627885, "rms_gain_error_db": -0.08883087292834922, "output_peak": 0.7894346589229868}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.208990093073728, "全帯域_10ms_rms_envelope_nrmse": 0.09661753297096547, "全帯域_10ms_envelope_correlation": 0.9924785097378259, "全帯域_10ms_modulation_mae_db": 0.08189317746997826, "全帯域_40ms_log_rms_mae_db": 0.6608955332012607, "全帯域_40ms_rms_envelope_nrmse": 0.045999063190696174, "全帯域_40ms_envelope_correlation": 0.9982590823016366, "全帯域_40ms_modulation_mae_db": 0.07671119568879682, "300–1000 Hz_10ms_log_rms_mae_db": 2.0001410026054898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14330104709137098, "300–1000 Hz_10ms_envelope_correlation": 0.9865159428587343, "300–1000 Hz_10ms_modulation_mae_db": 0.4281530919527567, "300–1000 Hz_40ms_log_rms_mae_db": 1.280459540427274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06787671908700026, "300–1000 Hz_40ms_envelope_correlation": 0.9971009408329796, "300–1000 Hz_40ms_modulation_mae_db": 0.25375597917002823, "1000–3000 Hz_10ms_log_rms_mae_db": 2.236641239676158, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22141548047299175, "1000–3000 Hz_10ms_envelope_correlation": 0.9660736831061162, "1000–3000 Hz_10ms_modulation_mae_db": 1.0897866187105618, "1000–3000 Hz_40ms_log_rms_mae_db": 1.5131988221740662, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1287171509085016, "1000–3000 Hz_40ms_envelope_correlation": 0.9879815849880031, "1000–3000 Hz_40ms_modulation_mae_db": 0.6470480786148332, "3000–6000 Hz_10ms_log_rms_mae_db": 2.6476912059171975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21794474463544464, "3000–6000 Hz_10ms_envelope_correlation": 0.9672660027892322, "3000–6000 Hz_10ms_modulation_mae_db": 0.3921819819479703, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7627330511774426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12368140433169722, "3000–6000 Hz_40ms_envelope_correlation": 0.9888583957660818, "3000–6000 Hz_40ms_modulation_mae_db": 0.3769654246275363, "mr_stft_db": 5.981431703223279, "band_mae_db": 0.1973738885060539, "rms_gain_error_db": -0.08576893775897139, "output_peak": 0.8216092716539378}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.212118749908393, "全帯域_10ms_rms_envelope_nrmse": 0.11212362400082243, "全帯域_10ms_envelope_correlation": 0.9898486545829275, "全帯域_10ms_modulation_mae_db": 0.44285539582667033, "全帯域_40ms_log_rms_mae_db": 0.6776231758327949, "全帯域_40ms_rms_envelope_nrmse": 0.04475148609686191, "全帯域_40ms_envelope_correlation": 0.9983000897451086, "全帯域_40ms_modulation_mae_db": 0.18523116873308867, "300–1000 Hz_10ms_log_rms_mae_db": 1.9534000043032709, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15836703512257927, "300–1000 Hz_10ms_envelope_correlation": 0.9834623864550806, "300–1000 Hz_10ms_modulation_mae_db": 0.5615702881303158, "300–1000 Hz_40ms_log_rms_mae_db": 1.2438316248619699, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07070774441960631, "300–1000 Hz_40ms_envelope_correlation": 0.9967854967557864, "300–1000 Hz_40ms_modulation_mae_db": 0.6433541749423456, "1000–3000 Hz_10ms_log_rms_mae_db": 2.213720036326071, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22014933584552238, "1000–3000 Hz_10ms_envelope_correlation": 0.9685887207414735, "1000–3000 Hz_10ms_modulation_mae_db": 0.9247665225902931, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3969433205733224, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12588894118816338, "1000–3000 Hz_40ms_envelope_correlation": 0.9898147064285682, "1000–3000 Hz_40ms_modulation_mae_db": 0.6039128812296917, "3000–6000 Hz_10ms_log_rms_mae_db": 2.553674463797037, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2056346091062057, "3000–6000 Hz_10ms_envelope_correlation": 0.971313515974894, "3000–6000 Hz_10ms_modulation_mae_db": 0.4688819489068799, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6981867654237834, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11738771776262026, "3000–6000 Hz_40ms_envelope_correlation": 0.9901825256757327, "3000–6000 Hz_40ms_modulation_mae_db": 0.4355190263516307, "mr_stft_db": 5.991447325780268, "band_mae_db": 0.18909715046363473, "rms_gain_error_db": -0.03308018537461062, "output_peak": 0.780853296497799}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2506638969975072, "全帯域_10ms_rms_envelope_nrmse": 0.08413688025453082, "全帯域_10ms_envelope_correlation": 0.9944124808516387, "全帯域_10ms_modulation_mae_db": 0.22858915398553778, "全帯域_40ms_log_rms_mae_db": 0.6445719410664383, "全帯域_40ms_rms_envelope_nrmse": 0.047918963429462134, "全帯域_40ms_envelope_correlation": 0.998137895818164, "全帯域_40ms_modulation_mae_db": 0.17037624433418985, "300–1000 Hz_10ms_log_rms_mae_db": 1.9095681560925901, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10202036137174311, "300–1000 Hz_10ms_envelope_correlation": 0.9931541236766382, "300–1000 Hz_10ms_modulation_mae_db": 0.4203914808346731, "300–1000 Hz_40ms_log_rms_mae_db": 1.281727601397543, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05234081542589933, "300–1000 Hz_40ms_envelope_correlation": 0.9981283602702197, "300–1000 Hz_40ms_modulation_mae_db": 0.5575841573109044, "1000–3000 Hz_10ms_log_rms_mae_db": 2.258659927943122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26449540285954987, "1000–3000 Hz_10ms_envelope_correlation": 0.9561788387101904, "1000–3000 Hz_10ms_modulation_mae_db": 1.025878742430416, "1000–3000 Hz_40ms_log_rms_mae_db": 1.549856216980302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18428832186817992, "1000–3000 Hz_40ms_envelope_correlation": 0.9787357150629145, "1000–3000 Hz_40ms_modulation_mae_db": 0.5573098142329336, "3000–6000 Hz_10ms_log_rms_mae_db": 2.55151984974134, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19135805972205006, "3000–6000 Hz_10ms_envelope_correlation": 0.9751918124600105, "3000–6000 Hz_10ms_modulation_mae_db": 0.7787151962303805, "3000–6000 Hz_40ms_log_rms_mae_db": 1.717646923533043, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11181657125547988, "3000–6000 Hz_40ms_envelope_correlation": 0.99187530940156, "3000–6000 Hz_40ms_modulation_mae_db": 0.5909668297806895, "mr_stft_db": 5.973861757884767, "band_mae_db": 0.4073190859300579, "rms_gain_error_db": 0.07967370848254604, "output_peak": 0.8005341856498914}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 62080, "output_samples": 62080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.439509588953443, "spectral_convergence": 0.3431776468564291, "spectral_convergence_status": "ok", "amplitude_floor": 2.580715109285093e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.828257974543036, "spectral_convergence": 0.3118845565385315, "spectral_convergence_status": "ok", "amplitude_floor": 1.9335078654426385e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.693418713867399, "spectral_convergence": 0.34299296165399734, "spectral_convergence_status": "ok", "amplitude_floor": 1.5577782358148932e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007926248856121452, "output_power": 0.000875580824464731, "delta_db": 0.43228538274832495, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003328913174136213, "output_power": 0.0032125239508652846, "delta_db": -0.15456093666169185, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00045950159700284816, "output_power": 0.0005325154904993925, "delta_db": 0.6404522049417661, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012582167228939335, "output_power": 0.00012119675864025262, "delta_db": -0.16264448122931444, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.488237987418478e-05, "output_power": 3.940968309127841e-05, "delta_db": 0.529968352384415, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.708544414240077e-05, "output_power": 6.442781978020682e-05, "delta_db": -0.17554864199630416, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3492433333269694, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.033132790708365646, "pitch": null, "output_rms_dbfs": -23.146473714202756, "output_peak": 0.7872102090423433, "reference_rms_dbfs": -23.179606504911124, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.987062092454626}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 62080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.00032928880743725145, "log_rms_mae_db": 1.2009425789612789, "rms_envelope_nrmse": 0.11028940402411004, "envelope_correlation": 0.9902201978755037, "rms_gain_error_db": 0.03313135009058112, "reference_mean_power": 0.004815032927514685, "output_mean_power": 0.00485190620183076}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.902918899290217e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008077922828075312, "output_power": 0.0008001861435657444, "reference_active": true, "delta_db": -0.04108673207151073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023728667544934464, "output_power": 0.00022819821735157165, "reference_active": true, "delta_db": -0.16961104157084658}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1018322593991373e-05, "output_power": 2.893367682648007e-05, "reference_active": true, "delta_db": 1.3880557291659357}], "active_band_count": 3, "active_band_mae_db": 0.5329178342694311}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0003041508818572066, "log_rms_mae_db": 0.5958220962575268, "rms_envelope_nrmse": 0.06002763196206556, "envelope_correlation": 0.9969394556492129, "rms_gain_error_db": 0.033133872101495675, "reference_mean_power": 0.004852582666148238, "output_mean_power": 0.004889746333975625}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.7242336092637014e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007426996521093883, "output_power": 0.0007402950587544501, "reference_active": true, "delta_db": -0.014083698786070645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010890208813537604, "output_power": 0.0001074552361446057, "reference_active": true, "delta_db": -0.0580862427361522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4730800018618504e-06, "output_power": 1.7310246609062703e-06, "reference_active": true, "delta_db": 0.7007692137325094}], "active_band_count": 3, "active_band_mae_db": 0.2576463850849107}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.0002965570632263201, "log_rms_mae_db": 2.0272067814386325, "rms_envelope_nrmse": 0.13394025949422217, "envelope_correlation": 0.9881868788697529, "rms_gain_error_db": -0.1511249233324132, "reference_mean_power": 0.002818219366063788, "output_mean_power": 0.0027218381853021164}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.2520173458640518e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044507178368220307, "output_power": 0.0004121224249894096, "reference_active": true, "delta_db": -0.33403815474181775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017793142616898364, "output_power": 0.00016082660685082996, "reference_active": true, "delta_db": -0.4389476053864396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6359305994808858e-05, "output_power": 2.3984977992346456e-05, "reference_active": true, "delta_db": 1.6617444850970826}], "active_band_count": 3, "active_band_mae_db": 0.8115767484084467}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0002772405007925501, "log_rms_mae_db": 1.3024318082151805, "rms_envelope_nrmse": 0.07260348043035553, "envelope_correlation": 0.9965249402524317, "rms_gain_error_db": -0.15112589074153845, "reference_mean_power": 0.0028402077631144488, "output_mean_power": 0.0027430739831348083}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.1541839932451843e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134314303572302, "output_power": 0.0003875806674645282, "reference_active": true, "delta_db": -0.2804138368132476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.103778868784845e-05, "output_power": 7.835829554224528e-05, "reference_active": true, "delta_db": -0.14602601131186663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3787553243482739e-06, "output_power": 1.5602761975077108e-06, "reference_active": true, "delta_db": 0.5371428068014001}], "active_band_count": 3, "active_band_mae_db": 0.32119421830883815}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 6.247605344635176e-05, "log_rms_mae_db": 2.2394615303248404, "rms_envelope_nrmse": 0.23599032459558733, "envelope_correlation": 0.9611372167956372, "rms_gain_error_db": -0.0670674356071358, "reference_mean_power": 0.00014490237542137544, "output_mean_power": 0.00014268185977149873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 6.946266094979904e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.0294766814481675e-05, "output_power": 2.91151916531319e-05, "reference_active": true, "delta_db": -0.17247960752340644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.606273878647975e-06, "output_power": 7.464924223942152e-06, "reference_active": true, "delta_db": -0.6178975899731218}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.077953085521739e-06, "output_power": 2.2816168597357775e-06, "reference_active": true, "delta_db": 3.2564285737441483}], "active_band_count": 3, "active_band_mae_db": 1.3489352570802255}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 5.585752960561645e-05, "log_rms_mae_db": 1.4826263911361364, "rms_envelope_nrmse": 0.15461467122955228, "envelope_correlation": 0.9826368173202202, "rms_gain_error_db": -0.06707286121718908, "reference_mean_power": 0.00014603285132237434, "output_mean_power": 0.00014379483237090442}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 6.45401857377675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9246286179300554e-05, "output_power": 2.8258936215728236e-05, "reference_active": true, "delta_db": -0.14914916152216048}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6077467608233e-06, "output_power": 3.2467793614471906e-06, "reference_active": true, "delta_db": -0.45783268776469405}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.605305892606176e-08, "output_power": 1.618332396420943e-07, "reference_active": true, "delta_db": 2.743014153016155}], "active_band_count": 3, "active_band_mae_db": 1.116665334101003}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 3.703669227124254e-05, "log_rms_mae_db": 2.476412354352291, "rms_envelope_nrmse": 0.20223854862658772, "envelope_correlation": 0.972361237131909, "rms_gain_error_db": -0.34837844226320036, "reference_mean_power": 5.2354216813681856e-05, "output_mean_power": 4.831854221063611e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 3.452458222743894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.744222722260824e-05, "output_power": 1.545194084210961e-05, "reference_active": true, "delta_db": -0.5261890303696932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.482818619951907e-06, "output_power": 5.244132346255582e-06, "reference_active": true, "delta_db": -1.543915748006347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0100258421135412e-06, "output_power": 7.597282728914062e-07, "reference_active": true, "delta_db": -1.236741968604813}], "active_band_count": 3, "active_band_mae_db": 1.1022822489936177}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 2.970153096248883e-05, "log_rms_mae_db": 1.6444654211421135, "rms_envelope_nrmse": 0.13929483685309307, "envelope_correlation": 0.9869822470697209, "rms_gain_error_db": -0.3483816763275497, "reference_mean_power": 5.276267962923645e-05, "output_mean_power": 4.869548279959039e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 3.130411452597279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7491368881094015e-05, "output_power": 1.5037911495136938e-05, "reference_active": true, "delta_db": -0.6563627437881858}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.132236890482271e-06, "output_power": 2.859853474332706e-06, "reference_active": true, "delta_db": -1.598414280855912}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.718357196867276e-08, "output_power": 4.876416485290543e-08, "reference_active": true, "delta_db": -1.3916229960030484}], "active_band_count": 3, "active_band_mae_db": 1.215466673549049}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8129567185312502, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a79ec3558124e9f4acd4aabda009c536d39fe54901a7dc813d4c3b34159694e5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9625343811455362, "standard_deviation": 0.007564170896185712, "minimum": 0.9517023346709502, "maximum": 0.9758471994346409, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.2266071798515563, "standard_deviation": 0.04051616103252452, "minimum": 2.148554566096926, "maximum": 2.265227632186249, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1774701457332755, "standard_deviation": 0.21340791622015187, "minimum": 0.9247665225902931, "maximum": 1.5009901059217006, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23714358581250106, "standard_deviation": 0.028950342028112638, "minimum": 0.18696463248198422, "maximum": 0.2811744470105949, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9852629184733138, "standard_deviation": 0.004745304491132585, "minimum": 0.9787357150629145, "maximum": 0.9929256042361614, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.4668241176487689, "standard_deviation": 0.05597452004190947, "minimum": 1.3969433205733224, "maximum": 1.549856216980302, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8911946955677073, "standard_deviation": 0.3023957137829765, "minimum": 0.5573098142329336, "maximum": 1.2130491404396935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14642750529623477, "standard_deviation": 0.02831416247906507, "minimum": 0.0990012293760191, "maximum": 0.18428832186817992, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9706109064394546, "standard_deviation": 0.004069247852909821, "minimum": 0.9645925335991706, "maximum": 0.9751918124600105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.562074739006758, "standard_deviation": 0.0566106488721187, "minimum": 2.476412354352291, "maximum": 2.6476912059171975, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6989386783650746, "standard_deviation": 0.3511684242616256, "minimum": 0.2299064258119826, "maximum": 1.26132905811776, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20735487483142234, "standard_deviation": 0.013572074821112244, "minimum": 0.1906163028342259, "maximum": 0.22734404382128573, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9888179295500414, "standard_deviation": 0.001983775297864057, "minimum": 0.9869822470697209, "maximum": 0.99187530940156, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7165890508129342, "standard_deviation": 0.05716300532672239, "minimum": 1.6444654211421135, "maximum": 1.8137425493503054, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6801768266021723, "standard_deviation": 0.39724472851652765, "minimum": 0.2372517638572751, "maximum": 1.3525343438005348, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12701124679703024, "standard_deviation": 0.011953291929599172, "minimum": 0.11181657125547988, "maximum": 0.13944258599276413, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9876526199711464, "standard_deviation": 0.004116207559352785, "minimum": 0.980928725365704, "maximum": 0.9931541236766382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9454696396182434, "standard_deviation": 0.05846596697423207, "minimum": 1.886641107374947, "maximum": 2.0272067814386325, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6044473064602702, "standard_deviation": 0.2197666045619907, "minimum": 0.3532009466457886, "maximum": 1.0050532897618125, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1362079453932651, "standard_deviation": 0.023225173435217877, "minimum": 0.10202036137174311, "maximum": 0.1729899858611219, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969986861220244, "standard_deviation": 0.0005586158659066086, "minimum": 0.9964080898970773, "maximum": 0.9981283602702197, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2669582078284984, "standard_deviation": 0.051762794249451564, "minimum": 1.2057516356783557, "maximum": 1.3630474672699395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5835768510859802, "standard_deviation": 0.27087157093810277, "minimum": 0.25375597917002823, "maximum": 1.0988473128953242, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07038910351387306, "standard_deviation": 0.008601550628924454, "minimum": 0.05234081542589933, "maximum": 0.07973346712095933, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.30733588753765384, "standard_deviation": 0.07685415626775999, "minimum": 0.18909715046363473, "maximum": 0.4073190859300579, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.988590032777166, "standard_deviation": 0.008495985257414193, "minimum": 5.973861757884767, "maximum": 6.001678796914727, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7923090337178129, "standard_deviation": 0.014195959131900974, "minimum": 0.775552833826749, "maximum": 0.8216092716539378, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06610852046741772, "standard_deviation": 0.11661289148300812, "minimum": -0.27134844707345945, "maximum": 0.07967370848254604, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9914583226756517, "standard_deviation": 0.00229797172130152, "minimum": 0.988120143999979, "maximum": 0.9944124808516387, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2263882821957248, "standard_deviation": 0.041814486927264714, "minimum": 1.1866144812903403, "maximum": 1.3066157023920653, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37426447072546276, "standard_deviation": 0.2608829083144218, "minimum": 0.08189317746997826, "maximum": 0.9150391232384815, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1026308672435427, "standard_deviation": 0.01396166969936101, "minimum": 0.08413688025453082, "maximum": 0.12313845282963132, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978696933754551, "standard_deviation": 0.0004782810014930056, "minimum": 0.9969394556492129, "maximum": 0.9983000897451086, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6535310584172845, "standard_deviation": 0.038799627647446364, "minimum": 0.5958220962575268, "maximum": 0.7061378434064581, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2668292628851797, "standard_deviation": 0.18192320168907974, "minimum": 0.07671119568879682, "maximum": 0.6546990101788807, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05170039383164457, "standard_deviation": 0.005345863821325109, "minimum": 0.04475148609686191, "maximum": 0.06002763196206556, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0032", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.3214943408966064, "predicted_mos": 2.3214943408966064, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "9b1b61b963fb9e06a8acb6a5f5a80b44a8a26cbeba00c0d1f2e3480fb1577363", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.830317399930209, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2349347080119533, "全帯域_10ms_rms_envelope_nrmse": 0.10457842512145162, "全帯域_10ms_envelope_correlation": 0.9913155705351114, "全帯域_10ms_modulation_mae_db": 0.11584851422624516, "全帯域_40ms_log_rms_mae_db": 0.5825612962490729, "全帯域_40ms_rms_envelope_nrmse": 0.052208937332277335, "全帯域_40ms_envelope_correlation": 0.997767130535382, "全帯域_40ms_modulation_mae_db": 0.26819915441753245, "300–1000 Hz_10ms_log_rms_mae_db": 1.5690837766666037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16385251495249767, "300–1000 Hz_10ms_envelope_correlation": 0.9826517623345054, "300–1000 Hz_10ms_modulation_mae_db": 0.33478220785764673, "300–1000 Hz_40ms_log_rms_mae_db": 1.0258366917058022, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.076475033523308, "300–1000 Hz_40ms_envelope_correlation": 0.9961391856407444, "300–1000 Hz_40ms_modulation_mae_db": 0.5192202443310454, "1000–3000 Hz_10ms_log_rms_mae_db": 1.697371632386017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23498236843786466, "1000–3000 Hz_10ms_envelope_correlation": 0.9653537380230673, "1000–3000 Hz_10ms_modulation_mae_db": 0.9290667018993884, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1723191789177234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15640644683861332, "1000–3000 Hz_40ms_envelope_correlation": 0.9847997295901373, "1000–3000 Hz_40ms_modulation_mae_db": 0.5762310574321492, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6923462688338125, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.180442047077151, "3000–6000 Hz_10ms_envelope_correlation": 0.9775590795089477, "3000–6000 Hz_10ms_modulation_mae_db": 0.6207133503466064, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1020777578458032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10009788241992859, "3000–6000 Hz_40ms_envelope_correlation": 0.9928043030925603, "3000–6000 Hz_40ms_modulation_mae_db": 0.5038448698549145, "mr_stft_db": 5.525413904696836, "band_mae_db": 0.25568137826285353, "rms_gain_error_db": 0.07443491549508573, "output_peak": 0.8662517903454611}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2829136786760251, "全帯域_10ms_rms_envelope_nrmse": 0.11262145773429827, "全帯域_10ms_envelope_correlation": 0.9910489209978247, "全帯域_10ms_modulation_mae_db": 0.19532214908065834, "全帯域_40ms_log_rms_mae_db": 0.6708111992564337, "全帯域_40ms_rms_envelope_nrmse": 0.07591925978237862, "全帯域_40ms_envelope_correlation": 0.996203021158069, "全帯域_40ms_modulation_mae_db": 0.18772242135736716, "300–1000 Hz_10ms_log_rms_mae_db": 1.5380573920200205, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14758080553072803, "300–1000 Hz_10ms_envelope_correlation": 0.9871482935510661, "300–1000 Hz_10ms_modulation_mae_db": 0.18706467529236193, "300–1000 Hz_40ms_log_rms_mae_db": 1.001539605427593, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09303848026719623, "300–1000 Hz_40ms_envelope_correlation": 0.9952603596319362, "300–1000 Hz_40ms_modulation_mae_db": 0.29160131698840014, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6839682058292387, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20853485998421323, "1000–3000 Hz_10ms_envelope_correlation": 0.9710666830204835, "1000–3000 Hz_10ms_modulation_mae_db": 0.8683267025745373, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1031928929662524, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12692767532932273, "1000–3000 Hz_40ms_envelope_correlation": 0.9889609608877852, "1000–3000 Hz_40ms_modulation_mae_db": 0.82783273145634, "3000–6000 Hz_10ms_log_rms_mae_db": 1.723950729299372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18810727348028736, "3000–6000 Hz_10ms_envelope_correlation": 0.9758795036247925, "3000–6000 Hz_10ms_modulation_mae_db": 0.3116358473604235, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1679711145386507, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12082561683061958, "3000–6000 Hz_40ms_envelope_correlation": 0.9894058347940342, "3000–6000 Hz_40ms_modulation_mae_db": 0.25550277110356323, "mr_stft_db": 5.533323203906771, "band_mae_db": 0.2524980868940205, "rms_gain_error_db": 0.2696254358484603, "output_peak": 0.776452777280326}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2836662704180013, "全帯域_10ms_rms_envelope_nrmse": 0.09274910578665557, "全帯域_10ms_envelope_correlation": 0.9930754210715068, "全帯域_10ms_modulation_mae_db": 0.2626361571494592, "全帯域_40ms_log_rms_mae_db": 0.6329253326186106, "全帯域_40ms_rms_envelope_nrmse": 0.04489610606248845, "全帯域_40ms_envelope_correlation": 0.9983722200408686, "全帯域_40ms_modulation_mae_db": 0.24046131456521833, "300–1000 Hz_10ms_log_rms_mae_db": 1.5418649657398271, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11692829047865064, "300–1000 Hz_10ms_envelope_correlation": 0.9913694323025346, "300–1000 Hz_10ms_modulation_mae_db": 0.4499838144959331, "300–1000 Hz_40ms_log_rms_mae_db": 1.0059871144408783, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06871058146377436, "300–1000 Hz_40ms_envelope_correlation": 0.9973266962773902, "300–1000 Hz_40ms_modulation_mae_db": 0.5842739784806649, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6937968029339368, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15889803759071108, "1000–3000 Hz_10ms_envelope_correlation": 0.9825562723139529, "1000–3000 Hz_10ms_modulation_mae_db": 0.7221250969597824, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0983962971288586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07997653953180153, "1000–3000 Hz_40ms_envelope_correlation": 0.9953574122034666, "1000–3000 Hz_40ms_modulation_mae_db": 0.6956445319662375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7252390945863585, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20581479154504623, "3000–6000 Hz_10ms_envelope_correlation": 0.9710367317922838, "3000–6000 Hz_10ms_modulation_mae_db": 0.5372139768757402, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1234486153760617, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10793203022337673, "3000–6000 Hz_40ms_envelope_correlation": 0.991503596893249, "3000–6000 Hz_40ms_modulation_mae_db": 0.6594042498871006, "mr_stft_db": 5.505604018806754, "band_mae_db": 0.18665183778713434, "rms_gain_error_db": -0.10253644164254143, "output_peak": 0.7869918573930884}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3268753425322886, "全帯域_10ms_rms_envelope_nrmse": 0.08235224000903012, "全帯域_10ms_envelope_correlation": 0.9945154874727113, "全帯域_10ms_modulation_mae_db": 0.21234545966324547, "全帯域_40ms_log_rms_mae_db": 0.7094401134395211, "全帯域_40ms_rms_envelope_nrmse": 0.03535988266787028, "全帯域_40ms_envelope_correlation": 0.9989682043317887, "全帯域_40ms_modulation_mae_db": 0.1192328770666496, "300–1000 Hz_10ms_log_rms_mae_db": 1.5305773615674327, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09756934345775646, "300–1000 Hz_10ms_envelope_correlation": 0.9939938786761269, "300–1000 Hz_10ms_modulation_mae_db": 0.4080026534198548, "300–1000 Hz_40ms_log_rms_mae_db": 1.0778913197362487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051680989765481675, "300–1000 Hz_40ms_envelope_correlation": 0.9985448904583182, "300–1000 Hz_40ms_modulation_mae_db": 0.8034440146454461, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6713151907369597, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20487445561284523, "1000–3000 Hz_10ms_envelope_correlation": 0.9719599200634325, "1000–3000 Hz_10ms_modulation_mae_db": 0.7275918348749123, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0692093685747188, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10750245072377479, "1000–3000 Hz_40ms_envelope_correlation": 0.9920805636800818, "1000–3000 Hz_40ms_modulation_mae_db": 0.7178800491505885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7040166189526154, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24070549257381696, "3000–6000 Hz_10ms_envelope_correlation": 0.962995994962492, "3000–6000 Hz_10ms_modulation_mae_db": 0.598025452958885, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1315759294126175, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14988875896294682, "3000–6000 Hz_40ms_envelope_correlation": 0.9853064629054971, "3000–6000 Hz_40ms_modulation_mae_db": 0.6935597864157169, "mr_stft_db": 5.5319416707799, "band_mae_db": 0.328653264499492, "rms_gain_error_db": -0.030351009352802615, "output_peak": 0.7827320836701088}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3029595206360873, "全帯域_10ms_rms_envelope_nrmse": 0.09832878543902236, "全帯域_10ms_envelope_correlation": 0.9924817072214341, "全帯域_10ms_modulation_mae_db": 0.09824093544724533, "全帯域_40ms_log_rms_mae_db": 0.74404782084662, "全帯域_40ms_rms_envelope_nrmse": 0.060734884556971694, "全帯域_40ms_envelope_correlation": 0.997092582127258, "全帯域_40ms_modulation_mae_db": 0.10770277108583659, "300–1000 Hz_10ms_log_rms_mae_db": 1.629206623906366, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12365469898536809, "300–1000 Hz_10ms_envelope_correlation": 0.9901235065033291, "300–1000 Hz_10ms_modulation_mae_db": 0.16555895888050443, "300–1000 Hz_40ms_log_rms_mae_db": 1.1301741132435872, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0753862098102071, "300–1000 Hz_40ms_envelope_correlation": 0.9962373435787756, "300–1000 Hz_40ms_modulation_mae_db": 0.3619660251825056, "1000–3000 Hz_10ms_log_rms_mae_db": 1.669155895928101, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20674065591401342, "1000–3000 Hz_10ms_envelope_correlation": 0.9701219731314408, "1000–3000 Hz_10ms_modulation_mae_db": 0.9174070548014038, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0926351316181027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11273728141608784, "1000–3000 Hz_40ms_envelope_correlation": 0.9911381127903052, "1000–3000 Hz_40ms_modulation_mae_db": 0.6717584876479031, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6847484961966404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.183950743127904, "3000–6000 Hz_10ms_envelope_correlation": 0.977450650591253, "3000–6000 Hz_10ms_modulation_mae_db": 0.7882985935109567, "3000–6000 Hz_40ms_log_rms_mae_db": 1.090980604677557, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09051470921298378, "3000–6000 Hz_40ms_envelope_correlation": 0.9942904640546872, "3000–6000 Hz_40ms_modulation_mae_db": 0.5983790989319275, "mr_stft_db": 5.517707230171879, "band_mae_db": 0.27530787158829984, "rms_gain_error_db": 0.0932796723718319, "output_peak": 0.7658804424876917}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.300726004239908, "全帯域_10ms_rms_envelope_nrmse": 0.11735741865301276, "全帯域_10ms_envelope_correlation": 0.9888548813089282, "全帯域_10ms_modulation_mae_db": 0.4419503185444639, "全帯域_40ms_log_rms_mae_db": 0.676697861692854, "全帯域_40ms_rms_envelope_nrmse": 0.060068102531163396, "全帯域_40ms_envelope_correlation": 0.9969366087603326, "全帯域_40ms_modulation_mae_db": 0.16181358950781963, "300–1000 Hz_10ms_log_rms_mae_db": 1.5551043549380472, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16361646345062383, "300–1000 Hz_10ms_envelope_correlation": 0.9823079782529868, "300–1000 Hz_10ms_modulation_mae_db": 0.45882827925605724, "300–1000 Hz_40ms_log_rms_mae_db": 0.9462841145934117, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07768877518863064, "300–1000 Hz_40ms_envelope_correlation": 0.995989286156191, "300–1000 Hz_40ms_modulation_mae_db": 0.652876702614642, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6807700094885087, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20973894327411752, "1000–3000 Hz_10ms_envelope_correlation": 0.9693311839045573, "1000–3000 Hz_10ms_modulation_mae_db": 0.8933192629906564, "1000–3000 Hz_40ms_log_rms_mae_db": 1.076996044854466, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1157601443324195, "1000–3000 Hz_40ms_envelope_correlation": 0.9902617016791881, "1000–3000 Hz_40ms_modulation_mae_db": 0.71635099551022, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6832081818468267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18977506191733345, "3000–6000 Hz_10ms_envelope_correlation": 0.9751318009151533, "3000–6000 Hz_10ms_modulation_mae_db": 0.5060326077841233, "3000–6000 Hz_40ms_log_rms_mae_db": 1.075094470399956, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11404361872794119, "3000–6000 Hz_40ms_envelope_correlation": 0.9905142331422351, "3000–6000 Hz_40ms_modulation_mae_db": 0.6106562916663525, "mr_stft_db": 5.529871236013516, "band_mae_db": 0.21602706358150314, "rms_gain_error_db": -0.06572197087968483, "output_peak": 0.7498117267642659}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.26472922915789, "全帯域_10ms_rms_envelope_nrmse": 0.08400724144319408, "全帯域_10ms_envelope_correlation": 0.9947219132574574, "全帯域_10ms_modulation_mae_db": 0.4574910528878591, "全帯域_40ms_log_rms_mae_db": 0.6610921103954537, "全帯域_40ms_rms_envelope_nrmse": 0.05304464366950185, "全帯域_40ms_envelope_correlation": 0.9981501064973526, "全帯域_40ms_modulation_mae_db": 0.4069898693032034, "300–1000 Hz_10ms_log_rms_mae_db": 1.6356240157740118, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1447188057160174, "300–1000 Hz_10ms_envelope_correlation": 0.987867139302573, "300–1000 Hz_10ms_modulation_mae_db": 0.626548755764336, "300–1000 Hz_40ms_log_rms_mae_db": 1.059886298864274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10373806141526337, "300–1000 Hz_40ms_envelope_correlation": 0.9947864188110265, "300–1000 Hz_40ms_modulation_mae_db": 0.9262179633758398, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7000300157466248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19382251705267786, "1000–3000 Hz_10ms_envelope_correlation": 0.9737679400275959, "1000–3000 Hz_10ms_modulation_mae_db": 0.8932291246615293, "1000–3000 Hz_40ms_log_rms_mae_db": 1.151881335409073, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13026378366190672, "1000–3000 Hz_40ms_envelope_correlation": 0.9876836356475565, "1000–3000 Hz_40ms_modulation_mae_db": 0.749800595362407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.719392166091148, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21561315021157895, "3000–6000 Hz_10ms_envelope_correlation": 0.9680890993643437, "3000–6000 Hz_10ms_modulation_mae_db": 0.8993654230958423, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1804220607683966, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11800496340967791, "3000–6000 Hz_40ms_envelope_correlation": 0.9909207891014328, "3000–6000 Hz_40ms_modulation_mae_db": 0.812506087245382, "mr_stft_db": 5.523728975787833, "band_mae_db": 0.2664042999504285, "rms_gain_error_db": -0.23165691619752196, "output_peak": 0.7516025234519611}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3108505841491576, "全帯域_10ms_rms_envelope_nrmse": 0.10896081389449286, "全帯域_10ms_envelope_correlation": 0.991487648426645, "全帯域_10ms_modulation_mae_db": 0.26791036105672433, "全帯域_40ms_log_rms_mae_db": 0.6461806080410801, "全帯域_40ms_rms_envelope_nrmse": 0.08142507515414522, "全帯域_40ms_envelope_correlation": 0.9957114498999601, "全帯域_40ms_modulation_mae_db": 0.35626233807194496, "300–1000 Hz_10ms_log_rms_mae_db": 1.6626816125891943, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15116318670189371, "300–1000 Hz_10ms_envelope_correlation": 0.9866371066303841, "300–1000 Hz_10ms_modulation_mae_db": 0.45289526949351505, "300–1000 Hz_40ms_log_rms_mae_db": 1.0779847365093491, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10946335216665085, "300–1000 Hz_40ms_envelope_correlation": 0.994015635441933, "300–1000 Hz_40ms_modulation_mae_db": 0.7151119619240222, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7898989827829797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23022536669362287, "1000–3000 Hz_10ms_envelope_correlation": 0.9631231305699799, "1000–3000 Hz_10ms_modulation_mae_db": 1.59050026721276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1582360104169207, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1517749857610702, "1000–3000 Hz_40ms_envelope_correlation": 0.984211958656049, "1000–3000 Hz_40ms_modulation_mae_db": 1.4032029679200768, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6696515628625614, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20961726547961618, "3000–6000 Hz_10ms_envelope_correlation": 0.9698812480585122, "3000–6000 Hz_10ms_modulation_mae_db": 0.5191169635355913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0427923600941886, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10964857273741305, "3000–6000 Hz_40ms_envelope_correlation": 0.9912056304734502, "3000–6000 Hz_40ms_modulation_mae_db": 0.6954361329171949, "mr_stft_db": 5.513141135341672, "band_mae_db": 0.2476099668781457, "rms_gain_error_db": -0.373195818177657, "output_peak": 0.7877194757216742}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 62080, "output_samples": 62080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.859235463338484, "spectral_convergence": 0.3267332503880347, "spectral_convergence_status": "ok", "amplitude_floor": 2.580715109285093e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.260441789876145, "spectral_convergence": 0.32399190595195687, "spectral_convergence_status": "ok", "amplitude_floor": 1.9335078654426385e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.456564460875876, "spectral_convergence": 0.3457145715420391, "spectral_convergence_status": "ok", "amplitude_floor": 1.5577782358148932e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007926248856121452, "output_power": 0.0007983245172418668, "delta_db": 0.031117635182888615, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003328913174136213, "output_power": 0.0033082937505201178, "delta_db": -0.026984033515603405, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00045950159700284816, "output_power": 0.0005493066747295144, "delta_db": 0.7752785112798667, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012582167228939335, "output_power": 0.00013779528790895304, "delta_db": 0.39478913481668054, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.488237987418478e-05, "output_power": 3.389656946139235e-05, "delta_db": -0.12450360119876254, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.708544414240077e-05, "output_power": 6.434084555677366e-05, "delta_db": -0.1814153535833193, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25568137826285353, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.07443491549508573, "pitch": null, "output_rms_dbfs": -23.105171589416038, "output_peak": 0.8662517903454611, "reference_rms_dbfs": -23.179606504911124, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.525413904696836}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 62080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.00032928880743725145, "log_rms_mae_db": 1.2349347080119533, "rms_envelope_nrmse": 0.10457842512145162, "envelope_correlation": 0.9913155705351114, "rms_gain_error_db": 0.07443224421884115, "reference_mean_power": 0.004815032927514685, "output_mean_power": 0.004898267353962774}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.902918899290217e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008077922828075312, "output_power": 0.0007984209077109071, "reference_active": true, "delta_db": -0.05067798843428831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023728667544934464, "output_power": 0.00022825881216660384, "reference_active": true, "delta_db": -0.1684579869675341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1018322593991373e-05, "output_power": 2.0405963085464153e-05, "reference_active": true, "delta_db": -0.1284095672769131}], "active_band_count": 3, "active_band_mae_db": 0.11584851422624516}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0003041508818572066, "log_rms_mae_db": 0.5825612962490729, "rms_envelope_nrmse": 0.052208937332277335, "envelope_correlation": 0.997767130535382, "rms_gain_error_db": 0.07443575883719493, "reference_mean_power": 0.004852582666148238, "output_mean_power": 0.004936470186099716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.7242336092637014e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007426996521093883, "output_power": 0.0007349313340841671, "reference_active": true, "delta_db": -0.04566456397144501}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010890208813537604, "output_power": 0.00010877114861846104, "reference_active": true, "delta_db": -0.005224924971611656}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4730800018618504e-06, "output_power": 1.2383845816875391e-06, "reference_active": true, "delta_db": -0.7537079743095407}], "active_band_count": 3, "active_band_mae_db": 0.26819915441753245}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.0002965570632263201, "log_rms_mae_db": 1.5690837766666037, "rms_envelope_nrmse": 0.16385251495249767, "envelope_correlation": 0.9826517623345054, "rms_gain_error_db": 0.09763223410103003, "reference_mean_power": 0.002818219366063788, "output_mean_power": 0.0028822922788615574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.2520173458640518e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044507178368220307, "output_power": 0.00042060628144648473, "reference_active": true, "delta_db": -0.24554307820059829}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017793142616898364, "output_power": 0.00016068580960111928, "reference_active": true, "delta_db": -0.44275134480031075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6359305994808858e-05, "output_power": 1.759422314882019e-05, "reference_active": true, "delta_db": 0.31605220057203115}], "active_band_count": 3, "active_band_mae_db": 0.33478220785764673}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0002772405007925501, "log_rms_mae_db": 1.0258366917058022, "rms_envelope_nrmse": 0.076475033523308, "envelope_correlation": 0.9961391856407444, "rms_gain_error_db": 0.09763200636654884, "reference_mean_power": 0.0028402077631144488, "output_mean_power": 0.002904780435225391}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.1541839932451843e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134314303572302, "output_power": 0.00038968577231485873, "reference_active": true, "delta_db": -0.25688940274814365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.103778868784845e-05, "output_power": 7.811192125108457e-05, "reference_active": true, "delta_db": -0.15970261836898775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3787553243482739e-06, "output_power": 1.0601817680278906e-06, "reference_active": true, "delta_db": -1.1410687118760048}], "active_band_count": 3, "active_band_mae_db": 0.5192202443310454}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 6.247605344635176e-05, "log_rms_mae_db": 1.697371632386017, "rms_envelope_nrmse": 0.23498236843786466, "envelope_correlation": 0.9653537380230673, "rms_gain_error_db": 0.37510153522831186, "reference_mean_power": 0.00014490237542137544, "output_mean_power": 0.00015797401588083543}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 6.946266094979904e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.0294766814481675e-05, "output_power": 2.8874505707002948e-05, "reference_active": true, "delta_db": -0.2085305556277928}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.606273878647975e-06, "output_power": 7.582962077425093e-06, "reference_active": true, "delta_db": -0.5497627850512461}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.077953085521739e-06, "output_power": 1.719849864331888e-06, "reference_active": true, "delta_db": 2.0289067650191264}], "active_band_count": 3, "active_band_mae_db": 0.9290667018993884}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 5.585752960561645e-05, "log_rms_mae_db": 1.1723191789177234, "rms_envelope_nrmse": 0.15640644683861332, "envelope_correlation": 0.9847997295901373, "rms_gain_error_db": 0.3751008460828832, "reference_mean_power": 0.00014603285132237434, "output_mean_power": 0.0001592064467276703}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 6.45401857377675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9246286179300554e-05, "output_power": 2.7765501069443017e-05, "reference_active": true, "delta_db": -0.22565209951875959}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6077467608233e-06, "output_power": 3.4869372325619365e-06, "reference_active": true, "delta_db": -0.14791915608123204}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.605305892606176e-08, "output_power": 1.1756512351429436e-07, "reference_active": true, "delta_db": 1.3551219166964559}], "active_band_count": 3, "active_band_mae_db": 0.5762310574321492}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 3.703669227124254e-05, "log_rms_mae_db": 1.6923462688338125, "rms_envelope_nrmse": 0.180442047077151, "envelope_correlation": 0.9775590795089477, "rms_gain_error_db": -0.16262193680524703, "reference_mean_power": 5.2354216813681856e-05, "output_mean_power": 5.043005868241771e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 3.452458222743894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.744222722260824e-05, "output_power": 1.5946868812357838e-05, "reference_active": true, "delta_db": -0.3892651831362165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.482818619951907e-06, "output_power": 6.20276422507991e-06, "reference_active": true, "delta_db": -0.8147994485838992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0100258421135412e-06, "output_power": 8.680104299849736e-07, "reference_active": true, "delta_db": -0.6580754193197037}], "active_band_count": 3, "active_band_mae_db": 0.6207133503466064}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 2.970153096248883e-05, "log_rms_mae_db": 1.1020777578458032, "rms_envelope_nrmse": 0.10009788241992859, "envelope_correlation": 0.9928043030925603, "rms_gain_error_db": -0.1626226326124787, "reference_mean_power": 5.276267962923645e-05, "output_mean_power": 5.082350124934283e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 3.130411452597279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7491368881094015e-05, "output_power": 1.5894479282203274e-05, "reference_active": true, "delta_db": -0.41577494261691383}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.132236890482271e-06, "output_power": 3.3558002005645695e-06, "reference_active": true, "delta_db": -0.9038911487551026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.718357196867276e-08, "output_power": 6.42800548334974e-08, "reference_active": true, "delta_db": -0.19186851819272704}], "active_band_count": 3, "active_band_mae_db": 0.5038448698549145}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.834465892533612, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "503694ddf2a2fab5788188e7bd43b53ca1e45218825b623af9c717d2721df168", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9709101051318137, "standard_deviation": 0.005849925407551978, "minimum": 0.9631231305699799, "maximum": 0.9825562723139529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6982883419790458, "standard_deviation": 0.038738621903597954, "minimum": 1.669155895928101, "maximum": 1.7898989827829797, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9426957557468713, "standard_deviation": 0.27407319370919175, "minimum": 0.7221250969597824, "maximum": 1.59050026721276, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20597715057000823, "standard_deviation": 0.02334813691549364, "minimum": 0.15889803759071108, "maximum": 0.23498236843786466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893117593918213, "standard_deviation": 0.0037339060203963924, "minimum": 0.984211958656049, "maximum": 0.9953574122034666, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1153582824857646, "standard_deviation": 0.039587890348515446, "minimum": 1.0692093685747188, "maximum": 1.1723191789177234, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7948376770557403, "standard_deviation": 0.25581371463985997, "minimum": 0.5762310574321492, "maximum": 1.4032029679200768, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12266866344937458, "standard_deviation": 0.02466991371309202, "minimum": 0.07997653953180153, "maximum": 0.15640644683861332, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9722530136022223, "standard_deviation": 0.005165044614181368, "minimum": 0.962995994962492, "maximum": 0.9775590795089477, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.700319139833667, "standard_deviation": 0.021031284940580205, "minimum": 1.6696515628625614, "maximum": 1.7252390945863585, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.597550276933521, "standard_deviation": 0.18044740232947692, "minimum": 0.3116358473604235, "maximum": 0.8993654230958423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20175322817659175, "standard_deviation": 0.02031360972513794, "minimum": 0.180442047077151, "maximum": 0.24070549257381696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9907439143071433, "standard_deviation": 0.002647205782053784, "minimum": 0.9853064629054971, "maximum": 0.9942904640546872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1142953641391538, "standard_deviation": 0.04625776182300809, "minimum": 1.0427923600941886, "maximum": 1.1804220607683966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.603661161002769, "standard_deviation": 0.16668288624344713, "minimum": 0.25550277110356323, "maximum": 0.812506087245382, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11386951906561096, "standard_deviation": 0.017529894589870285, "minimum": 0.09051470921298378, "maximum": 0.14988875896294682, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9877623871941883, "standard_deviation": 0.0040521695882469455, "minimum": 0.9823079782529868, "maximum": 0.9939938786761269, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5827750129001878, "standard_deviation": 0.05166510831638756, "minimum": 1.5305773615674327, "maximum": 1.6626816125891943, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.38545807680752614, "standard_deviation": 0.1526017755593615, "minimum": 0.16555895888050443, "maximum": 0.626548755764336, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13863551365919197, "standard_deviation": 0.023652154746309568, "minimum": 0.09756934345775646, "maximum": 0.16385251495249767, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960374769995394, "standard_deviation": 0.001427221252862712, "minimum": 0.994015635441933, "maximum": 0.9985448904583182, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.040697999315143, "standard_deviation": 0.057347157222574496, "minimum": 0.9462841145934117, "maximum": 1.1301741132435872, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6068390259428207, "standard_deviation": 0.2144836841960668, "minimum": 0.29160131698840014, "maximum": 0.9262179633758398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08202268545006403, "standard_deviation": 0.01903878791589315, "minimum": 0.051680989765481675, "maximum": 0.10946335216665085, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25360422118023473, "standard_deviation": 0.04175332831071103, "minimum": 0.18665183778713434, "maximum": 0.328653264499492, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.522591421938145, "standard_deviation": 0.009758436199327387, "minimum": 5.505604018806754, "maximum": 5.533323203906771, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7834303346393221, "standard_deviation": 0.036627210097952535, "minimum": 0.7498117267642659, "maximum": 0.8662517903454611, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04576526656685373, "standard_deviation": 0.19959894832414843, "minimum": -0.373195818177657, "maximum": 0.2696254358484603, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9921876937864524, "standard_deviation": 0.0019409313063344865, "minimum": 0.9888548813089282, "maximum": 0.9947219132574574, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.288456917227664, "standard_deviation": 0.02883355082879257, "minimum": 1.2349347080119533, "maximum": 1.3268753425322886, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2564681185069876, "standard_deviation": 0.13388597998762186, "minimum": 0.09824093544724533, "maximum": 0.4574910528878591, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1001194360101447, "standard_deviation": 0.013007170572704757, "minimum": 0.08235224000903012, "maximum": 0.11735741865301276, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974001654188764, "standard_deviation": 0.0011145221897705994, "minimum": 0.9957114498999601, "maximum": 0.9989682043317887, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6654695428174557, "standard_deviation": 0.0486777936802886, "minimum": 0.5825612962490729, "maximum": 0.74404782084662, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.23104804192194656, "standard_deviation": 0.1085555599495696, "minimum": 0.10770277108583659, "maximum": 0.4069898693032034, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05795711146959961, "standard_deviation": 0.015231334505169752, "minimum": 0.03535988266787028, "maximum": 0.08142507515414522, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0032", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.3582923412323, "predicted_mos": 2.3582923412323, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "210848545d26868a786d0b8eaac9090569d0795ba9668cd1bf56c483be6f3172", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7572018997743726, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1970582921828103, "全帯域_10ms_rms_envelope_nrmse": 0.11359586213054478, "全帯域_10ms_envelope_correlation": 0.9900763132951886, "全帯域_10ms_modulation_mae_db": 0.4348697218508524, "全帯域_40ms_log_rms_mae_db": 0.6827991868699677, "全帯域_40ms_rms_envelope_nrmse": 0.06316382173793356, "全帯域_40ms_envelope_correlation": 0.9974056313464128, "全帯域_40ms_modulation_mae_db": 0.7322764249655913, "300–1000 Hz_10ms_log_rms_mae_db": 1.6510030410115255, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16894195415681976, "300–1000 Hz_10ms_envelope_correlation": 0.9823174214256012, "300–1000 Hz_10ms_modulation_mae_db": 0.5544405077900226, "300–1000 Hz_40ms_log_rms_mae_db": 1.047900720494063, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10836244739758913, "300–1000 Hz_40ms_envelope_correlation": 0.9937792377075878, "300–1000 Hz_40ms_modulation_mae_db": 1.3051412107098164, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8443859200640074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21449257660347432, "1000–3000 Hz_10ms_envelope_correlation": 0.9694414293132896, "1000–3000 Hz_10ms_modulation_mae_db": 1.1376061724374529, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2694907015285477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.143962476559189, "1000–3000 Hz_40ms_envelope_correlation": 0.9857732170101977, "1000–3000 Hz_40ms_modulation_mae_db": 0.8788124672845324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9590992817076973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18851215382984665, "3000–6000 Hz_10ms_envelope_correlation": 0.9760659198297221, "3000–6000 Hz_10ms_modulation_mae_db": 0.3633558103372699, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3066596654951876, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09970858621348136, "3000–6000 Hz_40ms_envelope_correlation": 0.992925991928235, "3000–6000 Hz_40ms_modulation_mae_db": 0.3268408710716297, "mr_stft_db": 5.68707396427359, "band_mae_db": 0.18897963808254215, "rms_gain_error_db": -0.2849062468053716, "output_peak": 0.8101435138165535}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2903314128917105, "全帯域_10ms_rms_envelope_nrmse": 0.09860022892801218, "全帯域_10ms_envelope_correlation": 0.9921419780688274, "全帯域_10ms_modulation_mae_db": 0.1546762017421809, "全帯域_40ms_log_rms_mae_db": 0.6423940375815846, "全帯域_40ms_rms_envelope_nrmse": 0.042573142444083535, "全帯域_40ms_envelope_correlation": 0.9984571267955309, "全帯域_40ms_modulation_mae_db": 0.3104171553850843, "300–1000 Hz_10ms_log_rms_mae_db": 1.6452749518803986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1426493528568061, "300–1000 Hz_10ms_envelope_correlation": 0.9865387135741398, "300–1000 Hz_10ms_modulation_mae_db": 0.3728796766187242, "300–1000 Hz_40ms_log_rms_mae_db": 1.014352496926271, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06701231813847815, "300–1000 Hz_40ms_envelope_correlation": 0.9969567790988507, "300–1000 Hz_40ms_modulation_mae_db": 0.5551257839108664, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8907245659095273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19606626764068055, "1000–3000 Hz_10ms_envelope_correlation": 0.9745293935001539, "1000–3000 Hz_10ms_modulation_mae_db": 0.4776498810605328, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3049369803705289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13024080111358796, "1000–3000 Hz_40ms_envelope_correlation": 0.9884627996998478, "1000–3000 Hz_40ms_modulation_mae_db": 0.38449673163747394, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9748207963906128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19846994128884776, "3000–6000 Hz_10ms_envelope_correlation": 0.9731024891816639, "3000–6000 Hz_10ms_modulation_mae_db": 0.42534802091231366, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3024284169690064, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10928045469134971, "3000–6000 Hz_40ms_envelope_correlation": 0.9912994459877528, "3000–6000 Hz_40ms_modulation_mae_db": 0.5489268059096962, "mr_stft_db": 5.679839700019227, "band_mae_db": 0.39993393093116136, "rms_gain_error_db": -0.023607196943365445, "output_peak": 0.7813732928357136}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3383272182688837, "全帯域_10ms_rms_envelope_nrmse": 0.09870820614420596, "全帯域_10ms_envelope_correlation": 0.9922473438735896, "全帯域_10ms_modulation_mae_db": 0.35532394199404843, "全帯域_40ms_log_rms_mae_db": 0.7165885674230426, "全帯域_40ms_rms_envelope_nrmse": 0.05232268318969757, "全帯域_40ms_envelope_correlation": 0.9979187733650824, "全帯域_40ms_modulation_mae_db": 0.1792946804392589, "300–1000 Hz_10ms_log_rms_mae_db": 1.6709432194400573, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12788339373000654, "300–1000 Hz_10ms_envelope_correlation": 0.9894589945083369, "300–1000 Hz_10ms_modulation_mae_db": 0.41984920599765313, "300–1000 Hz_40ms_log_rms_mae_db": 1.0829711343173511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07118725547455904, "300–1000 Hz_40ms_envelope_correlation": 0.9969635271848989, "300–1000 Hz_40ms_modulation_mae_db": 0.508814479308073, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9178597352823747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21092083163662828, "1000–3000 Hz_10ms_envelope_correlation": 0.9690986414144082, "1000–3000 Hz_10ms_modulation_mae_db": 1.253209672712809, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2901146653702054, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12423432248594404, "1000–3000 Hz_40ms_envelope_correlation": 0.988803011071243, "1000–3000 Hz_40ms_modulation_mae_db": 0.7789579335360804, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9851489720221291, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18549295030906462, "3000–6000 Hz_10ms_envelope_correlation": 0.9762561459380595, "3000–6000 Hz_10ms_modulation_mae_db": 0.7046904463787732, "3000–6000 Hz_40ms_log_rms_mae_db": 1.399926723031312, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11741929814346422, "3000–6000 Hz_40ms_envelope_correlation": 0.990072609900591, "3000–6000 Hz_40ms_modulation_mae_db": 0.4850047310114, "mr_stft_db": 5.6972403631250925, "band_mae_db": 0.14454990396400574, "rms_gain_error_db": -0.1616741203926517, "output_peak": 0.8773471408607579}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.267687618249766, "全帯域_10ms_rms_envelope_nrmse": 0.15157780541399182, "全帯域_10ms_envelope_correlation": 0.9818096588892166, "全帯域_10ms_modulation_mae_db": 0.7744681357419262, "全帯域_40ms_log_rms_mae_db": 0.6885022829325523, "全帯域_40ms_rms_envelope_nrmse": 0.09234541978406019, "全帯域_40ms_envelope_correlation": 0.992896858793525, "全帯域_40ms_modulation_mae_db": 0.6596450152830924, "300–1000 Hz_10ms_log_rms_mae_db": 1.7327563733115119, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2080537572790258, "300–1000 Hz_10ms_envelope_correlation": 0.9719189529609776, "300–1000 Hz_10ms_modulation_mae_db": 1.048049398929054, "300–1000 Hz_40ms_log_rms_mae_db": 1.1217708983120882, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13660017074194394, "300–1000 Hz_40ms_envelope_correlation": 0.9874060987106231, "300–1000 Hz_40ms_modulation_mae_db": 0.6203444800428201, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8708500400734893, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19079709114092308, "1000–3000 Hz_10ms_envelope_correlation": 0.975679262675127, "1000–3000 Hz_10ms_modulation_mae_db": 0.7052206106914579, "1000–3000 Hz_40ms_log_rms_mae_db": 1.261739015143224, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11518742233754296, "1000–3000 Hz_40ms_envelope_correlation": 0.990884859222423, "1000–3000 Hz_40ms_modulation_mae_db": 0.6472311505587681, "3000–6000 Hz_10ms_log_rms_mae_db": 1.901562950870028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1909449090753523, "3000–6000 Hz_10ms_envelope_correlation": 0.9749861302870746, "3000–6000 Hz_10ms_modulation_mae_db": 0.42890879932745524, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2993600031331125, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1170021358264799, "3000–6000 Hz_40ms_envelope_correlation": 0.9899821759449239, "3000–6000 Hz_40ms_modulation_mae_db": 0.7586878886442969, "mr_stft_db": 5.684750922951463, "band_mae_db": 0.3281852277848935, "rms_gain_error_db": 0.05518869147121393, "output_peak": 0.9579679947805501}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3372776760819445, "全帯域_10ms_rms_envelope_nrmse": 0.09831427160511436, "全帯域_10ms_envelope_correlation": 0.9923380942312806, "全帯域_10ms_modulation_mae_db": 0.2830158929950388, "全帯域_40ms_log_rms_mae_db": 0.7232196630638424, "全帯域_40ms_rms_envelope_nrmse": 0.052064173983167525, "全帯域_40ms_envelope_correlation": 0.9979800214103874, "全帯域_40ms_modulation_mae_db": 0.1979929727611025, "300–1000 Hz_10ms_log_rms_mae_db": 1.715015788629165, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13249059828665946, "300–1000 Hz_10ms_envelope_correlation": 0.9893665826710467, "300–1000 Hz_10ms_modulation_mae_db": 0.49866993231956735, "300–1000 Hz_40ms_log_rms_mae_db": 1.1572445078617226, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08229438406676146, "300–1000 Hz_40ms_envelope_correlation": 0.9966634018638764, "300–1000 Hz_40ms_modulation_mae_db": 0.5123047631107127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9142672958807352, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19351383161459856, "1000–3000 Hz_10ms_envelope_correlation": 0.9740157236484377, "1000–3000 Hz_10ms_modulation_mae_db": 1.062583537429014, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2610983616184066, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11097328370943302, "1000–3000 Hz_40ms_envelope_correlation": 0.9911386172882851, "1000–3000 Hz_40ms_modulation_mae_db": 0.7166854847102965, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9873661230116593, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21146465819762847, "3000–6000 Hz_10ms_envelope_correlation": 0.971395613036208, "3000–6000 Hz_10ms_modulation_mae_db": 0.7105693290210354, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3251584654057205, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12442071280461774, "3000–6000 Hz_40ms_envelope_correlation": 0.9900421231998412, "3000–6000 Hz_40ms_modulation_mae_db": 0.7475130823418036, "mr_stft_db": 5.666707997075229, "band_mae_db": 0.37298446779264, "rms_gain_error_db": -0.17098831493223277, "output_peak": 0.7840312260073582}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2259952723066823, "全帯域_10ms_rms_envelope_nrmse": 0.09876287860578635, "全帯域_10ms_envelope_correlation": 0.9927228811316442, "全帯域_10ms_modulation_mae_db": 0.31200384746232057, "全帯域_40ms_log_rms_mae_db": 0.6154521554216398, "全帯域_40ms_rms_envelope_nrmse": 0.05757693844704299, "全帯域_40ms_envelope_correlation": 0.9980265035857364, "全帯域_40ms_modulation_mae_db": 0.5792813286859987, "300–1000 Hz_10ms_log_rms_mae_db": 1.5938958796305425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14003229952708424, "300–1000 Hz_10ms_envelope_correlation": 0.9888655180817337, "300–1000 Hz_10ms_modulation_mae_db": 0.488584178539156, "300–1000 Hz_40ms_log_rms_mae_db": 1.026698189464547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09133729736651215, "300–1000 Hz_40ms_envelope_correlation": 0.9966342462235427, "300–1000 Hz_40ms_modulation_mae_db": 0.8824613313321844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8861571669991528, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20510516562253506, "1000–3000 Hz_10ms_envelope_correlation": 0.9707857747392657, "1000–3000 Hz_10ms_modulation_mae_db": 0.8934808228530953, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2648669699752486, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12959740828979818, "1000–3000 Hz_40ms_envelope_correlation": 0.9877699593843597, "1000–3000 Hz_40ms_modulation_mae_db": 0.7725312377413852, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9773400403357506, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2025720373532663, "3000–6000 Hz_10ms_envelope_correlation": 0.9716551667725768, "3000–6000 Hz_10ms_modulation_mae_db": 0.5912224217192913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.30320626025947, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13442909896017244, "3000–6000 Hz_40ms_envelope_correlation": 0.9870998336059, "3000–6000 Hz_40ms_modulation_mae_db": 0.363880590200858, "mr_stft_db": 5.692123352073399, "band_mae_db": 0.16351310051440193, "rms_gain_error_db": -0.2828675542405279, "output_peak": 0.7547808635153072}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2661408081165952, "全帯域_10ms_rms_envelope_nrmse": 0.09485073154369322, "全帯域_10ms_envelope_correlation": 0.9927732649553543, "全帯域_10ms_modulation_mae_db": 0.3548936595635282, "全帯域_40ms_log_rms_mae_db": 0.6831042081521703, "全帯域_40ms_rms_envelope_nrmse": 0.046464618630521946, "全帯域_40ms_envelope_correlation": 0.9982873820873306, "全帯域_40ms_modulation_mae_db": 0.3377513164446368, "300–1000 Hz_10ms_log_rms_mae_db": 1.7815276781022833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12939640344006764, "300–1000 Hz_10ms_envelope_correlation": 0.9892561075598474, "300–1000 Hz_10ms_modulation_mae_db": 0.35088785058340494, "300–1000 Hz_40ms_log_rms_mae_db": 1.1427344287500414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07026971708841204, "300–1000 Hz_40ms_envelope_correlation": 0.9971458617949022, "300–1000 Hz_40ms_modulation_mae_db": 0.3877151002137332, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9024255243279782, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2609614381897848, "1000–3000 Hz_10ms_envelope_correlation": 0.9550279229223119, "1000–3000 Hz_10ms_modulation_mae_db": 1.0818153505510038, "1000–3000 Hz_40ms_log_rms_mae_db": 1.330048614930591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18485083314060466, "1000–3000 Hz_40ms_envelope_correlation": 0.9765825955988828, "1000–3000 Hz_40ms_modulation_mae_db": 1.1450637483630015, "3000–6000 Hz_10ms_log_rms_mae_db": 2.060871213491286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21390741026275908, "3000–6000 Hz_10ms_envelope_correlation": 0.9691690132775592, "3000–6000 Hz_10ms_modulation_mae_db": 0.206213628569331, "3000–6000 Hz_40ms_log_rms_mae_db": 1.435663783312423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11677300110029439, "3000–6000 Hz_40ms_envelope_correlation": 0.990260281175572, "3000–6000 Hz_40ms_modulation_mae_db": 0.21526250132510158, "mr_stft_db": 5.690851973512642, "band_mae_db": 0.1722814178380462, "rms_gain_error_db": -0.1191453226713325, "output_peak": 0.750898324258791}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2434701067449525, "全帯域_10ms_rms_envelope_nrmse": 0.09361155683117153, "全帯域_10ms_envelope_correlation": 0.9930041966015667, "全帯域_10ms_modulation_mae_db": 0.15520521664947168, "全帯域_40ms_log_rms_mae_db": 0.6324709049668721, "全帯域_40ms_rms_envelope_nrmse": 0.05691418601311321, "全帯域_40ms_envelope_correlation": 0.9974177065574392, "全帯域_40ms_modulation_mae_db": 0.2789986068507885, "300–1000 Hz_10ms_log_rms_mae_db": 1.6165518309513534, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13734135165602968, "300–1000 Hz_10ms_envelope_correlation": 0.9882208379705968, "300–1000 Hz_10ms_modulation_mae_db": 0.32699449227912464, "300–1000 Hz_40ms_log_rms_mae_db": 1.0412933721464113, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09111372949668571, "300–1000 Hz_40ms_envelope_correlation": 0.9952896740244621, "300–1000 Hz_40ms_modulation_mae_db": 0.36072459602573453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.907912755028565, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1979189065582674, "1000–3000 Hz_10ms_envelope_correlation": 0.9734692651627577, "1000–3000 Hz_10ms_modulation_mae_db": 0.49006859883424164, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3037135567904088, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1262695250267042, "1000–3000 Hz_40ms_envelope_correlation": 0.9901078721301412, "1000–3000 Hz_40ms_modulation_mae_db": 0.35614486197532064, "3000–6000 Hz_10ms_log_rms_mae_db": 1.950316650031816, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.203420952950633, "3000–6000 Hz_10ms_envelope_correlation": 0.9714868521226453, "3000–6000 Hz_10ms_modulation_mae_db": 0.8514291355608649, "3000–6000 Hz_40ms_log_rms_mae_db": 1.334583013229611, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13620153712496758, "3000–6000 Hz_40ms_envelope_correlation": 0.9864022473306739, "3000–6000 Hz_40ms_modulation_mae_db": 0.6687471999515493, "mr_stft_db": 5.680104228633667, "band_mae_db": 0.2263818965247408, "rms_gain_error_db": -0.13884379917516249, "output_peak": 0.7768291203871266}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 62080, "output_samples": 62080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.097322531583025, "spectral_convergence": 0.32256909525912736, "spectral_convergence_status": "ok", "amplitude_floor": 2.580715109285093e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.442574233641787, "spectral_convergence": 0.2884289894036608, "spectral_convergence_status": "ok", "amplitude_floor": 1.9335078654426385e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.521325127595958, "spectral_convergence": 0.30772561546784094, "spectral_convergence_status": "ok", "amplitude_floor": 1.5577782358148932e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007926248856121452, "output_power": 0.0007803015451113369, "delta_db": -0.06805236636643247, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003328913174136213, "output_power": 0.0030607777602367505, "delta_db": -0.36470670796224675, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00045950159700284816, "output_power": 0.00042668749516632334, "delta_db": -0.32177109844094615, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012582167228939335, "output_power": 0.00013153198479589682, "delta_db": 0.19275920523101117, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.488237987418478e-05, "output_power": 3.563524587888569e-05, "delta_db": 0.09273652413791555, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.708544414240077e-05, "output_power": 6.855095225260797e-05, "delta_db": 0.09385192635670062, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18897963808254215, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2849062468053716, "pitch": null, "output_rms_dbfs": -23.464512751716494, "output_peak": 0.8101435138165535, "reference_rms_dbfs": -23.179606504911124, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.68707396427359}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 62080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.00032928880743725145, "log_rms_mae_db": 1.1970582921828103, "rms_envelope_nrmse": 0.11359586213054478, "envelope_correlation": 0.9900763132951886, "rms_gain_error_db": -0.28491131433246014, "reference_mean_power": 0.004815032927514685, "output_mean_power": 0.004509289621449081}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.902918899290217e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008077922828075312, "output_power": 0.0007696029436847305, "reference_active": true, "delta_db": -0.2103297966586083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023728667544934464, "output_power": 0.00021131059394316012, "reference_active": true, "delta_db": -0.5035208089386817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1018322593991373e-05, "output_power": 1.8345189295877484e-05, "reference_active": true, "delta_db": -0.5907585599552672}], "active_band_count": 3, "active_band_mae_db": 0.4348697218508524}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0003041508818572066, "log_rms_mae_db": 0.6827991868699677, "rms_envelope_nrmse": 0.06316382173793356, "envelope_correlation": 0.9974056313464128, "rms_gain_error_db": -0.28491764784708296, "reference_mean_power": 0.004852582666148238, "output_mean_power": 0.004544448412302129}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.7242336092637014e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007426996521093883, "output_power": 0.000707291033326272, "reference_active": true, "delta_db": -0.21215068132029616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010890208813537604, "output_power": 9.599439841517539e-05, "reference_active": true, "delta_db": -0.5479031589854624}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4730800018618504e-06, "output_power": 1.0581537861822221e-06, "reference_active": true, "delta_db": -1.4367754345910153}], "active_band_count": 3, "active_band_mae_db": 0.7322764249655913}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.0002965570632263201, "log_rms_mae_db": 1.6510030410115255, "rms_envelope_nrmse": 0.16894195415681976, "envelope_correlation": 0.9823174214256012, "rms_gain_error_db": -0.5058143552541399, "reference_mean_power": 0.002818219366063788, "output_mean_power": 0.002508380173509725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.2520173458640518e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044507178368220307, "output_power": 0.0004166545772879055, "reference_active": true, "delta_db": -0.2865390482409646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017793142616898364, "output_power": 0.00014852979543362576, "reference_active": true, "delta_db": -0.7843907684722484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6359305994808858e-05, "output_power": 1.4273343091016137e-05, "reference_active": true, "delta_db": -0.5923917066568548}], "active_band_count": 3, "active_band_mae_db": 0.5544405077900226}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0002772405007925501, "log_rms_mae_db": 1.047900720494063, "rms_envelope_nrmse": 0.10836244739758913, "envelope_correlation": 0.9937792377075878, "rms_gain_error_db": -0.5058146599753562, "reference_mean_power": 0.0028402077631144488, "output_mean_power": 0.002527950956394057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.1541839932451843e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134314303572302, "output_power": 0.00038726470227320423, "reference_active": true, "delta_db": -0.2839557550341142}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.103778868784845e-05, "output_power": 6.728771082319827e-05, "reference_active": true, "delta_db": -0.8075182819791238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3787553243482739e-06, "output_power": 7.196018033897218e-07, "reference_active": true, "delta_db": -2.8239495951162112}], "active_band_count": 3, "active_band_mae_db": 1.3051412107098164}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 6.247605344635176e-05, "log_rms_mae_db": 1.8443859200640074, "rms_envelope_nrmse": 0.21449257660347432, "envelope_correlation": 0.9694414293132896, "rms_gain_error_db": 0.21201595179309088, "reference_mean_power": 0.00014490237542137544, "output_mean_power": 0.0001521518019808032}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 6.946266094979904e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.0294766814481675e-05, "output_power": 2.8831967048854212e-05, "reference_active": true, "delta_db": -0.21493341026977741}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.606273878647975e-06, "output_power": 7.753266932958307e-06, "reference_active": true, "delta_db": -0.45330426267517826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.077953085521739e-06, "output_power": 2.0279532749364542e-06, "reference_active": true, "delta_db": 2.744580844367403}], "active_band_count": 3, "active_band_mae_db": 1.1376061724374529}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 5.585752960561645e-05, "log_rms_mae_db": 1.2694907015285477, "rms_envelope_nrmse": 0.143962476559189, "envelope_correlation": 0.9857732170101977, "rms_gain_error_db": 0.2120128516848421, "reference_mean_power": 0.00014603285132237434, "output_mean_power": 0.00015333872582740308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 6.45401857377675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9246286179300554e-05, "output_power": 2.7860812069901145e-05, "reference_active": true, "delta_db": -0.2107695449381006}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6077467608233e-06, "output_power": 3.507339567188024e-06, "reference_active": true, "delta_db": -0.12258230788009068}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.605305892606176e-08, "output_power": 1.4624292598016417e-07, "reference_active": true, "delta_db": 2.3030855490354063}], "active_band_count": 3, "active_band_mae_db": 0.8788124672845324}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 3.703669227124254e-05, "log_rms_mae_db": 1.9590992817076973, "rms_envelope_nrmse": 0.18851215382984665, "envelope_correlation": 0.9760659198297221, "rms_gain_error_db": 0.10062112385661778, "reference_mean_power": 5.2354216813681856e-05, "output_mean_power": 5.358136579003595e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 3.452458222743894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.744222722260824e-05, "output_power": 1.73386102601509e-05, "reference_active": true, "delta_db": -0.025876552385893124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.482818619951907e-06, "output_power": 7.092571447340393e-06, "reference_active": true, "delta_db": -0.2326149887897675}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0100258421135412e-06, "output_power": 1.2231791640750973e-06, "reference_active": true, "delta_db": 0.831575889836149}], "active_band_count": 3, "active_band_mae_db": 0.3633558103372699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 2.970153096248883e-05, "log_rms_mae_db": 1.3066596654951876, "rms_envelope_nrmse": 0.09970858621348136, "envelope_correlation": 0.992925991928235, "rms_gain_error_db": 0.10061977225899188, "reference_mean_power": 5.276267962923645e-05, "output_mean_power": 5.3999385904247945e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 3.130411452597279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7491368881094015e-05, "output_power": 1.7597712510733544e-05, "reference_active": true, "delta_db": 0.026324196079340837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.132236890482271e-06, "output_power": 3.89689496504044e-06, "reference_active": true, "delta_db": -0.2546651034968149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.718357196867276e-08, "output_power": 7.892533278405764e-08, "reference_active": true, "delta_db": 0.6995333136387332}], "active_band_count": 3, "active_band_mae_db": 0.3268408710716297}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8259247924806483, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4e6259df28fef544ceae820226112b38b55e076c82cb2144a012771f25b3a9e3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9702559266719689, "standard_deviation": 0.006615050609719433, "minimum": 0.9550279229223119, "maximum": 0.975679262675127, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8918228754457287, "standard_deviation": 0.02471668482914879, "minimum": 1.8443859200640074, "maximum": 1.9178597352823747, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8877043308212009, "standard_deviation": 0.298543945885151, "minimum": 0.4776498810605328, "maximum": 1.253209672712809, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2087220136258615, "standard_deviation": 0.022708211041402575, "minimum": 0.19079709114092308, "maximum": 0.2609614381897848, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9874403664256726, "standard_deviation": 0.0047258945185362335, "minimum": 0.9765825955988828, "maximum": 0.9911386172882851, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2857511082158952, "standard_deviation": 0.025514543655185115, "minimum": 1.2610983616184066, "maximum": 1.330048614930591, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7099904519758573, "standard_deviation": 0.2567917239429689, "minimum": 0.35614486197532064, "maximum": 1.1450637483630015, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1331645090828505, "standard_deviation": 0.023152474850103615, "minimum": 0.11097328370943302, "maximum": 0.18485083314060466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9730146663056887, "standard_deviation": 0.002544028223761494, "minimum": 0.9691690132775592, "maximum": 0.9762561459380595, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9745657534826226, "standard_deviation": 0.04454353901055868, "minimum": 1.901562950870028, "maximum": 2.060871213491286, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5352171989782918, "standard_deviation": 0.2150576412032173, "minimum": 0.206213628569331, "maximum": 0.8514291355608649, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19934812665842477, "standard_deviation": 0.010464457421550624, "minimum": 0.18549295030906462, "maximum": 0.21390741026275908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9897605886341863, "standard_deviation": 0.002111916737893791, "minimum": 0.9864022473306739, "maximum": 0.992925991928235, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3383732913544804, "standard_deviation": 0.051408458730535796, "minimum": 1.2993600031331125, "maximum": 1.435663783312423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5143579588070419, "standard_deviation": 0.20256049203694232, "minimum": 0.21526250132510158, "maximum": 0.7586878886442969, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1194043531081034, "standard_deviation": 0.012189596856162012, "minimum": 0.09970858621348136, "maximum": 0.13620153712496758, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.985742891094035, "standard_deviation": 0.0060785655799692855, "minimum": 0.9719189529609776, "maximum": 0.9894589945083369, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6758710953696045, "standard_deviation": 0.06296154670385744, "minimum": 1.5938958796305425, "maximum": 1.7815276781022833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5075444053820883, "standard_deviation": 0.23230144314021955, "minimum": 0.32699449227912464, "maximum": 1.048049398929054, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1483486388665624, "standard_deviation": 0.02736285404195725, "minimum": 0.12788339373000654, "maximum": 0.2080537572790258, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.995104853326093, "standard_deviation": 0.0033140473187877268, "minimum": 0.9874060987106231, "maximum": 0.9971458617949022, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.079370718534062, "standard_deviation": 0.05520356435151938, "minimum": 1.014352496926271, "maximum": 1.1572445078617226, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6415789680817426, "standard_deviation": 0.31257650276472465, "minimum": 0.36072459602573453, "maximum": 1.3051412107098164, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0897721649713677, "standard_deviation": 0.023428865985213255, "minimum": 0.06701231813847815, "maximum": 0.13660017074194394, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24960119792905394, "standard_deviation": 0.10186700320372416, "minimum": 0.14454990396400574, "maximum": 0.39993393093116136, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.684836562708039, "standard_deviation": 0.009442748885090916, "minimum": 5.666707997075229, "maximum": 5.6972403631250925, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8116714345577698, "standard_deviation": 0.07125825251399764, "minimum": 0.750898324258791, "maximum": 0.9579679947805501, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1408554829611788, "standard_deviation": 0.11644453677591649, "minimum": -0.2849062468053716, "maximum": 0.05518869147121393, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9908892163808335, "standard_deviation": 0.0037795965697949834, "minimum": 0.9818096588892166, "maximum": 0.9930041966015667, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2707860506054183, "standard_deviation": 0.050120287177402356, "minimum": 1.1970582921828103, "maximum": 1.3383272182688837, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3530570772499209, "standard_deviation": 0.1961706860380267, "minimum": 0.1546762017421809, "maximum": 0.7744681357419262, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10600269265031503, "standard_deviation": 0.019391405819901876, "minimum": 0.09361155683117153, "maximum": 0.15157780541399182, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972987504926806, "standard_deviation": 0.0018165758869896163, "minimum": 0.992896858793525, "maximum": 0.9984571267955309, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.673066375801459, "standard_deviation": 0.039190868614759775, "minimum": 0.6154521554216398, "maximum": 0.7232196630638424, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40945718760194416, "standard_deviation": 0.21557512549324992, "minimum": 0.1792946804392589, "maximum": 0.7322764249655913, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.057928123028702566, "standard_deviation": 0.01533487375624891, "minimum": 0.042573142444083535, "maximum": 0.09234541978406019, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0032", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.422255754470825, "predicted_mos": 2.422255754470825, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "b6272458f9b7aff2c7acd8a219e3e942967215e5e68ce5cba86cec6e1aba56cf", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7148603997193277, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2568628593491251, "全帯域_10ms_rms_envelope_nrmse": 0.0888222372372317, "全帯域_10ms_envelope_correlation": 0.993638613516637, "全帯域_10ms_modulation_mae_db": 0.08856275595490332, "全帯域_40ms_log_rms_mae_db": 0.6506287559115675, "全帯域_40ms_rms_envelope_nrmse": 0.044356085366454405, "全帯域_40ms_envelope_correlation": 0.9983264794475439, "全帯域_40ms_modulation_mae_db": 0.19748598949816162, "300–1000 Hz_10ms_log_rms_mae_db": 1.5566060822987358, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12808867655568937, "300–1000 Hz_10ms_envelope_correlation": 0.9891928186397763, "300–1000 Hz_10ms_modulation_mae_db": 0.20572465726234657, "300–1000 Hz_40ms_log_rms_mae_db": 1.0431082541872483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06907800991669988, "300–1000 Hz_40ms_envelope_correlation": 0.9967913559916058, "300–1000 Hz_40ms_modulation_mae_db": 0.16420546734375624, "1000–3000 Hz_10ms_log_rms_mae_db": 1.833277094749268, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20313707624146457, "1000–3000 Hz_10ms_envelope_correlation": 0.9715798782210762, "1000–3000 Hz_10ms_modulation_mae_db": 1.2020989596331262, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1915983887972599, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11267950088030167, "1000–3000 Hz_40ms_envelope_correlation": 0.9908197059538716, "1000–3000 Hz_40ms_modulation_mae_db": 0.9508265733688578, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9407700312299987, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20737762901221732, "3000–6000 Hz_10ms_envelope_correlation": 0.9705113163135954, "3000–6000 Hz_10ms_modulation_mae_db": 0.7154616840512172, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2872792528932064, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12649855916133565, "3000–6000 Hz_40ms_envelope_correlation": 0.9892510216567372, "3000–6000 Hz_40ms_modulation_mae_db": 0.7971698475654586, "mr_stft_db": 5.64523337286198, "band_mae_db": 0.26015986951755915, "rms_gain_error_db": -0.0030449690200812264, "output_peak": 0.7997237499946956}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1864097817028396, "全帯域_10ms_rms_envelope_nrmse": 0.09537343470528639, "全帯域_10ms_envelope_correlation": 0.9927242409457284, "全帯域_10ms_modulation_mae_db": 0.27662436461167245, "全帯域_40ms_log_rms_mae_db": 0.639793446773134, "全帯域_40ms_rms_envelope_nrmse": 0.04744196866643745, "全帯域_40ms_envelope_correlation": 0.998258645796011, "全帯域_40ms_modulation_mae_db": 0.3256684899079656, "300–1000 Hz_10ms_log_rms_mae_db": 1.6502089387117564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14469058655294412, "300–1000 Hz_10ms_envelope_correlation": 0.9867289616964214, "300–1000 Hz_10ms_modulation_mae_db": 0.4145279776460819, "300–1000 Hz_40ms_log_rms_mae_db": 1.0606805958569159, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07538735557131769, "300–1000 Hz_40ms_envelope_correlation": 0.9970662185434759, "300–1000 Hz_40ms_modulation_mae_db": 0.8438260407723055, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8435352528396944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21519142746860986, "1000–3000 Hz_10ms_envelope_correlation": 0.96806332688599, "1000–3000 Hz_10ms_modulation_mae_db": 1.1348984286378994, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1628389345063743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11207590438616244, "1000–3000 Hz_40ms_envelope_correlation": 0.9908931351288205, "1000–3000 Hz_40ms_modulation_mae_db": 0.8962316691129445, "3000–6000 Hz_10ms_log_rms_mae_db": 1.897355313887419, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19452739703121, "3000–6000 Hz_10ms_envelope_correlation": 0.9740197876270719, "3000–6000 Hz_10ms_modulation_mae_db": 0.9401924053006869, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2574606299525848, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11272062317417973, "3000–6000 Hz_40ms_envelope_correlation": 0.9913351118781094, "3000–6000 Hz_40ms_modulation_mae_db": 0.7590054947732471, "mr_stft_db": 5.647634339371604, "band_mae_db": 0.16285928674326478, "rms_gain_error_db": -0.1371909022248435, "output_peak": 0.7875235292978616}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.303503244012356, "全帯域_10ms_rms_envelope_nrmse": 0.09153189070228746, "全帯域_10ms_envelope_correlation": 0.9935043579536321, "全帯域_10ms_modulation_mae_db": 0.31217940981336306, "全帯域_40ms_log_rms_mae_db": 0.698799085036404, "全帯域_40ms_rms_envelope_nrmse": 0.043756028693570474, "全帯域_40ms_envelope_correlation": 0.9987986807493412, "全帯域_40ms_modulation_mae_db": 0.46559511175225055, "300–1000 Hz_10ms_log_rms_mae_db": 1.6579007114398447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13663401887544818, "300–1000 Hz_10ms_envelope_correlation": 0.9882444253831145, "300–1000 Hz_10ms_modulation_mae_db": 0.5475870349997299, "300–1000 Hz_40ms_log_rms_mae_db": 1.0400277176697978, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06515068937788031, "300–1000 Hz_40ms_envelope_correlation": 0.9980466293836884, "300–1000 Hz_40ms_modulation_mae_db": 0.4683918031009855, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8453658303830431, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17799336296410057, "1000–3000 Hz_10ms_envelope_correlation": 0.9785214304312163, "1000–3000 Hz_10ms_modulation_mae_db": 0.43080042938302743, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2014317915359576, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11470823813061845, "1000–3000 Hz_40ms_envelope_correlation": 0.9907108400163421, "1000–3000 Hz_40ms_modulation_mae_db": 0.41444812242551704, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9575593824492326, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21816180631509802, "3000–6000 Hz_10ms_envelope_correlation": 0.9670953991132237, "3000–6000 Hz_10ms_modulation_mae_db": 0.4057940667246765, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2707274428581117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13284875023430578, "3000–6000 Hz_40ms_envelope_correlation": 0.987771105003682, "3000–6000 Hz_40ms_modulation_mae_db": 0.5795806021464864, "mr_stft_db": 5.633899978381588, "band_mae_db": 0.24900761785094694, "rms_gain_error_db": -0.2032568279947841, "output_peak": 0.7685248240777739}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2334465881316037, "全帯域_10ms_rms_envelope_nrmse": 0.0880246298509282, "全帯域_10ms_envelope_correlation": 0.9938545665326151, "全帯域_10ms_modulation_mae_db": 0.4127892397494714, "全帯域_40ms_log_rms_mae_db": 0.6170805601946292, "全帯域_40ms_rms_envelope_nrmse": 0.045418795004309946, "全帯域_40ms_envelope_correlation": 0.9983179665049859, "全帯域_40ms_modulation_mae_db": 0.39358140070544295, "300–1000 Hz_10ms_log_rms_mae_db": 1.6470373182749274, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1041961645560244, "300–1000 Hz_10ms_envelope_correlation": 0.9930704621568126, "300–1000 Hz_10ms_modulation_mae_db": 0.7322839204478377, "300–1000 Hz_40ms_log_rms_mae_db": 1.0883924542870231, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0646291803626057, "300–1000 Hz_40ms_envelope_correlation": 0.9973174461593559, "300–1000 Hz_40ms_modulation_mae_db": 0.8594222749408061, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8604007759628391, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2059728252190935, "1000–3000 Hz_10ms_envelope_correlation": 0.9704299161983133, "1000–3000 Hz_10ms_modulation_mae_db": 1.1695618941285426, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1976364235890427, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11916696299396552, "1000–3000 Hz_40ms_envelope_correlation": 0.9896833369778564, "1000–3000 Hz_40ms_modulation_mae_db": 1.1809580897317655, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9710045100916649, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2233051347153568, "3000–6000 Hz_10ms_envelope_correlation": 0.9654850285388545, "3000–6000 Hz_10ms_modulation_mae_db": 0.8752069238536265, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2815093435753973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13931566000353038, "3000–6000 Hz_40ms_envelope_correlation": 0.9863945745727292, "3000–6000 Hz_40ms_modulation_mae_db": 1.105652937777043, "mr_stft_db": 5.630342493563552, "band_mae_db": 0.38583567981800243, "rms_gain_error_db": 0.06293188150899472, "output_peak": 0.7608535436881348}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1744740217576388, "全帯域_10ms_rms_envelope_nrmse": 0.08587784028743256, "全帯域_10ms_envelope_correlation": 0.9940406055056906, "全帯域_10ms_modulation_mae_db": 0.30999456361346783, "全帯域_40ms_log_rms_mae_db": 0.5979351290343122, "全帯域_40ms_rms_envelope_nrmse": 0.045027024636550436, "全帯域_40ms_envelope_correlation": 0.9983131614878048, "全帯域_40ms_modulation_mae_db": 0.19077157779779744, "300–1000 Hz_10ms_log_rms_mae_db": 1.58582004721253, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13194304377347726, "300–1000 Hz_10ms_envelope_correlation": 0.9887888952512682, "300–1000 Hz_10ms_modulation_mae_db": 0.3042792762760855, "300–1000 Hz_40ms_log_rms_mae_db": 1.0120012558883638, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06995090211169804, "300–1000 Hz_40ms_envelope_correlation": 0.9971794801117116, "300–1000 Hz_40ms_modulation_mae_db": 0.6308753749988354, "1000–3000 Hz_10ms_log_rms_mae_db": 1.746025499198716, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20102127102944253, "1000–3000 Hz_10ms_envelope_correlation": 0.9721253504302108, "1000–3000 Hz_10ms_modulation_mae_db": 0.926450116712501, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1608491344660097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12227712485652709, "1000–3000 Hz_40ms_envelope_correlation": 0.9891738332117614, "1000–3000 Hz_40ms_modulation_mae_db": 0.8307202261493902, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9665166942797012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1981435801817764, "3000–6000 Hz_10ms_envelope_correlation": 0.9742098649505609, "3000–6000 Hz_10ms_modulation_mae_db": 0.7964854997153211, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2889219593272983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12437817168520814, "3000–6000 Hz_40ms_envelope_correlation": 0.9894431729370583, "3000–6000 Hz_40ms_modulation_mae_db": 0.7046691317646814, "mr_stft_db": 5.63950120660613, "band_mae_db": 0.27850662463575976, "rms_gain_error_db": -0.07238806450664788, "output_peak": 0.7764926340896393}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3046328354520702, "全帯域_10ms_rms_envelope_nrmse": 0.09170195723325997, "全帯域_10ms_envelope_correlation": 0.9932747332308975, "全帯域_10ms_modulation_mae_db": 0.4052811886506329, "全帯域_40ms_log_rms_mae_db": 0.7204482762410609, "全帯域_40ms_rms_envelope_nrmse": 0.04390993122094666, "全帯域_40ms_envelope_correlation": 0.9984086445469191, "全帯域_40ms_modulation_mae_db": 0.5204504503919588, "300–1000 Hz_10ms_log_rms_mae_db": 1.694047320425374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13505572571491714, "300–1000 Hz_10ms_envelope_correlation": 0.9880620182807407, "300–1000 Hz_10ms_modulation_mae_db": 0.5118923761368928, "300–1000 Hz_40ms_log_rms_mae_db": 0.9672938586697236, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05554883641893999, "300–1000 Hz_40ms_envelope_correlation": 0.9979186768149125, "300–1000 Hz_40ms_modulation_mae_db": 0.9497630594245953, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8741299194265957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23862370845542533, "1000–3000 Hz_10ms_envelope_correlation": 0.9621860037144958, "1000–3000 Hz_10ms_modulation_mae_db": 0.9398196888882957, "1000–3000 Hz_40ms_log_rms_mae_db": 1.204025007586778, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13723613701964377, "1000–3000 Hz_40ms_envelope_correlation": 0.9871727363581667, "1000–3000 Hz_40ms_modulation_mae_db": 0.5662560259131576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9655550001153765, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21067559463997304, "3000–6000 Hz_10ms_envelope_correlation": 0.9693706551786652, "3000–6000 Hz_10ms_modulation_mae_db": 0.37186506663882984, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2845333212744892, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11887942923242473, "3000–6000 Hz_40ms_envelope_correlation": 0.9903164663625071, "3000–6000 Hz_40ms_modulation_mae_db": 0.2960775055653597, "mr_stft_db": 5.6402695835688945, "band_mae_db": 0.3033788651940048, "rms_gain_error_db": 0.036406475311354256, "output_peak": 0.7863621719832254}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2585961403654613, "全帯域_10ms_rms_envelope_nrmse": 0.10223678893342734, "全帯域_10ms_envelope_correlation": 0.9915691788768416, "全帯域_10ms_modulation_mae_db": 0.4112196612224439, "全帯域_40ms_log_rms_mae_db": 0.6611499464894166, "全帯域_40ms_rms_envelope_nrmse": 0.0547940534059951, "全帯域_40ms_envelope_correlation": 0.9974357127832854, "全帯域_40ms_modulation_mae_db": 0.2595520307074995, "300–1000 Hz_10ms_log_rms_mae_db": 1.7100420263819955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15015213229513572, "300–1000 Hz_10ms_envelope_correlation": 0.985150324932461, "300–1000 Hz_10ms_modulation_mae_db": 0.7566783142221559, "300–1000 Hz_40ms_log_rms_mae_db": 1.0801625075240369, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07719016859008397, "300–1000 Hz_40ms_envelope_correlation": 0.9959106296387882, "300–1000 Hz_40ms_modulation_mae_db": 0.3277988779868049, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8871439528547278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21169514957706076, "1000–3000 Hz_10ms_envelope_correlation": 0.9701256282686898, "1000–3000 Hz_10ms_modulation_mae_db": 0.5805764436315538, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2420589780414242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14627598686808352, "1000–3000 Hz_40ms_envelope_correlation": 0.9870148727114253, "1000–3000 Hz_40ms_modulation_mae_db": 0.5753613051896228, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9641543409767699, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2116337355804159, "3000–6000 Hz_10ms_envelope_correlation": 0.9696626330975215, "3000–6000 Hz_10ms_modulation_mae_db": 0.3879785159726061, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3138413710264953, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11674447366333873, "3000–6000 Hz_40ms_envelope_correlation": 0.9901659221021438, "3000–6000 Hz_40ms_modulation_mae_db": 0.28644586814826006, "mr_stft_db": 5.656729438973168, "band_mae_db": 0.29016366379644265, "rms_gain_error_db": -0.002787958296179933, "output_peak": 0.7538472587552435}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2506499395178197, "全帯域_10ms_rms_envelope_nrmse": 0.09991545356281949, "全帯域_10ms_envelope_correlation": 0.9928903219752451, "全帯域_10ms_modulation_mae_db": 0.12243656565899626, "全帯域_40ms_log_rms_mae_db": 0.6778404901179096, "全帯域_40ms_rms_envelope_nrmse": 0.06114962026730227, "全帯域_40ms_envelope_correlation": 0.998069438656744, "全帯域_40ms_modulation_mae_db": 0.20150157038538544, "300–1000 Hz_10ms_log_rms_mae_db": 1.5427021724886774, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13616582293991258, "300–1000 Hz_10ms_envelope_correlation": 0.9893810777458258, "300–1000 Hz_10ms_modulation_mae_db": 0.4073975304681598, "300–1000 Hz_40ms_log_rms_mae_db": 1.0175672451031388, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08650169832462991, "300–1000 Hz_40ms_envelope_correlation": 0.996995279156981, "300–1000 Hz_40ms_modulation_mae_db": 0.394182812926631, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8233233341233512, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22760074810441264, "1000–3000 Hz_10ms_envelope_correlation": 0.9638165143601648, "1000–3000 Hz_10ms_modulation_mae_db": 0.5593002373467884, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1637528198055669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13921648691018676, "1000–3000 Hz_40ms_envelope_correlation": 0.9868217668555249, "1000–3000 Hz_40ms_modulation_mae_db": 0.5680600027241572, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9060473047649318, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19236428969201247, "3000–6000 Hz_10ms_envelope_correlation": 0.9746508062094905, "3000–6000 Hz_10ms_modulation_mae_db": 0.665958693046953, "3000–6000 Hz_40ms_log_rms_mae_db": 1.257333331204318, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11259644279704958, "3000–6000 Hz_40ms_envelope_correlation": 0.9914367373429533, "3000–6000 Hz_40ms_modulation_mae_db": 0.5849955446646675, "mr_stft_db": 5.641250530393074, "band_mae_db": 0.25201257402972405, "rms_gain_error_db": -0.3432974414508572, "output_peak": 0.7371725697615376}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 62080, "output_samples": 62080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.066869514764872, "spectral_convergence": 0.315895670270802, "spectral_convergence_status": "ok", "amplitude_floor": 2.580715109285093e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.387767075958217, "spectral_convergence": 0.295707366510141, "spectral_convergence_status": "ok", "amplitude_floor": 1.9335078654426385e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.48106352786285, "spectral_convergence": 0.3351674565819231, "spectral_convergence_status": "ok", "amplitude_floor": 1.5577782358148932e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007926248856121452, "output_power": 0.0008385484497033998, "delta_db": 0.24460456840354883, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003328913174136213, "output_power": 0.0032181478817017734, "delta_db": -0.14696470779113657, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00045950159700284816, "output_power": 0.0005221990963761833, "delta_db": 0.555490908556758, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012582167228939335, "output_power": 0.00012526643503340547, "delta_db": -0.019207350921058693, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.488237987418478e-05, "output_power": 3.783835472801817e-05, "delta_db": 0.3532613701500238, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.708544414240077e-05, "output_power": 6.345783750291187e-05, "delta_db": -0.24143031128282916, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.26015986951755915, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0030449690200812264, "pitch": null, "output_rms_dbfs": -23.182651473931205, "output_peak": 0.7997237499946956, "reference_rms_dbfs": -23.179606504911124, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.64523337286198}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 62080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.00032928880743725145, "log_rms_mae_db": 1.2568628593491251, "rms_envelope_nrmse": 0.0888222372372317, "envelope_correlation": 0.993638613516637, "rms_gain_error_db": -0.003045191019465889, "reference_mean_power": 0.004815032927514685, "output_mean_power": 0.004811657900605821}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.902918899290217e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008077922828075312, "output_power": 0.0007820051898537861, "reference_active": true, "delta_db": -0.1409006454179121}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023728667544934464, "output_power": 0.00023306785891047654, "reference_active": true, "delta_db": -0.07790965094470366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1018322593991373e-05, "output_power": 2.124642431378974e-05, "reference_active": true, "delta_db": 0.04687797150209418}], "active_band_count": 3, "active_band_mae_db": 0.08856275595490332}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0003041508818572066, "log_rms_mae_db": 0.6506287559115675, "rms_envelope_nrmse": 0.044356085366454405, "envelope_correlation": 0.9983264794475439, "rms_gain_error_db": -0.0030388999046394114, "reference_mean_power": 0.004852582666148238, "output_mean_power": 0.00484918834374462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.7242336092637014e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007426996521093883, "output_power": 0.0007200110249987583, "reference_active": true, "delta_db": -0.13474073972347875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010890208813537604, "output_power": 0.00010958988330243987, "reference_active": true, "delta_db": 0.02734257285886596}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4730800018618504e-06, "output_power": 1.3341014865704512e-06, "reference_active": true, "delta_db": -0.4303746559121402}], "active_band_count": 3, "active_band_mae_db": 0.19748598949816162}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.0002965570632263201, "log_rms_mae_db": 1.5566060822987358, "rms_envelope_nrmse": 0.12808867655568937, "envelope_correlation": 0.9891928186397763, "rms_gain_error_db": -0.10354542387001192, "reference_mean_power": 0.002818219366063788, "output_mean_power": 0.0027518214560975983}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.2520173458640518e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044507178368220307, "output_power": 0.0004331746345784265, "reference_active": true, "delta_db": -0.11767044387192992}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017793142616898364, "output_power": 0.00017013901633421998, "reference_active": true, "delta_db": -0.19448742171269437}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6359305994808858e-05, "output_power": 1.7549570266031744e-05, "reference_active": true, "delta_db": 0.3050161062024155}], "active_band_count": 3, "active_band_mae_db": 0.20572465726234657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0002772405007925501, "log_rms_mae_db": 1.0431082541872483, "rms_envelope_nrmse": 0.06907800991669988, "envelope_correlation": 0.9967913559916058, "rms_gain_error_db": -0.103546048919409, "reference_mean_power": 0.0028402077631144488, "output_mean_power": 0.0027732914022107085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.1541839932451843e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134314303572302, "output_power": 0.0004058526974823385, "reference_active": true, "delta_db": -0.08035053146203913}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.103778868784845e-05, "output_power": 7.993615540947435e-05, "reference_active": true, "delta_db": -0.05944325036966781}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3787553243482739e-06, "output_power": 1.2711738217930437e-06, "reference_active": true, "delta_db": -0.3528226201995618}], "active_band_count": 3, "active_band_mae_db": 0.16420546734375624}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 6.247605344635176e-05, "log_rms_mae_db": 1.833277094749268, "rms_envelope_nrmse": 0.20313707624146457, "envelope_correlation": 0.9715798782210762, "rms_gain_error_db": 0.027132694252207015, "reference_mean_power": 0.00014490237542137544, "output_mean_power": 0.0001458104916893721}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 6.946266094979904e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.0294766814481675e-05, "output_power": 2.7446873596906282e-05, "reference_active": true, "delta_db": -0.4287473175382573}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.606273878647975e-06, "output_power": 7.055837068078986e-06, "reference_active": true, "delta_db": -0.8626661897824911}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.077953085521739e-06, "output_power": 1.8369113480127356e-06, "reference_active": true, "delta_db": 2.31488337157863}], "active_band_count": 3, "active_band_mae_db": 1.2020989596331262}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 5.585752960561645e-05, "log_rms_mae_db": 1.1915983887972599, "rms_envelope_nrmse": 0.11267950088030167, "envelope_correlation": 0.9908197059538716, "rms_gain_error_db": 0.02713054177782781, "reference_mean_power": 0.00014603285132237434, "output_mean_power": 0.00014694797955370754}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 6.45401857377675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9246286179300554e-05, "output_power": 2.6256708917070616e-05, "reference_active": true, "delta_db": -0.4683043573043616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6077467608233e-06, "output_power": 3.3260794591056156e-06, "reference_active": true, "delta_db": -0.3530342525313457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.605305892606176e-08, "output_power": 1.373663706166649e-07, "reference_active": true, "delta_db": 2.031141110270866}], "active_band_count": 3, "active_band_mae_db": 0.9508265733688578}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 3.703669227124254e-05, "log_rms_mae_db": 1.9407700312299987, "rms_envelope_nrmse": 0.20737762901221732, "envelope_correlation": 0.9705113163135954, "rms_gain_error_db": -0.36488801432109885, "reference_mean_power": 5.2354216813681856e-05, "output_mean_power": 4.813520943624642e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 3.452458222743894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.744222722260824e-05, "output_power": 1.5559442385524763e-05, "reference_active": true, "delta_db": -0.49607910991245235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.482818619951907e-06, "output_power": 6.480602981371928e-06, "reference_active": true, "delta_db": -0.6244980199778145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0100258421135412e-06, "output_power": 1.2791258561953715e-06, "reference_active": true, "delta_db": 1.0258079222633845}], "active_band_count": 3, "active_band_mae_db": 0.7154616840512172}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 2.970153096248883e-05, "log_rms_mae_db": 1.2872792528932064, "rms_envelope_nrmse": 0.12649855916133565, "envelope_correlation": 0.9892510216567372, "rms_gain_error_db": -0.36489002075895866, "reference_mean_power": 5.276267962923645e-05, "output_mean_power": 4.851073352983238e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 3.130411452597279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7491368881094015e-05, "output_power": 1.566596835834109e-05, "reference_active": true, "delta_db": -0.47866553825528796}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.132236890482271e-06, "output_power": 3.6777337885543463e-06, "reference_active": true, "delta_db": -0.5060492075475812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.718357196867276e-08, "output_power": 9.288435508963872e-08, "reference_active": true, "delta_db": 1.4067947968935068}], "active_band_count": 3, "active_band_mae_db": 0.7971698475654586}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8289880875501082, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3b8807775ac022e624c7960350e8f5f15664a398c78cbd95a5bdcab204ed4ab0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9696060060637697, "standard_deviation": 0.005093853470077073, "minimum": 0.9621860037144958, "maximum": 0.9785214304312163, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8391502074422794, "standard_deviation": 0.04308401055836605, "minimum": 1.746025499198716, "maximum": 1.8871439528547278, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8679382747952168, "standard_deviation": 0.3050578924374661, "minimum": 0.43080042938302743, "maximum": 1.2020989596331262, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2101544461324512, "standard_deviation": 0.01823300339443349, "minimum": 0.17799336296410057, "maximum": 0.23862370845542533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9890362784017211, "standard_deviation": 0.0017861860712428074, "minimum": 0.9868217668555249, "maximum": 0.9908931351288205, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1905239347910517, "standard_deviation": 0.027704070224925964, "minimum": 1.1608491344660097, "maximum": 1.2420589780414242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7478577518269266, "standard_deviation": 0.25742590520947095, "minimum": 0.41444812242551704, "maximum": 1.1809580897317655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12545454275568615, "standard_deviation": 0.013464407958119836, "minimum": 0.11207590438616244, "maximum": 0.14627598686808352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9706256863786229, "standard_deviation": 0.00342204608013142, "minimum": 0.9654850285388545, "maximum": 0.9746508062094905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9461203222243868, "standard_deviation": 0.02898649393592823, "minimum": 1.897355313887419, "maximum": 1.9710045100916649, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6448678569129896, "standard_deviation": 0.22877557005831053, "minimum": 0.37186506663882984, "maximum": 0.9401924053006869, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2070236458960075, "standard_deviation": 0.011172472442602688, "minimum": 0.19236428969201247, "maximum": 0.2233051347153568, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.98951426398199, "standard_deviation": 0.0017289548263812792, "minimum": 0.9863945745727292, "maximum": 0.9914367373429533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2802008315139877, "standard_deviation": 0.018547680331210566, "minimum": 1.257333331204318, "maximum": 1.3138413710264953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6391996165506505, "standard_deviation": 0.26987655881784595, "minimum": 0.28644586814826006, "maximum": 1.105652937777043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12299776374392159, "standard_deviation": 0.009616564426050168, "minimum": 0.11259644279704958, "maximum": 0.13931566000353038, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9885773730108026, "standard_deviation": 0.002292691303577629, "minimum": 0.985150324932461, "maximum": 0.9930704621568126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.63054557715423, "standard_deviation": 0.062040276312250495, "minimum": 1.5427021724886774, "maximum": 1.7100420263819955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.48504638593241123, "standard_deviation": 0.1931845563253563, "minimum": 0.20572465726234657, "maximum": 0.7566783142221559, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1333657714079436, "standard_deviation": 0.013691232196716108, "minimum": 0.1041961645560244, "maximum": 0.15015213229513572, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997153214475065, "standard_deviation": 0.0006675386791015662, "minimum": 0.9959106296387882, "maximum": 0.9980466293836884, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0386542361482811, "standard_deviation": 0.03955296306106843, "minimum": 0.9672938586697236, "maximum": 1.0883924542870231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.57980821393684, "standard_deviation": 0.2854742002659259, "minimum": 0.16420546734375624, "maximum": 0.9497630594245953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07042960508423193, "standard_deviation": 0.009362553147098175, "minimum": 0.05554883641893999, "maximum": 0.08650169832462991, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27274052269821303, "standard_deviation": 0.06243979254341634, "minimum": 0.16285928674326478, "maximum": 0.38583567981800243, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.641857617964999, "standard_deviation": 0.0081967109449785, "minimum": 5.630342493563552, "maximum": 5.656729438973168, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.771312535206014, "standard_deviation": 0.020392762741935532, "minimum": 0.7371725697615376, "maximum": 0.7997237499946956, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08282847583413061, "standard_deviation": 0.13817498915717746, "minimum": -0.3432974414508572, "maximum": 0.06293188150899472, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931870773171609, "standard_deviation": 0.0007933801399398034, "minimum": 0.9915691788768416, "maximum": 0.9940406055056906, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2460719262861144, "standard_deviation": 0.04761036098312039, "minimum": 1.1744740217576388, "maximum": 1.3046328354520702, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2923859686593689, "standard_deviation": 0.12690868347118808, "minimum": 0.08856275595490332, "maximum": 0.4127892397494714, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09293552906408414, "standard_deviation": 0.0058004327030183895, "minimum": 0.08587784028743256, "maximum": 0.10223678893342734, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982410912465794, "standard_deviation": 0.0003846672912818884, "minimum": 0.9974357127832854, "maximum": 0.9987986807493412, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6579594612248043, "standard_deviation": 0.04077098006299214, "minimum": 0.5979351290343122, "maximum": 0.7204482762410609, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.31932582764330775, "standard_deviation": 0.12885570033180585, "minimum": 0.19077157779779744, "maximum": 0.5204504503919588, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048231688407695844, "standard_deviation": 0.006352422578896494, "minimum": 0.043756028693570474, "maximum": 0.06114962026730227, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8230-279154-0033", "language": "en", "style": "normal", "speaker": "librispeech_8230", "source_utterance_id": "8230-279154-0033", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8230-279154-0033.wav", "source_file_sha256": "0fb755fb9e1ea9eb4c58a08419056517596ca9a9fd59409480bb6663c34f8290", "decoded_pcm_sha256": "bd287f891d96a4cd6b7a04337bf6be93b04db8cbec30940ae4945f8b4efde0f4", "original_source_sha256": "1bcf539698173ec1996fa39e5b4c5faf7a8ddf5b5f9636d269ca583befc17eb6", "corpus_entry_id": "librispeech_8230-279154-0033"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.739860773086548, "predicted_mos": 3.739860773086548, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "07d541eb914552fe1bce57ecc08bb98484f854cd0d6939f87f8b557592a63ec7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.672668313000031, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8230-279154-0033", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "07d541eb914552fe1bce57ecc08bb98484f854cd0d6939f87f8b557592a63ec7", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "88f96bf577b1f1b95250b099c6716d41e7e2bed368df2791fdb21adff8013cf5"}, "rows": [{"case": "librispeech_8230-279154-0033", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.641920804977417, "predicted_mos": 2.641920804977417, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d31ad2623450d12eca9fbf672c4668294af4b809b889ea6f23f1d3acb97ff150", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.123391476001416, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8230-279154-0033", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d31ad2623450d12eca9fbf672c4668294af4b809b889ea6f23f1d3acb97ff150", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1029678183740816, "全帯域_10ms_rms_envelope_nrmse": 0.11011490302020228, "全帯域_10ms_envelope_correlation": 0.987076586156713, "全帯域_10ms_modulation_mae_db": 0.2304840474539109, "全帯域_40ms_log_rms_mae_db": 0.6025510297057284, "全帯域_40ms_rms_envelope_nrmse": 0.05568887566121783, "全帯域_40ms_envelope_correlation": 0.9965762710733819, "全帯域_40ms_modulation_mae_db": 0.20065872162204687, "300–1000 Hz_10ms_log_rms_mae_db": 1.4108987218912057, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14850734076644187, "300–1000 Hz_10ms_envelope_correlation": 0.9818030537215683, "300–1000 Hz_10ms_modulation_mae_db": 0.6469417779501434, "300–1000 Hz_40ms_log_rms_mae_db": 0.8961694634571479, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09276786347914669, "300–1000 Hz_40ms_envelope_correlation": 0.9934365311744063, "300–1000 Hz_40ms_modulation_mae_db": 0.6575430359814902, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5350435141191443, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15893260711832294, "1000–3000 Hz_10ms_envelope_correlation": 0.9801938809310399, "1000–3000 Hz_10ms_modulation_mae_db": 1.3900517642536432, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0409772535227155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0934449171757039, "1000–3000 Hz_40ms_envelope_correlation": 0.992839338204695, "1000–3000 Hz_40ms_modulation_mae_db": 1.4559725041496439, "3000–6000 Hz_10ms_log_rms_mae_db": 2.304759785527561, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36599015203741275, "3000–6000 Hz_10ms_envelope_correlation": 0.9048159407310922, "3000–6000 Hz_10ms_modulation_mae_db": 1.580263349481774, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8268334503849872, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2976237639499564, "3000–6000 Hz_40ms_envelope_correlation": 0.9310856657643958, "3000–6000 Hz_40ms_modulation_mae_db": 1.2948626456008534, "mr_stft_db": 5.752445989094503, "band_mae_db": 0.23423869999691446, "rms_gain_error_db": -0.12347810343270077, "output_peak": 0.9730240897532715}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1545613624190516, "全帯域_10ms_rms_envelope_nrmse": 0.1086703852388622, "全帯域_10ms_envelope_correlation": 0.9876735224551358, "全帯域_10ms_modulation_mae_db": 1.1045767117762677, "全帯域_40ms_log_rms_mae_db": 0.6726428777717276, "全帯域_40ms_rms_envelope_nrmse": 0.06134087415863202, "全帯域_40ms_envelope_correlation": 0.9957906283388382, "全帯域_40ms_modulation_mae_db": 0.8825810208657083, "300–1000 Hz_10ms_log_rms_mae_db": 1.3379857803330102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13388716418020727, "300–1000 Hz_10ms_envelope_correlation": 0.9844789359510391, "300–1000 Hz_10ms_modulation_mae_db": 1.0638479145744075, "300–1000 Hz_40ms_log_rms_mae_db": 0.8087436157526616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07381567174918859, "300–1000 Hz_40ms_envelope_correlation": 0.9949161118124155, "300–1000 Hz_40ms_modulation_mae_db": 0.8872259394127422, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5782341166863454, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19385566047680639, "1000–3000 Hz_10ms_envelope_correlation": 0.9706128183805854, "1000–3000 Hz_10ms_modulation_mae_db": 2.7943934580132628, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0916067177011732, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12615421524142872, "1000–3000 Hz_40ms_envelope_correlation": 0.986695987032766, "1000–3000 Hz_40ms_modulation_mae_db": 3.0188927881375363, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2800656106025845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36603786763619617, "3000–6000 Hz_10ms_envelope_correlation": 0.9032596036900455, "3000–6000 Hz_10ms_modulation_mae_db": 1.5385073579791069, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8107243354100637, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29642978408569637, "3000–6000 Hz_40ms_envelope_correlation": 0.9303492895915112, "3000–6000 Hz_40ms_modulation_mae_db": 1.5658068695180525, "mr_stft_db": 5.745307073099155, "band_mae_db": 0.11369562896636297, "rms_gain_error_db": 0.07506948532443694, "output_peak": 0.8584763344194376}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1419636280650811, "全帯域_10ms_rms_envelope_nrmse": 0.1016214790121247, "全帯域_10ms_envelope_correlation": 0.988985580159986, "全帯域_10ms_modulation_mae_db": 0.9566287600888738, "全帯域_40ms_log_rms_mae_db": 0.6100679446364166, "全帯域_40ms_rms_envelope_nrmse": 0.04723568294428085, "全帯域_40ms_envelope_correlation": 0.9974086557294225, "全帯域_40ms_modulation_mae_db": 0.731684346630005, "300–1000 Hz_10ms_log_rms_mae_db": 1.3573513963367267, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13696354378176162, "300–1000 Hz_10ms_envelope_correlation": 0.9838602921438283, "300–1000 Hz_10ms_modulation_mae_db": 0.8233357027682887, "300–1000 Hz_40ms_log_rms_mae_db": 0.8460487202577591, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06930817385829868, "300–1000 Hz_40ms_envelope_correlation": 0.9958185029640917, "300–1000 Hz_40ms_modulation_mae_db": 0.6624028057852952, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5486524430501358, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17642239203726512, "1000–3000 Hz_10ms_envelope_correlation": 0.9755584582829139, "1000–3000 Hz_10ms_modulation_mae_db": 1.3037120975003325, "1000–3000 Hz_40ms_log_rms_mae_db": 1.094520072982186, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11679077208548048, "1000–3000 Hz_40ms_envelope_correlation": 0.9886553949858948, "1000–3000 Hz_40ms_modulation_mae_db": 1.1493611739994043, "3000–6000 Hz_10ms_log_rms_mae_db": 2.233813636425733, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3640975799054463, "3000–6000 Hz_10ms_envelope_correlation": 0.9035110927419918, "3000–6000 Hz_10ms_modulation_mae_db": 1.4246183633783265, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7635650979122244, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2977164042063138, "3000–6000 Hz_40ms_envelope_correlation": 0.9291717306912237, "3000–6000 Hz_40ms_modulation_mae_db": 1.4106805002681533, "mr_stft_db": 5.74328494563476, "band_mae_db": 0.1905308324938848, "rms_gain_error_db": -0.023448016127439842, "output_peak": 0.7561610329502063}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.074244871700482, "全帯域_10ms_rms_envelope_nrmse": 0.08605733448771441, "全帯域_10ms_envelope_correlation": 0.9921472268912045, "全帯域_10ms_modulation_mae_db": 0.254378726004761, "全帯域_40ms_log_rms_mae_db": 0.596331998861021, "全帯域_40ms_rms_envelope_nrmse": 0.043572071162648766, "全帯域_40ms_envelope_correlation": 0.9979118598616908, "全帯域_40ms_modulation_mae_db": 0.4219562678730742, "300–1000 Hz_10ms_log_rms_mae_db": 1.4319001109084077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12323765729484604, "300–1000 Hz_10ms_envelope_correlation": 0.9872854421408709, "300–1000 Hz_10ms_modulation_mae_db": 0.6127456029950793, "300–1000 Hz_40ms_log_rms_mae_db": 0.8927722047521497, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08257397171910141, "300–1000 Hz_40ms_envelope_correlation": 0.9942539945782761, "300–1000 Hz_40ms_modulation_mae_db": 0.6472090113536609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6084483500581248, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17825520031846656, "1000–3000 Hz_10ms_envelope_correlation": 0.9752499004519074, "1000–3000 Hz_10ms_modulation_mae_db": 1.1051827060640906, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1451245286416825, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11533161858544597, "1000–3000 Hz_40ms_envelope_correlation": 0.9895750682005672, "1000–3000 Hz_40ms_modulation_mae_db": 1.0618312507544536, "3000–6000 Hz_10ms_log_rms_mae_db": 2.264673816847324, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37901953475393757, "3000–6000 Hz_10ms_envelope_correlation": 0.9004533099719595, "3000–6000 Hz_10ms_modulation_mae_db": 2.0343318074333356, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8449366896196409, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31099263928334436, "3000–6000 Hz_40ms_envelope_correlation": 0.9267984201105753, "3000–6000 Hz_40ms_modulation_mae_db": 2.247558272153816, "mr_stft_db": 5.736829498326416, "band_mae_db": 0.26396496196506325, "rms_gain_error_db": -0.09923308635502515, "output_peak": 0.7660090997554181}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1012777449522833, "全帯域_10ms_rms_envelope_nrmse": 0.10123657192919874, "全帯域_10ms_envelope_correlation": 0.9890489463774496, "全帯域_10ms_modulation_mae_db": 0.3833849327975776, "全帯域_40ms_log_rms_mae_db": 0.5941024472941336, "全帯域_40ms_rms_envelope_nrmse": 0.0500061347704824, "全帯域_40ms_envelope_correlation": 0.9971064621635839, "全帯域_40ms_modulation_mae_db": 0.46535826013105047, "300–1000 Hz_10ms_log_rms_mae_db": 1.377768479227168, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13413092323453343, "300–1000 Hz_10ms_envelope_correlation": 0.9846105097632095, "300–1000 Hz_10ms_modulation_mae_db": 0.2459251377848212, "300–1000 Hz_40ms_log_rms_mae_db": 0.8241987070938982, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06805969902092221, "300–1000 Hz_40ms_envelope_correlation": 0.9960886465262992, "300–1000 Hz_40ms_modulation_mae_db": 0.14384180277016914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6275814836335543, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1800105691845456, "1000–3000 Hz_10ms_envelope_correlation": 0.9745442472339766, "1000–3000 Hz_10ms_modulation_mae_db": 2.5175909938877727, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1351434258120014, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10074668338824319, "1000–3000 Hz_40ms_envelope_correlation": 0.9916319138524451, "1000–3000 Hz_40ms_modulation_mae_db": 2.513823862484935, "3000–6000 Hz_10ms_log_rms_mae_db": 2.246337809796685, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3655753422768753, "3000–6000 Hz_10ms_envelope_correlation": 0.9035872243532628, "3000–6000 Hz_10ms_modulation_mae_db": 1.586287775743262, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7924357295529592, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2999678277816641, "3000–6000 Hz_40ms_envelope_correlation": 0.9287894706621131, "3000–6000 Hz_40ms_modulation_mae_db": 1.3304524858887874, "mr_stft_db": 5.757329732596403, "band_mae_db": 0.16400601395729023, "rms_gain_error_db": -0.042782875086596964, "output_peak": 0.9062011427185773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0996807828407784, "全帯域_10ms_rms_envelope_nrmse": 0.09615503912437186, "全帯域_10ms_envelope_correlation": 0.9901080326722168, "全帯域_10ms_modulation_mae_db": 0.5313135752337418, "全帯域_40ms_log_rms_mae_db": 0.5794399632175642, "全帯域_40ms_rms_envelope_nrmse": 0.04169634467965907, "全帯域_40ms_envelope_correlation": 0.9979926322639274, "全帯域_40ms_modulation_mae_db": 0.30934837040053964, "300–1000 Hz_10ms_log_rms_mae_db": 1.41157930786639, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13318434850466504, "300–1000 Hz_10ms_envelope_correlation": 0.9847398934809466, "300–1000 Hz_10ms_modulation_mae_db": 0.512643926527472, "300–1000 Hz_40ms_log_rms_mae_db": 0.8573867492751646, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07137563370167088, "300–1000 Hz_40ms_envelope_correlation": 0.9955257606450957, "300–1000 Hz_40ms_modulation_mae_db": 1.2065295786764298, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5706207257150129, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16714922127624177, "1000–3000 Hz_10ms_envelope_correlation": 0.9781415645703375, "1000–3000 Hz_10ms_modulation_mae_db": 1.4118429968404713, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1159667180600479, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10017016878680471, "1000–3000 Hz_40ms_envelope_correlation": 0.9919006920449382, "1000–3000 Hz_40ms_modulation_mae_db": 1.0437095808497692, "3000–6000 Hz_10ms_log_rms_mae_db": 2.317889546979261, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3655674981202447, "3000–6000 Hz_10ms_envelope_correlation": 0.9032282346293021, "3000–6000 Hz_10ms_modulation_mae_db": 1.4916820773852113, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8484576400637687, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3000928358998537, "3000–6000 Hz_40ms_envelope_correlation": 0.9284641218805075, "3000–6000 Hz_40ms_modulation_mae_db": 1.2840912825108792, "mr_stft_db": 5.762055782321515, "band_mae_db": 0.15738493421374297, "rms_gain_error_db": -0.04220991989395045, "output_peak": 0.7764842955442616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0596005922865668, "全帯域_10ms_rms_envelope_nrmse": 0.10001114820566354, "全帯域_10ms_envelope_correlation": 0.9893280492729681, "全帯域_10ms_modulation_mae_db": 0.8118538258734395, "全帯域_40ms_log_rms_mae_db": 0.5093150871444091, "全帯域_40ms_rms_envelope_nrmse": 0.05285348931489451, "全帯域_40ms_envelope_correlation": 0.9967385604695794, "全帯域_40ms_modulation_mae_db": 0.6020209217748879, "300–1000 Hz_10ms_log_rms_mae_db": 1.3655866101055272, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12938569817726375, "300–1000 Hz_10ms_envelope_correlation": 0.985546711185572, "300–1000 Hz_10ms_modulation_mae_db": 0.9821107015978982, "300–1000 Hz_40ms_log_rms_mae_db": 0.8058576072871161, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07310251182388201, "300–1000 Hz_40ms_envelope_correlation": 0.9951489575035857, "300–1000 Hz_40ms_modulation_mae_db": 0.6578579980442413, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6029481560288341, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1797363844685296, "1000–3000 Hz_10ms_envelope_correlation": 0.9747648834020015, "1000–3000 Hz_10ms_modulation_mae_db": 1.7167266300043806, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1489598519370687, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10060739786460675, "1000–3000 Hz_40ms_envelope_correlation": 0.9920808551282824, "1000–3000 Hz_40ms_modulation_mae_db": 1.9632837811742008, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2917927420371162, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3705744816382009, "3000–6000 Hz_10ms_envelope_correlation": 0.9066218129799928, "3000–6000 Hz_10ms_modulation_mae_db": 1.2236071237554678, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7703525706535566, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3025165011988333, "3000–6000 Hz_40ms_envelope_correlation": 0.932337652463518, "3000–6000 Hz_40ms_modulation_mae_db": 1.4986595434981547, "mr_stft_db": 5.753298109839303, "band_mae_db": 0.21389637178573354, "rms_gain_error_db": -0.008567324520307144, "output_peak": 0.7980410594490037}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1063550213789592, "全帯域_10ms_rms_envelope_nrmse": 0.10681494532063297, "全帯域_10ms_envelope_correlation": 0.9878414821468455, "全帯域_10ms_modulation_mae_db": 1.0127207604507318, "全帯域_40ms_log_rms_mae_db": 0.5507834010799414, "全帯域_40ms_rms_envelope_nrmse": 0.057456151010812063, "全帯域_40ms_envelope_correlation": 0.9961789827128139, "全帯域_40ms_modulation_mae_db": 0.9947405455372577, "300–1000 Hz_10ms_log_rms_mae_db": 1.4456241517710007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13734033719577662, "300–1000 Hz_10ms_envelope_correlation": 0.9837435021091337, "300–1000 Hz_10ms_modulation_mae_db": 1.0657756257846724, "300–1000 Hz_40ms_log_rms_mae_db": 0.9062333961666698, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06872071665326003, "300–1000 Hz_40ms_envelope_correlation": 0.9958327646752002, "300–1000 Hz_40ms_modulation_mae_db": 0.9429330767186, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5740937525568164, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1548369525221577, "1000–3000 Hz_10ms_envelope_correlation": 0.9815035714364907, "1000–3000 Hz_10ms_modulation_mae_db": 1.2158348713554494, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1284134472643732, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09534413959343267, "1000–3000 Hz_40ms_envelope_correlation": 0.9930953277498977, "1000–3000 Hz_40ms_modulation_mae_db": 1.4025171632288702, "3000–6000 Hz_10ms_log_rms_mae_db": 2.234094353634246, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35478713102352327, "3000–6000 Hz_10ms_envelope_correlation": 0.9125927547718321, "3000–6000 Hz_10ms_modulation_mae_db": 1.2946796525691506, "3000–6000 Hz_40ms_log_rms_mae_db": 1.763715620694708, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2937983849001057, "3000–6000 Hz_40ms_envelope_correlation": 0.9345892464866759, "3000–6000 Hz_40ms_modulation_mae_db": 1.4302431736782877, "mr_stft_db": 5.754758197776451, "band_mae_db": 0.24729327412676252, "rms_gain_error_db": -0.029133913848155592, "output_peak": 0.8247824033418368}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9712974398540615, "spectral_convergence": 0.3066359032869522, "spectral_convergence_status": "ok", "amplitude_floor": 2.5091999427961322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.491183732882799, "spectral_convergence": 0.289863796312718, "spectral_convergence_status": "ok", "amplitude_floor": 1.1904358876245626e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.794856794546647, "spectral_convergence": 0.3150523065210095, "spectral_convergence_status": "ok", "amplitude_floor": 8.234850781853941e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001792776939061309, "output_power": 0.0018907982972792773, "delta_db": 0.23118945411549016, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0038916933601920887, "output_power": 0.0038242439670518377, "delta_db": -0.07593023162186309, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002833146281043118, "output_power": 0.002567504592351914, "delta_db": -0.4275776952902646, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00044617096434630785, "output_power": 0.00042342296960093975, "delta_db": -0.22726890429718738, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.1707528478122786e-05, "output_power": 5.3927031787743594e-05, "delta_db": 0.18252737011744372, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2019368729027054e-05, "output_power": 3.400216886614856e-05, "delta_db": 0.2609385445392379, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23423869999691446, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12347810343270077, "pitch": null, "output_rms_dbfs": -20.558185253316715, "output_peak": 0.9730240897532715, "reference_rms_dbfs": -20.434707149884012, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.752445989094503}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003179454955178215, "log_rms_mae_db": 1.1029678183740816, "rms_envelope_nrmse": 0.11011490302020228, "envelope_correlation": 0.987076586156713, "rms_gain_error_db": -0.12347754370450661, "reference_mean_power": 0.009055057750136907, "output_mean_power": 0.008801232019288908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4928013782154505e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002203466496672066, "output_power": 0.0020501387329097554, "reference_active": true, "delta_db": -0.31323200761113623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002920043469136947, "output_power": 0.00030078342409404384, "reference_active": true, "delta_db": 0.12864582389972848}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.410659541472349e-05, "output_power": 7.848999340093145e-05, "reference_active": true, "delta_db": 0.249574310850868}], "active_band_count": 3, "active_band_mae_db": 0.2304840474539109}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002752533875730688, "log_rms_mae_db": 0.6025510297057284, "rms_envelope_nrmse": 0.05568887566121783, "envelope_correlation": 0.9965762710733819, "rms_gain_error_db": -0.12347484184250271, "reference_mean_power": 0.009100511091100074, "output_mean_power": 0.008845416743517643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.115339465742665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001970497325835552, "output_power": 0.0018512899234841885, "reference_active": true, "delta_db": -0.2710141253184713}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014191960918623396, "output_power": 0.00013751111858540318, "reference_active": true, "delta_db": -0.13704591685113332}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.976197244793298e-06, "output_power": 3.802562034197253e-06, "reference_active": true, "delta_db": -0.19391612269653602}], "active_band_count": 3, "active_band_mae_db": 0.20065872162204687}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028904791382782896, "log_rms_mae_db": 1.4108987218912057, "rms_envelope_nrmse": 0.14850734076644187, "envelope_correlation": 0.9818030537215683, "rms_gain_error_db": -0.3813455348722094, "reference_mean_power": 0.00500094100859156, "output_mean_power": 0.004580545253657622}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3504006393160654e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001701857204603173, "output_power": 0.0013846695877747705, "reference_active": true, "delta_db": -0.8957696385527864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00029604441935142547, "output_power": 0.00025005016232926524, "reference_active": true, "delta_db": -0.7332973791224255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.422561443844994e-05, "output_power": 7.974980558148791e-05, "reference_active": true, "delta_db": 0.3117583161752183}], "active_band_count": 3, "active_band_mae_db": 0.6469417779501434}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002500254339817177, "log_rms_mae_db": 0.8961694634571479, "rms_envelope_nrmse": 0.09276786347914669, "envelope_correlation": 0.9934365311744063, "rms_gain_error_db": -0.3813189143794213, "reference_mean_power": 0.00502600851498449, "output_mean_power": 0.004603533719695993}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0757929049004627e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015562264230646289, "output_power": 0.0012744610132500687, "reference_active": true, "delta_db": -0.8674623029516503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001439985744956168, "output_power": 0.00011550239250838725, "reference_active": true, "delta_db": -0.957672125804636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7762361828128385e-06, "output_power": 4.616749665641213e-06, "reference_active": true, "delta_db": -0.14749467918818443}], "active_band_count": 3, "active_band_mae_db": 0.6575430359814902}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010495143398294122, "log_rms_mae_db": 1.5350435141191443, "rms_envelope_nrmse": 0.15893260711832294, "envelope_correlation": 0.9801938809310399, "rms_gain_error_db": -0.1400149865726305, "reference_mean_power": 0.0004146271168721916, "output_mean_power": 0.0004014728704734877}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0723455289073609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010117430747443104, "output_power": 0.0001015289282803103, "reference_active": true, "delta_db": 0.015195614979635473}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2721317302684375e-05, "output_power": 1.676880074703818e-05, "reference_active": true, "delta_db": 1.1996991912801305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6903995233241656e-06, "output_power": 5.313036824592913e-06, "reference_active": true, "delta_db": 2.9552604865011634}], "active_band_count": 3, "active_band_mae_db": 1.3900517642536432}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.780703381807892e-05, "log_rms_mae_db": 1.0409772535227155, "rms_envelope_nrmse": 0.0934449171757039, "envelope_correlation": 0.992839338204695, "rms_gain_error_db": -0.14005561414671938, "reference_mean_power": 0.0004167096352122121, "output_mean_power": 0.0004034855453467914}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9169450940862202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.13104680727751e-05, "output_power": 9.432139548624185e-05, "reference_active": true, "delta_db": 0.1408964847415632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.750575155696485e-06, "output_power": 8.172036886010743e-06, "reference_active": true, "delta_db": 1.52619034508886}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0115145688994735e-07, "output_power": 3.746331902859549e-07, "reference_active": true, "delta_db": 2.7008306826185087}], "active_band_count": 3, "active_band_mae_db": 1.4559725041496439}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4613391249073974e-05, "log_rms_mae_db": 2.304759785527561, "rms_envelope_nrmse": 0.36599015203741275, "envelope_correlation": 0.9048159407310922, "rms_gain_error_db": 0.3418022905835514, "reference_mean_power": 2.714804754964509e-05, "output_mean_power": 2.937100655744314e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3627974020623764e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.8436688435136e-06, "output_power": 4.267967024815658e-06, "reference_active": true, "delta_db": -0.549533879166469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.21772084491833e-06, "output_power": 2.0728816098744147e-06, "reference_active": true, "delta_db": -0.2933237994105381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.650596314175208e-07, "output_power": 1.1410426558804226e-06, "reference_active": true, "delta_db": 3.897932369868315}], "active_band_count": 3, "active_band_mae_db": 1.580263349481774}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2268327401512992e-05, "log_rms_mae_db": 1.8268334503849872, "rms_envelope_nrmse": 0.2976237639499564, "envelope_correlation": 0.9310856657643958, "rms_gain_error_db": 0.34178563319373123, "reference_mean_power": 2.7284479980131438e-05, "output_mean_power": 2.951849724308694e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2818152654967464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7409144698248785e-06, "output_power": 4.386619951763432e-06, "reference_active": true, "delta_db": -0.337321107395964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1748833712187593e-06, "output_power": 1.1880005126542513e-06, "reference_active": true, "delta_db": 0.04821871021769368}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.424838593381526e-08, "output_power": 7.665578239909344e-08, "reference_active": true, "delta_db": 3.4990481191889025}], "active_band_count": 3, "active_band_mae_db": 1.2948626456008534}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8607439070805702, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8e96a9c9117bbfe5f27b1409a0279527cbb8bd77f2e7ff8f68cd965fc2f9c5d7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9763211655861566, "standard_deviation": 0.003485953841722537, "minimum": 0.9706128183805854, "maximum": 0.9815035714364907, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5807028177309959, "standard_deviation": 0.030997005125266607, "minimum": 1.5350435141191443, "maximum": 1.6275814836335543, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6819169397399252, "standard_deviation": 0.6311198473205442, "minimum": 1.1051827060640906, "maximum": 2.7943934580132628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17364987342529195, "standard_deviation": 0.012691145347726874, "minimum": 0.1548369525221577, "maximum": 0.19385566047680639, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9908093221499358, "standard_deviation": 0.002265065091561817, "minimum": 0.986695987032766, "maximum": 0.9930953277498977, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1125890019901559, "standard_deviation": 0.035950180322251266, "minimum": 1.0409772535227155, "maximum": 1.1489598519370687, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.7011740130973516, "standard_deviation": 0.7329314749797735, "minimum": 1.0437095808497692, "maximum": 3.0188927881375363, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10607373909014331, "standard_deviation": 0.011780799456943733, "minimum": 0.0934449171757039, "maximum": 0.12615421524142872, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9047587467336848, "standard_deviation": 0.00360360448480238, "minimum": 0.9004533099719595, "maximum": 0.9125927547718321, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.271678412731314, "standard_deviation": 0.03216540019868793, "minimum": 2.233813636425733, "maximum": 2.317889546979261, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5217471884657043, "standard_deviation": 0.24535647739599328, "minimum": 1.2236071237554678, "maximum": 2.0343318074333356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3664561984239796, "standard_deviation": 0.006745218278250675, "minimum": 0.35478713102352327, "maximum": 0.37901953475393757, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9301981997063151, "standard_deviation": 0.0024574277332714035, "minimum": 0.9267984201105753, "maximum": 0.9345892464866759, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8026276417864886, "standard_deviation": 0.03530986284549249, "minimum": 1.7635650979122244, "maximum": 1.8484576400637687, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.507794346639623, "standard_deviation": 0.3146376492544319, "minimum": 1.2840912825108792, "maximum": 2.247558272153816, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.29989226766322097, "standard_deviation": 0.005198289268578637, "minimum": 0.2937983849001057, "maximum": 0.31099263928334436, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.984508542562021, "standard_deviation": 0.0015665680340074312, "minimum": 0.9818030537215683, "maximum": 0.9872854421408709, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3923368198049295, "standard_deviation": 0.03822010861899801, "minimum": 1.3379857803330102, "maximum": 1.4456241517710007, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7441657987478478, "standard_deviation": 0.29205799280915773, "minimum": 0.2459251377848212, "maximum": 1.0657756257846724, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13457962664193696, "standard_deviation": 0.007226247851519458, "minimum": 0.12323765729484604, "maximum": 0.14850734076644187, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9951276587349214, "standard_deviation": 0.0009043411196291499, "minimum": 0.9934365311744063, "maximum": 0.9960886465262992, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8546763080053208, "standard_deviation": 0.04024009428222745, "minimum": 0.8058576072871161, "maximum": 0.9062333961666698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7256929060928285, "standard_deviation": 0.30762968526707357, "minimum": 0.14384180277016914, "maximum": 1.2065295786764298, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07496553025068381, "standard_deviation": 0.008545556093342396, "minimum": 0.06805969902092221, "maximum": 0.09276786347914669, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19812633968821936, "standard_deviation": 0.051151321021030975, "minimum": 0.11369562896636297, "maximum": 0.26396496196506325, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.750663666086063, "standard_deviation": 0.008242933904946816, "minimum": 5.736829498326416, "maximum": 5.762055782321515, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8323974322415015, "standard_deviation": 0.07605165609263938, "minimum": 0.7561610329502063, "maximum": 0.9730240897532715, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03672296924246737, "standard_deviation": 0.05979119441365522, "minimum": -0.12347810343270077, "maximum": 0.07506948532443694, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9890261782665649, "standard_deviation": 0.0016052920608812574, "minimum": 0.987076586156713, "maximum": 0.9921472268912045, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1050814777521605, "standard_deviation": 0.03134527663654144, "minimum": 1.0596005922865668, "maximum": 1.1545613624190516, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.660667667459913, "standard_deviation": 0.3535925196655297, "minimum": 0.2304840474539109, "maximum": 1.1045767117762677, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10133522579234633, "standard_deviation": 0.00777365306676653, "minimum": 0.08605733448771441, "maximum": 0.11011490302020228, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969630065766547, "standard_deviation": 0.0007900035118592791, "minimum": 0.9957906283388382, "maximum": 0.9979926322639274, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5894043437138677, "standard_deviation": 0.047213334976620364, "minimum": 0.5093150871444091, "maximum": 0.6726428777717276, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5760435568543213, "standard_deviation": 0.2783126424522286, "minimum": 0.20065872162204687, "maximum": 0.9947405455372577, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05123120296282844, "standard_deviation": 0.006871253612890044, "minimum": 0.04169634467965907, "maximum": 0.06134087415863202, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0033", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5940372943878174, "predicted_mos": 2.5940372943878174, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "34bab7f2f3d3b45d4f5ddebcfbe95ca848ce31b9fbfc26f9fb6e6402ac65ee4e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6601690999232233, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1332556613708753, "全帯域_10ms_rms_envelope_nrmse": 0.1135511921575956, "全帯域_10ms_envelope_correlation": 0.9862494798593358, "全帯域_10ms_modulation_mae_db": 0.3000035405639196, "全帯域_40ms_log_rms_mae_db": 0.636167979321465, "全帯域_40ms_rms_envelope_nrmse": 0.059929327153292765, "全帯域_40ms_envelope_correlation": 0.9960073947784951, "全帯域_40ms_modulation_mae_db": 0.20582998824853296, "300–1000 Hz_10ms_log_rms_mae_db": 1.465135892291334, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1541409444533549, "300–1000 Hz_10ms_envelope_correlation": 0.9801265597742488, "300–1000 Hz_10ms_modulation_mae_db": 0.6242723751164377, "300–1000 Hz_40ms_log_rms_mae_db": 0.9032453966157822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09527014446212408, "300–1000 Hz_40ms_envelope_correlation": 0.9928054735223888, "300–1000 Hz_40ms_modulation_mae_db": 0.6154752262733113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4994080894008754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15710021234327382, "1000–3000 Hz_10ms_envelope_correlation": 0.9809069437636081, "1000–3000 Hz_10ms_modulation_mae_db": 1.0881342525125068, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9589430087041324, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09138175698400515, "1000–3000 Hz_40ms_envelope_correlation": 0.9936579819216742, "1000–3000 Hz_40ms_modulation_mae_db": 1.1575746811434158, "3000–6000 Hz_10ms_log_rms_mae_db": 2.140760485252674, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3542327492607513, "3000–6000 Hz_10ms_envelope_correlation": 0.909869417132063, "3000–6000 Hz_10ms_modulation_mae_db": 1.385211637604118, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6709953555611852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29162109911968104, "3000–6000 Hz_40ms_envelope_correlation": 0.9327526107654707, "3000–6000 Hz_40ms_modulation_mae_db": 1.152117585886711, "mr_stft_db": 5.6464352073379445, "band_mae_db": 0.21022020314409215, "rms_gain_error_db": -0.1281011139704966, "output_peak": 0.8166118997930552}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1232601649078984, "全帯域_10ms_rms_envelope_nrmse": 0.11206446461270589, "全帯域_10ms_envelope_correlation": 0.9869120464020378, "全帯域_10ms_modulation_mae_db": 1.1861696744674959, "全帯域_40ms_log_rms_mae_db": 0.641940022313381, "全帯域_40ms_rms_envelope_nrmse": 0.06200077215880357, "全帯域_40ms_envelope_correlation": 0.9957081254957693, "全帯域_40ms_modulation_mae_db": 1.053096100673332, "300–1000 Hz_10ms_log_rms_mae_db": 1.3505296345329125, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1372887791840886, "300–1000 Hz_10ms_envelope_correlation": 0.983697000355896, "300–1000 Hz_10ms_modulation_mae_db": 1.074361425399402, "300–1000 Hz_40ms_log_rms_mae_db": 0.8016968924946926, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07845323504313277, "300–1000 Hz_40ms_envelope_correlation": 0.9942257469594568, "300–1000 Hz_40ms_modulation_mae_db": 0.9456830169206846, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5327200758197295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17601025487016858, "1000–3000 Hz_10ms_envelope_correlation": 0.9756921582013383, "1000–3000 Hz_10ms_modulation_mae_db": 2.1281347351961446, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9931609330696904, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10643747919408507, "1000–3000 Hz_40ms_envelope_correlation": 0.9906020880318415, "1000–3000 Hz_40ms_modulation_mae_db": 2.238561767949749, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1293957908938284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3620995345184136, "3000–6000 Hz_10ms_envelope_correlation": 0.9044152364696209, "3000–6000 Hz_10ms_modulation_mae_db": 1.4687880443479397, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7065862715002782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29357599832527953, "3000–6000 Hz_40ms_envelope_correlation": 0.9306936109277393, "3000–6000 Hz_40ms_modulation_mae_db": 1.4390502447683922, "mr_stft_db": 5.63741997712134, "band_mae_db": 0.1494922823063507, "rms_gain_error_db": 0.08051415399793221, "output_peak": 0.8644714210702228}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1326685175407913, "全帯域_10ms_rms_envelope_nrmse": 0.10119924528057107, "全帯域_10ms_envelope_correlation": 0.989112026391878, "全帯域_10ms_modulation_mae_db": 1.0198861666466759, "全帯域_40ms_log_rms_mae_db": 0.6081360034302198, "全帯域_40ms_rms_envelope_nrmse": 0.04622616844804342, "全帯域_40ms_envelope_correlation": 0.9975246291936919, "全帯域_40ms_modulation_mae_db": 0.8000374687854239, "300–1000 Hz_10ms_log_rms_mae_db": 1.3630353157378585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13410018083247935, "300–1000 Hz_10ms_envelope_correlation": 0.9844456168712624, "300–1000 Hz_10ms_modulation_mae_db": 0.7198323041295135, "300–1000 Hz_40ms_log_rms_mae_db": 0.8071875795015081, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062410385244527224, "300–1000 Hz_40ms_envelope_correlation": 0.9964671010190653, "300–1000 Hz_40ms_modulation_mae_db": 0.5355603551172502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5311229960914494, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17176789895466313, "1000–3000 Hz_10ms_envelope_correlation": 0.9769677733466806, "1000–3000 Hz_10ms_modulation_mae_db": 0.834421017124324, "1000–3000 Hz_40ms_log_rms_mae_db": 1.030314979628695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11015774070513537, "1000–3000 Hz_40ms_envelope_correlation": 0.9902650494967272, "1000–3000 Hz_40ms_modulation_mae_db": 0.4419324928637109, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1216930549735658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3558235077436865, "3000–6000 Hz_10ms_envelope_correlation": 0.9066817383303472, "3000–6000 Hz_10ms_modulation_mae_db": 1.2314338548287258, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6389366039128443, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29316031049170793, "3000–6000 Hz_40ms_envelope_correlation": 0.9301222536714323, "3000–6000 Hz_40ms_modulation_mae_db": 1.1173360961182486, "mr_stft_db": 5.636585461612967, "band_mae_db": 0.17444193212565237, "rms_gain_error_db": 0.0018509462653614592, "output_peak": 0.8024858006763529}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0838441133697676, "全帯域_10ms_rms_envelope_nrmse": 0.09114210108727908, "全帯域_10ms_envelope_correlation": 0.9911570446038196, "全帯域_10ms_modulation_mae_db": 0.3128019401896083, "全帯域_40ms_log_rms_mae_db": 0.5975256889905884, "全帯域_40ms_rms_envelope_nrmse": 0.046571706194652854, "全帯域_40ms_envelope_correlation": 0.9975647534348283, "全帯域_40ms_modulation_mae_db": 0.47493905296090455, "300–1000 Hz_10ms_log_rms_mae_db": 1.4514670528725486, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12477198207301687, "300–1000 Hz_10ms_envelope_correlation": 0.9868663097975795, "300–1000 Hz_10ms_modulation_mae_db": 0.557171862312226, "300–1000 Hz_40ms_log_rms_mae_db": 0.830316666882049, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08215812322837099, "300–1000 Hz_40ms_envelope_correlation": 0.9942157100446175, "300–1000 Hz_40ms_modulation_mae_db": 0.6527434818453169, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5527302608342473, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17203030512356404, "1000–3000 Hz_10ms_envelope_correlation": 0.9772363000847172, "1000–3000 Hz_10ms_modulation_mae_db": 1.0498784575638582, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0269555500380225, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10598645002841796, "1000–3000 Hz_40ms_envelope_correlation": 0.9916705527626227, "1000–3000 Hz_40ms_modulation_mae_db": 0.9561599381356863, "3000–6000 Hz_10ms_log_rms_mae_db": 2.137119391661223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36906770317069876, "3000–6000 Hz_10ms_envelope_correlation": 0.9043045033777835, "3000–6000 Hz_10ms_modulation_mae_db": 1.9277612211215542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7416542710403078, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30546961627117264, "3000–6000 Hz_40ms_envelope_correlation": 0.9279707962075332, "3000–6000 Hz_40ms_modulation_mae_db": 2.1363261796349464, "mr_stft_db": 5.62586733232897, "band_mae_db": 0.2582980274346, "rms_gain_error_db": -0.08858359651613196, "output_peak": 0.7728337601002891}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.083807722995954, "全帯域_10ms_rms_envelope_nrmse": 0.09472079385252974, "全帯域_10ms_envelope_correlation": 0.9904126311140341, "全帯域_10ms_modulation_mae_db": 0.22698699784223872, "全帯域_40ms_log_rms_mae_db": 0.5733632202143571, "全帯域_40ms_rms_envelope_nrmse": 0.04317577298780158, "全帯域_40ms_envelope_correlation": 0.997847577870662, "全帯域_40ms_modulation_mae_db": 0.22451767431434608, "300–1000 Hz_10ms_log_rms_mae_db": 1.3903195362715777, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12884690806898666, "300–1000 Hz_10ms_envelope_correlation": 0.9857983236151366, "300–1000 Hz_10ms_modulation_mae_db": 0.14484964818610405, "300–1000 Hz_40ms_log_rms_mae_db": 0.7926192986137582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06657779842660844, "300–1000 Hz_40ms_envelope_correlation": 0.9962302250608858, "300–1000 Hz_40ms_modulation_mae_db": 0.18972589706929635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.562569439504965, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1826059236403557, "1000–3000 Hz_10ms_envelope_correlation": 0.9739142274664178, "1000–3000 Hz_10ms_modulation_mae_db": 2.514927009794633, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0224549063609607, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0985022329483695, "1000–3000 Hz_40ms_envelope_correlation": 0.9919753308141016, "1000–3000 Hz_40ms_modulation_mae_db": 2.422780864036745, "3000–6000 Hz_10ms_log_rms_mae_db": 2.105066244577498, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35416481082515455, "3000–6000 Hz_10ms_envelope_correlation": 0.9086470095394727, "3000–6000 Hz_10ms_modulation_mae_db": 1.2436899886694188, "3000–6000 Hz_40ms_log_rms_mae_db": 1.671913199819973, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2928168314172417, "3000–6000 Hz_40ms_envelope_correlation": 0.9311639648936808, "3000–6000 Hz_40ms_modulation_mae_db": 1.1224937364827028, "mr_stft_db": 5.64938123163405, "band_mae_db": 0.12662207550578963, "rms_gain_error_db": -0.043942998790682045, "output_peak": 0.8296419750743593}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0902165615648198, "全帯域_10ms_rms_envelope_nrmse": 0.09396972022130055, "全帯域_10ms_envelope_correlation": 0.9905605108031352, "全帯域_10ms_modulation_mae_db": 0.393975194320551, "全帯域_40ms_log_rms_mae_db": 0.5655377304040645, "全帯域_40ms_rms_envelope_nrmse": 0.042659148654430566, "全帯域_40ms_envelope_correlation": 0.9978786492247913, "全帯域_40ms_modulation_mae_db": 0.5517972057518487, "300–1000 Hz_10ms_log_rms_mae_db": 1.4137101160662973, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13399506910487397, "300–1000 Hz_10ms_envelope_correlation": 0.9844817982980553, "300–1000 Hz_10ms_modulation_mae_db": 0.6975632517980638, "300–1000 Hz_40ms_log_rms_mae_db": 0.8378908209791509, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07271788813545063, "300–1000 Hz_40ms_envelope_correlation": 0.9952103428913459, "300–1000 Hz_40ms_modulation_mae_db": 1.4219104537823553, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5182999605111613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15392672195760343, "1000–3000 Hz_10ms_envelope_correlation": 0.981609100021775, "1000–3000 Hz_10ms_modulation_mae_db": 0.9349867717186373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9997884517197368, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09176376067932798, "1000–3000 Hz_40ms_envelope_correlation": 0.9933931529726587, "1000–3000 Hz_40ms_modulation_mae_db": 0.5546823425801773, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1410856379194736, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3585268969373585, "3000–6000 Hz_10ms_envelope_correlation": 0.906065202301814, "3000–6000 Hz_10ms_modulation_mae_db": 1.4308885655236754, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6807212819301398, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29513754144862925, "3000–6000 Hz_40ms_envelope_correlation": 0.9297982026083303, "3000–6000 Hz_40ms_modulation_mae_db": 1.2946365114058733, "mr_stft_db": 5.647570645926614, "band_mae_db": 0.17456824003076457, "rms_gain_error_db": -0.021870621024998205, "output_peak": 0.8996233121509668}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0825165178208978, "全帯域_10ms_rms_envelope_nrmse": 0.10067872666749685, "全帯域_10ms_envelope_correlation": 0.9891579805170383, "全帯域_10ms_modulation_mae_db": 0.8146857445177532, "全帯域_40ms_log_rms_mae_db": 0.5370160466584685, "全帯域_40ms_rms_envelope_nrmse": 0.054187372908358254, "全帯域_40ms_envelope_correlation": 0.9965737708783277, "全帯域_40ms_modulation_mae_db": 0.6037817990095392, "300–1000 Hz_10ms_log_rms_mae_db": 1.3839194300684663, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13197438205459616, "300–1000 Hz_10ms_envelope_correlation": 0.9850067057524495, "300–1000 Hz_10ms_modulation_mae_db": 0.9234095275469393, "300–1000 Hz_40ms_log_rms_mae_db": 0.8111071995152193, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07702905059029352, "300–1000 Hz_40ms_envelope_correlation": 0.9946874292951803, "300–1000 Hz_40ms_modulation_mae_db": 0.651960234604681, "1000–3000 Hz_10ms_log_rms_mae_db": 1.577315187940081, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1772933886520885, "1000–3000 Hz_10ms_envelope_correlation": 0.9758895254661997, "1000–3000 Hz_10ms_modulation_mae_db": 1.642255267277423, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0289899310471304, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09728915942342102, "1000–3000 Hz_40ms_envelope_correlation": 0.9933287155650141, "1000–3000 Hz_40ms_modulation_mae_db": 1.4835321297875597, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1711982464227875, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3702544221908248, "3000–6000 Hz_10ms_envelope_correlation": 0.9064024589460953, "3000–6000 Hz_10ms_modulation_mae_db": 1.2324298669625797, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7043289176546086, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30351481264890695, "3000–6000 Hz_40ms_envelope_correlation": 0.9311864427431856, "3000–6000 Hz_40ms_modulation_mae_db": 1.5686656499176257, "mr_stft_db": 5.642022573500796, "band_mae_db": 0.2648986958323414, "rms_gain_error_db": -0.02979956579528567, "output_peak": 0.8093175833766265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1345489391353532, "全帯域_10ms_rms_envelope_nrmse": 0.11220200140312227, "全帯域_10ms_envelope_correlation": 0.9866072193242509, "全帯域_10ms_modulation_mae_db": 1.1853487124646425, "全帯域_40ms_log_rms_mae_db": 0.5720342406720945, "全帯域_40ms_rms_envelope_nrmse": 0.05829978650867613, "全帯域_40ms_envelope_correlation": 0.9960653909030994, "全帯域_40ms_modulation_mae_db": 1.125320192131804, "300–1000 Hz_10ms_log_rms_mae_db": 1.4574904353120992, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14375339576667917, "300–1000 Hz_10ms_envelope_correlation": 0.98214625125732, "300–1000 Hz_10ms_modulation_mae_db": 1.2014470604098164, "300–1000 Hz_40ms_log_rms_mae_db": 0.894647550758053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06965691375420091, "300–1000 Hz_40ms_envelope_correlation": 0.9956820030999944, "300–1000 Hz_40ms_modulation_mae_db": 1.050942849258224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5271973012455111, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1469381328715516, "1000–3000 Hz_10ms_envelope_correlation": 0.9835348742037432, "1000–3000 Hz_10ms_modulation_mae_db": 0.9071184296755627, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0120188446724745, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08701303180604435, "1000–3000 Hz_40ms_envelope_correlation": 0.9945228066532051, "1000–3000 Hz_40ms_modulation_mae_db": 1.0422852776952254, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1071598988569202, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3447077850034388, "3000–6000 Hz_10ms_envelope_correlation": 0.9160694414746178, "3000–6000 Hz_10ms_modulation_mae_db": 1.198247662733853, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6910851299524003, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.28926757421601107, "3000–6000 Hz_40ms_envelope_correlation": 0.9350573560485115, "3000–6000 Hz_40ms_modulation_mae_db": 1.366719247162016, "mr_stft_db": 5.646681267165043, "band_mae_db": 0.19862511723891063, "rms_gain_error_db": -0.01948223383003715, "output_peak": 0.8210166349552753}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.852327116004161, "spectral_convergence": 0.30752552474137373, "spectral_convergence_status": "ok", "amplitude_floor": 2.5091999427961322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.389437862758481, "spectral_convergence": 0.29020775676491045, "spectral_convergence_status": "ok", "amplitude_floor": 1.1904358876245626e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.69754064325119, "spectral_convergence": 0.3152421699719848, "spectral_convergence_status": "ok", "amplitude_floor": 8.234850781853941e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001792776939061309, "output_power": 0.0018753552125551978, "delta_db": 0.1955728280176647, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0038916933601920887, "output_power": 0.003832539988547221, "delta_db": -0.06651918440422623, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002833146281043118, "output_power": 0.0025824200140605142, "delta_db": -0.4024211970868011, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00044617096434630785, "output_power": 0.00040999712187855786, "delta_db": -0.3672049605953084, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.1707528478122786e-05, "output_power": 5.096059783734106e-05, "delta_db": -0.06319265590915504, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2019368729027054e-05, "output_power": 3.327007690161912e-05, "delta_db": 0.16641039285139744, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21022020314409215, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1281011139704966, "pitch": null, "output_rms_dbfs": -20.562808263854507, "output_peak": 0.8166118997930552, "reference_rms_dbfs": -20.434707149884012, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6464352073379445}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003179454955178215, "log_rms_mae_db": 1.1332556613708753, "rms_envelope_nrmse": 0.1135511921575956, "envelope_correlation": 0.9862494798593358, "rms_gain_error_db": -0.12810072125793162, "reference_mean_power": 0.009055057750136907, "output_mean_power": 0.008791867864299006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4928013782154505e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002203466496672066, "output_power": 0.002079693016034523, "reference_active": true, "delta_db": -0.25107218138378473}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002920043469136947, "output_power": 0.00030278039230912144, "reference_active": true, "delta_db": 0.15738430746939866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.410659541472349e-05, "output_power": 8.298742224984336e-05, "reference_active": true, "delta_db": 0.4915541328385754}], "active_band_count": 3, "active_band_mae_db": 0.3000035405639196}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002752533875730688, "log_rms_mae_db": 0.636167979321465, "rms_envelope_nrmse": 0.059929327153292765, "envelope_correlation": 0.9960073947784951, "rms_gain_error_db": -0.12811771857486032, "reference_mean_power": 0.009100511091100074, "output_mean_power": 0.00883596549858938}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.115339465742665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001970497325835552, "output_power": 0.0018824471817630467, "reference_active": true, "delta_db": -0.19853050334489758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014191960918623396, "output_power": 0.00013497331886706235, "reference_active": true, "delta_db": -0.2179447963310136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.976197244793298e-06, "output_power": 4.164562522725382e-06, "reference_active": true, "delta_db": 0.20101466506968768}], "active_band_count": 3, "active_band_mae_db": 0.20582998824853296}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028904791382782896, "log_rms_mae_db": 1.465135892291334, "rms_envelope_nrmse": 0.1541409444533549, "envelope_correlation": 0.9801265597742488, "rms_gain_error_db": -0.36064040626870186, "reference_mean_power": 0.00500094100859156, "output_mean_power": 0.004602435289208413}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3504006393160654e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001701857204603173, "output_power": 0.0014117612506930111, "reference_active": true, "delta_db": -0.8116186009765043}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00029604441935142547, "output_power": 0.0002538062818146801, "reference_active": true, "delta_db": -0.6685451186726659}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.422561443844994e-05, "output_power": 8.12492090615958e-05, "reference_active": true, "delta_db": 0.392653405700143}], "active_band_count": 3, "active_band_mae_db": 0.6242723751164377}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002500254339817177, "log_rms_mae_db": 0.9032453966157822, "rms_envelope_nrmse": 0.09527014446212408, "envelope_correlation": 0.9928054735223888, "rms_gain_error_db": -0.3606424415884246, "reference_mean_power": 0.00502600851498449, "output_mean_power": 0.004625503094865756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0757929049004627e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015562264230646289, "output_power": 0.0013005860783961722, "reference_active": true, "delta_db": -0.7793368393386415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001439985744956168, "output_power": 0.00011382175861681747, "reference_active": true, "delta_db": -1.0213290140352524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7762361828128385e-06, "output_power": 4.826827483182511e-06, "reference_active": true, "delta_db": 0.04575982544604018}], "active_band_count": 3, "active_band_mae_db": 0.6154752262733113}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010495143398294122, "log_rms_mae_db": 1.4994080894008754, "rms_envelope_nrmse": 0.15710021234327382, "envelope_correlation": 0.9809069437636081, "rms_gain_error_db": -0.2701512460867917, "reference_mean_power": 0.0004146271168721916, "output_mean_power": 0.0003896211978503736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0723455289073609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010117430747443104, "output_power": 9.789136730539938e-05, "reference_active": true, "delta_db": -0.14325845789631736}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2721317302684375e-05, "output_power": 1.575706277721084e-05, "reference_active": true, "delta_db": 0.9294318009155838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6903995233241656e-06, "output_power": 4.4564396257440976e-06, "reference_active": true, "delta_db": 2.191712498725619}], "active_band_count": 3, "active_band_mae_db": 1.0881342525125068}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.780703381807892e-05, "log_rms_mae_db": 0.9589430087041324, "rms_envelope_nrmse": 0.09138175698400515, "envelope_correlation": 0.9936579819216742, "rms_gain_error_db": -0.27018676750948717, "reference_mean_power": 0.0004167096352122121, "output_mean_power": 0.0003915749179794147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9169450940862202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.13104680727751e-05, "output_power": 9.03586825359141e-05, "reference_active": true, "delta_db": -0.045506789376245}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.750575155696485e-06, "output_power": 7.416041640996126e-06, "reference_active": true, "delta_db": 1.1046087597726233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0115145688994735e-07, "output_power": 3.4338714413075105e-07, "reference_active": true, "delta_db": 2.322608494281379}], "active_band_count": 3, "active_band_mae_db": 1.1575746811434158}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4613391249073974e-05, "log_rms_mae_db": 2.140760485252674, "rms_envelope_nrmse": 0.3542327492607513, "envelope_correlation": 0.909869417132063, "rms_gain_error_db": 0.23673127194092403, "reference_mean_power": 2.714804754964509e-05, "output_mean_power": 2.866894611657915e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3627974020623764e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.8436688435136e-06, "output_power": 4.533649495312131e-06, "reference_active": true, "delta_db": -0.28726502242137986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.21772084491833e-06, "output_power": 2.001579643255136e-06, "reference_active": true, "delta_db": -0.4453400308984615}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.650596314175208e-07, "output_power": 1.0228492452108864e-06, "reference_active": true, "delta_db": 3.4230298594925124}], "active_band_count": 3, "active_band_mae_db": 1.385211637604118}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2268327401512992e-05, "log_rms_mae_db": 1.6709953555611852, "rms_envelope_nrmse": 0.29162109911968104, "envelope_correlation": 0.9327526107654707, "rms_gain_error_db": 0.2367180615837466, "reference_mean_power": 2.7284479980131438e-05, "output_mean_power": 2.881293417495883e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2818152654967464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7409144698248785e-06, "output_power": 4.70349383966956e-06, "reference_active": true, "delta_db": -0.03441540792953439}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1748833712187593e-06, "output_power": 1.1284685100566254e-06, "reference_active": true, "delta_db": -0.17505312449265448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.424838593381526e-08, "output_power": 7.233167650497332e-08, "reference_active": true, "delta_db": 3.2468842252379444}], "active_band_count": 3, "active_band_mae_db": 1.152117585886711}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8574521716128022, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "486f449599af5a2212db7c527ffa5bf93813f2a79c22f240d9df3b5ac9db3fc4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.97821886281931, "standard_deviation": 0.003377447327280491, "minimum": 0.9739142274664178, "maximum": 0.9835348742037432, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5376704139185025, "standard_deviation": 0.025170132898101216, "minimum": 1.4994080894008754, "maximum": 1.577315187940081, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3874819926078863, "standard_deviation": 0.6357864544760384, "minimum": 0.834421017124324, "maximum": 2.514927009794633, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1672091048016586, "standard_deviation": 0.01281609108042345, "minimum": 0.1469381328715516, "maximum": 0.1826059236403557, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9924269597772306, "standard_deviation": 0.0015322632970071338, "minimum": 0.9902650494967272, "maximum": 0.9945228066532051, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0090783256551052, "standard_deviation": 0.024517799267990765, "minimum": 0.9589430087041324, "maximum": 1.030314979628695, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2871886867740336, "standard_deviation": 0.7241361514194062, "minimum": 0.4419324928637109, "maximum": 2.422780864036745, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0985664514711008, "standard_deviation": 0.008318490453249621, "minimum": 0.08701303180604435, "maximum": 0.11015774070513537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9078068759464768, "standard_deviation": 0.0038409721705532206, "minimum": 0.9043045033777835, "maximum": 0.9160694414746178, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.131684843819746, "standard_deviation": 0.0213028631587709, "minimum": 2.105066244577498, "maximum": 2.1711982464227875, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3898063552239832, "standard_deviation": 0.24060528956762212, "minimum": 1.198247662733853, "maximum": 1.9277612211215542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3586096762062908, "standard_deviation": 0.008426963797745725, "minimum": 0.3447077850034388, "maximum": 0.3702544221908248, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9310931547332355, "standard_deviation": 0.0021047806474012378, "minimum": 0.9279707962075332, "maximum": 0.9350573560485115, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6882776289214672, "standard_deviation": 0.030490993728985036, "minimum": 1.6389366039128443, "maximum": 1.7416542710403078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3996681564220643, "standard_deviation": 0.3386341806743553, "minimum": 1.1173360961182486, "maximum": 2.1363261796349464, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.29557047299232875, "standard_deviation": 0.0057815771615796815, "minimum": 0.28926757421601107, "maximum": 0.30546961627117264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9840710707152436, "standard_deviation": 0.002117450763256088, "minimum": 0.9801265597742488, "maximum": 0.9868663097975795, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4094509266441366, "standard_deviation": 0.04445880977264944, "minimum": 1.3505296345329125, "maximum": 1.465135892291334, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7428634318623129, "standard_deviation": 0.329834369500205, "minimum": 0.14484964818610405, "maximum": 1.2014470604098164, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13610895519225946, "standard_deviation": 0.00919506672729998, "minimum": 0.12477198207301687, "maximum": 0.1541409444533549, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949405039866168, "standard_deviation": 0.001211558539791988, "minimum": 0.9928054735223888, "maximum": 0.9964671010190653, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8348389256700266, "standard_deviation": 0.04224920089638085, "minimum": 0.7926192986137582, "maximum": 0.9032453966157822, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7580001893588899, "standard_deviation": 0.3735675589454867, "minimum": 0.18972589706929635, "maximum": 1.4219104537823553, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07553419236058856, "standard_deviation": 0.010275247342386504, "minimum": 0.062410385244527224, "maximum": 0.09527014446212408, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19464582170231268, "standard_deviation": 0.04887717021089259, "minimum": 0.12662207550578963, "maximum": 0.2648986958323414, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.641495462078465, "standard_deviation": 0.00788869757512678, "minimum": 5.62586733232897, "maximum": 5.64938123163405, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8270002983996434, "standard_deviation": 0.039054067925199984, "minimum": 0.7728337601002891, "maximum": 0.8996233121509668, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.031176878708042245, "standard_deviation": 0.061694876615045684, "minimum": -0.1281011139704966, "maximum": 0.08051415399793221, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9887711173769412, "standard_deviation": 0.0019402607484092131, "minimum": 0.9862494798593358, "maximum": 0.9911570446038196, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1080147748382947, "standard_deviation": 0.02483814151342481, "minimum": 1.0825165178208978, "maximum": 1.1345489391353532, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6799822463766106, "standard_deviation": 0.41602036488153765, "minimum": 0.22698699784223872, "maximum": 1.1861696744674959, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10244103066032513, "standard_deviation": 0.009063280958153046, "minimum": 0.09114210108727908, "maximum": 0.1135511921575956, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968962864724581, "standard_deviation": 0.0009030039758128099, "minimum": 0.9957081254957693, "maximum": 0.9978786492247913, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5914651165005799, "standard_deviation": 0.036258751085988264, "minimum": 0.5370160466584685, "maximum": 0.641940022313381, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6299149352344664, "standard_deviation": 0.3440199682207098, "minimum": 0.20582998824853296, "maximum": 1.125320192131804, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05163125687675739, "standard_deviation": 0.007875783878394332, "minimum": 0.042659148654430566, "maximum": 0.06200077215880357, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0033", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.8607916831970215, "predicted_mos": 2.8607916831970215, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "98ae8f867f2b0c16e26630f9cf134ea4264f6c9f8f4be4c78bd74db4c59eec29", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.683097300119698, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0781575564351318, "全帯域_10ms_rms_envelope_nrmse": 0.10674618467580563, "全帯域_10ms_envelope_correlation": 0.9880477312341708, "全帯域_10ms_modulation_mae_db": 0.5573617763537109, "全帯域_40ms_log_rms_mae_db": 0.5974549290072539, "全帯域_40ms_rms_envelope_nrmse": 0.05567696725425579, "全帯域_40ms_envelope_correlation": 0.9965051456434507, "全帯域_40ms_modulation_mae_db": 0.35303588677159486, "300–1000 Hz_10ms_log_rms_mae_db": 1.643709208608756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1349102939609995, "300–1000 Hz_10ms_envelope_correlation": 0.9842330074743434, "300–1000 Hz_10ms_modulation_mae_db": 0.6814101416904682, "300–1000 Hz_40ms_log_rms_mae_db": 1.0171937494956371, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06046972934927935, "300–1000 Hz_40ms_envelope_correlation": 0.9966033100874838, "300–1000 Hz_40ms_modulation_mae_db": 0.862528511622387, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8946831930420343, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17945311495018, "1000–3000 Hz_10ms_envelope_correlation": 0.9746750763089274, "1000–3000 Hz_10ms_modulation_mae_db": 1.339029572946344, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3223271650362056, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11182564150433041, "1000–3000 Hz_40ms_envelope_correlation": 0.9897294657872162, "1000–3000 Hz_40ms_modulation_mae_db": 1.025208877519472, "3000–6000 Hz_10ms_log_rms_mae_db": 2.916371100821989, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4189524815611366, "3000–6000 Hz_10ms_envelope_correlation": 0.8850845101480866, "3000–6000 Hz_10ms_modulation_mae_db": 2.12690950983367, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2862304855261204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33205295604783763, "3000–6000 Hz_40ms_envelope_correlation": 0.9228936219977167, "3000–6000 Hz_40ms_modulation_mae_db": 2.1547025654521836, "mr_stft_db": 6.136418025520398, "band_mae_db": 0.3255933436147187, "rms_gain_error_db": 0.07451814841325712, "output_peak": 0.8930737011712051}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0930180747331697, "全帯域_10ms_rms_envelope_nrmse": 0.11210897577504565, "全帯域_10ms_envelope_correlation": 0.9867388333454167, "全帯域_10ms_modulation_mae_db": 0.8800604514007246, "全帯域_40ms_log_rms_mae_db": 0.5396688898082923, "全帯域_40ms_rms_envelope_nrmse": 0.055126714851461235, "全帯域_40ms_envelope_correlation": 0.996870180124145, "全帯域_40ms_modulation_mae_db": 0.9953304236560058, "300–1000 Hz_10ms_log_rms_mae_db": 1.5595314072453266, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15772851345466474, "300–1000 Hz_10ms_envelope_correlation": 0.9802730088343402, "300–1000 Hz_10ms_modulation_mae_db": 0.6456078460325836, "300–1000 Hz_40ms_log_rms_mae_db": 0.9529357516651618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10651764835233558, "300–1000 Hz_40ms_envelope_correlation": 0.9920418968527341, "300–1000 Hz_40ms_modulation_mae_db": 0.7053555894899368, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9393609095843296, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17964333078259562, "1000–3000 Hz_10ms_envelope_correlation": 0.9746127603626817, "1000–3000 Hz_10ms_modulation_mae_db": 1.2999974883416674, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3364331208135702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.102954609258307, "1000–3000 Hz_40ms_envelope_correlation": 0.9913471847484548, "1000–3000 Hz_40ms_modulation_mae_db": 1.3009370770950393, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8958362191867195, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4093442282309912, "3000–6000 Hz_10ms_envelope_correlation": 0.8772915132260517, "3000–6000 Hz_10ms_modulation_mae_db": 1.8201062619059256, "3000–6000 Hz_40ms_log_rms_mae_db": 2.260668054614296, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32694907508655097, "3000–6000 Hz_40ms_envelope_correlation": 0.9145915604404681, "3000–6000 Hz_40ms_modulation_mae_db": 1.7273426993492225, "mr_stft_db": 6.107567498677162, "band_mae_db": 0.35939091187916544, "rms_gain_error_db": -0.1771046649249685, "output_peak": 0.7419400984160112}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0396807163567527, "全帯域_10ms_rms_envelope_nrmse": 0.09114596029916179, "全帯域_10ms_envelope_correlation": 0.9911272641094219, "全帯域_10ms_modulation_mae_db": 0.31440244771440595, "全帯域_40ms_log_rms_mae_db": 0.5228504219958582, "全帯域_40ms_rms_envelope_nrmse": 0.04102091506458537, "全帯域_40ms_envelope_correlation": 0.9980787682514369, "全帯域_40ms_modulation_mae_db": 0.11734302342249237, "300–1000 Hz_10ms_log_rms_mae_db": 1.6058065732581204, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1187711853557817, "300–1000 Hz_10ms_envelope_correlation": 0.9883936528608506, "300–1000 Hz_10ms_modulation_mae_db": 0.713603437818389, "300–1000 Hz_40ms_log_rms_mae_db": 0.9755741635153078, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06908719299905154, "300–1000 Hz_40ms_envelope_correlation": 0.9964111502149481, "300–1000 Hz_40ms_modulation_mae_db": 0.7537226773230715, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8706374943381594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17492444517534367, "1000–3000 Hz_10ms_envelope_correlation": 0.9760465024383965, "1000–3000 Hz_10ms_modulation_mae_db": 1.9604340325977863, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3191722239774686, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11510942469627895, "1000–3000 Hz_40ms_envelope_correlation": 0.9893609352961132, "1000–3000 Hz_40ms_modulation_mae_db": 1.921634277128592, "3000–6000 Hz_10ms_log_rms_mae_db": 2.883548779378078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4226259545231949, "3000–6000 Hz_10ms_envelope_correlation": 0.88077504485925, "3000–6000 Hz_10ms_modulation_mae_db": 2.1079108105369486, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2744745101833708, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.338244491452273, "3000–6000 Hz_40ms_envelope_correlation": 0.9177887101066502, "3000–6000 Hz_40ms_modulation_mae_db": 2.018311788457408, "mr_stft_db": 6.120849458869472, "band_mae_db": 0.35725648809816096, "rms_gain_error_db": -0.06024881526554767, "output_peak": 0.8237901918395514}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1352589391294825, "全帯域_10ms_rms_envelope_nrmse": 0.10462163849731476, "全帯域_10ms_envelope_correlation": 0.9884809961887335, "全帯域_10ms_modulation_mae_db": 0.8077000181838784, "全帯域_40ms_log_rms_mae_db": 0.5769316954018421, "全帯域_40ms_rms_envelope_nrmse": 0.059410529770857756, "全帯域_40ms_envelope_correlation": 0.9959895614600351, "全帯域_40ms_modulation_mae_db": 0.9818343764357419, "300–1000 Hz_10ms_log_rms_mae_db": 1.66239592948634, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1295161370613948, "300–1000 Hz_10ms_envelope_correlation": 0.985636618665472, "300–1000 Hz_10ms_modulation_mae_db": 0.4047077257859862, "300–1000 Hz_40ms_log_rms_mae_db": 1.0388539281851727, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0688590434339868, "300–1000 Hz_40ms_envelope_correlation": 0.9959108559851586, "300–1000 Hz_40ms_modulation_mae_db": 0.4653346550663377, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9025673585863498, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18364846719801972, "1000–3000 Hz_10ms_envelope_correlation": 0.9734785674036834, "1000–3000 Hz_10ms_modulation_mae_db": 1.6705494577804585, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3205678820349835, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11821442579068289, "1000–3000 Hz_40ms_envelope_correlation": 0.988608487406102, "1000–3000 Hz_40ms_modulation_mae_db": 1.7521148467091059, "3000–6000 Hz_10ms_log_rms_mae_db": 2.937790013928846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.43139970015081924, "3000–6000 Hz_10ms_envelope_correlation": 0.872539768167994, "3000–6000 Hz_10ms_modulation_mae_db": 2.568620614310867, "3000–6000 Hz_40ms_log_rms_mae_db": 2.319929636594963, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.34332912673523225, "3000–6000 Hz_40ms_envelope_correlation": 0.9125783877991842, "3000–6000 Hz_40ms_modulation_mae_db": 2.6481091636595546, "mr_stft_db": 6.136580501002111, "band_mae_db": 0.27840317041368323, "rms_gain_error_db": 0.030668795721829872, "output_peak": 0.8428658751342872}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1491940839576626, "全帯域_10ms_rms_envelope_nrmse": 0.110707602812115, "全帯域_10ms_envelope_correlation": 0.9870641539928315, "全帯域_10ms_modulation_mae_db": 0.922297692068113, "全帯域_40ms_log_rms_mae_db": 0.5850937731367579, "全帯域_40ms_rms_envelope_nrmse": 0.054224321002624976, "全帯域_40ms_envelope_correlation": 0.996630325152809, "全帯域_40ms_modulation_mae_db": 0.9748882436481546, "300–1000 Hz_10ms_log_rms_mae_db": 1.6279386741195179, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12698246447724026, "300–1000 Hz_10ms_envelope_correlation": 0.9861474147019887, "300–1000 Hz_10ms_modulation_mae_db": 0.45428518295311865, "300–1000 Hz_40ms_log_rms_mae_db": 0.9619330919380203, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07060401488294875, "300–1000 Hz_40ms_envelope_correlation": 0.9955979392228148, "300–1000 Hz_40ms_modulation_mae_db": 0.5070294332612557, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9444050736454122, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2078422459108824, "1000–3000 Hz_10ms_envelope_correlation": 0.9673565106785543, "1000–3000 Hz_10ms_modulation_mae_db": 2.565062893088737, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4351235501352364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1361376322770173, "1000–3000 Hz_40ms_envelope_correlation": 0.9851519751984905, "1000–3000 Hz_40ms_modulation_mae_db": 2.8288805454558976, "3000–6000 Hz_10ms_log_rms_mae_db": 2.9972768497040416, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.4180336265215109, "3000–6000 Hz_10ms_envelope_correlation": 0.8811264593064718, "3000–6000 Hz_10ms_modulation_mae_db": 1.605183265305098, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3833901762745886, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3347007496487639, "3000–6000 Hz_40ms_envelope_correlation": 0.9177340787804108, "3000–6000 Hz_40ms_modulation_mae_db": 1.4315052827311288, "mr_stft_db": 6.132093431595794, "band_mae_db": 0.26712612194901914, "rms_gain_error_db": 0.047888327663938024, "output_peak": 1.2137374262569574}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 4, "at_or_above_full_scale_fraction": 4.1666666666666665e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 4.1666666666666665e-05, "longest_full_scale_run_samples": 4, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0674927944331831, "全帯域_10ms_rms_envelope_nrmse": 0.10219229561349126, "全帯域_10ms_envelope_correlation": 0.9888277932968662, "全帯域_10ms_modulation_mae_db": 0.40640587084433016, "全帯域_40ms_log_rms_mae_db": 0.57076316551285, "全帯域_40ms_rms_envelope_nrmse": 0.04646144714486184, "全帯域_40ms_envelope_correlation": 0.9975296838134909, "全帯域_40ms_modulation_mae_db": 0.6402884809698789, "300–1000 Hz_10ms_log_rms_mae_db": 1.6584989595223298, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15426221440089768, "300–1000 Hz_10ms_envelope_correlation": 0.9798074940851605, "300–1000 Hz_10ms_modulation_mae_db": 1.036936751110434, "300–1000 Hz_40ms_log_rms_mae_db": 1.0580418028835752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0889532629526823, "300–1000 Hz_40ms_envelope_correlation": 0.9935119366840327, "300–1000 Hz_40ms_modulation_mae_db": 1.6649699398197153, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8754775714574305, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18249390501982973, "1000–3000 Hz_10ms_envelope_correlation": 0.9737990969522752, "1000–3000 Hz_10ms_modulation_mae_db": 0.9373380041725984, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3151359585138644, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11232671145865737, "1000–3000 Hz_40ms_envelope_correlation": 0.9896757577184897, "1000–3000 Hz_40ms_modulation_mae_db": 0.9456902244188603, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8546111566468957, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.40261602310312744, "3000–6000 Hz_10ms_envelope_correlation": 0.8804150453868217, "3000–6000 Hz_10ms_modulation_mae_db": 1.6148103355620806, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2560444950013117, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.32786146689901824, "3000–6000 Hz_40ms_envelope_correlation": 0.9133373742324482, "3000–6000 Hz_40ms_modulation_mae_db": 1.5337128902219679, "mr_stft_db": 6.131204087734511, "band_mae_db": 0.2792813757382369, "rms_gain_error_db": -0.0662017712228138, "output_peak": 0.766247720583072}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0770532660253387, "全帯域_10ms_rms_envelope_nrmse": 0.09475131501296281, "全帯域_10ms_envelope_correlation": 0.9904081236304982, "全帯域_10ms_modulation_mae_db": 0.4549264186533948, "全帯域_40ms_log_rms_mae_db": 0.6016442184296109, "全帯域_40ms_rms_envelope_nrmse": 0.05120761189030544, "全帯域_40ms_envelope_correlation": 0.9969426763407198, "全帯域_40ms_modulation_mae_db": 0.6129496091597324, "300–1000 Hz_10ms_log_rms_mae_db": 1.7339945775924965, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13280754243425352, "300–1000 Hz_10ms_envelope_correlation": 0.9849880322297435, "300–1000 Hz_10ms_modulation_mae_db": 0.7482478655819746, "300–1000 Hz_40ms_log_rms_mae_db": 1.088312637424039, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08108114516469193, "300–1000 Hz_40ms_envelope_correlation": 0.9943343785937802, "300–1000 Hz_40ms_modulation_mae_db": 0.4653234258852539, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8833396473666646, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1964013121162372, "1000–3000 Hz_10ms_envelope_correlation": 0.9696719234338987, "1000–3000 Hz_10ms_modulation_mae_db": 1.6980617151440487, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3629631498593557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13918207290212803, "1000–3000 Hz_40ms_envelope_correlation": 0.9837766865019539, "1000–3000 Hz_40ms_modulation_mae_db": 1.9028289889616439, "3000–6000 Hz_10ms_log_rms_mae_db": 2.8721181527599757, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.42188466531716085, "3000–6000 Hz_10ms_envelope_correlation": 0.8831559938145332, "3000–6000 Hz_10ms_modulation_mae_db": 2.1973114806607024, "3000–6000 Hz_40ms_log_rms_mae_db": 2.2496909666913427, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.33779295677489, "3000–6000 Hz_40ms_envelope_correlation": 0.91997911645169, "3000–6000 Hz_40ms_modulation_mae_db": 2.331974635674649, "mr_stft_db": 6.1274373939780515, "band_mae_db": 0.2697428104333003, "rms_gain_error_db": -0.025293623527013216, "output_peak": 0.8805301027521643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0980391875106308, "全帯域_10ms_rms_envelope_nrmse": 0.10114272290164839, "全帯域_10ms_envelope_correlation": 0.9891034489962803, "全帯域_10ms_modulation_mae_db": 0.47880463809132, "全帯域_40ms_log_rms_mae_db": 0.5759668603764008, "全帯域_40ms_rms_envelope_nrmse": 0.054994715941551114, "全帯域_40ms_envelope_correlation": 0.996495538727714, "全帯域_40ms_modulation_mae_db": 0.5740111113947043, "300–1000 Hz_10ms_log_rms_mae_db": 1.6650111605839775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13003587414358406, "300–1000 Hz_10ms_envelope_correlation": 0.98541541775269, "300–1000 Hz_10ms_modulation_mae_db": 0.6086949790424517, "300–1000 Hz_40ms_log_rms_mae_db": 0.9777589353927966, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07386408545421802, "300–1000 Hz_40ms_envelope_correlation": 0.9950600687875031, "300–1000 Hz_40ms_modulation_mae_db": 0.4649193900352375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8520304714646605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16657693855994685, "1000–3000 Hz_10ms_envelope_correlation": 0.9783146498264024, "1000–3000 Hz_10ms_modulation_mae_db": 0.8738546969612906, "1000–3000 Hz_40ms_log_rms_mae_db": 1.281175549629654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10426366589170588, "1000–3000 Hz_40ms_envelope_correlation": 0.9912957636424737, "1000–3000 Hz_40ms_modulation_mae_db": 0.6373833602625313, "3000–6000 Hz_10ms_log_rms_mae_db": 2.894095253380769, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.43549311381452127, "3000–6000 Hz_10ms_envelope_correlation": 0.8685406584291644, "3000–6000 Hz_10ms_modulation_mae_db": 1.8155641623061678, "3000–6000 Hz_40ms_log_rms_mae_db": 2.3139260365754346, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.35483667078476183, "3000–6000 Hz_40ms_envelope_correlation": 0.9052417023347623, "3000–6000 Hz_40ms_modulation_mae_db": 1.9284980114674797, "mr_stft_db": 6.122764612036419, "band_mae_db": 0.31053411406460524, "rms_gain_error_db": -0.016249405346252997, "output_peak": 0.8250750992999929}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.419453217909683, "spectral_convergence": 0.3312826358101457, "spectral_convergence_status": "ok", "amplitude_floor": 2.5091999427961322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.929717988113287, "spectral_convergence": 0.2983170231006276, "spectral_convergence_status": "ok", "amplitude_floor": 1.1904358876245626e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.060082870538223, "spectral_convergence": 0.32449514293677584, "spectral_convergence_status": "ok", "amplitude_floor": 8.234850781853941e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001792776939061309, "output_power": 0.002022420159261786, "delta_db": 0.5234512849924904, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0038916933601920887, "output_power": 0.00380497368663162, "delta_db": -0.09786955654095249, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002833146281043118, "output_power": 0.0028612726804476, "delta_db": 0.04290249620629883, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00044617096434630785, "output_power": 0.0004213243412047173, "delta_db": -0.24884753723172517, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.1707528478122786e-05, "output_power": 5.754265471041598e-05, "delta_db": 0.4643611422688357, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2019368729027054e-05, "output_power": 3.6561631730137864e-05, "delta_db": 0.5761280444480098, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3255933436147187, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.07451814841325712, "pitch": null, "output_rms_dbfs": -20.360189001470754, "output_peak": 0.8930737011712051, "reference_rms_dbfs": -20.434707149884012, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.136418025520398}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003179454955178215, "log_rms_mae_db": 1.0781575564351318, "rms_envelope_nrmse": 0.10674618467580563, "envelope_correlation": 0.9880477312341708, "rms_gain_error_db": 0.07451803843161955, "reference_mean_power": 0.009055057750136907, "output_mean_power": 0.009211768777176828}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4928013782154505e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002203466496672066, "output_power": 0.0022834143220941744, "reference_active": true, "delta_db": 0.15478269197395167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002920043469136947, "output_power": 0.00033457305148950847, "reference_active": true, "delta_db": 0.5910164075744792}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.410659541472349e-05, "output_power": 9.17245277493946e-05, "reference_active": true, "delta_db": 0.9262862295127019}], "active_band_count": 3, "active_band_mae_db": 0.5573617763537109}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002752533875730688, "log_rms_mae_db": 0.5974549290072539, "rms_envelope_nrmse": 0.05567696725425579, "envelope_correlation": 0.9965051456434507, "rms_gain_error_db": 0.07446122198175825, "reference_mean_power": 0.009100511091100074, "output_mean_power": 0.009257887637890075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.115339465742665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001970497325835552, "output_power": 0.0020740466825120755, "reference_active": true, "delta_db": 0.22242677415449083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014191960918623396, "output_power": 0.0001697081804121109, "reference_active": true, "delta_db": 0.7766037049624147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.976197244793298e-06, "output_power": 4.031583305591117e-06, "reference_active": true, "delta_db": 0.06007718119787898}], "active_band_count": 3, "active_band_mae_db": 0.35303588677159486}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028904791382782896, "log_rms_mae_db": 1.643709208608756, "rms_envelope_nrmse": 0.1349102939609995, "envelope_correlation": 0.9842330074743434, "rms_gain_error_db": -0.0652867146987219, "reference_mean_power": 0.00500094100859156, "output_mean_power": 0.004926325004274226}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3504006393160654e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001701857204603173, "output_power": 0.0016188014572126232, "reference_active": true, "delta_db": -0.21729530863276808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00029604441935142547, "output_power": 0.00028312504736854107, "reference_active": true, "delta_db": -0.19378586717247165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.422561443844994e-05, "output_power": 5.096110970709117e-05, "reference_active": true, "delta_db": -1.633149249266165}], "active_band_count": 3, "active_band_mae_db": 0.6814101416904682}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002500254339817177, "log_rms_mae_db": 1.0171937494956371, "rms_envelope_nrmse": 0.06046972934927935, "envelope_correlation": 0.9966033100874838, "rms_gain_error_db": -0.06538216458566891, "reference_mean_power": 0.00502600851498449, "output_mean_power": 0.004950909680599941}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0757929049004627e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015562264230646289, "output_power": 0.0015096983455751198, "reference_active": true, "delta_db": -0.13182605777685422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001439985744956168, "output_power": 0.00014126285604733425, "reference_active": true, "delta_db": -0.0833021029287203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7762361828128385e-06, "output_power": 2.765922816640559e-06, "reference_active": true, "delta_db": -2.3724573741615864}], "active_band_count": 3, "active_band_mae_db": 0.862528511622387}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010495143398294122, "log_rms_mae_db": 1.8946831930420343, "rms_envelope_nrmse": 0.17945311495018, "envelope_correlation": 0.9746750763089274, "rms_gain_error_db": -0.1470645308530296, "reference_mean_power": 0.0004146271168721916, "output_mean_power": 0.00040082172128274966}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0723455289073609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010117430747443104, "output_power": 0.00010739568113280146, "reference_active": true, "delta_db": 0.2591657638279354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2721317302684375e-05, "output_power": 1.243624011753755e-05, "reference_active": true, "delta_db": -0.09842986423908358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6903995233241656e-06, "output_power": 6.248360617491512e-06, "reference_active": true, "delta_db": 3.6594930907720133}], "active_band_count": 3, "active_band_mae_db": 1.339029572946344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.780703381807892e-05, "log_rms_mae_db": 1.3223271650362056, "rms_envelope_nrmse": 0.11182564150433041, "envelope_correlation": 0.9897294657872162, "rms_gain_error_db": -0.1470819478955409, "reference_mean_power": 0.0004167096352122121, "output_mean_power": 0.0004028332847024389}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9169450940862202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.13104680727751e-05, "output_power": 9.825979286196553e-05, "reference_active": true, "delta_db": 0.3185527522597854}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.750575155696485e-06, "output_power": 5.817319834360054e-06, "reference_active": true, "delta_db": 0.05011658099717406}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0115145688994735e-07, "output_power": 3.7516206038690717e-07, "reference_active": true, "delta_db": 2.7069572993014566}], "active_band_count": 3, "active_band_mae_db": 1.025208877519472}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4613391249073974e-05, "log_rms_mae_db": 2.916371100821989, "rms_envelope_nrmse": 0.4189524815611366, "envelope_correlation": 0.8850845101480866, "rms_gain_error_db": 0.7591826887830413, "reference_mean_power": 2.714804754964509e-05, "output_mean_power": 3.233380911235913e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3627974020623764e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.8436688435136e-06, "output_power": 4.455213541964538e-06, "reference_active": true, "delta_db": -0.3630591827901325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.21772084491833e-06, "output_power": 2.7122728446819945e-06, "reference_active": true, "delta_db": 0.8742649728244543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.650596314175208e-07, "output_power": 1.5200192235429176e-06, "reference_active": true, "delta_db": 5.1434043738864235}], "active_band_count": 3, "active_band_mae_db": 2.12690950983367}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2268327401512992e-05, "log_rms_mae_db": 2.2862304855261204, "rms_envelope_nrmse": 0.33205295604783763, "envelope_correlation": 0.9228936219977167, "rms_gain_error_db": 0.759151426921081, "reference_mean_power": 2.7284479980131438e-05, "output_mean_power": 3.249606865421563e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2818152654967464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7409144698248785e-06, "output_power": 4.607261414320123e-06, "reference_active": true, "delta_db": -0.12419265682454612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1748833712187593e-06, "output_power": 1.5956160448365202e-06, "reference_active": true, "delta_db": 1.3293363770719093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.424838593381526e-08, "output_power": 1.0856703443786786e-07, "reference_active": true, "delta_db": 5.010578662460096}], "active_band_count": 3, "active_band_mae_db": 2.1547025654521836}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8427984800129962, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4b52d50390ccbc0b9c5090b303c013822ae334d41efd3ab4b1237f73926b4be8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9734943859256024, "standard_deviation": 0.0034799602833373755, "minimum": 0.9673565106785543, "maximum": 0.9783146498264024, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8953127149356301, "standard_deviation": 0.032562765026152594, "minimum": 1.8520304714646605, "maximum": 1.9444050736454122, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.5430409826291163, "standard_deviation": 0.5573422596573556, "minimum": 0.8738546969612906, "maximum": 2.565062893088737, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1838729699641294, "standard_deviation": 0.012817908968563933, "minimum": 0.16657693855994685, "maximum": 0.2078422459108824, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9886182820374119, "standard_deviation": 0.0027510196063502417, "minimum": 0.9837766865019539, "maximum": 0.9913471847484548, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3366123250000421, "standard_deviation": 0.04582416239182755, "minimum": 1.281175549629654, "maximum": 1.4351235501352364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5393347746938928, "standard_deviation": 0.7044322651952459, "minimum": 0.6373833602625313, "maximum": 2.8288805454558976, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11750177297238848, "standard_deviation": 0.013466514252006716, "minimum": 0.102954609258307, "maximum": 0.13918207290212803, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.8786161241672967, "standard_deviation": 0.00556572390934991, "minimum": 0.8685406584291644, "maximum": 0.8850845101480866, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.9064559407259143, "standard_deviation": 0.04469873837729792, "minimum": 2.8546111566468957, "maximum": 2.9972768497040416, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.9820520550526826, "standard_deviation": 0.32908197570347586, "minimum": 1.605183265305098, "maximum": 2.568620614310867, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.4200437241528078, "standard_deviation": 0.010698024447682268, "minimum": 0.40261602310312744, "maximum": 0.43549311381452127, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9155180690179163, "standard_deviation": 0.005401395612662072, "minimum": 0.9052417023347623, "maximum": 0.9228936219977167, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 2.2930442951826784, "standard_deviation": 0.04473609777950087, "minimum": 2.2496909666913427, "maximum": 2.3833901762745886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.9717696296266993, "standard_deviation": 0.4085181709341226, "minimum": 1.4315052827311288, "maximum": 2.6481091636595546, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.33697093667866596, "standard_deviation": 0.009060661650539278, "minimum": 0.32694907508655097, "maximum": 0.35483667078476183, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9843618308255737, "standard_deviation": 0.0029288759479431394, "minimum": 0.9798074940851605, "maximum": 0.9883936528608506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.644610811302108, "standard_deviation": 0.050621663836986845, "minimum": 1.5595314072453266, "maximum": 1.7339945775924965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6616867412519258, "standard_deviation": 0.19402136574522003, "minimum": 0.4047077257859862, "maximum": 1.036936751110434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13562677816110202, "standard_deviation": 0.013478347380643182, "minimum": 0.1187711853557817, "maximum": 0.15772851345466474, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949339420535569, "standard_deviation": 0.0015641335193824798, "minimum": 0.9920418968527341, "maximum": 0.9966033100874838, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.008825507562464, "standard_deviation": 0.049433257026279195, "minimum": 0.9529357516651618, "maximum": 1.088312637424039, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7361479528128994, "standard_deviation": 0.4057946992653223, "minimum": 0.4649193900352375, "maximum": 1.6649699398197153, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07742951532364928, "standard_deviation": 0.01455154414930121, "minimum": 0.06046972934927935, "maximum": 0.10651764835233558, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.30591604202386125, "standard_deviation": 0.038141559695284336, "minimum": 0.26712612194901914, "maximum": 0.35939091187916544, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.12686437617674, "standard_deviation": 0.009691944676142357, "minimum": 6.107567498677162, "maximum": 6.136580501002111, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8734075269316552, "standard_deviation": 0.14675515796455724, "minimum": 0.7419400984160112, "maximum": 1.2137374262569574, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.024002876060946397, "standard_deviation": 0.07969601539080264, "minimum": -0.1771046649249685, "maximum": 0.07451814841325712, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9887247930992773, "standard_deviation": 0.0015108685321244617, "minimum": 0.9867388333454167, "maximum": 0.9911272641094219, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0922368273226688, "standard_deviation": 0.035722116568354695, "minimum": 1.0396807163567527, "maximum": 1.1491940839576626, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6027449141637347, "standard_deviation": 0.2336010049080467, "minimum": 0.31440244771440595, "maximum": 0.922297692068113, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10292708694844316, "standard_deviation": 0.007286607942526924, "minimum": 0.09114596029916179, "maximum": 0.11210897577504565, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968802349392252, "standard_deviation": 0.0006553846494919624, "minimum": 0.9959895614600351, "maximum": 0.9980787682514369, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5712967442086083, "standard_deviation": 0.027246998542180947, "minimum": 0.5228504219958582, "maximum": 0.6016442184296109, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6562101444322881, "standard_deviation": 0.31932148916407543, "minimum": 0.11734302342249237, "maximum": 0.9953304236560058, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05226540286506294, "standard_deviation": 0.0058968339707193675, "minimum": 0.04102091506458537, "maximum": 0.059410529770857756, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0033", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8205626010894775, "predicted_mos": 2.8205626010894775, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5138361e4144eb61bd975bc440ff0f8dcf1bdbd5c9abb97944828413844581e9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6949108000844717, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0983672125846078, "全帯域_10ms_rms_envelope_nrmse": 0.09838133833615458, "全帯域_10ms_envelope_correlation": 0.9896887528001508, "全帯域_10ms_modulation_mae_db": 0.5279058905711423, "全帯域_40ms_log_rms_mae_db": 0.5852334093089598, "全帯域_40ms_rms_envelope_nrmse": 0.04948337634912549, "全帯域_40ms_envelope_correlation": 0.9971877910554132, "全帯域_40ms_modulation_mae_db": 0.4859622349069585, "300–1000 Hz_10ms_log_rms_mae_db": 1.3268669140034002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12461090259651487, "300–1000 Hz_10ms_envelope_correlation": 0.9867729893990854, "300–1000 Hz_10ms_modulation_mae_db": 0.6715484686731537, "300–1000 Hz_40ms_log_rms_mae_db": 0.8350550075290515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0732316255249978, "300–1000 Hz_40ms_envelope_correlation": 0.9953743067603857, "300–1000 Hz_40ms_modulation_mae_db": 0.5317535769703415, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4405520801806708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1457673985845852, "1000–3000 Hz_10ms_envelope_correlation": 0.9833875343620152, "1000–3000 Hz_10ms_modulation_mae_db": 1.0076078113707045, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9659514147587195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08099430970718301, "1000–3000 Hz_40ms_envelope_correlation": 0.9946272709050987, "1000–3000 Hz_40ms_modulation_mae_db": 0.9803707448574737, "3000–6000 Hz_10ms_log_rms_mae_db": 2.006585433697559, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3564712215751425, "3000–6000 Hz_10ms_envelope_correlation": 0.9074697560738949, "3000–6000 Hz_10ms_modulation_mae_db": 1.5080005435026917, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6343817011594501, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2936924173383557, "3000–6000 Hz_40ms_envelope_correlation": 0.9307950385901437, "3000–6000 Hz_40ms_modulation_mae_db": 1.4912710974111347, "mr_stft_db": 5.602586651619862, "band_mae_db": 0.1502664556694333, "rms_gain_error_db": -0.045351455992162976, "output_peak": 0.8934724104359386}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1623755821979607, "全帯域_10ms_rms_envelope_nrmse": 0.1000478433149072, "全帯域_10ms_envelope_correlation": 0.9893108705117023, "全帯域_10ms_modulation_mae_db": 0.9244969638338459, "全帯域_40ms_log_rms_mae_db": 0.5620043946840928, "全帯域_40ms_rms_envelope_nrmse": 0.054919695266482986, "全帯域_40ms_envelope_correlation": 0.9965121504864585, "全帯域_40ms_modulation_mae_db": 1.0784175303825685, "300–1000 Hz_10ms_log_rms_mae_db": 1.3670574504486903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13240057893876772, "300–1000 Hz_10ms_envelope_correlation": 0.9849771301474172, "300–1000 Hz_10ms_modulation_mae_db": 0.7094763614019958, "300–1000 Hz_40ms_log_rms_mae_db": 0.8115248339789366, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06991101314439428, "300–1000 Hz_40ms_envelope_correlation": 0.995774979119261, "300–1000 Hz_40ms_modulation_mae_db": 0.6779328042465997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4864802300023683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1786810700187458, "1000–3000 Hz_10ms_envelope_correlation": 0.9749839552643244, "1000–3000 Hz_10ms_modulation_mae_db": 1.5243561198136544, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9964167257462335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11082735594081972, "1000–3000 Hz_40ms_envelope_correlation": 0.9902079977873175, "1000–3000 Hz_40ms_modulation_mae_db": 1.4251360423848185, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0043151253514737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3602747256423247, "3000–6000 Hz_10ms_envelope_correlation": 0.9090962001110641, "3000–6000 Hz_10ms_modulation_mae_db": 1.5473862383465689, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6279661444002684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2946633160124117, "3000–6000 Hz_40ms_envelope_correlation": 0.9333139168086311, "3000–6000 Hz_40ms_modulation_mae_db": 1.7156588892353506, "mr_stft_db": 5.604098889832911, "band_mae_db": 0.1953084184200714, "rms_gain_error_db": -0.04760209704443055, "output_peak": 0.8851296127633431}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.118245749673918, "全帯域_10ms_rms_envelope_nrmse": 0.10663566643879539, "全帯域_10ms_envelope_correlation": 0.9878649150477145, "全帯域_10ms_modulation_mae_db": 0.9964400478636911, "全帯域_40ms_log_rms_mae_db": 0.5560810702187347, "全帯域_40ms_rms_envelope_nrmse": 0.052699999055886265, "全帯域_40ms_envelope_correlation": 0.9968203396904711, "全帯域_40ms_modulation_mae_db": 0.8431287834067646, "300–1000 Hz_10ms_log_rms_mae_db": 1.3251057524642733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13167394395351892, "300–1000 Hz_10ms_envelope_correlation": 0.9854775432889772, "300–1000 Hz_10ms_modulation_mae_db": 0.493734901710365, "300–1000 Hz_40ms_log_rms_mae_db": 0.8489335978611717, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07525218129598348, "300–1000 Hz_40ms_envelope_correlation": 0.9955237064264438, "300–1000 Hz_40ms_modulation_mae_db": 0.8697178689601809, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3910233811970674, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16492382510402467, "1000–3000 Hz_10ms_envelope_correlation": 0.9787189311588447, "1000–3000 Hz_10ms_modulation_mae_db": 1.8918095759027445, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9981421324419867, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1092267768515582, "1000–3000 Hz_40ms_envelope_correlation": 0.9903316091087725, "1000–3000 Hz_40ms_modulation_mae_db": 1.924282200118646, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0044960874486506, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35609169642780475, "3000–6000 Hz_10ms_envelope_correlation": 0.9100047400286411, "3000–6000 Hz_10ms_modulation_mae_db": 1.4890176871208933, "3000–6000 Hz_40ms_log_rms_mae_db": 1.610347381503021, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29503723046454283, "3000–6000 Hz_40ms_envelope_correlation": 0.9321360068424586, "3000–6000 Hz_40ms_modulation_mae_db": 1.3647312909959017, "mr_stft_db": 5.595875818596583, "band_mae_db": 0.1786484880800067, "rms_gain_error_db": -0.06918501612799742, "output_peak": 1.0581456815335315}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 2.0833333333333333e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 2.0833333333333333e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1066718304162149, "全帯域_10ms_rms_envelope_nrmse": 0.09838606109824213, "全帯域_10ms_envelope_correlation": 0.9896506555918176, "全帯域_10ms_modulation_mae_db": 0.6296398281989176, "全帯域_40ms_log_rms_mae_db": 0.539057513859524, "全帯域_40ms_rms_envelope_nrmse": 0.0489006555917868, "全帯域_40ms_envelope_correlation": 0.997247515730028, "全帯域_40ms_modulation_mae_db": 0.44449410869128375, "300–1000 Hz_10ms_log_rms_mae_db": 1.33929989354243, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1479891690620203, "300–1000 Hz_10ms_envelope_correlation": 0.9819703920462788, "300–1000 Hz_10ms_modulation_mae_db": 0.9495548779115902, "300–1000 Hz_40ms_log_rms_mae_db": 0.8352881277677165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09985982823171062, "300–1000 Hz_40ms_envelope_correlation": 0.9921219424953456, "300–1000 Hz_40ms_modulation_mae_db": 0.6560949446319123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4555556383656445, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15689002906597138, "1000–3000 Hz_10ms_envelope_correlation": 0.9808901212190393, "1000–3000 Hz_10ms_modulation_mae_db": 0.551964418171805, "1000–3000 Hz_40ms_log_rms_mae_db": 0.955211392102452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09419449204587062, "1000–3000 Hz_40ms_envelope_correlation": 0.9930859494577393, "1000–3000 Hz_40ms_modulation_mae_db": 0.6253241942297215, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9938929121265776, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35556431199119165, "3000–6000 Hz_10ms_envelope_correlation": 0.9089809282149379, "3000–6000 Hz_10ms_modulation_mae_db": 1.3443100550040452, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6432128930701466, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2989968513713406, "3000–6000 Hz_40ms_envelope_correlation": 0.929051628579198, "3000–6000 Hz_40ms_modulation_mae_db": 1.5068779861227626, "mr_stft_db": 5.624624930518738, "band_mae_db": 0.25867856925915694, "rms_gain_error_db": -0.06151690522665076, "output_peak": 0.7789720717845042}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1749678542985822, "全帯域_10ms_rms_envelope_nrmse": 0.09788335269653012, "全帯域_10ms_envelope_correlation": 0.9897730533501804, "全帯域_10ms_modulation_mae_db": 0.4641707518999894, "全帯域_40ms_log_rms_mae_db": 0.6515773379401684, "全帯域_40ms_rms_envelope_nrmse": 0.05305935624088439, "全帯域_40ms_envelope_correlation": 0.9967278469402928, "全帯域_40ms_modulation_mae_db": 0.4399142586031761, "300–1000 Hz_10ms_log_rms_mae_db": 1.3694577491043238, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12212520555423433, "300–1000 Hz_10ms_envelope_correlation": 0.9872935329064556, "300–1000 Hz_10ms_modulation_mae_db": 0.63125691394995, "300–1000 Hz_40ms_log_rms_mae_db": 0.8280490152358192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06952835720442216, "300–1000 Hz_40ms_envelope_correlation": 0.9958597686470829, "300–1000 Hz_40ms_modulation_mae_db": 0.3157432087361541, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4462198627565332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14694536591923366, "1000–3000 Hz_10ms_envelope_correlation": 0.983261166974501, "1000–3000 Hz_10ms_modulation_mae_db": 1.1384178343356621, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9378117904743377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08664861892786124, "1000–3000 Hz_40ms_envelope_correlation": 0.9941789511387972, "1000–3000 Hz_40ms_modulation_mae_db": 1.1601291682655281, "3000–6000 Hz_10ms_log_rms_mae_db": 1.977340545315745, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3522886224330326, "3000–6000 Hz_10ms_envelope_correlation": 0.907118255375955, "3000–6000 Hz_10ms_modulation_mae_db": 1.4579386580552052, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6193189320118184, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2995891074054223, "3000–6000 Hz_40ms_envelope_correlation": 0.9259638127411844, "3000–6000 Hz_40ms_modulation_mae_db": 1.2119967696175802, "mr_stft_db": 5.6032884599516315, "band_mae_db": 0.16938951922025738, "rms_gain_error_db": -0.030784977057782728, "output_peak": 0.8112387401420125}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0812476150588202, "全帯域_10ms_rms_envelope_nrmse": 0.10329801251976195, "全帯域_10ms_envelope_correlation": 0.9887254199457641, "全帯域_10ms_modulation_mae_db": 0.8631845248262914, "全帯域_40ms_log_rms_mae_db": 0.5865159941122485, "全帯域_40ms_rms_envelope_nrmse": 0.042864336607334, "全帯域_40ms_envelope_correlation": 0.9979024961619934, "全帯域_40ms_modulation_mae_db": 0.9991917748368818, "300–1000 Hz_10ms_log_rms_mae_db": 1.3287234493065097, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12376920258126929, "300–1000 Hz_10ms_envelope_correlation": 0.9867628862206062, "300–1000 Hz_10ms_modulation_mae_db": 0.4622825911766238, "300–1000 Hz_40ms_log_rms_mae_db": 0.7742919426916403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058277173306203765, "300–1000 Hz_40ms_envelope_correlation": 0.9968690720617741, "300–1000 Hz_40ms_modulation_mae_db": 0.24112997149260043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4736975400751502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14918613675251338, "1000–3000 Hz_10ms_envelope_correlation": 0.9825734393505237, "1000–3000 Hz_10ms_modulation_mae_db": 1.2148168841676117, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9704205405785187, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08919481716535185, "1000–3000 Hz_40ms_envelope_correlation": 0.9934799414577069, "1000–3000 Hz_40ms_modulation_mae_db": 1.1386123806258388, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9779512140111075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35394632647396035, "3000–6000 Hz_10ms_envelope_correlation": 0.9096885937909647, "3000–6000 Hz_10ms_modulation_mae_db": 1.520986326345133, "3000–6000 Hz_40ms_log_rms_mae_db": 1.628374146731718, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29717354287061803, "3000–6000 Hz_40ms_envelope_correlation": 0.9299194047483534, "3000–6000 Hz_40ms_modulation_mae_db": 1.4168280911332183, "mr_stft_db": 5.60865521512882, "band_mae_db": 0.13505840443324346, "rms_gain_error_db": 0.023182698313183453, "output_peak": 1.047912188663673}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 2.0833333333333333e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 2.0833333333333333e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1210762054111387, "全帯域_10ms_rms_envelope_nrmse": 0.11011287378502868, "全帯域_10ms_envelope_correlation": 0.9870283183411891, "全帯域_10ms_modulation_mae_db": 1.2106726706106246, "全帯域_40ms_log_rms_mae_db": 0.5802996921353588, "全帯域_40ms_rms_envelope_nrmse": 0.05350682007761508, "全帯域_40ms_envelope_correlation": 0.9966770412669618, "全帯域_40ms_modulation_mae_db": 0.9967481537951189, "300–1000 Hz_10ms_log_rms_mae_db": 1.3473447351422043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1485910611229555, "300–1000 Hz_10ms_envelope_correlation": 0.9809027510405284, "300–1000 Hz_10ms_modulation_mae_db": 1.2920761953932065, "300–1000 Hz_40ms_log_rms_mae_db": 0.7955539656601563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07633740696701045, "300–1000 Hz_40ms_envelope_correlation": 0.9947551814761836, "300–1000 Hz_40ms_modulation_mae_db": 0.7262401797802496, "1000–3000 Hz_10ms_log_rms_mae_db": 1.472471341039786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15752118780999894, "1000–3000 Hz_10ms_envelope_correlation": 0.9806620442551445, "1000–3000 Hz_10ms_modulation_mae_db": 1.3308901518914955, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9685130436790743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08917514278420066, "1000–3000 Hz_40ms_envelope_correlation": 0.9938350360894106, "1000–3000 Hz_40ms_modulation_mae_db": 1.444811196147979, "3000–6000 Hz_10ms_log_rms_mae_db": 2.002483253138146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35403512426524736, "3000–6000 Hz_10ms_envelope_correlation": 0.9087865564128813, "3000–6000 Hz_10ms_modulation_mae_db": 1.2738802186224363, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5971154355607298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2974565843204519, "3000–6000 Hz_40ms_envelope_correlation": 0.9290950941892478, "3000–6000 Hz_40ms_modulation_mae_db": 1.2004106726251635, "mr_stft_db": 5.609355245337092, "band_mae_db": 0.15994327327098182, "rms_gain_error_db": -0.048392324067866976, "output_peak": 0.8275988518939432}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1360631178233098, "全帯域_10ms_rms_envelope_nrmse": 0.1074926253235365, "全帯域_10ms_envelope_correlation": 0.9876722007909707, "全帯域_10ms_modulation_mae_db": 0.8795671336660348, "全帯域_40ms_log_rms_mae_db": 0.5891902201823519, "全帯域_40ms_rms_envelope_nrmse": 0.0625216708282628, "全帯域_40ms_envelope_correlation": 0.995553683782717, "全帯域_40ms_modulation_mae_db": 0.8996301227953633, "300–1000 Hz_10ms_log_rms_mae_db": 1.433862283269939, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14179051602839668, "300–1000 Hz_10ms_envelope_correlation": 0.9832845084879118, "300–1000 Hz_10ms_modulation_mae_db": 0.4544029062811701, "300–1000 Hz_40ms_log_rms_mae_db": 0.8581981903013154, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09288030087943337, "300–1000 Hz_40ms_envelope_correlation": 0.9930257649277237, "300–1000 Hz_40ms_modulation_mae_db": 0.4390525074988792, "1000–3000 Hz_10ms_log_rms_mae_db": 1.460331286782886, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1572672330953189, "1000–3000 Hz_10ms_envelope_correlation": 0.9811536756902444, "1000–3000 Hz_10ms_modulation_mae_db": 0.479218701565948, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0257186910932758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1054029640284769, "1000–3000 Hz_40ms_envelope_correlation": 0.9917109197352798, "1000–3000 Hz_40ms_modulation_mae_db": 0.4088235942953742, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0164185525827345, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3594016605427518, "3000–6000 Hz_10ms_envelope_correlation": 0.9089504875063826, "3000–6000 Hz_10ms_modulation_mae_db": 1.3020989676369144, "3000–6000 Hz_40ms_log_rms_mae_db": 1.642332831638994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30263304875220426, "3000–6000 Hz_40ms_envelope_correlation": 0.9290786531821433, "3000–6000 Hz_40ms_modulation_mae_db": 1.1737347521129091, "mr_stft_db": 5.606615639162341, "band_mae_db": 0.2576947172991895, "rms_gain_error_db": -0.10544034099602348, "output_peak": 0.8691591248160226}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.739423356827746, "spectral_convergence": 0.315534881504197, "spectral_convergence_status": "ok", "amplitude_floor": 2.5091999427961322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.309676896819282, "spectral_convergence": 0.2972791091923517, "spectral_convergence_status": "ok", "amplitude_floor": 1.1904358876245626e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.758659701212556, "spectral_convergence": 0.3276454395450388, "spectral_convergence_status": "ok", "amplitude_floor": 8.234850781853941e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001792776939061309, "output_power": 0.0019178521561628646, "delta_db": 0.29288868005783336, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0038916933601920887, "output_power": 0.003862682886812788, "delta_db": -0.032495578975698186, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002833146281043118, "output_power": 0.0026549878444452027, "delta_db": -0.2820646154231492, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00044617096434630785, "output_power": 0.00043637385629813585, "delta_db": -0.09642580532774647, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.1707528478122786e-05, "output_power": 4.953029143208556e-05, "delta_db": -0.18682896472576066, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2019368729027054e-05, "output_power": 3.2099796146727326e-05, "delta_db": 0.0108950895064119, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1502664556694333, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.045351455992162976, "pitch": null, "output_rms_dbfs": -20.480058605876174, "output_peak": 0.8934724104359386, "reference_rms_dbfs": -20.434707149884012, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.602586651619862}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003179454955178215, "log_rms_mae_db": 1.0983672125846078, "rms_envelope_nrmse": 0.09838133833615458, "envelope_correlation": 0.9896887528001508, "rms_gain_error_db": -0.04535111177113602, "reference_mean_power": 0.009055057750136907, "output_mean_power": 0.00896099248825194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4928013782154505e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002203466496672066, "output_power": 0.0021313131340414955, "reference_active": true, "delta_db": -0.14459190267369695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002920043469136947, "output_power": 0.00030014672741960247, "reference_active": true, "delta_db": 0.11944295880370241}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.410659541472349e-05, "output_power": 0.00010042113890120328, "reference_active": true, "delta_db": 1.3196828102360278}], "active_band_count": 3, "active_band_mae_db": 0.5279058905711423}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002752533875730688, "log_rms_mae_db": 0.5852334093089598, "rms_envelope_nrmse": 0.04948337634912549, "envelope_correlation": 0.9971877910554132, "rms_gain_error_db": -0.04536363892480298, "reference_mean_power": 0.009100511091100074, "output_mean_power": 0.009005947675726302}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.115339465742665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001970497325835552, "output_power": 0.0019408286878617198, "reference_active": true, "delta_db": -0.06588646837928523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014191960918623396, "output_power": 0.00015354224760334568, "reference_active": true, "delta_db": 0.3418548712407419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.976197244793298e-06, "output_power": 5.063868934687078e-06, "reference_active": true, "delta_db": 1.0501453651008483}], "active_band_count": 3, "active_band_mae_db": 0.4859622349069585}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028904791382782896, "log_rms_mae_db": 1.3268669140034002, "rms_envelope_nrmse": 0.12461090259651487, "envelope_correlation": 0.9867729893990854, "rms_gain_error_db": -0.20479135396159076, "reference_mean_power": 0.00500094100859156, "output_mean_power": 0.004770595522368141}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3504006393160654e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001701857204603173, "output_power": 0.001475467357461617, "reference_active": true, "delta_db": -0.6199351170856471}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00029604441935142547, "output_power": 0.00022977493247875477, "reference_active": true, "delta_db": -1.1005423157946335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.422561443844994e-05, "output_power": 7.942744612840391e-05, "reference_active": true, "delta_db": 0.2941679731391805}], "active_band_count": 3, "active_band_mae_db": 0.6715484686731537}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002500254339817177, "log_rms_mae_db": 0.8350550075290515, "rms_envelope_nrmse": 0.0732316255249978, "envelope_correlation": 0.9953743067603857, "rms_gain_error_db": -0.20478077217020596, "reference_mean_power": 0.00502600851498449, "output_mean_power": 0.004794520090734615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0757929049004627e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015562264230646289, "output_power": 0.0013555261944487099, "reference_active": true, "delta_db": -0.5996487051129942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001439985744956168, "output_power": 0.00011457327775892288, "reference_active": true, "delta_db": -0.9927485510311705}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7762361828128385e-06, "output_power": 4.7793863820722226e-06, "reference_active": true, "delta_db": 0.002863474766860001}], "active_band_count": 3, "active_band_mae_db": 0.5317535769703415}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010495143398294122, "log_rms_mae_db": 1.4405520801806708, "rms_envelope_nrmse": 0.1457673985845852, "envelope_correlation": 0.9833875343620152, "rms_gain_error_db": -0.12082998483998272, "reference_mean_power": 0.0004146271168721916, "output_mean_power": 0.0004032503038777014}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0723455289073609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010117430747443104, "output_power": 9.46151802481183e-05, "reference_active": true, "delta_db": -0.2910941931750401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2721317302684375e-05, "output_power": 1.4237212402237808e-05, "reference_active": true, "delta_db": 0.48892879056570643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6903995233241656e-06, "output_power": 4.509172216368627e-06, "reference_active": true, "delta_db": 2.242800450371367}], "active_band_count": 3, "active_band_mae_db": 1.0076078113707045}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.780703381807892e-05, "log_rms_mae_db": 0.9659514147587195, "rms_envelope_nrmse": 0.08099430970718301, "envelope_correlation": 0.9946272709050987, "rms_gain_error_db": -0.12083709202858389, "reference_mean_power": 0.0004167096352122121, "output_mean_power": 0.00040527501747317506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9169450940862202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.13104680727751e-05, "output_power": 8.698923839732745e-05, "reference_active": true, "delta_db": -0.21055040500150735}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.750575155696485e-06, "output_power": 7.124772203775954e-06, "reference_active": true, "delta_db": 0.9305969973105009}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0115145688994735e-07, "output_power": 3.0445258484817264e-07, "reference_active": true, "delta_db": 1.7999648322604127}], "active_band_count": 3, "active_band_mae_db": 0.9803707448574737}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4613391249073974e-05, "log_rms_mae_db": 2.006585433697559, "rms_envelope_nrmse": 0.3564712215751425, "envelope_correlation": 0.9074697560738949, "rms_gain_error_db": 0.18581120996289102, "reference_mean_power": 2.714804754964509e-05, "output_mean_power": 2.8334771987266508e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3627974020623764e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.8436688435136e-06, "output_power": 4.402213015312898e-06, "reference_active": true, "delta_db": -0.41503389765499027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.21772084491833e-06, "output_power": 2.2800119660514168e-06, "reference_active": true, "delta_db": 0.1203024774912593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.650596314175208e-07, "output_power": 1.1651321030728702e-06, "reference_active": true, "delta_db": 3.988665255361826}], "active_band_count": 3, "active_band_mae_db": 1.5080005435026917}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2268327401512992e-05, "log_rms_mae_db": 1.6343817011594501, "rms_envelope_nrmse": 0.2936924173383557, "envelope_correlation": 0.9307950385901437, "rms_gain_error_db": 0.18580738904162297, "reference_mean_power": 2.7284479980131438e-05, "output_mean_power": 2.8477143243617646e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2818152654967464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7409144698248785e-06, "output_power": 4.518934048746899e-06, "reference_active": true, "delta_db": -0.20826117245262457}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1748833712187593e-06, "output_power": 1.2973220455768748e-06, "reference_active": true, "delta_db": 0.4305304114481317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.424838593381526e-08, "output_power": 8.28213451031222e-08, "reference_active": true, "delta_db": 3.835021708332648}], "active_band_count": 3, "active_band_mae_db": 1.4912710974111347}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8537796846829551, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1db1025ec4ee6922e68f11cfacc8068398ea2172eb753930470e0379f498dccd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9807038585343296, "standard_deviation": 0.0027808858463790036, "minimum": 0.9749839552643244, "maximum": 0.9833875343620152, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.453291420050013, "standard_deviation": 0.02935027131380749, "minimum": 1.3910233811970674, "maximum": 1.4864802300023683, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.142385187152453, "standard_deviation": 0.47130310739411224, "minimum": 0.479218701565948, "maximum": 1.8918095759027445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15714778079379899, "standard_deviation": 0.010832225018142908, "minimum": 0.1457673985845852, "maximum": 0.1786810700187458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9926822094600154, "standard_deviation": 0.0017213923095420079, "minimum": 0.9902079977873175, "maximum": 0.9946272709050987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9772732163593248, "standard_deviation": 0.02788947788517928, "minimum": 0.9378117904743377, "maximum": 1.0257186910932758, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1384361901156725, "standard_deviation": 0.4802811918450209, "minimum": 0.4088235942953742, "maximum": 1.924282200118646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09570805968141528, "standard_deviation": 0.011283997056022966, "minimum": 0.08099430970718301, "maximum": 0.11082735594081972, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9087619396893402, "standard_deviation": 0.000997892020481275, "minimum": 0.907118255375955, "maximum": 0.9100047400286411, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9979353904589991, "standard_deviation": 0.013942401582434344, "minimum": 1.977340545315745, "maximum": 2.0164185525827345, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.430452336829236, "standard_deviation": 0.10722404537772807, "minimum": 1.2738802186224363, "maximum": 1.5473862383465689, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.356009211168932, "standard_deviation": 0.0027286001669651087, "minimum": 0.3522886224330326, "maximum": 0.3602747256423247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.92991919446017, "standard_deviation": 0.0022357122848035947, "minimum": 0.9259638127411844, "maximum": 0.9333139168086311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6253811832595184, "standard_deviation": 0.015873139993980018, "minimum": 1.5971154355607298, "maximum": 1.6432128930701466, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3851886936567526, "standard_deviation": 0.18729007152863658, "minimum": 1.1737347521129091, "maximum": 1.7156588892353506, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.29740526231691844, "standard_deviation": 0.002966466781905457, "minimum": 0.2936924173383557, "maximum": 0.30263304875220426, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9846802166921576, "standard_deviation": 0.0023853995369887463, "minimum": 0.9809027510405284, "maximum": 0.9872935329064556, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3547147784102214, "standard_deviation": 0.03635874356382731, "minimum": 1.3251057524642733, "maximum": 1.433862283269939, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7080416520622569, "standard_deviation": 0.287181427351834, "minimum": 0.4544029062811701, "maximum": 1.2920761953932065, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1341188224797097, "standard_deviation": 0.010763555565901542, "minimum": 0.12212520555423433, "maximum": 0.1485910611229555, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.994913090239275, "standard_deviation": 0.001577879584582882, "minimum": 0.9921219424953456, "maximum": 0.9968690720617741, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8233618351282259, "standard_deviation": 0.028004258935267298, "minimum": 0.7742919426916403, "maximum": 0.8581981903013154, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5572081327896148, "standard_deviation": 0.21494405111324671, "minimum": 0.24112997149260043, "maximum": 0.8697178689601809, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07690973581926949, "standard_deviation": 0.013357761188959863, "minimum": 0.058277173306203765, "maximum": 0.09985982823171062, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18812348070654256, "standard_deviation": 0.046833624010273606, "minimum": 0.13505840443324346, "maximum": 0.25867856925915694, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.606887606268497, "standard_deviation": 0.008318557511545505, "minimum": 5.595875818596583, "maximum": 5.624624930518738, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8964535852541211, "standard_deviation": 0.10404023947737796, "minimum": 0.7789720717845042, "maximum": 1.0581456815335315, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04813630227496643, "standard_deviation": 0.03645234926164846, "minimum": -0.10544034099602348, "maximum": 0.023182698313183453, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9887142732974362, "standard_deviation": 0.0010659173410706407, "minimum": 0.9870283183411891, "maximum": 0.9897730533501804, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.124876895933069, "standard_deviation": 0.0317051288374892, "minimum": 1.0812476150588202, "maximum": 1.1749678542985822, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8120097264338171, "standard_deviation": 0.2529112351064882, "minimum": 0.4641707518999894, "maximum": 1.2106726706106246, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10277972168911956, "standard_deviation": 0.004798975097385255, "minimum": 0.09788335269653012, "maximum": 0.11011287378502868, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.996828608139292, "standard_deviation": 0.000677864576276524, "minimum": 0.995553683782717, "maximum": 0.9979024961619934, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5812449540551798, "standard_deviation": 0.03346118630032871, "minimum": 0.539057513859524, "maximum": 0.6515773379401684, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7734358709272644, "standard_deviation": 0.2716905929085888, "minimum": 0.4399142586031761, "maximum": 1.0784175303825685, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05224448875217222, "standard_deviation": 0.005631980069497721, "minimum": 0.042864336607334, "maximum": 0.0625216708282628, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0033", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.673271894454956, "predicted_mos": 2.673271894454956, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "50df2e96b6146c85aff4f4922fd4fb7aa308b1054bfdc04eb68fff7874ccd5ff", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5585734001360834, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1048439767547764, "全帯域_10ms_rms_envelope_nrmse": 0.11101402023226531, "全帯域_10ms_envelope_correlation": 0.9868604654425986, "全帯域_10ms_modulation_mae_db": 0.2282669833131815, "全帯域_40ms_log_rms_mae_db": 0.6066650993154348, "全帯域_40ms_rms_envelope_nrmse": 0.056148157820140224, "全帯域_40ms_envelope_correlation": 0.9965172536895851, "全帯域_40ms_modulation_mae_db": 0.20502854262306972, "300–1000 Hz_10ms_log_rms_mae_db": 1.434415157044287, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14847768526868996, "300–1000 Hz_10ms_envelope_correlation": 0.9817993245736957, "300–1000 Hz_10ms_modulation_mae_db": 0.6482556087047676, "300–1000 Hz_40ms_log_rms_mae_db": 0.9185817467045608, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09256105930124799, "300–1000 Hz_40ms_envelope_correlation": 0.9934595587214327, "300–1000 Hz_40ms_modulation_mae_db": 0.6558442059840844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5648167740464847, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15909591955110355, "1000–3000 Hz_10ms_envelope_correlation": 0.9801484409041942, "1000–3000 Hz_10ms_modulation_mae_db": 1.3922066265098438, "1000–3000 Hz_40ms_log_rms_mae_db": 1.086069532000467, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09354146234981259, "1000–3000 Hz_40ms_envelope_correlation": 0.9928139443437426, "1000–3000 Hz_40ms_modulation_mae_db": 1.4570790512145724, "3000–6000 Hz_10ms_log_rms_mae_db": 2.347871249963629, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3664463047837453, "3000–6000 Hz_10ms_envelope_correlation": 0.9048131459839912, "3000–6000 Hz_10ms_modulation_mae_db": 1.5749545319593465, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8789312886369074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29797439005857995, "3000–6000 Hz_40ms_envelope_correlation": 0.9311452690677295, "3000–6000 Hz_40ms_modulation_mae_db": 1.2949718840564268, "mr_stft_db": 5.769553363616155, "band_mae_db": 0.23426322041297507, "rms_gain_error_db": -0.12341657414756428, "output_peak": 0.9658179514113899}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1426000321992842, "全帯域_10ms_rms_envelope_nrmse": 0.10895402685292643, "全帯域_10ms_envelope_correlation": 0.9876130631403452, "全帯域_10ms_modulation_mae_db": 1.1078435673530809, "全帯域_40ms_log_rms_mae_db": 0.6567522142144596, "全帯域_40ms_rms_envelope_nrmse": 0.06153277906352629, "全帯域_40ms_envelope_correlation": 0.9957663411221849, "全帯域_40ms_modulation_mae_db": 0.886895210699361, "300–1000 Hz_10ms_log_rms_mae_db": 1.3575729539957029, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13406883313205295, "300–1000 Hz_10ms_envelope_correlation": 0.9844355925758903, "300–1000 Hz_10ms_modulation_mae_db": 1.0643490532509623, "300–1000 Hz_40ms_log_rms_mae_db": 0.8271886448757855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0739285478636723, "300–1000 Hz_40ms_envelope_correlation": 0.9948969256944613, "300–1000 Hz_40ms_modulation_mae_db": 0.8881348415208291, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6132520535009593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1940055722514078, "1000–3000 Hz_10ms_envelope_correlation": 0.9705714818891814, "1000–3000 Hz_10ms_modulation_mae_db": 2.7976267336376464, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1406893553024702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12620028746500192, "1000–3000 Hz_40ms_envelope_correlation": 0.9866886386118804, "1000–3000 Hz_40ms_modulation_mae_db": 3.022197385072437, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3195853293734303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3663211283543467, "3000–6000 Hz_10ms_envelope_correlation": 0.9033136383100753, "3000–6000 Hz_10ms_modulation_mae_db": 1.5329487468361542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.858568241760234, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2966213377984049, "3000–6000 Hz_40ms_envelope_correlation": 0.9304488989361772, "3000–6000 Hz_40ms_modulation_mae_db": 1.5616252707757996, "mr_stft_db": 5.762553337755903, "band_mae_db": 0.11116745832536852, "rms_gain_error_db": 0.07543788245750482, "output_peak": 0.8686685917584048}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1140364374376914, "全帯域_10ms_rms_envelope_nrmse": 0.1014762833471163, "全帯域_10ms_envelope_correlation": 0.9890149909507421, "全帯域_10ms_modulation_mae_db": 0.9485914900542763, "全帯域_40ms_log_rms_mae_db": 0.601884630781093, "全帯域_40ms_rms_envelope_nrmse": 0.04700943566985249, "全帯域_40ms_envelope_correlation": 0.9974318643988389, "全帯域_40ms_modulation_mae_db": 0.7211877395826732, "300–1000 Hz_10ms_log_rms_mae_db": 1.3772460941620275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13697301731531547, "300–1000 Hz_10ms_envelope_correlation": 0.9838523232754115, "300–1000 Hz_10ms_modulation_mae_db": 0.8237026007609655, "300–1000 Hz_40ms_log_rms_mae_db": 0.8591701157622883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0691944046132703, "300–1000 Hz_40ms_envelope_correlation": 0.9958261387561993, "300–1000 Hz_40ms_modulation_mae_db": 0.6628123006161979, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5811191505018163, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17676311485494858, "1000–3000 Hz_10ms_envelope_correlation": 0.9754606078709549, "1000–3000 Hz_10ms_modulation_mae_db": 1.3077424180687398, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1375394605383589, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11702709775110938, "1000–3000 Hz_40ms_envelope_correlation": 0.9885992769415358, "1000–3000 Hz_40ms_modulation_mae_db": 1.153641647227852, "3000–6000 Hz_10ms_log_rms_mae_db": 2.270100854478018, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36453610363494465, "3000–6000 Hz_10ms_envelope_correlation": 0.9034866687615197, "3000–6000 Hz_10ms_modulation_mae_db": 1.4216977707144156, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8123973526730197, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2979828118837486, "3000–6000 Hz_40ms_envelope_correlation": 0.9292377214966572, "3000–6000 Hz_40ms_modulation_mae_db": 1.4127298418393515, "mr_stft_db": 5.759300534414472, "band_mae_db": 0.19205358314613494, "rms_gain_error_db": -0.022845007703552012, "output_peak": 0.7549906912467685}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0743856709155515, "全帯域_10ms_rms_envelope_nrmse": 0.0861864669875795, "全帯域_10ms_envelope_correlation": 0.9921234697008198, "全帯域_10ms_modulation_mae_db": 0.25885527758228166, "全帯域_40ms_log_rms_mae_db": 0.5940183744984514, "全帯域_40ms_rms_envelope_nrmse": 0.04347505971718275, "全帯域_40ms_envelope_correlation": 0.9979223241222547, "全帯域_40ms_modulation_mae_db": 0.436260569856689, "300–1000 Hz_10ms_log_rms_mae_db": 1.4573005264346273, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12340794397849364, "300–1000 Hz_10ms_envelope_correlation": 0.9872375112536855, "300–1000 Hz_10ms_modulation_mae_db": 0.6134240387152727, "300–1000 Hz_40ms_log_rms_mae_db": 0.9157593034111302, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08264122169227205, "300–1000 Hz_40ms_envelope_correlation": 0.9942292333383175, "300–1000 Hz_40ms_modulation_mae_db": 0.6456854904654364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6410205900283594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17827317615947916, "1000–3000 Hz_10ms_envelope_correlation": 0.9752349559053178, "1000–3000 Hz_10ms_modulation_mae_db": 1.1074207149966313, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1897632203921582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11526086638279052, "1000–3000 Hz_40ms_envelope_correlation": 0.9895745887471001, "1000–3000 Hz_40ms_modulation_mae_db": 1.0626296387290315, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3077851226981143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3795210210120559, "3000–6000 Hz_10ms_envelope_correlation": 0.9004568272912161, "3000–6000 Hz_10ms_modulation_mae_db": 2.0429605161924194, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8968334384572376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3114181718700836, "3000–6000 Hz_40ms_envelope_correlation": 0.9268536978989111, "3000–6000 Hz_40ms_modulation_mae_db": 2.2504483975094085, "mr_stft_db": 5.754879353298225, "band_mae_db": 0.26460215996887115, "rms_gain_error_db": -0.0993896683875555, "output_peak": 0.7650964147389925}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0753997126657429, "全帯域_10ms_rms_envelope_nrmse": 0.10095002323703306, "全帯域_10ms_envelope_correlation": 0.9891137787201237, "全帯域_10ms_modulation_mae_db": 0.38585164279935497, "全帯域_40ms_log_rms_mae_db": 0.5836594612225207, "全帯域_40ms_rms_envelope_nrmse": 0.0499802891432103, "全帯域_40ms_envelope_correlation": 0.9971143587564014, "全帯域_40ms_modulation_mae_db": 0.4767818665266068, "300–1000 Hz_10ms_log_rms_mae_db": 1.3996410101325274, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1341827106474718, "300–1000 Hz_10ms_envelope_correlation": 0.9845911608272214, "300–1000 Hz_10ms_modulation_mae_db": 0.2474586840393456, "300–1000 Hz_40ms_log_rms_mae_db": 0.8452552746757833, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06793869533662236, "300–1000 Hz_40ms_envelope_correlation": 0.996095367948618, "300–1000 Hz_40ms_modulation_mae_db": 0.1455207550499082, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6564394794871298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18015237391949546, "1000–3000 Hz_10ms_envelope_correlation": 0.9745003534382803, "1000–3000 Hz_10ms_modulation_mae_db": 2.5201205237456366, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1781237720408586, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1007960261872913, "1000–3000 Hz_40ms_envelope_correlation": 0.9916149394169785, "1000–3000 Hz_40ms_modulation_mae_db": 2.516054149663672, "3000–6000 Hz_10ms_log_rms_mae_db": 2.287102346167092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36607698655748605, "3000–6000 Hz_10ms_envelope_correlation": 0.9035417457121426, "3000–6000 Hz_10ms_modulation_mae_db": 1.5849277057800488, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8416727528105015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3003295334292098, "3000–6000 Hz_40ms_envelope_correlation": 0.9288229154217105, "3000–6000 Hz_40ms_modulation_mae_db": 1.3309639881910489, "mr_stft_db": 5.774339869261873, "band_mae_db": 0.16344298425336048, "rms_gain_error_db": -0.04429578488926681, "output_peak": 0.9037959935855174}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1093058632571446, "全帯域_10ms_rms_envelope_nrmse": 0.09606061094391044, "全帯域_10ms_envelope_correlation": 0.9901273739166947, "全帯域_10ms_modulation_mae_db": 0.5269495577953017, "全帯域_40ms_log_rms_mae_db": 0.5812607676622015, "全帯域_40ms_rms_envelope_nrmse": 0.04172161731106885, "全帯域_40ms_envelope_correlation": 0.9979900874975214, "全帯域_40ms_modulation_mae_db": 0.31514360365445415, "300–1000 Hz_10ms_log_rms_mae_db": 1.436956631960093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1333515387584654, "300–1000 Hz_10ms_envelope_correlation": 0.9846959652067734, "300–1000 Hz_10ms_modulation_mae_db": 0.5104074234867646, "300–1000 Hz_40ms_log_rms_mae_db": 0.8839212446310979, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07147829045505023, "300–1000 Hz_40ms_envelope_correlation": 0.9955046912529943, "300–1000 Hz_40ms_modulation_mae_db": 1.2046391076524736, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5994585152698422, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16740209699762704, "1000–3000 Hz_10ms_envelope_correlation": 0.9780626262139873, "1000–3000 Hz_10ms_modulation_mae_db": 1.4141780500605525, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1630669831592813, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10031019918182552, "1000–3000 Hz_40ms_envelope_correlation": 0.9918594982476887, "1000–3000 Hz_40ms_modulation_mae_db": 1.045698310968515, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3641274230330276, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36609041620174193, "3000–6000 Hz_10ms_envelope_correlation": 0.9031787802437986, "3000–6000 Hz_10ms_modulation_mae_db": 1.4861352786757067, "3000–6000 Hz_40ms_log_rms_mae_db": 1.9035640037176096, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3004496266178786, "3000–6000 Hz_40ms_envelope_correlation": 0.9285048934929726, "3000–6000 Hz_40ms_modulation_mae_db": 1.2757408572468414, "mr_stft_db": 5.7792468970548105, "band_mae_db": 0.15575750639916405, "rms_gain_error_db": -0.04168288628972143, "output_peak": 0.7749933803497092}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0534382694211062, "全帯域_10ms_rms_envelope_nrmse": 0.10037186976437809, "全帯域_10ms_envelope_correlation": 0.9892532918090384, "全帯域_10ms_modulation_mae_db": 0.8138793460080033, "全帯域_40ms_log_rms_mae_db": 0.5001578328602808, "全帯域_40ms_rms_envelope_nrmse": 0.053282307843914335, "全帯域_40ms_envelope_correlation": 0.9966855727406014, "全帯域_40ms_modulation_mae_db": 0.5997478596666633, "300–1000 Hz_10ms_log_rms_mae_db": 1.38631115950554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12952047458961471, "300–1000 Hz_10ms_envelope_correlation": 0.9855122870856745, "300–1000 Hz_10ms_modulation_mae_db": 0.9826141564747392, "300–1000 Hz_40ms_log_rms_mae_db": 0.828677429118519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07317916238033882, "300–1000 Hz_40ms_envelope_correlation": 0.9951315933934325, "300–1000 Hz_40ms_modulation_mae_db": 0.6584205923162058, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6367768245176495, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17979251585806474, "1000–3000 Hz_10ms_envelope_correlation": 0.9747315788519887, "1000–3000 Hz_10ms_modulation_mae_db": 1.7161356451698184, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1923094955023477, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10049871359083948, "1000–3000 Hz_40ms_envelope_correlation": 0.9920745036432919, "1000–3000 Hz_40ms_modulation_mae_db": 1.9647466816909473, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3308952571591948, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3709894232100972, "3000–6000 Hz_10ms_envelope_correlation": 0.9066778509114174, "3000–6000 Hz_10ms_modulation_mae_db": 1.2342932869821372, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8223320341365017, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3028006045381248, "3000–6000 Hz_40ms_envelope_correlation": 0.9324725795797038, "3000–6000 Hz_40ms_modulation_mae_db": 1.5033987446717798, "mr_stft_db": 5.772004488359472, "band_mae_db": 0.21714889545712188, "rms_gain_error_db": -0.005720659816518842, "output_peak": 0.8065810096249595}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.097498098965913, "全帯域_10ms_rms_envelope_nrmse": 0.10696200796960316, "全帯域_10ms_envelope_correlation": 0.9878090304085783, "全帯域_10ms_modulation_mae_db": 1.0177386406953264, "全帯域_40ms_log_rms_mae_db": 0.5513674669056723, "全帯域_40ms_rms_envelope_nrmse": 0.05744773223137391, "全帯域_40ms_envelope_correlation": 0.9961799925863213, "全帯域_40ms_modulation_mae_db": 0.9984239957810944, "300–1000 Hz_10ms_log_rms_mae_db": 1.4655622007034947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1373907000074968, "300–1000 Hz_10ms_envelope_correlation": 0.9837251837963906, "300–1000 Hz_10ms_modulation_mae_db": 1.0665104416643418, "300–1000 Hz_40ms_log_rms_mae_db": 0.9295032808372132, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06864698064711104, "300–1000 Hz_40ms_envelope_correlation": 0.995833935927734, "300–1000 Hz_40ms_modulation_mae_db": 0.9428745248469319, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6055400488198408, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1549111837506004, "1000–3000 Hz_10ms_envelope_correlation": 0.9814733108001694, "1000–3000 Hz_10ms_modulation_mae_db": 1.2183773188037763, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1733332098755842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09529069614907495, "1000–3000 Hz_40ms_envelope_correlation": 0.9930890911651831, "1000–3000 Hz_40ms_modulation_mae_db": 1.4043291749889502, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2710942429980747, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3552239825683285, "3000–6000 Hz_10ms_envelope_correlation": 0.9126127537200917, "3000–6000 Hz_10ms_modulation_mae_db": 1.3061217685607824, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8098299189427132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2940851712000354, "3000–6000 Hz_40ms_envelope_correlation": 0.9346949032856208, "3000–6000 Hz_40ms_modulation_mae_db": 1.4383086366359727, "mr_stft_db": 5.7724074175107845, "band_mae_db": 0.2489235985481603, "rms_gain_error_db": -0.02877434376982643, "output_peak": 0.8242850427569186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.990135014512973, "spectral_convergence": 0.3065165426705528, "spectral_convergence_status": "ok", "amplitude_floor": 2.5091999427961322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.507901607214368, "spectral_convergence": 0.2892657106999899, "spectral_convergence_status": "ok", "amplitude_floor": 1.1904358876245626e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8106234691211265, "spectral_convergence": 0.31448447559653886, "spectral_convergence_status": "ok", "amplitude_floor": 8.234850781853941e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001792776939061309, "output_power": 0.001886621046646034, "delta_db": 0.22158417879982845, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0038916933601920887, "output_power": 0.003826828371002896, "delta_db": -0.07299628362595018, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002833146281043118, "output_power": 0.0025685685863025528, "delta_db": -0.4257783178589948, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00044617096434630785, "output_power": 0.00042388554071307884, "delta_db": -0.22252701633070585, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.1707528478122786e-05, "output_power": 5.403573829714696e-05, "delta_db": 0.19127309850106713, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2019368729027054e-05, "output_power": 3.408433366729358e-05, "delta_db": 0.2714204273613039, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23426322041297507, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12341657414756428, "pitch": null, "output_rms_dbfs": -20.558123724031574, "output_peak": 0.9658179514113899, "reference_rms_dbfs": -20.434707149884012, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.769553363616155}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003179454955178215, "log_rms_mae_db": 1.1048439767547764, "rms_envelope_nrmse": 0.11101402023226531, "envelope_correlation": 0.9868604654425986, "rms_gain_error_db": -0.12341609665998302, "reference_mean_power": 0.009055057750136907, "output_mean_power": 0.008801356546204212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4928013782154505e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002203466496672066, "output_power": 0.0020477704331425114, "reference_active": true, "delta_db": -0.31825183403110074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002920043469136947, "output_power": 0.00030007601603027507, "reference_active": true, "delta_db": 0.1184196864712292}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.410659541472349e-05, "output_power": 7.846388441863713e-05, "reference_active": true, "delta_db": 0.24812942943721453}], "active_band_count": 3, "active_band_mae_db": 0.2282669833131815}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002752533875730688, "log_rms_mae_db": 0.6066650993154348, "rms_envelope_nrmse": 0.056148157820140224, "envelope_correlation": 0.9965172536895851, "rms_gain_error_db": -0.12341389548063388, "reference_mean_power": 0.009100511091100074, "output_mean_power": 0.008845540875823002}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.115339465742665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001970497325835552, "output_power": 0.0018491311486283463, "reference_active": true, "delta_db": -0.27608135467239786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014191960918623396, "output_power": 0.000137362950939451, "reference_active": true, "delta_db": -0.1417279443248243}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.976197244793298e-06, "output_power": 3.7996210687464454e-06, "reference_active": true, "delta_db": -0.19727632887198704}], "active_band_count": 3, "active_band_mae_db": 0.20502854262306972}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028904791382782896, "log_rms_mae_db": 1.434415157044287, "rms_envelope_nrmse": 0.14847768526868996, "envelope_correlation": 0.9817993245736957, "rms_gain_error_db": -0.3790682691877808, "reference_mean_power": 0.00500094100859156, "output_mean_power": 0.004582947737286724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3504006393160654e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001701857204603173, "output_power": 0.0013846151758818378, "reference_active": true, "delta_db": -0.8959403020046907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00029604441935142547, "output_power": 0.0002501009147410514, "reference_active": true, "delta_db": -0.7324159857407786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.422561443844994e-05, "output_power": 7.983528044231528e-05, "reference_active": true, "delta_db": 0.31641053836883337}], "active_band_count": 3, "active_band_mae_db": 0.6482556087047676}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002500254339817177, "log_rms_mae_db": 0.9185817467045608, "rms_envelope_nrmse": 0.09256105930124799, "envelope_correlation": 0.9934595587214327, "rms_gain_error_db": -0.3790418473200703, "reference_mean_power": 0.00502600851498449, "output_mean_power": 0.004605948050059954}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0757929049004627e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015562264230646289, "output_power": 0.001274445370596933, "reference_active": true, "delta_db": -0.8675156083063101}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001439985744956168, "output_power": 0.00011552073100345242, "reference_active": true, "delta_db": -0.956982644413434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7762361828128385e-06, "output_power": 4.621493619749181e-06, "reference_active": true, "delta_db": -0.1430343652325094}], "active_band_count": 3, "active_band_mae_db": 0.6558442059840844}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010495143398294122, "log_rms_mae_db": 1.5648167740464847, "rms_envelope_nrmse": 0.15909591955110355, "envelope_correlation": 0.9801484409041942, "rms_gain_error_db": -0.1345151464904824, "reference_mean_power": 0.0004146271168721916, "output_mean_power": 0.0004019816117491756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0723455289073609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010117430747443104, "output_power": 0.00010153940142335581, "reference_active": true, "delta_db": 0.015643585201631165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2721317302684375e-05, "output_power": 1.67780569360159e-05, "reference_active": true, "delta_db": 1.2020957864763004}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6903995233241656e-06, "output_power": 5.317467303367633e-06, "reference_active": true, "delta_db": 2.9588805078515996}], "active_band_count": 3, "active_band_mae_db": 1.3922066265098438}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.780703381807892e-05, "log_rms_mae_db": 1.086069532000467, "rms_envelope_nrmse": 0.09354146234981259, "envelope_correlation": 0.9928139443437426, "rms_gain_error_db": -0.13455607596803304, "reference_mean_power": 0.0004167096352122121, "output_mean_power": 0.0004039968089741029}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9169450940862202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.13104680727751e-05, "output_power": 9.433778496776669e-05, "reference_active": true, "delta_db": 0.1416510582945989}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.750575155696485e-06, "output_power": 8.177218836405289e-06, "reference_active": true, "delta_db": 1.5289433664806813}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0115145688994735e-07, "output_power": 3.7461697728019706e-07, "reference_active": true, "delta_db": 2.7006427288684365}], "active_band_count": 3, "active_band_mae_db": 1.4570790512145724}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4613391249073974e-05, "log_rms_mae_db": 2.347871249963629, "rms_envelope_nrmse": 0.3664463047837453, "envelope_correlation": 0.9048131459839912, "rms_gain_error_db": 0.3567958399915585, "reference_mean_power": 2.714804754964509e-05, "output_mean_power": 2.947258203368454e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3627974020623764e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.8436688435136e-06, "output_power": 4.282761815970279e-06, "reference_active": true, "delta_db": -0.5345052127431449}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.21772084491833e-06, "output_power": 2.079776114886347e-06, "reference_active": true, "delta_db": -0.2789029227971145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.650596314175208e-07, "output_power": 1.144601179495254e-06, "reference_active": true, "delta_db": 3.9114554603377796}], "active_band_count": 3, "active_band_mae_db": 1.5749545319593465}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2268327401512992e-05, "log_rms_mae_db": 1.8789312886369074, "rms_envelope_nrmse": 0.29797439005857995, "envelope_correlation": 0.9311452690677295, "rms_gain_error_db": 0.356778713834405, "reference_mean_power": 2.7284479980131438e-05, "output_mean_power": 2.96205795978487e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2818152654967464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7409144698248785e-06, "output_power": 4.4005239044803735e-06, "reference_active": true, "delta_db": -0.3235773583920827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1748833712187593e-06, "output_power": 1.1915234388512165e-06, "reference_active": true, "delta_db": 0.061078328835136614}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.424838593381526e-08, "output_power": 7.667717524427991e-08, "reference_active": true, "delta_db": 3.500259964942061}], "active_band_count": 3, "active_band_mae_db": 1.2949718840564268}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8603299370347411, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "77fabeadaa09fa7ee481273625e54574a8c9321918f37cb7fd90f1a15b78013c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9762729194842593, "standard_deviation": 0.003485236553241047, "minimum": 0.9705714818891814, "maximum": 0.9814733108001694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.61230292952151, "standard_deviation": 0.03119223806623684, "minimum": 1.5648167740464847, "maximum": 1.6564394794871298, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6842260038740808, "standard_deviation": 0.6312187193665927, "minimum": 1.1074207149966313, "maximum": 2.7976267336376464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17379949416784085, "standard_deviation": 0.01268997313550291, "minimum": 0.1549111837506004, "maximum": 0.1940055722514078, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9907893101396752, "standard_deviation": 0.0022663414667311974, "minimum": 0.9866886386118804, "maximum": 0.9930890911651831, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1576118786014407, "standard_deviation": 0.035340932786746236, "minimum": 1.086069532000467, "maximum": 1.1923094955023477, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.703297004944497, "standard_deviation": 0.7332359876670759, "minimum": 1.045698310968515, "maximum": 3.022197385072437, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1061156686322182, "standard_deviation": 0.011801505891098012, "minimum": 0.09354146234981259, "maximum": 0.12620028746500192, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9047601763667816, "standard_deviation": 0.0036166627617082244, "minimum": 0.9004568272912161, "maximum": 0.9126127537200917, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3123202282338227, "standard_deviation": 0.03481428300955363, "minimum": 2.270100854478018, "maximum": 2.3641274230330276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5230049507126264, "standard_deviation": 0.24465751941708347, "minimum": 1.2342932869821372, "maximum": 2.0429605161924194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3669006707903433, "standard_deviation": 0.006759169939022887, "minimum": 0.3552239825683285, "maximum": 0.3795210210120559, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9302726098974353, "standard_deviation": 0.0024835327914093787, "minimum": 0.9268536978989111, "maximum": 0.9346949032856208, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8530161288918405, "standard_deviation": 0.03734508055579023, "minimum": 1.8098299189427132, "maximum": 1.9035640037176096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5085234526158287, "standard_deviation": 0.31593955081293307, "minimum": 1.2757408572468414, "maximum": 2.2504483975094085, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3002077059245082, "standard_deviation": 0.005247534911584707, "minimum": 0.2940851712000354, "maximum": 0.3114181718700836, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9844811685743429, "standard_deviation": 0.0015528862551160902, "minimum": 0.9817993245736957, "maximum": 0.9872375112536855, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4143757167422875, "standard_deviation": 0.039612346573643924, "minimum": 1.3575729539957029, "maximum": 1.4655622007034947, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7445902508871449, "standard_deviation": 0.2921004656812953, "minimum": 0.2474586840393456, "maximum": 1.0665104416643418, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1346716129622001, "standard_deviation": 0.007161872852591823, "minimum": 0.12340794397849364, "maximum": 0.14847768526868996, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9951221806291488, "standard_deviation": 0.0009030152224462834, "minimum": 0.9934595587214327, "maximum": 0.996095367948618, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8760071300020473, "standard_deviation": 0.041695856890646546, "minimum": 0.8271886448757855, "maximum": 0.9295032808372132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7254914773065084, "standard_deviation": 0.30689776435535493, "minimum": 0.1455207550499082, "maximum": 1.2046391076524736, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07494604528619814, "standard_deviation": 0.008515108400631268, "minimum": 0.06793869533662236, "maximum": 0.09256105930124799, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19841992581389456, "standard_deviation": 0.05245844208521418, "minimum": 0.11116745832536852, "maximum": 0.26460215996887115, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.768035657658961, "standard_deviation": 0.00829844098505737, "minimum": 5.754879353298225, "maximum": 5.7792468970548105, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8330286344340825, "standard_deviation": 0.07441193587279128, "minimum": 0.7549906912467685, "maximum": 0.9658179514113899, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.03633588031831256, "standard_deviation": 0.060143428843796636, "minimum": -0.12341657414756428, "maximum": 0.07543788245750482, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9889894330111175, "standard_deviation": 0.0016485092503275428, "minimum": 0.9868604654425986, "maximum": 0.9921234697008198, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0964385077021512, "standard_deviation": 0.027917981805493278, "minimum": 1.0534382694211062, "maximum": 1.1426000321992842, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6609970632001009, "standard_deviation": 0.35368811361610647, "minimum": 0.2282669833131815, "maximum": 1.1078435673530809, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10149691366685154, "standard_deviation": 0.007942467352641598, "minimum": 0.0861864669875795, "maximum": 0.11101402023226531, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969509743642136, "standard_deviation": 0.0008051216950294228, "minimum": 0.9957663411221849, "maximum": 0.9979900874975214, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5844707309325143, "standard_deviation": 0.04524071666771962, "minimum": 0.5001578328602808, "maximum": 0.6567522142144596, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5799336735488265, "standard_deviation": 0.27559150478304956, "minimum": 0.20502854262306972, "maximum": 0.9984239957810944, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05132467235003364, "standard_deviation": 0.00700060581227217, "minimum": 0.04172161731106885, "maximum": 0.06153277906352629, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8230-279154-0033", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7200400829315186, "predicted_mos": 2.7200400829315186, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "adf92cfccbece857df5ff349db32ebed20feaf272dac624a4d195d47a12f09b5", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6897774999961257, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0704506612541707, "全帯域_10ms_rms_envelope_nrmse": 0.09568196609963586, "全帯域_10ms_envelope_correlation": 0.9902368012948909, "全帯域_10ms_modulation_mae_db": 0.3628183129971277, "全帯域_40ms_log_rms_mae_db": 0.5642899307267729, "全帯域_40ms_rms_envelope_nrmse": 0.04274677885400865, "全帯域_40ms_envelope_correlation": 0.9979665826600087, "全帯域_40ms_modulation_mae_db": 0.12778861118375742, "300–1000 Hz_10ms_log_rms_mae_db": 1.37557208606782, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11994226090484933, "300–1000 Hz_10ms_envelope_correlation": 0.9881997233162632, "300–1000 Hz_10ms_modulation_mae_db": 0.6529350683504558, "300–1000 Hz_40ms_log_rms_mae_db": 0.8018877105383593, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07186273070144888, "300–1000 Hz_40ms_envelope_correlation": 0.9961156354144624, "300–1000 Hz_40ms_modulation_mae_db": 1.0282614788560551, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5768250025189612, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1836798364129774, "1000–3000 Hz_10ms_envelope_correlation": 0.9735218875433035, "1000–3000 Hz_10ms_modulation_mae_db": 1.7625051269701713, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0943191377132462, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10448149433269029, "1000–3000 Hz_40ms_envelope_correlation": 0.99135620482397, "1000–3000 Hz_40ms_modulation_mae_db": 1.647196251012656, "3000–6000 Hz_10ms_log_rms_mae_db": 2.279920846054336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3814423667434252, "3000–6000 Hz_10ms_envelope_correlation": 0.9068249581088884, "3000–6000 Hz_10ms_modulation_mae_db": 1.5316834181122971, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8597058678454808, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.31760046866984293, "3000–6000 Hz_40ms_envelope_correlation": 0.9306419814138092, "3000–6000 Hz_40ms_modulation_mae_db": 1.711061901023914, "mr_stft_db": 5.733515775278367, "band_mae_db": 0.29682247676268847, "rms_gain_error_db": -0.08744835095657785, "output_peak": 0.7766753515388393}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0901161153669916, "全帯域_10ms_rms_envelope_nrmse": 0.09502242038467186, "全帯域_10ms_envelope_correlation": 0.9904161849111464, "全帯域_10ms_modulation_mae_db": 0.741850291964826, "全帯域_40ms_log_rms_mae_db": 0.5993975688065034, "全帯域_40ms_rms_envelope_nrmse": 0.052580445834147456, "全帯域_40ms_envelope_correlation": 0.9969255115176354, "全帯域_40ms_modulation_mae_db": 0.7537463416537724, "300–1000 Hz_10ms_log_rms_mae_db": 1.4214406883180304, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14199737076040778, "300–1000 Hz_10ms_envelope_correlation": 0.983163321614202, "300–1000 Hz_10ms_modulation_mae_db": 1.0425295918786208, "300–1000 Hz_40ms_log_rms_mae_db": 0.8725544685996088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08476900931822885, "300–1000 Hz_40ms_envelope_correlation": 0.994360691104441, "300–1000 Hz_40ms_modulation_mae_db": 0.717315791163537, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5958585142390997, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1603659375864706, "1000–3000 Hz_10ms_envelope_correlation": 0.9804306358413866, "1000–3000 Hz_10ms_modulation_mae_db": 1.071032486292122, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1359880775289075, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10791325821435496, "1000–3000 Hz_40ms_envelope_correlation": 0.9906144517216471, "1000–3000 Hz_40ms_modulation_mae_db": 1.0365014138171738, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2588621771818493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36757653963057907, "3000–6000 Hz_10ms_envelope_correlation": 0.8985789758044698, "3000–6000 Hz_10ms_modulation_mae_db": 1.767721112305603, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8336687149248074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3050309963499807, "3000–6000 Hz_40ms_envelope_correlation": 0.9232601739047782, "3000–6000 Hz_40ms_modulation_mae_db": 1.6142883428036534, "mr_stft_db": 5.725729816310257, "band_mae_db": 0.18049457860276855, "rms_gain_error_db": -0.11214078697143864, "output_peak": 0.7757711216942065}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1009581802361172, "全帯域_10ms_rms_envelope_nrmse": 0.10945967048506737, "全帯域_10ms_envelope_correlation": 0.9871683019581908, "全帯域_10ms_modulation_mae_db": 0.8266712553149023, "全帯域_40ms_log_rms_mae_db": 0.5611397748058599, "全帯域_40ms_rms_envelope_nrmse": 0.05415842563906859, "全帯域_40ms_envelope_correlation": 0.9966149550375494, "全帯域_40ms_modulation_mae_db": 0.5725012610868719, "300–1000 Hz_10ms_log_rms_mae_db": 1.458935962770002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15225983725109665, "300–1000 Hz_10ms_envelope_correlation": 0.9802450140606653, "300–1000 Hz_10ms_modulation_mae_db": 0.8760635202790983, "300–1000 Hz_40ms_log_rms_mae_db": 0.8493908447526135, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08395123107505155, "300–1000 Hz_40ms_envelope_correlation": 0.9941974110212987, "300–1000 Hz_40ms_modulation_mae_db": 0.4120066622002903, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6065781452360608, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1588705578527977, "1000–3000 Hz_10ms_envelope_correlation": 0.9802232566542305, "1000–3000 Hz_10ms_modulation_mae_db": 1.7563780827322095, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1241206748855754, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10193191628695147, "1000–3000 Hz_40ms_envelope_correlation": 0.9915381585921107, "1000–3000 Hz_40ms_modulation_mae_db": 1.5521162932402912, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2797716236476795, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3702357791522299, "3000–6000 Hz_10ms_envelope_correlation": 0.8988188178784505, "3000–6000 Hz_10ms_modulation_mae_db": 1.7170442394662386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.812701445341004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30539230803023376, "3000–6000 Hz_40ms_envelope_correlation": 0.9243952952745644, "3000–6000 Hz_40ms_modulation_mae_db": 1.3992105100053884, "mr_stft_db": 5.728519438903315, "band_mae_db": 0.19973523002408247, "rms_gain_error_db": -0.0648296236274762, "output_peak": 0.8925261812346643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0759193743010094, "全帯域_10ms_rms_envelope_nrmse": 0.10306576592385845, "全帯域_10ms_envelope_correlation": 0.9886472321020041, "全帯域_10ms_modulation_mae_db": 0.3976385276355863, "全帯域_40ms_log_rms_mae_db": 0.5126396418541683, "全帯域_40ms_rms_envelope_nrmse": 0.046453063675974744, "全帯域_40ms_envelope_correlation": 0.9975462573360027, "全帯域_40ms_modulation_mae_db": 0.4259647844641566, "300–1000 Hz_10ms_log_rms_mae_db": 1.4169870084821603, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1265185174497539, "300–1000 Hz_10ms_envelope_correlation": 0.9866155141308846, "300–1000 Hz_10ms_modulation_mae_db": 0.8652512494682855, "300–1000 Hz_40ms_log_rms_mae_db": 0.8481608828406862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06927417901281734, "300–1000 Hz_40ms_envelope_correlation": 0.9962873199915189, "300–1000 Hz_40ms_modulation_mae_db": 1.105317875011572, "1000–3000 Hz_10ms_log_rms_mae_db": 1.551629741002709, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1549874138623159, "1000–3000 Hz_10ms_envelope_correlation": 0.9813810196992596, "1000–3000 Hz_10ms_modulation_mae_db": 0.5165861697331499, "1000–3000 Hz_40ms_log_rms_mae_db": 1.134980318677282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09183356262367307, "1000–3000 Hz_40ms_envelope_correlation": 0.9934626936745817, "1000–3000 Hz_40ms_modulation_mae_db": 0.4012938419941974, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2760540290351647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3642897738339056, "3000–6000 Hz_10ms_envelope_correlation": 0.9060839363041425, "3000–6000 Hz_10ms_modulation_mae_db": 1.2856478432045495, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7984567231682993, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2990870015173017, "3000–6000 Hz_40ms_envelope_correlation": 0.9308114251480469, "3000–6000 Hz_40ms_modulation_mae_db": 1.1964150454783384, "mr_stft_db": 5.733583698456184, "band_mae_db": 0.24969464960692309, "rms_gain_error_db": -0.07449814070694392, "output_peak": 0.9133117256055817}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0961951925691504, "全帯域_10ms_rms_envelope_nrmse": 0.12065699600274452, "全帯域_10ms_envelope_correlation": 0.9847169459408192, "全帯域_10ms_modulation_mae_db": 1.5996971865298775, "全帯域_40ms_log_rms_mae_db": 0.5904268980481708, "全帯域_40ms_rms_envelope_nrmse": 0.06054333686081602, "全帯域_40ms_envelope_correlation": 0.9958653061841225, "全帯域_40ms_modulation_mae_db": 2.0224360423785437, "300–1000 Hz_10ms_log_rms_mae_db": 1.4247944536884158, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16910614658205855, "300–1000 Hz_10ms_envelope_correlation": 0.9754452764655825, "300–1000 Hz_10ms_modulation_mae_db": 1.4341809974179431, "300–1000 Hz_40ms_log_rms_mae_db": 0.8490043259171731, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08586472765653064, "300–1000 Hz_40ms_envelope_correlation": 0.9931669879639178, "300–1000 Hz_40ms_modulation_mae_db": 1.5591026629667535, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6008352305449702, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16353608511956025, "1000–3000 Hz_10ms_envelope_correlation": 0.9790151573832905, "1000–3000 Hz_10ms_modulation_mae_db": 1.174357207225478, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1286811426174312, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10242184101862069, "1000–3000 Hz_40ms_envelope_correlation": 0.9913712334789213, "1000–3000 Hz_40ms_modulation_mae_db": 0.9698648728433943, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3059541265119576, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3852432949307441, "3000–6000 Hz_10ms_envelope_correlation": 0.8970073921286126, "3000–6000 Hz_10ms_modulation_mae_db": 1.7566399145299876, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8232896685736726, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.3154237177444958, "3000–6000 Hz_40ms_envelope_correlation": 0.9244097776400327, "3000–6000 Hz_40ms_modulation_mae_db": 1.8330485759138835, "mr_stft_db": 5.752321289093625, "band_mae_db": 0.18050986787226644, "rms_gain_error_db": 0.04153306110690631, "output_peak": 1.0644612278367667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 2, "at_or_above_full_scale_fraction": 2.0833333333333333e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 2.0833333333333333e-05, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.087461112246649, "全帯域_10ms_rms_envelope_nrmse": 0.09663724525888576, "全帯域_10ms_envelope_correlation": 0.9900477343007676, "全帯域_10ms_modulation_mae_db": 0.5924342840382673, "全帯域_40ms_log_rms_mae_db": 0.5379728523492989, "全帯域_40ms_rms_envelope_nrmse": 0.04804505993404443, "全帯域_40ms_envelope_correlation": 0.9974006881116793, "全帯域_40ms_modulation_mae_db": 0.7214030592331122, "300–1000 Hz_10ms_log_rms_mae_db": 1.4491808960881927, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13432504454432062, "300–1000 Hz_10ms_envelope_correlation": 0.9847229682632599, "300–1000 Hz_10ms_modulation_mae_db": 1.0810620477186272, "300–1000 Hz_40ms_log_rms_mae_db": 0.893586587955346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07541166531710371, "300–1000 Hz_40ms_envelope_correlation": 0.9953540569768301, "300–1000 Hz_40ms_modulation_mae_db": 1.0219994190555994, "1000–3000 Hz_10ms_log_rms_mae_db": 1.573722536218957, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16229187994725938, "1000–3000 Hz_10ms_envelope_correlation": 0.9793781617852482, "1000–3000 Hz_10ms_modulation_mae_db": 1.1186928249576502, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0932227269329275, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10928181852986461, "1000–3000 Hz_40ms_envelope_correlation": 0.9900666388436957, "1000–3000 Hz_40ms_modulation_mae_db": 1.2643211452839311, "3000–6000 Hz_10ms_log_rms_mae_db": 2.218167056083123, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.35869923216681293, "3000–6000 Hz_10ms_envelope_correlation": 0.9090108363726729, "3000–6000 Hz_10ms_modulation_mae_db": 1.3252308623295488, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7961111575433066, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2963014125572778, "3000–6000 Hz_40ms_envelope_correlation": 0.932049287747907, "3000–6000 Hz_40ms_modulation_mae_db": 1.3049288830228198, "mr_stft_db": 5.7372616011583455, "band_mae_db": 0.21411248950797077, "rms_gain_error_db": -0.08889336052316323, "output_peak": 0.7872720091692161}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.105264804252642, "全帯域_10ms_rms_envelope_nrmse": 0.10297182508963826, "全帯域_10ms_envelope_correlation": 0.9888977887887875, "全帯域_10ms_modulation_mae_db": 0.7000305034835188, "全帯域_40ms_log_rms_mae_db": 0.5884992682139436, "全帯域_40ms_rms_envelope_nrmse": 0.056632091546057875, "全帯域_40ms_envelope_correlation": 0.9963854941114718, "全帯域_40ms_modulation_mae_db": 0.8793286418544021, "300–1000 Hz_10ms_log_rms_mae_db": 1.4269560038362126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13938341001993465, "300–1000 Hz_10ms_envelope_correlation": 0.9832960163994872, "300–1000 Hz_10ms_modulation_mae_db": 0.512659803356821, "300–1000 Hz_40ms_log_rms_mae_db": 0.8758791081542843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07748807459319809, "300–1000 Hz_40ms_envelope_correlation": 0.9944800153943618, "300–1000 Hz_40ms_modulation_mae_db": 0.5668594189581632, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5673854991476308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16390439952555466, "1000–3000 Hz_10ms_envelope_correlation": 0.9790912793755594, "1000–3000 Hz_10ms_modulation_mae_db": 1.257151108146321, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1181849953800103, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10580675552885135, "1000–3000 Hz_40ms_envelope_correlation": 0.9911430590666447, "1000–3000 Hz_40ms_modulation_mae_db": 0.9589562782875225, "3000–6000 Hz_10ms_log_rms_mae_db": 2.257362490331547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.37000882101302845, "3000–6000 Hz_10ms_envelope_correlation": 0.8979968559316754, "3000–6000 Hz_10ms_modulation_mae_db": 1.717717050669133, "3000–6000 Hz_40ms_log_rms_mae_db": 1.803008208400932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.30305845556748084, "3000–6000 Hz_40ms_envelope_correlation": 0.9248057347938367, "3000–6000 Hz_40ms_modulation_mae_db": 1.687597836853983, "mr_stft_db": 5.734933786131534, "band_mae_db": 0.1484813602683757, "rms_gain_error_db": 0.0322491297767262, "output_peak": 0.9073610952014757}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.070044328536437, "全帯域_10ms_rms_envelope_nrmse": 0.09602187839321516, "全帯域_10ms_envelope_correlation": 0.9902093040809937, "全帯域_10ms_modulation_mae_db": 0.20658066512755954, "全帯域_40ms_log_rms_mae_db": 0.548901625612528, "全帯域_40ms_rms_envelope_nrmse": 0.052001614775339336, "全帯域_40ms_envelope_correlation": 0.9969987128967077, "全帯域_40ms_modulation_mae_db": 0.356211795032521, "300–1000 Hz_10ms_log_rms_mae_db": 1.4030015431956049, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13412078168362848, "300–1000 Hz_10ms_envelope_correlation": 0.9850904201046153, "300–1000 Hz_10ms_modulation_mae_db": 0.37629198588923657, "300–1000 Hz_40ms_log_rms_mae_db": 0.8648566176466832, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07657965580967556, "300–1000 Hz_40ms_envelope_correlation": 0.9956059359389137, "300–1000 Hz_40ms_modulation_mae_db": 0.6150844155345015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5880957107103582, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1785776051153554, "1000–3000 Hz_10ms_envelope_correlation": 0.9749658726636998, "1000–3000 Hz_10ms_modulation_mae_db": 2.0856511481058946, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0953215780993135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1137134702426055, "1000–3000 Hz_40ms_envelope_correlation": 0.9894690271923272, "1000–3000 Hz_40ms_modulation_mae_db": 2.2643826745423112, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2885565583107654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.36954123756913704, "3000–6000 Hz_10ms_envelope_correlation": 0.9010719049109123, "3000–6000 Hz_10ms_modulation_mae_db": 1.2400707580710566, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7837625340996368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.29826607811024264, "3000–6000 Hz_40ms_envelope_correlation": 0.9292766315378642, "3000–6000 Hz_40ms_modulation_mae_db": 1.3449785869963862, "mr_stft_db": 5.732320562240164, "band_mae_db": 0.24289776832210597, "rms_gain_error_db": -0.11275977539010659, "output_peak": 0.9850146334555627}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.956904291312623, "spectral_convergence": 0.31010939122781295, "spectral_convergence_status": "ok", "amplitude_floor": 2.5091999427961322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.457709763624732, "spectral_convergence": 0.29245276775290263, "spectral_convergence_status": "ok", "amplitude_floor": 1.1904358876245626e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.785933270897748, "spectral_convergence": 0.3163164980646443, "spectral_convergence_status": "ok", "amplitude_floor": 8.234850781853941e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001792776939061309, "output_power": 0.0019477053940723556, "delta_db": 0.35997009892343396, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0038916933601920887, "output_power": 0.003743419050475414, "delta_db": -0.1687016722997423, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002833146281043118, "output_power": 0.002661523900025846, "delta_db": -0.27138628181424634, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00044617096434630785, "output_power": 0.0004242186457853063, "delta_db": -0.21911550853474632, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.1707528478122786e-05, "output_power": 5.3210795114412476e-05, "delta_db": 0.12445968702414939, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2019368729027054e-05, "output_power": 3.7080273315056755e-05, "delta_db": 0.6373016119798125, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.29682247676268847, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08744835095657785, "pitch": null, "output_rms_dbfs": -20.52215550084059, "output_peak": 0.7766753515388393, "reference_rms_dbfs": -20.434707149884012, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.733515775278367}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003179454955178215, "log_rms_mae_db": 1.0704506612541707, "rms_envelope_nrmse": 0.09568196609963586, "envelope_correlation": 0.9902368012948909, "rms_gain_error_db": -0.08744745248557592, "reference_mean_power": 0.009055057750136907, "output_mean_power": 0.00887455283204736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4928013782154505e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002203466496672066, "output_power": 0.0021565555358351285, "reference_active": true, "delta_db": -0.0934580490229641}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002920043469136947, "output_power": 0.0002868253336749134, "reference_active": true, "delta_db": -0.07771809119418255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.410659541472349e-05, "output_power": 9.153448477859873e-05, "reference_active": true, "delta_db": 0.9172787987742365}], "active_band_count": 3, "active_band_mae_db": 0.3628183129971277}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002752533875730688, "log_rms_mae_db": 0.5642899307267729, "rms_envelope_nrmse": 0.04274677885400865, "envelope_correlation": 0.9979665826600087, "rms_gain_error_db": -0.08748400135211529, "reference_mean_power": 0.009100511091100074, "output_mean_power": 0.008919025039056203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.115339465742665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001970497325835552, "output_power": 0.0019434748418947318, "reference_active": true, "delta_db": -0.05996926719364631}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014191960918623396, "output_power": 0.00013549596234042083, "reference_active": true, "delta_db": -0.20116052730156275}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.976197244793298e-06, "output_power": 4.089700682242799e-06, "reference_active": true, "delta_db": 0.12223603905606326}], "active_band_count": 3, "active_band_mae_db": 0.12778861118375742}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028904791382782896, "log_rms_mae_db": 1.37557208606782, "rms_envelope_nrmse": 0.11994226090484933, "envelope_correlation": 0.9881997233162632, "rms_gain_error_db": -0.2967292381783613, "reference_mean_power": 0.00500094100859156, "output_mean_power": 0.004670665955459929}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3504006393160654e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001701857204603173, "output_power": 0.0014965921614030298, "reference_active": true, "delta_db": -0.5581965130728423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00029604441935142547, "output_power": 0.00022941313899649414, "reference_active": true, "delta_db": -1.1073859145380889}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.422561443844994e-05, "output_power": 7.941016147142973e-05, "reference_active": true, "delta_db": 0.2932227774404361}], "active_band_count": 3, "active_band_mae_db": 0.6529350683504558}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002500254339817177, "log_rms_mae_db": 0.8018877105383593, "rms_envelope_nrmse": 0.07186273070144888, "envelope_correlation": 0.9961156354144624, "rms_gain_error_db": -0.2967633561030898, "reference_mean_power": 0.00502600851498449, "output_mean_power": 0.004694041062762464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0757929049004627e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015562264230646289, "output_power": 0.0013747380894988965, "reference_active": true, "delta_db": -0.5385281918354052}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001439985744956168, "output_power": 0.00010641359916318554, "reference_active": true, "delta_db": -1.3136106052580454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7762361828128385e-06, "output_power": 3.5960128119634444e-06, "reference_active": true, "delta_db": -1.2326456394747147}], "active_band_count": 3, "active_band_mae_db": 1.0282614788560551}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010495143398294122, "log_rms_mae_db": 1.5768250025189612, "rms_envelope_nrmse": 0.1836798364129774, "envelope_correlation": 0.9735218875433035, "rms_gain_error_db": -0.24006630557010494, "reference_mean_power": 0.0004146271168721916, "output_mean_power": 0.000392329596187298}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0723455289073609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010117430747443104, "output_power": 0.0001070361995289533, "reference_active": true, "delta_db": 0.24460440078139054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2721317302684375e-05, "output_power": 1.5918196330295902e-05, "reference_active": true, "delta_db": 0.973617716724412}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6903995233241656e-06, "output_power": 6.866668849399019e-06, "reference_active": true, "delta_db": 4.069293263404711}], "active_band_count": 3, "active_band_mae_db": 1.7625051269701713}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.780703381807892e-05, "log_rms_mae_db": 1.0943191377132462, "rms_envelope_nrmse": 0.10448149433269029, "envelope_correlation": 0.99135620482397, "rms_gain_error_db": -0.24008347197861463, "reference_mean_power": 0.0004167096352122121, "output_mean_power": 0.0003942985637931673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9169450940862202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.13104680727751e-05, "output_power": 0.00010006694510367254, "reference_active": true, "delta_db": 0.3977007256406565}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.750575155696485e-06, "output_power": 7.821140171013025e-06, "reference_active": true, "delta_db": 1.3355878568837676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0115145688994735e-07, "output_power": 4.210689326629621e-07, "reference_active": true, "delta_db": 3.208300170513544}], "active_band_count": 3, "active_band_mae_db": 1.647196251012656}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4613391249073974e-05, "log_rms_mae_db": 2.279920846054336, "rms_envelope_nrmse": 0.3814423667434252, "envelope_correlation": 0.9068249581088884, "rms_gain_error_db": 0.756062336360991, "reference_mean_power": 2.714804754964509e-05, "output_mean_power": 3.2310586012063555e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3627974020623764e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.8436688435136e-06, "output_power": 5.141566837770553e-06, "reference_active": true, "delta_db": 0.2592104255178428}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.21772084491833e-06, "output_power": 2.636619957537692e-06, "reference_active": true, "delta_db": 0.7514065654850395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.650596314175208e-07, "output_power": 1.0615781459717405e-06, "reference_active": true, "delta_db": 3.58443326333401}], "active_band_count": 3, "active_band_mae_db": 1.5316834181122971}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2268327401512992e-05, "log_rms_mae_db": 1.8597058678454808, "rms_envelope_nrmse": 0.31760046866984293, "envelope_correlation": 0.9306419814138092, "rms_gain_error_db": 0.756025701218477, "reference_mean_power": 2.7284479980131438e-05, "output_mean_power": 3.247268883765631e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2818152654967464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7409144698248785e-06, "output_power": 5.332855711508127e-06, "reference_active": true, "delta_db": 0.5109771300296374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1748833712187593e-06, "output_power": 1.4451229264656476e-06, "reference_active": true, "delta_db": 0.8991003401439314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.424838593381526e-08, "output_power": 8.071438314892727e-08, "reference_active": true, "delta_db": 3.723108232898174}], "active_band_count": 3, "active_band_mae_db": 1.711061901023914}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8577196242519318, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6ea95c0a32789cdfa4acc5cc953d5bdb9c98cc34b2ecc78675ce49d237d80909", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9785009088682473, "standard_deviation": 0.002768575248825857, "minimum": 0.9735218875433035, "maximum": 0.9813810196992596, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5826162974523432, "standard_deviation": 0.01857902784943412, "minimum": 1.551629741002709, "maximum": 1.6065781452360608, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3427942692703745, "standard_deviation": 0.49877659756521125, "minimum": 0.5165861697331499, "maximum": 2.0856511481058946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1657767144277864, "standard_deviation": 0.009988816208391918, "minimum": 0.1549874138623159, "maximum": 0.1836798364129774, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9911276834242373, "standard_deviation": 0.0011876086825694592, "minimum": 0.9894690271923272, "maximum": 0.9934626936745817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1156023314793369, "standard_deviation": 0.018541469552013308, "minimum": 1.0932227269329275, "maximum": 1.1359880775289075, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2618290963776846, "standard_deviation": 0.5613965911977342, "minimum": 0.4012938419941974, "maximum": 2.2643826745423112, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10467301459720149, "standard_deviation": 0.0064611483785167526, "minimum": 0.09183356262367307, "maximum": 0.1137134702426055, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.901924209679978, "standard_deviation": 0.004670585957011052, "minimum": 0.8970073921286126, "maximum": 0.9090108363726729, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.270581113394553, "standard_deviation": 0.02628229831465414, "minimum": 2.218167056083123, "maximum": 2.3059541265119576, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5427193998360518, "standard_deviation": 0.22756667065807362, "minimum": 1.2400707580710566, "maximum": 1.767721112305603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.37087963062998275, "standard_deviation": 0.008648279150969263, "minimum": 0.35869923216681293, "maximum": 0.3852432949307441, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.927456288432605, "standard_deviation": 0.0035676841773758625, "minimum": 0.9232601739047782, "maximum": 0.932049287747907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8138380399871425, "standard_deviation": 0.024379109927755487, "minimum": 1.7837625340996368, "maximum": 1.8597058678454808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5114412102622956, "standard_deviation": 0.2289986800993527, "minimum": 1.1964150454783384, "maximum": 1.8330485759138835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.30502005481835703, "standard_deviation": 0.007813190536073034, "minimum": 0.2963014125572778, "maximum": 0.31760046866984293, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.98334728179437, "standard_deviation": 0.003986778121930651, "minimum": 0.9754452764655825, "maximum": 0.9881997233162632, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.422108580305805, "standard_deviation": 0.025849229202215766, "minimum": 1.37557208606782, "maximum": 1.458935962770002, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8551217830448861, "standard_deviation": 0.34005442337248853, "minimum": 0.37629198588923657, "maximum": 1.4341809974179431, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13970667114950625, "standard_deviation": 0.015362227859817534, "minimum": 0.11994226090484933, "maximum": 0.16910614658205855, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9949460067257181, "standard_deviation": 0.0010726608904516152, "minimum": 0.9931669879639178, "maximum": 0.9962873199915189, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8569150683005943, "standard_deviation": 0.027349415923527447, "minimum": 0.8018877105383593, "maximum": 0.893586587955346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.878243465468309, "standard_deviation": 0.37182760903447587, "minimum": 0.4120066622002903, "maximum": 1.5591026629667535, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07815015918550683, "standard_deviation": 0.0061608805648159315, "minimum": 0.06927417901281734, "maximum": 0.08586472765653064, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2140935526208977, "standard_deviation": 0.04737249404830811, "minimum": 0.1484813602683757, "maximum": 0.29682247676268847, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.734773245946474, "standard_deviation": 0.007964718076797097, "minimum": 5.725729816310257, "maximum": 5.752321289093625, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.8877991682170392, "standard_deviation": 0.10469051653163385, "minimum": 0.7757711216942065, "maximum": 1.0644612278367667, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.058348480911509244, "standard_deviation": 0.06108665915381275, "minimum": -0.11275977539010659, "maximum": 0.04153306110690631, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9887925366722001, "standard_deviation": 0.0019851467928323468, "minimum": 0.9847169459408192, "maximum": 0.9904161849111464, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.087051221095396, "standard_deviation": 0.01366456794853105, "minimum": 1.070044328536437, "maximum": 1.105264804252642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6784651283864581, "standard_deviation": 0.428233756255079, "minimum": 0.20658066512755954, "maximum": 1.5996971865298775, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10243972095471465, "standard_deviation": 0.008923203256283516, "minimum": 0.09502242038467186, "maximum": 0.12065699600274452, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969629384818972, "standard_deviation": 0.0006762500579550599, "minimum": 0.9958653061841225, "maximum": 0.9979665826600087, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5629084450521558, "standard_deviation": 0.02950442021140049, "minimum": 0.5126396418541683, "maximum": 0.5993975688065034, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7324225671108922, "standard_deviation": 0.574816772848578, "minimum": 0.12778861118375742, "maximum": 2.0224360423785437, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05164510213993214, "standard_deviation": 0.005737908194975546, "minimum": 0.04274677885400865, "maximum": 0.06054333686081602, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8455-210777-0010", "language": "en", "style": "normal", "speaker": "librispeech_8455", "source_utterance_id": "8455-210777-0010", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 49600, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8455-210777-0010.wav", "source_file_sha256": "f0a0102d6ed44efb60116d73a6762edb24b8b46d6f89a6e3832f45f8515d2fe7", "decoded_pcm_sha256": "10ec70ae1221ff8684ca0b1b6aa22fd7ae9343c0b701f90c315735ff27f3d5ad", "original_source_sha256": "28029242eb37055c0eb70761eebf5b7be98a5da6c30127b3b282bf7e11663a68", "corpus_entry_id": "librispeech_8455-210777-0010"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.212151050567627, "predicted_mos": 4.212151050567627, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "f34cb77e42a0793b339c70dd0c299eed04c5bb174b9dfd9f23bc25ae24fb4676", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6221239250007784, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8455-210777-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f34cb77e42a0793b339c70dd0c299eed04c5bb174b9dfd9f23bc25ae24fb4676", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "153385cc5de0b72725a7f9cb203194181b7d234e40b71aae096375c5bfee23b6"}, "rows": [{"case": "librispeech_8455-210777-0010", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.530282974243164, "predicted_mos": 2.530282974243164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "69111ac895e78d70cfaa778cad1ef3a946c87ab00177af78b3596ec886f3423f", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1896559209999396, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8455-210777-0010", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "69111ac895e78d70cfaa778cad1ef3a946c87ab00177af78b3596ec886f3423f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.040969108939195, "全帯域_10ms_rms_envelope_nrmse": 0.09105722474171787, "全帯域_10ms_envelope_correlation": 0.9932320362988254, "全帯域_10ms_modulation_mae_db": 0.23665821563238013, "全帯域_40ms_log_rms_mae_db": 0.538252919412101, "全帯域_40ms_rms_envelope_nrmse": 0.053708885084871254, "全帯域_40ms_envelope_correlation": 0.99764590797095, "全帯域_40ms_modulation_mae_db": 0.3895514738289674, "300–1000 Hz_10ms_log_rms_mae_db": 1.7539079480432436, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18728573972637316, "300–1000 Hz_10ms_envelope_correlation": 0.9767545848046819, "300–1000 Hz_10ms_modulation_mae_db": 0.9457324661335346, "300–1000 Hz_40ms_log_rms_mae_db": 1.0099735415397035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09535681419977593, "300–1000 Hz_40ms_envelope_correlation": 0.9943076663178699, "300–1000 Hz_40ms_modulation_mae_db": 0.6413194798741431, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4422490384310993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13405749186246546, "1000–3000 Hz_10ms_envelope_correlation": 0.9893806260275272, "1000–3000 Hz_10ms_modulation_mae_db": 0.8273040655749, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9241833392304853, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06411833609743071, "1000–3000 Hz_40ms_envelope_correlation": 0.997693281825309, "1000–3000 Hz_40ms_modulation_mae_db": 0.7563551341019088, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6154625851453945, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21367696232882727, "3000–6000 Hz_10ms_envelope_correlation": 0.9711606291342304, "3000–6000 Hz_10ms_modulation_mae_db": 0.33461696730840784, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1296608528777465, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10598264050131755, "3000–6000 Hz_40ms_envelope_correlation": 0.9933919252332766, "3000–6000 Hz_40ms_modulation_mae_db": 0.34696812152185985, "mr_stft_db": 5.767175628097834, "band_mae_db": 0.3013264676782901, "rms_gain_error_db": -0.11239544929845824, "output_peak": 0.7749833151950233}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0898885353413161, "全帯域_10ms_rms_envelope_nrmse": 0.0817755357232203, "全帯域_10ms_envelope_correlation": 0.9945850108608515, "全帯域_10ms_modulation_mae_db": 0.7152601917735887, "全帯域_40ms_log_rms_mae_db": 0.5971155447227434, "全帯域_40ms_rms_envelope_nrmse": 0.03630729202359064, "全帯域_40ms_envelope_correlation": 0.9990302152958441, "全帯域_40ms_modulation_mae_db": 0.9151055019525214, "300–1000 Hz_10ms_log_rms_mae_db": 1.7788388096342855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14748673014584818, "300–1000 Hz_10ms_envelope_correlation": 0.9860262964480269, "300–1000 Hz_10ms_modulation_mae_db": 1.1495404473267203, "300–1000 Hz_40ms_log_rms_mae_db": 1.0937038130199528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06970446403139409, "300–1000 Hz_40ms_envelope_correlation": 0.9975636364564421, "300–1000 Hz_40ms_modulation_mae_db": 1.2610295922559676, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4071960854290542, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12898911330472088, "1000–3000 Hz_10ms_envelope_correlation": 0.9897986336586297, "1000–3000 Hz_10ms_modulation_mae_db": 0.3536322382150668, "1000–3000 Hz_40ms_log_rms_mae_db": 0.943709967791194, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.059290303635858035, "1000–3000 Hz_40ms_envelope_correlation": 0.9979926959586736, "1000–3000 Hz_40ms_modulation_mae_db": 0.34647373329274017, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5773611108338879, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18851575880080423, "3000–6000 Hz_10ms_envelope_correlation": 0.9758780073562706, "3000–6000 Hz_10ms_modulation_mae_db": 0.44992116339532445, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0855526615732491, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09880847897332207, "3000–6000 Hz_40ms_envelope_correlation": 0.9930699418204227, "3000–6000 Hz_40ms_modulation_mae_db": 0.6496187422133185, "mr_stft_db": 5.791058578769274, "band_mae_db": 0.3690564499937133, "rms_gain_error_db": -0.12386885210704049, "output_peak": 0.7685535691322024}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1106028523644202, "全帯域_10ms_rms_envelope_nrmse": 0.12549043057942816, "全帯域_10ms_envelope_correlation": 0.9871049324392688, "全帯域_10ms_modulation_mae_db": 0.6752543845601385, "全帯域_40ms_log_rms_mae_db": 0.6092621813281967, "全帯域_40ms_rms_envelope_nrmse": 0.06547952675162513, "全帯域_40ms_envelope_correlation": 0.996406146158758, "全帯域_40ms_modulation_mae_db": 1.104112818323147, "300–1000 Hz_10ms_log_rms_mae_db": 1.801357551860771, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20393346169370316, "300–1000 Hz_10ms_envelope_correlation": 0.9722219291869154, "300–1000 Hz_10ms_modulation_mae_db": 1.3884995421999335, "300–1000 Hz_40ms_log_rms_mae_db": 0.9948943134524596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09328951276188008, "300–1000 Hz_40ms_envelope_correlation": 0.994203180171316, "300–1000 Hz_40ms_modulation_mae_db": 1.5369258891266746, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4103958406695667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1983340353153974, "1000–3000 Hz_10ms_envelope_correlation": 0.9760108245397718, "1000–3000 Hz_10ms_modulation_mae_db": 1.7915423673239925, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8779732050927215, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10343921736050295, "1000–3000 Hz_40ms_envelope_correlation": 0.9933830813859598, "1000–3000 Hz_40ms_modulation_mae_db": 2.0346752801753927, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5870121191055373, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20664469401219038, "3000–6000 Hz_10ms_envelope_correlation": 0.9716032259422516, "3000–6000 Hz_10ms_modulation_mae_db": 0.4045306936508979, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0904346905078854, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11721400173009738, "3000–6000 Hz_40ms_envelope_correlation": 0.9905922399570771, "3000–6000 Hz_40ms_modulation_mae_db": 0.4248362109703799, "mr_stft_db": 5.782867219835722, "band_mae_db": 0.1618154241626572, "rms_gain_error_db": -0.042958135859827065, "output_peak": 0.6715041018530243}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0744320916620549, "全帯域_10ms_rms_envelope_nrmse": 0.10498200627673726, "全帯域_10ms_envelope_correlation": 0.9909870177123009, "全帯域_10ms_modulation_mae_db": 0.9116887259499594, "全帯域_40ms_log_rms_mae_db": 0.5909454606873314, "全帯域_40ms_rms_envelope_nrmse": 0.051256142812013086, "全帯域_40ms_envelope_correlation": 0.9977846022049349, "全帯域_40ms_modulation_mae_db": 1.2934955028474013, "300–1000 Hz_10ms_log_rms_mae_db": 1.805725105296312, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17835465177612334, "300–1000 Hz_10ms_envelope_correlation": 0.979858806957103, "300–1000 Hz_10ms_modulation_mae_db": 1.424950818341597, "300–1000 Hz_40ms_log_rms_mae_db": 1.0924529046606033, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09646583715745782, "300–1000 Hz_40ms_envelope_correlation": 0.9941211801105972, "300–1000 Hz_40ms_modulation_mae_db": 1.4526956491464784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4155289877487918, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20980964803537963, "1000–3000 Hz_10ms_envelope_correlation": 0.9727782966644687, "1000–3000 Hz_10ms_modulation_mae_db": 1.125183107629548, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8809955476395622, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1005152478240812, "1000–3000 Hz_40ms_envelope_correlation": 0.9943561244563113, "1000–3000 Hz_40ms_modulation_mae_db": 1.136567332957584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7026753298381558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23819211127939377, "3000–6000 Hz_10ms_envelope_correlation": 0.9617174537821139, "3000–6000 Hz_10ms_modulation_mae_db": 0.9146569146258808, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1572099104566822, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12009760294983379, "3000–6000 Hz_40ms_envelope_correlation": 0.9898312618581774, "3000–6000 Hz_40ms_modulation_mae_db": 0.8434838323735695, "mr_stft_db": 5.779215612632029, "band_mae_db": 0.2485902112926005, "rms_gain_error_db": -0.03221322257391455, "output_peak": 0.783469754628068}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1118931137226364, "全帯域_10ms_rms_envelope_nrmse": 0.1365950004530208, "全帯域_10ms_envelope_correlation": 0.9860931763328379, "全帯域_10ms_modulation_mae_db": 1.1045106622114667, "全帯域_40ms_log_rms_mae_db": 0.5686163999928766, "全帯域_40ms_rms_envelope_nrmse": 0.08250229897218456, "全帯域_40ms_envelope_correlation": 0.9963687023953431, "全帯域_40ms_modulation_mae_db": 1.4389991124401182, "300–1000 Hz_10ms_log_rms_mae_db": 1.872021882013338, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.24265310799854856, "300–1000 Hz_10ms_envelope_correlation": 0.9624443172369423, "300–1000 Hz_10ms_modulation_mae_db": 1.8016847190339114, "300–1000 Hz_40ms_log_rms_mae_db": 1.0234211991880329, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14327081422969046, "300–1000 Hz_40ms_envelope_correlation": 0.9899874297158601, "300–1000 Hz_40ms_modulation_mae_db": 1.8247195722700813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3692498962206159, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17910185501645273, "1000–3000 Hz_10ms_envelope_correlation": 0.9824957872578387, "1000–3000 Hz_10ms_modulation_mae_db": 1.3123004818207076, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8955507775159617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11309707462103316, "1000–3000 Hz_40ms_envelope_correlation": 0.9953029859924454, "1000–3000 Hz_40ms_modulation_mae_db": 1.2748505211347079, "3000–6000 Hz_10ms_log_rms_mae_db": 1.683803830572245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22884208524599986, "3000–6000 Hz_10ms_envelope_correlation": 0.9641199083482976, "3000–6000 Hz_10ms_modulation_mae_db": 0.893445874542386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1021264132850077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09678750425189357, "3000–6000 Hz_40ms_envelope_correlation": 0.993293021525401, "3000–6000 Hz_40ms_modulation_mae_db": 0.7287320089329121, "mr_stft_db": 5.775120778936382, "band_mae_db": 0.4366022644725536, "rms_gain_error_db": -0.45328229467021397, "output_peak": 0.8129292782678422}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0379307959934079, "全帯域_10ms_rms_envelope_nrmse": 0.12075274389469473, "全帯域_10ms_envelope_correlation": 0.9885273143389268, "全帯域_10ms_modulation_mae_db": 0.5208410699186861, "全帯域_40ms_log_rms_mae_db": 0.5397280414886431, "全帯域_40ms_rms_envelope_nrmse": 0.06444306322551599, "全帯域_40ms_envelope_correlation": 0.9973592885238776, "全帯域_40ms_modulation_mae_db": 0.7073772140544277, "300–1000 Hz_10ms_log_rms_mae_db": 1.7566038335015184, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16958650658775035, "300–1000 Hz_10ms_envelope_correlation": 0.9822927728113015, "300–1000 Hz_10ms_modulation_mae_db": 0.8080935801339585, "300–1000 Hz_40ms_log_rms_mae_db": 1.082248336369589, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09667839808584189, "300–1000 Hz_40ms_envelope_correlation": 0.9958671577913626, "300–1000 Hz_40ms_modulation_mae_db": 1.4567034796767355, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4088941774923276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2156604841255932, "1000–3000 Hz_10ms_envelope_correlation": 0.9711285028177634, "1000–3000 Hz_10ms_modulation_mae_db": 2.4253982019585525, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9481580360926031, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09735516733507658, "1000–3000 Hz_40ms_envelope_correlation": 0.9946319333080865, "1000–3000 Hz_40ms_modulation_mae_db": 2.401292228283998, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6119659922599943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20735947564139615, "3000–6000 Hz_10ms_envelope_correlation": 0.9703746479586279, "3000–6000 Hz_10ms_modulation_mae_db": 0.8878071268626279, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1281807884403023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11935251637148406, "3000–6000 Hz_40ms_envelope_correlation": 0.9898194128550982, "3000–6000 Hz_40ms_modulation_mae_db": 0.9276852899157878, "mr_stft_db": 5.790516992841582, "band_mae_db": 0.38322327860463684, "rms_gain_error_db": -0.2958393981718104, "output_peak": 0.7822406857784382}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1199113705156385, "全帯域_10ms_rms_envelope_nrmse": 0.12836257955434588, "全帯域_10ms_envelope_correlation": 0.9864705827573276, "全帯域_10ms_modulation_mae_db": 0.89224942905631, "全帯域_40ms_log_rms_mae_db": 0.5641273720241077, "全帯域_40ms_rms_envelope_nrmse": 0.058047482741353906, "全帯域_40ms_envelope_correlation": 0.9973649537575284, "全帯域_40ms_modulation_mae_db": 0.9775894552256622, "300–1000 Hz_10ms_log_rms_mae_db": 1.7929577462783308, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1813679224517125, "300–1000 Hz_10ms_envelope_correlation": 0.9781340283919919, "300–1000 Hz_10ms_modulation_mae_db": 1.2747003744497933, "300–1000 Hz_40ms_log_rms_mae_db": 0.9903841589846515, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08631543683072483, "300–1000 Hz_40ms_envelope_correlation": 0.9951403123914955, "300–1000 Hz_40ms_modulation_mae_db": 1.0724162394812502, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4084788320165091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2006281299566203, "1000–3000 Hz_10ms_envelope_correlation": 0.9753167887972377, "1000–3000 Hz_10ms_modulation_mae_db": 1.6856220392346817, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9503536443154508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10408959044706755, "1000–3000 Hz_40ms_envelope_correlation": 0.994090728548407, "1000–3000 Hz_40ms_modulation_mae_db": 1.8452004444709849, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7226076220225377, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21547556596272682, "3000–6000 Hz_10ms_envelope_correlation": 0.9685852143548481, "3000–6000 Hz_10ms_modulation_mae_db": 0.7640687821876124, "3000–6000 Hz_40ms_log_rms_mae_db": 1.192516139142238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12506384316186123, "3000–6000 Hz_40ms_envelope_correlation": 0.988940591001228, "3000–6000 Hz_40ms_modulation_mae_db": 0.7755003265127333, "mr_stft_db": 5.795139303776614, "band_mae_db": 0.2860449803879465, "rms_gain_error_db": -0.16342918685091704, "output_peak": 0.8551547729738457}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0788860159168168, "全帯域_10ms_rms_envelope_nrmse": 0.11804786718868111, "全帯域_10ms_envelope_correlation": 0.9889055141156622, "全帯域_10ms_modulation_mae_db": 0.5361077480891946, "全帯域_40ms_log_rms_mae_db": 0.5538880289689956, "全帯域_40ms_rms_envelope_nrmse": 0.05561844286582487, "全帯域_40ms_envelope_correlation": 0.9980872745749669, "全帯域_40ms_modulation_mae_db": 0.8022919182674673, "300–1000 Hz_10ms_log_rms_mae_db": 1.7378714291710944, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1657614798115543, "300–1000 Hz_10ms_envelope_correlation": 0.982471577400784, "300–1000 Hz_10ms_modulation_mae_db": 0.7116112566446322, "300–1000 Hz_40ms_log_rms_mae_db": 0.9824277680256999, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09054591223471387, "300–1000 Hz_40ms_envelope_correlation": 0.9957535212978318, "300–1000 Hz_40ms_modulation_mae_db": 0.9691335063007543, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4040150265944213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19899194732793307, "1000–3000 Hz_10ms_envelope_correlation": 0.9756538378689359, "1000–3000 Hz_10ms_modulation_mae_db": 1.060459452519394, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8206955814699524, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0698374638626476, "1000–3000 Hz_40ms_envelope_correlation": 0.9969311075509733, "1000–3000 Hz_40ms_modulation_mae_db": 0.5082638271546616, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5647084216489688, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19945170496446324, "3000–6000 Hz_10ms_envelope_correlation": 0.9732271418134398, "3000–6000 Hz_10ms_modulation_mae_db": 0.46756691168291464, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0530156184878923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10623600534509793, "3000–6000 Hz_40ms_envelope_correlation": 0.9920924749883658, "3000–6000 Hz_40ms_modulation_mae_db": 0.531979360884793, "mr_stft_db": 5.774811914667869, "band_mae_db": 0.30456365895035903, "rms_gain_error_db": -0.2584782778269219, "output_peak": 0.7479628890029212}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49600, "output_samples": 49600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.860476554507568, "spectral_convergence": 0.46842934393838326, "spectral_convergence_status": "ok", "amplitude_floor": 1.4924605057201793e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.541355201540868, "spectral_convergence": 0.4417844547889929, "spectral_convergence_status": "ok", "amplitude_floor": 8.127484488587521e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8996951282450665, "spectral_convergence": 0.5040715817446784, "spectral_convergence_status": "ok", "amplitude_floor": 5.495951303547364e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185024372796483, "output_power": 0.000768168967145255, "delta_db": -0.2756321861298808, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004929662452993321, "output_power": 0.0005298210370654395, "delta_db": 0.31312015387974923, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001304221454098428, "output_power": 0.001157936619277519, "delta_db": -0.5166655142438219, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007846801909808332, "output_power": 0.0008286161239739701, "delta_db": 0.23660690833644246, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003122815376510157, "output_power": 0.00032213868346249534, "delta_db": 0.13496570315755016, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001061689290478197, "output_power": 0.00011457616203727171, "delta_db": 0.33096834032229616, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3013264676782901, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11239544929845824, "pitch": null, "output_rms_dbfs": -24.293102661972547, "output_peak": 0.7749833151950233, "reference_rms_dbfs": -24.180707212674086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.767175628097834}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.0002690918479937382, "log_rms_mae_db": 1.040969108939195, "rms_envelope_nrmse": 0.09105722474171787, "envelope_correlation": 0.9932320362988254, "rms_gain_error_db": -0.11239494441863435, "reference_mean_power": 0.0038249880205223005, "output_mean_power": 0.0037272676921907633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 2.1583633605855604e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008154057878800693, "output_power": 0.0007263913715566114, "reference_active": true, "delta_db": -0.5020311304385443}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.087495849252009e-05, "output_power": 9.499964446337954e-05, "reference_active": true, "delta_db": 0.1927775444123112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.599429116228097e-05, "output_power": 2.5903674963311953e-05, "reference_active": true, "delta_db": -0.015165972046284894}], "active_band_count": 3, "active_band_mae_db": 0.23665821563238013}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.0002456731948415248, "log_rms_mae_db": 0.538252919412101, "rms_envelope_nrmse": 0.053708885084871254, "envelope_correlation": 0.99764590797095, "rms_gain_error_db": -0.11239536578200732, "reference_mean_power": 0.0038624153279842384, "output_mean_power": 0.0037637384461903602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 2.0216648233438394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007562450561775054, "output_power": 0.0006708122892212789, "reference_active": true, "delta_db": -0.5206153830095125}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1399930780791825e-05, "output_power": 4.363171457400308e-05, "reference_active": true, "delta_db": 0.22802664581940305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0921358976760216e-06, "output_power": 1.2030340555222085e-06, "reference_active": true, "delta_db": 0.4200123926579866}], "active_band_count": 3, "active_band_mae_db": 0.3895514738289674}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00017000794064843354, "log_rms_mae_db": 1.7539079480432436, "rms_envelope_nrmse": 0.18728573972637316, "envelope_correlation": 0.9767545848046819, "rms_gain_error_db": -0.28686873810232383, "reference_mean_power": 0.001461911046007658, "output_mean_power": 0.0013684661956400136}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.2056298552980857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006038941034712422, "output_power": 0.0005041570822581096, "reference_active": true, "delta_db": -0.783949166851502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.14042207191201e-05, "output_power": 7.713585503235984e-05, "reference_active": true, "delta_db": 0.33532414158183455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.020128231954835e-05, "output_power": 3.000346045780369e-05, "reference_active": true, "delta_db": 1.7179240899672676}], "active_band_count": 3, "active_band_mae_db": 0.9457324661335346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00015858554889566586, "log_rms_mae_db": 1.0099735415397035, "rms_envelope_nrmse": 0.09535681419977593, "envelope_correlation": 0.9943076663178699, "rms_gain_error_db": -0.28686720698258084, "reference_mean_power": 0.0014762175799508475, "output_mean_power": 0.001381858748041634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.148107626412446e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000571116387197927, "output_power": 0.0004862764489142858, "reference_active": true, "delta_db": -0.6984138511818705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.637477969141665e-05, "output_power": 3.726174367538651e-05, "reference_active": true, "delta_db": 0.104628023460307}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.172790453481755e-06, "output_power": 1.5181408928891104e-06, "reference_active": true, "delta_db": 1.1209165649802517}], "active_band_count": 3, "active_band_mae_db": 0.6413194798741431}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00019136572750236326, "log_rms_mae_db": 1.4422490384310993, "rms_envelope_nrmse": 0.13405749186246546, "envelope_correlation": 0.9893806260275272, "rms_gain_error_db": 0.15116663205292835, "reference_mean_power": 0.0009651549113833473, "output_mean_power": 0.000999340960018857}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 7.99902404373172e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003359743599084894, "output_power": 0.0003488290020472737, "reference_active": true, "delta_db": 0.16306450305992432}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9966531082920368e-05, "output_power": 3.349369592885124e-05, "reference_active": true, "delta_db": 0.4832660122709519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.661292025781732e-06, "output_power": 1.0165216701579005e-05, "reference_active": true, "delta_db": 1.835581681393824}], "active_band_count": 3, "active_band_mae_db": 0.8273040655749}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00016758862552795117, "log_rms_mae_db": 0.9241833392304853, "rms_envelope_nrmse": 0.06411833609743071, "envelope_correlation": 0.997693281825309, "rms_gain_error_db": 0.15116566047957167, "reference_mean_power": 0.0009746009072787481, "output_mean_power": 0.001009121309896272}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 7.731649944214162e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031280752850392277, "output_power": 0.00032987325319117555, "reference_active": true, "delta_db": 0.23069906565670972}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5385931815426458e-05, "output_power": 1.7513257380264508e-05, "reference_active": true, "delta_db": 0.5624312676840957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1192579703922553e-07, "output_power": 4.3817218007439986e-07, "reference_active": true, "delta_db": 1.475935068964921}], "active_band_count": 3, "active_band_mae_db": 0.7563551341019088}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 5.61189760356505e-05, "log_rms_mae_db": 1.6154625851453945, "rms_envelope_nrmse": 0.21367696232882727, "envelope_correlation": 0.9711606291342304, "rms_gain_error_db": 0.33341328598766207, "reference_mean_power": 0.00014982208037105655, "output_mean_power": 0.00016177713940690195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.5994961899965245e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.519717012759751e-05, "output_power": 6.014900741273174e-05, "reference_active": true, "delta_db": 0.3731165224466291}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5683811710113524e-05, "output_power": 1.5031424663054667e-05, "reference_active": true, "delta_db": -0.1845147527117601}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1546072478197644e-06, "output_power": 3.4959664766489564e-06, "reference_active": true, "delta_db": 0.4462196267668344}], "active_band_count": 3, "active_band_mae_db": 0.33461696730840784}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 4.51769085602287e-05, "log_rms_mae_db": 1.1296608528777465, "rms_envelope_nrmse": 0.10598264050131755, "envelope_correlation": 0.9933919252332766, "rms_gain_error_db": 0.3334141015270203, "reference_mean_power": 0.0001512883740521334, "output_mean_power": 0.00016336046672890206}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.4928738695410609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.9695472955753735e-05, "output_power": 5.413625261713369e-05, "reference_active": true, "delta_db": 0.37171361865307007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.547268075821193e-06, "output_power": 7.957025485291944e-06, "reference_active": true, "delta_db": -0.3107657629894027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7628269396099532e-07, "output_power": 1.9144859217398074e-07, "reference_active": true, "delta_db": 0.3584249829231068}], "active_band_count": 3, "active_band_mae_db": 0.34696812152185985}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8567092220855897, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3d67c9adec6736ad77872ea73b017f222e36d25364139d436dc44576bcb10c21", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9790704122040217, "standard_deviation": 0.007279293795513442, "minimum": 0.9711285028177634, "maximum": 0.9897986336586297, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4082509855752983, "standard_deviation": 0.019835192607727984, "minimum": 1.3692498962206159, "maximum": 1.4422490384310993, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3226802442846055, "standard_deviation": 0.6391813435625843, "minimum": 0.3536322382150668, "maximum": 2.4253982019585525, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18319658811807032, "standard_deviation": 0.03362507945581718, "minimum": 0.12898911330472088, "maximum": 0.2156604841255932, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955477423782708, "standard_deviation": 0.001757516448026437, "minimum": 0.9933830813859598, "maximum": 0.9979926959586736, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9052025123934914, "standard_deviation": 0.0451804118707287, "minimum": 0.8206955814699524, "maximum": 0.9503536443154508, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2879598126964973, "standard_deviation": 0.7471495468017165, "minimum": 0.34647373329274017, "maximum": 2.401292228283998, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08896780014796223, "standard_deviation": 0.021004160302066543, "minimum": 0.059290303635858035, "maximum": 0.11309707462103316, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9695832785862599, "standard_deviation": 0.004672282762337774, "minimum": 0.9617174537821139, "maximum": 0.9758780073562706, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6331996264283402, "standard_deviation": 0.06104938183796366, "minimum": 1.5647084216489688, "maximum": 1.7226076220225377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6395768042820065, "standard_deviation": 0.2481471220048016, "minimum": 0.33461696730840784, "maximum": 0.9146569146258808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21226979477947522, "standard_deviation": 0.015780526144957147, "minimum": 0.18851575880080423, "maximum": 0.23819211127939377, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9913788586548808, "standard_deviation": 0.0017918989234224708, "minimum": 0.988940591001228, "maximum": 0.9933919252332766, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1173371343463756, "standard_deviation": 0.044100206576899934, "minimum": 1.0530156184878923, "maximum": 1.192516139142238, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6536004866656693, "standard_deviation": 0.20453659632586313, "minimum": 0.34696812152185985, "maximum": 0.9276852899157878, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11119282416061345, "standard_deviation": 0.010604155005342519, "minimum": 0.09678750425189357, "maximum": 0.12506384316186123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9775255391547184, "standard_deviation": 0.007381763119345337, "minimum": 0.9624443172369423, "maximum": 0.9860262964480269, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7874105382248617, "standard_deviation": 0.04195042658205613, "minimum": 1.7378714291710944, "maximum": 1.872021882013338, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1881016505330102, "standard_deviation": 0.3609351705117972, "minimum": 0.7116112566446322, "maximum": 1.8016847190339114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1845537000239517, "standard_deviation": 0.028681209538256053, "minimum": 0.14748673014584818, "maximum": 0.24265310799854856, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946180105315969, "standard_deviation": 0.0021959957655402276, "minimum": 0.9899874297158601, "maximum": 0.9975636364564421, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0336882544050865, "standard_deviation": 0.04794175779555524, "minimum": 0.9824277680256999, "maximum": 1.0937038130199528, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.2768679260165108, "standard_deviation": 0.37257816638665564, "minimum": 0.6413194798741431, "maximum": 1.8247195722700813, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09645339869143488, "standard_deviation": 0.020895372082353147, "minimum": 0.06970446403139409, "maximum": 0.14327081422969046, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3114028419428446, "standard_deviation": 0.08558217442706836, "minimum": 0.1618154241626572, "maximum": 0.4366022644725536, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.781988253694664, "standard_deviation": 0.009669512877725953, "minimum": 5.767175628097834, "maximum": 5.795139303776614, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7745997958539207, "standard_deviation": 0.05273960805040868, "minimum": 0.6715041018530243, "maximum": 0.8551547729738457, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.18530810216988797, "standard_deviation": 0.14275477887787558, "minimum": -0.45328229467021397, "maximum": -0.03221322257391455, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9894881981070002, "standard_deviation": 0.003158101455027683, "minimum": 0.9860931763328379, "maximum": 0.9945850108608515, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0830642355569358, "standard_deviation": 0.03138416464569235, "minimum": 1.0379307959934079, "maximum": 1.1199113705156385, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6990713033989655, "standard_deviation": 0.2725234868599794, "minimum": 0.23665821563238013, "maximum": 1.1045106622114667, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11338292355148076, "standard_deviation": 0.019100850726139637, "minimum": 0.0817755357232203, "maximum": 0.1365950004530208, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975058863602754, "standard_deviation": 0.0008703733467065905, "minimum": 0.9963687023953431, "maximum": 0.9990302152958441, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5702419935781244, "standard_deviation": 0.026552594173299995, "minimum": 0.538252919412101, "maximum": 0.6092621813281967, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.953565374617464, "standard_deviation": 0.3332787778241739, "minimum": 0.3895514738289674, "maximum": 1.4389991124401182, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05842039180962243, "standard_deviation": 0.013290953584086822, "minimum": 0.03630729202359064, "maximum": 0.08250229897218456, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0010", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.475074529647827, "predicted_mos": 2.475074529647827, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "f5242700646aa3357f601951377169472ab31fc2ecee3706ca9ebbb6715265bc", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.687189199961722, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0308360882556955, "全帯域_10ms_rms_envelope_nrmse": 0.08835041789385728, "全帯域_10ms_envelope_correlation": 0.9936620279507402, "全帯域_10ms_modulation_mae_db": 0.2836415771920862, "全帯域_40ms_log_rms_mae_db": 0.5437351545875014, "全帯域_40ms_rms_envelope_nrmse": 0.05429969339504889, "全帯域_40ms_envelope_correlation": 0.9976263209989932, "全帯域_40ms_modulation_mae_db": 0.2747285683807777, "300–1000 Hz_10ms_log_rms_mae_db": 1.7461244576406072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18181261346423466, "300–1000 Hz_10ms_envelope_correlation": 0.9781137845534891, "300–1000 Hz_10ms_modulation_mae_db": 0.8363692624295519, "300–1000 Hz_40ms_log_rms_mae_db": 0.9966874292073349, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0946664809556412, "300–1000 Hz_40ms_envelope_correlation": 0.9943350971606282, "300–1000 Hz_40ms_modulation_mae_db": 0.4072032061328133, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3603968347273623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12718456809313095, "1000–3000 Hz_10ms_envelope_correlation": 0.9901235407161822, "1000–3000 Hz_10ms_modulation_mae_db": 0.5471291583491725, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7841620936659022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.056065453307287016, "1000–3000 Hz_40ms_envelope_correlation": 0.9980630953561934, "1000–3000 Hz_40ms_modulation_mae_db": 0.3521155726443194, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5611970985968069, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2048687287156367, "3000–6000 Hz_10ms_envelope_correlation": 0.9735602248775524, "3000–6000 Hz_10ms_modulation_mae_db": 0.4114139324783359, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0281033784240998, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09461033257577618, "3000–6000 Hz_40ms_envelope_correlation": 0.9949258995251715, "3000–6000 Hz_40ms_modulation_mae_db": 0.2768463708013269, "mr_stft_db": 5.7351690762635, "band_mae_db": 0.23578764623457404, "rms_gain_error_db": -0.12772891854389767, "output_peak": 0.7914895279544425}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.148758671349912, "全帯域_10ms_rms_envelope_nrmse": 0.08261131843111799, "全帯域_10ms_envelope_correlation": 0.994475061931948, "全帯域_10ms_modulation_mae_db": 0.7034175867522351, "全帯域_40ms_log_rms_mae_db": 0.6160581964441086, "全帯域_40ms_rms_envelope_nrmse": 0.03724788486887327, "全帯域_40ms_envelope_correlation": 0.9989777202492266, "全帯域_40ms_modulation_mae_db": 0.8822705423063439, "300–1000 Hz_10ms_log_rms_mae_db": 1.748410124118083, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1518499869883277, "300–1000 Hz_10ms_envelope_correlation": 0.9851043027401243, "300–1000 Hz_10ms_modulation_mae_db": 1.10049085725463, "300–1000 Hz_40ms_log_rms_mae_db": 1.0460592718209234, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07345056249942944, "300–1000 Hz_40ms_envelope_correlation": 0.9971359103853563, "300–1000 Hz_40ms_modulation_mae_db": 1.1450962890566136, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3609529350287017, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13116543568619932, "1000–3000 Hz_10ms_envelope_correlation": 0.989523055721123, "1000–3000 Hz_10ms_modulation_mae_db": 0.3956310792568716, "1000–3000 Hz_40ms_log_rms_mae_db": 0.833052337751679, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0611614965159683, "1000–3000 Hz_40ms_envelope_correlation": 0.9979631783632377, "1000–3000 Hz_40ms_modulation_mae_db": 0.369438110869671, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5590632896598127, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18514666056896956, "3000–6000 Hz_10ms_envelope_correlation": 0.9767035294251902, "3000–6000 Hz_10ms_modulation_mae_db": 0.420600444236503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0579754274055946, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0981342147717088, "3000–6000 Hz_40ms_envelope_correlation": 0.9931467647349627, "3000–6000 Hz_40ms_modulation_mae_db": 0.586348251863786, "mr_stft_db": 5.765598188367029, "band_mae_db": 0.37170919980635375, "rms_gain_error_db": -0.12660004498436903, "output_peak": 0.7713139514108605}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1158149144780525, "全帯域_10ms_rms_envelope_nrmse": 0.12615921416876077, "全帯域_10ms_envelope_correlation": 0.9869232107431282, "全帯域_10ms_modulation_mae_db": 0.4862658339446373, "全帯域_40ms_log_rms_mae_db": 0.6154381359705288, "全帯域_40ms_rms_envelope_nrmse": 0.06684879122880125, "全帯域_40ms_envelope_correlation": 0.9962365796467045, "全帯域_40ms_modulation_mae_db": 0.8506278689684597, "300–1000 Hz_10ms_log_rms_mae_db": 1.7850110955785055, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2016102986416594, "300–1000 Hz_10ms_envelope_correlation": 0.9728464311332516, "300–1000 Hz_10ms_modulation_mae_db": 1.2081257256792899, "300–1000 Hz_40ms_log_rms_mae_db": 0.9677750073063336, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09480478102912572, "300–1000 Hz_40ms_envelope_correlation": 0.9939939164250947, "300–1000 Hz_40ms_modulation_mae_db": 1.3180003670393716, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3610025114712705, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19600525537432623, "1000–3000 Hz_10ms_envelope_correlation": 0.9763708035676717, "1000–3000 Hz_10ms_modulation_mae_db": 1.4551158579034815, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7917520393648615, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10523838379958905, "1000–3000 Hz_40ms_envelope_correlation": 0.9931052724830863, "1000–3000 Hz_40ms_modulation_mae_db": 1.6948102506640286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5609063288185139, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19894611077292776, "3000–6000 Hz_10ms_envelope_correlation": 0.9734928280833559, "3000–6000 Hz_10ms_modulation_mae_db": 0.5947888494978449, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0356025541631915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11420406921870876, "3000–6000 Hz_40ms_envelope_correlation": 0.9909663103895252, "3000–6000 Hz_40ms_modulation_mae_db": 0.59959380841637, "mr_stft_db": 5.74792304461165, "band_mae_db": 0.1418369105425593, "rms_gain_error_db": -0.05880160521057422, "output_peak": 0.6708976494335318}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0863319196323404, "全帯域_10ms_rms_envelope_nrmse": 0.10130837133786656, "全帯域_10ms_envelope_correlation": 0.9915886186424803, "全帯域_10ms_modulation_mae_db": 0.8183056796854173, "全帯域_40ms_log_rms_mae_db": 0.578511858171383, "全帯域_40ms_rms_envelope_nrmse": 0.05052671672635328, "全帯域_40ms_envelope_correlation": 0.9978476274857444, "全帯域_40ms_modulation_mae_db": 1.1536129489493427, "300–1000 Hz_10ms_log_rms_mae_db": 1.761905542880294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17727196841629886, "300–1000 Hz_10ms_envelope_correlation": 0.9802563280312746, "300–1000 Hz_10ms_modulation_mae_db": 1.5765939176425354, "300–1000 Hz_40ms_log_rms_mae_db": 1.0104832040379672, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10160971526352695, "300–1000 Hz_40ms_envelope_correlation": 0.9935452934464095, "300–1000 Hz_40ms_modulation_mae_db": 1.5911963756535703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3706620583320082, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2083513031429922, "1000–3000 Hz_10ms_envelope_correlation": 0.9734817667345866, "1000–3000 Hz_10ms_modulation_mae_db": 0.7799644093554717, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8298188793176139, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10048288751927145, "1000–3000 Hz_40ms_envelope_correlation": 0.9949418354052674, "1000–3000 Hz_40ms_modulation_mae_db": 0.7328435858954743, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6625971242399267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2220870674032964, "3000–6000 Hz_10ms_envelope_correlation": 0.9664494270448185, "3000–6000 Hz_10ms_modulation_mae_db": 0.840742577754734, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1146860149564572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1104814880325298, "3000–6000 Hz_40ms_envelope_correlation": 0.991294118447039, "3000–6000 Hz_40ms_modulation_mae_db": 0.7511650331768808, "mr_stft_db": 5.745314882360016, "band_mae_db": 0.26145706137944785, "rms_gain_error_db": -0.04504100403333483, "output_peak": 0.7784620461053855}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1125708699517214, "全帯域_10ms_rms_envelope_nrmse": 0.13602351500188772, "全帯域_10ms_envelope_correlation": 0.9860491577378946, "全帯域_10ms_modulation_mae_db": 1.258246415326847, "全帯域_40ms_log_rms_mae_db": 0.5789303648953308, "全帯域_40ms_rms_envelope_nrmse": 0.07886335255944198, "全帯域_40ms_envelope_correlation": 0.9966830919315186, "全帯域_40ms_modulation_mae_db": 1.479583634764289, "300–1000 Hz_10ms_log_rms_mae_db": 1.8705959280192292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23600569589657983, "300–1000 Hz_10ms_envelope_correlation": 0.9640873943196727, "300–1000 Hz_10ms_modulation_mae_db": 1.689802125953265, "300–1000 Hz_40ms_log_rms_mae_db": 0.9717350711918652, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1317368069036485, "300–1000 Hz_40ms_envelope_correlation": 0.9913259604087692, "300–1000 Hz_40ms_modulation_mae_db": 1.66119484626595, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3039504741215677, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18039267810799045, "1000–3000 Hz_10ms_envelope_correlation": 0.9826447417081837, "1000–3000 Hz_10ms_modulation_mae_db": 1.257817131662891, "1000–3000 Hz_40ms_log_rms_mae_db": 0.810800728282026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11147282324750957, "1000–3000 Hz_40ms_envelope_correlation": 0.9960994635889632, "1000–3000 Hz_40ms_modulation_mae_db": 1.2571507446286212, "3000–6000 Hz_10ms_log_rms_mae_db": 1.644077201238301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.218213271075183, "3000–6000 Hz_10ms_envelope_correlation": 0.9672905045825806, "3000–6000 Hz_10ms_modulation_mae_db": 0.7631378793976707, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0562057428321234, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08764335033553773, "3000–6000 Hz_40ms_envelope_correlation": 0.9945338748275103, "3000–6000 Hz_40ms_modulation_mae_db": 0.48290115603521205, "mr_stft_db": 5.744265120012776, "band_mae_db": 0.4481825901116143, "rms_gain_error_db": -0.4328417894255659, "output_peak": 0.8105984946475797}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0415481481836746, "全帯域_10ms_rms_envelope_nrmse": 0.12292749356324184, "全帯域_10ms_envelope_correlation": 0.9881841304685168, "全帯域_10ms_modulation_mae_db": 0.629474460608017, "全帯域_40ms_log_rms_mae_db": 0.5328079035829769, "全帯域_40ms_rms_envelope_nrmse": 0.06560619978786178, "全帯域_40ms_envelope_correlation": 0.9973584625193418, "全帯域_40ms_modulation_mae_db": 0.866549362448095, "300–1000 Hz_10ms_log_rms_mae_db": 1.7455011235412656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1645874761485304, "300–1000 Hz_10ms_envelope_correlation": 0.9836142923997223, "300–1000 Hz_10ms_modulation_mae_db": 0.9310441927530458, "300–1000 Hz_40ms_log_rms_mae_db": 1.0271725086212176, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09323189234390927, "300–1000 Hz_40ms_envelope_correlation": 0.9965125174660888, "300–1000 Hz_40ms_modulation_mae_db": 1.3841572618176787, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3403669333462895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2153635091360674, "1000–3000 Hz_10ms_envelope_correlation": 0.9712524991168342, "1000–3000 Hz_10ms_modulation_mae_db": 2.4822428954507894, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8274665217179343, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09986580361969948, "1000–3000 Hz_40ms_envelope_correlation": 0.9943631146849352, "1000–3000 Hz_40ms_modulation_mae_db": 2.48753763172818, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5823504717394086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20109575255358056, "3000–6000 Hz_10ms_envelope_correlation": 0.9721393754915207, "3000–6000 Hz_10ms_modulation_mae_db": 0.7550463312318255, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0885380442684485, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11903754517772121, "3000–6000 Hz_40ms_envelope_correlation": 0.9899934491868981, "3000–6000 Hz_40ms_modulation_mae_db": 0.7234402124925827, "mr_stft_db": 5.757247119098821, "band_mae_db": 0.39523831089395123, "rms_gain_error_db": -0.31733130392887776, "output_peak": 0.7744621926335259}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1160029103886038, "全帯域_10ms_rms_envelope_nrmse": 0.12731763123523002, "全帯域_10ms_envelope_correlation": 0.9866715957201296, "全帯域_10ms_modulation_mae_db": 0.8067129066685149, "全帯域_40ms_log_rms_mae_db": 0.5697365279336565, "全帯域_40ms_rms_envelope_nrmse": 0.055732950387628405, "全帯域_40ms_envelope_correlation": 0.9975307960182263, "全帯域_40ms_modulation_mae_db": 0.8606586662697094, "300–1000 Hz_10ms_log_rms_mae_db": 1.7455022869123904, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17865826307777552, "300–1000 Hz_10ms_envelope_correlation": 0.9787984970540723, "300–1000 Hz_10ms_modulation_mae_db": 1.2083944593184535, "300–1000 Hz_40ms_log_rms_mae_db": 0.9621681003994195, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08574030084429748, "300–1000 Hz_40ms_envelope_correlation": 0.9950813133120303, "300–1000 Hz_40ms_modulation_mae_db": 1.0258058239602132, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3427569797116583, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19762142341517813, "1000–3000 Hz_10ms_envelope_correlation": 0.9762467999220824, "1000–3000 Hz_10ms_modulation_mae_db": 1.441330718844896, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8535300493405507, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10600946327169, "1000–3000 Hz_40ms_envelope_correlation": 0.9940626336724496, "1000–3000 Hz_40ms_modulation_mae_db": 1.5536295276285224, "3000–6000 Hz_10ms_log_rms_mae_db": 1.700487853107045, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21104287313327, "3000–6000 Hz_10ms_envelope_correlation": 0.9696806480387223, "3000–6000 Hz_10ms_modulation_mae_db": 0.6137649228480798, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1150601357418064, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11857722958323237, "3000–6000 Hz_40ms_envelope_correlation": 0.989970300643261, "3000–6000 Hz_40ms_modulation_mae_db": 0.5166253503140492, "mr_stft_db": 5.761282956129986, "band_mae_db": 0.2541486177873899, "rms_gain_error_db": -0.14280354322733943, "output_peak": 0.8565730876404621}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1150677935407625, "全帯域_10ms_rms_envelope_nrmse": 0.11942785223901033, "全帯域_10ms_envelope_correlation": 0.9885512082272779, "全帯域_10ms_modulation_mae_db": 0.6098475237334708, "全帯域_40ms_log_rms_mae_db": 0.5985775535845764, "全帯域_40ms_rms_envelope_nrmse": 0.06022032017932629, "全帯域_40ms_envelope_correlation": 0.9975040178907987, "全帯域_40ms_modulation_mae_db": 0.9527031290433422, "300–1000 Hz_10ms_log_rms_mae_db": 1.7546250329794812, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16282431431635191, "300–1000 Hz_10ms_envelope_correlation": 0.9827616947334968, "300–1000 Hz_10ms_modulation_mae_db": 0.8002826984867838, "300–1000 Hz_40ms_log_rms_mae_db": 1.0007426568490945, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09289463189067686, "300–1000 Hz_40ms_envelope_correlation": 0.9949058974588456, "300–1000 Hz_40ms_modulation_mae_db": 1.0104937750963583, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3660617370219976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20053643764407172, "1000–3000 Hz_10ms_envelope_correlation": 0.9751162882090271, "1000–3000 Hz_10ms_modulation_mae_db": 1.0212328477403678, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7830481864125309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07688892158166284, "1000–3000 Hz_40ms_envelope_correlation": 0.9963013821794029, "1000–3000 Hz_40ms_modulation_mae_db": 0.6356444365929345, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5396472304046414, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19335977033963891, "3000–6000 Hz_10ms_envelope_correlation": 0.9747787660269027, "3000–6000 Hz_10ms_modulation_mae_db": 0.515903946633486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0280034857116704, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10345334892962402, "3000–6000 Hz_40ms_envelope_correlation": 0.9924623047480968, "3000–6000 Hz_40ms_modulation_mae_db": 0.49499608674299855, "mr_stft_db": 5.749940029389769, "band_mae_db": 0.3182690811928946, "rms_gain_error_db": -0.24186420671766298, "output_peak": 0.7591120456044138}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49600, "output_samples": 49600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.82016242940866, "spectral_convergence": 0.46310544991367303, "spectral_convergence_status": "ok", "amplitude_floor": 1.4924605057201793e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5308485778737735, "spectral_convergence": 0.4407012482526226, "spectral_convergence_status": "ok", "amplitude_floor": 8.127484488587521e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.854496221508069, "spectral_convergence": 0.5077170345460413, "spectral_convergence_status": "ok", "amplitude_floor": 5.495951303547364e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185024372796483, "output_power": 0.0007803041055242217, "delta_db": -0.20756085025342322, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004929662452993321, "output_power": 0.0005174319505202624, "delta_db": 0.21036059129925694, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001304221454098428, "output_power": 0.0011733859460967032, "delta_db": -0.45910457689426054, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007846801909808332, "output_power": 0.0008003094802063757, "delta_db": 0.08565272323954516, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003122815376510157, "output_power": 0.00032269854057205625, "delta_db": 0.14250692147955674, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001061689290478197, "output_power": 0.00011401223465021735, "delta_db": 0.30954021424140177, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23578764623457404, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12772891854389767, "pitch": null, "output_rms_dbfs": -24.308436131217984, "output_peak": 0.7914895279544425, "reference_rms_dbfs": -24.180707212674086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7351690762635}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.0002690918479937382, "log_rms_mae_db": 1.0308360882556955, "rms_envelope_nrmse": 0.08835041789385728, "envelope_correlation": 0.9936620279507402, "rms_gain_error_db": -0.12772862019661288, "reference_mean_power": 0.0038249880205223005, "output_mean_power": 0.0037141309979579294}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 2.1583633605855604e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008154057878800693, "output_power": 0.0007144771046835308, "reference_active": true, "delta_db": -0.5738547337292939}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.087495849252009e-05, "output_power": 9.183967330899704e-05, "reference_active": true, "delta_db": 0.045861046879384186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.599429116228097e-05, "output_power": 2.4646603817844382e-05, "reference_active": true, "delta_db": -0.23120895096758054}], "active_band_count": 3, "active_band_mae_db": 0.2836415771920862}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.0002456731948415248, "log_rms_mae_db": 0.5437351545875014, "rms_envelope_nrmse": 0.05429969339504889, "envelope_correlation": 0.9976263209989932, "rms_gain_error_db": -0.12772934493031732, "reference_mean_power": 0.0038624153279842384, "output_mean_power": 0.0037504729493954203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 2.0216648233438394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007562450561775054, "output_power": 0.0006591529821304634, "reference_active": true, "delta_db": -0.5967632729663955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1399930780791825e-05, "output_power": 4.215273665997286e-05, "reference_active": true, "delta_db": 0.07826160351890873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0921358976760216e-06, "output_power": 1.1302974950757704e-06, "reference_active": true, "delta_db": 0.14916082865702873}], "active_band_count": 3, "active_band_mae_db": 0.2747285683807777}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00017000794064843354, "log_rms_mae_db": 1.7461244576406072, "rms_envelope_nrmse": 0.18181261346423466, "envelope_correlation": 0.9781137845534891, "rms_gain_error_db": -0.27214711733208213, "reference_mean_power": 0.001461911046007658, "output_mean_power": 0.001373112864002421}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.2056298552980857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006038941034712422, "output_power": 0.0005018108394487332, "reference_active": true, "delta_db": -0.8042075097063868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.14042207191201e-05, "output_power": 7.5654883289879e-05, "reference_active": true, "delta_db": 0.2511308185932607}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.020128231954835e-05, "output_power": 2.8232925620404183e-05, "reference_active": true, "delta_db": 1.4537694589890084}], "active_band_count": 3, "active_band_mae_db": 0.8363692624295519}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00015858554889566586, "log_rms_mae_db": 0.9966874292073349, "rms_envelope_nrmse": 0.0946664809556412, "envelope_correlation": 0.9943350971606282, "rms_gain_error_db": -0.2721452133104562, "reference_mean_power": 0.0014762175799508475, "output_mean_power": 0.0013865510102754135}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.148107626412446e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000571116387197927, "output_power": 0.00048220037369098875, "reference_active": true, "delta_db": -0.7349707906640691}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.637477969141665e-05, "output_power": 3.6540369572287716e-05, "reference_active": true, "delta_db": 0.01972563887733368}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.172790453481755e-06, "output_power": 1.3059054701317323e-06, "reference_active": true, "delta_db": 0.4669131888570371}], "active_band_count": 3, "active_band_mae_db": 0.4072032061328133}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00019136572750236326, "log_rms_mae_db": 1.3603968347273623, "rms_envelope_nrmse": 0.12718456809313095, "envelope_correlation": 0.9901235407161822, "rms_gain_error_db": 0.025508883939420295, "reference_mean_power": 0.0009651549113833473, "output_mean_power": 0.000970840562933086}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 7.99902404373172e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003359743599084894, "output_power": 0.00033411540711023426, "reference_active": true, "delta_db": -0.024096324201682305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9966531082920368e-05, "output_power": 3.277251645909664e-05, "reference_active": true, "delta_db": 0.38873318198328555}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.661292025781732e-06, "output_power": 8.839226652735308e-06, "reference_active": true, "delta_db": 1.2285579688625499}], "active_band_count": 3, "active_band_mae_db": 0.5471291583491725}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00016758862552795117, "log_rms_mae_db": 0.7841620936659022, "rms_envelope_nrmse": 0.056065453307287016, "envelope_correlation": 0.9980630953561934, "rms_gain_error_db": 0.025508191408106985, "reference_mean_power": 0.0009746009072787481, "output_mean_power": 0.0009803420481196085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 7.731649944214162e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031280752850392277, "output_power": 0.0003156061473391088, "reference_active": true, "delta_db": 0.03868256848022043}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5385931815426458e-05, "output_power": 1.688150916063782e-05, "reference_active": true, "delta_db": 0.40287465282486457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1192579703922553e-07, "output_power": 3.5936042080985307e-07, "reference_active": true, "delta_db": 0.6147894966278732}], "active_band_count": 3, "active_band_mae_db": 0.3521155726443194}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 5.61189760356505e-05, "log_rms_mae_db": 1.5611970985968069, "rms_envelope_nrmse": 0.2048687287156367, "envelope_correlation": 0.9735602248775524, "rms_gain_error_db": 0.32565603533723897, "reference_mean_power": 0.00014982208037105655, "output_mean_power": 0.00016148843536759597}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.5994961899965245e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.519717012759751e-05, "output_power": 6.21011063044894e-05, "reference_active": true, "delta_db": 0.5118252431231303}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5683811710113524e-05, "output_power": 1.549995868437075e-05, "reference_active": true, "delta_db": -0.051210792494486214}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1546072478197644e-06, "output_power": 3.6818481498921083e-06, "reference_active": true, "delta_db": 0.6712057618173912}], "active_band_count": 3, "active_band_mae_db": 0.4114139324783359}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 4.51769085602287e-05, "log_rms_mae_db": 1.0281033784240998, "rms_envelope_nrmse": 0.09461033257577618, "envelope_correlation": 0.9949258995251715, "rms_gain_error_db": 0.3256571986309813, "reference_mean_power": 0.0001512883740521334, "output_mean_power": 0.00016306895017484662}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.4928738695410609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.9695472955753735e-05, "output_power": 5.5986325922035185e-05, "reference_active": true, "delta_db": 0.5176513987382421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.547268075821193e-06, "output_power": 8.34054624570126e-06, "reference_active": true, "delta_db": -0.10632830620094542}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7628269396099532e-07, "output_power": 1.8486965023019363e-07, "reference_active": true, "delta_db": 0.20655940746479312}], "active_band_count": 3, "active_band_mae_db": 0.2768463708013269}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8584619635077169, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "049d5d8d3e5fa5bca920c78b890b4479de340a28b25e60f4f1cbd1e83db55847", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9793449369619613, "standard_deviation": 0.007239285876011184, "minimum": 0.9712524991168342, "maximum": 0.9901235407161822, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.350768807970107, "standard_deviation": 0.021696292940654845, "minimum": 1.3039504741215677, "maximum": 1.3706620583320082, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1725580123204926, "standard_deviation": 0.6596998308351842, "minimum": 0.3956310792568716, "maximum": 2.4822428954507894, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18207757632499455, "standard_deviation": 0.03419003683289021, "minimum": 0.12718456809313095, "maximum": 0.2153635091360674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995612496966692, "standard_deviation": 0.0018103602119926548, "minimum": 0.9931052724830863, "maximum": 0.9980630953561934, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8142038544816372, "standard_deviation": 0.02594056553580487, "minimum": 0.7830481864125309, "maximum": 0.8535300493405507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1353962325814688, "standard_deviation": 0.7505231113410131, "minimum": 0.3521155726443194, "maximum": 2.48753763172818, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08964815410783472, "standard_deviation": 0.021748369924366656, "minimum": 0.056065453307287016, "maximum": 0.11147282324750957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9717619129463304, "standard_deviation": 0.0036339885553969896, "minimum": 0.9664494270448185, "maximum": 0.9767035294251902, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6012908247255568, "standard_deviation": 0.05929277914272655, "minimum": 1.5396472304046414, "maximum": 1.700487853107045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.61442486050981, "standard_deviation": 0.16123779657689769, "minimum": 0.4114139324783359, "maximum": 0.840742577754734, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20434502932031284, "standard_deviation": 0.012428094992982348, "minimum": 0.18514666056896956, "maximum": 0.2220870674032964, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992161627812808, "standard_deviation": 0.0019277642466002673, "minimum": 0.989970300643261, "maximum": 0.9949258995251715, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0655218479379238, "standard_deviation": 0.036346692427157976, "minimum": 1.0280034857116704, "maximum": 1.1150601357418064, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5539895337304008, "standard_deviation": 0.14997940135484006, "minimum": 0.2768463708013269, "maximum": 0.7511650331768808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10576769732810486, "standard_deviation": 0.011650690835437624, "minimum": 0.08764335033553773, "maximum": 0.11903754517772121, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.978197840620638, "standard_deviation": 0.006868285662103718, "minimum": 0.9640873943196727, "maximum": 0.9851043027401243, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.769709448958732, "standard_deviation": 0.0428995593883734, "minimum": 1.7455011235412656, "maximum": 1.8705959280192292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1688879049396945, "standard_deviation": 0.3269692921936331, "minimum": 0.8002826984867838, "maximum": 1.689802125953265, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18182757711871977, "standard_deviation": 0.026496425526681116, "minimum": 0.1518499869883277, "maximum": 0.23600569589657983, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946044882579028, "standard_deviation": 0.00180011472324315, "minimum": 0.9913259604087692, "maximum": 0.9971359103853563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9978529061792696, "standard_deviation": 0.029792024013679187, "minimum": 0.9621681003994195, "maximum": 1.0460592718209234, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.192893493127821, "standard_deviation": 0.3983836357822937, "minimum": 0.4072032061328133, "maximum": 1.66119484626595, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09601689646628193, "standard_deviation": 0.01665977662326539, "minimum": 0.07345056249942944, "maximum": 0.1317368069036485, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.30332867724359813, "standard_deviation": 0.09938556981503652, "minimum": 0.1418369105425593, "maximum": 0.4481825901116143, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7508425520291935, "standard_deviation": 0.009975419633331196, "minimum": 5.7351690762635, "maximum": 5.765598188367029, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7766136244287752, "standard_deviation": 0.0524922907981956, "minimum": 0.6708976494335318, "maximum": 0.8565730876404621, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.18662655200895273, "standard_deviation": 0.1342360006533701, "minimum": -0.4328417894255659, "maximum": -0.04504100403333483, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9895131264277645, "standard_deviation": 0.003286648569001843, "minimum": 0.9860491577378946, "maximum": 0.994475061931948, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0958664144725954, "standard_deviation": 0.040565762694952004, "minimum": 1.0308360882556955, "maximum": 1.148758671349912, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6994889979889032, "standard_deviation": 0.28514920407732025, "minimum": 0.2836415771920862, "maximum": 1.258246415326847, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11301572673387156, "standard_deviation": 0.0196972811947617, "minimum": 0.08261131843111799, "maximum": 0.13602351500188772, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974705770925693, "standard_deviation": 0.0008104760961346713, "minimum": 0.9962365796467045, "maximum": 0.9989777202492266, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5792244618962579, "standard_deviation": 0.030583140443001387, "minimum": 0.5328079035829769, "maximum": 0.6160581964441086, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.915091840141295, "standard_deviation": 0.3375489844125671, "minimum": 0.2747285683807777, "maximum": 1.479583634764289, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05866823864166689, "standard_deviation": 0.012410389474167243, "minimum": 0.03724788486887327, "maximum": 0.07886335255944198, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0010", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5503838062286377, "predicted_mos": 2.5503838062286377, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "1fc8411f7750b12edb125c9dd043487f8bb8616376688025404b9eaffba57ce9", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7029288997873664, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0267104681443844, "全帯域_10ms_rms_envelope_nrmse": 0.12204105579534902, "全帯域_10ms_envelope_correlation": 0.9882105434425928, "全帯域_10ms_modulation_mae_db": 1.1052510378256648, "全帯域_40ms_log_rms_mae_db": 0.5209272780332432, "全帯域_40ms_rms_envelope_nrmse": 0.05267070719658996, "全帯域_40ms_envelope_correlation": 0.9978899300995125, "全帯域_40ms_modulation_mae_db": 1.1953325455108466, "300–1000 Hz_10ms_log_rms_mae_db": 1.9973112727844997, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22792240556934015, "300–1000 Hz_10ms_envelope_correlation": 0.9659540216960241, "300–1000 Hz_10ms_modulation_mae_db": 1.2658219785118459, "300–1000 Hz_40ms_log_rms_mae_db": 1.2211126639168253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10671371986670977, "300–1000 Hz_40ms_envelope_correlation": 0.9922728280220824, "300–1000 Hz_40ms_modulation_mae_db": 1.490635353138848, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6744488046765589, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16077908463298826, "1000–3000 Hz_10ms_envelope_correlation": 0.9852012677084545, "1000–3000 Hz_10ms_modulation_mae_db": 0.8683180414836628, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0859748648856624, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0778936907441462, "1000–3000 Hz_40ms_envelope_correlation": 0.996907788683052, "1000–3000 Hz_40ms_modulation_mae_db": 0.7259201982099972, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0081345699985493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20543906139077814, "3000–6000 Hz_10ms_envelope_correlation": 0.9715394736924472, "3000–6000 Hz_10ms_modulation_mae_db": 0.449398844287495, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4129548782328138, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11163595478463496, "3000–6000 Hz_40ms_envelope_correlation": 0.991272993175437, "3000–6000 Hz_40ms_modulation_mae_db": 0.3511134430221387, "mr_stft_db": 5.9961972730103135, "band_mae_db": 0.26148621225221086, "rms_gain_error_db": 0.11076995309390993, "output_peak": 0.7844780861779458}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0642182052229034, "全帯域_10ms_rms_envelope_nrmse": 0.09489090340116711, "全帯域_10ms_envelope_correlation": 0.9928806289966425, "全帯域_10ms_modulation_mae_db": 0.4312668320601503, "全帯域_40ms_log_rms_mae_db": 0.5514978641540733, "全帯域_40ms_rms_envelope_nrmse": 0.0566329207527217, "全帯域_40ms_envelope_correlation": 0.9974691759004921, "全帯域_40ms_modulation_mae_db": 0.7443485552837433, "300–1000 Hz_10ms_log_rms_mae_db": 1.9136963403570422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20097881917926108, "300–1000 Hz_10ms_envelope_correlation": 0.9731422175751094, "300–1000 Hz_10ms_modulation_mae_db": 1.2756317108845476, "300–1000 Hz_40ms_log_rms_mae_db": 1.1682155383811905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09413365467675676, "300–1000 Hz_40ms_envelope_correlation": 0.9943301913610486, "300–1000 Hz_40ms_modulation_mae_db": 1.2084828225378386, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6642102937407226, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.216519318082874, "1000–3000 Hz_10ms_envelope_correlation": 0.9754564449536904, "1000–3000 Hz_10ms_modulation_mae_db": 2.6028222701514347, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0859099086655217, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12946822858661622, "1000–3000 Hz_40ms_envelope_correlation": 0.9926086562550491, "1000–3000 Hz_40ms_modulation_mae_db": 2.5232592179950726, "3000–6000 Hz_10ms_log_rms_mae_db": 2.029300098269719, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19118216643031946, "3000–6000 Hz_10ms_envelope_correlation": 0.976043804042558, "3000–6000 Hz_10ms_modulation_mae_db": 0.6393254813071534, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3648661630777075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10127360717233992, "3000–6000 Hz_40ms_envelope_correlation": 0.9932910573850509, "3000–6000 Hz_40ms_modulation_mae_db": 0.6781435486292139, "mr_stft_db": 6.0015139681431675, "band_mae_db": 0.34577550319104905, "rms_gain_error_db": 0.06154797293431703, "output_peak": 0.8359716481406215}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0284516305184297, "全帯域_10ms_rms_envelope_nrmse": 0.1235304153631194, "全帯域_10ms_envelope_correlation": 0.987542807571841, "全帯域_10ms_modulation_mae_db": 0.663270027091503, "全帯域_40ms_log_rms_mae_db": 0.49821071943113476, "全帯域_40ms_rms_envelope_nrmse": 0.059330816215293194, "全帯域_40ms_envelope_correlation": 0.9970493314490478, "全帯域_40ms_modulation_mae_db": 0.768324155860892, "300–1000 Hz_10ms_log_rms_mae_db": 1.9506563966723045, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2011618138806263, "300–1000 Hz_10ms_envelope_correlation": 0.9732558743582357, "300–1000 Hz_10ms_modulation_mae_db": 0.6563999073826667, "300–1000 Hz_40ms_log_rms_mae_db": 1.140562736684522, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12287434740927715, "300–1000 Hz_40ms_envelope_correlation": 0.9897164805012839, "300–1000 Hz_40ms_modulation_mae_db": 0.5213354093024573, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6005672214089972, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19508932852778257, "1000–3000 Hz_10ms_envelope_correlation": 0.9767466750749026, "1000–3000 Hz_10ms_modulation_mae_db": 1.407915860126405, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0039744088599887, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0710741826570007, "1000–3000 Hz_40ms_envelope_correlation": 0.9968997875387208, "1000–3000 Hz_40ms_modulation_mae_db": 1.5121920867641911, "3000–6000 Hz_10ms_log_rms_mae_db": 2.110393323735939, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21783964926172608, "3000–6000 Hz_10ms_envelope_correlation": 0.9690357778284376, "3000–6000 Hz_10ms_modulation_mae_db": 0.5723151306426915, "3000–6000 Hz_40ms_log_rms_mae_db": 1.409444660517048, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11139415606965981, "3000–6000 Hz_40ms_envelope_correlation": 0.9919488416445038, "3000–6000 Hz_40ms_modulation_mae_db": 0.3335010597365132, "mr_stft_db": 6.018664501682477, "band_mae_db": 0.1154810325886048, "rms_gain_error_db": -0.0022329830322743954, "output_peak": 0.7702772930820337}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1200715843152305, "全帯域_10ms_rms_envelope_nrmse": 0.10061375897477652, "全帯域_10ms_envelope_correlation": 0.9919604932514923, "全帯域_10ms_modulation_mae_db": 0.9282285516987617, "全帯域_40ms_log_rms_mae_db": 0.5426537754071927, "全帯域_40ms_rms_envelope_nrmse": 0.0533686085329397, "全帯域_40ms_envelope_correlation": 0.9980236633311718, "全帯域_40ms_modulation_mae_db": 1.1968480566615696, "300–1000 Hz_10ms_log_rms_mae_db": 1.9033183301667733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16341411245441465, "300–1000 Hz_10ms_envelope_correlation": 0.9826142153750389, "300–1000 Hz_10ms_modulation_mae_db": 0.7646022702426531, "300–1000 Hz_40ms_log_rms_mae_db": 1.135161826638704, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09710087254001952, "300–1000 Hz_40ms_envelope_correlation": 0.9942804554577442, "300–1000 Hz_40ms_modulation_mae_db": 0.6294430822549223, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6797685712812427, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1716601197303345, "1000–3000 Hz_10ms_envelope_correlation": 0.9823735993920667, "1000–3000 Hz_10ms_modulation_mae_db": 1.2653061879483807, "1000–3000 Hz_40ms_log_rms_mae_db": 1.093826093670432, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09710637556914556, "1000–3000 Hz_40ms_envelope_correlation": 0.9952361988810572, "1000–3000 Hz_40ms_modulation_mae_db": 1.4407304282740014, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0229068001694075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20235693284845252, "3000–6000 Hz_10ms_envelope_correlation": 0.9717164498194726, "3000–6000 Hz_10ms_modulation_mae_db": 0.4013941808834522, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3694032589079494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1142366521691833, "3000–6000 Hz_40ms_envelope_correlation": 0.990878481510625, "3000–6000 Hz_40ms_modulation_mae_db": 0.41959243393355883, "mr_stft_db": 6.005170110280649, "band_mae_db": 0.3290622593552753, "rms_gain_error_db": -0.21431571898855373, "output_peak": 0.78803475508326}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.069079597088298, "全帯域_10ms_rms_envelope_nrmse": 0.11411252125708121, "全帯域_10ms_envelope_correlation": 0.989449516474267, "全帯域_10ms_modulation_mae_db": 0.7000651933236398, "全帯域_40ms_log_rms_mae_db": 0.572561064907221, "全帯域_40ms_rms_envelope_nrmse": 0.06430879911362954, "全帯域_40ms_envelope_correlation": 0.9968366712384882, "全帯域_40ms_modulation_mae_db": 0.6524719626939424, "300–1000 Hz_10ms_log_rms_mae_db": 1.9182089169927734, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2107783632685918, "300–1000 Hz_10ms_envelope_correlation": 0.970349327005644, "300–1000 Hz_10ms_modulation_mae_db": 0.83931486973056, "300–1000 Hz_40ms_log_rms_mae_db": 1.1254868747324756, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.126298583411218, "300–1000 Hz_40ms_envelope_correlation": 0.9894892832263851, "300–1000 Hz_40ms_modulation_mae_db": 0.7257383007494026, "1000–3000 Hz_10ms_log_rms_mae_db": 1.694935437439004, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16609623789266698, "1000–3000 Hz_10ms_envelope_correlation": 0.9837160369689061, "1000–3000 Hz_10ms_modulation_mae_db": 1.2934469390131764, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0759144315947191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09553595312527806, "1000–3000 Hz_40ms_envelope_correlation": 0.9956064167847868, "1000–3000 Hz_40ms_modulation_mae_db": 1.18094338966534, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9835852274651737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22135965901045748, "3000–6000 Hz_10ms_envelope_correlation": 0.9662285156106434, "3000–6000 Hz_10ms_modulation_mae_db": 1.0083312528027912, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3718513005232646, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11415555919246538, "3000–6000 Hz_40ms_envelope_correlation": 0.9907046911864874, "3000–6000 Hz_40ms_modulation_mae_db": 1.0965376338480708, "mr_stft_db": 5.996627658793685, "band_mae_db": 0.2367165736042607, "rms_gain_error_db": -0.1981599841259499, "output_peak": 0.7363317484852404}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0182523340277136, "全帯域_10ms_rms_envelope_nrmse": 0.1039694338559413, "全帯域_10ms_envelope_correlation": 0.9914044222864584, "全帯域_10ms_modulation_mae_db": 0.794362561406227, "全帯域_40ms_log_rms_mae_db": 0.5689082361144712, "全帯域_40ms_rms_envelope_nrmse": 0.05253527113348546, "全帯域_40ms_envelope_correlation": 0.9981485497068658, "全帯域_40ms_modulation_mae_db": 0.5349430092062497, "300–1000 Hz_10ms_log_rms_mae_db": 1.9587280991418385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1973129701290917, "300–1000 Hz_10ms_envelope_correlation": 0.9741473969160855, "300–1000 Hz_10ms_modulation_mae_db": 0.6746109501652037, "300–1000 Hz_40ms_log_rms_mae_db": 1.1262277306148385, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08629986144384698, "300–1000 Hz_40ms_envelope_correlation": 0.994892867119806, "300–1000 Hz_40ms_modulation_mae_db": 0.44282354822746317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6542511619439488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19729729088818318, "1000–3000 Hz_10ms_envelope_correlation": 0.9762080156562833, "1000–3000 Hz_10ms_modulation_mae_db": 1.1638630289152392, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0876232565057735, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11256433154023428, "1000–3000 Hz_40ms_envelope_correlation": 0.9930727816827418, "1000–3000 Hz_40ms_modulation_mae_db": 1.2341409993574632, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0024379163693444, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20405141838506394, "3000–6000 Hz_10ms_envelope_correlation": 0.9715825509935928, "3000–6000 Hz_10ms_modulation_mae_db": 0.4735195860700676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3678785692496584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09101988036062109, "3000–6000 Hz_40ms_envelope_correlation": 0.9941063060715185, "3000–6000 Hz_40ms_modulation_mae_db": 0.7659263525620018, "mr_stft_db": 6.002598426622474, "band_mae_db": 0.27564477707387236, "rms_gain_error_db": -0.22242342688020794, "output_peak": 0.6343188338802748}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0819291042411634, "全帯域_10ms_rms_envelope_nrmse": 0.1067964908787189, "全帯域_10ms_envelope_correlation": 0.9906416299279325, "全帯域_10ms_modulation_mae_db": 0.39522469801927923, "全帯域_40ms_log_rms_mae_db": 0.6268522766754654, "全帯域_40ms_rms_envelope_nrmse": 0.05610174710654963, "全帯域_40ms_envelope_correlation": 0.9974051208856772, "全帯域_40ms_modulation_mae_db": 0.45586479329951385, "300–1000 Hz_10ms_log_rms_mae_db": 2.0636058104515422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20721603842498781, "300–1000 Hz_10ms_envelope_correlation": 0.9712974872970362, "300–1000 Hz_10ms_modulation_mae_db": 0.8923127964626872, "300–1000 Hz_40ms_log_rms_mae_db": 1.21600214907589, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11095123883505667, "300–1000 Hz_40ms_envelope_correlation": 0.9919503096405244, "300–1000 Hz_40ms_modulation_mae_db": 0.8553121756714263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5992372556385648, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18562508342401599, "1000–3000 Hz_10ms_envelope_correlation": 0.9790522603258757, "1000–3000 Hz_10ms_modulation_mae_db": 1.8238274343066667, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0504532412414773, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10480957532974065, "1000–3000 Hz_40ms_envelope_correlation": 0.9932544706557486, "1000–3000 Hz_40ms_modulation_mae_db": 2.031015652973901, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1198082374069305, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23416333168925374, "3000–6000 Hz_10ms_envelope_correlation": 0.9639008327199566, "3000–6000 Hz_10ms_modulation_mae_db": 1.078069832676604, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4691374518298614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13692134878132595, "3000–6000 Hz_40ms_envelope_correlation": 0.9873601841821421, "3000–6000 Hz_40ms_modulation_mae_db": 1.0383106140405687, "mr_stft_db": 6.015293552728977, "band_mae_db": 0.437736246456806, "rms_gain_error_db": -0.10578986971989451, "output_peak": 0.7871636288944205}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.088543397954277, "全帯域_10ms_rms_envelope_nrmse": 0.13154099056819907, "全帯域_10ms_envelope_correlation": 0.9870104741616526, "全帯域_10ms_modulation_mae_db": 1.1219369747385872, "全帯域_40ms_log_rms_mae_db": 0.5727043977413598, "全帯域_40ms_rms_envelope_nrmse": 0.06896195465927978, "全帯域_40ms_envelope_correlation": 0.9968459403568536, "全帯域_40ms_modulation_mae_db": 1.3724802341447384, "300–1000 Hz_10ms_log_rms_mae_db": 2.014706197245838, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1831159824657634, "300–1000 Hz_10ms_envelope_correlation": 0.97770344237141, "300–1000 Hz_10ms_modulation_mae_db": 0.4603891486144507, "300–1000 Hz_40ms_log_rms_mae_db": 1.199988197019096, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09617787445858354, "300–1000 Hz_40ms_envelope_correlation": 0.9937476210811417, "300–1000 Hz_40ms_modulation_mae_db": 0.7396421381512134, "1000–3000 Hz_10ms_log_rms_mae_db": 1.566158258136073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2622410973126319, "1000–3000 Hz_10ms_envelope_correlation": 0.9677253815075314, "1000–3000 Hz_10ms_modulation_mae_db": 3.373825436082999, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0186977810045517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14403926837990527, "1000–3000 Hz_40ms_envelope_correlation": 0.9941385514395642, "1000–3000 Hz_40ms_modulation_mae_db": 3.7670450926471997, "3000–6000 Hz_10ms_log_rms_mae_db": 2.03471200954468, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22649787103463492, "3000–6000 Hz_10ms_envelope_correlation": 0.9658337159003004, "3000–6000 Hz_10ms_modulation_mae_db": 0.5946786132391965, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4143008611499837, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12927571747818328, "3000–6000 Hz_40ms_envelope_correlation": 0.9885293576897808, "3000–6000 Hz_40ms_modulation_mae_db": 0.6482734408110945, "mr_stft_db": 5.987128983242935, "band_mae_db": 0.43815543927726447, "rms_gain_error_db": 0.20909876937142208, "output_peak": 0.8197818050200455}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49600, "output_samples": 49600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.109873115378427, "spectral_convergence": 0.4936148455140882, "spectral_convergence_status": "ok", "amplitude_floor": 1.4924605057201793e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.836676566998002, "spectral_convergence": 0.4526942330875065, "spectral_convergence_status": "ok", "amplitude_floor": 8.127484488587521e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.042042136654511, "spectral_convergence": 0.4848408714878478, "spectral_convergence_status": "ok", "amplitude_floor": 5.495951303547364e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185024372796483, "output_power": 0.0008141729016762648, "delta_db": -0.023033334115569452, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004929662452993321, "output_power": 0.0005973867031398444, "delta_db": 0.8343836867895114, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001304221454098428, "output_power": 0.0012360735037986168, "delta_db": -0.2330704287602688, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007846801909808332, "output_power": 0.0008522276843175685, "delta_db": 0.35862948885619106, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003122815376510157, "output_power": 0.000313824117849915, "delta_db": 0.021400075783699388, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001061689290478197, "output_power": 0.00010379045375556644, "delta_db": -0.09840025920802527, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.26148621225221086, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.11076995309390993, "pitch": null, "output_rms_dbfs": -24.069937259580175, "output_peak": 0.7844780861779458, "reference_rms_dbfs": -24.180707212674086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9961972730103135}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.0002690918479937382, "log_rms_mae_db": 1.0267104681443844, "rms_envelope_nrmse": 0.12204105579534902, "envelope_correlation": 0.9882105434425928, "rms_gain_error_db": 0.11076998882477726, "reference_mean_power": 0.0038249880205223005, "output_mean_power": 0.0039238019457395815}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 2.1583633605855604e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008154057878800693, "output_power": 0.0008845010439639582, "reference_active": true, "delta_db": 0.3532456001918576}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.087495849252009e-05, "output_power": 9.513578032404384e-05, "reference_active": true, "delta_db": 0.19899659204584474}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.599429116228097e-05, "output_power": 4.911668768233717e-05, "reference_active": true, "delta_db": 2.763510921239292}], "active_band_count": 3, "active_band_mae_db": 1.1052510378256648}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.0002456731948415248, "log_rms_mae_db": 0.5209272780332432, "rms_envelope_nrmse": 0.05267070719658996, "envelope_correlation": 0.9978899300995125, "rms_gain_error_db": 0.11076958235334437, "reference_mean_power": 0.0038624153279842384, "output_mean_power": 0.003962195771448963}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 2.0216648233438394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007562450561775054, "output_power": 0.00084021304783167, "reference_active": true, "delta_db": 0.4572687295369782}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1399930780791825e-05, "output_power": 4.729015661797977e-05, "reference_active": true, "delta_db": 0.5777113734005602}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0921358976760216e-06, "output_power": 1.9650718557869462e-06, "reference_active": true, "delta_db": 2.551017533595001}], "active_band_count": 3, "active_band_mae_db": 1.1953325455108466}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00017000794064843354, "log_rms_mae_db": 1.9973112727844997, "rms_envelope_nrmse": 0.22792240556934015, "envelope_correlation": 0.9659540216960241, "rms_gain_error_db": 0.05497317263117836, "reference_mean_power": 0.001461911046007658, "output_mean_power": 0.0014805335894535311}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.2056298552980857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006038941034712422, "output_power": 0.0006545197286258346, "reference_active": true, "delta_db": 0.34961952548560077}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.14042207191201e-05, "output_power": 7.292981786454885e-05, "reference_active": true, "delta_db": 0.09181245177361566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.020128231954835e-05, "output_power": 4.3750430814729654e-05, "reference_active": true, "delta_db": 3.3560339582763214}], "active_band_count": 3, "active_band_mae_db": 1.2658219785118459}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00015858554889566586, "log_rms_mae_db": 1.2211126639168253, "rms_envelope_nrmse": 0.10671371986670977, "envelope_correlation": 0.9922728280220824, "rms_gain_error_db": 0.05497193647621228, "reference_mean_power": 0.0014762175799508475, "output_mean_power": 0.001495021941541516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.148107626412446e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000571116387197927, "output_power": 0.0006468648307869893, "reference_active": true, "delta_db": 0.5408891807384417}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.637477969141665e-05, "output_power": 3.987072988110451e-05, "reference_active": true, "delta_db": 0.39853814370974566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.172790453481755e-06, "output_power": 2.6452595785298475e-06, "reference_active": true, "delta_db": 3.5324787349683566}], "active_band_count": 3, "active_band_mae_db": 1.490635353138848}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00019136572750236326, "log_rms_mae_db": 1.6744488046765589, "rms_envelope_nrmse": 0.16077908463298826, "envelope_correlation": 0.9852012677084545, "rms_gain_error_db": 0.2540111311258283, "reference_mean_power": 0.0009651549113833473, "output_mean_power": 0.001023288605302913}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 7.99902404373172e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003359743599084894, "output_power": 0.0003598576298397007, "reference_active": true, "delta_db": 0.29824579997658934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9966531082920368e-05, "output_power": 3.8641140458324027e-05, "reference_active": true, "delta_db": 1.1041346369151583}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.661292025781732e-06, "output_power": 8.786498555232278e-06, "reference_active": true, "delta_db": 1.2025736875592408}], "active_band_count": 3, "active_band_mae_db": 0.8683180414836628}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00016758862552795117, "log_rms_mae_db": 1.0859748648856624, "rms_envelope_nrmse": 0.0778936907441462, "envelope_correlation": 0.996907788683052, "rms_gain_error_db": 0.25401060652074703, "reference_mean_power": 0.0009746009072787481, "output_mean_power": 0.0010333034323354545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 7.731649944214162e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031280752850392277, "output_power": 0.0003396041825378363, "reference_active": true, "delta_db": 0.35695833432652263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5385931815426458e-05, "output_power": 1.958109586850596e-05, "reference_active": true, "delta_db": 1.047131902359468}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1192579703922553e-07, "output_power": 2.610262941744152e-07, "reference_active": true, "delta_db": -0.7736703579440007}], "active_band_count": 3, "active_band_mae_db": 0.7259201982099972}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 5.61189760356505e-05, "log_rms_mae_db": 2.0081345699985493, "rms_envelope_nrmse": 0.20543906139077814, "envelope_correlation": 0.9715394736924472, "rms_gain_error_db": 0.0930487734784445, "reference_mean_power": 0.00014982208037105655, "output_mean_power": 0.00015306669341701425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.5994961899965245e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.519717012759751e-05, "output_power": 4.977805613329182e-05, "reference_active": true, "delta_db": -0.44878879591619564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5683811710113524e-05, "output_power": 1.5172585478707197e-05, "reference_active": true, "delta_db": -0.14392026915294165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1546072478197644e-06, "output_power": 3.753998057578733e-06, "reference_active": true, "delta_db": 0.7554874677933477}], "active_band_count": 3, "active_band_mae_db": 0.449398844287495}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 4.51769085602287e-05, "log_rms_mae_db": 1.4129548782328138, "rms_envelope_nrmse": 0.11163595478463496, "envelope_correlation": 0.991272993175437, "rms_gain_error_db": 0.09304856705897117, "reference_mean_power": 0.0001512883740521334, "output_mean_power": 0.00015456473445426536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.4928738695410609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.9695472955753735e-05, "output_power": 4.52152630696891e-05, "reference_active": true, "delta_db": -0.4103176619243832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.547268075821193e-06, "output_power": 8.060211685800381e-06, "reference_active": true, "delta_db": -0.25480877493899684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7628269396099532e-07, "output_power": 1.9276628070053087e-07, "reference_active": true, "delta_db": 0.38821389220303604}], "active_band_count": 3, "active_band_mae_db": 0.3511134430221387}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.841586937221625, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b5a3fc388b7ba870e34f0edb8608378e01256cf67bf140619cce4fc3ba141391", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9783099601984638, "standard_deviation": 0.005616529746748244, "minimum": 0.9677253815075314, "maximum": 0.9852012677084545, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.641697125533139, "standard_deviation": 0.046631537378464906, "minimum": 1.566158258136073, "maximum": 1.694935437439004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.7249156497534957, "standard_deviation": 0.8490197319691379, "minimum": 0.8683180414836628, "maximum": 3.373825436082999, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19441344506143465, "standard_deviation": 0.033006745964865955, "minimum": 0.16077908463298826, "maximum": 0.2622410973126319, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9947155814900901, "standard_deviation": 0.001700709605358867, "minimum": 0.9926086562550491, "maximum": 0.996907788683052, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0627967483035157, "standard_deviation": 0.03459942478794174, "minimum": 1.0039744088599887, "maximum": 1.093826093670432, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.8019058832358958, "standard_deviation": 0.9647125702081835, "minimum": 0.7259201982099972, "maximum": 3.7670450926471997, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10406145074150837, "standard_deviation": 0.02452566479393935, "minimum": 0.0710741826570007, "maximum": 0.14403926837990527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9694851400759261, "standard_deviation": 0.003998913278414901, "minimum": 0.9639008327199566, "maximum": 0.976043804042558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.038909772869968, "standard_deviation": 0.04979764274338543, "minimum": 1.9835852274651737, "maximum": 2.1198082374069305, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6521291152386814, "standard_deviation": 0.25467322393092684, "minimum": 0.4013941808834522, "maximum": 1.078069832676604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21286126125633578, "standard_deviation": 0.014391311329761888, "minimum": 0.19118216643031946, "maximum": 0.23416333168925374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9910114891056931, "standard_deviation": 0.00224646254450044, "minimum": 0.9873601841821421, "maximum": 0.9941063060715185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3974796429360359, "standard_deviation": 0.03622429614350138, "minimum": 1.3648661630777075, "maximum": 1.4691374518298614, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.666424815822895, "standard_deviation": 0.2939507135572532, "minimum": 0.3335010597365132, "maximum": 1.0965376338480708, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11373910950105172, "standard_deviation": 0.01444545193840876, "minimum": 0.09101988036062109, "maximum": 0.13692134878132595, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.973557997824323, "standard_deviation": 0.004971881781421137, "minimum": 0.9659540216960241, "maximum": 0.9826142153750389, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9650289204765765, "standard_deviation": 0.056199896529152304, "minimum": 1.9033183301667733, "maximum": 2.0636058104515422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8536354539993269, "standard_deviation": 0.2886423357967942, "minimum": 0.4603891486144507, "maximum": 1.2756317108845476, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19898756317150962, "standard_deviation": 0.01917151933403358, "minimum": 0.16341411245441465, "maximum": 0.22792240556934015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9925850045512521, "standard_deviation": 0.0021013720065664447, "minimum": 0.9894892832263851, "maximum": 0.994892867119806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1665947146329427, "standard_deviation": 0.040560521501678895, "minimum": 1.1254868747324756, "maximum": 1.2211126639168253, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8266766037541965, "standard_deviation": 0.3555033092172748, "minimum": 0.44282354822746317, "maximum": 1.490635353138848, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10506876908018353, "standard_deviation": 0.014239968664669037, "minimum": 0.08629986144384698, "maximum": 0.126298583411218, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3050072554749179, "standard_deviation": 0.10752153263643638, "minimum": 0.1154810325886048, "maximum": 0.43815543927726447, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.002899309313085, "standard_deviation": 0.010297564712668695, "minimum": 5.987128983242935, "maximum": 6.018664501682477, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7695447248454803, "standard_deviation": 0.06233697903211306, "minimum": 0.6343188338802748, "maximum": 0.8359716481406215, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04518816091840393, "standard_deviation": 0.16445231268146676, "minimum": -0.22242342688020794, "maximum": 0.20909876937142208, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9898875645141099, "standard_deviation": 0.0021681051790105087, "minimum": 0.9870104741616526, "maximum": 0.9928806289966425, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0621570401890499, "standard_deviation": 0.035483550199965165, "minimum": 1.0182523340277136, "maximum": 1.1200715843152305, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7674507345204766, "standard_deviation": 0.2760200722063598, "minimum": 0.39522469801927923, "maximum": 1.1219369747385872, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11218694626179407, "standard_deviation": 0.012726468020979412, "minimum": 0.09489090340116711, "maximum": 0.13154099056819907, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974585478710136, "standard_deviation": 0.000522902972988243, "minimum": 0.9968366712384882, "maximum": 0.9981485497068658, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5567894515580202, "standard_deviation": 0.03873550473076157, "minimum": 0.49821071943113476, "maximum": 0.6268522766754654, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.865076664082687, "standard_deviation": 0.3427158872238159, "minimum": 0.45586479329951385, "maximum": 1.3724802341447384, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05798885308881112, "standard_deviation": 0.0059392418225138, "minimum": 0.05253527113348546, "maximum": 0.06896195465927978, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0010", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.798771858215332, "predicted_mos": 2.798771858215332, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "d0e2f4ed9881f4d4c4adb11a4a157f3c444e38cf1c18217ee61c8ecb2b06668e", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.669159899931401, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.06664939012384, "全帯域_10ms_rms_envelope_nrmse": 0.13106587820860627, "全帯域_10ms_envelope_correlation": 0.9864965878854722, "全帯域_10ms_modulation_mae_db": 0.9222395216411111, "全帯域_40ms_log_rms_mae_db": 0.5385744341209846, "全帯域_40ms_rms_envelope_nrmse": 0.06777665997111931, "全帯域_40ms_envelope_correlation": 0.9964772246837253, "全帯域_40ms_modulation_mae_db": 1.3386511532423222, "300–1000 Hz_10ms_log_rms_mae_db": 1.6463121009599364, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23661203793280713, "300–1000 Hz_10ms_envelope_correlation": 0.9630387218045102, "300–1000 Hz_10ms_modulation_mae_db": 1.2488489081621468, "300–1000 Hz_40ms_log_rms_mae_db": 0.9790845474304869, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10221990516075984, "300–1000 Hz_40ms_envelope_correlation": 0.9928831373669277, "300–1000 Hz_40ms_modulation_mae_db": 1.0704186048513062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3086789880677736, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2055786013172014, "1000–3000 Hz_10ms_envelope_correlation": 0.9783927399079245, "1000–3000 Hz_10ms_modulation_mae_db": 2.8758140427014216, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8235332017431689, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11484619838902846, "1000–3000 Hz_40ms_envelope_correlation": 0.9949513006985181, "1000–3000 Hz_40ms_modulation_mae_db": 3.2767402873086104, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4778959802166984, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21044602554108974, "3000–6000 Hz_10ms_envelope_correlation": 0.9694455698804104, "3000–6000 Hz_10ms_modulation_mae_db": 1.0402292160754347, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0376974679964028, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11967072118914539, "3000–6000 Hz_40ms_envelope_correlation": 0.9899390813849602, "3000–6000 Hz_40ms_modulation_mae_db": 1.096434407452509, "mr_stft_db": 5.708289731041774, "band_mae_db": 0.41500117381726503, "rms_gain_error_db": 0.10251517847232869, "output_peak": 0.9020851596256374}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0917484176696342, "全帯域_10ms_rms_envelope_nrmse": 0.09877865030016574, "全帯域_10ms_envelope_correlation": 0.9925539181656069, "全帯域_10ms_modulation_mae_db": 0.6628571090550405, "全帯域_40ms_log_rms_mae_db": 0.4819935841622385, "全帯域_40ms_rms_envelope_nrmse": 0.05555874899831156, "全帯域_40ms_envelope_correlation": 0.998184673968549, "全帯域_40ms_modulation_mae_db": 0.5060541594455713, "300–1000 Hz_10ms_log_rms_mae_db": 1.70161783800566, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1839202154278166, "300–1000 Hz_10ms_envelope_correlation": 0.9783383413320293, "300–1000 Hz_10ms_modulation_mae_db": 1.0440114276938857, "300–1000 Hz_40ms_log_rms_mae_db": 0.9846549185520642, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11813590500578944, "300–1000 Hz_40ms_envelope_correlation": 0.9919737880944189, "300–1000 Hz_40ms_modulation_mae_db": 1.0378628169143413, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3018627288648643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13638341255936023, "1000–3000 Hz_10ms_envelope_correlation": 0.9894967080407263, "1000–3000 Hz_10ms_modulation_mae_db": 0.5271964779152393, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8163096528201155, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08046800334847812, "1000–3000 Hz_40ms_envelope_correlation": 0.9972861691250636, "1000–3000 Hz_40ms_modulation_mae_db": 0.9808921128002449, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3835338850187702, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16960220329902312, "3000–6000 Hz_10ms_envelope_correlation": 0.9807993570419375, "3000–6000 Hz_10ms_modulation_mae_db": 0.6808132220165831, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9489978230364391, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08072492016631515, "3000–6000 Hz_40ms_envelope_correlation": 0.9955485650306, "3000–6000 Hz_40ms_modulation_mae_db": 0.7520056657317089, "mr_stft_db": 5.7144910521477525, "band_mae_db": 0.2259989379923454, "rms_gain_error_db": -0.27455190087798387, "output_peak": 0.7090304610373271}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1811569802486206, "全帯域_10ms_rms_envelope_nrmse": 0.11447736471269936, "全帯域_10ms_envelope_correlation": 0.9905507976637182, "全帯域_10ms_modulation_mae_db": 0.824249550756413, "全帯域_40ms_log_rms_mae_db": 0.6094027330497364, "全帯域_40ms_rms_envelope_nrmse": 0.07858115314838747, "全帯域_40ms_envelope_correlation": 0.9964143852669934, "全帯域_40ms_modulation_mae_db": 1.075042427264873, "300–1000 Hz_10ms_log_rms_mae_db": 1.7586026015488248, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2189519850781924, "300–1000 Hz_10ms_envelope_correlation": 0.9687454126201622, "300–1000 Hz_10ms_modulation_mae_db": 1.8666753666060256, "300–1000 Hz_40ms_log_rms_mae_db": 1.0273415867841844, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12586577653066627, "300–1000 Hz_40ms_envelope_correlation": 0.9909479669947747, "300–1000 Hz_40ms_modulation_mae_db": 1.690735903748146, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2465842845333903, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15198964258614603, "1000–3000 Hz_10ms_envelope_correlation": 0.9884004589824597, "1000–3000 Hz_10ms_modulation_mae_db": 1.676560505439112, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8077454871139055, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09075953297218683, "1000–3000 Hz_40ms_envelope_correlation": 0.998262915696455, "1000–3000 Hz_40ms_modulation_mae_db": 1.7305884957036568, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4355394364422882, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1847309849232985, "3000–6000 Hz_10ms_envelope_correlation": 0.9767237234393888, "3000–6000 Hz_10ms_modulation_mae_db": 0.8773602619195687, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0329822973480187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10245892160091301, "3000–6000 Hz_40ms_envelope_correlation": 0.993234363011092, "3000–6000 Hz_40ms_modulation_mae_db": 1.081839729235644, "mr_stft_db": 5.682638446488048, "band_mae_db": 0.45029663177412765, "rms_gain_error_db": -0.41431407564775286, "output_peak": 0.7711851304237586}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1288111967753518, "全帯域_10ms_rms_envelope_nrmse": 0.11333493818998862, "全帯域_10ms_envelope_correlation": 0.9906633868448748, "全帯域_10ms_modulation_mae_db": 0.7196794029975436, "全帯域_40ms_log_rms_mae_db": 0.5948757727668647, "全帯域_40ms_rms_envelope_nrmse": 0.07453759677500381, "全帯域_40ms_envelope_correlation": 0.9969334806678872, "全帯域_40ms_modulation_mae_db": 1.0981976780688336, "300–1000 Hz_10ms_log_rms_mae_db": 1.6788260725402966, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18718297663683642, "300–1000 Hz_10ms_envelope_correlation": 0.9778671495804517, "300–1000 Hz_10ms_modulation_mae_db": 0.9349525887570064, "300–1000 Hz_40ms_log_rms_mae_db": 1.0185596872565503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10842078633269944, "300–1000 Hz_40ms_envelope_correlation": 0.9941431406764142, "300–1000 Hz_40ms_modulation_mae_db": 1.001235663975241, "1000–3000 Hz_10ms_log_rms_mae_db": 1.30113509028881, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17943972894825735, "1000–3000 Hz_10ms_envelope_correlation": 0.9836960377513984, "1000–3000 Hz_10ms_modulation_mae_db": 1.6349589918595615, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8125298040242965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1160039358907761, "1000–3000 Hz_40ms_envelope_correlation": 0.9963808410163575, "1000–3000 Hz_40ms_modulation_mae_db": 1.191383276716001, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4197552301006189, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18967417175944262, "3000–6000 Hz_10ms_envelope_correlation": 0.9752858459536201, "3000–6000 Hz_10ms_modulation_mae_db": 0.5076040222050686, "3000–6000 Hz_40ms_log_rms_mae_db": 0.991032473180601, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10163449649734768, "3000–6000 Hz_40ms_envelope_correlation": 0.9930012754422305, "3000–6000 Hz_40ms_modulation_mae_db": 0.7860289667627046, "mr_stft_db": 5.701471247292287, "band_mae_db": 0.5428709103027131, "rms_gain_error_db": -0.3824690283998495, "output_peak": 0.6961028570452451}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0949869798848622, "全帯域_10ms_rms_envelope_nrmse": 0.10491357962480441, "全帯域_10ms_envelope_correlation": 0.991015217508798, "全帯域_10ms_modulation_mae_db": 0.2800863401908274, "全帯域_40ms_log_rms_mae_db": 0.6183823245383012, "全帯域_40ms_rms_envelope_nrmse": 0.05054920392433554, "全帯域_40ms_envelope_correlation": 0.9980352130684744, "全帯域_40ms_modulation_mae_db": 0.5099527238571433, "300–1000 Hz_10ms_log_rms_mae_db": 1.7478871832795018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19794565882309736, "300–1000 Hz_10ms_envelope_correlation": 0.9738426798166456, "300–1000 Hz_10ms_modulation_mae_db": 0.593255507768974, "300–1000 Hz_40ms_log_rms_mae_db": 0.9968239124069912, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0942057545404905, "300–1000 Hz_40ms_envelope_correlation": 0.9940801864820109, "300–1000 Hz_40ms_modulation_mae_db": 0.5081730814683917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2468808796414683, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15039776750911207, "1000–3000 Hz_10ms_envelope_correlation": 0.9860626657579519, "1000–3000 Hz_10ms_modulation_mae_db": 1.6130740112504982, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7807601361217309, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07325620405445397, "1000–3000 Hz_40ms_envelope_correlation": 0.9968211158531035, "1000–3000 Hz_40ms_modulation_mae_db": 1.2433867772312321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4214011133813278, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20373818651870068, "3000–6000 Hz_10ms_envelope_correlation": 0.9725759817413473, "3000–6000 Hz_10ms_modulation_mae_db": 0.8300667296440882, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9719166477733386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10042359855942366, "3000–6000 Hz_40ms_envelope_correlation": 0.993206129832464, "3000–6000 Hz_40ms_modulation_mae_db": 1.0397259652589632, "mr_stft_db": 5.685033423142556, "band_mae_db": 0.2087534225407075, "rms_gain_error_db": -0.14785853700179108, "output_peak": 0.7909732261122513}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0839556619222084, "全帯域_10ms_rms_envelope_nrmse": 0.13242459761429304, "全帯域_10ms_envelope_correlation": 0.9855915919901159, "全帯域_10ms_modulation_mae_db": 1.2390667569536518, "全帯域_40ms_log_rms_mae_db": 0.5712136444928883, "全帯域_40ms_rms_envelope_nrmse": 0.045653481817008505, "全帯域_40ms_envelope_correlation": 0.9984815324809815, "全帯域_40ms_modulation_mae_db": 1.5542179490297878, "300–1000 Hz_10ms_log_rms_mae_db": 1.7175105363259555, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18500600976831622, "300–1000 Hz_10ms_envelope_correlation": 0.9780123177286908, "300–1000 Hz_10ms_modulation_mae_db": 1.5066741726643207, "300–1000 Hz_40ms_log_rms_mae_db": 1.0883515059811244, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08493941884840371, "300–1000 Hz_40ms_envelope_correlation": 0.996797287490533, "300–1000 Hz_40ms_modulation_mae_db": 1.2163888902235556, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3084099079999592, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21724328936217266, "1000–3000 Hz_10ms_envelope_correlation": 0.9709141127086026, "1000–3000 Hz_10ms_modulation_mae_db": 2.7893811599664446, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8082652261362387, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09809527883467883, "1000–3000 Hz_40ms_envelope_correlation": 0.9940646806735993, "1000–3000 Hz_40ms_modulation_mae_db": 3.3992862303659503, "3000–6000 Hz_10ms_log_rms_mae_db": 1.435790817850728, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19698103339189832, "3000–6000 Hz_10ms_envelope_correlation": 0.9733118951097666, "3000–6000 Hz_10ms_modulation_mae_db": 0.7979444448463869, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9571995657035978, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10113139034443884, "3000–6000 Hz_40ms_envelope_correlation": 0.9926966766576429, "3000–6000 Hz_40ms_modulation_mae_db": 0.779650574438651, "mr_stft_db": 5.710077938974753, "band_mae_db": 0.33223970910616096, "rms_gain_error_db": -0.1621423043616383, "output_peak": 0.8725923695659267}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.192082751876634, "全帯域_10ms_rms_envelope_nrmse": 0.09706514648081184, "全帯域_10ms_envelope_correlation": 0.9924692442334896, "全帯域_10ms_modulation_mae_db": 0.3861482678991927, "全帯域_40ms_log_rms_mae_db": 0.6120191191599657, "全帯域_40ms_rms_envelope_nrmse": 0.053276502857666976, "全帯域_40ms_envelope_correlation": 0.997938713467304, "全帯域_40ms_modulation_mae_db": 0.39539643163611354, "300–1000 Hz_10ms_log_rms_mae_db": 1.5958261599115393, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14194887451687094, "300–1000 Hz_10ms_envelope_correlation": 0.9869749647354308, "300–1000 Hz_10ms_modulation_mae_db": 1.0329081196592689, "300–1000 Hz_40ms_log_rms_mae_db": 0.9106974700222424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07946133007739221, "300–1000 Hz_40ms_envelope_correlation": 0.9963051092476433, "300–1000 Hz_40ms_modulation_mae_db": 0.7495669904797931, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2764805404204282, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17726417558725757, "1000–3000 Hz_10ms_envelope_correlation": 0.980958945568275, "1000–3000 Hz_10ms_modulation_mae_db": 1.4421191076138413, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8289784548829173, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08782609677364503, "1000–3000 Hz_40ms_envelope_correlation": 0.9960431826284828, "1000–3000 Hz_40ms_modulation_mae_db": 1.225830575802618, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3784962703059187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1976757114246836, "3000–6000 Hz_10ms_envelope_correlation": 0.9743158836716757, "3000–6000 Hz_10ms_modulation_mae_db": 1.1654704133102316, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9440567164290748, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10253185063457178, "3000–6000 Hz_40ms_envelope_correlation": 0.9929556201988126, "3000–6000 Hz_40ms_modulation_mae_db": 1.1949205000782646, "mr_stft_db": 5.698558649350955, "band_mae_db": 0.33863148109484875, "rms_gain_error_db": -0.19210347592025198, "output_peak": 0.7169506115394136}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1411788438937287, "全帯域_10ms_rms_envelope_nrmse": 0.12819091187987217, "全帯域_10ms_envelope_correlation": 0.9874252296844077, "全帯域_10ms_modulation_mae_db": 1.2554116141481866, "全帯域_40ms_log_rms_mae_db": 0.6008743351892832, "全帯域_40ms_rms_envelope_nrmse": 0.08528982037661237, "全帯域_40ms_envelope_correlation": 0.9945010331460253, "全帯域_40ms_modulation_mae_db": 1.795050260058065, "300–1000 Hz_10ms_log_rms_mae_db": 1.8035535366582571, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.24291004281244544, "300–1000 Hz_10ms_envelope_correlation": 0.9640104224645226, "300–1000 Hz_10ms_modulation_mae_db": 1.5915598887236675, "300–1000 Hz_40ms_log_rms_mae_db": 1.057765881413724, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.16410983981716926, "300–1000 Hz_40ms_envelope_correlation": 0.983600697438732, "300–1000 Hz_40ms_modulation_mae_db": 1.9001496632758395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2847405269831393, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18208832322560567, "1000–3000 Hz_10ms_envelope_correlation": 0.9797707255011013, "1000–3000 Hz_10ms_modulation_mae_db": 2.0323912700549314, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7652825678427515, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07849513360936534, "1000–3000 Hz_40ms_envelope_correlation": 0.9961946318847229, "1000–3000 Hz_40ms_modulation_mae_db": 2.434509933310482, "3000–6000 Hz_10ms_log_rms_mae_db": 1.482819348551865, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1874649702128858, "3000–6000 Hz_10ms_envelope_correlation": 0.9757968367750331, "3000–6000 Hz_10ms_modulation_mae_db": 0.612383165723763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.061978525869166, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10533698011066235, "3000–6000 Hz_40ms_envelope_correlation": 0.9921843415614924, "3000–6000 Hz_40ms_modulation_mae_db": 0.400273639657781, "mr_stft_db": 5.68963502228146, "band_mae_db": 0.16867464310016153, "rms_gain_error_db": 0.1664476537718462, "output_peak": 0.775185056507921}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49600, "output_samples": 49600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7219149869266, "spectral_convergence": 0.5065853408770649, "spectral_convergence_status": "ok", "amplitude_floor": 1.4924605057201793e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.476368789924308, "spectral_convergence": 0.48876983503182075, "spectral_convergence_status": "ok", "amplitude_floor": 8.127484488587521e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.926585416274418, "spectral_convergence": 0.5377844294704234, "spectral_convergence_status": "ok", "amplitude_floor": 5.495951303547364e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185024372796483, "output_power": 0.0007883122353586796, "delta_db": -0.16321709681520483, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004929662452993321, "output_power": 0.0005817386113669353, "delta_db": 0.7191070690129988, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001304221454098428, "output_power": 0.0012035655885092361, "delta_db": -0.3488157775291723, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007846801909808332, "output_power": 0.0009478775841309801, "delta_db": 0.8205956387030666, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003122815376510157, "output_power": 0.0002823548241302785, "delta_db": -0.43751096923162697, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001061689290478197, "output_power": 0.00010618752188124272, "delta_db": 0.0007604916115208861, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.41500117381726503, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.10251517847232869, "pitch": null, "output_rms_dbfs": -24.078192034201756, "output_peak": 0.9020851596256374, "reference_rms_dbfs": -24.180707212674086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.708289731041774}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.0002690918479937382, "log_rms_mae_db": 1.06664939012384, "rms_envelope_nrmse": 0.13106587820860627, "envelope_correlation": 0.9864965878854722, "rms_gain_error_db": 0.10251565980751559, "reference_mean_power": 0.0038249880205223005, "output_mean_power": 0.0039163513346982546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 2.1583633605855604e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008154057878800693, "output_power": 0.0008699550987542725, "reference_active": true, "delta_db": 0.2812304801112344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.087495849252009e-05, "output_power": 0.00010010601809259928, "reference_active": true, "delta_db": 0.42015961326911955}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.599429116228097e-05, "output_power": 4.182258055350045e-05, "reference_active": true, "delta_db": 2.0653284715429794}], "active_band_count": 3, "active_band_mae_db": 0.9222395216411111}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.0002456731948415248, "log_rms_mae_db": 0.5385744341209846, "rms_envelope_nrmse": 0.06777665997111931, "envelope_correlation": 0.9964772246837253, "rms_gain_error_db": 0.1025160275697319, "reference_mean_power": 0.0038624153279842384, "output_mean_power": 0.0039546729622878}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 2.0216648233438394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007562450561775054, "output_power": 0.0008192861283179222, "reference_active": true, "delta_db": 0.34773053152928973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1399930780791825e-05, "output_power": 5.0120153281520073e-05, "reference_active": true, "delta_db": 0.8301277552215468}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0921358976760216e-06, "output_power": 2.0993565264122452e-06, "reference_active": true, "delta_db": 2.83809517297613}], "active_band_count": 3, "active_band_mae_db": 1.3386511532423222}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00017000794064843354, "log_rms_mae_db": 1.6463121009599364, "rms_envelope_nrmse": 0.23661203793280713, "envelope_correlation": 0.9630387218045102, "rms_gain_error_db": -0.04915248553987075, "reference_mean_power": 0.001461911046007658, "output_mean_power": 0.0014454587386605437}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.2056298552980857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006038941034712422, "output_power": 0.0005640440571117722, "reference_active": true, "delta_db": -0.29647761335242706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.14042207191201e-05, "output_power": 6.757890953484458e-05, "reference_active": true, "delta_db": -0.23912704133464202}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.020128231954835e-05, "output_power": 4.231293443839093e-05, "reference_active": true, "delta_db": 3.210942069799371}], "active_band_count": 3, "active_band_mae_db": 1.2488489081621468}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00015858554889566586, "log_rms_mae_db": 0.9790845474304869, "rms_envelope_nrmse": 0.10221990516075984, "envelope_correlation": 0.9928831373669277, "rms_gain_error_db": -0.049152405515942334, "reference_mean_power": 0.0014762175799508475, "output_mean_power": 0.00145960429415287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.148107626412446e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000571116387197927, "output_power": 0.0005541202886060042, "reference_active": true, "delta_db": -0.1312056992827616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.637477969141665e-05, "output_power": 3.581554515267584e-05, "reference_active": true, "delta_db": -0.06728805647971434}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.172790453481755e-06, "output_power": 2.346911040627725e-06, "reference_active": true, "delta_db": 3.012762058791443}], "active_band_count": 3, "active_band_mae_db": 1.0704186048513062}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00019136572750236326, "log_rms_mae_db": 1.3086789880677736, "rms_envelope_nrmse": 0.2055786013172014, "envelope_correlation": 0.9783927399079245, "rms_gain_error_db": 0.5380259829415778, "reference_mean_power": 0.0009651549113833473, "output_mean_power": 0.0010924451056719766}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 7.99902404373172e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003359743599084894, "output_power": 0.0004006757603564672, "reference_active": true, "delta_db": 0.7648693455335023}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9966531082920368e-05, "output_power": 4.9932917311640366e-05, "reference_active": true, "delta_db": 2.21750468235671}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.661292025781732e-06, "output_power": 2.4437978772107905e-05, "reference_active": true, "delta_db": 5.645068100214053}], "active_band_count": 3, "active_band_mae_db": 2.8758140427014216}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00016758862552795117, "log_rms_mae_db": 0.8235332017431689, "rms_envelope_nrmse": 0.11484619838902846, "envelope_correlation": 0.9949513006985181, "rms_gain_error_db": 0.5380262835541991, "reference_mean_power": 0.0009746009072787481, "output_mean_power": 0.0011031369703243146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 7.731649944214162e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031280752850392277, "output_power": 0.000384219767872326, "reference_active": true, "delta_db": 0.8930250837810421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5385931815426458e-05, "output_power": 2.7738175255034414e-05, "reference_active": true, "delta_db": 2.5595408430630506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1192579703922553e-07, "output_power": 1.354617827896931e-06, "reference_active": true, "delta_db": 6.377654935081737}], "active_band_count": 3, "active_band_mae_db": 3.2767402873086104}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 5.61189760356505e-05, "log_rms_mae_db": 1.4778959802166984, "rms_envelope_nrmse": 0.21044602554108974, "envelope_correlation": 0.9694455698804104, "rms_gain_error_db": -0.21156458099492936, "reference_mean_power": 0.00014982208037105655, "output_mean_power": 0.00014269848591871257}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.5994961899965245e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.519717012759751e-05, "output_power": 5.106792334634892e-05, "reference_active": true, "delta_db": -0.33768614873579095}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5683811710113524e-05, "output_power": 1.864400720434719e-05, "reference_active": true, "delta_db": 0.7508764228339229}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1546072478197644e-06, "output_power": 5.036835930990241e-06, "reference_active": true, "delta_db": 2.0321250766565906}], "active_band_count": 3, "active_band_mae_db": 1.0402292160754347}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 4.51769085602287e-05, "log_rms_mae_db": 1.0376974679964028, "rms_envelope_nrmse": 0.11967072118914539, "envelope_correlation": 0.9899390813849602, "rms_gain_error_db": -0.21156507945865946, "reference_mean_power": 0.0001512883740521334, "output_mean_power": 0.00014409504515643356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.4928738695410609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.9695472955753735e-05, "output_power": 4.64554552159487e-05, "reference_active": true, "delta_db": -0.29280108097239343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.547268075821193e-06, "output_power": 1.0492771711136815e-05, "reference_active": true, "delta_db": 0.8906289876883227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7628269396099532e-07, "output_power": 2.8628394565676183e-07, "reference_active": true, "delta_db": 2.1058731536968107}], "active_band_count": 3, "active_band_mae_db": 1.096434407452509}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8514792900052502, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6813227491cf5eb81d428fba436bb4bf7d7e3e57c760ff55453f443b5a52f218", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9822115492773049, "standard_deviation": 0.006072807782071493, "minimum": 0.9709141127086026, "maximum": 0.9894967080407263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2843466183499792, "standard_deviation": 0.02577783882309929, "minimum": 1.2465842845333903, "maximum": 1.3086789880677736, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.8239369458501313, "standard_deviation": 0.7569862732268131, "minimum": 0.5271964779152393, "maximum": 2.8758140427014216, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17504811763688913, "standard_deviation": 0.02782752250144998, "minimum": 0.13638341255936023, "maximum": 0.21724328936217266, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962506046970379, "standard_deviation": 0.0013080435394724867, "minimum": 0.9940646806735993, "maximum": 0.998262915696455, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8054255663356407, "standard_deviation": 0.021662502251417786, "minimum": 0.7652825678427515, "maximum": 0.8289784548829173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.9353272111548496, "standard_deviation": 0.9768924752019389, "minimum": 0.9808921128002449, "maximum": 3.3992862303659503, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09246879798407659, "standard_deviation": 0.016120928929237853, "minimum": 0.07325620405445397, "maximum": 0.1160039358907761, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9747818867016474, "standard_deviation": 0.003316304005667953, "minimum": 0.9694455698804104, "maximum": 0.9807993570419375, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4294040102335268, "standard_deviation": 0.03805243745646679, "minimum": 1.3784962703059187, "maximum": 1.482819348551865, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8139839344676406, "standard_deviation": 0.21761962618920694, "minimum": 0.5076040222050686, "maximum": 1.1654704133102316, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1925391608838778, "standard_deviation": 0.012610741275291541, "minimum": 0.16960220329902312, "maximum": 0.21044602554108974, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9928457566399118, "standard_deviation": 0.0015367297759796433, "minimum": 0.9899390813849602, "maximum": 0.9955485650306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9932326896670798, "standard_deviation": 0.045380515335320455, "minimum": 0.9440567164290748, "maximum": 1.061978525869166, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8913599310770284, "standard_deviation": 0.26094841931808316, "minimum": 0.400273639657781, "maximum": 1.1949205000782646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10173910988785223, "standard_deviation": 0.010551512603193883, "minimum": 0.08072492016631515, "maximum": 0.11967072118914539, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9738537512603055, "standard_deviation": 0.008155892370114102, "minimum": 0.9630387218045102, "maximum": 0.9869749647354308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7062670036537464, "standard_deviation": 0.06614781116691351, "minimum": 1.5958261599115393, "maximum": 1.8035535366582571, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.227360747504412, "standard_deviation": 0.4103766059709661, "minimum": 0.593255507768974, "maximum": 1.8666753666060256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1993097251245478, "standard_deviation": 0.03286093596899873, "minimum": 0.14194887451687094, "maximum": 0.24291004281244544, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9925914142239318, "standard_deviation": 0.004142215935916477, "minimum": 0.983600697438732, "maximum": 0.996797287490533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0079099387309212, "standard_deviation": 0.05397397527832654, "minimum": 0.9106974700222424, "maximum": 1.0883515059811244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.146816451867077, "standard_deviation": 0.45868965578246507, "minimum": 0.5081730814683917, "maximum": 1.9001496632758395, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10966983953917132, "standard_deviation": 0.027039768601694693, "minimum": 0.07946133007739221, "maximum": 0.16410983981716926, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3353083637160412, "standard_deviation": 0.1300590068297726, "minimum": 0.16867464310016153, "maximum": 0.5428709103027131, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.698774438839948, "standard_deviation": 0.011984250810298594, "minimum": 5.682638446488048, "maximum": 5.7144910521477525, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7792631089821851, "standard_deviation": 0.07535223740659683, "minimum": 0.6961028570452451, "maximum": 0.9020851596256374, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1630595612456366, "standard_deviation": 0.20835419557082382, "minimum": -0.41431407564775286, "maximum": 0.1664476537718462, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9895957467470604, "standard_deviation": 0.0027103986556410607, "minimum": 0.9855915919901159, "maximum": 0.9925539181656069, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1225712777993602, "standard_deviation": 0.046258085275409914, "minimum": 1.06664939012384, "maximum": 1.192082751876634, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7862173204552458, "standard_deviation": 0.35477019510440055, "minimum": 0.2800863401908274, "maximum": 1.2554116141481866, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11503138337640517, "standard_deviation": 0.014269317396952167, "minimum": 0.09706514648081184, "maximum": 0.13242459761429304, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971207820937426, "standard_deviation": 0.0013255743912453052, "minimum": 0.9945010331460253, "maximum": 0.9984815324809815, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5784169934350328, "standard_deviation": 0.04690181513868544, "minimum": 0.4819935841622385, "maximum": 0.6183823245383012, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0340703478253388, "standard_deviation": 0.5223038520183346, "minimum": 0.39539643163611354, "maximum": 1.795050260058065, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0639028959835557, "standard_deviation": 0.014613408253861425, "minimum": 0.045653481817008505, "maximum": 0.08528982037661237, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0010", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.5688564777374268, "predicted_mos": 2.5688564777374268, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "c34e6afae94eaded96537deff3e2734269ccb3678344c9564770a1e776349414", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.792829900048673, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0389921316662662, "全帯域_10ms_rms_envelope_nrmse": 0.0911518353487137, "全帯域_10ms_envelope_correlation": 0.9932176066400141, "全帯域_10ms_modulation_mae_db": 0.23444472843569794, "全帯域_40ms_log_rms_mae_db": 0.5292015620534017, "全帯域_40ms_rms_envelope_nrmse": 0.05377702060306877, "全帯域_40ms_envelope_correlation": 0.9976385901223987, "全帯域_40ms_modulation_mae_db": 0.401767336845817, "300–1000 Hz_10ms_log_rms_mae_db": 1.78261957150753, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18721683779204304, "300–1000 Hz_10ms_envelope_correlation": 0.9767621693730923, "300–1000 Hz_10ms_modulation_mae_db": 0.9449639983085083, "300–1000 Hz_40ms_log_rms_mae_db": 1.0510537968912101, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09507517105066385, "300–1000 Hz_40ms_envelope_correlation": 0.9943321659279132, "300–1000 Hz_40ms_modulation_mae_db": 0.6399568775883693, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4852577944391085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1341175125963419, "1000–3000 Hz_10ms_envelope_correlation": 0.9893800596204558, "1000–3000 Hz_10ms_modulation_mae_db": 0.8296681667665968, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9836875423724742, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0642137137602752, "1000–3000 Hz_40ms_envelope_correlation": 0.9976924580447909, "1000–3000 Hz_40ms_modulation_mae_db": 0.7598536306587835, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6521577469845032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2142674764169496, "3000–6000 Hz_10ms_envelope_correlation": 0.9710809380364765, "3000–6000 Hz_10ms_modulation_mae_db": 0.34361197137351507, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1859526986399074, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10676681803125772, "3000–6000 Hz_40ms_envelope_correlation": 0.9933440452396317, "3000–6000 Hz_40ms_modulation_mae_db": 0.355140367970228, "mr_stft_db": 5.777068687631839, "band_mae_db": 0.30599698071133763, "rms_gain_error_db": -0.11165377592152953, "output_peak": 0.778308374912595}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.068124465431986, "全帯域_10ms_rms_envelope_nrmse": 0.08178382665717943, "全帯域_10ms_envelope_correlation": 0.9945806569697028, "全帯域_10ms_modulation_mae_db": 0.7205854888285826, "全帯域_40ms_log_rms_mae_db": 0.5651964902922058, "全帯域_40ms_rms_envelope_nrmse": 0.036134400678116754, "全帯域_40ms_envelope_correlation": 0.9990367952754078, "全帯域_40ms_modulation_mae_db": 0.9129750450679245, "300–1000 Hz_10ms_log_rms_mae_db": 1.8059087074035143, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1474492225837557, "300–1000 Hz_10ms_envelope_correlation": 0.9860197218521337, "300–1000 Hz_10ms_modulation_mae_db": 1.1457848733636824, "300–1000 Hz_40ms_log_rms_mae_db": 1.1355950853725392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06957195697806316, "300–1000 Hz_40ms_envelope_correlation": 0.997558677314363, "300–1000 Hz_40ms_modulation_mae_db": 1.2576419981735503, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4435202710626145, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12894148194568822, "1000–3000 Hz_10ms_envelope_correlation": 0.9897999728606385, "1000–3000 Hz_10ms_modulation_mae_db": 0.3531951753715741, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9954418567872965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0591696089956101, "1000–3000 Hz_40ms_envelope_correlation": 0.9979927385714124, "1000–3000 Hz_40ms_modulation_mae_db": 0.3471114868028457, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6062192525747985, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18847854503671474, "3000–6000 Hz_10ms_envelope_correlation": 0.9759127933759215, "3000–6000 Hz_10ms_modulation_mae_db": 0.4533628365983308, "3000–6000 Hz_40ms_log_rms_mae_db": 1.132127715564964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09879241052820185, "3000–6000 Hz_40ms_envelope_correlation": 0.9930845705504774, "3000–6000 Hz_40ms_modulation_mae_db": 0.6537721643150425, "mr_stft_db": 5.801267639687556, "band_mae_db": 0.3699935289405083, "rms_gain_error_db": -0.1222660420627354, "output_peak": 0.7670765337701684}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0904673157276785, "全帯域_10ms_rms_envelope_nrmse": 0.12563153034606866, "全帯域_10ms_envelope_correlation": 0.9870752074148128, "全帯域_10ms_modulation_mae_db": 0.6751770404034496, "全帯域_40ms_log_rms_mae_db": 0.603249625953429, "全帯域_40ms_rms_envelope_nrmse": 0.06561555752063442, "全帯域_40ms_envelope_correlation": 0.996389156874599, "全帯域_40ms_modulation_mae_db": 1.1031275493249961, "300–1000 Hz_10ms_log_rms_mae_db": 1.8194247374817605, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2040036850562569, "300–1000 Hz_10ms_envelope_correlation": 0.9721993947597476, "300–1000 Hz_10ms_modulation_mae_db": 1.3861661324741847, "300–1000 Hz_40ms_log_rms_mae_db": 1.0240537952450255, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09319557763426464, "300–1000 Hz_40ms_envelope_correlation": 0.9942066425425153, "300–1000 Hz_40ms_modulation_mae_db": 1.5343825136057694, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4484061218630544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19846091724695952, "1000–3000 Hz_10ms_envelope_correlation": 0.9759864711605984, "1000–3000 Hz_10ms_modulation_mae_db": 1.7948797742649554, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9320180973999566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10358716843044558, "1000–3000 Hz_40ms_envelope_correlation": 0.9933646712104434, "1000–3000 Hz_40ms_modulation_mae_db": 2.0385688205163888, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6253335114332585, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20676786341581305, "3000–6000 Hz_10ms_envelope_correlation": 0.9716134597634789, "3000–6000 Hz_10ms_modulation_mae_db": 0.3961350628251646, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1463746020475787, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11731802372958339, "3000–6000 Hz_40ms_envelope_correlation": 0.9906038597822706, "3000–6000 Hz_40ms_modulation_mae_db": 0.4162235245466818, "mr_stft_db": 5.795078764228691, "band_mae_db": 0.16504083203562478, "rms_gain_error_db": -0.04142526150454518, "output_peak": 0.6718593671138687}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0495916720287224, "全帯域_10ms_rms_envelope_nrmse": 0.10492845716031811, "全帯域_10ms_envelope_correlation": 0.9909958407441396, "全帯域_10ms_modulation_mae_db": 0.9102314489421923, "全帯域_40ms_log_rms_mae_db": 0.5662926939673922, "全帯域_40ms_rms_envelope_nrmse": 0.051110710487834284, "全帯域_40ms_envelope_correlation": 0.9977964723551123, "全帯域_40ms_modulation_mae_db": 1.290548260833922, "300–1000 Hz_10ms_log_rms_mae_db": 1.8373405751664011, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17838228295190986, "300–1000 Hz_10ms_envelope_correlation": 0.9798611269976872, "300–1000 Hz_10ms_modulation_mae_db": 1.4252747366962446, "300–1000 Hz_40ms_log_rms_mae_db": 1.148910198340057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09640675462782491, "300–1000 Hz_40ms_envelope_correlation": 0.9941344023594925, "300–1000 Hz_40ms_modulation_mae_db": 1.4574310968682596, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4480544290257735, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20973900047787936, "1000–3000 Hz_10ms_envelope_correlation": 0.9727853950234573, "1000–3000 Hz_10ms_modulation_mae_db": 1.1255661393213863, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9230349265364899, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10034037158436308, "1000–3000 Hz_40ms_envelope_correlation": 0.9943617064134648, "1000–3000 Hz_40ms_modulation_mae_db": 1.1371148578273116, "3000–6000 Hz_10ms_log_rms_mae_db": 1.737344131674539, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2385717953426653, "3000–6000 Hz_10ms_envelope_correlation": 0.9616456158750317, "3000–6000 Hz_10ms_modulation_mae_db": 0.9190845225001242, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2029581560362363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1205855567932181, "3000–6000 Hz_40ms_envelope_correlation": 0.9897733127311403, "3000–6000 Hz_40ms_modulation_mae_db": 0.8413640100600106, "mr_stft_db": 5.788533260834523, "band_mae_db": 0.25062911060321247, "rms_gain_error_db": -0.031727708966625266, "output_peak": 0.7859903563703798}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.099615260298822, "全帯域_10ms_rms_envelope_nrmse": 0.1366603405450498, "全帯域_10ms_envelope_correlation": 0.9860990681868436, "全帯域_10ms_modulation_mae_db": 1.106592693676542, "全帯域_40ms_log_rms_mae_db": 0.5741485559936393, "全帯域_40ms_rms_envelope_nrmse": 0.08276133866282413, "全帯域_40ms_envelope_correlation": 0.9963564228742221, "全帯域_40ms_modulation_mae_db": 1.4410269224240098, "300–1000 Hz_10ms_log_rms_mae_db": 1.8953822684793433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2425382201177979, "300–1000 Hz_10ms_envelope_correlation": 0.9624584677092101, "300–1000 Hz_10ms_modulation_mae_db": 1.8003239093135086, "300–1000 Hz_40ms_log_rms_mae_db": 1.072989556524007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1430299501339824, "300–1000 Hz_40ms_envelope_correlation": 0.9900097333023617, "300–1000 Hz_40ms_modulation_mae_db": 1.8233673139895121, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4127205914637055, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17899357080722902, "1000–3000 Hz_10ms_envelope_correlation": 0.9824937757876759, "1000–3000 Hz_10ms_modulation_mae_db": 1.3131979486746723, "1000–3000 Hz_40ms_log_rms_mae_db": 0.955492278917321, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11288512213861311, "1000–3000 Hz_40ms_envelope_correlation": 0.9953041051744416, "1000–3000 Hz_40ms_modulation_mae_db": 1.2754771318864586, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7177940189923075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22887234037594886, "3000–6000 Hz_10ms_envelope_correlation": 0.9641336386195976, "3000–6000 Hz_10ms_modulation_mae_db": 0.8910945092772294, "3000–6000 Hz_40ms_log_rms_mae_db": 1.153255705487329, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09668323097471777, "3000–6000 Hz_40ms_envelope_correlation": 0.9933073665452552, "3000–6000 Hz_40ms_modulation_mae_db": 0.727367755072597, "mr_stft_db": 5.785740095134854, "band_mae_db": 0.4378730600214038, "rms_gain_error_db": -0.4558381669688472, "output_peak": 0.811408890827318}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0293102672042675, "全帯域_10ms_rms_envelope_nrmse": 0.12075862787590062, "全帯域_10ms_envelope_correlation": 0.9885263375230599, "全帯域_10ms_modulation_mae_db": 0.5231872875153797, "全帯域_40ms_log_rms_mae_db": 0.546102154273636, "全帯域_40ms_rms_envelope_nrmse": 0.06440618549132324, "全帯域_40ms_envelope_correlation": 0.997364506613039, "全帯域_40ms_modulation_mae_db": 0.6985432372738988, "300–1000 Hz_10ms_log_rms_mae_db": 1.7849620916346063, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16962552557145066, "300–1000 Hz_10ms_envelope_correlation": 0.9822597384269446, "300–1000 Hz_10ms_modulation_mae_db": 0.8034969181108274, "300–1000 Hz_40ms_log_rms_mae_db": 1.1217037110653312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0965993524837276, "300–1000 Hz_40ms_envelope_correlation": 0.9958499087633654, "300–1000 Hz_40ms_modulation_mae_db": 1.4528437722614524, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4498888235737624, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21571240656676627, "1000–3000 Hz_10ms_envelope_correlation": 0.9711041408222192, "1000–3000 Hz_10ms_modulation_mae_db": 2.425714709160308, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0115641975747311, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09733179294997883, "1000–3000 Hz_40ms_envelope_correlation": 0.9946172240330288, "1000–3000 Hz_40ms_modulation_mae_db": 2.4014587569574988, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6424797748718694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20747305359959126, "3000–6000 Hz_10ms_envelope_correlation": 0.9703496929005472, "3000–6000 Hz_10ms_modulation_mae_db": 0.8900786664099595, "3000–6000 Hz_40ms_log_rms_mae_db": 1.176879534008345, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11947672915441267, "3000–6000 Hz_40ms_envelope_correlation": 0.9897877266404856, "3000–6000 Hz_40ms_modulation_mae_db": 0.9340471932629315, "mr_stft_db": 5.801723169891353, "band_mae_db": 0.3853072869279228, "rms_gain_error_db": -0.2957818653938979, "output_peak": 0.7790726124280938}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.10511577017212, "全帯域_10ms_rms_envelope_nrmse": 0.127998377785055, "全帯域_10ms_envelope_correlation": 0.9865546112136262, "全帯域_10ms_modulation_mae_db": 0.8663539155051728, "全帯域_40ms_log_rms_mae_db": 0.548775237553862, "全帯域_40ms_rms_envelope_nrmse": 0.05834950962962328, "全帯域_40ms_envelope_correlation": 0.9973469888730613, "全帯域_40ms_modulation_mae_db": 0.945810592870255, "300–1000 Hz_10ms_log_rms_mae_db": 1.8215779547392248, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18141765771395404, "300–1000 Hz_10ms_envelope_correlation": 0.9781150293676046, "300–1000 Hz_10ms_modulation_mae_db": 1.2742924783616114, "300–1000 Hz_40ms_log_rms_mae_db": 1.0295346899400453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08609315962646669, "300–1000 Hz_40ms_envelope_correlation": 0.9951547522925904, "300–1000 Hz_40ms_modulation_mae_db": 1.0703649992653868, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4395985492131023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20058676283138002, "1000–3000 Hz_10ms_envelope_correlation": 0.9753128137076378, "1000–3000 Hz_10ms_modulation_mae_db": 1.685248188089506, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0030096427097466, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10396114135684281, "1000–3000 Hz_40ms_envelope_correlation": 0.9940886262456303, "1000–3000 Hz_40ms_modulation_mae_db": 1.8447540458802643, "3000–6000 Hz_10ms_log_rms_mae_db": 1.758281203260976, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2154924396870819, "3000–6000 Hz_10ms_envelope_correlation": 0.9686131624793871, "3000–6000 Hz_10ms_modulation_mae_db": 0.7629201020297257, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2468418979576639, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12515037464682688, "3000–6000 Hz_40ms_envelope_correlation": 0.9889445764518977, "3000–6000 Hz_40ms_modulation_mae_db": 0.7750198580124995, "mr_stft_db": 5.805658131946603, "band_mae_db": 0.2922444316684503, "rms_gain_error_db": -0.1671335919826197, "output_peak": 0.854511358461779}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0595996002546713, "全帯域_10ms_rms_envelope_nrmse": 0.1180687533450585, "全帯域_10ms_envelope_correlation": 0.9888904813927337, "全帯域_10ms_modulation_mae_db": 0.5284122866726086, "全帯域_40ms_log_rms_mae_db": 0.5466664427426019, "全帯域_40ms_rms_envelope_nrmse": 0.05528273183415686, "全帯域_40ms_envelope_correlation": 0.998106015564431, "全帯域_40ms_modulation_mae_db": 0.7931829599084915, "300–1000 Hz_10ms_log_rms_mae_db": 1.7637841449756668, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16570699194139843, "300–1000 Hz_10ms_envelope_correlation": 0.9824710419895957, "300–1000 Hz_10ms_modulation_mae_db": 0.7101226694173811, "300–1000 Hz_40ms_log_rms_mae_db": 1.0250375819629938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09035334706650767, "300–1000 Hz_40ms_envelope_correlation": 0.9957622228700683, "300–1000 Hz_40ms_modulation_mae_db": 0.9679358754912272, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4357411761172745, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19910278233144824, "1000–3000 Hz_10ms_envelope_correlation": 0.9756331717926549, "1000–3000 Hz_10ms_modulation_mae_db": 1.0650234242828398, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8649505302215098, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0699522844004511, "1000–3000 Hz_40ms_envelope_correlation": 0.9969212677119657, "1000–3000 Hz_40ms_modulation_mae_db": 0.5121397030577657, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5942159477597595, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19953815237309416, "3000–6000 Hz_10ms_envelope_correlation": 0.9732388445520995, "3000–6000 Hz_10ms_modulation_mae_db": 0.47683044201763963, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0868462101329481, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10615589041435165, "3000–6000 Hz_40ms_envelope_correlation": 0.9921251152856567, "3000–6000 Hz_40ms_modulation_mae_db": 0.5364897678086157, "mr_stft_db": 5.78317204987646, "band_mae_db": 0.30548354467394256, "rms_gain_error_db": -0.25599926454832717, "output_peak": 0.7513452809950255}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49600, "output_samples": 49600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.870240221955103, "spectral_convergence": 0.46792139751717227, "spectral_convergence_status": "ok", "amplitude_floor": 1.4924605057201793e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.551347716587647, "spectral_convergence": 0.44137757810320133, "spectral_convergence_status": "ok", "amplitude_floor": 8.127484488587521e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.90961812435277, "spectral_convergence": 0.5037411109549662, "spectral_convergence_status": "ok", "amplitude_floor": 5.495951303547364e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185024372796483, "output_power": 0.0007662843806170188, "delta_db": -0.28630003657458053, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004929662452993321, "output_power": 0.0005304489626018898, "delta_db": 0.3182642139303917, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001304221454098428, "output_power": 0.0011586585255105265, "delta_db": -0.5139587840298641, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007846801909808332, "output_power": 0.0008292264432879042, "delta_db": 0.23980453792382384, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003122815376510157, "output_power": 0.0003225275804795018, "delta_db": 0.14020549492111023, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001061689290478197, "output_power": 0.00011474725847515312, "delta_db": 0.33744881688825534, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.30599698071133763, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11165377592152953, "pitch": null, "output_rms_dbfs": -24.292360988595618, "output_peak": 0.778308374912595, "reference_rms_dbfs": -24.180707212674086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.777068687631839}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.0002690918479937382, "log_rms_mae_db": 1.0389921316662662, "rms_envelope_nrmse": 0.0911518353487137, "envelope_correlation": 0.9932176066400141, "rms_gain_error_db": -0.11165339676057963, "reference_mean_power": 0.0038249880205223005, "output_mean_power": 0.003727904168758025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 2.1583633605855604e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008154057878800693, "output_power": 0.0007260007917571297, "reference_active": true, "delta_db": -0.5043669548808886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.087495849252009e-05, "output_power": 9.493881558003037e-05, "reference_active": true, "delta_db": 0.18999583824537566}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.599429116228097e-05, "output_power": 2.5940649155393308e-05, "reference_active": true, "delta_db": -0.008971392180829492}], "active_band_count": 3, "active_band_mae_db": 0.23444472843569794}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.0002456731948415248, "log_rms_mae_db": 0.5292015620534017, "rms_envelope_nrmse": 0.05377702060306877, "envelope_correlation": 0.9976385901223987, "rms_gain_error_db": -0.11165355914738594, "reference_mean_power": 0.0038624153279842384, "output_mean_power": 0.003764381375061063}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 2.0216648233438394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007562450561775054, "output_power": 0.0006703872159068503, "reference_active": true, "delta_db": -0.523368247148456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1399930780791825e-05, "output_power": 4.368909622744245e-05, "reference_active": true, "delta_db": 0.23373445792275072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0921358976760216e-06, "output_power": 1.210867472234134e-06, "reference_active": true, "delta_db": 0.4481993054662444}], "active_band_count": 3, "active_band_mae_db": 0.401767336845817}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00017000794064843354, "log_rms_mae_db": 1.78261957150753, "rms_envelope_nrmse": 0.18721683779204304, "envelope_correlation": 0.9767621693730923, "rms_gain_error_db": -0.28366225317471583, "reference_mean_power": 0.001461911046007658, "output_mean_power": 0.0013694769352820037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.2056298552980857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006038941034712422, "output_power": 0.0005048166814909691, "reference_active": true, "delta_db": -0.7782709151823619}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.14042207191201e-05, "output_power": 7.715246057465769e-05, "reference_active": true, "delta_db": 0.3362589752260341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.020128231954835e-05, "output_power": 3.002030833682513e-05, "reference_active": true, "delta_db": 1.720362104517129}], "active_band_count": 3, "active_band_mae_db": 0.9449639983085083}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00015858554889566586, "log_rms_mae_db": 1.0510537968912101, "rms_envelope_nrmse": 0.09507517105066385, "envelope_correlation": 0.9943321659279132, "rms_gain_error_db": -0.283660974135233, "reference_mean_power": 0.0014762175799508475, "output_mean_power": 0.0013828792990626228}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.148107626412446e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000571116387197927, "output_power": 0.00048692047250780345, "reference_active": true, "delta_db": -0.6926658688865633}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.637477969141665e-05, "output_power": 3.727843816718898e-05, "reference_active": true, "delta_db": 0.10657337041546508}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.172790453481755e-06, "output_power": 1.5180412102207227e-06, "reference_active": true, "delta_db": 1.1206313934630794}], "active_band_count": 3, "active_band_mae_db": 0.6399568775883693}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00019136572750236326, "log_rms_mae_db": 1.4852577944391085, "rms_envelope_nrmse": 0.1341175125963419, "envelope_correlation": 0.9893800596204558, "rms_gain_error_db": 0.15469157516927218, "reference_mean_power": 0.0009651549113833473, "output_mean_power": 0.0010001524025159652}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 7.99902404373172e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003359743599084894, "output_power": 0.0003491179715903224, "reference_active": true, "delta_db": 0.166660704201137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9966531082920368e-05, "output_power": 3.348331918041131e-05, "reference_active": true, "delta_db": 0.4819203074448224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.661292025781732e-06, "output_power": 1.0176555889171056e-05, "reference_active": true, "delta_db": 1.840423488653831}], "active_band_count": 3, "active_band_mae_db": 0.8296681667665968}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00016758862552795117, "log_rms_mae_db": 0.9836875423724742, "rms_envelope_nrmse": 0.0642137137602752, "envelope_correlation": 0.9976924580447909, "rms_gain_error_db": 0.1546903849818064, "reference_mean_power": 0.0009746009072787481, "output_mean_power": 0.0010099406429804744}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 7.731649944214162e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031280752850392277, "output_power": 0.0003301635309331587, "reference_active": true, "delta_db": 0.23451903547422703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5385931815426458e-05, "output_power": 1.751302048371441e-05, "reference_active": true, "delta_db": 0.562372521582159}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1192579703922553e-07, "output_power": 4.3885214656342095e-07, "reference_active": true, "delta_db": 1.4826693349199647}], "active_band_count": 3, "active_band_mae_db": 0.7598536306587835}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 5.61189760356505e-05, "log_rms_mae_db": 1.6521577469845032, "rms_envelope_nrmse": 0.2142674764169496, "envelope_correlation": 0.9710809380364765, "rms_gain_error_db": 0.34301357370820545, "reference_mean_power": 0.00014982208037105655, "output_mean_power": 0.00016213515108425813}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.5994961899965245e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.519717012759751e-05, "output_power": 6.0282470042653634e-05, "reference_active": true, "delta_db": 0.38274226293709435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5683811710113524e-05, "output_power": 1.5067545624782138e-05, "reference_active": true, "delta_db": -0.17409104602677386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1546072478197644e-06, "output_power": 3.5184027973320318e-06, "reference_active": true, "delta_db": 0.4740026051566771}], "active_band_count": 3, "active_band_mae_db": 0.34361197137351507}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 4.51769085602287e-05, "log_rms_mae_db": 1.1859526986399074, "rms_envelope_nrmse": 0.10676681803125772, "envelope_correlation": 0.9933440452396317, "rms_gain_error_db": 0.34301422737490567, "reference_mean_power": 0.0001512883740521334, "output_mean_power": 0.00016372197619620457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.4928738695410609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.9695472955753735e-05, "output_power": 5.4264175957856806e-05, "reference_active": true, "delta_db": 0.3819638413695256}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.547268075821193e-06, "output_power": 7.976888398839709e-06, "reference_active": true, "delta_db": -0.2999380928019057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7628269396099532e-07, "output_power": 1.9255801297936465e-07, "reference_active": true, "delta_db": 0.38351916973925265}], "active_band_count": 3, "active_band_mae_db": 0.355140367970228}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.856406892273357, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "188ffd385ae9c9b55947079ab9c8e501d44dce486f11c6b167e5c4e4daba550d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9790619750969172, "standard_deviation": 0.0072853956813444965, "minimum": 0.9711041408222192, "maximum": 0.9897999728606385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4453984695947995, "standard_deviation": 0.02008305216981389, "minimum": 1.4127205914637055, "maximum": 1.4852577944391085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.32406169074148, "standard_deviation": 0.6391278434778506, "minimum": 0.3531951753715741, "maximum": 2.425714709160308, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18320680435046155, "standard_deviation": 0.03363722376288945, "minimum": 0.12894148194568822, "maximum": 0.21571240656676627, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955428496756472, "standard_deviation": 0.001760313138082844, "minimum": 0.9933646712104434, "maximum": 0.9979927385714124, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9586498840649407, "standard_deviation": 0.05002208511281009, "minimum": 0.8649505302215098, "maximum": 1.0115641975747311, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2895598041984146, "standard_deviation": 0.7466297315181315, "minimum": 0.3471114868028457, "maximum": 2.4014587569574988, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08893015045207248, "standard_deviation": 0.020949366664295685, "minimum": 0.0591696089956101, "maximum": 0.11288512213861311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9695735182003176, "standard_deviation": 0.0046907714651622376, "minimum": 0.9616456158750317, "maximum": 0.9759127933759215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6667281984440014, "standard_deviation": 0.06257503380398549, "minimum": 1.5942159477597595, "maximum": 1.758281203260976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6416397641289612, "standard_deviation": 0.2470793713030718, "minimum": 0.34361197137351507, "maximum": 0.9190845225001242, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21243270828098237, "standard_deviation": 0.01587001600920104, "minimum": 0.18847854503671474, "maximum": 0.2385717953426653, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991371321653352, "standard_deviation": 0.0018001123167942978, "minimum": 0.9889445764518977, "maximum": 0.9933440452396317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1664045649843715, "standard_deviation": 0.048368688535071314, "minimum": 1.0868462101329481, "maximum": 1.2468418979576639, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6549280801310758, "standard_deviation": 0.20466588943616637, "minimum": 0.355140367970228, "maximum": 0.9340471932629315, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11136612928407125, "standard_deviation": 0.010678391403461713, "minimum": 0.09668323097471777, "maximum": 0.12515037464682688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.977518336309502, "standard_deviation": 0.007375548642596078, "minimum": 0.9624584677092101, "maximum": 0.9860197218521337, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8138750064235059, "standard_deviation": 0.04082999035351973, "minimum": 1.7637841449756668, "maximum": 1.8953822684793433, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1863032145057435, "standard_deviation": 0.3615425333458323, "minimum": 0.7101226694173811, "maximum": 1.8003239093135086, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1845425529660708, "standard_deviation": 0.02866134776807477, "minimum": 0.1474492225837557, "maximum": 0.2425382201177979, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9946260631715838, "standard_deviation": 0.0021870696592089477, "minimum": 0.9900097333023617, "maximum": 0.997558677314363, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0761098019176512, "standard_deviation": 0.052156606313596116, "minimum": 1.0240537952450255, "maximum": 1.148910198340057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.2754905559054408, "standard_deviation": 0.3727577754919987, "minimum": 0.6399568775883693, "maximum": 1.8233673139895121, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0962906587001876, "standard_deviation": 0.0208698577905697, "minimum": 0.06957195697806316, "maximum": 0.1430299501339824, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.31407109694780033, "standard_deviation": 0.08483520616815543, "minimum": 0.16504083203562478, "maximum": 0.4378730600214038, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.792280224903985, "standard_deviation": 0.010197379280748715, "minimum": 5.777068687631839, "maximum": 5.805658131946603, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7749465968599035, "standard_deviation": 0.052175907536362136, "minimum": 0.6718593671138687, "maximum": 0.854511358461779, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.18522820966864093, "standard_deviation": 0.14362995143747614, "minimum": -0.4558381669688472, "maximum": -0.031727708966625266, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9894924762606165, "standard_deviation": 0.0031466890532676228, "minimum": 0.9860990681868436, "maximum": 0.9945806569697028, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.067602060348067, "standard_deviation": 0.028355949496161558, "minimum": 1.0293102672042675, "maximum": 1.10511577017212, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6956231112474531, "standard_deviation": 0.27137795990463487, "minimum": 0.23444472843569794, "maximum": 1.106592693676542, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.11337271863291798, "standard_deviation": 0.01907143489972567, "minimum": 0.08178382665717943, "maximum": 0.1366603405450498, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975043685690339, "standard_deviation": 0.0008798909787815316, "minimum": 0.9963564228742221, "maximum": 0.9990367952754078, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5599540953537709, "standard_deviation": 0.02262353190040775, "minimum": 0.5292015620534017, "maximum": 0.603249625953429, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9483727380686644, "standard_deviation": 0.33172212079347346, "minimum": 0.401767336845817, "maximum": 1.4410269224240098, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05842968186344771, "standard_deviation": 0.013424863985397611, "minimum": 0.036134400678116754, "maximum": 0.08276133866282413, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0010", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7464699745178223, "predicted_mos": 2.7464699745178223, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "898bffed1f302290b1e606140199534489f5848d37e77c7c2253e9d5b44d6eb2", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8433202002197504, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0277205133769196, "全帯域_10ms_rms_envelope_nrmse": 0.09489118415641386, "全帯域_10ms_envelope_correlation": 0.9927159569496926, "全帯域_10ms_modulation_mae_db": 0.500658601530951, "全帯域_40ms_log_rms_mae_db": 0.5298463034124455, "全帯域_40ms_rms_envelope_nrmse": 0.045499223181426386, "全帯域_40ms_envelope_correlation": 0.9985036592438596, "全帯域_40ms_modulation_mae_db": 0.6151779842779583, "300–1000 Hz_10ms_log_rms_mae_db": 1.7206809043752405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16509118167865763, "300–1000 Hz_10ms_envelope_correlation": 0.9830798790058664, "300–1000 Hz_10ms_modulation_mae_db": 1.2953388503568377, "300–1000 Hz_40ms_log_rms_mae_db": 0.9759781914455924, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09751698746605479, "300–1000 Hz_40ms_envelope_correlation": 0.9954269222206781, "300–1000 Hz_40ms_modulation_mae_db": 1.1575823730154013, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3671969710890928, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1446321736870782, "1000–3000 Hz_10ms_envelope_correlation": 0.9870743854088984, "1000–3000 Hz_10ms_modulation_mae_db": 1.1207185809420912, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8758074146267617, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05931122434890588, "1000–3000 Hz_40ms_envelope_correlation": 0.997839932836114, "1000–3000 Hz_40ms_modulation_mae_db": 1.0238210384670883, "3000–6000 Hz_10ms_log_rms_mae_db": 1.617975650749739, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20954887310096632, "3000–6000 Hz_10ms_envelope_correlation": 0.9718825785693891, "3000–6000 Hz_10ms_modulation_mae_db": 0.635916972155564, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1298885721328629, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11303499757125882, "3000–6000 Hz_40ms_envelope_correlation": 0.9920188227580062, "3000–6000 Hz_40ms_modulation_mae_db": 1.0115288959401556, "mr_stft_db": 5.764549687701648, "band_mae_db": 0.22515649624874845, "rms_gain_error_db": -0.14663704050355197, "output_peak": 0.7868893595971587}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0895699092474387, "全帯域_10ms_rms_envelope_nrmse": 0.12073803996209946, "全帯域_10ms_envelope_correlation": 0.9880483177187581, "全帯域_10ms_modulation_mae_db": 1.6829580380674376, "全帯域_40ms_log_rms_mae_db": 0.6249996300727956, "全帯域_40ms_rms_envelope_nrmse": 0.05461098142580674, "全帯域_40ms_envelope_correlation": 0.9976095703123917, "全帯域_40ms_modulation_mae_db": 1.773583467880546, "300–1000 Hz_10ms_log_rms_mae_db": 1.7386228798914691, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17497045811339382, "300–1000 Hz_10ms_envelope_correlation": 0.9797063174642273, "300–1000 Hz_10ms_modulation_mae_db": 1.9660145396193347, "300–1000 Hz_40ms_log_rms_mae_db": 1.0144690946186232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09628390893543874, "300–1000 Hz_40ms_envelope_correlation": 0.993978831743094, "300–1000 Hz_40ms_modulation_mae_db": 1.8527789729479258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3866669993861354, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17361222071655286, "1000–3000 Hz_10ms_envelope_correlation": 0.9814358152097137, "1000–3000 Hz_10ms_modulation_mae_db": 1.962664822160671, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8981917759498247, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07688936084059317, "1000–3000 Hz_40ms_envelope_correlation": 0.9965622479347932, "1000–3000 Hz_40ms_modulation_mae_db": 1.9028686536018469, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5830587318012594, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19650980637601406, "3000–6000 Hz_10ms_envelope_correlation": 0.973427315126437, "3000–6000 Hz_10ms_modulation_mae_db": 0.5216472653927852, "3000–6000 Hz_40ms_log_rms_mae_db": 1.102050273826088, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11058687032562292, "3000–6000 Hz_40ms_envelope_correlation": 0.9916199485423539, "3000–6000 Hz_40ms_modulation_mae_db": 0.45928947283285443, "mr_stft_db": 5.75198258378851, "band_mae_db": 0.21929983461195304, "rms_gain_error_db": -0.1306223911780906, "output_peak": 0.9265347283430998}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0339561005049014, "全帯域_10ms_rms_envelope_nrmse": 0.09419425600791488, "全帯域_10ms_envelope_correlation": 0.9927070459989207, "全帯域_10ms_modulation_mae_db": 0.34202124327811106, "全帯域_40ms_log_rms_mae_db": 0.5585061257599161, "全帯域_40ms_rms_envelope_nrmse": 0.04759069485101106, "全帯域_40ms_envelope_correlation": 0.9980983008361111, "全帯域_40ms_modulation_mae_db": 0.6198188852027346, "300–1000 Hz_10ms_log_rms_mae_db": 1.7768571815428986, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16906503186441207, "300–1000 Hz_10ms_envelope_correlation": 0.9810710546866037, "300–1000 Hz_10ms_modulation_mae_db": 0.5311542709330077, "300–1000 Hz_40ms_log_rms_mae_db": 1.0892782907036311, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10411352067816533, "300–1000 Hz_40ms_envelope_correlation": 0.9927159541479671, "300–1000 Hz_40ms_modulation_mae_db": 0.5556981529641113, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3055823892398295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1658312677482075, "1000–3000 Hz_10ms_envelope_correlation": 0.9829645955061652, "1000–3000 Hz_10ms_modulation_mae_db": 1.212112128721116, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8056653219854566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0696485615118595, "1000–3000 Hz_40ms_envelope_correlation": 0.9970507523466675, "1000–3000 Hz_40ms_modulation_mae_db": 1.5552546254897421, "3000–6000 Hz_10ms_log_rms_mae_db": 1.654295546171263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21977001903616714, "3000–6000 Hz_10ms_envelope_correlation": 0.9674240546218039, "3000–6000 Hz_10ms_modulation_mae_db": 1.0126765456252116, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1777272477748204, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13274336612545906, "3000–6000 Hz_40ms_envelope_correlation": 0.9876042670008485, "3000–6000 Hz_40ms_modulation_mae_db": 1.0556594606789518, "mr_stft_db": 5.752135759305279, "band_mae_db": 0.29698615882853846, "rms_gain_error_db": -0.059698303505794574, "output_peak": 0.6999996950878233}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9892952462495034, "全帯域_10ms_rms_envelope_nrmse": 0.09639999365068175, "全帯域_10ms_envelope_correlation": 0.9924208995967604, "全帯域_10ms_modulation_mae_db": 1.060531849991968, "全帯域_40ms_log_rms_mae_db": 0.44151240952099885, "全帯域_40ms_rms_envelope_nrmse": 0.05381785901530751, "全帯域_40ms_envelope_correlation": 0.9976822382236145, "全帯域_40ms_modulation_mae_db": 1.1968049067053803, "300–1000 Hz_10ms_log_rms_mae_db": 1.7230618995355138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15151940593319618, "300–1000 Hz_10ms_envelope_correlation": 0.9847701863702873, "300–1000 Hz_10ms_modulation_mae_db": 0.6050925957437513, "300–1000 Hz_40ms_log_rms_mae_db": 0.9477108382141443, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09147146802718001, "300–1000 Hz_40ms_envelope_correlation": 0.9944213001395458, "300–1000 Hz_40ms_modulation_mae_db": 0.6026626979114785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4634018914965794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20830967285383417, "1000–3000 Hz_10ms_envelope_correlation": 0.9731322566874582, "1000–3000 Hz_10ms_modulation_mae_db": 2.290142418885621, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9323435214840587, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1044295921012684, "1000–3000 Hz_40ms_envelope_correlation": 0.9935532964504311, "1000–3000 Hz_40ms_modulation_mae_db": 3.148470206268261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6167319959095978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20554488451592492, "3000–6000 Hz_10ms_envelope_correlation": 0.9721022596105763, "3000–6000 Hz_10ms_modulation_mae_db": 0.5935662027326357, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1500219166680556, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11806923135532654, "3000–6000 Hz_40ms_envelope_correlation": 0.9905513366970042, "3000–6000 Hz_40ms_modulation_mae_db": 0.6177772099198043, "mr_stft_db": 5.76501886183178, "band_mae_db": 0.28308133152663567, "rms_gain_error_db": -0.13170883117344112, "output_peak": 0.7764468832671847}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0906451958233205, "全帯域_10ms_rms_envelope_nrmse": 0.11323399131820462, "全帯域_10ms_envelope_correlation": 0.9894411529617732, "全帯域_10ms_modulation_mae_db": 0.42997112455226166, "全帯域_40ms_log_rms_mae_db": 0.608587331683318, "全帯域_40ms_rms_envelope_nrmse": 0.04113828866240096, "全帯域_40ms_envelope_correlation": 0.9985528355299048, "全帯域_40ms_modulation_mae_db": 0.6315010472117674, "300–1000 Hz_10ms_log_rms_mae_db": 1.695753628099276, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16150325374026592, "300–1000 Hz_10ms_envelope_correlation": 0.9826333294756965, "300–1000 Hz_10ms_modulation_mae_db": 0.3899146345685564, "300–1000 Hz_40ms_log_rms_mae_db": 0.9368544748745448, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07003020303898376, "300–1000 Hz_40ms_envelope_correlation": 0.9967595730113409, "300–1000 Hz_40ms_modulation_mae_db": 0.2656323372781551, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4136560636779674, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.207864484254684, "1000–3000 Hz_10ms_envelope_correlation": 0.9737710487771444, "1000–3000 Hz_10ms_modulation_mae_db": 2.0986300543600036, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8491676419046934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08409827810990451, "1000–3000 Hz_40ms_envelope_correlation": 0.9956494769665727, "1000–3000 Hz_40ms_modulation_mae_db": 2.645940534418186, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5799804834927063, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19442754083999095, "3000–6000 Hz_10ms_envelope_correlation": 0.9758044325970223, "3000–6000 Hz_10ms_modulation_mae_db": 0.16876465604550125, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0953494550790839, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11280583843566828, "3000–6000 Hz_40ms_envelope_correlation": 0.9919720493678004, "3000–6000 Hz_40ms_modulation_mae_db": 0.34579687816282734, "mr_stft_db": 5.758838187752107, "band_mae_db": 0.3219568701462946, "rms_gain_error_db": -0.017584132016928416, "output_peak": 0.6821196311387423}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.092019022414691, "全帯域_10ms_rms_envelope_nrmse": 0.10304711630442258, "全帯域_10ms_envelope_correlation": 0.9913451843991589, "全帯域_10ms_modulation_mae_db": 1.0145049889619815, "全帯域_40ms_log_rms_mae_db": 0.5627178379846487, "全帯域_40ms_rms_envelope_nrmse": 0.05450486153904839, "全帯域_40ms_envelope_correlation": 0.9976788647209136, "全帯域_40ms_modulation_mae_db": 0.9860883605604563, "300–1000 Hz_10ms_log_rms_mae_db": 1.7086333768141226, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16248476507392695, "300–1000 Hz_10ms_envelope_correlation": 0.9824798026762638, "300–1000 Hz_10ms_modulation_mae_db": 0.9898531786446515, "300–1000 Hz_40ms_log_rms_mae_db": 1.0452209718234784, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08513171698059815, "300–1000 Hz_40ms_envelope_correlation": 0.995314163921258, "300–1000 Hz_40ms_modulation_mae_db": 0.9331409181369495, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4818624512696887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18424901625480442, "1000–3000 Hz_10ms_envelope_correlation": 0.9795007172578558, "1000–3000 Hz_10ms_modulation_mae_db": 0.9651332171313411, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9708044413394498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12094909978299337, "1000–3000 Hz_40ms_envelope_correlation": 0.9918031801304242, "1000–3000 Hz_40ms_modulation_mae_db": 0.7829151010790206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.589383838963809, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21492811032365533, "3000–6000 Hz_10ms_envelope_correlation": 0.9692226729713951, "3000–6000 Hz_10ms_modulation_mae_db": 0.3624416229058805, "3000–6000 Hz_40ms_log_rms_mae_db": 1.083145836852919, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13929149139216607, "3000–6000 Hz_40ms_envelope_correlation": 0.9865816514083139, "3000–6000 Hz_40ms_modulation_mae_db": 0.5137280400665535, "mr_stft_db": 5.767507210033448, "band_mae_db": 0.22498181503445633, "rms_gain_error_db": -0.1488108842837916, "output_peak": 0.7390453084156321}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0800187471291782, "全帯域_10ms_rms_envelope_nrmse": 0.09434096690308746, "全帯域_10ms_envelope_correlation": 0.9928493788167361, "全帯域_10ms_modulation_mae_db": 0.37081485616482146, "全帯域_40ms_log_rms_mae_db": 0.5734298106176143, "全帯域_40ms_rms_envelope_nrmse": 0.04905235319412522, "全帯域_40ms_envelope_correlation": 0.9980650579932804, "全帯域_40ms_modulation_mae_db": 0.5863623145312794, "300–1000 Hz_10ms_log_rms_mae_db": 1.7735489744360722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15258657874661227, "300–1000 Hz_10ms_envelope_correlation": 0.9845445928350673, "300–1000 Hz_10ms_modulation_mae_db": 0.9136933649271258, "300–1000 Hz_40ms_log_rms_mae_db": 1.0220902015045517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07462245530831497, "300–1000 Hz_40ms_envelope_correlation": 0.9962252571443938, "300–1000 Hz_40ms_modulation_mae_db": 1.0422994811403237, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3405520613775228, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18128447560584676, "1000–3000 Hz_10ms_envelope_correlation": 0.9807949668759275, "1000–3000 Hz_10ms_modulation_mae_db": 1.8675887557569446, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8593580248743842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10381778949087656, "1000–3000 Hz_40ms_envelope_correlation": 0.993873240283754, "1000–3000 Hz_40ms_modulation_mae_db": 2.041672225405174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5983287183531731, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2233009976388634, "3000–6000 Hz_10ms_envelope_correlation": 0.9659969116727231, "3000–6000 Hz_10ms_modulation_mae_db": 0.912880999116387, "3000–6000 Hz_40ms_log_rms_mae_db": 1.107726508697115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10874695620280958, "3000–6000 Hz_40ms_envelope_correlation": 0.9915650034195439, "3000–6000 Hz_40ms_modulation_mae_db": 1.225612096569036, "mr_stft_db": 5.775799586798019, "band_mae_db": 0.24662322145720594, "rms_gain_error_db": 0.048862790589362715, "output_peak": 0.7571881126701809}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0398013629730414, "全帯域_10ms_rms_envelope_nrmse": 0.08598179549263506, "全帯域_10ms_envelope_correlation": 0.9939686219468873, "全帯域_10ms_modulation_mae_db": 0.323304856894486, "全帯域_40ms_log_rms_mae_db": 0.5210004950890185, "全帯域_40ms_rms_envelope_nrmse": 0.039752311452520304, "全帯域_40ms_envelope_correlation": 0.9987159960721659, "全帯域_40ms_modulation_mae_db": 0.2599891312101379, "300–1000 Hz_10ms_log_rms_mae_db": 1.8237584641319446, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.15370797115146154, "300–1000 Hz_10ms_envelope_correlation": 0.9844845653797583, "300–1000 Hz_10ms_modulation_mae_db": 0.7022415556583498, "300–1000 Hz_40ms_log_rms_mae_db": 1.0994785160410965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06897399059767013, "300–1000 Hz_40ms_envelope_correlation": 0.9972293356273334, "300–1000 Hz_40ms_modulation_mae_db": 0.42757170137029155, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3730478729416935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16641977704141703, "1000–3000 Hz_10ms_envelope_correlation": 0.9834632860576137, "1000–3000 Hz_10ms_modulation_mae_db": 1.368837727030978, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8731979431059683, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07593197054010818, "1000–3000 Hz_40ms_envelope_correlation": 0.996633214128558, "1000–3000 Hz_40ms_modulation_mae_db": 1.679160187009545, "3000–6000 Hz_10ms_log_rms_mae_db": 1.625332654536297, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20427106170440776, "3000–6000 Hz_10ms_envelope_correlation": 0.9714621018881007, "3000–6000 Hz_10ms_modulation_mae_db": 1.141256765820432, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0967389384228359, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11172711003283543, "3000–6000 Hz_40ms_envelope_correlation": 0.9918708999492227, "3000–6000 Hz_40ms_modulation_mae_db": 1.6396949752300392, "mr_stft_db": 5.752937705752338, "band_mae_db": 0.20913966356095634, "rms_gain_error_db": -0.06217341600756421, "output_peak": 0.8854537397004429}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49600, "output_samples": 49600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.860268494915011, "spectral_convergence": 0.4658823065826125, "spectral_convergence_status": "ok", "amplitude_floor": 1.4924605057201793e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.534903424638718, "spectral_convergence": 0.44334103816449133, "spectral_convergence_status": "ok", "amplitude_floor": 8.127484488587521e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.898477143551216, "spectral_convergence": 0.4930539654813434, "spectral_convergence_status": "ok", "amplitude_floor": 5.495951303547364e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185024372796483, "output_power": 0.0008565848790881582, "delta_db": 0.19750426759112877, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004929662452993321, "output_power": 0.0005033045252343559, "delta_db": 0.09013652171351068, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001304221454098428, "output_power": 0.0011273508095194851, "delta_db": -0.6329225893476099, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007846801909808332, "output_power": 0.0007682817671943558, "delta_db": -0.09172162445855078, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003122815376510157, "output_power": 0.0003268468233647217, "delta_db": 0.1979795926818613, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001061689290478197, "output_power": 0.0001096641999795341, "delta_db": 0.1406743816998292, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22515649624874845, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14663704050355197, "pitch": null, "output_rms_dbfs": -24.327344253177642, "output_peak": 0.7868893595971587, "reference_rms_dbfs": -24.180707212674086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.764549687701648}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.0002690918479937382, "log_rms_mae_db": 1.0277205133769196, "rms_envelope_nrmse": 0.09489118415641386, "envelope_correlation": 0.9927159569496926, "rms_gain_error_db": -0.14663688574301653, "reference_mean_power": 0.0038249880205223005, "output_mean_power": 0.003697995605667632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 2.1583633605855604e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008154057878800693, "output_power": 0.0007398310141378216, "reference_active": true, "delta_db": -0.42241256585218706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.087495849252009e-05, "output_power": 8.137979613054149e-05, "reference_active": true, "delta_db": -0.47927627960291086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.599429116228097e-05, "output_power": 2.984741163231903e-05, "reference_active": true, "delta_db": 0.6002869591377549}], "active_band_count": 3, "active_band_mae_db": 0.500658601530951}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.0002456731948415248, "log_rms_mae_db": 0.5298463034124455, "rms_envelope_nrmse": 0.045499223181426386, "envelope_correlation": 0.9985036592438596, "rms_gain_error_db": -0.1466370217047323, "reference_mean_power": 0.0038624153279842384, "output_mean_power": 0.0037341801820974267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 2.0216648233438394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007562450561775054, "output_power": 0.0006857928072797288, "reference_active": true, "delta_db": -0.4246962270110233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1399930780791825e-05, "output_power": 3.602374385429767e-05, "reference_active": true, "delta_db": -0.6041076906807218}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0921358976760216e-06, "output_power": 1.3181040237126876e-06, "reference_active": true, "delta_db": 0.8167300351421297}], "active_band_count": 3, "active_band_mae_db": 0.6151779842779583}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00017000794064843354, "log_rms_mae_db": 1.7206809043752405, "rms_envelope_nrmse": 0.16509118167865763, "envelope_correlation": 0.9830798790058664, "rms_gain_error_db": -0.491068642792198, "reference_mean_power": 0.001461911046007658, "output_mean_power": 0.00130561185039979}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.2056298552980857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006038941034712422, "output_power": 0.000491843553634348, "reference_active": true, "delta_db": -0.8913380543521718}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.14042207191201e-05, "output_power": 5.461736441736481e-05, "reference_active": true, "delta_db": -1.163931445203823}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.020128231954835e-05, "output_power": 3.0793116370729555e-05, "reference_active": true, "delta_db": 1.8307470515145186}], "active_band_count": 3, "active_band_mae_db": 1.2953388503568377}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00015858554889566586, "log_rms_mae_db": 0.9759781914455924, "rms_envelope_nrmse": 0.09751698746605479, "envelope_correlation": 0.9954269222206781, "rms_gain_error_db": -0.4910687420308338, "reference_mean_power": 0.0014762175799508475, "output_mean_power": 0.0013183887811603887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.148107626412446e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000571116387197927, "output_power": 0.0004709173988349867, "reference_active": true, "delta_db": -0.8377988519441566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.637477969141665e-05, "output_power": 2.631925989050308e-05, "reference_active": true, "delta_db": -1.4052669908463666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.172790453481755e-06, "output_power": 1.5566413037162007e-06, "reference_active": true, "delta_db": 1.2296812762556808}], "active_band_count": 3, "active_band_mae_db": 1.1575823730154013}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00019136572750236326, "log_rms_mae_db": 1.3671969710890928, "rms_envelope_nrmse": 0.1446321736870782, "envelope_correlation": 0.9870743854088984, "rms_gain_error_db": -0.09182336608913379, "reference_mean_power": 0.0009651549113833473, "output_mean_power": 0.0009449627483599344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 7.99902404373172e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003359743599084894, "output_power": 0.0003309334846976438, "reference_active": true, "delta_db": -0.06565422790672848}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9966531082920368e-05, "output_power": 2.3598398260203546e-05, "reference_active": true, "delta_db": -1.037539457902328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.661292025781732e-06, "output_power": 1.1206104779636528e-05, "reference_active": true, "delta_db": 2.258962057017217}], "active_band_count": 3, "active_band_mae_db": 1.1207185809420912}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00016758862552795117, "log_rms_mae_db": 0.8758074146267617, "rms_envelope_nrmse": 0.05931122434890588, "envelope_correlation": 0.997839932836114, "rms_gain_error_db": -0.09182311010098632, "reference_mean_power": 0.0009746009072787481, "output_mean_power": 0.0009542111792819434}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 7.731649944214162e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031280752850392277, "output_power": 0.00030768442401389845, "reference_active": true, "delta_db": -0.07171685503187378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5385931815426458e-05, "output_power": 1.1777622682139738e-05, "reference_active": true, "delta_db": -1.1606616669675676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1192579703922553e-07, "output_power": 4.7638683676965607e-07, "reference_active": true, "delta_db": 1.8390845934018232}], "active_band_count": 3, "active_band_mae_db": 1.0238210384670883}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 5.61189760356505e-05, "log_rms_mae_db": 1.617975650749739, "rms_envelope_nrmse": 0.20954887310096632, "envelope_correlation": 0.9718825785693891, "rms_gain_error_db": 0.2967678057636347, "reference_mean_power": 0.00014982208037105655, "output_mean_power": 0.00016041781763727523}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.5994961899965245e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.519717012759751e-05, "output_power": 5.956021267365041e-05, "reference_active": true, "delta_db": 0.33039427130271387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5683811710113524e-05, "output_power": 1.5169312793878528e-05, "reference_active": true, "delta_db": -0.14485713141175136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1546072478197644e-06, "output_power": 4.387278947634966e-06, "reference_active": true, "delta_db": 1.4324995137522267}], "active_band_count": 3, "active_band_mae_db": 0.635916972155564}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 4.51769085602287e-05, "log_rms_mae_db": 1.1298885721328629, "rms_envelope_nrmse": 0.11303499757125882, "envelope_correlation": 0.9920188227580062, "rms_gain_error_db": 0.29676694511029067, "reference_mean_power": 0.0001512883740521334, "output_mean_power": 0.00016198777863519673}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.4928738695410609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.9695472955753735e-05, "output_power": 5.382469673749642e-05, "reference_active": true, "delta_db": 0.346647634007321}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.547268075821193e-06, "output_power": 7.980834424370523e-06, "reference_active": true, "delta_db": -0.2977902460603651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7628269396099532e-07, "output_power": 3.056501122172857e-07, "reference_active": true, "delta_db": 2.3901488077527806}], "active_band_count": 3, "active_band_mae_db": 1.0115288959401556}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8508319143833675, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c8dfefc6e8a225e6ae0d10ffc40f50858b080670b74fca7e639d788661c23473", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9802671339725971, "standard_deviation": 0.0047684644686537625, "minimum": 0.9731322566874582, "maximum": 0.9870743854088984, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3914958375598137, "standard_deviation": 0.05947166193132586, "minimum": 1.3055823892398295, "maximum": 1.4818624512696887, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6107284631235959, "standard_deviation": 0.501967089601701, "minimum": 0.9651332171313411, "maximum": 2.290142418885621, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17902538602030313, "standard_deviation": 0.02160302356091368, "minimum": 0.1446321736870782, "maximum": 0.20830967285383417, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953706676346643, "standard_deviation": 0.0020799612278151343, "minimum": 0.9918031801304242, "maximum": 0.997839932836114, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8830670106588246, "standard_deviation": 0.051000272595785066, "minimum": 0.8056653219854566, "maximum": 0.9708044413394498, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.8475128214673582, "standard_deviation": 0.7834295477895222, "minimum": 0.7829151010790206, "maximum": 3.148470206268261, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08688448459081369, "standard_deviation": 0.020839352727479565, "minimum": 0.05931122434890588, "maximum": 0.12094909978299337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9709152908821809, "standard_deviation": 0.0032093981430758745, "minimum": 0.9659969116727231, "maximum": 0.9758044325970223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6081359524972307, "standard_deviation": 0.02526026611511725, "minimum": 1.5799804834927063, "maximum": 1.654295546171263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6686438787242996, "standard_deviation": 0.3323784329501466, "minimum": 0.16876465604550125, "maximum": 1.141256765820432, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20853766169199872, "standard_deviation": 0.010398541976469678, "minimum": 0.19442754083999095, "maximum": 0.2233009976388634, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9904729973928867, "standard_deviation": 0.0021539718535730527, "minimum": 0.9865816514083139, "maximum": 0.9920188227580062, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1178310936817226, "standard_deviation": 0.032235047892257875, "minimum": 1.083145836852919, "maximum": 1.1777272477748204, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8586358786750279, "standard_deviation": 0.4482306694355892, "minimum": 0.34579687816282734, "maximum": 1.6396949752300392, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11837573268014334, "standard_deviation": 0.01134600537705388, "minimum": 0.10874695620280958, "maximum": 0.13929149139216607, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9828462159867213, "standard_deviation": 0.0017934273990510656, "minimum": 0.9797063174642273, "maximum": 0.9847701863702873, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7451146636033172, "standard_deviation": 0.04291822323558879, "minimum": 1.695753628099276, "maximum": 1.8237584641319446, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9241628738064519, "standard_deviation": 0.5097034407720477, "minimum": 0.3899146345685564, "maximum": 1.9660145396193347, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16136608078774078, "standard_deviation": 0.008385215583572145, "minimum": 0.15151940593319618, "maximum": 0.17497045811339382, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9952589172444514, "standard_deviation": 0.0015074334678116593, "minimum": 0.9927159541479671, "maximum": 0.9972293356273334, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0163850724032077, "standard_deviation": 0.06066259453208562, "minimum": 0.9368544748745448, "maximum": 1.0994785160410965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8546708293455796, "standard_deviation": 0.5087253125226558, "minimum": 0.2656323372781551, "maximum": 1.8527789729479258, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08601803137905074, "standard_deviation": 0.013477789272560204, "minimum": 0.06897399059767013, "maximum": 0.10411352067816533, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25340317392684863, "standard_deviation": 0.04183947414272183, "minimum": 0.20913966356095634, "maximum": 0.3219568701462946, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.7610961978703905, "standard_deviation": 0.008615696901816159, "minimum": 5.75198258378851, "maximum": 5.775799586798019, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7817096822775331, "standard_deviation": 0.0851954945106724, "minimum": 0.6821196311387423, "maximum": 0.9265347283430998, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08104652600997497, "standard_deviation": 0.07134864729053521, "minimum": -0.1488108842837916, "maximum": 0.048862790589362715, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.991687069798586, "standard_deviation": 0.0019854993475343008, "minimum": 0.9880483177187581, "maximum": 0.9939686219468873, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0553782622148744, "standard_deviation": 0.03816503450234415, "minimum": 0.9892952462495034, "maximum": 1.092019022414691, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7155956949302523, "standard_deviation": 0.49051296308536896, "minimum": 0.323304856894486, "maximum": 1.6829580380674376, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10035341797443245, "standard_deviation": 0.011431734856955447, "minimum": 0.08598179549263506, "maximum": 0.12073803996209946, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981133153665303, "standard_deviation": 0.00043746975502210106, "minimum": 0.9976095703123917, "maximum": 0.9987159960721659, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5525749930175945, "standard_deviation": 0.05707655122188506, "minimum": 0.44151240952099885, "maximum": 0.6249996300727956, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8336657621975325, "standard_deviation": 0.4725538246405099, "minimum": 0.2599891312101379, "maximum": 1.773583467880546, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04824582166520582, "standard_deviation": 0.005879974097988668, "minimum": 0.039752311452520304, "maximum": 0.05461098142580674, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8455-210777-0049", "language": "en", "style": "normal", "speaker": "librispeech_8455", "source_utterance_id": "8455-210777-0049", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 65760, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8455-210777-0049.wav", "source_file_sha256": "711261a38f32a1be6e9d29cc2bdb11e662337a22b6047df851d627a075bef6d9", "decoded_pcm_sha256": "6bab2226481307e41ffb714c528db253ee0755e0775a2890f32c07e3f5832d55", "original_source_sha256": "2c4c3f4fa85e1e234b108aeada818530820e7a515dc7d91f2a2c075af7105fed", "corpus_entry_id": "librispeech_8455-210777-0049"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.054941177368164, "predicted_mos": 4.054941177368164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "a6a2a45322093cf40be0045ad24596c6bdb50624a3eab13bb2c8e0ba7bf60176", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9626630049970117, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8455-210777-0049", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "a6a2a45322093cf40be0045ad24596c6bdb50624a3eab13bb2c8e0ba7bf60176", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "2d76f8265f7a52a2814eb8e00d8edc511804ac1369740fd998f7cd19435e2fe8"}, "rows": [{"case": "librispeech_8455-210777-0049", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.2878174781799316, "predicted_mos": 3.2878174781799316, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "92c5c93215022d96dc2ec16169ba217e0b7af7249ed56cff9aa4364967001d8e", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.085738866997417, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8455-210777-0049", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "92c5c93215022d96dc2ec16169ba217e0b7af7249ed56cff9aa4364967001d8e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1499672663213145, "全帯域_10ms_rms_envelope_nrmse": 0.07042627003476232, "全帯域_10ms_envelope_correlation": 0.9962150642202571, "全帯域_10ms_modulation_mae_db": 0.4061766294878984, "全帯域_40ms_log_rms_mae_db": 0.6468180228887991, "全帯域_40ms_rms_envelope_nrmse": 0.04454296664243185, "全帯域_40ms_envelope_correlation": 0.9985783755473815, "全帯域_40ms_modulation_mae_db": 0.8152575849006599, "300–1000 Hz_10ms_log_rms_mae_db": 1.5924590380145058, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09054496090754321, "300–1000 Hz_10ms_envelope_correlation": 0.9950314843445371, "300–1000 Hz_10ms_modulation_mae_db": 1.078224969195414, "300–1000 Hz_40ms_log_rms_mae_db": 0.8590994866527286, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05630891115857844, "300–1000 Hz_40ms_envelope_correlation": 0.9983718798769187, "300–1000 Hz_40ms_modulation_mae_db": 1.3838990570211245, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5334422393870342, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1501829855805913, "1000–3000 Hz_10ms_envelope_correlation": 0.9850507843409985, "1000–3000 Hz_10ms_modulation_mae_db": 0.7388795143562031, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9483586068251514, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09832274060409507, "1000–3000 Hz_40ms_envelope_correlation": 0.9936092194285684, "1000–3000 Hz_40ms_modulation_mae_db": 0.574310194283536, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6546381057882693, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1557233244440516, "3000–6000 Hz_10ms_envelope_correlation": 0.9853849980820946, "3000–6000 Hz_10ms_modulation_mae_db": 0.2889069959128358, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0779675970709321, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08836595672541142, "3000–6000 Hz_40ms_envelope_correlation": 0.9951624237231788, "3000–6000 Hz_40ms_modulation_mae_db": 0.1345208956048436, "mr_stft_db": 5.6408865309456955, "band_mae_db": 0.19932873202707702, "rms_gain_error_db": -0.14826744697184016, "output_peak": 0.4690082973669654}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1852150906370411, "全帯域_10ms_rms_envelope_nrmse": 0.0678817704457893, "全帯域_10ms_envelope_correlation": 0.9964000804420946, "全帯域_10ms_modulation_mae_db": 0.289218257967249, "全帯域_40ms_log_rms_mae_db": 0.5976216975732348, "全帯域_40ms_rms_envelope_nrmse": 0.03716590486312935, "全帯域_40ms_envelope_correlation": 0.9989705171937409, "全帯域_40ms_modulation_mae_db": 0.3936492824517172, "300–1000 Hz_10ms_log_rms_mae_db": 1.6236395809382131, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08603404293518897, "300–1000 Hz_10ms_envelope_correlation": 0.9952258310706648, "300–1000 Hz_10ms_modulation_mae_db": 1.0205438735227679, "300–1000 Hz_40ms_log_rms_mae_db": 0.8812905079761443, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046154991373695206, "300–1000 Hz_40ms_envelope_correlation": 0.9986976474492065, "300–1000 Hz_40ms_modulation_mae_db": 1.0987846207629752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.614385310829931, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14182257072509633, "1000–3000 Hz_10ms_envelope_correlation": 0.9875837702746301, "1000–3000 Hz_10ms_modulation_mae_db": 1.0058134515585528, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0130279846057597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0803893619735819, "1000–3000 Hz_40ms_envelope_correlation": 0.9970129499297153, "1000–3000 Hz_40ms_modulation_mae_db": 0.6927455861722304, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6082442933401495, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2463560141114081, "3000–6000 Hz_10ms_envelope_correlation": 0.9635769090948326, "3000–6000 Hz_10ms_modulation_mae_db": 0.2653038460809961, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0016007508857818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15263717811432126, "3000–6000 Hz_40ms_envelope_correlation": 0.98729362495768, "3000–6000 Hz_40ms_modulation_mae_db": 0.3719192135589404, "mr_stft_db": 5.634745060251763, "band_mae_db": 0.2497456433562942, "rms_gain_error_db": -0.11221868721288988, "output_peak": 0.4974551213630518}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.147608976779188, "全帯域_10ms_rms_envelope_nrmse": 0.07213259204584391, "全帯域_10ms_envelope_correlation": 0.9958421427943612, "全帯域_10ms_modulation_mae_db": 0.2585240741185785, "全帯域_40ms_log_rms_mae_db": 0.6418051203976168, "全帯域_40ms_rms_envelope_nrmse": 0.03865013077779634, "全帯域_40ms_envelope_correlation": 0.9987828808695741, "全帯域_40ms_modulation_mae_db": 0.3329122068959293, "300–1000 Hz_10ms_log_rms_mae_db": 1.638407714350404, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09318821062133238, "300–1000 Hz_10ms_envelope_correlation": 0.9943376074462505, "300–1000 Hz_10ms_modulation_mae_db": 0.6086249621235645, "300–1000 Hz_40ms_log_rms_mae_db": 0.8271994416029794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04519070654427531, "300–1000 Hz_40ms_envelope_correlation": 0.9987221203390003, "300–1000 Hz_40ms_modulation_mae_db": 0.7738994907021862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5470857104006304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1301636609306806, "1000–3000 Hz_10ms_envelope_correlation": 0.9894761094212347, "1000–3000 Hz_10ms_modulation_mae_db": 0.9115456828859244, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0240563677220365, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07689243119807915, "1000–3000 Hz_40ms_envelope_correlation": 0.9970220240164986, "1000–3000 Hz_40ms_modulation_mae_db": 0.9128652709327953, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6203981802459722, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22611087778303052, "3000–6000 Hz_10ms_envelope_correlation": 0.9751929518143236, "3000–6000 Hz_10ms_modulation_mae_db": 0.35284921620618315, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0650335586479145, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16276360330471604, "3000–6000 Hz_40ms_envelope_correlation": 0.9886189037203232, "3000–6000 Hz_40ms_modulation_mae_db": 0.4563740774516582, "mr_stft_db": 5.637305563633272, "band_mae_db": 0.23122831257382695, "rms_gain_error_db": -0.04299617586641996, "output_peak": 0.49358558130769625}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1254670985092095, "全帯域_10ms_rms_envelope_nrmse": 0.06375800390460856, "全帯域_10ms_envelope_correlation": 0.9968165978131449, "全帯域_10ms_modulation_mae_db": 0.47704297825115444, "全帯域_40ms_log_rms_mae_db": 0.6134228577243234, "全帯域_40ms_rms_envelope_nrmse": 0.0363871807286154, "全帯域_40ms_envelope_correlation": 0.998991644898541, "全帯域_40ms_modulation_mae_db": 0.71510605797313, "300–1000 Hz_10ms_log_rms_mae_db": 1.6598713515196077, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08896173031834896, "300–1000 Hz_10ms_envelope_correlation": 0.9948282985736463, "300–1000 Hz_10ms_modulation_mae_db": 0.9120961728092469, "300–1000 Hz_40ms_log_rms_mae_db": 0.9671910225969268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048998846286137675, "300–1000 Hz_40ms_envelope_correlation": 0.9984320809908391, "300–1000 Hz_40ms_modulation_mae_db": 0.9252565831858711, "1000–3000 Hz_10ms_log_rms_mae_db": 1.543936519663955, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14096020583421875, "1000–3000 Hz_10ms_envelope_correlation": 0.9875263704884453, "1000–3000 Hz_10ms_modulation_mae_db": 0.46799138146710334, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9699124942316273, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08458450416436814, "1000–3000 Hz_40ms_envelope_correlation": 0.9962216739662918, "1000–3000 Hz_40ms_modulation_mae_db": 0.3741049836890426, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6674777997714327, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18058737966714622, "3000–6000 Hz_10ms_envelope_correlation": 0.981010574839178, "3000–6000 Hz_10ms_modulation_mae_db": 0.6529638416607476, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0515803474682168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09520629908232782, "3000–6000 Hz_40ms_envelope_correlation": 0.9947122821307116, "3000–6000 Hz_40ms_modulation_mae_db": 0.9226877469579385, "mr_stft_db": 5.648829705066068, "band_mae_db": 0.12341279838221834, "rms_gain_error_db": -0.09777217070551919, "output_peak": 0.5579569832469641}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1687550458393212, "全帯域_10ms_rms_envelope_nrmse": 0.07078362902595754, "全帯域_10ms_envelope_correlation": 0.9959884351910041, "全帯域_10ms_modulation_mae_db": 0.246702499128829, "全帯域_40ms_log_rms_mae_db": 0.6537664793702991, "全帯域_40ms_rms_envelope_nrmse": 0.04528967453168105, "全帯域_40ms_envelope_correlation": 0.9982940236977178, "全帯域_40ms_modulation_mae_db": 0.3664689381347435, "300–1000 Hz_10ms_log_rms_mae_db": 1.6412132939093245, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07494891683173836, "300–1000 Hz_10ms_envelope_correlation": 0.9963754021742698, "300–1000 Hz_10ms_modulation_mae_db": 0.6894592465107673, "300–1000 Hz_40ms_log_rms_mae_db": 0.9160878440742197, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039866367906966106, "300–1000 Hz_40ms_envelope_correlation": 0.9990182616749447, "300–1000 Hz_40ms_modulation_mae_db": 0.8123550563281062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5830509691309864, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13270052002383873, "1000–3000 Hz_10ms_envelope_correlation": 0.9892707785396361, "1000–3000 Hz_10ms_modulation_mae_db": 0.5023141030415511, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0154482949100094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08199121325289185, "1000–3000 Hz_40ms_envelope_correlation": 0.9967528803465071, "1000–3000 Hz_40ms_modulation_mae_db": 0.44967839258424097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7179528155834343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2839392292205081, "3000–6000 Hz_10ms_envelope_correlation": 0.9657887631285151, "3000–6000 Hz_10ms_modulation_mae_db": 0.5013252675737532, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1155794498881964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21738637476850464, "3000–6000 Hz_40ms_envelope_correlation": 0.982991281790332, "3000–6000 Hz_40ms_modulation_mae_db": 0.5286637772654462, "mr_stft_db": 5.6345317498361736, "band_mae_db": 0.26518507084524784, "rms_gain_error_db": -0.031144750663707456, "output_peak": 0.4571573692205349}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1726653090626933, "全帯域_10ms_rms_envelope_nrmse": 0.07654157536571893, "全帯域_10ms_envelope_correlation": 0.995309661536819, "全帯域_10ms_modulation_mae_db": 0.1197536622132487, "全帯域_40ms_log_rms_mae_db": 0.6003917995590188, "全帯域_40ms_rms_envelope_nrmse": 0.03647630571811154, "全帯域_40ms_envelope_correlation": 0.9989036976196182, "全帯域_40ms_modulation_mae_db": 0.23462449132968213, "300–1000 Hz_10ms_log_rms_mae_db": 1.6632757925752064, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09492134215730162, "300–1000 Hz_10ms_envelope_correlation": 0.9940914838441108, "300–1000 Hz_10ms_modulation_mae_db": 0.2938124510072016, "300–1000 Hz_40ms_log_rms_mae_db": 1.0957494770672935, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051940348426747385, "300–1000 Hz_40ms_envelope_correlation": 0.9982133365998748, "300–1000 Hz_40ms_modulation_mae_db": 0.36148506198516867, "1000–3000 Hz_10ms_log_rms_mae_db": 1.508168811593022, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12083051280051833, "1000–3000 Hz_10ms_envelope_correlation": 0.9905123696580668, "1000–3000 Hz_10ms_modulation_mae_db": 0.5956844566748064, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9416312031937688, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06637984645965826, "1000–3000 Hz_40ms_envelope_correlation": 0.9973779899934062, "1000–3000 Hz_40ms_modulation_mae_db": 0.5377411230258812, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7082873342722635, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2404871017838656, "3000–6000 Hz_10ms_envelope_correlation": 0.9648577665236514, "3000–6000 Hz_10ms_modulation_mae_db": 1.4385384462533086, "3000–6000 Hz_40ms_log_rms_mae_db": 1.057693777960156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12304367467342398, "3000–6000 Hz_40ms_envelope_correlation": 0.9904538236125422, "3000–6000 Hz_40ms_modulation_mae_db": 1.815672697757308, "mr_stft_db": 5.637582616318416, "band_mae_db": 0.09985540997913606, "rms_gain_error_db": -0.04513710426309121, "output_peak": 0.5290247976581935}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1197169772188416, "全帯域_10ms_rms_envelope_nrmse": 0.05661336689231801, "全帯域_10ms_envelope_correlation": 0.9974762704259551, "全帯域_10ms_modulation_mae_db": 0.1769157614288869, "全帯域_40ms_log_rms_mae_db": 0.5706476953534356, "全帯域_40ms_rms_envelope_nrmse": 0.027394758521591928, "全帯域_40ms_envelope_correlation": 0.9994347217722234, "全帯域_40ms_modulation_mae_db": 0.41801859239937, "300–1000 Hz_10ms_log_rms_mae_db": 1.6666965283631008, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08336339478700373, "300–1000 Hz_10ms_envelope_correlation": 0.9955022282683427, "300–1000 Hz_10ms_modulation_mae_db": 0.6680755997966533, "300–1000 Hz_40ms_log_rms_mae_db": 0.9310508699551232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043746518378930135, "300–1000 Hz_40ms_envelope_correlation": 0.998815947644119, "300–1000 Hz_40ms_modulation_mae_db": 1.0177585595780718, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5435381037829738, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13576474973748734, "1000–3000 Hz_10ms_envelope_correlation": 0.9886502198247891, "1000–3000 Hz_10ms_modulation_mae_db": 0.5474265336304119, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9670821673969248, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07951356109541496, "1000–3000 Hz_40ms_envelope_correlation": 0.9969947745071008, "1000–3000 Hz_40ms_modulation_mae_db": 0.36965506545851734, "3000–6000 Hz_10ms_log_rms_mae_db": 1.578501905859237, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16677412135394576, "3000–6000 Hz_10ms_envelope_correlation": 0.9831506900324685, "3000–6000 Hz_10ms_modulation_mae_db": 0.3727635028344558, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9630994145226426, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08963071269787258, "3000–6000 Hz_40ms_envelope_correlation": 0.9951208145883507, "3000–6000 Hz_40ms_modulation_mae_db": 0.3595628386411615, "mr_stft_db": 5.63671170940144, "band_mae_db": 0.13506469245624517, "rms_gain_error_db": -0.07352069054208574, "output_peak": 0.4912694139574274}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.169634483186168, "全帯域_10ms_rms_envelope_nrmse": 0.060465682263544214, "全帯域_10ms_envelope_correlation": 0.9971167146234093, "全帯域_10ms_modulation_mae_db": 0.6003567972836106, "全帯域_40ms_log_rms_mae_db": 0.5367561255235517, "全帯域_40ms_rms_envelope_nrmse": 0.030756605952439408, "全帯域_40ms_envelope_correlation": 0.9992710405884966, "全帯域_40ms_modulation_mae_db": 0.7619125363164082, "300–1000 Hz_10ms_log_rms_mae_db": 1.6889843608555144, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07916606200238115, "300–1000 Hz_10ms_envelope_correlation": 0.9958967228858175, "300–1000 Hz_10ms_modulation_mae_db": 0.7982153490315618, "300–1000 Hz_40ms_log_rms_mae_db": 0.8978268147172883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04260436703153366, "300–1000 Hz_40ms_envelope_correlation": 0.9988048783060881, "300–1000 Hz_40ms_modulation_mae_db": 0.8704150450928879, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5608537747431614, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15219402348861458, "1000–3000 Hz_10ms_envelope_correlation": 0.9865505541360504, "1000–3000 Hz_10ms_modulation_mae_db": 0.9418178022308702, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9400640476077979, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0990088580064769, "1000–3000 Hz_40ms_envelope_correlation": 0.9959622578133748, "1000–3000 Hz_40ms_modulation_mae_db": 0.769994427517867, "3000–6000 Hz_10ms_log_rms_mae_db": 1.636232315172854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2075592475452819, "3000–6000 Hz_10ms_envelope_correlation": 0.9765435076059292, "3000–6000 Hz_10ms_modulation_mae_db": 0.739268746309692, "3000–6000 Hz_40ms_log_rms_mae_db": 1.01990039005896, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11660975998819476, "3000–6000 Hz_40ms_envelope_correlation": 0.9932703679488698, "3000–6000 Hz_40ms_modulation_mae_db": 0.7504807855055868, "mr_stft_db": 5.638994083590112, "band_mae_db": 0.13451384715104195, "rms_gain_error_db": -0.07484616694570205, "output_peak": 0.45786735304907356}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65760, "output_samples": 65760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.792555893670449, "spectral_convergence": 0.22485594090757458, "spectral_convergence_status": "ok", "amplitude_floor": 1.6529112798953643e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.440955644997781, "spectral_convergence": 0.2191384868479481, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394789070711673e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.689148054168854, "spectral_convergence": 0.22649607673584476, "spectral_convergence_status": "ok", "amplitude_floor": 8.202976886987361e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008081818875357184, "output_power": 0.0008081576184127498, "delta_db": -0.00013041747909860602, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008371388318692298, "output_power": 0.0008298092171487824, "delta_db": -0.03819233291940813, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0012771265272884344, "output_power": 0.00118964119776983, "delta_db": -0.3081793017991781, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003871568194673651, "output_power": 0.0003620095565479341, "delta_db": -0.2916687797393486, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001322399975596578, "output_power": 0.0001376988770529563, "delta_db": 0.17567566010836863, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.960478257270517e-05, "output_power": 1.795351298321478e-05, "delta_db": -0.38212590011706005, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19932873202707702, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14826744697184016, "pitch": null, "output_rms_dbfs": -24.755688267583334, "output_peak": 0.4690082973669654, "reference_rms_dbfs": -24.60742082061149, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6408865309456955}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00020667206862955213, "log_rms_mae_db": 1.1499672663213145, "rms_envelope_nrmse": 0.07042627003476232, "envelope_correlation": 0.9962150642202571, "rms_gain_error_db": -0.14826713806953534, "reference_mean_power": 0.003465660170380573, "output_mean_power": 0.0033493401948143037}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 2.300680394398606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009732960931433549, "output_power": 0.0009266883591066574, "reference_active": true, "delta_db": -0.21311272574433548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.170752162168669e-05, "output_power": 7.331131583407937e-05, "reference_active": true, "delta_db": -0.470910229957473}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.07791133597542e-05, "output_power": 1.8372841103881025e-05, "reference_active": true, "delta_db": -0.5345069327618868}], "active_band_count": 3, "active_band_mae_db": 0.4061766294878984}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00018941444655877122, "log_rms_mae_db": 0.6468180228887991, "rms_envelope_nrmse": 0.04454296664243185, "envelope_correlation": 0.9985783755473815, "rms_gain_error_db": -0.14827214083000254, "reference_mean_power": 0.003491154294575285, "output_mean_power": 0.0033739747582562864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 2.1871865789273095e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008996086963166702, "output_power": 0.0008548897301462895, "reference_active": true, "delta_db": -0.22143545111029647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8292682982878605e-05, "output_power": 3.537542395678079e-05, "reference_active": true, "delta_db": -0.344141429592732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1507764875014336e-06, "output_power": 7.463995904951352e-07, "reference_active": true, "delta_db": -1.8801958739989513}], "active_band_count": 3, "active_band_mae_db": 0.8152575849006599}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.0001875778806757321, "log_rms_mae_db": 1.5924590380145058, "rms_envelope_nrmse": 0.09054496090754321, "envelope_correlation": 0.9950314843445371, "rms_gain_error_db": -0.2502548828482334, "reference_mean_power": 0.0016622862050937872, "output_mean_power": 0.0015692072735974355}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 1.477365390744301e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007590270471846978, "output_power": 0.0007191151858363802, "reference_active": true, "delta_db": -0.23458791812877333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.577918818080543e-05, "output_power": 4.354452397360237e-05, "reference_active": true, "delta_db": -1.0753864195557035}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4662085193890421e-05, "output_power": 9.412948479678979e-06, "reference_active": true, "delta_db": -1.9247005699017652}], "active_band_count": 3, "active_band_mae_db": 1.078224969195414}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00016891518288067464, "log_rms_mae_db": 0.8590994866527286, "rms_envelope_nrmse": 0.05630891115857844, "envelope_correlation": 0.9983718798769187, "rms_gain_error_db": -0.250257523748467, "reference_mean_power": 0.001674521427963224, "output_mean_power": 0.0015807564296949441}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 1.422523294383883e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007062170888212691, "output_power": 0.000667949771013717, "reference_active": true, "delta_db": -0.24194416974640073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.54990695717255e-05, "output_power": 2.0366307465045196e-05, "reference_active": true, "delta_db": -0.9761203795949424}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.502016839919997e-07, "output_power": 4.835624102420127e-07, "reference_active": true, "delta_db": -2.933632621722031}], "active_band_count": 3, "active_band_mae_db": 1.3838990570211245}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00010800393678736649, "log_rms_mae_db": 1.5334422393870342, "rms_envelope_nrmse": 0.1501829855805913, "envelope_correlation": 0.9850507843409985, "rms_gain_error_db": -0.22153571498301214, "reference_mean_power": 0.0004089931043607793, "output_mean_power": 0.00038865334900459376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 4.3909884369057556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019183671841687967, "output_power": 0.00017992490222621844, "reference_active": true, "delta_db": -0.27840461364364394}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4527862162762071e-05, "output_power": 1.540509572557027e-05, "reference_active": true, "delta_db": 0.2546269061575208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.245905579917347e-06, "output_power": 5.5959993485256365e-06, "reference_active": true, "delta_db": -1.6836070232674447}], "active_band_count": 3, "active_band_mae_db": 0.7388795143562031}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 8.815867833009974e-05, "log_rms_mae_db": 0.9483586068251514, "rms_envelope_nrmse": 0.09832274060409507, "envelope_correlation": 0.9936092194285684, "rms_gain_error_db": -0.2215369079395325, "reference_mean_power": 0.0004120037826983822, "output_mean_power": 0.00039151419487530354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 4.2438330741066236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018201492488703638, "output_power": 0.00016921400646924555, "reference_active": true, "delta_db": -0.31670692437465336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.209318154228534e-06, "output_power": 8.376683843231501e-06, "reference_active": true, "delta_db": 0.6517793261895711}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163481758524237e-07, "output_power": 2.659015698232949e-07, "reference_active": true, "delta_db": -0.7544443322863835}], "active_band_count": 3, "active_band_mae_db": 0.574310194283536}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 8.95305280668978e-05, "log_rms_mae_db": 1.6546381057882693, "rms_envelope_nrmse": 0.1557233244440516, "envelope_correlation": 0.9853849980820946, "rms_gain_error_db": -0.06312291858213934, "reference_mean_power": 8.45768795231113e-05, "output_mean_power": 8.335647981752318e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 3.626876619728336e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.450635181337156e-05, "output_power": 1.4150648350100553e-05, "reference_active": true, "delta_db": -0.10781867234115311}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.817293180171653e-06, "output_power": 8.121852159560365e-06, "reference_active": true, "delta_db": 0.16598679785469228}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650085525655336e-06, "output_power": 1.7939414599518167e-06, "reference_active": true, "delta_db": 0.5929155175426619}], "active_band_count": 3, "active_band_mae_db": 0.2889069959128358}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 5.6185612647784654e-05, "log_rms_mae_db": 1.0779675970709321, "rms_envelope_nrmse": 0.08836595672541142, "envelope_correlation": 0.9951624237231788, "rms_gain_error_db": -0.06312403292002568, "reference_mean_power": 8.519942751057583e-05, "output_mean_power": 8.397002322078595e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 3.342409970638096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5109841043230135e-05, "output_power": 1.5167819408958997e-05, "reference_active": true, "delta_db": 0.016632536798499752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.127479817162529e-06, "output_power": 4.445771487940252e-06, "reference_active": true, "delta_db": 0.3226217796166799}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2069317938633786e-07, "output_power": 1.224936469733618e-07, "reference_active": true, "delta_db": 0.06430837039935113}], "active_band_count": 3, "active_band_mae_db": 0.1345208956048436}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8857032051759609, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "37bc16c0e0e2cb983ca04b2d7be4b907cd7be69b91eff6ed8dac6b4a683f6edd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9880776195854813, "standard_deviation": 0.0017597363541774296, "minimum": 0.9850507843409985, "maximum": 0.9905123696580668, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.554307679941462, "standard_deviation": 0.03233355869668882, "minimum": 1.508168811593022, "maximum": 1.614385310829931, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7139341157306779, "standard_deviation": 0.21506668092666192, "minimum": 0.46799138146710334, "maximum": 1.0058134515585528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13807740364013074, "standard_deviation": 0.010436903672270702, "minimum": 0.12083051280051833, "maximum": 0.15219402348861458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963692212501829, "standard_deviation": 0.0012072201312162867, "minimum": 0.9936092194285684, "maximum": 0.9973779899934062, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9774476458116345, "standard_deviation": 0.03499617860045606, "minimum": 0.9400640476077979, "maximum": 1.0240563677220365, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5851368804580138, "standard_deviation": 0.19450938101364812, "minimum": 0.36965506545851734, "maximum": 0.9128652709327953, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08338531459432078, "standard_deviation": 0.010862208847074907, "minimum": 0.06637984645965826, "maximum": 0.0990088580064769, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9744382701401241, "standard_deviation": 0.008689098721430342, "minimum": 0.9635769090948326, "maximum": 0.9853849980820946, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6489665937542015, "standard_deviation": 0.048188282433684615, "minimum": 1.578501905859237, "maximum": 1.7179528155834343, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5764899828539964, "standard_deviation": 0.38770071702113473, "minimum": 0.2653038460809961, "maximum": 1.4385384462533086, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21344216198865473, "standard_deviation": 0.04402677579152595, "minimum": 0.1557233244440516, "maximum": 0.2839392292205081, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9909529403089985, "standard_deviation": 0.004425082531974011, "minimum": 0.982991281790332, "maximum": 0.9951624237231788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0440569108128501, "standard_deviation": 0.047641778564677405, "minimum": 0.9630994145226426, "maximum": 1.1155794498881964, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6674852540928604, "standard_deviation": 0.5237797810683633, "minimum": 0.1345208956048436, "maximum": 1.815672697757308, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13070544491934655, "standard_deviation": 0.044752743474146525, "minimum": 0.08836595672541142, "maximum": 0.21738637476850464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9951611323259548, "standard_deviation": 0.0007640124450866556, "minimum": 0.9940914838441108, "maximum": 0.9963754021742698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6468184575657348, "standard_deviation": 0.029781797632981295, "minimum": 1.5924590380145058, "maximum": 1.6889843608555144, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7586315779996472, "standard_deviation": 0.2528128176248487, "minimum": 0.2938124510072016, "maximum": 1.078224969195414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0863910825701048, "standard_deviation": 0.006920442420401602, "minimum": 0.07494891683173836, "maximum": 0.09492134215730162, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986345191101239, "standard_deviation": 0.0002694468813398323, "minimum": 0.9982133365998748, "maximum": 0.9990182616749447, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.921936933080338, "standard_deviation": 0.08240420880283465, "minimum": 0.8271994416029794, "maximum": 1.0957494770672935, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.905481684332049, "standard_deviation": 0.2933537084670025, "minimum": 0.36148506198516867, "maximum": 1.3838990570211245, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04685138213835799, "standard_deviation": 0.005340740704909144, "minimum": 0.039866367906966106, "maximum": 0.05630891115857844, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17979181334638594, "standard_deviation": 0.0641783039454056, "minimum": 0.09985540997913606, "maximum": 0.26518507084524784, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.638698377380368, "standard_deviation": 0.004592547733857081, "minimum": 5.6345317498361736, "maximum": 5.648829705066068, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4941656146462383, "standard_deviation": 0.03510008395780274, "minimum": 0.4571573692205349, "maximum": 0.5579569832469641, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07823789914640696, "standard_deviation": 0.039694514755281976, "minimum": -0.14826744697184016, "maximum": -0.031144750663707456, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963956208808806, "standard_deviation": 0.0007120461133658917, "minimum": 0.995309661536819, "maximum": 0.9974762704259551, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1548787809442222, "standard_deviation": 0.023376747536659217, "minimum": 1.1197169772188416, "maximum": 1.1852150906370411, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32183633248493193, "standard_deviation": 0.16094936612800134, "minimum": 0.1197536622132487, "maximum": 0.6003567972836106, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06732536124731785, "standard_deviation": 0.0065950302760275515, "minimum": 0.05661336689231801, "maximum": 0.07654157536571893, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989033627734116, "standard_deviation": 0.00036277888161787843, "minimum": 0.9982940236977178, "maximum": 0.9994347217722234, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.607653724798785, "standard_deviation": 0.04037170512316123, "minimum": 0.5367561255235517, "maximum": 0.6537664793702991, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.504743711300205, "standard_deviation": 0.22305687209547817, "minimum": 0.23462449132968213, "maximum": 0.8152575849006599, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037082940966974606, "standard_deviation": 0.006096848228357351, "minimum": 0.027394758521591928, "maximum": 0.04528967453168105, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0049", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.3138678073883057, "predicted_mos": 3.3138678073883057, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "ddddf8954168121a5cf280521676cdb5d2ba9e92de7cc2d98f061419248994c0", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.679866600316018, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1756432133211077, "全帯域_10ms_rms_envelope_nrmse": 0.07212075851634708, "全帯域_10ms_envelope_correlation": 0.9959980950338898, "全帯域_10ms_modulation_mae_db": 0.4388940456291679, "全帯域_40ms_log_rms_mae_db": 0.6241985052116763, "全帯域_40ms_rms_envelope_nrmse": 0.04520178173594781, "全帯域_40ms_envelope_correlation": 0.998505732218071, "全帯域_40ms_modulation_mae_db": 0.8550220491591954, "300–1000 Hz_10ms_log_rms_mae_db": 1.5676308364231326, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0944545322048716, "300–1000 Hz_10ms_envelope_correlation": 0.9944902057288159, "300–1000 Hz_10ms_modulation_mae_db": 1.2636552041559592, "300–1000 Hz_40ms_log_rms_mae_db": 0.77603003605092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057773183993989834, "300–1000 Hz_40ms_envelope_correlation": 0.9981950069689404, "300–1000 Hz_40ms_modulation_mae_db": 1.6075848454195636, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5296974856203165, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1519774955431344, "1000–3000 Hz_10ms_envelope_correlation": 0.9847674933138261, "1000–3000 Hz_10ms_modulation_mae_db": 0.6641516172509446, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8970971667907176, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10013678933825698, "1000–3000 Hz_40ms_envelope_correlation": 0.9934764857463217, "1000–3000 Hz_40ms_modulation_mae_db": 0.5022775428383852, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6356323997543523, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17043423581672953, "3000–6000 Hz_10ms_envelope_correlation": 0.9834639296605545, "3000–6000 Hz_10ms_modulation_mae_db": 0.5665887131996623, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0249991921313926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10601086108057522, "3000–6000 Hz_40ms_envelope_correlation": 0.9936864183405282, "3000–6000 Hz_40ms_modulation_mae_db": 0.5679633504686038, "mr_stft_db": 5.609759221350841, "band_mae_db": 0.2323674519597717, "rms_gain_error_db": -0.14067534542599391, "output_peak": 0.46880427115040585}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1941353533804784, "全帯域_10ms_rms_envelope_nrmse": 0.06834332762685692, "全帯域_10ms_envelope_correlation": 0.9963776218479569, "全帯域_10ms_modulation_mae_db": 0.2519142917853453, "全帯域_40ms_log_rms_mae_db": 0.5818133129690564, "全帯域_40ms_rms_envelope_nrmse": 0.03759556079689382, "全帯域_40ms_envelope_correlation": 0.9989781466067659, "全帯域_40ms_modulation_mae_db": 0.3447806710739692, "300–1000 Hz_10ms_log_rms_mae_db": 1.6272498102922337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.092156940298166, "300–1000 Hz_10ms_envelope_correlation": 0.994542705780317, "300–1000 Hz_10ms_modulation_mae_db": 1.0198730986614148, "300–1000 Hz_40ms_log_rms_mae_db": 0.8467756060220551, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049306171735248955, "300–1000 Hz_40ms_envelope_correlation": 0.9985487658257324, "300–1000 Hz_40ms_modulation_mae_db": 1.1412969872520693, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5935121981435851, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1446805400825925, "1000–3000 Hz_10ms_envelope_correlation": 0.9873736328450297, "1000–3000 Hz_10ms_modulation_mae_db": 1.0031146447432588, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9267023978970991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08345618225783295, "1000–3000 Hz_40ms_envelope_correlation": 0.9970867157078449, "1000–3000 Hz_40ms_modulation_mae_db": 0.671512133787084, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5969997444211201, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23554731164092835, "3000–6000 Hz_10ms_envelope_correlation": 0.9664107471867591, "3000–6000 Hz_10ms_modulation_mae_db": 0.2200295975405496, "3000–6000 Hz_40ms_log_rms_mae_db": 0.954465731238905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14000800239782876, "3000–6000 Hz_40ms_envelope_correlation": 0.9888065584248873, "3000–6000 Hz_40ms_modulation_mae_db": 0.3922671156392963, "mr_stft_db": 5.604359087474904, "band_mae_db": 0.2347125563506247, "rms_gain_error_db": -0.12551473076338301, "output_peak": 0.4971513270170031}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1527011946151768, "全帯域_10ms_rms_envelope_nrmse": 0.07450120808721253, "全帯域_10ms_envelope_correlation": 0.9955609975256334, "全帯域_10ms_modulation_mae_db": 0.25398756385631444, "全帯域_40ms_log_rms_mae_db": 0.6580503759312413, "全帯域_40ms_rms_envelope_nrmse": 0.04093901043895027, "全帯域_40ms_envelope_correlation": 0.9986283333776005, "全帯域_40ms_modulation_mae_db": 0.30241837068225236, "300–1000 Hz_10ms_log_rms_mae_db": 1.6157917186234962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09304557831095397, "300–1000 Hz_10ms_envelope_correlation": 0.994370853158609, "300–1000 Hz_10ms_modulation_mae_db": 0.5975523045207485, "300–1000 Hz_40ms_log_rms_mae_db": 0.8015246103140522, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04466374225704288, "300–1000 Hz_40ms_envelope_correlation": 0.9987774209604919, "300–1000 Hz_40ms_modulation_mae_db": 0.7658192284721914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5201138259287712, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13371113789377756, "1000–3000 Hz_10ms_envelope_correlation": 0.9888114757805482, "1000–3000 Hz_10ms_modulation_mae_db": 0.8671063941174766, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9720791459627123, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07763317750388829, "1000–3000 Hz_40ms_envelope_correlation": 0.9969064716594939, "1000–3000 Hz_40ms_modulation_mae_db": 0.8111857286205838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6008056547274332, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2420575045951906, "3000–6000 Hz_10ms_envelope_correlation": 0.9736109470096256, "3000–6000 Hz_10ms_modulation_mae_db": 0.23247883680752182, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0103646390453713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1837696028916858, "3000–6000 Hz_40ms_envelope_correlation": 0.9868268943567644, "3000–6000 Hz_40ms_modulation_mae_db": 0.4816864568927808, "mr_stft_db": 5.6080644231317045, "band_mae_db": 0.26195280040611885, "rms_gain_error_db": -0.04465676814289481, "output_peak": 0.49953748514862323}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1572789919875186, "全帯域_10ms_rms_envelope_nrmse": 0.06427046727762951, "全帯域_10ms_envelope_correlation": 0.9967589104727583, "全帯域_10ms_modulation_mae_db": 0.423821755971517, "全帯域_40ms_log_rms_mae_db": 0.6357643106409726, "全帯域_40ms_rms_envelope_nrmse": 0.03667280717777515, "全帯域_40ms_envelope_correlation": 0.9989690101781956, "全帯域_40ms_modulation_mae_db": 0.681453605317083, "300–1000 Hz_10ms_log_rms_mae_db": 1.6259107719479982, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0913570197383758, "300–1000 Hz_10ms_envelope_correlation": 0.994536601277729, "300–1000 Hz_10ms_modulation_mae_db": 0.9444022156662143, "300–1000 Hz_40ms_log_rms_mae_db": 0.9147645255469412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050485330156807263, "300–1000 Hz_40ms_envelope_correlation": 0.9983276770485717, "300–1000 Hz_40ms_modulation_mae_db": 1.0295914308651348, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4914731958442946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14923015878297788, "1000–3000 Hz_10ms_envelope_correlation": 0.9857615297278887, "1000–3000 Hz_10ms_modulation_mae_db": 0.4683398133446714, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9016009573401649, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08534262934797293, "1000–3000 Hz_40ms_envelope_correlation": 0.9959626231279081, "1000–3000 Hz_40ms_modulation_mae_db": 0.40100187413240906, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6608140232787547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18001028677326644, "3000–6000 Hz_10ms_envelope_correlation": 0.9814856111825798, "3000–6000 Hz_10ms_modulation_mae_db": 0.44509984521390306, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0199141754408392, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09586120831116669, "3000–6000 Hz_40ms_envelope_correlation": 0.9949023368760841, "3000–6000 Hz_40ms_modulation_mae_db": 0.6762642878369811, "mr_stft_db": 5.624609510162305, "band_mae_db": 0.12646069454639045, "rms_gain_error_db": -0.09642667822479578, "output_peak": 0.5525008193056972}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.196340679619091, "全帯域_10ms_rms_envelope_nrmse": 0.07182673849425392, "全帯域_10ms_envelope_correlation": 0.9958747255808725, "全帯域_10ms_modulation_mae_db": 0.17268420986599056, "全帯域_40ms_log_rms_mae_db": 0.6534092998529921, "全帯域_40ms_rms_envelope_nrmse": 0.04759387908843381, "全帯域_40ms_envelope_correlation": 0.9981268495594141, "全帯域_40ms_modulation_mae_db": 0.2686757963944328, "300–1000 Hz_10ms_log_rms_mae_db": 1.6711408217645074, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0764176394892456, "300–1000 Hz_10ms_envelope_correlation": 0.9962658024526261, "300–1000 Hz_10ms_modulation_mae_db": 0.726079801331705, "300–1000 Hz_40ms_log_rms_mae_db": 0.9159873573774465, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04180013451278473, "300–1000 Hz_40ms_envelope_correlation": 0.9989555760651816, "300–1000 Hz_40ms_modulation_mae_db": 0.8620964132794361, "1000–3000 Hz_10ms_log_rms_mae_db": 1.574945573014348, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13003939171279913, "1000–3000 Hz_10ms_envelope_correlation": 0.9896691476357197, "1000–3000 Hz_10ms_modulation_mae_db": 0.5349548524529396, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9487019306084455, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07983408986028603, "1000–3000 Hz_40ms_envelope_correlation": 0.9968932936958875, "1000–3000 Hz_40ms_modulation_mae_db": 0.43169787578876745, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6966810332982938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3093492332658391, "3000–6000 Hz_10ms_envelope_correlation": 0.9622455108222768, "3000–6000 Hz_10ms_modulation_mae_db": 0.7722144477200225, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0787353349249076, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24661750451455458, "3000–6000 Hz_40ms_envelope_correlation": 0.9797005208672535, "3000–6000 Hz_40ms_modulation_mae_db": 0.881796507699203, "mr_stft_db": 5.601602979711044, "band_mae_db": 0.28286181724633547, "rms_gain_error_db": -0.04726894618348492, "output_peak": 0.4545607122428992}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.196118703449998, "全帯域_10ms_rms_envelope_nrmse": 0.07797993335339189, "全帯域_10ms_envelope_correlation": 0.9951287619320202, "全帯域_10ms_modulation_mae_db": 0.14933124761056168, "全帯域_40ms_log_rms_mae_db": 0.5928978201839947, "全帯域_40ms_rms_envelope_nrmse": 0.036147437429619886, "全帯域_40ms_envelope_correlation": 0.9989178388267868, "全帯域_40ms_modulation_mae_db": 0.26818461671675103, "300–1000 Hz_10ms_log_rms_mae_db": 1.6042935247224703, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09378105497297365, "300–1000 Hz_10ms_envelope_correlation": 0.9942288202804994, "300–1000 Hz_10ms_modulation_mae_db": 0.22687850516678446, "300–1000 Hz_40ms_log_rms_mae_db": 0.9905870364384346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05084546134622919, "300–1000 Hz_40ms_envelope_correlation": 0.9982753331340148, "300–1000 Hz_40ms_modulation_mae_db": 0.21700369957070742, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4922459876895091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13020251874258787, "1000–3000 Hz_10ms_envelope_correlation": 0.9888642145047966, "1000–3000 Hz_10ms_modulation_mae_db": 0.5336121449480881, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8547091054235926, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07130361148493836, "1000–3000 Hz_40ms_envelope_correlation": 0.9968330227441115, "1000–3000 Hz_40ms_modulation_mae_db": 0.4528135667967372, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6694832099668766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2556291253127143, "3000–6000 Hz_10ms_envelope_correlation": 0.9609117030062952, "3000–6000 Hz_10ms_modulation_mae_db": 1.4730964156501036, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9901819387741201, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12172885488554025, "3000–6000 Hz_40ms_envelope_correlation": 0.9908146230282672, "3000–6000 Hz_40ms_modulation_mae_db": 1.6864273958290978, "mr_stft_db": 5.60589644496984, "band_mae_db": 0.07735871589701897, "rms_gain_error_db": -0.03745679373678127, "output_peak": 0.5326227786155173}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.168456171614968, "全帯域_10ms_rms_envelope_nrmse": 0.05822267173767104, "全帯域_10ms_envelope_correlation": 0.9973328965710715, "全帯域_10ms_modulation_mae_db": 0.1557826269453627, "全帯域_40ms_log_rms_mae_db": 0.6024769884341118, "全帯域_40ms_rms_envelope_nrmse": 0.02920102696235501, "全帯域_40ms_envelope_correlation": 0.9993547850310714, "全帯域_40ms_modulation_mae_db": 0.4125171440941902, "300–1000 Hz_10ms_log_rms_mae_db": 1.637810159197062, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08415229415602209, "300–1000 Hz_10ms_envelope_correlation": 0.9954169074729207, "300–1000 Hz_10ms_modulation_mae_db": 0.7401827065756158, "300–1000 Hz_40ms_log_rms_mae_db": 0.8515880497568518, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04510586697553922, "300–1000 Hz_40ms_envelope_correlation": 0.9987365667684555, "300–1000 Hz_40ms_modulation_mae_db": 1.1257759133653022, "1000–3000 Hz_10ms_log_rms_mae_db": 1.551386982888433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13672848234729634, "1000–3000 Hz_10ms_envelope_correlation": 0.9883992657603401, "1000–3000 Hz_10ms_modulation_mae_db": 0.583846670117667, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9004051294959283, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07922631889867283, "1000–3000 Hz_40ms_envelope_correlation": 0.9969368384280618, "1000–3000 Hz_40ms_modulation_mae_db": 0.38561426433193896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5690466087230068, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18426551496966306, "3000–6000 Hz_10ms_envelope_correlation": 0.9795309520467209, "3000–6000 Hz_10ms_modulation_mae_db": 0.3481083675820269, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9233282135684803, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10530405077226823, "3000–6000 Hz_40ms_envelope_correlation": 0.9935396788470198, "3000–6000 Hz_40ms_modulation_mae_db": 0.33317453913146683, "mr_stft_db": 5.606249109680309, "band_mae_db": 0.15534732317814107, "rms_gain_error_db": -0.07948511677746357, "output_peak": 0.49386014644267706}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.19058847575248, "全帯域_10ms_rms_envelope_nrmse": 0.0626917255682155, "全帯域_10ms_envelope_correlation": 0.9969200600074446, "全帯域_10ms_modulation_mae_db": 0.7099919594923468, "全帯域_40ms_log_rms_mae_db": 0.5582669569889651, "全帯域_40ms_rms_envelope_nrmse": 0.0320378891209369, "全帯域_40ms_envelope_correlation": 0.9992320173375906, "全帯域_40ms_modulation_mae_db": 0.8471800812443863, "300–1000 Hz_10ms_log_rms_mae_db": 1.6486825696416179, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07767095037541434, "300–1000 Hz_10ms_envelope_correlation": 0.9960731140852525, "300–1000 Hz_10ms_modulation_mae_db": 0.9024181312060281, "300–1000 Hz_40ms_log_rms_mae_db": 0.8138955800668066, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039340332303057474, "300–1000 Hz_40ms_envelope_correlation": 0.9990174327638393, "300–1000 Hz_40ms_modulation_mae_db": 0.9673701557070545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5759908017272453, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1587031169282983, "1000–3000 Hz_10ms_envelope_correlation": 0.9853795388767228, "1000–3000 Hz_10ms_modulation_mae_db": 1.0649208115702675, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9185891228094955, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10153569666755732, "1000–3000 Hz_40ms_envelope_correlation": 0.9959135670572408, "1000–3000 Hz_40ms_modulation_mae_db": 0.8376626985611489, "3000–6000 Hz_10ms_log_rms_mae_db": 1.658647632626501, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22020982354418006, "3000–6000 Hz_10ms_envelope_correlation": 0.9748245986299486, "3000–6000 Hz_10ms_modulation_mae_db": 0.8577002670215078, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0006105941477363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12378391106941788, "3000–6000 Hz_40ms_envelope_correlation": 0.9934074814525711, "3000–6000 Hz_40ms_modulation_mae_db": 0.9232421700980532, "mr_stft_db": 5.614030024120176, "band_mae_db": 0.14737233687659193, "rms_gain_error_db": -0.08791243489337172, "output_peak": 0.4589933489538027}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65760, "output_samples": 65760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.766593101552539, "spectral_convergence": 0.22595469296781043, "spectral_convergence_status": "ok", "amplitude_floor": 1.6529112798953643e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.416180164699007, "spectral_convergence": 0.22020357985891933, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394789070711673e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6465043978009755, "spectral_convergence": 0.22633580409240067, "spectral_convergence_status": "ok", "amplitude_floor": 8.202976886987361e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008081818875357184, "output_power": 0.000806145831969023, "delta_db": -0.010955001957120064, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008371388318692298, "output_power": 0.0008302699750845088, "delta_db": -0.035781548722441485, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0012771265272884344, "output_power": 0.0011951821742654687, "delta_db": -0.28799818587370235, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003871568194673651, "output_power": 0.0003595000635730307, "delta_db": -0.3218794190930091, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001322399975596578, "output_power": 0.00014223248888104397, "delta_db": 0.3163597713702553, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.960478257270517e-05, "output_power": 1.779258100099661e-05, "delta_db": -0.42123078474210196, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2323674519597717, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14067534542599391, "pitch": null, "output_rms_dbfs": -24.748096166037485, "output_peak": 0.46880427115040585, "reference_rms_dbfs": -24.60742082061149, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.609759221350841}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00020667206862955213, "log_rms_mae_db": 1.1756432133211077, "rms_envelope_nrmse": 0.07212075851634708, "envelope_correlation": 0.9959980950338898, "rms_gain_error_db": -0.1406737042871597, "reference_mean_power": 0.003465660170380573, "output_mean_power": 0.003355201480463096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 2.300680394398606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009732960931433549, "output_power": 0.0009250840193948, "reference_active": true, "delta_db": -0.22063801412620215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.170752162168669e-05, "output_power": 7.306536463433436e-05, "reference_active": true, "delta_db": -0.4855048155223663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.07791133597542e-05, "output_power": 1.805398518073234e-05, "reference_active": true, "delta_db": -0.6105393072389351}], "active_band_count": 3, "active_band_mae_db": 0.4388940456291679}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00018941444655877122, "log_rms_mae_db": 0.6241985052116763, "rms_envelope_nrmse": 0.04520178173594781, "envelope_correlation": 0.998505732218071, "rms_gain_error_db": -0.14067694068634362, "reference_mean_power": 0.003491154294575285, "output_mean_power": 0.0033798805286158326}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 2.1871865789273095e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008996086963166702, "output_power": 0.0008533792107771723, "reference_active": true, "delta_db": -0.22911586233916043}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8292682982878605e-05, "output_power": 3.5356367774485065e-05, "reference_active": true, "delta_db": -0.34648153573590845}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1507764875014336e-06, "output_power": 7.278537182203559e-07, "reference_active": true, "delta_db": -1.9894687494025178}], "active_band_count": 3, "active_band_mae_db": 0.8550220491591954}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.0001875778806757321, "log_rms_mae_db": 1.5676308364231326, "rms_envelope_nrmse": 0.0944545322048716, "envelope_correlation": 0.9944902057288159, "rms_gain_error_db": -0.23782563054935354, "reference_mean_power": 0.0016622862050937872, "output_mean_power": 0.0015737046849911723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 1.477365390744301e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007590270471846978, "output_power": 0.0007212744190855211, "reference_active": true, "delta_db": -0.22156722172179444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.577918818080543e-05, "output_power": 4.2237736737545545e-05, "reference_active": true, "delta_db": -1.2077155022566755}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4662085193890421e-05, "output_power": 8.511918364922451e-06, "reference_active": true, "delta_db": -2.361682888489408}], "active_band_count": 3, "active_band_mae_db": 1.2636552041559592}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00016891518288067464, "log_rms_mae_db": 0.77603003605092, "rms_envelope_nrmse": 0.057773183993989834, "envelope_correlation": 0.9981950069689404, "rms_gain_error_db": -0.23782751385561107, "reference_mean_power": 0.001674521427963224, "output_mean_power": 0.0015852872179776636}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 1.422523294383883e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007062170888212691, "output_power": 0.0006697679223379223, "reference_active": true, "delta_db": -0.23013878543396088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.54990695717255e-05, "output_power": 1.9828213601569657e-05, "reference_active": true, "delta_db": -1.0924074513315971}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.502016839919997e-07, "output_power": 4.2441914468408816e-07, "reference_active": true, "delta_db": -3.5002082994931327}], "active_band_count": 3, "active_band_mae_db": 1.6075848454195636}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00010800393678736649, "log_rms_mae_db": 1.5296974856203165, "rms_envelope_nrmse": 0.1519774955431344, "envelope_correlation": 0.9847674933138261, "rms_gain_error_db": -0.25620354749236235, "reference_mean_power": 0.0004089931043607793, "output_mean_power": 0.0003855632488611935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 4.3909884369057556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019183671841687967, "output_power": 0.0001804467244409261, "reference_active": true, "delta_db": -0.2658273378537136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4527862162762071e-05, "output_power": 1.5497806896365548e-05, "reference_active": true, "delta_db": 0.2806853455078213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.245905579917347e-06, "output_power": 5.910771093508172e-06, "reference_active": true, "delta_db": -1.4459421683912992}], "active_band_count": 3, "active_band_mae_db": 0.6641516172509446}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 8.815867833009974e-05, "log_rms_mae_db": 0.8970971667907176, "rms_envelope_nrmse": 0.10013678933825698, "envelope_correlation": 0.9934764857463217, "rms_gain_error_db": -0.25620434107829204, "reference_mean_power": 0.0004120037826983822, "output_mean_power": 0.0003884013844714495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 4.2438330741066236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018201492488703638, "output_power": 0.00016974621481164964, "reference_active": true, "delta_db": -0.3030690210600242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.209318154228534e-06, "output_power": 8.420061786959337e-06, "reference_active": true, "delta_db": 0.674210866076519}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163481758524237e-07, "output_power": 2.800335463872722e-07, "reference_active": true, "delta_db": -0.5295527413786124}], "active_band_count": 3, "active_band_mae_db": 0.5022775428383852}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 8.95305280668978e-05, "log_rms_mae_db": 1.6356323997543523, "rms_envelope_nrmse": 0.17043423581672953, "envelope_correlation": 0.9834639296605545, "rms_gain_error_db": 0.18764555136778555, "reference_mean_power": 8.45768795231113e-05, "output_mean_power": 8.831128677669033e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 3.626876619728336e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.450635181337156e-05, "output_power": 1.5037265634894452e-05, "reference_active": true, "delta_db": 0.15610665729238493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.817293180171653e-06, "output_power": 8.68998554097324e-06, "reference_active": true, "delta_db": 0.4596265377319935}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650085525655336e-06, "output_power": 2.0087228030246387e-06, "reference_active": true, "delta_db": 1.0840329445746084}], "active_band_count": 3, "active_band_mae_db": 0.5665887131996623}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 5.6185612647784654e-05, "log_rms_mae_db": 1.0249991921313926, "rms_envelope_nrmse": 0.10601086108057522, "envelope_correlation": 0.9936864183405282, "rms_gain_error_db": 0.18764529875650504, "reference_mean_power": 8.519942751057583e-05, "output_mean_power": 8.896131757138144e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 3.342409970638096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5109841043230135e-05, "output_power": 1.6195691455573174e-05, "reference_active": true, "delta_db": 0.30139598873321377}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.127479817162529e-06, "output_power": 4.7607118453279005e-06, "reference_active": true, "delta_db": 0.6198693715149219}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2069317938633786e-07, "output_power": 1.445257384455367e-07, "reference_active": true, "delta_db": 0.7826246911576757}], "active_band_count": 3, "active_band_mae_db": 0.5679633504686038}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8873709442151496, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ab16558b5b465b104450a64cef20291a7a7c5ccbefeac607758d8e794c3e7947", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.987378287305609, "standard_deviation": 0.0018507764463947874, "minimum": 0.9847674933138261, "maximum": 0.9896691476357197, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.541170756357063, "standard_deviation": 0.03897926625271709, "minimum": 1.4914731958442946, "maximum": 1.5935121981435851, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7150058685681642, "standard_deviation": 0.23131203060098757, "minimum": 0.4683398133446714, "maximum": 1.0649208115702675, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14190910525418302, "standard_deviation": 0.010801255064337777, "minimum": 0.13003939171279913, "maximum": 0.1587031169282983, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962511272708587, "standard_deviation": 0.0012097627442768642, "minimum": 0.9934764857463217, "maximum": 0.9970867157078449, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9149856195410195, "standard_deviation": 0.03564051324435141, "minimum": 0.8547091054235926, "maximum": 0.9720791459627123, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5617207106071318, "standard_deviation": 0.18500005833641445, "minimum": 0.38561426433193896, "maximum": 0.8376626985611489, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08480856191992572, "standard_deviation": 0.010734739258337794, "minimum": 0.07130361148493836, "maximum": 0.10153569666755732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9728104999430951, "standard_deviation": 0.008724121767298402, "minimum": 0.9609117030062952, "maximum": 0.9834639296605545, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6360137883495423, "standard_deviation": 0.0433828739309531, "minimum": 1.5690466087230068, "maximum": 1.6966810332982938, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6144145613419122, "standard_deviation": 0.41828073905968954, "minimum": 0.2200295975405496, "maximum": 1.4730964156501036, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22468787948981395, "standard_deviation": 0.04650361328574727, "minimum": 0.17043423581672953, "maximum": 0.3093492332658391, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9902105640241718, "standard_deviation": 0.005065907461394024, "minimum": 0.9797005208672535, "maximum": 0.9949023368760841, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.000324977408969, "standard_deviation": 0.04686769742601411, "minimum": 0.9233282135684803, "maximum": 1.0787353349249076, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7428527279494354, "standard_deviation": 0.4368970955068063, "minimum": 0.33317453913146683, "maximum": 1.6864273958290978, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14038549949037968, "standard_deviation": 0.050970553494057935, "minimum": 0.09586120831116669, "maximum": 0.24661750451455458, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949906262795962, "standard_deviation": 0.0008107517748393581, "minimum": 0.9942288202804994, "maximum": 0.9962658024526261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.624813776576565, "standard_deviation": 0.03085524944689358, "minimum": 1.5676308364231326, "maximum": 1.6711408217645074, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8026302459105588, "standard_deviation": 0.31009263072393195, "minimum": 0.22687850516678446, "maximum": 1.2636552041559592, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08787950119325288, "standard_deviation": 0.007412520286199139, "minimum": 0.0764176394892456, "maximum": 0.0944545322048716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986042224419034, "standard_deviation": 0.00031552295758962044, "minimum": 0.9981950069689404, "maximum": 0.9990174327638393, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8638941001966884, "standard_deviation": 0.07162101495712028, "minimum": 0.77603003605092, "maximum": 0.9905870364384346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9645673342414325, "standard_deviation": 0.3934792737216787, "minimum": 0.21700369957070742, "maximum": 1.6075848454195636, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04741502791008745, "standard_deviation": 0.005876004301363517, "minimum": 0.039340332303057474, "maximum": 0.057773183993989834, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18980421205762416, "standard_deviation": 0.07303390357750437, "minimum": 0.07735871589701897, "maximum": 0.28286181724633547, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.60932135007514, "standard_deviation": 0.007201557112611024, "minimum": 5.601602979711044, "maximum": 5.624609510162305, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4947538611095782, "standard_deviation": 0.034570591106392164, "minimum": 0.4545607122428992, "maximum": 0.5525008193056972, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08242460176852112, "standard_deviation": 0.03811478250629416, "minimum": -0.14067534542599391, "maximum": -0.03745679373678127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9962440086214559, "standard_deviation": 0.0007399634784435626, "minimum": 0.9951287619320202, "maximum": 0.9973328965710715, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1789078479676025, "standard_deviation": 0.01790020997989285, "minimum": 1.1527011946151768, "maximum": 1.196340679619091, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3195509626445758, "standard_deviation": 0.19412708744459267, "minimum": 0.14933124761056168, "maximum": 0.7099919594923468, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06874460383269729, "standard_deviation": 0.0066209094026054216, "minimum": 0.05822267173767104, "maximum": 0.07797993335339189, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998839089141937, "standard_deviation": 0.0004009956960694686, "minimum": 0.9981268495594141, "maximum": 0.9993547850310714, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6133596962766263, "standard_deviation": 0.035432390435433773, "minimum": 0.5582669569889651, "maximum": 0.6580503759312413, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.49752904183528257, "standard_deviation": 0.2556463180435747, "minimum": 0.26818461671675103, "maximum": 0.8550220491591954, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03817367409386408, "standard_deviation": 0.006217835516295445, "minimum": 0.02920102696235501, "maximum": 0.04759387908843381, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0049", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.0257577896118164, "predicted_mos": 3.0257577896118164, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "df35d89ed76801f97d86999b9ea67d765c5bdfefeec23ab6d5d6381fff3e53df", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6513347001746297, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1252314255699405, "全帯域_10ms_rms_envelope_nrmse": 0.0650828243063788, "全帯域_10ms_envelope_correlation": 0.9966128048864085, "全帯域_10ms_modulation_mae_db": 0.3498252096902423, "全帯域_40ms_log_rms_mae_db": 0.5422248170788184, "全帯域_40ms_rms_envelope_nrmse": 0.03246249705077503, "全帯域_40ms_envelope_correlation": 0.9991316428902525, "全帯域_40ms_modulation_mae_db": 0.509615653409007, "300–1000 Hz_10ms_log_rms_mae_db": 1.8560384747247483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08009039866589882, "300–1000 Hz_10ms_envelope_correlation": 0.995900547253063, "300–1000 Hz_10ms_modulation_mae_db": 0.4499194508192071, "300–1000 Hz_40ms_log_rms_mae_db": 1.0341508873831005, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04238126947536929, "300–1000 Hz_40ms_envelope_correlation": 0.9989411726224818, "300–1000 Hz_40ms_modulation_mae_db": 0.6252907855860185, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7314637239866026, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1424279503974409, "1000–3000 Hz_10ms_envelope_correlation": 0.9865495012103304, "1000–3000 Hz_10ms_modulation_mae_db": 0.7634750840663242, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1208560244813452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07938675245723582, "1000–3000 Hz_40ms_envelope_correlation": 0.9959078872039023, "1000–3000 Hz_40ms_modulation_mae_db": 0.6494526206549973, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0503851140021587, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2438544683112232, "3000–6000 Hz_10ms_envelope_correlation": 0.9653355499671378, "3000–6000 Hz_10ms_modulation_mae_db": 1.4026819322485098, "3000–6000 Hz_40ms_log_rms_mae_db": 1.31840504682015, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1395389004055832, "3000–6000 Hz_40ms_envelope_correlation": 0.9884569649622943, "3000–6000 Hz_40ms_modulation_mae_db": 1.6623260604373413, "mr_stft_db": 5.833249913301414, "band_mae_db": 0.11060455247724367, "rms_gain_error_db": -0.03935576225995302, "output_peak": 0.46265975913479407}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0818823596160112, "全帯域_10ms_rms_envelope_nrmse": 0.06168892522599833, "全帯域_10ms_envelope_correlation": 0.9970149945496461, "全帯域_10ms_modulation_mae_db": 0.08357916362176969, "全帯域_40ms_log_rms_mae_db": 0.5004421452749234, "全帯域_40ms_rms_envelope_nrmse": 0.03354055535972375, "全帯域_40ms_envelope_correlation": 0.9991476368784438, "全帯域_40ms_modulation_mae_db": 0.2837360351080442, "300–1000 Hz_10ms_log_rms_mae_db": 1.8473009273491034, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07734735179665458, "300–1000 Hz_10ms_envelope_correlation": 0.9961221759853833, "300–1000 Hz_10ms_modulation_mae_db": 0.5879214362946703, "300–1000 Hz_40ms_log_rms_mae_db": 1.1193807990046272, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04268725781563217, "300–1000 Hz_40ms_envelope_correlation": 0.9988492023822886, "300–1000 Hz_40ms_modulation_mae_db": 0.7464213598367696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7807069519934442, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15907029315337204, "1000–3000 Hz_10ms_envelope_correlation": 0.98501677944396, "1000–3000 Hz_10ms_modulation_mae_db": 0.8333886686985771, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1297174413870275, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11317016766515747, "1000–3000 Hz_40ms_envelope_correlation": 0.9936832673169911, "1000–3000 Hz_40ms_modulation_mae_db": 0.6455231975445922, "3000–6000 Hz_10ms_log_rms_mae_db": 2.060149016494125, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24657572690234034, "3000–6000 Hz_10ms_envelope_correlation": 0.9636279347440565, "3000–6000 Hz_10ms_modulation_mae_db": 0.9613108723429454, "3000–6000 Hz_40ms_log_rms_mae_db": 1.344674625080607, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12826122059985062, "3000–6000 Hz_40ms_envelope_correlation": 0.9897943639577517, "3000–6000 Hz_40ms_modulation_mae_db": 0.8467320829507584, "mr_stft_db": 5.844654543400487, "band_mae_db": 0.18691287213953353, "rms_gain_error_db": -0.09237949686055749, "output_peak": 0.4527500086615875}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1108015292909865, "全帯域_10ms_rms_envelope_nrmse": 0.0647663984822856, "全帯域_10ms_envelope_correlation": 0.9967235436725802, "全帯域_10ms_modulation_mae_db": 0.34364475460637306, "全帯域_40ms_log_rms_mae_db": 0.5541770894926449, "全帯域_40ms_rms_envelope_nrmse": 0.03504799476501178, "全帯域_40ms_envelope_correlation": 0.9990921274720489, "全帯域_40ms_modulation_mae_db": 0.6031552019011516, "300–1000 Hz_10ms_log_rms_mae_db": 1.8041471174881425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08279474183037383, "300–1000 Hz_10ms_envelope_correlation": 0.9955206986391995, "300–1000 Hz_10ms_modulation_mae_db": 0.7651398051918251, "300–1000 Hz_40ms_log_rms_mae_db": 1.12835381464099, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04271257475379825, "300–1000 Hz_40ms_envelope_correlation": 0.9988140137404288, "300–1000 Hz_40ms_modulation_mae_db": 1.1429965370376045, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7620591739248708, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15245911833185405, "1000–3000 Hz_10ms_envelope_correlation": 0.9877908015098593, "1000–3000 Hz_10ms_modulation_mae_db": 1.2746563937846311, "1000–3000 Hz_40ms_log_rms_mae_db": 1.100540726226292, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10565110677448247, "1000–3000 Hz_40ms_envelope_correlation": 0.9964092255621555, "1000–3000 Hz_40ms_modulation_mae_db": 0.9571796543874295, "3000–6000 Hz_10ms_log_rms_mae_db": 2.059508360820887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19681954467012747, "3000–6000 Hz_10ms_envelope_correlation": 0.9764369752444578, "3000–6000 Hz_10ms_modulation_mae_db": 0.44253474967613843, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2959025109396463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1000605069045005, "3000–6000 Hz_40ms_envelope_correlation": 0.9938636893264231, "3000–6000 Hz_40ms_modulation_mae_db": 0.8178863423022001, "mr_stft_db": 5.846572074111194, "band_mae_db": 0.218258352381951, "rms_gain_error_db": -0.10306708995300944, "output_peak": 0.4836602852424764}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.061891343296375, "全帯域_10ms_rms_envelope_nrmse": 0.05231921131451748, "全帯域_10ms_envelope_correlation": 0.997840539919872, "全帯域_10ms_modulation_mae_db": 0.3348339698822287, "全帯域_40ms_log_rms_mae_db": 0.4689343585570472, "全帯域_40ms_rms_envelope_nrmse": 0.02551763494857327, "全帯域_40ms_envelope_correlation": 0.9994988933295942, "全帯域_40ms_modulation_mae_db": 0.40328829474327166, "300–1000 Hz_10ms_log_rms_mae_db": 1.7053921619946337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07628534011709398, "300–1000 Hz_10ms_envelope_correlation": 0.9962159582084457, "300–1000 Hz_10ms_modulation_mae_db": 0.6709219458494807, "300–1000 Hz_40ms_log_rms_mae_db": 0.9586537351846136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03683195970409041, "300–1000 Hz_40ms_envelope_correlation": 0.9991436022362368, "300–1000 Hz_40ms_modulation_mae_db": 0.7541594211258041, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7235138378791812, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12802235534524692, "1000–3000 Hz_10ms_envelope_correlation": 0.9891729525104255, "1000–3000 Hz_10ms_modulation_mae_db": 0.7345931752345903, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1556409695104428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06909997408028211, "1000–3000 Hz_40ms_envelope_correlation": 0.9969256060665617, "1000–3000 Hz_40ms_modulation_mae_db": 0.4046774910622499, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0840982508664845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16453031259552328, "3000–6000 Hz_10ms_envelope_correlation": 0.9837258006141267, "3000–6000 Hz_10ms_modulation_mae_db": 1.0961606253876166, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3429320005240535, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08664498900223805, "3000–6000 Hz_40ms_envelope_correlation": 0.9956026508571787, "3000–6000 Hz_40ms_modulation_mae_db": 1.279062836117435, "mr_stft_db": 5.83996179046618, "band_mae_db": 0.13513850726657928, "rms_gain_error_db": -0.06404864746454698, "output_peak": 0.5508964565246081}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1790780267327134, "全帯域_10ms_rms_envelope_nrmse": 0.059150050866500624, "全帯域_10ms_envelope_correlation": 0.997202212182217, "全帯域_10ms_modulation_mae_db": 0.34525426643230356, "全帯域_40ms_log_rms_mae_db": 0.6279730155704276, "全帯域_40ms_rms_envelope_nrmse": 0.03612482273697658, "全帯域_40ms_envelope_correlation": 0.9989162246200162, "全帯域_40ms_modulation_mae_db": 0.3186097406695747, "300–1000 Hz_10ms_log_rms_mae_db": 1.824158004977616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08269626442453841, "300–1000 Hz_10ms_envelope_correlation": 0.9956126380944933, "300–1000 Hz_10ms_modulation_mae_db": 0.5215912534359587, "300–1000 Hz_40ms_log_rms_mae_db": 0.998983599385722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04779763598379367, "300–1000 Hz_40ms_envelope_correlation": 0.9986059054001881, "300–1000 Hz_40ms_modulation_mae_db": 0.46535472973147757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7313116018180563, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1318330856504049, "1000–3000 Hz_10ms_envelope_correlation": 0.9884322657718397, "1000–3000 Hz_10ms_modulation_mae_db": 0.6712597020551767, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0840338103789262, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0811276855358254, "1000–3000 Hz_40ms_envelope_correlation": 0.9954582759459746, "1000–3000 Hz_40ms_modulation_mae_db": 0.5304421015330745, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0226977956064953, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18458434494012843, "3000–6000 Hz_10ms_envelope_correlation": 0.9805189734909304, "3000–6000 Hz_10ms_modulation_mae_db": 0.7029108408455831, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3215265587040854, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11196350501298775, "3000–6000 Hz_40ms_envelope_correlation": 0.9928777696119534, "3000–6000 Hz_40ms_modulation_mae_db": 1.131163071326921, "mr_stft_db": 5.847503950985959, "band_mae_db": 0.1426652975161571, "rms_gain_error_db": -0.03213792975587998, "output_peak": 0.49710415310403194}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0946988577074004, "全帯域_10ms_rms_envelope_nrmse": 0.07023631555913311, "全帯域_10ms_envelope_correlation": 0.9960619498379132, "全帯域_10ms_modulation_mae_db": 0.30267909936017867, "全帯域_40ms_log_rms_mae_db": 0.5820528727949237, "全帯域_40ms_rms_envelope_nrmse": 0.04153094596889747, "全帯域_40ms_envelope_correlation": 0.9985809428813694, "全帯域_40ms_modulation_mae_db": 0.149453049669718, "300–1000 Hz_10ms_log_rms_mae_db": 1.7989142371121725, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08567680068678325, "300–1000 Hz_10ms_envelope_correlation": 0.9952849580973426, "300–1000 Hz_10ms_modulation_mae_db": 0.4264112414916917, "300–1000 Hz_40ms_log_rms_mae_db": 1.07383787738172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043537737196282636, "300–1000 Hz_40ms_envelope_correlation": 0.9988874856812905, "300–1000 Hz_40ms_modulation_mae_db": 0.558186661312332, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8282362054918357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13416050968888538, "1000–3000 Hz_10ms_envelope_correlation": 0.9884237538140701, "1000–3000 Hz_10ms_modulation_mae_db": 0.5898397540876085, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0817703296488603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07218390984785027, "1000–3000 Hz_40ms_envelope_correlation": 0.9971415172781655, "1000–3000 Hz_40ms_modulation_mae_db": 0.5395015072483423, "3000–6000 Hz_10ms_log_rms_mae_db": 1.985890473838433, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25395696980333504, "3000–6000 Hz_10ms_envelope_correlation": 0.9604631609094564, "3000–6000 Hz_10ms_modulation_mae_db": 0.8873815076577013, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3380682018803565, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17282840767167956, "3000–6000 Hz_40ms_envelope_correlation": 0.9811237826565439, "3000–6000 Hz_40ms_modulation_mae_db": 1.1089211043366596, "mr_stft_db": 5.863156632841814, "band_mae_db": 0.1537589836466423, "rms_gain_error_db": -0.040860771839342995, "output_peak": 0.48963871584265406}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1113887015276651, "全帯域_10ms_rms_envelope_nrmse": 0.06722845489702932, "全帯域_10ms_envelope_correlation": 0.9966354717339356, "全帯域_10ms_modulation_mae_db": 0.48620716517094315, "全帯域_40ms_log_rms_mae_db": 0.5588513589809201, "全帯域_40ms_rms_envelope_nrmse": 0.03832296972633278, "全帯域_40ms_envelope_correlation": 0.9990846609032542, "全帯域_40ms_modulation_mae_db": 0.3592822007906009, "300–1000 Hz_10ms_log_rms_mae_db": 1.7432192012330139, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08791369436122518, "300–1000 Hz_10ms_envelope_correlation": 0.9953164495474633, "300–1000 Hz_10ms_modulation_mae_db": 0.6774049194786982, "300–1000 Hz_40ms_log_rms_mae_db": 1.000512629470653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05346892347300149, "300–1000 Hz_40ms_envelope_correlation": 0.9985569381509536, "300–1000 Hz_40ms_modulation_mae_db": 0.5689634661779278, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7848635506969337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1504812258497988, "1000–3000 Hz_10ms_envelope_correlation": 0.9869522445756671, "1000–3000 Hz_10ms_modulation_mae_db": 0.4891945297190949, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1362286831314286, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10091648881060879, "1000–3000 Hz_40ms_envelope_correlation": 0.995677232784037, "1000–3000 Hz_40ms_modulation_mae_db": 0.4965484734999097, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0749311890264472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22012586681547963, "3000–6000 Hz_10ms_envelope_correlation": 0.9708427730516377, "3000–6000 Hz_10ms_modulation_mae_db": 1.5624292989890425, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2846766283045739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13553402627193079, "3000–6000 Hz_40ms_envelope_correlation": 0.9894758621773735, "3000–6000 Hz_40ms_modulation_mae_db": 2.0464735340326796, "mr_stft_db": 5.828099035075188, "band_mae_db": 0.2568191156841407, "rms_gain_error_db": -0.16746143330243893, "output_peak": 0.4562695019639601}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0810812975165098, "全帯域_10ms_rms_envelope_nrmse": 0.06572594921776852, "全帯域_10ms_envelope_correlation": 0.9965749375629402, "全帯域_10ms_modulation_mae_db": 0.19827452558676475, "全帯域_40ms_log_rms_mae_db": 0.5333759998067316, "全帯域_40ms_rms_envelope_nrmse": 0.03082841307485073, "全帯域_40ms_envelope_correlation": 0.9992620729990498, "全帯域_40ms_modulation_mae_db": 0.3882658906615836, "300–1000 Hz_10ms_log_rms_mae_db": 1.7915040371262587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08866841239168875, "300–1000 Hz_10ms_envelope_correlation": 0.9948484070667731, "300–1000 Hz_10ms_modulation_mae_db": 0.655258781441332, "300–1000 Hz_40ms_log_rms_mae_db": 1.0125314672162224, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04268851328036377, "300–1000 Hz_40ms_envelope_correlation": 0.9988053148659696, "300–1000 Hz_40ms_modulation_mae_db": 0.8872257457593404, "1000–3000 Hz_10ms_log_rms_mae_db": 1.831590473899292, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16256204103192096, "1000–3000 Hz_10ms_envelope_correlation": 0.9829316380014496, "1000–3000 Hz_10ms_modulation_mae_db": 0.44665966435045706, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1863733399527352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10103996414124913, "1000–3000 Hz_40ms_envelope_correlation": 0.9939946358094103, "1000–3000 Hz_40ms_modulation_mae_db": 0.10741146945704243, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9927712499407935, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1957231923332598, "3000–6000 Hz_10ms_envelope_correlation": 0.9767460439482787, "3000–6000 Hz_10ms_modulation_mae_db": 1.3879969415511184, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3060518244578454, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10381153317446373, "3000–6000 Hz_40ms_envelope_correlation": 0.9935478039708129, "3000–6000 Hz_40ms_modulation_mae_db": 1.7221473108609135, "mr_stft_db": 5.845538083291598, "band_mae_db": 0.1413130115277518, "rms_gain_error_db": -0.0739229445663057, "output_peak": 0.46725153322527063}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65760, "output_samples": 65760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.023786815149948, "spectral_convergence": 0.22366320179284085, "spectral_convergence_status": "ok", "amplitude_floor": 1.6529112798953643e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.706164546984498, "spectral_convergence": 0.2164611571503158, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394789070711673e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.769798377769795, "spectral_convergence": 0.2139028138606716, "spectral_convergence_status": "ok", "amplitude_floor": 8.202976886987361e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008081818875357184, "output_power": 0.0008422701681596174, "delta_db": 0.1794230587841207, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008371388318692298, "output_power": 0.0008318825862261487, "delta_db": -0.027354544411135814, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0012771265272884344, "output_power": 0.0012353800246252258, "delta_db": -0.14433350983040813, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003871568194673651, "output_power": 0.0003700072360685704, "delta_db": -0.1967669597171283, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001322399975596578, "output_power": 0.00013148394485677474, "delta_db": -0.024901069916163848, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.960478257270517e-05, "output_power": 1.9198938299010475e-05, "delta_db": -0.09084817220450522, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11060455247724367, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03935576225995302, "pitch": null, "output_rms_dbfs": -24.646776582871446, "output_peak": 0.46265975913479407, "reference_rms_dbfs": -24.60742082061149, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.833249913301414}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00020667206862955213, "log_rms_mae_db": 1.1252314255699405, "rms_envelope_nrmse": 0.0650828243063788, "envelope_correlation": 0.9966128048864085, "rms_gain_error_db": -0.03935508249392154, "reference_mean_power": 0.003465660170380573, "output_mean_power": 0.0034343967692612824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 2.300680394398606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009732960931433549, "output_power": 0.0009505263117663153, "reference_active": true, "delta_db": -0.10280836913763258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.170752162168669e-05, "output_power": 7.697030025596997e-05, "reference_active": true, "delta_db": -0.25938856825029766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.07791133597542e-05, "output_power": 1.7737775062737186e-05, "reference_active": true, "delta_db": -0.6872786916827966}], "active_band_count": 3, "active_band_mae_db": 0.3498252096902423}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00018941444655877122, "log_rms_mae_db": 0.5422248170788184, "rms_envelope_nrmse": 0.03246249705077503, "envelope_correlation": 0.9991316428902525, "rms_gain_error_db": -0.03934876412057483, "reference_mean_power": 0.003491154294575285, "output_mean_power": 0.0034596659466335747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 2.1871865789273095e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008996086963166702, "output_power": 0.0008806660345369111, "reference_active": true, "delta_db": -0.09242398183346609}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8292682982878605e-05, "output_power": 3.5240777614207145e-05, "reference_active": true, "delta_db": -0.36070313445746527}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1507764875014336e-06, "output_power": 8.982950271213e-07, "reference_active": true, "delta_db": -1.0757198439360898}], "active_band_count": 3, "active_band_mae_db": 0.509615653409007}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.0001875778806757321, "log_rms_mae_db": 1.8560384747247483, "rms_envelope_nrmse": 0.08009039866589882, "envelope_correlation": 0.995900547253063, "rms_gain_error_db": -0.12602818811338634, "reference_mean_power": 0.0016622862050937872, "output_mean_power": 0.0016147414066195408}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 1.477365390744301e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007590270471846978, "output_power": 0.0007101517459252956, "reference_active": true, "delta_db": -0.2890609272085561}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.577918818080543e-05, "output_power": 4.9151268410259755e-05, "reference_active": true, "delta_db": -0.5493745935718141}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4662085193890421e-05, "output_power": 1.3033572001810384e-05, "reference_active": true, "delta_db": -0.511322831677251}], "active_band_count": 3, "active_band_mae_db": 0.4499194508192071}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00016891518288067464, "log_rms_mae_db": 1.0341508873831005, "rms_envelope_nrmse": 0.04238126947536929, "envelope_correlation": 0.9989411726224818, "rms_gain_error_db": -0.1260310211544413, "reference_mean_power": 0.001674521427963224, "output_mean_power": 0.0016266256158935552}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 1.422523294383883e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007062170888212691, "output_power": 0.0006604053740781292, "reference_active": true, "delta_db": -0.2912762358108845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.54990695717255e-05, "output_power": 2.2266024921741733e-05, "reference_active": true, "delta_db": -0.5888164304920134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.502016839919997e-07, "output_power": 7.555058416798007e-07, "reference_active": true, "delta_db": -0.9957796904551574}], "active_band_count": 3, "active_band_mae_db": 0.6252907855860185}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00010800393678736649, "log_rms_mae_db": 1.7314637239866026, "rms_envelope_nrmse": 0.1424279503974409, "envelope_correlation": 0.9865495012103304, "rms_gain_error_db": -0.19556143726088712, "reference_mean_power": 0.0004089931043607793, "output_mean_power": 0.00039098477129292697}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 4.3909884369057556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019183671841687967, "output_power": 0.00019220096340144915, "reference_active": true, "delta_db": 0.008238235036998876}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4527862162762071e-05, "output_power": 1.2217375428871773e-05, "reference_active": true, "delta_db": -0.7522379114589037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.245905579917347e-06, "output_power": 5.797535905641775e-06, "reference_active": true, "delta_db": -1.5299491057030703}], "active_band_count": 3, "active_band_mae_db": 0.7634750840663242}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 8.815867833009974e-05, "log_rms_mae_db": 1.1208560244813452, "rms_envelope_nrmse": 0.07938675245723582, "envelope_correlation": 0.9959078872039023, "rms_gain_error_db": -0.19556311236253304, "reference_mean_power": 0.0004120037826983822, "output_mean_power": 0.000393862734848691}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 4.2438330741066236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018201492488703638, "output_power": 0.0001821399326778659, "reference_active": true, "delta_db": 0.0029817092787270717}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.209318154228534e-06, "output_power": 6.042505239836418e-06, "reference_active": true, "delta_db": -0.7667715610084032}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163481758524237e-07, "output_power": 2.4115977798570374e-07, "reference_active": true, "delta_db": -1.1786045916778618}], "active_band_count": 3, "active_band_mae_db": 0.6494526206549973}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 8.95305280668978e-05, "log_rms_mae_db": 2.0503851140021587, "rms_envelope_nrmse": 0.2438544683112232, "envelope_correlation": 0.9653355499671378, "rms_gain_error_db": 0.2064203987068877, "reference_mean_power": 8.45768795231113e-05, "output_mean_power": 8.869388891817884e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 3.626876619728336e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.450635181337156e-05, "output_power": 1.2764452735533137e-05, "reference_active": true, "delta_db": -0.5555600636984079}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.817293180171653e-06, "output_power": 4.953551436590468e-06, "reference_active": true, "delta_db": -1.9813972309495802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650085525655336e-06, "output_power": 1.0651428711462446e-06, "reference_active": true, "delta_db": -1.6710885020975417}], "active_band_count": 3, "active_band_mae_db": 1.4026819322485098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 5.6185612647784654e-05, "log_rms_mae_db": 1.31840504682015, "rms_envelope_nrmse": 0.1395389004055832, "envelope_correlation": 0.9884569649622943, "rms_gain_error_db": 0.20641932995380563, "reference_mean_power": 8.519942751057583e-05, "output_mean_power": 8.9346719133023e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 3.342409970638096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5109841043230135e-05, "output_power": 1.2767977811700373e-05, "reference_active": true, "delta_db": -0.7313777626271392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.127479817162529e-06, "output_power": 2.6003662199092137e-06, "reference_active": true, "delta_db": -2.0065044260066296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2069317938633786e-07, "output_power": 7.190732428116976e-08, "reference_active": true, "delta_db": -2.249095992678255}], "active_band_count": 3, "active_band_mae_db": 1.6623260604373413}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8631158984866242, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "10664cc4e74159ed79068d14e8e6b50aa33671bec01e04d9dbbb3605bce226fc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9869087421047003, "standard_deviation": 0.0020718828097851484, "minimum": 0.9829316380014496, "maximum": 0.9891729525104255, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.771718189961277, "standard_deviation": 0.042598868801105395, "minimum": 1.7235138378791812, "maximum": 1.831590473899292, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7253833714995574, "standard_deviation": 0.259029869763566, "minimum": 0.44665966435045706, "maximum": 1.2746563937846311, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1451270724311155, "standard_deviation": 0.012970678908254247, "minimum": 0.12802235534524692, "maximum": 0.16256204103192096, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9956497059958997, "standard_deviation": 0.0012615461595142025, "minimum": 0.9936832673169911, "maximum": 0.9971415172781655, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.124395165589632, "standard_deviation": 0.03588359861644438, "minimum": 1.0817703296488603, "maximum": 1.1863733399527352, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5413420644234548, "standard_deviation": 0.24040564872838266, "minimum": 0.10741146945704243, "maximum": 0.9571796543874295, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09032200616408644, "standard_deviation": 0.016765687683232656, "minimum": 0.06909997408028211, "maximum": 0.11317016766515747, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9722121514962603, "standard_deviation": 0.00845883415728226, "minimum": 0.9604631609094564, "maximum": 0.9837258006141267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.041303931324478, "standard_deviation": 0.036846037453141876, "minimum": 1.985890473838433, "maximum": 2.0840982508664845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0554258460873318, "standard_deviation": 0.3830145464143116, "minimum": 0.44253474967613843, "maximum": 1.5624292989890425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21327130329642716, "standard_deviation": 0.03278511297330734, "minimum": 0.16453031259552328, "maximum": 0.25395696980333504, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905928609400414, "standard_deviation": 0.0045647654944364355, "minimum": 0.9811237826565439, "maximum": 0.9956026508571787, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3190296745889147, "standard_deviation": 0.022301495349897938, "minimum": 1.2846766283045739, "maximum": 1.344674625080607, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3268390427956134, "standard_deviation": 0.4414730845464889, "minimum": 0.8178863423022001, "maximum": 2.0464735340326796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12233038613040428, "standard_deviation": 0.027452580626732427, "minimum": 0.08664498900223805, "maximum": 0.17282840767167956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9956027291115205, "standard_deviation": 0.0004618803130815755, "minimum": 0.9948484070667731, "maximum": 0.9962159582084457, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.796334270250711, "standard_deviation": 0.05087668278105543, "minimum": 1.7053921619946337, "maximum": 1.8560384747247483, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.594321104250358, "standard_deviation": 0.11954627606918275, "minimum": 0.4264112414916917, "maximum": 0.7651398051918251, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08268412553428209, "standard_deviation": 0.004602508555173297, "minimum": 0.07628534011709398, "maximum": 0.08866841239168875, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9988254543849797, "standard_deviation": 0.00018510993956509427, "minimum": 0.9985569381509536, "maximum": 0.9991436022362368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.040800601208456, "standard_deviation": 0.06082190256496467, "minimum": 0.9586537351846136, "maximum": 1.12835381464099, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7185748383209094, "standard_deviation": 0.2176330023892231, "minimum": 0.46535472973147757, "maximum": 1.1429965370376045, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04401323396029146, "standard_deviation": 0.004831395539580335, "minimum": 0.03683195970409041, "maximum": 0.05346892347300149, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1681838365799999, "standard_deviation": 0.04884834813988491, "minimum": 0.11060455247724367, "maximum": 0.2568191156841407, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.843592002934229, "standard_deviation": 0.010502002800944774, "minimum": 5.828099035075188, "maximum": 5.863156632841814, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.48252880171242285, "standard_deviation": 0.031943944990403095, "minimum": 0.4527500086615875, "maximum": 0.5508964565246081, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07665425950025431, "standard_deviation": 0.04473937275534772, "minimum": -0.16746143330243893, "maximum": -0.03213792975587998, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968333067931892, "standard_deviation": 0.0005271451621963449, "minimum": 0.9960619498379132, "maximum": 0.997840539919872, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1057566926572002, "standard_deviation": 0.03593199259921293, "minimum": 1.061891343296375, "maximum": 1.1790780267327134, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30553726929385044, "standard_deviation": 0.11908569297609614, "minimum": 0.08357916362176969, "maximum": 0.48620716517094315, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06327476623370147, "standard_deviation": 0.005543973742296439, "minimum": 0.05231921131451748, "maximum": 0.07023631555913311, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990892752467537, "standard_deviation": 0.0002649056097998807, "minimum": 0.9985809428813694, "maximum": 0.9994988933295942, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5460039571945546, "standard_deviation": 0.04846155415608421, "minimum": 0.4689343585570472, "maximum": 0.6279730155704276, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.376925758369119, "standard_deviation": 0.1382286728886372, "minimum": 0.149453049669718, "maximum": 0.6031552019011516, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03417197920389267, "standard_deviation": 0.0048603975373204145, "minimum": 0.02551763494857327, "maximum": 0.04153094596889747, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0049", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8617522716522217, "predicted_mos": 2.8617522716522217, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "312619183f5f93055cecf0ca6b0cecc6297493ea476e3832a77dc890bdc8a9bf", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7448702002875507, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1410721058106232, "全帯域_10ms_rms_envelope_nrmse": 0.07397076009640614, "全帯域_10ms_envelope_correlation": 0.9958453709791275, "全帯域_10ms_modulation_mae_db": 0.34455199643689016, "全帯域_40ms_log_rms_mae_db": 0.6358036881221782, "全帯域_40ms_rms_envelope_nrmse": 0.043458638502132725, "全帯域_40ms_envelope_correlation": 0.9987129297770986, "全帯域_40ms_modulation_mae_db": 0.5003797994721628, "300–1000 Hz_10ms_log_rms_mae_db": 1.5285239105468162, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09872950028856177, "300–1000 Hz_10ms_envelope_correlation": 0.9939877376422319, "300–1000 Hz_10ms_modulation_mae_db": 0.8950808151822994, "300–1000 Hz_40ms_log_rms_mae_db": 0.8286643259253601, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06222808283653661, "300–1000 Hz_40ms_envelope_correlation": 0.997890683388768, "300–1000 Hz_40ms_modulation_mae_db": 1.2065744373783363, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4226671723982345, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15269977892127012, "1000–3000 Hz_10ms_envelope_correlation": 0.9848246572897094, "1000–3000 Hz_10ms_modulation_mae_db": 0.6869837728126118, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8895467078666945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08465542891777611, "1000–3000 Hz_40ms_envelope_correlation": 0.9958091296219633, "1000–3000 Hz_40ms_modulation_mae_db": 0.7192345925301188, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5267924848216399, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2589145285678151, "3000–6000 Hz_10ms_envelope_correlation": 0.961029836541783, "3000–6000 Hz_10ms_modulation_mae_db": 1.0322618515260638, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9609607329141295, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1405151820415482, "3000–6000 Hz_40ms_envelope_correlation": 0.9922678055842924, "3000–6000 Hz_40ms_modulation_mae_db": 1.0293588426651266, "mr_stft_db": 5.555405402621475, "band_mae_db": 0.30719476509067506, "rms_gain_error_db": -0.17056696641880667, "output_peak": 0.5014919492881926}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2200150609569698, "全帯域_10ms_rms_envelope_nrmse": 0.06447349592417051, "全帯域_10ms_envelope_correlation": 0.9966816625747832, "全帯域_10ms_modulation_mae_db": 0.32997410114998305, "全帯域_40ms_log_rms_mae_db": 0.5943662809982267, "全帯域_40ms_rms_envelope_nrmse": 0.02922584020309723, "全帯域_40ms_envelope_correlation": 0.9993082418304206, "全帯域_40ms_modulation_mae_db": 0.2177605536246127, "300–1000 Hz_10ms_log_rms_mae_db": 1.5251930814330972, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0804230284837451, "300–1000 Hz_10ms_envelope_correlation": 0.9957484834813886, "300–1000 Hz_10ms_modulation_mae_db": 0.30038155620490786, "300–1000 Hz_40ms_log_rms_mae_db": 0.837178448251514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03868936969509079, "300–1000 Hz_40ms_envelope_correlation": 0.9989917472084278, "300–1000 Hz_40ms_modulation_mae_db": 0.2714233945191839, "1000–3000 Hz_10ms_log_rms_mae_db": 1.506661857076617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16076714306605894, "1000–3000 Hz_10ms_envelope_correlation": 0.983572218264522, "1000–3000 Hz_10ms_modulation_mae_db": 1.3196882342909535, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9232046926308956, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09571236055696027, "1000–3000 Hz_40ms_envelope_correlation": 0.9950527418744949, "1000–3000 Hz_40ms_modulation_mae_db": 0.869305323781791, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5291422234139962, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18759461638522582, "3000–6000 Hz_10ms_envelope_correlation": 0.9786190572026452, "3000–6000 Hz_10ms_modulation_mae_db": 0.649057850195058, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0149240784122762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09805448281929534, "3000–6000 Hz_40ms_envelope_correlation": 0.9940153341419753, "3000–6000 Hz_40ms_modulation_mae_db": 0.8201948708157349, "mr_stft_db": 5.561096123478003, "band_mae_db": 0.15148767997971999, "rms_gain_error_db": -0.049163309968751595, "output_peak": 0.5367306586772563}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0904970393568951, "全帯域_10ms_rms_envelope_nrmse": 0.06934606687879138, "全帯域_10ms_envelope_correlation": 0.9961650756891939, "全帯域_10ms_modulation_mae_db": 0.1566029886562336, "全帯域_40ms_log_rms_mae_db": 0.5799746478628812, "全帯域_40ms_rms_envelope_nrmse": 0.03898918900213407, "全帯域_40ms_envelope_correlation": 0.9987606459718104, "全帯域_40ms_modulation_mae_db": 0.3820039617778715, "300–1000 Hz_10ms_log_rms_mae_db": 1.5423300375486222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08711124693629235, "300–1000 Hz_10ms_envelope_correlation": 0.9950952481433477, "300–1000 Hz_10ms_modulation_mae_db": 0.8930099563333771, "300–1000 Hz_40ms_log_rms_mae_db": 0.9180977457326047, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04205102546919624, "300–1000 Hz_40ms_envelope_correlation": 0.9989403993498324, "300–1000 Hz_40ms_modulation_mae_db": 1.1294472944303011, "1000–3000 Hz_10ms_log_rms_mae_db": 1.448986159095751, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13052259862827065, "1000–3000 Hz_10ms_envelope_correlation": 0.9896435863138694, "1000–3000 Hz_10ms_modulation_mae_db": 0.39834971691406684, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9044744506267053, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08084786068406169, "1000–3000 Hz_40ms_envelope_correlation": 0.9968417612657485, "1000–3000 Hz_40ms_modulation_mae_db": 0.49625770401516656, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5092071230632567, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23581260565826279, "3000–6000 Hz_10ms_envelope_correlation": 0.9718096456264145, "3000–6000 Hz_10ms_modulation_mae_db": 0.7059368681872527, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9732350018864815, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17686978310476376, "3000–6000 Hz_40ms_envelope_correlation": 0.9848920074063581, "3000–6000 Hz_40ms_modulation_mae_db": 0.6733779180330725, "mr_stft_db": 5.548036706375168, "band_mae_db": 0.15769597100237306, "rms_gain_error_db": -0.060582253852647414, "output_peak": 0.508518345546825}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2052501967010867, "全帯域_10ms_rms_envelope_nrmse": 0.06320101055503617, "全帯域_10ms_envelope_correlation": 0.9968584687779639, "全帯域_10ms_modulation_mae_db": 0.23090791620201423, "全帯域_40ms_log_rms_mae_db": 0.6334330091808525, "全帯域_40ms_rms_envelope_nrmse": 0.03368634567606548, "全帯域_40ms_envelope_correlation": 0.9991329743370599, "全帯域_40ms_modulation_mae_db": 0.4274878203238233, "300–1000 Hz_10ms_log_rms_mae_db": 1.5363157938380427, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08580683116236078, "300–1000 Hz_10ms_envelope_correlation": 0.9952311234837823, "300–1000 Hz_10ms_modulation_mae_db": 0.43700920912044444, "300–1000 Hz_40ms_log_rms_mae_db": 0.8723852534877182, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04190553870189524, "300–1000 Hz_40ms_envelope_correlation": 0.9989303911789882, "300–1000 Hz_40ms_modulation_mae_db": 0.6618151492226745, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4447543231535402, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14764304306217005, "1000–3000 Hz_10ms_envelope_correlation": 0.9870198468319591, "1000–3000 Hz_10ms_modulation_mae_db": 0.48990259028265476, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9198261949832975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0978966602825808, "1000–3000 Hz_40ms_envelope_correlation": 0.9954896603905807, "1000–3000 Hz_40ms_modulation_mae_db": 0.4831870664032489, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5137406530985267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1769244512830097, "3000–6000 Hz_10ms_envelope_correlation": 0.9830933348705677, "3000–6000 Hz_10ms_modulation_mae_db": 0.7781435552116376, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9644927862829322, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11347985088459103, "3000–6000 Hz_40ms_envelope_correlation": 0.9935123559257323, "3000–6000 Hz_40ms_modulation_mae_db": 0.9199312840445901, "mr_stft_db": 5.5587575085623095, "band_mae_db": 0.1837925463242811, "rms_gain_error_db": -0.08987547225065347, "output_peak": 0.48053540390527083}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.222032246592358, "全帯域_10ms_rms_envelope_nrmse": 0.06242492801665638, "全帯域_10ms_envelope_correlation": 0.9969572272926431, "全帯域_10ms_modulation_mae_db": 0.5292411224667948, "全帯域_40ms_log_rms_mae_db": 0.6629861735277551, "全帯域_40ms_rms_envelope_nrmse": 0.03287072895054177, "全帯域_40ms_envelope_correlation": 0.9992016723839082, "全帯域_40ms_modulation_mae_db": 0.5851319512030072, "300–1000 Hz_10ms_log_rms_mae_db": 1.5819568365840657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08517003801313536, "300–1000 Hz_10ms_envelope_correlation": 0.9954418958150222, "300–1000 Hz_10ms_modulation_mae_db": 0.6525374981650992, "300–1000 Hz_40ms_log_rms_mae_db": 0.8917965391446628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046736028411165004, "300–1000 Hz_40ms_envelope_correlation": 0.998803387491777, "300–1000 Hz_40ms_modulation_mae_db": 0.6488085425752987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4435743820812168, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14063335384977887, "1000–3000 Hz_10ms_envelope_correlation": 0.9886756910895368, "1000–3000 Hz_10ms_modulation_mae_db": 1.5638475774078415, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9101301901293022, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0899277959161736, "1000–3000 Hz_40ms_envelope_correlation": 0.9968604122383546, "1000–3000 Hz_40ms_modulation_mae_db": 1.3726051251892304, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5035050755512955, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23010031349759297, "3000–6000 Hz_10ms_envelope_correlation": 0.9677035824639935, "3000–6000 Hz_10ms_modulation_mae_db": 0.5841938937785724, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9550240886690123, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12045062273893005, "3000–6000 Hz_40ms_envelope_correlation": 0.991390812120559, "3000–6000 Hz_40ms_modulation_mae_db": 0.8569873679444439, "mr_stft_db": 5.56584178804133, "band_mae_db": 0.30605885726144627, "rms_gain_error_db": -0.10219714878609035, "output_peak": 0.5008548926963492}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1436532598501925, "全帯域_10ms_rms_envelope_nrmse": 0.07090775132445817, "全帯域_10ms_envelope_correlation": 0.9961264637248678, "全帯域_10ms_modulation_mae_db": 0.17693924711982997, "全帯域_40ms_log_rms_mae_db": 0.591386243257756, "全帯域_40ms_rms_envelope_nrmse": 0.041226523864931496, "全帯域_40ms_envelope_correlation": 0.9987814692177969, "全帯域_40ms_modulation_mae_db": 0.25254986345916325, "300–1000 Hz_10ms_log_rms_mae_db": 1.6563215438504357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09073666660578557, "300–1000 Hz_10ms_envelope_correlation": 0.9949046333657005, "300–1000 Hz_10ms_modulation_mae_db": 1.2739651360905415, "300–1000 Hz_40ms_log_rms_mae_db": 0.9740832209861773, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053996023216539756, "300–1000 Hz_40ms_envelope_correlation": 0.9984392966127869, "300–1000 Hz_40ms_modulation_mae_db": 1.4419603916588823, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4642796644932008, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11950724084812461, "1000–3000 Hz_10ms_envelope_correlation": 0.9905466131654866, "1000–3000 Hz_10ms_modulation_mae_db": 0.4475560759644331, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9164740065168143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07216640323015813, "1000–3000 Hz_40ms_envelope_correlation": 0.996547978942104, "1000–3000 Hz_40ms_modulation_mae_db": 0.5396168672460907, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4598468281243062, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22287512643555524, "3000–6000 Hz_10ms_envelope_correlation": 0.9716595651402414, "3000–6000 Hz_10ms_modulation_mae_db": 0.38121257109988077, "3000–6000 Hz_40ms_log_rms_mae_db": 0.89104836627551, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09955103907448977, "3000–6000 Hz_40ms_envelope_correlation": 0.994615401255477, "3000–6000 Hz_40ms_modulation_mae_db": 0.24442835021167875, "mr_stft_db": 5.546889559184206, "band_mae_db": 0.17894561406986398, "rms_gain_error_db": -0.14244498574540265, "output_peak": 0.4845678405266213}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.222369831760785, "全帯域_10ms_rms_envelope_nrmse": 0.07358338586125626, "全帯域_10ms_envelope_correlation": 0.9957358795425387, "全帯域_10ms_modulation_mae_db": 0.23407171649234182, "全帯域_40ms_log_rms_mae_db": 0.6461916822232597, "全帯域_40ms_rms_envelope_nrmse": 0.04029369831777674, "全帯域_40ms_envelope_correlation": 0.9987608698249002, "全帯域_40ms_modulation_mae_db": 0.21345188959869557, "300–1000 Hz_10ms_log_rms_mae_db": 1.565779577951218, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08545721465231443, "300–1000 Hz_10ms_envelope_correlation": 0.9953407719801409, "300–1000 Hz_10ms_modulation_mae_db": 0.44271636086484056, "300–1000 Hz_40ms_log_rms_mae_db": 0.8783836738234505, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048958977207482754, "300–1000 Hz_40ms_envelope_correlation": 0.9985751367494485, "300–1000 Hz_40ms_modulation_mae_db": 0.4533330562862288, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5329515266534213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13160887634084006, "1000–3000 Hz_10ms_envelope_correlation": 0.9896869470213054, "1000–3000 Hz_10ms_modulation_mae_db": 0.537468679854034, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0532674323167497, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08641087478289877, "1000–3000 Hz_40ms_envelope_correlation": 0.9964609273679372, "1000–3000 Hz_40ms_modulation_mae_db": 0.4860053963501698, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5558900034896133, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24094909026673927, "3000–6000 Hz_10ms_envelope_correlation": 0.9658658965285813, "3000–6000 Hz_10ms_modulation_mae_db": 0.39434397217447953, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0205339860663614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13795027004263558, "3000–6000 Hz_40ms_envelope_correlation": 0.988532480706263, "3000–6000 Hz_40ms_modulation_mae_db": 0.48389652080723106, "mr_stft_db": 5.548089086725067, "band_mae_db": 0.21670793534116148, "rms_gain_error_db": -0.10475821664909088, "output_peak": 0.49723281998966673}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.220554288956778, "全帯域_10ms_rms_envelope_nrmse": 0.06395976199726053, "全帯域_10ms_envelope_correlation": 0.996787026877908, "全帯域_10ms_modulation_mae_db": 0.2781212340840234, "全帯域_40ms_log_rms_mae_db": 0.6233830011392663, "全帯域_40ms_rms_envelope_nrmse": 0.03391309449596492, "全帯域_40ms_envelope_correlation": 0.9991287261858589, "全帯域_40ms_modulation_mae_db": 0.2796815917103684, "300–1000 Hz_10ms_log_rms_mae_db": 1.6392617423517122, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08771988142994788, "300–1000 Hz_10ms_envelope_correlation": 0.9950411078611652, "300–1000 Hz_10ms_modulation_mae_db": 0.5502384569719017, "300–1000 Hz_40ms_log_rms_mae_db": 0.9234529044117825, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050803121157046674, "300–1000 Hz_40ms_envelope_correlation": 0.998403606676104, "300–1000 Hz_40ms_modulation_mae_db": 0.4635080625132764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4826576770986943, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15063843348174247, "1000–3000 Hz_10ms_envelope_correlation": 0.9872710755696189, "1000–3000 Hz_10ms_modulation_mae_db": 1.2577994618860062, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9814614576835354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09700753984483727, "1000–3000 Hz_40ms_envelope_correlation": 0.9966494829056378, "1000–3000 Hz_40ms_modulation_mae_db": 0.8894565078757397, "3000–6000 Hz_10ms_log_rms_mae_db": 1.485000968671803, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17376710085832375, "3000–6000 Hz_10ms_envelope_correlation": 0.9816820455132521, "3000–6000 Hz_10ms_modulation_mae_db": 0.6325521300575948, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9295663955896762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08419827369316871, "3000–6000 Hz_40ms_envelope_correlation": 0.9955900551990344, "3000–6000 Hz_40ms_modulation_mae_db": 0.7498660932437939, "mr_stft_db": 5.550911500513469, "band_mae_db": 0.2489467243839449, "rms_gain_error_db": -0.09385460874958335, "output_peak": 0.5042433687489517}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65760, "output_samples": 65760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.626121397736435, "spectral_convergence": 0.22554774860765192, "spectral_convergence_status": "ok", "amplitude_floor": 1.6529112798953643e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.347529573717584, "spectral_convergence": 0.22416229821315006, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394789070711673e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.692565236410405, "spectral_convergence": 0.22027585646141976, "spectral_convergence_status": "ok", "amplitude_floor": 8.202976886987361e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008081818875357184, "output_power": 0.0008146778420325098, "delta_db": 0.034767912574408304, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008371388318692298, "output_power": 0.0008458007144995332, "delta_db": 0.04470559926018203, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0012771265272884344, "output_power": 0.0011758520643443224, "delta_db": -0.3588123979536762, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003871568194673651, "output_power": 0.00036492985434630426, "delta_db": -0.25677519650596553, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001322399975596578, "output_power": 0.00010853986996288862, "delta_db": -0.8577353561538591, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.960478257270517e-05, "output_power": 1.833685358620698e-05, "delta_db": -0.29037212809595897, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.30719476509067506, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.17056696641880667, "pitch": null, "output_rms_dbfs": -24.7779877870303, "output_peak": 0.5014919492881926, "reference_rms_dbfs": -24.60742082061149, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.555405402621475}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00020667206862955213, "log_rms_mae_db": 1.1410721058106232, "rms_envelope_nrmse": 0.07397076009640614, "envelope_correlation": 0.9958453709791275, "rms_gain_error_db": -0.17056526635553, "reference_mean_power": 0.003465660170380573, "output_mean_power": 0.003332187635476522}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 2.300680394398606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009732960931433549, "output_power": 0.0009088979387933054, "reference_active": true, "delta_db": -0.2972986157291083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.170752162168669e-05, "output_power": 7.174414790249061e-05, "reference_active": true, "delta_db": -0.5647555565187036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.07791133597542e-05, "output_power": 1.9974082101060473e-05, "reference_active": true, "delta_db": -0.17160181706285854}], "active_band_count": 3, "active_band_mae_db": 0.34455199643689016}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00018941444655877122, "log_rms_mae_db": 0.6358036881221782, "rms_envelope_nrmse": 0.043458638502132725, "envelope_correlation": 0.9987129297770986, "rms_gain_error_db": -0.1705695783910583, "reference_mean_power": 0.003491154294575285, "output_mean_power": 0.003356696574859631}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 2.1871865789273095e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008996086963166702, "output_power": 0.0008408633021186605, "reference_active": true, "delta_db": -0.29328246075846776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8292682982878605e-05, "output_power": 3.38412110416954e-05, "reference_active": true, "delta_db": -0.5366990001626764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1507764875014336e-06, "output_power": 9.85996074974645e-07, "reference_active": true, "delta_db": -0.6711579374953445}], "active_band_count": 3, "active_band_mae_db": 0.5003797994721628}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.0001875778806757321, "log_rms_mae_db": 1.5285239105468162, "rms_envelope_nrmse": 0.09872950028856177, "envelope_correlation": 0.9939877376422319, "rms_gain_error_db": -0.2555017416994191, "reference_mean_power": 0.0016622862050937872, "output_mean_power": 0.0015673126058345032}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 1.477365390744301e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007590270471846978, "output_power": 0.0007029334273746936, "reference_active": true, "delta_db": -0.3334305553332842}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.577918818080543e-05, "output_power": 4.183745129612925e-05, "reference_active": true, "delta_db": -1.2490696939027792}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4662085193890421e-05, "output_power": 1.1374217247282863e-05, "reference_active": true, "delta_db": -1.1027421963108348}], "active_band_count": 3, "active_band_mae_db": 0.8950808151822994}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00016891518288067464, "log_rms_mae_db": 0.8286643259253601, "rms_envelope_nrmse": 0.06222808283653661, "envelope_correlation": 0.997890683388768, "rms_gain_error_db": -0.2555057921098546, "reference_mean_power": 0.001674521427963224, "output_mean_power": 0.0015788473050124623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 1.422523294383883e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007062170888212691, "output_power": 0.0006514478696590872, "reference_active": true, "delta_db": -0.350585540705449}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.54990695717255e-05, "output_power": 1.9462926551981997e-05, "reference_active": true, "delta_db": -1.1731619014185761}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.502016839919997e-07, "output_power": 5.864327402549314e-07, "reference_active": true, "delta_db": -2.095975870010984}], "active_band_count": 3, "active_band_mae_db": 1.2065744373783363}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00010800393678736649, "log_rms_mae_db": 1.4226671723982345, "rms_envelope_nrmse": 0.15269977892127012, "envelope_correlation": 0.9848246572897094, "rms_gain_error_db": -0.3197766827906723, "reference_mean_power": 0.0004089931043607793, "output_mean_power": 0.0003799603836978634}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 4.3909884369057556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019183671841687967, "output_power": 0.00016737508485804948, "reference_active": true, "delta_db": -0.5924092651756568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4527862162762071e-05, "output_power": 1.4923206063556859e-05, "reference_active": true, "delta_db": 0.11660425176448128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.245905579917347e-06, "output_power": 6.040106238638288e-06, "reference_active": true, "delta_db": -1.351937801497697}], "active_band_count": 3, "active_band_mae_db": 0.6869837728126118}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 8.815867833009974e-05, "log_rms_mae_db": 0.8895467078666945, "rms_envelope_nrmse": 0.08465542891777611, "envelope_correlation": 0.9958091296219633, "rms_gain_error_db": -0.31977747046372407, "reference_mean_power": 0.0004120037826983822, "output_mean_power": 0.00038275727707286585}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 4.2438330741066236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018201492488703638, "output_power": 0.00015905731006558605, "reference_active": true, "delta_db": -0.5855336728161075}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.209318154228534e-06, "output_power": 7.154707490529388e-06, "reference_active": true, "delta_db": -0.0330230880367645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163481758524237e-07, "output_power": 2.219476918201968e-07, "reference_active": true, "delta_db": -1.539147016737484}], "active_band_count": 3, "active_band_mae_db": 0.7192345925301188}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 8.95305280668978e-05, "log_rms_mae_db": 1.5267924848216399, "rms_envelope_nrmse": 0.2589145285678151, "envelope_correlation": 0.961029836541783, "rms_gain_error_db": -0.8326584369634376, "reference_mean_power": 8.45768795231113e-05, "output_mean_power": 6.982095971353386e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 3.626876619728336e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.450635181337156e-05, "output_power": 1.3942296018979168e-05, "reference_active": true, "delta_db": -0.17223906603824107}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.817293180171653e-06, "output_power": 5.988465319606236e-06, "reference_active": true, "delta_db": -1.1574086125255003}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650085525655336e-06, "output_power": 1.0418445582062193e-06, "reference_active": true, "delta_db": -1.7671378760144498}], "active_band_count": 3, "active_band_mae_db": 1.0322618515260638}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 5.6185612647784654e-05, "log_rms_mae_db": 0.9609607329141295, "rms_envelope_nrmse": 0.1405151820415482, "envelope_correlation": 0.9922678055842924, "rms_gain_error_db": -0.832660796597978, "reference_mean_power": 8.519942751057583e-05, "output_mean_power": 7.033485507255316e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 3.342409970638096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5109841043230135e-05, "output_power": 1.4137283946368546e-05, "reference_active": true, "delta_db": -0.28893914688710065}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.127479817162529e-06, "output_power": 3.1586707245565418e-06, "reference_active": true, "delta_db": -1.161806030346392}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2069317938633786e-07, "output_power": 8.278460727544567e-08, "reference_active": true, "delta_db": -1.6373313507618867}], "active_band_count": 3, "active_band_mae_db": 1.0293588426651266}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8891178114199442, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3648451392e05735b9931a97e568ca8ce47a6a653c8daf4dd1e010e72a89b8e1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.987655079443251, "standard_deviation": 0.0024703176558621602, "minimum": 0.983572218264522, "maximum": 0.9905466131654866, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4683165952563344, "standard_deviation": 0.03682951116399695, "minimum": 1.4226671723982345, "maximum": 1.5329515266534213, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8376995136765752, "standard_deviation": 0.4652410646558947, "minimum": 0.39834971691406684, "maximum": 1.5638475774078415, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14175255852478197, "standard_deviation": 0.013732993153552439, "minimum": 0.11950724084812461, "maximum": 0.16076714306605894, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962140118258527, "standard_deviation": 0.0006773057691318484, "minimum": 0.9950527418744949, "maximum": 0.9968604122383546, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9372981415942494, "standard_deviation": 0.05402711673784812, "minimum": 0.8895467078666945, "maximum": 1.0532674323167497, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7319585729239445, "standard_deviation": 0.30919020797710395, "minimum": 0.4831870664032489, "maximum": 1.3726051251892304, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08807811552693083, "standard_deviation": 0.008926646331709403, "minimum": 0.07216640323015813, "maximum": 0.0978966602825808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9726828704859349, "standard_deviation": 0.007869185393870177, "minimum": 0.961029836541783, "maximum": 0.9830933348705677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5103906700293046, "standard_deviation": 0.02915520315972644, "minimum": 1.4598468281243062, "maximum": 1.5558900034896133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6447128365288175, "standard_deviation": 0.20965708064173683, "minimum": 0.38121257109988077, "maximum": 1.0322618515260638, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21586722911906558, "standard_deviation": 0.032115818277499486, "minimum": 0.17376710085832375, "maximum": 0.2589145285678151, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9918520315424615, "standard_deviation": 0.0035635562767090197, "minimum": 0.9848920074063581, "maximum": 0.9955900551990344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9637231795120474, "standard_deviation": 0.04220982706938986, "minimum": 0.89104836627551, "maximum": 1.0205339860663614, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7222551559707089, "standard_deviation": 0.25338229901622383, "minimum": 0.24442835021167875, "maximum": 1.0293588426651266, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1213836880499278, "standard_deviation": 0.02972777709574254, "minimum": 0.08419827369316871, "maximum": 0.17686978310476376, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9950988752215975, "standard_deviation": 0.0005198027793356666, "minimum": 0.9939877376422319, "maximum": 0.9957484834813886, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5719603155130013, "standard_deviation": 0.05071884914339456, "minimum": 1.5251930814330972, "maximum": 1.6563215438504357, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6806173736166764, "standard_deviation": 0.3210418290559644, "minimum": 0.30038155620490786, "maximum": 1.2739651360905415, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0876443009465179, "standard_deviation": 0.005334023187076063, "minimum": 0.0804230284837451, "maximum": 0.09872950028856177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986218310820166, "standard_deviation": 0.0003749654801323275, "minimum": 0.997890683388768, "maximum": 0.9989917472084278, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8905052639704087, "standard_deviation": 0.04777499790169483, "minimum": 0.8286643259253601, "maximum": 0.9740832209861773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7846087910730228, "standard_deviation": 0.42068112362139337, "minimum": 0.2714233945191839, "maximum": 1.4419603916588823, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04817102083686914, "standard_deviation": 0.007632187417616473, "minimum": 0.03868936969509079, "maximum": 0.06222808283653661, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21885376168168325, "standard_deviation": 0.062586472689576, "minimum": 0.15148767997971999, "maximum": 0.30719476509067506, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.554378459437628, "standard_deviation": 0.0070167459754225375, "minimum": 5.546889559184206, "maximum": 5.56584178804133, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5017719099223916, "standard_deviation": 0.017080600710531958, "minimum": 0.48053540390527083, "maximum": 0.5367306586772563, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1016803703026283, "standard_deviation": 0.039757393128385624, "minimum": -0.17056696641880667, "maximum": -0.049163309968751595, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963946469323782, "standard_deviation": 0.0004822967524639043, "minimum": 0.9957358795425387, "maximum": 0.9969572272926431, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1831805037482108, "standard_deviation": 0.051006528830477724, "minimum": 1.0904970393568951, "maximum": 1.222369831760785, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2850512903260139, "standard_deviation": 0.11889532897129051, "minimum": 0.1566029886562336, "maximum": 0.5292411224667948, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06773339508175444, "standard_deviation": 0.004771851146943131, "minimum": 0.06242492801665638, "maximum": 0.07397076009640614, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989734411911066, "standard_deviation": 0.0002417002352847012, "minimum": 0.9987129297770986, "maximum": 0.9993082418304206, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.620940590789022, "standard_deviation": 0.029389619336168238, "minimum": 0.5799746478628812, "maximum": 0.6629861735277551, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3573059288962131, "standard_deviation": 0.1389124014548102, "minimum": 0.21345188959869557, "maximum": 0.5851319512030072, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03670800737658056, "standard_deviation": 0.0049526109961860005, "minimum": 0.02922584020309723, "maximum": 0.043458638502132725, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0049", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.193633794784546, "predicted_mos": 3.193633794784546, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "e5ff6d4d6d5e72832a136ffd7a9b56f3e4aae3237bb5c35a323fb7cb956573d5", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6521053998731077, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1424951926310587, "全帯域_10ms_rms_envelope_nrmse": 0.07043035764922938, "全帯域_10ms_envelope_correlation": 0.9962154457916561, "全帯域_10ms_modulation_mae_db": 0.40357015844979854, "全帯域_40ms_log_rms_mae_db": 0.632555979920343, "全帯域_40ms_rms_envelope_nrmse": 0.04455185214120575, "全帯域_40ms_envelope_correlation": 0.9985786731169287, "全帯域_40ms_modulation_mae_db": 0.8172150494449469, "300–1000 Hz_10ms_log_rms_mae_db": 1.6190297246401981, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09065138783869592, "300–1000 Hz_10ms_envelope_correlation": 0.995018006163542, "300–1000 Hz_10ms_modulation_mae_db": 1.078517567328657, "300–1000 Hz_40ms_log_rms_mae_db": 0.8962995974627929, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05635353366489691, "300–1000 Hz_40ms_envelope_correlation": 0.9983672074323201, "300–1000 Hz_40ms_modulation_mae_db": 1.3850321795105325, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5607089120348032, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15026138973968298, "1000–3000 Hz_10ms_envelope_correlation": 0.9850244215311231, "1000–3000 Hz_10ms_modulation_mae_db": 0.73976324365418, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9903012094506009, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09825952495238735, "1000–3000 Hz_40ms_envelope_correlation": 0.9936051541401144, "1000–3000 Hz_40ms_modulation_mae_db": 0.5752330453412663, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6904257206359092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1557949606753236, "3000–6000 Hz_10ms_envelope_correlation": 0.9853698355851638, "3000–6000 Hz_10ms_modulation_mae_db": 0.29119596918964646, "3000–6000 Hz_40ms_log_rms_mae_db": 1.128425130484682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08825566542858328, "3000–6000 Hz_40ms_envelope_correlation": 0.9951637444988418, "3000–6000 Hz_40ms_modulation_mae_db": 0.13730095985202712, "mr_stft_db": 5.649965516475579, "band_mae_db": 0.20102742622369452, "rms_gain_error_db": -0.14870058853217286, "output_peak": 0.47017803920192675}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1719460806790771, "全帯域_10ms_rms_envelope_nrmse": 0.06798217662719587, "全帯域_10ms_envelope_correlation": 0.9963880829143078, "全帯域_10ms_modulation_mae_db": 0.29480942156075046, "全帯域_40ms_log_rms_mae_db": 0.5843336751129095, "全帯域_40ms_rms_envelope_nrmse": 0.03719136476531889, "全帯域_40ms_envelope_correlation": 0.9989677719360787, "全帯域_40ms_modulation_mae_db": 0.40124915730368826, "300–1000 Hz_10ms_log_rms_mae_db": 1.6427041519639538, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0862027269844794, "300–1000 Hz_10ms_envelope_correlation": 0.9952051226495248, "300–1000 Hz_10ms_modulation_mae_db": 1.020527631552562, "300–1000 Hz_40ms_log_rms_mae_db": 0.914546278324243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046231059171016865, "300–1000 Hz_40ms_envelope_correlation": 0.998690539785116, "300–1000 Hz_40ms_modulation_mae_db": 1.099618835271902, "1000–3000 Hz_10ms_log_rms_mae_db": 1.643182530932935, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14185260924026777, "1000–3000 Hz_10ms_envelope_correlation": 0.9875534160467837, "1000–3000 Hz_10ms_modulation_mae_db": 1.0062290999807786, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0689630827536332, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08018014479234851, "1000–3000 Hz_40ms_envelope_correlation": 0.9970104743653223, "1000–3000 Hz_40ms_modulation_mae_db": 0.6932865459834602, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6335718705880842, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24656021451384352, "3000–6000 Hz_10ms_envelope_correlation": 0.9634679094456785, "3000–6000 Hz_10ms_modulation_mae_db": 0.26177817717562396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0469492498978505, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15236601991043994, "3000–6000 Hz_40ms_envelope_correlation": 0.9872820348811618, "3000–6000 Hz_40ms_modulation_mae_db": 0.3690966149443282, "mr_stft_db": 5.646274134396172, "band_mae_db": 0.244369936716926, "rms_gain_error_db": -0.11171055058705297, "output_peak": 0.4970031111793775}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1328020792791227, "全帯域_10ms_rms_envelope_nrmse": 0.07221048235459683, "全帯域_10ms_envelope_correlation": 0.9958332877076789, "全帯域_10ms_modulation_mae_db": 0.2566473403129964, "全帯域_40ms_log_rms_mae_db": 0.6301272069843876, "全帯域_40ms_rms_envelope_nrmse": 0.03871908175312984, "全帯域_40ms_envelope_correlation": 0.9987786485285234, "全帯域_40ms_modulation_mae_db": 0.3350948721376484, "300–1000 Hz_10ms_log_rms_mae_db": 1.6551446448413905, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09335454235958841, "300–1000 Hz_10ms_envelope_correlation": 0.9943186262346889, "300–1000 Hz_10ms_modulation_mae_db": 0.6090864792930608, "300–1000 Hz_40ms_log_rms_mae_db": 0.853988963390062, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045334116194799046, "300–1000 Hz_40ms_envelope_correlation": 0.9987154604634331, "300–1000 Hz_40ms_modulation_mae_db": 0.7754157584939785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5819313202150995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13016681242164876, "1000–3000 Hz_10ms_envelope_correlation": 0.9894583424453152, "1000–3000 Hz_10ms_modulation_mae_db": 0.9113216828886094, "1000–3000 Hz_40ms_log_rms_mae_db": 1.071605706460852, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07671888340470978, "1000–3000 Hz_40ms_envelope_correlation": 0.9970221605495837, "1000–3000 Hz_40ms_modulation_mae_db": 0.9131706441673995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6595203174889572, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22641465807595465, "3000–6000 Hz_10ms_envelope_correlation": 0.975191892840807, "3000–6000 Hz_10ms_modulation_mae_db": 0.3518737403937548, "3000–6000 Hz_40ms_log_rms_mae_db": 1.11330884842658, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16300476390550173, "3000–6000 Hz_40ms_envelope_correlation": 0.9886390544644694, "3000–6000 Hz_40ms_modulation_mae_db": 0.4533744174367886, "mr_stft_db": 5.649716476019921, "band_mae_db": 0.23068704539603427, "rms_gain_error_db": -0.04330543672742121, "output_peak": 0.49303211195328506}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1100349352366141, "全帯域_10ms_rms_envelope_nrmse": 0.063785354150421, "全帯域_10ms_envelope_correlation": 0.9968135963701137, "全帯域_10ms_modulation_mae_db": 0.47108997609587905, "全帯域_40ms_log_rms_mae_db": 0.5806324817357373, "全帯域_40ms_rms_envelope_nrmse": 0.03624891818750078, "全帯域_40ms_envelope_correlation": 0.9989999280837969, "全帯域_40ms_modulation_mae_db": 0.7104422176814201, "300–1000 Hz_10ms_log_rms_mae_db": 1.6852330912727735, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08913573790365849, "300–1000 Hz_10ms_envelope_correlation": 0.994808936922584, "300–1000 Hz_10ms_modulation_mae_db": 0.9127862157767748, "300–1000 Hz_40ms_log_rms_mae_db": 1.0089279426759812, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04911432033607802, "300–1000 Hz_40ms_envelope_correlation": 0.9984248659673561, "300–1000 Hz_40ms_modulation_mae_db": 0.9260693309656557, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5773197388246827, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14100493146674017, "1000–3000 Hz_10ms_envelope_correlation": 0.9874909152407235, "1000–3000 Hz_10ms_modulation_mae_db": 0.4658364788979008, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0209541100634463, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08443313415559342, "1000–3000 Hz_40ms_envelope_correlation": 0.9962096560403275, "1000–3000 Hz_40ms_modulation_mae_db": 0.3715510377377919, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7013990798541485, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18130872609931986, "3000–6000 Hz_10ms_envelope_correlation": 0.9809029937199677, "3000–6000 Hz_10ms_modulation_mae_db": 0.648474045188491, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1038070012078782, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0959539949676806, "3000–6000 Hz_40ms_envelope_correlation": 0.9946558156496212, "3000–6000 Hz_40ms_modulation_mae_db": 0.9203738258203081, "mr_stft_db": 5.659248678812304, "band_mae_db": 0.12029578666482098, "rms_gain_error_db": -0.09761077485792077, "output_peak": 0.5568621427350434}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1570247886083003, "全帯域_10ms_rms_envelope_nrmse": 0.07083687978434981, "全帯域_10ms_envelope_correlation": 0.9959822163525984, "全帯域_10ms_modulation_mae_db": 0.2419784037657119, "全帯域_40ms_log_rms_mae_db": 0.648974220252245, "全帯域_40ms_rms_envelope_nrmse": 0.04539133566237327, "全帯域_40ms_envelope_correlation": 0.9982858625704393, "全帯域_40ms_modulation_mae_db": 0.35929644614019884, "300–1000 Hz_10ms_log_rms_mae_db": 1.6549768349826577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07514874799139741, "300–1000 Hz_10ms_envelope_correlation": 0.9963565052114182, "300–1000 Hz_10ms_modulation_mae_db": 0.6901322137973153, "300–1000 Hz_40ms_log_rms_mae_db": 0.948908076144733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039991052520364494, "300–1000 Hz_40ms_envelope_correlation": 0.999012013011323, "300–1000 Hz_40ms_modulation_mae_db": 0.814703081649914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6112035114232262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13279478320440344, "1000–3000 Hz_10ms_envelope_correlation": 0.9892287149443438, "1000–3000 Hz_10ms_modulation_mae_db": 0.5034189637718044, "1000–3000 Hz_40ms_log_rms_mae_db": 1.053305171826503, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08193745932841977, "1000–3000 Hz_40ms_envelope_correlation": 0.9967321357660097, "1000–3000 Hz_40ms_modulation_mae_db": 0.4496019524993307, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7497296040776928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28470710375642616, "3000–6000 Hz_10ms_envelope_correlation": 0.9657068266139945, "3000–6000 Hz_10ms_modulation_mae_db": 0.49850291866360097, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1595651682022532, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21817761871888688, "3000–6000 Hz_40ms_envelope_correlation": 0.9829418220820169, "3000–6000 Hz_40ms_modulation_mae_db": 0.5272771975776658, "mr_stft_db": 5.64818434066737, "band_mae_db": 0.26622063422377606, "rms_gain_error_db": -0.03075047719531154, "output_peak": 0.45796474567775475}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1573924427341105, "全帯域_10ms_rms_envelope_nrmse": 0.0766941910266784, "全帯域_10ms_envelope_correlation": 0.9952909366574785, "全帯域_10ms_modulation_mae_db": 0.12173575754805628, "全帯域_40ms_log_rms_mae_db": 0.5893511868664371, "全帯域_40ms_rms_envelope_nrmse": 0.036521163194384275, "全帯域_40ms_envelope_correlation": 0.9989011658819417, "全帯域_40ms_modulation_mae_db": 0.24440960539958323, "300–1000 Hz_10ms_log_rms_mae_db": 1.6892482894189547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09503825552901685, "300–1000 Hz_10ms_envelope_correlation": 0.9940746327819716, "300–1000 Hz_10ms_modulation_mae_db": 0.2957080078652976, "300–1000 Hz_40ms_log_rms_mae_db": 1.1393872159446985, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05201765481344418, "300–1000 Hz_40ms_envelope_correlation": 0.9982057627991883, "300–1000 Hz_40ms_modulation_mae_db": 0.36329471194586976, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5349138299537894, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12096462369670799, "1000–3000 Hz_10ms_envelope_correlation": 0.9904783919846593, "1000–3000 Hz_10ms_modulation_mae_db": 0.5950229529535275, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9862055359498533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06638259810160702, "1000–3000 Hz_40ms_envelope_correlation": 0.9973630130221516, "1000–3000 Hz_40ms_modulation_mae_db": 0.5365282756618707, "3000–6000 Hz_10ms_log_rms_mae_db": 1.740691453677047, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24065563504166287, "3000–6000 Hz_10ms_envelope_correlation": 0.9648350523449134, "3000–6000 Hz_10ms_modulation_mae_db": 1.4327519343499773, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1080062496863554, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12290089397068925, "3000–6000 Hz_40ms_envelope_correlation": 0.9904766937719942, "3000–6000 Hz_40ms_modulation_mae_db": 1.8106058173084085, "mr_stft_db": 5.650069102586166, "band_mae_db": 0.09744581994541114, "rms_gain_error_db": -0.04538926663724494, "output_peak": 0.5300746604459067}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1138135942077747, "全帯域_10ms_rms_envelope_nrmse": 0.05681972463707614, "全帯域_10ms_envelope_correlation": 0.9974564553181182, "全帯域_10ms_modulation_mae_db": 0.17675253840027935, "全帯域_40ms_log_rms_mae_db": 0.5642777818042788, "全帯域_40ms_rms_envelope_nrmse": 0.02749099101387179, "全帯域_40ms_envelope_correlation": 0.9994292468634565, "全帯域_40ms_modulation_mae_db": 0.41626486014336644, "300–1000 Hz_10ms_log_rms_mae_db": 1.692552709668942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08348536012393033, "300–1000 Hz_10ms_envelope_correlation": 0.9954882498598809, "300–1000 Hz_10ms_modulation_mae_db": 0.6691524524687368, "300–1000 Hz_40ms_log_rms_mae_db": 0.9786635639680207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043805252729657136, "300–1000 Hz_40ms_envelope_correlation": 0.998811573104249, "300–1000 Hz_40ms_modulation_mae_db": 1.0188949654176211, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5630777014634194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13586080362481215, "1000–3000 Hz_10ms_envelope_correlation": 0.9886133942605154, "1000–3000 Hz_10ms_modulation_mae_db": 0.5464911946243431, "1000–3000 Hz_40ms_log_rms_mae_db": 1.005111037002322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07941371834767401, "1000–3000 Hz_40ms_envelope_correlation": 0.9969856365587927, "1000–3000 Hz_40ms_modulation_mae_db": 0.36851472570069327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6166182530433422, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16668386970931484, "3000–6000 Hz_10ms_envelope_correlation": 0.9831644278294447, "3000–6000 Hz_10ms_modulation_mae_db": 0.37168247024540907, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0176654493784139, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08936365387472017, "3000–6000 Hz_40ms_envelope_correlation": 0.9951333651755445, "3000–6000 Hz_40ms_modulation_mae_db": 0.3579305701848343, "mr_stft_db": 5.649539365875029, "band_mae_db": 0.13267049412188334, "rms_gain_error_db": -0.07294692703380275, "output_peak": 0.491659645797489}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1516076634873191, "全帯域_10ms_rms_envelope_nrmse": 0.06073371340090228, "全帯域_10ms_envelope_correlation": 0.9970911913437244, "全帯域_10ms_modulation_mae_db": 0.5978296928150358, "全帯域_40ms_log_rms_mae_db": 0.5296141222136963, "全帯域_40ms_rms_envelope_nrmse": 0.030826790554425002, "全帯域_40ms_envelope_correlation": 0.9992680801813384, "全帯域_40ms_modulation_mae_db": 0.7660405248106025, "300–1000 Hz_10ms_log_rms_mae_db": 1.7090681133769654, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07933458887144933, "300–1000 Hz_10ms_envelope_correlation": 0.9958783023896094, "300–1000 Hz_10ms_modulation_mae_db": 0.7993214313765927, "300–1000 Hz_40ms_log_rms_mae_db": 0.925539530843933, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.042715112175444735, "300–1000 Hz_40ms_envelope_correlation": 0.9987972534527726, "300–1000 Hz_40ms_modulation_mae_db": 0.872423403571449, "1000–3000 Hz_10ms_log_rms_mae_db": 1.589916793821843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15225176583277567, "1000–3000 Hz_10ms_envelope_correlation": 0.986515077712372, "1000–3000 Hz_10ms_modulation_mae_db": 0.9416610534122478, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9828057195185541, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09890922831968792, "1000–3000 Hz_40ms_envelope_correlation": 0.9959516519761256, "1000–3000 Hz_40ms_modulation_mae_db": 0.7696017028444965, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6677573782337873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20795003639348097, "3000–6000 Hz_10ms_envelope_correlation": 0.9765222221841967, "3000–6000 Hz_10ms_modulation_mae_db": 0.7390760806015145, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0657036473000912, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11680930052040629, "3000–6000 Hz_40ms_envelope_correlation": 0.9932993658727426, "3000–6000 Hz_40ms_modulation_mae_db": 0.7513301330006446, "mr_stft_db": 5.649396433477558, "band_mae_db": 0.1352488795755403, "rms_gain_error_db": -0.07443327705919144, "output_peak": 0.45723972788115363}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65760, "output_samples": 65760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.799707202222829, "spectral_convergence": 0.22419277362714507, "spectral_convergence_status": "ok", "amplitude_floor": 1.6529112798953643e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.449695245903603, "spectral_convergence": 0.21873088524159084, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394789070711673e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.700494101300303, "spectral_convergence": 0.2262889351415317, "spectral_convergence_status": "ok", "amplitude_floor": 8.202976886987361e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008081818875357184, "output_power": 0.0008067536471112012, "delta_db": -0.007681756746432826, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008371388318692298, "output_power": 0.000830074615584715, "delta_db": -0.036803548054581796, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0012771265272884344, "output_power": 0.0011898961264383218, "delta_db": -0.307248750188979, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003871568194673651, "output_power": 0.00036234268942938173, "delta_db": -0.28767409943961675, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001322399975596578, "output_power": 0.00013792351791272814, "delta_db": 0.18275493299716075, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.960478257270517e-05, "output_power": 1.7945761147064415e-05, "delta_db": -0.3840014699153959, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20102742622369452, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14870058853217286, "pitch": null, "output_rms_dbfs": -24.756121409143663, "output_peak": 0.47017803920192675, "reference_rms_dbfs": -24.60742082061149, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.649965516475579}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00020667206862955213, "log_rms_mae_db": 1.1424951926310587, "rms_envelope_nrmse": 0.07043035764922938, "envelope_correlation": 0.9962154457916561, "rms_gain_error_db": -0.14870066836835688, "reference_mean_power": 0.003465660170380573, "output_mean_power": 0.0033490058668308375}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 2.300680394398606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009732960931433549, "output_power": 0.0009269732856036655, "reference_active": true, "delta_db": -0.21177761686158703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.170752162168669e-05, "output_power": 7.324276901155855e-05, "reference_active": true, "delta_db": -0.47497282691271914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.07791133597542e-05, "output_power": 1.84175140234806e-05, "reference_active": true, "delta_db": -0.5239600315750893}], "active_band_count": 3, "active_band_mae_db": 0.40357015844979854}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00018941444655877122, "log_rms_mae_db": 0.632555979920343, "rms_envelope_nrmse": 0.04455185214120575, "envelope_correlation": 0.9985786731169287, "rms_gain_error_db": -0.14870677090054515, "reference_mean_power": 0.003491154294575285, "output_mean_power": 0.0033736371169616032}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 2.1871865789273095e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008996086963166702, "output_power": 0.0008550995102003813, "reference_active": true, "delta_db": -0.2203698734070822}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8292682982878605e-05, "output_power": 3.541061139960422e-05, "reference_active": true, "delta_db": -0.3398237097395745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1507764875014336e-06, "output_power": 7.444676373588311e-07, "reference_active": true, "delta_db": -1.8914515651881842}], "active_band_count": 3, "active_band_mae_db": 0.8172150494449469}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.0001875778806757321, "log_rms_mae_db": 1.6190297246401981, "rms_envelope_nrmse": 0.09065138783869592, "envelope_correlation": 0.995018006163542, "rms_gain_error_db": -0.24921129255658966, "reference_mean_power": 0.0016622862050937872, "output_mean_power": 0.0015695843924224356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 1.477365390744301e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007590270471846978, "output_power": 0.0007189755786055403, "reference_active": true, "delta_db": -0.23543112846955555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.577918818080543e-05, "output_power": 4.353356815564873e-05, "reference_active": true, "delta_db": -1.076479243523423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4662085193890421e-05, "output_power": 9.415242400478234e-06, "reference_active": true, "delta_db": -1.923642329992992}], "active_band_count": 3, "active_band_mae_db": 1.078517567328657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00016891518288067464, "log_rms_mae_db": 0.8962995974627929, "rms_envelope_nrmse": 0.05635353366489691, "envelope_correlation": 0.9983672074323201, "rms_gain_error_db": -0.24921439218687488, "reference_mean_power": 0.001674521427963224, "output_mean_power": 0.0015811361570542799}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 1.422523294383883e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007062170888212691, "output_power": 0.0006678297811345794, "reference_active": true, "delta_db": -0.24272440241952467}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.54990695717255e-05, "output_power": 2.0358369095892325e-05, "reference_active": true, "delta_db": -0.9778135004509959}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.502016839919997e-07, "output_power": 4.834593148043921e-07, "reference_active": true, "delta_db": -2.934558635661077}], "active_band_count": 3, "active_band_mae_db": 1.3850321795105325}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00010800393678736649, "log_rms_mae_db": 1.5607089120348032, "rms_envelope_nrmse": 0.15026138973968298, "envelope_correlation": 0.9850244215311231, "rms_gain_error_db": -0.21713575514860267, "reference_mean_power": 0.0004089931043607793, "output_mean_power": 0.0003890473041994198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 4.3909884369057556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019183671841687967, "output_power": 0.0001798634799081363, "reference_active": true, "delta_db": -0.27988745052179337}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4527862162762071e-05, "output_power": 1.540846798066017e-05, "reference_active": true, "delta_db": 0.2555774951883276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.245905579917347e-06, "output_power": 5.595718763478809e-06, "reference_active": true, "delta_db": -1.683824785252419}], "active_band_count": 3, "active_band_mae_db": 0.73976324365418}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 8.815867833009974e-05, "log_rms_mae_db": 0.9903012094506009, "rms_envelope_nrmse": 0.09825952495238735, "envelope_correlation": 0.9936051541401144, "rms_gain_error_db": -0.2171371664055725, "reference_mean_power": 0.0004120037826983822, "output_mean_power": 0.0003919110302428401}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 4.2438330741066236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018201492488703638, "output_power": 0.00016916339194374228, "reference_active": true, "delta_db": -0.31800616063554193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.209318154228534e-06, "output_power": 8.379996140286535e-06, "reference_active": true, "delta_db": 0.6534962681537025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163481758524237e-07, "output_power": 2.659167313714896e-07, "reference_active": true, "delta_db": -0.7541967072345543}], "active_band_count": 3, "active_band_mae_db": 0.5752330453412663}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 8.95305280668978e-05, "log_rms_mae_db": 1.6904257206359092, "rms_envelope_nrmse": 0.1557949606753236, "envelope_correlation": 0.9853698355851638, "rms_gain_error_db": -0.05485804610056355, "reference_mean_power": 8.45768795231113e-05, "output_mean_power": 8.351526300686285e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 3.626876619728336e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.450635181337156e-05, "output_power": 1.4159297127298337e-05, "reference_active": true, "delta_db": -0.1051651057123776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.817293180171653e-06, "output_power": 8.130075983310952e-06, "reference_active": true, "delta_db": 0.17038204443524407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650085525655336e-06, "output_power": 1.7960597939650566e-06, "reference_active": true, "delta_db": 0.5980407574213178}], "active_band_count": 3, "active_band_mae_db": 0.29119596918964646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 5.6185612647784654e-05, "log_rms_mae_db": 1.128425130484682, "rms_envelope_nrmse": 0.08825566542858328, "envelope_correlation": 0.9951637444988418, "rms_gain_error_db": -0.05485944135061813, "reference_mean_power": 8.519942751057583e-05, "output_mean_power": 8.412996968831499e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 3.342409970638096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5109841043230135e-05, "output_power": 1.517775986677329e-05, "reference_active": true, "delta_db": 0.019477818538900134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.127479817162529e-06, "output_power": 4.450431764626828e-06, "reference_active": true, "delta_db": 0.32717188391802604}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2069317938633786e-07, "output_power": 1.225202983384849e-07, "reference_active": true, "delta_db": 0.06525317709915518}], "active_band_count": 3, "active_band_mae_db": 0.13730095985202712}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8850524516905915, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "128067b5ab009ac74c2c86211f79b73d46ca3bc3a0d881816f13a9c4328fc58d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9880453342707295, "standard_deviation": 0.0017589122785741935, "minimum": 0.9850244215311231, "maximum": 0.9904783919846593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5827817923337248, "standard_deviation": 0.033175537418514915, "minimum": 1.5349138299537894, "maximum": 1.643182530932935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.713718083772924, "standard_deviation": 0.21546315601210136, "minimum": 0.4658364788979008, "maximum": 1.0062290999807786, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13814471490337987, "standard_deviation": 0.010422451699547275, "minimum": 0.12096462369670799, "maximum": 0.15225176583277567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963599853023034, "standard_deviation": 0.0012057032253198275, "minimum": 0.9936051541401144, "maximum": 0.9973630130221516, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0224064466282206, "standard_deviation": 0.03732775490873368, "minimum": 0.9828057195185541, "maximum": 1.071605706460852, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5846859912420386, "standard_deviation": 0.19519324264695506, "minimum": 0.36851472570069327, "maximum": 0.9131706441673995, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08327933642530347, "standard_deviation": 0.010855668394223798, "minimum": 0.06638259810160702, "maximum": 0.09890922831968792, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9743951450705208, "standard_deviation": 0.008710767056795621, "minimum": 0.9634679094456785, "maximum": 0.9853698355851638, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6824642096998712, "standard_deviation": 0.047518153071850554, "minimum": 1.6166182530433422, "maximum": 1.7497296040776928, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5744169169760023, "standard_deviation": 0.38613151790152295, "minimum": 0.26177817717562396, "maximum": 1.4327519343499773, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21375940053316583, "standard_deviation": 0.04416820096915378, "minimum": 0.1557949606753236, "maximum": 0.28470710375642616, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9909489870495491, "standard_deviation": 0.004434574195224524, "minimum": 0.9829418220820169, "maximum": 0.9951637444988418, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.092928843073013, "standard_deviation": 0.04626215065626362, "minimum": 1.0176654493784139, "maximum": 1.1595651682022532, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6659111920156257, "standard_deviation": 0.5222402856583425, "minimum": 0.13730095985202712, "maximum": 1.8106058173084085, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13085398891211353, "standard_deviation": 0.04493874524058973, "minimum": 0.08825566542858328, "maximum": 0.21817761871888688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9951435477766525, "standard_deviation": 0.0007638755515226347, "minimum": 0.9940746327819716, "maximum": 0.9963565052114182, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6684946950207293, "standard_deviation": 0.030257862546476468, "minimum": 1.6190297246401981, "maximum": 1.7090681133769654, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7594039999323746, "standard_deviation": 0.2523296533169833, "minimum": 0.2957080078652976, "maximum": 1.078517567328657, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08654391845027702, "standard_deviation": 0.006901580840773711, "minimum": 0.07514874799139741, "maximum": 0.09503825552901685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986280845019697, "standard_deviation": 0.000269638054046691, "minimum": 0.9982057627991883, "maximum": 0.999012013011323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.958282646094308, "standard_deviation": 0.08748153086169798, "minimum": 0.853988963390062, "maximum": 1.1393872159446985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9069315333533653, "standard_deviation": 0.2930492353915913, "minimum": 0.36329471194586976, "maximum": 1.3850321795105325, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04694526270071267, "standard_deviation": 0.005320718135860293, "minimum": 0.039991052520364494, "maximum": 0.05635353366489691, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17849575285851083, "standard_deviation": 0.06457912567223427, "minimum": 0.09744581994541114, "maximum": 0.26622063422377606, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.650299256038762, "standard_deviation": 0.0038296427685224964, "minimum": 5.646274134396172, "maximum": 5.659248678812304, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4942517731089921, "standard_deviation": 0.03481645488711685, "minimum": 0.45723972788115363, "maximum": 0.5568621427350434, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0781059123287648, "standard_deviation": 0.0397445635207338, "minimum": -0.14870058853217286, "maximum": -0.03075047719531154, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9963839015569594, "standard_deviation": 0.0007094023400040185, "minimum": 0.9952909366574785, "maximum": 0.9974564553181182, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.142139597107922, "standard_deviation": 0.021885012520283843, "minimum": 1.1100349352366141, "maximum": 1.1719460806790771, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3205516611185635, "standard_deviation": 0.15926756244492257, "minimum": 0.12173575754805628, "maximum": 0.5978296928150358, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06743660995380621, "standard_deviation": 0.0065497373248156856, "minimum": 0.05681972463707614, "maximum": 0.0766941910266784, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9989011721453129, "standard_deviation": 0.0003635727899823218, "minimum": 0.9982858625704393, "maximum": 0.9994292468634565, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5949833318612543, "standard_deviation": 0.03987276082340644, "minimum": 0.5296141222136963, "maximum": 0.648974220252245, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5062515916326819, "standard_deviation": 0.221829136270984, "minimum": 0.24440960539958323, "maximum": 0.8172150494449469, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0371176871590262, "standard_deviation": 0.006090381417824649, "minimum": 0.02749099101387179, "maximum": 0.04539133566237327, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0049", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.2439448833465576, "predicted_mos": 3.2439448833465576, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "53a84078099a42e2c281cacbc663e2fb1273d5bd67ab137e52564342ef811327", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.741926500108093, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1151317012558581, "全帯域_10ms_rms_envelope_nrmse": 0.06931052905136285, "全帯域_10ms_envelope_correlation": 0.9964561520040263, "全帯域_10ms_modulation_mae_db": 0.5449277823903543, "全帯域_40ms_log_rms_mae_db": 0.5434667584093794, "全帯域_40ms_rms_envelope_nrmse": 0.03957365894089186, "全帯域_40ms_envelope_correlation": 0.9990596362536371, "全帯域_40ms_modulation_mae_db": 0.46466400446404127, "300–1000 Hz_10ms_log_rms_mae_db": 1.6295707518279214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0970505755888776, "300–1000 Hz_10ms_envelope_correlation": 0.9944928521856335, "300–1000 Hz_10ms_modulation_mae_db": 0.8802960654637219, "300–1000 Hz_40ms_log_rms_mae_db": 0.8496258329313819, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06142529562409952, "300–1000 Hz_40ms_envelope_correlation": 0.9982725571308847, "300–1000 Hz_40ms_modulation_mae_db": 0.7332804719542759, "1000–3000 Hz_10ms_log_rms_mae_db": 1.562871078671309, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12293505763378264, "1000–3000 Hz_10ms_envelope_correlation": 0.9911507676614472, "1000–3000 Hz_10ms_modulation_mae_db": 0.579211659400222, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9554159742305606, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0766098026272461, "1000–3000 Hz_40ms_envelope_correlation": 0.997527921699611, "1000–3000 Hz_40ms_modulation_mae_db": 0.4492324796085511, "3000–6000 Hz_10ms_log_rms_mae_db": 1.677376116441453, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20460167196288992, "3000–6000 Hz_10ms_envelope_correlation": 0.9745689293504916, "3000–6000 Hz_10ms_modulation_mae_db": 0.576637318081935, "3000–6000 Hz_40ms_log_rms_mae_db": 1.060120888641842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09525142289175215, "3000–6000 Hz_40ms_envelope_correlation": 0.9943183740046538, "3000–6000 Hz_40ms_modulation_mae_db": 0.49375833960732624, "mr_stft_db": 5.6046104187672405, "band_mae_db": 0.22500730701135618, "rms_gain_error_db": -0.18303237960794216, "output_peak": 0.4605061752222936}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0859089605517642, "全帯域_10ms_rms_envelope_nrmse": 0.06316420559157569, "全帯域_10ms_envelope_correlation": 0.996835645328587, "全帯域_10ms_modulation_mae_db": 0.23744023180151838, "全帯域_40ms_log_rms_mae_db": 0.5693163631281581, "全帯域_40ms_rms_envelope_nrmse": 0.03544081416551762, "全帯域_40ms_envelope_correlation": 0.998996434339903, "全帯域_40ms_modulation_mae_db": 0.26291512012791607, "300–1000 Hz_10ms_log_rms_mae_db": 1.6199776273194815, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09179561350523578, "300–1000 Hz_10ms_envelope_correlation": 0.9944667363670946, "300–1000 Hz_10ms_modulation_mae_db": 0.5205405135052084, "300–1000 Hz_40ms_log_rms_mae_db": 0.8960674523784479, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05470023440807416, "300–1000 Hz_40ms_envelope_correlation": 0.9979951051367038, "300–1000 Hz_40ms_modulation_mae_db": 0.5808999528602542, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4888761919843414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1436869628076018, "1000–3000 Hz_10ms_envelope_correlation": 0.9878537686226048, "1000–3000 Hz_10ms_modulation_mae_db": 1.0094172228980944, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9698844041210033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09552520498856028, "1000–3000 Hz_40ms_envelope_correlation": 0.9958370906665318, "1000–3000 Hz_40ms_modulation_mae_db": 0.760066208179332, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6326873633330676, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21303893954871025, "3000–6000 Hz_10ms_envelope_correlation": 0.9723255968022625, "3000–6000 Hz_10ms_modulation_mae_db": 1.2092465754741886, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0479311329990497, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11651207938686298, "3000–6000 Hz_40ms_envelope_correlation": 0.9917224835364344, "3000–6000 Hz_40ms_modulation_mae_db": 1.3310200745089913, "mr_stft_db": 5.619372958920309, "band_mae_db": 0.1773935273251179, "rms_gain_error_db": -0.07073002457621712, "output_peak": 0.47106551988775525}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1414049074476897, "全帯域_10ms_rms_envelope_nrmse": 0.05910331977054195, "全帯域_10ms_envelope_correlation": 0.9972290652275151, "全帯域_10ms_modulation_mae_db": 0.09574260054055998, "全帯域_40ms_log_rms_mae_db": 0.5627921788190954, "全帯域_40ms_rms_envelope_nrmse": 0.029323365355178616, "全帯域_40ms_envelope_correlation": 0.9993247995528322, "全帯域_40ms_modulation_mae_db": 0.13951547132133432, "300–1000 Hz_10ms_log_rms_mae_db": 1.5928927613372488, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07520432640472338, "300–1000 Hz_10ms_envelope_correlation": 0.9962978970702383, "300–1000 Hz_10ms_modulation_mae_db": 0.6027473188244702, "300–1000 Hz_40ms_log_rms_mae_db": 0.9142011101731461, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03871539480074628, "300–1000 Hz_40ms_envelope_correlation": 0.9990082404927295, "300–1000 Hz_40ms_modulation_mae_db": 0.7191320058807545, "1000–3000 Hz_10ms_log_rms_mae_db": 1.530166138998256, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1471560817183244, "1000–3000 Hz_10ms_envelope_correlation": 0.987109652366138, "1000–3000 Hz_10ms_modulation_mae_db": 0.592934831723997, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9737974748486581, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0928733100336002, "1000–3000 Hz_40ms_envelope_correlation": 0.9961866107037695, "1000–3000 Hz_40ms_modulation_mae_db": 0.46387004951499294, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6408821049796483, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19049900494573493, "3000–6000 Hz_10ms_envelope_correlation": 0.9779389611844526, "3000–6000 Hz_10ms_modulation_mae_db": 0.5017751562841136, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0074026190789407, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09356702795450252, "3000–6000 Hz_40ms_envelope_correlation": 0.994542485006479, "3000–6000 Hz_40ms_modulation_mae_db": 0.7253675028729818, "mr_stft_db": 5.611075415027824, "band_mae_db": 0.18279731130656432, "rms_gain_error_db": -0.057341022919812364, "output_peak": 0.4730960631164397}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.098982961582503, "全帯域_10ms_rms_envelope_nrmse": 0.06371870437630275, "全帯域_10ms_envelope_correlation": 0.9968070589041355, "全帯域_10ms_modulation_mae_db": 0.272587699313197, "全帯域_40ms_log_rms_mae_db": 0.5733189318257809, "全帯域_40ms_rms_envelope_nrmse": 0.032579782165524686, "全帯域_40ms_envelope_correlation": 0.9992009736471845, "全帯域_40ms_modulation_mae_db": 0.24430420356338503, "300–1000 Hz_10ms_log_rms_mae_db": 1.557292402038054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08471346895224466, "300–1000 Hz_10ms_envelope_correlation": 0.9953321025572698, "300–1000 Hz_10ms_modulation_mae_db": 0.7373674111707406, "300–1000 Hz_40ms_log_rms_mae_db": 0.8379432271452141, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041517439816183346, "300–1000 Hz_40ms_envelope_correlation": 0.9989178840978935, "300–1000 Hz_40ms_modulation_mae_db": 0.837896490823252, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5787431900243616, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13909784078382514, "1000–3000 Hz_10ms_envelope_correlation": 0.9884725922251942, "1000–3000 Hz_10ms_modulation_mae_db": 0.45231074200103655, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9958916375614452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09581333193999439, "1000–3000 Hz_40ms_envelope_correlation": 0.9954534633440553, "1000–3000 Hz_40ms_modulation_mae_db": 0.45015108342513327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6810295489756542, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20085623131343855, "3000–6000 Hz_10ms_envelope_correlation": 0.9762121867824108, "3000–6000 Hz_10ms_modulation_mae_db": 0.29726308918952404, "3000–6000 Hz_40ms_log_rms_mae_db": 1.076217474424424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1088043802325302, "3000–6000 Hz_40ms_envelope_correlation": 0.9928609058256916, "3000–6000 Hz_40ms_modulation_mae_db": 0.4239310549298427, "mr_stft_db": 5.618191324598993, "band_mae_db": 0.163230389401022, "rms_gain_error_db": -0.08728447359240185, "output_peak": 0.4731622651954874}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0206089363263895, "全帯域_10ms_rms_envelope_nrmse": 0.06406288525931363, "全帯域_10ms_envelope_correlation": 0.9969138282521737, "全帯域_10ms_modulation_mae_db": 0.4711549755047537, "全帯域_40ms_log_rms_mae_db": 0.4988276816572331, "全帯域_40ms_rms_envelope_nrmse": 0.03603120134393363, "全帯域_40ms_envelope_correlation": 0.9991617543550967, "全帯域_40ms_modulation_mae_db": 0.5272814570744105, "300–1000 Hz_10ms_log_rms_mae_db": 1.6370247750687115, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08507304807958047, "300–1000 Hz_10ms_envelope_correlation": 0.9954074223494804, "300–1000 Hz_10ms_modulation_mae_db": 0.8794855138281422, "300–1000 Hz_40ms_log_rms_mae_db": 0.9372828380209716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047709950019962304, "300–1000 Hz_40ms_envelope_correlation": 0.9986858530193333, "300–1000 Hz_40ms_modulation_mae_db": 0.875117419017588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5583357756576872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14153338533292448, "1000–3000 Hz_10ms_envelope_correlation": 0.9872914198784015, "1000–3000 Hz_10ms_modulation_mae_db": 0.4610334933932923, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9770615350992664, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08646830120771104, "1000–3000 Hz_40ms_envelope_correlation": 0.9958336460180781, "1000–3000 Hz_40ms_modulation_mae_db": 0.39585845944572995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6719169827357738, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1776737839630987, "3000–6000 Hz_10ms_envelope_correlation": 0.9822729740666882, "3000–6000 Hz_10ms_modulation_mae_db": 0.7362751537909359, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0961604053762048, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10442845400725564, "3000–6000 Hz_40ms_envelope_correlation": 0.9953652900256984, "3000–6000 Hz_40ms_modulation_mae_db": 0.8526005702539617, "mr_stft_db": 5.609646512682214, "band_mae_db": 0.26829297859094836, "rms_gain_error_db": -0.15514703261971946, "output_peak": 0.5205706063060321}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.109981214292303, "全帯域_10ms_rms_envelope_nrmse": 0.06285905663101356, "全帯域_10ms_envelope_correlation": 0.996846084123519, "全帯域_10ms_modulation_mae_db": 0.6294106009470114, "全帯域_40ms_log_rms_mae_db": 0.5633673191762982, "全帯域_40ms_rms_envelope_nrmse": 0.030315089770438564, "全帯域_40ms_envelope_correlation": 0.9992518688206733, "全帯域_40ms_modulation_mae_db": 0.7507827295177939, "300–1000 Hz_10ms_log_rms_mae_db": 1.635130506985385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.083670618281441, "300–1000 Hz_10ms_envelope_correlation": 0.9954130708451597, "300–1000 Hz_10ms_modulation_mae_db": 0.9678674982981864, "300–1000 Hz_40ms_log_rms_mae_db": 0.9000807766485471, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0418080325289819, "300–1000 Hz_40ms_envelope_correlation": 0.9988530777700824, "300–1000 Hz_40ms_modulation_mae_db": 1.0976574493675926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5935182760263473, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1264577571858096, "1000–3000 Hz_10ms_envelope_correlation": 0.9893650644318095, "1000–3000 Hz_10ms_modulation_mae_db": 0.9327275631439775, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9797368641474916, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.071417627728197, "1000–3000 Hz_40ms_envelope_correlation": 0.9964869584739167, "1000–3000 Hz_40ms_modulation_mae_db": 0.5409777300277129, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6290811369392277, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22154355732902214, "3000–6000 Hz_10ms_envelope_correlation": 0.9715632822602105, "3000–6000 Hz_10ms_modulation_mae_db": 1.0114072085620542, "3000–6000 Hz_40ms_log_rms_mae_db": 0.993016525235042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11288851058741908, "3000–6000 Hz_40ms_envelope_correlation": 0.995045884102259, "3000–6000 Hz_40ms_modulation_mae_db": 1.018871429516736, "mr_stft_db": 5.622475256197454, "band_mae_db": 0.2014292623196157, "rms_gain_error_db": -0.03911903383919147, "output_peak": 0.5188037637958631}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1416180599307089, "全帯域_10ms_rms_envelope_nrmse": 0.06387266575974729, "全帯域_10ms_envelope_correlation": 0.9969162821309638, "全帯域_10ms_modulation_mae_db": 0.29539773331298885, "全帯域_40ms_log_rms_mae_db": 0.5567130280933363, "全帯域_40ms_rms_envelope_nrmse": 0.03847952699337826, "全帯域_40ms_envelope_correlation": 0.9989874435963628, "全帯域_40ms_modulation_mae_db": 0.21205387393351782, "300–1000 Hz_10ms_log_rms_mae_db": 1.688739027183284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08127660493217431, "300–1000 Hz_10ms_envelope_correlation": 0.9959030908796301, "300–1000 Hz_10ms_modulation_mae_db": 0.7208512516478626, "300–1000 Hz_40ms_log_rms_mae_db": 0.9184906124636395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0460913819262926, "300–1000 Hz_40ms_envelope_correlation": 0.998871995804697, "300–1000 Hz_40ms_modulation_mae_db": 0.6075920271674402, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5221057866769887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1487465656125878, "1000–3000 Hz_10ms_envelope_correlation": 0.9881928139212801, "1000–3000 Hz_10ms_modulation_mae_db": 0.8223505926320392, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9186365290368639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10128571432033415, "1000–3000 Hz_40ms_envelope_correlation": 0.9966191396623554, "1000–3000 Hz_40ms_modulation_mae_db": 0.6151225464791841, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6042214435886102, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20176371580298175, "3000–6000 Hz_10ms_envelope_correlation": 0.9755191047377244, "3000–6000 Hz_10ms_modulation_mae_db": 0.3063227400427785, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0607222445097984, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15364667362330375, "3000–6000 Hz_40ms_envelope_correlation": 0.9853091753287387, "3000–6000 Hz_40ms_modulation_mae_db": 0.14490875706744427, "mr_stft_db": 5.617261818416423, "band_mae_db": 0.26420985736836466, "rms_gain_error_db": -0.14860538971959097, "output_peak": 0.4864572635692901}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1092306859265058, "全帯域_10ms_rms_envelope_nrmse": 0.06331452524974136, "全帯域_10ms_envelope_correlation": 0.9967973219491101, "全帯域_10ms_modulation_mae_db": 0.1452928342455211, "全帯域_40ms_log_rms_mae_db": 0.5537075649089984, "全帯域_40ms_rms_envelope_nrmse": 0.03455066650975363, "全帯域_40ms_envelope_correlation": 0.9990177690125038, "全帯域_40ms_modulation_mae_db": 0.04153288354118541, "300–1000 Hz_10ms_log_rms_mae_db": 1.6030016312385047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0796304145046442, "300–1000 Hz_10ms_envelope_correlation": 0.995833769362064, "300–1000 Hz_10ms_modulation_mae_db": 0.32726563379902923, "300–1000 Hz_40ms_log_rms_mae_db": 0.9181409429257111, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0428409957324578, "300–1000 Hz_40ms_envelope_correlation": 0.9987668272722123, "300–1000 Hz_40ms_modulation_mae_db": 0.5080388676114908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5140102076085278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12732941892853258, "1000–3000 Hz_10ms_envelope_correlation": 0.9900653764814643, "1000–3000 Hz_10ms_modulation_mae_db": 1.2278620706647947, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9719029949606105, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0750716063984668, "1000–3000 Hz_40ms_envelope_correlation": 0.9973534370258517, "1000–3000 Hz_40ms_modulation_mae_db": 1.2018130470451778, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6332358102993607, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17477306509253962, "3000–6000 Hz_10ms_envelope_correlation": 0.9826386297550349, "3000–6000 Hz_10ms_modulation_mae_db": 1.1249660071313434, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9943342726445438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10153936151400374, "3000–6000 Hz_40ms_envelope_correlation": 0.9953778532310403, "3000–6000 Hz_40ms_modulation_mae_db": 1.31202168222318, "mr_stft_db": 5.61223115903224, "band_mae_db": 0.17033489584624117, "rms_gain_error_db": -0.04321187316629545, "output_peak": 0.5191400907842805}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65760, "output_samples": 65760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.756353889628668, "spectral_convergence": 0.22901663898538044, "spectral_convergence_status": "ok", "amplitude_floor": 1.6529112798953643e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.412440339518027, "spectral_convergence": 0.22057943959159027, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394789070711673e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.645037027155026, "spectral_convergence": 0.21685111165456983, "spectral_convergence_status": "ok", "amplitude_floor": 8.202976886987361e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008081818875357184, "output_power": 0.00083192805624699, "delta_db": 0.12576657770093383, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008371388318692298, "output_power": 0.0008300682276684709, "delta_db": -0.03683696971810892, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0012771265272884344, "output_power": 0.001163311657821607, "delta_db": -0.40537845469982187, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003871568194673651, "output_power": 0.00034778174830078887, "delta_db": -0.465801270785436, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001322399975596578, "output_power": 0.00012644535584647174, "delta_db": -0.19459949578624092, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.960478257270517e-05, "output_power": 1.906320520200611e-05, "delta_db": -0.12166107337759557, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22500730701135618, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.18303237960794216, "pitch": null, "output_rms_dbfs": -24.79045320021943, "output_peak": 0.4605061752222936, "reference_rms_dbfs": -24.60742082061149, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6046104187672405}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00020667206862955213, "log_rms_mae_db": 1.1151317012558581, "rms_envelope_nrmse": 0.06931052905136285, "envelope_correlation": 0.9964561520040263, "rms_gain_error_db": -0.18303209341087728, "reference_mean_power": 0.003465660170380573, "output_mean_power": 0.0033226359969230753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 2.300680394398606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009732960931433549, "output_power": 0.0009070872321269255, "reference_active": true, "delta_db": -0.30595926105542315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.170752162168669e-05, "output_power": 6.433970989341003e-05, "reference_active": true, "delta_db": -1.0378293919701922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.07791133597542e-05, "output_power": 1.9432449284909894e-05, "reference_active": true, "delta_db": -0.2909946941454478}], "active_band_count": 3, "active_band_mae_db": 0.5449277823903543}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00018941444655877122, "log_rms_mae_db": 0.5434667584093794, "rms_envelope_nrmse": 0.03957365894089186, "envelope_correlation": 0.9990596362536371, "rms_gain_error_db": -0.18302930126496714, "reference_mean_power": 0.003491154294575285, "output_mean_power": 0.0033470801571433384}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 2.1871865789273095e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008996086963166702, "output_power": 0.0008391391556369573, "reference_active": true, "delta_db": -0.30219658540661826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8292682982878605e-05, "output_power": 3.0410879689337035e-05, "reference_active": true, "delta_db": -1.0008681325111732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1507764875014336e-06, "output_power": 1.126933399063622e-06, "reference_active": true, "delta_db": -0.09092729547433225}], "active_band_count": 3, "active_band_mae_db": 0.46466400446404127}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.0001875778806757321, "log_rms_mae_db": 1.6295707518279214, "rms_envelope_nrmse": 0.0970505755888776, "envelope_correlation": 0.9944928521856335, "rms_gain_error_db": -0.311708777631492, "reference_mean_power": 0.0016622862050937872, "output_mean_power": 0.0015471589110116735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 1.477365390744301e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007590270471846978, "output_power": 0.0006937487464687151, "reference_active": true, "delta_db": -0.39055040422485654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.577918818080543e-05, "output_power": 3.9739358159795936e-05, "reference_active": true, "delta_db": -1.472513407520907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4662085193890421e-05, "output_power": 1.2257821707938246e-05, "reference_active": true, "delta_db": -0.7778243846454019}], "active_band_count": 3, "active_band_mae_db": 0.8802960654637219}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00016891518288067464, "log_rms_mae_db": 0.8496258329313819, "rms_envelope_nrmse": 0.06142529562409952, "envelope_correlation": 0.9982725571308847, "rms_gain_error_db": -0.3117121847977412, "reference_mean_power": 0.001674521427963224, "output_mean_power": 0.0015585455191146039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 1.422523294383883e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007062170888212691, "output_power": 0.0006458754061198356, "reference_active": true, "delta_db": -0.3878947464317991}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.54990695717255e-05, "output_power": 1.86665846559624e-05, "reference_active": true, "delta_db": -1.354594696359349}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.502016839919997e-07, "output_power": 8.552254149938794e-07, "reference_active": true, "delta_db": -0.4573519730716799}], "active_band_count": 3, "active_band_mae_db": 0.7332804719542759}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00010800393678736649, "log_rms_mae_db": 1.562871078671309, "rms_envelope_nrmse": 0.12293505763378264, "envelope_correlation": 0.9911507676614472, "rms_gain_error_db": -0.43818581446979027, "reference_mean_power": 0.0004089931043607793, "output_mean_power": 0.0003697408237028163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 4.3909884369057556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019183671841687967, "output_power": 0.00017070977116528973, "reference_active": true, "delta_db": -0.5067335654632655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4527862162762071e-05, "output_power": 1.2294009246951422e-05, "reference_active": true, "delta_db": -0.7250817526373708}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.245905579917347e-06, "output_power": 7.339329597025505e-06, "reference_active": true, "delta_db": -0.5058196601000295}], "active_band_count": 3, "active_band_mae_db": 0.579211659400222}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 8.815867833009974e-05, "log_rms_mae_db": 0.9554159742305606, "rms_envelope_nrmse": 0.0766098026272461, "envelope_correlation": 0.997527921699611, "rms_gain_error_db": -0.4381870090516276, "reference_mean_power": 0.0004120037826983822, "output_mean_power": 0.00037246245586488875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 4.2438330741066236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018201492488703638, "output_power": 0.00016274409314245996, "reference_active": true, "delta_db": -0.4860176614294067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.209318154228534e-06, "output_power": 6.158129521798644e-06, "reference_active": true, "delta_db": -0.6844537275657094}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163481758524237e-07, "output_power": 3.0369855780783714e-07, "reference_active": true, "delta_db": -0.17722604983053714}], "active_band_count": 3, "active_band_mae_db": 0.4492324796085511}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 8.95305280668978e-05, "log_rms_mae_db": 1.677376116441453, "rms_envelope_nrmse": 0.20460167196288992, "envelope_correlation": 0.9745689293504916, "rms_gain_error_db": -0.09564490572244727, "reference_mean_power": 8.45768795231113e-05, "output_mean_power": 8.27345991676564e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 3.626876619728336e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.450635181337156e-05, "output_power": 1.4079234819874791e-05, "reference_active": true, "delta_db": -0.12979153569478843}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.817293180171653e-06, "output_power": 6.535185320841607e-06, "reference_active": true, "delta_db": -0.7779849251346115}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650085525655336e-06, "output_power": 1.8911686702655447e-06, "reference_active": true, "delta_db": 0.822135493416405}], "active_band_count": 3, "active_band_mae_db": 0.576637318081935}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 5.6185612647784654e-05, "log_rms_mae_db": 1.060120888641842, "rms_envelope_nrmse": 0.09525142289175215, "envelope_correlation": 0.9943183740046538, "rms_gain_error_db": -0.09564637449855738, "reference_mean_power": 8.519942751057583e-05, "output_mean_power": 8.334355843119689e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 3.342409970638096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5109841043230135e-05, "output_power": 1.4691191558696733e-05, "reference_active": true, "delta_db": -0.12202874001467726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.127479817162529e-06, "output_power": 3.607098746071887e-06, "reference_active": true, "delta_db": -0.5852692679421825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2069317938633786e-07, "output_power": 1.44238244833899e-07, "reference_active": true, "delta_db": 0.7739770108651189}], "active_band_count": 3, "active_band_mae_db": 0.49375833960732624}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8794164674168261, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b477bb830ffaa74e0ecc34a60479c0e917be734f49854bace536f80673ef9841", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9886876819485424, "standard_deviation": 0.0014066157462874122, "minimum": 0.987109652366138, "maximum": 0.9911507676614472, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5435783307059774, "standard_deviation": 0.03551279989112258, "minimum": 1.4888761919843414, "maximum": 1.5935182760263473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7597310219821818, "standard_deviation": 0.28270998340835324, "minimum": 0.45231074200103655, "maximum": 1.2278620706647947, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13711788375042355, "standard_deviation": 0.010093606481549848, "minimum": 0.12293505763378264, "maximum": 0.1487465656125878, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9964122834492712, "standard_deviation": 0.0007382137523248385, "minimum": 0.9954534633440553, "maximum": 0.997527921699611, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9677909267507374, "standard_deviation": 0.022819052767718343, "minimum": 0.9186365290368639, "maximum": 0.9958916375614452, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6096364504657268, "standard_deviation": 0.26623304505139006, "minimum": 0.39585845944572995, "maximum": 1.2018130470451778, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08688311240551375, "standard_deviation": 0.011225976554893855, "minimum": 0.071417627728197, "maximum": 0.10128571432033415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9766299581174095, "standard_deviation": 0.0041320411966030155, "minimum": 0.9715632822602105, "maximum": 0.9826386297550349, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6463038134115995, "standard_deviation": 0.027465617321784774, "minimum": 1.6042214435886102, "maximum": 1.6810295489756542, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7204866560696092, "standard_deviation": 0.3598712344821196, "minimum": 0.29726308918952404, "maximum": 1.2092465754741886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19809374624480197, "standard_deviation": 0.01627976247910612, "minimum": 0.17477306509253962, "maximum": 0.22154355732902214, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9930678063826244, "standard_deviation": 0.003387270132404657, "minimum": 0.9853091753287387, "maximum": 0.9953778532310403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0419881953637307, "standard_deviation": 0.03907251950994169, "minimum": 0.993016525235042, "maximum": 1.0961604053762048, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.787809926372558, "standard_deviation": 0.4248626020404663, "minimum": 0.14490875706744427, "maximum": 1.3310200745089913, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11082973877470376, "standard_deviation": 0.0190559034354713, "minimum": 0.09356702795450252, "maximum": 0.15364667362330375, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9953933677020713, "standard_deviation": 0.000648719365803162, "minimum": 0.9944667363670946, "maximum": 0.9962978970702383, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.620453685374824, "standard_deviation": 0.038340154229217535, "minimum": 1.557292402038054, "maximum": 1.688739027183284, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7045526508171702, "standard_deviation": 0.21338586834103537, "minimum": 0.32726563379902923, "maximum": 0.9678674982981864, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08480183378111517, "standard_deviation": 0.006882572280337381, "minimum": 0.07520432640472338, "maximum": 0.0970505755888776, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986714425905671, "standard_deviation": 0.0003531950155294364, "minimum": 0.9979951051367038, "maximum": 0.9990082404927295, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8964790990858824, "standard_deviation": 0.03498484442329735, "minimum": 0.8379432271452141, "maximum": 0.9372828380209716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7449518355853311, "standard_deviation": 0.1899978999750393, "minimum": 0.5080388676114908, "maximum": 1.0976574493675926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.046851090607099743, "standard_deviation": 0.007667016366530562, "minimum": 0.03871539480074628, "maximum": 0.06142529562409952, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20658694114615378, "standard_deviation": 0.04158665707490822, "minimum": 0.163230389401022, "maximum": 0.26829297859094836, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.614358107955337, "standard_deviation": 0.005936310412172344, "minimum": 5.6046104187672405, "maximum": 5.622475256197454, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4903502184846802, "standard_deviation": 0.02513559246456622, "minimum": 0.4605061752222936, "maximum": 0.5205706063060321, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09805890375514636, "standard_deviation": 0.05610895030648654, "minimum": -0.18303237960794216, "maximum": -0.03911903383919147, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9968501797400038, "standard_deviation": 0.00021115806682337434, "minimum": 0.9964561520040263, "maximum": 0.9972290652275151, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.102858428414215, "standard_deviation": 0.03835435476220429, "minimum": 1.0206089363263895, "maximum": 1.1416180599307089, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3364943072569881, "standard_deviation": 0.19178933554441893, "minimum": 0.09574260054055998, "maximum": 0.6294106009470114, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0636757364611999, "standard_deviation": 0.002775080354674265, "minimum": 0.05910331977054195, "maximum": 0.06931052905136285, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991250849472741, "standard_deviation": 0.00012781693163242217, "minimum": 0.9989874435963628, "maximum": 0.9993247995528322, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.552688728252285, "standard_deviation": 0.023668802987323663, "minimum": 0.4988276816572331, "maximum": 0.5733189318257809, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.33038121794294806, "standard_deviation": 0.23279800639782838, "minimum": 0.04153288354118541, "maximum": 0.7507827295177939, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.034536763155577105, "standard_deviation": 0.0036410039874199863, "minimum": 0.029323365355178616, "maximum": 0.03957365894089186, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8455-210777-0060", "language": "en", "style": "normal", "speaker": "librispeech_8455", "source_utterance_id": "8455-210777-0060", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8455-210777-0060.wav", "source_file_sha256": "db1a20b3dbf4771d5b641e8b2e6f58d492d8d6963e16a5b96155b89e45bfc0bb", "decoded_pcm_sha256": "71c64b069cc8f43d09cfadd1defcda3beccfdf73919c8518a13182fa2c801f93", "original_source_sha256": "ef57db07976e02027d19da092bac4747c7928a93199441010b967c12ce13e724", "corpus_entry_id": "librispeech_8455-210777-0060"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.653318405151367, "predicted_mos": 3.653318405151367, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "71c3ff8ad7d01f0df67fe46cab8bf2c04b190bcf82e266a5d48e996f38896307", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.313606667001295, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8455-210777-0060", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "71c3ff8ad7d01f0df67fe46cab8bf2c04b190bcf82e266a5d48e996f38896307", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "9b6304a33d078fd527310e09782e4424d1a63e37c4d42866c9aa87d2920ab687"}, "rows": [{"case": "librispeech_8455-210777-0060", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.737190008163452, "predicted_mos": 2.737190008163452, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9428b2e0f322bb5ae8ecf958d3a479a7960f81f00762ba1fc00929f113a30cb3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1463426490008715, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8455-210777-0060", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9428b2e0f322bb5ae8ecf958d3a479a7960f81f00762ba1fc00929f113a30cb3", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9104510624729654, "全帯域_10ms_rms_envelope_nrmse": 0.08417083498794058, "全帯域_10ms_envelope_correlation": 0.9927287535964421, "全帯域_10ms_modulation_mae_db": 0.24150419721784858, "全帯域_40ms_log_rms_mae_db": 0.4935453792001787, "全帯域_40ms_rms_envelope_nrmse": 0.04292148059996806, "全帯域_40ms_envelope_correlation": 0.9983118757098035, "全帯域_40ms_modulation_mae_db": 0.3115507140285802, "300–1000 Hz_10ms_log_rms_mae_db": 1.5480527157859494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1314913512237657, "300–1000 Hz_10ms_envelope_correlation": 0.9873303306916109, "300–1000 Hz_10ms_modulation_mae_db": 0.4339498537670659, "300–1000 Hz_40ms_log_rms_mae_db": 0.8751593033650278, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07524844510848339, "300–1000 Hz_40ms_envelope_correlation": 0.9965125547395393, "300–1000 Hz_40ms_modulation_mae_db": 0.714812813644722, "1000–3000 Hz_10ms_log_rms_mae_db": 1.634539873182801, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17035814256427675, "1000–3000 Hz_10ms_envelope_correlation": 0.9790386138556463, "1000–3000 Hz_10ms_modulation_mae_db": 0.20390822793071417, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9915847457195485, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08782310382230694, "1000–3000 Hz_40ms_envelope_correlation": 0.9944662214049005, "1000–3000 Hz_40ms_modulation_mae_db": 0.16847246881527314, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9635019872804704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2518536938997106, "3000–6000 Hz_10ms_envelope_correlation": 0.9573816108472117, "3000–6000 Hz_10ms_modulation_mae_db": 0.8380580414290676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5599278096235032, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17733950848467023, "3000–6000 Hz_40ms_envelope_correlation": 0.9775611705170473, "3000–6000 Hz_40ms_modulation_mae_db": 0.962610988137008, "mr_stft_db": 5.8527036063182445, "band_mae_db": 0.21470366641864044, "rms_gain_error_db": -0.16766133504739653, "output_peak": 0.7019069019072526}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9217640843247793, "全帯域_10ms_rms_envelope_nrmse": 0.07900176520731059, "全帯域_10ms_envelope_correlation": 0.99344280806324, "全帯域_10ms_modulation_mae_db": 0.20384059110897215, "全帯域_40ms_log_rms_mae_db": 0.47849154787937276, "全帯域_40ms_rms_envelope_nrmse": 0.03989628435665014, "全帯域_40ms_envelope_correlation": 0.9983082250023024, "全帯域_40ms_modulation_mae_db": 0.13210562976273674, "300–1000 Hz_10ms_log_rms_mae_db": 1.4668154510599412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1193905440871195, "300–1000 Hz_10ms_envelope_correlation": 0.9891107136055859, "300–1000 Hz_10ms_modulation_mae_db": 0.4942394182792882, "300–1000 Hz_40ms_log_rms_mae_db": 0.8661277236749606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07338178292299512, "300–1000 Hz_40ms_envelope_correlation": 0.9958846886110957, "300–1000 Hz_40ms_modulation_mae_db": 0.3524147806977333, "1000–3000 Hz_10ms_log_rms_mae_db": 1.652842048408045, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1725500856945796, "1000–3000 Hz_10ms_envelope_correlation": 0.978526751339506, "1000–3000 Hz_10ms_modulation_mae_db": 0.8538833858065651, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9863178326030879, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09381717198474084, "1000–3000 Hz_40ms_envelope_correlation": 0.9936749588642881, "1000–3000 Hz_40ms_modulation_mae_db": 1.1967305543818385, "3000–6000 Hz_10ms_log_rms_mae_db": 1.975265430338827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25596470594672766, "3000–6000 Hz_10ms_envelope_correlation": 0.9547790407494773, "3000–6000 Hz_10ms_modulation_mae_db": 0.8099599385086532, "3000–6000 Hz_40ms_log_rms_mae_db": 1.534524136830691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16880490577179932, "3000–6000 Hz_40ms_envelope_correlation": 0.9789140814151296, "3000–6000 Hz_40ms_modulation_mae_db": 0.8624039233356348, "mr_stft_db": 5.8534296264607875, "band_mae_db": 0.23042506487002487, "rms_gain_error_db": -0.09837236867762664, "output_peak": 0.6908241884614985}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9342746777743325, "全帯域_10ms_rms_envelope_nrmse": 0.0668076500778224, "全帯域_10ms_envelope_correlation": 0.9953010939231844, "全帯域_10ms_modulation_mae_db": 0.09795061461584613, "全帯域_40ms_log_rms_mae_db": 0.4954596205528988, "全帯域_40ms_rms_envelope_nrmse": 0.03465104885682261, "全帯域_40ms_envelope_correlation": 0.9986958553527407, "全帯域_40ms_modulation_mae_db": 0.19754597551976882, "300–1000 Hz_10ms_log_rms_mae_db": 1.3950782516931022, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09547287718248955, "300–1000 Hz_10ms_envelope_correlation": 0.9929882239474072, "300–1000 Hz_10ms_modulation_mae_db": 0.2600430799567423, "300–1000 Hz_40ms_log_rms_mae_db": 0.7435786007836517, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049294782853074745, "300–1000 Hz_40ms_envelope_correlation": 0.998087722104182, "300–1000 Hz_40ms_modulation_mae_db": 0.1754383383471393, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6803471682241584, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15816873273478102, "1000–3000 Hz_10ms_envelope_correlation": 0.982477096516794, "1000–3000 Hz_10ms_modulation_mae_db": 0.7976875639602629, "1000–3000 Hz_40ms_log_rms_mae_db": 1.048915606319047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07907429957142818, "1000–3000 Hz_40ms_envelope_correlation": 0.9963377762140689, "1000–3000 Hz_40ms_modulation_mae_db": 1.1523129410720703, "3000–6000 Hz_10ms_log_rms_mae_db": 1.943028691539311, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25575223253722695, "3000–6000 Hz_10ms_envelope_correlation": 0.9535350482997269, "3000–6000 Hz_10ms_modulation_mae_db": 0.4358668907473606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5330161243438536, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17969741949419557, "3000–6000 Hz_40ms_envelope_correlation": 0.9755320675129319, "3000–6000 Hz_40ms_modulation_mae_db": 0.3780576416771521, "mr_stft_db": 5.860288248214057, "band_mae_db": 0.235193666750202, "rms_gain_error_db": -0.07242587795252198, "output_peak": 0.8333885168337546}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.912950143917812, "全帯域_10ms_rms_envelope_nrmse": 0.07937337162672113, "全帯域_10ms_envelope_correlation": 0.9933328833661074, "全帯域_10ms_modulation_mae_db": 0.4920705081945573, "全帯域_40ms_log_rms_mae_db": 0.4664781891528198, "全帯域_40ms_rms_envelope_nrmse": 0.03748389003879646, "全帯域_40ms_envelope_correlation": 0.9984474247693597, "全帯域_40ms_modulation_mae_db": 0.33435388393818516, "300–1000 Hz_10ms_log_rms_mae_db": 1.4481390636304414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11845588957089948, "300–1000 Hz_10ms_envelope_correlation": 0.9891930689460435, "300–1000 Hz_10ms_modulation_mae_db": 0.6372410285864354, "300–1000 Hz_40ms_log_rms_mae_db": 0.8079716011379812, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06053246135394734, "300–1000 Hz_40ms_envelope_correlation": 0.99712113224471, "300–1000 Hz_40ms_modulation_mae_db": 0.47315752404048156, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5743321255030793, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15366945379750643, "1000–3000 Hz_10ms_envelope_correlation": 0.9829560262863847, "1000–3000 Hz_10ms_modulation_mae_db": 0.8999102465769898, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9954823725856642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09205451168215682, "1000–3000 Hz_40ms_envelope_correlation": 0.9937047850893612, "1000–3000 Hz_40ms_modulation_mae_db": 0.6754162287953521, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9314332102430403, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26185719804745555, "3000–6000 Hz_10ms_envelope_correlation": 0.9514213995593589, "3000–6000 Hz_10ms_modulation_mae_db": 0.5322211141115227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.543808279954469, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1815456274740106, "3000–6000 Hz_40ms_envelope_correlation": 0.9750038981013571, "3000–6000 Hz_40ms_modulation_mae_db": 0.5159502374902568, "mr_stft_db": 5.857439066625385, "band_mae_db": 0.19989841918405915, "rms_gain_error_db": -0.06506340410029333, "output_peak": 0.7319885083038893}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8949746563067443, "全帯域_10ms_rms_envelope_nrmse": 0.06299066291167033, "全帯域_10ms_envelope_correlation": 0.9958273417840094, "全帯域_10ms_modulation_mae_db": 0.1776054987431892, "全帯域_40ms_log_rms_mae_db": 0.4414101912410563, "全帯域_40ms_rms_envelope_nrmse": 0.030729624869492617, "全帯域_40ms_envelope_correlation": 0.9989777673067683, "全帯域_40ms_modulation_mae_db": 0.20755744971548382, "300–1000 Hz_10ms_log_rms_mae_db": 1.4727790369759497, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09632952499907883, "300–1000 Hz_10ms_envelope_correlation": 0.9928544416646842, "300–1000 Hz_10ms_modulation_mae_db": 0.36257118636892605, "300–1000 Hz_40ms_log_rms_mae_db": 0.8412854530206718, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04827609736393298, "300–1000 Hz_40ms_envelope_correlation": 0.9981458548987645, "300–1000 Hz_40ms_modulation_mae_db": 0.5570850842218035, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5675502027673298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13152099132977138, "1000–3000 Hz_10ms_envelope_correlation": 0.9876614687005666, "1000–3000 Hz_10ms_modulation_mae_db": 0.5535120522313598, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9442902877698737, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07000895273381953, "1000–3000 Hz_40ms_envelope_correlation": 0.9966073490367696, "1000–3000 Hz_40ms_modulation_mae_db": 0.637184314509699, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0393703680057147, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2648853897404669, "3000–6000 Hz_10ms_envelope_correlation": 0.950038451004547, "3000–6000 Hz_10ms_modulation_mae_db": 0.5477565070422079, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5814426740347538, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.181011392544741, "3000–6000 Hz_40ms_envelope_correlation": 0.9751956615231677, "3000–6000 Hz_40ms_modulation_mae_db": 0.4411966590975301, "mr_stft_db": 5.848936624481862, "band_mae_db": 0.21275998131115625, "rms_gain_error_db": -0.06254738384614611, "output_peak": 0.7100982360525459}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9125666779917794, "全帯域_10ms_rms_envelope_nrmse": 0.08429322767948233, "全帯域_10ms_envelope_correlation": 0.9924557221106902, "全帯域_10ms_modulation_mae_db": 0.20184014019044919, "全帯域_40ms_log_rms_mae_db": 0.5061057691600712, "全帯域_40ms_rms_envelope_nrmse": 0.050656036054191195, "全帯域_40ms_envelope_correlation": 0.9970762893330336, "全帯域_40ms_modulation_mae_db": 0.3021714556096721, "300–1000 Hz_10ms_log_rms_mae_db": 1.5275984440191748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13193660576286786, "300–1000 Hz_10ms_envelope_correlation": 0.9865861682869308, "300–1000 Hz_10ms_modulation_mae_db": 0.661108863835658, "300–1000 Hz_40ms_log_rms_mae_db": 0.857204190680265, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07516331284910377, "300–1000 Hz_40ms_envelope_correlation": 0.995405416828735, "300–1000 Hz_40ms_modulation_mae_db": 0.4163709604006116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.630577107560849, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16133990285661023, "1000–3000 Hz_10ms_envelope_correlation": 0.9813886924740357, "1000–3000 Hz_10ms_modulation_mae_db": 0.317932528567356, "1000–3000 Hz_40ms_log_rms_mae_db": 1.030649534525466, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0935745773286473, "1000–3000 Hz_40ms_envelope_correlation": 0.9938967234034712, "1000–3000 Hz_40ms_modulation_mae_db": 0.24907055938306674, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9349231076277316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2567974898855428, "3000–6000 Hz_10ms_envelope_correlation": 0.9529224150571765, "3000–6000 Hz_10ms_modulation_mae_db": 0.9883425773246604, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5150797536825389, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1843931249985307, "3000–6000 Hz_40ms_envelope_correlation": 0.9742954523965223, "3000–6000 Hz_40ms_modulation_mae_db": 0.7535775609620363, "mr_stft_db": 5.855101332202191, "band_mae_db": 0.21149775375114155, "rms_gain_error_db": -0.03212351041791948, "output_peak": 0.9062173915735947}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9272517223096556, "全帯域_10ms_rms_envelope_nrmse": 0.06716791758753964, "全帯域_10ms_envelope_correlation": 0.9953184049277641, "全帯域_10ms_modulation_mae_db": 0.37207610839244837, "全帯域_40ms_log_rms_mae_db": 0.4478947039813065, "全帯域_40ms_rms_envelope_nrmse": 0.03494786405001547, "全帯域_40ms_envelope_correlation": 0.9987640392131651, "全帯域_40ms_modulation_mae_db": 0.5125163231527541, "300–1000 Hz_10ms_log_rms_mae_db": 1.5150439708306922, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09834293489019463, "300–1000 Hz_10ms_envelope_correlation": 0.9927428442339346, "300–1000 Hz_10ms_modulation_mae_db": 0.3408374533081025, "300–1000 Hz_40ms_log_rms_mae_db": 0.7822442671092394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052286722936589446, "300–1000 Hz_40ms_envelope_correlation": 0.998107067901535, "300–1000 Hz_40ms_modulation_mae_db": 0.30435514796656504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6459757220464475, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1650151703456137, "1000–3000 Hz_10ms_envelope_correlation": 0.9804040222105923, "1000–3000 Hz_10ms_modulation_mae_db": 0.4076167582160988, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0101063198423097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09076193768195408, "1000–3000 Hz_40ms_envelope_correlation": 0.9941127432147225, "1000–3000 Hz_40ms_modulation_mae_db": 0.6255385908570497, "3000–6000 Hz_10ms_log_rms_mae_db": 1.876989171682031, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24320590255583988, "3000–6000 Hz_10ms_envelope_correlation": 0.9593344457691352, "3000–6000 Hz_10ms_modulation_mae_db": 0.3289336589426635, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4320537950095722, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16057796700995475, "3000–6000 Hz_40ms_envelope_correlation": 0.9810575943127533, "3000–6000 Hz_40ms_modulation_mae_db": 0.17088489292171608, "mr_stft_db": 5.844005049020127, "band_mae_db": 0.2074099416918214, "rms_gain_error_db": -0.1079126770433079, "output_peak": 0.8257652580309083}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.910460067860749, "全帯域_10ms_rms_envelope_nrmse": 0.06857022850892225, "全帯域_10ms_envelope_correlation": 0.9950183048634448, "全帯域_10ms_modulation_mae_db": 0.2545785886038067, "全帯域_40ms_log_rms_mae_db": 0.450825279747021, "全帯域_40ms_rms_envelope_nrmse": 0.03325119758958329, "全帯域_40ms_envelope_correlation": 0.998758292815619, "全帯域_40ms_modulation_mae_db": 0.16869072412380967, "300–1000 Hz_10ms_log_rms_mae_db": 1.44351197339929, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10265627252494801, "300–1000 Hz_10ms_envelope_correlation": 0.9918870834397963, "300–1000 Hz_10ms_modulation_mae_db": 0.33757923193895367, "300–1000 Hz_40ms_log_rms_mae_db": 0.7537478484285471, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04883736591311692, "300–1000 Hz_40ms_envelope_correlation": 0.9981408893645619, "300–1000 Hz_40ms_modulation_mae_db": 0.15075074236742733, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6592334949907654, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14828100316192072, "1000–3000 Hz_10ms_envelope_correlation": 0.9843035614715857, "1000–3000 Hz_10ms_modulation_mae_db": 0.08282375684919417, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9914302162916518, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08199157806590897, "1000–3000 Hz_40ms_envelope_correlation": 0.9949976262793093, "1000–3000 Hz_40ms_modulation_mae_db": 0.4361503910787991, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9595141309870183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2620342914593822, "3000–6000 Hz_10ms_envelope_correlation": 0.9519688065609218, "3000–6000 Hz_10ms_modulation_mae_db": 0.8058431346854477, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5144942204798206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18104922783492525, "3000–6000 Hz_40ms_envelope_correlation": 0.975447790761737, "3000–6000 Hz_40ms_modulation_mae_db": 0.5970440901743391, "mr_stft_db": 5.856126874124737, "band_mae_db": 0.07813898977321254, "rms_gain_error_db": -0.041617345326288156, "output_peak": 0.7384877724022041}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.808857862360924, "spectral_convergence": 0.26536735011721263, "spectral_convergence_status": "ok", "amplitude_floor": 2.2239575932173076e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.579525377252404, "spectral_convergence": 0.27145704353762534, "spectral_convergence_status": "ok", "amplitude_floor": 1.640120571472942e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.169727579341406, "spectral_convergence": 0.27827777838601775, "spectral_convergence_status": "ok", "amplitude_floor": 1.3411038650227006e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0016608291202231148, "output_power": 0.0016845874735474586, "delta_db": 0.06168616032604348, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001899519997387598, "output_power": 0.0018723020446431047, "delta_db": -0.06267958422507439, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002727484270807038, "output_power": 0.002487606363192547, "delta_db": -0.39980595674633473, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007661816102082716, "output_power": 0.0007273826289787169, "delta_db": -0.22568798673127086, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000247481283821024, "output_power": 0.00024344040098741608, "delta_db": -0.07149705570261716, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.389711909369124e-05, "output_power": 7.114883662411874e-05, "delta_db": 0.4668652547805019, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21470366641864044, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16766133504739653, "pitch": null, "output_rms_dbfs": -21.49570185101997, "output_peak": 0.7019069019072526, "reference_rms_dbfs": -21.328040515972575, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8527036063182445}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031196729436307625, "log_rms_mae_db": 0.9104510624729654, "rms_envelope_nrmse": 0.08417083498794058, "envelope_correlation": 0.9927287535964421, "rms_gain_error_db": -0.16763878709742444, "reference_mean_power": 0.00737113406357093, "output_mean_power": 0.007092027833232396}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.772697507541185e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010414087648688783, "output_power": 0.0009957747598935964, "reference_active": true, "delta_db": -0.1946011460490961}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001272580988246754, "output_power": 0.00012839762712866864, "reference_active": true, "delta_db": 0.03871566967449384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4550125163182702e-05, "output_power": 2.7489908114466323e-05, "reference_active": true, "delta_db": 0.4911957759299559}], "active_band_count": 3, "active_band_mae_db": 0.24150419721784858}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002933828481800726, "log_rms_mae_db": 0.4935453792001787, "rms_envelope_nrmse": 0.04292148059996806, "envelope_correlation": 0.9983118757098035, "rms_gain_error_db": -0.167772869409826, "reference_mean_power": 0.007404046443755548, "output_mean_power": 0.007123474063367486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.623838097183376e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009238145152881926, "output_power": 0.0008836541786091562, "reference_active": true, "delta_db": -0.19302446265859893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0904498001922555e-05, "output_power": 5.836100282272889e-05, "reference_active": true, "delta_db": -0.18526622115384028}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8155579307174739e-06, "output_power": 2.0636985391469362e-06, "reference_active": true, "delta_db": 0.5563614582733014}], "active_band_count": 3, "active_band_mae_db": 0.3115507140285802}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027755394478816403, "log_rms_mae_db": 1.5480527157859494, "rms_envelope_nrmse": 0.1314913512237657, "envelope_correlation": 0.9873303306916109, "rms_gain_error_db": -0.34033148854163475, "reference_mean_power": 0.003209968762055962, "output_mean_power": 0.0029680257003896846}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3498130551666851e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006624323096451465, "output_power": 0.0006270011265757288, "reference_active": true, "delta_db": -0.23873185527068336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.187213693276335e-05, "output_power": 5.9469760818509657e-05, "reference_active": true, "delta_db": -0.8226436586249225}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2558954262897386e-05, "output_power": 1.1882452150023535e-05, "reference_active": true, "delta_db": -0.24047404740559203}], "active_band_count": 3, "active_band_mae_db": 0.4339498537670659}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025197665775423135, "log_rms_mae_db": 0.8751593033650278, "rms_envelope_nrmse": 0.07524844510848339, "envelope_correlation": 0.9965125547395393, "rms_gain_error_db": -0.34073195419424007, "reference_mean_power": 0.0032255924683410116, "output_mean_power": 0.002982196807468034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.287517318914353e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006130037756394711, "output_power": 0.0005772341599946105, "reference_active": true, "delta_db": -0.26111125266628293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.4809964114174026e-05, "output_power": 2.7575246042625045e-05, "reference_active": true, "delta_db": -1.0118417940613678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.243410741089396e-07, "output_power": 7.562819100846381e-07, "reference_active": true, "delta_db": -0.8714853942065154}], "active_band_count": 3, "active_band_mae_db": 0.714812813644722}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017139073768520627, "log_rms_mae_db": 1.634539873182801, "rms_envelope_nrmse": 0.17035814256427675, "envelope_correlation": 0.9790386138556463, "rms_gain_error_db": -0.20643035685605315, "reference_mean_power": 0.0008077036407054461, "output_mean_power": 0.0007702097371691833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.727013621661235e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024375238055045323, "output_power": 0.0002340918392112294, "reference_active": true, "delta_db": -0.17562592015837553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.980069758207981e-05, "output_power": 2.9626724187678982e-05, "reference_active": true, "delta_db": -0.02542795860234602}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.347390229447928e-06, "output_power": 8.076072088938782e-06, "reference_active": true, "delta_db": 0.41067080503142095}], "active_band_count": 3, "active_band_mae_db": 0.20390822793071417}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015289190214292506, "log_rms_mae_db": 0.9915847457195485, "rms_envelope_nrmse": 0.08782310382230694, "envelope_correlation": 0.9944662214049005, "rms_gain_error_db": -0.2060277690649912, "reference_mean_power": 0.0008114918980720446, "output_mean_power": 0.000773893878287698}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4774424074646414e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022848364185803233, "output_power": 0.00022093127824340429, "reference_active": true, "delta_db": -0.1459790721548544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2508904456720676e-05, "output_power": 1.2947538005361593e-05, "reference_active": true, "delta_db": 0.14967919109556105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.746475959138801e-07, "output_power": 6.03083585896319e-07, "reference_active": true, "delta_db": 0.209759143195404}], "active_band_count": 3, "active_band_mae_db": 0.16847246881527314}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.106734387114872e-05, "log_rms_mae_db": 1.9635019872804704, "rms_envelope_nrmse": 0.2518536938997106, "envelope_correlation": 0.9573816108472117, "rms_gain_error_db": 0.23178530430806416, "reference_mean_power": 0.0001388420753372346, "output_mean_power": 0.00014645345683858465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.164366134779655e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.596504488102423e-05, "output_power": 4.8943178547013256e-05, "reference_active": true, "delta_db": 0.27264482080712743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.50200265064541e-05, "output_power": 1.5652515277407418e-05, "reference_active": true, "delta_db": 0.17913437258073184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4297410894022624e-06, "output_power": 3.9066049066597605e-06, "reference_active": true, "delta_db": 2.0623949308993437}], "active_band_count": 3, "active_band_mae_db": 0.8380580414290676}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.91579507177982e-05, "log_rms_mae_db": 1.5599278096235032, "rms_envelope_nrmse": 0.17733950848467023, "envelope_correlation": 0.9775611705170473, "rms_gain_error_db": 0.23166597142674053, "reference_mean_power": 0.00013952517981136655, "output_mean_power": 0.00014716996548469163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0998934517742833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.388750580480172e-05, "output_power": 4.8506125663040947e-05, "reference_active": true, "delta_db": 0.4345568762827786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1884324982671665e-06, "output_power": 7.988435462556854e-06, "reference_active": true, "delta_db": 0.4582753201295806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0513859982120003e-07, "output_power": 3.2474873176169787e-07, "reference_active": true, "delta_db": 1.9950007679986645}], "active_band_count": 3, "active_band_mae_db": 0.962610988137008}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.866396603879473, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6539aff373e920c1d929f8642aa823b46967b427e8c7816fb50f94877bbee023", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9820945291068889, "standard_deviation": 0.002980480564804662, "minimum": 0.978526751339506, "maximum": 0.9876614687005666, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6306747178354344, "standard_deviation": 0.03996893154465147, "minimum": 1.5675502027673298, "maximum": 1.6803471682241584, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5146593150173175, "standard_deviation": 0.3112907461009283, "minimum": 0.08282375684919417, "maximum": 0.8999102465769898, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15761293531063247, "standard_deviation": 0.013298639004921217, "minimum": 0.13152099132977138, "maximum": 0.1725500856945796, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9947247729383615, "standard_deviation": 0.001164345394996211, "minimum": 0.9936749588642881, "maximum": 0.9966073490367696, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9998471144570811, "standard_deviation": 0.03137223889577904, "minimum": 0.9442902877698737, "maximum": 1.048915606319047, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6426095061116436, "standard_deviation": 0.37617417253179664, "minimum": 0.16847246881527314, "maximum": 1.1967305543818385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08613826660887033, "standard_deviation": 0.00845773267106147, "minimum": 0.07000895273381953, "maximum": 0.09381717198474084, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9539226522309444, "standard_deviation": 0.003123569695797654, "minimum": 0.950038451004547, "maximum": 0.9593344457691352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9530032622130182, "standard_deviation": 0.04594606913715892, "minimum": 1.876989171682031, "maximum": 2.0393703680057147, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.660872732848948, "standard_deviation": 0.2305373357514887, "minimum": 0.3289336589426635, "maximum": 0.9883425773246604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2565438630090441, "standard_deviation": 0.006854432409911775, "minimum": 0.24320590255583988, "maximum": 0.2648853897404669, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9766259645675808, "standard_deviation": 0.0023434494952147052, "minimum": 0.9742954523965223, "maximum": 0.9810575943127533, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5267933492449004, "standard_deviation": 0.04429849572475918, "minimum": 1.4320537950095722, "maximum": 1.5814426740347538, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5852157492244592, "standard_deviation": 0.2639994071871235, "minimum": 0.17088489292171608, "maximum": 0.962610988137008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17680239670160344, "standard_deviation": 0.008033236680453373, "minimum": 0.16057796700995475, "maximum": 0.1843931249985307, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9903366093519992, "standard_deviation": 0.002604350764419815, "minimum": 0.9865861682869308, "maximum": 0.9929882239474072, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4771273634243176, "standard_deviation": 0.05048765179740393, "minimum": 1.3950782516931022, "maximum": 1.5480527157859494, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4409462645051465, "standard_deviation": 0.14612492067970223, "minimum": 0.2600430799567423, "maximum": 0.661108863835658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11175950003017045, "standard_deviation": 0.015426078095988184, "minimum": 0.09547287718248955, "maximum": 0.13193660576286786, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971756658366404, "standard_deviation": 0.0011222027893760132, "minimum": 0.995405416828735, "maximum": 0.9981458548987645, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8159148735250431, "standard_deviation": 0.05164962657646755, "minimum": 0.7435786007836517, "maximum": 0.8751593033650278, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3930481739608105, "standard_deviation": 0.19010585887005216, "minimum": 0.15075074236742733, "maximum": 0.714812813644722, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06037762141265546, "standard_deviation": 0.012402802624665418, "minimum": 0.04827609736393298, "maximum": 0.07524844510848339, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19875343546878227, "standard_deviation": 0.05009716844748632, "minimum": 0.07813898977321254, "maximum": 0.235193666750202, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.8535038034309235, "standard_deviation": 0.005108520361908822, "minimum": 5.844005049020127, "maximum": 5.860288248214057, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7673345966957059, "standard_deviation": 0.07797234426600864, "minimum": 0.6908241884614985, "maximum": 0.9062173915735947, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08096548780143752, "standard_deviation": 0.04333286985033784, "minimum": -0.16766133504739653, "maximum": -0.03212351041791948, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941781640793603, "standard_deviation": 0.001326089272107191, "minimum": 0.9924557221106902, "maximum": 0.9958273417840094, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9155866366198522, "standard_deviation": 0.01205380065646296, "minimum": 0.8949746563067443, "maximum": 0.9342746777743325, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2551832808833897, "standard_deviation": 0.12316595114175324, "minimum": 0.09795061461584613, "maximum": 0.4920705081945573, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07404695732342614, "standard_deviation": 0.008555405370567908, "minimum": 0.06299066291167033, "maximum": 0.08429322767948233, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984174711878491, "standard_deviation": 0.0005916622324359206, "minimum": 0.9970762893330336, "maximum": 0.9989777673067683, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47252633511434067, "standard_deviation": 0.024533782787054443, "minimum": 0.4414101912410563, "maximum": 0.5061057691600712, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.27081151948137383, "standard_deviation": 0.12197123181070606, "minimum": 0.13210562976273674, "maximum": 0.5125163231527541, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03806717830193998, "standard_deviation": 0.00637139664940364, "minimum": 0.030729624869492617, "maximum": 0.050656036054191195, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0060", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.8335232734680176, "predicted_mos": 2.8335232734680176, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b90be6d4c6d07f9881244052d4905fffd0d9b1227e9d50972d3dfca2575bb71b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.725978699978441, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9317202402206062, "全帯域_10ms_rms_envelope_nrmse": 0.0829794811004922, "全帯域_10ms_envelope_correlation": 0.9929344078914716, "全帯域_10ms_modulation_mae_db": 0.2991753704687456, "全帯域_40ms_log_rms_mae_db": 0.5198328275532064, "全帯域_40ms_rms_envelope_nrmse": 0.04456643296181306, "全帯域_40ms_envelope_correlation": 0.99813207014292, "全帯域_40ms_modulation_mae_db": 0.3267460254250749, "300–1000 Hz_10ms_log_rms_mae_db": 1.529793077222072, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12464587578830107, "300–1000 Hz_10ms_envelope_correlation": 0.988650711064659, "300–1000 Hz_10ms_modulation_mae_db": 0.3652540313225648, "300–1000 Hz_40ms_log_rms_mae_db": 0.8236635690140558, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07257044353755576, "300–1000 Hz_40ms_envelope_correlation": 0.9967307356786274, "300–1000 Hz_40ms_modulation_mae_db": 0.6465271066315448, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6205583213706682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1622408787931458, "1000–3000 Hz_10ms_envelope_correlation": 0.9812589780229702, "1000–3000 Hz_10ms_modulation_mae_db": 0.15382585979549, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9374283802666219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08452166258983682, "1000–3000 Hz_40ms_envelope_correlation": 0.9952979981231629, "1000–3000 Hz_40ms_modulation_mae_db": 0.10346582785389206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9898563424751703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2514757973893459, "3000–6000 Hz_10ms_envelope_correlation": 0.957643789570061, "3000–6000 Hz_10ms_modulation_mae_db": 0.984676323885138, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5426589569029647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17719808921520072, "3000–6000 Hz_40ms_envelope_correlation": 0.9776617914966287, "3000–6000 Hz_40ms_modulation_mae_db": 1.1100801579080037, "mr_stft_db": 5.818055979760665, "band_mae_db": 0.22430575909815775, "rms_gain_error_db": -0.16489812096885065, "output_peak": 0.7003670491704703}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9415186208565599, "全帯域_10ms_rms_envelope_nrmse": 0.07848527780711559, "全帯域_10ms_envelope_correlation": 0.9935214477683499, "全帯域_10ms_modulation_mae_db": 0.24115169337754705, "全帯域_40ms_log_rms_mae_db": 0.47805720049649214, "全帯域_40ms_rms_envelope_nrmse": 0.0392550770508134, "全帯域_40ms_envelope_correlation": 0.9983559619770055, "全帯域_40ms_modulation_mae_db": 0.0835173929607121, "300–1000 Hz_10ms_log_rms_mae_db": 1.4427804285625005, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12057151539129464, "300–1000 Hz_10ms_envelope_correlation": 0.9888392247172784, "300–1000 Hz_10ms_modulation_mae_db": 0.5454332722628362, "300–1000 Hz_40ms_log_rms_mae_db": 0.8014037433450979, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07194512887473543, "300–1000 Hz_40ms_envelope_correlation": 0.9959824822309126, "300–1000 Hz_40ms_modulation_mae_db": 0.3554408116057363, "1000–3000 Hz_10ms_log_rms_mae_db": 1.610608114410692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16336693261027474, "1000–3000 Hz_10ms_envelope_correlation": 0.9810115877845997, "1000–3000 Hz_10ms_modulation_mae_db": 1.0028275262540332, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9487388262728997, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09155056592144513, "1000–3000 Hz_40ms_envelope_correlation": 0.9943396979717102, "1000–3000 Hz_40ms_modulation_mae_db": 1.3535089621716943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9730586608645815, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24727676367192616, "3000–6000 Hz_10ms_envelope_correlation": 0.9573679221963702, "3000–6000 Hz_10ms_modulation_mae_db": 0.808397027024197, "3000–6000 Hz_40ms_log_rms_mae_db": 1.525417794442897, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16244953901196685, "3000–6000 Hz_40ms_envelope_correlation": 0.9802186200833704, "3000–6000 Hz_40ms_modulation_mae_db": 0.9060984647461129, "mr_stft_db": 5.8127918051461185, "band_mae_db": 0.2514599628826159, "rms_gain_error_db": -0.09438177544833207, "output_peak": 0.6814438996865496}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9436847497752391, "全帯域_10ms_rms_envelope_nrmse": 0.07083800019633868, "全帯域_10ms_envelope_correlation": 0.9947090097588211, "全帯域_10ms_modulation_mae_db": 0.18977976286980178, "全帯域_40ms_log_rms_mae_db": 0.4982397850066988, "全帯域_40ms_rms_envelope_nrmse": 0.03536969646530513, "全帯域_40ms_envelope_correlation": 0.9986380807374695, "全帯域_40ms_modulation_mae_db": 0.2635900020399144, "300–1000 Hz_10ms_log_rms_mae_db": 1.421596319951194, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10024224756465634, "300–1000 Hz_10ms_envelope_correlation": 0.9922513210440008, "300–1000 Hz_10ms_modulation_mae_db": 0.34694414176488153, "300–1000 Hz_40ms_log_rms_mae_db": 0.7518237628947883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051456077407801296, "300–1000 Hz_40ms_envelope_correlation": 0.9978805920936504, "300–1000 Hz_40ms_modulation_mae_db": 0.18137273944881893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.655996842322294, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15461025937842252, "1000–3000 Hz_10ms_envelope_correlation": 0.9838079191982576, "1000–3000 Hz_10ms_modulation_mae_db": 1.0931310529437182, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0193561048017856, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08353318984827093, "1000–3000 Hz_40ms_envelope_correlation": 0.9964198957538764, "1000–3000 Hz_40ms_modulation_mae_db": 1.5165674932722606, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9523478675421153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24260666552427348, "3000–6000 Hz_10ms_envelope_correlation": 0.9581650914597669, "3000–6000 Hz_10ms_modulation_mae_db": 0.4132604549298981, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5312336646606937, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16614119545037617, "3000–6000 Hz_40ms_envelope_correlation": 0.9791371232414634, "3000–6000 Hz_40ms_modulation_mae_db": 0.36655151876522485, "mr_stft_db": 5.821717506733152, "band_mae_db": 0.24946704212615742, "rms_gain_error_db": -0.07241456683463379, "output_peak": 0.8318390399712114}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9340329636823538, "全帯域_10ms_rms_envelope_nrmse": 0.08022487549286406, "全帯域_10ms_envelope_correlation": 0.9931846418651833, "全帯域_10ms_modulation_mae_db": 0.487544320496347, "全帯域_40ms_log_rms_mae_db": 0.48361311624469233, "全帯域_40ms_rms_envelope_nrmse": 0.03831621922571722, "全帯域_40ms_envelope_correlation": 0.9983591701907305, "全帯域_40ms_modulation_mae_db": 0.30364699868032796, "300–1000 Hz_10ms_log_rms_mae_db": 1.4282082198036008, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12068065583130733, "300–1000 Hz_10ms_envelope_correlation": 0.9887771271791457, "300–1000 Hz_10ms_modulation_mae_db": 0.5930880974944701, "300–1000 Hz_40ms_log_rms_mae_db": 0.7667971274236722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061222612994648096, "300–1000 Hz_40ms_envelope_correlation": 0.9969807844632025, "300–1000 Hz_40ms_modulation_mae_db": 0.3976346793879049, "1000–3000 Hz_10ms_log_rms_mae_db": 1.553357127183441, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14516937206789424, "1000–3000 Hz_10ms_envelope_correlation": 0.9851057755543176, "1000–3000 Hz_10ms_modulation_mae_db": 0.6193965613420508, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9680578073533724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09386196641694146, "1000–3000 Hz_40ms_envelope_correlation": 0.9938966791453903, "1000–3000 Hz_40ms_modulation_mae_db": 0.3612791061695874, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9769870901571334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2582064603620362, "3000–6000 Hz_10ms_envelope_correlation": 0.9529490072507247, "3000–6000 Hz_10ms_modulation_mae_db": 0.596457244416703, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5683020507355827, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17956883367375562, "3000–6000 Hz_40ms_envelope_correlation": 0.9755853816253208, "3000–6000 Hz_40ms_modulation_mae_db": 0.5461488375388359, "mr_stft_db": 5.8249060520631915, "band_mae_db": 0.2196389598566831, "rms_gain_error_db": -0.054464537019814116, "output_peak": 0.7318638443553246}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9149436651586347, "全帯域_10ms_rms_envelope_nrmse": 0.0622877907245597, "全帯域_10ms_envelope_correlation": 0.9959459218409844, "全帯域_10ms_modulation_mae_db": 0.11465021843581245, "全帯域_40ms_log_rms_mae_db": 0.4592555781358192, "全帯域_40ms_rms_envelope_nrmse": 0.03111913646789407, "全帯域_40ms_envelope_correlation": 0.9989840043456009, "全帯域_40ms_modulation_mae_db": 0.3431715050571178, "300–1000 Hz_10ms_log_rms_mae_db": 1.453430352536332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09507189812001876, "300–1000 Hz_10ms_envelope_correlation": 0.9930543947913681, "300–1000 Hz_10ms_modulation_mae_db": 0.31874013202581636, "300–1000 Hz_40ms_log_rms_mae_db": 0.8066730144195192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049150177355506135, "300–1000 Hz_40ms_envelope_correlation": 0.9980947520862268, "300–1000 Hz_40ms_modulation_mae_db": 0.5978810763670613, "1000–3000 Hz_10ms_log_rms_mae_db": 1.550164038553497, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1273451554054877, "1000–3000 Hz_10ms_envelope_correlation": 0.9888121626036348, "1000–3000 Hz_10ms_modulation_mae_db": 0.8919527501850714, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9306619891482021, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06926880595461746, "1000–3000 Hz_40ms_envelope_correlation": 0.9971621969844464, "1000–3000 Hz_40ms_modulation_mae_db": 1.0264943753267683, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0414843060649908, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26090276783984423, "3000–6000 Hz_10ms_envelope_correlation": 0.9514002039397761, "3000–6000 Hz_10ms_modulation_mae_db": 0.5003052390882198, "3000–6000 Hz_40ms_log_rms_mae_db": 1.559063490597153, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17537967409114374, "3000–6000 Hz_40ms_envelope_correlation": 0.976801674049559, "3000–6000 Hz_40ms_modulation_mae_db": 0.40401228103073855, "mr_stft_db": 5.8188097530595675, "band_mae_db": 0.2461532285055519, "rms_gain_error_db": -0.07915484408113566, "output_peak": 0.6971281587903202}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9237374151508234, "全帯域_10ms_rms_envelope_nrmse": 0.0861556908390627, "全帯域_10ms_envelope_correlation": 0.9921199640259717, "全帯域_10ms_modulation_mae_db": 0.28933046954265934, "全帯域_40ms_log_rms_mae_db": 0.5157851675563356, "全帯域_40ms_rms_envelope_nrmse": 0.05063436876776459, "全帯域_40ms_envelope_correlation": 0.9970809465057248, "全帯域_40ms_modulation_mae_db": 0.22641502260579258, "300–1000 Hz_10ms_log_rms_mae_db": 1.4886475587963646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13319163511631518, "300–1000 Hz_10ms_envelope_correlation": 0.9863435945787856, "300–1000 Hz_10ms_modulation_mae_db": 0.6693819578069329, "300–1000 Hz_40ms_log_rms_mae_db": 0.7964753685539373, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07420648672626322, "300–1000 Hz_40ms_envelope_correlation": 0.9955337434577083, "300–1000 Hz_40ms_modulation_mae_db": 0.4624444838643586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6187513034972618, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15463256631526937, "1000–3000 Hz_10ms_envelope_correlation": 0.9833665308098324, "1000–3000 Hz_10ms_modulation_mae_db": 0.5422462193685557, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0013888477394721, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09277541939371864, "1000–3000 Hz_40ms_envelope_correlation": 0.9945672512748299, "1000–3000 Hz_40ms_modulation_mae_db": 0.5584123373408633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9610252762247014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24726368203846058, "3000–6000 Hz_10ms_envelope_correlation": 0.956457199481023, "3000–6000 Hz_10ms_modulation_mae_db": 0.8954880997770572, "3000–6000 Hz_40ms_log_rms_mae_db": 1.513260247287136, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17505677734663372, "3000–6000 Hz_40ms_envelope_correlation": 0.9771780766392275, "3000–6000 Hz_40ms_modulation_mae_db": 0.7659062439077454, "mr_stft_db": 5.817212982071766, "band_mae_db": 0.24846973896320554, "rms_gain_error_db": -0.035354748537933335, "output_peak": 0.9091585034081376}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9315303553443438, "全帯域_10ms_rms_envelope_nrmse": 0.06774134912581799, "全帯域_10ms_envelope_correlation": 0.9952187164964321, "全帯域_10ms_modulation_mae_db": 0.2508156856656445, "全帯域_40ms_log_rms_mae_db": 0.4599046356967747, "全帯域_40ms_rms_envelope_nrmse": 0.035171322221035105, "全帯域_40ms_envelope_correlation": 0.9987246623075156, "全帯域_40ms_modulation_mae_db": 0.39244438345044275, "300–1000 Hz_10ms_log_rms_mae_db": 1.5114299684329235, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0996682985978271, "300–1000 Hz_10ms_envelope_correlation": 0.9924629111524783, "300–1000 Hz_10ms_modulation_mae_db": 0.42827814745988163, "300–1000 Hz_40ms_log_rms_mae_db": 0.7600295946148637, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051235076296820586, "300–1000 Hz_40ms_envelope_correlation": 0.9980919553961117, "300–1000 Hz_40ms_modulation_mae_db": 0.3770939133306908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.603975881246663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14749828552770902, "1000–3000 Hz_10ms_envelope_correlation": 0.9848297129101781, "1000–3000 Hz_10ms_modulation_mae_db": 0.8542284371010472, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9575634160424428, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07831166980985535, "1000–3000 Hz_40ms_envelope_correlation": 0.996314981351507, "1000–3000 Hz_40ms_modulation_mae_db": 1.0017416061681768, "3000–6000 Hz_10ms_log_rms_mae_db": 1.861909210516415, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24295114112479702, "3000–6000 Hz_10ms_envelope_correlation": 0.9591847986414894, "3000–6000 Hz_10ms_modulation_mae_db": 0.23801299709510315, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3884739026447719, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15860951861217532, "3000–6000 Hz_40ms_envelope_correlation": 0.9813377974429917, "3000–6000 Hz_40ms_modulation_mae_db": 0.20254518492417595, "mr_stft_db": 5.802810806757594, "band_mae_db": 0.22886849245056803, "rms_gain_error_db": -0.1007821280524179, "output_peak": 0.8277133552144611}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9266940738288589, "全帯域_10ms_rms_envelope_nrmse": 0.06918018317653035, "全帯域_10ms_envelope_correlation": 0.994929177772659, "全帯域_10ms_modulation_mae_db": 0.29542608308903573, "全帯域_40ms_log_rms_mae_db": 0.46114722950024617, "全帯域_40ms_rms_envelope_nrmse": 0.03467397247630473, "全帯域_40ms_envelope_correlation": 0.9986489787161932, "全帯域_40ms_modulation_mae_db": 0.23647153226598613, "300–1000 Hz_10ms_log_rms_mae_db": 1.4427229890350484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10568070936634939, "300–1000 Hz_10ms_envelope_correlation": 0.991384169274476, "300–1000 Hz_10ms_modulation_mae_db": 0.393475505693411, "300–1000 Hz_40ms_log_rms_mae_db": 0.7540993898191594, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05137624109837862, "300–1000 Hz_40ms_envelope_correlation": 0.9979071499093427, "300–1000 Hz_40ms_modulation_mae_db": 0.2174530919402944, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6320167774348304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13086139989705542, "1000–3000 Hz_10ms_envelope_correlation": 0.9877607012948537, "1000–3000 Hz_10ms_modulation_mae_db": 0.35667977669406775, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9530010950024725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07059873196797348, "1000–3000 Hz_40ms_envelope_correlation": 0.9963852099463791, "1000–3000 Hz_40ms_modulation_mae_db": 0.681783167254749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.985028384181302, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2528873491983504, "3000–6000 Hz_10ms_envelope_correlation": 0.9548656425972674, "3000–6000 Hz_10ms_modulation_mae_db": 0.7262590793021116, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5113176399975363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17415484850457438, "3000–6000 Hz_40ms_envelope_correlation": 0.9770775197237825, "3000–6000 Hz_40ms_modulation_mae_db": 0.49368902653951213, "mr_stft_db": 5.818779472335637, "band_mae_db": 0.11258042773570183, "rms_gain_error_db": -0.043219029556998564, "output_peak": 0.7434608912180856}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.770821256875502, "spectral_convergence": 0.2629687871554964, "spectral_convergence_status": "ok", "amplitude_floor": 2.2239575932173076e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.55756849913307, "spectral_convergence": 0.26768053556385113, "spectral_convergence_status": "ok", "amplitude_floor": 1.640120571472942e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.12577818327342, "spectral_convergence": 0.27633958257457364, "spectral_convergence_status": "ok", "amplitude_floor": 1.3411038650227006e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0016608291202231148, "output_power": 0.0016907079807003812, "delta_db": 0.07743651719056542, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001899519997387598, "output_power": 0.0018767180474533692, "delta_db": -0.05244839584817268, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002727484270807038, "output_power": 0.0024976779488318553, "delta_db": -0.3822581510677252, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007661816102082716, "output_power": 0.0007118141746751292, "delta_db": -0.3196509183740275, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000247481283821024, "output_power": 0.0002433380665555559, "delta_db": -0.07332307236866419, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.389711909369124e-05, "output_power": 7.072175472246516e-05, "delta_db": 0.4407174997397916, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22430575909815775, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16489812096885065, "pitch": null, "output_rms_dbfs": -21.492938636941425, "output_peak": 0.7003670491704703, "reference_rms_dbfs": -21.328040515972575, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.818055979760665}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031196729436307625, "log_rms_mae_db": 0.9317202402206062, "rms_envelope_nrmse": 0.0829794811004922, "envelope_correlation": 0.9929344078914716, "rms_gain_error_db": -0.1648675156715555, "reference_mean_power": 0.00737113406357093, "output_mean_power": 0.007096554762980148}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.772697507541185e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010414087648688783, "output_power": 0.000998508458069629, "reference_active": true, "delta_db": -0.18269480530521548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001272580988246754, "output_power": 0.00013089673817116506, "reference_active": true, "delta_db": 0.12243393662047129}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4550125163182702e-05, "output_power": 2.8138014400431044e-05, "reference_active": true, "delta_db": 0.59239736948055}], "active_band_count": 3, "active_band_mae_db": 0.2991753704687456}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002933828481800726, "log_rms_mae_db": 0.5198328275532064, "rms_envelope_nrmse": 0.04456643296181306, "envelope_correlation": 0.99813207014292, "rms_gain_error_db": -0.16499751362637688, "reference_mean_power": 0.007404046443755548, "output_mean_power": 0.007128027769244034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.623838097183376e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009238145152881926, "output_power": 0.0008866154405539382, "reference_active": true, "delta_db": -0.1784949140778709}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0904498001922555e-05, "output_power": 5.928660763237597e-05, "reference_active": true, "delta_db": -0.11692767110774896}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8155579307174739e-06, "output_power": 2.1256495248961927e-06, "reference_active": true, "delta_db": 0.6848154910896049}], "active_band_count": 3, "active_band_mae_db": 0.3267460254250749}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027755394478816403, "log_rms_mae_db": 1.529793077222072, "rms_envelope_nrmse": 0.12464587578830107, "envelope_correlation": 0.988650711064659, "rms_gain_error_db": -0.3239224219939849, "reference_mean_power": 0.003209968762055962, "output_mean_power": 0.0029792610847434982}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3498130551666851e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006624323096451465, "output_power": 0.0006348196583225214, "reference_active": true, "delta_db": -0.18491139691060765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.187213693276335e-05, "output_power": 5.959379934741065e-05, "reference_active": true, "delta_db": -0.8135948332650234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2558954262897386e-05, "output_power": 1.2280834888035211e-05, "reference_active": true, "delta_db": -0.09725586379206344}], "active_band_count": 3, "active_band_mae_db": 0.3652540313225648}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025197665775423135, "log_rms_mae_db": 0.8236635690140558, "rms_envelope_nrmse": 0.07257044353755576, "envelope_correlation": 0.9967307356786274, "rms_gain_error_db": -0.3243627532486962, "reference_mean_power": 0.0032255924683410116, "output_mean_power": 0.0029934583579202144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.287517318914353e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006130037756394711, "output_power": 0.0005848469875761734, "reference_active": true, "delta_db": -0.20420892162188986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.4809964114174026e-05, "output_power": 2.8342539708109917e-05, "reference_active": true, "delta_db": -0.892648118084636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.243410741089396e-07, "output_power": 7.613070013547485e-07, "reference_active": true, "delta_db": -0.8427242801881086}], "active_band_count": 3, "active_band_mae_db": 0.6465271066315448}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017139073768520627, "log_rms_mae_db": 1.6205583213706682, "rms_envelope_nrmse": 0.1622408787931458, "envelope_correlation": 0.9812589780229702, "rms_gain_error_db": -0.29579158386749127, "reference_mean_power": 0.0008077036407054461, "output_mean_power": 0.0007545236936055733}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.727013621661235e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024375238055045323, "output_power": 0.00023186388991641631, "reference_active": true, "delta_db": -0.21715748198212798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.980069758207981e-05, "output_power": 2.8931137385920543e-05, "reference_active": true, "delta_db": -0.12860922623098303}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.347390229447928e-06, "output_power": 7.545780946668788e-06, "reference_active": true, "delta_db": 0.115710871173359}], "active_band_count": 3, "active_band_mae_db": 0.15382585979549}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015289190214292506, "log_rms_mae_db": 0.9374283802666219, "rms_envelope_nrmse": 0.08452166258983682, "envelope_correlation": 0.9952979981231629, "rms_gain_error_db": -0.29539370629314277, "reference_mean_power": 0.0008114918980720446, "output_mean_power": 0.0007581319814896587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4774424074646414e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022848364185803233, "output_power": 0.00021843951413666243, "reference_active": true, "delta_db": -0.19523910593270088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2508904456720676e-05, "output_power": 1.2268562522635486e-05, "reference_active": true, "delta_db": -0.08425594892938022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.746475959138801e-07, "output_power": 5.705731768926109e-07, "reference_active": true, "delta_db": -0.030902428699595114}], "active_band_count": 3, "active_band_mae_db": 0.10346582785389206}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.106734387114872e-05, "log_rms_mae_db": 1.9898563424751703, "rms_envelope_nrmse": 0.2514757973893459, "envelope_correlation": 0.957643789570061, "rms_gain_error_db": 0.22927866300615268, "reference_mean_power": 0.0001388420753372346, "output_mean_power": 0.00014636895188238167}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.164366134779655e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.596504488102423e-05, "output_power": 5.0051837204247584e-05, "reference_active": true, "delta_db": 0.36992334829085705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.50200265064541e-05, "output_power": 1.6055814648716718e-05, "reference_active": true, "delta_db": 0.2896164684679913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4297410894022624e-06, "output_power": 4.121059772069534e-06, "reference_active": true, "delta_db": 2.294489154896566}], "active_band_count": 3, "active_band_mae_db": 0.984676323885138}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.91579507177982e-05, "log_rms_mae_db": 1.5426589569029647, "rms_envelope_nrmse": 0.17719808921520072, "envelope_correlation": 0.9776617914966287, "rms_gain_error_db": 0.22915472655946065, "reference_mean_power": 0.00013952517981136655, "output_mean_power": 0.0001470848911850302}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0998934517742833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.388750580480172e-05, "output_power": 4.977898926956416e-05, "reference_active": true, "delta_db": 0.5470517437573447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1884324982671665e-06, "output_power": 8.215331332877979e-06, "reference_active": true, "delta_db": 0.579908847468794}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0513859982120003e-07, "output_power": 3.407025430021923e-07, "reference_active": true, "delta_db": 2.2032798824978723}], "active_band_count": 3, "active_band_mae_db": 1.1100801579080037}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8648108600409495, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cd24d13cf429b646834b58bf1b99e1a6fe99f03e55852994eedb98b36c39aa56", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9844941710223305, "standard_deviation": 0.0027810402356284888, "minimum": 0.9810115877845997, "maximum": 0.9888121626036348, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6056785507524185, "standard_deviation": 0.03677058063091047, "minimum": 1.550164038553497, "maximum": 1.655996842322294, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6892860229605042, "standard_deviation": 0.3280959413067999, "minimum": 0.15382585979549, "maximum": 1.0931310529437182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14821560624940733, "standard_deviation": 0.013389111854690128, "minimum": 0.1273451554054877, "maximum": 0.16336693261027474, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955479888189127, "standard_deviation": 0.001187207910175452, "minimum": 0.9938966791453903, "maximum": 0.9971621969844464, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9645245583284086, "standard_deviation": 0.030912946140586147, "minimum": 0.9306619891482021, "maximum": 1.0193561048017856, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.825406609444749, "standard_deviation": 0.486481867060263, "minimum": 0.10346582785389206, "maximum": 1.5165674932722606, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08305275148783242, "standard_deviation": 0.009662507673067564, "minimum": 0.06926880595461746, "maximum": 0.09386196641694146, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9560042068920599, "standard_deviation": 0.002708878411889044, "minimum": 0.9514002039397761, "maximum": 0.9591847986414894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9677121422533013, "standard_deviation": 0.05045817984107381, "minimum": 1.861909210516415, "maximum": 2.0414843060649908, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6453570581898035, "standard_deviation": 0.2546873461046345, "minimum": 0.23801299709510315, "maximum": 0.984676323885138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25044632839362924, "standard_deviation": 0.006701691550740102, "minimum": 0.24260666552427348, "maximum": 0.26090276783984423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.978124748037793, "standard_deviation": 0.0019323930156737066, "minimum": 0.9755853816253208, "maximum": 0.9813377974429917, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.517465968408592, "standard_deviation": 0.055929256141322124, "minimum": 1.3884739026447719, "maximum": 1.5683020507355827, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5993789644200438, "standard_deviation": 0.30389883515936095, "minimum": 0.20254518492417595, "maximum": 1.1100801579080037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1710698094882283, "standard_deviation": 0.007631108176254291, "minimum": 0.15860951861217532, "maximum": 0.17956883367375562, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990220431725274, "standard_deviation": 0.002391162513874985, "minimum": 0.9863435945787856, "maximum": 0.9930543947913681, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4648261142925045, "standard_deviation": 0.04013227382816649, "minimum": 1.421596319951194, "maximum": 1.529793077222072, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4575744107288493, "standard_deviation": 0.1287020825262258, "minimum": 0.31874013202581636, "maximum": 0.6693819578069329, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11246910447200872, "standard_deviation": 0.014004475979641426, "minimum": 0.09507189812001876, "maximum": 0.13319163511631518, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971502744144728, "standard_deviation": 0.0010047512900141018, "minimum": 0.9955337434577083, "maximum": 0.9980947520862268, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7826206962606368, "standard_deviation": 0.027595148240686554, "minimum": 0.7518237628947883, "maximum": 0.8236635690140558, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4044809878220512, "standard_deviation": 0.16357971920889697, "minimum": 0.18137273944881893, "maximum": 0.6465271066315448, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06039528053646364, "standard_deviation": 0.010984935128082985, "minimum": 0.049150177355506135, "maximum": 0.07420648672626322, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22261795145233018, "standard_deviation": 0.04618906557504389, "minimum": 0.11258042773570183, "maximum": 0.2514599628826159, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.816885544740962, "standard_deviation": 0.006668690386521258, "minimum": 5.802810806757594, "maximum": 5.8249060520631915, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.76537184272682, "standard_deviation": 0.08151122204513365, "minimum": 0.6814438996865496, "maximum": 0.9091585034081376, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08058371881251451, "standard_deviation": 0.04123114054336021, "minimum": -0.16489812096885065, "maximum": -0.035354748537933335, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.994070410927484, "standard_deviation": 0.0013181225405508338, "minimum": 0.9921199640259717, "maximum": 0.9959459218409844, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9309827605021774, "standard_deviation": 0.00933987460047497, "minimum": 0.9149436651586347, "maximum": 0.9436847497752391, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.27098420049319916, "standard_deviation": 0.10754284259908303, "minimum": 0.11465021843581245, "maximum": 0.487544320496347, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07473658105784767, "standard_deviation": 0.00838786294191245, "minimum": 0.0622877907245597, "maximum": 0.0861556908390627, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9983654843653951, "standard_deviation": 0.0005819798530980995, "minimum": 0.9970809465057248, "maximum": 0.9989840043456009, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.48447944252378317, "standard_deviation": 0.02462811871998983, "minimum": 0.4592555781358192, "maximum": 0.5198328275532064, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2720003578106711, "standard_deviation": 0.09449684489571715, "minimum": 0.0835173929607121, "maximum": 0.39244438345044275, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03863827820458091, "standard_deviation": 0.0062620169408241395, "minimum": 0.03111913646789407, "maximum": 0.05063436876776459, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0060", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.8637821674346924, "predicted_mos": 2.8637821674346924, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7a576194f3499c96708f54a81ba5405fa3e8787324a3392bb5dc45b78bcea7c2", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7960896999575198, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9003726573023283, "全帯域_10ms_rms_envelope_nrmse": 0.07954331046043907, "全帯域_10ms_envelope_correlation": 0.9933189788445342, "全帯域_10ms_modulation_mae_db": 0.20331438183302467, "全帯域_40ms_log_rms_mae_db": 0.46287220439386834, "全帯域_40ms_rms_envelope_nrmse": 0.03473207711487078, "全帯域_40ms_envelope_correlation": 0.9986848856105425, "全帯域_40ms_modulation_mae_db": 0.20808408835100542, "300–1000 Hz_10ms_log_rms_mae_db": 1.6578890873915673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11828548670211272, "300–1000 Hz_10ms_envelope_correlation": 0.9892010415910282, "300–1000 Hz_10ms_modulation_mae_db": 0.7520747410136958, "300–1000 Hz_40ms_log_rms_mae_db": 0.9513280175955693, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05233047518552493, "300–1000 Hz_40ms_envelope_correlation": 0.9978788726203535, "300–1000 Hz_40ms_modulation_mae_db": 0.4152296428107762, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8322230162689748, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15790693941259032, "1000–3000 Hz_10ms_envelope_correlation": 0.9823739158155828, "1000–3000 Hz_10ms_modulation_mae_db": 0.4429909695215344, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1161195568701536, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09202098277073587, "1000–3000 Hz_40ms_envelope_correlation": 0.9937584412847066, "1000–3000 Hz_40ms_modulation_mae_db": 0.3294893528648488, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3202017571942766, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2661321832782402, "3000–6000 Hz_10ms_envelope_correlation": 0.9506313084593677, "3000–6000 Hz_10ms_modulation_mae_db": 0.7766428501109881, "3000–6000 Hz_40ms_log_rms_mae_db": 1.752612207892187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.187903796582517, "3000–6000 Hz_40ms_envelope_correlation": 0.9736672969822688, "3000–6000 Hz_40ms_modulation_mae_db": 0.6156768504190399, "mr_stft_db": 6.064912601381979, "band_mae_db": 0.14850863160611746, "rms_gain_error_db": -0.06596722188247806, "output_peak": 0.7589347074483878}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.924853374966505, "全帯域_10ms_rms_envelope_nrmse": 0.07606895620768314, "全帯域_10ms_envelope_correlation": 0.9939273056776187, "全帯域_10ms_modulation_mae_db": 0.2608695411128866, "全帯域_40ms_log_rms_mae_db": 0.47720634107814086, "全帯域_40ms_rms_envelope_nrmse": 0.041671895404425584, "全帯域_40ms_envelope_correlation": 0.9981407442473874, "全帯域_40ms_modulation_mae_db": 0.5254042980345988, "300–1000 Hz_10ms_log_rms_mae_db": 1.621269140732137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10840856969201926, "300–1000 Hz_10ms_envelope_correlation": 0.9910447880000948, "300–1000 Hz_10ms_modulation_mae_db": 0.1681925313654354, "300–1000 Hz_40ms_log_rms_mae_db": 0.9633809985882851, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06598773246099408, "300–1000 Hz_40ms_envelope_correlation": 0.9966828350880695, "300–1000 Hz_40ms_modulation_mae_db": 0.43010797280227103, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9250068447085076, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19094276239764227, "1000–3000 Hz_10ms_envelope_correlation": 0.9735529201545927, "1000–3000 Hz_10ms_modulation_mae_db": 0.2951290026256112, "1000–3000 Hz_40ms_log_rms_mae_db": 1.18923969890345, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10801484893651833, "1000–3000 Hz_40ms_envelope_correlation": 0.9912638404120533, "1000–3000 Hz_40ms_modulation_mae_db": 0.20104285385879783, "3000–6000 Hz_10ms_log_rms_mae_db": 2.403861596627954, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2706727350017079, "3000–6000 Hz_10ms_envelope_correlation": 0.948698756918988, "3000–6000 Hz_10ms_modulation_mae_db": 0.5692260819813123, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7891046621212041, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19182544880912858, "3000–6000 Hz_40ms_envelope_correlation": 0.9724437401336813, "3000–6000 Hz_40ms_modulation_mae_db": 0.7747377909859919, "mr_stft_db": 6.080108087882764, "band_mae_db": 0.2065401114117924, "rms_gain_error_db": -0.09857487124989292, "output_peak": 0.740833745759561}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9265826868884371, "全帯域_10ms_rms_envelope_nrmse": 0.07050983451629333, "全帯域_10ms_envelope_correlation": 0.9947480040225976, "全帯域_10ms_modulation_mae_db": 0.38792998672741535, "全帯域_40ms_log_rms_mae_db": 0.44668064855612205, "全帯域_40ms_rms_envelope_nrmse": 0.03615583183382948, "全帯域_40ms_envelope_correlation": 0.9985600421457391, "全帯域_40ms_modulation_mae_db": 0.2454466299997852, "300–1000 Hz_10ms_log_rms_mae_db": 1.6862306769110798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10407158363594111, "300–1000 Hz_10ms_envelope_correlation": 0.9918343222925446, "300–1000 Hz_10ms_modulation_mae_db": 0.6197267323397874, "300–1000 Hz_40ms_log_rms_mae_db": 0.9262895478935939, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05647175397008456, "300–1000 Hz_40ms_envelope_correlation": 0.997737407484259, "300–1000 Hz_40ms_modulation_mae_db": 0.4968642931645886, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8211146067098656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16512865811856378, "1000–3000 Hz_10ms_envelope_correlation": 0.980407647630746, "1000–3000 Hz_10ms_modulation_mae_db": 0.5713098077703681, "1000–3000 Hz_40ms_log_rms_mae_db": 1.106210221798151, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0948697493082235, "1000–3000 Hz_40ms_envelope_correlation": 0.9932098662851088, "1000–3000 Hz_40ms_modulation_mae_db": 0.7047059698797714, "3000–6000 Hz_10ms_log_rms_mae_db": 2.352263787075156, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.264863694312695, "3000–6000 Hz_10ms_envelope_correlation": 0.9510374489360566, "3000–6000 Hz_10ms_modulation_mae_db": 0.30545325062220974, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7411640495527465, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18326406854744692, "3000–6000 Hz_40ms_envelope_correlation": 0.9749509962279037, "3000–6000 Hz_40ms_modulation_mae_db": 0.18008365158151063, "mr_stft_db": 6.064179196673188, "band_mae_db": 0.19933347483608324, "rms_gain_error_db": -0.06493432530698759, "output_peak": 0.7051449312466399}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9646990984264217, "全帯域_10ms_rms_envelope_nrmse": 0.07388401383242743, "全帯域_10ms_envelope_correlation": 0.9942453220348548, "全帯域_10ms_modulation_mae_db": 0.11875560478696841, "全帯域_40ms_log_rms_mae_db": 0.4938932725702816, "全帯域_40ms_rms_envelope_nrmse": 0.03362806012518275, "全帯域_40ms_envelope_correlation": 0.9987975986361375, "全帯域_40ms_modulation_mae_db": 0.12831984753777195, "300–1000 Hz_10ms_log_rms_mae_db": 1.6213833578080166, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11263931114345604, "300–1000 Hz_10ms_envelope_correlation": 0.990230498668907, "300–1000 Hz_10ms_modulation_mae_db": 0.33949960908090177, "300–1000 Hz_40ms_log_rms_mae_db": 0.9415740823429406, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051608906485367895, "300–1000 Hz_40ms_envelope_correlation": 0.997962973736735, "300–1000 Hz_40ms_modulation_mae_db": 0.533178504793228, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8552401571412098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14411207447588184, "1000–3000 Hz_10ms_envelope_correlation": 0.9850088822438485, "1000–3000 Hz_10ms_modulation_mae_db": 0.4479895321311483, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1746038002447174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0861089138591524, "1000–3000 Hz_40ms_envelope_correlation": 0.9944752175354723, "1000–3000 Hz_40ms_modulation_mae_db": 0.7986194461526824, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2842541108565735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26966632838220317, "3000–6000 Hz_10ms_envelope_correlation": 0.9490167356895198, "3000–6000 Hz_10ms_modulation_mae_db": 0.792179811414279, "3000–6000 Hz_40ms_log_rms_mae_db": 1.707717511665933, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18206702853433113, "3000–6000 Hz_40ms_envelope_correlation": 0.9751404208993992, "3000–6000 Hz_40ms_modulation_mae_db": 0.3860140742888314, "mr_stft_db": 6.068307219506054, "band_mae_db": 0.19660551123822767, "rms_gain_error_db": -0.08112385788959153, "output_peak": 0.7025315215673451}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9432624198713644, "全帯域_10ms_rms_envelope_nrmse": 0.08056440489150474, "全帯域_10ms_envelope_correlation": 0.9931310696962641, "全帯域_10ms_modulation_mae_db": 0.3375368689816322, "全帯域_40ms_log_rms_mae_db": 0.4980033951587774, "全帯域_40ms_rms_envelope_nrmse": 0.03768614220668777, "全帯域_40ms_envelope_correlation": 0.998449041696651, "全帯域_40ms_modulation_mae_db": 0.21715228261279665, "300–1000 Hz_10ms_log_rms_mae_db": 1.624343086292617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11414390405160307, "300–1000 Hz_10ms_envelope_correlation": 0.9899615931379018, "300–1000 Hz_10ms_modulation_mae_db": 0.6847385229256089, "300–1000 Hz_40ms_log_rms_mae_db": 0.9361472699856604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05851395755227634, "300–1000 Hz_40ms_envelope_correlation": 0.9973295315912175, "300–1000 Hz_40ms_modulation_mae_db": 0.07659186507022621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8225658977327075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18021014840040406, "1000–3000 Hz_10ms_envelope_correlation": 0.9766167206035207, "1000–3000 Hz_10ms_modulation_mae_db": 0.12295503438322247, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1276639333983647, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10767891733923278, "1000–3000 Hz_40ms_envelope_correlation": 0.9917548825715193, "1000–3000 Hz_40ms_modulation_mae_db": 0.37015682777346853, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3569798123061854, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2599324642607737, "3000–6000 Hz_10ms_envelope_correlation": 0.9527795428586942, "3000–6000 Hz_10ms_modulation_mae_db": 0.14387989441562263, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7592134081376198, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1737723488719883, "3000–6000 Hz_40ms_envelope_correlation": 0.9774384256370194, "3000–6000 Hz_40ms_modulation_mae_db": 0.1753961127259165, "mr_stft_db": 6.071767074104829, "band_mae_db": 0.17051664070824077, "rms_gain_error_db": -0.07484891165062965, "output_peak": 0.6914723057684972}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9161232353746296, "全帯域_10ms_rms_envelope_nrmse": 0.06737350455054414, "全帯域_10ms_envelope_correlation": 0.9952584116158734, "全帯域_10ms_modulation_mae_db": 0.2187040523998307, "全帯域_40ms_log_rms_mae_db": 0.45774076441659656, "全帯域_40ms_rms_envelope_nrmse": 0.0355798037406338, "全帯域_40ms_envelope_correlation": 0.9986750754896189, "全帯域_40ms_modulation_mae_db": 0.3028371339052568, "300–1000 Hz_10ms_log_rms_mae_db": 1.6019238146575734, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10356378502213076, "300–1000 Hz_10ms_envelope_correlation": 0.9920018399879807, "300–1000 Hz_10ms_modulation_mae_db": 0.40196231671393995, "300–1000 Hz_40ms_log_rms_mae_db": 0.9273891505394473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05990866736813131, "300–1000 Hz_40ms_envelope_correlation": 0.9975119574882304, "300–1000 Hz_40ms_modulation_mae_db": 0.8855623993022269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.77019105311328, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15988192649773308, "1000–3000 Hz_10ms_envelope_correlation": 0.9817088664685693, "1000–3000 Hz_10ms_modulation_mae_db": 0.639279868407954, "1000–3000 Hz_40ms_log_rms_mae_db": 1.093948607300663, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08427366853249434, "1000–3000 Hz_40ms_envelope_correlation": 0.99514508861553, "1000–3000 Hz_40ms_modulation_mae_db": 0.7510438390472212, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2940075944777987, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26446174024537117, "3000–6000 Hz_10ms_envelope_correlation": 0.9516106548175912, "3000–6000 Hz_10ms_modulation_mae_db": 0.6295369166257861, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6947065882176977, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18220784224302158, "3000–6000 Hz_40ms_envelope_correlation": 0.9754576152137376, "3000–6000 Hz_40ms_modulation_mae_db": 0.6287673227517294, "mr_stft_db": 6.055981529468042, "band_mae_db": 0.19388008424943126, "rms_gain_error_db": -0.0953285863903707, "output_peak": 0.6993452891432758}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8796819545735499, "全帯域_10ms_rms_envelope_nrmse": 0.07438629887454569, "全帯域_10ms_envelope_correlation": 0.9942541422199966, "全帯域_10ms_modulation_mae_db": 0.2472811979668652, "全帯域_40ms_log_rms_mae_db": 0.45733385126236675, "全帯域_40ms_rms_envelope_nrmse": 0.03766248437713878, "全帯域_40ms_envelope_correlation": 0.9985987819466244, "全帯域_40ms_modulation_mae_db": 0.08466175293114299, "300–1000 Hz_10ms_log_rms_mae_db": 1.6206739171916953, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12893016908216345, "300–1000 Hz_10ms_envelope_correlation": 0.9873867608411416, "300–1000 Hz_10ms_modulation_mae_db": 0.47820858463126364, "300–1000 Hz_40ms_log_rms_mae_db": 0.9738962538865614, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06438271704178382, "300–1000 Hz_40ms_envelope_correlation": 0.9971444844610831, "300–1000 Hz_40ms_modulation_mae_db": 0.4993148681959109, "1000–3000 Hz_10ms_log_rms_mae_db": 1.771862879847098, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1534158560610431, "1000–3000 Hz_10ms_envelope_correlation": 0.9836780639448512, "1000–3000 Hz_10ms_modulation_mae_db": 0.718503325091118, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1051897419225618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09233418826442585, "1000–3000 Hz_40ms_envelope_correlation": 0.9947019407018644, "1000–3000 Hz_40ms_modulation_mae_db": 0.9302609125987599, "3000–6000 Hz_10ms_log_rms_mae_db": 2.355957554580156, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26350001150376645, "3000–6000 Hz_10ms_envelope_correlation": 0.9525964732066727, "3000–6000 Hz_10ms_modulation_mae_db": 0.5595639183234898, "3000–6000 Hz_40ms_log_rms_mae_db": 1.75409693922712, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18123549796220734, "3000–6000 Hz_40ms_envelope_correlation": 0.9761428856190861, "3000–6000 Hz_40ms_modulation_mae_db": 0.5986992214582924, "mr_stft_db": 6.061428830050981, "band_mae_db": 0.2576107929838663, "rms_gain_error_db": -0.1214733387108735, "output_peak": 0.7289298189648545}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.89904916781122, "全帯域_10ms_rms_envelope_nrmse": 0.07173286183886159, "全帯域_10ms_envelope_correlation": 0.994537028588545, "全帯域_10ms_modulation_mae_db": 0.2710124351193448, "全帯域_40ms_log_rms_mae_db": 0.4432364043793568, "全帯域_40ms_rms_envelope_nrmse": 0.039844959641721335, "全帯域_40ms_envelope_correlation": 0.9981951246021237, "全帯域_40ms_modulation_mae_db": 0.4696810621417343, "300–1000 Hz_10ms_log_rms_mae_db": 1.5797531836915648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.104184418407955, "300–1000 Hz_10ms_envelope_correlation": 0.9916169946158033, "300–1000 Hz_10ms_modulation_mae_db": 0.39509666625531065, "300–1000 Hz_40ms_log_rms_mae_db": 0.9287841766817458, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05794572161800706, "300–1000 Hz_40ms_envelope_correlation": 0.9972841988658341, "300–1000 Hz_40ms_modulation_mae_db": 0.2899239753298825, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8126234998216744, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16125343044243087, "1000–3000 Hz_10ms_envelope_correlation": 0.981225615714184, "1000–3000 Hz_10ms_modulation_mae_db": 0.7611072129749603, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1551493290635642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09433624903193236, "1000–3000 Hz_40ms_envelope_correlation": 0.9934570642932747, "1000–3000 Hz_40ms_modulation_mae_db": 1.048241819883037, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3005075648580484, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25895689666848304, "3000–6000 Hz_10ms_envelope_correlation": 0.9537857603282274, "3000–6000 Hz_10ms_modulation_mae_db": 0.5369539570100922, "3000–6000 Hz_40ms_log_rms_mae_db": 1.740937471453975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17769255485156954, "3000–6000 Hz_40ms_envelope_correlation": 0.9767582877590347, "3000–6000 Hz_40ms_modulation_mae_db": 0.2842493979981883, "mr_stft_db": 6.061812271567604, "band_mae_db": 0.1590911343115098, "rms_gain_error_db": -0.028720222888429875, "output_peak": 0.7084695576826906}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.063917327009687, "spectral_convergence": 0.27792232155002816, "spectral_convergence_status": "ok", "amplitude_floor": 2.2239575932173076e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.858106609835506, "spectral_convergence": 0.2674006153308016, "spectral_convergence_status": "ok", "amplitude_floor": 1.640120571472942e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.272713867300746, "spectral_convergence": 0.28137186471409886, "spectral_convergence_status": "ok", "amplitude_floor": 1.3411038650227006e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0016608291202231148, "output_power": 0.0016457840543110348, "delta_db": -0.03952100745375785, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001899519997387598, "output_power": 0.0018778305563085086, "delta_db": -0.04987468314933227, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002727484270807038, "output_power": 0.002638181777312131, "delta_db": -0.14457538509004772, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007661816102082716, "output_power": 0.0007779149835637708, "delta_db": 0.06600412703303797, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000247481283821024, "output_power": 0.00024281730778433967, "delta_db": -0.08262720677426968, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.389711909369124e-05, "output_power": 7.183336574526368e-05, "delta_db": 0.5084493801362593, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14850863160611746, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06596722188247806, "pitch": null, "output_rms_dbfs": -21.394007737855052, "output_peak": 0.7589347074483878, "reference_rms_dbfs": -21.328040515972575, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.064912601381979}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031196729436307625, "log_rms_mae_db": 0.9003726573023283, "rms_envelope_nrmse": 0.07954331046043907, "envelope_correlation": 0.9933189788445342, "rms_gain_error_db": -0.06589212213394093, "reference_mean_power": 0.00737113406357093, "output_mean_power": 0.007260141711964899}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.772697507541185e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010414087648688783, "output_power": 0.0010046679670046525, "reference_active": true, "delta_db": -0.15598673167810823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001272580988246754, "output_power": 0.00012448750936853707, "reference_active": true, "delta_db": -0.09559652733517476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4550125163182702e-05, "output_power": 2.6661814302152754e-05, "reference_active": true, "delta_db": 0.358359886485791}], "active_band_count": 3, "active_band_mae_db": 0.20331438183302467}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002933828481800726, "log_rms_mae_db": 0.46287220439386834, "rms_envelope_nrmse": 0.03473207711487078, "envelope_correlation": 0.9986848856105425, "rms_gain_error_db": -0.06547382791053483, "reference_mean_power": 0.007404046443755548, "output_mean_power": 0.0072932609297502535}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.623838097183376e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009238145152881926, "output_power": 0.0008950870305095869, "reference_active": true, "delta_db": -0.13719517344405624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0904498001922555e-05, "output_power": 5.995733124625019e-05, "reference_active": true, "delta_db": -0.068070741491786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8155579307174739e-06, "output_power": 1.6485855006790732e-06, "reference_active": true, "delta_db": -0.41898635011717394}], "active_band_count": 3, "active_band_mae_db": 0.20808408835100542}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027755394478816403, "log_rms_mae_db": 1.6578890873915673, "rms_envelope_nrmse": 0.11828548670211272, "envelope_correlation": 0.9892010415910282, "rms_gain_error_db": -0.10535746741380414, "reference_mean_power": 0.003209968762055962, "output_mean_power": 0.003133033649501487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3498130551666851e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006624323096451465, "output_power": 0.0006243984084496636, "reference_active": true, "delta_db": -0.2567971928121297}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.187213693276335e-05, "output_power": 5.9516313958101844e-05, "reference_active": true, "delta_db": -0.8192453159684849}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2558954262897386e-05, "output_power": 1.6480548067944653e-05, "reference_active": true, "delta_db": 1.180181714260473}], "active_band_count": 3, "active_band_mae_db": 0.7520747410136958}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025197665775423135, "log_rms_mae_db": 0.9513280175955693, "rms_envelope_nrmse": 0.05233047518552493, "envelope_correlation": 0.9978788726203535, "rms_gain_error_db": -0.10538058252570429, "reference_mean_power": 0.0032255924683410116, "output_mean_power": 0.0031482661371621265}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.287517318914353e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006130037756394711, "output_power": 0.0005805168710757713, "reference_active": true, "delta_db": -0.236483036571785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.4809964114174026e-05, "output_power": 2.814918914098428e-05, "reference_active": true, "delta_db": -0.9223768638602454}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.243410741089396e-07, "output_power": 9.060441257751229e-07, "reference_active": true, "delta_db": -0.08682902800029813}], "active_band_count": 3, "active_band_mae_db": 0.4152296428107762}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017139073768520627, "log_rms_mae_db": 1.8322230162689748, "rms_envelope_nrmse": 0.15790693941259032, "envelope_correlation": 0.9823739158155828, "rms_gain_error_db": 0.08373957824656746, "reference_mean_power": 0.0008077036407054461, "output_mean_power": 0.0008234286968594424}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.727013621661235e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024375238055045323, "output_power": 0.00023881723724606725, "reference_active": true, "delta_db": -0.08883195920734062}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.980069758207981e-05, "output_power": 3.27381849134102e-05, "reference_active": true, "delta_db": 0.4082816710698215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.347390229447928e-06, "output_power": 8.898545680622214e-06, "reference_active": true, "delta_db": 0.831859278287441}], "active_band_count": 3, "active_band_mae_db": 0.4429909695215344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015289190214292506, "log_rms_mae_db": 1.1161195568701536, "rms_envelope_nrmse": 0.09202098277073587, "envelope_correlation": 0.9937584412847066, "rms_gain_error_db": 0.08383971430214571, "reference_mean_power": 0.0008114918980720446, "output_mean_power": 0.0008273097824273606}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4774424074646414e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022848364185803233, "output_power": 0.00022585816098582995, "reference_active": true, "delta_db": -0.05019324834681572}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2508904456720676e-05, "output_power": 1.461460993810135e-05, "reference_active": true, "delta_db": 0.6756795320701076}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.746475959138801e-07, "output_power": 6.104654955100035e-07, "reference_active": true, "delta_db": 0.26259527817762324}], "active_band_count": 3, "active_band_mae_db": 0.3294893528648488}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.106734387114872e-05, "log_rms_mae_db": 2.3202017571942766, "rms_envelope_nrmse": 0.2661321832782402, "envelope_correlation": 0.9506313084593677, "rms_gain_error_db": 0.06297085960007277, "reference_mean_power": 0.0001388420753372346, "output_mean_power": 0.00014086989231339883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.164366134779655e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.596504488102423e-05, "output_power": 4.730922727022339e-05, "reference_active": true, "delta_db": 0.12518166010954085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.50200265064541e-05, "output_power": 1.3353920183381888e-05, "reference_active": true, "delta_db": -0.5106192296664237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4297410894022624e-06, "output_power": 3.5889942634261155e-06, "reference_active": true, "delta_db": 1.6941276605569997}], "active_band_count": 3, "active_band_mae_db": 0.7766428501109881}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.91579507177982e-05, "log_rms_mae_db": 1.752612207892187, "rms_envelope_nrmse": 0.187903796582517, "envelope_correlation": 0.9736672969822688, "rms_gain_error_db": 0.06290726215800434, "reference_mean_power": 0.00013952517981136655, "output_mean_power": 0.00014156090065634206}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0998934517742833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.388750580480172e-05, "output_power": 4.672701567929772e-05, "reference_active": true, "delta_db": 0.2722714484936132}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1884324982671665e-06, "output_power": 6.767448534129474e-06, "reference_active": true, "delta_db": -0.2620923724416766}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0513859982120003e-07, "output_power": 2.7753263556402495e-07, "reference_active": true, "delta_db": 1.3126667303218298}], "active_band_count": 3, "active_band_mae_db": 0.6156768504190399}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8495777583851023, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ab1737903effd98b0012bbb2c353d19f046827fcd272346ebe87b86e3f1ce0d6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9805715790719869, "standard_deviation": 0.003765233136685634, "minimum": 0.9735529201545927, "maximum": 0.9850088822438485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8263534944179147, "standard_deviation": 0.049215452811214075, "minimum": 1.77019105311328, "maximum": 1.9250068447085076, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.49990809411323966, "standard_deviation": 0.2172867534533371, "minimum": 0.12295503438322247, "maximum": 0.7611072129749603, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16410647447578616, "standard_deviation": 0.014923834134964346, "minimum": 0.14411207447588184, "maximum": 0.19094276239764227, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934707927124411, "standard_deviation": 0.0013780866760573283, "minimum": 0.9912638404120533, "maximum": 0.99514508861553, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1335156111877032, "standard_deviation": 0.035278044114023466, "minimum": 1.093948607300663, "maximum": 1.18923969890345, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6416951277573233, "standard_deviation": 0.30577374790973827, "minimum": 0.20104285385879783, "maximum": 1.048241819883037, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09495468975533944, "standard_deviation": 0.008788539936890145, "minimum": 0.08427366853249434, "maximum": 0.10801484893651833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9512695851518898, "standard_deviation": 0.001798167443468356, "minimum": 0.948698756918988, "maximum": 0.9537857603282274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3335042222470186, "standard_deviation": 0.04072315753307351, "minimum": 2.2842541108565735, "maximum": 2.403861596627954, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5391795850629726, "standard_deviation": 0.2206404558892148, "minimum": 0.14387989441562263, "maximum": 0.792179811414279, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2647732567066551, "standard_deviation": 0.004133182212523995, "minimum": 0.25895689666848304, "maximum": 0.2706727350017079, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9752499585590164, "standard_deviation": 0.0016206674832782394, "minimum": 0.9724437401336813, "maximum": 0.9774384256370194, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7424441047835604, "standard_deviation": 0.029728080184647173, "minimum": 1.6947065882176977, "maximum": 1.7891046621212041, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.45545305277618753, "standard_deviation": 0.22883692582005266, "minimum": 0.1753961127259165, "maximum": 0.7747377909859919, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18249607330027628, "standard_deviation": 0.005579348391375505, "minimum": 0.1737723488719883, "maximum": 0.19182544880912858, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9904097298919252, "standard_deviation": 0.001570135783594574, "minimum": 0.9873867608411416, "maximum": 0.9920018399879807, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6266832830845313, "standard_deviation": 0.03258867240668104, "minimum": 1.5797531836915648, "maximum": 1.6862306769110798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4799374630407429, "standard_deviation": 0.1948376534884398, "minimum": 0.1681925313654354, "maximum": 0.7520747410136958, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11177840346717269, "standard_deviation": 0.008773021696741786, "minimum": 0.10356378502213076, "maximum": 0.12893016908216345, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974415326669728, "standard_deviation": 0.0004237294294742761, "minimum": 0.9966828350880695, "maximum": 0.997962973736735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9435986871892255, "standard_deviation": 0.017770250769421212, "minimum": 0.9262895478935939, "maximum": 0.9738962538865614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4533466901836388, "standard_deviation": 0.22923304494213048, "minimum": 0.07659186507022621, "maximum": 0.8855623993022269, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05839374146027125, "standard_deviation": 0.0051065493171128535, "minimum": 0.051608906485367895, "maximum": 0.06598773246099408, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19151079766815862, "standard_deviation": 0.033867027876521176, "minimum": 0.14850863160611746, "maximum": 0.2576107929838663, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.066062101329431, "standard_deviation": 0.00737650588133899, "minimum": 6.055981529468042, "maximum": 6.080108087882764, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7169577346976566, "standard_deviation": 0.023466917056768847, "minimum": 0.6914723057684972, "maximum": 0.7589347074483878, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07887141699615673, "standard_deviation": 0.027700297455169498, "minimum": -0.1214733387108735, "maximum": -0.028720222888429875, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941775328375355, "standard_deviation": 0.0007099548849543189, "minimum": 0.9931310696962641, "maximum": 0.9952584116158734, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.919328074401807, "standard_deviation": 0.026905041384499208, "minimum": 0.8796819545735499, "maximum": 0.9646990984264217, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.25567550861599597, "standard_deviation": 0.08227204808971517, "minimum": 0.11875560478696841, "maximum": 0.38792998672741535, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07425789814653738, "standard_deviation": 0.004455534001928482, "minimum": 0.06737350455054414, "maximum": 0.08056440489150474, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998512661796853, "standard_deviation": 0.00023606593522785524, "minimum": 0.9981407442473874, "maximum": 0.9987975986361375, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46712086022693877, "standard_deviation": 0.020571485997622856, "minimum": 0.4432364043793568, "maximum": 0.4980033951587774, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2726983869392615, "standard_deviation": 0.15481637501713597, "minimum": 0.08466175293114299, "maximum": 0.5254042980345988, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037120156805561286, "standard_deviation": 0.002671479191316375, "minimum": 0.03362806012518275, "maximum": 0.041671895404425584, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0060", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.7924842834472656, "predicted_mos": 2.7924842834472656, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1b3732ce1ffeda3a4e751aed05d686e1d67f953156100e61f99aea452d5c72e8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.959058200009167, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8737167795230544, "全帯域_10ms_rms_envelope_nrmse": 0.07844697982080083, "全帯域_10ms_envelope_correlation": 0.9934963178201375, "全帯域_10ms_modulation_mae_db": 0.3020024316152155, "全帯域_40ms_log_rms_mae_db": 0.45250700808289634, "全帯域_40ms_rms_envelope_nrmse": 0.036606572740452495, "全帯域_40ms_envelope_correlation": 0.9985458400114232, "全帯域_40ms_modulation_mae_db": 0.4315875277587371, "300–1000 Hz_10ms_log_rms_mae_db": 1.3909156168437236, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11811173442486503, "300–1000 Hz_10ms_envelope_correlation": 0.9892336351809342, "300–1000 Hz_10ms_modulation_mae_db": 0.4304021565281266, "300–1000 Hz_40ms_log_rms_mae_db": 0.7689501821821285, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05575836805097109, "300–1000 Hz_40ms_envelope_correlation": 0.9975452439911062, "300–1000 Hz_40ms_modulation_mae_db": 0.42968163212896693, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4453866606860009, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16560950454487597, "1000–3000 Hz_10ms_envelope_correlation": 0.9802782733318328, "1000–3000 Hz_10ms_modulation_mae_db": 0.7038929130078367, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8863621553208508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09291974043898751, "1000–3000 Hz_40ms_envelope_correlation": 0.9938160485956631, "1000–3000 Hz_40ms_modulation_mae_db": 0.9603363109104096, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8397978110344435, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26137687593028897, "3000–6000 Hz_10ms_envelope_correlation": 0.9529770832314888, "3000–6000 Hz_10ms_modulation_mae_db": 0.3723117600562918, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5073631800687115, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17904273890385006, "3000–6000 Hz_40ms_envelope_correlation": 0.9763309254402006, "3000–6000 Hz_40ms_modulation_mae_db": 0.6231656326621108, "mr_stft_db": 5.770433588737923, "band_mae_db": 0.2216499152531267, "rms_gain_error_db": -0.07720304766803837, "output_peak": 0.7102495011274653}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9407660863250735, "全帯域_10ms_rms_envelope_nrmse": 0.07420462286087816, "全帯域_10ms_envelope_correlation": 0.9943751082577765, "全帯域_10ms_modulation_mae_db": 0.20317490666482207, "全帯域_40ms_log_rms_mae_db": 0.49735244678319773, "全帯域_40ms_rms_envelope_nrmse": 0.041140283700315666, "全帯域_40ms_envelope_correlation": 0.9983761888564399, "全帯域_40ms_modulation_mae_db": 0.39489185193563553, "300–1000 Hz_10ms_log_rms_mae_db": 1.3897250334709823, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10275277213601652, "300–1000 Hz_10ms_envelope_correlation": 0.9921600830305444, "300–1000 Hz_10ms_modulation_mae_db": 0.3664802118731248, "300–1000 Hz_40ms_log_rms_mae_db": 0.7509259846877592, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0637154902407382, "300–1000 Hz_40ms_envelope_correlation": 0.9971562249461272, "300–1000 Hz_40ms_modulation_mae_db": 0.21165289560287834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5040198044938151, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15958576066831956, "1000–3000 Hz_10ms_envelope_correlation": 0.981870954715646, "1000–3000 Hz_10ms_modulation_mae_db": 0.4090024668729361, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9322835951195267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09377717029912781, "1000–3000 Hz_40ms_envelope_correlation": 0.9939543037701641, "1000–3000 Hz_40ms_modulation_mae_db": 0.4665026123114475, "3000–6000 Hz_10ms_log_rms_mae_db": 1.892870685879045, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2580593178742308, "3000–6000 Hz_10ms_envelope_correlation": 0.9524885784590569, "3000–6000 Hz_10ms_modulation_mae_db": 0.5555497701249673, "3000–6000 Hz_40ms_log_rms_mae_db": 1.534229175310263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.175001445012313, "3000–6000 Hz_40ms_envelope_correlation": 0.9772772475837245, "3000–6000 Hz_40ms_modulation_mae_db": 0.5006226637587827, "mr_stft_db": 5.774783347845755, "band_mae_db": 0.2694976130640297, "rms_gain_error_db": -0.15074086034814238, "output_peak": 0.7250256858020359}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9298128615270769, "全帯域_10ms_rms_envelope_nrmse": 0.06443311273428387, "全帯域_10ms_envelope_correlation": 0.9956792166274925, "全帯域_10ms_modulation_mae_db": 0.08857765355103404, "全帯域_40ms_log_rms_mae_db": 0.47185273817860157, "全帯域_40ms_rms_envelope_nrmse": 0.03139289146746469, "全帯域_40ms_envelope_correlation": 0.998996205150656, "全帯域_40ms_modulation_mae_db": 0.12002586710704398, "300–1000 Hz_10ms_log_rms_mae_db": 1.4100012876716623, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10329360106076864, "300–1000 Hz_10ms_envelope_correlation": 0.9918954283260172, "300–1000 Hz_10ms_modulation_mae_db": 0.542062468640164, "300–1000 Hz_40ms_log_rms_mae_db": 0.7637986821101801, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051640623725170844, "300–1000 Hz_40ms_envelope_correlation": 0.9980709735544556, "300–1000 Hz_40ms_modulation_mae_db": 0.5078289904876261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4663193779456434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14306153771114516, "1000–3000 Hz_10ms_envelope_correlation": 0.9856250083303847, "1000–3000 Hz_10ms_modulation_mae_db": 0.5063045279322678, "1000–3000 Hz_40ms_log_rms_mae_db": 0.927017033156816, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08403623070453191, "1000–3000 Hz_40ms_envelope_correlation": 0.9953349611822633, "1000–3000 Hz_40ms_modulation_mae_db": 0.45644263529892476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8050631442397427, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23991888307015038, "3000–6000 Hz_10ms_envelope_correlation": 0.9594094868019084, "3000–6000 Hz_10ms_modulation_mae_db": 0.40679443078331334, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4739107785821666, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1568313419192656, "3000–6000 Hz_40ms_envelope_correlation": 0.9814146538722337, "3000–6000 Hz_40ms_modulation_mae_db": 0.38013343968915314, "mr_stft_db": 5.766691505347812, "band_mae_db": 0.23489122603584922, "rms_gain_error_db": -0.09251424864459555, "output_peak": 0.6967321593413385}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.926441645516427, "全帯域_10ms_rms_envelope_nrmse": 0.07022851529498832, "全帯域_10ms_envelope_correlation": 0.9949449127072296, "全帯域_10ms_modulation_mae_db": 0.32994783829358415, "全帯域_40ms_log_rms_mae_db": 0.48007382216484246, "全帯域_40ms_rms_envelope_nrmse": 0.03724422756156577, "全帯域_40ms_envelope_correlation": 0.998684584996573, "全帯域_40ms_modulation_mae_db": 0.22759892313784222, "300–1000 Hz_10ms_log_rms_mae_db": 1.3460734109291121, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1014315501400746, "300–1000 Hz_10ms_envelope_correlation": 0.9922631154869462, "300–1000 Hz_10ms_modulation_mae_db": 0.20983313267420103, "300–1000 Hz_40ms_log_rms_mae_db": 0.7560382962219192, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055866874031454926, "300–1000 Hz_40ms_envelope_correlation": 0.9977914945724984, "300–1000 Hz_40ms_modulation_mae_db": 0.36706985504200834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4576117068389778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15130097122647732, "1000–3000 Hz_10ms_envelope_correlation": 0.9862079582482992, "1000–3000 Hz_10ms_modulation_mae_db": 0.7411129270572264, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9234191749654994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1037296842843845, "1000–3000 Hz_40ms_envelope_correlation": 0.9954036052923216, "1000–3000 Hz_40ms_modulation_mae_db": 0.6568749753403765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8770034047767423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25682651251102456, "3000–6000 Hz_10ms_envelope_correlation": 0.9538794567251042, "3000–6000 Hz_10ms_modulation_mae_db": 1.038002438424932, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4680984912366926, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16799979195893894, "3000–6000 Hz_40ms_envelope_correlation": 0.9787883793639482, "3000–6000 Hz_40ms_modulation_mae_db": 0.7506496667327324, "mr_stft_db": 5.769925533321071, "band_mae_db": 0.3335907588235374, "rms_gain_error_db": -0.13590988477104432, "output_peak": 0.6759522409963813}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9024162242111669, "全帯域_10ms_rms_envelope_nrmse": 0.07642374533223621, "全帯域_10ms_envelope_correlation": 0.9938169784222375, "全帯域_10ms_modulation_mae_db": 0.1998592778004056, "全帯域_40ms_log_rms_mae_db": 0.4792520025061496, "全帯域_40ms_rms_envelope_nrmse": 0.034054353496565895, "全帯域_40ms_envelope_correlation": 0.9987070795818469, "全帯域_40ms_modulation_mae_db": 0.15770781779241297, "300–1000 Hz_10ms_log_rms_mae_db": 1.3840121177717335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10515767070360173, "300–1000 Hz_10ms_envelope_correlation": 0.991472470718172, "300–1000 Hz_10ms_modulation_mae_db": 0.35024836253967323, "300–1000 Hz_40ms_log_rms_mae_db": 0.6952756197179072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04262064967544029, "300–1000 Hz_40ms_envelope_correlation": 0.9985364788500287, "300–1000 Hz_40ms_modulation_mae_db": 0.16353358791107922, "1000–3000 Hz_10ms_log_rms_mae_db": 1.46595188479876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15208893770959786, "1000–3000 Hz_10ms_envelope_correlation": 0.984475363366076, "1000–3000 Hz_10ms_modulation_mae_db": 0.8227697191585817, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9061498197386646, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08805725366071593, "1000–3000 Hz_40ms_envelope_correlation": 0.9959169868267478, "1000–3000 Hz_40ms_modulation_mae_db": 0.754266867573818, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8320798338601443, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23738670123134845, "3000–6000 Hz_10ms_envelope_correlation": 0.9608458547423502, "3000–6000 Hz_10ms_modulation_mae_db": 0.4734069377920423, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4682701577641615, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16450438276768004, "3000–6000 Hz_40ms_envelope_correlation": 0.9798048478150153, "3000–6000 Hz_40ms_modulation_mae_db": 0.16618035540530365, "mr_stft_db": 5.780526736404959, "band_mae_db": 0.3033636697890652, "rms_gain_error_db": -0.04985794693575533, "output_peak": 0.7632348245282824}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9688383107220756, "全帯域_10ms_rms_envelope_nrmse": 0.09125321467794302, "全帯域_10ms_envelope_correlation": 0.9911687826153637, "全帯域_10ms_modulation_mae_db": 0.2989402732815183, "全帯域_40ms_log_rms_mae_db": 0.5150114200392992, "全帯域_40ms_rms_envelope_nrmse": 0.040345424794855964, "全帯域_40ms_envelope_correlation": 0.9981519662633197, "全帯域_40ms_modulation_mae_db": 0.10158324097321121, "300–1000 Hz_10ms_log_rms_mae_db": 1.4592571281978015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1435193532583454, "300–1000 Hz_10ms_envelope_correlation": 0.9841242474816025, "300–1000 Hz_10ms_modulation_mae_db": 0.3090513392266492, "300–1000 Hz_40ms_log_rms_mae_db": 0.8023839275461057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06069770673674675, "300–1000 Hz_40ms_envelope_correlation": 0.9970323420177457, "300–1000 Hz_40ms_modulation_mae_db": 0.28246125847705633, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5072569641463534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15754849943561808, "1000–3000 Hz_10ms_envelope_correlation": 0.9821795733967921, "1000–3000 Hz_10ms_modulation_mae_db": 0.8354186504898035, "1000–3000 Hz_40ms_log_rms_mae_db": 0.918826425036589, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08647381060482745, "1000–3000 Hz_40ms_envelope_correlation": 0.9943830902037556, "1000–3000 Hz_40ms_modulation_mae_db": 0.5118771924105374, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8110181429193586, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24469011231558932, "3000–6000 Hz_10ms_envelope_correlation": 0.9592455697982145, "3000–6000 Hz_10ms_modulation_mae_db": 0.4716025483031781, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4583595881608031, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1704389935828375, "3000–6000 Hz_40ms_envelope_correlation": 0.978821473292881, "3000–6000 Hz_40ms_modulation_mae_db": 0.4199734862080395, "mr_stft_db": 5.762618376362124, "band_mae_db": 0.13812471058382256, "rms_gain_error_db": -0.029986579486771088, "output_peak": 0.8808852156079888}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9500338101546045, "全帯域_10ms_rms_envelope_nrmse": 0.07455886801364772, "全帯域_10ms_envelope_correlation": 0.9941429786556669, "全帯域_10ms_modulation_mae_db": 0.11796005304746367, "全帯域_40ms_log_rms_mae_db": 0.49652416601831717, "全帯域_40ms_rms_envelope_nrmse": 0.037395391514710205, "全帯域_40ms_envelope_correlation": 0.9984932906339848, "全帯域_40ms_modulation_mae_db": 0.27544343869902566, "300–1000 Hz_10ms_log_rms_mae_db": 1.425681541030306, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10117268029835715, "300–1000 Hz_10ms_envelope_correlation": 0.9921226157608585, "300–1000 Hz_10ms_modulation_mae_db": 0.1850294440158168, "300–1000 Hz_40ms_log_rms_mae_db": 0.7769384207351144, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052371619290189235, "300–1000 Hz_40ms_envelope_correlation": 0.9978206425829699, "300–1000 Hz_40ms_modulation_mae_db": 0.2376956203140749, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5292479987483998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13373631828015173, "1000–3000 Hz_10ms_envelope_correlation": 0.9876344790100263, "1000–3000 Hz_10ms_modulation_mae_db": 0.4381691266232198, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9473309162879888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.071701130624203, "1000–3000 Hz_40ms_envelope_correlation": 0.9969589533092453, "1000–3000 Hz_40ms_modulation_mae_db": 0.5414233616040094, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8136031657177925, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26134629240035845, "3000–6000 Hz_10ms_envelope_correlation": 0.9525637317486826, "3000–6000 Hz_10ms_modulation_mae_db": 1.061902478098774, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4748742731388131, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17183948810563962, "3000–6000 Hz_40ms_envelope_correlation": 0.9779988907753898, "3000–6000 Hz_40ms_modulation_mae_db": 0.972191613331563, "mr_stft_db": 5.773548315936812, "band_mae_db": 0.1643442558452354, "rms_gain_error_db": -0.08365259100372592, "output_peak": 0.7180058232164088}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9266987169931065, "全帯域_10ms_rms_envelope_nrmse": 0.07048229241367057, "全帯域_10ms_envelope_correlation": 0.9947533667037874, "全帯域_10ms_modulation_mae_db": 0.15441498872209578, "全帯域_40ms_log_rms_mae_db": 0.42863033704469145, "全帯域_40ms_rms_envelope_nrmse": 0.03434351943707369, "全帯域_40ms_envelope_correlation": 0.998704167557351, "全帯域_40ms_modulation_mae_db": 0.15995514833003663, "300–1000 Hz_10ms_log_rms_mae_db": 1.4117642984172405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10365712130007698, "300–1000 Hz_10ms_envelope_correlation": 0.9917345534982996, "300–1000 Hz_10ms_modulation_mae_db": 0.7017246206311124, "300–1000 Hz_40ms_log_rms_mae_db": 0.7633503439048783, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055086061573732616, "300–1000 Hz_40ms_envelope_correlation": 0.9975345565672779, "300–1000 Hz_40ms_modulation_mae_db": 0.6424565231728506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5180539907146795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16818248276813458, "1000–3000 Hz_10ms_envelope_correlation": 0.9812755811288686, "1000–3000 Hz_10ms_modulation_mae_db": 0.9949954953566803, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9689934407948189, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09878864414315171, "1000–3000 Hz_40ms_envelope_correlation": 0.9953047804822475, "1000–3000 Hz_40ms_modulation_mae_db": 1.1126815411233555, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8242074600937503, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24406971324500734, "3000–6000 Hz_10ms_envelope_correlation": 0.9580982552635952, "3000–6000 Hz_10ms_modulation_mae_db": 0.869882768929085, "3000–6000 Hz_40ms_log_rms_mae_db": 1.454870776759419, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16851147826114093, "3000–6000 Hz_40ms_envelope_correlation": 0.9785643574865391, "3000–6000 Hz_40ms_modulation_mae_db": 0.6568933891203215, "mr_stft_db": 5.780563203985679, "band_mae_db": 0.21208157277396597, "rms_gain_error_db": -0.06353739468014227, "output_peak": 0.7907625998412375}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.668949602676919, "spectral_convergence": 0.27275560669276483, "spectral_convergence_status": "ok", "amplitude_floor": 2.2239575932173076e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.508016754668958, "spectral_convergence": 0.27328501125687427, "spectral_convergence_status": "ok", "amplitude_floor": 1.640120571472942e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.134334408867894, "spectral_convergence": 0.2868833539470091, "spectral_convergence_status": "ok", "amplitude_floor": 1.3411038650227006e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0016608291202231148, "output_power": 0.0016859340045845001, "delta_db": 0.06515619328090813, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001899519997387598, "output_power": 0.0018603360830120401, "delta_db": -0.09052460285070299, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002727484270807038, "output_power": 0.0026468515115952913, "delta_db": -0.13032676666007018, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007661816102082716, "output_power": 0.0007202128703726975, "delta_db": -0.26870845843539515, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000247481283821024, "output_power": 0.00024627312705529915, "delta_db": -0.021253353984458028, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.389711909369124e-05, "output_power": 7.601062251534743e-05, "delta_db": 0.7539301163072256, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2216499152531267, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07720304766803837, "pitch": null, "output_rms_dbfs": -21.405243563640614, "output_peak": 0.7102495011274653, "reference_rms_dbfs": -21.328040515972575, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.770433588737923}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031196729436307625, "log_rms_mae_db": 0.8737167795230544, "rms_envelope_nrmse": 0.07844697982080083, "envelope_correlation": 0.9934963178201375, "rms_gain_error_db": -0.07711935294535485, "reference_mean_power": 0.00737113406357093, "output_mean_power": 0.007241397283755916}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.772697507541185e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010414087648688783, "output_power": 0.0009620669273762844, "reference_active": true, "delta_db": -0.344159432110701}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001272580988246754, "output_power": 0.00013185078921131252, "reference_active": true, "delta_db": 0.15397302871340698}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4550125163182702e-05, "output_power": 2.696753145812691e-05, "reference_active": true, "delta_db": 0.40787483402153846}], "active_band_count": 3, "active_band_mae_db": 0.3020024316152155}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002933828481800726, "log_rms_mae_db": 0.45250700808289634, "rms_envelope_nrmse": 0.036606572740452495, "envelope_correlation": 0.9985458400114232, "rms_gain_error_db": -0.07699458294846069, "reference_mean_power": 0.007404046443755548, "output_mean_power": 0.007273939357099706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.623838097183376e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009238145152881926, "output_power": 0.0008544111516703258, "reference_active": true, "delta_db": -0.33917873666725173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0904498001922555e-05, "output_power": 6.435453781454968e-05, "reference_active": true, "delta_db": 0.23929807687570043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8155579307174739e-06, "output_power": 1.539506524984328e-06, "reference_active": true, "delta_db": -0.7162857697332593}], "active_band_count": 3, "active_band_mae_db": 0.4315875277587371}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027755394478816403, "log_rms_mae_db": 1.3909156168437236, "rms_envelope_nrmse": 0.11811173442486503, "envelope_correlation": 0.9892336351809342, "rms_gain_error_db": -0.09709270820717152, "reference_mean_power": 0.003209968762055962, "output_mean_power": 0.0031390015868665764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3498130551666851e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006624323096451465, "output_power": 0.0006319140951244204, "reference_active": true, "delta_db": -0.20483464081785416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.187213693276335e-05, "output_power": 6.596604703902396e-05, "reference_active": true, "delta_db": -0.3724009755442259}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2558954262897386e-05, "output_power": 1.0655072117051399e-05, "reference_active": true, "delta_db": -0.7139708532222997}], "active_band_count": 3, "active_band_mae_db": 0.4304021565281266}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025197665775423135, "log_rms_mae_db": 0.7689501821821285, "rms_envelope_nrmse": 0.05575836805097109, "envelope_correlation": 0.9975452439911062, "rms_gain_error_db": -0.09720773485953897, "reference_mean_power": 0.0032255924683410116, "output_mean_power": 0.0031541963357552393}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.287517318914353e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006130037756394711, "output_power": 0.0005824813711580564, "reference_active": true, "delta_db": -0.22181109080433}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.4809964114174026e-05, "output_power": 3.136411759363267e-05, "reference_active": true, "delta_db": -0.4527050207783546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.243410741089396e-07, "output_power": 8.023786667110905e-07, "reference_active": true, "delta_db": -0.614528784804216}], "active_band_count": 3, "active_band_mae_db": 0.42968163212896693}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017139073768520627, "log_rms_mae_db": 1.4453866606860009, "rms_envelope_nrmse": 0.16560950454487597, "envelope_correlation": 0.9802782733318328, "rms_gain_error_db": -0.2253323896088732, "reference_mean_power": 0.0008077036407054461, "output_mean_power": 0.0007668647962912913}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.727013621661235e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024375238055045323, "output_power": 0.000202972955885765, "reference_active": true, "delta_db": -0.7951068945023658}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.980069758207981e-05, "output_power": 2.6661032525554608e-05, "reference_active": true, "delta_db": -0.48349465554819815}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.347390229447928e-06, "output_power": 6.06492434134667e-06, "reference_active": true, "delta_db": -0.8330771889729466}], "active_band_count": 3, "active_band_mae_db": 0.7038929130078367}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015289190214292506, "log_rms_mae_db": 0.8863621553208508, "rms_envelope_nrmse": 0.09291974043898751, "envelope_correlation": 0.9938160485956631, "rms_gain_error_db": -0.22554790803958297, "reference_mean_power": 0.0008114918980720446, "output_mean_power": 0.0007704232798741237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4774424074646414e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022848364185803233, "output_power": 0.00018948938067710802, "reference_active": true, "delta_db": -0.8127023613121953}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2508904456720676e-05, "output_power": 1.1938317207103046e-05, "reference_active": true, "delta_db": -0.20276161249887556}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.746475959138801e-07, "output_power": 3.73978898600368e-07, "reference_active": true, "delta_db": -1.865544958920158}], "active_band_count": 3, "active_band_mae_db": 0.9603363109104096}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.106734387114872e-05, "log_rms_mae_db": 1.8397978110344435, "rms_envelope_nrmse": 0.26137687593028897, "envelope_correlation": 0.9529770832314888, "rms_gain_error_db": 0.10183538780710004, "reference_mean_power": 0.0001388420753372346, "output_mean_power": 0.00014213617872594557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.164366134779655e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.596504488102423e-05, "output_power": 4.885036006716447e-05, "reference_active": true, "delta_db": 0.2644008066743537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.50200265064541e-05, "output_power": 1.5054277406115879e-05, "reference_active": true, "delta_db": 0.00989215457669022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4297410894022624e-06, "output_power": 2.950015039078334e-06, "reference_active": true, "delta_db": 0.8426423189178316}], "active_band_count": 3, "active_band_mae_db": 0.3723117600562918}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.91579507177982e-05, "log_rms_mae_db": 1.5073631800687115, "rms_envelope_nrmse": 0.17904273890385006, "envelope_correlation": 0.9763309254402006, "rms_gain_error_db": 0.10182316988641044, "reference_mean_power": 0.00013952517981136655, "output_mean_power": 0.00014283508838777972}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0998934517742833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.388750580480172e-05, "output_power": 4.798360683548938e-05, "reference_active": true, "delta_db": 0.3875199002461329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1884324982671665e-06, "output_power": 7.4557153174133224e-06, "reference_active": true, "delta_db": 0.15855118010874736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0513859982120003e-07, "output_power": 2.782210393432554e-07, "reference_active": true, "delta_db": 1.3234258176314522}], "active_band_count": 3, "active_band_mae_db": 0.6231656326621108}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8664227334700114, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "81e85ebb86b4563900fb03a4b4fadcebe0781d9213273a8c971031c7aae0a514", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9836933989409907, "standard_deviation": 0.0026543656882178387, "minimum": 0.9802782733318328, "maximum": 0.9876344790100263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4867310485465788, "standard_deviation": 0.03142959584707239, "minimum": 1.4453866606860009, "maximum": 1.5292479987483998, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.681458228312319, "standard_deviation": 0.2104895357856029, "minimum": 0.4090024668729361, "maximum": 0.9949954953566803, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15388925154304003, "standard_deviation": 0.01146934754312317, "minimum": 0.13373631828015173, "maximum": 0.16818248276813458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995134091207801, "standard_deviation": 0.0010541283392828638, "minimum": 0.9938160485956631, "maximum": 0.9969589533092453, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9262978200525942, "standard_deviation": 0.02497587840208861, "minimum": 0.8863621553208508, "maximum": 0.9689934407948189, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6825506870716098, "standard_deviation": 0.24343607039778256, "minimum": 0.45644263529892476, "maximum": 1.1126815411233555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08993545809499123, "standard_deviation": 0.009820725245949164, "minimum": 0.071701130624203, "maximum": 0.1037296842843845, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9561885020963001, "standard_deviation": 0.0035361761608393486, "minimum": 0.9524885784590569, "maximum": 0.9608458547423502, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8369554560651273, "standard_deviation": 0.031981928887041364, "minimum": 1.8050631442397427, "maximum": 1.892870685879045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.656181641564073, "standard_deviation": 0.28696530104738377, "minimum": 0.3723117600562918, "maximum": 1.061902478098774, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2504593010722498, "standard_deviation": 0.009943593110476413, "minimum": 0.23738670123134845, "maximum": 0.26137687593028897, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9786250969537416, "standard_deviation": 0.0015452419610219345, "minimum": 0.9763309254402006, "maximum": 0.9814146538722337, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.479997052627629, "standard_deviation": 0.02707639746322192, "minimum": 1.454870776759419, "maximum": 1.534229175310263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5587262808635008, "standard_deviation": 0.24790362564572696, "minimum": 0.16618035540530365, "maximum": 0.972191613331563, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1692712075639582, "standard_deviation": 0.006718256546991375, "minimum": 0.1568313419192656, "maximum": 0.17904273890385006, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9906257686854218, "standard_deviation": 0.0028041228213337055, "minimum": 0.9841242474816025, "maximum": 0.9922631154869462, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4021788042915704, "standard_deviation": 0.033216145172025015, "minimum": 1.3460734109291121, "maximum": 1.4592571281978015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3868539670161085, "standard_deviation": 0.1711338136572647, "minimum": 0.1850294440158168, "maximum": 0.7017246206311124, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10988706041526325, "standard_deviation": 0.014647103581143708, "minimum": 0.10117268029835715, "maximum": 0.1435193532583454, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9976859946352763, "standard_deviation": 0.00048571808027564117, "minimum": 0.9970323420177457, "maximum": 0.9985364788500287, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7597076821382491, "standard_deviation": 0.030415368808116804, "minimum": 0.6952756197179072, "maximum": 0.8023839275461057, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3552975453920676, "standard_deviation": 0.16385544071906147, "minimum": 0.16353358791107922, "maximum": 0.6424565231728506, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054719674165555496, "standard_deviation": 0.006331001720085751, "minimum": 0.04262064967544029, "maximum": 0.0637154902407382, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23469296527107902, "standard_deviation": 0.06618546128603574, "minimum": 0.13812471058382256, "maximum": 0.3335907588235374, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.772386325992766, "standard_deviation": 0.006301202326406373, "minimum": 5.762618376362124, "maximum": 5.780563203985679, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.7451060063076423, "standard_deviation": 0.0658179955168608, "minimum": 0.6759522409963813, "maximum": 0.8808852156079888, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08542531919227692, "standard_deviation": 0.040976274637207315, "minimum": -0.15074086034814238, "maximum": -0.029986579486771088, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9940472077262115, "standard_deviation": 0.0013483585236304131, "minimum": 0.9911687826153637, "maximum": 0.9956792166274925, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9273405543715731, "standard_deviation": 0.02908449834132363, "minimum": 0.8737167795230544, "maximum": 0.9688383107220756, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.21185967787201737, "standard_deviation": 0.09038685429722336, "minimum": 0.08857765355103404, "maximum": 0.32994783829358415, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07500391889355609, "standard_deviation": 0.007871636805613962, "minimum": 0.06443311273428387, "maximum": 0.09125321467794302, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985824153814493, "standard_deviation": 0.00025329885222553434, "minimum": 0.9981519662633197, "maximum": 0.998996205150656, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.47765049260224945, "standard_deviation": 0.027301081158291555, "minimum": 0.42863033704469145, "maximum": 0.5150114200392992, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.23359922696674315, "standard_deviation": 0.12450571114483763, "minimum": 0.10158324097321121, "maximum": 0.4315875277587371, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.036565333089125546, "standard_deviation": 0.0032594146834358574, "minimum": 0.03139289146746469, "maximum": 0.041140283700315666, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0060", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.7700517177581787, "predicted_mos": 2.7700517177581787, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1280cee06630df722c9994055b5eaacf6e22e55d70b6ff336b9bbc89ef53d194", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.0019855997525156, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9067840456866977, "全帯域_10ms_rms_envelope_nrmse": 0.08413839589013439, "全帯域_10ms_envelope_correlation": 0.9927348935976251, "全帯域_10ms_modulation_mae_db": 0.24917255981992767, "全帯域_40ms_log_rms_mae_db": 0.48533916498102075, "全帯域_40ms_rms_envelope_nrmse": 0.042828980755003174, "全帯域_40ms_envelope_correlation": 0.9983231510487769, "全帯域_40ms_modulation_mae_db": 0.3115923033515806, "300–1000 Hz_10ms_log_rms_mae_db": 1.568556228530137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13155829340603875, "300–1000 Hz_10ms_envelope_correlation": 0.9873112759865411, "300–1000 Hz_10ms_modulation_mae_db": 0.4327813274440147, "300–1000 Hz_40ms_log_rms_mae_db": 0.9089995389649721, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07524060170936374, "300–1000 Hz_40ms_envelope_correlation": 0.9965101150318066, "300–1000 Hz_40ms_modulation_mae_db": 0.7161775137316794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.654226599281332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17045847222326432, "1000–3000 Hz_10ms_envelope_correlation": 0.9790038307812714, "1000–3000 Hz_10ms_modulation_mae_db": 0.20762783447759467, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0174818345377266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08768518812884393, "1000–3000 Hz_40ms_envelope_correlation": 0.9944724614493116, "1000–3000 Hz_40ms_modulation_mae_db": 0.17225083540894437, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9831736377407203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2519275433279854, "3000–6000 Hz_10ms_envelope_correlation": 0.9574698231772956, "3000–6000 Hz_10ms_modulation_mae_db": 0.8398933962611332, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5872259956373909, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17728080500889165, "3000–6000 Hz_40ms_envelope_correlation": 0.9776708375703896, "3000–6000 Hz_40ms_modulation_mae_db": 0.9626982100473694, "mr_stft_db": 5.862784283178711, "band_mae_db": 0.2111715775231998, "rms_gain_error_db": -0.16752888077049427, "output_peak": 0.7028618643703605}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9082072996987612, "全帯域_10ms_rms_envelope_nrmse": 0.07891558494348301, "全帯域_10ms_envelope_correlation": 0.9934573620477397, "全帯域_10ms_modulation_mae_db": 0.20601266135200072, "全帯域_40ms_log_rms_mae_db": 0.4640979643849827, "全帯域_40ms_rms_envelope_nrmse": 0.03988140545951194, "全帯域_40ms_envelope_correlation": 0.9983102415322164, "全帯域_40ms_modulation_mae_db": 0.13138690809491518, "300–1000 Hz_10ms_log_rms_mae_db": 1.4846167695048433, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1195246337109784, "300–1000 Hz_10ms_envelope_correlation": 0.9890825893773134, "300–1000 Hz_10ms_modulation_mae_db": 0.4991718102266292, "300–1000 Hz_40ms_log_rms_mae_db": 0.8938863505262619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07343443275021656, "300–1000 Hz_40ms_envelope_correlation": 0.9958754468126257, "300–1000 Hz_40ms_modulation_mae_db": 0.35162192306816603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6782399017410774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17274346967716198, "1000–3000 Hz_10ms_envelope_correlation": 0.9784646982763123, "1000–3000 Hz_10ms_modulation_mae_db": 0.8510346443303516, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0129868264350752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09380188183728508, "1000–3000 Hz_40ms_envelope_correlation": 0.9936610747641315, "1000–3000 Hz_40ms_modulation_mae_db": 1.1959770230517912, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9955748543146892, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25597533585867177, "3000–6000 Hz_10ms_envelope_correlation": 0.9548638179286404, "3000–6000 Hz_10ms_modulation_mae_db": 0.810046279471207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5635767250028603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1685767060704909, "3000–6000 Hz_40ms_envelope_correlation": 0.9790316827999045, "3000–6000 Hz_40ms_modulation_mae_db": 0.8651619862754725, "mr_stft_db": 5.86281757889396, "band_mae_db": 0.22771916617977817, "rms_gain_error_db": -0.09864039707570506, "output_peak": 0.6943580575714674}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9190283150668126, "全帯域_10ms_rms_envelope_nrmse": 0.06686820080549094, "全帯域_10ms_envelope_correlation": 0.9952915970997447, "全帯域_10ms_modulation_mae_db": 0.09401297491410354, "全帯域_40ms_log_rms_mae_db": 0.4938226402554599, "全帯域_40ms_rms_envelope_nrmse": 0.0347344191596898, "全帯域_40ms_envelope_correlation": 0.9986885206645723, "全帯域_40ms_modulation_mae_db": 0.20385612909337947, "300–1000 Hz_10ms_log_rms_mae_db": 1.4035760537184883, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09558535741512957, "300–1000 Hz_10ms_envelope_correlation": 0.9929699320186425, "300–1000 Hz_10ms_modulation_mae_db": 0.2620043699146901, "300–1000 Hz_40ms_log_rms_mae_db": 0.7583720481783878, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049320119178486194, "300–1000 Hz_40ms_envelope_correlation": 0.9980836637009911, "300–1000 Hz_40ms_modulation_mae_db": 0.17450659462512463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.697466366017696, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1581198841856501, "1000–3000 Hz_10ms_envelope_correlation": 0.982466647172946, "1000–3000 Hz_10ms_modulation_mae_db": 0.7966476832479028, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0687036837769492, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07878039780767554, "1000–3000 Hz_40ms_envelope_correlation": 0.9963488723488249, "1000–3000 Hz_40ms_modulation_mae_db": 1.1541247376576595, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9662566918600517, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2556888005739741, "3000–6000 Hz_10ms_envelope_correlation": 0.9535950966718033, "3000–6000 Hz_10ms_modulation_mae_db": 0.43572071289883657, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5632840770611662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17931838045547385, "3000–6000 Hz_40ms_envelope_correlation": 0.9756456098555094, "3000–6000 Hz_40ms_modulation_mae_db": 0.38082292689133007, "mr_stft_db": 5.870784395906047, "band_mae_db": 0.2338919984027128, "rms_gain_error_db": -0.07211778333286123, "output_peak": 0.8362565822148796}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9015347273744473, "全帯域_10ms_rms_envelope_nrmse": 0.0794217352097761, "全帯域_10ms_envelope_correlation": 0.9933246652341526, "全帯域_10ms_modulation_mae_db": 0.49257768332104285, "全帯域_40ms_log_rms_mae_db": 0.45680398139458755, "全帯域_40ms_rms_envelope_nrmse": 0.03747471341429585, "全帯域_40ms_envelope_correlation": 0.9984480023060393, "全帯域_40ms_modulation_mae_db": 0.34050973656345396, "300–1000 Hz_10ms_log_rms_mae_db": 1.4683712017638557, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11856134089078027, "300–1000 Hz_10ms_envelope_correlation": 0.9891693293683014, "300–1000 Hz_10ms_modulation_mae_db": 0.6419025863181292, "300–1000 Hz_40ms_log_rms_mae_db": 0.8410154268397679, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06057319189569012, "300–1000 Hz_40ms_envelope_correlation": 0.9971128189307941, "300–1000 Hz_40ms_modulation_mae_db": 0.4746782609203239, "1000–3000 Hz_10ms_log_rms_mae_db": 1.596242015023838, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15392436686964087, "1000–3000 Hz_10ms_envelope_correlation": 0.9828911704436915, "1000–3000 Hz_10ms_modulation_mae_db": 0.9085984066564131, "1000–3000 Hz_40ms_log_rms_mae_db": 1.021138335050534, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09222643406924444, "1000–3000 Hz_40ms_envelope_correlation": 0.9936698570393683, "1000–3000 Hz_40ms_modulation_mae_db": 0.6837772003428966, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9528025723818698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2616474245418496, "3000–6000 Hz_10ms_envelope_correlation": 0.9515329796179546, "3000–6000 Hz_10ms_modulation_mae_db": 0.5305611145079469, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5695733395929075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18100833055024554, "3000–6000 Hz_40ms_envelope_correlation": 0.975160410001809, "3000–6000 Hz_40ms_modulation_mae_db": 0.5161346633452063, "mr_stft_db": 5.867014650725984, "band_mae_db": 0.19761734236591919, "rms_gain_error_db": -0.06492244760876256, "output_peak": 0.7301450497113738}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8876009701864536, "全帯域_10ms_rms_envelope_nrmse": 0.06324136703154605, "全帯域_10ms_envelope_correlation": 0.9957961827801458, "全帯域_10ms_modulation_mae_db": 0.17834025961618052, "全帯域_40ms_log_rms_mae_db": 0.4440435984708424, "全帯域_40ms_rms_envelope_nrmse": 0.030897398621832852, "全帯域_40ms_envelope_correlation": 0.9989687477488255, "全帯域_40ms_modulation_mae_db": 0.20446676237191608, "300–1000 Hz_10ms_log_rms_mae_db": 1.4938968221324216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09646075966047776, "300–1000 Hz_10ms_envelope_correlation": 0.9928320700557095, "300–1000 Hz_10ms_modulation_mae_db": 0.36638303446846443, "300–1000 Hz_40ms_log_rms_mae_db": 0.8652324799987984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04830589482246317, "300–1000 Hz_40ms_envelope_correlation": 0.9981405496170784, "300–1000 Hz_40ms_modulation_mae_db": 0.5579922574462507, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5856540540688417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13162155343195456, "1000–3000 Hz_10ms_envelope_correlation": 0.9876319094208359, "1000–3000 Hz_10ms_modulation_mae_db": 0.5514416301974123, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9646008356868214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07000186743822795, "1000–3000 Hz_40ms_envelope_correlation": 0.9965950502096763, "1000–3000 Hz_40ms_modulation_mae_db": 0.6356927615747888, "3000–6000 Hz_10ms_log_rms_mae_db": 2.063417445774094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26482110995146246, "3000–6000 Hz_10ms_envelope_correlation": 0.9501013477874, "3000–6000 Hz_10ms_modulation_mae_db": 0.547293465154203, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6152612516077705, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18050798686158281, "3000–6000 Hz_40ms_envelope_correlation": 0.9753421744032401, "3000–6000 Hz_40ms_modulation_mae_db": 0.43998408697518565, "mr_stft_db": 5.857477070183279, "band_mae_db": 0.20770411710422607, "rms_gain_error_db": -0.06359292998312627, "output_peak": 0.7120561359949141}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.902960107310545, "全帯域_10ms_rms_envelope_nrmse": 0.08425989932934187, "全帯域_10ms_envelope_correlation": 0.9924605400809957, "全帯域_10ms_modulation_mae_db": 0.19260603926390094, "全帯域_40ms_log_rms_mae_db": 0.4935720180261551, "全帯域_40ms_rms_envelope_nrmse": 0.05056583676116702, "全帯域_40ms_envelope_correlation": 0.9970863211866343, "全帯域_40ms_modulation_mae_db": 0.2895097889506504, "300–1000 Hz_10ms_log_rms_mae_db": 1.551032860202254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13207002798726292, "300–1000 Hz_10ms_envelope_correlation": 0.9865565225489016, "300–1000 Hz_10ms_modulation_mae_db": 0.6676996061533503, "300–1000 Hz_40ms_log_rms_mae_db": 0.8873105177034512, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07525161786496956, "300–1000 Hz_40ms_envelope_correlation": 0.9953928265475865, "300–1000 Hz_40ms_modulation_mae_db": 0.42020833310248257, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6494539868615195, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16146110222200494, "1000–3000 Hz_10ms_envelope_correlation": 0.9813437388012632, "1000–3000 Hz_10ms_modulation_mae_db": 0.325009696239096, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0558310544020963, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09343965592826575, "1000–3000 Hz_40ms_envelope_correlation": 0.9938959015528963, "1000–3000 Hz_40ms_modulation_mae_db": 0.24601399666810178, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9551754772668477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25670769834989743, "3000–6000 Hz_10ms_envelope_correlation": 0.9529713896281106, "3000–6000 Hz_10ms_modulation_mae_db": 0.9871652120472142, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5440913457198415, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18396999454984714, "3000–6000 Hz_40ms_envelope_correlation": 0.974397367193228, "3000–6000 Hz_40ms_modulation_mae_db": 0.7545682757525379, "mr_stft_db": 5.864133473415791, "band_mae_db": 0.20951168193853675, "rms_gain_error_db": -0.03205484664554251, "output_peak": 0.9036767975438428}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9116682905391564, "全帯域_10ms_rms_envelope_nrmse": 0.06720012404544891, "全帯域_10ms_envelope_correlation": 0.9953130837257596, "全帯域_10ms_modulation_mae_db": 0.377963044666511, "全帯域_40ms_log_rms_mae_db": 0.4487726255039731, "全帯域_40ms_rms_envelope_nrmse": 0.03495801567344042, "全帯域_40ms_envelope_correlation": 0.9987625425128106, "全帯域_40ms_modulation_mae_db": 0.5163844539491631, "300–1000 Hz_10ms_log_rms_mae_db": 1.5346121518835802, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09844615855288795, "300–1000 Hz_10ms_envelope_correlation": 0.9927229104092864, "300–1000 Hz_10ms_modulation_mae_db": 0.3452754885325864, "300–1000 Hz_40ms_log_rms_mae_db": 0.810094194736288, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052258347019922706, "300–1000 Hz_40ms_envelope_correlation": 0.9981056691754292, "300–1000 Hz_40ms_modulation_mae_db": 0.30678999115616506, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6674450008969364, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16512686921840136, "1000–3000 Hz_10ms_envelope_correlation": 0.9803635438332494, "1000–3000 Hz_10ms_modulation_mae_db": 0.40341283223059343, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0302677703341845, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09068136419138485, "1000–3000 Hz_40ms_envelope_correlation": 0.9941068971496968, "1000–3000 Hz_40ms_modulation_mae_db": 0.6278389400693026, "3000–6000 Hz_10ms_log_rms_mae_db": 1.89706867563595, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24333683345401916, "3000–6000 Hz_10ms_envelope_correlation": 0.9593931152003878, "3000–6000 Hz_10ms_modulation_mae_db": 0.3334776770531476, "3000–6000 Hz_40ms_log_rms_mae_db": 1.460507817586508, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16047199860167555, "3000–6000 Hz_40ms_envelope_correlation": 0.9811562248744444, "3000–6000 Hz_40ms_modulation_mae_db": 0.17316425910234437, "mr_stft_db": 5.854555544219893, "band_mae_db": 0.20508730352167617, "rms_gain_error_db": -0.10769253834274685, "output_peak": 0.8257984219155763}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9030838777055579, "全帯域_10ms_rms_envelope_nrmse": 0.06850133398306894, "全帯域_10ms_envelope_correlation": 0.9950271831939671, "全帯域_10ms_modulation_mae_db": 0.2566447787386545, "全帯域_40ms_log_rms_mae_db": 0.4383974002434569, "全帯域_40ms_rms_envelope_nrmse": 0.0332643275177401, "全帯域_40ms_envelope_correlation": 0.9987566599995444, "全帯域_40ms_modulation_mae_db": 0.16927664589746136, "300–1000 Hz_10ms_log_rms_mae_db": 1.4633865045109393, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10278671292559452, "300–1000 Hz_10ms_envelope_correlation": 0.9918642088557794, "300–1000 Hz_10ms_modulation_mae_db": 0.3413669244780995, "300–1000 Hz_40ms_log_rms_mae_db": 0.7798609390752534, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04892866754598781, "300–1000 Hz_40ms_envelope_correlation": 0.9981316967678595, "300–1000 Hz_40ms_modulation_mae_db": 0.15289568484848828, "1000–3000 Hz_10ms_log_rms_mae_db": 1.678116692404867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1485387929650477, "1000–3000 Hz_10ms_envelope_correlation": 0.984244565145798, "1000–3000 Hz_10ms_modulation_mae_db": 0.08986169010387553, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0103295648794552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0821453205594269, "1000–3000 Hz_40ms_envelope_correlation": 0.994971230956588, "1000–3000 Hz_40ms_modulation_mae_db": 0.43670588059265336, "3000–6000 Hz_10ms_log_rms_mae_db": 1.982293488852888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2621246028129338, "3000–6000 Hz_10ms_envelope_correlation": 0.9520211209099481, "3000–6000 Hz_10ms_modulation_mae_db": 0.8061964156852254, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5396885274276217, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1808298240995891, "3000–6000 Hz_40ms_envelope_correlation": 0.9755623767149085, "3000–6000 Hz_40ms_modulation_mae_db": 0.5964062787030374, "mr_stft_db": 5.86694072138532, "band_mae_db": 0.07329441320572348, "rms_gain_error_db": -0.04140601244711288, "output_peak": 0.7382578471958418}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.820331265604816, "spectral_convergence": 0.26519626431158316, "spectral_convergence_status": "ok", "amplitude_floor": 2.2239575932173076e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.588609421625495, "spectral_convergence": 0.27133237423706463, "spectral_convergence_status": "ok", "amplitude_floor": 1.640120571472942e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.179412162305822, "spectral_convergence": 0.27825853385583194, "spectral_convergence_status": "ok", "amplitude_floor": 1.3411038650227006e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0016608291202231148, "output_power": 0.0016822669791708147, "delta_db": 0.05569969386728607, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001899519997387598, "output_power": 0.001873118797468141, "delta_db": -0.06078547797871306, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002727484270807038, "output_power": 0.002488122911668912, "delta_db": -0.39890424308253025, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007661816102082716, "output_power": 0.0007280957615201693, "delta_db": -0.221432210013133, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000247481283821024, "output_power": 0.00024392321495227414, "delta_db": -0.06289224802057297, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.389711909369124e-05, "output_power": 7.115621471536057e-05, "delta_db": 0.4673155921769633, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2111715775231998, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.16752888077049427, "pitch": null, "output_rms_dbfs": -21.49556939674307, "output_peak": 0.7028618643703605, "reference_rms_dbfs": -21.328040515972575, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.862784283178711}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031196729436307625, "log_rms_mae_db": 0.9067840456866977, "rms_envelope_nrmse": 0.08413839589013439, "envelope_correlation": 0.9927348935976251, "rms_gain_error_db": -0.16750645406268938, "reference_mean_power": 0.00737113406357093, "output_mean_power": 0.007092243936338336}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.772697507541185e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010414087648688783, "output_power": 0.0009951921934848085, "reference_active": true, "delta_db": -0.19714267877991393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001272580988246754, "output_power": 0.00012857654800572677, "reference_active": true, "delta_db": 0.04476330945148341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4550125163182702e-05, "output_power": 2.7581309376842652e-05, "reference_active": true, "delta_db": 0.5056116912283857}], "active_band_count": 3, "active_band_mae_db": 0.24917255981992767}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002933828481800726, "log_rms_mae_db": 0.48533916498102075, "rms_envelope_nrmse": 0.042828980755003174, "envelope_correlation": 0.9983231510487769, "rms_gain_error_db": -0.1676461054824985, "reference_mean_power": 0.007404046443755548, "output_mean_power": 0.00712368198973193}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.623838097183376e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009238145152881926, "output_power": 0.0008830897785116177, "reference_active": true, "delta_db": -0.19579923775033733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0904498001922555e-05, "output_power": 5.844598332432393e-05, "reference_active": true, "delta_db": -0.17894698107394164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8155579307174739e-06, "output_power": 2.0654428372387964e-06, "reference_active": true, "delta_db": 0.5600306912304629}], "active_band_count": 3, "active_band_mae_db": 0.3115923033515806}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027755394478816403, "log_rms_mae_db": 1.568556228530137, "rms_envelope_nrmse": 0.13155829340603875, "envelope_correlation": 0.9873112759865411, "rms_gain_error_db": -0.3390177064611613, "reference_mean_power": 0.003209968762055962, "output_mean_power": 0.0029689236921890487}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3498130551666851e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006624323096451465, "output_power": 0.0006265366207981177, "reference_active": true, "delta_db": -0.24195046285844432}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.187213693276335e-05, "output_power": 5.941532460356986e-05, "reference_active": true, "delta_db": -0.8266208352963873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2558954262897386e-05, "output_power": 1.191176756498735e-05, "reference_active": true, "delta_db": -0.22977268417721247}], "active_band_count": 3, "active_band_mae_db": 0.4327813274440147}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025197665775423135, "log_rms_mae_db": 0.9089995389649721, "rms_envelope_nrmse": 0.07524060170936374, "envelope_correlation": 0.9965101150318066, "rms_gain_error_db": -0.33942184272128706, "reference_mean_power": 0.0032255924683410116, "output_mean_power": 0.002983096565530119}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.287517318914353e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006130037756394711, "output_power": 0.0005768175267150109, "reference_active": true, "delta_db": -0.26424701416209956}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.4809964114174026e-05, "output_power": 2.755594059041825e-05, "reference_active": true, "delta_db": -1.0148833580902936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.243410741089396e-07, "output_power": 7.566447706728606e-07, "reference_active": true, "delta_db": -0.8694021689426447}], "active_band_count": 3, "active_band_mae_db": 0.7161775137316794}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017139073768520627, "log_rms_mae_db": 1.654226599281332, "rms_envelope_nrmse": 0.17045847222326432, "envelope_correlation": 0.9790038307812714, "rms_gain_error_db": -0.20181711008661363, "reference_mean_power": 0.0008077036407054461, "output_mean_power": 0.0007710283189277528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.727013621661235e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024375238055045323, "output_power": 0.00023393945803721074, "reference_active": true, "delta_db": -0.17845386374888478}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.980069758207981e-05, "output_power": 2.9660178514874067e-05, "reference_active": true, "delta_db": -0.020526696915065968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.347390229447928e-06, "output_power": 8.100715888249307e-06, "reference_active": true, "delta_db": 0.4239029427688332}], "active_band_count": 3, "active_band_mae_db": 0.20762783447759467}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015289190214292506, "log_rms_mae_db": 1.0174818345377266, "rms_envelope_nrmse": 0.08768518812884393, "envelope_correlation": 0.9944724614493116, "rms_gain_error_db": -0.20141807212317653, "reference_mean_power": 0.0008114918980720446, "output_mean_power": 0.000774715742329204}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4774424074646414e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022848364185803233, "output_power": 0.00022081595729578333, "reference_active": true, "delta_db": -0.14824657937654584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2508904456720676e-05, "output_power": 1.2957861856212059e-05, "reference_active": true, "delta_db": 0.15314070237712213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.746475959138801e-07, "output_power": 6.03862577759414e-07, "reference_active": true, "delta_db": 0.21536522447316517}], "active_band_count": 3, "active_band_mae_db": 0.17225083540894437}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.106734387114872e-05, "log_rms_mae_db": 1.9831736377407203, "rms_envelope_nrmse": 0.2519275433279854, "envelope_correlation": 0.9574698231772956, "rms_gain_error_db": 0.24523881504891293, "reference_mean_power": 0.0001388420753372346, "output_mean_power": 0.00014690784163833497}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.164366134779655e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.596504488102423e-05, "output_power": 4.901644834888035e-05, "reference_active": true, "delta_db": 0.2791415128798306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.50200265064541e-05, "output_power": 1.5649281541491278e-05, "reference_active": true, "delta_db": 0.17823704794634462}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4297410894022624e-06, "output_power": 3.906520978857012e-06, "reference_active": true, "delta_db": 2.0623016279572246}], "active_band_count": 3, "active_band_mae_db": 0.8398933962611332}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.91579507177982e-05, "log_rms_mae_db": 1.5872259956373909, "rms_envelope_nrmse": 0.17728080500889165, "envelope_correlation": 0.9776708375703896, "rms_gain_error_db": 0.24511923136841507, "reference_mean_power": 0.00013952517981136655, "output_mean_power": 0.00014762656479063842}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0998934517742833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.388750580480172e-05, "output_power": 4.856998496066449e-05, "reference_active": true, "delta_db": 0.44027069081416115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1884324982671665e-06, "output_power": 7.985422094347575e-06, "reference_active": true, "delta_db": 0.45663678141581626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0513859982120003e-07, "output_power": 3.244636898162774e-07, "reference_active": true, "delta_db": 1.991187157912131}], "active_band_count": 3, "active_band_mae_db": 0.9626982100473694}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.866043559559662, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5743d68592089957100aad25d07da7558c7b80aee87d35952e71cf315fcc058b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9820512629844209, "standard_deviation": 0.0029840397695744605, "minimum": 0.9784646982763123, "maximum": 0.9876319094208359, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6508555770370137, "standard_deviation": 0.039978659098378715, "minimum": 1.5856540540688417, "maximum": 1.697466366017696, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.516704302185405, "standard_deviation": 0.30989339246373093, "minimum": 0.08986169010387553, "maximum": 0.9085984066564131, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15774931384914073, "standard_deviation": 0.013292394242409324, "minimum": 0.13162155343195456, "maximum": 0.17274346967716198, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9947151681838117, "standard_deviation": 0.0011693954617731372, "minimum": 0.9936610747641315, "maximum": 0.9965950502096763, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0226674881378552, "standard_deviation": 0.031450238986999016, "minimum": 0.9646008356868214, "maximum": 1.0687036837769492, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6440476719207673, "standard_deviation": 0.3762083891719165, "minimum": 0.17225083540894437, "maximum": 1.1959770230517912, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08609526374504431, "standard_deviation": 0.008473405495975786, "minimum": 0.07000186743822795, "maximum": 0.09380188183728508, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9539935863651925, "standard_deviation": 0.003123524654665614, "minimum": 0.9501013477874, "maximum": 0.9593931152003878, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9744703554783887, "standard_deviation": 0.04689532834351005, "minimum": 1.89706867563595, "maximum": 2.063417445774094, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6612942841348642, "standard_deviation": 0.2297973643522965, "minimum": 0.3334776770531476, "maximum": 0.9871652120472142, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2565286686088492, "standard_deviation": 0.0067877957819033035, "minimum": 0.24333683345401916, "maximum": 0.26482110995146246, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9767458354266793, "standard_deviation": 0.002334293229851288, "minimum": 0.974397367193228, "maximum": 0.9811562248744444, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5554011349545083, "standard_deviation": 0.04520655263085826, "minimum": 1.460507817586508, "maximum": 1.6152612516077705, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5861175858865605, "standard_deviation": 0.26378811628804794, "minimum": 0.17316425910234437, "maximum": 0.9626982100473694, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17649550327472457, "standard_deviation": 0.0079207266707446, "minimum": 0.16047199860167555, "maximum": 0.18396999454984714, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9903136048275594, "standard_deviation": 0.0026066475068173777, "minimum": 0.9865565225489016, "maximum": 0.9929699320186425, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.496006074030815, "standard_deviation": 0.05383453179933203, "minimum": 1.4035760537184883, "maximum": 1.568556228530137, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4445731434419955, "standard_deviation": 0.147261180433641, "minimum": 0.2620043699146901, "maximum": 0.6676996061533503, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11187416056864377, "standard_deviation": 0.015419733390520352, "minimum": 0.09558535741512957, "maximum": 0.13207002798726292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971690983230215, "standard_deviation": 0.0011244052506316248, "minimum": 0.9953928265475865, "maximum": 0.9981405496170784, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8430964370028975, "standard_deviation": 0.0555982429750806, "minimum": 0.7583720481783878, "maximum": 0.9089995389649721, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.39435881986233506, "standard_deviation": 0.19033683382998173, "minimum": 0.15289568484848828, "maximum": 0.7161775137316794, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.060414109098387486, "standard_deviation": 0.012407647255423872, "minimum": 0.04830589482246317, "maximum": 0.07525161786496956, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19574970003022157, "standard_deviation": 0.050896242393133326, "minimum": 0.07329441320572348, "maximum": 0.2338919984027128, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.8633134647386225, "standard_deviation": 0.005273025755747034, "minimum": 5.854555544219893, "maximum": 5.870784395906047, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.767926344564782, "standard_deviation": 0.07701687182031501, "minimum": 0.6943580575714674, "maximum": 0.9036767975438428, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08099447952579396, "standard_deviation": 0.043284146905525435, "minimum": -0.16752888077049427, "maximum": -0.03205484664554251, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941756884700162, "standard_deviation": 0.0013173470293064051, "minimum": 0.9924605400809957, "maximum": 0.9957961827801458, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9051084541960539, "standard_deviation": 0.009080609276686876, "minimum": 0.8876009701864536, "maximum": 0.9190283150668126, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2559162502115402, "standard_deviation": 0.12517377517006145, "minimum": 0.09401297491410354, "maximum": 0.49257768332104285, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07406833015478627, "standard_deviation": 0.008490817102511903, "minimum": 0.06324136703154605, "maximum": 0.08425989932934187, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9984180233749274, "standard_deviation": 0.0005861230317108045, "minimum": 0.9970863211866343, "maximum": 0.9989687477488255, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4656061741575598, "standard_deviation": 0.022475485094897947, "minimum": 0.4383974002434569, "maximum": 0.4938226402554599, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.27087284103406506, "standard_deviation": 0.12294585280345471, "minimum": 0.13138690809491518, "maximum": 0.5163844539491631, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.038075637170335144, "standard_deviation": 0.0062994373938404765, "minimum": 0.030897398621832852, "maximum": 0.05056583676116702, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8455-210777-0060", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.7670865058898926, "predicted_mos": 2.7670865058898926, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "226716b7810ff3d977d828e0d24be82947dfd76095cc7e30aa0cc9b028b670ad", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.183567200321704, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8746065075057192, "全帯域_10ms_rms_envelope_nrmse": 0.06868814058894328, "全帯域_10ms_envelope_correlation": 0.9950203447642524, "全帯域_10ms_modulation_mae_db": 0.410032375474324, "全帯域_40ms_log_rms_mae_db": 0.4256405906177888, "全帯域_40ms_rms_envelope_nrmse": 0.032111815522868285, "全帯域_40ms_envelope_correlation": 0.9988814505096804, "全帯域_40ms_modulation_mae_db": 0.18614071159320064, "300–1000 Hz_10ms_log_rms_mae_db": 1.4602119759767305, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0944404599802527, "300–1000 Hz_10ms_envelope_correlation": 0.9931569756100257, "300–1000 Hz_10ms_modulation_mae_db": 0.5855252376803806, "300–1000 Hz_40ms_log_rms_mae_db": 0.7721723509153626, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046783667840955605, "300–1000 Hz_40ms_envelope_correlation": 0.998323993240208, "300–1000 Hz_40ms_modulation_mae_db": 0.27409913203124714, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6191114569576406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16916696162301725, "1000–3000 Hz_10ms_envelope_correlation": 0.9793310131296767, "1000–3000 Hz_10ms_modulation_mae_db": 1.2676768678355976, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9795464786972048, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09575959864246994, "1000–3000 Hz_40ms_envelope_correlation": 0.9932882378905983, "1000–3000 Hz_40ms_modulation_mae_db": 1.328802184911649, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8796167160310622, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24555980188569104, "3000–6000 Hz_10ms_envelope_correlation": 0.9581912445620986, "3000–6000 Hz_10ms_modulation_mae_db": 0.3400402730053762, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4725876451542923, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1764988491512142, "3000–6000 Hz_40ms_envelope_correlation": 0.9768663858261284, "3000–6000 Hz_40ms_modulation_mae_db": 0.3849993007718613, "mr_stft_db": 5.8367195401024015, "band_mae_db": 0.14141081541383804, "rms_gain_error_db": -0.06711440299818996, "output_peak": 0.6985828037473223}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9158756688701033, "全帯域_10ms_rms_envelope_nrmse": 0.0795065654039573, "全帯域_10ms_envelope_correlation": 0.9932897657693626, "全帯域_10ms_modulation_mae_db": 0.37977774428573624, "全帯域_40ms_log_rms_mae_db": 0.5017994647705349, "全帯域_40ms_rms_envelope_nrmse": 0.03528974172695284, "全帯域_40ms_envelope_correlation": 0.998606960359299, "全帯域_40ms_modulation_mae_db": 0.11045103005029731, "300–1000 Hz_10ms_log_rms_mae_db": 1.4811855007486214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11966374722130334, "300–1000 Hz_10ms_envelope_correlation": 0.9889531293944367, "300–1000 Hz_10ms_modulation_mae_db": 0.3406360821717532, "300–1000 Hz_40ms_log_rms_mae_db": 0.8235190171346619, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05493905273265485, "300–1000 Hz_40ms_envelope_correlation": 0.9976671838984225, "300–1000 Hz_40ms_modulation_mae_db": 0.30212491479491604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5671090245655266, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14213553830855177, "1000–3000 Hz_10ms_envelope_correlation": 0.9854396445615892, "1000–3000 Hz_10ms_modulation_mae_db": 0.4560329290404317, "1000–3000 Hz_40ms_log_rms_mae_db": 0.969127333950378, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07405708500030624, "1000–3000 Hz_40ms_envelope_correlation": 0.9959019843227309, "1000–3000 Hz_40ms_modulation_mae_db": 0.5335575565560361, "3000–6000 Hz_10ms_log_rms_mae_db": 1.946669995128769, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2515133598746858, "3000–6000 Hz_10ms_envelope_correlation": 0.9567064512219651, "3000–6000 Hz_10ms_modulation_mae_db": 0.8011000015384365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5263845021934561, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18151794370130878, "3000–6000 Hz_40ms_envelope_correlation": 0.9758668765691131, "3000–6000 Hz_40ms_modulation_mae_db": 0.625026953845614, "mr_stft_db": 5.842142263307115, "band_mae_db": 0.1741968229965084, "rms_gain_error_db": -0.04927105394300558, "output_peak": 0.6666608620475961}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8969358695538796, "全帯域_10ms_rms_envelope_nrmse": 0.07448149351120911, "全帯域_10ms_envelope_correlation": 0.9942082757260594, "全帯域_10ms_modulation_mae_db": 0.17221681995775207, "全帯域_40ms_log_rms_mae_db": 0.4284431707393805, "全帯域_40ms_rms_envelope_nrmse": 0.03389700191239719, "全帯域_40ms_envelope_correlation": 0.998866634912813, "全帯域_40ms_modulation_mae_db": 0.12075179067581272, "300–1000 Hz_10ms_log_rms_mae_db": 1.401127963824412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10350252345306811, "300–1000 Hz_10ms_envelope_correlation": 0.9919880253045383, "300–1000 Hz_10ms_modulation_mae_db": 0.34878574611878027, "300–1000 Hz_40ms_log_rms_mae_db": 0.7685627906387296, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058223358100455015, "300–1000 Hz_40ms_envelope_correlation": 0.9976465776724291, "300–1000 Hz_40ms_modulation_mae_db": 0.5145044922034172, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5988198246693563, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15939108929252366, "1000–3000 Hz_10ms_envelope_correlation": 0.9818296501156234, "1000–3000 Hz_10ms_modulation_mae_db": 0.49282415050539613, "1000–3000 Hz_40ms_log_rms_mae_db": 0.913912189116821, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08299752952873901, "1000–3000 Hz_40ms_envelope_correlation": 0.9953254829749131, "1000–3000 Hz_40ms_modulation_mae_db": 0.429862982323949, "3000–6000 Hz_10ms_log_rms_mae_db": 1.911168142509713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26695552895990704, "3000–6000 Hz_10ms_envelope_correlation": 0.9503539883233931, "3000–6000 Hz_10ms_modulation_mae_db": 0.5476907740894479, "3000–6000 Hz_40ms_log_rms_mae_db": 1.539478022465762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18432808606445705, "3000–6000 Hz_40ms_envelope_correlation": 0.9745938423248259, "3000–6000 Hz_40ms_modulation_mae_db": 0.37272710655012425, "mr_stft_db": 5.836288750584516, "band_mae_db": 0.2667102881034349, "rms_gain_error_db": -0.11044548384252598, "output_peak": 0.683683578434196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9282910309339817, "全帯域_10ms_rms_envelope_nrmse": 0.08917428685127975, "全帯域_10ms_envelope_correlation": 0.9915938123763296, "全帯域_10ms_modulation_mae_db": 0.47349750156958903, "全帯域_40ms_log_rms_mae_db": 0.4535518570310767, "全帯域_40ms_rms_envelope_nrmse": 0.04334286560332888, "全帯域_40ms_envelope_correlation": 0.9978948806218136, "全帯域_40ms_modulation_mae_db": 0.23894972834014658, "300–1000 Hz_10ms_log_rms_mae_db": 1.4452200190035573, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12309439042382221, "300–1000 Hz_10ms_envelope_correlation": 0.9882968294177673, "300–1000 Hz_10ms_modulation_mae_db": 0.8297656818917662, "300–1000 Hz_40ms_log_rms_mae_db": 0.7782517692396341, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059242382574174135, "300–1000 Hz_40ms_envelope_correlation": 0.9971711801637133, "300–1000 Hz_40ms_modulation_mae_db": 0.38026030664092364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5725126825238238, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1877304954021466, "1000–3000 Hz_10ms_envelope_correlation": 0.9746059080295069, "1000–3000 Hz_10ms_modulation_mae_db": 0.5902603729944069, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9703251153942302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1014215491134151, "1000–3000 Hz_40ms_envelope_correlation": 0.9924293606512804, "1000–3000 Hz_40ms_modulation_mae_db": 0.5707835197639692, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8845306133610904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24507709409636155, "3000–6000 Hz_10ms_envelope_correlation": 0.9587027819556386, "3000–6000 Hz_10ms_modulation_mae_db": 0.5425731126648695, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4830529648691517, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1612061116700954, "3000–6000 Hz_40ms_envelope_correlation": 0.980946088281277, "3000–6000 Hz_40ms_modulation_mae_db": 0.4561024764896267, "mr_stft_db": 5.833999408395268, "band_mae_db": 0.13911489337119953, "rms_gain_error_db": -0.04901600824184768, "output_peak": 0.7462323795030621}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8770569599166795, "全帯域_10ms_rms_envelope_nrmse": 0.07724595862473833, "全帯域_10ms_envelope_correlation": 0.9937081379442224, "全帯域_10ms_modulation_mae_db": 0.35792844082552944, "全帯域_40ms_log_rms_mae_db": 0.44366475690944274, "全帯域_40ms_rms_envelope_nrmse": 0.03517475477020682, "全帯域_40ms_envelope_correlation": 0.9986470963208972, "全帯域_40ms_modulation_mae_db": 0.1618510409083685, "300–1000 Hz_10ms_log_rms_mae_db": 1.4373113589456639, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12091491701671282, "300–1000 Hz_10ms_envelope_correlation": 0.9887255802186731, "300–1000 Hz_10ms_modulation_mae_db": 0.3344187270628772, "300–1000 Hz_40ms_log_rms_mae_db": 0.7911794265218238, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06162524947091259, "300–1000 Hz_40ms_envelope_correlation": 0.9970194356198852, "300–1000 Hz_40ms_modulation_mae_db": 0.08157091304175183, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6044859871920056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15349002719274724, "1000–3000 Hz_10ms_envelope_correlation": 0.9830154989400239, "1000–3000 Hz_10ms_modulation_mae_db": 0.4857054188491065, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9711045737539621, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08789119892618577, "1000–3000 Hz_40ms_envelope_correlation": 0.9942362348657865, "1000–3000 Hz_40ms_modulation_mae_db": 0.4823729865019797, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9617020654072685, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2618319191410256, "3000–6000 Hz_10ms_envelope_correlation": 0.9522612133510449, "3000–6000 Hz_10ms_modulation_mae_db": 0.6508460789258121, "3000–6000 Hz_40ms_log_rms_mae_db": 1.535367946792513, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18118350390688132, "3000–6000 Hz_40ms_envelope_correlation": 0.9754692498758911, "3000–6000 Hz_40ms_modulation_mae_db": 0.33984328079655074, "mr_stft_db": 5.8338234180167525, "band_mae_db": 0.16706255510507537, "rms_gain_error_db": -0.06136889421451129, "output_peak": 0.7494240215868911}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9380555596209939, "全帯域_10ms_rms_envelope_nrmse": 0.07507200509450196, "全帯域_10ms_envelope_correlation": 0.9941705647846879, "全帯域_10ms_modulation_mae_db": 0.16897207800186761, "全帯域_40ms_log_rms_mae_db": 0.48533723639686727, "全帯域_40ms_rms_envelope_nrmse": 0.0435981718217375, "全帯域_40ms_envelope_correlation": 0.9980852867698509, "全帯域_40ms_modulation_mae_db": 0.23849630406811398, "300–1000 Hz_10ms_log_rms_mae_db": 1.476847940867281, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.120651642093078, "300–1000 Hz_10ms_envelope_correlation": 0.9890693717729013, "300–1000 Hz_10ms_modulation_mae_db": 0.46104117507927356, "300–1000 Hz_40ms_log_rms_mae_db": 0.8259866274968675, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07739634701280551, "300–1000 Hz_40ms_envelope_correlation": 0.9956469993013883, "300–1000 Hz_40ms_modulation_mae_db": 0.4565286172524265, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5847158851317902, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1498336074533106, "1000–3000 Hz_10ms_envelope_correlation": 0.9841961050288169, "1000–3000 Hz_10ms_modulation_mae_db": 0.5379437595313468, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9858684349108178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09582601091531509, "1000–3000 Hz_40ms_envelope_correlation": 0.9937925554387109, "1000–3000 Hz_40ms_modulation_mae_db": 0.6257466642685067, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9312580188418111, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25901961791088224, "3000–6000 Hz_10ms_envelope_correlation": 0.9560129318946696, "3000–6000 Hz_10ms_modulation_mae_db": 0.6648110109174196, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4914239835468293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1797903894789911, "3000–6000 Hz_40ms_envelope_correlation": 0.9777096300594414, "3000–6000 Hz_40ms_modulation_mae_db": 0.8254714397696779, "mr_stft_db": 5.85591663442445, "band_mae_db": 0.2721720901427993, "rms_gain_error_db": -0.12617014222434814, "output_peak": 0.6698170967397191}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9436105990600588, "全帯域_10ms_rms_envelope_nrmse": 0.07359144698711052, "全帯域_10ms_envelope_correlation": 0.9942973877203051, "全帯域_10ms_modulation_mae_db": 0.36271758074237787, "全帯域_40ms_log_rms_mae_db": 0.4837726293785044, "全帯域_40ms_rms_envelope_nrmse": 0.03513634192764823, "全帯域_40ms_envelope_correlation": 0.9986844362097409, "全帯域_40ms_modulation_mae_db": 0.36295088536096004, "300–1000 Hz_10ms_log_rms_mae_db": 1.46244730207087, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11441239006402928, "300–1000 Hz_10ms_envelope_correlation": 0.9899582053509961, "300–1000 Hz_10ms_modulation_mae_db": 0.5318415802033579, "300–1000 Hz_40ms_log_rms_mae_db": 0.7920908823406038, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055067119040053014, "300–1000 Hz_40ms_envelope_correlation": 0.9977247147461366, "300–1000 Hz_40ms_modulation_mae_db": 0.801350012336353, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6005562330370047, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16595663866380808, "1000–3000 Hz_10ms_envelope_correlation": 0.9806698353626732, "1000–3000 Hz_10ms_modulation_mae_db": 0.3919401181390933, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0072291234834159, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10012717244966851, "1000–3000 Hz_40ms_envelope_correlation": 0.9935304843276649, "1000–3000 Hz_40ms_modulation_mae_db": 0.557617898777892, "3000–6000 Hz_10ms_log_rms_mae_db": 1.946809379222947, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2585996739354429, "3000–6000 Hz_10ms_envelope_correlation": 0.9537804238857395, "3000–6000 Hz_10ms_modulation_mae_db": 1.0679090396979476, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5175136752718585, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1721655755586317, "3000–6000 Hz_40ms_envelope_correlation": 0.9780539284695358, "3000–6000 Hz_40ms_modulation_mae_db": 1.1911420267637667, "mr_stft_db": 5.8371357994934145, "band_mae_db": 0.2665769462282091, "rms_gain_error_db": -0.08412164718532825, "output_peak": 0.685023675521127}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8637427005849894, "全帯域_10ms_rms_envelope_nrmse": 0.06676758733667375, "全帯域_10ms_envelope_correlation": 0.9952795760319758, "全帯域_10ms_modulation_mae_db": 0.6224025961375886, "全帯域_40ms_log_rms_mae_db": 0.41039589732570964, "全帯域_40ms_rms_envelope_nrmse": 0.03253365787252169, "全帯域_40ms_envelope_correlation": 0.9987937570258255, "全帯域_40ms_modulation_mae_db": 0.47177068040390774, "300–1000 Hz_10ms_log_rms_mae_db": 1.416402813542937, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10445763413980619, "300–1000 Hz_10ms_envelope_correlation": 0.9916628204837502, "300–1000 Hz_10ms_modulation_mae_db": 0.8569012495860765, "300–1000 Hz_40ms_log_rms_mae_db": 0.7400929745680959, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049334643605399006, "300–1000 Hz_40ms_envelope_correlation": 0.9980485435813563, "300–1000 Hz_40ms_modulation_mae_db": 0.4499542182736296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5721108489224414, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14471602378970946, "1000–3000 Hz_10ms_envelope_correlation": 0.9849982450564988, "1000–3000 Hz_10ms_modulation_mae_db": 0.1268141083383065, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9419422982571459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07578024091371194, "1000–3000 Hz_40ms_envelope_correlation": 0.995991451672385, "1000–3000 Hz_40ms_modulation_mae_db": 0.21720855780139958, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9695267715145959, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25168561306663323, "3000–6000 Hz_10ms_envelope_correlation": 0.9551416675672523, "3000–6000 Hz_10ms_modulation_mae_db": 0.7018504301500146, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5683984597597362, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17147197790191254, "3000–6000 Hz_40ms_envelope_correlation": 0.9777366300822464, "3000–6000 Hz_40ms_modulation_mae_db": 0.3685949161824162, "mr_stft_db": 5.8183693553713125, "band_mae_db": 0.1651926333750643, "rms_gain_error_db": -0.00911483708150154, "output_peak": 0.6980475389614901}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.7978510223469515, "spectral_convergence": 0.2591975555268488, "spectral_convergence_status": "ok", "amplitude_floor": 2.2239575932173076e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.574779848085773, "spectral_convergence": 0.26434463867684943, "spectral_convergence_status": "ok", "amplitude_floor": 1.640120571472942e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.137527749874481, "spectral_convergence": 0.28267581426559424, "spectral_convergence_status": "ok", "amplitude_floor": 1.3411038650227006e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0016608291202231148, "output_power": 0.0016989334162011002, "delta_db": 0.09851407580397868, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001899519997387598, "output_power": 0.0018900474288959046, "delta_db": -0.02171167487371669, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002727484270807038, "output_power": 0.002623428929675965, "delta_db": -0.1689295133892514, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007661816102082716, "output_power": 0.000729212399700505, "delta_db": -0.21477679000226887, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000247481283821024, "output_power": 0.0002429247929625039, "delta_db": -0.08070519011344315, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.389711909369124e-05, "output_power": 6.789910387572373e-05, "delta_db": 0.2638276483003694, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14141081541383804, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06711440299818996, "pitch": null, "output_rms_dbfs": -21.39515491897076, "output_peak": 0.6985828037473223, "reference_rms_dbfs": -21.328040515972575, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8367195401024015}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031196729436307625, "log_rms_mae_db": 0.8746065075057192, "rms_envelope_nrmse": 0.06868814058894328, "envelope_correlation": 0.9950203447642524, "rms_gain_error_db": -0.06707509268490845, "reference_mean_power": 0.00737113406357093, "output_mean_power": 0.0072581643982770885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.772697507541185e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010414087648688783, "output_power": 0.0010198045265857414, "reference_active": true, "delta_db": -0.09104293202973529}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001272580988246754, "output_power": 0.00013051524865085242, "reference_active": true, "delta_db": 0.10975824311427482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4550125163182702e-05, "output_power": 3.111596725195377e-05, "reference_active": true, "delta_db": 1.0292959512789617}], "active_band_count": 3, "active_band_mae_db": 0.410032375474324}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002933828481800726, "log_rms_mae_db": 0.4256405906177888, "rms_envelope_nrmse": 0.032111815522868285, "envelope_correlation": 0.9988814505096804, "rms_gain_error_db": -0.06673591269324643, "reference_mean_power": 0.007404046443755548, "output_mean_power": 0.007291141774044292}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.623838097183376e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009238145152881926, "output_power": 0.0009079778327647855, "reference_active": true, "delta_db": -0.0750953589701425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0904498001922555e-05, "output_power": 6.213453816812738e-05, "reference_active": true, "delta_db": 0.08683706747367499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8155579307174739e-06, "output_power": 1.9891112869583396e-06, "reference_active": true, "delta_db": 0.39648970833578445}], "active_band_count": 3, "active_band_mae_db": 0.18614071159320064}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027755394478816403, "log_rms_mae_db": 1.4602119759767305, "rms_envelope_nrmse": 0.0944404599802527, "envelope_correlation": 0.9931569756100257, "rms_gain_error_db": -0.10045135966107273, "reference_mean_power": 0.003209968762055962, "output_mean_power": 0.0031365749530849196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3498130551666851e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006624323096451465, "output_power": 0.0006583152443429106, "reference_active": true, "delta_db": -0.02707594692828125}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.187213693276335e-05, "output_power": 6.571980162285483e-05, "reference_active": true, "delta_db": -0.38864313841477405}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2558954262897386e-05, "output_power": 1.710169440357914e-05, "reference_active": true, "delta_db": 1.3408566276980864}], "active_band_count": 3, "active_band_mae_db": 0.5855252376803806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025197665775423135, "log_rms_mae_db": 0.7721723509153626, "rms_envelope_nrmse": 0.046783667840955605, "envelope_correlation": 0.998323993240208, "rms_gain_error_db": -0.10021208293066466, "reference_mean_power": 0.0032255924683410116, "output_mean_power": 0.0031520150907510727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.287517318914353e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006130037756394711, "output_power": 0.0006090435693817805, "reference_active": true, "delta_db": -0.028147874181267775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.4809964114174026e-05, "output_power": 3.2984736998145456e-05, "reference_active": true, "delta_db": -0.23390549958572437}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.243410741089396e-07, "output_power": 1.0516147140096183e-06, "reference_active": true, "delta_db": 0.5602440223267492}], "active_band_count": 3, "active_band_mae_db": 0.27409913203124714}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017139073768520627, "log_rms_mae_db": 1.6191114569576406, "rms_envelope_nrmse": 0.16916696162301725, "envelope_correlation": 0.9793310131296767, "rms_gain_error_db": -0.19218327602333068, "reference_mean_power": 0.0008077036407054461, "output_mean_power": 0.0007727405680824905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.727013621661235e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024375238055045323, "output_power": 0.00020938140714166895, "reference_active": true, "delta_db": -0.660107516846191}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.980069758207981e-05, "output_power": 3.555046304704047e-05, "reference_active": true, "delta_db": 0.7661883154825178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.347390229447928e-06, "output_power": 1.2700083709731838e-05, "reference_active": true, "delta_db": 2.376734771178084}], "active_band_count": 3, "active_band_mae_db": 1.2676768678355976}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015289190214292506, "log_rms_mae_db": 0.9795464786972048, "rms_envelope_nrmse": 0.09575959864246994, "envelope_correlation": 0.9932882378905983, "rms_gain_error_db": -0.19183419314826722, "reference_mean_power": 0.0008114918980720446, "output_mean_power": 0.0007764272493009849}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4774424074646414e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022848364185803233, "output_power": 0.00019983263027287932, "reference_active": true, "delta_db": -0.581887077169957}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2508904456720676e-05, "output_power": 1.6246513602210312e-05, "reference_active": true, "delta_db": 1.1354090311618825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.746475959138801e-07, "output_power": 9.689751485441704e-07, "reference_active": true, "delta_db": 2.269110446403107}], "active_band_count": 3, "active_band_mae_db": 1.328802184911649}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.106734387114872e-05, "log_rms_mae_db": 1.8796167160310622, "rms_envelope_nrmse": 0.24555980188569104, "envelope_correlation": 0.9581912445620986, "rms_gain_error_db": 0.08366369957141634, "reference_mean_power": 0.0001388420753372346, "output_mean_power": 0.00014154269702364394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.164366134779655e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.596504488102423e-05, "output_power": 4.686317475983448e-05, "reference_active": true, "delta_db": 0.08404018162354682}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.50200265064541e-05, "output_power": 1.3248280714823796e-05, "reference_active": true, "delta_db": -0.5451117738017861}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4297410894022624e-06, "output_power": 2.220563064520776e-06, "reference_active": true, "delta_db": -0.39096886359079586}], "active_band_count": 3, "active_band_mae_db": 0.3400402730053762}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.91579507177982e-05, "log_rms_mae_db": 1.4725876451542923, "rms_envelope_nrmse": 0.1764988491512142, "envelope_correlation": 0.9768663858261284, "rms_gain_error_db": 0.08379963795004589, "reference_mean_power": 0.00013952517981136655, "output_mean_power": 0.0001422435408758391}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0998934517742833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.388750580480172e-05, "output_power": 4.5296095083505246e-05, "reference_active": true, "delta_db": 0.13719863757656925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1884324982671665e-06, "output_power": 6.595754415370382e-06, "reference_active": true, "delta_db": -0.37369722062599503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0513859982120003e-07, "output_power": 1.7686307946273022e-07, "reference_active": true, "delta_db": -0.6441020441130197}], "active_band_count": 3, "active_band_mae_db": 0.3849993007718613}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8761401220800145, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "83764a5d93e3ae50553f215e4eabe273b7a9d39df31f589a4af2a75255076db2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9817607375280512, "standard_deviation": 0.0035819231727299895, "minimum": 0.9746059080295069, "maximum": 0.9854396445615892, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5899277428749488, "standard_deviation": 0.018609709334441057, "minimum": 1.5671090245655266, "maximum": 1.6191114569576406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5436497156542107, "standard_deviation": 0.3244348129142492, "minimum": 0.1268141083383065, "maximum": 1.2676768678355976, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15905254771572683, "standard_deviation": 0.015024545413165476, "minimum": 0.14213553830855177, "maximum": 0.1877304954021466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9943119740180087, "standard_deviation": 0.001301345750112972, "minimum": 0.9924293606512804, "maximum": 0.995991451672385, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.967381943445497, "standard_deviation": 0.0283076992111125, "minimum": 0.913912189116821, "maximum": 1.0072291234834159, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5932440438631728, "standard_deviation": 0.3224519505129415, "minimum": 0.21720855780139958, "maximum": 1.328802184911649, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08923254818622645, "standard_deviation": 0.010725634508175862, "minimum": 0.07405708500030624, "maximum": 0.1014215491134151, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9551438378452253, "standard_deviation": 0.0028849595357971413, "minimum": 0.9503539883233931, "maximum": 0.9587027819556386, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.928910212752157, "standard_deviation": 0.03397581364041496, "minimum": 1.8796167160310622, "maximum": 1.9695267715145959, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6646025901236655, "standard_deviation": 0.21272537158365115, "minimum": 0.3400402730053762, "maximum": 1.0679090396979476, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2550303261088287, "standard_deviation": 0.00783253869796701, "minimum": 0.24507709409636155, "maximum": 0.26695552895990704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9771553289360573, "standard_deviation": 0.0019628758847376497, "minimum": 0.9745938423248259, "maximum": 0.980946088281277, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5167759000066998, "standard_deviation": 0.032406715972804166, "minimum": 1.4725876451542923, "maximum": 1.5683984597597362, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5704884376462047, "standard_deviation": 0.30089384774555844, "minimum": 0.33984328079655074, "maximum": 1.1911420267637667, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1760203046791865, "standard_deviation": 0.00751523181292146, "minimum": 0.1612061116700954, "maximum": 0.18432808606445705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.990226367194136, "standard_deviation": 0.001803269423182064, "minimum": 0.9882968294177673, "maximum": 0.9931569756100257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4475943593725091, "standard_deviation": 0.028302869620360264, "minimum": 1.401127963824412, "maximum": 1.4811855007486214, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5361144349742832, "standard_deviation": 0.21085401076194438, "minimum": 0.3344187270628772, "maximum": 0.8569012495860765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11264221304900908, "standard_deviation": 0.010529530839013103, "minimum": 0.0944404599802527, "maximum": 0.12309439042382221, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974060785279424, "standard_deviation": 0.0008265162137854647, "minimum": 0.9956469993013883, "maximum": 0.998323993240208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7864819798569724, "standard_deviation": 0.028614242479886403, "minimum": 0.7400929745680959, "maximum": 0.8259866274968675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4075490758218331, "standard_deviation": 0.2092999935284298, "minimum": 0.08157091304175183, "maximum": 0.801350012336353, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.057826477547176215, "standard_deviation": 0.009331811427142664, "minimum": 0.046783667840955605, "maximum": 0.07739634701280551, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1990546305920161, "standard_deviation": 0.0587808689330505, "minimum": 0.13911489337119953, "maximum": 0.2721720901427993, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.836799396211903, "standard_deviation": 0.010356213539791556, "minimum": 5.8183693553713125, "maximum": 5.85591663442445, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6996839945676755, "standard_deviation": 0.031843838099689124, "minimum": 0.6666608620475961, "maximum": 0.7494240215868911, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0695778087164073, "standard_deviation": 0.037126732439878185, "minimum": -0.12617014222434814, "maximum": -0.00911483708150154, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939459831396493, "standard_deviation": 0.0011461327068658235, "minimum": 0.9915938123763296, "maximum": 0.9952795760319758, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9047718620058007, "standard_deviation": 0.03097362272134188, "minimum": 0.8637427005849894, "maximum": 0.9436105990600588, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3684431421243456, "standard_deviation": 0.1492010992815087, "minimum": 0.16897207800186761, "maximum": 0.6224025961375886, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07556593554980175, "standard_deviation": 0.006903124013716235, "minimum": 0.06676758733667375, "maximum": 0.08917428685127975, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985575628412401, "standard_deviation": 0.00036741608993103367, "minimum": 0.9978948806218136, "maximum": 0.9988814505096804, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45407570039616313, "standard_deviation": 0.032991163984377424, "minimum": 0.41039589732570964, "maximum": 0.5017994647705349, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.23642027142510097, "standard_deviation": 0.12464038030695473, "minimum": 0.11045103005029731, "maximum": 0.47177068040390774, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.036385543894707675, "standard_deviation": 0.004534758562517682, "minimum": 0.032111815522868285, "maximum": 0.0435981718217375, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8463-294828-0003", "language": "en", "style": "normal", "speaker": "librispeech_8463", "source_utterance_id": "8463-294828-0003", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8463-294828-0003.wav", "source_file_sha256": "fabd13a8ede5e3cc4eaa31b48aab9c5e703a4cb10e1678da1670ca707feb5ec6", "decoded_pcm_sha256": "cd223f40e942ab9ac205813cbe938225779742060f5fe247eb3e2fd12df52eba", "original_source_sha256": "3790b43d50403978fd6ba4d90f598e8e6afddfb630a3bf5b4110bd432517f1e9", "corpus_entry_id": "librispeech_8463-294828-0003"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.328335285186768, "predicted_mos": 4.328335285186768, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cc028d09fdf0b9dcddce85841840fe9d1e3bddfcaccd48cb8cb2ea05dd1a0fdb", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4663278219995846, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8463-294828-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "cc028d09fdf0b9dcddce85841840fe9d1e3bddfcaccd48cb8cb2ea05dd1a0fdb", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "299a537dc8c17a869309de103cb4a613416f5a2ddbabf2579d3eae9635449db9"}, "rows": [{"case": "librispeech_8463-294828-0003", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.50744891166687, "predicted_mos": 2.50744891166687, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9e3fc248fc4a94bc515f23faa01289078ae5700e0be4927e3c2d0488c4b083dc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.283196587002749, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8463-294828-0003", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "9e3fc248fc4a94bc515f23faa01289078ae5700e0be4927e3c2d0488c4b083dc", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9409603432180674, "全帯域_10ms_rms_envelope_nrmse": 0.08181188787303807, "全帯域_10ms_envelope_correlation": 0.9931216106312178, "全帯域_10ms_modulation_mae_db": 0.7763739515979259, "全帯域_40ms_log_rms_mae_db": 0.47925387592455526, "全帯域_40ms_rms_envelope_nrmse": 0.04026670110075603, "全帯域_40ms_envelope_correlation": 0.998164752213635, "全帯域_40ms_modulation_mae_db": 0.9203057145282454, "300–1000 Hz_10ms_log_rms_mae_db": 1.5321024921007833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0857758626118949, "300–1000 Hz_10ms_envelope_correlation": 0.9937283940288572, "300–1000 Hz_10ms_modulation_mae_db": 0.5203090941011252, "300–1000 Hz_40ms_log_rms_mae_db": 0.8271824428652707, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04298196212186707, "300–1000 Hz_40ms_envelope_correlation": 0.9983155320209899, "300–1000 Hz_40ms_modulation_mae_db": 0.3145211737053974, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4722974226447056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14917519282526792, "1000–3000 Hz_10ms_envelope_correlation": 0.9815311314522039, "1000–3000 Hz_10ms_modulation_mae_db": 1.1256745615491066, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8874636211347843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08354570877522628, "1000–3000 Hz_40ms_envelope_correlation": 0.9938695688220645, "1000–3000 Hz_40ms_modulation_mae_db": 1.416016269143664, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5505791350782796, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1887326405449941, "3000–6000 Hz_10ms_envelope_correlation": 0.9678504356311881, "3000–6000 Hz_10ms_modulation_mae_db": 0.9210636761542954, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9187018286905407, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11829424995608259, "3000–6000 Hz_40ms_envelope_correlation": 0.9860669089727837, "3000–6000 Hz_40ms_modulation_mae_db": 1.2678481613938983, "mr_stft_db": 5.684599560609076, "band_mae_db": 0.11500911757576053, "rms_gain_error_db": -0.047406335781754266, "output_peak": 0.4845462845904768}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9729106931564656, "全帯域_10ms_rms_envelope_nrmse": 0.08309881216190194, "全帯域_10ms_envelope_correlation": 0.9930013214159361, "全帯域_10ms_modulation_mae_db": 0.40086780464662164, "全帯域_40ms_log_rms_mae_db": 0.519086229615509, "全帯域_40ms_rms_envelope_nrmse": 0.04365680362075308, "全帯域_40ms_envelope_correlation": 0.9979823916930751, "全帯域_40ms_modulation_mae_db": 0.8154516633061903, "300–1000 Hz_10ms_log_rms_mae_db": 1.5682388716721138, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08982903384818901, "300–1000 Hz_10ms_envelope_correlation": 0.9932083505927745, "300–1000 Hz_10ms_modulation_mae_db": 1.0841346304195034, "300–1000 Hz_40ms_log_rms_mae_db": 0.8393442263992617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0463819180607835, "300–1000 Hz_40ms_envelope_correlation": 0.9981809101115452, "300–1000 Hz_40ms_modulation_mae_db": 0.37643662244378806, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5226820212091807, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1460490107808108, "1000–3000 Hz_10ms_envelope_correlation": 0.9824575850681528, "1000–3000 Hz_10ms_modulation_mae_db": 0.6108672353794651, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8825596042717615, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08345885755968824, "1000–3000 Hz_40ms_envelope_correlation": 0.994150390923239, "1000–3000 Hz_40ms_modulation_mae_db": 0.7701710782639792, "3000–6000 Hz_10ms_log_rms_mae_db": 1.553085797583856, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1959753657153711, "3000–6000 Hz_10ms_envelope_correlation": 0.9656121694922098, "3000–6000 Hz_10ms_modulation_mae_db": 1.2370565953868045, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9196986890314516, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1270644312790349, "3000–6000 Hz_40ms_envelope_correlation": 0.9845044754808913, "3000–6000 Hz_40ms_modulation_mae_db": 1.4515359597373314, "mr_stft_db": 5.701242177291424, "band_mae_db": 0.17858842385885898, "rms_gain_error_db": -0.11038472042316144, "output_peak": 0.4606524774439077}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9545739043475746, "全帯域_10ms_rms_envelope_nrmse": 0.08397423740634208, "全帯域_10ms_envelope_correlation": 0.9927420593522751, "全帯域_10ms_modulation_mae_db": 0.2808273992156099, "全帯域_40ms_log_rms_mae_db": 0.5174155754657245, "全帯域_40ms_rms_envelope_nrmse": 0.0450239238449303, "全帯域_40ms_envelope_correlation": 0.9976756100147409, "全帯域_40ms_modulation_mae_db": 0.1360425133198571, "300–1000 Hz_10ms_log_rms_mae_db": 1.5565288814608818, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10206477827412859, "300–1000 Hz_10ms_envelope_correlation": 0.9911032615211861, "300–1000 Hz_10ms_modulation_mae_db": 1.0958548172433358, "300–1000 Hz_40ms_log_rms_mae_db": 0.8929758400307082, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06007471110753196, "300–1000 Hz_40ms_envelope_correlation": 0.9966541554028725, "300–1000 Hz_40ms_modulation_mae_db": 0.723915698037764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.559901909764981, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14986254043373715, "1000–3000 Hz_10ms_envelope_correlation": 0.9814156302802707, "1000–3000 Hz_10ms_modulation_mae_db": 0.44341409862596776, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9211166034247948, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07979917276173457, "1000–3000 Hz_40ms_envelope_correlation": 0.9945602762024669, "1000–3000 Hz_40ms_modulation_mae_db": 0.8651678228874548, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5065988370056136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18969328481777542, "3000–6000 Hz_10ms_envelope_correlation": 0.9695839324140142, "3000–6000 Hz_10ms_modulation_mae_db": 1.157960230992094, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9177931869055809, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13771513564454055, "3000–6000 Hz_40ms_envelope_correlation": 0.9822056121046815, "3000–6000 Hz_40ms_modulation_mae_db": 1.309399243464205, "mr_stft_db": 5.6914084274935774, "band_mae_db": 0.17171432210380455, "rms_gain_error_db": -0.025311617909254097, "output_peak": 0.4561345456495364}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8864906337957362, "全帯域_10ms_rms_envelope_nrmse": 0.07809620207829665, "全帯域_10ms_envelope_correlation": 0.9937792101560129, "全帯域_10ms_modulation_mae_db": 0.1492722743348619, "全帯域_40ms_log_rms_mae_db": 0.4641716641789501, "全帯域_40ms_rms_envelope_nrmse": 0.040614013587884815, "全帯域_40ms_envelope_correlation": 0.998204610471483, "全帯域_40ms_modulation_mae_db": 0.33349870929585923, "300–1000 Hz_10ms_log_rms_mae_db": 1.5314259843798215, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09436701093666836, "300–1000 Hz_10ms_envelope_correlation": 0.9924406861610383, "300–1000 Hz_10ms_modulation_mae_db": 1.2017814092686538, "300–1000 Hz_40ms_log_rms_mae_db": 0.8684104184870819, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05495042677839348, "300–1000 Hz_40ms_envelope_correlation": 0.9972984927674805, "300–1000 Hz_40ms_modulation_mae_db": 0.6601206024888725, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4516455847034961, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12945165641152714, "1000–3000 Hz_10ms_envelope_correlation": 0.9861920383079402, "1000–3000 Hz_10ms_modulation_mae_db": 0.5135430004318831, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8333372401646011, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07276535925978594, "1000–3000 Hz_40ms_envelope_correlation": 0.9954498747016636, "1000–3000 Hz_40ms_modulation_mae_db": 0.6013951815994512, "3000–6000 Hz_10ms_log_rms_mae_db": 1.546519567354451, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20329961288040657, "3000–6000 Hz_10ms_envelope_correlation": 0.9629452573109601, "3000–6000 Hz_10ms_modulation_mae_db": 0.276688974594923, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9386592851506371, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12656684588167605, "3000–6000 Hz_40ms_envelope_correlation": 0.9834881263106313, "3000–6000 Hz_40ms_modulation_mae_db": 0.1984693149097048, "mr_stft_db": 5.67586263190423, "band_mae_db": 0.1469118808542931, "rms_gain_error_db": -0.08664991333501966, "output_peak": 0.472557640320544}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9449160155343583, "全帯域_10ms_rms_envelope_nrmse": 0.08202938987181947, "全帯域_10ms_envelope_correlation": 0.9930960730585622, "全帯域_10ms_modulation_mae_db": 0.2229007125048171, "全帯域_40ms_log_rms_mae_db": 0.5229287782687607, "全帯域_40ms_rms_envelope_nrmse": 0.04368925006186054, "全帯域_40ms_envelope_correlation": 0.997862894217535, "全帯域_40ms_modulation_mae_db": 0.3918135952905606, "300–1000 Hz_10ms_log_rms_mae_db": 1.5335915378421585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09163573155924525, "300–1000 Hz_10ms_envelope_correlation": 0.9928344628496611, "300–1000 Hz_10ms_modulation_mae_db": 0.47496581388936887, "300–1000 Hz_40ms_log_rms_mae_db": 0.8314149113576671, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048955986126022075, "300–1000 Hz_40ms_envelope_correlation": 0.9978039192373969, "300–1000 Hz_40ms_modulation_mae_db": 0.30701664046253824, "1000–3000 Hz_10ms_log_rms_mae_db": 1.445835548999191, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1433180767091899, "1000–3000 Hz_10ms_envelope_correlation": 0.9832702997403233, "1000–3000 Hz_10ms_modulation_mae_db": 0.43127159807954113, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8324631138868454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08432046883200374, "1000–3000 Hz_40ms_envelope_correlation": 0.9942206966374432, "1000–3000 Hz_40ms_modulation_mae_db": 0.751663910263134, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5409945371604554, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17657864716143304, "3000–6000 Hz_10ms_envelope_correlation": 0.972896687460763, "3000–6000 Hz_10ms_modulation_mae_db": 0.7153519665572176, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9016615775897309, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.117942022986709, "3000–6000 Hz_40ms_envelope_correlation": 0.986272550129584, "3000–6000 Hz_40ms_modulation_mae_db": 0.8530364745126388, "mr_stft_db": 5.683648531421441, "band_mae_db": 0.1484110755616818, "rms_gain_error_db": -0.06337017308118563, "output_peak": 0.4730125034568989}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9354489083455862, "全帯域_10ms_rms_envelope_nrmse": 0.09261907709066672, "全帯域_10ms_envelope_correlation": 0.9911676514911495, "全帯域_10ms_modulation_mae_db": 0.07698266500425553, "全帯域_40ms_log_rms_mae_db": 0.5115600683719764, "全帯域_40ms_rms_envelope_nrmse": 0.052037293495889196, "全帯域_40ms_envelope_correlation": 0.996896552033618, "全帯域_40ms_modulation_mae_db": 0.24939208657246303, "300–1000 Hz_10ms_log_rms_mae_db": 1.5735987200429067, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11540744401223116, "300–1000 Hz_10ms_envelope_correlation": 0.9886853170974981, "300–1000 Hz_10ms_modulation_mae_db": 1.4135702550963607, "300–1000 Hz_40ms_log_rms_mae_db": 0.8421907764156596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06837015053576961, "300–1000 Hz_40ms_envelope_correlation": 0.9956832007799197, "300–1000 Hz_40ms_modulation_mae_db": 0.9071128533677498, "1000–3000 Hz_10ms_log_rms_mae_db": 1.494991617103928, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14356405380122567, "1000–3000 Hz_10ms_envelope_correlation": 0.9831626349182516, "1000–3000 Hz_10ms_modulation_mae_db": 0.4102673071899501, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8092497530056892, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07876314589348024, "1000–3000 Hz_40ms_envelope_correlation": 0.9950065684912803, "1000–3000 Hz_40ms_modulation_mae_db": 0.6304528724632582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4966772492244262, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19883801122146186, "3000–6000 Hz_10ms_envelope_correlation": 0.9644727976513433, "3000–6000 Hz_10ms_modulation_mae_db": 0.25059727213531674, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8816189612298339, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1251563100202795, "3000–6000 Hz_40ms_envelope_correlation": 0.9837494209643802, "3000–6000 Hz_40ms_modulation_mae_db": 0.49551116837451703, "mr_stft_db": 5.683446772515754, "band_mae_db": 0.16185568510027745, "rms_gain_error_db": -0.03098129705331191, "output_peak": 0.4705600581259536}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9084936048587655, "全帯域_10ms_rms_envelope_nrmse": 0.07908261268061866, "全帯域_10ms_envelope_correlation": 0.9936100193779, "全帯域_10ms_modulation_mae_db": 0.23830610815746414, "全帯域_40ms_log_rms_mae_db": 0.4583256293552771, "全帯域_40ms_rms_envelope_nrmse": 0.041417787371615895, "全帯域_40ms_envelope_correlation": 0.9981139016362276, "全帯域_40ms_modulation_mae_db": 0.5181935383610715, "300–1000 Hz_10ms_log_rms_mae_db": 1.5573248396484203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09613062863617315, "300–1000 Hz_10ms_envelope_correlation": 0.9921228559541119, "300–1000 Hz_10ms_modulation_mae_db": 1.588981967910995, "300–1000 Hz_40ms_log_rms_mae_db": 0.8238309998881223, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053101228141157544, "300–1000 Hz_40ms_envelope_correlation": 0.9974336455298803, "300–1000 Hz_40ms_modulation_mae_db": 1.0015476302706658, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4790249539389566, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1459813577004193, "1000–3000 Hz_10ms_envelope_correlation": 0.9824650460899816, "1000–3000 Hz_10ms_modulation_mae_db": 0.37051334325455526, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8389959920950696, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07572087739378039, "1000–3000 Hz_40ms_envelope_correlation": 0.9952840519416205, "1000–3000 Hz_40ms_modulation_mae_db": 0.7367343452731449, "3000–6000 Hz_10ms_log_rms_mae_db": 1.516694758415329, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17682063684476607, "3000–6000 Hz_10ms_envelope_correlation": 0.9717271034284868, "3000–6000 Hz_10ms_modulation_mae_db": 0.3248966985134859, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8838613787877466, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10977332493886693, "3000–6000 Hz_40ms_envelope_correlation": 0.9875306349557056, "3000–6000 Hz_40ms_modulation_mae_db": 0.4931562948284756, "mr_stft_db": 5.682733371506682, "band_mae_db": 0.08668582414533721, "rms_gain_error_db": -0.08062420637037755, "output_peak": 0.4612274296832572}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9531015219494158, "全帯域_10ms_rms_envelope_nrmse": 0.08187879309197099, "全帯域_10ms_envelope_correlation": 0.9931706912706544, "全帯域_10ms_modulation_mae_db": 0.3348572001525882, "全帯域_40ms_log_rms_mae_db": 0.5127128966382815, "全帯域_40ms_rms_envelope_nrmse": 0.0463969645925157, "全帯域_40ms_envelope_correlation": 0.9976544621446864, "全帯域_40ms_modulation_mae_db": 0.21035382360953592, "300–1000 Hz_10ms_log_rms_mae_db": 1.5506590772830597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10122389665017076, "300–1000 Hz_10ms_envelope_correlation": 0.9913518909663092, "300–1000 Hz_10ms_modulation_mae_db": 0.9811751774981944, "300–1000 Hz_40ms_log_rms_mae_db": 0.8976487140086368, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06232353575891298, "300–1000 Hz_40ms_envelope_correlation": 0.9965879812003324, "300–1000 Hz_40ms_modulation_mae_db": 0.661982643223982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4658847426916686, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13638039659790335, "1000–3000 Hz_10ms_envelope_correlation": 0.9845606331056733, "1000–3000 Hz_10ms_modulation_mae_db": 0.304359859425148, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8437461354268531, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08219589789223924, "1000–3000 Hz_40ms_envelope_correlation": 0.9939392201193173, "1000–3000 Hz_40ms_modulation_mae_db": 0.2659070554498714, "3000–6000 Hz_10ms_log_rms_mae_db": 1.531623049854187, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17494102883757917, "3000–6000 Hz_10ms_envelope_correlation": 0.9726352541097245, "3000–6000 Hz_10ms_modulation_mae_db": 0.4069716976605813, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9373775554905339, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11820673797864183, "3000–6000 Hz_40ms_envelope_correlation": 0.9856431210923312, "3000–6000 Hz_40ms_modulation_mae_db": 0.37179932732001014, "mr_stft_db": 5.687932164335185, "band_mae_db": 0.1387651736977462, "rms_gain_error_db": -0.09738801292483354, "output_peak": 0.4523830749575902}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.729032503835048, "spectral_convergence": 0.3092110676085762, "spectral_convergence_status": "ok", "amplitude_floor": 1.2198198453862723e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.477340939680744, "spectral_convergence": 0.3025458978821403, "spectral_convergence_status": "ok", "amplitude_floor": 7.927151445605076e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.847425238311437, "spectral_convergence": 0.31917146449019773, "spectral_convergence_status": "ok", "amplitude_floor": 6.454878545778745e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00013983470011737857, "output_power": 0.0001485087462528693, "delta_db": 0.2613707630023353, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042159782923140474, "output_power": 0.00041892069441587976, "delta_db": -0.027665513054054677, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017365691446010552, "output_power": 0.001724052354052564, "delta_db": -0.03141630353573062, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006597865121791135, "output_power": 0.0006401710684368383, "delta_db": -0.1310739017680377, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001856647778551521, "output_power": 0.00017768375627231215, "delta_db": -0.19081795581517805, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000145726779943178, "output_power": 0.00014413463139043383, "delta_db": -0.04771026827922689, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11500911757576053, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.047406335781754266, "pitch": null, "output_rms_dbfs": -24.876530265403936, "output_peak": 0.4845462845904768, "reference_rms_dbfs": -24.82912392962218, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.684599560609076}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017345649672524565, "log_rms_mae_db": 0.9409603432180674, "rms_envelope_nrmse": 0.08181188787303807, "envelope_correlation": 0.9931216106312178, "rms_gain_error_db": -0.04758337927204971, "reference_mean_power": 0.003291322552458033, "output_mean_power": 0.0032554580823069466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4377088226614287e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006890430591645026, "output_power": 0.0006825676869902069, "reference_active": true, "delta_db": -0.04100637596307692}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.716540294963627e-05, "output_power": 9.090108033145976e-05, "reference_active": true, "delta_db": -0.28942611341123947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.855953967229281e-05, "output_power": 1.8025290248448324e-05, "reference_active": true, "delta_db": -1.9986893654194613}], "active_band_count": 3, "active_band_mae_db": 0.7763739515979259}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015649665907765232, "log_rms_mae_db": 0.47925387592455526, "rms_envelope_nrmse": 0.04026670110075603, "envelope_correlation": 0.998164752213635, "rms_gain_error_db": -0.047719668319157464, "reference_mean_power": 0.003302205038848876, "output_mean_power": 0.003266119487755009}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.335462563779621e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377316706455328, "output_power": 0.0006342174659163545, "reference_active": true, "delta_db": -0.023997871638760604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.741968368909027e-05, "output_power": 4.346433228000672e-05, "reference_active": true, "delta_db": -0.3782564053806123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7622617238454117e-06, "output_power": 1.0237740611106685e-06, "reference_active": true, "delta_db": -2.3586628665653633}], "active_band_count": 3, "active_band_mae_db": 0.9203057145282454}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014746435244218596, "log_rms_mae_db": 1.5321024921007833, "rms_envelope_nrmse": 0.0857758626118949, "envelope_correlation": 0.9937283940288572, "rms_gain_error_db": -0.038584419793625126, "reference_mean_power": 0.001906578877331925, "output_mean_power": 0.0018897151082278977}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.936060821906154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045087668804334914, "output_power": 0.00043909050375932006, "reference_active": true, "delta_db": -0.11503736601296952}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.461610811295139e-05, "output_power": 5.356388030243518e-05, "reference_active": true, "delta_db": -0.08448718667298585}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.1974528143292685e-06, "output_power": 8.479173108964339e-06, "reference_active": true, "delta_db": 1.3614027296174203}], "active_band_count": 3, "active_band_mae_db": 0.5203090941011252}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013246473438636124, "log_rms_mae_db": 0.8271824428652707, "rms_envelope_nrmse": 0.04298196212186707, "envelope_correlation": 0.9983155320209899, "rms_gain_error_db": -0.03864745520078431, "reference_mean_power": 0.0019129188838057705, "output_mean_power": 0.0018959715179195627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.520895123968316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004268128119473447, "output_power": 0.00041666750643181447, "reference_active": true, "delta_db": -0.10447813897716403}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.526792747799959e-05, "output_power": 2.4717450220507838e-05, "reference_active": true, "delta_db": -0.09565953561063197}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.02211562944459e-07, "output_power": 5.959767154322404e-07, "reference_active": true, "delta_db": 0.7434258465283962}], "active_band_count": 3, "active_band_mae_db": 0.3145211737053974}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.50803654710503e-05, "log_rms_mae_db": 1.4722974226447056, "rms_envelope_nrmse": 0.14917519282526792, "envelope_correlation": 0.9815311314522039, "rms_gain_error_db": -0.16710318176146297, "reference_mean_power": 0.0007310287661738595, "output_mean_power": 0.0007034352842836295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3756975679714823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020778711952139343, "output_power": 0.00019692730565829295, "reference_active": true, "delta_db": -0.2331268368796564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.101230284766401e-05, "output_power": 3.704097587123143e-05, "reference_active": true, "delta_db": -0.4423173571787862}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.417981343795666e-05, "output_power": 7.612241674362703e-06, "reference_active": true, "delta_db": -2.7015794905888773}], "active_band_count": 3, "active_band_mae_db": 1.1256745615491066}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.582207074769948e-05, "log_rms_mae_db": 0.8874636211347843, "rms_envelope_nrmse": 0.08354570877522628, "envelope_correlation": 0.9938695688220645, "rms_gain_error_db": -0.16728709191564267, "reference_mean_power": 0.000733846607513954, "output_mean_power": 0.000706116860636713}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9980296656839734e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019557456921819033, "output_power": 0.00018345578797689756, "reference_active": true, "delta_db": -0.27780964148470666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6985486820992048e-05, "output_power": 1.577025638425285e-05, "reference_active": true, "delta_db": -0.32239244787577387}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.92201473631864e-07, "output_power": 4.2836620016781226e-07, "reference_active": true, "delta_db": -3.6478467180705114}], "active_band_count": 3, "active_band_mae_db": 1.416016269143664}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011116745339746556, "log_rms_mae_db": 1.5505791350782796, "rms_envelope_nrmse": 0.1887326405449941, "envelope_correlation": 0.9678504356311881, "rms_gain_error_db": -0.2296136474519051, "reference_mean_power": 0.00017647199152424748, "output_mean_power": 0.00016738418501283865}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.117031117126669e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.030239739448223e-05, "output_power": 3.9117309757951445e-05, "reference_active": true, "delta_db": -0.12961901880409832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8894394185439072e-05, "output_power": 2.277053112794008e-05, "reference_active": true, "delta_db": -1.0344043250572477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.877164672472975e-06, "output_power": 5.4507109995949e-06, "reference_active": true, "delta_db": -1.5991676846015401}], "active_band_count": 3, "active_band_mae_db": 0.9210636761542954}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.815724618424621e-05, "log_rms_mae_db": 0.9187018286905407, "rms_envelope_nrmse": 0.11829424995608259, "envelope_correlation": 0.9860669089727837, "rms_gain_error_db": -0.2295902688919687, "reference_mean_power": 0.0001772156182174664, "output_mean_power": 0.00016809042188990802}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0001638026368737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.251019833717301e-05, "output_power": 4.0732777627536917e-05, "reference_active": true, "delta_db": -0.18549104962279328}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.608059024163373e-05, "output_power": 1.1183852280971642e-05, "reference_active": true, "delta_db": -1.5771056340750529}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.447224316204941e-07, "output_power": 4.029749041807663e-07, "reference_active": true, "delta_db": -2.040947800483849}], "active_band_count": 3, "active_band_mae_db": 1.2678481613938983}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8848887091307917, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "23bdbb8a466692152825b6b37da01619c8dcdf41dd3f792ba1c7e5a896f388b8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9831318748703497, "standard_deviation": 0.0015961229959915529, "minimum": 0.9814156302802707, "maximum": 0.9861920383079402, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4865329751320133, "standard_deviation": 0.038396584799744914, "minimum": 1.445835548999191, "maximum": 1.559901909764981, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5262388754919521, "standard_deviation": 0.2589691535733409, "minimum": 0.304359859425148, "maximum": 1.1256745615491066, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14297278565751015, "standard_deviation": 0.006878565006346473, "minimum": 0.12945165641152714, "maximum": 0.14986254043373715, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945600809798869, "standard_deviation": 0.0006165017529101182, "minimum": 0.9938695688220645, "maximum": 0.9954498747016636, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8561165079262998, "standard_deviation": 0.03708657872348028, "minimum": 0.8092497530056892, "maximum": 0.9211166034247948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7546885669179947, "standard_deviation": 0.32248409533829575, "minimum": 0.2659070554498714, "maximum": 1.416016269143664, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08007118604599232, "standard_deviation": 0.0041439229476548335, "minimum": 0.07276535925978594, "maximum": 0.08432046883200374, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9684654546873362, "standard_deviation": 0.0038516373174521594, "minimum": 0.9629452573109601, "maximum": 0.972896687460763, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5303466164595747, "standard_deviation": 0.021337269717695555, "minimum": 1.4966772492244262, "maximum": 1.553085797583856, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6613233889993398, "standard_deviation": 0.4040894983085117, "minimum": 0.25059727213531674, "maximum": 1.2370565953868045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1881099035029734, "standard_deviation": 0.010983212087180316, "minimum": 0.17494102883757917, "maximum": 0.20329961288040657, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9849326062513736, "standard_deviation": 0.0017502211902498499, "minimum": 0.9822056121046815, "maximum": 0.9875306349557056, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.912421557859507, "standard_deviation": 0.021754257227771447, "minimum": 0.8816189612298339, "maximum": 0.9386592851506371, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8050944930675976, "standard_deviation": 0.4836668458056406, "minimum": 0.1984693149097048, "maximum": 1.4515359597373314, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12258988233572893, "standard_deviation": 0.008406874891072677, "minimum": 0.10977332493886693, "maximum": 0.13771513564454055, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9919344023964296, "standard_deviation": 0.0015824920129713, "minimum": 0.9886853170974981, "maximum": 0.9937283940288572, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5504338005537681, "standard_deviation": 0.0165586274089912, "minimum": 1.5314259843798215, "maximum": 1.5735987200429067, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.045096645678442, "standard_deviation": 0.3900131151079384, "minimum": 0.47496581388936887, "maximum": 1.588981967910995, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09705429831608764, "standard_deviation": 0.009228616364498411, "minimum": 0.0857758626118949, "maximum": 0.11540744401223116, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972447296313022, "standard_deviation": 0.0008938514682482682, "minimum": 0.9956832007799197, "maximum": 0.9983155320209899, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8528747911815511, "standard_deviation": 0.02955755492378279, "minimum": 0.8238309998881223, "maximum": 0.8976487140086368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6190817330000948, "standard_deviation": 0.26532334234260513, "minimum": 0.30701664046253824, "maximum": 1.0015476302706658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05464248982880478, "standard_deviation": 0.008585434114512283, "minimum": 0.04298196212186707, "maximum": 0.06837015053576961, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14349268786221997, "standard_deviation": 0.030383114065083667, "minimum": 0.08668582414533721, "maximum": 0.17858842385885898, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.686359204634671, "standard_deviation": 0.007480668333115112, "minimum": 5.67586263190423, "maximum": 5.701242177291424, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4663842517785206, "standard_deviation": 0.010622595325619832, "minimum": 0.4523830749575902, "maximum": 0.4845462845904768, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06776453460986226, "standard_deviation": 0.031179395137353193, "minimum": -0.11038472042316144, "maximum": -0.025311617909254097, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9929610795942134, "standard_deviation": 0.0007968119149473262, "minimum": 0.9911676514911495, "maximum": 0.9937792101560129, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9371119531507461, "standard_deviation": 0.027521791059837496, "minimum": 0.8864906337957362, "maximum": 0.9729106931564656, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.31004851445176806, "standard_deviation": 0.21383815745050902, "minimum": 0.07698266500425553, "maximum": 0.7763739515979259, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08282387653183182, "standard_deviation": 0.004411694644915827, "minimum": 0.07809620207829665, "maximum": 0.09261907709066672, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978193968031253, "standard_deviation": 0.0004280141747425383, "minimum": 0.996896552033618, "maximum": 0.998204610471483, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4981818397273793, "standard_deviation": 0.02649244380307766, "minimum": 0.4583256293552771, "maximum": 0.5229287782687607, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44688145553547287, "standard_deviation": 0.2859860826055159, "minimum": 0.1360425133198571, "maximum": 0.9203057145282454, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0441378422095257, "standard_deviation": 0.003845675460460774, "minimum": 0.04026670110075603, "maximum": 0.052037293495889196, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0003", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5972633361816406, "predicted_mos": 2.5972633361816406, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "86125e5814e18d3fea9ac00e0ef25e99a0390c006ba09da15d58149f64a10d6b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.933993299957365, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9376874020027899, "全帯域_10ms_rms_envelope_nrmse": 0.08224094065095809, "全帯域_10ms_envelope_correlation": 0.9930558605898437, "全帯域_10ms_modulation_mae_db": 0.7721766162933119, "全帯域_40ms_log_rms_mae_db": 0.4823649066600757, "全帯域_40ms_rms_envelope_nrmse": 0.041882324354478494, "全帯域_40ms_envelope_correlation": 0.9980279873027654, "全帯域_40ms_modulation_mae_db": 0.8950433190134736, "300–1000 Hz_10ms_log_rms_mae_db": 1.508225341888425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08748558937438708, "300–1000 Hz_10ms_envelope_correlation": 0.9934736435597059, "300–1000 Hz_10ms_modulation_mae_db": 0.5099709475229661, "300–1000 Hz_40ms_log_rms_mae_db": 0.7959654248854312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04345943019646762, "300–1000 Hz_40ms_envelope_correlation": 0.9982772895742912, "300–1000 Hz_40ms_modulation_mae_db": 0.3785666654558372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4379525989858968, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14680069045883462, "1000–3000 Hz_10ms_envelope_correlation": 0.9821578692967378, "1000–3000 Hz_10ms_modulation_mae_db": 1.1579038155125871, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8069506891812824, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0835537831072033, "1000–3000 Hz_40ms_envelope_correlation": 0.993930914522391, "1000–3000 Hz_40ms_modulation_mae_db": 1.3795266223085247, "3000–6000 Hz_10ms_log_rms_mae_db": 1.535979701605536, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1883504587848392, "3000–6000 Hz_10ms_envelope_correlation": 0.9682041418469345, "3000–6000 Hz_10ms_modulation_mae_db": 1.112139235801173, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8885906607237541, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11774398290378776, "3000–6000 Hz_40ms_envelope_correlation": 0.9866564568667997, "3000–6000 Hz_40ms_modulation_mae_db": 1.4227830290547523, "mr_stft_db": 5.629142804437407, "band_mae_db": 0.1472236162332723, "rms_gain_error_db": -0.058840222746140494, "output_peak": 0.47923137180606123}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9534828288674407, "全帯域_10ms_rms_envelope_nrmse": 0.08316730481063134, "全帯域_10ms_envelope_correlation": 0.9929879102862288, "全帯域_10ms_modulation_mae_db": 0.3524618273514502, "全帯域_40ms_log_rms_mae_db": 0.5081529672002342, "全帯域_40ms_rms_envelope_nrmse": 0.042924317276413486, "全帯域_40ms_envelope_correlation": 0.9980604228525362, "全帯域_40ms_modulation_mae_db": 0.8009092565219079, "300–1000 Hz_10ms_log_rms_mae_db": 1.5109969057117272, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09058871087378162, "300–1000 Hz_10ms_envelope_correlation": 0.9930481810759119, "300–1000 Hz_10ms_modulation_mae_db": 1.1687791730717494, "300–1000 Hz_40ms_log_rms_mae_db": 0.7949964414949143, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0452124957664175, "300–1000 Hz_40ms_envelope_correlation": 0.998223748655764, "300–1000 Hz_40ms_modulation_mae_db": 0.48165141236193093, "1000–3000 Hz_10ms_log_rms_mae_db": 1.548503537186905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1473685933930528, "1000–3000 Hz_10ms_envelope_correlation": 0.9822768265243682, "1000–3000 Hz_10ms_modulation_mae_db": 0.6560718494770409, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8625489070469158, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08264116061670307, "1000–3000 Hz_40ms_envelope_correlation": 0.994504385757012, "1000–3000 Hz_40ms_modulation_mae_db": 0.7703166088427212, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5357278754983765, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19767178814168299, "3000–6000 Hz_10ms_envelope_correlation": 0.9653223610685387, "3000–6000 Hz_10ms_modulation_mae_db": 1.3269763333257554, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9254084559079274, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12823698334384015, "3000–6000 Hz_40ms_envelope_correlation": 0.9847366113055427, "3000–6000 Hz_40ms_modulation_mae_db": 1.5233427878361796, "mr_stft_db": 5.638795545190607, "band_mae_db": 0.21022235667737413, "rms_gain_error_db": -0.11236778816946516, "output_peak": 0.4602506794176572}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9571657972592846, "全帯域_10ms_rms_envelope_nrmse": 0.08508940622337295, "全帯域_10ms_envelope_correlation": 0.9925499125662383, "全帯域_10ms_modulation_mae_db": 0.3524630075312281, "全帯域_40ms_log_rms_mae_db": 0.5189030686295093, "全帯域_40ms_rms_envelope_nrmse": 0.04693694265586262, "全帯域_40ms_envelope_correlation": 0.9974698574360812, "全帯域_40ms_modulation_mae_db": 0.21941455292314863, "300–1000 Hz_10ms_log_rms_mae_db": 1.5608171481111632, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10411412825350262, "300–1000 Hz_10ms_envelope_correlation": 0.9907589423423026, "300–1000 Hz_10ms_modulation_mae_db": 0.9759126381888407, "300–1000 Hz_40ms_log_rms_mae_db": 0.8917599471247174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06039993921627854, "300–1000 Hz_40ms_envelope_correlation": 0.9966188741279967, "300–1000 Hz_40ms_modulation_mae_db": 0.6041308587677418, "1000–3000 Hz_10ms_log_rms_mae_db": 1.527292980779023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14623710685298877, "1000–3000 Hz_10ms_envelope_correlation": 0.9823670347233661, "1000–3000 Hz_10ms_modulation_mae_db": 0.418275453341501, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8579783242784322, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08021177497045603, "1000–3000 Hz_40ms_envelope_correlation": 0.9945602127426811, "1000–3000 Hz_40ms_modulation_mae_db": 0.7466579329888839, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4803899668862648, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19266815798377526, "3000–6000 Hz_10ms_envelope_correlation": 0.9690305441605311, "3000–6000 Hz_10ms_modulation_mae_db": 1.1931739354120887, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9110338698748282, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1433434572396182, "3000–6000 Hz_40ms_envelope_correlation": 0.981213502542009, "3000–6000 Hz_40ms_modulation_mae_db": 1.3361461119112368, "mr_stft_db": 5.636182161192262, "band_mae_db": 0.16176168765212456, "rms_gain_error_db": -0.015494140805430299, "output_peak": 0.4543540347489898}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9009504205789426, "全帯域_10ms_rms_envelope_nrmse": 0.08269311984343158, "全帯域_10ms_envelope_correlation": 0.9929866736382523, "全帯域_10ms_modulation_mae_db": 0.2618634411614036, "全帯域_40ms_log_rms_mae_db": 0.47764798164794764, "全帯域_40ms_rms_envelope_nrmse": 0.04619289832528386, "全帯域_40ms_envelope_correlation": 0.9976108978143845, "全帯域_40ms_modulation_mae_db": 0.2778253346706297, "300–1000 Hz_10ms_log_rms_mae_db": 1.5364385752738527, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09233540780079068, "300–1000 Hz_10ms_envelope_correlation": 0.9927731219460736, "300–1000 Hz_10ms_modulation_mae_db": 1.2004594839143083, "300–1000 Hz_40ms_log_rms_mae_db": 0.8262674708722454, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05372658545018496, "300–1000 Hz_40ms_envelope_correlation": 0.9974295992624165, "300–1000 Hz_40ms_modulation_mae_db": 0.6532860023427872, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4649660418751358, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1292937473476906, "1000–3000 Hz_10ms_envelope_correlation": 0.986174907336189, "1000–3000 Hz_10ms_modulation_mae_db": 0.4244358489690854, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8095644667935872, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0701121374394469, "1000–3000 Hz_40ms_envelope_correlation": 0.995712353278136, "1000–3000 Hz_40ms_modulation_mae_db": 0.4563955192347519, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5327943245682845, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2197112286902588, "3000–6000 Hz_10ms_envelope_correlation": 0.9586592490160181, "3000–6000 Hz_10ms_modulation_mae_db": 0.6714418112669375, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9317931540366549, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14752647629658114, "3000–6000 Hz_40ms_envelope_correlation": 0.9793097216979033, "3000–6000 Hz_40ms_modulation_mae_db": 0.7573236577947955, "mr_stft_db": 5.634420054037505, "band_mae_db": 0.18172924419280864, "rms_gain_error_db": -0.07248453604727774, "output_peak": 0.4812574439076773}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9328644900075526, "全帯域_10ms_rms_envelope_nrmse": 0.07810432780084361, "全帯域_10ms_envelope_correlation": 0.9937577895140368, "全帯域_10ms_modulation_mae_db": 0.1533412930132891, "全帯域_40ms_log_rms_mae_db": 0.5166248294428416, "全帯域_40ms_rms_envelope_nrmse": 0.04102471672688234, "全帯域_40ms_envelope_correlation": 0.9981395827374253, "全帯域_40ms_modulation_mae_db": 0.3304511179032505, "300–1000 Hz_10ms_log_rms_mae_db": 1.549624615558035, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08865688156864784, "300–1000 Hz_10ms_envelope_correlation": 0.993298379682909, "300–1000 Hz_10ms_modulation_mae_db": 0.44934673471617104, "300–1000 Hz_40ms_log_rms_mae_db": 0.8060137108603921, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047627532942872225, "300–1000 Hz_40ms_envelope_correlation": 0.9979279683231458, "300–1000 Hz_40ms_modulation_mae_db": 0.2449339511713168, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4533168413061797, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1417033128876498, "1000–3000 Hz_10ms_envelope_correlation": 0.983571403112601, "1000–3000 Hz_10ms_modulation_mae_db": 0.31893446589865015, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8707131127433135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08203332966774199, "1000–3000 Hz_40ms_envelope_correlation": 0.9944320445012075, "1000–3000 Hz_40ms_modulation_mae_db": 0.5619047129797293, "3000–6000 Hz_10ms_log_rms_mae_db": 1.512308495794295, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17082706926945562, "3000–6000 Hz_10ms_envelope_correlation": 0.9741477573097835, "3000–6000 Hz_10ms_modulation_mae_db": 0.47472772164507976, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8850996861888928, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10943198845925917, "3000–6000 Hz_40ms_envelope_correlation": 0.9878721412817989, "3000–6000 Hz_40ms_modulation_mae_db": 0.628189180998433, "mr_stft_db": 5.63289889535418, "band_mae_db": 0.1462542862341598, "rms_gain_error_db": -0.06994975295074166, "output_peak": 0.47515904031981093}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.932251300695899, "全帯域_10ms_rms_envelope_nrmse": 0.09242399656282738, "全帯域_10ms_envelope_correlation": 0.9912092258940148, "全帯域_10ms_modulation_mae_db": 0.03891569560214584, "全帯域_40ms_log_rms_mae_db": 0.5173349442947565, "全帯域_40ms_rms_envelope_nrmse": 0.05335353422193355, "全帯域_40ms_envelope_correlation": 0.9967404049349351, "全帯域_40ms_modulation_mae_db": 0.20720244723552864, "300–1000 Hz_10ms_log_rms_mae_db": 1.5639525223534902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11479851336771675, "300–1000 Hz_10ms_envelope_correlation": 0.9888349097714402, "300–1000 Hz_10ms_modulation_mae_db": 1.4157599032765837, "300–1000 Hz_40ms_log_rms_mae_db": 0.8286400168853195, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06798035389095217, "300–1000 Hz_40ms_envelope_correlation": 0.9957505364826496, "300–1000 Hz_40ms_modulation_mae_db": 0.9286513855904887, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4902269760108788, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1471334742367445, "1000–3000 Hz_10ms_envelope_correlation": 0.9822706616691004, "1000–3000 Hz_10ms_modulation_mae_db": 0.3619756819418831, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7857636169384898, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08368138691104314, "1000–3000 Hz_40ms_envelope_correlation": 0.9942352160956048, "1000–3000 Hz_40ms_modulation_mae_db": 0.5592563849300517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4706935294333519, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19053052541262036, "3000–6000 Hz_10ms_envelope_correlation": 0.9670951687736871, "3000–6000 Hz_10ms_modulation_mae_db": 0.4723441048860085, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8669012107164995, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11489565623100342, "3000–6000 Hz_40ms_envelope_correlation": 0.9864484082045949, "3000–6000 Hz_40ms_modulation_mae_db": 0.6699099323249312, "mr_stft_db": 5.622177046992128, "band_mae_db": 0.17035551190959683, "rms_gain_error_db": -0.03170446841588047, "output_peak": 0.46839306642372214}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9303121167643544, "全帯域_10ms_rms_envelope_nrmse": 0.07853505427289381, "全帯域_10ms_envelope_correlation": 0.9937189090248667, "全帯域_10ms_modulation_mae_db": 0.25546098046193394, "全帯域_40ms_log_rms_mae_db": 0.4710230932867929, "全帯域_40ms_rms_envelope_nrmse": 0.041085047515126365, "全帯域_40ms_envelope_correlation": 0.9981707347331217, "全帯域_40ms_modulation_mae_db": 0.5061243318377752, "300–1000 Hz_10ms_log_rms_mae_db": 1.5454871846544715, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09372091351010577, "300–1000 Hz_10ms_envelope_correlation": 0.992516938054476, "300–1000 Hz_10ms_modulation_mae_db": 1.4192156486044933, "300–1000 Hz_40ms_log_rms_mae_db": 0.786785949467276, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05107797482433983, "300–1000 Hz_40ms_envelope_correlation": 0.9976317601012537, "300–1000 Hz_40ms_modulation_mae_db": 0.9082578089917982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4695755749860062, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14309445803476825, "1000–3000 Hz_10ms_envelope_correlation": 0.9832304467169858, "1000–3000 Hz_10ms_modulation_mae_db": 0.3578938234767827, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8051891099631227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07494838229404312, "1000–3000 Hz_40ms_envelope_correlation": 0.9954715282779415, "1000–3000 Hz_40ms_modulation_mae_db": 0.7566900330437925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5001623103726247, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17276659764145696, "3000–6000 Hz_10ms_envelope_correlation": 0.973075509572239, "3000–6000 Hz_10ms_modulation_mae_db": 0.4970980465820582, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8787609888195236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1071617451484703, "3000–6000 Hz_40ms_envelope_correlation": 0.9883661979950743, "3000–6000 Hz_40ms_modulation_mae_db": 0.6515896515347763, "mr_stft_db": 5.627577575063992, "band_mae_db": 0.11660318217960776, "rms_gain_error_db": -0.09024376170242579, "output_peak": 0.4646404707024186}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9714881787049869, "全帯域_10ms_rms_envelope_nrmse": 0.08415465131282666, "全帯域_10ms_envelope_correlation": 0.9927738627555339, "全帯域_10ms_modulation_mae_db": 0.41021710056158817, "全帯域_40ms_log_rms_mae_db": 0.5294006944223267, "全帯域_40ms_rms_envelope_nrmse": 0.04996912539813144, "全帯域_40ms_envelope_correlation": 0.9972524700604068, "全帯域_40ms_modulation_mae_db": 0.3199869844013026, "300–1000 Hz_10ms_log_rms_mae_db": 1.535441786795655, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09990242598740354, "300–1000 Hz_10ms_envelope_correlation": 0.991563640607128, "300–1000 Hz_10ms_modulation_mae_db": 0.8414722400468363, "300–1000 Hz_40ms_log_rms_mae_db": 0.8591397338590662, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061008178547234775, "300–1000 Hz_40ms_envelope_correlation": 0.9967143330958526, "300–1000 Hz_40ms_modulation_mae_db": 0.6212695896080799, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4697956813635005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14190671418576797, "1000–3000 Hz_10ms_envelope_correlation": 0.9832981090905417, "1000–3000 Hz_10ms_modulation_mae_db": 0.5402430416489471, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8272171668482179, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09093673362985522, "1000–3000 Hz_40ms_envelope_correlation": 0.9925692039908413, "1000–3000 Hz_40ms_modulation_mae_db": 0.525314009068015, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5230025735397894, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17692388181158808, "3000–6000 Hz_10ms_envelope_correlation": 0.9720300459817188, "3000–6000 Hz_10ms_modulation_mae_db": 0.31256505175632293, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9206278209410168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11716642295620373, "3000–6000 Hz_40ms_envelope_correlation": 0.9860313100056087, "3000–6000 Hz_40ms_modulation_mae_db": 0.30530361707857173, "mr_stft_db": 5.634279081943366, "band_mae_db": 0.1506878059510001, "rms_gain_error_db": -0.09622764971115498, "output_peak": 0.45017075648280674}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.694915432926012, "spectral_convergence": 0.3092408564178794, "spectral_convergence_status": "ok", "amplitude_floor": 1.2198198453862723e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.417443240701169, "spectral_convergence": 0.3013140539297196, "spectral_convergence_status": "ok", "amplitude_floor": 7.927151445605076e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7750697396850414, "spectral_convergence": 0.3181831625092783, "spectral_convergence_status": "ok", "amplitude_floor": 6.454878545778745e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00013983470011737857, "output_power": 0.0001478550904270046, "delta_db": 0.2422132583797213, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042159782923140474, "output_power": 0.000418786968223317, "delta_db": -0.029052071888345084, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017365691446010552, "output_power": 0.0017223674771500244, "delta_db": -0.035662639360108966, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006597865121791135, "output_power": 0.0006420145917340129, "delta_db": -0.11858534302339724, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001856647778551521, "output_power": 0.0001737659949056129, "delta_db": -0.28764731022657997, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000145726779943178, "output_power": 0.0001401268183204044, "delta_db": -0.17018107452148112, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1472236162332723, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.058840222746140494, "pitch": null, "output_rms_dbfs": -24.887964152368323, "output_peak": 0.47923137180606123, "reference_rms_dbfs": -24.82912392962218, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.629142804437407}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017345649672524565, "log_rms_mae_db": 0.9376874020027899, "rms_envelope_nrmse": 0.08224094065095809, "envelope_correlation": 0.9930558605898437, "rms_gain_error_db": -0.059114614346660016, "reference_mean_power": 0.003291322552458033, "output_mean_power": 0.0032468257691315107}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4377088226614287e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006890430591645026, "output_power": 0.000683463408302175, "reference_active": true, "delta_db": -0.03531094369651278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.716540294963627e-05, "output_power": 8.846131369594097e-05, "reference_active": true, "delta_db": -0.4075827147290021}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.855953967229281e-05, "output_power": 1.8551865373965397e-05, "reference_active": true, "delta_db": -1.873636190454421}], "active_band_count": 3, "active_band_mae_db": 0.7721766162933119}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015649665907765232, "log_rms_mae_db": 0.4823649066600757, "rms_envelope_nrmse": 0.041882324354478494, "envelope_correlation": 0.9980279873027654, "rms_gain_error_db": -0.0594350963837145, "reference_mean_power": 0.003302205038848876, "output_mean_power": 0.0032573207519085976}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.335462563779621e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377316706455328, "output_power": 0.0006344205064345454, "reference_active": true, "delta_db": -0.022607729150079352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.741968368909027e-05, "output_power": 4.251360287556155e-05, "reference_active": true, "delta_db": -0.47430741281166666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7622617238454117e-06, "output_power": 1.0647531338326943e-06, "reference_active": true, "delta_db": -2.1882148150786747}], "active_band_count": 3, "active_band_mae_db": 0.8950433190134736}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014746435244218596, "log_rms_mae_db": 1.508225341888425, "rms_envelope_nrmse": 0.08748558937438708, "envelope_correlation": 0.9934736435597059, "rms_gain_error_db": -0.044109477745015176, "reference_mean_power": 0.001906578877331925, "output_mean_power": 0.0018873125571030318}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.936060821906154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045087668804334914, "output_power": 0.0004406157909355379, "reference_active": true, "delta_db": -0.09997723806451046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.461610811295139e-05, "output_power": 5.194075075373106e-05, "reference_active": true, "delta_db": -0.21812526884488947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.1974528143292685e-06, "output_power": 8.192081464778807e-06, "reference_active": true, "delta_db": 1.2118103356594985}], "active_band_count": 3, "active_band_mae_db": 0.5099709475229661}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013246473438636124, "log_rms_mae_db": 0.7959654248854312, "rms_envelope_nrmse": 0.04345943019646762, "envelope_correlation": 0.9982772895742912, "rms_gain_error_db": -0.044260776770833794, "reference_mean_power": 0.0019129188838057705, "output_mean_power": 0.0018935225292034605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.520895123968316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004268128119473447, "output_power": 0.0004179572476112546, "reference_active": true, "delta_db": -0.09105586937753958}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.526792747799959e-05, "output_power": 2.4068925357628767e-05, "reference_active": true, "delta_db": -0.21112921653110922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.02211562944459e-07, "output_power": 6.084686383364877e-07, "reference_active": true, "delta_db": 0.8335149104588628}], "active_band_count": 3, "active_band_mae_db": 0.3785666654558372}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.50803654710503e-05, "log_rms_mae_db": 1.4379525989858968, "rms_envelope_nrmse": 0.14680069045883462, "envelope_correlation": 0.9821578692967378, "rms_gain_error_db": -0.18405448672206076, "reference_mean_power": 0.0007310287661738595, "output_mean_power": 0.0007006949995897048}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3756975679714823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020778711952139343, "output_power": 0.00019857341717738916, "reference_active": true, "delta_db": -0.19697513165290212}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.101230284766401e-05, "output_power": 3.570009728421456e-05, "reference_active": true, "delta_db": -0.6024475609603203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.417981343795666e-05, "output_power": 7.660227035404128e-06, "reference_active": true, "delta_db": -2.6742887539245386}], "active_band_count": 3, "active_band_mae_db": 1.1579038155125871}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.582207074769948e-05, "log_rms_mae_db": 0.8069506891812824, "rms_envelope_nrmse": 0.0835537831072033, "envelope_correlation": 0.993930914522391, "rms_gain_error_db": -0.18451855876686202, "reference_mean_power": 0.000733846607513954, "output_mean_power": 0.0007033207572159118}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9980296656839734e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019557456921819033, "output_power": 0.00018460637132781875, "reference_active": true, "delta_db": -0.2506569654839679}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6985486820992048e-05, "output_power": 1.5009227367678407e-05, "reference_active": true, "delta_db": -0.5371966212303586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.92201473631864e-07, "output_power": 4.586984152527179e-07, "reference_active": true, "delta_db": -3.350726280211248}], "active_band_count": 3, "active_band_mae_db": 1.3795266223085247}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011116745339746556, "log_rms_mae_db": 1.535979701605536, "rms_envelope_nrmse": 0.1883504587848392, "envelope_correlation": 0.9682041418469345, "rms_gain_error_db": -0.324236225155266, "reference_mean_power": 0.00017647199152424748, "output_mean_power": 0.00016377671818074155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.117031117126669e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.030239739448223e-05, "output_power": 3.8132899429091474e-05, "reference_active": true, "delta_db": -0.24031052942770723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8894394185439072e-05, "output_power": 2.2371396758164978e-05, "reference_active": true, "delta_db": -1.1112049313964316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.877164672472975e-06, "output_power": 4.9874631493199355e-06, "reference_active": true, "delta_db": -1.9849022465793797}], "active_band_count": 3, "active_band_mae_db": 1.112139235801173}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.815724618424621e-05, "log_rms_mae_db": 0.8885906607237541, "rms_envelope_nrmse": 0.11774398290378776, "envelope_correlation": 0.9866564568667997, "rms_gain_error_db": -0.32416614933718957, "reference_mean_power": 0.0001772156182174664, "output_mean_power": 0.00016446950266762282}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0001638026368737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.251019833717301e-05, "output_power": 3.946981816308142e-05, "reference_active": true, "delta_db": -0.32228005621126865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.608059024163373e-05, "output_power": 1.1042695989261307e-05, "reference_active": true, "delta_db": -1.632268695856744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.447224316204941e-07, "output_power": 3.7843626473567357e-07, "reference_active": true, "delta_db": -2.3138003350962437}], "active_band_count": 3, "active_band_mae_db": 1.4227830290547523}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.889012450551646, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5d5073c83e1b4d3abdeb6eec6d178ccd592fed5f32f24e9c5fa2a6c441b63635", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9831684073087361, "standard_deviation": 0.0013355773548866174, "minimum": 0.9821578692967378, "maximum": 0.986174907336189, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4827037790616906, "standard_deviation": 0.037596086136007116, "minimum": 1.4379525989858968, "maximum": 1.548503537186905, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5294667475333097, "standard_deviation": 0.2767896285353739, "minimum": 0.31893446589865015, "maximum": 1.1579038155125871, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14294226217468717, "standard_deviation": 0.00599779259892016, "minimum": 0.1292937473476906, "maximum": 0.1473685933930528, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9944269823957269, "standard_deviation": 0.000963714531418798, "minimum": 0.9925692039908413, "maximum": 0.995712353278136, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8282406742241702, "standard_deviation": 0.03163198532437252, "minimum": 0.7857636169384898, "maximum": 0.8707131127433135, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7195077279245587, "standard_deviation": 0.29222901460259104, "minimum": 0.4563955192347519, "maximum": 1.3795266223085247, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08101483607956159, "standard_deviation": 0.006233534463434612, "minimum": 0.0701121374394469, "maximum": 0.09093673362985522, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9684455972161814, "standard_deviation": 0.004993571965762268, "minimum": 0.9586592490160181, "maximum": 0.9741477573097835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5113823472123151, "standard_deviation": 0.025450488441205475, "minimum": 1.4706935294333519, "maximum": 1.535979701605536, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.757558280084428, "standard_deviation": 0.39176165784413414, "minimum": 0.31256505175632293, "maximum": 1.3269763333257554, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18868121346695965, "standard_deviation": 0.015916737224898277, "minimum": 0.17082706926945562, "maximum": 0.2197112286902588, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9850792937374164, "standard_deviation": 0.0032120692991042137, "minimum": 0.9793097216979033, "maximum": 0.9883661979950743, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9010269809011371, "standard_deviation": 0.02418390214463914, "minimum": 0.8669012107164995, "maximum": 0.9317931540366549, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9118234960667095, "standard_deviation": 0.44936685384707614, "minimum": 0.30530361707857173, "maximum": 1.5233427878361796, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12318833907234548, "standard_deviation": 0.015141355347715504, "minimum": 0.1071617451484703, "maximum": 0.14752647629658114, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9920334696299934, "standard_deviation": 0.0015838700010221943, "minimum": 0.9888349097714402, "maximum": 0.9934736435597059, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5388730100433525, "standard_deviation": 0.02071551517185651, "minimum": 1.508225341888425, "maximum": 1.5639525223534902, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9976145961677436, "standard_deviation": 0.3753630038212426, "minimum": 0.44934673471617104, "maximum": 1.4192156486044933, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09645032134204198, "standard_deviation": 0.009313215602954122, "minimum": 0.08748558937438708, "maximum": 0.11479851336771675, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973217637029212, "standard_deviation": 0.0008891085051193194, "minimum": 0.9957505364826496, "maximum": 0.9982772895742912, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8236960869311702, "standard_deviation": 0.03622879590194564, "minimum": 0.786785949467276, "maximum": 0.8917599471247174, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6025934592862476, "standard_deviation": 0.2376674351641184, "minimum": 0.2449339511713168, "maximum": 0.9286513855904887, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05381156135434345, "standard_deviation": 0.00864196716964553, "minimum": 0.04345943019646762, "maximum": 0.06798035389095217, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16060471137874302, "standard_deviation": 0.027836361910613422, "minimum": 0.11660318217960776, "maximum": 0.21022235667737413, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.631934145526431, "standard_deviation": 0.005341293525533066, "minimum": 5.622177046992128, "maximum": 5.638795545190607, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46668210797614296, "standard_deviation": 0.011432084390929588, "minimum": 0.45017075648280674, "maximum": 0.4812574439076773, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06841404006856458, "standard_deviation": 0.032603153481824576, "minimum": -0.11236778816946516, "maximum": -0.015494140805430299, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9928800180336269, "standard_deviation": 0.0007964154124402892, "minimum": 0.9912092258940148, "maximum": 0.9937577895140368, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9395253168601563, "standard_deviation": 0.02138485700246155, "minimum": 0.9009504205789426, "maximum": 0.9714881787049869, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32461249524704383, "standard_deviation": 0.2167516317486053, "minimum": 0.03891569560214584, "maximum": 0.7721766162933119, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08330110018472318, "standard_deviation": 0.004446297700974885, "minimum": 0.07810432780084361, "maximum": 0.09242399656282738, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997684044733957, "standard_deviation": 0.0005116190611657467, "minimum": 0.9967404049349351, "maximum": 0.9981707347331217, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5026815606980606, "standard_deviation": 0.02222555536817317, "minimum": 0.4710230932867929, "maximum": 0.5294006944223267, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44461966806337705, "standard_deviation": 0.26647941593261554, "minimum": 0.20720244723552864, "maximum": 0.8950433190134736, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045421113309264016, "standard_deviation": 0.004525923637258105, "minimum": 0.04102471672688234, "maximum": 0.05335353422193355, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0003", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.8715994358062744, "predicted_mos": 1.8715994358062744, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "36cc9e070e70267cd542844a3e3270e85fe8220aaea093b5579c70c1388c01bf", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.068758599925786, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.957900314169872, "全帯域_10ms_rms_envelope_nrmse": 0.08546056414630993, "全帯域_10ms_envelope_correlation": 0.9925280012920177, "全帯域_10ms_modulation_mae_db": 0.17195092911754908, "全帯域_40ms_log_rms_mae_db": 0.5230335284328074, "全帯域_40ms_rms_envelope_nrmse": 0.041995065985404735, "全帯域_40ms_envelope_correlation": 0.9980751380931185, "全帯域_40ms_modulation_mae_db": 0.3021121461289667, "300–1000 Hz_10ms_log_rms_mae_db": 1.7253229937201127, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10319191405651182, "300–1000 Hz_10ms_envelope_correlation": 0.9909092421879793, "300–1000 Hz_10ms_modulation_mae_db": 1.0436547973950638, "300–1000 Hz_40ms_log_rms_mae_db": 1.0093239401267822, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056250135327573725, "300–1000 Hz_40ms_envelope_correlation": 0.9971135229119421, "300–1000 Hz_40ms_modulation_mae_db": 0.7023999796868564, "1000–3000 Hz_10ms_log_rms_mae_db": 1.705142942995269, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15190163235565707, "1000–3000 Hz_10ms_envelope_correlation": 0.9810886643892522, "1000–3000 Hz_10ms_modulation_mae_db": 0.42960825320739043, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0221800208013945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09417336600052628, "1000–3000 Hz_40ms_envelope_correlation": 0.9925576672226532, "1000–3000 Hz_40ms_modulation_mae_db": 0.4915132480685151, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7148047694315094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19448344248865576, "3000–6000 Hz_10ms_envelope_correlation": 0.9661022307288126, "3000–6000 Hz_10ms_modulation_mae_db": 0.6802598225266229, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0429817755866582, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12767726983073602, "3000–6000 Hz_40ms_envelope_correlation": 0.9841089505261285, "3000–6000 Hz_40ms_modulation_mae_db": 0.9209343380333227, "mr_stft_db": 5.89372435877339, "band_mae_db": 0.15863933561228946, "rms_gain_error_db": -0.08538615720729324, "output_peak": 0.47148644916452714}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9795915622238801, "全帯域_10ms_rms_envelope_nrmse": 0.09621514519554995, "全帯域_10ms_envelope_correlation": 0.9904691870473745, "全帯域_10ms_modulation_mae_db": 0.2326173917005223, "全帯域_40ms_log_rms_mae_db": 0.5412527032186049, "全帯域_40ms_rms_envelope_nrmse": 0.05126559109402668, "全帯域_40ms_envelope_correlation": 0.9970257586171702, "全帯域_40ms_modulation_mae_db": 0.7636215112372744, "300–1000 Hz_10ms_log_rms_mae_db": 1.7270818407520327, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10582368863981528, "300–1000 Hz_10ms_envelope_correlation": 0.9904516182720025, "300–1000 Hz_10ms_modulation_mae_db": 1.6924055056119702, "300–1000 Hz_40ms_log_rms_mae_db": 1.0235061077194096, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06329754624984567, "300–1000 Hz_40ms_envelope_correlation": 0.9963555281394212, "300–1000 Hz_40ms_modulation_mae_db": 0.9245644957998307, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6897501099386503, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15815648888586958, "1000–3000 Hz_10ms_envelope_correlation": 0.9791740004419843, "1000–3000 Hz_10ms_modulation_mae_db": 0.13813415455886077, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0130839397361355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09401483495239324, "1000–3000 Hz_40ms_envelope_correlation": 0.9919900684544696, "1000–3000 Hz_40ms_modulation_mae_db": 0.3950528599288848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8040681819717694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22520865981059804, "3000–6000 Hz_10ms_envelope_correlation": 0.9542232746210879, "3000–6000 Hz_10ms_modulation_mae_db": 0.6471502011903411, "3000–6000 Hz_40ms_log_rms_mae_db": 1.076223778210076, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14346537555011346, "3000–6000 Hz_40ms_envelope_correlation": 0.9786154910583954, "3000–6000 Hz_40ms_modulation_mae_db": 0.8550929750999289, "mr_stft_db": 5.897444271886321, "band_mae_db": 0.13698704910356965, "rms_gain_error_db": -0.06689750407843677, "output_peak": 0.45592381674521704}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9363917040603518, "全帯域_10ms_rms_envelope_nrmse": 0.08287874349110781, "全帯域_10ms_envelope_correlation": 0.992949829593858, "全帯域_10ms_modulation_mae_db": 0.18718128381014007, "全帯域_40ms_log_rms_mae_db": 0.49156956824387654, "全帯域_40ms_rms_envelope_nrmse": 0.04082412716561446, "全帯域_40ms_envelope_correlation": 0.9981431759974199, "全帯域_40ms_modulation_mae_db": 0.40495582252392887, "300–1000 Hz_10ms_log_rms_mae_db": 1.8528108624126012, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10626113666145462, "300–1000 Hz_10ms_envelope_correlation": 0.9903492083896778, "300–1000 Hz_10ms_modulation_mae_db": 1.0039676022339032, "300–1000 Hz_40ms_log_rms_mae_db": 1.0815275458267979, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06268382668202446, "300–1000 Hz_40ms_envelope_correlation": 0.9963803160446255, "300–1000 Hz_40ms_modulation_mae_db": 0.7461556201347852, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6007904072914059, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1481701391499938, "1000–3000 Hz_10ms_envelope_correlation": 0.9818425364780733, "1000–3000 Hz_10ms_modulation_mae_db": 0.28493591929991796, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9691191898470702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08197620242482133, "1000–3000 Hz_40ms_envelope_correlation": 0.994247738008983, "1000–3000 Hz_40ms_modulation_mae_db": 0.4703000327407468, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7864207609833531, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.201300849393387, "3000–6000 Hz_10ms_envelope_correlation": 0.9632712323777636, "3000–6000 Hz_10ms_modulation_mae_db": 1.0056856700334622, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0640369177522977, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.131693870420191, "3000–6000 Hz_40ms_envelope_correlation": 0.9822775208635834, "3000–6000 Hz_40ms_modulation_mae_db": 1.3689631349716314, "mr_stft_db": 5.910892892429355, "band_mae_db": 0.12676262784481174, "rms_gain_error_db": -0.06737340912595614, "output_peak": 0.4609419285717677}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.926963115844601, "全帯域_10ms_rms_envelope_nrmse": 0.08274502560704781, "全帯域_10ms_envelope_correlation": 0.9930141268746263, "全帯域_10ms_modulation_mae_db": 0.3245013993281674, "全帯域_40ms_log_rms_mae_db": 0.48490059839706573, "全帯域_40ms_rms_envelope_nrmse": 0.03960939698484546, "全帯域_40ms_envelope_correlation": 0.9983128438407538, "全帯域_40ms_modulation_mae_db": 0.8663923071699658, "300–1000 Hz_10ms_log_rms_mae_db": 1.8154964786073293, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09851900525245702, "300–1000 Hz_10ms_envelope_correlation": 0.9917423238891744, "300–1000 Hz_10ms_modulation_mae_db": 0.8298633400824396, "300–1000 Hz_40ms_log_rms_mae_db": 1.1062929532008905, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052784362300817185, "300–1000 Hz_40ms_envelope_correlation": 0.9975050207819456, "300–1000 Hz_40ms_modulation_mae_db": 0.38609648338195557, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7050475716397089, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14612779586110217, "1000–3000 Hz_10ms_envelope_correlation": 0.9824944941687238, "1000–3000 Hz_10ms_modulation_mae_db": 0.4495294789993574, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0377585881521278, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08930146391480745, "1000–3000 Hz_40ms_envelope_correlation": 0.9932875126198495, "1000–3000 Hz_40ms_modulation_mae_db": 1.1307752700731746, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8025600620494933, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2203409073025549, "3000–6000 Hz_10ms_envelope_correlation": 0.9558422873219476, "3000–6000 Hz_10ms_modulation_mae_db": 1.2586643482083224, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1082613492525821, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15735448259890006, "3000–6000 Hz_40ms_envelope_correlation": 0.9746617068337597, "3000–6000 Hz_40ms_modulation_mae_db": 1.6355890768277035, "mr_stft_db": 5.901673885014772, "band_mae_db": 0.15393813777520185, "rms_gain_error_db": -0.08628410373169595, "output_peak": 0.4646545579736079}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9314154004318628, "全帯域_10ms_rms_envelope_nrmse": 0.09094264046767359, "全帯域_10ms_envelope_correlation": 0.9915234434877116, "全帯域_10ms_modulation_mae_db": 0.17814353684334275, "全帯域_40ms_log_rms_mae_db": 0.47481965834098117, "全帯域_40ms_rms_envelope_nrmse": 0.03911022414801816, "全帯域_40ms_envelope_correlation": 0.9983561768263458, "全帯域_40ms_modulation_mae_db": 0.35760097749240716, "300–1000 Hz_10ms_log_rms_mae_db": 1.8873607561312087, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12222466084341779, "300–1000 Hz_10ms_envelope_correlation": 0.9872213942014959, "300–1000 Hz_10ms_modulation_mae_db": 1.7425421359797484, "300–1000 Hz_40ms_log_rms_mae_db": 1.116963431708604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05889883316741197, "300–1000 Hz_40ms_envelope_correlation": 0.9968716173215212, "300–1000 Hz_40ms_modulation_mae_db": 0.9080413822713425, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6905410125050087, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1603445702889144, "1000–3000 Hz_10ms_envelope_correlation": 0.9787784261186316, "1000–3000 Hz_10ms_modulation_mae_db": 0.7807787694389349, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9857727508696873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09638680867022856, "1000–3000 Hz_40ms_envelope_correlation": 0.9920696998145624, "1000–3000 Hz_40ms_modulation_mae_db": 1.1697941052105298, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8196161294993654, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21247496905917487, "3000–6000 Hz_10ms_envelope_correlation": 0.9594916016851136, "3000–6000 Hz_10ms_modulation_mae_db": 0.34056182360503967, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1257764178422833, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14291919789546992, "3000–6000 Hz_40ms_envelope_correlation": 0.9788613955642484, "3000–6000 Hz_40ms_modulation_mae_db": 0.4619392536196792, "mr_stft_db": 5.90471692228694, "band_mae_db": 0.18578935057857515, "rms_gain_error_db": -0.08513243135496168, "output_peak": 0.5106213931910609}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9191909775075247, "全帯域_10ms_rms_envelope_nrmse": 0.08375440762897364, "全帯域_10ms_envelope_correlation": 0.9928261499483282, "全帯域_10ms_modulation_mae_db": 0.2064945002304576, "全帯域_40ms_log_rms_mae_db": 0.5098672927702416, "全帯域_40ms_rms_envelope_nrmse": 0.04363049197485214, "全帯域_40ms_envelope_correlation": 0.9978412760290517, "全帯域_40ms_modulation_mae_db": 0.21126169002107595, "300–1000 Hz_10ms_log_rms_mae_db": 1.7858512532089605, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09418406857684704, "300–1000 Hz_10ms_envelope_correlation": 0.9924478495160503, "300–1000 Hz_10ms_modulation_mae_db": 1.1821486056547885, "300–1000 Hz_40ms_log_rms_mae_db": 1.0843673236188818, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04571071341116569, "300–1000 Hz_40ms_envelope_correlation": 0.9980877795349812, "300–1000 Hz_40ms_modulation_mae_db": 0.7926116138273955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6599168113057488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14749875454050995, "1000–3000 Hz_10ms_envelope_correlation": 0.981970627425599, "1000–3000 Hz_10ms_modulation_mae_db": 0.3618403520174665, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9486508167339085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08527766429079729, "1000–3000 Hz_40ms_envelope_correlation": 0.9933827995754783, "1000–3000 Hz_40ms_modulation_mae_db": 0.18537641265554994, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8188389859786243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2146252265524051, "3000–6000 Hz_10ms_envelope_correlation": 0.9586508411265613, "3000–6000 Hz_10ms_modulation_mae_db": 0.5738495546843981, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0371434890558449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14021461101122332, "3000–6000 Hz_40ms_envelope_correlation": 0.9796926643472064, "3000–6000 Hz_40ms_modulation_mae_db": 1.01931844582795, "mr_stft_db": 5.9005040532655935, "band_mae_db": 0.1590446323572259, "rms_gain_error_db": 0.024491912159363104, "output_peak": 0.4976244254767966}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9135404482280007, "全帯域_10ms_rms_envelope_nrmse": 0.07524184517804164, "全帯域_10ms_envelope_correlation": 0.9941922660547288, "全帯域_10ms_modulation_mae_db": 0.24567605715461358, "全帯域_40ms_log_rms_mae_db": 0.488156026051868, "全帯域_40ms_rms_envelope_nrmse": 0.04085339500603585, "全帯域_40ms_envelope_correlation": 0.9981211533406575, "全帯域_40ms_modulation_mae_db": 0.32787356284923147, "300–1000 Hz_10ms_log_rms_mae_db": 1.799544280322323, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.089860604206398, "300–1000 Hz_10ms_envelope_correlation": 0.9931712418406143, "300–1000 Hz_10ms_modulation_mae_db": 0.6259071992896129, "300–1000 Hz_40ms_log_rms_mae_db": 1.0855870616937262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05696913466060782, "300–1000 Hz_40ms_envelope_correlation": 0.9971066669831913, "300–1000 Hz_40ms_modulation_mae_db": 0.32722436111995085, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5880101671628613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15243678104362263, "1000–3000 Hz_10ms_envelope_correlation": 0.9806615312868024, "1000–3000 Hz_10ms_modulation_mae_db": 0.34228118279480313, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9415241661144219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08968722795230978, "1000–3000 Hz_40ms_envelope_correlation": 0.9927319951102227, "1000–3000 Hz_40ms_modulation_mae_db": 0.6365836675792179, "3000–6000 Hz_10ms_log_rms_mae_db": 1.786323891611409, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18845624724192744, "3000–6000 Hz_10ms_envelope_correlation": 0.9688451312514513, "3000–6000 Hz_10ms_modulation_mae_db": 0.45046488926117606, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0330892332229527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12219652519230102, "3000–6000 Hz_40ms_envelope_correlation": 0.9850303204753019, "3000–6000 Hz_40ms_modulation_mae_db": 0.553047760253664, "mr_stft_db": 5.911251718944379, "band_mae_db": 0.10939791418301176, "rms_gain_error_db": -0.056763516869187736, "output_peak": 0.48747042423871034}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9279757119166249, "全帯域_10ms_rms_envelope_nrmse": 0.08494812719630726, "全帯域_10ms_envelope_correlation": 0.9925980830461649, "全帯域_10ms_modulation_mae_db": 0.28326231526878876, "全帯域_40ms_log_rms_mae_db": 0.5150099569779996, "全帯域_40ms_rms_envelope_nrmse": 0.044998829705565395, "全帯域_40ms_envelope_correlation": 0.9976815004746439, "全帯域_40ms_modulation_mae_db": 0.33127857146485, "300–1000 Hz_10ms_log_rms_mae_db": 1.7863945529522418, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09867672167294769, "300–1000 Hz_10ms_envelope_correlation": 0.9917394434620689, "300–1000 Hz_10ms_modulation_mae_db": 1.278121420517708, "300–1000 Hz_40ms_log_rms_mae_db": 1.096460063619134, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05476332195149142, "300–1000 Hz_40ms_envelope_correlation": 0.9972560994070617, "300–1000 Hz_40ms_modulation_mae_db": 1.0503323884083595, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6637355849441031, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14496910621869735, "1000–3000 Hz_10ms_envelope_correlation": 0.9825179835094894, "1000–3000 Hz_10ms_modulation_mae_db": 0.7725621035609338, "1000–3000 Hz_40ms_log_rms_mae_db": 0.988186707188034, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08309796913781133, "1000–3000 Hz_40ms_envelope_correlation": 0.9937475550156256, "1000–3000 Hz_40ms_modulation_mae_db": 1.0343240621499976, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7792466192367096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19259070653585972, "3000–6000 Hz_10ms_envelope_correlation": 0.9665143573081693, "3000–6000 Hz_10ms_modulation_mae_db": 0.8882291858958444, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0816221600350917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13015036440234273, "3000–6000 Hz_40ms_envelope_correlation": 0.9825462062576662, "3000–6000 Hz_40ms_modulation_mae_db": 1.2219891120784794, "mr_stft_db": 5.882911598428778, "band_mae_db": 0.11626343272947759, "rms_gain_error_db": 0.009878128491659038, "output_peak": 0.4636205732530508}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.990856629833176, "spectral_convergence": 0.31420096916783374, "spectral_convergence_status": "ok", "amplitude_floor": 1.2198198453862723e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.728186291261336, "spectral_convergence": 0.30538951306262296, "spectral_convergence_status": "ok", "amplitude_floor": 7.927151445605076e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.96213015522566, "spectral_convergence": 0.3298116969134701, "spectral_convergence_status": "ok", "amplitude_floor": 6.454878545778745e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00013983470011737857, "output_power": 0.00015053484837186797, "delta_db": 0.3202209407264711, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042159782923140474, "output_power": 0.0004252012517920571, "delta_db": 0.03696168347908179, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017365691446010552, "output_power": 0.0017053687882500638, "delta_db": -0.07873769961865161, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006597865121791135, "output_power": 0.0006172817437938352, "delta_db": -0.28920000317814637, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001856647778551521, "output_power": 0.00018725631209791165, "delta_db": 0.037069436642792344, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000145726779943178, "output_power": 0.0001395001721665414, "delta_db": -0.18964625002859337, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15863933561228946, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08538615720729324, "pitch": null, "output_rms_dbfs": -24.914510086829477, "output_peak": 0.47148644916452714, "reference_rms_dbfs": -24.82912392962218, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.89372435877339}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017345649672524565, "log_rms_mae_db": 0.957900314169872, "rms_envelope_nrmse": 0.08546056414630993, "envelope_correlation": 0.9925280012920177, "rms_gain_error_db": -0.08519839872491565, "reference_mean_power": 0.003291322552458033, "output_mean_power": 0.00322738373335975}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4377088226614287e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006890430591645026, "output_power": 0.00066567132976648, "reference_active": true, "delta_db": -0.14986509920617597}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.716540294963627e-05, "output_power": 9.310799801208801e-05, "reference_active": true, "delta_db": -0.1852466740721826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.855953967229281e-05, "output_power": 2.7395365669919255e-05, "reference_active": true, "delta_db": -0.18074101407428877}], "active_band_count": 3, "active_band_mae_db": 0.17195092911754908}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015649665907765232, "log_rms_mae_db": 0.5230335284328074, "rms_envelope_nrmse": 0.041995065985404735, "envelope_correlation": 0.9980751380931185, "rms_gain_error_db": -0.08512337280511088, "reference_mean_power": 0.003302205038848876, "output_mean_power": 0.0032381107503563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.335462563779621e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377316706455328, "output_power": 0.0006175157486590943, "reference_active": true, "delta_db": -0.13989947199303412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.741968368909027e-05, "output_power": 4.39762701778996e-05, "reference_active": true, "delta_db": -0.32740260559980405}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7622617238454117e-06, "output_power": 1.592820998347417e-06, "reference_active": true, "delta_db": -0.4390343607940619}], "active_band_count": 3, "active_band_mae_db": 0.3021121461289667}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014746435244218596, "log_rms_mae_db": 1.7253229937201127, "rms_envelope_nrmse": 0.10319191405651182, "envelope_correlation": 0.9909092421879793, "rms_gain_error_db": -0.0700653004526151, "reference_mean_power": 0.001906578877331925, "output_mean_power": 0.0018760665805834045}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.936060821906154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045087668804334914, "output_power": 0.0004418491970775469, "reference_active": true, "delta_db": -0.0878371115438081}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.461610811295139e-05, "output_power": 4.755917062942513e-05, "reference_active": true, "delta_db": -0.6008647705605172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.1974528143292685e-06, "output_power": 1.087525577376882e-05, "reference_active": true, "delta_db": 2.4422625100808664}], "active_band_count": 3, "active_band_mae_db": 1.0436547973950638}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013246473438636124, "log_rms_mae_db": 1.0093239401267822, "rms_envelope_nrmse": 0.056250135327573725, "envelope_correlation": 0.9971135229119421, "rms_gain_error_db": -0.06983797766353211, "reference_mean_power": 0.0019129188838057705, "output_mean_power": 0.001882403651648929}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.520895123968316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004268128119473447, "output_power": 0.00041922251292210753, "reference_active": true, "delta_db": -0.07792850565156878}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.526792747799959e-05, "output_power": 2.1454805737217616e-05, "reference_active": true, "delta_db": -0.7104503518438793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.02211562944459e-07, "output_power": 6.804070672094981e-07, "reference_active": true, "delta_db": 1.3188210815651213}], "active_band_count": 3, "active_band_mae_db": 0.7023999796868564}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.50803654710503e-05, "log_rms_mae_db": 1.705142942995269, "rms_envelope_nrmse": 0.15190163235565707, "envelope_correlation": 0.9810886643892522, "rms_gain_error_db": -0.2668490206729386, "reference_mean_power": 0.0007310287661738595, "output_mean_power": 0.0006874633734078164}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3756975679714823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020778711952139343, "output_power": 0.00018173129093576384, "reference_active": true, "delta_db": -0.5818891102565875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.101230284766401e-05, "output_power": 4.4183703265438696e-05, "reference_active": true, "delta_db": 0.32347957855094317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.417981343795666e-05, "output_power": 1.2981500961466918e-05, "reference_active": true, "delta_db": -0.3834560708146405}], "active_band_count": 3, "active_band_mae_db": 0.42960825320739043}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.582207074769948e-05, "log_rms_mae_db": 1.0221800208013945, "rms_envelope_nrmse": 0.09417336600052628, "envelope_correlation": 0.9925576672226532, "rms_gain_error_db": -0.2671797539611346, "reference_mean_power": 0.000733846607513954, "output_mean_power": 0.0006900607335283627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9980296656839734e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019557456921819033, "output_power": 0.0001722378789041005, "reference_active": true, "delta_db": -0.5518371377546203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6985486820992048e-05, "output_power": 1.920605456621967e-05, "reference_active": true, "delta_db": 0.5336015970387286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.92201473631864e-07, "output_power": 9.071722706898282e-07, "reference_active": true, "delta_db": -0.3891010094121965}], "active_band_count": 3, "active_band_mae_db": 0.4915132480685151}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011116745339746556, "log_rms_mae_db": 1.7148047694315094, "rms_envelope_nrmse": 0.19448344248865576, "envelope_correlation": 0.9661022307288126, "rms_gain_error_db": -0.29885843553482383, "reference_mean_power": 0.00017647199152424748, "output_mean_power": 0.00016473654118199897}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.117031117126669e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.030239739448223e-05, "output_power": 3.772891359257319e-05, "reference_active": true, "delta_db": -0.2865658115268598}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8894394185439072e-05, "output_power": 2.179584434433277e-05, "reference_active": true, "delta_db": -1.2243989554266186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.877164672472975e-06, "output_power": 6.972498900387412e-06, "reference_active": true, "delta_db": -0.52981470062639}], "active_band_count": 3, "active_band_mae_db": 0.6802598225266229}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.815724618424621e-05, "log_rms_mae_db": 1.0429817755866582, "rms_envelope_nrmse": 0.12767726983073602, "envelope_correlation": 0.9841089505261285, "rms_gain_error_db": -0.29978081934097656, "reference_mean_power": 0.0001772156182174664, "output_mean_power": 0.0001653955848786278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0001638026368737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.251019833717301e-05, "output_power": 3.920465704573789e-05, "reference_active": true, "delta_db": -0.35155472142217487}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.608059024163373e-05, "output_power": 1.1479090760698629e-05, "reference_active": true, "delta_db": -1.463944958850849}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.447224316204941e-07, "output_power": 5.183730890310793e-07, "reference_active": true, "delta_db": -0.9473033338269445}], "active_band_count": 3, "active_band_mae_db": 0.9209343380333227}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8808765044046947, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7445a01fe6743d02e7e72cf3e93565359bf497e6226156afb6fde8d8c938bc4d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9810660329773195, "standard_deviation": 0.0014415753574711977, "minimum": 0.9787784261186316, "maximum": 0.9825179835094894, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6628668259728445, "standard_deviation": 0.0455467034806869, "minimum": 1.5880101671628613, "maximum": 1.705142942995269, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4449587767347081, "standard_deviation": 0.2260074466658575, "minimum": 0.13813415455886077, "maximum": 0.7807787694389349, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15120065854304587, "standard_deviation": 0.005624796911805352, "minimum": 0.14496910621869735, "maximum": 0.1603445702889144, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9930018794777306, "standard_deviation": 0.0008015845442494625, "minimum": 0.9919900684544696, "maximum": 0.994247738008983, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9882845224303475, "standard_deviation": 0.03452243406171407, "minimum": 0.9415241661144219, "maximum": 1.0377585881521278, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.689214957300827, "standard_deviation": 0.3732476203651709, "minimum": 0.18537641265554994, "maximum": 1.1697941052105298, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0892394421679619, "standard_deviation": 0.0054064143629714715, "minimum": 0.08197620242482133, "maximum": 0.09638680867022856, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9616176195526134, "standard_deviation": 0.005350749804918898, "minimum": 0.9542232746210879, "maximum": 0.9688451312514513, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7889849250952792, "standard_deviation": 0.03349309256695556, "minimum": 1.7148047694315094, "maximum": 1.8196161294993654, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7306081869256509, "standard_deviation": 0.3033256630908772, "minimum": 0.34056182360503967, "maximum": 1.2586643482083224, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20618512604807038, "standard_deviation": 0.013799510050968652, "minimum": 0.18845624724192744, "maximum": 0.22520865981059804, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9807242819907862, "standard_deviation": 0.003411457691920249, "minimum": 0.9746617068337597, "maximum": 0.9850303204753019, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0711418901197234, "standard_deviation": 0.03369149143225059, "minimum": 1.0330892332229527, "maximum": 1.1257764178422833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.004609262089045, "standard_deviation": 0.39751276198427615, "minimum": 0.4619392536196792, "maximum": 1.6355890768277035, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13695896211265968, "standard_deviation": 0.011228737881093099, "minimum": 0.12219652519230102, "maximum": 0.15735448259890006, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9910040402198829, "standard_deviation": 0.0018094435275276524, "minimum": 0.9872213942014959, "maximum": 0.9931712418406143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.797482877263351, "standard_deviation": 0.055920298659820886, "minimum": 1.7253229937201127, "maximum": 1.8873607561312087, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1748263258456544, "standard_deviation": 0.39053503283951274, "minimum": 0.6259071992896129, "maximum": 1.7425421359797484, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10234272498873115, "standard_deviation": 0.00981188040798832, "minimum": 0.089860604206398, "maximum": 0.12222466084341779, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9970845688905863, "standard_deviation": 0.0005715619736147999, "minimum": 0.9963555281394212, "maximum": 0.9980877795349812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0755035534392783, "standard_deviation": 0.0385562512757798, "minimum": 1.0093239401267822, "maximum": 1.116963431708604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7296782905788095, "standard_deviation": 0.2556803001801577, "minimum": 0.32722436111995085, "maximum": 1.0503323884083595, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05641973421886724, "standard_deviation": 0.005652590382019403, "minimum": 0.04571071341116569, "maximum": 0.06329754624984567, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1433528100230204, "standard_deviation": 0.025621828475934628, "minimum": 0.10939791418301176, "maximum": 0.18578935057857515, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.900389962628691, "standard_deviation": 0.009314151458325403, "minimum": 5.882911598428778, "maximum": 5.911251718944379, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.47654294607684233, "standard_deviation": 0.019739767707736486, "minimum": 0.45592381674521704, "maximum": 0.5106213931910609, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.05168338521456367, "standard_deviation": 0.043993303305614066, "minimum": -0.08628410373169595, "maximum": 0.024491912159363104, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9925126359181012, "standard_deviation": 0.0011034881183022126, "minimum": 0.9904691870473745, "maximum": 0.9941922660547288, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9366211542978398, "standard_deviation": 0.021814027463569412, "minimum": 0.9135404482280007, "maximum": 0.9795915622238801, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.22872842668169768, "standard_deviation": 0.05405259528243768, "minimum": 0.17195092911754908, "maximum": 0.3245013993281674, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08527331236387645, "standard_deviation": 0.00617679291488254, "minimum": 0.07524184517804164, "maximum": 0.09621514519554995, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979446279023951, "standard_deviation": 0.00043372298542946713, "minimum": 0.9970257586171702, "maximum": 0.9983561768263458, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5035761665541807, "standard_deviation": 0.022443812146038557, "minimum": 0.47481965834098117, "maximum": 0.5412527032186049, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.44563707361096255, "standard_deviation": 0.23604833774682205, "minimum": 0.21126169002107595, "maximum": 0.8663923071699658, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.042785890258045356, "standard_deviation": 0.003951941033996788, "minimum": 0.03911022414801816, "maximum": 0.05126559109402668, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0003", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.1082029342651367, "predicted_mos": 2.1082029342651367, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fdd1b0e7c6874893653037a08e0bf35f8657d5713452f9865c24e6ffbdc2ee75", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.085428099613637, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9648912532559036, "全帯域_10ms_rms_envelope_nrmse": 0.08993271316270245, "全帯域_10ms_envelope_correlation": 0.9916844480891471, "全帯域_10ms_modulation_mae_db": 0.15130596104735541, "全帯域_40ms_log_rms_mae_db": 0.5116944638473273, "全帯域_40ms_rms_envelope_nrmse": 0.044574745286793345, "全帯域_40ms_envelope_correlation": 0.9977290013433064, "全帯域_40ms_modulation_mae_db": 0.30246918516858384, "300–1000 Hz_10ms_log_rms_mae_db": 1.4133000421664332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10429726799133177, "300–1000 Hz_10ms_envelope_correlation": 0.9907183844152296, "300–1000 Hz_10ms_modulation_mae_db": 1.4974808501766192, "300–1000 Hz_40ms_log_rms_mae_db": 0.7574781644875599, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05008762182959766, "300–1000 Hz_40ms_envelope_correlation": 0.9976751512348028, "300–1000 Hz_40ms_modulation_mae_db": 0.7534076545154592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4428797245482634, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1458419570814115, "1000–3000 Hz_10ms_envelope_correlation": 0.9823243293523626, "1000–3000 Hz_10ms_modulation_mae_db": 0.36489024982459356, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8277099503785413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08243169960077563, "1000–3000 Hz_40ms_envelope_correlation": 0.9939282116219775, "1000–3000 Hz_40ms_modulation_mae_db": 0.6940166345387837, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4544319680665105, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17598345262102855, "3000–6000 Hz_10ms_envelope_correlation": 0.973223318716707, "3000–6000 Hz_10ms_modulation_mae_db": 0.2930433492405469, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9105981728156715, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11170815506275698, "3000–6000 Hz_40ms_envelope_correlation": 0.9878102007654462, "3000–6000 Hz_40ms_modulation_mae_db": 0.28920702350152483, "mr_stft_db": 5.606729185527371, "band_mae_db": 0.135026046298187, "rms_gain_error_db": -0.00872176722872879, "output_peak": 0.45406127413081815}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.950244734732273, "全帯域_10ms_rms_envelope_nrmse": 0.07964668370940338, "全帯域_10ms_envelope_correlation": 0.9935072229078694, "全帯域_10ms_modulation_mae_db": 0.3074622736377027, "全帯域_40ms_log_rms_mae_db": 0.5261763835875157, "全帯域_40ms_rms_envelope_nrmse": 0.04013764945466935, "全帯域_40ms_envelope_correlation": 0.9982180382204452, "全帯域_40ms_modulation_mae_db": 0.47572863743305077, "300–1000 Hz_10ms_log_rms_mae_db": 1.4189223824091346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08491244899500942, "300–1000 Hz_10ms_envelope_correlation": 0.9938563814813155, "300–1000 Hz_10ms_modulation_mae_db": 0.832863824560356, "300–1000 Hz_40ms_log_rms_mae_db": 0.7625428734799903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04449138992186963, "300–1000 Hz_40ms_envelope_correlation": 0.9981974932193854, "300–1000 Hz_40ms_modulation_mae_db": 0.42147985567955143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4008125066119619, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1327757669858622, "1000–3000 Hz_10ms_envelope_correlation": 0.985810946521714, "1000–3000 Hz_10ms_modulation_mae_db": 0.5800745635986467, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8220816927384654, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07659331652887548, "1000–3000 Hz_40ms_envelope_correlation": 0.9954706766854221, "1000–3000 Hz_40ms_modulation_mae_db": 0.6161002879612779, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4708495818372607, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18371216121427675, "3000–6000 Hz_10ms_envelope_correlation": 0.9708453043214993, "3000–6000 Hz_10ms_modulation_mae_db": 0.5088500990518505, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8968728372668386, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11992212914321568, "3000–6000 Hz_40ms_envelope_correlation": 0.9860084937818793, "3000–6000 Hz_40ms_modulation_mae_db": 0.5443888348731346, "mr_stft_db": 5.602490354390192, "band_mae_db": 0.1622265789246345, "rms_gain_error_db": -0.0663431028250466, "output_peak": 0.4678454842942461}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9459362365917653, "全帯域_10ms_rms_envelope_nrmse": 0.0862144886082648, "全帯域_10ms_envelope_correlation": 0.9923483644655783, "全帯域_10ms_modulation_mae_db": 0.21309106516276863, "全帯域_40ms_log_rms_mae_db": 0.5010343099238791, "全帯域_40ms_rms_envelope_nrmse": 0.04069547076216232, "全帯域_40ms_envelope_correlation": 0.9981108745658375, "全帯域_40ms_modulation_mae_db": 0.5503933945112519, "300–1000 Hz_10ms_log_rms_mae_db": 1.4363898301375753, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10192711090724839, "300–1000 Hz_10ms_envelope_correlation": 0.9911644005733319, "300–1000 Hz_10ms_modulation_mae_db": 1.4316187310927144, "300–1000 Hz_40ms_log_rms_mae_db": 0.8178814940841279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050634319653540205, "300–1000 Hz_40ms_envelope_correlation": 0.9976317755059996, "300–1000 Hz_40ms_modulation_mae_db": 0.9885067390172306, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3948901136637384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14593130565040613, "1000–3000 Hz_10ms_envelope_correlation": 0.9825100441408335, "1000–3000 Hz_10ms_modulation_mae_db": 0.5134647663414974, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8380911901148975, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07632879494745105, "1000–3000 Hz_40ms_envelope_correlation": 0.9952158597111765, "1000–3000 Hz_40ms_modulation_mae_db": 0.8843945867332313, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4246302410172271, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18302835261879102, "3000–6000 Hz_10ms_envelope_correlation": 0.9697501118381164, "3000–6000 Hz_10ms_modulation_mae_db": 1.1029014740335394, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8568921352551457, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1185222064282336, "3000–6000 Hz_40ms_envelope_correlation": 0.9858267454632473, "3000–6000 Hz_40ms_modulation_mae_db": 1.4935353496398065, "mr_stft_db": 5.604330325402842, "band_mae_db": 0.12827842152751553, "rms_gain_error_db": -0.03446025757063803, "output_peak": 0.45559633149003936}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9158661109769793, "全帯域_10ms_rms_envelope_nrmse": 0.08204652097861981, "全帯域_10ms_envelope_correlation": 0.993244794170192, "全帯域_10ms_modulation_mae_db": 0.36641946315847856, "全帯域_40ms_log_rms_mae_db": 0.5035535058725731, "全帯域_40ms_rms_envelope_nrmse": 0.045792574106361277, "全帯域_40ms_envelope_correlation": 0.997849506061546, "全帯域_40ms_modulation_mae_db": 0.7276198376541445, "300–1000 Hz_10ms_log_rms_mae_db": 1.458520246700467, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10169577866566354, "300–1000 Hz_10ms_envelope_correlation": 0.9912942530760142, "300–1000 Hz_10ms_modulation_mae_db": 1.1341994156693267, "300–1000 Hz_40ms_log_rms_mae_db": 0.810279425387214, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0634360917848587, "300–1000 Hz_40ms_envelope_correlation": 0.9964797107679335, "300–1000 Hz_40ms_modulation_mae_db": 0.5264027622031496, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3832703844292213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14896538025721193, "1000–3000 Hz_10ms_envelope_correlation": 0.9818411735988206, "1000–3000 Hz_10ms_modulation_mae_db": 0.5017727224779872, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8296291270015854, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08914479243084533, "1000–3000 Hz_40ms_envelope_correlation": 0.993410185405472, "1000–3000 Hz_40ms_modulation_mae_db": 0.8978456719747602, "3000–6000 Hz_10ms_log_rms_mae_db": 1.473088000117768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.182365497766021, "3000–6000 Hz_10ms_envelope_correlation": 0.9699001860750871, "3000–6000 Hz_10ms_modulation_mae_db": 0.8400210240449283, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8929605422380759, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11846655182700813, "3000–6000 Hz_40ms_envelope_correlation": 0.9854666620327649, "3000–6000 Hz_40ms_modulation_mae_db": 1.0435686909117359, "mr_stft_db": 5.605912080876692, "band_mae_db": 0.1597726159351782, "rms_gain_error_db": -0.13802755788768145, "output_peak": 0.44486334419056267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9232962032716245, "全帯域_10ms_rms_envelope_nrmse": 0.08062099833022321, "全帯域_10ms_envelope_correlation": 0.9934730892337629, "全帯域_10ms_modulation_mae_db": 0.33271054637462716, "全帯域_40ms_log_rms_mae_db": 0.47717178453301623, "全帯域_40ms_rms_envelope_nrmse": 0.0412315841402088, "全帯域_40ms_envelope_correlation": 0.9982987690209839, "全帯域_40ms_modulation_mae_db": 0.48994037180665134, "300–1000 Hz_10ms_log_rms_mae_db": 1.49972192414256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10367597941998566, "300–1000 Hz_10ms_envelope_correlation": 0.990961946239287, "300–1000 Hz_10ms_modulation_mae_db": 0.6914045787630559, "300–1000 Hz_40ms_log_rms_mae_db": 0.866884808205903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059641620079961974, "300–1000 Hz_40ms_envelope_correlation": 0.9969619694075611, "300–1000 Hz_40ms_modulation_mae_db": 0.2826139068038323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3920655547446215, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1296677414039322, "1000–3000 Hz_10ms_envelope_correlation": 0.986137710258704, "1000–3000 Hz_10ms_modulation_mae_db": 0.36845216502271577, "1000–3000 Hz_40ms_log_rms_mae_db": 0.78064931384218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06249694232901655, "1000–3000 Hz_40ms_envelope_correlation": 0.9967292464050762, "1000–3000 Hz_40ms_modulation_mae_db": 0.348219667592274, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4682134677179817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18150840212878058, "3000–6000 Hz_10ms_envelope_correlation": 0.9704102805678669, "3000–6000 Hz_10ms_modulation_mae_db": 0.8575862274273155, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8874460895019614, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12216576558413061, "3000–6000 Hz_40ms_envelope_correlation": 0.9852317609541018, "3000–6000 Hz_40ms_modulation_mae_db": 1.191260934539619, "mr_stft_db": 5.60591706910657, "band_mae_db": 0.17136593242373177, "rms_gain_error_db": -0.132365981839703, "output_peak": 0.47221784912824905}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9177680357515802, "全帯域_10ms_rms_envelope_nrmse": 0.08569195013119041, "全帯域_10ms_envelope_correlation": 0.9924892265280656, "全帯域_10ms_modulation_mae_db": 0.7852759464949534, "全帯域_40ms_log_rms_mae_db": 0.48875328774312726, "全帯域_40ms_rms_envelope_nrmse": 0.04397629432147636, "全帯域_40ms_envelope_correlation": 0.9978812484739878, "全帯域_40ms_modulation_mae_db": 1.1681338545263527, "300–1000 Hz_10ms_log_rms_mae_db": 1.4053079882642867, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09361780649319197, "300–1000 Hz_10ms_envelope_correlation": 0.9925344132067715, "300–1000 Hz_10ms_modulation_mae_db": 0.7259896273211758, "300–1000 Hz_40ms_log_rms_mae_db": 0.7625257979131191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052436685712463994, "300–1000 Hz_40ms_envelope_correlation": 0.9975036357106193, "300–1000 Hz_40ms_modulation_mae_db": 0.24353627419430335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4411311926743842, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13359391594872366, "1000–3000 Hz_10ms_envelope_correlation": 0.9853567611429155, "1000–3000 Hz_10ms_modulation_mae_db": 0.32105475746419115, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8185823725749584, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06744686153646073, "1000–3000 Hz_40ms_envelope_correlation": 0.9962764199924355, "1000–3000 Hz_40ms_modulation_mae_db": 0.72608905789152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4487092516149285, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20432168371492695, "3000–6000 Hz_10ms_envelope_correlation": 0.9622154501174619, "3000–6000 Hz_10ms_modulation_mae_db": 1.1666670151200735, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8816768715111607, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13174197219813913, "3000–6000 Hz_40ms_envelope_correlation": 0.9826909075355549, "3000–6000 Hz_40ms_modulation_mae_db": 1.4506174472498035, "mr_stft_db": 5.5957925665544055, "band_mae_db": 0.16395436988620682, "rms_gain_error_db": -0.09081098229223586, "output_peak": 0.47620172244267145}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9674333355665397, "全帯域_10ms_rms_envelope_nrmse": 0.08469774989277212, "全帯域_10ms_envelope_correlation": 0.9926355593332818, "全帯域_10ms_modulation_mae_db": 0.11138121420202611, "全帯域_40ms_log_rms_mae_db": 0.5013471255761648, "全帯域_40ms_rms_envelope_nrmse": 0.041916964594716916, "全帯域_40ms_envelope_correlation": 0.9980399362699774, "全帯域_40ms_modulation_mae_db": 0.5146230002500792, "300–1000 Hz_10ms_log_rms_mae_db": 1.4822362496919357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10213361210558804, "300–1000 Hz_10ms_envelope_correlation": 0.9911203598485994, "300–1000 Hz_10ms_modulation_mae_db": 1.6134940989288282, "300–1000 Hz_40ms_log_rms_mae_db": 0.8410299342078358, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05763598174548413, "300–1000 Hz_40ms_envelope_correlation": 0.997001500682565, "300–1000 Hz_40ms_modulation_mae_db": 0.8658931472701861, "1000–3000 Hz_10ms_log_rms_mae_db": 1.43815753244091, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1465049993123896, "1000–3000 Hz_10ms_envelope_correlation": 0.9821743983300952, "1000–3000 Hz_10ms_modulation_mae_db": 0.31022601469123917, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8182953673796219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08794716547456753, "1000–3000 Hz_40ms_envelope_correlation": 0.9930900957568517, "1000–3000 Hz_40ms_modulation_mae_db": 0.5052519572166819, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4840816244566088, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18224092492225652, "3000–6000 Hz_10ms_envelope_correlation": 0.9701068231456812, "3000–6000 Hz_10ms_modulation_mae_db": 0.6273880111273149, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9176077326552531, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11285058012259472, "3000–6000 Hz_40ms_envelope_correlation": 0.9874065551088587, "3000–6000 Hz_40ms_modulation_mae_db": 0.6726986550548325, "mr_stft_db": 5.606788645088216, "band_mae_db": 0.1211395334510021, "rms_gain_error_db": -0.06923958227398434, "output_peak": 0.4706837747756567}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8999039334421485, "全帯域_10ms_rms_envelope_nrmse": 0.08191245759204492, "全帯域_10ms_envelope_correlation": 0.9931000514214492, "全帯域_10ms_modulation_mae_db": 0.10549263051226936, "全帯域_40ms_log_rms_mae_db": 0.48622078194045865, "全帯域_40ms_rms_envelope_nrmse": 0.0417733146917618, "全帯域_40ms_envelope_correlation": 0.99801392621188, "全帯域_40ms_modulation_mae_db": 0.35949767932550447, "300–1000 Hz_10ms_log_rms_mae_db": 1.4735215486423558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10109993449181301, "300–1000 Hz_10ms_envelope_correlation": 0.9912723520477139, "300–1000 Hz_10ms_modulation_mae_db": 1.48397478221369, "300–1000 Hz_40ms_log_rms_mae_db": 0.8338462206641255, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051256116762388816, "300–1000 Hz_40ms_envelope_correlation": 0.9975919230940903, "300–1000 Hz_40ms_modulation_mae_db": 0.6468759027912027, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3578727493795673, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14470287910896545, "1000–3000 Hz_10ms_envelope_correlation": 0.9827012039666108, "1000–3000 Hz_10ms_modulation_mae_db": 0.2099864150602122, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8185107027468805, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08978183816185219, "1000–3000 Hz_40ms_envelope_correlation": 0.99290999829801, "1000–3000 Hz_40ms_modulation_mae_db": 0.5501368942152913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4549003869974948, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1795336973343409, "3000–6000 Hz_10ms_envelope_correlation": 0.9711429027946012, "3000–6000 Hz_10ms_modulation_mae_db": 0.4702617069039983, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8639171455499969, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1098032205575194, "3000–6000 Hz_40ms_envelope_correlation": 0.987605505120429, "3000–6000 Hz_40ms_modulation_mae_db": 0.5070883512247129, "mr_stft_db": 5.600108247344341, "band_mae_db": 0.09798496701659021, "rms_gain_error_db": -0.033495503540057744, "output_peak": 0.4592708617870995}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.604177979005115, "spectral_convergence": 0.3149668230807602, "spectral_convergence_status": "ok", "amplitude_floor": 1.2198198453862723e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.379813466040545, "spectral_convergence": 0.3000764157735365, "spectral_convergence_status": "ok", "amplitude_floor": 7.927151445605076e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.836196111536454, "spectral_convergence": 0.31780163823925006, "spectral_convergence_status": "ok", "amplitude_floor": 6.454878545778745e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00013983470011737857, "output_power": 0.00014958900060950734, "delta_db": 0.2928470532280342, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042159782923140474, "output_power": 0.0004190163616123666, "delta_db": -0.026673845830006933, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017365691446010552, "output_power": 0.0017363908125459602, "delta_db": -0.0004460093320850646, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006597865121791135, "output_power": 0.0006329490519422345, "delta_db": -0.18034679449278887, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001856647778551521, "output_power": 0.00019754215370384337, "delta_db": 0.26930262058626525, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000145726779943178, "output_power": 0.00014709346009359335, "delta_db": 0.040539954319941576, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.135026046298187, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.00872176722872879, "pitch": null, "output_rms_dbfs": -24.837845696850913, "output_peak": 0.45406127413081815, "reference_rms_dbfs": -24.82912392962218, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.606729185527371}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017345649672524565, "log_rms_mae_db": 0.9648912532559036, "rms_envelope_nrmse": 0.08993271316270245, "envelope_correlation": 0.9916844480891471, "rms_gain_error_db": -0.008685866351927703, "reference_mean_power": 0.003291322552458033, "output_mean_power": 0.0032847465032242416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4377088226614287e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006890430591645026, "output_power": 0.0006912153063737312, "reference_active": true, "delta_db": 0.013669842961756665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.716540294963627e-05, "output_power": 0.00010597480179081449, "reference_active": true, "delta_db": 0.3769095691107196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.855953967229281e-05, "output_power": 2.8979110456956148e-05, "reference_active": true, "delta_db": 0.06333847106958995}], "active_band_count": 3, "active_band_mae_db": 0.15130596104735541}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015649665907765232, "log_rms_mae_db": 0.5116944638473273, "rms_envelope_nrmse": 0.044574745286793345, "envelope_correlation": 0.9977290013433064, "rms_gain_error_db": -0.007956813840747553, "reference_mean_power": 0.003302205038848876, "output_mean_power": 0.003296160528272241}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.335462563779621e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377316706455328, "output_power": 0.0006450885024249596, "reference_active": true, "delta_db": 0.04981316229572959}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.741968368909027e-05, "output_power": 5.017244769692067e-05, "reference_active": true, "delta_db": 0.2450663618607623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7622617238454117e-06, "output_power": 1.5304445718259291e-06, "reference_active": true, "delta_db": -0.6125280313492596}], "active_band_count": 3, "active_band_mae_db": 0.30246918516858384}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014746435244218596, "log_rms_mae_db": 1.4133000421664332, "rms_envelope_nrmse": 0.10429726799133177, "envelope_correlation": 0.9907183844152296, "rms_gain_error_db": -0.014685925169529416, "reference_mean_power": 0.001906578877331925, "output_mean_power": 0.0019001425564512072}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.936060821906154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045087668804334914, "output_power": 0.0004520216391553744, "reference_active": true, "delta_db": 0.011014445404381143}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.461610811295139e-05, "output_power": 5.8468357036919184e-05, "reference_active": true, "delta_db": 0.2960014078379864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.1974528143292685e-06, "output_power": 1.6246353495795352e-05, "reference_active": true, "delta_db": 4.1854266972874905}], "active_band_count": 3, "active_band_mae_db": 1.4974808501766192}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013246473438636124, "log_rms_mae_db": 0.7574781644875599, "rms_envelope_nrmse": 0.05008762182959766, "envelope_correlation": 0.9976751512348028, "rms_gain_error_db": -0.013487780109575964, "reference_mean_power": 0.0019129188838057705, "output_mean_power": 0.001906987192990187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.520895123968316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004268128119473447, "output_power": 0.0004331888876200164, "reference_active": true, "delta_db": 0.06439859876812218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.526792747799959e-05, "output_power": 2.6335511217981797e-05, "reference_active": true, "delta_db": 0.17972131438440356}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.02211562944459e-07, "output_power": 7.989084464379564e-07, "reference_active": true, "delta_db": 2.016103050393852}], "active_band_count": 3, "active_band_mae_db": 0.7534076545154592}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.50803654710503e-05, "log_rms_mae_db": 1.4428797245482634, "rms_envelope_nrmse": 0.1458419570814115, "envelope_correlation": 0.9823243293523626, "rms_gain_error_db": -0.12626073128058762, "reference_mean_power": 0.0007310287661738595, "output_mean_power": 0.0007100818201323548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3756975679714823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020778711952139343, "output_power": 0.0001943614497073284, "reference_active": true, "delta_db": -0.2900849292453773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.101230284766401e-05, "output_power": 4.424019284236901e-05, "reference_active": true, "delta_db": 0.32902855704280576}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.417981343795666e-05, "output_power": 1.270909981569914e-05, "reference_active": true, "delta_db": -0.47555726318559766}], "active_band_count": 3, "active_band_mae_db": 0.36489024982459356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.582207074769948e-05, "log_rms_mae_db": 0.8277099503785413, "rms_envelope_nrmse": 0.08243169960077563, "envelope_correlation": 0.9939282116219775, "rms_gain_error_db": -0.12704621605123112, "reference_mean_power": 0.000733846607513954, "output_mean_power": 0.000712690006846014}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9980296656839734e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019557456921819033, "output_power": 0.00018310004178420544, "reference_active": true, "delta_db": -0.28623938911970687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6985486820992048e-05, "output_power": 1.8683163465554634e-05, "reference_active": true, "delta_db": 0.4137241491583664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.92201473631864e-07, "output_power": 7.217573765634209e-07, "reference_active": true, "delta_db": -1.3820863653382776}], "active_band_count": 3, "active_band_mae_db": 0.6940166345387837}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011116745339746556, "log_rms_mae_db": 1.4544319680665105, "rms_envelope_nrmse": 0.17598345262102855, "envelope_correlation": 0.973223318716707, "rms_gain_error_db": 0.11923192912430258, "reference_mean_power": 0.00017647199152424748, "output_mean_power": 0.00018138400197161914}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.117031117126669e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.030239739448223e-05, "output_power": 4.370144965641595e-05, "reference_active": true, "delta_db": 0.35164962560259694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8894394185439072e-05, "output_power": 3.1711718685635026e-05, "reference_active": true, "delta_db": 0.4040618694099606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.877164672472975e-06, "output_power": 7.656460904048858e-06, "reference_active": true, "delta_db": -0.12341855270908321}], "active_band_count": 3, "active_band_mae_db": 0.2930433492405469}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.815724618424621e-05, "log_rms_mae_db": 0.9105981728156715, "rms_envelope_nrmse": 0.11170815506275698, "envelope_correlation": 0.9878102007654462, "rms_gain_error_db": 0.11938421187205506, "reference_mean_power": 0.0001772156182174664, "output_mean_power": 0.00018215471417890386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0001638026368737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.251019833717301e-05, "output_power": 4.7091248804717375e-05, "reference_active": true, "delta_db": 0.4444707634356331}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.608059024163373e-05, "output_power": 1.6864263148574508e-05, "reference_active": true, "delta_db": 0.20665384700343267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.447224316204941e-07, "output_power": 6.133708530024064e-07, "reference_active": true, "delta_db": -0.21649646006550868}], "active_band_count": 3, "active_band_mae_db": 0.28920702350152483}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8881234679789329, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b5e44edbba8901b034b222d1f45ca63e881fe5942833b5910f3f40238e6619fa", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.983607070914007, "standard_deviation": 0.0018190677253630025, "minimum": 0.9818411735988206, "maximum": 0.986137710258704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4063849698115836, "standard_deviation": 0.031178317014576646, "minimum": 1.3578727493795673, "maximum": 1.4428797245482634, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3962402068101354, "standard_deviation": 0.12430983354170756, "minimum": 0.2099864150602122, "maximum": 0.5800745635986467, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14099799321861284, "standard_deviation": 0.007617030541388791, "minimum": 0.1296677414039322, "maximum": 0.14896538025721193, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946288367345527, "standard_deviation": 0.0014869747731207576, "minimum": 0.99290999829801, "maximum": 0.9967292464050762, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8191937145971413, "standard_deviation": 0.01704698215505003, "minimum": 0.78064931384218, "maximum": 0.8380911901148975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6527568447654775, "standard_deviation": 0.18791324061914097, "minimum": 0.348219667592274, "maximum": 0.8978456719747602, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07902142637623055, "standard_deviation": 0.010210239186286352, "minimum": 0.06249694232901655, "maximum": 0.08978183816185219, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9696992971971277, "standard_deviation": 0.0032191067406455273, "minimum": 0.9622154501174619, "maximum": 0.973223318716707, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4598630652282225, "standard_deviation": 0.018407121038484464, "minimum": 1.4246302410172271, "maximum": 1.4840816244566088, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7333398633686959, "standard_deviation": 0.31072698618052963, "minimum": 0.2930433492405469, "maximum": 1.1666670151200735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1840867715400528, "standard_deviation": 0.008536680840623282, "minimum": 0.17598345262102855, "maximum": 0.20432168371492695, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9860058538452853, "standard_deviation": 0.001677047553341704, "minimum": 0.9826909075355549, "maximum": 0.9878102007654462, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8884964408492629, "standard_deviation": 0.020964697326993992, "minimum": 0.8568921352551457, "maximum": 0.9176077326552531, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8990456608743962, "standard_deviation": 0.457707677366485, "minimum": 0.28920702350152483, "maximum": 1.4935353496398065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11814757261544978, "standard_deviation": 0.007005071254890807, "minimum": 0.1098032205575194, "maximum": 0.13174197219813913, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916153113610329, "standard_deviation": 0.0010533961720062767, "minimum": 0.9907183844152296, "maximum": 0.9938563814813155, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4484900265193437, "standard_deviation": 0.03508027750031018, "minimum": 1.4053079882642867, "maximum": 1.49972192414256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1763782385907209, "standard_deviation": 0.3801824867886236, "minimum": 0.6914045787630559, "maximum": 1.6134940989288282, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09916999238372898, "standard_deviation": 0.006624480607808555, "minimum": 0.08491244899500942, "maximum": 0.10429726799133177, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973803949528697, "standard_deviation": 0.0005355275463507995, "minimum": 0.9964797107679335, "maximum": 0.9981974932193854, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8065585898037344, "standard_deviation": 0.04140281560970505, "minimum": 0.7574781644875599, "maximum": 0.866884808205903, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5910895303093644, "standard_deviation": 0.27040156759418166, "minimum": 0.24353627419430335, "maximum": 0.9885067390172306, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05370247843627064, "standard_deviation": 0.0060991912279919425, "minimum": 0.04449138992186963, "maximum": 0.0634360917848587, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14246855818288076, "standard_deviation": 0.025848393930232933, "minimum": 0.09798496701659021, "maximum": 0.17136593242373177, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.603508559286329, "standard_deviation": 0.003882560601025556, "minimum": 5.5957925665544055, "maximum": 5.606788645088216, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4625925802799179, "standard_deviation": 0.010810083173708577, "minimum": 0.44486334419056267, "maximum": 0.47620172244267145, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07168309193225947, "standard_deviation": 0.046739893356311984, "minimum": -0.13802755788768145, "maximum": -0.00872176722872879, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9928103445186682, "standard_deviation": 0.0006338410797138939, "minimum": 0.9916844480891471, "maximum": 0.9935072229078694, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9356674804486018, "standard_deviation": 0.024858474526610535, "minimum": 0.8999039334421485, "maximum": 0.9674333355665397, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29664238757377265, "standard_deviation": 0.22168607454516046, "minimum": 0.10549263051226936, "maximum": 0.7852759464949534, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08384544530065263, "standard_deviation": 0.0034184881116808897, "minimum": 0.07964668370940338, "maximum": 0.08993271316270245, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980176625209956, "standard_deviation": 0.00019219800543385362, "minimum": 0.9977290013433064, "maximum": 0.9982987690209839, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4994939553780078, "standard_deviation": 0.015459207380221909, "minimum": 0.47717178453301623, "maximum": 0.5261763835875157, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5735507450844524, "standard_deviation": 0.27180320813515135, "minimum": 0.30246918516858384, "maximum": 1.1681338545263527, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04251232466976877, "standard_deviation": 0.0020226484928120435, "minimum": 0.04013764945466935, "maximum": 0.045792574106361277, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0003", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.5285613536834717, "predicted_mos": 2.5285613536834717, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d1e65d792ce57ca4039f0dc9ea415d4181c9fa4355a120f94d5c02d24b77c7aa", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9632480000145733, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9421548689077764, "全帯域_10ms_rms_envelope_nrmse": 0.08186026543099417, "全帯域_10ms_envelope_correlation": 0.9931134654454774, "全帯域_10ms_modulation_mae_db": 0.7757145525685596, "全帯域_40ms_log_rms_mae_db": 0.4804512384901464, "全帯域_40ms_rms_envelope_nrmse": 0.04026722225211556, "全帯域_40ms_envelope_correlation": 0.9981646322956464, "全帯域_40ms_modulation_mae_db": 0.9192860412258236, "300–1000 Hz_10ms_log_rms_mae_db": 1.5364359674204748, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0857926479665321, "300–1000 Hz_10ms_envelope_correlation": 0.9937261122505927, "300–1000 Hz_10ms_modulation_mae_db": 0.5205459002364461, "300–1000 Hz_40ms_log_rms_mae_db": 0.8323093228998802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04299646087554328, "300–1000 Hz_40ms_envelope_correlation": 0.9983145685008717, "300–1000 Hz_40ms_modulation_mae_db": 0.31455344045598227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4757849347048502, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14921531175703176, "1000–3000 Hz_10ms_envelope_correlation": 0.9815203632064907, "1000–3000 Hz_10ms_modulation_mae_db": 1.1257556432006257, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8898865325805888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08353778160251471, "1000–3000 Hz_40ms_envelope_correlation": 0.993869529349745, "1000–3000 Hz_40ms_modulation_mae_db": 1.4160478488776391, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5543209476162059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18875756157656676, "3000–6000 Hz_10ms_envelope_correlation": 0.9678424705026432, "3000–6000 Hz_10ms_modulation_mae_db": 0.9211483639580388, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9238260023236716, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11830012612432211, "3000–6000 Hz_40ms_envelope_correlation": 0.9860654960313446, "3000–6000 Hz_40ms_modulation_mae_db": 1.2679728022739576, "mr_stft_db": 5.684365651539192, "band_mae_db": 0.11424863914267806, "rms_gain_error_db": -0.0472928028158083, "output_peak": 0.48447264011670993}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9676128600174929, "全帯域_10ms_rms_envelope_nrmse": 0.08300294920658423, "全帯域_10ms_envelope_correlation": 0.9930180176657812, "全帯域_10ms_modulation_mae_db": 0.40058680665984786, "全帯域_40ms_log_rms_mae_db": 0.5163582094505196, "全帯域_40ms_rms_envelope_nrmse": 0.043584121051387266, "全帯域_40ms_envelope_correlation": 0.9979899873467655, "全帯域_40ms_modulation_mae_db": 0.8143341220518305, "300–1000 Hz_10ms_log_rms_mae_db": 1.573988995192098, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08983803429323889, "300–1000 Hz_10ms_envelope_correlation": 0.99320699526104, "300–1000 Hz_10ms_modulation_mae_db": 1.0843430939336216, "300–1000 Hz_40ms_log_rms_mae_db": 0.8457626199333823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04638221303154676, "300–1000 Hz_40ms_envelope_correlation": 0.9981808819276516, "300–1000 Hz_40ms_modulation_mae_db": 0.376463036518357, "1000–3000 Hz_10ms_log_rms_mae_db": 1.526325278660568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1460781036914019, "1000–3000 Hz_10ms_envelope_correlation": 0.9824497162407041, "1000–3000 Hz_10ms_modulation_mae_db": 0.611041808141094, "1000–3000 Hz_40ms_log_rms_mae_db": 0.887804887674129, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08347486536399949, "1000–3000 Hz_40ms_envelope_correlation": 0.9941469392939546, "1000–3000 Hz_40ms_modulation_mae_db": 0.7707181636965585, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5578533281281886, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19598893031281875, "3000–6000 Hz_10ms_envelope_correlation": 0.9656057868322313, "3000–6000 Hz_10ms_modulation_mae_db": 1.2373113929677195, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9241050071806893, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12705820112670646, "3000–6000 Hz_40ms_envelope_correlation": 0.9845037258159678, "3000–6000 Hz_40ms_modulation_mae_db": 1.4519240764433314, "mr_stft_db": 5.702148985048559, "band_mae_db": 0.17705872445523604, "rms_gain_error_db": -0.11052119419209208, "output_peak": 0.4606441335764795}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9540962494532835, "全帯域_10ms_rms_envelope_nrmse": 0.08396462957493919, "全帯域_10ms_envelope_correlation": 0.992743755835261, "全帯域_10ms_modulation_mae_db": 0.28019585420605986, "全帯域_40ms_log_rms_mae_db": 0.5178632336782021, "全帯域_40ms_rms_envelope_nrmse": 0.04501903856451598, "全帯域_40ms_envelope_correlation": 0.9976760794030215, "全帯域_40ms_modulation_mae_db": 0.13561693279734124, "300–1000 Hz_10ms_log_rms_mae_db": 1.559451257719869, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10207298774121656, "300–1000 Hz_10ms_envelope_correlation": 0.9911018455878491, "300–1000 Hz_10ms_modulation_mae_db": 1.096266570896826, "300–1000 Hz_40ms_log_rms_mae_db": 0.8970555669576686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06007108265244835, "300–1000 Hz_40ms_envelope_correlation": 0.9966545874639517, "300–1000 Hz_40ms_modulation_mae_db": 0.7241326770230896, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5624584380965527, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1498858685334793, "1000–3000 Hz_10ms_envelope_correlation": 0.9814092083492761, "1000–3000 Hz_10ms_modulation_mae_db": 0.4439608337860929, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9270609477481845, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07980567485587077, "1000–3000 Hz_40ms_envelope_correlation": 0.9945585019071905, "1000–3000 Hz_40ms_modulation_mae_db": 0.8663839070965261, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5086480198152026, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18971573544918613, "3000–6000 Hz_10ms_envelope_correlation": 0.969574928588177, "3000–6000 Hz_10ms_modulation_mae_db": 1.157771083931116, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9204970001064897, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13771909687569414, "3000–6000 Hz_40ms_envelope_correlation": 0.9822022405831918, "3000–6000 Hz_40ms_modulation_mae_db": 1.3091796983448172, "mr_stft_db": 5.691008569139637, "band_mae_db": 0.1712179517340728, "rms_gain_error_db": -0.025236781468994402, "output_peak": 0.4561276982484861}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8827343736805295, "全帯域_10ms_rms_envelope_nrmse": 0.07818899994342368, "全帯域_10ms_envelope_correlation": 0.9937639072461653, "全帯域_10ms_modulation_mae_db": 0.1491324113348351, "全帯域_40ms_log_rms_mae_db": 0.46345536149831135, "全帯域_40ms_rms_envelope_nrmse": 0.04067439230898135, "全帯域_40ms_envelope_correlation": 0.9981985351381665, "全帯域_40ms_modulation_mae_db": 0.3346997553538557, "300–1000 Hz_10ms_log_rms_mae_db": 1.535764420874985, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09438292965279119, "300–1000 Hz_10ms_envelope_correlation": 0.9924378754013937, "300–1000 Hz_10ms_modulation_mae_db": 1.2020669978307403, "300–1000 Hz_40ms_log_rms_mae_db": 0.8745595634760686, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054954513185385294, "300–1000 Hz_40ms_envelope_correlation": 0.9972977805323466, "300–1000 Hz_40ms_modulation_mae_db": 0.6601429062765999, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4578070357980186, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1294816596083627, "1000–3000 Hz_10ms_envelope_correlation": 0.9861850717748499, "1000–3000 Hz_10ms_modulation_mae_db": 0.5135978212936722, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8408125989157627, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07278031047207563, "1000–3000 Hz_40ms_envelope_correlation": 0.9954469768824227, "1000–3000 Hz_40ms_modulation_mae_db": 0.6019207803704729, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5508241557200815, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20332287015000716, "3000–6000 Hz_10ms_envelope_correlation": 0.9629357349124924, "3000–6000 Hz_10ms_modulation_mae_db": 0.2764557302648425, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9424779905225454, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12657441863166666, "3000–6000 Hz_40ms_envelope_correlation": 0.9834844436294964, "3000–6000 Hz_40ms_modulation_mae_db": 0.19814429257401603, "mr_stft_db": 5.677196587665204, "band_mae_db": 0.1465680416430328, "rms_gain_error_db": -0.08647703246514768, "output_peak": 0.4726934251489727}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9364402750880262, "全帯域_10ms_rms_envelope_nrmse": 0.0820598795008865, "全帯域_10ms_envelope_correlation": 0.9930908534462566, "全帯域_10ms_modulation_mae_db": 0.2234162648047363, "全帯域_40ms_log_rms_mae_db": 0.5180282628781923, "全帯域_40ms_rms_envelope_nrmse": 0.04370551061501973, "全帯域_40ms_envelope_correlation": 0.9978611980427419, "全帯域_40ms_modulation_mae_db": 0.3926669918610615, "300–1000 Hz_10ms_log_rms_mae_db": 1.5383138459775607, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0916574198720311, "300–1000 Hz_10ms_envelope_correlation": 0.992831018878071, "300–1000 Hz_10ms_modulation_mae_db": 0.47519255450084774, "300–1000 Hz_40ms_log_rms_mae_db": 0.8363725617727628, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048963098640142776, "300–1000 Hz_40ms_envelope_correlation": 0.9978032592753174, "300–1000 Hz_40ms_modulation_mae_db": 0.30690785370613555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.450769470125865, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14333440860832217, "1000–3000 Hz_10ms_envelope_correlation": 0.9832652257651884, "1000–3000 Hz_10ms_modulation_mae_db": 0.4317531128970864, "1000–3000 Hz_40ms_log_rms_mae_db": 0.837045428163603, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08432651254498479, "1000–3000 Hz_40ms_envelope_correlation": 0.9942183045600601, "1000–3000 Hz_40ms_modulation_mae_db": 0.7523039697710515, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5443207805949828, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17661739114788574, "3000–6000 Hz_10ms_envelope_correlation": 0.9728853115982619, "3000–6000 Hz_10ms_modulation_mae_db": 0.7151298095927067, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9054519614401279, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11796207526334064, "3000–6000 Hz_40ms_envelope_correlation": 0.986268096284686, "3000–6000 Hz_40ms_modulation_mae_db": 0.8527434017056713, "mr_stft_db": 5.683322870080357, "band_mae_db": 0.14749220757639883, "rms_gain_error_db": -0.06338022692005069, "output_peak": 0.47296427541709024}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9317712231366486, "全帯域_10ms_rms_envelope_nrmse": 0.09263630282875109, "全帯域_10ms_envelope_correlation": 0.9911642707046634, "全帯域_10ms_modulation_mae_db": 0.07646583563759012, "全帯域_40ms_log_rms_mae_db": 0.510887186326845, "全帯域_40ms_rms_envelope_nrmse": 0.0520602203271033, "全帯域_40ms_envelope_correlation": 0.9968936850409884, "全帯域_40ms_modulation_mae_db": 0.25043041971630503, "300–1000 Hz_10ms_log_rms_mae_db": 1.575678947505682, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1154333684926704, "300–1000 Hz_10ms_envelope_correlation": 0.9886800746922926, "300–1000 Hz_10ms_modulation_mae_db": 1.4140148293919885, "300–1000 Hz_40ms_log_rms_mae_db": 0.8452343400649073, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06839051399086075, "300–1000 Hz_40ms_envelope_correlation": 0.9956805688274661, "300–1000 Hz_40ms_modulation_mae_db": 0.9074855718105436, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4969425379752876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14357567259874665, "1000–3000 Hz_10ms_envelope_correlation": 0.9831590793754819, "1000–3000 Hz_10ms_modulation_mae_db": 0.41055510357080016, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8139274832056788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.078762254435358, "1000–3000 Hz_40ms_envelope_correlation": 0.9950057258440007, "1000–3000 Hz_40ms_modulation_mae_db": 0.6308413876893365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5012999711702382, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19887280651436023, "3000–6000 Hz_10ms_envelope_correlation": 0.9644618200688005, "3000–6000 Hz_10ms_modulation_mae_db": 0.2504277795085446, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8868005628128326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1251815416813431, "3000–6000 Hz_40ms_envelope_correlation": 0.9837439901420425, "3000–6000 Hz_40ms_modulation_mae_db": 0.4955001671940839, "mr_stft_db": 5.682548526873027, "band_mae_db": 0.16126031362093227, "rms_gain_error_db": -0.030914721508889152, "output_peak": 0.4706291697126192}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9056628040787152, "全帯域_10ms_rms_envelope_nrmse": 0.07907547861492918, "全帯域_10ms_envelope_correlation": 0.9936111880429206, "全帯域_10ms_modulation_mae_db": 0.23804612824897411, "全帯域_40ms_log_rms_mae_db": 0.4570502173913438, "全帯域_40ms_rms_envelope_nrmse": 0.04143137914493523, "全帯域_40ms_envelope_correlation": 0.9981126632832852, "全帯域_40ms_modulation_mae_db": 0.5179379738466228, "300–1000 Hz_10ms_log_rms_mae_db": 1.5617545396365766, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09613969638034174, "300–1000 Hz_10ms_envelope_correlation": 0.9921211613381837, "300–1000 Hz_10ms_modulation_mae_db": 1.5892445179628092, "300–1000 Hz_40ms_log_rms_mae_db": 0.8278244276152931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053103411125286756, "300–1000 Hz_40ms_envelope_correlation": 0.997433243315111, "300–1000 Hz_40ms_modulation_mae_db": 1.0017467490709882, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4859671960893697, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14601705703968146, "1000–3000 Hz_10ms_envelope_correlation": 0.9824550761236072, "1000–3000 Hz_10ms_modulation_mae_db": 0.37071397074307527, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8456731582771605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0757455282092101, "1000–3000 Hz_40ms_envelope_correlation": 0.9952791844540118, "1000–3000 Hz_40ms_modulation_mae_db": 0.7369195812445923, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5203403414803378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1768425332058255, "3000–6000 Hz_10ms_envelope_correlation": 0.9717205614675842, "3000–6000 Hz_10ms_modulation_mae_db": 0.32516102591880974, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8892077308928563, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10977436698757793, "3000–6000 Hz_40ms_envelope_correlation": 0.9875301932851782, "3000–6000 Hz_40ms_modulation_mae_db": 0.49349796992535816, "mr_stft_db": 5.682096798026644, "band_mae_db": 0.08545058495107705, "rms_gain_error_db": -0.0806677347413928, "output_peak": 0.46123686553034265}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9528483497075271, "全帯域_10ms_rms_envelope_nrmse": 0.08186980616197498, "全帯域_10ms_envelope_correlation": 0.9931723135130388, "全帯域_10ms_modulation_mae_db": 0.33521999308970046, "全帯域_40ms_log_rms_mae_db": 0.5119912893225407, "全帯域_40ms_rms_envelope_nrmse": 0.04637713611661896, "全帯域_40ms_envelope_correlation": 0.9976566927700695, "全帯域_40ms_modulation_mae_db": 0.2111159933457738, "300–1000 Hz_10ms_log_rms_mae_db": 1.554998182728879, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10123679009912392, "300–1000 Hz_10ms_envelope_correlation": 0.9913493452114094, "300–1000 Hz_10ms_modulation_mae_db": 0.9812919664949181, "300–1000 Hz_40ms_log_rms_mae_db": 0.8998223911307904, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06232254857068314, "300–1000 Hz_40ms_envelope_correlation": 0.9965877546563537, "300–1000 Hz_40ms_modulation_mae_db": 0.6618191799476284, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4696838264737253, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13642029480629025, "1000–3000 Hz_10ms_envelope_correlation": 0.9845509636295758, "1000–3000 Hz_10ms_modulation_mae_db": 0.3038225320968945, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8476504768742882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.082228515770426, "1000–3000 Hz_40ms_envelope_correlation": 0.9939336171525023, "1000–3000 Hz_40ms_modulation_mae_db": 0.264302835412519, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5354715958837797, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17496898123737142, "3000–6000 Hz_10ms_envelope_correlation": 0.9726258596612063, "3000–6000 Hz_10ms_modulation_mae_db": 0.40675570558235413, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9426629736454735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11821557954019364, "3000–6000 Hz_40ms_envelope_correlation": 0.985639421499034, "3000–6000 Hz_40ms_modulation_mae_db": 0.3714655940447347, "mr_stft_db": 5.687653734169047, "band_mae_db": 0.13789002752229562, "rms_gain_error_db": -0.09741829003608185, "output_peak": 0.4523323893748546}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.729577203039871, "spectral_convergence": 0.309237797397153, "spectral_convergence_status": "ok", "amplitude_floor": 1.2198198453862723e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.476992507596816, "spectral_convergence": 0.30252977321265606, "spectral_convergence_status": "ok", "amplitude_floor": 7.927151445605076e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.84652724398089, "spectral_convergence": 0.3191559994776914, "spectral_convergence_status": "ok", "amplitude_floor": 6.454878545778745e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00013983470011737857, "output_power": 0.00014842124093899625, "delta_db": 0.25881102988734456, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042159782923140474, "output_power": 0.00041894784673977565, "delta_db": -0.027384034424323425, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017365691446010552, "output_power": 0.0017241055186331283, "delta_db": -0.03128238226631002, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006597865121791135, "output_power": 0.0006402159856346902, "delta_db": -0.130769192483207, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001856647778551521, "output_power": 0.00017770885912645823, "delta_db": -0.19020443541090282, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000145726779943178, "output_power": 0.0001441568528821305, "delta_db": -0.04704076038398051, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11424863914267806, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0472928028158083, "pitch": null, "output_rms_dbfs": -24.87641673243799, "output_peak": 0.48447264011670993, "reference_rms_dbfs": -24.82912392962218, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.684365651539192}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017345649672524565, "log_rms_mae_db": 0.9421548689077764, "rms_envelope_nrmse": 0.08186026543099417, "envelope_correlation": 0.9931134654454774, "rms_gain_error_db": -0.04746980437422288, "reference_mean_power": 0.003291322552458033, "output_mean_power": 0.003255543218814351}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4377088226614287e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006890430591645026, "output_power": 0.0006825091808939039, "reference_active": true, "delta_db": -0.04137864620507506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.716540294963627e-05, "output_power": 9.09017136768801e-05, "reference_active": true, "delta_db": -0.2893958544240006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.855953967229281e-05, "output_power": 1.8034922791415903e-05, "reference_active": true, "delta_db": -1.996369157076603}], "active_band_count": 3, "active_band_mae_db": 0.7757145525685596}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015649665907765232, "log_rms_mae_db": 0.4804512384901464, "rms_envelope_nrmse": 0.04026722225211556, "envelope_correlation": 0.9981646322956464, "rms_gain_error_db": -0.047607603660935, "reference_mean_power": 0.003302205038848876, "output_mean_power": 0.003266203767270803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.335462563779621e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377316706455328, "output_power": 0.0006341511250975862, "reference_active": true, "delta_db": -0.02445217887636355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.741968368909027e-05, "output_power": 4.346669165860425e-05, "reference_active": true, "delta_db": -0.37802066328085215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7622617238454117e-06, "output_power": 1.0245469866709613e-06, "reference_active": true, "delta_db": -2.355385281520255}], "active_band_count": 3, "active_band_mae_db": 0.9192860412258236}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014746435244218596, "log_rms_mae_db": 1.5364359674204748, "rms_envelope_nrmse": 0.0857926479665321, "envelope_correlation": 0.9937261122505927, "rms_gain_error_db": -0.0384422824590206, "reference_mean_power": 0.001906578877331925, "output_mean_power": 0.0018897769564611324}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.936060821906154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045087668804334914, "output_power": 0.00043906663473617416, "reference_active": true, "delta_db": -0.11527345553701501}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.461610811295139e-05, "output_power": 5.356095666007498e-05, "reference_active": true, "delta_db": -0.08472424126843064}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.1974528143292685e-06, "output_power": 8.479636376353948e-06, "reference_active": true, "delta_db": 1.3616400039038927}], "active_band_count": 3, "active_band_mae_db": 0.5205459002364461}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013246473438636124, "log_rms_mae_db": 0.8323093228998802, "rms_envelope_nrmse": 0.04299646087554328, "envelope_correlation": 0.9983145685008717, "rms_gain_error_db": -0.03850565824865285, "reference_mean_power": 0.0019129188838057705, "output_mean_power": 0.0018960334223145471}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.520895123968316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004268128119473447, "output_power": 0.00041664359472827993, "reference_active": true, "delta_db": -0.10472737892815812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.526792747799959e-05, "output_power": 2.471715431863043e-05, "reference_active": true, "delta_db": -0.09571152694493854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.02211562944459e-07, "output_power": 5.959486622854719e-07, "reference_active": true, "delta_db": 0.7432214154948502}], "active_band_count": 3, "active_band_mae_db": 0.31455344045598227}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.50803654710503e-05, "log_rms_mae_db": 1.4757849347048502, "rms_envelope_nrmse": 0.14921531175703176, "envelope_correlation": 0.9815203632064907, "rms_gain_error_db": -0.16674484982390453, "reference_mean_power": 0.0007310287661738595, "output_mean_power": 0.0007034933264043327}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3756975679714823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020778711952139343, "output_power": 0.00019691371710390846, "reference_active": true, "delta_db": -0.2334265229888981}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.101230284766401e-05, "output_power": 3.704658604112967e-05, "reference_active": true, "delta_db": -0.4416596311661837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.417981343795666e-05, "output_power": 7.611187825187723e-06, "reference_active": true, "delta_db": -2.702180775446795}], "active_band_count": 3, "active_band_mae_db": 1.1257556432006257}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.582207074769948e-05, "log_rms_mae_db": 0.8898865325805888, "rms_envelope_nrmse": 0.08353778160251471, "envelope_correlation": 0.993869529349745, "rms_gain_error_db": -0.16693070470820373, "reference_mean_power": 0.000733846607513954, "output_mean_power": 0.0007061748078021151}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9980296656839734e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019557456921819033, "output_power": 0.00018344378295782386, "reference_active": true, "delta_db": -0.2780938453352277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6985486820992048e-05, "output_power": 1.5771878475819525e-05, "reference_active": true, "delta_db": -0.32194576573438133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.92201473631864e-07, "output_power": 4.283408302813008e-07, "reference_active": true, "delta_db": -3.6481039355633085}], "active_band_count": 3, "active_band_mae_db": 1.4160478488776391}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011116745339746556, "log_rms_mae_db": 1.5543209476162059, "rms_envelope_nrmse": 0.18875756157656676, "envelope_correlation": 0.9678424705026432, "rms_gain_error_db": -0.2289317241845962, "reference_mean_power": 0.00017647199152424748, "output_mean_power": 0.00016741046951261195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.117031117126669e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.030239739448223e-05, "output_power": 3.9111189611597154e-05, "reference_active": true, "delta_db": -0.1302985526867506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8894394185439072e-05, "output_power": 2.2771025357317126e-05, "reference_active": true, "delta_db": -1.034310063416751}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.877164672472975e-06, "output_power": 5.451126706571423e-06, "reference_active": true, "delta_db": -1.5988364757706148}], "active_band_count": 3, "active_band_mae_db": 0.9211483639580388}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.815724618424621e-05, "log_rms_mae_db": 0.9238260023236716, "rms_envelope_nrmse": 0.11830012612432211, "envelope_correlation": 0.9860654960313446, "rms_gain_error_db": -0.22891011065016081, "reference_mean_power": 0.0001772156182174664, "output_mean_power": 0.00016811674896603652}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0001638026368737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.251019833717301e-05, "output_power": 4.0727210302903264e-05, "reference_active": true, "delta_db": -0.18608468054088015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.608059024163373e-05, "output_power": 1.1184110657072049e-05, "reference_active": true, "delta_db": -1.5770053019035641}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.447224316204941e-07, "output_power": 4.0298598105100183e-07, "reference_active": true, "delta_db": -2.040828424377428}], "active_band_count": 3, "active_band_mae_db": 1.2679728022739576}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8848421962962835, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "66e981a27f0f6675e4523b31099eb61624975d8616ba1b7a8923fa091ab847ff", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9831243380581468, "standard_deviation": 0.0015965068604091124, "minimum": 0.9814092083492761, "maximum": 0.9861850717748499, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4907173397405296, "standard_deviation": 0.037455370124348505, "minimum": 1.450769470125865, "maximum": 1.5624584380965527, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5264001032161676, "standard_deviation": 0.2589838596234817, "minimum": 0.3038225320968945, "maximum": 1.1257556432006257, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14300104708041453, "standard_deviation": 0.006877540132102083, "minimum": 0.1294816596083627, "maximum": 0.1498858685334793, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994557347430486, "standard_deviation": 0.0006163315275765308, "minimum": 0.993869529349745, "maximum": 0.9954469768824227, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8612326891799245, "standard_deviation": 0.03689143490943264, "minimum": 0.8139274832056788, "maximum": 0.9270609477481845, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.754929809269837, "standard_deviation": 0.3228454164525137, "minimum": 0.264302835412519, "maximum": 1.4160478488776391, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08008268040680494, "standard_deviation": 0.004140654129677152, "minimum": 0.07278031047207563, "maximum": 0.08432651254498479, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9684565592039246, "standard_deviation": 0.0038515826912555676, "minimum": 0.9629357349124924, "maximum": 0.9728853115982619, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5341348925511271, "standard_deviation": 0.02162047305108449, "minimum": 1.5012999711702382, "maximum": 1.5578533281281886, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6612701114655165, "standard_deviation": 0.4041560477889536, "minimum": 0.2504277795085446, "maximum": 1.2373113929677195, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1881358511992527, "standard_deviation": 0.01098090278293743, "minimum": 0.17496898123737142, "maximum": 0.20332287015000716, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9849297009088677, "standard_deviation": 0.0017510311203752186, "minimum": 0.9822022405831918, "maximum": 0.9875301932851782, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9168786536155857, "standard_deviation": 0.02149643254610987, "minimum": 0.8868005628128326, "maximum": 0.9426629736454735, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8050535003132463, "standard_deviation": 0.4837916827503164, "minimum": 0.19814429257401603, "maximum": 1.4519240764433314, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12259817577885558, "standard_deviation": 0.00840613932160922, "minimum": 0.10977436698757793, "maximum": 0.13771909687569414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991931803577604, "standard_deviation": 0.0015833075420534388, "minimum": 0.9886800746922926, "maximum": 0.9937261122505927, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5545482696320154, "standard_deviation": 0.016235166310853928, "minimum": 1.535764420874985, "maximum": 1.575678947505682, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0453708039060245, "standard_deviation": 0.3900568572212041, "minimum": 0.47519255450084774, "maximum": 1.5892445179628092, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09706923431224324, "standard_deviation": 0.009230904275190499, "minimum": 0.0857926479665321, "maximum": 0.1154333684926704, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972440805623838, "standard_deviation": 0.0008942453358205567, "minimum": 0.9956805688274661, "maximum": 0.9983145685008717, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8573675992313441, "standard_deviation": 0.029023407042040296, "minimum": 0.8278244276152931, "maximum": 0.8998223911307904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6191564268511656, "standard_deviation": 0.26544068426731365, "minimum": 0.30690785370613555, "maximum": 1.0017467490709882, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.054647980258987136, "standard_deviation": 0.008586073597714195, "minimum": 0.04299646087554328, "maximum": 0.06839051399086075, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.14264831133071543, "standard_deviation": 0.03044033432841097, "minimum": 0.08545058495107705, "maximum": 0.17705872445523604, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.686292715317709, "standard_deviation": 0.007582034236975522, "minimum": 5.677196587665204, "maximum": 5.702148985048559, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.46638757464069436, "standard_deviation": 0.010626165023852273, "minimum": 0.4523323893748546, "maximum": 0.48447264011670993, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06773859801855713, "standard_deviation": 0.03123403452011659, "minimum": -0.11052119419209208, "maximum": -0.025236781468994402, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9929597214874455, "standard_deviation": 0.0007955981301380352, "minimum": 0.9911642707046634, "maximum": 0.9937639072461653, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.93416512550875, "standard_deviation": 0.02776768714633205, "minimum": 0.8827343736805295, "maximum": 0.9676128600174929, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.30984723081878796, "standard_deviation": 0.21371237424339812, "minimum": 0.07646583563759012, "maximum": 0.7757145525685596, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08283228890781039, "standard_deviation": 0.004400837326771237, "minimum": 0.07818899994342368, "maximum": 0.09263630282875109, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978191841650856, "standard_deviation": 0.00042824250679403015, "minimum": 0.9968936850409884, "maximum": 0.9981985351381665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.49701062487951264, "standard_deviation": 0.02581495645949068, "minimum": 0.4570502173913438, "maximum": 0.5180282628781923, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4470110287748268, "standard_deviation": 0.285312950201916, "minimum": 0.13561693279734124, "maximum": 0.9192860412258236, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04413987754758467, "standard_deviation": 0.0038424439810455166, "minimum": 0.04026722225211556, "maximum": 0.0520602203271033, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0003", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.294252634048462, "predicted_mos": 2.294252634048462, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "46dc2811e9b32fc008d74f789e7b113ed2bb5174784cf88893ba1a9d54b7035e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.235983300022781, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9184713108564956, "全帯域_10ms_rms_envelope_nrmse": 0.07814774287088315, "全帯域_10ms_envelope_correlation": 0.9937726701804593, "全帯域_10ms_modulation_mae_db": 0.4938902166859959, "全帯域_40ms_log_rms_mae_db": 0.48773291471525704, "全帯域_40ms_rms_envelope_nrmse": 0.04613639567980992, "全帯域_40ms_envelope_correlation": 0.997632548252877, "全帯域_40ms_modulation_mae_db": 0.80969759370733, "300–1000 Hz_10ms_log_rms_mae_db": 1.5326831153131637, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08958486793614878, "300–1000 Hz_10ms_envelope_correlation": 0.9931557261656622, "300–1000 Hz_10ms_modulation_mae_db": 0.9376057517897088, "300–1000 Hz_40ms_log_rms_mae_db": 0.8605365229232562, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05636295221642846, "300–1000 Hz_40ms_envelope_correlation": 0.997076238541066, "300–1000 Hz_40ms_modulation_mae_db": 0.41123412888898364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4147532019629534, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1386319931852083, "1000–3000 Hz_10ms_envelope_correlation": 0.984543350198674, "1000–3000 Hz_10ms_modulation_mae_db": 1.1127063949454674, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7911063540858536, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08249299581244857, "1000–3000 Hz_40ms_envelope_correlation": 0.9946954358432123, "1000–3000 Hz_40ms_modulation_mae_db": 1.6096583835544351, "3000–6000 Hz_10ms_log_rms_mae_db": 1.520927791268521, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1815974114845873, "3000–6000 Hz_10ms_envelope_correlation": 0.9704184952358426, "3000–6000 Hz_10ms_modulation_mae_db": 0.44042834033062844, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8609270063181492, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1087148327035663, "3000–6000 Hz_40ms_envelope_correlation": 0.9877628627284633, "3000–6000 Hz_40ms_modulation_mae_db": 0.7188886503520852, "mr_stft_db": 5.653146484820219, "band_mae_db": 0.15641156101811457, "rms_gain_error_db": -0.0716901460833532, "output_peak": 0.45476422804121286}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9392109951143277, "全帯域_10ms_rms_envelope_nrmse": 0.0890655229681863, "全帯域_10ms_envelope_correlation": 0.9918367032479934, "全帯域_10ms_modulation_mae_db": 0.07369039700086216, "全帯域_40ms_log_rms_mae_db": 0.484077669963611, "全帯域_40ms_rms_envelope_nrmse": 0.04774011529345797, "全帯域_40ms_envelope_correlation": 0.9974104822752076, "全帯域_40ms_modulation_mae_db": 0.36908923555766054, "300–1000 Hz_10ms_log_rms_mae_db": 1.5637694292512594, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10482774246369926, "300–1000 Hz_10ms_envelope_correlation": 0.9906316488666478, "300–1000 Hz_10ms_modulation_mae_db": 1.6070515790631739, "300–1000 Hz_40ms_log_rms_mae_db": 0.8707594148690331, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05747328126328102, "300–1000 Hz_40ms_envelope_correlation": 0.9969638138672332, "300–1000 Hz_40ms_modulation_mae_db": 0.9273702848499163, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4605230693423417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15101229019610304, "1000–3000 Hz_10ms_envelope_correlation": 0.9810233805240551, "1000–3000 Hz_10ms_modulation_mae_db": 0.3593314634267292, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8032471836416892, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08352114955291663, "1000–3000 Hz_40ms_envelope_correlation": 0.9937480635515583, "1000–3000 Hz_40ms_modulation_mae_db": 0.38587447238816885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5171874151238236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19568221868763463, "3000–6000 Hz_10ms_envelope_correlation": 0.9655375361564368, "3000–6000 Hz_10ms_modulation_mae_db": 0.30026778369814017, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9028963729216041, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11549110712783367, "3000–6000 Hz_40ms_envelope_correlation": 0.986145557547913, "3000–6000 Hz_40ms_modulation_mae_db": 0.4071029194963635, "mr_stft_db": 5.664275036910077, "band_mae_db": 0.10640773779060585, "rms_gain_error_db": -0.05439756382319333, "output_peak": 0.4783740915256224}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9472524971786636, "全帯域_10ms_rms_envelope_nrmse": 0.08418729985583412, "全帯域_10ms_envelope_correlation": 0.9927439624528257, "全帯域_10ms_modulation_mae_db": 0.1550799323257394, "全帯域_40ms_log_rms_mae_db": 0.4846349648350883, "全帯域_40ms_rms_envelope_nrmse": 0.04007947044276624, "全帯域_40ms_envelope_correlation": 0.9982444205161014, "全帯域_40ms_modulation_mae_db": 0.36267730117490266, "300–1000 Hz_10ms_log_rms_mae_db": 1.5410597187523065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10001724326956236, "300–1000 Hz_10ms_envelope_correlation": 0.9914584486876836, "300–1000 Hz_10ms_modulation_mae_db": 0.48482641959992306, "300–1000 Hz_40ms_log_rms_mae_db": 0.8705671212159958, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05156348677624652, "300–1000 Hz_40ms_envelope_correlation": 0.9975681215962754, "300–1000 Hz_40ms_modulation_mae_db": 0.32912849083497114, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4851402210689315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12984867435312705, "1000–3000 Hz_10ms_envelope_correlation": 0.986434098223823, "1000–3000 Hz_10ms_modulation_mae_db": 0.32423652313400114, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8325715517003147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06802307236184105, "1000–3000 Hz_40ms_envelope_correlation": 0.9965691534816669, "1000–3000 Hz_40ms_modulation_mae_db": 0.6352701556548263, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5417326184764715, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19642037267628845, "3000–6000 Hz_10ms_envelope_correlation": 0.966202346720989, "3000–6000 Hz_10ms_modulation_mae_db": 0.4906249434522887, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9398877096515026, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12579437630094656, "3000–6000 Hz_40ms_envelope_correlation": 0.9842465432909981, "3000–6000 Hz_40ms_modulation_mae_db": 0.44756589282035647, "mr_stft_db": 5.665469751898562, "band_mae_db": 0.16360809114659572, "rms_gain_error_db": -0.07786797142058696, "output_peak": 0.44733847696121865}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9479063791954714, "全帯域_10ms_rms_envelope_nrmse": 0.0832857354015628, "全帯域_10ms_envelope_correlation": 0.9929485616642608, "全帯域_10ms_modulation_mae_db": 0.23388078248059493, "全帯域_40ms_log_rms_mae_db": 0.4907223204492765, "全帯域_40ms_rms_envelope_nrmse": 0.043394807086053554, "全帯域_40ms_envelope_correlation": 0.9979921457729184, "全帯域_40ms_modulation_mae_db": 0.21594846140816903, "300–1000 Hz_10ms_log_rms_mae_db": 1.517738610684842, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09177254613154684, "300–1000 Hz_10ms_envelope_correlation": 0.9928696910717265, "300–1000 Hz_10ms_modulation_mae_db": 1.3475453881415056, "300–1000 Hz_40ms_log_rms_mae_db": 0.8015869109693979, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04701033495952985, "300–1000 Hz_40ms_envelope_correlation": 0.9980766188768724, "300–1000 Hz_40ms_modulation_mae_db": 0.7325387715376689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4707985406208914, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1498590660389874, "1000–3000 Hz_10ms_envelope_correlation": 0.9817032285419353, "1000–3000 Hz_10ms_modulation_mae_db": 0.38451004255662863, "1000–3000 Hz_40ms_log_rms_mae_db": 0.848387587371201, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09367127510910622, "1000–3000 Hz_40ms_envelope_correlation": 0.9927598339563004, "1000–3000 Hz_40ms_modulation_mae_db": 0.8035936906290188, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5170126882026034, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18330862993042926, "3000–6000 Hz_10ms_envelope_correlation": 0.9706190951475117, "3000–6000 Hz_10ms_modulation_mae_db": 0.7487412516752293, "3000–6000 Hz_40ms_log_rms_mae_db": 0.926081015897651, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12765994463806796, "3000–6000 Hz_40ms_envelope_correlation": 0.983818409706019, "3000–6000 Hz_40ms_modulation_mae_db": 0.796571558956515, "mr_stft_db": 5.66390565569736, "band_mae_db": 0.1377742766444446, "rms_gain_error_db": -0.10595815068581221, "output_peak": 0.4415157741283311}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9538051134089521, "全帯域_10ms_rms_envelope_nrmse": 0.08515557442430036, "全帯域_10ms_envelope_correlation": 0.992555933524193, "全帯域_10ms_modulation_mae_db": 0.30129014549594535, "全帯域_40ms_log_rms_mae_db": 0.4924619412628829, "全帯域_40ms_rms_envelope_nrmse": 0.041525337645079874, "全帯域_40ms_envelope_correlation": 0.9980830736500578, "全帯域_40ms_modulation_mae_db": 0.48780057719272607, "300–1000 Hz_10ms_log_rms_mae_db": 1.5994582477081056, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09927444023927996, "300–1000 Hz_10ms_envelope_correlation": 0.9915865783518847, "300–1000 Hz_10ms_modulation_mae_db": 1.5643084759050794, "300–1000 Hz_40ms_log_rms_mae_db": 0.8671310404732919, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050110584580363965, "300–1000 Hz_40ms_envelope_correlation": 0.9976910008813069, "300–1000 Hz_40ms_modulation_mae_db": 1.0202875058679575, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4641306538022292, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14492670615368164, "1000–3000 Hz_10ms_envelope_correlation": 0.9825559356269936, "1000–3000 Hz_10ms_modulation_mae_db": 0.44676155850412796, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8350382045147918, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08806357528934881, "1000–3000 Hz_40ms_envelope_correlation": 0.9930777894917119, "1000–3000 Hz_40ms_modulation_mae_db": 0.7737249769422802, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5631997225528103, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20162865189324328, "3000–6000 Hz_10ms_envelope_correlation": 0.9638865710868397, "3000–6000 Hz_10ms_modulation_mae_db": 1.3486886201477961, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9128253869417483, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13594539649785486, "3000–6000 Hz_40ms_envelope_correlation": 0.9825778889916705, "3000–6000 Hz_40ms_modulation_mae_db": 1.7474004779733887, "mr_stft_db": 5.678115385399988, "band_mae_db": 0.1715228970397292, "rms_gain_error_db": -0.07104507920811862, "output_peak": 0.4562527372505318}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9423616933589645, "全帯域_10ms_rms_envelope_nrmse": 0.08867646322627752, "全帯域_10ms_envelope_correlation": 0.9919480634026169, "全帯域_10ms_modulation_mae_db": 0.3058992066290282, "全帯域_40ms_log_rms_mae_db": 0.5037413469588354, "全帯域_40ms_rms_envelope_nrmse": 0.04734256299058991, "全帯域_40ms_envelope_correlation": 0.9975205593158041, "全帯域_40ms_modulation_mae_db": 0.6932768199567562, "300–1000 Hz_10ms_log_rms_mae_db": 1.5200227590291595, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10721229970536904, "300–1000 Hz_10ms_envelope_correlation": 0.9901762711279949, "300–1000 Hz_10ms_modulation_mae_db": 1.615601295734085, "300–1000 Hz_40ms_log_rms_mae_db": 0.8821498358957874, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.059531063407397895, "300–1000 Hz_40ms_envelope_correlation": 0.9967525823827159, "300–1000 Hz_40ms_modulation_mae_db": 0.9047208709958808, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4858403962215827, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15820085580371113, "1000–3000 Hz_10ms_envelope_correlation": 0.9793404491617633, "1000–3000 Hz_10ms_modulation_mae_db": 0.5094947488503471, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8494774565750052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08656050279279968, "1000–3000 Hz_40ms_envelope_correlation": 0.9937272918426813, "1000–3000 Hz_40ms_modulation_mae_db": 0.9612601992267812, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5359621690550462, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18443867945274015, "3000–6000 Hz_10ms_envelope_correlation": 0.9692252714091677, "3000–6000 Hz_10ms_modulation_mae_db": 1.0036221411947344, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8719877034026263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11437046113279849, "3000–6000 Hz_40ms_envelope_correlation": 0.9867240678058303, "3000–6000 Hz_40ms_modulation_mae_db": 1.3358107868404605, "mr_stft_db": 5.6605155153924125, "band_mae_db": 0.14929682769620126, "rms_gain_error_db": -0.08452513159587464, "output_peak": 0.4636137161763562}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9350170093289912, "全帯域_10ms_rms_envelope_nrmse": 0.08902205015779491, "全帯域_10ms_envelope_correlation": 0.9918715978081153, "全帯域_10ms_modulation_mae_db": 0.49458190675077335, "全帯域_40ms_log_rms_mae_db": 0.5051505172289953, "全帯域_40ms_rms_envelope_nrmse": 0.047129513144466194, "全帯域_40ms_envelope_correlation": 0.9975306794474819, "全帯域_40ms_modulation_mae_db": 0.6549970830239933, "300–1000 Hz_10ms_log_rms_mae_db": 1.5554691297618344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0969940177056708, "300–1000 Hz_10ms_envelope_correlation": 0.9919809796817696, "300–1000 Hz_10ms_modulation_mae_db": 1.1252868917661047, "300–1000 Hz_40ms_log_rms_mae_db": 0.8958973433921613, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056013459262481205, "300–1000 Hz_40ms_envelope_correlation": 0.9971368393397908, "300–1000 Hz_40ms_modulation_mae_db": 0.8363898670907214, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4599262769952293, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1464293739396753, "1000–3000 Hz_10ms_envelope_correlation": 0.9823300068208705, "1000–3000 Hz_10ms_modulation_mae_db": 0.789418726820888, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8118837160563842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07659451220639368, "1000–3000 Hz_40ms_envelope_correlation": 0.9951153971318483, "1000–3000 Hz_40ms_modulation_mae_db": 0.9936685798978236, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5406714509637633, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20165117872956437, "3000–6000 Hz_10ms_envelope_correlation": 0.9631465647179583, "3000–6000 Hz_10ms_modulation_mae_db": 1.087909688319387, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9267306726115395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13433767258481352, "3000–6000 Hz_40ms_envelope_correlation": 0.9817794262758603, "3000–6000 Hz_40ms_modulation_mae_db": 1.4796023027999814, "mr_stft_db": 5.683454978621739, "band_mae_db": 0.19311470143598375, "rms_gain_error_db": -0.08337068548525318, "output_peak": 0.46050684170848155}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9009792767465645, "全帯域_10ms_rms_envelope_nrmse": 0.08861060957886992, "全帯域_10ms_envelope_correlation": 0.9919354903412557, "全帯域_10ms_modulation_mae_db": 0.5970982301966047, "全帯域_40ms_log_rms_mae_db": 0.475212878418843, "全帯域_40ms_rms_envelope_nrmse": 0.04795595564924287, "全帯域_40ms_envelope_correlation": 0.9973635649985926, "全帯域_40ms_modulation_mae_db": 0.3743544036193622, "300–1000 Hz_10ms_log_rms_mae_db": 1.5695040538974951, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0989725598833652, "300–1000 Hz_10ms_envelope_correlation": 0.9916507570320583, "300–1000 Hz_10ms_modulation_mae_db": 1.6307417600332066, "300–1000 Hz_40ms_log_rms_mae_db": 0.8387644254499547, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05001569264607454, "300–1000 Hz_40ms_envelope_correlation": 0.9976812023938406, "300–1000 Hz_40ms_modulation_mae_db": 0.9071753455998101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4979420817233053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14289332093311397, "1000–3000 Hz_10ms_envelope_correlation": 0.983023164647129, "1000–3000 Hz_10ms_modulation_mae_db": 0.5775517850808488, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8037394466584455, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.078594184276586, "1000–3000 Hz_40ms_envelope_correlation": 0.994459692373335, "1000–3000 Hz_40ms_modulation_mae_db": 0.801914519606966, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5321922481011943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20836132586196632, "3000–6000 Hz_10ms_envelope_correlation": 0.9644652907123464, "3000–6000 Hz_10ms_modulation_mae_db": 1.1218547562931664, "3000–6000 Hz_40ms_log_rms_mae_db": 0.882387289362573, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14150141242450895, "3000–6000 Hz_40ms_envelope_correlation": 0.9823029165273921, "3000–6000 Hz_40ms_modulation_mae_db": 1.3329720360760495, "mr_stft_db": 5.659066361659581, "band_mae_db": 0.19842118917024773, "rms_gain_error_db": 0.005227417963872614, "output_peak": 0.45673720265755385}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.728265631472044, "spectral_convergence": 0.3049348151190286, "spectral_convergence_status": "ok", "amplitude_floor": 1.2198198453862723e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.426368467180578, "spectral_convergence": 0.2957231697115501, "spectral_convergence_status": "ok", "amplitude_floor": 7.927151445605076e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.804805355808034, "spectral_convergence": 0.3202029271138393, "spectral_convergence_status": "ok", "amplitude_floor": 6.454878545778745e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00013983470011737857, "output_power": 0.00014802810287444998, "delta_db": 0.24729217881121224, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042159782923140474, "output_power": 0.00042746240477922184, "delta_db": 0.059995585756123154, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017365691446010552, "output_power": 0.0017063526922355822, "delta_db": -0.07623278187596122, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006597865121791135, "output_power": 0.0006134853496090463, "delta_db": -0.3159923716697946, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001856647778551521, "output_power": 0.00018770732179298604, "delta_db": 0.04751690933672002, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000145726779943178, "output_power": 0.00015229418547163347, "delta_db": 0.19143953865887606, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15641156101811457, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0716901460833532, "pitch": null, "output_rms_dbfs": -24.900814075705537, "output_peak": 0.45476422804121286, "reference_rms_dbfs": -24.82912392962218, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.653146484820219}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017345649672524565, "log_rms_mae_db": 0.9184713108564956, "rms_envelope_nrmse": 0.07814774287088315, "envelope_correlation": 0.9937726701804593, "rms_gain_error_db": -0.07146667592864207, "reference_mean_power": 0.003291322552458033, "output_mean_power": 0.0032376043733059034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4377088226614287e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006890430591645026, "output_power": 0.0006506080569145018, "reference_active": true, "delta_db": -0.24926925224198007}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.716540294963627e-05, "output_power": 9.423989268589622e-05, "reference_active": true, "delta_db": -0.13276873152758734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.855953967229281e-05, "output_power": 2.2171135516173026e-05, "reference_active": true, "delta_db": -1.0996326662884204}], "active_band_count": 3, "active_band_mae_db": 0.4938902166859959}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015649665907765232, "log_rms_mae_db": 0.48773291471525704, "rms_envelope_nrmse": 0.04613639567980992, "envelope_correlation": 0.997632548252877, "rms_gain_error_db": -0.07086769605209572, "reference_mean_power": 0.003302205038848876, "output_mean_power": 0.003248757283371237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.335462563779621e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377316706455328, "output_power": 0.0006060883582419454, "reference_active": true, "delta_db": -0.2210204304090166}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.741968368909027e-05, "output_power": 4.4869160121308296e-05, "reference_active": true, "delta_db": -0.2401071239641852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7622617238454117e-06, "output_power": 1.1201440744731573e-06, "reference_active": true, "delta_db": -1.9679652267487882}], "active_band_count": 3, "active_band_mae_db": 0.80969759370733}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014746435244218596, "log_rms_mae_db": 1.5326831153131637, "rms_envelope_nrmse": 0.08958486793614878, "envelope_correlation": 0.9931557261656622, "rms_gain_error_db": -0.05236343395906652, "reference_mean_power": 0.001906578877331925, "output_mean_power": 0.001883729043590727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.936060821906154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045087668804334914, "output_power": 0.0004305454500707746, "reference_active": true, "delta_db": -0.2003877724747133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.461610811295139e-05, "output_power": 5.170298693063934e-05, "reference_active": true, "delta_db": -0.23805116110090563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.1974528143292685e-06, "output_power": 1.0706587263266384e-05, "reference_active": true, "delta_db": 2.374378321793508}], "active_band_count": 3, "active_band_mae_db": 0.9376057517897088}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013246473438636124, "log_rms_mae_db": 0.8605365229232562, "rms_envelope_nrmse": 0.05636295221642846, "envelope_correlation": 0.997076238541066, "rms_gain_error_db": -0.051832248344698226, "reference_mean_power": 0.0019129188838057705, "output_mean_power": 0.0018902242460006268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.520895123968316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004268128119473447, "output_power": 0.0004104157648138136, "reference_active": true, "delta_db": -0.1701341301531314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.526792747799959e-05, "output_power": 2.4163522361422203e-05, "reference_active": true, "delta_db": -0.19409379271519284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.02211562944459e-07, "output_power": 6.135276699625495e-07, "reference_active": true, "delta_db": 0.8694744637986267}], "active_band_count": 3, "active_band_mae_db": 0.41123412888898364}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.50803654710503e-05, "log_rms_mae_db": 1.4147532019629534, "rms_envelope_nrmse": 0.1386319931852083, "envelope_correlation": 0.984543350198674, "rms_gain_error_db": -0.2952286918222176, "reference_mean_power": 0.0007310287661738595, "output_mean_power": 0.0006829856795076953}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3756975679714823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020778711952139343, "output_power": 0.00018182568599265498, "reference_active": true, "delta_db": -0.5796338789806345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.101230284766401e-05, "output_power": 3.292160789812046e-05, "reference_active": true, "delta_db": -0.9543311758963133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.417981343795666e-05, "output_power": 9.359552914501412e-06, "reference_active": true, "delta_db": -1.8041541299594548}], "active_band_count": 3, "active_band_mae_db": 1.1127063949454674}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.582207074769948e-05, "log_rms_mae_db": 0.7911063540858536, "rms_envelope_nrmse": 0.08249299581244857, "envelope_correlation": 0.9946954358432123, "rms_gain_error_db": -0.29529463411603657, "reference_mean_power": 0.000733846607513954, "output_mean_power": 0.0006856079225973313}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9980296656839734e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019557456921819033, "output_power": 0.00016913272923820084, "reference_active": true, "delta_db": -0.6308472533196468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6985486820992048e-05, "output_power": 1.3744660417076277e-05, "reference_active": true, "delta_db": -0.91943984296872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.92201473631864e-07, "output_power": 4.6637049002768685e-07, "reference_active": true, "delta_db": -3.2786880543749386}], "active_band_count": 3, "active_band_mae_db": 1.6096583835544351}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011116745339746556, "log_rms_mae_db": 1.520927791268521, "rms_envelope_nrmse": 0.1815974114845873, "envelope_correlation": 0.9704184952358426, "rms_gain_error_db": 0.00443473903944201, "reference_mean_power": 0.00017647199152424748, "output_mean_power": 0.00017665228553515814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.117031117126669e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.030239739448223e-05, "output_power": 4.3045226083115965e-05, "reference_active": true, "delta_db": 0.28594112185782505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8894394185439072e-05, "output_power": 2.7593093662625095e-05, "reference_active": true, "delta_db": -0.20013198133628957}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.877164672472975e-06, "output_power": 6.499032858573032e-06, "reference_active": true, "delta_db": -0.8352119177977708}], "active_band_count": 3, "active_band_mae_db": 0.44042834033062844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.815724618424621e-05, "log_rms_mae_db": 0.8609270063181492, "rms_envelope_nrmse": 0.1087148327035663, "envelope_correlation": 0.9877628627284633, "rms_gain_error_db": 0.004361053477145832, "reference_mean_power": 0.0001772156182174664, "output_mean_power": 0.00017739366214499067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0001638026368737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.251019833717301e-05, "output_power": 4.5802636125357975e-05, "reference_active": true, "delta_db": 0.3239734287951252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.608059024163373e-05, "output_power": 1.4350300321320982e-05, "reference_active": true, "delta_db": -0.49440995553696043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.447224316204941e-07, "output_power": 4.7374487422742655e-07, "reference_active": true, "delta_db": -1.3382825667241696}], "active_band_count": 3, "active_band_mae_db": 0.7188886503520852}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8838122581758697, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c4a78758b0c47f3948712a477a69a3d910d93359f8f565873d5a1bd7890bfd9c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9826192017181554, "standard_deviation": 0.0021594468712437306, "minimum": 0.9793404491617633, "maximum": 0.986434098223823, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.467381805217183, "standard_deviation": 0.02531135409050422, "minimum": 1.4147532019629534, "maximum": 1.4979420817233053, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5630014054148798, "standard_deviation": 0.2673617233256214, "minimum": 0.32423652313400114, "maximum": 1.1127063949454674, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14522528507545096, "standard_deviation": 0.00853744258621995, "minimum": 0.12984867435312705, "maximum": 0.15820085580371113, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942690822090392, "standard_deviation": 0.0012239685064293028, "minimum": 0.9927598339563004, "maximum": 0.9965691534816669, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8219314375754607, "standard_deviation": 0.022280365319466994, "minimum": 0.7911063540858536, "maximum": 0.8494774565750052, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8706206222375374, "standard_deviation": 0.35426054204389795, "minimum": 0.38587447238816885, "maximum": 1.6096583835544351, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08219015842518007, "standard_deviation": 0.00785788810174014, "minimum": 0.06802307236184105, "maximum": 0.09367127510910622, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9666876463983866, "standard_deviation": 0.0029927884637074393, "minimum": 0.9631465647179583, "maximum": 0.9706190951475117, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5336107629680291, "standard_deviation": 0.01560652742402257, "minimum": 1.5170126882026034, "maximum": 1.5631997225528103, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8177671906389213, "standard_deviation": 0.3786899821143939, "minimum": 0.30026778369814017, "maximum": 1.3486886201477961, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1941360585895567, "standard_deviation": 0.00993603083094935, "minimum": 0.1815974114845873, "maximum": 0.20836132586196632, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9844197091092683, "standard_deviation": 0.0022264979211679874, "minimum": 0.9817794262758603, "maximum": 0.9877628627284633, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9029653946384242, "standard_deviation": 0.028560471697520608, "minimum": 0.8609270063181492, "maximum": 0.9398877096515026, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0332393281644001, "standard_deviation": 0.5043444675403238, "minimum": 0.4071029194963635, "maximum": 1.7474004779733887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1254769004262988, "standard_deviation": 0.011672863077050287, "minimum": 0.1087148327035663, "maximum": 0.14150141242450895, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916887626231785, "standard_deviation": 0.0010072604586760835, "minimum": 0.9901762711279949, "maximum": 0.9931557261656622, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5499631330497707, "standard_deviation": 0.027685115687693652, "minimum": 1.517738610684842, "maximum": 1.5994582477081056, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.2891209452540984, "standard_deviation": 0.41417903187639843, "minimum": 0.48482641959992306, "maximum": 1.6307417600332066, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09858196466683028, "standard_deviation": 0.005928406685439284, "minimum": 0.08958486793614878, "maximum": 0.10721229970536904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973683022348876, "standard_deviation": 0.0004512331069023227, "minimum": 0.9967525823827159, "maximum": 0.9980766188768724, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8609240768986098, "standard_deviation": 0.029066451940025557, "minimum": 0.8015869109693979, "maximum": 0.8958973433921613, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7586056582082388, "standard_deviation": 0.2541734442767239, "minimum": 0.32912849083497114, "maximum": 1.0202875058679575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05351010688897543, "standard_deviation": 0.004410510432268083, "minimum": 0.04701033495952985, "maximum": 0.059531063407397895, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.15956966024274033, "standard_deviation": 0.029786462186217794, "minimum": 0.10640773779060585, "maximum": 0.19842118917024773, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.665993646299992, "standard_deviation": 0.010017621776590447, "minimum": 5.653146484820219, "maximum": 5.683454978621739, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.45738788355616355, "standard_deviation": 0.01102218385371513, "minimum": 0.4415157741283311, "maximum": 0.4783740915256224, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06795341379228995, "standard_deviation": 0.03298850307281073, "minimum": -0.10595815068581221, "maximum": 0.005227417963872614, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9924516228277149, "standard_deviation": 0.0006889927891329386, "minimum": 0.9918367032479934, "maximum": 0.9937726701804593, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9356255343985538, "standard_deviation": 0.01760574065823355, "minimum": 0.9009792767465645, "maximum": 0.9538051134089521, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.331926352195693, "standard_deviation": 0.18212814785790138, "minimum": 0.07369039700086216, "maximum": 0.5970982301966047, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08576887481046364, "standard_deviation": 0.003874932550037005, "minimum": 0.07814774287088315, "maximum": 0.0890655229681863, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9977221842786301, "standard_deviation": 0.00033529907856138483, "minimum": 0.9973635649985926, "maximum": 0.9982444205161014, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4904668192290987, "standard_deviation": 0.010077262697678288, "minimum": 0.475212878418843, "maximum": 0.5051505172289953, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4959801844551125, "standard_deviation": 0.20346267587976613, "minimum": 0.21594846140816903, "maximum": 0.80969759370733, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045163019741433316, "standard_deviation": 0.0030754479062099407, "minimum": 0.04007947044276624, "maximum": 0.04795595564924287, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8463-294828-0014", "language": "en", "style": "normal", "speaker": "librispeech_8463", "source_utterance_id": "8463-294828-0014", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 91600, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8463-294828-0014.wav", "source_file_sha256": "19262df0247f7490d6eb26c12b01eb091436ae2b13b8e24c14e533b83686c986", "decoded_pcm_sha256": "c608aedb889e9a3c3821520a449396c5b4cd1264e1d081a38be40b0b66f8b285", "original_source_sha256": "b61f11aa0c2b3d1adca49e6199328637c6ecc84c003dab03fc8c97b95cab4c63", "corpus_entry_id": "librispeech_8463-294828-0014"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.6556341648101807, "predicted_mos": 3.6556341648101807, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "1703405f5a666ca2ab8bcb794a4ce99e1c5f5977c71ded9ac683e4496e61c71b", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.323729218001972, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8463-294828-0014", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "1703405f5a666ca2ab8bcb794a4ce99e1c5f5977c71ded9ac683e4496e61c71b", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "57c0273c8992ab87674c0b415b90f57ef6597f45937ac6ef94c308d8b0bb1aaf"}, "rows": [{"case": "librispeech_8463-294828-0014", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.779000997543335, "predicted_mos": 2.779000997543335, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "84d67632246b72cc92be30d6e584cde59deb74a4dd77c6e248fa28114aac8cb7", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.4331756020001194, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8463-294828-0014", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "84d67632246b72cc92be30d6e584cde59deb74a4dd77c6e248fa28114aac8cb7", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8237900961393593, "全帯域_10ms_rms_envelope_nrmse": 0.06888436242421275, "全帯域_10ms_envelope_correlation": 0.9957085257137671, "全帯域_10ms_modulation_mae_db": 0.45234008315555085, "全帯域_40ms_log_rms_mae_db": 0.43050556065231926, "全帯域_40ms_rms_envelope_nrmse": 0.0359898067213902, "全帯域_40ms_envelope_correlation": 0.998812439375806, "全帯域_40ms_modulation_mae_db": 0.59272385414067, "300–1000 Hz_10ms_log_rms_mae_db": 1.4150836034582555, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08293628116813814, "300–1000 Hz_10ms_envelope_correlation": 0.9948957654627206, "300–1000 Hz_10ms_modulation_mae_db": 0.38038317686634543, "300–1000 Hz_40ms_log_rms_mae_db": 0.7453870349180723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046100758605174444, "300–1000 Hz_40ms_envelope_correlation": 0.9983848067058712, "300–1000 Hz_40ms_modulation_mae_db": 0.6188784522672263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4573658140410073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11385277290697875, "1000–3000 Hz_10ms_envelope_correlation": 0.9901679045277655, "1000–3000 Hz_10ms_modulation_mae_db": 0.3325916880985152, "1000–3000 Hz_40ms_log_rms_mae_db": 0.894299473259644, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06803937545741119, "1000–3000 Hz_40ms_envelope_correlation": 0.996768654140322, "1000–3000 Hz_40ms_modulation_mae_db": 0.47333743665160055, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4229431611047743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1513560796802832, "3000–6000 Hz_10ms_envelope_correlation": 0.979853332693383, "3000–6000 Hz_10ms_modulation_mae_db": 0.34152209795073013, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8410108967641878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0855493020406172, "3000–6000 Hz_40ms_envelope_correlation": 0.9928784526455626, "3000–6000 Hz_40ms_modulation_mae_db": 0.28609279272105853, "mr_stft_db": 5.504106923737197, "band_mae_db": 0.20013708784363593, "rms_gain_error_db": -0.0924017384091841, "output_peak": 0.5359342638480773}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8585053822333484, "全帯域_10ms_rms_envelope_nrmse": 0.08022559484601191, "全帯域_10ms_envelope_correlation": 0.9941486312460386, "全帯域_10ms_modulation_mae_db": 0.28655787675791694, "全帯域_40ms_log_rms_mae_db": 0.470630410125582, "全帯域_40ms_rms_envelope_nrmse": 0.04058727468631933, "全帯域_40ms_envelope_correlation": 0.9984849616570962, "全帯域_40ms_modulation_mae_db": 0.6764301628274193, "300–1000 Hz_10ms_log_rms_mae_db": 1.398326585623096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08652644935805151, "300–1000 Hz_10ms_envelope_correlation": 0.9944760768997156, "300–1000 Hz_10ms_modulation_mae_db": 0.2055341702746709, "300–1000 Hz_40ms_log_rms_mae_db": 0.7560658000189573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043058836005729906, "300–1000 Hz_40ms_envelope_correlation": 0.9986704796988846, "300–1000 Hz_40ms_modulation_mae_db": 0.5809677916036016, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4655204510368525, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12293375904448713, "1000–3000 Hz_10ms_envelope_correlation": 0.9879587826148614, "1000–3000 Hz_10ms_modulation_mae_db": 0.30971788170067277, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8842449899959085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06757686608680782, "1000–3000 Hz_40ms_envelope_correlation": 0.9961657822075004, "1000–3000 Hz_40ms_modulation_mae_db": 0.23046327896230254, "3000–6000 Hz_10ms_log_rms_mae_db": 1.401723058913467, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15515482944580294, "3000–6000 Hz_10ms_envelope_correlation": 0.9787050836218599, "3000–6000 Hz_10ms_modulation_mae_db": 0.21368282556542348, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8149505323466718, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08851924699692726, "3000–6000 Hz_40ms_envelope_correlation": 0.9924268141822365, "3000–6000 Hz_40ms_modulation_mae_db": 0.16766198730432477, "mr_stft_db": 5.49964613295406, "band_mae_db": 0.1734077564409824, "rms_gain_error_db": -0.09174457907787545, "output_peak": 0.5135357278801437}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8289824066949407, "全帯域_10ms_rms_envelope_nrmse": 0.06932135992620075, "全帯域_10ms_envelope_correlation": 0.9956106649899017, "全帯域_10ms_modulation_mae_db": 0.3316366204596584, "全帯域_40ms_log_rms_mae_db": 0.41111980001284903, "全帯域_40ms_rms_envelope_nrmse": 0.03305092130222472, "全帯域_40ms_envelope_correlation": 0.9989719137598088, "全帯域_40ms_modulation_mae_db": 0.4369270445516937, "300–1000 Hz_10ms_log_rms_mae_db": 1.427819140326855, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07586453558260649, "300–1000 Hz_10ms_envelope_correlation": 0.9957069467705606, "300–1000 Hz_10ms_modulation_mae_db": 0.2785580901561153, "300–1000 Hz_40ms_log_rms_mae_db": 0.7680635055005873, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0375806988281093, "300–1000 Hz_40ms_envelope_correlation": 0.9989122342496434, "300–1000 Hz_40ms_modulation_mae_db": 0.41682253195824653, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4198470485242818, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11022442302076041, "1000–3000 Hz_10ms_envelope_correlation": 0.9903142811260187, "1000–3000 Hz_10ms_modulation_mae_db": 0.18659852330703006, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8270683556221259, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06233014888906299, "1000–3000 Hz_40ms_envelope_correlation": 0.9966767601234489, "1000–3000 Hz_40ms_modulation_mae_db": 0.22632864674237543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4226282980881138, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17382662457470602, "3000–6000 Hz_10ms_envelope_correlation": 0.9742170803551066, "3000–6000 Hz_10ms_modulation_mae_db": 0.6587561073602555, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8491677541599, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11073163751494165, "3000–6000 Hz_40ms_envelope_correlation": 0.9897190214225818, "3000–6000 Hz_40ms_modulation_mae_db": 0.597632755093745, "mr_stft_db": 5.520790289270141, "band_mae_db": 0.19323819738284395, "rms_gain_error_db": -0.075835293295488, "output_peak": 0.5532965916133972}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8318657085955763, "全帯域_10ms_rms_envelope_nrmse": 0.07115816306186497, "全帯域_10ms_envelope_correlation": 0.9954435693319146, "全帯域_10ms_modulation_mae_db": 0.3581081683198944, "全帯域_40ms_log_rms_mae_db": 0.4317521342113571, "全帯域_40ms_rms_envelope_nrmse": 0.04275190644768448, "全帯域_40ms_envelope_correlation": 0.9983124506861156, "全帯域_40ms_modulation_mae_db": 0.26349766072105063, "300–1000 Hz_10ms_log_rms_mae_db": 1.4442460474065215, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08675884362083625, "300–1000 Hz_10ms_envelope_correlation": 0.9943954484486007, "300–1000 Hz_10ms_modulation_mae_db": 0.21664309187160855, "300–1000 Hz_40ms_log_rms_mae_db": 0.8331000195316165, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0570251279820191, "300–1000 Hz_40ms_envelope_correlation": 0.9974654451830003, "300–1000 Hz_40ms_modulation_mae_db": 0.0755148472165542, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3923633896102794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1099430242216707, "1000–3000 Hz_10ms_envelope_correlation": 0.9905951324556742, "1000–3000 Hz_10ms_modulation_mae_db": 0.5464854484292128, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8191204764186341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06098894076766291, "1000–3000 Hz_40ms_envelope_correlation": 0.9971916578717342, "1000–3000 Hz_40ms_modulation_mae_db": 0.6331359595407277, "3000–6000 Hz_10ms_log_rms_mae_db": 1.427402327888486, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1511255031313956, "3000–6000 Hz_10ms_envelope_correlation": 0.9802046676224055, "3000–6000 Hz_10ms_modulation_mae_db": 0.8218749355375671, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8258729207013055, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08158572295925022, "3000–6000 Hz_40ms_envelope_correlation": 0.9943106405163198, "3000–6000 Hz_40ms_modulation_mae_db": 0.75776686619129, "mr_stft_db": 5.502222308266614, "band_mae_db": 0.2140937909966373, "rms_gain_error_db": -0.10478574469587026, "output_peak": 0.5431993947408312}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.839613179157479, "全帯域_10ms_rms_envelope_nrmse": 0.06799493185533038, "全帯域_10ms_envelope_correlation": 0.995824735956942, "全帯域_10ms_modulation_mae_db": 0.3050800647655119, "全帯域_40ms_log_rms_mae_db": 0.43924066813969265, "全帯域_40ms_rms_envelope_nrmse": 0.038504621750471604, "全帯域_40ms_envelope_correlation": 0.9986239348888423, "全帯域_40ms_modulation_mae_db": 0.45004556190763784, "300–1000 Hz_10ms_log_rms_mae_db": 1.4399099623191727, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07997148586409943, "300–1000 Hz_10ms_envelope_correlation": 0.9952483700193469, "300–1000 Hz_10ms_modulation_mae_db": 0.4421206310315304, "300–1000 Hz_40ms_log_rms_mae_db": 0.8227636649606538, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0461762162205874, "300–1000 Hz_40ms_envelope_correlation": 0.9983691394752114, "300–1000 Hz_40ms_modulation_mae_db": 0.3351019070306244, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4430288933464508, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10158574417489127, "1000–3000 Hz_10ms_envelope_correlation": 0.9918718939924533, "1000–3000 Hz_10ms_modulation_mae_db": 0.21108830371721873, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8553286742195785, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04960306150733273, "1000–3000 Hz_40ms_envelope_correlation": 0.9980797137721392, "1000–3000 Hz_40ms_modulation_mae_db": 0.193109384286536, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3870404301884005, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15837576757292235, "3000–6000 Hz_10ms_envelope_correlation": 0.9779473553082607, "3000–6000 Hz_10ms_modulation_mae_db": 0.7846074316230007, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8247706791449041, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09554348520798119, "3000–6000 Hz_40ms_envelope_correlation": 0.9914280857986877, "3000–6000 Hz_40ms_modulation_mae_db": 0.5750830536722124, "mr_stft_db": 5.497353042534024, "band_mae_db": 0.2072380906840489, "rms_gain_error_db": -0.08761339566243839, "output_peak": 0.5376177493418046}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8364990959780512, "全帯域_10ms_rms_envelope_nrmse": 0.06802231065835879, "全帯域_10ms_envelope_correlation": 0.9957661260483189, "全帯域_10ms_modulation_mae_db": 0.3209167060162984, "全帯域_40ms_log_rms_mae_db": 0.4217500236712957, "全帯域_40ms_rms_envelope_nrmse": 0.03539013422740402, "全帯域_40ms_envelope_correlation": 0.9987849281508989, "全帯域_40ms_modulation_mae_db": 0.3475662974402369, "300–1000 Hz_10ms_log_rms_mae_db": 1.4205733881397644, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07908305205290558, "300–1000 Hz_10ms_envelope_correlation": 0.9953311399271847, "300–1000 Hz_10ms_modulation_mae_db": 0.21627309973525902, "300–1000 Hz_40ms_log_rms_mae_db": 0.7637152147468114, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045146760047435376, "300–1000 Hz_40ms_envelope_correlation": 0.9983970919052869, "300–1000 Hz_40ms_modulation_mae_db": 0.2469202796216392, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3896232683050291, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11569533227392682, "1000–3000 Hz_10ms_envelope_correlation": 0.9893928963045252, "1000–3000 Hz_10ms_modulation_mae_db": 0.5632890038467558, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8257630331136945, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06608327659829946, "1000–3000 Hz_40ms_envelope_correlation": 0.996410321274258, "1000–3000 Hz_40ms_modulation_mae_db": 0.6507176556234064, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4413434541158023, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14825095035428615, "3000–6000 Hz_10ms_envelope_correlation": 0.9806214704984451, "3000–6000 Hz_10ms_modulation_mae_db": 0.45963656513933704, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8230499734349573, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07962651615325078, "3000–6000 Hz_40ms_envelope_correlation": 0.9939964730165315, "3000–6000 Hz_40ms_modulation_mae_db": 0.4068918358666371, "mr_stft_db": 5.506318765569234, "band_mae_db": 0.14421193732589246, "rms_gain_error_db": -0.056536748954784374, "output_peak": 0.5321994200032626}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8260152779736865, "全帯域_10ms_rms_envelope_nrmse": 0.06388553106166886, "全帯域_10ms_envelope_correlation": 0.9964712744393446, "全帯域_10ms_modulation_mae_db": 0.4209578987720095, "全帯域_40ms_log_rms_mae_db": 0.4363645724408269, "全帯域_40ms_rms_envelope_nrmse": 0.035629572249840064, "全帯域_40ms_envelope_correlation": 0.9990073516439455, "全帯域_40ms_modulation_mae_db": 0.452653016040306, "300–1000 Hz_10ms_log_rms_mae_db": 1.4810342175059328, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07419896325047776, "300–1000 Hz_10ms_envelope_correlation": 0.9960494175959789, "300–1000 Hz_10ms_modulation_mae_db": 0.3430545607816078, "300–1000 Hz_40ms_log_rms_mae_db": 0.811953722450833, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043662271224743164, "300–1000 Hz_40ms_envelope_correlation": 0.9987055416273669, "300–1000 Hz_40ms_modulation_mae_db": 0.6245555658680244, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4685620733690559, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12204433883106795, "1000–3000 Hz_10ms_envelope_correlation": 0.9882095009461417, "1000–3000 Hz_10ms_modulation_mae_db": 0.33898810463589973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8783641455684934, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07219161879606943, "1000–3000 Hz_40ms_envelope_correlation": 0.9957483570141343, "1000–3000 Hz_40ms_modulation_mae_db": 0.4333801606400259, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4409735310487621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15858524489632955, "3000–6000 Hz_10ms_envelope_correlation": 0.9783651802945114, "3000–6000 Hz_10ms_modulation_mae_db": 0.7843888601860604, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8702157730888198, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08790571351010412, "3000–6000 Hz_40ms_envelope_correlation": 0.9936412589886194, "3000–6000 Hz_40ms_modulation_mae_db": 0.7966163875627079, "mr_stft_db": 5.517295339798316, "band_mae_db": 0.21354228926534738, "rms_gain_error_db": -0.14618344019297697, "output_peak": 0.5194902684032043}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8356344544774593, "全帯域_10ms_rms_envelope_nrmse": 0.06872970224215928, "全帯域_10ms_envelope_correlation": 0.9957114664148313, "全帯域_10ms_modulation_mae_db": 0.10607161447006304, "全帯域_40ms_log_rms_mae_db": 0.4480521323673224, "全帯域_40ms_rms_envelope_nrmse": 0.03592351001237135, "全帯域_40ms_envelope_correlation": 0.9988034095895469, "全帯域_40ms_modulation_mae_db": 0.2734700574024291, "300–1000 Hz_10ms_log_rms_mae_db": 1.479651286626603, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07873979854663955, "300–1000 Hz_10ms_envelope_correlation": 0.9954090463000581, "300–1000 Hz_10ms_modulation_mae_db": 0.18326228362570549, "300–1000 Hz_40ms_log_rms_mae_db": 0.7986652206978488, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04510159720443142, "300–1000 Hz_40ms_envelope_correlation": 0.9984663263797467, "300–1000 Hz_40ms_modulation_mae_db": 0.19556638215526814, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4417222833941346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1147602126402464, "1000–3000 Hz_10ms_envelope_correlation": 0.9895161633536128, "1000–3000 Hz_10ms_modulation_mae_db": 0.2522026938661674, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8433073116890702, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06523747065594278, "1000–3000 Hz_40ms_envelope_correlation": 0.9964191935169834, "1000–3000 Hz_40ms_modulation_mae_db": 0.2071606420503841, "3000–6000 Hz_10ms_log_rms_mae_db": 1.444438643849094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15837395283482536, "3000–6000 Hz_10ms_envelope_correlation": 0.9787748075021915, "3000–6000 Hz_10ms_modulation_mae_db": 1.0377197404499505, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8469751562365447, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08924234938892757, "3000–6000 Hz_40ms_envelope_correlation": 0.9938709901766398, "3000–6000 Hz_40ms_modulation_mae_db": 1.16356222734519, "mr_stft_db": 5.49752328771697, "band_mae_db": 0.22128328874524472, "rms_gain_error_db": -0.08804591026966027, "output_peak": 0.5304003862172045}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 91600, "output_samples": 91600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5892472665166295, "spectral_convergence": 0.2597526402976283, "spectral_convergence_status": "ok", "amplitude_floor": 1.4112644553101518e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.326628354183193, "spectral_convergence": 0.2529959314783747, "spectral_convergence_status": "ok", "amplitude_floor": 1.1975531593072457e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5964451505117685, "spectral_convergence": 0.26107468430814323, "spectral_convergence_status": "ok", "amplitude_floor": 1.114415465542996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.564075826256522e-05, "output_power": 6.275969326866997e-05, "delta_db": 0.5228777250205187, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010000129923479963, "output_power": 0.0009781109475367417, "delta_db": -0.09617522680659038, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018259392170112694, "output_power": 0.0018036109432331311, "delta_db": -0.05343454694978731, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005127219920465177, "output_power": 0.0004889503819131302, "delta_db": -0.20617156229663106, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00033320627903786135, "output_power": 0.00031247891983837245, "delta_db": -0.2789245208701837, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002817706196064759, "output_power": 0.00027897919223767326, "delta_db": -0.04323894511810428, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20013708784363593, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0924017384091841, "pitch": null, "output_rms_dbfs": -24.061725017657317, "output_peak": 0.5359342638480773, "reference_rms_dbfs": -23.96932327924813, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.504106923737197}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 91600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00020825955187282386, "log_rms_mae_db": 0.8237900961393593, "rms_envelope_nrmse": 0.06888436242421275, "envelope_correlation": 0.9957085257137671, "rms_gain_error_db": -0.09238576483120148, "reference_mean_power": 0.004012756910556895, "output_mean_power": 0.003928296634029046}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 2.2120307480210933e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008889321025104637, "output_power": 0.0008513607347526479, "reference_active": true, "delta_db": -0.18754974015085438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022980725896418495, "output_power": 0.00021126421836042194, "reference_active": true, "delta_db": -0.3653779551212929}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1131932747773025e-05, "output_power": 3.4179827110353887e-05, "reference_active": true, "delta_db": -0.8040925541945054}], "active_band_count": 3, "active_band_mae_db": 0.45234008315555085}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.0001984160011662782, "log_rms_mae_db": 0.43050556065231926, "rms_envelope_nrmse": 0.0359898067213902, "envelope_correlation": 0.998812439375806, "rms_gain_error_db": -0.09240959189156253, "reference_mean_power": 0.0040337408532874226, "output_mean_power": 0.003948817243155761}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 2.0115882799069226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008024963383362849, "output_power": 0.0007707041557857889, "reference_active": true, "delta_db": -0.17555358672360938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011463438169489824, "output_power": 0.00010312321705571267, "reference_active": true, "delta_db": -0.45958439950042196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4945953602736586e-06, "output_power": 2.6859228774569513e-06, "reference_active": true, "delta_db": -1.1430335761979784}], "active_band_count": 3, "active_band_mae_db": 0.59272385414067}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00019054595180557928, "log_rms_mae_db": 1.4150836034582555, "rms_envelope_nrmse": 0.08293628116813814, "envelope_correlation": 0.9948957654627206, "rms_gain_error_db": -0.08383850690991496, "reference_mean_power": 0.0025757888112863028, "output_mean_power": 0.0025265412992953157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.7595886306573375e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007279305516452219, "output_power": 0.000709155034250302, "reference_active": true, "delta_db": -0.11348757067165444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019010480993234813, "output_power": 0.00016904663804806613, "reference_active": true, "delta_db": -0.5098656723469376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6294378423488108e-05, "output_power": 2.333905568322953e-05, "reference_active": true, "delta_db": -0.5177962875804442}], "active_band_count": 3, "active_band_mae_db": 0.38038317686634543}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.00018019065505981728, "log_rms_mae_db": 0.7453870349180723, "rms_envelope_nrmse": 0.046100758605174444, "envelope_correlation": 0.9983848067058712, "rms_gain_error_db": -0.08383460460145711, "reference_mean_power": 0.0025893611436675294, "output_mean_power": 0.0025398564191153462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.6269545971301712e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006780414214466183, "output_power": 0.0006614477545447039, "reference_active": true, "delta_db": -0.10760679191810102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.548437806715927e-05, "output_power": 8.331108664742104e-05, "reference_active": true, "delta_db": -0.5922952456755906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9327510117604047e-06, "output_power": 2.246993734135949e-06, "reference_active": true, "delta_db": -1.1567333192079872}], "active_band_count": 3, "active_band_mae_db": 0.6188784522672263}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.327720993461315e-05, "log_rms_mae_db": 1.4573658140410073, "rms_envelope_nrmse": 0.11385277290697875, "envelope_correlation": 0.9901679045277655, "rms_gain_error_db": -0.2765198314461534, "reference_mean_power": 0.0007122971869413385, "output_mean_power": 0.0006683581466462924}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 4.689927289528067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023362430343497353, "output_power": 0.00021757022474449192, "reference_active": true, "delta_db": -0.309185591581464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.797902128989837e-05, "output_power": 3.573481644847603e-05, "reference_active": true, "delta_db": -0.26452213188791873}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.042184778585592e-05, "output_power": 9.452310770181072e-06, "reference_active": true, "delta_db": -0.4240673408261628}], "active_band_count": 3, "active_band_mae_db": 0.3325916880985152}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 8.747046012456969e-05, "log_rms_mae_db": 0.894299473259644, "rms_envelope_nrmse": 0.06803937545741119, "envelope_correlation": 0.996768654140322, "rms_gain_error_db": -0.2766581268520318, "reference_mean_power": 0.0007159674763911086, "output_mean_power": 0.0006717806368811551}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 4.328159646639031e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021700970165722394, "output_power": 0.00020282821935151292, "reference_active": true, "delta_db": -0.2935077193527849}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.702508733600319e-05, "output_power": 1.6779541346764924e-05, "reference_active": true, "delta_db": -0.06309262835336381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.679451250427792e-07, "output_power": 6.794399388468699e-07, "reference_active": true, "delta_db": -1.0634119622486529}], "active_band_count": 3, "active_band_mae_db": 0.47333743665160055}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.039773028293347e-05, "log_rms_mae_db": 1.4229431611047743, "rms_envelope_nrmse": 0.1513560796802832, "envelope_correlation": 0.979853332693383, "rms_gain_error_db": -0.0483838549027785, "reference_mean_power": 0.0003237352885604671, "output_mean_power": 0.00032014863647525984}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.2480235329901466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.70846061617512e-05, "output_power": 5.673725945488424e-05, "reference_active": true, "delta_db": -0.02650654609467537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9595191589145483e-05, "output_power": 1.9934174213103893e-05, "reference_active": true, "delta_db": 0.07448735336793884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.074204859780537e-06, "output_power": 4.1021426387187786e-06, "reference_active": true, "delta_db": -0.9235723943895762}], "active_band_count": 3, "active_band_mae_db": 0.34152209795073013}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 6.399293414479695e-05, "log_rms_mae_db": 0.8410108967641878, "rms_envelope_nrmse": 0.0855493020406172, "envelope_correlation": 0.9928784526455626, "rms_gain_error_db": -0.048460191489175464, "reference_mean_power": 0.0003253818281247376, "output_mean_power": 0.00032177127821326785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.1005868407421181e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4497636322434906e-05, "output_power": 5.4437068491289674e-05, "reference_active": true, "delta_db": -0.004829365922044011}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.385752117634814e-06, "output_power": 9.51205405596293e-06, "reference_active": true, "delta_db": 0.058052295702680334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7591737704275347e-07, "output_power": 3.130060486405888e-07, "reference_active": true, "delta_db": -0.7953967165384512}], "active_band_count": 3, "active_band_mae_db": 0.28609279272105853}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.877223232083865, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "53a915e5c8e9175102e4ea57d0351e497867abe83fdd26308a5569383cae632b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897533194151316, "standard_deviation": 0.0012802595663432286, "minimum": 0.9879587826148614, "maximum": 0.9918718939924533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4347541527033862, "standard_deviation": 0.031107760392731387, "minimum": 1.3896232683050291, "maximum": 1.4685620733690559, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.34262020595018405, "standard_deviation": 0.1420617647256273, "minimum": 0.18659852330703006, "maximum": 0.5632890038467558, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11387995088925368, "standard_deviation": 0.006894197102944556, "minimum": 0.10158574417489127, "maximum": 0.12293375904448713, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966825549900651, "standard_deviation": 0.000707302267084837, "minimum": 0.9957483570141343, "maximum": 0.9980797137721392, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8534370574858936, "standard_deviation": 0.029240879436502974, "minimum": 0.8191204764186341, "maximum": 0.894299473259644, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.38095414556216983, "standard_deviation": 0.19261269807139003, "minimum": 0.193109384286536, "maximum": 0.6507176556234064, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06400634484482366, "standard_deviation": 0.006774014582388552, "minimum": 0.04960306150733273, "maximum": 0.07219161879606943, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9785861222370205, "standard_deviation": 0.0019986879084639597, "minimum": 0.9742170803551066, "maximum": 0.9806214704984451, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4235616131496125, "standard_deviation": 0.020290543432676278, "minimum": 1.3870404301884005, "maximum": 1.444438643849094, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6377735704765406, "standard_deviation": 0.2768953984725944, "minimum": 0.21368282556542348, "maximum": 1.0377197404499505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1568811190613189, "standard_deviation": 0.007888778571558931, "minimum": 0.14825095035428615, "maximum": 0.17382662457470602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927839670933973, "standard_deviation": 0.0015623362864140113, "minimum": 0.9897190214225818, "maximum": 0.9943106405163198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8370017107346615, "standard_deviation": 0.018209429589497842, "minimum": 0.8149505323466718, "maximum": 0.8702157730888198, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5939134882196457, "standard_deviation": 0.3171807776854657, "minimum": 0.16766198730432477, "maximum": 1.16356222734519, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0898379967215, "standard_deviation": 0.009751862087192243, "minimum": 0.07962651615325078, "maximum": 0.11073163751494165, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9951890264280208, "standard_deviation": 0.0005737976408834634, "minimum": 0.9943954484486007, "maximum": 0.9960494175959789, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.438330528925775, "standard_deviation": 0.029604628901309955, "minimum": 1.398326585623096, "maximum": 1.4810342175059328, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.28322863804285536, "standard_deviation": 0.09504360375656429, "minimum": 0.18326228362570549, "maximum": 0.4421206310315304, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08050992618046934, "standard_deviation": 0.004598165245116519, "minimum": 0.07419896325047776, "maximum": 0.08675884362083625, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984213831531265, "standard_deviation": 0.0004316454441169831, "minimum": 0.9974654451830003, "maximum": 0.9989122342496434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7874642728531726, "standard_deviation": 0.033275105271162814, "minimum": 0.7453870349180723, "maximum": 0.8331000195316165, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.38679096971514815, "standard_deviation": 0.2086344623307225, "minimum": 0.0755148472165542, "maximum": 0.6245555658680244, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04548153326477876, "standard_deviation": 0.005424539336849066, "minimum": 0.0375806988281093, "maximum": 0.0570251279820191, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19589405483557912, "standard_deviation": 0.025684132723653885, "minimum": 0.14421193732589246, "maximum": 0.22128328874524472, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.50565701123082, "standard_deviation": 0.008866861922994714, "minimum": 5.497353042534024, "maximum": 5.520790289270141, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5332092252559908, "standard_deviation": 0.012617064708313482, "minimum": 0.5135357278801437, "maximum": 0.5532965916133972, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09289335631978474, "standard_deviation": 0.025765368364771953, "minimum": -0.14618344019297697, "maximum": -0.056536748954784374, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955856242676324, "standard_deviation": 0.0006536521702344925, "minimum": 0.9941486312460386, "maximum": 0.9964712744393446, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8351132001562376, "standard_deviation": 0.010880875156209228, "minimum": 0.8237900961393593, "maximum": 0.8585053822333484, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32270862908961295, "standard_deviation": 0.1044321512820931, "minimum": 0.10607161447006304, "maximum": 0.45234008315555085, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06977774450947596, "standard_deviation": 0.00469264000242607, "minimum": 0.06388553106166886, "maximum": 0.08022559484601191, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987251737190075, "standard_deviation": 0.0002378784818177689, "minimum": 0.9983124506861156, "maximum": 0.9990073516439455, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.43617691270265563, "standard_deviation": 0.017829442193661175, "minimum": 0.41111980001284903, "maximum": 0.470630410125582, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4366642068789304, "standard_deviation": 0.144832178497218, "minimum": 0.26349766072105063, "maximum": 0.6764301628274193, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.037228468424713224, "standard_deviation": 0.0031632613197753114, "minimum": 0.03305092130222472, "maximum": 0.04275190644768448, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0014", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.475177764892578, "predicted_mos": 2.475177764892578, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "0531a557033c790ff53d69224fbb452816b1e1f76e2f63d59ade6e3654c13677", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9535029996186495, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8387719200385727, "全帯域_10ms_rms_envelope_nrmse": 0.06928352846813671, "全帯域_10ms_envelope_correlation": 0.9956557897827184, "全帯域_10ms_modulation_mae_db": 0.3827181758291262, "全帯域_40ms_log_rms_mae_db": 0.43229366017215914, "全帯域_40ms_rms_envelope_nrmse": 0.035505776341888604, "全帯域_40ms_envelope_correlation": 0.9988465830131633, "全帯域_40ms_modulation_mae_db": 0.5266486562780334, "300–1000 Hz_10ms_log_rms_mae_db": 1.4214141342046833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08255539697789854, "300–1000 Hz_10ms_envelope_correlation": 0.9949332728275956, "300–1000 Hz_10ms_modulation_mae_db": 0.29942673388996094, "300–1000 Hz_40ms_log_rms_mae_db": 0.7215655501665419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04459439659231298, "300–1000 Hz_40ms_envelope_correlation": 0.9984820569917022, "300–1000 Hz_40ms_modulation_mae_db": 0.5319645095776028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.461238711755153, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11392679353123143, "1000–3000 Hz_10ms_envelope_correlation": 0.9901272303284759, "1000–3000 Hz_10ms_modulation_mae_db": 0.3078851185032709, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8670612204311741, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06687483464307632, "1000–3000 Hz_40ms_envelope_correlation": 0.9968743881888192, "1000–3000 Hz_40ms_modulation_mae_db": 0.5090706905928122, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4443047639719677, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14904423971136105, "3000–6000 Hz_10ms_envelope_correlation": 0.9804579686724403, "3000–6000 Hz_10ms_modulation_mae_db": 0.34695880614572344, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8687471729830932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08499265848782823, "3000–6000 Hz_40ms_envelope_correlation": 0.993032482069486, "3000–6000 Hz_40ms_modulation_mae_db": 0.31506678507131664, "mr_stft_db": 5.46430149997295, "band_mae_db": 0.19799921168973997, "rms_gain_error_db": -0.09076923036205559, "output_peak": 0.5361557004650915}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8636919166146789, "全帯域_10ms_rms_envelope_nrmse": 0.08272771173257883, "全帯域_10ms_envelope_correlation": 0.9937500065935589, "全帯域_10ms_modulation_mae_db": 0.23818116176000223, "全帯域_40ms_log_rms_mae_db": 0.47803072195684987, "全帯域_40ms_rms_envelope_nrmse": 0.04194789431773518, "全帯域_40ms_envelope_correlation": 0.9983470499110161, "全帯域_40ms_modulation_mae_db": 0.6067150950019481, "300–1000 Hz_10ms_log_rms_mae_db": 1.3972690516269601, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08992150922017869, "300–1000 Hz_10ms_envelope_correlation": 0.9939905903177912, "300–1000 Hz_10ms_modulation_mae_db": 0.17441425057493856, "300–1000 Hz_40ms_log_rms_mae_db": 0.7620557147494076, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04676167226009155, "300–1000 Hz_40ms_envelope_correlation": 0.9983553003895077, "300–1000 Hz_40ms_modulation_mae_db": 0.5042547698911001, "1000–3000 Hz_10ms_log_rms_mae_db": 1.471146884127763, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12278608117922439, "1000–3000 Hz_10ms_envelope_correlation": 0.9880200610357713, "1000–3000 Hz_10ms_modulation_mae_db": 0.2711143691281323, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8726463315421389, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0697244203796671, "1000–3000 Hz_40ms_envelope_correlation": 0.9959655841908029, "1000–3000 Hz_40ms_modulation_mae_db": 0.269169992628924, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4090934810613938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15636905921569483, "3000–6000 Hz_10ms_envelope_correlation": 0.9783827000271853, "3000–6000 Hz_10ms_modulation_mae_db": 0.3575797723160914, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8088575129615801, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08947951287738105, "3000–6000 Hz_40ms_envelope_correlation": 0.992284410353375, "3000–6000 Hz_40ms_modulation_mae_db": 0.3279805035606564, "mr_stft_db": 5.455504567003186, "band_mae_db": 0.17960052917220826, "rms_gain_error_db": -0.08233091246065126, "output_peak": 0.5114691255884032}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8372970596712898, "全帯域_10ms_rms_envelope_nrmse": 0.06956961204005488, "全帯域_10ms_envelope_correlation": 0.9955788908453856, "全帯域_10ms_modulation_mae_db": 0.38433537226262754, "全帯域_40ms_log_rms_mae_db": 0.417591626068676, "全帯域_40ms_rms_envelope_nrmse": 0.033456392792029575, "全帯域_40ms_envelope_correlation": 0.9989432849254556, "全帯域_40ms_modulation_mae_db": 0.5015116704591901, "300–1000 Hz_10ms_log_rms_mae_db": 1.4508088456173833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07627488256138959, "300–1000 Hz_10ms_envelope_correlation": 0.9956538570058138, "300–1000 Hz_10ms_modulation_mae_db": 0.3386876028183237, "300–1000 Hz_40ms_log_rms_mae_db": 0.7528218866618509, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038587372478063896, "300–1000 Hz_40ms_envelope_correlation": 0.998839760493837, "300–1000 Hz_40ms_modulation_mae_db": 0.4780911529742588, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4445804217153189, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1138101477712268, "1000–3000 Hz_10ms_envelope_correlation": 0.9896753167165496, "1000–3000 Hz_10ms_modulation_mae_db": 0.12718480505850135, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8099918384496752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06448834073276676, "1000–3000 Hz_40ms_envelope_correlation": 0.9964639998849651, "1000–3000 Hz_40ms_modulation_mae_db": 0.30142948821089416, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4516923194464597, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1678603427778609, "3000–6000 Hz_10ms_envelope_correlation": 0.9760118736436085, "3000–6000 Hz_10ms_modulation_mae_db": 0.6868978122879169, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8599974734603286, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10468037307869807, "3000–6000 Hz_40ms_envelope_correlation": 0.9909346616057547, "3000–6000 Hz_40ms_modulation_mae_db": 0.6088947640626896, "mr_stft_db": 5.484244416283601, "band_mae_db": 0.1924062803557931, "rms_gain_error_db": -0.07492639930991879, "output_peak": 0.5521351471988357}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.81983778399532, "全帯域_10ms_rms_envelope_nrmse": 0.07082238109113809, "全帯域_10ms_envelope_correlation": 0.9954847417566658, "全帯域_10ms_modulation_mae_db": 0.2761816000898419, "全帯域_40ms_log_rms_mae_db": 0.4249490648604041, "全帯域_40ms_rms_envelope_nrmse": 0.041397125796346405, "全帯域_40ms_envelope_correlation": 0.9984262390268298, "全帯域_40ms_modulation_mae_db": 0.20707994204018365, "300–1000 Hz_10ms_log_rms_mae_db": 1.4312371750937602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08548295124596147, "300–1000 Hz_10ms_envelope_correlation": 0.9945570030633936, "300–1000 Hz_10ms_modulation_mae_db": 0.14627452158574747, "300–1000 Hz_40ms_log_rms_mae_db": 0.8075790606788783, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05445172531955506, "300–1000 Hz_40ms_envelope_correlation": 0.9976906073786133, "300–1000 Hz_40ms_modulation_mae_db": 0.07217106136784975, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4165052435349295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11239631887749395, "1000–3000 Hz_10ms_envelope_correlation": 0.9902179899110366, "1000–3000 Hz_10ms_modulation_mae_db": 0.47696252830846647, "1000–3000 Hz_40ms_log_rms_mae_db": 0.788573813215356, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06184288505961243, "1000–3000 Hz_40ms_envelope_correlation": 0.9971937595607983, "1000–3000 Hz_40ms_modulation_mae_db": 0.5977382851896901, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4332082091075748, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15059388372563695, "3000–6000 Hz_10ms_envelope_correlation": 0.9802941764901685, "3000–6000 Hz_10ms_modulation_mae_db": 0.7729827519157687, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8325222825153735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.081696692701147, "3000–6000 Hz_40ms_envelope_correlation": 0.9941908085840617, "3000–6000 Hz_40ms_modulation_mae_db": 0.796988288156841, "mr_stft_db": 5.470909388520301, "band_mae_db": 0.21678766578512465, "rms_gain_error_db": -0.10573060518262081, "output_peak": 0.5422626436134419}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8393972368414045, "全帯域_10ms_rms_envelope_nrmse": 0.0659590744672181, "全帯域_10ms_envelope_correlation": 0.9960870790947174, "全帯域_10ms_modulation_mae_db": 0.35588811873029286, "全帯域_40ms_log_rms_mae_db": 0.44276282938454614, "全帯域_40ms_rms_envelope_nrmse": 0.03696530789095805, "全帯域_40ms_envelope_correlation": 0.9987485672833315, "全帯域_40ms_modulation_mae_db": 0.5161076242414958, "300–1000 Hz_10ms_log_rms_mae_db": 1.4547131686796044, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07487436804606207, "300–1000 Hz_10ms_envelope_correlation": 0.9958410822948613, "300–1000 Hz_10ms_modulation_mae_db": 0.33563968411698064, "300–1000 Hz_40ms_log_rms_mae_db": 0.7892188329698091, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04479192605521803, "300–1000 Hz_40ms_envelope_correlation": 0.9984605951893168, "300–1000 Hz_40ms_modulation_mae_db": 0.4192176527765438, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4684108082162626, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10195632135165765, "1000–3000 Hz_10ms_envelope_correlation": 0.9918725820421245, "1000–3000 Hz_10ms_modulation_mae_db": 0.38400141704888174, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8313497274499716, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04919637083463172, "1000–3000 Hz_40ms_envelope_correlation": 0.9982008484604632, "1000–3000 Hz_40ms_modulation_mae_db": 0.4704351814900021, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4022636499722079, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1509872273862681, "3000–6000 Hz_10ms_envelope_correlation": 0.9800436046980022, "3000–6000 Hz_10ms_modulation_mae_db": 0.8591779893232986, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8413549208464829, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09059206529163508, "3000–6000 Hz_40ms_envelope_correlation": 0.992402247785291, "3000–6000 Hz_40ms_modulation_mae_db": 0.6810167244843001, "mr_stft_db": 5.4647094999889125, "band_mae_db": 0.21294720983494722, "rms_gain_error_db": -0.09010159363839339, "output_peak": 0.537891730823039}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8447480116129256, "全帯域_10ms_rms_envelope_nrmse": 0.07073174797375559, "全帯域_10ms_envelope_correlation": 0.995420602887112, "全帯域_10ms_modulation_mae_db": 0.3100592565128255, "全帯域_40ms_log_rms_mae_db": 0.4258072754338517, "全帯域_40ms_rms_envelope_nrmse": 0.037250619492614795, "全帯域_40ms_envelope_correlation": 0.9986535500711121, "全帯域_40ms_modulation_mae_db": 0.3245089670702434, "300–1000 Hz_10ms_log_rms_mae_db": 1.441919998883582, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.083587071057654, "300–1000 Hz_10ms_envelope_correlation": 0.9947806039578623, "300–1000 Hz_10ms_modulation_mae_db": 0.18493282370552264, "300–1000 Hz_40ms_log_rms_mae_db": 0.7955650962547343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049006746446299175, "300–1000 Hz_40ms_envelope_correlation": 0.9981056593416412, "300–1000 Hz_40ms_modulation_mae_db": 0.2516023226394489, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4139303231796694, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11495819174437868, "1000–3000 Hz_10ms_envelope_correlation": 0.9895895605513881, "1000–3000 Hz_10ms_modulation_mae_db": 0.5969238103430575, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8365673700710796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06423322097966967, "1000–3000 Hz_40ms_envelope_correlation": 0.9967133249750434, "1000–3000 Hz_40ms_modulation_mae_db": 0.6816757560465572, "3000–6000 Hz_10ms_log_rms_mae_db": 1.453861612358319, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1504399875825207, "3000–6000 Hz_10ms_envelope_correlation": 0.9800489311833065, "3000–6000 Hz_10ms_modulation_mae_db": 0.4764747393112225, "3000–6000 Hz_40ms_log_rms_mae_db": 0.822919992052922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08260170516584452, "3000–6000 Hz_40ms_envelope_correlation": 0.9935163003122075, "3000–6000 Hz_40ms_modulation_mae_db": 0.3844985771588405, "mr_stft_db": 5.47367352839652, "band_mae_db": 0.15281945969165267, "rms_gain_error_db": -0.05821855759546091, "output_peak": 0.5322243646208106}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8242243231891656, "全帯域_10ms_rms_envelope_nrmse": 0.062480845792399105, "全帯域_10ms_envelope_correlation": 0.9966161233855662, "全帯域_10ms_modulation_mae_db": 0.39119586833923387, "全帯域_40ms_log_rms_mae_db": 0.43700894296188275, "全帯域_40ms_rms_envelope_nrmse": 0.034290427728413835, "全帯域_40ms_envelope_correlation": 0.9990803572880249, "全帯域_40ms_modulation_mae_db": 0.40643496087175784, "300–1000 Hz_10ms_log_rms_mae_db": 1.443841706912496, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07069742044447182, "300–1000 Hz_10ms_envelope_correlation": 0.9964120421156871, "300–1000 Hz_10ms_modulation_mae_db": 0.3294443860698519, "300–1000 Hz_40ms_log_rms_mae_db": 0.7509565641014267, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041042811454255174, "300–1000 Hz_40ms_envelope_correlation": 0.9988569707130867, "300–1000 Hz_40ms_modulation_mae_db": 0.6080435664585859, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5219465176767808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12205975133764461, "1000–3000 Hz_10ms_envelope_correlation": 0.988244734664218, "1000–3000 Hz_10ms_modulation_mae_db": 0.2992651182185468, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8777185684711263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07027270464635471, "1000–3000 Hz_40ms_envelope_correlation": 0.9960444456019378, "1000–3000 Hz_40ms_modulation_mae_db": 0.41271336145970494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4628418982405185, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1608692356238239, "3000–6000 Hz_10ms_envelope_correlation": 0.9775215502640547, "3000–6000 Hz_10ms_modulation_mae_db": 0.6199195337135436, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8668309882989745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0853580525536596, "3000–6000 Hz_40ms_envelope_correlation": 0.9938062458817338, "3000–6000 Hz_40ms_modulation_mae_db": 0.6404143844487685, "mr_stft_db": 5.480078869369369, "band_mae_db": 0.2029111333832002, "rms_gain_error_db": -0.1409133738497702, "output_peak": 0.5264648361743363}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.846008778801731, "全帯域_10ms_rms_envelope_nrmse": 0.06928624417076161, "全帯域_10ms_envelope_correlation": 0.9956372715928189, "全帯域_10ms_modulation_mae_db": 0.09456150686090509, "全帯域_40ms_log_rms_mae_db": 0.44775282518004694, "全帯域_40ms_rms_envelope_nrmse": 0.03671439542390115, "全帯域_40ms_envelope_correlation": 0.9987420860297087, "全帯域_40ms_modulation_mae_db": 0.22548329249172064, "300–1000 Hz_10ms_log_rms_mae_db": 1.4936915956658343, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07821700437794757, "300–1000 Hz_10ms_envelope_correlation": 0.9954597091423174, "300–1000 Hz_10ms_modulation_mae_db": 0.1648052911816121, "300–1000 Hz_40ms_log_rms_mae_db": 0.7432193645746513, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04496603835306491, "300–1000 Hz_40ms_envelope_correlation": 0.9984568122373195, "300–1000 Hz_40ms_modulation_mae_db": 0.15486638897648433, "1000–3000 Hz_10ms_log_rms_mae_db": 1.479939162067536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11707664246060137, "1000–3000 Hz_10ms_envelope_correlation": 0.9891097620457914, "1000–3000 Hz_10ms_modulation_mae_db": 0.2279196386452164, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8480778724503505, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0667391034582844, "1000–3000 Hz_40ms_envelope_correlation": 0.9962905897596203, "1000–3000 Hz_40ms_modulation_mae_db": 0.3116899577743184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4712010073982682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1601025917761575, "3000–6000 Hz_10ms_envelope_correlation": 0.9784152981269078, "3000–6000 Hz_10ms_modulation_mae_db": 1.0058273954695833, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8608070346822836, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09121880745882173, "3000–6000 Hz_40ms_envelope_correlation": 0.9937271449069782, "3000–6000 Hz_40ms_modulation_mae_db": 1.1644057707677027, "mr_stft_db": 5.459397746274175, "band_mae_db": 0.22815401748051745, "rms_gain_error_db": -0.08665274108635654, "output_peak": 0.5342260803766763}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 91600, "output_samples": 91600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.5489843643627745, "spectral_convergence": 0.2605071025392575, "spectral_convergence_status": "ok", "amplitude_floor": 1.4112644553101518e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.287969579278841, "spectral_convergence": 0.25423314997382823, "spectral_convergence_status": "ok", "amplitude_floor": 1.1975531593072457e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.555950556277235, "spectral_convergence": 0.2600507192963697, "spectral_convergence_status": "ok", "amplitude_floor": 1.114415465542996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.564075826256522e-05, "output_power": 6.24367369232617e-05, "delta_db": 0.5004715797973235, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010000129923479963, "output_power": 0.0009810223672825635, "delta_db": -0.08326733073502733, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018259392170112694, "output_power": 0.001804244184646908, "delta_db": -0.05191002201853475, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005127219920465177, "output_power": 0.00048808881669390003, "delta_db": -0.21383088892405194, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00033320627903786135, "output_power": 0.00031390921565008277, "delta_db": -0.2590911101491752, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002817706196064759, "output_power": 0.0002766643959018127, "delta_db": -0.07942433851432688, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19799921168973997, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09076923036205559, "pitch": null, "output_rms_dbfs": -24.060092509610186, "output_peak": 0.5361557004650915, "reference_rms_dbfs": -23.96932327924813, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.46430149997295}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 91600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00020825955187282386, "log_rms_mae_db": 0.8387719200385727, "rms_envelope_nrmse": 0.06928352846813671, "envelope_correlation": 0.9956557897827184, "rms_gain_error_db": -0.09075150654418036, "reference_mean_power": 0.004012756910556895, "output_mean_power": 0.00392977513759015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 2.2120307480210933e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008889321025104637, "output_power": 0.0008502748336235272, "reference_active": true, "delta_db": -0.19309265358134292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022980725896418495, "output_power": 0.00021083930662776936, "reference_active": true, "delta_db": -0.3741216339676431}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1131932747773025e-05, "output_power": 3.598198291851102e-05, "reference_active": true, "delta_db": -0.5809402399383924}], "active_band_count": 3, "active_band_mae_db": 0.3827181758291262}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.0001984160011662782, "log_rms_mae_db": 0.43229366017215914, "rms_envelope_nrmse": 0.035505776341888604, "envelope_correlation": 0.9988465830131633, "rms_gain_error_db": -0.09076885311536811, "reference_mean_power": 0.0040337408532874226, "output_mean_power": 0.0039503093647126186}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 2.0115882799069226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008024963383362849, "output_power": 0.000770059993749875, "reference_active": true, "delta_db": -0.17918498006869457}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011463438169489824, "output_power": 0.00010235562586913543, "reference_active": true, "delta_db": -0.49203174664562543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4945953602736586e-06, "output_power": 2.8348101022137652e-06, "reference_active": true, "delta_db": -0.9087292421197801}], "active_band_count": 3, "active_band_mae_db": 0.5266486562780334}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00019054595180557928, "log_rms_mae_db": 1.4214141342046833, "rms_envelope_nrmse": 0.08255539697789854, "envelope_correlation": 0.9949332728275956, "rms_gain_error_db": -0.07594429569946859, "reference_mean_power": 0.0025757888112863028, "output_mean_power": 0.002531137993386716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.7595886306573375e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007279305516452219, "output_power": 0.0007105964788776634, "reference_active": true, "delta_db": -0.10466896182440517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019010480993234813, "output_power": 0.00016973210220627593, "reference_active": true, "delta_db": -0.4922911511433545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6294378423488108e-05, "output_power": 2.4531883865044254e-05, "reference_active": true, "delta_db": -0.3013200887021231}], "active_band_count": 3, "active_band_mae_db": 0.29942673388996094}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.00018019065505981728, "log_rms_mae_db": 0.7215655501665419, "rms_envelope_nrmse": 0.04459439659231298, "envelope_correlation": 0.9984820569917022, "rms_gain_error_db": -0.07594019771446246, "reference_mean_power": 0.0025893611436675294, "output_mean_power": 0.0025444774528787626}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.6269545971301712e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006780414214466183, "output_power": 0.0006628366304531308, "reference_active": true, "delta_db": -0.09849724760777823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.548437806715927e-05, "output_power": 8.347224153911056e-05, "reference_active": true, "delta_db": -0.583902476118224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9327510117604047e-06, "output_power": 2.3764343038097252e-06, "reference_active": true, "delta_db": -0.913493805006806}], "active_band_count": 3, "active_band_mae_db": 0.5319645095776028}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.327720993461315e-05, "log_rms_mae_db": 1.461238711755153, "rms_envelope_nrmse": 0.11392679353123143, "envelope_correlation": 0.9901272303284759, "rms_gain_error_db": -0.2706418994900166, "reference_mean_power": 0.0007122971869413385, "output_mean_power": 0.0006692633442983755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 4.689927289528067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023362430343497353, "output_power": 0.00021889766770035542, "reference_active": true, "delta_db": -0.2827688521947962}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.797902128989837e-05, "output_power": 3.4993359803415606e-05, "reference_active": true, "delta_db": -0.3555812665584188}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.042184778585592e-05, "output_power": 9.759199638295347e-06, "reference_active": true, "delta_db": -0.2853052367565975}], "active_band_count": 3, "active_band_mae_db": 0.3078851185032709}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 8.747046012456969e-05, "log_rms_mae_db": 0.8670612204311741, "rms_envelope_nrmse": 0.06687483464307632, "envelope_correlation": 0.9968743881888192, "rms_gain_error_db": -0.270752400335616, "reference_mean_power": 0.0007159674763911086, "output_mean_power": 0.000672694775006086}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 4.328159646639031e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021700970165722394, "output_power": 0.00020408140518479785, "reference_active": true, "delta_db": -0.2667571398569934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.702508733600319e-05, "output_power": 1.6200445233607836e-05, "reference_active": true, "delta_db": -0.215623981171426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.679451250427792e-07, "output_power": 6.823531066042717e-07, "reference_active": true, "delta_db": -1.044830950750017}], "active_band_count": 3, "active_band_mae_db": 0.5090706905928122}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.039773028293347e-05, "log_rms_mae_db": 1.4443047639719677, "rms_envelope_nrmse": 0.14904423971136105, "envelope_correlation": 0.9804579686724403, "rms_gain_error_db": -0.0869047972427448, "reference_mean_power": 0.0003237352885604671, "output_mean_power": 0.00031732154654006637}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.2480235329901466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.70846061617512e-05, "output_power": 5.634586702618996e-05, "reference_active": true, "delta_db": -0.05656943099659425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9595191589145483e-05, "output_power": 1.9590896480051483e-05, "reference_active": true, "delta_db": -0.0009520430734871517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.074204859780537e-06, "output_power": 4.046061711798828e-06, "reference_active": true, "delta_db": -0.983354944367089}], "active_band_count": 3, "active_band_mae_db": 0.34695880614572344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 6.399293414479695e-05, "log_rms_mae_db": 0.8687471729830932, "rms_envelope_nrmse": 0.08499265848782823, "envelope_correlation": 0.993032482069486, "rms_gain_error_db": -0.08697242089907693, "reference_mean_power": 0.0003253818281247376, "output_mean_power": 0.0003189304992981227}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.1005868407421181e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4497636322434906e-05, "output_power": 5.411772699503306e-05, "reference_active": true, "delta_db": -0.030381189609469242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.385752117634814e-06, "output_power": 9.251211170305516e-06, "reference_active": true, "delta_db": -0.06270485718930693}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7591737704275347e-07, "output_power": 3.089448576998586e-07, "reference_active": true, "delta_db": -0.8521143084151738}], "active_band_count": 3, "active_band_mae_db": 0.31506678507131664}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8802243108213782, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9b1c91aea6c917a99d98fb18915fd168d6a5b6cc12888998f479d91f9d6ffcd6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9896071546619194, "standard_deviation": 0.0012190844970587573, "minimum": 0.9880200610357713, "maximum": 0.9918725820421245, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4597122590341767, "standard_deviation": 0.03521927305470213, "minimum": 1.4139303231796694, "maximum": 1.5219465176767808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.33640710065675916, "standard_deviation": 0.14730419096256334, "minimum": 0.12718480505850135, "maximum": 0.5969238103430575, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11487128103168236, "standard_deviation": 0.006482162567446801, "minimum": 0.10195632135165765, "maximum": 0.12278608117922439, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967183675778062, "standard_deviation": 0.0007286646696957499, "minimum": 0.9959655841908029, "maximum": 0.9982008484604632, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.841498342760109, "standard_deviation": 0.031395694443808755, "minimum": 0.788573813215356, "maximum": 0.8777185684711263, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4442403391741129, "standard_deviation": 0.14853304594829383, "minimum": 0.269169992628924, "maximum": 0.6816757560465572, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0641714850917579, "standard_deviation": 0.006673415993740992, "minimum": 0.04919637083463172, "maximum": 0.07027270464635471, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9788970128882092, "standard_deviation": 0.001592066507462177, "minimum": 0.9760118736436085, "maximum": 0.9804579686724403, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4410583676945885, "standard_deviation": 0.02466649872193635, "minimum": 1.4022636499722079, "maximum": 1.4712010073982682, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6407273500603935, "standard_deviation": 0.23740451355368972, "minimum": 0.34695880614572344, "maximum": 1.0058273954695833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15578332097491548, "standard_deviation": 0.006704149820183969, "minimum": 0.14904423971136105, "maximum": 0.1678603427778609, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992986787687361, "standard_deviation": 0.0010689127140833903, "minimum": 0.9909346616057547, "maximum": 0.9941908085840617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8452546722251297, "standard_deviation": 0.022288927817666414, "minimum": 0.8088575129615801, "maximum": 0.8687471729830932, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6149082247138895, "standard_deviation": 0.2842229441923141, "minimum": 0.31506678507131664, "maximum": 1.1644057707677027, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08882748345187691, "standard_deviation": 0.007337990740242723, "minimum": 0.081696692701147, "maximum": 0.10468037307869807, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9952035200906653, "standard_deviation": 0.0007816211866163033, "minimum": 0.9939905903177912, "maximum": 0.9964120421156871, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.441861959585538, "standard_deviation": 0.02791559046637804, "minimum": 1.3972690516269601, "maximum": 1.4936915956658343, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.24670316174286727, "standard_deviation": 0.08604912382754097, "minimum": 0.14627452158574747, "maximum": 0.3386876028183237, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08020132549144547, "standard_deviation": 0.006295549891616245, "minimum": 0.07069742044447182, "maximum": 0.08992150922017869, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984059703418781, "standard_deviation": 0.0003793662933075985, "minimum": 0.9976906073786133, "maximum": 0.9988569707130867, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7653727587696626, "standard_deviation": 0.029393348758791234, "minimum": 0.7215655501665419, "maximum": 0.8075790606788783, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3775264280827343, "standard_deviation": 0.1940460328873954, "minimum": 0.07217106136784975, "maximum": 0.6080435664585859, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.045525336119857596, "standard_deviation": 0.00482959460955594, "minimum": 0.038587372478063896, "maximum": 0.05445172531955506, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19795318842414794, "standard_deviation": 0.023666955225780606, "minimum": 0.15281945969165267, "maximum": 0.22815401748051745, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.469102439476127, "standard_deviation": 0.009966741386026896, "minimum": 5.455504567003186, "maximum": 5.484244416283601, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5341037036075793, "standard_deviation": 0.011859200226959814, "minimum": 0.5114691255884032, "maximum": 0.5521351471988357, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09120542668565343, "standard_deviation": 0.024307240984169892, "minimum": -0.1409133738497702, "maximum": -0.05821855759546091, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955288132423179, "standard_deviation": 0.0008195907517074604, "minimum": 0.9937500065935589, "maximum": 0.9966161233855662, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.839247128845636, "standard_deviation": 0.013522392821526914, "minimum": 0.81983778399532, "maximum": 0.8636919166146789, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3041401325481069, "standard_deviation": 0.10138529218839334, "minimum": 0.09456150686090509, "maximum": 0.39119586833923387, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07010764321700537, "standard_deviation": 0.005821002454185343, "minimum": 0.062480845792399105, "maximum": 0.08272771173257883, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987234646935803, "standard_deviation": 0.0002469087492063444, "minimum": 0.9983470499110161, "maximum": 0.9990803572880249, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4382746182523021, "standard_deviation": 0.018855439521405788, "minimum": 0.417591626068676, "maximum": 0.47803072195684987, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4143112760568216, "standard_deviation": 0.14848407612703357, "minimum": 0.20707994204018365, "maximum": 0.6067150950019481, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03719099247298595, "standard_deviation": 0.0030659831883021052, "minimum": 0.033456392792029575, "maximum": 0.04194789431773518, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0014", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.5378358364105225, "predicted_mos": 2.5378358364105225, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "824819a74625481b87bafb213ce4d19f51fc34ab037d5d1765a62f4905a594c6", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 12.871587599627674, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8554980214390965, "全帯域_10ms_rms_envelope_nrmse": 0.0800082925326661, "全帯域_10ms_envelope_correlation": 0.9941108190156774, "全帯域_10ms_modulation_mae_db": 0.22924958379389745, "全帯域_40ms_log_rms_mae_db": 0.45395469238554476, "全帯域_40ms_rms_envelope_nrmse": 0.04080019383875991, "全帯域_40ms_envelope_correlation": 0.9983587436930712, "全帯域_40ms_modulation_mae_db": 0.2440155392099271, "300–1000 Hz_10ms_log_rms_mae_db": 1.7832454852226203, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1067653164155368, "300–1000 Hz_10ms_envelope_correlation": 0.9914671772952389, "300–1000 Hz_10ms_modulation_mae_db": 0.7138403470900833, "300–1000 Hz_40ms_log_rms_mae_db": 1.0013027654155617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06021105620452401, "300–1000 Hz_40ms_envelope_correlation": 0.997123717130426, "300–1000 Hz_40ms_modulation_mae_db": 0.3290157533468916, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6322421150985964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11535333369476275, "1000–3000 Hz_10ms_envelope_correlation": 0.9894889646382525, "1000–3000 Hz_10ms_modulation_mae_db": 0.5983344671383826, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9681556352890618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0637228920090621, "1000–3000 Hz_40ms_envelope_correlation": 0.9967266720068106, "1000–3000 Hz_40ms_modulation_mae_db": 0.8120242613909543, "3000–6000 Hz_10ms_log_rms_mae_db": 1.566493462635116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17260614016325826, "3000–6000 Hz_10ms_envelope_correlation": 0.9743450257609665, "3000–6000 Hz_10ms_modulation_mae_db": 1.702301304926955, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9331115958498675, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09913465331474296, "3000–6000 Hz_40ms_envelope_correlation": 0.991815501659275, "3000–6000 Hz_40ms_modulation_mae_db": 1.7932251688392575, "mr_stft_db": 5.713600841023035, "band_mae_db": 0.2584157898278401, "rms_gain_error_db": -0.0449450293609604, "output_peak": 0.5323458346142188}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8288415256094962, "全帯域_10ms_rms_envelope_nrmse": 0.07641987031075688, "全帯域_10ms_envelope_correlation": 0.9947572700239526, "全帯域_10ms_modulation_mae_db": 0.5491490707729314, "全帯域_40ms_log_rms_mae_db": 0.44416759173429954, "全帯域_40ms_rms_envelope_nrmse": 0.03986536986114584, "全帯域_40ms_envelope_correlation": 0.99860819541078, "全帯域_40ms_modulation_mae_db": 0.8778774895312228, "300–1000 Hz_10ms_log_rms_mae_db": 1.6559252126250417, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08910072533084326, "300–1000 Hz_10ms_envelope_correlation": 0.994103011781052, "300–1000 Hz_10ms_modulation_mae_db": 0.32767280966907936, "300–1000 Hz_40ms_log_rms_mae_db": 0.9577133964286279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04635536951764492, "300–1000 Hz_40ms_envelope_correlation": 0.9983789050079271, "300–1000 Hz_40ms_modulation_mae_db": 0.6794539740158854, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6532941598221438, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11572174049818228, "1000–3000 Hz_10ms_envelope_correlation": 0.9898612035558966, "1000–3000 Hz_10ms_modulation_mae_db": 0.8739091783356684, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9869720496390267, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0662568130884565, "1000–3000 Hz_40ms_envelope_correlation": 0.9970296405455559, "1000–3000 Hz_40ms_modulation_mae_db": 1.2123872947858991, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6461026021742564, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17826103020931866, "3000–6000 Hz_10ms_envelope_correlation": 0.9726197342440189, "3000–6000 Hz_10ms_modulation_mae_db": 1.1627798198393606, "3000–6000 Hz_40ms_log_rms_mae_db": 0.98973570597247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10271683946712744, "3000–6000 Hz_40ms_envelope_correlation": 0.9912801531204807, "3000–6000 Hz_40ms_modulation_mae_db": 1.5345102387248364, "mr_stft_db": 5.70789471756338, "band_mae_db": 0.24968892797097375, "rms_gain_error_db": -0.12574588659352448, "output_peak": 0.5489207743155798}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8193312456073294, "全帯域_10ms_rms_envelope_nrmse": 0.06542692840005956, "全帯域_10ms_envelope_correlation": 0.996280086068742, "全帯域_10ms_modulation_mae_db": 0.4891241491202945, "全帯域_40ms_log_rms_mae_db": 0.4118928273384515, "全帯域_40ms_rms_envelope_nrmse": 0.03705846236046492, "全帯域_40ms_envelope_correlation": 0.9989072305912402, "全帯域_40ms_modulation_mae_db": 0.680176020297742, "300–1000 Hz_10ms_log_rms_mae_db": 1.6166849179660483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08355325027515381, "300–1000 Hz_10ms_envelope_correlation": 0.9949075390608297, "300–1000 Hz_10ms_modulation_mae_db": 0.35332690841061626, "300–1000 Hz_40ms_log_rms_mae_db": 0.9072394584076817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052413268368203744, "300–1000 Hz_40ms_envelope_correlation": 0.9979989394716606, "300–1000 Hz_40ms_modulation_mae_db": 0.6061079571451461, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5885275099054905, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12116585012056877, "1000–3000 Hz_10ms_envelope_correlation": 0.9887021330969525, "1000–3000 Hz_10ms_modulation_mae_db": 0.8047444863244398, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9322982985738878, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07504434039568161, "1000–3000 Hz_40ms_envelope_correlation": 0.9958073748771333, "1000–3000 Hz_40ms_modulation_mae_db": 1.012916699949744, "3000–6000 Hz_10ms_log_rms_mae_db": 1.577780733635216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16528823893116917, "3000–6000 Hz_10ms_envelope_correlation": 0.9761342672014366, "3000–6000 Hz_10ms_modulation_mae_db": 0.9243859747267528, "3000–6000 Hz_40ms_log_rms_mae_db": 0.914159216809801, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0946435843884844, "3000–6000 Hz_40ms_envelope_correlation": 0.9919735043579593, "3000–6000 Hz_40ms_modulation_mae_db": 1.0349928265413102, "mr_stft_db": 5.703609244669349, "band_mae_db": 0.2903144577675634, "rms_gain_error_db": -0.15237685634391943, "output_peak": 0.5352880541404852}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8136329128610471, "全帯域_10ms_rms_envelope_nrmse": 0.06809623690991935, "全帯域_10ms_envelope_correlation": 0.995849823725841, "全帯域_10ms_modulation_mae_db": 0.8263684764705573, "全帯域_40ms_log_rms_mae_db": 0.4620340715843189, "全帯域_40ms_rms_envelope_nrmse": 0.039825952458794805, "全帯域_40ms_envelope_correlation": 0.9985667754432617, "全帯域_40ms_modulation_mae_db": 0.9621943207080634, "300–1000 Hz_10ms_log_rms_mae_db": 1.6841545398067623, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08940653026942665, "300–1000 Hz_10ms_envelope_correlation": 0.9940754411546555, "300–1000 Hz_10ms_modulation_mae_db": 0.721127486667888, "300–1000 Hz_40ms_log_rms_mae_db": 0.9688846601313335, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05940024010314428, "300–1000 Hz_40ms_envelope_correlation": 0.9972818201084339, "300–1000 Hz_40ms_modulation_mae_db": 0.970905419982491, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6123111919189577, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12301774002697734, "1000–3000 Hz_10ms_envelope_correlation": 0.9880381812782231, "1000–3000 Hz_10ms_modulation_mae_db": 0.6452059150552804, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0125494860275817, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07401388484865869, "1000–3000 Hz_40ms_envelope_correlation": 0.9955418940245505, "1000–3000 Hz_40ms_modulation_mae_db": 0.6616371715546998, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6163976103610103, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17833544430208922, "3000–6000 Hz_10ms_envelope_correlation": 0.9725927950131121, "3000–6000 Hz_10ms_modulation_mae_db": 1.472742898952032, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9435036752673915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10089150347026267, "3000–6000 Hz_40ms_envelope_correlation": 0.9916684378297393, "3000–6000 Hz_40ms_modulation_mae_db": 1.5776699640966534, "mr_stft_db": 5.694235753841279, "band_mae_db": 0.24661949994193477, "rms_gain_error_db": -0.11401550094213932, "output_peak": 0.5193279042992052}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8176180119747551, "全帯域_10ms_rms_envelope_nrmse": 0.06651026496070457, "全帯域_10ms_envelope_correlation": 0.996126301758405, "全帯域_10ms_modulation_mae_db": 0.6695625009559354, "全帯域_40ms_log_rms_mae_db": 0.43171092135413464, "全帯域_40ms_rms_envelope_nrmse": 0.038464541999462046, "全帯域_40ms_envelope_correlation": 0.9987664436699512, "全帯域_40ms_modulation_mae_db": 0.9118367118902363, "300–1000 Hz_10ms_log_rms_mae_db": 1.652373253467967, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07852593874516715, "300–1000 Hz_10ms_envelope_correlation": 0.9954919446813324, "300–1000 Hz_10ms_modulation_mae_db": 0.4452647767312053, "300–1000 Hz_40ms_log_rms_mae_db": 0.9642474810274932, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04713040978001054, "300–1000 Hz_40ms_envelope_correlation": 0.9983780762222966, "300–1000 Hz_40ms_modulation_mae_db": 0.729298759673198, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6060800753966262, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11898097628771143, "1000–3000 Hz_10ms_envelope_correlation": 0.9895154090211601, "1000–3000 Hz_10ms_modulation_mae_db": 0.9764761436458644, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9924837594014906, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0744606092914218, "1000–3000 Hz_40ms_envelope_correlation": 0.9963644364233787, "1000–3000 Hz_40ms_modulation_mae_db": 1.1972934846352155, "3000–6000 Hz_10ms_log_rms_mae_db": 1.637466987694015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17323833563401417, "3000–6000 Hz_10ms_envelope_correlation": 0.9734117317460367, "3000–6000 Hz_10ms_modulation_mae_db": 0.5632421060588916, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9686528909966262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09487962774705877, "3000–6000 Hz_40ms_envelope_correlation": 0.9914212257654864, "3000–6000 Hz_40ms_modulation_mae_db": 0.637031262282907, "mr_stft_db": 5.707622317777681, "band_mae_db": 0.2561774397912614, "rms_gain_error_db": -0.14096820366075483, "output_peak": 0.552544282101526}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8283223274987301, "全帯域_10ms_rms_envelope_nrmse": 0.07000561593528987, "全帯域_10ms_envelope_correlation": 0.9955127662719163, "全帯域_10ms_modulation_mae_db": 0.395046515608674, "全帯域_40ms_log_rms_mae_db": 0.4410694799418849, "全帯域_40ms_rms_envelope_nrmse": 0.03878346696998454, "全帯域_40ms_envelope_correlation": 0.998536086600141, "全帯域_40ms_modulation_mae_db": 0.51719905218961, "300–1000 Hz_10ms_log_rms_mae_db": 1.725021137809427, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.081260996677714, "300–1000 Hz_10ms_envelope_correlation": 0.995065742827726, "300–1000 Hz_10ms_modulation_mae_db": 0.09497630023711283, "300–1000 Hz_40ms_log_rms_mae_db": 0.9968477633487617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04751083928937265, "300–1000 Hz_40ms_envelope_correlation": 0.9982030032457058, "300–1000 Hz_40ms_modulation_mae_db": 0.26480060463084126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.639316468703814, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1125819491991731, "1000–3000 Hz_10ms_envelope_correlation": 0.9905278593529031, "1000–3000 Hz_10ms_modulation_mae_db": 0.4946109675496046, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9721473554067198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06302257669284886, "1000–3000 Hz_40ms_envelope_correlation": 0.9974973117437828, "1000–3000 Hz_40ms_modulation_mae_db": 0.9151420442392665, "3000–6000 Hz_10ms_log_rms_mae_db": 1.614898479018531, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16830126775208087, "3000–6000 Hz_10ms_envelope_correlation": 0.975603415204703, "3000–6000 Hz_10ms_modulation_mae_db": 1.4112428286973548, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9279662020903927, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09542276372215026, "3000–6000 Hz_40ms_envelope_correlation": 0.9924498250174786, "3000–6000 Hz_40ms_modulation_mae_db": 1.4832327228586226, "mr_stft_db": 5.70713213768206, "band_mae_db": 0.24897496008449907, "rms_gain_error_db": -0.06364230777177149, "output_peak": 0.569810028240914}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8202603055553286, "全帯域_10ms_rms_envelope_nrmse": 0.07193405016784706, "全帯域_10ms_envelope_correlation": 0.9952464419037622, "全帯域_10ms_modulation_mae_db": 0.20725686382739764, "全帯域_40ms_log_rms_mae_db": 0.4133450739940347, "全帯域_40ms_rms_envelope_nrmse": 0.03818345297145783, "全帯域_40ms_envelope_correlation": 0.9985636337726141, "全帯域_40ms_modulation_mae_db": 0.16349066816277394, "300–1000 Hz_10ms_log_rms_mae_db": 1.706805747635708, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08689702125393232, "300–1000 Hz_10ms_envelope_correlation": 0.9943849962737507, "300–1000 Hz_10ms_modulation_mae_db": 0.062479790894899266, "300–1000 Hz_40ms_log_rms_mae_db": 0.9937466403461185, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05028216490807334, "300–1000 Hz_40ms_envelope_correlation": 0.9980327353566609, "300–1000 Hz_40ms_modulation_mae_db": 0.04278825477161977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5433426274691824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1101826434095582, "1000–3000 Hz_10ms_envelope_correlation": 0.9905299121253551, "1000–3000 Hz_10ms_modulation_mae_db": 0.17249408294105373, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9181046669693275, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06766900810492184, "1000–3000 Hz_40ms_envelope_correlation": 0.996395938996712, "1000–3000 Hz_40ms_modulation_mae_db": 0.34916510470393497, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6172924590547504, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1752429035959437, "3000–6000 Hz_10ms_envelope_correlation": 0.9728345416068848, "3000–6000 Hz_10ms_modulation_mae_db": 1.245317069738322, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9218040131390125, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10085320547963084, "3000–6000 Hz_40ms_envelope_correlation": 0.9903663789750259, "3000–6000 Hz_40ms_modulation_mae_db": 1.262977238684744, "mr_stft_db": 5.708866904160405, "band_mae_db": 0.1699390654047034, "rms_gain_error_db": -0.03521839629558429, "output_peak": 0.5275069243495719}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7984487878784056, "全帯域_10ms_rms_envelope_nrmse": 0.06592243150297081, "全帯域_10ms_envelope_correlation": 0.9960400856043479, "全帯域_10ms_modulation_mae_db": 0.10506897936926092, "全帯域_40ms_log_rms_mae_db": 0.3946922869956277, "全帯域_40ms_rms_envelope_nrmse": 0.03267430794700025, "全帯域_40ms_envelope_correlation": 0.9989937797877739, "全帯域_40ms_modulation_mae_db": 0.12591785572974065, "300–1000 Hz_10ms_log_rms_mae_db": 1.6721062646212554, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07845904509047734, "300–1000 Hz_10ms_envelope_correlation": 0.9954094947588684, "300–1000 Hz_10ms_modulation_mae_db": 0.31610442269509237, "300–1000 Hz_40ms_log_rms_mae_db": 0.8875991417927401, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04014475112007622, "300–1000 Hz_40ms_envelope_correlation": 0.9987544931240698, "300–1000 Hz_40ms_modulation_mae_db": 0.22140103453725038, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5403390559520593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10806309566079689, "1000–3000 Hz_10ms_envelope_correlation": 0.9909899151340815, "1000–3000 Hz_10ms_modulation_mae_db": 0.817610022042464, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8976745255202642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06144810002196429, "1000–3000 Hz_40ms_envelope_correlation": 0.9972027597399686, "1000–3000 Hz_40ms_modulation_mae_db": 0.8825381728963911, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5972278420083452, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1704217819570103, "3000–6000 Hz_10ms_envelope_correlation": 0.9745695047642372, "3000–6000 Hz_10ms_modulation_mae_db": 1.0270877594064562, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8880464221027886, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08706965270201548, "3000–6000 Hz_40ms_envelope_correlation": 0.9934392782053677, "3000–6000 Hz_40ms_modulation_mae_db": 1.0252717603773678, "mr_stft_db": 5.710321352823013, "band_mae_db": 0.27559154396776003, "rms_gain_error_db": -0.06549736914203522, "output_peak": 0.5454872519836395}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 91600, "output_samples": 91600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.829721452067407, "spectral_convergence": 0.2728116873857294, "spectral_convergence_status": "ok", "amplitude_floor": 1.4112644553101518e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.575052321107012, "spectral_convergence": 0.265918938028395, "spectral_convergence_status": "ok", "amplitude_floor": 1.1975531593072457e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.736028749894684, "spectral_convergence": 0.2677833164182978, "spectral_convergence_status": "ok", "amplitude_floor": 1.114415465542996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.564075826256522e-05, "output_power": 6.301000504917583e-05, "delta_db": 0.5401647456649854, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010000129923479963, "output_power": 0.001051871441031418, "delta_db": 0.21957021434488538, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018259392170112694, "output_power": 0.0017791838282398037, "delta_db": -0.1126549394459252, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005127219920465177, "output_power": 0.0004931323361288829, "delta_db": -0.16918464492684754, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00033320627903786135, "output_power": 0.0003192352477449045, "delta_db": -0.18602339671040483, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002817706196064759, "output_power": 0.0002615808985017485, "delta_db": -0.3228967978739921, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2584157898278401, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0449450293609604, "pitch": null, "output_rms_dbfs": -24.014268308609093, "output_peak": 0.5323458346142188, "reference_rms_dbfs": -23.96932327924813, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.713600841023035}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 91600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00020825955187282386, "log_rms_mae_db": 0.8554980214390965, "rms_envelope_nrmse": 0.0800082925326661, "envelope_correlation": 0.9941108190156774, "rms_gain_error_db": -0.0449334793485868, "reference_mean_power": 0.004012756910556895, "output_mean_power": 0.003971453696138742}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 2.2120307480210933e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008889321025104637, "output_power": 0.0008813544610514682, "reference_active": true, "delta_db": -0.03717983335138528}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022980725896418495, "output_power": 0.00021134255613612586, "reference_active": true, "delta_db": -0.3637678690481105}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1131932747773025e-05, "output_power": 4.393991595951728e-05, "reference_active": true, "delta_db": 0.2868010489821966}], "active_band_count": 3, "active_band_mae_db": 0.22924958379389745}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.0001984160011662782, "log_rms_mae_db": 0.45395469238554476, "rms_envelope_nrmse": 0.04080019383875991, "envelope_correlation": 0.9983587436930712, "rms_gain_error_db": -0.04493670712097244, "reference_mean_power": 0.0040337408532874226, "output_mean_power": 0.003992218684524169}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 2.0115882799069226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008024963383362849, "output_power": 0.0008014301309965742, "reference_active": true, "delta_db": -0.005773930928382166}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011463438169489824, "output_power": 0.0001067421920334354, "reference_active": true, "delta_db": -0.309787757256695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4945953602736586e-06, "output_power": 3.1750341660195293e-06, "reference_active": true, "delta_db": -0.41648492944470417}], "active_band_count": 3, "active_band_mae_db": 0.2440155392099271}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00019054595180557928, "log_rms_mae_db": 1.7832454852226203, "rms_envelope_nrmse": 0.1067653164155368, "envelope_correlation": 0.9914671772952389, "rms_gain_error_db": -0.009585102316738528, "reference_mean_power": 0.0025757888112863028, "output_mean_power": 0.0025701101818776124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.7595886306573375e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007279305516452219, "output_power": 0.000729229898577579, "reference_active": true, "delta_db": 0.007745191715534406}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019010480993234813, "output_power": 0.00017374419997169882, "reference_active": true, "delta_db": -0.3908278965877553}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6294378423488108e-05, "output_power": 3.927875414183769e-05, "reference_active": true, "delta_db": 1.74294795296696}], "active_band_count": 3, "active_band_mae_db": 0.7138403470900833}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.00018019065505981728, "log_rms_mae_db": 1.0013027654155617, "rms_envelope_nrmse": 0.06021105620452401, "envelope_correlation": 0.997123717130426, "rms_gain_error_db": -0.009587432573413527, "reference_mean_power": 0.0025893611436675294, "output_mean_power": 0.002583651206167721}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.6269545971301712e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006780414214466183, "output_power": 0.0006851047850888433, "reference_active": true, "delta_db": 0.04500775182716807}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.548437806715927e-05, "output_power": 8.776509960273547e-05, "reference_active": true, "delta_db": -0.36610473648189595}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9327510117604047e-06, "output_power": 3.348641617807541e-06, "reference_active": true, "delta_db": 0.5759347717316108}], "active_band_count": 3, "active_band_mae_db": 0.3290157533468916}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.327720993461315e-05, "log_rms_mae_db": 1.6322421150985964, "rms_envelope_nrmse": 0.11535333369476275, "envelope_correlation": 0.9894889646382525, "rms_gain_error_db": -0.1550321349439242, "reference_mean_power": 0.0007122971869413385, "output_mean_power": 0.000687318471643407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 4.689927289528067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023362430343497353, "output_power": 0.00022189574471062867, "reference_active": true, "delta_db": -0.22369045656480432}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.797902128989837e-05, "output_power": 3.133053069346508e-05, "reference_active": true, "delta_db": -0.8357601762377713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.042184778585592e-05, "output_power": 8.79811124580044e-06, "reference_active": true, "delta_db": -0.7355527686125719}], "active_band_count": 3, "active_band_mae_db": 0.5983344671383826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 8.747046012456969e-05, "log_rms_mae_db": 0.9681556352890618, "rms_envelope_nrmse": 0.0637228920090621, "envelope_correlation": 0.9967266720068106, "rms_gain_error_db": -0.15504080445715185, "reference_mean_power": 0.0007159674763911086, "output_mean_power": 0.0006908586729000991}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 4.328159646639031e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021700970165722394, "output_power": 0.00020575602462172604, "reference_active": true, "delta_db": -0.2312658950167039}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.702508733600319e-05, "output_power": 1.413780950810203e-05, "reference_active": true, "delta_db": -0.8070722274253208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.679451250427792e-07, "output_power": 6.290987207156614e-07, "reference_active": true, "delta_db": -1.3977346617308382}], "active_band_count": 3, "active_band_mae_db": 0.8120242613909543}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.039773028293347e-05, "log_rms_mae_db": 1.566493462635116, "rms_envelope_nrmse": 0.17260614016325826, "envelope_correlation": 0.9743450257609665, "rms_gain_error_db": -0.3710810245202244, "reference_mean_power": 0.0003237352885604671, "output_mean_power": 0.0002972226756981404}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.2480235329901466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.70846061617512e-05, "output_power": 5.1456974324829594e-05, "reference_active": true, "delta_db": -0.45074763048715966}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9595191589145483e-05, "output_power": 1.612394660002393e-05, "reference_active": true, "delta_db": -0.846781629178742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.074204859780537e-06, "output_power": 2.1107195437006773e-06, "reference_active": true, "delta_db": -3.8093746551149636}], "active_band_count": 3, "active_band_mae_db": 1.702301304926955}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 6.399293414479695e-05, "log_rms_mae_db": 0.9331115958498675, "rms_envelope_nrmse": 0.09913465331474296, "envelope_correlation": 0.991815501659275, "rms_gain_error_db": -0.37120310504677373, "reference_mean_power": 0.0003253818281247376, "output_mean_power": 0.0002987259729919243}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.1005868407421181e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4497636322434906e-05, "output_power": 4.794599468703662e-05, "reference_active": true, "delta_db": -0.5562533349098124}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.385752117634814e-06, "output_power": 7.816836140777993e-06, "reference_active": true, "delta_db": -0.7943807187398473}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7591737704275347e-07, "output_power": 1.4865799001078656e-07, "reference_active": true, "delta_db": -4.029041452868112}], "active_band_count": 3, "active_band_mae_db": 1.7932251688392575}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8707143079366993, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "eb8d11e6c338f7bc53669a4c76799daffc95868ae5384b6e1a636fcdda461ade", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.989706697275353, "standard_deviation": 0.000992802931762422, "minimum": 0.9880381812782231, "maximum": 0.9909899151340815, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6019316505333587, "standard_deviation": 0.04223301497525902, "minimum": 1.5403390559520593, "maximum": 1.6532941598221438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6729231578790947, "standard_deviation": 0.2561827365325182, "minimum": 0.17249408294105373, "maximum": 0.9764761436458644, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11563341611221634, "standard_deviation": 0.005247592609075813, "minimum": 0.10806309566079689, "maximum": 0.12301774002697734, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9965707535447366, "standard_deviation": 0.0006768894778494946, "minimum": 0.9955418940245505, "maximum": 0.9974973117437828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.96004822210342, "standard_deviation": 0.03993790994692167, "minimum": 0.8976745255202642, "maximum": 1.0125494860275817, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8803880292695132, "standard_deviation": 0.2838893726345559, "minimum": 0.34916510470393497, "maximum": 1.2123872947858991, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06820477805662697, "standard_deviation": 0.005559259813138555, "minimum": 0.06144810002196429, "maximum": 0.07504434039568161, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9740138769426745, "standard_deviation": 0.0013710052067565982, "minimum": 0.9725927950131121, "maximum": 0.9761342672014366, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.609207522072655, "standard_deviation": 0.027425070748796406, "minimum": 1.566493462635116, "maximum": 1.6461026021742564, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1886374702932656, "standard_deviation": 0.3555774566767666, "minimum": 0.5632421060588916, "maximum": 1.702301304926955, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17271189281811056, "standard_deviation": 0.004612412702413742, "minimum": 0.16528823893116917, "maximum": 0.17833544430208922, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9918017881163517, "standard_deviation": 0.0008965101082605896, "minimum": 0.9903663789750259, "maximum": 0.9934392782053677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9358724652785437, "standard_deviation": 0.031760012329801024, "minimum": 0.8880464221027886, "maximum": 0.98973570597247, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2936138978007123, "standard_deviation": 0.3770595519840399, "minimum": 0.637031262282907, "maximum": 1.7932251688392575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0969514787864341, "standard_deviation": 0.005047761884369242, "minimum": 0.08706965270201548, "maximum": 0.10271683946712744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9943631684791817, "standard_deviation": 0.0012926492372782243, "minimum": 0.9914671772952389, "maximum": 0.9954919446813324, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6870395698943537, "standard_deviation": 0.05132507073787533, "minimum": 1.6166849179660483, "maximum": 1.7832454852226203, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.37934910529949706, "standard_deviation": 0.24543351364901977, "minimum": 0.062479790894899266, "maximum": 0.721127486667888, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08674610300728142, "standard_deviation": 0.009180265563654515, "minimum": 0.07845904509047734, "maximum": 0.1067653164155368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980189612083976, "standard_deviation": 0.0005579082566666341, "minimum": 0.997123717130426, "maximum": 0.9987544931240698, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9596976633622898, "standard_deviation": 0.041968604634506856, "minimum": 0.8875991417927401, "maximum": 1.0013027654155617, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48047146976291544, "standard_deviation": 0.31303945139040545, "minimum": 0.04278825477161977, "maximum": 0.970905419982491, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05043101241138122, "standard_deviation": 0.0067811421434009036, "minimum": 0.04014475112007622, "maximum": 0.06021105620452401, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.249465210594567, "standard_deviation": 0.03547475845570185, "minimum": 0.1699390654047034, "maximum": 0.2903144577675634, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.706660408692525, "standard_deviation": 0.005766822607392972, "minimum": 5.694235753841279, "maximum": 5.713600841023035, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5414038817556425, "standard_deviation": 0.016054020570844277, "minimum": 0.5193279042992052, "maximum": 0.569810028240914, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09280119376383618, "standard_deviation": 0.04567970874291809, "minimum": -0.15237685634391943, "maximum": -0.03521839629558429, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954904492965806, "standard_deviation": 0.0007521449185319631, "minimum": 0.9941108190156774, "maximum": 0.996280086068742, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8227441423030235, "standard_deviation": 0.01629410104158356, "minimum": 0.7984487878784056, "maximum": 0.8554980214390965, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4338532674898686, "standard_deviation": 0.2474985409118537, "minimum": 0.10506897936926092, "maximum": 0.8263684764705573, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07054046134002678, "standard_deviation": 0.005293210764610429, "minimum": 0.06542692840005956, "maximum": 0.0800082925326661, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9986626111211041, "standard_deviation": 0.00021063797635128978, "minimum": 0.9983587436930712, "maximum": 0.9989937797877739, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4316083681660371, "standard_deviation": 0.023168286198308796, "minimum": 0.3946922869956277, "maximum": 0.4620340715843189, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5603384572149146, "standard_deviation": 0.34813007832152737, "minimum": 0.12591785572974065, "maximum": 0.9621943207080634, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03820696855088376, "standard_deviation": 0.002519208166750112, "minimum": 0.03267430794700025, "maximum": 0.04080019383875991, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0014", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.338702917098999, "predicted_mos": 2.338702917098999, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "aabfc9605b55c270e16ea9e78043c579a6f4e224f6b08c48738e471476cd4444", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.6865735999308527, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8429217028249015, "全帯域_10ms_rms_envelope_nrmse": 0.0680008516088581, "全帯域_10ms_envelope_correlation": 0.9957821482650732, "全帯域_10ms_modulation_mae_db": 0.24661162730485542, "全帯域_40ms_log_rms_mae_db": 0.43202746523018076, "全帯域_40ms_rms_envelope_nrmse": 0.034312186835629074, "全帯域_40ms_envelope_correlation": 0.9988822204432637, "全帯域_40ms_modulation_mae_db": 0.4072972326558486, "300–1000 Hz_10ms_log_rms_mae_db": 1.378905350751675, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07976169285276245, "300–1000 Hz_10ms_envelope_correlation": 0.9952427042411239, "300–1000 Hz_10ms_modulation_mae_db": 0.4115998648184656, "300–1000 Hz_40ms_log_rms_mae_db": 0.8092162543609015, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041944106229267045, "300–1000 Hz_40ms_envelope_correlation": 0.9986178806979206, "300–1000 Hz_40ms_modulation_mae_db": 0.38920495172366604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3647110277045018, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10036117869983811, "1000–3000 Hz_10ms_envelope_correlation": 0.9921328419112964, "1000–3000 Hz_10ms_modulation_mae_db": 0.3852544256420288, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7855792224912868, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05747073260586687, "1000–3000 Hz_40ms_envelope_correlation": 0.9974263909608585, "1000–3000 Hz_40ms_modulation_mae_db": 0.5869687867161496, "3000–6000 Hz_10ms_log_rms_mae_db": 1.347249121769675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14665032280241475, "3000–6000 Hz_10ms_envelope_correlation": 0.9815314766695558, "3000–6000 Hz_10ms_modulation_mae_db": 0.34828534461650373, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7967707155140565, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07815256332461751, "3000–6000 Hz_40ms_envelope_correlation": 0.995033678932814, "3000–6000 Hz_40ms_modulation_mae_db": 0.34429444660162273, "mr_stft_db": 5.434595819009947, "band_mae_db": 0.2101786265221676, "rms_gain_error_db": -0.06674816652080302, "output_peak": 0.5569887868377952}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8570484309259239, "全帯域_10ms_rms_envelope_nrmse": 0.0725144365231212, "全帯域_10ms_envelope_correlation": 0.9954148677512392, "全帯域_10ms_modulation_mae_db": 0.5202649701516529, "全帯域_40ms_log_rms_mae_db": 0.453729848297837, "全帯域_40ms_rms_envelope_nrmse": 0.04255174014110374, "全帯域_40ms_envelope_correlation": 0.9985199213812926, "全帯域_40ms_modulation_mae_db": 0.7036344675629113, "300–1000 Hz_10ms_log_rms_mae_db": 1.3834903149515876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08695151473631474, "300–1000 Hz_10ms_envelope_correlation": 0.9945461904472376, "300–1000 Hz_10ms_modulation_mae_db": 0.470805630407094, "300–1000 Hz_40ms_log_rms_mae_db": 0.7444307184186582, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055066304752861485, "300–1000 Hz_40ms_envelope_correlation": 0.9978683540832867, "300–1000 Hz_40ms_modulation_mae_db": 0.7775024169728098, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3628310091584974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11219300356728377, "1000–3000 Hz_10ms_envelope_correlation": 0.9902973283145078, "1000–3000 Hz_10ms_modulation_mae_db": 0.573844910875803, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8228015376052328, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06598107819361637, "1000–3000 Hz_40ms_envelope_correlation": 0.9967830599357592, "1000–3000 Hz_40ms_modulation_mae_db": 0.48318806069376824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3610995703321553, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15349751929918065, "3000–6000 Hz_10ms_envelope_correlation": 0.9793459363901097, "3000–6000 Hz_10ms_modulation_mae_db": 0.4313815974686801, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8283178154065108, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08367007109002962, "3000–6000 Hz_40ms_envelope_correlation": 0.9936358005017691, "3000–6000 Hz_40ms_modulation_mae_db": 0.541028566566036, "mr_stft_db": 5.433782044597218, "band_mae_db": 0.24432232713931654, "rms_gain_error_db": -0.16308635548388972, "output_peak": 0.5405773080474057}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8496895263485286, "全帯域_10ms_rms_envelope_nrmse": 0.07604746726840243, "全帯域_10ms_envelope_correlation": 0.9949786998102386, "全帯域_10ms_modulation_mae_db": 0.4496898757162218, "全帯域_40ms_log_rms_mae_db": 0.4563511648472474, "全帯域_40ms_rms_envelope_nrmse": 0.04855659470464923, "全帯域_40ms_envelope_correlation": 0.9980293092296916, "全帯域_40ms_modulation_mae_db": 0.580061646086623, "300–1000 Hz_10ms_log_rms_mae_db": 1.297410595789358, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09589818275973208, "300–1000 Hz_10ms_envelope_correlation": 0.9933873274856431, "300–1000 Hz_10ms_modulation_mae_db": 0.3816551388667582, "300–1000 Hz_40ms_log_rms_mae_db": 0.7391880560235246, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06772961616352093, "300–1000 Hz_40ms_envelope_correlation": 0.9967134082428089, "300–1000 Hz_40ms_modulation_mae_db": 0.6167052183780205, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3277183856707948, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1054589040995258, "1000–3000 Hz_10ms_envelope_correlation": 0.9912306300254032, "1000–3000 Hz_10ms_modulation_mae_db": 0.14595961355005646, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7780761861485289, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05738209774172291, "1000–3000 Hz_40ms_envelope_correlation": 0.9973622389592678, "1000–3000 Hz_40ms_modulation_mae_db": 0.211576480543979, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3002223009870713, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15136685523219823, "3000–6000 Hz_10ms_envelope_correlation": 0.9814721982691191, "3000–6000 Hz_10ms_modulation_mae_db": 0.39598736887073055, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8005723428598203, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09558325745860327, "3000–6000 Hz_40ms_envelope_correlation": 0.9935752725004245, "3000–6000 Hz_40ms_modulation_mae_db": 0.5945207421413371, "mr_stft_db": 5.428560734377686, "band_mae_db": 0.2882872450578862, "rms_gain_error_db": -0.18228054693350515, "output_peak": 0.5391269184375072}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8262717154767412, "全帯域_10ms_rms_envelope_nrmse": 0.07199863093584145, "全帯域_10ms_envelope_correlation": 0.995325416518658, "全帯域_10ms_modulation_mae_db": 0.20861967361775133, "全帯域_40ms_log_rms_mae_db": 0.45279461210926986, "全帯域_40ms_rms_envelope_nrmse": 0.042426227240851065, "全帯域_40ms_envelope_correlation": 0.9983374615123451, "全帯域_40ms_modulation_mae_db": 0.26010569852204085, "300–1000 Hz_10ms_log_rms_mae_db": 1.3565461527757876, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08333681383412778, "300–1000 Hz_10ms_envelope_correlation": 0.9948425905860864, "300–1000 Hz_10ms_modulation_mae_db": 0.15552132432808563, "300–1000 Hz_40ms_log_rms_mae_db": 0.7933244840219293, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04930542615275998, "300–1000 Hz_40ms_envelope_correlation": 0.9981376145793706, "300–1000 Hz_40ms_modulation_mae_db": 0.3194773612966805, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3247263449123186, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11101410820210762, "1000–3000 Hz_10ms_envelope_correlation": 0.9905139899325478, "1000–3000 Hz_10ms_modulation_mae_db": 0.28091573217801585, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8020025167330441, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07103513839273622, "1000–3000 Hz_40ms_envelope_correlation": 0.996175966192631, "1000–3000 Hz_40ms_modulation_mae_db": 0.2897053409469039, "3000–6000 Hz_10ms_log_rms_mae_db": 1.373442014030199, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1623581143122221, "3000–6000 Hz_10ms_envelope_correlation": 0.9771021658453326, "3000–6000 Hz_10ms_modulation_mae_db": 0.31322973374077573, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8666768145940252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09949382138362274, "3000–6000 Hz_40ms_envelope_correlation": 0.9911395921629154, "3000–6000 Hz_40ms_modulation_mae_db": 0.35374638294527205, "mr_stft_db": 5.429137954040935, "band_mae_db": 0.24304073201368762, "rms_gain_error_db": -0.11007121991085526, "output_peak": 0.5150230365208661}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8754260889577178, "全帯域_10ms_rms_envelope_nrmse": 0.07555977321843049, "全帯域_10ms_envelope_correlation": 0.9947929265850568, "全帯域_10ms_modulation_mae_db": 0.43507792428855635, "全帯域_40ms_log_rms_mae_db": 0.4501593069003331, "全帯域_40ms_rms_envelope_nrmse": 0.03962701938161587, "全帯域_40ms_envelope_correlation": 0.9985173785816104, "全帯域_40ms_modulation_mae_db": 0.6130445512096467, "300–1000 Hz_10ms_log_rms_mae_db": 1.357865665019843, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07966649404889521, "300–1000 Hz_10ms_envelope_correlation": 0.9952844139878937, "300–1000 Hz_10ms_modulation_mae_db": 0.3002845989988035, "300–1000 Hz_40ms_log_rms_mae_db": 0.7695937472762211, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04196437928465103, "300–1000 Hz_40ms_envelope_correlation": 0.9986650426286942, "300–1000 Hz_40ms_modulation_mae_db": 0.6222604004748294, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3920020890610747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11786134625713367, "1000–3000 Hz_10ms_envelope_correlation": 0.9891440418371084, "1000–3000 Hz_10ms_modulation_mae_db": 0.3732863652655489, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8390278650636552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07107521264266686, "1000–3000 Hz_40ms_envelope_correlation": 0.9960275443871596, "1000–3000 Hz_40ms_modulation_mae_db": 0.3934518012412677, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3735620185592847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16218091937593093, "3000–6000 Hz_10ms_envelope_correlation": 0.9767590626705434, "3000–6000 Hz_10ms_modulation_mae_db": 0.3432945296324532, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8178866851774449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08978160210548876, "3000–6000 Hz_40ms_envelope_correlation": 0.9921992735068813, "3000–6000 Hz_40ms_modulation_mae_db": 0.37117663772480775, "mr_stft_db": 5.432229417957174, "band_mae_db": 0.22760379672035047, "rms_gain_error_db": -0.08738125318199405, "output_peak": 0.5265015771642484}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8297871040024277, "全帯域_10ms_rms_envelope_nrmse": 0.06414617804731344, "全帯域_10ms_envelope_correlation": 0.9962292797038282, "全帯域_10ms_modulation_mae_db": 0.5577376372963946, "全帯域_40ms_log_rms_mae_db": 0.4216422192482545, "全帯域_40ms_rms_envelope_nrmse": 0.033457586766940856, "全帯域_40ms_envelope_correlation": 0.9989109847115458, "全帯域_40ms_modulation_mae_db": 0.6111355616189028, "300–1000 Hz_10ms_log_rms_mae_db": 1.394589364816414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07177135831036072, "300–1000 Hz_10ms_envelope_correlation": 0.9961435027221386, "300–1000 Hz_10ms_modulation_mae_db": 0.5725446037028762, "300–1000 Hz_40ms_log_rms_mae_db": 0.7924033358260306, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04292712027651109, "300–1000 Hz_40ms_envelope_correlation": 0.9985307440047834, "300–1000 Hz_40ms_modulation_mae_db": 0.6244770557424407, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3416298826559976, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1076675475644415, "1000–3000 Hz_10ms_envelope_correlation": 0.9909676434635093, "1000–3000 Hz_10ms_modulation_mae_db": 0.6957560670027978, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7816440712946174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05813412046523157, "1000–3000 Hz_40ms_envelope_correlation": 0.9974534357471797, "1000–3000 Hz_40ms_modulation_mae_db": 0.923774511660162, "3000–6000 Hz_10ms_log_rms_mae_db": 1.302049911426378, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15197305637889041, "3000–6000 Hz_10ms_envelope_correlation": 0.9797294042351395, "3000–6000 Hz_10ms_modulation_mae_db": 0.5667616445590612, "3000–6000 Hz_40ms_log_rms_mae_db": 0.76247518209114, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08771494109990507, "3000–6000 Hz_40ms_envelope_correlation": 0.9928486440893413, "3000–6000 Hz_40ms_modulation_mae_db": 0.5159709983358716, "mr_stft_db": 5.43599479085685, "band_mae_db": 0.21603113317863054, "rms_gain_error_db": -0.0547343529393366, "output_peak": 0.5695439562658142}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8340365019684286, "全帯域_10ms_rms_envelope_nrmse": 0.07114681898017426, "全帯域_10ms_envelope_correlation": 0.9954844884418447, "全帯域_10ms_modulation_mae_db": 0.5263613548577148, "全帯域_40ms_log_rms_mae_db": 0.4370848447840462, "全帯域_40ms_rms_envelope_nrmse": 0.038929955421374604, "全帯域_40ms_envelope_correlation": 0.9986775796573412, "全帯域_40ms_modulation_mae_db": 0.8090544638537315, "300–1000 Hz_10ms_log_rms_mae_db": 1.3586772400133695, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08563595835236648, "300–1000 Hz_10ms_envelope_correlation": 0.9946317269387175, "300–1000 Hz_10ms_modulation_mae_db": 0.4574990870101927, "300–1000 Hz_40ms_log_rms_mae_db": 0.7857650553392658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.054849979976789184, "300–1000 Hz_40ms_envelope_correlation": 0.997780958180911, "300–1000 Hz_40ms_modulation_mae_db": 0.6984689218274694, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3729854605788063, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11345988716131136, "1000–3000 Hz_10ms_envelope_correlation": 0.9898184341467275, "1000–3000 Hz_10ms_modulation_mae_db": 0.1908756574703102, "1000–3000 Hz_40ms_log_rms_mae_db": 0.813630920633073, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0599713833935792, "1000–3000 Hz_40ms_envelope_correlation": 0.9970906725985039, "1000–3000 Hz_40ms_modulation_mae_db": 0.39433566095272665, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3567335354030599, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15021355406432974, "3000–6000 Hz_10ms_envelope_correlation": 0.9803086978910934, "3000–6000 Hz_10ms_modulation_mae_db": 0.7278310072916884, "3000–6000 Hz_40ms_log_rms_mae_db": 0.801278847651294, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07660057469963925, "3000–6000 Hz_40ms_envelope_correlation": 0.9948569120742577, "3000–6000 Hz_40ms_modulation_mae_db": 0.7618551066534889, "mr_stft_db": 5.44010601257035, "band_mae_db": 0.22084244598658684, "rms_gain_error_db": -0.12634308488505733, "output_peak": 0.6115871988074393}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8234043342743508, "全帯域_10ms_rms_envelope_nrmse": 0.0678301176858133, "全帯域_10ms_envelope_correlation": 0.9958594019700232, "全帯域_10ms_modulation_mae_db": 0.4803426871776482, "全帯域_40ms_log_rms_mae_db": 0.4372167381564708, "全帯域_40ms_rms_envelope_nrmse": 0.036859670102475504, "全帯域_40ms_envelope_correlation": 0.9987701312195197, "全帯域_40ms_modulation_mae_db": 0.6361403766220408, "300–1000 Hz_10ms_log_rms_mae_db": 1.379581117840713, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07991185124726988, "300–1000 Hz_10ms_envelope_correlation": 0.9952688999718664, "300–1000 Hz_10ms_modulation_mae_db": 0.5504199851069272, "300–1000 Hz_40ms_log_rms_mae_db": 0.7436476387334638, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04273666583400038, "300–1000 Hz_40ms_envelope_correlation": 0.9986276462464898, "300–1000 Hz_40ms_modulation_mae_db": 0.673792834087609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.294126806600143, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10100432217297724, "1000–3000 Hz_10ms_envelope_correlation": 0.992199852320084, "1000–3000 Hz_10ms_modulation_mae_db": 0.12897607440648404, "1000–3000 Hz_40ms_log_rms_mae_db": 0.756697425819619, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05464499149912496, "1000–3000 Hz_40ms_envelope_correlation": 0.9979285282072027, "1000–3000 Hz_40ms_modulation_mae_db": 0.27151094133516956, "3000–6000 Hz_10ms_log_rms_mae_db": 1.332804335929086, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15759486836691194, "3000–6000 Hz_10ms_envelope_correlation": 0.9781243843492923, "3000–6000 Hz_10ms_modulation_mae_db": 0.5025900240695828, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7868319162997252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09316814479939077, "3000–6000 Hz_40ms_envelope_correlation": 0.9917138707905203, "3000–6000 Hz_40ms_modulation_mae_db": 0.5287876836800326, "mr_stft_db": 5.425019296243022, "band_mae_db": 0.20478837461874097, "rms_gain_error_db": -0.09972495204385348, "output_peak": 0.515145985962732}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 91600, "output_samples": 91600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.471842882522276, "spectral_convergence": 0.2586398647568361, "spectral_convergence_status": "ok", "amplitude_floor": 1.4112644553101518e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.236619740101511, "spectral_convergence": 0.2530208844343124, "spectral_convergence_status": "ok", "amplitude_floor": 1.1975531593072457e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5953248344060516, "spectral_convergence": 0.2668376318212157, "spectral_convergence_status": "ok", "amplitude_floor": 1.114415465542996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.564075826256522e-05, "output_power": 6.253703170165351e-05, "delta_db": 0.5074422391044062, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010000129923479963, "output_power": 0.0009797326956383912, "delta_db": -0.08898040902709604, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018259392170112694, "output_power": 0.0018291865056922364, "delta_db": 0.007716724585266656, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005127219920465177, "output_power": 0.0004973095434586674, "delta_db": -0.13255152415720162, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00033320627903786135, "output_power": 0.00031634086996156174, "delta_db": -0.22557872153818048, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002817706196064759, "output_power": 0.000263036182509183, "delta_db": -0.29880214072085465, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2101786265221676, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.06674816652080302, "pitch": null, "output_rms_dbfs": -24.036071445768936, "output_peak": 0.5569887868377952, "reference_rms_dbfs": -23.96932327924813, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.434595819009947}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 91600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00020825955187282386, "log_rms_mae_db": 0.8429217028249015, "rms_envelope_nrmse": 0.0680008516088581, "envelope_correlation": 0.9957821482650732, "rms_gain_error_db": -0.06673580092030565, "reference_mean_power": 0.004012756910556895, "output_mean_power": 0.003951566284043822}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 2.2120307480210933e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008889321025104637, "output_power": 0.0008472035838024522, "reference_active": true, "delta_db": -0.2088080617020974}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022980725896418495, "output_power": 0.00020603172470031852, "reference_active": true, "delta_db": -0.4742964466242817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1131932747773025e-05, "output_power": 4.1672749400270236e-05, "reference_active": true, "delta_db": 0.05673037358818713}], "active_band_count": 3, "active_band_mae_db": 0.24661162730485542}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.0001984160011662782, "log_rms_mae_db": 0.43202746523018076, "rms_envelope_nrmse": 0.034312186835629074, "envelope_correlation": 0.9988822204432637, "rms_gain_error_db": -0.06673577215729949, "reference_mean_power": 0.0040337408532874226, "output_mean_power": 0.003972230268435422}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 2.0115882799069226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008024963383362849, "output_power": 0.0007645648727291518, "reference_active": true, "delta_db": -0.21028718606923535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011463438169489824, "output_power": 0.00010469187715014434, "reference_active": true, "delta_db": -0.39401905980440555}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4945953602736586e-06, "output_power": 3.0313658292145496e-06, "reference_active": true, "delta_db": -0.6175854520939049}], "active_band_count": 3, "active_band_mae_db": 0.4072972326558486}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00019054595180557928, "log_rms_mae_db": 1.378905350751675, "rms_envelope_nrmse": 0.07976169285276245, "envelope_correlation": 0.9952427042411239, "rms_gain_error_db": -0.04078158779236343, "reference_mean_power": 0.0025757888112863028, "output_mean_power": 0.002551714571166099}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.7595886306573375e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007279305516452219, "output_power": 0.0007010608811182008, "reference_active": true, "delta_db": -0.16334212989303293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019010480993234813, "output_power": 0.00017035974878012569, "reference_active": true, "delta_db": -0.47626114323303825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6294378423488108e-05, "output_power": 3.015661149510209e-05, "reference_active": true, "delta_db": 0.5951963213293257}], "active_band_count": 3, "active_band_mae_db": 0.4115998648184656}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.00018019065505981728, "log_rms_mae_db": 0.8092162543609015, "rms_envelope_nrmse": 0.041944106229267045, "envelope_correlation": 0.9986178806979206, "rms_gain_error_db": -0.040781744921863435, "reference_mean_power": 0.0025893611436675294, "output_mean_power": 0.0025651599588838235}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.6269545971301712e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006780414214466183, "output_power": 0.0006530264576032469, "reference_active": true, "delta_db": -0.16325448431259687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.548437806715927e-05, "output_power": 8.721296818685273e-05, "reference_active": true, "delta_db": -0.39351256262519346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9327510117604047e-06, "output_power": 2.5479471319641253e-06, "reference_active": true, "delta_db": -0.6108478082332078}], "active_band_count": 3, "active_band_mae_db": 0.38920495172366604}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.327720993461315e-05, "log_rms_mae_db": 1.3647110277045018, "rms_envelope_nrmse": 0.10036117869983811, "envelope_correlation": 0.9921328419112964, "rms_gain_error_db": -0.1665323490650334, "reference_mean_power": 0.0007122971869413385, "output_mean_power": 0.0006855008447197039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 4.689927289528067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023362430343497353, "output_power": 0.00021529710263182422, "reference_active": true, "delta_db": -0.35479834183626224}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.797902128989837e-05, "output_power": 3.284694995928671e-05, "reference_active": true, "delta_db": -0.630487200201342}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.042184778585592e-05, "output_power": 1.0020674434222024e-05, "reference_active": true, "delta_db": -0.17047773488848206}], "active_band_count": 3, "active_band_mae_db": 0.3852544256420288}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 8.747046012456969e-05, "log_rms_mae_db": 0.7855792224912868, "rms_envelope_nrmse": 0.05747073260586687, "envelope_correlation": 0.9974263909608585, "rms_gain_error_db": -0.16655162288022834, "reference_mean_power": 0.0007159674763911086, "output_mean_power": 0.0006890300014182466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 4.328159646639031e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021700970165722394, "output_power": 0.00020093222250225473, "reference_active": true, "delta_db": -0.3342956191306096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.702508733600319e-05, "output_power": 1.4969486177059562e-05, "reference_active": true, "delta_db": -0.558824548761305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.679451250427792e-07, "output_power": 7.107447313110727e-07, "reference_active": true, "delta_db": -0.867786192256534}], "active_band_count": 3, "active_band_mae_db": 0.5869687867161496}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.039773028293347e-05, "log_rms_mae_db": 1.347249121769675, "rms_envelope_nrmse": 0.14665032280241475, "envelope_correlation": 0.9815314766695558, "rms_gain_error_db": -0.3120638513326798, "reference_mean_power": 0.0003237352885604671, "output_mean_power": 0.0003012892642368484}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.2480235329901466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.70846061617512e-05, "output_power": 5.438015508377684e-05, "reference_active": true, "delta_db": -0.2107856732155684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9595191589145483e-05, "output_power": 1.7573772060650264e-05, "reference_active": true, "delta_db": -0.4728452496987418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.074204859780537e-06, "output_power": 4.6692323502674404e-06, "reference_active": true, "delta_db": -0.36122511093520093}], "active_band_count": 3, "active_band_mae_db": 0.34828534461650373}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 6.399293414479695e-05, "log_rms_mae_db": 0.7967707155140565, "rms_envelope_nrmse": 0.07815256332461751, "envelope_correlation": 0.995033678932814, "rms_gain_error_db": -0.3121133743855095, "reference_mean_power": 0.0003253818281247376, "output_mean_power": 0.0003028181887168249}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.1005868407421181e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4497636322434906e-05, "output_power": 5.1625565547702285e-05, "reference_active": true, "delta_db": -0.23512844117029907}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.385752117634814e-06, "output_power": 8.348939836604786e-06, "reference_active": true, "delta_db": -0.5083774860510376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7591737704275347e-07, "output_power": 3.516856601501205e-07, "reference_active": true, "delta_db": -0.28937741258353167}], "active_band_count": 3, "active_band_mae_db": 0.34429444660162273}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8962901445863074, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d662a41e6ad65ba77511d2c85d4781ea66c0d2ad75b41e995147346ed4fe3ae2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9907880952438981, "standard_deviation": 0.0010682840971846708, "minimum": 0.9891440418371084, "maximum": 0.992199852320084, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3475913757927667, "standard_deviation": 0.03151925677773541, "minimum": 1.294126806600143, "maximum": 1.3920020890610747, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3468586057988806, "standard_deviation": 0.20421880139485926, "minimum": 0.12897607440648404, "maximum": 0.6957560670027978, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.10862753721557739, "standard_deviation": 0.006146206884200406, "minimum": 0.10036117869983811, "maximum": 0.11786134625713367, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970309796235703, "standard_deviation": 0.0006604222862377317, "minimum": 0.9960275443871596, "maximum": 0.9979285282072027, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7974324682236321, "standard_deviation": 0.02694126991071496, "minimum": 0.756697425819619, "maximum": 0.8390278650636552, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.44431394801126584, "standard_deviation": 0.22827309714037441, "minimum": 0.211576480543979, "maximum": 0.923774511660162, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06196184436681812, "standard_deviation": 0.006489841417896852, "minimum": 0.05464499149912496, "maximum": 0.07107521264266686, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9792966657900233, "standard_deviation": 0.0018350057897110938, "minimum": 0.9767590626705434, "maximum": 0.9815314766695558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3433953510546135, "standard_deviation": 0.02928811340201792, "minimum": 1.3002223009870713, "maximum": 1.3735620185592847, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.45367015628118446, "standard_deviation": 0.1399915017831696, "minimum": 0.31322973374077573, "maximum": 0.7278310072916884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15447940122900983, "standard_deviation": 0.005701119656430683, "minimum": 0.14665032280241475, "maximum": 0.1623581143122221, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9931253805698654, "standard_deviation": 0.0014128344395438546, "minimum": 0.9911395921629154, "maximum": 0.995033678932814, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8076012899492522, "standard_deviation": 0.030936096814206185, "minimum": 0.76247518209114, "maximum": 0.8666768145940252, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5014225705810587, "standard_deviation": 0.142669960418911, "minimum": 0.34429444660162273, "maximum": 0.7618551066534889, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08802062199516211, "standard_deviation": 0.008152498280320974, "minimum": 0.07660057469963925, "maximum": 0.09949382138362274, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9949184195475884, "standard_deviation": 0.0007958389907016857, "minimum": 0.9933873274856431, "maximum": 0.9961435027221386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3633832252448435, "standard_deviation": 0.03007882376425034, "minimum": 1.297410595789358, "maximum": 1.394589364816414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4125412791549004, "standard_deviation": 0.13608988619768947, "minimum": 0.15552132432808563, "maximum": 0.5725446037028762, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08286673326772867, "standard_deviation": 0.007031715692415994, "minimum": 0.07177135831036072, "maximum": 0.09589818275973208, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981177060830331, "standard_deviation": 0.0006685181455332125, "minimum": 0.9967134082428089, "maximum": 0.9986650426286942, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7721961612499993, "standard_deviation": 0.026969781054456417, "minimum": 0.7391880560235246, "maximum": 0.8092162543609015, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5902361450629406, "standard_deviation": 0.15601610505922145, "minimum": 0.3194773612966805, "maximum": 0.7775024169728098, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04956544983379514, "standard_deviation": 0.009217615716657374, "minimum": 0.041944106229267045, "maximum": 0.06772961616352093, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23188683515467085, "standard_deviation": 0.026878978068959097, "minimum": 0.20478837461874097, "maximum": 0.2882872450578862, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.432428258706648, "standard_deviation": 0.004764679895339432, "minimum": 5.425019296243022, "maximum": 5.44010601257035, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.546811846005476, "standard_deviation": 0.03238133278070975, "minimum": 0.5150230365208661, "maximum": 0.6115871988074393, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11129624148741182, "standard_deviation": 0.044462657407117814, "minimum": -0.18228054693350515, "maximum": -0.0547343529393366, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954834036307453, "standard_deviation": 0.0004702730198382246, "minimum": 0.9947929265850568, "maximum": 0.9962292797038282, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8423231755973775, "standard_deviation": 0.017762928312080657, "minimum": 0.8234043342743508, "maximum": 0.8754260889577178, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4280882188013494, "standard_deviation": 0.13055340838779775, "minimum": 0.20861967361775133, "maximum": 0.5577376372963946, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07090553428349433, "standard_deviation": 0.004062499675866074, "minimum": 0.06414617804731344, "maximum": 0.07604746726840243, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9985806233420763, "standard_deviation": 0.0002966895352498505, "minimum": 0.9980293092296916, "maximum": 0.9989109847115458, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44262577494670496, "standard_deviation": 0.012446683382143234, "minimum": 0.4216422192482545, "maximum": 0.4563511648472474, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5775592497664682, "standard_deviation": 0.17116220047626526, "minimum": 0.26010569852204085, "maximum": 0.8090544638537315, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03959012257432999, "standard_deviation": 0.004936300589816062, "minimum": 0.033457586766940856, "maximum": 0.04855659470464923, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0014", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.803248882293701, "predicted_mos": 2.803248882293701, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "f0b741a4c406cdc45e1e9b48482c6c9574add341147d484418b56e8336fe4b30", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.695357700344175, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8204545284283772, "全帯域_10ms_rms_envelope_nrmse": 0.0688790831220481, "全帯域_10ms_envelope_correlation": 0.9957091102864258, "全帯域_10ms_modulation_mae_db": 0.4528384096303089, "全帯域_40ms_log_rms_mae_db": 0.4287175672714079, "全帯域_40ms_rms_envelope_nrmse": 0.03599188185672352, "全帯域_40ms_envelope_correlation": 0.9988122246491989, "全帯域_40ms_modulation_mae_db": 0.5936237695557376, "300–1000 Hz_10ms_log_rms_mae_db": 1.4171390099155707, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08294554433695009, "300–1000 Hz_10ms_envelope_correlation": 0.9948947128772926, "300–1000 Hz_10ms_modulation_mae_db": 0.3804945479175356, "300–1000 Hz_40ms_log_rms_mae_db": 0.7481023810809877, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04610514821299589, "300–1000 Hz_40ms_envelope_correlation": 0.9983845647512296, "300–1000 Hz_40ms_modulation_mae_db": 0.6190864697590758, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4593138317215946, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11385699362727889, "1000–3000 Hz_10ms_envelope_correlation": 0.9901664313305489, "1000–3000 Hz_10ms_modulation_mae_db": 0.33259378684857555, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8957063338518877, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06803298280112165, "1000–3000 Hz_40ms_envelope_correlation": 0.9967685373699922, "1000–3000 Hz_40ms_modulation_mae_db": 0.47339816743698454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.42614099814294, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1513619591434766, "3000–6000 Hz_10ms_envelope_correlation": 0.9798506583357064, "3000–6000 Hz_10ms_modulation_mae_db": 0.3414206500664201, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8442726553411123, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08554591784444984, "3000–6000 Hz_40ms_envelope_correlation": 0.9928774070080326, "3000–6000 Hz_40ms_modulation_mae_db": 0.28594305283893345, "mr_stft_db": 5.503721699504539, "band_mae_db": 0.1987610817889697, "rms_gain_error_db": -0.09241860283307396, "output_peak": 0.5358252616616783}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8585193466404187, "全帯域_10ms_rms_envelope_nrmse": 0.08023663396888774, "全帯域_10ms_envelope_correlation": 0.9941470198967022, "全帯域_10ms_modulation_mae_db": 0.28726097852021, "全帯域_40ms_log_rms_mae_db": 0.4702553601521473, "全帯域_40ms_rms_envelope_nrmse": 0.040599827134462065, "全帯域_40ms_envelope_correlation": 0.9984838512640216, "全帯域_40ms_modulation_mae_db": 0.6777314304085551, "300–1000 Hz_10ms_log_rms_mae_db": 1.4005181484574016, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08653550333188408, "300–1000 Hz_10ms_envelope_correlation": 0.9944749577844257, "300–1000 Hz_10ms_modulation_mae_db": 0.20565435646875918, "300–1000 Hz_40ms_log_rms_mae_db": 0.7569055484999009, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043062700894321196, "300–1000 Hz_40ms_envelope_correlation": 0.9986702675849806, "300–1000 Hz_40ms_modulation_mae_db": 0.5812267487458851, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4675473772544412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12293996870446898, "1000–3000 Hz_10ms_envelope_correlation": 0.9879572300190393, "1000–3000 Hz_10ms_modulation_mae_db": 0.30989306278011824, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8859354472126671, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06757406347743448, "1000–3000 Hz_40ms_envelope_correlation": 0.9961657425730328, "1000–3000 Hz_40ms_modulation_mae_db": 0.23064657965381452, "3000–6000 Hz_10ms_log_rms_mae_db": 1.403559624996863, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15516592462860354, "3000–6000 Hz_10ms_envelope_correlation": 0.9787017686516605, "3000–6000 Hz_10ms_modulation_mae_db": 0.21400597743683256, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8173316305448809, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08852480251081427, "3000–6000 Hz_40ms_envelope_correlation": 0.9924252656363334, "3000–6000 Hz_40ms_modulation_mae_db": 0.16814857176895578, "mr_stft_db": 5.499544420645731, "band_mae_db": 0.1721072181036463, "rms_gain_error_db": -0.09174781131241969, "output_peak": 0.5137030733440797}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8312812746764398, "全帯域_10ms_rms_envelope_nrmse": 0.06932832620054395, "全帯域_10ms_envelope_correlation": 0.9956097538818942, "全帯域_10ms_modulation_mae_db": 0.3311275300997705, "全帯域_40ms_log_rms_mae_db": 0.41317785998955225, "全帯域_40ms_rms_envelope_nrmse": 0.03305436539933669, "全帯域_40ms_envelope_correlation": 0.99897167187411, "全帯域_40ms_modulation_mae_db": 0.4362833283725605, "300–1000 Hz_10ms_log_rms_mae_db": 1.4296232479271314, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07586557029262268, "300–1000 Hz_10ms_envelope_correlation": 0.9957068331490128, "300–1000 Hz_10ms_modulation_mae_db": 0.27842051299318876, "300–1000 Hz_40ms_log_rms_mae_db": 0.7693587408007998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037575921223666195, "300–1000 Hz_40ms_envelope_correlation": 0.9989125138261371, "300–1000 Hz_40ms_modulation_mae_db": 0.4167286774967996, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4223367403599734, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11023444486958107, "1000–3000 Hz_10ms_envelope_correlation": 0.990312395118935, "1000–3000 Hz_10ms_modulation_mae_db": 0.1865975984613449, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8288053333332265, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06233367897163653, "1000–3000 Hz_40ms_envelope_correlation": 0.9966761676972337, "1000–3000 Hz_40ms_modulation_mae_db": 0.22620324605026232, "3000–6000 Hz_10ms_log_rms_mae_db": 1.424266559781547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17383207145319637, "3000–6000 Hz_10ms_envelope_correlation": 0.974214375176385, "3000–6000 Hz_10ms_modulation_mae_db": 0.6591743844505961, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8504642472830837, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1107273853597853, "3000–6000 Hz_40ms_envelope_correlation": 0.989718181571117, "3000–6000 Hz_40ms_modulation_mae_db": 0.5982307149278046, "mr_stft_db": 5.519928102660224, "band_mae_db": 0.19200601710758827, "rms_gain_error_db": -0.07582960466799933, "output_peak": 0.5532071625777708}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8287910931862612, "全帯域_10ms_rms_envelope_nrmse": 0.0711625817340183, "全帯域_10ms_envelope_correlation": 0.9954429715579649, "全帯域_10ms_modulation_mae_db": 0.3582758055737439, "全帯域_40ms_log_rms_mae_db": 0.43189551072158433, "全帯域_40ms_rms_envelope_nrmse": 0.04275289005732821, "全帯域_40ms_envelope_correlation": 0.9983124094173546, "全帯域_40ms_modulation_mae_db": 0.26364595869260826, "300–1000 Hz_10ms_log_rms_mae_db": 1.4466682245387428, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08676410743759444, "300–1000 Hz_10ms_envelope_correlation": 0.9943948566101479, "300–1000 Hz_10ms_modulation_mae_db": 0.21668597507184553, "300–1000 Hz_40ms_log_rms_mae_db": 0.834693609169993, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05702530254964689, "300–1000 Hz_40ms_envelope_correlation": 0.9974655087176585, "300–1000 Hz_40ms_modulation_mae_db": 0.07556947567088972, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3943430107271872, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1099506093057764, "1000–3000 Hz_10ms_envelope_correlation": 0.9905935871359768, "1000–3000 Hz_10ms_modulation_mae_db": 0.546436172339797, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8206023656477484, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06098831873445822, "1000–3000 Hz_40ms_envelope_correlation": 0.9971913643187182, "1000–3000 Hz_40ms_modulation_mae_db": 0.6331809977369687, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4306084594150394, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15113734353475144, "3000–6000 Hz_10ms_envelope_correlation": 0.9802001596716873, "3000–6000 Hz_10ms_modulation_mae_db": 0.8221763006889725, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8291929728786954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08159062952278888, "3000–6000 Hz_40ms_envelope_correlation": 0.9943080045805496, "3000–6000 Hz_40ms_modulation_mae_db": 0.7583618693815734, "mr_stft_db": 5.501360599521519, "band_mae_db": 0.2130670428421232, "rms_gain_error_db": -0.10476411931254664, "output_peak": 0.543218422801965}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8394702777223496, "全帯域_10ms_rms_envelope_nrmse": 0.06799847791061397, "全帯域_10ms_envelope_correlation": 0.9958242294422902, "全帯域_10ms_modulation_mae_db": 0.30597174288895507, "全帯域_40ms_log_rms_mae_db": 0.4381252158737537, "全帯域_40ms_rms_envelope_nrmse": 0.038505961635079855, "全帯域_40ms_envelope_correlation": 0.998623695432725, "全帯域_40ms_modulation_mae_db": 0.4512931273701504, "300–1000 Hz_10ms_log_rms_mae_db": 1.4410844346896388, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07997991991278379, "300–1000 Hz_10ms_envelope_correlation": 0.9952473886145412, "300–1000 Hz_10ms_modulation_mae_db": 0.4421956637386866, "300–1000 Hz_40ms_log_rms_mae_db": 0.8230350227042716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04618065134429219, "300–1000 Hz_40ms_envelope_correlation": 0.9983688385663718, "300–1000 Hz_40ms_modulation_mae_db": 0.33524044012248116, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4456384533315132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10159283410032058, "1000–3000 Hz_10ms_envelope_correlation": 0.9918705118194201, "1000–3000 Hz_10ms_modulation_mae_db": 0.21086852746869653, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8572417573603182, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.04960377099634432, "1000–3000 Hz_40ms_envelope_correlation": 0.9980793676685026, "1000–3000 Hz_40ms_modulation_mae_db": 0.19305331556446945, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3908813365754014, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15838590438461322, "3000–6000 Hz_10ms_envelope_correlation": 0.977943847740865, "3000–6000 Hz_10ms_modulation_mae_db": 0.7849425710362818, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8277291847312896, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09554530319757795, "3000–6000 Hz_40ms_envelope_correlation": 0.9914264529985504, "3000–6000 Hz_40ms_modulation_mae_db": 0.575734708346897, "mr_stft_db": 5.497239158846855, "band_mae_db": 0.20577751274482647, "rms_gain_error_db": -0.087632859673715, "output_peak": 0.5375524700220461}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8353290478041002, "全帯域_10ms_rms_envelope_nrmse": 0.0680138849159102, "全帯域_10ms_envelope_correlation": 0.995767199610516, "全帯域_10ms_modulation_mae_db": 0.3217769596882672, "全帯域_40ms_log_rms_mae_db": 0.4222837002277448, "全帯域_40ms_rms_envelope_nrmse": 0.03538744204256405, "全帯域_40ms_envelope_correlation": 0.9987851186523881, "全帯域_40ms_modulation_mae_db": 0.34843665779933897, "300–1000 Hz_10ms_log_rms_mae_db": 1.4223990649085156, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07909112464099179, "300–1000 Hz_10ms_envelope_correlation": 0.9953302957445874, "300–1000 Hz_10ms_modulation_mae_db": 0.21620887627933327, "300–1000 Hz_40ms_log_rms_mae_db": 0.7641522940995648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04515424165364139, "300–1000 Hz_40ms_envelope_correlation": 0.9983966218092111, "300–1000 Hz_40ms_modulation_mae_db": 0.24721679769997917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3927360854166533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11569683746894904, "1000–3000 Hz_10ms_envelope_correlation": 0.9893920176546102, "1000–3000 Hz_10ms_modulation_mae_db": 0.5630533683916078, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8291585504755291, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06607702576351987, "1000–3000 Hz_40ms_envelope_correlation": 0.9964102241890738, "1000–3000 Hz_40ms_modulation_mae_db": 0.6505252881946765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4428128298544274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14826243883314516, "3000–6000 Hz_10ms_envelope_correlation": 0.9806177398887816, "3000–6000 Hz_10ms_modulation_mae_db": 0.45938608224437233, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8237281497292691, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07962773431887461, "3000–6000 Hz_40ms_envelope_correlation": 0.9939951481171792, "3000–6000 Hz_40ms_modulation_mae_db": 0.40736391173838166, "mr_stft_db": 5.504857833533272, "band_mae_db": 0.14290647148691168, "rms_gain_error_db": -0.0565355095963084, "output_peak": 0.5322272020771863}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8219505124312512, "全帯域_10ms_rms_envelope_nrmse": 0.06389343524955216, "全帯域_10ms_envelope_correlation": 0.9964703489209336, "全帯域_10ms_modulation_mae_db": 0.42095469288908854, "全帯域_40ms_log_rms_mae_db": 0.4353937886720335, "全帯域_40ms_rms_envelope_nrmse": 0.03564019227315479, "全帯域_40ms_envelope_correlation": 0.9990066294252271, "全帯域_40ms_modulation_mae_db": 0.45260889958773554, "300–1000 Hz_10ms_log_rms_mae_db": 1.4837308540350658, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07420603117831548, "300–1000 Hz_10ms_envelope_correlation": 0.9960485459627167, "300–1000 Hz_10ms_modulation_mae_db": 0.34312814614544207, "300–1000 Hz_40ms_log_rms_mae_db": 0.8118450912532426, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043661019265149044, "300–1000 Hz_40ms_envelope_correlation": 0.9987055031228833, "300–1000 Hz_40ms_modulation_mae_db": 0.624721352035784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4714706806444118, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12205845055605781, "1000–3000 Hz_10ms_envelope_correlation": 0.98820649947416, "1000–3000 Hz_10ms_modulation_mae_db": 0.33904847933542975, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8793174500774051, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07220245738466591, "1000–3000 Hz_40ms_envelope_correlation": 0.995746717810013, "1000–3000 Hz_40ms_modulation_mae_db": 0.43334353130494335, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4435190834426306, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15859442936462956, "3000–6000 Hz_10ms_envelope_correlation": 0.9783610923931035, "3000–6000 Hz_10ms_modulation_mae_db": 0.7848359018738619, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8713143263147438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08790869573446032, "3000–6000 Hz_40ms_envelope_correlation": 0.9936386701188269, "3000–6000 Hz_40ms_modulation_mae_db": 0.7973259824584761, "mr_stft_db": 5.516188031249219, "band_mae_db": 0.2118506036941469, "rms_gain_error_db": -0.14620836721492952, "output_peak": 0.5195451944106615}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8323723268366408, "全帯域_10ms_rms_envelope_nrmse": 0.06872648455378597, "全帯域_10ms_envelope_correlation": 0.9957118020003415, "全帯域_10ms_modulation_mae_db": 0.10600493528271161, "全帯域_40ms_log_rms_mae_db": 0.4454046031745244, "全帯域_40ms_rms_envelope_nrmse": 0.03591835153550854, "全帯域_40ms_envelope_correlation": 0.9988037298045727, "全帯域_40ms_modulation_mae_db": 0.2732904868158728, "300–1000 Hz_10ms_log_rms_mae_db": 1.481576706424441, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07874614222295627, "300–1000 Hz_10ms_envelope_correlation": 0.995408398595054, "300–1000 Hz_10ms_modulation_mae_db": 0.1832647294155727, "300–1000 Hz_40ms_log_rms_mae_db": 0.8008004159671565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04510274238828414, "300–1000 Hz_40ms_envelope_correlation": 0.9984663444538998, "300–1000 Hz_40ms_modulation_mae_db": 0.1957747446442181, "1000–3000 Hz_10ms_log_rms_mae_db": 1.445023945409587, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1147667391302128, "1000–3000 Hz_10ms_envelope_correlation": 0.9895146464273842, "1000–3000 Hz_10ms_modulation_mae_db": 0.25242264957009014, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8458079773395663, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0652357711552557, "1000–3000 Hz_40ms_envelope_correlation": 0.9964189424414706, "1000–3000 Hz_40ms_modulation_mae_db": 0.20732431073839305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4467764691720437, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1583825325127647, "3000–6000 Hz_10ms_envelope_correlation": 0.9787711819449058, "3000–6000 Hz_10ms_modulation_mae_db": 1.0382547287799528, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8489592850840283, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0892361149989285, "3000–6000 Hz_40ms_envelope_correlation": 0.993870020163607, "3000–6000 Hz_40ms_modulation_mae_db": 1.1644088060178315, "mr_stft_db": 5.4964635132431665, "band_mae_db": 0.22030623626284696, "rms_gain_error_db": -0.08801894785913003, "output_peak": 0.5305659563371626}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 91600, "output_samples": 91600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.58888518220821, "spectral_convergence": 0.2597228845660181, "spectral_convergence_status": "ok", "amplitude_floor": 1.4112644553101518e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.326468803804368, "spectral_convergence": 0.25296619711491797, "spectral_convergence_status": "ok", "amplitude_floor": 1.1975531593072457e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.595811112501041, "spectral_convergence": 0.26104685642068376, "spectral_convergence_status": "ok", "amplitude_floor": 1.114415465542996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.564075826256522e-05, "output_power": 6.265227784331508e-05, "delta_db": 0.5154382537329449, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010000129923479963, "output_power": 0.0009781228059635454, "delta_db": -0.09612257410784476, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018259392170112694, "output_power": 0.001803637445220007, "delta_db": -0.05337073284318584, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005127219920465177, "output_power": 0.0004889667639150053, "delta_db": -0.2060260568609627, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00033320627903786135, "output_power": 0.00031249575903285874, "delta_db": -0.27869048997280194, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002817706196064759, "output_power": 0.0002789997850392881, "delta_db": -0.042918383216078076, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1987610817889697, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09241860283307396, "pitch": null, "output_rms_dbfs": -24.061741882081208, "output_peak": 0.5358252616616783, "reference_rms_dbfs": -23.96932327924813, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.503721699504539}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 91600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00020825955187282386, "log_rms_mae_db": 0.8204545284283772, "rms_envelope_nrmse": 0.0688790831220481, "envelope_correlation": 0.9957091102864258, "rms_gain_error_db": -0.09240263367595077, "reference_mean_power": 0.004012756910556895, "output_mean_power": 0.003928281375788356}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 2.2120307480210933e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008889321025104637, "output_power": 0.0008513487667279843, "reference_active": true, "delta_db": -0.18761079163457822}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022980725896418495, "output_power": 0.00021126466913366361, "reference_active": true, "delta_db": -0.3653686886149188}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1131932747773025e-05, "output_power": 3.4168470774138926e-05, "reference_active": true, "delta_db": -0.8055357486414298}], "active_band_count": 3, "active_band_mae_db": 0.4528384096303089}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.0001984160011662782, "log_rms_mae_db": 0.4287175672714079, "rms_envelope_nrmse": 0.03599188185672352, "envelope_correlation": 0.9988122246491989, "rms_gain_error_db": -0.092426481289982, "reference_mean_power": 0.0040337408532874226, "output_mean_power": 0.00394880188652085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 2.0115882799069226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008024963383362849, "output_power": 0.0007706861778192544, "reference_active": true, "delta_db": -0.1756548943741364}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011463438169489824, "output_power": 0.00010312642285956406, "reference_active": true, "delta_db": -0.45944939195098755}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4945953602736586e-06, "output_power": 2.684232891547949e-06, "reference_active": true, "delta_db": -1.1457670223420888}], "active_band_count": 3, "active_band_mae_db": 0.5936237695557376}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00019054595180557928, "log_rms_mae_db": 1.4171390099155707, "rms_envelope_nrmse": 0.08294554433695009, "envelope_correlation": 0.9948947128772926, "rms_gain_error_db": -0.083780811638951, "reference_mean_power": 0.0025757888112863028, "output_mean_power": 0.0025265748641825534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.7595886306573375e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007279305516452219, "output_power": 0.0007091478499143572, "reference_active": true, "delta_db": -0.11353156857212825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019010480993234813, "output_power": 0.00016904378910070184, "reference_active": true, "delta_db": -0.5099388647250834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6294378423488108e-05, "output_power": 2.3337889965304963e-05, "reference_active": true, "delta_db": -0.5180132104553952}], "active_band_count": 3, "active_band_mae_db": 0.3804945479175356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.00018019065505981728, "log_rms_mae_db": 0.7481023810809877, "rms_envelope_nrmse": 0.04610514821299589, "envelope_correlation": 0.9983845647512296, "rms_gain_error_db": -0.08377691071559598, "reference_mean_power": 0.0025893611436675294, "output_mean_power": 0.0025398901600827716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.6269545971301712e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006780414214466183, "output_power": 0.0006614422346955062, "reference_active": true, "delta_db": -0.10764303438825334}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.548437806715927e-05, "output_power": 8.33101939999591e-05, "reference_active": true, "delta_db": -0.5923417789713983}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9327510117604047e-06, "output_power": 2.24671370074013e-06, "reference_active": true, "delta_db": -1.1572745959175756}], "active_band_count": 3, "active_band_mae_db": 0.6190864697590758}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.327720993461315e-05, "log_rms_mae_db": 1.4593138317215946, "rms_envelope_nrmse": 0.11385699362727889, "envelope_correlation": 0.9901664313305489, "rms_gain_error_db": -0.27636020932681854, "reference_mean_power": 0.0007122971869413385, "output_mean_power": 0.0006683827121678167}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 4.689927289528067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023362430343497353, "output_power": 0.0002175710296854955, "reference_active": true, "delta_db": -0.30916952408923903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.797902128989837e-05, "output_power": 3.5732651272959725e-05, "reference_active": true, "delta_db": -0.26478527924974704}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.042184778585592e-05, "output_power": 9.452834844233962e-06, "reference_active": true, "delta_db": -0.4238265572067406}], "active_band_count": 3, "active_band_mae_db": 0.33259378684857555}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 8.747046012456969e-05, "log_rms_mae_db": 0.8957063338518877, "rms_envelope_nrmse": 0.06803298280112165, "envelope_correlation": 0.9967685373699922, "rms_gain_error_db": -0.27649857952126233, "reference_mean_power": 0.0007159674763911086, "output_mean_power": 0.000671805316627439}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 4.328159646639031e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021700970165722394, "output_power": 0.0002028305304632936, "reference_active": true, "delta_db": -0.2934582342575471}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.702508733600319e-05, "output_power": 1.6778389070273934e-05, "reference_active": true, "delta_db": -0.06339087518541901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.679451250427792e-07, "output_power": 6.794503534951007e-07, "reference_active": true, "delta_db": -1.0633453928679875}], "active_band_count": 3, "active_band_mae_db": 0.47339816743698454}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.039773028293347e-05, "log_rms_mae_db": 1.42614099814294, "rms_envelope_nrmse": 0.1513619591434766, "envelope_correlation": 0.9798506583357064, "rms_gain_error_db": -0.0480809868728729, "reference_mean_power": 0.0003237352885604671, "output_mean_power": 0.00032017096376052776}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.2480235329901466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.70846061617512e-05, "output_power": 5.673419554347542e-05, "reference_active": true, "delta_db": -0.026741079055980416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9595191589145483e-05, "output_power": 1.993021132679303e-05, "reference_active": true, "delta_db": 0.07362389610422915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.074204859780537e-06, "output_power": 4.101836066468137e-06, "reference_active": true, "delta_db": -0.9238969750390507}], "active_band_count": 3, "active_band_mae_db": 0.3414206500664201}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 6.399293414479695e-05, "log_rms_mae_db": 0.8442726553411123, "rms_envelope_nrmse": 0.08554591784444984, "envelope_correlation": 0.9928774070080326, "rms_gain_error_db": -0.0481573776369498, "reference_mean_power": 0.0003253818281247376, "output_mean_power": 0.00032179371464784016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.1005868407421181e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4497636322434906e-05, "output_power": 5.443389998366679e-05, "reference_active": true, "delta_db": -0.005082154226612463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.385752117634814e-06, "output_power": 9.509890977832637e-06, "reference_active": true, "delta_db": 0.0570645808309443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7591737704275347e-07, "output_power": 3.129854577612288e-07, "reference_active": true, "delta_db": -0.7956824234592437}], "active_band_count": 3, "active_band_mae_db": 0.28594305283893345}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8771834304744682, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a7fe6d02388b0a7d928488dede063cd11b94f187160f35587e01fce8ce02e41c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9897516648725093, "standard_deviation": 0.001280505007654778, "minimum": 0.9879572300190393, "maximum": 0.9918705118194201, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4373012656081703, "standard_deviation": 0.031056358232852275, "minimum": 1.3927360854166533, "maximum": 1.4714706806444118, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3426142056494575, "standard_deviation": 0.14200263437675148, "minimum": 0.1865975984613449, "maximum": 0.5630533683916078, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1138871097203307, "standard_deviation": 0.0068947396645008315, "minimum": 0.10159283410032058, "maximum": 0.12293996870446898, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966821330085047, "standard_deviation": 0.0007075053636444843, "minimum": 0.995746717810013, "maximum": 0.9980793676685026, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8553219019122935, "standard_deviation": 0.028862170006626473, "minimum": 0.8206023656477484, "maximum": 0.8957063338518877, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.380959429585064, "standard_deviation": 0.1925660470434302, "minimum": 0.19305331556446945, "maximum": 0.6505252881946765, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06400600866055459, "standard_deviation": 0.006774514469065947, "minimum": 0.04960377099634432, "maximum": 0.07220245738466591, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9785826029753869, "standard_deviation": 0.001998373913841503, "minimum": 0.974214375176385, "maximum": 0.9806177398887816, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4260706701726114, "standard_deviation": 0.01993691763722078, "minimum": 1.3908813365754014, "maximum": 1.4467764691720437, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6380245745721613, "standard_deviation": 0.27706605252331623, "minimum": 0.21400597743683256, "maximum": 1.0382547287799528, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15689032548189757, "standard_deviation": 0.007887275345760732, "minimum": 0.14826243883314516, "maximum": 0.17383207145319637, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927823937742745, "standard_deviation": 0.0015620015366838512, "minimum": 0.989718181571117, "maximum": 0.9943080045805496, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.839124056488388, "standard_deviation": 0.01784243132011288, "minimum": 0.8173316305448809, "maximum": 0.8713143263147438, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5944397021848566, "standard_deviation": 0.3173898030178788, "minimum": 0.16814857176895578, "maximum": 1.1644088060178315, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08983832293595996, "standard_deviation": 0.009750013275529976, "minimum": 0.07962773431887461, "maximum": 0.1107273853597853, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9951882486672223, "standard_deviation": 0.0005739089396845833, "minimum": 0.9943948566101479, "maximum": 0.9960485459627167, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4403424613620635, "standard_deviation": 0.02972402697940816, "minimum": 1.4005181484574016, "maximum": 1.4837308540350658, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.28325660100379546, "standard_deviation": 0.09507317634044513, "minimum": 0.1832647294155727, "maximum": 0.4421956637386866, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.08051674291926234, "standard_deviation": 0.004599195572170815, "minimum": 0.07420603117831548, "maximum": 0.08676410743759444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984212703540465, "standard_deviation": 0.0004316618983207637, "minimum": 0.9974655087176585, "maximum": 0.9989125138261371, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7886116379469896, "standard_deviation": 0.03297636968223321, "minimum": 0.7481023810809877, "maximum": 0.834693609169993, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3869455882718891, "standard_deviation": 0.20865481974568145, "minimum": 0.07556947567088972, "maximum": 0.624721352035784, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.045483465941499615, "standard_deviation": 0.0054254764957787594, "minimum": 0.037575921223666195, "maximum": 0.05702530254964689, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19459777300388242, "standard_deviation": 0.02570867505032475, "minimum": 0.14290647148691168, "maximum": 0.22030623626284696, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.5049129199005655, "standard_deviation": 0.00866622949538137, "minimum": 5.4964635132431665, "maximum": 5.519928102660224, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5332305929040688, "standard_deviation": 0.012541088145181402, "minimum": 0.5137030733440797, "maximum": 0.5532071625777708, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09289447780876532, "standard_deviation": 0.025772189819263647, "minimum": -0.14620836721492952, "maximum": -0.0565355095963084, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9955853044496336, "standard_deviation": 0.0006540339785183403, "minimum": 0.9941470198967022, "maximum": 0.9964703489209336, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8335210509657298, "standard_deviation": 0.011933727575667819, "minimum": 0.8204545284283772, "maximum": 0.8585193466404187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.32302638182163196, "standard_deviation": 0.10448455883632879, "minimum": 0.10600493528271161, "maximum": 0.4528384096303089, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.06977986345692005, "standard_deviation": 0.004695330648714832, "minimum": 0.06389343524955216, "maximum": 0.08023663396888774, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9987249163149498, "standard_deviation": 0.0002379163063013403, "minimum": 0.9983124094173546, "maximum": 0.9990066294252271, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4356567007603435, "standard_deviation": 0.01709087088204093, "minimum": 0.41317785998955225, "maximum": 0.4702553601521473, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4371142073253199, "standard_deviation": 0.14522242113924697, "minimum": 0.26364595869260826, "maximum": 0.6777314304085551, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03723136399176972, "standard_deviation": 0.003164488306551675, "minimum": 0.03305436539933669, "maximum": 0.04275289005732821, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0014", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.164921760559082, "predicted_mos": 2.164921760559082, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "66a214b7323acbf6bb088e843a34932f12c6f9c9b14a68798fc755d1dc18e9e6", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.715053599793464, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8305243136409391, "全帯域_10ms_rms_envelope_nrmse": 0.06670847420061062, "全帯域_10ms_envelope_correlation": 0.9960978580741917, "全帯域_10ms_modulation_mae_db": 0.5928932527440894, "全帯域_40ms_log_rms_mae_db": 0.406731755141378, "全帯域_40ms_rms_envelope_nrmse": 0.0345738979559748, "全帯域_40ms_envelope_correlation": 0.9990586855627756, "全帯域_40ms_modulation_mae_db": 0.6667830918941293, "300–1000 Hz_10ms_log_rms_mae_db": 1.4338402952716733, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07833409180388298, "300–1000 Hz_10ms_envelope_correlation": 0.9954656516081144, "300–1000 Hz_10ms_modulation_mae_db": 0.4397286450996598, "300–1000 Hz_40ms_log_rms_mae_db": 0.8032277607921974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04452383643211352, "300–1000 Hz_40ms_envelope_correlation": 0.9985159679247742, "300–1000 Hz_40ms_modulation_mae_db": 0.5759256200994701, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4191165075786203, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10428968375266355, "1000–3000 Hz_10ms_envelope_correlation": 0.9920187604393522, "1000–3000 Hz_10ms_modulation_mae_db": 0.6825757602293231, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8360841302691889, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0625241465838079, "1000–3000 Hz_40ms_envelope_correlation": 0.9975654712800408, "1000–3000 Hz_40ms_modulation_mae_db": 0.7647150372498098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.404111601772511, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14635005028872308, "3000–6000 Hz_10ms_envelope_correlation": 0.9822998717673412, "3000–6000 Hz_10ms_modulation_mae_db": 0.9691345938605015, "3000–6000 Hz_40ms_log_rms_mae_db": 0.85092886169183, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08573574100060351, "3000–6000 Hz_40ms_envelope_correlation": 0.9946720536597325, "3000–6000 Hz_40ms_modulation_mae_db": 1.0393251028611927, "mr_stft_db": 5.496446687657912, "band_mae_db": 0.2798785358704318, "rms_gain_error_db": -0.14282812564903846, "output_peak": 0.5337264746135173}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8433742083470764, "全帯域_10ms_rms_envelope_nrmse": 0.07162941840932467, "全帯域_10ms_envelope_correlation": 0.9954458166544748, "全帯域_10ms_modulation_mae_db": 0.5524722602484283, "全帯域_40ms_log_rms_mae_db": 0.43715312943500406, "全帯域_40ms_rms_envelope_nrmse": 0.03838220977846644, "全帯域_40ms_envelope_correlation": 0.9987558573504731, "全帯域_40ms_modulation_mae_db": 0.7118420271659867, "300–1000 Hz_10ms_log_rms_mae_db": 1.4626425702180301, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08150062074270586, "300–1000 Hz_10ms_envelope_correlation": 0.9951232463525823, "300–1000 Hz_10ms_modulation_mae_db": 0.4175669117448823, "300–1000 Hz_40ms_log_rms_mae_db": 0.8196722927027017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047704602283522274, "300–1000 Hz_40ms_envelope_correlation": 0.9983305564115691, "300–1000 Hz_40ms_modulation_mae_db": 0.6299231438390261, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4690599720165132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11732247678400406, "1000–3000 Hz_10ms_envelope_correlation": 0.9894511852923861, "1000–3000 Hz_10ms_modulation_mae_db": 0.47935184865368763, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8727032846383611, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06628535960309018, "1000–3000 Hz_40ms_envelope_correlation": 0.996902436951756, "1000–3000 Hz_40ms_modulation_mae_db": 0.5849798367508809, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4572155242123819, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16007959758048243, "3000–6000 Hz_10ms_envelope_correlation": 0.9777525721717706, "3000–6000 Hz_10ms_modulation_mae_db": 1.0411845943533373, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8528418045298352, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08698200011961324, "3000–6000 Hz_40ms_envelope_correlation": 0.9935244665893381, "3000–6000 Hz_40ms_modulation_mae_db": 0.9399681586869318, "mr_stft_db": 5.485975811579121, "band_mae_db": 0.25038768923048144, "rms_gain_error_db": -0.13801881212754472, "output_peak": 0.5586894220361461}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.804692395555694, "全帯域_10ms_rms_envelope_nrmse": 0.06662381267867544, "全帯域_10ms_envelope_correlation": 0.9960366344605452, "全帯域_10ms_modulation_mae_db": 0.3133083761776387, "全帯域_40ms_log_rms_mae_db": 0.4007635180148117, "全帯域_40ms_rms_envelope_nrmse": 0.03140748907616064, "全帯域_40ms_envelope_correlation": 0.9991840964965987, "全帯域_40ms_modulation_mae_db": 0.48515246839319137, "300–1000 Hz_10ms_log_rms_mae_db": 1.372165801191619, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07651125259384918, "300–1000 Hz_10ms_envelope_correlation": 0.9956798530543475, "300–1000 Hz_10ms_modulation_mae_db": 0.24056746606854318, "300–1000 Hz_40ms_log_rms_mae_db": 0.752599956718706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03950662213242991, "300–1000 Hz_40ms_envelope_correlation": 0.9988588501246951, "300–1000 Hz_40ms_modulation_mae_db": 0.4302558567445514, "1000–3000 Hz_10ms_log_rms_mae_db": 1.448748869891818, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11898318853688009, "1000–3000 Hz_10ms_envelope_correlation": 0.9890224792450596, "1000–3000 Hz_10ms_modulation_mae_db": 0.3698542691569267, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8721623151794554, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07184504169660524, "1000–3000 Hz_40ms_envelope_correlation": 0.9960831430432102, "1000–3000 Hz_40ms_modulation_mae_db": 0.5024480388292889, "3000–6000 Hz_10ms_log_rms_mae_db": 1.360476039642474, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14596432816315516, "3000–6000 Hz_10ms_envelope_correlation": 0.9817975416743092, "3000–6000 Hz_10ms_modulation_mae_db": 0.6512478719310746, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8071162825005568, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08251154576983015, "3000–6000 Hz_40ms_envelope_correlation": 0.9944272145516287, "3000–6000 Hz_40ms_modulation_mae_db": 0.728706962282688, "mr_stft_db": 5.502127266238557, "band_mae_db": 0.24473259581386256, "rms_gain_error_db": -0.11634931556397436, "output_peak": 0.509944716627356}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8471859528831146, "全帯域_10ms_rms_envelope_nrmse": 0.07135260469410937, "全帯域_10ms_envelope_correlation": 0.9953432862908881, "全帯域_10ms_modulation_mae_db": 0.2758269379032026, "全帯域_40ms_log_rms_mae_db": 0.429223764329701, "全帯域_40ms_rms_envelope_nrmse": 0.03676503406925965, "全帯域_40ms_envelope_correlation": 0.9987072500816511, "全帯域_40ms_modulation_mae_db": 0.4608376430018947, "300–1000 Hz_10ms_log_rms_mae_db": 1.5316709888755473, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08566137928572418, "300–1000 Hz_10ms_envelope_correlation": 0.9945430508108772, "300–1000 Hz_10ms_modulation_mae_db": 0.27213789626604595, "300–1000 Hz_40ms_log_rms_mae_db": 0.837347013383441, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04245017570183581, "300–1000 Hz_40ms_envelope_correlation": 0.9986488049104152, "300–1000 Hz_40ms_modulation_mae_db": 0.4069548760535621, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4284487235446548, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11780322924451608, "1000–3000 Hz_10ms_envelope_correlation": 0.9890241712318073, "1000–3000 Hz_10ms_modulation_mae_db": 0.16348335267277675, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8514467213360352, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0674881298512192, "1000–3000 Hz_40ms_envelope_correlation": 0.9962983568696864, "1000–3000 Hz_40ms_modulation_mae_db": 0.40391085497844564, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3965907511998084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.148044202312053, "3000–6000 Hz_10ms_envelope_correlation": 0.9813849162082873, "3000–6000 Hz_10ms_modulation_mae_db": 0.74244328580494, "3000–6000 Hz_40ms_log_rms_mae_db": 0.817542855152745, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08940980288492575, "3000–6000 Hz_40ms_envelope_correlation": 0.9927234919773594, "3000–6000 Hz_40ms_modulation_mae_db": 0.6620116708073068, "mr_stft_db": 5.499670712361005, "band_mae_db": 0.16176996002259708, "rms_gain_error_db": -0.07339009293745971, "output_peak": 0.5375432340362811}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8166620954828925, "全帯域_10ms_rms_envelope_nrmse": 0.06661508964055658, "全帯域_10ms_envelope_correlation": 0.995968555048945, "全帯域_10ms_modulation_mae_db": 0.3590311801320478, "全帯域_40ms_log_rms_mae_db": 0.41177611521452007, "全帯域_40ms_rms_envelope_nrmse": 0.03606862869866116, "全帯域_40ms_envelope_correlation": 0.9987750191823702, "全帯域_40ms_modulation_mae_db": 0.4355116553586504, "300–1000 Hz_10ms_log_rms_mae_db": 1.4387380175135132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07757908860694154, "300–1000 Hz_10ms_envelope_correlation": 0.9955253638776455, "300–1000 Hz_10ms_modulation_mae_db": 0.2844756267093136, "300–1000 Hz_40ms_log_rms_mae_db": 0.8109976611039903, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04787072121999266, "300–1000 Hz_40ms_envelope_correlation": 0.9982247817686106, "300–1000 Hz_40ms_modulation_mae_db": 0.4116426843216104, "1000–3000 Hz_10ms_log_rms_mae_db": 1.408032377194555, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11347543776664705, "1000–3000 Hz_10ms_envelope_correlation": 0.9898284837955176, "1000–3000 Hz_10ms_modulation_mae_db": 0.29830841176611006, "1000–3000 Hz_40ms_log_rms_mae_db": 0.862032338452907, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06641830825939153, "1000–3000 Hz_40ms_envelope_correlation": 0.9964137558144481, "1000–3000 Hz_40ms_modulation_mae_db": 0.3579103238167834, "3000–6000 Hz_10ms_log_rms_mae_db": 1.390885576874198, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15380769774413341, "3000–6000 Hz_10ms_envelope_correlation": 0.979121680589543, "3000–6000 Hz_10ms_modulation_mae_db": 0.4496388304997348, "3000–6000 Hz_40ms_log_rms_mae_db": 0.781433997349098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08111789120265724, "3000–6000 Hz_40ms_envelope_correlation": 0.993778282851517, "3000–6000 Hz_40ms_modulation_mae_db": 0.549902511605791, "mr_stft_db": 5.49560582637777, "band_mae_db": 0.17234431904886527, "rms_gain_error_db": -0.08054776131262281, "output_peak": 0.5163949856744044}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.839841579569095, "全帯域_10ms_rms_envelope_nrmse": 0.07084334618963135, "全帯域_10ms_envelope_correlation": 0.995443772501359, "全帯域_10ms_modulation_mae_db": 0.22971888090251188, "全帯域_40ms_log_rms_mae_db": 0.39508356934509076, "全帯域_40ms_rms_envelope_nrmse": 0.03563923520065332, "全帯域_40ms_envelope_correlation": 0.9988322001222298, "全帯域_40ms_modulation_mae_db": 0.42996135702802657, "300–1000 Hz_10ms_log_rms_mae_db": 1.4746817221175645, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08634543932266946, "300–1000 Hz_10ms_envelope_correlation": 0.9944679922925018, "300–1000 Hz_10ms_modulation_mae_db": 0.3405390934811598, "300–1000 Hz_40ms_log_rms_mae_db": 0.7679406226117175, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048972965070146986, "300–1000 Hz_40ms_envelope_correlation": 0.9981906575968998, "300–1000 Hz_40ms_modulation_mae_db": 0.3358926457640616, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4246702567732759, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.10434876180955907, "1000–3000 Hz_10ms_envelope_correlation": 0.9913954940384399, "1000–3000 Hz_10ms_modulation_mae_db": 0.28138031140647374, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8305265307608699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057171512233894134, "1000–3000 Hz_40ms_envelope_correlation": 0.9973480001436436, "1000–3000 Hz_40ms_modulation_mae_db": 0.3574901199951559, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4043203397121873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1449669620858629, "3000–6000 Hz_10ms_envelope_correlation": 0.981578220050874, "3000–6000 Hz_10ms_modulation_mae_db": 0.46125977320483297, "3000–6000 Hz_40ms_log_rms_mae_db": 0.822860616587573, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07947927768090672, "3000–6000 Hz_40ms_envelope_correlation": 0.9941859197702791, "3000–6000 Hz_40ms_modulation_mae_db": 0.28051314683445855, "mr_stft_db": 5.480450715097997, "band_mae_db": 0.19419384733770487, "rms_gain_error_db": -0.09081853538394416, "output_peak": 0.5878993314231252}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7868749306405064, "全帯域_10ms_rms_envelope_nrmse": 0.06800720261113045, "全帯域_10ms_envelope_correlation": 0.995826325691869, "全帯域_10ms_modulation_mae_db": 0.38819869090937825, "全帯域_40ms_log_rms_mae_db": 0.4078648025657174, "全帯域_40ms_rms_envelope_nrmse": 0.03760671287893366, "全帯域_40ms_envelope_correlation": 0.9987046722792384, "全帯域_40ms_modulation_mae_db": 0.5811950839739369, "300–1000 Hz_10ms_log_rms_mae_db": 1.4570958452267597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07664767309079641, "300–1000 Hz_10ms_envelope_correlation": 0.9956102378634011, "300–1000 Hz_10ms_modulation_mae_db": 0.20705874161597193, "300–1000 Hz_40ms_log_rms_mae_db": 0.7836625423126042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04596527865587908, "300–1000 Hz_40ms_envelope_correlation": 0.9983342557603258, "300–1000 Hz_40ms_modulation_mae_db": 0.35225009358618964, "1000–3000 Hz_10ms_log_rms_mae_db": 1.41681382505667, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11151446239992767, "1000–3000 Hz_10ms_envelope_correlation": 0.9907753229669821, "1000–3000 Hz_10ms_modulation_mae_db": 0.7520694955935675, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8431666701352734, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06558042691672358, "1000–3000 Hz_40ms_envelope_correlation": 0.9972762531698984, "1000–3000 Hz_40ms_modulation_mae_db": 1.0258468323167584, "3000–6000 Hz_10ms_log_rms_mae_db": 1.366495385882323, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15493411930002005, "3000–6000 Hz_10ms_envelope_correlation": 0.9795644616234135, "3000–6000 Hz_10ms_modulation_mae_db": 1.0752377778806252, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7859713000346903, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0841742499601006, "3000–6000 Hz_40ms_envelope_correlation": 0.9945073499926367, "3000–6000 Hz_40ms_modulation_mae_db": 1.20781516946192, "mr_stft_db": 5.485289546434732, "band_mae_db": 0.2315624128158232, "rms_gain_error_db": -0.10194202202562307, "output_peak": 0.5483584441455563}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8244153912001804, "全帯域_10ms_rms_envelope_nrmse": 0.062189333723341525, "全帯域_10ms_envelope_correlation": 0.996458771747904, "全帯域_10ms_modulation_mae_db": 0.4344498451768686, "全帯域_40ms_log_rms_mae_db": 0.4199859178210235, "全帯域_40ms_rms_envelope_nrmse": 0.03197963588989047, "全帯域_40ms_envelope_correlation": 0.9990077162022168, "全帯域_40ms_modulation_mae_db": 0.5972820060905696, "300–1000 Hz_10ms_log_rms_mae_db": 1.425990731304859, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07019166935878506, "300–1000 Hz_10ms_envelope_correlation": 0.9963188024169876, "300–1000 Hz_10ms_modulation_mae_db": 0.2822899621861618, "300–1000 Hz_40ms_log_rms_mae_db": 0.7188120834373289, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039075515783738246, "300–1000 Hz_40ms_envelope_correlation": 0.9987931456447054, "300–1000 Hz_40ms_modulation_mae_db": 0.5246899146839723, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4358334745054961, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11230906952254427, "1000–3000 Hz_10ms_envelope_correlation": 0.9901277323191557, "1000–3000 Hz_10ms_modulation_mae_db": 0.2816453898283748, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8127148910649272, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0593010495452181, "1000–3000 Hz_40ms_envelope_correlation": 0.9973235100150161, "1000–3000 Hz_40ms_modulation_mae_db": 0.3946982268222174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4310552760241366, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14535593529612065, "3000–6000 Hz_10ms_envelope_correlation": 0.9815296570172687, "3000–6000 Hz_10ms_modulation_mae_db": 1.0337517757414103, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8309210228463221, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07937502950786343, "3000–6000 Hz_40ms_envelope_correlation": 0.9942952773532974, "3000–6000 Hz_40ms_modulation_mae_db": 1.2984932037938963, "mr_stft_db": 5.493758081207161, "band_mae_db": 0.19853450689958854, "rms_gain_error_db": -0.054547550986089294, "output_peak": 0.5216063564211952}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 91600, "output_samples": 91600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.586975219173038, "spectral_convergence": 0.26165668682903326, "spectral_convergence_status": "ok", "amplitude_floor": 1.4112644553101518e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.295926709587348, "spectral_convergence": 0.25541290278099105, "spectral_convergence_status": "ok", "amplitude_floor": 1.1975531593072457e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.606438134213348, "spectral_convergence": 0.2662839931521121, "spectral_convergence_status": "ok", "amplitude_floor": 1.114415465542996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.564075826256522e-05, "output_power": 6.138493876070019e-05, "delta_db": 0.4266878706494296, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010000129923479963, "output_power": 0.0010133935278550405, "delta_db": 0.0577248382747498, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018259392170112694, "output_power": 0.001760158275891859, "delta_db": -0.159345944400471, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005127219920465177, "output_power": 0.0004830390179669901, "delta_db": -0.2589973317029804, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00033320627903786135, "output_power": 0.0002990879278648258, "delta_db": -0.4691429290022158, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002817706196064759, "output_power": 0.0002625176312442529, "delta_db": -0.3073723011927442, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2798785358704318, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14282812564903846, "pitch": null, "output_rms_dbfs": -24.11215140489717, "output_peak": 0.5337264746135173, "reference_rms_dbfs": -23.96932327924813, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.496446687657912}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 91600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00020825955187282386, "log_rms_mae_db": 0.8305243136409391, "rms_envelope_nrmse": 0.06670847420061062, "envelope_correlation": 0.9960978580741917, "rms_gain_error_db": -0.14281312201148952, "reference_mean_power": 0.004012756910556895, "output_mean_power": 0.003882947683246203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 2.2120307480210933e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008889321025104637, "output_power": 0.0008475914557544315, "reference_active": true, "delta_db": -0.20682020302374415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022980725896418495, "output_power": 0.00021267165467466322, "reference_active": true, "delta_db": -0.33654132630619676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1131932747773025e-05, "output_power": 3.0949047971081776e-05, "reference_active": true, "delta_db": -1.2353182289023272}], "active_band_count": 3, "active_band_mae_db": 0.5928932527440894}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.0001984160011662782, "log_rms_mae_db": 0.406731755141378, "rms_envelope_nrmse": 0.0345738979559748, "envelope_correlation": 0.9990586855627756, "rms_gain_error_db": -0.14282081247709025, "reference_mean_power": 0.0040337408532874226, "output_mean_power": 0.003903245901661234}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 2.0115882799069226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008024963383362849, "output_power": 0.0007657064920687584, "reference_active": true, "delta_db": -0.20380730174270673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011463438169489824, "output_power": 0.0001042786086576902, "reference_active": true, "delta_db": -0.41119664917642795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4945953602736586e-06, "output_power": 2.5401675486554903e-06, "reference_active": true, "delta_db": -1.3853453247632532}], "active_band_count": 3, "active_band_mae_db": 0.6667830918941293}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00019054595180557928, "log_rms_mae_db": 1.4338402952716733, "rms_envelope_nrmse": 0.07833409180388298, "envelope_correlation": 0.9954656516081144, "rms_gain_error_db": -0.10016110620437758, "reference_mean_power": 0.0025757888112863028, "output_mean_power": 0.0025170633248100724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.7595886306573375e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007279305516452219, "output_power": 0.0007027094183705824, "reference_active": true, "delta_db": -0.15314172942420878}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019010480993234813, "output_power": 0.0001775359746485958, "reference_active": true, "delta_db": -0.2970673659030952}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6294378423488108e-05, "output_power": 2.152609362862313e-05, "reference_active": true, "delta_db": -0.8689768399716755}], "active_band_count": 3, "active_band_mae_db": 0.4397286450996598}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.00018019065505981728, "log_rms_mae_db": 0.8032277607921974, "rms_envelope_nrmse": 0.04452383643211352, "envelope_correlation": 0.9985159679247742, "rms_gain_error_db": -0.10016056216692328, "reference_mean_power": 0.0025893611436675294, "output_mean_power": 0.00253032653814977}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.6269545971301712e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006780414214466183, "output_power": 0.0006533974955354539, "reference_active": true, "delta_db": -0.1607876016367275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.548437806715927e-05, "output_power": 8.76084652162673e-05, "reference_active": true, "delta_db": -0.37386251546960936}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9327510117604047e-06, "output_power": 2.2282428363419786e-06, "reference_active": true, "delta_db": -1.1931267431920736}], "active_band_count": 3, "active_band_mae_db": 0.5759256200994701}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.327720993461315e-05, "log_rms_mae_db": 1.4191165075786203, "rms_envelope_nrmse": 0.10428968375266355, "envelope_correlation": 0.9920187604393522, "rms_gain_error_db": -0.3005609328340887, "reference_mean_power": 0.0007122971869413385, "output_mean_power": 0.0006646685593168326}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 4.689927289528067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023362430343497353, "output_power": 0.00021428083136083737, "reference_active": true, "delta_db": -0.3753469684568938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.797902128989837e-05, "output_power": 3.397314887458409e-05, "reference_active": true, "delta_db": -0.4840796662601343}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.042184778585592e-05, "output_power": 7.927105922110278e-06, "reference_active": true, "delta_db": -1.1883006459709413}], "active_band_count": 3, "active_band_mae_db": 0.6825757602293231}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 8.747046012456969e-05, "log_rms_mae_db": 0.8360841302691889, "rms_envelope_nrmse": 0.0625241465838079, "envelope_correlation": 0.9975654712800408, "rms_gain_error_db": -0.3006010163174118, "reference_mean_power": 0.0007159674763911086, "output_mean_power": 0.0006680872641655747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 4.328159646639031e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021700970165722394, "output_power": 0.00020027676756674832, "reference_active": true, "delta_db": -0.3484857655093635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.702508733600319e-05, "output_power": 1.4563536306734827e-05, "reference_active": true, "delta_db": -0.6782250561657698}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.679451250427792e-07, "output_power": 6.482594243132968e-07, "reference_active": true, "delta_db": -1.2674342900742963}], "active_band_count": 3, "active_band_mae_db": 0.7647150372498098}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.039773028293347e-05, "log_rms_mae_db": 1.404111601772511, "rms_envelope_nrmse": 0.14635005028872308, "envelope_correlation": 0.9822998717673412, "rms_gain_error_db": -0.4114011126913797, "reference_mean_power": 0.0003237352885604671, "output_mean_power": 0.00029447601725508295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.2480235329901466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.70846061617512e-05, "output_power": 5.3980460352981275e-05, "reference_active": true, "delta_db": -0.24282425154477263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9595191589145483e-05, "output_power": 1.7386102053498813e-05, "reference_active": true, "delta_db": -0.5194728955392353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.074204859780537e-06, "output_power": 3.0964019584140346e-06, "reference_active": true, "delta_db": -2.1451066344974965}], "active_band_count": 3, "active_band_mae_db": 0.9691345938605015}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 6.399293414479695e-05, "log_rms_mae_db": 0.85092886169183, "rms_envelope_nrmse": 0.08573574100060351, "envelope_correlation": 0.9946720536597325, "rms_gain_error_db": -0.41154798163225975, "reference_mean_power": 0.0003253818281247376, "output_mean_power": 0.0002959637331688749}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.1005868407421181e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4497636322434906e-05, "output_power": 5.114057327120271e-05, "reference_active": true, "delta_db": -0.27612074381815366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.385752117634814e-06, "output_power": 8.180567372474225e-06, "reference_active": true, "delta_db": -0.5968565439602921}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7591737704275347e-07, "output_power": 2.2417779835962893e-07, "reference_active": true, "delta_db": -2.2449980208051326}], "active_band_count": 3, "active_band_mae_db": 1.0393251028611927}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8892738170571406, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9e083006f66a0dba5d65d6f1154747ccdb7fa3812d524bcc1f216a944f5c7474", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9902054536660876, "standard_deviation": 0.0011045977226959922, "minimum": 0.9890224792450596, "maximum": 0.9920187604393522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4313405008202005, "standard_deviation": 0.019648955303069208, "minimum": 1.408032377194555, "maximum": 1.4690599720165132, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.413583604913405, "standard_deviation": 0.2083579009579792, "minimum": 0.16348335267277675, "maximum": 0.7520694955935675, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11250578872709273, "standard_deviation": 0.005722426843388714, "minimum": 0.10428968375266355, "maximum": 0.11898318853688009, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969013659109625, "standard_deviation": 0.0005644387223561833, "minimum": 0.9960831430432102, "maximum": 0.9975654712800408, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8476046102296273, "standard_deviation": 0.021102862234608354, "minimum": 0.8127148910649272, "maximum": 0.8727032846383611, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5489999088449176, "standard_deviation": 0.2376758621736083, "minimum": 0.3574901199951559, "maximum": 1.0258468323167584, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06457674683624373, "standard_deviation": 0.004713294119856617, "minimum": 0.057171512233894134, "maximum": 0.07184504169660524, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9806286151378509, "standard_deviation": 0.0016089039221785673, "minimum": 0.9777525721717706, "maximum": 0.9822998717673412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4013938119150025, "standard_deviation": 0.03170730782547874, "minimum": 1.360476039642474, "maximum": 1.4572155242123819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.802987312909557, "standard_deviation": 0.26187202856632946, "minimum": 0.4496388304997348, "maximum": 1.0752377778806252, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14993786159631883, "standard_deviation": 0.005615593412186809, "minimum": 0.1449669620858629, "maximum": 0.16007959758048243, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940142570932236, "standard_deviation": 0.0006454748439114991, "minimum": 0.9927234919773594, "maximum": 0.9946720536597325, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8187020925865813, "standard_deviation": 0.026636362599094493, "minimum": 0.781433997349098, "maximum": 0.8528418045298352, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8383419907917731, "standard_deviation": 0.34567826318040684, "minimum": 0.28051314683445855, "maximum": 1.2984932037938963, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08359819226581258, "standard_deviation": 0.0036304109497683605, "minimum": 0.07937502950786343, "maximum": 0.08940980288492575, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9953417747845572, "standard_deviation": 0.0006140280662954606, "minimum": 0.9944679922925018, "maximum": 0.9963188024169876, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4496032464649458, "standard_deviation": 0.04552995861236939, "minimum": 1.372165801191619, "maximum": 1.5316709888755473, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.3105455428964673, "standard_deviation": 0.08246304395518048, "minimum": 0.20705874161597193, "maximum": 0.4397286450996598, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07909640185066932, "standard_deviation": 0.0052953433388291515, "minimum": 0.07019166935878506, "maximum": 0.08634543932266946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9984871275177494, "standard_deviation": 0.00025729364282991356, "minimum": 0.9981906575968998, "maximum": 0.9988588501246951, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7867824916328359, "standard_deviation": 0.03900275334587231, "minimum": 0.7188120834373289, "maximum": 0.837347013383441, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.45844185438655544, "standard_deviation": 0.1065883896432076, "minimum": 0.3358926457640616, "maximum": 0.6299231438390261, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.044508714659957316, "standard_deviation": 0.0038237283078142717, "minimum": 0.039075515783738246, "maximum": 0.048972965070146986, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21667548337991935, "standard_deviation": 0.04132510220620419, "minimum": 0.16176996002259708, "maximum": 0.2798785358704318, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.492415580869282, "standard_deviation": 0.007658725830008819, "minimum": 5.480450715097997, "maximum": 5.502127266238557, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5392703706221977, "standard_deviation": 0.025504847905927004, "minimum": 0.509944716627356, "maximum": 0.5878993314231252, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09980527699828706, "standard_deviation": 0.03113828101610249, "minimum": -0.14282812564903846, "maximum": -0.054547550986089294, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9958276275587721, "standard_deviation": 0.00038961966652273117, "minimum": 0.9953432862908881, "maximum": 0.996458771747904, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8241963584149373, "standard_deviation": 0.02074069146031086, "minimum": 0.7868749306405064, "maximum": 0.8471859528831146, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3932374280242707, "standard_deviation": 0.12818752530376593, "minimum": 0.22971888090251188, "maximum": 0.5928932527440894, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0679961602684225, "standard_deviation": 0.003201359921402848, "minimum": 0.062189333723341525, "maximum": 0.07162941840932467, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9988781871596942, "standard_deviation": 0.0001812606207924803, "minimum": 0.9987046722792384, "maximum": 0.9991840964965987, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4135728214834058, "standard_deviation": 0.014296109669456356, "minimum": 0.39508356934509076, "maximum": 0.43715312943500406, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5460706666132982, "standard_deviation": 0.10862012541286814, "minimum": 0.42996135702802657, "maximum": 0.7118420271659867, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03530285544350002, "standard_deviation": 0.002517843604355606, "minimum": 0.03140748907616064, "maximum": 0.03838220977846644, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8463-294828-0022", "language": "en", "style": "normal", "speaker": "librispeech_8463", "source_utterance_id": "8463-294828-0022", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 37680, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8463-294828-0022.wav", "source_file_sha256": "957720b155bf575e4435aae7d0c07af8fc27331f2948e1aa862d2257bec5cf8e", "decoded_pcm_sha256": "276e5d5f0e0b54af2b651a409e0f7fbcd7b42508e69037a0becabc956b358b07", "original_source_sha256": "4b0b7f62860ac9f04945b08f79183b52c26f80eedd14ff829e2ff2aaa5143e3f", "corpus_entry_id": "librispeech_8463-294828-0022"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.7197670936584473, "predicted_mos": 3.7197670936584473, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "c3bed8029386dc2619bf5e69fe8c860e5459e1bcf14b2f0b07c54ce70672f5c3", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.818807222000032, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8463-294828-0022", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c3bed8029386dc2619bf5e69fe8c860e5459e1bcf14b2f0b07c54ce70672f5c3", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "908baa31161703eb40e83b7043052a4a50ebf6e6dc303d833ee950422de3dd86"}, "rows": [{"case": "librispeech_8463-294828-0022", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 1.8345232009887695, "predicted_mos": 1.8345232009887695, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "12ea9984229ea2317a30dc6680db70c72853e81874a17e995879a7d1fdedca93", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.1760948299997835, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8463-294828-0022", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "12ea9984229ea2317a30dc6680db70c72853e81874a17e995879a7d1fdedca93", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8740516433023192, "全帯域_10ms_rms_envelope_nrmse": 0.0890145861272344, "全帯域_10ms_envelope_correlation": 0.9928204590997016, "全帯域_10ms_modulation_mae_db": 0.2405873559442165, "全帯域_40ms_log_rms_mae_db": 0.45687006471892566, "全帯域_40ms_rms_envelope_nrmse": 0.04753311382342967, "全帯域_40ms_envelope_correlation": 0.9979948783886864, "全帯域_40ms_modulation_mae_db": 0.4512186107982606, "300–1000 Hz_10ms_log_rms_mae_db": 1.4491199147872624, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11082622798702739, "300–1000 Hz_10ms_envelope_correlation": 0.9902986035469615, "300–1000 Hz_10ms_modulation_mae_db": 0.5269450478107255, "300–1000 Hz_40ms_log_rms_mae_db": 0.836368989989499, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062087753182505, "300–1000 Hz_40ms_envelope_correlation": 0.9970422410467376, "300–1000 Hz_40ms_modulation_mae_db": 0.17935872453318857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5001102913060922, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18056445413290145, "1000–3000 Hz_10ms_envelope_correlation": 0.973230657355384, "1000–3000 Hz_10ms_modulation_mae_db": 0.4961071202979399, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8747555684177639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1038781053979894, "1000–3000 Hz_40ms_envelope_correlation": 0.9902274613216157, "1000–3000 Hz_40ms_modulation_mae_db": 0.46222398929078157, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4824014458390344, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15282346931611474, "3000–6000 Hz_10ms_envelope_correlation": 0.9832819220386303, "3000–6000 Hz_10ms_modulation_mae_db": 1.0254654195625676, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7614583007324865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07361424720626428, "3000–6000 Hz_40ms_envelope_correlation": 0.996984516048118, "3000–6000 Hz_40ms_modulation_mae_db": 1.4468391871121362, "mr_stft_db": 5.675231355703725, "band_mae_db": 0.21057578653364714, "rms_gain_error_db": -0.14536616910923905, "output_peak": 0.4406212946059799}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8746157589232326, "全帯域_10ms_rms_envelope_nrmse": 0.08093682489025547, "全帯域_10ms_envelope_correlation": 0.993957385642759, "全帯域_10ms_modulation_mae_db": 0.38281888061299063, "全帯域_40ms_log_rms_mae_db": 0.4505255911577799, "全帯域_40ms_rms_envelope_nrmse": 0.036851265421212635, "全帯域_40ms_envelope_correlation": 0.9987034873255892, "全帯域_40ms_modulation_mae_db": 0.3681549872114527, "300–1000 Hz_10ms_log_rms_mae_db": 1.3769652787369673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09832895935891242, "300–1000 Hz_10ms_envelope_correlation": 0.9922240389230498, "300–1000 Hz_10ms_modulation_mae_db": 0.12375895052266452, "300–1000 Hz_40ms_log_rms_mae_db": 0.777475898607097, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046882457282772225, "300–1000 Hz_40ms_envelope_correlation": 0.998162162585464, "300–1000 Hz_40ms_modulation_mae_db": 0.11411588738404242, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5402541839203068, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18320839214369955, "1000–3000 Hz_10ms_envelope_correlation": 0.9724486897208283, "1000–3000 Hz_10ms_modulation_mae_db": 0.9317448762707844, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9781207529141761, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1114791088092096, "1000–3000 Hz_40ms_envelope_correlation": 0.9887911499122235, "1000–3000 Hz_40ms_modulation_mae_db": 0.46698172237919144, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4783481500487372, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13544666074595102, "3000–6000 Hz_10ms_envelope_correlation": 0.9868970028470565, "3000–6000 Hz_10ms_modulation_mae_db": 1.2851724268573483, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8353969169666571, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.074161860362616, "3000–6000 Hz_40ms_envelope_correlation": 0.9965745997536786, "3000–6000 Hz_40ms_modulation_mae_db": 1.55737665025001, "mr_stft_db": 5.685496211210111, "band_mae_db": 0.19554084459292467, "rms_gain_error_db": -0.08208332881814091, "output_peak": 0.4108855357054596}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8520543242234748, "全帯域_10ms_rms_envelope_nrmse": 0.09629944750880663, "全帯域_10ms_envelope_correlation": 0.9914414247830072, "全帯域_10ms_modulation_mae_db": 0.6561331843300292, "全帯域_40ms_log_rms_mae_db": 0.43036504134353915, "全帯域_40ms_rms_envelope_nrmse": 0.052885767598234525, "全帯域_40ms_envelope_correlation": 0.9973042176004406, "全帯域_40ms_modulation_mae_db": 0.8440417460011845, "300–1000 Hz_10ms_log_rms_mae_db": 1.4892079574847954, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12115385621999143, "300–1000 Hz_10ms_envelope_correlation": 0.9881519405592345, "300–1000 Hz_10ms_modulation_mae_db": 0.29033016649632293, "300–1000 Hz_40ms_log_rms_mae_db": 0.9000627832119765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07847277440525402, "300–1000 Hz_40ms_envelope_correlation": 0.9947079692677785, "300–1000 Hz_40ms_modulation_mae_db": 0.4235154939753216, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4606689766829066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1630250531710534, "1000–3000 Hz_10ms_envelope_correlation": 0.978873329435822, "1000–3000 Hz_10ms_modulation_mae_db": 0.9073693235163128, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9170637585116864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09243648288706471, "1000–3000 Hz_40ms_envelope_correlation": 0.9936010685787466, "1000–3000 Hz_40ms_modulation_mae_db": 0.7681716192050968, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3985038845728883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15225462033220166, "3000–6000 Hz_10ms_envelope_correlation": 0.9832025723516029, "3000–6000 Hz_10ms_modulation_mae_db": 0.9325313596301946, "3000–6000 Hz_40ms_log_rms_mae_db": 0.72417778638129, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07205977489258458, "3000–6000 Hz_40ms_envelope_correlation": 0.9969295693142464, "3000–6000 Hz_40ms_modulation_mae_db": 1.2716596979394774, "mr_stft_db": 5.6737772344101165, "band_mae_db": 0.26390920662690337, "rms_gain_error_db": -0.11726079131796192, "output_peak": 0.4392268051438854}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.904987933907682, "全帯域_10ms_rms_envelope_nrmse": 0.08946411046955571, "全帯域_10ms_envelope_correlation": 0.9926032800727181, "全帯域_10ms_modulation_mae_db": 0.48943928297933237, "全帯域_40ms_log_rms_mae_db": 0.4573084694421502, "全帯域_40ms_rms_envelope_nrmse": 0.04620818752427189, "全帯域_40ms_envelope_correlation": 0.9978905665828594, "全帯域_40ms_modulation_mae_db": 0.5891483071597157, "300–1000 Hz_10ms_log_rms_mae_db": 1.4104937863442824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10933554022969795, "300–1000 Hz_10ms_envelope_correlation": 0.9905241170890979, "300–1000 Hz_10ms_modulation_mae_db": 0.33732834081339497, "300–1000 Hz_40ms_log_rms_mae_db": 0.7641090089495657, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06517966328573502, "300–1000 Hz_40ms_envelope_correlation": 0.9966017822192759, "300–1000 Hz_40ms_modulation_mae_db": 0.46476131909958046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.516803877445471, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1580660474914079, "1000–3000 Hz_10ms_envelope_correlation": 0.9802378867636535, "1000–3000 Hz_10ms_modulation_mae_db": 0.9015013073954593, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9329258439928135, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09378997498629493, "1000–3000 Hz_40ms_envelope_correlation": 0.9924623332201078, "1000–3000 Hz_40ms_modulation_mae_db": 1.066470412791078, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4387043992426682, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16139923423408617, "3000–6000 Hz_10ms_envelope_correlation": 0.9804625581237053, "3000–6000 Hz_10ms_modulation_mae_db": 1.0007957994623928, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7309294966406922, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06572835492820006, "3000–6000 Hz_40ms_envelope_correlation": 0.9965685383132561, "3000–6000 Hz_40ms_modulation_mae_db": 1.1944016211637936, "mr_stft_db": 5.680168254954087, "band_mae_db": 0.15037601576393111, "rms_gain_error_db": -0.0534621295557314, "output_peak": 0.45007915594262266}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9648093965345872, "全帯域_10ms_rms_envelope_nrmse": 0.08508505595103846, "全帯域_10ms_envelope_correlation": 0.9933251967421586, "全帯域_10ms_modulation_mae_db": 0.29686443180365757, "全帯域_40ms_log_rms_mae_db": 0.45840127117900087, "全帯域_40ms_rms_envelope_nrmse": 0.03817869585622913, "全帯域_40ms_envelope_correlation": 0.9986355804679369, "全帯域_40ms_modulation_mae_db": 0.7152586412311854, "300–1000 Hz_10ms_log_rms_mae_db": 1.4406937693838189, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10806324577537581, "300–1000 Hz_10ms_envelope_correlation": 0.9906025139925851, "300–1000 Hz_10ms_modulation_mae_db": 0.8309768695586892, "300–1000 Hz_40ms_log_rms_mae_db": 0.7396001538222683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052850679819698966, "300–1000 Hz_40ms_envelope_correlation": 0.9976455388870413, "300–1000 Hz_40ms_modulation_mae_db": 1.0126171760323126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5508178245473445, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15615602969666603, "1000–3000 Hz_10ms_envelope_correlation": 0.9803777563341242, "1000–3000 Hz_10ms_modulation_mae_db": 0.8499207850081949, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9145077066619126, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08397729484993503, "1000–3000 Hz_40ms_envelope_correlation": 0.9944870272200313, "1000–3000 Hz_40ms_modulation_mae_db": 0.5137605027866275, "3000–6000 Hz_10ms_log_rms_mae_db": 1.420157093002312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1376592492548488, "3000–6000 Hz_10ms_envelope_correlation": 0.986128633335813, "3000–6000 Hz_10ms_modulation_mae_db": 1.3217303470806216, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8154529640288721, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06850246266639277, "3000–6000 Hz_40ms_envelope_correlation": 0.9968031616159203, "3000–6000 Hz_40ms_modulation_mae_db": 1.6951254029895235, "mr_stft_db": 5.689098435032655, "band_mae_db": 0.13362407396524856, "rms_gain_error_db": -0.09771927838251564, "output_peak": 0.4192015514544245}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8822889202081734, "全帯域_10ms_rms_envelope_nrmse": 0.08742271907519437, "全帯域_10ms_envelope_correlation": 0.9929396573460255, "全帯域_10ms_modulation_mae_db": 0.49908674951226595, "全帯域_40ms_log_rms_mae_db": 0.42913995252727016, "全帯域_40ms_rms_envelope_nrmse": 0.0402473232770119, "全帯域_40ms_envelope_correlation": 0.9984440144878043, "全帯域_40ms_modulation_mae_db": 0.48451898198134735, "300–1000 Hz_10ms_log_rms_mae_db": 1.392455263157526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11756619525993443, "300–1000 Hz_10ms_envelope_correlation": 0.9888455469353548, "300–1000 Hz_10ms_modulation_mae_db": 0.3952246615755059, "300–1000 Hz_40ms_log_rms_mae_db": 0.7474864753138848, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06800681860526017, "300–1000 Hz_40ms_envelope_correlation": 0.9960025485787061, "300–1000 Hz_40ms_modulation_mae_db": 0.38945735835531997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.579007480662638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1575945599497557, "1000–3000 Hz_10ms_envelope_correlation": 0.9801973972598594, "1000–3000 Hz_10ms_modulation_mae_db": 0.5430881955108441, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9773126301023003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08813455582113705, "1000–3000 Hz_40ms_envelope_correlation": 0.9941140290155295, "1000–3000 Hz_40ms_modulation_mae_db": 0.3728255822757484, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4205248539755342, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13934837021760102, "3000–6000 Hz_10ms_envelope_correlation": 0.9864757302157335, "3000–6000 Hz_10ms_modulation_mae_db": 1.251907370539245, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8120556652886098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07400457602222264, "3000–6000 Hz_40ms_envelope_correlation": 0.9970624939525463, "3000–6000 Hz_40ms_modulation_mae_db": 1.5938058569636733, "mr_stft_db": 5.6649584373979005, "band_mae_db": 0.2079539177922314, "rms_gain_error_db": -0.08698855478758193, "output_peak": 0.4361240294083825}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9286106371372671, "全帯域_10ms_rms_envelope_nrmse": 0.08428372547330576, "全帯域_10ms_envelope_correlation": 0.9934442605829465, "全帯域_10ms_modulation_mae_db": 0.40169851883840363, "全帯域_40ms_log_rms_mae_db": 0.4596666962731485, "全帯域_40ms_rms_envelope_nrmse": 0.04627492397776154, "全帯域_40ms_envelope_correlation": 0.9979175039236651, "全帯域_40ms_modulation_mae_db": 0.39371180052400273, "300–1000 Hz_10ms_log_rms_mae_db": 1.4436740302721254, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1163265787095338, "300–1000 Hz_10ms_envelope_correlation": 0.9891175323087709, "300–1000 Hz_10ms_modulation_mae_db": 0.10218542253445177, "300–1000 Hz_40ms_log_rms_mae_db": 0.8110253233786735, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06995272885885463, "300–1000 Hz_40ms_envelope_correlation": 0.9958490581271547, "300–1000 Hz_40ms_modulation_mae_db": 0.19856868898980432, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5949286094876, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1831061956271178, "1000–3000 Hz_10ms_envelope_correlation": 0.972521241474342, "1000–3000 Hz_10ms_modulation_mae_db": 1.1275084687944352, "1000–3000 Hz_40ms_log_rms_mae_db": 1.03239938561134, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10955662265136373, "1000–3000 Hz_40ms_envelope_correlation": 0.989367534236851, "1000–3000 Hz_40ms_modulation_mae_db": 1.0669000928857193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4119978999053124, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15780833266011624, "3000–6000 Hz_10ms_envelope_correlation": 0.9814204416158303, "3000–6000 Hz_10ms_modulation_mae_db": 1.2132827537484692, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7484952493868265, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06264826198287306, "3000–6000 Hz_40ms_envelope_correlation": 0.9972357798485131, "3000–6000 Hz_40ms_modulation_mae_db": 1.7303053885648094, "mr_stft_db": 5.689261712659331, "band_mae_db": 0.18753025563847314, "rms_gain_error_db": -0.09604151043144879, "output_peak": 0.4215982312083892}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9098328426024725, "全帯域_10ms_rms_envelope_nrmse": 0.08398820930978307, "全帯域_10ms_envelope_correlation": 0.9936355453363651, "全帯域_10ms_modulation_mae_db": 1.417768783244469, "全帯域_40ms_log_rms_mae_db": 0.43278646149261774, "全帯域_40ms_rms_envelope_nrmse": 0.04358978060349051, "全帯域_40ms_envelope_correlation": 0.9983627000026691, "全帯域_40ms_modulation_mae_db": 1.4204632095891843, "300–1000 Hz_10ms_log_rms_mae_db": 1.325711852917549, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1005008431923143, "300–1000 Hz_10ms_envelope_correlation": 0.99202524334587, "300–1000 Hz_10ms_modulation_mae_db": 0.9069627805775712, "300–1000 Hz_40ms_log_rms_mae_db": 0.7506438905023444, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05635792927756264, "300–1000 Hz_40ms_envelope_correlation": 0.9975130504282168, "300–1000 Hz_40ms_modulation_mae_db": 0.877155406271997, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5533617795006056, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18343938791467054, "1000–3000 Hz_10ms_envelope_correlation": 0.9738148122059446, "1000–3000 Hz_10ms_modulation_mae_db": 1.5777589281795883, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0076397886369517, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10474835606524832, "1000–3000 Hz_40ms_envelope_correlation": 0.9928112682222734, "1000–3000 Hz_40ms_modulation_mae_db": 1.3412386630343602, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4661453324256346, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15765634316032218, "3000–6000 Hz_10ms_envelope_correlation": 0.9813630474727263, "3000–6000 Hz_10ms_modulation_mae_db": 0.587173697955177, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8085480405106206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06452938547790499, "3000–6000 Hz_40ms_envelope_correlation": 0.9967368238132286, "3000–6000 Hz_40ms_modulation_mae_db": 0.7799837039718759, "mr_stft_db": 5.680686506354749, "band_mae_db": 0.27190637395289513, "rms_gain_error_db": -0.15736405553625407, "output_peak": 0.3956467144024004}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37680, "output_samples": 37680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.70995969442468, "spectral_convergence": 0.34856323012419227, "spectral_convergence_status": "ok", "amplitude_floor": 1.2997261015401668e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.413383684094471, "spectral_convergence": 0.3160272025386081, "spectral_convergence_status": "ok", "amplitude_floor": 1.1046780177944747e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9023506885920245, "spectral_convergence": 0.3353816632070949, "spectral_convergence_status": "ok", "amplitude_floor": 9.696590055390529e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.751691690693756e-05, "output_power": 8.230748321044358e-05, "delta_db": 0.2604283096008581, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016017256752476519, "output_power": 0.0015388111201342152, "delta_db": -0.1740282121002373, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006838056632308089, "output_power": 0.000669023316238268, "delta_db": -0.09491439574265367, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003361315194748372, "output_power": 0.0003425627910214336, "delta_db": 0.08230949649656151, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00035354415012418854, "output_power": 0.0003275437363189736, "delta_db": -0.33174356886403533, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00021020525507420865, "output_power": 0.0001952722342800335, "delta_db": -0.3200307363975369, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.21057578653364714, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14536616910923905, "pitch": null, "output_rms_dbfs": -25.00928969124741, "output_peak": 0.4406212946059799, "reference_rms_dbfs": -24.863923522138165, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.675231355703725}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00018834951637393047, "log_rms_mae_db": 0.8740516433023192, "rms_envelope_nrmse": 0.0890145861272344, "envelope_correlation": 0.9928204590997016, "rms_gain_error_db": -0.1453611309703585, "reference_mean_power": 0.00326985830085679, "output_mean_power": 0.0031622253788999317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.479011996203747e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005558024994761568, "output_power": 0.000558603658251455, "reference_active": true, "delta_db": 0.021832800414470655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011456565813557757, "output_power": 0.00010366442552151536, "reference_active": true, "delta_db": -0.43424709007688533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.821714301732997e-05, "output_power": 3.59492637070858e-05, "reference_active": true, "delta_db": -0.26568217734129357}], "active_band_count": 3, "active_band_mae_db": 0.2405873559442165}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00017471448177334785, "log_rms_mae_db": 0.45687006471892566, "rms_envelope_nrmse": 0.04753311382342967, "envelope_correlation": 0.9979948783886864, "rms_gain_error_db": -0.14535830715178755, "reference_mean_power": 0.0033120735889998944, "output_mean_power": 0.0032030531620127795}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 7.763101583949748e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004881183543297566, "output_power": 0.0004898637693192927, "reference_active": true, "delta_db": 0.015501815116565012}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.883974713754685e-05, "output_power": 5.1595453185498124e-05, "reference_active": true, "delta_db": -0.5705936628376158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7824879244994614e-06, "output_power": 1.493724694745738e-06, "reference_active": true, "delta_db": -0.7675603544406011}], "active_band_count": 3, "active_band_mae_db": 0.4512186107982606}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00017587844894922502, "log_rms_mae_db": 1.4491199147872624, "rms_envelope_nrmse": 0.11082622798702739, "envelope_correlation": 0.9902986035469615, "rms_gain_error_db": -0.1904319190205639, "reference_mean_power": 0.002090015296145697, "output_mean_power": 0.002000351299495804}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 6.927253047980857e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041075223634869497, "output_power": 0.0004078181246308756, "reference_active": true, "delta_db": -0.031134134497446748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.156045953139752e-05, "output_power": 7.483773598418037e-05, "reference_active": true, "delta_db": -0.3735902342627011}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2217147797459886e-05, "output_power": 2.912724317397035e-05, "reference_active": true, "delta_db": 1.1761107746720287}], "active_band_count": 3, "active_band_mae_db": 0.5269450478107255}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00016316895420677254, "log_rms_mae_db": 0.836368989989499, "rms_envelope_nrmse": 0.062087753182505, "envelope_correlation": 0.9970422410467376, "rms_gain_error_db": -0.19043154648295108, "reference_mean_power": 0.002117036355483854, "output_mean_power": 0.0020262132989380542}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 5.84659163450443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037072755253225205, "output_power": 0.00036888781451403286, "reference_active": true, "delta_db": -0.021605546691458765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.09979708848843e-05, "output_power": 3.660953145672406e-05, "reference_active": true, "delta_db": -0.49168192043507286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.223097583568823e-06, "output_power": 1.2300987545925254e-06, "reference_active": true, "delta_db": 0.024788706473034116}], "active_band_count": 3, "active_band_mae_db": 0.17935872453318857}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0001038394087983997, "log_rms_mae_db": 1.5001102913060922, "rms_envelope_nrmse": 0.18056445413290145, "envelope_correlation": 0.973230657355384, "rms_gain_error_db": -0.08454133982740444, "reference_mean_power": 0.0005521302842545539, "output_mean_power": 0.000541486252195982}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 2.3748882259523475e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000136307073642641, "output_power": 0.0001448449023910563, "reference_active": true, "delta_db": 0.26384821345903026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.293151120982887e-05, "output_power": 2.9090230064498747e-05, "reference_active": true, "delta_db": -0.5386450477142022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.586128694071059e-05, "output_power": 1.3544270286886427e-05, "reference_active": true, "delta_db": -0.6858280997205872}], "active_band_count": 3, "active_band_mae_db": 0.4961071202979399}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 8.776193320418374e-05, "log_rms_mae_db": 0.8747555684177639, "rms_envelope_nrmse": 0.1038781053979894, "envelope_correlation": 0.9902274613216157, "rms_gain_error_db": -0.08452448629715617, "reference_mean_power": 0.0005592400075632049, "output_mean_power": 0.0005484610418204148}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.9436382041482247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012615724913962246, "output_power": 0.00013495596652956845, "reference_active": true, "delta_db": 0.2927987925489043}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.519569809811637e-05, "output_power": 1.2810023350128955e-05, "reference_active": true, "delta_db": -0.7417073520613128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.080674445150989e-07, "output_power": 8.373393935825831e-07, "reference_active": true, "delta_db": -0.35216582326212753}], "active_band_count": 3, "active_band_mae_db": 0.46222398929078157}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 6.310171076800209e-05, "log_rms_mae_db": 1.4824014458390344, "rms_envelope_nrmse": 0.15282346931611474, "envelope_correlation": 0.9832819220386303, "rms_gain_error_db": -0.37746806874727123, "reference_mean_power": 0.00026303793349419837, "output_mean_power": 0.0002411412884357489}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 1.3356760090821468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.8507771208283054e-05, "output_power": 5.3240454166664954e-05, "reference_active": true, "delta_db": -0.4097180287734869}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1981263949052112e-05, "output_power": 1.1829783329140723e-05, "reference_active": true, "delta_db": -0.055258455551289457}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.256106518668102e-06, "output_power": 5.0732511953462004e-06, "reference_active": true, "delta_db": -2.6114197743629264}], "active_band_count": 3, "active_band_mae_db": 1.0254654195625676}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 5.861628173421479e-05, "log_rms_mae_db": 0.7614583007324865, "rms_envelope_nrmse": 0.07361424720626428, "envelope_correlation": 0.996984516048118, "rms_gain_error_db": -0.3774484536877803, "reference_mean_power": 0.00026641634082957415, "output_mean_power": 0.00024423956269484205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.157398705126346e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.177699204483558e-05, "output_power": 4.684074956657531e-05, "reference_active": true, "delta_db": -0.43512980508570864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.036512028004977e-06, "output_power": 6.36399135480841e-06, "reference_active": true, "delta_db": -0.43627853367854746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.588547226531691e-07, "output_power": 2.514131578954731e-07, "reference_active": true, "delta_db": -3.4691092225721523}], "active_band_count": 3, "active_band_mae_db": 1.4468391871121362}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8814954107416156, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "67a37fdd7f87d9f1bfac49f08dcc7c20f6690d5a2d5cf99e45b54dfc8a749480", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9764627213187447, "standard_deviation": 0.0037501158525561186, "minimum": 0.9724486897208283, "maximum": 0.9803777563341242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5369941279441206, "standard_deviation": 0.0433997051734042, "minimum": 1.4606689766829066, "maximum": 1.5949286094876, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9168748756216949, "standard_deviation": 0.33867955713259323, "minimum": 0.4961071202979399, "maximum": 1.5777589281795883, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17064501501590906, "standard_deviation": 0.012938039335137595, "minimum": 0.15615602969666603, "maximum": 0.18343938791467054, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919827339659223, "standard_deviation": 0.0022186178168044314, "minimum": 0.9887911499122235, "maximum": 0.9944870272200313, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.954340679356118, "standard_deviation": 0.05317864594242406, "minimum": 0.8747555684177639, "maximum": 1.03239938561134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7573215730810754, "standard_deviation": 0.360758671063336, "minimum": 0.3728255822757484, "maximum": 1.3412386630343602, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09850006268353036, "standard_deviation": 0.010253738791317326, "minimum": 0.08397729484993503, "maximum": 0.1114791088092096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9836539885001373, "standard_deviation": 0.002545803710888556, "minimum": 0.9804625581237053, "maximum": 0.9868970028470565, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4395978823765152, "standard_deviation": 0.03214141542298809, "minimum": 1.3985038845728883, "maximum": 1.4824014458390344, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.077257396854502, "standard_deviation": 0.24543187494834398, "minimum": 0.587173697955177, "maximum": 1.3217303470806216, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14929953499015522, "standard_deviation": 0.010255404861014075, "minimum": 0.13544666074595102, "maximum": 0.16139923423408617, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9968619353324384, "standard_deviation": 0.00023523078556086534, "minimum": 0.9965685383132561, "maximum": 0.9972357798485131, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7795643024920069, "standard_deviation": 0.04313955172219032, "minimum": 0.72417778638129, "maximum": 0.8353969169666571, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4086871886194123, "standard_deviation": 0.31696801974240796, "minimum": 0.7799837039718759, "maximum": 1.7303053885648094, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06940611544238229, "standard_deviation": 0.004664559696513673, "minimum": 0.06264826198287306, "maximum": 0.074161860362616, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9902236920876155, "standard_deviation": 0.0014590297781390022, "minimum": 0.9881519405592345, "maximum": 0.9922240389230498, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4160402316355407, "standard_deviation": 0.05083908371840316, "minimum": 1.325711852917549, "maximum": 1.4892079574847954, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4392140299861657, "standard_deviation": 0.29941516805682744, "minimum": 0.10218542253445177, "maximum": 0.9069627805775712, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11026268084159845, "standard_deviation": 0.008039649546037402, "minimum": 0.09832895935891242, "maximum": 0.12115385621999143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966905438925469, "standard_deviation": 0.0011337549199124581, "minimum": 0.9947079692677785, "maximum": 0.998162162585464, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7908465654719137, "standard_deviation": 0.05532226945485038, "minimum": 0.7396001538222683, "maximum": 0.9000627832119765, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4574437568301959, "standard_deviation": 0.32785533178449666, "minimum": 0.11411588738404242, "maximum": 1.0126171760323126, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06247385058970533, "standard_deviation": 0.010163554401122137, "minimum": 0.046882457282772225, "maximum": 0.07847277440525402, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2026770593582818, "standard_deviation": 0.04842722693624387, "minimum": 0.13362407396524856, "maximum": 0.27190637395289513, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6798347684653345, "standard_deviation": 0.008338908404425712, "minimum": 5.6649584373979005, "maximum": 5.689261712659331, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42667291473394303, "standard_deviation": 0.018058528999295845, "minimum": 0.3956467144024004, "maximum": 0.45007915594262266, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10453572724235921, "standard_deviation": 0.034140520625796485, "minimum": -0.15736405553625407, "maximum": -0.0534621295557314, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930209012007102, "standard_deviation": 0.0007785105033857306, "minimum": 0.9914414247830072, "maximum": 0.993957385642759, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8989064321049011, "standard_deviation": 0.035978109758125834, "minimum": 0.8520543242234748, "maximum": 0.9648093965345872, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5480496484081706, "standard_deviation": 0.37421783165922823, "minimum": 0.2405873559442165, "maximum": 1.417768783244469, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08706183485064672, "standard_deviation": 0.004684874723952309, "minimum": 0.08093682489025547, "maximum": 0.09629944750880663, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981566185974563, "standard_deviation": 0.0004681639943266217, "minimum": 0.9973042176004406, "maximum": 0.9987034873255892, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.446882943516804, "standard_deviation": 0.013649515408194372, "minimum": 0.42913995252727016, "maximum": 0.4596666962731485, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6583145355620417, "standard_deviation": 0.34860796313491355, "minimum": 0.3681549872114527, "maximum": 1.4204632095891843, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04397113226020523, "standard_deviation": 0.005354935162041051, "minimum": 0.036851265421212635, "maximum": 0.052885767598234525, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0022", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 1.7132189273834229, "predicted_mos": 1.7132189273834229, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "db86ff31c46f5a8a0d54b2ea5f0ab21ce68e35e1baa91102c808530dc197e229", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8838633997365832, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8977866091192278, "全帯域_10ms_rms_envelope_nrmse": 0.09117415640415916, "全帯域_10ms_envelope_correlation": 0.992439292594727, "全帯域_10ms_modulation_mae_db": 0.20312011520948295, "全帯域_40ms_log_rms_mae_db": 0.4782796845106746, "全帯域_40ms_rms_envelope_nrmse": 0.05253948885925617, "全帯域_40ms_envelope_correlation": 0.997456009061027, "全帯域_40ms_modulation_mae_db": 0.38467926748850284, "300–1000 Hz_10ms_log_rms_mae_db": 1.4762312203208163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11032976070732563, "300–1000 Hz_10ms_envelope_correlation": 0.990350929942063, "300–1000 Hz_10ms_modulation_mae_db": 0.39501060134858457, "300–1000 Hz_40ms_log_rms_mae_db": 0.87998199989028, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06571642363386025, "300–1000 Hz_40ms_envelope_correlation": 0.9965749071582175, "300–1000 Hz_40ms_modulation_mae_db": 0.2364835320085117, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5647788662899187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19003676083533963, "1000–3000 Hz_10ms_envelope_correlation": 0.9703521290017787, "1000–3000 Hz_10ms_modulation_mae_db": 0.37598995683637604, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8945430885797699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11273976486979558, "1000–3000 Hz_40ms_envelope_correlation": 0.9885161767124695, "1000–3000 Hz_40ms_modulation_mae_db": 0.3358868233989703, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4638517779336064, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15231397927701482, "3000–6000 Hz_10ms_envelope_correlation": 0.9835027898204907, "3000–6000 Hz_10ms_modulation_mae_db": 1.1164775452636018, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7341240160833438, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07491413233902475, "3000–6000 Hz_40ms_envelope_correlation": 0.9969955950725871, "3000–6000 Hz_40ms_modulation_mae_db": 1.5380919658159475, "mr_stft_db": 5.6254848614417865, "band_mae_db": 0.19212392080196602, "rms_gain_error_db": -0.14061619731699807, "output_peak": 0.44353874471699173}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.883283161858277, "全帯域_10ms_rms_envelope_nrmse": 0.07903393524468269, "全帯域_10ms_envelope_correlation": 0.994219464317598, "全帯域_10ms_modulation_mae_db": 0.30473182483715283, "全帯域_40ms_log_rms_mae_db": 0.447398716614361, "全帯域_40ms_rms_envelope_nrmse": 0.03530497500959228, "全帯域_40ms_envelope_correlation": 0.9987856021507469, "全帯域_40ms_modulation_mae_db": 0.394390680871207, "300–1000 Hz_10ms_log_rms_mae_db": 1.367433048042963, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09756144012733671, "300–1000 Hz_10ms_envelope_correlation": 0.9923395023825128, "300–1000 Hz_10ms_modulation_mae_db": 0.08884242293431481, "300–1000 Hz_40ms_log_rms_mae_db": 0.7840487932020653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04634339055286151, "300–1000 Hz_40ms_envelope_correlation": 0.9981884180015721, "300–1000 Hz_40ms_modulation_mae_db": 0.10150673383678684, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5198883068187985, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1709493715328798, "1000–3000 Hz_10ms_envelope_correlation": 0.9760801536691102, "1000–3000 Hz_10ms_modulation_mae_db": 0.736578318227088, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9692291590230724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10318863097346415, "1000–3000 Hz_40ms_envelope_correlation": 0.9904458984407194, "1000–3000 Hz_40ms_modulation_mae_db": 0.3685481367291237, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5197519787243334, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13835653671424789, "3000–6000 Hz_10ms_envelope_correlation": 0.9863665375805221, "3000–6000 Hz_10ms_modulation_mae_db": 1.197204564365195, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9079853759250187, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07957772477217759, "3000–6000 Hz_40ms_envelope_correlation": 0.9960348448597361, "3000–6000 Hz_40ms_modulation_mae_db": 1.6288784730486237, "mr_stft_db": 5.637824905718897, "band_mae_db": 0.1939724023837116, "rms_gain_error_db": -0.06523652113856532, "output_peak": 0.4170853049972313}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8604210502897156, "全帯域_10ms_rms_envelope_nrmse": 0.09565349093536728, "全帯域_10ms_envelope_correlation": 0.9915407380771749, "全帯域_10ms_modulation_mae_db": 0.7222756686423916, "全帯域_40ms_log_rms_mae_db": 0.4127595657426533, "全帯域_40ms_rms_envelope_nrmse": 0.051933247687374756, "全帯域_40ms_envelope_correlation": 0.9973749006537281, "全帯域_40ms_modulation_mae_db": 0.9266366136087579, "300–1000 Hz_10ms_log_rms_mae_db": 1.5268481172828132, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12154628613330705, "300–1000 Hz_10ms_envelope_correlation": 0.9880787151480802, "300–1000 Hz_10ms_modulation_mae_db": 0.44987102368503074, "300–1000 Hz_40ms_log_rms_mae_db": 0.8894643114221407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07999803140654388, "300–1000 Hz_40ms_envelope_correlation": 0.9944792786071232, "300–1000 Hz_40ms_modulation_mae_db": 0.5577404012386485, "1000–3000 Hz_10ms_log_rms_mae_db": 1.458241438192873, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15779767921724247, "1000–3000 Hz_10ms_envelope_correlation": 0.9803281624519947, "1000–3000 Hz_10ms_modulation_mae_db": 1.0111340161462055, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8877178938801602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0882391695977783, "1000–3000 Hz_40ms_envelope_correlation": 0.9943517584149094, "1000–3000 Hz_40ms_modulation_mae_db": 0.9296672226440776, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4352011181913387, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.152314286233315, "3000–6000 Hz_10ms_envelope_correlation": 0.9833086790732825, "3000–6000 Hz_10ms_modulation_mae_db": 0.9657897913108618, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7272100011064218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0752788826329657, "3000–6000 Hz_40ms_envelope_correlation": 0.9966927471683418, "3000–6000 Hz_40ms_modulation_mae_db": 1.3575022385756954, "mr_stft_db": 5.649416338455755, "band_mae_db": 0.27486019696709457, "rms_gain_error_db": -0.10422838372619692, "output_peak": 0.43833648051276963}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9184469699007388, "全帯域_10ms_rms_envelope_nrmse": 0.08886243950950361, "全帯域_10ms_envelope_correlation": 0.9927036695922665, "全帯域_10ms_modulation_mae_db": 0.44927399152204317, "全帯域_40ms_log_rms_mae_db": 0.44125162164017157, "全帯域_40ms_rms_envelope_nrmse": 0.044820593694334165, "全帯域_40ms_envelope_correlation": 0.9980181594333708, "全帯域_40ms_modulation_mae_db": 0.6438199355902109, "300–1000 Hz_10ms_log_rms_mae_db": 1.3825069909871341, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11018430844126251, "300–1000 Hz_10ms_envelope_correlation": 0.9903441184580615, "300–1000 Hz_10ms_modulation_mae_db": 0.2688939282126279, "300–1000 Hz_40ms_log_rms_mae_db": 0.7392466703921345, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06218632051824551, "300–1000 Hz_40ms_envelope_correlation": 0.9969012475174523, "300–1000 Hz_40ms_modulation_mae_db": 0.47687013993747884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5007727617375162, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15153683915542493, "1000–3000 Hz_10ms_envelope_correlation": 0.9817416986170151, "1000–3000 Hz_10ms_modulation_mae_db": 0.9775136171302709, "1000–3000 Hz_40ms_log_rms_mae_db": 0.884993256444944, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08840284524817593, "1000–3000 Hz_40ms_envelope_correlation": 0.9932368021060335, "1000–3000 Hz_40ms_modulation_mae_db": 1.1752254004812888, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4565587173379633, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16379519615147223, "3000–6000 Hz_10ms_envelope_correlation": 0.9799083675809696, "3000–6000 Hz_10ms_modulation_mae_db": 1.1829586542832473, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7055147526692781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07494175448435236, "3000–6000 Hz_40ms_envelope_correlation": 0.9956098439050779, "3000–6000 Hz_40ms_modulation_mae_db": 1.344596389870639, "mr_stft_db": 5.628931316343722, "band_mae_db": 0.13275003855265988, "rms_gain_error_db": -0.05472957199189474, "output_peak": 0.45623465568276467}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9589141308404493, "全帯域_10ms_rms_envelope_nrmse": 0.0830105085629263, "全帯域_10ms_envelope_correlation": 0.9936528596762925, "全帯域_10ms_modulation_mae_db": 0.256501037097848, "全帯域_40ms_log_rms_mae_db": 0.4365871950811423, "全帯域_40ms_rms_envelope_nrmse": 0.03723891025470678, "全帯域_40ms_envelope_correlation": 0.9987092556318423, "全帯域_40ms_modulation_mae_db": 0.686192235443151, "300–1000 Hz_10ms_log_rms_mae_db": 1.354949583165341, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10778229882244368, "300–1000 Hz_10ms_envelope_correlation": 0.9906431062547701, "300–1000 Hz_10ms_modulation_mae_db": 0.7929036464891812, "300–1000 Hz_40ms_log_rms_mae_db": 0.757138341980865, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0557532832076099, "300–1000 Hz_40ms_envelope_correlation": 0.9973432213624125, "300–1000 Hz_40ms_modulation_mae_db": 0.972143920314485, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5107133130305943, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15585258533603474, "1000–3000 Hz_10ms_envelope_correlation": 0.9807373106573736, "1000–3000 Hz_10ms_modulation_mae_db": 0.9039753795601984, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8925307264706572, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08802759688840787, "1000–3000 Hz_40ms_envelope_correlation": 0.9942382945123331, "1000–3000 Hz_40ms_modulation_mae_db": 0.6624193135029748, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4194092228065012, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13942169077514244, "3000–6000 Hz_10ms_envelope_correlation": 0.98575100790464, "3000–6000 Hz_10ms_modulation_mae_db": 1.205949491582033, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7939344461790132, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07011393319896322, "3000–6000 Hz_40ms_envelope_correlation": 0.9966629328887701, "3000–6000 Hz_40ms_modulation_mae_db": 1.6309326008366132, "mr_stft_db": 5.635358284951317, "band_mae_db": 0.1551595139515719, "rms_gain_error_db": -0.1001960197032513, "output_peak": 0.4235843406969138}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.892502621075453, "全帯域_10ms_rms_envelope_nrmse": 0.08592710079550288, "全帯域_10ms_envelope_correlation": 0.9932024883333472, "全帯域_10ms_modulation_mae_db": 0.5381567222895405, "全帯域_40ms_log_rms_mae_db": 0.4482057458596649, "全帯域_40ms_rms_envelope_nrmse": 0.040322915636055556, "全帯域_40ms_envelope_correlation": 0.9984727278903781, "全帯域_40ms_modulation_mae_db": 0.5446019190757175, "300–1000 Hz_10ms_log_rms_mae_db": 1.4011073960404619, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11822159865970457, "300–1000 Hz_10ms_envelope_correlation": 0.9887184113990956, "300–1000 Hz_10ms_modulation_mae_db": 0.41808330165225643, "300–1000 Hz_40ms_log_rms_mae_db": 0.7711353679316665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0684524261072191, "300–1000 Hz_40ms_envelope_correlation": 0.9959612226110629, "300–1000 Hz_40ms_modulation_mae_db": 0.36842761620965064, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5634146775465487, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15291369583096256, "1000–3000 Hz_10ms_envelope_correlation": 0.9814804093983261, "1000–3000 Hz_10ms_modulation_mae_db": 0.7798765652975724, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9558991761531221, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08914286115290085, "1000–3000 Hz_40ms_envelope_correlation": 0.9939877098872871, "1000–3000 Hz_40ms_modulation_mae_db": 0.7167654349117644, "3000–6000 Hz_10ms_log_rms_mae_db": 1.423347540855732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14017397455328312, "3000–6000 Hz_10ms_envelope_correlation": 0.9864145460205085, "3000–6000 Hz_10ms_modulation_mae_db": 1.257093853201899, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7732057247143825, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07772481059249693, "3000–6000 Hz_40ms_envelope_correlation": 0.9967833867076944, "3000–6000 Hz_40ms_modulation_mae_db": 1.651947028676963, "mr_stft_db": 5.619344394929336, "band_mae_db": 0.19325412671706177, "rms_gain_error_db": -0.10396431894090663, "output_peak": 0.4331639834369158}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.929633907102086, "全帯域_10ms_rms_envelope_nrmse": 0.08378126133322779, "全帯域_10ms_envelope_correlation": 0.9935503553737874, "全帯域_10ms_modulation_mae_db": 0.4578254696290376, "全帯域_40ms_log_rms_mae_db": 0.4678419746509928, "全帯域_40ms_rms_envelope_nrmse": 0.04656953587303623, "全帯域_40ms_envelope_correlation": 0.9979310150820819, "全帯域_40ms_modulation_mae_db": 0.4299587402223726, "300–1000 Hz_10ms_log_rms_mae_db": 1.4270732228862484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11850083059897322, "300–1000 Hz_10ms_envelope_correlation": 0.9887496809796451, "300–1000 Hz_10ms_modulation_mae_db": 0.08875493841507337, "300–1000 Hz_40ms_log_rms_mae_db": 0.7893684723211473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07280596528703767, "300–1000 Hz_40ms_envelope_correlation": 0.9955642628636254, "300–1000 Hz_40ms_modulation_mae_db": 0.06256598181324347, "1000–3000 Hz_10ms_log_rms_mae_db": 1.607615019851406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18284248916015106, "1000–3000 Hz_10ms_envelope_correlation": 0.9725920128258614, "1000–3000 Hz_10ms_modulation_mae_db": 0.9349599300119626, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9794991478191509, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10867865879519735, "1000–3000 Hz_40ms_envelope_correlation": 0.9894390588980209, "1000–3000 Hz_40ms_modulation_mae_db": 0.9248351494097435, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4216464614566189, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1576988273426667, "3000–6000 Hz_10ms_envelope_correlation": 0.9815667554500793, "3000–6000 Hz_10ms_modulation_mae_db": 1.2543182222543001, "3000–6000 Hz_40ms_log_rms_mae_db": 0.728230903398044, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06573635323313283, "3000–6000 Hz_40ms_envelope_correlation": 0.9971377372957958, "3000–6000 Hz_40ms_modulation_mae_db": 1.8035204414951613, "mr_stft_db": 5.637460506184261, "band_mae_db": 0.21130917971264393, "rms_gain_error_db": -0.11194282424114674, "output_peak": 0.42308491674676413}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9249913761980715, "全帯域_10ms_rms_envelope_nrmse": 0.0837973051606007, "全帯域_10ms_envelope_correlation": 0.9936183267227736, "全帯域_10ms_modulation_mae_db": 1.468693645805563, "全帯域_40ms_log_rms_mae_db": 0.4440960025641629, "全帯域_40ms_rms_envelope_nrmse": 0.042702174403510394, "全帯域_40ms_envelope_correlation": 0.9983807243469188, "全帯域_40ms_modulation_mae_db": 1.3731168427199751, "300–1000 Hz_10ms_log_rms_mae_db": 1.3417851826939735, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0996593023058216, "300–1000 Hz_10ms_envelope_correlation": 0.9920768687136874, "300–1000 Hz_10ms_modulation_mae_db": 0.9380628094311624, "300–1000 Hz_40ms_log_rms_mae_db": 0.7809920821605473, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05340507529883528, "300–1000 Hz_40ms_envelope_correlation": 0.9976843939327317, "300–1000 Hz_40ms_modulation_mae_db": 0.7952671167242192, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5333773272393416, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18138305088764306, "1000–3000 Hz_10ms_envelope_correlation": 0.9744070711069064, "1000–3000 Hz_10ms_modulation_mae_db": 1.6017030349785122, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9740778751278681, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1043845341640325, "1000–3000 Hz_40ms_envelope_correlation": 0.9927865140016479, "1000–3000 Hz_40ms_modulation_mae_db": 1.4366119494737075, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4539608805387771, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15658653514815907, "3000–6000 Hz_10ms_envelope_correlation": 0.9817319725950121, "3000–6000 Hz_10ms_modulation_mae_db": 0.8140618903030256, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7723482108604784, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06564871937342703, "3000–6000 Hz_40ms_envelope_correlation": 0.9969284710535545, "3000–6000 Hz_40ms_modulation_mae_db": 1.053805424302649, "mr_stft_db": 5.624805853903392, "band_mae_db": 0.2802804369898407, "rms_gain_error_db": -0.1410337196646212, "output_peak": 0.4000423250758906}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37680, "output_samples": 37680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.681085814863723, "spectral_convergence": 0.3493855962023904, "spectral_convergence_status": "ok", "amplitude_floor": 1.2997261015401668e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.366768574019465, "spectral_convergence": 0.31544792811359246, "spectral_convergence_status": "ok", "amplitude_floor": 1.1046780177944747e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.828600195442172, "spectral_convergence": 0.33668165332148026, "spectral_convergence_status": "ok", "amplitude_floor": 9.696590055390529e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.751691690693756e-05, "output_power": 8.067914038200633e-05, "delta_db": 0.1736477113542104, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016017256752476519, "output_power": 0.0015438696279387793, "delta_db": -0.15977513708161775, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006838056632308089, "output_power": 0.0006715192102340891, "delta_db": -0.07874252173037716, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003361315194748372, "output_power": 0.0003404551137869067, "delta_db": 0.05550623141094762, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00035354415012418854, "output_power": 0.00033026888837577105, "delta_db": -0.2957599094780145, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00021020525507420865, "output_power": 0.00019218184910610884, "delta_db": -0.38931201375662866, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19212392080196602, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14061619731699807, "pitch": null, "output_rms_dbfs": -25.004539719455167, "output_peak": 0.44353874471699173, "reference_rms_dbfs": -24.863923522138165, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6254848614417865}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00018834951637393047, "log_rms_mae_db": 0.8977866091192278, "rms_envelope_nrmse": 0.09117415640415916, "envelope_correlation": 0.992439292594727, "rms_gain_error_db": -0.14061099387324613, "reference_mean_power": 0.00326985830085679, "output_mean_power": 0.003165685985098408}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.479011996203747e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005558024994761568, "output_power": 0.0005601076243110149, "reference_active": true, "delta_db": 0.03350988929906093}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011456565813557757, "output_power": 0.00010597648155356192, "reference_active": true, "delta_db": -0.33844957574487544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.821714301732997e-05, "output_power": 3.6184129538489606e-05, "reference_active": true, "delta_db": -0.23740088058451242}], "active_band_count": 3, "active_band_mae_db": 0.20312011520948295}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00017471448177334785, "log_rms_mae_db": 0.4782796845106746, "rms_envelope_nrmse": 0.05253948885925617, "envelope_correlation": 0.997456009061027, "rms_gain_error_db": -0.14060755163755873, "reference_mean_power": 0.0033120735889998944, "output_mean_power": 0.0032065589050171103}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 7.763101583949748e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004881183543297566, "output_power": 0.0004939273050525283, "reference_active": true, "delta_db": 0.051378968879036105}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.883974713754685e-05, "output_power": 5.317388642603379e-05, "reference_active": true, "delta_db": -0.4397239406661734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7824879244994614e-06, "output_power": 1.5301468307384279e-06, "reference_active": true, "delta_db": -0.662934892920299}], "active_band_count": 3, "active_band_mae_db": 0.38467926748850284}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00017587844894922502, "log_rms_mae_db": 1.4762312203208163, "rms_envelope_nrmse": 0.11032976070732563, "envelope_correlation": 0.990350929942063, "rms_gain_error_db": -0.17296386758973795, "reference_mean_power": 0.002090015296145697, "output_mean_power": 0.002008413249835383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 6.927253047980857e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041075223634869497, "output_power": 0.00040766816281918004, "reference_active": true, "delta_db": -0.032731404471252744}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.156045953139752e-05, "output_power": 7.498027870547057e-05, "reference_active": true, "delta_db": -0.3653261365038609}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2217147797459886e-05, "output_power": 2.6630893048500517e-05, "reference_active": true, "delta_db": 0.7869742630706401}], "active_band_count": 3, "active_band_mae_db": 0.39501060134858457}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00016316895420677254, "log_rms_mae_db": 0.87998199989028, "rms_envelope_nrmse": 0.06571642363386025, "envelope_correlation": 0.9965749071582175, "rms_gain_error_db": -0.17296378129624423, "reference_mean_power": 0.002117036355483854, "output_mean_power": 0.0020343793459608774}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 5.84659163450443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037072755253225205, "output_power": 0.0003704253259448045, "reference_active": true, "delta_db": -0.003541923678630392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.09979708848843e-05, "output_power": 3.7642434199004086e-05, "reference_active": true, "delta_db": -0.3708466278355616}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.223097583568823e-06, "output_power": 1.1322827830770938e-06, "reference_active": true, "delta_db": -0.3350620445113431}], "active_band_count": 3, "active_band_mae_db": 0.2364835320085117}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0001038394087983997, "log_rms_mae_db": 1.5647788662899187, "rms_envelope_nrmse": 0.19003676083533963, "envelope_correlation": 0.9703521290017787, "rms_gain_error_db": -0.10460070037672999, "reference_mean_power": 0.0005521302842545539, "output_mean_power": 0.0005389909817317801}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 2.3748882259523475e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000136307073642641, "output_power": 0.00014780663291439367, "reference_active": true, "delta_db": 0.35175529640171566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.293151120982887e-05, "output_power": 3.005985173848006e-05, "reference_active": true, "delta_db": -0.3962482649155668}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.586128694071059e-05, "output_power": 1.4532548831978217e-05, "reference_active": true, "delta_db": -0.37996630919184565}], "active_band_count": 3, "active_band_mae_db": 0.37598995683637604}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 8.776193320418374e-05, "log_rms_mae_db": 0.8945430885797699, "rms_envelope_nrmse": 0.11273976486979558, "envelope_correlation": 0.9885161767124695, "rms_gain_error_db": -0.1045763332787421, "reference_mean_power": 0.0005592400075632049, "output_mean_power": 0.0005459345747145147}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.9436382041482247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012615724913962246, "output_power": 0.00013866241638398302, "reference_active": true, "delta_db": 0.41046553481902387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.519569809811637e-05, "output_power": 1.3381366955773672e-05, "reference_active": true, "delta_db": -0.552201761100279}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.080674445150989e-07, "output_power": 9.175239794072699e-07, "reference_active": true, "delta_db": 0.04499317427760823}], "active_band_count": 3, "active_band_mae_db": 0.3358868233989703}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 6.310171076800209e-05, "log_rms_mae_db": 1.4638517779336064, "rms_envelope_nrmse": 0.15231397927701482, "envelope_correlation": 0.9835027898204907, "rms_gain_error_db": -0.40289982124470913, "reference_mean_power": 0.00026303793349419837, "output_mean_power": 0.00023973332107557845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 1.3356760090821468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.8507771208283054e-05, "output_power": 5.3038498324870114e-05, "reference_active": true, "delta_db": -0.4262233501363738}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1981263949052112e-05, "output_power": 1.1643635923502047e-05, "reference_active": true, "delta_db": -0.12414018508404281}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.256106518668102e-06, "output_power": 4.858715095761784e-06, "reference_active": true, "delta_db": -2.7990691005703883}], "active_band_count": 3, "active_band_mae_db": 1.1164775452636018}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 5.861628173421479e-05, "log_rms_mae_db": 0.7341240160833438, "rms_envelope_nrmse": 0.07491413233902475, "envelope_correlation": 0.9969955950725871, "rms_gain_error_db": -0.40289229020744804, "reference_mean_power": 0.00026641634082957415, "output_mean_power": 0.00024281282962354496}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.157398705126346e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.177699204483558e-05, "output_power": 4.6541673561683954e-05, "reference_active": true, "delta_db": -0.4629482099508223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.036512028004977e-06, "output_power": 6.308819851379395e-06, "reference_active": true, "delta_db": -0.4740930808804158}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.588547226531691e-07, "output_power": 2.396489136665733e-07, "reference_active": true, "delta_db": -3.677234606616604}], "active_band_count": 3, "active_band_mae_db": 1.5380919658159475}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8816144916944673, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5fdb8dd1c72e751b5fa1a88dd3d517a482ab78a403e5e26fb37ebee920684ca3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9772148684660458, "standard_deviation": 0.004446933453831569, "minimum": 0.9703521290017787, "maximum": 0.9817416986170151, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5323502138383747, "standard_deviation": 0.045973851424174955, "minimum": 1.458241438192873, "maximum": 1.607615019851406, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9152163522735233, "standard_deviation": 0.3434810853588364, "minimum": 0.37598995683637604, "maximum": 1.6017030349785122, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16791405899445977, "standard_deviation": 0.01532478543349941, "minimum": 0.15153683915542493, "maximum": 0.19003676083533963, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9921252766216776, "standard_deviation": 0.0023184085093605135, "minimum": 0.9885161767124695, "maximum": 0.9943517584149094, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.929811290437343, "standard_deviation": 0.04322194639784029, "minimum": 0.884993256444944, "maximum": 0.9794991478191509, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8187449288189563, "standard_deviation": 0.3785429752581502, "minimum": 0.3358868233989703, "maximum": 1.4366119494737075, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09785050771121906, "standard_deviation": 0.010447670473395596, "minimum": 0.08802759688840787, "maximum": 0.11273976486979558, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.983568832003188, "standard_deviation": 0.002436054783097661, "minimum": 0.9799083675809696, "maximum": 0.9864145460205085, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.449215962230609, "standard_deviation": 0.03333826028606475, "minimum": 1.4194092228065012, "maximum": 1.5197519787243334, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1242317515705205, "standard_deviation": 0.15647721092493447, "minimum": 0.8140618903030256, "maximum": 1.257093853201899, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15008262827441265, "standard_deviation": 0.009618726270004471, "minimum": 0.13835653671424789, "maximum": 0.16379519615147223, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9966056948689447, "standard_deviation": 0.0005208911781979564, "minimum": 0.9956098439050779, "maximum": 0.9971377372957958, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7678191788669975, "standard_deviation": 0.06393259774759419, "minimum": 0.7055147526692781, "maximum": 0.9079853759250187, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5011593203277864, "standard_deviation": 0.2372314432938613, "minimum": 1.053805424302649, "maximum": 1.8035204414951613, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07299203882831756, "standard_deviation": 0.005255591351796982, "minimum": 0.06564871937342703, "maximum": 0.07957772477217759, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9901626666597395, "standard_deviation": 0.001564208105598277, "minimum": 0.9880787151480802, "maximum": 0.9923395023825128, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4097418451774688, "standard_deviation": 0.06398852340022346, "minimum": 1.3417851826939735, "maximum": 1.5268481172828132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4300528340210289, "standard_deviation": 0.3048137198690226, "minimum": 0.08875493841507337, "maximum": 0.9380628094311624, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11047322822452188, "standard_deviation": 0.00876757565442766, "minimum": 0.09756144012733671, "maximum": 0.12154628613330705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965871190067748, "standard_deviation": 0.0012141816464884734, "minimum": 0.9944792786071232, "maximum": 0.9981884180015721, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7989220049126058, "standard_deviation": 0.05539689757705431, "minimum": 0.7392466703921345, "maximum": 0.8894643114221407, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.446375680260378, "standard_deviation": 0.3224603141148646, "minimum": 0.06256598181324347, "maximum": 0.972143920314485, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06308261450152663, "standard_deviation": 0.01098587099309498, "minimum": 0.04634339055286151, "maximum": 0.07999803140654388, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2042137270095688, "standard_deviation": 0.051702375053418625, "minimum": 0.13275003855265988, "maximum": 0.2802804369898407, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.632328307741059, "standard_deviation": 0.009571350234933504, "minimum": 5.619344394929336, "maximum": 5.649416338455755, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4293838439832802, "standard_deviation": 0.017325686853026906, "minimum": 0.4000423250758906, "maximum": 0.45623465568276467, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10274344459044762, "standard_deviation": 0.03095481815948582, "minimum": -0.1410337196646212, "maximum": -0.05472957199189474, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931158993359959, "standard_deviation": 0.0008505232081427667, "minimum": 0.9915407380771749, "maximum": 0.994219464317598, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9082474782980023, "standard_deviation": 0.03088409527563128, "minimum": 0.8604210502897156, "maximum": 0.9589141308404493, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5500723093791324, "standard_deviation": 0.4069192247898663, "minimum": 0.20312011520948295, "maximum": 1.468693645805563, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0864050247432463, "standard_deviation": 0.005265603732675004, "minimum": 0.07903393524468269, "maximum": 0.09565349093536728, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981410492812617, "standard_deviation": 0.0005373535411885699, "minimum": 0.9973749006537281, "maximum": 0.9987856021507469, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.44705256333297794, "standard_deviation": 0.01977266570838858, "minimum": 0.4127595657426533, "maximum": 0.4782796845106746, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6729245293774868, "standard_deviation": 0.3363305579617944, "minimum": 0.38467926748850284, "maximum": 1.3731168427199751, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04392898017723329, "standard_deviation": 0.006320216110285468, "minimum": 0.03530497500959228, "maximum": 0.05253948885925617, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0022", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.6920504570007324, "predicted_mos": 1.6920504570007324, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "6389279e8d70e677d8c6edc55c32e33b2fb1da340adab5b6030a9cb56f3e67c5", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.040504700038582, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8612638835979949, "全帯域_10ms_rms_envelope_nrmse": 0.08186736916678636, "全帯域_10ms_envelope_correlation": 0.9939204294418261, "全帯域_10ms_modulation_mae_db": 0.8216293452297606, "全帯域_40ms_log_rms_mae_db": 0.47372762018871534, "全帯域_40ms_rms_envelope_nrmse": 0.040957595947494, "全帯域_40ms_envelope_correlation": 0.9985199102964882, "全帯域_40ms_modulation_mae_db": 0.8549996529674627, "300–1000 Hz_10ms_log_rms_mae_db": 1.4230870098861284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11518081296448877, "300–1000 Hz_10ms_envelope_correlation": 0.9899789603373476, "300–1000 Hz_10ms_modulation_mae_db": 0.3819440558958389, "300–1000 Hz_40ms_log_rms_mae_db": 0.8362201721919739, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07677221739485231, "300–1000 Hz_40ms_envelope_correlation": 0.9958110052459791, "300–1000 Hz_40ms_modulation_mae_db": 0.45164109951428716, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6085420418189638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18763272552370158, "1000–3000 Hz_10ms_envelope_correlation": 0.9718069724948951, "1000–3000 Hz_10ms_modulation_mae_db": 0.7992160274097234, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9576628491162452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09999866689690125, "1000–3000 Hz_40ms_envelope_correlation": 0.9912053771361082, "1000–3000 Hz_40ms_modulation_mae_db": 0.5042292344041198, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8384190382465304, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1815048768339431, "3000–6000 Hz_10ms_envelope_correlation": 0.9756232741149072, "3000–6000 Hz_10ms_modulation_mae_db": 1.6719566278340885, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0149234936794194, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.078140137990075, "3000–6000 Hz_40ms_envelope_correlation": 0.9951648628096672, "3000–6000 Hz_40ms_modulation_mae_db": 1.9630011118211677, "mr_stft_db": 5.898899932713216, "band_mae_db": 0.18412467352516412, "rms_gain_error_db": -0.13710881612574946, "output_peak": 0.45244511569105095}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8739374538301848, "全帯域_10ms_rms_envelope_nrmse": 0.08329419088215169, "全帯域_10ms_envelope_correlation": 0.9936793412858056, "全帯域_10ms_modulation_mae_db": 0.6102188894021784, "全帯域_40ms_log_rms_mae_db": 0.4241895600966414, "全帯域_40ms_rms_envelope_nrmse": 0.038739368693500954, "全帯域_40ms_envelope_correlation": 0.9986721194557397, "全帯域_40ms_modulation_mae_db": 0.5348581078647802, "300–1000 Hz_10ms_log_rms_mae_db": 1.4155350390817234, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11457107003926412, "300–1000 Hz_10ms_envelope_correlation": 0.9895255696977542, "300–1000 Hz_10ms_modulation_mae_db": 0.31870968082318, "300–1000 Hz_40ms_log_rms_mae_db": 0.8002977969490235, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06331685428230592, "300–1000 Hz_40ms_envelope_correlation": 0.9967527604724429, "300–1000 Hz_40ms_modulation_mae_db": 0.5326515417565681, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6988074583360133, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1939941705453501, "1000–3000 Hz_10ms_envelope_correlation": 0.9692974700731696, "1000–3000 Hz_10ms_modulation_mae_db": 0.6580588130199235, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0484991063230515, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10307695845087582, "1000–3000 Hz_40ms_envelope_correlation": 0.9915394492404752, "1000–3000 Hz_40ms_modulation_mae_db": 0.30052805156232565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.783772166183202, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15292663368320314, "3000–6000 Hz_10ms_envelope_correlation": 0.9824959869315969, "3000–6000 Hz_10ms_modulation_mae_db": 1.3110970220794915, "3000–6000 Hz_40ms_log_rms_mae_db": 1.04470580548262, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08779909638380429, "3000–6000 Hz_40ms_envelope_correlation": 0.9939397252545992, "3000–6000 Hz_40ms_modulation_mae_db": 1.433321723510052, "mr_stft_db": 5.902036124468032, "band_mae_db": 0.2518463576478391, "rms_gain_error_db": -0.12553921211758165, "output_peak": 0.4032602590631034}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9111422376568596, "全帯域_10ms_rms_envelope_nrmse": 0.09869286265030075, "全帯域_10ms_envelope_correlation": 0.9909987732306356, "全帯域_10ms_modulation_mae_db": 0.44385915278484483, "全帯域_40ms_log_rms_mae_db": 0.45287902515784534, "全帯域_40ms_rms_envelope_nrmse": 0.04992683481926282, "全帯域_40ms_envelope_correlation": 0.9974886005299538, "全帯域_40ms_modulation_mae_db": 0.4224567271401201, "300–1000 Hz_10ms_log_rms_mae_db": 1.3870585996205012, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11969071189078767, "300–1000 Hz_10ms_envelope_correlation": 0.9886373284709064, "300–1000 Hz_10ms_modulation_mae_db": 0.6968825452527829, "300–1000 Hz_40ms_log_rms_mae_db": 0.77950910684302, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06003326514782183, "300–1000 Hz_40ms_envelope_correlation": 0.996949739012684, "300–1000 Hz_40ms_modulation_mae_db": 0.8258399474039111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7279488381993766, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16263317447083064, "1000–3000 Hz_10ms_envelope_correlation": 0.9793346191816436, "1000–3000 Hz_10ms_modulation_mae_db": 1.6089685060110288, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0728262582400518, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08981005424784647, "1000–3000 Hz_40ms_envelope_correlation": 0.9945533646509521, "1000–3000 Hz_40ms_modulation_mae_db": 1.768337755448572, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8221137561817657, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19607080509606697, "3000–6000 Hz_10ms_envelope_correlation": 0.9716415838761723, "3000–6000 Hz_10ms_modulation_mae_db": 1.4712432015239234, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0398338492942527, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10019536379147032, "3000–6000 Hz_40ms_envelope_correlation": 0.9920884095369397, "3000–6000 Hz_40ms_modulation_mae_db": 1.824343804166995, "mr_stft_db": 5.905424883613165, "band_mae_db": 0.2341533483987914, "rms_gain_error_db": 0.03486537255090445, "output_peak": 0.4266895077937465}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8977416209844684, "全帯域_10ms_rms_envelope_nrmse": 0.09512850115552988, "全帯域_10ms_envelope_correlation": 0.9917095969231025, "全帯域_10ms_modulation_mae_db": 0.42070005401344, "全帯域_40ms_log_rms_mae_db": 0.49959371997076074, "全帯域_40ms_rms_envelope_nrmse": 0.0501481400418959, "全帯域_40ms_envelope_correlation": 0.9975160065294871, "全帯域_40ms_modulation_mae_db": 0.338771737464306, "300–1000 Hz_10ms_log_rms_mae_db": 1.4518350316717783, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11845398302002555, "300–1000 Hz_10ms_envelope_correlation": 0.9887792846832558, "300–1000 Hz_10ms_modulation_mae_db": 0.5503629947928722, "300–1000 Hz_40ms_log_rms_mae_db": 0.8418360858265533, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06496831214657761, "300–1000 Hz_40ms_envelope_correlation": 0.9963653547531844, "300–1000 Hz_40ms_modulation_mae_db": 0.5841260575314966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.665485758020937, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18953391100221095, "1000–3000 Hz_10ms_envelope_correlation": 0.9703884938165409, "1000–3000 Hz_10ms_modulation_mae_db": 0.325443361245605, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0081442224969244, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0823318268420368, "1000–3000 Hz_40ms_envelope_correlation": 0.9938078818141017, "1000–3000 Hz_40ms_modulation_mae_db": 0.454233355340697, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8327765004351084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1498596149823689, "3000–6000 Hz_10ms_envelope_correlation": 0.9835344986993265, "3000–6000 Hz_10ms_modulation_mae_db": 1.419897961104323, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0127161394753283, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0804860110540122, "3000–6000 Hz_40ms_envelope_correlation": 0.9949430297334294, "3000–6000 Hz_40ms_modulation_mae_db": 1.5948393393047555, "mr_stft_db": 5.910402704976652, "band_mae_db": 0.11857445218910133, "rms_gain_error_db": 0.06643754907868843, "output_peak": 0.42729608489452425}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8631279964649072, "全帯域_10ms_rms_envelope_nrmse": 0.0977407805741356, "全帯域_10ms_envelope_correlation": 0.991108370887219, "全帯域_10ms_modulation_mae_db": 0.202710560146918, "全帯域_40ms_log_rms_mae_db": 0.4041812975057553, "全帯域_40ms_rms_envelope_nrmse": 0.04290983645444251, "全帯域_40ms_envelope_correlation": 0.9981405127120345, "全帯域_40ms_modulation_mae_db": 0.09734183115354177, "300–1000 Hz_10ms_log_rms_mae_db": 1.37770843523311, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1201061283526353, "300–1000 Hz_10ms_envelope_correlation": 0.9883711305410776, "300–1000 Hz_10ms_modulation_mae_db": 0.6825309713514883, "300–1000 Hz_40ms_log_rms_mae_db": 0.8124598843430157, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06138887036681465, "300–1000 Hz_40ms_envelope_correlation": 0.996754547904942, "300–1000 Hz_40ms_modulation_mae_db": 0.8115897869140096, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6275240654679148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16068642100145708, "1000–3000 Hz_10ms_envelope_correlation": 0.9787504065862562, "1000–3000 Hz_10ms_modulation_mae_db": 0.9870625001059486, "1000–3000 Hz_40ms_log_rms_mae_db": 0.949004412912161, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08582921459443021, "1000–3000 Hz_40ms_envelope_correlation": 0.9933726002417366, "1000–3000 Hz_40ms_modulation_mae_db": 0.7048819628793774, "3000–6000 Hz_10ms_log_rms_mae_db": 1.793917258032255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16627195790491608, "3000–6000 Hz_10ms_envelope_correlation": 0.9793685879801367, "3000–6000 Hz_10ms_modulation_mae_db": 1.0641410063991599, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0016472762967588, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07488761278056595, "3000–6000 Hz_40ms_envelope_correlation": 0.9958993536307785, "3000–6000 Hz_40ms_modulation_mae_db": 1.2686603960151321, "mr_stft_db": 5.914248399347194, "band_mae_db": 0.1512472194455947, "rms_gain_error_db": -0.032307100235206734, "output_peak": 0.45014599106718084}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9017363715223933, "全帯域_10ms_rms_envelope_nrmse": 0.09318744717348214, "全帯域_10ms_envelope_correlation": 0.9919581256278113, "全帯域_10ms_modulation_mae_db": 0.14259984499585301, "全帯域_40ms_log_rms_mae_db": 0.43937572183411927, "全帯域_40ms_rms_envelope_nrmse": 0.04762032283265878, "全帯域_40ms_envelope_correlation": 0.9977822302171396, "全帯域_40ms_modulation_mae_db": 0.37659299604163204, "300–1000 Hz_10ms_log_rms_mae_db": 1.4338892911387686, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11328032035607809, "300–1000 Hz_10ms_envelope_correlation": 0.9896913702765425, "300–1000 Hz_10ms_modulation_mae_db": 0.6466501001710417, "300–1000 Hz_40ms_log_rms_mae_db": 0.8389118609051384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0659715729713885, "300–1000 Hz_40ms_envelope_correlation": 0.9963667446306164, "300–1000 Hz_40ms_modulation_mae_db": 0.6322336341947844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.668059184457887, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16626086501623916, "1000–3000 Hz_10ms_envelope_correlation": 0.9772607352619496, "1000–3000 Hz_10ms_modulation_mae_db": 0.6412332132836261, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9612212299419257, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07711508970349934, "1000–3000 Hz_40ms_envelope_correlation": 0.9947389280339212, "1000–3000 Hz_40ms_modulation_mae_db": 0.2837658134805125, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8268339994413136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15300616409706863, "3000–6000 Hz_10ms_envelope_correlation": 0.9826017698990289, "3000–6000 Hz_10ms_modulation_mae_db": 1.190129533365684, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0135902033940456, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06727196062627937, "3000–6000 Hz_40ms_envelope_correlation": 0.9966956333634434, "3000–6000 Hz_40ms_modulation_mae_db": 1.5785202773540552, "mr_stft_db": 5.904843895869213, "band_mae_db": 0.15079521873867344, "rms_gain_error_db": -0.0853097038765244, "output_peak": 0.46747391727418014}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.92643985516369, "全帯域_10ms_rms_envelope_nrmse": 0.09558658752876853, "全帯域_10ms_envelope_correlation": 0.9916352947572877, "全帯域_10ms_modulation_mae_db": 0.3945691415214528, "全帯域_40ms_log_rms_mae_db": 0.5502295514583914, "全帯域_40ms_rms_envelope_nrmse": 0.05044591304364106, "全帯域_40ms_envelope_correlation": 0.9975133813321805, "全帯域_40ms_modulation_mae_db": 0.8025874109038682, "300–1000 Hz_10ms_log_rms_mae_db": 1.4335182892287133, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11590060205273038, "300–1000 Hz_10ms_envelope_correlation": 0.9891996329222583, "300–1000 Hz_10ms_modulation_mae_db": 1.064031705828589, "300–1000 Hz_40ms_log_rms_mae_db": 0.8429143748530585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0645937312957573, "300–1000 Hz_40ms_envelope_correlation": 0.996429054191651, "300–1000 Hz_40ms_modulation_mae_db": 1.2895573499294635, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7368760380530144, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17766385155240264, "1000–3000 Hz_10ms_envelope_correlation": 0.9743773855097472, "1000–3000 Hz_10ms_modulation_mae_db": 1.6241331348905401, "1000–3000 Hz_40ms_log_rms_mae_db": 1.142711577277162, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1064376933327015, "1000–3000 Hz_40ms_envelope_correlation": 0.9898124619769398, "1000–3000 Hz_40ms_modulation_mae_db": 1.1829868370468948, "3000–6000 Hz_10ms_log_rms_mae_db": 1.864597309084661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15276544901565944, "3000–6000 Hz_10ms_envelope_correlation": 0.982673341995963, "3000–6000 Hz_10ms_modulation_mae_db": 1.4008075504850082, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9872007545159964, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06578416863600436, "3000–6000 Hz_40ms_envelope_correlation": 0.9965244940875052, "3000–6000 Hz_40ms_modulation_mae_db": 1.7583401957685156, "mr_stft_db": 5.909939669285045, "band_mae_db": 0.11725621530332349, "rms_gain_error_db": 0.06513189034735906, "output_peak": 0.4103513950394331}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9349306974721187, "全帯域_10ms_rms_envelope_nrmse": 0.08943404335912224, "全帯域_10ms_envelope_correlation": 0.9927312639321655, "全帯域_10ms_modulation_mae_db": 1.0866691447021568, "全帯域_40ms_log_rms_mae_db": 0.46963177334084016, "全帯域_40ms_rms_envelope_nrmse": 0.04053593323389649, "全帯域_40ms_envelope_correlation": 0.9986107821163634, "全帯域_40ms_modulation_mae_db": 1.002868047895175, "300–1000 Hz_10ms_log_rms_mae_db": 1.445312248263109, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10645162610764508, "300–1000 Hz_10ms_envelope_correlation": 0.9910845954022862, "300–1000 Hz_10ms_modulation_mae_db": 0.9831163934213838, "300–1000 Hz_40ms_log_rms_mae_db": 0.8741974585936683, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06151552224741009, "300–1000 Hz_40ms_envelope_correlation": 0.9970819538563264, "300–1000 Hz_40ms_modulation_mae_db": 0.8733402636811253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7268802581476692, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17123671871829366, "1000–3000 Hz_10ms_envelope_correlation": 0.976244933862007, "1000–3000 Hz_10ms_modulation_mae_db": 1.1055254424997205, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0605417067737037, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08168347417571332, "1000–3000 Hz_40ms_envelope_correlation": 0.9949307650262278, "1000–3000 Hz_40ms_modulation_mae_db": 1.0053984637063034, "3000–6000 Hz_10ms_log_rms_mae_db": 1.860405268371603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18722991442487574, "3000–6000 Hz_10ms_envelope_correlation": 0.9737165889116894, "3000–6000 Hz_10ms_modulation_mae_db": 1.525342017887077, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0389635730712448, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08375606591126808, "3000–6000 Hz_40ms_envelope_correlation": 0.9948533341251378, "3000–6000 Hz_40ms_modulation_mae_db": 1.9215230339961937, "mr_stft_db": 5.898929333898271, "band_mae_db": 0.24214332075988101, "rms_gain_error_db": -0.15252860648298072, "output_peak": 0.39613719489475335}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37680, "output_samples": 37680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.930063486034504, "spectral_convergence": 0.3458636656638923, "spectral_convergence_status": "ok", "amplitude_floor": 1.2997261015401668e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.710119578533145, "spectral_convergence": 0.32309776466976203, "spectral_convergence_status": "ok", "amplitude_floor": 1.1046780177944747e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.056516733572002, "spectral_convergence": 0.3460010901086772, "spectral_convergence_status": "ok", "amplitude_floor": 9.696590055390529e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.751691690693756e-05, "output_power": 7.992741120881545e-05, "delta_db": 0.13299255592254225, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016017256752476519, "output_power": 0.001536682316756539, "delta_db": -0.18004043567739994, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006838056632308089, "output_power": 0.000617723639183511, "delta_db": -0.44138471940025914, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003361315194748372, "output_power": 0.0003525287574186011, "delta_db": 0.2068531148531753, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00035354415012418854, "output_power": 0.0003631384411529765, "delta_db": 0.11628569306159212, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00021020525507420865, "output_power": 0.00021152549561390772, "delta_db": 0.027191522236016015, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.18412467352516412, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.13710881612574946, "pitch": null, "output_rms_dbfs": -25.00103233826392, "output_peak": 0.45244511569105095, "reference_rms_dbfs": -24.863923522138165, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.898899932713216}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00018834951637393047, "log_rms_mae_db": 0.8612638835979949, "rms_envelope_nrmse": 0.08186736916678636, "envelope_correlation": 0.9939204294418261, "rms_gain_error_db": -0.13710410306135573, "reference_mean_power": 0.00326985830085679, "output_mean_power": 0.003168243281825394}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.479011996203747e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005558024994761568, "output_power": 0.0005682696994475785, "reference_active": true, "delta_db": 0.09634004272120354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011456565813557757, "output_power": 0.00010064485470667698, "reference_active": true, "delta_db": -0.5626287706227112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.821714301732997e-05, "output_power": 2.5215425165082745e-05, "reference_active": true, "delta_db": -1.805919222345367}], "active_band_count": 3, "active_band_mae_db": 0.8216293452297606}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00017471448177334785, "log_rms_mae_db": 0.47372762018871534, "rms_envelope_nrmse": 0.040957595947494, "envelope_correlation": 0.9985199102964882, "rms_gain_error_db": -0.13709183117365692, "reference_mean_power": 0.0033120735889998944, "output_mean_power": 0.0032091557441424846}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 7.763101583949748e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004881183543297566, "output_power": 0.0005011766558047023, "reference_active": true, "delta_db": 0.11465695491066845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.883974713754685e-05, "output_power": 5.2416087284282e-05, "reference_active": true, "delta_db": -0.5020619874853016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7824879244994614e-06, "output_power": 1.1381476204713307e-06, "reference_active": true, "delta_db": -1.9482800165064182}], "active_band_count": 3, "active_band_mae_db": 0.8549996529674627}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00017587844894922502, "log_rms_mae_db": 1.4230870098861284, "rms_envelope_nrmse": 0.11518081296448877, "envelope_correlation": 0.9899789603373476, "rms_gain_error_db": -0.30497564734505744, "reference_mean_power": 0.002090015296145697, "output_mean_power": 0.0019482823665619015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 6.927253047980857e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041075223634869497, "output_power": 0.00038438487164543653, "reference_active": true, "delta_db": -0.2881364996816214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.156045953139752e-05, "output_power": 7.410425957039717e-05, "reference_active": true, "delta_db": -0.4163649175241374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2217147797459886e-05, "output_power": 2.007036307961351e-05, "reference_active": true, "delta_db": -0.44133075048175796}], "active_band_count": 3, "active_band_mae_db": 0.3819440558958389}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00016316895420677254, "log_rms_mae_db": 0.8362201721919739, "rms_envelope_nrmse": 0.07677221739485231, "envelope_correlation": 0.9958110052459791, "rms_gain_error_db": -0.3049735859456042, "reference_mean_power": 0.002117036355483854, "output_mean_power": 0.0019734719483250787}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 5.84659163450443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037072755253225205, "output_power": 0.00035058268134042516, "reference_active": true, "delta_db": -0.2426440577089238}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.09979708848843e-05, "output_power": 3.657379910571555e-05, "reference_active": true, "delta_db": -0.49592287639658333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.223097583568823e-06, "output_power": 1.0612689833132156e-06, "reference_active": true, "delta_db": -0.6163563644373543}], "active_band_count": 3, "active_band_mae_db": 0.45164109951428716}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0001038394087983997, "log_rms_mae_db": 1.6085420418189638, "rms_envelope_nrmse": 0.18763272552370158, "envelope_correlation": 0.9718069724948951, "rms_gain_error_db": 0.13287172699082023, "reference_mean_power": 0.0005521302842545539, "output_mean_power": 0.0005692836900484796}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 2.3748882259523475e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000136307073642641, "output_power": 0.00016130378775176357, "reference_active": true, "delta_db": 0.7312617156501356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.293151120982887e-05, "output_power": 3.5692024152959536e-05, "reference_active": true, "delta_db": 0.3495951748729062}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.586128694071059e-05, "output_power": 1.171276156989108e-05, "reference_active": true, "delta_db": -1.3167911917061286}], "active_band_count": 3, "active_band_mae_db": 0.7992160274097234}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 8.776193320418374e-05, "log_rms_mae_db": 0.9576628491162452, "rms_envelope_nrmse": 0.09999866689690125, "envelope_correlation": 0.9912053771361082, "rms_gain_error_db": 0.1328978829797216, "reference_mean_power": 0.0005592400075632049, "output_mean_power": 0.0005766177687005888}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.9436382041482247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012615724913962246, "output_power": 0.00014936438264653528, "reference_active": true, "delta_db": 0.7333483755987852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.519569809811637e-05, "output_power": 1.65974930890929e-05, "reference_active": true, "delta_db": 0.38321839902191723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.080674445150989e-07, "output_power": 8.289073928477145e-07, "reference_active": true, "delta_db": -0.39612092859165704}], "active_band_count": 3, "active_band_mae_db": 0.5042292344041198}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 6.310171076800209e-05, "log_rms_mae_db": 1.8384190382465304, "rms_envelope_nrmse": 0.1815048768339431, "envelope_correlation": 0.9756232741149072, "rms_gain_error_db": 0.05857040396262965, "reference_mean_power": 0.00026303793349419837, "output_mean_power": 0.0002666093796211985}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 1.3356760090821468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.8507771208283054e-05, "output_power": 6.196786369366957e-05, "reference_active": true, "delta_db": 0.24952969919709}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1981263949052112e-05, "output_power": 1.1125176295362856e-05, "reference_active": true, "delta_db": -0.321957340798834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.256106518668102e-06, "output_power": 3.3265193915247792e-06, "reference_active": true, "delta_db": -4.444382843506341}], "active_band_count": 3, "active_band_mae_db": 1.6719566278340885}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 5.861628173421479e-05, "log_rms_mae_db": 1.0149234936794194, "rms_envelope_nrmse": 0.078140137990075, "envelope_correlation": 0.9951648628096672, "rms_gain_error_db": 0.058661554914562146, "reference_mean_power": 0.00026641634082957415, "output_mean_power": 0.0002700393255083089}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.157398705126346e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.177699204483558e-05, "output_power": 5.3361082971831763e-05, "reference_active": true, "delta_db": 0.1308781833628023}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.036512028004977e-06, "output_power": 6.080530311357569e-06, "reference_active": true, "delta_db": -0.634159766053316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.588547226531691e-07, "output_power": 1.7175223085182983e-07, "reference_active": true, "delta_db": -5.123965386047384}], "active_band_count": 3, "active_band_mae_db": 1.9630011118211677}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8699107205088563, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "38e958097c4ad9d1065e15fe9c390504d38c8780705efbbf96c64978eb7eefb7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.974682627098276, "standard_deviation": 0.003837360757563498, "minimum": 0.9692974700731696, "maximum": 0.9793346191816436, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.682515455312722, "standard_deviation": 0.04817830505727727, "minimum": 1.6085420418189638, "maximum": 1.7368760380530144, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9687051248082645, "standard_deviation": 0.463654905650225, "minimum": 0.325443361245605, "maximum": 1.6241331348905401, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17620522972881072, "standard_deviation": 0.012958317775902293, "minimum": 0.16068642100145708, "maximum": 0.1939941705453501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929951035150579, "standard_deviation": 0.0019075135868033206, "minimum": 0.9898124619769398, "maximum": 0.9949307650262278, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0250764203851532, "standard_deviation": 0.06826441291186384, "minimum": 0.949004412912161, "maximum": 1.142711577277162, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7755451842336003, "standard_deviation": 0.5146168439710813, "minimum": 0.2837658134805125, "maximum": 1.768337755448572, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0907853722805006, "standard_deviation": 0.011004293916420013, "minimum": 0.07711508970349934, "maximum": 0.1064376933327015, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9789569540511026, "standard_deviation": 0.004672622736397045, "minimum": 0.9716415838761723, "maximum": 0.9835344986993265, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.827854411997055, "standard_deviation": 0.028471560260540218, "minimum": 1.783772166183202, "maximum": 1.864597309084661, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.3818268650848444, "standard_deviation": 0.19170353480533428, "minimum": 1.0641410063991599, "maximum": 1.6719566278340885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16745442700476276, "standard_deviation": 0.018337833396145278, "minimum": 0.1498596149823689, "maximum": 0.19607080509606697, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9950136053176875, "standard_deviation": 0.0014957123786893723, "minimum": 0.9920884095369397, "maximum": 0.9966956333634434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0191976369012083, "standard_deviation": 0.02032151997410314, "minimum": 0.9872007545159964, "maximum": 1.04470580548262, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6678187352421086, "standard_deviation": 0.2425487208097745, "minimum": 1.2686603960151321, "maximum": 1.9630011118211677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07979005214668494, "standard_deviation": 0.01118901676763652, "minimum": 0.06578416863600436, "maximum": 0.10019536379147032, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9894084840414286, "standard_deviation": 0.0008733204281915875, "minimum": 0.9883711305410776, "maximum": 0.9910845954022862, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4209929930154792, "standard_deviation": 0.026525258151666856, "minimum": 1.37770843523311, "maximum": 1.4518350316717783, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6655285559421471, "standard_deviation": 0.26056585488729456, "minimum": 0.31870968082318, "maximum": 1.064031705828589, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11545440684795687, "standard_deviation": 0.004398265752523826, "minimum": 0.10645162610764508, "maximum": 0.1201061283526353, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965638950084783, "standard_deviation": 0.0004057780534524737, "minimum": 0.9958110052459791, "maximum": 0.9970819538563264, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8282933425631814, "standard_deviation": 0.029506189512532584, "minimum": 0.77950910684302, "maximum": 0.8741974585936683, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7501224601157057, "standard_deviation": 0.26552409132430616, "minimum": 0.45164109951428716, "maximum": 1.2895573499294635, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06482004323161603, "standard_deviation": 0.005238856422887354, "minimum": 0.06003326514782183, "maximum": 0.07677221739485231, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.18126760075104606, "standard_deviation": 0.05522674957917706, "minimum": 0.11725621530332349, "maximum": 0.2518463576478391, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.905590618021348, "standard_deviation": 0.005598278658834837, "minimum": 5.898899932713216, "maximum": 5.914248399347194, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42922493321474653, "standard_deviation": 0.025565530168492848, "minimum": 0.39613719489475335, "maximum": 0.46747391727418014, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.04579482835763638, "standard_deviation": 0.09199360705406932, "minimum": -0.15252860648298072, "maximum": 0.06643754907868843, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9922176495107317, "standard_deviation": 0.0011135892869575366, "minimum": 0.9909987732306356, "maximum": 0.9939204294418261, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8962900145865771, "standard_deviation": 0.02798126931186529, "minimum": 0.8612638835979949, "maximum": 0.9349306974721187, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5153695165995755, "standard_deviation": 0.3149957759897458, "minimum": 0.14259984499585301, "maximum": 1.0866691447021568, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09186647281128466, "standard_deviation": 0.0063999436362261524, "minimum": 0.08186736916678636, "maximum": 0.09869286265030075, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980304428986734, "standard_deviation": 0.0005192626968134864, "minimum": 0.9974886005299538, "maximum": 0.9986721194557397, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4642260336941336, "standard_deviation": 0.0458593542907444, "minimum": 0.4041812975057553, "maximum": 0.5502295514583914, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5538095639288607, "standard_deviation": 0.3065666797968133, "minimum": 0.09734183115354177, "maximum": 1.002868047895175, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.045160493133349064, "standard_deviation": 0.004883328111409088, "minimum": 0.038739368693500954, "maximum": 0.05044591304364106, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0022", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 1.7603106498718262, "predicted_mos": 1.7603106498718262, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "7e74435412121bef647bd21e03067213caf4c9b9bca0ceba2965cc196068074b", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.182432399597019, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8423467207074666, "全帯域_10ms_rms_envelope_nrmse": 0.08286831053951639, "全帯域_10ms_envelope_correlation": 0.9936173288462087, "全帯域_10ms_modulation_mae_db": 0.575583930214777, "全帯域_40ms_log_rms_mae_db": 0.412473041930808, "全帯域_40ms_rms_envelope_nrmse": 0.045930567606246545, "全帯域_40ms_envelope_correlation": 0.9978562004758454, "全帯域_40ms_modulation_mae_db": 0.28084370183224977, "300–1000 Hz_10ms_log_rms_mae_db": 1.2775413923598564, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09643927669366656, "300–1000 Hz_10ms_envelope_correlation": 0.9925594770865896, "300–1000 Hz_10ms_modulation_mae_db": 0.11756475625394537, "300–1000 Hz_40ms_log_rms_mae_db": 0.8583496382597617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06033225985470289, "300–1000 Hz_40ms_envelope_correlation": 0.9968665979437561, "300–1000 Hz_40ms_modulation_mae_db": 0.3569827570803421, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4447315075076477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16854734004337227, "1000–3000 Hz_10ms_envelope_correlation": 0.9767597711430406, "1000–3000 Hz_10ms_modulation_mae_db": 0.31048179590027264, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8780627321510837, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08142348279594083, "1000–3000 Hz_40ms_envelope_correlation": 0.9944197106398305, "1000–3000 Hz_40ms_modulation_mae_db": 0.2969183877979183, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3387397140341872, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14780739364361473, "3000–6000 Hz_10ms_envelope_correlation": 0.9842048575222735, "3000–6000 Hz_10ms_modulation_mae_db": 0.7673848133221984, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6901503531889774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06678927376498964, "3000–6000 Hz_40ms_envelope_correlation": 0.9974676951418837, "3000–6000 Hz_40ms_modulation_mae_db": 1.4226428645601414, "mr_stft_db": 5.607283232640086, "band_mae_db": 0.17359917409968006, "rms_gain_error_db": -0.03129983607749159, "output_peak": 0.41883054759158844}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9243546449257753, "全帯域_10ms_rms_envelope_nrmse": 0.08391536466602123, "全帯域_10ms_envelope_correlation": 0.9936557372158787, "全帯域_10ms_modulation_mae_db": 0.4244898582729322, "全帯域_40ms_log_rms_mae_db": 0.5075970098663793, "全帯域_40ms_rms_envelope_nrmse": 0.04014594540682914, "全帯域_40ms_envelope_correlation": 0.9986604387754633, "全帯域_40ms_modulation_mae_db": 0.2905457798469707, "300–1000 Hz_10ms_log_rms_mae_db": 1.2818223762019139, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10602578590251405, "300–1000 Hz_10ms_envelope_correlation": 0.9912059393397592, "300–1000 Hz_10ms_modulation_mae_db": 0.4548120422701319, "300–1000 Hz_40ms_log_rms_mae_db": 0.6634498202258621, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055995545648026716, "300–1000 Hz_40ms_envelope_correlation": 0.9977205620591053, "300–1000 Hz_40ms_modulation_mae_db": 0.37178966166406696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5062804015738596, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1562700560535518, "1000–3000 Hz_10ms_envelope_correlation": 0.9802129216676807, "1000–3000 Hz_10ms_modulation_mae_db": 0.43943110548978964, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9503226287319907, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08821426941508816, "1000–3000 Hz_40ms_envelope_correlation": 0.9935490360567713, "1000–3000 Hz_40ms_modulation_mae_db": 0.4361572746701438, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3479302249296659, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14209034080413493, "3000–6000 Hz_10ms_envelope_correlation": 0.9850573153751145, "3000–6000 Hz_10ms_modulation_mae_db": 0.5887335392482993, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8168345726779577, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07215947245888722, "3000–6000 Hz_40ms_envelope_correlation": 0.996217203437983, "3000–6000 Hz_40ms_modulation_mae_db": 0.602794730661126, "mr_stft_db": 5.594773716375801, "band_mae_db": 0.25411876320564086, "rms_gain_error_db": -0.15308014730829783, "output_peak": 0.4109293153023039}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9006156605657033, "全帯域_10ms_rms_envelope_nrmse": 0.08978858851184861, "全帯域_10ms_envelope_correlation": 0.9925181045189313, "全帯域_10ms_modulation_mae_db": 0.35082820050240165, "全帯域_40ms_log_rms_mae_db": 0.4505403334463459, "全帯域_40ms_rms_envelope_nrmse": 0.04496405155289624, "全帯域_40ms_envelope_correlation": 0.9980093898570663, "全帯域_40ms_modulation_mae_db": 0.607110181722008, "300–1000 Hz_10ms_log_rms_mae_db": 1.3072989387154819, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11750051885366415, "300–1000 Hz_10ms_envelope_correlation": 0.9889230595341745, "300–1000 Hz_10ms_modulation_mae_db": 1.1475529818579122, "300–1000 Hz_40ms_log_rms_mae_db": 0.8168765169257528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06456203791848639, "300–1000 Hz_40ms_envelope_correlation": 0.996513179939787, "300–1000 Hz_40ms_modulation_mae_db": 1.1589488253030569, "1000–3000 Hz_10ms_log_rms_mae_db": 1.419443387923771, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15170338050787605, "1000–3000 Hz_10ms_envelope_correlation": 0.9811217700477244, "1000–3000 Hz_10ms_modulation_mae_db": 0.7898120814373454, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8654259266478203, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07885443853352546, "1000–3000 Hz_40ms_envelope_correlation": 0.9944019881727733, "1000–3000 Hz_40ms_modulation_mae_db": 0.5539447658422639, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3102987936204826, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15400192366566054, "3000–6000 Hz_10ms_envelope_correlation": 0.9825020289311219, "3000–6000 Hz_10ms_modulation_mae_db": 0.8357403964152382, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7358631964076884, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06807160105493722, "3000–6000 Hz_40ms_envelope_correlation": 0.996948658648048, "3000–6000 Hz_40ms_modulation_mae_db": 1.0186404980592763, "mr_stft_db": 5.59458682563001, "band_mae_db": 0.16059499546864595, "rms_gain_error_db": -0.07666097806678691, "output_peak": 0.4187202614400675}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9025380741811906, "全帯域_10ms_rms_envelope_nrmse": 0.10514930464357004, "全帯域_10ms_envelope_correlation": 0.9897976072026187, "全帯域_10ms_modulation_mae_db": 0.7350057564318763, "全帯域_40ms_log_rms_mae_db": 0.47707742639762096, "全帯域_40ms_rms_envelope_nrmse": 0.059242468607673585, "全帯域_40ms_envelope_correlation": 0.9964636985586831, "全帯域_40ms_modulation_mae_db": 0.5801855629781144, "300–1000 Hz_10ms_log_rms_mae_db": 1.4262399847882408, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1435033672338665, "300–1000 Hz_10ms_envelope_correlation": 0.9835875957604019, "300–1000 Hz_10ms_modulation_mae_db": 0.8678974827214841, "300–1000 Hz_40ms_log_rms_mae_db": 0.8462692048313585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09115341619188376, "300–1000 Hz_40ms_envelope_correlation": 0.9928661984828694, "300–1000 Hz_40ms_modulation_mae_db": 1.0885654037805692, "1000–3000 Hz_10ms_log_rms_mae_db": 1.410089117837158, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1577980627916858, "1000–3000 Hz_10ms_envelope_correlation": 0.979512405295551, "1000–3000 Hz_10ms_modulation_mae_db": 0.7186626715588175, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8855492110763733, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08731953389393171, "1000–3000 Hz_40ms_envelope_correlation": 0.9930450121380551, "1000–3000 Hz_40ms_modulation_mae_db": 0.6366266130294989, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2949141614030146, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13551194880851913, "3000–6000 Hz_10ms_envelope_correlation": 0.9863179466529118, "3000–6000 Hz_10ms_modulation_mae_db": 0.8449702816609563, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7451419981352612, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06323066131253517, "3000–6000 Hz_40ms_envelope_correlation": 0.9970286248850081, "3000–6000 Hz_40ms_modulation_mae_db": 1.0684434571447694, "mr_stft_db": 5.596091636218453, "band_mae_db": 0.23389467817962195, "rms_gain_error_db": 0.03877851481496736, "output_peak": 0.4139734815708307}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9396989604204551, "全帯域_10ms_rms_envelope_nrmse": 0.0852274035885422, "全帯域_10ms_envelope_correlation": 0.9933062718949971, "全帯域_10ms_modulation_mae_db": 0.759709772638613, "全帯域_40ms_log_rms_mae_db": 0.48298744671963517, "全帯域_40ms_rms_envelope_nrmse": 0.04116645538207454, "全帯域_40ms_envelope_correlation": 0.9983885448500064, "全帯域_40ms_modulation_mae_db": 0.5608974138656094, "300–1000 Hz_10ms_log_rms_mae_db": 1.393473210625295, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10501179334980809, "300–1000 Hz_10ms_envelope_correlation": 0.991131270009549, "300–1000 Hz_10ms_modulation_mae_db": 0.3692864822686784, "300–1000 Hz_40ms_log_rms_mae_db": 0.7915832208443531, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05855937577644095, "300–1000 Hz_40ms_envelope_correlation": 0.9970506633338037, "300–1000 Hz_40ms_modulation_mae_db": 0.4515672460567764, "1000–3000 Hz_10ms_log_rms_mae_db": 1.519159273813593, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17969541363692773, "1000–3000 Hz_10ms_envelope_correlation": 0.9753081205080321, "1000–3000 Hz_10ms_modulation_mae_db": 0.8338818942411722, "1000–3000 Hz_40ms_log_rms_mae_db": 0.962125377328227, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11101581871631948, "1000–3000 Hz_40ms_envelope_correlation": 0.9919063169654421, "1000–3000 Hz_40ms_modulation_mae_db": 0.7284206797891359, "3000–6000 Hz_10ms_log_rms_mae_db": 1.305818908303078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.12309221652100776, "3000–6000 Hz_10ms_envelope_correlation": 0.9887039045625557, "3000–6000 Hz_10ms_modulation_mae_db": 1.0021445531358664, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7222501717330165, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0619731215072169, "3000–6000 Hz_40ms_envelope_correlation": 0.9970462099333904, "3000–6000 Hz_40ms_modulation_mae_db": 1.3448041498661067, "mr_stft_db": 5.5956740148135795, "band_mae_db": 0.23000859606661275, "rms_gain_error_db": -0.0994211890987976, "output_peak": 0.4033081443949301}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9328499273333974, "全帯域_10ms_rms_envelope_nrmse": 0.10554143831575975, "全帯域_10ms_envelope_correlation": 0.9898503622966555, "全帯域_10ms_modulation_mae_db": 0.2934636042952836, "全帯域_40ms_log_rms_mae_db": 0.48436033657927635, "全帯域_40ms_rms_envelope_nrmse": 0.06233514726915551, "全帯域_40ms_envelope_correlation": 0.9964245146687674, "全帯域_40ms_modulation_mae_db": 0.5255509569355389, "300–1000 Hz_10ms_log_rms_mae_db": 1.3798841457861108, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12354384540826036, "300–1000 Hz_10ms_envelope_correlation": 0.9877534413808478, "300–1000 Hz_10ms_modulation_mae_db": 0.7804196667339229, "300–1000 Hz_40ms_log_rms_mae_db": 0.8181556313573604, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07094884156973445, "300–1000 Hz_40ms_envelope_correlation": 0.9957718147662296, "300–1000 Hz_40ms_modulation_mae_db": 0.8485610849115949, "1000–3000 Hz_10ms_log_rms_mae_db": 1.471352244785141, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1891952258355239, "1000–3000 Hz_10ms_envelope_correlation": 0.9727192374001565, "1000–3000 Hz_10ms_modulation_mae_db": 0.7435727363145183, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9399265870031618, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1253494827188124, "1000–3000 Hz_40ms_envelope_correlation": 0.9892374425851302, "1000–3000 Hz_40ms_modulation_mae_db": 0.6147545651062346, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2810136893972683, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14466386815091942, "3000–6000 Hz_10ms_envelope_correlation": 0.9844596748272799, "3000–6000 Hz_10ms_modulation_mae_db": 1.0848726039261007, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6988966741305739, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06577422503393195, "3000–6000 Hz_40ms_envelope_correlation": 0.9968939997628548, "3000–6000 Hz_40ms_modulation_mae_db": 1.3827053756701069, "mr_stft_db": 5.606770149070116, "band_mae_db": 0.34194298444930604, "rms_gain_error_db": -0.1869311636103617, "output_peak": 0.44189232627254926}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8998261227413468, "全帯域_10ms_rms_envelope_nrmse": 0.08890368569196946, "全帯域_10ms_envelope_correlation": 0.9926706012343126, "全帯域_10ms_modulation_mae_db": 0.5259626874711331, "全帯域_40ms_log_rms_mae_db": 0.4184893681401267, "全帯域_40ms_rms_envelope_nrmse": 0.04128778768564209, "全帯域_40ms_envelope_correlation": 0.9982713969119288, "全帯域_40ms_modulation_mae_db": 0.5844383431983914, "300–1000 Hz_10ms_log_rms_mae_db": 1.4225595200299466, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1191999647012663, "300–1000 Hz_10ms_envelope_correlation": 0.9888567034066791, "300–1000 Hz_10ms_modulation_mae_db": 0.5960768757034385, "300–1000 Hz_40ms_log_rms_mae_db": 0.803633593923453, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06857441150132679, "300–1000 Hz_40ms_envelope_correlation": 0.9961120266865799, "300–1000 Hz_40ms_modulation_mae_db": 0.5623468600956715, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5297762092171747, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17306999150976066, "1000–3000 Hz_10ms_envelope_correlation": 0.9761597570957933, "1000–3000 Hz_10ms_modulation_mae_db": 1.1025569267448911, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9213895046654714, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09265249793436398, "1000–3000 Hz_40ms_envelope_correlation": 0.9938662868983619, "1000–3000 Hz_40ms_modulation_mae_db": 1.2161739061731913, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3867281956073383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13626854552402873, "3000–6000 Hz_10ms_envelope_correlation": 0.9861949241294183, "3000–6000 Hz_10ms_modulation_mae_db": 0.8450441829490343, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8342816129119799, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07800891130782155, "3000–6000 Hz_40ms_envelope_correlation": 0.995295197429525, "3000–6000 Hz_40ms_modulation_mae_db": 1.071270916517565, "mr_stft_db": 5.608069273300828, "band_mae_db": 0.2163562807057667, "rms_gain_error_db": 0.008119312981463683, "output_peak": 0.4473917742522982}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9500571866132438, "全帯域_10ms_rms_envelope_nrmse": 0.09161504557835205, "全帯域_10ms_envelope_correlation": 0.992186229509039, "全帯域_10ms_modulation_mae_db": 0.5516358527897384, "全帯域_40ms_log_rms_mae_db": 0.4816417648236691, "全帯域_40ms_rms_envelope_nrmse": 0.04891387528348291, "全帯域_40ms_envelope_correlation": 0.9975630015198945, "全帯域_40ms_modulation_mae_db": 0.5219673412706131, "300–1000 Hz_10ms_log_rms_mae_db": 1.4251770544556115, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1079255226747035, "300–1000 Hz_10ms_envelope_correlation": 0.990609180983148, "300–1000 Hz_10ms_modulation_mae_db": 0.12418347518068013, "300–1000 Hz_40ms_log_rms_mae_db": 0.8372143375094299, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056823261116673664, "300–1000 Hz_40ms_envelope_correlation": 0.9972114472747854, "300–1000 Hz_40ms_modulation_mae_db": 0.13636075096323233, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5155733298373588, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15526502165979258, "1000–3000 Hz_10ms_envelope_correlation": 0.9804104845576271, "1000–3000 Hz_10ms_modulation_mae_db": 0.6818836296401442, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9642565762154858, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09142039245190968, "1000–3000 Hz_40ms_envelope_correlation": 0.9928817716169053, "1000–3000 Hz_40ms_modulation_mae_db": 0.6129032906827119, "3000–6000 Hz_10ms_log_rms_mae_db": 1.323102952224887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1593039599427185, "3000–6000 Hz_10ms_envelope_correlation": 0.9816900756470568, "3000–6000 Hz_10ms_modulation_mae_db": 0.6876377689463581, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7311793262002882, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09670378299020926, "3000–6000 Hz_40ms_envelope_correlation": 0.992967398826452, "3000–6000 Hz_40ms_modulation_mae_db": 0.8648679415581392, "mr_stft_db": 5.608868829494287, "band_mae_db": 0.16004748407691088, "rms_gain_error_db": -0.02677098835006442, "output_peak": 0.41188229149887123}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37680, "output_samples": 37680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.575847183615388, "spectral_convergence": 0.33239537330191987, "spectral_convergence_status": "ok", "amplitude_floor": 1.2997261015401668e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.337260368817061, "spectral_convergence": 0.31841658089590025, "spectral_convergence_status": "ok", "amplitude_floor": 1.1046780177944747e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.908742145487809, "spectral_convergence": 0.3573500967725501, "spectral_convergence_status": "ok", "amplitude_floor": 9.696590055390529e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.751691690693756e-05, "output_power": 8.132293404594312e-05, "delta_db": 0.20816548016402425, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016017256752476519, "output_power": 0.0016192811462613958, "delta_db": 0.04734122035225967, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006838056632308089, "output_power": 0.0006864351756657728, "delta_db": 0.016668371136159182, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003361315194748372, "output_power": 0.0003119063276768103, "delta_db": -0.32485053293307026, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00035354415012418854, "output_power": 0.00034749889769146826, "delta_db": -0.0749022434428383, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00021020525507420865, "output_power": 0.0001930531309400388, "delta_db": -0.3696671965697286, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.17359917409968006, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.03129983607749159, "pitch": null, "output_rms_dbfs": -24.89522335821566, "output_peak": 0.41883054759158844, "reference_rms_dbfs": -24.863923522138165, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.607283232640086}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00018834951637393047, "log_rms_mae_db": 0.8423467207074666, "rms_envelope_nrmse": 0.08286831053951639, "envelope_correlation": 0.9936173288462087, "rms_gain_error_db": -0.03129531360689054, "reference_mean_power": 0.00326985830085679, "output_mean_power": 0.003246380354633285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.479011996203747e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005558024994761568, "output_power": 0.0005796454477996772, "reference_active": true, "delta_db": 0.1824193402292604}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011456565813557757, "output_power": 0.00010592023730599819, "reference_active": true, "delta_db": -0.3407550920417623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.821714301732997e-05, "output_power": 2.8966797395374445e-05, "reference_active": true, "delta_db": -1.2035773583733085}], "active_band_count": 3, "active_band_mae_db": 0.575583930214777}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00017471448177334785, "log_rms_mae_db": 0.412473041930808, "rms_envelope_nrmse": 0.045930567606246545, "envelope_correlation": 0.9978562004758454, "rms_gain_error_db": -0.03128795704097169, "reference_mean_power": 0.0033120735889998944, "output_mean_power": 0.0032882981023955623}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 7.763101583949748e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004881183543297566, "output_power": 0.0005111061198494715, "reference_active": true, "delta_db": 0.199859427169447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.883974713754685e-05, "output_power": 5.4984647744343935e-05, "reference_active": true, "delta_db": -0.29429350569880025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7824879244994614e-06, "output_power": 1.645086654120971e-06, "reference_active": true, "delta_db": -0.34837817262850207}], "active_band_count": 3, "active_band_mae_db": 0.28084370183224977}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00017587844894922502, "log_rms_mae_db": 1.2775413923598564, "rms_envelope_nrmse": 0.09643927669366656, "envelope_correlation": 0.9925594770865896, "rms_gain_error_db": 0.028148354858211086, "reference_mean_power": 0.002090015296145697, "output_mean_power": 0.002103605511808964}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 6.927253047980857e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041075223634869497, "output_power": 0.0004355900692702104, "reference_active": true, "delta_db": 0.25498033365596684}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.156045953139752e-05, "output_power": 8.322020487345454e-05, "reference_active": true, "delta_db": 0.08749116581227703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2217147797459886e-05, "output_power": 2.216491280763663e-05, "reference_active": true, "delta_db": -0.010222769293592306}], "active_band_count": 3, "active_band_mae_db": 0.11756475625394537}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00016316895420677254, "log_rms_mae_db": 0.8583496382597617, "rms_envelope_nrmse": 0.06033225985470289, "envelope_correlation": 0.9968665979437561, "rms_gain_error_db": 0.02814889861536976, "reference_mean_power": 0.002117036355483854, "output_mean_power": 0.0021308025409653352}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 5.84659163450443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037072755253225205, "output_power": 0.00039630513238721426, "reference_active": true, "delta_db": 0.28974832676640927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.09979708848843e-05, "output_power": 4.221650193636846e-05, "reference_active": true, "delta_db": 0.12719882089726448}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.223097583568823e-06, "output_power": 1.4218743523160898e-06, "reference_active": true, "delta_db": 0.6540011235773525}], "active_band_count": 3, "active_band_mae_db": 0.3569827570803421}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0001038394087983997, "log_rms_mae_db": 1.4447315075076477, "rms_envelope_nrmse": 0.16854734004337227, "envelope_correlation": 0.9767597711430406, "rms_gain_error_db": -0.23095717019254242, "reference_mean_power": 0.0005521302842545539, "output_mean_power": 0.0005235351576592502}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 2.3748882259523475e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000136307073642641, "output_power": 0.0001337443521013032, "reference_active": true, "delta_db": -0.08242942859009325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.293151120982887e-05, "output_power": 3.075262174746526e-05, "reference_active": true, "delta_db": -0.2972951420050744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.586128694071059e-05, "output_power": 1.8009866914259278e-05, "reference_active": true, "delta_db": 0.5517208171056502}], "active_band_count": 3, "active_band_mae_db": 0.31048179590027264}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 8.776193320418374e-05, "log_rms_mae_db": 0.8780627321510837, "rms_envelope_nrmse": 0.08142348279594083, "envelope_correlation": 0.9944197106398305, "rms_gain_error_db": -0.23093027742864117, "reference_mean_power": 0.0005592400075632049, "output_mean_power": 0.0005302799481811656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.9436382041482247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012615724913962246, "output_power": 0.000125612372378978, "reference_active": true, "delta_db": -0.018797926333221696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.519569809811637e-05, "output_power": 1.427716405476079e-05, "reference_active": true, "delta_db": -0.2707870668483241}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.080674445150989e-07, "output_power": 1.042881882944772e-06, "reference_active": true, "delta_db": 0.6011701702122092}], "active_band_count": 3, "active_band_mae_db": 0.2969183877979183}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 6.310171076800209e-05, "log_rms_mae_db": 1.3387397140341872, "rms_envelope_nrmse": 0.14780739364361473, "envelope_correlation": 0.9842048575222735, "rms_gain_error_db": -0.3333618696527568, "reference_mean_power": 0.00026303793349419837, "output_mean_power": 0.0002436027557639886}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 1.3356760090821468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.8507771208283054e-05, "output_power": 5.615311041173918e-05, "reference_active": true, "delta_db": -0.17839736897913194}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1981263949052112e-05, "output_power": 1.0778026560583334e-05, "reference_active": true, "delta_db": -0.45963386339907075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.256106518668102e-06, "output_power": 6.309806237193575e-06, "reference_active": true, "delta_db": -1.6641232075883925}], "active_band_count": 3, "active_band_mae_db": 0.7673848133221984}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 5.861628173421479e-05, "log_rms_mae_db": 0.6901503531889774, "rms_envelope_nrmse": 0.06678927376498964, "envelope_correlation": 0.9974676951418837, "rms_gain_error_db": -0.33336946687229885, "reference_mean_power": 0.00026641634082957415, "output_mean_power": 0.00024673110989691295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.157398705126346e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.177699204483558e-05, "output_power": 4.962607002458167e-05, "reference_active": true, "delta_db": -0.18426932720337477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.036512028004977e-06, "output_power": 5.845491431180913e-06, "reference_active": true, "delta_db": -0.8053640615057529}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.588547226531691e-07, "output_power": 2.627056449129625e-07, "reference_active": true, "delta_db": -3.278295204971297}], "active_band_count": 3, "active_band_mae_db": 1.4226428645601414}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8842320099548557, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "70edcbec27c4a624349fd98cb8d319cb079690cadb973dbdefea414e3b9b6e9f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9777755584644507, "standard_deviation": 0.002985551265081127, "minimum": 0.9727192374001565, "maximum": 0.9811217700477244, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4770506840619628, "standard_deviation": 0.047465684236236524, "minimum": 1.410089117837158, "maximum": 1.5297762092171747, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7025353551658688, "standard_deviation": 0.24207049985120332, "minimum": 0.31048179590027264, "maximum": 1.1025569267448911, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16644306150481136, "standard_deviation": 0.013427385776588115, "minimum": 0.15170338050787605, "maximum": 0.1891952258355239, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929134456341587, "standard_deviation": 0.0017047033414642429, "minimum": 0.9892374425851302, "maximum": 0.9944197106398305, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9208823179774517, "standard_deviation": 0.039585975607575874, "minimum": 0.8654259266478203, "maximum": 0.9642565762154858, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6369874353863874, "standard_deviation": 0.2693291343877323, "minimum": 0.2969183877979183, "maximum": 1.2161739061731913, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09453123955748646, "standard_deviation": 0.015785167945247043, "minimum": 0.07885443853352546, "maximum": 0.1253494827188124, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9848913409559665, "standard_deviation": 0.0022332624549316004, "minimum": 0.9816900756470568, "maximum": 0.9887039045625557, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3235683299399903, "standard_deviation": 0.033670969690313544, "minimum": 1.2810136893972683, "maximum": 1.3867281956073383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8320660174505066, "standard_deviation": 0.15921243868452106, "minimum": 0.5887335392482993, "maximum": 1.0848726039261007, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14284252463257546, "standard_deviation": 0.011408475859162838, "minimum": 0.12309221652100776, "maximum": 0.1593039599427185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9962331235081432, "standard_deviation": 0.0014789843778284968, "minimum": 0.992967398826452, "maximum": 0.9974676951418837, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7468247381732179, "standard_deviation": 0.05214107559045884, "minimum": 0.6901503531889774, "maximum": 0.8342816129119799, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0970212417546539, "standard_deviation": 0.28153141012955496, "minimum": 0.602794730661126, "maximum": 1.4226428645601414, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07158888117881612, "standard_deviation": 0.011363104735044453, "minimum": 0.0619731215072169, "maximum": 0.09670378299020926, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9893283334376436, "standard_deviation": 0.0027948902307431463, "minimum": 0.9835875957604019, "maximum": 0.9925594770865896, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3642495778703072, "standard_deviation": 0.06503779784799474, "minimum": 1.2775413923598564, "maximum": 1.4262399847882408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5572242203737742, "standard_deviation": 0.3631145505387589, "minimum": 0.11756475625394537, "maximum": 1.1475529818579122, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11489375935221868, "standard_deviation": 0.014560664767454072, "minimum": 0.09643927669366656, "maximum": 0.1435033672338665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962640613108645, "standard_deviation": 0.001505878045126698, "minimum": 0.9928661984828694, "maximum": 0.9977205620591053, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8044414954846664, "standard_deviation": 0.06108197303150319, "minimum": 0.6634498202258621, "maximum": 0.8583496382597617, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6218903237319138, "standard_deviation": 0.37006134613862884, "minimum": 0.13636075096323233, "maximum": 1.1589488253030569, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06586864369715945, "standard_deviation": 0.011572939392691751, "minimum": 0.055995545648026716, "maximum": 0.09115341619188376, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.22132036953152315, "standard_deviation": 0.0604779069387366, "minimum": 0.16004748407691088, "maximum": 0.34194298444930604, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6015147096928946, "standard_deviation": 0.006707148566544511, "minimum": 5.59458682563001, "maximum": 5.608868829494287, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4208660177904299, "standard_deviation": 0.015536593585261594, "minimum": 0.4033081443949301, "maximum": 0.4473917742522982, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06590830933942113, "standard_deviation": 0.07805958339540138, "minimum": -0.1869311636103617, "maximum": 0.03877851481496736, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9922002803398302, "standard_deviation": 0.0015568724555493113, "minimum": 0.9897976072026187, "maximum": 0.9936557372158787, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9115359121860723, "standard_deviation": 0.03384867456914503, "minimum": 0.8423467207074666, "maximum": 0.9500571866132438, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5270849578270944, "standard_deviation": 0.16753364480716054, "minimum": 0.2934636042952836, "maximum": 0.759709772638613, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09162614269194747, "standard_deviation": 0.008979336962350763, "minimum": 0.08286831053951639, "maximum": 0.10554143831575975, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997704648202207, "standard_deviation": 0.0008465941493865193, "minimum": 0.9964245146687674, "maximum": 0.9986604387754633, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4643958409879827, "standard_deviation": 0.03394148489533762, "minimum": 0.412473041930808, "maximum": 0.5075970098663793, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4939424102061869, "standard_deviation": 0.13173732416804357, "minimum": 0.28084370183224977, "maximum": 0.607110181722008, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.047998287349250074, "standard_deviation": 0.008448681430070907, "minimum": 0.04014594540682914, "maximum": 0.06233514726915551, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0022", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 1.8468389511108398, "predicted_mos": 1.8468389511108398, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "d48c43b1e19d34d802039930cb90ec770785f4bec52633c7851fe50843622c80", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.280651899985969, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8750656469663882, "全帯域_10ms_rms_envelope_nrmse": 0.08904470475777665, "全帯域_10ms_envelope_correlation": 0.9928150327400733, "全帯域_10ms_modulation_mae_db": 0.2411373115910199, "全帯域_40ms_log_rms_mae_db": 0.459412623834744, "全帯域_40ms_rms_envelope_nrmse": 0.04753659100904025, "全帯域_40ms_envelope_correlation": 0.9979941580804033, "全帯域_40ms_modulation_mae_db": 0.45158629411018997, "300–1000 Hz_10ms_log_rms_mae_db": 1.4520588265819026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11083080864779098, "300–1000 Hz_10ms_envelope_correlation": 0.9902975826314666, "300–1000 Hz_10ms_modulation_mae_db": 0.5268602842445943, "300–1000 Hz_40ms_log_rms_mae_db": 0.8385406225247343, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.062086027914702156, "300–1000 Hz_40ms_envelope_correlation": 0.9970422980305513, "300–1000 Hz_40ms_modulation_mae_db": 0.17924891839140336, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5009171290963848, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18057461728174104, "1000–3000 Hz_10ms_envelope_correlation": 0.9732271325766214, "1000–3000 Hz_10ms_modulation_mae_db": 0.4960774867864905, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8767495699952426, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10387860746169038, "1000–3000 Hz_40ms_envelope_correlation": 0.9902263439494391, "1000–3000 Hz_40ms_modulation_mae_db": 0.4622068173747765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4840375048742742, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15283238543776875, "3000–6000 Hz_10ms_envelope_correlation": 0.9832786792512418, "3000–6000 Hz_10ms_modulation_mae_db": 1.0255173829096227, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7632782527462122, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0736015262440006, "3000–6000 Hz_40ms_envelope_correlation": 0.9969840939896161, "3000–6000 Hz_40ms_modulation_mae_db": 1.4469488918118945, "mr_stft_db": 5.675415141281245, "band_mae_db": 0.2096918514307783, "rms_gain_error_db": -0.14530873858433813, "output_peak": 0.4406077154287219}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8709631913612631, "全帯域_10ms_rms_envelope_nrmse": 0.08095953767987003, "全帯域_10ms_envelope_correlation": 0.9939540712334312, "全帯域_10ms_modulation_mae_db": 0.38314861765047475, "全帯域_40ms_log_rms_mae_db": 0.4493883297949708, "全帯域_40ms_rms_envelope_nrmse": 0.03686165719607963, "全帯域_40ms_envelope_correlation": 0.9987027501848678, "全帯域_40ms_modulation_mae_db": 0.3680852100574274, "300–1000 Hz_10ms_log_rms_mae_db": 1.3813368237504453, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09833723007151794, "300–1000 Hz_10ms_envelope_correlation": 0.9922227506137937, "300–1000 Hz_10ms_modulation_mae_db": 0.1240313711378373, "300–1000 Hz_40ms_log_rms_mae_db": 0.7808868606700556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.046884203141367396, "300–1000 Hz_40ms_envelope_correlation": 0.998162058153699, "300–1000 Hz_40ms_modulation_mae_db": 0.11403990055655461, "1000–3000 Hz_10ms_log_rms_mae_db": 1.54124962331108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18321631496917637, "1000–3000 Hz_10ms_envelope_correlation": 0.9724453677557605, "1000–3000 Hz_10ms_modulation_mae_db": 0.9318147951372217, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9780087608699535, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11147950346685138, "1000–3000 Hz_40ms_envelope_correlation": 0.9887897816605414, "1000–3000 Hz_40ms_modulation_mae_db": 0.46703337744943035, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4802047625336479, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1354539780431956, "3000–6000 Hz_10ms_envelope_correlation": 0.9868941565607495, "3000–6000 Hz_10ms_modulation_mae_db": 1.285258012763818, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8375447834420163, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07415443202649707, "3000–6000 Hz_40ms_envelope_correlation": 0.9965734280703711, "3000–6000 Hz_40ms_modulation_mae_db": 1.5574499168630094, "mr_stft_db": 5.683551571825128, "band_mae_db": 0.19463190216618156, "rms_gain_error_db": -0.08204070557020954, "output_peak": 0.4108322590231054}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8496957633812482, "全帯域_10ms_rms_envelope_nrmse": 0.0963145810244772, "全帯域_10ms_envelope_correlation": 0.9914385470183961, "全帯域_10ms_modulation_mae_db": 0.6551041602337142, "全帯域_40ms_log_rms_mae_db": 0.428554708646161, "全帯域_40ms_rms_envelope_nrmse": 0.052878290260644036, "全帯域_40ms_envelope_correlation": 0.9973048226581094, "全帯域_40ms_modulation_mae_db": 0.8435106173406574, "300–1000 Hz_10ms_log_rms_mae_db": 1.4931274644185137, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12116337847565468, "300–1000 Hz_10ms_envelope_correlation": 0.9881500784812002, "300–1000 Hz_10ms_modulation_mae_db": 0.29036653313128874, "300–1000 Hz_40ms_log_rms_mae_db": 0.9026684659209556, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07847603424249959, "300–1000 Hz_40ms_envelope_correlation": 0.9947075924636078, "300–1000 Hz_40ms_modulation_mae_db": 0.42355258412378677, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4633095104056115, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16302881834766442, "1000–3000 Hz_10ms_envelope_correlation": 0.9788708333325595, "1000–3000 Hz_10ms_modulation_mae_db": 0.9073250681557686, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9190289687910191, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09242722899184783, "1000–3000 Hz_40ms_envelope_correlation": 0.9936008524475182, "1000–3000 Hz_40ms_modulation_mae_db": 0.7681190149075618, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4008941620311566, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15226265586139662, "3000–6000 Hz_10ms_envelope_correlation": 0.9831996347394554, "3000–6000 Hz_10ms_modulation_mae_db": 0.9327020152946481, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7262276345235723, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07205645038411644, "3000–6000 Hz_40ms_envelope_correlation": 0.9969284054498159, "3000–6000 Hz_40ms_modulation_mae_db": 1.2718088146896995, "mr_stft_db": 5.674477764580774, "band_mae_db": 0.26339357115443024, "rms_gain_error_db": -0.11716204580164752, "output_peak": 0.43917166709392874}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9037152931410142, "全帯域_10ms_rms_envelope_nrmse": 0.08946176205002304, "全帯域_10ms_envelope_correlation": 0.9926038979052552, "全帯域_10ms_modulation_mae_db": 0.49018300494301875, "全帯域_40ms_log_rms_mae_db": 0.45673269812614825, "全帯域_40ms_rms_envelope_nrmse": 0.04621971406484924, "全帯域_40ms_envelope_correlation": 0.9978895286120733, "全帯域_40ms_modulation_mae_db": 0.5895392492580936, "300–1000 Hz_10ms_log_rms_mae_db": 1.4131015582181374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10934209531757177, "300–1000 Hz_10ms_envelope_correlation": 0.9905229316761079, "300–1000 Hz_10ms_modulation_mae_db": 0.3374046022099859, "300–1000 Hz_40ms_log_rms_mae_db": 0.7665842019741843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06518201334339248, "300–1000 Hz_40ms_envelope_correlation": 0.9966013883493785, "300–1000 Hz_40ms_modulation_mae_db": 0.4648982693073463, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5202653569268514, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1580839056066402, "1000–3000 Hz_10ms_envelope_correlation": 0.9802339355256989, "1000–3000 Hz_10ms_modulation_mae_db": 0.9014435690530004, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9365073672390797, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09381000029310417, "1000–3000 Hz_40ms_envelope_correlation": 0.9924592217678064, "1000–3000 Hz_40ms_modulation_mae_db": 1.0663726088118377, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4384020754426772, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16141244414146724, "3000–6000 Hz_10ms_envelope_correlation": 0.9804593371364079, "3000–6000 Hz_10ms_modulation_mae_db": 1.00088409881264, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7309229700501114, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06573748362463297, "3000–6000 Hz_40ms_envelope_correlation": 0.9965669812203758, "3000–6000 Hz_40ms_modulation_mae_db": 1.1944339146904102, "mr_stft_db": 5.678590586994708, "band_mae_db": 0.14977846866006236, "rms_gain_error_db": -0.053370068042567845, "output_peak": 0.450117553678003}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9619898693176738, "全帯域_10ms_rms_envelope_nrmse": 0.0850681721809039, "全帯域_10ms_envelope_correlation": 0.9933279671951879, "全帯域_10ms_modulation_mae_db": 0.2966207987654872, "全帯域_40ms_log_rms_mae_db": 0.4560807102523621, "全帯域_40ms_rms_envelope_nrmse": 0.038163413119224075, "全帯域_40ms_envelope_correlation": 0.998636847170182, "全帯域_40ms_modulation_mae_db": 0.7148032615519785, "300–1000 Hz_10ms_log_rms_mae_db": 1.442872781334026, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10806872332778762, "300–1000 Hz_10ms_envelope_correlation": 0.9906013217978081, "300–1000 Hz_10ms_modulation_mae_db": 0.831061520874568, "300–1000 Hz_40ms_log_rms_mae_db": 0.7391424098658813, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052845411321286555, "300–1000 Hz_40ms_envelope_correlation": 0.9976458164504783, "300–1000 Hz_40ms_modulation_mae_db": 1.012690183481108, "1000–3000 Hz_10ms_log_rms_mae_db": 1.549853827715494, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15616464160330223, "1000–3000 Hz_10ms_envelope_correlation": 0.9803745539791423, "1000–3000 Hz_10ms_modulation_mae_db": 0.8500012067434497, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9149983538774494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08397420532599631, "1000–3000 Hz_40ms_envelope_correlation": 0.9944862045813306, "1000–3000 Hz_40ms_modulation_mae_db": 0.5138551970234803, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4199888588081562, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1376663880431292, "3000–6000 Hz_10ms_envelope_correlation": 0.9861264564270974, "3000–6000 Hz_10ms_modulation_mae_db": 1.3218212140837913, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8150324846197506, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06849458375446997, "3000–6000 Hz_40ms_envelope_correlation": 0.9968025672610787, "3000–6000 Hz_40ms_modulation_mae_db": 1.695229659347336, "mr_stft_db": 5.68762009438704, "band_mae_db": 0.13203395049991276, "rms_gain_error_db": -0.09775507875042672, "output_peak": 0.41900347101301205}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8841029079584788, "全帯域_10ms_rms_envelope_nrmse": 0.08740445282444585, "全帯域_10ms_envelope_correlation": 0.9929426319395112, "全帯域_10ms_modulation_mae_db": 0.49957824629374326, "全帯域_40ms_log_rms_mae_db": 0.4308375868196739, "全帯域_40ms_rms_envelope_nrmse": 0.040247650707465554, "全帯域_40ms_envelope_correlation": 0.9984440024253397, "全帯域_40ms_modulation_mae_db": 0.48516072905244423, "300–1000 Hz_10ms_log_rms_mae_db": 1.393704147278264, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11757318656343482, "300–1000 Hz_10ms_envelope_correlation": 0.9888441519529515, "300–1000 Hz_10ms_modulation_mae_db": 0.3954257284592786, "300–1000 Hz_40ms_log_rms_mae_db": 0.7508734342187817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06801049722831178, "300–1000 Hz_40ms_envelope_correlation": 0.9960019978957583, "300–1000 Hz_40ms_modulation_mae_db": 0.3895130692190458, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5799152330582238, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15760221641243036, "1000–3000 Hz_10ms_envelope_correlation": 0.9801942683878144, "1000–3000 Hz_10ms_modulation_mae_db": 0.5429910965334933, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9800142307955161, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08813588998939935, "1000–3000 Hz_40ms_envelope_correlation": 0.994112356240323, "1000–3000 Hz_40ms_modulation_mae_db": 0.37269494667337916, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4220049303980011, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13935156220141973, "3000–6000 Hz_10ms_envelope_correlation": 0.9864740049787437, "3000–6000 Hz_10ms_modulation_mae_db": 1.2520046423385913, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8153621364288931, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07399774387082646, "3000–6000 Hz_40ms_envelope_correlation": 0.9970615023762069, "3000–6000 Hz_40ms_modulation_mae_db": 1.5939009199055365, "mr_stft_db": 5.663565747606445, "band_mae_db": 0.2066835328321651, "rms_gain_error_db": -0.0869825797530287, "output_peak": 0.43609414556431086}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9292752029439584, "全帯域_10ms_rms_envelope_nrmse": 0.08427486909451455, "全帯域_10ms_envelope_correlation": 0.993445530139077, "全帯域_10ms_modulation_mae_db": 0.4015407173065273, "全帯域_40ms_log_rms_mae_db": 0.4611702606897175, "全帯域_40ms_rms_envelope_nrmse": 0.046275420052021454, "全帯域_40ms_envelope_correlation": 0.9979173055025614, "全帯域_40ms_modulation_mae_db": 0.39362151688658575, "300–1000 Hz_10ms_log_rms_mae_db": 1.4464727242765971, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11632996360699568, "300–1000 Hz_10ms_envelope_correlation": 0.9891165878892151, "300–1000 Hz_10ms_modulation_mae_db": 0.10207436911175499, "300–1000 Hz_40ms_log_rms_mae_db": 0.8111082131508321, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06995100966893657, "300–1000 Hz_40ms_envelope_correlation": 0.9958489702490996, "300–1000 Hz_40ms_modulation_mae_db": 0.19849391355124335, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5958515955928698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1831142265147022, "1000–3000 Hz_10ms_envelope_correlation": 0.972517574856072, "1000–3000 Hz_10ms_modulation_mae_db": 1.12758052471634, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0320459908562716, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10955772744462608, "1000–3000 Hz_40ms_envelope_correlation": 0.9893656435916646, "1000–3000 Hz_40ms_modulation_mae_db": 1.0670437822125594, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4148143233154074, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15782130219698798, "3000–6000 Hz_10ms_envelope_correlation": 0.9814167000574541, "3000–6000 Hz_10ms_modulation_mae_db": 1.2133680649531553, "3000–6000 Hz_40ms_log_rms_mae_db": 0.751403375945219, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06264658937442319, "3000–6000 Hz_40ms_envelope_correlation": 0.9972345547560072, "3000–6000 Hz_40ms_modulation_mae_db": 1.7304268129565494, "mr_stft_db": 5.686981971021994, "band_mae_db": 0.18672642130247843, "rms_gain_error_db": -0.09597931597673115, "output_peak": 0.4216331154206302}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9119647897087378, "全帯域_10ms_rms_envelope_nrmse": 0.08398100819021931, "全帯域_10ms_envelope_correlation": 0.9936369218586003, "全帯域_10ms_modulation_mae_db": 1.4181297857338595, "全帯域_40ms_log_rms_mae_db": 0.4327517449181065, "全帯域_40ms_rms_envelope_nrmse": 0.04358355602947852, "全帯域_40ms_envelope_correlation": 0.9983635446509163, "全帯域_40ms_modulation_mae_db": 1.420158641448083, "300–1000 Hz_10ms_log_rms_mae_db": 1.3272658607278431, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10049682509409046, "300–1000 Hz_10ms_envelope_correlation": 0.9920253929979763, "300–1000 Hz_10ms_modulation_mae_db": 0.9070197224799169, "300–1000 Hz_40ms_log_rms_mae_db": 0.75112857813042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05634442934908841, "300–1000 Hz_40ms_envelope_correlation": 0.997513807172454, "300–1000 Hz_40ms_modulation_mae_db": 0.8771450391630293, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5574343028189377, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1834428407729445, "1000–3000 Hz_10ms_envelope_correlation": 0.97381211217877, "1000–3000 Hz_10ms_modulation_mae_db": 1.57766054258647, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0095416671798942, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10474044906393155, "1000–3000 Hz_40ms_envelope_correlation": 0.9928105901386874, "1000–3000 Hz_40ms_modulation_mae_db": 1.3411295565540904, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4677218823347353, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1576710514213991, "3000–6000 Hz_10ms_envelope_correlation": 0.9813593212628681, "3000–6000 Hz_10ms_modulation_mae_db": 0.58718997194784, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8112383429044192, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06453480434495242, "3000–6000 Hz_40ms_envelope_correlation": 0.9967354403388751, "3000–6000 Hz_40ms_modulation_mae_db": 0.7800121177788801, "mr_stft_db": 5.680654976818478, "band_mae_db": 0.2702654391463496, "rms_gain_error_db": -0.15743485065016571, "output_peak": 0.3957195080705708}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37680, "output_samples": 37680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.71158117615012, "spectral_convergence": 0.3485257011270288, "spectral_convergence_status": "ok", "amplitude_floor": 1.2997261015401668e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.412851650069928, "spectral_convergence": 0.3159964797261885, "spectral_convergence_status": "ok", "amplitude_floor": 1.1046780177944747e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.901812597623684, "spectral_convergence": 0.33536373326628804, "spectral_convergence_status": "ok", "amplitude_floor": 9.696590055390529e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.751691690693756e-05, "output_power": 8.222316155944788e-05, "delta_db": 0.25597680684405744, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016017256752476519, "output_power": 0.0015388482862075388, "delta_db": -0.173923320567887, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006838056632308089, "output_power": 0.0006690579952975963, "delta_db": -0.09468928350205358, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003361315194748372, "output_power": 0.00034257570783199963, "delta_db": 0.08247325023577443, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00035354415012418854, "output_power": 0.0003275696103272473, "delta_db": -0.3314005155868782, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00021020525507420865, "output_power": 0.0001952876484414108, "delta_db": -0.31968793184801897, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2096918514307783, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.14530873858433813, "pitch": null, "output_rms_dbfs": -25.009232260722506, "output_peak": 0.4406077154287219, "reference_rms_dbfs": -24.863923522138165, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.675415141281245}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00018834951637393047, "log_rms_mae_db": 0.8750656469663882, "rms_envelope_nrmse": 0.08904470475777665, "envelope_correlation": 0.9928150327400733, "rms_gain_error_db": -0.14530372159391225, "reference_mean_power": 0.00326985830085679, "output_mean_power": 0.003162267180625409}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.479011996203747e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005558024994761568, "output_power": 0.0005586452427242071, "reference_active": true, "delta_db": 0.022156092870461136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011456565813557757, "output_power": 0.00010366596261973152, "reference_active": true, "delta_db": -0.43418269495570544}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.821714301732997e-05, "output_power": 3.593775162753154e-05, "reference_active": true, "delta_db": -0.26707314694689316}], "active_band_count": 3, "active_band_mae_db": 0.2411373115910199}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00017471448177334785, "log_rms_mae_db": 0.459412623834744, "rms_envelope_nrmse": 0.04753659100904025, "envelope_correlation": 0.9979941580804033, "rms_gain_error_db": -0.1453009555772169, "reference_mean_power": 0.0033120735889998944, "output_mean_power": 0.0032030954608129898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 7.763101583949748e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004881183543297566, "output_power": 0.0004898896122119223, "reference_active": true, "delta_db": 0.015730922275153873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.883974713754685e-05, "output_power": 5.1601002631308626e-05, "reference_active": true, "delta_db": -0.5701265743748636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7824879244994614e-06, "output_power": 1.4932635278814334e-06, "reference_active": true, "delta_db": -0.7689013856805524}], "active_band_count": 3, "active_band_mae_db": 0.45158629411018997}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00017587844894922502, "log_rms_mae_db": 1.4520588265819026, "rms_envelope_nrmse": 0.11083080864779098, "envelope_correlation": 0.9902975826314666, "rms_gain_error_db": -0.19031123299325198, "reference_mean_power": 0.002090015296145697, "output_mean_power": 0.002000406887999915}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 6.927253047980857e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041075223634869497, "output_power": 0.00040780314279658915, "reference_active": true, "delta_db": -0.031293682273356195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.156045953139752e-05, "output_power": 7.484359419075959e-05, "reference_active": true, "delta_db": -0.3732502871511235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2217147797459886e-05, "output_power": 2.9126747603920297e-05, "reference_active": true, "delta_db": 1.1760368833093033}], "active_band_count": 3, "active_band_mae_db": 0.5268602842445943}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00016316895420677254, "log_rms_mae_db": 0.8385406225247343, "rms_envelope_nrmse": 0.062086027914702156, "envelope_correlation": 0.9970422980305513, "rms_gain_error_db": -0.1903108680040501, "reference_mean_power": 0.002117036355483854, "output_mean_power": 0.0020262696026090285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 5.84659163450443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037072755253225205, "output_power": 0.00036887647084511807, "reference_active": true, "delta_db": -0.021739098637856367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.09979708848843e-05, "output_power": 3.6612382523036915e-05, "reference_active": true, "delta_db": -0.49134371502618934}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.223097583568823e-06, "output_power": 1.2300634165840186e-06, "reference_active": true, "delta_db": 0.02466394151016433}], "active_band_count": 3, "active_band_mae_db": 0.17924891839140336}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0001038394087983997, "log_rms_mae_db": 1.5009171290963848, "rms_envelope_nrmse": 0.18057461728174104, "envelope_correlation": 0.9732271325766214, "rms_gain_error_db": -0.08430343452795411, "reference_mean_power": 0.0005521302842545539, "output_mean_power": 0.0005415159154735133}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 2.3748882259523475e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000136307073642641, "output_power": 0.0001448463270944792, "reference_active": true, "delta_db": 0.2638909307229254}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.293151120982887e-05, "output_power": 2.909148700734717e-05, "reference_active": true, "delta_db": -0.53845739998887}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.586128694071059e-05, "output_power": 1.354409554840529e-05, "reference_active": true, "delta_db": -0.685884129647676}], "active_band_count": 3, "active_band_mae_db": 0.4960774867864905}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 8.776193320418374e-05, "log_rms_mae_db": 0.8767495699952426, "rms_envelope_nrmse": 0.10387860746169038, "envelope_correlation": 0.9902263439494391, "rms_gain_error_db": -0.08428669511898855, "reference_mean_power": 0.0005592400075632049, "output_mean_power": 0.0005484910727725137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.9436382041482247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012615724913962246, "output_power": 0.0001349574556910971, "reference_active": true, "delta_db": 0.29284671418460795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.519569809811637e-05, "output_power": 1.2810738051767982e-05, "reference_active": true, "delta_db": -0.7414650556102056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.080674445150989e-07, "output_power": 8.373118501616061e-07, "reference_active": true, "delta_db": -0.35230868232951607}], "active_band_count": 3, "active_band_mae_db": 0.4622068173747765}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 6.310171076800209e-05, "log_rms_mae_db": 1.4840375048742742, "rms_envelope_nrmse": 0.15283238543776875, "envelope_correlation": 0.9832786792512418, "rms_gain_error_db": -0.3771513867776829, "reference_mean_power": 0.00026303793349419837, "output_mean_power": 0.00024115887279052274}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 1.3356760090821468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.8507771208283054e-05, "output_power": 5.323986158951688e-05, "reference_active": true, "delta_db": -0.40976636690691304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1981263949052112e-05, "output_power": 1.1830102742673405e-05, "reference_active": true, "delta_db": -0.055141194179635915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.256106518668102e-06, "output_power": 5.072988584429767e-06, "reference_active": true, "delta_db": -2.611644587642319}], "active_band_count": 3, "active_band_mae_db": 1.0255173829096227}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 5.861628173421479e-05, "log_rms_mae_db": 0.7632782527462122, "rms_envelope_nrmse": 0.0736015262440006, "envelope_correlation": 0.9969840939896161, "rms_gain_error_db": -0.3771319462323675, "reference_mean_power": 0.00026641634082957415, "output_mean_power": 0.00024425736316495563}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.157398705126346e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.177699204483558e-05, "output_power": 4.6840417243764745e-05, "reference_active": true, "delta_db": -0.43516061724737454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.036512028004977e-06, "output_power": 6.364037024886726e-06, "reference_active": true, "delta_db": -0.4362473674014005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.588547226531691e-07, "output_power": 2.5139408569772055e-07, "reference_active": true, "delta_db": -3.4694386907869084}], "active_band_count": 3, "active_band_mae_db": 1.4469488918118945}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8814744056035166, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ab68c34796c9fe8406b1f4d0551ddf9ea6092f41f3e47436a50ad87c849201b2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9764594723240548, "standard_deviation": 0.003750160520235863, "minimum": 0.9724453677557605, "maximum": 0.9803745539791423, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5385995723656816, "standard_deviation": 0.04292503349405297, "minimum": 1.4633095104056115, "maximum": 1.5958515955928698, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9168617862140293, "standard_deviation": 0.3386778288751378, "minimum": 0.4960774867864905, "maximum": 1.57766054258647, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17065344768857516, "standard_deviation": 0.012936566806440033, "minimum": 0.15616464160330223, "maximum": 0.1834428407729445, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919813742971639, "standard_deviation": 0.0022188268326922803, "minimum": 0.9887897816605414, "maximum": 0.9944862045813306, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9558618637005533, "standard_deviation": 0.05267040990056313, "minimum": 0.8767495699952426, "maximum": 1.0320459908562716, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7573069126258895, "standard_deviation": 0.3607456974664692, "minimum": 0.37269494667337916, "maximum": 1.3411295565540904, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09850045150468088, "standard_deviation": 0.010253233634827816, "minimum": 0.08397420532599631, "maximum": 0.11147950346685138, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9836510363017523, "standard_deviation": 0.0025463772288196475, "minimum": 0.9804593371364079, "maximum": 0.9868941565607495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4410085624672568, "standard_deviation": 0.0320858145738995, "minimum": 1.4008941620311566, "maximum": 1.4840375048742742, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0773431753880134, "standard_deviation": 0.24544729455095096, "minimum": 0.58718997194784, "maximum": 1.3218212140837913, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14930897091834552, "standard_deviation": 0.010258637439117637, "minimum": 0.1354539780431956, "maximum": 0.16141244414146724, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9968608716827934, "standard_deviation": 0.00023536115797523962, "minimum": 0.9965669812203758, "maximum": 0.9972345547560072, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7813762475825243, "standard_deviation": 0.04333794984465847, "minimum": 0.7262276345235723, "maximum": 0.8375447834420163, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4087763810054144, "standard_deviation": 0.3169934299893959, "minimum": 0.7800121177788801, "maximum": 1.7304268129565494, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06940295170298989, "standard_deviation": 0.004659335374665211, "minimum": 0.06264658937442319, "maximum": 0.07415443202649707, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9902225997550649, "standard_deviation": 0.0014593855867524191, "minimum": 0.9881500784812002, "maximum": 0.9922227506137937, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4187425233232163, "standard_deviation": 0.05129778029667289, "minimum": 1.3272658607278431, "maximum": 1.4931274644185137, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4392805164561531, "standard_deviation": 0.2994065234986711, "minimum": 0.10207436911175499, "maximum": 0.9070197224799169, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1102677763881055, "standard_deviation": 0.00804143223109475, "minimum": 0.09833723007151794, "maximum": 0.12116337847565468, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9966904910956284, "standard_deviation": 0.001134005550975654, "minimum": 0.9947075924636078, "maximum": 0.998162058153699, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7926165983069806, "standard_deviation": 0.05567877023576289, "minimum": 0.7391424098658813, "maximum": 0.9026684659209556, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4574477347241897, "standard_deviation": 0.32790245026066633, "minimum": 0.11403990055655461, "maximum": 1.012690183481108, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06247245327619812, "standard_deviation": 0.01016598345264755, "minimum": 0.046884203141367396, "maximum": 0.07847603424249959, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.20165064214904477, "standard_deviation": 0.04843164830378677, "minimum": 0.13203395049991276, "maximum": 0.2702654391463496, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.678857231814477, "standard_deviation": 0.007864896440379302, "minimum": 5.663565747606445, "maximum": 5.68762009438704, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42664742941153533, "standard_deviation": 0.01805571801811185, "minimum": 0.3957195080705708, "maximum": 0.450117553678003, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10450417289113942, "standard_deviation": 0.03416645117662113, "minimum": -0.15743485065016571, "maximum": -0.053370068042567845, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930205750036915, "standard_deviation": 0.0007792973554045758, "minimum": 0.9914385470183961, "maximum": 0.9939540712334312, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8983465830973454, "standard_deviation": 0.03602174477410975, "minimum": 0.8496957633812482, "maximum": 0.9619898693176738, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5481803303147306, "standard_deviation": 0.3742166805968311, "minimum": 0.2411373115910199, "maximum": 1.4181297857338595, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08706363597527882, "standard_deviation": 0.004688793465738461, "minimum": 0.08095953767987003, "maximum": 0.0963145810244772, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981566199105566, "standard_deviation": 0.000468255822710417, "minimum": 0.9973048226581094, "maximum": 0.9987027501848678, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4468660828852355, "standard_deviation": 0.013846505617242587, "minimum": 0.428554708646161, "maximum": 0.4611702606897175, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6583081899631824, "standard_deviation": 0.3483920837768036, "minimum": 0.3680852100574274, "maximum": 1.420158641448083, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.043970786554850344, "standard_deviation": 0.005354632022226533, "minimum": 0.03686165719607963, "maximum": 0.052878290260644036, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8463-294828-0022", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 1.5589680671691895, "predicted_mos": 1.5589680671691895, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "1735a5c6f6942d940b9b1e4c9a48171e48e4afdeab369dc1fd38627d9e70e847", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.878524400293827, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8019990227939667, "全帯域_10ms_rms_envelope_nrmse": 0.07423505181109886, "全帯域_10ms_envelope_correlation": 0.9949148707863806, "全帯域_10ms_modulation_mae_db": 0.4919253018974552, "全帯域_40ms_log_rms_mae_db": 0.3909371817819696, "全帯域_40ms_rms_envelope_nrmse": 0.03873097046890168, "全帯域_40ms_envelope_correlation": 0.9985414895613467, "全帯域_40ms_modulation_mae_db": 0.37074299927869686, "300–1000 Hz_10ms_log_rms_mae_db": 1.2176139067108027, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08473895909584421, "300–1000 Hz_10ms_envelope_correlation": 0.9942253685725069, "300–1000 Hz_10ms_modulation_mae_db": 0.13308570431015548, "300–1000 Hz_40ms_log_rms_mae_db": 0.7754808871517814, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05111756514810471, "300–1000 Hz_40ms_envelope_correlation": 0.9977475532998035, "300–1000 Hz_40ms_modulation_mae_db": 0.22812533169575966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.496761737185916, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16227696654648247, "1000–3000 Hz_10ms_envelope_correlation": 0.9786859159999142, "1000–3000 Hz_10ms_modulation_mae_db": 0.9922873662772717, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9103600740106192, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07992384666447734, "1000–3000 Hz_40ms_envelope_correlation": 0.9950769004641488, "1000–3000 Hz_40ms_modulation_mae_db": 0.7915047164558677, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4250538130311692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1404638035226796, "3000–6000 Hz_10ms_envelope_correlation": 0.9865913546423132, "3000–6000 Hz_10ms_modulation_mae_db": 1.3768898128566907, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7828473760584087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07635354483193703, "3000–6000 Hz_40ms_envelope_correlation": 0.9971754181255628, "3000–6000 Hz_40ms_modulation_mae_db": 1.5865061664052644, "mr_stft_db": 5.650239488833155, "band_mae_db": 0.24386307526095172, "rms_gain_error_db": -0.07858693759885836, "output_peak": 0.43873006283747423}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8967022573128451, "全帯域_10ms_rms_envelope_nrmse": 0.08574082011423417, "全帯域_10ms_envelope_correlation": 0.993282015174722, "全帯域_10ms_modulation_mae_db": 0.7039506699249166, "全帯域_40ms_log_rms_mae_db": 0.4593520320492321, "全帯域_40ms_rms_envelope_nrmse": 0.04111974654145323, "全帯域_40ms_envelope_correlation": 0.9984534062869509, "全帯域_40ms_modulation_mae_db": 0.585959055398258, "300–1000 Hz_10ms_log_rms_mae_db": 1.3499670379539483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10048569356946652, "300–1000 Hz_10ms_envelope_correlation": 0.9919574930901519, "300–1000 Hz_10ms_modulation_mae_db": 0.16412884856725618, "300–1000 Hz_40ms_log_rms_mae_db": 0.7794088429457489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050325020542526404, "300–1000 Hz_40ms_envelope_correlation": 0.9980057114860533, "300–1000 Hz_40ms_modulation_mae_db": 0.29137401286884973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6560618248002594, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15367666954743744, "1000–3000 Hz_10ms_envelope_correlation": 0.9807383329780961, "1000–3000 Hz_10ms_modulation_mae_db": 1.0854467977482534, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0545935555957529, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08238883431095384, "1000–3000 Hz_40ms_envelope_correlation": 0.994253659952693, "1000–3000 Hz_40ms_modulation_mae_db": 1.2378389884123964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4097574056841502, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1605647412938385, "3000–6000 Hz_10ms_envelope_correlation": 0.9807015495970741, "3000–6000 Hz_10ms_modulation_mae_db": 0.8206084801002923, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8189087420987348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07492349530698238, "3000–6000 Hz_40ms_envelope_correlation": 0.9957283703080377, "3000–6000 Hz_40ms_modulation_mae_db": 1.4582321283590869, "mr_stft_db": 5.672567266308828, "band_mae_db": 0.1366017598793883, "rms_gain_error_db": -0.1122443809300746, "output_peak": 0.4450817235656047}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9013101807626561, "全帯域_10ms_rms_envelope_nrmse": 0.07825957635011385, "全帯域_10ms_envelope_correlation": 0.9946776937775345, "全帯域_10ms_modulation_mae_db": 0.6088113400678198, "全帯域_40ms_log_rms_mae_db": 0.4106235785006553, "全帯域_40ms_rms_envelope_nrmse": 0.043969954076204944, "全帯域_40ms_envelope_correlation": 0.9985190956040959, "全帯域_40ms_modulation_mae_db": 0.8239626660868602, "300–1000 Hz_10ms_log_rms_mae_db": 1.36237441624849, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11165509319106132, "300–1000 Hz_10ms_envelope_correlation": 0.9902565312180832, "300–1000 Hz_10ms_modulation_mae_db": 0.2250290859192334, "300–1000 Hz_40ms_log_rms_mae_db": 0.7774205357780107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06892742873136677, "300–1000 Hz_40ms_envelope_correlation": 0.9963714815446565, "300–1000 Hz_40ms_modulation_mae_db": 0.23054243958122125, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5553276514194831, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18219662483349855, "1000–3000 Hz_10ms_envelope_correlation": 0.9733324516969257, "1000–3000 Hz_10ms_modulation_mae_db": 0.6400610334341786, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9698243595477595, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11325808051124933, "1000–3000 Hz_40ms_envelope_correlation": 0.9896913091257281, "1000–3000 Hz_40ms_modulation_mae_db": 0.5972449464682263, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4389148767952367, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14758018714006604, "3000–6000 Hz_10ms_envelope_correlation": 0.98371492486862, "3000–6000 Hz_10ms_modulation_mae_db": 1.128246671888807, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8292272592759145, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06748113885929755, "3000–6000 Hz_40ms_envelope_correlation": 0.9964758118721166, "3000–6000 Hz_40ms_modulation_mae_db": 1.3201348822789936, "mr_stft_db": 5.668477394492012, "band_mae_db": 0.2372585481089784, "rms_gain_error_db": -0.19715236540687336, "output_peak": 0.41742253486732017}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9073046196372584, "全帯域_10ms_rms_envelope_nrmse": 0.09620959352335184, "全帯域_10ms_envelope_correlation": 0.9918259973518685, "全帯域_10ms_modulation_mae_db": 0.6051803368036673, "全帯域_40ms_log_rms_mae_db": 0.472615969814671, "全帯域_40ms_rms_envelope_nrmse": 0.05602391788602675, "全帯域_40ms_envelope_correlation": 0.9974349679641223, "全帯域_40ms_modulation_mae_db": 0.7596845311674763, "300–1000 Hz_10ms_log_rms_mae_db": 1.4013052655950786, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13060237326730545, "300–1000 Hz_10ms_envelope_correlation": 0.986596519446588, "300–1000 Hz_10ms_modulation_mae_db": 1.3057906486615254, "300–1000 Hz_40ms_log_rms_mae_db": 0.7887635879264066, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08066665277297275, "300–1000 Hz_40ms_envelope_correlation": 0.994954443911858, "300–1000 Hz_40ms_modulation_mae_db": 1.2611595790338956, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5407502367679298, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17409449904156366, "1000–3000 Hz_10ms_envelope_correlation": 0.9760431753993636, "1000–3000 Hz_10ms_modulation_mae_db": 1.276701193911782, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9480151464893126, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1019001321575776, "1000–3000 Hz_40ms_envelope_correlation": 0.9923936806080201, "1000–3000 Hz_40ms_modulation_mae_db": 1.4756268779069757, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3810244244292171, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.13499484820611288, "3000–6000 Hz_10ms_envelope_correlation": 0.9867816839675637, "3000–6000 Hz_10ms_modulation_mae_db": 1.061756327493157, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7787508676529455, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06880275415596167, "3000–6000 Hz_40ms_envelope_correlation": 0.9969244542109824, "3000–6000 Hz_40ms_modulation_mae_db": 1.0719173467290315, "mr_stft_db": 5.676989954754364, "band_mae_db": 0.24891170444672536, "rms_gain_error_db": -0.22685114236382664, "output_peak": 0.40532492641254686}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8908560067784761, "全帯域_10ms_rms_envelope_nrmse": 0.09243321114088758, "全帯域_10ms_envelope_correlation": 0.9920493585759874, "全帯域_10ms_modulation_mae_db": 0.4882294654907682, "全帯域_40ms_log_rms_mae_db": 0.44921900366937656, "全帯域_40ms_rms_envelope_nrmse": 0.052653093653187676, "全帯域_40ms_envelope_correlation": 0.9971727113449815, "全帯域_40ms_modulation_mae_db": 0.43250824833305135, "300–1000 Hz_10ms_log_rms_mae_db": 1.3472735852623798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12440613651800023, "300–1000 Hz_10ms_envelope_correlation": 0.9876770494932464, "300–1000 Hz_10ms_modulation_mae_db": 1.2107072691289509, "300–1000 Hz_40ms_log_rms_mae_db": 0.7684187069534737, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06837670848501684, "300–1000 Hz_40ms_envelope_correlation": 0.9959957203369839, "300–1000 Hz_40ms_modulation_mae_db": 0.9154880701433603, "1000–3000 Hz_10ms_log_rms_mae_db": 1.468634020838899, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1748829896793209, "1000–3000 Hz_10ms_envelope_correlation": 0.9753036557334169, "1000–3000 Hz_10ms_modulation_mae_db": 0.47409960500628134, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9251269335530471, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10144402003420355, "1000–3000 Hz_40ms_envelope_correlation": 0.9917552928316881, "1000–3000 Hz_40ms_modulation_mae_db": 0.3269673776451964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.494864668626884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1421884934610664, "3000–6000 Hz_10ms_envelope_correlation": 0.9852485634747331, "3000–6000 Hz_10ms_modulation_mae_db": 0.9361262018928013, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8273753187814653, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06689945296354016, "3000–6000 Hz_40ms_envelope_correlation": 0.9971194857942115, "3000–6000 Hz_40ms_modulation_mae_db": 1.2921302942187314, "mr_stft_db": 5.6644565621114, "band_mae_db": 0.2437821825418359, "rms_gain_error_db": -0.01843847420864288, "output_peak": 0.4271549530466242}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8521313401019299, "全帯域_10ms_rms_envelope_nrmse": 0.0851387837753776, "全帯域_10ms_envelope_correlation": 0.9933634304068911, "全帯域_10ms_modulation_mae_db": 0.37939776988756874, "全帯域_40ms_log_rms_mae_db": 0.377088257322434, "全帯域_40ms_rms_envelope_nrmse": 0.03859226689703745, "全帯域_40ms_envelope_correlation": 0.9986509078912076, "全帯域_40ms_modulation_mae_db": 0.23302252229812978, "300–1000 Hz_10ms_log_rms_mae_db": 1.2525011200340819, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0978251066444249, "300–1000 Hz_10ms_envelope_correlation": 0.9925113436027966, "300–1000 Hz_10ms_modulation_mae_db": 0.1567703990718564, "300–1000 Hz_40ms_log_rms_mae_db": 0.7167984847967785, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05276886391866539, "300–1000 Hz_40ms_envelope_correlation": 0.9979473547131675, "300–1000 Hz_40ms_modulation_mae_db": 0.1128994501281046, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5725082049609844, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17303776096876777, "1000–3000 Hz_10ms_envelope_correlation": 0.9753481139561119, "1000–3000 Hz_10ms_modulation_mae_db": 0.6312449276450403, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9857858104112402, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09394686612640298, "1000–3000 Hz_40ms_envelope_correlation": 0.9919340206736574, "1000–3000 Hz_40ms_modulation_mae_db": 0.3817707891676758, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3798454999046577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16263822553635193, "3000–6000 Hz_10ms_envelope_correlation": 0.9801652196946785, "3000–6000 Hz_10ms_modulation_mae_db": 0.7076188791239396, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8324911215575412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07694211035588637, "3000–6000 Hz_40ms_envelope_correlation": 0.9952624199739835, "3000–6000 Hz_40ms_modulation_mae_db": 0.966081019824473, "mr_stft_db": 5.656389760866106, "band_mae_db": 0.11162490248815791, "rms_gain_error_db": -0.10650102291514796, "output_peak": 0.42353747788732177}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9167878604155579, "全帯域_10ms_rms_envelope_nrmse": 0.09510710362632047, "全帯域_10ms_envelope_correlation": 0.9917347733517119, "全帯域_10ms_modulation_mae_db": 0.5868062860508202, "全帯域_40ms_log_rms_mae_db": 0.4980492257134642, "全帯域_40ms_rms_envelope_nrmse": 0.054229443342478305, "全帯域_40ms_envelope_correlation": 0.9972646612004569, "全帯域_40ms_modulation_mae_db": 0.3780497624796173, "300–1000 Hz_10ms_log_rms_mae_db": 1.4516997975947874, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12840841676578457, "300–1000 Hz_10ms_envelope_correlation": 0.9867389568366723, "300–1000 Hz_10ms_modulation_mae_db": 1.1734943285187496, "300–1000 Hz_40ms_log_rms_mae_db": 0.8129015839621531, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06996353149061738, "300–1000 Hz_40ms_envelope_correlation": 0.9959963294821557, "300–1000 Hz_40ms_modulation_mae_db": 0.7411160152191698, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5359330289228643, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15344812365694815, "1000–3000 Hz_10ms_envelope_correlation": 0.980940314950142, "1000–3000 Hz_10ms_modulation_mae_db": 0.9616717812129997, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9862651533621689, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08995372010542368, "1000–3000 Hz_40ms_envelope_correlation": 0.993251778923282, "1000–3000 Hz_40ms_modulation_mae_db": 0.9413231756495817, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4195853320493244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1329252165142658, "3000–6000 Hz_10ms_envelope_correlation": 0.9872034110802329, "3000–6000 Hz_10ms_modulation_mae_db": 1.178984866650497, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7561936485946842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.06053162740824376, "3000–6000 Hz_40ms_envelope_correlation": 0.9977814230810907, "3000–6000 Hz_40ms_modulation_mae_db": 1.530977027103056, "mr_stft_db": 5.651187264841245, "band_mae_db": 0.2070416576001454, "rms_gain_error_db": -0.13811970590563324, "output_peak": 0.4283316813613276}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.898302373451155, "全帯域_10ms_rms_envelope_nrmse": 0.08306962968238098, "全帯域_10ms_envelope_correlation": 0.9935906462189692, "全帯域_10ms_modulation_mae_db": 0.26013933603660794, "全帯域_40ms_log_rms_mae_db": 0.44192881827754926, "全帯域_40ms_rms_envelope_nrmse": 0.04062555953997876, "全帯域_40ms_envelope_correlation": 0.9983184733171487, "全帯域_40ms_modulation_mae_db": 0.20064537373202332, "300–1000 Hz_10ms_log_rms_mae_db": 1.3577329970295977, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11223815451938479, "300–1000 Hz_10ms_envelope_correlation": 0.9900274600035801, "300–1000 Hz_10ms_modulation_mae_db": 0.5886433628361846, "300–1000 Hz_40ms_log_rms_mae_db": 0.7394639782430669, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06212724290257301, "300–1000 Hz_40ms_envelope_correlation": 0.996750153826372, "300–1000 Hz_40ms_modulation_mae_db": 0.8943272272704977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5943961459917406, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1918181179797546, "1000–3000 Hz_10ms_envelope_correlation": 0.970610253057344, "1000–3000 Hz_10ms_modulation_mae_db": 0.3936690622147611, "1000–3000 Hz_40ms_log_rms_mae_db": 1.003303906040525, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11881253788701422, "1000–3000 Hz_40ms_envelope_correlation": 0.9891066250503561, "1000–3000 Hz_40ms_modulation_mae_db": 0.40895534079440005, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4086020673811852, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1294592239273676, "3000–6000 Hz_10ms_envelope_correlation": 0.9874863518011414, "3000–6000 Hz_10ms_modulation_mae_db": 1.0140397943076744, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7637596637331916, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.05493366497295444, "3000–6000 Hz_40ms_envelope_correlation": 0.9976243766568074, "3000–6000 Hz_40ms_modulation_mae_db": 1.0955874273651, "mr_stft_db": 5.655470078579462, "band_mae_db": 0.2371395644065146, "rms_gain_error_db": -0.002691462405460432, "output_peak": 0.4111384463194404}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37680, "output_samples": 37680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.677249372655074, "spectral_convergence": 0.34457619810224993, "spectral_convergence_status": "ok", "amplitude_floor": 1.2997261015401668e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.395391647885816, "spectral_convergence": 0.3171798313842587, "spectral_convergence_status": "ok", "amplitude_floor": 1.1046780177944747e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.878077445958579, "spectral_convergence": 0.35058357159941445, "spectral_convergence_status": "ok", "amplitude_floor": 9.696590055390529e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.751691690693756e-05, "output_power": 8.219921659101459e-05, "delta_db": 0.25471187346328206, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016017256752476519, "output_power": 0.001623502610168114, "delta_db": 0.05864853933253553, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006838056632308089, "output_power": 0.0006567278208514582, "delta_db": -0.17547278649866566, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003361315194748372, "output_power": 0.00032316930317241024, "delta_db": -0.17079136912512582, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00035354415012418854, "output_power": 0.0003337687648987728, "delta_db": -0.24997963998234812, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00021020525507420865, "output_power": 0.00018504874529748626, "delta_db": -0.5535742431637529, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.24386307526095172, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07858693759885836, "pitch": null, "output_rms_dbfs": -24.94251045973703, "output_peak": 0.43873006283747423, "reference_rms_dbfs": -24.863923522138165, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.650239488833155}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00018834951637393047, "log_rms_mae_db": 0.8019990227939667, "rms_envelope_nrmse": 0.07423505181109886, "envelope_correlation": 0.9949148707863806, "rms_gain_error_db": -0.07858487288787845, "reference_mean_power": 0.00326985830085679, "output_mean_power": 0.003211222851476177}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.479011996203747e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005558024994761568, "output_power": 0.0005349324147855156, "reference_active": true, "delta_db": -0.16621580275492545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011456565813557757, "output_power": 0.00011180964103427366, "reference_active": true, "delta_db": -0.1057520113376052}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.821714301732997e-05, "output_power": 2.896525847987756e-05, "reference_active": true, "delta_db": -1.2038080915998348}], "active_band_count": 3, "active_band_mae_db": 0.4919253018974552}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00017471448177334785, "log_rms_mae_db": 0.3909371817819696, "rms_envelope_nrmse": 0.03873097046890168, "envelope_correlation": 0.9985414895613467, "rms_gain_error_db": -0.07858436334877357, "reference_mean_power": 0.0033120735889998944, "output_mean_power": 0.0032526815121725525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 7.763101583949748e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004881183543297566, "output_power": 0.00046757198598492057, "reference_active": true, "delta_db": -0.18676655360717337}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.883974713754685e-05, "output_power": 5.794883265137488e-05, "reference_active": true, "delta_db": -0.06626105934217437}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7824879244994614e-06, "output_power": 1.4625356395309465e-06, "reference_active": true, "delta_db": -0.8592013848867428}], "active_band_count": 3, "active_band_mae_db": 0.37074299927869686}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00017587844894922502, "log_rms_mae_db": 1.2176139067108027, "rms_envelope_nrmse": 0.08473895909584421, "envelope_correlation": 0.9942253685725069, "rms_gain_error_db": -0.04324168498020406, "reference_mean_power": 0.002090015296145697, "output_mean_power": 0.002069308759202283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 6.927253047980857e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041075223634869497, "output_power": 0.0003927842306387861, "reference_active": true, "delta_db": -0.1942589307352399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.156045953139752e-05, "output_power": 7.867618423919781e-05, "reference_active": true, "delta_db": -0.1563637536926748}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2217147797459886e-05, "output_power": 2.196973716606391e-05, "reference_active": true, "delta_db": -0.048634428502551785}], "active_band_count": 3, "active_band_mae_db": 0.13308570431015548}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00016316895420677254, "log_rms_mae_db": 0.7754808871517814, "rms_envelope_nrmse": 0.05111756514810471, "envelope_correlation": 0.9977475532998035, "rms_gain_error_db": -0.04324282758399303, "reference_mean_power": 0.002117036355483854, "output_mean_power": 0.0020960615596774157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 5.84659163450443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037072755253225205, "output_power": 0.0003512493927113094, "reference_active": true, "delta_db": -0.23439282020636695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.09979708848843e-05, "output_power": 4.038571932050989e-05, "reference_active": true, "delta_db": -0.06534540070136505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.223097583568823e-06, "output_power": 1.336364362496476e-06, "reference_active": true, "delta_db": 0.38463777417954703}], "active_band_count": 3, "active_band_mae_db": 0.22812533169575966}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0001038394087983997, "log_rms_mae_db": 1.496761737185916, "rms_envelope_nrmse": 0.16227696654648247, "envelope_correlation": 0.9786859159999142, "rms_gain_error_db": -0.2964532910705634, "reference_mean_power": 0.0005521302842545539, "output_mean_power": 0.0005156989411105603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 2.3748882259523475e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000136307073642641, "output_power": 0.00012529305508656257, "reference_active": true, "delta_db": -0.36591395092821943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.293151120982887e-05, "output_power": 3.228240423489489e-05, "reference_active": true, "delta_db": -0.08645789341625568}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.586128694071059e-05, "output_power": 8.869301050755503e-06, "reference_active": true, "delta_db": -2.5244902544873398}], "active_band_count": 3, "active_band_mae_db": 0.9922873662772717}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 8.776193320418374e-05, "log_rms_mae_db": 0.9103600740106192, "rms_envelope_nrmse": 0.07992384666447734, "envelope_correlation": 0.9950769004641488, "rms_gain_error_db": -0.2964686890918436, "reference_mean_power": 0.0005592400075632049, "output_mean_power": 0.000522337689899596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.9436382041482247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012615724913962246, "output_power": 0.00011390749463392763, "reference_active": true, "delta_db": -0.44359910858557866}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.519569809811637e-05, "output_power": 1.4778334637864077e-05, "reference_active": true, "delta_db": -0.12095160166488526}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.080674445150989e-07, "output_power": 5.985793936376856e-07, "reference_active": true, "delta_db": -1.8099634391171395}], "active_band_count": 3, "active_band_mae_db": 0.7915047164558677}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 6.310171076800209e-05, "log_rms_mae_db": 1.4250538130311692, "rms_envelope_nrmse": 0.1404638035226796, "envelope_correlation": 0.9865913546423132, "rms_gain_error_db": -0.4368633010269715, "reference_mean_power": 0.00026303793349419837, "output_mean_power": 0.00023786582712620078}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 1.3356760090821468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.8507771208283054e-05, "output_power": 5.1153527482659876e-05, "reference_active": true, "delta_db": -0.5833796695658469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1981263949052112e-05, "output_power": 1.1331000790124416e-05, "reference_active": true, "delta_db": -0.24234366013840497}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.256106518668102e-06, "output_power": 4.324478406897423e-06, "reference_active": true, "delta_db": -3.3049461088658205}], "active_band_count": 3, "active_band_mae_db": 1.3768898128566907}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 5.861628173421479e-05, "log_rms_mae_db": 0.7828473760584087, "rms_envelope_nrmse": 0.07635354483193703, "envelope_correlation": 0.9971754181255628, "rms_gain_error_db": -0.43685683064877084, "reference_mean_power": 0.00026641634082957415, "output_mean_power": 0.00024092128783104644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.157398705126346e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.177699204483558e-05, "output_power": 4.449036034010261e-05, "reference_active": true, "delta_db": -0.6587089350157121}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.036512028004977e-06, "output_power": 6.068631291983948e-06, "reference_active": true, "delta_db": -0.6426668223837878}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.588547226531691e-07, "output_power": 2.5204880988707564e-07, "reference_active": true, "delta_db": -3.4581427418162933}], "active_band_count": 3, "active_band_mae_db": 1.5865061664052644}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8903658442010568, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "07bf973a45254974f81d69723ff70e13235e0bd7f51b96bc5df81eccca868a66", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9763752767214142, "standard_deviation": 0.003586141897388847, "minimum": 0.970610253057344, "maximum": 0.980940314950142, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5525466063610096, "standard_deviation": 0.05786859865847689, "minimum": 1.468634020838899, "maximum": 1.6560618248002594, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8068977209313211, "standard_deviation": 0.3155109498190147, "minimum": 0.3936690622147611, "maximum": 1.276701193911782, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1706789690317217, "standard_deviation": 0.013474830576808127, "minimum": 0.15344812365694815, "maximum": 0.1918181179797546, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9921829084536966, "standard_deviation": 0.0020621243396412694, "minimum": 0.9891066250503561, "maximum": 0.9950769004641488, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9729093673763032, "standard_deviation": 0.0459271612442794, "minimum": 0.9103600740106192, "maximum": 1.0545935555957529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.77015402656254, "standard_deviation": 0.42321792470314296, "minimum": 0.3269673776451964, "maximum": 1.4756268779069757, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09770350472466283, "standard_deviation": 0.013856231766463672, "minimum": 0.07992384666447734, "maximum": 0.11881253788701422, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9847366323907946, "standard_deviation": 0.0029216886361483796, "minimum": 0.9801652196946785, "maximum": 0.9874863518011414, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4197060109877282, "standard_deviation": 0.03655488180691671, "minimum": 1.3798454999046577, "maximum": 1.494864668626884, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0280338792892323, "standard_deviation": 0.2104418461993474, "minimum": 0.7076188791239396, "maximum": 1.3768898128566907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1438518424502186, "standard_deviation": 0.012336821993558528, "minimum": 0.1294592239273676, "maximum": 0.16263822553635193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9967614700028491, "standard_deviation": 0.000886973257660881, "minimum": 0.9952624199739835, "maximum": 0.9977814230810907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7986942497191107, "standard_deviation": 0.03158053408212332, "minimum": 0.7561936485946842, "maximum": 0.8324911215575412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2901957865354672, "standard_deviation": 0.22849820997069045, "minimum": 0.966081019824473, "maximum": 1.5865061664052644, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06835847360685042, "standard_deviation": 0.007794506716119015, "minimum": 0.05493366497295444, "maximum": 0.07694211035588637, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9899988402829532, "standard_deviation": 0.002820159531158626, "minimum": 0.986596519446588, "maximum": 0.9942253685725069, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.342558515803646, "standard_deviation": 0.07538903174069704, "minimum": 1.2176139067108027, "maximum": 1.4516997975947874, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.619706205876739, "standard_deviation": 0.5267243973941651, "minimum": 0.13308570431015548, "maximum": 1.3057906486615254, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.111294991696409, "standard_deviation": 0.016221974713562393, "minimum": 0.08473895909584421, "maximum": 0.13060237326730545, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9967210935751314, "standard_deviation": 0.0011021506605580122, "minimum": 0.994954443911858, "maximum": 0.9980057114860533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7698320759696774, "standard_deviation": 0.029595785616084256, "minimum": 0.7167984847967785, "maximum": 0.8129015839621531, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5843790157426073, "standard_deviation": 0.4223338963624706, "minimum": 0.1128994501281046, "maximum": 1.2611595790338956, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0630341267489804, "standard_deviation": 0.010901642682004602, "minimum": 0.050325020542526404, "maximum": 0.08066665277297275, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2082779243415872, "standard_deviation": 0.053893595914262606, "minimum": 0.11162490248815791, "maximum": 0.24891170444672536, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6619722213483215, "standard_deviation": 0.010097316091634776, "minimum": 5.650239488833155, "maximum": 5.676989954754364, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4245902257872075, "standard_deviation": 0.013336303786747727, "minimum": 0.40532492641254686, "maximum": 0.4450817235656047, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11007318646681469, "standard_deviation": 0.07832450364589438, "minimum": -0.22685114236382664, "maximum": -0.002691462405460432, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931798482055081, "standard_deviation": 0.001234382465927903, "minimum": 0.9917347733517119, "maximum": 0.9949148707863806, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8831742076567306, "standard_deviation": 0.037919120287635895, "minimum": 0.8019990227939667, "maximum": 0.9167878604155579, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.515555063269953, "standard_deviation": 0.1426184321631475, "minimum": 0.26013933603660794, "maximum": 0.7039506699249166, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.08627422125297067, "standard_deviation": 0.007889268167721506, "minimum": 0.07423505181109886, "maximum": 0.09620959352335184, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980444641462889, "standard_deviation": 0.0006349273498170648, "minimum": 0.9971727113449815, "maximum": 0.9986509078912076, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.437476758391169, "standard_deviation": 0.04153107830893663, "minimum": 0.377088257322434, "maximum": 0.4980492257134642, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.47307189484676415, "standard_deviation": 0.23021740540240274, "minimum": 0.20064537373202332, "maximum": 0.8239626660868602, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0457431190506586, "standard_deviation": 0.007333624037592405, "minimum": 0.03859226689703745, "maximum": 0.05602391788602675, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8555-284447-0007", "language": "en", "style": "normal", "speaker": "librispeech_8555", "source_utterance_id": "8555-284447-0007", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8555-284447-0007.wav", "source_file_sha256": "91d354d4aa509bec5988516fc33b4f1384fcd530112a1b2b5e79934c68555da5", "decoded_pcm_sha256": "b2e9b095c084b8ff6903b45eed724068bad7267cf8c1f692479d68f2b6adbaed", "original_source_sha256": "73d10187e60efe5f030a76fa33d1eab2abae7e38817b356dd61e30b743edf22f", "corpus_entry_id": "librispeech_8555-284447-0007"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.832388162612915, "predicted_mos": 3.832388162612915, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "29dbbe93727c4ac865cf172a2782845313b7658e5ea40b5c03de76939b47ae36", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.15777803599849, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8555-284447-0007", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "29dbbe93727c4ac865cf172a2782845313b7658e5ea40b5c03de76939b47ae36", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "93bf4a5143f192f1bec4014f5ed1882d65fcf3ffef7d30a7c6262721decfe4b4"}, "rows": [{"case": "librispeech_8555-284447-0007", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 3.05607008934021, "predicted_mos": 3.05607008934021, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d65ffeeed82a9fbbccea1a374406c4dc79e7efde6f6d49e739eb27e1b3571faf", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.099918375002744, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8555-284447-0007", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d65ffeeed82a9fbbccea1a374406c4dc79e7efde6f6d49e739eb27e1b3571faf", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6465420788328848, "全帯域_10ms_rms_envelope_nrmse": 0.051222659115566165, "全帯域_10ms_envelope_correlation": 0.9971633228040917, "全帯域_10ms_modulation_mae_db": 0.46685137727540726, "全帯域_40ms_log_rms_mae_db": 0.36694636089423815, "全帯域_40ms_rms_envelope_nrmse": 0.030498311876346187, "全帯域_40ms_envelope_correlation": 0.9990428166090031, "全帯域_40ms_modulation_mae_db": 0.4281157427234122, "300–1000 Hz_10ms_log_rms_mae_db": 1.0454812542353251, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06323217528626236, "300–1000 Hz_10ms_envelope_correlation": 0.9964903688437875, "300–1000 Hz_10ms_modulation_mae_db": 0.6620390110481893, "300–1000 Hz_40ms_log_rms_mae_db": 0.5834191089989853, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03979808065161025, "300–1000 Hz_40ms_envelope_correlation": 0.9987318984134331, "300–1000 Hz_40ms_modulation_mae_db": 0.7271213570602427, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3972264437390975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1315543679170226, "1000–3000 Hz_10ms_envelope_correlation": 0.9865642946318481, "1000–3000 Hz_10ms_modulation_mae_db": 0.719618102210911, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8157346849118638, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08001836299967628, "1000–3000 Hz_40ms_envelope_correlation": 0.9950177043591861, "1000–3000 Hz_40ms_modulation_mae_db": 0.7572608624931888, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5024563329985572, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17234367786857824, "3000–6000 Hz_10ms_envelope_correlation": 0.9787445316915236, "3000–6000 Hz_10ms_modulation_mae_db": 0.08843400600334395, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9315844055123333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1076359719731544, "3000–6000 Hz_40ms_envelope_correlation": 0.9908170303957525, "3000–6000 Hz_40ms_modulation_mae_db": 0.23915218065105345, "mr_stft_db": 5.411014492535955, "band_mae_db": 0.1624295027365605, "rms_gain_error_db": -0.1092278079026414, "output_peak": 0.2568853525188661}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7001259192556637, "全帯域_10ms_rms_envelope_nrmse": 0.0487999050352813, "全帯域_10ms_envelope_correlation": 0.9973326526294504, "全帯域_10ms_modulation_mae_db": 0.3230492661331857, "全帯域_40ms_log_rms_mae_db": 0.409556964734469, "全帯域_40ms_rms_envelope_nrmse": 0.027281849175493995, "全帯域_40ms_envelope_correlation": 0.9991501773639022, "全帯域_40ms_modulation_mae_db": 0.3492836874394489, "300–1000 Hz_10ms_log_rms_mae_db": 1.0398883937158294, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06518069759436775, "300–1000 Hz_10ms_envelope_correlation": 0.9961035750436739, "300–1000 Hz_10ms_modulation_mae_db": 0.49185114050361395, "300–1000 Hz_40ms_log_rms_mae_db": 0.5754208981798821, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03908157264343196, "300–1000 Hz_40ms_envelope_correlation": 0.9986360190020335, "300–1000 Hz_40ms_modulation_mae_db": 0.5860366401069655, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4015905749893993, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12592585623000038, "1000–3000 Hz_10ms_envelope_correlation": 0.9879127953740721, "1000–3000 Hz_10ms_modulation_mae_db": 0.8184852898047811, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8493852112130318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07483901510934528, "1000–3000 Hz_40ms_envelope_correlation": 0.995953124684916, "1000–3000 Hz_40ms_modulation_mae_db": 0.6896374241663444, "3000–6000 Hz_10ms_log_rms_mae_db": 1.548148084587727, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19920935926742267, "3000–6000 Hz_10ms_envelope_correlation": 0.9714281122983915, "3000–6000 Hz_10ms_modulation_mae_db": 0.8657816991712267, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9272646161511323, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10957966799275647, "3000–6000 Hz_40ms_envelope_correlation": 0.9904975812338584, "3000–6000 Hz_40ms_modulation_mae_db": 0.663176078524771, "mr_stft_db": 5.4378931502627665, "band_mae_db": 0.1224553254326735, "rms_gain_error_db": -0.06680514643037674, "output_peak": 0.24621310883540196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7175625550811573, "全帯域_10ms_rms_envelope_nrmse": 0.047672620006097045, "全帯域_10ms_envelope_correlation": 0.9975091778081645, "全帯域_10ms_modulation_mae_db": 0.557509956066849, "全帯域_40ms_log_rms_mae_db": 0.3813619713831973, "全帯域_40ms_rms_envelope_nrmse": 0.026568298628479228, "全帯域_40ms_envelope_correlation": 0.9992592208784591, "全帯域_40ms_modulation_mae_db": 0.6448083038409702, "300–1000 Hz_10ms_log_rms_mae_db": 1.0469410088292628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06014124013643179, "300–1000 Hz_10ms_envelope_correlation": 0.9967291847303955, "300–1000 Hz_10ms_modulation_mae_db": 0.47392904184573714, "300–1000 Hz_40ms_log_rms_mae_db": 0.556778488681733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03727536512522939, "300–1000 Hz_40ms_envelope_correlation": 0.9987964887492681, "300–1000 Hz_40ms_modulation_mae_db": 0.6148317693836048, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4006161340191277, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12512912687718925, "1000–3000 Hz_10ms_envelope_correlation": 0.9877321576708568, "1000–3000 Hz_10ms_modulation_mae_db": 0.5277075876631995, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8524906702747417, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07927671502005493, "1000–3000 Hz_40ms_envelope_correlation": 0.994865793534807, "1000–3000 Hz_40ms_modulation_mae_db": 0.513691716006191, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5257249283485532, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19771944932228372, "3000–6000 Hz_10ms_envelope_correlation": 0.9718179381864975, "3000–6000 Hz_10ms_modulation_mae_db": 0.8118409035326525, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9528738961184506, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1210517509196408, "3000–6000 Hz_40ms_envelope_correlation": 0.9885838406132661, "3000–6000 Hz_40ms_modulation_mae_db": 0.8530085981201014, "mr_stft_db": 5.418345991136474, "band_mae_db": 0.21776271790595134, "rms_gain_error_db": -0.0933685510509844, "output_peak": 0.24136599288207372}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.70427867437316, "全帯域_10ms_rms_envelope_nrmse": 0.04851556850465897, "全帯域_10ms_envelope_correlation": 0.9973663982991946, "全帯域_10ms_modulation_mae_db": 0.33121243194257916, "全帯域_40ms_log_rms_mae_db": 0.3657851197884685, "全帯域_40ms_rms_envelope_nrmse": 0.025365574995818936, "全帯域_40ms_envelope_correlation": 0.9992802642934523, "全帯域_40ms_modulation_mae_db": 0.43987170322767016, "300–1000 Hz_10ms_log_rms_mae_db": 1.0188509272163775, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06099948991305552, "300–1000 Hz_10ms_envelope_correlation": 0.9965739577916406, "300–1000 Hz_10ms_modulation_mae_db": 0.49184232778467485, "300–1000 Hz_40ms_log_rms_mae_db": 0.5665442200234836, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03594189220329102, "300–1000 Hz_40ms_envelope_correlation": 0.99883383450096, "300–1000 Hz_40ms_modulation_mae_db": 0.5481830994230604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4509023657405906, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13822289645934038, "1000–3000 Hz_10ms_envelope_correlation": 0.984903600130897, "1000–3000 Hz_10ms_modulation_mae_db": 0.6977260270183147, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8732990262771508, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07679192720726648, "1000–3000 Hz_40ms_envelope_correlation": 0.995093410636407, "1000–3000 Hz_40ms_modulation_mae_db": 0.5055280957680565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5588821343958847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19610605181096286, "3000–6000 Hz_10ms_envelope_correlation": 0.9723202174321882, "3000–6000 Hz_10ms_modulation_mae_db": 1.0120410931116477, "3000–6000 Hz_40ms_log_rms_mae_db": 0.978817912909497, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11934670349194343, "3000–6000 Hz_40ms_envelope_correlation": 0.9889272943237628, "3000–6000 Hz_40ms_modulation_mae_db": 1.3504582692470646, "mr_stft_db": 5.428404059198868, "band_mae_db": 0.19432732915086834, "rms_gain_error_db": -0.07253891858575558, "output_peak": 0.24355496246650982}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6942641959876972, "全帯域_10ms_rms_envelope_nrmse": 0.048064475647828175, "全帯域_10ms_envelope_correlation": 0.9974256179714295, "全帯域_10ms_modulation_mae_db": 0.511233736765237, "全帯域_40ms_log_rms_mae_db": 0.37584431526222395, "全帯域_40ms_rms_envelope_nrmse": 0.02705633456677672, "全帯域_40ms_envelope_correlation": 0.99917288396756, "全帯域_40ms_modulation_mae_db": 0.40545780925282787, "300–1000 Hz_10ms_log_rms_mae_db": 1.0431844298603332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06331718983745681, "300–1000 Hz_10ms_envelope_correlation": 0.9962814004749212, "300–1000 Hz_10ms_modulation_mae_db": 0.6388633381543835, "300–1000 Hz_40ms_log_rms_mae_db": 0.5691570185613419, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04024068319129186, "300–1000 Hz_40ms_envelope_correlation": 0.998475384366021, "300–1000 Hz_40ms_modulation_mae_db": 0.6570423331216783, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4402065605458185, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13598930632153064, "1000–3000 Hz_10ms_envelope_correlation": 0.9855186120735794, "1000–3000 Hz_10ms_modulation_mae_db": 1.2165396901509753, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8256984813945931, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07270360365389122, "1000–3000 Hz_40ms_envelope_correlation": 0.9958547963309687, "1000–3000 Hz_40ms_modulation_mae_db": 1.0544940310480837, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5639230540024043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18330727290854232, "3000–6000 Hz_10ms_envelope_correlation": 0.976580029735115, "3000–6000 Hz_10ms_modulation_mae_db": 0.11596262300688942, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9791373369664327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10876233878424216, "3000–6000 Hz_40ms_envelope_correlation": 0.9909892030868949, "3000–6000 Hz_40ms_modulation_mae_db": 0.19493259253010264, "mr_stft_db": 5.449342606004431, "band_mae_db": 0.17930415747559622, "rms_gain_error_db": -0.0727289658160376, "output_peak": 0.2524024138892328}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7140270557041154, "全帯域_10ms_rms_envelope_nrmse": 0.049590233587049046, "全帯域_10ms_envelope_correlation": 0.997294535067988, "全帯域_10ms_modulation_mae_db": 0.413437149525491, "全帯域_40ms_log_rms_mae_db": 0.3855345837302965, "全帯域_40ms_rms_envelope_nrmse": 0.02960421877748115, "全帯域_40ms_envelope_correlation": 0.9990359192736082, "全帯域_40ms_modulation_mae_db": 0.5951756207482691, "300–1000 Hz_10ms_log_rms_mae_db": 1.0374999719465814, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05871405498261709, "300–1000 Hz_10ms_envelope_correlation": 0.9968806694268976, "300–1000 Hz_10ms_modulation_mae_db": 0.459154284293223, "300–1000 Hz_40ms_log_rms_mae_db": 0.6333130224906252, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.038316584041490034, "300–1000 Hz_40ms_envelope_correlation": 0.9986940466528308, "300–1000 Hz_40ms_modulation_mae_db": 0.633924536546851, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4641307121641522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14874587446551987, "1000–3000 Hz_10ms_envelope_correlation": 0.9826986071155359, "1000–3000 Hz_10ms_modulation_mae_db": 0.5574946319492075, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8694004533444553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0911811187028091, "1000–3000 Hz_40ms_envelope_correlation": 0.9934049543016048, "1000–3000 Hz_40ms_modulation_mae_db": 0.4991826718144923, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5717515565424143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1907116202179954, "3000–6000 Hz_10ms_envelope_correlation": 0.9739799263244943, "3000–6000 Hz_10ms_modulation_mae_db": 0.6193325155146918, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9913458715304171, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1069900951689418, "3000–6000 Hz_40ms_envelope_correlation": 0.9909293518194615, "3000–6000 Hz_40ms_modulation_mae_db": 0.6747304541107252, "mr_stft_db": 5.447461261732573, "band_mae_db": 0.1506186962179971, "rms_gain_error_db": -0.08861620524215388, "output_peak": 0.2586837193680693}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.723985819018738, "全帯域_10ms_rms_envelope_nrmse": 0.04945184132181581, "全帯域_10ms_envelope_correlation": 0.9973394554721512, "全帯域_10ms_modulation_mae_db": 0.6323732818345053, "全帯域_40ms_log_rms_mae_db": 0.4045454513110299, "全帯域_40ms_rms_envelope_nrmse": 0.027430246792150528, "全帯域_40ms_envelope_correlation": 0.9992277630137013, "全帯域_40ms_modulation_mae_db": 0.6434881890200236, "300–1000 Hz_10ms_log_rms_mae_db": 1.0650681799141772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06106832397152481, "300–1000 Hz_10ms_envelope_correlation": 0.9966481156686269, "300–1000 Hz_10ms_modulation_mae_db": 0.8651548407763651, "300–1000 Hz_40ms_log_rms_mae_db": 0.5904588806437804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03591147581095481, "300–1000 Hz_40ms_envelope_correlation": 0.9989236671671824, "300–1000 Hz_40ms_modulation_mae_db": 0.7292804506606663, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4236181569683781, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1324379927513283, "1000–3000 Hz_10ms_envelope_correlation": 0.9865079328867042, "1000–3000 Hz_10ms_modulation_mae_db": 0.7501661190400076, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8259499164867601, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07629963073827303, "1000–3000 Hz_40ms_envelope_correlation": 0.9957156426800007, "1000–3000 Hz_40ms_modulation_mae_db": 0.4642130632910891, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5671414529011776, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18092696187212068, "3000–6000 Hz_10ms_envelope_correlation": 0.9776583702343679, "3000–6000 Hz_10ms_modulation_mae_db": 1.3238370104317463, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9753605932844363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10803600383887803, "3000–6000 Hz_40ms_envelope_correlation": 0.9928040720053457, "3000–6000 Hz_40ms_modulation_mae_db": 1.4885629053600191, "mr_stft_db": 5.437546980713191, "band_mae_db": 0.20999244855631316, "rms_gain_error_db": -0.0982220355145032, "output_peak": 0.24342292929612303}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7106529690074828, "全帯域_10ms_rms_envelope_nrmse": 0.051639240700148144, "全帯域_10ms_envelope_correlation": 0.997041571391535, "全帯域_10ms_modulation_mae_db": 0.643113182269936, "全帯域_40ms_log_rms_mae_db": 0.38917853566072047, "全帯域_40ms_rms_envelope_nrmse": 0.03145229434758499, "全帯域_40ms_envelope_correlation": 0.9988940652437154, "全帯域_40ms_modulation_mae_db": 0.8178746643357256, "300–1000 Hz_10ms_log_rms_mae_db": 1.022836887224212, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06264403452955493, "300–1000 Hz_10ms_envelope_correlation": 0.9963836458631247, "300–1000 Hz_10ms_modulation_mae_db": 0.5272621864006785, "300–1000 Hz_40ms_log_rms_mae_db": 0.5638642416552027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04123673389216009, "300–1000 Hz_40ms_envelope_correlation": 0.998426982661142, "300–1000 Hz_40ms_modulation_mae_db": 0.8461166155665275, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3816521733474536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13221257843023299, "1000–3000 Hz_10ms_envelope_correlation": 0.986325532471352, "1000–3000 Hz_10ms_modulation_mae_db": 0.5471996748639681, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7957686009265127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06734299563443742, "1000–3000 Hz_40ms_envelope_correlation": 0.9965020265694567, "1000–3000 Hz_40ms_modulation_mae_db": 0.6972580237130614, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5595663461635059, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18296273457940304, "3000–6000 Hz_10ms_envelope_correlation": 0.9759588515255577, "3000–6000 Hz_10ms_modulation_mae_db": 1.2165553976412082, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9486753629084016, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10788454425712515, "3000–6000 Hz_40ms_envelope_correlation": 0.9909879572126093, "3000–6000 Hz_40ms_modulation_mae_db": 1.3704743426164985, "mr_stft_db": 5.428194278570385, "band_mae_db": 0.14108751863197652, "rms_gain_error_db": -0.09115639799356436, "output_peak": 0.2578921118912043}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.294510444594626, "spectral_convergence": 0.13912295047929518, "spectral_convergence_status": "ok", "amplitude_floor": 9.671435620644006e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.347483765827426, "spectral_convergence": 0.1343213515046306, "spectral_convergence_status": "ok", "amplitude_floor": 9.322400769888516e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.591049267185815, "spectral_convergence": 0.1368477484194902, "spectral_convergence_status": "ok", "amplitude_floor": 9.453157582642776e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00016624637287256398, "output_power": 0.00017377893010995154, "delta_db": 0.1924494037562393, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013640231584289317, "output_power": 0.0013335988075405222, "delta_db": -0.09796545360183828, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005975940179232537, "output_power": 0.0005681741412044989, "delta_db": -0.21924777239638532, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.8168186983374496e-05, "output_power": 1.8300028743167125e-05, "delta_db": 0.03140180953689394, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.463537251271816e-05, "output_power": 2.2933376127743594e-05, "delta_db": -0.31091139725542116, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0343523773544458e-05, "output_power": 1.0055609485827826e-05, "delta_db": -0.12260117987258497, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1624295027365605, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1092278079026414, "pitch": null, "output_rms_dbfs": -26.72264998007352, "output_peak": 0.2568853525188661, "reference_rms_dbfs": -26.613422172170882, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.411014492535955}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001446111949915804, "log_rms_mae_db": 0.6465420788328848, "rms_envelope_nrmse": 0.051222659115566165, "envelope_correlation": 0.9971633228040917, "rms_gain_error_db": -0.10922964933242407, "reference_mean_power": 0.002182828397481346, "output_mean_power": 0.002128612607845538}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.790920693994897e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003230276115117931, "output_power": 0.0003044542582208458, "reference_active": true, "delta_db": -0.25717593563658947}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.477236494661316e-05, "output_power": 4.1893596334157815e-05, "reference_active": true, "delta_db": -0.2886239135854172}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.548345752003965e-06, "output_power": 7.842467154035802e-06, "reference_active": true, "delta_db": -0.8547542826042152}], "active_band_count": 3, "active_band_mae_db": 0.46685137727540726}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014111493703772434, "log_rms_mae_db": 0.36694636089423815, "rms_envelope_nrmse": 0.030498311876346187, "envelope_correlation": 0.9990428166090031, "rms_gain_error_db": -0.10923509304745577, "reference_mean_power": 0.002193693032291648, "output_mean_power": 0.002139204711924521}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.190420923463322e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028912450266098784, "output_power": 0.0002729527543154183, "reference_active": true, "delta_db": -0.2499741767657409}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2978791346744758e-05, "output_power": 2.1659800183941586e-05, "reference_active": true, "delta_db": -0.25672735799431073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.072854493661425e-07, "output_power": 5.07724716954055e-07, "reference_active": true, "delta_db": -0.777645693410185}], "active_band_count": 3, "active_band_mae_db": 0.4281157427234122}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012712872255296568, "log_rms_mae_db": 1.0454812542353251, "rms_envelope_nrmse": 0.06323217528626236, "envelope_correlation": 0.9964903688437875, "rms_gain_error_db": -0.16580875277010726, "reference_mean_power": 0.0014813690704446063, "output_mean_power": 0.0014258780958251096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.409297669270013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003163260619212235, "output_power": 0.00029731004471441865, "reference_active": true, "delta_db": -0.26925392430893363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6701975061022124e-05, "output_power": 3.2446362298882654e-05, "reference_active": true, "delta_db": -0.535234225337427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.301526036454895e-06, "output_power": 4.800459904753393e-06, "reference_active": true, "delta_db": -1.1816288834982072}], "active_band_count": 3, "active_band_mae_db": 0.6620390110481893}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012343588785799389, "log_rms_mae_db": 0.5834191089989853, "rms_envelope_nrmse": 0.03979808065161025, "envelope_correlation": 0.9987318984134331, "rms_gain_error_db": -0.16581967449554866, "reference_mean_power": 0.0014887936512772218, "output_mean_power": 0.0014330209536322306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.99328908394483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002875949770086079, "output_power": 0.0002713910558439919, "reference_active": true, "delta_db": -0.2518576596203763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97680708793848e-05, "output_power": 1.776293653552111e-05, "reference_active": true, "delta_db": -0.46449525449704665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.520482053574084e-07, "output_power": 3.2261431483542233e-07, "reference_active": true, "delta_db": -1.4650111570633055}], "active_band_count": 3, "active_band_mae_db": 0.7271213570602427}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.4084091617788044e-05, "log_rms_mae_db": 1.3972264437390975, "rms_envelope_nrmse": 0.1315543679170226, "envelope_correlation": 0.9865642946318481, "rms_gain_error_db": -0.21343452184833, "reference_mean_power": 3.6694438258439554e-05, "output_mean_power": 3.493468176618313e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6589829493109366e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3179349061495639e-05, "output_power": 1.1619475271132615e-05, "reference_active": true, "delta_db": -0.5470744463237327}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9890642401085437e-06, "output_power": 2.9503253869617016e-06, "reference_active": true, "delta_db": -0.056653325649725444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.136200949313584e-07, "output_power": 5.687337654372453e-07, "reference_active": true, "delta_db": -1.555126534659275}], "active_band_count": 3, "active_band_mae_db": 0.719618102210911}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.151462172211685e-05, "log_rms_mae_db": 0.8157346849118638, "rms_envelope_nrmse": 0.08001836299967628, "envelope_correlation": 0.9950177043591861, "rms_gain_error_db": -0.21358702194652135, "reference_mean_power": 3.687647047180041e-05, "output_mean_power": 3.5106751478408376e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4709890047082147e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2743108138238886e-05, "output_power": 1.1162590503571434e-05, "reference_active": true, "delta_db": -0.575103754880038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4131557808728274e-06, "output_power": 1.5050585923170645e-06, "reference_active": true, "delta_db": 0.2736336800916798}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.7131268313465536e-08, "output_power": 4.116890825633987e-08, "reference_active": true, "delta_db": -1.4230451525078487}], "active_band_count": 3, "active_band_mae_db": 0.7572608624931888}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.703026630874025e-05, "log_rms_mae_db": 1.5024563329985572, "rms_envelope_nrmse": 0.17234367786857824, "envelope_correlation": 0.9787445316915236, "rms_gain_error_db": -0.034099211195405696, "reference_mean_power": 1.1971789360385644e-05, "output_mean_power": 1.1878159312951834e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.043568625201122e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.248382965898838e-06, "output_power": 4.148040807734691e-06, "reference_active": true, "delta_db": -0.10380638590738403}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.589400168019184e-06, "output_power": 2.5329310827920192e-06, "reference_active": true, "delta_db": -0.09575798387824645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.486231293405906e-07, "output_power": 6.585157945404209e-07, "reference_active": true, "delta_db": 0.06573764822440134}], "active_band_count": 3, "active_band_mae_db": 0.08843400600334395}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.3691778537856407e-05, "log_rms_mae_db": 0.9315844055123333, "rms_envelope_nrmse": 0.1076359719731544, "envelope_correlation": 0.9908170303957525, "rms_gain_error_db": -0.03409454555522056, "reference_mean_power": 1.2031928101806957e-05, "output_mean_power": 1.1937840540748287e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.457061965350652e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.507382710758712e-06, "output_power": 4.436451826165293e-06, "reference_active": true, "delta_db": -0.06888664477808785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3712782786962106e-06, "output_power": 1.3845621096865834e-06, "reference_active": true, "delta_db": 0.0418684565609133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.274779912953651e-08, "output_power": 4.915683401681015e-08, "reference_active": true, "delta_db": 0.6067014406141592}], "active_band_count": 3, "active_band_mae_db": 0.23915218065105345}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9435416389966792, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5ebd7dace12e37824a4a7850ad28a75d83d07445fe7a8da9897de5dc079c821f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9860204415443556, "standard_deviation": 0.0016763579191206633, "minimum": 0.9826986071155359, "maximum": 0.9879127953740721, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4199928901892522, "standard_deviation": 0.029344514143806587, "minimum": 1.3816521733474536, "maximum": 1.4641307121641522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7293671403376707, "standard_deviation": 0.22367056737510427, "minimum": 0.5277075876631995, "maximum": 1.2165396901509753, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13377724993152057, "standard_deviation": 0.007505560924445012, "minimum": 0.12512912687718925, "maximum": 0.14874587446551987, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953009316371684, "standard_deviation": 0.0009452206120290855, "minimum": 0.9934049543016048, "maximum": 0.9965020265694567, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8384658806036387, "standard_deviation": 0.02711257779857772, "minimum": 0.7957686009265127, "maximum": 0.8732990262771508, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6476582360375635, "standard_deviation": 0.19842077254407342, "minimum": 0.4642130632910891, "maximum": 1.0544940310480837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07730667113321922, "standard_deviation": 0.006882338874028614, "minimum": 0.06734299563443742, "maximum": 0.0911811187028091, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9748109971785169, "standard_deviation": 0.0028113100053569775, "minimum": 0.9714281122983915, "maximum": 0.9787445316915236, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5496992362425281, "standard_deviation": 0.023855528374842127, "minimum": 1.5024563329985572, "maximum": 1.5717515565424143, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7567231560516758, "standard_deviation": 0.4612472891165699, "minimum": 0.08843400600334395, "maximum": 1.3238370104317463, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18791089098091363, "standard_deviation": 0.009529344818272178, "minimum": 0.17234367786857824, "maximum": 0.19920935926742267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905670413363689, "standard_deviation": 0.0013190981999456185, "minimum": 0.9885838406132661, "maximum": 0.9928040720053457, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9606324994226376, "standard_deviation": 0.02388586756083919, "minimum": 0.9272646161511323, "maximum": 0.9913458715304171, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8543119276450419, "standard_deviation": 0.5066480706803015, "minimum": 0.19493259253010264, "maximum": 1.4885629053600191, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11116088455333528, "standard_deviation": 0.0056493983850712136, "minimum": 0.1069900951689418, "maximum": 0.1210517509196408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9965113647303835, "standard_deviation": 0.00025139017635809224, "minimum": 0.9961035750436739, "maximum": 0.9968806694268976, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0399688816177624, "standard_deviation": 0.014466150433079174, "minimum": 1.0188509272163775, "maximum": 1.0650681799141772, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5762620213508581, "standard_deviation": 0.13916834821190305, "minimum": 0.459154284293223, "maximum": 0.8651548407763651, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06191215078140888, "standard_deviation": 0.0020648274340965135, "minimum": 0.05871405498261709, "maximum": 0.06518069759436775, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986897901891088, "standard_deviation": 0.0001717772651339009, "minimum": 0.998426982661142, "maximum": 0.9989236671671824, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5798694849043793, "standard_deviation": 0.024149588296499033, "minimum": 0.556778488681733, "maximum": 0.6333130224906252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6678171002336996, "standard_deviation": 0.09573627470799871, "minimum": 0.5481830994230604, "maximum": 0.8461166155665275, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03847529844493243, "standard_deviation": 0.001975412944866788, "minimum": 0.03591147581095481, "maximum": 0.04123673389216009, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17224721201349208, "standard_deviation": 0.03390751029796216, "minimum": 0.1224553254326735, "maximum": 0.21776271790595134, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.43227535251933, "standard_deviation": 0.013410795479627233, "minimum": 5.411014492535955, "maximum": 5.449342606004431, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2500525738934351, "standard_deviation": 0.007216379813744591, "minimum": 0.24136599288207372, "maximum": 0.2586837193680693, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08658300356700215, "standard_deviation": 0.01463694955654966, "minimum": -0.1092278079026414, "maximum": -0.06680514643037674, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973090914305006, "standard_deviation": 0.00014706729604951077, "minimum": 0.997041571391535, "maximum": 0.9975091778081645, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7014299084076124, "standard_deviation": 0.02416201707572484, "minimum": 0.6465420788328848, "maximum": 0.723985819018738, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4848475477266488, "standard_deviation": 0.12420814886424064, "minimum": 0.3230492661331857, "maximum": 0.643113182269936, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04936956798980558, "standard_deviation": 0.0014281632870979926, "minimum": 0.047672620006097045, "maximum": 0.051639240700148144, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991328888304252, "standard_deviation": 0.00013262823125970735, "minimum": 0.9988940652437154, "maximum": 0.9992802642934523, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.38484416284558043, "standard_deviation": 0.01601262075428381, "minimum": 0.3657851197884685, "maximum": 0.409556964734469, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5405094650735435, "standard_deviation": 0.15989695940851975, "minimum": 0.3492836874394489, "maximum": 0.8178746643357256, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028157141145016468, "standard_deviation": 0.00211311203673461, "minimum": 0.025365574995818936, "maximum": 0.03145229434758499, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284447-0007", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 3.164064407348633, "predicted_mos": 3.164064407348633, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7885f129b1f7f70a9e1a9520b57e622db99f538cc14ee01b0ac7462bcb36849a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.75492210034281, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6569237960574047, "全帯域_10ms_rms_envelope_nrmse": 0.05105657569767949, "全帯域_10ms_envelope_correlation": 0.9971999973777944, "全帯域_10ms_modulation_mae_db": 0.3848826558450071, "全帯域_40ms_log_rms_mae_db": 0.37175083978117573, "全帯域_40ms_rms_envelope_nrmse": 0.02980872306090521, "全帯域_40ms_envelope_correlation": 0.9991158508303405, "全帯域_40ms_modulation_mae_db": 0.33878055258488177, "300–1000 Hz_10ms_log_rms_mae_db": 1.0569978715202102, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06501187509995847, "300–1000 Hz_10ms_envelope_correlation": 0.996262388145528, "300–1000 Hz_10ms_modulation_mae_db": 0.5660734474471497, "300–1000 Hz_40ms_log_rms_mae_db": 0.6010646582971552, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.039572444808142676, "300–1000 Hz_40ms_envelope_correlation": 0.9987415560314927, "300–1000 Hz_40ms_modulation_mae_db": 0.6343542872934631, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3572689061243333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12401058247342837, "1000–3000 Hz_10ms_envelope_correlation": 0.9882702226981721, "1000–3000 Hz_10ms_modulation_mae_db": 0.9722213761935548, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7944652343077884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07031725091837979, "1000–3000 Hz_40ms_envelope_correlation": 0.9964910071400049, "1000–3000 Hz_40ms_modulation_mae_db": 0.8914230473123247, "3000–6000 Hz_10ms_log_rms_mae_db": 1.457212279290237, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16326983541778708, "3000–6000 Hz_10ms_envelope_correlation": 0.9809150970935498, "3000–6000 Hz_10ms_modulation_mae_db": 0.18720257196957465, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9227947309638281, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10251214700141707, "3000–6000 Hz_40ms_envelope_correlation": 0.9918406051966874, "3000–6000 Hz_40ms_modulation_mae_db": 0.12490728206979822, "mr_stft_db": 5.3109380639040475, "band_mae_db": 0.19058231893093844, "rms_gain_error_db": -0.1150341097731428, "output_peak": 0.25394118274332667}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.704713631490058, "全帯域_10ms_rms_envelope_nrmse": 0.0485150522711774, "全帯域_10ms_envelope_correlation": 0.9973778385970762, "全帯域_10ms_modulation_mae_db": 0.3885244024887007, "全帯域_40ms_log_rms_mae_db": 0.41632520054866756, "全帯域_40ms_rms_envelope_nrmse": 0.02755006301860136, "全帯域_40ms_envelope_correlation": 0.9991437772870607, "全帯域_40ms_modulation_mae_db": 0.4354795890040173, "300–1000 Hz_10ms_log_rms_mae_db": 1.0744739426950822, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06563504918420247, "300–1000 Hz_10ms_envelope_correlation": 0.9960600924959736, "300–1000 Hz_10ms_modulation_mae_db": 0.5530433932214555, "300–1000 Hz_40ms_log_rms_mae_db": 0.6024577922291124, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03990028610163465, "300–1000 Hz_40ms_envelope_correlation": 0.9985798714979597, "300–1000 Hz_40ms_modulation_mae_db": 0.6372732676633921, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3769609603030908, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12366614031182128, "1000–3000 Hz_10ms_envelope_correlation": 0.9885400426854379, "1000–3000 Hz_10ms_modulation_mae_db": 1.0071531913672465, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8289547724243239, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07369361987294443, "1000–3000 Hz_40ms_envelope_correlation": 0.9963220957400284, "1000–3000 Hz_40ms_modulation_mae_db": 0.9080746165594711, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4758916409444278, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1985985362806428, "3000–6000 Hz_10ms_envelope_correlation": 0.9716883405303648, "3000–6000 Hz_10ms_modulation_mae_db": 0.8986439618134549, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9266661364554181, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10827288467707527, "3000–6000 Hz_40ms_envelope_correlation": 0.9908637602150371, "3000–6000 Hz_40ms_modulation_mae_db": 0.640812306029267, "mr_stft_db": 5.334302869496514, "band_mae_db": 0.14569345254342828, "rms_gain_error_db": -0.0713914965070877, "output_peak": 0.24278157820863944}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7065607341986655, "全帯域_10ms_rms_envelope_nrmse": 0.04903287525024672, "全帯域_10ms_envelope_correlation": 0.9973681464670157, "全帯域_10ms_modulation_mae_db": 0.4803949895504838, "全帯域_40ms_log_rms_mae_db": 0.3885125620791271, "全帯域_40ms_rms_envelope_nrmse": 0.027414218275329787, "全帯域_40ms_envelope_correlation": 0.9992181697529119, "全帯域_40ms_modulation_mae_db": 0.5501122699490237, "300–1000 Hz_10ms_log_rms_mae_db": 1.0372263262298056, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06207804813813945, "300–1000 Hz_10ms_envelope_correlation": 0.9965192083905856, "300–1000 Hz_10ms_modulation_mae_db": 0.4213739095997265, "300–1000 Hz_40ms_log_rms_mae_db": 0.5547473121333006, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03752630927030991, "300–1000 Hz_40ms_envelope_correlation": 0.998800147268599, "300–1000 Hz_40ms_modulation_mae_db": 0.5650152661613471, "1000–3000 Hz_10ms_log_rms_mae_db": 1.31445007306559, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11826577465501195, "1000–3000 Hz_10ms_envelope_correlation": 0.9892799271867235, "1000–3000 Hz_10ms_modulation_mae_db": 0.6073052013241387, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7563288048699218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0719658041875105, "1000–3000 Hz_40ms_envelope_correlation": 0.9961242484625245, "1000–3000 Hz_40ms_modulation_mae_db": 0.585449171710056, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4733107030809267, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18989858595289386, "3000–6000 Hz_10ms_envelope_correlation": 0.974169352982278, "3000–6000 Hz_10ms_modulation_mae_db": 0.6941261126510027, "3000–6000 Hz_40ms_log_rms_mae_db": 0.970530434433304, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11324183717463585, "3000–6000 Hz_40ms_envelope_correlation": 0.9902216636957102, "3000–6000 Hz_40ms_modulation_mae_db": 0.6432084366827721, "mr_stft_db": 5.3188827538790315, "band_mae_db": 0.23475357701271848, "rms_gain_error_db": -0.09916681926271513, "output_peak": 0.24301556688024475}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7199499319366507, "全帯域_10ms_rms_envelope_nrmse": 0.050278314722259854, "全帯域_10ms_envelope_correlation": 0.9971713172787119, "全帯域_10ms_modulation_mae_db": 0.4516268024270296, "全帯域_40ms_log_rms_mae_db": 0.36880131662995613, "全帯域_40ms_rms_envelope_nrmse": 0.027338015968297758, "全帯域_40ms_envelope_correlation": 0.9991560208716189, "全帯域_40ms_modulation_mae_db": 0.541906167924819, "300–1000 Hz_10ms_log_rms_mae_db": 1.0550496775271903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06370396114039714, "300–1000 Hz_10ms_envelope_correlation": 0.9962542157821589, "300–1000 Hz_10ms_modulation_mae_db": 0.6140914927937368, "300–1000 Hz_40ms_log_rms_mae_db": 0.5748414555247106, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03734960737974151, "300–1000 Hz_40ms_envelope_correlation": 0.998727862712523, "300–1000 Hz_40ms_modulation_mae_db": 0.6993700220470652, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3660917844283886, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13354079519727927, "1000–3000 Hz_10ms_envelope_correlation": 0.9859514948365155, "1000–3000 Hz_10ms_modulation_mae_db": 0.7839336987327429, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8015885448490704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07245849152123571, "1000–3000 Hz_40ms_envelope_correlation": 0.9957491824394598, "1000–3000 Hz_40ms_modulation_mae_db": 0.5971371644292006, "3000–6000 Hz_10ms_log_rms_mae_db": 1.464079729357112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19322627339182152, "3000–6000 Hz_10ms_envelope_correlation": 0.9731513158367667, "3000–6000 Hz_10ms_modulation_mae_db": 0.8759757911104286, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9194608123812077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11445501584857411, "3000–6000 Hz_40ms_envelope_correlation": 0.989909136450297, "3000–6000 Hz_40ms_modulation_mae_db": 1.1595069415042232, "mr_stft_db": 5.31392866385801, "band_mae_db": 0.19650877326626104, "rms_gain_error_db": -0.07521633791083951, "output_peak": 0.246520416587255}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.702291746569604, "全帯域_10ms_rms_envelope_nrmse": 0.049648731650430995, "全帯域_10ms_envelope_correlation": 0.9972715557347732, "全帯域_10ms_modulation_mae_db": 0.4720770782197943, "全帯域_40ms_log_rms_mae_db": 0.39477062427192944, "全帯域_40ms_rms_envelope_nrmse": 0.028363057713451822, "全帯域_40ms_envelope_correlation": 0.9991107091563318, "全帯域_40ms_modulation_mae_db": 0.4511374364755385, "300–1000 Hz_10ms_log_rms_mae_db": 1.0619231246751084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06419521525582635, "300–1000 Hz_10ms_envelope_correlation": 0.9962046625071391, "300–1000 Hz_10ms_modulation_mae_db": 0.6134242544832716, "300–1000 Hz_40ms_log_rms_mae_db": 0.5926714347306163, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0406630377233521, "300–1000 Hz_40ms_envelope_correlation": 0.9984766478983051, "300–1000 Hz_40ms_modulation_mae_db": 0.7611647650150486, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4078231812157194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13564142042807703, "1000–3000 Hz_10ms_envelope_correlation": 0.985765619138798, "1000–3000 Hz_10ms_modulation_mae_db": 1.2145232255081646, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8264320772480629, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07388714269790136, "1000–3000 Hz_40ms_envelope_correlation": 0.9959820122307044, "1000–3000 Hz_40ms_modulation_mae_db": 1.0189341032030972, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4397245829303975, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17192816814529033, "3000–6000 Hz_10ms_envelope_correlation": 0.9789889402471047, "3000–6000 Hz_10ms_modulation_mae_db": 0.2951455010115606, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9206727013849798, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10165968447779751, "3000–6000 Hz_40ms_envelope_correlation": 0.991903506664096, "3000–6000 Hz_40ms_modulation_mae_db": 0.41192138872793643, "mr_stft_db": 5.341019117317248, "band_mae_db": 0.19790615756711352, "rms_gain_error_db": -0.08440786536178471, "output_peak": 0.2527440140424124}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7219684293086782, "全帯域_10ms_rms_envelope_nrmse": 0.04954165732614243, "全帯域_10ms_envelope_correlation": 0.9972932695589061, "全帯域_10ms_modulation_mae_db": 0.39252202976283174, "全帯域_40ms_log_rms_mae_db": 0.38457365371394214, "全帯域_40ms_rms_envelope_nrmse": 0.02864308254271797, "全帯域_40ms_envelope_correlation": 0.9990986294581394, "全帯域_40ms_modulation_mae_db": 0.5115233891775465, "300–1000 Hz_10ms_log_rms_mae_db": 1.050904275657913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05871847141320476, "300–1000 Hz_10ms_envelope_correlation": 0.9968689643377906, "300–1000 Hz_10ms_modulation_mae_db": 0.44116275166143964, "300–1000 Hz_40ms_log_rms_mae_db": 0.6027923479494749, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0370139395222676, "300–1000 Hz_40ms_envelope_correlation": 0.9987811670431388, "300–1000 Hz_40ms_modulation_mae_db": 0.5577126390933856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4038361181921044, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1407481562167969, "1000–3000 Hz_10ms_envelope_correlation": 0.9848550923177057, "1000–3000 Hz_10ms_modulation_mae_db": 0.799439193245351, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8523663831539501, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08532504521751505, "1000–3000 Hz_40ms_envelope_correlation": 0.99470438780703, "1000–3000 Hz_40ms_modulation_mae_db": 0.6801256329707243, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5089505823700857, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19194477979653005, "3000–6000 Hz_10ms_envelope_correlation": 0.9738576838094474, "3000–6000 Hz_10ms_modulation_mae_db": 0.4632307189573171, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9665426453406376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10836283724222358, "3000–6000 Hz_40ms_envelope_correlation": 0.9908660227700536, "3000–6000 Hz_40ms_modulation_mae_db": 0.4334025453545993, "mr_stft_db": 5.337629842291001, "band_mae_db": 0.15278604325606485, "rms_gain_error_db": -0.08340327470063813, "output_peak": 0.24219877161413753}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7196962682936308, "全帯域_10ms_rms_envelope_nrmse": 0.05043144906365736, "全帯域_10ms_envelope_correlation": 0.9972220519147298, "全帯域_10ms_modulation_mae_db": 0.7164334911921735, "全帯域_40ms_log_rms_mae_db": 0.4120771638270752, "全帯域_40ms_rms_envelope_nrmse": 0.028659978088987362, "全帯域_40ms_envelope_correlation": 0.9991375975070425, "全帯域_40ms_modulation_mae_db": 0.7289837106007268, "300–1000 Hz_10ms_log_rms_mae_db": 1.089802633396136, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06230523962929406, "300–1000 Hz_10ms_envelope_correlation": 0.9964851908073382, "300–1000 Hz_10ms_modulation_mae_db": 0.9814242905731029, "300–1000 Hz_40ms_log_rms_mae_db": 0.60629146026504, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0370348297437665, "300–1000 Hz_40ms_envelope_correlation": 0.9988215438211003, "300–1000 Hz_40ms_modulation_mae_db": 0.8289143526149512, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3647050523344264, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12765214201944702, "1000–3000 Hz_10ms_envelope_correlation": 0.9877393263758691, "1000–3000 Hz_10ms_modulation_mae_db": 0.8237548942838117, "1000–3000 Hz_40ms_log_rms_mae_db": 0.76872317049148, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07307180029009656, "1000–3000 Hz_40ms_envelope_correlation": 0.9964337255639998, "1000–3000 Hz_40ms_modulation_mae_db": 0.5859829365752546, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5054051521570522, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1764655788395653, "3000–6000 Hz_10ms_envelope_correlation": 0.9792603636039472, "3000–6000 Hz_10ms_modulation_mae_db": 1.2213979917087903, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9944077107493684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10870154174072794, "3000–6000 Hz_40ms_envelope_correlation": 0.9932938117988116, "3000–6000 Hz_40ms_modulation_mae_db": 1.2528615288248242, "mr_stft_db": 5.334759115306302, "band_mae_db": 0.24737879909442606, "rms_gain_error_db": -0.09919114452063858, "output_peak": 0.24261199589822613}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7225335092723179, "全帯域_10ms_rms_envelope_nrmse": 0.051098009256326994, "全帯域_10ms_envelope_correlation": 0.9970814870581557, "全帯域_10ms_modulation_mae_db": 0.6315665132680315, "全帯域_40ms_log_rms_mae_db": 0.3952712657803114, "全帯域_40ms_rms_envelope_nrmse": 0.030914555864526596, "全帯域_40ms_envelope_correlation": 0.9989084040829331, "全帯域_40ms_modulation_mae_db": 0.671087495144099, "300–1000 Hz_10ms_log_rms_mae_db": 1.0442434741410518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06238595298407729, "300–1000 Hz_10ms_envelope_correlation": 0.9964021550565555, "300–1000 Hz_10ms_modulation_mae_db": 0.48018199750822754, "300–1000 Hz_40ms_log_rms_mae_db": 0.5799790434415303, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04084245984424412, "300–1000 Hz_40ms_envelope_correlation": 0.998445472189703, "300–1000 Hz_40ms_modulation_mae_db": 0.6859212636979358, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3231297960108637, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.127461275327963, "1000–3000 Hz_10ms_envelope_correlation": 0.9873736462926272, "1000–3000 Hz_10ms_modulation_mae_db": 0.5505551743191484, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7227158396410341, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06455838008960592, "1000–3000 Hz_40ms_envelope_correlation": 0.9969137820219129, "1000–3000 Hz_40ms_modulation_mae_db": 0.6747988318930309, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5049264637944284, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18614815490272757, "3000–6000 Hz_10ms_envelope_correlation": 0.9751899028366507, "3000–6000 Hz_10ms_modulation_mae_db": 1.3568764649746043, "3000–6000 Hz_40ms_log_rms_mae_db": 0.927468948187442, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10872369871576723, "3000–6000 Hz_40ms_envelope_correlation": 0.9911302675796534, "3000–6000 Hz_40ms_modulation_mae_db": 1.334471398460532, "mr_stft_db": 5.320955205193038, "band_mae_db": 0.16303349298572686, "rms_gain_error_db": -0.08072638490949094, "output_peak": 0.25644206110335127}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.213652328050275, "spectral_convergence": 0.13904079524472315, "spectral_convergence_status": "ok", "amplitude_floor": 9.671435620644006e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.241628873592589, "spectral_convergence": 0.13441137065033384, "spectral_convergence_status": "ok", "amplitude_floor": 9.322400769888516e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.47753299006928, "spectral_convergence": 0.13668488798913367, "spectral_convergence_status": "ok", "amplitude_floor": 9.453157582642776e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00016624637287256398, "output_power": 0.0001727842878298011, "delta_db": 0.1675206855645307, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013640231584289317, "output_power": 0.0013323641859048282, "delta_db": -0.10198793555879052, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005975940179232537, "output_power": 0.0005686460397555227, "delta_db": -0.21564222506477435, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.8168186983374496e-05, "output_power": 1.8175885437258295e-05, "delta_db": 0.0018398577089704684, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.463537251271816e-05, "output_power": 2.2129653988779287e-05, "delta_db": -0.4658451013969447, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0343523773544458e-05, "output_power": 9.89925943405385e-06, "delta_db": -0.19065810829161992, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.19058231893093844, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1150341097731428, "pitch": null, "output_rms_dbfs": -26.728456281944023, "output_peak": 0.25394118274332667, "reference_rms_dbfs": -26.613422172170882, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.3109380639040475}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001446111949915804, "log_rms_mae_db": 0.6569237960574047, "rms_envelope_nrmse": 0.05105657569767949, "envelope_correlation": 0.9971999973777944, "rms_gain_error_db": -0.11503597395378473, "reference_mean_power": 0.002182828397481346, "output_mean_power": 0.0021257686487361595}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.790920693994897e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003230276115117931, "output_power": 0.0003041584284142701, "reference_active": true, "delta_db": -0.2613979068678151}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.477236494661316e-05, "output_power": 4.2675393338689894e-05, "reference_active": true, "delta_db": -0.20832502412468118}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.548345752003965e-06, "output_power": 8.155219093817928e-06, "reference_active": true, "delta_db": -0.6849250365425248}], "active_band_count": 3, "active_band_mae_db": 0.3848826558450071}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014111493703772434, "log_rms_mae_db": 0.37175083978117573, "rms_envelope_nrmse": 0.02980872306090521, "envelope_correlation": 0.9991158508303405, "rms_gain_error_db": -0.11505004914019304, "reference_mean_power": 0.002193693032291648, "output_mean_power": 0.0021363423551839546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.190420923463322e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028912450266098784, "output_power": 0.00027251712865227997, "reference_active": true, "delta_db": -0.25691094331530595}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2978791346744758e-05, "output_power": 2.2115578823968398e-05, "reference_active": true, "delta_db": -0.1662887114746972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.072854493661425e-07, "output_power": 5.297594044777663e-07, "reference_active": true, "delta_db": -0.5931420029646423}], "active_band_count": 3, "active_band_mae_db": 0.33878055258488177}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012712872255296568, "log_rms_mae_db": 1.0569978715202102, "rms_envelope_nrmse": 0.06501187509995847, "envelope_correlation": 0.996262388145528, "rms_gain_error_db": -0.16219182098335144, "reference_mean_power": 0.0014813690704446063, "output_mean_power": 0.0014270661035473}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.409297669270013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003163260619212235, "output_power": 0.00029599120873596927, "reference_active": true, "delta_db": -0.28856162433565763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6701975061022124e-05, "output_power": 3.329695121216894e-05, "reference_active": true, "delta_db": -0.4228496600646922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.301526036454895e-06, "output_power": 5.020706442644278e-06, "reference_active": true, "delta_db": -0.9868090579410991}], "active_band_count": 3, "active_band_mae_db": 0.5660734474471497}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012343588785799389, "log_rms_mae_db": 0.6010646582971552, "rms_envelope_nrmse": 0.039572444808142676, "envelope_correlation": 0.9987415560314927, "rms_gain_error_db": -0.16220661799244193, "reference_mean_power": 0.0014887936512772218, "output_mean_power": 0.00143421363283986}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.99328908394483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002875949770086079, "output_power": 0.0002700483512364163, "reference_active": true, "delta_db": -0.273397665377313}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97680708793848e-05, "output_power": 1.8221791964193885e-05, "reference_active": true, "delta_db": -0.35373205533775753}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.520482053574084e-07, "output_power": 3.369701531521404e-07, "reference_active": true, "delta_db": -1.2759331411653188}], "active_band_count": 3, "active_band_mae_db": 0.6343542872934631}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.4084091617788044e-05, "log_rms_mae_db": 1.3572689061243333, "rms_envelope_nrmse": 0.12401058247342837, "envelope_correlation": 0.9882702226981721, "rms_gain_error_db": -0.2721891992505082, "reference_mean_power": 3.6694438258439554e-05, "output_mean_power": 3.4465241320647814e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6589829493109366e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3179349061495639e-05, "output_power": 1.1815483102259852e-05, "reference_active": true, "delta_db": -0.4744247721319057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9890642401085437e-06, "output_power": 2.7999574259416486e-06, "reference_active": true, "delta_db": -0.28383821036249773}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.136200949313584e-07, "output_power": 4.949730481211584e-07, "reference_active": true, "delta_db": -2.158401146086261}], "active_band_count": 3, "active_band_mae_db": 0.9722213761935548}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.151462172211685e-05, "log_rms_mae_db": 0.7944652343077884, "rms_envelope_nrmse": 0.07031725091837979, "envelope_correlation": 0.9964910071400049, "rms_gain_error_db": -0.2724009942107423, "reference_mean_power": 3.687647047180041e-05, "output_mean_power": 3.463452594475404e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4709890047082147e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2743108138238886e-05, "output_power": 1.1372103747063863e-05, "reference_active": true, "delta_db": -0.4943455547236508}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4131557808728274e-06, "output_power": 1.4119935895028984e-06, "reference_active": true, "delta_db": -0.003573144362843307}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.7131268313465536e-08, "output_power": 3.4612966433916135e-08, "reference_active": true, "delta_db": -2.17635044285048}], "active_band_count": 3, "active_band_mae_db": 0.8914230473123247}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.703026630874025e-05, "log_rms_mae_db": 1.457212279290237, "rms_envelope_nrmse": 0.16326983541778708, "envelope_correlation": 0.9809150970935498, "rms_gain_error_db": -0.15957727751022543, "reference_mean_power": 1.1971789360385644e-05, "output_mean_power": 1.1539881235125518e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.043568625201122e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.248382965898838e-06, "output_power": 4.1674220885393895e-06, "reference_active": true, "delta_db": -0.08356169561752236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.589400168019184e-06, "output_power": 2.5321151709664087e-06, "reference_active": true, "delta_db": -0.09715716559898283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.486231293405906e-07, "output_power": 5.941601844237947e-07, "reference_active": true, "delta_db": -0.3808888546922187}], "active_band_count": 3, "active_band_mae_db": 0.18720257196957465}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.3691778537856407e-05, "log_rms_mae_db": 0.9227947309638281, "rms_envelope_nrmse": 0.10251214700141707, "envelope_correlation": 0.9918406051966874, "rms_gain_error_db": -0.1595725313548792, "reference_mean_power": 1.2031928101806957e-05, "output_mean_power": 1.1597863016358583e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.457061965350652e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.507382710758712e-06, "output_power": 4.471787888978851e-06, "reference_active": true, "delta_db": -0.0344323910263196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3712782786962106e-06, "output_power": 1.3389545814127775e-06, "reference_active": true, "delta_db": -0.10359751169554543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.274779912953651e-08, "output_power": 4.514222430070655e-08, "reference_active": true, "delta_db": 0.2366919434875296}], "active_band_count": 3, "active_band_mae_db": 0.12490728206979822}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9420562395702293, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "854f6fddfb07beab5fd26ccc5fc94d2125ecd16165491071b923d9ab35c24dc2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9872219214414811, "standard_deviation": 0.001544537625640825, "minimum": 0.9848550923177057, "maximum": 0.9892799271867235, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3642832339593145, "standard_deviation": 0.03348876434956501, "minimum": 1.31445007306559, "maximum": 1.4078231812157194, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8448607443717698, "standard_deviation": 0.21662311001610873, "minimum": 0.5505551743191484, "maximum": 1.2145232255081646, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1288732858287281, "standard_deviation": 0.00732286384168951, "minimum": 0.11826577465501195, "maximum": 0.1407481562167969, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960900551757081, "standard_deviation": 0.0006612057483843872, "minimum": 0.99470438780703, "maximum": 0.9969137820219129, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7939468533732039, "standard_deviation": 0.0428911560353218, "minimum": 0.7227158396410341, "maximum": 0.8523663831539501, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.742740688081645, "standard_deviation": 0.171036877910451, "minimum": 0.585449171710056, "maximum": 1.0189341032030972, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07315969184939866, "standard_deviation": 0.005766862175438924, "minimum": 0.06455838008960592, "maximum": 0.08532504521751505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9759026246175138, "standard_deviation": 0.0033583267893961067, "minimum": 0.9716883405303648, "maximum": 0.9809150970935498, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4786876417405834, "standard_deviation": 0.025488311212892953, "minimum": 1.4397245829303975, "maximum": 1.5089505823700857, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7490748892745916, "standard_deviation": 0.4204530317072507, "minimum": 0.18720257196957465, "maximum": 1.3568764649746043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18393498909090733, "standard_deviation": 0.012147729551492535, "minimum": 0.16326983541778708, "maximum": 0.1985985362806428, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9912535967962932, "standard_deviation": 0.0010769861905544585, "minimum": 0.989909136450297, "maximum": 0.9932938117988116, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9435680149870231, "standard_deviation": 0.029084153144003196, "minimum": 0.9194608123812077, "maximum": 0.9944077107493684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.750136478456744, "standard_deviation": 0.4457370639300682, "minimum": 0.12490728206979822, "maximum": 1.334471398460532, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10824120585977731, "standard_deviation": 0.0044755972218889556, "minimum": 0.10165968447779751, "maximum": 0.11445501584857411, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9963821096903837, "standard_deviation": 0.00024863160099033657, "minimum": 0.9960600924959736, "maximum": 0.9968689643377906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0588276657303122, "standard_deviation": 0.016790960303168588, "minimum": 1.0372263262298056, "maximum": 1.089802633396136, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5838469421610138, "standard_deviation": 0.1767260750577539, "minimum": 0.4213739095997265, "maximum": 0.9814242905731029, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0630042266056375, "standard_deviation": 0.0021698028708349976, "minimum": 0.05871847141320476, "maximum": 0.06563504918420247, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986717835578527, "standard_deviation": 0.00014958312688406015, "minimum": 0.998445472189703, "maximum": 0.9988215438211003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5893556880713675, "standard_deviation": 0.018061195963787222, "minimum": 0.5547473121333006, "maximum": 0.60629146026504, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6712157329483237, "standard_deviation": 0.09307519865262807, "minimum": 0.5577126390933856, "maximum": 0.8289143526149512, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.038737864299182384, "standard_deviation": 0.0016669360425616972, "minimum": 0.0370139395222676, "maximum": 0.04084245984424412, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1910803268320847, "standard_deviation": 0.036773873041478715, "minimum": 0.14569345254342828, "maximum": 0.24737879909442606, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.326551953905649, "standard_deviation": 0.011666397759736017, "minimum": 5.3109380639040475, "maximum": 5.341019117317248, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.24753194838469916, "standard_deviation": 0.005907255156150154, "minimum": 0.24219877161413753, "maximum": 0.25644206110335127, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08856717911829218, "standard_deviation": 0.014660015192167853, "minimum": -0.1150341097731428, "maximum": -0.0713914965070877, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9972482079983953, "standard_deviation": 0.00010034330267227022, "minimum": 0.9970814870581557, "maximum": 0.9973778385970762, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7068297558908763, "standard_deviation": 0.021808327946153963, "minimum": 0.6569237960574047, "maximum": 0.7225335092723179, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.48975349534425655, "standard_deviation": 0.1219177795265921, "minimum": 0.3848826558450071, "maximum": 0.7164334911921735, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04995033315474016, "standard_deviation": 0.0009288792821711934, "minimum": 0.0485150522711774, "maximum": 0.051098009256326994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991111448682974, "standard_deviation": 8.985352836461041e-05, "minimum": 0.9989084040829331, "maximum": 0.9992181697529119, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3915103283290231, "standard_deviation": 0.01701590618490513, "minimum": 0.36880131662995613, "maximum": 0.41632520054866756, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5286263263575816, "standard_deviation": 0.12654250387639163, "minimum": 0.33878055258488177, "maximum": 0.7289837106007268, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02858646181660223, "standard_deviation": 0.0012523896479612157, "minimum": 0.027338015968297758, "maximum": 0.030914555864526596, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284447-0007", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.817594289779663, "predicted_mos": 2.817594289779663, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b331092a6b240b4066444c94f6cabfe23d2572418438f80a5ebe7f413ffac952", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8405271996743977, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6982968215171205, "全帯域_10ms_rms_envelope_nrmse": 0.04723170877936086, "全帯域_10ms_envelope_correlation": 0.997513304047904, "全帯域_10ms_modulation_mae_db": 0.38706626400225747, "全帯域_40ms_log_rms_mae_db": 0.35460095925315954, "全帯域_40ms_rms_envelope_nrmse": 0.02557915968608533, "全帯域_40ms_envelope_correlation": 0.9992681570033535, "全帯域_40ms_modulation_mae_db": 0.4902990503231421, "300–1000 Hz_10ms_log_rms_mae_db": 1.2076387016931154, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06262992333905265, "300–1000 Hz_10ms_envelope_correlation": 0.9963865140735093, "300–1000 Hz_10ms_modulation_mae_db": 0.46698804411513795, "300–1000 Hz_40ms_log_rms_mae_db": 0.6929535508163384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03821619390371587, "300–1000 Hz_40ms_envelope_correlation": 0.9986659782377234, "300–1000 Hz_40ms_modulation_mae_db": 0.8535599092351651, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7777602016376246, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15270502794755872, "1000–3000 Hz_10ms_envelope_correlation": 0.981711561617701, "1000–3000 Hz_10ms_modulation_mae_db": 0.6064334838032833, "1000–3000 Hz_40ms_log_rms_mae_db": 1.01595475141074, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08237481855213516, "1000–3000 Hz_40ms_envelope_correlation": 0.9947089799121519, "1000–3000 Hz_40ms_modulation_mae_db": 0.5356663535108863, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0633522763334216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2324162862499927, "3000–6000 Hz_10ms_envelope_correlation": 0.9613060761532994, "3000–6000 Hz_10ms_modulation_mae_db": 0.3684656764874017, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3325371486226845, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15084620733496096, "3000–6000 Hz_40ms_envelope_correlation": 0.9822188504098999, "3000–6000 Hz_40ms_modulation_mae_db": 0.3907214620433366, "mr_stft_db": 5.828563075778537, "band_mae_db": 0.13436604828713425, "rms_gain_error_db": -0.07260151051542385, "output_peak": 0.2526660436143688}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6961538463668835, "全帯域_10ms_rms_envelope_nrmse": 0.04866908383950332, "全帯域_10ms_envelope_correlation": 0.9974050184232963, "全帯域_10ms_modulation_mae_db": 0.8543207489818831, "全帯域_40ms_log_rms_mae_db": 0.36311025715635503, "全帯域_40ms_rms_envelope_nrmse": 0.028021042722971994, "全帯域_40ms_envelope_correlation": 0.9991523517790512, "全帯域_40ms_modulation_mae_db": 0.7797718933381338, "300–1000 Hz_10ms_log_rms_mae_db": 1.190415801190732, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06182825413690693, "300–1000 Hz_10ms_envelope_correlation": 0.9966198444268252, "300–1000 Hz_10ms_modulation_mae_db": 1.008484106904177, "300–1000 Hz_40ms_log_rms_mae_db": 0.6571027553386781, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03727296870738063, "300–1000 Hz_40ms_envelope_correlation": 0.9988954072758738, "300–1000 Hz_40ms_modulation_mae_db": 1.0136413628710514, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7274215891266125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15379951116880128, "1000–3000 Hz_10ms_envelope_correlation": 0.981362358654632, "1000–3000 Hz_10ms_modulation_mae_db": 1.0283230588579833, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0559097061917189, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09154534851322017, "1000–3000 Hz_40ms_envelope_correlation": 0.9930459007018535, "1000–3000 Hz_40ms_modulation_mae_db": 0.7502008430448518, "3000–6000 Hz_10ms_log_rms_mae_db": 2.127540733078889, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25529635468672296, "3000–6000 Hz_10ms_envelope_correlation": 0.9529394631713765, "3000–6000 Hz_10ms_modulation_mae_db": 1.3224167996872447, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3919027466945668, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1553167521743166, "3000–6000 Hz_40ms_envelope_correlation": 0.9814747752303483, "3000–6000 Hz_40ms_modulation_mae_db": 1.4151799953191855, "mr_stft_db": 5.821629174477116, "band_mae_db": 0.15158583079554508, "rms_gain_error_db": -0.08941306677317165, "output_peak": 0.2572263642524707}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7232765157302983, "全帯域_10ms_rms_envelope_nrmse": 0.05155855526163505, "全帯域_10ms_envelope_correlation": 0.9970422220344827, "全帯域_10ms_modulation_mae_db": 0.6636775658966295, "全帯域_40ms_log_rms_mae_db": 0.4017679373308202, "全帯域_40ms_rms_envelope_nrmse": 0.027012101290964603, "全帯域_40ms_envelope_correlation": 0.9992026586561471, "全帯域_40ms_modulation_mae_db": 0.7419637437264849, "300–1000 Hz_10ms_log_rms_mae_db": 1.2795639160381518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06192305500324314, "300–1000 Hz_10ms_envelope_correlation": 0.9965344130510609, "300–1000 Hz_10ms_modulation_mae_db": 0.8451857030244784, "300–1000 Hz_40ms_log_rms_mae_db": 0.6720213241141295, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03385134012535177, "300–1000 Hz_40ms_envelope_correlation": 0.9990561817301182, "300–1000 Hz_40ms_modulation_mae_db": 0.9241917420297087, "1000–3000 Hz_10ms_log_rms_mae_db": 1.809986170313116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.152014123913332, "1000–3000 Hz_10ms_envelope_correlation": 0.9818884771299702, "1000–3000 Hz_10ms_modulation_mae_db": 0.9443098087740269, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1130272545801014, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09005408323573769, "1000–3000 Hz_40ms_envelope_correlation": 0.9933469494180686, "1000–3000 Hz_40ms_modulation_mae_db": 0.7399341251694357, "3000–6000 Hz_10ms_log_rms_mae_db": 2.153553302241826, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2328255613058661, "3000–6000 Hz_10ms_envelope_correlation": 0.9614444062092754, "3000–6000 Hz_10ms_modulation_mae_db": 0.85268551180652, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4474494102248838, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14272683212566528, "3000–6000 Hz_40ms_envelope_correlation": 0.9841411785130099, "3000–6000 Hz_40ms_modulation_mae_db": 0.9127118936943521, "mr_stft_db": 5.8113815155073665, "band_mae_db": 0.08476123020396871, "rms_gain_error_db": -0.0825289629533925, "output_peak": 0.24929646314868448}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6813957042298792, "全帯域_10ms_rms_envelope_nrmse": 0.04343463427516702, "全帯域_10ms_envelope_correlation": 0.9979772123239642, "全帯域_10ms_modulation_mae_db": 0.41764561621192264, "全帯域_40ms_log_rms_mae_db": 0.37520477565647725, "全帯域_40ms_rms_envelope_nrmse": 0.024420730524917333, "全帯域_40ms_envelope_correlation": 0.9994203199955191, "全帯域_40ms_modulation_mae_db": 0.5196425699008997, "300–1000 Hz_10ms_log_rms_mae_db": 1.1966709421700494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05232932115400163, "300–1000 Hz_10ms_envelope_correlation": 0.9975776805675771, "300–1000 Hz_10ms_modulation_mae_db": 0.5466015588044678, "300–1000 Hz_40ms_log_rms_mae_db": 0.6654085688255643, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0326270941072666, "300–1000 Hz_40ms_envelope_correlation": 0.999128371681536, "300–1000 Hz_40ms_modulation_mae_db": 0.5920045696349314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7166354933120636, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14481777327926423, "1000–3000 Hz_10ms_envelope_correlation": 0.9834185021403052, "1000–3000 Hz_10ms_modulation_mae_db": 0.9419072814310816, "1000–3000 Hz_40ms_log_rms_mae_db": 1.048559616062298, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08659613988520805, "1000–3000 Hz_40ms_envelope_correlation": 0.9936824546668431, "1000–3000 Hz_40ms_modulation_mae_db": 1.19600672642171, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0847037753048956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22539040043268105, "3000–6000 Hz_10ms_envelope_correlation": 0.9635397052573473, "3000–6000 Hz_10ms_modulation_mae_db": 1.1814324638270026, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3763706575115873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13358327828120944, "3000–6000 Hz_40ms_envelope_correlation": 0.9861247134781169, "3000–6000 Hz_40ms_modulation_mae_db": 1.3404485848380439, "mr_stft_db": 5.836192655557173, "band_mae_db": 0.13603505021631132, "rms_gain_error_db": -0.10075075663037675, "output_peak": 0.24661114537571643}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6812962065038423, "全帯域_10ms_rms_envelope_nrmse": 0.04707889961856508, "全帯域_10ms_envelope_correlation": 0.9975738663615952, "全帯域_10ms_modulation_mae_db": 0.5789281414397333, "全帯域_40ms_log_rms_mae_db": 0.3837634715661001, "全帯域_40ms_rms_envelope_nrmse": 0.026559742568530147, "全帯域_40ms_envelope_correlation": 0.999258665786615, "全帯域_40ms_modulation_mae_db": 0.5839414766037138, "300–1000 Hz_10ms_log_rms_mae_db": 1.194270310501828, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05663512102550528, "300–1000 Hz_10ms_envelope_correlation": 0.9971463421419665, "300–1000 Hz_10ms_modulation_mae_db": 0.9476349532559424, "300–1000 Hz_40ms_log_rms_mae_db": 0.6690288241068894, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03268474622400153, "300–1000 Hz_40ms_envelope_correlation": 0.9991526221893686, "300–1000 Hz_40ms_modulation_mae_db": 0.9832165338450465, "1000–3000 Hz_10ms_log_rms_mae_db": 1.754800691242688, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14140718909615296, "1000–3000 Hz_10ms_envelope_correlation": 0.9843161808535299, "1000–3000 Hz_10ms_modulation_mae_db": 0.7718687718442657, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1076521813459557, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09061991200784353, "1000–3000 Hz_40ms_envelope_correlation": 0.9932339669022241, "1000–3000 Hz_40ms_modulation_mae_db": 0.78375599576352, "3000–6000 Hz_10ms_log_rms_mae_db": 2.11547309134743, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2288329995762622, "3000–6000 Hz_10ms_envelope_correlation": 0.9627324548850918, "3000–6000 Hz_10ms_modulation_mae_db": 0.8950385297016431, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3477536165363362, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1415615936306387, "3000–6000 Hz_40ms_envelope_correlation": 0.9844207069035278, "3000–6000 Hz_40ms_modulation_mae_db": 1.111231923294068, "mr_stft_db": 5.8359465143896285, "band_mae_db": 0.13925411899971593, "rms_gain_error_db": -0.08815487373264543, "output_peak": 0.2507888263643189}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6566162111332797, "全帯域_10ms_rms_envelope_nrmse": 0.04599476339686515, "全帯域_10ms_envelope_correlation": 0.9976339487120797, "全帯域_10ms_modulation_mae_db": 0.5481691643822405, "全帯域_40ms_log_rms_mae_db": 0.38306867720284904, "全帯域_40ms_rms_envelope_nrmse": 0.026369984303476125, "全帯域_40ms_envelope_correlation": 0.9992011429884196, "全帯域_40ms_modulation_mae_db": 0.548748732882944, "300–1000 Hz_10ms_log_rms_mae_db": 1.177350569393807, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06118706798133563, "300–1000 Hz_10ms_envelope_correlation": 0.9965426665673353, "300–1000 Hz_10ms_modulation_mae_db": 0.669786228807323, "300–1000 Hz_40ms_log_rms_mae_db": 0.6481755088284528, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.037076589804906274, "300–1000 Hz_40ms_envelope_correlation": 0.9987304332335665, "300–1000 Hz_40ms_modulation_mae_db": 0.6263284810268813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7456245522498568, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14561816557195867, "1000–3000 Hz_10ms_envelope_correlation": 0.9834496264081553, "1000–3000 Hz_10ms_modulation_mae_db": 0.9463009953595588, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0366496367397493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0855152289871559, "1000–3000 Hz_40ms_envelope_correlation": 0.9941721624877434, "1000–3000 Hz_40ms_modulation_mae_db": 1.1105904376830384, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0940323739557933, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22669826560594286, "3000–6000 Hz_10ms_envelope_correlation": 0.9635534117133777, "3000–6000 Hz_10ms_modulation_mae_db": 0.6319186670067882, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3784650146424249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14291986458435502, "3000–6000 Hz_40ms_envelope_correlation": 0.9841599716744784, "3000–6000 Hz_40ms_modulation_mae_db": 0.8236612610707065, "mr_stft_db": 5.818247495355387, "band_mae_db": 0.13231583597031432, "rms_gain_error_db": -0.06527128656083062, "output_peak": 0.2519517234573}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.702498709370405, "全帯域_10ms_rms_envelope_nrmse": 0.05459072730782693, "全帯域_10ms_envelope_correlation": 0.9966441916795921, "全帯域_10ms_modulation_mae_db": 0.6321027373224549, "全帯域_40ms_log_rms_mae_db": 0.3701776574795119, "全帯域_40ms_rms_envelope_nrmse": 0.031193133871737656, "全帯域_40ms_envelope_correlation": 0.9988656780866537, "全帯域_40ms_modulation_mae_db": 0.7496413172945027, "300–1000 Hz_10ms_log_rms_mae_db": 1.204217128157188, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06745856662794604, "300–1000 Hz_10ms_envelope_correlation": 0.99577392961122, "300–1000 Hz_10ms_modulation_mae_db": 0.9182427327785909, "300–1000 Hz_40ms_log_rms_mae_db": 0.663505250730403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041186101234993354, "300–1000 Hz_40ms_envelope_correlation": 0.9984085368000614, "300–1000 Hz_40ms_modulation_mae_db": 1.028319041909114, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7599927483978894, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15242552777262025, "1000–3000 Hz_10ms_envelope_correlation": 0.98162262165189, "1000–3000 Hz_10ms_modulation_mae_db": 0.8067585816333714, "1000–3000 Hz_40ms_log_rms_mae_db": 1.057090528794421, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08754188307715406, "1000–3000 Hz_40ms_envelope_correlation": 0.9934956314190809, "1000–3000 Hz_40ms_modulation_mae_db": 0.5987147644244217, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1388228690934645, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23323163584663525, "3000–6000 Hz_10ms_envelope_correlation": 0.9611363835856866, "3000–6000 Hz_10ms_modulation_mae_db": 1.0628087846034286, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4178659008932024, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13906949633429858, "3000–6000 Hz_40ms_envelope_correlation": 0.9848393488117078, "3000–6000 Hz_40ms_modulation_mae_db": 1.2813116305034966, "mr_stft_db": 5.8388135958571965, "band_mae_db": 0.15245005681972953, "rms_gain_error_db": -0.06937760205948781, "output_peak": 0.25371234856516134}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7060508876025233, "全帯域_10ms_rms_envelope_nrmse": 0.054178905735321824, "全帯域_10ms_envelope_correlation": 0.996734536540117, "全帯域_10ms_modulation_mae_db": 0.8356821066346717, "全帯域_40ms_log_rms_mae_db": 0.36774713070675824, "全帯域_40ms_rms_envelope_nrmse": 0.028492926905702785, "全帯域_40ms_envelope_correlation": 0.9991128469007248, "全帯域_40ms_modulation_mae_db": 0.855852734081246, "300–1000 Hz_10ms_log_rms_mae_db": 1.2241139097583562, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06685754528682376, "300–1000 Hz_10ms_envelope_correlation": 0.9959134199847371, "300–1000 Hz_10ms_modulation_mae_db": 1.0617713399210764, "300–1000 Hz_40ms_log_rms_mae_db": 0.6788019211032037, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036837964627362836, "300–1000 Hz_40ms_envelope_correlation": 0.9988386348812893, "300–1000 Hz_40ms_modulation_mae_db": 1.0541615085811957, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7822934450415941, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16226668789824159, "1000–3000 Hz_10ms_envelope_correlation": 0.9796229208690272, "1000–3000 Hz_10ms_modulation_mae_db": 1.091359549892345, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1218065412995275, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10562084795108678, "1000–3000 Hz_40ms_envelope_correlation": 0.9913351782090087, "1000–3000 Hz_40ms_modulation_mae_db": 1.0397884972512397, "3000–6000 Hz_10ms_log_rms_mae_db": 2.075592718847698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22799656575222674, "3000–6000 Hz_10ms_envelope_correlation": 0.9627374209723417, "3000–6000 Hz_10ms_modulation_mae_db": 1.2359522718653035, "3000–6000 Hz_40ms_log_rms_mae_db": 1.37591539004893, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14647730997982392, "3000–6000 Hz_40ms_envelope_correlation": 0.9834538987435828, "3000–6000 Hz_40ms_modulation_mae_db": 1.3897766224553039, "mr_stft_db": 5.8419999748169085, "band_mae_db": 0.15697774164345515, "rms_gain_error_db": -0.08888296175748389, "output_peak": 0.25407055202335366}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.714296239809384, "spectral_convergence": 0.14309376548697883, "spectral_convergence_status": "ok", "amplitude_floor": 9.671435620644006e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.848162348251399, "spectral_convergence": 0.14396600560812065, "spectral_convergence_status": "ok", "amplitude_floor": 9.322400769888516e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.923230639274829, "spectral_convergence": 0.14257034364741983, "spectral_convergence_status": "ok", "amplitude_floor": 9.453157582642776e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00016624637287256398, "output_power": 0.00017360162084196628, "delta_db": 0.18801596940578283, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013640231584289317, "output_power": 0.0013347213688619187, "delta_db": -0.09431130279524517, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005975940179232537, "output_power": 0.0005853573588580213, "delta_db": -0.08985158938653343, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.8168186983374496e-05, "output_power": 1.7814675087703928e-05, "delta_db": -0.08533685011952499, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.463537251271816e-05, "output_power": 2.2927341043672946e-05, "delta_db": -0.312054425051972, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0343523773544458e-05, "output_power": 1.0431124580175054e-05, "delta_db": 0.03662615296374715, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.13436604828713425, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07260151051542385, "pitch": null, "output_rms_dbfs": -26.686023682686304, "output_peak": 0.2526660436143688, "reference_rms_dbfs": -26.613422172170882, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.828563075778537}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001446111949915804, "log_rms_mae_db": 0.6982968215171205, "rms_envelope_nrmse": 0.04723170877936086, "envelope_correlation": 0.997513304047904, "rms_gain_error_db": -0.07260896557409689, "reference_mean_power": 0.002182828397481346, "output_mean_power": 0.002146637434160615}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.790920693994897e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003230276115117931, "output_power": 0.0003155547018250191, "reference_active": true, "delta_db": -0.10164990586290334}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.477236494661316e-05, "output_power": 4.10203477232277e-05, "reference_active": true, "delta_db": -0.3801069760975059}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.548345752003965e-06, "output_power": 8.165521858300717e-06, "reference_active": true, "delta_db": -0.6794419100463631}], "active_band_count": 3, "active_band_mae_db": 0.38706626400225747}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014111493703772434, "log_rms_mae_db": 0.35460095925315954, "rms_envelope_nrmse": 0.02557915968608533, "envelope_correlation": 0.9992681570033535, "rms_gain_error_db": -0.07265000949943051, "reference_mean_power": 0.002193693032291648, "output_mean_power": 0.0021573015468415403}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.190420923463322e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028912450266098784, "output_power": 0.00028450718635449047, "reference_active": true, "delta_db": -0.06991658030586197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2978791346744758e-05, "output_power": 2.1336268341530525e-05, "reference_active": true, "delta_db": -0.3220871691049456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.072854493661425e-07, "output_power": 4.737001642803015e-07, "reference_active": true, "delta_db": -1.0788934015586187}], "active_band_count": 3, "active_band_mae_db": 0.4902990503231421}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012712872255296568, "log_rms_mae_db": 1.2076387016931154, "rms_envelope_nrmse": 0.06262992333905265, "envelope_correlation": 0.9963865140735093, "rms_gain_error_db": -0.1172341173464355, "reference_mean_power": 0.0014813690704446063, "output_mean_power": 0.001441915668276911}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.409297669270013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003163260619212235, "output_power": 0.0003092363704227344, "reference_active": true, "delta_db": -0.09844407445074187}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6701975061022124e-05, "output_power": 3.2518326827383146e-05, "reference_active": true, "delta_db": -0.5256124413114283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.301526036454895e-06, "output_power": 5.269324850127242e-06, "reference_active": true, "delta_db": -0.7769076165832438}], "active_band_count": 3, "active_band_mae_db": 0.46698804411513795}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012343588785799389, "log_rms_mae_db": 0.6929535508163384, "rms_envelope_nrmse": 0.03821619390371587, "envelope_correlation": 0.9986659782377234, "rms_gain_error_db": -0.11726354424859284, "reference_mean_power": 0.0014887936512772218, "output_mean_power": 0.001449132690689684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.99328908394483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002875949770086079, "output_power": 0.000282984138359106, "reference_active": true, "delta_db": -0.07019203187128524}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97680708793848e-05, "output_power": 1.7666655371785686e-05, "reference_active": true, "delta_db": -0.48809952358623127}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.520482053574084e-07, "output_power": 2.850663350040904e-07, "reference_active": true, "delta_db": -2.002388172247979}], "active_band_count": 3, "active_band_mae_db": 0.8535599092351651}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.4084091617788044e-05, "log_rms_mae_db": 1.7777602016376246, "rms_envelope_nrmse": 0.15270502794755872, "envelope_correlation": 0.981711561617701, "rms_gain_error_db": -0.216792943263282, "reference_mean_power": 3.6694438258439554e-05, "output_mean_power": 3.4907677041097485e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6589829493109366e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3179349061495639e-05, "output_power": 1.1994495896731882e-05, "reference_active": true, "delta_db": -0.4091196042894103}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9890642401085437e-06, "output_power": 2.421763278575623e-06, "reference_active": true, "delta_db": -0.9140355921105336}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.136200949313584e-07, "output_power": 7.257835837659386e-07, "reference_active": true, "delta_db": -0.49614525500990614}], "active_band_count": 3, "active_band_mae_db": 0.6064334838032833}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.151462172211685e-05, "log_rms_mae_db": 1.01595475141074, "rms_envelope_nrmse": 0.08237481855213516, "envelope_correlation": 0.9947089799121519, "rms_gain_error_db": -0.21671346037177977, "reference_mean_power": 3.687647047180041e-05, "output_mean_power": 3.508148760707854e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4709890047082147e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2743108138238886e-05, "output_power": 1.146081829467712e-05, "reference_active": true, "delta_db": -0.46059741482237465}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4131557808728274e-06, "output_power": 1.2982278392048932e-06, "reference_active": true, "delta_db": -0.36839121537176883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.7131268313465536e-08, "output_power": 4.7760935244388853e-08, "reference_active": true, "delta_db": -0.7780104303385156}], "active_band_count": 3, "active_band_mae_db": 0.5356663535108863}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.703026630874025e-05, "log_rms_mae_db": 2.0633522763334216, "rms_envelope_nrmse": 0.2324162862499927, "envelope_correlation": 0.9613060761532994, "rms_gain_error_db": -0.06722991723029306, "reference_mean_power": 1.1971789360385644e-05, "output_mean_power": 1.1787890019460435e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.043568625201122e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.248382965898838e-06, "output_power": 4.036787072566373e-06, "reference_active": true, "delta_db": -0.22187816457343346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.589400168019184e-06, "output_power": 2.1138565887245662e-06, "reference_active": true, "delta_db": -0.8812365190103999}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.486231293405906e-07, "output_power": 6.489640895526258e-07, "reference_active": true, "delta_db": 0.0022823458783718907}], "active_band_count": 3, "active_band_mae_db": 0.3684656764874017}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.3691778537856407e-05, "log_rms_mae_db": 1.3325371486226845, "rms_envelope_nrmse": 0.15084620733496096, "envelope_correlation": 0.9822188504098999, "rms_gain_error_db": -0.06722526312680933, "reference_mean_power": 1.2031928101806957e-05, "output_mean_power": 1.1847117662156448e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.457061965350652e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.507382710758712e-06, "output_power": 4.202468446171364e-06, "reference_active": true, "delta_db": -0.3041997324523903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3712782786962106e-06, "output_power": 1.1486154151736934e-06, "reference_active": true, "delta_db": -0.7695095625624627}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.274779912953651e-08, "output_power": 4.178960352905745e-08, "reference_active": true, "delta_db": -0.09845509111515668}], "active_band_count": 3, "active_band_mae_db": 0.3907214620433366}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9279366287289614, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "514b20784dd8848b6dfde7e350bd71f5d998d0d22a553594d4bce92e51b34f48", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9821740311656513, "standard_deviation": 0.0014895399178154968, "minimum": 0.9796229208690272, "maximum": 0.9843161808535299, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7593143614151807, "standard_deviation": 0.03043212243650163, "minimum": 1.7166354933120636, "maximum": 1.809986170313116, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8921576914494895, "standard_deviation": 0.15572388800321926, "minimum": 0.6064334838032833, "maximum": 1.091359549892345, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1506317508309912, "standard_deviation": 0.0065341350558353805, "minimum": 0.14140718909615296, "maximum": 0.16226668789824159, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9933776529646219, "standard_deviation": 0.0009866140438323956, "minimum": 0.9913351782090087, "maximum": 0.9947089799121519, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.069581277053064, "standard_deviation": 0.039296606544607676, "minimum": 1.01595475141074, "maximum": 1.1218065412995275, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8443322179086379, "standard_deviation": 0.2425469798854834, "minimum": 0.5356663535108863, "maximum": 1.19600672642171, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08998353277619267, "standard_deviation": 0.006996623725712516, "minimum": 0.08237481855213516, "maximum": 0.10562084795108678, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9611736652434746, "standard_deviation": 0.0034647366987240664, "minimum": 0.9529394631713765, "maximum": 0.9635534117133777, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1066338925254273, "standard_deviation": 0.03213669401039835, "minimum": 2.0633522763334216, "maximum": 2.153553302241826, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9438398381231665, "standard_deviation": 0.3247625456870378, "minimum": 0.3684656764874017, "maximum": 1.3224167996872447, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23283600868204124, "standard_deviation": 0.009542036332640624, "minimum": 0.22539040043268105, "maximum": 0.25529635468672296, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9838541804705839, "standard_deviation": 0.0014696357323314502, "minimum": 0.9814747752303483, "maximum": 0.9861247134781169, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.383532485646827, "standard_deviation": 0.03654715346156552, "minimum": 1.3325371486226845, "maximum": 1.4474494102248838, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0831304216523117, "standard_deviation": 0.3558418232449122, "minimum": 0.3907214620433366, "maximum": 1.4151799953191855, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14406266680565855, "standard_deviation": 0.006788609311557796, "minimum": 0.13358327828120944, "maximum": 0.1553167521743166, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.996561851303029, "standard_deviation": 0.0005916139741294948, "minimum": 0.99577392961122, "maximum": 0.9975776805675771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2092801598629035, "standard_deviation": 0.03150470694389142, "minimum": 1.177350569393807, "maximum": 1.2795639160381518, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8080868334513992, "standard_deviation": 0.22082611525166723, "minimum": 0.46698804411513795, "maximum": 1.0617713399210764, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.061356106819351885, "standard_deviation": 0.0049781227561641155, "minimum": 0.05232932115400163, "maximum": 0.06745856662794604, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9988595207536921, "standard_deviation": 0.0002552583270110768, "minimum": 0.9984085368000614, "maximum": 0.9991526221893686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6683747129829574, "standard_deviation": 0.01359698561093746, "minimum": 0.6481755088284528, "maximum": 0.6929535508163384, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8844278936416368, "standard_deviation": 0.1815135500525997, "minimum": 0.5920045696349314, "maximum": 1.0541615085811957, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.036219124841872355, "standard_deviation": 0.002973426511619289, "minimum": 0.0326270941072666, "maximum": 0.041186101234993354, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13596823911702177, "standard_deviation": 0.02269620791919722, "minimum": 0.08476123020396871, "maximum": 0.15697774164345515, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.829096750217413, "standard_deviation": 0.010991989315569816, "minimum": 5.8113815155073665, "maximum": 5.8419999748169085, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2520404333501718, "standard_deviation": 0.003224523878646885, "minimum": 0.24661114537571643, "maximum": 0.2572263642524707, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08212262762285157, "standard_deviation": 0.012068068585202537, "minimum": -0.10075075663037675, "maximum": -0.06527128656083062, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973155375153789, "standard_deviation": 0.00046537545305423846, "minimum": 0.9966441916795921, "maximum": 0.9979772123239642, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.693198112806779, "standard_deviation": 0.020036845282485387, "minimum": 0.6566162111332797, "maximum": 0.7232765157302983, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6146990431089742, "standard_deviation": 0.17125852550320603, "minimum": 0.38706626400225747, "maximum": 0.8543207489818831, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04909215977678066, "standard_deviation": 0.003990833833223422, "minimum": 0.04343463427516702, "maximum": 0.05459072730782693, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991852276495604, "standard_deviation": 0.00015880253040542608, "minimum": 0.9988656780866537, "maximum": 0.9994203199955191, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37493010829400386, "standard_deviation": 0.014598303509822686, "minimum": 0.35460095925315954, "maximum": 0.4017679373308202, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6587326897688834, "standard_deviation": 0.1384117750217523, "minimum": 0.4902990503231421, "maximum": 0.855852734081246, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.027206102734298247, "standard_deviation": 0.0020624609089057223, "minimum": 0.024420730524917333, "maximum": 0.031193133871737656, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284447-0007", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 3.097654104232788, "predicted_mos": 3.097654104232788, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "56d5353d258028e6cf74dad4839c87378507799740b481c7f9ea4c06e2d40ddf", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.2553161000832915, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7293994575042108, "全帯域_10ms_rms_envelope_nrmse": 0.053527627572371624, "全帯域_10ms_envelope_correlation": 0.9968095555368692, "全帯域_10ms_modulation_mae_db": 0.3048018570924571, "全帯域_40ms_log_rms_mae_db": 0.4102253624275308, "全帯域_40ms_rms_envelope_nrmse": 0.03216353037960718, "全帯域_40ms_envelope_correlation": 0.9988340161649127, "全帯域_40ms_modulation_mae_db": 0.40657706890463663, "300–1000 Hz_10ms_log_rms_mae_db": 1.0063390587841257, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06601723439657356, "300–1000 Hz_10ms_envelope_correlation": 0.9960158454259898, "300–1000 Hz_10ms_modulation_mae_db": 0.22295412035102902, "300–1000 Hz_40ms_log_rms_mae_db": 0.5695642982107194, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04142048409700128, "300–1000 Hz_40ms_envelope_correlation": 0.9984665380917317, "300–1000 Hz_40ms_modulation_mae_db": 0.22172110513074492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2865650182623254, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14068912724301386, "1000–3000 Hz_10ms_envelope_correlation": 0.9850130174601966, "1000–3000 Hz_10ms_modulation_mae_db": 0.8947188102031897, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7751908622515639, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09098928676471697, "1000–3000 Hz_40ms_envelope_correlation": 0.9940010379846639, "1000–3000 Hz_40ms_modulation_mae_db": 0.5744707649582286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4155550513946407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17646950425740696, "3000–6000 Hz_10ms_envelope_correlation": 0.9778128764823277, "3000–6000 Hz_10ms_modulation_mae_db": 0.6760806839419174, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9177682813907129, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09862953350338426, "3000–6000 Hz_40ms_envelope_correlation": 0.9923440005712459, "3000–6000 Hz_40ms_modulation_mae_db": 0.7007529667838153, "mr_stft_db": 5.330775600584114, "band_mae_db": 0.1521767030669446, "rms_gain_error_db": -0.09007239866701794, "output_peak": 0.23857044476392023}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7658396671646794, "全帯域_10ms_rms_envelope_nrmse": 0.05493639757237985, "全帯域_10ms_envelope_correlation": 0.9967144489801517, "全帯域_10ms_modulation_mae_db": 0.577878231102655, "全帯域_40ms_log_rms_mae_db": 0.4128331476227238, "全帯域_40ms_rms_envelope_nrmse": 0.032829517481132296, "全帯域_40ms_envelope_correlation": 0.9988701032161806, "全帯域_40ms_modulation_mae_db": 0.7121118286727359, "300–1000 Hz_10ms_log_rms_mae_db": 0.9686110948615898, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06552380066294307, "300–1000 Hz_10ms_envelope_correlation": 0.9961465683774601, "300–1000 Hz_10ms_modulation_mae_db": 0.6585720885523871, "300–1000 Hz_40ms_log_rms_mae_db": 0.554267565381653, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04194629793282168, "300–1000 Hz_40ms_envelope_correlation": 0.9984968161537067, "300–1000 Hz_40ms_modulation_mae_db": 0.8763648349035801, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2914084696252455, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14413649751843202, "1000–3000 Hz_10ms_envelope_correlation": 0.9836671167466404, "1000–3000 Hz_10ms_modulation_mae_db": 0.4141138910131444, "1000–3000 Hz_40ms_log_rms_mae_db": 0.788667399053831, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08048593419065941, "1000–3000 Hz_40ms_envelope_correlation": 0.9948573150515978, "1000–3000 Hz_40ms_modulation_mae_db": 0.16051693935184133, "3000–6000 Hz_10ms_log_rms_mae_db": 1.448399101179836, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19627216587340934, "3000–6000 Hz_10ms_envelope_correlation": 0.9723762956275802, "3000–6000 Hz_10ms_modulation_mae_db": 0.6620153649420577, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9174220619683675, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1134730326282559, "3000–6000 Hz_40ms_envelope_correlation": 0.9903982988307936, "3000–6000 Hz_40ms_modulation_mae_db": 0.5170704614984651, "mr_stft_db": 5.316637327258845, "band_mae_db": 0.2239074482811986, "rms_gain_error_db": -0.11760960386033348, "output_peak": 0.2549716129700771}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6972384671229177, "全帯域_10ms_rms_envelope_nrmse": 0.05025918415718034, "全帯域_10ms_envelope_correlation": 0.9972268721077723, "全帯域_10ms_modulation_mae_db": 1.004129745023378, "全帯域_40ms_log_rms_mae_db": 0.39043486541131417, "全帯域_40ms_rms_envelope_nrmse": 0.02815595536434084, "全帯域_40ms_envelope_correlation": 0.9991605087902264, "全帯域_40ms_modulation_mae_db": 1.0576314875241069, "300–1000 Hz_10ms_log_rms_mae_db": 0.9873948384210517, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06213419216168691, "300–1000 Hz_10ms_envelope_correlation": 0.9965371356366318, "300–1000 Hz_10ms_modulation_mae_db": 0.9354433943517696, "300–1000 Hz_40ms_log_rms_mae_db": 0.5764258897853665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03864567574214668, "300–1000 Hz_40ms_envelope_correlation": 0.998742206211513, "300–1000 Hz_40ms_modulation_mae_db": 1.1914566059717788, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2688212787297348, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12196026112815089, "1000–3000 Hz_10ms_envelope_correlation": 0.9883972895424471, "1000–3000 Hz_10ms_modulation_mae_db": 0.3764644116542308, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8042521348211397, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07816759661684101, "1000–3000 Hz_40ms_envelope_correlation": 0.9950800138050695, "1000–3000 Hz_40ms_modulation_mae_db": 0.48021593209824337, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3634078917919819, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1977316667217976, "3000–6000 Hz_10ms_envelope_correlation": 0.9719858380204907, "3000–6000 Hz_10ms_modulation_mae_db": 0.7008777452389204, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8881396190754535, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13263027480192346, "3000–6000 Hz_40ms_envelope_correlation": 0.9866353254292787, "3000–6000 Hz_40ms_modulation_mae_db": 0.5885876002088518, "mr_stft_db": 5.301424032547698, "band_mae_db": 0.18611123464465443, "rms_gain_error_db": -0.09756011471830794, "output_peak": 0.24626954082489333}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7005146808223796, "全帯域_10ms_rms_envelope_nrmse": 0.0476329184684592, "全帯域_10ms_envelope_correlation": 0.9975040490545765, "全帯域_10ms_modulation_mae_db": 0.6317922408146908, "全帯域_40ms_log_rms_mae_db": 0.37630688689172076, "全帯域_40ms_rms_envelope_nrmse": 0.027210562343693383, "全帯域_40ms_envelope_correlation": 0.9991996977394701, "全帯域_40ms_modulation_mae_db": 0.6659101745372334, "300–1000 Hz_10ms_log_rms_mae_db": 0.9643780179044793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06273325744191359, "300–1000 Hz_10ms_envelope_correlation": 0.996463844409138, "300–1000 Hz_10ms_modulation_mae_db": 0.7298682566331612, "300–1000 Hz_40ms_log_rms_mae_db": 0.5653951600176109, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03925400918490795, "300–1000 Hz_40ms_envelope_correlation": 0.9986873516581573, "300–1000 Hz_40ms_modulation_mae_db": 0.9007675380842833, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2607401093899415, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11904276349385068, "1000–3000 Hz_10ms_envelope_correlation": 0.9890174980447711, "1000–3000 Hz_10ms_modulation_mae_db": 1.116467940320727, "1000–3000 Hz_40ms_log_rms_mae_db": 0.727159322035903, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06623363680905398, "1000–3000 Hz_40ms_envelope_correlation": 0.9966608596791148, "1000–3000 Hz_40ms_modulation_mae_db": 0.8291004490483772, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4005591048100532, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.188130420180379, "3000–6000 Hz_10ms_envelope_correlation": 0.974581108682065, "3000–6000 Hz_10ms_modulation_mae_db": 1.082940231816293, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9432421832003642, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11818347615740515, "3000–6000 Hz_40ms_envelope_correlation": 0.9892513695326451, "3000–6000 Hz_40ms_modulation_mae_db": 1.200028348431749, "mr_stft_db": 5.310709852944279, "band_mae_db": 0.19425925763550647, "rms_gain_error_db": -0.0877700518961805, "output_peak": 0.24160040643760852}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6862815424378074, "全帯域_10ms_rms_envelope_nrmse": 0.04748152132521177, "全帯域_10ms_envelope_correlation": 0.9975191510191084, "全帯域_10ms_modulation_mae_db": 0.7164321949186512, "全帯域_40ms_log_rms_mae_db": 0.370708708943049, "全帯域_40ms_rms_envelope_nrmse": 0.02474649252463262, "全帯域_40ms_envelope_correlation": 0.9993530010164214, "全帯域_40ms_modulation_mae_db": 0.8129173605848418, "300–1000 Hz_10ms_log_rms_mae_db": 0.9247147486909627, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05302133032932262, "300–1000 Hz_10ms_envelope_correlation": 0.9974550595088932, "300–1000 Hz_10ms_modulation_mae_db": 1.040408171353192, "300–1000 Hz_40ms_log_rms_mae_db": 0.5187832991485279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.030845271269693678, "300–1000 Hz_40ms_envelope_correlation": 0.9991791849956948, "300–1000 Hz_40ms_modulation_mae_db": 0.96470749568302, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2599623343414768, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1274136070516129, "1000–3000 Hz_10ms_envelope_correlation": 0.9876092496634024, "1000–3000 Hz_10ms_modulation_mae_db": 0.7385848708564707, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7540885756122263, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07413175174593992, "1000–3000 Hz_40ms_envelope_correlation": 0.9960536380250691, "1000–3000 Hz_40ms_modulation_mae_db": 0.5515099303412053, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4070290458272732, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18699164208870148, "3000–6000 Hz_10ms_envelope_correlation": 0.9753329149175229, "3000–6000 Hz_10ms_modulation_mae_db": 1.2714143306286894, "3000–6000 Hz_40ms_log_rms_mae_db": 0.901891710065428, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11479117642757038, "3000–6000 Hz_40ms_envelope_correlation": 0.9906757622478841, "3000–6000 Hz_40ms_modulation_mae_db": 1.459464251063136, "mr_stft_db": 5.31362201799786, "band_mae_db": 0.16516807019600852, "rms_gain_error_db": -0.08219034987242738, "output_peak": 0.23891209531577026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6998057317696433, "全帯域_10ms_rms_envelope_nrmse": 0.05004438891622072, "全帯域_10ms_envelope_correlation": 0.9972153283183138, "全帯域_10ms_modulation_mae_db": 0.296165810773518, "全帯域_40ms_log_rms_mae_db": 0.3859013195415994, "全帯域_40ms_rms_envelope_nrmse": 0.027648474391638526, "全帯域_40ms_envelope_correlation": 0.9991539071242287, "全帯域_40ms_modulation_mae_db": 0.2170606979277112, "300–1000 Hz_10ms_log_rms_mae_db": 0.9508273219734439, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06108146763400297, "300–1000 Hz_10ms_envelope_correlation": 0.9965825824670226, "300–1000 Hz_10ms_modulation_mae_db": 0.2738507927122661, "300–1000 Hz_40ms_log_rms_mae_db": 0.5002988850314994, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03439777193049863, "300–1000 Hz_40ms_envelope_correlation": 0.9989696465417393, "300–1000 Hz_40ms_modulation_mae_db": 0.18323549899744015, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2936345603084405, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13532786405226324, "1000–3000 Hz_10ms_envelope_correlation": 0.9860482333704473, "1000–3000 Hz_10ms_modulation_mae_db": 1.054158221854702, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8123592547546301, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08294382526985332, "1000–3000 Hz_40ms_envelope_correlation": 0.9950075725672216, "1000–3000 Hz_40ms_modulation_mae_db": 1.0251083114974848, "3000–6000 Hz_10ms_log_rms_mae_db": 1.37517593284982, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17711307217427716, "3000–6000 Hz_10ms_envelope_correlation": 0.9775032178571357, "3000–6000 Hz_10ms_modulation_mae_db": 0.6576458767792794, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8592528944579557, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10115128564803368, "3000–6000 Hz_40ms_envelope_correlation": 0.9922082782042991, "3000–6000 Hz_40ms_modulation_mae_db": 0.6565738480948401, "mr_stft_db": 5.305391031302145, "band_mae_db": 0.189245487702952, "rms_gain_error_db": -0.07928500118551843, "output_peak": 0.2547916620656505}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6978536025886586, "全帯域_10ms_rms_envelope_nrmse": 0.05220674448212049, "全帯域_10ms_envelope_correlation": 0.996990194930646, "全帯域_10ms_modulation_mae_db": 0.6283403359856597, "全帯域_40ms_log_rms_mae_db": 0.3708941074048786, "全帯域_40ms_rms_envelope_nrmse": 0.03263860998601572, "全帯域_40ms_envelope_correlation": 0.9988177091080319, "全帯域_40ms_modulation_mae_db": 0.5752806806893539, "300–1000 Hz_10ms_log_rms_mae_db": 0.9477998840979984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06549639963355548, "300–1000 Hz_10ms_envelope_correlation": 0.9961395674650279, "300–1000 Hz_10ms_modulation_mae_db": 0.8613723040497359, "300–1000 Hz_40ms_log_rms_mae_db": 0.5395076509962392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04246932801685479, "300–1000 Hz_40ms_envelope_correlation": 0.998441543852751, "300–1000 Hz_40ms_modulation_mae_db": 0.8977838168753688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.280023245992324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13859110733800953, "1000–3000 Hz_10ms_envelope_correlation": 0.9849495692863962, "1000–3000 Hz_10ms_modulation_mae_db": 0.48843279777721715, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7788370381710862, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08516138615387767, "1000–3000 Hz_40ms_envelope_correlation": 0.9941735710324309, "1000–3000 Hz_40ms_modulation_mae_db": 0.2968810681739925, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3466808357470992, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17590574005245002, "3000–6000 Hz_10ms_envelope_correlation": 0.977782198956544, "3000–6000 Hz_10ms_modulation_mae_db": 0.7774607305416249, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8624487168398589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10990575850482097, "3000–6000 Hz_40ms_envelope_correlation": 0.9906263838803979, "3000–6000 Hz_40ms_modulation_mae_db": 0.8927678018629918, "mr_stft_db": 5.30998561210716, "band_mae_db": 0.17289293202158837, "rms_gain_error_db": -0.09764941539241291, "output_peak": 0.24276326184797464}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6566705322081441, "全帯域_10ms_rms_envelope_nrmse": 0.049225547019193, "全帯域_10ms_envelope_correlation": 0.9972961785586162, "全帯域_10ms_modulation_mae_db": 0.5108189501011485, "全帯域_40ms_log_rms_mae_db": 0.3511230522240795, "全帯域_40ms_rms_envelope_nrmse": 0.028056165784900753, "全帯域_40ms_envelope_correlation": 0.9991099520679074, "全帯域_40ms_modulation_mae_db": 0.5635861334684816, "300–1000 Hz_10ms_log_rms_mae_db": 0.9688743965178527, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06031038754537285, "300–1000 Hz_10ms_envelope_correlation": 0.9966579312366584, "300–1000 Hz_10ms_modulation_mae_db": 0.5195907619704652, "300–1000 Hz_40ms_log_rms_mae_db": 0.5479477845578661, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03571817325054869, "300–1000 Hz_40ms_envelope_correlation": 0.998845834202433, "300–1000 Hz_40ms_modulation_mae_db": 0.5472896419146424, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2991391222263278, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13752806690361902, "1000–3000 Hz_10ms_envelope_correlation": 0.9857619805108528, "1000–3000 Hz_10ms_modulation_mae_db": 1.1092733607021008, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7748862538873731, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08356526089434246, "1000–3000 Hz_40ms_envelope_correlation": 0.9951772772608637, "1000–3000 Hz_40ms_modulation_mae_db": 1.015354720610209, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3519836023721015, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17728304612321122, "3000–6000 Hz_10ms_envelope_correlation": 0.9780048112981888, "3000–6000 Hz_10ms_modulation_mae_db": 1.1167897756656902, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8436453819034366, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10353162515239739, "3000–6000 Hz_40ms_envelope_correlation": 0.992740639207017, "3000–6000 Hz_40ms_modulation_mae_db": 1.2525226047456597, "mr_stft_db": 5.3116154144104515, "band_mae_db": 0.23819698210392606, "rms_gain_error_db": -0.07602086903600662, "output_peak": 0.26315919348935984}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.184406770737218, "spectral_convergence": 0.13797977999195368, "spectral_convergence_status": "ok", "amplitude_floor": 9.671435620644006e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.24207176689079, "spectral_convergence": 0.14084006330038384, "spectral_convergence_status": "ok", "amplitude_floor": 9.322400769888516e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.565848264124336, "spectral_convergence": 0.1422680923161591, "spectral_convergence_status": "ok", "amplitude_floor": 9.453157582642776e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00016624637287256398, "output_power": 0.00017303459509731043, "delta_db": 0.17380762299972102, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013640231584289317, "output_power": 0.0013320805946158815, "delta_db": -0.10291242184029165, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005975940179232537, "output_power": 0.0005805254038329371, "delta_db": -0.12585012176775556, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.8168186983374496e-05, "output_power": 1.818803605986531e-05, "delta_db": 0.004742156652016348, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.463537251271816e-05, "output_power": 2.2195152211004112e-05, "delta_db": -0.45301005951872264, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0343523773544458e-05, "output_power": 1.0218678448468911e-05, "delta_db": -0.05273783562316053, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1521767030669446, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09007239866701794, "pitch": null, "output_rms_dbfs": -26.703494570837897, "output_peak": 0.23857044476392023, "reference_rms_dbfs": -26.613422172170882, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.330775600584114}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001446111949915804, "log_rms_mae_db": 0.7293994575042108, "rms_envelope_nrmse": 0.053527627572371624, "envelope_correlation": 0.9968095555368692, "rms_gain_error_db": -0.09007326390253088, "reference_mean_power": 0.002182828397481346, "output_mean_power": 0.0021380224872312517}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.790920693994897e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003230276115117931, "output_power": 0.00030978326790264066, "reference_active": true, "delta_db": -0.1818168938955558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.477236494661316e-05, "output_power": 4.0424733033665036e-05, "reference_active": true, "delta_db": -0.4436287461361735}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.548345752003965e-06, "output_power": 8.933716027112743e-06, "reference_active": true, "delta_db": -0.28895993124564207}], "active_band_count": 3, "active_band_mae_db": 0.3048018570924571}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014111493703772434, "log_rms_mae_db": 0.4102253624275308, "rms_envelope_nrmse": 0.03216353037960718, "envelope_correlation": 0.9988340161649127, "rms_gain_error_db": -0.09012852213234877, "reference_mean_power": 0.002193693032291648, "output_mean_power": 0.002148636769965743}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.190420923463322e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028912450266098784, "output_power": 0.00027784731928098773, "reference_active": true, "delta_db": -0.17278687732109568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2978791346744758e-05, "output_power": 2.1180440927959265e-05, "reference_active": true, "delta_db": -0.3539218477671984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.072854493661425e-07, "output_power": 5.177148559407812e-07, "reference_active": true, "delta_db": -0.6930224816256159}], "active_band_count": 3, "active_band_mae_db": 0.40657706890463663}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012712872255296568, "log_rms_mae_db": 1.0063390587841257, "rms_envelope_nrmse": 0.06601723439657356, "envelope_correlation": 0.9960158454259898, "rms_gain_error_db": -0.13762863484930274, "reference_mean_power": 0.0014813690704446063, "output_mean_power": 0.0014351602902318752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.409297669270013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003163260619212235, "output_power": 0.00030457710841037166, "reference_active": true, "delta_db": -0.16437715400365926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6701975061022124e-05, "output_power": 3.2732699936440924e-05, "reference_active": true, "delta_db": -0.4970760648230499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.301526036454895e-06, "output_power": 6.312285729084509e-06, "reference_active": true, "delta_db": 0.007409142226377888}], "active_band_count": 3, "active_band_mae_db": 0.22295412035102902}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012343588785799389, "log_rms_mae_db": 0.5695642982107194, "rms_envelope_nrmse": 0.04142048409700128, "envelope_correlation": 0.9984665380917317, "rms_gain_error_db": -0.1376489766171417, "reference_mean_power": 0.0014887936512772218, "output_mean_power": 0.0014423465181629851}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.99328908394483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002875949770086079, "output_power": 0.0002777619226601754, "reference_active": true, "delta_db": -0.15108586923887937}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97680708793848e-05, "output_power": 1.803068154218204e-05, "reference_active": true, "delta_db": -0.3995214664028355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.520482053574084e-07, "output_power": 4.402801963132196e-07, "reference_active": true, "delta_db": -0.11455597975051995}], "active_band_count": 3, "active_band_mae_db": 0.22172110513074492}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.4084091617788044e-05, "log_rms_mae_db": 1.2865650182623254, "rms_envelope_nrmse": 0.14068912724301386, "envelope_correlation": 0.9850130174601966, "rms_gain_error_db": -0.3276140074083492, "reference_mean_power": 3.6694438258439554e-05, "output_mean_power": 3.402818951605739e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6589829493109366e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3179349061495639e-05, "output_power": 1.1434930229994919e-05, "reference_active": true, "delta_db": -0.6166044152724767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9890642401085437e-06, "output_power": 2.6372631951670007e-06, "reference_active": true, "delta_db": -0.543817749352928}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.136200949313584e-07, "output_power": 5.728596584588066e-07, "reference_active": true, "delta_db": -1.5237342659841642}], "active_band_count": 3, "active_band_mae_db": 0.8947188102031897}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.151462172211685e-05, "log_rms_mae_db": 0.7751908622515639, "rms_envelope_nrmse": 0.09098928676471697, "envelope_correlation": 0.9940010379846639, "rms_gain_error_db": -0.327602287937144, "reference_mean_power": 3.687647047180041e-05, "output_mean_power": 3.419708739670215e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4709890047082147e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2743108138238886e-05, "output_power": 1.0874610315815543e-05, "reference_active": true, "delta_db": -0.6886166529169058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4131557808728274e-06, "output_power": 1.34403754804025e-06, "reference_active": true, "delta_db": -0.2177863778196634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.7131268313465536e-08, "output_power": 4.733397086050965e-08, "reference_active": true, "delta_db": -0.8170092641381165}], "active_band_count": 3, "active_band_mae_db": 0.5744707649582286}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.703026630874025e-05, "log_rms_mae_db": 1.4155550513946407, "rms_envelope_nrmse": 0.17646950425740696, "envelope_correlation": 0.9778128764823277, "rms_gain_error_db": -0.007711492678286962, "reference_mean_power": 1.1971789360385644e-05, "output_mean_power": 1.1950550672211336e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.043568625201122e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.248382965898838e-06, "output_power": 4.174103717248729e-06, "reference_active": true, "delta_db": -0.07660422724359575}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.589400168019184e-06, "output_power": 2.3875441745391885e-06, "reference_active": true, "delta_db": -0.35247756163402594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.486231293405906e-07, "output_power": 4.4882434409263703e-07, "reference_active": true, "delta_db": -1.5991602629481305}], "active_band_count": 3, "active_band_mae_db": 0.6760806839419174}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.3691778537856407e-05, "log_rms_mae_db": 0.9177682813907129, "rms_envelope_nrmse": 0.09862953350338426, "envelope_correlation": 0.9923440005712459, "rms_gain_error_db": -0.007707562328040787, "reference_mean_power": 1.2031928101806957e-05, "output_mean_power": 1.2010593593360285e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.457061965350652e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.507382710758712e-06, "output_power": 4.3063625589866095e-06, "reference_active": true, "delta_db": -0.19813843593979497}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3712782786962106e-06, "output_power": 1.2267517635995016e-06, "reference_active": true, "delta_db": -0.4836890575521041}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.274779912953651e-08, "output_power": 3.082269574326543e-08, "reference_active": true, "delta_db": -1.4204314068595467}], "active_band_count": 3, "active_band_mae_db": 0.7007529667838153}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.941219362627238, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d1de7ebb5b0c1e045dd642c4258327375ef336b7987e6d51d191a619819da63e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9863079943281443, "standard_deviation": 0.0018617179185340437, "minimum": 0.9836671167466404, "maximum": 0.9890174980447711, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.280036767359477, "standard_deviation": 0.015226221952523052, "minimum": 1.2599623343414768, "maximum": 1.2991391222263278, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7740267880477228, "standard_deviation": 0.3146675308987895, "minimum": 0.3764644116542308, "maximum": 1.116467940320727, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13308616184111902, "standard_deviation": 0.009166365204609938, "minimum": 0.11904276349385068, "maximum": 0.14413649751843202, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951264106757539, "standard_deviation": 0.0008847862058988285, "minimum": 0.9940010379846639, "maximum": 0.9966608596791148, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7769301050734692, "standard_deviation": 0.027121912664456082, "minimum": 0.727159322035903, "maximum": 0.8123592547546301, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6166447645099478, "standard_deviation": 0.31755425852584646, "minimum": 0.16051693935184133, "maximum": 1.0251083114974848, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0802098348056606, "standard_deviation": 0.007521134082399528, "minimum": 0.06623363680905398, "maximum": 0.09098928676471697, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9756724077302319, "standard_deviation": 0.0024953138314606884, "minimum": 0.9719858380204907, "maximum": 0.9780048112981888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3885988207466007, "standard_deviation": 0.035250090184613184, "minimum": 1.3466808357470992, "maximum": 1.448399101179836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.868153092444309, "standard_deviation": 0.24798541104838204, "minimum": 0.6576458767792794, "maximum": 1.2714143306286894, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1844871571839541, "standard_deviation": 0.009087268149517852, "minimum": 0.17590574005245002, "maximum": 0.1977316667217976, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906100072379452, "standard_deviation": 0.0019926364376756426, "minimum": 0.9866353254292787, "maximum": 0.992740639207017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8917263561126971, "standard_deviation": 0.03450055210704381, "minimum": 0.8436453819034366, "maximum": 0.9432421832003642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9084709853361885, "standard_deviation": 0.3524072471210048, "minimum": 0.5170704614984651, "maximum": 1.459464251063136, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11153702035297389, "standard_deviation": 0.010984745936099924, "minimum": 0.09862953350338426, "maximum": 0.13263027480192346, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9964998168158528, "standard_deviation": 0.00045289828475201735, "minimum": 0.9960158454259898, "maximum": 0.9974550595088932, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.964867420156438, "standard_deviation": 0.024952394756147047, "minimum": 0.9247147486909627, "maximum": 1.0063390587841257, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6552574862467508, "standard_deviation": 0.2989670290507891, "minimum": 0.22295412035102902, "maximum": 1.040408171353192, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.062039758725671384, "standard_deviation": 0.004236544796315229, "minimum": 0.05302133032932262, "maximum": 0.06601723439657356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9987286402134659, "standard_deviation": 0.0002622872200063421, "minimum": 0.998441543852751, "maximum": 0.9991791849956948, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5465238166411853, "standard_deviation": 0.02618239837515731, "minimum": 0.5002988850314994, "maximum": 0.5764258897853665, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7229158171951073, "standard_deviation": 0.3658738811197571, "minimum": 0.18323549899744015, "maximum": 1.1914566059717788, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03808712642805917, "standard_deviation": 0.004115293834714247, "minimum": 0.030845271269693678, "maximum": 0.04246932801685479, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.19024476445659738, "standard_deviation": 0.02889858182296447, "minimum": 0.1521767030669446, "maximum": 0.23819698210392606, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.312520111144069, "standard_deviation": 0.008749648955023368, "minimum": 5.301424032547698, "maximum": 5.330775600584114, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.24762977721440682, "standard_deviation": 0.008994553317519678, "minimum": 0.23857044476392023, "maximum": 0.26315919348935984, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09101972557852565, "standard_deviation": 0.013340787302805284, "minimum": -0.11760960386033348, "maximum": -0.07602086903600662, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9971594723132567, "standard_deviation": 0.00029847121777782667, "minimum": 0.9967144489801517, "maximum": 0.9975191510191084, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7042004602023051, "standard_deviation": 0.03192896921469928, "minimum": 0.6566705322081441, "maximum": 0.7658396671646794, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5837949207265198, "standard_deviation": 0.22818857544055882, "minimum": 0.296165810773518, "maximum": 1.004129745023378, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.050664291189142126, "standard_deviation": 0.002693016637118269, "minimum": 0.04748152132521177, "maximum": 0.05493639757237985, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9990623619034225, "standard_deviation": 0.0001973287893966158, "minimum": 0.9988177091080319, "maximum": 0.9993530010164214, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.383553431308362, "standard_deviation": 0.020877355057615263, "minimum": 0.3511230522240795, "maximum": 0.4128331476227238, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6263844290386377, "standard_deviation": 0.25413619606433385, "minimum": 0.2170606979277112, "maximum": 1.0576314875241069, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.029181163531995166, "standard_deviation": 0.0029850656427378415, "minimum": 0.02474649252463262, "maximum": 0.032829517481132296, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284447-0007", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 3.0624167919158936, "predicted_mos": 3.0624167919158936, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "dd4bd490550a7f6b3db510a20e3e9b96643386f42831a15e1e98bfa02f46658a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.081785999704152, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6477581392168691, "全帯域_10ms_rms_envelope_nrmse": 0.05120262169267319, "全帯域_10ms_envelope_correlation": 0.9971657381874006, "全帯域_10ms_modulation_mae_db": 0.4658734122827763, "全帯域_40ms_log_rms_mae_db": 0.3685794232381202, "全帯域_40ms_rms_envelope_nrmse": 0.030498244270000942, "全帯域_40ms_envelope_correlation": 0.9990429854267997, "全帯域_40ms_modulation_mae_db": 0.42688442079732963, "300–1000 Hz_10ms_log_rms_mae_db": 1.0476679477110953, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0632603411413291, "300–1000 Hz_10ms_envelope_correlation": 0.9964867084615351, "300–1000 Hz_10ms_modulation_mae_db": 0.6621079253057838, "300–1000 Hz_40ms_log_rms_mae_db": 0.5846218341946022, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0398122889295454, "300–1000 Hz_40ms_envelope_correlation": 0.9987303654591166, "300–1000 Hz_40ms_modulation_mae_db": 0.7271485553932356, "1000–3000 Hz_10ms_log_rms_mae_db": 1.401114254850345, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13159765315148234, "1000–3000 Hz_10ms_envelope_correlation": 0.9865532845086208, "1000–3000 Hz_10ms_modulation_mae_db": 0.7187771252070495, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8188975672153672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08000226962854183, "1000–3000 Hz_40ms_envelope_correlation": 0.9950167286551651, "1000–3000 Hz_40ms_modulation_mae_db": 0.7564064278661097, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5093369237714456, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1724896034268708, "3000–6000 Hz_10ms_envelope_correlation": 0.97871591234077, "3000–6000 Hz_10ms_modulation_mae_db": 0.08834264406701318, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9358909154116238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10778103231282438, "3000–6000 Hz_40ms_envelope_correlation": 0.9907942838251249, "3000–6000 Hz_40ms_modulation_mae_db": 0.24090819932023821, "mr_stft_db": 5.414832487480187, "band_mae_db": 0.1610937901791115, "rms_gain_error_db": -0.10926176821494371, "output_peak": 0.2568336391036624}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6985549911792731, "全帯域_10ms_rms_envelope_nrmse": 0.048807186511825335, "全帯域_10ms_envelope_correlation": 0.9973318041834274, "全帯域_10ms_modulation_mae_db": 0.32178795470205057, "全帯域_40ms_log_rms_mae_db": 0.4073579223904502, "全帯域_40ms_rms_envelope_nrmse": 0.027284433726550095, "全帯域_40ms_envelope_correlation": 0.9991500843410178, "全帯域_40ms_modulation_mae_db": 0.3491107205563235, "300–1000 Hz_10ms_log_rms_mae_db": 1.0426650932999093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06520150493262132, "300–1000 Hz_10ms_envelope_correlation": 0.9961006827628336, "300–1000 Hz_10ms_modulation_mae_db": 0.49187013665032, "300–1000 Hz_40ms_log_rms_mae_db": 0.5767764243813159, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03908252781708716, "300–1000 Hz_40ms_envelope_correlation": 0.9986355355495605, "300–1000 Hz_40ms_modulation_mae_db": 0.5861106340371269, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4046508010904435, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12596244683684782, "1000–3000 Hz_10ms_envelope_correlation": 0.9879043376387843, "1000–3000 Hz_10ms_modulation_mae_db": 0.8179724261032041, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8530872481525827, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07484085262126251, "1000–3000 Hz_40ms_envelope_correlation": 0.9959507081662555, "1000–3000 Hz_40ms_modulation_mae_db": 0.689222830365607, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5508951767044408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19927651747271477, "3000–6000 Hz_10ms_envelope_correlation": 0.9714133460220297, "3000–6000 Hz_10ms_modulation_mae_db": 0.862283359390816, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9303247941754531, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1095777257122546, "3000–6000 Hz_40ms_envelope_correlation": 0.9904950912183659, "3000–6000 Hz_40ms_modulation_mae_db": 0.6593887338294493, "mr_stft_db": 5.441618686054585, "band_mae_db": 0.12152991062212833, "rms_gain_error_db": -0.06685596935331761, "output_peak": 0.24628086906121907}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7197066462493498, "全帯域_10ms_rms_envelope_nrmse": 0.04768979142006625, "全帯域_10ms_envelope_correlation": 0.9975071833905066, "全帯域_10ms_modulation_mae_db": 0.5589185427475133, "全帯域_40ms_log_rms_mae_db": 0.3816157951329032, "全帯域_40ms_rms_envelope_nrmse": 0.026577186851995314, "全帯域_40ms_envelope_correlation": 0.9992585441815393, "全帯域_40ms_modulation_mae_db": 0.6468606125886375, "300–1000 Hz_10ms_log_rms_mae_db": 1.048813387533959, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06016338869651006, "300–1000 Hz_10ms_envelope_correlation": 0.9967263491220201, "300–1000 Hz_10ms_modulation_mae_db": 0.4739929825493314, "300–1000 Hz_40ms_log_rms_mae_db": 0.5571484297148828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03728051003563356, "300–1000 Hz_40ms_envelope_correlation": 0.9987957147950441, "300–1000 Hz_40ms_modulation_mae_db": 0.6150231241188504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4050984796490156, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12518220171525749, "1000–3000 Hz_10ms_envelope_correlation": 0.9877199144949057, "1000–3000 Hz_10ms_modulation_mae_db": 0.5268979122388296, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8573612933394781, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0792720335244929, "1000–3000 Hz_40ms_envelope_correlation": 0.9948632668062262, "1000–3000 Hz_40ms_modulation_mae_db": 0.5129674956889163, "3000–6000 Hz_10ms_log_rms_mae_db": 1.52934145494968, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1977014415255288, "3000–6000 Hz_10ms_envelope_correlation": 0.9718212550826246, "3000–6000 Hz_10ms_modulation_mae_db": 0.8096729972691827, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9555209005246735, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12094847867469082, "3000–6000 Hz_40ms_envelope_correlation": 0.9885945854384155, "3000–6000 Hz_40ms_modulation_mae_db": 0.8511469837977031, "mr_stft_db": 5.422283621816306, "band_mae_db": 0.21702930253294886, "rms_gain_error_db": -0.09334011017810255, "output_peak": 0.24114249989438555}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7013806425468976, "全帯域_10ms_rms_envelope_nrmse": 0.048567595156509206, "全帯域_10ms_envelope_correlation": 0.9973605619015244, "全帯域_10ms_modulation_mae_db": 0.3296768411067686, "全帯域_40ms_log_rms_mae_db": 0.36334216464839675, "全帯域_40ms_rms_envelope_nrmse": 0.025384623341991527, "全帯域_40ms_envelope_correlation": 0.999279006253902, "全帯域_40ms_modulation_mae_db": 0.4396581080433599, "300–1000 Hz_10ms_log_rms_mae_db": 1.0217311837143928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06102179758244277, "300–1000 Hz_10ms_envelope_correlation": 0.9965710737859504, "300–1000 Hz_10ms_modulation_mae_db": 0.49144958165683955, "300–1000 Hz_40ms_log_rms_mae_db": 0.5681303009610974, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03594515242309333, "300–1000 Hz_40ms_envelope_correlation": 0.9988332139886807, "300–1000 Hz_40ms_modulation_mae_db": 0.5477911663580527, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4543231623793784, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13828652594608162, "1000–3000 Hz_10ms_envelope_correlation": 0.9848901377312147, "1000–3000 Hz_10ms_modulation_mae_db": 0.6972034605736548, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8773589329362318, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07680890758940086, "1000–3000 Hz_40ms_envelope_correlation": 0.9950897856463327, "1000–3000 Hz_40ms_modulation_mae_db": 0.5056511822049389, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5634757750008024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19616846393206658, "3000–6000 Hz_10ms_envelope_correlation": 0.9723022988390686, "3000–6000 Hz_10ms_modulation_mae_db": 1.00919593613447, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9825094786893407, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11936855484628935, "3000–6000 Hz_40ms_envelope_correlation": 0.9889145318012548, "3000–6000 Hz_40ms_modulation_mae_db": 1.347488549146581, "mr_stft_db": 5.431580360873862, "band_mae_db": 0.1933956695258188, "rms_gain_error_db": -0.07251339443403405, "output_peak": 0.2437025128165458}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6902951119751174, "全帯域_10ms_rms_envelope_nrmse": 0.048101835939715044, "全帯域_10ms_envelope_correlation": 0.9974213572801142, "全帯域_10ms_modulation_mae_db": 0.5109976612146655, "全帯域_40ms_log_rms_mae_db": 0.3762489958935811, "全帯域_40ms_rms_envelope_nrmse": 0.027074549245087592, "全帯域_40ms_envelope_correlation": 0.9991714892515996, "全帯域_40ms_modulation_mae_db": 0.40667645957959625, "300–1000 Hz_10ms_log_rms_mae_db": 1.0457561166330633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06333730873866215, "300–1000 Hz_10ms_envelope_correlation": 0.9962785924476651, "300–1000 Hz_10ms_modulation_mae_db": 0.6385271069126824, "300–1000 Hz_40ms_log_rms_mae_db": 0.5707914217526496, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04024412554951129, "300–1000 Hz_40ms_envelope_correlation": 0.9984745872403102, "300–1000 Hz_40ms_modulation_mae_db": 0.6568781246600331, "1000–3000 Hz_10ms_log_rms_mae_db": 1.443386606896581, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1360155243867557, "1000–3000 Hz_10ms_envelope_correlation": 0.9855110607074852, "1000–3000 Hz_10ms_modulation_mae_db": 1.216506584382279, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8277995891996701, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07267472198938672, "1000–3000 Hz_40ms_envelope_correlation": 0.9958544175129028, "1000–3000 Hz_40ms_modulation_mae_db": 1.0549184959481188, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5695389898308176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18340509046681766, "3000–6000 Hz_10ms_envelope_correlation": 0.9765734624213658, "3000–6000 Hz_10ms_modulation_mae_db": 0.11604736006624193, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9839664937527247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10882489746552813, "3000–6000 Hz_40ms_envelope_correlation": 0.9909927323101796, "3000–6000 Hz_40ms_modulation_mae_db": 0.19289751294120708, "mr_stft_db": 5.452677544044509, "band_mae_db": 0.1789202447480748, "rms_gain_error_db": -0.07264497433574385, "output_peak": 0.2525130040110156}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7103332693638565, "全帯域_10ms_rms_envelope_nrmse": 0.04958440331111651, "全帯域_10ms_envelope_correlation": 0.9972950372083085, "全帯域_10ms_modulation_mae_db": 0.4135398405953284, "全帯域_40ms_log_rms_mae_db": 0.38494051013542224, "全帯域_40ms_rms_envelope_nrmse": 0.029603536933537667, "全帯域_40ms_envelope_correlation": 0.9990359219925744, "全帯域_40ms_modulation_mae_db": 0.5957986951500239, "300–1000 Hz_10ms_log_rms_mae_db": 1.040544512126367, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05873570066788641, "300–1000 Hz_10ms_envelope_correlation": 0.9968780750683914, "300–1000 Hz_10ms_modulation_mae_db": 0.4597544293917563, "300–1000 Hz_40ms_log_rms_mae_db": 0.6354178170724976, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03832022284248649, "300–1000 Hz_40ms_envelope_correlation": 0.9986934735342496, "300–1000 Hz_40ms_modulation_mae_db": 0.6341002576268039, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4681504548025115, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14878428212114173, "1000–3000 Hz_10ms_envelope_correlation": 0.9826875109108966, "1000–3000 Hz_10ms_modulation_mae_db": 0.5570175476549618, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8703643498101888, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09115823990375457, "1000–3000 Hz_40ms_envelope_correlation": 0.9934047924190883, "1000–3000 Hz_40ms_modulation_mae_db": 0.49850239039381633, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5754836179301488, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19076120163302643, "3000–6000 Hz_10ms_envelope_correlation": 0.9739756288606773, "3000–6000 Hz_10ms_modulation_mae_db": 0.6180792418956264, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9933170941345663, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10699695280591631, "3000–6000 Hz_40ms_envelope_correlation": 0.9909305869787183, "3000–6000 Hz_40ms_modulation_mae_db": 0.6721193248675448, "mr_stft_db": 5.450475494214904, "band_mae_db": 0.15094952684960175, "rms_gain_error_db": -0.0886454758540836, "output_peak": 0.258317747330818}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7246022810689405, "全帯域_10ms_rms_envelope_nrmse": 0.04945661977990415, "全帯域_10ms_envelope_correlation": 0.9973388770521364, "全帯域_10ms_modulation_mae_db": 0.6332834519583885, "全帯域_40ms_log_rms_mae_db": 0.4033941288660735, "全帯域_40ms_rms_envelope_nrmse": 0.027434371533721243, "全帯域_40ms_envelope_correlation": 0.9992275663918586, "全帯域_40ms_modulation_mae_db": 0.6443347550186055, "300–1000 Hz_10ms_log_rms_mae_db": 1.0666658411189545, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.061089877920461305, "300–1000 Hz_10ms_envelope_correlation": 0.9966454521661651, "300–1000 Hz_10ms_modulation_mae_db": 0.8648883159909885, "300–1000 Hz_40ms_log_rms_mae_db": 0.5918027810057839, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03591468008967326, "300–1000 Hz_40ms_envelope_correlation": 0.9989231883319911, "300–1000 Hz_40ms_modulation_mae_db": 0.7291737538400151, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4275423836611154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13249880091829053, "1000–3000 Hz_10ms_envelope_correlation": 0.9864927749468212, "1000–3000 Hz_10ms_modulation_mae_db": 0.7492450333376469, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8285063173363807, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07631008931492017, "1000–3000 Hz_40ms_envelope_correlation": 0.9957104555328716, "1000–3000 Hz_40ms_modulation_mae_db": 0.4636381601819725, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5704819749623486, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18086126392254867, "3000–6000 Hz_10ms_envelope_correlation": 0.9776601255338886, "3000–6000 Hz_10ms_modulation_mae_db": 1.320948320737356, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9765733305738227, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1078541925733587, "3000–6000 Hz_40ms_envelope_correlation": 0.9928113020831744, "3000–6000 Hz_40ms_modulation_mae_db": 1.4859872194863646, "mr_stft_db": 5.440637904516412, "band_mae_db": 0.2091322110732046, "rms_gain_error_db": -0.09830124358230896, "output_peak": 0.24320149937206045}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7106894469627731, "全帯域_10ms_rms_envelope_nrmse": 0.05167063985500401, "全帯域_10ms_envelope_correlation": 0.9970377864138513, "全帯域_10ms_modulation_mae_db": 0.6419538745972959, "全帯域_40ms_log_rms_mae_db": 0.39162381625786774, "全帯域_40ms_rms_envelope_nrmse": 0.03145812430947215, "全帯域_40ms_envelope_correlation": 0.9988936002471008, "全帯域_40ms_modulation_mae_db": 0.8177696266368031, "300–1000 Hz_10ms_log_rms_mae_db": 1.0246541751936975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0626673962482837, "300–1000 Hz_10ms_envelope_correlation": 0.9963804355438244, "300–1000 Hz_10ms_modulation_mae_db": 0.5269926194402803, "300–1000 Hz_40ms_log_rms_mae_db": 0.565103599610773, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041238906114531515, "300–1000 Hz_40ms_envelope_correlation": 0.9984263086734736, "300–1000 Hz_40ms_modulation_mae_db": 0.8460783303521716, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3872270859811717, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13228127322772032, "1000–3000 Hz_10ms_envelope_correlation": 0.9863102243466976, "1000–3000 Hz_10ms_modulation_mae_db": 0.5462375259586555, "1000–3000 Hz_40ms_log_rms_mae_db": 0.800321551095379, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06736067842884444, "1000–3000 Hz_40ms_envelope_correlation": 0.9964976775541583, "1000–3000 Hz_40ms_modulation_mae_db": 0.6971052720452046, "3000–6000 Hz_10ms_log_rms_mae_db": 1.562872634053464, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1829632557208333, "3000–6000 Hz_10ms_envelope_correlation": 0.9759590916441231, "3000–6000 Hz_10ms_modulation_mae_db": 1.2139285572771223, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9502490482948363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10780923533548173, "3000–6000 Hz_40ms_envelope_correlation": 0.9909944878217853, "3000–6000 Hz_40ms_modulation_mae_db": 1.3682241743556451, "mr_stft_db": 5.43127029551733, "band_mae_db": 0.14045343916690303, "rms_gain_error_db": -0.09115316800479568, "output_peak": 0.2578912505124017}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.296519068503451, "spectral_convergence": 0.13896561255739784, "spectral_convergence_status": "ok", "amplitude_floor": 9.671435620644006e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.35207453783983, "spectral_convergence": 0.13429910566837772, "spectral_convergence_status": "ok", "amplitude_floor": 9.322400769888516e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.595903856097283, "spectral_convergence": 0.1368748391462512, "spectral_convergence_status": "ok", "amplitude_floor": 9.453157582642776e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00016624637287256398, "output_power": 0.00017368346498190023, "delta_db": 0.1900629596428187, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013640231584289317, "output_power": 0.0013336392474748646, "delta_db": -0.09783376051728755, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005975940179232537, "output_power": 0.0005681845856653844, "delta_db": -0.21916793894904388, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.8168186983374496e-05, "output_power": 1.8305176048163964e-05, "delta_db": 0.032623191132367654, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.463537251271816e-05, "output_power": 2.2946603004898686e-05, "delta_db": -0.30840731594195564, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0343523773544458e-05, "output_power": 1.0065184948248952e-05, "delta_db": -0.11846757489119543, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.1610937901791115, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10926176821494371, "pitch": null, "output_rms_dbfs": -26.722683940385824, "output_peak": 0.2568336391036624, "reference_rms_dbfs": -26.613422172170882, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.414832487480187}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001446111949915804, "log_rms_mae_db": 0.6477581392168691, "rms_envelope_nrmse": 0.05120262169267319, "envelope_correlation": 0.9971657381874006, "rms_gain_error_db": -0.10926361509647183, "reference_mean_power": 0.002182828397481346, "output_mean_power": 0.0021285959602311026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.790920693994897e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003230276115117931, "output_power": 0.00030444798950819197, "reference_active": true, "delta_db": -0.257265357783558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.477236494661316e-05, "output_power": 4.188539439102776e-05, "reference_active": true, "delta_db": -0.2894742601037542}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.548345752003965e-06, "output_power": 7.84946531620508e-06, "reference_active": true, "delta_db": -0.8508806189610164}], "active_band_count": 3, "active_band_mae_db": 0.4658734122827763}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014111493703772434, "log_rms_mae_db": 0.3685794232381202, "rms_envelope_nrmse": 0.030498244270000942, "envelope_correlation": 0.9990429854267997, "rms_gain_error_db": -0.10926892300580524, "reference_mean_power": 0.002193693032291648, "output_mean_power": 0.0021391880483638596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.190420923463322e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028912450266098784, "output_power": 0.0002729488112059527, "reference_active": true, "delta_db": -0.25003691594037086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2978791346744758e-05, "output_power": 2.165473946956628e-05, "reference_active": true, "delta_db": -0.25774218597367515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.072854493661425e-07, "output_power": 5.082828535752341e-07, "reference_active": true, "delta_db": -0.7728741604779428}], "active_band_count": 3, "active_band_mae_db": 0.42688442079732963}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012712872255296568, "log_rms_mae_db": 1.0476679477110953, "rms_envelope_nrmse": 0.0632603411413291, "envelope_correlation": 0.9964867084615351, "rms_gain_error_db": -0.1656879883336522, "reference_mean_power": 0.0014813690704446063, "output_mean_power": 0.0014259177458243648}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.409297669270013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003163260619212235, "output_power": 0.0002973003113495914, "reference_active": true, "delta_db": -0.2693961063813472}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6701975061022124e-05, "output_power": 3.2447301559837915e-05, "reference_active": true, "delta_db": -0.5351085071006351}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.301526036454895e-06, "output_power": 4.800249585157638e-06, "reference_active": true, "delta_db": -1.181819162435369}], "active_band_count": 3, "active_band_mae_db": 0.6621079253057838}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012343588785799389, "log_rms_mae_db": 0.5846218341946022, "rms_envelope_nrmse": 0.0398122889295454, "envelope_correlation": 0.9987303654591166, "rms_gain_error_db": -0.16569892347807672, "reference_mean_power": 0.0014887936512772218, "output_mean_power": 0.0014330607978280522}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.99328908394483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002875949770086079, "output_power": 0.00027138541826945376, "reference_active": true, "delta_db": -0.2519478760402139}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97680708793848e-05, "output_power": 1.7762994719496412e-05, "reference_active": true, "delta_db": -0.4644810288435508}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.520482053574084e-07, "output_power": 3.2261389852700003e-07, "reference_active": true, "delta_db": -1.4650167612959422}], "active_band_count": 3, "active_band_mae_db": 0.7271485553932356}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.4084091617788044e-05, "log_rms_mae_db": 1.401114254850345, "rms_envelope_nrmse": 0.13159765315148234, "envelope_correlation": 0.9865532845086208, "rms_gain_error_db": -0.2119628631603834, "reference_mean_power": 3.6694438258439554e-05, "output_mean_power": 3.494652180602504e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6589829493109366e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3179349061495639e-05, "output_power": 1.1619513559526936e-05, "reference_active": true, "delta_db": -0.5470601355131415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9890642401085437e-06, "output_power": 2.9508829370861257e-06, "reference_active": true, "delta_db": -0.05583267696272444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.136200949313584e-07, "output_power": 5.689548580419262e-07, "reference_active": true, "delta_db": -1.5534385631452827}], "active_band_count": 3, "active_band_mae_db": 0.7187771252070495}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.151462172211685e-05, "log_rms_mae_db": 0.8188975672153672, "rms_envelope_nrmse": 0.08000226962854183, "envelope_correlation": 0.9950167286551651, "rms_gain_error_db": -0.21211566550855934, "reference_mean_power": 3.687647047180041e-05, "output_mean_power": 3.511864739190181e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4709890047082147e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2743108138238886e-05, "output_power": 1.1162747091007606e-05, "reference_active": true, "delta_db": -0.5750428330156516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4131557808728274e-06, "output_power": 1.5051203510189812e-06, "reference_active": true, "delta_db": 0.27381188520148025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.7131268313465536e-08, "output_power": 4.119432670275483e-08, "reference_active": true, "delta_db": -1.4203645653811972}], "active_band_count": 3, "active_band_mae_db": 0.7564064278661097}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.703026630874025e-05, "log_rms_mae_db": 1.5093369237714456, "rms_envelope_nrmse": 0.1724896034268708, "envelope_correlation": 0.97871591234077, "rms_gain_error_db": -0.030391843553271885, "reference_mean_power": 1.1971789360385644e-05, "output_mean_power": 1.1888303467828566e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.043568625201122e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.248382965898838e-06, "output_power": 4.150491792886269e-06, "reference_active": true, "delta_db": -0.10124099414781157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.589400168019184e-06, "output_power": 2.5344939371494505e-06, "reference_active": true, "delta_db": -0.09307915176015905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.486231293405906e-07, "output_power": 6.592698423257627e-07, "reference_active": true, "delta_db": 0.07070778629306892}], "active_band_count": 3, "active_band_mae_db": 0.08834264406701318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.3691778537856407e-05, "log_rms_mae_db": 0.9358909154116238, "rms_envelope_nrmse": 0.10778103231282438, "envelope_correlation": 0.9907942838251249, "rms_gain_error_db": -0.030387235050259656, "reference_mean_power": 1.2031928101806957e-05, "output_mean_power": 1.1948035507240548e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.457061965350652e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.507382710758712e-06, "output_power": 4.4395040602491026e-06, "reference_active": true, "delta_db": -0.06589976992361794}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3712782786962106e-06, "output_power": 1.3854223597690585e-06, "reference_active": true, "delta_db": 0.044565958153477817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.274779912953651e-08, "output_power": 4.921977759645335e-08, "reference_active": true, "delta_db": 0.6122588698836189}], "active_band_count": 3, "active_band_mae_db": 0.24090819932023821}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9434937798487317, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "547413d7cd96801508a1e21b9721101faf80145e3c1dc424f809e1e884c087cb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9860086556606782, "standard_deviation": 0.001676417875046533, "minimum": 0.9826875109108966, "maximum": 0.9879043376387843, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4239366536638203, "standard_deviation": 0.028945212407889193, "minimum": 1.3872270859811717, "maximum": 1.4681504548025115, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7287322019320351, "standard_deviation": 0.22390333273517865, "minimum": 0.5268979122388296, "maximum": 1.216506584382279, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1338260885379472, "standard_deviation": 0.007503374457313445, "minimum": 0.12518220171525749, "maximum": 0.14878428212114173, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995298479036625, "standard_deviation": 0.0009442054138185065, "minimum": 0.9934047924190883, "maximum": 0.9964976775541583, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8417121061356598, "standard_deviation": 0.026906107810504162, "minimum": 0.800321551095379, "maximum": 0.8773589329362318, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6473015318368356, "standard_deviation": 0.19866596120062585, "minimum": 0.4636381601819725, "maximum": 1.0549184959481188, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07730347412507549, "standard_deviation": 0.006873278535246422, "minimum": 0.06736067842884444, "maximum": 0.09115823990375457, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9748026400930685, "standard_deviation": 0.002809772219616387, "minimum": 0.9714133460220297, "maximum": 0.97871591234077, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5539283184003934, "standard_deviation": 0.023162804338551272, "minimum": 1.5093369237714456, "maximum": 1.5754836179301488, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7548123021047286, "standard_deviation": 0.46004143767763006, "minimum": 0.08834264406701318, "maximum": 1.320948320737356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18795335476255087, "standard_deviation": 0.009514082692957802, "minimum": 0.1724896034268708, "maximum": 0.19927651747271477, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9905659501846273, "standard_deviation": 0.0013207666341450424, "minimum": 0.9885945854384155, "maximum": 0.9928113020831744, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9635440069446302, "standard_deviation": 0.02373154263675951, "minimum": 0.9303247941754531, "maximum": 0.9933170941345663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8522700872180917, "standard_deviation": 0.505857358989024, "minimum": 0.19289751294120708, "maximum": 1.4859872194863646, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.111145133715793, "standard_deviation": 0.005632223441900967, "minimum": 0.10699695280591631, "maximum": 0.12094847867469082, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9965084211697981, "standard_deviation": 0.00025149923418463157, "minimum": 0.9961006827628336, "maximum": 0.9968780750683914, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0423122821664297, "standard_deviation": 0.014215400079174946, "minimum": 1.0217311837143928, "maximum": 1.0666658411189545, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5761978872372477, "standard_deviation": 0.13904118327950687, "minimum": 0.4597544293917563, "maximum": 0.8648883159909885, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0619346644910246, "standard_deviation": 0.002065075217935138, "minimum": 0.05873570066788641, "maximum": 0.06520150493262132, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9986890484465534, "standard_deviation": 0.00017179680847033345, "minimum": 0.9984263086734736, "maximum": 0.9989231883319911, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5812240760867002, "standard_deviation": 0.02449487018942749, "minimum": 0.5571484297148828, "maximum": 0.6354178170724976, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6677879932982862, "standard_deviation": 0.09575846243529476, "minimum": 0.5477911663580527, "maximum": 0.8460783303521716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.038479801725195245, "standard_deviation": 0.0019760118039712157, "minimum": 0.03591468008967326, "maximum": 0.041238906114531515, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.17156301183722394, "standard_deviation": 0.033837973666755845, "minimum": 0.12152991062212833, "maximum": 0.21702930253294886, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.435672049314762, "standard_deviation": 0.013190936918111757, "minimum": 5.414832487480187, "maximum": 5.452677544044509, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.24998537776276358, "standard_deviation": 0.007197559666435982, "minimum": 0.24114249989438555, "maximum": 0.258317747330818, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08658951299466625, "standard_deviation": 0.01465713039339504, "minimum": -0.10926176821494371, "maximum": -0.06685596935331761, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9973072932021587, "standard_deviation": 0.00014649452643311864, "minimum": 0.9970377864138513, "maximum": 0.9975071833905066, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7004150660703846, "standard_deviation": 0.02402581268897405, "minimum": 0.6477581392168691, "maximum": 0.7246022810689405, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.48450394740059843, "standard_deviation": 0.12478335388049534, "minimum": 0.32178795470205057, "maximum": 0.6419538745972959, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.04938508670835171, "standard_deviation": 0.0014190062125121591, "minimum": 0.04768979142006625, "maximum": 0.05167063985500401, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.999132399760799, "standard_deviation": 0.00013235849029287792, "minimum": 0.9988936002471008, "maximum": 0.999279006253902, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3846378445703519, "standard_deviation": 0.015619493258390252, "minimum": 0.36334216464839675, "maximum": 0.4073579223904502, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.540886674796335, "standard_deviation": 0.16019881171710207, "minimum": 0.3491107205563235, "maximum": 0.8177696266368031, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.028164383776544566, "standard_deviation": 0.0021080805329008294, "minimum": 0.025384623341991527, "maximum": 0.03145812430947215, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284447-0007", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 3.2054636478424072, "predicted_mos": 3.2054636478424072, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "db0dfa0f4197bacacdfbb4df060b6738bd324e39ce94b95282514011b54eae53", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.646076099947095, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6938344463749831, "全帯域_10ms_rms_envelope_nrmse": 0.05526299503613511, "全帯域_10ms_envelope_correlation": 0.9966221109094929, "全帯域_10ms_modulation_mae_db": 0.45198103050253485, "全帯域_40ms_log_rms_mae_db": 0.34630001245237846, "全帯域_40ms_rms_envelope_nrmse": 0.02962799518968627, "全帯域_40ms_envelope_correlation": 0.999064592678049, "全帯域_40ms_modulation_mae_db": 0.46118951866224633, "300–1000 Hz_10ms_log_rms_mae_db": 1.0286594610736983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06505963266891254, "300–1000 Hz_10ms_envelope_correlation": 0.9961624369266349, "300–1000 Hz_10ms_modulation_mae_db": 0.5251784325252568, "300–1000 Hz_40ms_log_rms_mae_db": 0.552229561513316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03706366186947246, "300–1000 Hz_40ms_envelope_correlation": 0.9988506153880528, "300–1000 Hz_40ms_modulation_mae_db": 0.6650701259127821, "1000–3000 Hz_10ms_log_rms_mae_db": 1.407235040165869, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1428216341015746, "1000–3000 Hz_10ms_envelope_correlation": 0.9840696249908355, "1000–3000 Hz_10ms_modulation_mae_db": 0.8657443142285031, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8357459150743828, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08440738567721344, "1000–3000 Hz_40ms_envelope_correlation": 0.9944074890777038, "1000–3000 Hz_40ms_modulation_mae_db": 0.5761151986312305, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5611796686832096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1875429653968853, "3000–6000 Hz_10ms_envelope_correlation": 0.9748459245995141, "3000–6000 Hz_10ms_modulation_mae_db": 0.6137934971351354, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9567278843859733, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11856263471795798, "3000–6000 Hz_40ms_envelope_correlation": 0.9888814975308227, "3000–6000 Hz_40ms_modulation_mae_db": 0.787433609232766, "mr_stft_db": 5.427478976223806, "band_mae_db": 0.152462268185697, "rms_gain_error_db": -0.10009196549144289, "output_peak": 0.25079681014410027}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7359519024320946, "全帯域_10ms_rms_envelope_nrmse": 0.051278060499014164, "全帯域_10ms_envelope_correlation": 0.9971037253024568, "全帯域_10ms_modulation_mae_db": 0.3464307361850672, "全帯域_40ms_log_rms_mae_db": 0.39990032913017104, "全帯域_40ms_rms_envelope_nrmse": 0.02858506331057908, "全帯域_40ms_envelope_correlation": 0.9991301190030648, "全帯域_40ms_modulation_mae_db": 0.3588941364341239, "300–1000 Hz_10ms_log_rms_mae_db": 1.061390974752533, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06101063535501452, "300–1000 Hz_10ms_envelope_correlation": 0.9966462033262727, "300–1000 Hz_10ms_modulation_mae_db": 0.36074311189411884, "300–1000 Hz_40ms_log_rms_mae_db": 0.5503000876877931, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.034497065909477716, "300–1000 Hz_40ms_envelope_correlation": 0.9990301521164003, "300–1000 Hz_40ms_modulation_mae_db": 0.42952612739282453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3359335983584288, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12964359144491144, "1000–3000 Hz_10ms_envelope_correlation": 0.9868072798199196, "1000–3000 Hz_10ms_modulation_mae_db": 0.37547011004827485, "1000–3000 Hz_40ms_log_rms_mae_db": 0.763683992085459, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07038132584832023, "1000–3000 Hz_40ms_envelope_correlation": 0.9960335795027032, "1000–3000 Hz_40ms_modulation_mae_db": 0.35498503540872234, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5394007898745423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19392788613922685, "3000–6000 Hz_10ms_envelope_correlation": 0.9730908196505362, "3000–6000 Hz_10ms_modulation_mae_db": 0.7478493450000706, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0072280330173207, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1304460620030164, "3000–6000 Hz_40ms_envelope_correlation": 0.986946118754848, "3000–6000 Hz_40ms_modulation_mae_db": 0.8101733105770602, "mr_stft_db": 5.421781875396966, "band_mae_db": 0.1710381978261831, "rms_gain_error_db": -0.09555027356482423, "output_peak": 0.24610673743937428}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6838463355598471, "全帯域_10ms_rms_envelope_nrmse": 0.04685019178771773, "全帯域_10ms_envelope_correlation": 0.9976024763746639, "全帯域_10ms_modulation_mae_db": 0.5637070208880844, "全帯域_40ms_log_rms_mae_db": 0.36274389035513255, "全帯域_40ms_rms_envelope_nrmse": 0.024162789425786223, "全帯域_40ms_envelope_correlation": 0.9994110172684818, "全帯域_40ms_modulation_mae_db": 0.6871306038092766, "300–1000 Hz_10ms_log_rms_mae_db": 1.0008985716582688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05618659876502845, "300–1000 Hz_10ms_envelope_correlation": 0.9971579045434763, "300–1000 Hz_10ms_modulation_mae_db": 0.6940006491484603, "300–1000 Hz_40ms_log_rms_mae_db": 0.5620378946152026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.032004865723744017, "300–1000 Hz_40ms_envelope_correlation": 0.9991563304304982, "300–1000 Hz_40ms_modulation_mae_db": 0.8139638711146043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3855194635542218, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13857625973664897, "1000–3000 Hz_10ms_envelope_correlation": 0.9854688487017272, "1000–3000 Hz_10ms_modulation_mae_db": 0.6775217078522053, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8192982106799847, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09179614301981853, "1000–3000 Hz_40ms_envelope_correlation": 0.993803263677469, "1000–3000 Hz_40ms_modulation_mae_db": 0.4241537165992009, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5294120714472768, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19184143307846635, "3000–6000 Hz_10ms_envelope_correlation": 0.9741372034408678, "3000–6000 Hz_10ms_modulation_mae_db": 0.62174725286717, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9405936454102978, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11775201741921247, "3000–6000 Hz_40ms_envelope_correlation": 0.9903873288968372, "3000–6000 Hz_40ms_modulation_mae_db": 0.6987278939994982, "mr_stft_db": 5.414216301797136, "band_mae_db": 0.2385361961239293, "rms_gain_error_db": -0.08877280609271564, "output_peak": 0.24604529374799824}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6931103224107987, "全帯域_10ms_rms_envelope_nrmse": 0.051859181485252405, "全帯域_10ms_envelope_correlation": 0.996978084359444, "全帯域_10ms_modulation_mae_db": 0.22937347748927092, "全帯域_40ms_log_rms_mae_db": 0.37981233171765194, "全帯域_40ms_rms_envelope_nrmse": 0.027236997034954132, "全帯域_40ms_envelope_correlation": 0.9991565857881946, "全帯域_40ms_modulation_mae_db": 0.31642364929955963, "300–1000 Hz_10ms_log_rms_mae_db": 1.0168969574572222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05878538125982505, "300–1000 Hz_10ms_envelope_correlation": 0.9968254721165763, "300–1000 Hz_10ms_modulation_mae_db": 0.49205618089068354, "300–1000 Hz_40ms_log_rms_mae_db": 0.5707943491622743, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.033944049161626896, "300–1000 Hz_40ms_envelope_correlation": 0.9989734414470441, "300–1000 Hz_40ms_modulation_mae_db": 0.4900791419585305, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3540042721557815, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1343020420857578, "1000–3000 Hz_10ms_envelope_correlation": 0.9860080951108262, "1000–3000 Hz_10ms_modulation_mae_db": 0.8416049267121006, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7841361420384724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07471864498394139, "1000–3000 Hz_40ms_envelope_correlation": 0.9957672347465737, "1000–3000 Hz_40ms_modulation_mae_db": 0.7132489511920307, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5586803956756587, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18915242133864382, "3000–6000 Hz_10ms_envelope_correlation": 0.9749366021944992, "3000–6000 Hz_10ms_modulation_mae_db": 0.8048847950687165, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9666404389963413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11041138609026295, "3000–6000 Hz_40ms_envelope_correlation": 0.9918029686157521, "3000–6000 Hz_40ms_modulation_mae_db": 0.8266553975988925, "mr_stft_db": 5.4093063099126155, "band_mae_db": 0.179711109517368, "rms_gain_error_db": -0.07120233070743273, "output_peak": 0.2514148643979124}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6978086036762924, "全帯域_10ms_rms_envelope_nrmse": 0.05196383018166332, "全帯域_10ms_envelope_correlation": 0.9969884817903703, "全帯域_10ms_modulation_mae_db": 0.8495924730666031, "全帯域_40ms_log_rms_mae_db": 0.39005314407918135, "全帯域_40ms_rms_envelope_nrmse": 0.02556257298408316, "全帯域_40ms_envelope_correlation": 0.9992936789911957, "全帯域_40ms_modulation_mae_db": 0.8567144668364918, "300–1000 Hz_10ms_log_rms_mae_db": 1.0329636316398265, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05999051461905804, "300–1000 Hz_10ms_envelope_correlation": 0.9967034947174148, "300–1000 Hz_10ms_modulation_mae_db": 0.8866007571773994, "300–1000 Hz_40ms_log_rms_mae_db": 0.5645030751464172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03225144739309132, "300–1000 Hz_40ms_envelope_correlation": 0.9991058103001066, "300–1000 Hz_40ms_modulation_mae_db": 1.1754543360080716, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3276019621962862, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12145431434739354, "1000–3000 Hz_10ms_envelope_correlation": 0.9884425590517473, "1000–3000 Hz_10ms_modulation_mae_db": 0.6467577058379238, "1000–3000 Hz_40ms_log_rms_mae_db": 0.768257752742205, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0651542301725054, "1000–3000 Hz_40ms_envelope_correlation": 0.9966126730017792, "1000–3000 Hz_40ms_modulation_mae_db": 0.38176501816624436, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5534419642060942, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18914743127691086, "3000–6000 Hz_10ms_envelope_correlation": 0.9742378233685248, "3000–6000 Hz_10ms_modulation_mae_db": 0.7468996879537517, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9847031727231629, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10929499875480558, "3000–6000 Hz_40ms_envelope_correlation": 0.9906938158344996, "3000–6000 Hz_40ms_modulation_mae_db": 0.7340955639220358, "mr_stft_db": 5.4253654802258096, "band_mae_db": 0.1436147706723415, "rms_gain_error_db": -0.08222867421545334, "output_peak": 0.2520344464535214}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.7256314033250484, "全帯域_10ms_rms_envelope_nrmse": 0.053844391584530873, "全帯域_10ms_envelope_correlation": 0.9968174274220101, "全帯域_10ms_modulation_mae_db": 0.6728034888420716, "全帯域_40ms_log_rms_mae_db": 0.3922274351292308, "全帯域_40ms_rms_envelope_nrmse": 0.029273149316184903, "全帯域_40ms_envelope_correlation": 0.9991130320773306, "全帯域_40ms_modulation_mae_db": 0.714380089807392, "300–1000 Hz_10ms_log_rms_mae_db": 1.0794005540648335, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06973984406563652, "300–1000 Hz_10ms_envelope_correlation": 0.9956208409275191, "300–1000 Hz_10ms_modulation_mae_db": 0.781159977234258, "300–1000 Hz_40ms_log_rms_mae_db": 0.6169252091561639, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04241147881065092, "300–1000 Hz_40ms_envelope_correlation": 0.9984920110303802, "300–1000 Hz_40ms_modulation_mae_db": 0.8231297181930631, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3830486946382157, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1361633558768811, "1000–3000 Hz_10ms_envelope_correlation": 0.9857823659212938, "1000–3000 Hz_10ms_modulation_mae_db": 0.7664960305105571, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8094562531442198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08227040565675947, "1000–3000 Hz_40ms_envelope_correlation": 0.9950260387457733, "1000–3000 Hz_40ms_modulation_mae_db": 0.6674512250750165, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5484734261623248, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1763767922997266, "3000–6000 Hz_10ms_envelope_correlation": 0.9777060040555492, "3000–6000 Hz_10ms_modulation_mae_db": 0.33426118447645803, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9741682776420353, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09857606959984157, "3000–6000 Hz_40ms_envelope_correlation": 0.9926237474289549, "3000–6000 Hz_40ms_modulation_mae_db": 0.44014850198540545, "mr_stft_db": 5.4261130513850615, "band_mae_db": 0.1834776699609811, "rms_gain_error_db": -0.10790936511596289, "output_peak": 0.2508217903518799}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6978274636330088, "全帯域_10ms_rms_envelope_nrmse": 0.04844530144424208, "全帯域_10ms_envelope_correlation": 0.9973688018640824, "全帯域_10ms_modulation_mae_db": 0.4195753665747479, "全帯域_40ms_log_rms_mae_db": 0.3838345284694931, "全帯域_40ms_rms_envelope_nrmse": 0.029747834926887463, "全帯域_40ms_envelope_correlation": 0.998974607805637, "全帯域_40ms_modulation_mae_db": 0.5898517000487128, "300–1000 Hz_10ms_log_rms_mae_db": 1.0283012770562105, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.05664511329151649, "300–1000 Hz_10ms_envelope_correlation": 0.9970322000028471, "300–1000 Hz_10ms_modulation_mae_db": 0.5197154540252004, "300–1000 Hz_40ms_log_rms_mae_db": 0.5872752513997271, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.036170783759397564, "300–1000 Hz_40ms_envelope_correlation": 0.9987798720806308, "300–1000 Hz_40ms_modulation_mae_db": 0.693280172643115, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4021599667750617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13254891033863875, "1000–3000 Hz_10ms_envelope_correlation": 0.9861238275974619, "1000–3000 Hz_10ms_modulation_mae_db": 0.27879735198827876, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8327082632960184, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07750897706955649, "1000–3000 Hz_40ms_envelope_correlation": 0.9949886887065292, "1000–3000 Hz_40ms_modulation_mae_db": 0.1474017069595651, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5519392451336909, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1927266567401881, "3000–6000 Hz_10ms_envelope_correlation": 0.9735446760640114, "3000–6000 Hz_10ms_modulation_mae_db": 0.6481494526443637, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9910213523932249, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11859636074256379, "3000–6000 Hz_40ms_envelope_correlation": 0.9896356722151505, "3000–6000 Hz_40ms_modulation_mae_db": 0.7631717678187669, "mr_stft_db": 5.428142192143299, "band_mae_db": 0.16082747347313284, "rms_gain_error_db": -0.07141378693924536, "output_peak": 0.24908418548700761}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.677526387543133, "全帯域_10ms_rms_envelope_nrmse": 0.04756280796458035, "全帯域_10ms_envelope_correlation": 0.9974813121116021, "全帯域_10ms_modulation_mae_db": 0.2905009148123177, "全帯域_40ms_log_rms_mae_db": 0.373904352379641, "全帯域_40ms_rms_envelope_nrmse": 0.029274520633229967, "全帯域_40ms_envelope_correlation": 0.999022777332066, "全帯域_40ms_modulation_mae_db": 0.5242076441130455, "300–1000 Hz_10ms_log_rms_mae_db": 1.0705972048140677, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.055727074436250246, "300–1000 Hz_10ms_envelope_correlation": 0.9971248635150454, "300–1000 Hz_10ms_modulation_mae_db": 0.38547686502682943, "300–1000 Hz_40ms_log_rms_mae_db": 0.5724641580277956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03322056736435238, "300–1000 Hz_40ms_envelope_correlation": 0.99897981707227, "300–1000 Hz_40ms_modulation_mae_db": 0.6121565895630696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3678996664818668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1420736433429707, "1000–3000 Hz_10ms_envelope_correlation": 0.9841243182074655, "1000–3000 Hz_10ms_modulation_mae_db": 0.6916214085784231, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8171379541309701, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08353112878887867, "1000–3000 Hz_40ms_envelope_correlation": 0.9943517924743167, "1000–3000 Hz_40ms_modulation_mae_db": 0.522856160856008, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5683853375415961, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19103846566508328, "3000–6000 Hz_10ms_envelope_correlation": 0.9737816459085848, "3000–6000 Hz_10ms_modulation_mae_db": 1.0310402736858675, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9438804743706082, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11309875944089229, "3000–6000 Hz_40ms_envelope_correlation": 0.9901464491116277, "3000–6000 Hz_40ms_modulation_mae_db": 1.0920584651506327, "mr_stft_db": 5.417066744243722, "band_mae_db": 0.10392606700414786, "rms_gain_error_db": -0.07325080294168602, "output_peak": 0.2423536881778799}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.335459053166848, "spectral_convergence": 0.14082040834874243, "spectral_convergence_status": "ok", "amplitude_floor": 9.671435620644006e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.361991023312295, "spectral_convergence": 0.14022290922258984, "spectral_convergence_status": "ok", "amplitude_floor": 9.322400769888516e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.584986852192272, "spectral_convergence": 0.14006698273812196, "spectral_convergence_status": "ok", "amplitude_floor": 9.453157582642776e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00016624637287256398, "output_power": 0.00017368317941902293, "delta_db": 0.1900558191540096, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013640231584289317, "output_power": 0.0013239867816950233, "delta_db": -0.12938094595142674, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005975940179232537, "output_power": 0.0005815725535860943, "delta_db": -0.11802338943931562, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.8168186983374496e-05, "output_power": 1.863769723357398e-05, "delta_db": 0.1108066150276603, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.463537251271816e-05, "output_power": 2.2953462567666576e-05, "delta_db": -0.3071092479573547, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0343523773544458e-05, "output_power": 1.0485961917501388e-05, "delta_db": 0.05939759158441507, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.152462268185697, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10009196549144289, "pitch": null, "output_rms_dbfs": -26.713514137662322, "output_peak": 0.25079681014410027, "reference_rms_dbfs": -26.613422172170882, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.427478976223806}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001446111949915804, "log_rms_mae_db": 0.6938344463749831, "rms_envelope_nrmse": 0.05526299503613511, "envelope_correlation": 0.9966221109094929, "rms_gain_error_db": -0.10009230381002125, "reference_mean_power": 0.002182828397481346, "output_mean_power": 0.002133095820237262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.790920693994897e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003230276115117931, "output_power": 0.00031160059228158823, "reference_active": true, "delta_db": -0.15641371719627584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.477236494661316e-05, "output_power": 4.18243779783375e-05, "reference_active": true, "delta_db": -0.2958054434960569}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.548345752003965e-06, "output_power": 7.75453469881181e-06, "reference_active": true, "delta_db": -0.9037239308152717}], "active_band_count": 3, "active_band_mae_db": 0.45198103050253485}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014111493703772434, "log_rms_mae_db": 0.34630001245237846, "rms_envelope_nrmse": 0.02962799518968627, "envelope_correlation": 0.999064592678049, "rms_gain_error_db": -0.10016345666571959, "reference_mean_power": 0.002193693032291648, "output_mean_power": 0.0021436777987693814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.190420923463322e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028912450266098784, "output_power": 0.00028074159323651806, "reference_active": true, "delta_db": -0.12778138478133189}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2978791346744758e-05, "output_power": 2.1370707066447215e-05, "reference_active": true, "delta_db": -0.3150829028022276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.072854493661425e-07, "output_power": 4.890153015672176e-07, "reference_active": true, "delta_db": -0.9407042684031796}], "active_band_count": 3, "active_band_mae_db": 0.46118951866224633}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012712872255296568, "log_rms_mae_db": 1.0286594610736983, "rms_envelope_nrmse": 0.06505963266891254, "envelope_correlation": 0.9961624369266349, "rms_gain_error_db": -0.14485224604720817, "reference_mean_power": 0.0014813690704446063, "output_mean_power": 0.0014327751751970314}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.409297669270013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003163260619212235, "output_power": 0.0003035356194659029, "reference_active": true, "delta_db": -0.1792531242984198}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6701975061022124e-05, "output_power": 3.257107517452324e-05, "reference_active": true, "delta_db": -0.5185734087184511}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.301526036454895e-06, "output_power": 5.148430548401554e-06, "reference_active": true, "delta_db": -0.8777087645588993}], "active_band_count": 3, "active_band_mae_db": 0.5251784325252568}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012343588785799389, "log_rms_mae_db": 0.552229561513316, "rms_envelope_nrmse": 0.03706366186947246, "envelope_correlation": 0.9988506153880528, "rms_gain_error_db": -0.1448964052178352, "reference_mean_power": 0.0014887936512772218, "output_mean_power": 0.001439941563333949}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.99328908394483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002875949770086079, "output_power": 0.0002774942171297568, "reference_active": true, "delta_db": -0.15527359577849653}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97680708793848e-05, "output_power": 1.7307153223006662e-05, "reference_active": true, "delta_db": -0.5773865100730492}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.520482053574084e-07, "output_power": 3.380101347991067e-07, "reference_active": true, "delta_db": -1.2625502718868007}], "active_band_count": 3, "active_band_mae_db": 0.6650701259127821}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.4084091617788044e-05, "log_rms_mae_db": 1.407235040165869, "rms_envelope_nrmse": 0.1428216341015746, "envelope_correlation": 0.9840696249908355, "rms_gain_error_db": -0.19969378479139283, "reference_mean_power": 3.6694438258439554e-05, "output_mean_power": 3.504538740113537e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6589829493109366e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3179349061495639e-05, "output_power": 1.1724819837087895e-05, "reference_active": true, "delta_db": -0.5078778254455665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9890642401085437e-06, "output_power": 2.700441674750944e-06, "reference_active": true, "delta_db": -0.441004472238059}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.136200949313584e-07, "output_power": 5.56655624820046e-07, "reference_active": true, "delta_db": -1.6483506450018834}], "active_band_count": 3, "active_band_mae_db": 0.8657443142285031}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.151462172211685e-05, "log_rms_mae_db": 0.8357459150743828, "rms_envelope_nrmse": 0.08440738567721344, "envelope_correlation": 0.9944074890777038, "rms_gain_error_db": -0.19976794168550946, "reference_mean_power": 3.687647047180041e-05, "output_mean_power": 3.5218637700542654e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4709890047082147e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2743108138238886e-05, "output_power": 1.1196781730340092e-05, "reference_active": true, "delta_db": -0.561821563829304}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4131557808728274e-06, "output_power": 1.3609574916291477e-06, "reference_active": true, "delta_db": -0.1634547885571897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.7131268313465536e-08, "output_power": 4.534891921696298e-08, "reference_active": true, "delta_db": -1.0030692435071977}], "active_band_count": 3, "active_band_mae_db": 0.5761151986312305}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.703026630874025e-05, "log_rms_mae_db": 1.5611796686832096, "rms_envelope_nrmse": 0.1875429653968853, "envelope_correlation": 0.9748459245995141, "rms_gain_error_db": -0.027296959100026715, "reference_mean_power": 1.1971789360385644e-05, "output_mean_power": 1.1896778371373909e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.043568625201122e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.248382965898838e-06, "output_power": 4.264197700987517e-06, "reference_active": true, "delta_db": 0.016136728337422343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.589400168019184e-06, "output_power": 2.3554702123919777e-06, "reference_active": true, "delta_db": -0.41121555386037467}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.486231293405906e-07, "output_power": 4.683705827631085e-07, "reference_active": true, "delta_db": -1.4140282092076093}], "active_band_count": 3, "active_band_mae_db": 0.6137934971351354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.3691778537856407e-05, "log_rms_mae_db": 0.9567278843859733, "rms_envelope_nrmse": 0.11856263471795798, "envelope_correlation": 0.9888814975308227, "rms_gain_error_db": -0.027300500364179386, "reference_mean_power": 1.2031928101806957e-05, "output_mean_power": 1.1956530555323067e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.457061965350652e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.507382710758712e-06, "output_power": 4.399597131674098e-06, "reference_active": true, "delta_db": -0.1051152427490016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3712782786962106e-06, "output_power": 1.2240185795452946e-06, "reference_active": true, "delta_db": -0.4933758668298587}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.274779912953651e-08, "output_power": 2.8479528055613876e-08, "reference_active": true, "delta_db": -1.7638097181194377}], "active_band_count": 3, "active_band_mae_db": 0.787433609232766}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9372831432786729, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "da25e129c055a0a781ad3211cafd3e9852dddfeac645f67e8088c31deb0d52cf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9858533649251596, "standard_deviation": 0.0014159900942435123, "minimum": 0.9840696249908355, "maximum": 0.9884425590517473, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3704253330407163, "standard_deviation": 0.029395759254157397, "minimum": 1.3276019621962862, "maximum": 1.407235040165869, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6430016944695334, "standard_deviation": 0.21115192121822038, "minimum": 0.27879735198827876, "maximum": 0.8657443142285031, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13469796890934713, "standard_deviation": 0.007003053790120943, "minimum": 0.12145431434739354, "maximum": 0.1428216341015746, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995123844991606, "standard_deviation": 0.0009516283245467816, "minimum": 0.993803263677469, "maximum": 0.9966126730017792, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8038030603989641, "standard_deviation": 0.028197088495042242, "minimum": 0.763683992085459, "maximum": 0.8357459150743828, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4734971266110023, "standard_deviation": 0.18489332947009787, "minimum": 0.1474017069595651, "maximum": 0.7132489511920307, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0787210301521242, "standard_deviation": 0.008536687382788257, "minimum": 0.0651542301725054, "maximum": 0.09179614301981853, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.974535087410261, "standard_deviation": 0.0014239643809835013, "minimum": 0.9730908196505362, "maximum": 0.9777060040555492, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5513641123405493, "standard_deviation": 0.012407444493366193, "minimum": 1.5294120714472768, "maximum": 1.5683853375415961, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6935781861039417, "standard_deviation": 0.19811833406457507, "minimum": 0.33426118447645803, "maximum": 1.0310402736858675, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1889692564918914, "standard_deviation": 0.005502036582497347, "minimum": 0.1763767922997266, "maximum": 0.19392788613922685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9901396997985616, "standard_deviation": 0.001746074893143963, "minimum": 0.986946118754848, "maximum": 0.9926237474289549, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9706204098673705, "standard_deviation": 0.023276908454971962, "minimum": 0.9405936454102978, "maximum": 1.0072280330173207, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7690580637856321, "standard_deviation": 0.17890927223285644, "minimum": 0.44014850198540545, "maximum": 1.0920584651506327, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11459228609606911, "standard_deviation": 0.009238223277343876, "minimum": 0.09857606959984157, "maximum": 0.1304460620030164, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9966591770094734, "standard_deviation": 0.0005290717201659987, "minimum": 0.9956208409275191, "maximum": 0.9971579045434763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.0398885790645824, "standard_deviation": 0.027577728543586155, "minimum": 1.0008985716582688, "maximum": 1.0794005540648335, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5806164284902758, "standard_deviation": 0.18812953286849088, "minimum": 0.36074311189411884, "maximum": 0.8866007571773994, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.06039309930765523, "standard_deviation": 0.00486650634142692, "minimum": 0.055727074436250246, "maximum": 0.06973984406563652, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9989210062331728, "standard_deviation": 0.0002126572075249666, "minimum": 0.9984920110303802, "maximum": 0.9991563304304982, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.5720661983385862, "standard_deviation": 0.021614960295982366, "minimum": 0.5503000876877931, "maximum": 0.6169252091561639, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7128325103482576, "standard_deviation": 0.23271464126710328, "minimum": 0.42952612739282453, "maximum": 1.1754543360080716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.03519548999897666, "standard_deviation": 0.00340825570194613, "minimum": 0.032004865723744017, "maximum": 0.04241147881065092, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.16669921909547258, "standard_deviation": 0.0384349771156279, "minimum": 0.10392606700414786, "maximum": 0.2385361961239293, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.421183866416052, "standard_deviation": 0.00693604537249374, "minimum": 5.4093063099126155, "maximum": 5.428142192143299, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.24858222702495925, "standard_deviation": 0.003411585583687184, "minimum": 0.2423536881778799, "maximum": 0.2520344464535214, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08630250063359539, "standard_deviation": 0.014067497740087616, "minimum": -0.10790936511596289, "maximum": -0.07120233070743273, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9971203025167653, "standard_deviation": 0.0003385636034811725, "minimum": 0.9966221109094929, "maximum": 0.9976024763746639, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.7006921081194009, "standard_deviation": 0.020023857157731684, "minimum": 0.677526387543133, "maximum": 0.7359519024320946, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4779955635450872, "standard_deviation": 0.2076215681626533, "minimum": 0.22937347748927092, "maximum": 0.8495924730666031, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.050883344997892, "standard_deviation": 0.003010955807684308, "minimum": 0.04685019178771773, "maximum": 0.05526299503613511, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9991458013680024, "standard_deviation": 0.00014376584759642988, "minimum": 0.998974607805637, "maximum": 0.9994110172684818, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37859700296411003, "standard_deviation": 0.01738722847547429, "minimum": 0.34630001245237846, "maximum": 0.39990032913017104, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.563598976126356, "standard_deviation": 0.18502509146047677, "minimum": 0.31642364929955963, "maximum": 0.8567144668364918, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0279338653526739, "standard_deviation": 0.0020875592382814155, "minimum": 0.024162789425786223, "maximum": 0.029747834926887463, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8555-284449-0020", "language": "en", "style": "normal", "speaker": "librispeech_8555", "source_utterance_id": "8555-284449-0020", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 81520, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8555-284449-0020.wav", "source_file_sha256": "715f323ae46e353a868649a2400d8829a17a304f4e0df36ea567c9950598dd60", "decoded_pcm_sha256": "fea5f67e0d192edbb42257fa9560e233f643cc196059371f1de72f17987367e7", "original_source_sha256": "5703c01afc67c6eb161bd9fe7fe92f5b373e9331629f60d1c163e5d893c97d7a", "corpus_entry_id": "librispeech_8555-284449-0020"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.627328634262085, "predicted_mos": 3.627328634262085, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "224f98950a81464aa9e795adc9b1038197065447a29b71942cef31644455cd7f", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.436567382002977, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8555-284449-0020", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "224f98950a81464aa9e795adc9b1038197065447a29b71942cef31644455cd7f", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "0ad90c10b2c7d8a7b3190141694c0f9202a9fa41f2cc5c03ef81a8fffa9a0703"}, "rows": [{"case": "librispeech_8555-284449-0020", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.811661958694458, "predicted_mos": 2.811661958694458, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "dfc5d91d054fe2e88c25a7be3108bef07791e036c69e9b921e815272dfd94810", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7456342130026314, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8555-284449-0020", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "dfc5d91d054fe2e88c25a7be3108bef07791e036c69e9b921e815272dfd94810", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8984980968729311, "全帯域_10ms_rms_envelope_nrmse": 0.0757319975442994, "全帯域_10ms_envelope_correlation": 0.9936000157630474, "全帯域_10ms_modulation_mae_db": 0.4452410963407658, "全帯域_40ms_log_rms_mae_db": 0.45542373327854946, "全帯域_40ms_rms_envelope_nrmse": 0.04637295616971104, "全帯域_40ms_envelope_correlation": 0.9973951653051819, "全帯域_40ms_modulation_mae_db": 0.4623753609250949, "300–1000 Hz_10ms_log_rms_mae_db": 1.6041880435415037, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1005547001987965, "300–1000 Hz_10ms_envelope_correlation": 0.9911605059880798, "300–1000 Hz_10ms_modulation_mae_db": 0.5156116930169382, "300–1000 Hz_40ms_log_rms_mae_db": 0.9694643298850862, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06559842446885696, "300–1000 Hz_40ms_envelope_correlation": 0.9960247012937518, "300–1000 Hz_40ms_modulation_mae_db": 0.538605481321965, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5240116473817982, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11331121358911182, "1000–3000 Hz_10ms_envelope_correlation": 0.9901284543871186, "1000–3000 Hz_10ms_modulation_mae_db": 0.4733522830984454, "1000–3000 Hz_40ms_log_rms_mae_db": 0.840271496940516, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05761599513586017, "1000–3000 Hz_40ms_envelope_correlation": 0.9974566530459952, "1000–3000 Hz_40ms_modulation_mae_db": 0.6792664247323733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7973136957282165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2608418939655096, "3000–6000 Hz_10ms_envelope_correlation": 0.9455690739679232, "3000–6000 Hz_10ms_modulation_mae_db": 1.090189772259477, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3254508212160252, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2055809243800066, "3000–6000 Hz_40ms_envelope_correlation": 0.9622751347720364, "3000–6000 Hz_40ms_modulation_mae_db": 1.4382119754189395, "mr_stft_db": 5.703355662919577, "band_mae_db": 0.2376696311615489, "rms_gain_error_db": -0.11886796586780021, "output_peak": 0.3165365036612825}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8973944237272801, "全帯域_10ms_rms_envelope_nrmse": 0.07156507045135126, "全帯域_10ms_envelope_correlation": 0.9942443469254082, "全帯域_10ms_modulation_mae_db": 0.20417647982973688, "全帯域_40ms_log_rms_mae_db": 0.4623724049916631, "全帯域_40ms_rms_envelope_nrmse": 0.03966671424784265, "全帯域_40ms_envelope_correlation": 0.998084125644277, "全帯域_40ms_modulation_mae_db": 0.33001287443506777, "300–1000 Hz_10ms_log_rms_mae_db": 1.5443743602962585, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09068605413716821, "300–1000 Hz_10ms_envelope_correlation": 0.9928668148789332, "300–1000 Hz_10ms_modulation_mae_db": 0.27852108896688854, "300–1000 Hz_40ms_log_rms_mae_db": 0.9796906338942293, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05152722696862573, "300–1000 Hz_40ms_envelope_correlation": 0.9976743260117392, "300–1000 Hz_40ms_modulation_mae_db": 0.5122310258300072, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6191704773061903, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12330245216225048, "1000–3000 Hz_10ms_envelope_correlation": 0.9883062527141745, "1000–3000 Hz_10ms_modulation_mae_db": 0.7718198689898212, "1000–3000 Hz_40ms_log_rms_mae_db": 0.835469235642051, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06400901634893408, "1000–3000 Hz_40ms_envelope_correlation": 0.9968579234082292, "1000–3000 Hz_40ms_modulation_mae_db": 0.9052647267266654, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8282282763141169, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2737842003637736, "3000–6000 Hz_10ms_envelope_correlation": 0.9420300344566508, "3000–6000 Hz_10ms_modulation_mae_db": 1.1785293887191888, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3517218148068202, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2179614485038444, "3000–6000 Hz_40ms_envelope_correlation": 0.959310937640275, "3000–6000 Hz_40ms_modulation_mae_db": 1.515625038301339, "mr_stft_db": 5.712904797000539, "band_mae_db": 0.18803792226380275, "rms_gain_error_db": -0.09764371203329665, "output_peak": 0.30330524184940444}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9098970096012338, "全帯域_10ms_rms_envelope_nrmse": 0.06996939936765426, "全帯域_10ms_envelope_correlation": 0.9945899085777945, "全帯域_10ms_modulation_mae_db": 0.7163412945094544, "全帯域_40ms_log_rms_mae_db": 0.49194861130128725, "全帯域_40ms_rms_envelope_nrmse": 0.03982670058438604, "全帯域_40ms_envelope_correlation": 0.9981900100808416, "全帯域_40ms_modulation_mae_db": 0.8423041980312372, "300–1000 Hz_10ms_log_rms_mae_db": 1.526965887508978, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08760669621475291, "300–1000 Hz_10ms_envelope_correlation": 0.9934825682442998, "300–1000 Hz_10ms_modulation_mae_db": 0.9408098490835796, "300–1000 Hz_40ms_log_rms_mae_db": 0.9455129715267828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05209750595411872, "300–1000 Hz_40ms_envelope_correlation": 0.9977435858454039, "300–1000 Hz_40ms_modulation_mae_db": 1.296102108136123, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6110058771541023, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11657323873715106, "1000–3000 Hz_10ms_envelope_correlation": 0.9896364650001551, "1000–3000 Hz_10ms_modulation_mae_db": 0.9927396397683707, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9085303901251088, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07002941728933934, "1000–3000 Hz_40ms_envelope_correlation": 0.9962512217170048, "1000–3000 Hz_40ms_modulation_mae_db": 1.3095042385263864, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8024792858222256, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25974875240875384, "3000–6000 Hz_10ms_envelope_correlation": 0.9445065683353724, "3000–6000 Hz_10ms_modulation_mae_db": 0.9331194762955634, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3594654089425988, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21143980619420782, "3000–6000 Hz_40ms_envelope_correlation": 0.9588010350868834, "3000–6000 Hz_40ms_modulation_mae_db": 1.1255549644377167, "mr_stft_db": 5.708264526977357, "band_mae_db": 0.27681808381021317, "rms_gain_error_db": -0.1295906555424584, "output_peak": 0.3064129861907982}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9148347795074426, "全帯域_10ms_rms_envelope_nrmse": 0.07451437576925501, "全帯域_10ms_envelope_correlation": 0.9938505670658222, "全帯域_10ms_modulation_mae_db": 0.7431887221409967, "全帯域_40ms_log_rms_mae_db": 0.49066977303086606, "全帯域_40ms_rms_envelope_nrmse": 0.04080998216934521, "全帯域_40ms_envelope_correlation": 0.998084984581824, "全帯域_40ms_modulation_mae_db": 0.8948724446356202, "300–1000 Hz_10ms_log_rms_mae_db": 1.640757109121831, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09410754473528311, "300–1000 Hz_10ms_envelope_correlation": 0.9922135298848926, "300–1000 Hz_10ms_modulation_mae_db": 0.7733934056033362, "300–1000 Hz_40ms_log_rms_mae_db": 1.0215374546246279, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053819943615958876, "300–1000 Hz_40ms_envelope_correlation": 0.9973152016382754, "300–1000 Hz_40ms_modulation_mae_db": 1.0031012161304325, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5927578919320315, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13335177883179927, "1000–3000 Hz_10ms_envelope_correlation": 0.9861854830161862, "1000–3000 Hz_10ms_modulation_mae_db": 0.9558322163799611, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8625229302702857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06959476793870116, "1000–3000 Hz_40ms_envelope_correlation": 0.9961380406710204, "1000–3000 Hz_40ms_modulation_mae_db": 1.0846027904889464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.806207360711787, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26576881889111553, "3000–6000 Hz_10ms_envelope_correlation": 0.9442078134788656, "3000–6000 Hz_10ms_modulation_mae_db": 0.5279334048736976, "3000–6000 Hz_40ms_log_rms_mae_db": 1.296874483781306, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20837921437889015, "3000–6000 Hz_40ms_envelope_correlation": 0.9617311026162289, "3000–6000 Hz_40ms_modulation_mae_db": 0.7569402673833875, "mr_stft_db": 5.718135196915949, "band_mae_db": 0.2368049479607605, "rms_gain_error_db": -0.1296677239683325, "output_peak": 0.3062611041741344}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8931072840677701, "全帯域_10ms_rms_envelope_nrmse": 0.07421697162011916, "全帯域_10ms_envelope_correlation": 0.9938661225228375, "全帯域_10ms_modulation_mae_db": 0.28730752447371316, "全帯域_40ms_log_rms_mae_db": 0.445722153724591, "全帯域_40ms_rms_envelope_nrmse": 0.04080184911977649, "全帯域_40ms_envelope_correlation": 0.9980536141904691, "全帯域_40ms_modulation_mae_db": 0.17989393119345443, "300–1000 Hz_10ms_log_rms_mae_db": 1.5451537694485094, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09464587296891445, "300–1000 Hz_10ms_envelope_correlation": 0.9921576290803903, "300–1000 Hz_10ms_modulation_mae_db": 0.44636090385084976, "300–1000 Hz_40ms_log_rms_mae_db": 0.9179884441353298, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05511114043330557, "300–1000 Hz_40ms_envelope_correlation": 0.9972433217535618, "300–1000 Hz_40ms_modulation_mae_db": 0.4609719350008225, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6177335761844074, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13215133534955684, "1000–3000 Hz_10ms_envelope_correlation": 0.9864983878733974, "1000–3000 Hz_10ms_modulation_mae_db": 1.2553103138136281, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9376884150069043, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0723988808673964, "1000–3000 Hz_40ms_envelope_correlation": 0.995897102669375, "1000–3000 Hz_40ms_modulation_mae_db": 1.4778141847045505, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7926747125835085, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26474813221244053, "3000–6000 Hz_10ms_envelope_correlation": 0.94444922425743, "3000–6000 Hz_10ms_modulation_mae_db": 0.9406762139902988, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3220757718659022, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20513387520230192, "3000–6000 Hz_40ms_envelope_correlation": 0.9628783150770281, "3000–6000 Hz_40ms_modulation_mae_db": 1.3435776941678608, "mr_stft_db": 5.707334087897042, "band_mae_db": 0.21111791210113706, "rms_gain_error_db": -0.12241486468798929, "output_peak": 0.31018511472968613}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8878800187749287, "全帯域_10ms_rms_envelope_nrmse": 0.06757820465497123, "全帯域_10ms_envelope_correlation": 0.9949056970950025, "全帯域_10ms_modulation_mae_db": 0.38301771151814873, "全帯域_40ms_log_rms_mae_db": 0.4584343898072032, "全帯域_40ms_rms_envelope_nrmse": 0.03650081683716654, "全帯域_40ms_envelope_correlation": 0.998421358139044, "全帯域_40ms_modulation_mae_db": 0.613052427934044, "300–1000 Hz_10ms_log_rms_mae_db": 1.5408802632587635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09026113071844179, "300–1000 Hz_10ms_envelope_correlation": 0.9928270407661626, "300–1000 Hz_10ms_modulation_mae_db": 0.34659892860746905, "300–1000 Hz_40ms_log_rms_mae_db": 0.9454269812295325, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05111850167048422, "300–1000 Hz_40ms_envelope_correlation": 0.9975932011341718, "300–1000 Hz_40ms_modulation_mae_db": 0.6228304474341283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6154810763827125, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11820577750307829, "1000–3000 Hz_10ms_envelope_correlation": 0.9898823318430965, "1000–3000 Hz_10ms_modulation_mae_db": 0.9640799599004802, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8519219751301521, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06983919891886724, "1000–3000 Hz_40ms_envelope_correlation": 0.9969226435059166, "1000–3000 Hz_40ms_modulation_mae_db": 1.0761370126323875, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8191151928451128, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26163730582950945, "3000–6000 Hz_10ms_envelope_correlation": 0.9489774359829597, "3000–6000 Hz_10ms_modulation_mae_db": 0.4919482018707571, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3556753774051118, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20626749891237078, "3000–6000 Hz_40ms_envelope_correlation": 0.9652721163318164, "3000–6000 Hz_40ms_modulation_mae_db": 0.7796446635401096, "mr_stft_db": 5.718612813912459, "band_mae_db": 0.2538727011216678, "rms_gain_error_db": -0.10236093430073277, "output_peak": 0.29915368259629593}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9117095694835904, "全帯域_10ms_rms_envelope_nrmse": 0.07330727594947466, "全帯域_10ms_envelope_correlation": 0.9940338650997737, "全帯域_10ms_modulation_mae_db": 0.41384156431129315, "全帯域_40ms_log_rms_mae_db": 0.4680500970002904, "全帯域_40ms_rms_envelope_nrmse": 0.04023085908922151, "全帯域_40ms_envelope_correlation": 0.99811604792279, "全帯域_40ms_modulation_mae_db": 0.5818041421870309, "300–1000 Hz_10ms_log_rms_mae_db": 1.526313532332377, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08728093688048776, "300–1000 Hz_10ms_envelope_correlation": 0.9934530643980015, "300–1000 Hz_10ms_modulation_mae_db": 0.5720009649823838, "300–1000 Hz_40ms_log_rms_mae_db": 0.8980508515323089, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04824677240365183, "300–1000 Hz_40ms_envelope_correlation": 0.9980363054231215, "300–1000 Hz_40ms_modulation_mae_db": 0.7673581297382887, "1000–3000 Hz_10ms_log_rms_mae_db": 1.628973121198808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12671183181206136, "1000–3000 Hz_10ms_envelope_correlation": 0.9878951770288459, "1000–3000 Hz_10ms_modulation_mae_db": 0.8688407174879721, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8983648091360991, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07347911236320857, "1000–3000 Hz_40ms_envelope_correlation": 0.9961165794448985, "1000–3000 Hz_40ms_modulation_mae_db": 1.0439426616176712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.749977763802212, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26129530530952944, "3000–6000 Hz_10ms_envelope_correlation": 0.9474012998747449, "3000–6000 Hz_10ms_modulation_mae_db": 0.3624118332609725, "3000–6000 Hz_40ms_log_rms_mae_db": 1.260215401718764, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2130520863412384, "3000–6000 Hz_40ms_envelope_correlation": 0.9612238892548999, "3000–6000 Hz_40ms_modulation_mae_db": 0.28759365499836304, "mr_stft_db": 5.708991204824699, "band_mae_db": 0.2453411443684509, "rms_gain_error_db": -0.11688939633300502, "output_peak": 0.3167907988202419}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9008108594035621, "全帯域_10ms_rms_envelope_nrmse": 0.07398644502685384, "全帯域_10ms_envelope_correlation": 0.9938872601024571, "全帯域_10ms_modulation_mae_db": 0.6274867180842731, "全帯域_40ms_log_rms_mae_db": 0.4510724505901869, "全帯域_40ms_rms_envelope_nrmse": 0.03766401550850481, "全帯域_40ms_envelope_correlation": 0.9983390497963566, "全帯域_40ms_modulation_mae_db": 0.853137923973256, "300–1000 Hz_10ms_log_rms_mae_db": 1.5113470850625927, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09195717817814528, "300–1000 Hz_10ms_envelope_correlation": 0.992581091277367, "300–1000 Hz_10ms_modulation_mae_db": 0.8467127398343305, "300–1000 Hz_40ms_log_rms_mae_db": 0.9260467892693804, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047697718090154426, "300–1000 Hz_40ms_envelope_correlation": 0.9979695554977946, "300–1000 Hz_40ms_modulation_mae_db": 1.1443546259811794, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6140980675673329, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13768489833314576, "1000–3000 Hz_10ms_envelope_correlation": 0.9854970637668404, "1000–3000 Hz_10ms_modulation_mae_db": 0.9257372457155154, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8947072737090543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09025552155154278, "1000–3000 Hz_40ms_envelope_correlation": 0.9936798027949646, "1000–3000 Hz_40ms_modulation_mae_db": 1.1127995287514063, "3000–6000 Hz_10ms_log_rms_mae_db": 1.816414382966876, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27263584896257675, "3000–6000 Hz_10ms_envelope_correlation": 0.9410163707348805, "3000–6000 Hz_10ms_modulation_mae_db": 0.578305760271299, "3000–6000 Hz_40ms_log_rms_mae_db": 1.335091282073849, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2119186971051289, "3000–6000 Hz_40ms_envelope_correlation": 0.9602290166784241, "3000–6000 Hz_40ms_modulation_mae_db": 0.8081816781285633, "mr_stft_db": 5.707399252664163, "band_mae_db": 0.25822122898008987, "rms_gain_error_db": -0.1087167931867234, "output_peak": 0.2742514716519653}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 81520, "output_samples": 81520, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.499982250422403, "spectral_convergence": 0.1880142850669522, "spectral_convergence_status": "ok", "amplitude_floor": 1.0222158521079025e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.621888497434717, "spectral_convergence": 0.1888974470497869, "spectral_convergence_status": "ok", "amplitude_floor": 7.564059658380168e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.988196240901611, "spectral_convergence": 0.19034039637325803, "spectral_convergence_status": "ok", "amplitude_floor": 6.35834815641291e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.932339290337003e-05, "output_power": 9.659335967848422e-05, "delta_db": 0.33982060651656887, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011480494900330098, "output_power": 0.0011309038655752638, "delta_db": -0.06534921453654388, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005518546578574125, "output_power": 0.0005154848624275594, "delta_db": -0.2960879606184026, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.085077476738271e-05, "output_power": 8.828366413424185e-05, "delta_db": -0.12448285418468737, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.0710168905088768e-05, "output_power": 1.9338440666910412e-05, "delta_db": -0.2976218854694596, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8163123695227053e-05, "output_power": 2.626729164532258e-05, "delta_db": -0.30265526564363104, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2376696311615489, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11886796586780021, "pitch": null, "output_rms_dbfs": -27.265654640035745, "output_peak": 0.3165365036612825, "reference_rms_dbfs": -27.14678667416795, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.703355662919577}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 81520, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.00013747273046535536, "log_rms_mae_db": 0.8984980968729311, "rms_envelope_nrmse": 0.0757319975442994, "envelope_correlation": 0.9936000157630474, "rms_gain_error_db": -0.11886790704179356, "reference_mean_power": 0.0019308462257719779, "output_mean_power": 0.0018787149740517866}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 6.283614364116833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041565087617635905, "output_power": 0.00039745084779094214, "reference_active": true, "delta_db": -0.19445272613485112}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.15171169386713e-05, "output_power": 7.820223307583879e-05, "reference_active": true, "delta_db": -0.18029656757542742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.519660659304176e-05, "output_power": 2.0195037114250525e-05, "reference_active": true, "delta_db": -0.960973995312019}], "active_band_count": 3, "active_band_mae_db": 0.4452410963407658}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012452330782545057, "log_rms_mae_db": 0.45542373327854946, "rms_envelope_nrmse": 0.04637295616971104, "envelope_correlation": 0.9973951653051819, "rms_gain_error_db": -0.11886797570367548, "reference_mean_power": 0.0019422928952156904, "output_mean_power": 0.0018898525630058043}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.257129910319434e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003707029677887353, "output_power": 0.00035280812439167976, "reference_active": true, "delta_db": -0.21487485518313315}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2635280867150454e-05, "output_power": 4.1400520567772025e-05, "reference_active": true, "delta_db": -0.12763326368220435}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.869258984918021e-06, "output_power": 1.469628888956358e-06, "reference_active": true, "delta_db": -1.044617963909947}], "active_band_count": 3, "active_band_mae_db": 0.4623753609250949}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.0001332544033329483, "log_rms_mae_db": 1.6041880435415037, "rms_envelope_nrmse": 0.1005547001987965, "envelope_correlation": 0.9911605059880798, "rms_gain_error_db": -0.18067151666692255, "reference_mean_power": 0.00116782876738011, "output_mean_power": 0.0011202424289319815}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 5.939192180731074e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134913428254024, "output_power": 0.0003909258146079487, "reference_active": true, "delta_db": -0.24372071441246426}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.995066727507785e-05, "output_power": 6.436902909300881e-05, "reference_active": true, "delta_db": -0.3611490316036894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9671389178533067e-05, "output_power": 1.5835744907744395e-05, "reference_active": true, "delta_db": -0.9419653330346611}], "active_band_count": 3, "active_band_mae_db": 0.5156116930169382}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012167319573637772, "log_rms_mae_db": 0.9694643298850862, "rms_envelope_nrmse": 0.06559842446885696, "envelope_correlation": 0.9960247012937518, "rms_gain_error_db": -0.18067131787489155, "reference_mean_power": 0.0011747523056415098, "output_mean_power": 0.0011268839005019048}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.189128511363634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037925439414056246, "output_power": 0.00035569170281394094, "reference_active": true, "delta_db": -0.27856886778898177}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.625302583005344e-05, "output_power": 3.3862117290011025e-05, "reference_active": true, "delta_db": -0.2963015079597966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7116807182635842e-06, "output_power": 1.3468776664741858e-06, "reference_active": true, "delta_db": -1.0409460682171165}], "active_band_count": 3, "active_band_mae_db": 0.538605481321965}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 4.762512740172365e-05, "log_rms_mae_db": 1.5240116473817982, "rms_envelope_nrmse": 0.11331121358911182, "envelope_correlation": 0.9901284543871186, "rms_gain_error_db": -0.1605228688474034, "reference_mean_power": 0.00010302415332771722, "output_mean_power": 9.928571488816078e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 3.1359809483594164e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4569632781045495e-05, "output_power": 1.394564779387376e-05, "reference_active": true, "delta_db": -0.19009913146563762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.842890886644979e-06, "output_power": 3.414555928412801e-06, "reference_active": true, "delta_db": -0.5132382287252136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.113490153313903e-06, "output_power": 9.440922487515958e-07, "reference_active": true, "delta_db": -0.716719489104485}], "active_band_count": 3, "active_band_mae_db": 0.4733522830984454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 4.327669210129085e-05, "log_rms_mae_db": 0.840271496940516, "rms_envelope_nrmse": 0.05761599513586017, "envelope_correlation": 0.9974566530459952, "rms_gain_error_db": -0.16052612944764622, "reference_mean_power": 0.00010363449577271808, "output_mean_power": 9.987383484760729e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 2.7762525532534898e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3720292251105881e-05, "output_power": 1.3128295526864062e-05, "reference_active": true, "delta_db": -0.1915501780350369}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7263230443470765e-06, "output_power": 1.49027577599807e-06, "reference_active": true, "delta_db": -0.6385542570675125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.698769643844551e-08, "output_power": 7.344245073125933e-08, "reference_active": true, "delta_db": -1.2076948390945705}], "active_band_count": 3, "active_band_mae_db": 0.6792664247323733}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 1.6513968336769973e-05, "log_rms_mae_db": 1.7973136957282165, "rms_envelope_nrmse": 0.2608418939655096, "envelope_correlation": 0.9455690739679232, "rms_gain_error_db": 0.1777936062279329, "reference_mean_power": 8.517905641506151e-06, "output_mean_power": 8.87385208651671e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 4.8245079516023535e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4329635827636286e-06, "output_power": 2.3347951556288364e-06, "reference_active": true, "delta_db": -0.17886824806710028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1930715407234243e-06, "output_power": 1.0256895450543652e-06, "reference_active": true, "delta_db": -0.6565055752417602}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.14650683405709e-07, "output_power": 1.7960115696515368e-07, "reference_active": true, "delta_db": -2.4351954934695703}], "active_band_count": 3, "active_band_mae_db": 1.090189772259477}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 1.2321016663379126e-05, "log_rms_mae_db": 1.3254508212160252, "rms_envelope_nrmse": 0.2055809243800066, "envelope_correlation": 0.9622751347720364, "rms_gain_error_db": 0.17780137752207884, "reference_mean_power": 8.568334809688713e-06, "output_mean_power": 8.926404562940412e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 4.130413403311084e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.518595705395907e-06, "output_power": 2.3342113459813e-06, "reference_active": true, "delta_db": -0.33018282764839013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.297939167002702e-07, "output_power": 5.474838874081491e-07, "reference_active": true, "delta_db": -0.6082711908568994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9684693476821222e-08, "output_power": 1.364314132891379e-08, "reference_active": true, "delta_db": -3.3761819077515294}], "active_band_count": 3, "active_band_mae_db": 1.4382119754189395}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9229256927927382, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9c5b4a9a921a4286c577fe217af50d372c1856048819624444b3d385ca71cbe5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9880037019537269, "standard_deviation": 0.0017995308331980547, "minimum": 0.9854970637668404, "maximum": 0.9901284543871186, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6029039668884228, "standard_deviation": 0.033466808539745076, "minimum": 1.5240116473817982, "maximum": 1.628973121198808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9009640306442743, "standard_deviation": 0.22108622420080104, "minimum": 0.4733522830984454, "maximum": 1.2553103138136281, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12516156578976936, "standard_deviation": 0.008792879862352034, "minimum": 0.11331121358911182, "maximum": 0.13768489833314576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961649959071756, "standard_deviation": 0.0011326967688129218, "minimum": 0.9936798027949646, "maximum": 0.9974566530459952, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8786845657450214, "standard_deviation": 0.036521676419832116, "minimum": 0.835469235642051, "maximum": 0.9376884150069043, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0861664460225484, "standard_deviation": 0.24028435052877065, "minimum": 0.6792664247323733, "maximum": 1.4778141847045505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07090273880173122, "standard_deviation": 0.009343232109730431, "minimum": 0.05761599513586017, "maximum": 0.09025552155154278, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9447697276361035, "standard_deviation": 0.002600831351606343, "minimum": 0.9410163707348805, "maximum": 0.9489774359829597, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8015513338467568, "standard_deviation": 0.023964805806846744, "minimum": 1.749977763802212, "maximum": 1.8282282763141169, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7628892564426568, "standard_deviation": 0.30788110367805877, "minimum": 0.3624118332609725, "maximum": 1.1785293887191888, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2650575322429011, "standard_deviation": 0.0054247360370364185, "minimum": 0.25974875240875384, "maximum": 0.2737842003637736, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9614651934321989, "standard_deviation": 0.00208878658532903, "minimum": 0.9588010350868834, "maximum": 0.9652721163318164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.325821295226297, "standard_deviation": 0.03367105482920417, "minimum": 1.260215401718764, "maximum": 1.3594654089425988, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0069162420470348, "standard_deviation": 0.4214392852808503, "minimum": 0.28759365499836304, "maximum": 1.515625038301339, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20996669387724862, "standard_deviation": 0.004443476272181657, "minimum": 0.20513387520230192, "maximum": 0.2179614485038444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925927805647659, "standard_deviation": 0.0007603117810361303, "minimum": 0.9911605059880798, "maximum": 0.9934825682442998, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5549975063213517, "standard_deviation": 0.04423535537552519, "minimum": 1.5113470850625927, "maximum": 1.640757109121831, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5900011967432219, "standard_deviation": 0.24064054141512342, "minimum": 0.27852108896688854, "maximum": 0.9408098490835796, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09213751425399874, "standard_deviation": 0.004322641907153402, "minimum": 0.08728093688048776, "maximum": 0.1005547001987965, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974500248247276, "standard_deviation": 0.0006395395922559071, "minimum": 0.9960247012937518, "maximum": 0.9980363054231215, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9504648070121597, "standard_deviation": 0.039123483801774495, "minimum": 0.8980508515323089, "maximum": 1.0215374546246279, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7931943711966183, "standard_deviation": 0.31715467818345594, "minimum": 0.4609719350008225, "maximum": 1.296102108136123, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.053152154200644536, "standard_deviation": 0.005617514336642896, "minimum": 0.047697718090154426, "maximum": 0.06559842446885696, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23848544647095887, "standard_deviation": 0.027877499021752756, "minimum": 0.18803792226380275, "maximum": 0.27681808381021317, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.710624692888973, "standard_deviation": 0.0054474847659113194, "minimum": 5.703355662919577, "maximum": 5.718612813912459, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.30411211295922613, "standard_deviation": 0.013516144118296776, "minimum": 0.2742514716519653, "maximum": 0.3167907988202419, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11576900574004229, "standard_deviation": 0.011935989966333426, "minimum": -0.1296677239683325, "maximum": -0.09764371203329665, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941222228940179, "standard_deviation": 0.0004346715001717348, "minimum": 0.9936000157630474, "maximum": 0.9949056970950025, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9017665051798425, "standard_deviation": 0.009518980187544155, "minimum": 0.8878800187749287, "maximum": 0.9148347795074426, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.47757513890104775, "standard_deviation": 0.19821723005408612, "minimum": 0.20417647982973688, "maximum": 0.7431887221409967, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07260871754799735, "standard_deviation": 0.002718199742598663, "minimum": 0.06757820465497123, "maximum": 0.0757319975442994, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.998085544457598, "standard_deviation": 0.0003085738058959088, "minimum": 0.9973951653051819, "maximum": 0.998421358139044, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46546170171557966, "standard_deviation": 0.017320883386721823, "minimum": 0.445722153724591, "maximum": 0.49194861130128725, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5946816629143508, "standard_deviation": 0.2614076639174667, "minimum": 0.17989393119345443, "maximum": 0.8948724446356202, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04023423671574428, "standard_deviation": 0.002912293723051171, "minimum": 0.03650081683716654, "maximum": 0.04637295616971104, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284449-0020", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.5837957859039307, "predicted_mos": 2.5837957859039307, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "821356037efee9535dfde4a54c41b14969ecd9d86f0dfdb22620eea85b20ae97", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7416461003012955, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9033874130712042, "全帯域_10ms_rms_envelope_nrmse": 0.07751333248758116, "全帯域_10ms_envelope_correlation": 0.9933091706145596, "全帯域_10ms_modulation_mae_db": 0.45585565972680014, "全帯域_40ms_log_rms_mae_db": 0.4514953766436081, "全帯域_40ms_rms_envelope_nrmse": 0.048155678207891676, "全帯域_40ms_envelope_correlation": 0.9972013039331721, "全帯域_40ms_modulation_mae_db": 0.4838506626351422, "300–1000 Hz_10ms_log_rms_mae_db": 1.5793723798840753, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10459714608474718, "300–1000 Hz_10ms_envelope_correlation": 0.9904229498123366, "300–1000 Hz_10ms_modulation_mae_db": 0.5508716288366361, "300–1000 Hz_40ms_log_rms_mae_db": 0.9588232342070842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06862504537513328, "300–1000 Hz_40ms_envelope_correlation": 0.9956308705324182, "300–1000 Hz_40ms_modulation_mae_db": 0.6486491459375133, "1000–3000 Hz_10ms_log_rms_mae_db": 1.536231261585706, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11555516590582941, "1000–3000 Hz_10ms_envelope_correlation": 0.9898041974441554, "1000–3000 Hz_10ms_modulation_mae_db": 0.5995950341140154, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8169195821338908, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05893700348167206, "1000–3000 Hz_40ms_envelope_correlation": 0.9974741709795418, "1000–3000 Hz_40ms_modulation_mae_db": 0.8332363786899352, "3000–6000 Hz_10ms_log_rms_mae_db": 1.659116488937328, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.247288848020163, "3000–6000 Hz_10ms_envelope_correlation": 0.950119600745237, "3000–6000 Hz_10ms_modulation_mae_db": 0.9194099203021485, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2611421497241468, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19856808162336176, "3000–6000 Hz_40ms_envelope_correlation": 0.9638642390169878, "3000–6000 Hz_40ms_modulation_mae_db": 1.2458279275942676, "mr_stft_db": 5.587784632436253, "band_mae_db": 0.25716687564729784, "rms_gain_error_db": -0.1258699827269314, "output_peak": 0.30975357127486236}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8700297494249928, "全帯域_10ms_rms_envelope_nrmse": 0.0700490331780173, "全帯域_10ms_envelope_correlation": 0.994474487248436, "全帯域_10ms_modulation_mae_db": 0.17503474374246095, "全帯域_40ms_log_rms_mae_db": 0.4428393323170246, "全帯域_40ms_rms_envelope_nrmse": 0.03894597412334393, "全帯域_40ms_envelope_correlation": 0.9981370563372342, "全帯域_40ms_modulation_mae_db": 0.28398079258724673, "300–1000 Hz_10ms_log_rms_mae_db": 1.525520391255163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0926175693005308, "300–1000 Hz_10ms_envelope_correlation": 0.9925047420127411, "300–1000 Hz_10ms_modulation_mae_db": 0.24629930935516522, "300–1000 Hz_40ms_log_rms_mae_db": 0.929681417894696, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0522292394822174, "300–1000 Hz_40ms_envelope_correlation": 0.9975445886669291, "300–1000 Hz_40ms_modulation_mae_db": 0.4948322055115948, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5865178896389307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11685022233338958, "1000–3000 Hz_10ms_envelope_correlation": 0.9895554325377799, "1000–3000 Hz_10ms_modulation_mae_db": 0.7684295144904206, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8404317086875405, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0613937064509441, "1000–3000 Hz_40ms_envelope_correlation": 0.9972098384509727, "1000–3000 Hz_40ms_modulation_mae_db": 0.9420032870483207, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6658508606706524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25577971460181265, "3000–6000 Hz_10ms_envelope_correlation": 0.947749781013256, "3000–6000 Hz_10ms_modulation_mae_db": 1.3604832694998266, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3021627717482447, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2105562194121333, "3000–6000 Hz_40ms_envelope_correlation": 0.9603385313331738, "3000–6000 Hz_40ms_modulation_mae_db": 1.6932599866225067, "mr_stft_db": 5.603546025590192, "band_mae_db": 0.18474339655890046, "rms_gain_error_db": -0.09001572331569908, "output_peak": 0.30321089435549736}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9217355109566557, "全帯域_10ms_rms_envelope_nrmse": 0.07227808485280703, "全帯域_10ms_envelope_correlation": 0.9941893682037262, "全帯域_10ms_modulation_mae_db": 0.7637019242160249, "全帯域_40ms_log_rms_mae_db": 0.49688887611593413, "全帯域_40ms_rms_envelope_nrmse": 0.04047955208789006, "全帯域_40ms_envelope_correlation": 0.9980939356647974, "全帯域_40ms_modulation_mae_db": 0.9098461008676906, "300–1000 Hz_10ms_log_rms_mae_db": 1.5573030297834645, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09142556561381943, "300–1000 Hz_10ms_envelope_correlation": 0.9927915149180114, "300–1000 Hz_10ms_modulation_mae_db": 1.0121687902656904, "300–1000 Hz_40ms_log_rms_mae_db": 0.9448484147825957, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052791723066090014, "300–1000 Hz_40ms_envelope_correlation": 0.9975883044604967, "300–1000 Hz_40ms_modulation_mae_db": 1.4024110605699083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5939894063860194, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12133908205707924, "1000–3000 Hz_10ms_envelope_correlation": 0.9889238918882497, "1000–3000 Hz_10ms_modulation_mae_db": 1.0180060543260712, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9163059187897705, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07534068585322036, "1000–3000 Hz_40ms_envelope_correlation": 0.9958915176466083, "1000–3000 Hz_40ms_modulation_mae_db": 1.4029766474705028, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6766952551999408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25187220538370086, "3000–6000 Hz_10ms_envelope_correlation": 0.9474777581349793, "3000–6000 Hz_10ms_modulation_mae_db": 0.8968331312875746, "3000–6000 Hz_40ms_log_rms_mae_db": 1.358058103168363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20803360110279426, "3000–6000 Hz_40ms_envelope_correlation": 0.9598283099605538, "3000–6000 Hz_40ms_modulation_mae_db": 1.1691366040859579, "mr_stft_db": 5.596813117646888, "band_mae_db": 0.2893917913372719, "rms_gain_error_db": -0.1232853410547941, "output_peak": 0.3091087059402362}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9337187065094713, "全帯域_10ms_rms_envelope_nrmse": 0.07628651521617126, "全帯域_10ms_envelope_correlation": 0.9934997863348763, "全帯域_10ms_modulation_mae_db": 0.7111359438789698, "全帯域_40ms_log_rms_mae_db": 0.5060830344596484, "全帯域_40ms_rms_envelope_nrmse": 0.042963058741992105, "全帯域_40ms_envelope_correlation": 0.9978031174001026, "全帯域_40ms_modulation_mae_db": 0.8845906893186033, "300–1000 Hz_10ms_log_rms_mae_db": 1.666971772187981, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09718590416362967, "300–1000 Hz_10ms_envelope_correlation": 0.9916336895938336, "300–1000 Hz_10ms_modulation_mae_db": 0.7965891937806427, "300–1000 Hz_40ms_log_rms_mae_db": 1.0441015718693567, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05688870112800207, "300–1000 Hz_40ms_envelope_correlation": 0.9969193612316415, "300–1000 Hz_40ms_modulation_mae_db": 1.08727828356863, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5899730299938775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1339439365823752, "1000–3000 Hz_10ms_envelope_correlation": 0.9861167757970362, "1000–3000 Hz_10ms_modulation_mae_db": 0.9705859912236603, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8765329279971971, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0707103113460852, "1000–3000 Hz_40ms_envelope_correlation": 0.9961233714905483, "1000–3000 Hz_40ms_modulation_mae_db": 1.177919475805762, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7081989613029833, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25174837824821295, "3000–6000 Hz_10ms_envelope_correlation": 0.9492389355950263, "3000–6000 Hz_10ms_modulation_mae_db": 0.6181479849339535, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3663465078614898, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2068343050565917, "3000–6000 Hz_40ms_envelope_correlation": 0.9614595246539825, "3000–6000 Hz_40ms_modulation_mae_db": 0.80027676429728, "mr_stft_db": 5.61109276788866, "band_mae_db": 0.2360480994156371, "rms_gain_error_db": -0.12065644218840614, "output_peak": 0.3091558948307806}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8991874884241753, "全帯域_10ms_rms_envelope_nrmse": 0.07475216690692738, "全帯域_10ms_envelope_correlation": 0.9937395370835266, "全帯域_10ms_modulation_mae_db": 0.2688736254550999, "全帯域_40ms_log_rms_mae_db": 0.4448643154658746, "全帯域_40ms_rms_envelope_nrmse": 0.04068946995719139, "全帯域_40ms_envelope_correlation": 0.9980149271727593, "全帯域_40ms_modulation_mae_db": 0.1602119441251109, "300–1000 Hz_10ms_log_rms_mae_db": 1.5768424897072135, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09753249095900611, "300–1000 Hz_10ms_envelope_correlation": 0.9916226934225484, "300–1000 Hz_10ms_modulation_mae_db": 0.4740413066320748, "300–1000 Hz_40ms_log_rms_mae_db": 0.9222789266003172, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05574965950045547, "300–1000 Hz_40ms_envelope_correlation": 0.9971202889993086, "300–1000 Hz_40ms_modulation_mae_db": 0.5166500907403323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6523267638819177, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1307874719418065, "1000–3000 Hz_10ms_envelope_correlation": 0.9868207525312118, "1000–3000 Hz_10ms_modulation_mae_db": 1.2216847274321456, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9514129761800788, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07213747063976324, "1000–3000 Hz_40ms_envelope_correlation": 0.996028030902645, "1000–3000 Hz_40ms_modulation_mae_db": 1.4290965205883275, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6424417847786665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25118797647879487, "3000–6000 Hz_10ms_envelope_correlation": 0.9492095602702468, "3000–6000 Hz_10ms_modulation_mae_db": 0.8460544418604465, "3000–6000 Hz_40ms_log_rms_mae_db": 1.30180615395672, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19895977797105555, "3000–6000 Hz_40ms_envelope_correlation": 0.9642187203740024, "3000–6000 Hz_40ms_modulation_mae_db": 1.2899692760366863, "mr_stft_db": 5.5955799538797635, "band_mae_db": 0.2107460803041971, "rms_gain_error_db": -0.1076233841676987, "output_peak": 0.3127741251185838}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.882442870039221, "全帯域_10ms_rms_envelope_nrmse": 0.06875564472250015, "全帯域_10ms_envelope_correlation": 0.9947494656702185, "全帯域_10ms_modulation_mae_db": 0.38651479430086305, "全帯域_40ms_log_rms_mae_db": 0.46604088787034204, "全帯域_40ms_rms_envelope_nrmse": 0.03813929640169123, "全帯域_40ms_envelope_correlation": 0.9983061218178854, "全帯域_40ms_modulation_mae_db": 0.6374204129495985, "300–1000 Hz_10ms_log_rms_mae_db": 1.5688515064661002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09338914581898325, "300–1000 Hz_10ms_envelope_correlation": 0.9923180716249095, "300–1000 Hz_10ms_modulation_mae_db": 0.37584072998594636, "300–1000 Hz_40ms_log_rms_mae_db": 0.9239841974501093, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05335228873241426, "300–1000 Hz_40ms_envelope_correlation": 0.9973741967689593, "300–1000 Hz_40ms_modulation_mae_db": 0.6929276859891962, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5997586379011148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11881821795127241, "1000–3000 Hz_10ms_envelope_correlation": 0.9897786248063333, "1000–3000 Hz_10ms_modulation_mae_db": 0.8927303607633289, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8822810708029278, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07095765748826302, "1000–3000 Hz_40ms_envelope_correlation": 0.9968581521465179, "1000–3000 Hz_40ms_modulation_mae_db": 1.057456173816571, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6815008033211094, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2543165347199845, "3000–6000 Hz_10ms_envelope_correlation": 0.9506200090365268, "3000–6000 Hz_10ms_modulation_mae_db": 0.30659052802377396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3533085017702327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20464049132913853, "3000–6000 Hz_40ms_envelope_correlation": 0.9644310245247965, "3000–6000 Hz_40ms_modulation_mae_db": 0.6038923118889791, "mr_stft_db": 5.6096050262079835, "band_mae_db": 0.24734599112302877, "rms_gain_error_db": -0.11648590464026351, "output_peak": 0.30764601881271575}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8943338888549549, "全帯域_10ms_rms_envelope_nrmse": 0.07272420189258852, "全帯域_10ms_envelope_correlation": 0.9941033657577383, "全帯域_10ms_modulation_mae_db": 0.33922084511029543, "全帯域_40ms_log_rms_mae_db": 0.4644038652622099, "全帯域_40ms_rms_envelope_nrmse": 0.04155480412460489, "全帯域_40ms_envelope_correlation": 0.9979475367576657, "全帯域_40ms_modulation_mae_db": 0.45731393343755916, "300–1000 Hz_10ms_log_rms_mae_db": 1.5610096309380441, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09067864336990078, "300–1000 Hz_10ms_envelope_correlation": 0.9927945949257141, "300–1000 Hz_10ms_modulation_mae_db": 0.5103015356613811, "300–1000 Hz_40ms_log_rms_mae_db": 0.9028636213954937, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.049883969182444664, "300–1000 Hz_40ms_envelope_correlation": 0.9977442970191743, "300–1000 Hz_40ms_modulation_mae_db": 0.700902783284767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6101010950860728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11983337999600362, "1000–3000 Hz_10ms_envelope_correlation": 0.989285797658949, "1000–3000 Hz_10ms_modulation_mae_db": 0.712293630035477, "1000–3000 Hz_40ms_log_rms_mae_db": 0.892276675058669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06708229084185494, "1000–3000 Hz_40ms_envelope_correlation": 0.9969681805964492, "1000–3000 Hz_40ms_modulation_mae_db": 0.8684391547825454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6308293461291359, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2500372866346569, "3000–6000 Hz_10ms_envelope_correlation": 0.950247946019574, "3000–6000 Hz_10ms_modulation_mae_db": 0.4058779339636027, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2752079380584285, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20640394981581608, "3000–6000 Hz_40ms_envelope_correlation": 0.9619643825972605, "3000–6000 Hz_40ms_modulation_mae_db": 0.3738845094979684, "mr_stft_db": 5.594414856470287, "band_mae_db": 0.23938403712719522, "rms_gain_error_db": -0.11460854560576686, "output_peak": 0.314710372513487}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9101417485353139, "全帯域_10ms_rms_envelope_nrmse": 0.07726379501136166, "全帯域_10ms_envelope_correlation": 0.9933293753914477, "全帯域_10ms_modulation_mae_db": 0.6346656198347801, "全帯域_40ms_log_rms_mae_db": 0.4617521266662726, "全帯域_40ms_rms_envelope_nrmse": 0.03937566650177845, "全帯域_40ms_envelope_correlation": 0.9981874941893698, "全帯域_40ms_modulation_mae_db": 0.835811183094132, "300–1000 Hz_10ms_log_rms_mae_db": 1.508901938649186, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09913622230614065, "300–1000 Hz_10ms_envelope_correlation": 0.9913453827193864, "300–1000 Hz_10ms_modulation_mae_db": 0.9490260293906049, "300–1000 Hz_40ms_log_rms_mae_db": 0.9142581211346292, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.051200607485478104, "300–1000 Hz_40ms_envelope_correlation": 0.9976352246280146, "300–1000 Hz_40ms_modulation_mae_db": 1.2993289712840481, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5956425735652842, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13359789743647155, "1000–3000 Hz_10ms_envelope_correlation": 0.9864751297896622, "1000–3000 Hz_10ms_modulation_mae_db": 0.8053186419386615, "1000–3000 Hz_40ms_log_rms_mae_db": 0.888255235212806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08172069552128806, "1000–3000 Hz_40ms_envelope_correlation": 0.9951054435338692, "1000–3000 Hz_40ms_modulation_mae_db": 0.9693818499320851, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6653647907329694, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2606742316318383, "3000–6000 Hz_10ms_envelope_correlation": 0.9453768644997882, "3000–6000 Hz_10ms_modulation_mae_db": 0.6369822529430039, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3106001303531418, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.210443654481572, "3000–6000 Hz_40ms_envelope_correlation": 0.9600454622092476, "3000–6000 Hz_40ms_modulation_mae_db": 0.8342916408637423, "mr_stft_db": 5.600690520549361, "band_mae_db": 0.2588656515408085, "rms_gain_error_db": -0.11257511474794647, "output_peak": 0.2756139779343096}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 81520, "output_samples": 81520, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.385079609693173, "spectral_convergence": 0.19019242901019465, "spectral_convergence_status": "ok", "amplitude_floor": 1.0222158521079025e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.492096843708652, "spectral_convergence": 0.18952291304542196, "spectral_convergence_status": "ok", "amplitude_floor": 7.564059658380168e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8861774439069325, "spectral_convergence": 0.19195552055286347, "spectral_convergence_status": "ok", "amplitude_floor": 6.35834815641291e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.932339290337003e-05, "output_power": 9.580116295245563e-05, "delta_db": 0.304055699452532, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011480494900330098, "output_power": 0.001130549234996492, "delta_db": -0.06671129544547987, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005518546578574125, "output_power": 0.0005149591596526336, "delta_db": -0.30051925099783217, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.085077476738271e-05, "output_power": 8.747527704851018e-05, "delta_db": -0.16443308646046212, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.0710168905088768e-05, "output_power": 1.9058188333287516e-05, "delta_db": -0.3610202652849248, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8163123695227053e-05, "output_power": 2.6004868201690015e-05, "delta_db": -0.346261656242556, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25716687564729784, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1258699827269314, "pitch": null, "output_rms_dbfs": -27.27265665689488, "output_peak": 0.30975357127486236, "reference_rms_dbfs": -27.14678667416795, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.587784632436253}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 81520, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.00013747273046535536, "log_rms_mae_db": 0.9033874130712042, "rms_envelope_nrmse": 0.07751333248758116, "envelope_correlation": 0.9933091706145596, "rms_gain_error_db": -0.1258699402334819, "reference_mean_power": 0.0019308462257719779, "output_mean_power": 0.0018756884042390677}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 6.283614364116833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041565087617635905, "output_power": 0.0003975851081938903, "reference_active": true, "delta_db": -0.19298591064672482}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.15171169386713e-05, "output_power": 7.72897879421876e-05, "reference_active": true, "delta_db": -0.23126695596865546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.519660659304176e-05, "output_power": 2.0277324156737757e-05, "reference_active": true, "delta_db": -0.94331411256502}], "active_band_count": 3, "active_band_mae_db": 0.45585565972680014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012452330782545057, "log_rms_mae_db": 0.4514953766436081, "rms_envelope_nrmse": 0.048155678207891676, "envelope_correlation": 0.9972013039331721, "rms_gain_error_db": -0.12586995838894507, "reference_mean_power": 0.0019422928952156904, "output_mean_power": 0.0018868080727171706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.257129910319434e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003707029677887353, "output_power": 0.00035337169572595157, "reference_active": true, "delta_db": -0.20794302378207355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2635280867150454e-05, "output_power": 4.092670500650682e-05, "reference_active": true, "delta_db": -0.1776234714864312}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.869258984918021e-06, "output_power": 1.462415992010598e-06, "reference_active": true, "delta_db": -1.0659854926369219}], "active_band_count": 3, "active_band_mae_db": 0.4838506626351422}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.0001332544033329483, "log_rms_mae_db": 1.5793723798840753, "rms_envelope_nrmse": 0.10459714608474718, "envelope_correlation": 0.9904229498123366, "rms_gain_error_db": -0.18283851551888186, "reference_mean_power": 0.00116782876738011, "output_mean_power": 0.0011196836010816328}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 5.939192180731074e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134913428254024, "output_power": 0.00038981133505524293, "reference_active": true, "delta_db": -0.25611957743304165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.995066727507785e-05, "output_power": 6.361894935104724e-05, "reference_active": true, "delta_db": -0.4120536909986522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9671389178533067e-05, "output_power": 1.568161794179647e-05, "reference_active": true, "delta_db": -0.9844416180782145}], "active_band_count": 3, "active_band_mae_db": 0.5508716288366361}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012167319573637772, "log_rms_mae_db": 0.9588232342070842, "rms_envelope_nrmse": 0.06862504537513328, "envelope_correlation": 0.9956308705324182, "rms_gain_error_db": -0.18283827100012143, "reference_mean_power": 0.0011747523056415098, "output_mean_power": 0.0011263217714426188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.189128511363634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037925439414056246, "output_power": 0.00035465952544393436, "reference_active": true, "delta_db": -0.2911899271689964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.625302583005344e-05, "output_power": 3.337200867599033e-05, "reference_active": true, "delta_db": -0.35961912645515304}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7116807182635842e-06, "output_power": 1.2703076691528066e-06, "reference_active": true, "delta_db": -1.2951383841883903}], "active_band_count": 3, "active_band_mae_db": 0.6486491459375133}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 4.762512740172365e-05, "log_rms_mae_db": 1.536231261585706, "rms_envelope_nrmse": 0.11555516590582941, "envelope_correlation": 0.9898041974441554, "rms_gain_error_db": -0.2058635577010811, "reference_mean_power": 0.00010302415332771722, "output_mean_power": 9.825455619412263e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 3.1359809483594164e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4569632781045495e-05, "output_power": 1.3826494600963575e-05, "reference_active": true, "delta_db": -0.22736517379159893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.842890886644979e-06, "output_power": 3.3549943662210546e-06, "reference_active": true, "delta_db": -0.5896625824378966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.113490153313903e-06, "output_power": 8.881997493372752e-07, "reference_active": true, "delta_db": -0.9817573461125506}], "active_band_count": 3, "active_band_mae_db": 0.5995950341140154}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 4.327669210129085e-05, "log_rms_mae_db": 0.8169195821338908, "rms_envelope_nrmse": 0.05893700348167206, "envelope_correlation": 0.9974741709795418, "rms_gain_error_db": -0.20586690297832777, "reference_mean_power": 0.00010363449577271808, "output_mean_power": 9.883656614729972e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 2.7762525532534898e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3720292251105881e-05, "output_power": 1.3024744132326334e-05, "reference_active": true, "delta_db": -0.2259416176837359}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7263230443470765e-06, "output_power": 1.4682144775790595e-06, "reference_active": true, "delta_db": -0.7033256569415102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.698769643844551e-08, "output_power": 6.755732709014844e-08, "reference_active": true, "delta_db": -1.5704418614445594}], "active_band_count": 3, "active_band_mae_db": 0.8332363786899352}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 1.6513968336769973e-05, "log_rms_mae_db": 1.659116488937328, "rms_envelope_nrmse": 0.247288848020163, "envelope_correlation": 0.950119600745237, "rms_gain_error_db": 0.011781066954772313, "reference_mean_power": 8.517905641506151e-06, "output_mean_power": 8.541043455434521e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 4.8245079516023535e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4329635827636286e-06, "output_power": 2.376331348103191e-06, "reference_active": true, "delta_db": -0.10228611169852228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1930715407234243e-06, "output_power": 1.1091275849613602e-06, "reference_active": true, "delta_db": -0.31684979554243253}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.14650683405709e-07, "output_power": 1.8361970866026165e-07, "reference_active": true, "delta_db": -2.3390938536654904}], "active_band_count": 3, "active_band_mae_db": 0.9194099203021485}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 1.2321016663379126e-05, "log_rms_mae_db": 1.2611421497241468, "rms_envelope_nrmse": 0.19856808162336176, "envelope_correlation": 0.9638642390169878, "rms_gain_error_db": 0.011788496723914361, "reference_mean_power": 8.568334809688713e-06, "output_mean_power": 8.591624306337348e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 4.130413403311084e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.518595705395907e-06, "output_power": 2.387364454887936e-06, "reference_active": true, "delta_db": -0.23239735051002486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.297939167002702e-07, "output_power": 5.852790236220056e-07, "reference_active": true, "delta_db": -0.31835502092084633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9684693476821222e-08, "output_power": 1.4251462278613012e-08, "reference_active": true, "delta_db": -3.186731411351932}], "active_band_count": 3, "active_band_mae_db": 1.2458279275942676}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9222912892232354, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "777523adee38e83f73665c467dcee3347013ee0c4efe8ed0731d9960ca959759", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9883450753066721, "standard_deviation": 0.0015881088596713295, "minimum": 0.9861167757970362, "maximum": 0.9898041974441554, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5955675822548654, "standard_deviation": 0.0318184219021091, "minimum": 1.536231261585706, "maximum": 1.6523267638819177, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8735804942904726, "standard_deviation": 0.19588990251547542, "minimum": 0.5995950341140154, "maximum": 1.2216847274321456, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12384067177552845, "standard_deviation": 0.007658746790140708, "minimum": 0.11555516590582941, "maximum": 0.1339439365823752, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9964573382183941, "standard_deviation": 0.000798697860675688, "minimum": 0.9951054435338692, "maximum": 0.9974741709795418, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.88305201185786, "standard_deviation": 0.041617807009170386, "minimum": 0.8169195821338908, "maximum": 0.9514129761800788, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0850636860167562, "standard_deviation": 0.23083802606766513, "minimum": 0.8332363786899352, "maximum": 1.4290965205883275, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06978497770288637, "standard_deviation": 0.007333771104935581, "minimum": 0.05893700348167206, "maximum": 0.08172069552128806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9487550569143293, "standard_deviation": 0.0017753756882884436, "minimum": 0.9453768644997882, "maximum": 0.9506200090365268, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6662497863840982, "standard_deviation": 0.023828822381315624, "minimum": 1.6308293461291359, "maximum": 1.7081989613029833, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7487974328517912, "standard_deviation": 0.3332425799274793, "minimum": 0.30659052802377396, "maximum": 1.3604832694998266, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25286314696489554, "standard_deviation": 0.004065901980747114, "minimum": 0.247288848020163, "maximum": 0.2606742316318383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9620187743337505, "standard_deviation": 0.0019233706863048463, "minimum": 0.9598283099605538, "maximum": 0.9644310245247965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.316079032080096, "standard_deviation": 0.03926284690477972, "minimum": 1.2611421497241468, "maximum": 1.3663465078614898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0013173776109237, "standard_deviation": 0.4259421164981383, "minimum": 0.3738845094979684, "maximum": 1.6932599866225067, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2055550100990579, "standard_deviation": 0.00463776804406734, "minimum": 0.19856808162336176, "maximum": 0.2105562194121333, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9919292048786852, "standard_deviation": 0.0008251294332517933, "minimum": 0.9904229498123366, "maximum": 0.9927945949257141, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5680966423589036, "standard_deviation": 0.04700147585791, "minimum": 1.508901938649186, "maximum": 1.666971772187981, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6143923154885177, "standard_deviation": 0.2752580718163287, "minimum": 0.24629930935516522, "maximum": 1.0121687902656904, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09582033595209473, "standard_deviation": 0.004701719417242203, "minimum": 0.09067864336990078, "maximum": 0.10459714608474718, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9971946415383678, "standard_deviation": 0.0006899891803218074, "minimum": 0.9956308705324182, "maximum": 0.9977442970191743, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9426049381667853, "standard_deviation": 0.044528391213699774, "minimum": 0.9028636213954937, "maximum": 1.0441015718693567, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8553725283607487, "standard_deviation": 0.3561369636567105, "minimum": 0.4948322055115948, "maximum": 1.4024110605699083, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0550901542440294, "standard_deviation": 0.0059231233143732034, "minimum": 0.049883969182444664, "maximum": 0.06862504537513328, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2404614903817921, "standard_deviation": 0.03177488731588002, "minimum": 0.18474339655890046, "maximum": 0.2893917913372719, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.599940862583674, "standard_deviation": 0.007914103377908778, "minimum": 5.587784632436253, "maximum": 5.61109276788866, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3052466950975591, "standard_deviation": 0.012448358822198902, "minimum": 0.2756139779343096, "maximum": 0.314710372513487, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11389005480593828, "standard_deviation": 0.0113125257546142, "minimum": -0.1258699827269314, "maximum": -0.09001572331569908, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9939243195380661, "standard_deviation": 0.0005388610640826094, "minimum": 0.9933091706145596, "maximum": 0.9947494656702185, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9018721719769986, "standard_deviation": 0.020476141988571058, "minimum": 0.8700297494249928, "maximum": 0.9337187065094713, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.46687539453316174, "standard_deviation": 0.21486652834535283, "minimum": 0.17503474374246095, "maximum": 0.7637019242160249, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07370284678349431, "standard_deviation": 0.003290449890029376, "minimum": 0.06875564472250015, "maximum": 0.07751333248758116, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9979614366591233, "standard_deviation": 0.00034314289290215765, "minimum": 0.9972013039331721, "maximum": 0.9983061218178854, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4667959768501143, "standard_deviation": 0.023206573018276326, "minimum": 0.4428393323170246, "maximum": 0.5060830344596484, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5816282148768854, "standard_deviation": 0.2824472501557462, "minimum": 0.1602119441251109, "maximum": 0.9098461008676906, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.041287937518297964, "standard_deviation": 0.003164767546644005, "minimum": 0.03813929640169123, "maximum": 0.048155678207891676, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284449-0020", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.283601760864258, "predicted_mos": 2.283601760864258, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "e798ae60b8173b1a2a38121c5ea67ab9192853310c365bbae735bb97ff903928", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8569195000454783, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8887606075036358, "全帯域_10ms_rms_envelope_nrmse": 0.06806118527283841, "全帯域_10ms_envelope_correlation": 0.9949147612150383, "全帯域_10ms_modulation_mae_db": 0.7297455182585405, "全帯域_40ms_log_rms_mae_db": 0.4473370048830174, "全帯域_40ms_rms_envelope_nrmse": 0.0389345674689342, "全帯域_40ms_envelope_correlation": 0.9982780848397411, "全帯域_40ms_modulation_mae_db": 1.0483396247799164, "300–1000 Hz_10ms_log_rms_mae_db": 1.9050060278528511, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09059724201823346, "300–1000 Hz_10ms_envelope_correlation": 0.9928588841673124, "300–1000 Hz_10ms_modulation_mae_db": 0.8977420523280454, "300–1000 Hz_40ms_log_rms_mae_db": 1.1512003395413424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.053214571815750136, "300–1000 Hz_40ms_envelope_correlation": 0.9974351147708239, "300–1000 Hz_40ms_modulation_mae_db": 1.2552256023166226, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8951318751969617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13505973363478738, "1000–3000 Hz_10ms_envelope_correlation": 0.9862657836767347, "1000–3000 Hz_10ms_modulation_mae_db": 1.1237898700672944, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0550185070775027, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07935043434424627, "1000–3000 Hz_40ms_envelope_correlation": 0.9954437149278065, "1000–3000 Hz_40ms_modulation_mae_db": 1.1775592753825117, "3000–6000 Hz_10ms_log_rms_mae_db": 2.253603181244452, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30955468181155515, "3000–6000 Hz_10ms_envelope_correlation": 0.9266295795617806, "3000–6000 Hz_10ms_modulation_mae_db": 0.5142054135033881, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7289345811116488, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23535611910790602, "3000–6000 Hz_40ms_envelope_correlation": 0.953526739871094, "3000–6000 Hz_40ms_modulation_mae_db": 0.6786468645055671, "mr_stft_db": 6.106352532954046, "band_mae_db": 0.2487114878446144, "rms_gain_error_db": -0.12902787982150032, "output_peak": 0.28215253679100566}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.906582400367084, "全帯域_10ms_rms_envelope_nrmse": 0.07699264894618914, "全帯域_10ms_envelope_correlation": 0.9935650373765474, "全帯域_10ms_modulation_mae_db": 0.9997609173348478, "全帯域_40ms_log_rms_mae_db": 0.4877935687921632, "全帯域_40ms_rms_envelope_nrmse": 0.04686304206177511, "全帯域_40ms_envelope_correlation": 0.9975596573075382, "全帯域_40ms_modulation_mae_db": 1.36048934064942, "300–1000 Hz_10ms_log_rms_mae_db": 1.842669885486153, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10451906087229573, "300–1000 Hz_10ms_envelope_correlation": 0.990616498936981, "300–1000 Hz_10ms_modulation_mae_db": 1.086534675370945, "300–1000 Hz_40ms_log_rms_mae_db": 1.1122154671557687, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06709437312555253, "300–1000 Hz_40ms_envelope_correlation": 0.9960453272171405, "300–1000 Hz_40ms_modulation_mae_db": 1.506430786272323, "1000–3000 Hz_10ms_log_rms_mae_db": 1.901964157309051, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15005315074138723, "1000–3000 Hz_10ms_envelope_correlation": 0.9826162011316213, "1000–3000 Hz_10ms_modulation_mae_db": 0.7873279156865026, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1141538658764005, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08798222994992527, "1000–3000 Hz_40ms_envelope_correlation": 0.993848401395353, "1000–3000 Hz_40ms_modulation_mae_db": 0.9666989485720686, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3205543835741658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3157759053510727, "3000–6000 Hz_10ms_envelope_correlation": 0.9204027701595212, "3000–6000 Hz_10ms_modulation_mae_db": 0.834211686838489, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7669771920826653, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23988209315578618, "3000–6000 Hz_40ms_envelope_correlation": 0.9493446226342545, "3000–6000 Hz_40ms_modulation_mae_db": 1.2971114251300078, "mr_stft_db": 6.099362753473041, "band_mae_db": 0.17670159252852577, "rms_gain_error_db": -0.1676592860487883, "output_peak": 0.3062409856849761}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8943908637205437, "全帯域_10ms_rms_envelope_nrmse": 0.07592315087457735, "全帯域_10ms_envelope_correlation": 0.9937068888223692, "全帯域_10ms_modulation_mae_db": 0.573430281380112, "全帯域_40ms_log_rms_mae_db": 0.4740604584311618, "全帯域_40ms_rms_envelope_nrmse": 0.044460627760304496, "全帯域_40ms_envelope_correlation": 0.9978091174696445, "全帯域_40ms_modulation_mae_db": 0.8747523245196227, "300–1000 Hz_10ms_log_rms_mae_db": 1.8632739535958933, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09571797181609594, "300–1000 Hz_10ms_envelope_correlation": 0.9922799888546452, "300–1000 Hz_10ms_modulation_mae_db": 0.7314770621542217, "300–1000 Hz_40ms_log_rms_mae_db": 1.193287818692408, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.058552256418643325, "300–1000 Hz_40ms_envelope_correlation": 0.9971981009686839, "300–1000 Hz_40ms_modulation_mae_db": 1.126461171011693, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8588841505130553, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14813959248256595, "1000–3000 Hz_10ms_envelope_correlation": 0.982958242079676, "1000–3000 Hz_10ms_modulation_mae_db": 0.5773759781017175, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0340139292839299, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08569280377679395, "1000–3000 Hz_40ms_envelope_correlation": 0.9940890388886137, "1000–3000 Hz_40ms_modulation_mae_db": 0.8464093690632429, "3000–6000 Hz_10ms_log_rms_mae_db": 2.381035209121421, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.352441207821484, "3000–6000 Hz_10ms_envelope_correlation": 0.9021347877340161, "3000–6000 Hz_10ms_modulation_mae_db": 0.7471519935740831, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8086791771892865, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26847548437043744, "3000–6000 Hz_40ms_envelope_correlation": 0.9373769829098897, "3000–6000 Hz_40ms_modulation_mae_db": 1.311449717038706, "mr_stft_db": 6.119110689509726, "band_mae_db": 0.2073396612866971, "rms_gain_error_db": -0.159673860412585, "output_peak": 0.3027837915021383}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9044259267066914, "全帯域_10ms_rms_envelope_nrmse": 0.07102674122431668, "全帯域_10ms_envelope_correlation": 0.9944778470258941, "全帯域_10ms_modulation_mae_db": 1.0502227521616974, "全帯域_40ms_log_rms_mae_db": 0.4669093136718396, "全帯域_40ms_rms_envelope_nrmse": 0.04009526218746074, "全帯域_40ms_envelope_correlation": 0.9982262242392818, "全帯域_40ms_modulation_mae_db": 1.1876606204711182, "300–1000 Hz_10ms_log_rms_mae_db": 1.7840243931647337, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0886602446675243, "300–1000 Hz_10ms_envelope_correlation": 0.9933790519344201, "300–1000 Hz_10ms_modulation_mae_db": 1.23899171156678, "300–1000 Hz_40ms_log_rms_mae_db": 1.1152100934197102, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0532615441821537, "300–1000 Hz_40ms_envelope_correlation": 0.9977037838547882, "300–1000 Hz_40ms_modulation_mae_db": 1.3765749582716407, "1000–3000 Hz_10ms_log_rms_mae_db": 1.924432875215619, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.147418250672046, "1000–3000 Hz_10ms_envelope_correlation": 0.983071453161507, "1000–3000 Hz_10ms_modulation_mae_db": 0.8109393318249586, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0971979489542494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08742836718039405, "1000–3000 Hz_40ms_envelope_correlation": 0.9936678090070488, "1000–3000 Hz_40ms_modulation_mae_db": 1.0840932483653232, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2953258572464383, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31667453303738496, "3000–6000 Hz_10ms_envelope_correlation": 0.9234597294683768, "3000–6000 Hz_10ms_modulation_mae_db": 0.7716982765182118, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7097660596461717, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2362301638177385, "3000–6000 Hz_40ms_envelope_correlation": 0.9535323251201202, "3000–6000 Hz_40ms_modulation_mae_db": 1.3276431709250434, "mr_stft_db": 6.097635326267896, "band_mae_db": 0.20855347275086059, "rms_gain_error_db": -0.14508432165990287, "output_peak": 0.3052490816161288}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.89919022168054, "全帯域_10ms_rms_envelope_nrmse": 0.0753143061029793, "全帯域_10ms_envelope_correlation": 0.9936164619835891, "全帯域_10ms_modulation_mae_db": 0.18005670065840143, "全帯域_40ms_log_rms_mae_db": 0.43453915830636997, "全帯域_40ms_rms_envelope_nrmse": 0.039356339202466624, "全帯域_40ms_envelope_correlation": 0.998121808324023, "全帯域_40ms_modulation_mae_db": 0.1978997083688888, "300–1000 Hz_10ms_log_rms_mae_db": 1.8217340053691475, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09708314526587607, "300–1000 Hz_10ms_envelope_correlation": 0.9916636305534222, "300–1000 Hz_10ms_modulation_mae_db": 0.20621025755008424, "300–1000 Hz_40ms_log_rms_mae_db": 1.1006921416966846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052776463548032164, "300–1000 Hz_40ms_envelope_correlation": 0.997395385589139, "300–1000 Hz_40ms_modulation_mae_db": 0.3258143306999382, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8768419530044063, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13202999863831003, "1000–3000 Hz_10ms_envelope_correlation": 0.9865271662300505, "1000–3000 Hz_10ms_modulation_mae_db": 0.6474309119131569, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0720450714330914, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07622792438472853, "1000–3000 Hz_40ms_envelope_correlation": 0.9952852719096386, "1000–3000 Hz_40ms_modulation_mae_db": 0.9770696025253697, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3728426671297056, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3245558604220461, "3000–6000 Hz_10ms_envelope_correlation": 0.9200704300072425, "3000–6000 Hz_10ms_modulation_mae_db": 0.24066573015920645, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7499909691116275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23409216457038198, "3000–6000 Hz_40ms_envelope_correlation": 0.9550083819707869, "3000–6000 Hz_40ms_modulation_mae_db": 0.670548445103352, "mr_stft_db": 6.104106476188438, "band_mae_db": 0.2131078027034082, "rms_gain_error_db": -0.09680812275201986, "output_peak": 0.3007566608460324}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8982475926315796, "全帯域_10ms_rms_envelope_nrmse": 0.06457939314738324, "全帯域_10ms_envelope_correlation": 0.9953354053119756, "全帯域_10ms_modulation_mae_db": 0.4740143441979354, "全帯域_40ms_log_rms_mae_db": 0.42347895421460585, "全帯域_40ms_rms_envelope_nrmse": 0.033046268203509394, "全帯域_40ms_envelope_correlation": 0.9987014786061283, "全帯域_40ms_modulation_mae_db": 0.6937844256071163, "300–1000 Hz_10ms_log_rms_mae_db": 1.789504618390612, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08336720205223556, "300–1000 Hz_10ms_envelope_correlation": 0.9939776851115718, "300–1000 Hz_10ms_modulation_mae_db": 0.6028168763591267, "300–1000 Hz_40ms_log_rms_mae_db": 1.0658554777887055, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04810074146589883, "300–1000 Hz_40ms_envelope_correlation": 0.9979468517202872, "300–1000 Hz_40ms_modulation_mae_db": 0.9438645069587822, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9255000908523805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13762174850697956, "1000–3000 Hz_10ms_envelope_correlation": 0.9852733619794456, "1000–3000 Hz_10ms_modulation_mae_db": 0.6435775370078107, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0288495198002052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07421954238477872, "1000–3000 Hz_40ms_envelope_correlation": 0.995511487026413, "1000–3000 Hz_40ms_modulation_mae_db": 0.8657298390848457, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3221820705118006, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31924328538910185, "3000–6000 Hz_10ms_envelope_correlation": 0.9260633108080436, "3000–6000 Hz_10ms_modulation_mae_db": 0.43621489551434617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7375736827406396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23674034389055074, "3000–6000 Hz_40ms_envelope_correlation": 0.9564852946417957, "3000–6000 Hz_40ms_modulation_mae_db": 0.9154393931123105, "mr_stft_db": 6.1163863548070445, "band_mae_db": 0.20717547373185194, "rms_gain_error_db": -0.0883302437495221, "output_peak": 0.2913841880911318}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9017773875702572, "全帯域_10ms_rms_envelope_nrmse": 0.07573371164518582, "全帯域_10ms_envelope_correlation": 0.9936494108127624, "全帯域_10ms_modulation_mae_db": 0.4167275585768129, "全帯域_40ms_log_rms_mae_db": 0.46839970982436147, "全帯域_40ms_rms_envelope_nrmse": 0.03788822076058495, "全帯域_40ms_envelope_correlation": 0.9984220937922731, "全帯域_40ms_modulation_mae_db": 0.6238821769808028, "300–1000 Hz_10ms_log_rms_mae_db": 1.8517936118873939, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09446499382040756, "300–1000 Hz_10ms_envelope_correlation": 0.9923654141238191, "300–1000 Hz_10ms_modulation_mae_db": 0.5533909352383819, "300–1000 Hz_40ms_log_rms_mae_db": 1.1462112536027842, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05379978915325977, "300–1000 Hz_40ms_envelope_correlation": 0.9976216253228077, "300–1000 Hz_40ms_modulation_mae_db": 0.8152563190032064, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8811250019236216, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14688881365780596, "1000–3000 Hz_10ms_envelope_correlation": 0.9832291071489834, "1000–3000 Hz_10ms_modulation_mae_db": 0.7294831937232967, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0800745527798994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08499562063935691, "1000–3000 Hz_40ms_envelope_correlation": 0.9940975643819135, "1000–3000 Hz_40ms_modulation_mae_db": 1.0122178905142938, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3383218564446504, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32469374567071263, "3000–6000 Hz_10ms_envelope_correlation": 0.9232559428328333, "3000–6000 Hz_10ms_modulation_mae_db": 0.5337293542595551, "3000–6000 Hz_40ms_log_rms_mae_db": 1.817119590591193, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.24437720968310644, "3000–6000 Hz_40ms_envelope_correlation": 0.9533939015102416, "3000–6000 Hz_40ms_modulation_mae_db": 0.6580174455074772, "mr_stft_db": 6.0916436888175545, "band_mae_db": 0.2243153822105941, "rms_gain_error_db": -0.13808254981073878, "output_peak": 0.2952255444557574}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9193652753942231, "全帯域_10ms_rms_envelope_nrmse": 0.07251141213392535, "全帯域_10ms_envelope_correlation": 0.9942505990511861, "全帯域_10ms_modulation_mae_db": 0.5869691633478543, "全帯域_40ms_log_rms_mae_db": 0.45007739983978184, "全帯域_40ms_rms_envelope_nrmse": 0.040677395040340396, "全帯域_40ms_envelope_correlation": 0.998182278434491, "全帯域_40ms_modulation_mae_db": 0.6629229097831518, "300–1000 Hz_10ms_log_rms_mae_db": 1.7988302769524558, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09210428432031997, "300–1000 Hz_10ms_envelope_correlation": 0.9929064786326103, "300–1000 Hz_10ms_modulation_mae_db": 0.7554417751753251, "300–1000 Hz_40ms_log_rms_mae_db": 1.1119998741151527, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057700950413787415, "300–1000 Hz_40ms_envelope_correlation": 0.9973115318670144, "300–1000 Hz_40ms_modulation_mae_db": 0.917327106885541, "1000–3000 Hz_10ms_log_rms_mae_db": 1.875953202263544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13562228636564755, "1000–3000 Hz_10ms_envelope_correlation": 0.9858252242328223, "1000–3000 Hz_10ms_modulation_mae_db": 1.2110733916721002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.011074032045198, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07704458119081604, "1000–3000 Hz_40ms_envelope_correlation": 0.9953152219688485, "1000–3000 Hz_40ms_modulation_mae_db": 1.5482700017352837, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3043183443642707, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.31904875783405395, "3000–6000 Hz_10ms_envelope_correlation": 0.9191317406888956, "3000–6000 Hz_10ms_modulation_mae_db": 0.26100417140971893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6965095310574299, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23865034010277328, "3000–6000 Hz_40ms_envelope_correlation": 0.950155704285968, "3000–6000 Hz_40ms_modulation_mae_db": 0.698100802612604, "mr_stft_db": 6.089760597156709, "band_mae_db": 0.25501095332045515, "rms_gain_error_db": -0.1477465321231141, "output_peak": 0.31508995859873085}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 81520, "output_samples": 81520, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.922287307427138, "spectral_convergence": 0.19191941707490232, "spectral_convergence_status": "ok", "amplitude_floor": 1.0222158521079025e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.098527633108243, "spectral_convergence": 0.19206010869363338, "spectral_convergence_status": "ok", "amplitude_floor": 7.564059658380168e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.298242658326755, "spectral_convergence": 0.19287277562492658, "spectral_convergence_status": "ok", "amplitude_floor": 6.35834815641291e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.932339290337003e-05, "output_power": 9.714869303138952e-05, "delta_db": 0.36471751123282814, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011480494900330098, "output_power": 0.0011220981233987161, "delta_db": -0.09929773946864033, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005518546578574125, "output_power": 0.0005177381623880082, "delta_db": -0.2771453457748365, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.085077476738271e-05, "output_power": 8.598164559625776e-05, "delta_db": -0.23922882494824355, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.0710168905088768e-05, "output_power": 1.957298825902881e-05, "delta_db": -0.2452650527769567, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8163123695227053e-05, "output_power": 2.9946237729484987e-05, "delta_db": 0.2666144528661812, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2487114878446144, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12902787982150032, "pitch": null, "output_rms_dbfs": -27.275814553989445, "output_peak": 0.28215253679100566, "reference_rms_dbfs": -27.14678667416795, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.106352532954046}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 81520, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.00013747273046535536, "log_rms_mae_db": 0.8887606075036358, "rms_envelope_nrmse": 0.06806118527283841, "envelope_correlation": 0.9949147612150383, "rms_gain_error_db": -0.12902762194960415, "reference_mean_power": 0.0019308462257719779, "output_mean_power": 0.0018743251185990195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 6.283614364116833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041565087617635905, "output_power": 0.0004008596336767026, "reference_active": true, "delta_db": -0.15736375055271926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.15171169386713e-05, "output_power": 7.102561868410338e-05, "reference_active": true, "delta_db": -0.5983378590686459}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.519660659304176e-05, "output_power": 1.811292382472797e-05, "reference_active": true, "delta_db": -1.4335349451542565}], "active_band_count": 3, "active_band_mae_db": 0.7297455182585405}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012452330782545057, "log_rms_mae_db": 0.4473370048830174, "rms_envelope_nrmse": 0.0389345674689342, "envelope_correlation": 0.9982780848397411, "rms_gain_error_db": -0.12902751602837192, "reference_mean_power": 0.0019422928952156904, "output_mean_power": 0.0018854367589591465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.257129910319434e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003707029677887353, "output_power": 0.0003580909361180516, "reference_active": true, "delta_db": -0.15032734506147305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2635280867150454e-05, "output_power": 3.710294854423918e-05, "reference_active": true, "delta_db": -0.6036070426540214}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.869258984918021e-06, "output_power": 1.0778567062384486e-06, "reference_active": true, "delta_db": -2.391084486624255}], "active_band_count": 3, "active_band_mae_db": 1.0483396247799164}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.0001332544033329483, "log_rms_mae_db": 1.9050060278528511, "rms_envelope_nrmse": 0.09059724201823346, "envelope_correlation": 0.9928588841673124, "rms_gain_error_db": -0.15443816943705044, "reference_mean_power": 0.00116782876738011, "output_mean_power": 0.001127029677295323}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 5.939192180731074e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134913428254024, "output_power": 0.00039606141593418874, "reference_active": true, "delta_db": -0.18703885501725548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.995066727507785e-05, "output_power": 5.808433980135813e-05, "reference_active": true, "delta_db": -0.8073280423975977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9671389178533067e-05, "output_power": 1.3302985098172676e-05, "reference_active": true, "delta_db": -1.6988592595692829}], "active_band_count": 3, "active_band_mae_db": 0.8977420523280454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012167319573637772, "log_rms_mae_db": 1.1512003395413424, "rms_envelope_nrmse": 0.053214571815750136, "envelope_correlation": 0.9974351147708239, "rms_gain_error_db": -0.1544381022980688, "reference_mean_power": 0.0011747523056415098, "output_mean_power": 0.0011337113533922704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.189128511363634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037925439414056246, "output_power": 0.000363060611306565, "reference_active": true, "delta_db": -0.18951486773405868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.625302583005344e-05, "output_power": 2.9570271163865633e-05, "reference_active": true, "delta_db": -0.8848895337379414}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7116807182635842e-06, "output_power": 9.210761095268798e-07, "reference_active": true, "delta_db": -2.691272405477868}], "active_band_count": 3, "active_band_mae_db": 1.2552256023166226}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 4.762512740172365e-05, "log_rms_mae_db": 1.8951318751969617, "rms_envelope_nrmse": 0.13505973363478738, "envelope_correlation": 0.9862657836767347, "rms_gain_error_db": -0.27976508771079917, "reference_mean_power": 0.00010302415332771722, "output_mean_power": 9.659675678650943e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 3.1359809483594164e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4569632781045495e-05, "output_power": 1.3747720265472745e-05, "reference_active": true, "delta_db": -0.2521791911304096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.842890886644979e-06, "output_power": 2.926673071509616e-06, "reference_active": true, "delta_db": -1.1828384185866136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.113490153313903e-06, "output_power": 7.129370435452025e-07, "reference_active": true, "delta_db": -1.93635200048486}], "active_band_count": 3, "active_band_mae_db": 1.1237898700672944}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 4.327669210129085e-05, "log_rms_mae_db": 1.0550185070775027, "rms_envelope_nrmse": 0.07935043434424627, "envelope_correlation": 0.9954437149278065, "rms_gain_error_db": -0.27976478435727625, "reference_mean_power": 0.00010363449577271808, "output_mean_power": 9.716902841467151e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 2.7762525532534898e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3720292251105881e-05, "output_power": 1.2813281683083477e-05, "reference_active": true, "delta_db": -0.2970298858051399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7263230443470765e-06, "output_power": 1.342525072883437e-06, "reference_active": true, "delta_db": -1.0919966259151388}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.698769643844551e-08, "output_power": 5.920406313073682e-08, "reference_active": true, "delta_db": -2.143651314427256}], "active_band_count": 3, "active_band_mae_db": 1.1775592753825117}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 1.6513968336769973e-05, "log_rms_mae_db": 2.253603181244452, "rms_envelope_nrmse": 0.30955468181155515, "envelope_correlation": 0.9266295795617806, "rms_gain_error_db": 0.4209551137012727, "reference_mean_power": 8.517905641506151e-06, "output_mean_power": 9.38487162180243e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 4.8245079516023535e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4329635827636286e-06, "output_power": 2.688084590747914e-06, "reference_active": true, "delta_db": 0.43307322969318895}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1930715407234243e-06, "output_power": 9.827136308408236e-07, "reference_active": true, "delta_db": -0.8423950623569456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.14650683405709e-07, "output_power": 3.3461351200565726e-07, "reference_active": true, "delta_db": 0.26714794846002976}], "active_band_count": 3, "active_band_mae_db": 0.5142054135033881}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 1.2321016663379126e-05, "log_rms_mae_db": 1.7289345811116488, "rms_envelope_nrmse": 0.23535611910790602, "envelope_correlation": 0.953526739871094, "rms_gain_error_db": 0.4209637876348318, "reference_mean_power": 8.568334809688713e-06, "output_mean_power": 9.440452405805265e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 4.130413403311084e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.518595705395907e-06, "output_power": 2.7196754879093255e-06, "reference_active": true, "delta_db": 0.33358628558629866}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.297939167002702e-07, "output_power": 5.085677203142601e-07, "reference_active": true, "delta_db": -0.9284967010591826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9684693476821222e-08, "output_power": 2.4839726408158835e-08, "reference_active": true, "delta_db": -0.77385760687122}], "active_band_count": 3, "active_band_mae_db": 0.6786468645055671}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9028609194053486, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "93cf7a64546600cad11ed7ae9b63f76a0a889be7bbd9fb1d672deb2adc59483e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9844708174551051, "standard_deviation": 0.0016542810410006448, "minimum": 0.9826162011316213, "maximum": 0.9865271662300505, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.89247916328483, "standard_deviation": 0.023855502366446842, "minimum": 1.8588841505130553, "maximum": 1.9255000908523805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8163747662496047, "standard_deviation": 0.2312190640016573, "minimum": 0.5773759781017175, "maximum": 1.2110733916721002, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1416041968374412, "standard_deviation": 0.007190917213536741, "minimum": 0.13202999863831003, "maximum": 0.15005315074138723, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946573136882044, "standard_deviation": 0.0007968624168986494, "minimum": 0.9936678090070488, "maximum": 0.995511487026413, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0615534284063095, "standard_deviation": 0.035675699597474554, "minimum": 1.011074032045198, "maximum": 1.1141538658764005, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0597560219053674, "standard_deviation": 0.22489213487265614, "minimum": 0.8464093690632429, "maximum": 1.5482700017352837, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08161768798137997, "standard_deviation": 0.0055048581100227995, "minimum": 0.07421954238477872, "maximum": 0.08798222994992527, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9201435364075887, "standard_deviation": 0.007775492643747723, "minimum": 0.9021347877340161, "maximum": 0.9266295795617806, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.323522946204613, "standard_deviation": 0.04144252371588814, "minimum": 2.253603181244452, "maximum": 2.381035209121421, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5423601902221248, "standard_deviation": 0.22741645233486416, "minimum": 0.24066573015920645, "maximum": 0.834211686838489, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3227484971671764, "standard_deviation": 0.012951938856184644, "minimum": 0.30955468181155515, "maximum": 0.352441207821484, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9511029941180189, "standard_deviation": 0.006019471671979763, "minimum": 0.9373769829098897, "maximum": 0.9564852946417957, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.751943847941333, "standard_deviation": 0.043563496364372664, "minimum": 1.6965095310574299, "maximum": 1.817119590591193, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9446196579918835, "standard_deviation": 0.31512858334929295, "minimum": 0.6580174455074772, "maximum": 1.3276431709250434, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.24172548983733508, "standard_deviation": 0.011271552746179794, "minimum": 0.23409216457038198, "maximum": 0.26847548437043744, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925059540393477, "standard_deviation": 0.0010394801640187237, "minimum": 0.990616498936981, "maximum": 0.9939776851115718, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.832104596587405, "standard_deviation": 0.041621807063790865, "minimum": 1.7840243931647337, "maximum": 1.9050060278528511, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7590756682178637, "standard_deviation": 0.32291866425589916, "minimum": 0.20621025755008424, "maximum": 1.23899171156678, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09331426810412358, "standard_deviation": 0.006282676021576462, "minimum": 0.08336720205223556, "maximum": 0.10451906087229573, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973322151638355, "standard_deviation": 0.0005717983490642691, "minimum": 0.9960453272171405, "maximum": 0.9979468517202872, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1245840582515696, "standard_deviation": 0.03838796965523747, "minimum": 1.0658554777887055, "maximum": 1.193287818692408, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0333693476774684, "standard_deviation": 0.3720882795809498, "minimum": 0.3258143306999382, "maximum": 1.506430786272323, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05556258626538473, "standard_deviation": 0.00565495818900256, "minimum": 0.04810074146589883, "maximum": 0.06709437312555253, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2176144782971259, "standard_deviation": 0.02507486877751508, "minimum": 0.17670159252852577, "maximum": 0.25501095332045515, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.103044802396807, "standard_deviation": 0.010673420914279301, "minimum": 6.089760597156709, "maximum": 6.119110689509726, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.29986034344823764, "standard_deviation": 0.010127298281500028, "minimum": 0.28215253679100566, "maximum": 0.31508995859873085, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.13405159954727142, "standard_deviation": 0.02832024575911909, "minimum": -0.1676592860487883, "maximum": -0.0883302437495221, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941895514499204, "standard_deviation": 0.0006728315337395697, "minimum": 0.9935650373765474, "maximum": 0.9953354053119756, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9015925344468194, "standard_deviation": 0.009120726144028472, "minimum": 0.8887606075036358, "maximum": 0.9193652753942231, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6263659044895252, "standard_deviation": 0.29288496774346107, "minimum": 0.18005670065840143, "maximum": 1.0502227521616974, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07251781866842441, "standard_deviation": 0.00438969958331759, "minimum": 0.06457939314738324, "maximum": 0.07699264894618914, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981625928766401, "standard_deviation": 0.00035165993379807596, "minimum": 0.9975596573075382, "maximum": 0.9987014786061283, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.45657444599541264, "standard_deviation": 0.021498996722176048, "minimum": 0.42347895421460585, "maximum": 0.4877935687921632, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8312163913950046, "standard_deviation": 0.36791069526681924, "minimum": 0.1978997083688888, "maximum": 1.36048934064942, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04016521533567199, "standard_deviation": 0.004168102678672274, "minimum": 0.033046268203509394, "maximum": 0.04686304206177511, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284449-0020", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.694197654724121, "predicted_mos": 2.694197654724121, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "bd6cc1e166e420bc9c91a302c443d18631a624e0cea8acd68b37d3a51313068e", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.048671300057322, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8724790574933957, "全帯域_10ms_rms_envelope_nrmse": 0.06735757132212408, "全帯域_10ms_envelope_correlation": 0.9949683481311322, "全帯域_10ms_modulation_mae_db": 0.25737273994373394, "全帯域_40ms_log_rms_mae_db": 0.442473822834613, "全帯域_40ms_rms_envelope_nrmse": 0.035661693820822446, "全帯域_40ms_envelope_correlation": 0.9985344083188017, "全帯域_40ms_modulation_mae_db": 0.34752839336805574, "300–1000 Hz_10ms_log_rms_mae_db": 1.418997866705479, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08374902750573691, "300–1000 Hz_10ms_envelope_correlation": 0.9940337796882041, "300–1000 Hz_10ms_modulation_mae_db": 0.3456160280078886, "300–1000 Hz_40ms_log_rms_mae_db": 0.8596410057341677, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04901824592213905, "300–1000 Hz_40ms_envelope_correlation": 0.9980202726178232, "300–1000 Hz_40ms_modulation_mae_db": 0.46803444884257966, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5618962131535374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1301396047349794, "1000–3000 Hz_10ms_envelope_correlation": 0.9873932681219543, "1000–3000 Hz_10ms_modulation_mae_db": 0.9531611671958039, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8873215597204035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07419627854699098, "1000–3000 Hz_40ms_envelope_correlation": 0.9963388521556968, "1000–3000 Hz_40ms_modulation_mae_db": 1.281313397140943, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6333279428834466, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2562439784112485, "3000–6000 Hz_10ms_envelope_correlation": 0.9459309130209246, "3000–6000 Hz_10ms_modulation_mae_db": 0.8507620534396466, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2629932356680595, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20957544634953496, "3000–6000 Hz_40ms_envelope_correlation": 0.959415323178677, "3000–6000 Hz_40ms_modulation_mae_db": 1.0768597307766958, "mr_stft_db": 5.578317904045133, "band_mae_db": 0.25597628943790385, "rms_gain_error_db": -0.10833625746769798, "output_peak": 0.3074214734918652}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9057969077424453, "全帯域_10ms_rms_envelope_nrmse": 0.06971346509184423, "全帯域_10ms_envelope_correlation": 0.9946257202552703, "全帯域_10ms_modulation_mae_db": 0.5662088221196483, "全帯域_40ms_log_rms_mae_db": 0.47029509898843225, "全帯域_40ms_rms_envelope_nrmse": 0.04293320595509742, "全帯域_40ms_envelope_correlation": 0.9978170587604287, "全帯域_40ms_modulation_mae_db": 0.6558484952333218, "300–1000 Hz_10ms_log_rms_mae_db": 1.4521348465147765, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08621096892230032, "300–1000 Hz_10ms_envelope_correlation": 0.9935873786984027, "300–1000 Hz_10ms_modulation_mae_db": 0.5416157963881136, "300–1000 Hz_40ms_log_rms_mae_db": 0.8810237306821034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05620072287687348, "300–1000 Hz_40ms_envelope_correlation": 0.9971557991222582, "300–1000 Hz_40ms_modulation_mae_db": 0.6232677646632646, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4719152966897002, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13030222967137814, "1000–3000 Hz_10ms_envelope_correlation": 0.9868975011965191, "1000–3000 Hz_10ms_modulation_mae_db": 0.47963116050098353, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8420773513414522, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07974846238612966, "1000–3000 Hz_40ms_envelope_correlation": 0.9949455671600084, "1000–3000 Hz_40ms_modulation_mae_db": 0.671159019666069, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6435790130342234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2837158059953604, "3000–6000 Hz_10ms_envelope_correlation": 0.9383568699041899, "3000–6000 Hz_10ms_modulation_mae_db": 0.27287612357625585, "3000–6000 Hz_40ms_log_rms_mae_db": 1.285462753333583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22367972096915617, "3000–6000 Hz_40ms_envelope_correlation": 0.9572680049381503, "3000–6000 Hz_40ms_modulation_mae_db": 0.551693806617236, "mr_stft_db": 5.576187219264873, "band_mae_db": 0.20736762433358594, "rms_gain_error_db": -0.12306455338662599, "output_peak": 0.2853501475841653}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.915194430388664, "全帯域_10ms_rms_envelope_nrmse": 0.07145352922781678, "全帯域_10ms_envelope_correlation": 0.9946086272160328, "全帯域_10ms_modulation_mae_db": 0.5508679256656129, "全帯域_40ms_log_rms_mae_db": 0.4632863014936725, "全帯域_40ms_rms_envelope_nrmse": 0.04033538490138565, "全帯域_40ms_envelope_correlation": 0.9984244782098077, "全帯域_40ms_modulation_mae_db": 0.7584441367436314, "300–1000 Hz_10ms_log_rms_mae_db": 1.4787763137268628, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08885440592179768, "300–1000 Hz_10ms_envelope_correlation": 0.993719070154525, "300–1000 Hz_10ms_modulation_mae_db": 0.4869403309217437, "300–1000 Hz_40ms_log_rms_mae_db": 0.961854333548262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055256786448942684, "300–1000 Hz_40ms_envelope_correlation": 0.9979450433582752, "300–1000 Hz_40ms_modulation_mae_db": 0.7102946662718258, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5236692621156658, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13181523671911985, "1000–3000 Hz_10ms_envelope_correlation": 0.9870081760170865, "1000–3000 Hz_10ms_modulation_mae_db": 1.2796892074115438, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8852049108902237, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07710592929045317, "1000–3000 Hz_40ms_envelope_correlation": 0.9959009095612422, "1000–3000 Hz_40ms_modulation_mae_db": 1.5709786586656733, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6263590162467603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2743605557672289, "3000–6000 Hz_10ms_envelope_correlation": 0.9387928095411224, "3000–6000 Hz_10ms_modulation_mae_db": 0.8377313259655209, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2482563166936593, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21880557839241113, "3000–6000 Hz_40ms_envelope_correlation": 0.9562612879504734, "3000–6000 Hz_40ms_modulation_mae_db": 1.2520081323248562, "mr_stft_db": 5.586453794105168, "band_mae_db": 0.33199721058422793, "rms_gain_error_db": -0.1833313662513788, "output_peak": 0.26360406421475224}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9092838992303492, "全帯域_10ms_rms_envelope_nrmse": 0.07434006318172103, "全帯域_10ms_envelope_correlation": 0.9937713217362243, "全帯域_10ms_modulation_mae_db": 0.33751766775273434, "全帯域_40ms_log_rms_mae_db": 0.44148317459728653, "全帯域_40ms_rms_envelope_nrmse": 0.03997733371064208, "全帯域_40ms_envelope_correlation": 0.9980319271685957, "全帯域_40ms_modulation_mae_db": 0.44147727878392057, "300–1000 Hz_10ms_log_rms_mae_db": 1.4802799364999104, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09879566396923198, "300–1000 Hz_10ms_envelope_correlation": 0.9913346539656837, "300–1000 Hz_10ms_modulation_mae_db": 0.4404449166276548, "300–1000 Hz_40ms_log_rms_mae_db": 0.9212150653294422, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05510746003924369, "300–1000 Hz_40ms_envelope_correlation": 0.9971256808418206, "300–1000 Hz_40ms_modulation_mae_db": 0.5998827502092637, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4575488676034576, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1302155144438921, "1000–3000 Hz_10ms_envelope_correlation": 0.9870003096183788, "1000–3000 Hz_10ms_modulation_mae_db": 0.6178368471735497, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7946446285493578, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07743137233346957, "1000–3000 Hz_40ms_envelope_correlation": 0.995407281670445, "1000–3000 Hz_40ms_modulation_mae_db": 0.7014422976958908, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6023706618213571, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25081888666610425, "3000–6000 Hz_10ms_envelope_correlation": 0.9484704240177463, "3000–6000 Hz_10ms_modulation_mae_db": 1.1926771608574775, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2092512175478771, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2103920120716265, "3000–6000 Hz_40ms_envelope_correlation": 0.9593015707644296, "3000–6000 Hz_40ms_modulation_mae_db": 1.5976171122141043, "mr_stft_db": 5.582177030727949, "band_mae_db": 0.25945408279447363, "rms_gain_error_db": -0.08929125035756189, "output_peak": 0.29895401809154226}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8600277491476535, "全帯域_10ms_rms_envelope_nrmse": 0.0671519914138989, "全帯域_10ms_envelope_correlation": 0.9951644628006081, "全帯域_10ms_modulation_mae_db": 0.4822559230854466, "全帯域_40ms_log_rms_mae_db": 0.43081576508884345, "全帯域_40ms_rms_envelope_nrmse": 0.039036257970339286, "全帯域_40ms_envelope_correlation": 0.9984002419948761, "全帯域_40ms_modulation_mae_db": 0.7708628999815913, "300–1000 Hz_10ms_log_rms_mae_db": 1.417717103638854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0907768665578557, "300–1000 Hz_10ms_envelope_correlation": 0.9930662307539458, "300–1000 Hz_10ms_modulation_mae_db": 0.5426992155379713, "300–1000 Hz_40ms_log_rms_mae_db": 0.8823218160114911, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05682923836542428, "300–1000 Hz_40ms_envelope_correlation": 0.9973498739360506, "300–1000 Hz_40ms_modulation_mae_db": 0.9022120509485395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.497086614868843, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12980255129089158, "1000–3000 Hz_10ms_envelope_correlation": 0.9874014792829922, "1000–3000 Hz_10ms_modulation_mae_db": 0.9984933941901731, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8766720879542982, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07883846484209453, "1000–3000 Hz_40ms_envelope_correlation": 0.9956256466377611, "1000–3000 Hz_40ms_modulation_mae_db": 1.180143236849549, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6062486694760731, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2658593673957911, "3000–6000 Hz_10ms_envelope_correlation": 0.9419370941358549, "3000–6000 Hz_10ms_modulation_mae_db": 0.8759776664321729, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2290327431535786, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2104836584986516, "3000–6000 Hz_40ms_envelope_correlation": 0.9591252211612687, "3000–6000 Hz_40ms_modulation_mae_db": 1.024875065674255, "mr_stft_db": 5.59632683833794, "band_mae_db": 0.3233599838571147, "rms_gain_error_db": -0.15644021302447692, "output_peak": 0.2834568021583782}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.884937864103056, "全帯域_10ms_rms_envelope_nrmse": 0.06914430212168357, "全帯域_10ms_envelope_correlation": 0.9947091612215953, "全帯域_10ms_modulation_mae_db": 0.6786711840937326, "全帯域_40ms_log_rms_mae_db": 0.44883555849163487, "全帯域_40ms_rms_envelope_nrmse": 0.03703993049311523, "全帯域_40ms_envelope_correlation": 0.9984402327864956, "全帯域_40ms_modulation_mae_db": 0.8783627276072856, "300–1000 Hz_10ms_log_rms_mae_db": 1.4569222859117839, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08896197212175652, "300–1000 Hz_10ms_envelope_correlation": 0.9931823009673504, "300–1000 Hz_10ms_modulation_mae_db": 0.8762345615305568, "300–1000 Hz_40ms_log_rms_mae_db": 0.8797999363367013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05067295990823775, "300–1000 Hz_40ms_envelope_correlation": 0.9978155860456943, "300–1000 Hz_40ms_modulation_mae_db": 1.146660066995129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4673570914259053, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1255407130061696, "1000–3000 Hz_10ms_envelope_correlation": 0.9879997050761609, "1000–3000 Hz_10ms_modulation_mae_db": 0.6309427855689638, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8379353728353481, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07122419366101151, "1000–3000 Hz_40ms_envelope_correlation": 0.9962714510904109, "1000–3000 Hz_40ms_modulation_mae_db": 0.7691276229750265, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6177728279717336, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2602707946264825, "3000–6000 Hz_10ms_envelope_correlation": 0.946365804381925, "3000–6000 Hz_10ms_modulation_mae_db": 0.6004282555694236, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2759482550025436, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2148683253007044, "3000–6000 Hz_40ms_envelope_correlation": 0.9590078237632073, "3000–6000 Hz_40ms_modulation_mae_db": 0.7055706309675518, "mr_stft_db": 5.587607416343752, "band_mae_db": 0.2830470115141062, "rms_gain_error_db": -0.12129950546867274, "output_peak": 0.2785068970157136}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8736311720368283, "全帯域_10ms_rms_envelope_nrmse": 0.07589967097062753, "全帯域_10ms_envelope_correlation": 0.9936324605900803, "全帯域_10ms_modulation_mae_db": 0.5426841968832181, "全帯域_40ms_log_rms_mae_db": 0.455294056383479, "全帯域_40ms_rms_envelope_nrmse": 0.0440058977219385, "全帯域_40ms_envelope_correlation": 0.9977537175907167, "全帯域_40ms_modulation_mae_db": 0.6821137562176599, "300–1000 Hz_10ms_log_rms_mae_db": 1.4297503350424716, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1024362909529327, "300–1000 Hz_10ms_envelope_correlation": 0.9909747167866061, "300–1000 Hz_10ms_modulation_mae_db": 0.610858310385248, "300–1000 Hz_40ms_log_rms_mae_db": 0.8932824329947661, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06542496377084717, "300–1000 Hz_40ms_envelope_correlation": 0.9962620760827957, "300–1000 Hz_40ms_modulation_mae_db": 0.7785826202187612, "1000–3000 Hz_10ms_log_rms_mae_db": 1.524388708122026, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12121688713655052, "1000–3000 Hz_10ms_envelope_correlation": 0.9888840295294206, "1000–3000 Hz_10ms_modulation_mae_db": 0.7407465365580842, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8706711128940731, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06261010715518459, "1000–3000 Hz_40ms_envelope_correlation": 0.9972773521634583, "1000–3000 Hz_40ms_modulation_mae_db": 1.0093783969396577, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5780344167145817, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24811551906242485, "3000–6000 Hz_10ms_envelope_correlation": 0.9521267315492262, "3000–6000 Hz_10ms_modulation_mae_db": 0.6281259878842669, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2786492360150439, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20901532983548368, "3000–6000 Hz_40ms_envelope_correlation": 0.9620751286586989, "3000–6000 Hz_40ms_modulation_mae_db": 0.8427900336656693, "mr_stft_db": 5.579204016571161, "band_mae_db": 0.18917925615306744, "rms_gain_error_db": -0.13285574178597515, "output_peak": 0.3160885431063026}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8993272022677138, "全帯域_10ms_rms_envelope_nrmse": 0.07231120127258614, "全帯域_10ms_envelope_correlation": 0.9941847629373168, "全帯域_10ms_modulation_mae_db": 0.4387350322120694, "全帯域_40ms_log_rms_mae_db": 0.4811817761322064, "全帯域_40ms_rms_envelope_nrmse": 0.038829424116277635, "全帯域_40ms_envelope_correlation": 0.998233634311509, "全帯域_40ms_modulation_mae_db": 0.6570957746939263, "300–1000 Hz_10ms_log_rms_mae_db": 1.5219581049763673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09603183336292861, "300–1000 Hz_10ms_envelope_correlation": 0.9919333449635293, "300–1000 Hz_10ms_modulation_mae_db": 0.4803724748693882, "300–1000 Hz_40ms_log_rms_mae_db": 0.969093851132234, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0534119839892953, "300–1000 Hz_40ms_envelope_correlation": 0.9974478788103421, "300–1000 Hz_40ms_modulation_mae_db": 0.6525243133322021, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4791575551514895, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12662813153396105, "1000–3000 Hz_10ms_envelope_correlation": 0.9881460319850763, "1000–3000 Hz_10ms_modulation_mae_db": 1.1261503612624295, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8314679984770468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0748819849651642, "1000–3000 Hz_40ms_envelope_correlation": 0.9962669359742954, "1000–3000 Hz_40ms_modulation_mae_db": 1.4025475475617502, "3000–6000 Hz_10ms_log_rms_mae_db": 1.563668812121274, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24644973689221936, "3000–6000 Hz_10ms_envelope_correlation": 0.94989326489231, "3000–6000 Hz_10ms_modulation_mae_db": 0.8723325630071505, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2063429371426002, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20753764945619854, "3000–6000 Hz_40ms_envelope_correlation": 0.9601116933245123, "3000–6000 Hz_40ms_modulation_mae_db": 1.1533801812730744, "mr_stft_db": 5.589736241384855, "band_mae_db": 0.30139244399802234, "rms_gain_error_db": -0.11280041176419481, "output_peak": 0.31306397422461996}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 81520, "output_samples": 81520, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.31554030639344, "spectral_convergence": 0.1837324169204444, "spectral_convergence_status": "ok", "amplitude_floor": 1.0222158521079025e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.472572350559447, "spectral_convergence": 0.1851625390972834, "spectral_convergence_status": "ok", "amplitude_floor": 7.564059658380168e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.946841055182515, "spectral_convergence": 0.19159428685132712, "spectral_convergence_status": "ok", "amplitude_floor": 6.35834815641291e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.932339290337003e-05, "output_power": 9.58029996231548e-05, "delta_db": 0.3041389602688967, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011480494900330098, "output_power": 0.001127464064305781, "delta_db": -0.078579015098797, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005518546578574125, "output_power": 0.0005247017957545968, "delta_db": -0.21912161965452076, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.085077476738271e-05, "output_power": 8.539847656794662e-05, "delta_db": -0.26878511949869605, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.0710168905088768e-05, "output_power": 1.8592537459621605e-05, "delta_db": -0.4684497570350021, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8163123695227053e-05, "output_power": 2.9468575936026128e-05, "delta_db": 0.19678326507151037, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.25597628943790385, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.10833625746769798, "pitch": null, "output_rms_dbfs": -27.255122931635643, "output_peak": 0.3074214734918652, "reference_rms_dbfs": -27.14678667416795, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.578317904045133}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 81520, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.00013747273046535536, "log_rms_mae_db": 0.8724790574933957, "rms_envelope_nrmse": 0.06735757132212408, "envelope_correlation": 0.9949683481311322, "rms_gain_error_db": -0.10833624122977849, "reference_mean_power": 0.0019308462257719779, "output_mean_power": 0.0018832763970197022}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 6.283614364116833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041565087617635905, "output_power": 0.00040478147921192573, "reference_active": true, "delta_db": -0.11508067185273009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.15171169386713e-05, "output_power": 7.852480683241869e-05, "reference_active": true, "delta_db": -0.16241934608405512}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.519660659304176e-05, "output_power": 2.2484344715264932e-05, "reference_active": true, "delta_db": -0.4946182018944167}], "active_band_count": 3, "active_band_mae_db": 0.25737273994373394}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012452330782545057, "log_rms_mae_db": 0.442473822834613, "rms_envelope_nrmse": 0.035661693820822446, "envelope_correlation": 0.9985344083188017, "rms_gain_error_db": -0.10833617724376737, "reference_mean_power": 0.0019422928952156904, "output_mean_power": 0.0018944410853272127}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.257129910319434e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003707029677887353, "output_power": 0.00036246437410880313, "reference_active": true, "delta_db": -0.0976073585849116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2635280867150454e-05, "output_power": 4.1595311098871684e-05, "reference_active": true, "delta_db": -0.10724751504523392}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.869258984918021e-06, "output_power": 1.5413328914601686e-06, "reference_active": true, "delta_db": -0.8377303064740217}], "active_band_count": 3, "active_band_mae_db": 0.34752839336805574}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.0001332544033329483, "log_rms_mae_db": 1.418997866705479, "rms_envelope_nrmse": 0.08374902750573691, "envelope_correlation": 0.9940337796882041, "rms_gain_error_db": -0.18858951900070137, "reference_mean_power": 0.00116782876738011, "output_mean_power": 0.0011182018777537658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 5.939192180731074e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134913428254024, "output_power": 0.0004012764520546802, "reference_active": true, "delta_db": -0.13022746306094535}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.995066727507785e-05, "output_power": 6.451410160799671e-05, "reference_active": true, "delta_db": -0.3513720780177126}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9671389178533067e-05, "output_power": 1.7310522478320213e-05, "reference_active": true, "delta_db": -0.5552485429450078}], "active_band_count": 3, "active_band_mae_db": 0.3456160280078886}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012167319573637772, "log_rms_mae_db": 0.8596410057341677, "rms_envelope_nrmse": 0.04901824592213905, "envelope_correlation": 0.9980202726178232, "rms_gain_error_db": -0.1885893412451859, "reference_mean_power": 0.0011747523056415098, "output_mean_power": 0.001124831246259275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.189128511363634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037925439414056246, "output_power": 0.00036891603504311056, "reference_active": true, "delta_db": -0.12003088963890296}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.625302583005344e-05, "output_power": 3.3686840447222487e-05, "reference_active": true, "delta_db": -0.3188398087783384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7116807182635842e-06, "output_power": 1.3705645473332955e-06, "reference_active": true, "delta_db": -0.9652326481104978}], "active_band_count": 3, "active_band_mae_db": 0.46803444884257966}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 4.762512740172365e-05, "log_rms_mae_db": 1.5618962131535374, "rms_envelope_nrmse": 0.1301396047349794, "envelope_correlation": 0.9873932681219543, "rms_gain_error_db": -0.32142686560410894, "reference_mean_power": 0.00010302415332771722, "output_mean_power": 9.567453663102407e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 3.1359809483594164e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4569632781045495e-05, "output_power": 1.3434258032423804e-05, "reference_active": true, "delta_db": -0.3523492021579625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.842890886644979e-06, "output_power": 3.1385028268094713e-06, "reference_active": true, "delta_db": -0.8793552932233907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.113490153313903e-06, "output_power": 7.654352882209568e-07, "reference_active": true, "delta_db": -1.6277790062060584}], "active_band_count": 3, "active_band_mae_db": 0.9531611671958039}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 4.327669210129085e-05, "log_rms_mae_db": 0.8873215597204035, "rms_envelope_nrmse": 0.07419627854699098, "envelope_correlation": 0.9963388521556968, "rms_gain_error_db": -0.321426863060161, "reference_mean_power": 0.00010363449577271808, "output_mean_power": 9.624133805121072e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 2.7762525532534898e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3720292251105881e-05, "output_power": 1.2643683012969847e-05, "reference_active": true, "delta_db": -0.35489763021419307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7263230443470765e-06, "output_power": 1.3568496590196833e-06, "reference_active": true, "delta_db": -1.045903380409354}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.698769643844551e-08, "output_power": 5.52589624667992e-08, "reference_active": true, "delta_db": -2.4431391807992817}], "active_band_count": 3, "active_band_mae_db": 1.281313397140943}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 1.6513968336769973e-05, "log_rms_mae_db": 1.6333279428834466, "rms_envelope_nrmse": 0.2562439784112485, "envelope_correlation": 0.9459309130209246, "rms_gain_error_db": -0.030863579065262877, "reference_mean_power": 8.517905641506151e-06, "output_mean_power": 8.457586863358586e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 4.8245079516023535e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4329635827636286e-06, "output_power": 2.2177572145179757e-06, "reference_active": true, "delta_db": -0.4022160763569943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1930715407234243e-06, "output_power": 1.0146297494941116e-06, "reference_active": true, "delta_db": -0.7035889434270765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.14650683405709e-07, "output_power": 2.2551766452036142e-07, "reference_active": true, "delta_db": -1.446481140534869}], "active_band_count": 3, "active_band_mae_db": 0.8507620534396466}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 1.2321016663379126e-05, "log_rms_mae_db": 1.2629932356680595, "rms_envelope_nrmse": 0.20957544634953496, "envelope_correlation": 0.959415323178677, "rms_gain_error_db": -0.03087023292062442, "reference_mean_power": 8.568334809688713e-06, "output_mean_power": 8.507645887316971e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 4.130413403311084e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.518595705395907e-06, "output_power": 2.240081684659011e-06, "reference_active": true, "delta_db": -0.5089460329685181}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.297939167002702e-07, "output_power": 5.366915172245273e-07, "reference_active": true, "delta_db": -0.6947373036664394}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9684693476821222e-08, "output_power": 1.8614140309277152e-08, "reference_active": true, "delta_db": -2.02689585569513}], "active_band_count": 3, "active_band_mae_db": 1.0768597307766958}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.925434039424382, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "918e3ed6f155d27eb630c7c4a58be9ee5c66abea595446d6e295e20594e72025", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9875913126034486, "standard_deviation": 0.000696068596385931, "minimum": 0.9868975011965191, "maximum": 0.9888840295294206, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4978774511413282, "standard_deviation": 0.03595522324182975, "minimum": 1.4575488676034576, "maximum": 1.5618962131535374, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8533314324826915, "standard_deviation": 0.27889853589178293, "minimum": 0.47963116050098353, "maximum": 1.2796892074115438, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12820760856711777, "standard_deviation": 0.003508923760506585, "minimum": 0.12121688713655052, "maximum": 0.13181523671911985, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960042495516648, "standard_deviation": 0.0007075458954696282, "minimum": 0.9949455671600084, "maximum": 0.9972773521634583, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8532493778327754, "standard_deviation": 0.032298801224429956, "minimum": 0.7946446285493578, "maximum": 0.8873215597204035, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0732612721868198, "standard_deviation": 0.3396164556479644, "minimum": 0.671159019666069, "maximum": 1.5709786586656733, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07450459914756227, "standard_deviation": 0.005530317672070329, "minimum": 0.06261010715518459, "maximum": 0.07974846238612966, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9452342389304125, "standard_deviation": 0.005086579647735509, "minimum": 0.9383568699041899, "maximum": 0.9521267315492262, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6089201700336813, "standard_deviation": 0.027336803573966, "minimum": 1.563668812121274, "maximum": 1.6435790130342234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7663638920914894, "standard_deviation": 0.2692259893592398, "minimum": 0.27287612357625585, "maximum": 1.1926771608574775, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26072933060210746, "standard_deviation": 0.013212906053309959, "minimum": 0.24644973689221936, "maximum": 0.2837158059953604, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9590707567174271, "standard_deviation": 0.0017497655603997428, "minimum": 0.9562612879504734, "maximum": 0.9620751286586989, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2494920868196182, "standard_deviation": 0.031458649962992385, "minimum": 1.2063429371426002, "maximum": 1.285462753333583, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0255993366891802, "standard_deviation": 0.3290343968691542, "minimum": 0.551693806617236, "maximum": 1.5976171122141043, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.21304471510922088, "standard_deviation": 0.005629156717167125, "minimum": 0.20753764945619854, "maximum": 0.22367972096915617, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992728934497281, "standard_deviation": 0.0011584155831415393, "minimum": 0.9909747167866061, "maximum": 0.9940337796882041, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4570670991270633, "standard_deviation": 0.035835359127175455, "minimum": 1.417717103638854, "maximum": 1.5219581049763673, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5405977042835707, "standard_deviation": 0.15682052184679227, "minimum": 0.3456160280078886, "maximum": 0.8762345615305568, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09197712866431755, "standard_deviation": 0.006475895657305928, "minimum": 0.08374902750573691, "maximum": 0.1024362909529327, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9973902763518825, "standard_deviation": 0.0005717550253238823, "minimum": 0.9962620760827957, "maximum": 0.9980202726178232, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.906029021471146, "standard_deviation": 0.040556554382556224, "minimum": 0.8596410057341677, "maximum": 0.969093851132234, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7351823351851957, "standard_deviation": 0.21001599295295587, "minimum": 0.46803444884257966, "maximum": 1.146660066995129, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05524029516512542, "standard_deviation": 0.004929643909356297, "minimum": 0.04901824592213905, "maximum": 0.06542496377084717, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2689717378340627, "standard_deviation": 0.05154870291792748, "minimum": 0.18917925615306744, "maximum": 0.33199721058422793, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.584501307597604, "standard_deviation": 0.0067773775336508885, "minimum": 5.576187219264873, "maximum": 5.59632683833794, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.2933057399859174, "standard_deviation": 0.018524459660092685, "minimum": 0.26360406421475224, "maximum": 0.3160885431063026, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12842741243832304, "standard_deviation": 0.029466289206742925, "minimum": -0.1833313662513788, "maximum": -0.08929125035756189, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9944581081110326, "standard_deviation": 0.0005478720209340567, "minimum": 0.9936324605900803, "maximum": 0.9951644628006081, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.8900847853012632, "standard_deviation": 0.02014941536311007, "minimum": 0.8600277491476535, "maximum": 0.915194430388664, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4817891864695245, "standard_deviation": 0.13490746087596964, "minimum": 0.25737273994373394, "maximum": 0.6786711840937326, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07092147432528778, "standard_deviation": 0.0031670547020607146, "minimum": 0.0671519914138989, "maximum": 0.07589967097062753, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982044623926539, "standard_deviation": 0.00030120246165829507, "minimum": 0.9977537175907167, "maximum": 0.9985344083188017, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.454208194251271, "standard_deviation": 0.016671816706423266, "minimum": 0.43081576508884345, "maximum": 0.4811817761322064, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.648966682828674, "standard_deviation": 0.175090022203492, "minimum": 0.34752839336805574, "maximum": 0.8783627276072856, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.03972739108620228, "standard_deviation": 0.002778780414820042, "minimum": 0.035661693820822446, "maximum": 0.0440058977219385, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284449-0020", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.8391854763031006, "predicted_mos": 2.8391854763031006, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "10b932dfedb8e85e5b4591e158e284a11c21339d60195c2d40c4d7a6d0d82be3", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.950995299965143, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8999334529007896, "全帯域_10ms_rms_envelope_nrmse": 0.07575331962161798, "全帯域_10ms_envelope_correlation": 0.9935964238146165, "全帯域_10ms_modulation_mae_db": 0.4451401717723293, "全帯域_40ms_log_rms_mae_db": 0.4558161858800259, "全帯域_40ms_rms_envelope_nrmse": 0.0463935739671112, "全帯域_40ms_envelope_correlation": 0.9973928423232157, "全帯域_40ms_modulation_mae_db": 0.4616766797309578, "300–1000 Hz_10ms_log_rms_mae_db": 1.6057559202771328, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10057288658599063, "300–1000 Hz_10ms_envelope_correlation": 0.9911569563706436, "300–1000 Hz_10ms_modulation_mae_db": 0.5156174648249227, "300–1000 Hz_40ms_log_rms_mae_db": 0.9706178993437595, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06560323136682752, "300–1000 Hz_40ms_envelope_correlation": 0.9960236936701048, "300–1000 Hz_40ms_modulation_mae_db": 0.5387461870309227, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5251863332439304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11334108922775432, "1000–3000 Hz_10ms_envelope_correlation": 0.990122932509228, "1000–3000 Hz_10ms_modulation_mae_db": 0.4734760011917934, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8410833191227145, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.057620600176892245, "1000–3000 Hz_40ms_envelope_correlation": 0.9974552164585353, "1000–3000 Hz_40ms_modulation_mae_db": 0.6797317149533445, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8009474426715513, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2610863941747794, "3000–6000 Hz_10ms_envelope_correlation": 0.9455316163568264, "3000–6000 Hz_10ms_modulation_mae_db": 1.0852507257119617, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3283103308073092, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20575819297319922, "3000–6000 Hz_40ms_envelope_correlation": 0.9622667550315543, "3000–6000 Hz_40ms_modulation_mae_db": 1.4333317316365897, "mr_stft_db": 5.703593118136491, "band_mae_db": 0.23642894774734904, "rms_gain_error_db": -0.11890284314867736, "output_peak": 0.3165971407690623}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8971494298698832, "全帯域_10ms_rms_envelope_nrmse": 0.07156214897986893, "全帯域_10ms_envelope_correlation": 0.9942449760529565, "全帯域_10ms_modulation_mae_db": 0.20196764797256472, "全帯域_40ms_log_rms_mae_db": 0.4626503324690797, "全帯域_40ms_rms_envelope_nrmse": 0.0396763623300277, "全帯域_40ms_envelope_correlation": 0.998083268203786, "全帯域_40ms_modulation_mae_db": 0.3279994204040743, "300–1000 Hz_10ms_log_rms_mae_db": 1.5461885374250344, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09070952924184351, "300–1000 Hz_10ms_envelope_correlation": 0.9928630612184475, "300–1000 Hz_10ms_modulation_mae_db": 0.27862241907401697, "300–1000 Hz_40ms_log_rms_mae_db": 0.9812538164194649, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05153633992771754, "300–1000 Hz_40ms_envelope_correlation": 0.9976733119812146, "300–1000 Hz_40ms_modulation_mae_db": 0.5123694348285911, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6216666712189998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12334040702087998, "1000–3000 Hz_10ms_envelope_correlation": 0.9882986196603801, "1000–3000 Hz_10ms_modulation_mae_db": 0.7719794066751217, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8381803806151256, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06401704504607189, "1000–3000 Hz_40ms_envelope_correlation": 0.9968557091834909, "1000–3000 Hz_40ms_modulation_mae_db": 0.9055942878811095, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8310575327392153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2741285123494563, "3000–6000 Hz_10ms_envelope_correlation": 0.9419742770027529, "3000–6000 Hz_10ms_modulation_mae_db": 1.173484617715071, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3544525802415133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2182240931218953, "3000–6000 Hz_40ms_envelope_correlation": 0.9592963785556958, "3000–6000 Hz_40ms_modulation_mae_db": 1.5106786617108696, "mr_stft_db": 5.713844733201376, "band_mae_db": 0.18800095145422677, "rms_gain_error_db": -0.09766246537433033, "output_peak": 0.3036371085544453}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.91117011976009, "全帯域_10ms_rms_envelope_nrmse": 0.06998518357916443, "全帯域_10ms_envelope_correlation": 0.9945875839518965, "全帯域_10ms_modulation_mae_db": 0.7169823745716822, "全帯域_40ms_log_rms_mae_db": 0.4916573384309007, "全帯域_40ms_rms_envelope_nrmse": 0.039843204184710324, "全帯域_40ms_envelope_correlation": 0.9981884022300825, "全帯域_40ms_modulation_mae_db": 0.8436059459344972, "300–1000 Hz_10ms_log_rms_mae_db": 1.52785557607385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08762384091710208, "300–1000 Hz_10ms_envelope_correlation": 0.9934794976547223, "300–1000 Hz_10ms_modulation_mae_db": 0.9409079287199646, "300–1000 Hz_40ms_log_rms_mae_db": 0.9455055536536412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05210050095327029, "300–1000 Hz_40ms_envelope_correlation": 0.9977426257297145, "300–1000 Hz_40ms_modulation_mae_db": 1.2963504079707222, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6119542837170304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11660084214773524, "1000–3000 Hz_10ms_envelope_correlation": 0.989629880622263, "1000–3000 Hz_10ms_modulation_mae_db": 0.9931251741727333, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9092768798633346, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07000931247868203, "1000–3000 Hz_40ms_envelope_correlation": 0.9962506285568251, "1000–3000 Hz_40ms_modulation_mae_db": 1.3104358770993076, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8049132171310398, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2599443395295112, "3000–6000 Hz_10ms_envelope_correlation": 0.9444633000323697, "3000–6000 Hz_10ms_modulation_mae_db": 0.9286204364220104, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3614049985301537, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21152278196783927, "3000–6000 Hz_40ms_envelope_correlation": 0.9587998632936456, "3000–6000 Hz_40ms_modulation_mae_db": 1.1212655465971708, "mr_stft_db": 5.710530086550161, "band_mae_db": 0.2760837482289206, "rms_gain_error_db": -0.12964680470614692, "output_peak": 0.3066149651488733}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9156859366171822, "全帯域_10ms_rms_envelope_nrmse": 0.07451657873203822, "全帯域_10ms_envelope_correlation": 0.9938501808068116, "全帯域_10ms_modulation_mae_db": 0.7431186211749797, "全帯域_40ms_log_rms_mae_db": 0.49119258857125253, "全帯域_40ms_rms_envelope_nrmse": 0.04079917252626632, "全帯域_40ms_envelope_correlation": 0.9980859283547484, "全帯域_40ms_modulation_mae_db": 0.8959735502400914, "300–1000 Hz_10ms_log_rms_mae_db": 1.6427071940760756, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09413866469703985, "300–1000 Hz_10ms_envelope_correlation": 0.9922082663991055, "300–1000 Hz_10ms_modulation_mae_db": 0.7735290376981189, "300–1000 Hz_40ms_log_rms_mae_db": 1.0229881692004108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05383316531275979, "300–1000 Hz_40ms_envelope_correlation": 0.9973135355262299, "300–1000 Hz_40ms_modulation_mae_db": 1.0032809340540862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5945624241550196, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1333885815916237, "1000–3000 Hz_10ms_envelope_correlation": 0.9861773688520185, "1000–3000 Hz_10ms_modulation_mae_db": 0.9561145279179848, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8640069097691178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0696071094183285, "1000–3000 Hz_40ms_envelope_correlation": 0.9961352223160386, "1000–3000 Hz_40ms_modulation_mae_db": 1.0849813290411767, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8085219895191622, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2659948533604083, "3000–6000 Hz_10ms_envelope_correlation": 0.9441821492294843, "3000–6000 Hz_10ms_modulation_mae_db": 0.5261890074935153, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2974525255859781, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2084977740495022, "3000–6000 Hz_40ms_envelope_correlation": 0.961750294586536, "3000–6000 Hz_40ms_modulation_mae_db": 0.755516169006026, "mr_stft_db": 5.719813309860985, "band_mae_db": 0.2354534507414728, "rms_gain_error_db": -0.1295592184046555, "output_peak": 0.30619571689250114}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8926949707057257, "全帯域_10ms_rms_envelope_nrmse": 0.07423997439759124, "全帯域_10ms_envelope_correlation": 0.9938625375332759, "全帯域_10ms_modulation_mae_db": 0.28727269826523744, "全帯域_40ms_log_rms_mae_db": 0.44633138528356164, "全帯域_40ms_rms_envelope_nrmse": 0.040824193810734925, "全帯域_40ms_envelope_correlation": 0.998051638086019, "全帯域_40ms_modulation_mae_db": 0.1804807668373513, "300–1000 Hz_10ms_log_rms_mae_db": 1.547600960998206, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09467281796354049, "300–1000 Hz_10ms_envelope_correlation": 0.992153183013847, "300–1000 Hz_10ms_modulation_mae_db": 0.44655701740444903, "300–1000 Hz_40ms_log_rms_mae_db": 0.920060634104395, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.055125337033899505, "300–1000 Hz_40ms_envelope_correlation": 0.9972418371796467, "300–1000 Hz_40ms_modulation_mae_db": 0.4613213986322366, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6201791715250897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13217424751446477, "1000–3000 Hz_10ms_envelope_correlation": 0.9864929481870677, "1000–3000 Hz_10ms_modulation_mae_db": 1.2557337192173526, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9396128885499353, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07238707294843086, "1000–3000 Hz_40ms_envelope_correlation": 0.9958966014671123, "1000–3000 Hz_40ms_modulation_mae_db": 1.4786368984867668, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7965584579256253, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26504715269248685, "3000–6000 Hz_10ms_envelope_correlation": 0.9443963223660166, "3000–6000 Hz_10ms_modulation_mae_db": 0.9356713234451671, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3254240003604005, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2053726883823929, "3000–6000 Hz_40ms_envelope_correlation": 0.9628559074988897, "3000–6000 Hz_40ms_modulation_mae_db": 1.3390242295514139, "mr_stft_db": 5.7093233256576825, "band_mae_db": 0.2103531252409525, "rms_gain_error_db": -0.12250502956907541, "output_peak": 0.3104893429270011}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8882703188801662, "全帯域_10ms_rms_envelope_nrmse": 0.06759301268649576, "全帯域_10ms_envelope_correlation": 0.9949041372635159, "全帯域_10ms_modulation_mae_db": 0.38427676213852696, "全帯域_40ms_log_rms_mae_db": 0.4576078900389482, "全帯域_40ms_rms_envelope_nrmse": 0.03646356244193789, "全帯域_40ms_envelope_correlation": 0.9984260209803512, "全帯域_40ms_modulation_mae_db": 0.6149460395238108, "300–1000 Hz_10ms_log_rms_mae_db": 1.5427773260224362, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09028231064566782, "300–1000 Hz_10ms_envelope_correlation": 0.9928234280840845, "300–1000 Hz_10ms_modulation_mae_db": 0.34667174002144147, "300–1000 Hz_40ms_log_rms_mae_db": 0.9469806967114811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05112564834234013, "300–1000 Hz_40ms_envelope_correlation": 0.9975921629355251, "300–1000 Hz_40ms_modulation_mae_db": 0.6229343429139264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6165870782278022, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11823292176954373, "1000–3000 Hz_10ms_envelope_correlation": 0.9898752077250773, "1000–3000 Hz_10ms_modulation_mae_db": 0.9643350385485207, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8520973557800553, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06981804355167988, "1000–3000 Hz_40ms_envelope_correlation": 0.9969217212517137, "1000–3000 Hz_40ms_modulation_mae_db": 1.076591071418423, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8204296198249341, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2617760967543631, "3000–6000 Hz_10ms_envelope_correlation": 0.9489998050333276, "3000–6000 Hz_10ms_modulation_mae_db": 0.49083049951880237, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3561470505949058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20637920842533206, "3000–6000 Hz_40ms_envelope_correlation": 0.9653086448582512, "3000–6000 Hz_40ms_modulation_mae_db": 0.7781351039583747, "mr_stft_db": 5.719442911428786, "band_mae_db": 0.24985510864913327, "rms_gain_error_db": -0.10268509484335428, "output_peak": 0.2991566646400965}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9122399296229611, "全帯域_10ms_rms_envelope_nrmse": 0.07331674853824986, "全帯域_10ms_envelope_correlation": 0.9940327376844784, "全帯域_10ms_modulation_mae_db": 0.4140798057162643, "全帯域_40ms_log_rms_mae_db": 0.4688111367468604, "全帯域_40ms_rms_envelope_nrmse": 0.04024982317471841, "全帯域_40ms_envelope_correlation": 0.9981147277085738, "全帯域_40ms_modulation_mae_db": 0.5822134981058096, "300–1000 Hz_10ms_log_rms_mae_db": 1.5282867791267007, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08730845406906172, "300–1000 Hz_10ms_envelope_correlation": 0.9934487645850925, "300–1000 Hz_10ms_modulation_mae_db": 0.5722192867838888, "300–1000 Hz_40ms_log_rms_mae_db": 0.8994587804138752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04825854109359816, "300–1000 Hz_40ms_envelope_correlation": 0.9980349776578152, "300–1000 Hz_40ms_modulation_mae_db": 0.7676972330964623, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6312077636158042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12673857290212084, "1000–3000 Hz_10ms_envelope_correlation": 0.9878886058422203, "1000–3000 Hz_10ms_modulation_mae_db": 0.8690361159481821, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8995681453547282, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07347020734533143, "1000–3000 Hz_40ms_envelope_correlation": 0.9961154053389809, "1000–3000 Hz_40ms_modulation_mae_db": 1.0445108711428657, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7528145170530174, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2615222174814308, "3000–6000 Hz_10ms_envelope_correlation": 0.9473889289955915, "3000–6000 Hz_10ms_modulation_mae_db": 0.3629343189404512, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2618383896134666, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2132136540871219, "3000–6000 Hz_40ms_envelope_correlation": 0.9612413155685483, "3000–6000 Hz_40ms_modulation_mae_db": 0.287729356174585, "mr_stft_db": 5.710143808470305, "band_mae_db": 0.24637483038191468, "rms_gain_error_db": -0.11711459786300644, "output_peak": 0.3163859799792882}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9012479602918066, "全帯域_10ms_rms_envelope_nrmse": 0.07395355643158288, "全帯域_10ms_envelope_correlation": 0.9938924892322748, "全帯域_10ms_modulation_mae_db": 0.6271030795440274, "全帯域_40ms_log_rms_mae_db": 0.4507110462441717, "全帯域_40ms_rms_envelope_nrmse": 0.03764115705074424, "全帯域_40ms_envelope_correlation": 0.9983407406644847, "全帯域_40ms_modulation_mae_db": 0.8520347272918173, "300–1000 Hz_10ms_log_rms_mae_db": 1.5132612579943576, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.091977494462148, "300–1000 Hz_10ms_envelope_correlation": 0.9925773178343943, "300–1000 Hz_10ms_modulation_mae_db": 0.8467334603304325, "300–1000 Hz_40ms_log_rms_mae_db": 0.9275950986828493, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04770116860039249, "300–1000 Hz_40ms_envelope_correlation": 0.99796849103661, "300–1000 Hz_40ms_modulation_mae_db": 1.1444585773174636, "1000–3000 Hz_10ms_log_rms_mae_db": 1.616025814371084, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1377043899951011, "1000–3000 Hz_10ms_envelope_correlation": 0.9854912320023727, "1000–3000 Hz_10ms_modulation_mae_db": 0.9263999668018122, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8962335689768858, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09024350374524415, "1000–3000 Hz_40ms_envelope_correlation": 0.9936787343768925, "1000–3000 Hz_40ms_modulation_mae_db": 1.1138919084670904, "3000–6000 Hz_10ms_log_rms_mae_db": 1.82008737862903, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27285447747629066, "3000–6000 Hz_10ms_envelope_correlation": 0.9409869096106795, "3000–6000 Hz_10ms_modulation_mae_db": 0.5767102523721084, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3379709791411414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21202727046013303, "3000–6000 Hz_40ms_envelope_correlation": 0.9602451068166563, "3000–6000 Hz_40ms_modulation_mae_db": 0.8065470998002605, "mr_stft_db": 5.709038577826765, "band_mae_db": 0.2572890675463551, "rms_gain_error_db": -0.10861274816224845, "output_peak": 0.2740743878191213}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 81520, "output_samples": 81520, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.499865264792756, "spectral_convergence": 0.18782523189050057, "spectral_convergence_status": "ok", "amplitude_floor": 1.0222158521079025e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.62148608070552, "spectral_convergence": 0.18881293426571485, "spectral_convergence_status": "ok", "amplitude_floor": 7.564059658380168e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.989428008911198, "spectral_convergence": 0.19031843359634726, "spectral_convergence_status": "ok", "amplitude_floor": 6.35834815641291e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.932339290337003e-05, "output_power": 9.650298616943972e-05, "delta_db": 0.33575541108455154, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011480494900330098, "output_power": 0.0011309368847536323, "delta_db": -0.06522241474339892, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005518546578574125, "output_power": 0.0005155070985053107, "delta_db": -0.29590062635714737, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.085077476738271e-05, "output_power": 8.829030188342162e-05, "delta_db": -0.12415633517314709, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.0710168905088768e-05, "output_power": 1.934725187849135e-05, "delta_db": -0.29564355168098544, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8163123695227053e-05, "output_power": 2.6271888235942754e-05, "delta_db": -0.3018953474448641, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23642894774734904, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11890284314867736, "pitch": null, "output_rms_dbfs": -27.265689517316623, "output_peak": 0.3165971407690623, "reference_rms_dbfs": -27.14678667416795, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.703593118136491}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 81520, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.00013747273046535536, "log_rms_mae_db": 0.8999334529007896, "rms_envelope_nrmse": 0.07575331962161798, "envelope_correlation": 0.9935964238146165, "rms_gain_error_db": -0.11890278968980417, "reference_mean_power": 0.0019308462257719779, "output_mean_power": 0.0018786998842238714}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 6.283614364116833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041565087617635905, "output_power": 0.0003974166013080523, "reference_active": true, "delta_db": -0.19482695352508378}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.15171169386713e-05, "output_power": 7.818209541781443e-05, "reference_active": true, "delta_db": -0.18141505225877305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.519660659304176e-05, "output_power": 2.020338799133461e-05, "reference_active": true, "delta_db": -0.9591785095331309}], "active_band_count": 3, "active_band_mae_db": 0.4451401717723293}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012452330782545057, "log_rms_mae_db": 0.4558161858800259, "rms_envelope_nrmse": 0.0463935739671112, "envelope_correlation": 0.9973928423232157, "rms_gain_error_db": -0.11890286795138731, "reference_mean_power": 0.0019422928952156904, "output_mean_power": 0.0018898373795435174}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.257129910319434e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003707029677887353, "output_power": 0.0003527785237859786, "reference_active": true, "delta_db": -0.21523924358979446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2635280867150454e-05, "output_power": 4.1386584805883325e-05, "reference_active": true, "delta_db": -0.12909538136809992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.869258984918021e-06, "output_power": 1.4709568580375453e-06, "reference_active": true, "delta_db": -1.040695414234979}], "active_band_count": 3, "active_band_mae_db": 0.4616766797309578}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.0001332544033329483, "log_rms_mae_db": 1.6057559202771328, "rms_envelope_nrmse": 0.10057288658599063, "envelope_correlation": 0.9911569563706436, "rms_gain_error_db": -0.18054027409514534, "reference_mean_power": 0.00116782876738011, "output_mean_power": 0.0011202762828548473}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 5.939192180731074e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134913428254024, "output_power": 0.0003909090189432143, "reference_active": true, "delta_db": -0.24390730790260218}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.995066727507785e-05, "output_power": 6.436975047464691e-05, "reference_active": true, "delta_db": -0.3611003606335549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9671389178533067e-05, "output_power": 1.5836184685316233e-05, "reference_active": true, "delta_db": -0.9418447259386107}], "active_band_count": 3, "active_band_mae_db": 0.5156174648249227}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012167319573637772, "log_rms_mae_db": 0.9706178993437595, "rms_envelope_nrmse": 0.06560323136682752, "envelope_correlation": 0.9960236936701048, "rms_gain_error_db": -0.18054007776325268, "reference_mean_power": 0.0011747523056415098, "output_mean_power": 0.0011269179544928446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.189128511363634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037925439414056246, "output_power": 0.00035567730138929627, "reference_active": true, "delta_db": -0.2787447106839806}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.625302583005344e-05, "output_power": 3.386277860473059e-05, "reference_active": true, "delta_db": -0.29621669267002093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7116807182635842e-06, "output_power": 1.3467749895802675e-06, "reference_active": true, "delta_db": -1.0412771577387665}], "active_band_count": 3, "active_band_mae_db": 0.5387461870309227}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 4.762512740172365e-05, "log_rms_mae_db": 1.5251863332439304, "rms_envelope_nrmse": 0.11334108922775432, "envelope_correlation": 0.990122932509228, "rms_gain_error_db": -0.15997969038231422, "reference_mean_power": 0.00010302415332771722, "output_mean_power": 9.929813347443106e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 3.1359809483594164e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4569632781045495e-05, "output_power": 1.394428924068765e-05, "reference_active": true, "delta_db": -0.1905222318490017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.842890886644979e-06, "output_power": 3.4146382229873383e-06, "reference_active": true, "delta_db": -0.5131335602128787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.113490153313903e-06, "output_power": 9.440807877458419e-07, "reference_active": true, "delta_db": -0.7167722115134999}], "active_band_count": 3, "active_band_mae_db": 0.4734760011917934}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 4.327669210129085e-05, "log_rms_mae_db": 0.8410833191227145, "rms_envelope_nrmse": 0.057620600176892245, "envelope_correlation": 0.9974552164585353, "rms_gain_error_db": -0.15998299313311512, "reference_mean_power": 0.00010363449577271808, "output_mean_power": 9.988632602605264e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 2.7762525532534898e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3720292251105881e-05, "output_power": 1.312688528715845e-05, "reference_active": true, "delta_db": -0.19201672163055447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7263230443470765e-06, "output_power": 1.4902250931573737e-06, "reference_active": true, "delta_db": -0.6387019589405202}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.698769643844551e-08, "output_power": 7.342923405298648e-08, "reference_active": true, "delta_db": -1.2084764642889587}], "active_band_count": 3, "active_band_mae_db": 0.6797317149533445}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 1.6513968336769973e-05, "log_rms_mae_db": 1.8009474426715513, "rms_envelope_nrmse": 0.2610863941747794, "envelope_correlation": 0.9455316163568264, "rms_gain_error_db": 0.1843127082148216, "reference_mean_power": 8.517905641506151e-06, "output_mean_power": 8.887182439370651e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 4.8245079516023535e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4329635827636286e-06, "output_power": 2.336755771740551e-06, "reference_active": true, "delta_db": -0.17522284284009598}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1930715407234243e-06, "output_power": 1.0268994803441415e-06, "reference_active": true, "delta_db": -0.6513855217466891}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.14650683405709e-07, "output_power": 1.7985159682918708e-07, "reference_active": true, "delta_db": -2.4291438125491003}], "active_band_count": 3, "active_band_mae_db": 1.0852507257119617}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 1.2321016663379126e-05, "log_rms_mae_db": 1.3283103308073092, "rms_envelope_nrmse": 0.20575819297319922, "envelope_correlation": 0.9622667550315543, "rms_gain_error_db": 0.18432043619856303, "reference_mean_power": 8.568334809688713e-06, "output_mean_power": 8.939813771280777e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 4.130413403311084e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.518595705395907e-06, "output_power": 2.3360190655644723e-06, "reference_active": true, "delta_db": -0.3268207550745503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.297939167002702e-07, "output_power": 5.482357808667785e-07, "reference_active": true, "delta_db": -0.6023108482479057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9684693476821222e-08, "output_power": 1.3659858783859214e-08, "reference_active": true, "delta_db": -3.3708635915873133}], "active_band_count": 3, "active_band_mae_db": 1.4333317316365897}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9228744736495437, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4134c532929d56783d0c4e2a9fdc0b15a4a982492b72f83857a7a7510e13b424", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9879970994250784, "standard_deviation": 0.0017995387774730085, "minimum": 0.9854912320023727, "maximum": 0.990122932509228, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.604671192509345, "standard_deviation": 0.033758644508826666, "minimum": 1.5251863332439304, "maximum": 1.6312077636158042, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9012749938091875, "standard_deviation": 0.22118554713989097, "minimum": 0.4734760011917934, "maximum": 1.2557337192173526, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12519013152115296, "standard_deviation": 0.008791202900749013, "minimum": 0.11334108922775432, "maximum": 0.1377043899951011, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961636548686987, "standard_deviation": 0.001132543494939828, "minimum": 0.9936787343768925, "maximum": 0.9974552164585353, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8800074310039872, "standard_deviation": 0.03655719181538868, "minimum": 0.8381803806151256, "maximum": 0.9396128885499353, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0867967448112603, "standard_deviation": 0.2404516972470769, "minimum": 0.6797317149533445, "maximum": 1.4786368984867668, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07089661183883263, "standard_deviation": 0.009337646121937658, "minimum": 0.057620600176892245, "maximum": 0.09024350374524415, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.944740413578381, "standard_deviation": 0.0026194384532329936, "minimum": 0.9409869096106795, "maximum": 0.9489998050333276, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8044162694366968, "standard_deviation": 0.023800932360704345, "minimum": 1.7528145170530174, "maximum": 1.8310575327392153, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.759961397702386, "standard_deviation": 0.3057606216048074, "minimum": 0.3629343189404512, "maximum": 1.173484617715071, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2652942554773408, "standard_deviation": 0.005460067466472954, "minimum": 0.2599443395295112, "maximum": 0.2741285123494563, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9614705332762221, "standard_deviation": 0.002096808302338426, "minimum": 0.9587998632936456, "maximum": 0.9653086448582512, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3278751068593586, "standard_deviation": 0.033857735949650615, "minimum": 1.2618383896134666, "maximum": 1.3614049985301537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0040284873044112, "standard_deviation": 0.4194949151520414, "minimum": 0.287729356174585, "maximum": 1.5106786617108696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.210124457933427, "standard_deviation": 0.004456773441859282, "minimum": 0.2053726883823929, "maximum": 0.2182240931218953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925888093950421, "standard_deviation": 0.0007604535255744824, "minimum": 0.9911569563706436, "maximum": 0.9934794976547223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.556804193999224, "standard_deviation": 0.044267037594783856, "minimum": 1.5132612579943576, "maximum": 1.6427071940760756, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5901072943571544, "standard_deviation": 0.2406303342231759, "minimum": 0.27862241907401697, "maximum": 0.9409079287199646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09216074982279926, "standard_deviation": 0.004322419171522519, "minimum": 0.08730845406906172, "maximum": 0.10057288658599063, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9974488294646077, "standard_deviation": 0.000639534794309008, "minimum": 0.9960236936701048, "maximum": 0.9980349776578152, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9518075810662345, "standard_deviation": 0.039069734749354, "minimum": 0.8994587804138752, "maximum": 1.0229881692004108, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7933948144805514, "standard_deviation": 0.3171465089173596, "minimum": 0.4613213986322366, "maximum": 1.2963504079707222, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.05316049157885068, "standard_deviation": 0.005617196081144525, "minimum": 0.04770116860039249, "maximum": 0.06560323136682752, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2374799037487906, "standard_deviation": 0.02752703877963264, "minimum": 0.18800095145422677, "maximum": 0.2760837482289206, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.711966233891569, "standard_deviation": 0.0055039599113341515, "minimum": 5.703593118136491, "maximum": 5.719813309860985, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.30414391334129864, "standard_deviation": 0.013547996194696783, "minimum": 0.2740743878191213, "maximum": 0.3165971407690623, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11583610025893684, "standard_deviation": 0.01189225665873719, "minimum": -0.12964680470614692, "maximum": -0.09766246537433033, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9941213832924782, "standard_deviation": 0.0004345287936800932, "minimum": 0.9935964238146165, "maximum": 0.9949041372635159, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9022990148310756, "standard_deviation": 0.00984877630017523, "minimum": 0.8882703188801662, "maximum": 0.9156859366171822, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.47749264514445156, "standard_deviation": 0.19862041887697, "minimum": 0.20196764797256472, "maximum": 0.7431186211749797, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07261506537082615, "standard_deviation": 0.002715946666491012, "minimum": 0.06759301268649576, "maximum": 0.07575331962161798, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980854460689077, "standard_deviation": 0.000310177803924206, "minimum": 0.9973928423232157, "maximum": 0.9984260209803512, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.4655972379581001, "standard_deviation": 0.017343317568596788, "minimum": 0.44633138528356164, "maximum": 0.4916573384309007, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5948663285085511, "standard_deviation": 0.26183655027101943, "minimum": 0.1804807668373513, "maximum": 0.8959735502400914, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04023638118578138, "standard_deviation": 0.0029279690734718487, "minimum": 0.03646356244193789, "maximum": 0.0463935739671112, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-284449-0020", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.6590051651000977, "predicted_mos": 2.6590051651000977, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "496233aef78249b4c1261ef3ff0941f4edd0c2f25ad66f840adc53b307bc5390", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7376317996531725, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9093251509148197, "全帯域_10ms_rms_envelope_nrmse": 0.07438773733722218, "全帯域_10ms_envelope_correlation": 0.9938915635182037, "全帯域_10ms_modulation_mae_db": 0.5655759487283735, "全帯域_40ms_log_rms_mae_db": 0.46491595118407375, "全帯域_40ms_rms_envelope_nrmse": 0.040521688969431216, "全帯域_40ms_envelope_correlation": 0.9981184964785952, "全帯域_40ms_modulation_mae_db": 0.641729960505749, "300–1000 Hz_10ms_log_rms_mae_db": 1.5753699887585604, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09809518090174253, "300–1000 Hz_10ms_envelope_correlation": 0.9917817065718594, "300–1000 Hz_10ms_modulation_mae_db": 0.7371413385940815, "300–1000 Hz_40ms_log_rms_mae_db": 0.959884263876519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.057807229585177276, "300–1000 Hz_40ms_envelope_correlation": 0.9972084892645733, "300–1000 Hz_40ms_modulation_mae_db": 0.814983532854666, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6431923010954836, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12759482244986012, "1000–3000 Hz_10ms_envelope_correlation": 0.9872855560241118, "1000–3000 Hz_10ms_modulation_mae_db": 1.0436313093145124, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9131489378280783, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0631714340581825, "1000–3000 Hz_40ms_envelope_correlation": 0.9966489473318839, "1000–3000 Hz_40ms_modulation_mae_db": 1.3433452751255253, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8064374064373037, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.262345012206337, "3000–6000 Hz_10ms_envelope_correlation": 0.9448744956745811, "3000–6000 Hz_10ms_modulation_mae_db": 0.2456143351967249, "3000–6000 Hz_40ms_log_rms_mae_db": 1.313406169559298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2064851399415895, "3000–6000 Hz_40ms_envelope_correlation": 0.961820378657476, "3000–6000 Hz_40ms_modulation_mae_db": 0.481779136791238, "mr_stft_db": 5.69340692213948, "band_mae_db": 0.158984971060071, "rms_gain_error_db": -0.12364931932239634, "output_peak": 0.2731347930961396}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.893497868187813, "全帯域_10ms_rms_envelope_nrmse": 0.07408906030455296, "全帯域_10ms_envelope_correlation": 0.9939511014410647, "全帯域_10ms_modulation_mae_db": 0.5676700311415336, "全帯域_40ms_log_rms_mae_db": 0.4930191547329577, "全帯域_40ms_rms_envelope_nrmse": 0.04347071375528038, "全帯域_40ms_envelope_correlation": 0.9978210359669107, "全帯域_40ms_modulation_mae_db": 0.6129699549816275, "300–1000 Hz_10ms_log_rms_mae_db": 1.6209940991401874, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09911988421619987, "300–1000 Hz_10ms_envelope_correlation": 0.9915306354288185, "300–1000 Hz_10ms_modulation_mae_db": 0.693008559975167, "300–1000 Hz_40ms_log_rms_mae_db": 0.9630851002587142, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0620934316358771, "300–1000 Hz_40ms_envelope_correlation": 0.9966137297370297, "300–1000 Hz_40ms_modulation_mae_db": 0.8126765028758789, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5697012565513357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11991057739238237, "1000–3000 Hz_10ms_envelope_correlation": 0.9888432665008403, "1000–3000 Hz_10ms_modulation_mae_db": 0.2515940278970427, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8228049227387354, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0639722830357578, "1000–3000 Hz_40ms_envelope_correlation": 0.9967040559896458, "1000–3000 Hz_40ms_modulation_mae_db": 0.2404967598393213, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8253315850730678, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2683448783360911, "3000–6000 Hz_10ms_envelope_correlation": 0.9439101183898081, "3000–6000 Hz_10ms_modulation_mae_db": 0.5314304003047357, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3467488977140718, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20932841585837503, "3000–6000 Hz_40ms_envelope_correlation": 0.9620765486330611, "3000–6000 Hz_40ms_modulation_mae_db": 0.5049630029739911, "mr_stft_db": 5.700929598652891, "band_mae_db": 0.21621375451610833, "rms_gain_error_db": -0.13754477097460058, "output_peak": 0.30790053321597477}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9098252910824915, "全帯域_10ms_rms_envelope_nrmse": 0.07549169354141971, "全帯域_10ms_envelope_correlation": 0.9936531876747567, "全帯域_10ms_modulation_mae_db": 0.36985658784381314, "全帯域_40ms_log_rms_mae_db": 0.469315978606053, "全帯域_40ms_rms_envelope_nrmse": 0.040851533012675634, "全帯域_40ms_envelope_correlation": 0.998034977881651, "全帯域_40ms_modulation_mae_db": 0.5012625823720587, "300–1000 Hz_10ms_log_rms_mae_db": 1.5401534547592617, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09505162489243241, "300–1000 Hz_10ms_envelope_correlation": 0.9922577585963726, "300–1000 Hz_10ms_modulation_mae_db": 0.4243009397813637, "300–1000 Hz_40ms_log_rms_mae_db": 0.9969776673221346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05513961625289663, "300–1000 Hz_40ms_envelope_correlation": 0.9974542490303534, "300–1000 Hz_40ms_modulation_mae_db": 0.5083361667119127, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6168024409291373, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13306485120434328, "1000–3000 Hz_10ms_envelope_correlation": 0.9862739856976642, "1000–3000 Hz_10ms_modulation_mae_db": 0.6613625713469647, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8523929086866842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06704845017087965, "1000–3000 Hz_40ms_envelope_correlation": 0.9964356952455043, "1000–3000 Hz_40ms_modulation_mae_db": 0.9027018385143833, "3000–6000 Hz_10ms_log_rms_mae_db": 1.79067247423038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26847743581027245, "3000–6000 Hz_10ms_envelope_correlation": 0.9424115079318466, "3000–6000 Hz_10ms_modulation_mae_db": 0.2752252440952679, "3000–6000 Hz_40ms_log_rms_mae_db": 1.32313508670994, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21542846845349226, "3000–6000 Hz_40ms_envelope_correlation": 0.9584855955175443, "3000–6000 Hz_40ms_modulation_mae_db": 0.2173297915183623, "mr_stft_db": 5.696013551364975, "band_mae_db": 0.22194835575520222, "rms_gain_error_db": -0.11364895911349959, "output_peak": 0.284721469500459}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8986452526568691, "全帯域_10ms_rms_envelope_nrmse": 0.0727203307769422, "全帯域_10ms_envelope_correlation": 0.9940866404221728, "全帯域_10ms_modulation_mae_db": 0.5897993234069587, "全帯域_40ms_log_rms_mae_db": 0.4432793947301485, "全帯域_40ms_rms_envelope_nrmse": 0.038886712020947255, "全帯域_40ms_envelope_correlation": 0.998219805770998, "全帯域_40ms_modulation_mae_db": 0.7435653939286757, "300–1000 Hz_10ms_log_rms_mae_db": 1.564196360772423, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08975661390112463, "300–1000 Hz_10ms_envelope_correlation": 0.9929196370021868, "300–1000 Hz_10ms_modulation_mae_db": 0.6334011916248691, "300–1000 Hz_40ms_log_rms_mae_db": 0.9348632592750796, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.050722954256662726, "300–1000 Hz_40ms_envelope_correlation": 0.9976426582127846, "300–1000 Hz_40ms_modulation_mae_db": 0.7979378371399989, "1000–3000 Hz_10ms_log_rms_mae_db": 1.667787004003924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1355101894018443, "1000–3000 Hz_10ms_envelope_correlation": 0.9863200403706646, "1000–3000 Hz_10ms_modulation_mae_db": 0.9216750192856381, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9281024875596927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0842815595341618, "1000–3000 Hz_40ms_envelope_correlation": 0.9950112150877065, "1000–3000 Hz_40ms_modulation_mae_db": 1.2130388594013741, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7695482445660846, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.26244356534177793, "3000–6000 Hz_10ms_envelope_correlation": 0.9481883254390131, "3000–6000 Hz_10ms_modulation_mae_db": 0.5856768975258386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3125315976720384, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.211892314855414, "3000–6000 Hz_40ms_envelope_correlation": 0.9627927594594515, "3000–6000 Hz_40ms_modulation_mae_db": 0.7411102138584047, "mr_stft_db": 5.703569548936461, "band_mae_db": 0.23956305626569385, "rms_gain_error_db": -0.11339350161784131, "output_peak": 0.2874137316287135}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8932145847193529, "全帯域_10ms_rms_envelope_nrmse": 0.07826146908198532, "全帯域_10ms_envelope_correlation": 0.9931058759410163, "全帯域_10ms_modulation_mae_db": 0.22860811449539145, "全帯域_40ms_log_rms_mae_db": 0.45711290230980545, "全帯域_40ms_rms_envelope_nrmse": 0.045861420172701166, "全帯域_40ms_envelope_correlation": 0.9974097341007155, "全帯域_40ms_modulation_mae_db": 0.5339026924738719, "300–1000 Hz_10ms_log_rms_mae_db": 1.5494232185857086, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1004343137162559, "300–1000 Hz_10ms_envelope_correlation": 0.991220548810496, "300–1000 Hz_10ms_modulation_mae_db": 0.3997831266441305, "300–1000 Hz_40ms_log_rms_mae_db": 0.9376382785134268, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06185000820694883, "300–1000 Hz_40ms_envelope_correlation": 0.9965603015464339, "300–1000 Hz_40ms_modulation_mae_db": 0.638976796052162, "1000–3000 Hz_10ms_log_rms_mae_db": 1.625862216825226, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12943014734157574, "1000–3000 Hz_10ms_envelope_correlation": 0.9870037401054712, "1000–3000 Hz_10ms_modulation_mae_db": 1.1106159841931234, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8826473544617113, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07034596852147292, "1000–3000 Hz_40ms_envelope_correlation": 0.9960021551908926, "1000–3000 Hz_40ms_modulation_mae_db": 1.2264243061133824, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8311942438626263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2733013906383311, "3000–6000 Hz_10ms_envelope_correlation": 0.9409626936797075, "3000–6000 Hz_10ms_modulation_mae_db": 0.43271858394816326, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3359999148139305, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21252390378314212, "3000–6000 Hz_40ms_envelope_correlation": 0.9602228013191547, "3000–6000 Hz_40ms_modulation_mae_db": 0.5801742172448513, "mr_stft_db": 5.694083598924144, "band_mae_db": 0.1539208341961312, "rms_gain_error_db": -0.10519799140910129, "output_peak": 0.28700006590537225}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8926517491960372, "全帯域_10ms_rms_envelope_nrmse": 0.07295303930708599, "全帯域_10ms_envelope_correlation": 0.993936039163972, "全帯域_10ms_modulation_mae_db": 0.34849565300988417, "全帯域_40ms_log_rms_mae_db": 0.4357997597879153, "全帯域_40ms_rms_envelope_nrmse": 0.03758993044146423, "全帯域_40ms_envelope_correlation": 0.9981756042683734, "全帯域_40ms_modulation_mae_db": 0.49509941524593204, "300–1000 Hz_10ms_log_rms_mae_db": 1.5423855907876673, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09356564286241935, "300–1000 Hz_10ms_envelope_correlation": 0.9921727577333428, "300–1000 Hz_10ms_modulation_mae_db": 0.5590981691807732, "300–1000 Hz_40ms_log_rms_mae_db": 0.9403137311594384, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05217011231017782, "300–1000 Hz_40ms_envelope_correlation": 0.9973182840532118, "300–1000 Hz_40ms_modulation_mae_db": 0.836113136108732, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6386073592076775, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13648885088230833, "1000–3000 Hz_10ms_envelope_correlation": 0.9855169613647053, "1000–3000 Hz_10ms_modulation_mae_db": 0.7895990359629477, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8784136634612628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07030950038042057, "1000–3000 Hz_40ms_envelope_correlation": 0.9960782037560957, "1000–3000 Hz_40ms_modulation_mae_db": 1.0478951457485792, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8305965716742896, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2753674257045364, "3000–6000 Hz_10ms_envelope_correlation": 0.939584144162551, "3000–6000 Hz_10ms_modulation_mae_db": 0.26763928707404405, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3380878745656983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2163734735766076, "3000–6000 Hz_40ms_envelope_correlation": 0.9583786252767587, "3000–6000 Hz_40ms_modulation_mae_db": 0.5092122024058235, "mr_stft_db": 5.706988537429592, "band_mae_db": 0.14366600580669878, "rms_gain_error_db": -0.04319114734906274, "output_peak": 0.29875545765502676}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9522438775426871, "全帯域_10ms_rms_envelope_nrmse": 0.07873028083738332, "全帯域_10ms_envelope_correlation": 0.9933876754468353, "全帯域_10ms_modulation_mae_db": 0.6046207225532771, "全帯域_40ms_log_rms_mae_db": 0.4988053009935252, "全帯域_40ms_rms_envelope_nrmse": 0.050309980028055964, "全帯域_40ms_envelope_correlation": 0.9972931275049051, "全帯域_40ms_modulation_mae_db": 0.682155080694872, "300–1000 Hz_10ms_log_rms_mae_db": 1.660296366751414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10459641450544846, "300–1000 Hz_10ms_envelope_correlation": 0.9910185317016815, "300–1000 Hz_10ms_modulation_mae_db": 0.5536356894071539, "300–1000 Hz_40ms_log_rms_mae_db": 1.0360035942228318, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07279937212296134, "300–1000 Hz_40ms_envelope_correlation": 0.9957543486221893, "300–1000 Hz_40ms_modulation_mae_db": 0.7227351172413723, "1000–3000 Hz_10ms_log_rms_mae_db": 1.596906194087209, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12925573843171792, "1000–3000 Hz_10ms_envelope_correlation": 0.9872213665171027, "1000–3000 Hz_10ms_modulation_mae_db": 0.7952553699105098, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8932483733168493, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08103606434643536, "1000–3000 Hz_40ms_envelope_correlation": 0.9949021805889722, "1000–3000 Hz_40ms_modulation_mae_db": 0.904218086598403, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8264818348030871, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28459163465742454, "3000–6000 Hz_10ms_envelope_correlation": 0.9385193878136272, "3000–6000 Hz_10ms_modulation_mae_db": 0.7744149570597173, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3544300167254146, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2235775115562089, "3000–6000 Hz_40ms_envelope_correlation": 0.9581605524221289, "3000–6000 Hz_40ms_modulation_mae_db": 0.7444099754280425, "mr_stft_db": 5.695260994194961, "band_mae_db": 0.24326239765878266, "rms_gain_error_db": -0.1966098077742663, "output_peak": 0.29562071068547013}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.8938030869427286, "全帯域_10ms_rms_envelope_nrmse": 0.07617769174871623, "全帯域_10ms_envelope_correlation": 0.9938111833907579, "全帯域_10ms_modulation_mae_db": 0.9216759470765101, "全帯域_40ms_log_rms_mae_db": 0.4588492464355892, "全帯域_40ms_rms_envelope_nrmse": 0.04430382760195778, "全帯域_40ms_envelope_correlation": 0.9979921149371875, "全帯域_40ms_modulation_mae_db": 1.3117872316876023, "300–1000 Hz_10ms_log_rms_mae_db": 1.568636525213572, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1065310569761812, "300–1000 Hz_10ms_envelope_correlation": 0.9903875590949107, "300–1000 Hz_10ms_modulation_mae_db": 1.2139615818781404, "300–1000 Hz_40ms_log_rms_mae_db": 0.9541349322999886, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06603663461870957, "300–1000 Hz_40ms_envelope_correlation": 0.9963766787397718, "300–1000 Hz_40ms_modulation_mae_db": 1.7366634408838852, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6137383352938208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12006839104539141, "1000–3000 Hz_10ms_envelope_correlation": 0.9890669961322369, "1000–3000 Hz_10ms_modulation_mae_db": 0.7704685816211997, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8827731656929921, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07033365439213829, "1000–3000 Hz_40ms_envelope_correlation": 0.9963394530839174, "1000–3000 Hz_40ms_modulation_mae_db": 0.9325137227283218, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8021819038576343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27301702533035116, "3000–6000 Hz_10ms_envelope_correlation": 0.93844902306886, "3000–6000 Hz_10ms_modulation_mae_db": 1.2058049559430934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3216460425532588, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21385758291405474, "3000–6000 Hz_40ms_envelope_correlation": 0.9578650290536112, "3000–6000 Hz_40ms_modulation_mae_db": 1.5584278109587955, "mr_stft_db": 5.713353246439477, "band_mae_db": 0.25692582103155365, "rms_gain_error_db": -0.1869522269929051, "output_peak": 0.2925822347879084}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 81520, "output_samples": 81520, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.518145348466053, "spectral_convergence": 0.18893475615939678, "spectral_convergence_status": "ok", "amplitude_floor": 1.0222158521079025e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.585832649162011, "spectral_convergence": 0.18881336310707264, "spectral_convergence_status": "ok", "amplitude_floor": 7.564059658380168e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.976242768790374, "spectral_convergence": 0.1911124973335043, "spectral_convergence_status": "ok", "amplitude_floor": 6.35834815641291e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.932339290337003e-05, "output_power": 9.574290144151112e-05, "delta_db": 0.3014137325953575, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011480494900330098, "output_power": 0.0011109288970684648, "delta_db": -0.1427434637783859, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005518546578574125, "output_power": 0.0005286271356528623, "delta_db": -0.18675259800747618, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.085077476738271e-05, "output_power": 9.071185628806748e-05, "delta_db": -0.006645810365642558, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.0710168905088768e-05, "output_power": 1.9814433791055743e-05, "delta_db": -0.19201974235205754, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8163123695227053e-05, "output_power": 2.8981060304042202e-05, "delta_db": 0.1243344792615062, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.158984971060071, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.12364931932239634, "pitch": null, "output_rms_dbfs": -27.270435993490345, "output_peak": 0.2731347930961396, "reference_rms_dbfs": -27.14678667416795, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.69340692213948}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 81520, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.00013747273046535536, "log_rms_mae_db": 0.9093251509148197, "rms_envelope_nrmse": 0.07438773733722218, "envelope_correlation": 0.9938915635182037, "rms_gain_error_db": -0.1236492915197987, "reference_mean_power": 0.0019308462257719779, "output_mean_power": 0.0018766477325958895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 6.283614364116833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041565087617635905, "output_power": 0.00039828018746824117, "reference_active": true, "delta_db": -0.1853999743901051}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.15171169386713e-05, "output_power": 7.83692346902179e-05, "reference_active": true, "delta_db": -0.17103205621421033}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.519660659304176e-05, "output_power": 1.850599641170649e-05, "reference_active": true, "delta_db": -1.340295815580805}], "active_band_count": 3, "active_band_mae_db": 0.5655759487283735}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012452330782545057, "log_rms_mae_db": 0.46491595118407375, "rms_envelope_nrmse": 0.040521688969431216, "envelope_correlation": 0.9981184964785952, "rms_gain_error_db": -0.12364966646793107, "reference_mean_power": 0.0019422928952156904, "output_mean_power": 0.0018877729331838926}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.257129910319434e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003707029677887353, "output_power": 0.00035461810159447877, "reference_active": true, "delta_db": -0.19265162404187508}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2635280867150454e-05, "output_power": 4.099477745896911e-05, "reference_active": true, "delta_db": -0.17040595139559883}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.869258984918021e-06, "output_power": 1.3045365341587375e-06, "reference_active": true, "delta_db": -1.5621323060797732}], "active_band_count": 3, "active_band_mae_db": 0.641729960505749}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.0001332544033329483, "log_rms_mae_db": 1.5753699887585604, "rms_envelope_nrmse": 0.09809518090174253, "envelope_correlation": 0.9917817065718594, "rms_gain_error_db": -0.2084162871636725, "reference_mean_power": 0.00116782876738011, "output_mean_power": 0.001113108605715697}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 5.939192180731074e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134913428254024, "output_power": 0.00039074430647074727, "reference_active": true, "delta_db": -0.24573762619400025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.995066727507785e-05, "output_power": 6.475346285271195e-05, "reference_active": true, "delta_db": -0.335288633264471}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9671389178533067e-05, "output_power": 1.3514352883086003e-05, "reference_active": true, "delta_db": -1.6303977563237733}], "active_band_count": 3, "active_band_mae_db": 0.7371413385940815}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012167319573637772, "log_rms_mae_db": 0.959884263876519, "rms_envelope_nrmse": 0.057807229585177276, "envelope_correlation": 0.9972084892645733, "rms_gain_error_db": -0.20841666382493573, "reference_mean_power": 0.0011747523056415098, "output_mean_power": 0.0011197076353219664}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.189128511363634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037925439414056246, "output_power": 0.0003562869909931552, "reference_active": true, "delta_db": -0.27130655969276074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.625302583005344e-05, "output_power": 3.308169265027761e-05, "reference_active": true, "delta_db": -0.397565380037718}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7116807182635842e-06, "output_power": 1.1371425032675923e-06, "reference_active": true, "delta_db": -1.776078658833519}], "active_band_count": 3, "active_band_mae_db": 0.814983532854666}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 4.762512740172365e-05, "log_rms_mae_db": 1.6431923010954836, "rms_envelope_nrmse": 0.12759482244986012, "envelope_correlation": 0.9872855560241118, "rms_gain_error_db": -0.057733302277447336, "reference_mean_power": 0.00010302415332771722, "output_mean_power": 0.00010166365605704798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 3.1359809483594164e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4569632781045495e-05, "output_power": 1.372808098898655e-05, "reference_active": true, "delta_db": -0.25838773135102905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.842890886644979e-06, "output_power": 3.009334100837289e-06, "reference_active": true, "delta_db": -1.0618764702371661}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.113490153313903e-06, "output_power": 7.338772218724348e-07, "reference_active": true, "delta_db": -1.810629726355342}], "active_band_count": 3, "active_band_mae_db": 1.0436313093145124}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 4.327669210129085e-05, "log_rms_mae_db": 0.9131489378280783, "rms_envelope_nrmse": 0.0631714340581825, "envelope_correlation": 0.9966489473318839, "rms_gain_error_db": -0.05773324156746651, "reference_mean_power": 0.00010363449577271808, "output_mean_power": 0.00010226593998447722}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 2.7762525532534898e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3720292251105881e-05, "output_power": 1.2822257650263443e-05, "reference_active": true, "delta_db": -0.29398862880280696}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7263230443470765e-06, "output_power": 1.325656501729142e-06, "reference_active": true, "delta_db": -1.146910617120634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.698769643844551e-08, "output_power": 5.3432191449168605e-08, "reference_active": true, "delta_db": -2.5891365794531342}], "active_band_count": 3, "active_band_mae_db": 1.3433452751255253}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 1.6513968336769973e-05, "log_rms_mae_db": 1.8064374064373037, "rms_envelope_nrmse": 0.262345012206337, "envelope_correlation": 0.9448744956745811, "rms_gain_error_db": 0.16350092355703078, "reference_mean_power": 8.517905641506151e-06, "output_mean_power": 8.84469613713991e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 4.8245079516023535e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4329635827636286e-06, "output_power": 2.4743303257211054e-06, "reference_active": true, "delta_db": 0.07322069591342592}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1930715407234243e-06, "output_power": 1.0828888743573659e-06, "reference_active": true, "delta_db": -0.4208259448228452}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.14650683405709e-07, "output_power": 2.9754252409076604e-07, "reference_active": true, "delta_db": -0.2427963648539035}], "active_band_count": 3, "active_band_mae_db": 0.2456143351967249}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 1.2321016663379126e-05, "log_rms_mae_db": 1.313406169559298, "rms_envelope_nrmse": 0.2064851399415895, "envelope_correlation": 0.961820378657476, "rms_gain_error_db": 0.16349646798866102, "reference_mean_power": 8.568334809688713e-06, "output_mean_power": 8.897050898772084e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 4.130413403311084e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.518595705395907e-06, "output_power": 2.537427357043728e-06, "reference_active": true, "delta_db": 0.03235159378863762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.297939167002702e-07, "output_power": 5.900585954234052e-07, "reference_active": true, "delta_db": -0.28303320244486313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9684693476821222e-08, "output_power": 2.2884282868385267e-08, "reference_active": true, "delta_db": -1.1299526141402132}], "active_band_count": 3, "active_band_mae_db": 0.481779136791238}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9209108206634217, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9e2b522dae2459ce7620b4acabbfaf99742a26dd316a4f6554dd9fd57ee15a0d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9871914890890996, "standard_deviation": 0.001236132485880122, "minimum": 0.9855169613647053, "maximum": 0.9890669961322369, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6215746384992267, "standard_deviation": 0.029988008046776257, "minimum": 1.5697012565513357, "maximum": 1.667787004003924, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7930252374414923, "standard_deviation": 0.26490601552559057, "minimum": 0.2515940278970427, "maximum": 1.1106159841931234, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12891544601867794, "standard_deviation": 0.006319442432219816, "minimum": 0.11991057739238237, "maximum": 0.13648885088230833, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960152382843273, "standard_deviation": 0.0006978591866867455, "minimum": 0.9949021805889722, "maximum": 0.9967040559896458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8816914767182507, "standard_deviation": 0.03298479257198983, "minimum": 0.8228049227387354, "maximum": 0.9281024875596927, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9763292492586613, "standard_deviation": 0.3406698623980005, "minimum": 0.2404967598393213, "maximum": 1.3433452751255253, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07131236430493111, "standard_deviation": 0.007596807594630771, "minimum": 0.0631714340581825, "maximum": 0.0842815595341618, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9421124620199994, "standard_deviation": 0.0034207615840150026, "minimum": 0.93844902306886, "maximum": 0.9481883254390131, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.810305533063059, "standard_deviation": 0.022242458701435614, "minimum": 1.7695482445660846, "maximum": 1.8311942438626263, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5398155826434481, "standard_deviation": 0.3257660000591626, "minimum": 0.2456143351967249, "maximum": 1.2058049559430934, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.27098604600314025, "standard_deviation": 0.007316366791527178, "minimum": 0.262345012206337, "maximum": 0.28459163465742454, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9599752862923984, "standard_deviation": 0.00201189454540441, "minimum": 0.9578650290536112, "maximum": 0.9627927594594515, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3307482000392064, "standard_deviation": 0.01545362410709785, "minimum": 1.3125315976720384, "maximum": 1.3544300167254146, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6671757938974386, "standard_deviation": 0.3966520032195069, "minimum": 0.2173297915183623, "maximum": 1.5584278109587955, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.21368335136736052, "standard_deviation": 0.005120530038477425, "minimum": 0.2064851399415895, "maximum": 0.2235775115562089, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9916611418674586, "standard_deviation": 0.0007984058441300455, "minimum": 0.9903875590949107, "maximum": 0.9929196370021868, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5776819505960993, "standard_deviation": 0.04212716622545042, "minimum": 1.5401534547592617, "maximum": 1.660296366751414, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6517913246357099, "standard_deviation": 0.2560255671784573, "minimum": 0.3997831266441305, "maximum": 1.2139615818781404, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09839384149647554, "standard_deviation": 0.005587609572825806, "minimum": 0.08975661390112463, "maximum": 0.1065310569761812, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968660924007935, "standard_deviation": 0.0006440264530529331, "minimum": 0.9957543486221893, "maximum": 0.9976426582127846, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9653626033660166, "standard_deviation": 0.03479986834344092, "minimum": 0.9348632592750796, "maximum": 1.0360035942228318, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8585528162335759, "standard_deviation": 0.371977472482301, "minimum": 0.5083361667119127, "maximum": 1.7366634408838852, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.059827419873676416, "standard_deviation": 0.007403029212637278, "minimum": 0.050722954256662726, "maximum": 0.07279937212296134, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2043106495362802, "standard_deviation": 0.04512272654304448, "minimum": 0.14366600580669878, "maximum": 0.25692582103155365, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.700450749760248, "standard_deviation": 0.007130935647662454, "minimum": 5.69340692213948, "maximum": 5.713353246439477, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.29089112455938304, "standard_deviation": 0.010399046015567091, "minimum": 0.2731347930961396, "maximum": 0.30790053321597477, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12752346556920915, "standard_deviation": 0.04843649810143589, "minimum": -0.1966098077742663, "maximum": -0.04319114734906274, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9937279083748474, "standard_deviation": 0.00033032580524014043, "minimum": 0.9931058759410163, "maximum": 0.9940866404221728, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.9054008576553498, "standard_deviation": 0.02022062899312504, "minimum": 0.8926517491960372, "maximum": 0.9522438775426871, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5245377910319677, "standard_deviation": 0.21193902811677653, "minimum": 0.22860811449539145, "maximum": 0.9216759470765101, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.07535141286691349, "standard_deviation": 0.0022609917106617963, "minimum": 0.0727203307769422, "maximum": 0.07873028083738332, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978831121136671, "standard_deviation": 0.00035158557898329074, "minimum": 0.9972931275049051, "maximum": 0.998219805770998, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.46513721109750855, "standard_deviation": 0.021944190017855775, "minimum": 0.4357997597879153, "maximum": 0.4988053009935252, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6903090389862987, "standard_deviation": 0.26613300550937385, "minimum": 0.49509941524593204, "maximum": 1.3117872316876023, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.0427244757503142, "standard_deviation": 0.004138386313523542, "minimum": 0.03758993044146423, "maximum": 0.050309980028055964, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_8555-292519-0001", "language": "en", "style": "normal", "speaker": "librispeech_8555", "source_utterance_id": "8555-292519-0001", "speaker_gender_as_distributed": "F", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_8555-292519-0001.wav", "source_file_sha256": "8f91b2100fa66d25a159ae2c8113239265f154577f2782c725a6c8d0148a63fa", "decoded_pcm_sha256": "c2e62e46b5a18b9f8f30423189b31cc00703ba64084bb8bca7ffe75ae8d6977a", "original_source_sha256": "b082de228dbad73be95d916e5149c02035758a390482845def17215c0eed0fa9", "corpus_entry_id": "librispeech_8555-292519-0001"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.648446798324585, "predicted_mos": 3.648446798324585, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d2fa49628c1e044eb66043373ed06871614341f1e9ee81f1f1b06bb2e0453f4b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.123368094999023, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_8555-292519-0001", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "d2fa49628c1e044eb66043373ed06871614341f1e9ee81f1f1b06bb2e0453f4b", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "1d71c61f160ff40a72fef1912367b5ac430b341d3c951361c0a870cdcd7e39fe"}, "rows": [{"case": "librispeech_8555-292519-0001", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.690948009490967, "predicted_mos": 2.690948009490967, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c4c67635e9d668d37aad27f2dc4eb50b8c2ff74d897bd0892582d8967e7a873a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.288388605000364, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_8555-292519-0001", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "c4c67635e9d668d37aad27f2dc4eb50b8c2ff74d897bd0892582d8967e7a873a", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6380997717588478, "全帯域_10ms_rms_envelope_nrmse": 0.038158195358670134, "全帯域_10ms_envelope_correlation": 0.9987969089195156, "全帯域_10ms_modulation_mae_db": 0.35033001592836904, "全帯域_40ms_log_rms_mae_db": 0.3636365058693796, "全帯域_40ms_rms_envelope_nrmse": 0.020297214098648726, "全帯域_40ms_envelope_correlation": 0.9996903928055794, "全帯域_40ms_modulation_mae_db": 0.31238733876636965, "300–1000 Hz_10ms_log_rms_mae_db": 0.9636654797191779, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07878543460071187, "300–1000 Hz_10ms_envelope_correlation": 0.9948907695148235, "300–1000 Hz_10ms_modulation_mae_db": 0.9199524011980827, "300–1000 Hz_40ms_log_rms_mae_db": 0.6742644965514509, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.052041405284765654, "300–1000 Hz_40ms_envelope_correlation": 0.997694150661097, "300–1000 Hz_40ms_modulation_mae_db": 1.3648699117718703, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5114070284687158, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12828394334928567, "1000–3000 Hz_10ms_envelope_correlation": 0.9885845416747081, "1000–3000 Hz_10ms_modulation_mae_db": 0.7341196229324091, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9890310322147075, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07906381516371147, "1000–3000 Hz_40ms_envelope_correlation": 0.9957760073986932, "1000–3000 Hz_40ms_modulation_mae_db": 1.2447797586930742, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4165686604900758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1892106407963467, "3000–6000 Hz_10ms_envelope_correlation": 0.979096215899544, "3000–6000 Hz_10ms_modulation_mae_db": 0.5948515468664383, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0474619131972434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1006877108509453, "3000–6000 Hz_40ms_envelope_correlation": 0.9939298790534328, "3000–6000 Hz_40ms_modulation_mae_db": 0.514228690731774, "mr_stft_db": 5.120048656580121, "band_mae_db": 0.11686562340142098, "rms_gain_error_db": -0.08221099090649187, "output_peak": 0.3589682719876802}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6584457194395965, "全帯域_10ms_rms_envelope_nrmse": 0.03527398482862463, "全帯域_10ms_envelope_correlation": 0.9989853870631585, "全帯域_10ms_modulation_mae_db": 0.12165567099093628, "全帯域_40ms_log_rms_mae_db": 0.3513342392383475, "全帯域_40ms_rms_envelope_nrmse": 0.019591998164119814, "全帯域_40ms_envelope_correlation": 0.9997329633188631, "全帯域_40ms_modulation_mae_db": 0.08295738374786879, "300–1000 Hz_10ms_log_rms_mae_db": 0.980424303943328, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07180229730692782, "300–1000 Hz_10ms_envelope_correlation": 0.9957505069448623, "300–1000 Hz_10ms_modulation_mae_db": 0.3648705934146263, "300–1000 Hz_40ms_log_rms_mae_db": 0.6940139651319526, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041761031862332806, "300–1000 Hz_40ms_envelope_correlation": 0.9985224433016007, "300–1000 Hz_40ms_modulation_mae_db": 0.29245106194824705, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5212692040886304, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14166074094958747, "1000–3000 Hz_10ms_envelope_correlation": 0.986457037730832, "1000–3000 Hz_10ms_modulation_mae_db": 0.3719564149964434, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9771569633554472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08320755743228611, "1000–3000 Hz_40ms_envelope_correlation": 0.9959319012878608, "1000–3000 Hz_40ms_modulation_mae_db": 0.6266025020503108, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5011560114798574, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18556300569735903, "3000–6000 Hz_10ms_envelope_correlation": 0.9795239899030787, "3000–6000 Hz_10ms_modulation_mae_db": 0.4417409810793254, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1818530355070018, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11335344923830383, "3000–6000 Hz_40ms_envelope_correlation": 0.9925487436118378, "3000–6000 Hz_40ms_modulation_mae_db": 0.3789453467182375, "mr_stft_db": 5.131246707359104, "band_mae_db": 0.15001811930186557, "rms_gain_error_db": -0.08291803201482957, "output_peak": 0.36255539383604246}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6714261192667004, "全帯域_10ms_rms_envelope_nrmse": 0.03581917871661302, "全帯域_10ms_envelope_correlation": 0.9989576769570285, "全帯域_10ms_modulation_mae_db": 0.3301646620099015, "全帯域_40ms_log_rms_mae_db": 0.38811965650076974, "全帯域_40ms_rms_envelope_nrmse": 0.020383108086488393, "全帯域_40ms_envelope_correlation": 0.9996961120461435, "全帯域_40ms_modulation_mae_db": 0.400875239496703, "300–1000 Hz_10ms_log_rms_mae_db": 0.9526262067331853, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07210445791140208, "300–1000 Hz_10ms_envelope_correlation": 0.9957107065279941, "300–1000 Hz_10ms_modulation_mae_db": 0.9297429255851558, "300–1000 Hz_40ms_log_rms_mae_db": 0.6987697677554294, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044694076817736295, "300–1000 Hz_40ms_envelope_correlation": 0.9983029526172642, "300–1000 Hz_40ms_modulation_mae_db": 1.409188104990296, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5522728046615242, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14942862465345222, "1000–3000 Hz_10ms_envelope_correlation": 0.9842121096378846, "1000–3000 Hz_10ms_modulation_mae_db": 1.128864347466713, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9898963233362266, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0832934992801311, "1000–3000 Hz_40ms_envelope_correlation": 0.9948269894711507, "1000–3000 Hz_40ms_modulation_mae_db": 1.3241215997458975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4513042824062223, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18509212949314016, "3000–6000 Hz_10ms_envelope_correlation": 0.980741348218697, "3000–6000 Hz_10ms_modulation_mae_db": 0.46526316371094584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.053063665295036, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10444420146809177, "3000–6000 Hz_40ms_envelope_correlation": 0.9940077352296863, "3000–6000 Hz_40ms_modulation_mae_db": 0.46731512277755627, "mr_stft_db": 5.134985492865678, "band_mae_db": 0.11518495222401066, "rms_gain_error_db": -0.0805248797102307, "output_peak": 0.36850025107638074}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6586353511721532, "全帯域_10ms_rms_envelope_nrmse": 0.04086518381413708, "全帯域_10ms_envelope_correlation": 0.9986247215792416, "全帯域_10ms_modulation_mae_db": 0.8584134171388403, "全帯域_40ms_log_rms_mae_db": 0.3799352436218043, "全帯域_40ms_rms_envelope_nrmse": 0.022345439204561085, "全帯域_40ms_envelope_correlation": 0.9996373685651099, "全帯域_40ms_modulation_mae_db": 0.8311784556190293, "300–1000 Hz_10ms_log_rms_mae_db": 1.0016701237619126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08623326982008511, "300–1000 Hz_10ms_envelope_correlation": 0.9938721146813545, "300–1000 Hz_10ms_modulation_mae_db": 0.783358268402598, "300–1000 Hz_40ms_log_rms_mae_db": 0.773894810684716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056417139789826154, "300–1000 Hz_40ms_envelope_correlation": 0.9972957377668091, "300–1000 Hz_40ms_modulation_mae_db": 0.7497430624559142, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4864786504782224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1395985337276606, "1000–3000 Hz_10ms_envelope_correlation": 0.9861910958819705, "1000–3000 Hz_10ms_modulation_mae_db": 1.0718151762577854, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9247956287924973, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07237779765542367, "1000–3000 Hz_40ms_envelope_correlation": 0.9962587085298967, "1000–3000 Hz_40ms_modulation_mae_db": 1.6253184039844728, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4120582025667787, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16096674205989878, "3000–6000 Hz_10ms_envelope_correlation": 0.9845016312285176, "3000–6000 Hz_10ms_modulation_mae_db": 1.067783889788007, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0335600235377622, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08903804100367066, "3000–6000 Hz_40ms_envelope_correlation": 0.9951768447974775, "3000–6000 Hz_40ms_modulation_mae_db": 1.257907484673725, "mr_stft_db": 5.12804466451552, "band_mae_db": 0.17317505933294655, "rms_gain_error_db": -0.08512818589416896, "output_peak": 0.3663642327187241}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6767993360849636, "全帯域_10ms_rms_envelope_nrmse": 0.037658702622284, "全帯域_10ms_envelope_correlation": 0.9988005034023544, "全帯域_10ms_modulation_mae_db": 0.41820203727385175, "全帯域_40ms_log_rms_mae_db": 0.38397324511147846, "全帯域_40ms_rms_envelope_nrmse": 0.0214388812769981, "全帯域_40ms_envelope_correlation": 0.9996235668265466, "全帯域_40ms_modulation_mae_db": 0.5372295400677475, "300–1000 Hz_10ms_log_rms_mae_db": 0.9610840711632567, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07722398790423664, "300–1000 Hz_10ms_envelope_correlation": 0.9951297927812812, "300–1000 Hz_10ms_modulation_mae_db": 0.8185389553822245, "300–1000 Hz_40ms_log_rms_mae_db": 0.6893146516231855, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04767403033454231, "300–1000 Hz_40ms_envelope_correlation": 0.9980888025420359, "300–1000 Hz_40ms_modulation_mae_db": 0.9595108189153527, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5267363874734337, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1334067112516823, "1000–3000 Hz_10ms_envelope_correlation": 0.987815077436563, "1000–3000 Hz_10ms_modulation_mae_db": 0.8003551828942563, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9744018864466669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07384766405586163, "1000–3000 Hz_40ms_envelope_correlation": 0.9966528271956482, "1000–3000 Hz_40ms_modulation_mae_db": 1.5860124405782632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4766677920510927, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18722566771188084, "3000–6000 Hz_10ms_envelope_correlation": 0.9795317218200131, "3000–6000 Hz_10ms_modulation_mae_db": 0.8353063907700257, "3000–6000 Hz_40ms_log_rms_mae_db": 1.117035671333962, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1015960059740716, "3000–6000 Hz_40ms_envelope_correlation": 0.9938205843807669, "3000–6000 Hz_40ms_modulation_mae_db": 0.8377474587164034, "mr_stft_db": 5.113900417158821, "band_mae_db": 0.08167142427882539, "rms_gain_error_db": -0.05592026238926259, "output_peak": 0.3620279275442604}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6728975279130378, "全帯域_10ms_rms_envelope_nrmse": 0.03744069308964996, "全帯域_10ms_envelope_correlation": 0.998851422217426, "全帯域_10ms_modulation_mae_db": 0.2421399963644283, "全帯域_40ms_log_rms_mae_db": 0.40177594600401373, "全帯域_40ms_rms_envelope_nrmse": 0.02142799532257964, "全帯域_40ms_envelope_correlation": 0.999659019042059, "全帯域_40ms_modulation_mae_db": 0.16142000266423942, "300–1000 Hz_10ms_log_rms_mae_db": 1.055072738621997, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06908294775437945, "300–1000 Hz_10ms_envelope_correlation": 0.9960736684068557, "300–1000 Hz_10ms_modulation_mae_db": 0.29155742757147024, "300–1000 Hz_40ms_log_rms_mae_db": 0.7568306025829489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044731104962695126, "300–1000 Hz_40ms_envelope_correlation": 0.9983041772667658, "300–1000 Hz_40ms_modulation_mae_db": 0.23279379379550003, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5085111247127734, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14465409690540265, "1000–3000 Hz_10ms_envelope_correlation": 0.9852299456725789, "1000–3000 Hz_10ms_modulation_mae_db": 0.4997082348183383, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9945588191672012, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08062317266600263, "1000–3000 Hz_40ms_envelope_correlation": 0.9954268437574114, "1000–3000 Hz_40ms_modulation_mae_db": 1.0063085019863327, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4491003079416847, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19391048437878372, "3000–6000 Hz_10ms_envelope_correlation": 0.978733549661697, "3000–6000 Hz_10ms_modulation_mae_db": 0.6710776050738446, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0868373801372713, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1387927504255722, "3000–6000 Hz_40ms_envelope_correlation": 0.9900685201592178, "3000–6000 Hz_40ms_modulation_mae_db": 0.45346105479068316, "mr_stft_db": 5.151501238255717, "band_mae_db": 0.17620508466616594, "rms_gain_error_db": -0.08938681081391178, "output_peak": 0.35651726406932277}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6263804589988093, "全帯域_10ms_rms_envelope_nrmse": 0.036772524710864396, "全帯域_10ms_envelope_correlation": 0.998855635858662, "全帯域_10ms_modulation_mae_db": 0.43730419757956424, "全帯域_40ms_log_rms_mae_db": 0.35018769845831643, "全帯域_40ms_rms_envelope_nrmse": 0.019871230041523396, "全帯域_40ms_envelope_correlation": 0.9996745989037372, "全帯域_40ms_modulation_mae_db": 0.564834748481661, "300–1000 Hz_10ms_log_rms_mae_db": 1.0142682696995728, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06916541767408793, "300–1000 Hz_10ms_envelope_correlation": 0.9960621715961874, "300–1000 Hz_10ms_modulation_mae_db": 0.8623064217142064, "300–1000 Hz_40ms_log_rms_mae_db": 0.692370077040053, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043729979647836155, "300–1000 Hz_40ms_envelope_correlation": 0.9983653791560588, "300–1000 Hz_40ms_modulation_mae_db": 0.937705685025028, "1000–3000 Hz_10ms_log_rms_mae_db": 1.435743525779879, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13635532937437522, "1000–3000 Hz_10ms_envelope_correlation": 0.9867894854012308, "1000–3000 Hz_10ms_modulation_mae_db": 0.24894958968695727, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9257261406388897, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07941909098434152, "1000–3000 Hz_40ms_envelope_correlation": 0.9953289526365393, "1000–3000 Hz_40ms_modulation_mae_db": 0.6678693635454295, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3986565726720321, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18816543623072038, "3000–6000 Hz_10ms_envelope_correlation": 0.979885085039872, "3000–6000 Hz_10ms_modulation_mae_db": 0.3874764749145803, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0585310836591972, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12003983598544381, "3000–6000 Hz_40ms_envelope_correlation": 0.9917959461385153, "3000–6000 Hz_40ms_modulation_mae_db": 1.1905465112824494, "mr_stft_db": 5.121213811036032, "band_mae_db": 0.11977426515804533, "rms_gain_error_db": -0.052886078744860274, "output_peak": 0.36443848947695245}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.649097342923667, "全帯域_10ms_rms_envelope_nrmse": 0.03700212465153945, "全帯域_10ms_envelope_correlation": 0.9988746920217679, "全帯域_10ms_modulation_mae_db": 0.5573815753911547, "全帯域_40ms_log_rms_mae_db": 0.3604219337961768, "全帯域_40ms_rms_envelope_nrmse": 0.019009854984258102, "全帯域_40ms_envelope_correlation": 0.9997432649033665, "全帯域_40ms_modulation_mae_db": 0.5588001819532075, "300–1000 Hz_10ms_log_rms_mae_db": 0.9908671233179059, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0830913142618963, "300–1000 Hz_10ms_envelope_correlation": 0.994281317512998, "300–1000 Hz_10ms_modulation_mae_db": 0.3785792146219203, "300–1000 Hz_40ms_log_rms_mae_db": 0.7137984588529013, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.048670512644630466, "300–1000 Hz_40ms_envelope_correlation": 0.9979784124427608, "300–1000 Hz_40ms_modulation_mae_db": 0.5740167957013624, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4708992954973108, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1252062338557622, "1000–3000 Hz_10ms_envelope_correlation": 0.9889018870238442, "1000–3000 Hz_10ms_modulation_mae_db": 0.5013573076392518, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9326962063275619, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07208279370456822, "1000–3000 Hz_40ms_envelope_correlation": 0.9961153081479834, "1000–3000 Hz_40ms_modulation_mae_db": 1.0723054249036565, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4209313693018517, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18477678975177392, "3000–6000 Hz_10ms_envelope_correlation": 0.9800181383597263, "3000–6000 Hz_10ms_modulation_mae_db": 0.6056372766106768, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0315130885430266, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0950104275300652, "3000–6000 Hz_40ms_envelope_correlation": 0.9945799453566277, "3000–6000 Hz_40ms_modulation_mae_db": 0.8779876042222767, "mr_stft_db": 5.138390438116637, "band_mae_db": 0.09185997739647829, "rms_gain_error_db": -0.0761289833813522, "output_peak": 0.3580687842890516}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.290727383258726, "spectral_convergence": 0.11792989063010173, "spectral_convergence_status": "ok", "amplitude_floor": 1.7050603882585003e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.095101695350676, "spectral_convergence": 0.11285494397808388, "spectral_convergence_status": "ok", "amplitude_floor": 1.5194442893884604e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.974316891130962, "spectral_convergence": 0.11321050101780726, "spectral_convergence_status": "ok", "amplitude_floor": 1.5327487413055003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00047003850181382925, "output_power": 0.00046780887695682774, "delta_db": -0.020649747887050016, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.004513086919531444, "output_power": 0.0044226942828750684, "delta_db": -0.08786777909770133, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003251325154635926, "output_power": 0.0003166453532030949, "delta_db": -0.1148728496851807, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.51153215194187e-05, "output_power": 8.125675379752698e-05, "delta_db": -0.20148275715898886, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6288307106843834e-05, "output_power": 2.6128687564772527e-05, "delta_db": -0.026450238529961406, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.4880972570108e-05, "output_power": 4.753891838078641e-05, "delta_db": 0.24987036804964358, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11686562340142098, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08221099090649187, "pitch": null, "output_rms_dbfs": -22.706672880722103, "output_peak": 0.3589682719876802, "reference_rms_dbfs": -22.624461889815606, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.120048656580121}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023914198234423542, "log_rms_mae_db": 0.6380997717588478, "rms_envelope_nrmse": 0.038158195358670134, "envelope_correlation": 0.9987969089195156, "rms_gain_error_db": -0.07710754923321897, "reference_mean_power": 0.005456771858924342, "output_mean_power": 0.005360743675766581}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.071388012833875e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045567084702756786, "output_power": 0.00044667450532916576, "reference_active": true, "delta_db": -0.08660079000574017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.44246106516196e-05, "output_power": 6.383577077871787e-05, "reference_active": true, "delta_db": -0.03987695987991382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7099940640562705e-05, "output_power": 1.382112430698243e-05, "reference_active": true, "delta_db": -0.924512297899453}], "active_band_count": 3, "active_band_mae_db": 0.35033001592836904}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022747909354312384, "log_rms_mae_db": 0.3636365058693796, "rms_envelope_nrmse": 0.020297214098648726, "envelope_correlation": 0.9996903928055794, "rms_gain_error_db": -0.07157718244614403, "reference_mean_power": 0.005416920568028677, "output_mean_power": 0.005328374595749897}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.96418119219355e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000410225191407457, "output_power": 0.00040259786703203236, "reference_active": true, "delta_db": -0.08150856661489858}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7757195967845804e-05, "output_power": 2.7775859846122516e-05, "reference_active": true, "delta_db": 0.0029192053391877614}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0255605769228834e-06, "output_power": 8.427289115779703e-07, "reference_active": true, "delta_db": -0.8527342443450227}], "active_band_count": 3, "active_band_mae_db": 0.31238733876636965}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.442623236865579e-05, "log_rms_mae_db": 0.9636654797191779, "rms_envelope_nrmse": 0.07878543460071187, "envelope_correlation": 0.9948907695148235, "rms_gain_error_db": -0.033209447497915816, "reference_mean_power": 0.0007990709247977912, "output_mean_power": 0.0007929839254758402}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.944736222458957e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022092122499658314, "output_power": 0.00021461391828181214, "reference_active": true, "delta_db": -0.12579559006119231}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3665816077667355e-05, "output_power": 2.1536943129840144e-05, "reference_active": true, "delta_db": -0.40937423747934326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5979447501439136e-06, "output_power": 3.353975379066058e-06, "reference_active": true, "delta_db": -2.224687376053713}], "active_band_count": 3, "active_band_mae_db": 0.9199524011980827}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.226666373654229e-05, "log_rms_mae_db": 0.6742644965514509, "rms_envelope_nrmse": 0.052041405284765654, "envelope_correlation": 0.997694150661097, "rms_gain_error_db": -0.03502565673397357, "reference_mean_power": 0.0008012273487091395, "output_mean_power": 0.0007947914732000735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.73006991688152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002092887788100005, "output_power": 0.00020219982342120413, "reference_active": true, "delta_db": -0.1496517191632934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0995901908054441e-05, "output_power": 9.96319509204422e-06, "reference_active": true, "delta_db": -0.42832222506022927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1412698290887665e-07, "output_power": 1.8427649755680609e-07, "reference_active": true, "delta_db": -3.5166357910920882}], "active_band_count": 3, "active_band_mae_db": 1.3648699117718703}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9464514079665034e-05, "log_rms_mae_db": 1.5114070284687158, "rms_envelope_nrmse": 0.12828394334928567, "envelope_correlation": 0.9885845416747081, "rms_gain_error_db": -0.24062515019295427, "reference_mean_power": 7.576176061225114e-05, "output_mean_power": 7.167827474737366e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.472428410826198e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.044189308875811e-05, "output_power": 2.7942649178371888e-05, "reference_active": true, "delta_db": -0.37204078199300117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3472076435688925e-06, "output_power": 2.870365434050227e-06, "reference_active": true, "delta_db": -0.6674546312367899}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0111506850143103e-06, "output_power": 7.736232884112874e-07, "reference_active": true, "delta_db": -1.1628634555674364}], "active_band_count": 3, "active_band_mae_db": 0.7341196229324091}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.608084029396227e-05, "log_rms_mae_db": 0.9890310322147075, "rms_envelope_nrmse": 0.07906381516371147, "envelope_correlation": 0.9957760073986932, "rms_gain_error_db": -0.2416619170143405, "reference_mean_power": 7.611087404608628e-05, "output_mean_power": 7.199138315361753e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.164443748057941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.887302600627445e-05, "output_power": 2.6519748313501088e-05, "reference_active": true, "delta_db": -0.36922903962678455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.617498060844069e-06, "output_power": 1.3204481281447209e-06, "reference_active": true, "delta_db": -0.8812242342169436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.602390292027545e-08, "output_power": 4.855461808621597e-08, "reference_active": true, "delta_db": -2.4838860022354945}], "active_band_count": 3, "active_band_mae_db": 1.2447797586930742}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.198647979957052e-05, "log_rms_mae_db": 1.4165686604900758, "rms_envelope_nrmse": 0.1892106407963467, "envelope_correlation": 0.979096215899544, "rms_gain_error_db": 0.11748614588677321, "reference_mean_power": 5.064740774931045e-05, "output_mean_power": 5.203623144534563e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.700608322524163e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.136344717386399e-05, "output_power": 4.2376519871034204e-05, "reference_active": true, "delta_db": 0.10508562732655327}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.648317567219929e-06, "output_power": 6.588240383214598e-06, "reference_active": true, "delta_db": -0.039423192588216714}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.888415633999808e-06, "output_power": 2.6654350145676003e-06, "reference_active": true, "delta_db": -1.640045820684545}], "active_band_count": 3, "active_band_mae_db": 0.5948515468664383}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.0010568231240365e-05, "log_rms_mae_db": 1.0474619131972434, "rms_envelope_nrmse": 0.1006877108509453, "envelope_correlation": 0.9939298790534328, "rms_gain_error_db": 0.11699744398909008, "reference_mean_power": 5.089623620868825e-05, "output_mean_power": 5.2285999175748404e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.379392951970345e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.821321246572876e-05, "output_power": 3.9096889805570764e-05, "reference_active": true, "delta_db": 0.09928661227370461}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.5895785128125065e-06, "output_power": 3.4043809855749053e-06, "reference_active": true, "delta_db": -0.2300530065614838}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2481135158983729e-07, "output_power": 1.7001357172263534e-07, "reference_active": true, "delta_db": -1.2133464533601337}], "active_band_count": 3, "active_band_mae_db": 0.514228690731774}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9556502979049213, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "efe852b9e9e76771e77abbbdab4b2a500b80ec55ed5baf0c3c357dcfc5b23bcf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9867726475574515, "standard_deviation": 0.0016158880804721433, "minimum": 0.9842121096378846, "maximum": 0.9889018870238442, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5016647526450613, "standard_deviation": 0.03632909159126926, "minimum": 1.435743525779879, "maximum": 1.5522728046615242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6696407345865194, "standard_deviation": 0.3199704812251862, "minimum": 0.24894958968695727, "maximum": 1.128864347466713, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13732427675840103, "standard_deviation": 0.008186045951723443, "minimum": 0.1252062338557622, "maximum": 0.14942862465345222, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.995789692303148, "standard_deviation": 0.00058090462263873, "minimum": 0.9948269894711507, "maximum": 0.9966528271956482, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9635328750348998, "standard_deviation": 0.03045458299064805, "minimum": 0.9247956287924973, "maximum": 0.9945588191672012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1441647494359297, "standard_deviation": 0.375716867475094, "minimum": 0.6266025020503108, "maximum": 1.6253184039844728, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0779894238677908, "standard_deviation": 0.00461410623293725, "minimum": 0.07208279370456822, "maximum": 0.0832934992801311, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9802539600163933, "standard_deviation": 0.0018196719505575462, "minimum": 0.978733549661697, "maximum": 0.9845016312285176, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4408053998636996, "standard_deviation": 0.035239306671487496, "minimum": 1.3986565726720321, "maximum": 1.5011560114798574, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6336421661017305, "standard_deviation": 0.22634139656729071, "minimum": 0.3874764749145803, "maximum": 1.067783889788007, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18436386201498794, "standard_deviation": 0.00990551947203599, "minimum": 0.16096674205989878, "maximum": 0.19391048437878372, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932410248409452, "standard_deviation": 0.0016720668548605014, "minimum": 0.9900685201592178, "maximum": 0.9951768447974775, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0762319826513127, "standard_deviation": 0.05131375130061398, "minimum": 1.0315130885430266, "maximum": 1.1818530355070018, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7472674092391383, "standard_deviation": 0.3458297889240078, "minimum": 0.3789453467182375, "maximum": 1.257907484673725, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10787030280952054, "standard_deviation": 0.015843363975397706, "minimum": 0.08903804100367066, "maximum": 0.1387927504255722, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9952213809957946, "standard_deviation": 0.0008258906623033834, "minimum": 0.9938721146813545, "maximum": 0.9960736684068557, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9899597896200419, "standard_deviation": 0.03376065600979192, "minimum": 0.9526262067331853, "maximum": 1.055072738621997, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6686132759862855, "standard_deviation": 0.2733578747675298, "minimum": 0.29155742757147024, "maximum": 0.9297429255851558, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.0759361409042159, "standard_deviation": 0.006454055214747452, "minimum": 0.06908294775437945, "maximum": 0.08623326982008511, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.998069006969299, "standard_deviation": 0.00040541660403753377, "minimum": 0.9972957377668091, "maximum": 0.9985224433016007, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7116571037778298, "standard_deviation": 0.0351842253041362, "minimum": 0.6742644965514509, "maximum": 0.773894810684716, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8150349043254463, "standard_deviation": 0.4417027617256134, "minimum": 0.23279379379550003, "maximum": 1.409188104990296, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04746491016804562, "standard_deviation": 0.004843538750647853, "minimum": 0.041761031862332806, "maximum": 0.056417139789826154, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12809431321996984, "standard_deviation": 0.03516805217949271, "minimum": 0.08167142427882539, "maximum": 0.17620508466616594, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.129916428235954, "standard_deviation": 0.011942137279643793, "minimum": 5.113900417158821, "maximum": 5.151501238255717, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36218007687480186, "standard_deviation": 0.004174789438911775, "minimum": 0.35651726406932277, "maximum": 0.36850025107638074, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0756380279818885, "standard_deviation": 0.013658373977470594, "minimum": -0.08938681081391178, "maximum": -0.052886078744860274, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9988433685023943, "standard_deviation": 0.00011099201661817938, "minimum": 0.9986247215792416, "maximum": 0.9989853870631585, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.656472703444722, "standard_deviation": 0.01777890030627077, "minimum": 0.6263804589988093, "maximum": 0.6767993360849636, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41444894658463083, "standard_deviation": 0.22190860091920594, "minimum": 0.12165567099093628, "maximum": 0.8584134171388403, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.037373823474047836, "standard_deviation": 0.0016994296029725854, "minimum": 0.03527398482862463, "maximum": 0.04086518381413708, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9996821608014257, "standard_deviation": 4.244211490467161e-05, "minimum": 0.9996235668265466, "maximum": 0.9997432649033665, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3724230585750358, "standard_deviation": 0.018739549335207033, "minimum": 0.35018769845831643, "maximum": 0.40177594600401373, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4312103613496032, "standard_deviation": 0.24332673102211522, "minimum": 0.08295738374786879, "maximum": 0.8311784556190293, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.020545715147397156, "standard_deviation": 0.0011101310884602841, "minimum": 0.019009854984258102, "maximum": 0.022345439204561085, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-292519-0001", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.740260362625122, "predicted_mos": 2.740260362625122, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4ac0c6bbaad778e5918649e4fa84313dcd4e387f705c137b4ac01e4e13f6aa02", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7716559004038572, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6417292573710137, "全帯域_10ms_rms_envelope_nrmse": 0.04020254801854928, "全帯域_10ms_envelope_correlation": 0.9986611340651491, "全帯域_10ms_modulation_mae_db": 0.3481488330748011, "全帯域_40ms_log_rms_mae_db": 0.36803393153001895, "全帯域_40ms_rms_envelope_nrmse": 0.020459454855384446, "全帯域_40ms_envelope_correlation": 0.9996860900813257, "全帯域_40ms_modulation_mae_db": 0.26189616972747004, "300–1000 Hz_10ms_log_rms_mae_db": 0.9294839976387437, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07975613203479362, "300–1000 Hz_10ms_envelope_correlation": 0.9947620721370536, "300–1000 Hz_10ms_modulation_mae_db": 0.8412278427131598, "300–1000 Hz_40ms_log_rms_mae_db": 0.6386414877884478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05276137647282908, "300–1000 Hz_40ms_envelope_correlation": 0.9976252916979167, "300–1000 Hz_40ms_modulation_mae_db": 1.2130514336808906, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4856036939834738, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12754994163596678, "1000–3000 Hz_10ms_envelope_correlation": 0.9889338893950596, "1000–3000 Hz_10ms_modulation_mae_db": 0.6700160786151844, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0055059415648302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0849702076582636, "1000–3000 Hz_40ms_envelope_correlation": 0.9953090381460024, "1000–3000 Hz_40ms_modulation_mae_db": 1.0908981557048163, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4648334918686259, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18595535016594805, "3000–6000 Hz_10ms_envelope_correlation": 0.9798467319628689, "3000–6000 Hz_10ms_modulation_mae_db": 0.6235333999775013, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1244866537050506, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10082510085114683, "3000–6000 Hz_40ms_envelope_correlation": 0.9939389108301365, "3000–6000 Hz_40ms_modulation_mae_db": 0.3573024049023199, "mr_stft_db": 5.060971611079032, "band_mae_db": 0.12675663681203878, "rms_gain_error_db": -0.08831541570684864, "output_peak": 0.3612008418803692}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6622992158209268, "全帯域_10ms_rms_envelope_nrmse": 0.03733526323668902, "全帯域_10ms_envelope_correlation": 0.9988633051443419, "全帯域_10ms_modulation_mae_db": 0.08901606071272035, "全帯域_40ms_log_rms_mae_db": 0.36509372460039136, "全帯域_40ms_rms_envelope_nrmse": 0.02053631591431116, "全帯域_40ms_envelope_correlation": 0.9997012845829607, "全帯域_40ms_modulation_mae_db": 0.12369154032458969, "300–1000 Hz_10ms_log_rms_mae_db": 0.9152838973938019, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07772968927520257, "300–1000 Hz_10ms_envelope_correlation": 0.995003478015633, "300–1000 Hz_10ms_modulation_mae_db": 0.43246001413705876, "300–1000 Hz_40ms_log_rms_mae_db": 0.61134182138229, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043964989440259644, "300–1000 Hz_40ms_envelope_correlation": 0.9983558953815189, "300–1000 Hz_40ms_modulation_mae_db": 0.3655894861979592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4803750272671434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1375655870072273, "1000–3000 Hz_10ms_envelope_correlation": 0.9873669697879154, "1000–3000 Hz_10ms_modulation_mae_db": 0.18714589179162142, "1000–3000 Hz_40ms_log_rms_mae_db": 0.949069939486143, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08204423383877162, "1000–3000 Hz_40ms_envelope_correlation": 0.9961677231902969, "1000–3000 Hz_40ms_modulation_mae_db": 0.4363245839125321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5425350012973083, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19311259407333337, "3000–6000 Hz_10ms_envelope_correlation": 0.977612712100971, "3000–6000 Hz_10ms_modulation_mae_db": 0.477692437887485, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2488260809755567, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12134878442267491, "3000–6000 Hz_40ms_envelope_correlation": 0.9908955117941618, "3000–6000 Hz_40ms_modulation_mae_db": 0.44068580822216985, "mr_stft_db": 5.067159214524707, "band_mae_db": 0.1757029029068926, "rms_gain_error_db": -0.08900905561716287, "output_peak": 0.3595704029059677}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6777263608518735, "全帯域_10ms_rms_envelope_nrmse": 0.036161071886400595, "全帯域_10ms_envelope_correlation": 0.9989292598945843, "全帯域_10ms_modulation_mae_db": 0.382987241599219, "全帯域_40ms_log_rms_mae_db": 0.3961748126247799, "全帯域_40ms_rms_envelope_nrmse": 0.02090833566961689, "全帯域_40ms_envelope_correlation": 0.999672871753168, "全帯域_40ms_modulation_mae_db": 0.4579976318562185, "300–1000 Hz_10ms_log_rms_mae_db": 0.9173029845576455, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.074897072735937, "300–1000 Hz_10ms_envelope_correlation": 0.9953755270663264, "300–1000 Hz_10ms_modulation_mae_db": 0.9832280519569551, "300–1000 Hz_40ms_log_rms_mae_db": 0.6487848283951667, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0455988736185451, "300–1000 Hz_40ms_envelope_correlation": 0.9982426143932124, "300–1000 Hz_40ms_modulation_mae_db": 1.4861747614038798, "1000–3000 Hz_10ms_log_rms_mae_db": 1.523520376784832, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14623937086800604, "1000–3000 Hz_10ms_envelope_correlation": 0.9849188696136706, "1000–3000 Hz_10ms_modulation_mae_db": 1.0583024006595743, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9973823325728094, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08583184127367965, "1000–3000 Hz_40ms_envelope_correlation": 0.9945502926196146, "1000–3000 Hz_40ms_modulation_mae_db": 1.092410179527109, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4826940346052804, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18607388190299623, "3000–6000 Hz_10ms_envelope_correlation": 0.9805286735636835, "3000–6000 Hz_10ms_modulation_mae_db": 0.5210204283518521, "3000–6000 Hz_40ms_log_rms_mae_db": 1.119129667410393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10784330729462885, "3000–6000 Hz_40ms_envelope_correlation": 0.9936007616202184, "3000–6000 Hz_40ms_modulation_mae_db": 0.5023000647427328, "mr_stft_db": 5.068537864479242, "band_mae_db": 0.10313038883333954, "rms_gain_error_db": -0.07640927694730495, "output_peak": 0.37069910245220383}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6423243887249018, "全帯域_10ms_rms_envelope_nrmse": 0.040384201525319866, "全帯域_10ms_envelope_correlation": 0.9986621628758822, "全帯域_10ms_modulation_mae_db": 0.8549494856777536, "全帯域_40ms_log_rms_mae_db": 0.36944746205461326, "全帯域_40ms_rms_envelope_nrmse": 0.021855090108733228, "全帯域_40ms_envelope_correlation": 0.9996534576356887, "全帯域_40ms_modulation_mae_db": 0.7839221274657273, "300–1000 Hz_10ms_log_rms_mae_db": 0.9342828439238902, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0857151330366288, "300–1000 Hz_10ms_envelope_correlation": 0.9939426969903374, "300–1000 Hz_10ms_modulation_mae_db": 0.9008231680720566, "300–1000 Hz_40ms_log_rms_mae_db": 0.6685878711863424, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05488355402368404, "300–1000 Hz_40ms_envelope_correlation": 0.9974364169200752, "300–1000 Hz_40ms_modulation_mae_db": 0.8413662536266289, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4647317295256148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13348017937007464, "1000–3000 Hz_10ms_envelope_correlation": 0.9874004070150292, "1000–3000 Hz_10ms_modulation_mae_db": 0.9189793873157676, "1000–3000 Hz_40ms_log_rms_mae_db": 0.925812109812361, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06747166200581742, "1000–3000 Hz_40ms_envelope_correlation": 0.9967753507634807, "1000–3000 Hz_40ms_modulation_mae_db": 1.1478781874979582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4208431975542979, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1625144044337614, "3000–6000 Hz_10ms_envelope_correlation": 0.9841928764807685, "3000–6000 Hz_10ms_modulation_mae_db": 0.9949204268585866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0917247454551497, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09350967236880002, "3000–6000 Hz_40ms_envelope_correlation": 0.9946291850374616, "3000–6000 Hz_40ms_modulation_mae_db": 1.1375766263633837, "mr_stft_db": 5.061658263595331, "band_mae_db": 0.18120665564160088, "rms_gain_error_db": -0.08750489591950686, "output_peak": 0.3687369445842839}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6820789613622888, "全帯域_10ms_rms_envelope_nrmse": 0.039484801376838755, "全帯域_10ms_envelope_correlation": 0.9986823387782404, "全帯域_10ms_modulation_mae_db": 0.4292729354671254, "全帯域_40ms_log_rms_mae_db": 0.3948568360153581, "全帯域_40ms_rms_envelope_nrmse": 0.02312370999518375, "全帯域_40ms_envelope_correlation": 0.9995598996188645, "全帯域_40ms_modulation_mae_db": 0.5661085374226879, "300–1000 Hz_10ms_log_rms_mae_db": 0.8835093491695841, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08536063960165828, "300–1000 Hz_10ms_envelope_correlation": 0.9940593301627771, "300–1000 Hz_10ms_modulation_mae_db": 0.9355343245877897, "300–1000 Hz_40ms_log_rms_mae_db": 0.6152202364826573, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05430430983129201, "300–1000 Hz_40ms_envelope_correlation": 0.9975305777845835, "300–1000 Hz_40ms_modulation_mae_db": 1.1508910688159018, "1000–3000 Hz_10ms_log_rms_mae_db": 1.473625457497078, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12974785221440632, "1000–3000 Hz_10ms_envelope_correlation": 0.9888234576298512, "1000–3000 Hz_10ms_modulation_mae_db": 0.6823331602279533, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9496399250544582, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07352440367289853, "1000–3000 Hz_40ms_envelope_correlation": 0.9970846869647486, "1000–3000 Hz_40ms_modulation_mae_db": 1.2976739956876153, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5099857549351403, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18443703196045494, "3000–6000 Hz_10ms_envelope_correlation": 0.9801509807498032, "3000–6000 Hz_10ms_modulation_mae_db": 0.7946510025783975, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1815552177279822, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10016803935679809, "3000–6000 Hz_40ms_envelope_correlation": 0.9940138774240534, "3000–6000 Hz_40ms_modulation_mae_db": 0.8567974368645471, "mr_stft_db": 5.053429966657586, "band_mae_db": 0.0834406998493561, "rms_gain_error_db": -0.060636909942834855, "output_peak": 0.3618295926401342}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.683914077044983, "全帯域_10ms_rms_envelope_nrmse": 0.03888751402280745, "全帯域_10ms_envelope_correlation": 0.9987535935551477, "全帯域_10ms_modulation_mae_db": 0.16602711363034614, "全帯域_40ms_log_rms_mae_db": 0.4163589601037868, "全帯域_40ms_rms_envelope_nrmse": 0.021492157515186416, "全帯域_40ms_envelope_correlation": 0.9996520578858099, "全帯域_40ms_modulation_mae_db": 0.2577817638273711, "300–1000 Hz_10ms_log_rms_mae_db": 0.9478725074749214, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0668299998336036, "300–1000 Hz_10ms_envelope_correlation": 0.9963292709055297, "300–1000 Hz_10ms_modulation_mae_db": 0.3889764166492586, "300–1000 Hz_40ms_log_rms_mae_db": 0.6443628013000253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041585946013697264, "300–1000 Hz_40ms_envelope_correlation": 0.9985310456593955, "300–1000 Hz_40ms_modulation_mae_db": 0.38001428344292165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4761670913565257, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1346390838936509, "1000–3000 Hz_10ms_envelope_correlation": 0.9874200982477722, "1000–3000 Hz_10ms_modulation_mae_db": 0.3808263359897519, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9473922951419452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07808852658121158, "1000–3000 Hz_40ms_envelope_correlation": 0.9959684670233709, "1000–3000 Hz_40ms_modulation_mae_db": 0.7094617655067874, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4793631353197745, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1902558728700309, "3000–6000 Hz_10ms_envelope_correlation": 0.979134093623969, "3000–6000 Hz_10ms_modulation_mae_db": 0.7571129515791934, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1517781030200855, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13430851690646736, "3000–6000 Hz_40ms_envelope_correlation": 0.9902411936527122, "3000–6000 Hz_40ms_modulation_mae_db": 0.4926907473725655, "mr_stft_db": 5.088538220477833, "band_mae_db": 0.16648679787716383, "rms_gain_error_db": -0.09015502463968533, "output_peak": 0.35866702162037595}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6326596357048605, "全帯域_10ms_rms_envelope_nrmse": 0.03854045331910187, "全帯域_10ms_envelope_correlation": 0.9987399968194275, "全帯域_10ms_modulation_mae_db": 0.4116994647109718, "全帯域_40ms_log_rms_mae_db": 0.35842270642231727, "全帯域_40ms_rms_envelope_nrmse": 0.021093469467670108, "全帯域_40ms_envelope_correlation": 0.9996269356443475, "全帯域_40ms_modulation_mae_db": 0.4743329945342564, "300–1000 Hz_10ms_log_rms_mae_db": 0.9252024948259528, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0689151779871333, "300–1000 Hz_10ms_envelope_correlation": 0.9960987595215096, "300–1000 Hz_10ms_modulation_mae_db": 0.7557947333820159, "300–1000 Hz_40ms_log_rms_mae_db": 0.6188318207484007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04258750812705453, "300–1000 Hz_40ms_envelope_correlation": 0.9984538158811972, "300–1000 Hz_40ms_modulation_mae_db": 0.832445476870043, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4419759787723874, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12847811734928294, "1000–3000 Hz_10ms_envelope_correlation": 0.9883141536549243, "1000–3000 Hz_10ms_modulation_mae_db": 0.2372515543085648, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9430028421352306, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07759144678880184, "1000–3000 Hz_40ms_envelope_correlation": 0.9955954756344426, "1000–3000 Hz_40ms_modulation_mae_db": 0.37957228400416576, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4456874706507714, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1965909073554423, "3000–6000 Hz_10ms_envelope_correlation": 0.978699921872651, "3000–6000 Hz_10ms_modulation_mae_db": 0.28956282307865217, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1251917881530642, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12999154878821295, "3000–6000 Hz_40ms_envelope_correlation": 0.9908934540334081, "3000–6000 Hz_40ms_modulation_mae_db": 0.9988319248394276, "mr_stft_db": 5.063754570384847, "band_mae_db": 0.12694837835585437, "rms_gain_error_db": -0.05052770239331327, "output_peak": 0.36356596374953654}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6569544457474245, "全帯域_10ms_rms_envelope_nrmse": 0.036785792386791145, "全帯域_10ms_envelope_correlation": 0.998889959813265, "全帯域_10ms_modulation_mae_db": 0.44786716707120283, "全帯域_40ms_log_rms_mae_db": 0.3652331248239588, "全帯域_40ms_rms_envelope_nrmse": 0.01894188028741663, "全帯域_40ms_envelope_correlation": 0.9997460368029877, "全帯域_40ms_modulation_mae_db": 0.49068172375112934, "300–1000 Hz_10ms_log_rms_mae_db": 0.9371241447459722, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0807621803583277, "300–1000 Hz_10ms_envelope_correlation": 0.9945975650772727, "300–1000 Hz_10ms_modulation_mae_db": 0.26961888372463366, "300–1000 Hz_40ms_log_rms_mae_db": 0.6580861512531327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04748783009437099, "300–1000 Hz_40ms_envelope_correlation": 0.9980685109381503, "300–1000 Hz_40ms_modulation_mae_db": 0.462408519387752, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4463179313679462, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11814542950832906, "1000–3000 Hz_10ms_envelope_correlation": 0.9901011073522704, "1000–3000 Hz_10ms_modulation_mae_db": 0.33886421039680403, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9223242173149035, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06628051671046307, "1000–3000 Hz_40ms_envelope_correlation": 0.9967404094473732, "1000–3000 Hz_40ms_modulation_mae_db": 0.7511238230857941, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4785526662823416, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18707678583477638, "3000–6000 Hz_10ms_envelope_correlation": 0.979762239507548, "3000–6000 Hz_10ms_modulation_mae_db": 0.41747502544001486, "3000–6000 Hz_40ms_log_rms_mae_db": 1.132230802098858, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09358460841680108, "3000–6000 Hz_40ms_envelope_correlation": 0.994910007304576, "3000–6000 Hz_40ms_modulation_mae_db": 0.7040956833063693, "mr_stft_db": 5.069698400857203, "band_mae_db": 0.10290453513431348, "rms_gain_error_db": -0.07627921623901829, "output_peak": 0.35737368290468446}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.254721485985054, "spectral_convergence": 0.1215231657364792, "spectral_convergence_status": "ok", "amplitude_floor": 1.7050603882585003e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.0317429564796905, "spectral_convergence": 0.11579341364078888, "spectral_convergence_status": "ok", "amplitude_floor": 1.5194442893884604e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.896450390772353, "spectral_convergence": 0.11445424018309354, "spectral_convergence_status": "ok", "amplitude_floor": 1.5327487413055003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00047003850181382925, "output_power": 0.0004660133130586106, "delta_db": -0.0373510955798132, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.004513086919531444, "output_power": 0.0044177947261416104, "delta_db": -0.09268165459441709, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003251325154635926, "output_power": 0.0003171943675842471, "delta_db": -0.1073493711770036, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.51153215194187e-05, "output_power": 8.055749570712728e-05, "delta_db": -0.2390178718212371, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6288307106843834e-05, "output_power": 2.585521958768627e-05, "delta_db": -0.07214389086628481, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.4880972570108e-05, "output_power": 4.71261382738821e-05, "delta_db": 0.21199593683347687, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.12675663681203878, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08831541570684864, "pitch": null, "output_rms_dbfs": -22.712777305522458, "output_peak": 0.3612008418803692, "reference_rms_dbfs": -22.624461889815606, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.060971611079032}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023914198234423542, "log_rms_mae_db": 0.6417292573710137, "rms_envelope_nrmse": 0.04020254801854928, "envelope_correlation": 0.9986611340651491, "rms_gain_error_db": -0.08038066074872832, "reference_mean_power": 0.005456771858924342, "output_mean_power": 0.005356705010239808}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.071388012833875e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045567084702756786, "output_power": 0.0004475328895869961, "reference_active": true, "delta_db": -0.0782628660933512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.44246106516196e-05, "output_power": 6.357119447453158e-05, "reference_active": true, "delta_db": -0.057914308762979934}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7099940640562705e-05, "output_power": 1.387291324436298e-05, "reference_active": true, "delta_db": -0.9082693243680722}], "active_band_count": 3, "active_band_mae_db": 0.3481488330748011}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022747909354312384, "log_rms_mae_db": 0.36803393153001895, "rms_envelope_nrmse": 0.020459454855384446, "envelope_correlation": 0.9996860900813257, "rms_gain_error_db": -0.07276271248699312, "reference_mean_power": 0.005416920568028677, "output_mean_power": 0.005326920263194763}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.96418119219355e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000410225191407457, "output_power": 0.00040372198633616753, "reference_active": true, "delta_db": -0.06939924982112382}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7757195967845804e-05, "output_power": 2.7746332921269354e-05, "reference_active": true, "delta_db": -0.0016999860511479523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0255605769228834e-06, "output_power": 8.699662219552553e-07, "reference_active": true, "delta_db": -0.7145892733101383}], "active_band_count": 3, "active_band_mae_db": 0.26189616972747004}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.442623236865579e-05, "log_rms_mae_db": 0.9294839976387437, "rms_envelope_nrmse": 0.07975613203479362, "envelope_correlation": 0.9947620721370536, "rms_gain_error_db": -0.023599353155950333, "reference_mean_power": 0.0007990709247977912, "output_mean_power": 0.000794740587921786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.944736222458957e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022092122499658314, "output_power": 0.0002161263776737981, "reference_active": true, "delta_db": -0.09529668045236256}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3665816077667355e-05, "output_power": 2.1827854155708206e-05, "reference_active": true, "delta_db": -0.3511044165740295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5979447501439136e-06, "output_power": 3.469767419015309e-06, "reference_active": true, "delta_db": -2.0772824311130873}], "active_band_count": 3, "active_band_mae_db": 0.8412278427131598}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.226666373654229e-05, "log_rms_mae_db": 0.6386414877884478, "rms_envelope_nrmse": 0.05276137647282908, "envelope_correlation": 0.9976252916979167, "rms_gain_error_db": -0.02363697833710711, "reference_mean_power": 0.0008012273487091395, "output_mean_power": 0.0007968784219084872}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.73006991688152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002092887788100005, "output_power": 0.00020371810431781207, "reference_active": true, "delta_db": -0.11716317675969115}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0995901908054441e-05, "output_power": 1.0118658708075737e-05, "reference_active": true, "delta_db": -0.36107909095348084}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1412698290887665e-07, "output_power": 2.0000567796636032e-07, "reference_active": true, "delta_db": -3.1609120333294998}], "active_band_count": 3, "active_band_mae_db": 1.2130514336808906}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9464514079665034e-05, "log_rms_mae_db": 1.4856036939834738, "rms_envelope_nrmse": 0.12754994163596678, "envelope_correlation": 0.9889338893950596, "rms_gain_error_db": -0.29944857936145, "reference_mean_power": 7.576176061225114e-05, "output_mean_power": 7.071396685982687e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.472428410826198e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.044189308875811e-05, "output_power": 2.7656161065448167e-05, "reference_active": true, "delta_db": -0.41679760582043346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3472076435688925e-06, "output_power": 2.938624632046435e-06, "reference_active": true, "delta_db": -0.5653853998695484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0111506850143103e-06, "output_power": 7.980486318586277e-07, "reference_active": true, "delta_db": -1.0278652301555715}], "active_band_count": 3, "active_band_mae_db": 0.6700160786151844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.608084029396227e-05, "log_rms_mae_db": 1.0055059415648302, "rms_envelope_nrmse": 0.0849702076582636, "envelope_correlation": 0.9953090381460024, "rms_gain_error_db": -0.2989400550781031, "reference_mean_power": 7.611087404608628e-05, "output_mean_power": 7.104813852880305e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.164443748057941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.887302600627445e-05, "output_power": 2.6268523926848175e-05, "reference_active": true, "delta_db": -0.41056632305436436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.617498060844069e-06, "output_power": 1.3649124446213595e-06, "reference_active": true, "delta_db": -0.737389752082372}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.602390292027545e-08, "output_power": 5.274062881514111e-08, "reference_active": true, "delta_db": -2.1247383919777123}], "active_band_count": 3, "active_band_mae_db": 1.0908981557048163}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.198647979957052e-05, "log_rms_mae_db": 1.4648334918686259, "rms_envelope_nrmse": 0.18595535016594805, "envelope_correlation": 0.9798467319628689, "rms_gain_error_db": 0.11944422909566899, "reference_mean_power": 5.064740774931045e-05, "output_mean_power": 5.2059698067273105e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.700608322524163e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.136344717386399e-05, "output_power": 4.293791414297785e-05, "reference_active": true, "delta_db": 0.16224218152708308}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.648317567219929e-06, "output_power": 6.970532311532303e-06, "reference_active": true, "delta_db": 0.20554188665885126}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.888415633999808e-06, "output_power": 2.751003021821185e-06, "reference_active": true, "delta_db": -1.5028161317465696}], "active_band_count": 3, "active_band_mae_db": 0.6235333999775013}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.0010568231240365e-05, "log_rms_mae_db": 1.1244866537050506, "rms_envelope_nrmse": 0.10082510085114683, "envelope_correlation": 0.9939389108301365, "rms_gain_error_db": 0.11953712350222498, "reference_mean_power": 5.089623620868825e-05, "output_mean_power": 5.231658407161901e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.379392951970345e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.821321246572876e-05, "output_power": 3.971349693237069e-05, "reference_active": true, "delta_db": 0.16724581107876832}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.5895785128125065e-06, "output_power": 3.5928056066276628e-06, "reference_active": true, "delta_db": 0.0039026292871854325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2481135158983729e-07, "output_power": 1.8270160359026505e-07, "reference_active": true, "delta_db": -0.9007587743410059}], "active_band_count": 3, "active_band_mae_db": 0.3573024049023199}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9573331155272664, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "73896002aa28f871d0b21db8e192ce9ad2333ff365c796949423ae5531726f6b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9879098690870616, "standard_deviation": 0.0015394097820794134, "minimum": 0.9849188696136706, "maximum": 0.9901011073522704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.474039660819375, "standard_deviation": 0.025383104852701344, "minimum": 1.4419759787723874, "maximum": 1.523520376784832, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5592148774131527, "standard_deviation": 0.32258475649232693, "minimum": 0.18714589179162142, "maximum": 1.0583024006595743, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.131980695230868, "standard_deviation": 0.008212500270316378, "minimum": 0.11814542950832906, "maximum": 0.14623937086800604, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960239304736662, "standard_deviation": 0.0008529934314818132, "minimum": 0.9945502926196146, "maximum": 0.9970846869647486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9550162003853351, "standard_deviation": 0.03054845061979648, "minimum": 0.9223242173149035, "maximum": 1.0055059415648302, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8631678718658473, "standard_deviation": 0.34361689097314746, "minimum": 0.37957228400416576, "maximum": 1.2976739956876153, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0769753548162384, "standard_deviation": 0.007428478814822404, "minimum": 0.06628051671046307, "maximum": 0.08583184127367965, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9799910287327829, "standard_deviation": 0.001930988713540158, "minimum": 0.977612712100971, "maximum": 0.9841928764807685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4780618440641926, "standard_deviation": 0.03721823364128081, "minimum": 1.4208431975542979, "maximum": 1.5425350012973083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6094960619689604, "standard_deviation": 0.22967417135731188, "minimum": 0.28956282307865217, "maximum": 0.9949204268585866, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18575210357459293, "standard_deviation": 0.010250854556143246, "minimum": 0.1625144044337614, "maximum": 0.1965909073554423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992890362712091, "standard_deviation": 0.0018880845771239868, "minimum": 0.9902411936527122, "maximum": 0.994910007304576, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1468653823182675, "standard_deviation": 0.04872369093430407, "minimum": 1.0917247454551497, "maximum": 1.2488260809755567, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6862850870766894, "standard_deviation": 0.285511852580759, "minimum": 0.3573024049023199, "maximum": 1.1375766263633837, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11019744730069125, "standard_deviation": 0.016239029596464852, "minimum": 0.09350967236880002, "maximum": 0.13430851690646736, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.995021087484555, "standard_deviation": 0.0008728460653347533, "minimum": 0.9939426969903374, "maximum": 0.9963292709055297, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9237577774663139, "standard_deviation": 0.01941737359989628, "minimum": 0.8835093491695841, "maximum": 0.9478725074749214, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.688457929402866, "standard_deviation": 0.2807027690006418, "minimum": 0.26961888372463366, "maximum": 0.9832280519569551, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07749575310791061, "standard_deviation": 0.006963024116874746, "minimum": 0.0668299998336036, "maximum": 0.0857151330366288, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980305210820062, "standard_deviation": 0.0004390106548976545, "minimum": 0.9974364169200752, "maximum": 0.9985310456593955, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.637982127317058, "standard_deviation": 0.02102132678915254, "minimum": 0.61134182138229, "maximum": 0.6685878711863424, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8414926604282471, "standard_deviation": 0.4196309748844093, "minimum": 0.3655894861979592, "maximum": 1.4861747614038798, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.047896798452716584, "standard_deviation": 0.005377962078574431, "minimum": 0.041585946013697264, "maximum": 0.05488355402368404, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.13332212442631997, "standard_deviation": 0.037028197171292326, "minimum": 0.0834406998493561, "maximum": 0.18120665564160088, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.066718514006973, "standard_deviation": 0.010229418703954336, "minimum": 5.053429966657586, "maximum": 5.088538220477833, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36270544409219446, "standard_deviation": 0.004760162459503508, "minimum": 0.35737368290468446, "maximum": 0.37069910245220383, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0773546871757094, "standard_deviation": 0.01475542939683741, "minimum": -0.09015502463968533, "maximum": -0.05052770239331327, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9987727188682547, "standard_deviation": 0.00010739464738725244, "minimum": 0.9986611340651491, "maximum": 0.9989292598945843, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.659960792828534, "standard_deviation": 0.019935179855987036, "minimum": 0.6326596357048605, "maximum": 0.683914077044983, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3912460377430175, "standard_deviation": 0.22806118448984472, "minimum": 0.08901606071272035, "maximum": 0.8549494856777536, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.03847270572156225, "standard_deviation": 0.0015734523848939656, "minimum": 0.036161071886400595, "maximum": 0.040384201525319866, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9996623292506441, "standard_deviation": 5.494014534025506e-05, "minimum": 0.9995598996188645, "maximum": 0.9997460368029877, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.37920269477190305, "standard_deviation": 0.020562165857492126, "minimum": 0.35842270642231727, "maximum": 0.4163589601037868, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.42705156111368126, "standard_deviation": 0.20767120455767415, "minimum": 0.12369154032458969, "maximum": 0.7839221274657273, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02105130172668783, "standard_deviation": 0.0012084627157216803, "minimum": 0.01894188028741663, "maximum": 0.02312370999518375, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-292519-0001", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 3.0599730014801025, "predicted_mos": 3.0599730014801025, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "32f3261d050ddd10a8e0a35effb5ee8c6def6f753ca1098ea0f342033298d382", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8612548001110554, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6293108325712407, "全帯域_10ms_rms_envelope_nrmse": 0.03551913108021514, "全帯域_10ms_envelope_correlation": 0.9989701053435268, "全帯域_10ms_modulation_mae_db": 0.5248714484336319, "全帯域_40ms_log_rms_mae_db": 0.3500668531229584, "全帯域_40ms_rms_envelope_nrmse": 0.01850456400529718, "全帯域_40ms_envelope_correlation": 0.9997589348366733, "全帯域_40ms_modulation_mae_db": 0.5750114770363577, "300–1000 Hz_10ms_log_rms_mae_db": 1.1907276591178193, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07177525378463294, "300–1000 Hz_10ms_envelope_correlation": 0.9957586048966681, "300–1000 Hz_10ms_modulation_mae_db": 0.22855690787142668, "300–1000 Hz_40ms_log_rms_mae_db": 0.9292966902935389, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.041888124896712654, "300–1000 Hz_40ms_envelope_correlation": 0.998513975939222, "300–1000 Hz_40ms_modulation_mae_db": 0.30973561684202283, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8349699162940707, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1738534320020236, "1000–3000 Hz_10ms_envelope_correlation": 0.9784429574509381, "1000–3000 Hz_10ms_modulation_mae_db": 0.21994959848720474, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1914563723338107, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10677873974601673, "1000–3000 Hz_40ms_envelope_correlation": 0.9915065174317171, "1000–3000 Hz_40ms_modulation_mae_db": 1.1151081793636244, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7695863507007215, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.207543808127908, "3000–6000 Hz_10ms_envelope_correlation": 0.974174839106954, "3000–6000 Hz_10ms_modulation_mae_db": 1.0945298434234239, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2135330649798857, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10807728763257411, "3000–6000 Hz_40ms_envelope_correlation": 0.9928213927920556, "3000–6000 Hz_40ms_modulation_mae_db": 1.0171430563314097, "mr_stft_db": 5.4140862359951605, "band_mae_db": 0.14025961469276335, "rms_gain_error_db": -0.08175272609761426, "output_peak": 0.36007435236694724}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6771654712196219, "全帯域_10ms_rms_envelope_nrmse": 0.038435132479697286, "全帯域_10ms_envelope_correlation": 0.9987589318809805, "全帯域_10ms_modulation_mae_db": 0.2451094723605484, "全帯域_40ms_log_rms_mae_db": 0.40442457626889083, "全帯域_40ms_rms_envelope_nrmse": 0.020328794496230113, "全帯域_40ms_envelope_correlation": 0.9996736644148776, "全帯域_40ms_modulation_mae_db": 0.07730449716378536, "300–1000 Hz_10ms_log_rms_mae_db": 1.1841460830397046, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07546873516994786, "300–1000 Hz_10ms_envelope_correlation": 0.9953032417311956, "300–1000 Hz_10ms_modulation_mae_db": 0.3086432455551849, "300–1000 Hz_40ms_log_rms_mae_db": 0.8967306413015187, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.045050725369629896, "300–1000 Hz_40ms_envelope_correlation": 0.9982930762125244, "300–1000 Hz_40ms_modulation_mae_db": 0.3930128009804932, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8456569676131915, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16106419781283804, "1000–3000 Hz_10ms_envelope_correlation": 0.9815174035987151, "1000–3000 Hz_10ms_modulation_mae_db": 0.6022019457922193, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2140031460488665, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08679891047322055, "1000–3000 Hz_40ms_envelope_correlation": 0.9944190549554464, "1000–3000 Hz_40ms_modulation_mae_db": 0.2508973001049551, "3000–6000 Hz_10ms_log_rms_mae_db": 1.760373462017691, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21392613027188848, "3000–6000 Hz_10ms_envelope_correlation": 0.9726985963246572, "3000–6000 Hz_10ms_modulation_mae_db": 0.9788865391002757, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2494560933277752, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11153527738955439, "3000–6000 Hz_40ms_envelope_correlation": 0.9923141693113946, "3000–6000 Hz_40ms_modulation_mae_db": 1.1823301633090963, "mr_stft_db": 5.410447089599669, "band_mae_db": 0.12959861723691066, "rms_gain_error_db": -0.06466947646084087, "output_peak": 0.36611895493845964}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6486010442317077, "全帯域_10ms_rms_envelope_nrmse": 0.034118640318722075, "全帯域_10ms_envelope_correlation": 0.9990081304212374, "全帯域_10ms_modulation_mae_db": 0.3944751235495842, "全帯域_40ms_log_rms_mae_db": 0.3702359031898319, "全帯域_40ms_rms_envelope_nrmse": 0.01844633232880853, "全帯域_40ms_envelope_correlation": 0.9997169322633621, "全帯域_40ms_modulation_mae_db": 0.31272116782789783, "300–1000 Hz_10ms_log_rms_mae_db": 1.1891011596104493, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07462930870823219, "300–1000 Hz_10ms_envelope_correlation": 0.9954519989540875, "300–1000 Hz_10ms_modulation_mae_db": 0.5801887376256409, "300–1000 Hz_40ms_log_rms_mae_db": 0.9147323416060216, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04416752638274271, "300–1000 Hz_40ms_envelope_correlation": 0.9983590244237512, "300–1000 Hz_40ms_modulation_mae_db": 0.7153918472636374, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7411566589179042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.150452927666127, "1000–3000 Hz_10ms_envelope_correlation": 0.9839208569277912, "1000–3000 Hz_10ms_modulation_mae_db": 0.5834773780049968, "1000–3000 Hz_40ms_log_rms_mae_db": 1.118442335951825, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07783545377877994, "1000–3000 Hz_40ms_envelope_correlation": 0.995485700120132, "1000–3000 Hz_40ms_modulation_mae_db": 0.974015423367739, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7869050446476313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18443091413081708, "3000–6000 Hz_10ms_envelope_correlation": 0.9796202152484644, "3000–6000 Hz_10ms_modulation_mae_db": 0.7283508515826496, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2773049315440759, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10773412239063429, "3000–6000 Hz_40ms_envelope_correlation": 0.9929238904526593, "3000–6000 Hz_40ms_modulation_mae_db": 1.3972083631278356, "mr_stft_db": 5.425475279719464, "band_mae_db": 0.14132084306340462, "rms_gain_error_db": -0.04022426071727857, "output_peak": 0.3620563002116267}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6776263569665818, "全帯域_10ms_rms_envelope_nrmse": 0.03739158698776603, "全帯域_10ms_envelope_correlation": 0.9988403486782821, "全帯域_10ms_modulation_mae_db": 0.6056500034031956, "全帯域_40ms_log_rms_mae_db": 0.39575431392989324, "全帯域_40ms_rms_envelope_nrmse": 0.019251348558444868, "全帯域_40ms_envelope_correlation": 0.9997176620512833, "全帯域_40ms_modulation_mae_db": 0.4947253459706446, "300–1000 Hz_10ms_log_rms_mae_db": 1.1991358652907806, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07315991708334597, "300–1000 Hz_10ms_envelope_correlation": 0.9955830464730657, "300–1000 Hz_10ms_modulation_mae_db": 0.8002135321167655, "300–1000 Hz_40ms_log_rms_mae_db": 0.9441922139467218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04332627940617735, "300–1000 Hz_40ms_envelope_correlation": 0.9983923285172417, "300–1000 Hz_40ms_modulation_mae_db": 1.0334654783888249, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7501265946896267, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14972894482960872, "1000–3000 Hz_10ms_envelope_correlation": 0.9844141356158671, "1000–3000 Hz_10ms_modulation_mae_db": 1.1107842426692576, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1356089973361398, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08995861536151872, "1000–3000 Hz_40ms_envelope_correlation": 0.9945957308196863, "1000–3000 Hz_40ms_modulation_mae_db": 1.9896255249030814, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8637947800671255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22849300910278977, "3000–6000 Hz_10ms_envelope_correlation": 0.9694862857596912, "3000–6000 Hz_10ms_modulation_mae_db": 1.0917823354348724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3809565292603696, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11758663972656472, "3000–6000 Hz_40ms_envelope_correlation": 0.9917078173032077, "3000–6000 Hz_40ms_modulation_mae_db": 1.1223965602757617, "mr_stft_db": 5.426315222925724, "band_mae_db": 0.11054681470986448, "rms_gain_error_db": -0.07016338687401337, "output_peak": 0.3651141993116804}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6578300813107668, "全帯域_10ms_rms_envelope_nrmse": 0.03535318132389771, "全帯域_10ms_envelope_correlation": 0.9989776290759215, "全帯域_10ms_modulation_mae_db": 0.4853103448907759, "全帯域_40ms_log_rms_mae_db": 0.34661986995633814, "全帯域_40ms_rms_envelope_nrmse": 0.01676230140381677, "全帯域_40ms_envelope_correlation": 0.999813354300383, "全帯域_40ms_modulation_mae_db": 0.4041768583073346, "300–1000 Hz_10ms_log_rms_mae_db": 1.178697990571064, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06250251468645145, "300–1000 Hz_10ms_envelope_correlation": 0.9967993543576398, "300–1000 Hz_10ms_modulation_mae_db": 0.7535886626056576, "300–1000 Hz_40ms_log_rms_mae_db": 0.9090928173813697, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03668589454331575, "300–1000 Hz_40ms_envelope_correlation": 0.9988629584397152, "300–1000 Hz_40ms_modulation_mae_db": 0.9258918810737717, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8315339475971704, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15779208698847125, "1000–3000 Hz_10ms_envelope_correlation": 0.9827623304726485, "1000–3000 Hz_10ms_modulation_mae_db": 0.7818588470347985, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2103653940797714, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09874150456806387, "1000–3000 Hz_40ms_envelope_correlation": 0.9935210239255932, "1000–3000 Hz_40ms_modulation_mae_db": 1.3556494929637992, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6991330335330423, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19642972519775992, "3000–6000 Hz_10ms_envelope_correlation": 0.9776452665635776, "3000–6000 Hz_10ms_modulation_mae_db": 1.1050241783178878, "3000–6000 Hz_40ms_log_rms_mae_db": 1.216810363415684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1103518038378367, "3000–6000 Hz_40ms_envelope_correlation": 0.9928055581319499, "3000–6000 Hz_40ms_modulation_mae_db": 1.059766311597542, "mr_stft_db": 5.429479978837224, "band_mae_db": 0.14328650121046912, "rms_gain_error_db": -0.08082221137606804, "output_peak": 0.36235268836350815}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6284440283860889, "全帯域_10ms_rms_envelope_nrmse": 0.034694752916324186, "全帯域_10ms_envelope_correlation": 0.9989911557287758, "全帯域_10ms_modulation_mae_db": 0.5937721320965956, "全帯域_40ms_log_rms_mae_db": 0.33879203150474346, "全帯域_40ms_rms_envelope_nrmse": 0.017861389844163886, "全帯域_40ms_envelope_correlation": 0.999748711303532, "全帯域_40ms_modulation_mae_db": 0.47675001449362703, "300–1000 Hz_10ms_log_rms_mae_db": 1.2154788842622835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08225848942076111, "300–1000 Hz_10ms_envelope_correlation": 0.9943982115397811, "300–1000 Hz_10ms_modulation_mae_db": 0.6732691543438443, "300–1000 Hz_40ms_log_rms_mae_db": 0.9028069362986061, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05215417393211698, "300–1000 Hz_40ms_envelope_correlation": 0.9976644410109982, "300–1000 Hz_40ms_modulation_mae_db": 0.7479085784428147, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7860334335991561, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17269995736565066, "1000–3000 Hz_10ms_envelope_correlation": 0.9787350414895979, "1000–3000 Hz_10ms_modulation_mae_db": 0.5897521052534073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1608063530658521, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09447051782602335, "1000–3000 Hz_40ms_envelope_correlation": 0.9934240169486709, "1000–3000 Hz_40ms_modulation_mae_db": 0.6345362374337856, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7516122956568343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1833399670470037, "3000–6000 Hz_10ms_envelope_correlation": 0.9798706078378274, "3000–6000 Hz_10ms_modulation_mae_db": 1.2907070370541092, "3000–6000 Hz_40ms_log_rms_mae_db": 1.242081140816407, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09753370826138867, "3000–6000 Hz_40ms_envelope_correlation": 0.9942036038439904, "3000–6000 Hz_40ms_modulation_mae_db": 1.3538939938460313, "mr_stft_db": 5.417926530665484, "band_mae_db": 0.16232601228710716, "rms_gain_error_db": -0.06329164627684475, "output_peak": 0.36013366974395117}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.650393424473668, "全帯域_10ms_rms_envelope_nrmse": 0.03483000368085634, "全帯域_10ms_envelope_correlation": 0.9989976835295818, "全帯域_10ms_modulation_mae_db": 0.5965056476790194, "全帯域_40ms_log_rms_mae_db": 0.36715761929917834, "全帯域_40ms_rms_envelope_nrmse": 0.01716023195987174, "全帯域_40ms_envelope_correlation": 0.9997883588205454, "全帯域_40ms_modulation_mae_db": 0.5346909286302419, "300–1000 Hz_10ms_log_rms_mae_db": 1.1850021092485044, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07328076470803244, "300–1000 Hz_10ms_envelope_correlation": 0.9955783714930222, "300–1000 Hz_10ms_modulation_mae_db": 0.9410014062906821, "300–1000 Hz_40ms_log_rms_mae_db": 0.9357377609454637, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04267276156907702, "300–1000 Hz_40ms_envelope_correlation": 0.9984470066138279, "300–1000 Hz_40ms_modulation_mae_db": 1.3445354339516646, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8515395970813093, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16087924750944968, "1000–3000 Hz_10ms_envelope_correlation": 0.9816771768029662, "1000–3000 Hz_10ms_modulation_mae_db": 0.9548848587525164, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2342145106042897, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09295914896440327, "1000–3000 Hz_40ms_envelope_correlation": 0.9938069485904574, "1000–3000 Hz_40ms_modulation_mae_db": 1.6773352010024147, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8001825789258175, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20905505072904804, "3000–6000 Hz_10ms_envelope_correlation": 0.973745092490359, "3000–6000 Hz_10ms_modulation_mae_db": 0.9987671430075663, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2557991304891742, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10001357304072143, "3000–6000 Hz_40ms_envelope_correlation": 0.9938745793332152, "3000–6000 Hz_40ms_modulation_mae_db": 0.9225785396484287, "mr_stft_db": 5.418896748078278, "band_mae_db": 0.049293845777053785, "rms_gain_error_db": -0.07622528823566196, "output_peak": 0.36087692170476454}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.633437672085786, "全帯域_10ms_rms_envelope_nrmse": 0.037392301370038676, "全帯域_10ms_envelope_correlation": 0.9988334352425576, "全帯域_10ms_modulation_mae_db": 0.39502768203419203, "全帯域_40ms_log_rms_mae_db": 0.34810434872770185, "全帯域_40ms_rms_envelope_nrmse": 0.02064633341441443, "全帯域_40ms_envelope_correlation": 0.999666687739673, "全帯域_40ms_modulation_mae_db": 0.6028774306677857, "300–1000 Hz_10ms_log_rms_mae_db": 1.1473327389855101, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08011435463565732, "300–1000 Hz_10ms_envelope_correlation": 0.9946881284100976, "300–1000 Hz_10ms_modulation_mae_db": 0.8579071136233876, "300–1000 Hz_40ms_log_rms_mae_db": 0.8606833548911871, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04802216879576174, "300–1000 Hz_40ms_envelope_correlation": 0.9980120534631354, "300–1000 Hz_40ms_modulation_mae_db": 1.265179890582088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7932933215769116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15631733659768707, "1000–3000 Hz_10ms_envelope_correlation": 0.9828298592733185, "1000–3000 Hz_10ms_modulation_mae_db": 0.9149800884914594, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1594256005336208, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08440049910747173, "1000–3000 Hz_40ms_envelope_correlation": 0.9951675420004031, "1000–3000 Hz_40ms_modulation_mae_db": 1.7248245334939594, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7608164482107735, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18882922955868567, "3000–6000 Hz_10ms_envelope_correlation": 0.9789172277340011, "3000–6000 Hz_10ms_modulation_mae_db": 0.3255272611128375, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2730048799574873, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10528607763486607, "3000–6000 Hz_40ms_envelope_correlation": 0.9932076947515116, "3000–6000 Hz_40ms_modulation_mae_db": 0.8347776706404386, "mr_stft_db": 5.405365486454305, "band_mae_db": 0.08006453151517941, "rms_gain_error_db": -0.07178037628158011, "output_peak": 0.36660386242138565}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.603151334710429, "spectral_convergence": 0.11620714622509129, "spectral_convergence_status": "ok", "amplitude_floor": 1.7050603882585003e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.480262909993648, "spectral_convergence": 0.11300405707300508, "spectral_convergence_status": "ok", "amplitude_floor": 1.5194442893884604e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.158844463281405, "spectral_convergence": 0.11052541818076506, "spectral_convergence_status": "ok", "amplitude_floor": 1.5327487413055003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00047003850181382925, "output_power": 0.0004649817303736685, "delta_db": -0.0469754400746142, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.004513086919531444, "output_power": 0.004427439758656256, "delta_db": -0.0832103713485756, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003251325154635926, "output_power": 0.0003164774902206098, "delta_db": -0.11717578250381502, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.51153215194187e-05, "output_power": 8.28158473605047e-05, "delta_db": -0.11894294153452332, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6288307106843834e-05, "output_power": 2.4416997547282927e-05, "delta_db": -0.3207036008624813, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.4880972570108e-05, "output_power": 4.650688103321771e-05, "delta_db": 0.15454955183257071, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.14025961469276335, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08175272609761426, "pitch": null, "output_rms_dbfs": -22.706214615913222, "output_peak": 0.36007435236694724, "reference_rms_dbfs": -22.624461889815606, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.4140862359951605}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023914198234423542, "log_rms_mae_db": 0.6293108325712407, "rms_envelope_nrmse": 0.03551913108021514, "envelope_correlation": 0.9989701053435268, "rms_gain_error_db": -0.07851235290051597, "reference_mean_power": 0.005456771858924342, "output_mean_power": 0.005359009927161647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.071388012833875e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045567084702756786, "output_power": 0.0004464507232275109, "reference_active": true, "delta_db": -0.08877713282584644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.44246106516196e-05, "output_power": 6.000266769224536e-05, "reference_active": true, "delta_db": -0.3088124322924321}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7099940640562705e-05, "output_power": 1.3040436383060017e-05, "reference_active": true, "delta_db": -1.177024780182617}], "active_band_count": 3, "active_band_mae_db": 0.5248714484336319}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022747909354312384, "log_rms_mae_db": 0.3500668531229584, "rms_envelope_nrmse": 0.01850456400529718, "envelope_correlation": 0.9997589348366733, "rms_gain_error_db": -0.0751260855318876, "reference_mean_power": 0.005416920568028677, "output_mean_power": 0.00532402221235862}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.96418119219355e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000410225191407457, "output_power": 0.00040377538942223604, "reference_active": true, "delta_db": -0.06882481660743811}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7757195967845804e-05, "output_power": 2.6951450963582605e-05, "reference_active": true, "delta_db": -0.12793440651929133}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0255605769228834e-06, "output_power": 7.213296717841827e-07, "reference_active": true, "delta_db": -1.5282752079823436}], "active_band_count": 3, "active_band_mae_db": 0.5750114770363577}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.442623236865579e-05, "log_rms_mae_db": 1.1907276591178193, "rms_envelope_nrmse": 0.07177525378463294, "envelope_correlation": 0.9957586048966681, "rms_gain_error_db": -0.010094174844165959, "reference_mean_power": 0.0007990709247977912, "output_mean_power": 0.0007972158252082865}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.944736222458957e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022092122499658314, "output_power": 0.0002133569348005447, "reference_active": true, "delta_db": -0.15130679277785827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3665816077667355e-05, "output_power": 2.1865187284313588e-05, "reference_active": true, "delta_db": -0.3436828340259535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5979447501439136e-06, "output_power": 5.357479255828191e-06, "reference_active": true, "delta_db": -0.19068109681046824}], "active_band_count": 3, "active_band_mae_db": 0.22855690787142668}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.226666373654229e-05, "log_rms_mae_db": 0.9292966902935389, "rms_envelope_nrmse": 0.041888124896712654, "envelope_correlation": 0.998513975939222, "rms_gain_error_db": -0.010622924126599924, "reference_mean_power": 0.0008012273487091395, "output_mean_power": 0.0007992699265737315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.73006991688152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002092887788100005, "output_power": 0.00020212846808498417, "reference_active": true, "delta_db": -0.1511845937974652}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0995901908054441e-05, "output_power": 1.0229314886672184e-05, "reference_active": true, "delta_db": -0.31384309343663735}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1412698290887665e-07, "output_power": 3.7214793375617424e-07, "reference_active": true, "delta_db": -0.46417916329196585}], "active_band_count": 3, "active_band_mae_db": 0.30973561684202283}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9464514079665034e-05, "log_rms_mae_db": 1.8349699162940707, "rms_envelope_nrmse": 0.1738534320020236, "envelope_correlation": 0.9784429574509381, "rms_gain_error_db": -0.12333374162398451, "reference_mean_power": 7.576176061225114e-05, "output_mean_power": 7.364049251247058e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.472428410826198e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.044189308875811e-05, "output_power": 3.117571073724024e-05, "reference_active": true, "delta_db": 0.10344706825439542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3472076435688925e-06, "output_power": 3.304092681003679e-06, "reference_active": true, "delta_db": -0.05630433654644781}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0111506850143103e-06, "output_power": 9.011687876513629e-07, "reference_active": true, "delta_db": -0.500097390660771}], "active_band_count": 3, "active_band_mae_db": 0.21994959848720474}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.608084029396227e-05, "log_rms_mae_db": 1.1914563723338107, "rms_envelope_nrmse": 0.10677873974601673, "envelope_correlation": 0.9915065174317171, "rms_gain_error_db": -0.12486338937786563, "reference_mean_power": 7.611087404608628e-05, "output_mean_power": 7.395377887774189e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.164443748057941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.887302600627445e-05, "output_power": 2.993947073629384e-05, "reference_active": true, "delta_db": 0.15751816678171332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.617498060844069e-06, "output_power": 1.4502490456645037e-06, "reference_active": true, "delta_db": -0.4740118032966444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.602390292027545e-08, "output_power": 4.605106579776165e-08, "reference_active": true, "delta_db": -2.7137945680125157}], "active_band_count": 3, "active_band_mae_db": 1.1151081793636244}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.198647979957052e-05, "log_rms_mae_db": 1.7695863507007215, "rms_envelope_nrmse": 0.207543808127908, "envelope_correlation": 0.974174839106954, "rms_gain_error_db": -0.10386971050300056, "reference_mean_power": 5.064740774931045e-05, "output_mean_power": 4.945045034005994e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.700608322524163e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.136344717386399e-05, "output_power": 4.055408664235654e-05, "reference_active": true, "delta_db": -0.08582100427368644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.648317567219929e-06, "output_power": 5.652824337233025e-06, "reference_active": true, "delta_db": -0.7044626614964123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.888415633999808e-06, "output_power": 2.1899898104307907e-06, "reference_active": true, "delta_db": -2.493305864500173}], "active_band_count": 3, "active_band_mae_db": 1.0945298434234239}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.0010568231240365e-05, "log_rms_mae_db": 1.2135330649798857, "rms_envelope_nrmse": 0.10807728763257411, "envelope_correlation": 0.9928213927920556, "rms_gain_error_db": -0.10411442022371073, "reference_mean_power": 5.089623620868825e-05, "output_mean_power": 4.969059823102312e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.379392951970345e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.821321246572876e-05, "output_power": 3.7281473436272056e-05, "reference_active": true, "delta_db": -0.10720480824351687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.5895785128125065e-06, "output_power": 3.0407666524990027e-06, "reference_active": true, "delta_db": -0.7206036305219594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2481135158983729e-07, "output_power": 1.3472747286388525e-07, "reference_active": true, "delta_db": -2.223620730228753}], "active_band_count": 3, "active_band_mae_db": 1.0171430563314097}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9477040384222986, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "377f004d6028cacbe063473b2a1fd73c4e5aed3992ce2de9938a477918585415", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9817874702039803, "standard_deviation": 0.0022063487883790937, "minimum": 0.9784429574509381, "maximum": 0.9844141356158671, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8042888046711676, "standard_deviation": 0.043096176083281174, "minimum": 1.7411566589179042, "maximum": 1.8515395970813093, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7197361330607325, "standard_deviation": 0.28011633860616114, "minimum": 0.21994959848720474, "maximum": 1.1107842426692576, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.160348516346482, "standard_deviation": 0.00901758942610239, "minimum": 0.14972894482960872, "maximum": 0.1738534320020236, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939908168490132, "standard_deviation": 0.0012487814353604962, "minimum": 0.9915065174317171, "maximum": 0.995485700120132, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.178040338744272, "standard_deviation": 0.04082999538970265, "minimum": 1.118442335951825, "maximum": 1.2342145106042897, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2152489865791698, "standard_deviation": 0.5889440330417306, "minimum": 0.2508973001049551, "maximum": 1.9896255249030814, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09149292372818726, "standard_deviation": 0.008931731663803028, "minimum": 0.07783545377877994, "maximum": 0.10677873974601673, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9757697663831915, "standard_deviation": 0.003791422323477718, "minimum": 0.9694862857596912, "maximum": 0.9798706078378274, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7740504992199546, "standard_deviation": 0.04690475241154574, "minimum": 1.6991330335330423, "maximum": 1.8637947800671255, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9516968986292028, "standard_deviation": 0.2984218166356171, "minimum": 0.3255272611128375, "maximum": 1.2907070370541092, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2015059792707376, "standard_deviation": 0.015964397780952776, "minimum": 0.1833399670470037, "maximum": 0.22849300910278977, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.992982338239998, "standard_deviation": 0.0007990881879036145, "minimum": 0.9917078173032077, "maximum": 0.9942036038439904, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2636182667238574, "standard_deviation": 0.05273757435499245, "minimum": 1.2135330649798857, "maximum": 1.3809565292603696, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.111261832347068, "standard_deviation": 0.1961651940020955, "minimum": 0.8347776706404386, "maximum": 1.3972083631278356, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10726481123926755, "standard_deviation": 0.006396172135039034, "minimum": 0.09753370826138867, "maximum": 0.11758663972656472, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9954451197319447, "standard_deviation": 0.0007229140931618932, "minimum": 0.9943982115397811, "maximum": 0.9967993543576398, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1862028112657645, "standard_deviation": 0.019368987276368238, "minimum": 1.1473327389855101, "maximum": 1.2154788842622835, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6429210950040737, "standard_deviation": 0.25644681850474116, "minimum": 0.22855690787142668, "maximum": 0.9410014062906821, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07414866727463265, "standard_deviation": 0.005932104900754064, "minimum": 0.06250251468645145, "maximum": 0.08225848942076111, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9983181080775521, "standard_deviation": 0.0003546830466894116, "minimum": 0.9976644410109982, "maximum": 0.9988629584397152, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9116590945830534, "standard_deviation": 0.026362979721339636, "minimum": 0.8606833548911871, "maximum": 0.9441922139467218, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8418901909406646, "standard_deviation": 0.37513760546583197, "minimum": 0.30973561684202283, "maximum": 1.3445354339516646, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04424595686194176, "standard_deviation": 0.004527505484402567, "minimum": 0.03668589454331575, "maximum": 0.05215417393211698, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.11958709756159408, "standard_deviation": 0.03774560889672081, "minimum": 0.049293845777053785, "maximum": 0.16232601228710716, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.418499071534413, "standard_deviation": 0.008346295719971565, "minimum": 5.405365486454305, "maximum": 5.429479978837224, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3629163686327904, "standard_deviation": 0.0026651694035629897, "minimum": 0.36007435236694724, "maximum": 0.36660386242138565, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.06861617153998774, "standard_deviation": 0.013319635180764983, "minimum": -0.08175272609761426, "maximum": -0.04022426071727857, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.998922177487608, "standard_deviation": 9.594588737078325e-05, "minimum": 0.9987589318809805, "maximum": 0.9990081304212374, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6503511139056828, "standard_deviation": 0.019750683454631077, "minimum": 0.6284440283860889, "maximum": 0.6776263569665818, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.48009023180594285, "standard_deviation": 0.12764657017292372, "minimum": 0.2451094723605484, "maximum": 0.6056500034031956, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.03596684126968969, "standard_deviation": 0.0015610333819256444, "minimum": 0.034118640318722075, "maximum": 0.038435132479697286, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9997355382162912, "standard_deviation": 5.182496012213288e-05, "minimum": 0.999666687739673, "maximum": 0.999813354300383, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.365144439499942, "standard_deviation": 0.0240985383107922, "minimum": 0.33879203150474346, "maximum": 0.40442457626889083, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43478221501220937, "standard_deviation": 0.17191458822426717, "minimum": 0.07730449716378536, "maximum": 0.6028774306677857, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.01862016200138094, "standard_deviation": 0.001395172730303745, "minimum": 0.01676230140381677, "maximum": 0.02064633341441443, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-292519-0001", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.966005802154541, "predicted_mos": 2.966005802154541, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "326883a706c2bbe22c4ab94103681d6c6e69be99ce7842b82d0cdcf4f4d71525", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.08140749996528, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6771988029140048, "全帯域_10ms_rms_envelope_nrmse": 0.04271343711973771, "全帯域_10ms_envelope_correlation": 0.9985550430776056, "全帯域_10ms_modulation_mae_db": 0.30897442245806034, "全帯域_40ms_log_rms_mae_db": 0.3793257076699182, "全帯域_40ms_rms_envelope_nrmse": 0.024248039960031578, "全帯域_40ms_envelope_correlation": 0.9996207172222924, "全帯域_40ms_modulation_mae_db": 0.3867222378932451, "300–1000 Hz_10ms_log_rms_mae_db": 0.9316411656542752, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08298830095657599, "300–1000 Hz_10ms_envelope_correlation": 0.9943338132957726, "300–1000 Hz_10ms_modulation_mae_db": 0.6715136139126514, "300–1000 Hz_40ms_log_rms_mae_db": 0.6440807882740524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05129650223325181, "300–1000 Hz_40ms_envelope_correlation": 0.9978019000733274, "300–1000 Hz_40ms_modulation_mae_db": 0.685697117016644, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3820798041741684, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12916010506731457, "1000–3000 Hz_10ms_envelope_correlation": 0.9885232225286538, "1000–3000 Hz_10ms_modulation_mae_db": 0.46107602667444136, "1000–3000 Hz_40ms_log_rms_mae_db": 0.893380619566842, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07022717404543476, "1000–3000 Hz_40ms_envelope_correlation": 0.9969374378702102, "1000–3000 Hz_40ms_modulation_mae_db": 0.6774053266386101, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3534989014770165, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22055958350163715, "3000–6000 Hz_10ms_envelope_correlation": 0.9715967512764275, "3000–6000 Hz_10ms_modulation_mae_db": 0.6806154973041466, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0132287719008588, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12146458738768316, "3000–6000 Hz_40ms_envelope_correlation": 0.9911428336622338, "3000–6000 Hz_40ms_modulation_mae_db": 0.7673557153183478, "mr_stft_db": 5.0329153776458595, "band_mae_db": 0.15431867342465774, "rms_gain_error_db": -0.11927322137644256, "output_peak": 0.36442500893866026}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6647123371950764, "全帯域_10ms_rms_envelope_nrmse": 0.03616853397139069, "全帯域_10ms_envelope_correlation": 0.9989344653733842, "全帯域_10ms_modulation_mae_db": 0.1918172024508575, "全帯域_40ms_log_rms_mae_db": 0.3901466461769882, "全帯域_40ms_rms_envelope_nrmse": 0.019189886801348793, "全帯域_40ms_envelope_correlation": 0.9997479178209123, "全帯域_40ms_modulation_mae_db": 0.2900449494256265, "300–1000 Hz_10ms_log_rms_mae_db": 0.9126780004468145, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08267157441052944, "300–1000 Hz_10ms_envelope_correlation": 0.9943414590248553, "300–1000 Hz_10ms_modulation_mae_db": 0.8923377925739375, "300–1000 Hz_40ms_log_rms_mae_db": 0.6252209443510639, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.047010982179150985, "300–1000 Hz_40ms_envelope_correlation": 0.9981054579152201, "300–1000 Hz_40ms_modulation_mae_db": 1.2298314300856983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.408955698353395, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13395817743206906, "1000–3000 Hz_10ms_envelope_correlation": 0.9876420263294939, "1000–3000 Hz_10ms_modulation_mae_db": 0.674678388532584, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9771697353669174, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0846610403145463, "1000–3000 Hz_40ms_envelope_correlation": 0.9952675931605172, "1000–3000 Hz_40ms_modulation_mae_db": 1.1115854896840658, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3018505186561231, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17263924197837816, "3000–6000 Hz_10ms_envelope_correlation": 0.983026065821044, "3000–6000 Hz_10ms_modulation_mae_db": 0.6462683610651712, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9758293726454316, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09912690681940965, "3000–6000 Hz_40ms_envelope_correlation": 0.9944166014745518, "3000–6000 Hz_40ms_modulation_mae_db": 0.783670792045411, "mr_stft_db": 5.035813176468797, "band_mae_db": 0.09714365543150945, "rms_gain_error_db": -0.08565796083103139, "output_peak": 0.36416840008612467}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6649884826948971, "全帯域_10ms_rms_envelope_nrmse": 0.03610857893444671, "全帯域_10ms_envelope_correlation": 0.9989486259560909, "全帯域_10ms_modulation_mae_db": 0.45469972368722616, "全帯域_40ms_log_rms_mae_db": 0.38245357300770366, "全帯域_40ms_rms_envelope_nrmse": 0.02144285387544227, "全帯域_40ms_envelope_correlation": 0.9996859552563039, "全帯域_40ms_modulation_mae_db": 0.5991264638245806, "300–1000 Hz_10ms_log_rms_mae_db": 0.915429403654229, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08075907082239824, "300–1000 Hz_10ms_envelope_correlation": 0.9946304636012597, "300–1000 Hz_10ms_modulation_mae_db": 1.2936152742110827, "300–1000 Hz_40ms_log_rms_mae_db": 0.6324486711700823, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05340037477291673, "300–1000 Hz_40ms_envelope_correlation": 0.9975987791313166, "300–1000 Hz_40ms_modulation_mae_db": 1.5570215413556385, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3260249540103572, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12082496694441788, "1000–3000 Hz_10ms_envelope_correlation": 0.9900989594349403, "1000–3000 Hz_10ms_modulation_mae_db": 1.108692036315733, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8967886896184548, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07353087337087329, "1000–3000 Hz_40ms_envelope_correlation": 0.9966275717360383, "1000–3000 Hz_40ms_modulation_mae_db": 1.5023767197886182, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3464501345525313, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16668877667955728, "3000–6000 Hz_10ms_envelope_correlation": 0.9834301727181283, "3000–6000 Hz_10ms_modulation_mae_db": 0.553411481874698, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0047998735360093, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0860874472426802, "3000–6000 Hz_40ms_envelope_correlation": 0.9954180432125562, "3000–6000 Hz_40ms_modulation_mae_db": 0.3922096043461039, "mr_stft_db": 5.022835540131183, "band_mae_db": 0.1907995709429868, "rms_gain_error_db": -0.08690038925490086, "output_peak": 0.35541140800967214}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6592141542638424, "全帯域_10ms_rms_envelope_nrmse": 0.0397681417639525, "全帯域_10ms_envelope_correlation": 0.9987187112424923, "全帯域_10ms_modulation_mae_db": 0.23061435183300327, "全帯域_40ms_log_rms_mae_db": 0.36150497227516737, "全帯域_40ms_rms_envelope_nrmse": 0.0212680168767276, "全帯域_40ms_envelope_correlation": 0.9996906034887089, "全帯域_40ms_modulation_mae_db": 0.35330243127399336, "300–1000 Hz_10ms_log_rms_mae_db": 0.9438599198913541, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07201147837796869, "300–1000 Hz_10ms_envelope_correlation": 0.9957439866649095, "300–1000 Hz_10ms_modulation_mae_db": 0.2598594082923751, "300–1000 Hz_40ms_log_rms_mae_db": 0.6877869830063871, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04413215747820744, "300–1000 Hz_40ms_envelope_correlation": 0.9983481967032392, "300–1000 Hz_40ms_modulation_mae_db": 0.5021289334256969, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3811164188583949, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14165546730605436, "1000–3000 Hz_10ms_envelope_correlation": 0.9859555187950518, "1000–3000 Hz_10ms_modulation_mae_db": 0.3129496145008413, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8997729095946234, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07540767049623928, "1000–3000 Hz_40ms_envelope_correlation": 0.9962202481465676, "1000–3000 Hz_40ms_modulation_mae_db": 0.5413958044224173, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2402073056545138, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20276768201488302, "3000–6000 Hz_10ms_envelope_correlation": 0.9756958884944192, "3000–6000 Hz_10ms_modulation_mae_db": 1.3507260680820832, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9152870859056256, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11739944559922083, "3000–6000 Hz_40ms_envelope_correlation": 0.9924976762985194, "3000–6000 Hz_40ms_modulation_mae_db": 1.4158236355497602, "mr_stft_db": 5.0324354076872355, "band_mae_db": 0.1326061861376537, "rms_gain_error_db": -0.10077982085815056, "output_peak": 0.3702920377690389}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6471117687502083, "全帯域_10ms_rms_envelope_nrmse": 0.03736155093519765, "全帯域_10ms_envelope_correlation": 0.9988058362865755, "全帯域_10ms_modulation_mae_db": 0.33040923075852385, "全帯域_40ms_log_rms_mae_db": 0.365566706446248, "全帯域_40ms_rms_envelope_nrmse": 0.019277597641770894, "全帯域_40ms_envelope_correlation": 0.9996874539115118, "全帯域_40ms_modulation_mae_db": 0.17539622114572526, "300–1000 Hz_10ms_log_rms_mae_db": 0.8952257121560131, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0800881133498363, "300–1000 Hz_10ms_envelope_correlation": 0.9947274158900815, "300–1000 Hz_10ms_modulation_mae_db": 0.44826577244445903, "300–1000 Hz_40ms_log_rms_mae_db": 0.6412948068238252, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04444225905641148, "300–1000 Hz_40ms_envelope_correlation": 0.9983264050632101, "300–1000 Hz_40ms_modulation_mae_db": 0.6025361661499099, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4184525885537085, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14448654694010174, "1000–3000 Hz_10ms_envelope_correlation": 0.9853476423279406, "1000–3000 Hz_10ms_modulation_mae_db": 0.23010477582611832, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9731162026372946, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08615327911620754, "1000–3000 Hz_40ms_envelope_correlation": 0.9948554994333463, "1000–3000 Hz_40ms_modulation_mae_db": 0.4614322582812351, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3299356360531205, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18234222721785592, "3000–6000 Hz_10ms_envelope_correlation": 0.9801325098732871, "3000–6000 Hz_10ms_modulation_mae_db": 0.8434538322498838, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0270520437246915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.100767430503309, "3000–6000 Hz_40ms_envelope_correlation": 0.9937162506039192, "3000–6000 Hz_40ms_modulation_mae_db": 0.8513198329569418, "mr_stft_db": 5.043542885889468, "band_mae_db": 0.10797578701739204, "rms_gain_error_db": -0.04339337802227826, "output_peak": 0.36222076153697674}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.667855605005358, "全帯域_10ms_rms_envelope_nrmse": 0.03746026412602044, "全帯域_10ms_envelope_correlation": 0.9988329852894142, "全帯域_10ms_modulation_mae_db": 0.37072325187793603, "全帯域_40ms_log_rms_mae_db": 0.3769009757490636, "全帯域_40ms_rms_envelope_nrmse": 0.01978067865978798, "全帯域_40ms_envelope_correlation": 0.9997007567010604, "全帯域_40ms_modulation_mae_db": 0.5384750832471304, "300–1000 Hz_10ms_log_rms_mae_db": 0.9641284783734688, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06854166759452461, "300–1000 Hz_10ms_envelope_correlation": 0.9961289633556064, "300–1000 Hz_10ms_modulation_mae_db": 0.4463742945120823, "300–1000 Hz_40ms_log_rms_mae_db": 0.6910709769213432, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04355432928621507, "300–1000 Hz_40ms_envelope_correlation": 0.9983777952124828, "300–1000 Hz_40ms_modulation_mae_db": 0.5500224312360599, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3545536091488495, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1138799936896637, "1000–3000 Hz_10ms_envelope_correlation": 0.9908120258565704, "1000–3000 Hz_10ms_modulation_mae_db": 0.14551841912149235, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8611167899411384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.05742953249783038, "1000–3000 Hz_40ms_envelope_correlation": 0.997550761622319, "1000–3000 Hz_40ms_modulation_mae_db": 0.33403012801986404, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3507705442552758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19451680484793266, "3000–6000 Hz_10ms_envelope_correlation": 0.9779237176727573, "3000–6000 Hz_10ms_modulation_mae_db": 0.882157022633485, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0257085571453661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1200039944801458, "3000–6000 Hz_40ms_envelope_correlation": 0.9923407737354675, "3000–6000 Hz_40ms_modulation_mae_db": 1.0562656078316868, "mr_stft_db": 5.041043658155186, "band_mae_db": 0.12023680723449708, "rms_gain_error_db": -0.07673411496202262, "output_peak": 0.36502286951469504}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6920043878481531, "全帯域_10ms_rms_envelope_nrmse": 0.037624907541773035, "全帯域_10ms_envelope_correlation": 0.9987930720409262, "全帯域_10ms_modulation_mae_db": 0.09914516407653506, "全帯域_40ms_log_rms_mae_db": 0.395898479819733, "全帯域_40ms_rms_envelope_nrmse": 0.021893418234546342, "全帯域_40ms_envelope_correlation": 0.9995976806712304, "全帯域_40ms_modulation_mae_db": 0.2076324247362713, "300–1000 Hz_10ms_log_rms_mae_db": 0.9330648140716095, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08167248343338412, "300–1000 Hz_10ms_envelope_correlation": 0.9946728546946467, "300–1000 Hz_10ms_modulation_mae_db": 0.05206517605595703, "300–1000 Hz_40ms_log_rms_mae_db": 0.6566249153067076, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05139442211848721, "300–1000 Hz_40ms_envelope_correlation": 0.9978765563241048, "300–1000 Hz_40ms_modulation_mae_db": 0.07128682188773411, "1000–3000 Hz_10ms_log_rms_mae_db": 1.390850874665753, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13810653841923826, "1000–3000 Hz_10ms_envelope_correlation": 0.9864634637395295, "1000–3000 Hz_10ms_modulation_mae_db": 0.47754539333798823, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9493741286069552, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0854051540115424, "1000–3000 Hz_40ms_envelope_correlation": 0.9946200466616709, "1000–3000 Hz_40ms_modulation_mae_db": 1.1612819338751357, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3560093823132924, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23002477483359215, "3000–6000 Hz_10ms_envelope_correlation": 0.9755062882983113, "3000–6000 Hz_10ms_modulation_mae_db": 1.3019322790415357, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0627024800436842, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16110224180922117, "3000–6000 Hz_40ms_envelope_correlation": 0.989922773350234, "3000–6000 Hz_40ms_modulation_mae_db": 1.1606515745132182, "mr_stft_db": 5.04523888892473, "band_mae_db": 0.1853913728481796, "rms_gain_error_db": -0.04555414118279556, "output_peak": 0.3573291036001644}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6772205760008518, "全帯域_10ms_rms_envelope_nrmse": 0.03914431550271216, "全帯域_10ms_envelope_correlation": 0.9987166020542348, "全帯域_10ms_modulation_mae_db": 0.8890978685681127, "全帯域_40ms_log_rms_mae_db": 0.37567070010432585, "全帯域_40ms_rms_envelope_nrmse": 0.01896210488661957, "全帯域_40ms_envelope_correlation": 0.9997217053440747, "全帯域_40ms_modulation_mae_db": 0.7731133354755921, "300–1000 Hz_10ms_log_rms_mae_db": 0.8653341252951043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07152591821239657, "300–1000 Hz_10ms_envelope_correlation": 0.9958026335500334, "300–1000 Hz_10ms_modulation_mae_db": 1.0149504600389074, "300–1000 Hz_40ms_log_rms_mae_db": 0.6111246811414368, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04018811849344713, "300–1000 Hz_40ms_envelope_correlation": 0.9986356764119901, "300–1000 Hz_40ms_modulation_mae_db": 1.3660356914218224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.378127252616148, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12278345237309475, "1000–3000 Hz_10ms_envelope_correlation": 0.9895064000086161, "1000–3000 Hz_10ms_modulation_mae_db": 0.9543707237329437, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9335078994908985, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07569920100700546, "1000–3000 Hz_40ms_envelope_correlation": 0.9960706110199181, "1000–3000 Hz_40ms_modulation_mae_db": 1.2171634621990408, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3441591144268066, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19302671423573609, "3000–6000 Hz_10ms_envelope_correlation": 0.9776413621719048, "3000–6000 Hz_10ms_modulation_mae_db": 1.1473383567858522, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0696161562852162, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10947168732765881, "3000–6000 Hz_40ms_envelope_correlation": 0.9926023571052025, "3000–6000 Hz_40ms_modulation_mae_db": 0.7471736221676163, "mr_stft_db": 5.032368734101343, "band_mae_db": 0.0899441455673916, "rms_gain_error_db": -0.0660672327243349, "output_peak": 0.36528242655715387}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.147849330463986, "spectral_convergence": 0.11894551368779101, "spectral_convergence_status": "ok", "amplitude_floor": 1.7050603882585003e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.988534319182073, "spectral_convergence": 0.11269375482154428, "spectral_convergence_status": "ok", "amplitude_floor": 1.5194442893884604e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.962362483291517, "spectral_convergence": 0.11541680530281487, "spectral_convergence_status": "ok", "amplitude_floor": 1.5327487413055003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00047003850181382925, "output_power": 0.00047113424313160134, "delta_db": 0.010112375766566909, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.004513086919531444, "output_power": 0.004377322917155195, "delta_db": -0.13265111968071677, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003251325154635926, "output_power": 0.0003155845538048159, "delta_db": -0.12944665431860353, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.51153215194187e-05, "output_power": 8.050281577837497e-05, "delta_db": -0.24196672893213322, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6288307106843834e-05, "output_power": 2.818967867430115e-05, "delta_db": 0.30327505845410696, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.4880972570108e-05, "output_power": 4.3774001570975254e-05, "delta_db": -0.10846010339581906, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15431867342465774, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11927322137644256, "pitch": null, "output_rms_dbfs": -22.74373511119205, "output_peak": 0.36442500893866026, "reference_rms_dbfs": -22.624461889815606, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.0329153776458595}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023914198234423542, "log_rms_mae_db": 0.6771988029140048, "rms_envelope_nrmse": 0.04271343711973771, "envelope_correlation": 0.9985550430776056, "rms_gain_error_db": -0.11759850044165812, "reference_mean_power": 0.005456771858924342, "output_mean_power": 0.005310995663207287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.071388012833875e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045567084702756786, "output_power": 0.000444249001401183, "reference_active": true, "delta_db": -0.11024783863103957}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.44246106516196e-05, "output_power": 5.8645253322402936e-05, "reference_active": true, "delta_db": -0.40818935992830707}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7099940640562705e-05, "output_power": 1.5564887854921702e-05, "reference_active": true, "delta_db": -0.4084860688148344}], "active_band_count": 3, "active_band_mae_db": 0.30897442245806034}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022747909354312384, "log_rms_mae_db": 0.3793257076699182, "rms_envelope_nrmse": 0.024248039960031578, "envelope_correlation": 0.9996207172222924, "rms_gain_error_db": -0.11371431209788635, "reference_mean_power": 0.005416920568028677, "output_mean_power": 0.005276926390022926}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.96418119219355e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000410225191407457, "output_power": 0.000402511910526263, "reference_active": true, "delta_db": -0.08243590440961213}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7757195967845804e-05, "output_power": 2.488721712889887e-05, "reference_active": true, "delta_db": -0.4739925474646077}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0255605769228834e-06, "output_power": 8.924573935719084e-07, "reference_active": true, "delta_db": -0.6037382618055155}], "active_band_count": 3, "active_band_mae_db": 0.3867222378932451}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.442623236865579e-05, "log_rms_mae_db": 0.9316411656542752, "rms_envelope_nrmse": 0.08298830095657599, "envelope_correlation": 0.9943338132957726, "rms_gain_error_db": -0.07901892963380072, "reference_mean_power": 0.0007990709247977912, "output_mean_power": 0.0007846634720492786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.944736222458957e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022092122499658314, "output_power": 0.0002113380280427489, "reference_active": true, "delta_db": -0.1925979191607165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3665816077667355e-05, "output_power": 1.9914594122296363e-05, "reference_active": true, "delta_db": -0.7495002549463775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5979447501439136e-06, "output_power": 4.373048973874854e-06, "reference_active": true, "delta_db": -1.0724426676308605}], "active_band_count": 3, "active_band_mae_db": 0.6715136139126514}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.226666373654229e-05, "log_rms_mae_db": 0.6440807882740524, "rms_envelope_nrmse": 0.05129650223325181, "envelope_correlation": 0.9978019000733274, "rms_gain_error_db": -0.08170856173151284, "reference_mean_power": 0.0008012273487091395, "output_mean_power": 0.0007862939040557417}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.73006991688152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002092887788100005, "output_power": 0.00019981546014655413, "reference_active": true, "delta_db": -0.2011685643060233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0995901908054441e-05, "output_power": 9.340298332314858e-06, "reference_active": true, "delta_db": -0.7087010898067001}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1412698290887665e-07, "output_power": 3.179884594313197e-07, "reference_active": true, "delta_db": -1.1472216969372087}], "active_band_count": 3, "active_band_mae_db": 0.685697117016644}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9464514079665034e-05, "log_rms_mae_db": 1.3820798041741684, "rms_envelope_nrmse": 0.12916010506731457, "envelope_correlation": 0.9885232225286538, "rms_gain_error_db": -0.27171985112108266, "reference_mean_power": 7.576176061225114e-05, "output_mean_power": 7.116690408469081e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.472428410826198e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.044189308875811e-05, "output_power": 2.7384294641612423e-05, "reference_active": true, "delta_db": -0.4597009744037242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3472076435688925e-06, "output_power": 3.7202794325642354e-06, "reference_active": true, "delta_db": 0.4589290665060925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0111506850143103e-06, "output_power": 1.1253190847791257e-06, "reference_active": true, "delta_db": 0.4645980391135074}], "active_band_count": 3, "active_band_mae_db": 0.46107602667444136}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.608084029396227e-05, "log_rms_mae_db": 0.893380619566842, "rms_envelope_nrmse": 0.07022717404543476, "envelope_correlation": 0.9969374378702102, "rms_gain_error_db": -0.27155764617795725, "reference_mean_power": 7.611087404608628e-05, "output_mean_power": 7.149751453878243e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.164443748057941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.887302600627445e-05, "output_power": 2.6141298872189203e-05, "reference_active": true, "delta_db": -0.43165139633993216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.617498060844069e-06, "output_power": 1.739017808443213e-06, "reference_active": true, "delta_db": 0.31460260743779195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.602390292027545e-08, "output_power": 6.39768635473282e-08, "reference_active": true, "delta_db": -1.2859619761381063}], "active_band_count": 3, "active_band_mae_db": 0.6774053266386101}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.198647979957052e-05, "log_rms_mae_db": 1.3534989014770165, "rms_envelope_nrmse": 0.22055958350163715, "envelope_correlation": 0.9715967512764275, "rms_gain_error_db": 0.10492881700735257, "reference_mean_power": 5.064740774931045e-05, "output_mean_power": 5.188598954129129e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.700608322524163e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.136344717386399e-05, "output_power": 4.181428994343222e-05, "reference_active": true, "delta_db": 0.04708001234458618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.648317567219929e-06, "output_power": 7.273747524998171e-06, "reference_active": true, "delta_db": 0.39046466403938285}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.888415633999808e-06, "output_power": 5.62604060341485e-06, "reference_active": true, "delta_db": 1.6043018155284707}], "active_band_count": 3, "active_band_mae_db": 0.6806154973041466}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.0010568231240365e-05, "log_rms_mae_db": 1.0132287719008588, "rms_envelope_nrmse": 0.12146458738768316, "envelope_correlation": 0.9911428336622338, "rms_gain_error_db": 0.10490594038197619, "reference_mean_power": 5.089623620868825e-05, "output_mean_power": 5.214062844442251e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.379392951970345e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.821321246572876e-05, "output_power": 4.045630197990125e-05, "reference_active": true, "delta_db": 0.24772633354277707}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.5895785128125065e-06, "output_power": 3.385947386248678e-06, "reference_active": true, "delta_db": -0.25363251463981623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2481135158983729e-07, "output_power": 3.4032124898245514e-07, "reference_active": true, "delta_db": 1.80070829777245}], "active_band_count": 3, "active_band_mae_db": 0.7673557153183478}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9598849068815002, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4a2d09d334031d0be0224d00d9542abc8188d29220fa67071c3896e7c0761d12", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9880436573775995, "standard_deviation": 0.0020193658354429197, "minimum": 0.9853476423279406, "maximum": 0.9908120258565704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3800201500475968, "standard_deviation": 0.02928840780581407, "minimum": 1.3260249540103572, "maximum": 1.4184525885537085, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5456169222552678, "standard_deviation": 0.3435802703675689, "minimum": 0.14551841912149235, "maximum": 1.108692036315733, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1306069060214943, "standard_deviation": 0.010829630206790737, "minimum": 0.1138799936896637, "maximum": 0.14448654694010174, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960187212063234, "standard_deviation": 0.001033631703465581, "minimum": 0.9946200466616709, "maximum": 0.997550761622319, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9230283718528906, "standard_deviation": 0.04172249833867567, "minimum": 0.8611167899411384, "maximum": 0.9771697353669174, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8758338903636234, "standard_deviation": 0.4247056045296576, "minimum": 0.33403012801986404, "maximum": 1.5023767197886182, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07606424060745992, "standard_deviation": 0.009645519364037708, "minimum": 0.05742953249783038, "maximum": 0.08615327911620754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9781190945407849, "standard_deviation": 0.003997245584704761, "minimum": 0.9715967512764275, "maximum": 0.9834301727181283, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.327860192173585, "standard_deviation": 0.03955048216421033, "minimum": 1.2402073056545138, "maximum": 1.3560093823132924, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.925737862379607, "standard_deviation": 0.3062979534671827, "minimum": 0.553411481874698, "maximum": 1.3507260680820832, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19532072566369654, "standard_deviation": 0.022056474710106008, "minimum": 0.16668877667955728, "maximum": 0.23002477483359215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927571636803356, "standard_deviation": 0.0017589920294948263, "minimum": 0.989922773350234, "maximum": 0.9954180432125562, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0117780426483605, "standard_deviation": 0.04930880988186822, "minimum": 0.9152870859056256, "maximum": 1.0696161562852162, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8968087980911357, "standard_deviation": 0.31008166517573793, "minimum": 0.3922096043461039, "maximum": 1.4158236355497602, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11442796764616608, "standard_deviation": 0.02241302615924496, "minimum": 0.0860874472426802, "maximum": 0.16110224180922117, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9950476987596457, "standard_deviation": 0.0007220464861126268, "minimum": 0.9943338132957726, "maximum": 0.9961289633556064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9201702024428586, "standard_deviation": 0.030470636821522817, "minimum": 0.8653341252951043, "maximum": 0.9641284783734688, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6348727240051816, "standard_deviation": 0.4134670631945831, "minimum": 0.05206517605595703, "maximum": 1.2936152742110827, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07753232589470174, "standard_deviation": 0.005826699668547904, "minimum": 0.06854166759452461, "maximum": 0.08298830095657599, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9981338458543614, "standard_deviation": 0.00035007319853945116, "minimum": 0.9975987791313166, "maximum": 0.9986356764119901, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.6487065958743623, "standard_deviation": 0.028509011591717696, "minimum": 0.6111246811414368, "maximum": 0.6910709769213432, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8205700165724005, "standard_deviation": 0.5083789418413139, "minimum": 0.07128682188773411, "maximum": 1.5570215413556385, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.046927393202260984, "standard_deviation": 0.004655788361241069, "minimum": 0.04018811849344713, "maximum": 0.05340037477291673, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.1348020248255335, "standard_deviation": 0.03862438774361831, "minimum": 0.0899441455673916, "maximum": 0.1907995709429868, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.035774208625475, "standard_deviation": 0.007329276341719791, "minimum": 5.022835540131183, "maximum": 5.04523888892473, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.3630190020015608, "standard_deviation": 0.004724120696277418, "minimum": 0.35541140800967214, "maximum": 0.3702920377690389, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07804503240149459, "standard_deviation": 0.026061985314242694, "minimum": -0.11927322137644256, "maximum": -0.04339337802227826, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9987881676650905, "standard_deviation": 0.00012749657683224462, "minimum": 0.9985550430776056, "maximum": 0.9989486259560909, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.668788264334049, "standard_deviation": 0.013492245660149194, "minimum": 0.6471117687502083, "maximum": 0.6920043878481531, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35943515196378184, "standard_deviation": 0.24070598421241174, "minimum": 0.09914516407653506, "maximum": 0.8890978685681127, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.03829371623690386, "standard_deviation": 0.0021952379227295283, "minimum": 0.03610857893444671, "maximum": 0.04271343711973771, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9996815988020118, "standard_deviation": 4.9651552826903764e-05, "minimum": 0.9995976806712304, "maximum": 0.9997479178209123, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3784334701561435, "standard_deviation": 0.011472273757082598, "minimum": 0.36150497227516737, "maximum": 0.395898479819733, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.41547664337777057, "standard_deviation": 0.20633650151083818, "minimum": 0.17539622114572526, "maximum": 0.7731133354755921, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.020757824617034376, "standard_deviation": 0.0018138223922924794, "minimum": 0.01896210488661957, "maximum": 0.024248039960031578, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-292519-0001", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.7219724655151367, "predicted_mos": 2.7219724655151367, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "a8978095215ffc8250826e3773bc2553676d62331d65972fb88b20a47e8ad614", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9476287998259068, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6374014985585941, "全帯域_10ms_rms_envelope_nrmse": 0.03828067769287547, "全帯域_10ms_envelope_correlation": 0.9987884472628831, "全帯域_10ms_modulation_mae_db": 0.3494616708750365, "全帯域_40ms_log_rms_mae_db": 0.36348848715503185, "全帯域_40ms_rms_envelope_nrmse": 0.020360118299348102, "全帯域_40ms_envelope_correlation": 0.9996877993317237, "全帯域_40ms_modulation_mae_db": 0.3118582830647805, "300–1000 Hz_10ms_log_rms_mae_db": 0.9662243976400913, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07891772128015563, "300–1000 Hz_10ms_envelope_correlation": 0.994873594857182, "300–1000 Hz_10ms_modulation_mae_db": 0.9165184789696456, "300–1000 Hz_40ms_log_rms_mae_db": 0.6766808359360472, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05211268303792586, "300–1000 Hz_40ms_envelope_correlation": 0.9976875407180542, "300–1000 Hz_40ms_modulation_mae_db": 1.3620192040352528, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5153902824274672, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12835739828430118, "1000–3000 Hz_10ms_envelope_correlation": 0.988562059430135, "1000–3000 Hz_10ms_modulation_mae_db": 0.7296485611973328, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9863665454518724, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0789280337819506, "1000–3000 Hz_40ms_envelope_correlation": 0.9957790830453703, "1000–3000 Hz_40ms_modulation_mae_db": 1.24167727261517, "3000–6000 Hz_10ms_log_rms_mae_db": 1.417153063414101, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18921947339162556, "3000–6000 Hz_10ms_envelope_correlation": 0.9791094123646624, "3000–6000 Hz_10ms_modulation_mae_db": 0.5945993596202265, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0401292097273889, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10057116741540777, "3000–6000 Hz_40ms_envelope_correlation": 0.9939537314811274, "3000–6000 Hz_40ms_modulation_mae_db": 0.5142415881731538, "mr_stft_db": 5.127497124482324, "band_mae_db": 0.11575018816002969, "rms_gain_error_db": -0.08213013422134707, "output_peak": 0.3588960418069398}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6603237914873713, "全帯域_10ms_rms_envelope_nrmse": 0.0353907541860454, "全帯域_10ms_envelope_correlation": 0.9989788979517925, "全帯域_10ms_modulation_mae_db": 0.12030822734415293, "全帯域_40ms_log_rms_mae_db": 0.3524652818628026, "全帯域_40ms_rms_envelope_nrmse": 0.019664731688497923, "全帯域_40ms_envelope_correlation": 0.9997312921472252, "全帯域_40ms_modulation_mae_db": 0.08412542842738198, "300–1000 Hz_10ms_log_rms_mae_db": 0.9828843654655642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07190463054272662, "300–1000 Hz_10ms_envelope_correlation": 0.9957389243197216, "300–1000 Hz_10ms_modulation_mae_db": 0.3621685424613699, "300–1000 Hz_40ms_log_rms_mae_db": 0.6964315521783533, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04180028587995138, "300–1000 Hz_40ms_envelope_correlation": 0.9985199583358416, "300–1000 Hz_40ms_modulation_mae_db": 0.2907744398112167, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5285478170815399, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14179094799005257, "1000–3000 Hz_10ms_envelope_correlation": 0.9864140198402719, "1000–3000 Hz_10ms_modulation_mae_db": 0.36628318587947034, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9834992559407764, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0831412640601188, "1000–3000 Hz_40ms_envelope_correlation": 0.9959185135052059, "1000–3000 Hz_40ms_modulation_mae_db": 0.6233342075427518, "3000–6000 Hz_10ms_log_rms_mae_db": 1.503421040048642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18558000229827146, "3000–6000 Hz_10ms_envelope_correlation": 0.9795169724489281, "3000–6000 Hz_10ms_modulation_mae_db": 0.4406007310651207, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1780638353566368, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11325957776983489, "3000–6000 Hz_40ms_envelope_correlation": 0.9925541773261953, "3000–6000 Hz_40ms_modulation_mae_db": 0.37793986170197186, "mr_stft_db": 5.1370003391196875, "band_mae_db": 0.15033045326205421, "rms_gain_error_db": -0.08338389805796784, "output_peak": 0.3625675854892713}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.672877915245448, "全帯域_10ms_rms_envelope_nrmse": 0.0358567794393038, "全帯域_10ms_envelope_correlation": 0.998955802531979, "全帯域_10ms_modulation_mae_db": 0.3288531988374828, "全帯域_40ms_log_rms_mae_db": 0.38845244131730766, "全帯域_40ms_rms_envelope_nrmse": 0.020422159762041044, "全帯域_40ms_envelope_correlation": 0.9996952657601686, "全帯域_40ms_modulation_mae_db": 0.40017686221154425, "300–1000 Hz_10ms_log_rms_mae_db": 0.9552509665279757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07222104899444665, "300–1000 Hz_10ms_envelope_correlation": 0.9956966809654664, "300–1000 Hz_10ms_modulation_mae_db": 0.9254511496231975, "300–1000 Hz_40ms_log_rms_mae_db": 0.7006435595015489, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.044735924037555594, "300–1000 Hz_40ms_envelope_correlation": 0.9982994861037092, "300–1000 Hz_40ms_modulation_mae_db": 1.406095298098444, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5575571744933172, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1496041268219665, "1000–3000 Hz_10ms_envelope_correlation": 0.9841789862826934, "1000–3000 Hz_10ms_modulation_mae_db": 1.1221934667037103, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9888563220955641, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08332569203677936, "1000–3000 Hz_40ms_envelope_correlation": 0.994822649795625, "1000–3000 Hz_40ms_modulation_mae_db": 1.3189781998719712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.44929241600382, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18503656681902544, "3000–6000 Hz_10ms_envelope_correlation": 0.9807701727445669, "3000–6000 Hz_10ms_modulation_mae_db": 0.4656076701937741, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0453061134839452, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10424900088536088, "3000–6000 Hz_40ms_envelope_correlation": 0.9940438843671136, "3000–6000 Hz_40ms_modulation_mae_db": 0.4676814237420277, "mr_stft_db": 5.142432563222143, "band_mae_db": 0.11626299376108999, "rms_gain_error_db": -0.08076122105303468, "output_peak": 0.3669966438435257}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6579124852013101, "全帯域_10ms_rms_envelope_nrmse": 0.040961795730435284, "全帯域_10ms_envelope_correlation": 0.9986170541847851, "全帯域_10ms_modulation_mae_db": 0.8541730153075496, "全帯域_40ms_log_rms_mae_db": 0.3801116180290117, "全帯域_40ms_rms_envelope_nrmse": 0.02238929003414156, "全帯域_40ms_envelope_correlation": 0.9996350919863692, "全帯域_40ms_modulation_mae_db": 0.8288069253037756, "300–1000 Hz_10ms_log_rms_mae_db": 1.004024952874275, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08637490295445843, "300–1000 Hz_10ms_envelope_correlation": 0.9938526409630586, "300–1000 Hz_10ms_modulation_mae_db": 0.7819669476982639, "300–1000 Hz_40ms_log_rms_mae_db": 0.7764851060886248, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056478722554715324, "300–1000 Hz_40ms_envelope_correlation": 0.9972903017627586, "300–1000 Hz_40ms_modulation_mae_db": 0.7488769708736832, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4920086620023467, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1397511697222558, "1000–3000 Hz_10ms_envelope_correlation": 0.98615528074503, "1000–3000 Hz_10ms_modulation_mae_db": 1.063741898973873, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9268905893688043, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07236861389382311, "1000–3000 Hz_40ms_envelope_correlation": 0.9962486474429966, "1000–3000 Hz_40ms_modulation_mae_db": 1.6207189883277886, "3000–6000 Hz_10ms_log_rms_mae_db": 1.415282104628687, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16100918516628263, "3000–6000 Hz_10ms_envelope_correlation": 0.9844956082310139, "3000–6000 Hz_10ms_modulation_mae_db": 1.066419138616409, "3000–6000 Hz_40ms_log_rms_mae_db": 1.030200221525841, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08895087541982279, "3000–6000 Hz_40ms_envelope_correlation": 0.9951831850384985, "3000–6000 Hz_40ms_modulation_mae_db": 1.256402400125544, "mr_stft_db": 5.133391171308092, "band_mae_db": 0.1720715087134469, "rms_gain_error_db": -0.08463310336170786, "output_peak": 0.36618850853877516}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6764716648468923, "全帯域_10ms_rms_envelope_nrmse": 0.03772056952658454, "全帯域_10ms_envelope_correlation": 0.9987956815143951, "全帯域_10ms_modulation_mae_db": 0.4145757120644384, "全帯域_40ms_log_rms_mae_db": 0.3830719342478977, "全帯域_40ms_rms_envelope_nrmse": 0.021465495215692208, "全帯域_40ms_envelope_correlation": 0.9996218611795605, "全帯域_40ms_modulation_mae_db": 0.5324458767072199, "300–1000 Hz_10ms_log_rms_mae_db": 0.9646414620467061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07735386601399284, "300–1000 Hz_10ms_envelope_correlation": 0.9951146178721093, "300–1000 Hz_10ms_modulation_mae_db": 0.8165933393346373, "300–1000 Hz_40ms_log_rms_mae_db": 0.6920321800965185, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04774311362055785, "300–1000 Hz_40ms_envelope_correlation": 0.998084236537436, "300–1000 Hz_40ms_modulation_mae_db": 0.957453158208862, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5308188926587445, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1335097029345191, "1000–3000 Hz_10ms_envelope_correlation": 0.9877819888610507, "1000–3000 Hz_10ms_modulation_mae_db": 0.7973400197264553, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9734786264895259, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07376881612244596, "1000–3000 Hz_40ms_envelope_correlation": 0.9966429637029532, "1000–3000 Hz_40ms_modulation_mae_db": 1.583965797939569, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4777596716726904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18722267179002738, "3000–6000 Hz_10ms_envelope_correlation": 0.9795467974370176, "3000–6000 Hz_10ms_modulation_mae_db": 0.8339775688615315, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1150955837062242, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10154175736784858, "3000–6000 Hz_40ms_envelope_correlation": 0.9938351336245239, "3000–6000 Hz_40ms_modulation_mae_db": 0.8369865380678917, "mr_stft_db": 5.120024443187551, "band_mae_db": 0.08230246061111025, "rms_gain_error_db": -0.055311706987824755, "output_peak": 0.3621940408215885}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6762849344827581, "全帯域_10ms_rms_envelope_nrmse": 0.037525114975969116, "全帯域_10ms_envelope_correlation": 0.9988458272096171, "全帯域_10ms_modulation_mae_db": 0.2417710400840453, "全帯域_40ms_log_rms_mae_db": 0.40372177019022903, "全帯域_40ms_rms_envelope_nrmse": 0.021496245406258614, "全帯域_40ms_envelope_correlation": 0.9996563821632813, "全帯域_40ms_modulation_mae_db": 0.16459252591536483, "300–1000 Hz_10ms_log_rms_mae_db": 1.0577800647694664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0692287802742882, "300–1000 Hz_10ms_envelope_correlation": 0.9960573171178629, "300–1000 Hz_10ms_modulation_mae_db": 0.2903811045860884, "300–1000 Hz_40ms_log_rms_mae_db": 0.7592334782537725, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04482958598297192, "300–1000 Hz_40ms_envelope_correlation": 0.9982967551205667, "300–1000 Hz_40ms_modulation_mae_db": 0.2345347333050126, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5128826678638507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14476425218203975, "1000–3000 Hz_10ms_envelope_correlation": 0.9852017221887567, "1000–3000 Hz_10ms_modulation_mae_db": 0.4933264224520177, "1000–3000 Hz_40ms_log_rms_mae_db": 0.995087441643147, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08059520427399243, "1000–3000 Hz_40ms_envelope_correlation": 0.9954201142903313, "1000–3000 Hz_40ms_modulation_mae_db": 0.9996874638087201, "3000–6000 Hz_10ms_log_rms_mae_db": 1.451470693976245, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19401021974348512, "3000–6000 Hz_10ms_envelope_correlation": 0.9786932871511227, "3000–6000 Hz_10ms_modulation_mae_db": 0.6695493695259999, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0829764816810767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13879325674443377, "3000–6000 Hz_40ms_envelope_correlation": 0.9900491719381292, "3000–6000 Hz_40ms_modulation_mae_db": 0.45188007738321384, "mr_stft_db": 5.158673860791656, "band_mae_db": 0.17396316638586629, "rms_gain_error_db": -0.08934842761536882, "output_peak": 0.3565980498726672}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6243896731276296, "全帯域_10ms_rms_envelope_nrmse": 0.03680900237778172, "全帯域_10ms_envelope_correlation": 0.9988529745787839, "全帯域_10ms_modulation_mae_db": 0.43522100146655146, "全帯域_40ms_log_rms_mae_db": 0.34875065235486263, "全帯域_40ms_rms_envelope_nrmse": 0.019896998227532294, "全帯域_40ms_envelope_correlation": 0.999673465125465, "全帯域_40ms_modulation_mae_db": 0.5649307334782542, "300–1000 Hz_10ms_log_rms_mae_db": 1.0173185549495694, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06936937469245938, "300–1000 Hz_10ms_envelope_correlation": 0.9960393418646973, "300–1000 Hz_10ms_modulation_mae_db": 0.8598545065197071, "300–1000 Hz_40ms_log_rms_mae_db": 0.6948445018709695, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.043809898681392775, "300–1000 Hz_40ms_envelope_correlation": 0.9983596739265415, "300–1000 Hz_40ms_modulation_mae_db": 0.9357464046427486, "1000–3000 Hz_10ms_log_rms_mae_db": 1.439057615307192, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1364305784797077, "1000–3000 Hz_10ms_envelope_correlation": 0.9867738188729324, "1000–3000 Hz_10ms_modulation_mae_db": 0.24709358898218206, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9251023515672026, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07931124818315162, "1000–3000 Hz_40ms_envelope_correlation": 0.9953374313693902, "1000–3000 Hz_40ms_modulation_mae_db": 0.6657035901357875, "3000–6000 Hz_10ms_log_rms_mae_db": 1.401596927960402, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18815993302023817, "3000–6000 Hz_10ms_envelope_correlation": 0.9798984379955947, "3000–6000 Hz_10ms_modulation_mae_db": 0.3873381643355797, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0543977933709783, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11990088366723002, "3000–6000 Hz_40ms_envelope_correlation": 0.991821139121583, "3000–6000 Hz_40ms_modulation_mae_db": 1.1901776477324721, "mr_stft_db": 5.128705893300199, "band_mae_db": 0.12074549933652352, "rms_gain_error_db": -0.05262233543569615, "output_peak": 0.3644296482627065}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6493822055020342, "全帯域_10ms_rms_envelope_nrmse": 0.03709113583938725, "全帯域_10ms_envelope_correlation": 0.9988688254496206, "全帯域_10ms_modulation_mae_db": 0.5548669086968471, "全帯域_40ms_log_rms_mae_db": 0.3621869570042753, "全帯域_40ms_rms_envelope_nrmse": 0.019046904007891944, "全帯域_40ms_envelope_correlation": 0.999741993915714, "全帯域_40ms_modulation_mae_db": 0.5583052928649458, "300–1000 Hz_10ms_log_rms_mae_db": 0.9936756140283118, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08325094636714726, "300–1000 Hz_10ms_envelope_correlation": 0.9942595235194697, "300–1000 Hz_10ms_modulation_mae_db": 0.37458579272483766, "300–1000 Hz_40ms_log_rms_mae_db": 0.7157304009945451, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04876852979134469, "300–1000 Hz_40ms_envelope_correlation": 0.9979702038297967, "300–1000 Hz_40ms_modulation_mae_db": 0.570698922865614, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4757740286013867, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12541667711544213, "1000–3000 Hz_10ms_envelope_correlation": 0.9888694212228235, "1000–3000 Hz_10ms_modulation_mae_db": 0.4963953746797893, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9315165548210127, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07216553740752198, "1000–3000 Hz_40ms_envelope_correlation": 0.9961068520224411, "1000–3000 Hz_40ms_modulation_mae_db": 1.0677724705440208, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4180703856793422, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18478190391851945, "3000–6000 Hz_10ms_envelope_correlation": 0.9800285484949065, "3000–6000 Hz_10ms_modulation_mae_db": 0.6058384650812778, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0251839866314494, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09491868404483622, "3000–6000 Hz_40ms_envelope_correlation": 0.9945962946197614, "3000–6000 Hz_40ms_modulation_mae_db": 0.8780318854145471, "mr_stft_db": 5.145460735105597, "band_mae_db": 0.09153918522306104, "rms_gain_error_db": -0.07604910163767072, "output_peak": 0.35877366702394997}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.298571567755372, "spectral_convergence": 0.11765552637920834, "spectral_convergence_status": "ok", "amplitude_floor": 1.7050603882585003e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1027264845798355, "spectral_convergence": 0.11300679169951228, "spectral_convergence_status": "ok", "amplitude_floor": 1.5194442893884604e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.981193321111764, "spectral_convergence": 0.11341217223240344, "spectral_convergence_status": "ok", "amplitude_floor": 1.5327487413055003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00047003850181382925, "output_power": 0.00046737328322888976, "delta_db": -0.02469550440658188, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.004513086919531444, "output_power": 0.004423077420750368, "delta_db": -0.08749156621152326, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003251325154635926, "output_power": 0.0003167148024663781, "delta_db": -0.11392042374973148, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.51153215194187e-05, "output_power": 8.129403404765497e-05, "delta_db": -0.1994906896442074, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6288307106843834e-05, "output_power": 2.6173669911618944e-05, "delta_db": -0.018979986412187642, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.4880972570108e-05, "output_power": 4.753949405238196e-05, "delta_db": 0.24992295853594648, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11575018816002969, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08213013422134707, "pitch": null, "output_rms_dbfs": -22.706592024036954, "output_peak": 0.3588960418069398, "reference_rms_dbfs": -22.624461889815606, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.127497124482324}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023914198234423542, "log_rms_mae_db": 0.6374014985585941, "rms_envelope_nrmse": 0.03828067769287547, "envelope_correlation": 0.9987884472628831, "rms_gain_error_db": -0.07701348656229134, "reference_mean_power": 0.005456771858924342, "output_mean_power": 0.005360859783925909}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.071388012833875e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045567084702756786, "output_power": 0.00044670908943797536, "reference_active": true, "delta_db": -0.0862645472604507}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.44246106516196e-05, "output_power": 6.38458127724738e-05, "reference_active": true, "delta_db": -0.039193825732732196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7099940640562705e-05, "output_power": 1.3826171477102536e-05, "reference_active": true, "delta_db": -0.9229266396319267}], "active_band_count": 3, "active_band_mae_db": 0.3494616708750365}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022747909354312384, "log_rms_mae_db": 0.36348848715503185, "rms_envelope_nrmse": 0.020360118299348102, "envelope_correlation": 0.9996877993317237, "rms_gain_error_db": -0.07143801784950439, "reference_mean_power": 0.005416920568028677, "output_mean_power": 0.0053285453400289465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.96418119219355e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000410225191407457, "output_power": 0.00040263742054069365, "reference_active": true, "delta_db": -0.08108191192575512}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7757195967845804e-05, "output_power": 2.7784774454598382e-05, "reference_active": true, "delta_db": 0.004312841445523302}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0255605769228834e-06, "output_power": 8.432246783819715e-07, "reference_active": true, "delta_db": -0.850180095823063}], "active_band_count": 3, "active_band_mae_db": 0.3118582830647805}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.442623236865579e-05, "log_rms_mae_db": 0.9662243976400913, "rms_envelope_nrmse": 0.07891772128015563, "envelope_correlation": 0.994873594857182, "rms_gain_error_db": -0.03229580490498805, "reference_mean_power": 0.0007990709247977912, "output_mean_power": 0.0007931507662103516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.944736222458957e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022092122499658314, "output_power": 0.00021465983720799772, "reference_active": true, "delta_db": -0.12486647030586878}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3665816077667355e-05, "output_power": 2.155528002339243e-05, "reference_active": true, "delta_db": -0.4056781588911712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5979447501439136e-06, "output_power": 3.3583621533516023e-06, "reference_active": true, "delta_db": -2.219010807711897}], "active_band_count": 3, "active_band_mae_db": 0.9165184789696456}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.226666373654229e-05, "log_rms_mae_db": 0.6766808359360472, "rms_envelope_nrmse": 0.05211268303792586, "envelope_correlation": 0.9976875407180542, "rms_gain_error_db": -0.03413972457025191, "reference_mean_power": 0.0008012273487091395, "output_mean_power": 0.0007949536219684475}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.73006991688152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002092887788100005, "output_power": 0.00020224130980756258, "reference_active": true, "delta_db": -0.14876074605237632}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0995901908054441e-05, "output_power": 9.972022836305826e-06, "reference_active": true, "delta_db": -0.4244759256268931}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1412698290887665e-07, "output_power": 1.8443843748276891e-07, "reference_active": true, "delta_db": -3.5128209404264887}], "active_band_count": 3, "active_band_mae_db": 1.3620192040352528}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9464514079665034e-05, "log_rms_mae_db": 1.5153902824274672, "rms_envelope_nrmse": 0.12835739828430118, "envelope_correlation": 0.988562059430135, "rms_gain_error_db": -0.23728697268942764, "reference_mean_power": 7.576176061225114e-05, "output_mean_power": 7.173339098677749e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.472428410826198e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.044189308875811e-05, "output_power": 2.7955840015100738e-05, "reference_active": true, "delta_db": -0.36999109950890596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3472076435688925e-06, "output_power": 2.8697013177029364e-06, "reference_active": true, "delta_db": -0.6684595744847717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0111506850143103e-06, "output_power": 7.758296613836246e-07, "reference_active": true, "delta_db": -1.1504950095983206}], "active_band_count": 3, "active_band_mae_db": 0.7296485611973328}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.608084029396227e-05, "log_rms_mae_db": 0.9863665454518724, "rms_envelope_nrmse": 0.0789280337819506, "envelope_correlation": 0.9957790830453703, "rms_gain_error_db": -0.23834389889340504, "reference_mean_power": 7.611087404608628e-05, "output_mean_power": 7.204640572358581e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.164443748057941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.887302600627445e-05, "output_power": 2.653193327458442e-05, "reference_active": true, "delta_db": -0.3672340561332018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.617498060844069e-06, "output_power": 1.320141023349412e-06, "reference_active": true, "delta_db": -0.882234417255032}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.602390292027545e-08, "output_power": 4.864775556288713e-08, "reference_active": true, "delta_db": -2.4755633444572758}], "active_band_count": 3, "active_band_mae_db": 1.24167727261517}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.198647979957052e-05, "log_rms_mae_db": 1.417153063414101, "rms_envelope_nrmse": 0.18921947339162556, "envelope_correlation": 0.9791094123646624, "rms_gain_error_db": 0.12124585960479774, "reference_mean_power": 5.064740774931045e-05, "output_mean_power": 5.208129903191258e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.700608322524163e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.136344717386399e-05, "output_power": 4.239388403230058e-05, "reference_active": true, "delta_db": 0.10686482364486716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.648317567219929e-06, "output_power": 6.588728417751613e-06, "reference_active": true, "delta_db": -0.039101493858102944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.888415633999808e-06, "output_power": 2.6667942157659222e-06, "reference_active": true, "delta_db": -1.6378317613577091}], "active_band_count": 3, "active_band_mae_db": 0.5945993596202265}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.0010568231240365e-05, "log_rms_mae_db": 1.0401292097273889, "rms_envelope_nrmse": 0.10057116741540777, "envelope_correlation": 0.9939537314811274, "rms_gain_error_db": 0.12074169527669627, "reference_mean_power": 5.089623620868825e-05, "output_mean_power": 5.233109676370268e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.379392951970345e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.821321246572876e-05, "output_power": 3.911274490839524e-05, "reference_active": true, "delta_db": 0.10104746534153769}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.5895785128125065e-06, "output_power": 3.4045004934834044e-06, "reference_active": true, "delta_db": -0.2299005538411073}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2481135158983729e-07, "output_power": 1.700750323014371e-07, "reference_active": true, "delta_db": -1.2117767453368165}], "active_band_count": 3, "active_band_mae_db": 0.5142415881731538}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9554428547673375, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9db56d6692cfe486c52adf2451a60a8865ef55ba536dfabdb035bccd88066833", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9867421621804617, "standard_deviation": 0.0016175127277108217, "minimum": 0.9841789862826934, "maximum": 0.9888694212228235, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5065046425544806, "standard_deviation": 0.03685237866354051, "minimum": 1.439057615307192, "maximum": 1.5575571744933172, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6645028148243539, "standard_deviation": 0.3188123380052786, "minimum": 0.24709358898218206, "maximum": 1.1221934667037103, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13745310669128558, "standard_deviation": 0.008188856931589298, "minimum": 0.12541667711544213, "maximum": 0.1496041268219665, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957845318967893, "standard_deviation": 0.0005771971290987006, "minimum": 0.994822649795625, "maximum": 0.9966429637029532, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9638497109222381, "standard_deviation": 0.030471819302183224, "minimum": 0.9251023515672026, "maximum": 0.995087441643147, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1402297488482223, "standard_deviation": 0.37557051198864994, "minimum": 0.6233342075427518, "maximum": 1.6207189883277886, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07795055121997298, "standard_deviation": 0.004594221748166222, "minimum": 0.07216553740752198, "maximum": 0.08332569203677936, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9802574046084765, "standard_deviation": 0.0018214627247917113, "minimum": 0.9786932871511227, "maximum": 0.9844956082310139, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4417557879229912, "standard_deviation": 0.03531479940801834, "minimum": 1.401596927960402, "maximum": 1.503421040048642, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6329913084124899, "standard_deviation": 0.2258886435962396, "minimum": 0.3873381643355797, "maximum": 1.066419138616409, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18437749451843438, "standard_deviation": 0.00990496166860104, "minimum": 0.16100918516628263, "maximum": 0.19401021974348512, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9932545896896166, "standard_deviation": 0.0016813485108092222, "minimum": 0.9900491719381292, "maximum": 0.9951831850384985, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0714191531854427, "standard_deviation": 0.0523745572198341, "minimum": 1.0251839866314494, "maximum": 1.1780638353566368, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7466676777926028, "standard_deviation": 0.3457207339029013, "minimum": 0.37793986170197186, "maximum": 1.256402400125544, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10777315041434686, "standard_deviation": 0.015865792428294126, "minimum": 0.08895087541982279, "maximum": 0.13879325674443377, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.995204080184946, "standard_deviation": 0.0008272319105533688, "minimum": 0.9938526409630586, "maximum": 0.9960573171178629, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9927250472877449, "standard_deviation": 0.033716687901448, "minimum": 0.9552509665279757, "maximum": 1.0577800647694664, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6659399827397183, "standard_deviation": 0.27310541774762315, "minimum": 0.2903811045860884, "maximum": 0.9254511496231975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07607765888995938, "standard_deviation": 0.006451796739751577, "minimum": 0.0692287802742882, "maximum": 0.08637490295445843, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980635195418381, "standard_deviation": 0.00040611146808930377, "minimum": 0.9972903017627586, "maximum": 0.9985199583358416, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7140102018650474, "standard_deviation": 0.03521917614998281, "minimum": 0.6766808359360472, "maximum": 0.7764851060886248, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8132748914801042, "standard_deviation": 0.4406618536124988, "minimum": 0.2345347333050126, "maximum": 1.406095298098444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.04753484294830192, "standard_deviation": 0.004846614121203626, "minimum": 0.04180028587995138, "maximum": 0.056478722554715324, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12787068193164772, "standard_deviation": 0.03445810086782678, "minimum": 0.08230246061111025, "maximum": 0.17396316638586629, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.136648266314657, "standard_deviation": 0.01213114829925643, "minimum": 5.120024443187551, "maximum": 5.158673860791656, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36208052320742806, "standard_deviation": 0.0037406268287989053, "minimum": 0.3565980498726672, "maximum": 0.3669966438435257, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.07552999104632724, "standard_deviation": 0.013836992485735026, "minimum": -0.08934842761536882, "maximum": -0.05262233543569615, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.998837938835482, "standard_deviation": 0.00011214068475186478, "minimum": 0.9986170541847851, "maximum": 0.9989788979517925, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6568805210565047, "standard_deviation": 0.018975526005474582, "minimum": 0.6243896731276296, "maximum": 0.6764716648468923, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.41240384683451303, "standard_deviation": 0.22082924745243643, "minimum": 0.12030822734415293, "maximum": 0.8541730153075496, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.037454478721047825, "standard_deviation": 0.001707961644488795, "minimum": 0.0353907541860454, "maximum": 0.040961795730435284, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9996803939511885, "standard_deviation": 4.2701823828371234e-05, "minimum": 0.9996218611795605, "maximum": 0.999741993915714, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.3727811427701774, "standard_deviation": 0.01909127390230758, "minimum": 0.34875065235486263, "maximum": 0.40372177019022903, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4306552409966584, "standard_deviation": 0.24175680489789464, "minimum": 0.08412542842738198, "maximum": 0.8288069253037756, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.02059274283017546, "standard_deviation": 0.001109935077807508, "minimum": 0.019046904007891944, "maximum": 0.02238929003414156, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_8555-292519-0001", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.630749225616455, "predicted_mos": 2.630749225616455, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fa49e5928a41c19a5aa21502519cda106a4ef28d341d3571cd6244813cab448e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.75075909961015, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6888493302121759, "全帯域_10ms_rms_envelope_nrmse": 0.04042110013050984, "全帯域_10ms_envelope_correlation": 0.9986786728520418, "全帯域_10ms_modulation_mae_db": 0.5563373044141372, "全帯域_40ms_log_rms_mae_db": 0.39667940439495525, "全帯域_40ms_rms_envelope_nrmse": 0.021937044039759195, "全帯域_40ms_envelope_correlation": 0.9996671621236308, "全帯域_40ms_modulation_mae_db": 0.4051160228986926, "300–1000 Hz_10ms_log_rms_mae_db": 0.9637190323994446, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07764422775025215, "300–1000 Hz_10ms_envelope_correlation": 0.9950197476245956, "300–1000 Hz_10ms_modulation_mae_db": 0.3885114720765553, "300–1000 Hz_40ms_log_rms_mae_db": 0.7049728378719549, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04222546497284757, "300–1000 Hz_40ms_envelope_correlation": 0.9984839547953053, "300–1000 Hz_40ms_modulation_mae_db": 0.374124003449993, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4651373791422928, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13229539806514629, "1000–3000 Hz_10ms_envelope_correlation": 0.9878454604208449, "1000–3000 Hz_10ms_modulation_mae_db": 1.1095095438270164, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9444330092040851, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07612482724259731, "1000–3000 Hz_40ms_envelope_correlation": 0.9961479201262444, "1000–3000 Hz_40ms_modulation_mae_db": 1.6700769926700663, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4294554113699087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18316090301919208, "3000–6000 Hz_10ms_envelope_correlation": 0.9803124208009539, "3000–6000 Hz_10ms_modulation_mae_db": 0.6342123652079282, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0550408773117372, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11212208047024871, "3000–6000 Hz_40ms_envelope_correlation": 0.9923866341656495, "3000–6000 Hz_40ms_modulation_mae_db": 0.5769916426678193, "mr_stft_db": 5.117658570387945, "band_mae_db": 0.11366026763239317, "rms_gain_error_db": -0.09891732790952575, "output_peak": 0.36369815689313867}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6488857170586951, "全帯域_10ms_rms_envelope_nrmse": 0.03543650872370175, "全帯域_10ms_envelope_correlation": 0.998957124766112, "全帯域_10ms_modulation_mae_db": 0.52734829282875, "全帯域_40ms_log_rms_mae_db": 0.3651167082268141, "全帯域_40ms_rms_envelope_nrmse": 0.018194636794491573, "全帯域_40ms_envelope_correlation": 0.9997505612771674, "全帯域_40ms_modulation_mae_db": 0.6632156085604037, "300–1000 Hz_10ms_log_rms_mae_db": 1.0404900122526317, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.0770831029873041, "300–1000 Hz_10ms_envelope_correlation": 0.9951017857700782, "300–1000 Hz_10ms_modulation_mae_db": 1.1387785846713154, "300–1000 Hz_40ms_log_rms_mae_db": 0.7442666177189026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04910134473772466, "300–1000 Hz_40ms_envelope_correlation": 0.9979516424986491, "300–1000 Hz_40ms_modulation_mae_db": 1.503020246619143, "1000–3000 Hz_10ms_log_rms_mae_db": 1.460053456067521, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13236238989410334, "1000–3000 Hz_10ms_envelope_correlation": 0.9882110730966127, "1000–3000 Hz_10ms_modulation_mae_db": 0.5025038553778002, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9332876728706467, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07870048442546122, "1000–3000 Hz_40ms_envelope_correlation": 0.9963354641061922, "1000–3000 Hz_40ms_modulation_mae_db": 0.943090801819405, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4082756313219509, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1697496152013169, "3000–6000 Hz_10ms_envelope_correlation": 0.9836231939217503, "3000–6000 Hz_10ms_modulation_mae_db": 0.2840262517139997, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0261166832277269, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09419011817382827, "3000–6000 Hz_40ms_envelope_correlation": 0.9950421833942017, "3000–6000 Hz_40ms_modulation_mae_db": 0.9459544263364048, "mr_stft_db": 5.10409925988162, "band_mae_db": 0.1629285548720345, "rms_gain_error_db": -0.06733486953386761, "output_peak": 0.36681695968900196}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6468552294740519, "全帯域_10ms_rms_envelope_nrmse": 0.037522834763644715, "全帯域_10ms_envelope_correlation": 0.9988128191198536, "全帯域_10ms_modulation_mae_db": 0.23780746314009601, "全帯域_40ms_log_rms_mae_db": 0.3524436267106287, "全帯域_40ms_rms_envelope_nrmse": 0.01720566869802357, "全帯域_40ms_envelope_correlation": 0.999769124885738, "全帯域_40ms_modulation_mae_db": 0.3460483888124181, "300–1000 Hz_10ms_log_rms_mae_db": 1.0159879251105857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06722424154744222, "300–1000 Hz_10ms_envelope_correlation": 0.9962892385809398, "300–1000 Hz_10ms_modulation_mae_db": 0.38352269269547384, "300–1000 Hz_40ms_log_rms_mae_db": 0.6970546330799539, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03969100754949866, "300–1000 Hz_40ms_envelope_correlation": 0.9986605750218909, "300–1000 Hz_40ms_modulation_mae_db": 0.450736693061081, "1000–3000 Hz_10ms_log_rms_mae_db": 1.473299899706402, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.11893484280421829, "1000–3000 Hz_10ms_envelope_correlation": 0.9899832477474252, "1000–3000 Hz_10ms_modulation_mae_db": 0.5516248673874253, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9708664280561602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06245600768583666, "1000–3000 Hz_40ms_envelope_correlation": 0.9971569653930019, "1000–3000 Hz_40ms_modulation_mae_db": 0.7424262363621571, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4839215045592495, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18317761455942289, "3000–6000 Hz_10ms_envelope_correlation": 0.9798892329092977, "3000–6000 Hz_10ms_modulation_mae_db": 0.482057135167105, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0880464800627905, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09713143110824765, "3000–6000 Hz_40ms_envelope_correlation": 0.9943485947293222, "3000–6000 Hz_40ms_modulation_mae_db": 1.1499712729467624, "mr_stft_db": 5.118350685134531, "band_mae_db": 0.09459155042793149, "rms_gain_error_db": -0.06528078279238377, "output_peak": 0.37121483931203847}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6411265870716618, "全帯域_10ms_rms_envelope_nrmse": 0.035150449205721516, "全帯域_10ms_envelope_correlation": 0.9989700382717845, "全帯域_10ms_modulation_mae_db": 0.4643850949685797, "全帯域_40ms_log_rms_mae_db": 0.36753577634497997, "全帯域_40ms_rms_envelope_nrmse": 0.019037149820379923, "全帯域_40ms_envelope_correlation": 0.9997238238751344, "全帯域_40ms_modulation_mae_db": 0.5191826567742507, "300–1000 Hz_10ms_log_rms_mae_db": 1.0209400302663347, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07928875440632581, "300–1000 Hz_10ms_envelope_correlation": 0.9947950129309687, "300–1000 Hz_10ms_modulation_mae_db": 0.6885357775396993, "300–1000 Hz_40ms_log_rms_mae_db": 0.7729801210784936, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0454898470057972, "300–1000 Hz_40ms_envelope_correlation": 0.9982289462950062, "300–1000 Hz_40ms_modulation_mae_db": 0.7860326473793058, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5009991596167824, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1545088738984529, "1000–3000 Hz_10ms_envelope_correlation": 0.9834906068821423, "1000–3000 Hz_10ms_modulation_mae_db": 0.6785546154332541, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9730278469837372, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08965718223014435, "1000–3000 Hz_40ms_envelope_correlation": 0.9948310539897282, "1000–3000 Hz_40ms_modulation_mae_db": 1.0397421245035594, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4353170648300542, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19957203187968328, "3000–6000 Hz_10ms_envelope_correlation": 0.9761293070128659, "3000–6000 Hz_10ms_modulation_mae_db": 0.6197333227810549, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0224826502665327, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09515598923811183, "3000–6000 Hz_40ms_envelope_correlation": 0.9948108593451227, "3000–6000 Hz_40ms_modulation_mae_db": 0.742310858837769, "mr_stft_db": 5.1164042899183935, "band_mae_db": 0.1425787725222751, "rms_gain_error_db": -0.0620973368637192, "output_peak": 0.3631680447721065}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6884448474534786, "全帯域_10ms_rms_envelope_nrmse": 0.0430184784514913, "全帯域_10ms_envelope_correlation": 0.9984642668294279, "全帯域_10ms_modulation_mae_db": 0.2695167502159872, "全帯域_40ms_log_rms_mae_db": 0.3741883718990235, "全帯域_40ms_rms_envelope_nrmse": 0.023857787195136544, "全帯域_40ms_envelope_correlation": 0.9995637218344229, "全帯域_40ms_modulation_mae_db": 0.10262036566786781, "300–1000 Hz_10ms_log_rms_mae_db": 0.9900484794753173, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.09388092582179944, "300–1000 Hz_10ms_envelope_correlation": 0.9927435776442196, "300–1000 Hz_10ms_modulation_mae_db": 0.1971336236994383, "300–1000 Hz_40ms_log_rms_mae_db": 0.7346732796802765, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0640519694880452, "300–1000 Hz_40ms_envelope_correlation": 0.9964904089240109, "300–1000 Hz_40ms_modulation_mae_db": 0.27633600973514094, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4794819096819154, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14441717524270997, "1000–3000 Hz_10ms_envelope_correlation": 0.9851996517729241, "1000–3000 Hz_10ms_modulation_mae_db": 0.7964610380669382, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9265742638339661, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07145349869985805, "1000–3000 Hz_40ms_envelope_correlation": 0.9963441917035073, "1000–3000 Hz_40ms_modulation_mae_db": 1.3910396220408359, "3000–6000 Hz_10ms_log_rms_mae_db": 1.457189636766758, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19678305778308552, "3000–6000 Hz_10ms_envelope_correlation": 0.976833841441053, "3000–6000 Hz_10ms_modulation_mae_db": 0.7637187411797338, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0329900969572454, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10885817818705289, "3000–6000 Hz_40ms_envelope_correlation": 0.9931193263608713, "3000–6000 Hz_40ms_modulation_mae_db": 0.6736073861236176, "mr_stft_db": 5.109624049031431, "band_mae_db": 0.1225355165937582, "rms_gain_error_db": -0.09295329395289123, "output_peak": 0.3673396712831851}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6679229297954893, "全帯域_10ms_rms_envelope_nrmse": 0.03491086729523431, "全帯域_10ms_envelope_correlation": 0.9990072396999972, "全帯域_10ms_modulation_mae_db": 0.6377832820080841, "全帯域_40ms_log_rms_mae_db": 0.36551938206755924, "全帯域_40ms_rms_envelope_nrmse": 0.01783306284739551, "全帯域_40ms_envelope_correlation": 0.999781116682226, "全帯域_40ms_modulation_mae_db": 0.4754576766277651, "300–1000 Hz_10ms_log_rms_mae_db": 0.932532479745549, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.07315061790293234, "300–1000 Hz_10ms_envelope_correlation": 0.9955767504963483, "300–1000 Hz_10ms_modulation_mae_db": 0.6271723879774497, "300–1000 Hz_40ms_log_rms_mae_db": 0.6782825964103866, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.04468229086511604, "300–1000 Hz_40ms_envelope_correlation": 0.9982937610716707, "300–1000 Hz_40ms_modulation_mae_db": 0.7656539373862753, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4845745282047746, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13734294502696473, "1000–3000 Hz_10ms_envelope_correlation": 0.9872455421613746, "1000–3000 Hz_10ms_modulation_mae_db": 0.5222670320383855, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9641715201590199, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08063395828296659, "1000–3000 Hz_40ms_envelope_correlation": 0.9961107919622029, "1000–3000 Hz_40ms_modulation_mae_db": 1.2933676199311297, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4400385883046969, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16749310855894392, "3000–6000 Hz_10ms_envelope_correlation": 0.98322969849634, "3000–6000 Hz_10ms_modulation_mae_db": 0.4049312944594294, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0620768683582222, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09476010627609208, "3000–6000 Hz_40ms_envelope_correlation": 0.9944782395970673, "3000–6000 Hz_40ms_modulation_mae_db": 0.44697689763882353, "mr_stft_db": 5.129323001456217, "band_mae_db": 0.07417766642097588, "rms_gain_error_db": -0.08114583758713163, "output_peak": 0.3604695904218543}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6792330012801906, "全帯域_10ms_rms_envelope_nrmse": 0.03979130047059768, "全帯域_10ms_envelope_correlation": 0.998665797589073, "全帯域_10ms_modulation_mae_db": 0.36128662533936406, "全帯域_40ms_log_rms_mae_db": 0.37911002131197846, "全帯域_40ms_rms_envelope_nrmse": 0.020476002065209657, "全帯域_40ms_envelope_correlation": 0.9996634635674518, "全帯域_40ms_modulation_mae_db": 0.24566635572837045, "300–1000 Hz_10ms_log_rms_mae_db": 1.015397663564826, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.08216479881635136, "300–1000 Hz_10ms_envelope_correlation": 0.9943972777963369, "300–1000 Hz_10ms_modulation_mae_db": 0.12486372036369237, "300–1000 Hz_40ms_log_rms_mae_db": 0.6910626842151809, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05272075721627472, "300–1000 Hz_40ms_envelope_correlation": 0.9976000540240697, "300–1000 Hz_40ms_modulation_mae_db": 0.2524369837160517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4947003221780626, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1329574400581506, "1000–3000 Hz_10ms_envelope_correlation": 0.9877423839093944, "1000–3000 Hz_10ms_modulation_mae_db": 0.9082842543948498, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9218988227864628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0683600800871127, "1000–3000 Hz_40ms_envelope_correlation": 0.9970510358621613, "1000–3000 Hz_40ms_modulation_mae_db": 1.3531829009397047, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4385621731815408, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16918121357093013, "3000–6000 Hz_10ms_envelope_correlation": 0.9831114327091673, "3000–6000 Hz_10ms_modulation_mae_db": 0.48433605257147877, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0325741550292586, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10203309051810765, "3000–6000 Hz_40ms_envelope_correlation": 0.9936437881935059, "3000–6000 Hz_40ms_modulation_mae_db": 1.0104412815793988, "mr_stft_db": 5.118326402600057, "band_mae_db": 0.24185496291272304, "rms_gain_error_db": -0.07005948671427664, "output_peak": 0.35903448880923894}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.6363792219266814, "全帯域_10ms_rms_envelope_nrmse": 0.0344149521015469, "全帯域_10ms_envelope_correlation": 0.9990228005137358, "全帯域_10ms_modulation_mae_db": 0.47868114114445515, "全帯域_40ms_log_rms_mae_db": 0.34867155285488466, "全帯域_40ms_rms_envelope_nrmse": 0.017612470596031795, "全帯域_40ms_envelope_correlation": 0.9997742944312551, "全帯域_40ms_modulation_mae_db": 0.2800563591325841, "300–1000 Hz_10ms_log_rms_mae_db": 0.9339890066769577, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.06359979265424028, "300–1000 Hz_10ms_envelope_correlation": 0.9966555177479661, "300–1000 Hz_10ms_modulation_mae_db": 0.16572871295197858, "300–1000 Hz_40ms_log_rms_mae_db": 0.6507407698241972, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.03931693041610075, "300–1000 Hz_40ms_envelope_correlation": 0.9986710503152596, "300–1000 Hz_40ms_modulation_mae_db": 0.1595327763806833, "1000–3000 Hz_10ms_log_rms_mae_db": 1.440709200224087, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.12249560186909247, "1000–3000 Hz_10ms_envelope_correlation": 0.9896013177046944, "1000–3000 Hz_10ms_modulation_mae_db": 0.36820562107939786, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9561930392456172, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06778811399385327, "1000–3000 Hz_40ms_envelope_correlation": 0.9969998680926099, "1000–3000 Hz_40ms_modulation_mae_db": 0.8747633287926136, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3970038625931396, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19997821593456816, "3000–6000 Hz_10ms_envelope_correlation": 0.9761150481277127, "3000–6000 Hz_10ms_modulation_mae_db": 0.9483252945276885, "3000–6000 Hz_40ms_log_rms_mae_db": 1.042067794244752, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09244141498132011, "3000–6000 Hz_40ms_envelope_correlation": 0.9947070006882177, "3000–6000 Hz_40ms_modulation_mae_db": 1.0325679904869514, "mr_stft_db": 5.115316429695246, "band_mae_db": 0.0791805927026963, "rms_gain_error_db": -0.0740635465541089, "output_peak": 0.36521965617277574}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.286227680848414, "spectral_convergence": 0.12293174558500909, "spectral_convergence_status": "ok", "amplitude_floor": 1.7050603882585003e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.0798439732051754, "spectral_convergence": 0.1141685823780798, "spectral_convergence_status": "ok", "amplitude_floor": 1.5194442893884604e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.986904057110247, "spectral_convergence": 0.11441193220952438, "spectral_convergence_status": "ok", "amplitude_floor": 1.5327487413055003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00047003850181382925, "output_power": 0.0004719711335903294, "delta_db": 0.017820040520773385, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.004513086919531444, "output_power": 0.004398042688121862, "delta_db": -0.11214257491139439, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003251325154635926, "output_power": 0.00031760946436952065, "delta_db": -0.10166968684241867, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.51153215194187e-05, "output_power": 8.106293684721046e-05, "delta_db": -0.2118541017187013, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6288307106843834e-05, "output_power": 2.7539670770631993e-05, "delta_db": 0.20196124416278494, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.4880972570108e-05, "output_power": 4.525990679552686e-05, "delta_db": 0.03651395763828636, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.11366026763239317, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.09891732790952575, "pitch": null, "output_rms_dbfs": -22.72337921772513, "output_peak": 0.36369815689313867, "reference_rms_dbfs": -22.624461889815606, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.117658570387945}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023914198234423542, "log_rms_mae_db": 0.6888493302121759, "rms_envelope_nrmse": 0.04042110013050984, "envelope_correlation": 0.9986786728520418, "rms_gain_error_db": -0.09602870950797057, "reference_mean_power": 0.005456771858924342, "output_mean_power": 0.00533743901526614}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.071388012833875e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045567084702756786, "output_power": 0.00044697803162488934, "reference_active": true, "delta_db": -0.08365065457920196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.44246106516196e-05, "output_power": 5.938139706411525e-05, "reference_active": true, "delta_db": -0.354013915722238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7099940640562705e-05, "output_power": 1.2878339461588335e-05, "reference_active": true, "delta_db": -1.2313473429409716}], "active_band_count": 3, "active_band_mae_db": 0.5563373044141372}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022747909354312384, "log_rms_mae_db": 0.39667940439495525, "rms_envelope_nrmse": 0.021937044039759195, "envelope_correlation": 0.9996671621236308, "rms_gain_error_db": -0.09149720627161156, "reference_mean_power": 0.005416920568028677, "output_mean_power": 0.005303990611408303}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.96418119219355e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000410225191407457, "output_power": 0.0004037724319740355, "reference_active": true, "delta_db": -0.06885662657336203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7757195967845804e-05, "output_power": 2.5650610557809226e-05, "reference_active": true, "delta_db": -0.3427788454648235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0255605769228834e-06, "output_power": 8.522952309260819e-07, "reference_active": true, "delta_db": -0.8037125966578922}], "active_band_count": 3, "active_band_mae_db": 0.4051160228986926}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.442623236865579e-05, "log_rms_mae_db": 0.9637190323994446, "rms_envelope_nrmse": 0.07764422775025215, "envelope_correlation": 0.9950197476245956, "rms_gain_error_db": -0.019961221143860043, "reference_mean_power": 0.0007990709247977912, "output_mean_power": 0.0007954066296744469}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.944736222458957e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022092122499658314, "output_power": 0.00021614362767050115, "reference_active": true, "delta_db": -0.09495006475882206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3665816077667355e-05, "output_power": 2.236095317638782e-05, "reference_active": true, "delta_db": -0.2463117280722376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5979447501439136e-06, "output_power": 4.630217003257394e-06, "reference_active": true, "delta_db": -0.8242726233986062}], "active_band_count": 3, "active_band_mae_db": 0.3885114720765553}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.226666373654229e-05, "log_rms_mae_db": 0.7049728378719549, "rms_envelope_nrmse": 0.04222546497284757, "envelope_correlation": 0.9984839547953053, "rms_gain_error_db": -0.021291422534878684, "reference_mean_power": 0.0008012273487091395, "output_mean_power": 0.0007973089195815629}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.73006991688152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002092887788100005, "output_power": 0.00020475470454894498, "reference_active": true, "delta_db": -0.09512054787558274}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0995901908054441e-05, "output_power": 1.0209377080037337e-05, "reference_active": true, "delta_db": -0.32231612308396723}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1412698290887665e-07, "output_power": 3.5207889654586793e-07, "reference_active": true, "delta_db": -0.7049353393904291}], "active_band_count": 3, "active_band_mae_db": 0.374124003449993}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9464514079665034e-05, "log_rms_mae_db": 1.4651373791422928, "rms_envelope_nrmse": 0.13229539806514629, "envelope_correlation": 0.9878454604208449, "rms_gain_error_db": -0.2431456067752788, "reference_mean_power": 7.576176061225114e-05, "output_mean_power": 7.163668785809027e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.472428410826198e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.044189308875811e-05, "output_power": 2.8237911999379632e-05, "reference_active": true, "delta_db": -0.32639075933554984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3472076435688925e-06, "output_power": 2.543608001107913e-06, "reference_active": true, "delta_db": -1.1923247217304098}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0111506850143103e-06, "output_power": 6.665528126757836e-07, "reference_active": true, "delta_db": -1.8098131504150896}], "active_band_count": 3, "active_band_mae_db": 1.1095095438270164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.608084029396227e-05, "log_rms_mae_db": 0.9444330092040851, "rms_envelope_nrmse": 0.07612482724259731, "envelope_correlation": 0.9961479201262444, "rms_gain_error_db": -0.24323189361433817, "reference_mean_power": 7.611087404608628e-05, "output_mean_power": 7.196536293810338e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.164443748057941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.887302600627445e-05, "output_power": 2.6740027971394394e-05, "reference_active": true, "delta_db": -0.33330444809886517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.617498060844069e-06, "output_power": 1.1417659764237897e-06, "reference_active": true, "delta_db": -1.512666713949058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.602390292027545e-08, "output_power": 4.15138642728685e-08, "reference_active": true, "delta_db": -3.164259815962276}], "active_band_count": 3, "active_band_mae_db": 1.6700769926700663}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.198647979957052e-05, "log_rms_mae_db": 1.4294554113699087, "rms_envelope_nrmse": 0.18316090301919208, "envelope_correlation": 0.9803124208009539, "rms_gain_error_db": 0.08772079696673288, "reference_mean_power": 5.064740774931045e-05, "output_mean_power": 5.168080880746392e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.700608322524163e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.136344717386399e-05, "output_power": 4.1606021402673444e-05, "reference_active": true, "delta_db": 0.025394630337663076}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.648317567219929e-06, "output_power": 6.89579145181084e-06, "reference_active": true, "delta_db": 0.15872362702483686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.888415633999808e-06, "output_power": 2.6177055681207727e-06, "reference_active": true, "delta_db": -1.7185188382612848}], "active_band_count": 3, "active_band_mae_db": 0.6342123652079282}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.0010568231240365e-05, "log_rms_mae_db": 1.0550408773117372, "rms_envelope_nrmse": 0.11212208047024871, "envelope_correlation": 0.9923866341656495, "rms_gain_error_db": 0.08765846110409486, "reference_mean_power": 5.089623620868825e-05, "output_mean_power": 5.193396888778479e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.379392951970345e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.821321246572876e-05, "output_power": 3.846505096824667e-05, "reference_active": true, "delta_db": 0.02852763230238607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.5895785128125065e-06, "output_power": 3.14266053180345e-06, "reference_active": true, "delta_db": -0.5774598551285054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2481135158983729e-07, "output_power": 1.7350799553389165e-07, "reference_active": true, "delta_db": -1.1249874405725666}], "active_band_count": 3, "active_band_mae_db": 0.5769916426678193}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9535631536889757, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8e66b6bf467a054ddefa1eaeb4da9683a54e8084c1bdede9dd6394791d89113e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9874149104619265, "standard_deviation": 0.0021593463090495628, "minimum": 0.9834906068821423, "maximum": 0.9899832477474252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4748694818527297, "standard_deviation": 0.01954884569311878, "minimum": 1.440709200224087, "maximum": 1.5009991596167824, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6796763534506335, "standard_deviation": 0.24501809073969336, "minimum": 0.36820562107939786, "maximum": 1.1095095438270164, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1344143333573548, "standard_deviation": 0.011369871569577314, "minimum": 0.11893484280421829, "maximum": 0.1545088738984529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963721614044561, "standard_deviation": 0.0007527608908841731, "minimum": 0.9948310539897282, "maximum": 0.9971569653930019, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9488065753924619, "standard_deviation": 0.020150206369737723, "minimum": 0.9218988227864628, "maximum": 0.9730278469837372, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1634612033824339, "standard_deviation": 0.3130056879422158, "minimum": 0.7424262363621571, "maximum": 1.6700769926700663, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07439676908097875, "standard_deviation": 0.008657378757884113, "minimum": 0.06245600768583666, "maximum": 0.08965718223014435, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9799055219273927, "standard_deviation": 0.0032376720605618477, "minimum": 0.9761150481277127, "maximum": 0.9836231939217503, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4362204841159123, "standard_deviation": 0.026993636071457397, "minimum": 1.3970038625931396, "maximum": 1.4839215045592495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5776675572010523, "standard_deviation": 0.210741615850294, "minimum": 0.2840262517139997, "maximum": 0.9483252945276885, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18363697006339286, "standard_deviation": 0.013921574041055273, "minimum": 0.16749310855894392, "maximum": 0.19997821593456816, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940670783092449, "standard_deviation": 0.0009306792223194127, "minimum": 0.9923866341656495, "maximum": 0.9950421833942017, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.045174450682283, "standard_deviation": 0.022090859608996982, "minimum": 1.0224826502665327, "maximum": 1.0880464800627905, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8223527195771934, "standard_deviation": 0.24837184253751443, "minimum": 0.44697689763882353, "maximum": 1.1499712729467624, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09958655111912615, "standard_deviation": 0.0073505552877061835, "minimum": 0.09244141498132011, "maximum": 0.11212208047024871, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9950723635739317, "standard_deviation": 0.0012071874209670036, "minimum": 0.9927435776442196, "maximum": 0.9966555177479661, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 0.9891380786864559, "standard_deviation": 0.04132006772045822, "minimum": 0.932532479745549, "maximum": 1.0404900122526317, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.46428087149695035, "standard_deviation": 0.3422069142859811, "minimum": 0.12486372036369237, "maximum": 1.1387785846713154, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07675455773583095, "standard_deviation": 0.00931027046690649, "minimum": 0.06359979265424028, "maximum": 0.09388092582179944, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9980475491182328, "standard_deviation": 0.0007253240853409056, "minimum": 0.9964904089240109, "maximum": 0.9986710503152596, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.7092541924849183, "standard_deviation": 0.039304933359431186, "minimum": 0.6507407698241972, "maximum": 0.7729801210784936, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5709841622159593, "standard_deviation": 0.44136620566299195, "minimum": 0.1595327763806833, "maximum": 1.503020246619143, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.047159951531425603, "standard_deviation": 0.008199365407623732, "minimum": 0.03931693041610075, "maximum": 0.0640519694880452, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.12893848551059847, "standard_deviation": 0.05479864729817254, "minimum": 0.07417766642097588, "maximum": 0.24185496291272304, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.11613783601318, "standard_deviation": 0.0073089982495123575, "minimum": 5.10409925988162, "maximum": 5.129323001456217, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.36462017591916746, "standard_deviation": 0.003918026839753705, "minimum": 0.35903448880923894, "maximum": 0.37121483931203847, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.0764815602384881, "standard_deviation": 0.013414023752341499, "minimum": -0.09891732790952575, "maximum": -0.0620973368637192, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9988223449552532, "standard_deviation": 0.00020276899155633667, "minimum": 0.9984642668294279, "maximum": 0.9990228005137358, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 0.6622121080340531, "standard_deviation": 0.0215319899197452, "minimum": 0.6363792219266814, "maximum": 0.6888493302121759, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4416432442574317, "standard_deviation": 0.14065852094086823, "minimum": 0.23780746314009601, "maximum": 0.6377832820080841, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.037583311392806004, "standard_deviation": 0.003166363331749869, "minimum": 0.0344149521015469, "maximum": 0.0430184784514913, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9997116585846283, "standard_deviation": 7.549781731088242e-05, "minimum": 0.9995637218344229, "maximum": 0.999781116682226, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.36865810547635297, "standard_deviation": 0.015170666684457225, "minimum": 0.34867155285488466, "maximum": 0.39667940439495525, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.37967042927529404, "standard_deviation": 0.1754388128400524, "minimum": 0.10262036566786781, "maximum": 0.6632156085604037, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.01951922775705347, "standard_deviation": 0.002372542984703516, "minimum": 0.01720566869802357, "maximum": 0.023857787195136544, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_908-157963-0018", "language": "en", "style": "normal", "speaker": "librispeech_908", "source_utterance_id": "908-157963-0018", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 68080, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_908-157963-0018.wav", "source_file_sha256": "7e87b3ae4060355e049838d40e87f3da92ef3b0d28b68ef08e791238f1baef1a", "decoded_pcm_sha256": "8ae8f9234bac8c63b967d3c7b5075657f9771ac253a59a383e523ecff9c3b006", "original_source_sha256": "53efff3282f4691482a4e735887fab9130c78ec10f3bdb79bf68f8c212898134", "corpus_entry_id": "librispeech_908-157963-0018"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.8767905235290527, "predicted_mos": 3.8767905235290527, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "4f7ac2354f36190dd460ce2e3d3db121238d43d090b0d66df1811f478e19b90d", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.5817255189977004, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_908-157963-0018", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "4f7ac2354f36190dd460ce2e3d3db121238d43d090b0d66df1811f478e19b90d", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "068cbe41c56c5de3cb2928395586f7d207582cd51ef6612040fe2872c397e9dd"}, "rows": [{"case": "librispeech_908-157963-0018", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.21039080619812, "predicted_mos": 2.21039080619812, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "b864da89f4eabb6b718101d02dc86c69bb1891421adfe9e4387432038ceac42c", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.067346167998039, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_908-157963-0018", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "b864da89f4eabb6b718101d02dc86c69bb1891421adfe9e4387432038ceac42c", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0499142042238927, "全帯域_10ms_rms_envelope_nrmse": 0.09823086793460886, "全帯域_10ms_envelope_correlation": 0.9899868709660282, "全帯域_10ms_modulation_mae_db": 0.27269707204922283, "全帯域_40ms_log_rms_mae_db": 0.520280567032123, "全帯域_40ms_rms_envelope_nrmse": 0.04596146767393709, "全帯域_40ms_envelope_correlation": 0.9977501469723025, "全帯域_40ms_modulation_mae_db": 0.2873418169379034, "300–1000 Hz_10ms_log_rms_mae_db": 1.6180190702074284, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12916747420601737, "300–1000 Hz_10ms_envelope_correlation": 0.9864557826757002, "300–1000 Hz_10ms_modulation_mae_db": 0.2542385386116102, "300–1000 Hz_40ms_log_rms_mae_db": 0.941548513883815, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06605294819124838, "300–1000 Hz_40ms_envelope_correlation": 0.9965729595366356, "300–1000 Hz_40ms_modulation_mae_db": 0.41750253041590984, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6076240104323005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1777763078417581, "1000–3000 Hz_10ms_envelope_correlation": 0.9752082489650673, "1000–3000 Hz_10ms_modulation_mae_db": 0.8272576649098956, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1496837315532327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10223216725538371, "1000–3000 Hz_40ms_envelope_correlation": 0.9916401820677322, "1000–3000 Hz_40ms_modulation_mae_db": 0.5719229879453839, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5418827032549032, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22362320773125025, "3000–6000 Hz_10ms_envelope_correlation": 0.9664867847201025, "3000–6000 Hz_10ms_modulation_mae_db": 0.42949813083350724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2584617619917302, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15155208898647898, "3000–6000 Hz_40ms_envelope_correlation": 0.9838983983163263, "3000–6000 Hz_40ms_modulation_mae_db": 0.5828358507588818, "mr_stft_db": 5.713041302141555, "band_mae_db": 0.15832997108538327, "rms_gain_error_db": -0.11084439914196895, "output_peak": 0.5056645344602257}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0131663502010733, "全帯域_10ms_rms_envelope_nrmse": 0.09708849377225867, "全帯域_10ms_envelope_correlation": 0.9905473522979499, "全帯域_10ms_modulation_mae_db": 0.14977043290083672, "全帯域_40ms_log_rms_mae_db": 0.4944872944403882, "全帯域_40ms_rms_envelope_nrmse": 0.05118179527953911, "全帯域_40ms_envelope_correlation": 0.9976316184577948, "全帯域_40ms_modulation_mae_db": 0.3370523649755821, "300–1000 Hz_10ms_log_rms_mae_db": 1.6677306811726085, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13079441550865267, "300–1000 Hz_10ms_envelope_correlation": 0.9867179688161944, "300–1000 Hz_10ms_modulation_mae_db": 0.42104217033737285, "300–1000 Hz_40ms_log_rms_mae_db": 1.00160438520819, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0736108064824261, "300–1000 Hz_40ms_envelope_correlation": 0.9964842460773902, "300–1000 Hz_40ms_modulation_mae_db": 0.5238999535310347, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5804617452581005, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15896768246888476, "1000–3000 Hz_10ms_envelope_correlation": 0.9803762258634582, "1000–3000 Hz_10ms_modulation_mae_db": 0.3361570713856021, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1413011674670666, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.094551177416791, "1000–3000 Hz_40ms_envelope_correlation": 0.9930350387416575, "1000–3000 Hz_40ms_modulation_mae_db": 0.34771875315080886, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5063143448532181, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19365726049492976, "3000–6000 Hz_10ms_envelope_correlation": 0.9753330071586451, "3000–6000 Hz_10ms_modulation_mae_db": 0.6030496859165778, "3000–6000 Hz_40ms_log_rms_mae_db": 1.191558545056662, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12357395447433436, "3000–6000 Hz_40ms_envelope_correlation": 0.9896373083110636, "3000–6000 Hz_40ms_modulation_mae_db": 0.725143515798884, "mr_stft_db": 5.717921988701494, "band_mae_db": 0.2881049664785244, "rms_gain_error_db": -0.20849314505583524, "output_peak": 0.4927416204629885}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1001499085311999, "全帯域_10ms_rms_envelope_nrmse": 0.09590939177998824, "全帯域_10ms_envelope_correlation": 0.9906317385782029, "全帯域_10ms_modulation_mae_db": 0.2725202204688994, "全帯域_40ms_log_rms_mae_db": 0.5906790934519254, "全帯域_40ms_rms_envelope_nrmse": 0.047974840968859754, "全帯域_40ms_envelope_correlation": 0.9977569385293316, "全帯域_40ms_modulation_mae_db": 0.5154596648234513, "300–1000 Hz_10ms_log_rms_mae_db": 1.7640271013701252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13091980637079698, "300–1000 Hz_10ms_envelope_correlation": 0.9864202410758401, "300–1000 Hz_10ms_modulation_mae_db": 0.7527579246204231, "300–1000 Hz_40ms_log_rms_mae_db": 0.9481591818659253, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07452144611481, "300–1000 Hz_40ms_envelope_correlation": 0.9959813346015092, "300–1000 Hz_40ms_modulation_mae_db": 0.6384371900423795, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6510155120245096, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16520984670761354, "1000–3000 Hz_10ms_envelope_correlation": 0.9788696230777861, "1000–3000 Hz_10ms_modulation_mae_db": 0.42474238843074663, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1746072596224668, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08996966178030297, "1000–3000 Hz_40ms_envelope_correlation": 0.9939970052794624, "1000–3000 Hz_40ms_modulation_mae_db": 0.6655792872485199, "3000–6000 Hz_10ms_log_rms_mae_db": 1.484059756309818, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1774788589080573, "3000–6000 Hz_10ms_envelope_correlation": 0.978462707832021, "3000–6000 Hz_10ms_modulation_mae_db": 0.44318479382515724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1568970834549293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11969903369852329, "3000–6000 Hz_40ms_envelope_correlation": 0.9899755205477733, "3000–6000 Hz_40ms_modulation_mae_db": 0.4140590221517466, "mr_stft_db": 5.743137590140673, "band_mae_db": 0.32319584780435795, "rms_gain_error_db": -0.16963432227304776, "output_peak": 0.4967593362680309}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0493093035974543, "全帯域_10ms_rms_envelope_nrmse": 0.08435461138988137, "全帯域_10ms_envelope_correlation": 0.9926191882519757, "全帯域_10ms_modulation_mae_db": 0.21019342755952178, "全帯域_40ms_log_rms_mae_db": 0.5608656485518981, "全帯域_40ms_rms_envelope_nrmse": 0.04741466122419095, "全帯域_40ms_envelope_correlation": 0.9974670164197929, "全帯域_40ms_modulation_mae_db": 0.28325358175464316, "300–1000 Hz_10ms_log_rms_mae_db": 1.6909077903019494, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11938943848856275, "300–1000 Hz_10ms_envelope_correlation": 0.988336609769946, "300–1000 Hz_10ms_modulation_mae_db": 0.35843123726917253, "300–1000 Hz_40ms_log_rms_mae_db": 0.9866651208735565, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06553407694702633, "300–1000 Hz_40ms_envelope_correlation": 0.9962016973427162, "300–1000 Hz_40ms_modulation_mae_db": 0.3843595427210101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.588099429578295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1593668833700831, "1000–3000 Hz_10ms_envelope_correlation": 0.98049734170078, "1000–3000 Hz_10ms_modulation_mae_db": 0.6428756771751559, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0897498836667758, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08768009080142197, "1000–3000 Hz_40ms_envelope_correlation": 0.9945122802644043, "1000–3000 Hz_40ms_modulation_mae_db": 0.6642968653147282, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5469166975545405, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2131922602908928, "3000–6000 Hz_10ms_envelope_correlation": 0.9693167369662747, "3000–6000 Hz_10ms_modulation_mae_db": 0.3664511669858796, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2202280823723828, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14633683759166444, "3000–6000 Hz_40ms_envelope_correlation": 0.9861935707641315, "3000–6000 Hz_40ms_modulation_mae_db": 0.4837032706053586, "mr_stft_db": 5.715573170495685, "band_mae_db": 0.28066961246785493, "rms_gain_error_db": -0.02022005979709116, "output_peak": 0.489219270731731}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0607829744977169, "全帯域_10ms_rms_envelope_nrmse": 0.08950613839178037, "全帯域_10ms_envelope_correlation": 0.9917464487977474, "全帯域_10ms_modulation_mae_db": 0.30000067888047965, "全帯域_40ms_log_rms_mae_db": 0.583985425669924, "全帯域_40ms_rms_envelope_nrmse": 0.04584140595255515, "全帯域_40ms_envelope_correlation": 0.9978104891127596, "全帯域_40ms_modulation_mae_db": 0.3739487818442302, "300–1000 Hz_10ms_log_rms_mae_db": 1.6639102359115994, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1337003557419574, "300–1000 Hz_10ms_envelope_correlation": 0.9855810455905475, "300–1000 Hz_10ms_modulation_mae_db": 0.5020892031979449, "300–1000 Hz_40ms_log_rms_mae_db": 0.9753665146791288, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07682372636040012, "300–1000 Hz_40ms_envelope_correlation": 0.9953778435878963, "300–1000 Hz_40ms_modulation_mae_db": 0.5590563497844899, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7451234657618808, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1679100010342424, "1000–3000 Hz_10ms_envelope_correlation": 0.9786731425596877, "1000–3000 Hz_10ms_modulation_mae_db": 0.7817838073983211, "1000–3000 Hz_40ms_log_rms_mae_db": 1.256019128165454, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0959613653188239, "1000–3000 Hz_40ms_envelope_correlation": 0.9938945436981403, "1000–3000 Hz_40ms_modulation_mae_db": 0.9071152522615762, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5257806407468386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1886274736403812, "3000–6000 Hz_10ms_envelope_correlation": 0.9756789607054962, "3000–6000 Hz_10ms_modulation_mae_db": 0.4256616787536595, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2355237225738762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1250732893406029, "3000–6000 Hz_40ms_envelope_correlation": 0.9892206627460833, "3000–6000 Hz_40ms_modulation_mae_db": 0.7335403919648843, "mr_stft_db": 5.7280173145441, "band_mae_db": 0.308225084105773, "rms_gain_error_db": -0.12145677398572788, "output_peak": 0.5235887164470976}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0584397692237677, "全帯域_10ms_rms_envelope_nrmse": 0.09419592412138648, "全帯域_10ms_envelope_correlation": 0.9907560947929599, "全帯域_10ms_modulation_mae_db": 0.08635797987894955, "全帯域_40ms_log_rms_mae_db": 0.5696734071180193, "全帯域_40ms_rms_envelope_nrmse": 0.05332853928377529, "全帯域_40ms_envelope_correlation": 0.9967774551990777, "全帯域_40ms_modulation_mae_db": 0.18136396632087573, "300–1000 Hz_10ms_log_rms_mae_db": 1.6805141062618916, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13051921413348713, "300–1000 Hz_10ms_envelope_correlation": 0.9860783003563516, "300–1000 Hz_10ms_modulation_mae_db": 0.3708609863401806, "300–1000 Hz_40ms_log_rms_mae_db": 0.9721223179863212, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0755221129294329, "300–1000 Hz_40ms_envelope_correlation": 0.9952079777294461, "300–1000 Hz_40ms_modulation_mae_db": 0.5234220718329853, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6182395957044131, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13570638382297961, "1000–3000 Hz_10ms_envelope_correlation": 0.9860455368278838, "1000–3000 Hz_10ms_modulation_mae_db": 0.7883127663612081, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1399100839027165, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08167875254766294, "1000–3000 Hz_40ms_envelope_correlation": 0.9951843670148566, "1000–3000 Hz_40ms_modulation_mae_db": 0.94951667791784, "3000–6000 Hz_10ms_log_rms_mae_db": 1.568255680058505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19433604585883416, "3000–6000 Hz_10ms_envelope_correlation": 0.9745304666577365, "3000–6000 Hz_10ms_modulation_mae_db": 0.69201696894812, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1962412592402414, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12397260105400496, "3000–6000 Hz_40ms_envelope_correlation": 0.989201197281598, "3000–6000 Hz_40ms_modulation_mae_db": 0.33546911385063954, "mr_stft_db": 5.7187922016636525, "band_mae_db": 0.24199981483610566, "rms_gain_error_db": -0.051102602794842925, "output_peak": 0.49224543795187303}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9979617971099147, "全帯域_10ms_rms_envelope_nrmse": 0.08615340249070157, "全帯域_10ms_envelope_correlation": 0.9922635092308546, "全帯域_10ms_modulation_mae_db": 0.10315692716272747, "全帯域_40ms_log_rms_mae_db": 0.5562372483799708, "全帯域_40ms_rms_envelope_nrmse": 0.04253212530439596, "全帯域_40ms_envelope_correlation": 0.997941344025211, "全帯域_40ms_modulation_mae_db": 0.09744255215839019, "300–1000 Hz_10ms_log_rms_mae_db": 1.7453962809727699, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12139671060839473, "300–1000 Hz_10ms_envelope_correlation": 0.9879815557432103, "300–1000 Hz_10ms_modulation_mae_db": 0.3992585725625429, "300–1000 Hz_40ms_log_rms_mae_db": 0.9996254137364163, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06787780618952477, "300–1000 Hz_40ms_envelope_correlation": 0.9959493380862823, "300–1000 Hz_40ms_modulation_mae_db": 0.44174776558949996, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6480396496746217, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15305495787014217, "1000–3000 Hz_10ms_envelope_correlation": 0.9830365168405703, "1000–3000 Hz_10ms_modulation_mae_db": 0.6010913595360223, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1825361734500357, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09124792492038117, "1000–3000 Hz_40ms_envelope_correlation": 0.9952284591671029, "1000–3000 Hz_40ms_modulation_mae_db": 0.7686030197788624, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5210326895536317, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20182803765600674, "3000–6000 Hz_10ms_envelope_correlation": 0.973562586520539, "3000–6000 Hz_10ms_modulation_mae_db": 0.5463923386562826, "3000–6000 Hz_40ms_log_rms_mae_db": 1.224324449164093, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1323940685721135, "3000–6000 Hz_40ms_envelope_correlation": 0.9883501178140234, "3000–6000 Hz_40ms_modulation_mae_db": 0.4606761129822418, "mr_stft_db": 5.727129029496339, "band_mae_db": 0.24568206086466626, "rms_gain_error_db": -0.011469870092907272, "output_peak": 0.47843770791729345}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.064585767485727, "全帯域_10ms_rms_envelope_nrmse": 0.09966107900343238, "全帯域_10ms_envelope_correlation": 0.9899513622950392, "全帯域_10ms_modulation_mae_db": 0.34032194756751905, "全帯域_40ms_log_rms_mae_db": 0.5166037966520429, "全帯域_40ms_rms_envelope_nrmse": 0.055566125559275176, "全帯域_40ms_envelope_correlation": 0.9970064164940688, "全帯域_40ms_modulation_mae_db": 0.4068060956314033, "300–1000 Hz_10ms_log_rms_mae_db": 1.6767606740393695, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12963692413267977, "300–1000 Hz_10ms_envelope_correlation": 0.986822005165173, "300–1000 Hz_10ms_modulation_mae_db": 0.4488509388594153, "300–1000 Hz_40ms_log_rms_mae_db": 0.9439334677292541, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07382600802950946, "300–1000 Hz_40ms_envelope_correlation": 0.9962537991013732, "300–1000 Hz_40ms_modulation_mae_db": 0.5505045927351883, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6698524088967075, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1581987840562569, "1000–3000 Hz_10ms_envelope_correlation": 0.9815016562183216, "1000–3000 Hz_10ms_modulation_mae_db": 0.805247000005513, "1000–3000 Hz_40ms_log_rms_mae_db": 1.169577890404464, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10119979943449846, "1000–3000 Hz_40ms_envelope_correlation": 0.99317522894897, "1000–3000 Hz_40ms_modulation_mae_db": 0.9744344545731458, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5184079576423473, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19253585730484885, "3000–6000 Hz_10ms_envelope_correlation": 0.9748019384874343, "3000–6000 Hz_10ms_modulation_mae_db": 0.7676707277731563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1693214363575914, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12384610873825518, "3000–6000 Hz_40ms_envelope_correlation": 0.9891473870032582, "3000–6000 Hz_40ms_modulation_mae_db": 0.5678801824510887, "mr_stft_db": 5.719229008530543, "band_mae_db": 0.2742090489015871, "rms_gain_error_db": -0.19619607603668185, "output_peak": 0.5036293341380343}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68080, "output_samples": 68080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.800888077741739, "spectral_convergence": 0.2973661374327058, "spectral_convergence_status": "ok", "amplitude_floor": 2.0850976744763412e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.4621077244510206, "spectral_convergence": 0.2710017105306715, "spectral_convergence_status": "ok", "amplitude_floor": 1.4381271492089952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8761281042319045, "spectral_convergence": 0.28564663233955545, "spectral_convergence_status": "ok", "amplitude_floor": 9.411979934974602e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005432476915893821, "output_power": 0.0005905606003633119, "delta_db": 0.3626657979676666, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030286578008489277, "output_power": 0.0028762945294621796, "delta_db": -0.22416851171450208, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038180407470981095, "output_power": 0.0003883337908280508, "delta_db": 0.0736462330041304, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002532984259990213, "output_power": 0.00024003048850352484, "delta_db": -0.23366082090781312, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013717204322211298, "output_power": 0.00013664702718472717, "delta_db": -0.016654197753762055, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.29065897621083e-05, "output_power": 8.36580000436477e-05, "delta_db": 0.03918426516442525, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15832997108538327, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11084439914196895, "pitch": null, "output_rms_dbfs": -23.64966419881867, "output_peak": 0.5056645344602257, "reference_rms_dbfs": -23.538819799676702, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.713041302141555}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00026228408676881105, "log_rms_mae_db": 1.0499142042238927, "rms_envelope_nrmse": 0.09823086793460886, "envelope_correlation": 0.9899868709660282, "rms_gain_error_db": -0.11084313715971653, "reference_mean_power": 0.004432288013980724, "output_mean_power": 0.004320596005210911}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 1.6030570640082068e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008005952323264084, "output_power": 0.0007788926773285643, "reference_active": true, "delta_db": -0.11935378594931509}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020547999455019126, "output_power": 0.00019190306359699353, "reference_active": true, "delta_db": -0.29687637538040323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.989786705158196e-05, "output_power": 4.376589199915959e-05, "reference_active": true, "delta_db": 0.4018610548179502}], "active_band_count": 3, "active_band_mae_db": 0.27269707204922283}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00021831174911177267, "log_rms_mae_db": 0.520280567032123, "rms_envelope_nrmse": 0.04596146767393709, "envelope_correlation": 0.9977501469723025, "rms_gain_error_db": -0.1108502011087762, "reference_mean_power": 0.004463740704660867, "output_mean_power": 0.004351249022327063}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 1.4245155353761532e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007405152195784611, "output_power": 0.0007255262341842578, "reference_active": true, "delta_db": -0.08880868227349537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001021109367695941, "output_power": 9.612458035090481e-05, "reference_active": true, "delta_db": -0.2623780362131349}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.434051823082556e-06, "output_power": 2.163943649482055e-06, "reference_active": true, "delta_db": -0.51083873232708}], "active_band_count": 3, "active_band_mae_db": 0.2873418169379034}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00023386465365005232, "log_rms_mae_db": 1.6180190702074284, "rms_envelope_nrmse": 0.12916747420601737, "envelope_correlation": 0.9864557826757002, "rms_gain_error_db": -0.21118366782969342, "reference_mean_power": 0.002112021838569365, "output_mean_power": 0.0020117779221477585}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 9.681983603336737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048709120906038954, "output_power": 0.0004759605600984407, "reference_active": true, "delta_db": -0.10039324649752293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014674692890756196, "output_power": 0.00012615453801118645, "reference_active": true, "delta_db": -0.6566614367326851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8361558797573682e-05, "output_power": 2.832461421584637e-05, "reference_active": true, "delta_db": -0.005660932604622729}], "active_band_count": 3, "active_band_mae_db": 0.2542385386116102}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00020000043006212289, "log_rms_mae_db": 0.941548513883815, "rms_envelope_nrmse": 0.06605294819124838, "envelope_correlation": 0.9965729595366356, "rms_gain_error_db": -0.21118774046636704, "reference_mean_power": 0.002127032769351675, "output_mean_power": 0.0020260744818741475}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.787606965551091e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004705886462934124, "output_power": 0.0004600134062360365, "reference_active": true, "delta_db": -0.09870956343018925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.123456963821095e-05, "output_power": 6.139969575289097e-05, "reference_active": true, "delta_db": -0.6452458573169872}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5465229502268847e-06, "output_power": 1.3756284599237502e-06, "reference_active": true, "delta_db": -0.5085521705005531}], "active_band_count": 3, "active_band_mae_db": 0.41750253041590984}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 7.720600674415526e-05, "log_rms_mae_db": 1.6076240104323005, "rms_envelope_nrmse": 0.1777763078417581, "envelope_correlation": 0.9752082489650673, "rms_gain_error_db": -0.23120583115737267, "reference_mean_power": 0.00033128809071676906, "output_mean_power": 0.0003141125155526387}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 2.2485323271997473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014154142624946028, "output_power": 0.00013412423433536424, "reference_active": true, "delta_db": -0.23376311724179782}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6917189688068767e-05, "output_power": 2.5560460239129982e-05, "reference_active": true, "delta_db": -0.22461045616055797}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.319943953798946e-06, "output_power": 1.0070549512651057e-05, "reference_active": true, "delta_db": 2.023399421327331}], "active_band_count": 3, "active_band_mae_db": 0.8272576649098956}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 6.30102051319373e-05, "log_rms_mae_db": 1.1496837315532327, "rms_envelope_nrmse": 0.10223216725538371, "envelope_correlation": 0.9916401820677322, "rms_gain_error_db": -0.23120837664449417, "reference_mean_power": 0.0003336427865109939, "output_mean_power": 0.00031634494712454885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 2.017085687613918e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013578457159768264, "output_power": 0.0001306453597963701, "reference_active": true, "delta_db": -0.1675643716033029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.310083041117988e-05, "output_power": 1.1917401842336946e-05, "reference_active": true, "delta_db": -0.4111724123477316}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7328752869332565e-07, "output_power": 4.850055391788836e-07, "reference_active": true, "delta_db": 1.1370321798851173}], "active_band_count": 3, "active_band_mae_db": 0.5719229879453839}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 5.974138539374114e-05, "log_rms_mae_db": 1.5418827032549032, "rms_envelope_nrmse": 0.22362320773125025, "envelope_correlation": 0.9664867847201025, "rms_gain_error_db": 0.08262638068025688, "reference_mean_power": 8.138813397574021e-05, "output_mean_power": 8.295140176202645e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 8.567729341398199e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0300307346839795e-05, "output_power": 4.103151786495308e-05, "reference_active": true, "delta_db": 0.07809224708416686}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.809515102528064e-06, "output_power": 5.187073920010182e-06, "reference_active": true, "delta_db": -0.49217447957310745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.571956353556118e-06, "output_power": 4.214333594214762e-06, "reference_active": true, "delta_db": 0.7182276658432474}], "active_band_count": 3, "active_band_mae_db": 0.42949813083350724}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 4.362120322552387e-05, "log_rms_mae_db": 1.2584617619917302, "rms_envelope_nrmse": 0.15155208898647898, "envelope_correlation": 0.9838983983163263, "rms_gain_error_db": 0.08262385332284955, "reference_mean_power": 8.196654666810843e-05, "output_mean_power": 8.354087573690493e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.099518469769939e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.90961297245674e-05, "output_power": 4.018522877243763e-05, "reference_active": true, "delta_db": 0.11932678035357647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1532747685680723e-06, "output_power": 2.7147094998690006e-06, "reference_active": true, "delta_db": -0.6503845303637489}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4027458558124955e-07, "output_power": 1.7573514274067403e-07, "reference_active": true, "delta_db": 0.97879624155932}], "active_band_count": 3, "active_band_mae_db": 0.5828358507588818}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8229435629646623, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dda8e56eca581b16ff50d2db8f9a31fdbe093b7750b834fed1c6ee481a7c7012", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9805260365066943, "standard_deviation": 0.0032115479469524556, "minimum": 0.9752082489650673, "maximum": 0.9860455368278838, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6385569771663535, "standard_deviation": 0.053280418819401404, "minimum": 1.5804617452581005, "maximum": 1.7451234657618808, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6509334669003081, "standard_deviation": 0.18658338604119018, "minimum": 0.3361570713856021, "maximum": 0.8272576649098956, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15952385589649506, "standard_deviation": 0.012230046875713716, "minimum": 0.13570638382297961, "maximum": 0.1777763078417581, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938333881477908, "standard_deviation": 0.0012037714742881788, "minimum": 0.9916401820677322, "maximum": 0.9952284591671029, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1629231647790266, "standard_deviation": 0.04744780030049143, "minimum": 1.0897498836667758, "maximum": 1.256019128165454, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7311484122738581, "standard_deviation": 0.21393572230657312, "minimum": 0.34771875315080886, "maximum": 0.9744344545731458, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09306511743440826, "standard_deviation": 0.006890327913135153, "minimum": 0.08167875254766294, "maximum": 0.10223216725538371, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9735216486310312, "standard_deviation": 0.0038205871897576535, "minimum": 0.9664867847201025, "maximum": 0.978462707832021, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5265813087467253, "standard_deviation": 0.02591601296404775, "minimum": 1.484059756309818, "maximum": 1.568255680058505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5342406864615425, "standard_deviation": 0.14316333472301307, "minimum": 0.3664511669858796, "maximum": 0.7676707277731563, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19815987523565015, "standard_deviation": 0.014505689635354702, "minimum": 0.1774788589080573, "maximum": 0.22362320773125025, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9882030203480322, "standard_deviation": 0.002094619524607998, "minimum": 0.9838983983163263, "maximum": 0.9899755205477733, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2065695425264384, "standard_deviation": 0.03426726361108797, "minimum": 1.1568970834549293, "maximum": 1.2584617619917302, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5379134325704656, "standard_deviation": 0.1421613788327941, "minimum": 0.33546911385063954, "maximum": 0.7335403919648843, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1308059978069972, "standard_deviation": 0.011818844929454828, "minimum": 0.11969903369852329, "maximum": 0.15155208898647898, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9867991886491203, "standard_deviation": 0.0009284283394370384, "minimum": 0.9855810455905475, "maximum": 0.988336609769946, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6884082425297176, "standard_deviation": 0.04652561317364511, "minimum": 1.6180190702074284, "maximum": 1.7640271013701252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4384411964748328, "standard_deviation": 0.14623451749720487, "minimum": 0.2542385386116102, "maximum": 0.7527579246204231, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1281905423988186, "standard_deviation": 0.00502382592448353, "minimum": 0.11938943848856275, "maximum": 0.1337003557419574, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960036495079061, "standard_deviation": 0.0004906350184811369, "minimum": 0.9952079777294461, "maximum": 0.9965729595366356, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9711281144953259, "standard_deviation": 0.024329064440055566, "minimum": 0.941548513883815, "maximum": 1.00160438520819, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5048662495815622, "standard_deviation": 0.08426611819802741, "minimum": 0.3843595427210101, "maximum": 0.6384371900423795, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07172111640554726, "standard_deviation": 0.004496127990127892, "minimum": 0.06553407694702633, "maximum": 0.07682372636040012, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.26505205081803157, "standard_deviation": 0.05126961240694872, "minimum": 0.15832997108538327, "maximum": 0.32319584780435795, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.722855200714255, "standard_deviation": 0.009707307509562045, "minimum": 5.713041302141555, "maximum": 5.743137590140673, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4977857447971593, "standard_deviation": 0.013453299768525311, "minimum": 0.47843770791729345, "maximum": 0.5235887164470976, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11117715614726287, "standard_deviation": 0.07750145455526823, "minimum": -0.20849314505583524, "maximum": -0.011469870092907272, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9910628206513447, "standard_deviation": 0.001019096809384599, "minimum": 0.9899513622950392, "maximum": 0.9926191882519757, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0492887593588434, "standard_deviation": 0.031569402040544065, "minimum": 0.9979617971099147, "maximum": 1.1001499085311999, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.21687733580851956, "standard_deviation": 0.0947746805890362, "minimum": 0.08635797987894955, "maximum": 0.34032194756751905, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09313748861050475, "standard_deviation": 0.005757194441568681, "minimum": 0.08435461138988137, "maximum": 0.09966107900343238, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9975176781512923, "standard_deviation": 0.00041427834306635215, "minimum": 0.9967774551990777, "maximum": 0.997941344025211, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5491015601620364, "standard_deviation": 0.03469779810814147, "minimum": 0.4944872944403882, "maximum": 0.5906790934519254, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3103336030558099, "standard_deviation": 0.1305822908498535, "minimum": 0.09744255215839019, "maximum": 0.5154596648234513, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04872512015581606, "standard_deviation": 0.0043196075319010445, "minimum": 0.04253212530439596, "maximum": 0.055566125559275176, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-157963-0018", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.151418447494507, "predicted_mos": 2.151418447494507, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "0a626a3179a79e509272bfc0dcabcb0de883c51e7fecd65a9ee51979e50b1335", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.8563574999570847, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0640979188855602, "全帯域_10ms_rms_envelope_nrmse": 0.10526400955622588, "全帯域_10ms_envelope_correlation": 0.9884725501148574, "全帯域_10ms_modulation_mae_db": 0.3748037126082118, "全帯域_40ms_log_rms_mae_db": 0.5206019377235194, "全帯域_40ms_rms_envelope_nrmse": 0.04846252478940844, "全帯域_40ms_envelope_correlation": 0.9974792580787394, "全帯域_40ms_modulation_mae_db": 0.16423261495962915, "300–1000 Hz_10ms_log_rms_mae_db": 1.6026485508683332, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1422120991658539, "300–1000 Hz_10ms_envelope_correlation": 0.9833939085800397, "300–1000 Hz_10ms_modulation_mae_db": 0.32158198879546235, "300–1000 Hz_40ms_log_rms_mae_db": 0.8912327700307899, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07389121089471438, "300–1000 Hz_40ms_envelope_correlation": 0.9954498833819675, "300–1000 Hz_40ms_modulation_mae_db": 0.25466929896239787, "1000–3000 Hz_10ms_log_rms_mae_db": 1.560766154029647, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17758954414151976, "1000–3000 Hz_10ms_envelope_correlation": 0.9753733364042496, "1000–3000 Hz_10ms_modulation_mae_db": 0.8148269788660419, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1145420105428443, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10347342619415129, "1000–3000 Hz_40ms_envelope_correlation": 0.9915683793830121, "1000–3000 Hz_40ms_modulation_mae_db": 0.5124153912354726, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4404121264766299, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2169178113413945, "3000–6000 Hz_10ms_envelope_correlation": 0.9686175659512096, "3000–6000 Hz_10ms_modulation_mae_db": 0.42851634189157645, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1843881753273935, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1443314452865318, "3000–6000 Hz_40ms_envelope_correlation": 0.9854417346479641, "3000–6000 Hz_40ms_modulation_mae_db": 0.5526569447165616, "mr_stft_db": 5.649954380181537, "band_mae_db": 0.2011974253817451, "rms_gain_error_db": -0.11001320471602213, "output_peak": 0.5079803290298774}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.01810608593851, "全帯域_10ms_rms_envelope_nrmse": 0.09716101409871633, "全帯域_10ms_envelope_correlation": 0.9903230050747117, "全帯域_10ms_modulation_mae_db": 0.05723195540210447, "全帯域_40ms_log_rms_mae_db": 0.50610882020246, "全帯域_40ms_rms_envelope_nrmse": 0.05121741205546224, "全帯域_40ms_envelope_correlation": 0.9973346777803234, "全帯域_40ms_modulation_mae_db": 0.2327310603846143, "300–1000 Hz_10ms_log_rms_mae_db": 1.6741887902035315, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13406695500922614, "300–1000 Hz_10ms_envelope_correlation": 0.9855286489054919, "300–1000 Hz_10ms_modulation_mae_db": 0.3520955041378678, "300–1000 Hz_40ms_log_rms_mae_db": 0.9798092027928487, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07592504713755081, "300–1000 Hz_40ms_envelope_correlation": 0.9955416632965289, "300–1000 Hz_40ms_modulation_mae_db": 0.48847906925983314, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5157024098987544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15469375159510768, "1000–3000 Hz_10ms_envelope_correlation": 0.9814251374210073, "1000–3000 Hz_10ms_modulation_mae_db": 0.28158608997802675, "1000–3000 Hz_40ms_log_rms_mae_db": 1.07046265917786, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09267865283751642, "1000–3000 Hz_40ms_envelope_correlation": 0.993271513590818, "1000–3000 Hz_40ms_modulation_mae_db": 0.39159330900891365, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4142496765520884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21288019152616772, "3000–6000 Hz_10ms_envelope_correlation": 0.971879203130602, "3000–6000 Hz_10ms_modulation_mae_db": 0.8065025054766192, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1030158672924917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14030815736610458, "3000–6000 Hz_40ms_envelope_correlation": 0.9877947507558744, "3000–6000 Hz_40ms_modulation_mae_db": 0.9808888833297663, "mr_stft_db": 5.653294085389644, "band_mae_db": 0.30016785381060135, "rms_gain_error_db": -0.1564307570580263, "output_peak": 0.5041555720715669}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.109809137057388, "全帯域_10ms_rms_envelope_nrmse": 0.09554799002573576, "全帯域_10ms_envelope_correlation": 0.9906813716638484, "全帯域_10ms_modulation_mae_db": 0.25705589325826356, "全帯域_40ms_log_rms_mae_db": 0.6027393282853839, "全帯域_40ms_rms_envelope_nrmse": 0.048693171227759575, "全帯域_40ms_envelope_correlation": 0.9976384298521438, "全帯域_40ms_modulation_mae_db": 0.49495089064059594, "300–1000 Hz_10ms_log_rms_mae_db": 1.7459775018517105, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13461292688333035, "300–1000 Hz_10ms_envelope_correlation": 0.9855085294373882, "300–1000 Hz_10ms_modulation_mae_db": 0.9101123930869198, "300–1000 Hz_40ms_log_rms_mae_db": 0.8802813547877631, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07672739533335299, "300–1000 Hz_40ms_envelope_correlation": 0.9955650632469588, "300–1000 Hz_40ms_modulation_mae_db": 0.7517772316437422, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5978744803528073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.164907060881193, "1000–3000 Hz_10ms_envelope_correlation": 0.9790113970978656, "1000–3000 Hz_10ms_modulation_mae_db": 0.3097127474875916, "1000–3000 Hz_40ms_log_rms_mae_db": 1.11936461222481, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08819554403618782, "1000–3000 Hz_40ms_envelope_correlation": 0.9943452583347773, "1000–3000 Hz_40ms_modulation_mae_db": 0.44399040725375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4155147955084533, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1735639263437889, "3000–6000 Hz_10ms_envelope_correlation": 0.9794737359355851, "3000–6000 Hz_10ms_modulation_mae_db": 0.3268134071629245, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1066539994062572, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11517663421124244, "3000–6000 Hz_40ms_envelope_correlation": 0.9907502226870819, "3000–6000 Hz_40ms_modulation_mae_db": 0.49296108963827745, "mr_stft_db": 5.673321020992213, "band_mae_db": 0.33853705198621226, "rms_gain_error_db": -0.16177080617809247, "output_peak": 0.5009873666775981}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0543297765543187, "全帯域_10ms_rms_envelope_nrmse": 0.08472456501192421, "全帯域_10ms_envelope_correlation": 0.9925438174108537, "全帯域_10ms_modulation_mae_db": 0.3203804254282746, "全帯域_40ms_log_rms_mae_db": 0.5567362581945947, "全帯域_40ms_rms_envelope_nrmse": 0.04674684423711674, "全帯域_40ms_envelope_correlation": 0.9975320760461199, "全帯域_40ms_modulation_mae_db": 0.4459127800143781, "300–1000 Hz_10ms_log_rms_mae_db": 1.691143074734755, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12432770553214484, "300–1000 Hz_10ms_envelope_correlation": 0.9873143117164463, "300–1000 Hz_10ms_modulation_mae_db": 0.5860051773724596, "300–1000 Hz_40ms_log_rms_mae_db": 0.9258427260890351, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0689827721126623, "300–1000 Hz_40ms_envelope_correlation": 0.9957786767132035, "300–1000 Hz_40ms_modulation_mae_db": 0.5992886214288492, "1000–3000 Hz_10ms_log_rms_mae_db": 1.521883348229333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15761837287464933, "1000–3000 Hz_10ms_envelope_correlation": 0.9810622147694208, "1000–3000 Hz_10ms_modulation_mae_db": 0.5837508802455843, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0386412339640085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0881153996924413, "1000–3000 Hz_40ms_envelope_correlation": 0.9945864529168454, "1000–3000 Hz_40ms_modulation_mae_db": 0.6859836277165794, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4629442943735338, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2081160917520424, "3000–6000 Hz_10ms_envelope_correlation": 0.9713669847390337, "3000–6000 Hz_10ms_modulation_mae_db": 0.5448256276665396, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1395490198475982, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1452854335336705, "3000–6000 Hz_40ms_envelope_correlation": 0.987101398941909, "3000–6000 Hz_40ms_modulation_mae_db": 0.3364338524635044, "mr_stft_db": 5.652674382913257, "band_mae_db": 0.31367814108134223, "rms_gain_error_db": -0.02807336723190252, "output_peak": 0.5019058878668406}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.066987438976075, "全帯域_10ms_rms_envelope_nrmse": 0.09229994170852832, "全帯域_10ms_envelope_correlation": 0.9911949904741829, "全帯域_10ms_modulation_mae_db": 0.27402100043208566, "全帯域_40ms_log_rms_mae_db": 0.5755328776229987, "全帯域_40ms_rms_envelope_nrmse": 0.04693866837717926, "全帯域_40ms_envelope_correlation": 0.9976717669835592, "全帯域_40ms_modulation_mae_db": 0.3829873759552065, "300–1000 Hz_10ms_log_rms_mae_db": 1.6600946156540308, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13845029727537655, "300–1000 Hz_10ms_envelope_correlation": 0.984408729086298, "300–1000 Hz_10ms_modulation_mae_db": 0.5165478403431573, "300–1000 Hz_40ms_log_rms_mae_db": 0.926872790772843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07625057027824524, "300–1000 Hz_40ms_envelope_correlation": 0.9953481298275482, "300–1000 Hz_40ms_modulation_mae_db": 0.5289671850101731, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6745235474713975, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16795178607474573, "1000–3000 Hz_10ms_envelope_correlation": 0.9786129016650239, "1000–3000 Hz_10ms_modulation_mae_db": 0.7200635155487237, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1848607970122704, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09315770125862237, "1000–3000 Hz_40ms_envelope_correlation": 0.9942686795962595, "1000–3000 Hz_40ms_modulation_mae_db": 0.8626174375164464, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4104796253744665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17956959782348367, "3000–6000 Hz_10ms_envelope_correlation": 0.9779725764965798, "3000–6000 Hz_10ms_modulation_mae_db": 0.2428117563858224, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1344284934089452, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11971247898417964, "3000–6000 Hz_40ms_envelope_correlation": 0.9899463416135654, "3000–6000 Hz_40ms_modulation_mae_db": 0.43865052046139236, "mr_stft_db": 5.65505480709546, "band_mae_db": 0.3258116287719372, "rms_gain_error_db": -0.11703573976639975, "output_peak": 0.5254739798762155}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0662370086222983, "全帯域_10ms_rms_envelope_nrmse": 0.09137452533017552, "全帯域_10ms_envelope_correlation": 0.9913140454980941, "全帯域_10ms_modulation_mae_db": 0.033764563214371, "全帯域_40ms_log_rms_mae_db": 0.5523576711469232, "全帯域_40ms_rms_envelope_nrmse": 0.05109586302710861, "全帯域_40ms_envelope_correlation": 0.9970772374819057, "全帯域_40ms_modulation_mae_db": 0.22542366659819388, "300–1000 Hz_10ms_log_rms_mae_db": 1.6928796792375642, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.130213893868748, "300–1000 Hz_10ms_envelope_correlation": 0.9862032092771724, "300–1000 Hz_10ms_modulation_mae_db": 0.3338444939741318, "300–1000 Hz_40ms_log_rms_mae_db": 0.9211588275874149, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07678320096183469, "300–1000 Hz_40ms_envelope_correlation": 0.9951269359907443, "300–1000 Hz_40ms_modulation_mae_db": 0.43385721373406555, "1000–3000 Hz_10ms_log_rms_mae_db": 1.571068424365743, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13369675118379046, "1000–3000 Hz_10ms_envelope_correlation": 0.9868831024065626, "1000–3000 Hz_10ms_modulation_mae_db": 1.0170474099811422, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0809387136108544, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07930091886996869, "1000–3000 Hz_40ms_envelope_correlation": 0.9960532615968462, "1000–3000 Hz_40ms_modulation_mae_db": 1.163431213098818, "3000–6000 Hz_10ms_log_rms_mae_db": 1.469632324204244, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19664783443309966, "3000–6000 Hz_10ms_envelope_correlation": 0.9740540666201921, "3000–6000 Hz_10ms_modulation_mae_db": 0.6712834368800769, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1253347118610524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12257675719594875, "3000–6000 Hz_40ms_envelope_correlation": 0.9894079992114199, "3000–6000 Hz_40ms_modulation_mae_db": 0.1637419360354456, "mr_stft_db": 5.651506814514444, "band_mae_db": 0.2717657852228946, "rms_gain_error_db": -0.0732316200135242, "output_peak": 0.49071263251009734}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9754843135464082, "全帯域_10ms_rms_envelope_nrmse": 0.08363055506672604, "全帯域_10ms_envelope_correlation": 0.9927094065291996, "全帯域_10ms_modulation_mae_db": 0.07984995064776775, "全帯域_40ms_log_rms_mae_db": 0.5433665282513221, "全帯域_40ms_rms_envelope_nrmse": 0.04271092072572642, "全帯域_40ms_envelope_correlation": 0.9979215823026246, "全帯域_40ms_modulation_mae_db": 0.17430885213671712, "300–1000 Hz_10ms_log_rms_mae_db": 1.6999270306510939, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12076456620290975, "300–1000 Hz_10ms_envelope_correlation": 0.9881455799455079, "300–1000 Hz_10ms_modulation_mae_db": 0.250031990242696, "300–1000 Hz_40ms_log_rms_mae_db": 0.9406725340158394, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06894210540875939, "300–1000 Hz_40ms_envelope_correlation": 0.9958324409075113, "300–1000 Hz_40ms_modulation_mae_db": 0.3134533602535397, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6054778347498413, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15491350212418742, "1000–3000 Hz_10ms_envelope_correlation": 0.9829282882490905, "1000–3000 Hz_10ms_modulation_mae_db": 0.6025985588139912, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1522911326082808, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09855923925580325, "1000–3000 Hz_40ms_envelope_correlation": 0.994431059764591, "1000–3000 Hz_40ms_modulation_mae_db": 0.6060971075288948, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4172088571907036, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.203242334821193, "3000–6000 Hz_10ms_envelope_correlation": 0.9733370201891876, "3000–6000 Hz_10ms_modulation_mae_db": 0.9322273269210105, "3000–6000 Hz_40ms_log_rms_mae_db": 1.167798020716975, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13739700833179688, "3000–6000 Hz_40ms_envelope_correlation": 0.9874688750176729, "3000–6000 Hz_40ms_modulation_mae_db": 0.7703139589248438, "mr_stft_db": 5.64372678298258, "band_mae_db": 0.2437679557427359, "rms_gain_error_db": -0.015931567943573203, "output_peak": 0.48348642374758316}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0791367340801905, "全帯域_10ms_rms_envelope_nrmse": 0.09789701694348248, "全帯域_10ms_envelope_correlation": 0.9902883135013594, "全帯域_10ms_modulation_mae_db": 0.4132814293559699, "全帯域_40ms_log_rms_mae_db": 0.523625240604613, "全帯域_40ms_rms_envelope_nrmse": 0.053019456071654675, "全帯域_40ms_envelope_correlation": 0.997277560812507, "全帯域_40ms_modulation_mae_db": 0.5412708722570003, "300–1000 Hz_10ms_log_rms_mae_db": 1.6296709578126405, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12550464177952267, "300–1000 Hz_10ms_envelope_correlation": 0.9876221169271534, "300–1000 Hz_10ms_modulation_mae_db": 0.7142012655765244, "300–1000 Hz_40ms_log_rms_mae_db": 0.875584874252232, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06906869290783202, "300–1000 Hz_40ms_envelope_correlation": 0.9967250146854735, "300–1000 Hz_40ms_modulation_mae_db": 0.4371847096089634, "1000–3000 Hz_10ms_log_rms_mae_db": 1.598061884175422, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15894735113865735, "1000–3000 Hz_10ms_envelope_correlation": 0.981487309112839, "1000–3000 Hz_10ms_modulation_mae_db": 0.7887411972775279, "1000–3000 Hz_40ms_log_rms_mae_db": 1.097167324544848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10094616324009785, "1000–3000 Hz_40ms_envelope_correlation": 0.9934576374757437, "1000–3000 Hz_40ms_modulation_mae_db": 1.0113973914432888, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4676289669654472, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19880973990763454, "3000–6000 Hz_10ms_envelope_correlation": 0.9733222829781227, "3000–6000 Hz_10ms_modulation_mae_db": 0.886933472991029, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1497354109300233, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12724717833066412, "3000–6000 Hz_40ms_envelope_correlation": 0.9885323478227283, "3000–6000 Hz_40ms_modulation_mae_db": 0.8810333466498852, "mr_stft_db": 5.6558749350494155, "band_mae_db": 0.29782446037601495, "rms_gain_error_db": -0.18775754747383813, "output_peak": 0.5042733125731095}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68080, "output_samples": 68080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.731802096914679, "spectral_convergence": 0.301770994055696, "spectral_convergence_status": "ok", "amplitude_floor": 2.0850976744763412e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.396784924944527, "spectral_convergence": 0.274963581062731, "spectral_convergence_status": "ok", "amplitude_floor": 1.4381271492089952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.821276118685404, "spectral_convergence": 0.2881170405124047, "spectral_convergence_status": "ok", "amplitude_floor": 9.411979934974602e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005432476915893821, "output_power": 0.0005870340605472386, "delta_db": 0.33665410800949813, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030286578008489277, "output_power": 0.002872809394852398, "delta_db": -0.22943394067542863, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038180407470981095, "output_power": 0.0004022463496392444, "delta_db": 0.22651552610260559, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002532984259990213, "output_power": 0.00023762207768727897, "delta_db": -0.27745702198946326, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013717204322211298, "output_power": 0.0001334156950284513, "delta_db": -0.1205868463366164, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.29065897621083e-05, "output_power": 8.32228841248187e-05, "delta_db": 0.016537109176858564, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2011974253817451, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11001320471602213, "pitch": null, "output_rms_dbfs": -23.648833004392724, "output_peak": 0.5079803290298774, "reference_rms_dbfs": -23.538819799676702, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.649954380181537}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00026228408676881105, "log_rms_mae_db": 1.0640979188855602, "rms_envelope_nrmse": 0.10526400955622588, "envelope_correlation": 0.9884725501148574, "rms_gain_error_db": -0.11001127664961832, "reference_mean_power": 0.004432288013980724, "output_mean_power": 0.004321423664226633}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 1.6030570640082068e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008005952323264084, "output_power": 0.0007748793691820641, "reference_active": true, "delta_db": -0.1417890133334229}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020547999455019126, "output_power": 0.00019394724966084317, "reference_active": true, "delta_db": -0.2508592022411186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.989786705158196e-05, "output_power": 4.7219998729479946e-05, "reference_active": true, "delta_db": 0.7317629222500941}], "active_band_count": 3, "active_band_mae_db": 0.3748037126082118}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00021831174911177267, "log_rms_mae_db": 0.5206019377235194, "rms_envelope_nrmse": 0.04846252478940844, "envelope_correlation": 0.9974792580787394, "rms_gain_error_db": -0.11001335910437798, "reference_mean_power": 0.004463740704660867, "output_mean_power": 0.004352087545253225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 1.4245155353761532e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007405152195784611, "output_power": 0.0007205706636000779, "reference_active": true, "delta_db": -0.11857412036201942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001021109367695941, "output_power": 9.751889696157528e-05, "reference_active": true, "delta_db": -0.19983480120980987}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.434051823082556e-06, "output_power": 2.3383037608718097e-06, "reference_active": true, "delta_db": -0.1742889233070581}], "active_band_count": 3, "active_band_mae_db": 0.16423261495962915}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00023386465365005232, "log_rms_mae_db": 1.6026485508683332, "rms_envelope_nrmse": 0.1422120991658539, "envelope_correlation": 0.9833939085800397, "rms_gain_error_db": -0.17773304892011654, "reference_mean_power": 0.002112021838569365, "output_mean_power": 0.0020273330465419444}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 9.681983603336737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048709120906038954, "output_power": 0.0004781245215324395, "reference_active": true, "delta_db": -0.08069273677438743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014674692890756196, "output_power": 0.000129520249360013, "reference_active": true, "delta_db": -0.5423134930516348}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8361558797573682e-05, "output_power": 3.0683441587537584e-05, "reference_active": true, "delta_db": 0.3417397365603648}], "active_band_count": 3, "active_band_mae_db": 0.32158198879546235}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00020000043006212289, "log_rms_mae_db": 0.8912327700307899, "rms_envelope_nrmse": 0.07389121089471438, "envelope_correlation": 0.9954498833819675, "rms_gain_error_db": -0.17773443395327554, "reference_mean_power": 0.002127032769351675, "output_mean_power": 0.00204174141119493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.787606965551091e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004705886462934124, "output_power": 0.0004615401004226311, "reference_active": true, "delta_db": -0.0843200469763293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.123456963821095e-05, "output_power": 6.340813593400746e-05, "reference_active": true, "delta_db": -0.5054581883767792}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5465229502268847e-06, "output_power": 1.6098274756631765e-06, "reference_active": true, "delta_db": 0.17422966153408503}], "active_band_count": 3, "active_band_mae_db": 0.25466929896239787}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 7.720600674415526e-05, "log_rms_mae_db": 1.560766154029647, "rms_envelope_nrmse": 0.17758954414151976, "envelope_correlation": 0.9753733364042496, "rms_gain_error_db": -0.2772114912671128, "reference_mean_power": 0.00033128809071676906, "output_mean_power": 0.0003108026226344708}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 2.2485323271997473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014154142624946028, "output_power": 0.00013346738299864033, "reference_active": true, "delta_db": -0.25508422488840915}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6917189688068767e-05, "output_power": 2.5414360865486318e-05, "reference_active": true, "delta_db": -0.2495052274015801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.319943953798946e-06, "output_power": 9.878758652721984e-06, "reference_active": true, "delta_db": 1.9398914843081363}], "active_band_count": 3, "active_band_mae_db": 0.8148269788660419}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 6.30102051319373e-05, "log_rms_mae_db": 1.1145420105428443, "rms_envelope_nrmse": 0.10347342619415129, "envelope_correlation": 0.9915683793830121, "rms_gain_error_db": -0.27721268785849446, "reference_mean_power": 0.0003336427865109939, "output_mean_power": 0.0003130116276596904}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 2.017085687613918e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013578457159768264, "output_power": 0.000130336004870652, "reference_active": true, "delta_db": -0.17786021784383949}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.310083041117988e-05, "output_power": 1.1748143420272676e-05, "reference_active": true, "delta_db": -0.47329585071817554}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7328752869332565e-07, "output_power": 4.577754425686512e-07, "reference_active": true, "delta_db": 0.8860901051444028}], "active_band_count": 3, "active_band_mae_db": 0.5124153912354726}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 5.974138539374114e-05, "log_rms_mae_db": 1.4404121264766299, "rms_envelope_nrmse": 0.2169178113413945, "envelope_correlation": 0.9686175659512096, "rms_gain_error_db": 0.07141994826790615, "reference_mean_power": 8.138813397574021e-05, "output_mean_power": 8.273763184280417e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 8.567729341398199e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0300307346839795e-05, "output_power": 4.1546475525090724e-05, "reference_active": true, "delta_db": 0.1322582930494494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.809515102528064e-06, "output_power": 5.214219409013197e-06, "reference_active": true, "delta_db": -0.4695058315970738}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.571956353556118e-06, "output_power": 4.181042995568555e-06, "reference_active": true, "delta_db": 0.6837849010282061}], "active_band_count": 3, "active_band_mae_db": 0.42851634189157645}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 4.362120322552387e-05, "log_rms_mae_db": 1.1843881753273935, "rms_envelope_nrmse": 0.1443314452865318, "envelope_correlation": 0.9854417346479641, "rms_gain_error_db": 0.07141922742578048, "reference_mean_power": 8.196654666810843e-05, "output_mean_power": 8.332562137428737e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.099518469769939e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.90961297245674e-05, "output_power": 4.0683001063305704e-05, "reference_active": true, "delta_db": 0.17279214985656888}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1532747685680723e-06, "output_power": 2.653013600844434e-06, "reference_active": true, "delta_db": -0.7502233931775439}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4027458558124955e-07, "output_power": 1.6614012138607282e-07, "reference_active": true, "delta_db": 0.734955291115572}], "active_band_count": 3, "active_band_mae_db": 0.5526569447165616}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8231091163625878, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6520e9358a2cb68ea6867091a5a54dfbbd9df7d41b6ada23e7774550ec45779f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9808479608907574, "standard_deviation": 0.0033780482156430835, "minimum": 0.9753733364042496, "maximum": 0.9868831024065626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5806697604091182, "standard_deviation": 0.05097254211254264, "minimum": 1.5157024098987544, "maximum": 1.6745235474713975, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6397909222748287, "standard_deviation": 0.25167326148909214, "minimum": 0.28158608997802675, "maximum": 1.0170474099811422, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1587897650017313, "standard_deviation": 0.012748415642852258, "minimum": 0.13369675118379046, "maximum": 0.17758954414151976, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939977803323616, "standard_deviation": 0.0012914335190142132, "minimum": 0.9915683793830121, "maximum": 0.9960532615968462, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.107283560460722, "standard_deviation": 0.04645502130898248, "minimum": 1.0386412339640085, "maximum": 1.1848607970122704, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7096907356002704, "standard_deviation": 0.27833375874054717, "minimum": 0.39159330900891365, "maximum": 1.163431213098818, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09305338067309862, "standard_deviation": 0.007914287138565084, "minimum": 0.07930091886996869, "maximum": 0.10347342619415129, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.973752929505064, "standard_deviation": 0.003516445227863822, "minimum": 0.9686175659512096, "maximum": 0.9794737359355851, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4372588333306957, "standard_deviation": 0.02608071562732726, "minimum": 1.4104796253744665, "maximum": 1.469632324204244, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6049892344219499, "standard_deviation": 0.2604656545439731, "minimum": 0.2428117563858224, "maximum": 0.9322273269210105, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19871844099360053, "standard_deviation": 0.015326341765475985, "minimum": 0.1735639263437889, "maximum": 0.2169178113413945, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9883054588372769, "standard_deviation": 0.0017133014141829767, "minimum": 0.9854417346479641, "maximum": 0.9907502226870819, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.138862962348842, "standard_deviation": 0.028166781413590393, "minimum": 1.1030158672924917, "maximum": 1.1843881753273935, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5770850665274596, "standard_deviation": 0.2798935345635609, "minimum": 0.1637419360354456, "maximum": 0.9808888833297663, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13150438665501735, "standard_deviation": 0.01177255301856899, "minimum": 0.11517663421124244, "maximum": 0.1452854335336705, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9860156292344372, "standard_deviation": 0.0016383398618584385, "minimum": 0.9833939085800397, "maximum": 0.9881455799455079, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6745662751267076, "standard_deviation": 0.04432062125741895, "minimum": 1.6026485508683332, "maximum": 1.7459775018517105, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.49805258169115235, "standard_deviation": 0.22850918202606613, "minimum": 0.250031990242696, "maximum": 0.9101123930869198, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13126913571463902, "standard_deviation": 0.007398828304222298, "minimum": 0.12076456620290975, "maximum": 0.1422120991658539, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.995670976006242, "standard_deviation": 0.0004819155713665443, "minimum": 0.9951269359907443, "maximum": 0.9967250146854735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9176818850410957, "standard_deviation": 0.0346864494352595, "minimum": 0.875584874252232, "maximum": 0.9798092027928487, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4759595862376955, "standard_deviation": 0.1571649420082914, "minimum": 0.25466929896239787, "maximum": 0.7517772316437422, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07332137437936898, "standard_deviation": 0.0036908381194892996, "minimum": 0.06894210540875939, "maximum": 0.07678320096183469, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.28659378779668543, "standard_deviation": 0.0457111231823608, "minimum": 0.2011974253817451, "maximum": 0.33853705198621226, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.654425901139819, "standard_deviation": 0.008509291517832671, "minimum": 5.64372678298258, "maximum": 5.673321020992213, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5023719380441111, "standard_deviation": 0.012344410291541985, "minimum": 0.48348642374758316, "maximum": 0.5254739798762155, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.10628057629767235, "standard_deviation": 0.06294425368367741, "minimum": -0.18775754747383813, "maximum": -0.015931567943573203, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9909409375333884, "standard_deviation": 0.001355861796583495, "minimum": 0.9884725501148574, "maximum": 0.9927094065291996, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0542735517075936, "standard_deviation": 0.04074221624261792, "minimum": 0.9754843135464082, "maximum": 1.109809137057388, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2262986162933811, "standard_deviation": 0.14937549423511062, "minimum": 0.033764563214371, "maximum": 0.4132814293559699, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09348745221768931, "standard_deviation": 0.007128350757342106, "minimum": 0.08363055506672604, "maximum": 0.10526400955622588, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974915736672404, "standard_deviation": 0.0002626317924968432, "minimum": 0.9970772374819057, "maximum": 0.9979215823026246, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5476335827539769, "standard_deviation": 0.03155451867343357, "minimum": 0.50610882020246, "maximum": 0.6027393282853839, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3327272641182919, "standard_deviation": 0.1512538661606698, "minimum": 0.16423261495962915, "maximum": 0.5412708722570003, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048610607563926996, "standard_deviation": 0.0032406770451455296, "minimum": 0.04271092072572642, "maximum": 0.053019456071654675, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-157963-0018", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.041738748550415, "predicted_mos": 2.041738748550415, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "1b6562d4cb8feea6e4fa3c7e237fa4d45a69b4f9aa9d20f6e718f8000d4c9abb", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.778285800013691, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0062212979956104, "全帯域_10ms_rms_envelope_nrmse": 0.09567098600521934, "全帯域_10ms_envelope_correlation": 0.9908867361051367, "全帯域_10ms_modulation_mae_db": 0.2632448664492354, "全帯域_40ms_log_rms_mae_db": 0.5611773209706499, "全帯域_40ms_rms_envelope_nrmse": 0.05714512491252644, "全帯域_40ms_envelope_correlation": 0.9969256910749341, "全帯域_40ms_modulation_mae_db": 0.3613049695688863, "300–1000 Hz_10ms_log_rms_mae_db": 1.9707068448757812, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1406145394677927, "300–1000 Hz_10ms_envelope_correlation": 0.9843827855378765, "300–1000 Hz_10ms_modulation_mae_db": 0.5384059995614119, "300–1000 Hz_40ms_log_rms_mae_db": 1.1858005448742175, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08664452034010117, "300–1000 Hz_40ms_envelope_correlation": 0.9944580094695556, "300–1000 Hz_40ms_modulation_mae_db": 0.3140435697346375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8924435564326474, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15949011673240163, "1000–3000 Hz_10ms_envelope_correlation": 0.9814016571443287, "1000–3000 Hz_10ms_modulation_mae_db": 0.4454498466002177, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2810407187588364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09706150722875745, "1000–3000 Hz_40ms_envelope_correlation": 0.9942976359957549, "1000–3000 Hz_40ms_modulation_mae_db": 0.45093138001758515, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9958647042245117, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22368208710808704, "3000–6000 Hz_10ms_envelope_correlation": 0.9692049378888316, "3000–6000 Hz_10ms_modulation_mae_db": 0.3616740863992953, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5230714344040546, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15939184573981616, "3000–6000 Hz_40ms_envelope_correlation": 0.9844651847106025, "3000–6000 Hz_40ms_modulation_mae_db": 0.3712262043489007, "mr_stft_db": 6.0030544558045165, "band_mae_db": 0.38175862682460987, "rms_gain_error_db": -0.2195217432134477, "output_peak": 0.5072807048034197}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0107223817040956, "全帯域_10ms_rms_envelope_nrmse": 0.08846424012104039, "全帯域_10ms_envelope_correlation": 0.9918492616636975, "全帯域_10ms_modulation_mae_db": 0.13402178767625697, "全帯域_40ms_log_rms_mae_db": 0.5083342250238431, "全帯域_40ms_rms_envelope_nrmse": 0.0428725758527988, "全帯域_40ms_envelope_correlation": 0.9979419397167109, "全帯域_40ms_modulation_mae_db": 0.1917493713281715, "300–1000 Hz_10ms_log_rms_mae_db": 1.9136722246995863, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12570574390133998, "300–1000 Hz_10ms_envelope_correlation": 0.9870327014818748, "300–1000 Hz_10ms_modulation_mae_db": 0.22455973384349218, "300–1000 Hz_40ms_log_rms_mae_db": 1.1198060442305189, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06365032567385323, "300–1000 Hz_40ms_envelope_correlation": 0.9965704113656986, "300–1000 Hz_40ms_modulation_mae_db": 0.21164457876849965, "1000–3000 Hz_10ms_log_rms_mae_db": 1.970165487610679, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1775501243463005, "1000–3000 Hz_10ms_envelope_correlation": 0.9753153486131085, "1000–3000 Hz_10ms_modulation_mae_db": 0.46679028625961805, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3868566371919115, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10559419898503629, "1000–3000 Hz_40ms_envelope_correlation": 0.9907370207247276, "1000–3000 Hz_40ms_modulation_mae_db": 0.1505363558337184, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9552856930428286, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19566184113819224, "3000–6000 Hz_10ms_envelope_correlation": 0.9743202992237078, "3000–6000 Hz_10ms_modulation_mae_db": 0.4703668416845284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4858355339140779, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1231092877947503, "3000–6000 Hz_40ms_envelope_correlation": 0.9896771126508497, "3000–6000 Hz_40ms_modulation_mae_db": 0.6660947395657191, "mr_stft_db": 5.9893830771632315, "band_mae_db": 0.1811851901279177, "rms_gain_error_db": -0.05961929368993195, "output_peak": 0.5099672681861457}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0269228480389483, "全帯域_10ms_rms_envelope_nrmse": 0.09158096612530288, "全帯域_10ms_envelope_correlation": 0.9913125894878653, "全帯域_10ms_modulation_mae_db": 0.4101053877543441, "全帯域_40ms_log_rms_mae_db": 0.5561260607844437, "全帯域_40ms_rms_envelope_nrmse": 0.05076183743129318, "全帯域_40ms_envelope_correlation": 0.9971176496809465, "全帯域_40ms_modulation_mae_db": 0.4116098067936003, "300–1000 Hz_10ms_log_rms_mae_db": 1.8393320557384387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1254569977662892, "300–1000 Hz_10ms_envelope_correlation": 0.987195272332902, "300–1000 Hz_10ms_modulation_mae_db": 0.36681934076509176, "300–1000 Hz_40ms_log_rms_mae_db": 1.1786826702148407, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07126206711030994, "300–1000 Hz_40ms_envelope_correlation": 0.9957790224548543, "300–1000 Hz_40ms_modulation_mae_db": 0.292409155483848, "1000–3000 Hz_10ms_log_rms_mae_db": 2.029947374932888, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1873334731317312, "1000–3000 Hz_10ms_envelope_correlation": 0.9727564617103194, "1000–3000 Hz_10ms_modulation_mae_db": 0.5784156760128197, "1000–3000 Hz_40ms_log_rms_mae_db": 1.38307380306306, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11111039484741436, "1000–3000 Hz_40ms_envelope_correlation": 0.9907094363628899, "1000–3000 Hz_40ms_modulation_mae_db": 0.3258557136940445, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9620823473106828, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19604763870718053, "3000–6000 Hz_10ms_envelope_correlation": 0.9742505504753212, "3000–6000 Hz_10ms_modulation_mae_db": 0.6012211851537147, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5151758804427133, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14431976445356032, "3000–6000 Hz_40ms_envelope_correlation": 0.9864408038627187, "3000–6000 Hz_40ms_modulation_mae_db": 0.603004279612977, "mr_stft_db": 5.983655286490639, "band_mae_db": 0.27678834214198894, "rms_gain_error_db": -0.07127464298546093, "output_peak": 0.5132912222087235}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0011276728337835, "全帯域_10ms_rms_envelope_nrmse": 0.09678397273693688, "全帯域_10ms_envelope_correlation": 0.9902339117093512, "全帯域_10ms_modulation_mae_db": 0.1308794623364725, "全帯域_40ms_log_rms_mae_db": 0.5101790749663306, "全帯域_40ms_rms_envelope_nrmse": 0.05241331092813044, "全帯域_40ms_envelope_correlation": 0.9968667600562333, "全帯域_40ms_modulation_mae_db": 0.13221383683718022, "300–1000 Hz_10ms_log_rms_mae_db": 1.8616938234556824, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13034560366728734, "300–1000 Hz_10ms_envelope_correlation": 0.9860432493182409, "300–1000 Hz_10ms_modulation_mae_db": 0.285904207807752, "300–1000 Hz_40ms_log_rms_mae_db": 1.0943827422047108, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0783059119798999, "300–1000 Hz_40ms_envelope_correlation": 0.9947139740750981, "300–1000 Hz_40ms_modulation_mae_db": 0.23975088129331892, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9759071409731213, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18198028412668846, "1000–3000 Hz_10ms_envelope_correlation": 0.973979497213537, "1000–3000 Hz_10ms_modulation_mae_db": 0.4482275006005079, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4573078234320611, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10238694454649189, "1000–3000 Hz_40ms_envelope_correlation": 0.9912833429948177, "1000–3000 Hz_40ms_modulation_mae_db": 0.3354781842483086, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0379854336928505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21860371781549184, "3000–6000 Hz_10ms_envelope_correlation": 0.9673257781064303, "3000–6000 Hz_10ms_modulation_mae_db": 0.37490787507758644, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5375225347173906, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14530295414429706, "3000–6000 Hz_40ms_envelope_correlation": 0.9854531857699803, "3000–6000 Hz_40ms_modulation_mae_db": 1.1380199170818663, "mr_stft_db": 5.982614525470093, "band_mae_db": 0.13431978333422057, "rms_gain_error_db": -0.006544082524262825, "output_peak": 0.497034909364557}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.967677678987769, "全帯域_10ms_rms_envelope_nrmse": 0.09122096170616473, "全帯域_10ms_envelope_correlation": 0.991340182536739, "全帯域_10ms_modulation_mae_db": 0.07789062356531971, "全帯域_40ms_log_rms_mae_db": 0.4984864054078231, "全帯域_40ms_rms_envelope_nrmse": 0.04944722997528048, "全帯域_40ms_envelope_correlation": 0.9972072776103691, "全帯域_40ms_modulation_mae_db": 0.12511751594686735, "300–1000 Hz_10ms_log_rms_mae_db": 1.842403869722962, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.130625918302701, "300–1000 Hz_10ms_envelope_correlation": 0.9859691451275038, "300–1000 Hz_10ms_modulation_mae_db": 0.40392427792491997, "300–1000 Hz_40ms_log_rms_mae_db": 1.0781180811951132, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07635178138257799, "300–1000 Hz_40ms_envelope_correlation": 0.9948393896322109, "300–1000 Hz_40ms_modulation_mae_db": 0.21196402271174264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9857761247675507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1873453849517056, "1000–3000 Hz_10ms_envelope_correlation": 0.9723689863786935, "1000–3000 Hz_10ms_modulation_mae_db": 0.2581514441084434, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3728128700759472, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10726404498673248, "1000–3000 Hz_40ms_envelope_correlation": 0.9906701034893356, "1000–3000 Hz_40ms_modulation_mae_db": 0.48130353022055594, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9848385788488687, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2160165521858233, "3000–6000 Hz_10ms_envelope_correlation": 0.9687981800548205, "3000–6000 Hz_10ms_modulation_mae_db": 0.44679314899674755, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5362442662238707, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13405034805368712, "3000–6000 Hz_40ms_envelope_correlation": 0.9878973090482978, "3000–6000 Hz_40ms_modulation_mae_db": 0.6601627075917299, "mr_stft_db": 5.995626724498117, "band_mae_db": 0.22904487964099288, "rms_gain_error_db": -0.009358266500113304, "output_peak": 0.5228771557956977}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0413191404681492, "全帯域_10ms_rms_envelope_nrmse": 0.10176205098039927, "全帯域_10ms_envelope_correlation": 0.9892541464566961, "全帯域_10ms_modulation_mae_db": 0.29327334675321565, "全帯域_40ms_log_rms_mae_db": 0.5455153669717107, "全帯域_40ms_rms_envelope_nrmse": 0.05037294688236484, "全帯域_40ms_envelope_correlation": 0.9972664839748846, "全帯域_40ms_modulation_mae_db": 0.2529598987617648, "300–1000 Hz_10ms_log_rms_mae_db": 1.8412568299389884, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13157837427762475, "300–1000 Hz_10ms_envelope_correlation": 0.9858967887564435, "300–1000 Hz_10ms_modulation_mae_db": 0.33809083837422277, "300–1000 Hz_40ms_log_rms_mae_db": 1.1121727961587677, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06829404671769183, "300–1000 Hz_40ms_envelope_correlation": 0.9962274340084595, "300–1000 Hz_40ms_modulation_mae_db": 0.2855406372806088, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9993009508317712, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17851378256732975, "1000–3000 Hz_10ms_envelope_correlation": 0.9751151836070749, "1000–3000 Hz_10ms_modulation_mae_db": 0.4164177456900864, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3827834340216494, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09813166124956973, "1000–3000 Hz_40ms_envelope_correlation": 0.9919825355903386, "1000–3000 Hz_40ms_modulation_mae_db": 0.46118779202967325, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0183150364217255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22652673580500995, "3000–6000 Hz_10ms_envelope_correlation": 0.968427050285406, "3000–6000 Hz_10ms_modulation_mae_db": 0.6776189626310284, "3000–6000 Hz_40ms_log_rms_mae_db": 1.5571402716252725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15757796757836143, "3000–6000 Hz_40ms_envelope_correlation": 0.9848769042497486, "3000–6000 Hz_40ms_modulation_mae_db": 0.6303695880297936, "mr_stft_db": 5.972386494975112, "band_mae_db": 0.27417854497838545, "rms_gain_error_db": -0.11449749245696386, "output_peak": 0.4966458742036678}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0166442431530716, "全帯域_10ms_rms_envelope_nrmse": 0.09984625308345135, "全帯域_10ms_envelope_correlation": 0.9896784708306532, "全帯域_10ms_modulation_mae_db": 0.206720576048773, "全帯域_40ms_log_rms_mae_db": 0.5557292202829232, "全帯域_40ms_rms_envelope_nrmse": 0.05045133138202069, "全帯域_40ms_envelope_correlation": 0.9972793758412578, "全帯域_40ms_modulation_mae_db": 0.1432345808780914, "300–1000 Hz_10ms_log_rms_mae_db": 1.9456168240557352, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14116604119902368, "300–1000 Hz_10ms_envelope_correlation": 0.9835700306222199, "300–1000 Hz_10ms_modulation_mae_db": 0.29230615270143323, "300–1000 Hz_40ms_log_rms_mae_db": 1.0480833359078277, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07640873751209278, "300–1000 Hz_40ms_envelope_correlation": 0.9949525261096109, "300–1000 Hz_40ms_modulation_mae_db": 0.29982365818944995, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9652885277535466, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16502689022391231, "1000–3000 Hz_10ms_envelope_correlation": 0.9796068994604775, "1000–3000 Hz_10ms_modulation_mae_db": 0.8641356948483946, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3917009171558334, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09463376535421712, "1000–3000 Hz_40ms_envelope_correlation": 0.9942940284832295, "1000–3000 Hz_40ms_modulation_mae_db": 0.5199317022604386, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0246039970426386, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24115886099921024, "3000–6000 Hz_10ms_envelope_correlation": 0.9602120628941777, "3000–6000 Hz_10ms_modulation_mae_db": 1.6620289801714037, "3000–6000 Hz_40ms_log_rms_mae_db": 1.607453079635225, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16357488296442008, "3000–6000 Hz_40ms_envelope_correlation": 0.9821750991903756, "3000–6000 Hz_40ms_modulation_mae_db": 1.3094770275928267, "mr_stft_db": 6.00307520601487, "band_mae_db": 0.3435567467289191, "rms_gain_error_db": -0.12230767841630141, "output_peak": 0.5137796245893281}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0644692067952173, "全帯域_10ms_rms_envelope_nrmse": 0.09285107833271389, "全帯域_10ms_envelope_correlation": 0.9910298306699106, "全帯域_10ms_modulation_mae_db": 0.2570473249485255, "全帯域_40ms_log_rms_mae_db": 0.530060967330371, "全帯域_40ms_rms_envelope_nrmse": 0.04598878036992122, "全帯域_40ms_envelope_correlation": 0.997608350400524, "全帯域_40ms_modulation_mae_db": 0.6780808208462549, "300–1000 Hz_10ms_log_rms_mae_db": 1.9792288729082568, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13452909577502262, "300–1000 Hz_10ms_envelope_correlation": 0.9852192082566662, "300–1000 Hz_10ms_modulation_mae_db": 0.8169307349632507, "300–1000 Hz_40ms_log_rms_mae_db": 1.171923086389329, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07603999101558978, "300–1000 Hz_40ms_envelope_correlation": 0.9951870204647676, "300–1000 Hz_40ms_modulation_mae_db": 0.7227785982160975, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9088915497462566, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1591665179170969, "1000–3000 Hz_10ms_envelope_correlation": 0.9807234496622497, "1000–3000 Hz_10ms_modulation_mae_db": 0.6408899393103168, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3114036168377403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09410224269394217, "1000–3000 Hz_40ms_envelope_correlation": 0.9937066177493653, "1000–3000 Hz_40ms_modulation_mae_db": 0.5726134438528857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9752140359144426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23425546064224642, "3000–6000 Hz_10ms_envelope_correlation": 0.9629307079340028, "3000–6000 Hz_10ms_modulation_mae_db": 0.6738997347134793, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4914511077919217, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13528726529941051, "3000–6000 Hz_40ms_envelope_correlation": 0.9875380766433758, "3000–6000 Hz_40ms_modulation_mae_db": 0.6552831165286971, "mr_stft_db": 5.974506379477513, "band_mae_db": 0.21543191846587306, "rms_gain_error_db": -0.03737578825004021, "output_peak": 0.500581506109294}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68080, "output_samples": 68080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.128932486785725, "spectral_convergence": 0.30152835364942765, "spectral_convergence_status": "ok", "amplitude_floor": 2.0850976744763412e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.822921506295696, "spectral_convergence": 0.2810241625927835, "spectral_convergence_status": "ok", "amplitude_floor": 1.4381271492089952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.057309374332126, "spectral_convergence": 0.2990178155818947, "spectral_convergence_status": "ok", "amplitude_floor": 9.411979934974602e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005432476915893821, "output_power": 0.0005884592613538393, "delta_db": 0.347185127414951, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030286578008489277, "output_power": 0.0027657140592072696, "delta_db": -0.3944292905640712, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038180407470981095, "output_power": 0.00040679181966029877, "delta_db": 0.27531651953646163, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002532984259990213, "output_power": 0.00022535785957525183, "delta_db": -0.507597819702476, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013717204322211298, "output_power": 0.00012931567679967958, "delta_db": -0.2561443051981815, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.29065897621083e-05, "output_power": 9.323455907867301e-05, "delta_db": 0.5098786985315178, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.38175862682460987, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2195217432134477, "pitch": null, "output_rms_dbfs": -23.75834154289015, "output_peak": 0.5072807048034197, "reference_rms_dbfs": -23.538819799676702, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0030544558045165}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00026228408676881105, "log_rms_mae_db": 1.0062212979956104, "rms_envelope_nrmse": 0.09567098600521934, "envelope_correlation": 0.9908867361051367, "rms_gain_error_db": -0.21952192644068858, "reference_mean_power": 0.004432288013980724, "output_mean_power": 0.0042138180666465505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 1.6030570640082068e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008005952323264084, "output_power": 0.0007549894351412702, "reference_active": true, "delta_db": -0.25472125061872525}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020547999455019126, "output_power": 0.00020506888818129743, "reference_active": true, "delta_db": -0.008697687153524966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.989786705158196e-05, "output_power": 3.534419647006749e-05, "reference_active": true, "delta_db": -0.526315661575456}], "active_band_count": 3, "active_band_mae_db": 0.2632448664492354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00021831174911177267, "log_rms_mae_db": 0.5611773209706499, "rms_envelope_nrmse": 0.05714512491252644, "envelope_correlation": 0.9969256910749341, "rms_gain_error_db": -0.2195097154238719, "reference_mean_power": 0.004463740704660867, "output_mean_power": 0.0042437323687126455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 1.4245155353761532e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007405152195784611, "output_power": 0.0006992822993201975, "reference_active": true, "delta_db": -0.24881453627753147}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001021109367695941, "output_power": 0.00010458406738157176, "reference_active": true, "delta_db": 0.10393267535116138}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.434051823082556e-06, "output_power": 2.056899396677631e-06, "reference_active": true, "delta_db": -0.7311676970779661}], "active_band_count": 3, "active_band_mae_db": 0.3613049695688863}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00023386465365005232, "log_rms_mae_db": 1.9707068448757812, "rms_envelope_nrmse": 0.1406145394677927, "envelope_correlation": 0.9843827855378765, "rms_gain_error_db": -0.3408321819538086, "reference_mean_power": 0.002112021838569365, "output_mean_power": 0.0019526085486477253}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 9.681983603336737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048709120906038954, "output_power": 0.00045022890030037646, "reference_active": true, "delta_db": -0.3417692254374155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014674692890756196, "output_power": 0.0001298669154084393, "reference_active": true, "delta_db": -0.5307049586302136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8361558797573682e-05, "output_power": 2.3903181835337567e-05, "reference_active": true, "delta_db": -0.7427438146166064}], "active_band_count": 3, "active_band_mae_db": 0.5384059995614119}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00020000043006212289, "log_rms_mae_db": 1.1858005448742175, "rms_envelope_nrmse": 0.08664452034010117, "envelope_correlation": 0.9944580094695556, "rms_gain_error_db": -0.34083855025935683, "reference_mean_power": 0.002127032769351675, "output_mean_power": 0.0019664835858630445}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.787606965551091e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004705886462934124, "output_power": 0.0004345255895240203, "reference_active": true, "delta_db": -0.3462608746306295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.123456963821095e-05, "output_power": 6.244834102100376e-05, "reference_active": true, "delta_db": -0.571698993063373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5465229502268847e-06, "output_power": 1.5379396194940611e-06, "reference_active": true, "delta_db": -0.02417084150991003}], "active_band_count": 3, "active_band_mae_db": 0.3140435697346375}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 7.720600674415526e-05, "log_rms_mae_db": 1.8924435564326474, "rms_envelope_nrmse": 0.15949011673240163, "envelope_correlation": 0.9814016571443287, "rms_gain_error_db": -0.44948923747675334, "reference_mean_power": 0.00033128809071676906, "output_mean_power": 0.00029871490984609435}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 2.2485323271997473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014154142624946028, "output_power": 0.00012183078329264203, "reference_active": true, "delta_db": -0.6512653103243777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6917189688068767e-05, "output_power": 2.478270463982348e-05, "reference_active": true, "delta_db": -0.3588101404352466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.319943953798946e-06, "output_power": 5.8625399877309315e-06, "reference_active": true, "delta_db": -0.3262740890410287}], "active_band_count": 3, "active_band_mae_db": 0.4454498466002177}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 6.30102051319373e-05, "log_rms_mae_db": 1.2810407187588364, "rms_envelope_nrmse": 0.09706150722875745, "envelope_correlation": 0.9942976359957549, "rms_gain_error_db": -0.44949638938202685, "reference_mean_power": 0.0003336427865109939, "output_mean_power": 0.0003008375898826539}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 2.017085687613918e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013578457159768264, "output_power": 0.00011710326238598376, "reference_active": true, "delta_db": -0.6428143214600402}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.310083041117988e-05, "output_power": 1.2533416920685453e-05, "reference_active": true, "delta_db": -0.1922933818603721}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7328752869332565e-07, "output_power": 3.3134073934526427e-07, "reference_active": true, "delta_db": -0.5176864367323429}], "active_band_count": 3, "active_band_mae_db": 0.45093138001758515}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 5.974138539374114e-05, "log_rms_mae_db": 1.9958647042245117, "rms_envelope_nrmse": 0.22368208710808704, "envelope_correlation": 0.9692049378888316, "rms_gain_error_db": 0.41944727940443016, "reference_mean_power": 8.138813397574021e-05, "output_mean_power": 8.964081971504434e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 8.567729341398199e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0300307346839795e-05, "output_power": 4.305521937695917e-05, "reference_active": true, "delta_db": 0.28717448146812224}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.809515102528064e-06, "output_power": 5.154772804358061e-06, "reference_active": true, "delta_db": -0.5193035644984946}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.571956353556118e-06, "output_power": 3.3500533416062765e-06, "reference_active": true, "delta_db": -0.27854421323126927}], "active_band_count": 3, "active_band_mae_db": 0.3616740863992953}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 4.362120322552387e-05, "log_rms_mae_db": 1.5230714344040546, "rms_envelope_nrmse": 0.15939184573981616, "envelope_correlation": 0.9844651847106025, "rms_gain_error_db": 0.41944286357750826, "reference_mean_power": 8.196654666810843e-05, "output_mean_power": 9.027779115665225e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.099518469769939e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.90961297245674e-05, "output_power": 4.1341855831744316e-05, "reference_active": true, "delta_db": 0.24256201112720027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1532747685680723e-06, "output_power": 2.7703256018326884e-06, "reference_active": true, "delta_db": -0.5623100019286937}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4027458558124955e-07, "output_power": 1.5061202729259596e-07, "reference_active": true, "delta_db": 0.3088065999908081}], "active_band_count": 3, "active_band_mae_db": 0.3712262043489007}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7923082941383462, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d1ed2662f7466b68e80ec3dbd80e2404db2de4caf6da658e9f29f15d3b13a775", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9764084354737237, "standard_deviation": 0.0036293038230633015, "minimum": 0.9723689863786935, "maximum": 0.9814016571443287, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9659650891310574, "standard_deviation": 0.045319878720197444, "minimum": 1.8924435564326474, "maximum": 2.029947374932888, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5148097666788005, "standard_deviation": 0.180900476676533, "minimum": 0.2581514441084434, "maximum": 0.8641356948483946, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1745508217496458, "standard_deviation": 0.011718128332652579, "minimum": 0.1591665179170969, "maximum": 0.1873453849517056, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9922100901738073, "standard_deviation": 0.001632016154210132, "minimum": 0.9906701034893356, "maximum": 0.9942976359957549, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.37087247756713, "standard_deviation": 0.05353394771684782, "minimum": 1.2810407187588364, "maximum": 1.4573078234320611, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4122297627696513, "standard_deviation": 0.1351312813935012, "minimum": 0.1505363558337184, "maximum": 0.5726134438528857, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10128559498652018, "standard_deviation": 0.0062756126507954755, "minimum": 0.09410224269394217, "maximum": 0.11111039484741436, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9681836958578373, "standard_deviation": 0.004896371986550877, "minimum": 0.9602120628941777, "maximum": 0.9743202992237078, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9942737283123186, "standard_deviation": 0.030283894619227245, "minimum": 1.9552856930428286, "maximum": 2.0379854336928505, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6585638518534729, "standard_deviation": 0.42421320834513015, "minimum": 0.3616740863992953, "maximum": 1.6620289801714037, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2189941118001552, "standard_deviation": 0.01640059303280208, "minimum": 0.19566184113819224, "maximum": 0.24115886099921024, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9860654595157436, "standard_deviation": 0.002335824153572367, "minimum": 0.9821750991903756, "maximum": 0.9896771126508497, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.5317367635943158, "standard_deviation": 0.038750666709201534, "minimum": 1.4858355339140779, "maximum": 1.607453079635225, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7542046975440638, "standard_deviation": 0.30866149975275464, "minimum": 0.3712262043489007, "maximum": 1.3094770275928267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1453267895035379, "standard_deviation": 0.014159763486122934, "minimum": 0.1231092877947503, "maximum": 0.16357488296442008, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.985663647679216, "standard_deviation": 0.0012382781742797525, "minimum": 0.9835700306222199, "maximum": 0.987195272332902, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.899238918174429, "standard_deviation": 0.060303872760594075, "minimum": 1.8393320557384387, "maximum": 1.9792288729082568, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4083676607426968, "standard_deviation": 0.190026884725693, "minimum": 0.22455973384349218, "maximum": 0.8169307349632507, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13250278929463516, "standard_deviation": 0.005974134538603824, "minimum": 0.1254569977662892, "maximum": 0.14116604119902368, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.995340973447532, "standard_deviation": 0.0007642129127400217, "minimum": 0.9944580094695556, "maximum": 0.9965704113656986, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.123621162646916, "standard_deviation": 0.05072069198892325, "minimum": 1.0480833359078277, "maximum": 1.1858005448742175, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.32224438770977537, "standard_deviation": 0.16666225810387442, "minimum": 0.21164457876849965, "maximum": 0.7227785982160975, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07461967271651458, "standard_deviation": 0.006948969212851019, "minimum": 0.06365032567385323, "maximum": 0.08664452034010117, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.25453300403036344, "standard_deviation": 0.08194291813265299, "minimum": 0.13431978333422057, "maximum": 0.38175862682460987, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.988037768736762, "standard_deviation": 0.011878588772051274, "minimum": 5.972386494975112, "maximum": 6.00307520601487, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5076822831576042, "standard_deviation": 0.009184569917876115, "minimum": 0.4966458742036678, "maximum": 0.5228771557956977, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.08006237350456527, "standard_deviation": 0.07085262082115772, "minimum": -0.2195217432134477, "maximum": -0.006544082524262825, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9906981411825062, "standard_deviation": 0.0008944899218452904, "minimum": 0.9892541464566961, "maximum": 0.9918492616636975, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0168880587470805, "standard_deviation": 0.02874381037554126, "minimum": 0.967677678987769, "maximum": 1.0644692067952173, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.22164792194151783, "standard_deviation": 0.10719546937577605, "minimum": 0.07789062356531971, "maximum": 0.4101053877543441, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09477256363640359, "standard_deviation": 0.004562821958111587, "minimum": 0.08846424012104039, "maximum": 0.10176205098039927, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972766910444826, "standard_deviation": 0.00035335572741030573, "minimum": 0.9968667600562333, "maximum": 0.9979419397167109, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.533201080217262, "standard_deviation": 0.02487318034360333, "minimum": 0.4984864054078231, "maximum": 0.5611773209706499, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.2870338501201021, "standard_deviation": 0.1908842705331362, "minimum": 0.12511751594686735, "maximum": 0.6780808208462549, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.04993164221679201, "standard_deviation": 0.004225066282142551, "minimum": 0.0428725758527988, "maximum": 0.05714512491252644, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-157963-0018", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.197321891784668, "predicted_mos": 2.197321891784668, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "c33c084ac75335fada80335601f16f7ad7c428571e2f1b910b1d0e3112c10a92", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.028890499845147, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0078324450111493, "全帯域_10ms_rms_envelope_nrmse": 0.09551069700421268, "全帯域_10ms_envelope_correlation": 0.9906418387505089, "全帯域_10ms_modulation_mae_db": 0.209399571276069, "全帯域_40ms_log_rms_mae_db": 0.5443723289904706, "全帯域_40ms_rms_envelope_nrmse": 0.05468639346546428, "全帯域_40ms_envelope_correlation": 0.9968837452231353, "全帯域_40ms_modulation_mae_db": 0.25376210326624066, "300–1000 Hz_10ms_log_rms_mae_db": 1.661859277410794, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12977197826906792, "300–1000 Hz_10ms_envelope_correlation": 0.9864392216341513, "300–1000 Hz_10ms_modulation_mae_db": 0.3203751711992339, "300–1000 Hz_40ms_log_rms_mae_db": 1.0392344532846092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07825735459334855, "300–1000 Hz_40ms_envelope_correlation": 0.9951223572052288, "300–1000 Hz_40ms_modulation_mae_db": 0.2989027837671685, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5593535137602628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1528548066560081, "1000–3000 Hz_10ms_envelope_correlation": 0.9820027808610622, "1000–3000 Hz_10ms_modulation_mae_db": 0.2803433729461205, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2325749615083097, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09218518311915892, "1000–3000 Hz_40ms_envelope_correlation": 0.9935568607219326, "1000–3000 Hz_40ms_modulation_mae_db": 0.33696027908178666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.357989281358904, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18787886747857777, "3000–6000 Hz_10ms_envelope_correlation": 0.9771261327927158, "3000–6000 Hz_10ms_modulation_mae_db": 0.9215779647528789, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1103110673128245, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1269387619819471, "3000–6000 Hz_40ms_envelope_correlation": 0.9891917189940709, "3000–6000 Hz_40ms_modulation_mae_db": 0.9948845427696673, "mr_stft_db": 5.62094096643979, "band_mae_db": 0.30495865048206916, "rms_gain_error_db": -0.15016440334867015, "output_peak": 0.5111893920939072}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0117507631345863, "全帯域_10ms_rms_envelope_nrmse": 0.08695797167430619, "全帯域_10ms_envelope_correlation": 0.9923810834749984, "全帯域_10ms_modulation_mae_db": 0.31169197804257903, "全帯域_40ms_log_rms_mae_db": 0.5152587990587733, "全帯域_40ms_rms_envelope_nrmse": 0.049501467741020766, "全帯域_40ms_envelope_correlation": 0.9976092094340889, "全帯域_40ms_modulation_mae_db": 0.4601156556658976, "300–1000 Hz_10ms_log_rms_mae_db": 1.5882179752239705, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12102151300503577, "300–1000 Hz_10ms_envelope_correlation": 0.988392905086158, "300–1000 Hz_10ms_modulation_mae_db": 1.2180963547727737, "300–1000 Hz_40ms_log_rms_mae_db": 0.8855182602975794, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0700706597118608, "300–1000 Hz_40ms_envelope_correlation": 0.9963581665488324, "300–1000 Hz_40ms_modulation_mae_db": 0.9095304404390893, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5206139621343668, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16554282872249892, "1000–3000 Hz_10ms_envelope_correlation": 0.9789342699577053, "1000–3000 Hz_10ms_modulation_mae_db": 0.5852539904511395, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0909071307031153, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10248789070296557, "1000–3000 Hz_40ms_envelope_correlation": 0.992130863619381, "1000–3000 Hz_40ms_modulation_mae_db": 0.38233867732595844, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2750854272705496, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17176460968317223, "3000–6000 Hz_10ms_envelope_correlation": 0.979941585385914, "3000–6000 Hz_10ms_modulation_mae_db": 0.10693251103763544, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0866333387926657, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11744534180729392, "3000–6000 Hz_40ms_envelope_correlation": 0.9901944098490195, "3000–6000 Hz_40ms_modulation_mae_db": 0.14764173239589765, "mr_stft_db": 5.606682181909982, "band_mae_db": 0.21554294899296558, "rms_gain_error_db": -0.1726547279527754, "output_peak": 0.47491502603645214}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.020353674164486, "全帯域_10ms_rms_envelope_nrmse": 0.09957583360356709, "全帯域_10ms_envelope_correlation": 0.9899487103754515, "全帯域_10ms_modulation_mae_db": 0.19792290888496913, "全帯域_40ms_log_rms_mae_db": 0.5768464439849943, "全帯域_40ms_rms_envelope_nrmse": 0.05917609631648914, "全帯域_40ms_envelope_correlation": 0.9964919539430203, "全帯域_40ms_modulation_mae_db": 0.2803993613013347, "300–1000 Hz_10ms_log_rms_mae_db": 1.578166819828947, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14163647402465968, "300–1000 Hz_10ms_envelope_correlation": 0.9835577426071097, "300–1000 Hz_10ms_modulation_mae_db": 0.4597448919644111, "300–1000 Hz_40ms_log_rms_mae_db": 0.8291127110608327, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07523907655622777, "300–1000 Hz_40ms_envelope_correlation": 0.995327001649078, "300–1000 Hz_40ms_modulation_mae_db": 0.3158826611972287, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4814145342896654, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14501863250068037, "1000–3000 Hz_10ms_envelope_correlation": 0.9843559608281012, "1000–3000 Hz_10ms_modulation_mae_db": 0.4046147090059584, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1215022365960614, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08627853674668388, "1000–3000 Hz_40ms_envelope_correlation": 0.9951532705842593, "1000–3000 Hz_40ms_modulation_mae_db": 0.5250284939177888, "3000–6000 Hz_10ms_log_rms_mae_db": 1.294360806789919, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14768426643002555, "3000–6000 Hz_10ms_envelope_correlation": 0.9856628037969489, "3000–6000 Hz_10ms_modulation_mae_db": 0.25169069113834913, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0447885644946981, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09975453827575083, "3000–6000 Hz_40ms_envelope_correlation": 0.9932656057658401, "3000–6000 Hz_40ms_modulation_mae_db": 0.4716078175091871, "mr_stft_db": 5.608618121260605, "band_mae_db": 0.2550253860050928, "rms_gain_error_db": -0.19209361191812452, "output_peak": 0.48413401909323683}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0113828286677997, "全帯域_10ms_rms_envelope_nrmse": 0.0905001711542962, "全帯域_10ms_envelope_correlation": 0.9915184534203365, "全帯域_10ms_modulation_mae_db": 0.20563446513446484, "全帯域_40ms_log_rms_mae_db": 0.5491208907692784, "全帯域_40ms_rms_envelope_nrmse": 0.051610251250060274, "全帯域_40ms_envelope_correlation": 0.9970973815284796, "全帯域_40ms_modulation_mae_db": 0.19828522185700578, "300–1000 Hz_10ms_log_rms_mae_db": 1.6521956657504004, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12291520088044548, "300–1000 Hz_10ms_envelope_correlation": 0.9876041796568702, "300–1000 Hz_10ms_modulation_mae_db": 0.2678390341539482, "300–1000 Hz_40ms_log_rms_mae_db": 0.9278618755705843, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0721237093245521, "300–1000 Hz_40ms_envelope_correlation": 0.9955261476609081, "300–1000 Hz_40ms_modulation_mae_db": 0.3176365587253526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5229965600328834, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14735955386701763, "1000–3000 Hz_10ms_envelope_correlation": 0.9835105046274607, "1000–3000 Hz_10ms_modulation_mae_db": 0.37008456372691406, "1000–3000 Hz_40ms_log_rms_mae_db": 1.171641427263285, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09192569216941214, "1000–3000 Hz_40ms_envelope_correlation": 0.9938368624549581, "1000–3000 Hz_40ms_modulation_mae_db": 0.35991454715976606, "3000–6000 Hz_10ms_log_rms_mae_db": 1.425242212208575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19016603637241, "3000–6000 Hz_10ms_envelope_correlation": 0.9768274376991941, "3000–6000 Hz_10ms_modulation_mae_db": 0.9989161024449907, "3000–6000 Hz_40ms_log_rms_mae_db": 1.155585773969185, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12463717920768666, "3000–6000 Hz_40ms_envelope_correlation": 0.9898322960729076, "3000–6000 Hz_40ms_modulation_mae_db": 1.229944562160534, "mr_stft_db": 5.616018494916712, "band_mae_db": 0.23009011532424836, "rms_gain_error_db": -0.1052810270406701, "output_peak": 0.4812897761542003}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0179626526434729, "全帯域_10ms_rms_envelope_nrmse": 0.088603770238819, "全帯域_10ms_envelope_correlation": 0.9918987802031893, "全帯域_10ms_modulation_mae_db": 0.15173801369539255, "全帯域_40ms_log_rms_mae_db": 0.5492471073273971, "全帯域_40ms_rms_envelope_nrmse": 0.049215867706257806, "全帯域_40ms_envelope_correlation": 0.9974189354725373, "全帯域_40ms_modulation_mae_db": 0.18846610716804188, "300–1000 Hz_10ms_log_rms_mae_db": 1.6582266189782084, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12109585137163657, "300–1000 Hz_10ms_envelope_correlation": 0.9880459903259688, "300–1000 Hz_10ms_modulation_mae_db": 0.5154470679691241, "300–1000 Hz_40ms_log_rms_mae_db": 0.9154023532931761, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07294627709230217, "300–1000 Hz_40ms_envelope_correlation": 0.9955368390188277, "300–1000 Hz_40ms_modulation_mae_db": 0.32372244086229585, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5450024082882197, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16400148643418522, "1000–3000 Hz_10ms_envelope_correlation": 0.9806245395047377, "1000–3000 Hz_10ms_modulation_mae_db": 0.9675128100330109, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1548995894577403, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09551737613884377, "1000–3000 Hz_40ms_envelope_correlation": 0.9951106525206561, "1000–3000 Hz_40ms_modulation_mae_db": 0.5807512634604254, "3000–6000 Hz_10ms_log_rms_mae_db": 1.347158048627695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19222930982355313, "3000–6000 Hz_10ms_envelope_correlation": 0.9755592893312459, "3000–6000 Hz_10ms_modulation_mae_db": 0.5250390900283597, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0828280295427801, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13158233329526614, "3000–6000 Hz_40ms_envelope_correlation": 0.9880745735121456, "3000–6000 Hz_40ms_modulation_mae_db": 0.8233728292042946, "mr_stft_db": 5.611734778554006, "band_mae_db": 0.28484581407413095, "rms_gain_error_db": -0.11572707073176437, "output_peak": 0.48643508914373074}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0559593269949863, "全帯域_10ms_rms_envelope_nrmse": 0.0971771118451495, "全帯域_10ms_envelope_correlation": 0.9902995181305894, "全帯域_10ms_modulation_mae_db": 0.3704196263448423, "全帯域_40ms_log_rms_mae_db": 0.5462191942458977, "全帯域_40ms_rms_envelope_nrmse": 0.05225900642578616, "全帯域_40ms_envelope_correlation": 0.9971355782565365, "全帯域_40ms_modulation_mae_db": 0.38550117555356117, "300–1000 Hz_10ms_log_rms_mae_db": 1.6239711103577252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14320158974924768, "300–1000 Hz_10ms_envelope_correlation": 0.9834140399625749, "300–1000 Hz_10ms_modulation_mae_db": 0.9934661967174793, "300–1000 Hz_40ms_log_rms_mae_db": 1.0197179462073107, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0905397256891189, "300–1000 Hz_40ms_envelope_correlation": 0.993285194350026, "300–1000 Hz_40ms_modulation_mae_db": 0.9047503360172476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5008223600358461, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15027346979674389, "1000–3000 Hz_10ms_envelope_correlation": 0.9831177966914201, "1000–3000 Hz_10ms_modulation_mae_db": 0.3839114771349951, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1097140041623292, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08812231927435697, "1000–3000 Hz_40ms_envelope_correlation": 0.9949061979664132, "1000–3000 Hz_40ms_modulation_mae_db": 0.4181874369080954, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3246332155661082, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17926683589279677, "3000–6000 Hz_10ms_envelope_correlation": 0.9782092150589258, "3000–6000 Hz_10ms_modulation_mae_db": 0.16630057363714215, "3000–6000 Hz_40ms_log_rms_mae_db": 1.046074414068643, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10774052279609361, "3000–6000 Hz_40ms_envelope_correlation": 0.9917739558875581, "3000–6000 Hz_40ms_modulation_mae_db": 0.34934760119233804, "mr_stft_db": 5.627400787528688, "band_mae_db": 0.20050228751877328, "rms_gain_error_db": -0.14087082706273252, "output_peak": 0.4900851595318592}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0401294998078465, "全帯域_10ms_rms_envelope_nrmse": 0.08977034414188387, "全帯域_10ms_envelope_correlation": 0.9916205771525496, "全帯域_10ms_modulation_mae_db": 0.24298852620555902, "全帯域_40ms_log_rms_mae_db": 0.5303896027517415, "全帯域_40ms_rms_envelope_nrmse": 0.04802175739686958, "全帯域_40ms_envelope_correlation": 0.9974243881633112, "全帯域_40ms_modulation_mae_db": 0.0707003498906198, "300–1000 Hz_10ms_log_rms_mae_db": 1.6954452549684336, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12417784497797091, "300–1000 Hz_10ms_envelope_correlation": 0.9874944046770485, "300–1000 Hz_10ms_modulation_mae_db": 0.26038101275895986, "300–1000 Hz_40ms_log_rms_mae_db": 1.0425864343918214, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06610077176581837, "300–1000 Hz_40ms_envelope_correlation": 0.9964847652042579, "300–1000 Hz_40ms_modulation_mae_db": 0.2048609025988877, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5441181843584855, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14586288555872332, "1000–3000 Hz_10ms_envelope_correlation": 0.983643216166474, "1000–3000 Hz_10ms_modulation_mae_db": 0.5954989316522569, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1091955590711597, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08511657106077113, "1000–3000 Hz_40ms_envelope_correlation": 0.9945499657918636, "1000–3000 Hz_40ms_modulation_mae_db": 0.39217155480604515, "3000–6000 Hz_10ms_log_rms_mae_db": 1.340567164051101, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20181974442028155, "3000–6000 Hz_10ms_envelope_correlation": 0.9721373013960848, "3000–6000 Hz_10ms_modulation_mae_db": 0.6492110557516079, "3000–6000 Hz_40ms_log_rms_mae_db": 1.039933808875451, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12683273321505828, "3000–6000 Hz_40ms_envelope_correlation": 0.9885500840939189, "3000–6000 Hz_40ms_modulation_mae_db": 0.9108818452807945, "mr_stft_db": 5.617479776539814, "band_mae_db": 0.21949889636920095, "rms_gain_error_db": -0.07151406839392677, "output_peak": 0.51383426788701}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0573391840892339, "全帯域_10ms_rms_envelope_nrmse": 0.08695441342804905, "全帯域_10ms_envelope_correlation": 0.9921572626605811, "全帯域_10ms_modulation_mae_db": 0.30338892728882905, "全帯域_40ms_log_rms_mae_db": 0.5092938138699507, "全帯域_40ms_rms_envelope_nrmse": 0.0437967188755647, "全帯域_40ms_envelope_correlation": 0.9979183520223005, "全帯域_40ms_modulation_mae_db": 0.19657419604610218, "300–1000 Hz_10ms_log_rms_mae_db": 1.7320419410562773, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1253123429674688, "300–1000 Hz_10ms_envelope_correlation": 0.9872060128983009, "300–1000 Hz_10ms_modulation_mae_db": 0.25698360257676733, "300–1000 Hz_40ms_log_rms_mae_db": 0.9331040090468578, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06681478926250671, "300–1000 Hz_40ms_envelope_correlation": 0.9963887475898906, "300–1000 Hz_40ms_modulation_mae_db": 0.3796896756552372, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5284399570002833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1699115079716935, "1000–3000 Hz_10ms_envelope_correlation": 0.9787325120295899, "1000–3000 Hz_10ms_modulation_mae_db": 0.5912953608027903, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1029154147096216, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11198397790182953, "1000–3000 Hz_40ms_envelope_correlation": 0.9917478404614566, "1000–3000 Hz_40ms_modulation_mae_db": 0.38190880242548614, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3492379386313176, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1861679792774145, "3000–6000 Hz_10ms_envelope_correlation": 0.9763634380465535, "3000–6000 Hz_10ms_modulation_mae_db": 0.265831505953394, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1125687945239229, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1251224044880121, "3000–6000 Hz_40ms_envelope_correlation": 0.9888704197449824, "3000–6000 Hz_40ms_modulation_mae_db": 0.2768516435092832, "mr_stft_db": 5.614583832903378, "band_mae_db": 0.26591881012680685, "rms_gain_error_db": -0.09047331695710387, "output_peak": 0.5011292365553129}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68080, "output_samples": 68080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.6489260398483765, "spectral_convergence": 0.2939167060023488, "spectral_convergence_status": "ok", "amplitude_floor": 2.0850976744763412e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.360178677469007, "spectral_convergence": 0.26820900174279716, "spectral_convergence_status": "ok", "amplitude_floor": 1.4381271492089952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8537181820019875, "spectral_convergence": 0.29684370081159117, "spectral_convergence_status": "ok", "amplitude_floor": 9.411979934974602e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005432476915893821, "output_power": 0.0006087618616670749, "delta_db": 0.49449546933506683, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030286578008489277, "output_power": 0.002789971388677787, "delta_db": -0.35650456865847413, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038180407470981095, "output_power": 0.0004169717798876536, "delta_db": 0.3826610449474077, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002532984259990213, "output_power": 0.00023522435927090618, "delta_db": -0.3215019685650412, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013717204322211298, "output_power": 0.00013775679616537424, "delta_db": 0.01847426119790073, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.29065897621083e-05, "output_power": 8.794284485857338e-05, "delta_db": 0.25611459018852434, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.30495865048206916, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15016440334867015, "pitch": null, "output_rms_dbfs": -23.688984203025374, "output_peak": 0.5111893920939072, "reference_rms_dbfs": -23.538819799676702, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.62094096643979}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00026228408676881105, "log_rms_mae_db": 1.0078324450111493, "rms_envelope_nrmse": 0.09551069700421268, "envelope_correlation": 0.9906418387505089, "rms_gain_error_db": -0.1501659411385029, "reference_mean_power": 0.004432288013980724, "output_mean_power": 0.0042816521295324286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 1.6030570640082068e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008005952323264084, "output_power": 0.0007575167381045552, "reference_active": true, "delta_db": -0.24020766023510248}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020547999455019126, "output_power": 0.0001974960425582624, "reference_active": true, "delta_db": -0.172111479122226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.989786705158196e-05, "output_power": 3.7963104585543034e-05, "reference_active": true, "delta_db": -0.21587957447087858}], "active_band_count": 3, "active_band_mae_db": 0.209399571276069}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00021831174911177267, "log_rms_mae_db": 0.5443723289904706, "rms_envelope_nrmse": 0.05468639346546428, "envelope_correlation": 0.9968837452231353, "rms_gain_error_db": -0.15015274773585813, "reference_mean_power": 0.004463740704660867, "output_mean_power": 0.0043120489675696665}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 1.4245155353761532e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007405152195784611, "output_power": 0.0007021365940535131, "reference_active": true, "delta_db": -0.23112380622786177}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001021109367695941, "output_power": 0.00010127659051858161, "reference_active": true, "delta_db": -0.03563188070282538}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.434051823082556e-06, "output_power": 2.1720847087101528e-06, "reference_active": true, "delta_db": -0.49453062286803473}], "active_band_count": 3, "active_band_mae_db": 0.25376210326624066}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00023386465365005232, "log_rms_mae_db": 1.661859277410794, "rms_envelope_nrmse": 0.12977197826906792, "envelope_correlation": 0.9864392216341513, "rms_gain_error_db": -0.24521751899208552, "reference_mean_power": 0.002112021838569365, "output_mean_power": 0.001996074068454773}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 9.681983603336737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048709120906038954, "output_power": 0.0004489120146696416, "reference_active": true, "delta_db": -0.3544906256570463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014674692890756196, "output_power": 0.00013173043366283157, "reference_active": true, "delta_db": -0.4688289963617948}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8361558797573682e-05, "output_power": 2.9275929041958937e-05, "reference_active": true, "delta_db": 0.13780589157886067}], "active_band_count": 3, "active_band_mae_db": 0.3203751711992339}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00020000043006212289, "log_rms_mae_db": 1.0392344532846092, "rms_envelope_nrmse": 0.07825735459334855, "envelope_correlation": 0.9951223572052288, "rms_gain_error_db": -0.24522174770312255, "reference_mean_power": 0.002127032769351675, "output_mean_power": 0.0020102589575947483}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.787606965551091e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004705886462934124, "output_power": 0.00043654799181651197, "reference_active": true, "delta_db": -0.3260945017129902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.123456963821095e-05, "output_power": 6.597557202328268e-05, "reference_active": true, "delta_db": -0.3330764052608255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5465229502268847e-06, "output_power": 1.6334660544511528e-06, "reference_active": true, "delta_db": 0.23753744432768986}], "active_band_count": 3, "active_band_mae_db": 0.2989027837671685}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 7.720600674415526e-05, "log_rms_mae_db": 1.5593535137602628, "rms_envelope_nrmse": 0.1528548066560081, "envelope_correlation": 0.9820027808610622, "rms_gain_error_db": -0.28200196762379526, "reference_mean_power": 0.00033128809071676906, "output_mean_power": 0.00031045998144992764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 2.2485323271997473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014154142624946028, "output_power": 0.00013338848340323063, "reference_active": true, "delta_db": -0.25765232731548454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6917189688068767e-05, "output_power": 2.5185040867778942e-05, "reference_active": true, "delta_db": -0.2888705511231081}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.319943953798946e-06, "output_power": 6.763382523454517e-06, "reference_active": true, "delta_db": 0.2945072403997689}], "active_band_count": 3, "active_band_mae_db": 0.2803433729461205}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 6.30102051319373e-05, "log_rms_mae_db": 1.2325749615083097, "rms_envelope_nrmse": 0.09218518311915892, "envelope_correlation": 0.9935568607219326, "rms_gain_error_db": -0.2820035179533734, "reference_mean_power": 0.0003336427865109939, "output_mean_power": 0.0003126665257129908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 2.017085687613918e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013578457159768264, "output_power": 0.00012856705419185985, "reference_active": true, "delta_db": -0.23720733226555618}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.310083041117988e-05, "output_power": 1.1995809385599505e-05, "reference_active": true, "delta_db": -0.3826926861552976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7328752869332565e-07, "output_power": 3.4115001819190253e-07, "reference_active": true, "delta_db": -0.3909808188245061}], "active_band_count": 3, "active_band_mae_db": 0.33696027908178666}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 5.974138539374114e-05, "log_rms_mae_db": 1.357989281358904, "rms_envelope_nrmse": 0.18787886747857777, "envelope_correlation": 0.9771261327927158, "rms_gain_error_db": 0.1914889188062493, "reference_mean_power": 8.138813397574021e-05, "output_mean_power": 8.505698478657734e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 8.567729341398199e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0300307346839795e-05, "output_power": 4.307961953195252e-05, "reference_active": true, "delta_db": 0.28963500803063374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.809515102528064e-06, "output_power": 5.6289532163978805e-06, "reference_active": true, "delta_db": -0.13712245895671615}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.571956353556118e-06, "output_power": 6.1193287602483485e-06, "reference_active": true, "delta_db": 2.3379764272712866}], "active_band_count": 3, "active_band_mae_db": 0.9215779647528789}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 4.362120322552387e-05, "log_rms_mae_db": 1.1103110673128245, "rms_envelope_nrmse": 0.1269387619819471, "envelope_correlation": 0.9891917189940709, "rms_gain_error_db": 0.19148820121811214, "reference_mean_power": 8.196654666810843e-05, "output_mean_power": 8.566145727183492e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.099518469769939e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.90961297245674e-05, "output_power": 4.306850517981416e-05, "reference_active": true, "delta_db": 0.4202603149465852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1532747685680723e-06, "output_power": 2.871757502756417e-06, "reference_active": true, "delta_db": -0.4061405136526366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4027458558124955e-07, "output_power": 2.3057078294460607e-07, "reference_active": true, "delta_db": 2.1582527997097802}], "active_band_count": 3, "active_band_mae_db": 0.9948845427696673}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.819901036924925, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ccfbb49c3f5d15de43645280d22fcab73446af1b978b07cca0b623560ff33907", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9818651975833189, "standard_deviation": 0.002191004615308192, "minimum": 0.9787325120295899, "maximum": 0.9843559608281012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5253451849875015, "standard_deviation": 0.025253422671032925, "minimum": 1.4814145342896654, "maximum": 1.5593535137602628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5223144019691482, "standard_deviation": 0.21613813449188446, "minimum": 0.2803433729461205, "maximum": 0.9675128100330109, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15510314643844386, "standard_deviation": 0.009876949129957692, "minimum": 0.14501863250068037, "maximum": 0.1699115079716935, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.993874064265115, "standard_deviation": 0.001328223323999565, "minimum": 0.9917478404614566, "maximum": 0.9951532705842593, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1366687904339527, "standard_deviation": 0.04735613428943777, "minimum": 1.0909071307031153, "maximum": 1.2325749615083097, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.422157631885669, "standard_deviation": 0.08534600955958582, "minimum": 0.33696027908178666, "maximum": 0.5807512634604254, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09420219338925273, "standard_deviation": 0.009087848018588291, "minimum": 0.08511657106077113, "maximum": 0.11198397790182953, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9777284004384479, "standard_deviation": 0.003911286616395368, "minimum": 0.9721373013960848, "maximum": 0.9856628037969489, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3392842618130212, "standard_deviation": 0.045127795292176885, "minimum": 1.2750854272705496, "maximum": 1.425242212208575, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4856874368430447, "standard_deviation": 0.3443932784362644, "minimum": 0.10693251103763544, "maximum": 0.9989161024449907, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18212220617227895, "standard_deviation": 0.016500051208173886, "minimum": 0.14768426643002555, "maximum": 0.20181974442028155, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9899691329900553, "standard_deviation": 0.0017569531727614231, "minimum": 0.9880745735121456, "maximum": 0.9932656057658401, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0848404739475215, "standard_deviation": 0.04062607199043434, "minimum": 1.039933808875451, "maximum": 1.155585773969185, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6505665717527496, "standard_deviation": 0.390460371276299, "minimum": 0.14764173239589765, "maximum": 1.229944562160534, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12000672688338859, "standard_deviation": 0.010976761053314481, "minimum": 0.09975453827575083, "maximum": 0.13158233329526614, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9865193121060227, "standard_deviation": 0.0019588780174589087, "minimum": 0.9834140399625749, "maximum": 0.988392905086158, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6487655829468446, "standard_deviation": 0.05161897371436141, "minimum": 1.578166819828947, "maximum": 1.7320419410562773, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5365416665140872, "standard_deviation": 0.3689727094412286, "minimum": 0.25698360257676733, "maximum": 1.2180963547727737, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1286415994056916, "standard_deviation": 0.008951796019165292, "minimum": 0.12102151300503577, "maximum": 0.14320158974924768, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9955036524033811, "standard_deviation": 0.0010403244879558802, "minimum": 0.993285194350026, "maximum": 0.9964847652042579, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9490672553940964, "standard_deviation": 0.07758209201785135, "minimum": 0.8291127110608327, "maximum": 1.0425864343918214, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4568719749078134, "standard_deviation": 0.2820610248352204, "minimum": 0.2048609025988877, "maximum": 0.9095304404390893, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07401154549946692, "standard_deviation": 0.007813152026563838, "minimum": 0.06610077176581837, "maximum": 0.0905397256891189, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.24704786361166098, "standard_deviation": 0.03665994096925308, "minimum": 0.20050228751877328, "maximum": 0.30495865048206916, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.615432367506622, "standard_deviation": 0.00671368585198016, "minimum": 5.606682181909982, "maximum": 5.627400787528688, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4928764958119637, "standard_deviation": 0.014269241428209092, "minimum": 0.47491502603645214, "maximum": 0.51383426788701, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12984738167572096, "standard_deviation": 0.04139934637965811, "minimum": -0.19209361191812452, "maximum": -0.07151406839392677, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9913082780210256, "standard_deviation": 0.0009001782152260529, "minimum": 0.9899487103754515, "maximum": 0.9923810834749984, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.027838796814195, "standard_deviation": 0.02033835770742047, "minimum": 1.0078324450111493, "maximum": 1.0573391840892339, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.24914800210908813, "standard_deviation": 0.07288411011463204, "minimum": 0.15173801369539255, "maximum": 0.3704196263448423, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09188128913628545, "standard_deviation": 0.004870365475331678, "minimum": 0.08695441342804905, "maximum": 0.09957583360356709, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972474430054262, "standard_deviation": 0.00044397314625538683, "minimum": 0.9964919539430203, "maximum": 0.9979183520223005, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5400935226248129, "standard_deviation": 0.021485684166253124, "minimum": 0.5092938138699507, "maximum": 0.5768464439849943, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.25422552134360044, "standard_deviation": 0.12233917196450246, "minimum": 0.0707003498906198, "maximum": 0.4601156556658976, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05103344489718909, "standard_deviation": 0.004607021317998278, "minimum": 0.0437967188755647, "maximum": 0.05917609631648914, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-157963-0018", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.2057111263275146, "predicted_mos": 2.2057111263275146, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "b0c83e6673802f5bc7248822e2e7313ee110fab69b107d6b0faf27768fcc60e0", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.134697100147605, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0433985026962846, "全帯域_10ms_rms_envelope_nrmse": 0.09834228021237369, "全帯域_10ms_envelope_correlation": 0.9899635252165002, "全帯域_10ms_modulation_mae_db": 0.27467820251139496, "全帯域_40ms_log_rms_mae_db": 0.5128625275609245, "全帯域_40ms_rms_envelope_nrmse": 0.04603147121304803, "全帯域_40ms_envelope_correlation": 0.9977428305400462, "全帯域_40ms_modulation_mae_db": 0.2884503513791085, "300–1000 Hz_10ms_log_rms_mae_db": 1.6370787415239951, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12924924292266177, "300–1000 Hz_10ms_envelope_correlation": 0.9864334631175399, "300–1000 Hz_10ms_modulation_mae_db": 0.25393278440830164, "300–1000 Hz_40ms_log_rms_mae_db": 0.9635793689433438, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06606672139386494, "300–1000 Hz_40ms_envelope_correlation": 0.9965667333471763, "300–1000 Hz_40ms_modulation_mae_db": 0.4173635175772133, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6269788678124177, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17787352914383356, "1000–3000 Hz_10ms_envelope_correlation": 0.9751768133648256, "1000–3000 Hz_10ms_modulation_mae_db": 0.8277377825380823, "1000–3000 Hz_40ms_log_rms_mae_db": 1.173167394667315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10225832564136395, "1000–3000 Hz_40ms_envelope_correlation": 0.9916317180215848, "1000–3000 Hz_40ms_modulation_mae_db": 0.5709735732007425, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5607898735746344, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22381507623018215, "3000–6000 Hz_10ms_envelope_correlation": 0.9664720517715054, "3000–6000 Hz_10ms_modulation_mae_db": 0.4272216115091724, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2872431721173543, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15162516101922693, "3000–6000 Hz_40ms_envelope_correlation": 0.9839140509643796, "3000–6000 Hz_40ms_modulation_mae_db": 0.5813875123956301, "mr_stft_db": 5.718602799523978, "band_mae_db": 0.15593454468068055, "rms_gain_error_db": -0.11080268170198138, "output_peak": 0.506013110814265}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0033679847011212, "全帯域_10ms_rms_envelope_nrmse": 0.09718593911282693, "全帯域_10ms_envelope_correlation": 0.9905255087669061, "全帯域_10ms_modulation_mae_db": 0.14808334290772382, "全帯域_40ms_log_rms_mae_db": 0.4822980110627714, "全帯域_40ms_rms_envelope_nrmse": 0.05122377996799309, "全帯域_40ms_envelope_correlation": 0.9976244994563075, "全帯域_40ms_modulation_mae_db": 0.33347458308394445, "300–1000 Hz_10ms_log_rms_mae_db": 1.6914320626617871, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13085281525517894, "300–1000 Hz_10ms_envelope_correlation": 0.9866993723103372, "300–1000 Hz_10ms_modulation_mae_db": 0.420768277595013, "300–1000 Hz_40ms_log_rms_mae_db": 1.0273463354610035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07358558898512256, "300–1000 Hz_40ms_envelope_correlation": 0.9964819839642468, "300–1000 Hz_40ms_modulation_mae_db": 0.5233818334328056, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5999199792898964, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15902283389925667, "1000–3000 Hz_10ms_envelope_correlation": 0.9803556131762556, "1000–3000 Hz_10ms_modulation_mae_db": 0.3370332758459908, "1000–3000 Hz_40ms_log_rms_mae_db": 1.165421319672218, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09451998518576796, "1000–3000 Hz_40ms_envelope_correlation": 0.993032668217911, "1000–3000 Hz_40ms_modulation_mae_db": 0.3463873870955028, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5250750884362665, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19382442005860084, "3000–6000 Hz_10ms_envelope_correlation": 0.9753370574423431, "3000–6000 Hz_10ms_modulation_mae_db": 0.5999450858351268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2181428320452852, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12375988893907039, "3000–6000 Hz_40ms_envelope_correlation": 0.9896451783203691, "3000–6000 Hz_40ms_modulation_mae_db": 0.7217581478191702, "mr_stft_db": 5.725011557639367, "band_mae_db": 0.2860178865547216, "rms_gain_error_db": -0.20819301559041414, "output_peak": 0.49328931484268956}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0886247376611518, "全帯域_10ms_rms_envelope_nrmse": 0.09589918598236609, "全帯域_10ms_envelope_correlation": 0.9906351929851106, "全帯域_10ms_modulation_mae_db": 0.2663990895413782, "全帯域_40ms_log_rms_mae_db": 0.5748386655234495, "全帯域_40ms_rms_envelope_nrmse": 0.047928545754442796, "全帯域_40ms_envelope_correlation": 0.997762550176773, "全帯域_40ms_modulation_mae_db": 0.5097902473773538, "300–1000 Hz_10ms_log_rms_mae_db": 1.7732461289081434, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13093593293814526, "300–1000 Hz_10ms_envelope_correlation": 0.9864112342622028, "300–1000 Hz_10ms_modulation_mae_db": 0.7521482617045079, "300–1000 Hz_40ms_log_rms_mae_db": 0.9684908683370276, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07446319561148382, "300–1000 Hz_40ms_envelope_correlation": 0.9959831974727513, "300–1000 Hz_40ms_modulation_mae_db": 0.637135220318263, "1000–3000 Hz_10ms_log_rms_mae_db": 1.67031377578897, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1652571233614978, "1000–3000 Hz_10ms_envelope_correlation": 0.978850568691638, "1000–3000 Hz_10ms_modulation_mae_db": 0.42289725366548847, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1998391036903873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08990581272743195, "1000–3000 Hz_40ms_envelope_correlation": 0.9939993972224519, "1000–3000 Hz_40ms_modulation_mae_db": 0.6641917718452389, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5027897685157046, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17742622463229651, "3000–6000 Hz_10ms_envelope_correlation": 0.9784744213282786, "3000–6000 Hz_10ms_modulation_mae_db": 0.43880085972625404, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1845957316896636, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11953705281870271, "3000–6000 Hz_40ms_envelope_correlation": 0.9899941095622704, "3000–6000 Hz_40ms_modulation_mae_db": 0.41280553163346084, "mr_stft_db": 5.750707787905458, "band_mae_db": 0.31944648832745254, "rms_gain_error_db": -0.16978541767842975, "output_peak": 0.4966866967780653}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0440672027385416, "全帯域_10ms_rms_envelope_nrmse": 0.084324832781273, "全帯域_10ms_envelope_correlation": 0.9926262119419546, "全帯域_10ms_modulation_mae_db": 0.20292674522114196, "全帯域_40ms_log_rms_mae_db": 0.553949167115635, "全帯域_40ms_rms_envelope_nrmse": 0.047437593531801166, "全帯域_40ms_envelope_correlation": 0.9974654623006819, "全帯域_40ms_modulation_mae_db": 0.27710953841067154, "300–1000 Hz_10ms_log_rms_mae_db": 1.7065729266723488, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11954438561547479, "300–1000 Hz_10ms_envelope_correlation": 0.9883089656484769, "300–1000 Hz_10ms_modulation_mae_db": 0.3560969933480234, "300–1000 Hz_40ms_log_rms_mae_db": 1.0118806295189706, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06564263143095804, "300–1000 Hz_40ms_envelope_correlation": 0.996191151250596, "300–1000 Hz_40ms_modulation_mae_db": 0.3829083832141606, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6072787618525202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15939802963719196, "1000–3000 Hz_10ms_envelope_correlation": 0.9804833146543472, "1000–3000 Hz_10ms_modulation_mae_db": 0.6415745156749368, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1119704238748591, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08761197595916975, "1000–3000 Hz_40ms_envelope_correlation": 0.9945157043637198, "1000–3000 Hz_40ms_modulation_mae_db": 0.6640906631592645, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5597925352610107, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21308591079261419, "3000–6000 Hz_10ms_envelope_correlation": 0.96932267236014, "3000–6000 Hz_10ms_modulation_mae_db": 0.3655533887110804, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2421437529601218, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14596271270585942, "3000–6000 Hz_40ms_envelope_correlation": 0.9862329678135217, "3000–6000 Hz_40ms_modulation_mae_db": 0.4817466851593042, "mr_stft_db": 5.721463560204817, "band_mae_db": 0.27763765658119127, "rms_gain_error_db": -0.01954714534279449, "output_peak": 0.48847787242564533}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0477899680876201, "全帯域_10ms_rms_envelope_nrmse": 0.08977481579092318, "全帯域_10ms_envelope_correlation": 0.9916972027624422, "全帯域_10ms_modulation_mae_db": 0.3002269288875166, "全帯域_40ms_log_rms_mae_db": 0.5793182232048449, "全帯域_40ms_rms_envelope_nrmse": 0.04601756440763344, "全帯域_40ms_envelope_correlation": 0.9977940706346551, "全帯域_40ms_modulation_mae_db": 0.37320325488536565, "300–1000 Hz_10ms_log_rms_mae_db": 1.6799259071594512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1338013071124532, "300–1000 Hz_10ms_envelope_correlation": 0.9855537499801109, "300–1000 Hz_10ms_modulation_mae_db": 0.5015597586403827, "300–1000 Hz_40ms_log_rms_mae_db": 1.000368652683027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07686888230497883, "300–1000 Hz_40ms_envelope_correlation": 0.9953670314141779, "300–1000 Hz_40ms_modulation_mae_db": 0.5591993389885935, "1000–3000 Hz_10ms_log_rms_mae_db": 1.763075718856944, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16800528104617932, "1000–3000 Hz_10ms_envelope_correlation": 0.9786414745775317, "1000–3000 Hz_10ms_modulation_mae_db": 0.7828357813286355, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2821100099062364, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09598419934259796, "1000–3000 Hz_40ms_envelope_correlation": 0.9938848681105824, "1000–3000 Hz_40ms_modulation_mae_db": 0.9076448971105618, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5471973031295136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18867989563821513, "3000–6000 Hz_10ms_envelope_correlation": 0.9756597478523602, "3000–6000 Hz_10ms_modulation_mae_db": 0.42293523234681035, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2645289304563954, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12487933439471599, "3000–6000 Hz_40ms_envelope_correlation": 0.9892390589338899, "3000–6000 Hz_40ms_modulation_mae_db": 0.7335362363100022, "mr_stft_db": 5.7344524368274525, "band_mae_db": 0.30408160142586377, "rms_gain_error_db": -0.12205815455229403, "output_peak": 0.5230022576094948}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0456579415722622, "全帯域_10ms_rms_envelope_nrmse": 0.09436152837661348, "全帯域_10ms_envelope_correlation": 0.9907236385749814, "全帯域_10ms_modulation_mae_db": 0.08844093661210338, "全帯域_40ms_log_rms_mae_db": 0.5615545906955849, "全帯域_40ms_rms_envelope_nrmse": 0.05347271818206056, "全帯域_40ms_envelope_correlation": 0.9967605405487104, "全帯域_40ms_modulation_mae_db": 0.1855515116964673, "300–1000 Hz_10ms_log_rms_mae_db": 1.7014539624002676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13062897663120335, "300–1000 Hz_10ms_envelope_correlation": 0.9860495952726966, "300–1000 Hz_10ms_modulation_mae_db": 0.3696421956781679, "300–1000 Hz_40ms_log_rms_mae_db": 1.0002778940933243, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07553430882272862, "300–1000 Hz_40ms_envelope_correlation": 0.9951997176867503, "300–1000 Hz_40ms_modulation_mae_db": 0.5229034228867101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.637024277555909, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13579832572267758, "1000–3000 Hz_10ms_envelope_correlation": 0.9860220039329507, "1000–3000 Hz_10ms_modulation_mae_db": 0.7860809491071014, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1644871697451222, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0816915900676387, "1000–3000 Hz_40ms_envelope_correlation": 0.9951771416564683, "1000–3000 Hz_40ms_modulation_mae_db": 0.9482032588488226, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5932915622829835, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19452406945742343, "3000–6000 Hz_10ms_envelope_correlation": 0.9745147699066298, "3000–6000 Hz_10ms_modulation_mae_db": 0.6865358375179386, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2289753328759778, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12422788080187647, "3000–6000 Hz_40ms_envelope_correlation": 0.9891810211533072, "3000–6000 Hz_40ms_modulation_mae_db": 0.33055606839597845, "mr_stft_db": 5.726808269038874, "band_mae_db": 0.23849203555464474, "rms_gain_error_db": -0.051468210112193956, "output_peak": 0.4920866165454402}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9906089088473503, "全帯域_10ms_rms_envelope_nrmse": 0.08624154760129797, "全帯域_10ms_envelope_correlation": 0.9922477823424695, "全帯域_10ms_modulation_mae_db": 0.11269940785212214, "全帯域_40ms_log_rms_mae_db": 0.5535000918417003, "全帯域_40ms_rms_envelope_nrmse": 0.0426178646647762, "全帯域_40ms_envelope_correlation": 0.9979330455610945, "全帯域_40ms_modulation_mae_db": 0.08927001359822058, "300–1000 Hz_10ms_log_rms_mae_db": 1.7595467657863735, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12161685356947624, "300–1000 Hz_10ms_envelope_correlation": 0.987940220040005, "300–1000 Hz_10ms_modulation_mae_db": 0.4016784619258224, "300–1000 Hz_40ms_log_rms_mae_db": 1.0211432254654125, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0680966787347655, "300–1000 Hz_40ms_envelope_correlation": 0.9959267789742858, "300–1000 Hz_40ms_modulation_mae_db": 0.4445569422585989, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6660989949694367, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15307311662045625, "1000–3000 Hz_10ms_envelope_correlation": 0.9830230983550464, "1000–3000 Hz_10ms_modulation_mae_db": 0.5999636940492673, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2034208693037949, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09115114437596893, "1000–3000 Hz_40ms_envelope_correlation": 0.9952344296656561, "1000–3000 Hz_40ms_modulation_mae_db": 0.7673073845120596, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5392735580221404, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20217295818158967, "3000–6000 Hz_10ms_envelope_correlation": 0.9735330414036232, "3000–6000 Hz_10ms_modulation_mae_db": 0.5525098667795116, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2476872741357408, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13265644788567993, "3000–6000 Hz_40ms_envelope_correlation": 0.9883550504651774, "3000–6000 Hz_40ms_modulation_mae_db": 0.4599293770573629, "mr_stft_db": 5.733315300799563, "band_mae_db": 0.24329072245824326, "rms_gain_error_db": -0.011506959683786839, "output_peak": 0.47810026198989153}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0521593376076062, "全帯域_10ms_rms_envelope_nrmse": 0.09974857183765677, "全帯域_10ms_envelope_correlation": 0.989931559013905, "全帯域_10ms_modulation_mae_db": 0.3372540485765188, "全帯域_40ms_log_rms_mae_db": 0.5193983671038548, "全帯域_40ms_rms_envelope_nrmse": 0.05562559471050153, "全帯域_40ms_envelope_correlation": 0.9969976056409714, "全帯域_40ms_modulation_mae_db": 0.40850306255205515, "300–1000 Hz_10ms_log_rms_mae_db": 1.6912628360661832, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12972011159067307, "300–1000 Hz_10ms_envelope_correlation": 0.9867988764141512, "300–1000 Hz_10ms_modulation_mae_db": 0.44663232486673904, "300–1000 Hz_40ms_log_rms_mae_db": 0.9674821421940359, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07386896619567056, "300–1000 Hz_40ms_envelope_correlation": 0.9962429558195192, "300–1000 Hz_40ms_modulation_mae_db": 0.5538723394935295, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6891486875511135, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1582176817864735, "1000–3000 Hz_10ms_envelope_correlation": 0.9814849252745312, "1000–3000 Hz_10ms_modulation_mae_db": 0.8035115137284065, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1909986211324846, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10110164653140624, "1000–3000 Hz_40ms_envelope_correlation": 0.9931777273438547, "1000–3000 Hz_40ms_modulation_mae_db": 0.9730628949482529, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5315766414530112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19258955026866487, "3000–6000 Hz_10ms_envelope_correlation": 0.9748133815615774, "3000–6000 Hz_10ms_modulation_mae_db": 0.7721639924997818, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1906545807610212, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12378713948530819, "3000–6000 Hz_40ms_envelope_correlation": 0.9891727758007792, "3000–6000 Hz_40ms_modulation_mae_db": 0.5713876437795951, "mr_stft_db": 5.725044701279065, "band_mae_db": 0.2727263388988332, "rms_gain_error_db": -0.19589121032738682, "output_peak": 0.5030283048694942}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68080, "output_samples": 68080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.804479117841189, "spectral_convergence": 0.2970584457240224, "spectral_convergence_status": "ok", "amplitude_floor": 2.0850976744763412e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.470562289129908, "spectral_convergence": 0.27034541455278216, "spectral_convergence_status": "ok", "amplitude_floor": 1.4381271492089952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8807669916008365, "spectral_convergence": 0.2851154362342357, "spectral_convergence_status": "ok", "amplitude_floor": 9.411979934974602e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005432476915893821, "output_power": 0.0005891590783391548, "delta_db": 0.35234684544527994, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030286578008489277, "output_power": 0.002877217648007006, "delta_db": -0.22277490975022446, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038180407470981095, "output_power": 0.0003885192962278006, "delta_db": 0.07572034390999513, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002532984259990213, "output_power": 0.00024014998561432225, "delta_db": -0.231499261257168, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013717204322211298, "output_power": 0.00013681105883265487, "delta_db": -0.011444035630998215, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.29065897621083e-05, "output_power": 8.370882360900178e-05, "delta_db": 0.04182187209041764, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.15593454468068055, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.11080268170198138, "pitch": null, "output_rms_dbfs": -23.64962248137868, "output_peak": 0.506013110814265, "reference_rms_dbfs": -23.538819799676702, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.718602799523978}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00026228408676881105, "log_rms_mae_db": 1.0433985026962846, "rms_envelope_nrmse": 0.09834228021237369, "envelope_correlation": 0.9899635252165002, "rms_gain_error_db": -0.11080196876063415, "reference_mean_power": 0.004432288013980724, "output_mean_power": 0.004320636961951346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 1.6030570640082068e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008005952323264084, "output_power": 0.0007786170798075117, "reference_active": true, "delta_db": -0.12089073277510712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020547999455019126, "output_power": 0.00019216835581724597, "reference_active": true, "delta_db": -0.29087671178414304}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.989786705158196e-05, "output_power": 4.3870884968370366e-05, "reference_active": true, "delta_db": 0.41226716297493465}], "active_band_count": 3, "active_band_mae_db": 0.27467820251139496}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00021831174911177267, "log_rms_mae_db": 0.5128625275609245, "rms_envelope_nrmse": 0.04603147121304803, "envelope_correlation": 0.9977428305400462, "rms_gain_error_db": -0.11081174361184701, "reference_mean_power": 0.004463740704660867, "output_mean_power": 0.00435128755352969}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 1.4245155353761532e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007405152195784611, "output_power": 0.0007253684755731571, "reference_active": true, "delta_db": -0.08975311599543259}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001021109367695941, "output_power": 9.624386132865973e-05, "reference_active": true, "delta_db": -0.25699221755995116}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.434051823082556e-06, "output_power": 2.1600770784506947e-06, "reference_active": true, "delta_db": -0.5186057205819418}], "active_band_count": 3, "active_band_mae_db": 0.2884503513791085}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00023386465365005232, "log_rms_mae_db": 1.6370787415239951, "rms_envelope_nrmse": 0.12924924292266177, "envelope_correlation": 0.9864334631175399, "rms_gain_error_db": -0.20969099581785772, "reference_mean_power": 0.002112021838569365, "output_mean_power": 0.0020124694899285936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 9.681983603336737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048709120906038954, "output_power": 0.0004758785027274918, "reference_active": true, "delta_db": -0.10114205088857409}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014674692890756196, "output_power": 0.0001261259796862194, "reference_active": true, "delta_db": -0.6576446853094493}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8361558797573682e-05, "output_power": 2.8341898279527854e-05, "reference_active": true, "delta_db": -0.0030116170268814388}], "active_band_count": 3, "active_band_mae_db": 0.25393278440830164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00020000043006212289, "log_rms_mae_db": 0.9635793689433438, "rms_envelope_nrmse": 0.06606672139386494, "envelope_correlation": 0.9965667333471763, "rms_gain_error_db": -0.2096957116405938, "reference_mean_power": 0.002127032769351675, "output_mean_power": 0.002026770664070404}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.787606965551091e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004705886462934124, "output_power": 0.00045992868957920723, "reference_active": true, "delta_db": -0.09950943935406781}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.123456963821095e-05, "output_power": 6.137963799976142e-05, "reference_active": true, "delta_db": -0.6466648211303248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5465229502268847e-06, "output_power": 1.3764636281984772e-06, "reference_active": true, "delta_db": -0.5059162922472472}], "active_band_count": 3, "active_band_mae_db": 0.4173635175772133}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 7.720600674415526e-05, "log_rms_mae_db": 1.6269788678124177, "rms_envelope_nrmse": 0.17787352914383356, "envelope_correlation": 0.9751768133648256, "rms_gain_error_db": -0.22801923870798052, "reference_mean_power": 0.00033128809071676906, "output_mean_power": 0.00031434307705408297}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 2.2485323271997473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014154142624946028, "output_power": 0.00013409747248155623, "reference_active": true, "delta_db": -0.23462975295229405}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6917189688068767e-05, "output_power": 2.557510553588488e-05, "reference_active": true, "delta_db": -0.22212280528821748}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.319943953798946e-06, "output_power": 1.007765080643506e-05, "reference_active": true, "delta_db": 2.0264607893737354}], "active_band_count": 3, "active_band_mae_db": 0.8277377825380823}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 6.30102051319373e-05, "log_rms_mae_db": 1.173167394667315, "rms_envelope_nrmse": 0.10225832564136395, "envelope_correlation": 0.9916317180215848, "rms_gain_error_db": -0.2280222859926319, "reference_mean_power": 0.0003336427865109939, "output_mean_power": 0.00031657711067316525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 2.017085687613918e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013578457159768264, "output_power": 0.0001306276978980176, "reference_active": true, "delta_db": -0.1681515324130552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.310083041117988e-05, "output_power": 1.192958044391206e-05, "reference_active": true, "delta_db": -0.40673654716554375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7328752869332565e-07, "output_power": 4.851172800881385e-07, "reference_active": true, "delta_db": 1.1380326400236287}], "active_band_count": 3, "active_band_mae_db": 0.5709735732007425}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 5.974138539374114e-05, "log_rms_mae_db": 1.5607898735746344, "rms_envelope_nrmse": 0.22381507623018215, "envelope_correlation": 0.9664720517715054, "rms_gain_error_db": 0.09177227945736875, "reference_mean_power": 8.138813397574021e-05, "output_mean_power": 8.312627493243498e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 8.567729341398199e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0300307346839795e-05, "output_power": 4.107447912481359e-05, "reference_active": true, "delta_db": 0.08263706487575459}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.809515102528064e-06, "output_power": 5.1929795804948325e-06, "reference_active": true, "delta_db": -0.4872327015040612}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.571956353556118e-06, "output_power": 4.208096110863657e-06, "reference_active": true, "delta_db": 0.7117950681477014}], "active_band_count": 3, "active_band_mae_db": 0.4272216115091724}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 4.362120322552387e-05, "log_rms_mae_db": 1.2872431721173543, "rms_envelope_nrmse": 0.15162516101922693, "envelope_correlation": 0.9839140509643796, "rms_gain_error_db": 0.0917689622638982, "reference_mean_power": 8.196654666810843e-05, "output_mean_power": 8.371697637568242e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.099518469769939e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.90961297245674e-05, "output_power": 4.0224948241199096e-05, "reference_active": true, "delta_db": 0.12361726898346448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1532747685680723e-06, "output_power": 2.7184473699366976e-06, "reference_active": true, "delta_db": -0.6444088641369682}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4027458558124955e-07, "output_power": 1.7562754666651142e-07, "reference_active": true, "delta_db": 0.9761364040664576}], "active_band_count": 3, "active_band_mae_db": 0.5813875123956301}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8228520690229705, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4e14cf4d41ca36292d236ef23b13537b2c1365c658cbbc16bf9f807af926a078", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9805047265033908, "standard_deviation": 0.003215156830611446, "minimum": 0.9751768133648256, "maximum": 0.9860220039329507, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6574798829596507, "standard_deviation": 0.05288049927960418, "minimum": 1.5999199792898964, "maximum": 1.763075718856944, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6502043457422386, "standard_deviation": 0.18647787619141976, "minimum": 0.3370332758459908, "maximum": 0.8277377825380823, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1595807401521958, "standard_deviation": 0.012235634816047777, "minimum": 0.13579832572267758, "maximum": 0.17787352914383356, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9938317068252787, "standard_deviation": 0.0012060992419847764, "minimum": 0.9916317180215848, "maximum": 0.9952344296656561, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1864268639990523, "standard_deviation": 0.04823603845846764, "minimum": 1.1119704238748591, "maximum": 1.2821100099062364, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7302327288400557, "standard_deviation": 0.21406412755972062, "minimum": 0.3463873870955028, "maximum": 0.9730628949482529, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09302808497891818, "standard_deviation": 0.006891671588783287, "minimum": 0.0816915900676387, "maximum": 0.10225832564136395, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9735158929533072, "standard_deviation": 0.003824359536164029, "minimum": 0.9664720517715054, "maximum": 0.9784744213282786, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5449732913344083, "standard_deviation": 0.027246811252721728, "minimum": 1.5027897685157046, "maximum": 1.5932915622829835, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5332082343657095, "standard_deviation": 0.14434569984604484, "minimum": 0.3655533887110804, "maximum": 0.7721639924997818, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19826476315744834, "standard_deviation": 0.014539555674269164, "minimum": 0.17742622463229651, "maximum": 0.22381507623018215, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9882167766267118, "standard_deviation": 0.002089294891061435, "minimum": 0.9839140509643796, "maximum": 0.9899941095622704, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.232996450880195, "standard_deviation": 0.03506883113926294, "minimum": 1.1845957316896636, "maximum": 1.2872431721173543, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5366384003188129, "standard_deviation": 0.14290354050689721, "minimum": 0.33055606839597845, "maximum": 0.7335362363100022, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.130804452256305, "standard_deviation": 0.011777010388101122, "minimum": 0.11953705281870271, "maximum": 0.15162516101922693, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9867744346306901, "standard_deviation": 0.0009245626000085688, "minimum": 0.9855537499801109, "maximum": 0.9883089656484769, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7050649163973188, "standard_deviation": 0.04352783119185362, "minimum": 1.6370787415239951, "maximum": 1.7732461289081434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.43780738227086974, "standard_deviation": 0.14622927771209587, "minimum": 0.25393278440830164, "maximum": 0.7521482617045079, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12829370320440833, "standard_deviation": 0.004977036002584123, "minimum": 0.11954438561547479, "maximum": 0.1338013071124532, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9959949437411879, "standard_deviation": 0.0004921674640195986, "minimum": 0.9951997176867503, "maximum": 0.9965667333471763, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9950711395870182, "standard_deviation": 0.025406888928469408, "minimum": 0.9635793689433438, "maximum": 1.0273463354610035, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5051651247712343, "standard_deviation": 0.08424782808117656, "minimum": 0.3829083832141606, "maximum": 0.637135220318263, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07176587168494661, "standard_deviation": 0.004451149641896772, "minimum": 0.06564263143095804, "maximum": 0.07686888230497883, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2622034093102039, "standard_deviation": 0.05093359961980482, "minimum": 0.15593454468068055, "maximum": 0.31944648832745254, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.729425801652322, "standard_deviation": 0.010137511213918987, "minimum": 5.718602799523978, "maximum": 5.750707787905458, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.49758555448437325, "standard_deviation": 0.013409945509468384, "minimum": 0.47810026198989153, "maximum": 0.5230022576094948, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11115659937366018, "standard_deviation": 0.07749412368071946, "minimum": -0.20819301559041414, "maximum": -0.011506959683786839, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9910438277005337, "standard_deviation": 0.0010227902068676875, "minimum": 0.989931559013905, "maximum": 0.9926262119419546, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0394593229889924, "standard_deviation": 0.03026560576220068, "minimum": 0.9906089088473503, "maximum": 1.0886247376611518, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2163385877637375, "standard_deviation": 0.09219924982713104, "minimum": 0.08844093661210338, "maximum": 0.3372540485765188, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09323483771191639, "standard_deviation": 0.00576640681789015, "minimum": 0.084324832781273, "maximum": 0.09974857183765677, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.997510075607405, "standard_deviation": 0.0004169594805795857, "minimum": 0.9967605405487104, "maximum": 0.9979330455610945, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5422149555135958, "standard_deviation": 0.03391634331433899, "minimum": 0.4822980110627714, "maximum": 0.5793182232048449, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3081690703728984, "standard_deviation": 0.13087246151945023, "minimum": 0.08927001359822058, "maximum": 0.5097902473773538, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.048794391554032104, "standard_deviation": 0.0043183756323821485, "minimum": 0.0426178646647762, "maximum": 0.05562559471050153, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-157963-0018", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.527702808380127, "predicted_mos": 2.527702808380127, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "b548bd2af64278c229985fc203a4ee258634f68336d9f1515d0380aa90e0967f", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6914892997592688, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9763894606419906, "全帯域_10ms_rms_envelope_nrmse": 0.08770542955899435, "全帯域_10ms_envelope_correlation": 0.9921771261851612, "全帯域_10ms_modulation_mae_db": 0.2286439422368569, "全帯域_40ms_log_rms_mae_db": 0.509395852925305, "全帯域_40ms_rms_envelope_nrmse": 0.04824662816780399, "全帯域_40ms_envelope_correlation": 0.9976620554291425, "全帯域_40ms_modulation_mae_db": 0.4902283466960669, "300–1000 Hz_10ms_log_rms_mae_db": 1.7095065518421766, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13180545544341, "300–1000 Hz_10ms_envelope_correlation": 0.9867177136512864, "300–1000 Hz_10ms_modulation_mae_db": 0.36350040141558476, "300–1000 Hz_40ms_log_rms_mae_db": 1.0145689499388477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08064368438848296, "300–1000 Hz_40ms_envelope_correlation": 0.9957468469239413, "300–1000 Hz_40ms_modulation_mae_db": 0.41545926530970395, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5969299581794176, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15659626026641976, "1000–3000 Hz_10ms_envelope_correlation": 0.9807489424371919, "1000–3000 Hz_10ms_modulation_mae_db": 0.22402878930182632, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1268530454389099, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09293037193457143, "1000–3000 Hz_40ms_envelope_correlation": 0.9927781358281482, "1000–3000 Hz_40ms_modulation_mae_db": 0.33256209612172055, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4864727405718547, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18931492405461547, "3000–6000 Hz_10ms_envelope_correlation": 0.9775981969459198, "3000–6000 Hz_10ms_modulation_mae_db": 0.20639614999739053, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1613729863067395, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11615632678678205, "3000–6000 Hz_40ms_envelope_correlation": 0.9917727519741469, "3000–6000 Hz_40ms_modulation_mae_db": 0.5171718526017233, "mr_stft_db": 5.695630970921338, "band_mae_db": 0.22314923808273032, "rms_gain_error_db": -0.15501327214714983, "output_peak": 0.5015192925856209}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0146158239383183, "全帯域_10ms_rms_envelope_nrmse": 0.10080380743199634, "全帯域_10ms_envelope_correlation": 0.9896076198406059, "全帯域_10ms_modulation_mae_db": 0.2282574571307666, "全帯域_40ms_log_rms_mae_db": 0.5622122715852971, "全帯域_40ms_rms_envelope_nrmse": 0.05150399493237772, "全帯域_40ms_envelope_correlation": 0.9973723095041014, "全帯域_40ms_modulation_mae_db": 0.15761766234682498, "300–1000 Hz_10ms_log_rms_mae_db": 1.7535252993365391, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1455135488962878, "300–1000 Hz_10ms_envelope_correlation": 0.9831443656276335, "300–1000 Hz_10ms_modulation_mae_db": 0.5031255222083463, "300–1000 Hz_40ms_log_rms_mae_db": 1.0212538019109467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08186002283969497, "300–1000 Hz_40ms_envelope_correlation": 0.9951995043899342, "300–1000 Hz_40ms_modulation_mae_db": 0.3806821486993148, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6463559676603858, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1711204129936486, "1000–3000 Hz_10ms_envelope_correlation": 0.9773958239444477, "1000–3000 Hz_10ms_modulation_mae_db": 0.7825034392397653, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1690709374452806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10556946441993603, "1000–3000 Hz_40ms_envelope_correlation": 0.9915195537978911, "1000–3000 Hz_40ms_modulation_mae_db": 0.9737417707978019, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4951790360884034, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18542301596311137, "3000–6000 Hz_10ms_envelope_correlation": 0.9765173815724437, "3000–6000 Hz_10ms_modulation_mae_db": 0.630408570414228, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1965606090647058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12891975137553383, "3000–6000 Hz_40ms_envelope_correlation": 0.9884895583760098, "3000–6000 Hz_40ms_modulation_mae_db": 0.7122600889732497, "mr_stft_db": 5.706332805518897, "band_mae_db": 0.27515511018033334, "rms_gain_error_db": -0.17208822542235905, "output_peak": 0.48795618533322677}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0362907523596079, "全帯域_10ms_rms_envelope_nrmse": 0.09156370207423499, "全帯域_10ms_envelope_correlation": 0.9913140928908546, "全帯域_10ms_modulation_mae_db": 0.23054041520456506, "全帯域_40ms_log_rms_mae_db": 0.5547535226334057, "全帯域_40ms_rms_envelope_nrmse": 0.05237303597795612, "全帯域_40ms_envelope_correlation": 0.9969688797967148, "全帯域_40ms_modulation_mae_db": 0.028441168442661754, "300–1000 Hz_10ms_log_rms_mae_db": 1.672915529464665, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13352612397344854, "300–1000 Hz_10ms_envelope_correlation": 0.9853567075262778, "300–1000 Hz_10ms_modulation_mae_db": 0.3229010026534405, "300–1000 Hz_40ms_log_rms_mae_db": 0.9714150781139429, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08199363831343702, "300–1000 Hz_40ms_envelope_correlation": 0.9941173771923568, "300–1000 Hz_40ms_modulation_mae_db": 0.24075143549855813, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6095358059252725, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16485698812298802, "1000–3000 Hz_10ms_envelope_correlation": 0.9786516093802062, "1000–3000 Hz_10ms_modulation_mae_db": 0.5140059313768309, "1000–3000 Hz_40ms_log_rms_mae_db": 1.157326932676047, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08312974453757296, "1000–3000 Hz_40ms_envelope_correlation": 0.9943847900177861, "1000–3000 Hz_40ms_modulation_mae_db": 0.5506223810276728, "3000–6000 Hz_10ms_log_rms_mae_db": 1.48995416136183, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22228479305071577, "3000–6000 Hz_10ms_envelope_correlation": 0.9660391984941997, "3000–6000 Hz_10ms_modulation_mae_db": 1.0334590590219204, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1791689165011774, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14096212819922502, "3000–6000 Hz_40ms_envelope_correlation": 0.9863861822693444, "3000–6000 Hz_40ms_modulation_mae_db": 0.5167993527577999, "mr_stft_db": 5.695731145295343, "band_mae_db": 0.19081584062805437, "rms_gain_error_db": -0.09049185627263923, "output_peak": 0.507013428908619}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9903978366673608, "全帯域_10ms_rms_envelope_nrmse": 0.09130225623434099, "全帯域_10ms_envelope_correlation": 0.9913755080224886, "全帯域_10ms_modulation_mae_db": 0.3905928840290753, "全帯域_40ms_log_rms_mae_db": 0.5030955298268949, "全帯域_40ms_rms_envelope_nrmse": 0.04843726649575204, "全帯域_40ms_envelope_correlation": 0.9974734011348672, "全帯域_40ms_modulation_mae_db": 0.3436715597455571, "300–1000 Hz_10ms_log_rms_mae_db": 1.7019543709254676, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11989469917415677, "300–1000 Hz_10ms_envelope_correlation": 0.9884097484106877, "300–1000 Hz_10ms_modulation_mae_db": 0.6172395550288498, "300–1000 Hz_40ms_log_rms_mae_db": 0.9732764062701437, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07338777834931937, "300–1000 Hz_40ms_envelope_correlation": 0.9956369596620326, "300–1000 Hz_40ms_modulation_mae_db": 0.4869535840195729, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6120581544792152, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15902803243200211, "1000–3000 Hz_10ms_envelope_correlation": 0.9801551402668045, "1000–3000 Hz_10ms_modulation_mae_db": 0.4850110440947312, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1693372205098347, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0943577430463547, "1000–3000 Hz_40ms_envelope_correlation": 0.992663558261609, "1000–3000 Hz_40ms_modulation_mae_db": 0.43660428571023674, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5295747859374158, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17073216732177196, "3000–6000 Hz_10ms_envelope_correlation": 0.9803801597226942, "3000–6000 Hz_10ms_modulation_mae_db": 0.1978436504627676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1820318216775458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11467441972077566, "3000–6000 Hz_40ms_envelope_correlation": 0.9907662576521395, "3000–6000 Hz_40ms_modulation_mae_db": 0.18783494646784052, "mr_stft_db": 5.701202663578548, "band_mae_db": 0.1583572737867576, "rms_gain_error_db": -0.11064118780736926, "output_peak": 0.49659741567707877}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9872847211354779, "全帯域_10ms_rms_envelope_nrmse": 0.09415032550988602, "全帯域_10ms_envelope_correlation": 0.9909189151629315, "全帯域_10ms_modulation_mae_db": 0.15153426774367085, "全帯域_40ms_log_rms_mae_db": 0.49555792379704916, "全帯域_40ms_rms_envelope_nrmse": 0.04642039477503888, "全帯域_40ms_envelope_correlation": 0.9978441226790719, "全帯域_40ms_modulation_mae_db": 0.3055814838405675, "300–1000 Hz_10ms_log_rms_mae_db": 1.628985525193065, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13159121944510804, "300–1000 Hz_10ms_envelope_correlation": 0.9860628141022567, "300–1000 Hz_10ms_modulation_mae_db": 0.6671273628019477, "300–1000 Hz_40ms_log_rms_mae_db": 0.9203842621616659, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07586562270150865, "300–1000 Hz_40ms_envelope_correlation": 0.9954953005051762, "300–1000 Hz_40ms_modulation_mae_db": 0.6602944923112699, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5619157750373287, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15561590280655346, "1000–3000 Hz_10ms_envelope_correlation": 0.9814415425429227, "1000–3000 Hz_10ms_modulation_mae_db": 0.4463974121363688, "1000–3000 Hz_40ms_log_rms_mae_db": 1.106399589945543, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08049705690482138, "1000–3000 Hz_40ms_envelope_correlation": 0.9955409771677429, "1000–3000 Hz_40ms_modulation_mae_db": 0.2571217281605007, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5134821470914535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17648733386491156, "3000–6000 Hz_10ms_envelope_correlation": 0.9788732173139654, "3000–6000 Hz_10ms_modulation_mae_db": 0.61065996657834, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1599862438795412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11383080296661452, "3000–6000 Hz_40ms_envelope_correlation": 0.9913719363763406, "3000–6000 Hz_40ms_modulation_mae_db": 0.37040632108774735, "mr_stft_db": 5.69174380869335, "band_mae_db": 0.279792460017616, "rms_gain_error_db": -0.15112536470111082, "output_peak": 0.4953390266345344}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 0.9847053505888258, "全帯域_10ms_rms_envelope_nrmse": 0.09113206754955686, "全帯域_10ms_envelope_correlation": 0.9913486577042115, "全帯域_10ms_modulation_mae_db": 0.34598039906104394, "全帯域_40ms_log_rms_mae_db": 0.5101134517054737, "全帯域_40ms_rms_envelope_nrmse": 0.046719165745030315, "全帯域_40ms_envelope_correlation": 0.9975471936183515, "全帯域_40ms_modulation_mae_db": 0.2255754883475524, "300–1000 Hz_10ms_log_rms_mae_db": 1.6539492881765463, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12105557677490565, "300–1000 Hz_10ms_envelope_correlation": 0.9879356859149405, "300–1000 Hz_10ms_modulation_mae_db": 0.15923827035144592, "300–1000 Hz_40ms_log_rms_mae_db": 0.983000589986524, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06711741483615688, "300–1000 Hz_40ms_envelope_correlation": 0.9960319900327655, "300–1000 Hz_40ms_modulation_mae_db": 0.2823266941957376, "1000–3000 Hz_10ms_log_rms_mae_db": 1.598854978041405, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15375914040773825, "1000–3000 Hz_10ms_envelope_correlation": 0.9817454058860023, "1000–3000 Hz_10ms_modulation_mae_db": 0.6179824387589513, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1332967672992769, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08723572994928498, "1000–3000 Hz_40ms_envelope_correlation": 0.9943193701576638, "1000–3000 Hz_40ms_modulation_mae_db": 0.42740455222578855, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5289706983909106, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18420521302221432, "3000–6000 Hz_10ms_envelope_correlation": 0.9778942026282372, "3000–6000 Hz_10ms_modulation_mae_db": 0.24405900116711832, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1889794938531908, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11581073283038341, "3000–6000 Hz_40ms_envelope_correlation": 0.9910709611642686, "3000–6000 Hz_40ms_modulation_mae_db": 0.32742473333070793, "mr_stft_db": 5.689743227518394, "band_mae_db": 0.22985845657197249, "rms_gain_error_db": -0.06021575948526023, "output_peak": 0.5020439805981183}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.089369766718143, "全帯域_10ms_rms_envelope_nrmse": 0.10294794252810782, "全帯域_10ms_envelope_correlation": 0.9890280569327167, "全帯域_10ms_modulation_mae_db": 0.32342786783234706, "全帯域_40ms_log_rms_mae_db": 0.5799514719187263, "全帯域_40ms_rms_envelope_nrmse": 0.05311919345644165, "全帯域_40ms_envelope_correlation": 0.9970112477485397, "全帯域_40ms_modulation_mae_db": 0.122408435975172, "300–1000 Hz_10ms_log_rms_mae_db": 1.6961835235122633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1271249610494512, "300–1000 Hz_10ms_envelope_correlation": 0.9868472493685361, "300–1000 Hz_10ms_modulation_mae_db": 0.23540348953069987, "300–1000 Hz_40ms_log_rms_mae_db": 0.9483017626022409, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06786961889900875, "300–1000 Hz_40ms_envelope_correlation": 0.9963014953891538, "300–1000 Hz_40ms_modulation_mae_db": 0.47789819874932604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6010779317218418, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1625992127540174, "1000–3000 Hz_10ms_envelope_correlation": 0.9799093576845462, "1000–3000 Hz_10ms_modulation_mae_db": 0.691680151581751, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1527503662730134, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08946232611594304, "1000–3000 Hz_40ms_envelope_correlation": 0.9946532439488426, "1000–3000 Hz_40ms_modulation_mae_db": 0.8900320376962435, "3000–6000 Hz_10ms_log_rms_mae_db": 1.493005307151695, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19771940256862564, "3000–6000 Hz_10ms_envelope_correlation": 0.9732194179283582, "3000–6000 Hz_10ms_modulation_mae_db": 1.071522780725898, "3000–6000 Hz_40ms_log_rms_mae_db": 1.175960644539261, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.129833056190086, "3000–6000 Hz_40ms_envelope_correlation": 0.9880599936592088, "3000–6000 Hz_40ms_modulation_mae_db": 1.1074364185025924, "mr_stft_db": 5.7056878603464405, "band_mae_db": 0.257097796775989, "rms_gain_error_db": -0.13281086121693858, "output_peak": 0.4936622091340569}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0543868159917478, "全帯域_10ms_rms_envelope_nrmse": 0.09864399455963944, "全帯域_10ms_envelope_correlation": 0.9898975523457212, "全帯域_10ms_modulation_mae_db": 0.12663321008125514, "全帯域_40ms_log_rms_mae_db": 0.5477292465007861, "全帯域_40ms_rms_envelope_nrmse": 0.053744293082993534, "全帯域_40ms_envelope_correlation": 0.9968188516292396, "全帯域_40ms_modulation_mae_db": 0.277547150700845, "300–1000 Hz_10ms_log_rms_mae_db": 1.68947685914717, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1402263706553144, "300–1000 Hz_10ms_envelope_correlation": 0.9839267087754735, "300–1000 Hz_10ms_modulation_mae_db": 0.42772300777928635, "300–1000 Hz_40ms_log_rms_mae_db": 0.9726678480415184, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08592658587998736, "300–1000 Hz_40ms_envelope_correlation": 0.9937701219403092, "300–1000 Hz_40ms_modulation_mae_db": 0.38169022132359776, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6702918420732722, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15814951957436946, "1000–3000 Hz_10ms_envelope_correlation": 0.9806930724374457, "1000–3000 Hz_10ms_modulation_mae_db": 0.39058518295300765, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1811940172116335, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08742423489661437, "1000–3000 Hz_40ms_envelope_correlation": 0.9943656992147917, "1000–3000 Hz_40ms_modulation_mae_db": 0.44208698567530885, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5233466876798474, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20556775386503343, "3000–6000 Hz_10ms_envelope_correlation": 0.9723105497395222, "3000–6000 Hz_10ms_modulation_mae_db": 0.5370291751366173, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1941239837872912, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1262337443352229, "3000–6000 Hz_40ms_envelope_correlation": 0.9892540706194674, "3000–6000 Hz_40ms_modulation_mae_db": 0.856611640498024, "mr_stft_db": 5.697209087571686, "band_mae_db": 0.2787252850132081, "rms_gain_error_db": -0.10138508216874853, "output_peak": 0.5069638488332249}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68080, "output_samples": 68080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.8038841264036325, "spectral_convergence": 0.2915182537797403, "spectral_convergence_status": "ok", "amplitude_floor": 2.0850976744763412e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.425850861630003, "spectral_convergence": 0.2712878402371983, "spectral_convergence_status": "ok", "amplitude_floor": 1.4381271492089952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.857157924730377, "spectral_convergence": 0.2880354921192789, "spectral_convergence_status": "ok", "amplitude_floor": 9.411979934974602e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005432476915893821, "output_power": 0.0005893271849804849, "delta_db": 0.3535858550030236, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030286578008489277, "output_power": 0.0028122715707827022, "delta_db": -0.3219294984361294, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038180407470981095, "output_power": 0.0003991057474630596, "delta_db": 0.19247423056780832, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002532984259990213, "output_power": 0.0002537835931174776, "delta_db": 0.008310508308266554, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013717204322211298, "output_power": 0.00012997953336943842, "delta_db": -0.23390634139063424, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.29065897621083e-05, "output_power": 8.738923768917916e-05, "delta_db": 0.22868899479051985, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.22314923808273032, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15501327214714983, "pitch": null, "output_rms_dbfs": -23.693833071823853, "output_peak": 0.5015192925856209, "reference_rms_dbfs": -23.538819799676702, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.695630970921338}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00026228408676881105, "log_rms_mae_db": 0.9763894606419906, "rms_envelope_nrmse": 0.08770542955899435, "envelope_correlation": 0.9921771261851612, "rms_gain_error_db": -0.15501044143108222, "reference_mean_power": 0.004432288013980724, "output_mean_power": 0.004276878663329372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 1.6030570640082068e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008005952323264084, "output_power": 0.0007421201919725397, "reference_active": true, "delta_db": -0.329387512348429}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020547999455019126, "output_power": 0.00021096159836407816, "reference_active": true, "delta_db": 0.11433861724037833}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.989786705158196e-05, "output_power": 3.773367539694732e-05, "reference_active": true, "delta_db": -0.24220569712176337}], "active_band_count": 3, "active_band_mae_db": 0.2286439422368569}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00021831174911177267, "log_rms_mae_db": 0.509395852925305, "rms_envelope_nrmse": 0.04824662816780399, "envelope_correlation": 0.9976620554291425, "rms_gain_error_db": -0.1549998345180502, "reference_mean_power": 0.004463740704660867, "output_mean_power": 0.004307239047755448}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 1.4245155353761532e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007405152195784611, "output_power": 0.0006914391225292853, "reference_active": true, "delta_db": -0.2978004001703311}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001021109367695941, "output_power": 0.00010580548085221209, "reference_active": true, "delta_db": 0.15435904818188548}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.434051823082556e-06, "output_power": 1.925206243009052e-06, "reference_active": true, "delta_db": -1.0185255917359841}], "active_band_count": 3, "active_band_mae_db": 0.4902283466960669}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00023386465365005232, "log_rms_mae_db": 1.7095065518421766, "rms_envelope_nrmse": 0.13180545544341, "envelope_correlation": 0.9867177136512864, "rms_gain_error_db": -0.3851828198977034, "reference_mean_power": 0.002112021838569365, "output_mean_power": 0.0019327697622502785}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 9.681983603336737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048709120906038954, "output_power": 0.00043007664672469793, "reference_active": true, "delta_db": -0.5406443067980767}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014674692890756196, "output_power": 0.0001326792106827337, "reference_active": true, "delta_db": -0.4376614197163323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8361558797573682e-05, "output_power": 2.7638250511217302e-05, "reference_active": true, "delta_db": -0.11219547773234537}], "active_band_count": 3, "active_band_mae_db": 0.36350040141558476}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00020000043006212289, "log_rms_mae_db": 1.0145689499388477, "rms_envelope_nrmse": 0.08064368438848296, "envelope_correlation": 0.9957468469239413, "rms_gain_error_db": -0.3851904816725344, "reference_mean_power": 0.002127032769351675, "output_mean_power": 0.0019465032473410184}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.787606965551091e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004705886462934124, "output_power": 0.0004154006265102521, "reference_active": true, "delta_db": -0.5417429770911959}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.123456963821095e-05, "output_power": 6.530393601749174e-05, "reference_active": true, "delta_db": -0.3775144688181712}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5465229502268847e-06, "output_power": 1.4343143196432075e-06, "reference_active": true, "delta_db": -0.3271203500197447}], "active_band_count": 3, "active_band_mae_db": 0.41545926530970395}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 7.720600674415526e-05, "log_rms_mae_db": 1.5969299581794176, "rms_envelope_nrmse": 0.15659626026641976, "envelope_correlation": 0.9807489424371919, "rms_gain_error_db": -0.10824551642578609, "reference_mean_power": 0.00033128809071676906, "output_mean_power": 0.0003231329699826601}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 2.2485323271997473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014154142624946028, "output_power": 0.0001415595951108051, "reference_active": true, "delta_db": 0.0005574431485580619}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6917189688068767e-05, "output_power": 2.4582045211746107e-05, "reference_active": true, "delta_db": -0.39411701936638477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.319943953798946e-06, "output_power": 6.736811837330231e-06, "reference_active": true, "delta_db": 0.2774119053905361}], "active_band_count": 3, "active_band_mae_db": 0.22402878930182632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 6.30102051319373e-05, "log_rms_mae_db": 1.1268530454389099, "rms_envelope_nrmse": 0.09293037193457143, "envelope_correlation": 0.9927781358281482, "rms_gain_error_db": -0.10824827113806737, "reference_mean_power": 0.0003336427865109939, "output_mean_power": 0.0003254294952205551}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 2.017085687613918e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013578457159768264, "output_power": 0.00013608830126632965, "reference_active": true, "delta_db": 0.009703665729610416}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.310083041117988e-05, "output_power": 1.1462615025594492e-05, "reference_active": true, "delta_db": -0.5801511800384258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7328752869332565e-07, "output_power": 3.3982892087409337e-07, "reference_active": true, "delta_db": -0.40783144259712545}], "active_band_count": 3, "active_band_mae_db": 0.33256209612172055}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 5.974138539374114e-05, "log_rms_mae_db": 1.4864727405718547, "rms_envelope_nrmse": 0.18931492405461547, "envelope_correlation": 0.9775981969459198, "rms_gain_error_db": 0.32542155621868124, "reference_mean_power": 8.138813397574021e-05, "output_mean_power": 8.772093396048808e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 8.567729341398199e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0300307346839795e-05, "output_power": 4.214931647078921e-05, "reference_active": true, "delta_db": 0.19482177865316333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.809515102528064e-06, "output_power": 6.368410677390551e-06, "reference_active": true, "delta_db": 0.39891176779885107}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.571956353556118e-06, "output_power": 3.592953808338589e-06, "reference_active": true, "delta_db": 0.025454903540157277}], "active_band_count": 3, "active_band_mae_db": 0.20639614999739053}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 4.362120322552387e-05, "log_rms_mae_db": 1.1613729863067395, "rms_envelope_nrmse": 0.11615632678678205, "envelope_correlation": 0.9917727519741469, "rms_gain_error_db": 0.32541949354933536, "reference_mean_power": 8.196654666810843e-05, "output_mean_power": 8.834431090937449e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.099518469769939e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.90961297245674e-05, "output_power": 4.051680307199469e-05, "reference_active": true, "delta_db": 0.1550140352190591}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1532747685680723e-06, "output_power": 3.630419439953311e-06, "reference_active": true, "delta_db": 0.6119498833693992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4027458558124955e-07, "output_power": 1.680483134662621e-07, "reference_active": true, "delta_db": 0.7845516392167117}], "active_band_count": 3, "active_band_mae_db": 0.5171718526017233}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8120231582979941, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "351b166002d073495730e9b01922a352cff6f36ab7a69aacf14c067649d04be7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9800926118224459, "standard_deviation": 0.0014505816525036933, "minimum": 0.9773958239444477, "maximum": 0.9817454058860023, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6121275516397673, "standard_deviation": 0.03297582615350213, "minimum": 1.5619157750373287, "maximum": 1.6702918420732722, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5190242986804041, "standard_deviation": 0.17687756931526805, "minimum": 0.22402878930182632, "maximum": 0.7825034392397653, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1602156836697171, "standard_deviation": 0.005704143227746035, "minimum": 0.15375914040773825, "maximum": 0.1711204129936486, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937781660493095, "standard_deviation": 0.0013214220901303208, "minimum": 0.9915195537978911, "maximum": 0.9955409771677429, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1495286095999422, "standard_deviation": 0.02531499130304543, "minimum": 1.106399589945543, "maximum": 1.1811940172116335, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5387719796769092, "standard_deviation": 0.2582250097070415, "minimum": 0.2571217281605007, "maximum": 0.9737417707978019, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09007583397563737, "standard_deviation": 0.007764483293429775, "minimum": 0.08049705690482138, "maximum": 0.10556946441993603, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9753540405431675, "standard_deviation": 0.004645646737330914, "minimum": 0.9660391984941997, "maximum": 0.9803801597226942, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5074981955341764, "standard_deviation": 0.018311179242772552, "minimum": 1.4864727405718547, "maximum": 1.5295747859374158, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.566422294188035, "standard_deviation": 0.3483742129330583, "minimum": 0.1978436504627676, "maximum": 1.071522780725898, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19146682546387495, "standard_deviation": 0.01662991527103185, "minimum": 0.17073216732177196, "maximum": 0.22228479305071577, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9896464640113658, "standard_deviation": 0.0019056387664830201, "minimum": 0.9863861822693444, "maximum": 0.9917727519741469, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1797730874511816, "standard_deviation": 0.01374570868549976, "minimum": 1.1599862438795412, "maximum": 1.1965606090647058, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5744931692774606, "standard_deviation": 0.30270499726453587, "minimum": 0.18783494646784052, "maximum": 1.1074364185025924, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12330262030057793, "standard_deviation": 0.009754175072784484, "minimum": 0.11383080296661452, "maximum": 0.14096212819922502, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9860501241721366, "standard_deviation": 0.0018376058674773083, "minimum": 0.9831443656276335, "maximum": 0.9884097484106877, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6883121184497367, "standard_deviation": 0.0375858890382424, "minimum": 1.628985525193065, "maximum": 1.7535252993365391, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.41203232647120014, "standard_deviation": 0.1777357482677007, "minimum": 0.15923827035144592, "maximum": 0.6671273628019477, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1313422444265103, "standard_deviation": 0.00877860023716545, "minimum": 0.11989469917415677, "maximum": 0.1455135488962878, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9952874495044587, "standard_deviation": 0.0008974663963447592, "minimum": 0.9937701219403092, "maximum": 0.9963014953891538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9756085873782288, "standard_deviation": 0.03269578058107807, "minimum": 0.9203842621616659, "maximum": 1.0212538019109467, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.4157570050133852, "standard_deviation": 0.1306853039428955, "minimum": 0.24075143549855813, "maximum": 0.6602944923112699, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07683304577594949, "standard_deviation": 0.006934983087619191, "minimum": 0.06711741483615688, "maximum": 0.08592658587998736, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.23661893263208267, "standard_deviation": 0.044707408182172156, "minimum": 0.1583572737867576, "maximum": 0.279792460017616, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.6979101961805, "standard_deviation": 0.006064311821816446, "minimum": 5.689743227518394, "maximum": 5.706332805518897, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.49888692346305996, "standard_deviation": 0.0066845374328045725, "minimum": 0.48795618533322677, "maximum": 0.507013428908619, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12172145115269693, "standard_deviation": 0.037658807139345093, "minimum": -0.17208822542235905, "maximum": -0.06021575948526023, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9907084411355864, "standard_deviation": 0.0010767705584159809, "minimum": 0.9890280569327167, "maximum": 0.9921771261851612, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.016680066005184, "standard_deviation": 0.04016926487530264, "minimum": 0.9763894606419906, "maximum": 1.089369766718143, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.2532013054149476, "standard_deviation": 0.09297512806022097, "minimum": 0.12663321008125514, "maximum": 0.3905928840290753, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.0947811906808446, "standard_deviation": 0.005400062246116251, "minimum": 0.08770542955899435, "maximum": 0.10294794252810782, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9973372576925036, "standard_deviation": 0.0003658594671238879, "minimum": 0.9968188516292396, "maximum": 0.9978441226790719, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.5328511588616173, "standard_deviation": 0.03190370109554811, "minimum": 0.49555792379704916, "maximum": 0.5799514719187263, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.24388391201190596, "standard_deviation": 0.14369207672360723, "minimum": 0.028441168442661754, "maximum": 0.4902283466960669, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.050070496579174284, "standard_deviation": 0.002944644937129587, "minimum": 0.04642039477503888, "maximum": 0.053744293082993534, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_908-31957-0002", "language": "en", "style": "normal", "speaker": "librispeech_908", "source_utterance_id": "908-31957-0002", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 76640, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_908-31957-0002.wav", "source_file_sha256": "762cc3ee94460e267478ba6714c3060aa1c1d7e9771bd290e7e74d6e44d8f2fb", "decoded_pcm_sha256": "5e0a5e522c82eea67f456aec58b62300d4c743b9610bb9f4f869f3b8bd8e7f17", "original_source_sha256": "77ad707e976d262395c265cdfda5b9a3da7439f329aac0f5216a404acefe3fe7", "corpus_entry_id": "librispeech_908-31957-0002"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 4.012834072113037, "predicted_mos": 4.012834072113037, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "8e4d04598117f3f4c4532ddfe11f136397793be0b75d1a2e03ad35405419f2f4", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.245331273999909, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_908-31957-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "8e4d04598117f3f4c4532ddfe11f136397793be0b75d1a2e03ad35405419f2f4", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "b69e3ce913f862c4a8715e3c55451386e805eade08afccb7a2d3eb6e55696791"}, "rows": [{"case": "librispeech_908-31957-0002", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.621971845626831, "predicted_mos": 2.621971845626831, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "87fa795e8d680be4c9fc8498399fb8b07ac55ea4ef1bba9f1f5fc866739eaeb3", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.324704118000227, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_908-31957-0002", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "87fa795e8d680be4c9fc8498399fb8b07ac55ea4ef1bba9f1f5fc866739eaeb3", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2469212238746687, "全帯域_10ms_rms_envelope_nrmse": 0.10062054611024285, "全帯域_10ms_envelope_correlation": 0.9902851347934329, "全帯域_10ms_modulation_mae_db": 0.32851707472797825, "全帯域_40ms_log_rms_mae_db": 0.6693581856040376, "全帯域_40ms_rms_envelope_nrmse": 0.06285643712191133, "全帯域_40ms_envelope_correlation": 0.9960835118285588, "全帯域_40ms_modulation_mae_db": 0.2400004430178778, "300–1000 Hz_10ms_log_rms_mae_db": 1.8611113933421648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12392021395339656, "300–1000 Hz_10ms_envelope_correlation": 0.9887122047423914, "300–1000 Hz_10ms_modulation_mae_db": 0.503279771550162, "300–1000 Hz_40ms_log_rms_mae_db": 1.1724680274836237, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06720320902680224, "300–1000 Hz_40ms_envelope_correlation": 0.996722819899952, "300–1000 Hz_40ms_modulation_mae_db": 0.3964786618954051, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8279790976593078, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1849491123664547, "1000–3000 Hz_10ms_envelope_correlation": 0.9765291927153296, "1000–3000 Hz_10ms_modulation_mae_db": 0.5928350601306742, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1533673374558377, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10452026981972154, "1000–3000 Hz_40ms_envelope_correlation": 0.992162938145309, "1000–3000 Hz_40ms_modulation_mae_db": 0.8738911173822883, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5685797169570928, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2416908441786205, "3000–6000 Hz_10ms_envelope_correlation": 0.9677733174996601, "3000–6000 Hz_10ms_modulation_mae_db": 1.4950750679335763, "3000–6000 Hz_40ms_log_rms_mae_db": 1.350406238048237, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.186438317852489, "3000–6000 Hz_40ms_envelope_correlation": 0.9819469575594265, "3000–6000 Hz_40ms_modulation_mae_db": 1.745269445003487, "mr_stft_db": 5.797025605893971, "band_mae_db": 0.2329155435079074, "rms_gain_error_db": -0.15091437346103323, "output_peak": 0.40263731799127794}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.29426099529805, "全帯域_10ms_rms_envelope_nrmse": 0.09930773196555436, "全帯域_10ms_envelope_correlation": 0.9906924914544969, "全帯域_10ms_modulation_mae_db": 1.3170352941965586, "全帯域_40ms_log_rms_mae_db": 0.6517908861771905, "全帯域_40ms_rms_envelope_nrmse": 0.05341074899038428, "全帯域_40ms_envelope_correlation": 0.9974748517759201, "全帯域_40ms_modulation_mae_db": 1.5233419390166525, "300–1000 Hz_10ms_log_rms_mae_db": 1.8892894858843587, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1219837697994006, "300–1000 Hz_10ms_envelope_correlation": 0.9893436305752386, "300–1000 Hz_10ms_modulation_mae_db": 1.1887656767213788, "300–1000 Hz_40ms_log_rms_mae_db": 1.271801796353332, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07056778636207077, "300–1000 Hz_40ms_envelope_correlation": 0.9967092586452816, "300–1000 Hz_40ms_modulation_mae_db": 1.1960186595128253, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8510311219077515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21240675840643036, "1000–3000 Hz_10ms_envelope_correlation": 0.9707958913202284, "1000–3000 Hz_10ms_modulation_mae_db": 0.9325559944208345, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1571456272253122, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13491099479019125, "1000–3000 Hz_40ms_envelope_correlation": 0.9903398787040767, "1000–3000 Hz_40ms_modulation_mae_db": 0.9072250982309855, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5673809863132973, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23135985643125626, "3000–6000 Hz_10ms_envelope_correlation": 0.9696503189870694, "3000–6000 Hz_10ms_modulation_mae_db": 1.7594484229562584, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3253792033888703, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13718133284828318, "3000–6000 Hz_40ms_envelope_correlation": 0.990219784442956, "3000–6000 Hz_40ms_modulation_mae_db": 2.195722348594385, "mr_stft_db": 5.807954033912329, "band_mae_db": 0.3880950191199147, "rms_gain_error_db": -0.1950666754912303, "output_peak": 0.3965476450260881}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.282801816666738, "全帯域_10ms_rms_envelope_nrmse": 0.09833541628360913, "全帯域_10ms_envelope_correlation": 0.9909393981703645, "全帯域_10ms_modulation_mae_db": 0.8671758594735981, "全帯域_40ms_log_rms_mae_db": 0.6846452707539277, "全帯域_40ms_rms_envelope_nrmse": 0.053361614631927734, "全帯域_40ms_envelope_correlation": 0.9975509963401525, "全帯域_40ms_modulation_mae_db": 0.9327412068549323, "300–1000 Hz_10ms_log_rms_mae_db": 1.7969555915804412, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12710226376642592, "300–1000 Hz_10ms_envelope_correlation": 0.9890838630632152, "300–1000 Hz_10ms_modulation_mae_db": 1.1011564057416663, "300–1000 Hz_40ms_log_rms_mae_db": 1.121339829597316, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07230479448278022, "300–1000 Hz_40ms_envelope_correlation": 0.997419862616923, "300–1000 Hz_40ms_modulation_mae_db": 0.7912973605508898, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7711040367968864, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17965346643732827, "1000–3000 Hz_10ms_envelope_correlation": 0.9778125518882937, "1000–3000 Hz_10ms_modulation_mae_db": 0.7349380305927696, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0866943013205608, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10392875820158132, "1000–3000 Hz_40ms_envelope_correlation": 0.9923998370197898, "1000–3000 Hz_40ms_modulation_mae_db": 0.6033880658838471, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5794356567827021, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2094379110742296, "3000–6000 Hz_10ms_envelope_correlation": 0.9758116319909167, "3000–6000 Hz_10ms_modulation_mae_db": 1.2340523143462958, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3738300186811467, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12167623163352591, "3000–6000 Hz_40ms_envelope_correlation": 0.9917673324199007, "3000–6000 Hz_40ms_modulation_mae_db": 1.6040977719320952, "mr_stft_db": 5.8031298957215505, "band_mae_db": 0.275662552634836, "rms_gain_error_db": -0.21103036484554016, "output_peak": 0.4154637300017902}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3189333591201353, "全帯域_10ms_rms_envelope_nrmse": 0.10604373295244149, "全帯域_10ms_envelope_correlation": 0.9890795111713535, "全帯域_10ms_modulation_mae_db": 0.08293297587127718, "全帯域_40ms_log_rms_mae_db": 0.6471567322520673, "全帯域_40ms_rms_envelope_nrmse": 0.05215032371045932, "全帯域_40ms_envelope_correlation": 0.997215863530019, "全帯域_40ms_modulation_mae_db": 0.13168065265699183, "300–1000 Hz_10ms_log_rms_mae_db": 1.8448656176776854, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13276833000778013, "300–1000 Hz_10ms_envelope_correlation": 0.9868839518184207, "300–1000 Hz_10ms_modulation_mae_db": 0.2197463995958322, "300–1000 Hz_40ms_log_rms_mae_db": 1.149910120582607, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06729196306811457, "300–1000 Hz_40ms_envelope_correlation": 0.9964616835453595, "300–1000 Hz_40ms_modulation_mae_db": 0.5216496696133605, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8336322238437326, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18618530070352923, "1000–3000 Hz_10ms_envelope_correlation": 0.9761389231726165, "1000–3000 Hz_10ms_modulation_mae_db": 1.8135981519001831, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1217664560330083, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09618479333338917, "1000–3000 Hz_40ms_envelope_correlation": 0.993563888405295, "1000–3000 Hz_40ms_modulation_mae_db": 1.571537509392111, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5755122464098505, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23329158346921677, "3000–6000 Hz_10ms_envelope_correlation": 0.969919214949262, "3000–6000 Hz_10ms_modulation_mae_db": 0.8221679650304639, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3525793310865293, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13869178672021953, "3000–6000 Hz_40ms_envelope_correlation": 0.9892410018819622, "3000–6000 Hz_40ms_modulation_mae_db": 1.0143129629715322, "mr_stft_db": 5.814451237581287, "band_mae_db": 0.17251445309685665, "rms_gain_error_db": -0.08583410174095571, "output_peak": 0.40249789932639946}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2903663261379983, "全帯域_10ms_rms_envelope_nrmse": 0.10793694472950051, "全帯域_10ms_envelope_correlation": 0.9889108884410615, "全帯域_10ms_modulation_mae_db": 0.5812184339884033, "全帯域_40ms_log_rms_mae_db": 0.7183572292006023, "全帯域_40ms_rms_envelope_nrmse": 0.06501390250331565, "全帯域_40ms_envelope_correlation": 0.9959990010602061, "全帯域_40ms_modulation_mae_db": 0.6743656061483497, "300–1000 Hz_10ms_log_rms_mae_db": 1.8870621034720536, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14780502200321866, "300–1000 Hz_10ms_envelope_correlation": 0.9839811560422721, "300–1000 Hz_10ms_modulation_mae_db": 0.8946316670649829, "300–1000 Hz_40ms_log_rms_mae_db": 1.2613436463229817, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09006100087837347, "300–1000 Hz_40ms_envelope_correlation": 0.9941355309349557, "300–1000 Hz_40ms_modulation_mae_db": 0.680373666036091, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7937953607827255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18972695620559962, "1000–3000 Hz_10ms_envelope_correlation": 0.9766394213095895, "1000–3000 Hz_10ms_modulation_mae_db": 0.9589867106302199, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1470074011216291, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1168338525132203, "1000–3000 Hz_40ms_envelope_correlation": 0.992621531605038, "1000–3000 Hz_40ms_modulation_mae_db": 0.6402812425382206, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5823042667426948, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2063992740795381, "3000–6000 Hz_10ms_envelope_correlation": 0.9766736859551552, "3000–6000 Hz_10ms_modulation_mae_db": 0.3194688552257847, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3355302562438311, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13754253808237124, "3000–6000 Hz_40ms_envelope_correlation": 0.9908370120835072, "3000–6000 Hz_40ms_modulation_mae_db": 0.23011050788682294, "mr_stft_db": 5.7905656943885395, "band_mae_db": 0.3747937907059217, "rms_gain_error_db": -0.19817860533559678, "output_peak": 0.4432509939936684}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2928787691723524, "全帯域_10ms_rms_envelope_nrmse": 0.1028877821965177, "全帯域_10ms_envelope_correlation": 0.9896920099648587, "全帯域_10ms_modulation_mae_db": 0.23810274843311416, "全帯域_40ms_log_rms_mae_db": 0.649217683295251, "全帯域_40ms_rms_envelope_nrmse": 0.051763737249485506, "全帯域_40ms_envelope_correlation": 0.9972248268538763, "全帯域_40ms_modulation_mae_db": 0.3119694920814626, "300–1000 Hz_10ms_log_rms_mae_db": 1.9053093864198256, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12460100346957077, "300–1000 Hz_10ms_envelope_correlation": 0.9884947571067655, "300–1000 Hz_10ms_modulation_mae_db": 0.6461356944457982, "300–1000 Hz_40ms_log_rms_mae_db": 1.2982873499536434, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0695491605720614, "300–1000 Hz_40ms_envelope_correlation": 0.9963170597627189, "300–1000 Hz_40ms_modulation_mae_db": 0.465571547946191, "1000–3000 Hz_10ms_log_rms_mae_db": 1.790405443705996, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17699737965580484, "1000–3000 Hz_10ms_envelope_correlation": 0.9786757754819773, "1000–3000 Hz_10ms_modulation_mae_db": 0.7099706909317455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0392589257106413, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09536161354609875, "1000–3000 Hz_40ms_envelope_correlation": 0.9940851736131903, "1000–3000 Hz_40ms_modulation_mae_db": 0.5948694507573632, "3000–6000 Hz_10ms_log_rms_mae_db": 1.587524191744661, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23224913977151349, "3000–6000 Hz_10ms_envelope_correlation": 0.9696842399866727, "3000–6000 Hz_10ms_modulation_mae_db": 0.5039777952519495, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3573894915513878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1373794816572046, "3000–6000 Hz_40ms_envelope_correlation": 0.9892001151720023, "3000–6000 Hz_40ms_modulation_mae_db": 0.5318492502944834, "mr_stft_db": 5.791554604043019, "band_mae_db": 0.17273150670888146, "rms_gain_error_db": -0.07188331847909396, "output_peak": 0.41238750183366396}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2777494952653825, "全帯域_10ms_rms_envelope_nrmse": 0.10772563070432258, "全帯域_10ms_envelope_correlation": 0.9888326474497359, "全帯域_10ms_modulation_mae_db": 0.3959083557036629, "全帯域_40ms_log_rms_mae_db": 0.6947164088387531, "全帯域_40ms_rms_envelope_nrmse": 0.060194684368604034, "全帯域_40ms_envelope_correlation": 0.9964572886553387, "全帯域_40ms_modulation_mae_db": 0.11476805366861585, "300–1000 Hz_10ms_log_rms_mae_db": 1.847374800824531, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13101413449001964, "300–1000 Hz_10ms_envelope_correlation": 0.9876768950291731, "300–1000 Hz_10ms_modulation_mae_db": 1.0064298383081751, "300–1000 Hz_40ms_log_rms_mae_db": 1.1904912595154535, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07721059872007038, "300–1000 Hz_40ms_envelope_correlation": 0.9960205865878774, "300–1000 Hz_40ms_modulation_mae_db": 1.216353169859905, "1000–3000 Hz_10ms_log_rms_mae_db": 1.841078230481998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19748029106604226, "1000–3000 Hz_10ms_envelope_correlation": 0.9734063787395061, "1000–3000 Hz_10ms_modulation_mae_db": 0.6009043125672852, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1530084653821633, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12666956445310545, "1000–3000 Hz_40ms_envelope_correlation": 0.9892551679301628, "1000–3000 Hz_40ms_modulation_mae_db": 1.0590253621358172, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6008667710137312, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1951706716126591, "3000–6000 Hz_10ms_envelope_correlation": 0.9785334677137848, "3000–6000 Hz_10ms_modulation_mae_db": 0.42897101625134587, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3922737999434396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10931425863838873, "3000–6000 Hz_40ms_envelope_correlation": 0.9931064635792066, "3000–6000 Hz_40ms_modulation_mae_db": 0.6857373742569037, "mr_stft_db": 5.7939468055996, "band_mae_db": 0.2501466233474679, "rms_gain_error_db": -0.15691408180903443, "output_peak": 0.39638642146969894}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2525426535102262, "全帯域_10ms_rms_envelope_nrmse": 0.099033734049535, "全帯域_10ms_envelope_correlation": 0.9909199629203697, "全帯域_10ms_modulation_mae_db": 0.5291888283108117, "全帯域_40ms_log_rms_mae_db": 0.6450304436317246, "全帯域_40ms_rms_envelope_nrmse": 0.055433240541220785, "全帯域_40ms_envelope_correlation": 0.9974604965305198, "全帯域_40ms_modulation_mae_db": 0.8223295015683808, "300–1000 Hz_10ms_log_rms_mae_db": 1.7902836414773833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13016926230940828, "300–1000 Hz_10ms_envelope_correlation": 0.9886517259191708, "300–1000 Hz_10ms_modulation_mae_db": 1.1976805973801197, "300–1000 Hz_40ms_log_rms_mae_db": 1.1636027320771545, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08165875690442262, "300–1000 Hz_40ms_envelope_correlation": 0.9964433622075219, "300–1000 Hz_40ms_modulation_mae_db": 1.2063207934411129, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8377304069062321, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2205566145762577, "1000–3000 Hz_10ms_envelope_correlation": 0.9684320275740903, "1000–3000 Hz_10ms_modulation_mae_db": 1.6026051689532606, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1770955302328832, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14610179065858234, "1000–3000 Hz_40ms_envelope_correlation": 0.9882919546044424, "1000–3000 Hz_40ms_modulation_mae_db": 1.6228579977778086, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6147617920199826, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20051628993367424, "3000–6000 Hz_10ms_envelope_correlation": 0.9781863656224217, "3000–6000 Hz_10ms_modulation_mae_db": 0.5734909160869576, "3000–6000 Hz_40ms_log_rms_mae_db": 1.326354247829685, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1195191866381864, "3000–6000 Hz_40ms_envelope_correlation": 0.9923550926222734, "3000–6000 Hz_40ms_modulation_mae_db": 0.4636683598424887, "mr_stft_db": 5.808847392334916, "band_mae_db": 0.3704899559545887, "rms_gain_error_db": -0.23635790614096197, "output_peak": 0.39445112735719884}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76640, "output_samples": 76640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.92878129983467, "spectral_convergence": 0.29994747517069503, "spectral_convergence_status": "ok", "amplitude_floor": 1.4500777065259659e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.609914244738574, "spectral_convergence": 0.2905199875328057, "spectral_convergence_status": "ok", "amplitude_floor": 1.1426990309647766e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.85238127310867, "spectral_convergence": 0.31285017231905343, "spectral_convergence_status": "ok", "amplitude_floor": 7.641152847432082e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003890931416418852, "output_power": 0.0004158421441394629, "delta_db": 0.28874925674215624, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013826376510160613, "output_power": 0.0012921081192472698, "delta_db": -0.29409523633010975, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006763873367355198, "output_power": 0.0006715751700058954, "delta_db": -0.03100837485586205, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00010128235718512909, "output_power": 9.997777964758817e-05, "delta_db": -0.05630312788944423, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.238121331616846e-05, "output_power": 7.137929764468667e-05, "delta_db": -0.0605358856468691, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017231931162072514, "output_power": 0.00014779298239373325, "delta_db": -0.6668013795830029, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2329155435079074, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15091437346103323, "pitch": null, "output_rms_dbfs": -25.688493347981442, "output_peak": 0.40263731799127794, "reference_rms_dbfs": -25.537578974520407, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.797025605893971}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00018226521943209611, "log_rms_mae_db": 1.2469212238746687, "rms_envelope_nrmse": 0.10062054611024285, "envelope_correlation": 0.9902851347934329, "rms_gain_error_db": -0.15090959386126773, "reference_mean_power": 0.0027970153309965396, "output_mean_power": 0.002701493254389924}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.029329446119484e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010840604762391139, "output_power": 0.0010550015571000192, "reference_active": true, "delta_db": -0.11800410148264082}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001764324474949557, "output_power": 0.00015764326162901124, "reference_active": true, "delta_db": -0.4890304694412048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.741521666729804e-05, "output_power": 4.345763361380891e-05, "reference_active": true, "delta_db": -0.37851665326008915}], "active_band_count": 3, "active_band_mae_db": 0.32851707472797825}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001676463927845064, "log_rms_mae_db": 0.6693581856040376, "rms_envelope_nrmse": 0.06285643712191133, "envelope_correlation": 0.9960835118285588, "rms_gain_error_db": -0.1509201299301743, "reference_mean_power": 0.0028146416719843875, "output_mean_power": 0.002718511035437602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.889944424347358e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010453379933036452, "output_power": 0.0010158834011362412, "reference_active": true, "delta_db": -0.12412871023733228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.750212405766423e-05, "output_power": 6.149184621627046e-05, "reference_active": true, "delta_db": -0.40499906394175367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9228813706071663e-06, "output_power": 2.79720202360761e-06, "reference_active": true, "delta_db": -0.19087355487454738}], "active_band_count": 3, "active_band_mae_db": 0.2400004430178778}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00017586243391937327, "log_rms_mae_db": 1.8611113933421648, "rms_envelope_nrmse": 0.12392021395339656, "envelope_correlation": 0.9887122047423914, "rms_gain_error_db": -0.1813813928735017, "reference_mean_power": 0.0014437040494073058, "output_mean_power": 0.001384650080685334}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 1.5100208608044745e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009242573938618176, "output_power": 0.0008886176071782743, "reference_active": true, "delta_db": -0.17078019367928515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001371496531392223, "output_power": 0.00011476590547378966, "reference_active": true, "delta_db": -0.7738182607562267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2803310723594984e-05, "output_power": 2.880007320915127e-05, "reference_active": true, "delta_db": -0.5652408602149741}], "active_band_count": 3, "active_band_mae_db": 0.503279771550162}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001539617967365963, "log_rms_mae_db": 1.1724680274836237, "rms_envelope_nrmse": 0.06720320902680224, "envelope_correlation": 0.996722819899952, "rms_gain_error_db": -0.18138303935794908, "reference_mean_power": 0.0014528118102568712, "output_mean_power": 0.0013933847650238753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.4301118553576922e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009152950896245062, "output_power": 0.000870447690654128, "reference_active": true, "delta_db": -0.21818455050637686}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.331274162853222e-05, "output_power": 4.79432347549232e-05, "reference_active": true, "delta_db": -0.46103684102157283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0543911334413026e-06, "output_power": 2.3104926506865845e-06, "reference_active": true, "delta_db": 0.5102145941582658}], "active_band_count": 3, "active_band_mae_db": 0.3964786618954051}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 8.12616340215947e-05, "log_rms_mae_db": 1.8279790976593078, "rms_envelope_nrmse": 0.1849491123664547, "envelope_correlation": 0.9765291927153296, "rms_gain_error_db": -0.0578440532896131, "reference_mean_power": 0.00015412407568599862, "output_mean_power": 0.00015208489410940086}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.103412885748503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011785014167056138, "output_power": 0.00011029730555393611, "reference_active": true, "delta_db": -0.2876520572005668}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.031054512020625e-06, "output_power": 9.56078720162676e-06, "reference_active": true, "delta_db": 0.7572107827270171}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6793453202028428e-06, "output_power": 3.1724363991257496e-06, "reference_active": true, "delta_db": 0.7336423404644388}], "active_band_count": 3, "active_band_mae_db": 0.5928350601306742}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 6.921465297386044e-05, "log_rms_mae_db": 1.1533673374558377, "rms_envelope_nrmse": 0.10452026981972154, "envelope_correlation": 0.992162938145309, "rms_gain_error_db": -0.057850265740672514, "reference_mean_power": 0.00015509566252518478, "output_mean_power": 0.00015304340717376858}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.0373126402115645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011306875179582822, "output_power": 0.00010685082896820561, "reference_active": true, "delta_db": -0.24564702023236992}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.441071013913094e-06, "output_power": 4.6182858157905225e-06, "reference_active": true, "delta_db": 1.2778717165004698}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5029543761566567e-07, "output_power": 1.9353577630377804e-07, "reference_active": true, "delta_db": 1.0981546154140251}], "active_band_count": 3, "active_band_mae_db": 0.8738911173822883}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00013414824565394521, "log_rms_mae_db": 1.5685797169570928, "rms_envelope_nrmse": 0.2416908441786205, "envelope_correlation": 0.9677733174996601, "rms_gain_error_db": -0.6760794053724044, "reference_mean_power": 0.0001730034207748911, "output_mean_power": 0.0001480630699824194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.1522167661145593e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001202536722048893, "output_power": 9.353201291179168e-05, "reference_active": true, "delta_db": -1.0913806642485722}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5558902161138425e-05, "output_power": 2.0072592438621676e-05, "reference_active": true, "delta_db": -1.049387288994159}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.530968744127951e-05, "output_power": 8.923199427551676e-06, "reference_active": true, "delta_db": -2.3444572505579977}], "active_band_count": 3, "active_band_mae_db": 1.4950750679335763}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 8.410768197622129e-05, "log_rms_mae_db": 1.350406238048237, "rms_envelope_nrmse": 0.186438317852489, "envelope_correlation": 0.9819469575594265, "rms_gain_error_db": -0.6760831340161735, "reference_mean_power": 0.00017409446216015204, "output_mean_power": 0.00014899669779975248}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.085538754133615e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012977220791713187, "output_power": 9.902652956391587e-05, "reference_active": true, "delta_db": -1.1743013445114159}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3304937813307329e-05, "output_power": 9.054313963327331e-06, "reference_active": true, "delta_db": -1.671572991980177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.046486321108266e-06, "output_power": 6.035873873048433e-07, "reference_active": true, "delta_db": -2.3899339985188672}], "active_band_count": 3, "active_band_mae_db": 1.745269445003487}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8511216325087698, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8d53cf49e10ce738b9dd088907463eea0627b5a98927f18260bc695bb4280edb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9748037702752039, "standard_deviation": 0.003601377188241762, "minimum": 0.9684320275740903, "maximum": 0.9786757754819773, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8183444902605788, "standard_deviation": 0.029038161046843935, "minimum": 1.7711040367968864, "maximum": 1.8510311219077515, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9932992650158716, "standard_deviation": 0.4645092696528985, "minimum": 0.5928350601306742, "maximum": 1.8135981519001831, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19349448492718085, "standard_deviation": 0.015630966124651802, "minimum": 0.17699737965580484, "maximum": 0.2205566145762577, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.991590046253413, "standard_deviation": 0.0020723218802861207, "minimum": 0.9882919546044424, "maximum": 0.9940851736131903, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1294180055602545, "standard_deviation": 0.04550346185360242, "minimum": 1.0392589257106413, "maximum": 1.1770955302328832, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9841344805123051, "standard_deviation": 0.4125560552562953, "minimum": 0.5948694507573632, "maximum": 1.6228579977778086, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11556395466448627, "standard_deviation": 0.0187952951954643, "minimum": 0.09536161354609875, "maximum": 0.14610179065858234, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9732790303381178, "standard_deviation": 0.004429126109617097, "minimum": 0.9677733174996601, "maximum": 0.9785334677137848, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5845457034980015, "standard_deviation": 0.01625924285724497, "minimum": 1.5673809863132973, "maximum": 1.6147617920199826, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8920815441353289, "standard_deviation": 0.5388134542220542, "minimum": 0.3194688552257847, "maximum": 1.7594484229562584, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2187644463188385, "standard_deviation": 0.017755006966017815, "minimum": 0.1951706716126591, "maximum": 0.2416908441786205, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9898342199701544, "standard_deviation": 0.0034829473935717313, "minimum": 0.9819469575594265, "maximum": 0.9931064635792066, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.351717823346641, "standard_deviation": 0.02319925456015837, "minimum": 1.3253792033888703, "maximum": 1.3922737999434396, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0588460025977748, "standard_deviation": 0.7092170959245372, "minimum": 0.23011050788682294, "maximum": 2.195722348594385, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13596789175883356, "standard_deviation": 0.023132709241620135, "minimum": 0.10931425863838873, "maximum": 0.186438317852489, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9878535230370808, "standard_deviation": 0.0017519903882252021, "minimum": 0.9839811560422721, "maximum": 0.9893436305752386, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8527815025848056, "standard_deviation": 0.04217474996498554, "minimum": 1.7902836414773833, "maximum": 1.9053093864198256, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8447282563510143, "standard_deviation": 0.35535870090280425, "minimum": 0.2197463995958322, "maximum": 1.1976805973801197, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12992049997490257, "standard_deviation": 0.008145628098599159, "minimum": 0.1219837697994006, "maximum": 0.14780502200321866, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962787705250737, "standard_deviation": 0.0009566862135045006, "minimum": 0.9941355309349557, "maximum": 0.997419862616923, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.203655595235764, "standard_deviation": 0.06474305505918106, "minimum": 1.121339829597316, "maximum": 1.2982873499536434, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8092579411069726, "standard_deviation": 0.3507397517320106, "minimum": 0.3964786618954051, "maximum": 1.216353169859905, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07448090875183697, "standard_deviation": 0.008024319959481207, "minimum": 0.06720320902680224, "maximum": 0.09006100087837347, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2796686806345468, "standard_deviation": 0.08867646321624525, "minimum": 0.17251445309685665, "maximum": 0.3880950191199147, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.800934408684402, "standard_deviation": 0.008936374966048615, "minimum": 5.7905656943885395, "maximum": 5.814451237581287, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40795282962497326, "standard_deviation": 0.01615198915585207, "minimum": 0.39445112735719884, "maximum": 0.4432509939936684, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.16327242841293083, "standard_deviation": 0.05903199716818469, "minimum": -0.23635790614096197, "maximum": -0.07188331847909396, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899190055457092, "standard_deviation": 0.0009046072280388211, "minimum": 0.9888326474497359, "maximum": 0.9909393981703645, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.282056829880694, "standard_deviation": 0.02335957348375733, "minimum": 1.2469212238746687, "maximum": 1.3189333591201353, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5425099463381755, "standard_deviation": 0.39245254093882215, "minimum": 0.08293297587127718, "maximum": 1.3170352941965586, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10273643987396545, "standard_deviation": 0.004003756295305685, "minimum": 0.09833541628360913, "maximum": 0.10793694472950051, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.996933354571824, "standard_deviation": 0.00064798969933424, "minimum": 0.9959990010602061, "maximum": 0.9975509963401525, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6700341049691942, "standard_deviation": 0.026920462738243372, "minimum": 0.6450304436317246, "maximum": 0.7183572292006023, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5938996118766579, "standard_deviation": 0.4906663526274652, "minimum": 0.11476805366861585, "maximum": 1.5233419390166525, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05677308613966358, "standard_deviation": 0.005180012298861754, "minimum": 0.051763737249485506, "maximum": 0.06501390250331565, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0002", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.7752702236175537, "predicted_mos": 2.7752702236175537, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "030eba93591f0efcc84d59cd3cfc4954d6dfc7e85ff53eb40134db6c029ee147", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.9409353001974523, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2451510428123511, "全帯域_10ms_rms_envelope_nrmse": 0.09962431358281418, "全帯域_10ms_envelope_correlation": 0.9905082733998363, "全帯域_10ms_modulation_mae_db": 0.37498460289735697, "全帯域_40ms_log_rms_mae_db": 0.6451891911715452, "全帯域_40ms_rms_envelope_nrmse": 0.060347550880144804, "全帯域_40ms_envelope_correlation": 0.9964567633482254, "全帯域_40ms_modulation_mae_db": 0.2977590804282411, "300–1000 Hz_10ms_log_rms_mae_db": 1.7952323857449917, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12302566585631114, "300–1000 Hz_10ms_envelope_correlation": 0.9889170759762244, "300–1000 Hz_10ms_modulation_mae_db": 0.547228276847842, "300–1000 Hz_40ms_log_rms_mae_db": 1.0511978006877722, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0683876169681865, "300–1000 Hz_40ms_envelope_correlation": 0.9966467671451339, "300–1000 Hz_40ms_modulation_mae_db": 0.4189427091285663, "1000–3000 Hz_10ms_log_rms_mae_db": 1.729155068444728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1826277811504006, "1000–3000 Hz_10ms_envelope_correlation": 0.9770804591269837, "1000–3000 Hz_10ms_modulation_mae_db": 0.581692439716043, "1000–3000 Hz_40ms_log_rms_mae_db": 1.012064528586672, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1008536437849661, "1000–3000 Hz_40ms_envelope_correlation": 0.9929403022913854, "1000–3000 Hz_40ms_modulation_mae_db": 0.9439164273357217, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5161630152210945, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24255017379173155, "3000–6000 Hz_10ms_envelope_correlation": 0.9679792360850971, "3000–6000 Hz_10ms_modulation_mae_db": 1.4923215444384101, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2583066928530042, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17845566295445633, "3000–6000 Hz_40ms_envelope_correlation": 0.9843884960667145, "3000–6000 Hz_40ms_modulation_mae_db": 1.7216807710870787, "mr_stft_db": 5.729606522516807, "band_mae_db": 0.3106352504089467, "rms_gain_error_db": -0.1622318736972802, "output_peak": 0.40391158292416174}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3040194337037505, "全帯域_10ms_rms_envelope_nrmse": 0.09968812319957181, "全帯域_10ms_envelope_correlation": 0.9906896189311415, "全帯域_10ms_modulation_mae_db": 1.207342717246859, "全帯域_40ms_log_rms_mae_db": 0.6702441495331236, "全帯域_40ms_rms_envelope_nrmse": 0.05482818577545102, "全帯域_40ms_envelope_correlation": 0.9974057541912764, "全帯域_40ms_modulation_mae_db": 1.3834198916866658, "300–1000 Hz_10ms_log_rms_mae_db": 1.8481334343907223, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12405635350788448, "300–1000 Hz_10ms_envelope_correlation": 0.9890326841452808, "300–1000 Hz_10ms_modulation_mae_db": 1.1836601880280833, "300–1000 Hz_40ms_log_rms_mae_db": 1.150210514135623, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07318889305090243, "300–1000 Hz_40ms_envelope_correlation": 0.9965152141831112, "300–1000 Hz_40ms_modulation_mae_db": 1.131637191570908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7720090996060522, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20593710361417927, "1000–3000 Hz_10ms_envelope_correlation": 0.9729981880558484, "1000–3000 Hz_10ms_modulation_mae_db": 0.9427733128067514, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0528672406263193, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12879837839427422, "1000–3000 Hz_40ms_envelope_correlation": 0.9918571745486617, "1000–3000 Hz_40ms_modulation_mae_db": 0.9338307885201242, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5185706702515025, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24479967050792495, "3000–6000 Hz_10ms_envelope_correlation": 0.9659612317453679, "3000–6000 Hz_10ms_modulation_mae_db": 1.837588736798996, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2787623954327199, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14756379830516367, "3000–6000 Hz_40ms_envelope_correlation": 0.9887127568245121, "3000–6000 Hz_40ms_modulation_mae_db": 2.2922149576083286, "mr_stft_db": 5.741764230883064, "band_mae_db": 0.4411084876386599, "rms_gain_error_db": -0.2138690920637546, "output_peak": 0.40047334544788016}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.313629941292072, "全帯域_10ms_rms_envelope_nrmse": 0.10037659804899447, "全帯域_10ms_envelope_correlation": 0.9905682300566043, "全帯域_10ms_modulation_mae_db": 0.863298794951889, "全帯域_40ms_log_rms_mae_db": 0.6891613643214064, "全帯域_40ms_rms_envelope_nrmse": 0.052861311119767426, "全帯域_40ms_envelope_correlation": 0.9976524440204418, "全帯域_40ms_modulation_mae_db": 0.8296757794972364, "300–1000 Hz_10ms_log_rms_mae_db": 1.7293101974913687, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12980819547954653, "300–1000 Hz_10ms_envelope_correlation": 0.9885188403704507, "300–1000 Hz_10ms_modulation_mae_db": 1.1013723773658943, "300–1000 Hz_40ms_log_rms_mae_db": 0.9721871042264355, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07191548597185467, "300–1000 Hz_40ms_envelope_correlation": 0.9974369345312964, "300–1000 Hz_40ms_modulation_mae_db": 0.7553260148437317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6869150845494545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17533198729122199, "1000–3000 Hz_10ms_envelope_correlation": 0.9788960665822781, "1000–3000 Hz_10ms_modulation_mae_db": 0.7756325785028664, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9425895493434703, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09909893391245916, "1000–3000 Hz_40ms_envelope_correlation": 0.9931699286442163, "1000–3000 Hz_40ms_modulation_mae_db": 0.6660588312964291, "3000–6000 Hz_10ms_log_rms_mae_db": 1.537831142819255, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21171613819792448, "3000–6000 Hz_10ms_envelope_correlation": 0.9747778112291092, "3000–6000 Hz_10ms_modulation_mae_db": 1.4835274073942866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2983460795972768, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11931885492415266, "3000–6000 Hz_40ms_envelope_correlation": 0.9918009125696633, "3000–6000 Hz_40ms_modulation_mae_db": 1.8818674828780655, "mr_stft_db": 5.740398015850879, "band_mae_db": 0.27733270822699946, "rms_gain_error_db": -0.2189256236727774, "output_peak": 0.41684649304212723}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3298459379388803, "全帯域_10ms_rms_envelope_nrmse": 0.11012575896475331, "全帯域_10ms_envelope_correlation": 0.9882253144724215, "全帯域_10ms_modulation_mae_db": 0.1408433173412649, "全帯域_40ms_log_rms_mae_db": 0.6708407062563465, "全帯域_40ms_rms_envelope_nrmse": 0.054248994540569806, "全帯域_40ms_envelope_correlation": 0.9969394119714693, "全帯域_40ms_modulation_mae_db": 0.3046452395797127, "300–1000 Hz_10ms_log_rms_mae_db": 1.7841778185947048, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13606276192791691, "300–1000 Hz_10ms_envelope_correlation": 0.9862706710433213, "300–1000 Hz_10ms_modulation_mae_db": 0.11779812681954094, "300–1000 Hz_40ms_log_rms_mae_db": 1.0471662059339606, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07018155131558168, "300–1000 Hz_40ms_envelope_correlation": 0.9961247108072451, "300–1000 Hz_40ms_modulation_mae_db": 0.5461042098583982, "1000–3000 Hz_10ms_log_rms_mae_db": 1.738770346188995, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18569616354276208, "1000–3000 Hz_10ms_envelope_correlation": 0.9764029974971101, "1000–3000 Hz_10ms_modulation_mae_db": 1.5519926335852314, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0017174003679559, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09664975371619336, "1000–3000 Hz_40ms_envelope_correlation": 0.9938064220773479, "1000–3000 Hz_40ms_modulation_mae_db": 1.3627048007765554, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5116383607296375, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23744052305436625, "3000–6000 Hz_10ms_envelope_correlation": 0.9691107389818489, "3000–6000 Hz_10ms_modulation_mae_db": 0.6709629996891571, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2767595942924277, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1404637281620456, "3000–6000 Hz_40ms_envelope_correlation": 0.9890962071734678, "3000–6000 Hz_40ms_modulation_mae_db": 0.7629971771751429, "mr_stft_db": 5.7397163295739695, "band_mae_db": 0.21016692861292866, "rms_gain_error_db": -0.05684210813915617, "output_peak": 0.4052856517043971}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2782762255142313, "全帯域_10ms_rms_envelope_nrmse": 0.11029749025390718, "全帯域_10ms_envelope_correlation": 0.9883058785148229, "全帯域_10ms_modulation_mae_db": 0.5185476511593151, "全帯域_40ms_log_rms_mae_db": 0.6908845406653324, "全帯域_40ms_rms_envelope_nrmse": 0.06528547417969717, "全帯域_40ms_envelope_correlation": 0.9958370413169814, "全帯域_40ms_modulation_mae_db": 0.44014254864188707, "300–1000 Hz_10ms_log_rms_mae_db": 1.8337194763319191, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14805174413104566, "300–1000 Hz_10ms_envelope_correlation": 0.9838656535371558, "300–1000 Hz_10ms_modulation_mae_db": 0.7737739765799322, "300–1000 Hz_40ms_log_rms_mae_db": 1.120360949598118, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08950022875442794, "300–1000 Hz_40ms_envelope_correlation": 0.9941344141671492, "300–1000 Hz_40ms_modulation_mae_db": 0.6272191410716844, "1000–3000 Hz_10ms_log_rms_mae_db": 1.730736108852428, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1843147140539549, "1000–3000 Hz_10ms_envelope_correlation": 0.9780937668019416, "1000–3000 Hz_10ms_modulation_mae_db": 0.7613471694133808, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0332445302506743, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1099130163326098, "1000–3000 Hz_40ms_envelope_correlation": 0.9937831349935556, "1000–3000 Hz_40ms_modulation_mae_db": 0.5155906991426217, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5168030314051213, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21028764429935481, "3000–6000 Hz_10ms_envelope_correlation": 0.9748418680027753, "3000–6000 Hz_10ms_modulation_mae_db": 0.2913821920142391, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1987902409008206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13295524291274957, "3000–6000 Hz_40ms_envelope_correlation": 0.9902706365127093, "3000–6000 Hz_40ms_modulation_mae_db": 0.3205423023577783, "mr_stft_db": 5.723472121385558, "band_mae_db": 0.35934712023266796, "rms_gain_error_db": -0.1720414153732795, "output_peak": 0.44289215446890096}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2985641958610044, "全帯域_10ms_rms_envelope_nrmse": 0.10283741989037032, "全帯域_10ms_envelope_correlation": 0.9897016799347722, "全帯域_10ms_modulation_mae_db": 0.2292259511057316, "全帯域_40ms_log_rms_mae_db": 0.649608602269151, "全帯域_40ms_rms_envelope_nrmse": 0.051491324149016866, "全帯域_40ms_envelope_correlation": 0.9972350202234281, "全帯域_40ms_modulation_mae_db": 0.3381686652247888, "300–1000 Hz_10ms_log_rms_mae_db": 1.8364612714960535, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1272671690110093, "300–1000 Hz_10ms_envelope_correlation": 0.9879539156544856, "300–1000 Hz_10ms_modulation_mae_db": 0.5953308264248532, "300–1000 Hz_40ms_log_rms_mae_db": 1.1771251523240733, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0696277216151098, "300–1000 Hz_40ms_envelope_correlation": 0.9962198760248492, "300–1000 Hz_40ms_modulation_mae_db": 0.3728037272214673, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7167904681667956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1752470796025878, "1000–3000 Hz_10ms_envelope_correlation": 0.9792728606780827, "1000–3000 Hz_10ms_modulation_mae_db": 0.9746694231657141, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9409257219085755, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09765949765626215, "1000–3000 Hz_40ms_envelope_correlation": 0.9940499381838059, "1000–3000 Hz_40ms_modulation_mae_db": 0.8841399893852238, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5356754819765224, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2418192989460181, "3000–6000 Hz_10ms_envelope_correlation": 0.9672825230955625, "3000–6000 Hz_10ms_modulation_mae_db": 0.40667112984671866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2601114230785986, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1366000064108104, "3000–6000 Hz_40ms_envelope_correlation": 0.989355716009955, "3000–6000 Hz_40ms_modulation_mae_db": 0.43166816918822565, "mr_stft_db": 5.728737953196908, "band_mae_db": 0.20326401804870783, "rms_gain_error_db": -0.05824249705737769, "output_peak": 0.4094186093133776}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3038446292898402, "全帯域_10ms_rms_envelope_nrmse": 0.10853942396449016, "全帯域_10ms_envelope_correlation": 0.9886574045225351, "全帯域_10ms_modulation_mae_db": 0.42156208993418876, "全帯域_40ms_log_rms_mae_db": 0.6952819341236366, "全帯域_40ms_rms_envelope_nrmse": 0.05947268338219202, "全帯域_40ms_envelope_correlation": 0.9965407033433057, "全帯域_40ms_modulation_mae_db": 0.16400787739297193, "300–1000 Hz_10ms_log_rms_mae_db": 1.7852265806929304, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13352685975936265, "300–1000 Hz_10ms_envelope_correlation": 0.9870197467476675, "300–1000 Hz_10ms_modulation_mae_db": 0.9862219500608275, "300–1000 Hz_40ms_log_rms_mae_db": 1.0486210322477403, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07703980840053598, "300–1000 Hz_40ms_envelope_correlation": 0.9958484086867391, "300–1000 Hz_40ms_modulation_mae_db": 1.2148162212457538, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7585721524474114, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19575775253117525, "1000–3000 Hz_10ms_envelope_correlation": 0.9741777918496622, "1000–3000 Hz_10ms_modulation_mae_db": 0.7892607586919264, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0160246416659109, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1244546925124832, "1000–3000 Hz_40ms_envelope_correlation": 0.9901129723252367, "1000–3000 Hz_40ms_modulation_mae_db": 1.3114662869775575, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5162166495404101, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19188443149352225, "3000–6000 Hz_10ms_envelope_correlation": 0.9789635225592764, "3000–6000 Hz_10ms_modulation_mae_db": 0.45375076216440213, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2666880696938718, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10307132509779006, "3000–6000 Hz_40ms_envelope_correlation": 0.9939710020151451, "3000–6000 Hz_40ms_modulation_mae_db": 0.6073621797397072, "mr_stft_db": 5.732136729973623, "band_mae_db": 0.299899927925181, "rms_gain_error_db": -0.15394549591868475, "output_peak": 0.4008625320783959}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2455331991585536, "全帯域_10ms_rms_envelope_nrmse": 0.10296809675716154, "全帯域_10ms_envelope_correlation": 0.9901803095118036, "全帯域_10ms_modulation_mae_db": 0.38432958812201273, "全帯域_40ms_log_rms_mae_db": 0.650528316417581, "全帯域_40ms_rms_envelope_nrmse": 0.05946987362727865, "全帯域_40ms_envelope_correlation": 0.9970260963457916, "全帯域_40ms_modulation_mae_db": 0.7381887708727354, "300–1000 Hz_10ms_log_rms_mae_db": 1.7597484254236126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13337886773306915, "300–1000 Hz_10ms_envelope_correlation": 0.9879423319731362, "300–1000 Hz_10ms_modulation_mae_db": 1.0204727977677441, "300–1000 Hz_40ms_log_rms_mae_db": 1.0508901734844194, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08054794877246817, "300–1000 Hz_40ms_envelope_correlation": 0.9965354007135272, "300–1000 Hz_40ms_modulation_mae_db": 1.0805213805544645, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7268210189904794, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2098069434197636, "1000–3000 Hz_10ms_envelope_correlation": 0.9729655245909931, "1000–3000 Hz_10ms_modulation_mae_db": 1.4199930548766566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0048091985077359, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13969342230488313, "1000–3000 Hz_40ms_envelope_correlation": 0.9910799176390668, "1000–3000 Hz_40ms_modulation_mae_db": 1.4877059199937879, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5605961674553808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20432270541164232, "3000–6000 Hz_10ms_envelope_correlation": 0.9769806002276866, "3000–6000 Hz_10ms_modulation_mae_db": 0.4065561628340819, "3000–6000 Hz_40ms_log_rms_mae_db": 1.248159024683173, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13223352306213904, "3000–6000 Hz_40ms_envelope_correlation": 0.9902310449095358, "3000–6000 Hz_40ms_modulation_mae_db": 0.16592279342948282, "mr_stft_db": 5.748621749202624, "band_mae_db": 0.40181038198883906, "rms_gain_error_db": -0.24783209280602098, "output_peak": 0.3946712361673668}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76640, "output_samples": 76640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.857121000785762, "spectral_convergence": 0.29795635037319473, "spectral_convergence_status": "ok", "amplitude_floor": 1.4500777065259659e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.539973603866942, "spectral_convergence": 0.28591021820072166, "spectral_convergence_status": "ok", "amplitude_floor": 1.1426990309647766e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7917249628977165, "spectral_convergence": 0.31070531911911264, "spectral_convergence_status": "ok", "amplitude_floor": 7.641152847432082e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003890931416418852, "output_power": 0.00042085058922207446, "delta_db": 0.3407436378610861, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013826376510160613, "output_power": 0.0012914880589080541, "delta_db": -0.2961798406596531, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006763873367355198, "output_power": 0.0006692140932915767, "delta_db": -0.04630389781936044, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00010128235718512909, "output_power": 9.59687368902052e-05, "delta_db": -0.23404021505020872, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.238121331616846e-05, "output_power": 6.921200775622958e-05, "delta_db": -0.19444411072335366, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017231931162072514, "output_power": 0.0001449185470001777, "delta_db": -0.7520998003400181, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3106352504089467, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.1622318736972802, "pitch": null, "output_rms_dbfs": -25.699810848217687, "output_peak": 0.40391158292416174, "reference_rms_dbfs": -25.537578974520407, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.729606522516807}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00018226521943209611, "log_rms_mae_db": 1.2451510428123511, "rms_envelope_nrmse": 0.09962431358281418, "envelope_correlation": 0.9905082733998363, "rms_gain_error_db": -0.16222760081045895, "reference_mean_power": 0.0027970153309965396, "output_mean_power": 0.002694462146615524}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.029329446119484e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010840604762391139, "output_power": 0.0010492714241853234, "reference_active": true, "delta_db": -0.1416566530164693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001764324474949557, "output_power": 0.00015312013858757902, "reference_active": true, "delta_db": -0.615461452469803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.741521666729804e-05, "output_power": 4.35646439686141e-05, "reference_active": true, "delta_db": -0.3678357032057986}], "active_band_count": 3, "active_band_mae_db": 0.37498460289735697}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001676463927845064, "log_rms_mae_db": 0.6451891911715452, "rms_envelope_nrmse": 0.060347550880144804, "envelope_correlation": 0.9964567633482254, "rms_gain_error_db": -0.1622386885289968, "reference_mean_power": 0.0028146416719843875, "output_mean_power": 0.0027114352915053415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.889944424347358e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010453379933036452, "output_power": 0.0010096666254521206, "reference_active": true, "delta_db": -0.1507873443569248}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.750212405766423e-05, "output_power": 5.975364526860574e-05, "reference_active": true, "delta_db": -0.529530342418484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9228813706071663e-06, "output_power": 2.783012992164468e-06, "reference_active": true, "delta_db": -0.2129595545093145}], "active_band_count": 3, "active_band_mae_db": 0.2977590804282411}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00017586243391937327, "log_rms_mae_db": 1.7952323857449917, "rms_envelope_nrmse": 0.12302566585631114, "envelope_correlation": 0.9889170759762244, "rms_gain_error_db": -0.1945264150364007, "reference_mean_power": 0.0014437040494073058, "output_mean_power": 0.0013804654227698536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 1.5100208608044745e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009242573938618176, "output_power": 0.0008836174751360234, "reference_active": true, "delta_db": -0.1952863727784718}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001371496531392223, "output_power": 0.00010982684060943631, "reference_active": true, "delta_db": -0.9648622315747584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2803310723594984e-05, "output_power": 2.9360541098886163e-05, "reference_active": true, "delta_db": -0.48153622619029607}], "active_band_count": 3, "active_band_mae_db": 0.547228276847842}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001539617967365963, "log_rms_mae_db": 1.0511978006877722, "rms_envelope_nrmse": 0.0683876169681865, "envelope_correlation": 0.9966467671451339, "rms_gain_error_db": -0.19452764700985434, "reference_mean_power": 0.0014528118102568712, "output_mean_power": 0.0013891738419337157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.4301118553576922e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009152950896245062, "output_power": 0.0008647227706889807, "reference_active": true, "delta_db": -0.2468423709029667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.331274162853222e-05, "output_power": 4.567704631309863e-05, "reference_active": true, "delta_db": -0.6713300418579626}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0543911334413026e-06, "output_power": 2.221000741192041e-06, "reference_active": true, "delta_db": 0.3386557146247695}], "active_band_count": 3, "active_band_mae_db": 0.4189427091285663}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 8.12616340215947e-05, "log_rms_mae_db": 1.729155068444728, "rms_envelope_nrmse": 0.1826277811504006, "envelope_correlation": 0.9770804591269837, "rms_gain_error_db": -0.20535863287887862, "reference_mean_power": 0.00015412407568599862, "output_mean_power": 0.00014700585142767196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.103412885748503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011785014167056138, "output_power": 0.00010797109709155744, "reference_active": true, "delta_db": -0.38022594686640543}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.031054512020625e-06, "output_power": 9.047211258784744e-06, "reference_active": true, "delta_db": 0.5174215776121212}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6793453202028428e-06, "output_power": 3.2566543729172433e-06, "reference_active": true, "delta_db": 0.8474297946696024}], "active_band_count": 3, "active_band_mae_db": 0.581692439716043}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 6.921465297386044e-05, "log_rms_mae_db": 1.012064528586672, "rms_envelope_nrmse": 0.1008536437849661, "envelope_correlation": 0.9929403022913854, "rms_gain_error_db": -0.2053609244713217, "reference_mean_power": 0.00015509566252518478, "output_mean_power": 0.00014793248744671318}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.0373126402115645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011306875179582822, "output_power": 0.00010465506637852142, "reference_active": true, "delta_db": -0.3358234038161333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.441071013913094e-06, "output_power": 4.33629833682538e-06, "reference_active": true, "delta_db": 1.0042551865288682}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5029543761566567e-07, "output_power": 2.1189117363056028e-07, "reference_active": true, "delta_db": 1.4916706916621636}], "active_band_count": 3, "active_band_mae_db": 0.9439164273357217}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00013414824565394521, "log_rms_mae_db": 1.5161630152210945, "rms_envelope_nrmse": 0.24255017379173155, "envelope_correlation": 0.9679792360850971, "rms_gain_error_db": -0.7561410075345418, "reference_mean_power": 0.0001730034207748911, "output_mean_power": 0.00014535855255078716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.1522167661145593e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001202536722048893, "output_power": 9.344291129232702e-05, "reference_active": true, "delta_db": -1.0955198657758454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5558902161138425e-05, "output_power": 2.08891494061986e-05, "reference_active": true, "delta_db": -0.8762143938537169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.530968744127951e-05, "output_power": 8.598907693711286e-06, "reference_active": true, "delta_db": -2.505230373685668}], "active_band_count": 3, "active_band_mae_db": 1.4923215444384101}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 8.410768197622129e-05, "log_rms_mae_db": 1.2583066928530042, "rms_envelope_nrmse": 0.17845566295445633, "envelope_correlation": 0.9843884960667145, "rms_gain_error_db": -0.7561432590879158, "reference_mean_power": 0.00017409446216015204, "output_mean_power": 0.00014627517648885973}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.085538754133615e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012977220791713187, "output_power": 9.915334942902119e-05, "reference_active": true, "delta_db": -1.1687430430926347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3304937813307329e-05, "output_power": 9.507911842332391e-06, "reference_active": true, "delta_db": -1.4592770273663267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.046486321108266e-06, "output_power": 5.834871977424089e-07, "reference_active": true, "delta_db": -2.5370222428022746}], "active_band_count": 3, "active_band_mae_db": 1.7216807710870787}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8531241691339807, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "46963729c7390cbf27765bb05d7f3c97409c359926bb8828347becc742f7a445", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9762359568978625, "standard_deviation": 0.002560871231037422, "minimum": 0.9729655245909931, "maximum": 0.9792728606780827, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7324711684057932, "standard_deviation": 0.02577751314508038, "minimum": 1.6869150845494545, "maximum": 1.7720090996060522, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9746701713448213, "standard_deviation": 0.33944532906395325, "minimum": 0.581692439716043, "maximum": 1.5519926335852314, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1893399406507557, "standard_deviation": 0.013169048772890872, "minimum": 0.1752470796025878, "maximum": 0.2098069434197636, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9925999738379095, "standard_deviation": 0.0014365610907554084, "minimum": 0.9901129723252367, "maximum": 0.9940499381838059, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0005303514071642, "standard_deviation": 0.03984262570874032, "minimum": 0.9409257219085755, "maximum": 1.0528672406263193, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0131767179285027, "standard_deviation": 0.34478780768685424, "minimum": 0.5155906991426217, "maximum": 1.4877059199937879, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11214016732676638, "standard_deviation": 0.01664927144917469, "minimum": 0.09664975371619336, "maximum": 0.13969342230488313, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9719871914908405, "standard_deviation": 0.004961621906991122, "minimum": 0.9659612317453679, "maximum": 0.9789635225592764, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5266868149248656, "standard_deviation": 0.0167555024936156, "minimum": 1.5116383607296375, "maximum": 1.5605961674553808, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8803451168975365, "standard_deviation": 0.6183160693900512, "minimum": 0.2913821920142391, "maximum": 1.837588736798996, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2231025732128106, "standard_deviation": 0.020791464387312067, "minimum": 0.19188443149352225, "maximum": 0.24479967050792495, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9897283465102128, "standard_deviation": 0.0027497095620971964, "minimum": 0.9843884960667145, "maximum": 0.9939710020151451, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2607404400664866, "standard_deviation": 0.029358378160792314, "minimum": 1.1987902409008206, "maximum": 1.2983460795972768, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0230319791829763, "standard_deviation": 0.8154553967048863, "minimum": 0.16592279342948282, "maximum": 2.2922149576083286, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13633276772866343, "standard_deviation": 0.021848892996322762, "minimum": 0.10307132509779006, "maximum": 0.17845566295445633, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9874401149309653, "standard_deviation": 0.0017204083825002267, "minimum": 0.9838656535371558, "maximum": 0.9890326841452808, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.796501198770788, "standard_deviation": 0.04104681117219069, "minimum": 1.7293101974913687, "maximum": 1.8481334343907223, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7907323149868397, "standard_deviation": 0.3570695037955267, "minimum": 0.11779812681954094, "maximum": 1.1836601880280833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13189720217576822, "standard_deviation": 0.00801031880122424, "minimum": 0.12302566585631114, "maximum": 0.14805174413104566, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961827157823814, "standard_deviation": 0.0009513446413568013, "minimum": 0.9941344141671492, "maximum": 0.9974369345312964, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0772198665797679, "standard_deviation": 0.06686153224280422, "minimum": 0.9721871042264355, "maximum": 1.1771251523240733, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7684213244368718, "standard_deviation": 0.33312023655190537, "minimum": 0.3728037272214673, "maximum": 1.2148162212457538, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0750486568561334, "standard_deviation": 0.007109098950397679, "minimum": 0.0683876169681865, "maximum": 0.08950022875442794, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3129456028853663, "standard_deviation": 0.08494795580882614, "minimum": 0.20326401804870783, "maximum": 0.4411084876386599, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.735556706572929, "standard_deviation": 0.008363209935763567, "minimum": 5.723472121385558, "maximum": 5.748621749202624, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4092952006433259, "standard_deviation": 0.015088537468645856, "minimum": 0.3946712361673668, "maximum": 0.44289215446890096, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.16049127484104142, "standard_deviation": 0.07098231967618363, "minimum": -0.24783209280602098, "maximum": -0.05684210813915617, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9896045886679922, "standard_deviation": 0.0010521831637373023, "minimum": 0.9882253144724215, "maximum": 0.9906896189311415, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2898580756963356, "standard_deviation": 0.031015313531061402, "minimum": 1.2451510428123511, "maximum": 1.3298459379388803, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5175168390948273, "standard_deviation": 0.3520718554681435, "minimum": 0.1408433173412649, "maximum": 1.207342717246859, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10430715308275787, "standard_deviation": 0.00463402484529641, "minimum": 0.09962431358281418, "maximum": 0.11029749025390718, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9968866543451149, "standard_deviation": 0.0005862301564931122, "minimum": 0.9958370413169814, "maximum": 0.9976524440204418, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6702173505947653, "standard_deviation": 0.020189878574320953, "minimum": 0.6451891911715452, "maximum": 0.6952819341236366, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5620009816655298, "standard_deviation": 0.402993373781936, "minimum": 0.16400787739297193, "maximum": 1.3834198916866658, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.057250674706764725, "standard_deviation": 0.004649515781152805, "minimum": 0.051491324149016866, "maximum": 0.06528547417969717, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0002", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 2.8628385066986084, "predicted_mos": 2.8628385066986084, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "38ffd4cc315633a2ea261765f6cb3c52e5a58f7448527444ff4bd1c34438d688", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.6377885001711547, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2551118836471618, "全帯域_10ms_rms_envelope_nrmse": 0.09834002123584673, "全帯域_10ms_envelope_correlation": 0.9905820561939896, "全帯域_10ms_modulation_mae_db": 0.2679313606702292, "全帯域_40ms_log_rms_mae_db": 0.6736172730366643, "全帯域_40ms_rms_envelope_nrmse": 0.05492950499862228, "全帯域_40ms_envelope_correlation": 0.9968719961374337, "全帯域_40ms_modulation_mae_db": 0.22217852571741362, "300–1000 Hz_10ms_log_rms_mae_db": 2.0668727990318225, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11589500332879793, "300–1000 Hz_10ms_envelope_correlation": 0.9902552401292328, "300–1000 Hz_10ms_modulation_mae_db": 0.33405189606918134, "300–1000 Hz_40ms_log_rms_mae_db": 1.3931638774036792, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06028839538338534, "300–1000 Hz_40ms_envelope_correlation": 0.9975523533182529, "300–1000 Hz_40ms_modulation_mae_db": 0.39367740779051313, "1000–3000 Hz_10ms_log_rms_mae_db": 2.104152161821826, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20594648618508377, "1000–3000 Hz_10ms_envelope_correlation": 0.9713057105090146, "1000–3000 Hz_10ms_modulation_mae_db": 0.6695656479158704, "1000–3000 Hz_40ms_log_rms_mae_db": 1.380909550290355, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13414556158076998, "1000–3000 Hz_40ms_envelope_correlation": 0.9882743464740659, "1000–3000 Hz_40ms_modulation_mae_db": 0.8354542748086704, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9123791604945208, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21276677843151534, "3000–6000 Hz_10ms_envelope_correlation": 0.977170129165449, "3000–6000 Hz_10ms_modulation_mae_db": 0.23184193683783294, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6666227147951584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13472451322938597, "3000–6000 Hz_40ms_envelope_correlation": 0.9915570796938241, "3000–6000 Hz_40ms_modulation_mae_db": 0.24913583040707213, "mr_stft_db": 6.060332366645203, "band_mae_db": 0.3178106197537289, "rms_gain_error_db": -0.07134322090230436, "output_peak": 0.41736400337429824}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2507691539155268, "全帯域_10ms_rms_envelope_nrmse": 0.10161604184082842, "全帯域_10ms_envelope_correlation": 0.9902512549057397, "全帯域_10ms_modulation_mae_db": 0.6968911580915081, "全帯域_40ms_log_rms_mae_db": 0.6409669772726506, "全帯域_40ms_rms_envelope_nrmse": 0.06256104799607534, "全帯域_40ms_envelope_correlation": 0.9963523465504602, "全帯域_40ms_modulation_mae_db": 0.8523059326703754, "300–1000 Hz_10ms_log_rms_mae_db": 2.126149295081047, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12475356275012514, "300–1000 Hz_10ms_envelope_correlation": 0.9889640876600196, "300–1000 Hz_10ms_modulation_mae_db": 0.8672078352024736, "300–1000 Hz_40ms_log_rms_mae_db": 1.3402063847126806, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07244196054139103, "300–1000 Hz_40ms_envelope_correlation": 0.9966886710566297, "300–1000 Hz_40ms_modulation_mae_db": 0.9316666579306082, "1000–3000 Hz_10ms_log_rms_mae_db": 2.163734716369255, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1973470313808775, "1000–3000 Hz_10ms_envelope_correlation": 0.9733903141465129, "1000–3000 Hz_10ms_modulation_mae_db": 0.7427149231206965, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4324064046024327, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1190096932237692, "1000–3000 Hz_40ms_envelope_correlation": 0.9904432644965202, "1000–3000 Hz_40ms_modulation_mae_db": 0.6038174716496484, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8962962334869238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22596348726587034, "3000–6000 Hz_10ms_envelope_correlation": 0.9708363131726142, "3000–6000 Hz_10ms_modulation_mae_db": 1.556102958940012, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6824230627722048, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1748656197650268, "3000–6000 Hz_40ms_envelope_correlation": 0.982405844724363, "3000–6000 Hz_40ms_modulation_mae_db": 1.8821463837086976, "mr_stft_db": 6.058866568250376, "band_mae_db": 0.2259727210155813, "rms_gain_error_db": -0.20287656516431085, "output_peak": 0.4196701124674556}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2246473544974184, "全帯域_10ms_rms_envelope_nrmse": 0.11439969980105917, "全帯域_10ms_envelope_correlation": 0.9877811013962915, "全帯域_10ms_modulation_mae_db": 1.0862945481314619, "全帯域_40ms_log_rms_mae_db": 0.6268373268761395, "全帯域_40ms_rms_envelope_nrmse": 0.07257005656892486, "全帯域_40ms_envelope_correlation": 0.9952661687698356, "全帯域_40ms_modulation_mae_db": 1.1643755023429516, "300–1000 Hz_10ms_log_rms_mae_db": 2.019729139235648, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13821034009767377, "300–1000 Hz_10ms_envelope_correlation": 0.9868573663630535, "300–1000 Hz_10ms_modulation_mae_db": 0.8564063414797953, "300–1000 Hz_40ms_log_rms_mae_db": 1.274037951081916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08079170197671917, "300–1000 Hz_40ms_envelope_correlation": 0.9964270767900424, "300–1000 Hz_40ms_modulation_mae_db": 0.7538930459022244, "1000–3000 Hz_10ms_log_rms_mae_db": 2.068187816676505, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2101079889568229, "1000–3000 Hz_10ms_envelope_correlation": 0.9701657720232413, "1000–3000 Hz_10ms_modulation_mae_db": 0.9034682453119167, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2825424673005237, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12839746126075088, "1000–3000 Hz_40ms_envelope_correlation": 0.9896409274513871, "1000–3000 Hz_40ms_modulation_mae_db": 0.8509458428298693, "3000–6000 Hz_10ms_log_rms_mae_db": 1.899048583555426, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27352742765227434, "3000–6000 Hz_10ms_envelope_correlation": 0.9576907583615236, "3000–6000 Hz_10ms_modulation_mae_db": 2.9392191982385083, "3000–6000 Hz_40ms_log_rms_mae_db": 1.67601837027236, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.22044086336129204, "3000–6000 Hz_40ms_envelope_correlation": 0.9730804575368442, "3000–6000 Hz_40ms_modulation_mae_db": 3.431912475476229, "mr_stft_db": 6.040823747681588, "band_mae_db": 0.3409460302349869, "rms_gain_error_db": -0.26336004404117175, "output_peak": 0.4034517437579762}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3120763583891475, "全帯域_10ms_rms_envelope_nrmse": 0.1118082617229084, "全帯域_10ms_envelope_correlation": 0.9879276079591933, "全帯域_10ms_modulation_mae_db": 1.4021325918784189, "全帯域_40ms_log_rms_mae_db": 0.6712361037537997, "全帯域_40ms_rms_envelope_nrmse": 0.05778853454745239, "全帯域_40ms_envelope_correlation": 0.996743062347039, "全帯域_40ms_modulation_mae_db": 1.517460466188897, "300–1000 Hz_10ms_log_rms_mae_db": 2.0980255053271146, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13703873898061594, "300–1000 Hz_10ms_envelope_correlation": 0.9860735407467266, "300–1000 Hz_10ms_modulation_mae_db": 1.2906557599048585, "300–1000 Hz_40ms_log_rms_mae_db": 1.405389793914294, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07788423967455972, "300–1000 Hz_40ms_envelope_correlation": 0.9953997331485492, "300–1000 Hz_40ms_modulation_mae_db": 1.0571228345870771, "1000–3000 Hz_10ms_log_rms_mae_db": 2.2360539482517208, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22738092906508764, "1000–3000 Hz_10ms_envelope_correlation": 0.964300189533034, "1000–3000 Hz_10ms_modulation_mae_db": 1.3768159019614439, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4398786148598213, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1356739305338133, "1000–3000 Hz_40ms_envelope_correlation": 0.9872988334307617, "1000–3000 Hz_40ms_modulation_mae_db": 1.5424419778762948, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9261969770368281, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29045329368186285, "3000–6000 Hz_10ms_envelope_correlation": 0.9526296370109231, "3000–6000 Hz_10ms_modulation_mae_db": 2.934688453679866, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6401902938446509, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17797838426403972, "3000–6000 Hz_40ms_envelope_correlation": 0.9853452752525138, "3000–6000 Hz_40ms_modulation_mae_db": 3.085549257316746, "mr_stft_db": 6.055786344794872, "band_mae_db": 0.3172093781984891, "rms_gain_error_db": -0.15054787218594434, "output_peak": 0.4305400298585476}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2299610718366842, "全帯域_10ms_rms_envelope_nrmse": 0.0932674615165133, "全帯域_10ms_envelope_correlation": 0.991670975261647, "全帯域_10ms_modulation_mae_db": 1.199424371954062, "全帯域_40ms_log_rms_mae_db": 0.6751529449960263, "全帯域_40ms_rms_envelope_nrmse": 0.05105639786579403, "全帯域_40ms_envelope_correlation": 0.9975020214795657, "全帯域_40ms_modulation_mae_db": 1.217024956733464, "300–1000 Hz_10ms_log_rms_mae_db": 1.9471925375212602, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11087773934505163, "300–1000 Hz_10ms_envelope_correlation": 0.991063892817634, "300–1000 Hz_10ms_modulation_mae_db": 1.1530423310195643, "300–1000 Hz_40ms_log_rms_mae_db": 1.3627254245828195, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06056419804838184, "300–1000 Hz_40ms_envelope_correlation": 0.9974392208668527, "300–1000 Hz_40ms_modulation_mae_db": 1.033943635615436, "1000–3000 Hz_10ms_log_rms_mae_db": 2.1142370705428433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21121289857385706, "1000–3000 Hz_10ms_envelope_correlation": 0.9694271838744217, "1000–3000 Hz_10ms_modulation_mae_db": 0.6501622648350597, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4293381696371095, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1289896822027509, "1000–3000 Hz_40ms_envelope_correlation": 0.9888093884066023, "1000–3000 Hz_40ms_modulation_mae_db": 1.0710721062295387, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8961574513061712, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2159765078642527, "3000–6000 Hz_10ms_envelope_correlation": 0.9735865206631084, "3000–6000 Hz_10ms_modulation_mae_db": 2.065565494862698, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6698764506752877, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14245205560920043, "3000–6000 Hz_40ms_envelope_correlation": 0.9888181792844006, "3000–6000 Hz_40ms_modulation_mae_db": 2.437553919037835, "mr_stft_db": 6.065215873167564, "band_mae_db": 0.24843049946974408, "rms_gain_error_db": -0.14526292001332156, "output_peak": 0.43503515973159657}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2855867019704559, "全帯域_10ms_rms_envelope_nrmse": 0.11293680640084276, "全帯域_10ms_envelope_correlation": 0.9880302162870052, "全帯域_10ms_modulation_mae_db": 0.5449290294719186, "全帯域_40ms_log_rms_mae_db": 0.6799953709756819, "全帯域_40ms_rms_envelope_nrmse": 0.0655353081172205, "全帯域_40ms_envelope_correlation": 0.9961962005056278, "全帯域_40ms_modulation_mae_db": 0.6224074118165639, "300–1000 Hz_10ms_log_rms_mae_db": 2.110613152967784, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14892140106408108, "300–1000 Hz_10ms_envelope_correlation": 0.9842507419341003, "300–1000 Hz_10ms_modulation_mae_db": 0.9079594836337536, "300–1000 Hz_40ms_log_rms_mae_db": 1.4326540610823224, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08839889045524074, "300–1000 Hz_40ms_envelope_correlation": 0.9951225205810637, "300–1000 Hz_40ms_modulation_mae_db": 0.8382366407483173, "1000–3000 Hz_10ms_log_rms_mae_db": 2.224003276529434, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2131553139996683, "1000–3000 Hz_10ms_envelope_correlation": 0.9688096813997636, "1000–3000 Hz_10ms_modulation_mae_db": 1.0299666215416425, "1000–3000 Hz_40ms_log_rms_mae_db": 1.440020262615004, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11279393723842418, "1000–3000 Hz_40ms_envelope_correlation": 0.990972126561233, "1000–3000 Hz_40ms_modulation_mae_db": 0.987568707245897, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9149863527296087, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25742255279685805, "3000–6000 Hz_10ms_envelope_correlation": 0.9626560718972823, "3000–6000 Hz_10ms_modulation_mae_db": 0.9064816737402869, "3000–6000 Hz_40ms_log_rms_mae_db": 1.685911198614677, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15969266105901864, "3000–6000 Hz_40ms_envelope_correlation": 0.9873838937475501, "3000–6000 Hz_40ms_modulation_mae_db": 0.9794963701699707, "mr_stft_db": 6.044865521927705, "band_mae_db": 0.2836147541516026, "rms_gain_error_db": -0.24716871031398774, "output_peak": 0.3997806861459116}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2594536660632214, "全帯域_10ms_rms_envelope_nrmse": 0.10542011192474275, "全帯域_10ms_envelope_correlation": 0.9897048958336355, "全帯域_10ms_modulation_mae_db": 0.498227514676541, "全帯域_40ms_log_rms_mae_db": 0.6915201997065734, "全帯域_40ms_rms_envelope_nrmse": 0.06090042978096689, "全帯域_40ms_envelope_correlation": 0.9969046544207041, "全帯域_40ms_modulation_mae_db": 0.555843857398471, "300–1000 Hz_10ms_log_rms_mae_db": 2.076557154280491, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13319212109811435, "300–1000 Hz_10ms_envelope_correlation": 0.9877517705234004, "300–1000 Hz_10ms_modulation_mae_db": 0.7910746390778512, "300–1000 Hz_40ms_log_rms_mae_db": 1.4378916630651548, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08086022368539326, "300–1000 Hz_40ms_envelope_correlation": 0.9962343758429416, "300–1000 Hz_40ms_modulation_mae_db": 0.7055721269377299, "1000–3000 Hz_10ms_log_rms_mae_db": 2.079697754479141, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18825794748122993, "1000–3000 Hz_10ms_envelope_correlation": 0.9758571411810381, "1000–3000 Hz_10ms_modulation_mae_db": 0.5573045739784624, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3920849974742548, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10956863879482936, "1000–3000 Hz_40ms_envelope_correlation": 0.9920053976046622, "1000–3000 Hz_40ms_modulation_mae_db": 0.2719346635623277, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9266967833837043, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19480783520621492, "3000–6000 Hz_10ms_envelope_correlation": 0.9784302817866732, "3000–6000 Hz_10ms_modulation_mae_db": 0.6431916483713943, "3000–6000 Hz_40ms_log_rms_mae_db": 1.62746587739298, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11786451968656422, "3000–6000 Hz_40ms_envelope_correlation": 0.9923379285820468, "3000–6000 Hz_40ms_modulation_mae_db": 0.8682494995557412, "mr_stft_db": 6.0581096059843915, "band_mae_db": 0.2792469762001207, "rms_gain_error_db": -0.25452471318627107, "output_peak": 0.4257344000159795}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.236556217924981, "全帯域_10ms_rms_envelope_nrmse": 0.10053112491616052, "全帯域_10ms_envelope_correlation": 0.9902495962844691, "全帯域_10ms_modulation_mae_db": 0.6585265768416441, "全帯域_40ms_log_rms_mae_db": 0.667470033974027, "全帯域_40ms_rms_envelope_nrmse": 0.05281926786416744, "全帯域_40ms_envelope_correlation": 0.9972716100760038, "全帯域_40ms_modulation_mae_db": 0.49327742095266264, "300–1000 Hz_10ms_log_rms_mae_db": 2.1835283842246556, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13416207231643298, "300–1000 Hz_10ms_envelope_correlation": 0.9869020095441872, "300–1000 Hz_10ms_modulation_mae_db": 1.0065615461907675, "300–1000 Hz_40ms_log_rms_mae_db": 1.503819456248929, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07496411234399865, "300–1000 Hz_40ms_envelope_correlation": 0.996107163801908, "300–1000 Hz_40ms_modulation_mae_db": 0.5336157478947648, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0993105118758417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19420141999238852, "1000–3000 Hz_10ms_envelope_correlation": 0.9748437514491571, "1000–3000 Hz_10ms_modulation_mae_db": 0.531491833794789, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3724983274468063, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11771390732675041, "1000–3000 Hz_40ms_envelope_correlation": 0.9916431727901399, "1000–3000 Hz_40ms_modulation_mae_db": 0.5097540533403017, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9816984249185754, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2031290668250773, "3000–6000 Hz_10ms_envelope_correlation": 0.9767174457410378, "3000–6000 Hz_10ms_modulation_mae_db": 0.7416818048067398, "3000–6000 Hz_40ms_log_rms_mae_db": 1.6646776991676682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1315725466609607, "3000–6000 Hz_40ms_envelope_correlation": 0.9901226131472272, "3000–6000 Hz_40ms_modulation_mae_db": 0.9436184415977159, "mr_stft_db": 6.052331105840735, "band_mae_db": 0.210724618434702, "rms_gain_error_db": -0.13222445678129757, "output_peak": 0.4055469156399558}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76640, "output_samples": 76640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.258305938289044, "spectral_convergence": 0.3076060401844692, "spectral_convergence_status": "ok", "amplitude_floor": 1.4500777065259659e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.937850037870944, "spectral_convergence": 0.3014095125353286, "spectral_convergence_status": "ok", "amplitude_floor": 1.1426990309647766e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.98484112377562, "spectral_convergence": 0.32302628140140416, "spectral_convergence_status": "ok", "amplitude_floor": 7.641152847432082e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003890931416418852, "output_power": 0.00042277508756035357, "delta_db": 0.36055812387259717, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013826376510160613, "output_power": 0.0013242228380461777, "delta_db": -0.18747305493108296, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006763873367355198, "output_power": 0.0006462878108176141, "delta_db": -0.19769503103124889, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00010128235718512909, "output_power": 9.899290672778257e-05, "delta_db": -0.09929723677119161, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.238121331616846e-05, "output_power": 6.343795011388905e-05, "delta_db": -0.5727671777924817, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017231931162072514, "output_power": 0.0001928595994564822, "delta_db": 0.48907309412377076, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.3178106197537289, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.07134322090230436, "pitch": null, "output_rms_dbfs": -25.60892219542271, "output_peak": 0.41736400337429824, "reference_rms_dbfs": -25.537578974520407, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.060332366645203}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00018226521943209611, "log_rms_mae_db": 1.2551118836471618, "rms_envelope_nrmse": 0.09834002123584673, "envelope_correlation": 0.9905820561939896, "rms_gain_error_db": -0.07134310766604948, "reference_mean_power": 0.0027970153309965396, "output_mean_power": 0.0027514431003822426}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.029329446119484e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010840604762391139, "output_power": 0.0010297114017869073, "reference_active": true, "delta_db": -0.22337989143516954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001764324474949557, "output_power": 0.0001726199508172169, "reference_active": true, "delta_db": -0.09487470220687426}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.741521666729804e-05, "output_power": 5.3023902514024754e-05, "reference_active": true, "delta_db": 0.48553948836864397}], "active_band_count": 3, "active_band_mae_db": 0.2679313606702292}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001676463927845064, "log_rms_mae_db": 0.6736172730366643, "rms_envelope_nrmse": 0.05492950499862228, "envelope_correlation": 0.9968719961374337, "rms_gain_error_db": -0.07133994181320233, "reference_mean_power": 0.0028146416719843875, "output_mean_power": 0.002768784270843488}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.889944424347358e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010453379933036452, "output_power": 0.0009943951675225305, "reference_active": true, "delta_db": -0.21697730229852785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.750212405766423e-05, "output_power": 6.826237246151526e-05, "reference_active": true, "delta_db": 0.048639393404838686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9228813706071663e-06, "output_power": 3.2055538529968563e-06, "reference_active": true, "delta_db": 0.40091888144887433}], "active_band_count": 3, "active_band_mae_db": 0.22217852571741362}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00017586243391937327, "log_rms_mae_db": 2.0668727990318225, "rms_envelope_nrmse": 0.11589500332879793, "envelope_correlation": 0.9902552401292328, "rms_gain_error_db": -0.20320078012948703, "reference_mean_power": 0.0014437040494073058, "output_mean_power": 0.0013777109069380896}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 1.5100208608044745e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009242573938618176, "output_power": 0.0008639226242272265, "reference_active": true, "delta_db": -0.2931808615432159}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001371496531392223, "output_power": 0.00012287235300922222, "reference_active": true, "delta_db": -0.4774053847598739}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2803310723594984e-05, "output_power": 3.4599882340126444e-05, "reference_active": true, "delta_db": 0.23156944190445436}], "active_band_count": 3, "active_band_mae_db": 0.33405189606918134}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001539617967365963, "log_rms_mae_db": 1.3931638774036792, "rms_envelope_nrmse": 0.06028839538338534, "envelope_correlation": 0.9975523533182529, "rms_gain_error_db": -0.2032039050859555, "reference_mean_power": 0.0014528118102568712, "output_mean_power": 0.0013864013454331528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.4301118553576922e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009152950896245062, "output_power": 0.0008525904734535629, "reference_active": true, "delta_db": -0.3082065678084423}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.331274162853222e-05, "output_power": 5.150013125953028e-05, "reference_active": true, "delta_db": -0.15022680923758408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0543911334413026e-06, "output_power": 2.426291722862869e-06, "reference_active": true, "delta_db": 0.722598846325513}], "active_band_count": 3, "active_band_mae_db": 0.39367740779051313}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 8.12616340215947e-05, "log_rms_mae_db": 2.104152161821826, "rms_envelope_nrmse": 0.20594648618508377, "envelope_correlation": 0.9713057105090146, "rms_gain_error_db": -0.3547303683564728, "reference_mean_power": 0.00015412407568599862, "output_mean_power": 0.0001420356778453077}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.103412885748503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011785014167056138, "output_power": 9.646691658877909e-05, "reference_active": true, "delta_db": -0.8695171172150156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.031054512020625e-06, "output_power": 8.203437650908934e-06, "reference_active": true, "delta_db": 0.09223307890487203}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6793453202028428e-06, "output_power": 3.409756491399779e-06, "reference_active": true, "delta_db": 1.0469467476277234}], "active_band_count": 3, "active_band_mae_db": 0.6695656479158704}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 6.921465297386044e-05, "log_rms_mae_db": 1.380909550290355, "rms_envelope_nrmse": 0.13414556158076998, "envelope_correlation": 0.9882743464740659, "rms_gain_error_db": -0.3547315900765048, "reference_mean_power": 0.00015509566252518478, "output_mean_power": 0.00014293102010532399}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.0373126402115645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011306875179582822, "output_power": 9.330480894811885e-05, "reference_active": true, "delta_db": -0.834385699261756}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.441071013913094e-06, "output_power": 4.2732335746220975e-06, "reference_active": true, "delta_db": 0.9406299665373417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5029543761566567e-07, "output_power": 1.7786091129780114e-07, "reference_active": true, "delta_db": 0.7313471586269136}], "active_band_count": 3, "active_band_mae_db": 0.8354542748086704}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00013414824565394521, "log_rms_mae_db": 1.9123791604945208, "rms_envelope_nrmse": 0.21276677843151534, "envelope_correlation": 0.977170129165449, "rms_gain_error_db": 0.44938372271417454, "reference_mean_power": 0.0001730034207748911, "output_mean_power": 0.00019186380927869778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.1522167661145593e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001202536722048893, "output_power": 0.0001346555095341403, "reference_active": true, "delta_db": 0.49125780239334116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5558902161138425e-05, "output_power": 2.4654086028311117e-05, "reference_active": true, "delta_db": -0.15653288430060824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.530968744127951e-05, "output_power": 1.5478890811367108e-05, "reference_active": true, "delta_db": 0.04773512381954946}], "active_band_count": 3, "active_band_mae_db": 0.23184193683783294}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 8.410768197622129e-05, "log_rms_mae_db": 1.6666227147951584, "rms_envelope_nrmse": 0.13472451322938597, "envelope_correlation": 0.9915570796938241, "rms_gain_error_db": 0.44938236198986675, "reference_mean_power": 0.00017409446216015204, "output_mean_power": 0.000193073732699447}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.085538754133615e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012977220791713187, "output_power": 0.0001433108109389836, "reference_active": true, "delta_db": 0.43097259817562233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3304937813307329e-05, "output_power": 1.2772662875941487e-05, "reference_active": true, "delta_db": -0.17731399541936738}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.046486321108266e-06, "output_power": 1.0805519357694304e-06, "reference_active": true, "delta_db": 0.1391208976262267}], "active_band_count": 3, "active_band_mae_db": 0.24913583040707213}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.839630629228609, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8feffa978af0150903beb6ea875006cb85afe796752e365873d495c33c9164be", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9710124680145229, "standard_deviation": 0.0037230307436008675, "minimum": 0.964300189533034, "maximum": 0.9758571411810381, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.1361721570683208, "standard_deviation": 0.06450560343333546, "minimum": 2.068187816676505, "maximum": 2.2360539482517208, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8076862515574852, "standard_deviation": 0.2852512762233758, "minimum": 0.531491833794789, "maximum": 1.3768159019614439, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20595125195437697, "standard_deviation": 0.012428292420245672, "minimum": 0.18825794748122993, "maximum": 0.22738092906508764, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9898859321519216, "standard_deviation": 0.00167245448505606, "minimum": 0.9872988334307617, "maximum": 0.9920053976046622, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3962098492782884, "standard_deviation": 0.05340895213357907, "minimum": 1.2825424673005237, "maximum": 1.440020262615004, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8341236371928186, "standard_deviation": 0.38840380208848746, "minimum": 0.2719346635623277, "maximum": 1.5424419778762948, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12328660152023227, "standard_deviation": 0.00983894826306092, "minimum": 0.10956863879482936, "maximum": 0.1356739305338133, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9687146447248265, "standard_deviation": 0.009820751682916597, "minimum": 0.9526296370109231, "maximum": 0.9784302817866732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9191824958639698, "standard_deviation": 0.02812500144836841, "minimum": 1.8961574513061712, "maximum": 1.9816984249185754, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5023466461846673, "standard_deviation": 1.0503269280185539, "minimum": 0.23184193683783294, "maximum": 2.9392191982385083, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23425586871549076, "standard_deviation": 0.035105567189030594, "minimum": 0.19480783520621492, "maximum": 0.29045329368186285, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9863814089960963, "standard_deviation": 0.006290595253449863, "minimum": 0.9730804575368442, "maximum": 0.9923379285820468, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.6641482084418733, "standard_deviation": 0.020379577200761407, "minimum": 1.62746587739298, "maximum": 1.685911198614677, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.734707772158751, "standard_deviation": 1.1572657366302685, "minimum": 0.24913583040707213, "maximum": 3.431912475476229, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15744889545443608, "standard_deviation": 0.03311373406486227, "minimum": 0.11786451968656422, "maximum": 0.22044086336129204, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9877648312147942, "standard_deviation": 0.00224568065520072, "minimum": 0.9842507419341003, "maximum": 0.991063892817634, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0785834959587275, "standard_deviation": 0.07126258407488041, "minimum": 1.9471925375212602, "maximum": 2.1835283842246556, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9008699790722807, "standard_deviation": 0.2836478524414293, "minimum": 0.33405189606918134, "maximum": 1.2906557599048585, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1303813723726116, "standard_deviation": 0.012498335664235553, "minimum": 0.11087773934505163, "maximum": 0.14892140106408108, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9963713894257801, "standard_deviation": 0.0008644639653944926, "minimum": 0.9951225205810637, "maximum": 0.9975523533182529, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3937360765114746, "standard_deviation": 0.06956347774408746, "minimum": 1.274037951081916, "maximum": 1.503819456248929, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7809660121758338, "standard_deviation": 0.23416803423860577, "minimum": 0.39367740779051313, "maximum": 1.0571228345870771, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07452421526363373, "standard_deviation": 0.009894390262138758, "minimum": 0.06028839538338534, "maximum": 0.08839889045524074, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2779944496823694, "standard_deviation": 0.046636456858951944, "minimum": 0.210724618434702, "maximum": 0.3409460302349869, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.054541391786555, "standard_deviation": 0.00816919943483685, "minimum": 6.040823747681588, "maximum": 6.065215873167564, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.41714038137396514, "standard_deviation": 0.013110664364343376, "minimum": 0.3997806861459116, "maximum": 0.43503515973159657, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.18341356282357615, "standard_deviation": 0.06928973758305802, "minimum": -0.26336004404117175, "maximum": -0.07134322090230436, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9895247130152464, "standard_deviation": 0.001445912833629996, "minimum": 0.9877811013962915, "maximum": 0.991670975261647, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2567703010305746, "standard_deviation": 0.029506263563213238, "minimum": 1.2246473544974184, "maximum": 1.3120763583891475, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.794294643964473, "standard_deviation": 0.39164099437139444, "minimum": 0.2679313606702292, "maximum": 1.4021325918784189, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10478994116986276, "standard_deviation": 0.007662086343604242, "minimum": 0.0932674615165133, "maximum": 0.11439969980105917, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9966385075358337, "standard_deviation": 0.0007013858999478373, "minimum": 0.9952661687698356, "maximum": 0.9975020214795657, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6658495288239454, "standard_deviation": 0.02130966948629802, "minimum": 0.6268373268761395, "maximum": 0.6915201997065734, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8306092592276, "standard_deviation": 0.43683555002611607, "minimum": 0.22217852571741362, "maximum": 1.517460466188897, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.059770068467402966, "standard_deviation": 0.007148867870390312, "minimum": 0.05105639786579403, "maximum": 0.07257005656892486, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0002", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.8804404735565186, "predicted_mos": 2.8804404735565186, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "94e48f05d2e7dd08433f60061a0feb3ccc037848853e05e1f2616f2d3e4348bb", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.056609200313687, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2167567475522054, "全帯域_10ms_rms_envelope_nrmse": 0.09885670342288604, "全帯域_10ms_envelope_correlation": 0.9906953585345758, "全帯域_10ms_modulation_mae_db": 0.6195808935317793, "全帯域_40ms_log_rms_mae_db": 0.6909375311690418, "全帯域_40ms_rms_envelope_nrmse": 0.05345664360146929, "全帯域_40ms_envelope_correlation": 0.9973557809865199, "全帯域_40ms_modulation_mae_db": 0.5749868982911868, "300–1000 Hz_10ms_log_rms_mae_db": 1.850558839037413, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12213475839678054, "300–1000 Hz_10ms_envelope_correlation": 0.9891882184346064, "300–1000 Hz_10ms_modulation_mae_db": 0.7669979885133328, "300–1000 Hz_40ms_log_rms_mae_db": 1.1828430436790116, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06929333753770224, "300–1000 Hz_40ms_envelope_correlation": 0.996696356995609, "300–1000 Hz_40ms_modulation_mae_db": 0.8821900565531751, "1000–3000 Hz_10ms_log_rms_mae_db": 1.687888139918333, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1776302974238055, "1000–3000 Hz_10ms_envelope_correlation": 0.9788008338381553, "1000–3000 Hz_10ms_modulation_mae_db": 0.9636850382474854, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0840852974300106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11688432066350322, "1000–3000 Hz_40ms_envelope_correlation": 0.9911762563612737, "1000–3000 Hz_40ms_modulation_mae_db": 0.6959619389649077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.471964962238915, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20053030345325573, "3000–6000 Hz_10ms_envelope_correlation": 0.9770354205283136, "3000–6000 Hz_10ms_modulation_mae_db": 1.3220495771083518, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2106299101555793, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1276724273136926, "3000–6000 Hz_40ms_envelope_correlation": 0.9907493799561993, "3000–6000 Hz_40ms_modulation_mae_db": 1.5821558854419473, "mr_stft_db": 5.712827283771034, "band_mae_db": 0.262242402154736, "rms_gain_error_db": -0.17455699294930574, "output_peak": 0.40053946903073623}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3104119843830027, "全帯域_10ms_rms_envelope_nrmse": 0.11102279415679533, "全帯域_10ms_envelope_correlation": 0.9879810356186776, "全帯域_10ms_modulation_mae_db": 0.3771956841381275, "全帯域_40ms_log_rms_mae_db": 0.6646825336579542, "全帯域_40ms_rms_envelope_nrmse": 0.06391694829457907, "全帯域_40ms_envelope_correlation": 0.9957477400174606, "全帯域_40ms_modulation_mae_db": 0.2785437451808049, "300–1000 Hz_10ms_log_rms_mae_db": 1.800216819253237, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14637875892307792, "300–1000 Hz_10ms_envelope_correlation": 0.9840458139319938, "300–1000 Hz_10ms_modulation_mae_db": 0.7447933621455083, "300–1000 Hz_40ms_log_rms_mae_db": 1.1448278765038071, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08018169183246095, "300–1000 Hz_40ms_envelope_correlation": 0.9950443232810263, "300–1000 Hz_40ms_modulation_mae_db": 0.6246840791240582, "1000–3000 Hz_10ms_log_rms_mae_db": 1.635622480215221, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18340818838358058, "1000–3000 Hz_10ms_envelope_correlation": 0.9782784377022951, "1000–3000 Hz_10ms_modulation_mae_db": 1.2051447316524346, "1000–3000 Hz_40ms_log_rms_mae_db": 0.962551829244571, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11364958136568362, "1000–3000 Hz_40ms_envelope_correlation": 0.9930482619969833, "1000–3000 Hz_40ms_modulation_mae_db": 1.251067451243476, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4949146039304881, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19155430666211914, "3000–6000 Hz_10ms_envelope_correlation": 0.9794980139628534, "3000–6000 Hz_10ms_modulation_mae_db": 0.3371521569054891, "3000–6000 Hz_40ms_log_rms_mae_db": 1.238031885345684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11790165570517627, "3000–6000 Hz_40ms_envelope_correlation": 0.9920706865474305, "3000–6000 Hz_40ms_modulation_mae_db": 0.49569423146614744, "mr_stft_db": 5.710176587119328, "band_mae_db": 0.2463878742678769, "rms_gain_error_db": -0.08261081444934926, "output_peak": 0.39922980553252885}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3383701078034165, "全帯域_10ms_rms_envelope_nrmse": 0.10830915880062414, "全帯域_10ms_envelope_correlation": 0.9885708956671866, "全帯域_10ms_modulation_mae_db": 0.19400197500973002, "全帯域_40ms_log_rms_mae_db": 0.7349784695740165, "全帯域_40ms_rms_envelope_nrmse": 0.06334378125979555, "全帯域_40ms_envelope_correlation": 0.9957691522000898, "全帯域_40ms_modulation_mae_db": 0.3525191252529596, "300–1000 Hz_10ms_log_rms_mae_db": 1.8390693227784356, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13267484851879122, "300–1000 Hz_10ms_envelope_correlation": 0.9869026118275117, "300–1000 Hz_10ms_modulation_mae_db": 0.23674378472546356, "300–1000 Hz_40ms_log_rms_mae_db": 1.1754010039577136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07385557765820731, "300–1000 Hz_40ms_envelope_correlation": 0.9957254396314368, "300–1000 Hz_40ms_modulation_mae_db": 0.4067318136839093, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6055098941368813, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1739285111355961, "1000–3000 Hz_10ms_envelope_correlation": 0.9797928086078347, "1000–3000 Hz_10ms_modulation_mae_db": 0.43423792602195493, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9529607345726435, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09813086164494736, "1000–3000 Hz_40ms_envelope_correlation": 0.9942801028399703, "1000–3000 Hz_40ms_modulation_mae_db": 0.8006084306253479, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4533478969834075, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2112056109312391, "3000–6000 Hz_10ms_envelope_correlation": 0.9759718740904522, "3000–6000 Hz_10ms_modulation_mae_db": 0.5845488171245495, "3000–6000 Hz_40ms_log_rms_mae_db": 1.153752482771238, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15396894985234685, "3000–6000 Hz_40ms_envelope_correlation": 0.98714122199863, "3000–6000 Hz_40ms_modulation_mae_db": 0.656924070700751, "mr_stft_db": 5.700052452439795, "band_mae_db": 0.2549075021773481, "rms_gain_error_db": -0.03348006909424142, "output_peak": 0.42594917135443094}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2839401241329573, "全帯域_10ms_rms_envelope_nrmse": 0.10340633523330497, "全帯域_10ms_envelope_correlation": 0.9897585537962508, "全帯域_10ms_modulation_mae_db": 1.3587006749005759, "全帯域_40ms_log_rms_mae_db": 0.6309788720395638, "全帯域_40ms_rms_envelope_nrmse": 0.059218955126361356, "全帯域_40ms_envelope_correlation": 0.9966257042432227, "全帯域_40ms_modulation_mae_db": 1.4692594534832157, "300–1000 Hz_10ms_log_rms_mae_db": 1.7494047638190593, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13113907618735698, "300–1000 Hz_10ms_envelope_correlation": 0.9874190462124346, "300–1000 Hz_10ms_modulation_mae_db": 1.2784645493473443, "300–1000 Hz_40ms_log_rms_mae_db": 1.1270945825998204, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07837681418128137, "300–1000 Hz_40ms_envelope_correlation": 0.9955671869879957, "300–1000 Hz_40ms_modulation_mae_db": 1.1572537508035767, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6225213349171361, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1926592111358062, "1000–3000 Hz_10ms_envelope_correlation": 0.9753158910529255, "1000–3000 Hz_10ms_modulation_mae_db": 1.2609653187518564, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0195599458926468, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12107149911475926, "1000–3000 Hz_40ms_envelope_correlation": 0.9911377537752142, "1000–3000 Hz_40ms_modulation_mae_db": 0.946918013340578, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4542767690109333, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20260213965627227, "3000–6000 Hz_10ms_envelope_correlation": 0.97656427438301, "3000–6000 Hz_10ms_modulation_mae_db": 1.4596592459807871, "3000–6000 Hz_40ms_log_rms_mae_db": 1.174967022155574, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11934092086600688, "3000–6000 Hz_40ms_envelope_correlation": 0.991966070031597, "3000–6000 Hz_40ms_modulation_mae_db": 1.6858172356985015, "mr_stft_db": 5.711871752828979, "band_mae_db": 0.32596801899905276, "rms_gain_error_db": -0.16789184519626632, "output_peak": 0.41994990293422896}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3062509163019806, "全帯域_10ms_rms_envelope_nrmse": 0.11281989627484558, "全帯域_10ms_envelope_correlation": 0.9876652827848191, "全帯域_10ms_modulation_mae_db": 0.1623418601619053, "全帯域_40ms_log_rms_mae_db": 0.6937379693714256, "全帯域_40ms_rms_envelope_nrmse": 0.0651989297642373, "全帯域_40ms_envelope_correlation": 0.9956868079977149, "全帯域_40ms_modulation_mae_db": 0.3938946337879708, "300–1000 Hz_10ms_log_rms_mae_db": 1.8582538286569992, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1386886981937782, "300–1000 Hz_10ms_envelope_correlation": 0.9857593026624026, "300–1000 Hz_10ms_modulation_mae_db": 0.3576234202411476, "300–1000 Hz_40ms_log_rms_mae_db": 1.222148506354458, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08253421498605001, "300–1000 Hz_40ms_envelope_correlation": 0.9948609892561523, "300–1000 Hz_40ms_modulation_mae_db": 0.12596069730531687, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6839981734654481, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17658490761234202, "1000–3000 Hz_10ms_envelope_correlation": 0.9789574236082788, "1000–3000 Hz_10ms_modulation_mae_db": 0.5309689643810297, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0564724335437927, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1133226263087414, "1000–3000 Hz_40ms_envelope_correlation": 0.9916209652673021, "1000–3000 Hz_40ms_modulation_mae_db": 0.9068523317882592, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4780049307593486, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22037022599781622, "3000–6000 Hz_10ms_envelope_correlation": 0.9729681918363529, "3000–6000 Hz_10ms_modulation_mae_db": 0.25219788622863004, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2121232597889806, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11316507633375453, "3000–6000 Hz_40ms_envelope_correlation": 0.9927567863229725, "3000–6000 Hz_40ms_modulation_mae_db": 0.4656054570647538, "mr_stft_db": 5.6970362623315465, "band_mae_db": 0.21600404296082854, "rms_gain_error_db": -0.12965917237641927, "output_peak": 0.43851633071635976}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2658091644795675, "全帯域_10ms_rms_envelope_nrmse": 0.10446150864417808, "全帯域_10ms_envelope_correlation": 0.9895490976341076, "全帯域_10ms_modulation_mae_db": 0.8220856614650461, "全帯域_40ms_log_rms_mae_db": 0.7125983649460927, "全帯域_40ms_rms_envelope_nrmse": 0.057594359625966526, "全帯域_40ms_envelope_correlation": 0.9968404416254112, "全帯域_40ms_modulation_mae_db": 0.6993603559309877, "300–1000 Hz_10ms_log_rms_mae_db": 1.7747923345699292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12538482122398267, "300–1000 Hz_10ms_envelope_correlation": 0.988584854641711, "300–1000 Hz_10ms_modulation_mae_db": 0.8680986479331848, "300–1000 Hz_40ms_log_rms_mae_db": 1.103612439086542, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06532155974359481, "300–1000 Hz_40ms_envelope_correlation": 0.9971341504638662, "300–1000 Hz_40ms_modulation_mae_db": 0.8333962937378782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5946056788872431, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16672676973187184, "1000–3000 Hz_10ms_envelope_correlation": 0.9809315410196717, "1000–3000 Hz_10ms_modulation_mae_db": 0.7149480562894409, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9726539852321181, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08821782518336857, "1000–3000 Hz_40ms_envelope_correlation": 0.9946088151105685, "1000–3000 Hz_40ms_modulation_mae_db": 1.1056244771769876, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4643738799503092, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22481994157067306, "3000–6000 Hz_10ms_envelope_correlation": 0.9710560398394565, "3000–6000 Hz_10ms_modulation_mae_db": 0.6681498611508196, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2123526153609863, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14946178883653488, "3000–6000 Hz_40ms_envelope_correlation": 0.9869674036950908, "3000–6000 Hz_40ms_modulation_mae_db": 0.544245982345025, "mr_stft_db": 5.706459862091328, "band_mae_db": 0.17933360195545378, "rms_gain_error_db": -0.1704336745991747, "output_peak": 0.40949906726358537}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2476499270993713, "全帯域_10ms_rms_envelope_nrmse": 0.10296339647899702, "全帯域_10ms_envelope_correlation": 0.9898273689586634, "全帯域_10ms_modulation_mae_db": 0.7396714980257567, "全帯域_40ms_log_rms_mae_db": 0.6257476572914773, "全帯域_40ms_rms_envelope_nrmse": 0.060088676615640615, "全帯域_40ms_envelope_correlation": 0.9964855557478635, "全帯域_40ms_modulation_mae_db": 0.9654504064924317, "300–1000 Hz_10ms_log_rms_mae_db": 1.672731764086352, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12409373146261674, "300–1000 Hz_10ms_envelope_correlation": 0.9886327358138335, "300–1000 Hz_10ms_modulation_mae_db": 0.47893252626259003, "300–1000 Hz_40ms_log_rms_mae_db": 0.9776504652467456, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07281079362743081, "300–1000 Hz_40ms_envelope_correlation": 0.9960260799968046, "300–1000 Hz_40ms_modulation_mae_db": 0.5205957541314438, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5798378103013349, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2006373038710556, "1000–3000 Hz_10ms_envelope_correlation": 0.9738407723178653, "1000–3000 Hz_10ms_modulation_mae_db": 1.3598424920475647, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9638412366590311, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14108382600206143, "1000–3000 Hz_40ms_envelope_correlation": 0.9882751624620526, "1000–3000 Hz_40ms_modulation_mae_db": 1.037075370805528, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4134485619593076, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20309295539985953, "3000–6000 Hz_10ms_envelope_correlation": 0.9818970762887623, "3000–6000 Hz_10ms_modulation_mae_db": 2.083381125556962, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1861269055426586, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.150680970759001, "3000–6000 Hz_40ms_envelope_correlation": 0.9936351890072689, "3000–6000 Hz_40ms_modulation_mae_db": 2.097388422731727, "mr_stft_db": 5.7050544686989175, "band_mae_db": 0.4636323815834383, "rms_gain_error_db": -0.16065439976464574, "output_peak": 0.4539568629934446}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3412326074212229, "全帯域_10ms_rms_envelope_nrmse": 0.11276558334573715, "全帯域_10ms_envelope_correlation": 0.9877628223367712, "全帯域_10ms_modulation_mae_db": 0.29349422546513315, "全帯域_40ms_log_rms_mae_db": 0.759326610542991, "全帯域_40ms_rms_envelope_nrmse": 0.059575208340685214, "全帯域_40ms_envelope_correlation": 0.9965892694386447, "全帯域_40ms_modulation_mae_db": 0.2513614192851693, "300–1000 Hz_10ms_log_rms_mae_db": 1.886887529342949, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14218953358948513, "300–1000 Hz_10ms_envelope_correlation": 0.9849617753985715, "300–1000 Hz_10ms_modulation_mae_db": 0.03866089186122373, "300–1000 Hz_40ms_log_rms_mae_db": 1.2587619909019163, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07642244557167202, "300–1000 Hz_40ms_envelope_correlation": 0.9955150751966685, "300–1000 Hz_40ms_modulation_mae_db": 0.5101578243832879, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6878422508318613, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.199753944324542, "1000–3000 Hz_10ms_envelope_correlation": 0.9762808592912258, "1000–3000 Hz_10ms_modulation_mae_db": 0.6507292321717857, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0604129061576297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14704374476825383, "1000–3000 Hz_40ms_envelope_correlation": 0.9895993617349103, "1000–3000 Hz_40ms_modulation_mae_db": 0.6735168215739221, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3900089300121692, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17516580874929363, "3000–6000 Hz_10ms_envelope_correlation": 0.9831511228647685, "3000–6000 Hz_10ms_modulation_mae_db": 1.0680067358394885, "3000–6000 Hz_40ms_log_rms_mae_db": 1.161116487566333, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11916167935875711, "3000–6000 Hz_40ms_envelope_correlation": 0.9928131113097501, "3000–6000 Hz_40ms_modulation_mae_db": 1.1017101034177423, "mr_stft_db": 5.713245050553556, "band_mae_db": 0.3955523371072263, "rms_gain_error_db": -0.16891687334340955, "output_peak": 0.4579838334414842}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76640, "output_samples": 76640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.779665867995752, "spectral_convergence": 0.29782350037053096, "spectral_convergence_status": "ok", "amplitude_floor": 1.4500777065259659e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.503537502865988, "spectral_convergence": 0.2916868216088581, "spectral_convergence_status": "ok", "amplitude_floor": 1.1426990309647766e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.855278480451361, "spectral_convergence": 0.31772528694658225, "spectral_convergence_status": "ok", "amplitude_floor": 7.641152847432082e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003890931416418852, "output_power": 0.0004091210207181087, "delta_db": 0.21798218436359637, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013826376510160613, "output_power": 0.001307063505163842, "delta_db": -0.2441169032585575, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006763873367355198, "output_power": 0.0006439082902873357, "delta_db": -0.21371451530503183, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00010128235718512909, "output_power": 9.397779621037314e-05, "delta_db": -0.32508543748339586, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.238121331616846e-05, "output_power": 6.714860007034269e-05, "delta_db": -0.32588896024352193, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017231931162072514, "output_power": 0.00016280483924888085, "delta_db": -0.2466664122743124, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.262242402154736, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.17455699294930574, "pitch": null, "output_rms_dbfs": -25.712135967469713, "output_peak": 0.40053946903073623, "reference_rms_dbfs": -25.537578974520407, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.712827283771034}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00018226521943209611, "log_rms_mae_db": 1.2167567475522054, "rms_envelope_nrmse": 0.09885670342288604, "envelope_correlation": 0.9906953585345758, "rms_gain_error_db": -0.1745565203339462, "reference_mean_power": 0.0027970153309965396, "output_mean_power": 0.002686823850459472}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.029329446119484e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010840604762391139, "output_power": 0.001022985369972657, "reference_active": true, "delta_db": -0.25184087989167425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001764324474949557, "output_power": 0.000153188008374104, "reference_active": true, "delta_db": -0.6135368889509752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.741521666729804e-05, "output_power": 3.772083069345077e-05, "reference_active": true, "delta_db": -0.9933649117526883}], "active_band_count": 3, "active_band_mae_db": 0.6195808935317793}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001676463927845064, "log_rms_mae_db": 0.6909375311690418, "rms_envelope_nrmse": 0.05345664360146929, "envelope_correlation": 0.9973557809865199, "rms_gain_error_db": -0.1745669483019549, "reference_mean_power": 0.0028146416719843875, "output_mean_power": 0.002703749290393649}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.889944424347358e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010453379933036452, "output_power": 0.0009811286233917752, "reference_active": true, "delta_db": -0.27530789340084355}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.750212405766423e-05, "output_power": 6.116343984011074e-05, "reference_active": true, "delta_db": -0.4282553657931165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9228813706071663e-06, "output_power": 2.310316318457037e-06, "reference_active": true, "delta_db": -1.0213974356796003}], "active_band_count": 3, "active_band_mae_db": 0.5749868982911868}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00017586243391937327, "log_rms_mae_db": 1.850558839037413, "rms_envelope_nrmse": 0.12213475839678054, "envelope_correlation": 0.9891882184346064, "rms_gain_error_db": -0.23247006767006384, "reference_mean_power": 0.0014437040494073058, "output_mean_power": 0.0013684570390720436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 1.5100208608044745e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009242573938618176, "output_power": 0.000859285133300642, "reference_active": true, "delta_db": -0.3165563549828377}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001371496531392223, "output_power": 0.00011657128302668022, "reference_active": true, "delta_db": -0.7060313707894177}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2803310723594984e-05, "output_power": 2.44386411160988e-05, "reference_active": true, "delta_db": -1.2784062397677431}], "active_band_count": 3, "active_band_mae_db": 0.7669979885133328}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001539617967365963, "log_rms_mae_db": 1.1828430436790116, "rms_envelope_nrmse": 0.06929333753770224, "envelope_correlation": 0.996696356995609, "rms_gain_error_db": -0.23247081778472573, "reference_mean_power": 0.0014528118102568712, "output_mean_power": 0.0013770898582865499}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.4301118553576922e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009152950896245062, "output_power": 0.0008463885417319109, "reference_active": true, "delta_db": -0.3399135714105255}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.331274162853222e-05, "output_power": 4.639015790825517e-05, "reference_active": true, "delta_db": -0.6040516603864714}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0543911334413026e-06, "output_power": 1.3881060217296956e-06, "reference_active": true, "delta_db": -1.7026049378625283}], "active_band_count": 3, "active_band_mae_db": 0.8821900565531751}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 8.12616340215947e-05, "log_rms_mae_db": 1.687888139918333, "rms_envelope_nrmse": 0.1776302974238055, "envelope_correlation": 0.9788008338381553, "rms_gain_error_db": -0.35663729914831843, "reference_mean_power": 0.00015412407568599862, "output_mean_power": 0.00014197332550979283}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.103412885748503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011785014167056138, "output_power": 9.897919999410372e-05, "reference_active": true, "delta_db": -0.7578616966116359}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.031054512020625e-06, "output_power": 6.032043533666336e-06, "reference_active": true, "delta_db": -1.243081065516593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6793453202028428e-06, "output_power": 3.2888186409537778e-06, "reference_active": true, "delta_db": 0.8901123526142272}], "active_band_count": 3, "active_band_mae_db": 0.9636850382474854}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 6.921465297386044e-05, "log_rms_mae_db": 1.0840852974300106, "rms_envelope_nrmse": 0.11688432066350322, "envelope_correlation": 0.9911762563612737, "rms_gain_error_db": -0.35664098858302606, "reference_mean_power": 0.00015509566252518478, "output_mean_power": 0.00014286819354343398}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.0373126402115645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011306875179582822, "output_power": 9.5157191285695e-05, "reference_active": true, "delta_db": -0.749009832030651}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.441071013913094e-06, "output_power": 2.9008527681482886e-06, "reference_active": true, "delta_db": -0.741679484306911}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5029543761566567e-07, "output_power": 1.7245089289793065e-07, "reference_active": true, "delta_db": 0.5971965005571609}], "active_band_count": 3, "active_band_mae_db": 0.6959619389649077}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00013414824565394521, "log_rms_mae_db": 1.471964962238915, "rms_envelope_nrmse": 0.20053030345325573, "envelope_correlation": 0.9770354205283136, "rms_gain_error_db": -0.23357945351627252, "reference_mean_power": 0.0001730034207748911, "output_mean_power": 0.00016394445981442632}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.1522167661145593e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001202536722048893, "output_power": 0.0001165869278316998, "reference_active": true, "delta_db": -0.1344848895667905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5558902161138425e-05, "output_power": 1.5208814082098302e-05, "reference_active": true, "delta_db": -2.254468445537917}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.530968744127951e-05, "output_power": 1.064747881689644e-05, "reference_active": true, "delta_db": -1.5771953962203482}], "active_band_count": 3, "active_band_mae_db": 1.3220495771083518}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 8.410768197622129e-05, "log_rms_mae_db": 1.2106299101555793, "rms_envelope_nrmse": 0.1276724273136926, "envelope_correlation": 0.9907493799561993, "rms_gain_error_db": -0.23358249486382393, "reference_mean_power": 0.00017409446216015204, "output_mean_power": 0.0001649782555741816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.085538754133615e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012977220791713187, "output_power": 0.00012059654994143105, "reference_active": true, "delta_db": -0.3184681012407567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3304937813307329e-05, "output_power": 7.028485833171301e-06, "reference_active": true, "delta_db": -2.7715107522623956}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.046486321108266e-06, "output_power": 7.146356470779222e-07, "reference_active": true, "delta_db": -1.656488802822689}], "active_band_count": 3, "active_band_mae_db": 1.5821558854419473}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.846159943687943, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b259d2835397e7b08afe3aca65f3f81e646abc0854d8f634d9d13829e356d9fb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9777748209297815, "standard_deviation": 0.002405263081223518, "minimum": 0.9738407723178653, "maximum": 0.9809315410196717, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6372282203341824, "standard_deviation": 0.04416025904311632, "minimum": 1.5798378103013349, "maximum": 1.687888139918333, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.890065219945444, "standard_deviation": 0.3560956780977285, "minimum": 0.43423792602195493, "maximum": 1.3598424920475647, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18391614170232495, "standard_deviation": 0.012510673388967693, "minimum": 0.16672676973187184, "maximum": 0.2006373038710556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9917183349435343, "standard_deviation": 0.0021932404288445007, "minimum": 0.9882751624620526, "maximum": 0.9946088151105685, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0090672960915552, "standard_deviation": 0.052514454493972094, "minimum": 0.9529607345726435, "maximum": 1.0840852974300106, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9272031044398759, "standard_deviation": 0.20122924597138908, "minimum": 0.6735168215739221, "maximum": 1.251067451243476, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11742553563141483, "standard_deviation": 0.01966507225836379, "minimum": 0.08821782518336857, "maximum": 0.14704374476825383, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9772677517242462, "standard_deviation": 0.004145884505298897, "minimum": 0.9710560398394565, "maximum": 0.9831511228647685, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4525425668556098, "standard_deviation": 0.03464799289763346, "minimum": 1.3900089300121692, "maximum": 1.4949146039304881, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9718931757368847, "standard_deviation": 0.6288676499147893, "minimum": 0.25219788622863004, "maximum": 2.083381125556962, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2036676615525661, "standard_deviation": 0.015827701347798203, "minimum": 0.17516580874929363, "maximum": 0.22481994157067306, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9910124811086174, "standard_deviation": 0.002580451724102269, "minimum": 0.9869674036950908, "maximum": 0.9936351890072689, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1936375710858793, "standard_deviation": 0.029289220182074378, "minimum": 1.153752482771238, "maximum": 1.238031885345684, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0786926736083244, "standard_deviation": 0.6369157165311123, "minimum": 0.4656054570647538, "maximum": 2.097388422731727, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13141918362815877, "standard_deviation": 0.017034205178132217, "minimum": 0.11316507633375453, "maximum": 0.15396894985234685, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9869367948653831, "standard_deviation": 0.0018723522332094497, "minimum": 0.9840458139319938, "maximum": 0.9891882184346064, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8039894001930468, "standard_deviation": 0.069990603643117, "minimum": 1.672731764086352, "maximum": 1.886887529342949, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5962893963787244, "standard_deviation": 0.39666371847377535, "minimum": 0.03866089186122373, "maximum": 1.2784645493473443, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13283552831198367, "standard_deviation": 0.008900388439628827, "minimum": 0.12213475839678054, "maximum": 0.14637875892307792, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958212002261948, "standard_deviation": 0.0007768674885510755, "minimum": 0.9948609892561523, "maximum": 0.9971341504638662, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.149042488541252, "standard_deviation": 0.08548909410281466, "minimum": 0.9776504652467456, "maximum": 1.2587619909019163, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6326212837153308, "standard_deviation": 0.31918991713570816, "minimum": 0.12596069730531687, "maximum": 1.1572537508035767, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07484955439229994, "standard_deviation": 0.005722137579715795, "minimum": 0.06532155974359481, "maximum": 0.08253421498605001, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2930035201507451, "standard_deviation": 0.09572092441365056, "minimum": 0.17933360195545378, "maximum": 0.4636323815834383, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.707090464979311, "standard_deviation": 0.006074988877697742, "minimum": 5.6970362623315465, "maximum": 5.713245050553556, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.42570305540834985, "standard_deviation": 0.022785565306054273, "minimum": 0.39922980553252885, "maximum": 0.4579838334414842, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1360254802216015, "standard_deviation": 0.051798664748207074, "minimum": -0.17455699294930574, "maximum": -0.03348006909424142, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9889763019163815, "standard_deviation": 0.0011317304568356436, "minimum": 0.9876652827848191, "maximum": 0.9906953585345758, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2888026973967155, "standard_deviation": 0.04377497953581485, "minimum": 1.2167567475522054, "maximum": 1.3412326074212229, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5708840590872567, "standard_deviation": 0.4032059003263328, "minimum": 0.1623418601619053, "maximum": 1.3587006749005759, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10682567204467103, "standard_deviation": 0.005164889759842832, "minimum": 0.09885670342288604, "maximum": 0.11281989627484558, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963875565321159, "standard_deviation": 0.0006014752549057216, "minimum": 0.9956868079977149, "maximum": 0.9973557809865199, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6891235010740704, "standard_deviation": 0.04720996311441462, "minimum": 0.6257476572914773, "maximum": 0.759326610542991, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6231720047130909, "standard_deviation": 0.4181554371481721, "minimum": 0.2513614192851693, "maximum": 1.4692594534832157, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.060299187828591866, "standard_deviation": 0.0038216877732416485, "minimum": 0.05345664360146929, "maximum": 0.0651989297642373, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0002", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.6298840045928955, "predicted_mos": 2.6298840045928955, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "bc8c27a1dde3a7758678b61dcab4a8141c8ab4353d73c46ee444fdf30acaea01", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.014546199701726, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2315243958393338, "全帯域_10ms_rms_envelope_nrmse": 0.10070968410137918, "全帯域_10ms_envelope_correlation": 0.9902654174401705, "全帯域_10ms_modulation_mae_db": 0.32783610928005286, "全帯域_40ms_log_rms_mae_db": 0.6728577880583877, "全帯域_40ms_rms_envelope_nrmse": 0.06292818272123227, "全帯域_40ms_envelope_correlation": 0.9960716554558857, "全帯域_40ms_modulation_mae_db": 0.2396501592571021, "300–1000 Hz_10ms_log_rms_mae_db": 1.8971741379708371, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12398074671886529, "300–1000 Hz_10ms_envelope_correlation": 0.9886991613787924, "300–1000 Hz_10ms_modulation_mae_db": 0.5031988314068604, "300–1000 Hz_40ms_log_rms_mae_db": 1.2317691352925075, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06720617201473919, "300–1000 Hz_40ms_envelope_correlation": 0.9967199043001629, "300–1000 Hz_40ms_modulation_mae_db": 0.3970445089781672, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8599482168261166, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18509464309669918, "1000–3000 Hz_10ms_envelope_correlation": 0.9764970846995527, "1000–3000 Hz_10ms_modulation_mae_db": 0.5935472408753415, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2060745165018119, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10464969259233829, "1000–3000 Hz_40ms_envelope_correlation": 0.992145271539563, "1000–3000 Hz_40ms_modulation_mae_db": 0.8744975677157019, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5776274462358952, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24172433142195732, "3000–6000 Hz_10ms_envelope_correlation": 0.96775689076365, "3000–6000 Hz_10ms_modulation_mae_db": 1.4955391245104301, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4089729327008031, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18643767623866822, "3000–6000 Hz_40ms_envelope_correlation": 0.9819388696229864, "3000–6000 Hz_40ms_modulation_mae_db": 1.7461533928339652, "mr_stft_db": 5.805469612608195, "band_mae_db": 0.23046935896841006, "rms_gain_error_db": -0.15044225481546847, "output_peak": 0.40244264626301435}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2741218999897244, "全帯域_10ms_rms_envelope_nrmse": 0.09936864735642514, "全帯域_10ms_envelope_correlation": 0.9906792344444191, "全帯域_10ms_modulation_mae_db": 1.3173586988894979, "全帯域_40ms_log_rms_mae_db": 0.6436416527802951, "全帯域_40ms_rms_envelope_nrmse": 0.05345318258552375, "全帯域_40ms_envelope_correlation": 0.9974682299840193, "全帯域_40ms_modulation_mae_db": 1.5246182084355764, "300–1000 Hz_10ms_log_rms_mae_db": 1.9376171043466526, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12209013543975081, "300–1000 Hz_10ms_envelope_correlation": 0.9893223180350917, "300–1000 Hz_10ms_modulation_mae_db": 1.1890657122156998, "300–1000 Hz_40ms_log_rms_mae_db": 1.336102115569688, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07063787614982585, "300–1000 Hz_40ms_envelope_correlation": 0.9966986949517136, "300–1000 Hz_40ms_modulation_mae_db": 1.1965482335186104, "1000–3000 Hz_10ms_log_rms_mae_db": 1.883054139500908, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21245875915034884, "1000–3000 Hz_10ms_envelope_correlation": 0.9707733061916062, "1000–3000 Hz_10ms_modulation_mae_db": 0.9340780957965379, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2112302684098994, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13491572165423496, "1000–3000 Hz_40ms_envelope_correlation": 0.9903303728737178, "1000–3000 Hz_40ms_modulation_mae_db": 0.9089780909734412, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5787448552834558, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23142080989284028, "3000–6000 Hz_10ms_envelope_correlation": 0.9696324536423707, "3000–6000 Hz_10ms_modulation_mae_db": 1.7603071738218725, "3000–6000 Hz_40ms_log_rms_mae_db": 1.371948897323564, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13720372726974195, "3000–6000 Hz_40ms_envelope_correlation": 0.9902120787000739, "3000–6000 Hz_40ms_modulation_mae_db": 2.1973997495728788, "mr_stft_db": 5.818468986388001, "band_mae_db": 0.3854633914621477, "rms_gain_error_db": -0.19467531426596738, "output_peak": 0.396568453643632}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2593615206879754, "全帯域_10ms_rms_envelope_nrmse": 0.09821428885397729, "全帯域_10ms_envelope_correlation": 0.9909603833897013, "全帯域_10ms_modulation_mae_db": 0.8701529869397855, "全帯域_40ms_log_rms_mae_db": 0.675642045423601, "全帯域_40ms_rms_envelope_nrmse": 0.053308098735753436, "全帯域_40ms_envelope_correlation": 0.9975536816594405, "全帯域_40ms_modulation_mae_db": 0.9361500559695841, "300–1000 Hz_10ms_log_rms_mae_db": 1.8275861776460025, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12713982516256364, "300–1000 Hz_10ms_envelope_correlation": 0.9890715472501316, "300–1000 Hz_10ms_modulation_mae_db": 1.102097666814985, "300–1000 Hz_40ms_log_rms_mae_db": 1.1825724744672867, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07225801075176655, "300–1000 Hz_40ms_envelope_correlation": 0.9974193953201358, "300–1000 Hz_40ms_modulation_mae_db": 0.7920210561805262, "1000–3000 Hz_10ms_log_rms_mae_db": 1.803763624324477, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1797314654477371, "1000–3000 Hz_10ms_envelope_correlation": 0.9777953031903093, "1000–3000 Hz_10ms_modulation_mae_db": 0.735283042191602, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1458070605201052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10394959316729371, "1000–3000 Hz_40ms_envelope_correlation": 0.9923967246439128, "1000–3000 Hz_40ms_modulation_mae_db": 0.6035292156174452, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5934175406727888, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20949715570231384, "3000–6000 Hz_10ms_envelope_correlation": 0.975806041677764, "3000–6000 Hz_10ms_modulation_mae_db": 1.2354342140745957, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4428057482651098, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1217070779743998, "3000–6000 Hz_40ms_envelope_correlation": 0.9917690331719796, "3000–6000 Hz_40ms_modulation_mae_db": 1.6055193386070696, "mr_stft_db": 5.812398790398603, "band_mae_db": 0.27479982921136425, "rms_gain_error_db": -0.21029699423237636, "output_peak": 0.415428216188438}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.307922996220721, "全帯域_10ms_rms_envelope_nrmse": 0.10600627435254491, "全帯域_10ms_envelope_correlation": 0.9890860757080956, "全帯域_10ms_modulation_mae_db": 0.08125223957313514, "全帯域_40ms_log_rms_mae_db": 0.6332014801846108, "全帯域_40ms_rms_envelope_nrmse": 0.052122646371103705, "全帯域_40ms_envelope_correlation": 0.9972180775312603, "全帯域_40ms_modulation_mae_db": 0.1318522713391654, "300–1000 Hz_10ms_log_rms_mae_db": 1.8850954618865725, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1328871632091492, "300–1000 Hz_10ms_envelope_correlation": 0.9868598109789058, "300–1000 Hz_10ms_modulation_mae_db": 0.21950195540318287, "300–1000 Hz_40ms_log_rms_mae_db": 1.2116284800147965, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06738399673297142, "300–1000 Hz_40ms_envelope_correlation": 0.9964504589528919, "300–1000 Hz_40ms_modulation_mae_db": 0.5218407488588391, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8644318133140212, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18624175853434313, "1000–3000 Hz_10ms_envelope_correlation": 0.9761239863632518, "1000–3000 Hz_10ms_modulation_mae_db": 1.8149743458728527, "1000–3000 Hz_40ms_log_rms_mae_db": 1.176601410562937, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09619564028118886, "1000–3000 Hz_40ms_envelope_correlation": 0.9935596623726577, "1000–3000 Hz_40ms_modulation_mae_db": 1.5726127328254806, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5897687519681776, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23341135443585126, "3000–6000 Hz_10ms_envelope_correlation": 0.969898003047646, "3000–6000 Hz_10ms_modulation_mae_db": 0.8256355516297971, "3000–6000 Hz_40ms_log_rms_mae_db": 1.420435350785129, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13879284592831287, "3000–6000 Hz_40ms_envelope_correlation": 0.9892330290337508, "3000–6000 Hz_40ms_modulation_mae_db": 1.0157420425040906, "mr_stft_db": 5.824678653084476, "band_mae_db": 0.17045950609392616, "rms_gain_error_db": -0.0857112644009676, "output_peak": 0.4029340208313238}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2670872059481255, "全帯域_10ms_rms_envelope_nrmse": 0.10783264557404612, "全帯域_10ms_envelope_correlation": 0.9889303711402514, "全帯域_10ms_modulation_mae_db": 0.5799509320207455, "全帯域_40ms_log_rms_mae_db": 0.7111177609466853, "全帯域_40ms_rms_envelope_nrmse": 0.06502871180159964, "全帯域_40ms_envelope_correlation": 0.995992905091533, "全帯域_40ms_modulation_mae_db": 0.6746407728166265, "300–1000 Hz_10ms_log_rms_mae_db": 1.9320382423575757, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14790858713948252, "300–1000 Hz_10ms_envelope_correlation": 0.9839563661012979, "300–1000 Hz_10ms_modulation_mae_db": 0.8936804259742813, "300–1000 Hz_40ms_log_rms_mae_db": 1.33116423709777, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09008272751522252, "300–1000 Hz_40ms_envelope_correlation": 0.99412987668936, "300–1000 Hz_40ms_modulation_mae_db": 0.6786592660391637, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8232270084050717, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1897809824417994, "1000–3000 Hz_10ms_envelope_correlation": 0.9766199994856507, "1000–3000 Hz_10ms_modulation_mae_db": 0.9591405606801077, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1900389523144912, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11684116083432583, "1000–3000 Hz_40ms_envelope_correlation": 0.9926137301408723, "1000–3000 Hz_40ms_modulation_mae_db": 0.640366394913285, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5966205412068084, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20642705664183222, "3000–6000 Hz_10ms_envelope_correlation": 0.9766612369331177, "3000–6000 Hz_10ms_modulation_mae_db": 0.32010162683134563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.40734451640921, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13753046465887814, "3000–6000 Hz_40ms_envelope_correlation": 0.9908322902153017, "3000–6000 Hz_40ms_modulation_mae_db": 0.2310578089358124, "mr_stft_db": 5.800273141122159, "band_mae_db": 0.3729107981850608, "rms_gain_error_db": -0.19728822086039077, "output_peak": 0.4430725886745638}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2782839294617214, "全帯域_10ms_rms_envelope_nrmse": 0.10286924315341464, "全帯域_10ms_envelope_correlation": 0.9896956997348907, "全帯域_10ms_modulation_mae_db": 0.2365036097679043, "全帯域_40ms_log_rms_mae_db": 0.6604200323461181, "全帯域_40ms_rms_envelope_nrmse": 0.05180193731132198, "全帯域_40ms_envelope_correlation": 0.9972210889079826, "全帯域_40ms_modulation_mae_db": 0.3102907364401079, "300–1000 Hz_10ms_log_rms_mae_db": 1.9443023511433863, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12468471007625191, "300–1000 Hz_10ms_envelope_correlation": 0.9884785251619012, "300–1000 Hz_10ms_modulation_mae_db": 0.6452702931086187, "300–1000 Hz_40ms_log_rms_mae_db": 1.3599285275067636, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06958151667827568, "300–1000 Hz_40ms_envelope_correlation": 0.9963118945712618, "300–1000 Hz_40ms_modulation_mae_db": 0.4651003955474639, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8200143663799682, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17707980568847997, "1000–3000 Hz_10ms_envelope_correlation": 0.9786553529092106, "1000–3000 Hz_10ms_modulation_mae_db": 0.7094595902677733, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0941819740158873, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09538496801539462, "1000–3000 Hz_40ms_envelope_correlation": 0.9940790359753682, "1000–3000 Hz_40ms_modulation_mae_db": 0.5962125786438582, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6046515505247299, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23234496322022516, "3000–6000 Hz_10ms_envelope_correlation": 0.9696679605229704, "3000–6000 Hz_10ms_modulation_mae_db": 0.505861475087917, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4285036801483004, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13747865307864918, "3000–6000 Hz_40ms_envelope_correlation": 0.9891913957235223, "3000–6000 Hz_40ms_modulation_mae_db": 0.5328938238304103, "mr_stft_db": 5.8011650080721795, "band_mae_db": 0.1697810202339007, "rms_gain_error_db": -0.07216213732435915, "output_peak": 0.4126314316400061}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2479903738690497, "全帯域_10ms_rms_envelope_nrmse": 0.10773696070825156, "全帯域_10ms_envelope_correlation": 0.988828050093541, "全帯域_10ms_modulation_mae_db": 0.39596597318551735, "全帯域_40ms_log_rms_mae_db": 0.6784055413145743, "全帯域_40ms_rms_envelope_nrmse": 0.060139295369413426, "全帯域_40ms_envelope_correlation": 0.9964618259036213, "全帯域_40ms_modulation_mae_db": 0.11454393855034391, "300–1000 Hz_10ms_log_rms_mae_db": 1.8860335000648327, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13102578141423318, "300–1000 Hz_10ms_envelope_correlation": 0.9876725258399597, "300–1000 Hz_10ms_modulation_mae_db": 1.0074054148320224, "300–1000 Hz_40ms_log_rms_mae_db": 1.2536267774274346, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07716442361725447, "300–1000 Hz_40ms_envelope_correlation": 0.9960229590659117, "300–1000 Hz_40ms_modulation_mae_db": 1.2172296210366327, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8867207432013375, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19757143438859698, "1000–3000 Hz_10ms_envelope_correlation": 0.973377930609439, "1000–3000 Hz_10ms_modulation_mae_db": 0.6018749122666728, "1000–3000 Hz_40ms_log_rms_mae_db": 1.226009156454444, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1267008075329038, "1000–3000 Hz_40ms_envelope_correlation": 0.9892440277007725, "1000–3000 Hz_40ms_modulation_mae_db": 1.0602431964709507, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6184095424200031, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19520886518425457, "3000–6000 Hz_10ms_envelope_correlation": 0.9785290192440397, "3000–6000 Hz_10ms_modulation_mae_db": 0.4280507003051948, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4602541417081583, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10931769976200639, "3000–6000 Hz_40ms_envelope_correlation": 0.9931089651068045, "3000–6000 Hz_40ms_modulation_mae_db": 0.6851396293514419, "mr_stft_db": 5.802898901594028, "band_mae_db": 0.24839690971401499, "rms_gain_error_db": -0.1562931193943643, "output_peak": 0.39654391112220966}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.238963422445742, "全帯域_10ms_rms_envelope_nrmse": 0.09910825195281991, "全帯域_10ms_envelope_correlation": 0.9909043090838151, "全帯域_10ms_modulation_mae_db": 0.5278674836878359, "全帯域_40ms_log_rms_mae_db": 0.6442542635905145, "全帯域_40ms_rms_envelope_nrmse": 0.05547807454986912, "全帯域_40ms_envelope_correlation": 0.9974542176510408, "全帯域_40ms_modulation_mae_db": 0.8195803891770789, "300–1000 Hz_10ms_log_rms_mae_db": 1.8227928728806908, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13016816282817867, "300–1000 Hz_10ms_envelope_correlation": 0.988646045653722, "300–1000 Hz_10ms_modulation_mae_db": 1.1975840158580173, "300–1000 Hz_40ms_log_rms_mae_db": 1.2237056738936072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0815815015650809, "300–1000 Hz_40ms_envelope_correlation": 0.9964464315694636, "300–1000 Hz_40ms_modulation_mae_db": 1.2066248407915847, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8768470566809252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2206424689366533, "1000–3000 Hz_10ms_envelope_correlation": 0.9683980729787607, "1000–3000 Hz_10ms_modulation_mae_db": 1.603839516533207, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2445609630531906, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14612088742506008, "1000–3000 Hz_40ms_envelope_correlation": 0.9882805684675514, "1000–3000 Hz_40ms_modulation_mae_db": 1.6235392978043957, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6260785955448493, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20067876576145266, "3000–6000 Hz_10ms_envelope_correlation": 0.9781622672646485, "3000–6000 Hz_10ms_modulation_mae_db": 0.5758778742401554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3903931240643228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11966901820605279, "3000–6000 Hz_40ms_envelope_correlation": 0.9923458515550698, "3000–6000 Hz_40ms_modulation_mae_db": 0.4655400142225621, "mr_stft_db": 5.8188020705981325, "band_mae_db": 0.3683466072498118, "rms_gain_error_db": -0.23614063072091226, "output_peak": 0.3944286477845814}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76640, "output_samples": 76640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9357701049499445, "spectral_convergence": 0.29967767952362273, "spectral_convergence_status": "ok", "amplitude_floor": 1.4500777065259659e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.619384355607419, "spectral_convergence": 0.29022985924907635, "spectral_convergence_status": "ok", "amplitude_floor": 1.1426990309647766e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.861254377267219, "spectral_convergence": 0.3126958666114974, "spectral_convergence_status": "ok", "amplitude_floor": 7.641152847432082e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003890931416418852, "output_power": 0.0004153564534575752, "delta_db": 0.2836738678357704, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013826376510160613, "output_power": 0.0012925460111910926, "delta_db": -0.29262367342773093, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006763873367355198, "output_power": 0.0006717527428349003, "delta_db": -0.029860198011105156, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00010128235718512909, "output_power": 0.00010004814675145573, "delta_db": -0.05324751937628619, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.238121331616846e-05, "output_power": 7.14170376368044e-05, "delta_db": -0.058240270612615386, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017231931162072514, "output_power": 0.00014784848837317156, "delta_db": -0.6651706245469523, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23046935896841006, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15044225481546847, "pitch": null, "output_rms_dbfs": -25.688021229335874, "output_peak": 0.40244264626301435, "reference_rms_dbfs": -25.537578974520407, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.805469612608195}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00018226521943209611, "log_rms_mae_db": 1.2315243958393338, "rms_envelope_nrmse": 0.10070968410137918, "envelope_correlation": 0.9902654174401705, "rms_gain_error_db": -0.15043842059073334, "reference_mean_power": 0.0027970153309965396, "output_mean_power": 0.0027017863597632364}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.029329446119484e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010840604762391139, "output_power": 0.0010545990209273241, "reference_active": true, "delta_db": -0.11966146963574455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001764324474949557, "output_power": 0.0001577491946112637, "reference_active": true, "delta_db": -0.48611308124985353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.741521666729804e-05, "output_power": 4.346546816380161e-05, "reference_active": true, "delta_db": -0.37773377695456045}], "active_band_count": 3, "active_band_mae_db": 0.32783610928005286}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001676463927845064, "log_rms_mae_db": 0.6728577880583877, "rms_envelope_nrmse": 0.06292818272123227, "envelope_correlation": 0.9960716554558857, "rms_gain_error_db": -0.1504472090380424, "reference_mean_power": 0.0028146416719843875, "output_mean_power": 0.002718807081258947}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.889944424347358e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010453379933036452, "output_power": 0.001015443923713281, "reference_active": true, "delta_db": -0.12600790144749058}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.750212405766423e-05, "output_power": 6.156459976734219e-05, "reference_active": true, "delta_db": -0.39986378326658156}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9228813706071663e-06, "output_power": 2.795782035298894e-06, "reference_active": true, "delta_db": -0.1930787930572342}], "active_band_count": 3, "active_band_mae_db": 0.2396501592571021}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00017586243391937327, "log_rms_mae_db": 1.8971741379708371, "rms_envelope_nrmse": 0.12398074671886529, "envelope_correlation": 0.9886991613787924, "rms_gain_error_db": -0.18017762260029907, "reference_mean_power": 0.0014437040494073058, "output_mean_power": 0.0013850339289029016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 1.5100208608044745e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009242573938618176, "output_power": 0.0008884465189249241, "reference_active": true, "delta_db": -0.17161643457218934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001371496531392223, "output_power": 0.00011475465759311072, "reference_active": true, "delta_db": -0.7742439213114821}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2803310723594984e-05, "output_power": 2.8810053443844297e-05, "reference_active": true, "delta_db": -0.5637361383369098}], "active_band_count": 3, "active_band_mae_db": 0.5031988314068604}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001539617967365963, "log_rms_mae_db": 1.2317691352925075, "rms_envelope_nrmse": 0.06720617201473919, "envelope_correlation": 0.9967199043001629, "rms_gain_error_db": -0.1801794750727374, "reference_mean_power": 0.0014528118102568712, "output_mean_power": 0.00139377096853515}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.4301118553576922e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009152950896245062, "output_power": 0.0008702824423713013, "reference_active": true, "delta_db": -0.2190091058608185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.331274162853222e-05, "output_power": 4.794703659498764e-05, "reference_active": true, "delta_db": -0.4606924644464132}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0543911334413026e-06, "output_power": 2.3111403911954214e-06, "reference_active": true, "delta_db": 0.5114319566272698}], "active_band_count": 3, "active_band_mae_db": 0.3970445089781672}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 8.12616340215947e-05, "log_rms_mae_db": 1.8599482168261166, "rms_envelope_nrmse": 0.18509464309669918, "envelope_correlation": 0.9764970846995527, "rms_gain_error_db": -0.05523638833703508, "reference_mean_power": 0.00015412407568599862, "output_mean_power": 0.00015217623893443248}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.103412885748503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011785014167056138, "output_power": 0.00011027876993821265, "reference_active": true, "delta_db": -0.28838195645445136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.031054512020625e-06, "output_power": 9.560797297666084e-06, "reference_active": true, "delta_db": 0.7572153688052969}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6793453202028428e-06, "output_power": 3.173460739656801e-06, "reference_active": true, "delta_db": 0.7350443973662762}], "active_band_count": 3, "active_band_mae_db": 0.5935472408753415}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 6.921465297386044e-05, "log_rms_mae_db": 1.2060745165018119, "rms_envelope_nrmse": 0.10464969259233829, "envelope_correlation": 0.992145271539563, "rms_gain_error_db": -0.05524392123304002, "reference_mean_power": 0.00015509566252518478, "output_mean_power": 0.00015313528113854806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.0373126402115645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011306875179582822, "output_power": 0.00010683161490616144, "reference_active": true, "delta_db": -0.24642804472758995}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.441071013913094e-06, "output_power": 4.6186723812572405e-06, "reference_active": true, "delta_db": 1.2782352198218154}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5029543761566567e-07, "output_power": 1.9356585096082252e-07, "reference_active": true, "delta_db": 1.0988294385977004}], "active_band_count": 3, "active_band_mae_db": 0.8744975677157019}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00013414824565394521, "log_rms_mae_db": 1.5776274462358952, "rms_envelope_nrmse": 0.24172433142195732, "envelope_correlation": 0.96775689076365, "rms_gain_error_db": -0.6742199779914432, "reference_mean_power": 0.0001730034207748911, "output_mean_power": 0.0001481264766071198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.1522167661145593e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001202536722048893, "output_power": 9.35508433185778e-05, "reference_active": true, "delta_db": -1.0905064054360223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5558902161138425e-05, "output_power": 2.0058401286098016e-05, "reference_active": true, "delta_db": -1.052458800022291}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.530968744127951e-05, "output_power": 8.924853737651462e-06, "reference_active": true, "delta_db": -2.3436521680729765}], "active_band_count": 3, "active_band_mae_db": 1.4955391245104301}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 8.410768197622129e-05, "log_rms_mae_db": 1.4089729327008031, "rms_envelope_nrmse": 0.18643767623866822, "envelope_correlation": 0.9819388696229864, "rms_gain_error_db": -0.6742244213216921, "reference_mean_power": 0.00017409446216015204, "output_mean_power": 0.00014906047971204367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.085538754133615e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012977220791713187, "output_power": 9.903998705809773e-05, "reference_active": true, "delta_db": -1.1737111876719761}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3304937813307329e-05, "output_power": 9.04841546163531e-06, "reference_active": true, "delta_db": -1.6744031583786942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.046486321108266e-06, "output_power": 6.035301528724786e-07, "reference_active": true, "delta_db": -2.390345832451225}], "active_band_count": 3, "active_band_mae_db": 1.7461533928339652}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8509299066988453, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "95a36a06d2f3ab26260a8403d1a3c46e3629c986c11a0af251d1e0822b2dcec5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9747801295534726, "standard_deviation": 0.003605532990874222, "minimum": 0.9683980729787607, "maximum": 0.9786553529092106, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8522508710791032, "standard_deviation": 0.03203168118912028, "minimum": 1.803763624324477, "maximum": 1.8867207432013375, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9940246630605118, "standard_deviation": 0.46487035496890117, "minimum": 0.5935472408753415, "maximum": 1.8149743458728527, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19357516471058223, "standard_deviation": 0.015625245410634234, "minimum": 0.17707980568847997, "maximum": 0.2206424689366533, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9915811742143019, "standard_deviation": 0.00207447015342108, "minimum": 0.9882805684675514, "maximum": 0.9940790359753682, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1868130377290957, "standard_deviation": 0.04812936778961741, "minimum": 1.0941819740158873, "maximum": 1.2445609630531906, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9849973843705697, "standard_deviation": 0.41267779812717403, "minimum": 0.5962125786438582, "maximum": 1.6235392978043957, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11559480893784252, "standard_deviation": 0.018785282121478574, "minimum": 0.09538496801539462, "maximum": 0.14612088742506008, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9732642341370259, "standard_deviation": 0.0044319370250975815, "minimum": 0.96775689076365, "maximum": 0.9785290192440397, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5981648529820887, "standard_deviation": 0.01742510295537523, "minimum": 1.5776274462358952, "maximum": 1.6260785955448493, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8933509675626636, "standard_deviation": 0.5387692830038626, "minimum": 0.32010162683134563, "maximum": 1.7603071738218725, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2188391627825909, "standard_deviation": 0.017753505337531644, "minimum": 0.19520886518425457, "maximum": 0.24172433142195732, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9898289391411861, "standard_deviation": 0.003485186202234116, "minimum": 0.9819388696229864, "maximum": 0.9931089651068045, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4163322989255747, "standard_deviation": 0.02821503394803601, "minimum": 1.371948897323564, "maximum": 1.4602541417081583, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.059930724982279, "standard_deviation": 0.7094186548843554, "minimum": 0.2310578089358124, "maximum": 2.1973997495728788, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13601714538958867, "standard_deviation": 0.023116686149478656, "minimum": 0.10931769976200639, "maximum": 0.18643767623866822, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9878382875499753, "standard_deviation": 0.0017558459316629252, "minimum": 0.9839563661012979, "maximum": 0.9893223180350917, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8915799810370688, "standard_deviation": 0.04700595113370164, "minimum": 1.8227928728806908, "maximum": 1.9443023511433863, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8447255394517085, "standard_deviation": 0.3556700741085812, "minimum": 0.21950195540318287, "maximum": 1.1975840158580173, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12998563899855942, "standard_deviation": 0.00815354730484583, "minimum": 0.12209013543975081, "maximum": 0.14790858713948252, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962749519276126, "standard_deviation": 0.0009572066776024018, "minimum": 0.99412987668936, "maximum": 0.9974193953201358, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2663121776587318, "standard_deviation": 0.06656097741663435, "minimum": 1.1825724744672867, "maximum": 1.3599285275067636, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8093835838688735, "standard_deviation": 0.35105170258631263, "minimum": 0.3970445089781672, "maximum": 1.2172296210366327, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07448702812814206, "standard_deviation": 0.008000341886183163, "minimum": 0.06720617201473919, "maximum": 0.09008272751522252, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.27757842763982954, "standard_deviation": 0.08875293079087844, "minimum": 0.1697810202339007, "maximum": 0.3854633914621477, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.810519395483222, "standard_deviation": 0.00934939211275839, "minimum": 5.800273141122159, "maximum": 5.824678653084476, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.40800623951847115, "standard_deviation": 0.016077343733095134, "minimum": 0.3944286477845814, "maximum": 0.4430725886745638, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1628762420018508, "standard_deviation": 0.05878919698927417, "minimum": -0.23614063072091226, "maximum": -0.07216213732435915, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899186926293606, "standard_deviation": 0.0008995601034227053, "minimum": 0.988828050093541, "maximum": 0.9909603833897013, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2631569680577992, "standard_deviation": 0.024491622519337414, "minimum": 1.2315243958393338, "maximum": 1.307922996220721, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5421110041680592, "standard_deviation": 0.39339439492765005, "minimum": 0.08125223957313514, "maximum": 1.3173586988894979, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10273074950660735, "standard_deviation": 0.003977599404870747, "minimum": 0.09821428885397729, "maximum": 0.10783264557404612, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9969302102730979, "standard_deviation": 0.0006497565197537297, "minimum": 0.995992905091533, "maximum": 0.9975536816594405, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6649425705805982, "standard_deviation": 0.02507193802077999, "minimum": 0.6332014801846108, "maximum": 0.7111177609466853, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5939158164981981, "standard_deviation": 0.4913566778435395, "minimum": 0.11454393855034391, "maximum": 1.5246182084355764, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05678251618072716, "standard_deviation": 0.005188105699190169, "minimum": 0.05180193731132198, "maximum": 0.06502871180159964, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0002", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.8682239055633545, "predicted_mos": 2.8682239055633545, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "608d7b1cd12fe451d8981359a004fc51be449c35cf3e5542ced31b97e763beef", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.7299736998975277, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1997366637279343, "全帯域_10ms_rms_envelope_nrmse": 0.10134160474281191, "全帯域_10ms_envelope_correlation": 0.9903249700363513, "全帯域_10ms_modulation_mae_db": 0.8504154789225614, "全帯域_40ms_log_rms_mae_db": 0.6519864326067074, "全帯域_40ms_rms_envelope_nrmse": 0.060703103699840666, "全帯域_40ms_envelope_correlation": 0.9966221648812225, "全帯域_40ms_modulation_mae_db": 1.146624028719348, "300–1000 Hz_10ms_log_rms_mae_db": 1.8246954004653828, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12861115246068974, "300–1000 Hz_10ms_envelope_correlation": 0.9880883811345957, "300–1000 Hz_10ms_modulation_mae_db": 0.8157981606249624, "300–1000 Hz_40ms_log_rms_mae_db": 1.2010689108328485, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08005283095891726, "300–1000 Hz_40ms_envelope_correlation": 0.9955659520675834, "300–1000 Hz_40ms_modulation_mae_db": 0.9759403245536188, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7913440379078909, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1891977817212109, "1000–3000 Hz_10ms_envelope_correlation": 0.9776245761512965, "1000–3000 Hz_10ms_modulation_mae_db": 1.1772272011305245, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1255505940522823, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1282212778194454, "1000–3000 Hz_40ms_envelope_correlation": 0.9915231902295648, "1000–3000 Hz_40ms_modulation_mae_db": 0.7003688237648936, "3000–6000 Hz_10ms_log_rms_mae_db": 1.579724463320163, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23296768508589255, "3000–6000 Hz_10ms_envelope_correlation": 0.9711864177849125, "3000–6000 Hz_10ms_modulation_mae_db": 1.7190022461780419, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2690868403007987, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14376267599170633, "3000–6000 Hz_40ms_envelope_correlation": 0.9920005668191323, "3000–6000 Hz_40ms_modulation_mae_db": 1.7383597317964312, "mr_stft_db": 5.789832928028219, "band_mae_db": 0.42744365168183823, "rms_gain_error_db": -0.2043642574551774, "output_peak": 0.4321052201801507}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2535681906888907, "全帯域_10ms_rms_envelope_nrmse": 0.10048964407689423, "全帯域_10ms_envelope_correlation": 0.990197060732371, "全帯域_10ms_modulation_mae_db": 0.8757371382576263, "全帯域_40ms_log_rms_mae_db": 0.7132535950134213, "全帯域_40ms_rms_envelope_nrmse": 0.05949083742171941, "全帯域_40ms_envelope_correlation": 0.9963374512464874, "全帯域_40ms_modulation_mae_db": 0.7967831685592747, "300–1000 Hz_10ms_log_rms_mae_db": 1.7853130767704422, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13047645230629873, "300–1000 Hz_10ms_envelope_correlation": 0.9873338534475962, "300–1000 Hz_10ms_modulation_mae_db": 1.226845904822013, "300–1000 Hz_40ms_log_rms_mae_db": 1.2055517558669364, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08145754400459952, "300–1000 Hz_40ms_envelope_correlation": 0.9947951560403216, "300–1000 Hz_40ms_modulation_mae_db": 1.0821028265127983, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8007063700380106, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20567350309374716, "1000–3000 Hz_10ms_envelope_correlation": 0.9723256889626487, "1000–3000 Hz_10ms_modulation_mae_db": 1.0577103982573228, "1000–3000 Hz_40ms_log_rms_mae_db": 1.154525374475116, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13722287548445736, "1000–3000 Hz_40ms_envelope_correlation": 0.989093073182332, "1000–3000 Hz_40ms_modulation_mae_db": 1.0260072103529554, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5765507638123601, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20903966320904585, "3000–6000 Hz_10ms_envelope_correlation": 0.9760823053619796, "3000–6000 Hz_10ms_modulation_mae_db": 1.0265674912488507, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2948400105823135, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12284891452648763, "3000–6000 Hz_40ms_envelope_correlation": 0.9933233719149064, "3000–6000 Hz_40ms_modulation_mae_db": 0.8394975892736344, "mr_stft_db": 5.796646346834702, "band_mae_db": 0.34424886635831226, "rms_gain_error_db": -0.0817959829446991, "output_peak": 0.4084778557066438}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2593986642356072, "全帯域_10ms_rms_envelope_nrmse": 0.10292039360106384, "全帯域_10ms_envelope_correlation": 0.989982567662722, "全帯域_10ms_modulation_mae_db": 0.510128039889964, "全帯域_40ms_log_rms_mae_db": 0.6997525852758648, "全帯域_40ms_rms_envelope_nrmse": 0.05813193608413078, "全帯域_40ms_envelope_correlation": 0.9969115017499871, "全帯域_40ms_modulation_mae_db": 0.4951246861544118, "300–1000 Hz_10ms_log_rms_mae_db": 1.8208202266058222, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1315776235242714, "300–1000 Hz_10ms_envelope_correlation": 0.9872394548150367, "300–1000 Hz_10ms_modulation_mae_db": 0.7186407272862899, "300–1000 Hz_40ms_log_rms_mae_db": 1.2180316879976811, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07804937032124702, "300–1000 Hz_40ms_envelope_correlation": 0.9954700133477788, "300–1000 Hz_40ms_modulation_mae_db": 0.8890783526322195, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8051045096337073, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19641526131289977, "1000–3000 Hz_10ms_envelope_correlation": 0.9738808933239341, "1000–3000 Hz_10ms_modulation_mae_db": 0.5808720062696519, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1215583258855963, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12012746736334962, "1000–3000 Hz_40ms_envelope_correlation": 0.990747693346622, "1000–3000 Hz_40ms_modulation_mae_db": 0.4817113677394915, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5756174752286833, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20406400034335156, "3000–6000 Hz_10ms_envelope_correlation": 0.977747504921359, "3000–6000 Hz_10ms_modulation_mae_db": 1.1995803391504565, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3523972305163336, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13670349304797771, "3000–6000 Hz_40ms_envelope_correlation": 0.9916124923730718, "3000–6000 Hz_40ms_modulation_mae_db": 1.4177989760879675, "mr_stft_db": 5.791703866494362, "band_mae_db": 0.31208817706285, "rms_gain_error_db": -0.19695246880175335, "output_peak": 0.4213604570199299}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1894425213892923, "全帯域_10ms_rms_envelope_nrmse": 0.09735140214111994, "全帯域_10ms_envelope_correlation": 0.9907733284357264, "全帯域_10ms_modulation_mae_db": 0.7866654104429044, "全帯域_40ms_log_rms_mae_db": 0.5761291414404818, "全帯域_40ms_rms_envelope_nrmse": 0.047289893881014754, "全帯域_40ms_envelope_correlation": 0.997662430024746, "全帯域_40ms_modulation_mae_db": 0.9940345345965275, "300–1000 Hz_10ms_log_rms_mae_db": 1.7971045065212916, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11730536535391545, "300–1000 Hz_10ms_envelope_correlation": 0.9897752683605602, "300–1000 Hz_10ms_modulation_mae_db": 1.1778916041356082, "300–1000 Hz_40ms_log_rms_mae_db": 1.1103663710315412, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06432393356951355, "300–1000 Hz_40ms_envelope_correlation": 0.9967535309585965, "300–1000 Hz_40ms_modulation_mae_db": 1.1488881646867677, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8274760663392202, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17076674240293255, "1000–3000 Hz_10ms_envelope_correlation": 0.9800686651879985, "1000–3000 Hz_10ms_modulation_mae_db": 0.5183918790291933, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1673062654092563, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09641499920911521, "1000–3000 Hz_40ms_envelope_correlation": 0.9936961536663694, "1000–3000 Hz_40ms_modulation_mae_db": 0.22799213767082072, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5458942318984779, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19228052889153988, "3000–6000 Hz_10ms_envelope_correlation": 0.9789225536981024, "3000–6000 Hz_10ms_modulation_mae_db": 0.7879061801426662, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2975349852302658, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11268765941590711, "3000–6000 Hz_40ms_envelope_correlation": 0.992750829250175, "3000–6000 Hz_40ms_modulation_mae_db": 1.0538112462119655, "mr_stft_db": 5.781054279143494, "band_mae_db": 0.24727391679599353, "rms_gain_error_db": -0.04413237253600211, "output_peak": 0.3945404742128976}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2116615795331194, "全帯域_10ms_rms_envelope_nrmse": 0.10019719320412765, "全帯域_10ms_envelope_correlation": 0.9902439412418172, "全帯域_10ms_modulation_mae_db": 0.22895527013130232, "全帯域_40ms_log_rms_mae_db": 0.6253847217607882, "全帯域_40ms_rms_envelope_nrmse": 0.04944773800632847, "全帯域_40ms_envelope_correlation": 0.9975117867337143, "全帯域_40ms_modulation_mae_db": 0.5014291104690902, "300–1000 Hz_10ms_log_rms_mae_db": 1.824624636194509, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12351065110140089, "300–1000 Hz_10ms_envelope_correlation": 0.9887880056677556, "300–1000 Hz_10ms_modulation_mae_db": 0.4745264535259292, "300–1000 Hz_40ms_log_rms_mae_db": 1.1773164903870057, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06172821329374016, "300–1000 Hz_40ms_envelope_correlation": 0.9972881341883657, "300–1000 Hz_40ms_modulation_mae_db": 0.5065309808461286, "1000–3000 Hz_10ms_log_rms_mae_db": 1.857174588998471, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1893513540474277, "1000–3000 Hz_10ms_envelope_correlation": 0.9767028080373996, "1000–3000 Hz_10ms_modulation_mae_db": 1.2986149544453252, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1604820558648419, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12642107144751782, "1000–3000 Hz_40ms_envelope_correlation": 0.99074260626449, "1000–3000 Hz_40ms_modulation_mae_db": 1.1139706370127807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5552473843374295, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20978951048259584, "3000–6000 Hz_10ms_envelope_correlation": 0.9761586121413707, "3000–6000 Hz_10ms_modulation_mae_db": 0.800259170422463, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3016661218260195, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11128868050934221, "3000–6000 Hz_40ms_envelope_correlation": 0.9934366102355253, "3000–6000 Hz_40ms_modulation_mae_db": 0.48257540344048006, "mr_stft_db": 5.778084321702896, "band_mae_db": 0.29038711104662157, "rms_gain_error_db": -0.09090827336019192, "output_peak": 0.42683582185286506}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.257050749481629, "全帯域_10ms_rms_envelope_nrmse": 0.1054626785379267, "全帯域_10ms_envelope_correlation": 0.9897196441576893, "全帯域_10ms_modulation_mae_db": 0.8927041289445118, "全帯域_40ms_log_rms_mae_db": 0.6347119660789611, "全帯域_40ms_rms_envelope_nrmse": 0.0626968965128752, "全帯域_40ms_envelope_correlation": 0.9966754100739399, "全帯域_40ms_modulation_mae_db": 0.9576124885696444, "300–1000 Hz_10ms_log_rms_mae_db": 1.8965078785706793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.14425213318500518, "300–1000 Hz_10ms_envelope_correlation": 0.9853902491627028, "300–1000 Hz_10ms_modulation_mae_db": 1.061785367681197, "300–1000 Hz_40ms_log_rms_mae_db": 1.2204576020536984, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08771911823181254, "300–1000 Hz_40ms_envelope_correlation": 0.9952949658013326, "300–1000 Hz_40ms_modulation_mae_db": 1.0964419788411426, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7923637510297774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17263691058614164, "1000–3000 Hz_10ms_envelope_correlation": 0.9796995633185479, "1000–3000 Hz_10ms_modulation_mae_db": 0.4216871441306878, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1471042841985528, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10144379960151294, "1000–3000 Hz_40ms_envelope_correlation": 0.9930867902533531, "1000–3000 Hz_40ms_modulation_mae_db": 0.33755935770509754, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5206839326737285, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2086512949275601, "3000–6000 Hz_10ms_envelope_correlation": 0.9754659320378841, "3000–6000 Hz_10ms_modulation_mae_db": 0.6844823712989734, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3068281242765727, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13267755095959374, "3000–6000 Hz_40ms_envelope_correlation": 0.9907015356355324, "3000–6000 Hz_40ms_modulation_mae_db": 0.7057675212308757, "mr_stft_db": 5.790924126727376, "band_mae_db": 0.3063192601382512, "rms_gain_error_db": -0.25868653679817083, "output_peak": 0.4130340057694775}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3186867261791937, "全帯域_10ms_rms_envelope_nrmse": 0.1040185257675488, "全帯域_10ms_envelope_correlation": 0.989645191770899, "全帯域_10ms_modulation_mae_db": 1.1498891129971165, "全帯域_40ms_log_rms_mae_db": 0.7491425250752045, "全帯域_40ms_rms_envelope_nrmse": 0.05411731413344254, "全帯域_40ms_envelope_correlation": 0.9972566475795342, "全帯域_40ms_modulation_mae_db": 0.8603141691062541, "300–1000 Hz_10ms_log_rms_mae_db": 1.9217427618053873, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13553814664178443, "300–1000 Hz_10ms_envelope_correlation": 0.9866592471906231, "300–1000 Hz_10ms_modulation_mae_db": 1.1563961509520178, "300–1000 Hz_40ms_log_rms_mae_db": 1.2737391059976777, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07875946945569685, "300–1000 Hz_40ms_envelope_correlation": 0.9957027913506002, "300–1000 Hz_40ms_modulation_mae_db": 1.232556483744563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.845725778463551, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18527965324320472, "1000–3000 Hz_10ms_envelope_correlation": 0.9766244335129639, "1000–3000 Hz_10ms_modulation_mae_db": 0.5569153007997566, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1464695816797315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10552485841618726, "1000–3000 Hz_40ms_envelope_correlation": 0.9927294495821637, "1000–3000 Hz_40ms_modulation_mae_db": 0.9037845033622478, "3000–6000 Hz_10ms_log_rms_mae_db": 1.542094852172097, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17037129382692034, "3000–6000 Hz_10ms_envelope_correlation": 0.9835746183344398, "3000–6000 Hz_10ms_modulation_mae_db": 0.5800421028615287, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2700568207360012, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08763854130502743, "3000–6000 Hz_40ms_envelope_correlation": 0.9955817184586353, "3000–6000 Hz_40ms_modulation_mae_db": 0.5087482177014132, "mr_stft_db": 5.771960956789854, "band_mae_db": 0.24640927393936105, "rms_gain_error_db": -0.1651940110660018, "output_peak": 0.40180877745651716}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.263160067075135, "全帯域_10ms_rms_envelope_nrmse": 0.1070498493783707, "全帯域_10ms_envelope_correlation": 0.9890828465351013, "全帯域_10ms_modulation_mae_db": 1.4570216936036886, "全帯域_40ms_log_rms_mae_db": 0.6467522399979224, "全帯域_40ms_rms_envelope_nrmse": 0.060891043722018735, "全帯域_40ms_envelope_correlation": 0.9965203586058489, "全帯域_40ms_modulation_mae_db": 1.6069979638302752, "300–1000 Hz_10ms_log_rms_mae_db": 1.7755631480533975, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13217619669156228, "300–1000 Hz_10ms_envelope_correlation": 0.9872318864314222, "300–1000 Hz_10ms_modulation_mae_db": 1.177424098141022, "300–1000 Hz_40ms_log_rms_mae_db": 1.1855180044124467, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08566277310621506, "300–1000 Hz_40ms_envelope_correlation": 0.9946285338704246, "300–1000 Hz_40ms_modulation_mae_db": 1.0889443680153679, "1000–3000 Hz_10ms_log_rms_mae_db": 1.819929176172741, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1885216058870205, "1000–3000 Hz_10ms_envelope_correlation": 0.9760859714987135, "1000–3000 Hz_10ms_modulation_mae_db": 0.5931525241274279, "1000–3000 Hz_40ms_log_rms_mae_db": 1.14469940251117, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1126850055547666, "1000–3000 Hz_40ms_envelope_correlation": 0.9920796822236907, "1000–3000 Hz_40ms_modulation_mae_db": 0.3459328840742371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5198642103308873, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24931015930161887, "3000–6000 Hz_10ms_envelope_correlation": 0.9662620601385519, "3000–6000 Hz_10ms_modulation_mae_db": 2.240445813591822, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3075415841244629, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14631283043702806, "3000–6000 Hz_40ms_envelope_correlation": 0.9915239103745597, "3000–6000 Hz_40ms_modulation_mae_db": 2.5339027918332655, "mr_stft_db": 5.787743686824203, "band_mae_db": 0.34468358615515493, "rms_gain_error_db": -0.19032596369606777, "output_peak": 0.4080846668693207}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76640, "output_samples": 76640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.941492581981919, "spectral_convergence": 0.3004060969249758, "spectral_convergence_status": "ok", "amplitude_floor": 1.4500777065259659e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.561273911833885, "spectral_convergence": 0.30038579612981553, "spectral_convergence_status": "ok", "amplitude_floor": 1.1426990309647766e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.866732290268854, "spectral_convergence": 0.31639767086312987, "spectral_convergence_status": "ok", "amplitude_floor": 7.641152847432082e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003890931416418852, "output_power": 0.00041419437600032224, "delta_db": 0.27150621904474326, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013826376510160613, "output_power": 0.0013228811493068322, "delta_db": -0.19187551186976443, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006763873367355198, "output_power": 0.0006310233008589387, "delta_db": -0.3015007209191353, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00010128235718512909, "output_power": 8.587562624081631e-05, "delta_db": -0.7166388316614363, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.238121331616846e-05, "output_power": 6.732611730139733e-05, "delta_db": -0.3144228943274013, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017231931162072514, "output_power": 0.00014436508729971407, "delta_db": -0.7687177322685488, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.42744365168183823, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.2043642574551774, "pitch": null, "output_rms_dbfs": -25.741943231975583, "output_peak": 0.4321052201801507, "reference_rms_dbfs": -25.537578974520407, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.789832928028219}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00018226521943209611, "log_rms_mae_db": 1.1997366637279343, "rms_envelope_nrmse": 0.10134160474281191, "envelope_correlation": 0.9903249700363513, "rms_gain_error_db": -0.20435939036707518, "reference_mean_power": 0.0027970153309965396, "output_mean_power": 0.0026684490055972378}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.029329446119484e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010840604762391139, "output_power": 0.0010123269749979685, "reference_active": true, "delta_db": -0.29732701324937927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001764324474949557, "output_power": 0.00013763145961854042, "reference_active": true, "delta_db": -1.0786074304583744}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.741521666729804e-05, "output_power": 3.6173167562645366e-05, "reference_active": true, "delta_db": -1.1753119930599303}], "active_band_count": 3, "active_band_mae_db": 0.8504154789225614}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001676463927845064, "log_rms_mae_db": 0.6519864326067074, "rms_envelope_nrmse": 0.060703103699840666, "envelope_correlation": 0.9966221648812225, "rms_gain_error_db": -0.2043610641978738, "reference_mean_power": 0.0028146416719843875, "output_mean_power": 0.0026852641073318168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.889944424347358e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010453379933036452, "output_power": 0.0009698248748974426, "reference_active": true, "delta_db": -0.325634162425168}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.750212405766423e-05, "output_power": 5.519022339214628e-05, "reference_active": true, "delta_db": -0.8745528683451658}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9228813706071663e-06, "output_power": 1.7451898710226292e-06, "reference_active": true, "delta_db": -2.239685055387711}], "active_band_count": 3, "active_band_mae_db": 1.146624028719348}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00017586243391937327, "log_rms_mae_db": 1.8246954004653828, "rms_envelope_nrmse": 0.12861115246068974, "envelope_correlation": 0.9880883811345957, "rms_gain_error_db": -0.25586799132857335, "reference_mean_power": 0.0014437040494073058, "output_mean_power": 0.0013611042043279878}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 1.5100208608044745e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009242573938618176, "output_power": 0.0008396118517737486, "reference_active": true, "delta_db": -0.4171437316443998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001371496531392223, "output_power": 0.00010679980695081005, "reference_active": true, "delta_db": -1.086242459104795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2803310723594984e-05, "output_power": 2.6394707197529705e-05, "reference_active": true, "delta_db": -0.9440082911256924}], "active_band_count": 3, "active_band_mae_db": 0.8157981606249624}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001539617967365963, "log_rms_mae_db": 1.2010689108328485, "rms_envelope_nrmse": 0.08005283095891726, "envelope_correlation": 0.9955659520675834, "rms_gain_error_db": -0.2558693378096284, "reference_mean_power": 0.0014528118102568712, "output_mean_power": 0.0013696904505951593}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.4301118553576922e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009152950896245062, "output_power": 0.0008219994778153019, "reference_active": true, "delta_db": -0.4668959081556087}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.331274162853222e-05, "output_power": 4.4145079515167154e-05, "reference_active": true, "delta_db": -0.819487134448213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0543911334413026e-06, "output_power": 1.4077947423165994e-06, "reference_active": true, "delta_db": -1.641437931057035}], "active_band_count": 3, "active_band_mae_db": 0.9759403245536188}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 8.12616340215947e-05, "log_rms_mae_db": 1.7913440379078909, "rms_envelope_nrmse": 0.1891977817212109, "envelope_correlation": 0.9776245761512965, "rms_gain_error_db": -0.6274249055302303, "reference_mean_power": 0.00015412407568599862, "output_mean_power": 0.00013339145031703405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.103412885748503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011785014167056138, "output_power": 9.305551893371606e-05, "reference_active": true, "delta_db": -1.0258797358301748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.031054512020625e-06, "output_power": 5.694948281873503e-06, "reference_active": true, "delta_db": -1.492827893279967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6793453202028428e-06, "output_power": 2.1219311712990548e-06, "reference_active": true, "delta_db": -1.0129739742814317}], "active_band_count": 3, "active_band_mae_db": 1.1772272011305245}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 6.921465297386044e-05, "log_rms_mae_db": 1.1255505940522823, "rms_envelope_nrmse": 0.1282212778194454, "envelope_correlation": 0.9915231902295648, "rms_gain_error_db": -0.6274277876824729, "reference_mean_power": 0.00015509566252518478, "output_mean_power": 0.00013423225112859367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.0373126402115645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011306875179582822, "output_power": 8.908336408247484e-05, "reference_active": true, "delta_db": -1.035459888043803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.441071013913094e-06, "output_power": 2.7661915118890273e-06, "reference_active": true, "delta_db": -0.948113909676833}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5029543761566567e-07, "output_power": 1.462825525527858e-07, "reference_active": true, "delta_db": -0.11753267357404477}], "active_band_count": 3, "active_band_mae_db": 0.7003688237648936}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00013414824565394521, "log_rms_mae_db": 1.579724463320163, "rms_envelope_nrmse": 0.23296768508589255, "envelope_correlation": 0.9711864177849125, "rms_gain_error_db": -0.7855647813085396, "reference_mean_power": 0.0001730034207748911, "output_mean_power": 0.00014437706613498657}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.1522167661145593e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001202536722048893, "output_power": 9.963317885525706e-05, "reference_active": true, "delta_db": -0.8169436040542464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5558902161138425e-05, "output_power": 1.2300651103217593e-05, "reference_active": true, "delta_db": -3.176140952061362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.530968744127951e-05, "output_power": 1.1710464074811477e-05, "reference_active": true, "delta_db": -1.1639221824185173}], "active_band_count": 3, "active_band_mae_db": 1.7190022461780419}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 8.410768197622129e-05, "log_rms_mae_db": 1.2690868403007987, "rms_envelope_nrmse": 0.14376267599170633, "envelope_correlation": 0.9920005668191323, "rms_gain_error_db": -0.7855687073188378, "reference_mean_power": 0.00017409446216015204, "output_mean_power": 0.00014528744485088217}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.085538754133615e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012977220791713187, "output_power": 0.0001036660343913568, "reference_active": true, "delta_db": -0.9754520820697982}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3304937813307329e-05, "output_power": 6.453881673514459e-06, "reference_active": true, "delta_db": -3.1419185030161296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.046486321108266e-06, "output_power": 8.127606985236439e-07, "reference_active": true, "delta_db": -1.097708610303366}], "active_band_count": 3, "active_band_mae_db": 1.7383597317964312}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8507983183339719, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6d934afc374a1b4504cf52aee0ad47250743f2eb110b34541ab16944b5af40d9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9766265749991878, "standard_deviation": 0.002635518521667444, "minimum": 0.9723256889626487, "maximum": 0.9800686651879985, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8174780348229211, "standard_deviation": 0.024575905949816866, "minimum": 1.7913440379078909, "maximum": 1.857174588998471, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7755714260237363, "standard_deviation": 0.3432772139323799, "minimum": 0.4216871441306878, "maximum": 1.2986149544453252, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1872303515368231, "standard_deviation": 0.01147900783574055, "minimum": 0.17076674240293255, "maximum": 0.20567350309374716, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9917123298435733, "standard_deviation": 0.0015005401576898229, "minimum": 0.989093073182332, "maximum": 0.9936961536663694, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1459619855095684, "standard_deviation": 0.01583424173110512, "minimum": 1.1215583258855963, "maximum": 1.1673062654092563, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6421658652103155, "standard_deviation": 0.3421759035974947, "minimum": 0.22799213767082072, "maximum": 1.1139706370127807, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11600766936204403, "standard_deviation": 0.014350596439364198, "minimum": 0.09641499920911521, "maximum": 0.13722287548445736, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9756750005523249, "standard_deviation": 0.005160843596762778, "minimum": 0.9662620601385519, "maximum": 0.9835746183344398, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5519596642217284, "standard_deviation": 0.024167572440349967, "minimum": 1.5198642103308873, "maximum": 1.579724463320163, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1297857143618504, "standard_deviation": 0.575977397054777, "minimum": 0.5800421028615287, "maximum": 2.240445813591822, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20955926700856561, "standard_deviation": 0.02388569469418422, "minimum": 0.17037129382692034, "maximum": 0.24931015930161887, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926163793826923, "standard_deviation": 0.0015227352760278342, "minimum": 0.9907015356355324, "maximum": 0.9955817184586353, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.299993964699096, "standard_deviation": 0.026012808449999883, "minimum": 1.2690868403007987, "maximum": 1.3523972305163336, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.160057684697004, "standard_deviation": 0.706290824931286, "minimum": 0.48257540344048006, "maximum": 2.5339027918332655, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12424004327413378, "standard_deviation": 0.019776142890690504, "minimum": 0.08763854130502743, "maximum": 0.14631283043702806, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9875632932762866, "standard_deviation": 0.0013349818618131266, "minimum": 0.9853902491627028, "maximum": 0.9897752683605602, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8307964543733641, "standard_deviation": 0.052124390424284094, "minimum": 1.7755631480533975, "maximum": 1.9217427618053873, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9761635583961299, "standard_deviation": 0.2745207210701731, "minimum": 0.4745264535259292, "maximum": 1.226845904822013, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.130430965158116, "standard_deviation": 0.00797276212261473, "minimum": 0.11730536535391545, "maximum": 0.14425213318500518, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9956873847031253, "standard_deviation": 0.0009128080674147799, "minimum": 0.9946285338704246, "maximum": 0.9972881341883657, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1990062410724796, "standard_deviation": 0.04623485445127244, "minimum": 1.1103663710315412, "maximum": 1.2737391059976777, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0025604349790758, "standard_deviation": 0.22568781845886882, "minimum": 0.5065309808461286, "maximum": 1.232556483744563, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07721915661771775, "standard_deviation": 0.009387342634614307, "minimum": 0.06172821329374016, "maximum": 0.08771911823181254, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.3148567303972979, "standard_deviation": 0.05899292980890487, "minimum": 0.24640927393936105, "maximum": 0.42744365168183823, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.785993814068139, "standard_deviation": 0.008212100347065964, "minimum": 5.771960956789854, "maximum": 5.796646346834702, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.4132809098834753, "standard_deviation": 0.012740903548625131, "minimum": 0.3945404742128976, "maximum": 0.4321052201801507, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.15404498333225802, "standard_deviation": 0.07372777237925086, "minimum": -0.25868653679817083, "maximum": -0.04413237253600211, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9899961938215847, "standard_deviation": 0.0005137855864758576, "minimum": 0.9890828465351013, "maximum": 0.9907733284357264, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2440881452888501, "standard_deviation": 0.0421319002400636, "minimum": 1.1894425213892923, "maximum": 1.3186867261791937, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8439395341487095, "standard_deviation": 0.371449280169184, "minimum": 0.22895527013130232, "maximum": 1.4570216936036886, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.10235391143123297, "standard_deviation": 0.003139265140877585, "minimum": 0.09735140214111994, "maximum": 0.1070498493783707, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.996937218861935, "standard_deviation": 0.0004869807950597016, "minimum": 0.9963374512464874, "maximum": 0.997662430024746, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.6621391509061689, "standard_deviation": 0.05534016744782971, "minimum": 0.5761291414404818, "maximum": 0.7491425250752045, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9198650187506032, "standard_deviation": 0.3595007900622373, "minimum": 0.4951246861544118, "maximum": 1.6069979638302752, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05659609543267132, "standard_deviation": 0.00569612609171517, "minimum": 0.047289893881014754, "maximum": 0.0626968965128752, "valid_count": 8, "planned_count": 8}}}]}, {"case": {"id": "librispeech_908-31957-0016", "language": "en", "style": "normal", "speaker": "librispeech_908", "source_utterance_id": "908-31957-0016", "speaker_gender_as_distributed": "M", "excerpt_start_sample": 0, "excerpt_samples": 96000, "kind": "natural_speech", "sr": 16000, "source_file": "audio/librispeech_908-31957-0016.wav", "source_file_sha256": "c34151bcb03f4127cf41442c82d28c1075768197b543e8188e677f6cfeda591a", "decoded_pcm_sha256": "a1a0fc0c8f705774ce9810918acda74ca07e00f2107a855ab611aa7ee1735b94", "original_source_sha256": "2a57896e33e6a87debcb741f00196495aa665bb2f0d7c205ac980a8417976d12", "corpus_entry_id": "librispeech_908-31957-0016"}, "original": {"method": "identity", "mos": {"status": "measured", "value": 3.5881359577178955, "predicted_mos": 3.5881359577178955, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f9cf81fa1b43b7486fc8af45ddc81003fab67c8d6aa84878c1f0b7b56caf6d9e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.7404068319992803, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "identity", "case": "librispeech_908-31957-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "f9cf81fa1b43b7486fc8af45ddc81003fab67c8d6aa84878c1f0b7b56caf6d9e", "measurement_environment": "historical_Linux_not_new_Windows"}}, "waveform_sha256": "e82f7cc1bcf52dc8247694fec36c8608076490306e8962865090852777303b82"}, "rows": [{"case": "librispeech_908-31957-0016", "method": "hnm", "status": "measured", "mos": {"status": "measured", "value": 2.5461785793304443, "predicted_mos": 2.5461785793304443, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "60875f38498b1f838cd08b9345912f980a5549dcbd78c7721687a2ead0d498a3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.989212537999265, "cpu_threads": 4, "reused_from": {"path": "experiments/2026-10-12/noise-envelope/results/run-001/results.json", "row_method": "candidate", "case": "librispeech_908-31957-0016", "results_sha256": "c50df39a5300056809a18c03e2793a1dbf00ab9a19810a3683b367351e8a0b21", "input_float32_sha256": "60875f38498b1f838cd08b9345912f980a5549dcbd78c7721687a2ead0d498a3", "measurement_environment": "historical_Linux_not_new_Windows"}}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4059876761830545, "全帯域_10ms_rms_envelope_nrmse": 0.0981011120893691, "全帯域_10ms_envelope_correlation": 0.9932524566502107, "全帯域_10ms_modulation_mae_db": 0.27473587640622354, "全帯域_40ms_log_rms_mae_db": 0.7240449753383732, "全帯域_40ms_rms_envelope_nrmse": 0.05001580836976093, "全帯域_40ms_envelope_correlation": 0.9984881874689653, "全帯域_40ms_modulation_mae_db": 0.503272567424196, "300–1000 Hz_10ms_log_rms_mae_db": 1.9660640351434835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12346654032716146, "300–1000 Hz_10ms_envelope_correlation": 0.9913486901258638, "300–1000 Hz_10ms_modulation_mae_db": 0.3618683310750425, "300–1000 Hz_40ms_log_rms_mae_db": 1.2090615510532392, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06997508333746985, "300–1000 Hz_40ms_envelope_correlation": 0.9979684062615616, "300–1000 Hz_40ms_modulation_mae_db": 0.9010179903450926, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8791711174648758, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2035732297963769, "1000–3000 Hz_10ms_envelope_correlation": 0.9712875566045869, "1000–3000 Hz_10ms_modulation_mae_db": 0.46986958255470146, "1000–3000 Hz_40ms_log_rms_mae_db": 1.151566486400229, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11380717349860321, "1000–3000 Hz_40ms_envelope_correlation": 0.9905832768262983, "1000–3000 Hz_40ms_modulation_mae_db": 0.30729077851970615, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7051296648918346, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1773046733481832, "3000–6000 Hz_10ms_envelope_correlation": 0.982173358540458, "3000–6000 Hz_10ms_modulation_mae_db": 0.8381651892269071, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2983195606202473, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10780807251129802, "3000–6000 Hz_40ms_envelope_correlation": 0.9940848632715326, "3000–6000 Hz_40ms_modulation_mae_db": 0.6550804432818612, "mr_stft_db": 5.909128455025859, "band_mae_db": 0.2092447068255303, "rms_gain_error_db": -0.20209450391375003, "output_peak": 0.6369810035476922}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4050749212838303, "全帯域_10ms_rms_envelope_nrmse": 0.09181973104359213, "全帯域_10ms_envelope_correlation": 0.9938851624381677, "全帯域_10ms_modulation_mae_db": 0.4521292633788388, "全帯域_40ms_log_rms_mae_db": 0.7036817158145812, "全帯域_40ms_rms_envelope_nrmse": 0.04466075316876635, "全帯域_40ms_envelope_correlation": 0.9985080471650651, "全帯域_40ms_modulation_mae_db": 0.6410841587116782, "300–1000 Hz_10ms_log_rms_mae_db": 1.907458301128903, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11606912182054631, "300–1000 Hz_10ms_envelope_correlation": 0.9916994236831911, "300–1000 Hz_10ms_modulation_mae_db": 0.9260886212942666, "300–1000 Hz_40ms_log_rms_mae_db": 1.198980811026478, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06314361242620962, "300–1000 Hz_40ms_envelope_correlation": 0.9975029707260857, "300–1000 Hz_40ms_modulation_mae_db": 0.9499724116946219, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9423011308693388, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21164012716416555, "1000–3000 Hz_10ms_envelope_correlation": 0.9686785343651619, "1000–3000 Hz_10ms_modulation_mae_db": 1.0382296197428735, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1637338599403435, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11465721702490889, "1000–3000 Hz_40ms_envelope_correlation": 0.9909825870467045, "1000–3000 Hz_40ms_modulation_mae_db": 0.7613928956843258, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7236145052226763, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16841885677640916, "3000–6000 Hz_10ms_envelope_correlation": 0.9835074501497898, "3000–6000 Hz_10ms_modulation_mae_db": 0.5168513532818179, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2973432767127546, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07661783717889006, "3000–6000 Hz_40ms_envelope_correlation": 0.9965561268840086, "3000–6000 Hz_40ms_modulation_mae_db": 0.17424785190906125, "mr_stft_db": 5.926604631011543, "band_mae_db": 0.22892393279596243, "rms_gain_error_db": -0.023246683712687156, "output_peak": 0.6576314128677838}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.456713907785052, "全帯域_10ms_rms_envelope_nrmse": 0.1042319207004773, "全帯域_10ms_envelope_correlation": 0.9921248405476999, "全帯域_10ms_modulation_mae_db": 0.37094817537443087, "全帯域_40ms_log_rms_mae_db": 0.7368398030768166, "全帯域_40ms_rms_envelope_nrmse": 0.05684875833053457, "全帯域_40ms_envelope_correlation": 0.9976316572283117, "全帯域_40ms_modulation_mae_db": 0.4280367328063693, "300–1000 Hz_10ms_log_rms_mae_db": 2.0385546269823878, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12530311265896504, "300–1000 Hz_10ms_envelope_correlation": 0.9903736260682922, "300–1000 Hz_10ms_modulation_mae_db": 0.8724374312740214, "300–1000 Hz_40ms_log_rms_mae_db": 1.2769925899019752, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07469151951233652, "300–1000 Hz_40ms_envelope_correlation": 0.9966588161295411, "300–1000 Hz_40ms_modulation_mae_db": 0.7131349160327138, "1000–3000 Hz_10ms_log_rms_mae_db": 1.847426367201928, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20719621335910032, "1000–3000 Hz_10ms_envelope_correlation": 0.9701647487877547, "1000–3000 Hz_10ms_modulation_mae_db": 0.53660186001256, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1208119975829829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11683475594932985, "1000–3000 Hz_40ms_envelope_correlation": 0.9900501185273086, "1000–3000 Hz_40ms_modulation_mae_db": 0.2933208090812749, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6614229098711601, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17177381998407196, "3000–6000 Hz_10ms_envelope_correlation": 0.9830698388062713, "3000–6000 Hz_10ms_modulation_mae_db": 1.0356390880251642, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2793798514280412, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08963593502407145, "3000–6000 Hz_40ms_envelope_correlation": 0.9959105613640438, "3000–6000 Hz_40ms_modulation_mae_db": 1.240229737665053, "mr_stft_db": 5.923573197725685, "band_mae_db": 0.1714334232838732, "rms_gain_error_db": -0.09068159006760822, "output_peak": 0.5489300931622776}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3880917369451402, "全帯域_10ms_rms_envelope_nrmse": 0.0950239994133573, "全帯域_10ms_envelope_correlation": 0.9937249317402717, "全帯域_10ms_modulation_mae_db": 0.3076051273081504, "全帯域_40ms_log_rms_mae_db": 0.721946979312864, "全帯域_40ms_rms_envelope_nrmse": 0.06023888237272892, "全帯域_40ms_envelope_correlation": 0.9976641402637231, "全帯域_40ms_modulation_mae_db": 0.3815484053826408, "300–1000 Hz_10ms_log_rms_mae_db": 1.9779342570414358, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12003390327257872, "300–1000 Hz_10ms_envelope_correlation": 0.9920076913759137, "300–1000 Hz_10ms_modulation_mae_db": 0.5031569335414455, "300–1000 Hz_40ms_log_rms_mae_db": 1.2880091405390017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0770797489583164, "300–1000 Hz_40ms_envelope_correlation": 0.9974321965039903, "300–1000 Hz_40ms_modulation_mae_db": 0.4507340395930943, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9313104771336702, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2256334186988044, "1000–3000 Hz_10ms_envelope_correlation": 0.9644321977873237, "1000–3000 Hz_10ms_modulation_mae_db": 0.3109198962869146, "1000–3000 Hz_40ms_log_rms_mae_db": 1.154005975967016, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13950801771308094, "1000–3000 Hz_40ms_envelope_correlation": 0.9866319284979755, "1000–3000 Hz_40ms_modulation_mae_db": 0.2662152780342721, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6736830833496907, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17202213488743656, "3000–6000 Hz_10ms_envelope_correlation": 0.9850622869404241, "3000–6000 Hz_10ms_modulation_mae_db": 1.4472437885205227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2630674760464078, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10012935159369535, "3000–6000 Hz_40ms_envelope_correlation": 0.9957367840899257, "3000–6000 Hz_40ms_modulation_mae_db": 1.350520370206904, "mr_stft_db": 5.901171455477218, "band_mae_db": 0.2674735565728448, "rms_gain_error_db": -0.2079647437168401, "output_peak": 0.540566675628449}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4230965643344744, "全帯域_10ms_rms_envelope_nrmse": 0.10330185508728422, "全帯域_10ms_envelope_correlation": 0.9922552842048037, "全帯域_10ms_modulation_mae_db": 0.653077365939471, "全帯域_40ms_log_rms_mae_db": 0.754560000730537, "全帯域_40ms_rms_envelope_nrmse": 0.053405245317498455, "全帯域_40ms_envelope_correlation": 0.9979014376071051, "全帯域_40ms_modulation_mae_db": 0.7127138070548605, "300–1000 Hz_10ms_log_rms_mae_db": 2.0169035269238096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11599479421407964, "300–1000 Hz_10ms_envelope_correlation": 0.9918956177027309, "300–1000 Hz_10ms_modulation_mae_db": 1.049275989015551, "300–1000 Hz_40ms_log_rms_mae_db": 1.2577005552823035, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05608276500334715, "300–1000 Hz_40ms_envelope_correlation": 0.9983768599114672, "300–1000 Hz_40ms_modulation_mae_db": 0.4599092010638184, "1000–3000 Hz_10ms_log_rms_mae_db": 1.877806253403034, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20855393115743168, "1000–3000 Hz_10ms_envelope_correlation": 0.9698777484691466, "1000–3000 Hz_10ms_modulation_mae_db": 0.1327094223337742, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1542570899526856, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12058142492200381, "1000–3000 Hz_40ms_envelope_correlation": 0.9894293255538691, "1000–3000 Hz_40ms_modulation_mae_db": 0.22540524009296958, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7060814578546442, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1563293678802443, "3000–6000 Hz_10ms_envelope_correlation": 0.9861542497559407, "3000–6000 Hz_10ms_modulation_mae_db": 0.7940832680471829, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2870617260664023, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10546685967595917, "3000–6000 Hz_40ms_envelope_correlation": 0.9940396213798465, "3000–6000 Hz_40ms_modulation_mae_db": 0.860767457383678, "mr_stft_db": 5.925725975020533, "band_mae_db": 0.2553565946412595, "rms_gain_error_db": -0.07783511396611789, "output_peak": 0.5611728099222824}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3991703101432615, "全帯域_10ms_rms_envelope_nrmse": 0.09053610611911521, "全帯域_10ms_envelope_correlation": 0.9940814063976791, "全帯域_10ms_modulation_mae_db": 0.20315323093068674, "全帯域_40ms_log_rms_mae_db": 0.6761908694162007, "全帯域_40ms_rms_envelope_nrmse": 0.04393649360368044, "全帯域_40ms_envelope_correlation": 0.998623237592188, "全帯域_40ms_modulation_mae_db": 0.2754002888194309, "300–1000 Hz_10ms_log_rms_mae_db": 1.9190526353694846, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1269469908484983, "300–1000 Hz_10ms_envelope_correlation": 0.9900675177863, "300–1000 Hz_10ms_modulation_mae_db": 0.724710150434471, "300–1000 Hz_40ms_log_rms_mae_db": 1.187833207143741, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06775913325324054, "300–1000 Hz_40ms_envelope_correlation": 0.9972123247840525, "300–1000 Hz_40ms_modulation_mae_db": 0.9455277750119823, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8576208518659638, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18090578928946416, "1000–3000 Hz_10ms_envelope_correlation": 0.9772053320398199, "1000–3000 Hz_10ms_modulation_mae_db": 0.6606091011738027, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0946021526016483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10101944779882464, "1000–3000 Hz_40ms_envelope_correlation": 0.9925815749770012, "1000–3000 Hz_40ms_modulation_mae_db": 0.4065006958728494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7065740412034236, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14577733062860226, "3000–6000 Hz_10ms_envelope_correlation": 0.9883389668794252, "3000–6000 Hz_10ms_modulation_mae_db": 0.4946781883358806, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2833193915014847, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07461198411380994, "3000–6000 Hz_40ms_envelope_correlation": 0.9971345451386708, "3000–6000 Hz_40ms_modulation_mae_db": 0.26452598687663836, "mr_stft_db": 5.9160713166014816, "band_mae_db": 0.1690165778106684, "rms_gain_error_db": -0.09343802364293705, "output_peak": 0.5970905783904203}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.391070733268462, "全帯域_10ms_rms_envelope_nrmse": 0.10299088677871161, "全帯域_10ms_envelope_correlation": 0.9923067742473679, "全帯域_10ms_modulation_mae_db": 0.8749108531585249, "全帯域_40ms_log_rms_mae_db": 0.754997248779359, "全帯域_40ms_rms_envelope_nrmse": 0.05704974020297972, "全帯域_40ms_envelope_correlation": 0.9975604885823995, "全帯域_40ms_modulation_mae_db": 1.1426926229691718, "300–1000 Hz_10ms_log_rms_mae_db": 1.9704133153811387, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12377303393181417, "300–1000 Hz_10ms_envelope_correlation": 0.9905312872161609, "300–1000 Hz_10ms_modulation_mae_db": 1.368241443704906, "300–1000 Hz_40ms_log_rms_mae_db": 1.219254103365691, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06679249274302461, "300–1000 Hz_40ms_envelope_correlation": 0.9972447443157039, "300–1000 Hz_40ms_modulation_mae_db": 1.3171363996211785, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8522811818514533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20007544632940735, "1000–3000 Hz_10ms_envelope_correlation": 0.972560355072258, "1000–3000 Hz_10ms_modulation_mae_db": 0.9008022246986901, "1000–3000 Hz_40ms_log_rms_mae_db": 1.190430180752129, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1247813560274414, "1000–3000 Hz_40ms_envelope_correlation": 0.989899332115833, "1000–3000 Hz_40ms_modulation_mae_db": 0.846359210224589, "3000–6000 Hz_10ms_log_rms_mae_db": 1.723378271543198, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18385522154382464, "3000–6000 Hz_10ms_envelope_correlation": 0.9809025461136391, "3000–6000 Hz_10ms_modulation_mae_db": 1.1365821085271461, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2906283362347168, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09169899817963499, "3000–6000 Hz_40ms_envelope_correlation": 0.9952583229443094, "3000–6000 Hz_40ms_modulation_mae_db": 1.0693252013823897, "mr_stft_db": 5.916537506024301, "band_mae_db": 0.17038865157130603, "rms_gain_error_db": -0.025275996395948845, "output_peak": 0.5383929065043945}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4047955843799045, "全帯域_10ms_rms_envelope_nrmse": 0.0967969935673573, "全帯域_10ms_envelope_correlation": 0.9936202020756447, "全帯域_10ms_modulation_mae_db": 0.6295042868652124, "全帯域_40ms_log_rms_mae_db": 0.7480143067689734, "全帯域_40ms_rms_envelope_nrmse": 0.05685334538119022, "全帯域_40ms_envelope_correlation": 0.998145590015803, "全帯域_40ms_modulation_mae_db": 0.8827579123141237, "300–1000 Hz_10ms_log_rms_mae_db": 2.0441907759911597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11533169115288303, "300–1000 Hz_10ms_envelope_correlation": 0.9927702363905347, "300–1000 Hz_10ms_modulation_mae_db": 0.6719277418309607, "300–1000 Hz_40ms_log_rms_mae_db": 1.2704472967727776, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06954766588606265, "300–1000 Hz_40ms_envelope_correlation": 0.9982016690477997, "300–1000 Hz_40ms_modulation_mae_db": 0.7695271789568349, "1000–3000 Hz_10ms_log_rms_mae_db": 1.845458973757698, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20563829130112915, "1000–3000 Hz_10ms_envelope_correlation": 0.9703901313801355, "1000–3000 Hz_10ms_modulation_mae_db": 0.6454093423741555, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1180269918015369, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11218521737812952, "1000–3000 Hz_40ms_envelope_correlation": 0.9910656671363601, "1000–3000 Hz_40ms_modulation_mae_db": 0.4739516047846409, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7003542054655911, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1651530142338695, "3000–6000 Hz_10ms_envelope_correlation": 0.9845892413164651, "3000–6000 Hz_10ms_modulation_mae_db": 0.3654179058571845, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2838557068621532, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09087069289991138, "3000–6000 Hz_40ms_envelope_correlation": 0.9960603772578018, "3000–6000 Hz_40ms_modulation_mae_db": 0.414852392612973, "mr_stft_db": 5.893050274616896, "band_mae_db": 0.2588832636339341, "rms_gain_error_db": -0.2508643360012344, "output_peak": 0.5531752528446299}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.171985847427049, "spectral_convergence": 0.34341336576963705, "spectral_convergence_status": "ok", "amplitude_floor": 2.3135422827537554e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.736627592022428, "spectral_convergence": 0.3074883245143213, "spectral_convergence_status": "ok", "amplitude_floor": 1.7511633455315343e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.8187719256281, "spectral_convergence": 0.3407771727116589, "spectral_convergence_status": "ok", "amplitude_floor": 1.6157176909573571e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00044579673236332844, "output_power": 0.0004831875524921701, "delta_db": 0.3497885696587338, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00284687947566611, "output_power": 0.002645179326145905, "delta_db": -0.3191396167926864, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00047388905696308834, "output_power": 0.0004655306790267565, "delta_db": -0.07728373310846222, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013682287067652582, "output_power": 0.00013303316606507183, "delta_db": -0.12198771095787063, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011121195656671286, "output_power": 0.00010807467709680418, "delta_db": -0.12427534784985919, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00022124255174898303, "output_power": 0.00020824248257397847, "delta_db": -0.26299326258556976, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2092447068255303, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.20209450391375003, "pitch": null, "output_rms_dbfs": -23.932696321283665, "output_peak": 0.6369810035476922, "reference_rms_dbfs": -23.730601817369916, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.909128455025859}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028051035499693775, "log_rms_mae_db": 1.4059876761830545, "rms_envelope_nrmse": 0.0981011120893691, "envelope_correlation": 0.9932524566502107, "rms_gain_error_db": -0.20209182514079554, "reference_mean_power": 0.004239368434074771, "output_mean_power": 0.004046615859303783}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.0558356273882855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018964431225846456, "output_power": 0.0017826113238626003, "reference_active": true, "delta_db": -0.2688316102695494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001936564415781366, "output_power": 0.00017929389661256205, "reference_active": true, "delta_db": -0.3346644157464928}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.246230567335058e-05, "output_power": 5.9367238916978685e-05, "reference_active": true, "delta_db": -0.22071160320262834}], "active_band_count": 3, "active_band_mae_db": 0.27473587640622354}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026381677718961605, "log_rms_mae_db": 0.7240449753383732, "rms_envelope_nrmse": 0.05001580836976093, "envelope_correlation": 0.9984881874689653, "rms_gain_error_db": -0.20208588027831037, "reference_mean_power": 0.004260663575263385, "output_mean_power": 0.004066948335288479}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.853150694908015e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017547087963691286, "output_power": 0.0016582272076222607, "reference_active": true, "delta_db": -0.24561016588068782}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.102196929549433e-05, "output_power": 7.271133084924882e-05, "reference_active": true, "delta_db": -0.47000701115550747}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0378089105429984e-06, "output_power": 2.5301158024571793e-06, "reference_active": true, "delta_db": -0.7942005252363928}], "active_band_count": 3, "active_band_mae_db": 0.503272567424196}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002736722962808348, "log_rms_mae_db": 1.9660640351434835, "rms_envelope_nrmse": 0.12346654032716146, "envelope_correlation": 0.9913486901258638, "rms_gain_error_db": -0.3770579532060129, "reference_mean_power": 0.0024581880191069405, "output_mean_power": 0.0022537684879900804}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.077905340514435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019022535341174436, "output_power": 0.0017637869260673387, "reference_active": true, "delta_db": -0.3282228052583962}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012849688146439824, "output_power": 0.0001126375748450908, "reference_active": true, "delta_db": -0.572092964744498}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.095154000220668e-05, "output_power": 3.9241112919389645e-05, "reference_active": true, "delta_db": -0.1852892232222333}], "active_band_count": 3, "active_band_mae_db": 0.3618683310750425}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025339024948019116, "log_rms_mae_db": 1.2090615510532392, "rms_envelope_nrmse": 0.06997508333746985, "envelope_correlation": 0.9979684062615616, "rms_gain_error_db": -0.3770594653234284, "reference_mean_power": 0.002470549817529662, "output_mean_power": 0.0022651015076178616}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.981046082661891e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017987268874748684, "output_power": 0.00167609216636552, "reference_active": true, "delta_db": -0.30667330264655157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5079379851283876e-05, "output_power": 4.7111210243774433e-05, "reference_active": true, "delta_db": -0.6786478080377861}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4721075346991203e-06, "output_power": 1.6645393723975575e-06, "reference_active": true, "delta_db": -1.7177328603509399}], "active_band_count": 3, "active_band_mae_db": 0.9010179903450926}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.470771529403075e-05, "log_rms_mae_db": 1.8791711174648758, "rms_envelope_nrmse": 0.2035732297963769, "envelope_correlation": 0.9712875566045869, "rms_gain_error_db": -0.005442224170396769, "reference_mean_power": 0.00019492290635367423, "output_mean_power": 0.00019467879784869322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7570632979474236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.555963644612216e-05, "output_power": 7.098497815590492e-05, "reference_active": true, "delta_db": -0.2712340662897513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.293127801504832e-05, "output_power": 1.1862925823612965e-05, "reference_active": true, "delta_db": -0.3744963403282965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505165714035109e-06, "output_power": 7.756149980973692e-06, "reference_active": true, "delta_db": 0.7638783410460565}], "active_band_count": 3, "active_band_mae_db": 0.46986958255470146}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.885515861839929e-05, "log_rms_mae_db": 1.151566486400229, "rms_envelope_nrmse": 0.11380717349860321, "envelope_correlation": 0.9905832768262983, "rms_gain_error_db": -0.005453095238252438, "reference_mean_power": 0.00019590175284412906, "output_mean_power": 0.00019565592873889786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6226758883463992e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.225076970752032e-05, "output_power": 6.72678831343261e-05, "reference_active": true, "delta_db": -0.310347171099318}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.071619770568594e-06, "output_power": 5.838209918659901e-06, "reference_active": true, "delta_db": -0.170248600348881}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6400879277669946e-07, "output_power": 4.0293921010867324e-07, "reference_active": true, "delta_db": 0.44127656411091953}], "active_band_count": 3, "active_band_mae_db": 0.30729077851970615}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012408236865898127, "log_rms_mae_db": 1.7051296648918346, "rms_envelope_nrmse": 0.1773046733481832, "envelope_correlation": 0.982173358540458, "rms_gain_error_db": -0.3897554685075518, "reference_mean_power": 0.00018911407706020183, "output_mean_power": 0.00017288141588223314}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8583468534278485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010082037892943086, "output_power": 9.310377205543534e-05, "reference_active": true, "delta_db": -0.34581048809947135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9440078487669303e-05, "output_power": 1.6944279343346993e-05, "reference_active": true, "delta_db": -0.5967491142078475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.009136082695238e-06, "output_power": 3.4879418328391088e-06, "reference_active": true, "delta_db": -1.5719359653734024}], "active_band_count": 3, "active_band_mae_db": 0.8381651892269071}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.970743476594613e-05, "log_rms_mae_db": 1.2983195606202473, "rms_envelope_nrmse": 0.10780807251129802, "envelope_correlation": 0.9940848632715326, "rms_gain_error_db": -0.38975726547840095, "reference_mean_power": 0.00019006483009233416, "output_mean_power": 0.00017375048886211734}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7984096251410197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010012152579842322, "output_power": 9.156093584624511e-05, "reference_active": true, "delta_db": -0.3881723644369529}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0693666398121268e-05, "output_power": 9.40829490258811e-06, "reference_active": true, "delta_db": -0.5561570974636594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.760055744312906e-07, "output_power": 2.1818589262377396e-07, "reference_active": true, "delta_db": -1.0209118679449711}], "active_band_count": 3, "active_band_mae_db": 0.6550804432818612}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.782525308547355, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "59297575fcfc585deadf8564b350f5e98cb7e4e0009e5cb853e41b01f91669e5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9705745755632734, "standard_deviation": 0.0035942397623338076, "minimum": 0.9644321977873237, "maximum": 0.9772053320398199, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8791720441934952, "standard_deviation": 0.03786723923963876, "minimum": 1.845458973757698, "maximum": 1.9423011308693388, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.586893881147184, "standard_deviation": 0.29514391810226587, "minimum": 0.1327094223337742, "maximum": 1.0382296197428735, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20540205588698493, "standard_deviation": 0.012496943351270984, "minimum": 0.18090578928946416, "maximum": 0.2256334186988044, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9901529763351687, "standard_deviation": 0.0017155120747699142, "minimum": 0.9866319284979755, "maximum": 0.9925815749770012, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1434293418748214, "standard_deviation": 0.030372893957814482, "minimum": 1.0946021526016483, "maximum": 1.190430180752129, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4475545640368285, "standard_deviation": 0.23470897102448568, "minimum": 0.22540524009296958, "maximum": 0.846359210224589, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11792182628904027, "standard_deviation": 0.01113205870381206, "minimum": 0.10101944779882464, "maximum": 0.13950801771308094, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9842247423128017, "standard_deviation": 0.0023475718887014186, "minimum": 0.9809025461136391, "maximum": 0.9883389668794252, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7000297674252773, "standard_deviation": 0.022011667049372137, "minimum": 1.6614229098711601, "maximum": 1.7236145052226763, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8285826112277257, "standard_deviation": 0.36741371053020777, "minimum": 0.3654179058571845, "maximum": 1.4472437885205227, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1675793024103302, "standard_deviation": 0.011983916888696291, "minimum": 0.14577733062860226, "maximum": 0.18385522154382464, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955976502912673, "standard_deviation": 0.0010982604852386633, "minimum": 0.9940396213798465, "maximum": 0.9971345451386708, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.285371915684026, "standard_deviation": 0.011220803489922375, "minimum": 1.2630674760464078, "maximum": 1.2983195606202473, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7536936801648197, "standard_deviation": 0.447786296738379, "minimum": 0.17424785190906125, "maximum": 1.350520370206904, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09210496639715879, "standard_deviation": 0.012207482770697936, "minimum": 0.07461198411380994, "maximum": 0.10780807251129802, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913367612936235, "standard_deviation": 0.0009362830656570991, "minimum": 0.9900675177863, "maximum": 0.9927702363905347, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9800714342452252, "standard_deviation": 0.05095963641233742, "minimum": 1.907458301128903, "maximum": 2.0441907759911597, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8097133302713331, "standard_deviation": 0.31793391469946974, "minimum": 0.3618683310750425, "maximum": 1.368241443704906, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12086489852831583, "standard_deviation": 0.00462817011775599, "minimum": 0.11533169115288303, "maximum": 0.1269469908484983, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975747484600253, "standard_deviation": 0.0005727175147712212, "minimum": 0.9966588161295411, "maximum": 0.9983768599114672, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2385349068856508, "standard_deviation": 0.039075079434517955, "minimum": 1.187833207143741, "maximum": 1.2880091405390017, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8133699890399171, "standard_deviation": 0.2843532242916788, "minimum": 0.4507340395930943, "maximum": 1.3171363996211785, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06813400264000091, "standard_deviation": 0.006551901856411065, "minimum": 0.05608276500334715, "maximum": 0.0770797489583164, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.21634008839192234, "standard_deviation": 0.04229690369972033, "minimum": 0.1690165778106684, "maximum": 0.2674735565728448, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.91398285143794, "standard_deviation": 0.012106331065965882, "minimum": 5.893050274616896, "maximum": 5.926604631011543, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5792425916084912, "standard_deviation": 0.04611550113777689, "minimum": 0.5383929065043945, "maximum": 0.6576314128677838, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12142512392714047, "standard_deviation": 0.08722987426988262, "minimum": -0.2508643360012344, "maximum": -0.023246683712687156, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931563822877307, "standard_deviation": 0.0008046549934295292, "minimum": 0.9921248405476999, "maximum": 0.9940814063976791, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4092501792903973, "standard_deviation": 0.021948718561136848, "minimum": 1.3880917369451402, "maximum": 1.456713907785052, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4707580224201923, "standard_deviation": 0.22951661616181201, "minimum": 0.20315323093068674, "maximum": 0.8749108531585249, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09785032559990803, "standard_deviation": 0.005290094493824102, "minimum": 0.09053610611911521, "maximum": 0.1042319207004773, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980653482404451, "standard_deviation": 0.00043478407234764227, "minimum": 0.9975604885823995, "maximum": 0.998623237592188, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7275344874047132, "standard_deviation": 0.027357715797543605, "minimum": 0.6761908694162007, "maximum": 0.754997248779359, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6209383119353089, "standard_deviation": 0.28719479708569046, "minimum": 0.2754002888194309, "maximum": 1.1426926229691718, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05287612834339245, "standard_deviation": 0.006083937645198669, "minimum": 0.04393649360368044, "maximum": 0.06023888237272892, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0016", "method": "A", "status": "measured", "mos": {"status": "measured", "value": 2.291398763656616, "predicted_mos": 2.291398763656616, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e70c00f266bfa1e884ad15209b61bd5d7932cf4f8df3f08ef291acbdfe752820", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 3.095109600108117, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.415692389832015, "全帯域_10ms_rms_envelope_nrmse": 0.09859399494599543, "全帯域_10ms_envelope_correlation": 0.993162615966573, "全帯域_10ms_modulation_mae_db": 0.15422687550328695, "全帯域_40ms_log_rms_mae_db": 0.7200382889030217, "全帯域_40ms_rms_envelope_nrmse": 0.05030757831452972, "全帯域_40ms_envelope_correlation": 0.9984409340809555, "全帯域_40ms_modulation_mae_db": 0.4071916882908921, "300–1000 Hz_10ms_log_rms_mae_db": 1.9331794175867518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12389777062758463, "300–1000 Hz_10ms_envelope_correlation": 0.9912579746073473, "300–1000 Hz_10ms_modulation_mae_db": 0.25279708769549186, "300–1000 Hz_40ms_log_rms_mae_db": 1.1381148682483873, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07040795936806118, "300–1000 Hz_40ms_envelope_correlation": 0.9978999908655819, "300–1000 Hz_40ms_modulation_mae_db": 0.6058445769145478, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8444350304841433, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20972373599537908, "1000–3000 Hz_10ms_envelope_correlation": 0.9693314362167601, "1000–3000 Hz_10ms_modulation_mae_db": 0.4827479028551429, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0428193917975297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11270865573741039, "1000–3000 Hz_40ms_envelope_correlation": 0.9907462751557929, "1000–3000 Hz_40ms_modulation_mae_db": 0.2607387298613099, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6734545069833389, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18408766753201342, "3000–6000 Hz_10ms_envelope_correlation": 0.9810729982710708, "3000–6000 Hz_10ms_modulation_mae_db": 1.0171425690959455, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2375836156375724, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11585805832867355, "3000–6000 Hz_40ms_envelope_correlation": 0.9934962371452437, "3000–6000 Hz_40ms_modulation_mae_db": 0.8946414532550447, "mr_stft_db": 5.867569099039219, "band_mae_db": 0.23966687377172927, "rms_gain_error_db": -0.19595111373580967, "output_peak": 0.6420696376278592}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4262792848085029, "全帯域_10ms_rms_envelope_nrmse": 0.09289976784018815, "全帯域_10ms_envelope_correlation": 0.9937447991356, "全帯域_10ms_modulation_mae_db": 0.33009474190247223, "全帯域_40ms_log_rms_mae_db": 0.7066244376845405, "全帯域_40ms_rms_envelope_nrmse": 0.047094773359999766, "全帯域_40ms_envelope_correlation": 0.9983405425178883, "全帯域_40ms_modulation_mae_db": 0.45621428623878746, "300–1000 Hz_10ms_log_rms_mae_db": 1.9193416002520691, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11816346272948382, "300–1000 Hz_10ms_envelope_correlation": 0.9914123222964883, "300–1000 Hz_10ms_modulation_mae_db": 0.7181419578050896, "300–1000 Hz_40ms_log_rms_mae_db": 1.1696744508481538, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06991615512226498, "300–1000 Hz_40ms_envelope_correlation": 0.9969386256850753, "300–1000 Hz_40ms_modulation_mae_db": 0.7098518033333688, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8940606768668533, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2074537251620182, "1000–3000 Hz_10ms_envelope_correlation": 0.9699678701369578, "1000–3000 Hz_10ms_modulation_mae_db": 1.065887514401556, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0565716262212175, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11248624169546526, "1000–3000 Hz_40ms_envelope_correlation": 0.9913416952658273, "1000–3000 Hz_40ms_modulation_mae_db": 0.7307873392818718, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6826847127153615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1730434536800711, "3000–6000 Hz_10ms_envelope_correlation": 0.9825902603381828, "3000–6000 Hz_10ms_modulation_mae_db": 0.4770670948122568, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2103156744221213, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07801081824084498, "3000–6000 Hz_40ms_envelope_correlation": 0.996393324799751, "3000–6000 Hz_40ms_modulation_mae_db": 0.35415167752813126, "mr_stft_db": 5.885159314448884, "band_mae_db": 0.2288857756453266, "rms_gain_error_db": -0.016496363780867534, "output_peak": 0.6590290810367669}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4716102716222583, "全帯域_10ms_rms_envelope_nrmse": 0.1033769181345634, "全帯域_10ms_envelope_correlation": 0.9922594037690827, "全帯域_10ms_modulation_mae_db": 0.4789218227621044, "全帯域_40ms_log_rms_mae_db": 0.752252863900696, "全帯域_40ms_rms_envelope_nrmse": 0.05723364579742491, "全帯域_40ms_envelope_correlation": 0.9976006418017721, "全帯域_40ms_modulation_mae_db": 0.6232014383680496, "300–1000 Hz_10ms_log_rms_mae_db": 2.0207080347893385, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12499553992222104, "300–1000 Hz_10ms_envelope_correlation": 0.9903998094066753, "300–1000 Hz_10ms_modulation_mae_db": 0.9391132900855862, "300–1000 Hz_40ms_log_rms_mae_db": 1.2497331944765648, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07652923052307942, "300–1000 Hz_40ms_envelope_correlation": 0.9964383859353918, "300–1000 Hz_40ms_modulation_mae_db": 0.8897949896886675, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8326327216446392, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19997434057386504, "1000–3000 Hz_10ms_envelope_correlation": 0.9720690034281212, "1000–3000 Hz_10ms_modulation_mae_db": 0.6562571749509379, "1000–3000 Hz_40ms_log_rms_mae_db": 1.059793052818498, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1094931399221386, "1000–3000 Hz_40ms_envelope_correlation": 0.991601350650013, "1000–3000 Hz_40ms_modulation_mae_db": 0.4447867518063994, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6309228772956037, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17404474924649602, "3000–6000 Hz_10ms_envelope_correlation": 0.9824317615446986, "3000–6000 Hz_10ms_modulation_mae_db": 0.8659963442489582, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2128315860775603, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08913444616533323, "3000–6000 Hz_40ms_envelope_correlation": 0.9956816330378466, "3000–6000 Hz_40ms_modulation_mae_db": 1.0207257590505079, "mr_stft_db": 5.885138079830907, "band_mae_db": 0.19623185225622977, "rms_gain_error_db": -0.09119603527346018, "output_peak": 0.5447946077240735}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.424662034833464, "全帯域_10ms_rms_envelope_nrmse": 0.09270745731961971, "全帯域_10ms_envelope_correlation": 0.9939638113890743, "全帯域_10ms_modulation_mae_db": 0.36695538338688954, "全帯域_40ms_log_rms_mae_db": 0.7335340701037855, "全帯域_40ms_rms_envelope_nrmse": 0.056396054349035066, "全帯域_40ms_envelope_correlation": 0.9979053416552126, "全帯域_40ms_modulation_mae_db": 0.44515230734317973, "300–1000 Hz_10ms_log_rms_mae_db": 1.9413208768218957, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1169670763590138, "300–1000 Hz_10ms_envelope_correlation": 0.9923330623442105, "300–1000 Hz_10ms_modulation_mae_db": 0.5225147110680516, "300–1000 Hz_40ms_log_rms_mae_db": 1.19401683675858, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07223784397780486, "300–1000 Hz_40ms_envelope_correlation": 0.9977409732250735, "300–1000 Hz_40ms_modulation_mae_db": 0.50625835228062, "1000–3000 Hz_10ms_log_rms_mae_db": 1.902862622262604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.217104215411044, "1000–3000 Hz_10ms_envelope_correlation": 0.9671941195386634, "1000–3000 Hz_10ms_modulation_mae_db": 0.4862902269409261, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0735802127663843, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1336931980704149, "1000–3000 Hz_40ms_envelope_correlation": 0.9878042816763153, "1000–3000 Hz_40ms_modulation_mae_db": 0.37967788317017215, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6245451589241884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.181454146903685, "3000–6000 Hz_10ms_envelope_correlation": 0.9837112449710576, "3000–6000 Hz_10ms_modulation_mae_db": 1.502008247100096, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2001013671976144, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10555275051383017, "3000–6000 Hz_40ms_envelope_correlation": 0.9955105820183121, "3000–6000 Hz_40ms_modulation_mae_db": 1.4002289908999697, "mr_stft_db": 5.858903461728748, "band_mae_db": 0.2760977822639888, "rms_gain_error_db": -0.18108480710414, "output_peak": 0.5251881740664939}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4705838881721371, "全帯域_10ms_rms_envelope_nrmse": 0.10717642206457817, "全帯域_10ms_envelope_correlation": 0.9916529569964266, "全帯域_10ms_modulation_mae_db": 0.7532665670058312, "全帯域_40ms_log_rms_mae_db": 0.755769125875707, "全帯域_40ms_rms_envelope_nrmse": 0.05433715038883701, "全帯域_40ms_envelope_correlation": 0.9977918379327659, "全帯域_40ms_modulation_mae_db": 0.7721602892263263, "300–1000 Hz_10ms_log_rms_mae_db": 1.9800331912922435, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11483555528256013, "300–1000 Hz_10ms_envelope_correlation": 0.9919645198123749, "300–1000 Hz_10ms_modulation_mae_db": 1.060716156702079, "300–1000 Hz_40ms_log_rms_mae_db": 1.1558681689451888, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05437172254355678, "300–1000 Hz_40ms_envelope_correlation": 0.9983589054398898, "300–1000 Hz_40ms_modulation_mae_db": 0.5546329625962488, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8739868245392357, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21302711686407044, "1000–3000 Hz_10ms_envelope_correlation": 0.9685302248430675, "1000–3000 Hz_10ms_modulation_mae_db": 0.254723446364602, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0937637284816706, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12253648270398067, "1000–3000 Hz_40ms_envelope_correlation": 0.9890735050434041, "1000–3000 Hz_40ms_modulation_mae_db": 0.18852542628856406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7105215224049621, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15963738008066025, "3000–6000 Hz_10ms_envelope_correlation": 0.9855054319590073, "3000–6000 Hz_10ms_modulation_mae_db": 0.6636794302761454, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2601130770070692, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10919985306273779, "3000–6000 Hz_40ms_envelope_correlation": 0.9935222920138818, "3000–6000 Hz_40ms_modulation_mae_db": 0.7123043696573843, "mr_stft_db": 5.884378588420538, "band_mae_db": 0.26572559366970067, "rms_gain_error_db": -0.04230751898010601, "output_peak": 0.5769695401623764}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4153102089824705, "全帯域_10ms_rms_envelope_nrmse": 0.09036594324398173, "全帯域_10ms_envelope_correlation": 0.9941452075931678, "全帯域_10ms_modulation_mae_db": 0.2545125991685591, "全帯域_40ms_log_rms_mae_db": 0.6810453171553879, "全帯域_40ms_rms_envelope_nrmse": 0.04570175591791711, "全帯域_40ms_envelope_correlation": 0.9985611518057032, "全帯域_40ms_modulation_mae_db": 0.3486665261562096, "300–1000 Hz_10ms_log_rms_mae_db": 1.8891406724129278, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12600425991621533, "300–1000 Hz_10ms_envelope_correlation": 0.9902722638051858, "300–1000 Hz_10ms_modulation_mae_db": 0.6800344840015896, "300–1000 Hz_40ms_log_rms_mae_db": 1.132713663713701, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06925960288686232, "300–1000 Hz_40ms_envelope_correlation": 0.997164694368893, "300–1000 Hz_40ms_modulation_mae_db": 0.8372105346206101, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8409757224106456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18668050944741474, "1000–3000 Hz_10ms_envelope_correlation": 0.9759791994075689, "1000–3000 Hz_10ms_modulation_mae_db": 0.5981003896632523, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0122503818525876, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11205203735908245, "1000–3000 Hz_40ms_envelope_correlation": 0.9916415783301739, "1000–3000 Hz_40ms_modulation_mae_db": 0.7084964128437051, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6880708895410577, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15204998455456548, "3000–6000 Hz_10ms_envelope_correlation": 0.9874856828503641, "3000–6000 Hz_10ms_modulation_mae_db": 0.7640552559794712, "3000–6000 Hz_40ms_log_rms_mae_db": 1.212797402002326, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07975032887312575, "3000–6000 Hz_40ms_envelope_correlation": 0.9968266911286652, "3000–6000 Hz_40ms_modulation_mae_db": 0.49550596849653566, "mr_stft_db": 5.872690932657335, "band_mae_db": 0.2246220289669171, "rms_gain_error_db": -0.12281599857139572, "output_peak": 0.5918032540826825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.42506618651192, "全帯域_10ms_rms_envelope_nrmse": 0.10193825861330705, "全帯域_10ms_envelope_correlation": 0.9924600339974237, "全帯域_10ms_modulation_mae_db": 0.7824670945739719, "全帯域_40ms_log_rms_mae_db": 0.7917800525711345, "全帯域_40ms_rms_envelope_nrmse": 0.05519791888907022, "全帯域_40ms_envelope_correlation": 0.997725980591325, "全帯域_40ms_modulation_mae_db": 1.0094230188252835, "300–1000 Hz_10ms_log_rms_mae_db": 1.924235674799606, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12342013900819841, "300–1000 Hz_10ms_envelope_correlation": 0.9905971465433991, "300–1000 Hz_10ms_modulation_mae_db": 1.2395582204263833, "300–1000 Hz_40ms_log_rms_mae_db": 1.1043803466105042, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06661756628624754, "300–1000 Hz_40ms_envelope_correlation": 0.9972800955093719, "300–1000 Hz_40ms_modulation_mae_db": 1.1521362609828782, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8424798123918174, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1950161215316595, "1000–3000 Hz_10ms_envelope_correlation": 0.9742836052957509, "1000–3000 Hz_10ms_modulation_mae_db": 0.7213734838630561, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1152119509275162, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11927274638321049, "1000–3000 Hz_40ms_envelope_correlation": 0.9913107954542193, "1000–3000 Hz_40ms_modulation_mae_db": 0.5761295450834193, "3000–6000 Hz_10ms_log_rms_mae_db": 1.688619825164568, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18438177230339423, "3000–6000 Hz_10ms_envelope_correlation": 0.9808282424850511, "3000–6000 Hz_10ms_modulation_mae_db": 1.1449537637316227, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2249257863691685, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09197553310851757, "3000–6000 Hz_40ms_envelope_correlation": 0.9952524324929148, "3000–6000 Hz_40ms_modulation_mae_db": 1.1148520738393566, "mr_stft_db": 5.885049293291228, "band_mae_db": 0.1918887298187292, "rms_gain_error_db": -0.045397163645607204, "output_peak": 0.5276404088328637}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4334404213703702, "全帯域_10ms_rms_envelope_nrmse": 0.09537339298459163, "全帯域_10ms_envelope_correlation": 0.993817884753297, "全帯域_10ms_modulation_mae_db": 0.5248336101565204, "全帯域_40ms_log_rms_mae_db": 0.7359826971996113, "全帯域_40ms_rms_envelope_nrmse": 0.05466911552011608, "全帯域_40ms_envelope_correlation": 0.9983244306470037, "全帯域_40ms_modulation_mae_db": 0.6122574746146104, "300–1000 Hz_10ms_log_rms_mae_db": 1.9477138443438764, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11539168731480877, "300–1000 Hz_10ms_envelope_correlation": 0.9926942328123691, "300–1000 Hz_10ms_modulation_mae_db": 0.5838670866159209, "300–1000 Hz_40ms_log_rms_mae_db": 1.1131545723775322, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06720863853918439, "300–1000 Hz_40ms_envelope_correlation": 0.9983377480230595, "300–1000 Hz_40ms_modulation_mae_db": 0.5746011652753383, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8602479480361878, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2047426126544477, "1000–3000 Hz_10ms_envelope_correlation": 0.9707510201646328, "1000–3000 Hz_10ms_modulation_mae_db": 0.6793673448005361, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0653700813411857, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10967953081935507, "1000–3000 Hz_40ms_envelope_correlation": 0.9917629554670752, "1000–3000 Hz_40ms_modulation_mae_db": 0.5611337983556997, "3000–6000 Hz_10ms_log_rms_mae_db": 1.62094996857608, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16401695214870912, "3000–6000 Hz_10ms_envelope_correlation": 0.9846948011986343, "3000–6000 Hz_10ms_modulation_mae_db": 0.44092484178783503, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2124716412525367, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09078110050620859, "3000–6000 Hz_40ms_envelope_correlation": 0.9958912313078979, "3000–6000 Hz_40ms_modulation_mae_db": 0.4958557505615651, "mr_stft_db": 5.858884286965984, "band_mae_db": 0.28280934208003455, "rms_gain_error_db": -0.2492687800116744, "output_peak": 0.5575721832315734}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.139046305229155, "spectral_convergence": 0.341028847629544, "spectral_convergence_status": "ok", "amplitude_floor": 2.3135422827537554e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.699051251449962, "spectral_convergence": 0.3075671160907145, "spectral_convergence_status": "ok", "amplitude_floor": 1.7511633455315343e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.764609740438542, "spectral_convergence": 0.3404425350353883, "spectral_convergence_status": "ok", "amplitude_floor": 1.6157176909573571e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00044579673236332844, "output_power": 0.0004829314021114903, "delta_db": 0.34748565035582024, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00284687947566611, "output_power": 0.0026551683338681452, "delta_db": -0.3027702148300445, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00047388905696308834, "output_power": 0.0004705668640821298, "delta_db": -0.030553379034358105, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013682287067652582, "output_power": 0.00012979594782140422, "delta_db": -0.2289756393099493, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011121195656671286, "output_power": 0.00010712043758348155, "delta_db": -0.16279143330580756, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00022124255174898303, "output_power": 0.00020338839363992077, "delta_db": -0.36542492579439595, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23966687377172927, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.19595111373580967, "pitch": null, "output_rms_dbfs": -23.926552931105725, "output_peak": 0.6420696376278592, "reference_rms_dbfs": -23.730601817369916, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.867569099039219}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028051035499693775, "log_rms_mae_db": 1.415692389832015, "rms_envelope_nrmse": 0.09859399494599543, "envelope_correlation": 0.993162615966573, "rms_gain_error_db": -0.19594986971739536, "reference_mean_power": 0.004239368434074771, "output_mean_power": 0.00405234278386122}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.0558356273882855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018964431225846456, "output_power": 0.001789538665013682, "reference_active": true, "delta_db": -0.25198735783356374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001936564415781366, "output_power": 0.00018811255778520942, "reference_active": true, "delta_db": -0.1261415885224927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.246230567335058e-05, "output_power": 6.125800420234578e-05, "reference_active": true, "delta_db": -0.08455168015380446}], "active_band_count": 3, "active_band_mae_db": 0.15422687550328695}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026381677718961605, "log_rms_mae_db": 0.7200382889030217, "rms_envelope_nrmse": 0.05030757831452972, "envelope_correlation": 0.9984409340809555, "rms_gain_error_db": -0.19594554745472936, "reference_mean_power": 0.004260663575263385, "output_mean_power": 0.004072702513505803}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.853150694908015e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017547087963691286, "output_power": 0.0016664785780529704, "reference_active": true, "delta_db": -0.22405317867753763}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.102196929549433e-05, "output_power": 7.735908413656793e-05, "reference_active": true, "delta_db": -0.20091475365088346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0378089105429984e-06, "output_power": 2.528714147897483e-06, "reference_active": true, "delta_db": -0.7966071325442552}], "active_band_count": 3, "active_band_mae_db": 0.4071916882908921}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002736722962808348, "log_rms_mae_db": 1.9331794175867518, "rms_envelope_nrmse": 0.12389777062758463, "envelope_correlation": 0.9912579746073473, "rms_gain_error_db": -0.37252079430201485, "reference_mean_power": 0.0024581880191069405, "output_mean_power": 0.0022561242741090826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.077905340514435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019022535341174436, "output_power": 0.0017689871672399772, "reference_active": true, "delta_db": -0.3154371721135927}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012849688146439824, "output_power": 0.0001189143924280734, "reference_active": true, "delta_db": -0.33658166489644703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.095154000220668e-05, "output_power": 4.19669570921205e-05, "reference_active": true, "delta_db": 0.10637242607643588}], "active_band_count": 3, "active_band_mae_db": 0.25279708769549186}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025339024948019116, "log_rms_mae_db": 1.1381148682483873, "rms_envelope_nrmse": 0.07040795936806118, "envelope_correlation": 0.9978999908655819, "rms_gain_error_db": -0.372521722962207, "reference_mean_power": 0.002470549817529662, "output_mean_power": 0.0022674694443750583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.981046082661891e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017987268874748684, "output_power": 0.0016818610560947703, "reference_active": true, "delta_db": -0.2917511207399175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5079379851283876e-05, "output_power": 5.01794822739819e-05, "reference_active": true, "delta_db": -0.40462865636933787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4721075346991203e-06, "output_power": 1.9096420224319445e-06, "reference_active": true, "delta_db": -1.1211539536343882}], "active_band_count": 3, "active_band_mae_db": 0.6058445769145478}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.470771529403075e-05, "log_rms_mae_db": 1.8444350304841433, "rms_envelope_nrmse": 0.20972373599537908, "envelope_correlation": 0.9693314362167601, "rms_gain_error_db": -0.0873097393256588, "reference_mean_power": 0.00019492290635367423, "output_mean_power": 0.0001910433408852287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7570632979474236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.555963644612216e-05, "output_power": 7.118233089688433e-05, "reference_active": true, "delta_db": -0.2591765458905144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.293127801504832e-05, "output_power": 1.2064439445685592e-05, "reference_active": true, "delta_db": -0.30134300955278576}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505165714035109e-06, "output_power": 7.98051245387825e-06, "reference_active": true, "delta_db": 0.8877241531221285}], "active_band_count": 3, "active_band_mae_db": 0.4827479028551429}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.885515861839929e-05, "log_rms_mae_db": 1.0428193917975297, "rms_envelope_nrmse": 0.11270865573741039, "envelope_correlation": 0.9907462751557929, "rms_gain_error_db": -0.08731738305480613, "reference_mean_power": 0.00019590175284412906, "output_mean_power": 0.00019200236738841443}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6226758883463992e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.225076970752032e-05, "output_power": 6.778115415960574e-05, "reference_active": true, "delta_db": -0.27733518496839865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.071619770568594e-06, "output_power": 5.837840608858618e-06, "reference_active": true, "delta_db": -0.1705233323029534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6400879277669946e-07, "output_power": 3.9314036954972687e-07, "reference_active": true, "delta_db": 0.3343576723125776}], "active_band_count": 3, "active_band_mae_db": 0.2607387298613099}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012408236865898127, "log_rms_mae_db": 1.6734545069833389, "rms_envelope_nrmse": 0.18408766753201342, "envelope_correlation": 0.9810729982710708, "rms_gain_error_db": -0.4841043326008648, "reference_mean_power": 0.00018911407706020183, "output_mean_power": 0.00016916613397661143}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8583468534278485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010082037892943086, "output_power": 9.230430139478322e-05, "reference_active": true, "delta_db": -0.3832638569614032}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9440078487669303e-05, "output_power": 1.7458244434728847e-05, "reference_active": true, "delta_db": -0.4669744421858486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.009136082695238e-06, "output_power": 3.0174764510695894e-06, "reference_active": true, "delta_db": -2.2011894081405843}], "active_band_count": 3, "active_band_mae_db": 1.0171425690959455}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.970743476594613e-05, "log_rms_mae_db": 1.2375836156375724, "rms_envelope_nrmse": 0.11585805832867355, "envelope_correlation": 0.9934962371452437, "rms_gain_error_db": -0.4841053229090704, "reference_mean_power": 0.00019006483009233416, "output_mean_power": 0.0001700165618541722}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7984096251410197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010012152579842322, "output_power": 9.084104609407869e-05, "reference_active": true, "delta_db": -0.4224533264030088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0693666398121268e-05, "output_power": 9.562514034582467e-06, "reference_active": true, "delta_db": -0.48554546071484794}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.760055744312906e-07, "output_power": 1.8336873808546407e-07, "reference_active": true, "delta_db": -1.7759255726472778}], "active_band_count": 3, "active_band_mae_db": 0.8946414532550447}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7817686417798577, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bc1ce63afb5a7e9732a92542903df28733e8533460dfbf5e44bae4714cc9a964", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9710133098789404, "standard_deviation": 0.0029571159962769947, "minimum": 0.9671941195386634, "maximum": 0.9759791994075689, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.861460169829516, "standard_deviation": 0.026284396184696832, "minimum": 1.8326327216446392, "maximum": 1.902862622262604, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6180934354800012, "standard_deviation": 0.23437614406227256, "minimum": 0.254723446364602, "maximum": 1.065887514401556, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20421529720498732, "standard_deviation": 0.009963935548260021, "minimum": 0.18668050944741474, "maximum": 0.217104215411044, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9906603046303526, "standard_deviation": 0.0014458772442077386, "minimum": 0.9878042816763153, "maximum": 0.9917629554670752, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0649200532758236, "standard_deviation": 0.031156171940806207, "minimum": 1.0122503818525876, "maximum": 1.1152119509275162, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.48128448583639266, "standard_deviation": 0.19829729316855668, "minimum": 0.18852542628856406, "maximum": 0.7307873392818718, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11649025408638222, "standard_deviation": 0.008329091467016348, "minimum": 0.1094931399221386, "maximum": 0.1336931980704149, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9835400529522582, "standard_deviation": 0.002279227651660355, "minimum": 0.9808282424850511, "maximum": 0.9874856828503641, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.664971182700645, "standard_deviation": 0.034402744746139434, "minimum": 1.62094996857608, "maximum": 1.7105215224049621, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8594784433790414, "standard_deviation": 0.3561920753737171, "minimum": 0.44092484178783503, "maximum": 1.502008247100096, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17158951330619931, "standard_deviation": 0.011995768458742803, "minimum": 0.15204998455456548, "maximum": 0.18438177230339423, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9953218029930642, "standard_deviation": 0.0012234584332557056, "minimum": 0.9934962371452437, "maximum": 0.9968266911286652, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2213925187457462, "standard_deviation": 0.019185837411911644, "minimum": 1.2001013671976144, "maximum": 1.2601130770070692, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.811033255411062, "standard_deviation": 0.36011927853748943, "minimum": 0.35415167752813126, "maximum": 1.4002289908999697, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09503286109990895, "standard_deviation": 0.013787066039614942, "minimum": 0.07801081824084498, "maximum": 0.11585805832867355, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913664164535063, "standard_deviation": 0.000909660106428572, "minimum": 0.9902722638051858, "maximum": 0.9926942328123691, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9444591640373385, "standard_deviation": 0.04022631196418151, "minimum": 1.8891406724129278, "maximum": 2.0207080347893385, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.749592874300024, "standard_deviation": 0.3172301637555257, "minimum": 0.25279708769549186, "maximum": 1.2395582204263833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12045943639501074, "standard_deviation": 0.004577700585449728, "minimum": 0.11483555528256013, "maximum": 0.12600425991621533, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997519927381542, "standard_deviation": 0.0006821580967953648, "minimum": 0.9964383859353918, "maximum": 0.9983589054398898, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1572070127473264, "standard_deviation": 0.04746276021528114, "minimum": 1.1043803466105042, "maximum": 1.2497331944765648, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.728791330711535, "standard_deviation": 0.2190887337275638, "minimum": 0.50625835228062, "maximum": 1.1521362609828782, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06831858990588269, "standard_deviation": 0.00642782944078541, "minimum": 0.05437172254355678, "maximum": 0.07652923052307942, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.238240997309082, "standard_deviation": 0.034565067262832816, "minimum": 0.1918887298187292, "maximum": 0.28280934208003455, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.874721632047856, "standard_deviation": 0.011793966201920945, "minimum": 5.858884286965984, "maximum": 5.885159314448884, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5781333608455862, "standard_deviation": 0.05029505937725286, "minimum": 0.5251881740664939, "maximum": 0.6590290810367669, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11806472263788259, "standard_deviation": 0.0839495479237987, "minimum": -0.2492687800116744, "maximum": -0.016496363780867534, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931508392000806, "standard_deviation": 0.0009229679765551844, "minimum": 0.9916529569964266, "maximum": 0.9941452075931678, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4353305857666423, "standard_deviation": 0.0228353103891332, "minimum": 1.4153102089824705, "maximum": 1.4716102716222583, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4556598368074544, "standard_deviation": 0.22540629823360525, "minimum": 0.15422687550328695, "maximum": 0.7824670945739719, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09780401939335316, "standard_deviation": 0.005950686362860079, "minimum": 0.09036594324398173, "maximum": 0.10717642206457817, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980863576290784, "standard_deviation": 0.0003698924165005256, "minimum": 0.9976006418017721, "maximum": 0.9985611518057032, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7346283566742355, "standard_deviation": 0.03358387177347911, "minimum": 0.6810453171553879, "maximum": 0.7917800525711345, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5842833786329173, "standard_deviation": 0.22070488634653915, "minimum": 0.3486665261562096, "maximum": 1.0094230188252835, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.052617249067116234, "standard_deviation": 0.004360126534456056, "minimum": 0.04570175591791711, "maximum": 0.05723364579742491, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0016", "method": "B", "status": "measured", "mos": {"status": "measured", "value": 1.9804251194000244, "predicted_mos": 1.9804251194000244, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f7dd2ea810bc11b80207531f6e256e165b8c71d976d6c714de29c11c365c2a51", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8775432999245822, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4504730621188764, "全帯域_10ms_rms_envelope_nrmse": 0.0898881832990418, "全帯域_10ms_envelope_correlation": 0.9941593693127317, "全帯域_10ms_modulation_mae_db": 0.3126959739164488, "全帯域_40ms_log_rms_mae_db": 0.7247702005268951, "全帯域_40ms_rms_envelope_nrmse": 0.0394138305514356, "全帯域_40ms_envelope_correlation": 0.9988919873569804, "全帯域_40ms_modulation_mae_db": 0.18989325572514104, "300–1000 Hz_10ms_log_rms_mae_db": 2.303579341203003, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11517644584624641, "300–1000 Hz_10ms_envelope_correlation": 0.9919600318806349, "300–1000 Hz_10ms_modulation_mae_db": 0.6947918193915616, "300–1000 Hz_40ms_log_rms_mae_db": 1.436119027175077, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05692807158192311, "300–1000 Hz_40ms_envelope_correlation": 0.9982427719875795, "300–1000 Hz_40ms_modulation_mae_db": 0.43605610512619647, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3270465237320224, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2345853364862597, "1000–3000 Hz_10ms_envelope_correlation": 0.9617122693714627, "1000–3000 Hz_10ms_modulation_mae_db": 0.4846124504747045, "1000–3000 Hz_40ms_log_rms_mae_db": 1.403928438585818, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12567771171487047, "1000–3000 Hz_40ms_envelope_correlation": 0.9884802996661891, "1000–3000 Hz_40ms_modulation_mae_db": 0.3704093764696352, "3000–6000 Hz_10ms_log_rms_mae_db": 2.26366004874737, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.181957931104452, "3000–6000 Hz_10ms_envelope_correlation": 0.9810738026891335, "3000–6000 Hz_10ms_modulation_mae_db": 0.6657440749127365, "3000–6000 Hz_40ms_log_rms_mae_db": 1.754887296229424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09644782156008024, "3000–6000 Hz_40ms_envelope_correlation": 0.995362138627729, "3000–6000 Hz_40ms_modulation_mae_db": 0.7554479740538632, "mr_stft_db": 6.221120745449583, "band_mae_db": 0.27116454162546316, "rms_gain_error_db": -0.08149069506187524, "output_peak": 0.6079773126361498}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4212529902697724, "全帯域_10ms_rms_envelope_nrmse": 0.10179097537565134, "全帯域_10ms_envelope_correlation": 0.9925647523960206, "全帯域_10ms_modulation_mae_db": 0.44373117719661764, "全帯域_40ms_log_rms_mae_db": 0.7784257581613439, "全帯域_40ms_rms_envelope_nrmse": 0.05992067505444729, "全帯域_40ms_envelope_correlation": 0.9974515469684734, "全帯域_40ms_modulation_mae_db": 0.8420524455170115, "300–1000 Hz_10ms_log_rms_mae_db": 2.2224020030046194, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12484226834264531, "300–1000 Hz_10ms_envelope_correlation": 0.9906501143133803, "300–1000 Hz_10ms_modulation_mae_db": 0.49902759754030973, "300–1000 Hz_40ms_log_rms_mae_db": 1.3065829255928059, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07495438799234691, "300–1000 Hz_40ms_envelope_correlation": 0.9969023010741216, "300–1000 Hz_40ms_modulation_mae_db": 0.8162556572493934, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3817403975068308, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.24051083234525258, "1000–3000 Hz_10ms_envelope_correlation": 0.9593643587810017, "1000–3000 Hz_10ms_modulation_mae_db": 0.3738271828702213, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4702956633615838, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14385843820052902, "1000–3000 Hz_40ms_envelope_correlation": 0.9856414666186658, "1000–3000 Hz_40ms_modulation_mae_db": 0.29512516374710623, "3000–6000 Hz_10ms_log_rms_mae_db": 2.327982067446647, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20483814003423412, "3000–6000 Hz_10ms_envelope_correlation": 0.9754694204847041, "3000–6000 Hz_10ms_modulation_mae_db": 1.3729417130898598, "3000–6000 Hz_40ms_log_rms_mae_db": 1.746627114801971, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0999510017883875, "3000–6000 Hz_40ms_envelope_correlation": 0.9942010430648835, "3000–6000 Hz_40ms_modulation_mae_db": 1.2547658776460549, "mr_stft_db": 6.220512452998258, "band_mae_db": 0.25197328726814416, "rms_gain_error_db": -0.1352956252286121, "output_peak": 0.6156328838464882}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3021947356933106, "全帯域_10ms_rms_envelope_nrmse": 0.08884666038107136, "全帯域_10ms_envelope_correlation": 0.9943110045432225, "全帯域_10ms_modulation_mae_db": 0.10766826863195328, "全帯域_40ms_log_rms_mae_db": 0.6634385342632314, "全帯域_40ms_rms_envelope_nrmse": 0.04450051217923501, "全帯域_40ms_envelope_correlation": 0.9985325507900793, "全帯域_40ms_modulation_mae_db": 0.03126577208305654, "300–1000 Hz_10ms_log_rms_mae_db": 2.2548286474593144, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10300535341077394, "300–1000 Hz_10ms_envelope_correlation": 0.9935243621439721, "300–1000 Hz_10ms_modulation_mae_db": 0.1426526474144472, "300–1000 Hz_40ms_log_rms_mae_db": 1.4129650139397034, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056418060556373086, "300–1000 Hz_40ms_envelope_correlation": 0.998030481985379, "300–1000 Hz_40ms_modulation_mae_db": 0.48982245184618795, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3239876277123623, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21529888095205035, "1000–3000 Hz_10ms_envelope_correlation": 0.9684301423583888, "1000–3000 Hz_10ms_modulation_mae_db": 1.1075035552085943, "1000–3000 Hz_40ms_log_rms_mae_db": 1.393753817885425, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12860239495918083, "1000–3000 Hz_40ms_envelope_correlation": 0.9898519875112938, "1000–3000 Hz_40ms_modulation_mae_db": 1.19603153962067, "3000–6000 Hz_10ms_log_rms_mae_db": 2.260592266250216, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17323150378257637, "3000–6000 Hz_10ms_envelope_correlation": 0.9829336214109817, "3000–6000 Hz_10ms_modulation_mae_db": 1.217732324382468, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7195554415124206, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09612532171225813, "3000–6000 Hz_40ms_envelope_correlation": 0.9953433157653708, "3000–6000 Hz_40ms_modulation_mae_db": 0.8770948753905835, "mr_stft_db": 6.210399937717338, "band_mae_db": 0.30177815496153876, "rms_gain_error_db": 0.0202732520935138, "output_peak": 0.5720625723206156}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.425727662464635, "全帯域_10ms_rms_envelope_nrmse": 0.10474195583443859, "全帯域_10ms_envelope_correlation": 0.9921291933728905, "全帯域_10ms_modulation_mae_db": 0.17082414599999704, "全帯域_40ms_log_rms_mae_db": 0.7659832583926467, "全帯域_40ms_rms_envelope_nrmse": 0.05242808508910831, "全帯域_40ms_envelope_correlation": 0.9981248332830991, "全帯域_40ms_modulation_mae_db": 0.45687630789865574, "300–1000 Hz_10ms_log_rms_mae_db": 2.2906368593112054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13720873578832488, "300–1000 Hz_10ms_envelope_correlation": 0.9884955410970007, "300–1000 Hz_10ms_modulation_mae_db": 0.4883552456818399, "300–1000 Hz_40ms_log_rms_mae_db": 1.4384159146690503, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07904334010476122, "300–1000 Hz_40ms_envelope_correlation": 0.9963580193798331, "300–1000 Hz_40ms_modulation_mae_db": 0.4968418250917477, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3676874196023077, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23895043341706154, "1000–3000 Hz_10ms_envelope_correlation": 0.9617171413570891, "1000–3000 Hz_10ms_modulation_mae_db": 0.9242171896312628, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4671911475666772, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14633204008208425, "1000–3000 Hz_40ms_envelope_correlation": 0.9881937132456721, "1000–3000 Hz_40ms_modulation_mae_db": 1.1691157943655222, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2951044800514238, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19205634996434198, "3000–6000 Hz_10ms_envelope_correlation": 0.9805827986822717, "3000–6000 Hz_10ms_modulation_mae_db": 1.191778792761875, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7019540382554978, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12257166468457725, "3000–6000 Hz_40ms_envelope_correlation": 0.994290743568392, "3000–6000 Hz_40ms_modulation_mae_db": 1.2594699424665496, "mr_stft_db": 6.227096649061063, "band_mae_db": 0.4228779546466395, "rms_gain_error_db": -0.14960704188757973, "output_peak": 0.7288457057049784}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.376729898783921, "全帯域_10ms_rms_envelope_nrmse": 0.11198076148148531, "全帯域_10ms_envelope_correlation": 0.990900828726366, "全帯域_10ms_modulation_mae_db": 0.2829097252171904, "全帯域_40ms_log_rms_mae_db": 0.7112125316286517, "全帯域_40ms_rms_envelope_nrmse": 0.06544267751801783, "全帯域_40ms_envelope_correlation": 0.9968379994719562, "全帯域_40ms_modulation_mae_db": 0.20640830988105638, "300–1000 Hz_10ms_log_rms_mae_db": 2.223116570512974, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13688249561961255, "300–1000 Hz_10ms_envelope_correlation": 0.9885965854459361, "300–1000 Hz_10ms_modulation_mae_db": 0.7191020775718978, "300–1000 Hz_40ms_log_rms_mae_db": 1.4320882262225199, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07726260088136619, "300–1000 Hz_40ms_envelope_correlation": 0.9966075511534186, "300–1000 Hz_40ms_modulation_mae_db": 0.38285232890573484, "1000–3000 Hz_10ms_log_rms_mae_db": 2.329623700678156, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21522619564768053, "1000–3000 Hz_10ms_envelope_correlation": 0.9681427583527193, "1000–3000 Hz_10ms_modulation_mae_db": 0.948196715121985, "1000–3000 Hz_40ms_log_rms_mae_db": 1.446112343669353, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13026758574016226, "1000–3000 Hz_40ms_envelope_correlation": 0.989082299301673, "1000–3000 Hz_40ms_modulation_mae_db": 1.0635350300494133, "3000–6000 Hz_10ms_log_rms_mae_db": 2.285915543146024, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18822898835294832, "3000–6000 Hz_10ms_envelope_correlation": 0.9808605071555466, "3000–6000 Hz_10ms_modulation_mae_db": 0.32695386962237427, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7409237978635104, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10703020482129062, "3000–6000 Hz_40ms_envelope_correlation": 0.9941638211500913, "3000–6000 Hz_40ms_modulation_mae_db": 0.23254326805600586, "mr_stft_db": 6.21387426036688, "band_mae_db": 0.2366609634038661, "rms_gain_error_db": -0.09132469322974754, "output_peak": 0.6872914050900651}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3616960951664583, "全帯域_10ms_rms_envelope_nrmse": 0.09873288801832587, "全帯域_10ms_envelope_correlation": 0.9929586891687133, "全帯域_10ms_modulation_mae_db": 0.5750255130086274, "全帯域_40ms_log_rms_mae_db": 0.703984550641216, "全帯域_40ms_rms_envelope_nrmse": 0.04941938120371105, "全帯域_40ms_envelope_correlation": 0.9982189192873405, "全帯域_40ms_modulation_mae_db": 0.7107207905215766, "300–1000 Hz_10ms_log_rms_mae_db": 2.248364184360253, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11856347968706196, "300–1000 Hz_10ms_envelope_correlation": 0.9914406628865072, "300–1000 Hz_10ms_modulation_mae_db": 0.8375464685378137, "300–1000 Hz_40ms_log_rms_mae_db": 1.4029276437577205, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05899746001086575, "300–1000 Hz_40ms_envelope_correlation": 0.9980491242162356, "300–1000 Hz_40ms_modulation_mae_db": 0.5831917882101686, "1000–3000 Hz_10ms_log_rms_mae_db": 2.353304952766934, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.207595592560489, "1000–3000 Hz_10ms_envelope_correlation": 0.9698686775947775, "1000–3000 Hz_10ms_modulation_mae_db": 0.5392296819706422, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4771536720261642, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12013757912065309, "1000–3000 Hz_40ms_envelope_correlation": 0.9896078207863289, "1000–3000 Hz_40ms_modulation_mae_db": 0.518617449585815, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2804023695665214, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17250867943634954, "3000–6000 Hz_10ms_envelope_correlation": 0.9835386341218748, "3000–6000 Hz_10ms_modulation_mae_db": 0.3852038929616893, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7140257859156929, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10329590045465971, "3000–6000 Hz_40ms_envelope_correlation": 0.9951502641491333, "3000–6000 Hz_40ms_modulation_mae_db": 0.5242680595559049, "mr_stft_db": 6.21966800753932, "band_mae_db": 0.2835776035608362, "rms_gain_error_db": -0.07468367230930237, "output_peak": 0.5677345445383957}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3398581183111122, "全帯域_10ms_rms_envelope_nrmse": 0.08587169592226719, "全帯域_10ms_envelope_correlation": 0.9948763029900454, "全帯域_10ms_modulation_mae_db": 0.5635735379301001, "全帯域_40ms_log_rms_mae_db": 0.662675160333679, "全帯域_40ms_rms_envelope_nrmse": 0.04631765412780253, "全帯域_40ms_envelope_correlation": 0.998703578264049, "全帯域_40ms_modulation_mae_db": 0.7154057845286541, "300–1000 Hz_10ms_log_rms_mae_db": 2.3034416785455925, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11820770262294239, "300–1000 Hz_10ms_envelope_correlation": 0.9918235625846293, "300–1000 Hz_10ms_modulation_mae_db": 0.2922053960278233, "300–1000 Hz_40ms_log_rms_mae_db": 1.4707470362388153, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06924113592776455, "300–1000 Hz_40ms_envelope_correlation": 0.9976291637960474, "300–1000 Hz_40ms_modulation_mae_db": 0.6123943123517788, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3465675840491307, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2234251735914721, "1000–3000 Hz_10ms_envelope_correlation": 0.9650581829941194, "1000–3000 Hz_10ms_modulation_mae_db": 0.6950622192732682, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4346415778197885, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12796869731793348, "1000–3000 Hz_40ms_envelope_correlation": 0.9880882301842168, "1000–3000 Hz_40ms_modulation_mae_db": 0.5102757300911303, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2285729727163837, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17134188919995857, "3000–6000 Hz_10ms_envelope_correlation": 0.9830375375345488, "3000–6000 Hz_10ms_modulation_mae_db": 0.6822182984139991, "3000–6000 Hz_40ms_log_rms_mae_db": 1.702194418807504, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09262308944144639, "3000–6000 Hz_40ms_envelope_correlation": 0.995027909518139, "3000–6000 Hz_40ms_modulation_mae_db": 0.5660570945247414, "mr_stft_db": 6.220635163167799, "band_mae_db": 0.20598149155772705, "rms_gain_error_db": -0.18604030091105306, "output_peak": 0.604291952259532}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.362590133583547, "全帯域_10ms_rms_envelope_nrmse": 0.0999321125656076, "全帯域_10ms_envelope_correlation": 0.9927520005185183, "全帯域_10ms_modulation_mae_db": 0.34647683191439854, "全帯域_40ms_log_rms_mae_db": 0.6786919467298341, "全帯域_40ms_rms_envelope_nrmse": 0.0481382137092442, "全帯域_40ms_envelope_correlation": 0.9982772021997388, "全帯域_40ms_modulation_mae_db": 0.3193375360390173, "300–1000 Hz_10ms_log_rms_mae_db": 2.1644977400700878, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1252017991073493, "300–1000 Hz_10ms_envelope_correlation": 0.9903317969030471, "300–1000 Hz_10ms_modulation_mae_db": 0.6915724364904813, "300–1000 Hz_40ms_log_rms_mae_db": 1.381571604943691, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06648014748036798, "300–1000 Hz_40ms_envelope_correlation": 0.9973111491962179, "300–1000 Hz_40ms_modulation_mae_db": 0.6543945822833308, "1000–3000 Hz_10ms_log_rms_mae_db": 2.3510463465791998, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23490806203162676, "1000–3000 Hz_10ms_envelope_correlation": 0.9613150274375316, "1000–3000 Hz_10ms_modulation_mae_db": 0.519201671388167, "1000–3000 Hz_40ms_log_rms_mae_db": 1.4409278081242562, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12067079530249299, "1000–3000 Hz_40ms_envelope_correlation": 0.9894923936923822, "1000–3000 Hz_40ms_modulation_mae_db": 0.4163433376198085, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2474791297542978, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16161720033748822, "3000–6000 Hz_10ms_envelope_correlation": 0.9852447644548641, "3000–6000 Hz_10ms_modulation_mae_db": 1.232774743378515, "3000–6000 Hz_40ms_log_rms_mae_db": 1.7004113505659606, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0880621154810916, "3000–6000 Hz_40ms_envelope_correlation": 0.9962306845740366, "3000–6000 Hz_40ms_modulation_mae_db": 0.856977231583563, "mr_stft_db": 6.214855010187804, "band_mae_db": 0.2968950213060903, "rms_gain_error_db": -0.05038914053691068, "output_peak": 0.6938941140431318}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.536816065482638, "spectral_convergence": 0.34933633837756783, "spectral_convergence_status": "ok", "amplitude_floor": 2.3135422827537554e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 6.127028465879409, "spectral_convergence": 0.3130306860199657, "spectral_convergence_status": "ok", "amplitude_floor": 1.7511633455315343e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.999517704986702, "spectral_convergence": 0.35469481168499, "spectral_convergence_status": "ok", "amplitude_floor": 1.6157176909573571e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00044579673236332844, "output_power": 0.0004939498685911008, "delta_db": 0.44545993399907546, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00284687947566611, "output_power": 0.0027048922029051696, "delta_db": -0.22219119415360825, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00047388905696308834, "output_power": 0.0005038099021765916, "delta_db": 0.26590019030114276, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013682287067652582, "output_power": 0.0001409014369362362, "delta_db": 0.12756724035581407, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011121195656671286, "output_power": 0.00010219165381360936, "delta_db": -0.36736053824748804, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00022124255174898303, "output_power": 0.00021135758772245793, "delta_db": -0.1985081526956503, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27116454162546316, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.08149069506187524, "pitch": null, "output_rms_dbfs": -23.812092512431793, "output_peak": 0.6079773126361498, "reference_rms_dbfs": -23.730601817369916, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.221120745449583}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028051035499693775, "log_rms_mae_db": 1.4504730621188764, "rms_envelope_nrmse": 0.0898881832990418, "envelope_correlation": 0.9941593693127317, "rms_gain_error_db": -0.08149254106118778, "reference_mean_power": 0.004239368434074771, "output_mean_power": 0.004160561132919462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.0558356273882855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018964431225846456, "output_power": 0.0018512606504359611, "reference_active": true, "delta_db": -0.10472251971972789}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001936564415781366, "output_power": 0.0001806782868170593, "reference_active": true, "delta_db": -0.30125983514441085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.246230567335058e-05, "output_power": 7.060387997861901e-05, "reference_active": true, "delta_db": 0.5321055668852077}], "active_band_count": 3, "active_band_mae_db": 0.3126959739164488}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026381677718961605, "log_rms_mae_db": 0.7247702005268951, "rms_envelope_nrmse": 0.0394138305514356, "envelope_correlation": 0.9988919873569804, "rms_gain_error_db": -0.08148547513396079, "reference_mean_power": 0.004260663575263385, "output_mean_power": 0.004181467213482128}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.853150694908015e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017547087963691286, "output_power": 0.0017253620768682323, "reference_active": true, "delta_db": -0.07324805113843667}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.102196929549433e-05, "output_power": 7.91714827131292e-05, "reference_active": true, "delta_db": -0.10034016380366288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0378089105429984e-06, "output_power": 3.3278949361801035e-06, "reference_active": true, "delta_db": 0.3960915522333236}], "active_band_count": 3, "active_band_mae_db": 0.18989325572514104}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002736722962808348, "log_rms_mae_db": 2.303579341203003, "rms_envelope_nrmse": 0.11517644584624641, "envelope_correlation": 0.9919600318806349, "rms_gain_error_db": -0.19937008000628265, "reference_mean_power": 0.0024581880191069405, "output_mean_power": 0.0023478918572543347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.077905340514435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019022535341174436, "output_power": 0.0018010576906829134, "reference_active": true, "delta_db": -0.23740775459940971}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012849688146439824, "output_power": 0.00011575775857147302, "reference_active": true, "delta_db": -0.45342478844170314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.095154000220668e-05, "output_power": 5.644487684015535e-05, "reference_active": true, "delta_db": 1.393542915133572}], "active_band_count": 3, "active_band_mae_db": 0.6947918193915616}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025339024948019116, "log_rms_mae_db": 1.436119027175077, "rms_envelope_nrmse": 0.05692807158192311, "envelope_correlation": 0.9982427719875795, "rms_gain_error_db": -0.19937086710882307, "reference_mean_power": 0.002470549817529662, "output_mean_power": 0.002359698567868613}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.981046082661891e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017987268874748684, "output_power": 0.0017130153071162747, "reference_active": true, "delta_db": -0.21203982825880677}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5079379851283876e-05, "output_power": 5.088573243597265e-05, "reference_active": true, "delta_db": -0.34393011792150174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4721075346991203e-06, "output_power": 2.939592936907034e-06, "reference_active": true, "delta_db": 0.752198369198281}], "active_band_count": 3, "active_band_mae_db": 0.43605610512619647}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.470771529403075e-05, "log_rms_mae_db": 2.3270465237320224, "rms_envelope_nrmse": 0.2345853364862597, "envelope_correlation": 0.9617122693714627, "rms_gain_error_db": -0.03506808149931953, "reference_mean_power": 0.00019492290635367423, "output_mean_power": 0.00019335529518185583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7570632979474236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.555963644612216e-05, "output_power": 6.817857337971298e-05, "reference_active": true, "delta_db": -0.44641950174336914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.293127801504832e-05, "output_power": 1.107548848391393e-05, "reference_active": true, "delta_db": -0.6727855903501067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505165714035109e-06, "output_power": 7.026217513320754e-06, "reference_active": true, "delta_db": 0.3346322593306379}], "active_band_count": 3, "active_band_mae_db": 0.4846124504747045}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.885515861839929e-05, "log_rms_mae_db": 1.403928438585818, "rms_envelope_nrmse": 0.12567771171487047, "envelope_correlation": 0.9884802996661891, "rms_gain_error_db": -0.03505629398574957, "reference_mean_power": 0.00019590175284412906, "output_mean_power": 0.0001943267970184236}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6226758883463992e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.225076970752032e-05, "output_power": 6.514535420222213e-05, "reference_active": true, "delta_db": -0.4495902838335319}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.071619770568594e-06, "output_power": 5.49465667637269e-06, "reference_active": true, "delta_db": -0.4336400495227639}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6400879277669946e-07, "output_power": 3.8362919804195434e-07, "reference_active": true, "delta_db": 0.22799779605260967}], "active_band_count": 3, "active_band_mae_db": 0.3704093764696352}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012408236865898127, "log_rms_mae_db": 2.26366004874737, "rms_envelope_nrmse": 0.181957931104452, "envelope_correlation": 0.9810738026891335, "rms_gain_error_db": -0.33368383674596547, "reference_mean_power": 0.00018911407706020183, "output_mean_power": 0.00017512795400234651}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8583468534278485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010082037892943086, "output_power": 9.648405320043663e-05, "reference_active": true, "delta_db": -0.19092785919439484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9440078487669303e-05, "output_power": 1.3252205997937215e-05, "reference_active": true, "delta_db": -1.6640983590007339}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.009136082695238e-06, "output_power": 4.847772487713053e-06, "reference_active": true, "delta_db": -0.14220600654308063}], "active_band_count": 3, "active_band_mae_db": 0.6657440749127365}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.970743476594613e-05, "log_rms_mae_db": 1.754887296229424, "rms_envelope_nrmse": 0.09644782156008024, "envelope_correlation": 0.995362138627729, "rms_gain_error_db": -0.33368986318228877, "reference_mean_power": 0.00019006483009233416, "output_mean_power": 0.0001760081488954873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7984096251410197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010012152579842322, "output_power": 9.561625132281224e-05, "reference_active": true, "delta_db": -0.19995746414537735}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0693666398121268e-05, "output_power": 7.309992118738954e-06, "reference_active": true, "delta_db": -1.65209722909202}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.760055744312906e-07, "output_power": 3.0363146596864316e-07, "reference_active": true, "delta_db": 0.4142892289241922}], "active_band_count": 3, "active_band_mae_db": 0.7554479740538632}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7602879339568933, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "db318a43f4494a4256a3a151a6859ffe26018731e7a09343a9173ff11d86a939", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9644510697808862, "standard_deviation": 0.003961699250358136, "minimum": 0.9593643587810017, "maximum": 0.9698686775947775, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.347625569078368, "standard_deviation": 0.020416003738491877, "minimum": 2.3239876277123623, "maximum": 2.3817403975068308, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6989813332423557, "standard_deviation": 0.2643768466107506, "minimum": 0.3738271828702213, "maximum": 1.1075035552085943, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2263125633789866, "standard_deviation": 0.012572638881943748, "minimum": 0.207595592560489, "maximum": 0.24051083234525258, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9885547763758027, "standard_deviation": 0.0013517194544292726, "minimum": 0.9856414666186658, "maximum": 0.9898519875112938, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.441750558629883, "standard_deviation": 0.030519002487317825, "minimum": 1.393753817885425, "maximum": 1.4771536720261642, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6924316776936376, "standard_deviation": 0.3816741394967761, "minimum": 0.29512516374710623, "maximum": 1.19603153962067, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1304394053047383, "standard_deviation": 0.009756916234869323, "minimum": 0.12013757912065309, "maximum": 0.14633204008208425, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9815926358167406, "standard_deviation": 0.0029357324119631766, "minimum": 0.9754694204847041, "maximum": 0.9852447644548641, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2737136097098603, "standard_deviation": 0.0306953470555572, "minimum": 2.2285729727163837, "maximum": 2.327982067446647, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8844184636904397, "standard_deviation": 0.4164923118533601, "minimum": 0.32695386962237427, "maximum": 1.3729417130898598, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1807225852765436, "standard_deviation": 0.013858901798282773, "minimum": 0.16161720033748822, "maximum": 0.20483814003423412, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9949712400522219, "standard_deviation": 0.000719216611434969, "minimum": 0.9941638211500913, "maximum": 0.9962306845740366, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.7225724054939975, "standard_deviation": 0.021952199198049367, "minimum": 1.7004113505659606, "maximum": 1.754887296229424, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7908280404096583, "standard_deviation": 0.3548019707502316, "minimum": 0.23254326805600586, "maximum": 1.2594699424665496, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10076338999297393, "standard_deviation": 0.010621205692396005, "minimum": 0.0880621154810916, "maximum": 0.12257166468457725, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9908528321568885, "standard_deviation": 0.0017159282723816127, "minimum": 0.9884955410970007, "maximum": 0.9935243621439721, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.251358378058381, "standard_deviation": 0.048083778105002466, "minimum": 2.1644977400700878, "maximum": 2.303579341203003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5456567110820219, "standard_deviation": 0.23631794010804258, "minimum": 0.1426526474144472, "maximum": 0.8375464685378137, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1223860350531196, "standard_deviation": 0.011360840160710385, "minimum": 0.10300535341077394, "maximum": 0.13720873578832488, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997391320348604, "standard_deviation": 0.0007121742596821172, "minimum": 0.9963580193798331, "maximum": 0.9982427719875795, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4101771740674227, "standard_deviation": 0.049612186386307516, "minimum": 1.3065829255928059, "maximum": 1.4707470362388153, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5589761313830673, "standard_deviation": 0.1381040836065521, "minimum": 0.38285232890573484, "maximum": 0.8162556572493934, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0674156505669711, "standard_deviation": 0.009219606053912678, "minimum": 0.056418060556373086, "maximum": 0.07904334010476122, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2838636272912881, "standard_deviation": 0.06468361644991924, "minimum": 0.20598149155772705, "maximum": 0.4228779546466395, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 6.218520278311005, "standard_deviation": 0.005224050828538383, "minimum": 6.210399937717338, "maximum": 6.227096649061063, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6347163113049197, "standard_deviation": 0.06040735134796105, "minimum": 0.5677345445383957, "maximum": 0.7288457057049784, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.09356973963394585, "standard_deviation": 0.06410309940064954, "minimum": -0.18604030091105306, "maximum": 0.0202732520935138, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930815176285634, "standard_deviation": 0.0013064508660712937, "minimum": 0.990900828726366, "maximum": 0.9948763029900454, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3800653370489542, "standard_deviation": 0.04938638681106471, "minimum": 1.3021947356933106, "maximum": 1.4504730621188764, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3503631467269166, "standard_deviation": 0.16977648285461241, "minimum": 0.10766826863195328, "maximum": 0.5750255130086274, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09772315410973614, "standard_deviation": 0.008909266004589814, "minimum": 0.08587169592226719, "maximum": 0.11198076148148531, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9981298272027146, "standard_deviation": 0.0006793725424027519, "minimum": 0.9968379994719562, "maximum": 0.9988919873569804, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7111477425846873, "standard_deviation": 0.0438329467149296, "minimum": 0.662675160333679, "maximum": 0.7784257581613439, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43399502527427114, "standard_deviation": 0.2950840446913162, "minimum": 0.03126577208305654, "maximum": 0.8420524455170115, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05069762867912523, "standard_deviation": 0.008439470031227187, "minimum": 0.0394138305514356, "maximum": 0.06544267751801783, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0016", "method": "C", "status": "measured", "mos": {"status": "measured", "value": 2.105447292327881, "predicted_mos": 2.105447292327881, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f83b5c16aa3c3cab85a4a5b02cc6616617e2ca7a62cab668e72e44e9d47d1417", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.9096126998774707, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4175312947094507, "全帯域_10ms_rms_envelope_nrmse": 0.09128900925908512, "全帯域_10ms_envelope_correlation": 0.993971670851414, "全帯域_10ms_modulation_mae_db": 0.4081143911559817, "全帯域_40ms_log_rms_mae_db": 0.769300251938301, "全帯域_40ms_rms_envelope_nrmse": 0.04658636302382812, "全帯域_40ms_envelope_correlation": 0.9984190987748623, "全帯域_40ms_modulation_mae_db": 0.33908249189297496, "300–1000 Hz_10ms_log_rms_mae_db": 1.9774650586182836, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1101888676957052, "300–1000 Hz_10ms_envelope_correlation": 0.9925365554907648, "300–1000 Hz_10ms_modulation_mae_db": 0.45426732333963155, "300–1000 Hz_40ms_log_rms_mae_db": 1.1858741811237759, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06135134323763531, "300–1000 Hz_40ms_envelope_correlation": 0.9977018433932854, "300–1000 Hz_40ms_modulation_mae_db": 0.6050807664916444, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6935630708034097, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22194419686641872, "1000–3000 Hz_10ms_envelope_correlation": 0.9655893581122387, "1000–3000 Hz_10ms_modulation_mae_db": 1.0968805072021555, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0765140938476025, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14241470958890065, "1000–3000 Hz_40ms_envelope_correlation": 0.9857263605987857, "1000–3000 Hz_40ms_modulation_mae_db": 0.7251301007720844, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4986790651896642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18523953795999168, "3000–6000 Hz_10ms_envelope_correlation": 0.9809907150800491, "3000–6000 Hz_10ms_modulation_mae_db": 0.5130048924396552, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1758517930902255, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10380526564316002, "3000–6000 Hz_40ms_envelope_correlation": 0.9954571270650501, "3000–6000 Hz_40ms_modulation_mae_db": 0.5470208023533555, "mr_stft_db": 5.7937853507667185, "band_mae_db": 0.23456306330667162, "rms_gain_error_db": -0.0780473092818486, "output_peak": 0.5334650781629787}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.481494897591341, "全帯域_10ms_rms_envelope_nrmse": 0.09851121386890888, "全帯域_10ms_envelope_correlation": 0.9929542511861883, "全帯域_10ms_modulation_mae_db": 0.6090085635043617, "全帯域_40ms_log_rms_mae_db": 0.7678678355746066, "全帯域_40ms_rms_envelope_nrmse": 0.04924867832177231, "全帯域_40ms_envelope_correlation": 0.9982180732804063, "全帯域_40ms_modulation_mae_db": 0.42246001991233983, "300–1000 Hz_10ms_log_rms_mae_db": 1.8363867211542964, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11403123882560676, "300–1000 Hz_10ms_envelope_correlation": 0.9921382372622598, "300–1000 Hz_10ms_modulation_mae_db": 0.62372698314983, "300–1000 Hz_40ms_log_rms_mae_db": 1.136465707221305, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.061583345409085354, "300–1000 Hz_40ms_envelope_correlation": 0.9979008098272609, "300–1000 Hz_40ms_modulation_mae_db": 0.12706928638428586, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7210200984955795, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20431804606758036, "1000–3000 Hz_10ms_envelope_correlation": 0.9708058139307412, "1000–3000 Hz_10ms_modulation_mae_db": 0.5910473260439653, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0569868606597086, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12568627177639066, "1000–3000 Hz_40ms_envelope_correlation": 0.9885108204287174, "1000–3000 Hz_40ms_modulation_mae_db": 0.3570247106064192, "3000–6000 Hz_10ms_log_rms_mae_db": 1.460013513105017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14115138646495, "3000–6000 Hz_10ms_envelope_correlation": 0.9892906346587226, "3000–6000 Hz_10ms_modulation_mae_db": 0.5022627089467154, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0805110518294332, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07182732743677213, "3000–6000 Hz_40ms_envelope_correlation": 0.9975327923411317, "3000–6000 Hz_40ms_modulation_mae_db": 0.464944009506098, "mr_stft_db": 5.796276068064343, "band_mae_db": 0.2642882363486186, "rms_gain_error_db": -0.06716567286354208, "output_peak": 0.7460171015925696}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3956538034663288, "全帯域_10ms_rms_envelope_nrmse": 0.09445613232589753, "全帯域_10ms_envelope_correlation": 0.9935581027919599, "全帯域_10ms_modulation_mae_db": 0.5219308566762839, "全帯域_40ms_log_rms_mae_db": 0.7740023178253514, "全帯域_40ms_rms_envelope_nrmse": 0.04985584739996026, "全帯域_40ms_envelope_correlation": 0.9982038895938404, "全帯域_40ms_modulation_mae_db": 0.742397254088658, "300–1000 Hz_10ms_log_rms_mae_db": 1.8427382643143042, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11059315483849608, "300–1000 Hz_10ms_envelope_correlation": 0.9927537520591516, "300–1000 Hz_10ms_modulation_mae_db": 0.6314460508127793, "300–1000 Hz_40ms_log_rms_mae_db": 1.1869475521462425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06642455652554329, "300–1000 Hz_40ms_envelope_correlation": 0.9976467780711532, "300–1000 Hz_40ms_modulation_mae_db": 0.5481160537587376, "1000–3000 Hz_10ms_log_rms_mae_db": 1.678246184759322, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2034874373243812, "1000–3000 Hz_10ms_envelope_correlation": 0.9710676139161335, "1000–3000 Hz_10ms_modulation_mae_db": 0.4653150641765625, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0168325334309853, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11633844656597893, "1000–3000 Hz_40ms_envelope_correlation": 0.9902367074330045, "1000–3000 Hz_40ms_modulation_mae_db": 0.4876644592299472, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4793985729165273, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18581656428920323, "3000–6000 Hz_10ms_envelope_correlation": 0.979830311896843, "3000–6000 Hz_10ms_modulation_mae_db": 0.23359353368804656, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1645610070833727, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09096708699015862, "3000–6000 Hz_40ms_envelope_correlation": 0.9952908434051185, "3000–6000 Hz_40ms_modulation_mae_db": 0.42552630604233804, "mr_stft_db": 5.780161727672048, "band_mae_db": 0.21489513333781107, "rms_gain_error_db": -0.08969327679903813, "output_peak": 0.6008716666603795}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4903343998978016, "全帯域_10ms_rms_envelope_nrmse": 0.10638024337342558, "全帯域_10ms_envelope_correlation": 0.9918105963874977, "全帯域_10ms_modulation_mae_db": 0.40195220986661995, "全帯域_40ms_log_rms_mae_db": 0.7217306079022696, "全帯域_40ms_rms_envelope_nrmse": 0.051652065384467434, "全帯域_40ms_envelope_correlation": 0.9980041842121086, "全帯域_40ms_modulation_mae_db": 0.638030221362875, "300–1000 Hz_10ms_log_rms_mae_db": 1.9032994739765978, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1216445872603086, "300–1000 Hz_10ms_envelope_correlation": 0.9908736450892374, "300–1000 Hz_10ms_modulation_mae_db": 0.7037668892577003, "300–1000 Hz_40ms_log_rms_mae_db": 1.0916881779192915, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.060705441108079274, "300–1000 Hz_40ms_envelope_correlation": 0.9977146062887493, "300–1000 Hz_40ms_modulation_mae_db": 0.601074789044401, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6970416937999317, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20517208599811174, "1000–3000 Hz_10ms_envelope_correlation": 0.9711132893450156, "1000–3000 Hz_10ms_modulation_mae_db": 0.45076581175069497, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0384713643867438, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11827049903013405, "1000–3000 Hz_40ms_envelope_correlation": 0.991246553096003, "1000–3000 Hz_40ms_modulation_mae_db": 0.2795675560621454, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5099963133496301, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1534400040427916, "3000–6000 Hz_10ms_envelope_correlation": 0.9863395782748693, "3000–6000 Hz_10ms_modulation_mae_db": 0.2764931246276204, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1674175310326913, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07454057809402387, "3000–6000 Hz_40ms_envelope_correlation": 0.9967110808664934, "3000–6000 Hz_40ms_modulation_mae_db": 0.09182860726111713, "mr_stft_db": 5.815879659607956, "band_mae_db": 0.2827575365097426, "rms_gain_error_db": -0.009152634546971572, "output_peak": 0.5905412861317445}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5124393608175477, "全帯域_10ms_rms_envelope_nrmse": 0.0993809936320483, "全帯域_10ms_envelope_correlation": 0.9928661583062552, "全帯域_10ms_modulation_mae_db": 0.07712791044457437, "全帯域_40ms_log_rms_mae_db": 0.8290811239516677, "全帯域_40ms_rms_envelope_nrmse": 0.04841098310487001, "全帯域_40ms_envelope_correlation": 0.9983303353065122, "全帯域_40ms_modulation_mae_db": 0.16393418601609322, "300–1000 Hz_10ms_log_rms_mae_db": 1.8457859823649003, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.10996266938973384, "300–1000 Hz_10ms_envelope_correlation": 0.9926503599742972, "300–1000 Hz_10ms_modulation_mae_db": 0.33227069379665836, "300–1000 Hz_40ms_log_rms_mae_db": 1.1238365147508376, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.05850623019327885, "300–1000 Hz_40ms_envelope_correlation": 0.9980502781366684, "300–1000 Hz_40ms_modulation_mae_db": 0.14893782915421394, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7315279400669332, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2235092828244344, "1000–3000 Hz_10ms_envelope_correlation": 0.9659552649033201, "1000–3000 Hz_10ms_modulation_mae_db": 1.0825301154214166, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0591740163031673, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13095682053970678, "1000–3000 Hz_40ms_envelope_correlation": 0.9897540377538737, "1000–3000 Hz_40ms_modulation_mae_db": 1.2957766421612626, "3000–6000 Hz_10ms_log_rms_mae_db": 1.508981575448325, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1984660019611448, "3000–6000 Hz_10ms_envelope_correlation": 0.9773186265823587, "3000–6000 Hz_10ms_modulation_mae_db": 1.048252652729125, "3000–6000 Hz_40ms_log_rms_mae_db": 1.150041530339463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12767398118452272, "3000–6000 Hz_40ms_envelope_correlation": 0.991156703666038, "3000–6000 Hz_40ms_modulation_mae_db": 1.130778125795161, "mr_stft_db": 5.815022511604677, "band_mae_db": 0.37756285112069127, "rms_gain_error_db": -0.1049008309135126, "output_peak": 0.5665113129122372}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3913213905697193, "全帯域_10ms_rms_envelope_nrmse": 0.09699864305312483, "全帯域_10ms_envelope_correlation": 0.9933122607256423, "全帯域_10ms_modulation_mae_db": 0.4868936797583872, "全帯域_40ms_log_rms_mae_db": 0.740510574037419, "全帯域_40ms_rms_envelope_nrmse": 0.05143797402458101, "全帯域_40ms_envelope_correlation": 0.9982488558215894, "全帯域_40ms_modulation_mae_db": 0.6468111309758668, "300–1000 Hz_10ms_log_rms_mae_db": 1.8558251342058856, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11485813797304943, "300–1000 Hz_10ms_envelope_correlation": 0.9923099267429921, "300–1000 Hz_10ms_modulation_mae_db": 0.5137903740669706, "300–1000 Hz_40ms_log_rms_mae_db": 1.148196062468709, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0651039483499925, "300–1000 Hz_40ms_envelope_correlation": 0.9979798338913994, "300–1000 Hz_40ms_modulation_mae_db": 0.5087445911355165, "1000–3000 Hz_10ms_log_rms_mae_db": 1.686151842232233, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18326912194981074, "1000–3000 Hz_10ms_envelope_correlation": 0.976890814525624, "1000–3000 Hz_10ms_modulation_mae_db": 0.8445111106942661, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0334469985184602, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10379853303846912, "1000–3000 Hz_40ms_envelope_correlation": 0.9929899243120425, "1000–3000 Hz_40ms_modulation_mae_db": 0.9338043513842499, "3000–6000 Hz_10ms_log_rms_mae_db": 1.49407241352203, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19281069961802141, "3000–6000 Hz_10ms_envelope_correlation": 0.9782681786142798, "3000–6000 Hz_10ms_modulation_mae_db": 0.0913695710805778, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1431168020918316, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11511138444280677, "3000–6000 Hz_40ms_envelope_correlation": 0.9921868573159717, "3000–6000 Hz_40ms_modulation_mae_db": 0.0678525261734534, "mr_stft_db": 5.797292082630277, "band_mae_db": 0.27949745837630313, "rms_gain_error_db": -0.16517123151342256, "output_peak": 0.5424016935237823}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4281422872972747, "全帯域_10ms_rms_envelope_nrmse": 0.10455453932749281, "全帯域_10ms_envelope_correlation": 0.9923090095998702, "全帯域_10ms_modulation_mae_db": 0.2788748613181213, "全帯域_40ms_log_rms_mae_db": 0.6371423036107426, "全帯域_40ms_rms_envelope_nrmse": 0.05200256421549755, "全帯域_40ms_envelope_correlation": 0.9983744669125929, "全帯域_40ms_modulation_mae_db": 0.31035235200991634, "300–1000 Hz_10ms_log_rms_mae_db": 1.9707354162171495, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13252122300162308, "300–1000 Hz_10ms_envelope_correlation": 0.9898038021098493, "300–1000 Hz_10ms_modulation_mae_db": 0.479797710786473, "300–1000 Hz_40ms_log_rms_mae_db": 1.191828805806992, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07304494908549064, "300–1000 Hz_40ms_envelope_correlation": 0.99762056975901, "300–1000 Hz_40ms_modulation_mae_db": 0.2988067764410609, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6880589210514536, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19420987494141786, "1000–3000 Hz_10ms_envelope_correlation": 0.9738540802464389, "1000–3000 Hz_10ms_modulation_mae_db": 0.6774366165825908, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9647440800321343, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10595308002161773, "1000–3000 Hz_40ms_envelope_correlation": 0.9926038556270239, "1000–3000 Hz_40ms_modulation_mae_db": 0.6868942223937341, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4745750385169716, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15409797566140407, "3000–6000 Hz_10ms_envelope_correlation": 0.9865553863134141, "3000–6000 Hz_10ms_modulation_mae_db": 0.8881028592556875, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1366984143190089, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08697439700318929, "3000–6000 Hz_40ms_envelope_correlation": 0.9956865219975513, "3000–6000 Hz_40ms_modulation_mae_db": 1.2503196312592761, "mr_stft_db": 5.803065471077416, "band_mae_db": 0.22603003859525464, "rms_gain_error_db": -0.21084626435519865, "output_peak": 0.5391387688672041}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.441058944586963, "全帯域_10ms_rms_envelope_nrmse": 0.10268408349692044, "全帯域_10ms_envelope_correlation": 0.9924789099094286, "全帯域_10ms_modulation_mae_db": 0.17233534521934277, "全帯域_40ms_log_rms_mae_db": 0.7152669379006277, "全帯域_40ms_rms_envelope_nrmse": 0.0553049275681753, "全帯域_40ms_envelope_correlation": 0.9979332791628376, "全帯域_40ms_modulation_mae_db": 0.3766478669073208, "300–1000 Hz_10ms_log_rms_mae_db": 1.9853666747548953, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1182947414223501, "300–1000 Hz_10ms_envelope_correlation": 0.9919404164920524, "300–1000 Hz_10ms_modulation_mae_db": 0.5092589708569715, "300–1000 Hz_40ms_log_rms_mae_db": 1.2403877777718608, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0746601052704734, "300–1000 Hz_40ms_envelope_correlation": 0.997265185434286, "300–1000 Hz_40ms_modulation_mae_db": 0.3310618450787864, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7462337618955273, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21471228655894045, "1000–3000 Hz_10ms_envelope_correlation": 0.9685495766780209, "1000–3000 Hz_10ms_modulation_mae_db": 0.5068375676525806, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0357034389223128, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13257378790738727, "1000–3000 Hz_40ms_envelope_correlation": 0.9891001293881028, "1000–3000 Hz_40ms_modulation_mae_db": 0.5903486427868746, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4851342058800079, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1785099495555876, "3000–6000 Hz_10ms_envelope_correlation": 0.9842760300963092, "3000–6000 Hz_10ms_modulation_mae_db": 1.2678284917835196, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1052690689872393, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10952682989480927, "3000–6000 Hz_40ms_envelope_correlation": 0.9950231890877858, "3000–6000 Hz_40ms_modulation_mae_db": 0.7999124366707698, "mr_stft_db": 5.80295156008093, "band_mae_db": 0.3213881714872103, "rms_gain_error_db": -0.1648077494237613, "output_peak": 0.560561183034126}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.988508989171783, "spectral_convergence": 0.3439801930336524, "spectral_convergence_status": "ok", "amplitude_floor": 2.3135422827537554e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.585522343797922, "spectral_convergence": 0.30479226177597574, "spectral_convergence_status": "ok", "amplitude_floor": 1.7511633455315343e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.807324719330449, "spectral_convergence": 0.3464014683052749, "spectral_convergence_status": "ok", "amplitude_floor": 1.6157176909573571e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00044579673236332844, "output_power": 0.000468071880470644, "delta_db": 0.21175670737652777, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00284687947566611, "output_power": 0.0027772269489713336, "delta_db": -0.1075771061637186, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00047388905696308834, "output_power": 0.0004807969854045567, "delta_db": 0.06285055882874796, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013682287067652582, "output_power": 0.0001295438146804684, "delta_db": -0.23742016515177697, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011121195656671286, "output_power": 0.0001004510970873489, "delta_db": -0.4419679711332087, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00022124255174898303, "output_power": 0.00020430926984885112, "delta_db": -0.34580587118604955, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.23456306330667162, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.0780473092818486, "pitch": null, "output_rms_dbfs": -23.808649126651765, "output_peak": 0.5334650781629787, "reference_rms_dbfs": -23.730601817369916, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7937853507667185}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028051035499693775, "log_rms_mae_db": 1.4175312947094507, "rms_envelope_nrmse": 0.09128900925908512, "envelope_correlation": 0.993971670851414, "rms_gain_error_db": -0.07804624096245807, "reference_mean_power": 0.004239368434074771, "output_mean_power": 0.004163864014592811}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.0558356273882855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018964431225846456, "output_power": 0.0018835326858964777, "reference_active": true, "delta_db": -0.029666608991418018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001936564415781366, "output_power": 0.00021817711550484889, "reference_active": true, "delta_db": 0.5177724823332877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.246230567335058e-05, "output_power": 5.344752549377022e-05, "reference_active": true, "delta_db": -0.6769040821432394}], "active_band_count": 3, "active_band_mae_db": 0.4081143911559817}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026381677718961605, "log_rms_mae_db": 0.769300251938301, "rms_envelope_nrmse": 0.04658636302382812, "envelope_correlation": 0.9984190987748623, "rms_gain_error_db": -0.07804216407148144, "reference_mean_power": 0.004260663575263385, "output_mean_power": 0.004184783811366359}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.853150694908015e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017547087963691286, "output_power": 0.0017542511819251979, "reference_active": true, "delta_db": -0.001132754036491628}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.102196929549433e-05, "output_power": 9.7611057279616e-05, "reference_active": true, "delta_db": 0.8089622207685736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0378089105429984e-06, "output_power": 2.8963110525119314e-06, "reference_active": true, "delta_db": -0.2071525008738596}], "active_band_count": 3, "active_band_mae_db": 0.33908249189297496}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002736722962808348, "log_rms_mae_db": 1.9774650586182836, "rms_envelope_nrmse": 0.1101888676957052, "envelope_correlation": 0.9925365554907648, "rms_gain_error_db": -0.10541606227059318, "reference_mean_power": 0.0024581880191069405, "output_mean_power": 0.0023992388844652416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.077905340514435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019022535341174436, "output_power": 0.0018858414485651627, "reference_active": true, "delta_db": -0.037632228456182984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012849688146439824, "output_power": 0.00015476167701630638, "reference_active": true, "delta_db": 0.8077083939133206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.095154000220668e-05, "output_power": 3.635164792731227e-05, "reference_active": true, "delta_db": -0.5174613476493912}], "active_band_count": 3, "active_band_mae_db": 0.45426732333963155}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025339024948019116, "log_rms_mae_db": 1.1858741811237759, "rms_envelope_nrmse": 0.06135134323763531, "envelope_correlation": 0.9977018433932854, "rms_gain_error_db": -0.10541654422397269, "reference_mean_power": 0.002470549817529662, "output_mean_power": 0.002411303970388848}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.981046082661891e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017987268874748684, "output_power": 0.001794349814354876, "reference_active": true, "delta_db": -0.010581124762881412}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5079379851283876e-05, "output_power": 7.130547576911532e-05, "reference_active": true, "delta_db": 1.1213384014345986}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4721075346991203e-06, "output_power": 2.112200207556902e-06, "reference_active": true, "delta_db": -0.6833227732774533}], "active_band_count": 3, "active_band_mae_db": 0.6050807664916444}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.470771529403075e-05, "log_rms_mae_db": 1.6935630708034097, "rms_envelope_nrmse": 0.22194419686641872, "envelope_correlation": 0.9655893581122387, "rms_gain_error_db": -0.32748818695466647, "reference_mean_power": 0.00019492290635367423, "output_mean_power": 0.00018076488096091376}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7570632979474236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.555963644612216e-05, "output_power": 6.521566969005499e-05, "reference_active": true, "delta_db": -0.6393790131071085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.293127801504832e-05, "output_power": 8.706752814383511e-06, "reference_active": true, "delta_db": -1.7178523392174252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505165714035109e-06, "output_power": 8.064907326242213e-06, "reference_active": true, "delta_db": 0.933410169281933}], "active_band_count": 3, "active_band_mae_db": 1.0968805072021555}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.885515861839929e-05, "log_rms_mae_db": 1.0765140938476025, "rms_envelope_nrmse": 0.14241470958890065, "envelope_correlation": 0.9857263605987857, "rms_gain_error_db": -0.3275185995009201, "reference_mean_power": 0.00019590175284412906, "output_mean_power": 0.00018167135773731346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6226758883463992e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.225076970752032e-05, "output_power": 6.261154871230709e-05, "reference_active": true, "delta_db": -0.621880318124513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.071619770568594e-06, "output_power": 4.6181917957242226e-06, "reference_active": true, "delta_db": -1.188326010446294}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6400879277669946e-07, "output_power": 3.9594081409948807e-07, "reference_active": true, "delta_db": 0.3651839737454462}], "active_band_count": 3, "active_band_mae_db": 0.7251301007720844}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012408236865898127, "log_rms_mae_db": 1.4986790651896642, "rms_envelope_nrmse": 0.18523953795999168, "envelope_correlation": 0.9809907150800491, "rms_gain_error_db": -0.5160724687061105, "reference_mean_power": 0.00018911407706020183, "output_mean_power": 0.00016792548476970339}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8583468534278485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010082037892943086, "output_power": 8.515532082070942e-05, "reference_active": true, "delta_db": -0.7333653590530514}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9440078487669303e-05, "output_power": 2.2648426055493935e-05, "reference_active": true, "delta_db": 0.6634001223230792}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.009136082695238e-06, "output_power": 5.175922320479164e-06, "reference_active": true, "delta_db": 0.14224919594283497}], "active_band_count": 3, "active_band_mae_db": 0.5130048924396552}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.970743476594613e-05, "log_rms_mae_db": 1.1758517930902255, "rms_envelope_nrmse": 0.10380526564316002, "envelope_correlation": 0.9954571270650501, "rms_gain_error_db": -0.5160760150218555, "reference_mean_power": 0.00019006483009233416, "output_mean_power": 0.0001687695763576269}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7984096251410197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010012152579842322, "output_power": 8.651378017376434e-05, "reference_active": true, "delta_db": -0.6344217070270219}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0693666398121268e-05, "output_power": 1.1906919849263602e-05, "reference_active": true, "delta_db": 0.466727985755615}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.760055744312906e-07, "output_power": 3.125425311104919e-07, "reference_active": true, "delta_db": 0.5399127142774295}], "active_band_count": 3, "active_band_mae_db": 0.5470208023533555}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.783755278328271, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7ccda977d31d6a2e626e6d1022c7e09209e122ab33eb37da1a295260009e5a49", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9704782264571916, "standard_deviation": 0.003807015618295686, "minimum": 0.9655893581122387, "maximum": 0.976890814525624, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7052304391380488, "standard_deviation": 0.02453082983461209, "minimum": 1.678246184759322, "maximum": 1.7462337618955273, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.714415514940529, "standard_deviation": 0.26456180675275415, "minimum": 0.45076581175069497, "maximum": 1.0968805072021555, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20632779156638695, "standard_deviation": 0.013634850331789435, "minimum": 0.18326912194981074, "maximum": 0.2235092828244344, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9900210485796942, "standard_deviation": 0.0023506232084970925, "minimum": 0.9857263605987857, "maximum": 0.9929899243120425, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0352341732626393, "standard_deviation": 0.03395731467516949, "minimum": 0.9647440800321343, "maximum": 1.0765140938476025, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6695263356745896, "standard_deviation": 0.3283059772238295, "minimum": 0.2795675560621454, "maximum": 1.2957766421612626, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12199901855857315, "standard_deviation": 0.013401161940606292, "minimum": 0.10379853303846912, "maximum": 0.14241470958890065, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9828586826896057, "standard_deviation": 0.004368497838386953, "minimum": 0.9773186265823587, "maximum": 0.9892906346587226, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4888563372410215, "standard_deviation": 0.017381372640880105, "minimum": 1.460013513105017, "maximum": 1.5099963133496301, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6026134793188684, "standard_deviation": 0.42166506468450055, "minimum": 0.0913695710805778, "maximum": 1.2678284917835196, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1736915149441368, "standard_deviation": 0.021165334986728776, "minimum": 0.14115138646495, "maximum": 0.1984660019611448, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9948806394681425, "standard_deviation": 0.0021606534388802956, "minimum": 0.991156703666038, "maximum": 0.9975327923411317, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1404333998466583, "standard_deviation": 0.032744535508225765, "minimum": 1.0805110518294332, "maximum": 1.1758517930902255, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5972728056326961, "standard_deviation": 0.4371095984466738, "minimum": 0.0678525261734534, "maximum": 1.2503196312592761, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09755335633618034, "standard_deviation": 0.019801503121562803, "minimum": 0.07182732743677213, "maximum": 0.12767398118452272, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9918758369025755, "standard_deviation": 0.0010258338168280734, "minimum": 0.9898038021098493, "maximum": 0.9927537520591516, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.902200340700789, "standard_deviation": 0.06598711151547276, "minimum": 1.8363867211542964, "maximum": 1.9853666747548953, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5310406245083769, "standard_deviation": 0.11790301471616041, "minimum": 0.33227069379665836, "maximum": 0.7037668892577003, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.11651182755085913, "standard_deviation": 0.007677562271934676, "minimum": 0.10996266938973384, "maximum": 0.13252122300162308, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9977349881002266, "standard_deviation": 0.0002478349723071031, "minimum": 0.997265185434286, "maximum": 0.9980502781366684, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1631530974011266, "standard_deviation": 0.046993728917679584, "minimum": 1.0916881779192915, "maximum": 1.2403877777718608, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.3961114921860808, "standard_deviation": 0.19590373177914405, "minimum": 0.12706928638428586, "maximum": 0.6050807664916444, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06517248989744734, "standard_deviation": 0.005921383660730391, "minimum": 0.05850623019327885, "maximum": 0.0746601052704734, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2751228111352879, "standard_deviation": 0.05413188468124327, "minimum": 0.21489513333781107, "maximum": 0.37756285112069127, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.800554303938046, "standard_deviation": 0.011630438148990868, "minimum": 5.780161727672048, "maximum": 5.815879659607956, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5849385113606278, "standard_deviation": 0.06940759133280107, "minimum": 0.5334650781629787, "maximum": 0.7460171015925696, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.11122312121216195, "standard_deviation": 0.06510438287936386, "minimum": -0.21084626435519865, "maximum": -0.009152634546971572, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9929076199697819, "standard_deviation": 0.0007039224077310924, "minimum": 0.9918105963874977, "maximum": 0.993971670851414, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4447470473670534, "standard_deviation": 0.045206392183727975, "minimum": 1.3913213905697193, "maximum": 1.5124393608175477, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.36952972724295907, "standard_deviation": 0.18106559970392627, "minimum": 0.07712791044457437, "maximum": 0.6090085635043617, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09928185729211295, "standard_deviation": 0.005109562906154393, "minimum": 0.09128900925908512, "maximum": 0.10638024337342558, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9982165228830937, "standard_deviation": 0.00017132205053886873, "minimum": 0.9979332791628376, "maximum": 0.9984190987748623, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7443627440926232, "standard_deviation": 0.05618835471352762, "minimum": 0.6371423036107426, "maximum": 0.8290811239516677, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.45496444039575556, "standard_deviation": 0.19970183822367318, "minimum": 0.16393418601609322, "maximum": 0.742397254088658, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.050562425380394, "standard_deviation": 0.0026506653379584784, "minimum": 0.04658636302382812, "maximum": 0.0553049275681753, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0016", "method": "D", "status": "measured", "mos": {"status": "measured", "value": 2.60896372795105, "predicted_mos": 2.60896372795105, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b7349f3e55b7eb6b42ac098e91be6f4075072d98bc03b0e2cf99530c4de84c5e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8565217000432312, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.373018126068295, "全帯域_10ms_rms_envelope_nrmse": 0.0983236802871702, "全帯域_10ms_envelope_correlation": 0.9932202814366207, "全帯域_10ms_modulation_mae_db": 0.27617304072531623, "全帯域_40ms_log_rms_mae_db": 0.7020748848882287, "全帯域_40ms_rms_envelope_nrmse": 0.05003537316253767, "全帯域_40ms_envelope_correlation": 0.9984869068359128, "全帯域_40ms_modulation_mae_db": 0.5071441906486555, "300–1000 Hz_10ms_log_rms_mae_db": 2.0069220387794835, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12351245581201103, "300–1000 Hz_10ms_envelope_correlation": 0.9913330696685966, "300–1000 Hz_10ms_modulation_mae_db": 0.3636277539236086, "300–1000 Hz_40ms_log_rms_mae_db": 1.2625004780502853, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0699037068194389, "300–1000 Hz_40ms_envelope_correlation": 0.997966183073165, "300–1000 Hz_40ms_modulation_mae_db": 0.9041084367595476, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9192842884753574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2038003355419595, "1000–3000 Hz_10ms_envelope_correlation": 0.9712471701453431, "1000–3000 Hz_10ms_modulation_mae_db": 0.46738196043550073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.209679401964605, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11394864802223695, "1000–3000 Hz_40ms_envelope_correlation": 0.9905701776889437, "1000–3000 Hz_40ms_modulation_mae_db": 0.30384761997790805, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7395426631689532, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17733657900237002, "3000–6000 Hz_10ms_envelope_correlation": 0.982145671270847, "3000–6000 Hz_10ms_modulation_mae_db": 0.8336328371950419, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3615631690564725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1076798903985608, "3000–6000 Hz_40ms_envelope_correlation": 0.9940724163696585, "3000–6000 Hz_40ms_modulation_mae_db": 0.6524819629659084, "mr_stft_db": 5.921200991826385, "band_mae_db": 0.20248515937635694, "rms_gain_error_db": -0.20218937869785292, "output_peak": 0.6373299929456128}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3790964862912534, "全帯域_10ms_rms_envelope_nrmse": 0.09191202874493458, "全帯域_10ms_envelope_correlation": 0.9938728840632947, "全帯域_10ms_modulation_mae_db": 0.44618888553461683, "全帯域_40ms_log_rms_mae_db": 0.672135408936242, "全帯域_40ms_rms_envelope_nrmse": 0.04472080997089459, "全帯域_40ms_envelope_correlation": 0.998504232120325, "全帯域_40ms_modulation_mae_db": 0.6320199888194077, "300–1000 Hz_10ms_log_rms_mae_db": 1.9391519060875015, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11619895664642754, "300–1000 Hz_10ms_envelope_correlation": 0.991679392549504, "300–1000 Hz_10ms_modulation_mae_db": 0.928852285978731, "300–1000 Hz_40ms_log_rms_mae_db": 1.2447083160677173, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06315784326067372, "300–1000 Hz_40ms_envelope_correlation": 0.9974995260689408, "300–1000 Hz_40ms_modulation_mae_db": 0.952986041916689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.982442663254644, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21171653160465598, "1000–3000 Hz_10ms_envelope_correlation": 0.9686449585489568, "1000–3000 Hz_10ms_modulation_mae_db": 1.035559302747026, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2245672513931882, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11456784998794033, "1000–3000 Hz_40ms_envelope_correlation": 0.9909692731016976, "1000–3000 Hz_40ms_modulation_mae_db": 0.7576700940893016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7522915921407478, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16844033550942844, "3000–6000 Hz_10ms_envelope_correlation": 0.9835121879261767, "3000–6000 Hz_10ms_modulation_mae_db": 0.5131704734867416, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3633522360751273, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07661224246952766, "3000–6000 Hz_40ms_envelope_correlation": 0.9965582677991074, "3000–6000 Hz_40ms_modulation_mae_db": 0.17147708644517098, "mr_stft_db": 5.940792625833793, "band_mae_db": 0.22342473080384403, "rms_gain_error_db": -0.02355550293714164, "output_peak": 0.6551854930908891}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4376880422355065, "全帯域_10ms_rms_envelope_nrmse": 0.10435520853460027, "全帯域_10ms_envelope_correlation": 0.9921055043246491, "全帯域_10ms_modulation_mae_db": 0.36919333421127726, "全帯域_40ms_log_rms_mae_db": 0.7314634660554595, "全帯域_40ms_rms_envelope_nrmse": 0.0568962417465809, "全帯域_40ms_envelope_correlation": 0.9976267551959764, "全帯域_40ms_modulation_mae_db": 0.4243831145582468, "300–1000 Hz_10ms_log_rms_mae_db": 2.0729071610782235, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12535603081315577, "300–1000 Hz_10ms_envelope_correlation": 0.9903606177928996, "300–1000 Hz_10ms_modulation_mae_db": 0.874059937746293, "300–1000 Hz_40ms_log_rms_mae_db": 1.3248762613407723, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07465980163353755, "300–1000 Hz_40ms_envelope_correlation": 0.9966567104856671, "300–1000 Hz_40ms_modulation_mae_db": 0.7140843925316721, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8935505087244417, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2074183734964128, "1000–3000 Hz_10ms_envelope_correlation": 0.9701193583835043, "1000–3000 Hz_10ms_modulation_mae_db": 0.533093709899201, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1830141817786384, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11705691635562777, "1000–3000 Hz_40ms_envelope_correlation": 0.9900145300149444, "1000–3000 Hz_40ms_modulation_mae_db": 0.2887768594110293, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6892507864381205, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1717805498971356, "3000–6000 Hz_10ms_envelope_correlation": 0.9830566464868915, "3000–6000 Hz_10ms_modulation_mae_db": 1.0369724848602029, "3000–6000 Hz_40ms_log_rms_mae_db": 1.334756207523761, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08954760476452688, "3000–6000 Hz_40ms_envelope_correlation": 0.9959005013187135, "3000–6000 Hz_40ms_modulation_mae_db": 1.2412988364858661, "mr_stft_db": 5.936388090680832, "band_mae_db": 0.16771875386028454, "rms_gain_error_db": -0.09014560360081231, "output_peak": 0.5441635697658971}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.370486659926694, "全帯域_10ms_rms_envelope_nrmse": 0.09504346974573828, "全帯域_10ms_envelope_correlation": 0.9937227580759426, "全帯域_10ms_modulation_mae_db": 0.3186885252679577, "全帯域_40ms_log_rms_mae_db": 0.7136828579265111, "全帯域_40ms_rms_envelope_nrmse": 0.06020214046785617, "全帯域_40ms_envelope_correlation": 0.9976682140661984, "全帯域_40ms_modulation_mae_db": 0.3878223616251298, "300–1000 Hz_10ms_log_rms_mae_db": 2.016021097774425, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12009993270085541, "300–1000 Hz_10ms_envelope_correlation": 0.9919898115964914, "300–1000 Hz_10ms_modulation_mae_db": 0.5063000403651773, "300–1000 Hz_40ms_log_rms_mae_db": 1.343652294737511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07706188701974724, "300–1000 Hz_40ms_envelope_correlation": 0.9974261334310437, "300–1000 Hz_40ms_modulation_mae_db": 0.4513874981666504, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9718764211961384, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22580774781345134, "1000–3000 Hz_10ms_envelope_correlation": 0.964355050169968, "1000–3000 Hz_10ms_modulation_mae_db": 0.3024960360786591, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2134146257183245, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1395939354520779, "1000–3000 Hz_40ms_envelope_correlation": 0.9865729993926762, "1000–3000 Hz_40ms_modulation_mae_db": 0.2650188640741334, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7032827040606229, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17226011838951336, "3000–6000 Hz_10ms_envelope_correlation": 0.9850620388885003, "3000–6000 Hz_10ms_modulation_mae_db": 1.4509022357905836, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3178704938702668, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10050209973277617, "3000–6000 Hz_40ms_envelope_correlation": 0.9957348951406318, "3000–6000 Hz_40ms_modulation_mae_db": 1.3532313140650267, "mr_stft_db": 5.913250759067367, "band_mae_db": 0.2612628269536747, "rms_gain_error_db": -0.20806586325825907, "output_peak": 0.5384247974221634}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4089487038004695, "全帯域_10ms_rms_envelope_nrmse": 0.10345421189154116, "全帯域_10ms_envelope_correlation": 0.9922324319954231, "全帯域_10ms_modulation_mae_db": 0.6539613739530176, "全帯域_40ms_log_rms_mae_db": 0.7361067142120111, "全帯域_40ms_rms_envelope_nrmse": 0.05336482531623636, "全帯域_40ms_envelope_correlation": 0.9979048405829879, "全帯域_40ms_modulation_mae_db": 0.7192633592675862, "300–1000 Hz_10ms_log_rms_mae_db": 2.0576620399125454, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1160713647443291, "300–1000 Hz_10ms_envelope_correlation": 0.9918780940187183, "300–1000 Hz_10ms_modulation_mae_db": 1.0520125783697196, "300–1000 Hz_40ms_log_rms_mae_db": 1.3068948159271143, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.056006457203576346, "300–1000 Hz_40ms_envelope_correlation": 0.9983756867879985, "300–1000 Hz_40ms_modulation_mae_db": 0.461916003542784, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9208484612723382, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20887812052339025, "1000–3000 Hz_10ms_envelope_correlation": 0.9698114784490658, "1000–3000 Hz_10ms_modulation_mae_db": 0.13070953479027733, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2071160563244632, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12085962841866363, "1000–3000 Hz_40ms_envelope_correlation": 0.989391153292861, "1000–3000 Hz_40ms_modulation_mae_db": 0.21566551828148542, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7349142901255343, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1565556792740827, "3000–6000 Hz_10ms_envelope_correlation": 0.9860951082417162, "3000–6000 Hz_10ms_modulation_mae_db": 0.7972776407805925, "3000–6000 Hz_40ms_log_rms_mae_db": 1.34904733792671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1054668409388772, "3000–6000 Hz_40ms_envelope_correlation": 0.994016936664368, "3000–6000 Hz_40ms_modulation_mae_db": 0.862395248013854, "mr_stft_db": 5.938624810811926, "band_mae_db": 0.2520620198258893, "rms_gain_error_db": -0.07795243916834238, "output_peak": 0.5592384394990679}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3808741393198065, "全帯域_10ms_rms_envelope_nrmse": 0.09056846515710198, "全帯域_10ms_envelope_correlation": 0.9940760814029966, "全帯域_10ms_modulation_mae_db": 0.2091542971261253, "全帯域_40ms_log_rms_mae_db": 0.6623833523245205, "全帯域_40ms_rms_envelope_nrmse": 0.043955366935102304, "全帯域_40ms_envelope_correlation": 0.998620487287675, "全帯域_40ms_modulation_mae_db": 0.2875252652566342, "300–1000 Hz_10ms_log_rms_mae_db": 1.9578854794948983, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12715652964069005, "300–1000 Hz_10ms_envelope_correlation": 0.9900305096644362, "300–1000 Hz_10ms_modulation_mae_db": 0.7292785995206353, "300–1000 Hz_40ms_log_rms_mae_db": 1.241892042205895, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.067860778577525, "300–1000 Hz_40ms_envelope_correlation": 0.9971989598777108, "300–1000 Hz_40ms_modulation_mae_db": 0.9502069827814705, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9047522893376339, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18116188264420363, "1000–3000 Hz_10ms_envelope_correlation": 0.977148323364804, "1000–3000 Hz_10ms_modulation_mae_db": 0.655623422385342, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1588123819465233, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10129816025420912, "1000–3000 Hz_40ms_envelope_correlation": 0.9925336425904429, "1000–3000 Hz_40ms_modulation_mae_db": 0.39930567880152174, "3000–6000 Hz_10ms_log_rms_mae_db": 1.738263000936671, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.14590269632379588, "3000–6000 Hz_10ms_envelope_correlation": 0.9883441467382537, "3000–6000 Hz_10ms_modulation_mae_db": 0.49577652951960793, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3475445994750868, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07488039901520467, "3000–6000 Hz_40ms_envelope_correlation": 0.99713283315553, "3000–6000 Hz_40ms_modulation_mae_db": 0.26546081753212236, "mr_stft_db": 5.9281821914252575, "band_mae_db": 0.16763274644025683, "rms_gain_error_db": -0.09256362262490675, "output_peak": 0.5988094796244725}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.372349860896093, "全帯域_10ms_rms_envelope_nrmse": 0.10306572120508173, "全帯域_10ms_envelope_correlation": 0.9922955217895907, "全帯域_10ms_modulation_mae_db": 0.8763199156573879, "全帯域_40ms_log_rms_mae_db": 0.7310389917115764, "全帯域_40ms_rms_envelope_nrmse": 0.05712120899711559, "全帯域_40ms_envelope_correlation": 0.9975537968309738, "全帯域_40ms_modulation_mae_db": 1.145295931049658, "300–1000 Hz_10ms_log_rms_mae_db": 2.0105069710984504, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1239140466455427, "300–1000 Hz_10ms_envelope_correlation": 0.9905077890751318, "300–1000 Hz_10ms_modulation_mae_db": 1.3702899993094686, "300–1000 Hz_40ms_log_rms_mae_db": 1.278041283770191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0668649448130298, "300–1000 Hz_40ms_envelope_correlation": 0.9972367213753033, "300–1000 Hz_40ms_modulation_mae_db": 1.3186450294426075, "1000–3000 Hz_10ms_log_rms_mae_db": 1.890757548598544, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.200089776174386, "1000–3000 Hz_10ms_envelope_correlation": 0.9725238923624235, "1000–3000 Hz_10ms_modulation_mae_db": 0.8916615389534034, "1000–3000 Hz_40ms_log_rms_mae_db": 1.24452950805851, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12460517190019312, "1000–3000 Hz_40ms_envelope_correlation": 0.9898825622118346, "1000–3000 Hz_40ms_modulation_mae_db": 0.8371523562366604, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7533810976103164, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18414011784705128, "3000–6000 Hz_10ms_envelope_correlation": 0.980870566109777, "3000–6000 Hz_10ms_modulation_mae_db": 1.1401354433407842, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3471849493304522, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09206839737766684, "3000–6000 Hz_40ms_envelope_correlation": 0.9952368294175598, "3000–6000 Hz_40ms_modulation_mae_db": 1.0734035779844306, "mr_stft_db": 5.928606835009153, "band_mae_db": 0.16548306207131291, "rms_gain_error_db": -0.023850236394374766, "output_peak": 0.5395762270215805}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3924271187236192, "全帯域_10ms_rms_envelope_nrmse": 0.09688270370778276, "全帯域_10ms_envelope_correlation": 0.9936091452118284, "全帯域_10ms_modulation_mae_db": 0.6350607678497794, "全帯域_40ms_log_rms_mae_db": 0.7369423400700486, "全帯域_40ms_rms_envelope_nrmse": 0.05693865095268912, "全帯域_40ms_envelope_correlation": 0.9981391868616472, "全帯域_40ms_modulation_mae_db": 0.8912105313223421, "300–1000 Hz_10ms_log_rms_mae_db": 2.081114192014001, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11536446237325713, "300–1000 Hz_10ms_envelope_correlation": 0.9927558768907285, "300–1000 Hz_10ms_modulation_mae_db": 0.6752871814520787, "300–1000 Hz_40ms_log_rms_mae_db": 1.3216003799928868, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06942362700480931, "300–1000 Hz_40ms_envelope_correlation": 0.9982030970623382, "300–1000 Hz_40ms_modulation_mae_db": 0.7730779629918768, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8777594159781237, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20574464144372062, "1000–3000 Hz_10ms_envelope_correlation": 0.970355873097787, "1000–3000 Hz_10ms_modulation_mae_db": 0.6387106879752581, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1716288635006302, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11217055887374348, "1000–3000 Hz_40ms_envelope_correlation": 0.9910461786621845, "1000–3000 Hz_40ms_modulation_mae_db": 0.4669633049394812, "3000–6000 Hz_10ms_log_rms_mae_db": 1.730152736410698, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16507169057638316, "3000–6000 Hz_10ms_envelope_correlation": 0.9845942475101318, "3000–6000 Hz_10ms_modulation_mae_db": 0.3622615178148035, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3362292763651449, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09063020432414068, "3000–6000 Hz_40ms_envelope_correlation": 0.9960699466276661, "3000–6000 Hz_40ms_modulation_mae_db": 0.41288857764715964, "mr_stft_db": 5.906654441936049, "band_mae_db": 0.2521350035947904, "rms_gain_error_db": -0.2512164409645542, "output_peak": 0.5559890267095262}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.1816256798350135, "spectral_convergence": 0.3434306454148203, "spectral_convergence_status": "ok", "amplitude_floor": 2.3135422827537554e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.748977814352, "spectral_convergence": 0.3065352913337704, "spectral_convergence_status": "ok", "amplitude_floor": 1.7511633455315343e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.83299948129214, "spectral_convergence": 0.3400716084111398, "spectral_convergence_status": "ok", "amplitude_floor": 1.6157176909573571e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00044579673236332844, "output_power": 0.00048115591785806905, "delta_db": 0.33148950831091345, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00284687947566611, "output_power": 0.002646171797416493, "delta_db": -0.3175104495299088, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00047388905696308834, "output_power": 0.00046600226420920516, "delta_db": -0.07288653252702787, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013682287067652582, "output_power": 0.0001333011276841209, "delta_db": -0.11324874684299109, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011121195656671286, "output_power": 0.00010823460289557788, "delta_db": -0.11785353325373879, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00022124255174898303, "output_power": 0.00020829384661514774, "delta_db": -0.26192218579356163, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.20248515937635694, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.20218937869785292, "pitch": null, "output_rms_dbfs": -23.93279119606777, "output_peak": 0.6373299929456128, "reference_rms_dbfs": -23.730601817369916, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.921200991826385}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028051035499693775, "log_rms_mae_db": 1.373018126068295, "rms_envelope_nrmse": 0.0983236802871702, "envelope_correlation": 0.9932202814366207, "rms_gain_error_db": -0.20218733977093034, "reference_mean_power": 0.004239368434074771, "output_mean_power": 0.004046526862831814}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.0558356273882855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018964431225846456, "output_power": 0.0017833316510625636, "reference_active": true, "delta_db": -0.26707704417264994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001936564415781366, "output_power": 0.00017907108681022561, "reference_active": true, "delta_db": -0.3400647804368692}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.246230567335058e-05, "output_power": 5.9358139697609976e-05, "reference_active": true, "delta_db": -0.22137729756642952}], "active_band_count": 3, "active_band_mae_db": 0.27617304072531623}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026381677718961605, "log_rms_mae_db": 0.7020748848882287, "rms_envelope_nrmse": 0.05003537316253767, "envelope_correlation": 0.9984869068359128, "rms_gain_error_db": -0.20218341922930533, "reference_mean_power": 0.004260663575263385, "output_mean_power": 0.004066856996016091}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.853150694908015e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017547087963691286, "output_power": 0.001658916660390163, "reference_active": true, "delta_db": -0.24380484450266762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.102196929549433e-05, "output_power": 7.260442263286002e-05, "reference_active": true, "delta_db": -0.47639718644096374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0378089105429984e-06, "output_power": 2.526023563989885e-06, "reference_active": true, "delta_db": -0.801230541002335}], "active_band_count": 3, "active_band_mae_db": 0.5071441906486555}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002736722962808348, "log_rms_mae_db": 2.0069220387794835, "rms_envelope_nrmse": 0.12351245581201103, "envelope_correlation": 0.9913330696685966, "rms_gain_error_db": -0.3754050692874691, "reference_mean_power": 0.0024581880191069405, "output_mean_power": 0.0022546264143111622}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.077905340514435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019022535341174436, "output_power": 0.001763439159236823, "reference_active": true, "delta_db": -0.3290791904875743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012849688146439824, "output_power": 0.00011250619979252376, "reference_active": true, "delta_db": -0.5771613231263432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.095154000220668e-05, "output_power": 3.92469546443742e-05, "reference_active": true, "delta_db": -0.18464274815690837}], "active_band_count": 3, "active_band_mae_db": 0.3636277539236086}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025339024948019116, "log_rms_mae_db": 1.2625004780502853, "rms_envelope_nrmse": 0.0699037068194389, "envelope_correlation": 0.997966183073165, "rms_gain_error_db": -0.37540684418862147, "reference_mean_power": 0.002470549817529662, "output_mean_power": 0.002265963610891067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.981046082661891e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017987268874748684, "output_power": 0.0016757752478509034, "reference_active": true, "delta_db": -0.3074945520905923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5079379851283876e-05, "output_power": 4.705215048452962e-05, "reference_active": true, "delta_db": -0.6840956447988232}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4721075346991203e-06, "output_power": 1.6633890834420001e-06, "reference_active": true, "delta_db": -1.720735113389227}], "active_band_count": 3, "active_band_mae_db": 0.9041084367595476}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.470771529403075e-05, "log_rms_mae_db": 1.9192842884753574, "rms_envelope_nrmse": 0.2038003355419595, "envelope_correlation": 0.9712471701453431, "rms_gain_error_db": 0.002690140658529534, "reference_mean_power": 0.00019492290635367423, "output_mean_power": 0.00019504368441918931}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7570632979474236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.555963644612216e-05, "output_power": 7.11313647390486e-05, "reference_active": true, "delta_db": -0.2622871842530739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.293127801504832e-05, "output_power": 1.1874755280907505e-05, "reference_active": true, "delta_db": -0.37016780593642085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505165714035109e-06, "output_power": 7.766537677518543e-06, "reference_active": true, "delta_db": 0.7696908911170075}], "active_band_count": 3, "active_band_mae_db": 0.46738196043550073}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.885515861839929e-05, "log_rms_mae_db": 1.209679401964605, "rms_envelope_nrmse": 0.11394864802223695, "envelope_correlation": 0.9905701776889437, "rms_gain_error_db": 0.002677458777163458, "reference_mean_power": 0.00019590175284412906, "output_mean_power": 0.00019602256501367315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6226758883463992e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.225076970752032e-05, "output_power": 6.74122547036848e-05, "reference_active": true, "delta_db": -0.30103625102769715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.071619770568594e-06, "output_power": 5.845416248604588e-06, "reference_active": true, "delta_db": -0.16489124008791672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6400879277669946e-07, "output_power": 4.033419663511015e-07, "reference_active": true, "delta_db": 0.44561536881811026}], "active_band_count": 3, "active_band_mae_db": 0.30384761997790805}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012408236865898127, "log_rms_mae_db": 1.7395426631689532, "rms_envelope_nrmse": 0.17733657900237002, "envelope_correlation": 0.982145671270847, "rms_gain_error_db": -0.3831584904940444, "reference_mean_power": 0.00018911407706020183, "output_mean_power": 0.00017314422409131784}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8583468534278485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010082037892943086, "output_power": 9.326012831681004e-05, "reference_active": true, "delta_db": -0.33852316718345904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9440078487669303e-05, "output_power": 1.694071551463185e-05, "reference_active": true, "delta_db": -0.5976626460868563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.009136082695238e-06, "output_power": 3.4937478699544084e-06, "reference_active": true, "delta_db": -1.5647126983148105}], "active_band_count": 3, "active_band_mae_db": 0.8336328371950419}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.970743476594613e-05, "log_rms_mae_db": 1.3615631690564725, "rms_envelope_nrmse": 0.1076798903985608, "envelope_correlation": 0.9940724163696585, "rms_gain_error_db": -0.383160783865888, "reference_mean_power": 0.00019006483009233416, "output_mean_power": 0.00017401459831521826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7984096251410197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010012152579842322, "output_power": 9.17086514449559e-05, "reference_active": true, "delta_db": -0.3811715211163322}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0693666398121268e-05, "output_power": 9.404880845228192e-06, "reference_active": true, "delta_db": -0.5577333398891433}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.760055744312906e-07, "output_power": 2.183050341497737e-07, "reference_active": true, "delta_db": -1.0185410278922498}], "active_band_count": 3, "active_band_mae_db": 0.6524819629659084}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7823003671340751, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "947b9a27f69f27f80211949d7f79de6699610c2a9ce2e14f6f4e72274e89709f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9705257630652315, "standard_deviation": 0.003599416756890217, "minimum": 0.964355050169968, "maximum": 0.977148323364804, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9201589496046525, "standard_deviation": 0.03810274727967113, "minimum": 1.8777594159781237, "maximum": 1.982442663254644, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5819045241580835, "standard_deviation": 0.29460807315739906, "minimum": 0.13070953479027733, "maximum": 1.035559302747026, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20557717615527252, "standard_deviation": 0.012482314799580828, "minimum": 0.18116188264420363, "maximum": 0.22580774781345134, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9901225646194481, "standard_deviation": 0.0017232560980079642, "minimum": 0.9865729993926762, "maximum": 0.9925336425904429, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2015952838356103, "standard_deviation": 0.028486106979222332, "minimum": 1.1588123819465233, "maximum": 1.24452950805851, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4418000369764401, "standard_deviation": 0.23401821532346634, "minimum": 0.21566551828148542, "maximum": 0.8371523562366604, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11801260865808655, "standard_deviation": 0.01108486029977197, "minimum": 0.10129816025420912, "maximum": 0.1395939354520779, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9842100766465367, "standard_deviation": 0.0023527751556622783, "minimum": 0.980870566109777, "maximum": 0.9883441467382537, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.730134858861458, "standard_deviation": 0.022681870668761615, "minimum": 1.6892507864381205, "maximum": 1.7533810976103164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8287661453485448, "standard_deviation": 0.3696481966571505, "minimum": 0.3622615178148035, "maximum": 1.4509022357905836, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16768597085247006, "standard_deviation": 0.011996187259976551, "minimum": 0.14590269632379588, "maximum": 0.18414011784705128, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955903283116544, "standard_deviation": 0.0011063419804333192, "minimum": 0.994016936664368, "maximum": 0.99713283315553, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3446935337028778, "standard_deviation": 0.01490873511927559, "minimum": 1.3178704938702668, "maximum": 1.3633522360751273, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7540796776424423, "standard_deviation": 0.4495986845995036, "minimum": 0.17147708644517098, "maximum": 1.3532313140650267, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09217345987766011, "standard_deviation": 0.012171505608814191, "minimum": 0.07488039901520467, "maximum": 0.1076798903985608, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913168951570632, "standard_deviation": 0.0009406685909211499, "minimum": 0.9900305096644362, "maximum": 0.9927558768907285, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.017771360779941, "standard_deviation": 0.051525883958958636, "minimum": 1.9391519060875015, "maximum": 2.081114192014001, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.812463547083214, "standard_deviation": 0.3177648042935201, "minimum": 0.3636277539236086, "maximum": 1.3702899993094686, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12095922242203358, "standard_deviation": 0.004653409341338493, "minimum": 0.11536446237325713, "maximum": 0.12715652964069005, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9975703772702709, "standard_deviation": 0.0005751285937121269, "minimum": 0.9966567104856671, "maximum": 0.9983756867879985, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2905207340115465, "standard_deviation": 0.03899032605879745, "minimum": 1.241892042205895, "maximum": 1.343652294737511, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8158015435166622, "standard_deviation": 0.2847902341889865, "minimum": 0.4513874981666504, "maximum": 1.3186450294426075, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06811738079154223, "standard_deviation": 0.0065531957878573735, "minimum": 0.056006457203576346, "maximum": 0.07706188701974724, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.2115255378658012, "standard_deviation": 0.04132618208391926, "minimum": 0.16548306207131291, "maximum": 0.2612628269536747, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.926712593323845, "standard_deviation": 0.012267492058997955, "minimum": 5.906654441936049, "maximum": 5.940792625833793, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.5785896282599012, "standard_deviation": 0.04618575484552696, "minimum": 0.5384247974221634, "maximum": 0.6551854930908891, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.1211923859557805, "standard_deviation": 0.08756664854536399, "minimum": -0.2512164409645542, "maximum": -0.02355550293714164, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.9931418260375433, "standard_deviation": 0.0008094525941288854, "minimum": 0.9921055043246491, "maximum": 0.9940760814029966, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3893611421577172, "standard_deviation": 0.023356486943269902, "minimum": 1.370486659926694, "maximum": 1.4376880422355065, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.47309251754068476, "standard_deviation": 0.22845433898461945, "minimum": 0.2091542971261253, "maximum": 0.8763199156573879, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09795068615924388, "standard_deviation": 0.005320640060389604, "minimum": 0.09056846515710198, "maximum": 0.10435520853460027, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980630524727121, "standard_deviation": 0.0004344860695483535, "minimum": 0.9975537968309738, "maximum": 0.998620487287675, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7107285020155747, "standard_deviation": 0.0294811497294116, "minimum": 0.6623833523245205, "maximum": 0.7369423400700486, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6243330928184575, "standard_deviation": 0.28655260328180937, "minimum": 0.2875252652566342, "maximum": 1.145295931049658, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05290432719362659, "standard_deviation": 0.006079804041931073, "minimum": 0.043955366935102304, "maximum": 0.06020214046785617, "valid_count": 8, "planned_count": 8}}}, {"case": "librispeech_908-31957-0016", "method": "E", "status": "measured", "mos": {"status": "measured", "value": 2.3308639526367188, "predicted_mos": 2.3308639526367188, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "92ad509b430ab31f86e348fc6d72d647ad7df2dd5e6203514eefb372afafee8d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 2.8865415998734534, "cpu_threads": 4}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3923912519064279, "全帯域_10ms_rms_envelope_nrmse": 0.09475447554012216, "全帯域_10ms_envelope_correlation": 0.9935447980400767, "全帯域_10ms_modulation_mae_db": 0.4630514147045619, "全帯域_40ms_log_rms_mae_db": 0.7222548642725874, "全帯域_40ms_rms_envelope_nrmse": 0.05319794459834126, "全帯域_40ms_envelope_correlation": 0.9979055918058868, "全帯域_40ms_modulation_mae_db": 0.7837524100313559, "300–1000 Hz_10ms_log_rms_mae_db": 1.8961153859852802, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1116265253593427, "300–1000 Hz_10ms_envelope_correlation": 0.9923094556384292, "300–1000 Hz_10ms_modulation_mae_db": 0.5303877867396192, "300–1000 Hz_40ms_log_rms_mae_db": 1.1501931699206065, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06111100446742565, "300–1000 Hz_40ms_envelope_correlation": 0.9976841849719627, "300–1000 Hz_40ms_modulation_mae_db": 0.5305529761933953, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9303002969146805, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20461041558196305, "1000–3000 Hz_10ms_envelope_correlation": 0.9710665900237486, "1000–3000 Hz_10ms_modulation_mae_db": 0.8258918122232637, "1000–3000 Hz_40ms_log_rms_mae_db": 1.137209737161202, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1184065678711874, "1000–3000 Hz_40ms_envelope_correlation": 0.9908708792528785, "1000–3000 Hz_40ms_modulation_mae_db": 0.4833600341272712, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6785219547170296, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16991091850458043, "3000–6000 Hz_10ms_envelope_correlation": 0.9832322888894682, "3000–6000 Hz_10ms_modulation_mae_db": 0.5648577004525405, "3000–6000 Hz_40ms_log_rms_mae_db": 1.297202434999711, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07822805980737609, "3000–6000 Hz_40ms_envelope_correlation": 0.9963986200901858, "3000–6000 Hz_40ms_modulation_mae_db": 0.5364462355129602, "mr_stft_db": 5.90872573735198, "band_mae_db": 0.2644059199205993, "rms_gain_error_db": 0.025841482826016348, "output_peak": 0.6677934236914191}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 1, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.395133852363035, "全帯域_10ms_rms_envelope_nrmse": 0.09489068246660044, "全帯域_10ms_envelope_correlation": 0.9934890306159178, "全帯域_10ms_modulation_mae_db": 0.294235232773905, "全帯域_40ms_log_rms_mae_db": 0.7548505835977745, "全帯域_40ms_rms_envelope_nrmse": 0.05582583369594926, "全帯域_40ms_envelope_correlation": 0.9977167911476476, "全帯域_40ms_modulation_mae_db": 0.3569996987936945, "300–1000 Hz_10ms_log_rms_mae_db": 2.0032372320571508, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.11617762555784344, "300–1000 Hz_10ms_envelope_correlation": 0.9917739060471418, "300–1000 Hz_10ms_modulation_mae_db": 0.711671238950007, "300–1000 Hz_40ms_log_rms_mae_db": 1.2841326311080883, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07469133622555982, "300–1000 Hz_40ms_envelope_correlation": 0.9966771805702674, "300–1000 Hz_40ms_modulation_mae_db": 0.897593608669025, "1000–3000 Hz_10ms_log_rms_mae_db": 1.872660392221911, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21402587739436546, "1000–3000 Hz_10ms_envelope_correlation": 0.9684971362033712, "1000–3000 Hz_10ms_modulation_mae_db": 1.2042248462100242, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1873577278732763, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12315548495104534, "1000–3000 Hz_40ms_envelope_correlation": 0.9904990119314417, "1000–3000 Hz_40ms_modulation_mae_db": 1.1169927626210485, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6940276199945643, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20759841653143257, "3000–6000 Hz_10ms_envelope_correlation": 0.9752915904021757, "3000–6000 Hz_10ms_modulation_mae_db": 1.167385583389944, "3000–6000 Hz_40ms_log_rms_mae_db": 1.258577187248423, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1028647436154831, "3000–6000 Hz_40ms_envelope_correlation": 0.9938146555320394, "3000–6000 Hz_40ms_modulation_mae_db": 1.4597798577039882, "mr_stft_db": 5.9011124134124175, "band_mae_db": 0.2600988293661901, "rms_gain_error_db": -0.08662274164354904, "output_peak": 0.5683885152309205}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 2, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.336206770587744, "全帯域_10ms_rms_envelope_nrmse": 0.09048288159367122, "全帯域_10ms_envelope_correlation": 0.9941867889270213, "全帯域_10ms_modulation_mae_db": 0.4552516413062353, "全帯域_40ms_log_rms_mae_db": 0.6383822188790179, "全帯域_40ms_rms_envelope_nrmse": 0.047122953799045526, "全帯域_40ms_envelope_correlation": 0.9985423003111756, "全帯域_40ms_modulation_mae_db": 0.44238304410793877, "300–1000 Hz_10ms_log_rms_mae_db": 1.9930037990913043, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.118556091384873, "300–1000 Hz_10ms_envelope_correlation": 0.9920983315077702, "300–1000 Hz_10ms_modulation_mae_db": 0.7192503455390495, "300–1000 Hz_40ms_log_rms_mae_db": 1.2288731204780274, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07310632933973046, "300–1000 Hz_40ms_envelope_correlation": 0.9976559332499805, "300–1000 Hz_40ms_modulation_mae_db": 1.0855167532704584, "1000–3000 Hz_10ms_log_rms_mae_db": 1.832742070967488, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2324299442363856, "1000–3000 Hz_10ms_envelope_correlation": 0.9627278390693048, "1000–3000 Hz_10ms_modulation_mae_db": 0.36540139115272147, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0846018844415943, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12411423396453193, "1000–3000 Hz_40ms_envelope_correlation": 0.9888399159039285, "1000–3000 Hz_40ms_modulation_mae_db": 0.46158573053864344, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6611422762543218, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18067918453819712, "3000–6000 Hz_10ms_envelope_correlation": 0.9841568244496359, "3000–6000 Hz_10ms_modulation_mae_db": 0.44192382681942105, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2470782782706682, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10527995477862037, "3000–6000 Hz_40ms_envelope_correlation": 0.9958170480653868, "3000–6000 Hz_40ms_modulation_mae_db": 0.6165223911050113, "mr_stft_db": 5.8940655016759775, "band_mae_db": 0.2980702504267188, "rms_gain_error_db": -0.1514864058348245, "output_peak": 0.5891060996657453}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 3, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.334347546685634, "全帯域_10ms_rms_envelope_nrmse": 0.09228144035982012, "全帯域_10ms_envelope_correlation": 0.9939969173465677, "全帯域_10ms_modulation_mae_db": 0.4801625273489634, "全帯域_40ms_log_rms_mae_db": 0.6987002582196749, "全帯域_40ms_rms_envelope_nrmse": 0.0579130692981603, "全帯域_40ms_envelope_correlation": 0.997743422124353, "全帯域_40ms_modulation_mae_db": 0.4443103574791925, "300–1000 Hz_10ms_log_rms_mae_db": 1.9171899321284862, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.123509341405114, "300–1000 Hz_10ms_envelope_correlation": 0.991509564689452, "300–1000 Hz_10ms_modulation_mae_db": 0.6509112334301623, "300–1000 Hz_40ms_log_rms_mae_db": 1.1785634420908022, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.08829586373538015, "300–1000 Hz_40ms_envelope_correlation": 0.9962502841000042, "300–1000 Hz_40ms_modulation_mae_db": 0.432844683192985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8276027282073655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2149009070448528, "1000–3000 Hz_10ms_envelope_correlation": 0.9676856623810265, "1000–3000 Hz_10ms_modulation_mae_db": 0.4153913434198772, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1293308541510085, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11725736363458969, "1000–3000 Hz_40ms_envelope_correlation": 0.9905500857129207, "1000–3000 Hz_40ms_modulation_mae_db": 0.18952346566997255, "3000–6000 Hz_10ms_log_rms_mae_db": 1.707494621116169, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17443915710322278, "3000–6000 Hz_10ms_envelope_correlation": 0.9855285220588164, "3000–6000 Hz_10ms_modulation_mae_db": 1.3552724896184347, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2769977955798917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1152679702858334, "3000–6000 Hz_40ms_envelope_correlation": 0.9947350165170138, "3000–6000 Hz_40ms_modulation_mae_db": 1.1422282566193447, "mr_stft_db": 5.887347074011923, "band_mae_db": 0.4065227714185082, "rms_gain_error_db": -0.17339189160032636, "output_peak": 0.5657334990768604}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 4, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3775525895601222, "全帯域_10ms_rms_envelope_nrmse": 0.10875863652434863, "全帯域_10ms_envelope_correlation": 0.9914116731753491, "全帯域_10ms_modulation_mae_db": 0.6603701932537555, "全帯域_40ms_log_rms_mae_db": 0.7314032396954544, "全帯域_40ms_rms_envelope_nrmse": 0.058134530994875575, "全帯域_40ms_envelope_correlation": 0.9974662247642778, "全帯域_40ms_modulation_mae_db": 0.8066771471744715, "300–1000 Hz_10ms_log_rms_mae_db": 2.007768796634292, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.13106423096219735, "300–1000 Hz_10ms_envelope_correlation": 0.9894164149111112, "300–1000 Hz_10ms_modulation_mae_db": 1.0299488453034378, "300–1000 Hz_40ms_log_rms_mae_db": 1.203032870409637, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07119581357190428, "300–1000 Hz_40ms_envelope_correlation": 0.996811910105082, "300–1000 Hz_40ms_modulation_mae_db": 0.9099203340390435, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8147434228142996, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1823624578774317, "1000–3000 Hz_10ms_envelope_correlation": 0.9776814207411466, "1000–3000 Hz_10ms_modulation_mae_db": 0.7237645806225702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0943058023516725, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11331113771718529, "1000–3000 Hz_40ms_envelope_correlation": 0.992218536637485, "1000–3000 Hz_40ms_modulation_mae_db": 0.9129798675829152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6856783501714752, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17203588134525707, "3000–6000 Hz_10ms_envelope_correlation": 0.9852109348128358, "3000–6000 Hz_10ms_modulation_mae_db": 1.3813198350783293, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2984139228467264, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11970401551080403, "3000–6000 Hz_40ms_envelope_correlation": 0.9947213823888387, "3000–6000 Hz_40ms_modulation_mae_db": 1.30372589912539, "mr_stft_db": 5.894764860695577, "band_mae_db": 0.4281486323886019, "rms_gain_error_db": -0.028611966320810436, "output_peak": 0.602515191088296}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 5, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4053762736468154, "全帯域_10ms_rms_envelope_nrmse": 0.1068290741676949, "全帯域_10ms_envelope_correlation": 0.9919022139248137, "全帯域_10ms_modulation_mae_db": 0.21327609919082036, "全帯域_40ms_log_rms_mae_db": 0.730018149863665, "全帯域_40ms_rms_envelope_nrmse": 0.05520315094833009, "全帯域_40ms_envelope_correlation": 0.9980513137044219, "全帯域_40ms_modulation_mae_db": 0.5643505525315166, "300–1000 Hz_10ms_log_rms_mae_db": 1.9596696811600247, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12898550340674664, "300–1000 Hz_10ms_envelope_correlation": 0.9900489625655304, "300–1000 Hz_10ms_modulation_mae_db": 0.3153965096934758, "300–1000 Hz_40ms_log_rms_mae_db": 1.2588051782598928, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.06808090485847113, "300–1000 Hz_40ms_envelope_correlation": 0.9976465920414913, "300–1000 Hz_40ms_modulation_mae_db": 0.3219506135627897, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9064207737017507, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2293742112282174, "1000–3000 Hz_10ms_envelope_correlation": 0.9633505859388012, "1000–3000 Hz_10ms_modulation_mae_db": 0.3652429478940591, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1955824219197628, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1360619702724785, "1000–3000 Hz_40ms_envelope_correlation": 0.9864880053171912, "1000–3000 Hz_40ms_modulation_mae_db": 0.2959737304141515, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7108134213280421, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.17164630958548263, "3000–6000 Hz_10ms_envelope_correlation": 0.9832843238406818, "3000–6000 Hz_10ms_modulation_mae_db": 0.8802288370602381, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3268655923690917, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09326897138479215, "3000–6000 Hz_40ms_envelope_correlation": 0.9950215862881874, "3000–6000 Hz_40ms_modulation_mae_db": 1.1173975761082824, "mr_stft_db": 5.896739248200753, "band_mae_db": 0.13069548367781073, "rms_gain_error_db": -0.19361118969492994, "output_peak": 0.5785223275913037}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 6, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4115462357751674, "全帯域_10ms_rms_envelope_nrmse": 0.10104107783066489, "全帯域_10ms_envelope_correlation": 0.992766403963859, "全帯域_10ms_modulation_mae_db": 0.15689941530796878, "全帯域_40ms_log_rms_mae_db": 0.7429949646531249, "全帯域_40ms_rms_envelope_nrmse": 0.05445470564569122, "全帯域_40ms_envelope_correlation": 0.9980656410499414, "全帯域_40ms_modulation_mae_db": 0.23109080218500685, "300–1000 Hz_10ms_log_rms_mae_db": 1.9968579301499054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.12437538547303291, "300–1000 Hz_10ms_envelope_correlation": 0.991356354175643, "300–1000 Hz_10ms_modulation_mae_db": 0.8859387567900804, "300–1000 Hz_40ms_log_rms_mae_db": 1.2307156576827918, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.07744627748089483, "300–1000 Hz_40ms_envelope_correlation": 0.9974154640640858, "300–1000 Hz_40ms_modulation_mae_db": 1.1134710516316908, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8871856484640939, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2214227321243095, "1000–3000 Hz_10ms_envelope_correlation": 0.9657249624003486, "1000–3000 Hz_10ms_modulation_mae_db": 0.5466277532005663, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1173799562581241, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1141907310937069, "1000–3000 Hz_40ms_envelope_correlation": 0.9913114094100682, "1000–3000 Hz_40ms_modulation_mae_db": 0.30607802924071964, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6982367604157407, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18610270349813096, "3000–6000 Hz_10ms_envelope_correlation": 0.9799030563242289, "3000–6000 Hz_10ms_modulation_mae_db": 1.5472879197574523, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2859706440364647, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08691065200731593, "3000–6000 Hz_40ms_envelope_correlation": 0.9955322732212666, "3000–6000 Hz_40ms_modulation_mae_db": 1.25719397052466, "mr_stft_db": 5.896500326180278, "band_mae_db": 0.26418802109159667, "rms_gain_error_db": -0.18403568831788036, "output_peak": 0.6570265481498008}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}, {"seed": 7, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3598256839096337, "全帯域_10ms_rms_envelope_nrmse": 0.08995362983066689, "全帯域_10ms_envelope_correlation": 0.9943528468239881, "全帯域_10ms_modulation_mae_db": 0.36249746437040037, "全帯域_40ms_log_rms_mae_db": 0.6567929380165706, "全帯域_40ms_rms_envelope_nrmse": 0.048501482040607155, "全帯域_40ms_envelope_correlation": 0.9985605884269578, "全帯域_40ms_modulation_mae_db": 0.5434273450594703, "300–1000 Hz_10ms_log_rms_mae_db": 1.9343203680738195, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1149571413142628, "300–1000 Hz_10ms_envelope_correlation": 0.9926688931920478, "300–1000 Hz_10ms_modulation_mae_db": 0.3630226464468073, "300–1000 Hz_40ms_log_rms_mae_db": 1.178092682989846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.0709155726894604, "300–1000 Hz_40ms_envelope_correlation": 0.9978971067449833, "300–1000 Hz_40ms_modulation_mae_db": 0.3691450989226149, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8077372634827489, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1967095726678059, "1000–3000 Hz_10ms_envelope_correlation": 0.9735489299489347, "1000–3000 Hz_10ms_modulation_mae_db": 0.6037789281706604, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1263568614573483, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11888898127952752, "1000–3000 Hz_40ms_envelope_correlation": 0.9898925854908249, "1000–3000 Hz_40ms_modulation_mae_db": 0.2470969663852313, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6781583289736535, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15906424666052735, "3000–6000 Hz_10ms_envelope_correlation": 0.9854005229424394, "3000–6000 Hz_10ms_modulation_mae_db": 0.49616119419065297, "3000–6000 Hz_40ms_log_rms_mae_db": 1.270270089257434, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08027086540708805, "3000–6000 Hz_40ms_envelope_correlation": 0.9962209022621406, "3000–6000 Hz_40ms_modulation_mae_db": 0.5012069114600756, "mr_stft_db": 5.890646171102898, "band_mae_db": 0.1888861491984566, "rms_gain_error_db": -0.18820780199472031, "output_peak": 0.5865404423342171}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.175334877839166, "spectral_convergence": 0.34793186936921106, "spectral_convergence_status": "ok", "amplitude_floor": 2.3135422827537554e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.727649144195115, "spectral_convergence": 0.30370006004736216, "spectral_convergence_status": "ok", "amplitude_floor": 1.7511633455315343e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.82319319002166, "spectral_convergence": 0.35773905196966926, "spectral_convergence_status": "ok", "amplitude_floor": 1.6157176909573571e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00044579673236332844, "output_power": 0.0005246674928743027, "delta_db": 0.7074727636947897, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00284687947566611, "output_power": 0.002831425346909622, "delta_db": -0.023639658745883703, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00047388905696308834, "output_power": 0.0004502240696687729, "delta_db": -0.22247970761681504, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013682287067652582, "output_power": 0.0001264608431541368, "delta_db": -0.3420262503906949, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011121195656671286, "output_power": 0.00010499778651315916, "delta_db": -0.24971337730329227, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00022124255174898303, "output_power": 0.00022334644028145517, "delta_db": 0.04110376177212026, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.2644059199205993, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.025841482826016348, "pitch": null, "output_rms_dbfs": -23.7047603345439, "output_peak": 0.6677934236914191, "reference_rms_dbfs": -23.730601817369916, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.90872573735198}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028051035499693775, "log_rms_mae_db": 1.3923912519064279, "rms_envelope_nrmse": 0.09475447554012216, "envelope_correlation": 0.9935447980400767, "rms_gain_error_db": 0.025846078571223507, "reference_mean_power": 0.004239368434074771, "output_mean_power": 0.0042646733241775096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.0558356273882855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018964431225846456, "output_power": 0.0019364453632104855, "reference_active": true, "delta_db": 0.0906542584395601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001936564415781366, "output_power": 0.00021874385845148514, "reference_active": true, "delta_db": 0.5290392095950394}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.246230567335058e-05, "output_power": 7.456999069970977e-05, "reference_active": true, "delta_db": 0.7694607760790864}], "active_band_count": 3, "active_band_mae_db": 0.4630514147045619}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026381677718961605, "log_rms_mae_db": 0.7222548642725874, "rms_envelope_nrmse": 0.05319794459834126, "envelope_correlation": 0.9979055918058868, "rms_gain_error_db": 0.02585169259096405, "reference_mean_power": 0.004260663575263385, "output_mean_power": 0.004286101117101475}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.853150694908015e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017547087963691286, "output_power": 0.0018102224058879715, "reference_active": true, "delta_db": 0.13526882828167897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.102196929549433e-05, "output_power": 9.751188203552686e-05, "reference_active": true, "delta_db": 0.80454743846676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0378089105429984e-06, "output_power": 4.204405039636068e-06, "reference_active": true, "delta_db": 1.4114409633456289}], "active_band_count": 3, "active_band_mae_db": 0.7837524100313559}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002736722962808348, "log_rms_mae_db": 1.8961153859852802, "rms_envelope_nrmse": 0.1116265253593427, "envelope_correlation": 0.9923094556384292, "rms_gain_error_db": -0.08239110677304028, "reference_mean_power": 0.0024581880191069405, "output_mean_power": 0.0024119926883921735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.077905340514435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019022535341174436, "output_power": 0.001864072441039073, "reference_active": true, "delta_db": -0.08805613859632413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012849688146439824, "output_power": 0.0001302164537987509, "reference_active": true, "delta_db": 0.0577327622399155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.095154000220668e-05, "output_power": 5.7122562851907795e-05, "reference_active": true, "delta_db": 1.4453744593826179}], "active_band_count": 3, "active_band_mae_db": 0.5303877867396192}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025339024948019116, "log_rms_mae_db": 1.1501931699206065, "rms_envelope_nrmse": 0.06111100446742565, "envelope_correlation": 0.9976841849719627, "rms_gain_error_db": -0.08239214524373983, "reference_mean_power": 0.002470549817529662, "output_mean_power": 0.002424121598912976}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.981046082661891e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017987268874748684, "output_power": 0.0017788791341185886, "reference_active": true, "delta_db": -0.04818785602972481}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5079379851283876e-05, "output_power": 5.9302139542543884e-05, "reference_active": true, "delta_db": 0.32081320603674385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4721075346991203e-06, "output_power": 3.2759189248418765e-06, "reference_active": true, "delta_db": 1.2226578665137173}], "active_band_count": 3, "active_band_mae_db": 0.5305529761933953}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.470771529403075e-05, "log_rms_mae_db": 1.9303002969146805, "rms_envelope_nrmse": 0.20461041558196305, "envelope_correlation": 0.9710665900237486, "rms_gain_error_db": -0.388650220458063, "reference_mean_power": 0.00019492290635367423, "output_mean_power": 0.00017823699689850444}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7570632979474236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.555963644612216e-05, "output_power": 6.299557897467494e-05, "reference_active": true, "delta_db": -0.789797878877947}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.293127801504832e-05, "output_power": 9.974239554915562e-06, "reference_active": true, "delta_db": -1.127616542757245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505165714035109e-06, "output_power": 7.4008988026445805e-06, "reference_active": true, "delta_db": 0.5602610150345989}], "active_band_count": 3, "active_band_mae_db": 0.8258918122232637}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.885515861839929e-05, "log_rms_mae_db": 1.137209737161202, "rms_envelope_nrmse": 0.1184065678711874, "envelope_correlation": 0.9908708792528785, "rms_gain_error_db": -0.3886523158461302, "reference_mean_power": 0.00019590175284412906, "output_mean_power": 0.00017913196514719764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6226758883463992e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.225076970752032e-05, "output_power": 5.994296446134071e-05, "reference_active": true, "delta_db": -0.8110426106882269}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.071619770568594e-06, "output_power": 5.346766994192811e-06, "reference_active": true, "delta_db": -0.5521330798804417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6400879277669946e-07, "output_power": 3.7136615018638624e-07, "reference_active": true, "delta_db": 0.08690441181314496}], "active_band_count": 3, "active_band_mae_db": 0.4833600341272712}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012408236865898127, "log_rms_mae_db": 1.6785219547170296, "rms_envelope_nrmse": 0.16991091850458043, "envelope_correlation": 0.9832322888894682, "rms_gain_error_db": -0.028984078416405677, "reference_mean_power": 0.00018911407706020183, "output_mean_power": 0.000187856163947407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8583468534278485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010082037892943086, "output_power": 0.00010372731832698924, "reference_active": true, "delta_db": 0.12344824796920341}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9440078487669303e-05, "output_power": 2.0503868712249373e-05, "reference_active": true, "delta_db": 0.23137798345231797}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.009136082695238e-06, "output_power": 6.81926420874124e-06, "reference_active": true, "delta_db": 1.3397468699361}], "active_band_count": 3, "active_band_mae_db": 0.5648577004525405}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.970743476594613e-05, "log_rms_mae_db": 1.297202434999711, "rms_envelope_nrmse": 0.07822805980737609, "envelope_correlation": 0.9963986200901858, "rms_gain_error_db": -0.028986555075893764, "reference_mean_power": 0.00019006483009233416, "output_mean_power": 0.00018880048527338287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7984096251410197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010012152579842322, "output_power": 0.00010414486496924233, "reference_active": true, "delta_db": 0.17110401818970664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0693666398121268e-05, "output_power": 1.1077816923569523e-05, "reference_active": true, "delta_db": 0.15327551912636272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.760055744312906e-07, "output_power": 3.710340530321318e-07, "reference_active": true, "delta_db": 1.2849591692228113}], "active_band_count": 3, "active_band_mae_db": 0.5364462355129602}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7726146233221414, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fafabf2c83187fe41b869d1e4ea8895719e404f0efc7519330a7cd0b34084f33", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9687853908383353, "standard_deviation": 0.005125904463498114, "minimum": 0.9627278390693048, "maximum": 0.9776814207411466, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8599240745967922, "standard_deviation": 0.045620246437670715, "minimum": 1.8077372634827489, "maximum": 1.9303002969146805, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6312904503617178, "standard_deviation": 0.2855296638706017, "minimum": 0.3652429478940591, "maximum": 1.2042248462100242, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21197951476941643, "standard_deviation": 0.016849488464778604, "minimum": 0.1823624578774317, "maximum": 0.2324299442363856, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9900838037070923, "standard_deviation": 0.0017560911330603162, "minimum": 0.9864880053171912, "maximum": 0.992218536637485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1340156557017487, "standard_deviation": 0.03968085136710436, "minimum": 1.0846018844415943, "maximum": 1.1955824219197628, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5016988233224942, "standard_deviation": 0.3364839136290854, "minimum": 0.18952346566997255, "maximum": 1.1169927626210485, "valid_count": 8, "planned_count": 8}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12067330884803157, "standard_deviation": 0.007280385229503336, "minimum": 0.11331113771718529, "maximum": 0.1360619702724785, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9827510079650352, "standard_deviation": 0.003526063574470527, "minimum": 0.9752915904021757, "maximum": 0.9855285220588164, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6892591666213745, "standard_deviation": 0.016657692863835585, "minimum": 1.6611422762543218, "maximum": 1.7108134213280421, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9793046732958766, "standard_deviation": 0.44154525676515, "minimum": 0.44192382681942105, "maximum": 1.5472879197574523, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17768460222085386, "standard_deviation": 0.01444555284104319, "minimum": 0.15906424666052735, "maximum": 0.20759841653143257, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9952826855456324, "standard_deviation": 0.0008710982695911192, "minimum": 0.9938146555320394, "maximum": 0.9963986200901858, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2826719930760513, "standard_deviation": 0.025231424140497886, "minimum": 1.2470782782706682, "maximum": 1.3268655923690917, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.991812637269964, "standard_deviation": 0.38060144070789537, "minimum": 0.5012069114600756, "maximum": 1.4597798577039882, "valid_count": 8, "planned_count": 8}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09772440409966414, "standard_deviation": 0.01556512143664059, "minimum": 0.07822805980737609, "maximum": 0.11970401551080403, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9913977353408907, "standard_deviation": 0.0011236707971034093, "minimum": 0.9894164149111112, "maximum": 0.9926688931920478, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.963520390660033, "standard_deviation": 0.043216113329944524, "minimum": 1.8961153859852802, "maximum": 2.007768796634292, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.65081592036158, "standard_deviation": 0.24414435347930447, "minimum": 0.3153965096934758, "maximum": 1.0299488453034378, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.12115648060792661, "standard_deviation": 0.006929824714002825, "minimum": 0.1116265253593427, "maximum": 0.13106423096219735, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9972548319809822, "standard_deviation": 0.0005947387287216482, "minimum": 0.9962502841000042, "maximum": 0.9978971067449833, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2140510941174616, "standard_deviation": 0.04498153038787537, "minimum": 1.1501931699206065, "maximum": 1.2841326311080883, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7076243899352503, "standard_deviation": 0.32835015481562085, "minimum": 0.3219506135627897, "maximum": 1.1134710516316908, "valid_count": 8, "planned_count": 8}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07310538779610334, "standard_deviation": 0.007834552954422045, "minimum": 0.06111100446742565, "maximum": 0.08829586373538015, "valid_count": 8, "planned_count": 8}, "band_mae_db": {"mean": 0.28012700718606026, "standard_deviation": 0.09983218553089293, "minimum": 0.13069548367781073, "maximum": 0.4281486323886019, "valid_count": 8, "planned_count": 8}, "mr_stft_db": {"mean": 5.896237666578976, "standard_deviation": 0.00651624955563827, "minimum": 5.887347074011923, "maximum": 5.90872573735198, "valid_count": 8, "planned_count": 8}, "output_peak": {"mean": 0.6019532558535703, "standard_deviation": 0.03920347330171535, "minimum": 0.5657334990768604, "maximum": 0.6677934236914191, "valid_count": 8, "planned_count": 8}, "rms_gain_error_db": {"mean": -0.12251577532262807, "standard_deviation": 0.08340819410319737, "minimum": -0.19361118969492994, "maximum": 0.025841482826016348, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_envelope_correlation": {"mean": 0.993206334102199, "standard_deviation": 0.0010830087661678992, "minimum": 0.9914116731753491, "maximum": 0.9943528468239881, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3765475255543225, "standard_deviation": 0.030098235620098647, "minimum": 1.334347546685634, "maximum": 1.4115462357751674, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_modulation_mae_db": {"mean": 0.38571799853207633, "standard_deviation": 0.16308254883334208, "minimum": 0.15689941530796878, "maximum": 0.6603701932537555, "valid_count": 8, "planned_count": 8}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.09737398728919866, "standard_deviation": 0.007309174083881105, "minimum": 0.08995362983066689, "maximum": 0.10875863652434863, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_envelope_correlation": {"mean": 0.9980064841668328, "standard_deviation": 0.00038818678677582206, "minimum": 0.9974662247642778, "maximum": 0.9985605884269578, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7094246521497336, "standard_deviation": 0.04173792933121016, "minimum": 0.6383822188790179, "maximum": 0.7548505835977745, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5216239196703308, "standard_deviation": 0.1985787976732194, "minimum": 0.23109080218500685, "maximum": 0.8066771471744715, "valid_count": 8, "planned_count": 8}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.05379420887762505, "standard_deviation": 0.004057205462815818, "minimum": 0.047122953799045526, "maximum": 0.058134530994875575, "valid_count": 8, "planned_count": 8}}}]}], "planned_recordings": 120, "failures": 0, "status": "completed", "schema_version": 3, "planned_cases": 120, "comparison_reference": "hnm", "method_labels": {"hnm": "実験時のHNM v2", "A": "A：元パワーで重ね合わせ", "B": "B：長い合成窓", "C": "C：短い合成窓", "D": "HNM v3（D：FFT端点の重み）", "E": "E：振幅乱数を固定", "world_harvest": "WORLD（Harvest）", "identity": "原音"}, "formal_base": {"status": "selected", "method": "hnm", "version": "hnm-v2", "reason": "固定120録音で音量変動の再現誤差が平均約7.2%減少。振幅超過の頻度・程度も踏まえて事後に採否を見直し、雑音包絡を分離する実装を採用。日本語12録音の補助評価も実施。予測MOSや知覚品質の改善を保証するものではない。", "comparison_reference": "hnm", "comparison_reference_meaning": "実験開始時点の唯一の自作ベース。過去結果は当時の版を保持する。既存比較対象はWORLD（Harvest）。", "previous_version": "hnm-v1", "selection_report": "experiments/2026-10-12/jvs-supplement/results/run-001/index.html"}, "index_methods": ["identity", "hnm", "D", "world_harvest"], "resynthesis": [{"case": "librispeech_1089-134686-0008", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0008", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0008", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0008", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0008", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0008", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0011", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0011", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0011", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0011", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0011", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0011", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0032", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0032", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0032", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0032", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0032", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0032", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0003", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0003", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0003", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0003", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0003", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0003", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0031", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0031", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0031", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0031", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0031", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0031", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0037", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0037", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0037", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0037", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0037", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1188-133604-0037", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-121726-0012", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-121726-0012", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-121726-0012", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-121726-0012", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-121726-0012", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-121726-0012", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-123852-0000", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-123852-0000", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-123852-0000", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-123852-0000", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-123852-0000", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-123852-0000", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-127105-0035", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-127105-0035", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-127105-0035", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-127105-0035", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-127105-0035", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_121-127105-0035", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135766-0005", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135766-0005", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135766-0005", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135766-0005", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135766-0005", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135766-0005", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0006", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0006", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0006", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0006", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0006", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0006", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0019", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0019", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0019", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0019", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0019", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1221-135767-0019", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0020", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0020", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0020", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0020", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0020", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0020", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0027", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0027", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0027", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0027", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0027", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0027", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0028", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0028", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0028", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0028", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0028", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1284-1180-0028", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0019", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0019", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0019", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0019", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0019", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0019", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0023", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0023", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0023", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0023", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0023", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0023", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0039", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0039", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0039", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0039", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0039", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1320-122617-0039", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141083-0010", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141083-0010", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141083-0010", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141083-0010", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141083-0010", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141083-0010", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0016", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0016", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0016", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0016", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0016", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0016", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0047", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0047", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0047", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0047", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0047", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1580-141084-0047", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0000", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0000", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0000", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0000", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0000", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0000", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0014", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0014", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0014", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0014", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0014", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1836-0014", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1837-0023", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1837-0023", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1837-0023", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1837-0023", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1837-0023", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1995-1837-0023", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0032", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0032", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0032", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0032", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0032", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0032", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0039", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0039", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0039", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0039", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0039", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0039", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0040", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0040", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0040", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0040", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0040", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2094-142345-0040", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0022", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0022", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0022", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0022", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0022", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0022", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0028", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0028", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0028", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0028", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0028", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0028", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0041", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0041", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0041", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0041", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0041", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2300-131720-0041", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-126133-0003", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-126133-0003", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-126133-0003", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-126133-0003", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-126133-0003", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-126133-0003", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0029", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0029", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0029", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0029", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0029", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0029", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0031", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0031", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0031", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0031", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0031", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_237-134500-0031", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123286-0008", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123286-0008", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123286-0008", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123286-0008", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123286-0008", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123286-0008", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123288-0003", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123288-0003", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123288-0003", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123288-0003", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123288-0003", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123288-0003", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123440-0020", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123440-0020", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123440-0020", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123440-0020", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123440-0020", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_260-123440-0020", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0000", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0000", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0000", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0000", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0000", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0000", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0038", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0038", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0038", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0038", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0038", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0038", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0042", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0042", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0042", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0042", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0042", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2830-3980-0042", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0010", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0010", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0010", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0010", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0010", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0010", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0019", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0019", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0019", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0019", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0019", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-960-0019", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-961-0011", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-961-0011", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-961-0011", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-961-0011", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-961-0011", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_2961-961-0011", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0001", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0001", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0001", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0001", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0001", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0001", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0004", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0004", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0004", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0004", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0004", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5694-0004", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5696-0010", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5696-0010", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5696-0010", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5696-0010", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5696-0010", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3570-5696-0010", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0034", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0034", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0034", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0034", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0034", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0034", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0041", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0041", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0041", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0041", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0041", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0041", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0054", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0054", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0054", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0054", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0054", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3575-170457-0054", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0000", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0000", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0000", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0000", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0000", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0000", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0007", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0007", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0007", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0007", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0007", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0007", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0017", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0017", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0017", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0017", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0017", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_3729-6852-0017", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13751-0001", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13751-0001", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13751-0001", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13751-0001", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13751-0001", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13751-0001", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0002", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0002", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0002", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0002", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0002", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0002", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0012", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0012", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0012", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0012", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0012", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4077-13754-0012", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0003", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0003", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0003", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0003", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0003", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0003", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0022", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0022", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0022", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0022", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0022", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2271-0022", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2273-0010", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2273-0010", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2273-0010", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2273-0010", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2273-0010", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4446-2273-0010", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0016", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0016", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0016", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0016", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0016", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0016", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0032", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0032", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0032", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0032", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0032", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0032", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0045", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0045", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0045", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0045", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0045", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4507-16021-0045", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0004", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0004", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0004", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0004", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0004", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0004", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0005", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0005", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0005", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0005", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0005", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29093-0005", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29095-0008", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29095-0008", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29095-0008", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29095-0008", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29095-0008", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4970-29095-0008", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-23283-0016", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-23283-0016", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-23283-0016", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-23283-0016", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-23283-0016", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-23283-0016", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41797-0000", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41797-0000", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41797-0000", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41797-0000", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41797-0000", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41797-0000", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41806-0000", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41806-0000", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41806-0000", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41806-0000", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41806-0000", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_4992-41806-0000", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28240-0011", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28240-0011", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28240-0011", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28240-0011", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28240-0011", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28240-0011", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0004", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0004", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0004", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0004", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0004", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0004", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0019", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0019", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0019", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0019", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0019", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5105-28241-0019", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0019", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0019", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0019", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0019", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0019", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0019", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0030", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0030", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0030", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0030", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0030", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0030", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0065", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0065", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0065", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0065", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0065", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5142-33396-0065", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0013", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0013", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0013", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0013", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0013", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0013", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0023", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0023", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0023", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0023", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0023", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0023", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0030", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0030", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0030", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0030", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0030", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5639-40744-0030", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32865-0017", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32865-0017", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32865-0017", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32865-0017", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32865-0017", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32865-0017", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0009", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0009", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0009", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0009", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0009", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0009", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0019", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0019", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0019", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0019", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0019", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_5683-32879-0019", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0042", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0042", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0042", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0042", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0042", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0042", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0043", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0043", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0043", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0043", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0043", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0043", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0054", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0054", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0054", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0054", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0054", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_61-70968-0054", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0003", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0003", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0003", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0003", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0003", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0003", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0044", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0044", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0044", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0044", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0044", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0044", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0056", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0056", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0056", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0056", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0056", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_672-122797-0056", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0014", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0014", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0014", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0014", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0014", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0014", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0038", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0038", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0038", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0038", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0038", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0038", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0041", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0041", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0041", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0041", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0041", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6829-68769-0041", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-75918-0013", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-75918-0013", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-75918-0013", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-75918-0013", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-75918-0013", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-75918-0013", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-76324-0024", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-76324-0024", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-76324-0024", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-76324-0024", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-76324-0024", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-76324-0024", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-81414-0026", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-81414-0026", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-81414-0026", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-81414-0026", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-81414-0026", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_6930-81414-0026", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0002", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0002", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0002", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0002", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0002", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0002", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0011", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0011", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0011", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0011", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0011", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-79740-0011", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-85628-0025", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-85628-0025", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-85628-0025", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-85628-0025", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-85628-0025", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7021-85628-0025", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0002", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0002", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0002", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0002", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0002", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0002", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0029", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0029", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0029", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0029", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0029", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75946-0029", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75947-0011", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75947-0011", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75947-0011", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75947-0011", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75947-0011", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7127-75947-0011", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0002", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0002", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0002", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0002", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0002", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0002", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0004", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0004", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0004", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0004", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0004", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0004", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0026", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0026", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0026", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0026", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0026", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7176-88083-0026", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0028", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0028", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0028", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0028", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0028", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0028", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0038", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0038", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0038", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0038", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0038", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0038", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0042", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0042", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0042", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0042", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0042", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_7729-102255-0042", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0010", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0010", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0010", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0010", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0010", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0010", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0013", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0013", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0013", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0013", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0013", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274381-0013", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274384-0010", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274384-0010", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274384-0010", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274384-0010", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274384-0010", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8224-274384-0010", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0018", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0018", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0018", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0018", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0018", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0018", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0032", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0032", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0032", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0032", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0032", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0032", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0033", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0033", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0033", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0033", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0033", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8230-279154-0033", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0010", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0010", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0010", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0010", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0010", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0010", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0049", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0049", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0049", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0049", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0049", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0049", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0060", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0060", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0060", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0060", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0060", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8455-210777-0060", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0003", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0003", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0003", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0003", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0003", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0003", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0014", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0014", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0014", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0014", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0014", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0014", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0022", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0022", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0022", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0022", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0022", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8463-294828-0022", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284447-0007", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284447-0007", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284447-0007", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284447-0007", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284447-0007", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284447-0007", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284449-0020", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284449-0020", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284449-0020", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284449-0020", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284449-0020", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-284449-0020", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-292519-0001", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-292519-0001", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-292519-0001", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-292519-0001", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-292519-0001", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_8555-292519-0001", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-157963-0018", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-157963-0018", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-157963-0018", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-157963-0018", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-157963-0018", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-157963-0018", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0002", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0002", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0002", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0002", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0002", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0002", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0016", "method": "hnm", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0016", "method": "A", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0016", "method": "B", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0016", "method": "C", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0016", "method": "D", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_908-31957-0016", "method": "E", "status": "measured", "detail_location": "parts"}, {"case": "librispeech_1089-134686-0008", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1089-134686-0011", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1089-134686-0032", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1188-133604-0003", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1188-133604-0031", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1188-133604-0037", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_121-121726-0012", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_121-123852-0000", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_121-127105-0035", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1221-135766-0005", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1221-135767-0006", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1221-135767-0019", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1284-1180-0020", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1284-1180-0027", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1284-1180-0028", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1320-122617-0019", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1320-122617-0023", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1320-122617-0039", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1580-141083-0010", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1580-141084-0016", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1580-141084-0047", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1995-1836-0000", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1995-1836-0014", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_1995-1837-0023", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2094-142345-0032", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2094-142345-0039", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2094-142345-0040", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2300-131720-0022", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2300-131720-0028", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2300-131720-0041", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_237-126133-0003", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_237-134500-0029", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_237-134500-0031", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_260-123286-0008", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_260-123288-0003", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_260-123440-0020", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2830-3980-0000", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2830-3980-0038", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2830-3980-0042", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2961-960-0010", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2961-960-0019", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_2961-961-0011", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3570-5694-0001", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3570-5694-0004", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3570-5696-0010", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3575-170457-0034", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3575-170457-0041", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3575-170457-0054", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3729-6852-0000", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3729-6852-0007", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_3729-6852-0017", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4077-13751-0001", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4077-13754-0002", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4077-13754-0012", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4446-2271-0003", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4446-2271-0022", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4446-2273-0010", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4507-16021-0016", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4507-16021-0032", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4507-16021-0045", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4970-29093-0004", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4970-29093-0005", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4970-29095-0008", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4992-23283-0016", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4992-41797-0000", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_4992-41806-0000", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5105-28240-0011", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5105-28241-0004", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5105-28241-0019", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5142-33396-0019", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5142-33396-0030", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5142-33396-0065", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5639-40744-0013", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5639-40744-0023", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5639-40744-0030", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5683-32865-0017", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5683-32879-0009", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_5683-32879-0019", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_61-70968-0042", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_61-70968-0043", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_61-70968-0054", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_672-122797-0003", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_672-122797-0044", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_672-122797-0056", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_6829-68769-0014", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_6829-68769-0038", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_6829-68769-0041", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_6930-75918-0013", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_6930-76324-0024", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_6930-81414-0026", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7021-79740-0002", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7021-79740-0011", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7021-85628-0025", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7127-75946-0002", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7127-75946-0029", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7127-75947-0011", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7176-88083-0002", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7176-88083-0004", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7176-88083-0026", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7729-102255-0028", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7729-102255-0038", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_7729-102255-0042", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8224-274381-0010", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8224-274381-0013", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8224-274384-0010", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8230-279154-0018", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8230-279154-0032", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8230-279154-0033", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8455-210777-0010", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8455-210777-0049", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8455-210777-0060", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8463-294828-0003", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8463-294828-0014", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8463-294828-0022", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8555-284447-0007", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8555-284449-0020", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_8555-292519-0001", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_908-157963-0018", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_908-31957-0002", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}, {"case": "librispeech_908-31957-0016", "method": "world_harvest", "status": "measured", "detail_location": "world_reference.rows"}], "world_reference": {"origin": "current_separate_execution", "source_sha256": "e89bdd8e40342c4b1ab7cc475c6579ba5510e655508c9dfeb227a2d09d4d33da", "rows": [{"case": "librispeech_1089-134686-0008", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.5213124752044678, "predicted_mos": 3.5213124752044678, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f96303c96d422dbe4a702f0c6625cea1e43b53058e27b0d06427b26ba1f24063", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 17.906363600399345}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.240989754664442, "全帯域_10ms_rms_envelope_nrmse": 0.18048308408464905, "全帯域_10ms_envelope_correlation": 0.9774595186767199, "全帯域_10ms_modulation_mae_db": 0.6483997725635912, "全帯域_40ms_log_rms_mae_db": 1.5046031967201743, "全帯域_40ms_rms_envelope_nrmse": 0.11917599811397661, "全帯域_40ms_envelope_correlation": 0.9930377346622904, "全帯域_40ms_modulation_mae_db": 0.6995550293295881, "300–1000 Hz_10ms_log_rms_mae_db": 2.497238045692429, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2128959489673064, "300–1000 Hz_10ms_envelope_correlation": 0.9784549512898939, "300–1000 Hz_10ms_modulation_mae_db": 1.0232331186343748, "300–1000 Hz_40ms_log_rms_mae_db": 1.935728968297973, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15985578522252253, "300–1000 Hz_40ms_envelope_correlation": 0.9913124569946296, "300–1000 Hz_40ms_modulation_mae_db": 1.0357417668112792, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7994694753815592, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18511664318842147, "1000–3000 Hz_10ms_envelope_correlation": 0.9766635613884972, "1000–3000 Hz_10ms_modulation_mae_db": 0.7854086335498517, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0114314013176475, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09230155124737914, "1000–3000 Hz_40ms_envelope_correlation": 0.9950606552066243, "1000–3000 Hz_40ms_modulation_mae_db": 0.8347061014250264, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1077184295008844, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28373988915198184, "3000–6000 Hz_10ms_envelope_correlation": 0.960181431807913, "3000–6000 Hz_10ms_modulation_mae_db": 0.8683233646557076, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2060140234302543, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19098721408951536, "3000–6000 Hz_40ms_envelope_correlation": 0.9850604939333478, "3000–6000 Hz_40ms_modulation_mae_db": 0.8622013052841236, "mr_stft_db": 6.07019779135516, "band_mae_db": 0.5614816299693691, "rms_gain_error_db": 0.6902483204273806, "output_peak": 0.5324006313712616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.465250015980011, "spectral_convergence": 0.7739551464958245, "spectral_convergence_status": "ok", "amplitude_floor": 1.4536470931091146e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.248909086440388, "spectral_convergence": 0.30087737319065355, "spectral_convergence_status": "ok", "amplitude_floor": 9.268471977763907e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.496434271645084, "spectral_convergence": 0.25578517967570497, "spectral_convergence_status": "ok", "amplitude_floor": 5.673364127394706e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031816153358243563, "output_power": 0.0003343662508675785, "delta_db": 0.21574764285832249, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009151577935124143, "output_power": 0.0011355866984161028, "delta_db": 0.9372431428839121, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00046151207587702967, "output_power": 0.0005242245661508033, "delta_db": 0.5533429973019023, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011338295265990407, "output_power": 0.00012647056465665913, "delta_db": 0.4744169509600294, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.877003374326731e-05, "output_power": 4.5079122763775016e-05, "delta_db": 0.654792771264559, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3868434770644032e-05, "output_power": 1.5680578307323923e-05, "delta_db": 0.5333462745474892, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5614816299693691, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6902483204273806, "pitch": null, "output_rms_dbfs": -26.61263142053238, "output_peak": 0.5324006313712616, "reference_rms_dbfs": -27.30287974095976, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.07019779135516}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016375851184643225, "log_rms_mae_db": 2.240989754664442, "rms_envelope_nrmse": 0.18048308408464905, "envelope_correlation": 0.9774595186767199, "rms_gain_error_db": 0.6903750223185207, "reference_mean_power": 0.0018623503698243853, "output_mean_power": 0.002183226990587197}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.48847287729811e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002826666045552883, "output_power": 0.00032850291687211273, "reference_active": true, "delta_db": 0.652647280309111}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.155794557866922e-05, "output_power": 6.32596272237281e-05, "reference_active": true, "delta_db": 0.11842511024311664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.629343961311588e-06, "output_power": 6.585141135521518e-06, "reference_active": true, "delta_db": -1.1741269271385457}], "active_band_count": 3, "active_band_mae_db": 0.6483997725635912}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014981717790874295, "log_rms_mae_db": 1.5046031967201743, "rms_envelope_nrmse": 0.11917599811397661, "envelope_correlation": 0.9930377346622904, "rms_gain_error_db": 0.6908583827929318, "reference_mean_power": 0.0018713077892111094, "output_mean_power": 0.0021939719135980486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.792361294396736e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025909451544035513, "output_power": 0.000300596826966303, "reference_active": true, "delta_db": 0.6452617209963074}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9672561350575063e-05, "output_power": 3.078773952988058e-05, "reference_active": true, "delta_db": 0.16022767233914825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.579324289571111e-07, "output_power": 5.627466518235074e-07, "reference_active": true, "delta_db": -1.2931756946533084}], "active_band_count": 3, "active_band_mae_db": 0.6995550293295881}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014318840727902982, "log_rms_mae_db": 2.497238045692429, "rms_envelope_nrmse": 0.2128959489673064, "envelope_correlation": 0.9784549512898939, "rms_gain_error_db": 0.9518818849977786, "reference_mean_power": 0.0010573416450716808, "output_mean_power": 0.0013164474464043279}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.91879361710777e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021964908416761717, "output_power": 0.00026662131041943563, "reference_active": true, "delta_db": 0.8416546180672675}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.090309412479679e-05, "output_power": 4.226228425141448e-05, "reference_active": true, "delta_db": 0.1419680514522213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.52813465615998e-06, "output_power": 2.800990685684993e-06, "reference_active": true, "delta_db": -2.086076686383636}], "active_band_count": 3, "active_band_mae_db": 1.0232331186343748}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013022613407710025, "log_rms_mae_db": 1.935728968297973, "rms_envelope_nrmse": 0.15985578522252253, "envelope_correlation": 0.9913124569946296, "rms_gain_error_db": 0.9518870474403398, "reference_mean_power": 0.001062644583935512, "output_mean_power": 0.0013230514643632145}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.561667844043168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020949277098297487, "output_power": 0.00025244096843282025, "reference_active": true, "delta_db": 0.8099079632114574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.887160269235181e-05, "output_power": 1.9446943576814674e-05, "reference_active": true, "delta_db": 0.13042569458829717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.293974094910667e-07, "output_power": 2.6071754366249306e-07, "reference_active": true, "delta_db": -2.166891642634083}], "active_band_count": 3, "active_band_mae_db": 1.0357417668112792}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.227046231131274e-05, "log_rms_mae_db": 1.7994694753815592, "rms_envelope_nrmse": 0.18511664318842147, "envelope_correlation": 0.9766635613884972, "rms_gain_error_db": 0.39308153486707864, "reference_mean_power": 0.00011346719055744512, "output_mean_power": 0.0001242162618136889}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5656809677774876e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2313251419233995e-05, "output_power": 2.4247684346266965e-05, "reference_active": true, "delta_db": 0.36107410829690256}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.900047334435587e-06, "output_power": 2.9268569947314023e-06, "reference_active": true, "delta_db": 0.03996416976020584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0009489505393011e-06, "output_power": 6.381064967202408e-07, "reference_active": true, "delta_db": -1.955187622592447}], "active_band_count": 3, "active_band_mae_db": 0.7854086335498517}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.589293372832859e-05, "log_rms_mae_db": 1.0114314013176475, "rms_envelope_nrmse": 0.09230155124737914, "envelope_correlation": 0.9950606552066243, "rms_gain_error_db": 0.3928795488933261, "reference_mean_power": 0.0001140327901646041, "output_mean_power": 0.00012482963645291887}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.160520396268688e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1030857564945278e-05, "output_power": 2.287390187741786e-05, "reference_active": true, "delta_db": 0.3648327171554816}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3790687257240336e-06, "output_power": 1.4328744384875372e-06, "reference_active": true, "delta_db": 0.16622225497145734}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.278096740186873e-08, "output_power": 4.620739557329924e-08, "reference_active": true, "delta_db": -1.9730633321481406}], "active_band_count": 3, "active_band_mae_db": 0.8347061014250264}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9795267482339013e-05, "log_rms_mae_db": 2.1077184295008844, "rms_envelope_nrmse": 0.28373988915198184, "envelope_correlation": 0.960181431807913, "rms_gain_error_db": 0.9038865620218096, "reference_mean_power": 2.1804807525571507e-05, "output_mean_power": 2.684979125736962e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7142973306691612e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0022096022650052e-05, "output_power": 1.2596408124498418e-05, "reference_active": true, "delta_db": 0.9928816392094707}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6747496282854573e-06, "output_power": 3.827378577514081e-06, "reference_active": true, "delta_db": 1.5561828588466096}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.037734768320834e-07, "output_power": 7.934930212816323e-07, "reference_active": true, "delta_db": -0.055905595911042286}], "active_band_count": 3, "active_band_mae_db": 0.8683233646557076}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.774754919111929e-05, "log_rms_mae_db": 1.2060140234302543, "rms_envelope_nrmse": 0.19098721408951536, "envelope_correlation": 0.9850604939333478, "rms_gain_error_db": 0.9038225874119116, "reference_mean_power": 2.1914321405186452e-05, "output_mean_power": 2.698424589079274e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.567366861761873e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0140844119742473e-05, "output_power": 1.3030445351672509e-05, "reference_active": true, "delta_db": 1.0888515218864996}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3436726086917478e-06, "output_power": 1.890143557114609e-06, "reference_active": true, "delta_db": 1.4820132626892892}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.915436539179574e-08, "output_power": 4.933282738292916e-08, "reference_active": true, "delta_db": 0.015739131276582358}], "active_band_count": 3, "active_band_mae_db": 0.8622013052841236}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9336871648363301, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d22e2b43a8758ececda020cd321ebabf2f52cbca588b9af876a6d1335cddec8a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9766635613884972, "standard_deviation": null, "minimum": 0.9766635613884972, "maximum": 0.9766635613884972, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7994694753815592, "standard_deviation": null, "minimum": 1.7994694753815592, "maximum": 1.7994694753815592, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7854086335498517, "standard_deviation": null, "minimum": 0.7854086335498517, "maximum": 0.7854086335498517, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18511664318842147, "standard_deviation": null, "minimum": 0.18511664318842147, "maximum": 0.18511664318842147, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950606552066243, "standard_deviation": null, "minimum": 0.9950606552066243, "maximum": 0.9950606552066243, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0114314013176475, "standard_deviation": null, "minimum": 1.0114314013176475, "maximum": 1.0114314013176475, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8347061014250264, "standard_deviation": null, "minimum": 0.8347061014250264, "maximum": 0.8347061014250264, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09230155124737914, "standard_deviation": null, "minimum": 0.09230155124737914, "maximum": 0.09230155124737914, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.960181431807913, "standard_deviation": null, "minimum": 0.960181431807913, "maximum": 0.960181431807913, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1077184295008844, "standard_deviation": null, "minimum": 2.1077184295008844, "maximum": 2.1077184295008844, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8683233646557076, "standard_deviation": null, "minimum": 0.8683233646557076, "maximum": 0.8683233646557076, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.28373988915198184, "standard_deviation": null, "minimum": 0.28373988915198184, "maximum": 0.28373988915198184, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9850604939333478, "standard_deviation": null, "minimum": 0.9850604939333478, "maximum": 0.9850604939333478, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2060140234302543, "standard_deviation": null, "minimum": 1.2060140234302543, "maximum": 1.2060140234302543, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8622013052841236, "standard_deviation": null, "minimum": 0.8622013052841236, "maximum": 0.8622013052841236, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19098721408951536, "standard_deviation": null, "minimum": 0.19098721408951536, "maximum": 0.19098721408951536, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9784549512898939, "standard_deviation": null, "minimum": 0.9784549512898939, "maximum": 0.9784549512898939, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.497238045692429, "standard_deviation": null, "minimum": 2.497238045692429, "maximum": 2.497238045692429, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0232331186343748, "standard_deviation": null, "minimum": 1.0232331186343748, "maximum": 1.0232331186343748, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2128959489673064, "standard_deviation": null, "minimum": 0.2128959489673064, "maximum": 0.2128959489673064, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9913124569946296, "standard_deviation": null, "minimum": 0.9913124569946296, "maximum": 0.9913124569946296, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.935728968297973, "standard_deviation": null, "minimum": 1.935728968297973, "maximum": 1.935728968297973, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0357417668112792, "standard_deviation": null, "minimum": 1.0357417668112792, "maximum": 1.0357417668112792, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15985578522252253, "standard_deviation": null, "minimum": 0.15985578522252253, "maximum": 0.15985578522252253, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5614816299693691, "standard_deviation": null, "minimum": 0.5614816299693691, "maximum": 0.5614816299693691, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.07019779135516, "standard_deviation": null, "minimum": 6.07019779135516, "maximum": 6.07019779135516, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5324006313712616, "standard_deviation": null, "minimum": 0.5324006313712616, "maximum": 0.5324006313712616, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6902483204273806, "standard_deviation": null, "minimum": 0.6902483204273806, "maximum": 0.6902483204273806, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9774595186767199, "standard_deviation": null, "minimum": 0.9774595186767199, "maximum": 0.9774595186767199, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.240989754664442, "standard_deviation": null, "minimum": 2.240989754664442, "maximum": 2.240989754664442, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6483997725635912, "standard_deviation": null, "minimum": 0.6483997725635912, "maximum": 0.6483997725635912, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18048308408464905, "standard_deviation": null, "minimum": 0.18048308408464905, "maximum": 0.18048308408464905, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9930377346622904, "standard_deviation": null, "minimum": 0.9930377346622904, "maximum": 0.9930377346622904, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.5046031967201743, "standard_deviation": null, "minimum": 1.5046031967201743, "maximum": 1.5046031967201743, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6995550293295881, "standard_deviation": null, "minimum": 0.6995550293295881, "maximum": 0.6995550293295881, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11917599811397661, "standard_deviation": null, "minimum": 0.11917599811397661, "maximum": 0.11917599811397661, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1089"}, {"case": "librispeech_1089-134686-0011", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.287036418914795, "predicted_mos": 3.287036418914795, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "30f94914bb8cefbdb6bc7a506f02b9becb52353da5c5ffa40e9e3aaac8744b5a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.061068599577993}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.0587356125330203, "全帯域_10ms_rms_envelope_nrmse": 0.15002587190244449, "全帯域_10ms_envelope_correlation": 0.986182042786642, "全帯域_10ms_modulation_mae_db": 0.8124086018834705, "全帯域_40ms_log_rms_mae_db": 1.5371463198857673, "全帯域_40ms_rms_envelope_nrmse": 0.10864340784581805, "全帯域_40ms_envelope_correlation": 0.9952701334879654, "全帯域_40ms_modulation_mae_db": 0.732821922194117, "300–1000 Hz_10ms_log_rms_mae_db": 2.3906976316564936, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1839453298917479, "300–1000 Hz_10ms_envelope_correlation": 0.984815673499722, "300–1000 Hz_10ms_modulation_mae_db": 1.318765266243872, "300–1000 Hz_40ms_log_rms_mae_db": 1.8186237463980828, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1444328051127505, "300–1000 Hz_40ms_envelope_correlation": 0.9931486943246239, "300–1000 Hz_40ms_modulation_mae_db": 1.1790968646114857, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8395673674099806, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.156245047296677, "1000–3000 Hz_10ms_envelope_correlation": 0.9859590293816912, "1000–3000 Hz_10ms_modulation_mae_db": 0.7756870962314921, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1010576057293981, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09651466652272403, "1000–3000 Hz_40ms_envelope_correlation": 0.9960447402861785, "1000–3000 Hz_40ms_modulation_mae_db": 1.0866487847498199, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1587871350191956, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20222650911388823, "3000–6000 Hz_10ms_envelope_correlation": 0.9712715306565866, "3000–6000 Hz_10ms_modulation_mae_db": 0.2916203809764502, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2801312279775214, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12586206443611095, "3000–6000 Hz_40ms_envelope_correlation": 0.9881186041343313, "3000–6000 Hz_40ms_modulation_mae_db": 0.46505470809337573, "mr_stft_db": 6.1343072313220475, "band_mae_db": 0.47683921306311056, "rms_gain_error_db": 0.6872053374207997, "output_peak": 0.6085349550346482}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.589808601698983, "spectral_convergence": 0.7311184870419255, "spectral_convergence_status": "ok", "amplitude_floor": 1.4412887028114478e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.262162617634965, "spectral_convergence": 0.3014193340820907, "spectral_convergence_status": "ok", "amplitude_floor": 8.082487629299873e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.550950474632194, "spectral_convergence": 0.27043867589121856, "spectral_convergence_status": "ok", "amplitude_floor": 6.705293681874596e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00036448625115270087, "output_power": 0.000377214089560555, "delta_db": 0.1490675510679002, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008505212390156939, "output_power": 0.0010189034002856335, "delta_db": 0.7844784838781209, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010532918945599765, "output_power": 0.001278512551611145, "delta_db": 0.8415625385238056, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002198202138287696, "output_power": 0.00024722921303351386, "delta_db": 0.510321604511674, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.872413022951283e-05, "output_power": 4.302951330303617e-05, "delta_db": 0.457847619157024, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.441903264074617e-05, "output_power": 2.3765814733815664e-05, "delta_db": -0.11775748124013896, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.47683921306311056, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6872053374207997, "pitch": null, "output_rms_dbfs": -25.245242761854776, "output_peak": 0.6085349550346482, "reference_rms_dbfs": -25.93244809927558, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.1343072313220475}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019142337741905198, "log_rms_mae_db": 2.0587356125330203, "rms_envelope_nrmse": 0.15002587190244449, "envelope_correlation": 0.986182042786642, "rms_gain_error_db": 0.6873150482753689, "reference_mean_power": 0.0025533227171819564, "output_mean_power": 0.002991143260669579}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1748402838347296e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007147512141697323, "output_power": 0.0008228861198735981, "reference_active": true, "delta_db": 0.6118483509381758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.175142428480134e-05, "output_power": 8.934822065406913e-05, "reference_active": true, "delta_db": 0.3859058033154005}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.431203959050675e-05, "output_power": 1.0274350182658826e-05, "reference_active": true, "delta_db": -1.4394716513968353}], "active_band_count": 3, "active_band_mae_db": 0.8124086018834705}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016640454830055276, "log_rms_mae_db": 1.5371463198857673, "rms_envelope_nrmse": 0.10864340784581805, "envelope_correlation": 0.9952701334879654, "rms_gain_error_db": 0.6878713845494954, "reference_mean_power": 0.002564981061322552, "output_mean_power": 0.0030051856146790915}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0626539543337418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006440432662381922, "output_power": 0.0007419053769754375, "reference_active": true, "delta_db": 0.6143347480477033}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1296244809426825e-05, "output_power": 4.6630808577698875e-05, "reference_active": true, "delta_db": 0.5276238442051864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.585289215752233e-07, "output_power": 6.731381619651771e-07, "reference_active": true, "delta_db": -1.0565071743294614}], "active_band_count": 3, "active_band_mae_db": 0.732821922194117}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016776043588712122, "log_rms_mae_db": 2.3906976316564936, "rms_envelope_nrmse": 0.1839453298917479, "envelope_correlation": 0.984815673499722, "rms_gain_error_db": 0.8845734782211152, "reference_mean_power": 0.0015150798695545982, "output_mean_power": 0.0018573462603462764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.503421266374213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005710315291680691, "output_power": 0.0006731824768132053, "reference_active": true, "delta_db": 0.7147271438548496}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462070521392158e-05, "output_power": 7.179265783906316e-05, "reference_active": true, "delta_db": 0.4570833844730396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.82379092220564e-06, "output_power": 5.1740495544105915e-06, "reference_active": true, "delta_db": -2.7844852704037266}], "active_band_count": 3, "active_band_mae_db": 1.318765266243872}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015135634256581334, "log_rms_mae_db": 1.8186237463980828, "rms_envelope_nrmse": 0.1444328051127505, "envelope_correlation": 0.9931486943246239, "rms_gain_error_db": 0.8845788189254868, "reference_mean_power": 0.0015225180811750088, "output_mean_power": 0.0018664671075940456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.758019890177841e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005253930575206644, "output_power": 0.0006173753632245614, "reference_active": true, "delta_db": 0.7006496490702553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0513708673138778e-05, "output_power": 3.3933783575697114e-05, "reference_active": true, "delta_db": 0.46137290124518204}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.334657094067365e-07, "output_power": 4.823485000660172e-07, "reference_active": true, "delta_db": -2.375268043519019}], "active_band_count": 3, "active_band_mae_db": 1.1790968646114857}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.526806573516121e-05, "log_rms_mae_db": 1.8395673674099806, "rms_envelope_nrmse": 0.156245047296677, "envelope_correlation": 0.9859590293816912, "rms_gain_error_db": 0.5246239847703599, "reference_mean_power": 0.00023682053730462446, "output_mean_power": 0.00026722787649150566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1631666261710027e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014015200472863968, "output_power": 0.00015693967203235442, "reference_active": true, "delta_db": 0.49133426427519156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3136457414027934e-05, "output_power": 1.3286157280446572e-05, "reference_active": true, "delta_db": 0.04921127020794825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.230915915111415e-06, "output_power": 2.1412830547887095e-06, "reference_active": true, "delta_db": -1.7865157542113366}], "active_band_count": 3, "active_band_mae_db": 0.7756870962314921}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.301677625713533e-05, "log_rms_mae_db": 1.1010576057293981, "rms_envelope_nrmse": 0.09651466652272403, "envelope_correlation": 0.9960447402861785, "rms_gain_error_db": 0.5247581174215947, "reference_mean_power": 0.0002379882918616051, "output_mean_power": 0.0002685538628808998}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0209734833253925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001318054605596598, "output_power": 0.00014737062887586658, "reference_active": true, "delta_db": 0.48477533803994727}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.132933188997116e-06, "output_power": 6.539724663190275e-06, "reference_active": true, "delta_db": 0.27891230362985164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2319303983633163e-07, "output_power": 1.2561883921142684e-07, "reference_active": true, "delta_db": -2.4962587125796603}], "active_band_count": 3, "active_band_mae_db": 1.0866487847498199}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.803319177623983e-05, "log_rms_mae_db": 2.1587871350191956, "rms_envelope_nrmse": 0.20222650911388823, "envelope_correlation": 0.9712715306565866, "rms_gain_error_db": 0.03534545999238115, "reference_mean_power": 3.0517013147816785e-05, "output_mean_power": 3.07663921156663e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.826326538605559e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1311662288856248e-05, "output_power": 1.2149864511860464e-05, "reference_active": true, "delta_db": 0.31045004236505913}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5475498743040963e-06, "output_power": 2.5994506900139014e-06, "reference_active": true, "delta_db": 0.08758888446003835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.732573657220471e-07, "output_power": 5.136501532574659e-07, "reference_active": true, "delta_db": -0.4768222161042531}], "active_band_count": 3, "active_band_mae_db": 0.2916203809764502}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.459177046887288e-05, "log_rms_mae_db": 1.2801312279775214, "rms_envelope_nrmse": 0.12586206443611095, "envelope_correlation": 0.9881186041343313, "rms_gain_error_db": 0.03542971423378162, "reference_mean_power": 3.0668913834731335e-05, "output_mean_power": 3.092013395750883e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6363060360497894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1094596811060155e-05, "output_power": 1.1947047964241756e-05, "reference_active": true, "delta_db": 0.3214908312275733}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1901156553217825e-06, "output_power": 1.214727500010828e-06, "reference_active": true, "delta_db": 0.08889695380612408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.41782955712918e-08, "output_power": 3.5215294327850166e-08, "reference_active": true, "delta_db": -0.9847763392464297}], "active_band_count": 3, "active_band_mae_db": 0.46505470809337573}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9081680881436462, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2607e7d5a4f121f7a3b693bd64c733026000833b62b4347604f0f5aa20137c85", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9859590293816912, "standard_deviation": null, "minimum": 0.9859590293816912, "maximum": 0.9859590293816912, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8395673674099806, "standard_deviation": null, "minimum": 1.8395673674099806, "maximum": 1.8395673674099806, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7756870962314921, "standard_deviation": null, "minimum": 0.7756870962314921, "maximum": 0.7756870962314921, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.156245047296677, "standard_deviation": null, "minimum": 0.156245047296677, "maximum": 0.156245047296677, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960447402861785, "standard_deviation": null, "minimum": 0.9960447402861785, "maximum": 0.9960447402861785, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1010576057293981, "standard_deviation": null, "minimum": 1.1010576057293981, "maximum": 1.1010576057293981, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0866487847498199, "standard_deviation": null, "minimum": 1.0866487847498199, "maximum": 1.0866487847498199, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09651466652272403, "standard_deviation": null, "minimum": 0.09651466652272403, "maximum": 0.09651466652272403, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9712715306565866, "standard_deviation": null, "minimum": 0.9712715306565866, "maximum": 0.9712715306565866, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1587871350191956, "standard_deviation": null, "minimum": 2.1587871350191956, "maximum": 2.1587871350191956, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.2916203809764502, "standard_deviation": null, "minimum": 0.2916203809764502, "maximum": 0.2916203809764502, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20222650911388823, "standard_deviation": null, "minimum": 0.20222650911388823, "maximum": 0.20222650911388823, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9881186041343313, "standard_deviation": null, "minimum": 0.9881186041343313, "maximum": 0.9881186041343313, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2801312279775214, "standard_deviation": null, "minimum": 1.2801312279775214, "maximum": 1.2801312279775214, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.46505470809337573, "standard_deviation": null, "minimum": 0.46505470809337573, "maximum": 0.46505470809337573, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12586206443611095, "standard_deviation": null, "minimum": 0.12586206443611095, "maximum": 0.12586206443611095, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.984815673499722, "standard_deviation": null, "minimum": 0.984815673499722, "maximum": 0.984815673499722, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3906976316564936, "standard_deviation": null, "minimum": 2.3906976316564936, "maximum": 2.3906976316564936, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.318765266243872, "standard_deviation": null, "minimum": 1.318765266243872, "maximum": 1.318765266243872, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1839453298917479, "standard_deviation": null, "minimum": 0.1839453298917479, "maximum": 0.1839453298917479, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9931486943246239, "standard_deviation": null, "minimum": 0.9931486943246239, "maximum": 0.9931486943246239, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.8186237463980828, "standard_deviation": null, "minimum": 1.8186237463980828, "maximum": 1.8186237463980828, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1790968646114857, "standard_deviation": null, "minimum": 1.1790968646114857, "maximum": 1.1790968646114857, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1444328051127505, "standard_deviation": null, "minimum": 0.1444328051127505, "maximum": 0.1444328051127505, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.47683921306311056, "standard_deviation": null, "minimum": 0.47683921306311056, "maximum": 0.47683921306311056, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.1343072313220475, "standard_deviation": null, "minimum": 6.1343072313220475, "maximum": 6.1343072313220475, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6085349550346482, "standard_deviation": null, "minimum": 0.6085349550346482, "maximum": 0.6085349550346482, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6872053374207997, "standard_deviation": null, "minimum": 0.6872053374207997, "maximum": 0.6872053374207997, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.986182042786642, "standard_deviation": null, "minimum": 0.986182042786642, "maximum": 0.986182042786642, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.0587356125330203, "standard_deviation": null, "minimum": 2.0587356125330203, "maximum": 2.0587356125330203, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8124086018834705, "standard_deviation": null, "minimum": 0.8124086018834705, "maximum": 0.8124086018834705, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15002587190244449, "standard_deviation": null, "minimum": 0.15002587190244449, "maximum": 0.15002587190244449, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9952701334879654, "standard_deviation": null, "minimum": 0.9952701334879654, "maximum": 0.9952701334879654, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.5371463198857673, "standard_deviation": null, "minimum": 1.5371463198857673, "maximum": 1.5371463198857673, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.732821922194117, "standard_deviation": null, "minimum": 0.732821922194117, "maximum": 0.732821922194117, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10864340784581805, "standard_deviation": null, "minimum": 0.10864340784581805, "maximum": 0.10864340784581805, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1089"}, {"case": "librispeech_1089-134686-0032", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3701257705688477, "predicted_mos": 3.3701257705688477, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65440, "input_sample_rate": 16000, "input_seconds": 4.09, "input_float32_sha256": "afbfb0bcc8e90d0533a29c97ad02e11a4844b098db7702fa204445cc70002ec2", "internal_samples": 65440, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.5537642003037035}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.056133323721495, "全帯域_10ms_rms_envelope_nrmse": 0.1816856307240048, "全帯域_10ms_envelope_correlation": 0.9791827019659012, "全帯域_10ms_modulation_mae_db": 1.0212226484556581, "全帯域_40ms_log_rms_mae_db": 2.501003503209866, "全帯域_40ms_rms_envelope_nrmse": 0.12667144720687293, "全帯域_40ms_envelope_correlation": 0.992516602995846, "全帯域_40ms_modulation_mae_db": 1.5566277291007662, "300–1000 Hz_10ms_log_rms_mae_db": 2.320842162638393, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19223489447033232, "300–1000 Hz_10ms_envelope_correlation": 0.9826668637344909, "300–1000 Hz_10ms_modulation_mae_db": 0.8222495568581488, "300–1000 Hz_40ms_log_rms_mae_db": 2.14079858724072, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1359596258554474, "300–1000 Hz_40ms_envelope_correlation": 0.9945816895968931, "300–1000 Hz_40ms_modulation_mae_db": 1.1713702057566582, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6130831469466924, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19560405984208448, "1000–3000 Hz_10ms_envelope_correlation": 0.9745305242611019, "1000–3000 Hz_10ms_modulation_mae_db": 0.4634038109578542, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8888607826727349, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10229024011931291, "1000–3000 Hz_40ms_envelope_correlation": 0.9946443261723266, "1000–3000 Hz_40ms_modulation_mae_db": 0.613413230375246, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7372343283761233, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2525527499194943, "3000–6000 Hz_10ms_envelope_correlation": 0.9574983618974391, "3000–6000 Hz_10ms_modulation_mae_db": 0.4959678282940101, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8697521851568458, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12662609573415468, "3000–6000 Hz_40ms_envelope_correlation": 0.9893038921940567, "3000–6000 Hz_40ms_modulation_mae_db": 0.2553195312978283, "mr_stft_db": 6.126034806565193, "band_mae_db": 0.502420970661778, "rms_gain_error_db": 0.7127098829923681, "output_peak": 0.44440785869587485}, "full_scale": {"sample_count": 65440, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65440, "output_samples": 65440, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.53930246140317, "spectral_convergence": 0.794550454456133, "spectral_convergence_status": "ok", "amplitude_floor": 1.239800719019687e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2251067612260655, "spectral_convergence": 0.3047952542388241, "spectral_convergence_status": "ok", "amplitude_floor": 7.959069981155994e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.613695197066342, "spectral_convergence": 0.2553169787623593, "spectral_convergence_status": "ok", "amplitude_floor": 5.70263552352201e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002917609391844852, "output_power": 0.00030784549288885663, "delta_db": 0.2330565123485445, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009442146735758444, "output_power": 0.0011418791956395741, "delta_db": 0.825494151356304, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008095018592590047, "output_power": 0.0009672568218154273, "delta_db": 0.7732395079669723, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.140726670679603e-05, "output_power": 8.147021530095506e-05, "delta_db": 0.572564543437014, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.8558923829027427e-05, "output_power": 3.0474540968339494e-05, "delta_db": 0.2819533453148589, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.629692823024492e-06, "output_power": 9.307158575232822e-06, "delta_db": 0.3282177635469743, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.502420970661778, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7127098829923681, "pitch": null, "output_rms_dbfs": -25.954684366074538, "output_peak": 0.44440785869587485, "reference_rms_dbfs": -26.667394249066906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.126034806565193}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65440, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00016079151656388168, "log_rms_mae_db": 3.056133323721495, "rms_envelope_nrmse": 0.1816856307240048, "envelope_correlation": 0.9791827019659012, "rms_gain_error_db": 0.7127114514170394, "reference_mean_power": 0.0021567095381351967, "output_mean_power": 0.0025413401870800737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.6620838184249172e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000592954618765589, "output_power": 0.00071469539376356, "reference_active": true, "delta_db": 0.8109952675844601}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014246995030046006, "output_power": 0.00012670918979877392, "reference_active": true, "delta_db": -0.5091515883225046}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.93963265073239e-05, "output_power": 1.2982770416856933e-05, "reference_active": true, "delta_db": -1.7435210894600095}], "active_band_count": 3, "active_band_mae_db": 1.0212226484556581}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00012458719833187296, "log_rms_mae_db": 2.501003503209866, "rms_envelope_nrmse": 0.12667144720687293, "envelope_correlation": 0.992516602995846, "rms_gain_error_db": 0.7127177751419482, "reference_mean_power": 0.0021726617564687806, "output_mean_power": 0.0025601410745349205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.5403145620848259e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005607371759141368, "output_power": 0.0006762758789334701, "reference_active": true, "delta_db": 0.8136454747218457}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.842815402518841e-05, "output_power": 5.692616023010044e-05, "reference_active": true, "delta_db": -0.7992293339432008}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9268319788025257e-06, "output_power": 9.531099593177746e-07, "reference_active": true, "delta_db": -3.0570083786372515}], "active_band_count": 3, "active_band_mae_db": 1.5566277291007662}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 0.00013740128966162354, "log_rms_mae_db": 2.320842162638393, "rms_envelope_nrmse": 0.19223489447033232, "envelope_correlation": 0.9826668637344909, "rms_gain_error_db": 0.8517245844911014, "reference_mean_power": 0.001344215530062727, "output_mean_power": 0.0016354654243191952}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 1.2437342579426783e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004237422220408671, "output_power": 0.0005225988476957588, "reference_active": true, "delta_db": 0.9106670816063102}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010753440950326828, "output_power": 0.00011050716725993685, "reference_active": true, "delta_db": 0.11842991716289732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.157936749643158e-05, "output_power": 8.316099372410943e-06, "reference_active": true, "delta_db": -1.437651671805239}], "active_band_count": 3, "active_band_mae_db": 0.8222495568581488}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 0.00011336523032200158, "log_rms_mae_db": 2.14079858724072, "rms_envelope_nrmse": 0.1359596258554474, "envelope_correlation": 0.9945816895968931, "rms_gain_error_db": 0.8517243182947447, "reference_mean_power": 0.0013541602801659725, "output_mean_power": 0.0016475647929374948}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 1.1658618871138855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004110855381392752, "output_power": 0.0004985945724240609, "reference_active": true, "delta_db": 0.8381534786729096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.54169136496196e-05, "output_power": 4.7678513881718665e-05, "reference_active": true, "delta_db": 0.2110509219132487}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3416624074278087e-06, "output_power": 7.605935349890365e-07, "reference_active": true, "delta_db": -2.464906216683816}], "active_band_count": 3, "active_band_mae_db": 1.1713702057566582}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 4.166087007180048e-05, "log_rms_mae_db": 1.6130831469466924, "rms_envelope_nrmse": 0.19560405984208448, "envelope_correlation": 0.9745305242611019, "rms_gain_error_db": 0.5108795171072404, "reference_mean_power": 7.742789466080943e-05, "output_mean_power": 8.709343151281063e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.015275941663665e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8994572701205476e-05, "output_power": 4.3586057248465475e-05, "reference_active": true, "delta_db": 0.48343419200083954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.4522009847535795e-06, "output_power": 8.055140273661915e-06, "reference_active": true, "delta_db": 0.33788548463509493}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.182412963771591e-06, "output_power": 1.037241792402274e-06, "reference_active": true, "delta_db": -0.5688917562376282}], "active_band_count": 3, "active_band_mae_db": 0.4634038109578542}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 3.308768412789438e-05, "log_rms_mae_db": 0.8888607826727349, "rms_envelope_nrmse": 0.10229024011931291, "envelope_correlation": 0.9946443261723266, "rms_gain_error_db": 0.5108799729798199, "reference_mean_power": 7.800066882867515e-05, "output_mean_power": 8.773771586365731e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 6.505338847209246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.845568227980419e-05, "output_power": 4.268667522499251e-05, "reference_active": true, "delta_db": 0.4533180883919438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.058614755593344e-06, "output_power": 3.498448656913553e-06, "reference_active": true, "delta_db": 0.5835072525453275}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905753913870012e-08, "output_power": 8.232772218994886e-08, "reference_active": true, "delta_db": -0.8034143501884665}], "active_band_count": 3, "active_band_mae_db": 0.613413230375246}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 817, "rms_floor": 1.92244611767394e-05, "log_rms_mae_db": 1.7372343283761233, "rms_envelope_nrmse": 0.2525527499194943, "envelope_correlation": 0.9574983618974391, "rms_gain_error_db": 0.30492558504791284, "reference_mean_power": 1.4177032937223143e-05, "output_mean_power": 1.5208203209024453e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 817, "frequency_resolution_hz": 0.24479804161566707, "power_floor": 7.931109844663442e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.831691262181727e-06, "output_power": 4.204168486252664e-06, "reference_active": true, "delta_db": 0.4028960387199313}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0928531053800205e-06, "output_power": 1.0802405820817104e-06, "reference_active": true, "delta_db": -0.050413020013059036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4875100105629565e-07, "output_power": 1.8876407303732168e-07, "reference_active": true, "delta_db": 1.0345944261490398}], "active_band_count": 3, "active_band_mae_db": 0.4959678282940101}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65440, "uncovered_tail_samples": 0, "metrics": {"frame_count": 811, "rms_floor": 1.4099069206239137e-05, "log_rms_mae_db": 0.8697521851568458, "rms_envelope_nrmse": 0.12662609573415468, "envelope_correlation": 0.9893038921940567, "rms_gain_error_db": 0.30492651034793655, "reference_mean_power": 1.4281884811465133e-05, "output_mean_power": 1.5320684776501388e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 811, "frequency_resolution_hz": 0.24660912453760792, "power_floor": 7.397985811629507e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.714499411475195e-06, "output_power": 4.0955125843046124e-06, "reference_active": true, "delta_db": 0.42407970854637106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.071953545728327e-07, "output_power": 5.380169678823022e-07, "reference_active": true, "delta_db": 0.25620705416184086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3628863245634334e-08, "output_power": 1.390038460892699e-08, "reference_active": true, "delta_db": 0.08567183118527308}], "active_band_count": 3, "active_band_mae_db": 0.2553195312978283}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9282090581218085, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b7371658def1acb7a692ac007ce696b25dc1f30ff7d9f0a915b7afbc5e8c3d9e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9745305242611019, "standard_deviation": null, "minimum": 0.9745305242611019, "maximum": 0.9745305242611019, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6130831469466924, "standard_deviation": null, "minimum": 1.6130831469466924, "maximum": 1.6130831469466924, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4634038109578542, "standard_deviation": null, "minimum": 0.4634038109578542, "maximum": 0.4634038109578542, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19560405984208448, "standard_deviation": null, "minimum": 0.19560405984208448, "maximum": 0.19560405984208448, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946443261723266, "standard_deviation": null, "minimum": 0.9946443261723266, "maximum": 0.9946443261723266, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8888607826727349, "standard_deviation": null, "minimum": 0.8888607826727349, "maximum": 0.8888607826727349, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.613413230375246, "standard_deviation": null, "minimum": 0.613413230375246, "maximum": 0.613413230375246, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10229024011931291, "standard_deviation": null, "minimum": 0.10229024011931291, "maximum": 0.10229024011931291, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9574983618974391, "standard_deviation": null, "minimum": 0.9574983618974391, "maximum": 0.9574983618974391, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7372343283761233, "standard_deviation": null, "minimum": 1.7372343283761233, "maximum": 1.7372343283761233, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4959678282940101, "standard_deviation": null, "minimum": 0.4959678282940101, "maximum": 0.4959678282940101, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2525527499194943, "standard_deviation": null, "minimum": 0.2525527499194943, "maximum": 0.2525527499194943, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9893038921940567, "standard_deviation": null, "minimum": 0.9893038921940567, "maximum": 0.9893038921940567, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8697521851568458, "standard_deviation": null, "minimum": 0.8697521851568458, "maximum": 0.8697521851568458, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.2553195312978283, "standard_deviation": null, "minimum": 0.2553195312978283, "maximum": 0.2553195312978283, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12662609573415468, "standard_deviation": null, "minimum": 0.12662609573415468, "maximum": 0.12662609573415468, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9826668637344909, "standard_deviation": null, "minimum": 0.9826668637344909, "maximum": 0.9826668637344909, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.320842162638393, "standard_deviation": null, "minimum": 2.320842162638393, "maximum": 2.320842162638393, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8222495568581488, "standard_deviation": null, "minimum": 0.8222495568581488, "maximum": 0.8222495568581488, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19223489447033232, "standard_deviation": null, "minimum": 0.19223489447033232, "maximum": 0.19223489447033232, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9945816895968931, "standard_deviation": null, "minimum": 0.9945816895968931, "maximum": 0.9945816895968931, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.14079858724072, "standard_deviation": null, "minimum": 2.14079858724072, "maximum": 2.14079858724072, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1713702057566582, "standard_deviation": null, "minimum": 1.1713702057566582, "maximum": 1.1713702057566582, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1359596258554474, "standard_deviation": null, "minimum": 0.1359596258554474, "maximum": 0.1359596258554474, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.502420970661778, "standard_deviation": null, "minimum": 0.502420970661778, "maximum": 0.502420970661778, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.126034806565193, "standard_deviation": null, "minimum": 6.126034806565193, "maximum": 6.126034806565193, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.44440785869587485, "standard_deviation": null, "minimum": 0.44440785869587485, "maximum": 0.44440785869587485, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7127098829923681, "standard_deviation": null, "minimum": 0.7127098829923681, "maximum": 0.7127098829923681, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9791827019659012, "standard_deviation": null, "minimum": 0.9791827019659012, "maximum": 0.9791827019659012, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.056133323721495, "standard_deviation": null, "minimum": 3.056133323721495, "maximum": 3.056133323721495, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0212226484556581, "standard_deviation": null, "minimum": 1.0212226484556581, "maximum": 1.0212226484556581, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1816856307240048, "standard_deviation": null, "minimum": 0.1816856307240048, "maximum": 0.1816856307240048, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.992516602995846, "standard_deviation": null, "minimum": 0.992516602995846, "maximum": 0.992516602995846, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.501003503209866, "standard_deviation": null, "minimum": 2.501003503209866, "maximum": 2.501003503209866, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.5566277291007662, "standard_deviation": null, "minimum": 1.5566277291007662, "maximum": 1.5566277291007662, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12667144720687293, "standard_deviation": null, "minimum": 0.12667144720687293, "maximum": 0.12667144720687293, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1089"}, {"case": "librispeech_1188-133604-0003", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2666280269622803, "predicted_mos": 3.2666280269622803, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e3a60819e5340abe659c2428e9f80eb92756efd4cd21e035051ca6c04c6cbe61", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.336922199930996}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.38317718455703, "全帯域_10ms_rms_envelope_nrmse": 0.15593132193000292, "全帯域_10ms_envelope_correlation": 0.9847025334214279, "全帯域_10ms_modulation_mae_db": 1.2435531206854222, "全帯域_40ms_log_rms_mae_db": 2.930644207900185, "全帯域_40ms_rms_envelope_nrmse": 0.11122772335121787, "全帯域_40ms_envelope_correlation": 0.9941744989457857, "全帯域_40ms_modulation_mae_db": 1.140247326214108, "300–1000 Hz_10ms_log_rms_mae_db": 2.3258988873786217, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20792630529150524, "300–1000 Hz_10ms_envelope_correlation": 0.9849647239591408, "300–1000 Hz_10ms_modulation_mae_db": 1.9760097182563323, "300–1000 Hz_40ms_log_rms_mae_db": 1.7158566789612162, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1800337063447132, "300–1000 Hz_40ms_envelope_correlation": 0.9909919544954189, "300–1000 Hz_40ms_modulation_mae_db": 1.8872975486859378, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5710054080714033, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16278129714031309, "1000–3000 Hz_10ms_envelope_correlation": 0.9833762145292265, "1000–3000 Hz_10ms_modulation_mae_db": 1.237828950605048, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8610278701877547, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08511024479710257, "1000–3000 Hz_40ms_envelope_correlation": 0.9961659629216623, "1000–3000 Hz_40ms_modulation_mae_db": 1.2237429076212767, "3000–6000 Hz_10ms_log_rms_mae_db": 1.47432192640508, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3301049939236072, "3000–6000 Hz_10ms_envelope_correlation": 0.93514347988451, "3000–6000 Hz_10ms_modulation_mae_db": 2.366682692057424, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1105033257099097, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2209135963826746, "3000–6000 Hz_40ms_envelope_correlation": 0.9701693982756104, "3000–6000 Hz_40ms_modulation_mae_db": 2.3126311406645814, "mr_stft_db": 6.0506433269779505, "band_mae_db": 0.40676376079722393, "rms_gain_error_db": 0.5753482889954727, "output_peak": 0.3795859754221811}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.859816553967492, "spectral_convergence": 0.7674836640222019, "spectral_convergence_status": "ok", "amplitude_floor": 8.79928282883349e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.570275255451226, "spectral_convergence": 0.32223372545489365, "spectral_convergence_status": "ok", "amplitude_floor": 4.444740709272264e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.721838171515131, "spectral_convergence": 0.32106580513114297, "spectral_convergence_status": "ok", "amplitude_floor": 3.930299196146418e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002477612487613689, "output_power": 0.00025165692855779804, "delta_db": 0.067755105118345, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004515434578601184, "output_power": 0.0005537261585857185, "delta_db": 0.8859548607929961, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015370997116088465, "output_power": 0.0001750991057844087, "delta_db": 0.5658188710310331, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.4065683577917706e-05, "output_power": 4.84034132865409e-05, "delta_db": 0.4077547648277729, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.662327944441642e-06, "output_power": 9.443696133308696e-06, "delta_db": 0.3750738295350874, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.476431620775013e-05, "output_power": 2.398854059809421e-05, "delta_db": -0.13822513347810939, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.40676376079722393, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.5753482889954727, "pitch": null, "output_rms_dbfs": -29.737455239464442, "output_peak": 0.3795859754221811, "reference_rms_dbfs": -30.312803528459913, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0506433269779505}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.467120737686064e-05, "log_rms_mae_db": 3.38317718455703, "rms_envelope_nrmse": 0.15593132193000292, "envelope_correlation": 0.9847025334214279, "rms_gain_error_db": 0.5753531255963417, "reference_mean_power": 0.0009312818437003785, "output_mean_power": 0.0010632036247827284}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.494393820063344e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023666657525761142, "output_power": 0.00028883659631260454, "reference_active": true, "delta_db": 0.8651529231162634}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5441929621447868e-05, "output_power": 1.4563898260738131e-05, "reference_active": true, "delta_db": -0.2542393235632382}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.985747798516416e-06, "output_power": 3.829007270830969e-06, "reference_active": true, "delta_db": -2.611267115376765}], "active_band_count": 3, "active_band_mae_db": 1.2435531206854222}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.334380857742655e-05, "log_rms_mae_db": 2.930644207900185, "rms_envelope_nrmse": 0.11122772335121787, "envelope_correlation": 0.9941744989457857, "rms_gain_error_db": 0.5754142121415718, "reference_mean_power": 0.0009359486205868301, "output_mean_power": 0.0010685465089638568}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.189856446091556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021037225251172397, "output_power": 0.00026133642681078765, "reference_active": true, "delta_db": 0.9421149172246588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.644252267790675e-06, "output_power": 7.1833151066812346e-06, "reference_active": true, "delta_db": -0.270100929442282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4342900916585017e-07, "output_power": 2.0653057952633986e-07, "reference_active": true, "delta_db": -2.208526131975383}], "active_band_count": 3, "active_band_mae_db": 1.140247326214108}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.589555428061273e-05, "log_rms_mae_db": 2.3258988873786217, "rms_envelope_nrmse": 0.20792630529150524, "envelope_correlation": 0.9849647239591408, "rms_gain_error_db": 1.0798135780374125, "reference_mean_power": 0.00038327408961782034, "output_mean_power": 0.0004914629900354162}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.028222657916465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015396875485141256, "output_power": 0.00021211637731787843, "reference_active": true, "delta_db": 1.3914160366469726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.75189127488202e-06, "output_power": 1.0288758119343368e-05, "reference_active": true, "delta_db": 0.7026104365486384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1315883357233815e-06, "output_power": 1.7089004907624814e-06, "reference_active": true, "delta_db": -3.834002681573386}], "active_band_count": 3, "active_band_mae_db": 1.9760097182563323}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.149058869949865e-05, "log_rms_mae_db": 1.7158566789612162, "rms_envelope_nrmse": 0.1800337063447132, "envelope_correlation": 0.9909919544954189, "rms_gain_error_db": 1.0798029907953144, "reference_mean_power": 0.00038520090394369153, "output_mean_power": 0.000493932492803236}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9052955697898775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014058163319658512, "output_power": 0.0001961659532328992, "reference_active": true, "delta_db": 1.4469504858840478}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.901478078483036e-06, "output_power": 4.5782432625253985e-06, "reference_active": true, "delta_db": 0.6946969390175901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3393802456709067e-07, "output_power": 1.0401028734299281e-07, "reference_active": true, "delta_db": -3.5202452211561757}], "active_band_count": 3, "active_band_mae_db": 1.8872975486859378}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.043858753489399e-05, "log_rms_mae_db": 1.5710054080714033, "rms_envelope_nrmse": 0.16278129714031309, "envelope_correlation": 0.9833762145292265, "rms_gain_error_db": 0.3621483204502371, "reference_mean_power": 4.084610004445957e-05, "output_mean_power": 4.43982067013265e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0558567323423266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3562756220605441e-05, "output_power": 1.4400864826126268e-05, "reference_active": true, "delta_db": 0.2604061815860652}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5216123885214767e-06, "output_power": 1.5819201433415647e-06, "reference_active": true, "delta_db": 0.16880520658589143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602773162915538e-07, "output_power": 3.09955131709657e-07, "reference_active": true, "delta_db": -3.284275463643187}], "active_band_count": 3, "active_band_mae_db": 1.237828950605048}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.7804457921821952e-05, "log_rms_mae_db": 0.8610278701877547, "rms_envelope_nrmse": 0.08511024479710257, "envelope_correlation": 0.9961659629216623, "rms_gain_error_db": 0.3621599186776056, "reference_mean_power": 4.105094002568101e-05, "output_mean_power": 4.462097938244322e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8992373996258767e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2457002317192863e-05, "output_power": 1.3301740896104754e-05, "reference_active": true, "delta_db": 0.28494938747809334}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.119435793981677e-07, "output_power": 8.251077391999123e-07, "reference_active": true, "delta_db": 0.0698480876076771}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.742155092447501e-08, "output_power": 1.743727665326634e-08, "reference_active": true, "delta_db": -3.3164312477780595}], "active_band_count": 3, "active_band_mae_db": 1.2237429076212767}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.7924378350629476e-05, "log_rms_mae_db": 1.47432192640508, "rms_envelope_nrmse": 0.3301049939236072, "envelope_correlation": 0.93514347988451, "rms_gain_error_db": -0.07176929856841396, "reference_mean_power": 2.620464364014996e-05, "output_mean_power": 2.577515752777127e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.347390510898905e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.954495563916905e-05, "output_power": 1.697828061574317e-05, "reference_active": true, "delta_db": -0.6114098176738831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.79596482520383e-06, "output_power": 1.3473890410665801e-06, "reference_active": true, "delta_db": -3.170386927266276}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.14743402003256e-06, "output_power": 5.344444715009179e-07, "reference_active": true, "delta_db": -3.318251331232113}], "active_band_count": 3, "active_band_mae_db": 2.366682692057424}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.7050501746546627e-05, "log_rms_mae_db": 1.1105033257099097, "rms_envelope_nrmse": 0.2209135963826746, "envelope_correlation": 0.9701693982756104, "rms_gain_error_db": -0.07176824465553504, "reference_mean_power": 2.633641752064706e-05, "output_mean_power": 2.5904777960954617e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2809655135480154e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9754827323518705e-05, "output_power": 1.6566203705561022e-05, "reference_active": true, "delta_db": -0.7645024053835925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4198245590923022e-06, "output_power": 6.673473530139406e-07, "reference_active": true, "delta_db": -3.278827418043045}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.51727649987637e-08, "output_power": 3.3466460815664125e-08, "reference_active": true, "delta_db": -2.8945635985671063}], "active_band_count": 3, "active_band_mae_db": 2.3126311406645814}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7460459931535739, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5235798cfedb21ec2c4ee4783894a81665f0822749dd69ef08f147ee009b83c6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9833762145292265, "standard_deviation": null, "minimum": 0.9833762145292265, "maximum": 0.9833762145292265, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5710054080714033, "standard_deviation": null, "minimum": 1.5710054080714033, "maximum": 1.5710054080714033, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.237828950605048, "standard_deviation": null, "minimum": 1.237828950605048, "maximum": 1.237828950605048, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16278129714031309, "standard_deviation": null, "minimum": 0.16278129714031309, "maximum": 0.16278129714031309, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961659629216623, "standard_deviation": null, "minimum": 0.9961659629216623, "maximum": 0.9961659629216623, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8610278701877547, "standard_deviation": null, "minimum": 0.8610278701877547, "maximum": 0.8610278701877547, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2237429076212767, "standard_deviation": null, "minimum": 1.2237429076212767, "maximum": 1.2237429076212767, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08511024479710257, "standard_deviation": null, "minimum": 0.08511024479710257, "maximum": 0.08511024479710257, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.93514347988451, "standard_deviation": null, "minimum": 0.93514347988451, "maximum": 0.93514347988451, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.47432192640508, "standard_deviation": null, "minimum": 1.47432192640508, "maximum": 1.47432192640508, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.366682692057424, "standard_deviation": null, "minimum": 2.366682692057424, "maximum": 2.366682692057424, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3301049939236072, "standard_deviation": null, "minimum": 0.3301049939236072, "maximum": 0.3301049939236072, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9701693982756104, "standard_deviation": null, "minimum": 0.9701693982756104, "maximum": 0.9701693982756104, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1105033257099097, "standard_deviation": null, "minimum": 1.1105033257099097, "maximum": 1.1105033257099097, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.3126311406645814, "standard_deviation": null, "minimum": 2.3126311406645814, "maximum": 2.3126311406645814, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2209135963826746, "standard_deviation": null, "minimum": 0.2209135963826746, "maximum": 0.2209135963826746, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9849647239591408, "standard_deviation": null, "minimum": 0.9849647239591408, "maximum": 0.9849647239591408, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3258988873786217, "standard_deviation": null, "minimum": 2.3258988873786217, "maximum": 2.3258988873786217, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.9760097182563323, "standard_deviation": null, "minimum": 1.9760097182563323, "maximum": 1.9760097182563323, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20792630529150524, "standard_deviation": null, "minimum": 0.20792630529150524, "maximum": 0.20792630529150524, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9909919544954189, "standard_deviation": null, "minimum": 0.9909919544954189, "maximum": 0.9909919544954189, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7158566789612162, "standard_deviation": null, "minimum": 1.7158566789612162, "maximum": 1.7158566789612162, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.8872975486859378, "standard_deviation": null, "minimum": 1.8872975486859378, "maximum": 1.8872975486859378, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1800337063447132, "standard_deviation": null, "minimum": 0.1800337063447132, "maximum": 0.1800337063447132, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.40676376079722393, "standard_deviation": null, "minimum": 0.40676376079722393, "maximum": 0.40676376079722393, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.0506433269779505, "standard_deviation": null, "minimum": 6.0506433269779505, "maximum": 6.0506433269779505, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.3795859754221811, "standard_deviation": null, "minimum": 0.3795859754221811, "maximum": 0.3795859754221811, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.5753482889954727, "standard_deviation": null, "minimum": 0.5753482889954727, "maximum": 0.5753482889954727, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9847025334214279, "standard_deviation": null, "minimum": 0.9847025334214279, "maximum": 0.9847025334214279, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.38317718455703, "standard_deviation": null, "minimum": 3.38317718455703, "maximum": 3.38317718455703, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.2435531206854222, "standard_deviation": null, "minimum": 1.2435531206854222, "maximum": 1.2435531206854222, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15593132193000292, "standard_deviation": null, "minimum": 0.15593132193000292, "maximum": 0.15593132193000292, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9941744989457857, "standard_deviation": null, "minimum": 0.9941744989457857, "maximum": 0.9941744989457857, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.930644207900185, "standard_deviation": null, "minimum": 2.930644207900185, "maximum": 2.930644207900185, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.140247326214108, "standard_deviation": null, "minimum": 1.140247326214108, "maximum": 1.140247326214108, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11122772335121787, "standard_deviation": null, "minimum": 0.11122772335121787, "maximum": 0.11122772335121787, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1188"}, {"case": "librispeech_1188-133604-0031", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2458810806274414, "predicted_mos": 3.2458810806274414, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68000, "input_sample_rate": 16000, "input_seconds": 4.25, "input_float32_sha256": "75eb84c266e0ef4431d5589a6a04feafeb5c3b36bc360c9c736856d64a8e93a2", "internal_samples": 68000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.849458199925721}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 5.609514033155247, "全帯域_10ms_rms_envelope_nrmse": 0.13779097105083732, "全帯域_10ms_envelope_correlation": 0.9904541392480073, "全帯域_10ms_modulation_mae_db": 1.1290953597070639, "全帯域_40ms_log_rms_mae_db": 5.373528647577003, "全帯域_40ms_rms_envelope_nrmse": 0.10646715208234203, "全帯域_40ms_envelope_correlation": 0.9963352049065383, "全帯域_40ms_modulation_mae_db": 0.9385956647466474, "300–1000 Hz_10ms_log_rms_mae_db": 3.023235401830885, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18010093837673913, "300–1000 Hz_10ms_envelope_correlation": 0.9885805591006955, "300–1000 Hz_10ms_modulation_mae_db": 0.7279345268260542, "300–1000 Hz_40ms_log_rms_mae_db": 2.710291058095824, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15068941177464945, "300–1000 Hz_40ms_envelope_correlation": 0.9942524078323821, "300–1000 Hz_40ms_modulation_mae_db": 1.0306577768224578, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5605688885885427, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17285661326069546, "1000–3000 Hz_10ms_envelope_correlation": 0.9868104905483519, "1000–3000 Hz_10ms_modulation_mae_db": 1.2625145304088567, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8789548638775316, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12769022052117568, "1000–3000 Hz_40ms_envelope_correlation": 0.9946183755274229, "1000–3000 Hz_40ms_modulation_mae_db": 1.3444032555197438, "3000–6000 Hz_10ms_log_rms_mae_db": 1.598113209459517, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29911190813040744, "3000–6000 Hz_10ms_envelope_correlation": 0.9468692092064666, "3000–6000 Hz_10ms_modulation_mae_db": 1.414732859974394, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8991468307083788, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1857322583987023, "3000–6000 Hz_40ms_envelope_correlation": 0.978793950375228, "3000–6000 Hz_40ms_modulation_mae_db": 1.6105479907915459, "mr_stft_db": 6.265942539044242, "band_mae_db": 0.5755075678453464, "rms_gain_error_db": 0.6647880759019567, "output_peak": 0.5150196716309051}, "full_scale": {"sample_count": 68000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68000, "output_samples": 68000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.164752590388816, "spectral_convergence": 0.8688399800098922, "spectral_convergence_status": "ok", "amplitude_floor": 1.0380731662877042e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.719691862044561, "spectral_convergence": 0.32621362173599233, "spectral_convergence_status": "ok", "amplitude_floor": 5.867181533968612e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.913383164699348, "spectral_convergence": 0.25382005253001694, "spectral_convergence_status": "ok", "amplitude_floor": 4.6154090777751923e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006114723941341076, "output_power": 0.0006363493892680899, "delta_db": 0.17318776715519565, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011102815674838913, "output_power": 0.00133464534362092, "delta_db": 0.799327457441093, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00031219826577364904, "output_power": 0.00040149047892100876, "delta_db": 1.0924476445680515, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.006894391321554e-05, "output_power": 0.00010565028207664703, "delta_db": 0.6929559038286969, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.882819827254369e-06, "output_power": 9.167528955352535e-06, "delta_db": -0.326285875155064, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.595718179316582e-06, "output_power": 6.09171966922055e-06, "delta_db": 0.3688407589239777, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5755075678453464, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6647880759019567, "pitch": null, "output_rms_dbfs": -26.032089605963723, "output_peak": 0.5150196716309051, "reference_rms_dbfs": -26.69687768186568, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.265942539044242}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.0001318685794969069, "log_rms_mae_db": 5.609514033155247, "rms_envelope_nrmse": 0.13779097105083732, "envelope_correlation": 0.9904541392480073, "rms_gain_error_db": 0.664796816046799, "reference_mean_power": 0.0021420152585098387, "output_mean_power": 0.0024963314130004755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 1.3134511495070254e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003582591828573965, "output_power": 0.00039725488105142045, "reference_active": true, "delta_db": 0.4487191107007513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.463795601086152e-05, "output_power": 5.1363926541998205e-05, "reference_active": true, "delta_db": 0.6095391632291696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.556588167851325e-06, "output_power": 1.2918989178248827e-05, "reference_active": true, "delta_db": 2.3290278051912705}], "active_band_count": 3, "active_band_mae_db": 1.1290953597070639}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 0.00012588557284836115, "log_rms_mae_db": 5.373528647577003, "rms_envelope_nrmse": 0.10646715208234203, "envelope_correlation": 0.9963352049065383, "rms_gain_error_db": 0.6650048342884907, "reference_mean_power": 0.0021571542863483557, "output_mean_power": 0.002514095042839533}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 1.2592546923162062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031562831458090785, "output_power": 0.0003554128693886613, "reference_active": true, "delta_db": 0.5155719317697087}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9876750850691092e-05, "output_power": 2.3483255150453946e-05, "reference_active": true, "delta_db": 0.7241290283219637}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.602361817361658e-07, "output_power": 9.490911503828453e-07, "reference_active": true, "delta_db": 1.5760860341482696}], "active_band_count": 3, "active_band_mae_db": 0.9385956647466474}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 0.00010444605382463925, "log_rms_mae_db": 3.023235401830885, "rms_envelope_nrmse": 0.18010093837673913, "envelope_correlation": 0.9885805591006955, "rms_gain_error_db": 0.9577895449156206, "reference_mean_power": 0.0005604503758559807, "output_mean_power": 0.000698740826496309}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 5.643164593552106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025020135526255573, "output_power": 0.0003207660112563883, "reference_active": true, "delta_db": 1.0789868589495182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.833230380435073e-05, "output_power": 4.9320017656981456e-05, "reference_active": true, "delta_db": 1.0945830189503722}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.791716397073579e-06, "output_power": 6.775731257314634e-06, "reference_active": true, "delta_db": -0.010233702578272273}], "active_band_count": 3, "active_band_mae_db": 0.7279345268260542}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 9.32306023645191e-05, "log_rms_mae_db": 2.710291058095824, "rms_envelope_nrmse": 0.15068941177464945, "envelope_correlation": 0.9942524078323821, "rms_gain_error_db": 0.9577824107697972, "reference_mean_power": 0.0005644391167624198, "output_mean_power": 0.0007037126283965411}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 5.412619397247975e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022898404327685652, "output_power": 0.0002945282566396906, "reference_active": true, "delta_db": 1.093217471027887}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5949246143041614e-05, "output_power": 2.0835689680937712e-05, "reference_active": true, "delta_db": 1.1606772006899049}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.408378605310095e-07, "output_power": 5.283726094827267e-07, "reference_active": true, "delta_db": -0.8380786587495815}], "active_band_count": 3, "active_band_mae_db": 1.0306577768224578}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 4.754025004833877e-05, "log_rms_mae_db": 1.5605688885885427, "rms_envelope_nrmse": 0.17285661326069546, "envelope_correlation": 0.9868104905483519, "rms_gain_error_db": 0.7161981435394714, "reference_mean_power": 8.438545283737632e-05, "output_mean_power": 9.951473701432126e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 6.948180578959798e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5558932008941216e-05, "output_power": 4.07555827063248e-05, "reference_active": true, "delta_db": 0.5923839815790741}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9367090695871985e-06, "output_power": 4.599935377693396e-06, "reference_active": true, "delta_db": 0.6761840975349944}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.57982090332121e-07, "output_power": 4.243863365399344e-07, "reference_active": true, "delta_db": -2.518975512112502}], "active_band_count": 3, "active_band_mae_db": 1.2625145304088567}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 4.260586736236329e-05, "log_rms_mae_db": 0.8789548638775316, "rms_envelope_nrmse": 0.12769022052117568, "envelope_correlation": 0.9946183755274229, "rms_gain_error_db": 0.7161991353402642, "reference_mean_power": 8.498589632737057e-05, "output_mean_power": 0.00010022285559873859}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 6.685624503092077e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2789374619144395e-05, "output_power": 3.773614572924064e-05, "reference_active": true, "delta_db": 0.6102440680303912}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.67867676799424e-06, "output_power": 1.9688281422513294e-06, "reference_active": true, "delta_db": 0.6924072844249985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.924150120657105e-08, "output_power": 3.159149757128517e-08, "reference_active": true, "delta_db": -2.730558414103842}], "active_band_count": 3, "active_band_mae_db": 1.3444032555197438}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 849, "rms_floor": 2.7803794970768208e-05, "log_rms_mae_db": 1.598113209459517, "rms_envelope_nrmse": 0.29911190813040744, "envelope_correlation": 0.9468692092064666, "rms_gain_error_db": 0.08474219677544778, "reference_mean_power": 9.160796502339738e-06, "output_mean_power": 9.341302917878405e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 849, "frequency_resolution_hz": 0.23557126030624262, "power_floor": 3.7361073674013695e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6569760918264083e-06, "output_power": 1.8447949476299525e-06, "reference_active": true, "delta_db": 0.46631858425742057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.026752210402781e-07, "output_power": 6.135534777624088e-07, "reference_active": true, "delta_db": 0.07769086291780165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8917938491313036e-07, "output_power": 8.069653661916333e-08, "reference_active": true, "delta_db": -3.70018913274796}], "active_band_count": 3, "active_band_mae_db": 1.414732859974394}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 843, "rms_floor": 2.058581204447328e-05, "log_rms_mae_db": 0.8991468307083788, "rms_envelope_nrmse": 0.1857322583987023, "envelope_correlation": 0.978793950375228, "rms_gain_error_db": 0.08474204284514632, "reference_mean_power": 9.225978949789671e-06, "output_mean_power": 9.407769401635056e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 843, "frequency_resolution_hz": 0.2372479240806643, "power_floor": 3.5247411287487305e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7104686599805127e-06, "output_power": 1.8485515190786156e-06, "reference_active": true, "delta_db": 0.337164376086018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2748792658937344e-07, "output_power": 3.039642967337651e-07, "reference_active": true, "delta_db": -0.3237271868575098}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.947253788526217e-08, "output_power": 7.453277935271517e-09, "reference_active": true, "delta_db": -4.17075240943111}], "active_band_count": 3, "active_band_mae_db": 1.6105479907915459}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8299198779902748, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f8245d99298394124222249532e83aaf7b22b88ce55b3383cf47b07f8cf3cb1a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9868104905483519, "standard_deviation": null, "minimum": 0.9868104905483519, "maximum": 0.9868104905483519, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5605688885885427, "standard_deviation": null, "minimum": 1.5605688885885427, "maximum": 1.5605688885885427, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2625145304088567, "standard_deviation": null, "minimum": 1.2625145304088567, "maximum": 1.2625145304088567, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17285661326069546, "standard_deviation": null, "minimum": 0.17285661326069546, "maximum": 0.17285661326069546, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946183755274229, "standard_deviation": null, "minimum": 0.9946183755274229, "maximum": 0.9946183755274229, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8789548638775316, "standard_deviation": null, "minimum": 0.8789548638775316, "maximum": 0.8789548638775316, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3444032555197438, "standard_deviation": null, "minimum": 1.3444032555197438, "maximum": 1.3444032555197438, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12769022052117568, "standard_deviation": null, "minimum": 0.12769022052117568, "maximum": 0.12769022052117568, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9468692092064666, "standard_deviation": null, "minimum": 0.9468692092064666, "maximum": 0.9468692092064666, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.598113209459517, "standard_deviation": null, "minimum": 1.598113209459517, "maximum": 1.598113209459517, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.414732859974394, "standard_deviation": null, "minimum": 1.414732859974394, "maximum": 1.414732859974394, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29911190813040744, "standard_deviation": null, "minimum": 0.29911190813040744, "maximum": 0.29911190813040744, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.978793950375228, "standard_deviation": null, "minimum": 0.978793950375228, "maximum": 0.978793950375228, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8991468307083788, "standard_deviation": null, "minimum": 0.8991468307083788, "maximum": 0.8991468307083788, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6105479907915459, "standard_deviation": null, "minimum": 1.6105479907915459, "maximum": 1.6105479907915459, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1857322583987023, "standard_deviation": null, "minimum": 0.1857322583987023, "maximum": 0.1857322583987023, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9885805591006955, "standard_deviation": null, "minimum": 0.9885805591006955, "maximum": 0.9885805591006955, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 3.023235401830885, "standard_deviation": null, "minimum": 3.023235401830885, "maximum": 3.023235401830885, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7279345268260542, "standard_deviation": null, "minimum": 0.7279345268260542, "maximum": 0.7279345268260542, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18010093837673913, "standard_deviation": null, "minimum": 0.18010093837673913, "maximum": 0.18010093837673913, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942524078323821, "standard_deviation": null, "minimum": 0.9942524078323821, "maximum": 0.9942524078323821, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.710291058095824, "standard_deviation": null, "minimum": 2.710291058095824, "maximum": 2.710291058095824, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0306577768224578, "standard_deviation": null, "minimum": 1.0306577768224578, "maximum": 1.0306577768224578, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15068941177464945, "standard_deviation": null, "minimum": 0.15068941177464945, "maximum": 0.15068941177464945, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5755075678453464, "standard_deviation": null, "minimum": 0.5755075678453464, "maximum": 0.5755075678453464, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.265942539044242, "standard_deviation": null, "minimum": 6.265942539044242, "maximum": 6.265942539044242, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5150196716309051, "standard_deviation": null, "minimum": 0.5150196716309051, "maximum": 0.5150196716309051, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6647880759019567, "standard_deviation": null, "minimum": 0.6647880759019567, "maximum": 0.6647880759019567, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9904541392480073, "standard_deviation": null, "minimum": 0.9904541392480073, "maximum": 0.9904541392480073, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 5.609514033155247, "standard_deviation": null, "minimum": 5.609514033155247, "maximum": 5.609514033155247, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1290953597070639, "standard_deviation": null, "minimum": 1.1290953597070639, "maximum": 1.1290953597070639, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.13779097105083732, "standard_deviation": null, "minimum": 0.13779097105083732, "maximum": 0.13779097105083732, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963352049065383, "standard_deviation": null, "minimum": 0.9963352049065383, "maximum": 0.9963352049065383, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 5.373528647577003, "standard_deviation": null, "minimum": 5.373528647577003, "maximum": 5.373528647577003, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9385956647466474, "standard_deviation": null, "minimum": 0.9385956647466474, "maximum": 0.9385956647466474, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10646715208234203, "standard_deviation": null, "minimum": 0.10646715208234203, "maximum": 0.10646715208234203, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1188"}, {"case": "librispeech_1188-133604-0037", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9190211296081543, "predicted_mos": 2.9190211296081543, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b984f8c52620d10fc842a64311f17ffd49f3801be697032d040809f14341a5b0", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.782677699811757}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.0243551585417117, "全帯域_10ms_rms_envelope_nrmse": 0.1302303127837023, "全帯域_10ms_envelope_correlation": 0.9889117538853047, "全帯域_10ms_modulation_mae_db": 0.4594066306223345, "全帯域_40ms_log_rms_mae_db": 2.7034858418194854, "全帯域_40ms_rms_envelope_nrmse": 0.09906240960673009, "全帯域_40ms_envelope_correlation": 0.9954631342909911, "全帯域_40ms_modulation_mae_db": 0.6979594669208286, "300–1000 Hz_10ms_log_rms_mae_db": 2.304475687447635, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17211838956421285, "300–1000 Hz_10ms_envelope_correlation": 0.9867542375446722, "300–1000 Hz_10ms_modulation_mae_db": 0.8571131973856237, "300–1000 Hz_40ms_log_rms_mae_db": 1.8879130282711218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14423810430936643, "300–1000 Hz_40ms_envelope_correlation": 0.9923340561961279, "300–1000 Hz_40ms_modulation_mae_db": 1.014380178302157, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4887488249761027, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14133060261371613, "1000–3000 Hz_10ms_envelope_correlation": 0.9914605658070736, "1000–3000 Hz_10ms_modulation_mae_db": 0.9081150668698835, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8995145682774864, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10940275436265819, "1000–3000 Hz_40ms_envelope_correlation": 0.9962592782206046, "1000–3000 Hz_40ms_modulation_mae_db": 0.9847497510187563, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6569992864967729, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28579431809991074, "3000–6000 Hz_10ms_envelope_correlation": 0.9512027579618225, "3000–6000 Hz_10ms_modulation_mae_db": 0.2970450842362676, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0185578764950263, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19331130222812434, "3000–6000 Hz_40ms_envelope_correlation": 0.9771510476110663, "3000–6000 Hz_40ms_modulation_mae_db": 0.2059229724915014, "mr_stft_db": 6.037902539016131, "band_mae_db": 0.46730993113884806, "rms_gain_error_db": 0.5953482263094706, "output_peak": 0.6797615360483316}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.972413307128258, "spectral_convergence": 0.8310824950106471, "spectral_convergence_status": "ok", "amplitude_floor": 1.0507333742302167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.45187533106667, "spectral_convergence": 0.3008664488768978, "spectral_convergence_status": "ok", "amplitude_floor": 5.595595188050952e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.689418978853465, "spectral_convergence": 0.24256259016048415, "spectral_convergence_status": "ok", "amplitude_floor": 5.269142735102521e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007416076491485813, "output_power": 0.0007557064841665485, "delta_db": 0.08178948003959845, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001352027771357195, "output_power": 0.0016333119264510666, "delta_db": 0.8208352096464406, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000371703582440024, "output_power": 0.00043699341043020045, "delta_db": 0.7027814134761907, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018903159721465413, "output_power": 0.00022154970110680182, "delta_db": 0.6893676455683155, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5427403014296353e-05, "output_power": 1.648501852928208e-05, "delta_db": 0.28796614764462813, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1014091243658427e-05, "output_power": 2.2111722872370183e-05, "delta_db": 0.22111969045791516, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.46730993113884806, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.5953482263094706, "pitch": null, "output_rms_dbfs": -25.10581806326412, "output_peak": 0.6797615360483316, "reference_rms_dbfs": -25.701166289573592, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.037902539016131}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018114550052554645, "log_rms_mae_db": 3.0243551585417117, "rms_envelope_nrmse": 0.1302303127837023, "envelope_correlation": 0.9889117538853047, "rms_gain_error_db": 0.5958602920749096, "reference_mean_power": 0.002692702060931978, "output_mean_power": 0.0030886900706730515}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0110233554726448e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033858489572855915, "output_power": 0.0003916696137748277, "reference_active": true, "delta_db": 0.6325229940912475}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.201391758025471e-05, "output_power": 4.5979246982453886e-05, "reference_active": true, "delta_db": 0.3916867526748159}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.1566777501091025e-06, "output_power": 7.764481809116948e-06, "reference_active": true, "delta_db": 0.35401014510094}], "active_band_count": 3, "active_band_mae_db": 0.4594066306223345}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015936923488846764, "log_rms_mae_db": 2.7034858418194854, "rms_envelope_nrmse": 0.09906240960673009, "envelope_correlation": 0.9954631342909911, "rms_gain_error_db": 0.59857176080616, "reference_mean_power": 0.002702794865208283, "output_mean_power": 0.0031022033411734736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.583684100944911e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031173336044584235, "output_power": 0.00036165586885077714, "reference_active": true, "delta_db": 0.6451223589410167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.830633609453067e-05, "output_power": 2.1142674797456758e-05, "reference_active": true, "delta_db": 0.6255849830240403}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.642076702713698e-07, "output_power": 5.610859813603983e-07, "reference_active": true, "delta_db": 0.8231710587974284}], "active_band_count": 3, "active_band_mae_db": 0.6979594669208286}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001259260583850441, "log_rms_mae_db": 2.304475687447635, "rms_envelope_nrmse": 0.17211838956421285, "envelope_correlation": 0.9867542375446722, "rms_gain_error_db": 0.8284924843243533, "reference_mean_power": 0.0007105481356967998, "output_mean_power": 0.0008598897115964031}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8068878391559104e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016083630318364556, "output_power": 0.00020875851114062654, "reference_active": true, "delta_db": 1.132601085160598}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8008354908350678e-05, "output_power": 2.1779618117844794e-05, "reference_active": true, "delta_db": 0.825762194553164}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.325041103743567e-06, "output_power": 2.0189845883107193e-06, "reference_active": true, "delta_db": -0.6129763124431089}], "active_band_count": 3, "active_band_mae_db": 0.8571131973856237}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010786919997015104, "log_rms_mae_db": 1.8879130282711218, "rms_envelope_nrmse": 0.14423810430936643, "envelope_correlation": 0.9923340561961279, "rms_gain_error_db": 0.8250218235223403, "reference_mean_power": 0.0007140165693307253, "output_mean_power": 0.0008633968749450939}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6527252040517484e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000151264236566862, "output_power": 0.00019570372911002909, "reference_active": true, "delta_db": 1.1186284133371505}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.688501064647705e-06, "output_power": 9.088770911413905e-06, "reference_active": true, "delta_db": 0.7266347803787288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2694181944274492e-07, "output_power": 1.7223713716314006e-07, "reference_active": true, "delta_db": -1.1978773411905923}], "active_band_count": 3, "active_band_mae_db": 1.014380178302157}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.418520257385562e-05, "log_rms_mae_db": 1.4887488249761027, "rms_envelope_nrmse": 0.14133060261371613, "envelope_correlation": 0.9914605658070736, "rms_gain_error_db": 0.6457297755654884, "reference_mean_power": 0.000153818152796868, "output_mean_power": 0.00017847630623445644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.418336470516266e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4748514150479256e-05, "output_power": 2.6895679203092873e-05, "reference_active": true, "delta_db": 0.36133386188689687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.277056498060538e-06, "output_power": 3.2714263326360234e-06, "reference_active": true, "delta_db": -0.007467837793179064}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.913659986768058e-07, "output_power": 5.182051368438415e-07, "reference_active": true, "delta_db": -2.3555435009295747}], "active_band_count": 3, "active_band_mae_db": 0.9081150668698835}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.421085502718712e-05, "log_rms_mae_db": 0.8995145682774864, "rms_envelope_nrmse": 0.10940275436265819, "envelope_correlation": 0.9962592782206046, "rms_gain_error_db": 0.6458110242542894, "reference_mean_power": 0.00015458728634615244, "output_mean_power": 0.00017937209310767464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1974511373265086e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3402826997507283e-05, "output_power": 2.5112622282140984e-05, "reference_active": true, "delta_db": 0.3062374232831978}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3826120793884328e-06, "output_power": 1.4327553564424927e-06, "reference_active": true, "delta_db": 0.15471693930065716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6109211314667126e-08, "output_power": 2.59691780947675e-08, "reference_active": true, "delta_db": -2.4932948904724137}], "active_band_count": 3, "active_band_mae_db": 0.9847497510187563}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.007885661618194e-05, "log_rms_mae_db": 1.6569992864967729, "rms_envelope_nrmse": 0.28579431809991074, "envelope_correlation": 0.9512027579618225, "rms_gain_error_db": 0.3068610453810405, "reference_mean_power": 2.0928545188569684e-05, "output_mean_power": 2.2460796067789737e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4882901242606597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1348414996445462e-05, "output_power": 1.2058235407335463e-05, "reference_active": true, "delta_db": 0.26348549155885786}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8308296968161944e-06, "output_power": 2.939071701699408e-06, "reference_active": true, "delta_db": 0.16296438342763098}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.670421987321005e-07, "output_power": 7.42372450993631e-07, "reference_active": true, "delta_db": 0.4646853777223139}], "active_band_count": 3, "active_band_mae_db": 0.2970450842362676}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3258654185194873e-05, "log_rms_mae_db": 1.0185578764950263, "rms_envelope_nrmse": 0.19331130222812434, "envelope_correlation": 0.9771510476110663, "rms_gain_error_db": 0.30688750155960504, "reference_mean_power": 2.103354073620274e-05, "output_mean_power": 2.257361621374892e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.390796824652454e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1287290435083056e-05, "output_power": 1.2158133862030863e-05, "reference_active": true, "delta_db": 0.32277220602392936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.514784832162418e-06, "output_power": 1.5898938654084568e-06, "reference_active": true, "delta_db": 0.2101718583015772}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.433812491322156e-08, "output_power": 4.52126338934955e-08, "reference_active": true, "delta_db": 0.08482485314899771}], "active_band_count": 3, "active_band_mae_db": 0.2059229724915014}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9128532385096347, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5267c3576b2138a0fa785f661205d6d5d8631e8e04b5c4c7a0e0602c91e19cbd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9914605658070736, "standard_deviation": null, "minimum": 0.9914605658070736, "maximum": 0.9914605658070736, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4887488249761027, "standard_deviation": null, "minimum": 1.4887488249761027, "maximum": 1.4887488249761027, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9081150668698835, "standard_deviation": null, "minimum": 0.9081150668698835, "maximum": 0.9081150668698835, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14133060261371613, "standard_deviation": null, "minimum": 0.14133060261371613, "maximum": 0.14133060261371613, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962592782206046, "standard_deviation": null, "minimum": 0.9962592782206046, "maximum": 0.9962592782206046, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8995145682774864, "standard_deviation": null, "minimum": 0.8995145682774864, "maximum": 0.8995145682774864, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9847497510187563, "standard_deviation": null, "minimum": 0.9847497510187563, "maximum": 0.9847497510187563, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10940275436265819, "standard_deviation": null, "minimum": 0.10940275436265819, "maximum": 0.10940275436265819, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9512027579618225, "standard_deviation": null, "minimum": 0.9512027579618225, "maximum": 0.9512027579618225, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6569992864967729, "standard_deviation": null, "minimum": 1.6569992864967729, "maximum": 1.6569992864967729, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.2970450842362676, "standard_deviation": null, "minimum": 0.2970450842362676, "maximum": 0.2970450842362676, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.28579431809991074, "standard_deviation": null, "minimum": 0.28579431809991074, "maximum": 0.28579431809991074, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9771510476110663, "standard_deviation": null, "minimum": 0.9771510476110663, "maximum": 0.9771510476110663, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0185578764950263, "standard_deviation": null, "minimum": 1.0185578764950263, "maximum": 1.0185578764950263, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.2059229724915014, "standard_deviation": null, "minimum": 0.2059229724915014, "maximum": 0.2059229724915014, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19331130222812434, "standard_deviation": null, "minimum": 0.19331130222812434, "maximum": 0.19331130222812434, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9867542375446722, "standard_deviation": null, "minimum": 0.9867542375446722, "maximum": 0.9867542375446722, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.304475687447635, "standard_deviation": null, "minimum": 2.304475687447635, "maximum": 2.304475687447635, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8571131973856237, "standard_deviation": null, "minimum": 0.8571131973856237, "maximum": 0.8571131973856237, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17211838956421285, "standard_deviation": null, "minimum": 0.17211838956421285, "maximum": 0.17211838956421285, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9923340561961279, "standard_deviation": null, "minimum": 0.9923340561961279, "maximum": 0.9923340561961279, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.8879130282711218, "standard_deviation": null, "minimum": 1.8879130282711218, "maximum": 1.8879130282711218, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.014380178302157, "standard_deviation": null, "minimum": 1.014380178302157, "maximum": 1.014380178302157, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14423810430936643, "standard_deviation": null, "minimum": 0.14423810430936643, "maximum": 0.14423810430936643, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.46730993113884806, "standard_deviation": null, "minimum": 0.46730993113884806, "maximum": 0.46730993113884806, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.037902539016131, "standard_deviation": null, "minimum": 6.037902539016131, "maximum": 6.037902539016131, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6797615360483316, "standard_deviation": null, "minimum": 0.6797615360483316, "maximum": 0.6797615360483316, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.5953482263094706, "standard_deviation": null, "minimum": 0.5953482263094706, "maximum": 0.5953482263094706, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9889117538853047, "standard_deviation": null, "minimum": 0.9889117538853047, "maximum": 0.9889117538853047, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.0243551585417117, "standard_deviation": null, "minimum": 3.0243551585417117, "maximum": 3.0243551585417117, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4594066306223345, "standard_deviation": null, "minimum": 0.4594066306223345, "maximum": 0.4594066306223345, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1302303127837023, "standard_deviation": null, "minimum": 0.1302303127837023, "maximum": 0.1302303127837023, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9954631342909911, "standard_deviation": null, "minimum": 0.9954631342909911, "maximum": 0.9954631342909911, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.7034858418194854, "standard_deviation": null, "minimum": 2.7034858418194854, "maximum": 2.7034858418194854, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6979594669208286, "standard_deviation": null, "minimum": 0.6979594669208286, "maximum": 0.6979594669208286, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09906240960673009, "standard_deviation": null, "minimum": 0.09906240960673009, "maximum": 0.09906240960673009, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1188"}, {"case": "librispeech_121-121726-0012", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.5604605674743652, "predicted_mos": 2.5604605674743652, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64720, "input_sample_rate": 16000, "input_seconds": 4.045, "input_float32_sha256": "b08b7aa45a3aa0d6f710e210b9ff5023728c64d8a59120cb7a330fc387182e19", "internal_samples": 64720, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.777967399917543}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.0605551084553402, "全帯域_10ms_rms_envelope_nrmse": 0.14144590945775157, "全帯域_10ms_envelope_correlation": 0.9898811036933927, "全帯域_10ms_modulation_mae_db": 0.6001631746443973, "全帯域_40ms_log_rms_mae_db": 0.9748404868695941, "全帯域_40ms_rms_envelope_nrmse": 0.10548854729298346, "全帯域_40ms_envelope_correlation": 0.9970959195406944, "全帯域_40ms_modulation_mae_db": 1.4213753979031816, "300–1000 Hz_10ms_log_rms_mae_db": 1.1832586298520942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1494033508982216, "300–1000 Hz_10ms_envelope_correlation": 0.991482066889388, "300–1000 Hz_10ms_modulation_mae_db": 0.4643077520009304, "300–1000 Hz_40ms_log_rms_mae_db": 1.0152656480631999, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12487208624646096, "300–1000 Hz_40ms_envelope_correlation": 0.9961903633919308, "300–1000 Hz_40ms_modulation_mae_db": 0.8408844517901061, "1000–3000 Hz_10ms_log_rms_mae_db": 0.9440755658449042, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15924183064730066, "1000–3000 Hz_10ms_envelope_correlation": 0.9872603143506243, "1000–3000 Hz_10ms_modulation_mae_db": 1.2297473529978484, "1000–3000 Hz_40ms_log_rms_mae_db": 0.725387857868538, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10696050938142931, "1000–3000 Hz_40ms_envelope_correlation": 0.9967156923427717, "1000–3000 Hz_40ms_modulation_mae_db": 1.7226932545111326, "3000–6000 Hz_10ms_log_rms_mae_db": 0.9999199902934749, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16434967263205666, "3000–6000 Hz_10ms_envelope_correlation": 0.9814103173269259, "3000–6000 Hz_10ms_modulation_mae_db": 0.9773933292802663, "3000–6000 Hz_40ms_log_rms_mae_db": 0.6952031299126661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.07755587834259646, "3000–6000 Hz_40ms_envelope_correlation": 0.9965085221001421, "3000–6000 Hz_40ms_modulation_mae_db": 1.5569929836839809, "mr_stft_db": 4.995319822394038, "band_mae_db": 0.5564233187611792, "rms_gain_error_db": 0.7342072962151174, "output_peak": 0.7857418710614744}, "full_scale": {"sample_count": 64720, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64720, "output_samples": 64720, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.691627034659898, "spectral_convergence": 0.556383360476704, "spectral_convergence_status": "ok", "amplitude_floor": 1.1807658352589429e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.4375358211178275, "spectral_convergence": 0.23551420708288687, "spectral_convergence_status": "ok", "amplitude_floor": 7.24953185483845e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.856796611404388, "spectral_convergence": 0.2831733482573517, "spectral_convergence_status": "ok", "amplitude_floor": 4.6531510695198173e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00031402968364977683, "output_power": 0.00033280775608264925, "delta_db": 0.25222737275729246, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005546801738700932, "output_power": 0.0006751061011562192, "delta_db": 0.8532938985407522, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010899709362407707, "output_power": 0.0013584895032309155, "delta_db": 0.9564136930422618, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008204698714335104, "output_power": 0.0009827597874958334, "delta_db": 0.7838473983149721, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0004732551323003661, "output_power": 0.0005193318232925066, "delta_db": 0.403496033916824, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00010423546715106559, "output_power": 0.00010640000998434757, "delta_db": 0.08926151599497334, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5564233187611792, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7342072962151174, "pitch": null, "output_rms_dbfs": -24.00674341138176, "output_peak": 0.7857418710614744, "reference_rms_dbfs": -24.740950707596877, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.995319822394038}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64720, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00017261805365043103, "log_rms_mae_db": 1.0605551084553402, "rms_envelope_nrmse": 0.14144590945775157, "envelope_correlation": 0.9898811036933927, "rms_gain_error_db": 0.7342073033408237, "reference_mean_power": 0.003360795516302892, "output_mean_power": 0.003979814403569756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.9828935159721717e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046042689591210996, "output_power": 0.0005717572024499781, "reference_active": true, "delta_db": 0.9405095905779073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011507848708673394, "output_power": 0.00011427462950816166, "reference_active": true, "delta_db": -0.030443217456692026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.734946402574847e-05, "output_power": 3.085536600553854e-05, "reference_active": true, "delta_db": -0.8295367158985927}], "active_band_count": 3, "active_band_mae_db": 0.6001631746443973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00013379439656886598, "log_rms_mae_db": 0.9748404868695941, "rms_envelope_nrmse": 0.10548854729298346, "envelope_correlation": 0.9970959195406944, "rms_gain_error_db": 0.7342073208144119, "reference_mean_power": 0.0033859385941668236, "output_mean_power": 0.004009588555899188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.884361456130501e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004231625539622664, "output_power": 0.0005205642778944982, "reference_active": true, "delta_db": 0.899671330714948}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.583961694519604e-05, "output_power": 5.307074407988329e-05, "reference_active": true, "delta_db": -0.2208725369355785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3554104634271355e-06, "output_power": 1.6269995090341651e-06, "reference_active": true, "delta_db": -3.143582326059018}], "active_band_count": 3, "active_band_mae_db": 1.4213753979031816}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00011786303332041996, "log_rms_mae_db": 1.1832586298520942, "rms_envelope_nrmse": 0.1494033508982216, "envelope_correlation": 0.991482066889388, "rms_gain_error_db": 0.8503127244543593, "reference_mean_power": 0.0013229704572187174, "output_mean_power": 0.0016090940119500464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 9.694412540428036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002357826301518165, "output_power": 0.00030522737132589043, "reference_active": true, "delta_db": 1.1211166853558785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.91091063998743e-05, "output_power": 6.200998409099559e-05, "reference_active": true, "delta_db": 0.20807226108420898}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.351021856959089e-05, "output_power": 1.3313398885241294e-05, "reference_active": true, "delta_db": -0.06373430956270375}], "active_band_count": 3, "active_band_mae_db": 0.4643077520009304}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010411406065660645, "log_rms_mae_db": 1.0152656480631999, "rms_envelope_nrmse": 0.12487208624646096, "envelope_correlation": 0.9961903633919308, "rms_gain_error_db": 0.8503127089077585, "reference_mean_power": 0.0013328679887454193, "output_mean_power": 0.0016211321122328647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 9.215200160073525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002182221456522984, "output_power": 0.0002796183706407188, "reference_active": true, "delta_db": 1.0766687908775199}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.729502442376393e-05, "output_power": 2.8547030841299186e-05, "reference_active": true, "delta_db": 0.19477456983560393}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1926400570958608e-06, "output_power": 8.941047300096997e-07, "reference_active": true, "delta_db": -1.2512099946571942}], "active_band_count": 3, "active_band_mae_db": 0.8408844517901061}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 0.00014037287410750346, "log_rms_mae_db": 0.9440755658449042, "rms_envelope_nrmse": 0.15924183064730066, "envelope_correlation": 0.9872603143506243, "rms_gain_error_db": 0.6916036501158734, "reference_mean_power": 0.001043588584426717, "output_mean_power": 0.0012237414903309264}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 7.131098310180307e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021758763787569653, "output_power": 0.00026278411805310045, "reference_active": true, "delta_db": 0.8196489660063422}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8027856042619025e-05, "output_power": 6.578219159235205e-05, "reference_active": true, "delta_db": 0.544718133290837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3022954991601597e-05, "output_power": 1.3479493341035137e-05, "reference_active": true, "delta_db": -2.3248749596963663}], "active_band_count": 3, "active_band_mae_db": 1.2297473529978484}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 0.00010015064346906524, "log_rms_mae_db": 0.725387857868538, "rms_envelope_nrmse": 0.10696050938142931, "envelope_correlation": 0.9967156923427717, "rms_gain_error_db": 0.6916036506924756, "reference_mean_power": 0.0010513959774729573, "output_mean_power": 0.001232896660400248}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 6.799409328431789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020213626575120905, "output_power": 0.00024059590025873924, "reference_active": true, "delta_db": 0.7564398439391171}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0065691390851235e-05, "output_power": 3.195580135869809e-05, "reference_active": true, "delta_db": 0.2647851753022357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1255469357025214e-06, "output_power": 8.180601263908953e-07, "reference_active": true, "delta_db": -4.146854744292045}], "active_band_count": 3, "active_band_mae_db": 1.7226932545111326}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 808, "rms_floor": 5.7697319934351994e-05, "log_rms_mae_db": 0.9999199902934749, "rms_envelope_nrmse": 0.16434967263205666, "envelope_correlation": 0.9814103173269259, "rms_gain_error_db": 0.3253379448080347, "reference_mean_power": 0.0001838054244420426, "output_mean_power": 0.00019810348633591326}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 808, "frequency_resolution_hz": 0.24752475247524752, "power_floor": 1.2570064040799192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.4129018402509514e-05, "output_power": 4.764672204751956e-05, "reference_active": true, "delta_db": 0.3330876064519394}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0833298446779821e-05, "output_power": 1.0400224920919997e-05, "reference_active": true, "delta_db": -0.17717976010648329}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.606324451113381e-06, "output_power": 2.6373226374035138e-06, "reference_active": true, "delta_db": -2.4219126212823765}], "active_band_count": 3, "active_band_mae_db": 0.9773933292802663}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64720, "uncovered_tail_samples": 0, "metrics": {"frame_count": 802, "rms_floor": 3.902816175045304e-05, "log_rms_mae_db": 0.6952031299126661, "rms_envelope_nrmse": 0.07755587834259646, "envelope_correlation": 0.9965085221001421, "rms_gain_error_db": 0.3253379460500577, "reference_mean_power": 0.00018518052639832005, "output_mean_power": 0.00019958555631792937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 802, "frequency_resolution_hz": 0.24937655860349128, "power_floor": 1.1869756093173816e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2060123324004214e-05, "output_power": 4.521043892410294e-05, "reference_active": true, "delta_db": 0.3136818164684573}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.372206568913873e-06, "output_power": 4.870831275520326e-06, "reference_active": true, "delta_db": -0.42549617581693006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0554113039123233e-07, "output_power": 1.640041401105103e-07, "reference_active": true, "delta_db": -3.931800958766555}], "active_band_count": 3, "active_band_mae_db": 1.5569929836839809}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9694277110724244, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1aa2881102172ce0f507739f5d169f90ad988c5452ea14a02376ceafe79bc10d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9872603143506243, "standard_deviation": null, "minimum": 0.9872603143506243, "maximum": 0.9872603143506243, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 0.9440755658449042, "standard_deviation": null, "minimum": 0.9440755658449042, "maximum": 0.9440755658449042, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2297473529978484, "standard_deviation": null, "minimum": 1.2297473529978484, "maximum": 1.2297473529978484, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15924183064730066, "standard_deviation": null, "minimum": 0.15924183064730066, "maximum": 0.15924183064730066, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9967156923427717, "standard_deviation": null, "minimum": 0.9967156923427717, "maximum": 0.9967156923427717, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.725387857868538, "standard_deviation": null, "minimum": 0.725387857868538, "maximum": 0.725387857868538, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.7226932545111326, "standard_deviation": null, "minimum": 1.7226932545111326, "maximum": 1.7226932545111326, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10696050938142931, "standard_deviation": null, "minimum": 0.10696050938142931, "maximum": 0.10696050938142931, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9814103173269259, "standard_deviation": null, "minimum": 0.9814103173269259, "maximum": 0.9814103173269259, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 0.9999199902934749, "standard_deviation": null, "minimum": 0.9999199902934749, "maximum": 0.9999199902934749, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9773933292802663, "standard_deviation": null, "minimum": 0.9773933292802663, "maximum": 0.9773933292802663, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16434967263205666, "standard_deviation": null, "minimum": 0.16434967263205666, "maximum": 0.16434967263205666, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9965085221001421, "standard_deviation": null, "minimum": 0.9965085221001421, "maximum": 0.9965085221001421, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.6952031299126661, "standard_deviation": null, "minimum": 0.6952031299126661, "maximum": 0.6952031299126661, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5569929836839809, "standard_deviation": null, "minimum": 1.5569929836839809, "maximum": 1.5569929836839809, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07755587834259646, "standard_deviation": null, "minimum": 0.07755587834259646, "maximum": 0.07755587834259646, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.991482066889388, "standard_deviation": null, "minimum": 0.991482066889388, "maximum": 0.991482066889388, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1832586298520942, "standard_deviation": null, "minimum": 1.1832586298520942, "maximum": 1.1832586298520942, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.4643077520009304, "standard_deviation": null, "minimum": 0.4643077520009304, "maximum": 0.4643077520009304, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1494033508982216, "standard_deviation": null, "minimum": 0.1494033508982216, "maximum": 0.1494033508982216, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961903633919308, "standard_deviation": null, "minimum": 0.9961903633919308, "maximum": 0.9961903633919308, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0152656480631999, "standard_deviation": null, "minimum": 1.0152656480631999, "maximum": 1.0152656480631999, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8408844517901061, "standard_deviation": null, "minimum": 0.8408844517901061, "maximum": 0.8408844517901061, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12487208624646096, "standard_deviation": null, "minimum": 0.12487208624646096, "maximum": 0.12487208624646096, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5564233187611792, "standard_deviation": null, "minimum": 0.5564233187611792, "maximum": 0.5564233187611792, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 4.995319822394038, "standard_deviation": null, "minimum": 4.995319822394038, "maximum": 4.995319822394038, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7857418710614744, "standard_deviation": null, "minimum": 0.7857418710614744, "maximum": 0.7857418710614744, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7342072962151174, "standard_deviation": null, "minimum": 0.7342072962151174, "maximum": 0.7342072962151174, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9898811036933927, "standard_deviation": null, "minimum": 0.9898811036933927, "maximum": 0.9898811036933927, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.0605551084553402, "standard_deviation": null, "minimum": 1.0605551084553402, "maximum": 1.0605551084553402, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6001631746443973, "standard_deviation": null, "minimum": 0.6001631746443973, "maximum": 0.6001631746443973, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14144590945775157, "standard_deviation": null, "minimum": 0.14144590945775157, "maximum": 0.14144590945775157, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9970959195406944, "standard_deviation": null, "minimum": 0.9970959195406944, "maximum": 0.9970959195406944, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9748404868695941, "standard_deviation": null, "minimum": 0.9748404868695941, "maximum": 0.9748404868695941, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4213753979031816, "standard_deviation": null, "minimum": 1.4213753979031816, "maximum": 1.4213753979031816, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10548854729298346, "standard_deviation": null, "minimum": 0.10548854729298346, "maximum": 0.10548854729298346, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_121"}, {"case": "librispeech_121-123852-0000", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.0239031314849854, "predicted_mos": 3.0239031314849854, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4add8619bb2809eabde4611b39e3bbdcd7c294b94ccea55f63a11e8d10afa085", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.748098500072956}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.1138944268204867, "全帯域_10ms_rms_envelope_nrmse": 0.1550989437634511, "全帯域_10ms_envelope_correlation": 0.9845773408245371, "全帯域_10ms_modulation_mae_db": 0.802698569559691, "全帯域_40ms_log_rms_mae_db": 0.7530668173984492, "全帯域_40ms_rms_envelope_nrmse": 0.10395258675394388, "全帯域_40ms_envelope_correlation": 0.9952793118535646, "全帯域_40ms_modulation_mae_db": 0.6404858931456028, "300–1000 Hz_10ms_log_rms_mae_db": 1.1483269659463122, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1397061436774599, "300–1000 Hz_10ms_envelope_correlation": 0.9925619802297997, "300–1000 Hz_10ms_modulation_mae_db": 0.8571078464275826, "300–1000 Hz_40ms_log_rms_mae_db": 0.8267109238613831, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11563945339990782, "300–1000 Hz_40ms_envelope_correlation": 0.9964486411956172, "300–1000 Hz_40ms_modulation_mae_db": 0.6544882318425763, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1019419289872834, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20362061447322755, "1000–3000 Hz_10ms_envelope_correlation": 0.9807153826022446, "1000–3000 Hz_10ms_modulation_mae_db": 0.6533992760793983, "1000–3000 Hz_40ms_log_rms_mae_db": 0.760950491271294, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14300044475122364, "1000–3000 Hz_40ms_envelope_correlation": 0.9946376757891301, "1000–3000 Hz_40ms_modulation_mae_db": 0.5554716427942176, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2381107807305898, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25505802718851395, "3000–6000 Hz_10ms_envelope_correlation": 0.9546396609416157, "3000–6000 Hz_10ms_modulation_mae_db": 0.992252937721947, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8851931220991701, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1524118446282344, "3000–6000 Hz_40ms_envelope_correlation": 0.983336715461545, "3000–6000 Hz_40ms_modulation_mae_db": 0.7871133840410979, "mr_stft_db": 4.473342332986152, "band_mae_db": 0.5847312021502086, "rms_gain_error_db": 0.6140951156325105, "output_peak": 0.7115662675182438}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.872344578381965, "spectral_convergence": 0.571920875206783, "spectral_convergence_status": "ok", "amplitude_floor": 1.0172154037614098e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.1518140555120455, "spectral_convergence": 0.3066094430156451, "spectral_convergence_status": "ok", "amplitude_floor": 7.045944280558953e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.395868365064446, "spectral_convergence": 0.32247758312735003, "spectral_convergence_status": "ok", "amplitude_floor": 6.386940858702166e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00026428484162924415, "output_power": 0.00025389983418775335, "delta_db": -0.17409837117850577, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008055137994793597, "output_power": 0.0009497294153446625, "delta_db": 0.7152690449189247, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006109316451161528, "output_power": 0.0007150497966080167, "delta_db": 0.6834366609382779, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027595006700553216, "output_power": 0.00035493086023603376, "delta_db": 1.0931325781247105, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003628265208503686, "output_power": 0.00041673770264906635, "delta_db": 0.6016376890369325, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001954243236049976, "output_power": 0.00020656650633371745, "delta_db": 0.24081286870389967, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5847312021502086, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6140951156325105, "pitch": null, "output_rms_dbfs": -25.38064380108097, "output_peak": 0.7115662675182438, "reference_rms_dbfs": -25.994738916713484, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.473342332986152}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017615675305240148, "log_rms_mae_db": 1.1138944268204867, "rms_envelope_nrmse": 0.1550989437634511, "envelope_correlation": 0.9845773408245371, "rms_gain_error_db": 0.6143257422230368, "reference_mean_power": 0.0025168568603362385, "output_mean_power": 0.0028992862181056456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0162973895242792e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002469993456433397, "output_power": 0.0002818019080378705, "reference_active": true, "delta_db": 0.5724812660748434}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010173002443957726, "output_power": 0.00013144269582566104, "reference_active": true, "delta_db": 1.1128730885413058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1084315450169838e-05, "output_power": 2.4901966276157542e-05, "reference_active": true, "delta_db": 0.7227413540629242}], "active_band_count": 3, "active_band_mae_db": 0.802698569559691}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016525538510559444, "log_rms_mae_db": 0.7530668173984492, "rms_envelope_nrmse": 0.10395258675394388, "envelope_correlation": 0.9952793118535646, "rms_gain_error_db": 0.6142181040019987, "reference_mean_power": 0.002527970677484516, "output_mean_power": 0.002912016574565256}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.625293642327984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023064746753792456, "output_power": 0.0002635517939872672, "reference_active": true, "delta_db": 0.5791728628166337}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.045652222183457e-05, "output_power": 7.685823713369185e-05, "reference_active": true, "delta_db": 1.0424725835550757}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2874374178901008e-06, "output_power": 2.450927384816352e-06, "reference_active": true, "delta_db": 0.299812233065099}], "active_band_count": 3, "active_band_mae_db": 0.6404858931456028}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001279910087199623, "log_rms_mae_db": 1.1483269659463122, "rms_envelope_nrmse": 0.1397061436774599, "envelope_correlation": 0.9925619802297997, "rms_gain_error_db": 0.7500137347712694, "reference_mean_power": 0.0009939686093891967, "output_mean_power": 0.0011813376422925767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.147046828130101e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013887130907316313, "output_power": 0.00016343973236689467, "reference_active": true, "delta_db": 0.7074511305895553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4048550736300918e-05, "output_power": 2.7116163556892286e-05, "reference_active": true, "delta_db": 0.5213933564198959}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.939409383024551e-06, "output_power": 6.728572366097428e-06, "reference_active": true, "delta_db": 1.3424790522732966}], "active_band_count": 3, "active_band_mae_db": 0.8571078464275826}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011859895721274409, "log_rms_mae_db": 0.8267109238613831, "rms_envelope_nrmse": 0.11563945339990782, "envelope_correlation": 0.9964486411956172, "rms_gain_error_db": 0.7499322274994391, "reference_mean_power": 0.0009986896907851761, "output_mean_power": 0.0011869263996541883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.009742136856967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012678804378904424, "output_power": 0.0001483125162938253, "reference_active": true, "delta_db": 0.6809950210381553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.298881565077983e-05, "output_power": 1.4757395399694682e-05, "reference_active": true, "delta_db": 0.5544016069067415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8039990598306815e-07, "output_power": 3.3157720809602415e-07, "reference_active": true, "delta_db": 0.7280680675828324}], "active_band_count": 3, "active_band_mae_db": 0.6544882318425763}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.31629258809413e-05, "log_rms_mae_db": 1.1019419289872834, "rms_envelope_nrmse": 0.20362061447322755, "envelope_correlation": 0.9807153826022446, "rms_gain_error_db": 0.9350131545419471, "reference_mean_power": 0.0004146029386445929, "output_mean_power": 0.0005142019170378203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.1875762004578613e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604229953230037e-05, "output_power": 0.0001064972725368351, "reference_active": true, "delta_db": 0.9262647670589247}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2897806527831664e-05, "output_power": 2.730812010497069e-05, "reference_active": true, "delta_db": 0.7649792273964111}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2516459896674564e-05, "output_power": 1.3316095329009129e-05, "reference_active": true, "delta_db": 0.2689538337828589}], "active_band_count": 3, "active_band_mae_db": 0.6533992760793983}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.786085823827924e-05, "log_rms_mae_db": 0.760950491271294, "rms_envelope_nrmse": 0.14300044475122364, "envelope_correlation": 0.9946376757891301, "rms_gain_error_db": 0.9350265473553123, "reference_mean_power": 0.0004166837249389974, "output_mean_power": 0.0005167841588369016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.035015228514449e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.632173919031572e-05, "output_power": 0.00010563451980475515, "reference_active": true, "delta_db": 0.8768568060273741}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4669499526204761e-05, "output_power": 1.7539992214306284e-05, "reference_active": true, "delta_db": 0.7761409881978165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2709360758910173e-06, "output_power": 1.2670156930972408e-06, "reference_active": true, "delta_db": -0.013417134157462209}], "active_band_count": 3, "active_band_mae_db": 0.5554716427942176}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011544071276927888, "log_rms_mae_db": 1.2381107807305898, "rms_envelope_nrmse": 0.25505802718851395, "envelope_correlation": 0.9546396609416157, "rms_gain_error_db": 0.34069010964535473, "reference_mean_power": 0.00033862428581899314, "output_mean_power": 0.0003662579945153683}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7742895295757852e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.9949103091747194e-05, "output_power": 7.535149088812815e-05, "reference_active": true, "delta_db": 0.9930915965804665}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5523159384577664e-05, "output_power": 5.908124689849793e-05, "reference_active": true, "delta_db": 1.132172572315918}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.055143212029728e-06, "output_power": 8.583319887461152e-06, "reference_active": true, "delta_db": 0.8514946442694566}], "active_band_count": 3, "active_band_mae_db": 0.992252937721947}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.669712818495073e-05, "log_rms_mae_db": 0.8851931220991701, "rms_envelope_nrmse": 0.1524118446282344, "envelope_correlation": 0.983336715461545, "rms_gain_error_db": 0.3407020180288606, "reference_mean_power": 0.00034032151520785135, "output_mean_power": 0.00036809473699791416}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6870888960608118e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.86684133644439e-05, "output_power": 8.825281618510181e-05, "reference_active": true, "delta_db": 1.0897155999549875}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.573388989930172e-05, "output_power": 3.2835244877269314e-05, "reference_active": true, "delta_db": 1.0583482097101813}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.935502106463895e-07, "output_power": 9.385266210618873e-07, "reference_active": true, "delta_db": 0.2132763424581249}], "active_band_count": 3, "active_band_mae_db": 0.7871133840410979}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9354514917024149, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "081730678e5433d301f04e58a103b2ffaf348a7c8101b750d8bc551835ea63be", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9807153826022446, "standard_deviation": null, "minimum": 0.9807153826022446, "maximum": 0.9807153826022446, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1019419289872834, "standard_deviation": null, "minimum": 1.1019419289872834, "maximum": 1.1019419289872834, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6533992760793983, "standard_deviation": null, "minimum": 0.6533992760793983, "maximum": 0.6533992760793983, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20362061447322755, "standard_deviation": null, "minimum": 0.20362061447322755, "maximum": 0.20362061447322755, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946376757891301, "standard_deviation": null, "minimum": 0.9946376757891301, "maximum": 0.9946376757891301, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.760950491271294, "standard_deviation": null, "minimum": 0.760950491271294, "maximum": 0.760950491271294, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5554716427942176, "standard_deviation": null, "minimum": 0.5554716427942176, "maximum": 0.5554716427942176, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14300044475122364, "standard_deviation": null, "minimum": 0.14300044475122364, "maximum": 0.14300044475122364, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9546396609416157, "standard_deviation": null, "minimum": 0.9546396609416157, "maximum": 0.9546396609416157, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2381107807305898, "standard_deviation": null, "minimum": 1.2381107807305898, "maximum": 1.2381107807305898, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.992252937721947, "standard_deviation": null, "minimum": 0.992252937721947, "maximum": 0.992252937721947, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25505802718851395, "standard_deviation": null, "minimum": 0.25505802718851395, "maximum": 0.25505802718851395, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.983336715461545, "standard_deviation": null, "minimum": 0.983336715461545, "maximum": 0.983336715461545, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8851931220991701, "standard_deviation": null, "minimum": 0.8851931220991701, "maximum": 0.8851931220991701, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7871133840410979, "standard_deviation": null, "minimum": 0.7871133840410979, "maximum": 0.7871133840410979, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1524118446282344, "standard_deviation": null, "minimum": 0.1524118446282344, "maximum": 0.1524118446282344, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9925619802297997, "standard_deviation": null, "minimum": 0.9925619802297997, "maximum": 0.9925619802297997, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.1483269659463122, "standard_deviation": null, "minimum": 1.1483269659463122, "maximum": 1.1483269659463122, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8571078464275826, "standard_deviation": null, "minimum": 0.8571078464275826, "maximum": 0.8571078464275826, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1397061436774599, "standard_deviation": null, "minimum": 0.1397061436774599, "maximum": 0.1397061436774599, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9964486411956172, "standard_deviation": null, "minimum": 0.9964486411956172, "maximum": 0.9964486411956172, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.8267109238613831, "standard_deviation": null, "minimum": 0.8267109238613831, "maximum": 0.8267109238613831, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6544882318425763, "standard_deviation": null, "minimum": 0.6544882318425763, "maximum": 0.6544882318425763, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11563945339990782, "standard_deviation": null, "minimum": 0.11563945339990782, "maximum": 0.11563945339990782, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5847312021502086, "standard_deviation": null, "minimum": 0.5847312021502086, "maximum": 0.5847312021502086, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 4.473342332986152, "standard_deviation": null, "minimum": 4.473342332986152, "maximum": 4.473342332986152, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7115662675182438, "standard_deviation": null, "minimum": 0.7115662675182438, "maximum": 0.7115662675182438, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6140951156325105, "standard_deviation": null, "minimum": 0.6140951156325105, "maximum": 0.6140951156325105, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9845773408245371, "standard_deviation": null, "minimum": 0.9845773408245371, "maximum": 0.9845773408245371, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.1138944268204867, "standard_deviation": null, "minimum": 1.1138944268204867, "maximum": 1.1138944268204867, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.802698569559691, "standard_deviation": null, "minimum": 0.802698569559691, "maximum": 0.802698569559691, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1550989437634511, "standard_deviation": null, "minimum": 0.1550989437634511, "maximum": 0.1550989437634511, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9952793118535646, "standard_deviation": null, "minimum": 0.9952793118535646, "maximum": 0.9952793118535646, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.7530668173984492, "standard_deviation": null, "minimum": 0.7530668173984492, "maximum": 0.7530668173984492, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6404858931456028, "standard_deviation": null, "minimum": 0.6404858931456028, "maximum": 0.6404858931456028, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10395258675394388, "standard_deviation": null, "minimum": 0.10395258675394388, "maximum": 0.10395258675394388, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_121"}, {"case": "librispeech_121-127105-0035", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.6218395233154297, "predicted_mos": 2.6218395233154297, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "96dec9e445f67804f9e666ee72d947d9a1c439efc952d936f4f4a5c1cc9ec955", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.167366699781269}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.153904136084599, "全帯域_10ms_rms_envelope_nrmse": 0.17273765842225097, "全帯域_10ms_envelope_correlation": 0.9841883203736189, "全帯域_10ms_modulation_mae_db": 0.9291085696927796, "全帯域_40ms_log_rms_mae_db": 0.8749158096696936, "全帯域_40ms_rms_envelope_nrmse": 0.1304414100389552, "全帯域_40ms_envelope_correlation": 0.9946094714869326, "全帯域_40ms_modulation_mae_db": 0.8274554521349144, "300–1000 Hz_10ms_log_rms_mae_db": 1.3870459688498893, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2077281298202589, "300–1000 Hz_10ms_envelope_correlation": 0.9890584485225797, "300–1000 Hz_10ms_modulation_mae_db": 1.404957850488002, "300–1000 Hz_40ms_log_rms_mae_db": 0.9954891948268215, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17669570982225843, "300–1000 Hz_40ms_envelope_correlation": 0.9962083263713774, "300–1000 Hz_40ms_modulation_mae_db": 1.322567081290651, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2286418193474475, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21146573683909578, "1000–3000 Hz_10ms_envelope_correlation": 0.9791524780033266, "1000–3000 Hz_10ms_modulation_mae_db": 0.7619248218324365, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8823370622667146, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15352134196280343, "1000–3000 Hz_40ms_envelope_correlation": 0.992633016851384, "1000–3000 Hz_40ms_modulation_mae_db": 0.774993759543503, "3000–6000 Hz_10ms_log_rms_mae_db": 1.273334696197599, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2273662184643886, "3000–6000 Hz_10ms_envelope_correlation": 0.9757511188776239, "3000–6000 Hz_10ms_modulation_mae_db": 0.4357426787503629, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9754675090761915, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15832497026675174, "3000–6000 Hz_40ms_envelope_correlation": 0.9899090379862084, "3000–6000 Hz_40ms_modulation_mae_db": 0.4321131347105074, "mr_stft_db": 4.58368322662508, "band_mae_db": 0.8423661742702553, "rms_gain_error_db": 0.8721977596570676, "output_peak": 0.6858167802729918}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 4.939450505975367, "spectral_convergence": 0.5907747567803078, "spectral_convergence_status": "ok", "amplitude_floor": 1.0971770400792783e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.274716887837492, "spectral_convergence": 0.3863547529460689, "spectral_convergence_status": "ok", "amplitude_floor": 6.752711905169626e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.536882286062382, "spectral_convergence": 0.41405278369087517, "spectral_convergence_status": "ok", "amplitude_floor": 5.858543156261471e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003996059347809583, "output_power": 0.0004798014962190804, "delta_db": 0.7942966838069434, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00030506165058519967, "output_power": 0.0003925345185375614, "delta_db": 1.09490237817359, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002850207180759584, "output_power": 0.000374149528130254, "delta_db": 1.1816877193620547, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017376514501848305, "output_power": 0.000210754822974387, "delta_db": 0.8381485479096167, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002541837569284722, "output_power": 0.00031705141130827287, "delta_db": 0.9598189628509186, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023122095798090296, "output_power": 0.00024130230560833997, "delta_db": 0.18534275351840765, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8423661742702553, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8721977596570676, "pitch": null, "output_rms_dbfs": -26.955969252818775, "output_peak": 0.6858167802729918, "reference_rms_dbfs": -27.828167012475838, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.58368322662508}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015167567220720444, "log_rms_mae_db": 1.153904136084599, "rms_envelope_nrmse": 0.17273765842225097, "envelope_correlation": 0.9841883203736189, "rms_gain_error_db": 0.8721978019799024, "reference_mean_power": 0.0016502333338478435, "output_mean_power": 0.0020172751357356612}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103350035549285e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021122178949431673, "output_power": 0.0002803094776021898, "reference_active": true, "delta_db": 1.2289906433741316}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018786794884032712, "output_power": 0.0002230636215185554, "reference_active": true, "delta_db": 0.7457605512000909}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.617925051494854e-05, "output_power": 3.1565644453513294e-05, "reference_active": true, "delta_db": 0.8125745145041162}], "active_band_count": 3, "active_band_mae_db": 0.9291085696927796}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014379702931337134, "log_rms_mae_db": 0.8749158096696936, "rms_envelope_nrmse": 0.1304414100389552, "envelope_correlation": 0.9946094714869326, "rms_gain_error_db": 0.8721979987487802, "reference_mean_power": 0.0016585326982667058, "output_mean_power": 0.0020274205210747795}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0955155560805602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019170493341265238, "output_power": 0.0002580123443290116, "reference_active": true, "delta_db": 1.2900719549008353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.87936190430084e-05, "output_power": 9.169138775642856e-05, "reference_active": true, "delta_db": 0.6583749749073728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1061472825588806e-06, "output_power": 2.3816650341422914e-06, "reference_active": true, "delta_db": 0.5339194265965354}], "active_band_count": 3, "active_band_mae_db": 0.8274554521349144}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010769717002798511, "log_rms_mae_db": 1.3870459688498893, "rms_envelope_nrmse": 0.2077281298202589, "envelope_correlation": 0.9890584485225797, "rms_gain_error_db": 1.2801765534020835, "reference_mean_power": 0.0005226470255969395, "output_mean_power": 0.0007018206434952699}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.88003979644585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021484485335499559, "output_power": 0.00029588650251100105, "reference_active": true, "delta_db": 1.390001998042446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0907744988019035e-05, "output_power": 5.884820096156539e-05, "reference_active": true, "delta_db": 1.5792765067254877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.643789476493196e-06, "output_power": 1.1514993192876929e-05, "reference_active": true, "delta_db": 1.2455950466960726}], "active_band_count": 3, "active_band_mae_db": 1.404957850488002}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.974444405364767e-05, "log_rms_mae_db": 0.9954891948268215, "rms_envelope_nrmse": 0.17669570982225843, "envelope_correlation": 0.9962083263713774, "rms_gain_error_db": 1.2801766682609559, "reference_mean_power": 0.0005252755682472565, "output_mean_power": 0.0007053503206075534}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.5461080823927677e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019511319942898067, "output_power": 0.000270261376752602, "reference_active": true, "delta_db": 1.4149733434009724}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8168801206395614e-05, "output_power": 2.6060054657056494e-05, "reference_active": true, "delta_db": 1.5664904912650681}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.358572422696516e-07, "output_power": 9.23458343357622e-07, "reference_active": true, "delta_db": 0.9862374092059125}], "active_band_count": 3, "active_band_mae_db": 1.322567081290651}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.224188331064728e-05, "log_rms_mae_db": 1.2286418193474475, "rms_envelope_nrmse": 0.21146573683909578, "envelope_correlation": 0.9791524780033266, "rms_gain_error_db": 0.944925534428453, "reference_mean_power": 0.00028515402410464236, "output_mean_power": 0.000354463938097407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.768616554938329e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001096694429350374, "output_power": 0.00014337724264144226, "reference_active": true, "delta_db": 1.1639458651816272}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.898846044442003e-05, "output_power": 6.340991481102156e-05, "reference_active": true, "delta_db": 1.120633785877089}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.522948535574358e-06, "output_power": 8.525293659291945e-06, "reference_active": true, "delta_db": 0.0011948144385933574}], "active_band_count": 3, "active_band_mae_db": 0.7619248218324365}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.641451530936745e-05, "log_rms_mae_db": 0.8823370622667146, "rms_envelope_nrmse": 0.15352134196280343, "envelope_correlation": 0.992633016851384, "rms_gain_error_db": 0.9449256610026695, "reference_mean_power": 0.0002865881294452572, "output_mean_power": 0.0003562466293603311}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.5110883907146213e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010245851234810692, "output_power": 0.00013462432502217846, "reference_active": true, "delta_db": 1.185754930270874}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5140429681693116e-05, "output_power": 3.2563333957209596e-05, "reference_active": true, "delta_db": 1.1235616714956684}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.446669785409335e-07, "output_power": 7.419858586986544e-07, "reference_active": true, "delta_db": -0.015664676863966637}], "active_band_count": 3, "active_band_mae_db": 0.774993759543503}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001420862345348913, "log_rms_mae_db": 1.273334696197599, "rms_envelope_nrmse": 0.2273662184643886, "envelope_correlation": 0.9757511188776239, "rms_gain_error_db": 0.7438371547374735, "reference_mean_power": 0.00023492613523189528, "output_mean_power": 0.00027881430339201047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4459776587443134e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000158142036380199, "output_power": 0.00018647870351564115, "reference_active": true, "delta_db": 0.7158191424431011}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.639730090455687e-05, "output_power": 0.00010624857133145536, "reference_active": true, "delta_db": 0.42258225074443745}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.267646990584481e-05, "output_power": 1.3178955575727077e-05, "reference_active": true, "delta_db": 0.16882664306355005}], "active_band_count": 3, "active_band_mae_db": 0.4357426787503629}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010454414213188805, "log_rms_mae_db": 0.9754675090761915, "rms_envelope_nrmse": 0.15832497026675174, "envelope_correlation": 0.9899090379862084, "rms_gain_error_db": 0.7438371601230375, "reference_mean_power": 0.00023610764979113905, "output_mean_power": 0.00028021654516511955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.2342047370765003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017362277943143474, "output_power": 0.00020252353288311597, "reference_active": true, "delta_db": 0.6686879042626508}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.8688155273395004e-05, "output_power": 5.422825848887582e-05, "reference_active": true, "delta_db": 0.46802337484883905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.341220903531162e-06, "output_power": 1.2928183298196338e-06, "reference_active": true, "delta_db": -0.15962812502003215}], "active_band_count": 3, "active_band_mae_db": 0.4321131347105074}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9361802812777776, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f89dee450855a271d564b51200fc66e30d30da6c8f8abf8812267ceba12ea840", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9791524780033266, "standard_deviation": null, "minimum": 0.9791524780033266, "maximum": 0.9791524780033266, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2286418193474475, "standard_deviation": null, "minimum": 1.2286418193474475, "maximum": 1.2286418193474475, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7619248218324365, "standard_deviation": null, "minimum": 0.7619248218324365, "maximum": 0.7619248218324365, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21146573683909578, "standard_deviation": null, "minimum": 0.21146573683909578, "maximum": 0.21146573683909578, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.992633016851384, "standard_deviation": null, "minimum": 0.992633016851384, "maximum": 0.992633016851384, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8823370622667146, "standard_deviation": null, "minimum": 0.8823370622667146, "maximum": 0.8823370622667146, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.774993759543503, "standard_deviation": null, "minimum": 0.774993759543503, "maximum": 0.774993759543503, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15352134196280343, "standard_deviation": null, "minimum": 0.15352134196280343, "maximum": 0.15352134196280343, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9757511188776239, "standard_deviation": null, "minimum": 0.9757511188776239, "maximum": 0.9757511188776239, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.273334696197599, "standard_deviation": null, "minimum": 1.273334696197599, "maximum": 1.273334696197599, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4357426787503629, "standard_deviation": null, "minimum": 0.4357426787503629, "maximum": 0.4357426787503629, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2273662184643886, "standard_deviation": null, "minimum": 0.2273662184643886, "maximum": 0.2273662184643886, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9899090379862084, "standard_deviation": null, "minimum": 0.9899090379862084, "maximum": 0.9899090379862084, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9754675090761915, "standard_deviation": null, "minimum": 0.9754675090761915, "maximum": 0.9754675090761915, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4321131347105074, "standard_deviation": null, "minimum": 0.4321131347105074, "maximum": 0.4321131347105074, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15832497026675174, "standard_deviation": null, "minimum": 0.15832497026675174, "maximum": 0.15832497026675174, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9890584485225797, "standard_deviation": null, "minimum": 0.9890584485225797, "maximum": 0.9890584485225797, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3870459688498893, "standard_deviation": null, "minimum": 1.3870459688498893, "maximum": 1.3870459688498893, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.404957850488002, "standard_deviation": null, "minimum": 1.404957850488002, "maximum": 1.404957850488002, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2077281298202589, "standard_deviation": null, "minimum": 0.2077281298202589, "maximum": 0.2077281298202589, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9962083263713774, "standard_deviation": null, "minimum": 0.9962083263713774, "maximum": 0.9962083263713774, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 0.9954891948268215, "standard_deviation": null, "minimum": 0.9954891948268215, "maximum": 0.9954891948268215, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.322567081290651, "standard_deviation": null, "minimum": 1.322567081290651, "maximum": 1.322567081290651, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17669570982225843, "standard_deviation": null, "minimum": 0.17669570982225843, "maximum": 0.17669570982225843, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8423661742702553, "standard_deviation": null, "minimum": 0.8423661742702553, "maximum": 0.8423661742702553, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 4.58368322662508, "standard_deviation": null, "minimum": 4.58368322662508, "maximum": 4.58368322662508, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6858167802729918, "standard_deviation": null, "minimum": 0.6858167802729918, "maximum": 0.6858167802729918, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8721977596570676, "standard_deviation": null, "minimum": 0.8721977596570676, "maximum": 0.8721977596570676, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9841883203736189, "standard_deviation": null, "minimum": 0.9841883203736189, "maximum": 0.9841883203736189, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.153904136084599, "standard_deviation": null, "minimum": 1.153904136084599, "maximum": 1.153904136084599, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9291085696927796, "standard_deviation": null, "minimum": 0.9291085696927796, "maximum": 0.9291085696927796, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17273765842225097, "standard_deviation": null, "minimum": 0.17273765842225097, "maximum": 0.17273765842225097, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9946094714869326, "standard_deviation": null, "minimum": 0.9946094714869326, "maximum": 0.9946094714869326, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.8749158096696936, "standard_deviation": null, "minimum": 0.8749158096696936, "maximum": 0.8749158096696936, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8274554521349144, "standard_deviation": null, "minimum": 0.8274554521349144, "maximum": 0.8274554521349144, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1304414100389552, "standard_deviation": null, "minimum": 0.1304414100389552, "maximum": 0.1304414100389552, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_121"}, {"case": "librispeech_1221-135766-0005", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.164355516433716, "predicted_mos": 3.164355516433716, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ec29312f5003cf9621b20e30b3b0f7d9e9fc74da0cd79bec496f077c6f0947a7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.96758639998734}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6911538344295032, "全帯域_10ms_rms_envelope_nrmse": 0.13743129240551902, "全帯域_10ms_envelope_correlation": 0.9884645971922038, "全帯域_10ms_modulation_mae_db": 0.9548454370385668, "全帯域_40ms_log_rms_mae_db": 1.3472680278740938, "全帯域_40ms_rms_envelope_nrmse": 0.11569521501042587, "全帯域_40ms_envelope_correlation": 0.9939325768272796, "全帯域_40ms_modulation_mae_db": 1.1281560454861628, "300–1000 Hz_10ms_log_rms_mae_db": 1.6822656941216447, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17645387846859353, "300–1000 Hz_10ms_envelope_correlation": 0.9897235518462959, "300–1000 Hz_10ms_modulation_mae_db": 0.826346777505761, "300–1000 Hz_40ms_log_rms_mae_db": 1.3615228775627004, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.16434480274482463, "300–1000 Hz_40ms_envelope_correlation": 0.9922728591865132, "300–1000 Hz_40ms_modulation_mae_db": 0.7817312446664378, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2618541059205066, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13876009080078283, "1000–3000 Hz_10ms_envelope_correlation": 0.9927995860164593, "1000–3000 Hz_10ms_modulation_mae_db": 1.1147807873450752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9065954143844566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1170689361455498, "1000–3000 Hz_40ms_envelope_correlation": 0.996602318181622, "1000–3000 Hz_40ms_modulation_mae_db": 0.9912231754355324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.321761225095384, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.184241777427222, "3000–6000 Hz_10ms_envelope_correlation": 0.9765042807021941, "3000–6000 Hz_10ms_modulation_mae_db": 0.7933291760019893, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8326836735605584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12110558992181542, "3000–6000 Hz_40ms_envelope_correlation": 0.9904451341802596, "3000–6000 Hz_40ms_modulation_mae_db": 0.7256788305884149, "mr_stft_db": 5.773151806982241, "band_mae_db": 0.6708582759190866, "rms_gain_error_db": 0.7514628279703164, "output_peak": 0.48125803049707666}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.103461928132851, "spectral_convergence": 0.5642011379716023, "spectral_convergence_status": "ok", "amplitude_floor": 6.643709912602353e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.353493336019013, "spectral_convergence": 0.30700064822576995, "spectral_convergence_status": "ok", "amplitude_floor": 5.274693968404162e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.862500156794859, "spectral_convergence": 0.3331600190867243, "spectral_convergence_status": "ok", "amplitude_floor": 3.8895875894681354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00010232532114960487, "output_power": 0.00011420927972446494, "delta_db": 0.47718276249167907, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003592345857428397, "output_power": 0.0004050770596205796, "delta_db": 0.5215950662253, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003905970862424533, "output_power": 0.0005056300106384115, "delta_db": 1.1210384346823956, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022506467673995083, "output_power": 0.00026817510994418, "delta_db": 0.7611112827616856, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011237312839533447, "output_power": 0.00013460025887175112, "delta_db": 0.7838342364127959, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00013068329846048222, "output_power": 0.00014199036476786458, "delta_db": 0.36038787294066393, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6708582759190866, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7514628279703164, "pitch": null, "output_rms_dbfs": -28.04188298758589, "output_peak": 0.48125803049707666, "reference_rms_dbfs": -28.793345815556208, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.773151806982241}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001036839322212551, "log_rms_mae_db": 1.6911538344295032, "rms_envelope_nrmse": 0.13743129240551902, "envelope_correlation": 0.9884645971922038, "rms_gain_error_db": 0.7514992107230019, "reference_mean_power": 0.0013213586135692834, "output_mean_power": 0.0015709798735957276}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.530367535346178e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002613603148630271, "output_power": 0.00031164850282220226, "reference_active": true, "delta_db": 0.7642540019070356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.246118337274924e-05, "output_power": 0.00010328566505541381, "reference_active": true, "delta_db": 0.9779048735295358}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.028948119628933e-05, "output_power": 1.3323895609696288e-05, "reference_active": true, "delta_db": 1.1223774356791292}], "active_band_count": 3, "active_band_mae_db": 0.9548454370385668}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.673473491549852e-05, "log_rms_mae_db": 1.3472680278740938, "rms_envelope_nrmse": 0.11569521501042587, "envelope_correlation": 0.9939325768272796, "rms_gain_error_db": 0.7516891220849093, "reference_mean_power": 0.0013278763848253465, "output_mean_power": 0.0015787979710966725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.825313121587468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002333069672738863, "output_power": 0.0002794382410610909, "reference_active": true, "delta_db": 0.783581305729195}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6274669036315644e-05, "output_power": 5.835554917049098e-05, "reference_active": true, "delta_db": 1.0073883916524344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.136183723659665e-07, "output_power": 8.856215958756167e-07, "reference_active": true, "delta_db": 1.5934984390768592}], "active_band_count": 3, "active_band_mae_db": 1.1281560454861628}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.601318795186366e-05, "log_rms_mae_db": 1.6822656941216447, "rms_envelope_nrmse": 0.17645387846859353, "envelope_correlation": 0.9897235518462959, "rms_gain_error_db": 1.0848151188286366, "reference_mean_power": 0.0005515675704872272, "output_mean_power": 0.0007080765891423346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3413004486315466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020691279435977171, "output_power": 0.0002540909586477613, "reference_active": true, "delta_db": 0.8920186586682893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313846735404118e-05, "output_power": 4.4515564345949146e-05, "reference_active": true, "delta_db": 1.2817946497931167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.722807599466502e-06, "output_power": 2.5380153924759267e-06, "reference_active": true, "delta_db": -0.30522702405587665}], "active_band_count": 3, "active_band_mae_db": 0.826346777505761}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.977903967560393e-05, "log_rms_mae_db": 1.3615228775627004, "rms_envelope_nrmse": 0.16434480274482463, "envelope_correlation": 0.9922728591865132, "rms_gain_error_db": 1.0848303594733364, "reference_mean_power": 0.0005543343450028261, "output_mean_power": 0.0007116309418708928}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0568367997569596e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001948094737446988, "output_power": 0.0002398288134572117, "reference_active": true, "delta_db": 0.9029128564997552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.636770712067895e-05, "output_power": 2.1891181153283408e-05, "reference_active": true, "delta_db": 1.2628134955891812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1494053604425814e-07, "output_power": 2.240087929132291e-07, "reference_active": true, "delta_db": 0.17946738191037678}], "active_band_count": 3, "active_band_mae_db": 0.7817312446664378}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.251077329795781e-05, "log_rms_mae_db": 1.2618541059205066, "rms_envelope_nrmse": 0.13876009080078283, "envelope_correlation": 0.9927995860164593, "rms_gain_error_db": 0.7996271464492535, "reference_mean_power": 0.0002789441343631021, "output_mean_power": 0.00033533582126744026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.041685297045403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011680629747545264, "output_power": 0.000136739565345666, "reference_active": true, "delta_db": 0.684279371848242}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1432581274087105e-05, "output_power": 1.4919882402252613e-05, "reference_active": true, "delta_db": 1.1562110260521576}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6649698962790864e-06, "output_power": 2.353919366492072e-06, "reference_active": true, "delta_db": 1.503851964134826}], "active_band_count": 3, "active_band_mae_db": 1.1147807873450752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.285244854844391e-05, "log_rms_mae_db": 0.9065954143844566, "rms_envelope_nrmse": 0.1170689361455498, "envelope_correlation": 0.996602318181622, "rms_gain_error_db": 0.799832179240103, "reference_mean_power": 0.00028031132153199156, "output_mean_power": 0.00033699531002089674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9116341431814158e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010971300365261617, "output_power": 0.00012952171841782944, "reference_active": true, "delta_db": 0.7208449268907773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3756201199309846e-06, "output_power": 6.913560205949184e-06, "reference_active": true, "delta_db": 1.0927317844922628}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.905227736918092e-08, "output_power": 1.2938196615034535e-07, "reference_active": true, "delta_db": 1.160092814923557}], "active_band_count": 3, "active_band_mae_db": 0.9912231754355324}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.076518340019586e-05, "log_rms_mae_db": 1.321761225095384, "rms_envelope_nrmse": 0.184241777427222, "envelope_correlation": 0.9765042807021941, "rms_gain_error_db": 0.49799921082627574, "reference_mean_power": 9.153664281853568e-05, "output_mean_power": 0.00010265849695184826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.939589252350226e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.46860955530461e-05, "output_power": 4.001885381827677e-05, "reference_active": true, "delta_db": 0.6210922967905822}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1944580647265704e-05, "output_power": 2.572748062602706e-05, "reference_active": true, "delta_db": 0.6906997377421147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8366553780699585e-06, "output_power": 6.185355897527822e-06, "reference_active": true, "delta_db": 1.0681954934732707}], "active_band_count": 3, "active_band_mae_db": 0.7933291760019893}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4119030940985276e-05, "log_rms_mae_db": 0.8326836735605584, "rms_envelope_nrmse": 0.12110558992181542, "envelope_correlation": 0.9904451341802596, "rms_gain_error_db": 0.49815738324443604, "reference_mean_power": 9.196260234115854e-05, "output_mean_power": 0.00010313996760938464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.164079440499503e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.035826325529754e-05, "output_power": 4.6674475363750676e-05, "reference_active": true, "delta_db": 0.631469761464465}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2217507839367818e-05, "output_power": 1.4691357422953517e-05, "reference_active": true, "delta_db": 0.8007929836028306}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.5236785056789705e-07, "output_power": 5.369935966968099e-07, "reference_active": true, "delta_db": 0.7447737466979489}], "active_band_count": 3, "active_band_mae_db": 0.7256788305884149}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8858420285414762, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "52a2672fae409e9cab2ecf5f28d336367f1421815c857367e028d0d75ff31e3f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9927995860164593, "standard_deviation": null, "minimum": 0.9927995860164593, "maximum": 0.9927995860164593, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2618541059205066, "standard_deviation": null, "minimum": 1.2618541059205066, "maximum": 1.2618541059205066, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1147807873450752, "standard_deviation": null, "minimum": 1.1147807873450752, "maximum": 1.1147807873450752, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13876009080078283, "standard_deviation": null, "minimum": 0.13876009080078283, "maximum": 0.13876009080078283, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996602318181622, "standard_deviation": null, "minimum": 0.996602318181622, "maximum": 0.996602318181622, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9065954143844566, "standard_deviation": null, "minimum": 0.9065954143844566, "maximum": 0.9065954143844566, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9912231754355324, "standard_deviation": null, "minimum": 0.9912231754355324, "maximum": 0.9912231754355324, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1170689361455498, "standard_deviation": null, "minimum": 0.1170689361455498, "maximum": 0.1170689361455498, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9765042807021941, "standard_deviation": null, "minimum": 0.9765042807021941, "maximum": 0.9765042807021941, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.321761225095384, "standard_deviation": null, "minimum": 1.321761225095384, "maximum": 1.321761225095384, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7933291760019893, "standard_deviation": null, "minimum": 0.7933291760019893, "maximum": 0.7933291760019893, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.184241777427222, "standard_deviation": null, "minimum": 0.184241777427222, "maximum": 0.184241777427222, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9904451341802596, "standard_deviation": null, "minimum": 0.9904451341802596, "maximum": 0.9904451341802596, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8326836735605584, "standard_deviation": null, "minimum": 0.8326836735605584, "maximum": 0.8326836735605584, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7256788305884149, "standard_deviation": null, "minimum": 0.7256788305884149, "maximum": 0.7256788305884149, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12110558992181542, "standard_deviation": null, "minimum": 0.12110558992181542, "maximum": 0.12110558992181542, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9897235518462959, "standard_deviation": null, "minimum": 0.9897235518462959, "maximum": 0.9897235518462959, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6822656941216447, "standard_deviation": null, "minimum": 1.6822656941216447, "maximum": 1.6822656941216447, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.826346777505761, "standard_deviation": null, "minimum": 0.826346777505761, "maximum": 0.826346777505761, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17645387846859353, "standard_deviation": null, "minimum": 0.17645387846859353, "maximum": 0.17645387846859353, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9922728591865132, "standard_deviation": null, "minimum": 0.9922728591865132, "maximum": 0.9922728591865132, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3615228775627004, "standard_deviation": null, "minimum": 1.3615228775627004, "maximum": 1.3615228775627004, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7817312446664378, "standard_deviation": null, "minimum": 0.7817312446664378, "maximum": 0.7817312446664378, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16434480274482463, "standard_deviation": null, "minimum": 0.16434480274482463, "maximum": 0.16434480274482463, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6708582759190866, "standard_deviation": null, "minimum": 0.6708582759190866, "maximum": 0.6708582759190866, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.773151806982241, "standard_deviation": null, "minimum": 5.773151806982241, "maximum": 5.773151806982241, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.48125803049707666, "standard_deviation": null, "minimum": 0.48125803049707666, "maximum": 0.48125803049707666, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7514628279703164, "standard_deviation": null, "minimum": 0.7514628279703164, "maximum": 0.7514628279703164, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9884645971922038, "standard_deviation": null, "minimum": 0.9884645971922038, "maximum": 0.9884645971922038, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6911538344295032, "standard_deviation": null, "minimum": 1.6911538344295032, "maximum": 1.6911538344295032, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9548454370385668, "standard_deviation": null, "minimum": 0.9548454370385668, "maximum": 0.9548454370385668, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.13743129240551902, "standard_deviation": null, "minimum": 0.13743129240551902, "maximum": 0.13743129240551902, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9939325768272796, "standard_deviation": null, "minimum": 0.9939325768272796, "maximum": 0.9939325768272796, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.3472680278740938, "standard_deviation": null, "minimum": 1.3472680278740938, "maximum": 1.3472680278740938, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1281560454861628, "standard_deviation": null, "minimum": 1.1281560454861628, "maximum": 1.1281560454861628, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11569521501042587, "standard_deviation": null, "minimum": 0.11569521501042587, "maximum": 0.11569521501042587, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1221"}, {"case": "librispeech_1221-135767-0006", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.1080844402313232, "predicted_mos": 3.1080844402313232, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "046332ed57a7c0afd4129e6ec4106f0ebf5daef015b1481e4f7b368ec6f84b75", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.8549290001392365}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7565922267652838, "全帯域_10ms_rms_envelope_nrmse": 0.15482796979253896, "全帯域_10ms_envelope_correlation": 0.9807812720376126, "全帯域_10ms_modulation_mae_db": 1.2672530906572501, "全帯域_40ms_log_rms_mae_db": 1.4689918624265184, "全帯域_40ms_rms_envelope_nrmse": 0.11881910348528579, "全帯域_40ms_envelope_correlation": 0.9918677671561223, "全帯域_40ms_modulation_mae_db": 1.2022686016929516, "300–1000 Hz_10ms_log_rms_mae_db": 1.8847681574666544, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19692179157951403, "300–1000 Hz_10ms_envelope_correlation": 0.9856212596104938, "300–1000 Hz_10ms_modulation_mae_db": 1.496766896547494, "300–1000 Hz_40ms_log_rms_mae_db": 1.4796252515666641, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1761988693771075, "300–1000 Hz_40ms_envelope_correlation": 0.9907827678436709, "300–1000 Hz_40ms_modulation_mae_db": 1.3024317426878447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2482720343444837, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1342235497132654, "1000–3000 Hz_10ms_envelope_correlation": 0.991877584505586, "1000–3000 Hz_10ms_modulation_mae_db": 0.6614989732370667, "1000–3000 Hz_40ms_log_rms_mae_db": 0.864143105461925, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10079348384612877, "1000–3000 Hz_40ms_envelope_correlation": 0.9973474245273052, "1000–3000 Hz_40ms_modulation_mae_db": 0.6953786169547321, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3227490773420425, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22258639160804375, "3000–6000 Hz_10ms_envelope_correlation": 0.9679180575264781, "3000–6000 Hz_10ms_modulation_mae_db": 0.3504230155117816, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7710911383961001, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12186581966527374, "3000–6000 Hz_40ms_envelope_correlation": 0.9903265911619179, "3000–6000 Hz_40ms_modulation_mae_db": 0.17267931707286102, "mr_stft_db": 5.856820326547596, "band_mae_db": 0.6425004547189024, "rms_gain_error_db": 0.7313949649742557, "output_peak": 0.3816178104408559}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2770944505367865, "spectral_convergence": 0.6158987016346089, "spectral_convergence_status": "ok", "amplitude_floor": 6.641091871383268e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.412007753592853, "spectral_convergence": 0.3713468372588585, "spectral_convergence_status": "ok", "amplitude_floor": 5.299135910844449e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.88135877551315, "spectral_convergence": 0.39514711745127384, "spectral_convergence_status": "ok", "amplitude_floor": 3.9942669914893874e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.486389753340235e-05, "output_power": 6.916324049439685e-05, "delta_db": -0.34397100418011056, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003919851371758182, "output_power": 0.0004809542757514585, "delta_db": 0.8883418975391606, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038865006012116416, "output_power": 0.0004950252563530847, "delta_db": 1.0506861812411359, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001490171424082218, "output_power": 0.0001713263053901819, "delta_db": 0.6058781853466113, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.76347124356667e-05, "output_power": 0.0001121706722007443, "delta_db": 0.6027507099273358, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002530434948901305, "output_power": 0.0002751266449599499, "delta_db": 0.3633747500790601, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6425004547189024, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7313949649742557, "pitch": null, "output_rms_dbfs": -27.948588908435465, "output_peak": 0.3816178104408559, "reference_rms_dbfs": -28.67998387340972, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.856820326547596}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.947854524209456e-05, "log_rms_mae_db": 1.7565922267652838, "rms_envelope_nrmse": 0.15482796979253896, "envelope_correlation": 0.9807812720376126, "rms_gain_error_db": 0.7347374876781905, "reference_mean_power": 0.0013552494150092414, "output_mean_power": 0.001605066303225388}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.097356675930481e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024105831474109917, "output_power": 0.0003147952397211147, "reference_active": true, "delta_db": 1.159060403642531}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7795131277493674e-05, "output_power": 3.6284227736148726e-05, "reference_active": true, "delta_db": 1.1574915453842851}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0822198872325408e-05, "output_power": 1.5234781950106896e-05, "reference_active": true, "delta_db": 1.4852073229449343}], "active_band_count": 3, "active_band_mae_db": 1.2672530906572501}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.267346213866395e-05, "log_rms_mae_db": 1.4689918624265184, "rms_envelope_nrmse": 0.11881910348528579, "envelope_correlation": 0.9918677671561223, "rms_gain_error_db": 0.7338336804187093, "reference_mean_power": 0.0013563897957452433, "output_mean_power": 0.0016060826179785286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.582417446963071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021668841148527672, "output_power": 0.00028496837676143973, "reference_active": true, "delta_db": 1.189609827349564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1881573874210085e-05, "output_power": 1.608671929788167e-05, "reference_active": true, "delta_db": 1.3159351114564752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.720949484930854e-07, "output_power": 8.66077423685897e-07, "reference_active": true, "delta_db": 1.1012608662728152}], "active_band_count": 3, "active_band_mae_db": 1.2022686016929516}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.299288573172057e-05, "log_rms_mae_db": 1.8847681574666544, "rms_envelope_nrmse": 0.19692179157951403, "envelope_correlation": 0.9856212596104938, "rms_gain_error_db": 1.1489695715328643, "reference_mean_power": 0.0005726237314993749, "output_mean_power": 0.0007460471916131736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0939920051603367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018116410089920085, "output_power": 0.00025439540119028225, "reference_active": true, "delta_db": 1.474371131517942}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.90482495868649e-05, "output_power": 4.2666108769263176e-05, "reference_active": true, "delta_db": 1.6696306873445443}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.620984223411989e-06, "output_power": 7.663766310438225e-06, "reference_active": true, "delta_db": 1.3462988707799959}], "active_band_count": 3, "active_band_mae_db": 1.496766896547494}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.559252308295269e-05, "log_rms_mae_db": 1.4796252515666641, "rms_envelope_nrmse": 0.1761988693771075, "envelope_correlation": 0.9907827678436709, "rms_gain_error_db": 1.1498464485566213, "reference_mean_power": 0.0005714831913152958, "output_mean_power": 0.0007447115784858869}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7768006225472703e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016765591938032763, "output_power": 0.00023786439139764234, "reference_active": true, "delta_db": 1.5191054091756246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3133776830395815e-05, "output_power": 2.0032661202097815e-05, "reference_active": true, "delta_db": 1.8334901371212147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.763861134744573e-07, "output_power": 5.412888256305669e-07, "reference_active": true, "delta_db": 0.5546996817666948}], "active_band_count": 3, "active_band_mae_db": 1.3024317426878447}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.686331455816738e-05, "log_rms_mae_db": 1.2482720343444837, "rms_envelope_nrmse": 0.1342235497132654, "envelope_correlation": 0.991877584505586, "rms_gain_error_db": 0.6823493895026599, "reference_mean_power": 0.00020459070686192204, "output_mean_power": 0.00023939817855671267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3670238086467597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.952541720107036e-05, "output_power": 8.170779119856436e-05, "reference_active": true, "delta_db": 0.7011986678856641}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.449912227458803e-06, "output_power": 8.991652027138054e-06, "reference_active": true, "delta_db": 0.8168833549402638}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5578562674598188e-06, "output_power": 1.3992196520139752e-06, "reference_active": true, "delta_db": -0.4664148968852722}], "active_band_count": 3, "active_band_mae_db": 0.6614989732370667}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.4935905625855925e-05, "log_rms_mae_db": 0.864143105461925, "rms_envelope_nrmse": 0.10079348384612877, "envelope_correlation": 0.9973474245273052, "rms_gain_error_db": 0.681602875818831, "reference_mean_power": 0.00020499017149025147, "output_mean_power": 0.00023982437776412152}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2857171527439997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.293887842389662e-05, "output_power": 7.391071034732099e-05, "reference_active": true, "delta_db": 0.697883767708317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.713520804403163e-06, "output_power": 4.4999436143899265e-06, "reference_active": true, "delta_db": 0.8342121064908301}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3334126499972412e-07, "output_power": 1.1737093743057314e-07, "reference_active": true, "delta_db": -0.5540399766650494}], "active_band_count": 3, "active_band_mae_db": 0.6953786169547321}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.417692344469317e-05, "log_rms_mae_db": 1.3227490773420425, "rms_envelope_nrmse": 0.22258639160804375, "envelope_correlation": 0.9679180575264781, "rms_gain_error_db": 0.2673091179080828, "reference_mean_power": 0.00010545678906217645, "output_mean_power": 0.00011215159531265797}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.569551408613973e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4047256300828896e-05, "output_power": 3.396356392669705e-05, "reference_active": true, "delta_db": -0.01068863880430938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.550080432854394e-06, "output_power": 7.166078916554032e-06, "reference_active": true, "delta_db": -0.22669991768149753}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.516411722896473e-06, "output_power": 4.2411903855682874e-06, "reference_active": true, "delta_db": 0.813880490049538}], "active_band_count": 3, "active_band_mae_db": 0.3504230155117816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.848412221971719e-05, "log_rms_mae_db": 0.7710911383961001, "rms_envelope_nrmse": 0.12186581966527374, "envelope_correlation": 0.9903265911619179, "rms_gain_error_db": 0.2672224015651522, "reference_mean_power": 0.0001059674543662489, "output_mean_power": 0.00011269242946011832}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.064372154195953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3923869911506206e-05, "output_power": 3.41844993000887e-05, "reference_active": true, "delta_db": 0.03323834033857215}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.204493449095079e-06, "output_power": 3.887382044920694e-06, "reference_active": true, "delta_db": -0.3405645574439781}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4577951814618017e-07, "output_power": 2.540792470807684e-07, "reference_active": true, "delta_db": 0.14423505343603277}], "active_band_count": 3, "active_band_mae_db": 0.17267931707286102}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9049380593691768, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b64131907cd9ab41869e3a06f60b87ae4f0974ac4e2bccb9d30c380c2f2a2f26", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.991877584505586, "standard_deviation": null, "minimum": 0.991877584505586, "maximum": 0.991877584505586, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2482720343444837, "standard_deviation": null, "minimum": 1.2482720343444837, "maximum": 1.2482720343444837, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6614989732370667, "standard_deviation": null, "minimum": 0.6614989732370667, "maximum": 0.6614989732370667, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1342235497132654, "standard_deviation": null, "minimum": 0.1342235497132654, "maximum": 0.1342235497132654, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973474245273052, "standard_deviation": null, "minimum": 0.9973474245273052, "maximum": 0.9973474245273052, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.864143105461925, "standard_deviation": null, "minimum": 0.864143105461925, "maximum": 0.864143105461925, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6953786169547321, "standard_deviation": null, "minimum": 0.6953786169547321, "maximum": 0.6953786169547321, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10079348384612877, "standard_deviation": null, "minimum": 0.10079348384612877, "maximum": 0.10079348384612877, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9679180575264781, "standard_deviation": null, "minimum": 0.9679180575264781, "maximum": 0.9679180575264781, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3227490773420425, "standard_deviation": null, "minimum": 1.3227490773420425, "maximum": 1.3227490773420425, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3504230155117816, "standard_deviation": null, "minimum": 0.3504230155117816, "maximum": 0.3504230155117816, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22258639160804375, "standard_deviation": null, "minimum": 0.22258639160804375, "maximum": 0.22258639160804375, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9903265911619179, "standard_deviation": null, "minimum": 0.9903265911619179, "maximum": 0.9903265911619179, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7710911383961001, "standard_deviation": null, "minimum": 0.7710911383961001, "maximum": 0.7710911383961001, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.17267931707286102, "standard_deviation": null, "minimum": 0.17267931707286102, "maximum": 0.17267931707286102, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12186581966527374, "standard_deviation": null, "minimum": 0.12186581966527374, "maximum": 0.12186581966527374, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9856212596104938, "standard_deviation": null, "minimum": 0.9856212596104938, "maximum": 0.9856212596104938, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8847681574666544, "standard_deviation": null, "minimum": 1.8847681574666544, "maximum": 1.8847681574666544, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.496766896547494, "standard_deviation": null, "minimum": 1.496766896547494, "maximum": 1.496766896547494, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19692179157951403, "standard_deviation": null, "minimum": 0.19692179157951403, "maximum": 0.19692179157951403, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9907827678436709, "standard_deviation": null, "minimum": 0.9907827678436709, "maximum": 0.9907827678436709, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4796252515666641, "standard_deviation": null, "minimum": 1.4796252515666641, "maximum": 1.4796252515666641, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3024317426878447, "standard_deviation": null, "minimum": 1.3024317426878447, "maximum": 1.3024317426878447, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1761988693771075, "standard_deviation": null, "minimum": 0.1761988693771075, "maximum": 0.1761988693771075, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6425004547189024, "standard_deviation": null, "minimum": 0.6425004547189024, "maximum": 0.6425004547189024, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.856820326547596, "standard_deviation": null, "minimum": 5.856820326547596, "maximum": 5.856820326547596, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.3816178104408559, "standard_deviation": null, "minimum": 0.3816178104408559, "maximum": 0.3816178104408559, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7313949649742557, "standard_deviation": null, "minimum": 0.7313949649742557, "maximum": 0.7313949649742557, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9807812720376126, "standard_deviation": null, "minimum": 0.9807812720376126, "maximum": 0.9807812720376126, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7565922267652838, "standard_deviation": null, "minimum": 1.7565922267652838, "maximum": 1.7565922267652838, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.2672530906572501, "standard_deviation": null, "minimum": 1.2672530906572501, "maximum": 1.2672530906572501, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15482796979253896, "standard_deviation": null, "minimum": 0.15482796979253896, "maximum": 0.15482796979253896, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9918677671561223, "standard_deviation": null, "minimum": 0.9918677671561223, "maximum": 0.9918677671561223, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.4689918624265184, "standard_deviation": null, "minimum": 1.4689918624265184, "maximum": 1.4689918624265184, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.2022686016929516, "standard_deviation": null, "minimum": 1.2022686016929516, "maximum": 1.2022686016929516, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11881910348528579, "standard_deviation": null, "minimum": 0.11881910348528579, "maximum": 0.11881910348528579, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1221"}, {"case": "librispeech_1221-135767-0019", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.074368715286255, "predicted_mos": 2.074368715286255, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60480, "input_sample_rate": 16000, "input_seconds": 3.78, "input_float32_sha256": "eb1f1065339cf811a440b9ec7771b2959a0479b0c5836ecc4666486bcdbeec74", "internal_samples": 60480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.939562799874693}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.703921256204552, "全帯域_10ms_rms_envelope_nrmse": 0.24423434826464613, "全帯域_10ms_envelope_correlation": 0.9833752297147289, "全帯域_10ms_modulation_mae_db": 2.552593507595887, "全帯域_40ms_log_rms_mae_db": 2.32927673703483, "全帯域_40ms_rms_envelope_nrmse": 0.21696584810822309, "全帯域_40ms_envelope_correlation": 0.9916292876029634, "全帯域_40ms_modulation_mae_db": 2.7550035889765225, "300–1000 Hz_10ms_log_rms_mae_db": 2.1802808956843864, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2889409448034763, "300–1000 Hz_10ms_envelope_correlation": 0.9807925646960378, "300–1000 Hz_10ms_modulation_mae_db": 2.8976601337105783, "300–1000 Hz_40ms_log_rms_mae_db": 1.8992616504126476, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.26578332737118676, "300–1000 Hz_40ms_envelope_correlation": 0.9873269038310033, "300–1000 Hz_40ms_modulation_mae_db": 2.550256197038312, "1000–3000 Hz_10ms_log_rms_mae_db": 1.453505969267793, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21728866765005625, "1000–3000 Hz_10ms_envelope_correlation": 0.9827859862866887, "1000–3000 Hz_10ms_modulation_mae_db": 0.9320262060931026, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0532523235733773, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17932615978150293, "1000–3000 Hz_40ms_envelope_correlation": 0.9920522688742893, "1000–3000 Hz_40ms_modulation_mae_db": 0.917145758900276, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4092598753277716, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.314494427437901, "3000–6000 Hz_10ms_envelope_correlation": 0.9672850410378031, "3000–6000 Hz_10ms_modulation_mae_db": 3.7546466519946655, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9153681634658684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.23554215988507535, "3000–6000 Hz_40ms_envelope_correlation": 0.9899116370410976, "3000–6000 Hz_40ms_modulation_mae_db": 3.60792199190193, "mr_stft_db": 6.039299157034946, "band_mae_db": 1.1970881762636945, "rms_gain_error_db": 1.4818695470872667, "output_peak": 0.49383870568827154}, "full_scale": {"sample_count": 60480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60480, "output_samples": 60480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.0895045031316695, "spectral_convergence": 0.5014396813171749, "spectral_convergence_status": "ok", "amplitude_floor": 9.224568988143785e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.909923841553883, "spectral_convergence": 0.45983822363832416, "spectral_convergence_status": "ok", "amplitude_floor": 8.440100575569802e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.118469126419284, "spectral_convergence": 0.5099684671746193, "spectral_convergence_status": "ok", "amplitude_floor": 6.303303867538867e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.6236728988818796e-05, "output_power": 4.9438290293668915e-05, "delta_db": 0.2907634089862769, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005640699493062965, "output_power": 0.0008385410936299391, "delta_db": 1.7219138713764148, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007824000374604723, "output_power": 0.0011271481067912403, "delta_db": 1.585521233364009, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028236131111099825, "output_power": 0.0003816949071347814, "delta_db": 1.3091117547596, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018611455363091946, "output_power": 0.00022525165440066152, "delta_db": 0.8288765435855667, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00035218826443925665, "output_power": 0.0004913705287161461, "delta_db": 1.4463422455102988, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.1970881762636945, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.4818695470872667, "pitch": null, "output_rms_dbfs": -25.067588602765056, "output_peak": 0.49383870568827154, "reference_rms_dbfs": -26.54945814985232, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.039299157034946}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00014780031156712295, "log_rms_mae_db": 2.703921256204552, "rms_envelope_nrmse": 0.24423434826464613, "envelope_correlation": 0.9833752297147289, "rms_gain_error_db": 1.4819112422326162, "reference_mean_power": 0.0022162793576100603, "output_mean_power": 0.003117565779387692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 1.2692207138635437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044041649987677594, "output_power": 0.0006724111391308254, "reference_active": true, "delta_db": 1.8377131820425021}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888390879465172, "output_power": 0.00023129639658072939, "reference_active": true, "delta_db": 2.890457909923267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.435323106110239e-05, "output_power": 4.780986665360112e-05, "reference_active": true, "delta_db": 2.929609430821892}], "active_band_count": 3, "active_band_mae_db": 2.552593507595887}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00013486753957748745, "log_rms_mae_db": 2.32927673703483, "rms_envelope_nrmse": 0.21696584810822309, "envelope_correlation": 0.9916292876029634, "rms_gain_error_db": 1.4820740826207852, "reference_mean_power": 0.0022339157834068563, "output_mean_power": 0.003142492174306244}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.176190391785113e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004109863651179866, "output_power": 0.000641031351188561, "reference_active": true, "delta_db": 1.9305185629714985}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3981406650238314e-05, "output_power": 9.664704251876521e-05, "reference_active": true, "delta_db": 2.529443715538486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0096166790138946e-06, "output_power": 2.4247187198360267e-06, "reference_active": true, "delta_db": 3.8050484884195828}], "active_band_count": 3, "active_band_mae_db": 2.7550035889765225}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00012936155891805085, "log_rms_mae_db": 2.1802808956843864, "rms_envelope_nrmse": 0.2889409448034763, "envelope_correlation": 0.9807925646960378, "rms_gain_error_db": 1.7464413273900734, "reference_mean_power": 0.0012072233727869219, "output_mean_power": 0.0018048111604515055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 8.078028680738071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004104434555454037, "output_power": 0.0006396438601863846, "reference_active": true, "delta_db": 1.9268490023492246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.258149083667088e-05, "output_power": 0.00022022183339987864, "reference_active": true, "delta_db": 3.763362039901728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4118213823838e-05, "output_power": 2.8187508552597985e-05, "reference_active": true, "delta_db": 3.0027693588807827}], "active_band_count": 3, "active_band_mae_db": 2.8976601337105783}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00011150674016193665, "log_rms_mae_db": 1.8992616504126476, "rms_envelope_nrmse": 0.26578332737118676, "envelope_correlation": 0.9873269038310033, "rms_gain_error_db": 1.7464378500183688, "reference_mean_power": 0.0012168889554241998, "output_mean_power": 0.0018192598476464144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.50453906071102e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038524435492848886, "output_power": 0.0006190965646515916, "reference_active": true, "delta_db": 2.060221105093258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.764413089442773e-05, "output_power": 0.00010937620737502482, "reference_active": true, "delta_db": 3.6091345099232535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0419619324101079e-06, "output_power": 1.644345787672283e-06, "reference_active": true, "delta_db": 1.9814129760984245}], "active_band_count": 3, "active_band_mae_db": 2.550256197038312}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 7.756656945350889e-05, "log_rms_mae_db": 1.453505969267793, "rms_envelope_nrmse": 0.21728866765005625, "envelope_correlation": 0.9827859862866887, "rms_gain_error_db": 1.1686261417805732, "reference_mean_power": 0.0003662329710023159, "output_mean_power": 0.00047931393386699343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 2.3479287649550626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015936395121522886, "output_power": 0.0002185222169792025, "reference_active": true, "delta_db": 1.371055089623769}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.480344306075996e-05, "output_power": 2.6175282149593253e-05, "reference_active": true, "delta_db": 0.23379400651342988}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.158977700048883e-06, "output_power": 9.41833127113317e-06, "reference_active": true, "delta_db": 1.191229522142109}], "active_band_count": 3, "active_band_mae_db": 0.9320262060931026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 6.756499548428438e-05, "log_rms_mae_db": 1.0532523235733773, "rms_envelope_nrmse": 0.17932615978150293, "envelope_correlation": 0.9920522688742893, "rms_gain_error_db": 1.1686446644285204, "reference_mean_power": 0.00036915132536056565, "output_mean_power": 0.00048313544275955026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 2.14579734589437e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015227225639161801, "output_power": 0.0002074922516033903, "reference_active": true, "delta_db": 1.3438110024343222}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5037342052746137e-05, "output_power": 1.603227694349565e-05, "reference_active": true, "delta_db": 0.27824127620829914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0212382710455496e-07, "output_power": 5.215528308598416e-07, "reference_active": true, "delta_db": 1.129384998058207}], "active_band_count": 3, "active_band_mae_db": 0.917145758900276}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 755, "rms_floor": 0.00011110909618382604, "log_rms_mae_db": 1.4092598753277716, "rms_envelope_nrmse": 0.314494427437901, "envelope_correlation": 0.9672850410378031, "rms_gain_error_db": 1.508799000045808, "reference_mean_power": 0.00024441441718797436, "output_mean_power": 0.0003459447306427106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 755, "frequency_resolution_hz": 0.26490066225165565, "power_floor": 3.469643399885664e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011046778972231385, "output_power": 0.0001692351237686559, "reference_active": true, "delta_db": 1.8525483877792268}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.9627372395634935e-05, "output_power": 9.45078583509844e-05, "reference_active": true, "delta_db": 2.000222500240956}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.408768678752184e-06, "output_power": 2.429037499302896e-05, "reference_active": true, "delta_db": 7.411169067963814}], "active_band_count": 3, "active_band_mae_db": 3.7546466519946655}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.166729119583493e-05, "log_rms_mae_db": 0.9153681634658684, "rms_envelope_nrmse": 0.23554215988507535, "envelope_correlation": 0.9899116370410976, "rms_gain_error_db": 1.5088183967391355, "reference_mean_power": 0.00024636420339127556, "output_mean_power": 0.00034870601994460137}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.3341266023063243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010489106418058692, "output_power": 0.00017126355903885883, "reference_active": true, "delta_db": 2.129264732777736}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0862352851100365e-05, "output_power": 5.589045923304037e-05, "reference_active": true, "delta_db": 2.5790864581352855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.747611403563906e-07, "output_power": 1.532131460078922e-06, "reference_active": true, "delta_db": 6.11541478479277}], "active_band_count": 3, "active_band_mae_db": 3.60792199190193}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.7439709020549667, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0ecd4314fe7042f9579e8890ff4ae25ee2544584ccecee03e4a51136858b734e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9827859862866887, "standard_deviation": null, "minimum": 0.9827859862866887, "maximum": 0.9827859862866887, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.453505969267793, "standard_deviation": null, "minimum": 1.453505969267793, "maximum": 1.453505969267793, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9320262060931026, "standard_deviation": null, "minimum": 0.9320262060931026, "maximum": 0.9320262060931026, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21728866765005625, "standard_deviation": null, "minimum": 0.21728866765005625, "maximum": 0.21728866765005625, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9920522688742893, "standard_deviation": null, "minimum": 0.9920522688742893, "maximum": 0.9920522688742893, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0532523235733773, "standard_deviation": null, "minimum": 1.0532523235733773, "maximum": 1.0532523235733773, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.917145758900276, "standard_deviation": null, "minimum": 0.917145758900276, "maximum": 0.917145758900276, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17932615978150293, "standard_deviation": null, "minimum": 0.17932615978150293, "maximum": 0.17932615978150293, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9672850410378031, "standard_deviation": null, "minimum": 0.9672850410378031, "maximum": 0.9672850410378031, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4092598753277716, "standard_deviation": null, "minimum": 1.4092598753277716, "maximum": 1.4092598753277716, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 3.7546466519946655, "standard_deviation": null, "minimum": 3.7546466519946655, "maximum": 3.7546466519946655, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.314494427437901, "standard_deviation": null, "minimum": 0.314494427437901, "maximum": 0.314494427437901, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9899116370410976, "standard_deviation": null, "minimum": 0.9899116370410976, "maximum": 0.9899116370410976, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9153681634658684, "standard_deviation": null, "minimum": 0.9153681634658684, "maximum": 0.9153681634658684, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 3.60792199190193, "standard_deviation": null, "minimum": 3.60792199190193, "maximum": 3.60792199190193, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23554215988507535, "standard_deviation": null, "minimum": 0.23554215988507535, "maximum": 0.23554215988507535, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9807925646960378, "standard_deviation": null, "minimum": 0.9807925646960378, "maximum": 0.9807925646960378, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1802808956843864, "standard_deviation": null, "minimum": 2.1802808956843864, "maximum": 2.1802808956843864, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.8976601337105783, "standard_deviation": null, "minimum": 2.8976601337105783, "maximum": 2.8976601337105783, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2889409448034763, "standard_deviation": null, "minimum": 0.2889409448034763, "maximum": 0.2889409448034763, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9873269038310033, "standard_deviation": null, "minimum": 0.9873269038310033, "maximum": 0.9873269038310033, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.8992616504126476, "standard_deviation": null, "minimum": 1.8992616504126476, "maximum": 1.8992616504126476, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 2.550256197038312, "standard_deviation": null, "minimum": 2.550256197038312, "maximum": 2.550256197038312, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.26578332737118676, "standard_deviation": null, "minimum": 0.26578332737118676, "maximum": 0.26578332737118676, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.1970881762636945, "standard_deviation": null, "minimum": 1.1970881762636945, "maximum": 1.1970881762636945, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.039299157034946, "standard_deviation": null, "minimum": 6.039299157034946, "maximum": 6.039299157034946, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.49383870568827154, "standard_deviation": null, "minimum": 0.49383870568827154, "maximum": 0.49383870568827154, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.4818695470872667, "standard_deviation": null, "minimum": 1.4818695470872667, "maximum": 1.4818695470872667, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9833752297147289, "standard_deviation": null, "minimum": 0.9833752297147289, "maximum": 0.9833752297147289, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.703921256204552, "standard_deviation": null, "minimum": 2.703921256204552, "maximum": 2.703921256204552, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 2.552593507595887, "standard_deviation": null, "minimum": 2.552593507595887, "maximum": 2.552593507595887, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.24423434826464613, "standard_deviation": null, "minimum": 0.24423434826464613, "maximum": 0.24423434826464613, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9916292876029634, "standard_deviation": null, "minimum": 0.9916292876029634, "maximum": 0.9916292876029634, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.32927673703483, "standard_deviation": null, "minimum": 2.32927673703483, "maximum": 2.32927673703483, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 2.7550035889765225, "standard_deviation": null, "minimum": 2.7550035889765225, "maximum": 2.7550035889765225, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.21696584810822309, "standard_deviation": null, "minimum": 0.21696584810822309, "maximum": 0.21696584810822309, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1221"}, {"case": "librispeech_1284-1180-0020", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.06990909576416, "predicted_mos": 3.06990909576416, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 93920, "input_sample_rate": 16000, "input_seconds": 5.87, "input_float32_sha256": "cef9940d48080ea67f945ea00fc6bf972da3bf83ec8fd0eb311081158d3b70b4", "internal_samples": 93920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.015785899944603}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.795238137967324, "全帯域_10ms_rms_envelope_nrmse": 0.19112649674733817, "全帯域_10ms_envelope_correlation": 0.9829056537211514, "全帯域_10ms_modulation_mae_db": 1.1800498416415464, "全帯域_40ms_log_rms_mae_db": 1.3659210015701961, "全帯域_40ms_rms_envelope_nrmse": 0.1692153305416878, "全帯域_40ms_envelope_correlation": 0.9895802883874522, "全帯域_40ms_modulation_mae_db": 1.2144568032059526, "300–1000 Hz_10ms_log_rms_mae_db": 1.787277254525518, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2412904457328695, "300–1000 Hz_10ms_envelope_correlation": 0.9868274625856098, "300–1000 Hz_10ms_modulation_mae_db": 1.4594768420202964, "300–1000 Hz_40ms_log_rms_mae_db": 1.57406364864355, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.22102719062487164, "300–1000 Hz_40ms_envelope_correlation": 0.9928716737706412, "300–1000 Hz_40ms_modulation_mae_db": 1.4435271761728183, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3923239026159324, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15324069284142672, "1000–3000 Hz_10ms_envelope_correlation": 0.9903873743550452, "1000–3000 Hz_10ms_modulation_mae_db": 0.8811488105515721, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0400304632302235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13059160282837598, "1000–3000 Hz_40ms_envelope_correlation": 0.9949768750859826, "1000–3000 Hz_40ms_modulation_mae_db": 0.8108390015477737, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6501622057589287, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20372941227230645, "3000–6000 Hz_10ms_envelope_correlation": 0.9659824780138592, "3000–6000 Hz_10ms_modulation_mae_db": 0.38253238730682987, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0346309472599131, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.122343314715708, "3000–6000 Hz_40ms_envelope_correlation": 0.987407053011333, "3000–6000 Hz_40ms_modulation_mae_db": 0.16168662748806215, "mr_stft_db": 5.686550483977761, "band_mae_db": 0.7475150308506798, "rms_gain_error_db": 1.1199180341102475, "output_peak": 0.5277819530169021}, "full_scale": {"sample_count": 93920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 93920, "output_samples": 93920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.83061725455828, "spectral_convergence": 0.4883755650280944, "spectral_convergence_status": "ok", "amplitude_floor": 2.214924872151779e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.398584253583207, "spectral_convergence": 0.3127900436553947, "spectral_convergence_status": "ok", "amplitude_floor": 1.3113821592368984e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.830449943791797, "spectral_convergence": 0.33460376804746994, "spectral_convergence_status": "ok", "amplitude_floor": 7.380112905156557e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00039502697710642294, "output_power": 0.00036855566361562514, "delta_db": -0.3012366597976051, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014539054117940992, "output_power": 0.001740477060459797, "delta_db": 0.7813215054023438, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0016903939251448888, "output_power": 0.002501189685133264, "delta_db": 1.7015870568797564, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.480236057553453e-05, "output_power": 7.47561914715912e-05, "delta_db": 0.6205634136922717, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.753736279208405e-05, "output_power": 0.00011649419636188392, "delta_db": 0.7713328074033328, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.937191536111839e-05, "output_power": 5.3013301310536995e-05, "delta_db": 0.3090487419287695, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7475150308506798, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1199180341102475, "pitch": null, "output_rms_dbfs": -23.13856738250836, "output_peak": 0.5277819530169021, "reference_rms_dbfs": -24.258485416618605, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.686550483977761}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 93920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002156956510733609, "log_rms_mae_db": 1.795238137967324, "rms_envelope_nrmse": 0.19112649674733817, "envelope_correlation": 0.9829056537211514, "rms_gain_error_db": 1.119918656904309, "reference_mean_power": 0.003754233895192234, "output_mean_power": 0.004858622892421096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.680528929094005e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006742966926403748, "output_power": 0.0008824009640231383, "reference_active": true, "delta_db": 1.1681494438944167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002358850904334285, "output_power": 0.0003307560434179063, "reference_active": true, "delta_db": 1.4680729681239313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0118677140216256e-05, "output_power": 3.708758924137906e-05, "reference_active": true, "delta_db": 0.9039271129062914}], "active_band_count": 3, "active_band_mae_db": 1.1800498416415464}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00016458512240675186, "log_rms_mae_db": 1.3659210015701961, "rms_envelope_nrmse": 0.1692153305416878, "envelope_correlation": 0.9895802883874522, "rms_gain_error_db": 1.1199381781160893, "reference_mean_power": 0.003773515049488898, "output_mean_power": 0.004883597966113195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.4929366354790985e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006129787155244603, "output_power": 0.0008191425280189695, "reference_active": true, "delta_db": 1.259140793388671}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073825851363778, "output_power": 0.00016161013070240885, "reference_active": true, "delta_db": 1.6417089205215099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.608951147124256e-06, "output_power": 3.095408825769411e-06, "reference_active": true, "delta_db": 0.7425206957076771}], "active_band_count": 3, "active_band_mae_db": 1.2144568032059526}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 0.0002116840322083979, "log_rms_mae_db": 1.787277254525518, "rms_envelope_nrmse": 0.2412904457328695, "envelope_correlation": 0.9868274625856098, "rms_gain_error_db": 1.5821819306605067, "reference_mean_power": 0.002333704816993391, "output_mean_power": 0.0033594185479420215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 1.3033890876287954e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006736793990758906, "output_power": 0.0009439809966975774, "reference_active": true, "delta_db": 1.4650998464000629}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002058984330594163, "output_power": 0.00031827747017383997, "reference_active": true, "delta_db": 1.8915285587948825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6276513462408744e-05, "output_power": 3.3246654619961895e-05, "reference_active": true, "delta_db": 1.0218021208659431}], "active_band_count": 3, "active_band_mae_db": 1.4594768420202964}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 0.00015464611695479593, "log_rms_mae_db": 1.57406364864355, "rms_envelope_nrmse": 0.22102719062487164, "envelope_correlation": 0.9928716737706412, "rms_gain_error_db": 1.5821817060744168, "reference_mean_power": 0.002345702993864749, "output_mean_power": 0.0033766900083778352}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 1.1648344391379404e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377113304270822, "output_power": 0.0009028857687413119, "reference_active": true, "delta_db": 1.5100867443167276}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.690162436721034e-05, "output_power": 0.00015146448171440342, "reference_active": true, "delta_db": 1.9397974589562148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0235059299775393e-06, "output_power": 3.7032289019037715e-06, "reference_active": true, "delta_db": 0.880697325245513}], "active_band_count": 3, "active_band_mae_db": 1.4435271761728183}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 4.7944659079237455e-05, "log_rms_mae_db": 1.3923239026159324, "rms_envelope_nrmse": 0.15324069284142672, "envelope_correlation": 0.9903873743550452, "rms_gain_error_db": 0.8765966345786146, "reference_mean_power": 0.0001071814340711088, "output_mean_power": 0.00013115330135256777}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 8.167234143423064e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.040560289520861e-05, "output_power": 5.006716943440308e-05, "reference_active": true, "delta_db": 0.931114478041044}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.346030743633636e-06, "output_power": 1.0973581009459934e-05, "reference_active": true, "delta_db": 1.1886839389683055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1788672389317654e-06, "output_power": 2.4580774938288423e-06, "reference_active": true, "delta_db": 0.5236480146453669}], "active_band_count": 3, "active_band_mae_db": 0.8811488105515721}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 3.945206980281183e-05, "log_rms_mae_db": 1.0400304632302235, "rms_envelope_nrmse": 0.13059160282837598, "envelope_correlation": 0.9949768750859826, "rms_gain_error_db": 0.8766004361596685, "reference_mean_power": 0.00010773182378082642, "output_mean_power": 0.00013182690491270484}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 7.380645247762918e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7294287822712847e-05, "output_power": 4.654847707161149e-05, "reference_active": true, "delta_db": 0.9626315850495515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.878816663714596e-06, "output_power": 5.070449429466135e-06, "reference_active": true, "delta_db": 1.1634720289603482}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8761762452219972e-07, "output_power": 2.013330033071485e-07, "reference_active": true, "delta_db": 0.3064133906334215}], "active_band_count": 3, "active_band_mae_db": 0.8108390015477737}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1173, "rms_floor": 5.718269674480757e-05, "log_rms_mae_db": 1.6501622057589287, "rms_envelope_nrmse": 0.20372941227230645, "envelope_correlation": 0.9659824780138592, "rms_gain_error_db": 0.37845087516083503, "reference_mean_power": 7.041944137047507e-05, "output_mean_power": 7.683120826486764e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1173, "frequency_resolution_hz": 0.17050298380221654, "power_floor": 5.931095881530669e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9173785649337023e-05, "output_power": 2.808374402501363e-05, "reference_active": true, "delta_db": -0.16537781791911152}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1741151998348071e-05, "output_power": 1.173233548530976e-05, "reference_active": true, "delta_db": -0.0032623726192238236}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.929169683988829e-06, "output_power": 2.4169414270714834e-06, "reference_active": true, "delta_db": 0.9789569713821542}], "active_band_count": 3, "active_band_mae_db": 0.38253238730682987}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 93920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1167, "rms_floor": 4.673628190489769e-05, "log_rms_mae_db": 1.0346309472599131, "rms_envelope_nrmse": 0.122343314715708, "envelope_correlation": 0.987407053011333, "rms_gain_error_db": 0.3784039119773758, "reference_mean_power": 7.077979037597618e-05, "output_mean_power": 7.722353236558989e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1167, "frequency_resolution_hz": 0.1713796058269066, "power_floor": 5.253277550211025e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7931175609497952e-05, "output_power": 2.6630213479473862e-05, "reference_active": true, "delta_db": -0.20714567379694165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.1737648370747885e-06, "output_power": 5.2139369137218065e-06, "reference_active": true, "delta_db": 0.0335908723781795}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7085501560216023e-07, "output_power": 1.8074240884934886e-07, "reference_active": true, "delta_db": 0.24432333628906527}], "active_band_count": 3, "active_band_mae_db": 0.16168662748806215}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9341290518082113, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "38a47f0f740b7a2ce97f87598d8c13dd31bba8278ad2af94bd42271ee2d1dd37", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9903873743550452, "standard_deviation": null, "minimum": 0.9903873743550452, "maximum": 0.9903873743550452, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3923239026159324, "standard_deviation": null, "minimum": 1.3923239026159324, "maximum": 1.3923239026159324, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8811488105515721, "standard_deviation": null, "minimum": 0.8811488105515721, "maximum": 0.8811488105515721, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15324069284142672, "standard_deviation": null, "minimum": 0.15324069284142672, "maximum": 0.15324069284142672, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949768750859826, "standard_deviation": null, "minimum": 0.9949768750859826, "maximum": 0.9949768750859826, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0400304632302235, "standard_deviation": null, "minimum": 1.0400304632302235, "maximum": 1.0400304632302235, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8108390015477737, "standard_deviation": null, "minimum": 0.8108390015477737, "maximum": 0.8108390015477737, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13059160282837598, "standard_deviation": null, "minimum": 0.13059160282837598, "maximum": 0.13059160282837598, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9659824780138592, "standard_deviation": null, "minimum": 0.9659824780138592, "maximum": 0.9659824780138592, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6501622057589287, "standard_deviation": null, "minimum": 1.6501622057589287, "maximum": 1.6501622057589287, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.38253238730682987, "standard_deviation": null, "minimum": 0.38253238730682987, "maximum": 0.38253238730682987, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20372941227230645, "standard_deviation": null, "minimum": 0.20372941227230645, "maximum": 0.20372941227230645, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.987407053011333, "standard_deviation": null, "minimum": 0.987407053011333, "maximum": 0.987407053011333, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0346309472599131, "standard_deviation": null, "minimum": 1.0346309472599131, "maximum": 1.0346309472599131, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.16168662748806215, "standard_deviation": null, "minimum": 0.16168662748806215, "maximum": 0.16168662748806215, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.122343314715708, "standard_deviation": null, "minimum": 0.122343314715708, "maximum": 0.122343314715708, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9868274625856098, "standard_deviation": null, "minimum": 0.9868274625856098, "maximum": 0.9868274625856098, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.787277254525518, "standard_deviation": null, "minimum": 1.787277254525518, "maximum": 1.787277254525518, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4594768420202964, "standard_deviation": null, "minimum": 1.4594768420202964, "maximum": 1.4594768420202964, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2412904457328695, "standard_deviation": null, "minimum": 0.2412904457328695, "maximum": 0.2412904457328695, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9928716737706412, "standard_deviation": null, "minimum": 0.9928716737706412, "maximum": 0.9928716737706412, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.57406364864355, "standard_deviation": null, "minimum": 1.57406364864355, "maximum": 1.57406364864355, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.4435271761728183, "standard_deviation": null, "minimum": 1.4435271761728183, "maximum": 1.4435271761728183, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.22102719062487164, "standard_deviation": null, "minimum": 0.22102719062487164, "maximum": 0.22102719062487164, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7475150308506798, "standard_deviation": null, "minimum": 0.7475150308506798, "maximum": 0.7475150308506798, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.686550483977761, "standard_deviation": null, "minimum": 5.686550483977761, "maximum": 5.686550483977761, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5277819530169021, "standard_deviation": null, "minimum": 0.5277819530169021, "maximum": 0.5277819530169021, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1199180341102475, "standard_deviation": null, "minimum": 1.1199180341102475, "maximum": 1.1199180341102475, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9829056537211514, "standard_deviation": null, "minimum": 0.9829056537211514, "maximum": 0.9829056537211514, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.795238137967324, "standard_deviation": null, "minimum": 1.795238137967324, "maximum": 1.795238137967324, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1800498416415464, "standard_deviation": null, "minimum": 1.1800498416415464, "maximum": 1.1800498416415464, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19112649674733817, "standard_deviation": null, "minimum": 0.19112649674733817, "maximum": 0.19112649674733817, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9895802883874522, "standard_deviation": null, "minimum": 0.9895802883874522, "maximum": 0.9895802883874522, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.3659210015701961, "standard_deviation": null, "minimum": 1.3659210015701961, "maximum": 1.3659210015701961, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.2144568032059526, "standard_deviation": null, "minimum": 1.2144568032059526, "maximum": 1.2144568032059526, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1692153305416878, "standard_deviation": null, "minimum": 0.1692153305416878, "maximum": 0.1692153305416878, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1284"}, {"case": "librispeech_1284-1180-0027", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.214627504348755, "predicted_mos": 3.214627504348755, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 52320, "input_sample_rate": 16000, "input_seconds": 3.27, "input_float32_sha256": "5c97ead3bba440e36b254ba44bbd953a146e26843702a9efb299ef700913ab8d", "internal_samples": 52320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.6829225001856685}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.1178300589609886, "全帯域_10ms_rms_envelope_nrmse": 0.24609399562920034, "全帯域_10ms_envelope_correlation": 0.9835074027117906, "全帯域_10ms_modulation_mae_db": 1.6911064996378553, "全帯域_40ms_log_rms_mae_db": 1.7725638455537733, "全帯域_40ms_rms_envelope_nrmse": 0.22975780441479232, "全帯域_40ms_envelope_correlation": 0.9902892460141871, "全帯域_40ms_modulation_mae_db": 1.5241847565607618, "300–1000 Hz_10ms_log_rms_mae_db": 2.001410852761706, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.307445438156663, "300–1000 Hz_10ms_envelope_correlation": 0.9866595951927567, "300–1000 Hz_10ms_modulation_mae_db": 2.178152707673901, "300–1000 Hz_40ms_log_rms_mae_db": 1.6678790007810185, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2949038889851758, "300–1000 Hz_40ms_envelope_correlation": 0.9913117407544487, "300–1000 Hz_40ms_modulation_mae_db": 1.8003388660219297, "1000–3000 Hz_10ms_log_rms_mae_db": 1.462346038966885, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17566126459658862, "1000–3000 Hz_10ms_envelope_correlation": 0.9894763043288372, "1000–3000 Hz_10ms_modulation_mae_db": 1.3186958619467972, "1000–3000 Hz_40ms_log_rms_mae_db": 1.049495855769314, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14236017248982993, "1000–3000 Hz_40ms_envelope_correlation": 0.9968099429493408, "1000–3000 Hz_40ms_modulation_mae_db": 1.3086271039327528, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5744661696970168, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.32593403481826944, "3000–6000 Hz_10ms_envelope_correlation": 0.9009562328163695, "3000–6000 Hz_10ms_modulation_mae_db": 0.8280377334649506, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0542378240781416, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18124532186442605, "3000–6000 Hz_40ms_envelope_correlation": 0.9644910684369602, "3000–6000 Hz_40ms_modulation_mae_db": 0.7647848749050432, "mr_stft_db": 5.680411817036379, "band_mae_db": 1.0066399830741957, "rms_gain_error_db": 1.6263047979776757, "output_peak": 0.4622224764758243}, "full_scale": {"sample_count": 52320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 52320, "output_samples": 52320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.167156766072967, "spectral_convergence": 0.5168433818592597, "spectral_convergence_status": "ok", "amplitude_floor": 1.2746778473169441e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.245449124677452, "spectral_convergence": 0.317965575856695, "spectral_convergence_status": "ok", "amplitude_floor": 1.090624511245042e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6286295603587195, "spectral_convergence": 0.3398499291292952, "spectral_convergence_status": "ok", "amplitude_floor": 7.935262595002381e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004730173956506593, "output_power": 0.00045337075345531065, "delta_db": -0.18423611868573075, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018195471619226037, "output_power": 0.00246618536409746, "delta_db": 1.3206239924237733, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018173989836957952, "output_power": 0.003144140605914414, "delta_db": 2.3805167858791236, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013671881984837986, "output_power": 0.00017912764371469793, "delta_db": 1.1733431221623558, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012324536624073353, "output_power": 0.00014705765951407977, "delta_db": 0.7671704998438533, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.679288082514536e-05, "output_power": 9.11756864265337e-05, "delta_db": 0.21394937945033626, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.0066399830741957, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.6263047979776757, "pitch": null, "output_rms_dbfs": -21.883541111839367, "output_peak": 0.4622224764758243, "reference_rms_dbfs": -23.509845909817045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.680411817036379}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 52320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00017439460044090568, "log_rms_mae_db": 2.1178300589609886, "rms_envelope_nrmse": 0.24609399562920034, "envelope_correlation": 0.9835074027117906, "rms_gain_error_db": 1.6263176702563489, "reference_mean_power": 0.004463526965610992, "output_mean_power": 0.006490974901482833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 1.3056377024951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008270090313661458, "output_power": 0.0012587546984932168, "reference_active": true, "delta_db": 1.824308523361626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010887958720618104, "output_power": 0.000168030214398062, "reference_active": true, "delta_db": 1.8844091586341305}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8937894537156724e-05, "output_power": 3.96211531468676e-05, "reference_active": true, "delta_db": 1.3646018169178098}], "active_band_count": 3, "active_band_mae_db": 1.6911064996378553}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015806536675822195, "log_rms_mae_db": 1.7725638455537733, "rms_envelope_nrmse": 0.22975780441479232, "envelope_correlation": 0.9902892460141871, "rms_gain_error_db": 1.6263639744219534, "reference_mean_power": 0.004504823508022685, "output_mean_power": 0.006551099235052479}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.0922276511075418e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000711888936919089, "output_power": 0.0011080090978577875, "reference_active": true, "delta_db": 1.9213108254330762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.550528203435593e-05, "output_power": 9.176253432195159e-05, "reference_active": true, "delta_db": 1.4638907867377229}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5248696894099073e-06, "output_power": 2.004324326647306e-06, "reference_active": true, "delta_db": 1.187352657511487}], "active_band_count": 3, "active_band_mae_db": 1.5241847565607618}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 0.00016816220894332027, "log_rms_mae_db": 2.001410852761706, "rms_envelope_nrmse": 0.307445438156663, "envelope_correlation": 0.9866595951927567, "rms_gain_error_db": 2.093462055431771, "reference_mean_power": 0.002573247092376177, "output_mean_power": 0.004167040258414629}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 9.221294414644421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007324784888809201, "output_power": 0.0011904648335494962, "reference_active": true, "delta_db": 2.1092169579146085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.576574656453582e-05, "output_power": 0.00015347159454310878, "reference_active": true, "delta_db": 2.527141330168262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.804896765293062e-05, "output_power": 2.794231280310755e-05, "reference_active": true, "delta_db": 1.898099834938832}], "active_band_count": 3, "active_band_mae_db": 2.178152707673901}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.0001473388036867024, "log_rms_mae_db": 1.6678790007810185, "rms_envelope_nrmse": 0.2949038889851758, "envelope_correlation": 0.9913117407544487, "rms_gain_error_db": 2.0934641168332266, "reference_mean_power": 0.002597106693874505, "output_mean_power": 0.004205679787873618}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.911179205548874e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006603877113347142, "output_power": 0.0010972450004929565, "reference_active": true, "delta_db": 2.205046274923226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.506710493994746e-05, "output_power": 7.58083972698221e-05, "reference_active": true, "delta_db": 2.258576545667233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1610111255482175e-06, "output_power": 1.4407073081730398e-06, "reference_active": true, "delta_db": 0.9373937774753304}], "active_band_count": 3, "active_band_mae_db": 1.8003388660219297}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 5.380259383983556e-05, "log_rms_mae_db": 1.462346038966885, "rms_envelope_nrmse": 0.17566126459658862, "envelope_correlation": 0.9894763043288372, "rms_gain_error_db": 1.037176838245641, "reference_mean_power": 0.00019859577428883629, "output_mean_power": 0.00025216667220939254}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 8.760825576371796e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.452705657171872e-05, "output_power": 7.158708657846594e-05, "reference_active": true, "delta_db": 1.1822263339761474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.928893294666125e-06, "output_power": 8.391846271399521e-06, "reference_active": true, "delta_db": 1.508838853074478}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9788817769833428e-06, "output_power": 2.64802504959357e-06, "reference_active": true, "delta_db": 1.2650223987897657}], "active_band_count": 3, "active_band_mae_db": 1.3186958619467972}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 5.051409218298903e-05, "log_rms_mae_db": 1.049495855769314, "rms_envelope_nrmse": 0.14236017248982993, "envelope_correlation": 0.9968099429493408, "rms_gain_error_db": 1.037278152306271, "reference_mean_power": 0.00020041602526910178, "output_mean_power": 0.00025448386966865}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 7.85711379260727e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.022793901144941e-05, "output_power": 6.60989450818515e-05, "reference_active": true, "delta_db": 1.1924917009430216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.619667417561746e-06, "output_power": 3.769032535356545e-06, "reference_active": true, "delta_db": 1.579837278799535}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1855266901360492e-07, "output_power": 1.54620320460442e-07, "reference_active": true, "delta_db": 1.1535523320557024}], "active_band_count": 3, "active_band_mae_db": 1.3086271039327528}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 653, "rms_floor": 6.112392927022402e-05, "log_rms_mae_db": 1.5744661696970168, "rms_envelope_nrmse": 0.32593403481826944, "envelope_correlation": 0.9009562328163695, "rms_gain_error_db": 0.28801531419817744, "reference_mean_power": 8.711996692518435e-05, "output_mean_power": 9.309347313653688e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 653, "frequency_resolution_hz": 0.30627871362940273, "power_floor": 4.203401537603793e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8248175978150172e-05, "output_power": 2.0554210389543688e-05, "reference_active": true, "delta_db": 0.5168133712989714}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.243431842943793e-05, "output_power": 1.3541658450350743e-05, "reference_active": true, "delta_db": 0.37049870775694077}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.767669752037568e-06, "output_power": 4.6855334835544465e-06, "reference_active": true, "delta_db": -1.5968011213389395}], "active_band_count": 3, "active_band_mae_db": 0.8280377334649506}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 52320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 3.626522691368819e-05, "log_rms_mae_db": 1.0542378240781416, "rms_envelope_nrmse": 0.18124532186442605, "envelope_correlation": 0.9644910684369602, "rms_gain_error_db": 0.2881867701920231, "reference_mean_power": 8.789733624317315e-05, "output_mean_power": 9.39278520266496e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 3.223393789064049e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.051966991658498e-05, "output_power": 2.2062197522774073e-05, "reference_active": true, "delta_db": 0.3147839815665379}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.880534452438109e-06, "output_power": 6.230681944397698e-06, "reference_active": true, "delta_db": 0.2511878380223661}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.871896444237529e-07, "output_power": 3.944029070112823e-07, "reference_active": true, "delta_db": -1.7283828051262256}], "active_band_count": 3, "active_band_mae_db": 0.7647848749050432}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9259565179114652, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8ed855221b4bed99867a4fc3de7e78810172269380f59fd6b159cf83dd48c0a6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9894763043288372, "standard_deviation": null, "minimum": 0.9894763043288372, "maximum": 0.9894763043288372, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.462346038966885, "standard_deviation": null, "minimum": 1.462346038966885, "maximum": 1.462346038966885, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3186958619467972, "standard_deviation": null, "minimum": 1.3186958619467972, "maximum": 1.3186958619467972, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17566126459658862, "standard_deviation": null, "minimum": 0.17566126459658862, "maximum": 0.17566126459658862, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968099429493408, "standard_deviation": null, "minimum": 0.9968099429493408, "maximum": 0.9968099429493408, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.049495855769314, "standard_deviation": null, "minimum": 1.049495855769314, "maximum": 1.049495855769314, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3086271039327528, "standard_deviation": null, "minimum": 1.3086271039327528, "maximum": 1.3086271039327528, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14236017248982993, "standard_deviation": null, "minimum": 0.14236017248982993, "maximum": 0.14236017248982993, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9009562328163695, "standard_deviation": null, "minimum": 0.9009562328163695, "maximum": 0.9009562328163695, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5744661696970168, "standard_deviation": null, "minimum": 1.5744661696970168, "maximum": 1.5744661696970168, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8280377334649506, "standard_deviation": null, "minimum": 0.8280377334649506, "maximum": 0.8280377334649506, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.32593403481826944, "standard_deviation": null, "minimum": 0.32593403481826944, "maximum": 0.32593403481826944, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9644910684369602, "standard_deviation": null, "minimum": 0.9644910684369602, "maximum": 0.9644910684369602, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0542378240781416, "standard_deviation": null, "minimum": 1.0542378240781416, "maximum": 1.0542378240781416, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7647848749050432, "standard_deviation": null, "minimum": 0.7647848749050432, "maximum": 0.7647848749050432, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18124532186442605, "standard_deviation": null, "minimum": 0.18124532186442605, "maximum": 0.18124532186442605, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9866595951927567, "standard_deviation": null, "minimum": 0.9866595951927567, "maximum": 0.9866595951927567, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.001410852761706, "standard_deviation": null, "minimum": 2.001410852761706, "maximum": 2.001410852761706, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.178152707673901, "standard_deviation": null, "minimum": 2.178152707673901, "maximum": 2.178152707673901, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.307445438156663, "standard_deviation": null, "minimum": 0.307445438156663, "maximum": 0.307445438156663, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9913117407544487, "standard_deviation": null, "minimum": 0.9913117407544487, "maximum": 0.9913117407544487, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6678790007810185, "standard_deviation": null, "minimum": 1.6678790007810185, "maximum": 1.6678790007810185, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.8003388660219297, "standard_deviation": null, "minimum": 1.8003388660219297, "maximum": 1.8003388660219297, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2949038889851758, "standard_deviation": null, "minimum": 0.2949038889851758, "maximum": 0.2949038889851758, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.0066399830741957, "standard_deviation": null, "minimum": 1.0066399830741957, "maximum": 1.0066399830741957, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.680411817036379, "standard_deviation": null, "minimum": 5.680411817036379, "maximum": 5.680411817036379, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4622224764758243, "standard_deviation": null, "minimum": 0.4622224764758243, "maximum": 0.4622224764758243, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.6263047979776757, "standard_deviation": null, "minimum": 1.6263047979776757, "maximum": 1.6263047979776757, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9835074027117906, "standard_deviation": null, "minimum": 0.9835074027117906, "maximum": 0.9835074027117906, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.1178300589609886, "standard_deviation": null, "minimum": 2.1178300589609886, "maximum": 2.1178300589609886, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.6911064996378553, "standard_deviation": null, "minimum": 1.6911064996378553, "maximum": 1.6911064996378553, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.24609399562920034, "standard_deviation": null, "minimum": 0.24609399562920034, "maximum": 0.24609399562920034, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9902892460141871, "standard_deviation": null, "minimum": 0.9902892460141871, "maximum": 0.9902892460141871, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.7725638455537733, "standard_deviation": null, "minimum": 1.7725638455537733, "maximum": 1.7725638455537733, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.5241847565607618, "standard_deviation": null, "minimum": 1.5241847565607618, "maximum": 1.5241847565607618, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.22975780441479232, "standard_deviation": null, "minimum": 0.22975780441479232, "maximum": 0.22975780441479232, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1284"}, {"case": "librispeech_1284-1180-0028", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.5149476528167725, "predicted_mos": 3.5149476528167725, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1ce700af15d0d1aefd2eef60a92f81951ad56bcfe23f2279b82ba5c7f138f1f1", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.653531800024211}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6994640221299797, "全帯域_10ms_rms_envelope_nrmse": 0.24202111842227497, "全帯域_10ms_envelope_correlation": 0.9846452948246364, "全帯域_10ms_modulation_mae_db": 1.5269397578709931, "全帯域_40ms_log_rms_mae_db": 1.4425048402634277, "全帯域_40ms_rms_envelope_nrmse": 0.2241784812383233, "全帯域_40ms_envelope_correlation": 0.9908847557469075, "全帯域_40ms_modulation_mae_db": 1.4263096012734044, "300–1000 Hz_10ms_log_rms_mae_db": 1.5677980942874388, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.29233899499352506, "300–1000 Hz_10ms_envelope_correlation": 0.9860953562671037, "300–1000 Hz_10ms_modulation_mae_db": 1.4604640268543314, "300–1000 Hz_40ms_log_rms_mae_db": 1.2857927447491846, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.27492134483614555, "300–1000 Hz_40ms_envelope_correlation": 0.9917969069827062, "300–1000 Hz_40ms_modulation_mae_db": 1.506903182132061, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4722367839999655, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2724137429387285, "1000–3000 Hz_10ms_envelope_correlation": 0.9761080711596366, "1000–3000 Hz_10ms_modulation_mae_db": 0.8572989123790636, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1579898002660707, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.2385027726624936, "1000–3000 Hz_40ms_envelope_correlation": 0.9857828680284175, "1000–3000 Hz_40ms_modulation_mae_db": 0.765486464486406, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6641417675659615, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22851776325390402, "3000–6000 Hz_10ms_envelope_correlation": 0.9665504127298079, "3000–6000 Hz_10ms_modulation_mae_db": 0.5976340629995192, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1343129482279941, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1546345841946546, "3000–6000 Hz_40ms_envelope_correlation": 0.9877980310861937, "3000–6000 Hz_40ms_modulation_mae_db": 0.7557811832917443, "mr_stft_db": 5.6671832941414175, "band_mae_db": 1.1039824425717997, "rms_gain_error_db": 1.5439480253991342, "output_peak": 0.6436996514869825}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.032713594001612, "spectral_convergence": 0.5129039588787699, "spectral_convergence_status": "ok", "amplitude_floor": 1.8586249352090494e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2848298861426315, "spectral_convergence": 0.3048909054121651, "spectral_convergence_status": "ok", "amplitude_floor": 1.6686015821705495e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.68400640228001, "spectral_convergence": 0.3332535065490331, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267872572435527e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003360135402914431, "output_power": 0.00035714484306984174, "delta_db": 0.2648760513953063, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011242144422279699, "output_power": 0.0013950122047450414, "delta_db": 0.9372884704161002, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015417458345753522, "output_power": 0.0025442564137720725, "delta_db": 2.1754809425508896, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.127065821964785e-05, "output_power": 0.00011725084505336701, "delta_db": 1.5918220429301544, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.295885949753132e-05, "output_power": 6.648075650981524e-05, "delta_db": 0.9875732907973067, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.948150516097124e-05, "output_power": 2.2714746459317572e-05, "delta_db": 0.6668538573410412, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.1039824425717997, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.5439480253991342, "pitch": null, "output_rms_dbfs": -23.465115739921607, "output_peak": 0.6436996514869825, "reference_rms_dbfs": -25.009063765320747, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6671832941414175}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025375764068356454, "log_rms_mae_db": 1.6994640221299797, "rms_envelope_nrmse": 0.24202111842227497, "envelope_correlation": 0.9846452948246364, "rms_gain_error_db": 1.5439529641018201, "reference_mean_power": 0.0031583120304531104, "output_mean_power": 0.004506613683441732}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1807904078754825e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008327873716353349, "output_power": 0.001267354310059727, "reference_active": true, "delta_db": 1.8236391520968533}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013733244323789595, "output_power": 0.0002190733485427226, "reference_active": true, "delta_db": 2.0281639998116603}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.747498790065721e-05, "output_power": 2.066895386578333e-05, "reference_active": true, "delta_db": 0.7290161217044655}], "active_band_count": 3, "active_band_mae_db": 1.5269397578709931}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002402786371599188, "log_rms_mae_db": 1.4425048402634277, "rms_envelope_nrmse": 0.2241784812383233, "envelope_correlation": 0.9908847557469075, "rms_gain_error_db": 1.5439659224506517, "reference_mean_power": 0.0031741738524038612, "output_mean_power": 0.004529260523077031}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0384864838590044e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000782161356346486, "output_power": 0.0012045974508853204, "reference_active": true, "delta_db": 1.8754558451345122}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.85949543312588e-05, "output_power": 9.626014336251072e-05, "reference_active": true, "delta_db": 2.1558628404067743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3154862476080315e-06, "output_power": 1.392667125664405e-06, "reference_active": true, "delta_db": 0.24761011827892696}], "active_band_count": 3, "active_band_mae_db": 1.4263096012734044}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002501327279402343, "log_rms_mae_db": 1.5677980942874388, "rms_envelope_nrmse": 0.29233899499352506, "envelope_correlation": 0.9860953562671037, "rms_gain_error_db": 1.9210182486064338, "reference_mean_power": 0.0021283926153372414, "output_mean_power": 0.0033124823149366755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0698576130619868e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007463968445819582, "output_power": 0.0012056377731433918, "reference_active": true, "delta_db": 2.0824705140263258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014378870760237378, "output_power": 0.00024176630575529148, "reference_active": true, "delta_db": 2.256709942075145}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4173849116031066e-05, "output_power": 1.4036752514893034e-05, "reference_active": true, "delta_db": -0.04221162446152428}], "active_band_count": 3, "active_band_mae_db": 1.4604640268543314}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00023719519351520602, "log_rms_mae_db": 1.2857927447491846, "rms_envelope_nrmse": 0.27492134483614555, "envelope_correlation": 0.9917969069827062, "rms_gain_error_db": 1.9210185439123717, "reference_mean_power": 0.0021390960955203144, "output_mean_power": 0.003329140693830073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.624199297765067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007211274534636877, "output_power": 0.0011750308734192378, "reference_active": true, "delta_db": 2.1203724810229123}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.136224424017926e-05, "output_power": 0.00010515055090276065, "reference_active": true, "delta_db": 2.339103174174299}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3649293472151981e-06, "output_power": 1.3843106519663719e-06, "reference_active": true, "delta_db": 0.06123389119897228}], "active_band_count": 3, "active_band_mae_db": 1.506903182132061}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.486004098842285e-05, "log_rms_mae_db": 1.4722367839999655, "rms_envelope_nrmse": 0.2724137429387285, "envelope_correlation": 0.9761080711596366, "rms_gain_error_db": 1.4776103608143156, "reference_mean_power": 9.73047769043704e-05, "output_mean_power": 0.00013673988091394022}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.517204404381221e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1360183676562794e-05, "output_power": 5.333146429837738e-05, "reference_active": true, "delta_db": 1.1040104932785213}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.821587959830417e-06, "output_power": 6.424213987109214e-06, "reference_active": true, "delta_db": 1.2462990465889472}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2148111012630244e-06, "output_power": 2.330748489189884e-06, "reference_active": true, "delta_db": 0.22158719726972287}], "active_band_count": 3, "active_band_mae_db": 0.8572989123790636}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5230962699738786e-05, "log_rms_mae_db": 1.1579898002660707, "rms_envelope_nrmse": 0.2385027726624936, "envelope_correlation": 0.9857828680284175, "rms_gain_error_db": 1.4776853278442847, "reference_mean_power": 9.77871242521371e-05, "output_mean_power": 0.0001374200832367084}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.922472441667658e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.88087102342988e-05, "output_power": 5.040438001373011e-05, "reference_active": true, "delta_db": 1.1353906751215224}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.146701619692133e-06, "output_power": 2.776473175244838e-06, "reference_active": true, "delta_db": 1.1172179794727908}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.612101873452723e-07, "output_power": 1.595906372174667e-07, "reference_active": true, "delta_db": -0.04385073886490455}], "active_band_count": 3, "active_band_mae_db": 0.765486464486406}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.5953225979159985e-05, "log_rms_mae_db": 1.6641417675659615, "rms_envelope_nrmse": 0.22851776325390402, "envelope_correlation": 0.9665504127298079, "rms_gain_error_db": 0.8389914639047862, "reference_mean_power": 3.136903007980474e-05, "output_mean_power": 3.8053993271355325e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9944851157580815e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0899878382466362e-05, "output_power": 1.3038302450407868e-05, "reference_active": true, "delta_db": 0.7779939893860437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5519027988888207e-06, "output_power": 3.170276571653159e-06, "reference_active": true, "delta_db": 0.942330230376477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4088478818809383e-06, "output_power": 1.4325899505394474e-06, "reference_active": true, "delta_db": 0.07257796923603664}], "active_band_count": 3, "active_band_mae_db": 0.5976340629995192}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9738091980327894e-05, "log_rms_mae_db": 1.1343129482279941, "rms_envelope_nrmse": 0.1546345841946546, "envelope_correlation": 0.9877980310861937, "rms_gain_error_db": 0.8393545922748434, "reference_mean_power": 3.151908471872331e-05, "output_mean_power": 3.823922279764568e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7532611532165288e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0780259770731776e-05, "output_power": 1.2871106544677985e-05, "reference_active": true, "delta_db": 0.7698665921095292}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1761748042676738e-06, "output_power": 1.5264354320997111e-06, "reference_active": true, "delta_db": 1.132065666429754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.367404555815231e-08, "output_power": 8.611486947144397e-08, "reference_active": true, "delta_db": -0.3654112913359493}], "active_band_count": 3, "active_band_mae_db": 0.7557811832917443}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.943399820438044, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d6ad5245261100e7f05bda144fe49f1fe78e8c5c08d69fe0cf2046f0a603ffb6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9761080711596366, "standard_deviation": null, "minimum": 0.9761080711596366, "maximum": 0.9761080711596366, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4722367839999655, "standard_deviation": null, "minimum": 1.4722367839999655, "maximum": 1.4722367839999655, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8572989123790636, "standard_deviation": null, "minimum": 0.8572989123790636, "maximum": 0.8572989123790636, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2724137429387285, "standard_deviation": null, "minimum": 0.2724137429387285, "maximum": 0.2724137429387285, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9857828680284175, "standard_deviation": null, "minimum": 0.9857828680284175, "maximum": 0.9857828680284175, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1579898002660707, "standard_deviation": null, "minimum": 1.1579898002660707, "maximum": 1.1579898002660707, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.765486464486406, "standard_deviation": null, "minimum": 0.765486464486406, "maximum": 0.765486464486406, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2385027726624936, "standard_deviation": null, "minimum": 0.2385027726624936, "maximum": 0.2385027726624936, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9665504127298079, "standard_deviation": null, "minimum": 0.9665504127298079, "maximum": 0.9665504127298079, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6641417675659615, "standard_deviation": null, "minimum": 1.6641417675659615, "maximum": 1.6641417675659615, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5976340629995192, "standard_deviation": null, "minimum": 0.5976340629995192, "maximum": 0.5976340629995192, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22851776325390402, "standard_deviation": null, "minimum": 0.22851776325390402, "maximum": 0.22851776325390402, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9877980310861937, "standard_deviation": null, "minimum": 0.9877980310861937, "maximum": 0.9877980310861937, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1343129482279941, "standard_deviation": null, "minimum": 1.1343129482279941, "maximum": 1.1343129482279941, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7557811832917443, "standard_deviation": null, "minimum": 0.7557811832917443, "maximum": 0.7557811832917443, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1546345841946546, "standard_deviation": null, "minimum": 0.1546345841946546, "maximum": 0.1546345841946546, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9860953562671037, "standard_deviation": null, "minimum": 0.9860953562671037, "maximum": 0.9860953562671037, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5677980942874388, "standard_deviation": null, "minimum": 1.5677980942874388, "maximum": 1.5677980942874388, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4604640268543314, "standard_deviation": null, "minimum": 1.4604640268543314, "maximum": 1.4604640268543314, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29233899499352506, "standard_deviation": null, "minimum": 0.29233899499352506, "maximum": 0.29233899499352506, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9917969069827062, "standard_deviation": null, "minimum": 0.9917969069827062, "maximum": 0.9917969069827062, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2857927447491846, "standard_deviation": null, "minimum": 1.2857927447491846, "maximum": 1.2857927447491846, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.506903182132061, "standard_deviation": null, "minimum": 1.506903182132061, "maximum": 1.506903182132061, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.27492134483614555, "standard_deviation": null, "minimum": 0.27492134483614555, "maximum": 0.27492134483614555, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.1039824425717997, "standard_deviation": null, "minimum": 1.1039824425717997, "maximum": 1.1039824425717997, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.6671832941414175, "standard_deviation": null, "minimum": 5.6671832941414175, "maximum": 5.6671832941414175, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6436996514869825, "standard_deviation": null, "minimum": 0.6436996514869825, "maximum": 0.6436996514869825, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.5439480253991342, "standard_deviation": null, "minimum": 1.5439480253991342, "maximum": 1.5439480253991342, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9846452948246364, "standard_deviation": null, "minimum": 0.9846452948246364, "maximum": 0.9846452948246364, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6994640221299797, "standard_deviation": null, "minimum": 1.6994640221299797, "maximum": 1.6994640221299797, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.5269397578709931, "standard_deviation": null, "minimum": 1.5269397578709931, "maximum": 1.5269397578709931, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.24202111842227497, "standard_deviation": null, "minimum": 0.24202111842227497, "maximum": 0.24202111842227497, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9908847557469075, "standard_deviation": null, "minimum": 0.9908847557469075, "maximum": 0.9908847557469075, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.4425048402634277, "standard_deviation": null, "minimum": 1.4425048402634277, "maximum": 1.4425048402634277, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4263096012734044, "standard_deviation": null, "minimum": 1.4263096012734044, "maximum": 1.4263096012734044, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2241784812383233, "standard_deviation": null, "minimum": 0.2241784812383233, "maximum": 0.2241784812383233, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1284"}, {"case": "librispeech_1320-122617-0019", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3203516006469727, "predicted_mos": 3.3203516006469727, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fdaa4bc68a834b0a69f2317c1b3d512c6fbc4e50e44442de376bc20d3f88d47b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 12.565119999926537}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.361746882983392, "全帯域_10ms_rms_envelope_nrmse": 0.160742810082838, "全帯域_10ms_envelope_correlation": 0.9783332710586251, "全帯域_10ms_modulation_mae_db": 0.5212464796443887, "全帯域_40ms_log_rms_mae_db": 1.7779521793846464, "全帯域_40ms_rms_envelope_nrmse": 0.1095748351724602, "全帯域_40ms_envelope_correlation": 0.9928347420956058, "全帯域_40ms_modulation_mae_db": 0.6006579777054807, "300–1000 Hz_10ms_log_rms_mae_db": 2.09150440928117, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21114355374785868, "300–1000 Hz_10ms_envelope_correlation": 0.9803693023027688, "300–1000 Hz_10ms_modulation_mae_db": 1.0109943627045581, "300–1000 Hz_40ms_log_rms_mae_db": 1.5845730069736896, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.16987534471813417, "300–1000 Hz_40ms_envelope_correlation": 0.9904193765398958, "300–1000 Hz_40ms_modulation_mae_db": 1.1970716513039399, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8462574248029953, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22064231401028783, "1000–3000 Hz_10ms_envelope_correlation": 0.9783028166024328, "1000–3000 Hz_10ms_modulation_mae_db": 0.3763028476917906, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0966499017189242, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1737311116679123, "1000–3000 Hz_40ms_envelope_correlation": 0.9898170516008914, "1000–3000 Hz_40ms_modulation_mae_db": 0.35396117457180626, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0389142688849704, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28287801527568923, "3000–6000 Hz_10ms_envelope_correlation": 0.9514353824129493, "3000–6000 Hz_10ms_modulation_mae_db": 0.15819270282054662, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2051643045549156, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14663466879585701, "3000–6000 Hz_40ms_envelope_correlation": 0.9866662564232346, "3000–6000 Hz_40ms_modulation_mae_db": 0.22337868721604504, "mr_stft_db": 5.930269276559595, "band_mae_db": 0.5910199837436511, "rms_gain_error_db": 0.6496570686968655, "output_peak": 0.8589965646002072}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.1266953972411855, "spectral_convergence": 0.6772961608885302, "spectral_convergence_status": "ok", "amplitude_floor": 1.884454455950008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.258319739538613, "spectral_convergence": 0.24833245617000133, "spectral_convergence_status": "ok", "amplitude_floor": 1.301578596614512e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.405792692898985, "spectral_convergence": 0.24823199239401322, "spectral_convergence_status": "ok", "amplitude_floor": 1.1393421242566376e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004137534659655858, "output_power": 0.004643520187912807, "delta_db": 0.5010569316859783, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026137240204796635, "output_power": 0.003151507354886851, "delta_db": 0.8125859589637698, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008071793462917449, "output_power": 0.0009445387536237597, "delta_db": 0.6824974050021925, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004252507832785635, "output_power": 0.0005453036306304836, "delta_db": 1.0799326700655052, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.026398390024539e-05, "output_power": 7.355096597190815e-05, "delta_db": 0.19855610617135205, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.066875275629503e-05, "output_power": 5.393731171771874e-05, "delta_db": 0.2714908305731082, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5910199837436511, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6496570686968655, "pitch": null, "output_rms_dbfs": -20.26301553299747, "output_peak": 0.8589965646002072, "reference_rms_dbfs": -20.91267260169434, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.930269276559595}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025953969251226393, "log_rms_mae_db": 2.361746882983392, "rms_envelope_nrmse": 0.160742810082838, "envelope_correlation": 0.9783332710586251, "rms_gain_error_db": 0.6506350794598665, "reference_mean_power": 0.008109529837298932, "output_mean_power": 0.009420179623655555}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5635786081597607e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013500334014131179, "output_power": 0.001551238383869229, "reference_active": true, "delta_db": 0.6033402880986519}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020457869490256438, "output_power": 0.00018813617472483935, "reference_active": true, "delta_db": -0.36388094231478296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3880284077345985e-05, "output_power": 2.953216065149233e-05, "reference_active": true, "delta_db": -0.5965182085197315}], "active_band_count": 3, "active_band_mae_db": 0.5212464796443887}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024049824298809385, "log_rms_mae_db": 1.7779521793846464, "rms_envelope_nrmse": 0.1095748351724602, "envelope_correlation": 0.9928347420956058, "rms_gain_error_db": 0.6521896233197788, "reference_mean_power": 0.008139876536629037, "output_mean_power": 0.009458816058520284}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2818780696338306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012006574574251973, "output_power": 0.001407111674211938, "reference_active": true, "delta_db": 0.6890944354581042}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.159020687440259e-05, "output_power": 8.649881526669713e-05, "reference_active": true, "delta_db": -0.24838880778759811}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2353163446294293e-06, "output_power": 1.831850066949306e-06, "reference_active": true, "delta_db": -0.8644906898707398}], "active_band_count": 3, "active_band_mae_db": 0.6006579777054807}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002074995834794702, "log_rms_mae_db": 2.09150440928117, "rms_envelope_nrmse": 0.21114355374785868, "envelope_correlation": 0.9803693023027688, "rms_gain_error_db": 1.0256116968777245, "reference_mean_power": 0.0020359851811281022, "output_mean_power": 0.002578313854933067}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.345568889858965e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045532158312677247, "output_power": 0.0005746996220844205, "reference_active": true, "delta_db": 1.0112267434946238}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.561332252218219e-05, "output_power": 0.00011078375995347112, "reference_active": true, "delta_db": 1.1193474914733388}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0669601144719856e-05, "output_power": 8.66776841698246e-06, "reference_active": true, "delta_db": -0.9024088531457115}], "active_band_count": 3, "active_band_mae_db": 1.0109943627045581}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001833599147850085, "log_rms_mae_db": 1.5845730069736896, "rms_envelope_nrmse": 0.16987534471813417, "envelope_correlation": 0.9904193765398958, "rms_gain_error_db": 1.0262179373628464, "reference_mean_power": 0.0020457391187242045, "output_mean_power": 0.002591027626265494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.75393473065512e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00043556686984963635, "output_power": 0.0005494862384369671, "reference_active": true, "delta_db": 1.0090198127301224}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.69121464262054e-05, "output_power": 5.030149545748529e-05, "reference_active": true, "delta_db": 1.344115967949287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.183033529622033e-07, "output_power": 5.401320853565532e-07, "reference_active": true, "delta_db": -1.2380791732324106}], "active_band_count": 3, "active_band_mae_db": 1.1970716513039399}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011093565856595723, "log_rms_mae_db": 1.8462574248029953, "rms_envelope_nrmse": 0.22064231401028783, "envelope_correlation": 0.9783028166024328, "rms_gain_error_db": 1.0052236374939616, "reference_mean_power": 0.00042557205365836593, "output_mean_power": 0.0005364082699298417}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.99964400169894e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011758808293094384, "output_power": 0.00013683247719439913, "reference_active": true, "delta_db": 0.6582587942417067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1475307548341578e-05, "output_power": 2.328546698716099e-05, "reference_active": true, "delta_db": 0.3514555998978061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0561557218273926e-06, "output_power": 4.91927414802455e-06, "reference_active": true, "delta_db": -0.11919414893585897}], "active_band_count": 3, "active_band_mae_db": 0.3763028476917906}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010031642073947198, "log_rms_mae_db": 1.0966499017189242, "rms_envelope_nrmse": 0.1737311116679123, "envelope_correlation": 0.9898170516008914, "rms_gain_error_db": 1.0053148779504404, "reference_mean_power": 0.0004276640413209328, "output_mean_power": 0.000539056420851786}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8247169971067077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011139646210262638, "output_power": 0.00013061014629697437, "reference_active": true, "delta_db": 0.691055178346261}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0378224135538677e-05, "output_power": 1.1068785433105939e-05, "reference_active": true, "delta_db": 0.27976922994575765}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.220864105203889e-07, "output_power": 3.154034866061741e-07, "reference_active": true, "delta_db": -0.09105911542340026}], "active_band_count": 3, "active_band_mae_db": 0.35396117457180626}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.985234496629778e-05, "log_rms_mae_db": 2.0389142688849704, "rms_envelope_nrmse": 0.28287801527568923, "envelope_correlation": 0.9514353824129493, "rms_gain_error_db": 0.22894460991511764, "reference_mean_power": 6.361685525644655e-05, "output_mean_power": 6.706048005361849e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.697638884988847e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.1912782580556425e-06, "output_power": 6.362468623136165e-06, "reference_active": true, "delta_db": 0.11845330413456648}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.016318108845029e-06, "output_power": 2.074888023694569e-06, "reference_active": true, "delta_db": 0.12435613338250284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.069070181696055e-07, "output_power": 6.401754666469084e-07, "reference_active": true, "delta_db": 0.23176867094457054}], "active_band_count": 3, "active_band_mae_db": 0.15819270282054662}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.357134521214995e-05, "log_rms_mae_db": 1.2051643045549156, "rms_envelope_nrmse": 0.14663466879585701, "envelope_correlation": 0.9866662564232346, "rms_gain_error_db": 0.22919562449053832, "reference_mean_power": 6.39232761364752e-05, "output_mean_power": 6.738738246952957e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.609918878818377e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.077093160223491e-06, "output_power": 6.2918815111682235e-06, "reference_active": true, "delta_db": 0.15084641368986162}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.511523340291838e-07, "output_power": 9.944385176687978e-07, "reference_active": true, "delta_db": 0.19327859565648736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7938163224025654e-08, "output_power": 3.0116115603039055e-08, "reference_active": true, "delta_db": 0.32601105230178606}], "active_band_count": 3, "active_band_mae_db": 0.22337868721604504}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9358777958428688, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "461ba0d3dad7fdb226362512ffdce6700a71ff51a82e4c809f2a05d30daf63c8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9783028166024328, "standard_deviation": null, "minimum": 0.9783028166024328, "maximum": 0.9783028166024328, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8462574248029953, "standard_deviation": null, "minimum": 1.8462574248029953, "maximum": 1.8462574248029953, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3763028476917906, "standard_deviation": null, "minimum": 0.3763028476917906, "maximum": 0.3763028476917906, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22064231401028783, "standard_deviation": null, "minimum": 0.22064231401028783, "maximum": 0.22064231401028783, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9898170516008914, "standard_deviation": null, "minimum": 0.9898170516008914, "maximum": 0.9898170516008914, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0966499017189242, "standard_deviation": null, "minimum": 1.0966499017189242, "maximum": 1.0966499017189242, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.35396117457180626, "standard_deviation": null, "minimum": 0.35396117457180626, "maximum": 0.35396117457180626, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1737311116679123, "standard_deviation": null, "minimum": 0.1737311116679123, "maximum": 0.1737311116679123, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9514353824129493, "standard_deviation": null, "minimum": 0.9514353824129493, "maximum": 0.9514353824129493, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0389142688849704, "standard_deviation": null, "minimum": 2.0389142688849704, "maximum": 2.0389142688849704, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.15819270282054662, "standard_deviation": null, "minimum": 0.15819270282054662, "maximum": 0.15819270282054662, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.28287801527568923, "standard_deviation": null, "minimum": 0.28287801527568923, "maximum": 0.28287801527568923, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9866662564232346, "standard_deviation": null, "minimum": 0.9866662564232346, "maximum": 0.9866662564232346, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2051643045549156, "standard_deviation": null, "minimum": 1.2051643045549156, "maximum": 1.2051643045549156, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.22337868721604504, "standard_deviation": null, "minimum": 0.22337868721604504, "maximum": 0.22337868721604504, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14663466879585701, "standard_deviation": null, "minimum": 0.14663466879585701, "maximum": 0.14663466879585701, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9803693023027688, "standard_deviation": null, "minimum": 0.9803693023027688, "maximum": 0.9803693023027688, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.09150440928117, "standard_deviation": null, "minimum": 2.09150440928117, "maximum": 2.09150440928117, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0109943627045581, "standard_deviation": null, "minimum": 1.0109943627045581, "maximum": 1.0109943627045581, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21114355374785868, "standard_deviation": null, "minimum": 0.21114355374785868, "maximum": 0.21114355374785868, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9904193765398958, "standard_deviation": null, "minimum": 0.9904193765398958, "maximum": 0.9904193765398958, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5845730069736896, "standard_deviation": null, "minimum": 1.5845730069736896, "maximum": 1.5845730069736896, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1970716513039399, "standard_deviation": null, "minimum": 1.1970716513039399, "maximum": 1.1970716513039399, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16987534471813417, "standard_deviation": null, "minimum": 0.16987534471813417, "maximum": 0.16987534471813417, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5910199837436511, "standard_deviation": null, "minimum": 0.5910199837436511, "maximum": 0.5910199837436511, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.930269276559595, "standard_deviation": null, "minimum": 5.930269276559595, "maximum": 5.930269276559595, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.8589965646002072, "standard_deviation": null, "minimum": 0.8589965646002072, "maximum": 0.8589965646002072, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6496570686968655, "standard_deviation": null, "minimum": 0.6496570686968655, "maximum": 0.6496570686968655, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9783332710586251, "standard_deviation": null, "minimum": 0.9783332710586251, "maximum": 0.9783332710586251, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.361746882983392, "standard_deviation": null, "minimum": 2.361746882983392, "maximum": 2.361746882983392, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5212464796443887, "standard_deviation": null, "minimum": 0.5212464796443887, "maximum": 0.5212464796443887, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.160742810082838, "standard_deviation": null, "minimum": 0.160742810082838, "maximum": 0.160742810082838, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9928347420956058, "standard_deviation": null, "minimum": 0.9928347420956058, "maximum": 0.9928347420956058, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.7779521793846464, "standard_deviation": null, "minimum": 1.7779521793846464, "maximum": 1.7779521793846464, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6006579777054807, "standard_deviation": null, "minimum": 0.6006579777054807, "maximum": 0.6006579777054807, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1095748351724602, "standard_deviation": null, "minimum": 0.1095748351724602, "maximum": 0.1095748351724602, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1320"}, {"case": "librispeech_1320-122617-0023", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.698338270187378, "predicted_mos": 3.698338270187378, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f21a786c5937fd008be43469e33657e53f6dd7fa3c236ef6cf7a2dff474717b4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 11.926399799995124}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.922913810526294, "全帯域_10ms_rms_envelope_nrmse": 0.1567807186995818, "全帯域_10ms_envelope_correlation": 0.9790240060184419, "全帯域_10ms_modulation_mae_db": 0.5092525218400649, "全帯域_40ms_log_rms_mae_db": 1.3736223645969803, "全帯域_40ms_rms_envelope_nrmse": 0.09877041767924061, "全帯域_40ms_envelope_correlation": 0.9952215394326875, "全帯域_40ms_modulation_mae_db": 0.43503497181173706, "300–1000 Hz_10ms_log_rms_mae_db": 2.1513014744790078, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21173643085683766, "300–1000 Hz_10ms_envelope_correlation": 0.9799732927743583, "300–1000 Hz_10ms_modulation_mae_db": 0.9225871348157774, "300–1000 Hz_40ms_log_rms_mae_db": 1.6793140404942428, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.16011615741209073, "300–1000 Hz_40ms_envelope_correlation": 0.994397724156473, "300–1000 Hz_40ms_modulation_mae_db": 1.1102564707196736, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7094297040078181, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20623542443048795, "1000–3000 Hz_10ms_envelope_correlation": 0.9748362661572422, "1000–3000 Hz_10ms_modulation_mae_db": 0.5990296056350156, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1193481966712657, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1061059988347249, "1000–3000 Hz_40ms_envelope_correlation": 0.9958398390788004, "1000–3000 Hz_40ms_modulation_mae_db": 0.34334573091279114, "3000–6000 Hz_10ms_log_rms_mae_db": 2.1068978234382887, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2177640608590946, "3000–6000 Hz_10ms_envelope_correlation": 0.9691908001800987, "3000–6000 Hz_10ms_modulation_mae_db": 0.7508224244609923, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2996789054239655, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12906556533673266, "3000–6000 Hz_40ms_envelope_correlation": 0.9887606446131555, "3000–6000 Hz_40ms_modulation_mae_db": 0.8631952678197804, "mr_stft_db": 6.0682677059269245, "band_mae_db": 0.4803067428900068, "rms_gain_error_db": 0.6286516417595468, "output_peak": 0.7840552585481659}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.47311794459842, "spectral_convergence": 0.7210638225824418, "spectral_convergence_status": "ok", "amplitude_floor": 1.2959192499433599e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2257585211502375, "spectral_convergence": 0.26368516672625736, "spectral_convergence_status": "ok", "amplitude_floor": 8.646137696849433e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.505926652032115, "spectral_convergence": 0.22173512944386803, "spectral_convergence_status": "ok", "amplitude_floor": 6.578152434885499e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0020500598531285057, "output_power": 0.0021589003295964825, "delta_db": 0.22466051818983068, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0025325885847670413, "output_power": 0.0030583963045236404, "delta_db": 0.8192911522057044, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009745321457083826, "output_power": 0.0012155323719559804, "delta_db": 0.9597036019294982, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029510461183058167, "output_power": 0.0003430739348478291, "delta_db": 0.6541172703452367, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.807836063095196e-05, "output_power": 3.8652205664673966e-05, "delta_db": 0.06496039065830483, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.519436277144243e-05, "output_power": 3.3928319866917165e-05, "delta_db": -0.15910752401146633, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4803067428900068, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6286516417595468, "pitch": null, "output_rms_dbfs": -21.64405588371444, "output_peak": 0.7840552585481659, "reference_rms_dbfs": -22.27270752547399, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0682677059269245}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019946925088635918, "log_rms_mae_db": 1.922913810526294, "rms_envelope_nrmse": 0.1567807186995818, "envelope_correlation": 0.9790240060184419, "rms_gain_error_db": 0.6342607891509657, "reference_mean_power": 0.0059224241180442505, "output_mean_power": 0.006853707784883978}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0772791897966612e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010245127053592495, "output_power": 0.0011758991722086501, "reference_active": true, "delta_db": 0.5985273602281199}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001329567656774176, "output_power": 0.00014672213504206597, "reference_active": true, "delta_db": 0.4278519618451574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.301896656350168e-05, "output_power": 2.583590306513087e-05, "reference_active": true, "delta_db": 0.5013782434469171}], "active_band_count": 3, "active_band_mae_db": 0.5092525218400649}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001930447992601615, "log_rms_mae_db": 1.3736223645969803, "rms_envelope_nrmse": 0.09877041767924061, "envelope_correlation": 0.9952215394326875, "rms_gain_error_db": 0.6360425085755982, "reference_mean_power": 0.005919423744261807, "output_mean_power": 0.006853046538426588}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.875987153432414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009473592447040891, "output_power": 0.0010987793024202098, "reference_active": true, "delta_db": 0.6439577257970902}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.123049684235515e-05, "output_power": 6.942030130188768e-05, "reference_active": true, "delta_db": 0.5451871084490477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.329909700774956e-06, "output_power": 1.2948699305038815e-06, "reference_active": true, "delta_db": -0.11596008118907333}], "active_band_count": 3, "active_band_mae_db": 0.43503497181173706}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015110036587973006, "log_rms_mae_db": 2.1513014744790078, "rms_envelope_nrmse": 0.21173643085683766, "envelope_correlation": 0.9799732927743583, "rms_gain_error_db": 1.1120381420907623, "reference_mean_power": 0.0025365732398871734, "output_mean_power": 0.003276809700834761}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.31579034521966e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006679887319597139, "output_power": 0.0008773244182273111, "reference_active": true, "delta_db": 1.1839108048137241}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.085773760663629e-05, "output_power": 0.00011389636840466793, "reference_active": true, "delta_db": 0.9814795827358974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6589878271689645e-05, "output_power": 1.4441293605613923e-05, "reference_active": true, "delta_db": -0.6023710168977104}], "active_band_count": 3, "active_band_mae_db": 0.9225871348157774}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013775289901134618, "log_rms_mae_db": 1.6793140404942428, "rms_envelope_nrmse": 0.16011615741209073, "envelope_correlation": 0.994397724156473, "rms_gain_error_db": 1.115254652841213, "reference_mean_power": 0.0025400163932975937, "output_mean_power": 0.003283688748449607}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2092894134690622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006291541070562532, "output_power": 0.0008290213731240054, "reference_active": true, "delta_db": 1.1980869132499834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.975083640469782e-05, "output_power": 5.107284463324335e-05, "reference_active": true, "delta_db": 1.0884377651658272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.7844516380393e-07, "output_power": 7.69328847741602e-07, "reference_active": true, "delta_db": -1.0442447337432104}], "active_band_count": 3, "active_band_mae_db": 1.1102564707196736}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010600794217574433, "log_rms_mae_db": 1.7094297040078181, "rms_envelope_nrmse": 0.20623542443048795, "envelope_correlation": 0.9748362661572422, "rms_gain_error_db": 0.6158795797219442, "reference_mean_power": 0.0002691478602162077, "output_mean_power": 0.00031015507326972997}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.02821083589085e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010735336162950012, "output_power": 0.00011888378346641469, "reference_active": true, "delta_db": 0.4430696967540712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.315168535027072e-06, "output_power": 8.536141617037778e-06, "reference_active": true, "delta_db": 0.1139055545422295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.408982675283079e-06, "output_power": 3.2051254543279295e-06, "reference_active": true, "delta_db": 1.240113565608746}], "active_band_count": 3, "active_band_mae_db": 0.5990296056350156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.461607013685591e-05, "log_rms_mae_db": 1.1193481966712657, "rms_envelope_nrmse": 0.1061059988347249, "envelope_correlation": 0.9958398390788004, "rms_gain_error_db": 0.6168920016872033, "reference_mean_power": 0.00026921330106685756, "output_mean_power": 0.0003103028136307839}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9187018796507895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.751052870210862e-05, "output_power": 0.00010845930912219, "reference_active": true, "delta_db": 0.4621532249500468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.818200149529557e-06, "output_power": 3.968680241857808e-06, "reference_active": true, "delta_db": 0.16787418726256254}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4850622077020166e-07, "output_power": 1.6283419977638216e-07, "reference_active": true, "delta_db": 0.4000097805257641}], "active_band_count": 3, "active_band_mae_db": 0.34334573091279114}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.633594187238404e-05, "log_rms_mae_db": 2.1068978234382887, "rms_envelope_nrmse": 0.2177640608590946, "envelope_correlation": 0.9691908001800987, "rms_gain_error_db": -0.08537019404802997, "reference_mean_power": 4.9508565949953196e-05, "output_mean_power": 4.854486832939557e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.164436282244953e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0281496224399232e-05, "output_power": 9.056270261857313e-06, "reference_active": true, "delta_db": -0.5510694589494026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.187880808785332e-06, "output_power": 3.1247902371547802e-06, "reference_active": true, "delta_db": -1.271734436628946}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.689088055617476e-07, "output_power": 6.280713651180742e-07, "reference_active": true, "delta_db": 0.429663377804628}], "active_band_count": 3, "active_band_mae_db": 0.7508224244609923}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.3561893402191595e-05, "log_rms_mae_db": 1.2996789054239655, "rms_envelope_nrmse": 0.12906556533673266, "envelope_correlation": 0.9887606446131555, "rms_gain_error_db": -0.08545774028747449, "reference_mean_power": 4.9751584691373346e-05, "output_mean_power": 4.87821732713502e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0334524009764534e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.035252748280077e-05, "output_power": 9.178081046128499e-06, "reference_active": true, "delta_db": -0.522945037256014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2970706912154704e-06, "output_power": 1.6187497802048789e-06, "reference_active": true, "delta_db": -1.5199463814228262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.66930165214842e-08, "output_power": 5.2956755599854956e-08, "reference_active": true, "delta_db": 0.5466943847805013}], "active_band_count": 3, "active_band_mae_db": 0.8631952678197804}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9240907489821762, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5348ff1144d8a07c022186b269eba4480790fe03b76e55c47396489dbe687987", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9748362661572422, "standard_deviation": null, "minimum": 0.9748362661572422, "maximum": 0.9748362661572422, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7094297040078181, "standard_deviation": null, "minimum": 1.7094297040078181, "maximum": 1.7094297040078181, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5990296056350156, "standard_deviation": null, "minimum": 0.5990296056350156, "maximum": 0.5990296056350156, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20623542443048795, "standard_deviation": null, "minimum": 0.20623542443048795, "maximum": 0.20623542443048795, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9958398390788004, "standard_deviation": null, "minimum": 0.9958398390788004, "maximum": 0.9958398390788004, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1193481966712657, "standard_deviation": null, "minimum": 1.1193481966712657, "maximum": 1.1193481966712657, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.34334573091279114, "standard_deviation": null, "minimum": 0.34334573091279114, "maximum": 0.34334573091279114, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1061059988347249, "standard_deviation": null, "minimum": 0.1061059988347249, "maximum": 0.1061059988347249, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9691908001800987, "standard_deviation": null, "minimum": 0.9691908001800987, "maximum": 0.9691908001800987, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.1068978234382887, "standard_deviation": null, "minimum": 2.1068978234382887, "maximum": 2.1068978234382887, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7508224244609923, "standard_deviation": null, "minimum": 0.7508224244609923, "maximum": 0.7508224244609923, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2177640608590946, "standard_deviation": null, "minimum": 0.2177640608590946, "maximum": 0.2177640608590946, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9887606446131555, "standard_deviation": null, "minimum": 0.9887606446131555, "maximum": 0.9887606446131555, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2996789054239655, "standard_deviation": null, "minimum": 1.2996789054239655, "maximum": 1.2996789054239655, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8631952678197804, "standard_deviation": null, "minimum": 0.8631952678197804, "maximum": 0.8631952678197804, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12906556533673266, "standard_deviation": null, "minimum": 0.12906556533673266, "maximum": 0.12906556533673266, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9799732927743583, "standard_deviation": null, "minimum": 0.9799732927743583, "maximum": 0.9799732927743583, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1513014744790078, "standard_deviation": null, "minimum": 2.1513014744790078, "maximum": 2.1513014744790078, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9225871348157774, "standard_deviation": null, "minimum": 0.9225871348157774, "maximum": 0.9225871348157774, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21173643085683766, "standard_deviation": null, "minimum": 0.21173643085683766, "maximum": 0.21173643085683766, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.994397724156473, "standard_deviation": null, "minimum": 0.994397724156473, "maximum": 0.994397724156473, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6793140404942428, "standard_deviation": null, "minimum": 1.6793140404942428, "maximum": 1.6793140404942428, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1102564707196736, "standard_deviation": null, "minimum": 1.1102564707196736, "maximum": 1.1102564707196736, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16011615741209073, "standard_deviation": null, "minimum": 0.16011615741209073, "maximum": 0.16011615741209073, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.4803067428900068, "standard_deviation": null, "minimum": 0.4803067428900068, "maximum": 0.4803067428900068, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.0682677059269245, "standard_deviation": null, "minimum": 6.0682677059269245, "maximum": 6.0682677059269245, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7840552585481659, "standard_deviation": null, "minimum": 0.7840552585481659, "maximum": 0.7840552585481659, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6286516417595468, "standard_deviation": null, "minimum": 0.6286516417595468, "maximum": 0.6286516417595468, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9790240060184419, "standard_deviation": null, "minimum": 0.9790240060184419, "maximum": 0.9790240060184419, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.922913810526294, "standard_deviation": null, "minimum": 1.922913810526294, "maximum": 1.922913810526294, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5092525218400649, "standard_deviation": null, "minimum": 0.5092525218400649, "maximum": 0.5092525218400649, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1567807186995818, "standard_deviation": null, "minimum": 0.1567807186995818, "maximum": 0.1567807186995818, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9952215394326875, "standard_deviation": null, "minimum": 0.9952215394326875, "maximum": 0.9952215394326875, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.3736223645969803, "standard_deviation": null, "minimum": 1.3736223645969803, "maximum": 1.3736223645969803, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43503497181173706, "standard_deviation": null, "minimum": 0.43503497181173706, "maximum": 0.43503497181173706, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09877041767924061, "standard_deviation": null, "minimum": 0.09877041767924061, "maximum": 0.09877041767924061, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1320"}, {"case": "librispeech_1320-122617-0039", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.0912604331970215, "predicted_mos": 3.0912604331970215, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "02045fc6e7cc4d9e81dbcb776e72eda1c59bb6099a99c31c65a59253a28b0f94", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 11.651091800071299}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7536569134247086, "全帯域_10ms_rms_envelope_nrmse": 0.16425446164976118, "全帯域_10ms_envelope_correlation": 0.9764993001596283, "全帯域_10ms_modulation_mae_db": 0.470646844726337, "全帯域_40ms_log_rms_mae_db": 1.2468848396864727, "全帯域_40ms_rms_envelope_nrmse": 0.11619726816366575, "全帯域_40ms_envelope_correlation": 0.9921412105386086, "全帯域_40ms_modulation_mae_db": 0.607076227717365, "300–1000 Hz_10ms_log_rms_mae_db": 2.077119811413454, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2448494869832453, "300–1000 Hz_10ms_envelope_correlation": 0.9738535514368375, "300–1000 Hz_10ms_modulation_mae_db": 0.987410825008757, "300–1000 Hz_40ms_log_rms_mae_db": 1.5698725991314821, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1989450674384473, "300–1000 Hz_40ms_envelope_correlation": 0.9869259959279382, "300–1000 Hz_40ms_modulation_mae_db": 1.143945095540544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7041409202837523, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21545180403600678, "1000–3000 Hz_10ms_envelope_correlation": 0.9650340174823716, "1000–3000 Hz_10ms_modulation_mae_db": 0.6988580828451217, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0647956389573765, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12764672418557468, "1000–3000 Hz_40ms_envelope_correlation": 0.9893204928973601, "1000–3000 Hz_40ms_modulation_mae_db": 0.6575900181208152, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9622863929731906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21950809498142287, "3000–6000 Hz_10ms_envelope_correlation": 0.9725826735240011, "3000–6000 Hz_10ms_modulation_mae_db": 1.2445312991173927, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2626837160298228, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13386119831958007, "3000–6000 Hz_40ms_envelope_correlation": 0.9916422908809223, "3000–6000 Hz_40ms_modulation_mae_db": 1.0490633639707223, "mr_stft_db": 5.95010534216549, "band_mae_db": 0.6735209215691184, "rms_gain_error_db": 0.7294971335797378, "output_peak": 0.7602949050445598}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.2259581921726745, "spectral_convergence": 0.7175664517919642, "spectral_convergence_status": "ok", "amplitude_floor": 2.2281588625700377e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.151061148304206, "spectral_convergence": 0.26129928078168974, "spectral_convergence_status": "ok", "amplitude_floor": 1.3568197926870596e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.473296686019589, "spectral_convergence": 0.25009995093246923, "spectral_convergence_status": "ok", "amplitude_floor": 1.1580179000740219e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0038028140043242973, "output_power": 0.004307004890952851, "delta_db": 0.5407028035640455, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002493613270814922, "output_power": 0.0031469089827399046, "delta_db": 1.0105508154815093, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006858506604602334, "output_power": 0.0008252007961354225, "delta_db": 0.8033007710925937, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002628229680145485, "output_power": 0.00029804616746442194, "delta_db": 0.5462022626595555, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001273968713381539, "output_power": 0.00014675585397207528, "delta_db": 0.6143667136234814, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.457810201861998e-05, "output_power": 6.160538061143509e-05, "delta_db": 0.5260021629935243, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6735209215691184, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7294971335797378, "pitch": null, "output_rms_dbfs": -20.56232425836516, "output_peak": 0.7602949050445598, "reference_rms_dbfs": -21.2918213919449, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.95010534216549}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00025738076998093343, "log_rms_mae_db": 1.7536569134247086, "rms_envelope_nrmse": 0.16425446164976118, "envelope_correlation": 0.9764993001596283, "rms_gain_error_db": 0.7298625580027376, "reference_mean_power": 0.007432639167823318, "output_mean_power": 0.008792842731082917}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.369590439777949e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014473681837183883, "output_power": 0.001841949705711431, "reference_active": true, "delta_db": 1.0469874598343638}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002579988221497507, "output_power": 0.00023995201295182197, "reference_active": true, "delta_db": -0.3149332569934129}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2677922934851034e-05, "output_power": 3.230371443834799e-05, "reference_active": true, "delta_db": -0.050019817351234314}], "active_band_count": 3, "active_band_mae_db": 0.470646844726337}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022253035095363898, "log_rms_mae_db": 1.2468848396864727, "rms_envelope_nrmse": 0.11619726816366575, "envelope_correlation": 0.9921412105386086, "rms_gain_error_db": 0.7300396037194241, "reference_mean_power": 0.0074694638279480755, "output_mean_power": 0.008836766690381893}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.0676309697735546e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013241614802002638, "output_power": 0.0016927960382042266, "reference_active": true, "delta_db": 1.066636838181482}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011975787910962369, "output_power": 0.00011152638543943777, "reference_active": true, "delta_db": -0.30926469144668794}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4293821094800366e-06, "output_power": 2.1926186992422452e-06, "reference_active": true, "delta_db": -0.44532715352392527}], "active_band_count": 3, "active_band_mae_db": 0.607076227717365}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017894181312046162, "log_rms_mae_db": 2.077119811413454, "rms_envelope_nrmse": 0.2448494869832453, "envelope_correlation": 0.9738535514368375, "rms_gain_error_db": 1.2044232232983538, "reference_mean_power": 0.0017811195275832714, "output_mean_power": 0.0023503654122483263}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1758328243827778e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007391946250824912, "output_power": 0.0010940107968348096, "reference_active": true, "delta_db": 1.7026280775634688}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015375544244981539, "output_power": 0.00018727900911110856, "reference_active": true, "delta_db": 0.8565860550051619}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0532547306567537e-05, "output_power": 9.59912360913565e-06, "reference_active": true, "delta_db": -0.4030183424576403}], "active_band_count": 3, "active_band_mae_db": 0.987410825008757}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016320284141686103, "log_rms_mae_db": 1.5698725991314821, "rms_envelope_nrmse": 0.1989450674384473, "envelope_correlation": 0.9869259959279382, "rms_gain_error_db": 1.2044966436175633, "reference_mean_power": 0.0017900337371132, "output_mean_power": 0.002362168537227335}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0791360125207298e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007228891058818755, "output_power": 0.0010658608918926744, "reference_active": true, "delta_db": 1.6862884768511015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980895739386466e-05, "output_power": 8.429737278496125e-05, "reference_active": true, "delta_db": 0.8190288778923672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.398569514368023e-07, "output_power": 7.592939941949427e-07, "reference_active": true, "delta_db": -0.926517931878163}], "active_band_count": 3, "active_band_mae_db": 1.143945095540544}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.442874453255196e-05, "log_rms_mae_db": 1.7041409202837523, "rms_envelope_nrmse": 0.21545180403600678, "envelope_correlation": 0.9650340174823716, "rms_gain_error_db": 0.5864341680892613, "reference_mean_power": 0.000318038298123639, "output_mean_power": 0.0003640179818946643}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6089659140343417e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.507437744605689e-05, "output_power": 0.00010307300199640739, "reference_active": true, "delta_db": 0.8334614532052164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3978123456671057e-05, "output_power": 2.8288802574032292e-05, "reference_active": true, "delta_db": 0.7179937283217606}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.6559576292684446e-06, "output_power": 5.2786264081659725e-06, "reference_active": true, "delta_db": 0.5451190670083883}], "active_band_count": 3, "active_band_mae_db": 0.6988580828451217}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.707405683354541e-05, "log_rms_mae_db": 1.0647956389573765, "rms_envelope_nrmse": 0.12764672418557468, "envelope_correlation": 0.9893204928973601, "rms_gain_error_db": 0.5874001531895992, "reference_mean_power": 0.00031952379334241086, "output_mean_power": 0.00036579959369847855}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4169476048907225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.318339905927379e-05, "output_power": 0.0001020686374142839, "reference_active": true, "delta_db": 0.8885565447550374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1060276940136043e-05, "output_power": 1.3236528618580114e-05, "reference_active": true, "delta_db": 0.7800810149117268}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55972585077424e-07, "output_power": 2.7454067927824487e-07, "reference_active": true, "delta_db": 0.30413249469568143}], "active_band_count": 3, "active_band_mae_db": 0.6575900181208152}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.0412968032519945e-05, "log_rms_mae_db": 1.9622863929731906, "rms_envelope_nrmse": 0.21950809498142287, "envelope_correlation": 0.9725826735240011, "rms_gain_error_db": 0.6210288150169984, "reference_mean_power": 8.301357548516487e-05, "output_mean_power": 9.577496485780419e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.362198065588504e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.2275109921127774e-05, "output_power": 3.840945911473258e-05, "reference_active": true, "delta_db": 0.7557046139471013}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8777051155174143e-06, "output_power": 2.8355678768698685e-06, "reference_active": true, "delta_db": -0.06406241071937915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.623133887825791e-07, "output_power": 1.2955262638490423e-06, "reference_active": true, "delta_db": 2.9138268726856973}], "active_band_count": 3, "active_band_mae_db": 1.2445312991173927}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.404038681520291e-05, "log_rms_mae_db": 1.2626837160298228, "rms_envelope_nrmse": 0.13386119831958007, "envelope_correlation": 0.9916422908809223, "rms_gain_error_db": 0.6235589007257402, "reference_mean_power": 8.336414632310516e-05, "output_mean_power": 9.623547567181628e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.1059391757915914e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.094184470310106e-05, "output_power": 3.728255977651518e-05, "reference_active": true, "delta_db": 0.809595207768517}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2229001330453808e-06, "output_power": 1.1904927797675432e-06, "reference_active": true, "delta_db": -0.11664226454630196}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7483618813517085e-08, "output_power": 7.918429818216488e-08, "reference_active": true, "delta_db": 2.2209526195973477}], "active_band_count": 3, "active_band_mae_db": 1.0490633639707223}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9399910659415127, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2033d6a62fa7bb429b4e1fb6f58d9d87941956d84d334bdc547a0046da981bbe", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9650340174823716, "standard_deviation": null, "minimum": 0.9650340174823716, "maximum": 0.9650340174823716, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7041409202837523, "standard_deviation": null, "minimum": 1.7041409202837523, "maximum": 1.7041409202837523, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6988580828451217, "standard_deviation": null, "minimum": 0.6988580828451217, "maximum": 0.6988580828451217, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21545180403600678, "standard_deviation": null, "minimum": 0.21545180403600678, "maximum": 0.21545180403600678, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9893204928973601, "standard_deviation": null, "minimum": 0.9893204928973601, "maximum": 0.9893204928973601, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0647956389573765, "standard_deviation": null, "minimum": 1.0647956389573765, "maximum": 1.0647956389573765, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6575900181208152, "standard_deviation": null, "minimum": 0.6575900181208152, "maximum": 0.6575900181208152, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12764672418557468, "standard_deviation": null, "minimum": 0.12764672418557468, "maximum": 0.12764672418557468, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9725826735240011, "standard_deviation": null, "minimum": 0.9725826735240011, "maximum": 0.9725826735240011, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9622863929731906, "standard_deviation": null, "minimum": 1.9622863929731906, "maximum": 1.9622863929731906, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2445312991173927, "standard_deviation": null, "minimum": 1.2445312991173927, "maximum": 1.2445312991173927, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21950809498142287, "standard_deviation": null, "minimum": 0.21950809498142287, "maximum": 0.21950809498142287, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9916422908809223, "standard_deviation": null, "minimum": 0.9916422908809223, "maximum": 0.9916422908809223, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2626837160298228, "standard_deviation": null, "minimum": 1.2626837160298228, "maximum": 1.2626837160298228, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0490633639707223, "standard_deviation": null, "minimum": 1.0490633639707223, "maximum": 1.0490633639707223, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13386119831958007, "standard_deviation": null, "minimum": 0.13386119831958007, "maximum": 0.13386119831958007, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9738535514368375, "standard_deviation": null, "minimum": 0.9738535514368375, "maximum": 0.9738535514368375, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.077119811413454, "standard_deviation": null, "minimum": 2.077119811413454, "maximum": 2.077119811413454, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.987410825008757, "standard_deviation": null, "minimum": 0.987410825008757, "maximum": 0.987410825008757, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2448494869832453, "standard_deviation": null, "minimum": 0.2448494869832453, "maximum": 0.2448494869832453, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9869259959279382, "standard_deviation": null, "minimum": 0.9869259959279382, "maximum": 0.9869259959279382, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5698725991314821, "standard_deviation": null, "minimum": 1.5698725991314821, "maximum": 1.5698725991314821, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.143945095540544, "standard_deviation": null, "minimum": 1.143945095540544, "maximum": 1.143945095540544, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1989450674384473, "standard_deviation": null, "minimum": 0.1989450674384473, "maximum": 0.1989450674384473, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6735209215691184, "standard_deviation": null, "minimum": 0.6735209215691184, "maximum": 0.6735209215691184, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.95010534216549, "standard_deviation": null, "minimum": 5.95010534216549, "maximum": 5.95010534216549, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7602949050445598, "standard_deviation": null, "minimum": 0.7602949050445598, "maximum": 0.7602949050445598, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7294971335797378, "standard_deviation": null, "minimum": 0.7294971335797378, "maximum": 0.7294971335797378, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9764993001596283, "standard_deviation": null, "minimum": 0.9764993001596283, "maximum": 0.9764993001596283, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7536569134247086, "standard_deviation": null, "minimum": 1.7536569134247086, "maximum": 1.7536569134247086, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.470646844726337, "standard_deviation": null, "minimum": 0.470646844726337, "maximum": 0.470646844726337, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16425446164976118, "standard_deviation": null, "minimum": 0.16425446164976118, "maximum": 0.16425446164976118, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9921412105386086, "standard_deviation": null, "minimum": 0.9921412105386086, "maximum": 0.9921412105386086, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2468848396864727, "standard_deviation": null, "minimum": 1.2468848396864727, "maximum": 1.2468848396864727, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.607076227717365, "standard_deviation": null, "minimum": 0.607076227717365, "maximum": 0.607076227717365, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11619726816366575, "standard_deviation": null, "minimum": 0.11619726816366575, "maximum": 0.11619726816366575, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1320"}, {"case": "librispeech_1580-141083-0010", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.770113468170166, "predicted_mos": 2.770113468170166, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85120, "input_sample_rate": 16000, "input_seconds": 5.32, "input_float32_sha256": "4918a3883f73e24550694949d648606faadff3e653031ceb89716537f1d82144", "internal_samples": 85120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.991529799997807}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3954792208276992, "全帯域_10ms_rms_envelope_nrmse": 0.1893093237378265, "全帯域_10ms_envelope_correlation": 0.9921666922982115, "全帯域_10ms_modulation_mae_db": 1.1012691246241229, "全帯域_40ms_log_rms_mae_db": 1.1012943361742404, "全帯域_40ms_rms_envelope_nrmse": 0.17546968742404034, "全帯域_40ms_envelope_correlation": 0.9958314178137503, "全帯域_40ms_modulation_mae_db": 1.1744771882815497, "300–1000 Hz_10ms_log_rms_mae_db": 1.749750103542597, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.30133428002890866, "300–1000 Hz_10ms_envelope_correlation": 0.9873389282161017, "300–1000 Hz_10ms_modulation_mae_db": 2.0420258653472545, "300–1000 Hz_40ms_log_rms_mae_db": 1.3534910847860262, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.28988935738195193, "300–1000 Hz_40ms_envelope_correlation": 0.9910579404073467, "300–1000 Hz_40ms_modulation_mae_db": 2.0221806038012615, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5501745295012614, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14128617955871023, "1000–3000 Hz_10ms_envelope_correlation": 0.9936362131935027, "1000–3000 Hz_10ms_modulation_mae_db": 0.8367374749113172, "1000–3000 Hz_40ms_log_rms_mae_db": 1.195038068590965, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11891705503379674, "1000–3000 Hz_40ms_envelope_correlation": 0.9974256580660045, "1000–3000 Hz_40ms_modulation_mae_db": 0.8330177506793154, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5044146468871653, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1407926624698095, "3000–6000 Hz_10ms_envelope_correlation": 0.9882989799228132, "3000–6000 Hz_10ms_modulation_mae_db": 0.38412475639050525, "3000–6000 Hz_40ms_log_rms_mae_db": 0.966243539502946, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09234474143695244, "3000–6000 Hz_40ms_envelope_correlation": 0.995722147905014, "3000–6000 Hz_40ms_modulation_mae_db": 0.23700209998029168, "mr_stft_db": 5.668538807880759, "band_mae_db": 1.2122008712169494, "rms_gain_error_db": 1.290583322722005, "output_peak": 0.9082551527175741}, "full_scale": {"sample_count": 85120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85120, "output_samples": 85120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.810284939920051, "spectral_convergence": 0.46854789130155433, "spectral_convergence_status": "ok", "amplitude_floor": 1.1374446115550809e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3963845809064, "spectral_convergence": 0.33030882456799854, "spectral_convergence_status": "ok", "amplitude_floor": 9.193083193108671e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.798946902815824, "spectral_convergence": 0.3788485832158709, "spectral_convergence_status": "ok", "amplitude_floor": 7.3161107874683e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.1849686170678433e-06, "output_power": 3.146080409266067e-06, "delta_db": 1.5832461483379912, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004950757600852583, "output_power": 0.000758982916844898, "delta_db": 1.8556033801170422, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008267786751374999, "output_power": 0.0012741359492729606, "delta_db": 1.8782650287270766, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007700805582515175, "output_power": 0.0009600258570159934, "delta_db": 0.9574677117489588, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006256626820185013, "output_power": 0.0007022406902744404, "delta_db": 0.5014573836917189, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00016725866300063496, "output_power": 0.000187544865317929, "delta_db": 0.4971655746789092, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.2122008712169494, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.290583322722005, "pitch": null, "output_rms_dbfs": -24.104886700189503, "output_peak": 0.9082551527175741, "reference_rms_dbfs": -25.39547002291151, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.668538807880759}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.0002036923290800657, "log_rms_mae_db": 1.3954792208276992, "rms_envelope_nrmse": 0.1893093237378265, "envelope_correlation": 0.9921666922982115, "rms_gain_error_db": 1.2905864446893172, "reference_mean_power": 0.0028897508304992817, "output_mean_power": 0.003889726285240141}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 1.5709970652253345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007599193486798772, "output_power": 0.0010043166055809243, "reference_active": true, "delta_db": 1.2110314103157969}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014055064928452022, "output_power": 0.00019597105632510276, "reference_active": true, "delta_db": 1.4435907736335671}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3577241130570316e-05, "output_power": 2.73786168546944e-05, "reference_active": true, "delta_db": 0.6491851899230044}], "active_band_count": 3, "active_band_mae_db": 1.1012691246241229}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.0001938989159680862, "log_rms_mae_db": 1.1012943361742404, "rms_envelope_nrmse": 0.17546968742404034, "envelope_correlation": 0.9958314178137503, "rms_gain_error_db": 1.2905991948224937, "reference_mean_power": 0.002906120097080002, "output_mean_power": 0.0039117714915203735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 1.4139945205650041e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006810740061577038, "output_power": 0.0009110820659452185, "reference_active": true, "delta_db": 1.2636319258543447}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.5174236911387e-05, "output_power": 9.073736921216602e-05, "reference_active": true, "delta_db": 1.437102284651264}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3046113483012314e-06, "output_power": 1.5767066066673837e-06, "reference_active": true, "delta_db": 0.8226973543390406}], "active_band_count": 3, "active_band_mae_db": 1.1744771882815497}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013759847516455276, "log_rms_mae_db": 1.749750103542597, "rms_envelope_nrmse": 0.30133428002890866, "envelope_correlation": 0.9873389282161017, "rms_gain_error_db": 2.0343104273846992, "reference_mean_power": 0.001146104301837711, "output_mean_power": 0.0018308602061806322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 6.878091269334419e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031946504747131925, "output_power": 0.0005055554738854196, "reference_active": true, "delta_db": 1.9934546793196182}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.126588772977069e-05, "output_power": 0.00012535373657866842, "reference_active": true, "delta_db": 2.4525558472076288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2614620971957042e-05, "output_power": 1.857295099639259e-05, "reference_active": true, "delta_db": 1.680067069514516}], "active_band_count": 3, "active_band_mae_db": 2.0420258653472545}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00012449215623942515, "log_rms_mae_db": 1.3534910847860262, "rms_envelope_nrmse": 0.28988935738195193, "envelope_correlation": 0.9910579404073467, "rms_gain_error_db": 2.034309229121372, "reference_mean_power": 0.001152608697675057, "output_mean_power": 0.0018412502354657617}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 6.232509709606391e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00029975403822873686, "output_power": 0.00048139141457742714, "reference_active": true, "delta_db": 2.0573329799838027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2561293827419036e-05, "output_power": 5.599950607559367e-05, "reference_active": true, "delta_db": 2.3548254316774067}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.743814177835147e-07, "output_power": 1.1334262404887758e-06, "reference_active": true, "delta_db": 1.654383399742574}], "active_band_count": 3, "active_band_mae_db": 2.0221806038012615}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00013997823549865974, "log_rms_mae_db": 1.5501745295012614, "rms_envelope_nrmse": 0.14128617955871023, "envelope_correlation": 0.9936362131935027, "rms_gain_error_db": 0.847236294060877, "reference_mean_power": 0.0010964326399058018, "output_mean_power": 0.0013326177237283057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 8.778805813988002e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045662233680262583, "output_power": 0.0005501378936390505, "reference_active": true, "delta_db": 0.8091440804060229}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.737497372188522e-05, "output_power": 6.138862398801217e-05, "reference_active": true, "delta_db": 1.1253891679753432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.406352772002978e-05, "output_power": 1.2317607349818752e-05, "reference_active": true, "delta_db": -0.5756791763525855}], "active_band_count": 3, "active_band_mae_db": 0.8367374749113172}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013340521153934576, "log_rms_mae_db": 1.195038068590965, "rms_envelope_nrmse": 0.11891705503379674, "envelope_correlation": 0.9974256580660045, "rms_gain_error_db": 0.8472533388181622, "reference_mean_power": 0.0011026403148293462, "output_mean_power": 0.001340167867971836}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 8.248821386475387e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042363838931095744, "output_power": 0.0005125778414416672, "reference_active": true, "delta_db": 0.8276451969292196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.073594132211124e-05, "output_power": 2.721803610679621e-05, "reference_active": true, "delta_db": 1.181330306255246}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.366449484579509e-07, "output_power": 6.580371915141571e-07, "reference_active": true, "delta_db": -0.4900777488534802}], "active_band_count": 3, "active_band_mae_db": 0.8330177506793154}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1063, "rms_floor": 0.00010989278407704884, "log_rms_mae_db": 1.5044146468871653, "rms_envelope_nrmse": 0.1407926624698095, "envelope_correlation": 0.9882989799228132, "rms_gain_error_db": 0.4330721844374405, "reference_mean_power": 0.0003075575386675287, "output_mean_power": 0.000339807996855553}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1063, "frequency_resolution_hz": 0.18814675446848542, "power_floor": 2.696003640873154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014789902573029864, "output_power": 0.00016068540073200027, "reference_active": true, "delta_db": 0.36011107065240344}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.17770472077717e-05, "output_power": 3.1956373652087844e-05, "reference_active": true, "delta_db": 0.024439513743225803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.120339140119136e-06, "output_power": 3.7237813045291688e-06, "reference_active": true, "delta_db": 0.7678236847758866}], "active_band_count": 3, "active_band_mae_db": 0.38412475639050525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010174719078202204, "log_rms_mae_db": 0.966243539502946, "rms_envelope_nrmse": 0.09234474143695244, "envelope_correlation": 0.995722147905014, "rms_gain_error_db": 0.4330773412044792, "reference_mean_power": 0.000309290771080255, "output_mean_power": 0.0003417233816349178}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.5256223542383704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014158879062551878, "output_power": 0.00015254731903348003, "reference_active": true, "delta_db": 0.3237570720485503}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.501151157810192e-05, "output_power": 1.5042355593998355e-05, "reference_active": true, "delta_db": 0.008914254215509369}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.631467090177325e-07, "output_power": 2.870988417092785e-07, "reference_active": true, "delta_db": 0.3783349736768154}], "active_band_count": 3, "active_band_mae_db": 0.23700209998029168}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9242058025370763, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "23e5854d64b4a9541f5be1844f6d51fbd589de3bbd8d5f99816ec07f4f6c2265", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9936362131935027, "standard_deviation": null, "minimum": 0.9936362131935027, "maximum": 0.9936362131935027, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5501745295012614, "standard_deviation": null, "minimum": 1.5501745295012614, "maximum": 1.5501745295012614, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8367374749113172, "standard_deviation": null, "minimum": 0.8367374749113172, "maximum": 0.8367374749113172, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14128617955871023, "standard_deviation": null, "minimum": 0.14128617955871023, "maximum": 0.14128617955871023, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974256580660045, "standard_deviation": null, "minimum": 0.9974256580660045, "maximum": 0.9974256580660045, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.195038068590965, "standard_deviation": null, "minimum": 1.195038068590965, "maximum": 1.195038068590965, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8330177506793154, "standard_deviation": null, "minimum": 0.8330177506793154, "maximum": 0.8330177506793154, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11891705503379674, "standard_deviation": null, "minimum": 0.11891705503379674, "maximum": 0.11891705503379674, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9882989799228132, "standard_deviation": null, "minimum": 0.9882989799228132, "maximum": 0.9882989799228132, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5044146468871653, "standard_deviation": null, "minimum": 1.5044146468871653, "maximum": 1.5044146468871653, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.38412475639050525, "standard_deviation": null, "minimum": 0.38412475639050525, "maximum": 0.38412475639050525, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1407926624698095, "standard_deviation": null, "minimum": 0.1407926624698095, "maximum": 0.1407926624698095, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995722147905014, "standard_deviation": null, "minimum": 0.995722147905014, "maximum": 0.995722147905014, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.966243539502946, "standard_deviation": null, "minimum": 0.966243539502946, "maximum": 0.966243539502946, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.23700209998029168, "standard_deviation": null, "minimum": 0.23700209998029168, "maximum": 0.23700209998029168, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09234474143695244, "standard_deviation": null, "minimum": 0.09234474143695244, "maximum": 0.09234474143695244, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9873389282161017, "standard_deviation": null, "minimum": 0.9873389282161017, "maximum": 0.9873389282161017, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.749750103542597, "standard_deviation": null, "minimum": 1.749750103542597, "maximum": 1.749750103542597, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.0420258653472545, "standard_deviation": null, "minimum": 2.0420258653472545, "maximum": 2.0420258653472545, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30133428002890866, "standard_deviation": null, "minimum": 0.30133428002890866, "maximum": 0.30133428002890866, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9910579404073467, "standard_deviation": null, "minimum": 0.9910579404073467, "maximum": 0.9910579404073467, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3534910847860262, "standard_deviation": null, "minimum": 1.3534910847860262, "maximum": 1.3534910847860262, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 2.0221806038012615, "standard_deviation": null, "minimum": 2.0221806038012615, "maximum": 2.0221806038012615, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.28988935738195193, "standard_deviation": null, "minimum": 0.28988935738195193, "maximum": 0.28988935738195193, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.2122008712169494, "standard_deviation": null, "minimum": 1.2122008712169494, "maximum": 1.2122008712169494, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.668538807880759, "standard_deviation": null, "minimum": 5.668538807880759, "maximum": 5.668538807880759, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.9082551527175741, "standard_deviation": null, "minimum": 0.9082551527175741, "maximum": 0.9082551527175741, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.290583322722005, "standard_deviation": null, "minimum": 1.290583322722005, "maximum": 1.290583322722005, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9921666922982115, "standard_deviation": null, "minimum": 0.9921666922982115, "maximum": 0.9921666922982115, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3954792208276992, "standard_deviation": null, "minimum": 1.3954792208276992, "maximum": 1.3954792208276992, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1012691246241229, "standard_deviation": null, "minimum": 1.1012691246241229, "maximum": 1.1012691246241229, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1893093237378265, "standard_deviation": null, "minimum": 0.1893093237378265, "maximum": 0.1893093237378265, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9958314178137503, "standard_deviation": null, "minimum": 0.9958314178137503, "maximum": 0.9958314178137503, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1012943361742404, "standard_deviation": null, "minimum": 1.1012943361742404, "maximum": 1.1012943361742404, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1744771882815497, "standard_deviation": null, "minimum": 1.1744771882815497, "maximum": 1.1744771882815497, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.17546968742404034, "standard_deviation": null, "minimum": 0.17546968742404034, "maximum": 0.17546968742404034, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1580"}, {"case": "librispeech_1580-141084-0016", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.1402225494384766, "predicted_mos": 2.1402225494384766, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 95360, "input_sample_rate": 16000, "input_seconds": 5.96, "input_float32_sha256": "16f4d5a6d9b71dad368c8527ac85f521d7dbf2fe7f59d13d1f0aa2092985cc14", "internal_samples": 95360, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.907293299678713}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6368931267467528, "全帯域_10ms_rms_envelope_nrmse": 0.178548156103909, "全帯域_10ms_envelope_correlation": 0.9848204706704333, "全帯域_10ms_modulation_mae_db": 1.0468577748374888, "全帯域_40ms_log_rms_mae_db": 1.220325209541365, "全帯域_40ms_rms_envelope_nrmse": 0.1450304620396809, "全帯域_40ms_envelope_correlation": 0.9930757014206789, "全帯域_40ms_modulation_mae_db": 1.0922473875553438, "300–1000 Hz_10ms_log_rms_mae_db": 2.101500519578928, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22409190992038178, "300–1000 Hz_10ms_envelope_correlation": 0.9888674471920963, "300–1000 Hz_10ms_modulation_mae_db": 1.571771367011029, "300–1000 Hz_40ms_log_rms_mae_db": 1.5825999945122506, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.20362443638696853, "300–1000 Hz_40ms_envelope_correlation": 0.9938489966182186, "300–1000 Hz_40ms_modulation_mae_db": 1.179535141617512, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6385452564272116, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18545240862024376, "1000–3000 Hz_10ms_envelope_correlation": 0.9895591416480922, "1000–3000 Hz_10ms_modulation_mae_db": 0.934091603200076, "1000–3000 Hz_40ms_log_rms_mae_db": 1.123762719123889, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1534155341289703, "1000–3000 Hz_40ms_envelope_correlation": 0.9960931029953155, "1000–3000 Hz_40ms_modulation_mae_db": 1.045499518297493, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6558854180505507, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22143622853418735, "3000–6000 Hz_10ms_envelope_correlation": 0.963908735885462, "3000–6000 Hz_10ms_modulation_mae_db": 1.0404482588391784, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9973335518168586, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13146817749729264, "3000–6000 Hz_40ms_envelope_correlation": 0.9865303712911786, "3000–6000 Hz_40ms_modulation_mae_db": 1.039678897349244, "mr_stft_db": 5.873136211415388, "band_mae_db": 0.9596554220195134, "rms_gain_error_db": 0.9491145858107237, "output_peak": 0.5960165909925491}, "full_scale": {"sample_count": 95360, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 95360, "output_samples": 95360, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.170760380594033, "spectral_convergence": 0.5588659807050176, "spectral_convergence_status": "ok", "amplitude_floor": 1.2569032294341943e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.582727149927807, "spectral_convergence": 0.3975140498533627, "spectral_convergence_status": "ok", "amplitude_floor": 9.073914887165124e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.865921103724321, "spectral_convergence": 0.43111145626250796, "spectral_convergence_status": "ok", "amplitude_floor": 7.969434026985316e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.036863249031963e-06, "output_power": 7.061716751674902e-06, "delta_db": 1.4675013393717709, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00014485054714414723, "output_power": 0.00017927678709412352, "delta_db": 0.9260392039165928, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004849180298355377, "output_power": 0.0006675660911655643, "delta_db": 1.3882593669283774, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006903353830986323, "output_power": 0.0008989873806690711, "delta_db": 1.146934617695907, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00029452981369653373, "output_power": 0.0003310634206517585, "delta_db": 0.5078193532404986, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004100383435072048, "output_power": 0.0004415321534623309, "delta_db": 0.3213786509639346, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9596554220195134, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.9491145858107237, "pitch": null, "output_rms_dbfs": -25.976547681410853, "output_peak": 0.5960165909925491, "reference_rms_dbfs": -26.925662267221572, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.873136211415388}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 95360, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.0001898595271537964, "log_rms_mae_db": 1.6368931267467528, "rms_envelope_nrmse": 0.178548156103909, "envelope_correlation": 0.9848204706704333, "rms_gain_error_db": 0.9491176306759086, "reference_mean_power": 0.002031411160407025, "output_mean_power": 0.0025276072778018187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 9.927152563814217e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006381525961065371, "output_power": 0.0007839041476611633, "reference_active": true, "delta_db": 0.8933842196283488}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010674365906282643, "output_power": 0.00013593912588364353, "reference_active": true, "delta_db": 1.0500238710214373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4567838102832594e-05, "output_power": 4.553955082648461e-05, "reference_active": true, "delta_db": 1.1971652338626806}], "active_band_count": 3, "active_band_mae_db": 1.0468577748374888}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00017019756464998765, "log_rms_mae_db": 1.220325209541365, "rms_envelope_nrmse": 0.1450304620396809, "envelope_correlation": 0.9930757014206789, "rms_gain_error_db": 0.9491219899127434, "reference_mean_power": 0.0020416854120747243, "output_mean_power": 0.002540393686450285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 8.390142156345837e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005509511464645055, "output_power": 0.0006858953245984699, "reference_active": true, "delta_db": 0.9514475138511294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.5897699076585876e-05, "output_power": 5.9841692382195874e-05, "reference_active": true, "delta_db": 1.1521295308561184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848237969563024e-06, "output_power": 2.421443833232048e-06, "reference_active": true, "delta_db": 1.1731651179587836}], "active_band_count": 3, "active_band_mae_db": 1.0922473875553438}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 9.246932648014587e-05, "log_rms_mae_db": 2.101500519578928, "rms_envelope_nrmse": 0.22409190992038178, "envelope_correlation": 0.9888674471920963, "rms_gain_error_db": 1.4324557671998202, "reference_mean_power": 0.0005236182881812637, "output_mean_power": 0.000728216279128583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.7855833380403224e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028934531457073826, "output_power": 0.00042047728894031763, "reference_active": true, "delta_db": 1.623260892705698}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6197786234688146e-05, "output_power": 5.373755568120267e-05, "reference_active": true, "delta_db": 1.7159589713774972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0243815325354375e-05, "output_power": 1.4062779400384212e-05, "reference_active": true, "delta_db": 1.3760942369498919}], "active_band_count": 3, "active_band_mae_db": 1.571771367011029}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 8.334521268786246e-05, "log_rms_mae_db": 1.5825999945122506, "rms_envelope_nrmse": 0.20362443638696853, "envelope_correlation": 0.9938489966182186, "rms_gain_error_db": 1.432458386637203, "reference_mean_power": 0.000526268698480611, "output_mean_power": 0.0007319027490605706}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.428386331422997e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002786455084699618, "output_power": 0.00040762714809227627, "reference_active": true, "delta_db": 1.652110534537141}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6309295594140896e-05, "output_power": 2.3705423692146134e-05, "reference_active": true, "delta_db": 1.624125178931319}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.788822138965803e-07, "output_power": 8.273871235774266e-07, "reference_active": true, "delta_db": 0.2623697113840757}], "active_band_count": 3, "active_band_mae_db": 1.179535141617512}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00017019739791537767, "log_rms_mae_db": 1.6385452564272116, "rms_envelope_nrmse": 0.18545240862024376, "envelope_correlation": 0.9895591416480922, "rms_gain_error_db": 1.075999990258998, "reference_mean_power": 0.0008446325483556021, "output_mean_power": 0.0010821010344332488}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 4.004634586327471e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028725854625925005, "output_power": 0.0003767645916879522, "reference_active": true, "delta_db": 1.177971227778107}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3436307686794414e-05, "output_power": 5.9980278387330114e-05, "reference_active": true, "delta_db": 1.4015557627676298}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1340506745101254e-05, "output_power": 1.1937331190336755e-05, "reference_active": true, "delta_db": 0.2227478190544913}], "active_band_count": 3, "active_band_mae_db": 0.934091603200076}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 0.00015333717053704127, "log_rms_mae_db": 1.123762719123889, "rms_envelope_nrmse": 0.1534155341289703, "envelope_correlation": 0.9960931029953155, "rms_gain_error_db": 1.0760028829216914, "reference_mean_power": 0.0008489024890862695, "output_mean_power": 0.0010875721936166901}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 3.579273675799131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002706567111490358, "output_power": 0.00035809755151670284, "reference_active": true, "delta_db": 1.2158255130542497}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.939557003916555e-05, "output_power": 2.6758196323809914e-05, "reference_active": true, "delta_db": 1.3975428770039835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.660068008029229e-07, "output_power": 8.640634641279625e-07, "reference_active": true, "delta_db": 0.5231301648342457}], "active_band_count": 3, "active_band_mae_db": 1.045499518297493}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1191, "rms_floor": 0.00010107776753147987, "log_rms_mae_db": 1.6558854180505507, "rms_envelope_nrmse": 0.22143622853418735, "envelope_correlation": 0.963908735885462, "rms_gain_error_db": 0.31136836188826145, "reference_mean_power": 0.00032790731816785513, "output_mean_power": 0.0003522799657609938}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1191, "frequency_resolution_hz": 0.16792611251049538, "power_floor": 3.172510420573334e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020494142716961824, "output_power": 0.00020416544271871937, "reference_active": true, "delta_db": -0.01647521437694402}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6580284356723e-05, "output_power": 4.266424417768645e-05, "reference_active": true, "delta_db": 0.6681697911297515}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.565019545039446e-06, "output_power": 1.6763158017426077e-05, "reference_active": true, "delta_db": 2.4366997710108396}], "active_band_count": 3, "active_band_mae_db": 1.0404482588391784}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 95360, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1185, "rms_floor": 7.865643818167715e-05, "log_rms_mae_db": 0.9973335518168586, "rms_envelope_nrmse": 0.13146817749729264, "envelope_correlation": 0.9865303712911786, "rms_gain_error_db": 0.3113694351716999, "reference_mean_power": 0.0003295653902210026, "output_mean_power": 0.0003540613662533725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1185, "frequency_resolution_hz": 0.16877637130801687, "power_floor": 2.853826866715288e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001952923309195638, "output_power": 0.00020043973207712143, "reference_active": true, "delta_db": 0.11298624571773198}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5170390000310775e-05, "output_power": 1.944900843652542e-05, "reference_active": true, "delta_db": 1.079007189387764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.363045655471935e-07, "output_power": 1.1475241101249575e-06, "reference_active": true, "delta_db": 1.9270432569422358}], "active_band_count": 3, "active_band_mae_db": 1.039678897349244}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8679517139009115, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "b6815243ebe084b1ec1a84d848b95f601977ba8d44f547844c31bbba532b312f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895591416480922, "standard_deviation": null, "minimum": 0.9895591416480922, "maximum": 0.9895591416480922, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6385452564272116, "standard_deviation": null, "minimum": 1.6385452564272116, "maximum": 1.6385452564272116, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.934091603200076, "standard_deviation": null, "minimum": 0.934091603200076, "maximum": 0.934091603200076, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18545240862024376, "standard_deviation": null, "minimum": 0.18545240862024376, "maximum": 0.18545240862024376, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960931029953155, "standard_deviation": null, "minimum": 0.9960931029953155, "maximum": 0.9960931029953155, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.123762719123889, "standard_deviation": null, "minimum": 1.123762719123889, "maximum": 1.123762719123889, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.045499518297493, "standard_deviation": null, "minimum": 1.045499518297493, "maximum": 1.045499518297493, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1534155341289703, "standard_deviation": null, "minimum": 0.1534155341289703, "maximum": 0.1534155341289703, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.963908735885462, "standard_deviation": null, "minimum": 0.963908735885462, "maximum": 0.963908735885462, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6558854180505507, "standard_deviation": null, "minimum": 1.6558854180505507, "maximum": 1.6558854180505507, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0404482588391784, "standard_deviation": null, "minimum": 1.0404482588391784, "maximum": 1.0404482588391784, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22143622853418735, "standard_deviation": null, "minimum": 0.22143622853418735, "maximum": 0.22143622853418735, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9865303712911786, "standard_deviation": null, "minimum": 0.9865303712911786, "maximum": 0.9865303712911786, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9973335518168586, "standard_deviation": null, "minimum": 0.9973335518168586, "maximum": 0.9973335518168586, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.039678897349244, "standard_deviation": null, "minimum": 1.039678897349244, "maximum": 1.039678897349244, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13146817749729264, "standard_deviation": null, "minimum": 0.13146817749729264, "maximum": 0.13146817749729264, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9888674471920963, "standard_deviation": null, "minimum": 0.9888674471920963, "maximum": 0.9888674471920963, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.101500519578928, "standard_deviation": null, "minimum": 2.101500519578928, "maximum": 2.101500519578928, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.571771367011029, "standard_deviation": null, "minimum": 1.571771367011029, "maximum": 1.571771367011029, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22409190992038178, "standard_deviation": null, "minimum": 0.22409190992038178, "maximum": 0.22409190992038178, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9938489966182186, "standard_deviation": null, "minimum": 0.9938489966182186, "maximum": 0.9938489966182186, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5825999945122506, "standard_deviation": null, "minimum": 1.5825999945122506, "maximum": 1.5825999945122506, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.179535141617512, "standard_deviation": null, "minimum": 1.179535141617512, "maximum": 1.179535141617512, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20362443638696853, "standard_deviation": null, "minimum": 0.20362443638696853, "maximum": 0.20362443638696853, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9596554220195134, "standard_deviation": null, "minimum": 0.9596554220195134, "maximum": 0.9596554220195134, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.873136211415388, "standard_deviation": null, "minimum": 5.873136211415388, "maximum": 5.873136211415388, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5960165909925491, "standard_deviation": null, "minimum": 0.5960165909925491, "maximum": 0.5960165909925491, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.9491145858107237, "standard_deviation": null, "minimum": 0.9491145858107237, "maximum": 0.9491145858107237, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9848204706704333, "standard_deviation": null, "minimum": 0.9848204706704333, "maximum": 0.9848204706704333, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6368931267467528, "standard_deviation": null, "minimum": 1.6368931267467528, "maximum": 1.6368931267467528, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0468577748374888, "standard_deviation": null, "minimum": 1.0468577748374888, "maximum": 1.0468577748374888, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.178548156103909, "standard_deviation": null, "minimum": 0.178548156103909, "maximum": 0.178548156103909, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9930757014206789, "standard_deviation": null, "minimum": 0.9930757014206789, "maximum": 0.9930757014206789, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.220325209541365, "standard_deviation": null, "minimum": 1.220325209541365, "maximum": 1.220325209541365, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0922473875553438, "standard_deviation": null, "minimum": 1.0922473875553438, "maximum": 1.0922473875553438, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1450304620396809, "standard_deviation": null, "minimum": 0.1450304620396809, "maximum": 0.1450304620396809, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1580"}, {"case": "librispeech_1580-141084-0047", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2041006088256836, "predicted_mos": 3.2041006088256836, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84000, "input_sample_rate": 16000, "input_seconds": 5.25, "input_float32_sha256": "58bdf37ff27a1292b7669f61bb9e17e4f2d748094212c236736316a0d8be7173", "internal_samples": 84000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.3736826996319}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5666432166798074, "全帯域_10ms_rms_envelope_nrmse": 0.18933161604391732, "全帯域_10ms_envelope_correlation": 0.9860227406044861, "全帯域_10ms_modulation_mae_db": 0.8373696556338684, "全帯域_40ms_log_rms_mae_db": 1.1891976715342731, "全帯域_40ms_rms_envelope_nrmse": 0.16089490540946522, "全帯域_40ms_envelope_correlation": 0.9937393406333174, "全帯域_40ms_modulation_mae_db": 1.0225836710150027, "300–1000 Hz_10ms_log_rms_mae_db": 1.942909875772633, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2535863413087768, "300–1000 Hz_10ms_envelope_correlation": 0.9876654595011595, "300–1000 Hz_10ms_modulation_mae_db": 2.1033485924910664, "300–1000 Hz_40ms_log_rms_mae_db": 1.5582535121810757, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2361766005299727, "300–1000 Hz_40ms_envelope_correlation": 0.9927128427151019, "300–1000 Hz_40ms_modulation_mae_db": 1.8381319324569312, "1000–3000 Hz_10ms_log_rms_mae_db": 1.496629710991574, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16485795493669103, "1000–3000 Hz_10ms_envelope_correlation": 0.9894267806655069, "1000–3000 Hz_10ms_modulation_mae_db": 0.9979650217149153, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9961774129693705, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12375010576795205, "1000–3000 Hz_40ms_envelope_correlation": 0.9971451586714767, "1000–3000 Hz_40ms_modulation_mae_db": 1.1056976312631215, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4131155562923652, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16320323401087833, "3000–6000 Hz_10ms_envelope_correlation": 0.9829564148139016, "3000–6000 Hz_10ms_modulation_mae_db": 0.25313980399944164, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8691196904889605, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0943991727499084, "3000–6000 Hz_40ms_envelope_correlation": 0.9955132638262528, "3000–6000 Hz_40ms_modulation_mae_db": 0.18517703067902014, "mr_stft_db": 5.9631784093881715, "band_mae_db": 1.0405125903256085, "rms_gain_error_db": 1.1156057709133003, "output_peak": 0.7865849114541438}, "full_scale": {"sample_count": 84000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84000, "output_samples": 84000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.591197557238011, "spectral_convergence": 0.6063840169513819, "spectral_convergence_status": "ok", "amplitude_floor": 1.02822042353332e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.339852717637016, "spectral_convergence": 0.37937366390109467, "spectral_convergence_status": "ok", "amplitude_floor": 8.928413196720056e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.958484953289489, "spectral_convergence": 0.4213207207510874, "spectral_convergence_status": "ok", "amplitude_floor": 7.736392334055093e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 1.4784881078912855e-05, "output_power": 1.868518008191087e-05, "delta_db": 1.0167945213901877, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002602315552138397, "output_power": 0.0003779253338700026, "delta_db": 1.6204604822348339, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005931268593866254, "output_power": 0.0008419447741081163, "delta_db": 1.5213601441015459, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005283196568297616, "output_power": 0.0006466033259956935, "delta_db": 0.877411647169214, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003079318069923152, "output_power": 0.00036603049498805753, "delta_db": 0.7506271863971286, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00026143236338356946, "output_power": 0.00029040327185070585, "delta_db": 0.45642156066074113, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.0405125903256085, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1156057709133003, "pitch": null, "output_rms_dbfs": -25.94894100093264, "output_peak": 0.7865849114541438, "reference_rms_dbfs": -27.06454677184594, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9631784093881715}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00019312594202620684, "log_rms_mae_db": 1.5666432166798074, "rms_envelope_nrmse": 0.18933161604391732, "envelope_correlation": 0.9860227406044861, "rms_gain_error_db": 1.1156075832076413, "reference_mean_power": 0.0019676998400409485, "output_mean_power": 0.002544014654214594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 4.0378624939923077e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018862042103601667, "output_power": 0.00023326516176190262, "reference_active": true, "delta_db": 0.9226117165045834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.862312285918816e-05, "output_power": 0.00012784998685389683, "reference_active": true, "delta_db": 1.1272193754359328}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3789377405749525e-05, "output_power": 1.5338131103226597e-05, "reference_active": true, "delta_db": 0.46227787496108896}], "active_band_count": 3, "active_band_mae_db": 0.8373696556338684}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00018244538675777041, "log_rms_mae_db": 1.1891976715342731, "rms_envelope_nrmse": 0.16089490540946522, "envelope_correlation": 0.9937393406333174, "rms_gain_error_db": 1.115616520865816, "reference_mean_power": 0.001979010526094432, "output_mean_power": 0.002558643365171755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 3.3102005537920435e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017836230285954771, "output_power": 0.00021457425855243056, "reference_active": true, "delta_db": 0.8027454957495621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3320761072042273e-05, "output_power": 5.766890192146199e-05, "reference_active": true, "delta_db": 1.2424560421571405}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.132775238046075e-07, "output_power": 1.029185639069913e-06, "reference_active": true, "delta_db": 1.022549475138305}], "active_band_count": 3, "active_band_mae_db": 1.0225836710150027}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00014606965842361196, "log_rms_mae_db": 1.942909875772633, "rms_envelope_nrmse": 0.2535863413087768, "envelope_correlation": 0.9876654595011595, "rms_gain_error_db": 1.67122923363587, "reference_mean_power": 0.0007714288246159527, "output_mean_power": 0.0011334928519333464}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 2.155891354220712e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.355627855804233e-05, "output_power": 0.00014939715355038278, "reference_active": true, "delta_db": 2.523632339203608}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.117824128471733e-05, "output_power": 3.657065354545028e-05, "reference_active": true, "delta_db": 2.3724282964575365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3214088476749303e-06, "output_power": 4.599608336132947e-06, "reference_active": true, "delta_db": 1.413985141812055}], "active_band_count": 3, "active_band_mae_db": 2.1033485924910664}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.0001346093108760191, "log_rms_mae_db": 1.5582535121810757, "rms_envelope_nrmse": 0.2361766005299727, "envelope_correlation": 0.9927128427151019, "rms_gain_error_db": 1.671230064879778, "reference_mean_power": 0.0007758659430163304, "output_mean_power": 0.0011400127150171812}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.945223621465206e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.346976557991944e-05, "output_power": 0.00013420229001572355, "reference_active": true, "delta_db": 2.6165127251531652}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.57892144723787e-06, "output_power": 1.5859382476847105e-05, "reference_active": true, "delta_db": 2.189696612071383}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.32297115664878e-07, "output_power": 2.7344026319379414e-07, "reference_active": true, "delta_db": 0.7081864601462463}], "active_band_count": 3, "active_band_mae_db": 1.8381319324569312}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00012444218003818797, "log_rms_mae_db": 1.496629710991574, "rms_envelope_nrmse": 0.16485795493669103, "envelope_correlation": 0.9894267806655069, "rms_gain_error_db": 0.8743220722007884, "reference_mean_power": 0.0006875503513213645, "output_mean_power": 0.000840885213154287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.3390858183888e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.254326674690693e-05, "output_power": 5.055804387029524e-05, "reference_active": true, "delta_db": 0.7495942762742552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4859458498211302e-05, "output_power": 1.985711587760586e-05, "reference_active": true, "delta_db": 1.2591318682522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.141405389451205e-06, "output_power": 3.941312553015798e-06, "reference_active": true, "delta_db": 0.985168920618291}], "active_band_count": 3, "active_band_mae_db": 0.9979650217149153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 0.00011369765125964215, "log_rms_mae_db": 0.9961774129693705, "rms_envelope_nrmse": 0.12375010576795205, "envelope_correlation": 0.9971451586714767, "rms_gain_error_db": 0.8743332190631208, "reference_mean_power": 0.0006915015836222027, "output_mean_power": 0.0008457198049127607}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.214722623247201e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7827908938064025e-05, "output_power": 4.507375826006913e-05, "reference_active": true, "delta_db": 0.7611143624993953}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.619804659862373e-06, "output_power": 8.23344665528029e-06, "reference_active": true, "delta_db": 1.6586045595553398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0150238495180175e-07, "output_power": 2.4775223895446314e-07, "reference_active": true, "delta_db": 0.8973739717346291}], "active_band_count": 3, "active_band_mae_db": 1.1056976312631215}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1049, "rms_floor": 0.00010848831624356859, "log_rms_mae_db": 1.4131155562923652, "rms_envelope_nrmse": 0.16320323401087833, "envelope_correlation": 0.9829564148139016, "rms_gain_error_db": 0.488034310883947, "reference_mean_power": 0.00023818463717296146, "output_mean_power": 0.0002665122509270507}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1049, "frequency_resolution_hz": 0.19065776930409914, "power_floor": 1.7158903163900278e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.877460217795428e-05, "output_power": 9.387970321003994e-05, "reference_active": true, "delta_db": 0.24282973164658775}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.954888518804953e-05, "output_power": 4.163902059768498e-05, "reference_active": true, "delta_db": 0.22366259778514117}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.296333400161532e-06, "output_power": 7.805439804002869e-06, "reference_active": true, "delta_db": 0.29292708256659605}], "active_band_count": 3, "active_band_mae_db": 0.25313980399944164}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1043, "rms_floor": 7.397564116979367e-05, "log_rms_mae_db": 0.8691196904889605, "rms_envelope_nrmse": 0.0943991727499084, "envelope_correlation": 0.9955132638262528, "rms_gain_error_db": 0.4880356349698044, "reference_mean_power": 0.0002395532500233675, "output_mean_power": 0.0002680437164346196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1043, "frequency_resolution_hz": 0.19175455417066156, "power_floor": 1.5691901914560294e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.470858674877874e-05, "output_power": 9.959063458995208e-05, "reference_active": true, "delta_db": 0.21829143623821623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.833295648941501e-05, "output_power": 1.938614324702524e-05, "reference_active": true, "delta_db": 0.24258909771665574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.840671065644946e-07, "output_power": 4.94732713221455e-07, "reference_active": true, "delta_db": 0.0946505580821885}], "active_band_count": 3, "active_band_mae_db": 0.18517703067902014}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9193645066923752, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "fc202488bfd2be033f15e31829c727643c19bf82975a7fa0581d9d7220e59d37", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9894267806655069, "standard_deviation": null, "minimum": 0.9894267806655069, "maximum": 0.9894267806655069, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.496629710991574, "standard_deviation": null, "minimum": 1.496629710991574, "maximum": 1.496629710991574, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9979650217149153, "standard_deviation": null, "minimum": 0.9979650217149153, "maximum": 0.9979650217149153, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16485795493669103, "standard_deviation": null, "minimum": 0.16485795493669103, "maximum": 0.16485795493669103, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971451586714767, "standard_deviation": null, "minimum": 0.9971451586714767, "maximum": 0.9971451586714767, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9961774129693705, "standard_deviation": null, "minimum": 0.9961774129693705, "maximum": 0.9961774129693705, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1056976312631215, "standard_deviation": null, "minimum": 1.1056976312631215, "maximum": 1.1056976312631215, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12375010576795205, "standard_deviation": null, "minimum": 0.12375010576795205, "maximum": 0.12375010576795205, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9829564148139016, "standard_deviation": null, "minimum": 0.9829564148139016, "maximum": 0.9829564148139016, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4131155562923652, "standard_deviation": null, "minimum": 1.4131155562923652, "maximum": 1.4131155562923652, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.25313980399944164, "standard_deviation": null, "minimum": 0.25313980399944164, "maximum": 0.25313980399944164, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16320323401087833, "standard_deviation": null, "minimum": 0.16320323401087833, "maximum": 0.16320323401087833, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9955132638262528, "standard_deviation": null, "minimum": 0.9955132638262528, "maximum": 0.9955132638262528, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8691196904889605, "standard_deviation": null, "minimum": 0.8691196904889605, "maximum": 0.8691196904889605, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.18517703067902014, "standard_deviation": null, "minimum": 0.18517703067902014, "maximum": 0.18517703067902014, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0943991727499084, "standard_deviation": null, "minimum": 0.0943991727499084, "maximum": 0.0943991727499084, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9876654595011595, "standard_deviation": null, "minimum": 0.9876654595011595, "maximum": 0.9876654595011595, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.942909875772633, "standard_deviation": null, "minimum": 1.942909875772633, "maximum": 1.942909875772633, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.1033485924910664, "standard_deviation": null, "minimum": 2.1033485924910664, "maximum": 2.1033485924910664, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2535863413087768, "standard_deviation": null, "minimum": 0.2535863413087768, "maximum": 0.2535863413087768, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9927128427151019, "standard_deviation": null, "minimum": 0.9927128427151019, "maximum": 0.9927128427151019, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5582535121810757, "standard_deviation": null, "minimum": 1.5582535121810757, "maximum": 1.5582535121810757, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.8381319324569312, "standard_deviation": null, "minimum": 1.8381319324569312, "maximum": 1.8381319324569312, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2361766005299727, "standard_deviation": null, "minimum": 0.2361766005299727, "maximum": 0.2361766005299727, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.0405125903256085, "standard_deviation": null, "minimum": 1.0405125903256085, "maximum": 1.0405125903256085, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.9631784093881715, "standard_deviation": null, "minimum": 5.9631784093881715, "maximum": 5.9631784093881715, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7865849114541438, "standard_deviation": null, "minimum": 0.7865849114541438, "maximum": 0.7865849114541438, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1156057709133003, "standard_deviation": null, "minimum": 1.1156057709133003, "maximum": 1.1156057709133003, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9860227406044861, "standard_deviation": null, "minimum": 0.9860227406044861, "maximum": 0.9860227406044861, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5666432166798074, "standard_deviation": null, "minimum": 1.5666432166798074, "maximum": 1.5666432166798074, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8373696556338684, "standard_deviation": null, "minimum": 0.8373696556338684, "maximum": 0.8373696556338684, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18933161604391732, "standard_deviation": null, "minimum": 0.18933161604391732, "maximum": 0.18933161604391732, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9937393406333174, "standard_deviation": null, "minimum": 0.9937393406333174, "maximum": 0.9937393406333174, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1891976715342731, "standard_deviation": null, "minimum": 1.1891976715342731, "maximum": 1.1891976715342731, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0225836710150027, "standard_deviation": null, "minimum": 1.0225836710150027, "maximum": 1.0225836710150027, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16089490540946522, "standard_deviation": null, "minimum": 0.16089490540946522, "maximum": 0.16089490540946522, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1580"}, {"case": "librispeech_1995-1836-0000", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2686283588409424, "predicted_mos": 3.2686283588409424, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7754f4d1b3bc9ec0ef493bf759dfc7f8ddd04951d53d851e177cd2b6d16bec5e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.155363900121301}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6056613731728178, "全帯域_10ms_rms_envelope_nrmse": 0.17427138436769415, "全帯域_10ms_envelope_correlation": 0.9855768713306386, "全帯域_10ms_modulation_mae_db": 1.2216730724899376, "全帯域_40ms_log_rms_mae_db": 1.1776904322337685, "全帯域_40ms_rms_envelope_nrmse": 0.15812434818862117, "全帯域_40ms_envelope_correlation": 0.991243735036081, "全帯域_40ms_modulation_mae_db": 0.9359485217697471, "300–1000 Hz_10ms_log_rms_mae_db": 1.7318810681541887, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22580590790911093, "300–1000 Hz_10ms_envelope_correlation": 0.9811701895201226, "300–1000 Hz_10ms_modulation_mae_db": 1.9012075760729739, "300–1000 Hz_40ms_log_rms_mae_db": 1.3444940306977664, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2117825551962507, "300–1000 Hz_40ms_envelope_correlation": 0.984995632259557, "300–1000 Hz_40ms_modulation_mae_db": 1.466464565126713, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3610448494377467, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2309989246910671, "1000–3000 Hz_10ms_envelope_correlation": 0.9847859008394831, "1000–3000 Hz_10ms_modulation_mae_db": 1.9589801561984947, "1000–3000 Hz_40ms_log_rms_mae_db": 1.046874287184546, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.20947712373057284, "1000–3000 Hz_40ms_envelope_correlation": 0.9911499777139577, "1000–3000 Hz_40ms_modulation_mae_db": 1.7652724516120377, "3000–6000 Hz_10ms_log_rms_mae_db": 1.445644042085642, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.34120458882859506, "3000–6000 Hz_10ms_envelope_correlation": 0.9338376847250924, "3000–6000 Hz_10ms_modulation_mae_db": 2.1300194979506104, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9185091997757932, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.21173901267750678, "3000–6000 Hz_40ms_envelope_correlation": 0.9752261264894855, "3000–6000 Hz_40ms_modulation_mae_db": 2.397599571382515, "mr_stft_db": 5.254967172523161, "band_mae_db": 0.9698868287388157, "rms_gain_error_db": 1.1013269797669467, "output_peak": 0.4095593649295868}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.839942618998417, "spectral_convergence": 0.4710358943154988, "spectral_convergence_status": "ok", "amplitude_floor": 1.1304927744850054e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.770735911974346, "spectral_convergence": 0.22527020756015514, "spectral_convergence_status": "ok", "amplitude_floor": 1.0019114511304393e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.15422298659672, "spectral_convergence": 0.25360468494289196, "spectral_convergence_status": "ok", "amplitude_floor": 8.812050137930174e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0013595771819321111, "output_power": 0.0016310306828820303, "delta_db": 0.7905826392986609, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002257060285350754, "output_power": 0.0029838474983984007, "delta_db": 1.2123346389287641, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006976179772550826, "output_power": 0.0009490817712801245, "delta_db": 1.3368596850403505, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001064894568705692, "output_power": 0.00014803965045887365, "delta_db": 1.4307143899383823, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.7425121182630722e-05, "output_power": 3.300561543257088e-05, "delta_db": 0.8043927989172244, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.591013773579702e-05, "output_power": 5.914720735785732e-05, "delta_db": 0.2444368203095129, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9698868287388157, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1013269797669467, "pitch": null, "output_rms_dbfs": -22.362611908554445, "output_peak": 0.4095593649295868, "reference_rms_dbfs": -23.46393888832139, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.254967172523161}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016450226550328756, "log_rms_mae_db": 1.6056613731728178, "rms_envelope_nrmse": 0.17427138436769415, "envelope_correlation": 0.9855768713306386, "rms_gain_error_db": 1.1031110212084352, "reference_mean_power": 0.004505860165227205, "output_mean_power": 0.005808831941303598}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3773993983732174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011178733037365022, "output_power": 0.0014112515426768478, "reference_active": true, "delta_db": 1.0121184491079687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011348254393479664, "output_power": 0.0001580344738798061, "reference_active": true, "delta_db": 1.4382277214660215}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1976543085088728e-05, "output_power": 2.9068771865721055e-05, "reference_active": true, "delta_db": 1.214673046895823}], "active_band_count": 3, "active_band_mae_db": 1.2216730724899376}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014948818281718847, "log_rms_mae_db": 1.1776904322337685, "rms_envelope_nrmse": 0.15812434818862117, "envelope_correlation": 0.991243735036081, "rms_gain_error_db": 1.1060885263798115, "reference_mean_power": 0.004512140106290299, "output_mean_power": 0.005820917299239264}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1895509906107845e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010130343468129068, "output_power": 0.0012944598075696867, "reference_active": true, "delta_db": 1.0646440000567485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.38575581348341e-05, "output_power": 7.903972997709044e-05, "reference_active": true, "delta_db": 1.665987890001882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.096976553551289e-06, "output_power": 2.1345923179554077e-06, "reference_active": true, "delta_db": 0.0772136752506104}], "active_band_count": 3, "active_band_mae_db": 0.9359485217697471}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012259006245866322, "log_rms_mae_db": 1.7318810681541887, "rms_envelope_nrmse": 0.22580590790911093, "envelope_correlation": 0.9811701895201226, "rms_gain_error_db": 1.3427166948973233, "reference_mean_power": 0.001387125934818555, "output_mean_power": 0.001889676930736567}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.931300849598275e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005535817649712056, "output_power": 0.0007651216585849501, "reference_active": true, "delta_db": 1.4054871988587636}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.308179193682493e-05, "output_power": 0.00012086925564188518, "reference_active": true, "delta_db": 2.1850666002719126}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.782827787780852e-06, "output_power": 9.406913534604906e-06, "reference_active": true, "delta_db": 2.1130689290882456}], "active_band_count": 3, "active_band_mae_db": 1.9012075760729739}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001104267603656376, "log_rms_mae_db": 1.3444940306977664, "rms_envelope_nrmse": 0.2117825551962507, "envelope_correlation": 0.984995632259557, "rms_gain_error_db": 1.3454404944615515, "reference_mean_power": 0.0013888741987160204, "output_mean_power": 0.001893245614453536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.150356442217087e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005149581742784205, "output_power": 0.0007273973315953825, "reference_active": true, "delta_db": 1.499997471582758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.177504958330739e-05, "output_power": 5.2868759189181975e-05, "reference_active": true, "delta_db": 2.211128808685194}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.976808620230845e-07, "output_power": 9.346648578075001e-07, "reference_active": true, "delta_db": 0.6882674151121875}], "active_band_count": 3, "active_band_mae_db": 1.466464565126713}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.111137261258481e-05, "log_rms_mae_db": 1.3610448494377467, "rms_envelope_nrmse": 0.2309989246910671, "envelope_correlation": 0.9847859008394831, "rms_gain_error_db": 1.4272993066174002, "reference_mean_power": 0.00011418247196706928, "output_mean_power": 0.00015860956410700334}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.04035362622924e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.425624857591467e-05, "output_power": 6.82396109136771e-05, "reference_active": true, "delta_db": 1.8806194416485502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.422825900093454e-06, "output_power": 6.258767781460053e-06, "reference_active": true, "delta_db": 2.6210402867875526}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0991841496467888e-06, "output_power": 1.5086848568258687e-06, "reference_active": true, "delta_db": 1.3752807401593818}], "active_band_count": 3, "active_band_mae_db": 1.9589801561984947}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.646632245602786e-05, "log_rms_mae_db": 1.046874287184546, "rms_envelope_nrmse": 0.20947712373057284, "envelope_correlation": 0.9911499777139577, "rms_gain_error_db": 1.4258215163443089, "reference_mean_power": 0.0001146281153210772, "output_mean_power": 0.00015917443008421278}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.508716025767634e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.069122651510939e-05, "output_power": 6.324550954368792e-05, "reference_active": true, "delta_db": 1.915289153050915}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6543344614858788e-06, "output_power": 3.0139237969342015e-06, "reference_active": true, "delta_db": 2.6050895091804787}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6696400281137344e-08, "output_power": 6.777965519361211e-08, "reference_active": true, "delta_db": 0.7754386926047184}], "active_band_count": 3, "active_band_mae_db": 1.7652724516120377}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.742254313282893e-05, "log_rms_mae_db": 1.445644042085642, "rms_envelope_nrmse": 0.34120458882859506, "envelope_correlation": 0.9338376847250924, "rms_gain_error_db": 0.6511573261694867, "reference_mean_power": 4.027652017807587e-05, "output_mean_power": 4.679157608551439e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.067374336554813e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.634873883610167e-05, "output_power": 1.9624381123596428e-05, "reference_active": true, "delta_db": 0.7931171345717882}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.842264617940254e-06, "output_power": 9.782496982748917e-06, "reference_active": true, "delta_db": 2.2386849872286803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8971365027113356e-06, "output_power": 4.110779835826922e-06, "reference_active": true, "delta_db": 3.358256372051363}], "active_band_count": 3, "active_band_mae_db": 2.1300194979506104}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.2287916076198396e-05, "log_rms_mae_db": 0.9185091997757932, "rms_envelope_nrmse": 0.21173901267750678, "envelope_correlation": 0.9752261264894855, "rms_gain_error_db": 0.6512389425660536, "reference_mean_power": 4.047626840683265e-05, "output_mean_power": 4.702451893813308e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8536334265621397e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7722714720556858e-05, "output_power": 2.290254990236687e-05, "reference_active": true, "delta_db": 1.1135359132331188}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2765957035699146e-06, "output_power": 5.906495420871839e-06, "reference_active": true, "delta_db": 2.5590701415247445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7512938438165208e-07, "output_power": 3.938930218442239e-07, "reference_active": true, "delta_db": 3.520192659389682}], "active_band_count": 3, "active_band_mae_db": 2.397599571382515}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.962428029106226, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "63940d4bfc799907724b7363f40b25e08a23af33b8e045e85a3c9d229d832137", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9847859008394831, "standard_deviation": null, "minimum": 0.9847859008394831, "maximum": 0.9847859008394831, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3610448494377467, "standard_deviation": null, "minimum": 1.3610448494377467, "maximum": 1.3610448494377467, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.9589801561984947, "standard_deviation": null, "minimum": 1.9589801561984947, "maximum": 1.9589801561984947, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2309989246910671, "standard_deviation": null, "minimum": 0.2309989246910671, "maximum": 0.2309989246910671, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9911499777139577, "standard_deviation": null, "minimum": 0.9911499777139577, "maximum": 0.9911499777139577, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.046874287184546, "standard_deviation": null, "minimum": 1.046874287184546, "maximum": 1.046874287184546, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.7652724516120377, "standard_deviation": null, "minimum": 1.7652724516120377, "maximum": 1.7652724516120377, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20947712373057284, "standard_deviation": null, "minimum": 0.20947712373057284, "maximum": 0.20947712373057284, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9338376847250924, "standard_deviation": null, "minimum": 0.9338376847250924, "maximum": 0.9338376847250924, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.445644042085642, "standard_deviation": null, "minimum": 1.445644042085642, "maximum": 1.445644042085642, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.1300194979506104, "standard_deviation": null, "minimum": 2.1300194979506104, "maximum": 2.1300194979506104, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.34120458882859506, "standard_deviation": null, "minimum": 0.34120458882859506, "maximum": 0.34120458882859506, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9752261264894855, "standard_deviation": null, "minimum": 0.9752261264894855, "maximum": 0.9752261264894855, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9185091997757932, "standard_deviation": null, "minimum": 0.9185091997757932, "maximum": 0.9185091997757932, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.397599571382515, "standard_deviation": null, "minimum": 2.397599571382515, "maximum": 2.397599571382515, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.21173901267750678, "standard_deviation": null, "minimum": 0.21173901267750678, "maximum": 0.21173901267750678, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9811701895201226, "standard_deviation": null, "minimum": 0.9811701895201226, "maximum": 0.9811701895201226, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7318810681541887, "standard_deviation": null, "minimum": 1.7318810681541887, "maximum": 1.7318810681541887, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.9012075760729739, "standard_deviation": null, "minimum": 1.9012075760729739, "maximum": 1.9012075760729739, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22580590790911093, "standard_deviation": null, "minimum": 0.22580590790911093, "maximum": 0.22580590790911093, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.984995632259557, "standard_deviation": null, "minimum": 0.984995632259557, "maximum": 0.984995632259557, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3444940306977664, "standard_deviation": null, "minimum": 1.3444940306977664, "maximum": 1.3444940306977664, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.466464565126713, "standard_deviation": null, "minimum": 1.466464565126713, "maximum": 1.466464565126713, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2117825551962507, "standard_deviation": null, "minimum": 0.2117825551962507, "maximum": 0.2117825551962507, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9698868287388157, "standard_deviation": null, "minimum": 0.9698868287388157, "maximum": 0.9698868287388157, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.254967172523161, "standard_deviation": null, "minimum": 5.254967172523161, "maximum": 5.254967172523161, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4095593649295868, "standard_deviation": null, "minimum": 0.4095593649295868, "maximum": 0.4095593649295868, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1013269797669467, "standard_deviation": null, "minimum": 1.1013269797669467, "maximum": 1.1013269797669467, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9855768713306386, "standard_deviation": null, "minimum": 0.9855768713306386, "maximum": 0.9855768713306386, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6056613731728178, "standard_deviation": null, "minimum": 1.6056613731728178, "maximum": 1.6056613731728178, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.2216730724899376, "standard_deviation": null, "minimum": 1.2216730724899376, "maximum": 1.2216730724899376, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17427138436769415, "standard_deviation": null, "minimum": 0.17427138436769415, "maximum": 0.17427138436769415, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.991243735036081, "standard_deviation": null, "minimum": 0.991243735036081, "maximum": 0.991243735036081, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1776904322337685, "standard_deviation": null, "minimum": 1.1776904322337685, "maximum": 1.1776904322337685, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9359485217697471, "standard_deviation": null, "minimum": 0.9359485217697471, "maximum": 0.9359485217697471, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.15812434818862117, "standard_deviation": null, "minimum": 0.15812434818862117, "maximum": 0.15812434818862117, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1995"}, {"case": "librispeech_1995-1836-0014", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.097271203994751, "predicted_mos": 3.097271203994751, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ccee75dba242dc7a8898d8fa0bc06ad90839b1297aa9f49dd346fb5f51084139", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.733298500068486}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.598098224326099, "全帯域_10ms_rms_envelope_nrmse": 0.2057057499888545, "全帯域_10ms_envelope_correlation": 0.9801557723903328, "全帯域_10ms_modulation_mae_db": 1.062864538989026, "全帯域_40ms_log_rms_mae_db": 1.143503482378612, "全帯域_40ms_rms_envelope_nrmse": 0.18315906320579797, "全帯域_40ms_envelope_correlation": 0.9890398590695366, "全帯域_40ms_modulation_mae_db": 0.9030771306038048, "300–1000 Hz_10ms_log_rms_mae_db": 1.7994380689029448, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2030821770922666, "300–1000 Hz_10ms_envelope_correlation": 0.985601881775929, "300–1000 Hz_10ms_modulation_mae_db": 0.661707091708304, "300–1000 Hz_40ms_log_rms_mae_db": 1.4115789571701438, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17938621457463186, "300–1000 Hz_40ms_envelope_correlation": 0.991756694121075, "300–1000 Hz_40ms_modulation_mae_db": 0.44266303997809914, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4880869491358297, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22352009282473043, "1000–3000 Hz_10ms_envelope_correlation": 0.9907548994228043, "1000–3000 Hz_10ms_modulation_mae_db": 1.668742012571047, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0698798588027003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.2019751153343628, "1000–3000 Hz_40ms_envelope_correlation": 0.9962134211496904, "1000–3000 Hz_40ms_modulation_mae_db": 1.346252803075379, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5895465215324918, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23018891690198956, "3000–6000 Hz_10ms_envelope_correlation": 0.9598508104254265, "3000–6000 Hz_10ms_modulation_mae_db": 0.7024064544026486, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9612671488000628, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1369055082625874, "3000–6000 Hz_40ms_envelope_correlation": 0.9847079648640874, "3000–6000 Hz_40ms_modulation_mae_db": 1.5298439796458236, "mr_stft_db": 5.244222160211188, "band_mae_db": 0.9527451292507764, "rms_gain_error_db": 1.2411420148328576, "output_peak": 0.595760250758433}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.024826611327636, "spectral_convergence": 0.5558402800541352, "spectral_convergence_status": "ok", "amplitude_floor": 1.7246496227499867e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.583002400206414, "spectral_convergence": 0.24148659924416882, "spectral_convergence_status": "ok", "amplitude_floor": 1.4526318235172953e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.124837469099512, "spectral_convergence": 0.26718408885702644, "spectral_convergence_status": "ok", "amplitude_floor": 1.3962213853324247e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0014645134106370877, "output_power": 0.0016640352762926946, "delta_db": 0.5546917587129797, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024771538682985465, "output_power": 0.0035849301423467638, "delta_db": 1.6052771361206684, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013427666217707404, "output_power": 0.0017798999317306982, "delta_db": 1.2239504929633012, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024269361162520585, "output_power": 0.0003459183001186946, "delta_db": 1.5391519347000289, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.4425202041437504e-05, "output_power": 5.2070971241190013e-05, "delta_db": 0.6896626671366324, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.4446805694883255e-05, "output_power": 3.3633748837385283e-05, "delta_db": -0.10373678587104722, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9527451292507764, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.2411420148328576, "pitch": null, "output_rms_dbfs": -21.27232742272588, "output_peak": 0.595760250758433, "reference_rms_dbfs": -22.513469437558737, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.244222160211188}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002110151907366841, "log_rms_mae_db": 1.598098224326099, "rms_envelope_nrmse": 0.2057057499888545, "envelope_correlation": 0.9801557723903328, "rms_gain_error_db": 1.244180470226546, "reference_mean_power": 0.005604891233518836, "output_mean_power": 0.007464233835144157}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2250047113785757e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011643562088722453, "output_power": 0.0016936059981065542, "reference_active": true, "delta_db": 1.627265197457014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011188245530999946, "output_power": 0.00012857452146924987, "reference_active": true, "delta_db": 0.6039292804316453}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8450792038001933e-05, "output_power": 3.5467801978607544e-05, "reference_active": true, "delta_db": 0.9573991390784192}], "active_band_count": 3, "active_band_mae_db": 1.062864538989026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020713167740975364, "log_rms_mae_db": 1.143503482378612, "rms_envelope_nrmse": 0.18315906320579797, "envelope_correlation": 0.9890398590695366, "rms_gain_error_db": 1.2442602389692314, "reference_mean_power": 0.005603505754865554, "output_mean_power": 0.007462525809913954}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.048317169639576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010458478077031374, "output_power": 0.0015504445052642816, "reference_active": true, "delta_db": 1.7098773586141114}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.131979407845485e-05, "output_power": 6.816957816419903e-05, "reference_active": true, "delta_db": 0.45989918821518677}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6292932861003563e-06, "output_power": 1.8447807254441034e-06, "reference_active": true, "delta_db": 0.5394548449821162}], "active_band_count": 3, "active_band_mae_db": 0.9030771306038048}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015831513319989338, "log_rms_mae_db": 1.7994380689029448, "rms_envelope_nrmse": 0.2030821770922666, "envelope_correlation": 0.985601881775929, "rms_gain_error_db": 1.2131798924118515, "reference_mean_power": 0.0022309508329157127, "output_mean_power": 0.0029499047172978515}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0339679622838414e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046453953265101344, "output_power": 0.0005384212166775769, "reference_active": true, "delta_db": 0.6409948640242362}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.019812862224435e-05, "output_power": 8.302152398818314e-05, "reference_active": true, "delta_db": 0.7286516662220703}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.545275737486218e-05, "output_power": 1.7805470749449218e-05, "reference_active": true, "delta_db": 0.6154747448786053}], "active_band_count": 3, "active_band_mae_db": 0.661707091708304}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014375880677218868, "log_rms_mae_db": 1.4115789571701438, "rms_envelope_nrmse": 0.17938621457463186, "envelope_correlation": 0.991756694121075, "rms_gain_error_db": 1.2070796518691882, "reference_mean_power": 0.0022214967634666704, "output_mean_power": 0.002933280870866718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.581275628581243e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00040407180276463134, "output_power": 0.0004704452495366576, "reference_active": true, "delta_db": 0.6605054222716901}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0351036816934105e-05, "output_power": 4.612171212421993e-05, "reference_active": true, "delta_db": 0.5805072237902973}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1841518228550588e-06, "output_power": 1.160672622188061e-06, "reference_active": true, "delta_db": -0.08697647387230997}], "active_band_count": 3, "active_band_mae_db": 0.44266303997809914}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.952734804355257e-05, "log_rms_mae_db": 1.4880869491358297, "rms_envelope_nrmse": 0.22352009282473043, "envelope_correlation": 0.9907548994228043, "rms_gain_error_db": 1.4722262773323704, "reference_mean_power": 0.00025078805819662737, "output_mean_power": 0.0003519893152553288}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4212041991971396e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.171024783262431e-05, "output_power": 0.00011448016833448028, "reference_active": true, "delta_db": 2.0314903591137172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.275733290040072e-06, "output_power": 1.0793994861520855e-05, "reference_active": true, "delta_db": 1.713054364362595}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0724449404458702e-06, "output_power": 1.4339801383373771e-06, "reference_active": true, "delta_db": 1.261681314236829}], "active_band_count": 3, "active_band_mae_db": 1.668742012571047}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.039214481073785e-05, "log_rms_mae_db": 1.0698798588027003, "rms_envelope_nrmse": 0.2019751153343628, "envelope_correlation": 0.9962134211496904, "rms_gain_error_db": 1.4716782167213287, "reference_mean_power": 0.0002518169764916092, "output_mean_power": 0.00035338883709768837}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3611130645010071e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.488974762253847e-05, "output_power": 0.00010541698416444221, "reference_active": true, "delta_db": 2.1073450243757046}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.711335389958727e-06, "output_power": 5.304066801626013e-06, "reference_active": true, "delta_db": 1.5507878220240365}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.486818337389537e-08, "output_power": 9.264190190175872e-08, "reference_active": true, "delta_db": 0.38062556282639637}], "active_band_count": 3, "active_band_mae_db": 1.346252803075379}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.8930089256612995e-05, "log_rms_mae_db": 1.5895465215324918, "rms_envelope_nrmse": 0.23018891690198956, "envelope_correlation": 0.9598508104254265, "rms_gain_error_db": 0.15471664990118245, "reference_mean_power": 2.8173794881918634e-05, "output_mean_power": 2.9195573682799803e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9380195589864167e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.123435668533653e-06, "output_power": 5.919987782546056e-06, "reference_active": true, "delta_db": -0.14674349157732225}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.722245443537651e-06, "output_power": 2.011279675815771e-06, "reference_active": true, "delta_db": -1.3145481452610432}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.909737577930936e-07, "output_power": 5.954821958389708e-07, "reference_active": true, "delta_db": -0.6459277263695802}], "active_band_count": 3, "active_band_mae_db": 0.7024064544026486}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3848385775447824e-05, "log_rms_mae_db": 0.9612671488000628, "rms_envelope_nrmse": 0.1369055082625874, "envelope_correlation": 0.9847079648640874, "rms_gain_error_db": 0.15527021729443893, "reference_mean_power": 2.8310170166312676e-05, "output_mean_power": 2.934063451322172e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.80878248957878e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.951898475340433e-06, "output_power": 5.6550001729406126e-06, "reference_active": true, "delta_db": -0.222228920688139}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2501096733423988e-06, "output_power": 8.566644139191624e-07, "reference_active": true, "delta_db": -1.6413738919792444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.567916579161296e-08, "output_power": 3.5061840957688897e-08, "reference_active": true, "delta_db": -2.7259291262700875}], "active_band_count": 3, "active_band_mae_db": 1.5298439796458236}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9585595075154684, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "37129eac4737e6be8c8abab6d1e8ef106db454877f04f3730d142fb4042300c6", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9907548994228043, "standard_deviation": null, "minimum": 0.9907548994228043, "maximum": 0.9907548994228043, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4880869491358297, "standard_deviation": null, "minimum": 1.4880869491358297, "maximum": 1.4880869491358297, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.668742012571047, "standard_deviation": null, "minimum": 1.668742012571047, "maximum": 1.668742012571047, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22352009282473043, "standard_deviation": null, "minimum": 0.22352009282473043, "maximum": 0.22352009282473043, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962134211496904, "standard_deviation": null, "minimum": 0.9962134211496904, "maximum": 0.9962134211496904, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0698798588027003, "standard_deviation": null, "minimum": 1.0698798588027003, "maximum": 1.0698798588027003, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.346252803075379, "standard_deviation": null, "minimum": 1.346252803075379, "maximum": 1.346252803075379, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2019751153343628, "standard_deviation": null, "minimum": 0.2019751153343628, "maximum": 0.2019751153343628, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9598508104254265, "standard_deviation": null, "minimum": 0.9598508104254265, "maximum": 0.9598508104254265, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5895465215324918, "standard_deviation": null, "minimum": 1.5895465215324918, "maximum": 1.5895465215324918, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7024064544026486, "standard_deviation": null, "minimum": 0.7024064544026486, "maximum": 0.7024064544026486, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23018891690198956, "standard_deviation": null, "minimum": 0.23018891690198956, "maximum": 0.23018891690198956, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9847079648640874, "standard_deviation": null, "minimum": 0.9847079648640874, "maximum": 0.9847079648640874, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9612671488000628, "standard_deviation": null, "minimum": 0.9612671488000628, "maximum": 0.9612671488000628, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5298439796458236, "standard_deviation": null, "minimum": 1.5298439796458236, "maximum": 1.5298439796458236, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1369055082625874, "standard_deviation": null, "minimum": 0.1369055082625874, "maximum": 0.1369055082625874, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.985601881775929, "standard_deviation": null, "minimum": 0.985601881775929, "maximum": 0.985601881775929, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7994380689029448, "standard_deviation": null, "minimum": 1.7994380689029448, "maximum": 1.7994380689029448, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.661707091708304, "standard_deviation": null, "minimum": 0.661707091708304, "maximum": 0.661707091708304, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2030821770922666, "standard_deviation": null, "minimum": 0.2030821770922666, "maximum": 0.2030821770922666, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.991756694121075, "standard_deviation": null, "minimum": 0.991756694121075, "maximum": 0.991756694121075, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4115789571701438, "standard_deviation": null, "minimum": 1.4115789571701438, "maximum": 1.4115789571701438, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.44266303997809914, "standard_deviation": null, "minimum": 0.44266303997809914, "maximum": 0.44266303997809914, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17938621457463186, "standard_deviation": null, "minimum": 0.17938621457463186, "maximum": 0.17938621457463186, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9527451292507764, "standard_deviation": null, "minimum": 0.9527451292507764, "maximum": 0.9527451292507764, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.244222160211188, "standard_deviation": null, "minimum": 5.244222160211188, "maximum": 5.244222160211188, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.595760250758433, "standard_deviation": null, "minimum": 0.595760250758433, "maximum": 0.595760250758433, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.2411420148328576, "standard_deviation": null, "minimum": 1.2411420148328576, "maximum": 1.2411420148328576, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9801557723903328, "standard_deviation": null, "minimum": 0.9801557723903328, "maximum": 0.9801557723903328, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.598098224326099, "standard_deviation": null, "minimum": 1.598098224326099, "maximum": 1.598098224326099, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.062864538989026, "standard_deviation": null, "minimum": 1.062864538989026, "maximum": 1.062864538989026, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2057057499888545, "standard_deviation": null, "minimum": 0.2057057499888545, "maximum": 0.2057057499888545, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9890398590695366, "standard_deviation": null, "minimum": 0.9890398590695366, "maximum": 0.9890398590695366, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.143503482378612, "standard_deviation": null, "minimum": 1.143503482378612, "maximum": 1.143503482378612, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9030771306038048, "standard_deviation": null, "minimum": 0.9030771306038048, "maximum": 0.9030771306038048, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18315906320579797, "standard_deviation": null, "minimum": 0.18315906320579797, "maximum": 0.18315906320579797, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1995"}, {"case": "librispeech_1995-1837-0023", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.212294816970825, "predicted_mos": 3.212294816970825, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0690f481d9ba0d385c9aef4bd8f16ff09982ad9549e49668003ed388caa9aa6b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.718819099944085}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4905871621424207, "全帯域_10ms_rms_envelope_nrmse": 0.1611856142099383, "全帯域_10ms_envelope_correlation": 0.9874757032999598, "全帯域_10ms_modulation_mae_db": 1.1541875349427233, "全帯域_40ms_log_rms_mae_db": 1.2278879449753275, "全帯域_40ms_rms_envelope_nrmse": 0.1457660062974229, "全帯域_40ms_envelope_correlation": 0.9929763972052873, "全帯域_40ms_modulation_mae_db": 0.7674157597839534, "300–1000 Hz_10ms_log_rms_mae_db": 1.6084401440069325, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21901970470786303, "300–1000 Hz_10ms_envelope_correlation": 0.9823986874643236, "300–1000 Hz_10ms_modulation_mae_db": 1.485905314406961, "300–1000 Hz_40ms_log_rms_mae_db": 1.3333796665013027, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.20048890154107035, "300–1000 Hz_40ms_envelope_correlation": 0.9878441141452312, "300–1000 Hz_40ms_modulation_mae_db": 1.3458927380107941, "1000–3000 Hz_10ms_log_rms_mae_db": 1.473146641256311, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23798857106173332, "1000–3000 Hz_10ms_envelope_correlation": 0.9895907193163905, "1000–3000 Hz_10ms_modulation_mae_db": 1.712289594591299, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1701060240507806, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.22632483120322278, "1000–3000 Hz_40ms_envelope_correlation": 0.9933210965239977, "1000–3000 Hz_40ms_modulation_mae_db": 1.4308708793653047, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3726473978864242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21980500485528806, "3000–6000 Hz_10ms_envelope_correlation": 0.9726722159475053, "3000–6000 Hz_10ms_modulation_mae_db": 0.28472814634574634, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9842638374249958, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.155650908014714, "3000–6000 Hz_40ms_envelope_correlation": 0.9878494763325266, "3000–6000 Hz_40ms_modulation_mae_db": 0.2602355440439659, "mr_stft_db": 4.982928619034148, "band_mae_db": 1.1201947117072577, "rms_gain_error_db": 1.0327047558741556, "output_peak": 0.48782654002256}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.2832503061015315, "spectral_convergence": 0.44412321893378426, "spectral_convergence_status": "ok", "amplitude_floor": 1.5362649016028353e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.532993173637835, "spectral_convergence": 0.21593292689035257, "spectral_convergence_status": "ok", "amplitude_floor": 1.0749003550909146e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.132542377363078, "spectral_convergence": 0.24185850093761913, "spectral_convergence_status": "ok", "amplitude_floor": 8.625899017329306e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001501246216341323, "output_power": 0.0018005790819343188, "delta_db": 0.7896027462035181, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002746582421816528, "output_power": 0.003443006223887445, "delta_db": 0.9814517153843193, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011276426514964012, "output_power": 0.0015394547065884546, "delta_db": 1.3519542164694718, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016504304929751857, "output_power": 0.0002450964041395996, "delta_db": 1.7173970070956432, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.934184208485853e-05, "output_power": 5.3343653433131524e-05, "delta_db": 1.322280657727157, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8643647097645308e-05, "output_power": 3.25744061577676e-05, "delta_db": 0.558481927363436, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.1201947117072577, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0327047558741556, "pitch": null, "output_rms_dbfs": -21.47882813511115, "output_peak": 0.48782654002256, "reference_rms_dbfs": -22.511532890985304, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.982928619034148}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018834661138467316, "log_rms_mae_db": 1.4905871621424207, "rms_envelope_nrmse": 0.1611856142099383, "envelope_correlation": 0.9874757032999598, "rms_gain_error_db": 1.0371960433605834, "reference_mean_power": 0.005606081823103195, "output_mean_power": 0.007118345060344876}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7511238572638392e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009260227506744139, "output_power": 0.001167615406784804, "reference_active": true, "delta_db": 1.006781602774856}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010953814679795108, "output_power": 0.00013647306459252726, "reference_active": true, "delta_db": 0.9548155490826713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.583845526426034e-05, "output_power": 3.6505902653336534e-05, "reference_active": true, "delta_db": 1.5009654529706429}], "active_band_count": 3, "active_band_mae_db": 1.1541875349427233}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015988223985812614, "log_rms_mae_db": 1.2278879449753275, "rms_envelope_nrmse": 0.1457660062974229, "envelope_correlation": 0.9929763972052873, "rms_gain_error_db": 1.0415600368053655, "reference_mean_power": 0.00558793457857638, "output_mean_power": 0.007102435797355573}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5615527006220852e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008311543470725633, "output_power": 0.0010539004467562624, "reference_active": true, "delta_db": 1.0311790796002112}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.497505864184299e-05, "output_power": 6.974560350275751e-05, "reference_active": true, "delta_db": 1.0335113524358233}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0542332226205543e-06, "output_power": 2.1697286988763605e-06, "reference_active": true, "delta_db": 0.23755684731582627}], "active_band_count": 3, "active_band_mae_db": 0.7674157597839534}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016200114302549476, "log_rms_mae_db": 1.6084401440069325, "rms_envelope_nrmse": 0.21901970470786303, "envelope_correlation": 0.9823986874643236, "rms_gain_error_db": 1.308457310679532, "reference_mean_power": 0.0020825301889685474, "output_mean_power": 0.002814731911335115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.598844168109503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038009245216565237, "output_power": 0.0004755274117597168, "reference_active": true, "delta_db": 0.9728631136765336}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.129961338774397e-05, "output_power": 9.250436836144045e-05, "reference_active": true, "delta_db": 1.7870450657847416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4399990501668035e-05, "output_power": 2.1288398048027014e-05, "reference_active": true, "delta_db": 1.697807763759608}], "active_band_count": 3, "active_band_mae_db": 1.485905314406961}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012179662863477574, "log_rms_mae_db": 1.3333796665013027, "rms_envelope_nrmse": 0.20048890154107035, "envelope_correlation": 0.9878441141452312, "rms_gain_error_db": 1.3125278751154659, "reference_mean_power": 0.0020746475669658963, "output_mean_power": 0.002806707263607189}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.889619852823582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035505972043036534, "output_power": 0.0004487968709523879, "reference_active": true, "delta_db": 1.017484135437521}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2260222284985983e-05, "output_power": 4.892940832847926e-05, "reference_active": true, "delta_db": 1.8090260863638916}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1886983564224358e-06, "output_power": 1.5710444088138593e-06, "reference_active": true, "delta_db": 1.2111679922309704}], "active_band_count": 3, "active_band_mae_db": 1.3458927380107941}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.2267966954040695e-05, "log_rms_mae_db": 1.473146641256311, "rms_envelope_nrmse": 0.23798857106173332, "envelope_correlation": 0.9895907193163905, "rms_gain_error_db": 1.5777673206743246, "reference_mean_power": 0.00015543877011727877, "output_mean_power": 0.0002235301365360364}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4080839081981722e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.101806003601554e-05, "output_power": 0.000114333529176807, "reference_active": true, "delta_db": 2.0680480450510736}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0132703192354966e-06, "output_power": 3.2242332176060136e-06, "reference_active": true, "delta_db": 2.0452435700738603}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.589135119213016e-07, "output_power": 9.606164116456764e-07, "reference_active": true, "delta_db": 1.0235771686489623}], "active_band_count": 3, "active_band_mae_db": 1.712289594591299}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.184196106800431e-05, "log_rms_mae_db": 1.1701060240507806, "rms_envelope_nrmse": 0.22632483120322278, "envelope_correlation": 0.9933210965239977, "rms_gain_error_db": 1.5917421792321096, "reference_mean_power": 0.00015383740434286893, "output_mean_power": 0.00022194029628976194}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3572636139953185e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.549215276795424e-05, "output_power": 0.00010618587090688, "reference_active": true, "delta_db": 2.0987746716706144}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.98253690709034e-07, "output_power": 1.6027768405244305e-06, "reference_active": true, "delta_db": 2.056321338989147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.830181704119583e-08, "output_power": 4.6796331177810004e-08, "reference_active": true, "delta_db": -0.13751662743615262}], "active_band_count": 3, "active_band_mae_db": 1.4308708793653047}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.781014610300033e-05, "log_rms_mae_db": 1.3726473978864242, "rms_envelope_nrmse": 0.21980500485528806, "envelope_correlation": 0.9726722159475053, "rms_gain_error_db": 0.6763832876103488, "reference_mean_power": 2.2945322907755735e-05, "output_mean_power": 2.6812203232861492e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.468236349213924e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8725122830509875e-06, "output_power": 4.113062682310141e-06, "reference_active": true, "delta_db": 0.2617252417377335}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.737803238621202e-07, "output_power": 5.953011763345322e-07, "reference_active": true, "delta_db": 0.15991089441397124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.433384337397866e-07, "output_power": 1.5834978750318758e-07, "reference_active": true, "delta_db": 0.4325483028855343}], "active_band_count": 3, "active_band_mae_db": 0.28472814634574634}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.4952988731037596e-05, "log_rms_mae_db": 0.9842638374249958, "rms_envelope_nrmse": 0.155650908014714, "envelope_correlation": 0.9878494763325266, "rms_gain_error_db": 0.674947604647897, "reference_mean_power": 2.2958960928334722e-05, "output_mean_power": 2.6819272279764933e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.967813463196935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5837220182078332e-06, "output_power": 3.834389553909431e-06, "reference_active": true, "delta_db": 0.29361917844259083}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.664756741318262e-07, "output_power": 2.588308104233032e-07, "reference_active": true, "delta_db": -0.1264159741058503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1431200548978765e-08, "output_power": 1.052021695253565e-08, "reference_active": true, "delta_db": -0.3606714795834566}], "active_band_count": 3, "active_band_mae_db": 0.2602355440439659}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9654596626077321, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7bf004ddd2b71e3d6034eacedfdce91f26dff74f40ac42e7ab363afc2233ec59", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895907193163905, "standard_deviation": null, "minimum": 0.9895907193163905, "maximum": 0.9895907193163905, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.473146641256311, "standard_deviation": null, "minimum": 1.473146641256311, "maximum": 1.473146641256311, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.712289594591299, "standard_deviation": null, "minimum": 1.712289594591299, "maximum": 1.712289594591299, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23798857106173332, "standard_deviation": null, "minimum": 0.23798857106173332, "maximum": 0.23798857106173332, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9933210965239977, "standard_deviation": null, "minimum": 0.9933210965239977, "maximum": 0.9933210965239977, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1701060240507806, "standard_deviation": null, "minimum": 1.1701060240507806, "maximum": 1.1701060240507806, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4308708793653047, "standard_deviation": null, "minimum": 1.4308708793653047, "maximum": 1.4308708793653047, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.22632483120322278, "standard_deviation": null, "minimum": 0.22632483120322278, "maximum": 0.22632483120322278, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9726722159475053, "standard_deviation": null, "minimum": 0.9726722159475053, "maximum": 0.9726722159475053, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3726473978864242, "standard_deviation": null, "minimum": 1.3726473978864242, "maximum": 1.3726473978864242, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.28472814634574634, "standard_deviation": null, "minimum": 0.28472814634574634, "maximum": 0.28472814634574634, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21980500485528806, "standard_deviation": null, "minimum": 0.21980500485528806, "maximum": 0.21980500485528806, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9878494763325266, "standard_deviation": null, "minimum": 0.9878494763325266, "maximum": 0.9878494763325266, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9842638374249958, "standard_deviation": null, "minimum": 0.9842638374249958, "maximum": 0.9842638374249958, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.2602355440439659, "standard_deviation": null, "minimum": 0.2602355440439659, "maximum": 0.2602355440439659, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.155650908014714, "standard_deviation": null, "minimum": 0.155650908014714, "maximum": 0.155650908014714, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9823986874643236, "standard_deviation": null, "minimum": 0.9823986874643236, "maximum": 0.9823986874643236, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6084401440069325, "standard_deviation": null, "minimum": 1.6084401440069325, "maximum": 1.6084401440069325, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.485905314406961, "standard_deviation": null, "minimum": 1.485905314406961, "maximum": 1.485905314406961, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21901970470786303, "standard_deviation": null, "minimum": 0.21901970470786303, "maximum": 0.21901970470786303, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9878441141452312, "standard_deviation": null, "minimum": 0.9878441141452312, "maximum": 0.9878441141452312, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3333796665013027, "standard_deviation": null, "minimum": 1.3333796665013027, "maximum": 1.3333796665013027, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3458927380107941, "standard_deviation": null, "minimum": 1.3458927380107941, "maximum": 1.3458927380107941, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20048890154107035, "standard_deviation": null, "minimum": 0.20048890154107035, "maximum": 0.20048890154107035, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.1201947117072577, "standard_deviation": null, "minimum": 1.1201947117072577, "maximum": 1.1201947117072577, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 4.982928619034148, "standard_deviation": null, "minimum": 4.982928619034148, "maximum": 4.982928619034148, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.48782654002256, "standard_deviation": null, "minimum": 0.48782654002256, "maximum": 0.48782654002256, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0327047558741556, "standard_deviation": null, "minimum": 1.0327047558741556, "maximum": 1.0327047558741556, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9874757032999598, "standard_deviation": null, "minimum": 0.9874757032999598, "maximum": 0.9874757032999598, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4905871621424207, "standard_deviation": null, "minimum": 1.4905871621424207, "maximum": 1.4905871621424207, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1541875349427233, "standard_deviation": null, "minimum": 1.1541875349427233, "maximum": 1.1541875349427233, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1611856142099383, "standard_deviation": null, "minimum": 0.1611856142099383, "maximum": 0.1611856142099383, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9929763972052873, "standard_deviation": null, "minimum": 0.9929763972052873, "maximum": 0.9929763972052873, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2278879449753275, "standard_deviation": null, "minimum": 1.2278879449753275, "maximum": 1.2278879449753275, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7674157597839534, "standard_deviation": null, "minimum": 0.7674157597839534, "maximum": 0.7674157597839534, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1457660062974229, "standard_deviation": null, "minimum": 0.1457660062974229, "maximum": 0.1457660062974229, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_1995"}, {"case": "librispeech_2094-142345-0032", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2500953674316406, "predicted_mos": 3.2500953674316406, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 51840, "input_sample_rate": 16000, "input_seconds": 3.24, "input_float32_sha256": "f322ecc2704a8246c85e0192659e898bf2de678ed369b5e0d9c9906b69eb11a3", "internal_samples": 51840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.112649900373071}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.7415602461611224, "全帯域_10ms_rms_envelope_nrmse": 0.1683058166867041, "全帯域_10ms_envelope_correlation": 0.9787522167333059, "全帯域_10ms_modulation_mae_db": 1.8080278027433867, "全帯域_40ms_log_rms_mae_db": 2.4487156295575527, "全帯域_40ms_rms_envelope_nrmse": 0.14768020170237187, "全帯域_40ms_envelope_correlation": 0.9859946774886983, "全帯域_40ms_modulation_mae_db": 1.5176476554062746, "300–1000 Hz_10ms_log_rms_mae_db": 2.3272556530982893, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2592204615801287, "300–1000 Hz_10ms_envelope_correlation": 0.9768923792092883, "300–1000 Hz_10ms_modulation_mae_db": 1.3158115285739707, "300–1000 Hz_40ms_log_rms_mae_db": 2.125946382046559, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2503210756587659, "300–1000 Hz_40ms_envelope_correlation": 0.9785353404820802, "300–1000 Hz_40ms_modulation_mae_db": 1.26345480246364, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4752565128742403, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17308092255004145, "1000–3000 Hz_10ms_envelope_correlation": 0.9916667945968779, "1000–3000 Hz_10ms_modulation_mae_db": 1.142070134785673, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0465795290280848, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15412366631300203, "1000–3000 Hz_40ms_envelope_correlation": 0.9953874515106043, "1000–3000 Hz_40ms_modulation_mae_db": 1.0331980136790324, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3219413333356724, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22014309752340494, "3000–6000 Hz_10ms_envelope_correlation": 0.9682099286647502, "3000–6000 Hz_10ms_modulation_mae_db": 0.4724896125444134, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8144272219413674, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1242771242201947, "3000–6000 Hz_40ms_envelope_correlation": 0.9895445024354463, "3000–6000 Hz_40ms_modulation_mae_db": 0.2877768149614847, "mr_stft_db": 5.687446956802574, "band_mae_db": 0.5690480601788667, "rms_gain_error_db": 0.8670349566551103, "output_peak": 0.6269872038563448}, "full_scale": {"sample_count": 51840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 51840, "output_samples": 51840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.1480859883132295, "spectral_convergence": 0.5381338128109191, "spectral_convergence_status": "ok", "amplitude_floor": 1.3508970735580905e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.126212136228658, "spectral_convergence": 0.2892744014205629, "spectral_convergence_status": "ok", "amplitude_floor": 9.199259876037329e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.788042745865832, "spectral_convergence": 0.31514132576526105, "spectral_convergence_status": "ok", "amplitude_floor": 7.633709769762664e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012491242347822075, "output_power": 0.0012683832377067441, "delta_db": 0.066448596142669, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0036291285647318994, "output_power": 0.004809494597188644, "delta_db": 1.222970875039065, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000626458342936399, "output_power": 0.00067622032673418, "delta_db": 0.33196023847533374, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0008149528003014777, "output_power": 0.001042548877088281, "delta_db": 1.0696396845318354, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002174945257645064, "output_power": 0.0002471249643410437, "delta_db": 0.5546828921588303, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023982804336185733, "output_power": 0.0002306966865065355, "delta_db": -0.16858607472546638, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5690480601788667, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8670349566551103, "pitch": null, "output_rms_dbfs": -20.822598830625665, "output_peak": 0.6269872038563448, "reference_rms_dbfs": -21.689633787280776, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.687446956802574}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 51840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00019883564266349852, "log_rms_mae_db": 2.7415602461611224, "rms_envelope_nrmse": 0.1683058166867041, "envelope_correlation": 0.9787522167333059, "rms_gain_error_db": 0.867067728349661, "reference_mean_power": 0.006787407573323676, "output_mean_power": 0.008287254984155935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.1392663194097437e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011382823319318316, "output_power": 0.0017008422312929014, "reference_active": true, "delta_db": 1.7441403576973282}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00037355144462451696, "output_power": 0.0005005768691517772, "reference_active": true, "delta_db": 1.2712035740694367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.066072767849799e-05, "output_power": 3.483643204696064e-05, "reference_active": true, "delta_db": -2.408739476463396}], "active_band_count": 3, "active_band_mae_db": 1.8080278027433867}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00017748806718771712, "log_rms_mae_db": 2.4487156295575527, "rms_envelope_nrmse": 0.14768020170237187, "envelope_correlation": 0.9859946774886983, "rms_gain_error_db": 0.8672480306512574, "reference_mean_power": 0.006850619249749789, "output_mean_power": 0.008364782127238601}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.7219585966772077e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000998012429960962, "output_power": 0.0015349398138756463, "reference_active": true, "delta_db": 1.8695540080419104}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018464455345700424, "output_power": 0.0002523322867132665, "reference_active": true, "delta_db": 1.356363218267054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.884647670701786e-06, "output_power": 3.5985759202479257e-06, "reference_active": true, "delta_db": -1.32702573990986}], "active_band_count": 3, "active_band_mae_db": 1.5176476554062746}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00015076905043222532, "log_rms_mae_db": 2.3272556530982893, "rms_envelope_nrmse": 0.2592204615801287, "envelope_correlation": 0.9768923792092883, "rms_gain_error_db": 1.563249050560453, "reference_mean_power": 0.0024151558510913072, "output_mean_power": 0.0034615456732204333}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 1.3965349131952106e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007806812222828868, "output_power": 0.0012549479789560807, "reference_active": true, "delta_db": 2.0615199008013523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024651476965397817, "output_power": 0.000354039205838456, "reference_active": true, "delta_db": 1.5720841323549875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0716739338834284e-05, "output_power": 3.301856293636659e-05, "reference_active": true, "delta_db": 0.31383055256557213}], "active_band_count": 3, "active_band_mae_db": 1.3158115285739707}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00011516776273837431, "log_rms_mae_db": 2.125946382046559, "rms_envelope_nrmse": 0.2503210756587659, "envelope_correlation": 0.9785353404820802, "rms_gain_error_db": 1.5632314760886308, "reference_mean_power": 0.0024377564716401134, "output_mean_power": 0.003493924094203295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 1.190139920080037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007614290342229629, "output_power": 0.0012223472587431258, "reference_active": true, "delta_db": 2.055651699292335}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011269364256369602, "output_power": 0.00015981812058336875, "reference_active": true, "delta_db": 1.5172660248782586}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5333706288121526e-06, "output_power": 3.714786537325797e-06, "reference_active": true, "delta_db": 0.21744668322032742}], "active_band_count": 3, "active_band_mae_db": 1.26345480246364}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 0.00014489100768405854, "log_rms_mae_db": 1.4752565128742403, "rms_envelope_nrmse": 0.17308092255004145, "envelope_correlation": 0.9916667945968779, "rms_gain_error_db": 1.0622094873380827, "reference_mean_power": 0.0008223932804627593, "output_mean_power": 0.0010502688911139522}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 6.859948035043385e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004581143880628001, "output_power": 0.0006418421294984803, "reference_active": true, "delta_db": 1.464542881574973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.242911398972428e-05, "output_power": 0.00012328946629775215, "reference_active": true, "delta_db": 1.251171826066243}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.706958400708153e-06, "output_power": 7.392935204978595e-06, "reference_active": true, "delta_db": -0.7104956967158034}], "active_band_count": 3, "active_band_mae_db": 1.142070134785673}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 0.00013523948700125142, "log_rms_mae_db": 1.0465795290280848, "rms_envelope_nrmse": 0.15412366631300203, "envelope_correlation": 0.9953874515106043, "rms_gain_error_db": 1.0622118317262261, "reference_mean_power": 0.000830089539077811, "output_mean_power": 0.0010600982657134452}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 6.395142215880047e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004454767193500814, "output_power": 0.0006258748233901997, "reference_active": true, "delta_db": 1.4766246915680332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.066351786105847e-05, "output_power": 5.461879750610017e-05, "reference_active": true, "delta_db": 1.2813718704522854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.874405735020567e-07, "output_power": 8.203128840330817e-07, "reference_active": true, "delta_db": -0.34159747901677834}], "active_band_count": 3, "active_band_mae_db": 1.0331980136790324}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 647, "rms_floor": 8.841298696090765e-05, "log_rms_mae_db": 1.3219413333356724, "rms_envelope_nrmse": 0.22014309752340494, "envelope_correlation": 0.9682099286647502, "rms_gain_error_db": -0.023447850394135092, "reference_mean_power": 0.0002494763103331033, "output_mean_power": 0.00024813300057210214}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 647, "frequency_resolution_hz": 0.3091190108191654, "power_floor": 2.730062412281665e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013367937231840985, "output_power": 0.0001277809947897283, "reference_active": true, "delta_db": -0.19598132850823516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8654341368924343e-05, "output_power": 2.3444768784899224e-05, "reference_active": true, "delta_db": -0.8714447645040362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.319126391736523e-06, "output_power": 9.017413748916464e-06, "reference_active": true, "delta_db": 0.35004274462096885}], "active_band_count": 3, "active_band_mae_db": 0.4724896125444134}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 51840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 641, "rms_floor": 7.608900761326353e-05, "log_rms_mae_db": 0.8144272219413674, "rms_envelope_nrmse": 0.1242771242201947, "envelope_correlation": 0.9895445024354463, "rms_gain_error_db": -0.023449059029312422, "reference_mean_power": 0.00025181099859470626, "output_mean_power": 0.000250455047960729}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 641, "frequency_resolution_hz": 0.31201248049922, "power_floor": 2.5627299576593646e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001307843821160996, "output_power": 0.0001247158863019741, "reference_active": true, "delta_db": -0.20634107515976843}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1887209147638657e-05, "output_power": 1.0683741500929174e-05, "reference_active": true, "delta_db": -0.46356532419755003}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.396397638044017e-07, "output_power": 6.687716564841477e-07, "reference_active": true, "delta_db": 0.1934240455271356}], "active_band_count": 3, "active_band_mae_db": 0.2877768149614847}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9180559343285492, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9d82e233ace30b801bd19800a0c3e4add5200beb9b35af2428a50e25a279642f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9916667945968779, "standard_deviation": null, "minimum": 0.9916667945968779, "maximum": 0.9916667945968779, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4752565128742403, "standard_deviation": null, "minimum": 1.4752565128742403, "maximum": 1.4752565128742403, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.142070134785673, "standard_deviation": null, "minimum": 1.142070134785673, "maximum": 1.142070134785673, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17308092255004145, "standard_deviation": null, "minimum": 0.17308092255004145, "maximum": 0.17308092255004145, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9953874515106043, "standard_deviation": null, "minimum": 0.9953874515106043, "maximum": 0.9953874515106043, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0465795290280848, "standard_deviation": null, "minimum": 1.0465795290280848, "maximum": 1.0465795290280848, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0331980136790324, "standard_deviation": null, "minimum": 1.0331980136790324, "maximum": 1.0331980136790324, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15412366631300203, "standard_deviation": null, "minimum": 0.15412366631300203, "maximum": 0.15412366631300203, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9682099286647502, "standard_deviation": null, "minimum": 0.9682099286647502, "maximum": 0.9682099286647502, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3219413333356724, "standard_deviation": null, "minimum": 1.3219413333356724, "maximum": 1.3219413333356724, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4724896125444134, "standard_deviation": null, "minimum": 0.4724896125444134, "maximum": 0.4724896125444134, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22014309752340494, "standard_deviation": null, "minimum": 0.22014309752340494, "maximum": 0.22014309752340494, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9895445024354463, "standard_deviation": null, "minimum": 0.9895445024354463, "maximum": 0.9895445024354463, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8144272219413674, "standard_deviation": null, "minimum": 0.8144272219413674, "maximum": 0.8144272219413674, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.2877768149614847, "standard_deviation": null, "minimum": 0.2877768149614847, "maximum": 0.2877768149614847, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1242771242201947, "standard_deviation": null, "minimum": 0.1242771242201947, "maximum": 0.1242771242201947, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9768923792092883, "standard_deviation": null, "minimum": 0.9768923792092883, "maximum": 0.9768923792092883, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3272556530982893, "standard_deviation": null, "minimum": 2.3272556530982893, "maximum": 2.3272556530982893, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3158115285739707, "standard_deviation": null, "minimum": 1.3158115285739707, "maximum": 1.3158115285739707, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2592204615801287, "standard_deviation": null, "minimum": 0.2592204615801287, "maximum": 0.2592204615801287, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9785353404820802, "standard_deviation": null, "minimum": 0.9785353404820802, "maximum": 0.9785353404820802, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.125946382046559, "standard_deviation": null, "minimum": 2.125946382046559, "maximum": 2.125946382046559, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.26345480246364, "standard_deviation": null, "minimum": 1.26345480246364, "maximum": 1.26345480246364, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2503210756587659, "standard_deviation": null, "minimum": 0.2503210756587659, "maximum": 0.2503210756587659, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5690480601788667, "standard_deviation": null, "minimum": 0.5690480601788667, "maximum": 0.5690480601788667, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.687446956802574, "standard_deviation": null, "minimum": 5.687446956802574, "maximum": 5.687446956802574, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6269872038563448, "standard_deviation": null, "minimum": 0.6269872038563448, "maximum": 0.6269872038563448, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8670349566551103, "standard_deviation": null, "minimum": 0.8670349566551103, "maximum": 0.8670349566551103, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9787522167333059, "standard_deviation": null, "minimum": 0.9787522167333059, "maximum": 0.9787522167333059, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.7415602461611224, "standard_deviation": null, "minimum": 2.7415602461611224, "maximum": 2.7415602461611224, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.8080278027433867, "standard_deviation": null, "minimum": 1.8080278027433867, "maximum": 1.8080278027433867, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1683058166867041, "standard_deviation": null, "minimum": 0.1683058166867041, "maximum": 0.1683058166867041, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9859946774886983, "standard_deviation": null, "minimum": 0.9859946774886983, "maximum": 0.9859946774886983, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.4487156295575527, "standard_deviation": null, "minimum": 2.4487156295575527, "maximum": 2.4487156295575527, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.5176476554062746, "standard_deviation": null, "minimum": 1.5176476554062746, "maximum": 1.5176476554062746, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.14768020170237187, "standard_deviation": null, "minimum": 0.14768020170237187, "maximum": 0.14768020170237187, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2094"}, {"case": "librispeech_2094-142345-0039", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3789925575256348, "predicted_mos": 3.3789925575256348, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "67385ddf327eff466c87b84a4e0d3987b8e0a2c8b30d80e659a1fe9ee1adfb3b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.848823399748653}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.0885183375439986, "全帯域_10ms_rms_envelope_nrmse": 0.1481906290911678, "全帯域_10ms_envelope_correlation": 0.9816492294635021, "全帯域_10ms_modulation_mae_db": 1.333103042084527, "全帯域_40ms_log_rms_mae_db": 1.792189332642539, "全帯域_40ms_rms_envelope_nrmse": 0.12408863372410707, "全帯域_40ms_envelope_correlation": 0.990395026916491, "全帯域_40ms_modulation_mae_db": 1.411483772881775, "300–1000 Hz_10ms_log_rms_mae_db": 1.9910301944068483, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22007561718574759, "300–1000 Hz_10ms_envelope_correlation": 0.9836324309779734, "300–1000 Hz_10ms_modulation_mae_db": 1.8104034167799512, "300–1000 Hz_40ms_log_rms_mae_db": 1.6721253128248295, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2071046479832843, "300–1000 Hz_40ms_envelope_correlation": 0.987272767688066, "300–1000 Hz_40ms_modulation_mae_db": 1.6811417756807936, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3815469089130552, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15004679434623552, "1000–3000 Hz_10ms_envelope_correlation": 0.9911443228964442, "1000–3000 Hz_10ms_modulation_mae_db": 0.6369133648608142, "1000–3000 Hz_40ms_log_rms_mae_db": 1.003861831057405, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12538358462912882, "1000–3000 Hz_40ms_envelope_correlation": 0.9957513500363392, "1000–3000 Hz_40ms_modulation_mae_db": 0.9167776506513811, "3000–6000 Hz_10ms_log_rms_mae_db": 1.291747729253853, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15357284067264224, "3000–6000 Hz_10ms_envelope_correlation": 0.9854755167096608, "3000–6000 Hz_10ms_modulation_mae_db": 0.28785633953317197, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7564930890436584, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08464736673037652, "3000–6000 Hz_40ms_envelope_correlation": 0.995464808031995, "3000–6000 Hz_40ms_modulation_mae_db": 0.21279736892878265, "mr_stft_db": 5.605806624992596, "band_mae_db": 0.6700436763575465, "rms_gain_error_db": 0.7840056757597991, "output_peak": 0.6323670275119686}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.244723560076733, "spectral_convergence": 0.5263777842756125, "spectral_convergence_status": "ok", "amplitude_floor": 1.4935395043887774e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.995750135325015, "spectral_convergence": 0.2849850654125343, "spectral_convergence_status": "ok", "amplitude_floor": 1.0494802571211602e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.576946179576038, "spectral_convergence": 0.3040907466683447, "spectral_convergence_status": "ok", "amplitude_floor": 9.412333460941768e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00214444778551191, "output_power": 0.0023764915787987543, "delta_db": 0.44620803488736144, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003131735025595309, "output_power": 0.003971384505077093, "delta_db": 1.0315692749392942, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006987919290953902, "output_power": 0.0008733836921996066, "delta_db": 0.9685719819740757, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004380829302633605, "output_power": 0.0005382474744996267, "delta_db": 0.89425669369046, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014837331900785606, "output_power": 0.00016989820698661283, "delta_db": 0.5883298392411658, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003193593810232492, "output_power": 0.000312713805419716, "delta_db": -0.09132623341292162, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6700436763575465, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7840056757597991, "pitch": null, "output_rms_dbfs": -20.839611055503312, "output_peak": 0.6323670275119686, "reference_rms_dbfs": -21.62361673126311, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.605806624992596}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020729159449367856, "log_rms_mae_db": 2.0885183375439986, "rms_envelope_nrmse": 0.1481906290911678, "envelope_correlation": 0.9816492294635021, "rms_gain_error_db": 0.7840306373130533, "reference_mean_power": 0.0068864212769201745, "output_mean_power": 0.008248911626594352}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4588043207823824e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006447788388285564, "output_power": 0.0007978069827473197, "reference_active": true, "delta_db": 0.9248705731854783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001410929626621107, "output_power": 0.0002154804167573283, "reference_active": true, "delta_db": 1.839024539843025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.775085636154685e-05, "output_power": 3.6882282572220204e-05, "reference_active": true, "delta_db": 1.2354140132250775}], "active_band_count": 3, "active_band_mae_db": 1.333103042084527}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001768834026970723, "log_rms_mae_db": 1.792189332642539, "rms_envelope_nrmse": 0.12408863372410707, "envelope_correlation": 0.990395026916491, "rms_gain_error_db": 0.7835099486525454, "reference_mean_power": 0.006918665749586704, "output_mean_power": 0.008286542165982048}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2539809640994155e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005542590263527943, "output_power": 0.0006935711233389693, "reference_active": true, "delta_db": 0.9737822858056265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.178340930501843e-05, "output_power": 0.00011023978799398983, "reference_active": true, "delta_db": 1.863142883443408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.482399562538733e-06, "output_power": 2.0451157487576943e-06, "reference_active": true, "delta_db": 1.3975261493962907}], "active_band_count": 3, "active_band_mae_db": 1.411483772881775}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016054516244843413, "log_rms_mae_db": 1.9910301944068483, "rms_envelope_nrmse": 0.22007561718574759, "envelope_correlation": 0.9836324309779734, "rms_gain_error_db": 1.357173384903523, "reference_mean_power": 0.0021921303983993327, "output_mean_power": 0.0029962891592644524}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.01765222997539e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005084502896759698, "output_power": 0.0007483977467969017, "reference_active": true, "delta_db": 1.6788397220036924}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.688692893438751e-05, "output_power": 0.00017563442187150998, "reference_active": true, "delta_db": 2.5834444540007864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1104523408685175e-05, "output_power": 2.7622829107944736e-05, "reference_active": true, "delta_db": 1.1689260743353749}], "active_band_count": 3, "active_band_mae_db": 1.8104034167799512}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013960343413021469, "log_rms_mae_db": 1.6721253128248295, "rms_envelope_nrmse": 0.2071046479832843, "envelope_correlation": 0.987272767688066, "rms_gain_error_db": 1.357163107585098, "reference_mean_power": 0.00220313483790695, "output_mean_power": 0.0030113233294171227}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.111026246795902e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004662873703819768, "output_power": 0.0006982087646495064, "reference_active": true, "delta_db": 1.7533164369414291}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.729339985490381e-05, "output_power": 8.77207138590849e-05, "reference_active": true, "delta_db": 2.683016213014147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4261322714398465e-06, "output_power": 1.6400952665249598e-06, "reference_active": true, "delta_db": 0.6070926770868049}], "active_band_count": 3, "active_band_mae_db": 1.6811417756807936}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.530788197178361e-05, "log_rms_mae_db": 1.3815469089130552, "rms_envelope_nrmse": 0.15004679434623552, "envelope_correlation": 0.9911443228964442, "rms_gain_error_db": 0.8336912475003576, "reference_mean_power": 0.0004376816454019562, "output_mean_power": 0.0005303071213631828}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8441025227565861e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.86794553375767e-05, "output_power": 0.00012178619688183499, "reference_active": true, "delta_db": 0.9137132488011326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2863459584597701e-05, "output_power": 1.534030340300914e-05, "reference_active": true, "delta_db": 0.764761628928809}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6929848290709045e-06, "output_power": 2.840929316993019e-06, "reference_active": true, "delta_db": 0.2322652168525009}], "active_band_count": 3, "active_band_mae_db": 0.6369133648608142}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.347481502172338e-05, "log_rms_mae_db": 1.003861831057405, "rms_envelope_nrmse": 0.12538358462912882, "envelope_correlation": 0.9957513500363392, "rms_gain_error_db": 0.8336976143287169, "reference_mean_power": 0.00043987971266632326, "output_mean_power": 0.0005329711414692277}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.705293511824543e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.166386964932698e-05, "output_power": 0.00011362023927224957, "reference_active": true, "delta_db": 0.9325751220342164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.453379972535193e-06, "output_power": 6.564671847870488e-06, "reference_active": true, "delta_db": 0.8054726276431395}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5473641354942359e-07, "output_power": 1.9535343267052827e-07, "reference_active": true, "delta_db": 1.0122852022767872}], "active_band_count": 3, "active_band_mae_db": 0.9167776506513811}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012710775449688844, "log_rms_mae_db": 1.291747729253853, "rms_envelope_nrmse": 0.15357284067264224, "envelope_correlation": 0.9854755167096608, "rms_gain_error_db": 0.07325243919964704, "reference_mean_power": 0.00030261254733777104, "output_mean_power": 0.0003077600014151722}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.325105138813031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3055217502518526e-05, "output_power": 4.351364488719183e-05, "reference_active": true, "delta_db": 0.04599675658739459}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2532169127733897e-05, "output_power": 1.2433615299373587e-05, "reference_active": true, "delta_db": -0.03428821255894224}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7715638592807256e-06, "output_power": 1.4792051389541622e-06, "reference_active": true, "delta_db": -0.783284049453179}], "active_band_count": 3, "active_band_mae_db": 0.28785633953317197}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010399786917892939, "log_rms_mae_db": 0.7564930890436584, "rms_envelope_nrmse": 0.08464736673037652, "envelope_correlation": 0.995464808031995, "rms_gain_error_db": 0.07325311300567851, "reference_mean_power": 0.0003041332985815958, "output_mean_power": 0.0003093066687000562}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.904406483303471e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.339234052233676e-05, "output_power": 4.406599048123423e-05, "reference_active": true, "delta_db": 0.06690460303416906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.902501195013482e-06, "output_power": 5.849977613156702e-06, "reference_active": true, "delta_db": -0.03881879561347974}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6308355063930231e-07, "output_power": 1.4425912294938964e-07, "reference_active": true, "delta_db": -0.5326687081386992}], "active_band_count": 3, "active_band_mae_db": 0.21279736892878265}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9491287938173346, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d4cf7fc3f81df6e20ee55839972a555f38bcc946643aa33207b268bab69753f4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9911443228964442, "standard_deviation": null, "minimum": 0.9911443228964442, "maximum": 0.9911443228964442, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3815469089130552, "standard_deviation": null, "minimum": 1.3815469089130552, "maximum": 1.3815469089130552, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6369133648608142, "standard_deviation": null, "minimum": 0.6369133648608142, "maximum": 0.6369133648608142, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15004679434623552, "standard_deviation": null, "minimum": 0.15004679434623552, "maximum": 0.15004679434623552, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957513500363392, "standard_deviation": null, "minimum": 0.9957513500363392, "maximum": 0.9957513500363392, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.003861831057405, "standard_deviation": null, "minimum": 1.003861831057405, "maximum": 1.003861831057405, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9167776506513811, "standard_deviation": null, "minimum": 0.9167776506513811, "maximum": 0.9167776506513811, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12538358462912882, "standard_deviation": null, "minimum": 0.12538358462912882, "maximum": 0.12538358462912882, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9854755167096608, "standard_deviation": null, "minimum": 0.9854755167096608, "maximum": 0.9854755167096608, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.291747729253853, "standard_deviation": null, "minimum": 1.291747729253853, "maximum": 1.291747729253853, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.28785633953317197, "standard_deviation": null, "minimum": 0.28785633953317197, "maximum": 0.28785633953317197, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15357284067264224, "standard_deviation": null, "minimum": 0.15357284067264224, "maximum": 0.15357284067264224, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.995464808031995, "standard_deviation": null, "minimum": 0.995464808031995, "maximum": 0.995464808031995, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7564930890436584, "standard_deviation": null, "minimum": 0.7564930890436584, "maximum": 0.7564930890436584, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.21279736892878265, "standard_deviation": null, "minimum": 0.21279736892878265, "maximum": 0.21279736892878265, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08464736673037652, "standard_deviation": null, "minimum": 0.08464736673037652, "maximum": 0.08464736673037652, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9836324309779734, "standard_deviation": null, "minimum": 0.9836324309779734, "maximum": 0.9836324309779734, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9910301944068483, "standard_deviation": null, "minimum": 1.9910301944068483, "maximum": 1.9910301944068483, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.8104034167799512, "standard_deviation": null, "minimum": 1.8104034167799512, "maximum": 1.8104034167799512, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22007561718574759, "standard_deviation": null, "minimum": 0.22007561718574759, "maximum": 0.22007561718574759, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.987272767688066, "standard_deviation": null, "minimum": 0.987272767688066, "maximum": 0.987272767688066, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6721253128248295, "standard_deviation": null, "minimum": 1.6721253128248295, "maximum": 1.6721253128248295, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.6811417756807936, "standard_deviation": null, "minimum": 1.6811417756807936, "maximum": 1.6811417756807936, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2071046479832843, "standard_deviation": null, "minimum": 0.2071046479832843, "maximum": 0.2071046479832843, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6700436763575465, "standard_deviation": null, "minimum": 0.6700436763575465, "maximum": 0.6700436763575465, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.605806624992596, "standard_deviation": null, "minimum": 5.605806624992596, "maximum": 5.605806624992596, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6323670275119686, "standard_deviation": null, "minimum": 0.6323670275119686, "maximum": 0.6323670275119686, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7840056757597991, "standard_deviation": null, "minimum": 0.7840056757597991, "maximum": 0.7840056757597991, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9816492294635021, "standard_deviation": null, "minimum": 0.9816492294635021, "maximum": 0.9816492294635021, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.0885183375439986, "standard_deviation": null, "minimum": 2.0885183375439986, "maximum": 2.0885183375439986, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.333103042084527, "standard_deviation": null, "minimum": 1.333103042084527, "maximum": 1.333103042084527, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1481906290911678, "standard_deviation": null, "minimum": 0.1481906290911678, "maximum": 0.1481906290911678, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.990395026916491, "standard_deviation": null, "minimum": 0.990395026916491, "maximum": 0.990395026916491, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.792189332642539, "standard_deviation": null, "minimum": 1.792189332642539, "maximum": 1.792189332642539, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.411483772881775, "standard_deviation": null, "minimum": 1.411483772881775, "maximum": 1.411483772881775, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12408863372410707, "standard_deviation": null, "minimum": 0.12408863372410707, "maximum": 0.12408863372410707, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2094"}, {"case": "librispeech_2094-142345-0040", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.049264669418335, "predicted_mos": 2.049264669418335, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 38160, "input_sample_rate": 16000, "input_seconds": 2.385, "input_float32_sha256": "9ac48612f18c25bc2b31b66bc89a4701304eb232a862ab460161f5c1e664d24d", "internal_samples": 38160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.648116500116885}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 4.469489841766993, "全帯域_10ms_rms_envelope_nrmse": 0.5138648160620384, "全帯域_10ms_envelope_correlation": 0.7560042721881168, "全帯域_10ms_modulation_mae_db": 4.988444972557986, "全帯域_40ms_log_rms_mae_db": 4.327279080471564, "全帯域_40ms_rms_envelope_nrmse": 0.47763659412950993, "全帯域_40ms_envelope_correlation": 0.7690815480663353, "全帯域_40ms_modulation_mae_db": 6.129924922803636, "300–1000 Hz_10ms_log_rms_mae_db": 3.6137347650650216, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.30162641333160645, "300–1000 Hz_10ms_envelope_correlation": 0.9612655016751539, "300–1000 Hz_10ms_modulation_mae_db": 1.586867179810267, "300–1000 Hz_40ms_log_rms_mae_db": 3.4949711711427196, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.26555420218446396, "300–1000 Hz_40ms_envelope_correlation": 0.9716082386587735, "300–1000 Hz_40ms_modulation_mae_db": 0.7244627985927202, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5725685760810553, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20968466110026188, "1000–3000 Hz_10ms_envelope_correlation": 0.9848066250182821, "1000–3000 Hz_10ms_modulation_mae_db": 1.0408504765132909, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9670075429159195, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1727888765450336, "1000–3000 Hz_40ms_envelope_correlation": 0.9925524903486735, "1000–3000 Hz_40ms_modulation_mae_db": 1.1989534114355436, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4595189505131914, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18822132898132396, "3000–6000 Hz_10ms_envelope_correlation": 0.9776125962424322, "3000–6000 Hz_10ms_modulation_mae_db": 0.6027798544135546, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8074910255060063, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11012812503358614, "3000–6000 Hz_40ms_envelope_correlation": 0.9927714231544535, "3000–6000 Hz_40ms_modulation_mae_db": 0.9111657536180905, "mr_stft_db": 5.970351602277762, "band_mae_db": 1.501409585363991, "rms_gain_error_db": -1.3996589159330592, "output_peak": 0.5510417530752456}, "full_scale": {"sample_count": 38160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 38160, "output_samples": 38160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.373189127695059, "spectral_convergence": 0.7493038986251587, "spectral_convergence_status": "ok", "amplitude_floor": 5.674431450769592e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.554057611837862, "spectral_convergence": 0.7111450796867929, "spectral_convergence_status": "ok", "amplitude_floor": 3.1139992269810527e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.9838080673003615, "spectral_convergence": 0.672758437520494, "spectral_convergence_status": "ok", "amplitude_floor": 1.28722240515293e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.003603182122246331, "output_power": 0.0011851085668529544, "delta_db": -4.829280764797827, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001490228689925631, "output_power": 0.0021338174932648926, "delta_db": 1.559043510082863, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00043834980790852295, "output_power": 0.0005193424123322915, "delta_db": 0.7363297032201992, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023807008265260595, "output_power": 0.0003096671525473244, "delta_db": 1.1419031797630919, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.11014495688624e-05, "output_power": 8.636248017544975e-05, "delta_db": 0.2729648934523326, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.691475709405268e-05, "output_power": 2.4159999367012644e-05, "delta_db": -0.4689354608676341, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.501409585363991, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -1.3996589159330592, "pitch": null, "output_rms_dbfs": -23.707476210629675, "output_peak": 0.5510417530752456, "reference_rms_dbfs": -22.307817294696616, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.970351602277762}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 38160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.0004033177834583336, "log_rms_mae_db": 4.469489841766993, "rms_envelope_nrmse": 0.5138648160620384, "envelope_correlation": 0.7560042721881168, "rms_gain_error_db": -1.3996156736920806, "reference_mean_power": 0.005890133913079112, "output_mean_power": 0.004267402440812927}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 4.702783438048151e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019529842144495992, "output_power": 0.0014859131903433936, "reference_active": true, "delta_db": -1.187052950735836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0010897406087412391, "output_power": 0.0003010919375115845, "reference_active": true, "delta_db": -5.586240086568138}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00042203241052064084, "output_power": 6.399434766827326e-05, "reference_active": true, "delta_db": -8.192041880369983}], "active_band_count": 3, "active_band_mae_db": 4.988444972557986}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0002641139400842597, "log_rms_mae_db": 4.327279080471564, "rms_envelope_nrmse": 0.47763659412950993, "envelope_correlation": 0.7690815480663353, "rms_gain_error_db": -1.399434609061612, "reference_mean_power": 0.0059649962331930016, "output_mean_power": 0.004321820377244693}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.959398727002631e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020291985866854774, "output_power": 0.0012718756051162711, "reference_active": true, "delta_db": -2.0287991363584847}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005646977153887027, "output_power": 0.00014977630495371865, "reference_active": true, "delta_db": -5.763729185549043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.714480412149099e-05, "output_power": 3.2372290415443016e-06, "reference_active": true, "delta_db": -10.59724644650338}], "active_band_count": 3, "active_band_mae_db": 6.129924922803636}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 0.00011878678262042868, "log_rms_mae_db": 3.6137347650650216, "rms_envelope_nrmse": 0.30162641333160645, "envelope_correlation": 0.9612655016751539, "rms_gain_error_db": 1.3189766963740421, "reference_mean_power": 0.0009978964081743123, "output_mean_power": 0.0013520200405798407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 1.0827926032603135e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006103304193321489, "output_power": 0.0006831371924043293, "reference_active": true, "delta_db": 0.4894291444252588}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011556410212235237, "output_power": 0.00015377823334772068, "reference_active": true, "delta_db": 1.2407191790180006}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8931134571049757e-05, "output_power": 3.803837575871102e-05, "reference_active": true, "delta_db": 3.0304532159875412}], "active_band_count": 3, "active_band_mae_db": 1.586867179810267}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 9.505044043492436e-05, "log_rms_mae_db": 3.4949711711427196, "rms_envelope_nrmse": 0.26555420218446396, "envelope_correlation": 0.9716082386587735, "rms_gain_error_db": 1.3189240148585193, "reference_mean_power": 0.0010106286473201321, "output_mean_power": 0.0013692539614873955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.735673127654544e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005396175923660179, "output_power": 0.0006051508630390302, "reference_active": true, "delta_db": 0.4977755709516303}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.461345336739454e-05, "output_power": 8.75800555989445e-05, "reference_active": true, "delta_db": 1.3208226335135}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3971113807678808e-06, "output_power": 2.601160994420119e-06, "reference_active": true, "delta_db": 0.35479019131303013}], "active_band_count": 3, "active_band_mae_db": 0.7244627985927202}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 9.972124238081888e-05, "log_rms_mae_db": 1.5725685760810553, "rms_envelope_nrmse": 0.20968466110026188, "envelope_correlation": 0.9848066250182821, "rms_gain_error_db": 1.0599125171456347, "reference_mean_power": 0.00028254365881671625, "output_mean_power": 0.0003606424265614206}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 3.4051840450675846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002046408997048573, "output_power": 0.0002538602477952735, "reference_active": true, "delta_db": 0.9360226311701042}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.521356834551269e-05, "output_power": 6.617951302046617e-05, "reference_active": true, "delta_db": 1.654547833168627}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018106896586635e-06, "output_power": 9.062957396367553e-06, "reference_active": true, "delta_db": 0.5319809652011414}], "active_band_count": 3, "active_band_mae_db": 1.0408504765132909}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 8.26289228894579e-05, "log_rms_mae_db": 0.9670075429159195, "rms_envelope_nrmse": 0.1727888765450336, "envelope_correlation": 0.9925524903486735, "rms_gain_error_db": 1.0599206047776828, "reference_mean_power": 0.0002861487173387815, "output_mean_power": 0.00036524465064889015}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 3.083140770966499e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019461067978585587, "output_power": 0.00024860021595451786, "reference_active": true, "delta_db": 1.063348319024061}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0239988981192927e-05, "output_power": 2.977683581800657e-05, "reference_active": true, "delta_db": 1.6766827461010523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.640717194960459e-07, "output_power": 6.272652210206751e-07, "reference_active": true, "delta_db": -0.8568291691815169}], "active_band_count": 3, "active_band_mae_db": 1.1989534114355436}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 476, "rms_floor": 4.3426949519701344e-05, "log_rms_mae_db": 1.4595189505131914, "rms_envelope_nrmse": 0.18822132898132396, "envelope_correlation": 0.9776125962424322, "rms_gain_error_db": -0.3670723871106263, "reference_mean_power": 4.5529138246552346e-05, "output_mean_power": 4.183908629103095e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 476, "frequency_resolution_hz": 0.42016806722689076, "power_floor": 6.179039674292901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3063171222465546e-05, "output_power": 2.8866098471390802e-05, "reference_active": true, "delta_db": -0.5895641736877176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2207009907392542e-05, "output_power": 1.0647032406011124e-05, "reference_active": true, "delta_db": -0.593807211363742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2513877801441335e-06, "output_power": 1.9496357810558754e-06, "reference_active": true, "delta_db": -0.6249681781892044}], "active_band_count": 3, "active_band_mae_db": 0.6027798544135546}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 38160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 3.683581645514392e-05, "log_rms_mae_db": 0.8074910255060063, "rms_envelope_nrmse": 0.11012812503358614, "envelope_correlation": 0.9927714231544535, "rms_gain_error_db": -0.36707447737870574, "reference_mean_power": 4.610980557969544e-05, "output_mean_power": 4.237267122300062e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 5.621286439624602e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.226124623783574e-05, "output_power": 2.826695524104083e-05, "reference_active": true, "delta_db": -0.5740210875653708}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.247687157348319e-06, "output_power": 6.270027041529698e-06, "reference_active": true, "delta_db": -0.6293002510671326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.30977962238882e-07, "output_power": 1.6238763368865243e-07, "reference_active": true, "delta_db": -1.5301759222217681}], "active_band_count": 3, "active_band_mae_db": 0.9111657536180905}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8587149880995496, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "07596d54ce76edb00417b68e9d125f454d77ad1ffa4f30ba1611b0d990ae614d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9848066250182821, "standard_deviation": null, "minimum": 0.9848066250182821, "maximum": 0.9848066250182821, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5725685760810553, "standard_deviation": null, "minimum": 1.5725685760810553, "maximum": 1.5725685760810553, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0408504765132909, "standard_deviation": null, "minimum": 1.0408504765132909, "maximum": 1.0408504765132909, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20968466110026188, "standard_deviation": null, "minimum": 0.20968466110026188, "maximum": 0.20968466110026188, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9925524903486735, "standard_deviation": null, "minimum": 0.9925524903486735, "maximum": 0.9925524903486735, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9670075429159195, "standard_deviation": null, "minimum": 0.9670075429159195, "maximum": 0.9670075429159195, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1989534114355436, "standard_deviation": null, "minimum": 1.1989534114355436, "maximum": 1.1989534114355436, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1727888765450336, "standard_deviation": null, "minimum": 0.1727888765450336, "maximum": 0.1727888765450336, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9776125962424322, "standard_deviation": null, "minimum": 0.9776125962424322, "maximum": 0.9776125962424322, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4595189505131914, "standard_deviation": null, "minimum": 1.4595189505131914, "maximum": 1.4595189505131914, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6027798544135546, "standard_deviation": null, "minimum": 0.6027798544135546, "maximum": 0.6027798544135546, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18822132898132396, "standard_deviation": null, "minimum": 0.18822132898132396, "maximum": 0.18822132898132396, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9927714231544535, "standard_deviation": null, "minimum": 0.9927714231544535, "maximum": 0.9927714231544535, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8074910255060063, "standard_deviation": null, "minimum": 0.8074910255060063, "maximum": 0.8074910255060063, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9111657536180905, "standard_deviation": null, "minimum": 0.9111657536180905, "maximum": 0.9111657536180905, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11012812503358614, "standard_deviation": null, "minimum": 0.11012812503358614, "maximum": 0.11012812503358614, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9612655016751539, "standard_deviation": null, "minimum": 0.9612655016751539, "maximum": 0.9612655016751539, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 3.6137347650650216, "standard_deviation": null, "minimum": 3.6137347650650216, "maximum": 3.6137347650650216, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.586867179810267, "standard_deviation": null, "minimum": 1.586867179810267, "maximum": 1.586867179810267, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30162641333160645, "standard_deviation": null, "minimum": 0.30162641333160645, "maximum": 0.30162641333160645, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9716082386587735, "standard_deviation": null, "minimum": 0.9716082386587735, "maximum": 0.9716082386587735, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 3.4949711711427196, "standard_deviation": null, "minimum": 3.4949711711427196, "maximum": 3.4949711711427196, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7244627985927202, "standard_deviation": null, "minimum": 0.7244627985927202, "maximum": 0.7244627985927202, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.26555420218446396, "standard_deviation": null, "minimum": 0.26555420218446396, "maximum": 0.26555420218446396, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.501409585363991, "standard_deviation": null, "minimum": 1.501409585363991, "maximum": 1.501409585363991, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.970351602277762, "standard_deviation": null, "minimum": 5.970351602277762, "maximum": 5.970351602277762, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5510417530752456, "standard_deviation": null, "minimum": 0.5510417530752456, "maximum": 0.5510417530752456, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": -1.3996589159330592, "standard_deviation": null, "minimum": -1.3996589159330592, "maximum": -1.3996589159330592, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.7560042721881168, "standard_deviation": null, "minimum": 0.7560042721881168, "maximum": 0.7560042721881168, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 4.469489841766993, "standard_deviation": null, "minimum": 4.469489841766993, "maximum": 4.469489841766993, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 4.988444972557986, "standard_deviation": null, "minimum": 4.988444972557986, "maximum": 4.988444972557986, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.5138648160620384, "standard_deviation": null, "minimum": 0.5138648160620384, "maximum": 0.5138648160620384, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.7690815480663353, "standard_deviation": null, "minimum": 0.7690815480663353, "maximum": 0.7690815480663353, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 4.327279080471564, "standard_deviation": null, "minimum": 4.327279080471564, "maximum": 4.327279080471564, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 6.129924922803636, "standard_deviation": null, "minimum": 6.129924922803636, "maximum": 6.129924922803636, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.47763659412950993, "standard_deviation": null, "minimum": 0.47763659412950993, "maximum": 0.47763659412950993, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2094"}, {"case": "librispeech_2300-131720-0022", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.6796398162841797, "predicted_mos": 3.6796398162841797, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "39529b002c0d624969f21ef1aa978d56bad2411dc282318dec663ecd1ff855b9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.976001400034875}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.2252338066668393, "全帯域_10ms_rms_envelope_nrmse": 0.17667634135147045, "全帯域_10ms_envelope_correlation": 0.9771195632419689, "全帯域_10ms_modulation_mae_db": 0.8683506935140025, "全帯域_40ms_log_rms_mae_db": 1.8410991846555957, "全帯域_40ms_rms_envelope_nrmse": 0.14056502018721112, "全帯域_40ms_envelope_correlation": 0.9889530936220693, "全帯域_40ms_modulation_mae_db": 1.1721344645583647, "300–1000 Hz_10ms_log_rms_mae_db": 1.905823381093285, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2700954144395827, "300–1000 Hz_10ms_envelope_correlation": 0.9757530559226999, "300–1000 Hz_10ms_modulation_mae_db": 1.3287995687470222, "300–1000 Hz_40ms_log_rms_mae_db": 1.6618563572167269, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.24016503689696306, "300–1000 Hz_40ms_envelope_correlation": 0.9852864392899878, "300–1000 Hz_40ms_modulation_mae_db": 1.170742840486991, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6611367650119837, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19643914857446362, "1000–3000 Hz_10ms_envelope_correlation": 0.9789890137554388, "1000–3000 Hz_10ms_modulation_mae_db": 0.7933515895218504, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0685699885962752, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13422290912156992, "1000–3000 Hz_40ms_envelope_correlation": 0.99298979034024, "1000–3000 Hz_40ms_modulation_mae_db": 1.0510362110957419, "3000–6000 Hz_10ms_log_rms_mae_db": 1.821117820813664, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.256703428350207, "3000–6000 Hz_10ms_envelope_correlation": 0.9662873780041957, "3000–6000 Hz_10ms_modulation_mae_db": 1.4073360003152786, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0523691136990858, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18267286002759772, "3000–6000 Hz_40ms_envelope_correlation": 0.9838164370879653, "3000–6000 Hz_40ms_modulation_mae_db": 1.191914906691311, "mr_stft_db": 5.8480022240065574, "band_mae_db": 0.7690017176639031, "rms_gain_error_db": 0.8198882400546248, "output_peak": 0.85505728031462}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.7825083607724, "spectral_convergence": 0.6464434184127078, "spectral_convergence_status": "ok", "amplitude_floor": 2.1903561449018732e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.144500840302881, "spectral_convergence": 0.2692825252967095, "spectral_convergence_status": "ok", "amplitude_floor": 2.0456135368211886e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.616997470944391, "spectral_convergence": 0.2743087699412636, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394418151882858e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0025307845601282375, "output_power": 0.002736209644497641, "delta_db": 0.3389419310851128, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0026710985360254404, "output_power": 0.003522047171456027, "delta_db": 1.201052589495176, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004653266401104453, "output_power": 0.0005967569725439074, "delta_db": 1.0803958527724065, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012444324831517497, "output_power": 0.00014960835762960165, "delta_db": 0.7998451648741782, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00016257063671523992, "output_power": 0.00018963177387450964, "delta_db": 0.6686900092256126, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.589634923518972e-05, "output_power": 4.0509674878198927e-05, "delta_db": 0.525084758530933, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7690017176639031, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8198882400546248, "pitch": null, "output_rms_dbfs": -21.40575655438733, "output_peak": 0.85505728031462, "reference_rms_dbfs": -22.225644794441955, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8480022240065574}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031508938028130336, "log_rms_mae_db": 2.2252338066668393, "rms_envelope_nrmse": 0.17667634135147045, "envelope_correlation": 0.9771195632419689, "rms_gain_error_db": 0.8201605010164339, "reference_mean_power": 0.0059947180836205155, "output_mean_power": 0.007240771029199161}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.905621652404067e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011843610245185533, "output_power": 0.001575718108583438, "reference_active": true, "delta_db": 1.2399441907873117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002763334576078544, "output_power": 0.00035976211396196163, "reference_active": true, "delta_db": 1.1458195533703421}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0009701982183084e-05, "output_power": 5.2599679769413e-05, "reference_active": true, "delta_db": 0.219288336384354}], "active_band_count": 3, "active_band_mae_db": 0.8683506935140025}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002456604014165908, "log_rms_mae_db": 1.8410991846555957, "rms_envelope_nrmse": 0.14056502018721112, "envelope_correlation": 0.9889530936220693, "rms_gain_error_db": 0.8199494293530094, "reference_mean_power": 0.006022035579915752, "output_mean_power": 0.0072734131944706725}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5974133279358877e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001059507007170014, "output_power": 0.001450797115137057, "reference_active": true, "delta_db": 1.3650285001655469}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014348331799351297, "output_power": 0.0001923340505366521, "reference_active": true, "delta_db": 1.2725476689609123}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4072600706208148e-06, "output_power": 2.783059948597118e-06, "reference_active": true, "delta_db": -0.8788272245486346}], "active_band_count": 3, "active_band_mae_db": 1.1721344645583647}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026889287764829997, "log_rms_mae_db": 1.905823381093285, "rms_envelope_nrmse": 0.2700954144395827, "envelope_correlation": 0.9757530559226999, "rms_gain_error_db": 1.5306665399662445, "reference_mean_power": 0.0016839373335741825, "output_mean_power": 0.002395480166951521}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1138357532326787e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006324020549599114, "output_power": 0.0009214038350967591, "reference_active": true, "delta_db": 1.6345674307352975}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002190350723697022, "output_power": 0.00031462624510491226, "reference_active": true, "delta_db": 1.5728128666068122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1194141233781235e-05, "output_power": 3.732285649943337e-05, "reference_active": true, "delta_db": 0.7790184088989569}], "active_band_count": 3, "active_band_mae_db": 1.3287995687470222}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001970017370857221, "log_rms_mae_db": 1.6618563572167269, "rms_envelope_nrmse": 0.24016503689696306, "envelope_correlation": 0.9852864392899878, "rms_gain_error_db": 1.5306653094676939, "reference_mean_power": 0.001692399182790771, "output_mean_power": 0.0024075168638690984}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.64481232375879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970882254118313, "output_power": 0.000886283601208731, "reference_active": true, "delta_db": 1.715342067108703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011301601543605377, "output_power": 0.00016475743985383242, "reference_active": true, "delta_db": 1.6370504341523262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.966963236644965e-06, "output_power": 2.859753255297479e-06, "reference_active": true, "delta_db": -0.15983602019994383}], "active_band_count": 3, "active_band_mae_db": 1.170742840486991}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.658803471872848e-05, "log_rms_mae_db": 1.6611367650119837, "rms_envelope_nrmse": 0.19643914857446362, "envelope_correlation": 0.9789890137554388, "rms_gain_error_db": 0.7855891659499423, "reference_mean_power": 0.00020257298157917204, "output_mean_power": 0.00024273949099232146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0818748871802829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.592086384233793e-05, "output_power": 8.318227461043868e-05, "reference_active": true, "delta_db": 1.0100790210534616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.925318155530915e-05, "output_power": 2.1559598310017793e-05, "reference_active": true, "delta_db": 0.49138158790726333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.819375654274508e-06, "output_power": 3.1198445800726027e-06, "reference_active": true, "delta_db": -0.8785941596048263}], "active_band_count": 3, "active_band_mae_db": 0.7933515895218504}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.479997453028037e-05, "log_rms_mae_db": 1.0685699885962752, "rms_envelope_nrmse": 0.13422290912156992, "envelope_correlation": 0.99298979034024, "rms_gain_error_db": 0.7856009687356673, "reference_mean_power": 0.00020359039158760131, "output_mean_power": 0.00024395929776324145}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.438814668171717e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.33475246376368e-05, "output_power": 7.941402912928496e-05, "reference_active": true, "delta_db": 0.9816758173292883}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.661338887813947e-06, "output_power": 1.0786441548441856e-05, "reference_active": true, "delta_db": 0.4784087838962648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1289641878321164e-07, "output_power": 2.1188401280336206e-07, "reference_active": true, "delta_db": -1.693024032061673}], "active_band_count": 3, "active_band_mae_db": 1.0510362110957419}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.558280881155462e-05, "log_rms_mae_db": 1.821117820813664, "rms_envelope_nrmse": 0.256703428350207, "envelope_correlation": 0.9662873780041957, "rms_gain_error_db": 0.6034969340327355, "reference_mean_power": 8.701244577979696e-05, "output_mean_power": 9.998412943675154e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.56825886194788e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4501869786005614e-05, "output_power": 2.789662213105969e-05, "reference_active": true, "delta_db": 0.5635239231174467}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.528335672589267e-06, "output_power": 7.102410376644453e-06, "reference_active": true, "delta_db": 1.9546714980444895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7809461115210634e-06, "output_power": 2.636525867303417e-06, "reference_active": true, "delta_db": 1.7038125797839005}], "active_band_count": 3, "active_band_mae_db": 1.4073360003152786}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.713713309096251e-05, "log_rms_mae_db": 1.0523691136990858, "rms_envelope_nrmse": 0.18267286002759772, "envelope_correlation": 0.9838164370879653, "rms_gain_error_db": 0.6035104006770715, "reference_mean_power": 8.74494743564098e-05, "output_mean_power": 0.00010048662116117019}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.348274544921528e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.135032255455729e-05, "output_power": 2.5365650959654793e-05, "reference_active": true, "delta_db": 0.7484157150483017}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4135743662495686e-06, "output_power": 3.718422412628115e-06, "reference_active": true, "delta_db": 1.8769803937525473}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6206121669861435e-07, "output_power": 2.017037424360957e-07, "reference_active": true, "delta_db": 0.9503486112730842}], "active_band_count": 3, "active_band_mae_db": 1.191914906691311}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.957404892697926, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d2723c7cd4850f281be81b06d98dd0e82ddcb90722e441fee7d5d5a896888852", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9789890137554388, "standard_deviation": null, "minimum": 0.9789890137554388, "maximum": 0.9789890137554388, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6611367650119837, "standard_deviation": null, "minimum": 1.6611367650119837, "maximum": 1.6611367650119837, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7933515895218504, "standard_deviation": null, "minimum": 0.7933515895218504, "maximum": 0.7933515895218504, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19643914857446362, "standard_deviation": null, "minimum": 0.19643914857446362, "maximum": 0.19643914857446362, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99298979034024, "standard_deviation": null, "minimum": 0.99298979034024, "maximum": 0.99298979034024, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0685699885962752, "standard_deviation": null, "minimum": 1.0685699885962752, "maximum": 1.0685699885962752, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0510362110957419, "standard_deviation": null, "minimum": 1.0510362110957419, "maximum": 1.0510362110957419, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13422290912156992, "standard_deviation": null, "minimum": 0.13422290912156992, "maximum": 0.13422290912156992, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9662873780041957, "standard_deviation": null, "minimum": 0.9662873780041957, "maximum": 0.9662873780041957, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.821117820813664, "standard_deviation": null, "minimum": 1.821117820813664, "maximum": 1.821117820813664, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.4073360003152786, "standard_deviation": null, "minimum": 1.4073360003152786, "maximum": 1.4073360003152786, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.256703428350207, "standard_deviation": null, "minimum": 0.256703428350207, "maximum": 0.256703428350207, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9838164370879653, "standard_deviation": null, "minimum": 0.9838164370879653, "maximum": 0.9838164370879653, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0523691136990858, "standard_deviation": null, "minimum": 1.0523691136990858, "maximum": 1.0523691136990858, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.191914906691311, "standard_deviation": null, "minimum": 1.191914906691311, "maximum": 1.191914906691311, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18267286002759772, "standard_deviation": null, "minimum": 0.18267286002759772, "maximum": 0.18267286002759772, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9757530559226999, "standard_deviation": null, "minimum": 0.9757530559226999, "maximum": 0.9757530559226999, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.905823381093285, "standard_deviation": null, "minimum": 1.905823381093285, "maximum": 1.905823381093285, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3287995687470222, "standard_deviation": null, "minimum": 1.3287995687470222, "maximum": 1.3287995687470222, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2700954144395827, "standard_deviation": null, "minimum": 0.2700954144395827, "maximum": 0.2700954144395827, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9852864392899878, "standard_deviation": null, "minimum": 0.9852864392899878, "maximum": 0.9852864392899878, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6618563572167269, "standard_deviation": null, "minimum": 1.6618563572167269, "maximum": 1.6618563572167269, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.170742840486991, "standard_deviation": null, "minimum": 1.170742840486991, "maximum": 1.170742840486991, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.24016503689696306, "standard_deviation": null, "minimum": 0.24016503689696306, "maximum": 0.24016503689696306, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7690017176639031, "standard_deviation": null, "minimum": 0.7690017176639031, "maximum": 0.7690017176639031, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.8480022240065574, "standard_deviation": null, "minimum": 5.8480022240065574, "maximum": 5.8480022240065574, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.85505728031462, "standard_deviation": null, "minimum": 0.85505728031462, "maximum": 0.85505728031462, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8198882400546248, "standard_deviation": null, "minimum": 0.8198882400546248, "maximum": 0.8198882400546248, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9771195632419689, "standard_deviation": null, "minimum": 0.9771195632419689, "maximum": 0.9771195632419689, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.2252338066668393, "standard_deviation": null, "minimum": 2.2252338066668393, "maximum": 2.2252338066668393, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8683506935140025, "standard_deviation": null, "minimum": 0.8683506935140025, "maximum": 0.8683506935140025, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17667634135147045, "standard_deviation": null, "minimum": 0.17667634135147045, "maximum": 0.17667634135147045, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9889530936220693, "standard_deviation": null, "minimum": 0.9889530936220693, "maximum": 0.9889530936220693, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.8410991846555957, "standard_deviation": null, "minimum": 1.8410991846555957, "maximum": 1.8410991846555957, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1721344645583647, "standard_deviation": null, "minimum": 1.1721344645583647, "maximum": 1.1721344645583647, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.14056502018721112, "standard_deviation": null, "minimum": 0.14056502018721112, "maximum": 0.14056502018721112, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2300"}, {"case": "librispeech_2300-131720-0028", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.4929301738739014, "predicted_mos": 3.4929301738739014, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e525bf723b59aa99dc719accdbc0bb7bacc5ece043f15f75144ee7c1dee4056e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.990819999948144}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.180625364239005, "全帯域_10ms_rms_envelope_nrmse": 0.2427272191856283, "全帯域_10ms_envelope_correlation": 0.964908785212153, "全帯域_10ms_modulation_mae_db": 1.663792879402396, "全帯域_40ms_log_rms_mae_db": 1.4790639109262764, "全帯域_40ms_rms_envelope_nrmse": 0.20626399962887804, "全帯域_40ms_envelope_correlation": 0.9796529617290423, "全帯域_40ms_modulation_mae_db": 1.3837825529827084, "300–1000 Hz_10ms_log_rms_mae_db": 2.6695938349599664, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.32666948945061236, "300–1000 Hz_10ms_envelope_correlation": 0.9719358124967044, "300–1000 Hz_10ms_modulation_mae_db": 1.5079521092735908, "300–1000 Hz_40ms_log_rms_mae_db": 2.034268269344837, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.29034763673328506, "300–1000 Hz_40ms_envelope_correlation": 0.9845520481878265, "300–1000 Hz_40ms_modulation_mae_db": 0.9952171290585342, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9883337184514618, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22233907503651684, "1000–3000 Hz_10ms_envelope_correlation": 0.9737648659492874, "1000–3000 Hz_10ms_modulation_mae_db": 1.0355891555938073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0724483978980641, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1492020387718465, "1000–3000 Hz_40ms_envelope_correlation": 0.9918276218008303, "1000–3000 Hz_40ms_modulation_mae_db": 1.1292748989043326, "3000–6000 Hz_10ms_log_rms_mae_db": 2.0231619792869076, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20378092797529448, "3000–6000 Hz_10ms_envelope_correlation": 0.9787862149788225, "3000–6000 Hz_10ms_modulation_mae_db": 1.2661236341399837, "3000–6000 Hz_40ms_log_rms_mae_db": 1.104048769023702, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13938711725326383, "3000–6000 Hz_40ms_envelope_correlation": 0.991072861797932, "3000–6000 Hz_40ms_modulation_mae_db": 1.0279647901057605, "mr_stft_db": 6.043663674224372, "band_mae_db": 0.8320738910779192, "rms_gain_error_db": 1.1030817575691738, "output_peak": 0.6943533334235241}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.07791467467061, "spectral_convergence": 0.5635348164424449, "spectral_convergence_status": "ok", "amplitude_floor": 1.711947899928738e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.423590707605254, "spectral_convergence": 0.38356039643071643, "spectral_convergence_status": "ok", "amplitude_floor": 1.4743066880059505e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.629485640397252, "spectral_convergence": 0.38411554644882134, "spectral_convergence_status": "ok", "amplitude_floor": 1.2829193968481007e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015568083457567275, "output_power": 0.001481524453615035, "delta_db": -0.21526326886947497, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001958578248681761, "output_power": 0.0031057045943246927, "delta_db": 2.0021921744997266, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00020973665094298736, "output_power": 0.0002554967263154094, "delta_db": 0.8571101093630212, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018883580517861778, "output_power": 0.0002340263280382, "delta_db": 0.9318037403313528, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010837478072549288, "output_power": 0.00012398295777033607, "delta_db": 0.5843376100564962, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.95174204959434e-05, "output_power": 8.722398565716854e-05, "delta_db": 0.40173644334744246, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8320738910779192, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1030817575691738, "pitch": null, "output_rms_dbfs": -22.767119170410098, "output_peak": 0.6943533334235241, "reference_rms_dbfs": -23.870200927979273, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.043663674224372}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023872023988588299, "log_rms_mae_db": 2.180625364239005, "rms_envelope_nrmse": 0.2427272191856283, "envelope_correlation": 0.964908785212153, "rms_gain_error_db": 1.1034570117734699, "reference_mean_power": 0.004104010412713017, "output_mean_power": 0.0052911997271728635}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.972155761177528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037154075128907306, "output_power": 0.00046816765410385183, "reference_active": true, "delta_db": 1.0039494986690476}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.499306084417706e-05, "output_power": 6.997967008477317e-05, "reference_active": true, "delta_db": 1.0466399805885533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6530449727349195e-05, "output_power": 3.253595697109306e-05, "reference_active": true, "delta_db": 2.9407891589495865}], "active_band_count": 3, "active_band_mae_db": 1.663792879402396}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00021497588341920653, "log_rms_mae_db": 1.4790639109262764, "rms_envelope_nrmse": 0.20626399962887804, "envelope_correlation": 0.9796529617290423, "rms_gain_error_db": 1.1034006136414836, "reference_mean_power": 0.00411739985369924, "output_mean_power": 0.005308393468213873}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.973189563381595e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033993681960372955, "output_power": 0.0004314734312411333, "reference_active": true, "delta_db": 1.0355585150092408}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4133967175911375e-05, "output_power": 3.2145056945282235e-05, "reference_active": true, "delta_db": 1.2448548173235237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.827774931478282e-07, "output_power": 1.35814045180472e-06, "reference_active": true, "delta_db": 1.8709343266153609}], "active_band_count": 3, "active_band_mae_db": 1.3837825529827084}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001937632671887944, "log_rms_mae_db": 2.6695938349599664, "rms_envelope_nrmse": 0.32666948945061236, "envelope_correlation": 0.9719358124967044, "rms_gain_error_db": 1.9016631412494847, "reference_mean_power": 0.0014622039835396937, "output_mean_power": 0.0022655532629586223}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.890315707731239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020584065582039213, "output_power": 0.0002599739761635552, "reference_active": true, "delta_db": 1.0139871963933755}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2632804837837524e-05, "output_power": 5.778564726841086e-05, "reference_active": true, "delta_db": 1.3207607617640615}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.198246611681044e-06, "output_power": 1.3571623878045123e-05, "reference_active": true, "delta_db": 2.1891083696633356}], "active_band_count": 3, "active_band_mae_db": 1.5079521092735908}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001823230477151681, "log_rms_mae_db": 2.034268269344837, "rms_envelope_nrmse": 0.29034763673328506, "envelope_correlation": 0.9845520481878265, "rms_gain_error_db": 1.8925237032003184, "reference_mean_power": 0.0014670926652510829, "output_mean_power": 0.002268349218391843}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5556250379109994e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020148633268320183, "output_power": 0.0002546774254515642, "reference_active": true, "delta_db": 1.0174485870035037}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9312508746169674e-05, "output_power": 2.7381394342437838e-05, "reference_active": true, "delta_db": 1.5161686648367385}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.933225179066708e-07, "output_power": 7.693761494075335e-07, "reference_active": true, "delta_db": 0.4520341353353601}], "active_band_count": 3, "active_band_mae_db": 0.9952171290585342}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.475794683802334e-05, "log_rms_mae_db": 1.9883337184514618, "rms_envelope_nrmse": 0.22233907503651684, "envelope_correlation": 0.9737648659492874, "rms_gain_error_db": 0.870462011216308, "reference_mean_power": 0.000205358190626383, "output_mean_power": 0.0002509332608851747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.94981889493768e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.427037048631366e-05, "output_power": 6.376449267322784e-05, "reference_active": true, "delta_db": 0.7001612229955024}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.925199890017768e-06, "output_power": 9.835435787670107e-06, "reference_active": true, "delta_db": 0.42175655952664864}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.574401681634841e-06, "output_power": 1.6300140281181866e-06, "reference_active": true, "delta_db": -1.984849684259271}], "active_band_count": 3, "active_band_mae_db": 1.0355891555938073}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.971385946015731e-05, "log_rms_mae_db": 1.0724483978980641, "rms_envelope_nrmse": 0.1492020387718465, "envelope_correlation": 0.9918276218008303, "rms_gain_error_db": 0.8690052624258797, "reference_mean_power": 0.00020637707638038015, "output_mean_power": 0.0002520936939225247}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.280870922430589e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.267995669092476e-05, "output_power": 6.148427640081625e-05, "reference_active": true, "delta_db": 0.6711865700682506}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.180545251002319e-06, "output_power": 4.643653561816749e-06, "reference_active": true, "delta_db": 0.4562688320505181}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0193067246169046e-07, "output_power": 1.1999561118058115e-07, "reference_active": true, "delta_db": -2.2603692945942293}], "active_band_count": 3, "active_band_mae_db": 1.1292748989043326}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.517166954195252e-05, "log_rms_mae_db": 2.0231619792869076, "rms_envelope_nrmse": 0.20378092797529448, "envelope_correlation": 0.9787862149788225, "rms_gain_error_db": 0.5412103096369295, "reference_mean_power": 0.00014718641611289376, "output_mean_power": 0.0001667204069213791}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4339203545791116e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.0687614509069e-05, "output_power": 7.614708113449631e-05, "reference_active": true, "delta_db": 0.3230993484292752}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.343067234178686e-06, "output_power": 1.481381891570491e-05, "reference_active": true, "delta_db": 2.001775574170429}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3595191848035655e-06, "output_power": 3.3126313746580535e-06, "reference_active": true, "delta_db": 1.4734959798202476}], "active_band_count": 3, "active_band_mae_db": 1.2661236341399837}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.54521457203503e-05, "log_rms_mae_db": 1.104048769023702, "rms_envelope_nrmse": 0.13938711725326383, "envelope_correlation": 0.991072861797932, "rms_gain_error_db": 0.5411784195065619, "reference_mean_power": 0.0001479201087162099, "output_mean_power": 0.00016755024194273216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3826641859038576e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.726675521598142e-05, "output_power": 7.398738642451203e-05, "reference_active": true, "delta_db": 0.4135720742720658}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.27086384707959e-06, "output_power": 7.414180142877593e-06, "reference_active": true, "delta_db": 2.395474073414813}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6847652723937028e-07, "output_power": 1.7948337367957276e-07, "reference_active": true, "delta_db": 0.27484822263040287}], "active_band_count": 3, "active_band_mae_db": 1.0279647901057605}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.874295398894212, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cb80b1f66e67d02b77455d56220e5f32b7bbc9702347a337781922ce0247514d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9737648659492874, "standard_deviation": null, "minimum": 0.9737648659492874, "maximum": 0.9737648659492874, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9883337184514618, "standard_deviation": null, "minimum": 1.9883337184514618, "maximum": 1.9883337184514618, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0355891555938073, "standard_deviation": null, "minimum": 1.0355891555938073, "maximum": 1.0355891555938073, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22233907503651684, "standard_deviation": null, "minimum": 0.22233907503651684, "maximum": 0.22233907503651684, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9918276218008303, "standard_deviation": null, "minimum": 0.9918276218008303, "maximum": 0.9918276218008303, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0724483978980641, "standard_deviation": null, "minimum": 1.0724483978980641, "maximum": 1.0724483978980641, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1292748989043326, "standard_deviation": null, "minimum": 1.1292748989043326, "maximum": 1.1292748989043326, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1492020387718465, "standard_deviation": null, "minimum": 0.1492020387718465, "maximum": 0.1492020387718465, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9787862149788225, "standard_deviation": null, "minimum": 0.9787862149788225, "maximum": 0.9787862149788225, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.0231619792869076, "standard_deviation": null, "minimum": 2.0231619792869076, "maximum": 2.0231619792869076, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2661236341399837, "standard_deviation": null, "minimum": 1.2661236341399837, "maximum": 1.2661236341399837, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20378092797529448, "standard_deviation": null, "minimum": 0.20378092797529448, "maximum": 0.20378092797529448, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991072861797932, "standard_deviation": null, "minimum": 0.991072861797932, "maximum": 0.991072861797932, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.104048769023702, "standard_deviation": null, "minimum": 1.104048769023702, "maximum": 1.104048769023702, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0279647901057605, "standard_deviation": null, "minimum": 1.0279647901057605, "maximum": 1.0279647901057605, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13938711725326383, "standard_deviation": null, "minimum": 0.13938711725326383, "maximum": 0.13938711725326383, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9719358124967044, "standard_deviation": null, "minimum": 0.9719358124967044, "maximum": 0.9719358124967044, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.6695938349599664, "standard_deviation": null, "minimum": 2.6695938349599664, "maximum": 2.6695938349599664, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.5079521092735908, "standard_deviation": null, "minimum": 1.5079521092735908, "maximum": 1.5079521092735908, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.32666948945061236, "standard_deviation": null, "minimum": 0.32666948945061236, "maximum": 0.32666948945061236, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9845520481878265, "standard_deviation": null, "minimum": 0.9845520481878265, "maximum": 0.9845520481878265, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.034268269344837, "standard_deviation": null, "minimum": 2.034268269344837, "maximum": 2.034268269344837, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9952171290585342, "standard_deviation": null, "minimum": 0.9952171290585342, "maximum": 0.9952171290585342, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.29034763673328506, "standard_deviation": null, "minimum": 0.29034763673328506, "maximum": 0.29034763673328506, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8320738910779192, "standard_deviation": null, "minimum": 0.8320738910779192, "maximum": 0.8320738910779192, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.043663674224372, "standard_deviation": null, "minimum": 6.043663674224372, "maximum": 6.043663674224372, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6943533334235241, "standard_deviation": null, "minimum": 0.6943533334235241, "maximum": 0.6943533334235241, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1030817575691738, "standard_deviation": null, "minimum": 1.1030817575691738, "maximum": 1.1030817575691738, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.964908785212153, "standard_deviation": null, "minimum": 0.964908785212153, "maximum": 0.964908785212153, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.180625364239005, "standard_deviation": null, "minimum": 2.180625364239005, "maximum": 2.180625364239005, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.663792879402396, "standard_deviation": null, "minimum": 1.663792879402396, "maximum": 1.663792879402396, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2427272191856283, "standard_deviation": null, "minimum": 0.2427272191856283, "maximum": 0.2427272191856283, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9796529617290423, "standard_deviation": null, "minimum": 0.9796529617290423, "maximum": 0.9796529617290423, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.4790639109262764, "standard_deviation": null, "minimum": 1.4790639109262764, "maximum": 1.4790639109262764, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.3837825529827084, "standard_deviation": null, "minimum": 1.3837825529827084, "maximum": 1.3837825529827084, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.20626399962887804, "standard_deviation": null, "minimum": 0.20626399962887804, "maximum": 0.20626399962887804, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2300"}, {"case": "librispeech_2300-131720-0041", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.4924471378326416, "predicted_mos": 3.4924471378326416, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "3c508594f7f16461da582a661dda6f50c031807edeb9a820bc767095ff8e86a3", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.099285599775612}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4526757930546577, "全帯域_10ms_rms_envelope_nrmse": 0.19909113377193297, "全帯域_10ms_envelope_correlation": 0.9748336083126272, "全帯域_10ms_modulation_mae_db": 1.1161537718058518, "全帯域_40ms_log_rms_mae_db": 1.0564195784534058, "全帯域_40ms_rms_envelope_nrmse": 0.1589780833155297, "全帯域_40ms_envelope_correlation": 0.9881563860058039, "全帯域_40ms_modulation_mae_db": 1.1291606965928735, "300–1000 Hz_10ms_log_rms_mae_db": 1.801194959544778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.30008669396204707, "300–1000 Hz_10ms_envelope_correlation": 0.9721712750016956, "300–1000 Hz_10ms_modulation_mae_db": 1.0253443838224645, "300–1000 Hz_40ms_log_rms_mae_db": 1.545173765576017, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2708305225050245, "300–1000 Hz_40ms_envelope_correlation": 0.9819783919647881, "300–1000 Hz_40ms_modulation_mae_db": 1.1696142442719217, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4326803581523786, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2025292616529522, "1000–3000 Hz_10ms_envelope_correlation": 0.9795818276027438, "1000–3000 Hz_10ms_modulation_mae_db": 0.6458511166325911, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9286440929125709, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11684380874125552, "1000–3000 Hz_40ms_envelope_correlation": 0.9963754419496409, "1000–3000 Hz_40ms_modulation_mae_db": 0.6569505861017975, "3000–6000 Hz_10ms_log_rms_mae_db": 1.562124864893028, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23869870565109172, "3000–6000 Hz_10ms_envelope_correlation": 0.9639511742198905, "3000–6000 Hz_10ms_modulation_mae_db": 1.0764813434960905, "3000–6000 Hz_40ms_log_rms_mae_db": 1.099642771519159, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1451900712977254, "3000–6000 Hz_40ms_envelope_correlation": 0.9864395869412581, "3000–6000 Hz_40ms_modulation_mae_db": 1.2175750488432, "mr_stft_db": 4.981527181003036, "band_mae_db": 0.7063741797591415, "rms_gain_error_db": 0.9529113930665977, "output_peak": 0.6646857006442194}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.170868816268207, "spectral_convergence": 0.6703320055689596, "spectral_convergence_status": "ok", "amplitude_floor": 1.419137961520426e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.22307692265299, "spectral_convergence": 0.31003915955630124, "spectral_convergence_status": "ok", "amplitude_floor": 1.2001614819603722e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.5506358040879125, "spectral_convergence": 0.29631885377157385, "spectral_convergence_status": "ok", "amplitude_floor": 7.977266337453151e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0015734976003854988, "output_power": 0.0016176072493422663, "delta_db": 0.12006999609309402, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002272380907829781, "output_power": 0.0031674630401085984, "delta_db": 1.4423042422221894, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00035321559220182435, "output_power": 0.0004784058781030509, "delta_db": 1.3175664048270126, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00024119869501752642, "output_power": 0.0002925131796341966, "delta_db": 0.8377048490361969, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012307274117421543, "output_power": 0.00013563284543783343, "delta_db": 0.42202999243769224, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.88440946982043e-05, "output_power": 3.9735800231631574e-05, "delta_db": 0.09856959393866302, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7063741797591415, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.9529113930665977, "pitch": null, "output_rms_dbfs": -22.41742463734385, "output_peak": 0.6646857006442194, "reference_rms_dbfs": -23.37033603041045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 4.981527181003036}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00019859470734653442, "log_rms_mae_db": 1.4526757930546577, "rms_envelope_nrmse": 0.19909113377193297, "envelope_correlation": 0.9748336083126272, "rms_gain_error_db": 0.952911393066484, "reference_mean_power": 0.004608354103444978, "output_mean_power": 0.005739010006199027}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.3747785986813366e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009662478923264104, "output_power": 0.0011693686785214507, "reference_active": true, "delta_db": 0.8286289750001401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.594410683545364e-05, "output_power": 0.00011984852171756453, "reference_active": true, "delta_db": 0.9661437729063503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7446486319031935e-05, "output_power": 2.4950360843754365e-05, "reference_active": true, "delta_db": 1.553688567511065}], "active_band_count": 3, "active_band_mae_db": 1.1161537718058518}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00016219647203140816, "log_rms_mae_db": 1.0564195784534058, "rms_envelope_nrmse": 0.1589780833155297, "envelope_correlation": 0.9881563860058039, "rms_gain_error_db": 0.9529113930664846, "reference_mean_power": 0.004645568268479526, "output_mean_power": 0.005785354636127955}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1771130170213754e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008502808342452142, "output_power": 0.0010342540751880982, "reference_active": true, "delta_db": 0.8506485081422825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.133174041000048e-05, "output_power": 5.394556587875723e-05, "reference_active": true, "delta_db": 1.1567206004786958}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2770439378674992e-06, "output_power": 1.7547576237680554e-06, "reference_active": true, "delta_db": 1.380112981157642}], "active_band_count": 3, "active_band_mae_db": 1.1291606965928735}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015290889340024391, "log_rms_mae_db": 1.801194959544778, "rms_envelope_nrmse": 0.30008669396204707, "envelope_correlation": 0.9721712750016956, "rms_gain_error_db": 1.7131582501717681, "reference_mean_power": 0.0016143224036409894, "output_mean_power": 0.002395003205325373}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 8.858385684123232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006875363814202012, "output_power": 0.000952854359209345, "reference_active": true, "delta_db": 1.4173084108968073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.173936122300265e-05, "output_power": 8.895950592316819e-05, "reference_active": true, "delta_db": 1.5863023038703643}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0160658568889886e-05, "output_power": 1.0331517164924913e-05, "reference_active": true, "delta_db": 0.07242243670022176}], "active_band_count": 3, "active_band_mae_db": 1.0253443838224645}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012027897252950328, "log_rms_mae_db": 1.545173765576017, "rms_envelope_nrmse": 0.2708305225050245, "envelope_correlation": 0.9819783919647881, "rms_gain_error_db": 1.7131582501717688, "reference_mean_power": 0.0016273586545452231, "output_mean_power": 0.0024143437426496694}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.855745033342305e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006306037982733972, "output_power": 0.0008756483894758985, "reference_active": true, "delta_db": 1.4257317096397029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.290965975606541e-05, "output_power": 3.436122349258579e-05, "reference_active": true, "delta_db": 1.7604997997184346}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0400487875049354e-06, "output_power": 9.65589622673031e-07, "reference_active": true, "delta_db": -0.32261122345762794}], "active_band_count": 3, "active_band_mae_db": 1.1696142442719217}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 9.957022642296078e-05, "log_rms_mae_db": 1.4326803581523786, "rms_envelope_nrmse": 0.2025292616529522, "envelope_correlation": 0.9795818276027438, "rms_gain_error_db": 0.7477762651812756, "reference_mean_power": 0.000306890935740228, "output_mean_power": 0.00036455384899411545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.0486708619890815e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.911544017889007e-05, "output_power": 8.34387526464534e-05, "reference_active": true, "delta_db": 0.817927247772593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0167211042917879e-05, "output_power": 1.2376245768549902e-05, "reference_active": true, "delta_db": 0.8538708664236585}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3309723892934117e-06, "output_power": 2.4780647663070735e-06, "reference_active": true, "delta_db": 0.2657552357015218}], "active_band_count": 3, "active_band_mae_db": 0.6458511166325911}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 8.985216395929287e-05, "log_rms_mae_db": 0.9286440929125709, "rms_envelope_nrmse": 0.11684380874125552, "envelope_correlation": 0.9963754419496409, "rms_gain_error_db": 0.7477762651812747, "reference_mean_power": 0.00030936919363315046, "output_mean_power": 0.00036749775625382556}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 9.09439257266203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.275291186820264e-05, "output_power": 7.569206927127821e-05, "reference_active": true, "delta_db": 0.8141649534492246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.365294966430701e-06, "output_power": 5.3593516807513105e-06, "reference_active": true, "delta_db": 0.890986617357628}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9190263566647066e-07, "output_power": 1.805140128735927e-07, "reference_active": true, "delta_db": -0.26570018749853974}], "active_band_count": 3, "active_band_mae_db": 0.6569505861017975}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.809631559896483e-05, "log_rms_mae_db": 1.562124864893028, "rms_envelope_nrmse": 0.23869870565109172, "envelope_correlation": 0.9639511742198905, "rms_gain_error_db": 0.26671255989715636, "reference_mean_power": 6.929937891252573e-05, "output_mean_power": 7.368864968209804e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 7.858105660801035e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.988342411251772e-05, "output_power": 4.258771762692418e-05, "reference_active": true, "delta_db": 0.28491929129725113}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1366606898498195e-05, "output_power": 8.65816643861722e-06, "reference_active": true, "delta_db": -1.1820491052621975}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.736645328964402e-06, "output_power": 2.6059057006753904e-06, "reference_active": true, "delta_db": 1.762475633928823}], "active_band_count": 3, "active_band_mae_db": 1.0764813434960905}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.651257368418169e-05, "log_rms_mae_db": 1.099642771519159, "rms_envelope_nrmse": 0.1451900712977254, "envelope_correlation": 0.9864395869412581, "rms_gain_error_db": 0.26671255989714543, "reference_mean_power": 6.985899704640885e-05, "output_mean_power": 7.428371280200712e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 7.431140169416232e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0602300645347785e-05, "output_power": 4.3560606742285e-05, "reference_active": true, "delta_db": 0.3054327760750057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067895142462218e-06, "output_power": 4.5452481516061545e-06, "reference_active": true, "delta_db": -1.2548046838843865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.231130378903163e-07, "output_power": 1.9932088948860594e-07, "reference_active": true, "delta_db": 2.0924876865702076}], "active_band_count": 3, "active_band_mae_db": 1.2175750488432}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9532691306412333, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "475b4efd3d8a3daf1b7fc9a74628b8f8cdf8c3cf16ecc95a42c9617309eff17a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9795818276027438, "standard_deviation": null, "minimum": 0.9795818276027438, "maximum": 0.9795818276027438, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4326803581523786, "standard_deviation": null, "minimum": 1.4326803581523786, "maximum": 1.4326803581523786, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6458511166325911, "standard_deviation": null, "minimum": 0.6458511166325911, "maximum": 0.6458511166325911, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2025292616529522, "standard_deviation": null, "minimum": 0.2025292616529522, "maximum": 0.2025292616529522, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9963754419496409, "standard_deviation": null, "minimum": 0.9963754419496409, "maximum": 0.9963754419496409, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9286440929125709, "standard_deviation": null, "minimum": 0.9286440929125709, "maximum": 0.9286440929125709, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6569505861017975, "standard_deviation": null, "minimum": 0.6569505861017975, "maximum": 0.6569505861017975, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11684380874125552, "standard_deviation": null, "minimum": 0.11684380874125552, "maximum": 0.11684380874125552, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9639511742198905, "standard_deviation": null, "minimum": 0.9639511742198905, "maximum": 0.9639511742198905, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.562124864893028, "standard_deviation": null, "minimum": 1.562124864893028, "maximum": 1.562124864893028, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0764813434960905, "standard_deviation": null, "minimum": 1.0764813434960905, "maximum": 1.0764813434960905, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23869870565109172, "standard_deviation": null, "minimum": 0.23869870565109172, "maximum": 0.23869870565109172, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9864395869412581, "standard_deviation": null, "minimum": 0.9864395869412581, "maximum": 0.9864395869412581, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.099642771519159, "standard_deviation": null, "minimum": 1.099642771519159, "maximum": 1.099642771519159, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2175750488432, "standard_deviation": null, "minimum": 1.2175750488432, "maximum": 1.2175750488432, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1451900712977254, "standard_deviation": null, "minimum": 0.1451900712977254, "maximum": 0.1451900712977254, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9721712750016956, "standard_deviation": null, "minimum": 0.9721712750016956, "maximum": 0.9721712750016956, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.801194959544778, "standard_deviation": null, "minimum": 1.801194959544778, "maximum": 1.801194959544778, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0253443838224645, "standard_deviation": null, "minimum": 1.0253443838224645, "maximum": 1.0253443838224645, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30008669396204707, "standard_deviation": null, "minimum": 0.30008669396204707, "maximum": 0.30008669396204707, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9819783919647881, "standard_deviation": null, "minimum": 0.9819783919647881, "maximum": 0.9819783919647881, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.545173765576017, "standard_deviation": null, "minimum": 1.545173765576017, "maximum": 1.545173765576017, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1696142442719217, "standard_deviation": null, "minimum": 1.1696142442719217, "maximum": 1.1696142442719217, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2708305225050245, "standard_deviation": null, "minimum": 0.2708305225050245, "maximum": 0.2708305225050245, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7063741797591415, "standard_deviation": null, "minimum": 0.7063741797591415, "maximum": 0.7063741797591415, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 4.981527181003036, "standard_deviation": null, "minimum": 4.981527181003036, "maximum": 4.981527181003036, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6646857006442194, "standard_deviation": null, "minimum": 0.6646857006442194, "maximum": 0.6646857006442194, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.9529113930665977, "standard_deviation": null, "minimum": 0.9529113930665977, "maximum": 0.9529113930665977, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9748336083126272, "standard_deviation": null, "minimum": 0.9748336083126272, "maximum": 0.9748336083126272, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4526757930546577, "standard_deviation": null, "minimum": 1.4526757930546577, "maximum": 1.4526757930546577, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1161537718058518, "standard_deviation": null, "minimum": 1.1161537718058518, "maximum": 1.1161537718058518, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19909113377193297, "standard_deviation": null, "minimum": 0.19909113377193297, "maximum": 0.19909113377193297, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9881563860058039, "standard_deviation": null, "minimum": 0.9881563860058039, "maximum": 0.9881563860058039, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0564195784534058, "standard_deviation": null, "minimum": 1.0564195784534058, "maximum": 1.0564195784534058, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1291606965928735, "standard_deviation": null, "minimum": 1.1291606965928735, "maximum": 1.1291606965928735, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1589780833155297, "standard_deviation": null, "minimum": 0.1589780833155297, "maximum": 0.1589780833155297, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2300"}, {"case": "librispeech_237-126133-0003", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.763366222381592, "predicted_mos": 2.763366222381592, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "4fd449ee8198e1c0d160e1b0f9ecadd3eec3a9194b1758f931f4cfa5785f99b7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.447810499928892}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8161910473673577, "全帯域_10ms_rms_envelope_nrmse": 0.186975560814521, "全帯域_10ms_envelope_correlation": 0.9891146532799888, "全帯域_10ms_modulation_mae_db": 0.9855877307003951, "全帯域_40ms_log_rms_mae_db": 1.4035686497468842, "全帯域_40ms_rms_envelope_nrmse": 0.1654830640381344, "全帯域_40ms_envelope_correlation": 0.995520512645089, "全帯域_40ms_modulation_mae_db": 0.8479805225854232, "300–1000 Hz_10ms_log_rms_mae_db": 1.9362418551421121, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21292450809936578, "300–1000 Hz_10ms_envelope_correlation": 0.9888345401834542, "300–1000 Hz_10ms_modulation_mae_db": 1.2833930234271465, "300–1000 Hz_40ms_log_rms_mae_db": 1.5397805315356226, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1943117495346633, "300–1000 Hz_40ms_envelope_correlation": 0.9939783838800023, "300–1000 Hz_40ms_modulation_mae_db": 0.9033833456269741, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6730754393693428, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23119457483472833, "1000–3000 Hz_10ms_envelope_correlation": 0.9859360531484173, "1000–3000 Hz_10ms_modulation_mae_db": 2.006613862443541, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2092286973785829, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.20376363536434247, "1000–3000 Hz_40ms_envelope_correlation": 0.9937459014729166, "1000–3000 Hz_40ms_modulation_mae_db": 1.9356060377618283, "3000–6000 Hz_10ms_log_rms_mae_db": 1.668956850578569, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22132625586543986, "3000–6000 Hz_10ms_envelope_correlation": 0.9742557762309456, "3000–6000 Hz_10ms_modulation_mae_db": 0.850234302702168, "3000–6000 Hz_40ms_log_rms_mae_db": 1.068446242455174, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1647977301943024, "3000–6000 Hz_40ms_envelope_correlation": 0.989927029705434, "3000–6000 Hz_40ms_modulation_mae_db": 1.07195952649323, "mr_stft_db": 5.754207439258583, "band_mae_db": 0.9563093402189873, "rms_gain_error_db": 1.2089647532583143, "output_peak": 0.2928557232132792}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.425420077685415, "spectral_convergence": 0.5307917440694261, "spectral_convergence_status": "ok", "amplitude_floor": 6.60007588392934e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.182108739387866, "spectral_convergence": 0.2658209754585509, "spectral_convergence_status": "ok", "amplitude_floor": 5.570424644238683e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.65509350070247, "spectral_convergence": 0.2920262192658276, "spectral_convergence_status": "ok", "amplitude_floor": 3.8555660806566813e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.162305217059986e-05, "output_power": 3.3944665100403076e-05, "delta_db": 0.3076774372654538, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000304634409424632, "output_power": 0.00037799885731121545, "delta_db": 0.9371153025646521, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004181285613649292, "output_power": 0.0005764117867576953, "delta_db": 1.3942301872035854, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001153043826104485, "output_power": 0.00015735775290724266, "delta_db": 1.3504233043337657, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.6981779049220075e-05, "output_power": 7.91348712413242e-05, "delta_db": 1.4263189568090082, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.999437610841641e-06, "output_power": 3.230345080978813e-06, "delta_db": 0.322090853137459, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9563093402189873, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.2089647532583143, "pitch": null, "output_rms_dbfs": -29.107739501385026, "output_peak": 0.2928557232132792, "reference_rms_dbfs": -30.31670425464334, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.754207439258583}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.667347842324586e-05, "log_rms_mae_db": 1.8161910473673577, "rms_envelope_nrmse": 0.186975560814521, "envelope_correlation": 0.9891146532799888, "rms_gain_error_db": 1.210861552731354, "reference_mean_power": 0.0009300339703354519, "output_mean_power": 0.0012290936278724054}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3130295372770493e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001632046380308234, "output_power": 0.0002099554680902585, "reference_active": true, "delta_db": 1.0939469331705594}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.195140055898603e-05, "output_power": 3.9831361381099143e-05, "reference_active": true, "delta_db": 0.9573525045696099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.743176048402797e-06, "output_power": 4.610909823246248e-06, "reference_active": true, "delta_db": 0.9054637543610158}], "active_band_count": 3, "active_band_mae_db": 0.9855877307003951}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.141277463219576e-05, "log_rms_mae_db": 1.4035686497468842, "rms_envelope_nrmse": 0.1654830640381344, "envelope_correlation": 0.995520512645089, "rms_gain_error_db": 1.2125760383225799, "reference_mean_power": 0.0009316560347730994, "output_mean_power": 0.0012317234372720906}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0476212509594803e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014768349458157853, "output_power": 0.00019076068627415097, "reference_active": true, "delta_db": 1.1115691580965594}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4986865819787958e-05, "output_power": 1.8723758318474807e-05, "reference_active": true, "delta_db": 0.9668220772212348}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9796240655945466e-07, "output_power": 3.5750902958349585e-07, "reference_active": true, "delta_db": -0.46555033243847543}], "active_band_count": 3, "active_band_mae_db": 0.8479805225854232}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.73108999878072e-05, "log_rms_mae_db": 1.9362418551421121, "rms_envelope_nrmse": 0.21292450809936578, "envelope_correlation": 0.9888345401834542, "rms_gain_error_db": 1.3908414241965805, "reference_mean_power": 0.0005508019269041907, "output_mean_power": 0.0007587166125900791}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0420364913089255e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.387221376866869e-05, "output_power": 0.00012557376926136224, "reference_active": true, "delta_db": 1.2636187050435563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.257079739763329e-05, "output_power": 3.0115839402388945e-05, "reference_active": true, "delta_db": 1.2524807010706338}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1515320396484854e-06, "output_power": 2.925201478107606e-06, "reference_active": true, "delta_db": 1.3340796641672494}], "active_band_count": 3, "active_band_mae_db": 1.2833930234271465}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091336805501339e-05, "log_rms_mae_db": 1.5397805315356226, "rms_envelope_nrmse": 0.1943117495346633, "envelope_correlation": 0.9939783838800023, "rms_gain_error_db": 1.392173877808596, "reference_mean_power": 0.0005517580103658142, "output_mean_power": 0.0007602668156777218}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.88495020416125e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.537091697949908e-05, "output_power": 0.0001144185939129108, "reference_active": true, "delta_db": 1.2718666017191596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0155340206288246e-05, "output_power": 1.350589160897533e-05, "reference_active": true, "delta_db": 1.238287830321969}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.398901151759038e-07, "output_power": 2.5119553378011154e-07, "reference_active": true, "delta_db": 0.19999560483979384}], "active_band_count": 3, "active_band_mae_db": 0.9033833456269741}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.045309710707616e-05, "log_rms_mae_db": 1.6730754393693428, "rms_envelope_nrmse": 0.23119457483472833, "envelope_correlation": 0.9859360531484173, "rms_gain_error_db": 1.4326655775478, "reference_mean_power": 0.00014468505517057894, "output_mean_power": 0.0002012288438486451}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.678016723697031e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.684581121209281e-05, "output_power": 5.4960821325053706e-05, "reference_active": true, "delta_db": 1.7366509155716925}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.802139012131204e-06, "output_power": 1.2364991202620665e-05, "reference_active": true, "delta_db": 1.9998012775930007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.902802552587548e-06, "output_power": 3.2190866334668235e-06, "reference_active": true, "delta_db": 2.2833893941659293}], "active_band_count": 3, "active_band_mae_db": 2.006613862443541}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.539248843957361e-05, "log_rms_mae_db": 1.2092286973785829, "rms_envelope_nrmse": 0.20376363536434247, "envelope_correlation": 0.9937459014729166, "rms_gain_error_db": 1.437060238746729, "reference_mean_power": 0.0001448248558759811, "output_mean_power": 0.0002016272045242102}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.24555278166651e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.61104008282169e-05, "output_power": 5.454545610594973e-05, "reference_active": true, "delta_db": 1.791262684945654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6607268103199764e-06, "output_power": 5.728435402902149e-06, "reference_active": true, "delta_db": 1.9446870018194826}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4925257777124562e-07, "output_power": 2.404410873436723e-07, "reference_active": true, "delta_db": 2.070868426520348}], "active_band_count": 3, "active_band_mae_db": 1.9356060377618283}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.667503736683173e-05, "log_rms_mae_db": 1.668956850578569, "rms_envelope_nrmse": 0.22132625586543986, "envelope_correlation": 0.9742557762309456, "rms_gain_error_db": 0.99019274368163, "reference_mean_power": 8.017666159457313e-06, "output_mean_power": 1.0070875879524317e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.8921445305362116e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.737362730986327e-06, "output_power": 3.3461036762889134e-06, "reference_active": true, "delta_db": 0.8720704302536592}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.579169756744011e-07, "output_power": 9.471716096903518e-07, "reference_active": true, "delta_db": 0.968070377117901}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.809426009650544e-07, "output_power": 1.5363303806476556e-07, "reference_active": true, "delta_db": -0.7105621007349439}], "active_band_count": 3, "active_band_mae_db": 0.850234302702168}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4241724067408686e-05, "log_rms_mae_db": 1.068446242455174, "rms_envelope_nrmse": 0.1647977301943024, "envelope_correlation": 0.989927029705434, "rms_gain_error_db": 0.9914504920184831, "reference_mean_power": 8.048582997885633e-06, "output_mean_power": 1.0112638347840034e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.450576990854383e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.750016160596199e-06, "output_power": 3.3402080475541733e-06, "reference_active": true, "delta_db": 0.8443827204713823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.547469624529576e-07, "output_power": 4.4691307090261963e-07, "reference_active": true, "delta_db": 1.0030437122495357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2784081824019638e-08, "output_power": 9.328768738510025e-09, "reference_active": true, "delta_db": -1.3684521467587722}], "active_band_count": 3, "active_band_mae_db": 1.07195952649323}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9388946487743, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4f3015a598ccacc8696a044ea6253ee688cd7a246e5cbe17c962bbee363a01b9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9859360531484173, "standard_deviation": null, "minimum": 0.9859360531484173, "maximum": 0.9859360531484173, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6730754393693428, "standard_deviation": null, "minimum": 1.6730754393693428, "maximum": 1.6730754393693428, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 2.006613862443541, "standard_deviation": null, "minimum": 2.006613862443541, "maximum": 2.006613862443541, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23119457483472833, "standard_deviation": null, "minimum": 0.23119457483472833, "maximum": 0.23119457483472833, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9937459014729166, "standard_deviation": null, "minimum": 0.9937459014729166, "maximum": 0.9937459014729166, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2092286973785829, "standard_deviation": null, "minimum": 1.2092286973785829, "maximum": 1.2092286973785829, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.9356060377618283, "standard_deviation": null, "minimum": 1.9356060377618283, "maximum": 1.9356060377618283, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20376363536434247, "standard_deviation": null, "minimum": 0.20376363536434247, "maximum": 0.20376363536434247, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9742557762309456, "standard_deviation": null, "minimum": 0.9742557762309456, "maximum": 0.9742557762309456, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.668956850578569, "standard_deviation": null, "minimum": 1.668956850578569, "maximum": 1.668956850578569, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.850234302702168, "standard_deviation": null, "minimum": 0.850234302702168, "maximum": 0.850234302702168, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22132625586543986, "standard_deviation": null, "minimum": 0.22132625586543986, "maximum": 0.22132625586543986, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.989927029705434, "standard_deviation": null, "minimum": 0.989927029705434, "maximum": 0.989927029705434, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.068446242455174, "standard_deviation": null, "minimum": 1.068446242455174, "maximum": 1.068446242455174, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.07195952649323, "standard_deviation": null, "minimum": 1.07195952649323, "maximum": 1.07195952649323, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1647977301943024, "standard_deviation": null, "minimum": 0.1647977301943024, "maximum": 0.1647977301943024, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9888345401834542, "standard_deviation": null, "minimum": 0.9888345401834542, "maximum": 0.9888345401834542, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9362418551421121, "standard_deviation": null, "minimum": 1.9362418551421121, "maximum": 1.9362418551421121, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.2833930234271465, "standard_deviation": null, "minimum": 1.2833930234271465, "maximum": 1.2833930234271465, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21292450809936578, "standard_deviation": null, "minimum": 0.21292450809936578, "maximum": 0.21292450809936578, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9939783838800023, "standard_deviation": null, "minimum": 0.9939783838800023, "maximum": 0.9939783838800023, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5397805315356226, "standard_deviation": null, "minimum": 1.5397805315356226, "maximum": 1.5397805315356226, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9033833456269741, "standard_deviation": null, "minimum": 0.9033833456269741, "maximum": 0.9033833456269741, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1943117495346633, "standard_deviation": null, "minimum": 0.1943117495346633, "maximum": 0.1943117495346633, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9563093402189873, "standard_deviation": null, "minimum": 0.9563093402189873, "maximum": 0.9563093402189873, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.754207439258583, "standard_deviation": null, "minimum": 5.754207439258583, "maximum": 5.754207439258583, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.2928557232132792, "standard_deviation": null, "minimum": 0.2928557232132792, "maximum": 0.2928557232132792, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.2089647532583143, "standard_deviation": null, "minimum": 1.2089647532583143, "maximum": 1.2089647532583143, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9891146532799888, "standard_deviation": null, "minimum": 0.9891146532799888, "maximum": 0.9891146532799888, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8161910473673577, "standard_deviation": null, "minimum": 1.8161910473673577, "maximum": 1.8161910473673577, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9855877307003951, "standard_deviation": null, "minimum": 0.9855877307003951, "maximum": 0.9855877307003951, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.186975560814521, "standard_deviation": null, "minimum": 0.186975560814521, "maximum": 0.186975560814521, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.995520512645089, "standard_deviation": null, "minimum": 0.995520512645089, "maximum": 0.995520512645089, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.4035686497468842, "standard_deviation": null, "minimum": 1.4035686497468842, "maximum": 1.4035686497468842, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8479805225854232, "standard_deviation": null, "minimum": 0.8479805225854232, "maximum": 0.8479805225854232, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1654830640381344, "standard_deviation": null, "minimum": 0.1654830640381344, "maximum": 0.1654830640381344, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_237"}, {"case": "librispeech_237-134500-0029", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.2438244819641113, "predicted_mos": 2.2438244819641113, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 33040, "input_sample_rate": 16000, "input_seconds": 2.065, "input_float32_sha256": "fa83ba5b7c62e106c3ae70d460441ad6986faa70e1b4de1182c1b28afcb74360", "internal_samples": 33040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.414957500062883}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8085721876496041, "全帯域_10ms_rms_envelope_nrmse": 0.16616952118314807, "全帯域_10ms_envelope_correlation": 0.9846592226415863, "全帯域_10ms_modulation_mae_db": 0.4180435349712857, "全帯域_40ms_log_rms_mae_db": 1.330262208005396, "全帯域_40ms_rms_envelope_nrmse": 0.11220467358295905, "全帯域_40ms_envelope_correlation": 0.9971560364111374, "全帯域_40ms_modulation_mae_db": 0.46642133166074135, "300–1000 Hz_10ms_log_rms_mae_db": 1.8645183914714725, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19243399483630927, "300–1000 Hz_10ms_envelope_correlation": 0.9818397675983442, "300–1000 Hz_10ms_modulation_mae_db": 0.6818873818585436, "300–1000 Hz_40ms_log_rms_mae_db": 1.412758803907012, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.13667543496692527, "300–1000 Hz_40ms_envelope_correlation": 0.9961653917675825, "300–1000 Hz_40ms_modulation_mae_db": 0.8657268551846341, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7637239027635987, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16350776539084996, "1000–3000 Hz_10ms_envelope_correlation": 0.9868276064655325, "1000–3000 Hz_10ms_modulation_mae_db": 0.5402001859638667, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1770624047187235, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10484513518113621, "1000–3000 Hz_40ms_envelope_correlation": 0.9975747044082315, "1000–3000 Hz_40ms_modulation_mae_db": 0.6869446023330318, "3000–6000 Hz_10ms_log_rms_mae_db": 1.57763936283589, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1446078720891013, "3000–6000 Hz_10ms_envelope_correlation": 0.9881224838167227, "3000–6000 Hz_10ms_modulation_mae_db": 0.40205106059061285, "3000–6000 Hz_40ms_log_rms_mae_db": 0.952990697283675, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0758148555186716, "3000–6000 Hz_40ms_envelope_correlation": 0.9981386799196099, "3000–6000 Hz_40ms_modulation_mae_db": 0.556741628745936, "mr_stft_db": 5.815583491855713, "band_mae_db": 0.6641020809994108, "rms_gain_error_db": 0.8099449423064227, "output_peak": 1.374589474784451}, "full_scale": {"sample_count": 33040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 18, "at_or_above_full_scale_fraction": 0.0005447941888619854, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0005447941888619854, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 33040, "output_samples": 33040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.813012758900422, "spectral_convergence": 0.6230580707334687, "spectral_convergence_status": "ok", "amplitude_floor": 1.5756171103077008e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.051188850060774, "spectral_convergence": 0.24442077464336812, "spectral_convergence_status": "ok", "amplitude_floor": 1.2505050051947005e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5825488666059435, "spectral_convergence": 0.2993365181439167, "spectral_convergence_status": "ok", "amplitude_floor": 9.121352204658788e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021685618865241799, "output_power": 0.00019315057571541393, "delta_db": -0.5027581361339584, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010795690773652512, "output_power": 0.001297502486406211, "delta_db": 0.7985776241104968, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002989928450689294, "output_power": 0.003752545760423229, "delta_db": 0.9866520121719068, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0020580355715967367, "output_power": 0.002442134164675808, "delta_db": 0.7431664236022433, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0009751360590152492, "output_power": 0.0011375089099733495, "delta_db": 0.6688959085018746, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.9392123549665832e-05, "output_power": 2.070530382319988e-05, "delta_db": 0.2845623814759845, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6641020809994108, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8099449423064227, "pitch": null, "output_rms_dbfs": -20.533735018615108, "output_peak": 1.374589474784451, "reference_rms_dbfs": -21.343679960921527, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.815583491855713}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 33040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00032988134947798775, "log_rms_mae_db": 1.8085721876496041, "rms_envelope_nrmse": 0.16616952118314807, "envelope_correlation": 0.9846592226415863, "rms_gain_error_db": 0.8099475607336902, "reference_mean_power": 0.007356724541990913, "output_mean_power": 0.008865010434525747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 2.980285612038863e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0020727352824950612, "output_power": 0.0025188116562142745, "reference_active": true, "delta_db": 0.8465185429373024}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00028987718896809884, "output_power": 0.00029914005721055334, "reference_active": true, "delta_db": 0.13660531204252707}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.012977878291207e-05, "output_power": 6.588731259820715e-05, "reference_active": true, "delta_db": -0.2710067499340275}], "active_band_count": 3, "active_band_mae_db": 0.4180435349712857}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0002960474071300256, "log_rms_mae_db": 1.330262208005396, "rms_envelope_nrmse": 0.11220467358295905, "envelope_correlation": 0.9971560364111374, "rms_gain_error_db": 0.8099500731399425, "reference_mean_power": 0.007465415825536721, "output_mean_power": 0.008995990960326605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 2.5597102021075326e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001877761726334087, "output_power": 0.0022966465424962084, "reference_active": true, "delta_db": 0.8745367897291967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012702101262043316, "output_power": 0.0001362600025290403, "reference_active": true, "delta_db": 0.3049282202109262}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.962314408759724e-06, "output_power": 3.7667694133517164e-06, "reference_active": true, "delta_db": -0.2197989850421012}], "active_band_count": 3, "active_band_mae_db": 0.46642133166074135}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.00021933697126022545, "log_rms_mae_db": 1.8645183914714725, "rms_envelope_nrmse": 0.19243399483630927, "envelope_correlation": 0.9818397675983442, "rms_gain_error_db": 0.982267701495163, "reference_mean_power": 0.003363603458339666, "output_mean_power": 0.004217271495222723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.3609823101859838e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008546335077009212, "output_power": 0.0011065273959099805, "reference_active": true, "delta_db": 1.1218225461533813}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016660612815788933, "output_power": 0.00018171816507020125, "reference_active": true, "delta_db": 0.37707371085124475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.188726605138031e-05, "output_power": 3.6932221058397016e-05, "reference_active": true, "delta_db": -0.5467658885710049}], "active_band_count": 3, "active_band_mae_db": 0.6818873818585436}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.0001893291578274055, "log_rms_mae_db": 1.412758803907012, "rms_envelope_nrmse": 0.13667543496692527, "envelope_correlation": 0.9961653917675825, "rms_gain_error_db": 0.9822708018025499, "reference_mean_power": 0.0034133051942474998, "output_mean_power": 0.004279590372218418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.15224102535942e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007711766819603955, "output_power": 0.0010003833303151753, "reference_active": true, "delta_db": 1.1301255699290589}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.304889773698623e-05, "output_power": 8.643494390723932e-05, "reference_active": true, "delta_db": 0.7307568741023258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527118205585199e-06, "output_power": 2.133024011322826e-06, "reference_active": true, "delta_db": -0.7362981215225174}], "active_band_count": 3, "active_band_mae_db": 0.8657268551846341}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 0.0002256735995145706, "log_rms_mae_db": 1.7637239027635987, "rms_envelope_nrmse": 0.16350776539084996, "envelope_correlation": 0.9868276064655325, "rms_gain_error_db": 0.733504047610122, "reference_mean_power": 0.002711357875845494, "output_mean_power": 0.0032102381331353897}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 1.6613986227009806e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012279896352023905, "output_power": 0.0014239668663509825, "reference_active": true, "delta_db": 0.6430518283873663}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011888327196960883, "output_power": 0.00013736334149307196, "reference_active": true, "delta_db": 0.627500974367398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.234146057131384e-05, "output_power": 2.9836979345523555e-05, "reference_active": true, "delta_db": -0.3500477551368356}], "active_band_count": 3, "active_band_mae_db": 0.5402001859638667}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 0.00020374269470838238, "log_rms_mae_db": 1.1770624047187235, "rms_envelope_nrmse": 0.10484513518113621, "envelope_correlation": 0.9975747044082315, "rms_gain_error_db": 0.7335021793937101, "reference_mean_power": 0.002751412553659036, "output_mean_power": 0.003257661328609386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 1.4886745667367597e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001137287164515336, "output_power": 0.0013217838634632605, "reference_active": true, "delta_db": 0.6529030793158235}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.907501205097428e-05, "output_power": 5.9113983338341595e-05, "reference_active": true, "delta_db": 0.808298099282627}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5268858664777347e-06, "output_power": 2.2010117623486725e-06, "reference_active": true, "delta_db": -0.5996326284006447}], "active_band_count": 3, "active_band_mae_db": 0.6869446023330318}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 412, "rms_floor": 4.718157231228245e-05, "log_rms_mae_db": 1.57763936283589, "rms_envelope_nrmse": 0.1446078720891013, "envelope_correlation": 0.9881224838167227, "rms_gain_error_db": 0.4752704746971481, "reference_mean_power": 9.509723574878595e-05, "output_mean_power": 0.00010609500575098944}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 412, "frequency_resolution_hz": 0.4854368932038835, "power_floor": 5.720070068605411e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.50238140104046e-05, "output_power": 3.780585151476063e-05, "reference_active": true, "delta_db": 0.33195586315556946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.975419310826897e-06, "output_power": 5.007699743415855e-06, "reference_active": true, "delta_db": 0.028085936930973585}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.471999704474668e-06, "output_power": 1.211425147143459e-06, "reference_active": true, "delta_db": -0.8461113816852954}], "active_band_count": 3, "active_band_mae_db": 0.40205106059061285}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 33040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 406, "rms_floor": 4.015479902310721e-05, "log_rms_mae_db": 0.952990697283675, "rms_envelope_nrmse": 0.0758148555186716, "envelope_correlation": 0.9981386799196099, "rms_gain_error_db": 0.475304353208909, "reference_mean_power": 9.64997641207967e-05, "output_mean_power": 0.00010766057304324203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 406, "frequency_resolution_hz": 0.4926108374384236, "power_floor": 5.087796481772889e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1685652820939687e-05, "output_power": 3.409196636023985e-05, "reference_active": true, "delta_db": 0.31789391894594815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.336227531732081e-06, "output_power": 2.321247304656813e-06, "reference_active": true, "delta_db": -0.027937251921568096}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1955736150290575e-07, "output_power": 8.813267470154988e-08, "reference_active": true, "delta_db": -1.324393715370292}], "active_band_count": 3, "active_band_mae_db": 0.556741628745936}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.943454724991852, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c05e2787598dc93097971e3c6568dd43d1f2766fc9e599b9b656847871b7e7b8", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9868276064655325, "standard_deviation": null, "minimum": 0.9868276064655325, "maximum": 0.9868276064655325, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7637239027635987, "standard_deviation": null, "minimum": 1.7637239027635987, "maximum": 1.7637239027635987, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5402001859638667, "standard_deviation": null, "minimum": 0.5402001859638667, "maximum": 0.5402001859638667, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16350776539084996, "standard_deviation": null, "minimum": 0.16350776539084996, "maximum": 0.16350776539084996, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9975747044082315, "standard_deviation": null, "minimum": 0.9975747044082315, "maximum": 0.9975747044082315, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1770624047187235, "standard_deviation": null, "minimum": 1.1770624047187235, "maximum": 1.1770624047187235, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6869446023330318, "standard_deviation": null, "minimum": 0.6869446023330318, "maximum": 0.6869446023330318, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10484513518113621, "standard_deviation": null, "minimum": 0.10484513518113621, "maximum": 0.10484513518113621, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9881224838167227, "standard_deviation": null, "minimum": 0.9881224838167227, "maximum": 0.9881224838167227, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.57763936283589, "standard_deviation": null, "minimum": 1.57763936283589, "maximum": 1.57763936283589, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.40205106059061285, "standard_deviation": null, "minimum": 0.40205106059061285, "maximum": 0.40205106059061285, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1446078720891013, "standard_deviation": null, "minimum": 0.1446078720891013, "maximum": 0.1446078720891013, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9981386799196099, "standard_deviation": null, "minimum": 0.9981386799196099, "maximum": 0.9981386799196099, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.952990697283675, "standard_deviation": null, "minimum": 0.952990697283675, "maximum": 0.952990697283675, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.556741628745936, "standard_deviation": null, "minimum": 0.556741628745936, "maximum": 0.556741628745936, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0758148555186716, "standard_deviation": null, "minimum": 0.0758148555186716, "maximum": 0.0758148555186716, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9818397675983442, "standard_deviation": null, "minimum": 0.9818397675983442, "maximum": 0.9818397675983442, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8645183914714725, "standard_deviation": null, "minimum": 1.8645183914714725, "maximum": 1.8645183914714725, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6818873818585436, "standard_deviation": null, "minimum": 0.6818873818585436, "maximum": 0.6818873818585436, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19243399483630927, "standard_deviation": null, "minimum": 0.19243399483630927, "maximum": 0.19243399483630927, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961653917675825, "standard_deviation": null, "minimum": 0.9961653917675825, "maximum": 0.9961653917675825, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.412758803907012, "standard_deviation": null, "minimum": 1.412758803907012, "maximum": 1.412758803907012, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8657268551846341, "standard_deviation": null, "minimum": 0.8657268551846341, "maximum": 0.8657268551846341, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13667543496692527, "standard_deviation": null, "minimum": 0.13667543496692527, "maximum": 0.13667543496692527, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6641020809994108, "standard_deviation": null, "minimum": 0.6641020809994108, "maximum": 0.6641020809994108, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.815583491855713, "standard_deviation": null, "minimum": 5.815583491855713, "maximum": 5.815583491855713, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.374589474784451, "standard_deviation": null, "minimum": 1.374589474784451, "maximum": 1.374589474784451, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8099449423064227, "standard_deviation": null, "minimum": 0.8099449423064227, "maximum": 0.8099449423064227, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9846592226415863, "standard_deviation": null, "minimum": 0.9846592226415863, "maximum": 0.9846592226415863, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8085721876496041, "standard_deviation": null, "minimum": 1.8085721876496041, "maximum": 1.8085721876496041, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4180435349712857, "standard_deviation": null, "minimum": 0.4180435349712857, "maximum": 0.4180435349712857, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16616952118314807, "standard_deviation": null, "minimum": 0.16616952118314807, "maximum": 0.16616952118314807, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9971560364111374, "standard_deviation": null, "minimum": 0.9971560364111374, "maximum": 0.9971560364111374, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.330262208005396, "standard_deviation": null, "minimum": 1.330262208005396, "maximum": 1.330262208005396, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.46642133166074135, "standard_deviation": null, "minimum": 0.46642133166074135, "maximum": 0.46642133166074135, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11220467358295905, "standard_deviation": null, "minimum": 0.11220467358295905, "maximum": 0.11220467358295905, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_237"}, {"case": "librispeech_237-134500-0031", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.094668388366699, "predicted_mos": 3.094668388366699, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1cd6e7be3b5996bf2a30dd190a95927e6dfcc0aca2e6abdbf31d8d922ef87b2d", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.749381300061941}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7539899665890415, "全帯域_10ms_rms_envelope_nrmse": 0.173815004736086, "全帯域_10ms_envelope_correlation": 0.9877700734911293, "全帯域_10ms_modulation_mae_db": 0.858054065874005, "全帯域_40ms_log_rms_mae_db": 1.2524545906507105, "全帯域_40ms_rms_envelope_nrmse": 0.14167006663224818, "全帯域_40ms_envelope_correlation": 0.996096890852907, "全帯域_40ms_modulation_mae_db": 0.9084764576838862, "300–1000 Hz_10ms_log_rms_mae_db": 1.9654737739758352, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1991718915889637, "300–1000 Hz_10ms_envelope_correlation": 0.9879351339302533, "300–1000 Hz_10ms_modulation_mae_db": 0.986979474376319, "300–1000 Hz_40ms_log_rms_mae_db": 1.4800508877600564, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17043432678201356, "300–1000 Hz_40ms_envelope_correlation": 0.9953292176550633, "300–1000 Hz_40ms_modulation_mae_db": 0.94105167668571, "1000–3000 Hz_10ms_log_rms_mae_db": 1.628983690818604, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15327219147856383, "1000–3000 Hz_10ms_envelope_correlation": 0.9891586963935364, "1000–3000 Hz_10ms_modulation_mae_db": 0.6147174770865398, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1638580152854452, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10829657827439536, "1000–3000 Hz_40ms_envelope_correlation": 0.9974144595035647, "1000–3000 Hz_40ms_modulation_mae_db": 0.5535516066748366, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5131544382551516, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2367007372373643, "3000–6000 Hz_10ms_envelope_correlation": 0.9554103099108027, "3000–6000 Hz_10ms_modulation_mae_db": 0.6417300166439496, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9446366606307245, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13532037625687238, "3000–6000 Hz_40ms_envelope_correlation": 0.9847978408441121, "3000–6000 Hz_40ms_modulation_mae_db": 0.5759440529358691, "mr_stft_db": 5.706925474291737, "band_mae_db": 0.6465066302953999, "rms_gain_error_db": 1.0317284975094152, "output_peak": 0.9103883664783726}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.310177525539367, "spectral_convergence": 0.5411751546439246, "spectral_convergence_status": "ok", "amplitude_floor": 2.051312383514402e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.165129682990491, "spectral_convergence": 0.25767597022448874, "spectral_convergence_status": "ok", "amplitude_floor": 1.425885904641783e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.645469214345349, "spectral_convergence": 0.3065152999592032, "spectral_convergence_status": "ok", "amplitude_floor": 8.483800566678623e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001304494586920238, "output_power": 0.00011748603476191219, "delta_db": -0.45456035009681306, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014301207186140645, "output_power": 0.0016928792865640307, "delta_db": 0.7325329270315394, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.004308309383381418, "output_power": 0.005716047240094601, "delta_db": 1.227889265430502, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.001304992137209625, "output_power": 0.0016118584764256848, "delta_db": 0.917190124536731, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002442775973124686, "output_power": 0.0002719742586321881, "delta_db": 0.4664416192445272, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0021123865382114e-05, "output_power": 9.837253899810225e-06, "delta_db": -0.08042549543228651, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6465066302953999, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0317284975094152, "pitch": null, "output_rms_dbfs": -20.259452913667133, "output_peak": 0.9103883664783726, "reference_rms_dbfs": -21.29118141117655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.706925474291737}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002631236493882886, "log_rms_mae_db": 1.7539899665890415, "rms_envelope_nrmse": 0.173815004736086, "envelope_correlation": 0.9877700734911293, "rms_gain_error_db": 1.031728787605343, "reference_mean_power": 0.007434364903539191, "output_mean_power": 0.009427938755832765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.450307530079474e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015621941098031631, "output_power": 0.0020419025500365393, "reference_active": true, "delta_db": 1.163000155254099}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031400948871223737, "output_power": 0.00039050168470531547, "reference_active": true, "delta_db": 0.9468014010837817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9119711962951405e-05, "output_power": 5.4662800216874595e-05, "reference_active": true, "delta_db": 0.46436064128413457}], "active_band_count": 3, "active_band_mae_db": 0.858054065874005}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00024033810696394756, "log_rms_mae_db": 1.2524545906507105, "rms_envelope_nrmse": 0.14167006663224818, "envelope_correlation": 0.996096890852907, "rms_gain_error_db": 1.031729950802597, "reference_mean_power": 0.007471745584669175, "output_mean_power": 0.00947534585033768}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.136699874330791e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014162642987993615, "output_power": 0.0018741311217673699, "reference_active": true, "delta_db": 1.2165566501673515}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001495506387635467, "output_power": 0.00019001739683941501, "reference_active": true, "delta_db": 1.040050918014222}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0546566248780076e-06, "output_power": 3.4028638927796146e-06, "reference_active": true, "delta_db": 0.4688218048700849}], "active_band_count": 3, "active_band_mae_db": 0.9084764576838862}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002303175773961387, "log_rms_mae_db": 1.9654737739758352, "rms_envelope_nrmse": 0.1991718915889637, "envelope_correlation": 0.9879351339302533, "rms_gain_error_db": 1.2357564053357428, "reference_mean_power": 0.004944923373977616, "output_mean_power": 0.0065725697965467724}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6955719701225893e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012875788279854436, "output_power": 0.0017109273377179386, "reference_active": true, "delta_db": 1.2345773880158921}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026650258378785487, "output_power": 0.0003558790018953264, "reference_active": true, "delta_db": 1.2560093992595722}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0302791253002337e-05, "output_power": 3.376896871217419e-05, "reference_active": true, "delta_db": 0.47035163585349304}], "active_band_count": 3, "active_band_mae_db": 0.986979474376319}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020459712012168156, "log_rms_mae_db": 1.4800508877600564, "rms_envelope_nrmse": 0.17043432678201356, "envelope_correlation": 0.9953292176550633, "rms_gain_error_db": 1.2357565514336153, "reference_mean_power": 0.004969790940376444, "output_mean_power": 0.00660562286983053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.46585734788702e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011840763589577269, "output_power": 0.0015797860889187655, "reference_active": true, "delta_db": 1.2521857508981773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012134699925967192, "output_power": 0.00016012773097741543, "reference_active": true, "delta_db": 1.204375083954138}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.482336680977429e-06, "output_power": 2.7009721386327705e-06, "reference_active": true, "delta_db": 0.36659419520481484}], "active_band_count": 3, "active_band_mae_db": 0.94105167668571}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014083569934759876, "log_rms_mae_db": 1.628983690818604, "rms_envelope_nrmse": 0.15327219147856383, "envelope_correlation": 0.9891586963935364, "rms_gain_error_db": 0.7626439217705057, "reference_mean_power": 0.0013181968247743324, "output_mean_power": 0.001571247694201317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.706116677050846e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003562978060875054, "output_power": 0.00042077841428633453, "reference_active": true, "delta_db": 0.7224030445590922}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927249258823322e-05, "output_power": 8.081162285559938e-05, "reference_active": true, "delta_db": 0.6691301347557173}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3444445693793863e-05, "output_power": 1.492123692888995e-05, "reference_active": true, "delta_db": 0.45261925194480973}], "active_band_count": 3, "active_band_mae_db": 0.6147174770865398}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010879087849669345, "log_rms_mae_db": 1.1638580152854452, "rms_envelope_nrmse": 0.10829657827439536, "envelope_correlation": 0.9974144595035647, "rms_gain_error_db": 0.762648823488441, "reference_mean_power": 0.0013248216670349005, "output_mean_power": 0.0015791460727904109}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.107952744901008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003355694425017771, "output_power": 0.0003960237639741017, "reference_active": true, "delta_db": 0.7193884077427897}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.409137696567018e-05, "output_power": 4.037741560447053e-05, "reference_active": true, "delta_db": 0.7349397513113618}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.097157965658087e-07, "output_power": 8.491125854865177e-07, "reference_active": true, "delta_db": 0.20632666097035823}], "active_band_count": 3, "active_band_mae_db": 0.5535516066748366}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.557718084358035e-05, "log_rms_mae_db": 1.5131544382551516, "rms_envelope_nrmse": 0.2367007372373643, "envelope_correlation": 0.9554103099108027, "rms_gain_error_db": 0.4822867309641755, "reference_mean_power": 2.3934708718272692e-05, "output_mean_power": 2.6745875560466396e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3699291941082093e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.018836822609012e-06, "output_power": 4.658992690847256e-06, "reference_active": true, "delta_db": 0.6419165628521055}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.745214583455981e-06, "output_power": 5.866356794319006e-06, "reference_active": true, "delta_db": 0.9211261697523282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3999346848964708e-06, "output_power": 1.521677102206151e-06, "reference_active": true, "delta_db": 0.36214731732741506}], "active_band_count": 3, "active_band_mae_db": 0.6417300166439496}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.6234588308267947e-05, "log_rms_mae_db": 0.9446366606307245, "rms_envelope_nrmse": 0.13532037625687238, "envelope_correlation": 0.9847978408441121, "rms_gain_error_db": 0.48232582925127043, "reference_mean_power": 2.405439007374668e-05, "output_mean_power": 2.687985565821331e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1165079820636279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.688364032412796e-06, "output_power": 4.51875655999985e-06, "reference_active": true, "delta_db": 0.8818516682714271}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313243490564173e-06, "output_power": 3.96209360990831e-06, "reference_active": true, "delta_db": 0.7767137873232182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.200263616160695e-07, "output_power": 1.2195603935180282e-07, "reference_active": true, "delta_db": 0.06926670321296197}], "active_band_count": 3, "active_band_mae_db": 0.5759440529358691}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9335193324845718, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f5837d33c19543ce31c99b62d28a7852c00dedc94d69012c725425f3685859f7", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9891586963935364, "standard_deviation": null, "minimum": 0.9891586963935364, "maximum": 0.9891586963935364, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.628983690818604, "standard_deviation": null, "minimum": 1.628983690818604, "maximum": 1.628983690818604, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6147174770865398, "standard_deviation": null, "minimum": 0.6147174770865398, "maximum": 0.6147174770865398, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15327219147856383, "standard_deviation": null, "minimum": 0.15327219147856383, "maximum": 0.15327219147856383, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974144595035647, "standard_deviation": null, "minimum": 0.9974144595035647, "maximum": 0.9974144595035647, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1638580152854452, "standard_deviation": null, "minimum": 1.1638580152854452, "maximum": 1.1638580152854452, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5535516066748366, "standard_deviation": null, "minimum": 0.5535516066748366, "maximum": 0.5535516066748366, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10829657827439536, "standard_deviation": null, "minimum": 0.10829657827439536, "maximum": 0.10829657827439536, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9554103099108027, "standard_deviation": null, "minimum": 0.9554103099108027, "maximum": 0.9554103099108027, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5131544382551516, "standard_deviation": null, "minimum": 1.5131544382551516, "maximum": 1.5131544382551516, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6417300166439496, "standard_deviation": null, "minimum": 0.6417300166439496, "maximum": 0.6417300166439496, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2367007372373643, "standard_deviation": null, "minimum": 0.2367007372373643, "maximum": 0.2367007372373643, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9847978408441121, "standard_deviation": null, "minimum": 0.9847978408441121, "maximum": 0.9847978408441121, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9446366606307245, "standard_deviation": null, "minimum": 0.9446366606307245, "maximum": 0.9446366606307245, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5759440529358691, "standard_deviation": null, "minimum": 0.5759440529358691, "maximum": 0.5759440529358691, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13532037625687238, "standard_deviation": null, "minimum": 0.13532037625687238, "maximum": 0.13532037625687238, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9879351339302533, "standard_deviation": null, "minimum": 0.9879351339302533, "maximum": 0.9879351339302533, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9654737739758352, "standard_deviation": null, "minimum": 1.9654737739758352, "maximum": 1.9654737739758352, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.986979474376319, "standard_deviation": null, "minimum": 0.986979474376319, "maximum": 0.986979474376319, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1991718915889637, "standard_deviation": null, "minimum": 0.1991718915889637, "maximum": 0.1991718915889637, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9953292176550633, "standard_deviation": null, "minimum": 0.9953292176550633, "maximum": 0.9953292176550633, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4800508877600564, "standard_deviation": null, "minimum": 1.4800508877600564, "maximum": 1.4800508877600564, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.94105167668571, "standard_deviation": null, "minimum": 0.94105167668571, "maximum": 0.94105167668571, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17043432678201356, "standard_deviation": null, "minimum": 0.17043432678201356, "maximum": 0.17043432678201356, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6465066302953999, "standard_deviation": null, "minimum": 0.6465066302953999, "maximum": 0.6465066302953999, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.706925474291737, "standard_deviation": null, "minimum": 5.706925474291737, "maximum": 5.706925474291737, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.9103883664783726, "standard_deviation": null, "minimum": 0.9103883664783726, "maximum": 0.9103883664783726, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0317284975094152, "standard_deviation": null, "minimum": 1.0317284975094152, "maximum": 1.0317284975094152, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9877700734911293, "standard_deviation": null, "minimum": 0.9877700734911293, "maximum": 0.9877700734911293, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7539899665890415, "standard_deviation": null, "minimum": 1.7539899665890415, "maximum": 1.7539899665890415, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.858054065874005, "standard_deviation": null, "minimum": 0.858054065874005, "maximum": 0.858054065874005, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.173815004736086, "standard_deviation": null, "minimum": 0.173815004736086, "maximum": 0.173815004736086, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.996096890852907, "standard_deviation": null, "minimum": 0.996096890852907, "maximum": 0.996096890852907, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2524545906507105, "standard_deviation": null, "minimum": 1.2524545906507105, "maximum": 1.2524545906507105, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9084764576838862, "standard_deviation": null, "minimum": 0.9084764576838862, "maximum": 0.9084764576838862, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.14167006663224818, "standard_deviation": null, "minimum": 0.14167006663224818, "maximum": 0.14167006663224818, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_237"}, {"case": "librispeech_260-123286-0008", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.0943896770477295, "predicted_mos": 3.0943896770477295, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59760, "input_sample_rate": 16000, "input_seconds": 3.735, "input_float32_sha256": "0956a32d934912ac9cb5d55664d8df521dd631377033e9dab42d9abad25785e1", "internal_samples": 59760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.272937199566513}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.810124329465423, "全帯域_10ms_rms_envelope_nrmse": 0.2269843871633539, "全帯域_10ms_envelope_correlation": 0.9818143536217451, "全帯域_10ms_modulation_mae_db": 0.9797381214721176, "全帯域_40ms_log_rms_mae_db": 1.4533263787645105, "全帯域_40ms_rms_envelope_nrmse": 0.2029226738195644, "全帯域_40ms_envelope_correlation": 0.988967499653238, "全帯域_40ms_modulation_mae_db": 1.0231963093140821, "300–1000 Hz_10ms_log_rms_mae_db": 2.123151647577371, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22043033524605754, "300–1000 Hz_10ms_envelope_correlation": 0.9853741270173771, "300–1000 Hz_10ms_modulation_mae_db": 1.1350704142886319, "300–1000 Hz_40ms_log_rms_mae_db": 1.6933248017645768, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.18148442518136362, "300–1000 Hz_40ms_envelope_correlation": 0.9945790616821423, "300–1000 Hz_40ms_modulation_mae_db": 0.9828299868221865, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6157638952089468, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22701535946128296, "1000–3000 Hz_10ms_envelope_correlation": 0.9728947490425887, "1000–3000 Hz_10ms_modulation_mae_db": 0.9865218396602463, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1466326202100106, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17749711035378943, "1000–3000 Hz_40ms_envelope_correlation": 0.9855661432701143, "1000–3000 Hz_40ms_modulation_mae_db": 1.2703220033944027, "3000–6000 Hz_10ms_log_rms_mae_db": 1.564792512769499, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21256451362121964, "3000–6000 Hz_10ms_envelope_correlation": 0.9732675226736189, "3000–6000 Hz_10ms_modulation_mae_db": 0.2615099922202047, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0326428959385767, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1236252735244945, "3000–6000 Hz_40ms_envelope_correlation": 0.9914239003302865, "3000–6000 Hz_40ms_modulation_mae_db": 0.44980600024937006, "mr_stft_db": 5.455049789472159, "band_mae_db": 0.9117516917798718, "rms_gain_error_db": 1.3115708722247459, "output_peak": 0.6388974486426948}, "full_scale": {"sample_count": 59760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59760, "output_samples": 59760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.272910085872795, "spectral_convergence": 0.5783702522657599, "spectral_convergence_status": "ok", "amplitude_floor": 1.6701466153856754e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.003545131675021, "spectral_convergence": 0.33030780027010476, "spectral_convergence_status": "ok", "amplitude_floor": 1.1197689318182202e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.088694150868659, "spectral_convergence": 0.351079971758263, "spectral_convergence_status": "ok", "amplitude_floor": 1.0416079453540003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0010765709682483164, "output_power": 0.0011740676382487464, "delta_db": 0.37650453342082124, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019015067355649686, "output_power": 0.002876078352318595, "delta_db": 1.7970284513413983, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009390993860770502, "output_power": 0.0012860358140499753, "delta_db": 1.365415065332292, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028731828855173817, "output_power": 0.0003813743435156755, "delta_db": 1.229882022337806, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.911907847527107e-05, "output_power": 6.524423447584529e-05, "delta_db": 0.4281448399754217, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.5637714608274427e-05, "output_power": 5.925470035763556e-05, "delta_db": 0.27353523827149223, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9117516917798718, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.3115708722247459, "pitch": null, "output_rms_dbfs": -22.334343525121426, "output_peak": 0.6388974486426948, "reference_rms_dbfs": -23.64591439734617, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.455049789472159}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00019933639001390812, "log_rms_mae_db": 1.810124329465423, "rms_envelope_nrmse": 0.2269843871633539, "envelope_correlation": 0.9818143536217451, "rms_gain_error_db": 1.3115708903616463, "reference_mean_power": 0.004325042032530898, "output_mean_power": 0.0058498862516716795}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 2.983288998803909e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014406782878671795, "output_power": 0.0020725442874359743, "reference_active": true, "delta_db": 1.5793680853157186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010200208961765019, "output_power": 0.00012389656636511246, "reference_active": true, "delta_db": 0.8445020180477805}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7770617096993514e-05, "output_power": 3.1269429410319336e-05, "reference_active": true, "delta_db": 0.515344261052854}], "active_band_count": 3, "active_band_mae_db": 0.9797381214721176}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00018454723874230657, "log_rms_mae_db": 1.4533263787645105, "rms_envelope_nrmse": 0.2029226738195644, "envelope_correlation": 0.988967499653238, "rms_gain_error_db": 1.3115709373123399, "reference_mean_power": 0.004360109839005934, "output_mean_power": 0.005897317687753312}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 2.740771226026708e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013018843976497208, "output_power": 0.0019042117219171674, "reference_active": true, "delta_db": 1.6514281199551366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649956473480408e-05, "output_power": 6.102582519697776e-05, "reference_active": true, "delta_db": 1.1806477306071894}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9034855452992526e-06, "output_power": 1.8021804554405664e-06, "reference_active": true, "delta_db": -0.2375130773799201}], "active_band_count": 3, "active_band_mae_db": 1.0231963093140821}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 0.00017949843182759416, "log_rms_mae_db": 2.123151647577371, "rms_envelope_nrmse": 0.22043033524605754, "envelope_correlation": 0.9853741270173771, "rms_gain_error_db": 1.2629731713865346, "reference_mean_power": 0.001722884360599687, "output_mean_power": 0.002304376543626723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.0185329105758934e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006952905258544651, "output_power": 0.0009005802360622348, "reference_active": true, "delta_db": 1.1235610014301054}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.45148729518806e-05, "output_power": 9.001015160100657e-05, "reference_active": true, "delta_db": 1.446316468076636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5243024554692387e-05, "output_power": 1.8475854206724518e-05, "reference_active": true, "delta_db": 0.8353337733591546}], "active_band_count": 3, "active_band_mae_db": 1.1350704142886319}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 0.00016126256915979383, "log_rms_mae_db": 1.6933248017645768, "rms_envelope_nrmse": 0.18148442518136362, "envelope_correlation": 0.9945790616821423, "rms_gain_error_db": 1.2629731218944582, "reference_mean_power": 0.0017368536839123, "output_mean_power": 0.0023230606387971506}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 9.447738359038237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006450702777726167, "output_power": 0.0008409800026345126, "reference_active": true, "delta_db": 1.1517863716291228}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9818892488309525e-05, "output_power": 4.396577534491933e-05, "reference_active": true, "delta_db": 1.6862322726284618}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0738004145337803e-06, "output_power": 1.1014649931022683e-06, "reference_active": true, "delta_db": 0.11047131620897514}], "active_band_count": 3, "active_band_mae_db": 0.9828299868221865}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 6.626476450042254e-05, "log_rms_mae_db": 1.6157638952089468, "rms_envelope_nrmse": 0.22701535946128296, "envelope_correlation": 0.9728947490425887, "rms_gain_error_db": 0.9191225070054443, "reference_mean_power": 0.00025607969857314127, "output_mean_power": 0.00031643710355993193}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 1.9689769875456006e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013521942671011384, "output_power": 0.00015977879185513565, "reference_active": true, "delta_db": 0.7248004265292984}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.463778228624424e-05, "output_power": 1.8379233391604652e-05, "reference_active": true, "delta_db": 0.9885210932548256}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5896173962011905e-06, "output_power": 4.195248414691538e-06, "reference_active": true, "delta_db": -1.2462439991966148}], "active_band_count": 3, "active_band_mae_db": 0.9865218396602463}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 5.564377890100535e-05, "log_rms_mae_db": 1.1466326202100106, "rms_envelope_nrmse": 0.17749711035378943, "envelope_correlation": 0.9855661432701143, "rms_gain_error_db": 0.9191225296001675, "reference_mean_power": 0.0002581560174430691, "output_mean_power": 0.00031900280774472704}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 1.774101433771277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012602347588101508, "output_power": 0.0001495363420939266, "reference_active": true, "delta_db": 0.7429529906452041}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.970334241514461e-06, "output_power": 8.896567746119404e-06, "reference_active": true, "delta_db": 1.059688863041093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.812565026245218e-07, "output_power": 2.40095955345079e-07, "reference_active": true, "delta_db": -2.008324156496911}], "active_band_count": 3, "active_band_mae_db": 1.2703220033944027}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 746, "rms_floor": 3.541043489510122e-05, "log_rms_mae_db": 1.564792512769499, "rms_envelope_nrmse": 0.21256451362121964, "envelope_correlation": 0.9732675226736189, "rms_gain_error_db": 0.41808716998401996, "reference_mean_power": 6.312530293815556e-05, "output_mean_power": 6.950438298551332e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 746, "frequency_resolution_hz": 0.2680965147453083, "power_floor": 4.34877201852775e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2352362117577493e-05, "output_power": 2.4628183912120994e-05, "reference_active": true, "delta_db": 0.4210896346739396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.837592049886246e-06, "output_power": 9.253923813984733e-06, "reference_active": true, "delta_db": 0.19991969386044495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1202466049559064e-06, "output_power": 2.041899259581167e-06, "reference_active": true, "delta_db": -0.16352064812622957}], "active_band_count": 3, "active_band_mae_db": 0.2615099922202047}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 740, "rms_floor": 3.3704303759041684e-05, "log_rms_mae_db": 1.0326428959385767, "rms_envelope_nrmse": 0.1236252735244945, "envelope_correlation": 0.9914239003302865, "rms_gain_error_db": 0.4180872057983539, "reference_mean_power": 6.363712790649308e-05, "output_mean_power": 7.006793061918938e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 740, "frequency_resolution_hz": 0.27027027027027023, "power_floor": 3.9430492064624054e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.36389645521954e-05, "output_power": 2.576927142776233e-05, "reference_active": true, "delta_db": 0.3747369055263977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.125341563464706e-06, "output_power": 4.295461737925393e-06, "reference_active": true, "delta_db": 0.17549942487587775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9054004218332769e-07, "output_power": 1.5851416962613199e-07, "reference_active": true, "delta_db": -0.7991816703458348}], "active_band_count": 3, "active_band_mae_db": 0.44980600024937006}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9020920292569358, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1c135f7867c38628de500c17ebe3db3c8d9f282e0d4e1b1a852225e4b82d807e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9728947490425887, "standard_deviation": null, "minimum": 0.9728947490425887, "maximum": 0.9728947490425887, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6157638952089468, "standard_deviation": null, "minimum": 1.6157638952089468, "maximum": 1.6157638952089468, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9865218396602463, "standard_deviation": null, "minimum": 0.9865218396602463, "maximum": 0.9865218396602463, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22701535946128296, "standard_deviation": null, "minimum": 0.22701535946128296, "maximum": 0.22701535946128296, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9855661432701143, "standard_deviation": null, "minimum": 0.9855661432701143, "maximum": 0.9855661432701143, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1466326202100106, "standard_deviation": null, "minimum": 1.1466326202100106, "maximum": 1.1466326202100106, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2703220033944027, "standard_deviation": null, "minimum": 1.2703220033944027, "maximum": 1.2703220033944027, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17749711035378943, "standard_deviation": null, "minimum": 0.17749711035378943, "maximum": 0.17749711035378943, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9732675226736189, "standard_deviation": null, "minimum": 0.9732675226736189, "maximum": 0.9732675226736189, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.564792512769499, "standard_deviation": null, "minimum": 1.564792512769499, "maximum": 1.564792512769499, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.2615099922202047, "standard_deviation": null, "minimum": 0.2615099922202047, "maximum": 0.2615099922202047, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21256451362121964, "standard_deviation": null, "minimum": 0.21256451362121964, "maximum": 0.21256451362121964, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9914239003302865, "standard_deviation": null, "minimum": 0.9914239003302865, "maximum": 0.9914239003302865, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0326428959385767, "standard_deviation": null, "minimum": 1.0326428959385767, "maximum": 1.0326428959385767, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.44980600024937006, "standard_deviation": null, "minimum": 0.44980600024937006, "maximum": 0.44980600024937006, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1236252735244945, "standard_deviation": null, "minimum": 0.1236252735244945, "maximum": 0.1236252735244945, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9853741270173771, "standard_deviation": null, "minimum": 0.9853741270173771, "maximum": 0.9853741270173771, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.123151647577371, "standard_deviation": null, "minimum": 2.123151647577371, "maximum": 2.123151647577371, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1350704142886319, "standard_deviation": null, "minimum": 1.1350704142886319, "maximum": 1.1350704142886319, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22043033524605754, "standard_deviation": null, "minimum": 0.22043033524605754, "maximum": 0.22043033524605754, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9945790616821423, "standard_deviation": null, "minimum": 0.9945790616821423, "maximum": 0.9945790616821423, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6933248017645768, "standard_deviation": null, "minimum": 1.6933248017645768, "maximum": 1.6933248017645768, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9828299868221865, "standard_deviation": null, "minimum": 0.9828299868221865, "maximum": 0.9828299868221865, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18148442518136362, "standard_deviation": null, "minimum": 0.18148442518136362, "maximum": 0.18148442518136362, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9117516917798718, "standard_deviation": null, "minimum": 0.9117516917798718, "maximum": 0.9117516917798718, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.455049789472159, "standard_deviation": null, "minimum": 5.455049789472159, "maximum": 5.455049789472159, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6388974486426948, "standard_deviation": null, "minimum": 0.6388974486426948, "maximum": 0.6388974486426948, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.3115708722247459, "standard_deviation": null, "minimum": 1.3115708722247459, "maximum": 1.3115708722247459, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9818143536217451, "standard_deviation": null, "minimum": 0.9818143536217451, "maximum": 0.9818143536217451, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.810124329465423, "standard_deviation": null, "minimum": 1.810124329465423, "maximum": 1.810124329465423, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9797381214721176, "standard_deviation": null, "minimum": 0.9797381214721176, "maximum": 0.9797381214721176, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2269843871633539, "standard_deviation": null, "minimum": 0.2269843871633539, "maximum": 0.2269843871633539, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.988967499653238, "standard_deviation": null, "minimum": 0.988967499653238, "maximum": 0.988967499653238, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.4533263787645105, "standard_deviation": null, "minimum": 1.4533263787645105, "maximum": 1.4533263787645105, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0231963093140821, "standard_deviation": null, "minimum": 1.0231963093140821, "maximum": 1.0231963093140821, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2029226738195644, "standard_deviation": null, "minimum": 0.2029226738195644, "maximum": 0.2029226738195644, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_260"}, {"case": "librispeech_260-123288-0003", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.242741584777832, "predicted_mos": 3.242741584777832, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "19476a9a94859b858d9e9a5c3d388298db20358c2ee3a42d564c78f57fbf009a", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.6326641999185085}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9676243176720014, "全帯域_10ms_rms_envelope_nrmse": 0.18236825163708267, "全帯域_10ms_envelope_correlation": 0.9794634545608338, "全帯域_10ms_modulation_mae_db": 0.48640058890942167, "全帯域_40ms_log_rms_mae_db": 1.190586679215879, "全帯域_40ms_rms_envelope_nrmse": 0.12594757093815664, "全帯域_40ms_envelope_correlation": 0.9940140214785564, "全帯域_40ms_modulation_mae_db": 0.4509729430568852, "300–1000 Hz_10ms_log_rms_mae_db": 2.199803020986773, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23192253618756173, "300–1000 Hz_10ms_envelope_correlation": 0.9786001791912491, "300–1000 Hz_10ms_modulation_mae_db": 0.7580154460869516, "300–1000 Hz_40ms_log_rms_mae_db": 1.4894352181578665, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.18289156931981201, "300–1000 Hz_40ms_envelope_correlation": 0.9913576696486652, "300–1000 Hz_40ms_modulation_mae_db": 0.7110389878957472, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8655679672096117, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2181578877158494, "1000–3000 Hz_10ms_envelope_correlation": 0.9703632603758876, "1000–3000 Hz_10ms_modulation_mae_db": 0.4072780527933307, "1000–3000 Hz_40ms_log_rms_mae_db": 1.188988222404884, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13109379222697617, "1000–3000 Hz_40ms_envelope_correlation": 0.9925127929670169, "1000–3000 Hz_40ms_modulation_mae_db": 0.5070160385522354, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9841896065369518, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22621672942635743, "3000–6000 Hz_10ms_envelope_correlation": 0.9677945718205075, "3000–6000 Hz_10ms_modulation_mae_db": 0.6604095007524969, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1585840210266962, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14121183273946553, "3000–6000 Hz_40ms_envelope_correlation": 0.9872006901440619, "3000–6000 Hz_40ms_modulation_mae_db": 0.7881792326333992, "mr_stft_db": 5.681439545869306, "band_mae_db": 0.6688960263469722, "rms_gain_error_db": 0.8128921909895845, "output_peak": 0.47081481637282324}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.3533211344365865, "spectral_convergence": 0.5902627526752735, "spectral_convergence_status": "ok", "amplitude_floor": 1.3116896377207775e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.253363768722724, "spectral_convergence": 0.3201559737830013, "spectral_convergence_status": "ok", "amplitude_floor": 8.457684224428572e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.437633734448608, "spectral_convergence": 0.3750372760164316, "spectral_convergence_status": "ok", "amplitude_floor": 6.717993311155654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008437417957183396, "output_power": 0.0009018135510040954, "delta_db": 0.28907193983059637, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011741200681074208, "output_power": 0.0014953039026351803, "delta_db": 1.0501695569656326, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004760668283187788, "output_power": 0.0006215273547084301, "delta_db": 1.1579232611242287, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001833223674641814, "output_power": 0.00022039994926352155, "delta_db": 0.7999603302376235, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.526890589901711e-05, "output_power": 7.423805433175913e-05, "delta_db": 0.5592024846207821, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6955820206525705e-05, "output_power": 3.8316665184825153e-05, "delta_db": 0.15704858530296936, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6688960263469722, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8128921909895845, "pitch": null, "output_rms_dbfs": -24.747478860042413, "output_peak": 0.47081481637282324, "reference_rms_dbfs": -25.560371051032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.681439545869306}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001544797211376954, "log_rms_mae_db": 1.9676243176720014, "rms_envelope_nrmse": 0.18236825163708267, "envelope_correlation": 0.9794634545608338, "rms_gain_error_db": 0.8129009290515371, "reference_mean_power": 0.002781784143506645, "output_mean_power": 0.0033543897333641495}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.487428460245723e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008814675226478596, "output_power": 0.0010767340905911051, "reference_active": true, "delta_db": 0.8690214800680708}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001997406840389717, "output_power": 0.0002173936924852022, "reference_active": true, "delta_db": 0.36780406361674267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0345220429658976e-05, "output_power": 4.2464861789090466e-05, "reference_active": true, "delta_db": 0.2223762230434516}], "active_band_count": 3, "active_band_mae_db": 0.48640058890942167}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013989081102856241, "log_rms_mae_db": 1.190586679215879, "rms_envelope_nrmse": 0.12594757093815664, "envelope_correlation": 0.9940140214785564, "rms_gain_error_db": 0.8128446284386381, "reference_mean_power": 0.002795409730019158, "output_mean_power": 0.003370776327993047}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3091173503801554e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008259246223406256, "output_power": 0.0010072210275432394, "reference_active": true, "delta_db": 0.8618437044197019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.583418527096058e-05, "output_power": 0.00010567793241422703, "reference_active": true, "delta_db": 0.42463852657342416}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329771424834818e-06, "output_power": 3.279221571233626e-06, "reference_active": true, "delta_db": -0.06643659817752949}], "active_band_count": 3, "active_band_mae_db": 0.4509729430568852}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000132208474335363, "log_rms_mae_db": 2.199803020986773, "rms_envelope_nrmse": 0.23192253618756173, "envelope_correlation": 0.9786001791912491, "rms_gain_error_db": 1.1903041292434793, "reference_mean_power": 0.001270855716483403, "output_mean_power": 0.0016715780502003055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.478174051796597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005667765898001897, "output_power": 0.0007174016012325226, "reference_active": true, "delta_db": 1.0235043788059537}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011084740591226797, "output_power": 0.0001329061690347832, "reference_active": true, "delta_db": 0.788196057599672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2190579526022385e-05, "output_power": 1.9949593655869317e-05, "reference_active": true, "delta_db": -0.46234590185522906}], "active_band_count": 3, "active_band_mae_db": 0.7580154460869516}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012422751968092996, "log_rms_mae_db": 1.4894352181578665, "rms_envelope_nrmse": 0.18289156931981201, "envelope_correlation": 0.9913576696486652, "rms_gain_error_db": 1.1902953858524643, "reference_mean_power": 0.0012772403358267607, "output_mean_power": 0.0016799724659766209}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.710312666254066e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005574547781532276, "output_power": 0.0007045629267518456, "reference_active": true, "delta_db": 1.0171014508465532}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.253668134122962e-05, "output_power": 6.198242785467036e-05, "reference_active": true, "delta_db": 0.7180594794870799}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7043530583570796e-06, "output_power": 1.5551201939834066e-06, "reference_active": true, "delta_db": -0.39795603335360813}], "active_band_count": 3, "active_band_mae_db": 0.7110389878957472}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.199705053096058e-05, "log_rms_mae_db": 1.8655679672096117, "rms_envelope_nrmse": 0.2181578877158494, "envelope_correlation": 0.9703632603758876, "rms_gain_error_db": 0.7638423754772048, "reference_mean_power": 0.00020286667816607547, "output_mean_power": 0.00024187721278236504}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3687094204678632e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.826983537989376e-05, "output_power": 9.712790253375385e-05, "reference_active": true, "delta_db": 0.9374958989994366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2413787132715612e-05, "output_power": 2.24085894817776e-05, "reference_active": true, "delta_db": -0.001007225096861989}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388555256663533e-06, "output_power": 6.8192365992741636e-06, "reference_active": true, "delta_db": 0.2833310342836934}], "active_band_count": 3, "active_band_mae_db": 0.4072780527933307}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.313537286332709e-05, "log_rms_mae_db": 1.188988222404884, "rms_envelope_nrmse": 0.13109379222697617, "envelope_correlation": 0.9925127929670169, "rms_gain_error_db": 0.7635743222660327, "reference_mean_power": 0.00020386380860129322, "output_mean_power": 0.00024305108587545948}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.180981609421061e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.558718162537124e-05, "output_power": 9.328133251611069e-05, "reference_active": true, "delta_db": 0.9134658895862628}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0576597086854303e-05, "output_power": 1.0748848453761466e-05, "reference_active": true, "delta_db": 0.07015979567396269}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.38427493679316e-07, "output_power": 4.757567788623346e-07, "reference_active": true, "delta_db": -0.5374224303964807}], "active_band_count": 3, "active_band_mae_db": 0.5070160385522354}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.0553751915640445e-05, "log_rms_mae_db": 1.9841896065369518, "rms_envelope_nrmse": 0.22621672942635743, "envelope_correlation": 0.9677945718205075, "rms_gain_error_db": 0.35444121452829846, "reference_mean_power": 5.25173391881524e-05, "output_mean_power": 5.698320019735774e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.578963632678691e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2240876569039177e-05, "output_power": 2.4876802952073653e-05, "reference_active": true, "delta_db": 0.48642666263698264}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.96335110256818e-06, "output_power": 9.191160767038822e-06, "reference_active": true, "delta_db": 0.6227449851921275}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4165297947437381e-06, "output_power": 1.7315354971359574e-06, "reference_active": true, "delta_db": 0.8720568544283808}], "active_band_count": 3, "active_band_mae_db": 0.6604095007524969}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.394748258818145e-05, "log_rms_mae_db": 1.1585840210266962, "rms_envelope_nrmse": 0.14121183273946553, "envelope_correlation": 0.9872006901440619, "rms_gain_error_db": 0.3496901362042329, "reference_mean_power": 5.256947927681562e-05, "output_mean_power": 5.6977408030885447e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.187575200111859e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.180909445738007e-05, "output_power": 2.4706150217246814e-05, "reference_active": true, "delta_db": 0.5416744420156208}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0063771970727005e-06, "output_power": 4.7597965571056596e-06, "reference_active": true, "delta_db": 0.7483655518670171}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0500724932970203e-07, "output_power": 1.344835152834696e-07, "reference_active": true, "delta_db": 1.0744977040175598}], "active_band_count": 3, "active_band_mae_db": 0.7881792326333992}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8830257309814082, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2ba998ace78e39e3443a4a67fa83fb4bfca4fe63432253bf917c557dc69ceea5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9703632603758876, "standard_deviation": null, "minimum": 0.9703632603758876, "maximum": 0.9703632603758876, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8655679672096117, "standard_deviation": null, "minimum": 1.8655679672096117, "maximum": 1.8655679672096117, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4072780527933307, "standard_deviation": null, "minimum": 0.4072780527933307, "maximum": 0.4072780527933307, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2181578877158494, "standard_deviation": null, "minimum": 0.2181578877158494, "maximum": 0.2181578877158494, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9925127929670169, "standard_deviation": null, "minimum": 0.9925127929670169, "maximum": 0.9925127929670169, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.188988222404884, "standard_deviation": null, "minimum": 1.188988222404884, "maximum": 1.188988222404884, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5070160385522354, "standard_deviation": null, "minimum": 0.5070160385522354, "maximum": 0.5070160385522354, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13109379222697617, "standard_deviation": null, "minimum": 0.13109379222697617, "maximum": 0.13109379222697617, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9677945718205075, "standard_deviation": null, "minimum": 0.9677945718205075, "maximum": 0.9677945718205075, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9841896065369518, "standard_deviation": null, "minimum": 1.9841896065369518, "maximum": 1.9841896065369518, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6604095007524969, "standard_deviation": null, "minimum": 0.6604095007524969, "maximum": 0.6604095007524969, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22621672942635743, "standard_deviation": null, "minimum": 0.22621672942635743, "maximum": 0.22621672942635743, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9872006901440619, "standard_deviation": null, "minimum": 0.9872006901440619, "maximum": 0.9872006901440619, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1585840210266962, "standard_deviation": null, "minimum": 1.1585840210266962, "maximum": 1.1585840210266962, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7881792326333992, "standard_deviation": null, "minimum": 0.7881792326333992, "maximum": 0.7881792326333992, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14121183273946553, "standard_deviation": null, "minimum": 0.14121183273946553, "maximum": 0.14121183273946553, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9786001791912491, "standard_deviation": null, "minimum": 0.9786001791912491, "maximum": 0.9786001791912491, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.199803020986773, "standard_deviation": null, "minimum": 2.199803020986773, "maximum": 2.199803020986773, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7580154460869516, "standard_deviation": null, "minimum": 0.7580154460869516, "maximum": 0.7580154460869516, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23192253618756173, "standard_deviation": null, "minimum": 0.23192253618756173, "maximum": 0.23192253618756173, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9913576696486652, "standard_deviation": null, "minimum": 0.9913576696486652, "maximum": 0.9913576696486652, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4894352181578665, "standard_deviation": null, "minimum": 1.4894352181578665, "maximum": 1.4894352181578665, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7110389878957472, "standard_deviation": null, "minimum": 0.7110389878957472, "maximum": 0.7110389878957472, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18289156931981201, "standard_deviation": null, "minimum": 0.18289156931981201, "maximum": 0.18289156931981201, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6688960263469722, "standard_deviation": null, "minimum": 0.6688960263469722, "maximum": 0.6688960263469722, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.681439545869306, "standard_deviation": null, "minimum": 5.681439545869306, "maximum": 5.681439545869306, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.47081481637282324, "standard_deviation": null, "minimum": 0.47081481637282324, "maximum": 0.47081481637282324, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8128921909895845, "standard_deviation": null, "minimum": 0.8128921909895845, "maximum": 0.8128921909895845, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9794634545608338, "standard_deviation": null, "minimum": 0.9794634545608338, "maximum": 0.9794634545608338, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.9676243176720014, "standard_deviation": null, "minimum": 1.9676243176720014, "maximum": 1.9676243176720014, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.48640058890942167, "standard_deviation": null, "minimum": 0.48640058890942167, "maximum": 0.48640058890942167, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18236825163708267, "standard_deviation": null, "minimum": 0.18236825163708267, "maximum": 0.18236825163708267, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9940140214785564, "standard_deviation": null, "minimum": 0.9940140214785564, "maximum": 0.9940140214785564, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.190586679215879, "standard_deviation": null, "minimum": 1.190586679215879, "maximum": 1.190586679215879, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4509729430568852, "standard_deviation": null, "minimum": 0.4509729430568852, "maximum": 0.4509729430568852, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12594757093815664, "standard_deviation": null, "minimum": 0.12594757093815664, "maximum": 0.12594757093815664, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_260"}, {"case": "librispeech_260-123440-0020", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.3918473720550537, "predicted_mos": 2.3918473720550537, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79920, "input_sample_rate": 16000, "input_seconds": 4.995, "input_float32_sha256": "da6d88df1d7bf1cd1cf43dbe99b8de50a5b025f531ec1d42337bc3f8e93e7c25", "internal_samples": 79920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.4982696999795735}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9688025348215767, "全帯域_10ms_rms_envelope_nrmse": 0.18786330255257722, "全帯域_10ms_envelope_correlation": 0.9870282461720522, "全帯域_10ms_modulation_mae_db": 0.5196825366712031, "全帯域_40ms_log_rms_mae_db": 1.4026687620766636, "全帯域_40ms_rms_envelope_nrmse": 0.16184140507019076, "全帯域_40ms_envelope_correlation": 0.9944656510986095, "全帯域_40ms_modulation_mae_db": 0.7376933111985697, "300–1000 Hz_10ms_log_rms_mae_db": 2.3324529912435032, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21248770601147943, "300–1000 Hz_10ms_envelope_correlation": 0.9830964936994864, "300–1000 Hz_10ms_modulation_mae_db": 0.8164361626446319, "300–1000 Hz_40ms_log_rms_mae_db": 1.800722151285623, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17432379150752192, "300–1000 Hz_40ms_envelope_correlation": 0.9927322616112779, "300–1000 Hz_40ms_modulation_mae_db": 0.8252706318708447, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6080284307072792, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2318803871748435, "1000–3000 Hz_10ms_envelope_correlation": 0.9813171345043871, "1000–3000 Hz_10ms_modulation_mae_db": 0.4849360659018191, "1000–3000 Hz_40ms_log_rms_mae_db": 0.969498798454549, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19006005895083272, "1000–3000 Hz_40ms_envelope_correlation": 0.9915141167014578, "1000–3000 Hz_40ms_modulation_mae_db": 0.7145130179889286, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6264426459783963, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22067127972221484, "3000–6000 Hz_10ms_envelope_correlation": 0.9741481292572616, "3000–6000 Hz_10ms_modulation_mae_db": 0.7414317617029783, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0451151220877424, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16612541880942075, "3000–6000 Hz_40ms_envelope_correlation": 0.9862646972963631, "3000–6000 Hz_40ms_modulation_mae_db": 0.7056592022114575, "mr_stft_db": 5.907885163925648, "band_mae_db": 0.8621408301731056, "rms_gain_error_db": 1.1493644893355526, "output_peak": 0.6142908173570054}, "full_scale": {"sample_count": 79920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79920, "output_samples": 79920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.458554644363497, "spectral_convergence": 0.5084695658154258, "spectral_convergence_status": "ok", "amplitude_floor": 1.5443736044991425e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.482210914314998, "spectral_convergence": 0.2748522825081294, "spectral_convergence_status": "ok", "amplitude_floor": 1.1257392826698756e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.782889933098447, "spectral_convergence": 0.33290654188057545, "spectral_convergence_status": "ok", "amplitude_floor": 1.1201855379193364e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001035002994445706, "output_power": 0.0012040338914252183, "delta_db": 0.6569710543058205, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0032595449236567883, "output_power": 0.004381762411208181, "delta_db": 1.2849185451632614, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007961593328208376, "output_power": 0.0010534509625114848, "delta_db": 1.2161433387467648, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023616410570225164, "output_power": 0.0003110790591236681, "delta_db": 1.1965688633825602, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.38734826098075e-05, "output_power": 6.46877655003906e-05, "delta_db": 0.794470985457328, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.3554110211771778e-05, "output_power": 1.3628505446560276e-05, "delta_db": 0.023772193982899238, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8621408301731056, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1493644893355526, "pitch": null, "output_rms_dbfs": -21.53128539891807, "output_peak": 0.6142908173570054, "reference_rms_dbfs": -22.68064988825362, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.907885163925648}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001944445221026879, "log_rms_mae_db": 1.9688025348215767, "rms_envelope_nrmse": 0.18786330255257722, "envelope_correlation": 0.9870282461720522, "rms_gain_error_db": 1.1493652721122711, "reference_mean_power": 0.005399702914122853, "output_mean_power": 0.0070356851001397895}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 2.8891675057826967e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000788720956533661, "output_power": 0.0009925732582297172, "reference_active": true, "delta_db": 0.9983919001389703}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001646344464707057, "output_power": 0.00018588996139830158, "reference_active": true, "delta_db": 0.5273522940012689}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.1066196116820634e-05, "output_power": 5.1459297175097894e-05, "reference_active": true, "delta_db": 0.03330341587336996}], "active_band_count": 3, "active_band_mae_db": 0.5196825366712031}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00017595749087279735, "log_rms_mae_db": 1.4026687620766636, "rms_envelope_nrmse": 0.16184140507019076, "envelope_correlation": 0.9944656510986095, "rms_gain_error_db": 1.1493680869715766, "reference_mean_power": 0.005432355653378944, "output_mean_power": 0.007078235434308427}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 2.6677500098154174e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007158752148089715, "output_power": 0.0008969606559133525, "reference_active": true, "delta_db": 0.9793606722550938}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.294187442575086e-05, "output_power": 8.430058238989667e-05, "reference_active": true, "delta_db": 0.6285365553993043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7590528739181784e-06, "output_power": 3.270093450752272e-06, "reference_active": true, "delta_db": -0.6051827059413111}], "active_band_count": 3, "active_band_mae_db": 0.7376933111985697}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.0001659932762873133, "log_rms_mae_db": 2.3324529912435032, "rms_envelope_nrmse": 0.21248770601147943, "envelope_correlation": 0.9830964936994864, "rms_gain_error_db": 1.1738970677369804, "reference_mean_power": 0.0017617711849040457, "output_mean_power": 0.002308549596338798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.325463642910379e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006157322295664234, "output_power": 0.0007908854207204428, "reference_active": true, "delta_db": 1.0872168310664374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015692982702212587, "output_power": 0.00020030115158456115, "reference_active": true, "delta_db": 1.0597795013435578}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0206834209715196e-05, "output_power": 3.750322915685476e-05, "reference_active": true, "delta_db": -0.30231215552390067}], "active_band_count": 3, "active_band_mae_db": 0.8164361626446319}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 0.00014319111528307625, "log_rms_mae_db": 1.800722151285623, "rms_envelope_nrmse": 0.17432379150752192, "envelope_correlation": 0.9927322616112779, "rms_gain_error_db": 1.1738970114284466, "reference_mean_power": 0.0017724260962696937, "output_mean_power": 0.0023225113062881426}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.2006657592514347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005739716253322745, "output_power": 0.0007384916595088897, "reference_active": true, "delta_db": 1.0945517148052468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.210701544963183e-05, "output_power": 9.379288285211402e-05, "reference_active": true, "delta_db": 1.1419236446019063}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.009713658025178e-06, "output_power": 3.7947197378252034e-06, "reference_active": true, "delta_db": -0.23933653620538087}], "active_band_count": 3, "active_band_mae_db": 0.8252706318708447}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 0.00012416177010932925, "log_rms_mae_db": 1.6080284307072792, "rms_envelope_nrmse": 0.2318803871748435, "envelope_correlation": 0.9813171345043871, "rms_gain_error_db": 1.1993305550520912, "reference_mean_power": 0.0002459141129048244, "output_mean_power": 0.00032412796975776124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 1.7078032497142614e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.578711086190898e-05, "output_power": 0.00011295014638729827, "reference_active": true, "delta_db": 0.7157972388381919}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5502167493014184e-05, "output_power": 1.7512236294168376e-05, "reference_active": true, "delta_db": 0.5294918366497884}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901592249182956e-06, "output_power": 5.623637106155379e-06, "reference_active": true, "delta_db": -0.2095191222174768}], "active_band_count": 3, "active_band_mae_db": 0.4849360659018191}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 9.745648796239735e-05, "log_rms_mae_db": 0.969498798454549, "rms_envelope_nrmse": 0.19006005895083272, "envelope_correlation": 0.9915141167014578, "rms_gain_error_db": 1.199338317037646, "reference_mean_power": 0.00024740043253105237, "output_mean_power": 0.00032608760142290407}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 1.5717362127850378e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.047075579844295e-05, "output_power": 0.00010587645381064217, "reference_active": true, "delta_db": 0.6829116830805029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.5496928110188975e-06, "output_power": 8.740319625024816e-06, "reference_active": true, "delta_db": 0.6359803364591528}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.070423026829593e-07, "output_power": 3.366470731778753e-07, "reference_active": true, "delta_db": -0.82464703442713}], "active_band_count": 3, "active_band_mae_db": 0.7145130179889286}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 998, "rms_floor": 3.6263821941295626e-05, "log_rms_mae_db": 1.6264426459783963, "rms_envelope_nrmse": 0.22067127972221484, "envelope_correlation": 0.9741481292572616, "rms_gain_error_db": 0.6531053322546297, "reference_mean_power": 2.0515613806993198e-05, "output_mean_power": 2.3844874905374467e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 998, "frequency_resolution_hz": 0.2004008016032064, "power_floor": 4.904304308942738e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.621343716362641e-06, "output_power": 1.791482217622042e-06, "reference_active": true, "delta_db": 0.4333740892525798}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.453801656353405e-07, "output_power": 7.990005404361337e-07, "reference_active": true, "delta_db": 0.30169240953777865}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.409211497093947e-07, "output_power": 1.7098265181522678e-07, "reference_active": true, "delta_db": -1.4892287863185762}], "active_band_count": 3, "active_band_mae_db": 0.7414317617029783}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 992, "rms_floor": 2.5114070553475023e-05, "log_rms_mae_db": 1.0451151220877424, "rms_envelope_nrmse": 0.16612541880942075, "envelope_correlation": 0.9862646972963631, "rms_gain_error_db": 0.6531118061983407, "reference_mean_power": 2.063955290066712e-05, "output_mean_power": 2.3988962518239788e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 992, "frequency_resolution_hz": 0.20161290322580647, "power_floor": 4.623298998188318e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6281370009575077e-06, "output_power": 1.7876340497365818e-06, "reference_active": true, "delta_db": 0.4058767207179057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7397133003103647e-07, "output_power": 3.81988739257185e-07, "reference_active": true, "delta_db": 0.09212251502509186}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7174687173869334e-08, "output_power": 1.1830170349005326e-08, "reference_active": true, "delta_db": -1.6189783708913748}], "active_band_count": 3, "active_band_mae_db": 0.7056592022114575}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9130540932550081, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "6de0e577bd3758999a4067264b2f292d3ad067d362ad07233d5740b77f2684e0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9813171345043871, "standard_deviation": null, "minimum": 0.9813171345043871, "maximum": 0.9813171345043871, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6080284307072792, "standard_deviation": null, "minimum": 1.6080284307072792, "maximum": 1.6080284307072792, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4849360659018191, "standard_deviation": null, "minimum": 0.4849360659018191, "maximum": 0.4849360659018191, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2318803871748435, "standard_deviation": null, "minimum": 0.2318803871748435, "maximum": 0.2318803871748435, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9915141167014578, "standard_deviation": null, "minimum": 0.9915141167014578, "maximum": 0.9915141167014578, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.969498798454549, "standard_deviation": null, "minimum": 0.969498798454549, "maximum": 0.969498798454549, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7145130179889286, "standard_deviation": null, "minimum": 0.7145130179889286, "maximum": 0.7145130179889286, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19006005895083272, "standard_deviation": null, "minimum": 0.19006005895083272, "maximum": 0.19006005895083272, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9741481292572616, "standard_deviation": null, "minimum": 0.9741481292572616, "maximum": 0.9741481292572616, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6264426459783963, "standard_deviation": null, "minimum": 1.6264426459783963, "maximum": 1.6264426459783963, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7414317617029783, "standard_deviation": null, "minimum": 0.7414317617029783, "maximum": 0.7414317617029783, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22067127972221484, "standard_deviation": null, "minimum": 0.22067127972221484, "maximum": 0.22067127972221484, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9862646972963631, "standard_deviation": null, "minimum": 0.9862646972963631, "maximum": 0.9862646972963631, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0451151220877424, "standard_deviation": null, "minimum": 1.0451151220877424, "maximum": 1.0451151220877424, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7056592022114575, "standard_deviation": null, "minimum": 0.7056592022114575, "maximum": 0.7056592022114575, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16612541880942075, "standard_deviation": null, "minimum": 0.16612541880942075, "maximum": 0.16612541880942075, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9830964936994864, "standard_deviation": null, "minimum": 0.9830964936994864, "maximum": 0.9830964936994864, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3324529912435032, "standard_deviation": null, "minimum": 2.3324529912435032, "maximum": 2.3324529912435032, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8164361626446319, "standard_deviation": null, "minimum": 0.8164361626446319, "maximum": 0.8164361626446319, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21248770601147943, "standard_deviation": null, "minimum": 0.21248770601147943, "maximum": 0.21248770601147943, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9927322616112779, "standard_deviation": null, "minimum": 0.9927322616112779, "maximum": 0.9927322616112779, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.800722151285623, "standard_deviation": null, "minimum": 1.800722151285623, "maximum": 1.800722151285623, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8252706318708447, "standard_deviation": null, "minimum": 0.8252706318708447, "maximum": 0.8252706318708447, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17432379150752192, "standard_deviation": null, "minimum": 0.17432379150752192, "maximum": 0.17432379150752192, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8621408301731056, "standard_deviation": null, "minimum": 0.8621408301731056, "maximum": 0.8621408301731056, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.907885163925648, "standard_deviation": null, "minimum": 5.907885163925648, "maximum": 5.907885163925648, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6142908173570054, "standard_deviation": null, "minimum": 0.6142908173570054, "maximum": 0.6142908173570054, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1493644893355526, "standard_deviation": null, "minimum": 1.1493644893355526, "maximum": 1.1493644893355526, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9870282461720522, "standard_deviation": null, "minimum": 0.9870282461720522, "maximum": 0.9870282461720522, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.9688025348215767, "standard_deviation": null, "minimum": 1.9688025348215767, "maximum": 1.9688025348215767, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5196825366712031, "standard_deviation": null, "minimum": 0.5196825366712031, "maximum": 0.5196825366712031, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18786330255257722, "standard_deviation": null, "minimum": 0.18786330255257722, "maximum": 0.18786330255257722, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9944656510986095, "standard_deviation": null, "minimum": 0.9944656510986095, "maximum": 0.9944656510986095, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.4026687620766636, "standard_deviation": null, "minimum": 1.4026687620766636, "maximum": 1.4026687620766636, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7376933111985697, "standard_deviation": null, "minimum": 0.7376933111985697, "maximum": 0.7376933111985697, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16184140507019076, "standard_deviation": null, "minimum": 0.16184140507019076, "maximum": 0.16184140507019076, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_260"}, {"case": "librispeech_2830-3980-0000", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.8135786056518555, "predicted_mos": 2.8135786056518555, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59680, "input_sample_rate": 16000, "input_seconds": 3.73, "input_float32_sha256": "938bbdbfde9d5661e478ead4194ea868721382d8469b9aa659f77f1572f56299", "internal_samples": 59680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.662320599891245}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.895233474358373, "全帯域_10ms_rms_envelope_nrmse": 0.2582627771576524, "全帯域_10ms_envelope_correlation": 0.9746230263119162, "全帯域_10ms_modulation_mae_db": 1.593155096980653, "全帯域_40ms_log_rms_mae_db": 2.4514533203528783, "全帯域_40ms_rms_envelope_nrmse": 0.2239873182130422, "全帯域_40ms_envelope_correlation": 0.9876913833214257, "全帯域_40ms_modulation_mae_db": 1.5465406873741447, "300–1000 Hz_10ms_log_rms_mae_db": 2.018368680309548, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2940968220152114, "300–1000 Hz_10ms_envelope_correlation": 0.9802342837456478, "300–1000 Hz_10ms_modulation_mae_db": 0.9353915968370021, "300–1000 Hz_40ms_log_rms_mae_db": 1.5894065976042466, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2664450232481203, "300–1000 Hz_40ms_envelope_correlation": 0.9895269742387528, "300–1000 Hz_40ms_modulation_mae_db": 0.8869238786503368, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4504131677714556, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17441398787230664, "1000–3000 Hz_10ms_envelope_correlation": 0.9831552080531504, "1000–3000 Hz_10ms_modulation_mae_db": 0.46487939628554026, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8224267932756476, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09726856522599456, "1000–3000 Hz_40ms_envelope_correlation": 0.9969984103053623, "1000–3000 Hz_40ms_modulation_mae_db": 0.6535405195241375, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5869177721103938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20284682842686957, "3000–6000 Hz_10ms_envelope_correlation": 0.9714022736431982, "3000–6000 Hz_10ms_modulation_mae_db": 0.5255778539992898, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8401505373596635, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09781168933141106, "3000–6000 Hz_40ms_envelope_correlation": 0.9937106927912145, "3000–6000 Hz_40ms_modulation_mae_db": 0.6943589006332654, "mr_stft_db": 5.874074039061959, "band_mae_db": 0.8007015341078209, "rms_gain_error_db": 1.4790928732439026, "output_peak": 1.2619740184987454}, "full_scale": {"sample_count": 59680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 57, "at_or_above_full_scale_fraction": 0.0009550938337801609, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0009718498659517426, "longest_full_scale_run_samples": 4, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59680, "output_samples": 59680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.801380433431309, "spectral_convergence": 0.6832860413541493, "spectral_convergence_status": "ok", "amplitude_floor": 3.189740694195829e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.254164136532527, "spectral_convergence": 0.32979817223131497, "spectral_convergence_status": "ok", "amplitude_floor": 2.205927306089595e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.566677547222043, "spectral_convergence": 0.349671456907061, "spectral_convergence_status": "ok", "amplitude_floor": 1.9576049396551346e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004284845572396228, "output_power": 0.004172562362161586, "delta_db": -0.1153233812313676, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.012674374150346739, "output_power": 0.019577328966320896, "delta_db": 1.888269150751696, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0031087438744962816, "output_power": 0.004698115156127125, "delta_db": 1.7933871508188461, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005542642121172655, "output_power": 0.0006354969660538347, "delta_db": 0.5939664366272325, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001593466145796079, "output_power": 0.00017013137381256317, "delta_db": 0.28441567907467574, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.162629787041768e-05, "output_power": 8.894783848380722e-05, "delta_db": -0.12884740614310752, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8007015341078209, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.4790928732439026, "pitch": null, "output_rms_dbfs": -15.325016632633986, "output_peak": 1.2619740184987454, "reference_rms_dbfs": -16.80410950587789, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.874074039061959}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00043785897255075944, "log_rms_mae_db": 2.895233474358373, "rms_envelope_nrmse": 0.2582627771576524, "envelope_correlation": 0.9746230263119162, "rms_gain_error_db": 1.4790934275966214, "reference_mean_power": 0.0209012125553907, "output_mean_power": 0.02938196415682878}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.005551294872497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00444581511703579, "output_power": 0.00703125008292103, "reference_active": true, "delta_db": 1.9908114667631884}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00039454721092167534, "output_power": 0.0005224307175700611, "reference_active": true, "delta_db": 1.2192972673118911}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.0001084016816849342, "output_power": 0.00015558641549669812, "reference_active": true, "delta_db": 1.569356556866879}], "active_band_count": 3, "active_band_mae_db": 1.593155096980653}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0004107294113879301, "log_rms_mae_db": 2.4514533203528783, "rms_envelope_nrmse": 0.2239873182130422, "envelope_correlation": 0.9876913833214257, "rms_gain_error_db": 1.4791045417977136, "reference_mean_power": 0.021070817398047652, "output_mean_power": 0.02962046265374122}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.012296407388435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00391054354874771, "output_power": 0.006271574668453259, "reference_active": true, "delta_db": 2.051394706255325}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019349823972573515, "output_power": 0.00024600921000426923, "reference_active": true, "delta_db": 1.042743478188892}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9805555260958754e-06, "output_power": 8.53669793879977e-06, "reference_active": true, "delta_db": 1.5454838776782172}], "active_band_count": 3, "active_band_mae_db": 1.5465406873741447}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.0003730035430310483, "log_rms_mae_db": 2.018368680309548, "rms_envelope_nrmse": 0.2940968220152114, "envelope_correlation": 0.9802342837456478, "rms_gain_error_db": 1.8227558348912627, "reference_mean_power": 0.007423998211444209, "output_mean_power": 0.011295707608817516}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.725628531347034e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002177971361103236, "output_power": 0.002942270238637368, "reference_active": true, "delta_db": 1.306303940909391}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00032785256002961345, "output_power": 0.00039430201405370733, "reference_active": true, "delta_db": 0.801504162151364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.051887548388713e-05, "output_power": 9.456585852698132e-05, "reference_active": true, "delta_db": 0.6983666874502517}], "active_band_count": 3, "active_band_mae_db": 0.9353915968370021}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00035212990311193784, "log_rms_mae_db": 1.5894065976042466, "rms_envelope_nrmse": 0.2664450232481203, "envelope_correlation": 0.9895269742387528, "rms_gain_error_db": 1.8227616644808633, "reference_mean_power": 0.007484242812710004, "output_mean_power": 0.011387385820539025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.215922155621447e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018646593945503163, "output_power": 0.0024759836834146393, "reference_active": true, "delta_db": 1.231482647991323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015694821766981044, "output_power": 0.0001875605934713217, "reference_active": true, "delta_db": 0.7738521010192655}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505413793795639e-06, "output_power": 7.565168662644796e-06, "reference_active": true, "delta_db": 0.6554368869404217}], "active_band_count": 3, "active_band_mae_db": 0.8869238786503368}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 0.00013449969207959865, "log_rms_mae_db": 1.4504131677714556, "rms_envelope_nrmse": 0.17441398787230664, "envelope_correlation": 0.9831552080531504, "rms_gain_error_db": 0.6067381120114064, "reference_mean_power": 0.0006066736643199691, "output_mean_power": 0.0006976361132741947}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 3.4525665594490304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019778914796660728, "output_power": 0.0002271105551956174, "reference_active": true, "delta_db": 0.6003485935001128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.140914614473558e-05, "output_power": 3.73961767865728e-05, "reference_active": true, "delta_db": 0.7577107389969423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0263986880205243e-05, "output_power": 1.0350801351870104e-05, "reference_active": true, "delta_db": 0.03657885635956564}], "active_band_count": 3, "active_band_mae_db": 0.46487939628554026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.00012636239720129053, "log_rms_mae_db": 0.8224267932756476, "rms_envelope_nrmse": 0.09726856522599456, "envelope_correlation": 0.9969984103053623, "rms_gain_error_db": 0.6067367509335447, "reference_mean_power": 0.0006115927691385325, "output_mean_power": 0.0007032925504169756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 3.048532457224497e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001773914005108231, "output_power": 0.00020485653843915723, "reference_active": true, "delta_db": 0.6251726745612349}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4831454807134908e-05, "output_power": 1.7852032757207023e-05, "reference_active": true, "delta_db": 0.8050392230573469}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.85797297398325e-07, "output_power": 5.489035764507329e-07, "reference_active": true, "delta_db": 0.5304096609538307}], "active_band_count": 3, "active_band_mae_db": 0.6535405195241375}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 745, "rms_floor": 5.5319997387638595e-05, "log_rms_mae_db": 1.5869177721103938, "rms_envelope_nrmse": 0.20284682842686957, "envelope_correlation": 0.9714022736431982, "rms_gain_error_db": -0.2502352322564971, "reference_mean_power": 9.493685130389184e-05, "output_mean_power": 8.962131263303663e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 745, "frequency_resolution_hz": 0.2684563758389262, "power_floor": 9.144674032958852e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.902158206464317e-05, "output_power": 2.4325579140453044e-05, "reference_active": true, "delta_db": -0.7665789491474018}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.982855070886748e-06, "output_power": 8.53346155924437e-06, "reference_active": true, "delta_db": -0.681295296996965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.211971373682015e-06, "output_power": 3.3087015707106125e-06, "reference_active": true, "delta_db": 0.12885931585350271}], "active_band_count": 3, "active_band_mae_db": 0.5255778539992898}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 4.1914797815135623e-05, "log_rms_mae_db": 0.8401505373596635, "rms_envelope_nrmse": 0.09781168933141106, "envelope_correlation": 0.9937106927912145, "rms_gain_error_db": -0.25022745885451103, "reference_mean_power": 9.570694886911743e-05, "output_mean_power": 9.034845395175917e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 8.518176746918335e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6494345211181707e-05, "output_power": 2.2770747442529847e-05, "reference_active": true, "delta_db": -0.6577590427793983}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.515847029872695e-06, "output_power": 3.984546748758052e-06, "reference_active": true, "delta_db": -0.5436029562148147}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6527459125158204e-07, "output_power": 2.1653304526035734e-07, "reference_active": true, "delta_db": -0.8817147029055833}], "active_band_count": 3, "active_band_mae_db": 0.6943589006332654}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9314442356349691, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0c887eebe97e1ba307bd5d79bd3cc29e65d687290849e599ea651063ad2d15e2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9831552080531504, "standard_deviation": null, "minimum": 0.9831552080531504, "maximum": 0.9831552080531504, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4504131677714556, "standard_deviation": null, "minimum": 1.4504131677714556, "maximum": 1.4504131677714556, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.46487939628554026, "standard_deviation": null, "minimum": 0.46487939628554026, "maximum": 0.46487939628554026, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17441398787230664, "standard_deviation": null, "minimum": 0.17441398787230664, "maximum": 0.17441398787230664, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969984103053623, "standard_deviation": null, "minimum": 0.9969984103053623, "maximum": 0.9969984103053623, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8224267932756476, "standard_deviation": null, "minimum": 0.8224267932756476, "maximum": 0.8224267932756476, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6535405195241375, "standard_deviation": null, "minimum": 0.6535405195241375, "maximum": 0.6535405195241375, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09726856522599456, "standard_deviation": null, "minimum": 0.09726856522599456, "maximum": 0.09726856522599456, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9714022736431982, "standard_deviation": null, "minimum": 0.9714022736431982, "maximum": 0.9714022736431982, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5869177721103938, "standard_deviation": null, "minimum": 1.5869177721103938, "maximum": 1.5869177721103938, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5255778539992898, "standard_deviation": null, "minimum": 0.5255778539992898, "maximum": 0.5255778539992898, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20284682842686957, "standard_deviation": null, "minimum": 0.20284682842686957, "maximum": 0.20284682842686957, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937106927912145, "standard_deviation": null, "minimum": 0.9937106927912145, "maximum": 0.9937106927912145, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8401505373596635, "standard_deviation": null, "minimum": 0.8401505373596635, "maximum": 0.8401505373596635, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6943589006332654, "standard_deviation": null, "minimum": 0.6943589006332654, "maximum": 0.6943589006332654, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09781168933141106, "standard_deviation": null, "minimum": 0.09781168933141106, "maximum": 0.09781168933141106, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9802342837456478, "standard_deviation": null, "minimum": 0.9802342837456478, "maximum": 0.9802342837456478, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.018368680309548, "standard_deviation": null, "minimum": 2.018368680309548, "maximum": 2.018368680309548, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9353915968370021, "standard_deviation": null, "minimum": 0.9353915968370021, "maximum": 0.9353915968370021, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2940968220152114, "standard_deviation": null, "minimum": 0.2940968220152114, "maximum": 0.2940968220152114, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9895269742387528, "standard_deviation": null, "minimum": 0.9895269742387528, "maximum": 0.9895269742387528, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5894065976042466, "standard_deviation": null, "minimum": 1.5894065976042466, "maximum": 1.5894065976042466, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8869238786503368, "standard_deviation": null, "minimum": 0.8869238786503368, "maximum": 0.8869238786503368, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2664450232481203, "standard_deviation": null, "minimum": 0.2664450232481203, "maximum": 0.2664450232481203, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8007015341078209, "standard_deviation": null, "minimum": 0.8007015341078209, "maximum": 0.8007015341078209, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.874074039061959, "standard_deviation": null, "minimum": 5.874074039061959, "maximum": 5.874074039061959, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.2619740184987454, "standard_deviation": null, "minimum": 1.2619740184987454, "maximum": 1.2619740184987454, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.4790928732439026, "standard_deviation": null, "minimum": 1.4790928732439026, "maximum": 1.4790928732439026, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9746230263119162, "standard_deviation": null, "minimum": 0.9746230263119162, "maximum": 0.9746230263119162, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.895233474358373, "standard_deviation": null, "minimum": 2.895233474358373, "maximum": 2.895233474358373, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.593155096980653, "standard_deviation": null, "minimum": 1.593155096980653, "maximum": 1.593155096980653, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2582627771576524, "standard_deviation": null, "minimum": 0.2582627771576524, "maximum": 0.2582627771576524, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9876913833214257, "standard_deviation": null, "minimum": 0.9876913833214257, "maximum": 0.9876913833214257, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.4514533203528783, "standard_deviation": null, "minimum": 2.4514533203528783, "maximum": 2.4514533203528783, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.5465406873741447, "standard_deviation": null, "minimum": 1.5465406873741447, "maximum": 1.5465406873741447, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2239873182130422, "standard_deviation": null, "minimum": 0.2239873182130422, "maximum": 0.2239873182130422, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2830"}, {"case": "librispeech_2830-3980-0038", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.4062492847442627, "predicted_mos": 3.4062492847442627, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88640, "input_sample_rate": 16000, "input_seconds": 5.54, "input_float32_sha256": "7a35dc3bbcbb7ad0675437684a08c3b9ddb282e94dc4ead8528f66e0dae21c4b", "internal_samples": 88640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.870874300133437}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.757855535778793, "全帯域_10ms_rms_envelope_nrmse": 0.23236053335156415, "全帯域_10ms_envelope_correlation": 0.971895420951942, "全帯域_10ms_modulation_mae_db": 0.7379400079956459, "全帯域_40ms_log_rms_mae_db": 3.157683743215217, "全帯域_40ms_rms_envelope_nrmse": 0.19648178239482247, "全帯域_40ms_envelope_correlation": 0.9850265032427773, "全帯域_40ms_modulation_mae_db": 0.861308124913227, "300–1000 Hz_10ms_log_rms_mae_db": 6.481436872081686, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18863679586870719, "300–1000 Hz_10ms_envelope_correlation": 0.9863776346278309, "300–1000 Hz_10ms_modulation_mae_db": 0.7556935501369794, "300–1000 Hz_40ms_log_rms_mae_db": 5.934844309932365, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14449083271727925, "300–1000 Hz_40ms_envelope_correlation": 0.9950156456637725, "300–1000 Hz_40ms_modulation_mae_db": 0.9137968347369713, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5716444890303907, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20548738827261762, "1000–3000 Hz_10ms_envelope_correlation": 0.9802060720407418, "1000–3000 Hz_10ms_modulation_mae_db": 0.8314267166542159, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9739657920306699, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12546920440094605, "1000–3000 Hz_40ms_envelope_correlation": 0.9960577558811475, "1000–3000 Hz_40ms_modulation_mae_db": 1.0528846350075691, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5489356633675069, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21440942184291675, "3000–6000 Hz_10ms_envelope_correlation": 0.9717463452162329, "3000–6000 Hz_10ms_modulation_mae_db": 0.8088234497220084, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8556861643504589, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1254452149806085, "3000–6000 Hz_40ms_envelope_correlation": 0.9910884766136152, "3000–6000 Hz_40ms_modulation_mae_db": 0.7712559411176588, "mr_stft_db": 6.196275502518749, "band_mae_db": 0.8246883992923619, "rms_gain_error_db": 1.035814668314929, "output_peak": 0.7432229697275571}, "full_scale": {"sample_count": 88640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88640, "output_samples": 88640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.698008107202456, "spectral_convergence": 0.6365482909260666, "spectral_convergence_status": "ok", "amplitude_floor": 1.5935254482600421e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.816094280954466, "spectral_convergence": 0.37001369049108357, "spectral_convergence_status": "ok", "amplitude_floor": 9.146897175620752e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.074724119399327, "spectral_convergence": 0.3889629055107131, "spectral_convergence_status": "ok", "amplitude_floor": 7.905477970735852e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007242342814613898, "output_power": 0.0005903695519527165, "delta_db": -0.8875512734825043, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001748392652036385, "output_power": 0.00271877193768136, "delta_db": 1.917338060071764, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009820880914630636, "output_power": 0.0011434281292238607, "delta_db": 0.6605842705660583, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002665715938534463, "output_power": 0.00031048186104400035, "delta_db": 0.6622236425156446, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.016487980563797e-05, "output_power": 6.800014771187227e-05, "delta_db": 0.531668026676121, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00015485832573279259, "output_power": 0.00016550498215873557, "delta_db": 0.2887651224420798, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8246883992923619, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.035814668314929, "pitch": null, "output_rms_dbfs": -23.013291248174884, "output_peak": 0.7432229697275571, "reference_rms_dbfs": -24.04910591648981, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.196275502518749}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00022148173042756648, "log_rms_mae_db": 3.757855535778793, "rms_envelope_nrmse": 0.23236053335156415, "envelope_correlation": 0.971895420951942, "rms_gain_error_db": 1.035851612684257, "reference_mean_power": 0.0039398267431541775, "output_mean_power": 0.00500106253429287}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.3094240146918987e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011389510990882513, "output_power": 0.001450640110591615, "reference_active": true, "delta_db": 1.0505460362355348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017715594495014008, "output_power": 0.00021288312034582993, "reference_active": true, "delta_db": 0.7978549643711326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5198627564616767e-05, "output_power": 2.7410615651719214e-05, "reference_active": true, "delta_db": 0.3654190233802701}], "active_band_count": 3, "active_band_mae_db": 0.7379400079956459}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.00020322184022683357, "log_rms_mae_db": 3.157683743215217, "rms_envelope_nrmse": 0.19648178239482247, "envelope_correlation": 0.9850265032427773, "rms_gain_error_db": 1.0367301624414265, "reference_mean_power": 0.0039603691212238695, "output_mean_power": 0.005028155289031176}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.1092095592585075e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010485738336262066, "output_power": 0.001336589983506931, "reference_active": true, "delta_db": 1.0539918614875277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.703453851742181e-05, "output_power": 0.00010703125653102351, "reference_active": true, "delta_db": 0.8981899310448367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.852514808001562e-06, "output_power": 2.142574571930413e-06, "reference_active": true, "delta_db": 0.6317425822073164}], "active_band_count": 3, "active_band_mae_db": 0.861308124913227}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00017913726163896074, "log_rms_mae_db": 6.481436872081686, "rms_envelope_nrmse": 0.18863679586870719, "envelope_correlation": 0.9863776346278309, "rms_gain_error_db": 0.8829750563861677, "reference_mean_power": 0.001348381907516537, "output_mean_power": 0.001652381873663375}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 1.724352265721046e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008753679492081999, "output_power": 0.0009794605328106462, "reference_active": true, "delta_db": 0.4879629958835096}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012951638097036193, "output_power": 0.00014941918571174625, "reference_active": true, "delta_db": 0.6208166472884589}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6972503859652748e-05, "output_power": 1.2999175746721364e-05, "reference_active": true, "delta_db": -1.1583010072389694}], "active_band_count": 3, "active_band_mae_db": 0.7556935501369794}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 0.0001658599445391014, "log_rms_mae_db": 5.934844309932365, "rms_envelope_nrmse": 0.14449083271727925, "envelope_correlation": 0.9950156456637725, "rms_gain_error_db": 0.8828284206490911, "reference_mean_power": 0.001355729112927364, "output_mean_power": 0.0016613294518365979}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 1.6447728018068126e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008253872294632201, "output_power": 0.0009253197234925983, "reference_active": true, "delta_db": 0.4963407428251016}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.001034181994634e-05, "output_power": 7.075574908468783e-05, "reference_active": true, "delta_db": 0.7153563263991625}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.081003459949387e-06, "output_power": 1.463199212693766e-06, "reference_active": true, "delta_db": -1.52969343498665}], "active_band_count": 3, "active_band_mae_db": 0.9137968347369713}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 0.00010266849739942118, "log_rms_mae_db": 1.5716444890303907, "rms_envelope_nrmse": 0.20548738827261762, "envelope_correlation": 0.9802060720407418, "rms_gain_error_db": 0.8071386710415314, "reference_mean_power": 0.0002406311059478081, "output_mean_power": 0.00028977814914829493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 2.4743465614681157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013528556332001226, "output_power": 0.00016699144523466412, "reference_active": true, "delta_db": 0.9144276898959253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.422298628989197e-05, "output_power": 1.929378387606415e-05, "reference_active": true, "delta_db": 1.3242661795703834}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7979674926628947e-06, "output_power": 2.6380561580102556e-06, "reference_active": true, "delta_db": -0.2555862804963386}], "active_band_count": 3, "active_band_mae_db": 0.8314267166542159}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 8.271272308600344e-05, "log_rms_mae_db": 0.9739657920306699, "rms_envelope_nrmse": 0.12546920440094605, "envelope_correlation": 0.9960577558811475, "rms_gain_error_db": 0.8071401743752653, "reference_mean_power": 0.00024194203098044546, "output_mean_power": 0.00029135692133954344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 2.355449255956094e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012689841334701795, "output_power": 0.000159017838256776, "reference_active": true, "delta_db": 0.9798965321259203}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.520296796235256e-06, "output_power": 8.949105305064092e-06, "reference_active": true, "delta_db": 1.3751225374520226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.927693257279988e-07, "output_power": 1.6020440183307768e-07, "reference_active": true, "delta_db": -0.8036348354447647}], "active_band_count": 3, "active_band_mae_db": 1.0528846350075691}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1107, "rms_floor": 5.887039168463902e-05, "log_rms_mae_db": 1.5489356633675069, "rms_envelope_nrmse": 0.21440942184291675, "envelope_correlation": 0.9717463452162329, "rms_gain_error_db": 0.43425080314034326, "reference_mean_power": 0.00013131711251995087, "output_mean_power": 0.0001451263942001883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1107, "frequency_resolution_hz": 0.18066847335140018, "power_floor": 6.872487837717039e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.115138542486786e-05, "output_power": 3.4922610352494784e-05, "reference_active": true, "delta_db": 0.49629332146954397}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.223707279035506e-06, "output_power": 3.749884654115879e-06, "reference_active": true, "delta_db": 0.656623092403184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5460080528529549e-06, "output_power": 2.0728465128481473e-06, "reference_active": true, "delta_db": 1.2735539352932972}], "active_band_count": 3, "active_band_mae_db": 0.8088234497220084}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1101, "rms_floor": 5.4692297362327305e-05, "log_rms_mae_db": 0.8556861643504589, "rms_envelope_nrmse": 0.1254452149806085, "envelope_correlation": 0.9910884766136152, "rms_gain_error_db": 0.43425186876446614, "reference_mean_power": 0.00013203257062264164, "output_mean_power": 0.00014591712553953503}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1101, "frequency_resolution_hz": 0.18165304268846502, "power_floor": 6.492145466124568e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.913804296383342e-05, "output_power": 3.337191370898454e-05, "reference_active": true, "delta_db": 0.5892073261162558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5333654526802403e-06, "output_power": 1.603013490444047e-06, "reference_active": true, "delta_db": 0.19291503044755393}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.807699107498543e-08, "output_power": 1.2532191320373436e-07, "reference_active": true, "delta_db": 1.5316454667891666}], "active_band_count": 3, "active_band_mae_db": 0.7712559411176588}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8036279281368577, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c885b2548705eadc1adcf4f2e0bb98db318280ea527a432fbca52ebf5cfcb21e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9802060720407418, "standard_deviation": null, "minimum": 0.9802060720407418, "maximum": 0.9802060720407418, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5716444890303907, "standard_deviation": null, "minimum": 1.5716444890303907, "maximum": 1.5716444890303907, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8314267166542159, "standard_deviation": null, "minimum": 0.8314267166542159, "maximum": 0.8314267166542159, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20548738827261762, "standard_deviation": null, "minimum": 0.20548738827261762, "maximum": 0.20548738827261762, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960577558811475, "standard_deviation": null, "minimum": 0.9960577558811475, "maximum": 0.9960577558811475, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9739657920306699, "standard_deviation": null, "minimum": 0.9739657920306699, "maximum": 0.9739657920306699, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0528846350075691, "standard_deviation": null, "minimum": 1.0528846350075691, "maximum": 1.0528846350075691, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12546920440094605, "standard_deviation": null, "minimum": 0.12546920440094605, "maximum": 0.12546920440094605, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9717463452162329, "standard_deviation": null, "minimum": 0.9717463452162329, "maximum": 0.9717463452162329, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5489356633675069, "standard_deviation": null, "minimum": 1.5489356633675069, "maximum": 1.5489356633675069, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8088234497220084, "standard_deviation": null, "minimum": 0.8088234497220084, "maximum": 0.8088234497220084, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21440942184291675, "standard_deviation": null, "minimum": 0.21440942184291675, "maximum": 0.21440942184291675, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9910884766136152, "standard_deviation": null, "minimum": 0.9910884766136152, "maximum": 0.9910884766136152, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8556861643504589, "standard_deviation": null, "minimum": 0.8556861643504589, "maximum": 0.8556861643504589, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7712559411176588, "standard_deviation": null, "minimum": 0.7712559411176588, "maximum": 0.7712559411176588, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1254452149806085, "standard_deviation": null, "minimum": 0.1254452149806085, "maximum": 0.1254452149806085, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9863776346278309, "standard_deviation": null, "minimum": 0.9863776346278309, "maximum": 0.9863776346278309, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 6.481436872081686, "standard_deviation": null, "minimum": 6.481436872081686, "maximum": 6.481436872081686, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7556935501369794, "standard_deviation": null, "minimum": 0.7556935501369794, "maximum": 0.7556935501369794, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18863679586870719, "standard_deviation": null, "minimum": 0.18863679586870719, "maximum": 0.18863679586870719, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9950156456637725, "standard_deviation": null, "minimum": 0.9950156456637725, "maximum": 0.9950156456637725, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 5.934844309932365, "standard_deviation": null, "minimum": 5.934844309932365, "maximum": 5.934844309932365, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9137968347369713, "standard_deviation": null, "minimum": 0.9137968347369713, "maximum": 0.9137968347369713, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14449083271727925, "standard_deviation": null, "minimum": 0.14449083271727925, "maximum": 0.14449083271727925, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8246883992923619, "standard_deviation": null, "minimum": 0.8246883992923619, "maximum": 0.8246883992923619, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.196275502518749, "standard_deviation": null, "minimum": 6.196275502518749, "maximum": 6.196275502518749, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7432229697275571, "standard_deviation": null, "minimum": 0.7432229697275571, "maximum": 0.7432229697275571, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.035814668314929, "standard_deviation": null, "minimum": 1.035814668314929, "maximum": 1.035814668314929, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.971895420951942, "standard_deviation": null, "minimum": 0.971895420951942, "maximum": 0.971895420951942, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.757855535778793, "standard_deviation": null, "minimum": 3.757855535778793, "maximum": 3.757855535778793, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7379400079956459, "standard_deviation": null, "minimum": 0.7379400079956459, "maximum": 0.7379400079956459, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.23236053335156415, "standard_deviation": null, "minimum": 0.23236053335156415, "maximum": 0.23236053335156415, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9850265032427773, "standard_deviation": null, "minimum": 0.9850265032427773, "maximum": 0.9850265032427773, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 3.157683743215217, "standard_deviation": null, "minimum": 3.157683743215217, "maximum": 3.157683743215217, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.861308124913227, "standard_deviation": null, "minimum": 0.861308124913227, "maximum": 0.861308124913227, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.19648178239482247, "standard_deviation": null, "minimum": 0.19648178239482247, "maximum": 0.19648178239482247, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2830"}, {"case": "librispeech_2830-3980-0042", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.4884068965911865, "predicted_mos": 3.4884068965911865, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48320, "input_sample_rate": 16000, "input_seconds": 3.02, "input_float32_sha256": "c8c6fb3834c1f9543548f7505740a12421c7b4da1ac5a6518d9ba1cbd47c614b", "internal_samples": 48320, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.736845800187439}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 4.234667456091732, "全帯域_10ms_rms_envelope_nrmse": 0.19118935077048477, "全帯域_10ms_envelope_correlation": 0.9846386833492439, "全帯域_10ms_modulation_mae_db": 0.7530794129428648, "全帯域_40ms_log_rms_mae_db": 3.655873283147236, "全帯域_40ms_rms_envelope_nrmse": 0.15898785317391945, "全帯域_40ms_envelope_correlation": 0.993945251754445, "全帯域_40ms_modulation_mae_db": 0.7865145189816131, "300–1000 Hz_10ms_log_rms_mae_db": 6.571539612404468, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16613204359657344, "300–1000 Hz_10ms_envelope_correlation": 0.9907641038881662, "300–1000 Hz_10ms_modulation_mae_db": 0.5749703805820353, "300–1000 Hz_40ms_log_rms_mae_db": 6.106051259016174, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1360198494111946, "300–1000 Hz_40ms_envelope_correlation": 0.996095170323478, "300–1000 Hz_40ms_modulation_mae_db": 0.5313285158279987, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6020884707769731, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23844087174655845, "1000–3000 Hz_10ms_envelope_correlation": 0.9772774102544709, "1000–3000 Hz_10ms_modulation_mae_db": 1.4926569095448876, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0581616921442307, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18601314227242333, "1000–3000 Hz_40ms_envelope_correlation": 0.9900668450905462, "1000–3000 Hz_40ms_modulation_mae_db": 1.373121824610389, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6473426572179457, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22958670023364003, "3000–6000 Hz_10ms_envelope_correlation": 0.9640753652664081, "3000–6000 Hz_10ms_modulation_mae_db": 0.9002500993578652, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0195966757134831, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15186609906183482, "3000–6000 Hz_40ms_envelope_correlation": 0.9833056804066591, "3000–6000 Hz_40ms_modulation_mae_db": 0.5532053616053081, "mr_stft_db": 6.201203669617407, "band_mae_db": 0.8062463505777774, "rms_gain_error_db": 0.8811734232056792, "output_peak": 0.7078521145037394}, "full_scale": {"sample_count": 48320, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48320, "output_samples": 48320, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.619620509587709, "spectral_convergence": 0.5789444649121723, "spectral_convergence_status": "ok", "amplitude_floor": 1.6806044167073843e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.916954011895786, "spectral_convergence": 0.3335288911934407, "spectral_convergence_status": "ok", "amplitude_floor": 1.0917469705551035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.067036487368729, "spectral_convergence": 0.36011341165799554, "spectral_convergence_status": "ok", "amplitude_floor": 1.0075473322757857e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0009427012142024427, "output_power": 0.0007930239888832567, "delta_db": -0.7508774158080561, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0024186117972299723, "output_power": 0.0032930749190156422, "delta_db": 1.3403544433549466, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0014180970240603914, "output_power": 0.0018172933967998278, "delta_db": 1.077191029282099, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00023968394187503082, "output_power": 0.0002979339813199382, "delta_db": 0.9448110162252468, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.504236566794763e-05, "output_power": 6.304412753515185e-05, "delta_db": 0.5894754773309003, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00019111878985838765, "output_power": 0.00018527913509019288, "delta_db": -0.13476872146541563, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8062463505777774, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8811734232056792, "pitch": null, "output_rms_dbfs": -21.904630995173427, "output_peak": 0.7078521145037394, "reference_rms_dbfs": -22.785804418379104, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.201203669617407}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48320, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00023057786100875665, "log_rms_mae_db": 4.234667456091732, "rms_envelope_nrmse": 0.19118935077048477, "envelope_correlation": 0.9846386833492439, "rms_gain_error_db": 0.8812748388165608, "reference_mean_power": 0.005273848963429562, "output_mean_power": 0.00646033697815944}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.9211277816293244e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014051118338624634, "output_power": 0.0020320119007378604, "reference_active": true, "delta_db": 1.6021535569335266}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001364841166272218, "output_power": 0.00015737126174545401, "reference_active": true, "delta_db": 0.6184331365854991}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.767297964659028e-05, "output_power": 3.8009759832352644e-05, "reference_active": true, "delta_db": 0.038651545309568984}], "active_band_count": 3, "active_band_mae_db": 0.7530794129428648}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00020448142520679275, "log_rms_mae_db": 3.655873283147236, "rms_envelope_nrmse": 0.15898785317391945, "envelope_correlation": 0.993945251754445, "rms_gain_error_db": 0.8826140879239164, "reference_mean_power": 0.005325056781288949, "output_mean_power": 0.006525077162853491}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.674469134143454e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013611842323655793, "output_power": 0.001983415306462718, "reference_active": true, "delta_db": 1.6349675076975485}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.483051409804231e-05, "output_power": 7.357482739970348e-05, "reference_active": true, "delta_db": 0.5494978643622505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6716298904493925e-06, "output_power": 1.7403954797880093e-06, "reference_active": true, "delta_db": 0.1750781848850403}], "active_band_count": 3, "active_band_mae_db": 0.7865145189816131}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 0.00020294603022145288, "log_rms_mae_db": 6.571539612404468, "rms_envelope_nrmse": 0.16613204359657344, "envelope_correlation": 0.9907641038881662, "rms_gain_error_db": 0.8852151378596741, "reference_mean_power": 0.002254450133685994, "output_mean_power": 0.002764153440014783}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.5930281169536517e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007344814515063544, "output_power": 0.0008541135377811381, "reference_active": true, "delta_db": 0.6553477287858847}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010026570767730039, "output_power": 0.00011620931371886578, "reference_active": true, "delta_db": 0.6408851302281932}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.614259763381322e-05, "output_power": 1.7817273642632214e-05, "reference_active": true, "delta_db": 0.42867828273202796}], "active_band_count": 3, "active_band_mae_db": 0.5749703805820353}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 0.00017314576936579135, "log_rms_mae_db": 6.106051259016174, "rms_envelope_nrmse": 0.1360198494111946, "envelope_correlation": 0.996095170323478, "rms_gain_error_db": 0.8850863308509677, "reference_mean_power": 0.002277104155750591, "output_mean_power": 0.002791846452075275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.4781762283606311e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007359452368517945, "output_power": 0.0008594701343999686, "reference_active": true, "delta_db": 0.6738529122683045}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.796294327084698e-05, "output_power": 5.55067990319132e-05, "reference_active": true, "delta_db": 0.6344035719384287}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4655779178142778e-06, "output_power": 1.3722587429195225e-06, "reference_active": true, "delta_db": -0.285729063277263}], "active_band_count": 3, "active_band_mae_db": 0.5313285158279987}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 8.794013668187323e-05, "log_rms_mae_db": 1.6020884707769731, "rms_envelope_nrmse": 0.23844087174655845, "envelope_correlation": 0.9772774102544709, "rms_gain_error_db": 1.1114349416032292, "reference_mean_power": 0.0002815001552852329, "output_mean_power": 0.00036359854187530745}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 2.4039983283954285e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014419558923291274, "output_power": 0.00018069123416247955, "reference_active": true, "delta_db": 0.979851081826407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9143053212884802e-05, "output_power": 2.443641610186169e-05, "reference_active": true, "delta_db": 1.0602630509050401}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.190405195331938e-06, "output_power": 7.345843626254057e-06, "reference_active": true, "delta_db": 2.437856595903216}], "active_band_count": 3, "active_band_mae_db": 1.4926569095448876}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 7.034799916708663e-05, "log_rms_mae_db": 1.0581616921442307, "rms_envelope_nrmse": 0.18601314227242333, "envelope_correlation": 0.9900668450905462, "rms_gain_error_db": 1.1114355616505072, "reference_mean_power": 0.00028432562712779955, "output_mean_power": 0.0003672481036340372}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 2.2325770901894182e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001384543016382475, "output_power": 0.00017533673558448815, "reference_active": true, "delta_db": 1.0256646321708676}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.792343358100607e-06, "output_power": 1.225835783407302e-05, "reference_active": true, "delta_db": 0.9754566156135197}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0751348414140544e-07, "output_power": 5.008279830755344e-07, "reference_active": true, "delta_db": 2.118244226046779}], "active_band_count": 3, "active_band_mae_db": 1.373121824610389}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 603, "rms_floor": 5.049105990445221e-05, "log_rms_mae_db": 1.6473426572179457, "rms_envelope_nrmse": 0.22958670023364003, "envelope_correlation": 0.9640753652664081, "rms_gain_error_db": 0.0008340972168916418, "reference_mean_power": 0.00010138440655269997, "output_mean_power": 0.0001014038801071716}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 603, "frequency_resolution_hz": 0.33167495854063017, "power_floor": 1.3079247184162883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1617157093216335e-05, "output_power": 5.4577880894910844e-05, "reference_active": true, "delta_db": 0.2422258790705386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0705364401401438e-05, "output_power": 1.565779628978869e-05, "reference_active": true, "delta_db": -1.2135223968465427}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.974689061207133e-06, "output_power": 3.962249129335642e-06, "reference_active": true, "delta_db": 1.245002022156514}], "active_band_count": 3, "active_band_mae_db": 0.9002500993578652}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48320, "uncovered_tail_samples": 0, "metrics": {"frame_count": 597, "rms_floor": 4.0590764349045424e-05, "log_rms_mae_db": 1.0195966757134831, "rms_envelope_nrmse": 0.15186609906183482, "envelope_correlation": 0.9833056804066591, "rms_gain_error_db": 0.0008298212955698223, "reference_mean_power": 0.00010240265187942684, "output_mean_power": 0.0001024222201732064}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 597, "frequency_resolution_hz": 0.33500837520938026, "power_floor": 1.2076196938321023e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.154744016518151e-05, "output_power": 6.479731161287413e-05, "reference_active": true, "delta_db": 0.22346992930114928}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.945675894844695e-06, "output_power": 7.1776058214488135e-06, "reference_active": true, "delta_db": -1.4165469839538793}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5378071346519667e-07, "output_power": 2.549285825103568e-07, "reference_active": true, "delta_db": 0.019599171560895735}], "active_band_count": 3, "active_band_mae_db": 0.5532053616053081}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8305856795817977, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a0403ba9a69fc2e93780cee70ba32c88131456ae5c3c9b7fa3be88cb4b6f2853", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9772774102544709, "standard_deviation": null, "minimum": 0.9772774102544709, "maximum": 0.9772774102544709, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6020884707769731, "standard_deviation": null, "minimum": 1.6020884707769731, "maximum": 1.6020884707769731, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.4926569095448876, "standard_deviation": null, "minimum": 1.4926569095448876, "maximum": 1.4926569095448876, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23844087174655845, "standard_deviation": null, "minimum": 0.23844087174655845, "maximum": 0.23844087174655845, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9900668450905462, "standard_deviation": null, "minimum": 0.9900668450905462, "maximum": 0.9900668450905462, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0581616921442307, "standard_deviation": null, "minimum": 1.0581616921442307, "maximum": 1.0581616921442307, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.373121824610389, "standard_deviation": null, "minimum": 1.373121824610389, "maximum": 1.373121824610389, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18601314227242333, "standard_deviation": null, "minimum": 0.18601314227242333, "maximum": 0.18601314227242333, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9640753652664081, "standard_deviation": null, "minimum": 0.9640753652664081, "maximum": 0.9640753652664081, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6473426572179457, "standard_deviation": null, "minimum": 1.6473426572179457, "maximum": 1.6473426572179457, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9002500993578652, "standard_deviation": null, "minimum": 0.9002500993578652, "maximum": 0.9002500993578652, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22958670023364003, "standard_deviation": null, "minimum": 0.22958670023364003, "maximum": 0.22958670023364003, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9833056804066591, "standard_deviation": null, "minimum": 0.9833056804066591, "maximum": 0.9833056804066591, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0195966757134831, "standard_deviation": null, "minimum": 1.0195966757134831, "maximum": 1.0195966757134831, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5532053616053081, "standard_deviation": null, "minimum": 0.5532053616053081, "maximum": 0.5532053616053081, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15186609906183482, "standard_deviation": null, "minimum": 0.15186609906183482, "maximum": 0.15186609906183482, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9907641038881662, "standard_deviation": null, "minimum": 0.9907641038881662, "maximum": 0.9907641038881662, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 6.571539612404468, "standard_deviation": null, "minimum": 6.571539612404468, "maximum": 6.571539612404468, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5749703805820353, "standard_deviation": null, "minimum": 0.5749703805820353, "maximum": 0.5749703805820353, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16613204359657344, "standard_deviation": null, "minimum": 0.16613204359657344, "maximum": 0.16613204359657344, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.996095170323478, "standard_deviation": null, "minimum": 0.996095170323478, "maximum": 0.996095170323478, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 6.106051259016174, "standard_deviation": null, "minimum": 6.106051259016174, "maximum": 6.106051259016174, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5313285158279987, "standard_deviation": null, "minimum": 0.5313285158279987, "maximum": 0.5313285158279987, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1360198494111946, "standard_deviation": null, "minimum": 0.1360198494111946, "maximum": 0.1360198494111946, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8062463505777774, "standard_deviation": null, "minimum": 0.8062463505777774, "maximum": 0.8062463505777774, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.201203669617407, "standard_deviation": null, "minimum": 6.201203669617407, "maximum": 6.201203669617407, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7078521145037394, "standard_deviation": null, "minimum": 0.7078521145037394, "maximum": 0.7078521145037394, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8811734232056792, "standard_deviation": null, "minimum": 0.8811734232056792, "maximum": 0.8811734232056792, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9846386833492439, "standard_deviation": null, "minimum": 0.9846386833492439, "maximum": 0.9846386833492439, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 4.234667456091732, "standard_deviation": null, "minimum": 4.234667456091732, "maximum": 4.234667456091732, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7530794129428648, "standard_deviation": null, "minimum": 0.7530794129428648, "maximum": 0.7530794129428648, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19118935077048477, "standard_deviation": null, "minimum": 0.19118935077048477, "maximum": 0.19118935077048477, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.993945251754445, "standard_deviation": null, "minimum": 0.993945251754445, "maximum": 0.993945251754445, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 3.655873283147236, "standard_deviation": null, "minimum": 3.655873283147236, "maximum": 3.655873283147236, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7865145189816131, "standard_deviation": null, "minimum": 0.7865145189816131, "maximum": 0.7865145189816131, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.15898785317391945, "standard_deviation": null, "minimum": 0.15898785317391945, "maximum": 0.15898785317391945, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2830"}, {"case": "librispeech_2961-960-0010", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.0297842025756836, "predicted_mos": 3.0297842025756836, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3ea229c11e0b56faafe45417aa802274c98affa882702876a413acf1d6fd9f98", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.776462700217962}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.440889540104777, "全帯域_10ms_rms_envelope_nrmse": 0.26070230052015636, "全帯域_10ms_envelope_correlation": 0.9821077760233202, "全帯域_10ms_modulation_mae_db": 1.936553305259901, "全帯域_40ms_log_rms_mae_db": 1.1264976797097292, "全帯域_40ms_rms_envelope_nrmse": 0.23508976827705677, "全帯域_40ms_envelope_correlation": 0.9914550771784518, "全帯域_40ms_modulation_mae_db": 1.652074626066854, "300–1000 Hz_10ms_log_rms_mae_db": 1.8221532901924942, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.30105804295981287, "300–1000 Hz_10ms_envelope_correlation": 0.9816704817640944, "300–1000 Hz_10ms_modulation_mae_db": 2.1011830621644516, "300–1000 Hz_40ms_log_rms_mae_db": 1.3566661997001617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2760504390022808, "300–1000 Hz_40ms_envelope_correlation": 0.9910397809124627, "300–1000 Hz_40ms_modulation_mae_db": 1.7866448877311083, "1000–3000 Hz_10ms_log_rms_mae_db": 1.190577800499666, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1480412373026798, "1000–3000 Hz_10ms_envelope_correlation": 0.9893589648643493, "1000–3000 Hz_10ms_modulation_mae_db": 0.665574816814868, "1000–3000 Hz_40ms_log_rms_mae_db": 0.6919877496086451, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09913124236876249, "1000–3000 Hz_40ms_envelope_correlation": 0.997559276549596, "1000–3000 Hz_40ms_modulation_mae_db": 0.7056497050471192, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3339566112781422, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18730335284020816, "3000–6000 Hz_10ms_envelope_correlation": 0.9730590002814421, "3000–6000 Hz_10ms_modulation_mae_db": 0.30885156072118103, "3000–6000 Hz_40ms_log_rms_mae_db": 0.800852601070396, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11096792226693732, "3000–6000 Hz_40ms_envelope_correlation": 0.9914265466329609, "3000–6000 Hz_40ms_modulation_mae_db": 0.5218303657561888, "mr_stft_db": 5.820219761211614, "band_mae_db": 0.8666744937238477, "rms_gain_error_db": 1.6487169250715654, "output_peak": 0.4023529676168881}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.051954518789347, "spectral_convergence": 0.5354500390950409, "spectral_convergence_status": "ok", "amplitude_floor": 9.509362929412646e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.329324495880521, "spectral_convergence": 0.33438299310062874, "spectral_convergence_status": "ok", "amplitude_floor": 8.503880220761315e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.079380268964975, "spectral_convergence": 0.37228813512409714, "spectral_convergence_status": "ok", "amplitude_floor": 5.089992363719204e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 2.8030381557821282e-05, "output_power": 2.9133302459691374e-05, "delta_db": 0.16760708131897428, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003997973981863815, "output_power": 0.0005309704239529487, "delta_db": 1.2323036726265877, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008734907128440196, "output_power": 0.00141199790359736, "delta_db": 2.0857576009259646, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001464675075572333, "output_power": 0.00017086862609404617, "delta_db": 0.669210361895503, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.350074484130352e-05, "output_power": 7.202059125405053e-05, "delta_db": 0.5467786295382308, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.7271042012588545e-05, "output_power": 1.9371243584816476e-05, "delta_db": 0.4983896160378259, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8666744937238477, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.6487169250715654, "pitch": null, "output_rms_dbfs": -26.50846445653798, "output_peak": 0.4023529676168881, "reference_rms_dbfs": -28.157181381609547, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.820219761211614}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012908667094909063, "log_rms_mae_db": 1.440889540104777, "rms_envelope_nrmse": 0.26070230052015636, "envelope_correlation": 0.9821077760233202, "rms_gain_error_db": 1.6540019387443912, "reference_mean_power": 0.001527852496322364, "output_mean_power": 0.002236050573894367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.179575522794606e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002748180655831064, "output_power": 0.000457188992332168, "reference_active": true, "delta_db": 2.210504870752787}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.543131824921543e-05, "output_power": 9.532331175615561e-05, "reference_active": true, "delta_db": 2.354439153862469}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2340545383743233e-05, "output_power": 1.6436371268677338e-05, "reference_active": true, "delta_db": 1.2447158911644474}], "active_band_count": 3, "active_band_mae_db": 1.936553305259901}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001146348873577052, "log_rms_mae_db": 1.1264976797097292, "rms_envelope_nrmse": 0.23508976827705677, "envelope_correlation": 0.9914550771784518, "rms_gain_error_db": 1.6573050992128286, "reference_mean_power": 0.0015234711554082644, "output_mean_power": 0.0022313348364342792}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.718780555754886e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002667517105928089, "output_power": 0.0004481549787931321, "reference_active": true, "delta_db": 2.253210124992397}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.819324290787445e-05, "output_power": 4.795202065791841e-05, "reference_active": true, "delta_db": 2.306618795361906}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.654413380387e-07, "output_power": 8.385933392750627e-07, "reference_active": true, "delta_db": 0.396394957846259}], "active_band_count": 3, "active_band_mae_db": 1.652074626066854}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001269978635547725, "log_rms_mae_db": 1.8221532901924942, "rms_envelope_nrmse": 0.30105804295981287, "envelope_correlation": 0.9816704817640944, "rms_gain_error_db": 1.9440698867666741, "reference_mean_power": 0.0011723709196259178, "output_mean_power": 0.001834307011018483}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.280903308109198e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025355561575728737, "output_power": 0.0004415381891225518, "reference_active": true, "delta_db": 2.4089503838700157}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.608107242392719e-05, "output_power": 9.916968973391171e-05, "reference_active": true, "delta_db": 2.475626445454796}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1840508111446508e-05, "output_power": 1.6416008767188666e-05, "reference_active": true, "delta_db": 1.418972357168543}], "active_band_count": 3, "active_band_mae_db": 2.1011830621644516}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.809451727166842e-05, "log_rms_mae_db": 1.3566661997001617, "rms_envelope_nrmse": 0.2760504390022808, "envelope_correlation": 0.9910397809124627, "rms_gain_error_db": 1.9480222255074973, "reference_mean_power": 0.0011689931508753605, "output_mean_power": 0.001830687386273418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.891072559770159e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025025709822561247, "output_power": 0.0004376133450231118, "reference_active": true, "delta_db": 2.427041528027117}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.836247458716607e-05, "output_power": 4.9790770471605945e-05, "reference_active": true, "delta_db": 2.4440472691627875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.847292373572026e-07, "output_power": 8.782222187191652e-07, "reference_active": true, "delta_db": 0.4888458660034201}], "active_band_count": 3, "active_band_mae_db": 1.7866448877311083}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.9423530315000806e-05, "log_rms_mae_db": 1.190577800499666, "rms_envelope_nrmse": 0.1480412373026798, "envelope_correlation": 0.9893589648643493, "rms_gain_error_db": 0.6643077906475441, "reference_mean_power": 0.00015673633956673046, "output_mean_power": 0.00018264192634797772}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.197939100277272e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.3248792094830553e-05, "output_power": 4.1098458344695665e-05, "reference_active": true, "delta_db": 0.9204965893125405}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.648467501564817e-06, "output_power": 4.38191304136255e-06, "reference_active": true, "delta_db": 0.7955327261628603}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1792733873632069e-06, "output_power": 1.2580097626935416e-06, "reference_active": true, "delta_db": 0.2806951349692032}], "active_band_count": 3, "active_band_mae_db": 0.665574816814868}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.3506304987563625e-05, "log_rms_mae_db": 0.6919877496086451, "rms_envelope_nrmse": 0.09913124236876249, "envelope_correlation": 0.997559276549596, "rms_gain_error_db": 0.6664882924149287, "reference_mean_power": 0.0001561782704543471, "output_mean_power": 0.00018208301597239882}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.897719153255603e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.156452210778437e-05, "output_power": 3.91320821303278e-05, "reference_active": true, "delta_db": 0.9333373792977185}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6348481818884734e-06, "output_power": 1.8989536657162099e-06, "reference_active": true, "delta_db": 0.6503693946312922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.181225496757285e-08, "output_power": 6.351471154289881e-08, "reference_active": true, "delta_db": -0.5332423412123473}], "active_band_count": 3, "active_band_mae_db": 0.7056497050471192}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.766512572532955e-05, "log_rms_mae_db": 1.3339566112781422, "rms_envelope_nrmse": 0.18730335284020816, "envelope_correlation": 0.9730590002814421, "rms_gain_error_db": 0.4527368317293652, "reference_mean_power": 2.7841403311343547e-05, "output_mean_power": 3.090045004847069e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1058323154967027e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.079425705940839e-06, "output_power": 5.944559800958546e-06, "reference_active": true, "delta_db": 0.683050878061407}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8241793691525221e-06, "output_power": 1.8573790206858268e-06, "reference_active": true, "delta_db": 0.07832996136642853}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.319091479762654e-07, "output_power": 5.120590843352039e-07, "reference_active": true, "delta_db": -0.16517384273570757}], "active_band_count": 3, "active_band_mae_db": 0.30885156072118103}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5244936872690136e-05, "log_rms_mae_db": 0.800852601070396, "rms_envelope_nrmse": 0.11096792226693732, "envelope_correlation": 0.9914265466329609, "rms_gain_error_db": 0.4546423178244591, "reference_mean_power": 2.7868742577273385e-05, "output_mean_power": 3.094436718812429e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.800642728490222e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.984173950209188e-06, "output_power": 5.778840851956617e-06, "reference_active": true, "delta_db": 0.6424754323954642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0071402238290061e-06, "output_power": 9.882866170665757e-07, "reference_active": true, "delta_db": -0.08207027068916935}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932914966264332e-08, "output_power": 3.2405596286155095e-08, "reference_active": true, "delta_db": -0.8409453941839328}], "active_band_count": 3, "active_band_mae_db": 0.5218303657561888}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8144026734195449, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7768f12cb22d8c05b96d8fd947c3e5a8affdc31fd2576dfff0b5f37998df3855", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9893589648643493, "standard_deviation": null, "minimum": 0.9893589648643493, "maximum": 0.9893589648643493, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.190577800499666, "standard_deviation": null, "minimum": 1.190577800499666, "maximum": 1.190577800499666, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.665574816814868, "standard_deviation": null, "minimum": 0.665574816814868, "maximum": 0.665574816814868, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1480412373026798, "standard_deviation": null, "minimum": 0.1480412373026798, "maximum": 0.1480412373026798, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.997559276549596, "standard_deviation": null, "minimum": 0.997559276549596, "maximum": 0.997559276549596, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.6919877496086451, "standard_deviation": null, "minimum": 0.6919877496086451, "maximum": 0.6919877496086451, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7056497050471192, "standard_deviation": null, "minimum": 0.7056497050471192, "maximum": 0.7056497050471192, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09913124236876249, "standard_deviation": null, "minimum": 0.09913124236876249, "maximum": 0.09913124236876249, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9730590002814421, "standard_deviation": null, "minimum": 0.9730590002814421, "maximum": 0.9730590002814421, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3339566112781422, "standard_deviation": null, "minimum": 1.3339566112781422, "maximum": 1.3339566112781422, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.30885156072118103, "standard_deviation": null, "minimum": 0.30885156072118103, "maximum": 0.30885156072118103, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18730335284020816, "standard_deviation": null, "minimum": 0.18730335284020816, "maximum": 0.18730335284020816, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9914265466329609, "standard_deviation": null, "minimum": 0.9914265466329609, "maximum": 0.9914265466329609, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.800852601070396, "standard_deviation": null, "minimum": 0.800852601070396, "maximum": 0.800852601070396, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5218303657561888, "standard_deviation": null, "minimum": 0.5218303657561888, "maximum": 0.5218303657561888, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11096792226693732, "standard_deviation": null, "minimum": 0.11096792226693732, "maximum": 0.11096792226693732, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9816704817640944, "standard_deviation": null, "minimum": 0.9816704817640944, "maximum": 0.9816704817640944, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8221532901924942, "standard_deviation": null, "minimum": 1.8221532901924942, "maximum": 1.8221532901924942, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.1011830621644516, "standard_deviation": null, "minimum": 2.1011830621644516, "maximum": 2.1011830621644516, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30105804295981287, "standard_deviation": null, "minimum": 0.30105804295981287, "maximum": 0.30105804295981287, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9910397809124627, "standard_deviation": null, "minimum": 0.9910397809124627, "maximum": 0.9910397809124627, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3566661997001617, "standard_deviation": null, "minimum": 1.3566661997001617, "maximum": 1.3566661997001617, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7866448877311083, "standard_deviation": null, "minimum": 1.7866448877311083, "maximum": 1.7866448877311083, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2760504390022808, "standard_deviation": null, "minimum": 0.2760504390022808, "maximum": 0.2760504390022808, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8666744937238477, "standard_deviation": null, "minimum": 0.8666744937238477, "maximum": 0.8666744937238477, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.820219761211614, "standard_deviation": null, "minimum": 5.820219761211614, "maximum": 5.820219761211614, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4023529676168881, "standard_deviation": null, "minimum": 0.4023529676168881, "maximum": 0.4023529676168881, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.6487169250715654, "standard_deviation": null, "minimum": 1.6487169250715654, "maximum": 1.6487169250715654, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9821077760233202, "standard_deviation": null, "minimum": 0.9821077760233202, "maximum": 0.9821077760233202, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.440889540104777, "standard_deviation": null, "minimum": 1.440889540104777, "maximum": 1.440889540104777, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.936553305259901, "standard_deviation": null, "minimum": 1.936553305259901, "maximum": 1.936553305259901, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.26070230052015636, "standard_deviation": null, "minimum": 0.26070230052015636, "maximum": 0.26070230052015636, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9914550771784518, "standard_deviation": null, "minimum": 0.9914550771784518, "maximum": 0.9914550771784518, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1264976797097292, "standard_deviation": null, "minimum": 1.1264976797097292, "maximum": 1.1264976797097292, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.652074626066854, "standard_deviation": null, "minimum": 1.652074626066854, "maximum": 1.652074626066854, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.23508976827705677, "standard_deviation": null, "minimum": 0.23508976827705677, "maximum": 0.23508976827705677, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2961"}, {"case": "librispeech_2961-960-0019", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.609520673751831, "predicted_mos": 2.609520673751831, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "1bc8b7b37cdfc47782929349bad0d721634ca50f20eca99f1562050250667f90", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.164687700103968}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.610284809957172, "全帯域_10ms_rms_envelope_nrmse": 0.263159904759041, "全帯域_10ms_envelope_correlation": 0.9792792432206423, "全帯域_10ms_modulation_mae_db": 1.982493561252338, "全帯域_40ms_log_rms_mae_db": 1.2162409117984039, "全帯域_40ms_rms_envelope_nrmse": 0.23030665493140898, "全帯域_40ms_envelope_correlation": 0.9904525283758927, "全帯域_40ms_modulation_mae_db": 1.9048963990563976, "300–1000 Hz_10ms_log_rms_mae_db": 1.9216525717896111, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.32035624015581066, "300–1000 Hz_10ms_envelope_correlation": 0.9747571534191366, "300–1000 Hz_10ms_modulation_mae_db": 2.2015442042623192, "300–1000 Hz_40ms_log_rms_mae_db": 1.470911866596956, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2855609935914065, "300–1000 Hz_40ms_envelope_correlation": 0.9879130927216508, "300–1000 Hz_40ms_modulation_mae_db": 2.05544020224876, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2421385868797723, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17629432080515947, "1000–3000 Hz_10ms_envelope_correlation": 0.9895256321808512, "1000–3000 Hz_10ms_modulation_mae_db": 1.307639475705027, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7530342310259796, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13544690034731255, "1000–3000 Hz_40ms_envelope_correlation": 0.9971114934226355, "1000–3000 Hz_40ms_modulation_mae_db": 1.5750702405138177, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2679993197076234, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1612644317225593, "3000–6000 Hz_10ms_envelope_correlation": 0.9875328682868096, "3000–6000 Hz_10ms_modulation_mae_db": 0.561183510176548, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7229291482393334, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11036848545056807, "3000–6000 Hz_40ms_envelope_correlation": 0.9960004759996465, "3000–6000 Hz_40ms_modulation_mae_db": 0.7895107803174454, "mr_stft_db": 5.83511148958662, "band_mae_db": 0.8688254183969759, "rms_gain_error_db": 1.5890787329600622, "output_peak": 0.6329470232829832}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2054963274452986, "spectral_convergence": 0.6123361094464325, "spectral_convergence_status": "ok", "amplitude_floor": 1.5171791511760096e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.36519427482918, "spectral_convergence": 0.3455109296116972, "spectral_convergence_status": "ok", "amplitude_floor": 1.2250623067014244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.93464386648538, "spectral_convergence": 0.4007084912371922, "spectral_convergence_status": "ok", "amplitude_floor": 7.883290232655995e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.642531119863568e-05, "output_power": 4.7424126108124017e-05, "delta_db": 0.09244513232368093, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003916996423883643, "output_power": 0.0004684180722748587, "delta_db": 0.7768046741047742, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0020347588099002725, "output_power": 0.003224483775328894, "delta_db": 1.9994725851118877, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007257385551015356, "output_power": 0.0009408425287809828, "delta_db": 1.127367447992054, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001922577058777525, "output_power": 0.00021817849067991434, "delta_db": 0.5492817728397623, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.752614415284284e-05, "output_power": 4.3761483128943315e-05, "delta_db": 0.6675808980096959, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8688254183969759, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.5890787329600622, "pitch": null, "output_rms_dbfs": -23.059998588610256, "output_peak": 0.6329470232829832, "reference_rms_dbfs": -24.64907732157032, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.83511148958662}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019822260308892205, "log_rms_mae_db": 1.610284809957172, "rms_envelope_nrmse": 0.263159904759041, "envelope_correlation": 0.9792792432206423, "rms_gain_error_db": 1.5891295494142832, "reference_mean_power": 0.003431222806852645, "output_mean_power": 0.004947227418754689}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9226628348981265e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006393448264747715, "output_power": 0.001036324965208794, "reference_active": true, "delta_db": 2.0976080524120384}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011197882141156391, "output_power": 0.00021877785851038767, "reference_active": true, "delta_db": 2.908674747709792}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.279682095777351e-05, "output_power": 5.315343041144313e-05, "reference_active": true, "delta_db": 0.9411978836351832}], "active_band_count": 3, "active_band_mae_db": 1.982493561252338}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001679501427730559, "log_rms_mae_db": 1.2162409117984039, "rms_envelope_nrmse": 0.23030665493140898, "envelope_correlation": 0.9904525283758927, "rms_gain_error_db": 1.5892467674431625, "reference_mean_power": 0.003447933525092644, "output_mean_power": 0.004971455552465624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8195786138583623e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005865504590263927, "output_power": 0.0009618121130492648, "reference_active": true, "delta_db": 2.147848632722652}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.169773206547347e-05, "output_power": 0.00011754043917090777, "reference_active": true, "delta_db": 2.7991810886963524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.232527996839891e-06, "output_power": 2.6641756915990584e-06, "reference_active": true, "delta_db": 0.7676594757501882}], "active_band_count": 3, "active_band_mae_db": 1.9048963990563976}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001900450506992256, "log_rms_mae_db": 1.9216525717896111, "rms_envelope_nrmse": 0.32035624015581066, "envelope_correlation": 0.9747571534191366, "rms_gain_error_db": 1.9571339218158332, "reference_mean_power": 0.002194955532035717, "output_mean_power": 0.0034446025472322324}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4050180992423441e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048796679990628384, "output_power": 0.0008879217096700847, "reference_active": true, "delta_db": 2.599843999603109}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011744031611151056, "output_power": 0.0002256816259344147, "reference_active": true, "delta_db": 2.836789905001607}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.011051692782722e-05, "output_power": 5.248777098592708e-05, "reference_active": true, "delta_db": 1.1679987081822414}], "active_band_count": 3, "active_band_mae_db": 2.2015442042623192}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014767968623269652, "log_rms_mae_db": 1.470911866596956, "rms_envelope_nrmse": 0.2855609935914065, "envelope_correlation": 0.9879130927216508, "rms_gain_error_db": 1.9571800577155285, "reference_mean_power": 0.002205773735941533, "output_mean_power": 0.0034616166177722425}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315971715150661e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004578823491754882, "output_power": 0.0008395706549116425, "reference_active": true, "delta_db": 2.633033481900032}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0620858906342295e-05, "output_power": 0.00011465013467328428, "reference_active": true, "delta_db": 2.767524838339564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.249804591930109e-06, "output_power": 2.6836200473072684e-06, "reference_active": true, "delta_db": 0.7657622865066837}], "active_band_count": 3, "active_band_mae_db": 2.05544020224876}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012415579655882008, "log_rms_mae_db": 1.2421385868797723, "rms_envelope_nrmse": 0.17629432080515947, "envelope_correlation": 0.9895256321808512, "rms_gain_error_db": 0.9562797187818699, "reference_mean_power": 0.0006561031712122296, "output_mean_power": 0.000817711506035258}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1091236930101443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020492597233413883, "output_power": 0.0002679804682169641, "reference_active": true, "delta_db": 1.165061371788926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1330216505908255e-06, "output_power": 1.3013895900945531e-05, "reference_active": true, "delta_db": 2.611337862397987}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.269143835584622e-06, "output_power": 4.415630317758621e-06, "reference_active": true, "delta_db": 0.14651919292816754}], "active_band_count": 3, "active_band_mae_db": 1.307639475705027}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011294582885045651, "log_rms_mae_db": 0.7530342310259796, "rms_envelope_nrmse": 0.13544690034731255, "envelope_correlation": 0.9971114934226355, "rms_gain_error_db": 0.9565158723280622, "reference_mean_power": 0.000659140146325869, "output_mean_power": 0.0008215412061890181}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0026600561194904e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019159735065259618, "output_power": 0.0002521413217816545, "reference_active": true, "delta_db": 1.1925452567564911}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0478497584910014e-06, "output_power": 5.546249682801996e-06, "reference_active": true, "delta_db": 2.6000586191810102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0860941276748228e-07, "output_power": 2.58579793984022e-07, "reference_active": true, "delta_db": 0.932606845603952}], "active_band_count": 3, "active_band_mae_db": 1.5750702405138177}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.245162939338534e-05, "log_rms_mae_db": 1.2679993197076234, "rms_envelope_nrmse": 0.1612644317225593, "envelope_correlation": 0.9875328682868096, "rms_gain_error_db": 0.6420799023962577, "reference_mean_power": 8.990510589252764e-05, "output_mean_power": 0.00010422990624885192}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.06772172679864e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1343959051909258e-05, "output_power": 1.3601822558311687e-05, "reference_active": true, "delta_db": 0.7883245481018168}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.893195477577164e-06, "output_power": 4.256103988039919e-06, "reference_active": true, "delta_db": 0.38706020178601447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973785070466098e-07, "output_power": 7.093293780105146e-07, "reference_active": true, "delta_db": -0.5081657806418125}], "active_band_count": 3, "active_band_mae_db": 0.561183510176548}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.719774514443707e-05, "log_rms_mae_db": 0.7229291482393334, "rms_envelope_nrmse": 0.11036848545056807, "envelope_correlation": 0.9960004759996465, "rms_gain_error_db": 0.6420904905456716, "reference_mean_power": 9.035589049479441e-05, "output_mean_power": 0.00010475277085219756}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9012156385948547e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1053455137857872e-05, "output_power": 1.3186522258045331e-05, "reference_active": true, "delta_db": 0.7663221924199712}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0905012137138186e-06, "output_power": 2.2476011042797734e-06, "reference_active": true, "delta_db": 0.3146881276683375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.495315305956309e-08, "output_power": 4.82889804538635e-08, "reference_active": true, "delta_db": -1.2875220208640274}], "active_band_count": 3, "active_band_mae_db": 0.7895107803174454}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8398020107404753, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "97c26a085393dc1f4324c64c28deb55c1eceed78c087998123a90e71a5b22d07", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9895256321808512, "standard_deviation": null, "minimum": 0.9895256321808512, "maximum": 0.9895256321808512, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2421385868797723, "standard_deviation": null, "minimum": 1.2421385868797723, "maximum": 1.2421385868797723, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.307639475705027, "standard_deviation": null, "minimum": 1.307639475705027, "maximum": 1.307639475705027, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17629432080515947, "standard_deviation": null, "minimum": 0.17629432080515947, "maximum": 0.17629432080515947, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971114934226355, "standard_deviation": null, "minimum": 0.9971114934226355, "maximum": 0.9971114934226355, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7530342310259796, "standard_deviation": null, "minimum": 0.7530342310259796, "maximum": 0.7530342310259796, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5750702405138177, "standard_deviation": null, "minimum": 1.5750702405138177, "maximum": 1.5750702405138177, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13544690034731255, "standard_deviation": null, "minimum": 0.13544690034731255, "maximum": 0.13544690034731255, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9875328682868096, "standard_deviation": null, "minimum": 0.9875328682868096, "maximum": 0.9875328682868096, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2679993197076234, "standard_deviation": null, "minimum": 1.2679993197076234, "maximum": 1.2679993197076234, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.561183510176548, "standard_deviation": null, "minimum": 0.561183510176548, "maximum": 0.561183510176548, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1612644317225593, "standard_deviation": null, "minimum": 0.1612644317225593, "maximum": 0.1612644317225593, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9960004759996465, "standard_deviation": null, "minimum": 0.9960004759996465, "maximum": 0.9960004759996465, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7229291482393334, "standard_deviation": null, "minimum": 0.7229291482393334, "maximum": 0.7229291482393334, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7895107803174454, "standard_deviation": null, "minimum": 0.7895107803174454, "maximum": 0.7895107803174454, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11036848545056807, "standard_deviation": null, "minimum": 0.11036848545056807, "maximum": 0.11036848545056807, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9747571534191366, "standard_deviation": null, "minimum": 0.9747571534191366, "maximum": 0.9747571534191366, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9216525717896111, "standard_deviation": null, "minimum": 1.9216525717896111, "maximum": 1.9216525717896111, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.2015442042623192, "standard_deviation": null, "minimum": 2.2015442042623192, "maximum": 2.2015442042623192, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.32035624015581066, "standard_deviation": null, "minimum": 0.32035624015581066, "maximum": 0.32035624015581066, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9879130927216508, "standard_deviation": null, "minimum": 0.9879130927216508, "maximum": 0.9879130927216508, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.470911866596956, "standard_deviation": null, "minimum": 1.470911866596956, "maximum": 1.470911866596956, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 2.05544020224876, "standard_deviation": null, "minimum": 2.05544020224876, "maximum": 2.05544020224876, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2855609935914065, "standard_deviation": null, "minimum": 0.2855609935914065, "maximum": 0.2855609935914065, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8688254183969759, "standard_deviation": null, "minimum": 0.8688254183969759, "maximum": 0.8688254183969759, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.83511148958662, "standard_deviation": null, "minimum": 5.83511148958662, "maximum": 5.83511148958662, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6329470232829832, "standard_deviation": null, "minimum": 0.6329470232829832, "maximum": 0.6329470232829832, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.5890787329600622, "standard_deviation": null, "minimum": 1.5890787329600622, "maximum": 1.5890787329600622, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9792792432206423, "standard_deviation": null, "minimum": 0.9792792432206423, "maximum": 0.9792792432206423, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.610284809957172, "standard_deviation": null, "minimum": 1.610284809957172, "maximum": 1.610284809957172, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.982493561252338, "standard_deviation": null, "minimum": 1.982493561252338, "maximum": 1.982493561252338, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.263159904759041, "standard_deviation": null, "minimum": 0.263159904759041, "maximum": 0.263159904759041, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9904525283758927, "standard_deviation": null, "minimum": 0.9904525283758927, "maximum": 0.9904525283758927, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2162409117984039, "standard_deviation": null, "minimum": 1.2162409117984039, "maximum": 1.2162409117984039, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.9048963990563976, "standard_deviation": null, "minimum": 1.9048963990563976, "maximum": 1.9048963990563976, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.23030665493140898, "standard_deviation": null, "minimum": 0.23030665493140898, "maximum": 0.23030665493140898, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2961"}, {"case": "librispeech_2961-961-0011", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.090876579284668, "predicted_mos": 3.090876579284668, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ed49c8b9b2fa0527686cb0a6855406ffeb3317b7b3f58e83d1a09777149ed2df", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 4.698569399770349}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.424775248654498, "全帯域_10ms_rms_envelope_nrmse": 0.27633540901620657, "全帯域_10ms_envelope_correlation": 0.9757032582179858, "全帯域_10ms_modulation_mae_db": 2.0619318545377383, "全帯域_40ms_log_rms_mae_db": 1.0307548355232459, "全帯域_40ms_rms_envelope_nrmse": 0.245454688236828, "全帯域_40ms_envelope_correlation": 0.9862935998185516, "全帯域_40ms_modulation_mae_db": 1.6721918325279652, "300–1000 Hz_10ms_log_rms_mae_db": 1.757949721875253, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.33116140624108326, "300–1000 Hz_10ms_envelope_correlation": 0.9758321730689208, "300–1000 Hz_10ms_modulation_mae_db": 2.4247440412388976, "300–1000 Hz_40ms_log_rms_mae_db": 1.2973247732689563, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.3021276992198481, "300–1000 Hz_40ms_envelope_correlation": 0.986318996948641, "300–1000 Hz_40ms_modulation_mae_db": 2.0645001508004035, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3549728955812104, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1696456468778868, "1000–3000 Hz_10ms_envelope_correlation": 0.9870588006457548, "1000–3000 Hz_10ms_modulation_mae_db": 0.8640239280627666, "1000–3000 Hz_40ms_log_rms_mae_db": 0.849718002790771, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11379338700522978, "1000–3000 Hz_40ms_envelope_correlation": 0.996832255488111, "1000–3000 Hz_40ms_modulation_mae_db": 0.6387062685646322, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4462955804352158, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18209144225688748, "3000–6000 Hz_10ms_envelope_correlation": 0.977884441511656, "3000–6000 Hz_10ms_modulation_mae_db": 0.6057905339588958, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8329495687524818, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09124426601889224, "3000–6000 Hz_40ms_envelope_correlation": 0.9946940315727296, "3000–6000 Hz_40ms_modulation_mae_db": 0.9056505444510435, "mr_stft_db": 5.737869121974467, "band_mae_db": 0.8826585039957453, "rms_gain_error_db": 1.5924236733864947, "output_peak": 0.611765584408913}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.2571814944271456, "spectral_convergence": 0.6478433166210718, "spectral_convergence_status": "ok", "amplitude_floor": 1.4064891899964367e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.219797003828924, "spectral_convergence": 0.3704259628920103, "spectral_convergence_status": "ok", "amplitude_floor": 1.2304331075199438e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.73662886766733, "spectral_convergence": 0.40936525897707204, "spectral_convergence_status": "ok", "amplitude_floor": 7.896502575224385e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.444718915891811e-05, "output_power": 5.009099660846174e-05, "delta_db": -0.3621579169737354, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025153246625132694, "output_power": 0.00032651265129574655, "delta_db": 1.1330596437656502, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010938925867664089, "output_power": 0.0017818288601680501, "delta_db": 2.1189130973687953, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031036369064007244, "output_power": 0.000368329675048494, "delta_db": 0.743658022994441, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001328835538454751, "output_power": 0.00014360705975204725, "delta_db": 0.33704556040005207, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6997042780401114e-05, "output_power": 4.248921329287142e-05, "delta_db": 0.6011167824717978, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8826585039957453, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.5924236733864947, "pitch": null, "output_rms_dbfs": -25.665728650330287, "output_peak": 0.611765584408913, "reference_rms_dbfs": -27.258152323716782, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.737869121974467}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019413100530357063, "log_rms_mae_db": 1.424775248654498, "rms_envelope_nrmse": 0.27633540901620657, "envelope_correlation": 0.9757032582179858, "rms_gain_error_db": 1.5925185033608078, "reference_mean_power": 0.0018816420370153617, "output_mean_power": 0.002715118927215187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1507655155627412e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005726194534917747, "output_power": 0.0009544854075404013, "reference_active": true, "delta_db": 2.2190319501592346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.421513884012437e-05, "output_power": 0.0001568089140908002, "reference_active": true, "delta_db": 2.2125005485963136}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.679657413963232e-05, "output_power": 2.51563146257085e-05, "reference_active": true, "delta_db": 1.7542630648576676}], "active_band_count": 3, "active_band_mae_db": 2.0619318545377383}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016416501002274435, "log_rms_mae_db": 1.0307548355232459, "rms_envelope_nrmse": 0.245454688236828, "envelope_correlation": 0.9862935998185516, "rms_gain_error_db": 1.5927531981364065, "reference_mean_power": 0.001890759403997248, "output_mean_power": 0.002728422290339245}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0781635397104837e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005383866955860211, "output_power": 0.0009103759271154164, "reference_active": true, "delta_db": 2.281264460195624}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8874025426333776e-05, "output_power": 6.678109835529801e-05, "reference_active": true, "delta_db": 2.3499404327312994}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650199092025111e-06, "output_power": 1.7102394781356934e-06, "reference_active": true, "delta_db": 0.38537060465697126}], "active_band_count": 3, "active_band_mae_db": 1.6721918325279652}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017990629858894173, "log_rms_mae_db": 1.757949721875253, "rms_envelope_nrmse": 0.33116140624108326, "envelope_correlation": 0.9758321730689208, "rms_gain_error_db": 2.013401376258156, "reference_mean_power": 0.0012781810161111924, "output_mean_power": 0.002032041160359469}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.355790711503392e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004472139003696596, "output_power": 0.0008125887944790783, "reference_active": true, "delta_db": 2.5935553525825346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.621773891011176e-05, "output_power": 0.00014312304307740424, "reference_active": true, "delta_db": 2.7365350084754296}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7010237353267386e-05, "output_power": 2.661488229731324e-05, "reference_active": true, "delta_db": 1.944141762658729}], "active_band_count": 3, "active_band_mae_db": 2.4247440412388976}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000149706333097475, "log_rms_mae_db": 1.2973247732689563, "rms_envelope_nrmse": 0.3021276992198481, "envelope_correlation": 0.986318996948641, "rms_gain_error_db": 2.0135250070066086, "reference_mean_power": 0.0012844845607787134, "output_mean_power": 0.0020421206136148505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.840745806614025e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004236215690036061, "output_power": 0.0007825250374980167, "reference_active": true, "delta_db": 2.665201776097603}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1600142098026355e-05, "output_power": 6.25348470912839e-05, "reference_active": true, "delta_db": 2.9643305671906446}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5445822829049627e-06, "output_power": 1.758765184044494e-06, "reference_active": true, "delta_db": 0.5639681091129631}], "active_band_count": 3, "active_band_mae_db": 2.0645001508004035}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010005798255669499, "log_rms_mae_db": 1.3549728955812104, "rms_envelope_nrmse": 0.1696456468778868, "envelope_correlation": 0.9870588006457548, "rms_gain_error_db": 0.7470641834398566, "reference_mean_power": 0.0003605696849433608, "output_mean_power": 0.000428248281386598}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5037729366163507e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001406674030835393, "output_power": 0.00017153078859158294, "reference_active": true, "delta_db": 0.8614861438429486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0544824834133773e-05, "output_power": 2.4281862463062417e-05, "reference_active": true, "delta_db": 0.7257955205396517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4458512826457347e-06, "output_power": 4.342857127426098e-06, "reference_active": true, "delta_db": 1.0047901198056994}], "active_band_count": 3, "active_band_mae_db": 0.8640239280627666}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.18211380942645e-05, "log_rms_mae_db": 0.849718002790771, "rms_envelope_nrmse": 0.11379338700522978, "envelope_correlation": 0.996832255488111, "rms_gain_error_db": 0.7470673153106967, "reference_mean_power": 0.00036238111197119544, "output_mean_power": 0.00043040002197104293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3815388264892723e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001355192968806531, "output_power": 0.00016533337770843615, "reference_active": true, "delta_db": 0.8635939860645985}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.971075792205542e-06, "output_power": 9.324716997097644e-06, "reference_active": true, "delta_db": 0.6811872135162466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.416322058192608e-07, "output_power": 2.2183039774602824e-07, "reference_active": true, "delta_db": -0.3713376061130513}], "active_band_count": 3, "active_band_mae_db": 0.6387062685646322}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.283283642007837e-05, "log_rms_mae_db": 1.4462955804352158, "rms_envelope_nrmse": 0.18209144225688748, "envelope_correlation": 0.977884441511656, "rms_gain_error_db": 0.250895183873248, "reference_mean_power": 6.230386432542126e-05, "output_mean_power": 6.600920503932746e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0221136919924677e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.16042156208526e-05, "output_power": 1.2140764797263977e-05, "reference_active": true, "delta_db": 0.19630255675742073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3772446197384082e-06, "output_power": 2.4419818874780376e-06, "reference_active": true, "delta_db": 0.11668565236494143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4509999616936e-07, "output_power": 6.293553520834044e-07, "reference_active": true, "delta_db": 1.5043833927543253}], "active_band_count": 3, "active_band_mae_db": 0.6057905339588958}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.758960619357764e-05, "log_rms_mae_db": 0.8329495687524818, "rms_envelope_nrmse": 0.09124426601889224, "envelope_correlation": 0.9946940315727296, "rms_gain_error_db": 0.2508925841429748, "reference_mean_power": 6.26170059672204e-05, "output_mean_power": 6.634093015510205e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.846748801507289e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1232117533806203e-05, "output_power": 1.1867133818491232e-05, "reference_active": true, "delta_db": 0.23884200366055067}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.067581168241283e-06, "output_power": 1.1267795254454687e-06, "reference_active": true, "delta_db": 0.2343804250046627}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.579800269976513e-08, "output_power": 4.324730053853953e-08, "reference_active": true, "delta_db": 2.2437292046879174}], "active_band_count": 3, "active_band_mae_db": 0.9056505444510435}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9245942695463392, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a6ebcbcb1b6b5e1fe5dddc0bdc91f8b539ab4bd369bf6fa64e095ba50cf16832", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9870588006457548, "standard_deviation": null, "minimum": 0.9870588006457548, "maximum": 0.9870588006457548, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3549728955812104, "standard_deviation": null, "minimum": 1.3549728955812104, "maximum": 1.3549728955812104, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8640239280627666, "standard_deviation": null, "minimum": 0.8640239280627666, "maximum": 0.8640239280627666, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1696456468778868, "standard_deviation": null, "minimum": 0.1696456468778868, "maximum": 0.1696456468778868, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996832255488111, "standard_deviation": null, "minimum": 0.996832255488111, "maximum": 0.996832255488111, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.849718002790771, "standard_deviation": null, "minimum": 0.849718002790771, "maximum": 0.849718002790771, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6387062685646322, "standard_deviation": null, "minimum": 0.6387062685646322, "maximum": 0.6387062685646322, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11379338700522978, "standard_deviation": null, "minimum": 0.11379338700522978, "maximum": 0.11379338700522978, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.977884441511656, "standard_deviation": null, "minimum": 0.977884441511656, "maximum": 0.977884441511656, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4462955804352158, "standard_deviation": null, "minimum": 1.4462955804352158, "maximum": 1.4462955804352158, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6057905339588958, "standard_deviation": null, "minimum": 0.6057905339588958, "maximum": 0.6057905339588958, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18209144225688748, "standard_deviation": null, "minimum": 0.18209144225688748, "maximum": 0.18209144225688748, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9946940315727296, "standard_deviation": null, "minimum": 0.9946940315727296, "maximum": 0.9946940315727296, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8329495687524818, "standard_deviation": null, "minimum": 0.8329495687524818, "maximum": 0.8329495687524818, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9056505444510435, "standard_deviation": null, "minimum": 0.9056505444510435, "maximum": 0.9056505444510435, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09124426601889224, "standard_deviation": null, "minimum": 0.09124426601889224, "maximum": 0.09124426601889224, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9758321730689208, "standard_deviation": null, "minimum": 0.9758321730689208, "maximum": 0.9758321730689208, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.757949721875253, "standard_deviation": null, "minimum": 1.757949721875253, "maximum": 1.757949721875253, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.4247440412388976, "standard_deviation": null, "minimum": 2.4247440412388976, "maximum": 2.4247440412388976, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.33116140624108326, "standard_deviation": null, "minimum": 0.33116140624108326, "maximum": 0.33116140624108326, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.986318996948641, "standard_deviation": null, "minimum": 0.986318996948641, "maximum": 0.986318996948641, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2973247732689563, "standard_deviation": null, "minimum": 1.2973247732689563, "maximum": 1.2973247732689563, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 2.0645001508004035, "standard_deviation": null, "minimum": 2.0645001508004035, "maximum": 2.0645001508004035, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3021276992198481, "standard_deviation": null, "minimum": 0.3021276992198481, "maximum": 0.3021276992198481, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8826585039957453, "standard_deviation": null, "minimum": 0.8826585039957453, "maximum": 0.8826585039957453, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.737869121974467, "standard_deviation": null, "minimum": 5.737869121974467, "maximum": 5.737869121974467, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.611765584408913, "standard_deviation": null, "minimum": 0.611765584408913, "maximum": 0.611765584408913, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.5924236733864947, "standard_deviation": null, "minimum": 1.5924236733864947, "maximum": 1.5924236733864947, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9757032582179858, "standard_deviation": null, "minimum": 0.9757032582179858, "maximum": 0.9757032582179858, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.424775248654498, "standard_deviation": null, "minimum": 1.424775248654498, "maximum": 1.424775248654498, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 2.0619318545377383, "standard_deviation": null, "minimum": 2.0619318545377383, "maximum": 2.0619318545377383, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.27633540901620657, "standard_deviation": null, "minimum": 0.27633540901620657, "maximum": 0.27633540901620657, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9862935998185516, "standard_deviation": null, "minimum": 0.9862935998185516, "maximum": 0.9862935998185516, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0307548355232459, "standard_deviation": null, "minimum": 1.0307548355232459, "maximum": 1.0307548355232459, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.6721918325279652, "standard_deviation": null, "minimum": 1.6721918325279652, "maximum": 1.6721918325279652, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.245454688236828, "standard_deviation": null, "minimum": 0.245454688236828, "maximum": 0.245454688236828, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_2961"}, {"case": "librispeech_3570-5694-0001", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.571845769882202, "predicted_mos": 3.571845769882202, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 90800, "input_sample_rate": 16000, "input_seconds": 5.675, "input_float32_sha256": "c524cc0043d165c2028d6f3d871a1843b11143625eccb9ab4228a589740d9d09", "internal_samples": 90800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.951771500054747}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8834028131118743, "全帯域_10ms_rms_envelope_nrmse": 0.16188988998043066, "全帯域_10ms_envelope_correlation": 0.9859773857944337, "全帯域_10ms_modulation_mae_db": 0.3806437888559399, "全帯域_40ms_log_rms_mae_db": 1.2343878424369903, "全帯域_40ms_rms_envelope_nrmse": 0.12859777155408925, "全帯域_40ms_envelope_correlation": 0.9939252882797247, "全帯域_40ms_modulation_mae_db": 0.5682544425442351, "300–1000 Hz_10ms_log_rms_mae_db": 1.902388192427346, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20295665019159811, "300–1000 Hz_10ms_envelope_correlation": 0.986499291597458, "300–1000 Hz_10ms_modulation_mae_db": 0.5521373740849673, "300–1000 Hz_40ms_log_rms_mae_db": 1.3750956233660963, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.180003002531014, "300–1000 Hz_40ms_envelope_correlation": 0.9918341402350339, "300–1000 Hz_40ms_modulation_mae_db": 0.8510375693714617, "1000–3000 Hz_10ms_log_rms_mae_db": 1.834487220178404, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16451678062870254, "1000–3000 Hz_10ms_envelope_correlation": 0.9854210033912691, "1000–3000 Hz_10ms_modulation_mae_db": 0.6148083743314352, "1000–3000 Hz_40ms_log_rms_mae_db": 1.116013182134496, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10662510718303948, "1000–3000 Hz_40ms_envelope_correlation": 0.9965252108157929, "1000–3000 Hz_40ms_modulation_mae_db": 0.761069308358267, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6033550422855631, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22285729847907557, "3000–6000 Hz_10ms_envelope_correlation": 0.9676436239057743, "3000–6000 Hz_10ms_modulation_mae_db": 1.1745524600825805, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9136875930098058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11821194807602131, "3000–6000 Hz_40ms_envelope_correlation": 0.9903902444282131, "3000–6000 Hz_40ms_modulation_mae_db": 1.5731673880372732, "mr_stft_db": 5.7192737696959, "band_mae_db": 0.6429418911951771, "rms_gain_error_db": 0.843541776840886, "output_peak": 0.5072865396846746}, "full_scale": {"sample_count": 90800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 90800, "output_samples": 90800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.380797400691957, "spectral_convergence": 0.5390612256974586, "spectral_convergence_status": "ok", "amplitude_floor": 1.4038593411999917e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.284459313433415, "spectral_convergence": 0.2285300915467378, "spectral_convergence_status": "ok", "amplitude_floor": 9.854110842773577e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.492564594962329, "spectral_convergence": 0.26420363791272633, "spectral_convergence_status": "ok", "amplitude_floor": 8.983114295610284e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0018717561089627995, "output_power": 0.0021374565195211763, "delta_db": 0.5764802976761016, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00201441754378434, "output_power": 0.002517691159988528, "delta_db": 0.9685295992686065, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010569467904903444, "output_power": 0.0013719912666860692, "delta_db": 1.1329822258890831, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00018132930505040446, "output_power": 0.00021205819709266865, "delta_db": 0.6798706763047936, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.873439242518033e-05, "output_power": 8.669273940294455e-05, "delta_db": 0.41818245946756477, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.113111558995142e-05, "output_power": 3.055160814488435e-05, "delta_db": -0.08160608856491283, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6429418911951771, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.843541776840886, "pitch": null, "output_rms_dbfs": -21.967859462073832, "output_peak": 0.5072865396846746, "reference_rms_dbfs": -22.81140123891472, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7192737696959}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 90800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.0001853404348361204, "log_rms_mae_db": 1.8834028131118743, "rms_envelope_nrmse": 0.16188988998043066, "envelope_correlation": 0.9859773857944337, "rms_gain_error_db": 0.8435421114829277, "reference_mean_power": 0.005238930533969212, "output_mean_power": 0.006362046676422222}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 2.4623434734036186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012515931525511833, "output_power": 0.0014274960015806923, "reference_active": true, "delta_db": 0.5711172199827552}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003238300007472361, "output_power": 0.0003528855362647159, "reference_active": true, "delta_db": 0.37316777431891357}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.578860153055394e-05, "output_power": 7.24167876917166e-05, "reference_active": true, "delta_db": -0.197646372266151}], "active_band_count": 3, "active_band_mae_db": 0.3806437888559399}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00016148298564882675, "log_rms_mae_db": 1.2343878424369903, "rms_envelope_nrmse": 0.12859777155408925, "envelope_correlation": 0.9939252882797247, "rms_gain_error_db": 0.8435423347435786, "reference_mean_power": 0.005266794777733929, "output_mean_power": 0.006395884754791863}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.1218089587296202e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001164760758537026, "output_power": 0.0013328014761707038, "reference_active": true, "delta_db": 0.5852873449105167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015381285066774075, "output_power": 0.0001730917713886738, "reference_active": true, "delta_db": 0.5128380126945741}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.591871326154896e-06, "output_power": 4.862877980291675e-06, "reference_active": true, "delta_db": -0.6066379700276143}], "active_band_count": 3, "active_band_mae_db": 0.5682544425442351}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 0.00014972999954250175, "log_rms_mae_db": 1.902388192427346, "rms_envelope_nrmse": 0.20295665019159811, "envelope_correlation": 0.986499291597458, "rms_gain_error_db": 1.1855676005115283, "reference_mean_power": 0.0025001891623953804, "output_mean_power": 0.003284956541053908}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.4970716182960256e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007649864504896549, "output_power": 0.0009308559781475269, "reference_active": true, "delta_db": 0.8522874926480085}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022996394368157333, "output_power": 0.00027631236674946914, "reference_active": true, "delta_db": 0.7974057500288468}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.80930068112851e-05, "output_power": 4.801866064551559e-05, "reference_active": true, "delta_db": -0.006718879578046536}], "active_band_count": 3, "active_band_mae_db": 0.5521373740849673}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 0.00013636056906360855, "log_rms_mae_db": 1.3750956233660963, "rms_envelope_nrmse": 0.180003002531014, "envelope_correlation": 0.9918341402350339, "rms_gain_error_db": 1.185567858013437, "reference_mean_power": 0.0025134871346063777, "output_mean_power": 0.0033024287191636224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3161092414230216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215738253307795, "output_power": 0.0008818110933503768, "reference_active": true, "delta_db": 0.8709478699057169}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011225926953226224, "output_power": 0.00013428186291735152, "reference_active": true, "delta_db": 0.7779514571465549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.841934016741924e-06, "output_power": 3.119813002071531e-06, "reference_active": true, "delta_db": -0.9042133810621131}], "active_band_count": 3, "active_band_mae_db": 0.8510375693714617}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.2746042839695326e-05, "log_rms_mae_db": 1.834487220178404, "rms_envelope_nrmse": 0.16451678062870254, "envelope_correlation": 0.9854210033912691, "rms_gain_error_db": 0.7023098972776666, "reference_mean_power": 0.00020292459461300377, "output_mean_power": 0.00023854245072239466}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 1.5147206216944168e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.937659859611645e-05, "output_power": 8.100453380534525e-05, "reference_active": true, "delta_db": 0.6729632353434645}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3925889957061755e-05, "output_power": 2.8492533095310284e-05, "reference_active": true, "delta_db": 0.7586296045634775}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8985142545041605e-06, "output_power": 5.3636373329913116e-06, "reference_active": true, "delta_db": -0.4128322830873635}], "active_band_count": 3, "active_band_mae_db": 0.6148083743314352}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.512933910060942e-05, "log_rms_mae_db": 1.116013182134496, "rms_envelope_nrmse": 0.10662510718303948, "envelope_correlation": 0.9965252108157929, "rms_gain_error_db": 0.7023016285359651, "reference_mean_power": 0.00020400334148028423, "output_mean_power": 0.00023981008547761275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 1.3444198629453317e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.558296247449876e-05, "output_power": 7.708016768162771e-05, "reference_active": true, "delta_db": 0.7015162028874997}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3197503778987225e-05, "output_power": 1.5298778543173233e-05, "reference_active": true, "delta_db": 0.6416496303427746}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.991435827054748e-07, "output_power": 4.019955844116073e-07, "reference_active": true, "delta_db": -0.9400420918445268}], "active_band_count": 3, "active_band_mae_db": 0.761069308358267}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1134, "rms_floor": 5.624711907970004e-05, "log_rms_mae_db": 1.6033550422855631, "rms_envelope_nrmse": 0.22285729847907557, "envelope_correlation": 0.9676436239057743, "rms_gain_error_db": 0.09733771724318255, "reference_mean_power": 5.558522396873647e-05, "output_mean_power": 5.68451126708319e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1134, "frequency_resolution_hz": 0.1763668430335097, "power_floor": 3.230968975305017e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3093182032279326e-05, "output_power": 1.304874664525083e-05, "reference_active": true, "delta_db": -0.014764069025901373}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8826443460154885e-06, "output_power": 6.9451304288116275e-06, "reference_active": true, "delta_db": 0.03925077767887153}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3224614880429907e-06, "output_power": 1.494499252693578e-06, "reference_active": true, "delta_db": -3.469642533542969}], "active_band_count": 3, "active_band_mae_db": 1.1745524600825805}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 90800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1128, "rms_floor": 4.3621221947891856e-05, "log_rms_mae_db": 0.9136875930098058, "rms_envelope_nrmse": 0.11821194807602131, "envelope_correlation": 0.9903902444282131, "rms_gain_error_db": 0.09733729998017325, "reference_mean_power": 5.588085792782989e-05, "output_mean_power": 5.71474419467811e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1128, "frequency_resolution_hz": 0.1773049645390071, "power_floor": 2.891660816172807e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3816996130233874e-05, "output_power": 1.3344587429178493e-05, "reference_active": true, "delta_db": -0.15108484585113452}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.814539997745632e-06, "output_power": 3.4428902802150024e-06, "reference_active": true, "delta_db": -0.4451899048710404}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.118413559303936e-07, "output_power": 1.2067323871950806e-07, "reference_active": true, "delta_db": -4.123227413389644}], "active_band_count": 3, "active_band_mae_db": 1.5731673880372732}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9325445154410407, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "dafc0fea2b2776173e3ed89ec1f58682b310696a960b152898cf39c6c7ef2331", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9854210033912691, "standard_deviation": null, "minimum": 0.9854210033912691, "maximum": 0.9854210033912691, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.834487220178404, "standard_deviation": null, "minimum": 1.834487220178404, "maximum": 1.834487220178404, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6148083743314352, "standard_deviation": null, "minimum": 0.6148083743314352, "maximum": 0.6148083743314352, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16451678062870254, "standard_deviation": null, "minimum": 0.16451678062870254, "maximum": 0.16451678062870254, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9965252108157929, "standard_deviation": null, "minimum": 0.9965252108157929, "maximum": 0.9965252108157929, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.116013182134496, "standard_deviation": null, "minimum": 1.116013182134496, "maximum": 1.116013182134496, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.761069308358267, "standard_deviation": null, "minimum": 0.761069308358267, "maximum": 0.761069308358267, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10662510718303948, "standard_deviation": null, "minimum": 0.10662510718303948, "maximum": 0.10662510718303948, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9676436239057743, "standard_deviation": null, "minimum": 0.9676436239057743, "maximum": 0.9676436239057743, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6033550422855631, "standard_deviation": null, "minimum": 1.6033550422855631, "maximum": 1.6033550422855631, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1745524600825805, "standard_deviation": null, "minimum": 1.1745524600825805, "maximum": 1.1745524600825805, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22285729847907557, "standard_deviation": null, "minimum": 0.22285729847907557, "maximum": 0.22285729847907557, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9903902444282131, "standard_deviation": null, "minimum": 0.9903902444282131, "maximum": 0.9903902444282131, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9136875930098058, "standard_deviation": null, "minimum": 0.9136875930098058, "maximum": 0.9136875930098058, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5731673880372732, "standard_deviation": null, "minimum": 1.5731673880372732, "maximum": 1.5731673880372732, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11821194807602131, "standard_deviation": null, "minimum": 0.11821194807602131, "maximum": 0.11821194807602131, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.986499291597458, "standard_deviation": null, "minimum": 0.986499291597458, "maximum": 0.986499291597458, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.902388192427346, "standard_deviation": null, "minimum": 1.902388192427346, "maximum": 1.902388192427346, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5521373740849673, "standard_deviation": null, "minimum": 0.5521373740849673, "maximum": 0.5521373740849673, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20295665019159811, "standard_deviation": null, "minimum": 0.20295665019159811, "maximum": 0.20295665019159811, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9918341402350339, "standard_deviation": null, "minimum": 0.9918341402350339, "maximum": 0.9918341402350339, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3750956233660963, "standard_deviation": null, "minimum": 1.3750956233660963, "maximum": 1.3750956233660963, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8510375693714617, "standard_deviation": null, "minimum": 0.8510375693714617, "maximum": 0.8510375693714617, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.180003002531014, "standard_deviation": null, "minimum": 0.180003002531014, "maximum": 0.180003002531014, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6429418911951771, "standard_deviation": null, "minimum": 0.6429418911951771, "maximum": 0.6429418911951771, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.7192737696959, "standard_deviation": null, "minimum": 5.7192737696959, "maximum": 5.7192737696959, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5072865396846746, "standard_deviation": null, "minimum": 0.5072865396846746, "maximum": 0.5072865396846746, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.843541776840886, "standard_deviation": null, "minimum": 0.843541776840886, "maximum": 0.843541776840886, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9859773857944337, "standard_deviation": null, "minimum": 0.9859773857944337, "maximum": 0.9859773857944337, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8834028131118743, "standard_deviation": null, "minimum": 1.8834028131118743, "maximum": 1.8834028131118743, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3806437888559399, "standard_deviation": null, "minimum": 0.3806437888559399, "maximum": 0.3806437888559399, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16188988998043066, "standard_deviation": null, "minimum": 0.16188988998043066, "maximum": 0.16188988998043066, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9939252882797247, "standard_deviation": null, "minimum": 0.9939252882797247, "maximum": 0.9939252882797247, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2343878424369903, "standard_deviation": null, "minimum": 1.2343878424369903, "maximum": 1.2343878424369903, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5682544425442351, "standard_deviation": null, "minimum": 0.5682544425442351, "maximum": 0.5682544425442351, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12859777155408925, "standard_deviation": null, "minimum": 0.12859777155408925, "maximum": 0.12859777155408925, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3570"}, {"case": "librispeech_3570-5694-0004", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3302276134490967, "predicted_mos": 3.3302276134490967, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 85280, "input_sample_rate": 16000, "input_seconds": 5.33, "input_float32_sha256": "19b31d21dbb35a4f301666eb6dbdb899b5d79e44a51ebb2cb5ca10de5b52b3da", "internal_samples": 85280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.956671600230038}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.563550489664066, "全帯域_10ms_rms_envelope_nrmse": 0.1482292294981937, "全帯域_10ms_envelope_correlation": 0.9871252581397193, "全帯域_10ms_modulation_mae_db": 0.35178500220423786, "全帯域_40ms_log_rms_mae_db": 1.0815808634117738, "全帯域_40ms_rms_envelope_nrmse": 0.10690368946006806, "全帯域_40ms_envelope_correlation": 0.995904274270271, "全帯域_40ms_modulation_mae_db": 0.3996249012003681, "300–1000 Hz_10ms_log_rms_mae_db": 2.0196904753888374, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19963060719241263, "300–1000 Hz_10ms_envelope_correlation": 0.9818946508412271, "300–1000 Hz_10ms_modulation_mae_db": 0.5974319048661044, "300–1000 Hz_40ms_log_rms_mae_db": 1.4472809276647223, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15848752423340168, "300–1000 Hz_40ms_envelope_correlation": 0.9916308957056967, "300–1000 Hz_40ms_modulation_mae_db": 0.6252358725017856, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6544840621954227, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18230360198299936, "1000–3000 Hz_10ms_envelope_correlation": 0.9828038679957186, "1000–3000 Hz_10ms_modulation_mae_db": 0.7702267858372934, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9838397830695211, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11665587062803359, "1000–3000 Hz_40ms_envelope_correlation": 0.9960196959700865, "1000–3000 Hz_40ms_modulation_mae_db": 0.5392535097769494, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7266301890266096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23158586771819314, "3000–6000 Hz_10ms_envelope_correlation": 0.973725049689442, "3000–6000 Hz_10ms_modulation_mae_db": 1.232065105885181, "3000–6000 Hz_40ms_log_rms_mae_db": 1.051486254315183, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15654053591637393, "3000–6000 Hz_40ms_envelope_correlation": 0.9907597277372762, "3000–6000 Hz_40ms_modulation_mae_db": 0.7395840582707919, "mr_stft_db": 5.879529282208196, "band_mae_db": 0.6588098373929481, "rms_gain_error_db": 0.6986563450595922, "output_peak": 0.6475635112427683}, "full_scale": {"sample_count": 85280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 85280, "output_samples": 85280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.534077756350342, "spectral_convergence": 0.599414736558218, "spectral_convergence_status": "ok", "amplitude_floor": 1.2882929606468862e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2691059265732365, "spectral_convergence": 0.22056588776650113, "spectral_convergence_status": "ok", "amplitude_floor": 9.933866418820512e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.835404163701009, "spectral_convergence": 0.2631210751805227, "spectral_convergence_status": "ok", "amplitude_floor": 8.99033450142914e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001234751326260816, "output_power": 0.001450682456767871, "delta_db": 0.6999285780417415, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018448232014630782, "output_power": 0.0021339949961921008, "delta_db": 0.6323864485269979, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006193820215877917, "output_power": 0.0007579328767333966, "delta_db": 0.8767215058578482, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00017436605640499536, "output_power": 0.0002057415455635499, "delta_db": 0.7186005267239425, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010535962705358282, "output_power": 0.0001265600831438268, "delta_db": 0.7962252680018718, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.506722876313705e-05, "output_power": 1.5883018869682167e-05, "delta_db": 0.2289966972052866, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6588098373929481, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6986563450595922, "pitch": null, "output_rms_dbfs": -23.287535485463295, "output_peak": 0.6475635112427683, "reference_rms_dbfs": -23.986191830522884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.879529282208196}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 85280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.0001748275799785616, "log_rms_mae_db": 1.563550489664066, "rms_envelope_nrmse": 0.1482292294981937, "envelope_correlation": 0.9871252581397193, "rms_gain_error_db": 0.69865668317676, "reference_mean_power": 0.003997498793754849, "output_mean_power": 0.004695199061403343}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.9383161701154944e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012108392795166332, "output_power": 0.0013793044387507784, "reference_active": true, "delta_db": 0.5657363283027445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002231683599850243, "output_power": 0.00023987549260027157, "reference_active": true, "delta_db": 0.3135325769146817}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.764859371015312e-05, "output_power": 7.044779153581753e-05, "reference_active": true, "delta_db": 0.17608610139528746}], "active_band_count": 3, "active_band_mae_db": 0.35178500220423786}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00017019095451879833, "log_rms_mae_db": 1.0815808634117738, "rms_envelope_nrmse": 0.10690368946006806, "envelope_correlation": 0.995904274270271, "rms_gain_error_db": 0.6986566487281312, "reference_mean_power": 0.00402014109400192, "output_mean_power": 0.00472179318003399}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.6741447857886242e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011176954288226882, "output_power": 0.0012769163937122842, "reference_active": true, "delta_db": 0.5783898795256844}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.194819962775497e-05, "output_power": 9.798452908853799e-05, "reference_active": true, "delta_db": 0.2761427970869444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3053605179616555e-06, "output_power": 3.578104646158704e-06, "reference_active": true, "delta_db": 0.3443420269884754}], "active_band_count": 3, "active_band_mae_db": 0.3996249012003681}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 0.00014662911909855806, "log_rms_mae_db": 2.0196904753888374, "rms_envelope_nrmse": 0.19963060719241263, "envelope_correlation": 0.9818946508412271, "rms_gain_error_db": 0.9839982568694849, "reference_mean_power": 0.0015423797799867744, "output_mean_power": 0.0019345998459158417}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 1.0119951767590758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000652626948773928, "output_power": 0.0008267810586271308, "reference_active": true, "delta_db": 1.0272551543391188}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016341376588393177, "output_power": 0.00018719727058731062, "reference_active": true, "delta_db": 0.5901087379640573}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.788084498799748e-05, "output_power": 4.5990552177108804e-05, "reference_active": true, "delta_db": -0.17493182229513715}], "active_band_count": 3, "active_band_mae_db": 0.5974319048661044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 0.00013732417437068112, "log_rms_mae_db": 1.4472809276647223, "rms_envelope_nrmse": 0.15848752423340168, "envelope_correlation": 0.9916308957056967, "rms_gain_error_db": 0.9839933623277097, "reference_mean_power": 0.0015511172002359399, "output_mean_power": 0.001945556959181769}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 8.798155816797883e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006408046711327789, "output_power": 0.000809235087255946, "reference_active": true, "delta_db": 1.0134903601522216}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.486060187757687e-05, "output_power": 8.250606722940593e-05, "reference_active": true, "delta_db": 0.422325718257447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5748897665955177e-06, "output_power": 3.230527934545551e-06, "reference_active": true, "delta_db": -0.43989153909568823}], "active_band_count": 3, "active_band_mae_db": 0.6252358725017856}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 8.659636175169924e-05, "log_rms_mae_db": 1.6544840621954227, "rms_envelope_nrmse": 0.18230360198299936, "envelope_correlation": 0.9828038679957186, "rms_gain_error_db": 0.7553905089301818, "reference_mean_power": 0.00020984803844209803, "output_mean_power": 0.00024971461725324044}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 2.0976407038326697e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011684508155422031, "output_power": 0.0001372772355300928, "reference_active": true, "delta_db": 0.6998808879411321}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.122844755814475e-05, "output_power": 4.9963778185277694e-05, "reference_active": true, "delta_db": 0.8345828972310015}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.871758765973707e-06, "output_power": 7.4197294062064715e-06, "reference_active": true, "delta_db": -0.7762165723397467}], "active_band_count": 3, "active_band_mae_db": 0.7702267858372934}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 7.401841529112696e-05, "log_rms_mae_db": 0.9838397830695211, "rms_envelope_nrmse": 0.11665587062803359, "envelope_correlation": 0.9960196959700865, "rms_gain_error_db": 0.7553971740301808, "reference_mean_power": 0.000211036168686384, "output_mean_power": 0.0002511288518971566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 1.8857384691636124e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011739951182146869, "output_power": 0.00013886711803105144, "reference_active": true, "delta_db": 0.7293313149319063}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97467891176206e-05, "output_power": 2.2291763090923072e-05, "reference_active": true, "delta_db": 0.5264793065607585}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.379135569148535e-07, "output_power": 5.869036914082833e-07, "reference_active": true, "delta_db": -0.36194990783818354}], "active_band_count": 3, "active_band_mae_db": 0.5392535097769494}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1065, "rms_floor": 4.122543101057287e-05, "log_rms_mae_db": 1.7266301890266096, "rms_envelope_nrmse": 0.23158586771819314, "envelope_correlation": 0.973725049689442, "rms_gain_error_db": 0.8371883608365365, "reference_mean_power": 4.5035535336143536e-05, "output_mean_power": 5.461025011891843e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1065, "frequency_resolution_hz": 0.18779342723004694, "power_floor": 3.018630059598249e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5795382838093414e-05, "output_power": 1.9006692897788365e-05, "reference_active": true, "delta_db": 0.8037640111110831}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.352602360298049e-06, "output_power": 5.7297300724477325e-06, "reference_active": true, "delta_db": 1.193851696110456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.848555477734869e-06, "output_power": 2.7333198161207077e-06, "reference_active": true, "delta_db": 1.6985796104340043}], "active_band_count": 3, "active_band_mae_db": 1.232065105885181}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 85280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1059, "rms_floor": 3.1688229259308314e-05, "log_rms_mae_db": 1.051486254315183, "rms_envelope_nrmse": 0.15654053591637393, "envelope_correlation": 0.9907597277372762, "rms_gain_error_db": 0.8371885801791824, "reference_mean_power": 4.529062388817587e-05, "output_mean_power": 5.4919574176951604e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1059, "frequency_resolution_hz": 0.18885741265344666, "power_floor": 2.7414916263445066e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5534436484200316e-05, "output_power": 1.9170319163343335e-05, "reference_active": true, "delta_db": 0.9133383970299119}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0764100435071083e-06, "output_power": 2.6474232525419837e-06, "reference_active": true, "delta_db": 1.0551025810929855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4701562809614912e-07, "output_power": 1.5573801083428735e-07, "reference_active": true, "delta_db": 0.2503111966894778}], "active_band_count": 3, "active_band_mae_db": 0.7395840582707919}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9501864125465062, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "bb9b481bbf4b970ddefb4f1711261357fe7aac0aa387f4fb0b58e134e7a43236", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9828038679957186, "standard_deviation": null, "minimum": 0.9828038679957186, "maximum": 0.9828038679957186, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6544840621954227, "standard_deviation": null, "minimum": 1.6544840621954227, "maximum": 1.6544840621954227, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7702267858372934, "standard_deviation": null, "minimum": 0.7702267858372934, "maximum": 0.7702267858372934, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18230360198299936, "standard_deviation": null, "minimum": 0.18230360198299936, "maximum": 0.18230360198299936, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9960196959700865, "standard_deviation": null, "minimum": 0.9960196959700865, "maximum": 0.9960196959700865, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9838397830695211, "standard_deviation": null, "minimum": 0.9838397830695211, "maximum": 0.9838397830695211, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5392535097769494, "standard_deviation": null, "minimum": 0.5392535097769494, "maximum": 0.5392535097769494, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11665587062803359, "standard_deviation": null, "minimum": 0.11665587062803359, "maximum": 0.11665587062803359, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.973725049689442, "standard_deviation": null, "minimum": 0.973725049689442, "maximum": 0.973725049689442, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7266301890266096, "standard_deviation": null, "minimum": 1.7266301890266096, "maximum": 1.7266301890266096, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.232065105885181, "standard_deviation": null, "minimum": 1.232065105885181, "maximum": 1.232065105885181, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23158586771819314, "standard_deviation": null, "minimum": 0.23158586771819314, "maximum": 0.23158586771819314, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9907597277372762, "standard_deviation": null, "minimum": 0.9907597277372762, "maximum": 0.9907597277372762, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.051486254315183, "standard_deviation": null, "minimum": 1.051486254315183, "maximum": 1.051486254315183, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7395840582707919, "standard_deviation": null, "minimum": 0.7395840582707919, "maximum": 0.7395840582707919, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15654053591637393, "standard_deviation": null, "minimum": 0.15654053591637393, "maximum": 0.15654053591637393, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9818946508412271, "standard_deviation": null, "minimum": 0.9818946508412271, "maximum": 0.9818946508412271, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0196904753888374, "standard_deviation": null, "minimum": 2.0196904753888374, "maximum": 2.0196904753888374, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5974319048661044, "standard_deviation": null, "minimum": 0.5974319048661044, "maximum": 0.5974319048661044, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19963060719241263, "standard_deviation": null, "minimum": 0.19963060719241263, "maximum": 0.19963060719241263, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9916308957056967, "standard_deviation": null, "minimum": 0.9916308957056967, "maximum": 0.9916308957056967, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4472809276647223, "standard_deviation": null, "minimum": 1.4472809276647223, "maximum": 1.4472809276647223, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6252358725017856, "standard_deviation": null, "minimum": 0.6252358725017856, "maximum": 0.6252358725017856, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15848752423340168, "standard_deviation": null, "minimum": 0.15848752423340168, "maximum": 0.15848752423340168, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6588098373929481, "standard_deviation": null, "minimum": 0.6588098373929481, "maximum": 0.6588098373929481, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.879529282208196, "standard_deviation": null, "minimum": 5.879529282208196, "maximum": 5.879529282208196, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6475635112427683, "standard_deviation": null, "minimum": 0.6475635112427683, "maximum": 0.6475635112427683, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6986563450595922, "standard_deviation": null, "minimum": 0.6986563450595922, "maximum": 0.6986563450595922, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9871252581397193, "standard_deviation": null, "minimum": 0.9871252581397193, "maximum": 0.9871252581397193, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.563550489664066, "standard_deviation": null, "minimum": 1.563550489664066, "maximum": 1.563550489664066, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35178500220423786, "standard_deviation": null, "minimum": 0.35178500220423786, "maximum": 0.35178500220423786, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1482292294981937, "standard_deviation": null, "minimum": 0.1482292294981937, "maximum": 0.1482292294981937, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.995904274270271, "standard_deviation": null, "minimum": 0.995904274270271, "maximum": 0.995904274270271, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0815808634117738, "standard_deviation": null, "minimum": 1.0815808634117738, "maximum": 1.0815808634117738, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3996249012003681, "standard_deviation": null, "minimum": 0.3996249012003681, "maximum": 0.3996249012003681, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10690368946006806, "standard_deviation": null, "minimum": 0.10690368946006806, "maximum": 0.10690368946006806, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3570"}, {"case": "librispeech_3570-5696-0010", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.1608128547668457, "predicted_mos": 3.1608128547668457, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e6e53e7b3274267d65d5d3b24cb40dc43a1681317fc6e4df245610c427fb5792", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.932281699962914}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6961907937156275, "全帯域_10ms_rms_envelope_nrmse": 0.15529783363984076, "全帯域_10ms_envelope_correlation": 0.9835761149876094, "全帯域_10ms_modulation_mae_db": 0.35003776961564964, "全帯域_40ms_log_rms_mae_db": 1.1855086667735952, "全帯域_40ms_rms_envelope_nrmse": 0.11861321015038859, "全帯域_40ms_envelope_correlation": 0.9956669624486734, "全帯域_40ms_modulation_mae_db": 0.6234428011572724, "300–1000 Hz_10ms_log_rms_mae_db": 1.9321351790419241, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22068684186754592, "300–1000 Hz_10ms_envelope_correlation": 0.9804265750097602, "300–1000 Hz_10ms_modulation_mae_db": 1.1322935150197553, "300–1000 Hz_40ms_log_rms_mae_db": 1.3931384329805239, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.19004577665166755, "300–1000 Hz_40ms_envelope_correlation": 0.9903230310630556, "300–1000 Hz_40ms_modulation_mae_db": 1.1836221052512175, "1000–3000 Hz_10ms_log_rms_mae_db": 1.693185240515656, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15998739474263957, "1000–3000 Hz_10ms_envelope_correlation": 0.9893039395583957, "1000–3000 Hz_10ms_modulation_mae_db": 0.5947670792877827, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0251643547729181, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1119094011414906, "1000–3000 Hz_40ms_envelope_correlation": 0.9970688628104184, "1000–3000 Hz_40ms_modulation_mae_db": 0.897895880805318, "3000–6000 Hz_10ms_log_rms_mae_db": 1.832752052959417, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2467981201653181, "3000–6000 Hz_10ms_envelope_correlation": 0.9577610232907419, "3000–6000 Hz_10ms_modulation_mae_db": 0.8990946498209994, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1613003936983752, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14732662997567988, "3000–6000 Hz_40ms_envelope_correlation": 0.9838621670622646, "3000–6000 Hz_40ms_modulation_mae_db": 0.724636548834393, "mr_stft_db": 5.826848885862649, "band_mae_db": 0.8426568708758664, "rms_gain_error_db": 0.8640993608237559, "output_peak": 0.7844574008980698}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.666005723603584, "spectral_convergence": 0.6093848605760777, "spectral_convergence_status": "ok", "amplitude_floor": 1.1745223847485646e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.10646395665953, "spectral_convergence": 0.20652643132355664, "spectral_convergence_status": "ok", "amplitude_floor": 9.481908348017771e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.708076977324833, "spectral_convergence": 0.24245843115943533, "spectral_convergence_status": "ok", "amplitude_floor": 8.395402116778918e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.004068927413639468, "output_power": 0.004834376407752977, "delta_db": 0.748605193829288, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0018176508169902323, "output_power": 0.0021481013837262434, "delta_db": 0.72544318871663, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007805177117872849, "output_power": 0.0011708595090540025, "delta_db": 1.7612202470797107, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000310938544728363, "output_power": 0.0003713762050304028, "delta_db": 0.7713951237714617, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.08647023559392e-05, "output_power": 9.31169598655304e-05, "delta_db": 0.6126979609784597, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.353447039170603e-05, "output_power": 4.813835181581441e-05, "delta_db": 0.436579510879648, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8426568708758664, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8640993608237559, "pitch": null, "output_rms_dbfs": -20.621828780206386, "output_peak": 0.7844574008980698, "reference_rms_dbfs": -21.485928141030136, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.826848885862649}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019068326661124743, "log_rms_mae_db": 1.6961907937156275, "rms_envelope_nrmse": 0.15529783363984076, "envelope_correlation": 0.9835761149876094, "rms_gain_error_db": 0.8659842077583358, "reference_mean_power": 0.007104543911662494, "output_mean_power": 0.008672306603621225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.7734542191089184e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001358301262786015, "output_power": 0.001614456860104571, "reference_active": true, "delta_db": 0.7503034027956923}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002660490997406596, "output_power": 0.00026523976768554373, "reference_active": true, "delta_db": -0.013231548372178856}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.883599774152297e-05, "output_power": 7.38017623384279e-05, "reference_active": true, "delta_db": -0.2865783576790778}], "active_band_count": 3, "active_band_mae_db": 0.35003776961564964}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001870958079240233, "log_rms_mae_db": 1.1855086667735952, "rms_envelope_nrmse": 0.11861321015038859, "envelope_correlation": 0.9956669624486734, "rms_gain_error_db": 0.8669785688417979, "reference_mean_power": 0.007117638239873263, "output_mean_power": 0.008690279962352837}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4206220628648622e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0012103399895685932, "output_power": 0.0014432459875806235, "reference_active": true, "delta_db": 0.7643297618313826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012555497766250794, "output_power": 0.0001227416300060479, "reference_active": true, "delta_db": -0.09842048710472216}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0918852450489625e-06, "output_power": 4.830520152997814e-06, "reference_active": true, "delta_db": -1.0075781545357123}], "active_band_count": 3, "active_band_mae_db": 0.6234428011572724}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013521834879092463, "log_rms_mae_db": 1.9321351790419241, "rms_envelope_nrmse": 0.22068684186754592, "envelope_correlation": 0.9804265750097602, "rms_gain_error_db": 1.3046010569315518, "reference_mean_power": 0.0018937311393695897, "output_mean_power": 0.0025572808484504842}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.102346197011796e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006150154421416123, "output_power": 0.0008644731578554432, "reference_active": true, "delta_db": 1.478654924345785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012427322589552618, "output_power": 0.00017218190886958683, "reference_active": true, "delta_db": 1.4160994631336976}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6589172324725154e-05, "output_power": 2.3686026069602074e-05, "reference_active": true, "delta_db": -0.5021261575797836}], "active_band_count": 3, "active_band_mae_db": 1.1322935150197553}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012428224153258587, "log_rms_mae_db": 1.3931384329805239, "rms_envelope_nrmse": 0.19004577665166755, "envelope_correlation": 0.9903230310630556, "rms_gain_error_db": 1.3042459915303506, "reference_mean_power": 0.0018973976433641784, "output_mean_power": 0.0025620225975457223}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.796182335233462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005643605996811222, "output_power": 0.0007963668929088611, "reference_active": true, "delta_db": 1.4955651068447475}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.691592416494353e-05, "output_power": 7.537095071590842e-05, "reference_active": true, "delta_db": 1.2197020137672248}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2464097368217303e-06, "output_power": 1.8532288538923713e-06, "reference_active": true, "delta_db": -0.8355991951416801}], "active_band_count": 3, "active_band_mae_db": 1.1836221052512175}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012748188769016831, "log_rms_mae_db": 1.693185240515656, "rms_envelope_nrmse": 0.15998739474263957, "envelope_correlation": 0.9893039395583957, "rms_gain_error_db": 0.753482347948951, "reference_mean_power": 0.00032033050063867626, "output_mean_power": 0.00038101890736520023}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0949985705721996e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.663152856065707e-05, "output_power": 8.238406357633081e-05, "reference_active": true, "delta_db": 0.9216343331576056}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2062707324852824e-05, "output_power": 1.1931489412670844e-05, "reference_active": true, "delta_db": -0.04750130535971049}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2084052775211794e-06, "output_power": 2.6593326282486574e-06, "reference_active": true, "delta_db": -0.8151655993460323}], "active_band_count": 3, "active_band_mae_db": 0.5947670792877827}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001243805164613915, "log_rms_mae_db": 1.0251643547729181, "rms_envelope_nrmse": 0.1119094011414906, "envelope_correlation": 0.9970688628104184, "rms_gain_error_db": 0.7439137242786671, "reference_mean_power": 0.0003214084172642619, "output_mean_power": 0.0003814596611050918}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.993543951674123e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.454820101360055e-05, "output_power": 7.867576337879849e-05, "reference_active": true, "delta_db": 0.8595682270282495}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.405216922862835e-06, "output_power": 5.333595471384095e-06, "reference_active": true, "delta_db": -0.05793054431626077}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4345993241652877e-07, "output_power": 2.281692910629128e-07, "reference_active": true, "delta_db": -1.7761888710714437}], "active_band_count": 3, "active_band_mae_db": 0.897895880805318}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.885927766129993e-05, "log_rms_mae_db": 1.832752052959417, "rms_envelope_nrmse": 0.2467981201653181, "envelope_correlation": 0.9577610232907419, "rms_gain_error_db": 0.3694021410114131, "reference_mean_power": 5.543232992418922e-05, "output_mean_power": 6.035362321844355e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.209399643069482e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1581941196726732e-05, "output_power": 2.44997618054659e-05, "reference_active": true, "delta_db": 0.5507135713239664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7792321883200776e-05, "output_power": 1.9534835185708942e-05, "reference_active": true, "delta_db": 0.4057712460768572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1497984756130906e-05, "output_power": 7.700905876799757e-06, "reference_active": true, "delta_db": -1.7407991320621745}], "active_band_count": 3, "active_band_mae_db": 0.8990946498209994}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.78543565194846e-05, "log_rms_mae_db": 1.1613003936983752, "rms_envelope_nrmse": 0.14732662997567988, "envelope_correlation": 0.9838621670622646, "rms_gain_error_db": 0.3694244070068484, "reference_mean_power": 5.5701179342093606e-05, "output_mean_power": 6.0646652071872445e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.455779259683724e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.849693387286164e-05, "output_power": 3.0220385667224368e-05, "reference_active": true, "delta_db": 0.25501867813415446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0110098860775756e-05, "output_power": 9.4536732245121e-06, "reference_active": true, "delta_db": -0.29154815913293963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1498909214748545e-06, "output_power": 7.905372966254738e-07, "reference_active": true, "delta_db": -1.6273428092360849}], "active_band_count": 3, "active_band_mae_db": 0.724636548834393}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9358839079723194, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7c26da7bdff3ecc0c1c602505c5bd0712a9744890f215fd55205daea1c3ae964", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9893039395583957, "standard_deviation": null, "minimum": 0.9893039395583957, "maximum": 0.9893039395583957, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.693185240515656, "standard_deviation": null, "minimum": 1.693185240515656, "maximum": 1.693185240515656, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5947670792877827, "standard_deviation": null, "minimum": 0.5947670792877827, "maximum": 0.5947670792877827, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15998739474263957, "standard_deviation": null, "minimum": 0.15998739474263957, "maximum": 0.15998739474263957, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970688628104184, "standard_deviation": null, "minimum": 0.9970688628104184, "maximum": 0.9970688628104184, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0251643547729181, "standard_deviation": null, "minimum": 1.0251643547729181, "maximum": 1.0251643547729181, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.897895880805318, "standard_deviation": null, "minimum": 0.897895880805318, "maximum": 0.897895880805318, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1119094011414906, "standard_deviation": null, "minimum": 0.1119094011414906, "maximum": 0.1119094011414906, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9577610232907419, "standard_deviation": null, "minimum": 0.9577610232907419, "maximum": 0.9577610232907419, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.832752052959417, "standard_deviation": null, "minimum": 1.832752052959417, "maximum": 1.832752052959417, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8990946498209994, "standard_deviation": null, "minimum": 0.8990946498209994, "maximum": 0.8990946498209994, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2467981201653181, "standard_deviation": null, "minimum": 0.2467981201653181, "maximum": 0.2467981201653181, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9838621670622646, "standard_deviation": null, "minimum": 0.9838621670622646, "maximum": 0.9838621670622646, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1613003936983752, "standard_deviation": null, "minimum": 1.1613003936983752, "maximum": 1.1613003936983752, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.724636548834393, "standard_deviation": null, "minimum": 0.724636548834393, "maximum": 0.724636548834393, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14732662997567988, "standard_deviation": null, "minimum": 0.14732662997567988, "maximum": 0.14732662997567988, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9804265750097602, "standard_deviation": null, "minimum": 0.9804265750097602, "maximum": 0.9804265750097602, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9321351790419241, "standard_deviation": null, "minimum": 1.9321351790419241, "maximum": 1.9321351790419241, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1322935150197553, "standard_deviation": null, "minimum": 1.1322935150197553, "maximum": 1.1322935150197553, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22068684186754592, "standard_deviation": null, "minimum": 0.22068684186754592, "maximum": 0.22068684186754592, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9903230310630556, "standard_deviation": null, "minimum": 0.9903230310630556, "maximum": 0.9903230310630556, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3931384329805239, "standard_deviation": null, "minimum": 1.3931384329805239, "maximum": 1.3931384329805239, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1836221052512175, "standard_deviation": null, "minimum": 1.1836221052512175, "maximum": 1.1836221052512175, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19004577665166755, "standard_deviation": null, "minimum": 0.19004577665166755, "maximum": 0.19004577665166755, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8426568708758664, "standard_deviation": null, "minimum": 0.8426568708758664, "maximum": 0.8426568708758664, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.826848885862649, "standard_deviation": null, "minimum": 5.826848885862649, "maximum": 5.826848885862649, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7844574008980698, "standard_deviation": null, "minimum": 0.7844574008980698, "maximum": 0.7844574008980698, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8640993608237559, "standard_deviation": null, "minimum": 0.8640993608237559, "maximum": 0.8640993608237559, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9835761149876094, "standard_deviation": null, "minimum": 0.9835761149876094, "maximum": 0.9835761149876094, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6961907937156275, "standard_deviation": null, "minimum": 1.6961907937156275, "maximum": 1.6961907937156275, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.35003776961564964, "standard_deviation": null, "minimum": 0.35003776961564964, "maximum": 0.35003776961564964, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15529783363984076, "standard_deviation": null, "minimum": 0.15529783363984076, "maximum": 0.15529783363984076, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9956669624486734, "standard_deviation": null, "minimum": 0.9956669624486734, "maximum": 0.9956669624486734, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1855086667735952, "standard_deviation": null, "minimum": 1.1855086667735952, "maximum": 1.1855086667735952, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6234428011572724, "standard_deviation": null, "minimum": 0.6234428011572724, "maximum": 0.6234428011572724, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11861321015038859, "standard_deviation": null, "minimum": 0.11861321015038859, "maximum": 0.11861321015038859, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3570"}, {"case": "librispeech_3575-170457-0034", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9313712120056152, "predicted_mos": 2.9313712120056152, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 55920, "input_sample_rate": 16000, "input_seconds": 3.495, "input_float32_sha256": "0985da111e813131ddadbb50aa0831f7ab8070326b230a4516da248c70ba23e4", "internal_samples": 55920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.544290399644524}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8122634033405418, "全帯域_10ms_rms_envelope_nrmse": 0.30828250753217445, "全帯域_10ms_envelope_correlation": 0.9072659979201808, "全帯域_10ms_modulation_mae_db": 3.450171747564825, "全帯域_40ms_log_rms_mae_db": 1.6231904502469203, "全帯域_40ms_rms_envelope_nrmse": 0.2961638950850278, "全帯域_40ms_envelope_correlation": 0.906704897215512, "全帯域_40ms_modulation_mae_db": 4.16317795793199, "300–1000 Hz_10ms_log_rms_mae_db": 2.175434233802299, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2843112005460939, "300–1000 Hz_10ms_envelope_correlation": 0.9608831906659929, "300–1000 Hz_10ms_modulation_mae_db": 2.3294843249342922, "300–1000 Hz_40ms_log_rms_mae_db": 1.9555984745242934, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.267994655868643, "300–1000 Hz_40ms_envelope_correlation": 0.9649760637976114, "300–1000 Hz_40ms_modulation_mae_db": 2.6523810818420057, "1000–3000 Hz_10ms_log_rms_mae_db": 1.308752954682663, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18306563090877778, "1000–3000 Hz_10ms_envelope_correlation": 0.9940736428675894, "1000–3000 Hz_10ms_modulation_mae_db": 1.2389792790596699, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0399261843318819, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1662866277597288, "1000–3000 Hz_40ms_envelope_correlation": 0.9973241773457772, "1000–3000 Hz_40ms_modulation_mae_db": 0.8838336327994658, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3791169088629263, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23860465722679713, "3000–6000 Hz_10ms_envelope_correlation": 0.9630393616875143, "3000–6000 Hz_10ms_modulation_mae_db": 1.2287200032506318, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9562223537877983, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1560013621006151, "3000–6000 Hz_40ms_envelope_correlation": 0.9862985805026563, "3000–6000 Hz_40ms_modulation_mae_db": 0.7798913013881498, "mr_stft_db": 5.778873252690896, "band_mae_db": 2.1640586227983425, "rms_gain_error_db": 0.7322980505933767, "output_peak": 0.956721634957942}, "full_scale": {"sample_count": 55920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 55920, "output_samples": 55920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.005665980398814, "spectral_convergence": 0.5563612401655464, "spectral_convergence_status": "ok", "amplitude_floor": 7.545858591823444e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.382698133841964, "spectral_convergence": 0.4834688177665211, "spectral_convergence_status": "ok", "amplitude_floor": 3.9963548992246115e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.94825564383191, "spectral_convergence": 0.4833958509381772, "spectral_convergence_status": "ok", "amplitude_floor": 1.7986227023732554e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001975478876015244, "output_power": 0.0003939986071533021, "delta_db": -7.001777035604486, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005013278749394044, "output_power": 0.00730986324250647, "delta_db": 1.6378739896226169, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015170067046631896, "output_power": 0.00210468389811862, "delta_db": 1.4219937827596527, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0018605244766058998, "output_power": 0.0024358563845286512, "delta_db": 1.170162913129738, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002538671104412511, "output_power": 0.0003278942376417552, "delta_db": 1.1112734468854535, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.356486516629775e-05, "output_power": 7.367907874680569e-05, "delta_db": 0.6412705687881061, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 2.1640586227983425, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7322980505933767, "pitch": null, "output_rms_dbfs": -18.98047665675289, "output_peak": 0.956721634957942, "reference_rms_dbfs": -19.712774707346266, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.778873252690896}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 55920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.0006292775937517623, "log_rms_mae_db": 1.8122634033405418, "rms_envelope_nrmse": 0.30828250753217445, "envelope_correlation": 0.9072659979201808, "rms_gain_error_db": 0.7322984474901485, "reference_mean_power": 0.010699017072596153, "output_mean_power": 0.012664082331169375}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 4.048774453403744e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014613887448305467, "output_power": 0.0014427408522058333, "reference_active": true, "delta_db": -0.05577428902284898}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0008261048603288555, "output_power": 0.0003418254432337735, "reference_active": true, "delta_db": -3.832307917028323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 0.00047302858802070836, "output_power": 0.0001068179121251522, "reference_active": true, "delta_db": -6.4624330366433025}], "active_band_count": 3, "active_band_mae_db": 3.450171747564825}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00037792225909995443, "log_rms_mae_db": 1.6231904502469203, "rms_envelope_nrmse": 0.2961638950850278, "envelope_correlation": 0.906704897215512, "rms_gain_error_db": 0.7323112191722541, "reference_mean_power": 0.01079168973829664, "output_mean_power": 0.01277381354939208}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 3.4667222612961444e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016650159461876711, "output_power": 0.0013144225099781157, "reference_active": true, "delta_db": -1.0268340924011057}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0005334276556675961, "output_power": 0.0001811723359177342, "reference_active": true, "delta_db": -4.68983644151776}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.906803616368198e-05, "output_power": 8.213632114527662e-06, "reference_active": true, "delta_db": -6.772863339877104}], "active_band_count": 3, "active_band_mae_db": 4.16317795793199}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00018013905998130842, "log_rms_mae_db": 2.175434233802299, "rms_envelope_nrmse": 0.2843112005460939, "envelope_correlation": 0.9608831906659929, "rms_gain_error_db": 1.4400587197195196, "reference_mean_power": 0.002043548171012202, "output_mean_power": 0.002847021530098794}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.1637162379955263e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006991910792272859, "output_power": 0.0009299960661928542, "reference_active": true, "delta_db": 1.2388523290123326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001302332633123258, "output_power": 0.00024762234332832754, "reference_active": true, "delta_db": 2.7906790614304184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.350140195021887e-05, "output_power": 6.62148104257528e-05, "reference_active": true, "delta_db": 2.9589215843601258}], "active_band_count": 3, "active_band_mae_db": 2.3294843249342922}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00014142899773968785, "log_rms_mae_db": 1.9555984745242934, "rms_envelope_nrmse": 0.267994655868643, "envelope_correlation": 0.9649760637976114, "rms_gain_error_db": 1.4400423942388834, "reference_mean_power": 0.002061256155357224, "output_mean_power": 0.002871681067612237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.0272195933213497e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006464643086637218, "output_power": 0.0008790001988276612, "reference_active": true, "delta_db": 1.3344442085825472}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.880640785962183e-05, "output_power": 0.00014414416382657104, "reference_active": true, "delta_db": 3.211681777189155}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.140839173745677e-06, "output_power": 4.695542426829919e-06, "reference_active": true, "delta_db": 3.4110172597543142}], "active_band_count": 3, "active_band_mae_db": 2.6523810818420057}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 0.00022296985761207562, "log_rms_mae_db": 1.308752954682663, "rms_envelope_nrmse": 0.18306563090877778, "envelope_correlation": 0.9940736428675894, "rms_gain_error_db": 1.2273223955797943, "reference_mean_power": 0.0018916055464765064, "output_mean_power": 0.0025093591182853273}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 1.550199090959336e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001029129669204316, "output_power": 0.001333539517880124, "reference_active": true, "delta_db": 1.125357911881186}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.51753044024429e-05, "output_power": 8.685200931711944e-05, "reference_active": true, "delta_db": 1.9703513243636903}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2165950116970791e-05, "output_power": 1.4036825194446153e-05, "reference_active": true, "delta_db": 0.6212286009341329}], "active_band_count": 3, "active_band_mae_db": 1.2389792790596699}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 0.00021618630891631995, "log_rms_mae_db": 1.0399261843318819, "rms_envelope_nrmse": 0.1662866277597288, "envelope_correlation": 0.9973241773457772, "rms_gain_error_db": 1.2273357309463946, "reference_mean_power": 0.0019079864948334626, "output_mean_power": 0.002531097468348412}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 1.466045308126094e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009690551708671025, "output_power": 0.0012627269779987356, "reference_active": true, "delta_db": 1.149609559272014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151940036741328e-05, "output_power": 3.007196730646507e-05, "reference_active": true, "delta_db": 1.4533167489297325}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.641667021707976e-07, "output_power": 7.727616636983598e-07, "reference_active": true, "delta_db": 0.048574590196651043}], "active_band_count": 3, "active_band_mae_db": 0.8838336327994658}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 698, "rms_floor": 3.608035305023674e-05, "log_rms_mae_db": 1.3791169088629263, "rms_envelope_nrmse": 0.23860465722679713, "envelope_correlation": 0.9630393616875143, "rms_gain_error_db": 0.7392850792078404, "reference_mean_power": 6.674409144257214e-05, "output_mean_power": 7.91300305689336e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 698, "frequency_resolution_hz": 0.28653295128939826, "power_floor": 5.317396933477194e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.505970371058983e-05, "output_power": 4.0152537433469924e-05, "reference_active": true, "delta_db": 0.5890475410295456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.310128917388118e-06, "output_power": 3.3430383780816642e-06, "reference_active": true, "delta_db": 1.6050514582895405}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.893247149634755e-07, "output_power": 1.3949060556070946e-06, "reference_active": true, "delta_db": 1.4920610104328094}], "active_band_count": 3, "active_band_mae_db": 1.2287200032506318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 55920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 692, "rms_floor": 2.62382889528218e-05, "log_rms_mae_db": 0.9562223537877983, "rms_envelope_nrmse": 0.1560013621006151, "envelope_correlation": 0.9862985805026563, "rms_gain_error_db": 0.7395731154991202, "reference_mean_power": 6.730512481108481e-05, "output_mean_power": 7.980046931648063e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 692, "frequency_resolution_hz": 0.28901734104046245, "power_floor": 4.920971050071525e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.323536847986801e-05, "output_power": 3.8431790344941164e-05, "reference_active": true, "delta_db": 0.6309011839279349}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2494288246616935e-06, "output_power": 1.6255519460534348e-06, "reference_active": true, "delta_db": 1.1428933132876171}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.101103941251684e-08, "output_power": 9.228521856411949e-08, "reference_active": true, "delta_db": 0.5658794069488977}], "active_band_count": 3, "active_band_mae_db": 0.7798913013881498}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9059353080695093, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "12fedaf5125e4197b5c032dee7805ab5847564d6edd5696cf02ccd3f50887fdc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9940736428675894, "standard_deviation": null, "minimum": 0.9940736428675894, "maximum": 0.9940736428675894, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.308752954682663, "standard_deviation": null, "minimum": 1.308752954682663, "maximum": 1.308752954682663, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2389792790596699, "standard_deviation": null, "minimum": 1.2389792790596699, "maximum": 1.2389792790596699, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18306563090877778, "standard_deviation": null, "minimum": 0.18306563090877778, "maximum": 0.18306563090877778, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9973241773457772, "standard_deviation": null, "minimum": 0.9973241773457772, "maximum": 0.9973241773457772, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0399261843318819, "standard_deviation": null, "minimum": 1.0399261843318819, "maximum": 1.0399261843318819, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8838336327994658, "standard_deviation": null, "minimum": 0.8838336327994658, "maximum": 0.8838336327994658, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1662866277597288, "standard_deviation": null, "minimum": 0.1662866277597288, "maximum": 0.1662866277597288, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9630393616875143, "standard_deviation": null, "minimum": 0.9630393616875143, "maximum": 0.9630393616875143, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3791169088629263, "standard_deviation": null, "minimum": 1.3791169088629263, "maximum": 1.3791169088629263, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2287200032506318, "standard_deviation": null, "minimum": 1.2287200032506318, "maximum": 1.2287200032506318, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23860465722679713, "standard_deviation": null, "minimum": 0.23860465722679713, "maximum": 0.23860465722679713, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9862985805026563, "standard_deviation": null, "minimum": 0.9862985805026563, "maximum": 0.9862985805026563, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9562223537877983, "standard_deviation": null, "minimum": 0.9562223537877983, "maximum": 0.9562223537877983, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7798913013881498, "standard_deviation": null, "minimum": 0.7798913013881498, "maximum": 0.7798913013881498, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1560013621006151, "standard_deviation": null, "minimum": 0.1560013621006151, "maximum": 0.1560013621006151, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9608831906659929, "standard_deviation": null, "minimum": 0.9608831906659929, "maximum": 0.9608831906659929, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.175434233802299, "standard_deviation": null, "minimum": 2.175434233802299, "maximum": 2.175434233802299, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.3294843249342922, "standard_deviation": null, "minimum": 2.3294843249342922, "maximum": 2.3294843249342922, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2843112005460939, "standard_deviation": null, "minimum": 0.2843112005460939, "maximum": 0.2843112005460939, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9649760637976114, "standard_deviation": null, "minimum": 0.9649760637976114, "maximum": 0.9649760637976114, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.9555984745242934, "standard_deviation": null, "minimum": 1.9555984745242934, "maximum": 1.9555984745242934, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 2.6523810818420057, "standard_deviation": null, "minimum": 2.6523810818420057, "maximum": 2.6523810818420057, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.267994655868643, "standard_deviation": null, "minimum": 0.267994655868643, "maximum": 0.267994655868643, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 2.1640586227983425, "standard_deviation": null, "minimum": 2.1640586227983425, "maximum": 2.1640586227983425, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.778873252690896, "standard_deviation": null, "minimum": 5.778873252690896, "maximum": 5.778873252690896, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.956721634957942, "standard_deviation": null, "minimum": 0.956721634957942, "maximum": 0.956721634957942, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7322980505933767, "standard_deviation": null, "minimum": 0.7322980505933767, "maximum": 0.7322980505933767, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9072659979201808, "standard_deviation": null, "minimum": 0.9072659979201808, "maximum": 0.9072659979201808, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8122634033405418, "standard_deviation": null, "minimum": 1.8122634033405418, "maximum": 1.8122634033405418, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 3.450171747564825, "standard_deviation": null, "minimum": 3.450171747564825, "maximum": 3.450171747564825, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.30828250753217445, "standard_deviation": null, "minimum": 0.30828250753217445, "maximum": 0.30828250753217445, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.906704897215512, "standard_deviation": null, "minimum": 0.906704897215512, "maximum": 0.906704897215512, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.6231904502469203, "standard_deviation": null, "minimum": 1.6231904502469203, "maximum": 1.6231904502469203, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 4.16317795793199, "standard_deviation": null, "minimum": 4.16317795793199, "maximum": 4.16317795793199, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2961638950850278, "standard_deviation": null, "minimum": 0.2961638950850278, "maximum": 0.2961638950850278, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3575"}, {"case": "librispeech_3575-170457-0041", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.993044137954712, "predicted_mos": 2.993044137954712, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6acfb6d158570b5230ae3cd085f3f7e7e62e398e88cab4264c993ac2955a6f21", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.428234599996358}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.979089348078164, "全帯域_10ms_rms_envelope_nrmse": 0.18485132654877726, "全帯域_10ms_envelope_correlation": 0.9847062791577162, "全帯域_10ms_modulation_mae_db": 1.1807959127350547, "全帯域_40ms_log_rms_mae_db": 1.7286106273027757, "全帯域_40ms_rms_envelope_nrmse": 0.16577391957846443, "全帯域_40ms_envelope_correlation": 0.9916391313749376, "全帯域_40ms_modulation_mae_db": 1.0534873981586075, "300–1000 Hz_10ms_log_rms_mae_db": 2.1332106689051984, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2312380112732935, "300–1000 Hz_10ms_envelope_correlation": 0.9864291529695415, "300–1000 Hz_10ms_modulation_mae_db": 1.789303808233292, "300–1000 Hz_40ms_log_rms_mae_db": 1.7814051019113375, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.21397138714215375, "300–1000 Hz_40ms_envelope_correlation": 0.9911659047795779, "300–1000 Hz_40ms_modulation_mae_db": 1.784175771076076, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3242010439156915, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1697292885605568, "1000–3000 Hz_10ms_envelope_correlation": 0.9893824816993092, "1000–3000 Hz_10ms_modulation_mae_db": 0.7605771621278089, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9139962663815533, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13883453532075107, "1000–3000 Hz_40ms_envelope_correlation": 0.9955451059448319, "1000–3000 Hz_40ms_modulation_mae_db": 0.6903689635263236, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5438490045804965, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24031566644658883, "3000–6000 Hz_10ms_envelope_correlation": 0.9680664773585538, "3000–6000 Hz_10ms_modulation_mae_db": 1.7159502879997206, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0202642765032701, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1563029750069277, "3000–6000 Hz_40ms_envelope_correlation": 0.9881961018600082, "3000–6000 Hz_40ms_modulation_mae_db": 1.4581776754024574, "mr_stft_db": 5.868275168279369, "band_mae_db": 1.0273752926542723, "rms_gain_error_db": 1.1680577651286246, "output_peak": 0.784122717392902}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.046725255081971, "spectral_convergence": 0.49124969320177153, "spectral_convergence_status": "ok", "amplitude_floor": 1.6781522163747868e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.547238574136857, "spectral_convergence": 0.3276779565486003, "spectral_convergence_status": "ok", "amplitude_floor": 1.6368745556803986e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.010861675619278, "spectral_convergence": 0.37115682101068986, "spectral_convergence_status": "ok", "amplitude_floor": 1.5514964797831996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006090282805006308, "output_power": 0.0008591993520403531, "delta_db": 1.4945648107557672, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005054165098664251, "output_power": 0.0063384429084815085, "delta_db": 0.9833315849223246, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0022275244303280044, "output_power": 0.0032770185603551213, "delta_db": 1.6765642529267508, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006076310340369887, "output_power": 0.0007274666401615393, "delta_db": 0.7817313559863969, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001299593081283461, "output_power": 0.0001580401587445104, "delta_db": 0.8496006663844771, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0003180661511978585, "output_power": 0.0003470270769686967, "delta_db": 0.3784590849499169, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.0273752926542723, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1680577651286246, "pitch": null, "output_rms_dbfs": -19.31547159041923, "output_peak": 0.784122717392902, "reference_rms_dbfs": -20.48352935554785, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.868275168279369}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002515751859318019, "log_rms_mae_db": 1.979089348078164, "rms_envelope_nrmse": 0.18485132654877726, "envelope_correlation": 0.9847062791577162, "rms_gain_error_db": 1.168068095883244, "reference_mean_power": 0.008953808913712473, "output_mean_power": 0.011716951474589567}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.40642035802602e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010131590544257714, "output_power": 0.001316018452618992, "reference_active": true, "delta_db": 1.1358434880552777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021577628078402175, "output_power": 0.0002878864001781302, "reference_active": true, "delta_db": 1.2521744605531673}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.554717092269403e-05, "output_power": 5.9415312505699086e-05, "reference_active": true, "delta_db": 1.154369789596719}], "active_band_count": 3, "active_band_mae_db": 1.1807959127350547}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002371340103862925, "log_rms_mae_db": 1.7286106273027757, "rms_envelope_nrmse": 0.16577391957846443, "envelope_correlation": 0.9916391313749376, "rms_gain_error_db": 1.1677383314308694, "reference_mean_power": 0.008997361160684636, "output_mean_power": 0.011773049956783308}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.083478820328128e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000878683378715502, "output_power": 0.0011552088906887695, "reference_active": true, "delta_db": 1.1882811133406155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.731034174913729e-05, "output_power": 0.00013880143085608844, "reference_active": true, "delta_db": 1.5423494535366489}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4907380289326384e-06, "output_power": 2.749863993487852e-06, "reference_active": true, "delta_db": 0.42983162759855803}], "active_band_count": 3, "active_band_mae_db": 1.0534873981586075}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019737503490519328, "log_rms_mae_db": 2.1332106689051984, "rms_envelope_nrmse": 0.2312380112732935, "envelope_correlation": 0.9864291529695415, "rms_gain_error_db": 1.4708054348023023, "reference_mean_power": 0.0030829546180243575, "output_mean_power": 0.004325613134821605}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6914511369831456e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008052884179451131, "output_power": 0.0011661279298275874, "reference_active": true, "delta_db": 1.6079474432439567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019776686451848026, "output_power": 0.0003326599139545849, "reference_active": true, "delta_db": 2.258469427511552}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9868777263914452e-05, "output_power": 2.807505376268605e-05, "reference_active": true, "delta_db": 1.501494553944367}], "active_band_count": 3, "active_band_mae_db": 1.789303808233292}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017438355068165795, "log_rms_mae_db": 1.7814051019113375, "rms_envelope_nrmse": 0.21397138714215375, "envelope_correlation": 0.9911659047795779, "rms_gain_error_db": 1.4704231946280502, "reference_mean_power": 0.0030981947388459344, "output_mean_power": 0.004346613570545638}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5303812378070619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007441051144552582, "output_power": 0.0010940406519149726, "reference_active": true, "delta_db": 1.673991700512614}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.887152997212749e-05, "output_power": 0.0001736605025350366, "reference_active": true, "delta_db": 2.4462979880991647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5498600424852708e-06, "output_power": 2.0583358782261288e-06, "reference_active": true, "delta_db": 1.2322376246164493}], "active_band_count": 3, "active_band_mae_db": 1.784175771076076}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011259893106980884, "log_rms_mae_db": 1.3242010439156915, "rms_envelope_nrmse": 0.1697292885605568, "envelope_correlation": 0.9893824816993092, "rms_gain_error_db": 0.9332782302048419, "reference_mean_power": 0.0006356182176644616, "output_mean_power": 0.0007879962653581093}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.9581273226652876e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002027625859711201, "output_power": 0.00023281695345951108, "reference_active": true, "delta_db": 0.6002678050414564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.693176540437241e-05, "output_power": 3.63001691456189e-05, "reference_active": true, "delta_db": 1.2964382590983183}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.128022173574573e-06, "output_power": 6.523287203264924e-06, "reference_active": true, "delta_db": -0.3850254222436516}], "active_band_count": 3, "active_band_mae_db": 0.7605771621278089}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001021231893168278, "log_rms_mae_db": 0.9139962663815533, "rms_envelope_nrmse": 0.13883453532075107, "envelope_correlation": 0.9955451059448319, "rms_gain_error_db": 0.9332590652010966, "reference_mean_power": 0.0006388062286465308, "output_mean_power": 0.0007919450499088099}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.69069653778161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018976599367877858, "output_power": 0.00021870662523074518, "reference_active": true, "delta_db": 0.6164355031362155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3827689740852905e-05, "output_power": 1.819953426504475e-05, "reference_active": true, "delta_db": 1.193106478337484}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.22079217642337e-07, "output_power": 3.9740879844484835e-07, "reference_active": true, "delta_db": -0.2615649091052714}], "active_band_count": 3, "active_band_mae_db": 0.6903689635263236}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.113996831641667e-05, "log_rms_mae_db": 1.5438490045804965, "rms_envelope_nrmse": 0.24031566644658883, "envelope_correlation": 0.9680664773585538, "rms_gain_error_db": 0.6780891778414716, "reference_mean_power": 0.0001805297653048898, "output_mean_power": 0.00021103657763569349}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6198199988921157e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.959931264840548e-05, "output_power": 0.00010785936549017763, "reference_active": true, "delta_db": 0.34601519818394477}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361167220741e-06, "output_power": 1.1883631915934841e-05, "reference_active": true, "delta_db": 1.0361918826920915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4041286689847167e-06, "output_power": 5.7216605097320825e-06, "reference_active": true, "delta_db": 3.765643783123126}], "active_band_count": 3, "active_band_mae_db": 1.7159502879997206}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.456026972891915e-05, "log_rms_mae_db": 1.0202642765032701, "rms_envelope_nrmse": 0.1563029750069277, "envelope_correlation": 0.9881961018600082, "rms_gain_error_db": 0.6780896478865494, "reference_mean_power": 0.00018143628465930542, "output_mean_power": 0.00021209630806828385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.558011664744216e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.473019444861243e-05, "output_power": 0.00010521235673097116, "reference_active": true, "delta_db": 0.4557832003615164}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179317052925175e-06, "output_power": 5.154474733715832e-06, "reference_active": true, "delta_db": 0.9107909612921594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8116784694632823e-07, "output_power": 3.621404271619951e-07, "reference_active": true, "delta_db": 3.0079588645536965}], "active_band_count": 3, "active_band_mae_db": 1.4581776754024574}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8813609392285575, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e7f16b985dc8c777b164c2785521c524a5365d4cf3ad52440cd5c4795319cc34", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9893824816993092, "standard_deviation": null, "minimum": 0.9893824816993092, "maximum": 0.9893824816993092, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3242010439156915, "standard_deviation": null, "minimum": 1.3242010439156915, "maximum": 1.3242010439156915, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7605771621278089, "standard_deviation": null, "minimum": 0.7605771621278089, "maximum": 0.7605771621278089, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1697292885605568, "standard_deviation": null, "minimum": 0.1697292885605568, "maximum": 0.1697292885605568, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955451059448319, "standard_deviation": null, "minimum": 0.9955451059448319, "maximum": 0.9955451059448319, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9139962663815533, "standard_deviation": null, "minimum": 0.9139962663815533, "maximum": 0.9139962663815533, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6903689635263236, "standard_deviation": null, "minimum": 0.6903689635263236, "maximum": 0.6903689635263236, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13883453532075107, "standard_deviation": null, "minimum": 0.13883453532075107, "maximum": 0.13883453532075107, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9680664773585538, "standard_deviation": null, "minimum": 0.9680664773585538, "maximum": 0.9680664773585538, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5438490045804965, "standard_deviation": null, "minimum": 1.5438490045804965, "maximum": 1.5438490045804965, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7159502879997206, "standard_deviation": null, "minimum": 1.7159502879997206, "maximum": 1.7159502879997206, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24031566644658883, "standard_deviation": null, "minimum": 0.24031566644658883, "maximum": 0.24031566644658883, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9881961018600082, "standard_deviation": null, "minimum": 0.9881961018600082, "maximum": 0.9881961018600082, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0202642765032701, "standard_deviation": null, "minimum": 1.0202642765032701, "maximum": 1.0202642765032701, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4581776754024574, "standard_deviation": null, "minimum": 1.4581776754024574, "maximum": 1.4581776754024574, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1563029750069277, "standard_deviation": null, "minimum": 0.1563029750069277, "maximum": 0.1563029750069277, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9864291529695415, "standard_deviation": null, "minimum": 0.9864291529695415, "maximum": 0.9864291529695415, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1332106689051984, "standard_deviation": null, "minimum": 2.1332106689051984, "maximum": 2.1332106689051984, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.789303808233292, "standard_deviation": null, "minimum": 1.789303808233292, "maximum": 1.789303808233292, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2312380112732935, "standard_deviation": null, "minimum": 0.2312380112732935, "maximum": 0.2312380112732935, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9911659047795779, "standard_deviation": null, "minimum": 0.9911659047795779, "maximum": 0.9911659047795779, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7814051019113375, "standard_deviation": null, "minimum": 1.7814051019113375, "maximum": 1.7814051019113375, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.784175771076076, "standard_deviation": null, "minimum": 1.784175771076076, "maximum": 1.784175771076076, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.21397138714215375, "standard_deviation": null, "minimum": 0.21397138714215375, "maximum": 0.21397138714215375, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.0273752926542723, "standard_deviation": null, "minimum": 1.0273752926542723, "maximum": 1.0273752926542723, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.868275168279369, "standard_deviation": null, "minimum": 5.868275168279369, "maximum": 5.868275168279369, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.784122717392902, "standard_deviation": null, "minimum": 0.784122717392902, "maximum": 0.784122717392902, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1680577651286246, "standard_deviation": null, "minimum": 1.1680577651286246, "maximum": 1.1680577651286246, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9847062791577162, "standard_deviation": null, "minimum": 0.9847062791577162, "maximum": 0.9847062791577162, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.979089348078164, "standard_deviation": null, "minimum": 1.979089348078164, "maximum": 1.979089348078164, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1807959127350547, "standard_deviation": null, "minimum": 1.1807959127350547, "maximum": 1.1807959127350547, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18485132654877726, "standard_deviation": null, "minimum": 0.18485132654877726, "maximum": 0.18485132654877726, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9916391313749376, "standard_deviation": null, "minimum": 0.9916391313749376, "maximum": 0.9916391313749376, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.7286106273027757, "standard_deviation": null, "minimum": 1.7286106273027757, "maximum": 1.7286106273027757, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0534873981586075, "standard_deviation": null, "minimum": 1.0534873981586075, "maximum": 1.0534873981586075, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16577391957846443, "standard_deviation": null, "minimum": 0.16577391957846443, "maximum": 0.16577391957846443, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3575"}, {"case": "librispeech_3575-170457-0054", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.4201161861419678, "predicted_mos": 2.4201161861419678, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "635f5b04774fc3df5dd49adf45f74a2183699964adf94060cee7c9779ca77321", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.267539599910378}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.9705587493684367, "全帯域_10ms_rms_envelope_nrmse": 0.24378325448686358, "全帯域_10ms_envelope_correlation": 0.9859645744184432, "全帯域_10ms_modulation_mae_db": 1.1061891653667313, "全帯域_40ms_log_rms_mae_db": 2.6250914956767013, "全帯域_40ms_rms_envelope_nrmse": 0.22277555056938156, "全帯域_40ms_envelope_correlation": 0.9929946488397918, "全帯域_40ms_modulation_mae_db": 0.8733346535162254, "300–1000 Hz_10ms_log_rms_mae_db": 1.8816258808901944, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.3078196151807395, "300–1000 Hz_10ms_envelope_correlation": 0.9844333116723385, "300–1000 Hz_10ms_modulation_mae_db": 1.19706413083946, "300–1000 Hz_40ms_log_rms_mae_db": 1.8118296101350373, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2845532755535066, "300–1000 Hz_40ms_envelope_correlation": 0.991781964362227, "300–1000 Hz_40ms_modulation_mae_db": 0.9193232815874032, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2852810289153564, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17894832924759804, "1000–3000 Hz_10ms_envelope_correlation": 0.9885259510172287, "1000–3000 Hz_10ms_modulation_mae_db": 0.8910152892691464, "1000–3000 Hz_40ms_log_rms_mae_db": 0.92725863110836, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13476814750294572, "1000–3000 Hz_40ms_envelope_correlation": 0.996341371120164, "1000–3000 Hz_40ms_modulation_mae_db": 0.6922849333112892, "3000–6000 Hz_10ms_log_rms_mae_db": 1.678464923379712, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22862482801611142, "3000–6000 Hz_10ms_envelope_correlation": 0.9718334458496704, "3000–6000 Hz_10ms_modulation_mae_db": 0.5172801490514268, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0262452719882358, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13924109777815916, "3000–6000 Hz_40ms_envelope_correlation": 0.9908055138023582, "3000–6000 Hz_40ms_modulation_mae_db": 0.6253325081866133, "mr_stft_db": 6.021655262303998, "band_mae_db": 1.0704370017135185, "rms_gain_error_db": 1.577895854573147, "output_peak": 1.0404653337314103}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.0416666666666666e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.0416666666666666e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.548487128177939, "spectral_convergence": 0.5692491565248073, "spectral_convergence_status": "ok", "amplitude_floor": 1.8779397470586322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.590656251212694, "spectral_convergence": 0.3355208665137138, "spectral_convergence_status": "ok", "amplitude_floor": 1.7710051963816244e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.925822407521361, "spectral_convergence": 0.3763280153019657, "spectral_convergence_status": "ok", "amplitude_floor": 1.1208016909489659e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004732055872254276, "output_power": 0.0005543625990109769, "delta_db": 0.687440584160447, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016454515092618504, "output_power": 0.0022661861485937197, "delta_db": 1.3901049237003236, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001811442607675158, "output_power": 0.0029667864111184387, "delta_db": 2.142617023859622, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004356719801952226, "output_power": 0.0005382047476242241, "delta_db": 0.9178789483166905, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011106135183657726, "output_power": 0.0001310635787210172, "delta_db": 0.7191906689969875, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.173475364573245e-05, "output_power": 7.03183554978773e-05, "delta_db": 0.5653898612470394, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.0704370017135185, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.577895854573147, "pitch": null, "output_rms_dbfs": -21.852915879190107, "output_peak": 1.0404653337314103, "reference_rms_dbfs": -23.430811733763257, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.021655262303998}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027737662806153396, "log_rms_mae_db": 2.9705587493684367, "rms_envelope_nrmse": 0.24378325448686358, "envelope_correlation": 0.9859645744184432, "rms_gain_error_db": 1.5786512664150842, "reference_mean_power": 0.004541375985905238, "output_mean_power": 0.006532096406039662}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.847489317868867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007100074703458764, "output_power": 0.0009171094737177281, "reference_active": true, "delta_db": 1.1115826154606687}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018446902300009587, "output_power": 0.0002239294436104382, "reference_active": true, "delta_db": 0.8418775339907812}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9341715429184444e-05, "output_power": 4.017873280483182e-05, "reference_active": true, "delta_db": 1.3651073466487438}], "active_band_count": 3, "active_band_mae_db": 1.1061891653667313}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002314838901045675, "log_rms_mae_db": 2.6250914956767013, "rms_envelope_nrmse": 0.22277555056938156, "envelope_correlation": 0.9929946488397918, "rms_gain_error_db": 1.5792411472604486, "reference_mean_power": 0.0045632674272876395, "output_mean_power": 0.0065644755604487666}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7204098415154255e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006818437378247786, "output_power": 0.0008762742034870698, "reference_active": true, "delta_db": 1.089551705309892}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.591872232322897e-05, "output_power": 9.403278936985182e-05, "reference_active": true, "delta_db": 0.9293042861754648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0593810422650415e-06, "output_power": 2.365110888022561e-06, "reference_active": true, "delta_db": 0.6011479690633196}], "active_band_count": 3, "active_band_mae_db": 0.8733346535162254}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002609159645197948, "log_rms_mae_db": 1.8816258808901944, "rms_envelope_nrmse": 0.3078196151807395, "envelope_correlation": 0.9844333116723385, "rms_gain_error_db": 1.986191888995856, "reference_mean_power": 0.0022069594688974205, "output_mean_power": 0.003486691686675627}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0859718062152698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048225083122535774, "output_power": 0.0006654315836302269, "reference_active": true, "delta_db": 1.3983042522220606}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00016124782840735576, "output_power": 0.00021755269419620587, "reference_active": true, "delta_db": 1.300705914927166}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2038625150435634e-05, "output_power": 2.7064669039909404e-05, "reference_active": true, "delta_db": 0.8921822253691538}], "active_band_count": 3, "active_band_mae_db": 1.19706413083946}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001981315811908706, "log_rms_mae_db": 1.8118296101350373, "rms_envelope_nrmse": 0.2845532755535066, "envelope_correlation": 0.991781964362227, "rms_gain_error_db": 1.9862404358258052, "reference_mean_power": 0.002217812362061406, "output_mean_power": 0.0035038769288240727}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0257217759681144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004916402889453352, "output_power": 0.0006746061136867949, "reference_active": true, "delta_db": 1.3740280746603482}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.925750439584527e-05, "output_power": 9.427314142217498e-05, "reference_active": true, "delta_db": 1.3392114086719278}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6222165494821603e-06, "output_power": 1.6055942032070188e-06, "reference_active": true, "delta_db": -0.044730361429933405}], "active_band_count": 3, "active_band_mae_db": 0.9193232815874032}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001406989062134312, "log_rms_mae_db": 1.2852810289153564, "rms_envelope_nrmse": 0.17894832924759804, "envelope_correlation": 0.9885259510172287, "rms_gain_error_db": 0.9042488073257587, "reference_mean_power": 0.0005036178738924359, "output_mean_power": 0.0006201917945138979}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.692987420204184e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020569637799863847, "output_power": 0.0002597721670336341, "reference_active": true, "delta_db": 1.0136597279183073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.72045822558081e-05, "output_power": 8.100928634574381e-05, "reference_active": true, "delta_db": 0.8113592039589874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.876518957296052e-06, "output_power": 8.359327460109571e-06, "reference_active": true, "delta_db": 0.8480269359301447}], "active_band_count": 3, "active_band_mae_db": 0.8910152892691464}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012225910578362228, "log_rms_mae_db": 0.92725863110836, "rms_envelope_nrmse": 0.13476814750294572, "envelope_correlation": 0.996341371120164, "rms_gain_error_db": 0.9042860123590958, "reference_mean_power": 0.0005061264034494051, "output_mean_power": 0.000623286320366984}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.5254708519662464e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002159112955909936, "output_power": 0.00027105202711860167, "reference_active": true, "delta_db": 0.9877729616007808}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9510380576050556e-05, "output_power": 3.6611128986850917e-05, "reference_active": true, "delta_db": 0.936383111916922}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.547001199869126e-07, "output_power": 8.251708777561711e-07, "reference_active": true, "delta_db": -0.15269872641616464}], "active_band_count": 3, "active_band_mae_db": 0.6922849333112892}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3844040204199326e-05, "log_rms_mae_db": 1.678464923379712, "rms_envelope_nrmse": 0.22862482801611142, "envelope_correlation": 0.9718334458496704, "rms_gain_error_db": 0.5743888752154287, "reference_mean_power": 4.7417249737341345e-05, "output_mean_power": 5.4122175877223566e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.272212516285404e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.356579821454215e-05, "output_power": 3.561609872497714e-05, "reference_active": true, "delta_db": 0.25749367062362677}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1452306980819945e-06, "output_power": 9.612245199188867e-06, "reference_active": true, "delta_db": 1.288085852899373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.623091261016208e-06, "output_power": 1.6254328470459587e-06, "reference_active": true, "delta_db": 0.006260923631280783}], "active_band_count": 3, "active_band_mae_db": 0.5172801490514268}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.685962479131491e-05, "log_rms_mae_db": 1.0262452719882358, "rms_envelope_nrmse": 0.13924109777815916, "envelope_correlation": 0.9908055138023582, "rms_gain_error_db": 0.574462405458513, "reference_mean_power": 4.7645706194829226e-05, "output_mean_power": 5.438385745316357e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.965193836309832e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.1852605209893166e-05, "output_power": 3.432007860862039e-05, "reference_active": true, "delta_db": 0.32403315040600955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.594578726064393e-06, "output_power": 4.846502784511876e-06, "reference_active": true, "delta_db": 1.297804669135152}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8710953646003778e-07, "output_power": 1.9838639851695628e-07, "reference_active": true, "delta_db": 0.25415970501867846}], "active_band_count": 3, "active_band_mae_db": 0.6253325081866133}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9353423608126337, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "967b2d4ac190bcf6fa6752af150972d99c16e3c8510a0a35432b533f5b875d13", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9885259510172287, "standard_deviation": null, "minimum": 0.9885259510172287, "maximum": 0.9885259510172287, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2852810289153564, "standard_deviation": null, "minimum": 1.2852810289153564, "maximum": 1.2852810289153564, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8910152892691464, "standard_deviation": null, "minimum": 0.8910152892691464, "maximum": 0.8910152892691464, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17894832924759804, "standard_deviation": null, "minimum": 0.17894832924759804, "maximum": 0.17894832924759804, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996341371120164, "standard_deviation": null, "minimum": 0.996341371120164, "maximum": 0.996341371120164, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.92725863110836, "standard_deviation": null, "minimum": 0.92725863110836, "maximum": 0.92725863110836, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6922849333112892, "standard_deviation": null, "minimum": 0.6922849333112892, "maximum": 0.6922849333112892, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13476814750294572, "standard_deviation": null, "minimum": 0.13476814750294572, "maximum": 0.13476814750294572, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9718334458496704, "standard_deviation": null, "minimum": 0.9718334458496704, "maximum": 0.9718334458496704, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.678464923379712, "standard_deviation": null, "minimum": 1.678464923379712, "maximum": 1.678464923379712, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5172801490514268, "standard_deviation": null, "minimum": 0.5172801490514268, "maximum": 0.5172801490514268, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22862482801611142, "standard_deviation": null, "minimum": 0.22862482801611142, "maximum": 0.22862482801611142, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9908055138023582, "standard_deviation": null, "minimum": 0.9908055138023582, "maximum": 0.9908055138023582, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0262452719882358, "standard_deviation": null, "minimum": 1.0262452719882358, "maximum": 1.0262452719882358, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6253325081866133, "standard_deviation": null, "minimum": 0.6253325081866133, "maximum": 0.6253325081866133, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13924109777815916, "standard_deviation": null, "minimum": 0.13924109777815916, "maximum": 0.13924109777815916, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9844333116723385, "standard_deviation": null, "minimum": 0.9844333116723385, "maximum": 0.9844333116723385, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8816258808901944, "standard_deviation": null, "minimum": 1.8816258808901944, "maximum": 1.8816258808901944, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.19706413083946, "standard_deviation": null, "minimum": 1.19706413083946, "maximum": 1.19706413083946, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3078196151807395, "standard_deviation": null, "minimum": 0.3078196151807395, "maximum": 0.3078196151807395, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.991781964362227, "standard_deviation": null, "minimum": 0.991781964362227, "maximum": 0.991781964362227, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.8118296101350373, "standard_deviation": null, "minimum": 1.8118296101350373, "maximum": 1.8118296101350373, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9193232815874032, "standard_deviation": null, "minimum": 0.9193232815874032, "maximum": 0.9193232815874032, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2845532755535066, "standard_deviation": null, "minimum": 0.2845532755535066, "maximum": 0.2845532755535066, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.0704370017135185, "standard_deviation": null, "minimum": 1.0704370017135185, "maximum": 1.0704370017135185, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.021655262303998, "standard_deviation": null, "minimum": 6.021655262303998, "maximum": 6.021655262303998, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.0404653337314103, "standard_deviation": null, "minimum": 1.0404653337314103, "maximum": 1.0404653337314103, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.577895854573147, "standard_deviation": null, "minimum": 1.577895854573147, "maximum": 1.577895854573147, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9859645744184432, "standard_deviation": null, "minimum": 0.9859645744184432, "maximum": 0.9859645744184432, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.9705587493684367, "standard_deviation": null, "minimum": 2.9705587493684367, "maximum": 2.9705587493684367, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1061891653667313, "standard_deviation": null, "minimum": 1.1061891653667313, "maximum": 1.1061891653667313, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.24378325448686358, "standard_deviation": null, "minimum": 0.24378325448686358, "maximum": 0.24378325448686358, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9929946488397918, "standard_deviation": null, "minimum": 0.9929946488397918, "maximum": 0.9929946488397918, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.6250914956767013, "standard_deviation": null, "minimum": 2.6250914956767013, "maximum": 2.6250914956767013, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8733346535162254, "standard_deviation": null, "minimum": 0.8733346535162254, "maximum": 0.8733346535162254, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.22277555056938156, "standard_deviation": null, "minimum": 0.22277555056938156, "maximum": 0.22277555056938156, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3575"}, {"case": "librispeech_3729-6852-0000", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.65273380279541, "predicted_mos": 3.65273380279541, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c8c8babe043f320847aa137c9a4f2c3a2e1cc691605d5c1da1208bcaf904e781", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.657127799931914}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6856352973932243, "全帯域_10ms_rms_envelope_nrmse": 0.2984840196289776, "全帯域_10ms_envelope_correlation": 0.9758199056573514, "全帯域_10ms_modulation_mae_db": 2.546797635226151, "全帯域_40ms_log_rms_mae_db": 1.4353648006842001, "全帯域_40ms_rms_envelope_nrmse": 0.27926284085659947, "全帯域_40ms_envelope_correlation": 0.9831503094249557, "全帯域_40ms_modulation_mae_db": 2.502725191020431, "300–1000 Hz_10ms_log_rms_mae_db": 1.9574674465686506, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.3677407605830239, "300–1000 Hz_10ms_envelope_correlation": 0.9788688306390979, "300–1000 Hz_10ms_modulation_mae_db": 2.837892691620677, "300–1000 Hz_40ms_log_rms_mae_db": 1.6605071420028477, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.3510777232901156, "300–1000 Hz_40ms_envelope_correlation": 0.9852033962744855, "300–1000 Hz_40ms_modulation_mae_db": 2.742425161713204, "1000–3000 Hz_10ms_log_rms_mae_db": 1.357730553315314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.23296486645053643, "1000–3000 Hz_10ms_envelope_correlation": 0.9818587123222057, "1000–3000 Hz_10ms_modulation_mae_db": 1.300981812105834, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9767944258713275, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.19950993205905435, "1000–3000 Hz_40ms_envelope_correlation": 0.9907790961936762, "1000–3000 Hz_40ms_modulation_mae_db": 1.2644360999504716, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5057948497084017, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25614656141277403, "3000–6000 Hz_10ms_envelope_correlation": 0.9524521213114002, "3000–6000 Hz_10ms_modulation_mae_db": 1.0748288376018427, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9339506156357462, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19146604695219968, "3000–6000 Hz_40ms_envelope_correlation": 0.9720627219905409, "3000–6000 Hz_40ms_modulation_mae_db": 1.4842169165032315, "mr_stft_db": 5.858805395420453, "band_mae_db": 1.2687921997965923, "rms_gain_error_db": 1.8440288374632898, "output_peak": 0.5491744251651889}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.195250000965488, "spectral_convergence": 0.5469080241964637, "spectral_convergence_status": "ok", "amplitude_floor": 1.174323222508143e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.310045615572468, "spectral_convergence": 0.3693644015833771, "spectral_convergence_status": "ok", "amplitude_floor": 1.1752576893073956e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.071120569723401, "spectral_convergence": 0.4024277710524309, "spectral_convergence_status": "ok", "amplitude_floor": 9.275961087692859e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.278493760532488e-05, "output_power": 0.00012079646164172753, "delta_db": 1.1457673312407592, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009129851997232178, "output_power": 0.001445840035572686, "delta_db": 1.9965650896679148, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007449145920476126, "output_power": 0.0012419455276689198, "delta_db": 2.2199606618941954, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00040544152038629675, "output_power": 0.0005538442916601808, "delta_db": 1.354594621123043, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011224761509504284, "output_power": 0.00012881097996262557, "delta_db": 0.5977576185363361, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.007390871044891e-05, "output_power": 5.3631788613356165e-05, "delta_db": 0.2981078763173049, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.2687921997965923, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.8440288374632898, "pitch": null, "output_rms_dbfs": -24.503997990309998, "output_peak": 0.5491744251651889, "reference_rms_dbfs": -26.348026827773285, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.858805395420453}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001795198121335632, "log_rms_mae_db": 1.6856352973932243, "rms_envelope_nrmse": 0.2984840196289776, "envelope_correlation": 0.9758199056573514, "rms_gain_error_db": 1.8448481014956821, "reference_mean_power": 0.002319729266015563, "output_mean_power": 0.0035474976120649353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2030894540936354e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042099620393065603, "output_power": 0.0007297444624143919, "reference_active": true, "delta_db": 2.3889262820154213}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010212932714007922, "output_power": 0.00020320422124001455, "reference_active": true, "delta_db": 2.987822548237413}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7753962605616007e-05, "output_power": 2.9899209645446907e-05, "reference_active": true, "delta_db": 2.2636440754256184}], "active_band_count": 3, "active_band_mae_db": 2.546797635226151}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017377061993537507, "log_rms_mae_db": 1.4353648006842001, "rms_envelope_nrmse": 0.27926284085659947, "envelope_correlation": 0.9831503094249557, "rms_gain_error_db": 1.845598247589194, "reference_mean_power": 0.002329667623369792, "output_mean_power": 0.00356331149506431}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1345793651623815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000381470345516612, "output_power": 0.0006758264826425764, "reference_active": true, "delta_db": 2.4837442313423326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.449258478788481e-05, "output_power": 0.00011148278794969887, "reference_active": true, "delta_db": 3.1087041235676467}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.451644818904224e-07, "output_power": 1.4691972772645523e-06, "reference_active": true, "delta_db": 1.9157272181513134}], "active_band_count": 3, "active_band_mae_db": 2.502725191020431}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001705289920117241, "log_rms_mae_db": 1.9574674465686506, "rms_envelope_nrmse": 0.3677407605830239, "envelope_correlation": 0.9788688306390979, "rms_gain_error_db": 2.376056369720064, "reference_mean_power": 0.0014327712363456494, "output_mean_power": 0.0024761815947854453}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.476686730813627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004044339102676846, "output_power": 0.0007390290714248715, "reference_active": true, "delta_db": 2.6181396030257584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.628475293024379e-05, "output_power": 0.0002071693610563465, "reference_active": true, "delta_db": 3.3276800628554732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2048085119397813e-05, "output_power": 2.1762253656536e-05, "reference_active": true, "delta_db": 2.567858408980799}], "active_band_count": 3, "active_band_mae_db": 2.837892691620677}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015393777861692204, "log_rms_mae_db": 1.6605071420028477, "rms_envelope_nrmse": 0.3510777232901156, "envelope_correlation": 0.9852033962744855, "rms_gain_error_db": 2.377423481745268, "reference_mean_power": 0.0014385306131629134, "output_mean_power": 0.0024869179481375146}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.958921932585308e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037465347734378165, "output_power": 0.000695114213249552, "reference_active": true, "delta_db": 2.6842640092121646}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.015607336171049e-05, "output_power": 0.00011067162081378, "reference_active": true, "delta_db": 3.437127414079545}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.018660953269709e-07, "output_power": 1.3022376790895244e-06, "reference_active": true, "delta_db": 2.105884061847903}], "active_band_count": 3, "active_band_mae_db": 2.742425161713204}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.828766409891504e-05, "log_rms_mae_db": 1.357730553315314, "rms_envelope_nrmse": 0.23296486645053643, "envelope_correlation": 0.9818587123222057, "rms_gain_error_db": 1.30179562301053, "reference_mean_power": 0.00046178379763570874, "output_mean_power": 0.0006231868107810439}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.102877053654468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.713589283902264e-05, "output_power": 0.00011209197361823884, "reference_active": true, "delta_db": 1.093774303539015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3909625300627975e-05, "output_power": 1.983427664700959e-05, "reference_active": true, "delta_db": 1.5410093538065004}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.457858491278464e-06, "output_power": 5.969561849031604e-06, "reference_active": true, "delta_db": 1.2681617789719866}], "active_band_count": 3, "active_band_mae_db": 1.300981812105834}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.324411735838957e-05, "log_rms_mae_db": 0.9767944258713275, "rms_envelope_nrmse": 0.19950993205905435, "envelope_correlation": 0.9907790961936762, "rms_gain_error_db": 1.3024272902108405, "reference_mean_power": 0.0004639006426222396, "output_mean_power": 0.0006261345996054742}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.97726981805197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.049771362210352e-05, "output_power": 0.00010392707835015195, "reference_active": true, "delta_db": 1.109451730851771}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.625262232859704e-06, "output_power": 9.558368987061314e-06, "reference_active": true, "delta_db": 1.5918071916992862}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8383435618714896e-07, "output_power": 3.6498070049904553e-07, "reference_active": true, "delta_db": 1.0920493773003572}], "active_band_count": 3, "active_band_mae_db": 1.2644360999504716}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.433462630283978e-05, "log_rms_mae_db": 1.5057948497084017, "rms_envelope_nrmse": 0.25614656141277403, "envelope_correlation": 0.9524521213114002, "rms_gain_error_db": 0.3296959891000348, "reference_mean_power": 4.2981917565965076e-05, "output_mean_power": 4.637195287177901e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5620745474492675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.382029340071467e-06, "output_power": 7.451780718732143e-06, "reference_active": true, "delta_db": 0.6730127011966809}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1967382600492452e-06, "output_power": 1.7726473202863834e-06, "reference_active": true, "delta_db": 1.7062316271872124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059624559400323e-07, "output_power": 6.146706981071684e-07, "reference_active": true, "delta_db": 0.8452421844216347}], "active_band_count": 3, "active_band_mae_db": 1.0748288376018427}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.1780456996026785e-05, "log_rms_mae_db": 0.9339506156357462, "rms_envelope_nrmse": 0.19146604695219968, "envelope_correlation": 0.9720627219905409, "rms_gain_error_db": 0.3296937872003918, "reference_mean_power": 4.317499760725885e-05, "output_mean_power": 4.65802377506554e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4490976093596605e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.797984044318825e-06, "output_power": 6.945063305586818e-06, "reference_active": true, "delta_db": 0.7839919289935504}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.94257411489492e-07, "output_power": 8.818732655706374e-07, "reference_active": true, "delta_db": 1.7143157005829939}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.104930954274029e-08, "output_power": 3.301202510548402e-08, "reference_active": true, "delta_db": 1.9543431199331505}], "active_band_count": 3, "active_band_mae_db": 1.4842169165032315}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9215473373163307, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9b8008e6d4dffe5b629be9eb6135b42810b5a2bfc9eebd0f72a9685969f85e4a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9818587123222057, "standard_deviation": null, "minimum": 0.9818587123222057, "maximum": 0.9818587123222057, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.357730553315314, "standard_deviation": null, "minimum": 1.357730553315314, "maximum": 1.357730553315314, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.300981812105834, "standard_deviation": null, "minimum": 1.300981812105834, "maximum": 1.300981812105834, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23296486645053643, "standard_deviation": null, "minimum": 0.23296486645053643, "maximum": 0.23296486645053643, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9907790961936762, "standard_deviation": null, "minimum": 0.9907790961936762, "maximum": 0.9907790961936762, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9767944258713275, "standard_deviation": null, "minimum": 0.9767944258713275, "maximum": 0.9767944258713275, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2644360999504716, "standard_deviation": null, "minimum": 1.2644360999504716, "maximum": 1.2644360999504716, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19950993205905435, "standard_deviation": null, "minimum": 0.19950993205905435, "maximum": 0.19950993205905435, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9524521213114002, "standard_deviation": null, "minimum": 0.9524521213114002, "maximum": 0.9524521213114002, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5057948497084017, "standard_deviation": null, "minimum": 1.5057948497084017, "maximum": 1.5057948497084017, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0748288376018427, "standard_deviation": null, "minimum": 1.0748288376018427, "maximum": 1.0748288376018427, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25614656141277403, "standard_deviation": null, "minimum": 0.25614656141277403, "maximum": 0.25614656141277403, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9720627219905409, "standard_deviation": null, "minimum": 0.9720627219905409, "maximum": 0.9720627219905409, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9339506156357462, "standard_deviation": null, "minimum": 0.9339506156357462, "maximum": 0.9339506156357462, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4842169165032315, "standard_deviation": null, "minimum": 1.4842169165032315, "maximum": 1.4842169165032315, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19146604695219968, "standard_deviation": null, "minimum": 0.19146604695219968, "maximum": 0.19146604695219968, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9788688306390979, "standard_deviation": null, "minimum": 0.9788688306390979, "maximum": 0.9788688306390979, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9574674465686506, "standard_deviation": null, "minimum": 1.9574674465686506, "maximum": 1.9574674465686506, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.837892691620677, "standard_deviation": null, "minimum": 2.837892691620677, "maximum": 2.837892691620677, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3677407605830239, "standard_deviation": null, "minimum": 0.3677407605830239, "maximum": 0.3677407605830239, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9852033962744855, "standard_deviation": null, "minimum": 0.9852033962744855, "maximum": 0.9852033962744855, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6605071420028477, "standard_deviation": null, "minimum": 1.6605071420028477, "maximum": 1.6605071420028477, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 2.742425161713204, "standard_deviation": null, "minimum": 2.742425161713204, "maximum": 2.742425161713204, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3510777232901156, "standard_deviation": null, "minimum": 0.3510777232901156, "maximum": 0.3510777232901156, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.2687921997965923, "standard_deviation": null, "minimum": 1.2687921997965923, "maximum": 1.2687921997965923, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.858805395420453, "standard_deviation": null, "minimum": 5.858805395420453, "maximum": 5.858805395420453, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5491744251651889, "standard_deviation": null, "minimum": 0.5491744251651889, "maximum": 0.5491744251651889, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.8440288374632898, "standard_deviation": null, "minimum": 1.8440288374632898, "maximum": 1.8440288374632898, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9758199056573514, "standard_deviation": null, "minimum": 0.9758199056573514, "maximum": 0.9758199056573514, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6856352973932243, "standard_deviation": null, "minimum": 1.6856352973932243, "maximum": 1.6856352973932243, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 2.546797635226151, "standard_deviation": null, "minimum": 2.546797635226151, "maximum": 2.546797635226151, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2984840196289776, "standard_deviation": null, "minimum": 0.2984840196289776, "maximum": 0.2984840196289776, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9831503094249557, "standard_deviation": null, "minimum": 0.9831503094249557, "maximum": 0.9831503094249557, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.4353648006842001, "standard_deviation": null, "minimum": 1.4353648006842001, "maximum": 1.4353648006842001, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 2.502725191020431, "standard_deviation": null, "minimum": 2.502725191020431, "maximum": 2.502725191020431, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.27926284085659947, "standard_deviation": null, "minimum": 0.27926284085659947, "maximum": 0.27926284085659947, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3729"}, {"case": "librispeech_3729-6852-0007", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.408066749572754, "predicted_mos": 3.408066749572754, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "d0fb7dd0f814821bbc7a2e528621d40f880081e543742825ed94b962c86ff933", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.456396600231528}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.479049665021457, "全帯域_10ms_rms_envelope_nrmse": 0.25365536319883697, "全帯域_10ms_envelope_correlation": 0.979995048771541, "全帯域_10ms_modulation_mae_db": 1.5843877883822612, "全帯域_40ms_log_rms_mae_db": 1.3213980569103323, "全帯域_40ms_rms_envelope_nrmse": 0.23816703335876543, "全帯域_40ms_envelope_correlation": 0.9859343177112152, "全帯域_40ms_modulation_mae_db": 1.5289601500765218, "300–1000 Hz_10ms_log_rms_mae_db": 1.6926081984342967, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.31406990889966346, "300–1000 Hz_10ms_envelope_correlation": 0.9785957009327035, "300–1000 Hz_10ms_modulation_mae_db": 1.7212884471541587, "300–1000 Hz_40ms_log_rms_mae_db": 1.470558891610803, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.29768701571360856, "300–1000 Hz_40ms_envelope_correlation": 0.9847559892691548, "300–1000 Hz_40ms_modulation_mae_db": 1.6987771897349317, "1000–3000 Hz_10ms_log_rms_mae_db": 1.295044618803617, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19484610989705156, "1000–3000 Hz_10ms_envelope_correlation": 0.9838251639472039, "1000–3000 Hz_10ms_modulation_mae_db": 1.2420292455354949, "1000–3000 Hz_40ms_log_rms_mae_db": 0.95199488937575, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16651290507530403, "1000–3000 Hz_40ms_envelope_correlation": 0.9920833475682405, "1000–3000 Hz_40ms_modulation_mae_db": 1.2238022188558357, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5893599376457566, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21893805317842127, "3000–6000 Hz_10ms_envelope_correlation": 0.9673526045819868, "3000–6000 Hz_10ms_modulation_mae_db": 0.8974982655683225, "3000–6000 Hz_40ms_log_rms_mae_db": 1.037827100497232, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15259367743822366, "3000–6000 Hz_40ms_envelope_correlation": 0.9861906818266203, "3000–6000 Hz_40ms_modulation_mae_db": 0.8353278846565345, "mr_stft_db": 5.707931389094264, "band_mae_db": 1.1686794105364402, "rms_gain_error_db": 1.6141535668039164, "output_peak": 0.4007062807195712}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.060904070308512, "spectral_convergence": 0.535339116470388, "spectral_convergence_status": "ok", "amplitude_floor": 1.0845908458793198e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.051330086941972, "spectral_convergence": 0.32371961798430376, "spectral_convergence_status": "ok", "amplitude_floor": 9.938577161724507e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.011560010032309, "spectral_convergence": 0.3602335586138735, "spectral_convergence_status": "ok", "amplitude_floor": 7.780261052298239e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.1610975793667e-05, "output_power": 9.441532607928389e-05, "delta_db": 0.6329392679203476, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005078113602376632, "output_power": 0.0007286591464208409, "delta_db": 1.5682200847051724, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000931370997501075, "output_power": 0.001455347896751386, "delta_db": 1.9384411252517726, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000354698831360673, "output_power": 0.00046444500748868585, "delta_db": 1.1707454124070091, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.667116250090391e-05, "output_power": 0.0001208333381315895, "delta_db": 0.9688983228206123, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.4087248189368409e-05, "output_power": 1.6676672154881742e-05, "delta_db": 0.7328322501137263, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.1686794105364402, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.6141535668039164, "pitch": null, "output_rms_dbfs": -25.405506072526247, "output_peak": 0.4007062807195712, "reference_rms_dbfs": -27.01965963933016, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.707931389094264}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001511941088699898, "log_rms_mae_db": 1.479049665021457, "rms_envelope_nrmse": 0.25365536319883697, "envelope_correlation": 0.979995048771541, "rms_gain_error_db": 1.615512157258964, "reference_mean_power": 0.0019872702200913737, "output_mean_power": 0.002882757696221966}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.92641346613152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000630498462430925, "output_power": 0.0009282094679809184, "reference_active": true, "delta_db": 1.6796196220824653}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.948331074534803e-05, "output_power": 0.00013292355926771616, "reference_active": true, "delta_db": 2.2332601314920657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3123506088131581e-05, "output_power": 1.5924955895358994e-05, "reference_active": true, "delta_db": 0.8402836115722525}], "active_band_count": 3, "active_band_mae_db": 1.5843877883822612}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013307215895407113, "log_rms_mae_db": 1.3213980569103323, "rms_envelope_nrmse": 0.23816703335876543, "envelope_correlation": 0.9859343177112152, "rms_gain_error_db": 1.6170442273199905, "reference_mean_power": 0.0019932283004473585, "output_mean_power": 0.0028924207436676224}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.959638607230672e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005794424432114469, "output_power": 0.0008609927055230332, "reference_active": true, "delta_db": 1.7198916865886458}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.108494970400102e-05, "output_power": 6.802863406964697e-05, "reference_active": true, "delta_db": 2.190089913063819}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.934091733283841e-07, "output_power": 9.272292056981707e-07, "reference_active": true, "delta_db": 0.6768988505771005}], "active_band_count": 3, "active_band_mae_db": 1.5289601500765218}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014956359127228178, "log_rms_mae_db": 1.6926081984342967, "rms_envelope_nrmse": 0.31406990889966346, "envelope_correlation": 0.9785957009327035, "rms_gain_error_db": 2.0148476299498554, "reference_mean_power": 0.001265981682968443, "output_mean_power": 0.002013317113274719}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.65319874591582e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005127917421071942, "output_power": 0.0008079508264415443, "reference_active": true, "delta_db": 1.9744390674043852}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.49871434628785e-05, "output_power": 0.00012823095596210953, "reference_active": true, "delta_db": 2.3300606991101165}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1228212970044594e-05, "output_power": 1.3685075538958859e-05, "reference_active": true, "delta_db": 0.859365574947975}], "active_band_count": 3, "active_band_mae_db": 1.7212884471541587}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012969159209010741, "log_rms_mae_db": 1.470558891610803, "rms_envelope_nrmse": 0.29768701571360856, "envelope_correlation": 0.9847559892691548, "rms_gain_error_db": 2.016091444264631, "reference_mean_power": 0.0012704534730935044, "output_mean_power": 0.0020210074257980743}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.90499086389905e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004798268169104879, "output_power": 0.0007618310570683383, "reference_active": true, "delta_db": 2.0077415680854958}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7762549535694356e-05, "output_power": 6.375879035473355e-05, "reference_active": true, "delta_db": 2.274787609641754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.708330714477498e-07, "output_power": 9.296951776949121e-07, "reference_active": true, "delta_db": 0.8138023914775451}], "active_band_count": 3, "active_band_mae_db": 1.6987771897349317}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.983632889928613e-05, "log_rms_mae_db": 1.295044618803617, "rms_envelope_nrmse": 0.19484610989705156, "envelope_correlation": 0.9838251639472039, "rms_gain_error_db": 1.1215682972739416, "reference_mean_power": 0.00040681650183919396, "output_mean_power": 0.0005266903857392344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8313223654566366e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016437493457690703, "output_power": 0.00021544418250807646, "reference_active": true, "delta_db": 1.1749917864609363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7703924674415462e-05, "output_power": 3.909285887613087e-05, "reference_active": true, "delta_db": 1.4955613400168255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.687130809571553e-06, "output_power": 5.97668731287768e-06, "reference_active": true, "delta_db": 1.055534610128723}], "active_band_count": 3, "active_band_mae_db": 1.2420292455354949}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.811887222434066e-05, "log_rms_mae_db": 0.95199488937575, "rms_envelope_nrmse": 0.16651290507530403, "envelope_correlation": 0.9920833475682405, "rms_gain_error_db": 1.1237361859018138, "reference_mean_power": 0.00040698573595692556, "output_mean_power": 0.0005271725725339543}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.593242154160197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00015388869283670126, "output_power": 0.00020354549053604211, "reference_active": true, "delta_db": 1.2145477453607745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4161267354242513e-05, "output_power": 1.9597661005878385e-05, "reference_active": true, "delta_db": 1.4110211908796058}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7920568451213675e-07, "output_power": 3.5522868498434115e-07, "reference_active": true, "delta_db": 1.045837720327127}], "active_band_count": 3, "active_band_mae_db": 1.2238022188558357}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4474245023095566e-05, "log_rms_mae_db": 1.5893599376457566, "rms_envelope_nrmse": 0.21893805317842127, "envelope_correlation": 0.9673526045819868, "rms_gain_error_db": 0.7330697629841101, "reference_mean_power": 1.6116938837494647e-05, "output_mean_power": 1.9080490466177167e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.1253846819517355e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.226560638885928e-06, "output_power": 2.915254647319224e-06, "reference_active": true, "delta_db": 1.1704196900171269}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.434590024496585e-06, "output_power": 1.712448894391728e-06, "reference_active": true, "delta_db": 0.7688981283655016}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.807607987621042e-07, "output_power": 5.718034164690079e-07, "reference_active": true, "delta_db": 0.7531769783223387}], "active_band_count": 3, "active_band_mae_db": 0.8974982655683225}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.920336454108668e-05, "log_rms_mae_db": 1.037827100497232, "rms_envelope_nrmse": 0.15259367743822366, "envelope_correlation": 0.9861906818266203, "rms_gain_error_db": 0.7332109407377096, "reference_mean_power": 1.619697385840346e-05, "output_mean_power": 1.9175865519365145e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.254945521069739e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4889493401597257e-06, "output_power": 3.227283242041197e-06, "reference_active": true, "delta_db": 1.1282102575862913}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.64830026091652e-07, "output_power": 9.170429307028025e-07, "reference_active": true, "delta_db": 0.7882473796909144}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8106205193126903e-08, "output_power": 5.510026671347372e-08, "reference_active": true, "delta_db": 0.5895260166923977}], "active_band_count": 3, "active_band_mae_db": 0.8353278846565345}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9378153469436933, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9e6d423968676be386ceed08d85540bb2e4c9be99cecae9a829290e5f0ae36b1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9838251639472039, "standard_deviation": null, "minimum": 0.9838251639472039, "maximum": 0.9838251639472039, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.295044618803617, "standard_deviation": null, "minimum": 1.295044618803617, "maximum": 1.295044618803617, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2420292455354949, "standard_deviation": null, "minimum": 1.2420292455354949, "maximum": 1.2420292455354949, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19484610989705156, "standard_deviation": null, "minimum": 0.19484610989705156, "maximum": 0.19484610989705156, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9920833475682405, "standard_deviation": null, "minimum": 0.9920833475682405, "maximum": 0.9920833475682405, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.95199488937575, "standard_deviation": null, "minimum": 0.95199488937575, "maximum": 0.95199488937575, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.2238022188558357, "standard_deviation": null, "minimum": 1.2238022188558357, "maximum": 1.2238022188558357, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16651290507530403, "standard_deviation": null, "minimum": 0.16651290507530403, "maximum": 0.16651290507530403, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9673526045819868, "standard_deviation": null, "minimum": 0.9673526045819868, "maximum": 0.9673526045819868, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5893599376457566, "standard_deviation": null, "minimum": 1.5893599376457566, "maximum": 1.5893599376457566, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8974982655683225, "standard_deviation": null, "minimum": 0.8974982655683225, "maximum": 0.8974982655683225, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21893805317842127, "standard_deviation": null, "minimum": 0.21893805317842127, "maximum": 0.21893805317842127, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9861906818266203, "standard_deviation": null, "minimum": 0.9861906818266203, "maximum": 0.9861906818266203, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.037827100497232, "standard_deviation": null, "minimum": 1.037827100497232, "maximum": 1.037827100497232, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8353278846565345, "standard_deviation": null, "minimum": 0.8353278846565345, "maximum": 0.8353278846565345, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15259367743822366, "standard_deviation": null, "minimum": 0.15259367743822366, "maximum": 0.15259367743822366, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9785957009327035, "standard_deviation": null, "minimum": 0.9785957009327035, "maximum": 0.9785957009327035, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6926081984342967, "standard_deviation": null, "minimum": 1.6926081984342967, "maximum": 1.6926081984342967, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.7212884471541587, "standard_deviation": null, "minimum": 1.7212884471541587, "maximum": 1.7212884471541587, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.31406990889966346, "standard_deviation": null, "minimum": 0.31406990889966346, "maximum": 0.31406990889966346, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9847559892691548, "standard_deviation": null, "minimum": 0.9847559892691548, "maximum": 0.9847559892691548, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.470558891610803, "standard_deviation": null, "minimum": 1.470558891610803, "maximum": 1.470558891610803, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.6987771897349317, "standard_deviation": null, "minimum": 1.6987771897349317, "maximum": 1.6987771897349317, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.29768701571360856, "standard_deviation": null, "minimum": 0.29768701571360856, "maximum": 0.29768701571360856, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.1686794105364402, "standard_deviation": null, "minimum": 1.1686794105364402, "maximum": 1.1686794105364402, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.707931389094264, "standard_deviation": null, "minimum": 5.707931389094264, "maximum": 5.707931389094264, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4007062807195712, "standard_deviation": null, "minimum": 0.4007062807195712, "maximum": 0.4007062807195712, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.6141535668039164, "standard_deviation": null, "minimum": 1.6141535668039164, "maximum": 1.6141535668039164, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.979995048771541, "standard_deviation": null, "minimum": 0.979995048771541, "maximum": 0.979995048771541, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.479049665021457, "standard_deviation": null, "minimum": 1.479049665021457, "maximum": 1.479049665021457, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.5843877883822612, "standard_deviation": null, "minimum": 1.5843877883822612, "maximum": 1.5843877883822612, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.25365536319883697, "standard_deviation": null, "minimum": 0.25365536319883697, "maximum": 0.25365536319883697, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9859343177112152, "standard_deviation": null, "minimum": 0.9859343177112152, "maximum": 0.9859343177112152, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.3213980569103323, "standard_deviation": null, "minimum": 1.3213980569103323, "maximum": 1.3213980569103323, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.5289601500765218, "standard_deviation": null, "minimum": 1.5289601500765218, "maximum": 1.5289601500765218, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.23816703335876543, "standard_deviation": null, "minimum": 0.23816703335876543, "maximum": 0.23816703335876543, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3729"}, {"case": "librispeech_3729-6852-0017", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.350425958633423, "predicted_mos": 3.350425958633423, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "620d489db1b7023c9aadb66be60226a6ff7ab539ea5f62de17730accdc1ea607", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.07194249983877}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.434597280008415, "全帯域_10ms_rms_envelope_nrmse": 0.26207129200183316, "全帯域_10ms_envelope_correlation": 0.9792540118743017, "全帯域_10ms_modulation_mae_db": 1.9969453791353875, "全帯域_40ms_log_rms_mae_db": 1.1910336194213043, "全帯域_40ms_rms_envelope_nrmse": 0.24413551941178216, "全帯域_40ms_envelope_correlation": 0.9862319552021894, "全帯域_40ms_modulation_mae_db": 1.848892853475183, "300–1000 Hz_10ms_log_rms_mae_db": 1.8482231098708484, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.33230370411182114, "300–1000 Hz_10ms_envelope_correlation": 0.9788016426824935, "300–1000 Hz_10ms_modulation_mae_db": 2.1002044372814743, "300–1000 Hz_40ms_log_rms_mae_db": 1.5539208684381771, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.3184700487891126, "300–1000 Hz_40ms_envelope_correlation": 0.9840772218650663, "300–1000 Hz_40ms_modulation_mae_db": 1.8217823097959345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2903308370526252, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18054020728079762, "1000–3000 Hz_10ms_envelope_correlation": 0.9896105272756461, "1000–3000 Hz_10ms_modulation_mae_db": 1.4596094306893266, "1000–3000 Hz_40ms_log_rms_mae_db": 0.952383489893268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14562102000414096, "1000–3000 Hz_40ms_envelope_correlation": 0.9971033801030112, "1000–3000 Hz_40ms_modulation_mae_db": 1.4836085438357098, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3928144404814153, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29119776415067555, "3000–6000 Hz_10ms_envelope_correlation": 0.9527966091296391, "3000–6000 Hz_10ms_modulation_mae_db": 1.590996262108823, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9626033253819248, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.19830871786807783, "3000–6000 Hz_40ms_envelope_correlation": 0.9833615200618033, "3000–6000 Hz_40ms_modulation_mae_db": 1.7853608448127802, "mr_stft_db": 5.7837114016744975, "band_mae_db": 1.1258815852324728, "rms_gain_error_db": 1.6276005836944845, "output_peak": 0.42559076989818584}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.9612911689712975, "spectral_convergence": 0.5573963113133694, "spectral_convergence_status": "ok", "amplitude_floor": 1.1114078093309192e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.28403870318323, "spectral_convergence": 0.3287649440339257, "spectral_convergence_status": "ok", "amplitude_floor": 9.683415126458649e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.105804332868963, "spectral_convergence": 0.36394244633805417, "spectral_convergence_status": "ok", "amplitude_floor": 7.055286564887976e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011089954817882694, "output_power": 0.00012886297610983742, "delta_db": 0.6519838044984104, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007963949709957278, "output_power": 0.0012636888568228891, "delta_db": 2.0051164698940274, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005473467336544822, "output_power": 0.0007855190623501094, "delta_db": 1.5689419788299732, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020656259415557, "output_power": 0.00027210168971903705, "delta_db": 1.1967955957873364, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.567254948983353e-05, "output_power": 0.00011878783656805883, "delta_db": 0.9398462555345939, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.697083983250908e-05, "output_power": 4.046875621906291e-05, "delta_db": 0.3926054068504952, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.1258815852324728, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.6276005836944845, "pitch": null, "output_rms_dbfs": -25.834544857841205, "output_peak": 0.42559076989818584, "reference_rms_dbfs": -27.46214544153569, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7837114016744975}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015211892717126925, "log_rms_mae_db": 1.434597280008415, "rms_envelope_nrmse": 0.26207129200183316, "envelope_correlation": 0.9792540118743017, "rms_gain_error_db": 1.628922477786888, "reference_mean_power": 0.0017946561941472319, "output_mean_power": 0.0026114006627961078}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.098233390179555e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00033819477659287355, "output_power": 0.0005551279320919463, "reference_active": true, "delta_db": 2.152261844082387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.872738107665891e-05, "output_power": 6.473038987270806e-05, "reference_active": true, "delta_db": 2.2309009380004206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.329707857377202e-06, "output_power": 1.0613387008138821e-05, "reference_active": true, "delta_db": 1.6076733553233553}], "active_band_count": 3, "active_band_mae_db": 1.9969453791353875}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012855597503212083, "log_rms_mae_db": 1.1910336194213043, "rms_envelope_nrmse": 0.24413551941178216, "envelope_correlation": 0.9862319552021894, "rms_gain_error_db": 1.6318245626836403, "reference_mean_power": 0.0017996791281982027, "output_mean_power": 0.002620460008542909}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.521732580259768e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030590519204748624, "output_power": 0.0005070111426678426, "reference_active": true, "delta_db": 2.1943065578796714}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7673272930744677e-05, "output_power": 2.839769590355989e-05, "reference_active": true, "delta_db": 2.0596611986258146}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.996266898550172e-07, "output_power": 5.381775361870284e-07, "reference_active": true, "delta_db": 1.2927108039200625}], "active_band_count": 3, "active_band_mae_db": 1.848892853475183}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013968182745477764, "log_rms_mae_db": 1.8482231098708484, "rms_envelope_nrmse": 0.33230370411182114, "envelope_correlation": 0.9788016426824935, "rms_gain_error_db": 2.1192347934331695, "reference_mean_power": 0.0010910330804496425, "output_mean_power": 0.0017773026894435625}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.057825806205699e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031975947498638787, "output_power": 0.0005474405536627231, "reference_active": true, "delta_db": 2.335135440353141}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.9465579420304694e-05, "output_power": 7.140605549637169e-05, "reference_active": true, "delta_db": 2.5751655975567727}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.299925023694004e-06, "output_power": 8.676940277899947e-06, "reference_active": true, "delta_db": 1.3903122739345082}], "active_band_count": 3, "active_band_mae_db": 2.1002044372814743}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012087571887341512, "log_rms_mae_db": 1.5539208684381771, "rms_envelope_nrmse": 0.3184700487891126, "envelope_correlation": 0.9840772218650663, "rms_gain_error_db": 2.1229638377822986, "reference_mean_power": 0.0010942126232004053, "output_mean_power": 0.001784013368405722}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.627912926828161e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002965811192999727, "output_power": 0.0005113868677329335, "reference_active": true, "delta_db": 2.3660607139761773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8971998770340482e-05, "output_power": 3.305927111735958e-05, "reference_active": true, "delta_db": 2.4118018629228315}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.113014123121411e-07, "output_power": 4.818464193281797e-07, "reference_active": true, "delta_db": 0.6874843524887941}], "active_band_count": 3, "active_band_mae_db": 1.8217823097959345}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.190065013136077e-05, "log_rms_mae_db": 1.2903308370526252, "rms_envelope_nrmse": 0.18054020728079762, "envelope_correlation": 0.9896105272756461, "rms_gain_error_db": 1.0650199480580211, "reference_mean_power": 0.00025666306625944256, "output_mean_power": 0.0003279936026685337}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2531384738435257e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.158137778593675e-05, "output_power": 9.620817229241255e-05, "reference_active": true, "delta_db": 1.2841191089014623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.324550913609082e-06, "output_power": 7.614450266198449e-06, "reference_active": true, "delta_db": 1.5535557028932125}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7772043437261862e-06, "output_power": 2.5342690168448856e-06, "reference_active": true, "delta_db": 1.541153480273305}], "active_band_count": 3, "active_band_mae_db": 1.4596094306893266}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.863962912697811e-05, "log_rms_mae_db": 0.952383489893268, "rms_envelope_nrmse": 0.14562102000414096, "envelope_correlation": 0.9971033801030112, "rms_gain_error_db": 1.0639532494927524, "reference_mean_power": 0.00025765686496881694, "output_mean_power": 0.0003291827304105644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.1689575138838273e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607374483640592e-05, "output_power": 8.924231884087316e-05, "reference_active": true, "delta_db": 1.3054192446435189}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.394936823991012e-06, "output_power": 3.3258121048370497e-06, "reference_active": true, "delta_db": 1.4260364801188028}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.833675392620637e-08, "output_power": 1.461008891178649e-07, "reference_active": true, "delta_db": 1.7193699067448074}], "active_band_count": 3, "active_band_mae_db": 1.4836085438357098}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.534983553210222e-05, "log_rms_mae_db": 1.3928144404814153, "rms_envelope_nrmse": 0.29119776415067555, "envelope_correlation": 0.9527966091296391, "rms_gain_error_db": 1.0591329023152598, "reference_mean_power": 1.9026430507347487e-05, "output_mean_power": 2.4281225901239504e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1975009508047096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.298926861031976e-06, "output_power": 6.568615157872188e-06, "reference_active": true, "delta_db": 0.9328589299851564}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588034222302733e-06, "output_power": 2.051683497195717e-06, "reference_active": true, "delta_db": 1.1125050791723408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.193680029927605e-07, "output_power": 7.858827736260078e-07, "reference_active": true, "delta_db": 2.7276247771689714}], "active_band_count": 3, "active_band_mae_db": 1.590996262108823}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.3831266720127666e-05, "log_rms_mae_db": 0.9626033253819248, "rms_envelope_nrmse": 0.19830871786807783, "envelope_correlation": 0.9833615200618033, "rms_gain_error_db": 1.0586427515799075, "reference_mean_power": 1.9104126133492904e-05, "output_mean_power": 2.437762836703006e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1095182205605233e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.38789012101126e-06, "output_power": 6.805140024266999e-06, "reference_active": true, "delta_db": 1.014183355114354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.477294263383285e-07, "output_power": 9.958507445209546e-07, "reference_active": true, "delta_db": 1.2444978007028338}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.443035434026362e-08, "output_power": 4.9850544225820805e-08, "reference_active": true, "delta_db": 3.0974013786211527}], "active_band_count": 3, "active_band_mae_db": 1.7853608448127802}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9246348540222148, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3cbda37be149fa5d2c4204dbf5a46a2e2d00a6af65949a368807e0c460ae4efd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9896105272756461, "standard_deviation": null, "minimum": 0.9896105272756461, "maximum": 0.9896105272756461, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2903308370526252, "standard_deviation": null, "minimum": 1.2903308370526252, "maximum": 1.2903308370526252, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.4596094306893266, "standard_deviation": null, "minimum": 1.4596094306893266, "maximum": 1.4596094306893266, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18054020728079762, "standard_deviation": null, "minimum": 0.18054020728079762, "maximum": 0.18054020728079762, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9971033801030112, "standard_deviation": null, "minimum": 0.9971033801030112, "maximum": 0.9971033801030112, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.952383489893268, "standard_deviation": null, "minimum": 0.952383489893268, "maximum": 0.952383489893268, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4836085438357098, "standard_deviation": null, "minimum": 1.4836085438357098, "maximum": 1.4836085438357098, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14562102000414096, "standard_deviation": null, "minimum": 0.14562102000414096, "maximum": 0.14562102000414096, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9527966091296391, "standard_deviation": null, "minimum": 0.9527966091296391, "maximum": 0.9527966091296391, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3928144404814153, "standard_deviation": null, "minimum": 1.3928144404814153, "maximum": 1.3928144404814153, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.590996262108823, "standard_deviation": null, "minimum": 1.590996262108823, "maximum": 1.590996262108823, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29119776415067555, "standard_deviation": null, "minimum": 0.29119776415067555, "maximum": 0.29119776415067555, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9833615200618033, "standard_deviation": null, "minimum": 0.9833615200618033, "maximum": 0.9833615200618033, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9626033253819248, "standard_deviation": null, "minimum": 0.9626033253819248, "maximum": 0.9626033253819248, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.7853608448127802, "standard_deviation": null, "minimum": 1.7853608448127802, "maximum": 1.7853608448127802, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19830871786807783, "standard_deviation": null, "minimum": 0.19830871786807783, "maximum": 0.19830871786807783, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9788016426824935, "standard_deviation": null, "minimum": 0.9788016426824935, "maximum": 0.9788016426824935, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8482231098708484, "standard_deviation": null, "minimum": 1.8482231098708484, "maximum": 1.8482231098708484, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.1002044372814743, "standard_deviation": null, "minimum": 2.1002044372814743, "maximum": 2.1002044372814743, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.33230370411182114, "standard_deviation": null, "minimum": 0.33230370411182114, "maximum": 0.33230370411182114, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9840772218650663, "standard_deviation": null, "minimum": 0.9840772218650663, "maximum": 0.9840772218650663, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5539208684381771, "standard_deviation": null, "minimum": 1.5539208684381771, "maximum": 1.5539208684381771, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.8217823097959345, "standard_deviation": null, "minimum": 1.8217823097959345, "maximum": 1.8217823097959345, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3184700487891126, "standard_deviation": null, "minimum": 0.3184700487891126, "maximum": 0.3184700487891126, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.1258815852324728, "standard_deviation": null, "minimum": 1.1258815852324728, "maximum": 1.1258815852324728, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.7837114016744975, "standard_deviation": null, "minimum": 5.7837114016744975, "maximum": 5.7837114016744975, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.42559076989818584, "standard_deviation": null, "minimum": 0.42559076989818584, "maximum": 0.42559076989818584, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.6276005836944845, "standard_deviation": null, "minimum": 1.6276005836944845, "maximum": 1.6276005836944845, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9792540118743017, "standard_deviation": null, "minimum": 0.9792540118743017, "maximum": 0.9792540118743017, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.434597280008415, "standard_deviation": null, "minimum": 1.434597280008415, "maximum": 1.434597280008415, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.9969453791353875, "standard_deviation": null, "minimum": 1.9969453791353875, "maximum": 1.9969453791353875, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.26207129200183316, "standard_deviation": null, "minimum": 0.26207129200183316, "maximum": 0.26207129200183316, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9862319552021894, "standard_deviation": null, "minimum": 0.9862319552021894, "maximum": 0.9862319552021894, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1910336194213043, "standard_deviation": null, "minimum": 1.1910336194213043, "maximum": 1.1910336194213043, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.848892853475183, "standard_deviation": null, "minimum": 1.848892853475183, "maximum": 1.848892853475183, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.24413551941178216, "standard_deviation": null, "minimum": 0.24413551941178216, "maximum": 0.24413551941178216, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_3729"}, {"case": "librispeech_4077-13751-0001", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9589743614196777, "predicted_mos": 2.9589743614196777, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5590f7b4bbdaaa779dc32708d731647555f176f6db1e8835b9967964b48358cc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.140682799741626}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 6.6434846641738226, "全帯域_10ms_rms_envelope_nrmse": 0.19181807696798714, "全帯域_10ms_envelope_correlation": 0.974231639485269, "全帯域_10ms_modulation_mae_db": 0.4241029533153693, "全帯域_40ms_log_rms_mae_db": 6.0541839142277, "全帯域_40ms_rms_envelope_nrmse": 0.13781044941440188, "全帯域_40ms_envelope_correlation": 0.9899108818658768, "全帯域_40ms_modulation_mae_db": 0.38867733144864447, "300–1000 Hz_10ms_log_rms_mae_db": 2.220080076562793, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2081268600531607, "300–1000 Hz_10ms_envelope_correlation": 0.9816780855026146, "300–1000 Hz_10ms_modulation_mae_db": 0.7685276193711118, "300–1000 Hz_40ms_log_rms_mae_db": 1.7386768378999218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15029282246421752, "300–1000 Hz_40ms_envelope_correlation": 0.9951459014222418, "300–1000 Hz_40ms_modulation_mae_db": 0.7859659632295246, "1000–3000 Hz_10ms_log_rms_mae_db": 1.788564888601988, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19827196645782827, "1000–3000 Hz_10ms_envelope_correlation": 0.9770303551704673, "1000–3000 Hz_10ms_modulation_mae_db": 0.38252972474285407, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0886221258488264, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1085319283636961, "1000–3000 Hz_40ms_envelope_correlation": 0.9951847388236021, "1000–3000 Hz_40ms_modulation_mae_db": 0.4657706866102771, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8927916065049437, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23610796691624344, "3000–6000 Hz_10ms_envelope_correlation": 0.9677201212017138, "3000–6000 Hz_10ms_modulation_mae_db": 0.7146276060022153, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1345024055629878, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12246042529035, "3000–6000 Hz_40ms_envelope_correlation": 0.9920642626982762, "3000–6000 Hz_40ms_modulation_mae_db": 0.5854074645295415, "mr_stft_db": 6.283768326826594, "band_mae_db": 0.5876349563603294, "rms_gain_error_db": 0.29509982663508955, "output_peak": 1.0006982926513655}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 1, "at_or_above_full_scale_fraction": 1.0416666666666666e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 1.0416666666666666e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.118312660551994, "spectral_convergence": 0.6658888128644567, "spectral_convergence_status": "ok", "amplitude_floor": 1.963677544918102e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.795371578002932, "spectral_convergence": 0.42419339925027266, "spectral_convergence_status": "ok", "amplitude_floor": 1.0999213683671372e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.937620741924857, "spectral_convergence": 0.422724173429604, "spectral_convergence_status": "ok", "amplitude_floor": 9.527550859021228e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.000972009721466654, "output_power": 0.0008197056905897835, "delta_db": -0.7401265843311122, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00078771634185205, "output_power": 0.0009089773021512739, "delta_db": 0.6218318332183401, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008277001014055567, "output_power": 0.001011431429972723, "delta_db": 0.8706343668966987, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015338014805248517, "output_power": 0.00017021712530100678, "delta_db": 0.45234099048986687, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000153079755450355, "output_power": 0.0001638517773199448, "delta_db": 0.2953339683430229, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000364753212040036, "output_power": 0.0004135740915894246, "delta_db": 0.5455419948829352, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5876349563603294, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.29509982663508955, "pitch": null, "output_rms_dbfs": -24.574537290591188, "output_peak": 1.0006982926513655, "reference_rms_dbfs": -24.869637117226276, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.283768326826594}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002718271425028404, "log_rms_mae_db": 6.6434846641738226, "rms_envelope_nrmse": 0.19181807696798714, "envelope_correlation": 0.974231639485269, "rms_gain_error_db": 0.3267100875028832, "reference_mean_power": 0.0032374876277477364, "output_mean_power": 0.003490431554641293}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2735378968683332e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004666640237447372, "output_power": 0.0005558610960371091, "reference_active": true, "delta_db": 0.7596195805500607}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002202584731702016, "output_power": 0.00020343054555843255, "reference_active": true, "delta_db": -0.34516460586786857}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.478672520125922e-05, "output_power": 3.61548041938279e-05, "reference_active": true, "delta_db": 0.16752467352817862}], "active_band_count": 3, "active_band_mae_db": 0.4241029533153693}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002399603663054526, "log_rms_mae_db": 6.0541839142277, "rms_envelope_nrmse": 0.13781044941440188, "envelope_correlation": 0.9899108818658768, "rms_gain_error_db": 0.3628812076990494, "reference_mean_power": 0.0031195469748277875, "output_mean_power": 0.003391404956025155}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1403707959359127e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046404851798205753, "output_power": 0.0005502463595829934, "reference_active": true, "delta_db": 0.7399378793602583}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.861370772885771e-05, "output_power": 9.268655996807129e-05, "reference_active": true, "delta_db": -0.26920524177874505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9585409286275983e-06, "output_power": 1.8890513536577739e-06, "reference_active": true, "delta_db": -0.15688887320693007}], "active_band_count": 3, "active_band_mae_db": 0.38867733144864447}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022360065406936124, "log_rms_mae_db": 2.220080076562793, "rms_envelope_nrmse": 0.2081268600531607, "envelope_correlation": 0.9816780855026146, "rms_gain_error_db": 0.8970627100173882, "reference_mean_power": 0.0012597240435666534, "output_mean_power": 0.0015487513200461267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2605416603296333e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001978275618555469, "output_power": 0.0002431285926669646, "reference_active": true, "delta_db": 0.8954923768319341}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.107555923758901e-05, "output_power": 9.317815891976433e-05, "reference_active": true, "delta_db": 0.6042417209530866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1829892137914497e-05, "output_power": 9.826433260040675e-06, "reference_active": true, "delta_db": -0.8058487603283147}], "active_band_count": 3, "active_band_mae_db": 0.7685276193711118}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019906826982741464, "log_rms_mae_db": 1.7386768378999218, "rms_envelope_nrmse": 0.15029282246421752, "envelope_correlation": 0.9951459014222418, "rms_gain_error_db": 0.9905233551319828, "reference_mean_power": 0.0011746644654148715, "output_mean_power": 0.0014755915742082025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9316122263011187e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214336491155288, "output_power": 0.00025605949917471604, "reference_active": true, "delta_db": 0.8171143859010442}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6617395760512454e-05, "output_power": 4.109117261735693e-05, "reference_active": true, "delta_db": 0.500610809669128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0023494047314195e-06, "output_power": 7.888635095759922e-07, "reference_active": true, "delta_db": -1.0401726941184017}], "active_band_count": 3, "active_band_mae_db": 0.7859659632295246}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.946276793614836e-05, "log_rms_mae_db": 1.788564888601988, "rms_envelope_nrmse": 0.19827196645782827, "envelope_correlation": 0.9770303551704673, "rms_gain_error_db": 0.526329722686827, "reference_mean_power": 0.00019501605476176557, "output_mean_power": 0.00022014221837007053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.789505238315189e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.753576467028983e-05, "output_power": 4.409052585266669e-05, "reference_active": true, "delta_db": 0.6990001089581553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.624079451059313e-05, "output_power": 1.7845347464217235e-05, "reference_active": true, "delta_db": 0.409177370568924}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.016250052718367e-06, "output_power": 3.979967950012605e-06, "reference_active": true, "delta_db": -0.03941169470148294}], "active_band_count": 3, "active_band_mae_db": 0.38252972474285407}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.236492865510781e-05, "log_rms_mae_db": 1.0886221258488264, "rms_envelope_nrmse": 0.1085319283636961, "envelope_correlation": 0.9951847388236021, "rms_gain_error_db": 0.5866062884932792, "reference_mean_power": 0.00017673870709398627, "output_mean_power": 0.00020229833227078644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.939463723397018e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.781437846450251e-05, "output_power": 4.4010239553462814e-05, "reference_active": true, "delta_db": 0.6589676597510362}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.77351369725072e-06, "output_power": 8.620237079672533e-06, "reference_active": true, "delta_db": 0.4490184210952093}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.249207604003898e-07, "output_power": 3.039798966646586e-07, "reference_active": true, "delta_db": -0.28932597898458556}], "active_band_count": 3, "active_band_mae_db": 0.4657706866102771}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014584429481341953, "log_rms_mae_db": 1.8927916065049437, "rms_envelope_nrmse": 0.23610796691624344, "envelope_correlation": 0.9677201212017138, "rms_gain_error_db": 0.42025567843685685, "reference_mean_power": 0.00024652309966206675, "output_mean_power": 0.0002715708724744139}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9096739107120075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.093194518056527e-05, "output_power": 8.187449205918055e-05, "reference_active": true, "delta_db": 0.6230674893636864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.832881668922571e-05, "output_power": 3.191841840392872e-05, "reference_active": true, "delta_db": 0.5181292922528835}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4378297018808954e-05, "output_power": 1.1414025727939374e-05, "reference_active": true, "delta_db": -1.0026860363900763}], "active_band_count": 3, "active_band_mae_db": 0.7146276060022153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.656067210642841e-05, "log_rms_mae_db": 1.1345024055629878, "rms_envelope_nrmse": 0.12246042529035, "envelope_correlation": 0.9920642626982762, "rms_gain_error_db": 0.43166663532474436, "reference_mean_power": 0.0002460055124763468, "output_mean_power": 0.00027171367871499705}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8079454178991016e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.471033966073099e-05, "output_power": 8.46255231302957e-05, "reference_active": true, "delta_db": 0.541206555069275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5789672835999803e-05, "output_power": 1.6359630082807148e-05, "reference_active": true, "delta_db": 0.1540034789360489}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1060211769697704e-06, "output_power": 8.662878165183521e-07, "reference_active": true, "delta_db": -1.0610123595833005}], "active_band_count": 3, "active_band_mae_db": 0.5854074645295415}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8308179464479468, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f2de45d63c50d6a064b6de42f9aea8460121224976b680d340de27c410525010", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9770303551704673, "standard_deviation": null, "minimum": 0.9770303551704673, "maximum": 0.9770303551704673, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.788564888601988, "standard_deviation": null, "minimum": 1.788564888601988, "maximum": 1.788564888601988, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.38252972474285407, "standard_deviation": null, "minimum": 0.38252972474285407, "maximum": 0.38252972474285407, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19827196645782827, "standard_deviation": null, "minimum": 0.19827196645782827, "maximum": 0.19827196645782827, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951847388236021, "standard_deviation": null, "minimum": 0.9951847388236021, "maximum": 0.9951847388236021, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0886221258488264, "standard_deviation": null, "minimum": 1.0886221258488264, "maximum": 1.0886221258488264, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4657706866102771, "standard_deviation": null, "minimum": 0.4657706866102771, "maximum": 0.4657706866102771, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1085319283636961, "standard_deviation": null, "minimum": 0.1085319283636961, "maximum": 0.1085319283636961, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9677201212017138, "standard_deviation": null, "minimum": 0.9677201212017138, "maximum": 0.9677201212017138, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8927916065049437, "standard_deviation": null, "minimum": 1.8927916065049437, "maximum": 1.8927916065049437, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7146276060022153, "standard_deviation": null, "minimum": 0.7146276060022153, "maximum": 0.7146276060022153, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23610796691624344, "standard_deviation": null, "minimum": 0.23610796691624344, "maximum": 0.23610796691624344, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9920642626982762, "standard_deviation": null, "minimum": 0.9920642626982762, "maximum": 0.9920642626982762, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1345024055629878, "standard_deviation": null, "minimum": 1.1345024055629878, "maximum": 1.1345024055629878, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5854074645295415, "standard_deviation": null, "minimum": 0.5854074645295415, "maximum": 0.5854074645295415, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12246042529035, "standard_deviation": null, "minimum": 0.12246042529035, "maximum": 0.12246042529035, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9816780855026146, "standard_deviation": null, "minimum": 0.9816780855026146, "maximum": 0.9816780855026146, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.220080076562793, "standard_deviation": null, "minimum": 2.220080076562793, "maximum": 2.220080076562793, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7685276193711118, "standard_deviation": null, "minimum": 0.7685276193711118, "maximum": 0.7685276193711118, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2081268600531607, "standard_deviation": null, "minimum": 0.2081268600531607, "maximum": 0.2081268600531607, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9951459014222418, "standard_deviation": null, "minimum": 0.9951459014222418, "maximum": 0.9951459014222418, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7386768378999218, "standard_deviation": null, "minimum": 1.7386768378999218, "maximum": 1.7386768378999218, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7859659632295246, "standard_deviation": null, "minimum": 0.7859659632295246, "maximum": 0.7859659632295246, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15029282246421752, "standard_deviation": null, "minimum": 0.15029282246421752, "maximum": 0.15029282246421752, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5876349563603294, "standard_deviation": null, "minimum": 0.5876349563603294, "maximum": 0.5876349563603294, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.283768326826594, "standard_deviation": null, "minimum": 6.283768326826594, "maximum": 6.283768326826594, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.0006982926513655, "standard_deviation": null, "minimum": 1.0006982926513655, "maximum": 1.0006982926513655, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.29509982663508955, "standard_deviation": null, "minimum": 0.29509982663508955, "maximum": 0.29509982663508955, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.974231639485269, "standard_deviation": null, "minimum": 0.974231639485269, "maximum": 0.974231639485269, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 6.6434846641738226, "standard_deviation": null, "minimum": 6.6434846641738226, "maximum": 6.6434846641738226, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4241029533153693, "standard_deviation": null, "minimum": 0.4241029533153693, "maximum": 0.4241029533153693, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19181807696798714, "standard_deviation": null, "minimum": 0.19181807696798714, "maximum": 0.19181807696798714, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9899108818658768, "standard_deviation": null, "minimum": 0.9899108818658768, "maximum": 0.9899108818658768, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 6.0541839142277, "standard_deviation": null, "minimum": 6.0541839142277, "maximum": 6.0541839142277, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.38867733144864447, "standard_deviation": null, "minimum": 0.38867733144864447, "maximum": 0.38867733144864447, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.13781044941440188, "standard_deviation": null, "minimum": 0.13781044941440188, "maximum": 0.13781044941440188, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4077"}, {"case": "librispeech_4077-13754-0002", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.831660509109497, "predicted_mos": 2.831660509109497, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "382ad7b8e6ca9540b55f87b30a018556d5d5b396772d7b9d6bd7c8d1c383d2be", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.409170099999756}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.1056241368163064, "全帯域_10ms_rms_envelope_nrmse": 0.20106252317747253, "全帯域_10ms_envelope_correlation": 0.9742144381353971, "全帯域_10ms_modulation_mae_db": 0.5674479625641718, "全帯域_40ms_log_rms_mae_db": 1.365822764901928, "全帯域_40ms_rms_envelope_nrmse": 0.13862323068734886, "全帯域_40ms_envelope_correlation": 0.9912326498511745, "全帯域_40ms_modulation_mae_db": 0.5138199901819868, "300–1000 Hz_10ms_log_rms_mae_db": 2.1462583743434163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22376518018080585, "300–1000 Hz_10ms_envelope_correlation": 0.9815343285389072, "300–1000 Hz_10ms_modulation_mae_db": 0.911023824733288, "300–1000 Hz_40ms_log_rms_mae_db": 1.5302355790559536, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17550925677337798, "300–1000 Hz_40ms_envelope_correlation": 0.9929780873216231, "300–1000 Hz_40ms_modulation_mae_db": 1.0602978977996305, "1000–3000 Hz_10ms_log_rms_mae_db": 2.017891388293485, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21831447715464913, "1000–3000 Hz_10ms_envelope_correlation": 0.9697445307375925, "1000–3000 Hz_10ms_modulation_mae_db": 0.6027306780421225, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2066082972315733, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13172872955116688, "1000–3000 Hz_40ms_envelope_correlation": 0.9912281730184089, "1000–3000 Hz_40ms_modulation_mae_db": 0.6011774820406994, "3000–6000 Hz_10ms_log_rms_mae_db": 2.3997920816764067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27354963316353353, "3000–6000 Hz_10ms_envelope_correlation": 0.9522409024328791, "3000–6000 Hz_10ms_modulation_mae_db": 0.19207920456895425, "3000–6000 Hz_40ms_log_rms_mae_db": 1.46704227950661, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1765215640931955, "3000–6000 Hz_40ms_envelope_correlation": 0.9799807300699933, "3000–6000 Hz_40ms_modulation_mae_db": 0.18748611875201274, "mr_stft_db": 6.131930869382451, "band_mae_db": 0.7034912538283921, "rms_gain_error_db": 0.8021684856268457, "output_peak": 0.8693246281398861}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.092677525947783, "spectral_convergence": 0.6510877887828338, "spectral_convergence_status": "ok", "amplitude_floor": 1.602383167696498e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.600227657483656, "spectral_convergence": 0.3964849120867476, "spectral_convergence_status": "ok", "amplitude_floor": 1.3018377539140758e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.702887424715916, "spectral_convergence": 0.4109125542724952, "spectral_convergence_status": "ok", "amplitude_floor": 8.15121073207966e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00018979827012669627, "output_power": 0.00018162011862926194, "delta_db": -0.19128294821206365, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007146011617002843, "output_power": 0.000854052135547305, "delta_db": 0.7742066508843872, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006180173350849692, "output_power": 0.0008375320299476238, "delta_db": 1.320007678511824, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00021309861976674557, "output_power": 0.00026824749822156936, "delta_db": 0.9995504344976387, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003708463313924785, "output_power": 0.00040294329352538115, "delta_db": 0.3604994479439111, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001752804045595543, "output_power": 0.00020011212474393112, "delta_db": 0.5754003629205271, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7034912538283921, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8021684856268457, "pitch": null, "output_rms_dbfs": -25.615356254459336, "output_peak": 0.8693246281398861, "reference_rms_dbfs": -26.41752474008618, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.131930869382451}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002801281523487693, "log_rms_mae_db": 2.1056241368163064, "rms_envelope_nrmse": 0.20106252317747253, "envelope_correlation": 0.9742144381353971, "rms_gain_error_db": 0.8021876277024913, "reference_mean_power": 0.0022835195719053173, "output_mean_power": 0.0027467776251712864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.817352658591746e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000450774313706133, "output_power": 0.000563026355321736, "reference_active": true, "delta_db": 0.9656956384766815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021888119814572043, "output_power": 0.00023490628838328633, "reference_active": true, "delta_db": 0.3068618553440742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.753791744185585e-05, "output_power": 3.4000987881751553e-05, "reference_active": true, "delta_db": -0.4297863938717599}], "active_band_count": 3, "active_band_mae_db": 0.5674479625641718}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022259441477048914, "log_rms_mae_db": 1.365822764901928, "rms_envelope_nrmse": 0.13862323068734886, "envelope_correlation": 0.9912326498511745, "rms_gain_error_db": 0.8021622612580547, "reference_mean_power": 0.002294705859570105, "output_mean_power": 0.002760217155125885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.266680044598684e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042660839170957063, "output_power": 0.0005324720339293898, "reference_active": true, "delta_db": 0.9626740894385399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012272337901795565, "output_power": 0.0001327438507905295, "reference_active": true, "delta_db": 0.34087107550960255}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.241371364832561e-06, "output_power": 3.068579013914168e-06, "reference_active": true, "delta_db": -0.23791480559781777}], "active_band_count": 3, "active_band_mae_db": 0.5138199901819868}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002244694917839089, "log_rms_mae_db": 2.1462583743434163, "rms_envelope_nrmse": 0.22376518018080585, "envelope_correlation": 0.9815343285389072, "rms_gain_error_db": 1.15306741955227, "reference_mean_power": 0.001082154267593255, "output_mean_power": 0.001411223885429004}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.600154760198564e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003310843525581766, "output_power": 0.00042951264561883287, "reference_active": true, "delta_db": 1.1303729879881288}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.047815482126218e-05, "output_power": 0.0001087644349810407, "reference_active": true, "delta_db": 0.7994317291480341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6938491072881828e-05, "output_power": 1.4078229603056746e-05, "reference_active": true, "delta_db": -0.8032667570637015}], "active_band_count": 3, "active_band_mae_db": 0.911023824733288}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018235367957614062, "log_rms_mae_db": 1.5302355790559536, "rms_envelope_nrmse": 0.17550925677337798, "envelope_correlation": 0.9929780873216231, "rms_gain_error_db": 1.1530452122395294, "reference_mean_power": 0.0010875823794426604, "output_mean_power": 0.0014182953660385723}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.047574386178106e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00032394555601725244, "output_power": 0.0004166840406980713, "reference_active": true, "delta_db": 1.0933484044152566}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.774463594481584e-05, "output_power": 5.836796492200288e-05, "reference_active": true, "delta_db": 0.8724996513730874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5157059923891315e-06, "output_power": 1.1458041841873874e-06, "reference_active": true, "delta_db": -1.2150456376105474}], "active_band_count": 3, "active_band_mae_db": 1.0602978977996305}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014040871531833862, "log_rms_mae_db": 2.017891388293485, "rms_envelope_nrmse": 0.21831447715464913, "envelope_correlation": 0.9697445307375925, "rms_gain_error_db": 0.6755748313646872, "reference_mean_power": 0.00036159326722347827, "output_mean_power": 0.00042245243582497726}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.2796709554971252e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010222442860363171, "output_power": 0.00011893820886935144, "reference_active": true, "delta_db": 0.6576670233601274}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.551583786781791e-05, "output_power": 4.83183360920826e-05, "reference_active": true, "delta_db": 0.2594942859600942}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1818852015306123e-05, "output_power": 9.626584390270393e-06, "reference_active": true, "delta_db": -0.8910307248061461}], "active_band_count": 3, "active_band_mae_db": 0.6027306780421225}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010845211232047094, "log_rms_mae_db": 1.2066082972315733, "rms_envelope_nrmse": 0.13172872955116688, "envelope_correlation": 0.9912281730184089, "rms_gain_error_db": 0.6755676375559911, "reference_mean_power": 0.00036337479954251316, "output_mean_power": 0.0004245331117086732}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.941433739824422e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.608975832570782e-05, "output_power": 0.0001124119142434541, "reference_active": true, "delta_db": 0.6813524230803474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4640366344588194e-05, "output_power": 2.505924382416519e-05, "reference_active": true, "delta_db": 0.0732080130885193}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.520284367741145e-07, "output_power": 7.477436262716712e-07, "reference_active": true, "delta_db": -1.0489720099532316}], "active_band_count": 3, "active_band_mae_db": 0.6011774820406994}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012084003773539908, "log_rms_mae_db": 2.3997920816764067, "rms_envelope_nrmse": 0.27354963316353353, "envelope_correlation": 0.9522409024328791, "rms_gain_error_db": 0.5581933255938798, "reference_mean_power": 0.00030833628390941923, "output_mean_power": 0.0003506258783201073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2964446771205807e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7472950448048514e-05, "output_power": 4.719750608158138e-05, "reference_active": true, "delta_db": -0.025271728977136285}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.978240331431912e-05, "output_power": 4.64232765279854e-05, "reference_active": true, "delta_db": -0.3034006949018162}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3740568823166797e-05, "output_power": 1.2979208012640479e-05, "reference_active": true, "delta_db": -0.24756518982791026}], "active_band_count": 3, "active_band_mae_db": 0.19207920456895425}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010527157253957595, "log_rms_mae_db": 1.46704227950661, "rms_envelope_nrmse": 0.1765215640931955, "envelope_correlation": 0.9799807300699933, "rms_gain_error_db": 0.5582366257415815, "reference_mean_power": 0.0003098592063705908, "output_mean_power": 0.00035236118898842813}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.056454468513196e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.721882315895364e-05, "output_power": 5.597723917624731e-05, "reference_active": true, "delta_db": -0.09527446212907603}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8898944201462896e-05, "output_power": 2.7175859728245422e-05, "reference_active": true, "delta_db": -0.26698684264020905}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2532553850736713e-06, "output_power": 1.1967952979731935e-06, "reference_active": true, "delta_db": -0.20019705148675315}], "active_band_count": 3, "active_band_mae_db": 0.18748611875201274}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.868091002037715, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "520a6cadf033783f40e426f99a16580b35641ec09383a96b5be90566bd1b0482", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9697445307375925, "standard_deviation": null, "minimum": 0.9697445307375925, "maximum": 0.9697445307375925, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.017891388293485, "standard_deviation": null, "minimum": 2.017891388293485, "maximum": 2.017891388293485, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6027306780421225, "standard_deviation": null, "minimum": 0.6027306780421225, "maximum": 0.6027306780421225, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21831447715464913, "standard_deviation": null, "minimum": 0.21831447715464913, "maximum": 0.21831447715464913, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9912281730184089, "standard_deviation": null, "minimum": 0.9912281730184089, "maximum": 0.9912281730184089, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2066082972315733, "standard_deviation": null, "minimum": 1.2066082972315733, "maximum": 1.2066082972315733, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6011774820406994, "standard_deviation": null, "minimum": 0.6011774820406994, "maximum": 0.6011774820406994, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13172872955116688, "standard_deviation": null, "minimum": 0.13172872955116688, "maximum": 0.13172872955116688, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9522409024328791, "standard_deviation": null, "minimum": 0.9522409024328791, "maximum": 0.9522409024328791, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.3997920816764067, "standard_deviation": null, "minimum": 2.3997920816764067, "maximum": 2.3997920816764067, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.19207920456895425, "standard_deviation": null, "minimum": 0.19207920456895425, "maximum": 0.19207920456895425, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.27354963316353353, "standard_deviation": null, "minimum": 0.27354963316353353, "maximum": 0.27354963316353353, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9799807300699933, "standard_deviation": null, "minimum": 0.9799807300699933, "maximum": 0.9799807300699933, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.46704227950661, "standard_deviation": null, "minimum": 1.46704227950661, "maximum": 1.46704227950661, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.18748611875201274, "standard_deviation": null, "minimum": 0.18748611875201274, "maximum": 0.18748611875201274, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1765215640931955, "standard_deviation": null, "minimum": 0.1765215640931955, "maximum": 0.1765215640931955, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9815343285389072, "standard_deviation": null, "minimum": 0.9815343285389072, "maximum": 0.9815343285389072, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1462583743434163, "standard_deviation": null, "minimum": 2.1462583743434163, "maximum": 2.1462583743434163, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.911023824733288, "standard_deviation": null, "minimum": 0.911023824733288, "maximum": 0.911023824733288, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22376518018080585, "standard_deviation": null, "minimum": 0.22376518018080585, "maximum": 0.22376518018080585, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9929780873216231, "standard_deviation": null, "minimum": 0.9929780873216231, "maximum": 0.9929780873216231, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5302355790559536, "standard_deviation": null, "minimum": 1.5302355790559536, "maximum": 1.5302355790559536, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0602978977996305, "standard_deviation": null, "minimum": 1.0602978977996305, "maximum": 1.0602978977996305, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17550925677337798, "standard_deviation": null, "minimum": 0.17550925677337798, "maximum": 0.17550925677337798, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7034912538283921, "standard_deviation": null, "minimum": 0.7034912538283921, "maximum": 0.7034912538283921, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.131930869382451, "standard_deviation": null, "minimum": 6.131930869382451, "maximum": 6.131930869382451, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.8693246281398861, "standard_deviation": null, "minimum": 0.8693246281398861, "maximum": 0.8693246281398861, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8021684856268457, "standard_deviation": null, "minimum": 0.8021684856268457, "maximum": 0.8021684856268457, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9742144381353971, "standard_deviation": null, "minimum": 0.9742144381353971, "maximum": 0.9742144381353971, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.1056241368163064, "standard_deviation": null, "minimum": 2.1056241368163064, "maximum": 2.1056241368163064, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5674479625641718, "standard_deviation": null, "minimum": 0.5674479625641718, "maximum": 0.5674479625641718, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20106252317747253, "standard_deviation": null, "minimum": 0.20106252317747253, "maximum": 0.20106252317747253, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9912326498511745, "standard_deviation": null, "minimum": 0.9912326498511745, "maximum": 0.9912326498511745, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.365822764901928, "standard_deviation": null, "minimum": 1.365822764901928, "maximum": 1.365822764901928, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.5138199901819868, "standard_deviation": null, "minimum": 0.5138199901819868, "maximum": 0.5138199901819868, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.13862323068734886, "standard_deviation": null, "minimum": 0.13862323068734886, "maximum": 0.13862323068734886, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4077"}, {"case": "librispeech_4077-13754-0012", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.476252317428589, "predicted_mos": 2.476252317428589, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3d1ee6d690c07cf529365907c4b4bf60c8ecab62ed6926e759883b8bd3645ac8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.597814099863172}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.4416480599087937, "全帯域_10ms_rms_envelope_nrmse": 0.1846568727492595, "全帯域_10ms_envelope_correlation": 0.9768280358150024, "全帯域_10ms_modulation_mae_db": 0.34816755457005627, "全帯域_40ms_log_rms_mae_db": 1.8088546353455488, "全帯域_40ms_rms_envelope_nrmse": 0.12379515772455253, "全帯域_40ms_envelope_correlation": 0.9914646007605331, "全帯域_40ms_modulation_mae_db": 0.40997175982204886, "300–1000 Hz_10ms_log_rms_mae_db": 1.921506873265857, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2136968768117633, "300–1000 Hz_10ms_envelope_correlation": 0.9858829399122698, "300–1000 Hz_10ms_modulation_mae_db": 0.886110350762765, "300–1000 Hz_40ms_log_rms_mae_db": 1.254159022516763, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.18103282369957738, "300–1000 Hz_40ms_envelope_correlation": 0.9934907349198911, "300–1000 Hz_40ms_modulation_mae_db": 0.8140498718995689, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8732980038596405, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18258089208413986, "1000–3000 Hz_10ms_envelope_correlation": 0.9758137757338016, "1000–3000 Hz_10ms_modulation_mae_db": 0.7368684911977743, "1000–3000 Hz_40ms_log_rms_mae_db": 1.119148361066003, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11144050567005619, "1000–3000 Hz_40ms_envelope_correlation": 0.9910876503384316, "1000–3000 Hz_40ms_modulation_mae_db": 0.7405250833732104, "3000–6000 Hz_10ms_log_rms_mae_db": 2.164463136257469, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23198447133549294, "3000–6000 Hz_10ms_envelope_correlation": 0.9691635935054452, "3000–6000 Hz_10ms_modulation_mae_db": 0.22545986352109518, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2883738437062107, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13458435958223863, "3000–6000 Hz_40ms_envelope_correlation": 0.989763363372159, "3000–6000 Hz_40ms_modulation_mae_db": 0.3610018653467937, "mr_stft_db": 6.10897189746088, "band_mae_db": 0.5777700052333012, "rms_gain_error_db": 0.60838740694397, "output_peak": 0.6271727081135676}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.055765394101695, "spectral_convergence": 0.628169287852535, "spectral_convergence_status": "ok", "amplitude_floor": 1.32832143223519e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.561962178943374, "spectral_convergence": 0.4479213812887227, "spectral_convergence_status": "ok", "amplitude_floor": 1.1133586355586426e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.709188119337572, "spectral_convergence": 0.4772630483964902, "spectral_convergence_status": "ok", "amplitude_floor": 6.710028542715064e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00015990772541983682, "output_power": 0.00015259277805729514, "delta_db": -0.2033546604772128, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005346827040387907, "output_power": 0.0006755300279882872, "delta_db": 1.015485228800477, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025601253900142847, "output_power": 0.00032802918403160785, "delta_db": 1.0765124691025465, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015659557404440672, "output_power": 0.00017925173466468904, "delta_db": 0.5868388397118656, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003554436518358531, "output_power": 0.00037443511651790004, "delta_db": 0.2260581021715069, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0004235024710497043, "output_power": 0.0004599313592058554, "delta_db": 0.3583707311361989, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5777700052333012, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.60838740694397, "pitch": null, "output_rms_dbfs": -26.635862596788705, "output_peak": 0.6271727081135676, "reference_rms_dbfs": -27.244250003732674, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.10897189746088}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018816516597127566, "log_rms_mae_db": 2.4416480599087937, "rms_envelope_nrmse": 0.1846568727492595, "envelope_correlation": 0.9768280358150024, "rms_gain_error_db": 0.6101541377654389, "reference_mean_power": 0.0018869472812584292, "output_mean_power": 0.002171576736407591}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.448172186664129e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039739520206635815, "output_power": 0.0004459914675946021, "reference_active": true, "delta_db": 0.50103930795033}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001684948117890412, "output_power": 0.00018285566940602713, "reference_active": true, "delta_db": 0.35521897275402026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9559595301379592e-05, "output_power": 1.8729898611192137e-05, "reference_active": true, "delta_db": -0.18824438300581858}], "active_band_count": 3, "active_band_mae_db": 0.34816755457005627}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016656172387320294, "log_rms_mae_db": 1.8088546353455488, "rms_envelope_nrmse": 0.12379515772455253, "envelope_correlation": 0.9914646007605331, "rms_gain_error_db": 0.6126106358183561, "reference_mean_power": 0.0018945898208991615, "output_mean_power": 0.002181605717956811}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.514104383617561e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003829793712831679, "output_power": 0.00043138899006473825, "reference_active": true, "delta_db": 0.5169367496265387}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.831616093630877e-05, "output_power": 8.574551561470333e-05, "reference_active": true, "delta_db": 0.39360026069036036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3812315351816923e-06, "output_power": 2.2124007573336025e-06, "reference_active": true, "delta_db": -0.31937826914924744}], "active_band_count": 3, "active_band_mae_db": 0.40997175982204886}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015092951870698086, "log_rms_mae_db": 1.921506873265857, "rms_envelope_nrmse": 0.2136968768117633, "envelope_correlation": 0.9858829399122698, "rms_gain_error_db": 1.2028569122007147, "reference_mean_power": 0.000631981580682194, "output_mean_power": 0.000833662203666006}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.167139764996573e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018037000728302603, "output_power": 0.00023293066713306307, "reference_active": true, "delta_db": 1.110623477279444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6944061726753776e-05, "output_power": 3.2463503240895994e-05, "reference_active": true, "delta_db": 0.809323193758342}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.521461134259169e-06, "output_power": 2.9708755624039766e-06, "reference_active": true, "delta_db": -0.7383843812505089}], "active_band_count": 3, "active_band_mae_db": 0.886110350762765}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013588907795973108, "log_rms_mae_db": 1.254159022516763, "rms_envelope_nrmse": 0.18103282369957738, "envelope_correlation": 0.9934907349198911, "rms_gain_error_db": 1.2054601254608086, "reference_mean_power": 0.0006345375227010602, "output_mean_power": 0.0008375356871779124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.990151794989553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001715162379513309, "output_power": 0.00022083904475288702, "reference_active": true, "delta_db": 1.0977061764314908}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2552077755571312e-05, "output_power": 1.4900791975278047e-05, "reference_active": true, "delta_db": 0.7449373084043012}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.057253626713498e-07, "output_power": 2.6630594135590863e-07, "reference_active": true, "delta_db": -0.599506130862915}], "active_band_count": 3, "active_band_mae_db": 0.8140498718995689}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.324369547066356e-05, "log_rms_mae_db": 1.8732980038596405, "rms_envelope_nrmse": 0.18258089208413986, "envelope_correlation": 0.9758137757338016, "rms_gain_error_db": 0.3611512237733016, "reference_mean_power": 0.00026216016454011753, "output_mean_power": 0.0002848930294189195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5456728460340458e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.502085853898132e-05, "output_power": 8.230430611790913e-05, "reference_active": true, "delta_db": 0.40240527954254435}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.151594827944158e-05, "output_power": 1.860959408272844e-05, "reference_active": true, "delta_db": -0.6302359135275695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.513878290774259e-06, "output_power": 6.4912899647937635e-06, "reference_active": true, "delta_db": -1.1779642805232093}], "active_band_count": 3, "active_band_mae_db": 0.7368684911977743}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.873751028776232e-05, "log_rms_mae_db": 1.119148361066003, "rms_envelope_nrmse": 0.11144050567005619, "envelope_correlation": 0.9910876503384316, "rms_gain_error_db": 0.36336596234958635, "reference_mean_power": 0.0002632887650860277, "output_mean_power": 0.00028626544243403625}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3635149275752304e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.257955032157826e-05, "output_power": 7.751262126952268e-05, "reference_active": true, "delta_db": 0.28558150684311345}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.405048683009261e-06, "output_power": 8.017086906073112e-06, "reference_active": true, "delta_db": -0.693444564936557}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.677553554297748e-07, "output_power": 5.767238831559403e-07, "reference_active": true, "delta_db": -1.242549178339961}], "active_band_count": 3, "active_band_mae_db": 0.7405250833732104}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015682729984940356, "log_rms_mae_db": 2.164463136257469, "rms_envelope_nrmse": 0.23198447133549294, "envelope_correlation": 0.9691635935054452, "rms_gain_error_db": 0.3493542119177021, "reference_mean_power": 0.0005797089728212459, "output_mean_power": 0.0006282687285645385}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2924452919576597e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020605682430557354, "output_power": 0.00020856075209804172, "reference_active": true, "delta_db": 0.052455819800176134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012645863441832237, "output_power": 0.00013502432801773542, "reference_active": true, "delta_db": 0.284635368723892}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.26990748699409e-06, "output_power": 8.573271153936973e-06, "reference_active": true, "delta_db": -0.3392884020392174}], "active_band_count": 3, "active_band_mae_db": 0.22545986352109518}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012704534221320322, "log_rms_mae_db": 1.2883738437062107, "rms_envelope_nrmse": 0.13458435958223863, "envelope_correlation": 0.989763363372159, "rms_gain_error_db": 0.35027247301142217, "reference_mean_power": 0.0005824704905358117, "output_mean_power": 0.0006313950533113864}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.040463159930301e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022945963060392102, "output_power": 0.0002340698151420751, "reference_active": true, "delta_db": 0.08639122048952715}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.01414803619856e-05, "output_power": 6.6184763324343e-05, "reference_active": true, "delta_db": 0.4158390626164127}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3285241811094974e-06, "output_power": 1.1622296009284325e-06, "reference_active": true, "delta_db": -0.5807753129344413}], "active_band_count": 3, "active_band_mae_db": 0.3610018653467937}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9083288186345833, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f6b615433a2d3a03c4a474509c23bce2e71cc2e9515cba7cfe508ad060e04adb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9758137757338016, "standard_deviation": null, "minimum": 0.9758137757338016, "maximum": 0.9758137757338016, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8732980038596405, "standard_deviation": null, "minimum": 1.8732980038596405, "maximum": 1.8732980038596405, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7368684911977743, "standard_deviation": null, "minimum": 0.7368684911977743, "maximum": 0.7368684911977743, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18258089208413986, "standard_deviation": null, "minimum": 0.18258089208413986, "maximum": 0.18258089208413986, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9910876503384316, "standard_deviation": null, "minimum": 0.9910876503384316, "maximum": 0.9910876503384316, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.119148361066003, "standard_deviation": null, "minimum": 1.119148361066003, "maximum": 1.119148361066003, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7405250833732104, "standard_deviation": null, "minimum": 0.7405250833732104, "maximum": 0.7405250833732104, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11144050567005619, "standard_deviation": null, "minimum": 0.11144050567005619, "maximum": 0.11144050567005619, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9691635935054452, "standard_deviation": null, "minimum": 0.9691635935054452, "maximum": 0.9691635935054452, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.164463136257469, "standard_deviation": null, "minimum": 2.164463136257469, "maximum": 2.164463136257469, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.22545986352109518, "standard_deviation": null, "minimum": 0.22545986352109518, "maximum": 0.22545986352109518, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23198447133549294, "standard_deviation": null, "minimum": 0.23198447133549294, "maximum": 0.23198447133549294, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.989763363372159, "standard_deviation": null, "minimum": 0.989763363372159, "maximum": 0.989763363372159, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2883738437062107, "standard_deviation": null, "minimum": 1.2883738437062107, "maximum": 1.2883738437062107, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3610018653467937, "standard_deviation": null, "minimum": 0.3610018653467937, "maximum": 0.3610018653467937, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13458435958223863, "standard_deviation": null, "minimum": 0.13458435958223863, "maximum": 0.13458435958223863, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9858829399122698, "standard_deviation": null, "minimum": 0.9858829399122698, "maximum": 0.9858829399122698, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.921506873265857, "standard_deviation": null, "minimum": 1.921506873265857, "maximum": 1.921506873265857, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.886110350762765, "standard_deviation": null, "minimum": 0.886110350762765, "maximum": 0.886110350762765, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2136968768117633, "standard_deviation": null, "minimum": 0.2136968768117633, "maximum": 0.2136968768117633, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9934907349198911, "standard_deviation": null, "minimum": 0.9934907349198911, "maximum": 0.9934907349198911, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.254159022516763, "standard_deviation": null, "minimum": 1.254159022516763, "maximum": 1.254159022516763, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.8140498718995689, "standard_deviation": null, "minimum": 0.8140498718995689, "maximum": 0.8140498718995689, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18103282369957738, "standard_deviation": null, "minimum": 0.18103282369957738, "maximum": 0.18103282369957738, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5777700052333012, "standard_deviation": null, "minimum": 0.5777700052333012, "maximum": 0.5777700052333012, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.10897189746088, "standard_deviation": null, "minimum": 6.10897189746088, "maximum": 6.10897189746088, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6271727081135676, "standard_deviation": null, "minimum": 0.6271727081135676, "maximum": 0.6271727081135676, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.60838740694397, "standard_deviation": null, "minimum": 0.60838740694397, "maximum": 0.60838740694397, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9768280358150024, "standard_deviation": null, "minimum": 0.9768280358150024, "maximum": 0.9768280358150024, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.4416480599087937, "standard_deviation": null, "minimum": 2.4416480599087937, "maximum": 2.4416480599087937, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.34816755457005627, "standard_deviation": null, "minimum": 0.34816755457005627, "maximum": 0.34816755457005627, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1846568727492595, "standard_deviation": null, "minimum": 0.1846568727492595, "maximum": 0.1846568727492595, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9914646007605331, "standard_deviation": null, "minimum": 0.9914646007605331, "maximum": 0.9914646007605331, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.8088546353455488, "standard_deviation": null, "minimum": 1.8088546353455488, "maximum": 1.8088546353455488, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.40997175982204886, "standard_deviation": null, "minimum": 0.40997175982204886, "maximum": 0.40997175982204886, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12379515772455253, "standard_deviation": null, "minimum": 0.12379515772455253, "maximum": 0.12379515772455253, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4077"}, {"case": "librispeech_4446-2271-0003", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.732762098312378, "predicted_mos": 2.732762098312378, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59200, "input_sample_rate": 16000, "input_seconds": 3.7, "input_float32_sha256": "86c572b3f96c558fa0f84cf29623c149edda888650d583156104dca5deb95a6b", "internal_samples": 59200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.864194400142878}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.602100048584684, "全帯域_10ms_rms_envelope_nrmse": 0.19213794056614766, "全帯域_10ms_envelope_correlation": 0.9764553709735636, "全帯域_10ms_modulation_mae_db": 1.0069512026727623, "全帯域_40ms_log_rms_mae_db": 1.2287411008357092, "全帯域_40ms_rms_envelope_nrmse": 0.1624209521830076, "全帯域_40ms_envelope_correlation": 0.9869566338305001, "全帯域_40ms_modulation_mae_db": 0.8490182604280255, "300–1000 Hz_10ms_log_rms_mae_db": 1.6305246684851018, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2461866649471397, "300–1000 Hz_10ms_envelope_correlation": 0.9737253576302066, "300–1000 Hz_10ms_modulation_mae_db": 1.2733851245492345, "300–1000 Hz_40ms_log_rms_mae_db": 1.3516519058701681, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.21812644102033285, "300–1000 Hz_40ms_envelope_correlation": 0.9834069606693465, "300–1000 Hz_40ms_modulation_mae_db": 1.1002240608553517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2537946974190346, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13427156852436525, "1000–3000 Hz_10ms_envelope_correlation": 0.9942884259958265, "1000–3000 Hz_10ms_modulation_mae_db": 0.535212429102983, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8764750898587593, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11403630034560205, "1000–3000 Hz_40ms_envelope_correlation": 0.9976503470351075, "1000–3000 Hz_40ms_modulation_mae_db": 0.43024992272364787, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6815642813029827, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3190394494559081, "3000–6000 Hz_10ms_envelope_correlation": 0.9527124474332355, "3000–6000 Hz_10ms_modulation_mae_db": 1.8180785766937912, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1996877634092744, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2216421638742134, "3000–6000 Hz_40ms_envelope_correlation": 0.9849923842095049, "3000–6000 Hz_40ms_modulation_mae_db": 1.5861664776803215, "mr_stft_db": 5.6613295154429695, "band_mae_db": 0.8812094577836941, "rms_gain_error_db": 1.0537958457013088, "output_peak": 0.6223962629342007}, "full_scale": {"sample_count": 59200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59200, "output_samples": 59200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.015814968486064, "spectral_convergence": 0.6215339974479761, "spectral_convergence_status": "ok", "amplitude_floor": 2.0296349328313604e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.216492980142983, "spectral_convergence": 0.27269018695461894, "spectral_convergence_status": "ok", "amplitude_floor": 1.6850709272784723e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7516805976998615, "spectral_convergence": 0.3254582530023453, "spectral_convergence_status": "ok", "amplitude_floor": 8.273401155700601e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185911287346583, "output_power": 0.0008261363076910861, "delta_db": 0.03984675515891038, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003085801603155809, "output_power": 0.004238122710261614, "delta_db": 1.3780552675176725, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013392891870332754, "output_power": 0.0016445660809570398, "delta_db": 0.8917696622375311, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00027936558999729056, "output_power": 0.000330697481213351, "delta_db": 0.7325797490707047, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.040755904577448e-06, "output_power": 8.679514708417797e-06, "delta_db": 0.9087615539694942, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.309787748977687e-06, "output_power": 8.583000274271478e-06, "delta_db": 1.3362437587478515, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8812094577836941, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0537958457013088, "pitch": null, "output_rms_dbfs": -21.513931082185273, "output_peak": 0.6223962629342007, "reference_rms_dbfs": -22.567726927886582, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.6613295154429695}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002672933594000982, "log_rms_mae_db": 1.602100048584684, "rms_envelope_nrmse": 0.19213794056614766, "envelope_correlation": 0.9764553709735636, "rms_gain_error_db": 1.053795955078298, "reference_mean_power": 0.0055438895852693145, "output_mean_power": 0.007066334105318819}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.3303693041440347e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007860073564586987, "output_power": 0.0009593028469999231, "reference_active": true, "delta_db": 0.8652912262266856}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024364364087793404, "output_power": 0.0003145949740261659, "reference_active": true, "delta_db": 1.1099669930960534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.963914121364001e-05, "output_power": 6.315151856309038e-05, "reference_active": true, "delta_db": 1.0455953886955482}], "active_band_count": 3, "active_band_mae_db": 1.0069512026727623}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00019127079187884833, "log_rms_mae_db": 1.2287411008357092, "rms_envelope_nrmse": 0.1624209521830076, "envelope_correlation": 0.9869566338305001, "rms_gain_error_db": 1.0537963613807948, "reference_mean_power": 0.005589268172913711, "output_mean_power": 0.0071241750762814715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.0930617045795464e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007030365316665854, "output_power": 0.0008739342161198672, "reference_active": true, "delta_db": 0.9450085038794692}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013359627830049334, "output_power": 0.0001739833573547877, "reference_active": true, "delta_db": 1.1471334734821717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7787329335262405e-06, "output_power": 3.0855890790842464e-06, "reference_active": true, "delta_db": 0.45491280392243605}], "active_band_count": 3, "active_band_mae_db": 0.8490182604280255}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 0.0002516933783769556, "log_rms_mae_db": 1.6305246684851018, "rms_envelope_nrmse": 0.2461866649471397, "envelope_correlation": 0.9737253576302066, "rms_gain_error_db": 1.423211392685635, "reference_mean_power": 0.003609667084718048, "output_mean_power": 0.005009429726090103}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.0389022511376814e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005770551731550911, "output_power": 0.0007922193407736346, "reference_active": true, "delta_db": 1.376281020846823}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00021321841152816334, "output_power": 0.000311103343077158, "reference_active": true, "delta_db": 1.640799744640284}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.0182675609825566e-05, "output_power": 6.0375574164087474e-05, "reference_active": true, "delta_db": 0.8030746081605962}], "active_band_count": 3, "active_band_mae_db": 1.2733851245492345}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 0.00017703062295034008, "log_rms_mae_db": 1.3516519058701681, "rms_envelope_nrmse": 0.21812644102033285, "envelope_correlation": 0.9834069606693465, "rms_gain_error_db": 1.4232113893666627, "reference_mean_power": 0.003639213964400563, "output_mean_power": 0.005050434339531984}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 8.593451292498866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005240710435115444, "output_power": 0.0007304810746846589, "reference_active": true, "delta_db": 1.4421880440782706}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012044100626957096, "output_power": 0.00017057135714488035, "reference_active": true, "delta_db": 1.5113172958678933}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.706489326091672e-06, "output_power": 2.9317234640970414e-06, "reference_active": true, "delta_db": 0.3471668426198914}], "active_band_count": 3, "active_band_mae_db": 1.1002240608553517}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 7.943457135434651e-05, "log_rms_mae_db": 1.2537946974190346, "rms_envelope_nrmse": 0.13427156852436525, "envelope_correlation": 0.9942884259958265, "rms_gain_error_db": 0.8205973100665388, "reference_mean_power": 0.00024301128670702193, "output_mean_power": 0.0002935527653880989}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 1.5901124046586967e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.178189042095757e-05, "output_power": 0.000105675902514858, "reference_active": true, "delta_db": 0.6121896700318801}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7997292458206206e-05, "output_power": 2.0659314388561764e-05, "reference_active": true, "delta_db": 0.599087306328071}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3356967361336817e-06, "output_power": 2.557716970010037e-06, "reference_active": true, "delta_db": 0.39436031094899804}], "active_band_count": 3, "active_band_mae_db": 0.535212429102983}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 6.195280015885064e-05, "log_rms_mae_db": 0.8764750898587593, "rms_envelope_nrmse": 0.11403630034560205, "envelope_correlation": 0.9976503470351075, "rms_gain_error_db": 0.8205981284070105, "reference_mean_power": 0.00024500023514433585, "output_mean_power": 0.0002959554310168025}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 1.4888167929489627e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.774520748189054e-05, "output_power": 0.00010105928773305728, "reference_active": true, "delta_db": 0.6135282776051226}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.213495209539377e-06, "output_power": 1.0445314618621514e-05, "reference_active": true, "delta_db": 0.5449711132962549}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5741557318464897e-07, "output_power": 1.5269424092458887e-07, "reference_active": true, "delta_db": -0.13225037726956634}], "active_band_count": 3, "active_band_mae_db": 0.43024992272364787}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 739, "rms_floor": 1.5353574631917574e-05, "log_rms_mae_db": 1.6815642813029827, "rms_envelope_nrmse": 0.3190394494559081, "envelope_correlation": 0.9527124474332355, "rms_gain_error_db": 1.3439291389531989, "reference_mean_power": 7.123238480443002e-06, "output_mean_power": 9.706672978449504e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 739, "frequency_resolution_hz": 0.2706359945872801, "power_floor": 4.409187309389394e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6167395840190636e-06, "output_power": 3.6723332901176442e-06, "reference_active": true, "delta_db": 1.4718158547582498}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.009142614437233e-07, "output_power": 1.1830857809799208e-06, "reference_active": true, "delta_db": 2.273513381579742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.642480310119981e-07, "output_power": 3.9165385834290954e-07, "reference_active": true, "delta_db": 1.7089064937433815}], "active_band_count": 3, "active_band_mae_db": 1.8180785766937912}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 733, "rms_floor": 1.282880604964887e-05, "log_rms_mae_db": 1.1996877634092744, "rms_envelope_nrmse": 0.2216421638742134, "envelope_correlation": 0.9849923842095049, "rms_gain_error_db": 1.3439468519124258, "reference_mean_power": 7.1814170088864185e-06, "output_mean_power": 9.785991431561787e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 733, "frequency_resolution_hz": 0.2728512960436562, "power_floor": 3.966042949113477e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6674623865712373e-06, "output_power": 3.8125009864701083e-06, "reference_active": true, "delta_db": 1.5511166049908098}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.255436860296439e-07, "output_power": 5.667817417422637e-07, "reference_active": true, "delta_db": 2.4080657492042428}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7820276752795505e-08, "output_power": 3.344207017535708e-08, "reference_active": true, "delta_db": 0.7993170788459116}], "active_band_count": 3, "active_band_mae_db": 1.5861664776803215}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9462722021845555, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "842856bab1d917d34143f10b0b904c82e2f06e82b21d697d6fa4ec5fad3767ef", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9942884259958265, "standard_deviation": null, "minimum": 0.9942884259958265, "maximum": 0.9942884259958265, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2537946974190346, "standard_deviation": null, "minimum": 1.2537946974190346, "maximum": 1.2537946974190346, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.535212429102983, "standard_deviation": null, "minimum": 0.535212429102983, "maximum": 0.535212429102983, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13427156852436525, "standard_deviation": null, "minimum": 0.13427156852436525, "maximum": 0.13427156852436525, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9976503470351075, "standard_deviation": null, "minimum": 0.9976503470351075, "maximum": 0.9976503470351075, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8764750898587593, "standard_deviation": null, "minimum": 0.8764750898587593, "maximum": 0.8764750898587593, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.43024992272364787, "standard_deviation": null, "minimum": 0.43024992272364787, "maximum": 0.43024992272364787, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11403630034560205, "standard_deviation": null, "minimum": 0.11403630034560205, "maximum": 0.11403630034560205, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9527124474332355, "standard_deviation": null, "minimum": 0.9527124474332355, "maximum": 0.9527124474332355, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6815642813029827, "standard_deviation": null, "minimum": 1.6815642813029827, "maximum": 1.6815642813029827, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.8180785766937912, "standard_deviation": null, "minimum": 1.8180785766937912, "maximum": 1.8180785766937912, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3190394494559081, "standard_deviation": null, "minimum": 0.3190394494559081, "maximum": 0.3190394494559081, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9849923842095049, "standard_deviation": null, "minimum": 0.9849923842095049, "maximum": 0.9849923842095049, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1996877634092744, "standard_deviation": null, "minimum": 1.1996877634092744, "maximum": 1.1996877634092744, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5861664776803215, "standard_deviation": null, "minimum": 1.5861664776803215, "maximum": 1.5861664776803215, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2216421638742134, "standard_deviation": null, "minimum": 0.2216421638742134, "maximum": 0.2216421638742134, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9737253576302066, "standard_deviation": null, "minimum": 0.9737253576302066, "maximum": 0.9737253576302066, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6305246684851018, "standard_deviation": null, "minimum": 1.6305246684851018, "maximum": 1.6305246684851018, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.2733851245492345, "standard_deviation": null, "minimum": 1.2733851245492345, "maximum": 1.2733851245492345, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2461866649471397, "standard_deviation": null, "minimum": 0.2461866649471397, "maximum": 0.2461866649471397, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9834069606693465, "standard_deviation": null, "minimum": 0.9834069606693465, "maximum": 0.9834069606693465, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3516519058701681, "standard_deviation": null, "minimum": 1.3516519058701681, "maximum": 1.3516519058701681, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1002240608553517, "standard_deviation": null, "minimum": 1.1002240608553517, "maximum": 1.1002240608553517, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.21812644102033285, "standard_deviation": null, "minimum": 0.21812644102033285, "maximum": 0.21812644102033285, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8812094577836941, "standard_deviation": null, "minimum": 0.8812094577836941, "maximum": 0.8812094577836941, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.6613295154429695, "standard_deviation": null, "minimum": 5.6613295154429695, "maximum": 5.6613295154429695, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6223962629342007, "standard_deviation": null, "minimum": 0.6223962629342007, "maximum": 0.6223962629342007, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0537958457013088, "standard_deviation": null, "minimum": 1.0537958457013088, "maximum": 1.0537958457013088, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9764553709735636, "standard_deviation": null, "minimum": 0.9764553709735636, "maximum": 0.9764553709735636, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.602100048584684, "standard_deviation": null, "minimum": 1.602100048584684, "maximum": 1.602100048584684, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0069512026727623, "standard_deviation": null, "minimum": 1.0069512026727623, "maximum": 1.0069512026727623, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19213794056614766, "standard_deviation": null, "minimum": 0.19213794056614766, "maximum": 0.19213794056614766, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9869566338305001, "standard_deviation": null, "minimum": 0.9869566338305001, "maximum": 0.9869566338305001, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2287411008357092, "standard_deviation": null, "minimum": 1.2287411008357092, "maximum": 1.2287411008357092, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8490182604280255, "standard_deviation": null, "minimum": 0.8490182604280255, "maximum": 0.8490182604280255, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1624209521830076, "standard_deviation": null, "minimum": 0.1624209521830076, "maximum": 0.1624209521830076, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4446"}, {"case": "librispeech_4446-2271-0022", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.611863136291504, "predicted_mos": 2.611863136291504, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37200, "input_sample_rate": 16000, "input_seconds": 2.325, "input_float32_sha256": "b56069bd39e7cc36e3adb48cb510f57044efa910cc41b9e7dc83d1d892232bd0", "internal_samples": 37200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 11.408129699993879}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4570037107522733, "全帯域_10ms_rms_envelope_nrmse": 0.16251508378926563, "全帯域_10ms_envelope_correlation": 0.9879309538745452, "全帯域_10ms_modulation_mae_db": 1.1613398883075632, "全帯域_40ms_log_rms_mae_db": 1.0891494077143213, "全帯域_40ms_rms_envelope_nrmse": 0.1345322237031603, "全帯域_40ms_envelope_correlation": 0.9945412585721628, "全帯域_40ms_modulation_mae_db": 1.1778675776423784, "300–1000 Hz_10ms_log_rms_mae_db": 1.4358307347282302, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1967041968834058, "300–1000 Hz_10ms_envelope_correlation": 0.9873014022886588, "300–1000 Hz_10ms_modulation_mae_db": 1.210352205454632, "300–1000 Hz_40ms_log_rms_mae_db": 1.0371400343714483, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1736628269836745, "300–1000 Hz_40ms_envelope_correlation": 0.9928540233599836, "300–1000 Hz_40ms_modulation_mae_db": 1.3146276538424375, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4162707066091376, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17493180015333296, "1000–3000 Hz_10ms_envelope_correlation": 0.9909558686981804, "1000–3000 Hz_10ms_modulation_mae_db": 1.6384557749499098, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0071848791797677, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14032749031417785, "1000–3000 Hz_40ms_envelope_correlation": 0.9977832797295171, "1000–3000 Hz_40ms_modulation_mae_db": 1.6877148576720995, "3000–6000 Hz_10ms_log_rms_mae_db": 1.44881240130884, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18536192320708422, "3000–6000 Hz_10ms_envelope_correlation": 0.9805540290222176, "3000–6000 Hz_10ms_modulation_mae_db": 1.7333032022587276, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9615727469759708, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11216536277476225, "3000–6000 Hz_40ms_envelope_correlation": 0.9944111056230861, "3000–6000 Hz_40ms_modulation_mae_db": 1.4526277391783644, "mr_stft_db": 5.8509905245849865, "band_mae_db": 0.7258940291641244, "rms_gain_error_db": 0.8983662960590083, "output_peak": 0.5715276287927277}, "full_scale": {"sample_count": 37200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37200, "output_samples": 37200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.068897819812746, "spectral_convergence": 0.54395518799377, "spectral_convergence_status": "ok", "amplitude_floor": 1.3482074212160965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.545547808483336, "spectral_convergence": 0.2302924945092162, "spectral_convergence_status": "ok", "amplitude_floor": 1.080252897815134e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.938525945458878, "spectral_convergence": 0.28118600730423615, "spectral_convergence_status": "ok", "amplitude_floor": 8.596818112459857e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0004510568589627121, "output_power": 0.0004827215910737964, "delta_db": 0.2946543301568766, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0019377343288488795, "output_power": 0.002433431053410981, "delta_db": 0.9892481257467044, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0013786956670426643, "output_power": 0.0017057154283994238, "delta_db": 0.9243816702656893, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00031975863120548915, "output_power": 0.0004082251047726174, "delta_db": 1.0607743300565458, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.4830953182024825e-05, "output_power": 1.7348867914746186e-05, "delta_db": 0.6810207660959078, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.03312086795025e-05, "output_power": 2.2319871669345303e-05, "delta_db": 0.4052849526630222, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7258940291641244, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8983662960590083, "pitch": null, "output_rms_dbfs": -22.950124352339863, "output_peak": 0.5715276287927277, "reference_rms_dbfs": -23.848490648398872, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.8509905245849865}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00019668187909197959, "log_rms_mae_db": 1.4570037107522733, "rms_envelope_nrmse": 0.16251508378926563, "envelope_correlation": 0.9879309538745452, "rms_gain_error_db": 0.8983663439779205, "reference_mean_power": 0.004131292024738224, "output_mean_power": 0.005080688033770329}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.0026327800860698e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011612269501705267, "output_power": 0.0013610125346044351, "reference_active": true, "delta_db": 0.6894501843318015}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017095886275955477, "output_power": 0.00024961532002867556, "reference_active": true, "delta_db": 1.6437961627544049}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.217923048678998e-05, "output_power": 5.497636029588305e-05, "reference_active": true, "delta_db": 1.150773317836483}], "active_band_count": 3, "active_band_mae_db": 1.1613398883075632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00018115923356648065, "log_rms_mae_db": 1.0891494077143213, "rms_envelope_nrmse": 0.1345322237031603, "envelope_correlation": 0.9945412585721628, "rms_gain_error_db": 0.8983668246396526, "reference_mean_power": 0.004185412442455258, "output_mean_power": 0.005147246221993545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.7728315536386115e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066252574844359, "output_power": 0.001245181626019052, "reference_active": true, "delta_db": 0.6737261064553837}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.440983264620606e-05, "output_power": 0.0001183185103170255, "reference_active": true, "delta_db": 1.4665965383786717}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.940091323355611e-06, "output_power": 4.052180937867786e-06, "reference_active": true, "delta_db": 1.3932800880930796}], "active_band_count": 3, "active_band_mae_db": 1.1778675776423784}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.000176021246464811, "log_rms_mae_db": 1.4358307347282302, "rms_envelope_nrmse": 0.1967041968834058, "envelope_correlation": 0.9873014022886588, "rms_gain_error_db": 1.156034135667934, "reference_mean_power": 0.0026395436088586166, "output_mean_power": 0.003444548111418038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.515899729654865e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009091516678957249, "output_power": 0.0011388834307314063, "reference_active": true, "delta_db": 0.9784293477189658}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001833276731517994, "output_power": 0.0003053280292944075, "reference_active": true, "delta_db": 2.215386483376252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7517429693718724e-05, "output_power": 4.149131085172993e-05, "reference_active": true, "delta_db": 0.4372407852686785}], "active_band_count": 3, "active_band_mae_db": 1.210352205454632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 0.00016136314602147107, "log_rms_mae_db": 1.0371400343714483, "rms_envelope_nrmse": 0.1736628269836745, "envelope_correlation": 0.9928540233599836, "rms_gain_error_db": 1.1560340215855227, "reference_mean_power": 0.002674122622482191, "output_mean_power": 0.0034896728950016857}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.348456388983372e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008380618007305507, "output_power": 0.00105246231119929, "reference_active": true, "delta_db": 0.989305068986947}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.026333057603631e-05, "output_power": 0.00014875366935178055, "reference_active": true, "delta_db": 2.1695633313772182}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498114091574834e-06, "output_power": 2.9930483001967656e-06, "reference_active": true, "delta_db": 0.785014561163147}], "active_band_count": 3, "active_band_mae_db": 1.3146276538424375}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 6.667691377903701e-05, "log_rms_mae_db": 1.4162707066091376, "rms_envelope_nrmse": 0.17493180015333296, "envelope_correlation": 0.9909558686981804, "rms_gain_error_db": 1.0455233567351154, "reference_mean_power": 0.0003102146198149334, "output_mean_power": 0.0003946522626626703}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 3.247013787334141e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018435630971779633, "output_power": 0.00023478494428099276, "reference_active": true, "delta_db": 1.0501223787097456}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0328682781544803e-05, "output_power": 2.4568990873354297e-05, "reference_active": true, "delta_db": 0.8227808000949056}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.580926194982421e-06, "output_power": 9.229957671412462e-06, "reference_active": true, "delta_db": 3.0424641460450785}], "active_band_count": 3, "active_band_mae_db": 1.6384557749499098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 6.131102329922278e-05, "log_rms_mae_db": 1.0071848791797677, "rms_envelope_nrmse": 0.14032749031417785, "envelope_correlation": 0.9977832797295171, "rms_gain_error_db": 1.0455243304659658, "reference_mean_power": 0.0003142782197045969, "output_mean_power": 0.0003998220277138838}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 3.052275331975406e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017569545994603763, "output_power": 0.00022384240763315981, "reference_active": true, "delta_db": 1.0518182914868621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.589536328721284e-06, "output_power": 1.068835870850582e-05, "reference_active": true, "delta_db": 0.9494129975300516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8479520444652144e-07, "output_power": 5.764000376554248e-07, "reference_active": true, "delta_db": 3.0619132839993854}], "active_band_count": 3, "active_band_mae_db": 1.6877148576720995}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 2.0872644535346135e-05, "log_rms_mae_db": 1.44881240130884, "rms_envelope_nrmse": 0.18536192320708422, "envelope_correlation": 0.9805540290222176, "rms_gain_error_db": 0.5452503927708504, "reference_mean_power": 1.984318614204444e-05, "output_mean_power": 2.2497613196647696e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 2.6220002221112295e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.352365504647456e-06, "output_power": 1.0389483812125325e-05, "reference_active": true, "delta_db": 0.4567249936816275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.791249366368675e-07, "output_power": 8.355944405379383e-07, "reference_active": true, "delta_db": -0.6884256962012218}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495952213974179e-07, "output_power": 6.349099863352621e-07, "reference_active": true, "delta_db": 4.054758916893333}], "active_band_count": 3, "active_band_mae_db": 1.7333032022587276}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 458, "rms_floor": 1.9522833601394496e-05, "log_rms_mae_db": 0.9615727469759708, "rms_envelope_nrmse": 0.11216536277476225, "envelope_correlation": 0.9944111056230861, "rms_gain_error_db": 0.5452601868266246, "reference_mean_power": 2.010291979320635e-05, "output_mean_power": 2.2792142871241396e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 458, "frequency_resolution_hz": 0.4366812227074236, "power_floor": 2.5131001811961146e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.175181026763155e-06, "output_power": 9.128571604637217e-06, "reference_active": true, "delta_db": 0.4790544830087896}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.283226898133603e-07, "output_power": 3.6757519432702094e-07, "reference_active": true, "delta_db": -0.6642488566280849}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.754857048422832e-08, "output_power": 3.678745401025764e-08, "reference_active": true, "delta_db": 3.214579877898218}], "active_band_count": 3, "active_band_mae_db": 1.4526277391783644}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9319895399954096, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9baa00c5a12bb00b8557433abd68482c78bb456689c2b2596030bf9a235d6f9f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9909558686981804, "standard_deviation": null, "minimum": 0.9909558686981804, "maximum": 0.9909558686981804, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4162707066091376, "standard_deviation": null, "minimum": 1.4162707066091376, "maximum": 1.4162707066091376, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6384557749499098, "standard_deviation": null, "minimum": 1.6384557749499098, "maximum": 1.6384557749499098, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17493180015333296, "standard_deviation": null, "minimum": 0.17493180015333296, "maximum": 0.17493180015333296, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9977832797295171, "standard_deviation": null, "minimum": 0.9977832797295171, "maximum": 0.9977832797295171, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0071848791797677, "standard_deviation": null, "minimum": 1.0071848791797677, "maximum": 1.0071848791797677, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.6877148576720995, "standard_deviation": null, "minimum": 1.6877148576720995, "maximum": 1.6877148576720995, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14032749031417785, "standard_deviation": null, "minimum": 0.14032749031417785, "maximum": 0.14032749031417785, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9805540290222176, "standard_deviation": null, "minimum": 0.9805540290222176, "maximum": 0.9805540290222176, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.44881240130884, "standard_deviation": null, "minimum": 1.44881240130884, "maximum": 1.44881240130884, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7333032022587276, "standard_deviation": null, "minimum": 1.7333032022587276, "maximum": 1.7333032022587276, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18536192320708422, "standard_deviation": null, "minimum": 0.18536192320708422, "maximum": 0.18536192320708422, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9944111056230861, "standard_deviation": null, "minimum": 0.9944111056230861, "maximum": 0.9944111056230861, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9615727469759708, "standard_deviation": null, "minimum": 0.9615727469759708, "maximum": 0.9615727469759708, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.4526277391783644, "standard_deviation": null, "minimum": 1.4526277391783644, "maximum": 1.4526277391783644, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11216536277476225, "standard_deviation": null, "minimum": 0.11216536277476225, "maximum": 0.11216536277476225, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9873014022886588, "standard_deviation": null, "minimum": 0.9873014022886588, "maximum": 0.9873014022886588, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4358307347282302, "standard_deviation": null, "minimum": 1.4358307347282302, "maximum": 1.4358307347282302, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.210352205454632, "standard_deviation": null, "minimum": 1.210352205454632, "maximum": 1.210352205454632, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1967041968834058, "standard_deviation": null, "minimum": 0.1967041968834058, "maximum": 0.1967041968834058, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9928540233599836, "standard_deviation": null, "minimum": 0.9928540233599836, "maximum": 0.9928540233599836, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0371400343714483, "standard_deviation": null, "minimum": 1.0371400343714483, "maximum": 1.0371400343714483, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3146276538424375, "standard_deviation": null, "minimum": 1.3146276538424375, "maximum": 1.3146276538424375, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1736628269836745, "standard_deviation": null, "minimum": 0.1736628269836745, "maximum": 0.1736628269836745, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7258940291641244, "standard_deviation": null, "minimum": 0.7258940291641244, "maximum": 0.7258940291641244, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.8509905245849865, "standard_deviation": null, "minimum": 5.8509905245849865, "maximum": 5.8509905245849865, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5715276287927277, "standard_deviation": null, "minimum": 0.5715276287927277, "maximum": 0.5715276287927277, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8983662960590083, "standard_deviation": null, "minimum": 0.8983662960590083, "maximum": 0.8983662960590083, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9879309538745452, "standard_deviation": null, "minimum": 0.9879309538745452, "maximum": 0.9879309538745452, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4570037107522733, "standard_deviation": null, "minimum": 1.4570037107522733, "maximum": 1.4570037107522733, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1613398883075632, "standard_deviation": null, "minimum": 1.1613398883075632, "maximum": 1.1613398883075632, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16251508378926563, "standard_deviation": null, "minimum": 0.16251508378926563, "maximum": 0.16251508378926563, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9945412585721628, "standard_deviation": null, "minimum": 0.9945412585721628, "maximum": 0.9945412585721628, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0891494077143213, "standard_deviation": null, "minimum": 1.0891494077143213, "maximum": 1.0891494077143213, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1778675776423784, "standard_deviation": null, "minimum": 1.1778675776423784, "maximum": 1.1778675776423784, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1345322237031603, "standard_deviation": null, "minimum": 0.1345322237031603, "maximum": 0.1345322237031603, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4446"}, {"case": "librispeech_4446-2273-0010", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.5266623497009277, "predicted_mos": 3.5266623497009277, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7ac83aedae8547fba48c0dcd2e0bcf3bd61328635a5a20426d806f106d3dbe82", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.742633199784905}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.340608484012953, "全帯域_10ms_rms_envelope_nrmse": 0.1767232626565126, "全帯域_10ms_envelope_correlation": 0.9809949333802, "全帯域_10ms_modulation_mae_db": 0.8333445165422334, "全帯域_40ms_log_rms_mae_db": 1.6707939636868465, "全帯域_40ms_rms_envelope_nrmse": 0.13599133207951483, "全帯域_40ms_envelope_correlation": 0.9914784330749082, "全帯域_40ms_modulation_mae_db": 0.7670380713529905, "300–1000 Hz_10ms_log_rms_mae_db": 1.8632107419193096, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20042965461507772, "300–1000 Hz_10ms_envelope_correlation": 0.9897707180808673, "300–1000 Hz_10ms_modulation_mae_db": 0.8839573690771604, "300–1000 Hz_40ms_log_rms_mae_db": 1.2769950193195738, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1760703858240272, "300–1000 Hz_40ms_envelope_correlation": 0.9950749752242416, "300–1000 Hz_40ms_modulation_mae_db": 0.986927125711072, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6472073090870925, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17118383859468114, "1000–3000 Hz_10ms_envelope_correlation": 0.9864559674577256, "1000–3000 Hz_10ms_modulation_mae_db": 0.903340822391411, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0318361485001548, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12374795713915558, "1000–3000 Hz_40ms_envelope_correlation": 0.9955571881900864, "1000–3000 Hz_40ms_modulation_mae_db": 0.8835282372788709, "3000–6000 Hz_10ms_log_rms_mae_db": 1.696274847991242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2229854910107463, "3000–6000 Hz_10ms_envelope_correlation": 0.9658099442202092, "3000–6000 Hz_10ms_modulation_mae_db": 0.46123192226620136, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1251447002210058, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11709475614328345, "3000–6000 Hz_40ms_envelope_correlation": 0.990523702520424, "3000–6000 Hz_40ms_modulation_mae_db": 0.7493328247988087, "mr_stft_db": 5.930424480748319, "band_mae_db": 0.7507994208990793, "rms_gain_error_db": 0.8299261924490764, "output_peak": 0.4508901122679371}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.470916323592949, "spectral_convergence": 0.6034273131933267, "spectral_convergence_status": "ok", "amplitude_floor": 9.383482434461407e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.552168411729018, "spectral_convergence": 0.4278800936094615, "spectral_convergence_status": "ok", "amplitude_floor": 6.715789372234631e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.768188706922991, "spectral_convergence": 0.4462262472654315, "spectral_convergence_status": "ok", "amplitude_floor": 5.217365851457555e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.0167397027219553e-05, "output_power": 2.703235923696953e-05, "delta_db": -0.4765388878755755, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003363761740772986, "output_power": 0.00047073589209047437, "delta_db": 1.4595208649690414, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005972956425120644, "output_power": 0.0007538690265529585, "delta_db": 1.0110655384162754, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016539661163734624, "output_power": 0.00020109721891842865, "delta_db": 0.8487945632062028, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011856598981285755, "output_power": 0.0001331380256248036, "delta_db": 0.5034198104716732, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00046450684077542097, "output_power": 0.0004870099118450176, "delta_db": 0.20545686045570682, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7507994208990793, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8299261924490764, "pitch": null, "output_rms_dbfs": -26.834253286809904, "output_peak": 0.4508901122679371, "reference_rms_dbfs": -27.66417947925898, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.930424480748319}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014203698766261565, "log_rms_mae_db": 2.340608484012953, "rms_envelope_nrmse": 0.1767232626565126, "envelope_correlation": 0.9809949333802, "rms_gain_error_db": 0.8299387332508217, "reference_mean_power": 0.0017137270951261038, "output_mean_power": 0.0020746055557126684}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.499341088840717e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046355332033754323, "output_power": 0.0005636417988919399, "reference_active": true, "delta_db": 0.8490349612033108}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011665881324257648, "output_power": 0.00015365570097794533, "reference_active": true, "delta_db": 1.1963112412399308}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5955876441113554e-05, "output_power": 2.337582586484621e-05, "reference_active": true, "delta_db": -0.4546873471834587}], "active_band_count": 3, "active_band_mae_db": 0.8333445165422334}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274207517780892, "log_rms_mae_db": 1.6707939636868465, "rms_envelope_nrmse": 0.13599133207951483, "envelope_correlation": 0.9914784330749082, "rms_gain_error_db": 0.8293651464703222, "reference_mean_power": 0.0017200719561396216, "output_mean_power": 0.0020820115280941512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.335969034308142e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004202066244366408, "output_power": 0.000511526698323839, "reference_active": true, "delta_db": 0.8540541129910546}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.868302162815529e-05, "output_power": 8.083083352461708e-05, "reference_active": true, "delta_db": 1.3906458938931716}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5858058061398606e-06, "output_power": 1.5653396386278844e-06, "reference_active": true, "delta_db": -0.05641420717474526}], "active_band_count": 3, "active_band_mae_db": 0.7670380713529905}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000125392198184175, "log_rms_mae_db": 1.8632107419193096, "rms_envelope_nrmse": 0.20042965461507772, "envelope_correlation": 0.9897707180808673, "rms_gain_error_db": 1.2402926502571647, "reference_mean_power": 0.0008371717774342565, "output_mean_power": 0.0011138939473448119}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.199403135704403e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003464696140419235, "output_power": 0.0004478481548533283, "reference_active": true, "delta_db": 1.1146563697528227}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.952130707503092e-05, "output_power": 0.0001254421559894561, "reference_active": true, "delta_db": 1.4651709509986965}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1976615680578699e-05, "output_power": 1.1779575361845622e-05, "reference_active": true, "delta_db": -0.07204478647996156}], "active_band_count": 3, "active_band_mae_db": 0.8839573690771604}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011133215930497508, "log_rms_mae_db": 1.2769950193195738, "rms_envelope_nrmse": 0.1760703858240272, "envelope_correlation": 0.9950749752242416, "rms_gain_error_db": 1.2402446587031855, "reference_mean_power": 0.0008413676993576814, "output_mean_power": 0.0011194644357253526}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.575111583882504e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003227285828337414, "output_power": 0.00041647814024182604, "reference_active": true, "delta_db": 1.1075478054835788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.553081813418887e-05, "output_power": 6.509765630133838e-05, "reference_active": true, "delta_db": 1.552598988979819}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.055134444571575e-06, "output_power": 1.1307621396787241e-06, "reference_active": true, "delta_db": 0.300634582669818}], "active_band_count": 3, "active_band_mae_db": 0.986927125711072}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.277196005950351e-05, "log_rms_mae_db": 1.6472073090870925, "rms_envelope_nrmse": 0.17118383859468114, "envelope_correlation": 0.9864559674577256, "rms_gain_error_db": 0.7975947337097543, "reference_mean_power": 0.0001860033831933617, "output_mean_power": 0.00022350143557077753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.45795784914854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.813551789415923e-05, "output_power": 0.00010717009530172542, "reference_active": true, "delta_db": 0.8492265609464622}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.039396685146323e-05, "output_power": 2.517319144803679e-05, "reference_active": true, "delta_db": 0.914365696738801}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.503164081364961e-06, "output_power": 3.621385435814889e-06, "reference_active": true, "delta_db": -0.9464302094889698}], "active_band_count": 3, "active_band_mae_db": 0.903340822391411}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5906277437135206e-05, "log_rms_mae_db": 1.0318361485001548, "rms_envelope_nrmse": 0.12374795713915558, "envelope_correlation": 0.9955571881900864, "rms_gain_error_db": 0.7976907779452309, "reference_mean_power": 0.00018692252747526107, "output_mean_power": 0.00022461084541944135}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.315230721785866e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.510469005068111e-05, "output_power": 0.00010296535808249391, "reference_active": true, "delta_db": 0.8273763981713784}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648548828731175e-06, "output_power": 1.2280824050063115e-05, "reference_active": true, "delta_db": 1.047655101456524}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.651744211074769e-07, "output_power": 3.054534454549407e-07, "reference_active": true, "delta_db": -0.77555321220871}], "active_band_count": 3, "active_band_mae_db": 0.8835282372788709}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.827493119721784e-05, "log_rms_mae_db": 1.696274847991242, "rms_envelope_nrmse": 0.2229854910107463, "envelope_correlation": 0.9658099442202092, "rms_gain_error_db": 0.3058205176983918, "reference_mean_power": 0.0003389197953959056, "output_mean_power": 0.0003636461444407948}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5434076582307222e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.258901619859077e-05, "output_power": 6.330377667168874e-05, "reference_active": true, "delta_db": 0.04931495404900657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4223152137238325e-05, "output_power": 1.4010108028147068e-05, "reference_active": true, "delta_db": -0.06554371446989475}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.429050160653041e-06, "output_power": 6.293539009851228e-06, "reference_active": true, "delta_db": -1.2688370982797028}], "active_band_count": 3, "active_band_mae_db": 0.46123192226620136}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.761964716205697e-05, "log_rms_mae_db": 1.1251447002210058, "rms_envelope_nrmse": 0.11709475614328345, "envelope_correlation": 0.990523702520424, "rms_gain_error_db": 0.2994752668710702, "reference_mean_power": 0.00033859490771126537, "output_mean_power": 0.00036276714654477816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4129997856527214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.165966182107609e-05, "output_power": 6.121715060514226e-05, "reference_active": true, "delta_db": -0.031280272700978655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.083964993529701e-06, "output_power": 6.930914064845311e-06, "reference_active": true, "delta_db": -0.09485892250893888}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.531889318612958e-07, "output_power": 4.0073095296858083e-07, "reference_active": true, "delta_db": -2.1218592791865087}], "active_band_count": 3, "active_band_mae_db": 0.7493328247988087}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9104556478738363, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c207b4ea5a92e50e0d46589626ad79ea52e1228a840cd683290df9e9725dec4f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9864559674577256, "standard_deviation": null, "minimum": 0.9864559674577256, "maximum": 0.9864559674577256, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6472073090870925, "standard_deviation": null, "minimum": 1.6472073090870925, "maximum": 1.6472073090870925, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.903340822391411, "standard_deviation": null, "minimum": 0.903340822391411, "maximum": 0.903340822391411, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17118383859468114, "standard_deviation": null, "minimum": 0.17118383859468114, "maximum": 0.17118383859468114, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955571881900864, "standard_deviation": null, "minimum": 0.9955571881900864, "maximum": 0.9955571881900864, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0318361485001548, "standard_deviation": null, "minimum": 1.0318361485001548, "maximum": 1.0318361485001548, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8835282372788709, "standard_deviation": null, "minimum": 0.8835282372788709, "maximum": 0.8835282372788709, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12374795713915558, "standard_deviation": null, "minimum": 0.12374795713915558, "maximum": 0.12374795713915558, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9658099442202092, "standard_deviation": null, "minimum": 0.9658099442202092, "maximum": 0.9658099442202092, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.696274847991242, "standard_deviation": null, "minimum": 1.696274847991242, "maximum": 1.696274847991242, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.46123192226620136, "standard_deviation": null, "minimum": 0.46123192226620136, "maximum": 0.46123192226620136, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2229854910107463, "standard_deviation": null, "minimum": 0.2229854910107463, "maximum": 0.2229854910107463, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.990523702520424, "standard_deviation": null, "minimum": 0.990523702520424, "maximum": 0.990523702520424, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1251447002210058, "standard_deviation": null, "minimum": 1.1251447002210058, "maximum": 1.1251447002210058, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7493328247988087, "standard_deviation": null, "minimum": 0.7493328247988087, "maximum": 0.7493328247988087, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11709475614328345, "standard_deviation": null, "minimum": 0.11709475614328345, "maximum": 0.11709475614328345, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9897707180808673, "standard_deviation": null, "minimum": 0.9897707180808673, "maximum": 0.9897707180808673, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8632107419193096, "standard_deviation": null, "minimum": 1.8632107419193096, "maximum": 1.8632107419193096, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8839573690771604, "standard_deviation": null, "minimum": 0.8839573690771604, "maximum": 0.8839573690771604, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20042965461507772, "standard_deviation": null, "minimum": 0.20042965461507772, "maximum": 0.20042965461507772, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9950749752242416, "standard_deviation": null, "minimum": 0.9950749752242416, "maximum": 0.9950749752242416, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2769950193195738, "standard_deviation": null, "minimum": 1.2769950193195738, "maximum": 1.2769950193195738, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.986927125711072, "standard_deviation": null, "minimum": 0.986927125711072, "maximum": 0.986927125711072, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1760703858240272, "standard_deviation": null, "minimum": 0.1760703858240272, "maximum": 0.1760703858240272, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7507994208990793, "standard_deviation": null, "minimum": 0.7507994208990793, "maximum": 0.7507994208990793, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.930424480748319, "standard_deviation": null, "minimum": 5.930424480748319, "maximum": 5.930424480748319, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4508901122679371, "standard_deviation": null, "minimum": 0.4508901122679371, "maximum": 0.4508901122679371, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8299261924490764, "standard_deviation": null, "minimum": 0.8299261924490764, "maximum": 0.8299261924490764, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9809949333802, "standard_deviation": null, "minimum": 0.9809949333802, "maximum": 0.9809949333802, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.340608484012953, "standard_deviation": null, "minimum": 2.340608484012953, "maximum": 2.340608484012953, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8333445165422334, "standard_deviation": null, "minimum": 0.8333445165422334, "maximum": 0.8333445165422334, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1767232626565126, "standard_deviation": null, "minimum": 0.1767232626565126, "maximum": 0.1767232626565126, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9914784330749082, "standard_deviation": null, "minimum": 0.9914784330749082, "maximum": 0.9914784330749082, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.6707939636868465, "standard_deviation": null, "minimum": 1.6707939636868465, "maximum": 1.6707939636868465, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7670380713529905, "standard_deviation": null, "minimum": 0.7670380713529905, "maximum": 0.7670380713529905, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.13599133207951483, "standard_deviation": null, "minimum": 0.13599133207951483, "maximum": 0.13599133207951483, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4446"}, {"case": "librispeech_4507-16021-0016", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.6437742710113525, "predicted_mos": 2.6437742710113525, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e318d4d1ca9b6ea7f94280d4db65a7127f577cb2f15430b011f865ae9848d979", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.656214800197631}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4342985756498565, "全帯域_10ms_rms_envelope_nrmse": 0.1993595598460629, "全帯域_10ms_envelope_correlation": 0.9877044970056253, "全帯域_10ms_modulation_mae_db": 1.319283116457344, "全帯域_40ms_log_rms_mae_db": 1.0830428309403974, "全帯域_40ms_rms_envelope_nrmse": 0.18122179456465135, "全帯域_40ms_envelope_correlation": 0.9930754117090344, "全帯域_40ms_modulation_mae_db": 1.4783520196296103, "300–1000 Hz_10ms_log_rms_mae_db": 2.02126173134384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2398036334315706, "300–1000 Hz_10ms_envelope_correlation": 0.9872316093366285, "300–1000 Hz_10ms_modulation_mae_db": 1.7098121815227805, "300–1000 Hz_40ms_log_rms_mae_db": 1.590954842586998, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2244159528210191, "300–1000 Hz_40ms_envelope_correlation": 0.9916913770518365, "300–1000 Hz_40ms_modulation_mae_db": 1.7197389310850693, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4211498300789374, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1715540581012039, "1000–3000 Hz_10ms_envelope_correlation": 0.9915800436360523, "1000–3000 Hz_10ms_modulation_mae_db": 1.0973803216343183, "1000–3000 Hz_40ms_log_rms_mae_db": 1.019254797724492, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1470102200594804, "1000–3000 Hz_40ms_envelope_correlation": 0.9968143434080312, "1000–3000 Hz_40ms_modulation_mae_db": 0.8332203626607101, "3000–6000 Hz_10ms_log_rms_mae_db": 1.2818527093695347, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18329737225037337, "3000–6000 Hz_10ms_envelope_correlation": 0.9736634844782628, "3000–6000 Hz_10ms_modulation_mae_db": 0.1706869570556362, "3000–6000 Hz_40ms_log_rms_mae_db": 0.838613978502151, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1062934123274322, "3000–6000 Hz_40ms_envelope_correlation": 0.9906944522047073, "3000–6000 Hz_40ms_modulation_mae_db": 0.47834466415037874, "mr_stft_db": 5.841513745776106, "band_mae_db": 1.030027895721248, "rms_gain_error_db": 1.2792004272604944, "output_peak": 0.3340404822926332}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.95410290798753, "spectral_convergence": 0.49841228590832454, "spectral_convergence_status": "ok", "amplitude_floor": 9.54303147153e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.537759473036151, "spectral_convergence": 0.2742089135049768, "spectral_convergence_status": "ok", "amplitude_floor": 7.553612828866807e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.032678856304638, "spectral_convergence": 0.31919775720611854, "spectral_convergence_status": "ok", "amplitude_floor": 6.230060381303045e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.459798737171228e-05, "output_power": 7.518325888692141e-05, "delta_db": 1.3894451293608643, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011571142287648336, "output_power": 0.0015077781171263934, "delta_db": 1.1496120192818258, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009701755098987012, "output_power": 0.0013610165539918282, "delta_db": 1.47013099992179, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.8727504132367385e-05, "output_power": 6.296459264140134e-05, "delta_db": 1.1132223098584575, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.5070297613490014e-05, "output_power": 3.013217357645833e-05, "delta_db": 0.7987097095447946, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.1228013015272054e-05, "output_power": 3.314736971758436e-05, "delta_db": 0.2590472063597568, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.030027895721248, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.2792004272604944, "pitch": null, "output_rms_dbfs": -25.128302113225672, "output_peak": 0.3340404822926332, "reference_rms_dbfs": -26.407502540486163, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.841513745776106}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014140982422473965, "log_rms_mae_db": 1.4342985756498565, "rms_envelope_nrmse": 0.1993595598460629, "envelope_correlation": 0.9877044970056253, "rms_gain_error_db": 1.279201026988655, "reference_mean_power": 0.0022888204000268467, "output_mean_power": 0.003072782482669744}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0460573675773498e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006367944533168533, "output_power": 0.0009386093041368684, "reference_active": true, "delta_db": 1.6848558301912346}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.21105688014902e-05, "output_power": 0.0001492789001764946, "reference_active": true, "delta_db": 2.096889625480648}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2041191998800966e-05, "output_power": 1.2539489921337194e-05, "reference_active": true, "delta_db": 0.17610389370014878}], "active_band_count": 3, "active_band_mae_db": 1.319283116457344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001274322977924003, "log_rms_mae_db": 1.0830428309403974, "rms_envelope_nrmse": 0.18122179456465135, "envelope_correlation": 0.9930754117090344, "rms_gain_error_db": 1.279203360762779, "reference_mean_power": 0.0023003276499839287, "output_mean_power": 0.0030882328318765154}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.201135437818204e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005685810358795958, "output_power": 0.0008490778463897963, "reference_active": true, "delta_db": 1.741551392965749}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.696081479723381e-05, "output_power": 7.82080923627931e-05, "reference_active": true, "delta_db": 2.2151606922237086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.36988853805451e-07, "output_power": 1.0460894496281912e-06, "reference_active": true, "delta_db": 0.47834397369937354}], "active_band_count": 3, "active_band_mae_db": 1.4783520196296103}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012682596688393775, "log_rms_mae_db": 2.02126173134384, "rms_envelope_nrmse": 0.2398036334315706, "envelope_correlation": 0.9872316093366285, "rms_gain_error_db": 1.5778721393853468, "reference_mean_power": 0.0016135328798776547, "output_mean_power": 0.002320411630701642}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.436393406849114e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006358276871928121, "output_power": 0.0010081345403967561, "reference_active": true, "delta_db": 2.0017905926616626}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.784078725837931e-05, "output_power": 0.00015417217431395048, "reference_active": true, "delta_db": 2.4430977795488977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0842784949487312e-05, "output_power": 1.269391721000847e-05, "reference_active": true, "delta_db": 0.6845481723577815}], "active_band_count": 3, "active_band_mae_db": 1.7098121815227805}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011190972053745482, "log_rms_mae_db": 1.590954842586998, "rms_envelope_nrmse": 0.2244159528210191, "envelope_correlation": 0.9916913770518365, "rms_gain_error_db": 1.5778731756850002, "reference_mean_power": 0.0016216465744125685, "output_mean_power": 0.002332080440981201}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.451026691855673e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005862792692297138, "output_power": 0.0009347992842317322, "reference_active": true, "delta_db": 2.0261383333321397}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.32096116317955e-05, "output_power": 7.686323522849761e-05, "reference_active": true, "delta_db": 2.5013829739033087}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.999199919742325e-07, "output_power": 1.1564711422457708e-06, "reference_active": true, "delta_db": 0.6316954860197604}], "active_band_count": 3, "active_band_mae_db": 1.7197389310850693}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.381727808324483e-05, "log_rms_mae_db": 1.4211498300789374, "rms_envelope_nrmse": 0.1715540581012039, "envelope_correlation": 0.9915800436360523, "rms_gain_error_db": 1.0710156086786755, "reference_mean_power": 6.869686019984948e-05, "output_mean_power": 8.791003417524653e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.573688907938692e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3586087172622807e-05, "output_power": 1.7560081919110758e-05, "reference_active": true, "delta_db": 1.1143214075732772}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.445561080668119e-06, "output_power": 3.364002439784246e-06, "reference_active": true, "delta_db": 1.3847778781226074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.901755177746097e-07, "output_power": 7.084111002690086e-07, "reference_active": true, "delta_db": 0.79304167920707}], "active_band_count": 3, "active_band_mae_db": 1.0973803216343183}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.1112752780143365e-05, "log_rms_mae_db": 1.019254797724492, "rms_envelope_nrmse": 0.1470102200594804, "envelope_correlation": 0.9968143434080312, "rms_gain_error_db": 1.0710381453018134, "reference_mean_power": 6.904136883729842e-05, "output_mean_power": 8.835135364553513e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.335283050026158e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2478970323107004e-05, "output_power": 1.6224117016200717e-05, "reference_active": true, "delta_db": 1.1398231810940689}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3575517045580512e-06, "output_power": 1.8449581324173497e-06, "reference_active": true, "delta_db": 1.3323013578905099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.9851801848052634e-08, "output_power": 3.959991988890795e-08, "reference_active": true, "delta_db": -0.02753654899755164}], "active_band_count": 3, "active_band_mae_db": 0.8332203626607101}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.653785524000496e-05, "log_rms_mae_db": 1.2818527093695347, "rms_envelope_nrmse": 0.18329737225037337, "envelope_correlation": 0.9736634844782628, "rms_gain_error_db": 0.2388672747054229, "reference_mean_power": 1.4641647362050651e-05, "output_mean_power": 1.5469514016687552e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3222330564047269e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.030474467255676e-06, "output_power": 6.0225225379120455e-06, "reference_active": true, "delta_db": -0.005730490987434388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.175765551110617e-06, "output_power": 2.1825839494405247e-06, "reference_active": true, "delta_db": 0.013588608190117122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1480409466014962e-06, "output_power": 1.0249040302113673e-06, "reference_active": true, "delta_db": -0.49274177198935704}], "active_band_count": 3, "active_band_mae_db": 0.1706869570556362}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.7197117785089643e-05, "log_rms_mae_db": 0.838613978502151, "rms_envelope_nrmse": 0.1062934123274322, "envelope_correlation": 0.9906944522047073, "rms_gain_error_db": 0.23888651755837778, "reference_mean_power": 1.4714471928104246e-05, "output_mean_power": 1.554652510640544e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.175274300154406e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.965217773921107e-06, "output_power": 5.823252917684011e-06, "reference_active": true, "delta_db": -0.10460650252416699}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4021233434065406e-06, "output_power": 1.3227003963697398e-06, "reference_active": true, "delta_db": -0.25324736078285887}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593565327256392e-08, "output_power": 7.486210322574363e-08, "reference_active": true, "delta_db": -1.0771801291441103}], "active_band_count": 3, "active_band_mae_db": 0.47834466415037874}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9212752626392743, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "00bb11d0a05333a4ad37b992fc47f868b2ac807117a06389fc7a4d3b40b37f45", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9915800436360523, "standard_deviation": null, "minimum": 0.9915800436360523, "maximum": 0.9915800436360523, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4211498300789374, "standard_deviation": null, "minimum": 1.4211498300789374, "maximum": 1.4211498300789374, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0973803216343183, "standard_deviation": null, "minimum": 1.0973803216343183, "maximum": 1.0973803216343183, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1715540581012039, "standard_deviation": null, "minimum": 0.1715540581012039, "maximum": 0.1715540581012039, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9968143434080312, "standard_deviation": null, "minimum": 0.9968143434080312, "maximum": 0.9968143434080312, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.019254797724492, "standard_deviation": null, "minimum": 1.019254797724492, "maximum": 1.019254797724492, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8332203626607101, "standard_deviation": null, "minimum": 0.8332203626607101, "maximum": 0.8332203626607101, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1470102200594804, "standard_deviation": null, "minimum": 0.1470102200594804, "maximum": 0.1470102200594804, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9736634844782628, "standard_deviation": null, "minimum": 0.9736634844782628, "maximum": 0.9736634844782628, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.2818527093695347, "standard_deviation": null, "minimum": 1.2818527093695347, "maximum": 1.2818527093695347, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.1706869570556362, "standard_deviation": null, "minimum": 0.1706869570556362, "maximum": 0.1706869570556362, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18329737225037337, "standard_deviation": null, "minimum": 0.18329737225037337, "maximum": 0.18329737225037337, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9906944522047073, "standard_deviation": null, "minimum": 0.9906944522047073, "maximum": 0.9906944522047073, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.838613978502151, "standard_deviation": null, "minimum": 0.838613978502151, "maximum": 0.838613978502151, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.47834466415037874, "standard_deviation": null, "minimum": 0.47834466415037874, "maximum": 0.47834466415037874, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1062934123274322, "standard_deviation": null, "minimum": 0.1062934123274322, "maximum": 0.1062934123274322, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9872316093366285, "standard_deviation": null, "minimum": 0.9872316093366285, "maximum": 0.9872316093366285, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.02126173134384, "standard_deviation": null, "minimum": 2.02126173134384, "maximum": 2.02126173134384, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.7098121815227805, "standard_deviation": null, "minimum": 1.7098121815227805, "maximum": 1.7098121815227805, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2398036334315706, "standard_deviation": null, "minimum": 0.2398036334315706, "maximum": 0.2398036334315706, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9916913770518365, "standard_deviation": null, "minimum": 0.9916913770518365, "maximum": 0.9916913770518365, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.590954842586998, "standard_deviation": null, "minimum": 1.590954842586998, "maximum": 1.590954842586998, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7197389310850693, "standard_deviation": null, "minimum": 1.7197389310850693, "maximum": 1.7197389310850693, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2244159528210191, "standard_deviation": null, "minimum": 0.2244159528210191, "maximum": 0.2244159528210191, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.030027895721248, "standard_deviation": null, "minimum": 1.030027895721248, "maximum": 1.030027895721248, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.841513745776106, "standard_deviation": null, "minimum": 5.841513745776106, "maximum": 5.841513745776106, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.3340404822926332, "standard_deviation": null, "minimum": 0.3340404822926332, "maximum": 0.3340404822926332, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.2792004272604944, "standard_deviation": null, "minimum": 1.2792004272604944, "maximum": 1.2792004272604944, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9877044970056253, "standard_deviation": null, "minimum": 0.9877044970056253, "maximum": 0.9877044970056253, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4342985756498565, "standard_deviation": null, "minimum": 1.4342985756498565, "maximum": 1.4342985756498565, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.319283116457344, "standard_deviation": null, "minimum": 1.319283116457344, "maximum": 1.319283116457344, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1993595598460629, "standard_deviation": null, "minimum": 0.1993595598460629, "maximum": 0.1993595598460629, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9930754117090344, "standard_deviation": null, "minimum": 0.9930754117090344, "maximum": 0.9930754117090344, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0830428309403974, "standard_deviation": null, "minimum": 1.0830428309403974, "maximum": 1.0830428309403974, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4783520196296103, "standard_deviation": null, "minimum": 1.4783520196296103, "maximum": 1.4783520196296103, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18122179456465135, "standard_deviation": null, "minimum": 0.18122179456465135, "maximum": 0.18122179456465135, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4507"}, {"case": "librispeech_4507-16021-0032", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.6104753017425537, "predicted_mos": 2.6104753017425537, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "196642202cc9252e58ee5a3dd297c65792c678f3ca4b9847b7eaaa4677f8c1aa", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.11933819996193}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4631420187617519, "全帯域_10ms_rms_envelope_nrmse": 0.2521498870954843, "全帯域_10ms_envelope_correlation": 0.9844065583490571, "全帯域_10ms_modulation_mae_db": 1.5167575686874872, "全帯域_40ms_log_rms_mae_db": 1.143652655819051, "全帯域_40ms_rms_envelope_nrmse": 0.2339279831236701, "全帯域_40ms_envelope_correlation": 0.990563842194975, "全帯域_40ms_modulation_mae_db": 1.3662830986900614, "300–1000 Hz_10ms_log_rms_mae_db": 2.0099884269069634, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2851191857692058, "300–1000 Hz_10ms_envelope_correlation": 0.9877688463416305, "300–1000 Hz_10ms_modulation_mae_db": 1.928294461202829, "300–1000 Hz_40ms_log_rms_mae_db": 1.5708309671830716, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.26914141810673536, "300–1000 Hz_40ms_envelope_correlation": 0.9927256355953081, "300–1000 Hz_40ms_modulation_mae_db": 1.664931391187638, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4671917092081013, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1771074862586564, "1000–3000 Hz_10ms_envelope_correlation": 0.9904994623547526, "1000–3000 Hz_10ms_modulation_mae_db": 1.3478395129253073, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0616125400846126, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14830058712595728, "1000–3000 Hz_40ms_envelope_correlation": 0.9964869719121882, "1000–3000 Hz_40ms_modulation_mae_db": 0.9386490935759104, "3000–6000 Hz_10ms_log_rms_mae_db": 1.369585514278038, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20551211988323104, "3000–6000 Hz_10ms_envelope_correlation": 0.9680330648514485, "3000–6000 Hz_10ms_modulation_mae_db": 0.7119387486838885, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8957212689992934, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.126379390037091, "3000–6000 Hz_40ms_envelope_correlation": 0.987817681478567, "3000–6000 Hz_40ms_modulation_mae_db": 0.6558161415798699, "mr_stft_db": 5.811421404661196, "band_mae_db": 1.0931208197120883, "rms_gain_error_db": 1.6203029393156783, "output_peak": 0.522971412403822}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.893071252667349, "spectral_convergence": 0.47089230779509733, "spectral_convergence_status": "ok", "amplitude_floor": 1.3235749190080935e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5112813805665235, "spectral_convergence": 0.32542155335360806, "spectral_convergence_status": "ok", "amplitude_floor": 1.1987597823804542e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.029911580749714, "spectral_convergence": 0.35330501835989386, "spectral_convergence_status": "ok", "amplitude_floor": 9.377214849682438e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.350500895926799e-05, "output_power": 5.7015328257780274e-05, "delta_db": 1.1745236662673488, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012608807022653607, "output_power": 0.0019151827020433226, "delta_db": 1.8153621261202804, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00036191686574782775, "output_power": 0.000480290692152518, "delta_db": 1.2289534805286573, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.9809147109830135e-05, "output_power": 3.7223064428098225e-05, "delta_db": 0.9646257370859311, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.6488056661262093e-05, "output_power": 2.1369352053304e-05, "delta_db": 1.1262188275898195, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.109371525887873e-05, "output_power": 6.469945829840575e-05, "delta_db": 0.24904108068049238, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.0931208197120883, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.6203029393156783, "pitch": null, "output_rms_dbfs": -25.890911325656745, "output_peak": 0.522971412403822, "reference_rms_dbfs": -27.511214264972427, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.811421404661196}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018136510927426692, "log_rms_mae_db": 1.4631420187617519, "rms_envelope_nrmse": 0.2521498870954843, "envelope_correlation": 0.9844065583490571, "rms_gain_error_db": 1.6207779247132472, "reference_mean_power": 0.0017749427149271913, "output_mean_power": 0.002577876654568124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.578419076835135e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003061888162786739, "output_power": 0.000433265413764526, "reference_active": true, "delta_db": 1.5076469823720815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.561280729204581e-05, "output_power": 8.114931805696654e-05, "reference_active": true, "delta_db": 1.641100560744839}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1447257663776473e-05, "output_power": 1.5807164564963673e-05, "reference_active": true, "delta_db": 1.4015251629455407}], "active_band_count": 3, "active_band_mae_db": 1.5167575686874872}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015531555176710683, "log_rms_mae_db": 1.143652655819051, "rms_envelope_nrmse": 0.2339279831236701, "envelope_correlation": 0.990563842194975, "rms_gain_error_db": 1.618916367408797, "reference_mean_power": 0.0017815693662331493, "output_mean_power": 0.0025863921484734036}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.067921014076438e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002861407755621019, "output_power": 0.00041192098160961247, "reference_active": true, "delta_db": 1.5823416380598436}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5605898871779113e-05, "output_power": 3.6293958237113125e-05, "reference_active": true, "delta_db": 1.5149430920485238}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.159270454376389e-07, "output_power": 1.1534993114705774e-06, "reference_active": true, "delta_db": 1.001564565961816}], "active_band_count": 3, "active_band_mae_db": 1.3662830986900614}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017559797263300874, "log_rms_mae_db": 2.0099884269069634, "rms_envelope_nrmse": 0.2851191857692058, "envelope_correlation": 0.9877688463416305, "rms_gain_error_db": 1.9202896745766493, "reference_mean_power": 0.0012870696988326355, "output_mean_power": 0.0020027697972268953}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.703155295091611e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025376682215949657, "output_power": 0.0003937983729952118, "reference_active": true, "delta_db": 1.9083907630568784}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.494529209431561e-05, "output_power": 7.415267429822891e-05, "reference_active": true, "delta_db": 2.174426111691506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.800066525463509e-06, "output_power": 1.302244722758088e-05, "reference_active": true, "delta_db": 1.7020665088601026}], "active_band_count": 3, "active_band_mae_db": 1.928294461202829}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014992083957937532, "log_rms_mae_db": 1.5708309671830716, "rms_envelope_nrmse": 0.26914141810673536, "envelope_correlation": 0.9927256355953081, "rms_gain_error_db": 1.9172732820479275, "reference_mean_power": 0.0012919551058812144, "output_mean_power": 0.002008976010669949}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4067631484646013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002445691472192921, "output_power": 0.0003830159200227949, "reference_active": true, "delta_db": 1.9481515652838544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8830927140807284e-05, "output_power": 3.0479926276729703e-05, "reference_active": true, "delta_db": 2.0914220918537962}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.050384653726169e-07, "output_power": 8.784860378233952e-07, "reference_active": true, "delta_db": 0.9552205164252635}], "active_band_count": 3, "active_band_mae_db": 1.664931391187638}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.2493337902299445e-05, "log_rms_mae_db": 1.4671917092081013, "rms_envelope_nrmse": 0.1771074862586564, "envelope_correlation": 0.9904994623547526, "rms_gain_error_db": 1.0592291123524276, "reference_mean_power": 4.199153922743893e-05, "output_mean_power": 5.3590117037177964e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0299126624198213e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.222152478948813e-06, "output_power": 5.720697799553099e-06, "reference_active": true, "delta_db": 1.3191509303790365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.960031981806193e-07, "output_power": 1.2169596973586416e-06, "reference_active": true, "delta_db": 1.3296663589913984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.633269352164066e-07, "output_power": 3.6304913762334537e-07, "reference_active": true, "delta_db": 1.3947012494054876}], "active_band_count": 3, "active_band_mae_db": 1.3478395129253073}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.820326553727612e-05, "log_rms_mae_db": 1.0616125400846126, "rms_envelope_nrmse": 0.14830058712595728, "envelope_correlation": 0.9964869719121882, "rms_gain_error_db": 1.0592674468930277, "reference_mean_power": 4.2190507757870256e-05, "output_mean_power": 5.3844518390014056e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.548028980657037e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.958931660389438e-06, "output_power": 5.4109858794184585e-06, "reference_active": true, "delta_db": 1.3569839562616188}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.7410748918107156e-07, "output_power": 4.983142393058147e-07, "reference_active": true, "delta_db": 1.2450689495528362}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8002833142836267e-08, "output_power": 1.8911687973981415e-08, "reference_active": true, "delta_db": 0.2138943749132763}], "active_band_count": 3, "active_band_mae_db": 0.9386490935759104}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.219318962627022e-05, "log_rms_mae_db": 1.369585514278038, "rms_envelope_nrmse": 0.20551211988323104, "envelope_correlation": 0.9680330648514485, "rms_gain_error_db": 0.3501425150872537, "reference_mean_power": 2.2095607497389627e-05, "output_mean_power": 2.3950809588452165e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.718240736767725e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.504526854650138e-06, "output_power": 9.598489976130895e-06, "reference_active": true, "delta_db": 0.04272413580583955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5665611692910136e-06, "output_power": 2.1229937113825285e-06, "reference_active": true, "delta_db": -0.8240291161052964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.91578688150447e-07, "output_power": 4.416784606982683e-07, "reference_active": true, "delta_db": -1.2690629941405296}], "active_band_count": 3, "active_band_mae_db": 0.7119387486838885}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.1857789632782756e-05, "log_rms_mae_db": 0.8957212689992934, "rms_envelope_nrmse": 0.126379390037091, "envelope_correlation": 0.987817681478567, "rms_gain_error_db": 0.3499818642946772, "reference_mean_power": 2.220381322589703e-05, "output_mean_power": 2.406721024696526e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6122764663239433e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.855212413255228e-06, "output_power": 9.60075098511309e-06, "reference_active": true, "delta_db": -0.11360783702769446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3142139667203495e-06, "output_power": 1.0433779131484771e-06, "reference_active": true, "delta_db": -1.002244392376267}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2621117296192714e-08, "output_power": 3.5032022478893674e-08, "reference_active": true, "delta_db": -0.8515961953356481}], "active_band_count": 3, "active_band_mae_db": 0.6558161415798699}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9141634363561588, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8583444791bceb5640764c80fd720acd008163e4767350b7c5f507f7853ca648", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9904994623547526, "standard_deviation": null, "minimum": 0.9904994623547526, "maximum": 0.9904994623547526, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4671917092081013, "standard_deviation": null, "minimum": 1.4671917092081013, "maximum": 1.4671917092081013, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.3478395129253073, "standard_deviation": null, "minimum": 1.3478395129253073, "maximum": 1.3478395129253073, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1771074862586564, "standard_deviation": null, "minimum": 0.1771074862586564, "maximum": 0.1771074862586564, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9964869719121882, "standard_deviation": null, "minimum": 0.9964869719121882, "maximum": 0.9964869719121882, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0616125400846126, "standard_deviation": null, "minimum": 1.0616125400846126, "maximum": 1.0616125400846126, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9386490935759104, "standard_deviation": null, "minimum": 0.9386490935759104, "maximum": 0.9386490935759104, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14830058712595728, "standard_deviation": null, "minimum": 0.14830058712595728, "maximum": 0.14830058712595728, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9680330648514485, "standard_deviation": null, "minimum": 0.9680330648514485, "maximum": 0.9680330648514485, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.369585514278038, "standard_deviation": null, "minimum": 1.369585514278038, "maximum": 1.369585514278038, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7119387486838885, "standard_deviation": null, "minimum": 0.7119387486838885, "maximum": 0.7119387486838885, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20551211988323104, "standard_deviation": null, "minimum": 0.20551211988323104, "maximum": 0.20551211988323104, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.987817681478567, "standard_deviation": null, "minimum": 0.987817681478567, "maximum": 0.987817681478567, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8957212689992934, "standard_deviation": null, "minimum": 0.8957212689992934, "maximum": 0.8957212689992934, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6558161415798699, "standard_deviation": null, "minimum": 0.6558161415798699, "maximum": 0.6558161415798699, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.126379390037091, "standard_deviation": null, "minimum": 0.126379390037091, "maximum": 0.126379390037091, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9877688463416305, "standard_deviation": null, "minimum": 0.9877688463416305, "maximum": 0.9877688463416305, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0099884269069634, "standard_deviation": null, "minimum": 2.0099884269069634, "maximum": 2.0099884269069634, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.928294461202829, "standard_deviation": null, "minimum": 1.928294461202829, "maximum": 1.928294461202829, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2851191857692058, "standard_deviation": null, "minimum": 0.2851191857692058, "maximum": 0.2851191857692058, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9927256355953081, "standard_deviation": null, "minimum": 0.9927256355953081, "maximum": 0.9927256355953081, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5708309671830716, "standard_deviation": null, "minimum": 1.5708309671830716, "maximum": 1.5708309671830716, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.664931391187638, "standard_deviation": null, "minimum": 1.664931391187638, "maximum": 1.664931391187638, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.26914141810673536, "standard_deviation": null, "minimum": 0.26914141810673536, "maximum": 0.26914141810673536, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.0931208197120883, "standard_deviation": null, "minimum": 1.0931208197120883, "maximum": 1.0931208197120883, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.811421404661196, "standard_deviation": null, "minimum": 5.811421404661196, "maximum": 5.811421404661196, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.522971412403822, "standard_deviation": null, "minimum": 0.522971412403822, "maximum": 0.522971412403822, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.6203029393156783, "standard_deviation": null, "minimum": 1.6203029393156783, "maximum": 1.6203029393156783, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9844065583490571, "standard_deviation": null, "minimum": 0.9844065583490571, "maximum": 0.9844065583490571, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4631420187617519, "standard_deviation": null, "minimum": 1.4631420187617519, "maximum": 1.4631420187617519, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.5167575686874872, "standard_deviation": null, "minimum": 1.5167575686874872, "maximum": 1.5167575686874872, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2521498870954843, "standard_deviation": null, "minimum": 0.2521498870954843, "maximum": 0.2521498870954843, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.990563842194975, "standard_deviation": null, "minimum": 0.990563842194975, "maximum": 0.990563842194975, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.143652655819051, "standard_deviation": null, "minimum": 1.143652655819051, "maximum": 1.143652655819051, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.3662830986900614, "standard_deviation": null, "minimum": 1.3662830986900614, "maximum": 1.3662830986900614, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2339279831236701, "standard_deviation": null, "minimum": 0.2339279831236701, "maximum": 0.2339279831236701, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4507"}, {"case": "librispeech_4507-16021-0045", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.89443302154541, "predicted_mos": 2.89443302154541, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 78240, "input_sample_rate": 16000, "input_seconds": 4.89, "input_float32_sha256": "dc31b08abaf94d152d21b41fabaf6d5606d8b79b3ba91b103fbd18594e36be11", "internal_samples": 78240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.707784700207412}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5352940289770767, "全帯域_10ms_rms_envelope_nrmse": 0.26324395779965315, "全帯域_10ms_envelope_correlation": 0.9877155891464352, "全帯域_10ms_modulation_mae_db": 2.1470562017134305, "全帯域_40ms_log_rms_mae_db": 1.1632453390605086, "全帯域_40ms_rms_envelope_nrmse": 0.24088981339440688, "全帯域_40ms_envelope_correlation": 0.9945981851901204, "全帯域_40ms_modulation_mae_db": 1.8347313785412005, "300–1000 Hz_10ms_log_rms_mae_db": 2.1588329146540772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2970106994705909, "300–1000 Hz_10ms_envelope_correlation": 0.9878070084217853, "300–1000 Hz_10ms_modulation_mae_db": 2.3262093765685554, "300–1000 Hz_40ms_log_rms_mae_db": 1.7089133923023865, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2750062472986385, "300–1000 Hz_40ms_envelope_correlation": 0.9944312924359399, "300–1000 Hz_40ms_modulation_mae_db": 1.9767089048660917, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3516006826294524, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16038800284287372, "1000–3000 Hz_10ms_envelope_correlation": 0.994704252134501, "1000–3000 Hz_10ms_modulation_mae_db": 1.0047931665738388, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9390668217962849, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13607881266082272, "1000–3000 Hz_40ms_envelope_correlation": 0.9985572593001849, "1000–3000 Hz_40ms_modulation_mae_db": 0.9574474882727674, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3974446665155067, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2718335439916743, "3000–6000 Hz_10ms_envelope_correlation": 0.9589886289535468, "3000–6000 Hz_10ms_modulation_mae_db": 0.9416757215441738, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9249822006843911, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16246043740962923, "3000–6000 Hz_40ms_envelope_correlation": 0.9880048658677707, "3000–6000 Hz_40ms_modulation_mae_db": 0.8923089081452261, "mr_stft_db": 5.899555993492766, "band_mae_db": 1.284104407489327, "rms_gain_error_db": 1.743609688656187, "output_peak": 0.5448144580463609}, "full_scale": {"sample_count": 78240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 78240, "output_samples": 78240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.898329977304764, "spectral_convergence": 0.5103757946774409, "spectral_convergence_status": "ok", "amplitude_floor": 1.658911626731477e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.7165135478181925, "spectral_convergence": 0.3638182833794153, "spectral_convergence_status": "ok", "amplitude_floor": 1.4097603217110054e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.083824455355342, "spectral_convergence": 0.40109661255610807, "spectral_convergence_status": "ok", "amplitude_floor": 8.986287240444543e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.8753665955686144e-05, "output_power": 7.16480186659506e-05, "delta_db": 1.6719690768567985, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00084480798498807, "output_power": 0.001122550350948138, "delta_db": 1.2344781989569051, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015560766761226633, "output_power": 0.002513657919635327, "delta_db": 2.0827518164689747, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.758941174067266e-05, "output_power": 7.169083982613474e-05, "delta_db": 0.9512102562254949, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.7471770650556e-05, "output_power": 9.880198090350986e-05, "delta_db": 1.0562216987743485, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 5.273577609828158e-05, "output_power": 6.207330674305207e-05, "delta_db": 0.7079953976534412, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.284104407489327, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.743609688656187, "pitch": null, "output_rms_dbfs": -24.04457218931291, "output_peak": 0.5448144580463609, "reference_rms_dbfs": -25.788181877969095, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.899555993492766}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 78240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.0002230061513489049, "log_rms_mae_db": 1.5352940289770767, "rms_envelope_nrmse": 0.26324395779965315, "envelope_correlation": 0.9877155891464352, "rms_gain_error_db": 1.7436100444169156, "reference_mean_power": 0.002640134243154231, "output_mean_power": 0.003944455078786766}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.2388906472962606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011616255671056531, "output_power": 0.0017677230818832568, "reference_active": true, "delta_db": 1.8234807053859998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014996209998372379, "output_power": 0.00027195376604617025, "reference_active": true, "delta_db": 2.5851356394705522}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.935590189308993e-05, "output_power": 6.28441845501098e-05, "reference_active": true, "delta_db": 2.0325522602837385}], "active_band_count": 3, "active_band_mae_db": 2.1470562017134305}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.0001943907377270206, "log_rms_mae_db": 1.1632453390605086, "rms_envelope_nrmse": 0.24088981339440688, "envelope_correlation": 0.9945981851901204, "rms_gain_error_db": 1.7436114634263178, "reference_mean_power": 0.0026564436783668454, "output_mean_power": 0.003968823255658013}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 2.080120853269114e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001087632600289277, "output_power": 0.0016648826388240793, "reference_active": true, "delta_db": 1.8490140807502136}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.148356435560039e-05, "output_power": 0.00013353950557905576, "reference_active": true, "delta_db": 2.714035645338014}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.554313443594102e-06, "output_power": 3.172405031977318e-06, "reference_active": true, "delta_db": 0.9411444095353739}], "active_band_count": 3, "active_band_mae_db": 1.8347313785412005}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 0.00021647011759154595, "log_rms_mae_db": 2.1588329146540772, "rms_envelope_nrmse": 0.2970106994705909, "envelope_correlation": 0.9878070084217853, "rms_gain_error_db": 1.9841726634769399, "reference_mean_power": 0.0021208071215793162, "output_mean_power": 0.003349025384940752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 2.0688695374548394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001066233690351074, "output_power": 0.0016839168378855812, "reference_active": true, "delta_db": 1.9846823853632842}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015588791655839424, "output_power": 0.0002949419636523461, "reference_active": true, "delta_db": 2.7692411460798967}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.732583628298963e-05, "output_power": 6.229884666614733e-05, "reference_active": true, "delta_db": 2.224704598262485}], "active_band_count": 3, "active_band_mae_db": 2.3262093765685554}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 0.00018533448652616975, "log_rms_mae_db": 1.7089133923023865, "rms_envelope_nrmse": 0.2750062472986385, "envelope_correlation": 0.9944312924359399, "rms_gain_error_db": 1.984173189176571, "reference_mean_power": 0.0021339103768736657, "output_mean_power": 0.003369717506927302}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.9317825081344165e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001010025588563974, "output_power": 0.0016024376083317827, "reference_active": true, "delta_db": 2.0044875247245293}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.797692317548564e-05, "output_power": 0.00014769018168646095, "reference_active": true, "delta_db": 2.7738553001926682}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604806628732506e-06, "output_power": 3.338104200751934e-06, "reference_active": true, "delta_db": 1.1517838896810775}], "active_band_count": 3, "active_band_mae_db": 1.9767089048660917}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 6.255230128136302e-05, "log_rms_mae_db": 1.3516006826294524, "rms_envelope_nrmse": 0.16038800284287372, "envelope_correlation": 0.994704252134501, "rms_gain_error_db": 1.0316760418186501, "reference_mean_power": 0.0001286629346191704, "output_mean_power": 0.00016316276523386782}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.8253387298901982e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.359313427297316e-05, "output_power": 9.3241700560408e-05, "reference_active": true, "delta_db": 1.0277288609263568}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.331803440637771e-06, "output_power": 9.428633647863804e-06, "reference_active": true, "delta_db": 1.0923794805067581}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8895709731947531e-06, "output_power": 2.3216156607841196e-06, "reference_active": true, "delta_db": 0.8942711582884015}], "active_band_count": 3, "active_band_mae_db": 1.0047931665738388}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 6.0793288183779695e-05, "log_rms_mae_db": 0.9390668217962849, "rms_envelope_nrmse": 0.13607881266082272, "envelope_correlation": 0.9985572593001849, "rms_gain_error_db": 1.031680910899803, "reference_mean_power": 0.00012945711895499159, "output_mean_power": 0.00016417008715032778}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.7711142240683522e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.969624557508824e-05, "output_power": 8.776669961335641e-05, "reference_active": true, "delta_db": 1.001203834347348}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.240093489969041e-06, "output_power": 4.147773125765728e-06, "reference_active": true, "delta_db": 1.0725745183406927}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.585555587494176e-07, "output_power": 1.9056269391168672e-07, "reference_active": true, "delta_db": 0.7985641121302618}], "active_band_count": 3, "active_band_mae_db": 0.9574474882727674}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 977, "rms_floor": 3.765201856958641e-05, "log_rms_mae_db": 1.3974446665155067, "rms_envelope_nrmse": 0.2718335439916743, "envelope_correlation": 0.9589886289535468, "rms_gain_error_db": 0.7644197936988755, "reference_mean_power": 2.5945814971953512e-05, "output_mean_power": 3.093921539877872e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 977, "frequency_resolution_hz": 0.2047082906857728, "power_floor": 1.30554638893879e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0948707154331005e-06, "output_power": 5.533819603376422e-06, "reference_active": true, "delta_db": 0.3589182942791746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.046923144145412e-06, "output_power": 1.4118594327567735e-06, "reference_active": true, "delta_db": 1.2987665895262825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.3665503073484234e-07, "output_power": 5.713111448548117e-07, "reference_active": true, "delta_db": 1.167342280827064}], "active_band_count": 3, "active_band_mae_db": 0.9416757215441738}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 78240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 971, "rms_floor": 2.0830708932641756e-05, "log_rms_mae_db": 0.9249822006843911, "rms_envelope_nrmse": 0.16246043740962923, "envelope_correlation": 0.9880048658677707, "rms_gain_error_db": 0.7643906205164747, "reference_mean_power": 2.6105327616819126e-05, "output_mean_power": 3.112921793385853e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 971, "frequency_resolution_hz": 0.20597322348094746, "power_floor": 1.2170827427715922e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.759181059225212e-06, "output_power": 5.186685603550717e-06, "reference_active": true, "delta_db": 0.37357696072554486}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.967295837314181e-07, "output_power": 8.0361872771694e-07, "reference_active": true, "delta_db": 1.2927247993898408}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240630896276443e-08, "output_power": 4.08970577439644e-08, "reference_active": true, "delta_db": 1.010624964320293}], "active_band_count": 3, "active_band_mae_db": 0.8923089081452261}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.902948662522257, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1be2396753c93463ca97ae2dff95cdc4fc82e9349b39487a826b81b870072f76", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.994704252134501, "standard_deviation": null, "minimum": 0.994704252134501, "maximum": 0.994704252134501, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3516006826294524, "standard_deviation": null, "minimum": 1.3516006826294524, "maximum": 1.3516006826294524, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0047931665738388, "standard_deviation": null, "minimum": 1.0047931665738388, "maximum": 1.0047931665738388, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16038800284287372, "standard_deviation": null, "minimum": 0.16038800284287372, "maximum": 0.16038800284287372, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9985572593001849, "standard_deviation": null, "minimum": 0.9985572593001849, "maximum": 0.9985572593001849, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9390668217962849, "standard_deviation": null, "minimum": 0.9390668217962849, "maximum": 0.9390668217962849, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9574474882727674, "standard_deviation": null, "minimum": 0.9574474882727674, "maximum": 0.9574474882727674, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13607881266082272, "standard_deviation": null, "minimum": 0.13607881266082272, "maximum": 0.13607881266082272, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9589886289535468, "standard_deviation": null, "minimum": 0.9589886289535468, "maximum": 0.9589886289535468, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3974446665155067, "standard_deviation": null, "minimum": 1.3974446665155067, "maximum": 1.3974446665155067, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9416757215441738, "standard_deviation": null, "minimum": 0.9416757215441738, "maximum": 0.9416757215441738, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2718335439916743, "standard_deviation": null, "minimum": 0.2718335439916743, "maximum": 0.2718335439916743, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9880048658677707, "standard_deviation": null, "minimum": 0.9880048658677707, "maximum": 0.9880048658677707, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9249822006843911, "standard_deviation": null, "minimum": 0.9249822006843911, "maximum": 0.9249822006843911, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8923089081452261, "standard_deviation": null, "minimum": 0.8923089081452261, "maximum": 0.8923089081452261, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16246043740962923, "standard_deviation": null, "minimum": 0.16246043740962923, "maximum": 0.16246043740962923, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9878070084217853, "standard_deviation": null, "minimum": 0.9878070084217853, "maximum": 0.9878070084217853, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1588329146540772, "standard_deviation": null, "minimum": 2.1588329146540772, "maximum": 2.1588329146540772, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.3262093765685554, "standard_deviation": null, "minimum": 2.3262093765685554, "maximum": 2.3262093765685554, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2970106994705909, "standard_deviation": null, "minimum": 0.2970106994705909, "maximum": 0.2970106994705909, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9944312924359399, "standard_deviation": null, "minimum": 0.9944312924359399, "maximum": 0.9944312924359399, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7089133923023865, "standard_deviation": null, "minimum": 1.7089133923023865, "maximum": 1.7089133923023865, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.9767089048660917, "standard_deviation": null, "minimum": 1.9767089048660917, "maximum": 1.9767089048660917, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2750062472986385, "standard_deviation": null, "minimum": 0.2750062472986385, "maximum": 0.2750062472986385, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.284104407489327, "standard_deviation": null, "minimum": 1.284104407489327, "maximum": 1.284104407489327, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.899555993492766, "standard_deviation": null, "minimum": 5.899555993492766, "maximum": 5.899555993492766, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5448144580463609, "standard_deviation": null, "minimum": 0.5448144580463609, "maximum": 0.5448144580463609, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.743609688656187, "standard_deviation": null, "minimum": 1.743609688656187, "maximum": 1.743609688656187, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9877155891464352, "standard_deviation": null, "minimum": 0.9877155891464352, "maximum": 0.9877155891464352, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5352940289770767, "standard_deviation": null, "minimum": 1.5352940289770767, "maximum": 1.5352940289770767, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 2.1470562017134305, "standard_deviation": null, "minimum": 2.1470562017134305, "maximum": 2.1470562017134305, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.26324395779965315, "standard_deviation": null, "minimum": 0.26324395779965315, "maximum": 0.26324395779965315, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9945981851901204, "standard_deviation": null, "minimum": 0.9945981851901204, "maximum": 0.9945981851901204, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1632453390605086, "standard_deviation": null, "minimum": 1.1632453390605086, "maximum": 1.1632453390605086, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.8347313785412005, "standard_deviation": null, "minimum": 1.8347313785412005, "maximum": 1.8347313785412005, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.24088981339440688, "standard_deviation": null, "minimum": 0.24088981339440688, "maximum": 0.24088981339440688, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4507"}, {"case": "librispeech_4970-29093-0004", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.651092052459717, "predicted_mos": 2.651092052459717, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 60000, "input_sample_rate": 16000, "input_seconds": 3.75, "input_float32_sha256": "a18d3cad1e23d01985d128741fa7495260286bd48dee2f4f84b6e60198edc5ea", "internal_samples": 60000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.145670800004154}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.2666993589010445, "全帯域_10ms_rms_envelope_nrmse": 0.13521062347880045, "全帯域_10ms_envelope_correlation": 0.9954036889906682, "全帯域_10ms_modulation_mae_db": 0.8130292489770098, "全帯域_40ms_log_rms_mae_db": 0.9281486361244177, "全帯域_40ms_rms_envelope_nrmse": 0.1204125781964991, "全帯域_40ms_envelope_correlation": 0.9978647758207928, "全帯域_40ms_modulation_mae_db": 1.1134996172722043, "300–1000 Hz_10ms_log_rms_mae_db": 1.2914135428151277, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1991193451379194, "300–1000 Hz_10ms_envelope_correlation": 0.9919346631882615, "300–1000 Hz_10ms_modulation_mae_db": 1.4043711609252185, "300–1000 Hz_40ms_log_rms_mae_db": 1.0721972660101289, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17683068458320989, "300–1000 Hz_40ms_envelope_correlation": 0.9968138139763737, "300–1000 Hz_40ms_modulation_mae_db": 1.5017933738761098, "1000–3000 Hz_10ms_log_rms_mae_db": 1.1373444609580456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.07446878199464697, "1000–3000 Hz_10ms_envelope_correlation": 0.9981125841639185, "1000–3000 Hz_10ms_modulation_mae_db": 0.19275099000351037, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7448644285361614, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.06210695533677542, "1000–3000 Hz_40ms_envelope_correlation": 0.9990577061483505, "1000–3000 Hz_40ms_modulation_mae_db": 0.27323489520666, "3000–6000 Hz_10ms_log_rms_mae_db": 1.386873984903204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18628319355011044, "3000–6000 Hz_10ms_envelope_correlation": 0.9744655031076884, "3000–6000 Hz_10ms_modulation_mae_db": 0.5506202409292166, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8571005755751636, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11370990160004561, "3000–6000 Hz_40ms_envelope_correlation": 0.9902006454685918, "3000–6000 Hz_40ms_modulation_mae_db": 0.6133064936556126, "mr_stft_db": 5.200904485770323, "band_mae_db": 0.666515657119418, "rms_gain_error_db": 0.8855284215470155, "output_peak": 0.8382972587289713}, "full_scale": {"sample_count": 60000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 60000, "output_samples": 60000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.683842363788228, "spectral_convergence": 0.3646753298345562, "spectral_convergence_status": "ok", "amplitude_floor": 1.4952467880734047e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.885957824650904, "spectral_convergence": 0.2779559379131615, "spectral_convergence_status": "ok", "amplitude_floor": 1.087083783284318e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.032913268871838, "spectral_convergence": 0.3352930559807361, "spectral_convergence_status": "ok", "amplitude_floor": 6.922347653241366e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.706318352359072e-05, "output_power": 6.555131565626936e-05, "delta_db": 0.6022541563090242, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006570402135178566, "output_power": 0.0008122442511319025, "delta_db": 0.9209469524510744, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010679696452407233, "output_power": 0.0014544207930701904, "delta_db": 1.3413116582278737, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007107211898605409, "output_power": 0.000812402876576628, "delta_db": 0.5807218854919849, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005187917577577873, "output_power": 0.0005696555934397157, "delta_db": 0.4061929840256452, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.297230867177819e-05, "output_power": 8.019855507004577e-05, "delta_db": -0.14766630621090515, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.666515657119418, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8855284215470155, "pitch": null, "output_rms_dbfs": -24.208484890391997, "output_peak": 0.8382972587289713, "reference_rms_dbfs": -25.09401331193901, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.200904485770323}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 60000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.0002227499093660225, "log_rms_mae_db": 1.2666993589010445, "rms_envelope_nrmse": 0.13521062347880045, "envelope_correlation": 0.9954036889906682, "rms_gain_error_db": 0.8855284336680762, "reference_mean_power": 0.003098689875324867, "output_mean_power": 0.0037995394361640213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 3.065525005901086e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018487969313489734, "output_power": 0.002277069589208374, "reference_active": true, "delta_db": 0.9048709160600295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000254937700664643, "output_power": 0.00030807459680425085, "reference_active": true, "delta_db": 0.8222182426629977}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5927620235575996e-05, "output_power": 4.232805475213791e-05, "reference_active": true, "delta_db": 0.7119985882080021}], "active_band_count": 3, "active_band_mae_db": 0.8130292489770098}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00020550660315895724, "log_rms_mae_db": 0.9281486361244177, "rms_envelope_nrmse": 0.1204125781964991, "envelope_correlation": 0.9978647758207928, "rms_gain_error_db": 0.8855284574716764, "reference_mean_power": 0.003123712899469445, "output_mean_power": 0.003830222091068751}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 2.850551559013711e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017535539962877898, "output_power": 0.00216871781386075, "reference_active": true, "delta_db": 0.9228390334828711}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011841366649455362, "output_power": 0.00014733729003381875, "reference_active": true, "delta_db": 0.9491084906439937}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7582705881956915e-06, "output_power": 2.465703457707588e-06, "reference_active": true, "delta_db": 1.4685513276897484}], "active_band_count": 3, "active_band_mae_db": 1.1134996172722043}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00016827060195170467, "log_rms_mae_db": 1.2914135428151277, "rms_envelope_nrmse": 0.1991193451379194, "envelope_correlation": 0.9919346631882615, "rms_gain_error_db": 1.305036853295432, "reference_mean_power": 0.001201501254742638, "output_mean_power": 0.0016226614327162876}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.2448894715706734e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007295274990436897, "output_power": 0.0010153766692098491, "reference_active": true, "delta_db": 1.4358551321983317}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014476368795830424, "output_power": 0.00020752121043116596, "reference_active": true, "delta_db": 1.564028533641959}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.785748904828138e-05, "output_power": 3.683536267019907e-05, "reference_active": true, "delta_db": 1.2132298169353648}], "active_band_count": 3, "active_band_mae_db": 1.4043711609252185}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.00012583494842815252, "log_rms_mae_db": 1.0721972660101289, "rms_envelope_nrmse": 0.17683068458320989, "envelope_correlation": 0.9968138139763737, "rms_gain_error_db": 1.3050368545034625, "reference_mean_power": 0.0012112038213410922, "output_mean_power": 0.001635765023829667}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.1451125517965788e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007097609595329509, "output_power": 0.0009909176726340173, "reference_active": true, "delta_db": 1.4492546669400999}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.709378020571589e-05, "output_power": 9.824883188868465e-05, "reference_active": true, "delta_db": 1.656451339778563}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7774397828761626e-06, "output_power": 2.4533658054445102e-06, "reference_active": true, "delta_db": 1.3996741149096668}], "active_band_count": 3, "active_band_mae_db": 1.5017933738761098}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 0.00015339527329010622, "log_rms_mae_db": 1.1373444609580456, "rms_envelope_nrmse": 0.07446878199464697, "envelope_correlation": 0.9981125841639185, "rms_gain_error_db": 0.4019710654833206, "reference_mean_power": 0.0009604249251386078, "output_mean_power": 0.0010535630453892405}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 1.606398277844274e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000996843421347699, "output_power": 0.0010873244788192147, "reference_active": true, "delta_db": 0.3773221827395379}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011073134134931949, "output_power": 0.00011569174827642448, "reference_active": true, "delta_db": 0.1903182317246467}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0381145318442363e-05, "output_power": 1.035580859705282e-05, "reference_active": true, "delta_db": -0.010612555546346537}], "active_band_count": 3, "active_band_mae_db": 0.19275099000351037}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 0.0001445669043392124, "log_rms_mae_db": 0.7448644285361614, "rms_envelope_nrmse": 0.06210695533677542, "envelope_correlation": 0.9990577061483505, "rms_gain_error_db": 0.40197108442550256, "reference_mean_power": 0.0009681806844812572, "output_mean_power": 0.00106207093150035}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 1.5548349296058167e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009718305477517134, "output_power": 0.0010582676561190323, "reference_active": true, "delta_db": 0.37004976757708363}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.313569881232493e-05, "output_power": 5.5479917171866336e-05, "reference_active": true, "delta_db": 0.18749407465178314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.152693877304456e-07, "output_power": 5.792275674845296e-07, "reference_active": true, "delta_db": -0.26216084339111323}], "active_band_count": 3, "active_band_mae_db": 0.27323489520666}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 749, "rms_floor": 4.2745097516618104e-05, "log_rms_mae_db": 1.386873984903204, "rms_envelope_nrmse": 0.18628319355011044, "envelope_correlation": 0.9744655031076884, "rms_gain_error_db": 0.30944432668244387, "reference_mean_power": 0.0001383637841793036, "output_mean_power": 0.00014858222716923988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 749, "frequency_resolution_hz": 0.26702269692923897, "power_floor": 9.868150581916701e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.732185415785845e-05, "output_power": 5.997685152787481e-05, "reference_active": true, "delta_db": 0.196634339925088}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6922402540887435e-05, "output_power": 1.4944815434317118e-05, "reference_active": true, "delta_db": -0.5397146564510195}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.249240297758961e-06, "output_power": 3.4415978269350936e-06, "reference_active": true, "delta_db": -0.9155117264115424}], "active_band_count": 3, "active_band_mae_db": 0.5506202409292166}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 60000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 743, "rms_floor": 3.734940150914352e-05, "log_rms_mae_db": 0.8571005755751636, "rms_envelope_nrmse": 0.11370990160004561, "envelope_correlation": 0.9902006454685918, "rms_gain_error_db": 0.3094444010942091, "reference_mean_power": 0.00013948111618043138, "output_mean_power": 0.00014978207894613852}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 743, "frequency_resolution_hz": 0.2691790040376851, "power_floor": 8.397728752995274e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.502560023531223e-05, "output_power": 5.70789770369983e-05, "reference_active": true, "delta_db": 0.15911392473307834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.667910609926144e-06, "output_power": 6.504185575795638e-06, "reference_active": true, "delta_db": -0.7148411742369174}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2898812052596812e-07, "output_power": 1.8332281370868135e-07, "reference_active": true, "delta_db": -0.965964381996842}], "active_band_count": 3, "active_band_mae_db": 0.6133064936556126}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9356695989169883, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ffb3ae734dcb60109735ea0576545f2b1d032f8f1e36dec66808e8edb320135e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9981125841639185, "standard_deviation": null, "minimum": 0.9981125841639185, "maximum": 0.9981125841639185, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.1373444609580456, "standard_deviation": null, "minimum": 1.1373444609580456, "maximum": 1.1373444609580456, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.19275099000351037, "standard_deviation": null, "minimum": 0.19275099000351037, "maximum": 0.19275099000351037, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.07446878199464697, "standard_deviation": null, "minimum": 0.07446878199464697, "maximum": 0.07446878199464697, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9990577061483505, "standard_deviation": null, "minimum": 0.9990577061483505, "maximum": 0.9990577061483505, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7448644285361614, "standard_deviation": null, "minimum": 0.7448644285361614, "maximum": 0.7448644285361614, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.27323489520666, "standard_deviation": null, "minimum": 0.27323489520666, "maximum": 0.27323489520666, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.06210695533677542, "standard_deviation": null, "minimum": 0.06210695533677542, "maximum": 0.06210695533677542, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9744655031076884, "standard_deviation": null, "minimum": 0.9744655031076884, "maximum": 0.9744655031076884, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.386873984903204, "standard_deviation": null, "minimum": 1.386873984903204, "maximum": 1.386873984903204, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5506202409292166, "standard_deviation": null, "minimum": 0.5506202409292166, "maximum": 0.5506202409292166, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18628319355011044, "standard_deviation": null, "minimum": 0.18628319355011044, "maximum": 0.18628319355011044, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9902006454685918, "standard_deviation": null, "minimum": 0.9902006454685918, "maximum": 0.9902006454685918, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8571005755751636, "standard_deviation": null, "minimum": 0.8571005755751636, "maximum": 0.8571005755751636, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6133064936556126, "standard_deviation": null, "minimum": 0.6133064936556126, "maximum": 0.6133064936556126, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11370990160004561, "standard_deviation": null, "minimum": 0.11370990160004561, "maximum": 0.11370990160004561, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9919346631882615, "standard_deviation": null, "minimum": 0.9919346631882615, "maximum": 0.9919346631882615, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.2914135428151277, "standard_deviation": null, "minimum": 1.2914135428151277, "maximum": 1.2914135428151277, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4043711609252185, "standard_deviation": null, "minimum": 1.4043711609252185, "maximum": 1.4043711609252185, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1991193451379194, "standard_deviation": null, "minimum": 0.1991193451379194, "maximum": 0.1991193451379194, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9968138139763737, "standard_deviation": null, "minimum": 0.9968138139763737, "maximum": 0.9968138139763737, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0721972660101289, "standard_deviation": null, "minimum": 1.0721972660101289, "maximum": 1.0721972660101289, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.5017933738761098, "standard_deviation": null, "minimum": 1.5017933738761098, "maximum": 1.5017933738761098, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17683068458320989, "standard_deviation": null, "minimum": 0.17683068458320989, "maximum": 0.17683068458320989, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.666515657119418, "standard_deviation": null, "minimum": 0.666515657119418, "maximum": 0.666515657119418, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.200904485770323, "standard_deviation": null, "minimum": 5.200904485770323, "maximum": 5.200904485770323, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.8382972587289713, "standard_deviation": null, "minimum": 0.8382972587289713, "maximum": 0.8382972587289713, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8855284215470155, "standard_deviation": null, "minimum": 0.8855284215470155, "maximum": 0.8855284215470155, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9954036889906682, "standard_deviation": null, "minimum": 0.9954036889906682, "maximum": 0.9954036889906682, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.2666993589010445, "standard_deviation": null, "minimum": 1.2666993589010445, "maximum": 1.2666993589010445, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8130292489770098, "standard_deviation": null, "minimum": 0.8130292489770098, "maximum": 0.8130292489770098, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.13521062347880045, "standard_deviation": null, "minimum": 0.13521062347880045, "maximum": 0.13521062347880045, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9978647758207928, "standard_deviation": null, "minimum": 0.9978647758207928, "maximum": 0.9978647758207928, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9281486361244177, "standard_deviation": null, "minimum": 0.9281486361244177, "maximum": 0.9281486361244177, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1134996172722043, "standard_deviation": null, "minimum": 1.1134996172722043, "maximum": 1.1134996172722043, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1204125781964991, "standard_deviation": null, "minimum": 0.1204125781964991, "maximum": 0.1204125781964991, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4970"}, {"case": "librispeech_4970-29093-0005", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.7696456909179688, "predicted_mos": 2.7696456909179688, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "c58c10b017a7421361aa141ac7dacd74872d9c7cd063e268474fc3b861ebb5b4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 11.602486299816519}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.769466241697947, "全帯域_10ms_rms_envelope_nrmse": 0.18128309139689466, "全帯域_10ms_envelope_correlation": 0.9892615353435668, "全帯域_10ms_modulation_mae_db": 0.7444196326302036, "全帯域_40ms_log_rms_mae_db": 1.2902919733509164, "全帯域_40ms_rms_envelope_nrmse": 0.15331865336162007, "全帯域_40ms_envelope_correlation": 0.9949535705435767, "全帯域_40ms_modulation_mae_db": 0.6139975254507667, "300–1000 Hz_10ms_log_rms_mae_db": 1.5034261808301392, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20918598703975716, "300–1000 Hz_10ms_envelope_correlation": 0.992504729296344, "300–1000 Hz_10ms_modulation_mae_db": 1.214249769719945, "300–1000 Hz_40ms_log_rms_mae_db": 1.3497971535972821, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.19354864115692005, "300–1000 Hz_40ms_envelope_correlation": 0.9958622620018229, "300–1000 Hz_40ms_modulation_mae_db": 1.0546461868883696, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2683317932838505, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1394609003520675, "1000–3000 Hz_10ms_envelope_correlation": 0.9919011204228837, "1000–3000 Hz_10ms_modulation_mae_db": 0.6498566233417341, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8780800663322775, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09311311375825246, "1000–3000 Hz_40ms_envelope_correlation": 0.9979696484948097, "1000–3000 Hz_40ms_modulation_mae_db": 0.5331059912108066, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6288864664429354, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.16154846228937886, "3000–6000 Hz_10ms_envelope_correlation": 0.9840937387462086, "3000–6000 Hz_10ms_modulation_mae_db": 0.13817278628688903, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0485185258305452, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10609255717404191, "3000–6000 Hz_40ms_envelope_correlation": 0.9935656063621127, "3000–6000 Hz_40ms_modulation_mae_db": 0.24087727560475936, "mr_stft_db": 5.463843352116619, "band_mae_db": 0.6948741072092696, "rms_gain_error_db": 1.0337744353822433, "output_peak": 1.0425958184988877}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 5, "at_or_above_full_scale_fraction": 5.208333333333334e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 6.25e-05, "longest_full_scale_run_samples": 1, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.251489771232532, "spectral_convergence": 0.6150746051088963, "spectral_convergence_status": "ok", "amplitude_floor": 1.6417184973675735e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.024738792669105, "spectral_convergence": 0.26382273993858135, "spectral_convergence_status": "ok", "amplitude_floor": 1.2466848673527415e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.115301492448221, "spectral_convergence": 0.31267041815620933, "spectral_convergence_status": "ok", "amplitude_floor": 1.127090317770565e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.270965260159295e-05, "output_power": 4.5780865061066405e-05, "delta_db": 0.3015795580099335, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004287949616170093, "output_power": 0.0005795551222167428, "delta_db": 1.3084507484679522, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015047786428193433, "output_power": 0.002090762997369407, "delta_db": 1.4283218660055939, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0009641149894002944, "output_power": 0.0010873170179391326, "delta_db": 0.5222735033686531, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00041492564533297956, "output_power": 0.0004768527432885436, "delta_db": 0.6041400731025062, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.594979221160373e-05, "output_power": 8.586119757974874e-05, "delta_db": -0.0044788943009788995, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6948741072092696, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0337744353822433, "pitch": null, "output_rms_dbfs": -23.599033430919462, "output_peak": 1.0425958184988877, "reference_rms_dbfs": -24.632807866301704, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.463843352116619}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027067935709258154, "log_rms_mae_db": 1.769466241697947, "rms_envelope_nrmse": 0.18128309139689466, "envelope_correlation": 0.9892615353435668, "rms_gain_error_db": 1.0348597071274503, "reference_mean_power": 0.0034432187972516922, "output_mean_power": 0.0043696896324670765}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.845747823685581e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011374442611865624, "output_power": 0.0013904623204689767, "reference_active": true, "delta_db": 0.8722910072652483}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011129046974770975, "output_power": 0.00013631742986941676, "reference_active": true, "delta_db": 0.8809341376911277}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.356136872122755e-05, "output_power": 2.6315031309151067e-05, "reference_active": true, "delta_db": 0.48003375293423467}], "active_band_count": 3, "active_band_mae_db": 0.7444196326302036}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026426858408099, "log_rms_mae_db": 1.2902919733509164, "rms_envelope_nrmse": 0.15331865336162007, "envelope_correlation": 0.9949535705435767, "rms_gain_error_db": 1.0354683939249487, "reference_mean_power": 0.0034555022621815964, "output_mean_power": 0.004385892886531566}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7242545404192642e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010676968681649734, "output_power": 0.0013095351120985534, "reference_active": true, "delta_db": 0.8866917867490536}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.454467282708388e-05, "output_power": 5.521491972156151e-05, "reference_active": true, "delta_db": 0.9326067160531739}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7651139324996653e-06, "output_power": 1.7743616826447054e-06, "reference_active": true, "delta_db": 0.022694073550072635}], "active_band_count": 3, "active_band_mae_db": 0.6139975254507667}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001806334260766054, "log_rms_mae_db": 1.5034261808301392, "rms_envelope_nrmse": 0.20918598703975716, "envelope_correlation": 0.992504729296344, "rms_gain_error_db": 1.3944138004181879, "reference_mean_power": 0.0015370357837470275, "output_mean_power": 0.0021189726853827187}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.498278279762008e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005426794452872498, "output_power": 0.0007443273100174955, "reference_active": true, "delta_db": 1.3722058152284555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.715116909567221e-05, "output_power": 7.397290302000292e-05, "reference_active": true, "delta_db": 1.120475438134499}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1787274842814488e-05, "output_power": 1.5361025695578007e-05, "reference_active": true, "delta_db": 1.1500680557968799}], "active_band_count": 3, "active_band_mae_db": 1.214249769719945}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017379577463928091, "log_rms_mae_db": 1.3497971535972821, "rms_envelope_nrmse": 0.19354864115692005, "envelope_correlation": 0.9958622620018229, "rms_gain_error_db": 1.3960413157294496, "reference_mean_power": 0.0015404251144981946, "output_mean_power": 0.0021244412337736696}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.036061947317895e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005182860743704318, "output_power": 0.0007090594762196784, "reference_active": true, "delta_db": 1.3611312541791922}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.185473992393753e-05, "output_power": 2.8610664682742178e-05, "reference_active": true, "delta_db": 1.1698230478899543}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.514034995645172e-07, "output_power": 9.849946163716489e-07, "reference_active": true, "delta_db": 0.6329842585959624}], "active_band_count": 3, "active_band_mae_db": 1.0546461868883696}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018667130520360042, "log_rms_mae_db": 1.2683317932838505, "rms_envelope_nrmse": 0.1394609003520675, "envelope_correlation": 0.9919011204228837, "rms_gain_error_db": 0.6065438131940135, "reference_mean_power": 0.001062265092647352, "output_mean_power": 0.0012214826189539348}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.90294470054441e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004039276312698453, "output_power": 0.00045706824165428803, "reference_active": true, "delta_db": 0.536774836291979}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.530941238269661e-05, "output_power": 5.2061399151104837e-05, "reference_active": true, "delta_db": 0.6032740574758575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.869428114247789e-06, "output_power": 5.867188704268846e-06, "reference_active": true, "delta_db": 0.8095209762573659}], "active_band_count": 3, "active_band_mae_db": 0.6498566233417341}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017791979118134677, "log_rms_mae_db": 0.8780800663322775, "rms_envelope_nrmse": 0.09311311375825246, "envelope_correlation": 0.9979696484948097, "rms_gain_error_db": 0.6065861135018052, "reference_mean_power": 0.0010675566768630375, "output_mean_power": 0.0012275792884568764}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.672011383574165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003902673215159807, "output_power": 0.000442442575485131, "reference_active": true, "delta_db": 0.544947239482616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8376752084991478e-05, "output_power": 2.1554004826261736e-05, "reference_active": true, "delta_db": 0.6925921932585686}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.2996790275380256e-07, "output_power": 3.9560173030765184e-07, "reference_active": true, "delta_db": -0.361778540891235}], "active_band_count": 3, "active_band_mae_db": 0.5331059912108066}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.103910936684106e-05, "log_rms_mae_db": 1.6288864664429354, "rms_envelope_nrmse": 0.16154846228937886, "envelope_correlation": 0.9840937387462086, "rms_gain_error_db": 0.40558207798516355, "reference_mean_power": 0.00018921068158653413, "output_mean_power": 0.0002077322178913386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9505568897447032e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.310547772849957e-05, "output_power": 9.491953461551187e-05, "reference_active": true, "delta_db": 0.08380367400275232}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9017487007684174e-05, "output_power": 1.9539177015114142e-05, "reference_active": true, "delta_db": 0.11753139174074353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.040878232439125e-06, "output_power": 4.799410280276701e-06, "reference_active": true, "delta_db": -0.21318329311717127}], "active_band_count": 3, "active_band_mae_db": 0.13817278628688903}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.917454226005249e-05, "log_rms_mae_db": 1.0485185258305452, "rms_envelope_nrmse": 0.10609255717404191, "envelope_correlation": 0.9935656063621127, "rms_gain_error_db": 0.4058600244041334, "reference_mean_power": 0.00019006200677954335, "output_mean_power": 0.00020868023297883603}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8235095603824036e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.910190725222932e-05, "output_power": 9.090857828774253e-05, "reference_active": true, "delta_db": 0.08717865584000743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.361035171170285e-06, "output_power": 9.768962814762735e-06, "reference_active": true, "delta_db": 0.18524579506640476}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6146687278338854e-07, "output_power": 3.258725388949762e-07, "reference_active": true, "delta_db": -0.45020737590786586}], "active_band_count": 3, "active_band_mae_db": 0.24087727560475936}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9334561245547912, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cec11512f0cd5c903cd59a12e54e47d5a89205e28d4925221494716e004c65ff", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9919011204228837, "standard_deviation": null, "minimum": 0.9919011204228837, "maximum": 0.9919011204228837, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2683317932838505, "standard_deviation": null, "minimum": 1.2683317932838505, "maximum": 1.2683317932838505, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6498566233417341, "standard_deviation": null, "minimum": 0.6498566233417341, "maximum": 0.6498566233417341, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1394609003520675, "standard_deviation": null, "minimum": 0.1394609003520675, "maximum": 0.1394609003520675, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9979696484948097, "standard_deviation": null, "minimum": 0.9979696484948097, "maximum": 0.9979696484948097, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8780800663322775, "standard_deviation": null, "minimum": 0.8780800663322775, "maximum": 0.8780800663322775, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5331059912108066, "standard_deviation": null, "minimum": 0.5331059912108066, "maximum": 0.5331059912108066, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09311311375825246, "standard_deviation": null, "minimum": 0.09311311375825246, "maximum": 0.09311311375825246, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9840937387462086, "standard_deviation": null, "minimum": 0.9840937387462086, "maximum": 0.9840937387462086, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6288864664429354, "standard_deviation": null, "minimum": 1.6288864664429354, "maximum": 1.6288864664429354, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.13817278628688903, "standard_deviation": null, "minimum": 0.13817278628688903, "maximum": 0.13817278628688903, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16154846228937886, "standard_deviation": null, "minimum": 0.16154846228937886, "maximum": 0.16154846228937886, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9935656063621127, "standard_deviation": null, "minimum": 0.9935656063621127, "maximum": 0.9935656063621127, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0485185258305452, "standard_deviation": null, "minimum": 1.0485185258305452, "maximum": 1.0485185258305452, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.24087727560475936, "standard_deviation": null, "minimum": 0.24087727560475936, "maximum": 0.24087727560475936, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10609255717404191, "standard_deviation": null, "minimum": 0.10609255717404191, "maximum": 0.10609255717404191, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.992504729296344, "standard_deviation": null, "minimum": 0.992504729296344, "maximum": 0.992504729296344, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5034261808301392, "standard_deviation": null, "minimum": 1.5034261808301392, "maximum": 1.5034261808301392, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.214249769719945, "standard_deviation": null, "minimum": 1.214249769719945, "maximum": 1.214249769719945, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20918598703975716, "standard_deviation": null, "minimum": 0.20918598703975716, "maximum": 0.20918598703975716, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9958622620018229, "standard_deviation": null, "minimum": 0.9958622620018229, "maximum": 0.9958622620018229, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3497971535972821, "standard_deviation": null, "minimum": 1.3497971535972821, "maximum": 1.3497971535972821, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0546461868883696, "standard_deviation": null, "minimum": 1.0546461868883696, "maximum": 1.0546461868883696, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19354864115692005, "standard_deviation": null, "minimum": 0.19354864115692005, "maximum": 0.19354864115692005, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6948741072092696, "standard_deviation": null, "minimum": 0.6948741072092696, "maximum": 0.6948741072092696, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.463843352116619, "standard_deviation": null, "minimum": 5.463843352116619, "maximum": 5.463843352116619, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.0425958184988877, "standard_deviation": null, "minimum": 1.0425958184988877, "maximum": 1.0425958184988877, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0337744353822433, "standard_deviation": null, "minimum": 1.0337744353822433, "maximum": 1.0337744353822433, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9892615353435668, "standard_deviation": null, "minimum": 0.9892615353435668, "maximum": 0.9892615353435668, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.769466241697947, "standard_deviation": null, "minimum": 1.769466241697947, "maximum": 1.769466241697947, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7444196326302036, "standard_deviation": null, "minimum": 0.7444196326302036, "maximum": 0.7444196326302036, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18128309139689466, "standard_deviation": null, "minimum": 0.18128309139689466, "maximum": 0.18128309139689466, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9949535705435767, "standard_deviation": null, "minimum": 0.9949535705435767, "maximum": 0.9949535705435767, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2902919733509164, "standard_deviation": null, "minimum": 1.2902919733509164, "maximum": 1.2902919733509164, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6139975254507667, "standard_deviation": null, "minimum": 0.6139975254507667, "maximum": 0.6139975254507667, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.15331865336162007, "standard_deviation": null, "minimum": 0.15331865336162007, "maximum": 0.15331865336162007, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4970"}, {"case": "librispeech_4970-29095-0008", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.8130173683166504, "predicted_mos": 3.8130173683166504, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 48640, "input_sample_rate": 16000, "input_seconds": 3.04, "input_float32_sha256": "2f57d69507417bd1c32339df6a173cdd725e6e91d22d152cc6a3dffa2fb68fa4", "internal_samples": 48640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.910142599605024}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4172944742405635, "全帯域_10ms_rms_envelope_nrmse": 0.17674167566951318, "全帯域_10ms_envelope_correlation": 0.9873519174125845, "全帯域_10ms_modulation_mae_db": 1.3751078391409068, "全帯域_40ms_log_rms_mae_db": 1.3093454427146183, "全帯域_40ms_rms_envelope_nrmse": 0.14106230958362817, "全帯域_40ms_envelope_correlation": 0.9952862815619237, "全帯域_40ms_modulation_mae_db": 1.4342740970189176, "300–1000 Hz_10ms_log_rms_mae_db": 1.3071696533862414, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20649278345214134, "300–1000 Hz_10ms_envelope_correlation": 0.9884888561167178, "300–1000 Hz_10ms_modulation_mae_db": 1.6059491013384173, "300–1000 Hz_40ms_log_rms_mae_db": 1.0742581801823414, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17890314801554416, "300–1000 Hz_40ms_envelope_correlation": 0.9948880927740814, "300–1000 Hz_40ms_modulation_mae_db": 1.7855856371637453, "1000–3000 Hz_10ms_log_rms_mae_db": 1.129860977148974, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16629202175425023, "1000–3000 Hz_10ms_envelope_correlation": 0.983972103373625, "1000–3000 Hz_10ms_modulation_mae_db": 0.47647855374937026, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7682869117397931, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0803921094755383, "1000–3000 Hz_40ms_envelope_correlation": 0.9969443620116427, "1000–3000 Hz_40ms_modulation_mae_db": 0.4160695970257546, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3656143433876675, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1805137834188653, "3000–6000 Hz_10ms_envelope_correlation": 0.9785749125726682, "3000–6000 Hz_10ms_modulation_mae_db": 0.5012568338674299, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8532116853000284, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0976200525617238, "3000–6000 Hz_40ms_envelope_correlation": 0.9941943039165083, "3000–6000 Hz_40ms_modulation_mae_db": 0.7027706585911911, "mr_stft_db": 5.621244928591408, "band_mae_db": 0.4909848893015754, "rms_gain_error_db": 0.968214286188964, "output_peak": 0.4088194960850355}, "full_scale": {"sample_count": 48640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 48640, "output_samples": 48640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.623069417626376, "spectral_convergence": 0.626893590118708, "spectral_convergence_status": "ok", "amplitude_floor": 9.047519636617716e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.03309082528585, "spectral_convergence": 0.26491150702264227, "spectral_convergence_status": "ok", "amplitude_floor": 7.308511681145121e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.207574542861999, "spectral_convergence": 0.30332141052931666, "spectral_convergence_status": "ok", "amplitude_floor": 4.4367419091404295e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00011152155483091157, "output_power": 0.00011155985489900451, "delta_db": 0.0014912500699049583, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042085142829602937, "output_power": 0.00050103092158411, "delta_db": 0.7573572414390448, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006661514585812611, "output_power": 0.0009322875761085681, "delta_db": 1.4597691359933846, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00015794681682280444, "output_power": 0.00016563558598571053, "delta_db": 0.2064277070657493, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.332154632901203e-05, "output_power": 5.884151478667612e-05, "delta_db": 0.4278110896206047, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2768891588196896e-05, "output_power": 3.347858105572773e-05, "delta_db": 0.09305291162076404, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4909848893015754, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.968214286188964, "pitch": null, "output_rms_dbfs": -27.44044106700514, "output_peak": 0.4088194960850355, "reference_rms_dbfs": -28.408655353194106, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.621244928591408}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 48640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00012692463938083332, "log_rms_mae_db": 1.4172944742405635, "rms_envelope_nrmse": 0.17674167566951318, "envelope_correlation": 0.9873519174125845, "rms_gain_error_db": 0.968214580902822, "reference_mean_power": 0.0014449383611437946, "output_mean_power": 0.0018058047008575847}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.053245386897797e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005746543771154043, "output_power": 0.000732656686855006, "reference_active": true, "delta_db": 1.0549379865610558}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014747895972213286, "output_power": 0.0002043752712498709, "reference_active": true, "delta_db": 1.4169828086116376}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.536231970329893e-05, "output_power": 3.711327203498747e-05, "reference_active": true, "delta_db": 1.653402722250027}], "active_band_count": 3, "active_band_mae_db": 1.3751078391409068}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 0.00010829935323771927, "log_rms_mae_db": 1.3093454427146183, "rms_envelope_nrmse": 0.14106230958362817, "envelope_correlation": 0.9952862815619237, "rms_gain_error_db": 0.9682153687981101, "reference_mean_power": 0.0014593630016451321, "output_mean_power": 0.0018238321561633143}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 9.289432894914324e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005498374692309951, "output_power": 0.000712382042755423, "reference_active": true, "delta_db": 1.1247863157465077}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.817099868184177e-05, "output_power": 8.7642473499515e-05, "reference_active": true, "delta_db": 1.7800810614025644}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4585911877580322e-06, "output_power": 2.0124684325909038e-06, "reference_active": true, "delta_db": 1.3979549139076801}], "active_band_count": 3, "active_band_mae_db": 1.4342740970189176}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 0.00011599347403670498, "log_rms_mae_db": 1.3071696533862414, "rms_envelope_nrmse": 0.20649278345214134, "envelope_correlation": 0.9884888561167178, "rms_gain_error_db": 1.263997697058405, "reference_mean_power": 0.0008943859828976575, "output_mean_power": 0.0011965332013480923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 8.128232732706927e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005209748261573112, "output_power": 0.000710725063892255, "reference_active": true, "delta_db": 1.348848927985429}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010680702813776803, "output_power": 0.00016108957874918766, "reference_active": true, "delta_db": 1.7846761466474048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2431239684221878e-05, "output_power": 3.305867925802607e-05, "reference_active": true, "delta_db": 1.6843222293824176}], "active_band_count": 3, "active_band_mae_db": 1.6059491013384173}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 9.903058317661472e-05, "log_rms_mae_db": 1.0742581801823414, "rms_envelope_nrmse": 0.17890314801554416, "envelope_correlation": 0.9948880927740814, "rms_gain_error_db": 1.2639977199814418, "reference_mean_power": 0.0009033149500834516, "output_mean_power": 0.0012084786159686188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 7.323650960940424e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005139241705701183, "output_power": 0.0007069349305649196, "reference_active": true, "delta_db": 1.3848039760050248}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.0770880999134286e-05, "output_power": 6.673788635853154e-05, "reference_active": true, "delta_db": 2.1402235206559617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2000817341803485e-06, "output_power": 1.8297163425905692e-06, "reference_active": true, "delta_db": 1.8317294148302496}], "active_band_count": 3, "active_band_mae_db": 1.7855856371637453}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 7.992748424465529e-05, "log_rms_mae_db": 1.129860977148974, "rms_envelope_nrmse": 0.16629202175425023, "envelope_correlation": 0.983972103373625, "rms_gain_error_db": 0.35058545765474874, "reference_mean_power": 0.00017342262158164605, "output_mean_power": 0.00018800278934237455}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 1.6075439307790634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.767193199363673e-05, "output_power": 0.0001056783605512957, "reference_active": true, "delta_db": 0.3421628886163788}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.270979881634546e-05, "output_power": 2.507663946818176e-05, "reference_active": true, "delta_db": 0.4305604918884537}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.175922545645147e-06, "output_power": 4.857621397449684e-06, "reference_active": true, "delta_db": 0.6567122807432781}], "active_band_count": 3, "active_band_mae_db": 0.47647855374937026}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 6.970659359658012e-05, "log_rms_mae_db": 0.7682869117397931, "rms_envelope_nrmse": 0.0803921094755383, "envelope_correlation": 0.9969443620116427, "rms_gain_error_db": 0.35058545139438224, "reference_mean_power": 0.00017515363198831316, "output_mean_power": 0.00018987933079007207}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 1.4903701525512362e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.997559182587721e-05, "output_power": 0.00010857043853521307, "reference_active": true, "delta_db": 0.3581760831586206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.524639480988942e-06, "output_power": 1.0848098276088175e-05, "reference_active": true, "delta_db": 0.5650506484875718}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.337301598444833e-07, "output_power": 3.096686569283628e-07, "reference_active": true, "delta_db": -0.3249820594310713}], "active_band_count": 3, "active_band_mae_db": 0.4160695970257546}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 607, "rms_floor": 2.3755576133519224e-05, "log_rms_mae_db": 1.3656143433876675, "rms_envelope_nrmse": 0.1805137834188653, "envelope_correlation": 0.9785749125726682, "rms_gain_error_db": 0.36790159917260146, "reference_mean_power": 3.0329423269641833e-05, "output_mean_power": 3.301066794003134e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 607, "frequency_resolution_hz": 0.32948929159802304, "power_floor": 3.170053749158532e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1550798308210337e-05, "output_power": 1.2331153736850488e-05, "reference_active": true, "delta_db": 0.2839171170919278}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.1283799417647094e-06, "output_power": 7.311158151895679e-06, "reference_active": true, "delta_db": 0.7664049602515459}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7104244593685833e-06, "output_power": 1.5408453638072146e-06, "reference_active": true, "delta_db": -0.45344842425881604}], "active_band_count": 3, "active_band_mae_db": 0.5012568338674299}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 48640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 601, "rms_floor": 2.1008424841575618e-05, "log_rms_mae_db": 0.8532116853000284, "rms_envelope_nrmse": 0.0976200525617238, "envelope_correlation": 0.9941943039165083, "rms_gain_error_db": 0.3679056417006877, "reference_mean_power": 3.0632047924156045e-05, "output_mean_power": 3.3340076881620586e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 601, "frequency_resolution_hz": 0.3327787021630616, "power_floor": 2.7930999787597993e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1345569218097283e-05, "output_power": 1.2119320148323244e-05, "reference_active": true, "delta_db": 0.28651968366215685}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6424993904119263e-06, "output_power": 3.4451506072470157e-06, "reference_active": true, "delta_db": 1.1519331636183652}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1775222693841551e-07, "output_power": 1.0092138836655312e-07, "reference_active": true, "delta_db": -0.6698591284930516}], "active_band_count": 3, "active_band_mae_db": 0.7027706585911911}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9626240651933529, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2bca96ec83d51f19ed20b0fce936973e0e1b5718542a65b92e041f26c2707830", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.983972103373625, "standard_deviation": null, "minimum": 0.983972103373625, "maximum": 0.983972103373625, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.129860977148974, "standard_deviation": null, "minimum": 1.129860977148974, "maximum": 1.129860977148974, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47647855374937026, "standard_deviation": null, "minimum": 0.47647855374937026, "maximum": 0.47647855374937026, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16629202175425023, "standard_deviation": null, "minimum": 0.16629202175425023, "maximum": 0.16629202175425023, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969443620116427, "standard_deviation": null, "minimum": 0.9969443620116427, "maximum": 0.9969443620116427, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7682869117397931, "standard_deviation": null, "minimum": 0.7682869117397931, "maximum": 0.7682869117397931, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4160695970257546, "standard_deviation": null, "minimum": 0.4160695970257546, "maximum": 0.4160695970257546, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0803921094755383, "standard_deviation": null, "minimum": 0.0803921094755383, "maximum": 0.0803921094755383, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9785749125726682, "standard_deviation": null, "minimum": 0.9785749125726682, "maximum": 0.9785749125726682, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3656143433876675, "standard_deviation": null, "minimum": 1.3656143433876675, "maximum": 1.3656143433876675, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5012568338674299, "standard_deviation": null, "minimum": 0.5012568338674299, "maximum": 0.5012568338674299, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1805137834188653, "standard_deviation": null, "minimum": 0.1805137834188653, "maximum": 0.1805137834188653, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9941943039165083, "standard_deviation": null, "minimum": 0.9941943039165083, "maximum": 0.9941943039165083, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8532116853000284, "standard_deviation": null, "minimum": 0.8532116853000284, "maximum": 0.8532116853000284, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7027706585911911, "standard_deviation": null, "minimum": 0.7027706585911911, "maximum": 0.7027706585911911, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0976200525617238, "standard_deviation": null, "minimum": 0.0976200525617238, "maximum": 0.0976200525617238, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9884888561167178, "standard_deviation": null, "minimum": 0.9884888561167178, "maximum": 0.9884888561167178, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3071696533862414, "standard_deviation": null, "minimum": 1.3071696533862414, "maximum": 1.3071696533862414, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.6059491013384173, "standard_deviation": null, "minimum": 1.6059491013384173, "maximum": 1.6059491013384173, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20649278345214134, "standard_deviation": null, "minimum": 0.20649278345214134, "maximum": 0.20649278345214134, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9948880927740814, "standard_deviation": null, "minimum": 0.9948880927740814, "maximum": 0.9948880927740814, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.0742581801823414, "standard_deviation": null, "minimum": 1.0742581801823414, "maximum": 1.0742581801823414, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7855856371637453, "standard_deviation": null, "minimum": 1.7855856371637453, "maximum": 1.7855856371637453, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17890314801554416, "standard_deviation": null, "minimum": 0.17890314801554416, "maximum": 0.17890314801554416, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.4909848893015754, "standard_deviation": null, "minimum": 0.4909848893015754, "maximum": 0.4909848893015754, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.621244928591408, "standard_deviation": null, "minimum": 5.621244928591408, "maximum": 5.621244928591408, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4088194960850355, "standard_deviation": null, "minimum": 0.4088194960850355, "maximum": 0.4088194960850355, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.968214286188964, "standard_deviation": null, "minimum": 0.968214286188964, "maximum": 0.968214286188964, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9873519174125845, "standard_deviation": null, "minimum": 0.9873519174125845, "maximum": 0.9873519174125845, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4172944742405635, "standard_deviation": null, "minimum": 1.4172944742405635, "maximum": 1.4172944742405635, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.3751078391409068, "standard_deviation": null, "minimum": 1.3751078391409068, "maximum": 1.3751078391409068, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17674167566951318, "standard_deviation": null, "minimum": 0.17674167566951318, "maximum": 0.17674167566951318, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9952862815619237, "standard_deviation": null, "minimum": 0.9952862815619237, "maximum": 0.9952862815619237, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.3093454427146183, "standard_deviation": null, "minimum": 1.3093454427146183, "maximum": 1.3093454427146183, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4342740970189176, "standard_deviation": null, "minimum": 1.4342740970189176, "maximum": 1.4342740970189176, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.14106230958362817, "standard_deviation": null, "minimum": 0.14106230958362817, "maximum": 0.14106230958362817, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4970"}, {"case": "librispeech_4992-23283-0016", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.134223222732544, "predicted_mos": 3.134223222732544, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 71920, "input_sample_rate": 16000, "input_seconds": 4.495, "input_float32_sha256": "f6659561c14057cd543fbb5d1bd79004b67d54b37e4dd886707bd8f736ce101f", "internal_samples": 71920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.546527999918908}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.5100929754408736, "全帯域_10ms_rms_envelope_nrmse": 0.19264904203850494, "全帯域_10ms_envelope_correlation": 0.9868449012950155, "全帯域_10ms_modulation_mae_db": 1.0199257687860432, "全帯域_40ms_log_rms_mae_db": 2.106827246469607, "全帯域_40ms_rms_envelope_nrmse": 0.1703480482291097, "全帯域_40ms_envelope_correlation": 0.9935096562070223, "全帯域_40ms_modulation_mae_db": 0.9772463959903036, "300–1000 Hz_10ms_log_rms_mae_db": 1.7508545136893616, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.225108692583789, "300–1000 Hz_10ms_envelope_correlation": 0.9893663922359449, "300–1000 Hz_10ms_modulation_mae_db": 1.3042145718684377, "300–1000 Hz_40ms_log_rms_mae_db": 1.4529457896072486, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.20675863791771362, "300–1000 Hz_40ms_envelope_correlation": 0.9944642303305421, "300–1000 Hz_40ms_modulation_mae_db": 1.8053821232626097, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3833317451075515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17808688981652432, "1000–3000 Hz_10ms_envelope_correlation": 0.9871581687411953, "1000–3000 Hz_10ms_modulation_mae_db": 1.5441580233794554, "1000–3000 Hz_40ms_log_rms_mae_db": 1.011215680585824, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14187594349266552, "1000–3000 Hz_40ms_envelope_correlation": 0.9943686518624733, "1000–3000 Hz_40ms_modulation_mae_db": 1.1918976317886807, "3000–6000 Hz_10ms_log_rms_mae_db": 1.392067575397672, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2613212591838259, "3000–6000 Hz_10ms_envelope_correlation": 0.9613866556118403, "3000–6000 Hz_10ms_modulation_mae_db": 1.495909298375265, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9658503242369579, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14639111728184662, "3000–6000 Hz_40ms_envelope_correlation": 0.9898797709042516, "3000–6000 Hz_40ms_modulation_mae_db": 1.9281459019168263, "mr_stft_db": 5.821895609556357, "band_mae_db": 0.9637123601277363, "rms_gain_error_db": 1.1549980192513072, "output_peak": 0.37772120499379946}, "full_scale": {"sample_count": 71920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 71920, "output_samples": 71920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.140813204212726, "spectral_convergence": 0.5493492911478389, "spectral_convergence_status": "ok", "amplitude_floor": 7.580332405627278e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.470379123493848, "spectral_convergence": 0.3030140719697655, "spectral_convergence_status": "ok", "amplitude_floor": 6.818562526348602e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.854494500962494, "spectral_convergence": 0.3285602634582767, "spectral_convergence_status": "ok", "amplitude_floor": 5.715517707749406e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020741335082478346, "output_power": 0.00019210131987572958, "delta_db": -0.3330635891759109, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006538184008046484, "output_power": 0.0008470094960795402, "delta_db": 1.1243114033747816, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004887436095925812, "output_power": 0.0007325862286935097, "delta_db": 1.7577765848291982, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 3.3579219734638266e-05, "output_power": 4.58129048270359e-05, "delta_db": 1.3491722924571892, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.100095928785279e-05, "output_power": 6.98490750433972e-05, "delta_db": 0.5882399478126676, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.8128919162253983e-05, "output_power": 2.0957667542776714e-05, "delta_db": 0.6297103431166704, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9637123601277363, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1549980192513072, "pitch": null, "output_rms_dbfs": -27.19349550909743, "output_peak": 0.37772120499379946, "reference_rms_dbfs": -28.34849352834874, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.821895609556357}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 71920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 0.00010701965100918563, "log_rms_mae_db": 2.5100929754408736, "rms_envelope_nrmse": 0.19264904203850494, "envelope_correlation": 0.9868449012950155, "rms_gain_error_db": 1.1549981541998942, "reference_mean_power": 0.0014643132072841782, "output_mean_power": 0.0019104417256407523}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 5.509210445243936e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00038836857080924806, "output_power": 0.0005444516400830879, "reference_active": true, "delta_db": 1.4671523353721723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3200931448411458e-05, "output_power": 3.139643014134081e-05, "reference_active": true, "delta_db": 1.3137484955516827}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.17358638897755e-06, "output_power": 9.781980723798352e-06, "reference_active": true, "delta_db": 0.2788764754342747}], "active_band_count": 3, "active_band_mae_db": 1.0199257687860432}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 0.00010161161302678278, "log_rms_mae_db": 2.106827246469607, "rms_envelope_nrmse": 0.1703480482291097, "envelope_correlation": 0.9935096562070223, "rms_gain_error_db": 1.154998697009757, "reference_mean_power": 0.0014741623423244587, "output_mean_power": 0.0019232918115709556}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 4.939959887293225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00035657053350397007, "output_power": 0.0005002753324436863, "reference_active": true, "delta_db": 1.4706363840970347}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.1722039793266e-06, "output_power": 1.089981008462252e-05, "reference_active": true, "delta_db": 1.2507973283034257}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.984441222185337e-07, "output_power": 6.654280736761503e-07, "reference_active": true, "delta_db": -0.21030547557045026}], "active_band_count": 3, "active_band_mae_db": 0.9772463959903036}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 9.15514590150253e-05, "log_rms_mae_db": 1.7508545136893616, "rms_envelope_nrmse": 0.225108692583789, "envelope_correlation": 0.9893663922359449, "rms_gain_error_db": 1.484870181825109, "reference_mean_power": 0.0006933001790048939, "output_mean_power": 0.0009759067694160222}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 3.8791214865659925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002719560122465178, "output_power": 0.0004029653136679186, "reference_active": true, "delta_db": 1.7076900028574429}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.187110727790427e-05, "output_power": 3.257945320339953e-05, "reference_active": true, "delta_db": 1.7307302025416293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.698791354460839e-06, "output_power": 6.00584472991769e-06, "reference_active": true, "delta_db": -0.4742235102062412}], "active_band_count": 3, "active_band_mae_db": 1.3042145718684377}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 8.797925167476037e-05, "log_rms_mae_db": 1.4529457896072486, "rms_envelope_nrmse": 0.20675863791771362, "envelope_correlation": 0.9944642303305421, "rms_gain_error_db": 1.4848700916582223, "reference_mean_power": 0.000697963583829786, "output_mean_power": 0.0009824710749407935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.5511880459540643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002583969788516632, "output_power": 0.00038377166663523636, "reference_active": true, "delta_db": 1.7178547655363474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.105856462043575e-06, "output_power": 1.2207316914903234e-05, "reference_active": true, "delta_db": 1.7782131044845075}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.388944870715119e-07, "output_power": 4.10602217701991e-07, "reference_active": true, "delta_db": -1.9200784997669742}], "active_band_count": 3, "active_band_mae_db": 1.8053821232626097}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 4.592612274041097e-05, "log_rms_mae_db": 1.3833317451075515, "rms_envelope_nrmse": 0.17808688981652432, "envelope_correlation": 0.9871581687411953, "rms_gain_error_db": 0.8956461588122082, "reference_mean_power": 8.288074215773485e-05, "output_mean_power": 0.00010186341858167401}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 4.287875385442721e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2411523150580774e-05, "output_power": 2.8913094294743064e-05, "reference_active": true, "delta_db": 1.1062319916358252}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.113877152498063e-06, "output_power": 4.607912212536328e-06, "reference_active": true, "delta_db": 1.7020272160055092}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.881539635700238e-07, "output_power": 1.503993998376041e-06, "reference_active": true, "delta_db": 1.8242148624970316}], "active_band_count": 3, "active_band_mae_db": 1.5441580233794554}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 4.2997336393594785e-05, "log_rms_mae_db": 1.011215680585824, "rms_envelope_nrmse": 0.14187594349266552, "envelope_correlation": 0.9943686518624733, "rms_gain_error_db": 0.8956473257944536, "reference_mean_power": 8.343814618382176e-05, "output_mean_power": 0.0001025485157704237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 3.9683446107037247e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.063395086340567e-05, "output_power": 2.6743422228610305e-05, "reference_active": true, "delta_db": 1.1263458908130448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4408129681959194e-06, "output_power": 2.0859364563220893e-06, "reference_active": true, "delta_db": 1.6069346567734506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.9896369199765786e-08, "output_power": 6.057733722818484e-08, "reference_active": true, "delta_db": 0.8424123477795462}], "active_band_count": 3, "active_band_mae_db": 1.1918976317886807}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 898, "rms_floor": 2.431018241105218e-05, "log_rms_mae_db": 1.392067575397672, "rms_envelope_nrmse": 0.2613212591838259, "envelope_correlation": 0.9613866556118403, "rms_gain_error_db": 0.6460415534064934, "reference_mean_power": 1.5917473583289094e-05, "output_mean_power": 1.8470484735397754e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 898, "frequency_resolution_hz": 0.22271714922048996, "power_floor": 8.500689890351627e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.331755201814068e-06, "output_power": 4.983074308582752e-06, "reference_active": true, "delta_db": 0.6083345791108659}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3672340908387822e-06, "output_power": 1.742592274113846e-06, "reference_active": true, "delta_db": 1.0535290561455486}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.790743563061608e-07, "output_power": 5.349415819577808e-07, "reference_active": true, "delta_db": 2.8258642598693804}], "active_band_count": 3, "active_band_mae_db": 1.495909298375265}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 71920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 892, "rms_floor": 1.765504974955162e-05, "log_rms_mae_db": 0.9658503242369579, "rms_envelope_nrmse": 0.14639111728184662, "envelope_correlation": 0.9898797709042516, "rms_gain_error_db": 0.6460379939222105, "reference_mean_power": 1.6024474415231737e-05, "output_mean_power": 1.85946322414276e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 892, "frequency_resolution_hz": 0.2242152466367713, "power_floor": 7.931595726775402e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.29977249297526e-06, "output_power": 5.191369948537157e-06, "reference_active": true, "delta_db": 0.8183650171554445}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.410418312822663e-07, "output_power": 9.958870896914132e-07, "reference_active": true, "delta_db": 1.2836737809114447}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9106902570925117e-08, "output_power": 4.460979010612775e-08, "reference_active": true, "delta_db": 3.68239890768359}], "active_band_count": 3, "active_band_mae_db": 1.9281459019168263}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9313477405238563, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5bc9ea9559c582063126648f3b1be4384d1cf734239f39763e4177c4c02bdf93", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9871581687411953, "standard_deviation": null, "minimum": 0.9871581687411953, "maximum": 0.9871581687411953, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3833317451075515, "standard_deviation": null, "minimum": 1.3833317451075515, "maximum": 1.3833317451075515, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.5441580233794554, "standard_deviation": null, "minimum": 1.5441580233794554, "maximum": 1.5441580233794554, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17808688981652432, "standard_deviation": null, "minimum": 0.17808688981652432, "maximum": 0.17808688981652432, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9943686518624733, "standard_deviation": null, "minimum": 0.9943686518624733, "maximum": 0.9943686518624733, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.011215680585824, "standard_deviation": null, "minimum": 1.011215680585824, "maximum": 1.011215680585824, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1918976317886807, "standard_deviation": null, "minimum": 1.1918976317886807, "maximum": 1.1918976317886807, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14187594349266552, "standard_deviation": null, "minimum": 0.14187594349266552, "maximum": 0.14187594349266552, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9613866556118403, "standard_deviation": null, "minimum": 0.9613866556118403, "maximum": 0.9613866556118403, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.392067575397672, "standard_deviation": null, "minimum": 1.392067575397672, "maximum": 1.392067575397672, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.495909298375265, "standard_deviation": null, "minimum": 1.495909298375265, "maximum": 1.495909298375265, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2613212591838259, "standard_deviation": null, "minimum": 0.2613212591838259, "maximum": 0.2613212591838259, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9898797709042516, "standard_deviation": null, "minimum": 0.9898797709042516, "maximum": 0.9898797709042516, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9658503242369579, "standard_deviation": null, "minimum": 0.9658503242369579, "maximum": 0.9658503242369579, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.9281459019168263, "standard_deviation": null, "minimum": 1.9281459019168263, "maximum": 1.9281459019168263, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14639111728184662, "standard_deviation": null, "minimum": 0.14639111728184662, "maximum": 0.14639111728184662, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9893663922359449, "standard_deviation": null, "minimum": 0.9893663922359449, "maximum": 0.9893663922359449, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7508545136893616, "standard_deviation": null, "minimum": 1.7508545136893616, "maximum": 1.7508545136893616, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3042145718684377, "standard_deviation": null, "minimum": 1.3042145718684377, "maximum": 1.3042145718684377, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.225108692583789, "standard_deviation": null, "minimum": 0.225108692583789, "maximum": 0.225108692583789, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9944642303305421, "standard_deviation": null, "minimum": 0.9944642303305421, "maximum": 0.9944642303305421, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4529457896072486, "standard_deviation": null, "minimum": 1.4529457896072486, "maximum": 1.4529457896072486, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.8053821232626097, "standard_deviation": null, "minimum": 1.8053821232626097, "maximum": 1.8053821232626097, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20675863791771362, "standard_deviation": null, "minimum": 0.20675863791771362, "maximum": 0.20675863791771362, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9637123601277363, "standard_deviation": null, "minimum": 0.9637123601277363, "maximum": 0.9637123601277363, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.821895609556357, "standard_deviation": null, "minimum": 5.821895609556357, "maximum": 5.821895609556357, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.37772120499379946, "standard_deviation": null, "minimum": 0.37772120499379946, "maximum": 0.37772120499379946, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1549980192513072, "standard_deviation": null, "minimum": 1.1549980192513072, "maximum": 1.1549980192513072, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9868449012950155, "standard_deviation": null, "minimum": 0.9868449012950155, "maximum": 0.9868449012950155, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.5100929754408736, "standard_deviation": null, "minimum": 2.5100929754408736, "maximum": 2.5100929754408736, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0199257687860432, "standard_deviation": null, "minimum": 1.0199257687860432, "maximum": 1.0199257687860432, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19264904203850494, "standard_deviation": null, "minimum": 0.19264904203850494, "maximum": 0.19264904203850494, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9935096562070223, "standard_deviation": null, "minimum": 0.9935096562070223, "maximum": 0.9935096562070223, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.106827246469607, "standard_deviation": null, "minimum": 2.106827246469607, "maximum": 2.106827246469607, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9772463959903036, "standard_deviation": null, "minimum": 0.9772463959903036, "maximum": 0.9772463959903036, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1703480482291097, "standard_deviation": null, "minimum": 0.1703480482291097, "maximum": 0.1703480482291097, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4992"}, {"case": "librispeech_4992-41797-0000", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 1.9364440441131592, "predicted_mos": 1.9364440441131592, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87760, "input_sample_rate": 16000, "input_seconds": 5.485, "input_float32_sha256": "0991c33f209ba7bf64995bbdd488915a4194a831ee77d4a6e0d01c5021e4aab2", "internal_samples": 87760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 7.394259400200099}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.947404884213386, "全帯域_10ms_rms_envelope_nrmse": 0.322148677944948, "全帯域_10ms_envelope_correlation": 0.9830213685601346, "全帯域_10ms_modulation_mae_db": 3.684360666527208, "全帯域_40ms_log_rms_mae_db": 1.6594615442122735, "全帯域_40ms_rms_envelope_nrmse": 0.3097020698082092, "全帯域_40ms_envelope_correlation": 0.9878660094974672, "全帯域_40ms_modulation_mae_db": 3.6188303288599024, "300–1000 Hz_10ms_log_rms_mae_db": 1.9882880831062093, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.3964370060696223, "300–1000 Hz_10ms_envelope_correlation": 0.9821670181268456, "300–1000 Hz_10ms_modulation_mae_db": 3.846301464451346, "300–1000 Hz_40ms_log_rms_mae_db": 2.201619086848617, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.3847588233697988, "300–1000 Hz_40ms_envelope_correlation": 0.9868343219293065, "300–1000 Hz_40ms_modulation_mae_db": 3.6256311892883293, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2409863729364496, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2114963860707595, "1000–3000 Hz_10ms_envelope_correlation": 0.9935802204631968, "1000–3000 Hz_10ms_modulation_mae_db": 1.636602653635455, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0743840706630052, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.2003013414358832, "1000–3000 Hz_40ms_envelope_correlation": 0.9961904032838, "1000–3000 Hz_40ms_modulation_mae_db": 1.5717857891200573, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5017621620025636, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18614392715269654, "3000–6000 Hz_10ms_envelope_correlation": 0.9803073008957011, "3000–6000 Hz_10ms_modulation_mae_db": 0.3046004239955619, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1383174859245524, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13239402834637712, "3000–6000 Hz_40ms_envelope_correlation": 0.9919237605812908, "3000–6000 Hz_40ms_modulation_mae_db": 0.5723167556005259, "mr_stft_db": 5.4331984868895935, "band_mae_db": 1.4031035769720244, "rms_gain_error_db": 2.14770213898895, "output_peak": 0.7176128955054474}, "full_scale": {"sample_count": 87760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87760, "output_samples": 87760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.681471724344478, "spectral_convergence": 0.45970724711276534, "spectral_convergence_status": "ok", "amplitude_floor": 1.7869460513029852e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.062936034131801, "spectral_convergence": 0.37234707786574733, "spectral_convergence_status": "ok", "amplitude_floor": 1.6383670687778497e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.555187702192501, "spectral_convergence": 0.4051648911340775, "spectral_convergence_status": "ok", "amplitude_floor": 9.631824221799184e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 9.013402842456697e-05, "output_power": 9.670209477524483e-05, "delta_db": 0.30547100202946975, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00191544195180521, "output_power": 0.0033363798011628487, "delta_db": 2.4100648820127124, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0015517446606579416, "output_power": 0.0025578401986190223, "delta_db": 2.170531486771525, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955960266218775, "output_power": 0.0007110881501995654, "delta_db": 1.5679562566117686, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000158155871967105, "output_power": 0.0002071602256566548, "delta_db": 1.1722105450136828, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.514760940569477e-06, "output_power": 1.141922446864146e-05, "delta_db": 0.7923872893929874, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.4031035769720244, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 2.14770213898895, "pitch": null, "output_rms_dbfs": -21.598568982712905, "output_peak": 0.7176128955054474, "reference_rms_dbfs": -23.74627112170186, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.4331984868895935}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002415922668587691, "log_rms_mae_db": 1.947404884213386, "rms_envelope_nrmse": 0.322148677944948, "envelope_correlation": 0.9830213685601346, "rms_gain_error_db": 2.1477024607166193, "reference_mean_power": 0.004224437839878649, "output_mean_power": 0.006926904022156411}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.932071872304039e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008459874532766979, "output_power": 0.0016049889555340103, "reference_active": true, "delta_db": 2.7810812610034903}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.267978041368795e-05, "output_power": 0.00026005884247529127, "reference_active": true, "delta_db": 4.480866290222108}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0604587197812338e-05, "output_power": 4.932616631500459e-05, "reference_active": true, "delta_db": 3.7911344483560248}], "active_band_count": 3, "active_band_mae_db": 3.684360666527208}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00018058591873130477, "log_rms_mae_db": 1.6594615442122735, "rms_envelope_nrmse": 0.3097020698082092, "envelope_correlation": 0.9878660094974672, "rms_gain_error_db": 2.147702971966298, "reference_mean_power": 0.004247690453399677, "output_mean_power": 0.006965032665998228}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8509413564750568e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008019972220007684, "output_power": 0.001550899249286186, "reference_active": true, "delta_db": 2.8641072180149374}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.130657727931199e-05, "output_power": 0.00011916676659011604, "reference_active": true, "delta_db": 4.601359453611517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0523532353898244e-06, "output_power": 2.2975447133520323e-06, "reference_active": true, "delta_db": 3.3910243149532544}], "active_band_count": 3, "active_band_mae_db": 3.6188303288599024}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.0002374784601360974, "log_rms_mae_db": 1.9882880831062093, "rms_envelope_nrmse": 0.3964370060696223, "envelope_correlation": 0.9821670181268456, "rms_gain_error_db": 2.640935161133436, "reference_mean_power": 0.0028785349502786395, "output_mean_power": 0.005287678275741346}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 1.6369199856429154e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008090138264371756, "output_power": 0.001642042568971858, "reference_active": true, "delta_db": 3.074284677759244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000105081165545653, "output_power": 0.0002868826258709203, "reference_active": true, "delta_db": 4.361793661108806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7476383653056024e-05, "output_power": 4.4950463629736824e-05, "reference_active": true, "delta_db": 4.1028260544859885}], "active_band_count": 3, "active_band_mae_db": 3.846301464451346}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.0001750973468816212, "log_rms_mae_db": 2.201619086848617, "rms_envelope_nrmse": 0.3847588233697988, "envelope_correlation": 0.9868343219293065, "rms_gain_error_db": 2.6409348854798327, "reference_mean_power": 0.002894380018498304, "output_mean_power": 0.005316784280374212}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.567165950582092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007701013013376472, "output_power": 0.0015907720038193579, "reference_active": true, "delta_db": 3.1506008190825368}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.715572645231493e-05, "output_power": 0.0001303162117429174, "reference_active": true, "delta_db": 4.414640069440272}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1197000044443696e-06, "output_power": 2.4003078633330876e-06, "reference_active": true, "delta_db": 3.3116526793421794}], "active_band_count": 3, "active_band_mae_db": 3.6256311892883293}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 0.00012170766433419248, "log_rms_mae_db": 1.2409863729364496, "rms_envelope_nrmse": 0.2114963860707595, "envelope_correlation": 0.9935802204631968, "rms_gain_error_db": 1.4637472303801888, "reference_mean_power": 0.0006352125631157451, "output_mean_power": 0.0008898028694572586}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 2.616269223692633e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010398636594123316, "output_power": 0.00013454662181297815, "reference_active": true, "delta_db": 1.1189639701679785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0847306945521593e-05, "output_power": 1.6876784375991712e-05, "reference_active": true, "delta_db": 1.919677721883399}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2384759945240008e-06, "output_power": 4.982614746787438e-06, "reference_active": true, "delta_db": 1.8711662688549877}], "active_band_count": 3, "active_band_mae_db": 1.636602653635455}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 0.00010363700534457545, "log_rms_mae_db": 1.0743840706630052, "rms_envelope_nrmse": 0.2003013414358832, "envelope_correlation": 0.9961904032838, "rms_gain_error_db": 1.4637473506661045, "reference_mean_power": 0.0006387091058518984, "output_mean_power": 0.0008947008354001382}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.5308592276752824e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.945508810428335e-05, "output_power": 0.00012935184932755264, "reference_active": true, "delta_db": 1.1414563569617575}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.423063898822258e-06, "output_power": 8.479884934881148e-06, "reference_active": true, "delta_db": 1.9414523760875102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8432082747665845e-07, "output_power": 2.684227212343493e-07, "reference_active": true, "delta_db": 1.6324486343109035}], "active_band_count": 3, "active_band_mae_db": 1.5717857891200573}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1096, "rms_floor": 2.9435709173680706e-05, "log_rms_mae_db": 1.5017621620025636, "rms_envelope_nrmse": 0.18614392715269654, "envelope_correlation": 0.9803073008957011, "rms_gain_error_db": 0.6658850573032694, "reference_mean_power": 1.4688872454843858e-05, "output_mean_power": 1.7122886011291027e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1096, "frequency_resolution_hz": 0.18248175182481752, "power_floor": 9.038086104701173e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.4018145980274633e-06, "output_power": 3.733355576904601e-06, "reference_active": true, "delta_db": 0.40388715033076705}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2285201909756879e-06, "output_power": 1.3569008103813936e-06, "reference_active": true, "delta_db": 0.4316580324754687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.73608399821043e-07, "output_power": 2.6872236760429427e-07, "reference_active": true, "delta_db": -0.0782560891804501}], "active_band_count": 3, "active_band_mae_db": 0.3046004239955619}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 2.1616481165467548e-05, "log_rms_mae_db": 1.1383174859245524, "rms_envelope_nrmse": 0.13239402834637712, "envelope_correlation": 0.9919237605812908, "rms_gain_error_db": 0.6658854049560667, "reference_mean_power": 1.4769720024935061e-05, "output_mean_power": 1.7217131772896314e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.51604517212682e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5641156211998326e-06, "output_power": 3.856853984294612e-06, "reference_active": true, "delta_db": 0.34281413063721167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790775026435671e-07, "output_power": 6.723567536691068e-07, "reference_active": true, "delta_db": 0.6486307884825661}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8657206259627155e-08, "output_power": 1.5786871647178547e-08, "reference_active": true, "delta_db": -0.7255053476817999}], "active_band_count": 3, "active_band_mae_db": 0.5723167556005259}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.933322696756075, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f6ddabf1b4e4b76d8f34d0459ca5797edec975dd78c82da9056c0915381bb5e2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9935802204631968, "standard_deviation": null, "minimum": 0.9935802204631968, "maximum": 0.9935802204631968, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2409863729364496, "standard_deviation": null, "minimum": 1.2409863729364496, "maximum": 1.2409863729364496, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.636602653635455, "standard_deviation": null, "minimum": 1.636602653635455, "maximum": 1.636602653635455, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2114963860707595, "standard_deviation": null, "minimum": 0.2114963860707595, "maximum": 0.2114963860707595, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9961904032838, "standard_deviation": null, "minimum": 0.9961904032838, "maximum": 0.9961904032838, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0743840706630052, "standard_deviation": null, "minimum": 1.0743840706630052, "maximum": 1.0743840706630052, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5717857891200573, "standard_deviation": null, "minimum": 1.5717857891200573, "maximum": 1.5717857891200573, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2003013414358832, "standard_deviation": null, "minimum": 0.2003013414358832, "maximum": 0.2003013414358832, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9803073008957011, "standard_deviation": null, "minimum": 0.9803073008957011, "maximum": 0.9803073008957011, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5017621620025636, "standard_deviation": null, "minimum": 1.5017621620025636, "maximum": 1.5017621620025636, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3046004239955619, "standard_deviation": null, "minimum": 0.3046004239955619, "maximum": 0.3046004239955619, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18614392715269654, "standard_deviation": null, "minimum": 0.18614392715269654, "maximum": 0.18614392715269654, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9919237605812908, "standard_deviation": null, "minimum": 0.9919237605812908, "maximum": 0.9919237605812908, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1383174859245524, "standard_deviation": null, "minimum": 1.1383174859245524, "maximum": 1.1383174859245524, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5723167556005259, "standard_deviation": null, "minimum": 0.5723167556005259, "maximum": 0.5723167556005259, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13239402834637712, "standard_deviation": null, "minimum": 0.13239402834637712, "maximum": 0.13239402834637712, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9821670181268456, "standard_deviation": null, "minimum": 0.9821670181268456, "maximum": 0.9821670181268456, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9882880831062093, "standard_deviation": null, "minimum": 1.9882880831062093, "maximum": 1.9882880831062093, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 3.846301464451346, "standard_deviation": null, "minimum": 3.846301464451346, "maximum": 3.846301464451346, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3964370060696223, "standard_deviation": null, "minimum": 0.3964370060696223, "maximum": 0.3964370060696223, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9868343219293065, "standard_deviation": null, "minimum": 0.9868343219293065, "maximum": 0.9868343219293065, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.201619086848617, "standard_deviation": null, "minimum": 2.201619086848617, "maximum": 2.201619086848617, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 3.6256311892883293, "standard_deviation": null, "minimum": 3.6256311892883293, "maximum": 3.6256311892883293, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.3847588233697988, "standard_deviation": null, "minimum": 0.3847588233697988, "maximum": 0.3847588233697988, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.4031035769720244, "standard_deviation": null, "minimum": 1.4031035769720244, "maximum": 1.4031035769720244, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.4331984868895935, "standard_deviation": null, "minimum": 5.4331984868895935, "maximum": 5.4331984868895935, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7176128955054474, "standard_deviation": null, "minimum": 0.7176128955054474, "maximum": 0.7176128955054474, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 2.14770213898895, "standard_deviation": null, "minimum": 2.14770213898895, "maximum": 2.14770213898895, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9830213685601346, "standard_deviation": null, "minimum": 0.9830213685601346, "maximum": 0.9830213685601346, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.947404884213386, "standard_deviation": null, "minimum": 1.947404884213386, "maximum": 1.947404884213386, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 3.684360666527208, "standard_deviation": null, "minimum": 3.684360666527208, "maximum": 3.684360666527208, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.322148677944948, "standard_deviation": null, "minimum": 0.322148677944948, "maximum": 0.322148677944948, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9878660094974672, "standard_deviation": null, "minimum": 0.9878660094974672, "maximum": 0.9878660094974672, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.6594615442122735, "standard_deviation": null, "minimum": 1.6594615442122735, "maximum": 1.6594615442122735, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 3.6188303288599024, "standard_deviation": null, "minimum": 3.6188303288599024, "maximum": 3.6188303288599024, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.3097020698082092, "standard_deviation": null, "minimum": 0.3097020698082092, "maximum": 0.3097020698082092, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4992"}, {"case": "librispeech_4992-41806-0000", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9902095794677734, "predicted_mos": 2.9902095794677734, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7f71c402ac902d3a609590b3e179a8aad18d115c2ec49822b9a825da1bcc1639", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.411279399879277}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5234645103312605, "全帯域_10ms_rms_envelope_nrmse": 0.21622614191162734, "全帯域_10ms_envelope_correlation": 0.9888937384421971, "全帯域_10ms_modulation_mae_db": 1.6615463754522724, "全帯域_40ms_log_rms_mae_db": 1.2565184063438237, "全帯域_40ms_rms_envelope_nrmse": 0.20250363972421825, "全帯域_40ms_envelope_correlation": 0.9930017925456827, "全帯域_40ms_modulation_mae_db": 1.6246222517496685, "300–1000 Hz_10ms_log_rms_mae_db": 1.6589027137822485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.258029801343065, "300–1000 Hz_10ms_envelope_correlation": 0.9873001970849293, "300–1000 Hz_10ms_modulation_mae_db": 1.9987004924432652, "300–1000 Hz_40ms_log_rms_mae_db": 1.4747479287031744, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.24426500246995672, "300–1000 Hz_40ms_envelope_correlation": 0.9912615181568517, "300–1000 Hz_40ms_modulation_mae_db": 1.7753361909088803, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4563805928617344, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20667883807096088, "1000–3000 Hz_10ms_envelope_correlation": 0.9898280214447631, "1000–3000 Hz_10ms_modulation_mae_db": 1.4796762118388473, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1601993758617637, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18846031239124875, "1000–3000 Hz_40ms_envelope_correlation": 0.9939432216754446, "1000–3000 Hz_40ms_modulation_mae_db": 1.5734862769905094, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6043471233270468, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2569740070467292, "3000–6000 Hz_10ms_envelope_correlation": 0.9642517037102821, "3000–6000 Hz_10ms_modulation_mae_db": 0.6720824947966041, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1952909107141028, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18096193808845518, "3000–6000 Hz_40ms_envelope_correlation": 0.9834787858304187, "3000–6000 Hz_40ms_modulation_mae_db": 0.8086018752478568, "mr_stft_db": 5.496919158859445, "band_mae_db": 1.0095639657194964, "rms_gain_error_db": 1.4401212491749584, "output_peak": 0.7581434197190311}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.734981528055353, "spectral_convergence": 0.41895039309161547, "spectral_convergence_status": "ok", "amplitude_floor": 1.726332693189446e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.165000720437141, "spectral_convergence": 0.2973785124352883, "spectral_convergence_status": "ok", "amplitude_floor": 1.4419419879485463e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.590775228085841, "spectral_convergence": 0.3373933650691259, "spectral_convergence_status": "ok", "amplitude_floor": 1.208114881083552e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024021305054769385, "output_power": 0.00024031905980415752, "delta_db": 0.0019161772904630458, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0023981297439394636, "output_power": 0.003289239796950438, "delta_db": 1.372228605484997, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0027873624928200164, "output_power": 0.00408255655616299, "delta_db": 1.6573875797439612, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0004955109173134605, "output_power": 0.0006740933590370442, "delta_db": 1.336668211297269, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00027155220884703173, "output_power": 0.0003506296098065537, "delta_db": 1.1099524864478116, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.6324739911259635e-05, "output_power": 4.150740540228722e-05, "delta_db": 0.5792307340524765, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.0095639657194964, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.4401212491749584, "pitch": null, "output_rms_dbfs": -20.615630494747084, "output_peak": 0.7581434197190311, "reference_rms_dbfs": -22.055751743922045, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.496919158859445}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024330354027503592, "log_rms_mae_db": 1.5234645103312605, "rms_envelope_nrmse": 0.21622614191162734, "envelope_correlation": 0.9888937384421971, "rms_gain_error_db": 1.4400706877277334, "reference_mean_power": 0.006233969602179056, "output_mean_power": 0.008685038521081854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8072421002875186e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017414213578018356, "output_power": 0.002793817208722042, "reference_active": true, "delta_db": 2.0529412149595356}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002371798239966289, "output_power": 0.0004031371602412213, "reference_active": true, "delta_db": 2.3037508979747785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.501630810521967e-05, "output_power": 7.513062577555878e-05, "reference_active": true, "delta_db": 0.627947013422503}], "active_band_count": 3, "active_band_mae_db": 1.6615463754522724}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022924128639936837, "log_rms_mae_db": 1.2565184063438237, "rms_envelope_nrmse": 0.20250363972421825, "envelope_correlation": 0.9930017925456827, "rms_gain_error_db": 1.4404078076164173, "reference_mean_power": 0.006262494274967633, "output_mean_power": 0.008725455794651709}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.530420990426951e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00159929971914277, "output_power": 0.002601707299061135, "reference_active": true, "delta_db": 2.1132857438728707}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010914154654466241, "output_power": 0.00018567311994469894, "reference_active": true, "delta_db": 2.3075893163022836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.773901654262189e-06, "output_power": 5.298740336001399e-06, "reference_active": true, "delta_db": 0.4529916950738508}], "active_band_count": 3, "active_band_mae_db": 1.6246222517496685}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00021113569197140015, "log_rms_mae_db": 1.6589027137822485, "rms_envelope_nrmse": 0.258029801343065, "envelope_correlation": 0.9873001970849293, "rms_gain_error_db": 1.7099760371027093, "reference_mean_power": 0.0043153114305736935, "output_mean_power": 0.006397491939573344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.429798497423617e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014993372451038876, "output_power": 0.0025207017943751618, "reference_active": true, "delta_db": 2.256221410848803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023063830674655453, "output_power": 0.00041183748841951086, "reference_active": true, "delta_db": 2.5179443572157947}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.9359539965921096e-05, "output_power": 7.864735066429179e-05, "reference_active": true, "delta_db": 1.2219357092651981}], "active_band_count": 3, "active_band_mae_db": 1.9987004924432652}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0001994421712728477, "log_rms_mae_db": 1.4747479287031744, "rms_envelope_nrmse": 0.24426500246995672, "envelope_correlation": 0.9912615181568517, "rms_gain_error_db": 1.7100448810955127, "reference_mean_power": 0.004336654014858995, "output_mean_power": 0.006429234447048715}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.221204429766169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014035050384262429, "output_power": 0.002380030868761936, "reference_active": true, "delta_db": 2.2936861380705005}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010608827780695941, "output_power": 0.00018401303086489334, "reference_active": true, "delta_db": 2.391811793087231}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7233786501496316e-06, "output_power": 5.473987815616701e-06, "reference_active": true, "delta_db": 0.640510641568909}], "active_band_count": 3, "active_band_mae_db": 1.7753361909088803}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011279772332784735, "log_rms_mae_db": 1.4563805928617344, "rms_envelope_nrmse": 0.20667883807096088, "envelope_correlation": 0.9898280214447631, "rms_gain_error_db": 1.3014289419888383, "reference_mean_power": 0.0007258838363645759, "output_mean_power": 0.0009795125844735285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4338411984908935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00026856860022608645, "output_power": 0.00035302950539142775, "reference_active": true, "delta_db": 1.1875576859696009}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.89737119770249e-05, "output_power": 3.1717625181422976e-05, "reference_active": true, "delta_db": 2.2314835893410265}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310520152307534e-06, "output_power": 6.7163883130610855e-06, "reference_active": true, "delta_db": 1.0199873602059144}], "active_band_count": 3, "active_band_mae_db": 1.4796762118388473}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010728134925657107, "log_rms_mae_db": 1.1601993758617637, "rms_envelope_nrmse": 0.18846031239124875, "envelope_correlation": 0.9939432216754446, "rms_gain_error_db": 1.3014246738971633, "reference_mean_power": 0.0007295243282152702, "output_mean_power": 0.0009844241214730777}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.220316674170232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000246564866107528, "output_power": 0.0003260944295629546, "reference_active": true, "delta_db": 1.2141218760390944}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.89658406854447e-06, "output_power": 1.4985202215617474e-05, "reference_active": true, "delta_db": 2.264393208570567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.477241032022853e-07, "output_power": 4.628381727620921e-07, "reference_active": true, "delta_db": 1.2419437463618666}], "active_band_count": 3, "active_band_mae_db": 1.5734862769905094}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.800534128114117e-05, "log_rms_mae_db": 1.6043471233270468, "rms_envelope_nrmse": 0.2569740070467292, "envelope_correlation": 0.9642517037102821, "rms_gain_error_db": 0.7369151893335967, "reference_mean_power": 3.947976781678993e-05, "output_mean_power": 4.6780634590643105e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2531812232892484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6595680637210926e-05, "output_power": 1.867844643339307e-05, "reference_active": true, "delta_db": 0.5134568246719792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.516181974676621e-06, "output_power": 9.942648164079155e-06, "reference_active": true, "delta_db": 0.672571385915762}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1029203509908326e-06, "output_power": 2.5459199913015207e-06, "reference_active": true, "delta_db": 0.8302192738020708}], "active_band_count": 3, "active_band_mae_db": 0.6720824947966041}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.0235226592584236e-05, "log_rms_mae_db": 1.1952909107141028, "rms_envelope_nrmse": 0.18096193808845518, "envelope_correlation": 0.9834787858304187, "rms_gain_error_db": 0.736914619638749, "reference_mean_power": 3.967785924701116e-05, "output_mean_power": 4.70153522660528e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.9337568551130464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.8381327852746327e-05, "output_power": 2.117932755570233e-05, "reference_active": true, "delta_db": 0.6153528585027528}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.503290667812924e-06, "output_power": 5.033537020929118e-06, "reference_active": true, "delta_db": 0.4834328744532968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6513948696073343e-07, "output_power": 2.2415731727517276e-07, "reference_active": true, "delta_db": 1.327019892787521}], "active_band_count": 3, "active_band_mae_db": 0.8086018752478568}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9389479027599792, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7ecb63a25cc07ff80fbd506b544893224232c736b8851206d8516953524f92a5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9898280214447631, "standard_deviation": null, "minimum": 0.9898280214447631, "maximum": 0.9898280214447631, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4563805928617344, "standard_deviation": null, "minimum": 1.4563805928617344, "maximum": 1.4563805928617344, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.4796762118388473, "standard_deviation": null, "minimum": 1.4796762118388473, "maximum": 1.4796762118388473, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20667883807096088, "standard_deviation": null, "minimum": 0.20667883807096088, "maximum": 0.20667883807096088, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9939432216754446, "standard_deviation": null, "minimum": 0.9939432216754446, "maximum": 0.9939432216754446, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1601993758617637, "standard_deviation": null, "minimum": 1.1601993758617637, "maximum": 1.1601993758617637, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.5734862769905094, "standard_deviation": null, "minimum": 1.5734862769905094, "maximum": 1.5734862769905094, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18846031239124875, "standard_deviation": null, "minimum": 0.18846031239124875, "maximum": 0.18846031239124875, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9642517037102821, "standard_deviation": null, "minimum": 0.9642517037102821, "maximum": 0.9642517037102821, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6043471233270468, "standard_deviation": null, "minimum": 1.6043471233270468, "maximum": 1.6043471233270468, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6720824947966041, "standard_deviation": null, "minimum": 0.6720824947966041, "maximum": 0.6720824947966041, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2569740070467292, "standard_deviation": null, "minimum": 0.2569740070467292, "maximum": 0.2569740070467292, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9834787858304187, "standard_deviation": null, "minimum": 0.9834787858304187, "maximum": 0.9834787858304187, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1952909107141028, "standard_deviation": null, "minimum": 1.1952909107141028, "maximum": 1.1952909107141028, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8086018752478568, "standard_deviation": null, "minimum": 0.8086018752478568, "maximum": 0.8086018752478568, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18096193808845518, "standard_deviation": null, "minimum": 0.18096193808845518, "maximum": 0.18096193808845518, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9873001970849293, "standard_deviation": null, "minimum": 0.9873001970849293, "maximum": 0.9873001970849293, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6589027137822485, "standard_deviation": null, "minimum": 1.6589027137822485, "maximum": 1.6589027137822485, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.9987004924432652, "standard_deviation": null, "minimum": 1.9987004924432652, "maximum": 1.9987004924432652, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.258029801343065, "standard_deviation": null, "minimum": 0.258029801343065, "maximum": 0.258029801343065, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9912615181568517, "standard_deviation": null, "minimum": 0.9912615181568517, "maximum": 0.9912615181568517, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4747479287031744, "standard_deviation": null, "minimum": 1.4747479287031744, "maximum": 1.4747479287031744, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7753361909088803, "standard_deviation": null, "minimum": 1.7753361909088803, "maximum": 1.7753361909088803, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.24426500246995672, "standard_deviation": null, "minimum": 0.24426500246995672, "maximum": 0.24426500246995672, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.0095639657194964, "standard_deviation": null, "minimum": 1.0095639657194964, "maximum": 1.0095639657194964, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.496919158859445, "standard_deviation": null, "minimum": 5.496919158859445, "maximum": 5.496919158859445, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7581434197190311, "standard_deviation": null, "minimum": 0.7581434197190311, "maximum": 0.7581434197190311, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.4401212491749584, "standard_deviation": null, "minimum": 1.4401212491749584, "maximum": 1.4401212491749584, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9888937384421971, "standard_deviation": null, "minimum": 0.9888937384421971, "maximum": 0.9888937384421971, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5234645103312605, "standard_deviation": null, "minimum": 1.5234645103312605, "maximum": 1.5234645103312605, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.6615463754522724, "standard_deviation": null, "minimum": 1.6615463754522724, "maximum": 1.6615463754522724, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.21622614191162734, "standard_deviation": null, "minimum": 0.21622614191162734, "maximum": 0.21622614191162734, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9930017925456827, "standard_deviation": null, "minimum": 0.9930017925456827, "maximum": 0.9930017925456827, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2565184063438237, "standard_deviation": null, "minimum": 1.2565184063438237, "maximum": 1.2565184063438237, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.6246222517496685, "standard_deviation": null, "minimum": 1.6246222517496685, "maximum": 1.6246222517496685, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.20250363972421825, "standard_deviation": null, "minimum": 0.20250363972421825, "maximum": 0.20250363972421825, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_4992"}, {"case": "librispeech_5105-28240-0011", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3335225582122803, "predicted_mos": 3.3335225582122803, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "5536023a7f920fcb628566fb06d372fed76aa65112dd7469a75b05d788e05220", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.222782500088215}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.157550666962975, "全帯域_10ms_rms_envelope_nrmse": 0.1731266430247638, "全帯域_10ms_envelope_correlation": 0.9773820384450446, "全帯域_10ms_modulation_mae_db": 0.8389160914358583, "全帯域_40ms_log_rms_mae_db": 2.6592694429353396, "全帯域_40ms_rms_envelope_nrmse": 0.08472505452104767, "全帯域_40ms_envelope_correlation": 0.9972188286868557, "全帯域_40ms_modulation_mae_db": 0.776143112204999, "300–1000 Hz_10ms_log_rms_mae_db": 2.3381238367923656, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18450732676922046, "300–1000 Hz_10ms_envelope_correlation": 0.9798330867600404, "300–1000 Hz_10ms_modulation_mae_db": 0.5347862526059289, "300–1000 Hz_40ms_log_rms_mae_db": 1.764908301139674, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.09768449092235579, "300–1000 Hz_40ms_envelope_correlation": 0.9969409950584233, "300–1000 Hz_40ms_modulation_mae_db": 0.48276385778498626, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6603251708813276, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2079433665725481, "1000–3000 Hz_10ms_envelope_correlation": 0.9757159690721923, "1000–3000 Hz_10ms_modulation_mae_db": 0.5566946251781071, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9153081906088874, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11818700256041784, "1000–3000 Hz_40ms_envelope_correlation": 0.9957679898139686, "1000–3000 Hz_40ms_modulation_mae_db": 0.6781679948752631, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5367807859792073, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21999154005477675, "3000–6000 Hz_10ms_envelope_correlation": 0.9621100074917744, "3000–6000 Hz_10ms_modulation_mae_db": 0.5317729470523368, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8386230661044621, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0990464602194822, "3000–6000 Hz_40ms_envelope_correlation": 0.9922850809172743, "3000–6000 Hz_40ms_modulation_mae_db": 0.6368455490983101, "mr_stft_db": 5.9950406637723015, "band_mae_db": 0.47530491801701436, "rms_gain_error_db": 0.5175770915517757, "output_peak": 0.7346024154981438}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.937084327864355, "spectral_convergence": 0.7399000176257152, "spectral_convergence_status": "ok", "amplitude_floor": 1.4461531006911672e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.316641174280327, "spectral_convergence": 0.2619720070583402, "spectral_convergence_status": "ok", "amplitude_floor": 7.398993822544586e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.731396489172223, "spectral_convergence": 0.3063947392206373, "spectral_convergence_status": "ok", "amplitude_floor": 6.518159095373565e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005975557583734567, "output_power": 0.0005980238486264864, "delta_db": 0.0034006773471272206, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005550914808547427, "output_power": 0.0006389876058428094, "delta_db": 0.6112787228069921, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0010492685783192576, "output_power": 0.0011840234917586902, "delta_db": 0.5247365158641724, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007612304608209731, "output_power": 0.000922066129686525, "delta_db": 0.8324591098728228, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00020798712521433106, "output_power": 0.00022954670690065833, "delta_db": 0.42834614548499805, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.709473434650137e-06, "output_power": 9.663906790996678e-06, "delta_db": 0.4516083367259737, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.47530491801701436, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.5175770915517757, "pitch": null, "output_rms_dbfs": -24.4583662976925, "output_peak": 0.7346024154981438, "reference_rms_dbfs": -24.975943389244275, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.9950406637723015}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002040524003006053, "log_rms_mae_db": 3.157550666962975, "rms_envelope_nrmse": 0.1731266430247638, "envelope_correlation": 0.9773820384450446, "rms_gain_error_db": 0.5175784952933878, "reference_mean_power": 0.0031824932955688964, "output_mean_power": 0.0035852987272351816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.3108768872499092e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006941576008860869, "output_power": 0.0007638098983745555, "reference_active": true, "delta_db": 0.41527198729347803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.971136227750714e-05, "output_power": 7.470220390786373e-05, "reference_active": true, "delta_db": 0.3002984514723507}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2294065042747786e-05, "output_power": 3.375258560350307e-05, "reference_active": true, "delta_db": 1.8011778355417463}], "active_band_count": 3, "active_band_mae_db": 0.8389160914358583}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019294226027584624, "log_rms_mae_db": 2.6592694429353396, "rms_envelope_nrmse": 0.08472505452104767, "envelope_correlation": 0.9972188286868557, "rms_gain_error_db": 0.5176068613655955, "reference_mean_power": 0.0031984676882902833, "output_mean_power": 0.003603318520170494}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.195649150719305e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000632939746621434, "output_power": 0.0007029171859549899, "reference_active": true, "delta_db": 0.455417928620716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.313010835721604e-05, "output_power": 3.5792363753933766e-05, "reference_active": true, "delta_db": 0.3356752415046963}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3632381122717001e-06, "output_power": 1.942250920213888e-06, "reference_active": true, "delta_db": 1.537336166489585}], "active_band_count": 3, "active_band_mae_db": 0.776143112204999}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015457760673902765, "log_rms_mae_db": 2.3381238367923656, "rms_envelope_nrmse": 0.18450732676922046, "envelope_correlation": 0.9798330867600404, "rms_gain_error_db": 0.6106081958843746, "reference_mean_power": 0.0011955559481795221, "output_mean_power": 0.0013760389403045317}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2205605502187216e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134156506493551, "output_power": 0.00044478498798286456, "reference_active": true, "delta_db": 0.31763207210764544}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.135405678849972e-05, "output_power": 4.5254299980897006e-05, "reference_active": true, "delta_db": 0.3914173156995099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1249926033986765e-05, "output_power": 1.3825492280573338e-05, "reference_active": true, "delta_db": 0.8953093700106313}], "active_band_count": 3, "active_band_mae_db": 0.5347862526059289}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014615313182534672, "log_rms_mae_db": 1.764908301139674, "rms_envelope_nrmse": 0.09768449092235579, "envelope_correlation": 0.9969409950584233, "rms_gain_error_db": 0.6106055747496169, "reference_mean_power": 0.0012015673286161978, "output_mean_power": 0.001382956973447173}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.16593985496473e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003798908177653278, "output_power": 0.00041087258080469807, "reference_active": true, "delta_db": 0.34048363523869574}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8446225039409073e-05, "output_power": 2.0073528343201816e-05, "reference_active": true, "delta_db": 0.3671621293203383}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.60382048073901e-07, "output_power": 9.017713553830079e-07, "reference_active": true, "delta_db": 0.7406458087959249}], "active_band_count": 3, "active_band_mae_db": 0.48276385778498626}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011311095186173674, "log_rms_mae_db": 1.6603251708813276, "rms_envelope_nrmse": 0.2079433665725481, "envelope_correlation": 0.9757159690721923, "rms_gain_error_db": 0.7526595707004378, "reference_mean_power": 0.0008539373095994977, "output_mean_power": 0.0010155281021882454}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.101259078601872e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000229953654454443, "output_power": 0.0002777647964972753, "reference_active": true, "delta_db": 0.8203688721260739}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8372953922112745e-05, "output_power": 6.70282315214182e-05, "reference_active": true, "delta_db": 0.6004609005037876}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3078079511035874e-05, "output_power": 1.2348623886904608e-05, "reference_active": true, "delta_db": -0.24925410290445965}], "active_band_count": 3, "active_band_mae_db": 0.5566946251781071}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.65333606126804e-05, "log_rms_mae_db": 0.9153081906088874, "rms_envelope_nrmse": 0.11818700256041784, "envelope_correlation": 0.9957679898139686, "rms_gain_error_db": 0.7526600955720654, "reference_mean_power": 0.0008582264011938477, "output_mean_power": 0.0010206289431170979}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.585136403392408e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021214290189767478, "output_power": 0.0002573535292815765, "reference_active": true, "delta_db": 0.8390162317222298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7723754282185457e-05, "output_power": 3.220124922564047e-05, "reference_active": true, "delta_db": 0.6502067917677824}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.120050210892624e-07, "output_power": 8.043945755312362e-07, "reference_active": true, "delta_db": -0.5452809611357771}], "active_band_count": 3, "active_band_mae_db": 0.6781679948752631}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.9261609894090753e-05, "log_rms_mae_db": 1.5367807859792073, "rms_envelope_nrmse": 0.21999154005477675, "envelope_correlation": 0.9621100074917744, "rms_gain_error_db": 0.2460906520958267, "reference_mean_power": 3.945170478466558e-05, "output_mean_power": 4.175176504243611e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.2228241150412126e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1970652723962295e-05, "output_power": 1.2010427687650047e-05, "reference_active": true, "delta_db": 0.014406409378574915}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3243801130743035e-06, "output_power": 3.2563286964879418e-06, "reference_active": true, "delta_db": -0.08982439162450533}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0798067724676823e-06, "output_power": 7.660149452874535e-07, "reference_active": true, "delta_db": -1.4910880401539304}], "active_band_count": 3, "active_band_mae_db": 0.5317729470523368}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.5144247671385576e-05, "log_rms_mae_db": 0.8386230661044621, "rms_envelope_nrmse": 0.0990464602194822, "envelope_correlation": 0.9922850809172743, "rms_gain_error_db": 0.24610474373240157, "reference_mean_power": 3.9649231901027684e-05, "output_mean_power": 4.196094427046432e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.976126261669901e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1283346332236661e-05, "output_power": 1.12253031996488e-05, "reference_active": true, "delta_db": -0.02239838337230681}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7283331039583737e-06, "output_power": 1.6851927475336905e-06, "reference_active": true, "delta_db": -0.1097786695734866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.035315524247478e-08, "output_power": 4.67140606528209e-08, "reference_active": true, "delta_db": -1.7783595943491373}], "active_band_count": 3, "active_band_mae_db": 0.6368455490983101}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8410229696458632, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e002b7c64ebe93dd5b55e15febb3207d7b8bf812ac0538bbeb8e06072dd3b56c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9757159690721923, "standard_deviation": null, "minimum": 0.9757159690721923, "maximum": 0.9757159690721923, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6603251708813276, "standard_deviation": null, "minimum": 1.6603251708813276, "maximum": 1.6603251708813276, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5566946251781071, "standard_deviation": null, "minimum": 0.5566946251781071, "maximum": 0.5566946251781071, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2079433665725481, "standard_deviation": null, "minimum": 0.2079433665725481, "maximum": 0.2079433665725481, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9957679898139686, "standard_deviation": null, "minimum": 0.9957679898139686, "maximum": 0.9957679898139686, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9153081906088874, "standard_deviation": null, "minimum": 0.9153081906088874, "maximum": 0.9153081906088874, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6781679948752631, "standard_deviation": null, "minimum": 0.6781679948752631, "maximum": 0.6781679948752631, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11818700256041784, "standard_deviation": null, "minimum": 0.11818700256041784, "maximum": 0.11818700256041784, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9621100074917744, "standard_deviation": null, "minimum": 0.9621100074917744, "maximum": 0.9621100074917744, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5367807859792073, "standard_deviation": null, "minimum": 1.5367807859792073, "maximum": 1.5367807859792073, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5317729470523368, "standard_deviation": null, "minimum": 0.5317729470523368, "maximum": 0.5317729470523368, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21999154005477675, "standard_deviation": null, "minimum": 0.21999154005477675, "maximum": 0.21999154005477675, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9922850809172743, "standard_deviation": null, "minimum": 0.9922850809172743, "maximum": 0.9922850809172743, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8386230661044621, "standard_deviation": null, "minimum": 0.8386230661044621, "maximum": 0.8386230661044621, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6368455490983101, "standard_deviation": null, "minimum": 0.6368455490983101, "maximum": 0.6368455490983101, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0990464602194822, "standard_deviation": null, "minimum": 0.0990464602194822, "maximum": 0.0990464602194822, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9798330867600404, "standard_deviation": null, "minimum": 0.9798330867600404, "maximum": 0.9798330867600404, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3381238367923656, "standard_deviation": null, "minimum": 2.3381238367923656, "maximum": 2.3381238367923656, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.5347862526059289, "standard_deviation": null, "minimum": 0.5347862526059289, "maximum": 0.5347862526059289, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18450732676922046, "standard_deviation": null, "minimum": 0.18450732676922046, "maximum": 0.18450732676922046, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9969409950584233, "standard_deviation": null, "minimum": 0.9969409950584233, "maximum": 0.9969409950584233, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.764908301139674, "standard_deviation": null, "minimum": 1.764908301139674, "maximum": 1.764908301139674, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.48276385778498626, "standard_deviation": null, "minimum": 0.48276385778498626, "maximum": 0.48276385778498626, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09768449092235579, "standard_deviation": null, "minimum": 0.09768449092235579, "maximum": 0.09768449092235579, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.47530491801701436, "standard_deviation": null, "minimum": 0.47530491801701436, "maximum": 0.47530491801701436, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.9950406637723015, "standard_deviation": null, "minimum": 5.9950406637723015, "maximum": 5.9950406637723015, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7346024154981438, "standard_deviation": null, "minimum": 0.7346024154981438, "maximum": 0.7346024154981438, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.5175770915517757, "standard_deviation": null, "minimum": 0.5175770915517757, "maximum": 0.5175770915517757, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9773820384450446, "standard_deviation": null, "minimum": 0.9773820384450446, "maximum": 0.9773820384450446, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.157550666962975, "standard_deviation": null, "minimum": 3.157550666962975, "maximum": 3.157550666962975, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8389160914358583, "standard_deviation": null, "minimum": 0.8389160914358583, "maximum": 0.8389160914358583, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1731266430247638, "standard_deviation": null, "minimum": 0.1731266430247638, "maximum": 0.1731266430247638, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9972188286868557, "standard_deviation": null, "minimum": 0.9972188286868557, "maximum": 0.9972188286868557, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.6592694429353396, "standard_deviation": null, "minimum": 2.6592694429353396, "maximum": 2.6592694429353396, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.776143112204999, "standard_deviation": null, "minimum": 0.776143112204999, "maximum": 0.776143112204999, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08472505452104767, "standard_deviation": null, "minimum": 0.08472505452104767, "maximum": 0.08472505452104767, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5105"}, {"case": "librispeech_5105-28241-0004", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3599960803985596, "predicted_mos": 3.3599960803985596, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "04ed42aaff9ca851782ac2bc41dbefe19ad7b117260a9e04b441298e525c44a4", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.956905700266361}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.501049141749944, "全帯域_10ms_rms_envelope_nrmse": 0.17259715006936158, "全帯域_10ms_envelope_correlation": 0.9820531427551581, "全帯域_10ms_modulation_mae_db": 0.44122710381486113, "全帯域_40ms_log_rms_mae_db": 2.8560785120873073, "全帯域_40ms_rms_envelope_nrmse": 0.11441514555974328, "全帯域_40ms_envelope_correlation": 0.9955717291806815, "全帯域_40ms_modulation_mae_db": 0.6702047721124863, "300–1000 Hz_10ms_log_rms_mae_db": 1.845279750340266, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18026068426543063, "300–1000 Hz_10ms_envelope_correlation": 0.9849853609093209, "300–1000 Hz_10ms_modulation_mae_db": 0.9808729915335563, "300–1000 Hz_40ms_log_rms_mae_db": 1.2858261267438724, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1302107423433625, "300–1000 Hz_40ms_envelope_correlation": 0.9954292926852527, "300–1000 Hz_40ms_modulation_mae_db": 1.0727155018067107, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5991445597194147, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17701884298129836, "1000–3000 Hz_10ms_envelope_correlation": 0.9837630804681579, "1000–3000 Hz_10ms_modulation_mae_db": 0.7287068638259012, "1000–3000 Hz_40ms_log_rms_mae_db": 1.031338018210178, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1088905476905698, "1000–3000 Hz_40ms_envelope_correlation": 0.996818899657202, "1000–3000 Hz_40ms_modulation_mae_db": 0.7912654346147309, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7631201408304242, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24969771071091626, "3000–6000 Hz_10ms_envelope_correlation": 0.9643281082475792, "3000–6000 Hz_10ms_modulation_mae_db": 0.6904652755516348, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0556360341261684, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1422251709813692, "3000–6000 Hz_40ms_envelope_correlation": 0.988472493756849, "3000–6000 Hz_40ms_modulation_mae_db": 0.5399496400570171, "mr_stft_db": 5.963894263834028, "band_mae_db": 0.6746301345791683, "rms_gain_error_db": 0.7370878727357424, "output_peak": 0.6832285937390005}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.943483607007766, "spectral_convergence": 0.6209668394562033, "spectral_convergence_status": "ok", "amplitude_floor": 1.7451878151186278e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.292526123253065, "spectral_convergence": 0.3213084146713685, "spectral_convergence_status": "ok", "amplitude_floor": 9.668633122862388e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.655673061241253, "spectral_convergence": 0.36732154678952783, "spectral_convergence_status": "ok", "amplitude_floor": 6.131486475721294e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.299449699297122e-05, "output_power": 6.84135388207219e-05, "delta_db": -0.8390724142169019, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007475936848096536, "output_power": 0.0009923140762426311, "delta_db": 1.2298352790145919, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001248382978500818, "output_power": 0.0014443907014958127, "delta_db": 0.6333684542225502, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.000559570006980701, "output_power": 0.000648529447733612, "delta_db": 0.6407527264110057, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00012219591005283435, "output_power": 0.00013864953470503336, "delta_db": 0.5486174632878337, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001408377861622848, "output_power": 0.00014599320298630173, "delta_db": 0.15613447032212668, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6746301345791683, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7370878727357424, "pitch": null, "output_rms_dbfs": -24.636574324544508, "output_peak": 0.6832285937390005, "reference_rms_dbfs": -25.37366219728025, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.963894263834028}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023508705421562093, "log_rms_mae_db": 3.501049141749944, "rms_envelope_nrmse": 0.17259715006936158, "envelope_correlation": 0.9820531427551581, "rms_gain_error_db": 0.737834518316163, "reference_mean_power": 0.0029034514839332195, "output_mean_power": 0.003441105797444503}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1197992634528758e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005958091728481294, "output_power": 0.000729409302048126, "reference_active": true, "delta_db": 0.8786411234512042}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.320217055945433e-05, "output_power": 9.775313835783445e-05, "reference_active": true, "delta_db": 0.2070468265114737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9969137080619498e-05, "output_power": 2.1093985075841305e-05, "reference_active": true, "delta_db": 0.23799336148190534}], "active_band_count": 3, "active_band_mae_db": 0.44122710381486113}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019484731874036573, "log_rms_mae_db": 2.8560785120873073, "rms_envelope_nrmse": 0.11441514555974328, "envelope_correlation": 0.9955717291806815, "rms_gain_error_db": 0.7391665206986457, "reference_mean_power": 0.002910052669271509, "output_mean_power": 0.003449987340463696}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0516863068209195e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005686142820757538, "output_power": 0.0006950003080376405, "reference_active": true, "delta_db": 0.8716723326623554}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3606311869244326e-05, "output_power": 4.749100542196519e-05, "reference_active": true, "delta_db": 0.37062007528630037}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3652354435919545e-06, "output_power": 1.1438665598004786e-06, "reference_active": true, "delta_db": -0.7683219083888033}], "active_band_count": 3, "active_band_mae_db": 0.6702047721124863}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020607260287182458, "log_rms_mae_db": 1.845279750340266, "rms_envelope_nrmse": 0.18026068426543063, "envelope_correlation": 0.9849853609093209, "rms_gain_error_db": 0.8534236589047444, "reference_mean_power": 0.001692101437139437, "output_mean_power": 0.002059533024391198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.505279098297708e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003919860661962002, "output_power": 0.0004794911884502671, "reference_active": true, "delta_db": 0.8751000103725906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.427449102286661e-05, "output_power": 5.4027235659397354e-05, "reference_active": true, "delta_db": 0.8645916986720013}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.188127359577471e-06, "output_power": 6.96521104303878e-06, "reference_active": true, "delta_db": -1.2029272655560772}], "active_band_count": 3, "active_band_mae_db": 0.9808729915335563}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016462000863024626, "log_rms_mae_db": 1.2858261267438724, "rms_envelope_nrmse": 0.1302107423433625, "envelope_correlation": 0.9954292926852527, "rms_gain_error_db": 0.8549473083129788, "reference_mean_power": 0.001695759255771571, "output_mean_power": 0.0020647093621539584}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.197777273221207e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003745482187229297, "output_power": 0.00045919419700181423, "reference_active": true, "delta_db": 0.8848865509935868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.969988957093567e-05, "output_power": 2.4071966227802606e-05, "reference_active": true, "delta_db": 0.870477737299038}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.294406999255848e-07, "output_power": 5.922527968197389e-07, "reference_active": true, "delta_db": -1.4627822171275078}], "active_band_count": 3, "active_band_mae_db": 1.0727155018067107}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010781730421004015, "log_rms_mae_db": 1.5991445597194147, "rms_envelope_nrmse": 0.17701884298129836, "envelope_correlation": 0.9837630804681579, "rms_gain_error_db": 0.7204093381305019, "reference_mean_power": 0.0005546344869318713, "output_mean_power": 0.0006547082358619057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0245820082367704e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001805463155146596, "output_power": 0.00020868395330160991, "reference_active": true, "delta_db": 0.62900425366608}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.244891302260328e-05, "output_power": 2.5164201906617976e-05, "reference_active": true, "delta_db": 0.4958784382737333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.880443381781143e-06, "output_power": 3.822394336396725e-06, "reference_active": true, "delta_db": -1.0612378995378902}], "active_band_count": 3, "active_band_mae_db": 0.7287068638259012}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010092628732279422, "log_rms_mae_db": 1.031338018210178, "rms_envelope_nrmse": 0.1088905476905698, "envelope_correlation": 0.996818899657202, "rms_gain_error_db": 0.7191097239659118, "reference_mean_power": 0.0005569522649053947, "output_mean_power": 0.0006572475059766223}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.880183972268188e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017516791642605334, "output_power": 0.0002018457656765336, "reference_active": true, "delta_db": 0.6156507901695377}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.918962387005773e-06, "output_power": 1.1191434339423897e-05, "reference_active": true, "delta_db": 0.5241950762279314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2471992219002067e-07, "output_power": 2.444071408211327e-07, "reference_active": true, "delta_db": -1.2339504374467234}], "active_band_count": 3, "active_band_mae_db": 0.7912654346147309}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001308702624055787, "log_rms_mae_db": 1.7631201408304242, "rms_envelope_nrmse": 0.24969771071091626, "envelope_correlation": 0.9643281082475792, "rms_gain_error_db": 0.29930796580999336, "reference_mean_power": 0.00016841175600957635, "output_mean_power": 0.00018042769493367195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.6472123519121696e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.520693965195547e-05, "output_power": 0.00010303762634406854, "reference_active": true, "delta_db": 0.34327240035469586}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1413816053513955e-05, "output_power": 2.0378631572747304e-05, "reference_active": true, "delta_db": -0.21519049999707818}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.37748433230818e-06, "output_power": 7.6185521788624635e-06, "reference_active": true, "delta_db": 1.5129329263031304}], "active_band_count": 3, "active_band_mae_db": 0.6904652755516348}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.575610534879353e-05, "log_rms_mae_db": 1.0556360341261684, "rms_envelope_nrmse": 0.1422251709813692, "envelope_correlation": 0.988472493756849, "rms_gain_error_db": 0.29920714261694714, "reference_mean_power": 0.00016921252708073642, "output_mean_power": 0.0001812813913117039}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.5742070644083465e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.764878819980257e-05, "output_power": 0.00010606718624736398, "reference_active": true, "delta_db": 0.3591418987360685}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.033680531591513e-05, "output_power": 9.949826363562953e-06, "reference_active": true, "delta_db": -0.16570834968344367}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.467040891819506e-07, "output_power": 4.4612733200698113e-07, "reference_active": true, "delta_db": 1.094998671751539}], "active_band_count": 3, "active_band_mae_db": 0.5399496400570171}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8756126549445858, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e055c6166da9e4aea2cd0e9854508ab68618f304c711c7a794f62ee053f6836f", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9837630804681579, "standard_deviation": null, "minimum": 0.9837630804681579, "maximum": 0.9837630804681579, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5991445597194147, "standard_deviation": null, "minimum": 1.5991445597194147, "maximum": 1.5991445597194147, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.7287068638259012, "standard_deviation": null, "minimum": 0.7287068638259012, "maximum": 0.7287068638259012, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17701884298129836, "standard_deviation": null, "minimum": 0.17701884298129836, "maximum": 0.17701884298129836, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.996818899657202, "standard_deviation": null, "minimum": 0.996818899657202, "maximum": 0.996818899657202, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.031338018210178, "standard_deviation": null, "minimum": 1.031338018210178, "maximum": 1.031338018210178, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7912654346147309, "standard_deviation": null, "minimum": 0.7912654346147309, "maximum": 0.7912654346147309, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1088905476905698, "standard_deviation": null, "minimum": 0.1088905476905698, "maximum": 0.1088905476905698, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9643281082475792, "standard_deviation": null, "minimum": 0.9643281082475792, "maximum": 0.9643281082475792, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7631201408304242, "standard_deviation": null, "minimum": 1.7631201408304242, "maximum": 1.7631201408304242, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6904652755516348, "standard_deviation": null, "minimum": 0.6904652755516348, "maximum": 0.6904652755516348, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24969771071091626, "standard_deviation": null, "minimum": 0.24969771071091626, "maximum": 0.24969771071091626, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.988472493756849, "standard_deviation": null, "minimum": 0.988472493756849, "maximum": 0.988472493756849, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0556360341261684, "standard_deviation": null, "minimum": 1.0556360341261684, "maximum": 1.0556360341261684, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5399496400570171, "standard_deviation": null, "minimum": 0.5399496400570171, "maximum": 0.5399496400570171, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1422251709813692, "standard_deviation": null, "minimum": 0.1422251709813692, "maximum": 0.1422251709813692, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9849853609093209, "standard_deviation": null, "minimum": 0.9849853609093209, "maximum": 0.9849853609093209, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.845279750340266, "standard_deviation": null, "minimum": 1.845279750340266, "maximum": 1.845279750340266, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9808729915335563, "standard_deviation": null, "minimum": 0.9808729915335563, "maximum": 0.9808729915335563, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18026068426543063, "standard_deviation": null, "minimum": 0.18026068426543063, "maximum": 0.18026068426543063, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9954292926852527, "standard_deviation": null, "minimum": 0.9954292926852527, "maximum": 0.9954292926852527, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2858261267438724, "standard_deviation": null, "minimum": 1.2858261267438724, "maximum": 1.2858261267438724, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0727155018067107, "standard_deviation": null, "minimum": 1.0727155018067107, "maximum": 1.0727155018067107, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1302107423433625, "standard_deviation": null, "minimum": 0.1302107423433625, "maximum": 0.1302107423433625, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6746301345791683, "standard_deviation": null, "minimum": 0.6746301345791683, "maximum": 0.6746301345791683, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.963894263834028, "standard_deviation": null, "minimum": 5.963894263834028, "maximum": 5.963894263834028, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6832285937390005, "standard_deviation": null, "minimum": 0.6832285937390005, "maximum": 0.6832285937390005, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7370878727357424, "standard_deviation": null, "minimum": 0.7370878727357424, "maximum": 0.7370878727357424, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9820531427551581, "standard_deviation": null, "minimum": 0.9820531427551581, "maximum": 0.9820531427551581, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.501049141749944, "standard_deviation": null, "minimum": 3.501049141749944, "maximum": 3.501049141749944, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.44122710381486113, "standard_deviation": null, "minimum": 0.44122710381486113, "maximum": 0.44122710381486113, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17259715006936158, "standard_deviation": null, "minimum": 0.17259715006936158, "maximum": 0.17259715006936158, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9955717291806815, "standard_deviation": null, "minimum": 0.9955717291806815, "maximum": 0.9955717291806815, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.8560785120873073, "standard_deviation": null, "minimum": 2.8560785120873073, "maximum": 2.8560785120873073, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6702047721124863, "standard_deviation": null, "minimum": 0.6702047721124863, "maximum": 0.6702047721124863, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11441514555974328, "standard_deviation": null, "minimum": 0.11441514555974328, "maximum": 0.11441514555974328, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5105"}, {"case": "librispeech_5105-28241-0019", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.79712176322937, "predicted_mos": 2.79712176322937, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 84640, "input_sample_rate": 16000, "input_seconds": 5.29, "input_float32_sha256": "cdd4631afe4836c620be21835fd24c3a80acf522f545fc2e37abc293fbff2b84", "internal_samples": 84640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.806664600037038}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 4.734158274276268, "全帯域_10ms_rms_envelope_nrmse": 0.17549019706750904, "全帯域_10ms_envelope_correlation": 0.9781104920928345, "全帯域_10ms_modulation_mae_db": 0.37908289562437525, "全帯域_40ms_log_rms_mae_db": 3.4004999950927925, "全帯域_40ms_rms_envelope_nrmse": 0.09417134488538689, "全帯域_40ms_envelope_correlation": 0.9957643642930932, "全帯域_40ms_modulation_mae_db": 0.3646024234647542, "300–1000 Hz_10ms_log_rms_mae_db": 2.0402771844626955, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1828340296621436, "300–1000 Hz_10ms_envelope_correlation": 0.9825355191819647, "300–1000 Hz_10ms_modulation_mae_db": 0.6954654058606797, "300–1000 Hz_40ms_log_rms_mae_db": 1.3899886491749016, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11332999162858351, "300–1000 Hz_40ms_envelope_correlation": 0.9961212874834674, "300–1000 Hz_40ms_modulation_mae_db": 0.7624743559747978, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8975990678609778, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18520028130595892, "1000–3000 Hz_10ms_envelope_correlation": 0.9794215131741008, "1000–3000 Hz_10ms_modulation_mae_db": 0.45543594833007406, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1894527087051798, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08577362157154772, "1000–3000 Hz_40ms_envelope_correlation": 0.9969124985989214, "1000–3000 Hz_40ms_modulation_mae_db": 0.5753936793814652, "3000–6000 Hz_10ms_log_rms_mae_db": 2.2168819719509116, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23576926480062513, "3000–6000 Hz_10ms_envelope_correlation": 0.9647800332703267, "3000–6000 Hz_10ms_modulation_mae_db": 0.4285305611399055, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3655711643740251, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1350133316900807, "3000–6000 Hz_40ms_envelope_correlation": 0.9884869666407731, "3000–6000 Hz_40ms_modulation_mae_db": 0.560563534284243, "mr_stft_db": 6.261905608122515, "band_mae_db": 0.570848932258247, "rms_gain_error_db": 0.4914708067936346, "output_peak": 0.6772026241035408}, "full_scale": {"sample_count": 84640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 84640, "output_samples": 84640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.533532405894333, "spectral_convergence": 0.7177258418245449, "spectral_convergence_status": "ok", "amplitude_floor": 1.2121984197282631e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.551548798498676, "spectral_convergence": 0.2929370394297837, "spectral_convergence_status": "ok", "amplitude_floor": 7.081048851832571e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.700635619974534, "spectral_convergence": 0.32319856142975323, "spectral_convergence_status": "ok", "amplitude_floor": 5.009180364479501e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.291800237989852e-05, "output_power": 5.9867269385900585e-05, "delta_db": -0.8564531252817921, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00025616304806087414, "output_power": 0.0002978796292977188, "delta_db": 0.6552432246856881, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038484147909669214, "output_power": 0.0004537145008021395, "delta_db": 0.7150078447920774, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002720810642199278, "output_power": 0.0003028368297993574, "delta_db": 0.4651037356050561, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.5951933766776914e-05, "output_power": 4.770799134136681e-05, "delta_db": 0.16287339508201015, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.6368160234296276e-05, "output_power": 2.312270307585866e-05, "delta_db": -0.570412268102858, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.570848932258247, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.4914708067936346, "pitch": null, "output_rms_dbfs": -29.26234402559419, "output_peak": 0.6772026241035408, "reference_rms_dbfs": -29.753814832387818, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.261905608122515}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 84640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00018085971712132593, "log_rms_mae_db": 4.734158274276268, "rms_envelope_nrmse": 0.17549019706750904, "envelope_correlation": 0.9781104920928345, "rms_gain_error_db": 0.49150983787523883, "reference_mean_power": 0.0010593154039372696, "output_mean_power": 0.0011862501254660833}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 5.491358535179395e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002450078451267542, "output_power": 0.00028738755924228134, "reference_active": true, "delta_db": 0.6928797332122019}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.385267752666162e-05, "output_power": 8.372372202271032e-05, "reference_active": true, "delta_db": -0.006684076634020027}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4530628910939782e-05, "output_power": 1.3137596649412404e-05, "reference_active": true, "delta_db": -0.4376848770269038}], "active_band_count": 3, "active_band_mae_db": 0.37908289562437525}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00014864066644098504, "log_rms_mae_db": 3.4004999950927925, "rms_envelope_nrmse": 0.09417134488538689, "envelope_correlation": 0.9957643642930932, "rms_gain_error_db": 0.49175367499186057, "reference_mean_power": 0.0010653029350414017, "output_mean_power": 0.0011930221062477367}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 4.840611714703517e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022860940218938906, "output_power": 0.00026403097852599895, "reference_active": true, "delta_db": 0.6256079727505843}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8392485325428115e-05, "output_power": 3.858372886046002e-05, "reference_active": true, "delta_db": 0.021579700620654875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1188041388515455e-06, "output_power": 1.0094669498143041e-06, "reference_active": true, "delta_db": -0.44661959702302345}], "active_band_count": 3, "active_band_mae_db": 0.3646024234647542}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00013930286763496605, "log_rms_mae_db": 2.0402771844626955, "rms_envelope_nrmse": 0.1828340296621436, "envelope_correlation": 0.9825355191819647, "rms_gain_error_db": 0.7238700255485093, "reference_mean_power": 0.0005230599489297363, "output_mean_power": 0.0006179288464345498}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 3.2510596296193775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001486768204336917, "output_power": 0.00017584219918199024, "reference_active": true, "delta_db": 0.7287984183891133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.417268122003869e-05, "output_power": 5.564928173723626e-05, "reference_active": true, "delta_db": 0.11679232218622689}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.810050335491182e-06, "output_power": 5.869124000331746e-06, "reference_active": true, "delta_db": -1.2408054770066992}], "active_band_count": 3, "active_band_mae_db": 0.6954654058606797}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 0.00011121771745577188, "log_rms_mae_db": 1.3899886491749016, "rms_envelope_nrmse": 0.11332999162858351, "envelope_correlation": 0.9961212874834674, "rms_gain_error_db": 0.7238699542301057, "reference_mean_power": 0.0005260459982805144, "output_mean_power": 0.0006214564739971019}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.9069199937213013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014215463322126283, "output_power": 0.00016581752572611313, "reference_active": true, "delta_db": 0.668694113179769}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2639933776267858e-05, "output_power": 2.3491667590707954e-05, "reference_active": true, "delta_db": 0.16038694737098108}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.874281484444505e-07, "output_power": 4.913524328029635e-07, "reference_active": true, "delta_db": -1.4583420073736433}], "active_band_count": 3, "active_band_mae_db": 0.7624743559747978}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 0.00010000130422823669, "log_rms_mae_db": 1.8975990678609778, "rms_envelope_nrmse": 0.18520028130595892, "envelope_correlation": 0.9794215131741008, "rms_gain_error_db": 0.46101217574760733, "reference_mean_power": 0.0002857956004291321, "output_mean_power": 0.00031780209550107493}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.149273337334192e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04716085544297e-05, "output_power": 0.00010023699279082097, "reference_active": true, "delta_db": 0.44515716922908255}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.048324869712114e-05, "output_power": 2.0260581305411477e-05, "reference_active": true, "delta_db": -0.04746936273344465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.217073445558843e-06, "output_power": 6.719686894798035e-06, "reference_active": true, "delta_db": -0.8736813130276949}], "active_band_count": 3, "active_band_mae_db": 0.45543594833007406}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 8.407177238322724e-05, "log_rms_mae_db": 1.1894527087051798, "rms_envelope_nrmse": 0.08577362157154772, "envelope_correlation": 0.9969124985989214, "rms_gain_error_db": 0.46101234298514204, "reference_mean_power": 0.00028742710472676696, "output_mean_power": 0.0003196163256758131}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 1.945560884297879e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.432816729085121e-05, "output_power": 9.201683485700801e-05, "reference_active": true, "delta_db": 0.3789462868849296}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0296492874768435e-05, "output_power": 9.845119794710027e-06, "reference_active": true, "delta_db": -0.1946831832663913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.97290149260778e-07, "output_power": 4.5806825507211664e-07, "reference_active": true, "delta_db": -1.1525515679930747}], "active_band_count": 3, "active_band_mae_db": 0.5753936793814652}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1057, "rms_floor": 5.528232384026309e-05, "log_rms_mae_db": 2.2168819719509116, "rms_envelope_nrmse": 0.23576926480062513, "envelope_correlation": 0.9647800332703267, "rms_gain_error_db": -0.23916988091117394, "reference_mean_power": 3.470766577035859e-05, "output_mean_power": 3.28479611606251e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1057, "frequency_resolution_hz": 0.1892147587511826, "power_floor": 2.6534989625639428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5332454345272327e-05, "output_power": 1.4104894901172266e-05, "reference_active": true, "delta_db": -0.36241825708208886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.828360450157823e-06, "output_power": 3.579684446284886e-06, "reference_active": true, "delta_db": -0.2916807626935396}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1450631921352025e-06, "output_power": 9.901025709074061e-07, "reference_active": true, "delta_db": -0.6314926636440882}], "active_band_count": 3, "active_band_mae_db": 0.4285305611399055}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 84640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1051, "rms_floor": 4.431598350735027e-05, "log_rms_mae_db": 1.3655711643740251, "rms_envelope_nrmse": 0.1350133316900807, "envelope_correlation": 0.9884869666407731, "rms_gain_error_db": -0.23916987201340828, "reference_mean_power": 3.49057787168493e-05, "output_mean_power": 3.303545889567609e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1051, "frequency_resolution_hz": 0.19029495718363465, "power_floor": 2.464962908732827e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.55740476492126e-05, "output_power": 1.4209755323853799e-05, "reference_active": true, "delta_db": -0.39814899172661977}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.044495719000328e-06, "output_power": 1.8703825232961824e-06, "reference_active": true, "delta_db": -0.38655769691674774}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.011419545771505e-07, "output_power": 8.226838100766238e-08, "reference_active": true, "delta_db": -0.8969839142093614}], "active_band_count": 3, "active_band_mae_db": 0.560563534284243}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9030893983166551, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7157859bd7fc1fa131de31b028e6100030b7814bf8f7d2f343fcc982fba7cae0", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9794215131741008, "standard_deviation": null, "minimum": 0.9794215131741008, "maximum": 0.9794215131741008, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8975990678609778, "standard_deviation": null, "minimum": 1.8975990678609778, "maximum": 1.8975990678609778, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.45543594833007406, "standard_deviation": null, "minimum": 0.45543594833007406, "maximum": 0.45543594833007406, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18520028130595892, "standard_deviation": null, "minimum": 0.18520028130595892, "maximum": 0.18520028130595892, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9969124985989214, "standard_deviation": null, "minimum": 0.9969124985989214, "maximum": 0.9969124985989214, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1894527087051798, "standard_deviation": null, "minimum": 1.1894527087051798, "maximum": 1.1894527087051798, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5753936793814652, "standard_deviation": null, "minimum": 0.5753936793814652, "maximum": 0.5753936793814652, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08577362157154772, "standard_deviation": null, "minimum": 0.08577362157154772, "maximum": 0.08577362157154772, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9647800332703267, "standard_deviation": null, "minimum": 0.9647800332703267, "maximum": 0.9647800332703267, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.2168819719509116, "standard_deviation": null, "minimum": 2.2168819719509116, "maximum": 2.2168819719509116, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4285305611399055, "standard_deviation": null, "minimum": 0.4285305611399055, "maximum": 0.4285305611399055, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23576926480062513, "standard_deviation": null, "minimum": 0.23576926480062513, "maximum": 0.23576926480062513, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9884869666407731, "standard_deviation": null, "minimum": 0.9884869666407731, "maximum": 0.9884869666407731, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3655711643740251, "standard_deviation": null, "minimum": 1.3655711643740251, "maximum": 1.3655711643740251, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.560563534284243, "standard_deviation": null, "minimum": 0.560563534284243, "maximum": 0.560563534284243, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1350133316900807, "standard_deviation": null, "minimum": 0.1350133316900807, "maximum": 0.1350133316900807, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9825355191819647, "standard_deviation": null, "minimum": 0.9825355191819647, "maximum": 0.9825355191819647, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0402771844626955, "standard_deviation": null, "minimum": 2.0402771844626955, "maximum": 2.0402771844626955, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6954654058606797, "standard_deviation": null, "minimum": 0.6954654058606797, "maximum": 0.6954654058606797, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1828340296621436, "standard_deviation": null, "minimum": 0.1828340296621436, "maximum": 0.1828340296621436, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9961212874834674, "standard_deviation": null, "minimum": 0.9961212874834674, "maximum": 0.9961212874834674, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3899886491749016, "standard_deviation": null, "minimum": 1.3899886491749016, "maximum": 1.3899886491749016, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7624743559747978, "standard_deviation": null, "minimum": 0.7624743559747978, "maximum": 0.7624743559747978, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11332999162858351, "standard_deviation": null, "minimum": 0.11332999162858351, "maximum": 0.11332999162858351, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.570848932258247, "standard_deviation": null, "minimum": 0.570848932258247, "maximum": 0.570848932258247, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.261905608122515, "standard_deviation": null, "minimum": 6.261905608122515, "maximum": 6.261905608122515, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6772026241035408, "standard_deviation": null, "minimum": 0.6772026241035408, "maximum": 0.6772026241035408, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.4914708067936346, "standard_deviation": null, "minimum": 0.4914708067936346, "maximum": 0.4914708067936346, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9781104920928345, "standard_deviation": null, "minimum": 0.9781104920928345, "maximum": 0.9781104920928345, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 4.734158274276268, "standard_deviation": null, "minimum": 4.734158274276268, "maximum": 4.734158274276268, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.37908289562437525, "standard_deviation": null, "minimum": 0.37908289562437525, "maximum": 0.37908289562437525, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17549019706750904, "standard_deviation": null, "minimum": 0.17549019706750904, "maximum": 0.17549019706750904, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9957643642930932, "standard_deviation": null, "minimum": 0.9957643642930932, "maximum": 0.9957643642930932, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 3.4004999950927925, "standard_deviation": null, "minimum": 3.4004999950927925, "maximum": 3.4004999950927925, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3646024234647542, "standard_deviation": null, "minimum": 0.3646024234647542, "maximum": 0.3646024234647542, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09417134488538689, "standard_deviation": null, "minimum": 0.09417134488538689, "maximum": 0.09417134488538689, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5105"}, {"case": "librispeech_5142-33396-0019", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.351886034011841, "predicted_mos": 2.351886034011841, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 79760, "input_sample_rate": 16000, "input_seconds": 4.985, "input_float32_sha256": "f83d2435d1daa0456cffba794605d2a5ca5db1cd79743079b5541ee9d6aa5926", "internal_samples": 79760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.380704999901354}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4686633316780033, "全帯域_10ms_rms_envelope_nrmse": 0.1850876296955398, "全帯域_10ms_envelope_correlation": 0.9843069234068695, "全帯域_10ms_modulation_mae_db": 1.2059792771190132, "全帯域_40ms_log_rms_mae_db": 1.0839060198802764, "全帯域_40ms_rms_envelope_nrmse": 0.16338027527549437, "全帯域_40ms_envelope_correlation": 0.9911956218725175, "全帯域_40ms_modulation_mae_db": 1.0211317877708632, "300–1000 Hz_10ms_log_rms_mae_db": 1.838658657440657, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.254691086932188, "300–1000 Hz_10ms_envelope_correlation": 0.98214629714914, "300–1000 Hz_10ms_modulation_mae_db": 1.8534493149241165, "300–1000 Hz_40ms_log_rms_mae_db": 1.535681548581083, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2356336698011448, "300–1000 Hz_40ms_envelope_correlation": 0.9881469187091526, "300–1000 Hz_40ms_modulation_mae_db": 1.798435058517553, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4174194956056456, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.145069725627311, "1000–3000 Hz_10ms_envelope_correlation": 0.9878925868538195, "1000–3000 Hz_10ms_modulation_mae_db": 0.9508346635047255, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0990253384981374, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11139059937713669, "1000–3000 Hz_40ms_envelope_correlation": 0.9947333000652118, "1000–3000 Hz_40ms_modulation_mae_db": 0.7944714867251049, "3000–6000 Hz_10ms_log_rms_mae_db": 1.476111868213913, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1523983805489182, "3000–6000 Hz_10ms_envelope_correlation": 0.9803931658288447, "3000–6000 Hz_10ms_modulation_mae_db": 0.6025318582656366, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9219403000650428, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.0869575774755947, "3000–6000 Hz_40ms_envelope_correlation": 0.9940416344995854, "3000–6000 Hz_40ms_modulation_mae_db": 0.26156598854339524, "mr_stft_db": 6.1246913955958355, "band_mae_db": 0.8363711782836719, "rms_gain_error_db": 1.0946609047572298, "output_peak": 0.561937832494117}, "full_scale": {"sample_count": 79760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 79760, "output_samples": 79760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.651895614860863, "spectral_convergence": 0.6060213511278199, "spectral_convergence_status": "ok", "amplitude_floor": 8.669731759960068e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.421605712285835, "spectral_convergence": 0.32337578923169213, "spectral_convergence_status": "ok", "amplitude_floor": 7.258845275738271e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.300572859640809, "spectral_convergence": 0.38809542001999237, "spectral_convergence_status": "ok", "amplitude_floor": 5.316334054535358e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.354150869521306e-05, "output_power": 6.555655751261562e-05, "delta_db": 0.8792553535418628, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0007507214796955326, "output_power": 0.0009813820292078781, "delta_db": 1.1635925893084125, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006075331721980808, "output_power": 0.0008706753199086545, "delta_db": 1.5628623812769415, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00029803976258038344, "output_power": 0.00034435922159601283, "delta_db": 0.6273750868448109, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0002296848467085225, "output_power": 0.00026014577690071345, "delta_db": 0.5408443625560406, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011626570661661749, "output_power": 0.00012299327477656593, "delta_db": 0.24429729617396226, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8363711782836719, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0946609047572298, "pitch": null, "output_rms_dbfs": -25.775559046951464, "output_peak": 0.561937832494117, "reference_rms_dbfs": -26.870219951708698, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.1246913955958355}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 79760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.000134521265514355, "log_rms_mae_db": 1.4686633316780033, "rms_envelope_nrmse": 0.1850876296955398, "envelope_correlation": 0.9843069234068695, "rms_gain_error_db": 1.094662639124471, "reference_mean_power": 0.0020578494395295825, "output_mean_power": 0.002647767583315344}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 5.569199266163003e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00039409678072813294, "output_power": 0.0004923235435605266, "reference_active": true, "delta_db": 0.9664771755981354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.428667833313107e-05, "output_power": 6.791099145015572e-05, "reference_active": true, "delta_db": 0.9724680182156047}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.373716545850661e-06, "output_power": 1.0853897165395415e-05, "reference_active": true, "delta_db": 1.6789926375432995}], "active_band_count": 3, "active_band_mae_db": 1.2059792771190132}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00012498981593367327, "log_rms_mae_db": 1.0839060198802764, "rms_envelope_nrmse": 0.16338027527549437, "envelope_correlation": 0.9911956218725175, "rms_gain_error_db": 1.0946661702872327, "reference_mean_power": 0.0020703160518770356, "output_mean_power": 0.002663810131662141}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 4.869724313408462e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003604621899100367, "output_power": 0.00045490850919600325, "reference_active": true, "delta_db": 1.0106434367317827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.585585870556879e-05, "output_power": 3.281572012376719e-05, "reference_active": true, "delta_db": 1.0352297324888007}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.932164327326652e-07, "output_power": 4.970314597013461e-07, "reference_active": true, "delta_db": 1.017522194092006}], "active_band_count": 3, "active_band_mae_db": 1.0211317877708632}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 0.00011089795767449635, "log_rms_mae_db": 1.838658657440657, "rms_envelope_nrmse": 0.254691086932188, "envelope_correlation": 0.98214629714914, "rms_gain_error_db": 1.5783897446846438, "reference_mean_power": 0.001090370195617981, "output_mean_power": 0.0015682415146885038}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 3.4652335334077796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023090422993351253, "output_power": 0.00033142336036885955, "reference_active": true, "delta_db": 1.5695122757132771}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.174073740387618e-05, "output_power": 5.967333547502682e-05, "reference_active": true, "delta_db": 1.552201964911645}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.679949697911218e-06, "output_power": 8.205490661070977e-06, "reference_active": true, "delta_db": 2.438633704147427}], "active_band_count": 3, "active_band_mae_db": 1.8534493149241165}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 0.00010160775009558359, "log_rms_mae_db": 1.535681548581083, "rms_envelope_nrmse": 0.2356336698011448, "envelope_correlation": 0.9881469187091526, "rms_gain_error_db": 1.5783896029343902, "reference_mean_power": 0.0010969783403958974, "output_mean_power": 0.0015777457279065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 3.031226473561925e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002111816957536322, "output_power": 0.0003082418561404381, "reference_active": true, "delta_db": 1.6423533833562791}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.975380679045621e-05, "output_power": 2.797737118012687e-05, "reference_active": true, "delta_db": 1.5115610306494245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.773024908316542e-07, "output_power": 4.646145934081558e-07, "reference_active": true, "delta_db": 2.241390761546955}], "active_band_count": 3, "active_band_mae_db": 1.798435058517553}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 7.327188751847531e-05, "log_rms_mae_db": 1.4174194956056456, "rms_envelope_nrmse": 0.145069725627311, "envelope_correlation": 0.9878925868538195, "rms_gain_error_db": 0.673933051052134, "reference_mean_power": 0.0003949163505892704, "output_mean_power": 0.0004612096867417923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 2.1879792460690968e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000156993183862756, "output_power": 0.0001689912059071144, "reference_active": true, "delta_db": 0.31983307901061736}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3083527056032412e-05, "output_power": 1.605016967723782e-05, "reference_active": true, "delta_db": 0.8875479114812463}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5497810931869906e-06, "output_power": 3.724047369062996e-06, "reference_active": true, "delta_db": 1.6451230000223127}], "active_band_count": 3, "active_band_mae_db": 0.9508346635047255}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 7.131903461227575e-05, "log_rms_mae_db": 1.0990253384981374, "rms_envelope_nrmse": 0.11139059937713669, "envelope_correlation": 0.9947333000652118, "rms_gain_error_db": 0.6739426836008287, "reference_mean_power": 0.0003973070610776639, "output_mean_power": 0.00046400274724780035}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 1.9730997112139496e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014395632169995676, "output_power": 0.00015558560760962765, "reference_active": true, "delta_db": 0.3373867901078165}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.343232783051936e-06, "output_power": 7.991244375496058e-06, "reference_active": true, "delta_db": 1.0030376240424417}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5197840492846542e-07, "output_power": 1.93232817680862e-07, "reference_active": true, "delta_db": 1.0429900460250565}], "active_band_count": 3, "active_band_mae_db": 0.7944714867251049}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 996, "rms_floor": 4.2584167929876904e-05, "log_rms_mae_db": 1.476111868213913, "rms_envelope_nrmse": 0.1523983805489182, "envelope_correlation": 0.9803931658288447, "rms_gain_error_db": 0.3331031654194585, "reference_mean_power": 0.0001744638441038751, "output_mean_power": 0.00018837174197469927}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 996, "frequency_resolution_hz": 0.2008032128514056, "power_floor": 1.0543054012178789e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.162385199300134e-05, "output_power": 7.634925744410855e-05, "reference_active": true, "delta_db": 0.2774714352285417}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.778561151416546e-06, "output_power": 1.0141356987775109e-05, "reference_active": true, "delta_db": 0.15821114647710915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4459128360402158e-06, "output_power": 1.983048849221042e-06, "reference_active": true, "delta_db": 1.3719129930912588}], "active_band_count": 3, "active_band_mae_db": 0.6025318582656366}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 79760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 990, "rms_floor": 3.853974797463785e-05, "log_rms_mae_db": 0.9219403000650428, "rms_envelope_nrmse": 0.0869575774755947, "envelope_correlation": 0.9940416344995854, "rms_gain_error_db": 0.33309770857958165, "reference_mean_power": 0.00017551968398739576, "output_mean_power": 0.00018951151311807161}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 990, "frequency_resolution_hz": 0.20202020202020202, "power_floor": 9.383868144458744e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.640153736830121e-05, "output_power": 7.107486510465801e-05, "reference_active": true, "delta_db": 0.29537909570091314}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.23065423970599e-06, "output_power": 4.529873190630553e-06, "reference_active": true, "delta_db": 0.2967851150303199}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2239857829027987e-07, "output_power": 1.2794689437417271e-07, "reference_active": true, "delta_db": 0.1925337548989527}], "active_band_count": 3, "active_band_mae_db": 0.26156598854339524}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9050155381460625, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7fff807e421c34649df96521bd7a157e3cf924415d3b72c0e7dea0204cb10342", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9878925868538195, "standard_deviation": null, "minimum": 0.9878925868538195, "maximum": 0.9878925868538195, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4174194956056456, "standard_deviation": null, "minimum": 1.4174194956056456, "maximum": 1.4174194956056456, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9508346635047255, "standard_deviation": null, "minimum": 0.9508346635047255, "maximum": 0.9508346635047255, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.145069725627311, "standard_deviation": null, "minimum": 0.145069725627311, "maximum": 0.145069725627311, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9947333000652118, "standard_deviation": null, "minimum": 0.9947333000652118, "maximum": 0.9947333000652118, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0990253384981374, "standard_deviation": null, "minimum": 1.0990253384981374, "maximum": 1.0990253384981374, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7944714867251049, "standard_deviation": null, "minimum": 0.7944714867251049, "maximum": 0.7944714867251049, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11139059937713669, "standard_deviation": null, "minimum": 0.11139059937713669, "maximum": 0.11139059937713669, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9803931658288447, "standard_deviation": null, "minimum": 0.9803931658288447, "maximum": 0.9803931658288447, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.476111868213913, "standard_deviation": null, "minimum": 1.476111868213913, "maximum": 1.476111868213913, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6025318582656366, "standard_deviation": null, "minimum": 0.6025318582656366, "maximum": 0.6025318582656366, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1523983805489182, "standard_deviation": null, "minimum": 0.1523983805489182, "maximum": 0.1523983805489182, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9940416344995854, "standard_deviation": null, "minimum": 0.9940416344995854, "maximum": 0.9940416344995854, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9219403000650428, "standard_deviation": null, "minimum": 0.9219403000650428, "maximum": 0.9219403000650428, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.26156598854339524, "standard_deviation": null, "minimum": 0.26156598854339524, "maximum": 0.26156598854339524, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0869575774755947, "standard_deviation": null, "minimum": 0.0869575774755947, "maximum": 0.0869575774755947, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.98214629714914, "standard_deviation": null, "minimum": 0.98214629714914, "maximum": 0.98214629714914, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.838658657440657, "standard_deviation": null, "minimum": 1.838658657440657, "maximum": 1.838658657440657, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.8534493149241165, "standard_deviation": null, "minimum": 1.8534493149241165, "maximum": 1.8534493149241165, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.254691086932188, "standard_deviation": null, "minimum": 0.254691086932188, "maximum": 0.254691086932188, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9881469187091526, "standard_deviation": null, "minimum": 0.9881469187091526, "maximum": 0.9881469187091526, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.535681548581083, "standard_deviation": null, "minimum": 1.535681548581083, "maximum": 1.535681548581083, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.798435058517553, "standard_deviation": null, "minimum": 1.798435058517553, "maximum": 1.798435058517553, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2356336698011448, "standard_deviation": null, "minimum": 0.2356336698011448, "maximum": 0.2356336698011448, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8363711782836719, "standard_deviation": null, "minimum": 0.8363711782836719, "maximum": 0.8363711782836719, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.1246913955958355, "standard_deviation": null, "minimum": 6.1246913955958355, "maximum": 6.1246913955958355, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.561937832494117, "standard_deviation": null, "minimum": 0.561937832494117, "maximum": 0.561937832494117, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0946609047572298, "standard_deviation": null, "minimum": 1.0946609047572298, "maximum": 1.0946609047572298, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9843069234068695, "standard_deviation": null, "minimum": 0.9843069234068695, "maximum": 0.9843069234068695, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4686633316780033, "standard_deviation": null, "minimum": 1.4686633316780033, "maximum": 1.4686633316780033, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.2059792771190132, "standard_deviation": null, "minimum": 1.2059792771190132, "maximum": 1.2059792771190132, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1850876296955398, "standard_deviation": null, "minimum": 0.1850876296955398, "maximum": 0.1850876296955398, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9911956218725175, "standard_deviation": null, "minimum": 0.9911956218725175, "maximum": 0.9911956218725175, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0839060198802764, "standard_deviation": null, "minimum": 1.0839060198802764, "maximum": 1.0839060198802764, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0211317877708632, "standard_deviation": null, "minimum": 1.0211317877708632, "maximum": 1.0211317877708632, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16338027527549437, "standard_deviation": null, "minimum": 0.16338027527549437, "maximum": 0.16338027527549437, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5142"}, {"case": "librispeech_5142-33396-0030", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.637596607208252, "predicted_mos": 2.637596607208252, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44240, "input_sample_rate": 16000, "input_seconds": 2.765, "input_float32_sha256": "332ff94efc89117003129d245d95cd4285b4c2af7f1f95c21f27f0b3c2bdcc1b", "internal_samples": 44240, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.302629699930549}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6434070855192064, "全帯域_10ms_rms_envelope_nrmse": 0.15631528028042785, "全帯域_10ms_envelope_correlation": 0.9915482495289027, "全帯域_10ms_modulation_mae_db": 1.1725079512210097, "全帯域_40ms_log_rms_mae_db": 1.2960180279998503, "全帯域_40ms_rms_envelope_nrmse": 0.1289703339242118, "全帯域_40ms_envelope_correlation": 0.9974323999331167, "全帯域_40ms_modulation_mae_db": 1.431702842256296, "300–1000 Hz_10ms_log_rms_mae_db": 1.5303233214057061, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18140433260672995, "300–1000 Hz_10ms_envelope_correlation": 0.9921126937367518, "300–1000 Hz_10ms_modulation_mae_db": 1.4231583242911, "300–1000 Hz_40ms_log_rms_mae_db": 1.1693201930426218, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1606593566018349, "300–1000 Hz_40ms_envelope_correlation": 0.9965719367596417, "300–1000 Hz_40ms_modulation_mae_db": 1.7862800604403544, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3758574093212579, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15720325720739428, "1000–3000 Hz_10ms_envelope_correlation": 0.9922840077641495, "1000–3000 Hz_10ms_modulation_mae_db": 1.2627399768033893, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0438006110668185, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13694234376781764, "1000–3000 Hz_40ms_envelope_correlation": 0.9959385833949231, "1000–3000 Hz_40ms_modulation_mae_db": 1.3528569262032246, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5084821315456112, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19669958100590876, "3000–6000 Hz_10ms_envelope_correlation": 0.9827788772311571, "3000–6000 Hz_10ms_modulation_mae_db": 0.8818462455900945, "3000–6000 Hz_40ms_log_rms_mae_db": 1.015471902415211, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14243761503135016, "3000–6000 Hz_40ms_envelope_correlation": 0.9941578180485486, "3000–6000 Hz_40ms_modulation_mae_db": 1.1199915464547159, "mr_stft_db": 5.941003308899415, "band_mae_db": 0.7874550311176106, "rms_gain_error_db": 0.944541355355807, "output_peak": 0.4601111301311507}, "full_scale": {"sample_count": 44240, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44240, "output_samples": 44240, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.535497948803641, "spectral_convergence": 0.6806833821414293, "spectral_convergence_status": "ok", "amplitude_floor": 7.692579600611731e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.30312472597795, "spectral_convergence": 0.2746719614291602, "spectral_convergence_status": "ok", "amplitude_floor": 5.999960941709396e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.984387251916652, "spectral_convergence": 0.321904146559581, "spectral_convergence_status": "ok", "amplitude_floor": 5.417405012540354e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 6.282816067608343e-05, "output_power": 7.965111315846133e-05, "delta_db": 1.030375039040954, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019406375287345015, "output_power": 0.0002097646922346779, "delta_db": 0.33787963538953825, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005567857807558596, "output_power": 0.0007259818301977071, "delta_db": 1.152376156627373, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00030647737825017173, "output_power": 0.00038876300388530314, "delta_db": 1.032865053291042, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00015415279416012865, "output_power": 0.00018819983112327145, "delta_db": 0.866678282695429, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.966178555848869e-05, "output_power": 5.326941559180169e-05, "delta_db": 0.3045560196613275, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7874550311176106, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.944541355355807, "pitch": null, "output_rms_dbfs": -27.83667834044207, "output_peak": 0.4601111301311507, "reference_rms_dbfs": -28.781219695797876, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.941003308899415}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44240, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 0.00013943874660832744, "log_rms_mae_db": 1.6434070855192064, "rms_envelope_nrmse": 0.15631528028042785, "envelope_correlation": 0.9915482495289027, "rms_gain_error_db": 0.9445417058582851, "reference_mean_power": 0.0013263679977567136, "output_mean_power": 0.0016486110460349848}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 5.047819293813421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030810693936871264, "output_power": 0.000379633871205642, "reference_active": true, "delta_db": 0.9066347368127463}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2023354078518551e-05, "output_power": 1.6058678931453817e-05, "reference_active": true, "delta_db": 1.2568417820647737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2193973896229425e-06, "output_power": 4.397231405215512e-06, "reference_active": true, "delta_db": 1.3540473347855095}], "active_band_count": 3, "active_band_mae_db": 1.1725079512210097}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 0.0001327212418155818, "log_rms_mae_db": 1.2960180279998503, "rms_envelope_nrmse": 0.1289703339242118, "envelope_correlation": 0.9974323999331167, "rms_gain_error_db": 0.9445436779645645, "reference_mean_power": 0.0013409420747202037, "output_mean_power": 0.0016667266731996426}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 4.5645799716966823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002745054731183271, "output_power": 0.0003396414855791977, "reference_active": true, "delta_db": 0.9246972392038623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.376005868866601e-06, "output_power": 7.596722389310884e-06, "reference_active": true, "delta_db": 1.5016652190591018}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7970463694996416e-07, "output_power": 2.7633372429615697e-07, "reference_active": true, "delta_db": 1.8687460685059238}], "active_band_count": 3, "active_band_mae_db": 1.431702842256296}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 9.713079411954574e-05, "log_rms_mae_db": 1.5303233214057061, "rms_envelope_nrmse": 0.18140433260672995, "envelope_correlation": 0.9921126937367518, "rms_gain_error_db": 1.1606516686051078, "reference_mean_power": 0.0005849439303269558, "output_mean_power": 0.0007641513870087468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 2.471278629444775e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014231281864882344, "output_power": 0.0001819312517598932, "reference_active": true, "delta_db": 1.0666328726810386}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.153370236673122e-06, "output_power": 1.1261658500741065e-05, "reference_active": true, "delta_db": 1.4026518996571913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4861340139926787e-06, "output_power": 2.249453366871354e-06, "reference_active": true, "delta_db": 1.8001902005350698}], "active_band_count": 3, "active_band_mae_db": 1.4231583242911}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 9.335285483371196e-05, "log_rms_mae_db": 1.1693201930426218, "rms_envelope_nrmse": 0.1606593566018349, "envelope_correlation": 0.9965719367596417, "rms_gain_error_db": 1.16065130649098, "reference_mean_power": 0.0005913717973798096, "output_mean_power": 0.000772548475328042}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 2.253861652661314e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012623197394568146, "output_power": 0.000163256435519584, "reference_active": true, "delta_db": 1.1170093651792936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1401147597866345e-06, "output_power": 4.436213776729334e-06, "reference_active": true, "delta_db": 1.5006694595169576}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.001702937868675e-08, "output_power": 1.5041756706368355e-07, "reference_active": true, "delta_db": 2.7411613566248114}], "active_band_count": 3, "active_band_mae_db": 1.7862800604403544}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 7.293407135281025e-05, "log_rms_mae_db": 1.3758574093212579, "rms_envelope_nrmse": 0.15720325720739428, "envelope_correlation": 0.9922840077641495, "rms_gain_error_db": 0.9272960168941566, "reference_mean_power": 0.0003354294563993909, "output_mean_power": 0.0004152702315782267}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 1.7752008532728783e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001256086055891369, "output_power": 0.00014520956176775925, "reference_active": true, "delta_db": 0.6297582032242411}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.726051831560619e-06, "output_power": 1.2392637441536271e-05, "reference_active": true, "delta_db": 1.5234595587297373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2668523428769663e-06, "output_power": 1.8459767031334846e-06, "reference_active": true, "delta_db": 1.6350021684561902}], "active_band_count": 3, "active_band_mae_db": 1.2627399768033893}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 7.108743338455805e-05, "log_rms_mae_db": 1.0438006110668185, "rms_envelope_nrmse": 0.13694234376781764, "envelope_correlation": 0.9959385833949231, "rms_gain_error_db": 0.9272963950463062, "reference_mean_power": 0.00033911524225685186, "output_mean_power": 0.0004198333651353675}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 1.626160013872791e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011427350869395253, "output_power": 0.00013223851866817703, "reference_active": true, "delta_db": 0.6341241332153749}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.179974955820808e-06, "output_power": 6.219510352870001e-06, "reference_active": true, "delta_db": 1.7258251535184375}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.39572209062608e-08, "output_power": 1.3892880873073925e-07, "reference_active": true, "delta_db": 1.6986214918758624}], "active_band_count": 3, "active_band_mae_db": 1.3528569262032246}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 552, "rms_floor": 4.804880216814636e-05, "log_rms_mae_db": 1.5084821315456112, "rms_envelope_nrmse": 0.19669958100590876, "envelope_correlation": 0.9827788772311571, "rms_gain_error_db": 0.8672977602640892, "reference_mean_power": 0.00012431296346075482, "output_mean_power": 0.0001517910606449348}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 552, "frequency_resolution_hz": 0.3623188405797101, "power_floor": 7.893026272799499e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.194673883780046e-05, "output_power": 4.993738859311856e-05, "reference_active": true, "delta_db": 0.757276256598214}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.051190096168371e-06, "output_power": 6.4074020187836285e-06, "reference_active": true, "delta_db": -0.41580449444479534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.501810525086e-07, "output_power": 1.333681615332873e-06, "reference_active": true, "delta_db": 1.4724579857272744}], "active_band_count": 3, "active_band_mae_db": 0.8818462455900945}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44240, "uncovered_tail_samples": 0, "metrics": {"frame_count": 546, "rms_floor": 4.532286462232191e-05, "log_rms_mae_db": 1.015471902415211, "rms_envelope_nrmse": 0.14243761503135016, "envelope_correlation": 0.9941578180485486, "rms_gain_error_db": 0.8672995506465841, "reference_mean_power": 0.00012567880182247303, "output_mean_power": 0.0001534588667393239}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 546, "frequency_resolution_hz": 0.3663003663003663, "power_floor": 7.206643574889899e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.830114378682998e-05, "output_power": 4.561980340659059e-05, "reference_active": true, "delta_db": 0.7594166590564343}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6851867496780266e-06, "output_power": 3.399549777997751e-06, "reference_active": true, "delta_db": -0.3503809623507841}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.995617022655143e-08, "output_power": 1.006587622441256e-07, "reference_active": true, "delta_db": 2.250177017956929}], "active_band_count": 3, "active_band_mae_db": 1.1199915464547159}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9274910365518517, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2c43b8e55fd421c1080067b3f4c603b4dce0b7326292400fbcc2a3cbe9bc20dc", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9922840077641495, "standard_deviation": null, "minimum": 0.9922840077641495, "maximum": 0.9922840077641495, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3758574093212579, "standard_deviation": null, "minimum": 1.3758574093212579, "maximum": 1.3758574093212579, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2627399768033893, "standard_deviation": null, "minimum": 1.2627399768033893, "maximum": 1.2627399768033893, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15720325720739428, "standard_deviation": null, "minimum": 0.15720325720739428, "maximum": 0.15720325720739428, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9959385833949231, "standard_deviation": null, "minimum": 0.9959385833949231, "maximum": 0.9959385833949231, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0438006110668185, "standard_deviation": null, "minimum": 1.0438006110668185, "maximum": 1.0438006110668185, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3528569262032246, "standard_deviation": null, "minimum": 1.3528569262032246, "maximum": 1.3528569262032246, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13694234376781764, "standard_deviation": null, "minimum": 0.13694234376781764, "maximum": 0.13694234376781764, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9827788772311571, "standard_deviation": null, "minimum": 0.9827788772311571, "maximum": 0.9827788772311571, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5084821315456112, "standard_deviation": null, "minimum": 1.5084821315456112, "maximum": 1.5084821315456112, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8818462455900945, "standard_deviation": null, "minimum": 0.8818462455900945, "maximum": 0.8818462455900945, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19669958100590876, "standard_deviation": null, "minimum": 0.19669958100590876, "maximum": 0.19669958100590876, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9941578180485486, "standard_deviation": null, "minimum": 0.9941578180485486, "maximum": 0.9941578180485486, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.015471902415211, "standard_deviation": null, "minimum": 1.015471902415211, "maximum": 1.015471902415211, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1199915464547159, "standard_deviation": null, "minimum": 1.1199915464547159, "maximum": 1.1199915464547159, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14243761503135016, "standard_deviation": null, "minimum": 0.14243761503135016, "maximum": 0.14243761503135016, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9921126937367518, "standard_deviation": null, "minimum": 0.9921126937367518, "maximum": 0.9921126937367518, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.5303233214057061, "standard_deviation": null, "minimum": 1.5303233214057061, "maximum": 1.5303233214057061, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4231583242911, "standard_deviation": null, "minimum": 1.4231583242911, "maximum": 1.4231583242911, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18140433260672995, "standard_deviation": null, "minimum": 0.18140433260672995, "maximum": 0.18140433260672995, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9965719367596417, "standard_deviation": null, "minimum": 0.9965719367596417, "maximum": 0.9965719367596417, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1693201930426218, "standard_deviation": null, "minimum": 1.1693201930426218, "maximum": 1.1693201930426218, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7862800604403544, "standard_deviation": null, "minimum": 1.7862800604403544, "maximum": 1.7862800604403544, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1606593566018349, "standard_deviation": null, "minimum": 0.1606593566018349, "maximum": 0.1606593566018349, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7874550311176106, "standard_deviation": null, "minimum": 0.7874550311176106, "maximum": 0.7874550311176106, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.941003308899415, "standard_deviation": null, "minimum": 5.941003308899415, "maximum": 5.941003308899415, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4601111301311507, "standard_deviation": null, "minimum": 0.4601111301311507, "maximum": 0.4601111301311507, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.944541355355807, "standard_deviation": null, "minimum": 0.944541355355807, "maximum": 0.944541355355807, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9915482495289027, "standard_deviation": null, "minimum": 0.9915482495289027, "maximum": 0.9915482495289027, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6434070855192064, "standard_deviation": null, "minimum": 1.6434070855192064, "maximum": 1.6434070855192064, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1725079512210097, "standard_deviation": null, "minimum": 1.1725079512210097, "maximum": 1.1725079512210097, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15631528028042785, "standard_deviation": null, "minimum": 0.15631528028042785, "maximum": 0.15631528028042785, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9974323999331167, "standard_deviation": null, "minimum": 0.9974323999331167, "maximum": 0.9974323999331167, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2960180279998503, "standard_deviation": null, "minimum": 1.2960180279998503, "maximum": 1.2960180279998503, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.431702842256296, "standard_deviation": null, "minimum": 1.431702842256296, "maximum": 1.431702842256296, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1289703339242118, "standard_deviation": null, "minimum": 0.1289703339242118, "maximum": 0.1289703339242118, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5142"}, {"case": "librispeech_5142-33396-0065", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9643993377685547, "predicted_mos": 2.9643993377685547, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83120, "input_sample_rate": 16000, "input_seconds": 5.195, "input_float32_sha256": "497d5dac96d63b68424c019252c9ff6a4fcaa4e22324190ccd6c8110800bbf35", "internal_samples": 83120, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.393089800141752}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.954397311262717, "全帯域_10ms_rms_envelope_nrmse": 0.19995561363629505, "全帯域_10ms_envelope_correlation": 0.9856113335981339, "全帯域_10ms_modulation_mae_db": 1.0651784829440223, "全帯域_40ms_log_rms_mae_db": 1.5371808469607653, "全帯域_40ms_rms_envelope_nrmse": 0.16322376336185712, "全帯域_40ms_envelope_correlation": 0.9948153033228482, "全帯域_40ms_modulation_mae_db": 1.0209181045845488, "300–1000 Hz_10ms_log_rms_mae_db": 1.4763191614425781, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22194950425862775, "300–1000 Hz_10ms_envelope_correlation": 0.9870927684669716, "300–1000 Hz_10ms_modulation_mae_db": 1.905736869941352, "300–1000 Hz_40ms_log_rms_mae_db": 1.1909643613283014, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.19492593099965394, "300–1000 Hz_40ms_envelope_correlation": 0.9936328563692569, "300–1000 Hz_40ms_modulation_mae_db": 1.7583867379971962, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3936281127425727, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18583427092286042, "1000–3000 Hz_10ms_envelope_correlation": 0.9894982915966917, "1000–3000 Hz_10ms_modulation_mae_db": 0.9944846300226146, "1000–3000 Hz_40ms_log_rms_mae_db": 1.011865438064268, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.15179104306703362, "1000–3000 Hz_40ms_envelope_correlation": 0.9965686666736996, "1000–3000 Hz_40ms_modulation_mae_db": 0.7725831350266765, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6826631600767874, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.27356071043129154, "3000–6000 Hz_10ms_envelope_correlation": 0.9609329539346882, "3000–6000 Hz_10ms_modulation_mae_db": 0.9794500184513368, "3000–6000 Hz_40ms_log_rms_mae_db": 1.151238349182671, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18484842363488582, "3000–6000 Hz_40ms_envelope_correlation": 0.9837276752610551, "3000–6000 Hz_40ms_modulation_mae_db": 1.126833557703818, "mr_stft_db": 5.763391047758637, "band_mae_db": 0.7288337297014321, "rms_gain_error_db": 1.1230371129869638, "output_peak": 0.5538989575903049}, "full_scale": {"sample_count": 83120, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83120, "output_samples": 83120, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.246557050986234, "spectral_convergence": 0.5642032871467355, "spectral_convergence_status": "ok", "amplitude_floor": 8.933262458231008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.279429757494296, "spectral_convergence": 0.3347041705010311, "spectral_convergence_status": "ok", "amplitude_floor": 6.540934930277787e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.764186334795381, "spectral_convergence": 0.3797167459667251, "spectral_convergence_status": "ok", "amplitude_floor": 6.044182898288457e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.147025669043969e-05, "output_power": 3.1791628817444696e-05, "delta_db": 0.0441249390238329, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00012755476455897642, "output_power": 0.0001364940580859136, "delta_db": 0.29417060436172504, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005321662836872518, "output_power": 0.0007422638230006726, "delta_db": 1.4451093849405159, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00022563854189948138, "output_power": 0.0002927078176851591, "delta_db": 1.130210371873455, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.229072811272528e-05, "output_power": 9.948064314264127e-05, "delta_db": 0.8238767932024332, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.381343402576783e-05, "output_power": 7.386915730587024e-05, "delta_db": 0.6355102848066306, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7288337297014321, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1230371129869638, "pitch": null, "output_rms_dbfs": -28.611899860103648, "output_peak": 0.5538989575903049, "reference_rms_dbfs": -29.734936973090612, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.763391047758637}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83120, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.00015201529585382603, "log_rms_mae_db": 1.954397311262717, "rms_envelope_nrmse": 0.19995561363629505, "envelope_correlation": 0.9856113335981339, "rms_gain_error_db": 1.1230374746969405, "reference_mean_power": 0.0010639579210132953, "output_mean_power": 0.0013779333125281583}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.575807359242935e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021400202307720183, "output_power": 0.0002859281700870868, "reference_active": true, "delta_db": 1.2583906584236029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3329698566280647e-05, "output_power": 3.0520775011071736e-05, "reference_active": true, "delta_db": 1.166864299359844}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.358098056371455e-06, "output_power": 6.397918540969874e-06, "reference_active": true, "delta_db": 0.7702804910486201}], "active_band_count": 3, "active_band_mae_db": 1.0651784829440223}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.0001490763524181648, "log_rms_mae_db": 1.5371808469607653, "rms_envelope_nrmse": 0.16322376336185712, "envelope_correlation": 0.9948153033228482, "rms_gain_error_db": 1.123039319061703, "reference_mean_power": 0.0010701430709341084, "output_mean_power": 0.0013859442969653816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 5.309268989226658e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019126377610844124, "output_power": 0.00025904102337707106, "reference_active": true, "delta_db": 1.3173582137967799}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0962051121960504e-05, "output_power": 1.3427519510704103e-05, "reference_active": true, "delta_db": 0.8810396900418216}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.152513485967136e-07, "output_power": 3.846737879706653e-07, "reference_active": true, "delta_db": 0.8643564099150448}], "active_band_count": 3, "active_band_mae_db": 1.0209181045845488}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 0.0001087945365454221, "log_rms_mae_db": 1.4763191614425781, "rms_envelope_nrmse": 0.22194950425862775, "envelope_correlation": 0.9870927684669716, "rms_gain_error_db": 1.327409430048137, "reference_mean_power": 0.0005739909463294839, "output_mean_power": 0.0007791946917294211}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 3.2669273354505384e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013940368795916017, "output_power": 0.00019651263998534408, "reference_active": true, "delta_db": 1.491162267947614}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.236537072685246e-06, "output_power": 1.7457682442233738e-05, "reference_active": true, "delta_db": 2.7647741164783217}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7105571121059603e-06, "output_power": 2.3947765759922463e-06, "reference_active": true, "delta_db": 1.4612742253981197}], "active_band_count": 3, "active_band_mae_db": 1.905736869941352}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 0.00010533749157464783, "log_rms_mae_db": 1.1909643613283014, "rms_envelope_nrmse": 0.19492593099965394, "envelope_correlation": 0.9936328563692569, "rms_gain_error_db": 1.3274092390376815, "reference_mean_power": 0.0005773280667275119, "output_mean_power": 0.0007837248097237663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 3.141212974645643e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001274169753045263, "output_power": 0.00018128153828545084, "reference_active": true, "delta_db": 1.5312628644808637}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.860144772024442e-06, "output_power": 7.056905983404153e-06, "reference_active": true, "delta_db": 2.62010738637617}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1444270061935373e-07, "output_power": 1.4824057608616655e-07, "reference_active": true, "delta_db": 1.1237899631345551}], "active_band_count": 3, "active_band_mae_db": 1.7583867379971962}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 8.389504112280413e-05, "log_rms_mae_db": 1.3936281127425727, "rms_envelope_nrmse": 0.18583427092286042, "envelope_correlation": 0.9894982915966917, "rms_gain_error_db": 1.079727864921335, "reference_mean_power": 0.00021157950445783077, "output_mean_power": 0.0002712978687878179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.2662436125548267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5392735637289177e-05, "output_power": 7.118559602697152e-05, "reference_active": true, "delta_db": 1.0893931184057286}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.790041823950801e-06, "output_power": 6.582260892568387e-06, "reference_active": true, "delta_db": 0.5569339103488672}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4111913003034675e-06, "output_power": 1.9199882751543923e-06, "reference_active": true, "delta_db": 1.3371268613132479}], "active_band_count": 3, "active_band_mae_db": 0.9944846300226146}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 7.722064559288104e-05, "log_rms_mae_db": 1.011865438064268, "rms_envelope_nrmse": 0.15179104306703362, "envelope_correlation": 0.9965686666736996, "rms_gain_error_db": 1.0797282650930031, "reference_mean_power": 0.0002128095577406816, "output_mean_power": 0.0002728751300392804}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 1.2078803728484267e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.0224253455435505e-05, "output_power": 6.464338146934802e-05, "reference_active": true, "delta_db": 1.096105761131184}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7667526770926532e-06, "output_power": 3.018638698253803e-06, "reference_active": true, "delta_db": 0.378407964611049}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.983894982848384e-08, "output_power": 1.2123375925206132e-07, "reference_active": true, "delta_db": 0.8432356793377966}], "active_band_count": 3, "active_band_mae_db": 0.7725831350266765}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.54681310079566e-05, "log_rms_mae_db": 1.6826631600767874, "rms_envelope_nrmse": 0.27356071043129154, "envelope_correlation": 0.9609329539346882, "rms_gain_error_db": 0.7314636515226209, "reference_mean_power": 0.00010955764718773985, "output_mean_power": 0.00012965493808436614}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 6.937978158746457e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.2971450195505076e-05, "output_power": 2.9842484678877003e-05, "reference_active": true, "delta_db": 1.1364656635186106}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1375769637538917e-05, "output_power": 1.3918255797673268e-05, "reference_active": true, "delta_db": 0.8760402505282696}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4470462261987434e-06, "output_power": 3.0284930939140656e-06, "reference_active": true, "delta_db": 0.9258441413071307}], "active_band_count": 3, "active_band_mae_db": 0.9794500184513368}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83120, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1032, "rms_floor": 5.762149573106217e-05, "log_rms_mae_db": 1.151238349182671, "rms_envelope_nrmse": 0.18484842363488582, "envelope_correlation": 0.9837276752610551, "rms_gain_error_db": 0.7314641155752546, "reference_mean_power": 0.00011019456862153096, "output_mean_power": 0.00013040871053026122}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1032, "frequency_resolution_hz": 0.19379844961240308, "power_floor": 6.599778128222787e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3073624841125754e-05, "output_power": 3.068454351818266e-05, "reference_active": true, "delta_db": 1.2380383977369807}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.235055585786147e-06, "output_power": 7.2476019344232745e-06, "reference_active": true, "delta_db": 0.6535400265061573}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9515593840599056e-07, "output_power": 2.7496283432032407e-07, "reference_active": true, "delta_db": 1.488922248868316}], "active_band_count": 3, "active_band_mae_db": 1.126833557703818}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9448022517648913, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7ae4a4f5ad606cebfdcde0298d70aafe0e55d276d2b2741176cb60774389fd77", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9894982915966917, "standard_deviation": null, "minimum": 0.9894982915966917, "maximum": 0.9894982915966917, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3936281127425727, "standard_deviation": null, "minimum": 1.3936281127425727, "maximum": 1.3936281127425727, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9944846300226146, "standard_deviation": null, "minimum": 0.9944846300226146, "maximum": 0.9944846300226146, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18583427092286042, "standard_deviation": null, "minimum": 0.18583427092286042, "maximum": 0.18583427092286042, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9965686666736996, "standard_deviation": null, "minimum": 0.9965686666736996, "maximum": 0.9965686666736996, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.011865438064268, "standard_deviation": null, "minimum": 1.011865438064268, "maximum": 1.011865438064268, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7725831350266765, "standard_deviation": null, "minimum": 0.7725831350266765, "maximum": 0.7725831350266765, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15179104306703362, "standard_deviation": null, "minimum": 0.15179104306703362, "maximum": 0.15179104306703362, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9609329539346882, "standard_deviation": null, "minimum": 0.9609329539346882, "maximum": 0.9609329539346882, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6826631600767874, "standard_deviation": null, "minimum": 1.6826631600767874, "maximum": 1.6826631600767874, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9794500184513368, "standard_deviation": null, "minimum": 0.9794500184513368, "maximum": 0.9794500184513368, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.27356071043129154, "standard_deviation": null, "minimum": 0.27356071043129154, "maximum": 0.27356071043129154, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9837276752610551, "standard_deviation": null, "minimum": 0.9837276752610551, "maximum": 0.9837276752610551, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.151238349182671, "standard_deviation": null, "minimum": 1.151238349182671, "maximum": 1.151238349182671, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.126833557703818, "standard_deviation": null, "minimum": 1.126833557703818, "maximum": 1.126833557703818, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18484842363488582, "standard_deviation": null, "minimum": 0.18484842363488582, "maximum": 0.18484842363488582, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9870927684669716, "standard_deviation": null, "minimum": 0.9870927684669716, "maximum": 0.9870927684669716, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.4763191614425781, "standard_deviation": null, "minimum": 1.4763191614425781, "maximum": 1.4763191614425781, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.905736869941352, "standard_deviation": null, "minimum": 1.905736869941352, "maximum": 1.905736869941352, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22194950425862775, "standard_deviation": null, "minimum": 0.22194950425862775, "maximum": 0.22194950425862775, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936328563692569, "standard_deviation": null, "minimum": 0.9936328563692569, "maximum": 0.9936328563692569, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1909643613283014, "standard_deviation": null, "minimum": 1.1909643613283014, "maximum": 1.1909643613283014, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7583867379971962, "standard_deviation": null, "minimum": 1.7583867379971962, "maximum": 1.7583867379971962, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19492593099965394, "standard_deviation": null, "minimum": 0.19492593099965394, "maximum": 0.19492593099965394, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7288337297014321, "standard_deviation": null, "minimum": 0.7288337297014321, "maximum": 0.7288337297014321, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.763391047758637, "standard_deviation": null, "minimum": 5.763391047758637, "maximum": 5.763391047758637, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5538989575903049, "standard_deviation": null, "minimum": 0.5538989575903049, "maximum": 0.5538989575903049, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1230371129869638, "standard_deviation": null, "minimum": 1.1230371129869638, "maximum": 1.1230371129869638, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9856113335981339, "standard_deviation": null, "minimum": 0.9856113335981339, "maximum": 0.9856113335981339, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.954397311262717, "standard_deviation": null, "minimum": 1.954397311262717, "maximum": 1.954397311262717, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0651784829440223, "standard_deviation": null, "minimum": 1.0651784829440223, "maximum": 1.0651784829440223, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19995561363629505, "standard_deviation": null, "minimum": 0.19995561363629505, "maximum": 0.19995561363629505, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9948153033228482, "standard_deviation": null, "minimum": 0.9948153033228482, "maximum": 0.9948153033228482, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.5371808469607653, "standard_deviation": null, "minimum": 1.5371808469607653, "maximum": 1.5371808469607653, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0209181045845488, "standard_deviation": null, "minimum": 1.0209181045845488, "maximum": 1.0209181045845488, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16322376336185712, "standard_deviation": null, "minimum": 0.16322376336185712, "maximum": 0.16322376336185712, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5142"}, {"case": "librispeech_5639-40744-0013", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2454609870910645, "predicted_mos": 3.2454609870910645, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "16886c5f7a0ea0eeb272187d5aff0196a77fb7ccdff0718f204b8b530606d6d8", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 8.92319200001657}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4866295144774988, "全帯域_10ms_rms_envelope_nrmse": 0.18171739831057646, "全帯域_10ms_envelope_correlation": 0.9819510968546322, "全帯域_10ms_modulation_mae_db": 0.7247778999266505, "全帯域_40ms_log_rms_mae_db": 0.9724368406968728, "全帯域_40ms_rms_envelope_nrmse": 0.12717976684872284, "全帯域_40ms_envelope_correlation": 0.9956440095018423, "全帯域_40ms_modulation_mae_db": 0.8428471364743281, "300–1000 Hz_10ms_log_rms_mae_db": 1.9443064366155234, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21087465062408256, "300–1000 Hz_10ms_envelope_correlation": 0.9841473494123829, "300–1000 Hz_10ms_modulation_mae_db": 1.1190780164937768, "300–1000 Hz_40ms_log_rms_mae_db": 1.3423899927166798, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.16327528580066777, "300–1000 Hz_40ms_envelope_correlation": 0.994971690192088, "300–1000 Hz_40ms_modulation_mae_db": 1.4855772335322526, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9053770628437272, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.26472504025145005, "1000–3000 Hz_10ms_envelope_correlation": 0.9752132686732475, "1000–3000 Hz_10ms_modulation_mae_db": 0.43987529227287453, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3095550028498297, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.21154203376829045, "1000–3000 Hz_40ms_envelope_correlation": 0.9897987800078218, "1000–3000 Hz_40ms_modulation_mae_db": 0.639750474139498, "3000–6000 Hz_10ms_log_rms_mae_db": 2.023863851027917, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21338170439363324, "3000–6000 Hz_10ms_envelope_correlation": 0.972731823879088, "3000–6000 Hz_10ms_modulation_mae_db": 0.5604962501288632, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2506295403314558, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1263165748883761, "3000–6000 Hz_40ms_envelope_correlation": 0.9916000561765136, "3000–6000 Hz_40ms_modulation_mae_db": 0.5917236558522453, "mr_stft_db": 5.721709973918941, "band_mae_db": 0.708615969339288, "rms_gain_error_db": 0.8730929003270607, "output_peak": 0.3689477954400511}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.445474931564364, "spectral_convergence": 0.669094374698896, "spectral_convergence_status": "ok", "amplitude_floor": 7.181372005990482e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.045577152716893, "spectral_convergence": 0.2677754910063521, "spectral_convergence_status": "ok", "amplitude_floor": 5.417622086660404e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.674077837475563, "spectral_convergence": 0.2893164873136046, "spectral_convergence_status": "ok", "amplitude_floor": 3.166881446630363e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001077391554152542, "output_power": 0.00010769474933985816, "delta_db": -0.0017903694223548546, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002614688899274717, "output_power": 0.0003248830191019589, "delta_db": 0.9430698926877389, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00015854777413607497, "output_power": 0.00020699208241182305, "delta_db": 1.1579358440124103, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.449556222199797e-05, "output_power": 8.778733600338139e-05, "delta_db": 1.3390203716575821, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.5950148027800597e-05, "output_power": 1.9095986147626103e-05, "delta_db": 0.7817737300968386, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.2322545231279567e-05, "output_power": 1.2243056681383336e-05, "delta_db": -0.02810560815880373, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.708615969339288, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8730929003270607, "pitch": null, "output_rms_dbfs": -31.199320741415747, "output_peak": 0.3689477954400511, "reference_rms_dbfs": -32.072413641742806, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.721709973918941}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001113800968506657, "log_rms_mae_db": 1.4866295144774988, "rms_envelope_nrmse": 0.18171739831057646, "envelope_correlation": 0.9819510968546322, "rms_gain_error_db": 0.873053314489411, "reference_mean_power": 0.0006210401213146882, "output_mean_power": 0.0007593202629142034}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.9949527250696107e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014723003466329858, "output_power": 0.000185616653165453, "reference_active": true, "delta_db": 1.0062052335229277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.0749733198322775e-05, "output_power": 3.8132862564942005e-05, "reference_active": true, "delta_db": 0.9345805624887489}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6784890109842875e-06, "output_power": 5.381185946206492e-06, "reference_active": true, "delta_db": -0.23354790376827486}], "active_band_count": 3, "active_band_mae_db": 0.7247778999266505}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.537149969198292e-05, "log_rms_mae_db": 0.9724368406968728, "rms_envelope_nrmse": 0.12717976684872284, "envelope_correlation": 0.9956440095018423, "rms_gain_error_db": 0.8730577328856368, "reference_mean_power": 0.0006241534506116662, "output_mean_power": 0.0007631275792301033}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.688739338888854e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013414910111714758, "output_power": 0.00017055110972798705, "reference_active": true, "delta_db": 1.0426678261121962}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.37620762285614e-05, "output_power": 1.778436324923484e-05, "reference_active": true, "delta_db": 1.1135436124714748}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.149987894354722e-07, "output_power": 3.809025084884302e-07, "reference_active": true, "delta_db": -0.3723299708393132}], "active_band_count": 3, "active_band_mae_db": 0.8428471364743281}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.440030149355929e-05, "log_rms_mae_db": 1.9443064366155234, "rms_envelope_nrmse": 0.21087465062408256, "envelope_correlation": 0.9841473494123829, "rms_gain_error_db": 1.1264223084594556, "reference_mean_power": 0.000315142987429675, "output_mean_power": 0.0004084603249210561}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.94905977338963e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010605229421605894, "output_power": 0.00014297584355348495, "reference_active": true, "delta_db": 1.2974259955900087}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.109806584151368e-05, "output_power": 3.082872644174115e-05, "reference_active": true, "delta_db": 1.6471294074617548}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3612767263836796e-06, "output_power": 3.0565844811736003e-06, "reference_active": true, "delta_db": -0.4126786464295662}], "active_band_count": 3, "active_band_mae_db": 1.1190780164937768}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.828793840107377e-05, "log_rms_mae_db": 1.3423899927166798, "rms_envelope_nrmse": 0.16327528580066777, "envelope_correlation": 0.994971690192088, "rms_gain_error_db": 1.126425326033768, "reference_mean_power": 0.00031672629889435445, "output_mean_power": 0.0004105127577225989}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7988521248624547e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.967318213226487e-05, "output_power": 0.00013485241738338803, "reference_active": true, "delta_db": 1.3128041246175794}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.491202888187961e-06, "output_power": 1.4298671088294958e-05, "reference_active": true, "delta_db": 1.7797441904878806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5805315595038687e-07, "output_power": 2.6153460905036784e-07, "reference_active": true, "delta_db": -1.3641833854912981}], "active_band_count": 3, "active_band_mae_db": 1.4855772335322526}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.58934180861454e-05, "log_rms_mae_db": 1.9053770628437272, "rms_envelope_nrmse": 0.26472504025145005, "envelope_correlation": 0.9752132686732475, "rms_gain_error_db": 1.3326737029552447, "reference_mean_power": 6.36645119573909e-05, "output_mean_power": 8.652961760841792e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.512623526912829e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3991998554391248e-05, "output_power": 1.7017316259952834e-05, "reference_active": true, "delta_db": 0.8501131851951672}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3957717792014314e-06, "output_power": 2.6451243576801633e-06, "reference_active": true, "delta_db": 0.43000649868080726}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.685934249147775e-07, "output_power": 7.756169371134814e-07, "reference_active": true, "delta_db": 0.03950619294264907}], "active_band_count": 3, "active_band_mae_db": 0.43987529227287453}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6382034637347546e-05, "log_rms_mae_db": 1.3095550028498297, "rms_envelope_nrmse": 0.21154203376829045, "envelope_correlation": 0.9897987800078218, "rms_gain_error_db": 1.332724076926227, "reference_mean_power": 6.398182095924481e-05, "output_mean_power": 8.696189678753132e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.275432891493173e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2825234350232933e-05, "output_power": 1.552438446000271e-05, "reference_active": true, "delta_db": 0.829490799246158}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.068139769619595e-06, "output_power": 1.2368912053402016e-06, "reference_active": true, "delta_db": 0.6370341629115782}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.783047579212978e-08, "output_power": 5.210556618951746e-08, "reference_active": true, "delta_db": -0.4527264602607579}], "active_band_count": 3, "active_band_mae_db": 0.639750474139498}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.0860899025143115e-05, "log_rms_mae_db": 2.023863851027917, "rms_envelope_nrmse": 0.21338170439363324, "envelope_correlation": 0.972731823879088, "rms_gain_error_db": 0.5181283043381412, "reference_mean_power": 1.1507041647425964e-05, "output_mean_power": 1.2965119241456956e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.593308449162533e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.586447862335724e-06, "output_power": 5.225114289768432e-06, "reference_active": true, "delta_db": 0.5661933336650051}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.405946680666416e-07, "output_power": 1.0789183861506486e-06, "reference_active": true, "delta_db": 1.0840196316755415}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.329768801508531e-07, "output_power": 3.3058755564518507e-07, "reference_active": true, "delta_db": -0.03127578504604285}], "active_band_count": 3, "active_band_mae_db": 0.5604962501288632}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.4734203651083813e-05, "log_rms_mae_db": 1.2506295403314558, "rms_envelope_nrmse": 0.1263165748883761, "envelope_correlation": 0.9916000561765136, "rms_gain_error_db": 0.5175627280725849, "reference_mean_power": 1.156194575111937e-05, "output_mean_power": 1.3025283964334868e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.1073798497935e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.414350374295127e-06, "output_power": 5.048678706841195e-06, "reference_active": true, "delta_db": 0.5831093261749758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.649548666103306e-07, "output_power": 5.933866827985789e-07, "reference_active": true, "delta_db": 1.0592699758351403}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1590180837860948e-08, "output_power": 2.0940021025740323e-08, "reference_active": true, "delta_db": -0.13279166554661972}], "active_band_count": 3, "active_band_mae_db": 0.5917236558522453}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9334566718892673, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "88a5ec88ae94fbb957add2ad65d2a83af9afd1292771818a3a9da336c0ab126b", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9752132686732475, "standard_deviation": null, "minimum": 0.9752132686732475, "maximum": 0.9752132686732475, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9053770628437272, "standard_deviation": null, "minimum": 1.9053770628437272, "maximum": 1.9053770628437272, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.43987529227287453, "standard_deviation": null, "minimum": 0.43987529227287453, "maximum": 0.43987529227287453, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26472504025145005, "standard_deviation": null, "minimum": 0.26472504025145005, "maximum": 0.26472504025145005, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9897987800078218, "standard_deviation": null, "minimum": 0.9897987800078218, "maximum": 0.9897987800078218, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3095550028498297, "standard_deviation": null, "minimum": 1.3095550028498297, "maximum": 1.3095550028498297, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.639750474139498, "standard_deviation": null, "minimum": 0.639750474139498, "maximum": 0.639750474139498, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.21154203376829045, "standard_deviation": null, "minimum": 0.21154203376829045, "maximum": 0.21154203376829045, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.972731823879088, "standard_deviation": null, "minimum": 0.972731823879088, "maximum": 0.972731823879088, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.023863851027917, "standard_deviation": null, "minimum": 2.023863851027917, "maximum": 2.023863851027917, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5604962501288632, "standard_deviation": null, "minimum": 0.5604962501288632, "maximum": 0.5604962501288632, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21338170439363324, "standard_deviation": null, "minimum": 0.21338170439363324, "maximum": 0.21338170439363324, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9916000561765136, "standard_deviation": null, "minimum": 0.9916000561765136, "maximum": 0.9916000561765136, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2506295403314558, "standard_deviation": null, "minimum": 1.2506295403314558, "maximum": 1.2506295403314558, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5917236558522453, "standard_deviation": null, "minimum": 0.5917236558522453, "maximum": 0.5917236558522453, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1263165748883761, "standard_deviation": null, "minimum": 0.1263165748883761, "maximum": 0.1263165748883761, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9841473494123829, "standard_deviation": null, "minimum": 0.9841473494123829, "maximum": 0.9841473494123829, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9443064366155234, "standard_deviation": null, "minimum": 1.9443064366155234, "maximum": 1.9443064366155234, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1190780164937768, "standard_deviation": null, "minimum": 1.1190780164937768, "maximum": 1.1190780164937768, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21087465062408256, "standard_deviation": null, "minimum": 0.21087465062408256, "maximum": 0.21087465062408256, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.994971690192088, "standard_deviation": null, "minimum": 0.994971690192088, "maximum": 0.994971690192088, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.3423899927166798, "standard_deviation": null, "minimum": 1.3423899927166798, "maximum": 1.3423899927166798, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.4855772335322526, "standard_deviation": null, "minimum": 1.4855772335322526, "maximum": 1.4855772335322526, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16327528580066777, "standard_deviation": null, "minimum": 0.16327528580066777, "maximum": 0.16327528580066777, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.708615969339288, "standard_deviation": null, "minimum": 0.708615969339288, "maximum": 0.708615969339288, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.721709973918941, "standard_deviation": null, "minimum": 5.721709973918941, "maximum": 5.721709973918941, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.3689477954400511, "standard_deviation": null, "minimum": 0.3689477954400511, "maximum": 0.3689477954400511, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8730929003270607, "standard_deviation": null, "minimum": 0.8730929003270607, "maximum": 0.8730929003270607, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9819510968546322, "standard_deviation": null, "minimum": 0.9819510968546322, "maximum": 0.9819510968546322, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4866295144774988, "standard_deviation": null, "minimum": 1.4866295144774988, "maximum": 1.4866295144774988, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7247778999266505, "standard_deviation": null, "minimum": 0.7247778999266505, "maximum": 0.7247778999266505, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18171739831057646, "standard_deviation": null, "minimum": 0.18171739831057646, "maximum": 0.18171739831057646, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9956440095018423, "standard_deviation": null, "minimum": 0.9956440095018423, "maximum": 0.9956440095018423, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9724368406968728, "standard_deviation": null, "minimum": 0.9724368406968728, "maximum": 0.9724368406968728, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8428471364743281, "standard_deviation": null, "minimum": 0.8428471364743281, "maximum": 0.8428471364743281, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12717976684872284, "standard_deviation": null, "minimum": 0.12717976684872284, "maximum": 0.12717976684872284, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5639"}, {"case": "librispeech_5639-40744-0023", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2355141639709473, "predicted_mos": 3.2355141639709473, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "fe2681cfeae6da9158ef53659446c2c54a2aa604552624e4fed309cc5fda080e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.039270900189877}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4895143870319172, "全帯域_10ms_rms_envelope_nrmse": 0.2263353816737757, "全帯域_10ms_envelope_correlation": 0.9798983385160985, "全帯域_10ms_modulation_mae_db": 1.1079801954872153, "全帯域_40ms_log_rms_mae_db": 1.1039690330010092, "全帯域_40ms_rms_envelope_nrmse": 0.18792477128489607, "全帯域_40ms_envelope_correlation": 0.9928687923539932, "全帯域_40ms_modulation_mae_db": 0.7929429409794798, "300–1000 Hz_10ms_log_rms_mae_db": 1.8850368452826471, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2797404442692956, "300–1000 Hz_10ms_envelope_correlation": 0.9819327049569908, "300–1000 Hz_10ms_modulation_mae_db": 1.4423261859841474, "300–1000 Hz_40ms_log_rms_mae_db": 1.543462966342258, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2490151672275235, "300–1000 Hz_40ms_envelope_correlation": 0.9918060610236759, "300–1000 Hz_40ms_modulation_mae_db": 1.1438181174776505, "1000–3000 Hz_10ms_log_rms_mae_db": 1.558019336108271, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21348338719125493, "1000–3000 Hz_10ms_envelope_correlation": 0.9798754526266059, "1000–3000 Hz_10ms_modulation_mae_db": 0.908851818564, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0085839507229655, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1460717803954614, "1000–3000 Hz_40ms_envelope_correlation": 0.9951885268082342, "1000–3000 Hz_40ms_modulation_mae_db": 1.1261126380391333, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7190619490838939, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.199952880840787, "3000–6000 Hz_10ms_envelope_correlation": 0.9702472241455977, "3000–6000 Hz_10ms_modulation_mae_db": 0.06244071518678779, "3000–6000 Hz_40ms_log_rms_mae_db": 1.056435183545237, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10036004757314014, "3000–6000 Hz_40ms_envelope_correlation": 0.9924622798947083, "3000–6000 Hz_40ms_modulation_mae_db": 0.1647712728625301, "mr_stft_db": 5.304715647163007, "band_mae_db": 0.7643908224803712, "rms_gain_error_db": 1.318050876678174, "output_peak": 0.27661458404729194}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.132128597061892, "spectral_convergence": 0.6374444545935363, "spectral_convergence_status": "ok", "amplitude_floor": 7.733796592545008e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.709182566175443, "spectral_convergence": 0.2958694286213384, "spectral_convergence_status": "ok", "amplitude_floor": 4.871370963739547e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.072835778251682, "spectral_convergence": 0.31970425526445506, "spectral_convergence_status": "ok", "amplitude_floor": 4.7483837933045734e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001233024392514864, "output_power": 0.00012322327099324737, "delta_db": -0.002789351166895293, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000393494821723947, "output_power": 0.0005508566121691881, "delta_db": 1.4609954527807956, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000358215617892152, "output_power": 0.0005215656160450157, "delta_db": 1.6316443622803642, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 5.947495410559829e-05, "output_power": 7.626024845069347e-05, "delta_db": 1.079641002597327, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 1.0477761937191392e-05, "output_power": 1.1137567977092869e-05, "delta_db": 0.26521841008896174, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 9.39686924640271e-06, "output_power": 9.718266837799487e-06, "delta_db": 0.1460563559678833, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7643908224803712, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.318050876678174, "pitch": null, "output_rms_dbfs": -28.88481562485235, "output_peak": 0.27661458404729194, "reference_rms_dbfs": -30.202866501530522, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.304715647163007}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.933633747331572e-05, "log_rms_mae_db": 1.4895143870319172, "rms_envelope_nrmse": 0.2263353816737757, "envelope_correlation": 0.9798983385160985, "rms_gain_error_db": 1.318051011778869, "reference_mean_power": 0.0009551583992700257, "output_mean_power": 0.0012938397815716095}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.1194144575506277e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002166069295810575, "output_power": 0.00029141730885107924, "reference_active": true, "delta_db": 1.2884299705526678}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.38201712334803e-05, "output_power": 3.069394210286501e-05, "reference_active": true, "delta_db": 1.1010779046091912}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.302931519380675e-06, "output_power": 1.029614682613161e-05, "reference_active": true, "delta_db": 0.9344327112997866}], "active_band_count": 3, "active_band_mae_db": 1.1079801954872153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.652181544798992e-05, "log_rms_mae_db": 1.1039690330010092, "rms_envelope_nrmse": 0.18792477128489607, "envelope_correlation": 0.9928687923539932, "rms_gain_error_db": 1.3180512780774936, "reference_mean_power": 0.0009599620519447541, "output_mean_power": 0.001300346799769647}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.73215395386014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001986653950491428, "output_power": 0.0002696619794747984, "reference_active": true, "delta_db": 1.3269749297133595}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916927878865493e-05, "output_power": 1.3804015410419389e-05, "reference_active": true, "delta_db": 1.0190499424349353}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.81959177520348e-07, "output_power": 5.775799733189048e-07, "reference_active": true, "delta_db": -0.03280395079014471}], "active_band_count": 3, "active_band_mae_db": 0.7929429409794798}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.027389787958093e-05, "log_rms_mae_db": 1.8850368452826471, "rms_envelope_nrmse": 0.2797404442692956, "envelope_correlation": 0.9819327049569908, "rms_gain_error_db": 1.7458054165821648, "reference_mean_power": 0.0006127358569067375, "output_mean_power": 0.0009159121862573151}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.366962347376557e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001936423107817767, "output_power": 0.0002710799417658436, "reference_active": true, "delta_db": 1.4609712712797025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9545968551344243e-05, "output_power": 2.9178940005414694e-05, "reference_active": true, "delta_db": 1.7401231538821347}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.863910891753386e-06, "output_power": 8.895288745932172e-06, "reference_active": true, "delta_db": 1.1258841327906044}], "active_band_count": 3, "active_band_mae_db": 1.4423261859841474}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.943169310162246e-05, "log_rms_mae_db": 1.543462966342258, "rms_envelope_nrmse": 0.2490151672275235, "envelope_correlation": 0.9918060610236759, "rms_gain_error_db": 1.7458055338520648, "reference_mean_power": 0.000615817475654304, "output_mean_power": 0.0009205185877968484}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.097820742933078e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018026482035346189, "output_power": 0.00025358531432555735, "reference_active": true, "delta_db": 1.4821311891041322}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.739141870245847e-06, "output_power": 1.2600301423053361e-05, "reference_active": true, "delta_db": 1.589121446953314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.763892611677605e-07, "output_power": 5.305126073603923e-07, "reference_active": true, "delta_db": -0.360201716375505}], "active_band_count": 3, "active_band_mae_db": 1.1438181174776505}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.05554473589972e-05, "log_rms_mae_db": 1.558019336108271, "rms_envelope_nrmse": 0.21348338719125493, "envelope_correlation": 0.9798754526266059, "rms_gain_error_db": 0.9766426605410774, "reference_mean_power": 5.214627438247735e-05, "output_mean_power": 6.529614645602529e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.666296423769008e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.72355480524535e-05, "output_power": 3.685018949611524e-05, "reference_active": true, "delta_db": 1.3130360661993998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.163900077209868e-06, "output_power": 5.7168165943236436e-06, "reference_active": true, "delta_db": 1.3765396047226954}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1939677607224186e-06, "output_power": 1.2041776955360926e-06, "reference_active": true, "delta_db": 0.03697978476990485}], "active_band_count": 3, "active_band_mae_db": 0.908851818564}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.291047760123656e-05, "log_rms_mae_db": 1.0085839507229655, "rms_envelope_nrmse": 0.1460717803954614, "envelope_correlation": 0.9951885268082342, "rms_gain_error_db": 0.9766430904049754, "reference_mean_power": 5.2408520364582374e-05, "output_mean_power": 6.562453023749667e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.33304878770606e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.583453225376265e-05, "output_power": 3.529099373938067e-05, "reference_active": true, "delta_db": 1.3546328473886502}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0765400661096766e-06, "output_power": 2.8251090803584154e-06, "reference_active": true, "delta_db": 1.3369490594855227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.119613733354675e-08, "output_power": 7.168026649777931e-08, "reference_active": true, "delta_db": 0.6867560072432266}], "active_band_count": 3, "active_band_mae_db": 1.1261126380391333}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.3449414786854306e-05, "log_rms_mae_db": 1.7190619490838939, "rms_envelope_nrmse": 0.199952880840787, "envelope_correlation": 0.9702472241455977, "rms_gain_error_db": 0.24665114764581222, "reference_mean_power": 6.886139457761324e-06, "output_mean_power": 7.288546494326098e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.997833564030985e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.1829042042252175e-06, "output_power": 2.2206981905999236e-06, "reference_active": true, "delta_db": 0.07454861400732497}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.111376843686143e-07, "output_power": 7.920665543139264e-07, "reference_active": true, "delta_db": -0.10332903386661695}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6537364416157584e-07, "output_power": 1.6501440083396388e-07, "reference_active": true, "delta_db": -0.009444497686421445}], "active_band_count": 3, "active_band_mae_db": 0.06244071518678779}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.0291390653962812e-05, "log_rms_mae_db": 1.056435183545237, "rms_envelope_nrmse": 0.10036004757314014, "envelope_correlation": 0.9924622798947083, "rms_gain_error_db": 0.24665122123955957, "reference_mean_power": 6.9207717017349184e-06, "output_mean_power": 7.3252026754458945e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.443349031906855e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9318994555375575e-06, "output_power": 1.9845533063170794e-06, "reference_active": true, "delta_db": 0.1167824872445873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.056544990756342e-07, "output_power": 4.057254235690255e-07, "reference_active": true, "delta_db": 0.0007592526097925925}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.969546554060793e-09, "output_power": 8.691821919162586e-09, "reference_active": true, "delta_db": 0.3767720787332104}], "active_band_count": 3, "active_band_mae_db": 0.1647712728625301}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9418607997443519, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "9db046d0319e0914767c5ba5d48543648a29287d575687da4337e694853ff3b4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9798754526266059, "standard_deviation": null, "minimum": 0.9798754526266059, "maximum": 0.9798754526266059, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.558019336108271, "standard_deviation": null, "minimum": 1.558019336108271, "maximum": 1.558019336108271, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.908851818564, "standard_deviation": null, "minimum": 0.908851818564, "maximum": 0.908851818564, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21348338719125493, "standard_deviation": null, "minimum": 0.21348338719125493, "maximum": 0.21348338719125493, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951885268082342, "standard_deviation": null, "minimum": 0.9951885268082342, "maximum": 0.9951885268082342, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0085839507229655, "standard_deviation": null, "minimum": 1.0085839507229655, "maximum": 1.0085839507229655, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1261126380391333, "standard_deviation": null, "minimum": 1.1261126380391333, "maximum": 1.1261126380391333, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1460717803954614, "standard_deviation": null, "minimum": 0.1460717803954614, "maximum": 0.1460717803954614, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9702472241455977, "standard_deviation": null, "minimum": 0.9702472241455977, "maximum": 0.9702472241455977, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7190619490838939, "standard_deviation": null, "minimum": 1.7190619490838939, "maximum": 1.7190619490838939, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.06244071518678779, "standard_deviation": null, "minimum": 0.06244071518678779, "maximum": 0.06244071518678779, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.199952880840787, "standard_deviation": null, "minimum": 0.199952880840787, "maximum": 0.199952880840787, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9924622798947083, "standard_deviation": null, "minimum": 0.9924622798947083, "maximum": 0.9924622798947083, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.056435183545237, "standard_deviation": null, "minimum": 1.056435183545237, "maximum": 1.056435183545237, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.1647712728625301, "standard_deviation": null, "minimum": 0.1647712728625301, "maximum": 0.1647712728625301, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10036004757314014, "standard_deviation": null, "minimum": 0.10036004757314014, "maximum": 0.10036004757314014, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9819327049569908, "standard_deviation": null, "minimum": 0.9819327049569908, "maximum": 0.9819327049569908, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8850368452826471, "standard_deviation": null, "minimum": 1.8850368452826471, "maximum": 1.8850368452826471, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4423261859841474, "standard_deviation": null, "minimum": 1.4423261859841474, "maximum": 1.4423261859841474, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2797404442692956, "standard_deviation": null, "minimum": 0.2797404442692956, "maximum": 0.2797404442692956, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9918060610236759, "standard_deviation": null, "minimum": 0.9918060610236759, "maximum": 0.9918060610236759, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.543462966342258, "standard_deviation": null, "minimum": 1.543462966342258, "maximum": 1.543462966342258, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1438181174776505, "standard_deviation": null, "minimum": 1.1438181174776505, "maximum": 1.1438181174776505, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2490151672275235, "standard_deviation": null, "minimum": 0.2490151672275235, "maximum": 0.2490151672275235, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7643908224803712, "standard_deviation": null, "minimum": 0.7643908224803712, "maximum": 0.7643908224803712, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.304715647163007, "standard_deviation": null, "minimum": 5.304715647163007, "maximum": 5.304715647163007, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.27661458404729194, "standard_deviation": null, "minimum": 0.27661458404729194, "maximum": 0.27661458404729194, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.318050876678174, "standard_deviation": null, "minimum": 1.318050876678174, "maximum": 1.318050876678174, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9798983385160985, "standard_deviation": null, "minimum": 0.9798983385160985, "maximum": 0.9798983385160985, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4895143870319172, "standard_deviation": null, "minimum": 1.4895143870319172, "maximum": 1.4895143870319172, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1079801954872153, "standard_deviation": null, "minimum": 1.1079801954872153, "maximum": 1.1079801954872153, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2263353816737757, "standard_deviation": null, "minimum": 0.2263353816737757, "maximum": 0.2263353816737757, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9928687923539932, "standard_deviation": null, "minimum": 0.9928687923539932, "maximum": 0.9928687923539932, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1039690330010092, "standard_deviation": null, "minimum": 1.1039690330010092, "maximum": 1.1039690330010092, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7929429409794798, "standard_deviation": null, "minimum": 0.7929429409794798, "maximum": 0.7929429409794798, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18792477128489607, "standard_deviation": null, "minimum": 0.18792477128489607, "maximum": 0.18792477128489607, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5639"}, {"case": "librispeech_5639-40744-0030", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.7054362297058105, "predicted_mos": 3.7054362297058105, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "969ebece04724b1d5cafc07244856bbaf62967afc52a0967ce65cc674ba2435e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.711177699733526}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4124749319170697, "全帯域_10ms_rms_envelope_nrmse": 0.23659793912301827, "全帯域_10ms_envelope_correlation": 0.9802583697217724, "全帯域_10ms_modulation_mae_db": 1.5351454921271321, "全帯域_40ms_log_rms_mae_db": 0.9958606363839658, "全帯域_40ms_rms_envelope_nrmse": 0.19990529032136814, "全帯域_40ms_envelope_correlation": 0.9919380912994381, "全帯域_40ms_modulation_mae_db": 1.4790741571398682, "300–1000 Hz_10ms_log_rms_mae_db": 1.7050032964290935, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.3088786435143803, "300–1000 Hz_10ms_envelope_correlation": 0.9807064693054536, "300–1000 Hz_10ms_modulation_mae_db": 1.8415921659259358, "300–1000 Hz_40ms_log_rms_mae_db": 1.2814382202031767, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2804518378129315, "300–1000 Hz_40ms_envelope_correlation": 0.9898850514898667, "300–1000 Hz_40ms_modulation_mae_db": 1.4552092369153955, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5671670609594852, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18587898041379852, "1000–3000 Hz_10ms_envelope_correlation": 0.9811311226583069, "1000–3000 Hz_10ms_modulation_mae_db": 0.47761815678363, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9042387823829512, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10382482188447438, "1000–3000 Hz_40ms_envelope_correlation": 0.9970693601510047, "1000–3000 Hz_40ms_modulation_mae_db": 0.5592608166583325, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7305357089865883, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24443617257075528, "3000–6000 Hz_10ms_envelope_correlation": 0.9533414500796799, "3000–6000 Hz_10ms_modulation_mae_db": 0.6452565722216345, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0693529930020413, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1275320162587051, "3000–6000 Hz_40ms_envelope_correlation": 0.9860191470768315, "3000–6000 Hz_40ms_modulation_mae_db": 0.8183261086415324, "mr_stft_db": 5.442311095281695, "band_mae_db": 0.7851038304281007, "rms_gain_error_db": 1.3682564027986461, "output_peak": 0.2932275318322672}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.214908021236132, "spectral_convergence": 0.6481573926001409, "spectral_convergence_status": "ok", "amplitude_floor": 6.716189184049609e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.787786387122548, "spectral_convergence": 0.3043839815900761, "spectral_convergence_status": "ok", "amplitude_floor": 4.979433430044932e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.324238877486403, "spectral_convergence": 0.33470810698961195, "spectral_convergence_status": "ok", "amplitude_floor": 3.630042469105695e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.462171010234058e-05, "output_power": 7.16768366970297e-05, "delta_db": -0.1748636710131844, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0003077822779549395, "output_power": 0.00046759730388586676, "delta_db": 1.816283887816084, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 5.338393512143839e-05, "output_power": 7.18167897647151e-05, "delta_db": 1.288154040744355, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.1515857114073014e-05, "output_power": 2.5458748379266082e-05, "delta_db": 0.7307839726691744, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.940885952831811e-05, "output_power": 3.299861540564598e-05, "delta_db": 0.5001753467521338, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.017666424897867e-06, "output_power": 8.396227351601062e-06, "delta_db": 0.2003620635736714, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7851038304281007, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.3682564027986461, "pitch": null, "output_rms_dbfs": -31.688058444797353, "output_peak": 0.2932275318322672, "reference_rms_dbfs": -33.056314847596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.442311095281695}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.650394450462669e-05, "log_rms_mae_db": 1.4124749319170697, "rms_envelope_nrmse": 0.23659793912301827, "envelope_correlation": 0.9802583697217724, "rms_gain_error_db": 1.36846096337061, "reference_mean_power": 0.0004951134756771775, "output_mean_power": 0.0006785015485245708}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.645854738282274e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010486873457313227, "output_power": 0.00014994373061286194, "reference_active": true, "delta_db": 1.552822844239567}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5071002541192514e-05, "output_power": 3.858884520251363e-05, "reference_active": true, "delta_db": 1.872900813007149}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.059292105678353e-06, "output_power": 6.638363608309723e-06, "reference_active": true, "delta_db": 1.1797128191346802}], "active_band_count": 3, "active_band_mae_db": 1.5351454921271321}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.020301765617229e-05, "log_rms_mae_db": 0.9958606363839658, "rms_envelope_nrmse": 0.19990529032136814, "envelope_correlation": 0.9919380912994381, "rms_gain_error_db": 1.3687372159967468, "reference_mean_power": 0.0004970274687625529, "output_mean_power": 0.000681167804455106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.472129124053938e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.182773781582673e-05, "output_power": 0.00013233635954466952, "reference_active": true, "delta_db": 1.587052978506078}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2533326164459841e-05, "output_power": 2.0448334551261848e-05, "reference_active": true, "delta_db": 2.1259160014659786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4723702388808576e-07, "output_power": 2.9210490944793214e-07, "reference_active": true, "delta_db": 0.7242534914475486}], "active_band_count": 3, "active_band_mae_db": 1.4790741571398682}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.432622539351727e-05, "log_rms_mae_db": 1.7050032964290935, "rms_envelope_nrmse": 0.3088786435143803, "envelope_correlation": 0.9807064693054536, "rms_gain_error_db": 1.9191405598646991, "reference_mean_power": 0.00026104952136561866, "output_mean_power": 0.0004061037101937124}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.392840173310503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.633554932680893e-05, "output_power": 0.0001548330700616743, "reference_active": true, "delta_db": 2.0607714698610047}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.160826072139944e-05, "output_power": 3.5175150678811524e-05, "reference_active": true, "delta_db": 2.1161615515556633}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.769645916048805e-06, "output_power": 6.505347300682898e-06, "reference_active": true, "delta_db": 1.34784347636114}], "active_band_count": 3, "active_band_mae_db": 1.8415921659259358}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.77451695930137e-05, "log_rms_mae_db": 1.2814382202031767, "rms_envelope_nrmse": 0.2804518378129315, "envelope_correlation": 0.9898850514898667, "rms_gain_error_db": 1.919206986788355, "reference_mean_power": 0.00026200159904973233, "output_mean_power": 0.0004075910514244845}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2678124187221727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.604417147270571e-05, "output_power": 0.00013873977685285995, "reference_active": true, "delta_db": 2.0747953456974026}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1040946271380912e-05, "output_power": 1.8368023745313676e-05, "reference_active": true, "delta_db": 2.2105613567823124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9369342076961927e-07, "output_power": 2.883149197417858e-07, "reference_active": true, "delta_db": -0.08027100826647071}], "active_band_count": 3, "active_band_mae_db": 1.4552092369153955}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.588832427367551e-05, "log_rms_mae_db": 1.5671670609594852, "rms_envelope_nrmse": 0.18587898041379852, "envelope_correlation": 0.9811311226583069, "rms_gain_error_db": 0.6846697750891546, "reference_mean_power": 4.0575597065913316e-05, "output_mean_power": 4.750418774394036e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.7364679678483124e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6458578452566544e-05, "output_power": 1.9361519173198037e-05, "reference_active": true, "delta_db": 0.7054710863511898}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.144001855335748e-06, "output_power": 3.255610406681835e-06, "reference_active": true, "delta_db": 0.1514963439723913}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.079905260013679e-07, "output_power": 7.95229739385068e-07, "reference_active": true, "delta_db": -0.575887040027309}], "active_band_count": 3, "active_band_mae_db": 0.47761815678363}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.068450322528164e-05, "log_rms_mae_db": 0.9042387823829512, "rms_envelope_nrmse": 0.10382482188447438, "envelope_correlation": 0.9970693601510047, "rms_gain_error_db": 0.6848357997009608, "reference_mean_power": 4.071142209359612e-05, "output_mean_power": 4.766502805586375e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.516451938680711e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5218014411010514e-05, "output_power": 1.782495548451371e-05, "reference_active": true, "delta_db": 0.6867046290301825}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2444589918408885e-06, "output_power": 1.3480947155432844e-06, "reference_active": true, "delta_db": 0.3473981624071938}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2414094224167626e-08, "output_power": 4.519393110624231e-08, "reference_active": true, "delta_db": -0.6436796585376212}], "active_band_count": 3, "active_band_mae_db": 0.5592608166583325}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.7354442023779843e-05, "log_rms_mae_db": 1.7305357089865883, "rms_envelope_nrmse": 0.24443617257075528, "envelope_correlation": 0.9533414500796799, "rms_gain_error_db": 0.08866843967989682, "reference_mean_power": 9.786447369332928e-06, "output_mean_power": 9.988307609033204e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.593718108385407e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.378843387910109e-06, "output_power": 2.428295676024521e-06, "reference_active": true, "delta_db": 0.08935715494620904}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.926849454092717e-07, "output_power": 8.255915544654055e-07, "reference_active": true, "delta_db": -0.3393296910855312}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6983089704559518e-07, "output_power": 1.2003514363620265e-07, "reference_active": true, "delta_db": -1.5070828706331634}], "active_band_count": 3, "active_band_mae_db": 0.6452565722216345}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.2351945663576383e-05, "log_rms_mae_db": 1.0693529930020413, "rms_envelope_nrmse": 0.1275320162587051, "envelope_correlation": 0.9860191470768315, "rms_gain_error_db": 0.08858694424679334, "reference_mean_power": 9.83170860207206e-06, "output_mean_power": 1.0034314127132195e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.2206731147886266e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.414723334288e-06, "output_power": 2.4231440344384895e-06, "reference_active": true, "delta_db": 0.015118509969662694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.701511911331394e-07, "output_power": 3.806975866886799e-07, "reference_active": true, "delta_db": 0.12200972486898051}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1816168225696604e-08, "output_power": 6.92933656938206e-09, "reference_active": true, "delta_db": -2.317850091085954}], "active_band_count": 3, "active_band_mae_db": 0.8183261086415324}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9429131875840915, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "07d809b501ace08a9bb8189aaf5e21797d2124fe566fc43345ad89e4b57faf34", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9811311226583069, "standard_deviation": null, "minimum": 0.9811311226583069, "maximum": 0.9811311226583069, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5671670609594852, "standard_deviation": null, "minimum": 1.5671670609594852, "maximum": 1.5671670609594852, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.47761815678363, "standard_deviation": null, "minimum": 0.47761815678363, "maximum": 0.47761815678363, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18587898041379852, "standard_deviation": null, "minimum": 0.18587898041379852, "maximum": 0.18587898041379852, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970693601510047, "standard_deviation": null, "minimum": 0.9970693601510047, "maximum": 0.9970693601510047, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9042387823829512, "standard_deviation": null, "minimum": 0.9042387823829512, "maximum": 0.9042387823829512, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5592608166583325, "standard_deviation": null, "minimum": 0.5592608166583325, "maximum": 0.5592608166583325, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10382482188447438, "standard_deviation": null, "minimum": 0.10382482188447438, "maximum": 0.10382482188447438, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9533414500796799, "standard_deviation": null, "minimum": 0.9533414500796799, "maximum": 0.9533414500796799, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7305357089865883, "standard_deviation": null, "minimum": 1.7305357089865883, "maximum": 1.7305357089865883, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6452565722216345, "standard_deviation": null, "minimum": 0.6452565722216345, "maximum": 0.6452565722216345, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24443617257075528, "standard_deviation": null, "minimum": 0.24443617257075528, "maximum": 0.24443617257075528, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9860191470768315, "standard_deviation": null, "minimum": 0.9860191470768315, "maximum": 0.9860191470768315, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0693529930020413, "standard_deviation": null, "minimum": 1.0693529930020413, "maximum": 1.0693529930020413, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8183261086415324, "standard_deviation": null, "minimum": 0.8183261086415324, "maximum": 0.8183261086415324, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1275320162587051, "standard_deviation": null, "minimum": 0.1275320162587051, "maximum": 0.1275320162587051, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9807064693054536, "standard_deviation": null, "minimum": 0.9807064693054536, "maximum": 0.9807064693054536, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7050032964290935, "standard_deviation": null, "minimum": 1.7050032964290935, "maximum": 1.7050032964290935, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.8415921659259358, "standard_deviation": null, "minimum": 1.8415921659259358, "maximum": 1.8415921659259358, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3088786435143803, "standard_deviation": null, "minimum": 0.3088786435143803, "maximum": 0.3088786435143803, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9898850514898667, "standard_deviation": null, "minimum": 0.9898850514898667, "maximum": 0.9898850514898667, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2814382202031767, "standard_deviation": null, "minimum": 1.2814382202031767, "maximum": 1.2814382202031767, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.4552092369153955, "standard_deviation": null, "minimum": 1.4552092369153955, "maximum": 1.4552092369153955, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2804518378129315, "standard_deviation": null, "minimum": 0.2804518378129315, "maximum": 0.2804518378129315, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7851038304281007, "standard_deviation": null, "minimum": 0.7851038304281007, "maximum": 0.7851038304281007, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.442311095281695, "standard_deviation": null, "minimum": 5.442311095281695, "maximum": 5.442311095281695, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.2932275318322672, "standard_deviation": null, "minimum": 0.2932275318322672, "maximum": 0.2932275318322672, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.3682564027986461, "standard_deviation": null, "minimum": 1.3682564027986461, "maximum": 1.3682564027986461, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9802583697217724, "standard_deviation": null, "minimum": 0.9802583697217724, "maximum": 0.9802583697217724, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4124749319170697, "standard_deviation": null, "minimum": 1.4124749319170697, "maximum": 1.4124749319170697, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.5351454921271321, "standard_deviation": null, "minimum": 1.5351454921271321, "maximum": 1.5351454921271321, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.23659793912301827, "standard_deviation": null, "minimum": 0.23659793912301827, "maximum": 0.23659793912301827, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9919380912994381, "standard_deviation": null, "minimum": 0.9919380912994381, "maximum": 0.9919380912994381, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 0.9958606363839658, "standard_deviation": null, "minimum": 0.9958606363839658, "maximum": 0.9958606363839658, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4790741571398682, "standard_deviation": null, "minimum": 1.4790741571398682, "maximum": 1.4790741571398682, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.19990529032136814, "standard_deviation": null, "minimum": 0.19990529032136814, "maximum": 0.19990529032136814, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5639"}, {"case": "librispeech_5683-32865-0017", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.375115394592285, "predicted_mos": 2.375115394592285, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 87280, "input_sample_rate": 16000, "input_seconds": 5.455, "input_float32_sha256": "e125557f3ffd5e62c5df52743f1403895eaf93be8e6943bd6e97558542511b34", "internal_samples": 87280, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 11.097287599928677}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.5529654022209405, "全帯域_10ms_rms_envelope_nrmse": 0.16440355771850168, "全帯域_10ms_envelope_correlation": 0.9833764607501585, "全帯域_10ms_modulation_mae_db": 0.5186826178104532, "全帯域_40ms_log_rms_mae_db": 2.2861498123606276, "全帯域_40ms_rms_envelope_nrmse": 0.1261375005891387, "全帯域_40ms_envelope_correlation": 0.9926664393698768, "全帯域_40ms_modulation_mae_db": 0.8240407649979566, "300–1000 Hz_10ms_log_rms_mae_db": 2.155335832608833, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2067790423098858, "300–1000 Hz_10ms_envelope_correlation": 0.9811160950929382, "300–1000 Hz_10ms_modulation_mae_db": 0.8811718416130027, "300–1000 Hz_40ms_log_rms_mae_db": 1.8581292737707136, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17817412435592372, "300–1000 Hz_40ms_envelope_correlation": 0.9880366187859074, "300–1000 Hz_40ms_modulation_mae_db": 0.9169414891130652, "1000–3000 Hz_10ms_log_rms_mae_db": 1.707229678220966, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.25277039836434895, "1000–3000 Hz_10ms_envelope_correlation": 0.9735599405977656, "1000–3000 Hz_10ms_modulation_mae_db": 0.9161499509853172, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2611277480467669, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18976128774809217, "1000–3000 Hz_40ms_envelope_correlation": 0.9896872742656379, "1000–3000 Hz_40ms_modulation_mae_db": 1.0753247244462028, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6474546379436537, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.359989214862954, "3000–6000 Hz_10ms_envelope_correlation": 0.9471876488708637, "3000–6000 Hz_10ms_modulation_mae_db": 1.9862954492993572, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1671355080917756, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.26745005491065293, "3000–6000 Hz_40ms_envelope_correlation": 0.9782462294693193, "3000–6000 Hz_40ms_modulation_mae_db": 1.8403735031009774, "mr_stft_db": 6.124047430994723, "band_mae_db": 0.7946518236943906, "rms_gain_error_db": 0.7618805801095283, "output_peak": 0.17061504377943698}, "full_scale": {"sample_count": 87280, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 87280, "output_samples": 87280, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.338548178923306, "spectral_convergence": 0.580490246173947, "spectral_convergence_status": "ok", "amplitude_floor": 4.3919411473735467e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.897342354600674, "spectral_convergence": 0.33966129134164164, "spectral_convergence_status": "ok", "amplitude_floor": 3.1709236310047874e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.136251759460188, "spectral_convergence": 0.3687529534162129, "spectral_convergence_status": "ok", "amplitude_floor": 2.989229099554426e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0001235363438035874, "output_power": 0.00013746064461344514, "delta_db": 0.4638363241530483, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 7.224896060742283e-05, "output_power": 9.047086717954544e-05, "delta_db": 0.9767714952870427, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.329659782840476e-05, "output_power": 5.3190444486497044e-05, "delta_db": 0.8937984793543706, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.747692740945149e-05, "output_power": 6.186503882888797e-05, "delta_db": 1.1496268512606176, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.909876260278539e-06, "output_power": 1.0716833304121477e-05, "delta_db": 0.8019480286708158, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.1191996093186054e-05, "output_power": 2.367908091457133e-05, "delta_db": 0.4819297634404491, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7946518236943906, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7618805801095283, "pitch": null, "output_rms_dbfs": -34.232177717731375, "output_peak": 0.17061504377943698, "reference_rms_dbfs": -34.9940582978409, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.124047430994723}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 87280, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 5.579882181126775e-05, "log_rms_mae_db": 2.5529654022209405, "rms_envelope_nrmse": 0.16440355771850168, "envelope_correlation": 0.9833764607501585, "rms_gain_error_db": 0.7618821732064283, "reference_mean_power": 0.0003169510345489487, "output_mean_power": 0.0003777290540079866}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 1.8938855835285014e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.978895728357633e-05, "output_power": 7.231103158782177e-05, "reference_active": true, "delta_db": 0.8258357780285366}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1930505918996815e-05, "output_power": 2.548305189924312e-05, "reference_active": true, "delta_db": 0.6520278810464064}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0863861439295844e-06, "output_power": 3.0313203925556404e-06, "reference_active": true, "delta_db": -0.0781841943564166}], "active_band_count": 3, "active_band_mae_db": 0.5186826178104532}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 5.258028604719465e-05, "log_rms_mae_db": 2.2861498123606276, "rms_envelope_nrmse": 0.1261375005891387, "envelope_correlation": 0.9926664393698768, "rms_gain_error_db": 0.7618960189540576, "reference_mean_power": 0.00031870366172679027, "output_mean_power": 0.0003798189730476075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 1.711395707762808e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.23137186055323e-05, "output_power": 6.402252576707607e-05, "reference_active": true, "delta_db": 0.8771721160351542}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1932036565608145e-05, "output_power": 1.3666427701812944e-05, "reference_active": true, "delta_db": 0.5894043261017742}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.569030812932547e-07, "output_power": 2.0380495001140257e-07, "reference_active": true, "delta_db": -1.0055458528569419}], "active_band_count": 3, "active_band_mae_db": 0.8240407649979566}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 4.2439491809696055e-05, "log_rms_mae_db": 2.155335832608833, "rms_envelope_nrmse": 0.2067790423098858, "envelope_correlation": 0.9811160950929382, "rms_gain_error_db": 1.0270002687929685, "reference_mean_power": 0.000101477720422738, "output_mean_power": 0.0001285496000036306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 8.361033224981284e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.73677738976729e-05, "output_power": 3.5509969179194996e-05, "reference_active": true, "delta_db": 1.1311082214376258}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1422451661963211e-05, "output_power": 1.3616956217236292e-05, "reference_active": true, "delta_db": 0.76320712207456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.10199254010919e-06, "output_power": 9.273819449984604e-07, "reference_active": true, "delta_db": -0.7492001813268223}], "active_band_count": 3, "active_band_mae_db": 0.8811718416130027}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.7895186903062846e-05, "log_rms_mae_db": 1.8581292737707136, "rms_envelope_nrmse": 0.17817412435592372, "envelope_correlation": 0.9880366187859074, "rms_gain_error_db": 1.0269994126773887, "reference_mean_power": 0.00010203935243697765, "output_mean_power": 0.0001292610368069936}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 7.765913024651851e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5071474654741866e-05, "output_power": 3.235676672826431e-05, "reference_active": true, "delta_db": 1.1078523888412493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.067872704197783e-06, "output_power": 7.229064790907684e-06, "reference_active": true, "delta_db": 0.7604565585703592}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0845180908068795e-07, "output_power": 8.850855467384106e-08, "reference_active": true, "delta_db": -0.8825155199275869}], "active_band_count": 3, "active_band_mae_db": 0.9169414891130652}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 3.774086107813168e-05, "log_rms_mae_db": 1.707229678220966, "rms_envelope_nrmse": 0.25277039836434895, "envelope_correlation": 0.9735599405977656, "rms_gain_error_db": 1.1433195878696774, "reference_mean_power": 4.068228265632881e-05, "output_mean_power": 5.2934311863354604e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 2.809144227115543e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.649297963098168e-06, "output_power": 1.191511854608372e-05, "reference_active": true, "delta_db": 1.3911750904487585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.584218334576925e-06, "output_power": 3.348530544676259e-06, "reference_active": true, "delta_db": 1.1252506144366508}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3501889520111653e-06, "output_power": 1.279947437522428e-06, "reference_active": true, "delta_db": -0.23202414807054222}], "active_band_count": 3, "active_band_mae_db": 0.9161499509853172}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 3.332899304538301e-05, "log_rms_mae_db": 1.2611277480467669, "rms_envelope_nrmse": 0.18976128774809217, "envelope_correlation": 0.9896872742656379, "rms_gain_error_db": 1.1433239301132034, "reference_mean_power": 4.0907280378011884e-05, "output_mean_power": 5.322712396337615e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 2.6173037083412648e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.99444690144364e-06, "output_power": 1.1059257518253631e-05, "reference_active": true, "delta_db": 1.4093754855251792}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1581434874555947e-06, "output_power": 1.4917838596215827e-06, "reference_active": true, "delta_db": 1.0994353462021758}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.731926648355553e-08, "output_power": 7.402761278583426e-08, "reference_active": true, "delta_db": -0.7171633416112535}], "active_band_count": 3, "active_band_mae_db": 1.0753247244462028}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1090, "rms_floor": 1.8891386922351935e-05, "log_rms_mae_db": 1.6474546379436537, "rms_envelope_nrmse": 0.359989214862954, "envelope_correlation": 0.9471876488708637, "rms_gain_error_db": 1.4819041489787375, "reference_mean_power": 6.700122358829966e-06, "output_mean_power": 9.424821828434844e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1090, "frequency_resolution_hz": 0.18348623853211007, "power_floor": 5.001365064934297e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.184828883572986e-06, "output_power": 3.1447675099487215e-06, "reference_active": true, "delta_db": 1.5817111536800637}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0698658461156927e-06, "output_power": 1.8254253396648932e-06, "reference_active": true, "delta_db": 2.3203475136871226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.241288004170751e-07, "output_power": 8.416291956614071e-07, "reference_active": true, "delta_db": 2.0568276805308856}], "active_band_count": 3, "active_band_mae_db": 1.9862954492993572}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 87280, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1084, "rms_floor": 1.2324616345992443e-05, "log_rms_mae_db": 1.1671355080917756, "rms_envelope_nrmse": 0.26745005491065293, "envelope_correlation": 0.9782462294693193, "rms_gain_error_db": 1.4819757820188761, "reference_mean_power": 6.7368909993733856e-06, "output_mean_power": 9.476699264911498e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1084, "frequency_resolution_hz": 0.18450184501845018, "power_floor": 4.500371991965458e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.395916806498352e-06, "output_power": 3.6560168296688695e-06, "reference_active": true, "delta_db": 1.8353645228861155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.187368503320664e-07, "output_power": 1.0626542426987355e-06, "reference_active": true, "delta_db": 2.3488599828010788}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.878047962583857e-08, "output_power": 6.636447551749829e-08, "reference_active": true, "delta_db": 1.3368960036157378}], "active_band_count": 3, "active_band_mae_db": 1.8403735031009774}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8710915954515703, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4cc0a15ff9314725427245228f0df9b2411a2bce7270d8d310674d18b8a7656a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9735599405977656, "standard_deviation": null, "minimum": 0.9735599405977656, "maximum": 0.9735599405977656, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.707229678220966, "standard_deviation": null, "minimum": 1.707229678220966, "maximum": 1.707229678220966, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9161499509853172, "standard_deviation": null, "minimum": 0.9161499509853172, "maximum": 0.9161499509853172, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25277039836434895, "standard_deviation": null, "minimum": 0.25277039836434895, "maximum": 0.25277039836434895, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9896872742656379, "standard_deviation": null, "minimum": 0.9896872742656379, "maximum": 0.9896872742656379, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2611277480467669, "standard_deviation": null, "minimum": 1.2611277480467669, "maximum": 1.2611277480467669, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0753247244462028, "standard_deviation": null, "minimum": 1.0753247244462028, "maximum": 1.0753247244462028, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18976128774809217, "standard_deviation": null, "minimum": 0.18976128774809217, "maximum": 0.18976128774809217, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9471876488708637, "standard_deviation": null, "minimum": 0.9471876488708637, "maximum": 0.9471876488708637, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6474546379436537, "standard_deviation": null, "minimum": 1.6474546379436537, "maximum": 1.6474546379436537, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.9862954492993572, "standard_deviation": null, "minimum": 1.9862954492993572, "maximum": 1.9862954492993572, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.359989214862954, "standard_deviation": null, "minimum": 0.359989214862954, "maximum": 0.359989214862954, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9782462294693193, "standard_deviation": null, "minimum": 0.9782462294693193, "maximum": 0.9782462294693193, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1671355080917756, "standard_deviation": null, "minimum": 1.1671355080917756, "maximum": 1.1671355080917756, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.8403735031009774, "standard_deviation": null, "minimum": 1.8403735031009774, "maximum": 1.8403735031009774, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.26745005491065293, "standard_deviation": null, "minimum": 0.26745005491065293, "maximum": 0.26745005491065293, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9811160950929382, "standard_deviation": null, "minimum": 0.9811160950929382, "maximum": 0.9811160950929382, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.155335832608833, "standard_deviation": null, "minimum": 2.155335832608833, "maximum": 2.155335832608833, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8811718416130027, "standard_deviation": null, "minimum": 0.8811718416130027, "maximum": 0.8811718416130027, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2067790423098858, "standard_deviation": null, "minimum": 0.2067790423098858, "maximum": 0.2067790423098858, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9880366187859074, "standard_deviation": null, "minimum": 0.9880366187859074, "maximum": 0.9880366187859074, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.8581292737707136, "standard_deviation": null, "minimum": 1.8581292737707136, "maximum": 1.8581292737707136, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9169414891130652, "standard_deviation": null, "minimum": 0.9169414891130652, "maximum": 0.9169414891130652, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17817412435592372, "standard_deviation": null, "minimum": 0.17817412435592372, "maximum": 0.17817412435592372, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7946518236943906, "standard_deviation": null, "minimum": 0.7946518236943906, "maximum": 0.7946518236943906, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.124047430994723, "standard_deviation": null, "minimum": 6.124047430994723, "maximum": 6.124047430994723, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.17061504377943698, "standard_deviation": null, "minimum": 0.17061504377943698, "maximum": 0.17061504377943698, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7618805801095283, "standard_deviation": null, "minimum": 0.7618805801095283, "maximum": 0.7618805801095283, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9833764607501585, "standard_deviation": null, "minimum": 0.9833764607501585, "maximum": 0.9833764607501585, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.5529654022209405, "standard_deviation": null, "minimum": 2.5529654022209405, "maximum": 2.5529654022209405, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.5186826178104532, "standard_deviation": null, "minimum": 0.5186826178104532, "maximum": 0.5186826178104532, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16440355771850168, "standard_deviation": null, "minimum": 0.16440355771850168, "maximum": 0.16440355771850168, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9926664393698768, "standard_deviation": null, "minimum": 0.9926664393698768, "maximum": 0.9926664393698768, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.2861498123606276, "standard_deviation": null, "minimum": 2.2861498123606276, "maximum": 2.2861498123606276, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8240407649979566, "standard_deviation": null, "minimum": 0.8240407649979566, "maximum": 0.8240407649979566, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1261375005891387, "standard_deviation": null, "minimum": 0.1261375005891387, "maximum": 0.1261375005891387, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5683"}, {"case": "librispeech_5683-32879-0009", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.7994840145111084, "predicted_mos": 2.7994840145111084, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 82160, "input_sample_rate": 16000, "input_seconds": 5.135, "input_float32_sha256": "dd78506d9df3543937ce2a0b1e39e743ec696b0c2e3bc5c6d64f455f5571da96", "internal_samples": 82160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.599890599958599}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5682592979827994, "全帯域_10ms_rms_envelope_nrmse": 0.17681254476843689, "全帯域_10ms_envelope_correlation": 0.9932212954777774, "全帯域_10ms_modulation_mae_db": 1.5105968977088817, "全帯域_40ms_log_rms_mae_db": 1.3693977748633348, "全帯域_40ms_rms_envelope_nrmse": 0.16348514433917674, "全帯域_40ms_envelope_correlation": 0.9963708230182586, "全帯域_40ms_modulation_mae_db": 1.658853424662441, "300–1000 Hz_10ms_log_rms_mae_db": 1.7436284567375646, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20173709422722497, "300–1000 Hz_10ms_envelope_correlation": 0.9905290342404652, "300–1000 Hz_10ms_modulation_mae_db": 0.6614722872958232, "300–1000 Hz_40ms_log_rms_mae_db": 1.5101584629834295, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.19033245581019143, "300–1000 Hz_40ms_envelope_correlation": 0.9931109713734342, "300–1000 Hz_40ms_modulation_mae_db": 0.5356593813772111, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4587031816196956, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19270082611021777, "1000–3000 Hz_10ms_envelope_correlation": 0.9954830678170365, "1000–3000 Hz_10ms_modulation_mae_db": 2.061674440427024, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1289863045633897, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1803022127767937, "1000–3000 Hz_40ms_envelope_correlation": 0.9978508172564613, "1000–3000 Hz_40ms_modulation_mae_db": 1.9081699456179126, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4591688214808782, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20274886945877554, "3000–6000 Hz_10ms_envelope_correlation": 0.9731895832844771, "3000–6000 Hz_10ms_modulation_mae_db": 0.46937787977287554, "3000–6000 Hz_40ms_log_rms_mae_db": 1.030806568004425, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12371526392748528, "3000–6000 Hz_40ms_envelope_correlation": 0.9900991587157704, "3000–6000 Hz_40ms_modulation_mae_db": 0.5810982673032535, "mr_stft_db": 5.625322652805832, "band_mae_db": 0.9258281017820377, "rms_gain_error_db": 1.1996137267960787, "output_peak": 0.9695785399929093}, "full_scale": {"sample_count": 82160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 82160, "output_samples": 82160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.650076918262663, "spectral_convergence": 0.33608170485376376, "spectral_convergence_status": "ok", "amplitude_floor": 2.2676184880983433e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.463268279007345, "spectral_convergence": 0.27555914787358715, "spectral_convergence_status": "ok", "amplitude_floor": 1.9761500134157035e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.76262276114749, "spectral_convergence": 0.29766026600908574, "spectral_convergence_status": "ok", "amplitude_floor": 1.298774161363914e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.367365175011865e-05, "output_power": 0.00010267572824034902, "delta_db": 0.8887906860969688, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016048781148481498, "output_power": 0.0022265436265175927, "delta_db": 1.4218915427018883, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001327868811244639, "output_power": 0.001685204945607457, "delta_db": 1.0349755460184802, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0026813637139198997, "output_power": 0.0035927933743408746, "delta_db": 1.2707651324155698, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003087131182449005, "output_power": 0.0003674527212322227, "delta_db": 0.7564638347956385, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00020800689998023804, "output_power": 0.00021691317359360672, "delta_db": 0.18208186866368004, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9258281017820377, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1996137267960787, "pitch": null, "output_rms_dbfs": -20.86632133767072, "output_peak": 0.9695785399929093, "reference_rms_dbfs": -22.065935064466796, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.625322652805832}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 82160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003747956436237944, "log_rms_mae_db": 1.5682592979827994, "rms_envelope_nrmse": 0.17681254476843689, "envelope_correlation": 0.9932212954777774, "rms_gain_error_db": 1.199611012768921, "reference_mean_power": 0.006220557891926456, "output_mean_power": 0.008199557910834574}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.814448626587233e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017308539978626828, "output_power": 0.0022863736257499314, "reference_active": true, "delta_db": 1.2088676618624392}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0006400481816254359, "output_power": 0.0008002699618365717, "reference_active": true, "delta_db": 0.9702384785230638}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.4706618104713675e-05, "output_power": 7.684936327544837e-05, "reference_active": true, "delta_db": 2.3526845527411413}], "active_band_count": 3, "active_band_mae_db": 1.5105968977088817}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.00031163820648484283, "log_rms_mae_db": 1.3693977748633348, "rms_envelope_nrmse": 0.16348514433917674, "envelope_correlation": 0.9963708230182586, "rms_gain_error_db": 1.1996013932218246, "reference_mean_power": 0.006257088451954998, "output_mean_power": 0.00824769198615719}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 6.546091034740867e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001677353066756132, "output_power": 0.002226558237134797, "reference_active": true, "delta_db": 1.230095718238464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00035132010859926165, "output_power": 0.00043502169550935346, "reference_active": true, "delta_db": 0.9280790840810915}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3827329510863976e-06, "output_power": 4.559465332494035e-06, "reference_active": true, "delta_db": 2.818385471667767}], "active_band_count": 3, "active_band_mae_db": 1.658853424662441}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.00015152624573583172, "log_rms_mae_db": 1.7436284567375646, "rms_envelope_nrmse": 0.20173709422722497, "envelope_correlation": 0.9905290342404652, "rms_gain_error_db": 1.3067999535885935, "reference_mean_power": 0.001672933915423699, "output_mean_power": 0.0022602619866159876}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 1.623496524404657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006256720367801861, "output_power": 0.0007959262663914125, "reference_active": true, "delta_db": 1.045260916596479}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015725509039345338, "output_power": 0.00018328327732791867, "reference_active": true, "delta_db": 0.665181293795692}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4941700779965039e-05, "output_power": 1.5914664961331197e-05, "reference_active": true, "delta_db": 0.2739746514952985}], "active_band_count": 3, "active_band_mae_db": 0.6614722872958232}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0001385613366624899, "log_rms_mae_db": 1.5101584629834295, "rms_envelope_nrmse": 0.19033245581019143, "envelope_correlation": 0.9931109713734342, "rms_gain_error_db": 1.3067999151595444, "reference_mean_power": 0.0016827742497549935, "output_mean_power": 0.0022735570126560215}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 1.5301470087365327e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005852793815004822, "output_power": 0.0007461710006646268, "reference_active": true, "delta_db": 1.054751416755291}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.151264176170612e-05, "output_power": 9.076741084123163e-05, "reference_active": true, "delta_db": 0.4670497877001286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3333886770336671e-06, "output_power": 1.3597981838685357e-06, "reference_active": true, "delta_db": 0.08517693967621391}], "active_band_count": 3, "active_band_mae_db": 0.5356593813772111}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 0.0003048971769130071, "log_rms_mae_db": 1.4587031816196956, "rms_envelope_nrmse": 0.19270082611021777, "envelope_correlation": 0.9954830678170365, "rms_gain_error_db": 1.3516184076188305, "reference_mean_power": 0.0025898505670984154, "output_mean_power": 0.003535383632925737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 4.739471399174337e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014413951986732973, "output_power": 0.0020374400644404154, "reference_active": true, "delta_db": 1.5030177084059582}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0004245147064825657, "output_power": 0.0005860796172838058, "reference_active": true, "delta_db": 1.4006387759730663}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3891971817365186e-05, "output_power": 5.086144300108242e-05, "reference_active": true, "delta_db": 3.2813668369020483}], "active_band_count": 3, "active_band_mae_db": 2.061674440427024}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 0.0002474836728453422, "log_rms_mae_db": 1.1289863045633897, "rms_envelope_nrmse": 0.1803022127767937, "envelope_correlation": 0.9978508172564613, "rms_gain_error_db": 1.3515932647369435, "reference_mean_power": 0.002605032676611572, "output_mean_power": 0.003556088017769176}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 4.638343817264169e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014427582782272192, "output_power": 0.0020401955692083483, "reference_active": true, "delta_db": 1.5047822517857723}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020918188683306615, "output_power": 0.0002869649373482219, "reference_active": true, "delta_db": 1.37304759903802}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0082267328987986e-06, "output_power": 3.867949030500676e-06, "reference_active": true, "delta_db": 2.846679986029945}], "active_band_count": 3, "active_band_mae_db": 1.9081699456179126}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1026, "rms_floor": 8.624763428459013e-05, "log_rms_mae_db": 1.4591688214808782, "rms_envelope_nrmse": 0.20274886945877554, "envelope_correlation": 0.9731895832844771, "rms_gain_error_db": 0.3488779167465498, "reference_mean_power": 0.00014375800668184303, "output_mean_power": 0.00015578291783788198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1026, "frequency_resolution_hz": 0.19493177387914232, "power_floor": 6.640530401894145e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.056434359733932e-05, "output_power": 2.4272431487812362e-05, "reference_active": true, "delta_db": 0.7199843228416509}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.844428212989113e-05, "output_power": 2.1012196651268408e-05, "reference_active": true, "delta_db": 0.5660969996186639}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.64438001039378e-06, "output_power": 3.748252916012812e-06, "reference_active": true, "delta_db": 0.12205231685831189}], "active_band_count": 3, "active_band_mae_db": 0.46937787977287554}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 82160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1020, "rms_floor": 6.581694648038094e-05, "log_rms_mae_db": 1.030806568004425, "rms_envelope_nrmse": 0.12371526392748528, "envelope_correlation": 0.9900991587157704, "rms_gain_error_db": 0.3489097538248276, "reference_mean_power": 0.00014459942402382801, "output_mean_power": 0.00015669586582476889}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1020, "frequency_resolution_hz": 0.196078431372549, "power_floor": 5.765239137842014e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0474191785118657e-05, "output_power": 2.373815199606146e-05, "reference_active": true, "delta_db": 0.6424013923633642}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1083544144974636e-05, "output_power": 1.2695599354652083e-05, "reference_active": true, "delta_db": 0.589745531968964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.842442602727836e-07, "output_power": 2.526835187303909e-07, "reference_active": true, "delta_db": -0.5111478775774324}], "active_band_count": 3, "active_band_mae_db": 0.5810982673032535}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9073019677691212, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7a857a284f1f644ffaf2dc69dc695cf247f7cee381ee284582271abf1e935134", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9954830678170365, "standard_deviation": null, "minimum": 0.9954830678170365, "maximum": 0.9954830678170365, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4587031816196956, "standard_deviation": null, "minimum": 1.4587031816196956, "maximum": 1.4587031816196956, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 2.061674440427024, "standard_deviation": null, "minimum": 2.061674440427024, "maximum": 2.061674440427024, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19270082611021777, "standard_deviation": null, "minimum": 0.19270082611021777, "maximum": 0.19270082611021777, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9978508172564613, "standard_deviation": null, "minimum": 0.9978508172564613, "maximum": 0.9978508172564613, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1289863045633897, "standard_deviation": null, "minimum": 1.1289863045633897, "maximum": 1.1289863045633897, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.9081699456179126, "standard_deviation": null, "minimum": 1.9081699456179126, "maximum": 1.9081699456179126, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1803022127767937, "standard_deviation": null, "minimum": 0.1803022127767937, "maximum": 0.1803022127767937, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9731895832844771, "standard_deviation": null, "minimum": 0.9731895832844771, "maximum": 0.9731895832844771, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4591688214808782, "standard_deviation": null, "minimum": 1.4591688214808782, "maximum": 1.4591688214808782, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.46937787977287554, "standard_deviation": null, "minimum": 0.46937787977287554, "maximum": 0.46937787977287554, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20274886945877554, "standard_deviation": null, "minimum": 0.20274886945877554, "maximum": 0.20274886945877554, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900991587157704, "standard_deviation": null, "minimum": 0.9900991587157704, "maximum": 0.9900991587157704, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.030806568004425, "standard_deviation": null, "minimum": 1.030806568004425, "maximum": 1.030806568004425, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5810982673032535, "standard_deviation": null, "minimum": 0.5810982673032535, "maximum": 0.5810982673032535, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12371526392748528, "standard_deviation": null, "minimum": 0.12371526392748528, "maximum": 0.12371526392748528, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9905290342404652, "standard_deviation": null, "minimum": 0.9905290342404652, "maximum": 0.9905290342404652, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7436284567375646, "standard_deviation": null, "minimum": 1.7436284567375646, "maximum": 1.7436284567375646, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6614722872958232, "standard_deviation": null, "minimum": 0.6614722872958232, "maximum": 0.6614722872958232, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20173709422722497, "standard_deviation": null, "minimum": 0.20173709422722497, "maximum": 0.20173709422722497, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9931109713734342, "standard_deviation": null, "minimum": 0.9931109713734342, "maximum": 0.9931109713734342, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5101584629834295, "standard_deviation": null, "minimum": 1.5101584629834295, "maximum": 1.5101584629834295, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.5356593813772111, "standard_deviation": null, "minimum": 0.5356593813772111, "maximum": 0.5356593813772111, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19033245581019143, "standard_deviation": null, "minimum": 0.19033245581019143, "maximum": 0.19033245581019143, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9258281017820377, "standard_deviation": null, "minimum": 0.9258281017820377, "maximum": 0.9258281017820377, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.625322652805832, "standard_deviation": null, "minimum": 5.625322652805832, "maximum": 5.625322652805832, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.9695785399929093, "standard_deviation": null, "minimum": 0.9695785399929093, "maximum": 0.9695785399929093, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1996137267960787, "standard_deviation": null, "minimum": 1.1996137267960787, "maximum": 1.1996137267960787, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9932212954777774, "standard_deviation": null, "minimum": 0.9932212954777774, "maximum": 0.9932212954777774, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5682592979827994, "standard_deviation": null, "minimum": 1.5682592979827994, "maximum": 1.5682592979827994, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.5105968977088817, "standard_deviation": null, "minimum": 1.5105968977088817, "maximum": 1.5105968977088817, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17681254476843689, "standard_deviation": null, "minimum": 0.17681254476843689, "maximum": 0.17681254476843689, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9963708230182586, "standard_deviation": null, "minimum": 0.9963708230182586, "maximum": 0.9963708230182586, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.3693977748633348, "standard_deviation": null, "minimum": 1.3693977748633348, "maximum": 1.3693977748633348, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.658853424662441, "standard_deviation": null, "minimum": 1.658853424662441, "maximum": 1.658853424662441, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16348514433917674, "standard_deviation": null, "minimum": 0.16348514433917674, "maximum": 0.16348514433917674, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5683"}, {"case": "librispeech_5683-32879-0019", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.7210288047790527, "predicted_mos": 2.7210288047790527, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "e642e74fabaa12db20970a7cc9487ceb37404af4477879eb411f2470ff812f3b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.866061800159514}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5319135546733968, "全帯域_10ms_rms_envelope_nrmse": 0.24732217596732523, "全帯域_10ms_envelope_correlation": 0.9831905414531211, "全帯域_10ms_modulation_mae_db": 1.3359471812232966, "全帯域_40ms_log_rms_mae_db": 1.2834973629088915, "全帯域_40ms_rms_envelope_nrmse": 0.2327953866708903, "全帯域_40ms_envelope_correlation": 0.9876477121370583, "全帯域_40ms_modulation_mae_db": 1.1061037434219079, "300–1000 Hz_10ms_log_rms_mae_db": 1.6980686663626579, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.28747947000133117, "300–1000 Hz_10ms_envelope_correlation": 0.9837480002451614, "300–1000 Hz_10ms_modulation_mae_db": 1.4598032574740802, "300–1000 Hz_40ms_log_rms_mae_db": 1.4215113008694098, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2774576427857794, "300–1000 Hz_40ms_envelope_correlation": 0.9865899197179276, "300–1000 Hz_40ms_modulation_mae_db": 1.3432031786168264, "1000–3000 Hz_10ms_log_rms_mae_db": 1.470031841776833, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17990152041227542, "1000–3000 Hz_10ms_envelope_correlation": 0.9894970744707565, "1000–3000 Hz_10ms_modulation_mae_db": 0.8695926132122002, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1856509445341676, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1519106539213596, "1000–3000 Hz_40ms_envelope_correlation": 0.9949550692148204, "1000–3000 Hz_40ms_modulation_mae_db": 0.7688685648612257, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4044125235826563, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19029105386764344, "3000–6000 Hz_10ms_envelope_correlation": 0.974097156465978, "3000–6000 Hz_10ms_modulation_mae_db": 0.5289776088812951, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0148002162600762, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11337337590541277, "3000–6000 Hz_40ms_envelope_correlation": 0.991756008010168, "3000–6000 Hz_40ms_modulation_mae_db": 0.32192570157044004, "mr_stft_db": 5.752477608914148, "band_mae_db": 0.9783008960679217, "rms_gain_error_db": 1.5505421764026344, "output_peak": 0.9941412019984518}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.697058389653356, "spectral_convergence": 0.44675914032498265, "spectral_convergence_status": "ok", "amplitude_floor": 2.005860984254184e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.569033188352327, "spectral_convergence": 0.34262927012954525, "spectral_convergence_status": "ok", "amplitude_floor": 1.6673734947378395e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.991341248736763, "spectral_convergence": 0.37654128732580405, "spectral_convergence_status": "ok", "amplitude_floor": 1.455049241978443e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 3.2140486659240935e-05, "output_power": 3.2941538503792266e-05, "delta_db": 0.10691430223705048, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002187325069648938, "output_power": 0.003670328397228752, "delta_db": 2.247915932750202, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018573993391017246, "output_power": 0.0025551329939206553, "delta_db": 1.3850822327731844, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.002022873257906857, "output_power": 0.0025311658081899157, "delta_db": 0.9735192214250064, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00023141111513331255, "output_power": 0.00027138209020165195, "delta_db": 0.6919696796621928, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.19984427950361e-05, "output_power": 4.673837285328194e-05, "delta_db": 0.4644040075598946, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9783008960679217, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.5505421764026344, "pitch": null, "output_rms_dbfs": -20.405917980654582, "output_peak": 0.9941412019984518, "reference_rms_dbfs": -21.95646015705721, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.752477608914148}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003217977906320387, "log_rms_mae_db": 1.5319135546733968, "rms_envelope_nrmse": 0.24732217596732523, "envelope_correlation": 0.9831905414531211, "rms_gain_error_db": 1.5505847955025995, "reference_mean_power": 0.006378025340653872, "output_mean_power": 0.009114749132573527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.934428693189062e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0026409373563640536, "output_power": 0.004448287755594087, "reference_active": true, "delta_db": 2.2643477382347323}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022353254716419522, "output_power": 0.0002838415546451871, "reference_active": true, "delta_db": 1.037352099057771}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.171045723198585e-05, "output_power": 3.730926339784959e-05, "reference_active": true, "delta_db": 0.7061417063773865}], "active_band_count": 3, "active_band_mae_db": 1.3359471812232966}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00029448546471938644, "log_rms_mae_db": 1.2834973629088915, "rms_envelope_nrmse": 0.2327953866708903, "envelope_correlation": 0.9876477121370583, "rms_gain_error_db": 1.5505941774867584, "reference_mean_power": 0.006407121641291136, "output_mean_power": 0.009156350039737035}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7183045790225306e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0025278777543949896, "output_power": 0.004261397314868041, "reference_active": true, "delta_db": 2.2679595981248384}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010431651037251879, "output_power": 0.00013223769897077528, "reference_active": true, "delta_db": 1.030022331184897}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4474914310652027e-06, "output_power": 2.4360615096269084e-06, "reference_active": true, "delta_db": -0.02032930095598813}], "active_band_count": 3, "active_band_mae_db": 1.1061037434219079}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002169046697405618, "log_rms_mae_db": 1.6980686663626579, "rms_envelope_nrmse": 0.28747947000133117, "envelope_correlation": 0.9837480002451614, "rms_gain_error_db": 1.8495059621922292, "reference_mean_power": 0.00266513165861851, "output_mean_power": 0.00408008550396644}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9887000298118345e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014141653900070267, "output_power": 0.002550454699742744, "reference_active": true, "delta_db": 2.5611740994945196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010731466350302803, "output_power": 0.00015186718729502685, "reference_active": true, "delta_db": 1.5080488140098234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0173327828420217e-05, "output_power": 2.166687271510038e-05, "reference_active": true, "delta_db": 0.3101868589178971}], "active_band_count": 3, "active_band_mae_db": 1.4598032574740802}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020963310569053025, "log_rms_mae_db": 1.4215113008694098, "rms_envelope_nrmse": 0.2774576427857794, "envelope_correlation": 0.9865899197179276, "rms_gain_error_db": 1.8492658503052688, "reference_mean_power": 0.0026772178601939312, "output_mean_power": 0.004098361835017174}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.883939001913369e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0013521293519634116, "output_power": 0.002435355541489054, "reference_active": true, "delta_db": 2.555441329940316}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.792726117783695e-05, "output_power": 6.681956507830682e-05, "reference_active": true, "delta_db": 1.4432103276094743}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6850826685161704e-06, "output_power": 1.6971373822777615e-06, "reference_active": true, "delta_db": 0.030957878300688736}], "active_band_count": 3, "active_band_mae_db": 1.3432031786168264}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002744641361828464, "log_rms_mae_db": 1.470031841776833, "rms_envelope_nrmse": 0.17990152041227542, "envelope_correlation": 0.9894970744707565, "rms_gain_error_db": 1.0153450652879856, "reference_mean_power": 0.0019381811578355007, "output_mean_power": 0.0024486621812608595}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1196235236571854e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006735796402683569, "output_power": 0.0009281744608790827, "reference_active": true, "delta_db": 1.3924066281545613}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013823323993445422, "output_power": 0.00016618809043928985, "reference_active": true, "delta_db": 0.7998741044774017}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8792319908524634e-05, "output_power": 2.0683789423868822e-05, "reference_active": true, "delta_db": 0.4164971070046377}], "active_band_count": 3, "active_band_mae_db": 0.8695926132122002}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002445180716567873, "log_rms_mae_db": 1.1856509445341676, "rms_envelope_nrmse": 0.1519106539213596, "envelope_correlation": 0.9949550692148204, "rms_gain_error_db": 1.015362554971633, "reference_mean_power": 0.0019467823981069799, "output_mean_power": 0.0024595387338395692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0571732642454036e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006649824803295284, "output_power": 0.0009128053892653649, "reference_active": true, "delta_db": 1.3756799201806431}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.927173921912536e-05, "output_power": 8.086620796539621e-05, "reference_active": true, "delta_db": 0.6721098665810634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7481239585012505e-06, "output_power": 1.8554694824583292e-06, "reference_active": true, "delta_db": 0.2588159078219706}], "active_band_count": 3, "active_band_mae_db": 0.7688685648612257}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.6882736857616036e-05, "log_rms_mae_db": 1.4044125235826563, "rms_envelope_nrmse": 0.19029105386764344, "envelope_correlation": 0.974097156465978, "rms_gain_error_db": 0.5019943636736115, "reference_mean_power": 6.564824297759055e-05, "output_mean_power": 7.369237329752712e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.001422573299629e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.567613726314846e-05, "output_power": 1.723259074060223e-05, "reference_active": true, "delta_db": 0.41111516424936795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.79643479102731e-06, "output_power": 6.237860411237682e-06, "reference_active": true, "delta_db": 0.3187469754856326}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.958485457614687e-07, "output_power": 1.2131105012315582e-06, "reference_active": true, "delta_db": 0.8570706869088847}], "active_band_count": 3, "active_band_mae_db": 0.5289776088812951}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.766393520730197e-05, "log_rms_mae_db": 1.0148002162600762, "rms_envelope_nrmse": 0.11337337590541277, "envelope_correlation": 0.991756008010168, "rms_gain_error_db": 0.5014176238074595, "reference_mean_power": 6.585532886735065e-05, "output_mean_power": 7.391501770188611e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6817222442955538e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5480772645031266e-05, "output_power": 1.7026941234649417e-05, "reference_active": true, "delta_db": 0.4134400462881668}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.979992150286795e-06, "output_power": 3.159069318818766e-06, "reference_active": true, "delta_db": 0.25344035559853806}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.719812912358681e-08, "output_power": 9.341074532343523e-08, "reference_active": true, "delta_db": 0.2988967028246154}], "active_band_count": 3, "active_band_mae_db": 0.32192570157044004}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8922957539708867, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ce70ac62aa8502f78f12d3e86f71eb61853581e3222f3eba638686eddaef552c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9894970744707565, "standard_deviation": null, "minimum": 0.9894970744707565, "maximum": 0.9894970744707565, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.470031841776833, "standard_deviation": null, "minimum": 1.470031841776833, "maximum": 1.470031841776833, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8695926132122002, "standard_deviation": null, "minimum": 0.8695926132122002, "maximum": 0.8695926132122002, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17990152041227542, "standard_deviation": null, "minimum": 0.17990152041227542, "maximum": 0.17990152041227542, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949550692148204, "standard_deviation": null, "minimum": 0.9949550692148204, "maximum": 0.9949550692148204, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1856509445341676, "standard_deviation": null, "minimum": 1.1856509445341676, "maximum": 1.1856509445341676, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7688685648612257, "standard_deviation": null, "minimum": 0.7688685648612257, "maximum": 0.7688685648612257, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1519106539213596, "standard_deviation": null, "minimum": 0.1519106539213596, "maximum": 0.1519106539213596, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.974097156465978, "standard_deviation": null, "minimum": 0.974097156465978, "maximum": 0.974097156465978, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4044125235826563, "standard_deviation": null, "minimum": 1.4044125235826563, "maximum": 1.4044125235826563, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5289776088812951, "standard_deviation": null, "minimum": 0.5289776088812951, "maximum": 0.5289776088812951, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19029105386764344, "standard_deviation": null, "minimum": 0.19029105386764344, "maximum": 0.19029105386764344, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.991756008010168, "standard_deviation": null, "minimum": 0.991756008010168, "maximum": 0.991756008010168, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0148002162600762, "standard_deviation": null, "minimum": 1.0148002162600762, "maximum": 1.0148002162600762, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.32192570157044004, "standard_deviation": null, "minimum": 0.32192570157044004, "maximum": 0.32192570157044004, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11337337590541277, "standard_deviation": null, "minimum": 0.11337337590541277, "maximum": 0.11337337590541277, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9837480002451614, "standard_deviation": null, "minimum": 0.9837480002451614, "maximum": 0.9837480002451614, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6980686663626579, "standard_deviation": null, "minimum": 1.6980686663626579, "maximum": 1.6980686663626579, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4598032574740802, "standard_deviation": null, "minimum": 1.4598032574740802, "maximum": 1.4598032574740802, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.28747947000133117, "standard_deviation": null, "minimum": 0.28747947000133117, "maximum": 0.28747947000133117, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9865899197179276, "standard_deviation": null, "minimum": 0.9865899197179276, "maximum": 0.9865899197179276, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4215113008694098, "standard_deviation": null, "minimum": 1.4215113008694098, "maximum": 1.4215113008694098, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3432031786168264, "standard_deviation": null, "minimum": 1.3432031786168264, "maximum": 1.3432031786168264, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2774576427857794, "standard_deviation": null, "minimum": 0.2774576427857794, "maximum": 0.2774576427857794, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9783008960679217, "standard_deviation": null, "minimum": 0.9783008960679217, "maximum": 0.9783008960679217, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.752477608914148, "standard_deviation": null, "minimum": 5.752477608914148, "maximum": 5.752477608914148, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.9941412019984518, "standard_deviation": null, "minimum": 0.9941412019984518, "maximum": 0.9941412019984518, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.5505421764026344, "standard_deviation": null, "minimum": 1.5505421764026344, "maximum": 1.5505421764026344, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9831905414531211, "standard_deviation": null, "minimum": 0.9831905414531211, "maximum": 0.9831905414531211, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5319135546733968, "standard_deviation": null, "minimum": 1.5319135546733968, "maximum": 1.5319135546733968, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.3359471812232966, "standard_deviation": null, "minimum": 1.3359471812232966, "maximum": 1.3359471812232966, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.24732217596732523, "standard_deviation": null, "minimum": 0.24732217596732523, "maximum": 0.24732217596732523, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9876477121370583, "standard_deviation": null, "minimum": 0.9876477121370583, "maximum": 0.9876477121370583, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2834973629088915, "standard_deviation": null, "minimum": 1.2834973629088915, "maximum": 1.2834973629088915, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1061037434219079, "standard_deviation": null, "minimum": 1.1061037434219079, "maximum": 1.1061037434219079, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2327953866708903, "standard_deviation": null, "minimum": 0.2327953866708903, "maximum": 0.2327953866708903, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_5683"}, {"case": "librispeech_61-70968-0042", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9082229137420654, "predicted_mos": 2.9082229137420654, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44560, "input_sample_rate": 16000, "input_seconds": 2.785, "input_float32_sha256": "064de4e6da5de549ee610690b856ae1d7e4c13c792de0872d51adc1ffc52e0a8", "internal_samples": 44560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 12.339299300219864}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.5193609137313575, "全帯域_10ms_rms_envelope_nrmse": 0.1793303753873902, "全帯域_10ms_envelope_correlation": 0.9597023931264064, "全帯域_10ms_modulation_mae_db": 0.29412602683420025, "全帯域_40ms_log_rms_mae_db": 1.7038308029659055, "全帯域_40ms_rms_envelope_nrmse": 0.08857553236650542, "全帯域_40ms_envelope_correlation": 0.9900274431891283, "全帯域_40ms_modulation_mae_db": 0.36033887947309323, "300–1000 Hz_10ms_log_rms_mae_db": 2.750755084354002, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19614061148242645, "300–1000 Hz_10ms_envelope_correlation": 0.9661775004325196, "300–1000 Hz_10ms_modulation_mae_db": 0.9137629916226176, "300–1000 Hz_40ms_log_rms_mae_db": 1.9245163135812344, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10841805816544639, "300–1000 Hz_40ms_envelope_correlation": 0.9892418796032276, "300–1000 Hz_40ms_modulation_mae_db": 0.9181829216326985, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6860825873690806, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19216768542088855, "1000–3000 Hz_10ms_envelope_correlation": 0.9668656341634102, "1000–3000 Hz_10ms_modulation_mae_db": 0.6459458073352863, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8697497320426328, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10315414288977966, "1000–3000 Hz_40ms_envelope_correlation": 0.9919069816690508, "1000–3000 Hz_40ms_modulation_mae_db": 0.6053278612281847, "3000–6000 Hz_10ms_log_rms_mae_db": 1.896810805358268, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19404901516321596, "3000–6000 Hz_10ms_envelope_correlation": 0.9765493802766944, "3000–6000 Hz_10ms_modulation_mae_db": 1.2334306172938414, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0901471412273809, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09396365177170107, "3000–6000 Hz_40ms_envelope_correlation": 0.9942861655387055, "3000–6000 Hz_40ms_modulation_mae_db": 1.620936759022638, "mr_stft_db": 6.067990275347488, "band_mae_db": 0.27529916726560705, "rms_gain_error_db": 0.24926068355518488, "output_peak": 0.45457884369797963}, "full_scale": {"sample_count": 44560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44560, "output_samples": 44560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.990006639303959, "spectral_convergence": 0.8725202439449792, "spectral_convergence_status": "ok", "amplitude_floor": 9.278908430047933e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.547365155796152, "spectral_convergence": 0.3803320791658842, "spectral_convergence_status": "ok", "amplitude_floor": 4.74996132181228e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.666599030942356, "spectral_convergence": 0.31280931580844284, "spectral_convergence_status": "ok", "amplitude_floor": 3.301834166389807e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005909300513360037, "output_power": 0.0005849310605059309, "delta_db": -0.044313927769516065, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.000598279616407713, "output_power": 0.0006687780782524738, "delta_db": 0.4837782247753778, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005543526355345429, "output_power": 0.0005825084639720739, "delta_db": 0.2151612351837534, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00011395641530102642, "output_power": 0.00013046618593845615, "delta_db": 0.5875918707477049, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011253089498164014, "output_power": 0.00012077038007587428, "delta_db": 0.306886601501072, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6482875672349115e-05, "output_power": 1.6429587832430005e-05, "delta_db": -0.014063143616218075, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.27529916726560705, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.24926068355518488, "pitch": null, "output_rms_dbfs": -26.769782594192968, "output_peak": 0.45457884369797963, "reference_rms_dbfs": -27.019043277748153, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.067990275347488}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 0.00011061650326899306, "log_rms_mae_db": 2.5193609137313575, "rms_envelope_nrmse": 0.1793303753873902, "envelope_correlation": 0.9597023931264064, "rms_gain_error_db": 0.24936530516997868, "reference_mean_power": 0.00199005080866686, "output_mean_power": 0.002107660688141492}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 8.439637043999617e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005970864143567599, "output_power": 0.0006452000623105665, "reference_active": true, "delta_db": 0.3365721096627425}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010123102438186569, "output_power": 9.52642408988952e-05, "reference_active": true, "delta_db": -0.26383720416760803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.240115070482855e-05, "output_power": 3.0364319847293978e-05, "reference_active": true, "delta_db": -0.2819687666722501}], "active_band_count": 3, "active_band_mae_db": 0.29412602683420025}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 9.74750993349889e-05, "log_rms_mae_db": 1.7038308029659055, "rms_envelope_nrmse": 0.08857553236650542, "envelope_correlation": 0.9900274431891283, "rms_gain_error_db": 0.2498444167900034, "reference_mean_power": 0.0020114637230104894, "output_mean_power": 0.002130574113840838}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 6.839204647135984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005172552173341314, "output_power": 0.0005594917079668693, "reference_active": true, "delta_db": 0.34088774479506745}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6641602045504645e-05, "output_power": 5.698198689807871e-05, "reference_active": true, "delta_db": 0.02602060131905677}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.529860845744804e-06, "output_power": 2.1462771619331704e-06, "reference_active": true, "delta_db": -0.7141082923051555}], "active_band_count": 3, "active_band_mae_db": 0.36033887947309323}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 8.308955229004579e-05, "log_rms_mae_db": 2.750755084354002, "rms_envelope_nrmse": 0.19614061148242645, "envelope_correlation": 0.9661775004325196, "rms_gain_error_db": 0.30473865070676515, "reference_mean_power": 0.000839400894204417, "output_mean_power": 0.00090041618446988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.694834819211473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00042315114528166547, "output_power": 0.00044356595640130357, "reference_active": true, "delta_db": 0.20462686086939802}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.939611213191948e-05, "output_power": 6.309530298297938e-05, "reference_active": true, "delta_db": -0.41338109924091393}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4277196064029695e-05, "output_power": 1.4889167794136037e-05, "reference_active": true, "delta_db": -2.1232810147575414}], "active_band_count": 3, "active_band_mae_db": 0.9137629916226176}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 8.008999734930271e-05, "log_rms_mae_db": 1.9245163135812344, "rms_envelope_nrmse": 0.10841805816544639, "envelope_correlation": 0.9892418796032276, "rms_gain_error_db": 0.3046234975927755, "reference_mean_power": 0.0008485551352947845, "output_mean_power": 0.0009102117044797}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 5.733703583211585e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003764127527272362, "output_power": 0.0003935141550356805, "reference_active": true, "delta_db": 0.1929603017079773}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8963437868750806e-05, "output_power": 3.671137192421955e-05, "reference_active": true, "delta_db": -0.2585665445684593}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.296630500499659e-06, "output_power": 1.3514156263141167e-06, "reference_active": true, "delta_db": -2.303021918621659}], "active_band_count": 3, "active_band_mae_db": 0.9181829216326985}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.2528786526254916e-05, "log_rms_mae_db": 1.6860825873690806, "rms_envelope_nrmse": 0.19216768542088855, "envelope_correlation": 0.9668656341634102, "rms_gain_error_db": 0.5130061129958564, "reference_mean_power": 0.0001368277838750774, "output_mean_power": 0.0001539837543951797}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 6.073610600879068e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.932287229550182e-05, "output_power": 4.1592919717732885e-05, "reference_active": true, "delta_db": 0.24374174932406376}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.8616503346595646e-06, "output_power": 7.475860966509396e-06, "reference_active": true, "delta_db": 0.37232633009284505}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.317401673918684e-06, "output_power": 3.1845312962785203e-06, "reference_active": true, "delta_db": -1.32176934258895}], "active_band_count": 3, "active_band_mae_db": 0.6459458073352863}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 3.268734932259802e-05, "log_rms_mae_db": 0.8697497320426328, "rms_envelope_nrmse": 0.10315414288977966, "envelope_correlation": 0.9919069816690508, "rms_gain_error_db": 0.5130197037422766, "reference_mean_power": 0.00013831808192096785, "output_mean_power": 0.00015566139861207306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 4.85870788941741e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.5019378943818874e-05, "output_power": 3.695071851396909e-05, "reference_active": true, "delta_db": 0.23314447996175358}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.168829067205367e-06, "output_power": 4.56596832991496e-06, "reference_active": true, "delta_db": 0.39518806555656044}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163313597471507e-07, "output_power": 2.406451672219654e-07, "reference_active": true, "delta_db": -1.1876510381662402}], "active_band_count": 3, "active_band_mae_db": 0.6053278612281847}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 556, "rms_floor": 4.9606629692154165e-05, "log_rms_mae_db": 1.896810805358268, "rms_envelope_nrmse": 0.19404901516321596, "envelope_correlation": 0.9765493802766944, "rms_gain_error_db": -0.024429562133111087, "reference_mean_power": 5.1065139870557587e-05, "output_mean_power": 5.077869899544625e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 556, "frequency_resolution_hz": 0.3597122302158273, "power_floor": 2.3044643706212192e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0452875124351046e-05, "output_power": 1.0547174261288092e-05, "reference_active": true, "delta_db": 0.039003594070499886}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.855628387226494e-06, "output_power": 2.9027123463836376e-06, "reference_active": true, "delta_db": 0.07102309613486359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6522678437727886e-06, "output_power": 7.228595692250583e-07, "reference_active": true, "delta_db": -3.590265161676161}], "active_band_count": 3, "active_band_mae_db": 1.2334306172938414}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 550, "rms_floor": 4.1718716562101114e-05, "log_rms_mae_db": 1.0901471412273809, "rms_envelope_nrmse": 0.09396365177170107, "envelope_correlation": 0.9942861655387055, "rms_gain_error_db": -0.024424497855420457, "reference_mean_power": 5.162171435861935e-05, "output_mean_power": 5.13322113354042e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 550, "frequency_resolution_hz": 0.36363636363636365, "power_floor": 2.066669918587524e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.0052154991285675e-05, "output_power": 9.975590488642124e-06, "reference_active": true, "delta_db": -0.033205637678930276}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.890809551942203e-06, "output_power": 1.601858876607869e-06, "reference_active": true, "delta_db": -0.7202353540260465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8144313206293327e-07, "output_power": 7.043744682207344e-08, "reference_active": true, "delta_db": -4.109369285362937}], "active_band_count": 3, "active_band_mae_db": 1.620936759022638}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.902556167519126, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5eccc8acc656370651256f4928b0f658f982d9d9dfbc24469275226b8e9916eb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9668656341634102, "standard_deviation": null, "minimum": 0.9668656341634102, "maximum": 0.9668656341634102, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6860825873690806, "standard_deviation": null, "minimum": 1.6860825873690806, "maximum": 1.6860825873690806, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6459458073352863, "standard_deviation": null, "minimum": 0.6459458073352863, "maximum": 0.6459458073352863, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19216768542088855, "standard_deviation": null, "minimum": 0.19216768542088855, "maximum": 0.19216768542088855, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9919069816690508, "standard_deviation": null, "minimum": 0.9919069816690508, "maximum": 0.9919069816690508, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8697497320426328, "standard_deviation": null, "minimum": 0.8697497320426328, "maximum": 0.8697497320426328, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6053278612281847, "standard_deviation": null, "minimum": 0.6053278612281847, "maximum": 0.6053278612281847, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10315414288977966, "standard_deviation": null, "minimum": 0.10315414288977966, "maximum": 0.10315414288977966, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9765493802766944, "standard_deviation": null, "minimum": 0.9765493802766944, "maximum": 0.9765493802766944, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.896810805358268, "standard_deviation": null, "minimum": 1.896810805358268, "maximum": 1.896810805358268, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.2334306172938414, "standard_deviation": null, "minimum": 1.2334306172938414, "maximum": 1.2334306172938414, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19404901516321596, "standard_deviation": null, "minimum": 0.19404901516321596, "maximum": 0.19404901516321596, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9942861655387055, "standard_deviation": null, "minimum": 0.9942861655387055, "maximum": 0.9942861655387055, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0901471412273809, "standard_deviation": null, "minimum": 1.0901471412273809, "maximum": 1.0901471412273809, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.620936759022638, "standard_deviation": null, "minimum": 1.620936759022638, "maximum": 1.620936759022638, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09396365177170107, "standard_deviation": null, "minimum": 0.09396365177170107, "maximum": 0.09396365177170107, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9661775004325196, "standard_deviation": null, "minimum": 0.9661775004325196, "maximum": 0.9661775004325196, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.750755084354002, "standard_deviation": null, "minimum": 2.750755084354002, "maximum": 2.750755084354002, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9137629916226176, "standard_deviation": null, "minimum": 0.9137629916226176, "maximum": 0.9137629916226176, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19614061148242645, "standard_deviation": null, "minimum": 0.19614061148242645, "maximum": 0.19614061148242645, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9892418796032276, "standard_deviation": null, "minimum": 0.9892418796032276, "maximum": 0.9892418796032276, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.9245163135812344, "standard_deviation": null, "minimum": 1.9245163135812344, "maximum": 1.9245163135812344, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9181829216326985, "standard_deviation": null, "minimum": 0.9181829216326985, "maximum": 0.9181829216326985, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10841805816544639, "standard_deviation": null, "minimum": 0.10841805816544639, "maximum": 0.10841805816544639, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.27529916726560705, "standard_deviation": null, "minimum": 0.27529916726560705, "maximum": 0.27529916726560705, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.067990275347488, "standard_deviation": null, "minimum": 6.067990275347488, "maximum": 6.067990275347488, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.45457884369797963, "standard_deviation": null, "minimum": 0.45457884369797963, "maximum": 0.45457884369797963, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.24926068355518488, "standard_deviation": null, "minimum": 0.24926068355518488, "maximum": 0.24926068355518488, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9597023931264064, "standard_deviation": null, "minimum": 0.9597023931264064, "maximum": 0.9597023931264064, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.5193609137313575, "standard_deviation": null, "minimum": 2.5193609137313575, "maximum": 2.5193609137313575, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.29412602683420025, "standard_deviation": null, "minimum": 0.29412602683420025, "maximum": 0.29412602683420025, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1793303753873902, "standard_deviation": null, "minimum": 0.1793303753873902, "maximum": 0.1793303753873902, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9900274431891283, "standard_deviation": null, "minimum": 0.9900274431891283, "maximum": 0.9900274431891283, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.7038308029659055, "standard_deviation": null, "minimum": 1.7038308029659055, "maximum": 1.7038308029659055, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.36033887947309323, "standard_deviation": null, "minimum": 0.36033887947309323, "maximum": 0.36033887947309323, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08857553236650542, "standard_deviation": null, "minimum": 0.08857553236650542, "maximum": 0.08857553236650542, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_61"}, {"case": "librispeech_61-70968-0043", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.924679756164551, "predicted_mos": 2.924679756164551, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "68a652cad6ff46256b8cf1de4156d3d3cfebd423f835f6cc9e7ba5c78e45ab93", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 9.70629319967702}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.7847815454458753, "全帯域_10ms_rms_envelope_nrmse": 0.20923831675706867, "全帯域_10ms_envelope_correlation": 0.9580877756861182, "全帯域_10ms_modulation_mae_db": 0.4022104834464398, "全帯域_40ms_log_rms_mae_db": 1.924645196563891, "全帯域_40ms_rms_envelope_nrmse": 0.11155728980148465, "全帯域_40ms_envelope_correlation": 0.9890655632416242, "全帯域_40ms_modulation_mae_db": 0.35683368769850926, "300–1000 Hz_10ms_log_rms_mae_db": 3.5039488404477686, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23385725518316278, "300–1000 Hz_10ms_envelope_correlation": 0.963311140659583, "300–1000 Hz_10ms_modulation_mae_db": 0.6486760773198498, "300–1000 Hz_40ms_log_rms_mae_db": 2.829614009891718, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15624199483389178, "300–1000 Hz_40ms_envelope_correlation": 0.9840767997680075, "300–1000 Hz_40ms_modulation_mae_db": 0.9007582076942882, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8094749703377397, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2647097083843851, "1000–3000 Hz_10ms_envelope_correlation": 0.9472324090548284, "1000–3000 Hz_10ms_modulation_mae_db": 0.3790609648893491, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0338884150163041, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12000460739710805, "1000–3000 Hz_40ms_envelope_correlation": 0.9892457777488166, "1000–3000 Hz_40ms_modulation_mae_db": 0.8204816249393776, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7243682254583388, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24248288191735223, "3000–6000 Hz_10ms_envelope_correlation": 0.9642810097538533, "3000–6000 Hz_10ms_modulation_mae_db": 0.7873136876197498, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0312659054746376, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14088661341971892, "3000–6000 Hz_40ms_envelope_correlation": 0.9885364228107062, "3000–6000 Hz_40ms_modulation_mae_db": 0.8397085999114952, "mr_stft_db": 6.247899529195297, "band_mae_db": 0.39212606757241697, "rms_gain_error_db": 0.23724336832387344, "output_peak": 0.571583486049074}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.9658122420920705, "spectral_convergence": 0.7903356972924794, "spectral_convergence_status": "ok", "amplitude_floor": 1.4385235391470026e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.809688732221383, "spectral_convergence": 0.4149562920238751, "spectral_convergence_status": "ok", "amplitude_floor": 9.243439915833013e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.968197613272436, "spectral_convergence": 0.4390275777840135, "spectral_convergence_status": "ok", "amplitude_floor": 6.315989368278398e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00029756845932849967, "output_power": 0.00026136382873588566, "delta_db": -0.5634141288939817, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0006886429842704082, "output_power": 0.0007795983099044888, "delta_db": 0.5387676158098318, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003825307900717409, "output_power": 0.00039584203280116525, "delta_db": 0.14855510773405062, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019304685978074288, "output_power": 0.00020267049587447203, "delta_db": 0.21127788570819045, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010892090215881134, "output_power": 0.00012131890520719713, "delta_db": 0.4681725273116482, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.08275993922433e-05, "output_power": 8.908746358626739e-05, "delta_db": 0.42256913997679896, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.39212606757241697, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.23724336832387344, "pitch": null, "output_rms_dbfs": -27.328561997172223, "output_peak": 0.571583486049074, "reference_rms_dbfs": -27.565805365496097, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.247899529195297}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017313921740261034, "log_rms_mae_db": 2.7847815454458753, "rms_envelope_nrmse": 0.20923831675706867, "envelope_correlation": 0.9580877756861182, "rms_gain_error_db": 0.2457176735217722, "reference_mean_power": 0.0017487207983142375, "output_mean_power": 0.0018505134358307947}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.577112847045013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002769005691966373, "output_power": 0.0003305932037003908, "reference_active": true, "delta_db": 0.7697007266588217}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.377497445156709e-05, "output_power": 3.596357662324633e-05, "reference_active": true, "delta_db": 0.2726784716489423}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4953137492438862e-05, "output_power": 1.552950281616104e-05, "reference_active": true, "delta_db": 0.1642522520315554}], "active_band_count": 3, "active_band_mae_db": 0.4022104834464398}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015508669364693045, "log_rms_mae_db": 1.924645196563891, "rms_envelope_nrmse": 0.11155728980148465, "envelope_correlation": 0.9890655632416242, "rms_gain_error_db": 0.2558965852040992, "reference_mean_power": 0.0017487025727625826, "output_mean_power": 0.0018548363890996404}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.09499315514524e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025296787730251264, "output_power": 0.0003032121117182572, "reference_active": true, "delta_db": 0.7868116854364298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6687984786311574e-05, "output_power": 1.762793465010326e-05, "reference_active": true, "delta_db": 0.23797536677983072}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.96993861330795e-07, "output_power": 6.896957535387569e-07, "reference_active": true, "delta_db": -0.04571401087926731}], "active_band_count": 3, "active_band_mae_db": 0.35683368769850926}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014635391485956695, "log_rms_mae_db": 3.5039488404477686, "rms_envelope_nrmse": 0.23385725518316278, "envelope_correlation": 0.963311140659583, "rms_gain_error_db": 0.4744985199977765, "reference_mean_power": 0.0008142561216173605, "output_mean_power": 0.0009082614558401043}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.2354521678013805e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001985819153870851, "output_power": 0.0002342399687466617, "reference_active": true, "delta_db": 0.7172130614158885}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.365393218996536e-05, "output_power": 2.189708800161831e-05, "reference_active": true, "delta_db": -0.3351698371269335}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.646479803369054e-06, "output_power": 5.410372363967763e-06, "reference_active": true, "delta_db": -0.8936453334167274}], "active_band_count": 3, "active_band_mae_db": 0.6486760773198498}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012544182666407365, "log_rms_mae_db": 2.829614009891718, "rms_envelope_nrmse": 0.15624199483389178, "envelope_correlation": 0.9840767997680075, "rms_gain_error_db": 0.48796427918426116, "reference_mean_power": 0.0008126713020357493, "output_mean_power": 0.000909308711172925}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.0135181965386276e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018492598004057708, "output_power": 0.00021523564279681727, "reference_active": true, "delta_db": 0.6591626267760086}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0725232186915858e-05, "output_power": 9.90550514750222e-06, "reference_active": true, "delta_db": -0.3453007485291799}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.105637817013119e-07, "output_power": 2.777151103923798e-07, "reference_active": true, "delta_db": -1.6978112477776761}], "active_band_count": 3, "active_band_mae_db": 0.9007582076942882}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.111876173726916e-05, "log_rms_mae_db": 1.8094749703377397, "rms_envelope_nrmse": 0.2647097083843851, "envelope_correlation": 0.9472324090548284, "rms_gain_error_db": 0.28080056939466935, "reference_mean_power": 0.0002425197498003479, "output_mean_power": 0.0002587183116576179}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.1105776313023488e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.8957411226134796e-05, "output_power": 4.185767548948891e-05, "reference_active": true, "delta_db": 0.31185017007746196}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.645635870343935e-06, "output_power": 5.4918137934317065e-06, "reference_active": true, "delta_db": -0.1199706024979885}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.039589927108413e-06, "output_power": 3.434006148258625e-06, "reference_active": true, "delta_db": -0.705362122092597}], "active_band_count": 3, "active_band_mae_db": 0.3790609648893491}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.178958171420995e-05, "log_rms_mae_db": 1.0338884150163041, "rms_envelope_nrmse": 0.12000460739710805, "envelope_correlation": 0.9892457777488166, "rms_gain_error_db": 0.3005782498576338, "reference_mean_power": 0.00024219152470715313, "output_mean_power": 0.00025954744993950167}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0514876259433796e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7077906526859964e-05, "output_power": 3.951247756328277e-05, "reference_active": true, "delta_db": 0.2761905697902661}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.355351339185207e-06, "output_power": 2.741851788578145e-06, "reference_active": true, "delta_db": -0.8769402670506837}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8063796091292985e-07, "output_power": 2.0764190869215338e-07, "reference_active": true, "delta_db": -1.3083140379771827}], "active_band_count": 3, "active_band_mae_db": 0.8204816249393776}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3400740982798506e-05, "log_rms_mae_db": 1.7243682254583388, "rms_envelope_nrmse": 0.24248288191735223, "envelope_correlation": 0.9642810097538533, "rms_gain_error_db": 0.4545014349088455, "reference_mean_power": 0.0001023804664102342, "output_mean_power": 0.00011367559791635816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.5067174911166034e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.7037143173191167e-05, "output_power": 3.235587042081419e-05, "reference_active": true, "delta_db": 0.7799228666237481}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.303852533122454e-06, "output_power": 5.57696227758234e-06, "reference_active": true, "delta_db": 1.125403246657108}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8801329619345497e-06, "output_power": 2.0885749188313786e-06, "reference_active": true, "delta_db": 0.4566149495783936}], "active_band_count": 3, "active_band_mae_db": 0.7873136876197498}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.152733550445532e-05, "log_rms_mae_db": 1.0312659054746376, "rms_envelope_nrmse": 0.14088661341971892, "envelope_correlation": 0.9885364228107062, "rms_gain_error_db": 0.45456100663573146, "reference_mean_power": 0.0001028923086971688, "output_mean_power": 0.00011424547631902898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.218443109204273e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.6964722520221083e-05, "output_power": 3.273589924050401e-05, "reference_active": true, "delta_db": 0.8422831974590501}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9782668045768803e-06, "output_power": 2.398146983557749e-06, "reference_active": true, "delta_db": 0.8359093412056178}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5223765481589483e-07, "output_power": 1.2543789546474427e-07, "reference_active": true, "delta_db": -0.8409332610698177}], "active_band_count": 3, "active_band_mae_db": 0.8397085999114952}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.807383855394936, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "eaaf2741d533ab14bd85abfbfb11b1360a1eb04a5576c3918e0470d0ca3507bf", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9472324090548284, "standard_deviation": null, "minimum": 0.9472324090548284, "maximum": 0.9472324090548284, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8094749703377397, "standard_deviation": null, "minimum": 1.8094749703377397, "maximum": 1.8094749703377397, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3790609648893491, "standard_deviation": null, "minimum": 0.3790609648893491, "maximum": 0.3790609648893491, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2647097083843851, "standard_deviation": null, "minimum": 0.2647097083843851, "maximum": 0.2647097083843851, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9892457777488166, "standard_deviation": null, "minimum": 0.9892457777488166, "maximum": 0.9892457777488166, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0338884150163041, "standard_deviation": null, "minimum": 1.0338884150163041, "maximum": 1.0338884150163041, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8204816249393776, "standard_deviation": null, "minimum": 0.8204816249393776, "maximum": 0.8204816249393776, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12000460739710805, "standard_deviation": null, "minimum": 0.12000460739710805, "maximum": 0.12000460739710805, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9642810097538533, "standard_deviation": null, "minimum": 0.9642810097538533, "maximum": 0.9642810097538533, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7243682254583388, "standard_deviation": null, "minimum": 1.7243682254583388, "maximum": 1.7243682254583388, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7873136876197498, "standard_deviation": null, "minimum": 0.7873136876197498, "maximum": 0.7873136876197498, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24248288191735223, "standard_deviation": null, "minimum": 0.24248288191735223, "maximum": 0.24248288191735223, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9885364228107062, "standard_deviation": null, "minimum": 0.9885364228107062, "maximum": 0.9885364228107062, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0312659054746376, "standard_deviation": null, "minimum": 1.0312659054746376, "maximum": 1.0312659054746376, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8397085999114952, "standard_deviation": null, "minimum": 0.8397085999114952, "maximum": 0.8397085999114952, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14088661341971892, "standard_deviation": null, "minimum": 0.14088661341971892, "maximum": 0.14088661341971892, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.963311140659583, "standard_deviation": null, "minimum": 0.963311140659583, "maximum": 0.963311140659583, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 3.5039488404477686, "standard_deviation": null, "minimum": 3.5039488404477686, "maximum": 3.5039488404477686, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.6486760773198498, "standard_deviation": null, "minimum": 0.6486760773198498, "maximum": 0.6486760773198498, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23385725518316278, "standard_deviation": null, "minimum": 0.23385725518316278, "maximum": 0.23385725518316278, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9840767997680075, "standard_deviation": null, "minimum": 0.9840767997680075, "maximum": 0.9840767997680075, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.829614009891718, "standard_deviation": null, "minimum": 2.829614009891718, "maximum": 2.829614009891718, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9007582076942882, "standard_deviation": null, "minimum": 0.9007582076942882, "maximum": 0.9007582076942882, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15624199483389178, "standard_deviation": null, "minimum": 0.15624199483389178, "maximum": 0.15624199483389178, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.39212606757241697, "standard_deviation": null, "minimum": 0.39212606757241697, "maximum": 0.39212606757241697, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.247899529195297, "standard_deviation": null, "minimum": 6.247899529195297, "maximum": 6.247899529195297, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.571583486049074, "standard_deviation": null, "minimum": 0.571583486049074, "maximum": 0.571583486049074, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.23724336832387344, "standard_deviation": null, "minimum": 0.23724336832387344, "maximum": 0.23724336832387344, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9580877756861182, "standard_deviation": null, "minimum": 0.9580877756861182, "maximum": 0.9580877756861182, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.7847815454458753, "standard_deviation": null, "minimum": 2.7847815454458753, "maximum": 2.7847815454458753, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4022104834464398, "standard_deviation": null, "minimum": 0.4022104834464398, "maximum": 0.4022104834464398, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20923831675706867, "standard_deviation": null, "minimum": 0.20923831675706867, "maximum": 0.20923831675706867, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9890655632416242, "standard_deviation": null, "minimum": 0.9890655632416242, "maximum": 0.9890655632416242, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.924645196563891, "standard_deviation": null, "minimum": 1.924645196563891, "maximum": 1.924645196563891, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35683368769850926, "standard_deviation": null, "minimum": 0.35683368769850926, "maximum": 0.35683368769850926, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11155728980148465, "standard_deviation": null, "minimum": 0.11155728980148465, "maximum": 0.11155728980148465, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_61"}, {"case": "librispeech_61-70968-0054", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9834561347961426, "predicted_mos": 2.9834561347961426, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "b0d0cf87656299af6d2325879b299a94b7631107df928cd3bacd7ae725008340", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 10.386688399594277}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.3307703023094573, "全帯域_10ms_rms_envelope_nrmse": 0.16525527690719713, "全帯域_10ms_envelope_correlation": 0.9652456574075062, "全帯域_10ms_modulation_mae_db": 0.23701553712060072, "全帯域_40ms_log_rms_mae_db": 1.501332856384297, "全帯域_40ms_rms_envelope_nrmse": 0.09103873579687824, "全帯域_40ms_envelope_correlation": 0.989703066618108, "全帯域_40ms_modulation_mae_db": 0.35910482225804996, "300–1000 Hz_10ms_log_rms_mae_db": 2.877381967976672, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17902222123824163, "300–1000 Hz_10ms_envelope_correlation": 0.9756240813696883, "300–1000 Hz_10ms_modulation_mae_db": 0.9786344871227094, "300–1000 Hz_40ms_log_rms_mae_db": 2.162876590144207, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10930333072501791, "300–1000 Hz_40ms_envelope_correlation": 0.9919406246196837, "300–1000 Hz_40ms_modulation_mae_db": 0.9717069283704757, "1000–3000 Hz_10ms_log_rms_mae_db": 1.8178600755768164, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.205035589571686, "1000–3000 Hz_10ms_envelope_correlation": 0.9626855500536228, "1000–3000 Hz_10ms_modulation_mae_db": 1.0473848102983898, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0199131600014033, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1053970013410068, "1000–3000 Hz_40ms_envelope_correlation": 0.9899839973474024, "1000–3000 Hz_40ms_modulation_mae_db": 1.3900868825334516, "3000–6000 Hz_10ms_log_rms_mae_db": 2.018734160634232, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.29452709644857206, "3000–6000 Hz_10ms_envelope_correlation": 0.9577383228273592, "3000–6000 Hz_10ms_modulation_mae_db": 2.116137637490385, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1425853194396949, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.20994455219765012, "3000–6000 Hz_40ms_envelope_correlation": 0.9814210769164817, "3000–6000 Hz_40ms_modulation_mae_db": 2.0480262163886698, "mr_stft_db": 6.131330430294795, "band_mae_db": 0.4527743783017047, "rms_gain_error_db": 0.3404626651378089, "output_peak": 0.5147063685391113}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.016271064114701, "spectral_convergence": 0.7101108640059155, "spectral_convergence_status": "ok", "amplitude_floor": 1.3149451373049662e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.623556369162211, "spectral_convergence": 0.36205897651352736, "spectral_convergence_status": "ok", "amplitude_floor": 7.412465414213296e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.754163857607471, "spectral_convergence": 0.34030757224576624, "spectral_convergence_status": "ok", "amplitude_floor": 4.876012126794439e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006137697087039973, "output_power": 0.0005892831067705281, "delta_db": -0.17681459777076372, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0009615845769051578, "output_power": 0.001092821711863833, "delta_db": 0.5556182590256689, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000542394172591287, "output_power": 0.000589563711742207, "delta_db": 0.3621572951430436, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013703842807085808, "output_power": 0.00015152353200127374, "delta_db": 0.43637716829866, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 8.620351571877408e-05, "output_power": 9.221326076538957e-05, "delta_db": 0.2926840101409501, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011238415255857088, "output_power": 0.00013803987871577553, "delta_db": 0.8929949394311418, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4527743783017047, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.3404626651378089, "pitch": null, "output_rms_dbfs": -25.761898768591635, "output_peak": 0.5147063685391113, "reference_rms_dbfs": -26.102361433729442, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.131330430294795}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013602681282863616, "log_rms_mae_db": 2.3307703023094573, "rms_envelope_nrmse": 0.16525527690719713, "envelope_correlation": 0.9652456574075062, "rms_gain_error_db": 0.3423583077761023, "reference_mean_power": 0.0024543453947075094, "output_mean_power": 0.0026556541210320406}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.153929939453048e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000440684439069062, "output_power": 0.0004901640658734828, "reference_active": true, "delta_db": 0.4621375429400803}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001235413948928314, "output_power": 0.00012681868237495616, "reference_active": true, "delta_db": 0.11370736024409359}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6241583124510384e-05, "output_power": 2.707136801426627e-05, "reference_active": true, "delta_db": 0.13520170817762828}], "active_band_count": 3, "active_band_mae_db": 0.23701553712060072}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011635584990380617, "log_rms_mae_db": 1.501332856384297, "rms_envelope_nrmse": 0.09103873579687824, "envelope_correlation": 0.989703066618108, "rms_gain_error_db": 0.3450011059521342, "reference_mean_power": 0.0024645680451820984, "output_mean_power": 0.002668338508574962}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.594353848464586e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003918219819660693, "output_power": 0.0004376489979095334, "reference_active": true, "delta_db": 0.48037141297934893}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.98714927012752e-05, "output_power": 6.312938087580759e-05, "reference_active": true, "delta_db": 0.2301144412003658}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.328965955891304e-06, "output_power": 1.4460946421141983e-06, "reference_active": true, "delta_db": 0.3668286125944352}], "active_band_count": 3, "active_band_mae_db": 0.35910482225804996}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010777091024361473, "log_rms_mae_db": 2.877381967976672, "rms_envelope_nrmse": 0.17902222123824163, "envelope_correlation": 0.9756240813696883, "rms_gain_error_db": 0.5007428141357338, "reference_mean_power": 0.0010415487104655132, "output_mean_power": 0.0011688367740959776}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.64672159263247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003552137892650671, "output_power": 0.0003872397280066812, "reference_active": true, "delta_db": 0.3749008982605265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.132330818484316e-05, "output_power": 6.567311061768052e-05, "reference_active": true, "delta_db": -0.35843891621818086}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2237174874701015e-05, "output_power": 7.3692756484817895e-06, "reference_active": true, "delta_db": -2.2025636468894207}], "active_band_count": 3, "active_band_mae_db": 0.9786344871227094}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.988831023316543e-05, "log_rms_mae_db": 2.162876590144207, "rms_envelope_nrmse": 0.10930333072501791, "envelope_correlation": 0.9919406246196837, "rms_gain_error_db": 0.5018758319782145, "reference_mean_power": 0.0010464471526710588, "output_mean_power": 0.001174640265188029}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.816907801707229e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003300382500563707, "output_power": 0.00035820036474308137, "reference_active": true, "delta_db": 0.35561748026815587}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.204099572239431e-05, "output_power": 3.0049624879018795e-05, "reference_active": true, "delta_db": -0.278669490715025}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.619198216109287e-07, "output_power": 5.097808145591604e-07, "reference_active": true, "delta_db": -2.280833814128246}], "active_band_count": 3, "active_band_mae_db": 0.9717069283704757}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5701896659278796e-05, "log_rms_mae_db": 1.8178600755768164, "rms_envelope_nrmse": 0.205035589571686, "envelope_correlation": 0.9626855500536228, "rms_gain_error_db": 0.34543577533196285, "reference_mean_power": 0.0001796467459408233, "output_mean_power": 0.00019451940500695068}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2519806248605502e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7558987061199216e-05, "output_power": 5.196681736054099e-05, "reference_active": true, "delta_db": 0.38493523652089073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1481328556544434e-05, "output_power": 9.805432787403297e-06, "reference_active": true, "delta_db": -0.6852537804728597}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.593622469144843e-06, "output_power": 5.953688146353315e-06, "reference_active": true, "delta_db": -2.071965413901419}], "active_band_count": 3, "active_band_mae_db": 1.0473848102983898}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.308273569685666e-05, "log_rms_mae_db": 1.0199131600014033, "rms_envelope_nrmse": 0.1053970013410068, "envelope_correlation": 0.9899839973474024, "rms_gain_error_db": 0.3466110905991284, "reference_mean_power": 0.0001804894756897876, "output_mean_power": 0.000195484799145948}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.07434863630674e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.420432318446027e-05, "output_power": 4.748959286323262e-05, "reference_active": true, "delta_db": 0.31133700870987824}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8496256649417895e-06, "output_power": 4.621395412606153e-06, "reference_active": true, "delta_db": -1.0235494625777666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.792037562074947e-07, "output_power": 3.0150460662484437e-07, "reference_active": true, "delta_db": -2.835374176312709}], "active_band_count": 3, "active_band_mae_db": 1.3900868825334516}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.825215268339244e-05, "log_rms_mae_db": 2.018734160634232, "rms_envelope_nrmse": 0.29452709644857206, "envelope_correlation": 0.9577383228273592, "rms_gain_error_db": 0.9320931659596704, "reference_mean_power": 0.0001267678324533094, "output_mean_power": 0.00015711526460940195}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.277972501807574e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.1540328186660286e-05, "output_power": 6.896945189587662e-05, "reference_active": true, "delta_db": 1.2650959501560621}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.007459575479445e-05, "output_power": 1.873573502907174e-05, "reference_active": true, "delta_db": 2.6944310673531096}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.777776179485975e-06, "output_power": 6.548243392072854e-06, "reference_active": true, "delta_db": 2.388885894961984}], "active_band_count": 3, "active_band_mae_db": 2.116137637490385}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.580061124327403e-05, "log_rms_mae_db": 1.1425853194396949, "rms_envelope_nrmse": 0.20994455219765012, "envelope_correlation": 0.9814210769164817, "rms_gain_error_db": 0.9350727908855991, "reference_mean_power": 0.0001273148740012305, "output_mean_power": 0.00015790156108716225}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.770926544449125e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2932423481689434e-05, "output_power": 7.419212841897836e-05, "reference_active": true, "delta_db": 1.4663605189289994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.69146274773148e-06, "output_power": 8.703099857379462e-06, "reference_active": true, "delta_db": 2.683656943594781}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.856201910676535e-07, "output_power": 4.520589000872381e-07, "reference_active": true, "delta_db": 1.994061186642229}], "active_band_count": 3, "active_band_mae_db": 2.0480262163886698}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8807136774873379, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "1d23c500c9aae8a0f675c49c7b7f9c062426c918bf2bb51a1150a480b71ebcae", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9626855500536228, "standard_deviation": null, "minimum": 0.9626855500536228, "maximum": 0.9626855500536228, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.8178600755768164, "standard_deviation": null, "minimum": 1.8178600755768164, "maximum": 1.8178600755768164, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0473848102983898, "standard_deviation": null, "minimum": 1.0473848102983898, "maximum": 1.0473848102983898, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.205035589571686, "standard_deviation": null, "minimum": 0.205035589571686, "maximum": 0.205035589571686, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9899839973474024, "standard_deviation": null, "minimum": 0.9899839973474024, "maximum": 0.9899839973474024, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0199131600014033, "standard_deviation": null, "minimum": 1.0199131600014033, "maximum": 1.0199131600014033, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3900868825334516, "standard_deviation": null, "minimum": 1.3900868825334516, "maximum": 1.3900868825334516, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1053970013410068, "standard_deviation": null, "minimum": 0.1053970013410068, "maximum": 0.1053970013410068, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9577383228273592, "standard_deviation": null, "minimum": 0.9577383228273592, "maximum": 0.9577383228273592, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.018734160634232, "standard_deviation": null, "minimum": 2.018734160634232, "maximum": 2.018734160634232, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.116137637490385, "standard_deviation": null, "minimum": 2.116137637490385, "maximum": 2.116137637490385, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29452709644857206, "standard_deviation": null, "minimum": 0.29452709644857206, "maximum": 0.29452709644857206, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9814210769164817, "standard_deviation": null, "minimum": 0.9814210769164817, "maximum": 0.9814210769164817, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1425853194396949, "standard_deviation": null, "minimum": 1.1425853194396949, "maximum": 1.1425853194396949, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.0480262163886698, "standard_deviation": null, "minimum": 2.0480262163886698, "maximum": 2.0480262163886698, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20994455219765012, "standard_deviation": null, "minimum": 0.20994455219765012, "maximum": 0.20994455219765012, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9756240813696883, "standard_deviation": null, "minimum": 0.9756240813696883, "maximum": 0.9756240813696883, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.877381967976672, "standard_deviation": null, "minimum": 2.877381967976672, "maximum": 2.877381967976672, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9786344871227094, "standard_deviation": null, "minimum": 0.9786344871227094, "maximum": 0.9786344871227094, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17902222123824163, "standard_deviation": null, "minimum": 0.17902222123824163, "maximum": 0.17902222123824163, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9919406246196837, "standard_deviation": null, "minimum": 0.9919406246196837, "maximum": 0.9919406246196837, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.162876590144207, "standard_deviation": null, "minimum": 2.162876590144207, "maximum": 2.162876590144207, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9717069283704757, "standard_deviation": null, "minimum": 0.9717069283704757, "maximum": 0.9717069283704757, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10930333072501791, "standard_deviation": null, "minimum": 0.10930333072501791, "maximum": 0.10930333072501791, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.4527743783017047, "standard_deviation": null, "minimum": 0.4527743783017047, "maximum": 0.4527743783017047, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.131330430294795, "standard_deviation": null, "minimum": 6.131330430294795, "maximum": 6.131330430294795, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5147063685391113, "standard_deviation": null, "minimum": 0.5147063685391113, "maximum": 0.5147063685391113, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.3404626651378089, "standard_deviation": null, "minimum": 0.3404626651378089, "maximum": 0.3404626651378089, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9652456574075062, "standard_deviation": null, "minimum": 0.9652456574075062, "maximum": 0.9652456574075062, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.3307703023094573, "standard_deviation": null, "minimum": 2.3307703023094573, "maximum": 2.3307703023094573, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.23701553712060072, "standard_deviation": null, "minimum": 0.23701553712060072, "maximum": 0.23701553712060072, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16525527690719713, "standard_deviation": null, "minimum": 0.16525527690719713, "maximum": 0.16525527690719713, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.989703066618108, "standard_deviation": null, "minimum": 0.989703066618108, "maximum": 0.989703066618108, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.501332856384297, "standard_deviation": null, "minimum": 1.501332856384297, "maximum": 1.501332856384297, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35910482225804996, "standard_deviation": null, "minimum": 0.35910482225804996, "maximum": 0.35910482225804996, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09103873579687824, "standard_deviation": null, "minimum": 0.09103873579687824, "maximum": 0.09103873579687824, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_61"}, {"case": "librispeech_672-122797-0003", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9875566959381104, "predicted_mos": 2.9875566959381104, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76160, "input_sample_rate": 16000, "input_seconds": 4.76, "input_float32_sha256": "e4e1a48e446f3b352badd3e869be6b69c0543f9bbfa3761c82a47362075ae55f", "internal_samples": 76160, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 11.547643600031734}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.64301678118109, "全帯域_10ms_rms_envelope_nrmse": 0.27230983720216817, "全帯域_10ms_envelope_correlation": 0.9458837409548216, "全帯域_10ms_modulation_mae_db": 1.674158727956535, "全帯域_40ms_log_rms_mae_db": 3.198925080938006, "全帯域_40ms_rms_envelope_nrmse": 0.18774306771387514, "全帯域_40ms_envelope_correlation": 0.9756424183285062, "全帯域_40ms_modulation_mae_db": 1.7731353218234105, "300–1000 Hz_10ms_log_rms_mae_db": 2.3392725846807316, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19881894356329963, "300–1000 Hz_10ms_envelope_correlation": 0.9845032226562794, "300–1000 Hz_10ms_modulation_mae_db": 1.3954486445956145, "300–1000 Hz_40ms_log_rms_mae_db": 1.606223671762501, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15000835359771073, "300–1000 Hz_40ms_envelope_correlation": 0.994741730258376, "300–1000 Hz_40ms_modulation_mae_db": 1.313883232966345, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6812241777147359, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21675461352345113, "1000–3000 Hz_10ms_envelope_correlation": 0.9669385921138944, "1000–3000 Hz_10ms_modulation_mae_db": 0.8263657944034973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8782714191428416, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09720821909394341, "1000–3000 Hz_40ms_envelope_correlation": 0.9935853124431681, "1000–3000 Hz_40ms_modulation_mae_db": 1.274815641117934, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6892524499531418, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3824384891395332, "3000–6000 Hz_10ms_envelope_correlation": 0.9306717894810205, "3000–6000 Hz_10ms_modulation_mae_db": 2.012120556490381, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9843335821265939, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.27296097648267303, "3000–6000 Hz_40ms_envelope_correlation": 0.9666358226151195, "3000–6000 Hz_40ms_modulation_mae_db": 2.661543780281365, "mr_stft_db": 6.255154204125897, "band_mae_db": 0.5946892592292009, "rms_gain_error_db": 0.526194829218358, "output_peak": 0.551424781396887}, "full_scale": {"sample_count": 76160, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76160, "output_samples": 76160, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.863714841683865, "spectral_convergence": 0.704699096489906, "spectral_convergence_status": "ok", "amplitude_floor": 1.3356779608968942e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.8114301518865705, "spectral_convergence": 0.5822701802777431, "spectral_convergence_status": "ok", "amplitude_floor": 9.962006407647762e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.090317618807256, "spectral_convergence": 0.5868476727667549, "spectral_convergence_status": "ok", "amplitude_floor": 4.291398708237255e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021802020252140593, "output_power": 0.0001999455378853034, "delta_db": -0.375850221491733, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00019115120810434436, "output_power": 0.0002294746905619859, "delta_db": 0.7935774572977101, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 4.836069030265609e-05, "output_power": 5.80976328297275e-05, "delta_db": 0.7966594618320249, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 4.9699402895957116e-05, "output_power": 5.56384959261049e-05, "delta_db": 0.49024210192127576, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001234531768639194, "output_power": 0.00012890878443048954, "delta_db": 0.18780242897726213, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00036807758285640346, "output_power": 0.00045534414505526006, "delta_db": 0.924003883855199, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5946892592292009, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.526194829218358, "pitch": null, "output_rms_dbfs": -29.479183921332503, "output_peak": 0.551424781396887, "reference_rms_dbfs": -30.00537875055086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.255154204125897}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76160, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.00018341452353274373, "log_rms_mae_db": 3.64301678118109, "rms_envelope_nrmse": 0.27230983720216817, "envelope_correlation": 0.9458837409548216, "rms_gain_error_db": 0.5262352586087612, "reference_mean_power": 0.0009998027766238483, "output_mean_power": 0.0011285943308396325}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 5.754282242130778e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018894414816227902, "output_power": 0.00020927565613285302, "reference_active": true, "delta_db": 0.44385266412126534}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013084274193993133, "output_power": 0.00010312813111798987, "reference_active": true, "delta_db": -1.0337248905999419}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.690431882761885e-05, "output_power": 1.631500948857932e-05, "reference_active": true, "delta_db": -3.5448986291483973}], "active_band_count": 3, "active_band_mae_db": 1.674158727956535}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.0001230241515816465, "log_rms_mae_db": 3.198925080938006, "rms_envelope_nrmse": 0.18774306771387514, "envelope_correlation": 0.9756424183285062, "rms_gain_error_db": 0.5264197464322382, "reference_mean_power": 0.001006100998286567, "output_mean_power": 0.0011357521158715086}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 5.061730772699165e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019045538591950462, "output_power": 0.00019911288365684376, "reference_active": true, "delta_db": 0.1930610343047393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.588390935994399e-05, "output_power": 5.502695351822577e-05, "reference_active": true, "delta_db": -1.3957422718227483}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5269059669432984e-06, "output_power": 1.070357397700078e-06, "reference_active": true, "delta_db": -3.7306026593427437}], "active_band_count": 3, "active_band_mae_db": 1.7731353218234105}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 5.489840871101384e-05, "log_rms_mae_db": 2.3392725846807316, "rms_envelope_nrmse": 0.19881894356329963, "envelope_correlation": 0.9845032226562794, "rms_gain_error_db": 0.9830211191800906, "reference_mean_power": 0.0001499835837122031, "output_mean_power": 0.00018808139545887468}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.245105068973029e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.560041827470393e-05, "output_power": 8.323614258353763e-05, "reference_active": true, "delta_db": 1.0340533698765926}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1382200906061474e-05, "output_power": 2.457904593246834e-05, "reference_active": true, "delta_db": 0.6051261533599862}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4492416985862286e-06, "output_power": 1.3624325275111766e-06, "reference_active": true, "delta_db": -2.5471664105502647}], "active_band_count": 3, "active_band_mae_db": 1.3954486445956145}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 4.550505877934166e-05, "log_rms_mae_db": 1.606223671762501, "rms_envelope_nrmse": 0.15000835359771073, "envelope_correlation": 0.994741730258376, "rms_gain_error_db": 0.9830112045714374, "reference_mean_power": 0.00015093488949896852, "output_mean_power": 0.00018927391338523545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 1.1354747567630587e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.429099822482352e-05, "output_power": 8.0380193298669e-05, "reference_active": true, "delta_db": 0.9699887729170986}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0916365304226932e-05, "output_power": 1.3007069663175803e-05, "reference_active": true, "delta_db": 0.761014058111968}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3162457402511593e-07, "output_power": 1.3922587206270047e-07, "reference_active": true, "delta_db": -2.2106468678699684}], "active_band_count": 3, "active_band_mae_db": 1.313883232966345}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 4.066436664003698e-05, "log_rms_mae_db": 1.6812241777147359, "rms_envelope_nrmse": 0.21675461352345113, "envelope_correlation": 0.9669385921138944, "rms_gain_error_db": 0.2871277050345119, "reference_mean_power": 0.0001020893317881212, "output_mean_power": 0.00010906694076433354}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 9.602503308503003e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.081225810114696e-05, "output_power": 3.117410753899333e-05, "reference_active": true, "delta_db": 0.05070502121446326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3389822486480084e-05, "output_power": 1.0247704112736197e-05, "reference_active": true, "delta_db": -1.1614824214928208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.310273392307641e-06, "output_power": 3.966668310741349e-06, "reference_active": true, "delta_db": -1.2669099405032083}], "active_band_count": 3, "active_band_mae_db": 0.8263657944034973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 3.8037796302268945e-05, "log_rms_mae_db": 0.8782714191428416, "rms_envelope_nrmse": 0.09720821909394341, "envelope_correlation": 0.9935853124431681, "rms_gain_error_db": 0.28713070626639464, "reference_mean_power": 0.00010273559922291735, "output_mean_power": 0.00010975745518101776}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 8.714815910547825e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9375203264151156e-05, "output_power": 2.9628053014821897e-05, "reference_active": true, "delta_db": 0.0372223265629261}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.005890574709935e-06, "output_power": 5.428279920421657e-06, "reference_active": true, "delta_db": -1.10801114917877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.15357493378345e-07, "output_power": 2.780912027133191e-07, "reference_active": true, "delta_db": -2.679213447612106}], "active_band_count": 3, "active_band_mae_db": 1.274815641117934}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001808188641270182, "log_rms_mae_db": 1.6892524499531418, "rms_envelope_nrmse": 0.3824384891395332, "envelope_correlation": 0.9306717894810205, "rms_gain_error_db": 0.8645673487141556, "reference_mean_power": 0.0004025082939999733, "output_mean_power": 0.0004911697012273081}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 3.5966286763117237e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001367199427375967, "output_power": 0.0001278480073936345, "reference_active": true, "delta_db": -0.2913790420488131}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.299777768885521e-05, "output_power": 5.325303865597519e-05, "reference_active": true, "delta_db": -2.421281766402033}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.495937881252574e-05, "output_power": 1.1610841227769998e-05, "reference_active": true, "delta_db": -3.3237008610202965}], "active_band_count": 3, "active_band_mae_db": 2.012120556490381}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76160, "uncovered_tail_samples": 0, "metrics": {"frame_count": 945, "rms_floor": 0.00011978792663441872, "log_rms_mae_db": 0.9843335821265939, "rms_envelope_nrmse": 0.27296097648267303, "envelope_correlation": 0.9666358226151195, "rms_gain_error_db": 0.8645703127245864, "reference_mean_power": 0.0004050621320160614, "output_mean_power": 0.0004942864162319462}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 945, "frequency_resolution_hz": 0.21164021164021163, "power_floor": 3.450538172499259e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016854041012359937, "output_power": 0.00013676630214397826, "reference_active": true, "delta_db": -0.9072494166649647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.211674623356507e-05, "output_power": 2.7711636301530284e-05, "reference_active": true, "delta_db": -2.743151233771728}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0452544406168747e-06, "output_power": 7.539182827687946e-07, "reference_active": true, "delta_db": -4.334230690407402}], "active_band_count": 3, "active_band_mae_db": 2.661543780281365}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8066947184434111, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f7b0017b672bb613257a04122b62c4f227f93bace59c535b4c9082bb3df135a4", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9669385921138944, "standard_deviation": null, "minimum": 0.9669385921138944, "maximum": 0.9669385921138944, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6812241777147359, "standard_deviation": null, "minimum": 1.6812241777147359, "maximum": 1.6812241777147359, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8263657944034973, "standard_deviation": null, "minimum": 0.8263657944034973, "maximum": 0.8263657944034973, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21675461352345113, "standard_deviation": null, "minimum": 0.21675461352345113, "maximum": 0.21675461352345113, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9935853124431681, "standard_deviation": null, "minimum": 0.9935853124431681, "maximum": 0.9935853124431681, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8782714191428416, "standard_deviation": null, "minimum": 0.8782714191428416, "maximum": 0.8782714191428416, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.274815641117934, "standard_deviation": null, "minimum": 1.274815641117934, "maximum": 1.274815641117934, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09720821909394341, "standard_deviation": null, "minimum": 0.09720821909394341, "maximum": 0.09720821909394341, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9306717894810205, "standard_deviation": null, "minimum": 0.9306717894810205, "maximum": 0.9306717894810205, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6892524499531418, "standard_deviation": null, "minimum": 1.6892524499531418, "maximum": 1.6892524499531418, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.012120556490381, "standard_deviation": null, "minimum": 2.012120556490381, "maximum": 2.012120556490381, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3824384891395332, "standard_deviation": null, "minimum": 0.3824384891395332, "maximum": 0.3824384891395332, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9666358226151195, "standard_deviation": null, "minimum": 0.9666358226151195, "maximum": 0.9666358226151195, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9843335821265939, "standard_deviation": null, "minimum": 0.9843335821265939, "maximum": 0.9843335821265939, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.661543780281365, "standard_deviation": null, "minimum": 2.661543780281365, "maximum": 2.661543780281365, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.27296097648267303, "standard_deviation": null, "minimum": 0.27296097648267303, "maximum": 0.27296097648267303, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9845032226562794, "standard_deviation": null, "minimum": 0.9845032226562794, "maximum": 0.9845032226562794, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3392725846807316, "standard_deviation": null, "minimum": 2.3392725846807316, "maximum": 2.3392725846807316, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3954486445956145, "standard_deviation": null, "minimum": 1.3954486445956145, "maximum": 1.3954486445956145, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19881894356329963, "standard_deviation": null, "minimum": 0.19881894356329963, "maximum": 0.19881894356329963, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.994741730258376, "standard_deviation": null, "minimum": 0.994741730258376, "maximum": 0.994741730258376, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.606223671762501, "standard_deviation": null, "minimum": 1.606223671762501, "maximum": 1.606223671762501, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.313883232966345, "standard_deviation": null, "minimum": 1.313883232966345, "maximum": 1.313883232966345, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15000835359771073, "standard_deviation": null, "minimum": 0.15000835359771073, "maximum": 0.15000835359771073, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5946892592292009, "standard_deviation": null, "minimum": 0.5946892592292009, "maximum": 0.5946892592292009, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.255154204125897, "standard_deviation": null, "minimum": 6.255154204125897, "maximum": 6.255154204125897, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.551424781396887, "standard_deviation": null, "minimum": 0.551424781396887, "maximum": 0.551424781396887, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.526194829218358, "standard_deviation": null, "minimum": 0.526194829218358, "maximum": 0.526194829218358, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9458837409548216, "standard_deviation": null, "minimum": 0.9458837409548216, "maximum": 0.9458837409548216, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.64301678118109, "standard_deviation": null, "minimum": 3.64301678118109, "maximum": 3.64301678118109, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.674158727956535, "standard_deviation": null, "minimum": 1.674158727956535, "maximum": 1.674158727956535, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.27230983720216817, "standard_deviation": null, "minimum": 0.27230983720216817, "maximum": 0.27230983720216817, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9756424183285062, "standard_deviation": null, "minimum": 0.9756424183285062, "maximum": 0.9756424183285062, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 3.198925080938006, "standard_deviation": null, "minimum": 3.198925080938006, "maximum": 3.198925080938006, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.7731353218234105, "standard_deviation": null, "minimum": 1.7731353218234105, "maximum": 1.7731353218234105, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18774306771387514, "standard_deviation": null, "minimum": 0.18774306771387514, "maximum": 0.18774306771387514, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_672"}, {"case": "librispeech_672-122797-0044", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.6653618812561035, "predicted_mos": 3.6653618812561035, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 59840, "input_sample_rate": 16000, "input_seconds": 3.74, "input_float32_sha256": "824dea4bb4c1680411ceb77036ee22c988b3fd4cb7b422a6a4eeaae58308fd75", "internal_samples": 59840, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 11.375604700297117}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.3598871324435513, "全帯域_10ms_rms_envelope_nrmse": 0.16982325292847594, "全帯域_10ms_envelope_correlation": 0.9817962968124091, "全帯域_10ms_modulation_mae_db": 1.4048699968980485, "全帯域_40ms_log_rms_mae_db": 2.1467343318699244, "全帯域_40ms_rms_envelope_nrmse": 0.11753566785661868, "全帯域_40ms_envelope_correlation": 0.9951620156414366, "全帯域_40ms_modulation_mae_db": 1.8176634428935605, "300–1000 Hz_10ms_log_rms_mae_db": 1.657064843188485, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19599376887948056, "300–1000 Hz_10ms_envelope_correlation": 0.9858978570024564, "300–1000 Hz_10ms_modulation_mae_db": 1.0145516484593473, "300–1000 Hz_40ms_log_rms_mae_db": 1.2900102648789864, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14682349087929547, "300–1000 Hz_40ms_envelope_correlation": 0.9960013475692071, "300–1000 Hz_40ms_modulation_mae_db": 0.9773286008399592, "1000–3000 Hz_10ms_log_rms_mae_db": 1.2287497184553327, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15697843324643718, "1000–3000 Hz_10ms_envelope_correlation": 0.9837722262560243, "1000–3000 Hz_10ms_modulation_mae_db": 0.953586763140435, "1000–3000 Hz_40ms_log_rms_mae_db": 0.7249906788852831, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09016695324185191, "1000–3000 Hz_40ms_envelope_correlation": 0.9966095650806329, "1000–3000 Hz_40ms_modulation_mae_db": 0.9864563321116314, "3000–6000 Hz_10ms_log_rms_mae_db": 1.398572386519484, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23861804282643412, "3000–6000 Hz_10ms_envelope_correlation": 0.9664093733948557, "3000–6000 Hz_10ms_modulation_mae_db": 0.8080858862469711, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8707967114778942, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.151708188747097, "3000–6000 Hz_40ms_envelope_correlation": 0.9864883001293013, "3000–6000 Hz_40ms_modulation_mae_db": 0.8262119187692308, "mr_stft_db": 5.912232206794208, "band_mae_db": 0.5455647064142397, "rms_gain_error_db": 0.7507997325728423, "output_peak": 0.3457906778043492}, "full_scale": {"sample_count": 59840, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 59840, "output_samples": 59840, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.677768929804718, "spectral_convergence": 0.7314352658580386, "spectral_convergence_status": "ok", "amplitude_floor": 9.91771656423115e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.2433789028427364, "spectral_convergence": 0.3139516812068547, "spectral_convergence_status": "ok", "amplitude_floor": 5.25712991724442e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.815548787735169, "spectral_convergence": 0.30861252908263137, "spectral_convergence_status": "ok", "amplitude_floor": 4.1776466342690114e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00025386719019706026, "output_power": 0.000253708867673959, "delta_db": -0.0027092925013917813, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004894966727833924, "output_power": 0.0006027136278729534, "delta_db": 0.9036126719627275, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005421484972995941, "output_power": 0.0006872096867777788, "delta_db": 1.029710142440417, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.5605939839496278e-05, "output_power": 1.7516419131501166e-05, "delta_db": 0.5015540002017806, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000107892990107672, "output_power": 0.00012884914838959111, "delta_db": 0.7708832334200467, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.406384293716839e-05, "output_power": 7.29649606617267e-05, "delta_db": -0.06491889795907403, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5455647064142397, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7507997325728423, "pitch": null, "output_rms_dbfs": -27.537568736292606, "output_peak": 0.3457906778043492, "reference_rms_dbfs": -28.28836846886545, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.912232206794208}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 59840, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.000110539633783132, "log_rms_mae_db": 2.3598871324435513, "rms_envelope_nrmse": 0.16982325292847594, "envelope_correlation": 0.9817962968124091, "rms_gain_error_db": 0.7508055309090063, "reference_mean_power": 0.00148505750597689, "output_mean_power": 0.0017653215558148397}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.0996976338580339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028388367321682654, "output_power": 0.00035153357959709873, "reference_active": true, "delta_db": 0.9282640051723295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.6121837601531416e-05, "output_power": 6.799831021456318e-05, "reference_active": true, "delta_db": 1.6859151817529914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3216272322760606e-05, "output_power": 1.9105220708269834e-05, "reference_active": true, "delta_db": 1.6004308037688246}], "active_band_count": 3, "active_band_mae_db": 1.4048699968980485}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 0.00010014294879668002, "log_rms_mae_db": 2.1467343318699244, "rms_envelope_nrmse": 0.11753566785661868, "envelope_correlation": 0.9951620156414366, "rms_gain_error_db": 0.7508692529513167, "reference_mean_power": 0.0014970488171277764, "output_mean_power": 0.0017796020105775026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.0259449632693504e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002533237576934496, "output_power": 0.00031722473997262106, "reference_active": true, "delta_db": 0.9769112852933238}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.399735365000864e-05, "output_power": 3.5360550278180926e-05, "reference_active": true, "delta_db": 1.6835566307981815}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.21988770898296e-07, "output_power": 1.1831363170025482e-06, "reference_active": true, "delta_db": 2.792522412589176}], "active_band_count": 3, "active_band_mae_db": 1.8176634428935605}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 0.00010594974849403162, "log_rms_mae_db": 1.657064843188485, "rms_envelope_nrmse": 0.19599376887948056, "envelope_correlation": 0.9858978570024564, "rms_gain_error_db": 1.0070435388167567, "reference_mean_power": 0.0008075829245111437, "output_mean_power": 0.0010183369014036896}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.2363198879746262e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00030947682521494476, "output_power": 0.00038590550906622995, "reference_active": true, "delta_db": 0.9585284549315353}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.465844834587486e-05, "output_power": 5.3015082770272806e-05, "reference_active": true, "delta_db": 1.8459032828303854}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.693327955984243e-06, "output_power": 6.334608571288323e-06, "reference_active": true, "delta_db": -0.23922320761612106}], "active_band_count": 3, "active_band_mae_db": 1.0145516484593473}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 9.562043532359368e-05, "log_rms_mae_db": 1.2900102648789864, "rms_envelope_nrmse": 0.14682349087929547, "envelope_correlation": 0.9960013475692071, "rms_gain_error_db": 1.0070410561578895, "reference_mean_power": 0.0008141211968758593, "output_mean_power": 0.001026580872257316}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.204812078547339e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002902058778585621, "output_power": 0.0003620545610554978, "reference_active": true, "delta_db": 0.9606781852934332}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7747321249654287e-05, "output_power": 2.760759518609584e-05, "reference_active": true, "delta_db": 1.9189576754066398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.319074590753967e-07, "output_power": 3.359324882350227e-07, "reference_active": true, "delta_db": 0.052349941819804686}], "active_band_count": 3, "active_band_mae_db": 0.9773286008399592}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 2.8704944305868236e-05, "log_rms_mae_db": 1.2287497184553327, "rms_envelope_nrmse": 0.15697843324643718, "envelope_correlation": 0.9837722262560243, "rms_gain_error_db": 0.5420133123111344, "reference_mean_power": 4.5961113590202676e-05, "output_mean_power": 5.207051509863134e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 2.8606659711814982e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3098445296769517e-05, "output_power": 1.4823780810199305e-05, "reference_active": true, "delta_db": 0.5373923403651565}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.2173629500624627e-06, "output_power": 3.47160167553079e-06, "reference_active": true, "delta_db": 0.33029832965059613}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2063471779341703e-06, "output_power": 7.62369214598485e-07, "reference_active": true, "delta_db": -1.9930696194055524}], "active_band_count": 3, "active_band_mae_db": 0.953586763140435}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 2.1311353675739154e-05, "log_rms_mae_db": 0.7249906788852831, "rms_envelope_nrmse": 0.09016695324185191, "envelope_correlation": 0.9966095650806329, "rms_gain_error_db": 0.5420225549282733, "reference_mean_power": 4.633078215253765e-05, "output_mean_power": 5.248943372680703e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 2.5755466765658237e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2064714839118615e-05, "output_power": 1.360517763282245e-05, "reference_active": true, "delta_db": 0.5218715500960313}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.637842198880982e-06, "output_power": 1.690594134295048e-06, "reference_active": true, "delta_db": 0.13767301322758874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.209021145285469e-07, "output_power": 7.119532112835717e-08, "reference_active": true, "delta_db": -2.299824433011274}], "active_band_count": 3, "active_band_mae_db": 0.9864563321116314}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 747, "rms_floor": 6.327125594838897e-05, "log_rms_mae_db": 1.398572386519484, "rms_envelope_nrmse": 0.23861804282643412, "envelope_correlation": 0.9664093733948557, "rms_gain_error_db": 0.38276190233589774, "reference_mean_power": 0.0001246130887459015, "output_mean_power": 0.0001360942733512594}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 747, "frequency_resolution_hz": 0.2677376171352075, "power_floor": 1.481947168400202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.119164755141333e-05, "output_power": 5.614942832037829e-05, "reference_active": true, "delta_db": -0.3734680759838954}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1929332977586818e-05, "output_power": 2.9182875959463123e-05, "reference_active": true, "delta_db": 1.2410266720987966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.860956874662985e-06, "output_power": 8.267248282707989e-06, "reference_active": true, "delta_db": 0.8097629106582211}], "active_band_count": 3, "active_band_mae_db": 0.8080858862469711}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 59840, "uncovered_tail_samples": 0, "metrics": {"frame_count": 741, "rms_floor": 5.46294552803286e-05, "log_rms_mae_db": 0.8707967114778942, "rms_envelope_nrmse": 0.151708188747097, "envelope_correlation": 0.9864883001293013, "rms_gain_error_db": 0.3827523889828617, "reference_mean_power": 0.00012561601350824365, "output_mean_power": 0.0001371893017282479}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 741, "frequency_resolution_hz": 0.2699055330634278, "power_floor": 1.3825123163038373e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.737559737198788e-05, "output_power": 5.398416679239176e-05, "reference_active": true, "delta_db": -0.26460817540190756}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1774546568382367e-05, "output_power": 1.4836484660032088e-05, "reference_active": true, "delta_db": 1.0038682021620036}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.943011691356486e-07, "output_power": 7.852763579588816e-07, "reference_active": true, "delta_db": 1.2101593787437817}], "active_band_count": 3, "active_band_mae_db": 0.8262119187692308}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8510893077550538, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "48ea99e58e16012460aae447a2476c83c62d0715d2d0410c2c152605a3574055", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9837722262560243, "standard_deviation": null, "minimum": 0.9837722262560243, "maximum": 0.9837722262560243, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.2287497184553327, "standard_deviation": null, "minimum": 1.2287497184553327, "maximum": 1.2287497184553327, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.953586763140435, "standard_deviation": null, "minimum": 0.953586763140435, "maximum": 0.953586763140435, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15697843324643718, "standard_deviation": null, "minimum": 0.15697843324643718, "maximum": 0.15697843324643718, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966095650806329, "standard_deviation": null, "minimum": 0.9966095650806329, "maximum": 0.9966095650806329, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.7249906788852831, "standard_deviation": null, "minimum": 0.7249906788852831, "maximum": 0.7249906788852831, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9864563321116314, "standard_deviation": null, "minimum": 0.9864563321116314, "maximum": 0.9864563321116314, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09016695324185191, "standard_deviation": null, "minimum": 0.09016695324185191, "maximum": 0.09016695324185191, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9664093733948557, "standard_deviation": null, "minimum": 0.9664093733948557, "maximum": 0.9664093733948557, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.398572386519484, "standard_deviation": null, "minimum": 1.398572386519484, "maximum": 1.398572386519484, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8080858862469711, "standard_deviation": null, "minimum": 0.8080858862469711, "maximum": 0.8080858862469711, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23861804282643412, "standard_deviation": null, "minimum": 0.23861804282643412, "maximum": 0.23861804282643412, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9864883001293013, "standard_deviation": null, "minimum": 0.9864883001293013, "maximum": 0.9864883001293013, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8707967114778942, "standard_deviation": null, "minimum": 0.8707967114778942, "maximum": 0.8707967114778942, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8262119187692308, "standard_deviation": null, "minimum": 0.8262119187692308, "maximum": 0.8262119187692308, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.151708188747097, "standard_deviation": null, "minimum": 0.151708188747097, "maximum": 0.151708188747097, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9858978570024564, "standard_deviation": null, "minimum": 0.9858978570024564, "maximum": 0.9858978570024564, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.657064843188485, "standard_deviation": null, "minimum": 1.657064843188485, "maximum": 1.657064843188485, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0145516484593473, "standard_deviation": null, "minimum": 1.0145516484593473, "maximum": 1.0145516484593473, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19599376887948056, "standard_deviation": null, "minimum": 0.19599376887948056, "maximum": 0.19599376887948056, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9960013475692071, "standard_deviation": null, "minimum": 0.9960013475692071, "maximum": 0.9960013475692071, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2900102648789864, "standard_deviation": null, "minimum": 1.2900102648789864, "maximum": 1.2900102648789864, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9773286008399592, "standard_deviation": null, "minimum": 0.9773286008399592, "maximum": 0.9773286008399592, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14682349087929547, "standard_deviation": null, "minimum": 0.14682349087929547, "maximum": 0.14682349087929547, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5455647064142397, "standard_deviation": null, "minimum": 0.5455647064142397, "maximum": 0.5455647064142397, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.912232206794208, "standard_deviation": null, "minimum": 5.912232206794208, "maximum": 5.912232206794208, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.3457906778043492, "standard_deviation": null, "minimum": 0.3457906778043492, "maximum": 0.3457906778043492, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7507997325728423, "standard_deviation": null, "minimum": 0.7507997325728423, "maximum": 0.7507997325728423, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9817962968124091, "standard_deviation": null, "minimum": 0.9817962968124091, "maximum": 0.9817962968124091, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.3598871324435513, "standard_deviation": null, "minimum": 2.3598871324435513, "maximum": 2.3598871324435513, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.4048699968980485, "standard_deviation": null, "minimum": 1.4048699968980485, "maximum": 1.4048699968980485, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16982325292847594, "standard_deviation": null, "minimum": 0.16982325292847594, "maximum": 0.16982325292847594, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9951620156414366, "standard_deviation": null, "minimum": 0.9951620156414366, "maximum": 0.9951620156414366, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.1467343318699244, "standard_deviation": null, "minimum": 2.1467343318699244, "maximum": 2.1467343318699244, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.8176634428935605, "standard_deviation": null, "minimum": 1.8176634428935605, "maximum": 1.8176634428935605, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.11753566785661868, "standard_deviation": null, "minimum": 0.11753566785661868, "maximum": 0.11753566785661868, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_672"}, {"case": "librispeech_672-122797-0056", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3703773021698, "predicted_mos": 3.3703773021698, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 83600, "input_sample_rate": 16000, "input_seconds": 5.225, "input_float32_sha256": "fbf5316b5ac3bfb3035dfe0bac8e0b47a4994206a2c3073672e87be23e7074ea", "internal_samples": 83600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.829133199993521}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.297147121393546, "全帯域_10ms_rms_envelope_nrmse": 0.18028056714453147, "全帯域_10ms_envelope_correlation": 0.9737303788937487, "全帯域_10ms_modulation_mae_db": 1.0119308112891348, "全帯域_40ms_log_rms_mae_db": 1.9599567953912247, "全帯域_40ms_rms_envelope_nrmse": 0.1205628821922213, "全帯域_40ms_envelope_correlation": 0.9919210650169662, "全帯域_40ms_modulation_mae_db": 0.6609232341445374, "300–1000 Hz_10ms_log_rms_mae_db": 1.8707251731857282, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19856612918096053, "300–1000 Hz_10ms_envelope_correlation": 0.9874242814802903, "300–1000 Hz_10ms_modulation_mae_db": 1.201979922760207, "300–1000 Hz_40ms_log_rms_mae_db": 1.4257727400742468, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17426298000439655, "300–1000 Hz_40ms_envelope_correlation": 0.9935804185527475, "300–1000 Hz_40ms_modulation_mae_db": 1.3028418945310374, "1000–3000 Hz_10ms_log_rms_mae_db": 1.480231404711509, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1858603394969169, "1000–3000 Hz_10ms_envelope_correlation": 0.9764694420340075, "1000–3000 Hz_10ms_modulation_mae_db": 0.8357580959084752, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8574513421298918, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10960471539381993, "1000–3000 Hz_40ms_envelope_correlation": 0.9929796453829887, "1000–3000 Hz_40ms_modulation_mae_db": 0.7618353960936016, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5425383858583896, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2821821090170082, "3000–6000 Hz_10ms_envelope_correlation": 0.9641929925583754, "3000–6000 Hz_10ms_modulation_mae_db": 1.436566722836633, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8202547778723038, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1792781511575996, "3000–6000 Hz_40ms_envelope_correlation": 0.9885966267309846, "3000–6000 Hz_40ms_modulation_mae_db": 1.329463903083538, "mr_stft_db": 5.975715130724843, "band_mae_db": 0.6709284025273717, "rms_gain_error_db": 0.6532263211513549, "output_peak": 0.4622516378636949}, "full_scale": {"sample_count": 83600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 83600, "output_samples": 83600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.946337796251571, "spectral_convergence": 0.692054547031774, "spectral_convergence_status": "ok", "amplitude_floor": 9.102684030105167e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.209153756113016, "spectral_convergence": 0.3982215003602221, "spectral_convergence_status": "ok", "amplitude_floor": 4.6649711782186323e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.77165383980994, "spectral_convergence": 0.4087239609973401, "spectral_convergence_status": "ok", "amplitude_floor": 4.1057122515023654e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00021084017487574348, "output_power": 0.0001996084448589787, "delta_db": -0.2377445666702135, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00047622157098200406, "output_power": 0.0005887264647500705, "delta_db": 0.9210449552852962, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 9.587972951197478e-05, "output_power": 0.00011952231555493415, "delta_db": 0.9572219815089804, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 7.411215555329107e-05, "output_power": 8.466858562498145e-05, "delta_db": 0.5783286010044741, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013227809814888984, "output_power": 0.0001417556284035387, "delta_db": 0.3005236941545705, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00014809096583739848, "output_power": 0.00018775833579612333, "delta_db": 1.0307066165406957, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6709284025273717, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6532263211513549, "pitch": null, "output_rms_dbfs": -28.787554784372006, "output_peak": 0.4622516378636949, "reference_rms_dbfs": -29.44078110552336, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.975715130724843}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 83600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011323786448874551, "log_rms_mae_db": 2.297147121393546, "rms_envelope_nrmse": 0.18028056714453147, "envelope_correlation": 0.9737303788937487, "rms_gain_error_db": 0.6532791487441586, "reference_mean_power": 0.0011384943575262435, "output_mean_power": 0.0013233014775965587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 5.62759551802579e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002928658523286973, "output_power": 0.000359951408662455, "reference_active": true, "delta_db": 0.8957514088670439}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4610498849864094e-05, "output_power": 5.241983422502601e-05, "reference_active": true, "delta_db": 0.700585634809678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2101686887119984e-05, "output_power": 8.687611129518361e-06, "reference_active": true, "delta_db": -1.4394553901906824}], "active_band_count": 3, "active_band_mae_db": 1.0119308112891348}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.563924945576787e-05, "log_rms_mae_db": 1.9599567953912247, "rms_envelope_nrmse": 0.1205628821922213, "envelope_correlation": 0.9919210650169662, "rms_gain_error_db": 0.653524368858197, "reference_mean_power": 0.0011449801518647084, "output_mean_power": 0.0013309152310314988}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 5.095010942049215e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002727818973842863, "output_power": 0.00033988767239879325, "reference_active": true, "delta_db": 0.9551986729625178}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1267992599082604e-05, "output_power": 2.5099261330609443e-05, "reference_active": true, "delta_db": 0.719344399001588}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.757168583210849e-07, "output_power": 5.362734344883387e-07, "reference_active": true, "delta_db": -0.3082266304695065}], "active_band_count": 3, "active_band_mae_db": 0.6609232341445374}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.979916609312644e-05, "log_rms_mae_db": 1.8707251731857282, "rms_envelope_nrmse": 0.19856612918096053, "envelope_correlation": 0.9874242814802903, "rms_gain_error_db": 1.196052644315496, "reference_mean_power": 0.00038382289364078325, "output_mean_power": 0.0005055174360985318}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 2.8601579771498626e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014700129208541836, "output_power": 0.00019109534271438702, "reference_active": true, "delta_db": 1.1392895071541307}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.957621157356578e-05, "output_power": 2.4181148317729647e-05, "reference_active": true, "delta_db": 0.9174827076324598}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3113428408893907e-06, "output_power": 1.6178848702552938e-06, "reference_active": true, "delta_db": -1.5491675534940308}], "active_band_count": 3, "active_band_mae_db": 1.201979922760207}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 6.59156861629162e-05, "log_rms_mae_db": 1.4257727400742468, "rms_envelope_nrmse": 0.17426298000439655, "envelope_correlation": 0.9935804185527475, "rms_gain_error_db": 1.1960243519385403, "reference_mean_power": 0.00038603855676420007, "output_mean_power": 0.0005084322831612446}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 2.667606930039739e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013762626674571402, "output_power": 0.00017853710279088484, "reference_active": true, "delta_db": 1.130271536350011}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.746730510582489e-06, "output_power": 1.2038954303374562e-05, "reference_active": true, "delta_db": 0.917298076613639}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5949673646594045e-07, "output_power": 1.0390972468329773e-07, "reference_active": true, "delta_db": -1.8609560706294621}], "active_band_count": 3, "active_band_mae_db": 1.3028418945310374}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 6.341026161507385e-05, "log_rms_mae_db": 1.480231404711509, "rms_envelope_nrmse": 0.1858603394969169, "envelope_correlation": 0.9764694420340075, "rms_gain_error_db": 0.4851189844951196, "reference_mean_power": 0.0001433729237288462, "output_mean_power": 0.00016031680168764214}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.1455083323281239e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.589033152322262e-05, "output_power": 6.302779822069453e-05, "reference_active": true, "delta_db": 0.5219545026325287}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4083924473542424e-05, "output_power": 1.36125749335402e-05, "reference_active": true, "delta_db": -0.14783404012698786}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.499188380560451e-06, "output_power": 3.6020523696064554e-06, "reference_active": true, "delta_db": -1.837485744965909}], "active_band_count": 3, "active_band_mae_db": 0.8357580959084752}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 5.482542724448801e-05, "log_rms_mae_db": 0.8574513421298918, "rms_envelope_nrmse": 0.10960471539381993, "envelope_correlation": 0.9929796453829887, "rms_gain_error_db": 0.4852446586175923, "reference_mean_power": 0.00014418079784839864, "output_mean_power": 0.00016122481612302852}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.042775196620521e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.2049448137923216e-05, "output_power": 5.841892166548432e-05, "reference_active": true, "delta_db": 0.501374070224148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.197217517258297e-06, "output_power": 5.774168991135661e-06, "reference_active": true, "delta_db": -0.3070724968088048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5366430302416474e-07, "output_power": 1.8053166733933638e-07, "reference_active": true, "delta_db": -1.4770596212478524}], "active_band_count": 3, "active_band_mae_db": 0.7618353960936016}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1044, "rms_floor": 0.00011141137328067575, "log_rms_mae_db": 1.5425383858583896, "rms_envelope_nrmse": 0.2821821090170082, "envelope_correlation": 0.9641929925583754, "rms_gain_error_db": 0.8448860626340975, "reference_mean_power": 0.00018590462562984253, "output_mean_power": 0.00022582852719887046}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1044, "frequency_resolution_hz": 0.19157088122605365, "power_floor": 1.861645592580093e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.396584441083375e-05, "output_power": 0.00011136619183906551, "reference_active": true, "delta_db": 0.7378334839925073}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7617019162087818e-05, "output_power": 2.4980893296468915e-05, "reference_active": true, "delta_db": 1.5167553761162869}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.489954892102086e-06, "output_power": 4.043249555193312e-06, "reference_active": true, "delta_db": -2.0551113084011043}], "active_band_count": 3, "active_band_mae_db": 1.436566722836633}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 83600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1038, "rms_floor": 9.054702623382987e-05, "log_rms_mae_db": 0.8202547778723038, "rms_envelope_nrmse": 0.1792781511575996, "envelope_correlation": 0.9885966267309846, "rms_gain_error_db": 0.8449031345205746, "reference_mean_power": 0.00018697687803543306, "output_mean_power": 0.00022713194374463755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1038, "frequency_resolution_hz": 0.19267822736030826, "power_floor": 1.7928912528322647e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.176227386223464e-05, "output_power": 0.00011003163487353469, "reference_active": true, "delta_db": 0.7885339903634253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.193273720945643e-06, "output_power": 1.0677014402111732e-05, "reference_active": true, "delta_db": 1.1499236452429384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.787097063305561e-07, "output_power": 2.3621802216070553e-07, "reference_active": true, "delta_db": -2.04993407364425}], "active_band_count": 3, "active_band_mae_db": 1.329463903083538}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8798319955359684, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0579959c404cd9665c571487c539849247b416d947654a227534578f766884d9", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9764694420340075, "standard_deviation": null, "minimum": 0.9764694420340075, "maximum": 0.9764694420340075, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.480231404711509, "standard_deviation": null, "minimum": 1.480231404711509, "maximum": 1.480231404711509, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8357580959084752, "standard_deviation": null, "minimum": 0.8357580959084752, "maximum": 0.8357580959084752, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1858603394969169, "standard_deviation": null, "minimum": 0.1858603394969169, "maximum": 0.1858603394969169, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929796453829887, "standard_deviation": null, "minimum": 0.9929796453829887, "maximum": 0.9929796453829887, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8574513421298918, "standard_deviation": null, "minimum": 0.8574513421298918, "maximum": 0.8574513421298918, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7618353960936016, "standard_deviation": null, "minimum": 0.7618353960936016, "maximum": 0.7618353960936016, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10960471539381993, "standard_deviation": null, "minimum": 0.10960471539381993, "maximum": 0.10960471539381993, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9641929925583754, "standard_deviation": null, "minimum": 0.9641929925583754, "maximum": 0.9641929925583754, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5425383858583896, "standard_deviation": null, "minimum": 1.5425383858583896, "maximum": 1.5425383858583896, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.436566722836633, "standard_deviation": null, "minimum": 1.436566722836633, "maximum": 1.436566722836633, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2821821090170082, "standard_deviation": null, "minimum": 0.2821821090170082, "maximum": 0.2821821090170082, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9885966267309846, "standard_deviation": null, "minimum": 0.9885966267309846, "maximum": 0.9885966267309846, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8202547778723038, "standard_deviation": null, "minimum": 0.8202547778723038, "maximum": 0.8202547778723038, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.329463903083538, "standard_deviation": null, "minimum": 1.329463903083538, "maximum": 1.329463903083538, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1792781511575996, "standard_deviation": null, "minimum": 0.1792781511575996, "maximum": 0.1792781511575996, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9874242814802903, "standard_deviation": null, "minimum": 0.9874242814802903, "maximum": 0.9874242814802903, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.8707251731857282, "standard_deviation": null, "minimum": 1.8707251731857282, "maximum": 1.8707251731857282, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.201979922760207, "standard_deviation": null, "minimum": 1.201979922760207, "maximum": 1.201979922760207, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19856612918096053, "standard_deviation": null, "minimum": 0.19856612918096053, "maximum": 0.19856612918096053, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9935804185527475, "standard_deviation": null, "minimum": 0.9935804185527475, "maximum": 0.9935804185527475, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4257727400742468, "standard_deviation": null, "minimum": 1.4257727400742468, "maximum": 1.4257727400742468, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3028418945310374, "standard_deviation": null, "minimum": 1.3028418945310374, "maximum": 1.3028418945310374, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17426298000439655, "standard_deviation": null, "minimum": 0.17426298000439655, "maximum": 0.17426298000439655, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6709284025273717, "standard_deviation": null, "minimum": 0.6709284025273717, "maximum": 0.6709284025273717, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.975715130724843, "standard_deviation": null, "minimum": 5.975715130724843, "maximum": 5.975715130724843, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4622516378636949, "standard_deviation": null, "minimum": 0.4622516378636949, "maximum": 0.4622516378636949, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6532263211513549, "standard_deviation": null, "minimum": 0.6532263211513549, "maximum": 0.6532263211513549, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9737303788937487, "standard_deviation": null, "minimum": 0.9737303788937487, "maximum": 0.9737303788937487, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.297147121393546, "standard_deviation": null, "minimum": 2.297147121393546, "maximum": 2.297147121393546, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.0119308112891348, "standard_deviation": null, "minimum": 1.0119308112891348, "maximum": 1.0119308112891348, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18028056714453147, "standard_deviation": null, "minimum": 0.18028056714453147, "maximum": 0.18028056714453147, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9919210650169662, "standard_deviation": null, "minimum": 0.9919210650169662, "maximum": 0.9919210650169662, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.9599567953912247, "standard_deviation": null, "minimum": 1.9599567953912247, "maximum": 1.9599567953912247, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6609232341445374, "standard_deviation": null, "minimum": 0.6609232341445374, "maximum": 0.6609232341445374, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1205628821922213, "standard_deviation": null, "minimum": 0.1205628821922213, "maximum": 0.1205628821922213, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_672"}, {"case": "librispeech_6829-68769-0014", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.123014211654663, "predicted_mos": 3.123014211654663, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 58480, "input_sample_rate": 16000, "input_seconds": 3.655, "input_float32_sha256": "7a8c86316362ec65990f0bdd2fe66f870b767850dd037c5d6a9acc2381af1acc", "internal_samples": 58480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.214939000084996}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4665842409350078, "全帯域_10ms_rms_envelope_nrmse": 0.19635081176296515, "全帯域_10ms_envelope_correlation": 0.9821659058072747, "全帯域_10ms_modulation_mae_db": 0.862422646818283, "全帯域_40ms_log_rms_mae_db": 1.0746207453312218, "全帯域_40ms_rms_envelope_nrmse": 0.16560027699365829, "全帯域_40ms_envelope_correlation": 0.9908842676321276, "全帯域_40ms_modulation_mae_db": 0.8222645502060454, "300–1000 Hz_10ms_log_rms_mae_db": 2.1938567883478126, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23444838528280892, "300–1000 Hz_10ms_envelope_correlation": 0.9830651608231707, "300–1000 Hz_10ms_modulation_mae_db": 1.1097467995556223, "300–1000 Hz_40ms_log_rms_mae_db": 1.7688608865982878, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2064124573137769, "300–1000 Hz_40ms_envelope_correlation": 0.9907293806696189, "300–1000 Hz_40ms_modulation_mae_db": 1.3071487246608215, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5405357486026519, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.14810552882298353, "1000–3000 Hz_10ms_envelope_correlation": 0.9863729130248339, "1000–3000 Hz_10ms_modulation_mae_db": 0.3760433689920059, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0194306223942433, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1104890426866901, "1000–3000 Hz_40ms_envelope_correlation": 0.9929341930563864, "1000–3000 Hz_40ms_modulation_mae_db": 0.3859938278637896, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5243240841826096, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.18962651803263586, "3000–6000 Hz_10ms_envelope_correlation": 0.9741632846131227, "3000–6000 Hz_10ms_modulation_mae_db": 1.0723411811358234, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8887604397497363, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.10710239291973556, "3000–6000 Hz_40ms_envelope_correlation": 0.9918779899631142, "3000–6000 Hz_40ms_modulation_mae_db": 1.3856192489212307, "mr_stft_db": 5.876737314925976, "band_mae_db": 0.751250123881538, "rms_gain_error_db": 1.0788276049119658, "output_peak": 0.6085572133562728}, "full_scale": {"sample_count": 58480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 58480, "output_samples": 58480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.214671698078705, "spectral_convergence": 0.524659691624377, "spectral_convergence_status": "ok", "amplitude_floor": 1.5267125480811954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.486464151751936, "spectral_convergence": 0.3801328641133489, "spectral_convergence_status": "ok", "amplitude_floor": 1.0684354905475814e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.929076094947287, "spectral_convergence": 0.43165652548167993, "spectral_convergence_status": "ok", "amplitude_floor": 7.412945744843815e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0002955479965644196, "output_power": 0.0002250221714372698, "delta_db": -1.1840270847629966, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00146613378566028, "output_power": 0.0017842100650310266, "delta_db": 0.8527238320205889, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0021080578166581473, "output_power": 0.003040963949164783, "delta_db": 1.5912875370030763, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0001787602167832836, "output_power": 0.00020603939838297517, "delta_db": 0.6167940044905861, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00014763682412872298, "output_power": 0.0001564327986310307, "delta_db": 0.25133120683112153, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001169621027158164, "output_power": 0.00011665717614694003, "delta_db": -0.01133707818085828, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.751250123881538, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0788276049119658, "pitch": null, "output_rms_dbfs": -22.573278386694522, "output_peak": 0.6085572133562728, "reference_rms_dbfs": -23.65210599160649, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.876737314925976}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 58480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.0002136685864137289, "log_rms_mae_db": 1.4665842409350078, "rms_envelope_nrmse": 0.19635081176296515, "envelope_correlation": 0.9821659058072747, "rms_gain_error_db": 1.0788307712880276, "reference_mean_power": 0.004318997149502425, "output_mean_power": 0.005536891261754427}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.9590597233654063e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007496792113024279, "output_power": 0.000971132015110202, "reference_active": true, "delta_db": 1.1240280368225306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00031885669741703006, "output_power": 0.00043930310417903247, "reference_active": true, "delta_db": 1.3916872851756295}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.908796664605564e-05, "output_power": 7.023566081169979e-05, "reference_active": true, "delta_db": 0.07155261845668909}], "active_band_count": 3, "active_band_mae_db": 0.862422646818283}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00017810705079567896, "log_rms_mae_db": 1.0746207453312218, "rms_envelope_nrmse": 0.16560027699365829, "envelope_correlation": 0.9908842676321276, "rms_gain_error_db": 1.078856630612781, "reference_mean_power": 0.0043546897567722925, "output_mean_power": 0.005582681903971896}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.684242583835521e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006964729892166045, "output_power": 0.0009094850190315625, "reference_active": true, "delta_db": 1.1589127161929815}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017819877477929594, "output_power": 0.0002383391747879882, "reference_active": true, "delta_db": 1.2629071765220214}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.2119939135052175e-06, "output_power": 5.266247617645504e-06, "reference_active": true, "delta_db": 0.04497375790313336}], "active_band_count": 3, "active_band_mae_db": 0.8222645502060454}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 0.00018825872292702335, "log_rms_mae_db": 2.1938567883478126, "rms_envelope_nrmse": 0.23444838528280892, "envelope_correlation": 0.9830651608231707, "rms_gain_error_db": 1.3955514959342663, "reference_mean_power": 0.003284629583795886, "output_mean_power": 0.004529409115028515}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 2.043578979045215e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008901806972697356, "output_power": 0.001220241944907664, "reference_active": true, "delta_db": 1.369677766805022}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002718123373610413, "output_power": 0.00036368593913118166, "reference_active": true, "delta_db": 1.2645734548532026}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.0768894288562865e-05, "output_power": 5.178242878542955e-05, "reference_active": true, "delta_db": -0.694989177008642}], "active_band_count": 3, "active_band_mae_db": 1.1097467995556223}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 0.00015953911997549416, "log_rms_mae_db": 1.7688608865982878, "rms_envelope_nrmse": 0.2064124573137769, "envelope_correlation": 0.9907293806696189, "rms_gain_error_db": 1.3955540289713397, "reference_mean_power": 0.0033118454961761057, "output_mean_power": 0.004566941734057996}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.8056141501282585e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008648498098004977, "output_power": 0.0011706233698172517, "reference_active": true, "delta_db": 1.31476495689934}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013850439438479803, "output_power": 0.00018169323962271875, "reference_active": true, "delta_db": 1.1787521587910763}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.4525026177578845e-06, "output_power": 4.6444606060483315e-06, "reference_active": true, "delta_db": -1.4279290582920483}], "active_band_count": 3, "active_band_mae_db": 1.3071487246608215}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 6.604508172479527e-05, "log_rms_mae_db": 1.5405357486026519, "rms_envelope_nrmse": 0.14810552882298353, "envelope_correlation": 0.9863729130248339, "rms_gain_error_db": 0.4991179529765399, "reference_mean_power": 0.00023215995390588967, "output_mean_power": 0.00026043485324587877}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 1.444500480390906e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.349295087979117e-05, "output_power": 8.530917650331281e-05, "reference_active": true, "delta_db": 0.6475006422749662}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7764968189144465e-05, "output_power": 2.981302465465674e-05, "reference_active": true, "delta_db": 0.3090885900991112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.892866752425807e-06, "output_power": 5.66464285503961e-06, "reference_active": true, "delta_db": -0.17154087460194017}], "active_band_count": 3, "active_band_mae_db": 0.3760433689920059}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.621650617144381e-05, "log_rms_mae_db": 1.0194306223942433, "rms_envelope_nrmse": 0.1104890426866901, "envelope_correlation": 0.9929341930563864, "rms_gain_error_db": 0.4992356151775653, "reference_mean_power": 0.00023404438013250112, "output_mean_power": 0.0002625558981251238}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 1.2566052761381804e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.967443215044184e-05, "output_power": 7.993387360225942e-05, "reference_active": true, "delta_db": 0.5965742154563932}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5348913604759878e-05, "output_power": 1.658547333958535e-05, "reference_active": true, "delta_db": 0.3365022897758725}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.927931708651043e-07, "output_power": 4.679227738918408e-07, "reference_active": true, "delta_db": -0.22490497835910317}], "active_band_count": 3, "active_band_mae_db": 0.3859938278637896}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 730, "rms_floor": 5.916637942334853e-05, "log_rms_mae_db": 1.5243240841826096, "rms_envelope_nrmse": 0.18962651803263586, "envelope_correlation": 0.9741632846131227, "rms_gain_error_db": 0.31019760589108286, "reference_mean_power": 0.00012513898539343656, "output_mean_power": 0.00013440406068816909}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 730, "frequency_resolution_hz": 0.273972602739726, "power_floor": 8.947237162820932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.53948723828488e-05, "output_power": 4.86580075187515e-05, "reference_active": true, "delta_db": 0.3014752165461746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2730550619395128e-05, "output_power": 1.4280742512219932e-05, "reference_active": true, "delta_db": 0.4990360062940085}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.684612122113266e-06, "output_power": 2.9386722607442155e-06, "reference_active": true, "delta_db": 2.4165123205672874}], "active_band_count": 3, "active_band_mae_db": 1.0723411811358234}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 58480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 724, "rms_floor": 5.1723901891273096e-05, "log_rms_mae_db": 0.8887604397497363, "rms_envelope_nrmse": 0.10710239291973556, "envelope_correlation": 0.9918779899631142, "rms_gain_error_db": 0.31030550505668114, "reference_mean_power": 0.0001261310627116738, "output_mean_power": 0.00013547295544837822}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 724, "frequency_resolution_hz": 0.27624309392265195, "power_floor": 8.199333001241026e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.1138157243798736e-05, "output_power": 4.529340539839645e-05, "reference_active": true, "delta_db": 0.41790140608651044}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.547615689423209e-06, "output_power": 8.068734081169381e-06, "reference_active": true, "delta_db": 0.907222221025667}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.061187881770097e-07, "output_power": 2.036881172484058e-07, "reference_active": true, "delta_db": 2.831734119651515}], "active_band_count": 3, "active_band_mae_db": 1.3856192489212307}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8923407940359006, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2a2b4e9a933a4892577aabdd16f317d8dd6b5d1c283cd958113e305695cedf0d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9863729130248339, "standard_deviation": null, "minimum": 0.9863729130248339, "maximum": 0.9863729130248339, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5405357486026519, "standard_deviation": null, "minimum": 1.5405357486026519, "maximum": 1.5405357486026519, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3760433689920059, "standard_deviation": null, "minimum": 0.3760433689920059, "maximum": 0.3760433689920059, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.14810552882298353, "standard_deviation": null, "minimum": 0.14810552882298353, "maximum": 0.14810552882298353, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929341930563864, "standard_deviation": null, "minimum": 0.9929341930563864, "maximum": 0.9929341930563864, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0194306223942433, "standard_deviation": null, "minimum": 1.0194306223942433, "maximum": 1.0194306223942433, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.3859938278637896, "standard_deviation": null, "minimum": 0.3859938278637896, "maximum": 0.3859938278637896, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1104890426866901, "standard_deviation": null, "minimum": 0.1104890426866901, "maximum": 0.1104890426866901, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9741632846131227, "standard_deviation": null, "minimum": 0.9741632846131227, "maximum": 0.9741632846131227, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5243240841826096, "standard_deviation": null, "minimum": 1.5243240841826096, "maximum": 1.5243240841826096, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.0723411811358234, "standard_deviation": null, "minimum": 1.0723411811358234, "maximum": 1.0723411811358234, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18962651803263586, "standard_deviation": null, "minimum": 0.18962651803263586, "maximum": 0.18962651803263586, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9918779899631142, "standard_deviation": null, "minimum": 0.9918779899631142, "maximum": 0.9918779899631142, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8887604397497363, "standard_deviation": null, "minimum": 0.8887604397497363, "maximum": 0.8887604397497363, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3856192489212307, "standard_deviation": null, "minimum": 1.3856192489212307, "maximum": 1.3856192489212307, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10710239291973556, "standard_deviation": null, "minimum": 0.10710239291973556, "maximum": 0.10710239291973556, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9830651608231707, "standard_deviation": null, "minimum": 0.9830651608231707, "maximum": 0.9830651608231707, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1938567883478126, "standard_deviation": null, "minimum": 2.1938567883478126, "maximum": 2.1938567883478126, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1097467995556223, "standard_deviation": null, "minimum": 1.1097467995556223, "maximum": 1.1097467995556223, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23444838528280892, "standard_deviation": null, "minimum": 0.23444838528280892, "maximum": 0.23444838528280892, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9907293806696189, "standard_deviation": null, "minimum": 0.9907293806696189, "maximum": 0.9907293806696189, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7688608865982878, "standard_deviation": null, "minimum": 1.7688608865982878, "maximum": 1.7688608865982878, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3071487246608215, "standard_deviation": null, "minimum": 1.3071487246608215, "maximum": 1.3071487246608215, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2064124573137769, "standard_deviation": null, "minimum": 0.2064124573137769, "maximum": 0.2064124573137769, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.751250123881538, "standard_deviation": null, "minimum": 0.751250123881538, "maximum": 0.751250123881538, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.876737314925976, "standard_deviation": null, "minimum": 5.876737314925976, "maximum": 5.876737314925976, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6085572133562728, "standard_deviation": null, "minimum": 0.6085572133562728, "maximum": 0.6085572133562728, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0788276049119658, "standard_deviation": null, "minimum": 1.0788276049119658, "maximum": 1.0788276049119658, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9821659058072747, "standard_deviation": null, "minimum": 0.9821659058072747, "maximum": 0.9821659058072747, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4665842409350078, "standard_deviation": null, "minimum": 1.4665842409350078, "maximum": 1.4665842409350078, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.862422646818283, "standard_deviation": null, "minimum": 0.862422646818283, "maximum": 0.862422646818283, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19635081176296515, "standard_deviation": null, "minimum": 0.19635081176296515, "maximum": 0.19635081176296515, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9908842676321276, "standard_deviation": null, "minimum": 0.9908842676321276, "maximum": 0.9908842676321276, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0746207453312218, "standard_deviation": null, "minimum": 1.0746207453312218, "maximum": 1.0746207453312218, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8222645502060454, "standard_deviation": null, "minimum": 0.8222645502060454, "maximum": 0.8222645502060454, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16560027699365829, "standard_deviation": null, "minimum": 0.16560027699365829, "maximum": 0.16560027699365829, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_6829"}, {"case": "librispeech_6829-68769-0038", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.2752301692962646, "predicted_mos": 2.2752301692962646, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 40080, "input_sample_rate": 16000, "input_seconds": 2.505, "input_float32_sha256": "31e06400297753ded7fee783f7e1fa459b00558e5013138bb2ed9a1dcd2b026b", "internal_samples": 40080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.114172500092536}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.590857030153452, "全帯域_10ms_rms_envelope_nrmse": 0.18135847941749125, "全帯域_10ms_envelope_correlation": 0.985586599879198, "全帯域_10ms_modulation_mae_db": 1.2555209388989657, "全帯域_40ms_log_rms_mae_db": 1.064954708474147, "全帯域_40ms_rms_envelope_nrmse": 0.1456151239672922, "全帯域_40ms_envelope_correlation": 0.9947517820149684, "全帯域_40ms_modulation_mae_db": 1.0644731804420298, "300–1000 Hz_10ms_log_rms_mae_db": 2.3080406296280915, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.23086459180385785, "300–1000 Hz_10ms_envelope_correlation": 0.9833306562074816, "300–1000 Hz_10ms_modulation_mae_db": 1.2096579928900013, "300–1000 Hz_40ms_log_rms_mae_db": 1.6345448909845088, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.20335928497371822, "300–1000 Hz_40ms_envelope_correlation": 0.9904519671688629, "300–1000 Hz_40ms_modulation_mae_db": 1.0495906804897728, "1000–3000 Hz_10ms_log_rms_mae_db": 1.600932053257719, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20052428308302533, "1000–3000 Hz_10ms_envelope_correlation": 0.968519885615217, "1000–3000 Hz_10ms_modulation_mae_db": 0.6305046358802973, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9880095618142967, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.106749164481851, "1000–3000 Hz_40ms_envelope_correlation": 0.9924547081066234, "1000–3000 Hz_40ms_modulation_mae_db": 0.7040232932846838, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6875200000757473, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.1985046336472933, "3000–6000 Hz_10ms_envelope_correlation": 0.9718148081136394, "3000–6000 Hz_10ms_modulation_mae_db": 0.3536467483223295, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1811602286700673, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11385417380360048, "3000–6000 Hz_40ms_envelope_correlation": 0.990306628989457, "3000–6000 Hz_40ms_modulation_mae_db": 0.3634546411109845, "mr_stft_db": 6.0016740586539115, "band_mae_db": 0.6363544763667003, "rms_gain_error_db": 1.0072354084085335, "output_peak": 0.4168886354198507}, "full_scale": {"sample_count": 40080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 40080, "output_samples": 40080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.36213921954254, "spectral_convergence": 0.6319299465823995, "spectral_convergence_status": "ok", "amplitude_floor": 1.4347408585118857e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.672488182815941, "spectral_convergence": 0.33315973520786824, "spectral_convergence_status": "ok", "amplitude_floor": 9.356428468567288e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.970394773603255, "spectral_convergence": 0.37961041213323465, "spectral_convergence_status": "ok", "amplitude_floor": 6.30553407186755e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00020558767458207216, "output_power": 0.00019167705098753257, "delta_db": -0.304269552249869, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002656133227213937, "output_power": 0.003414057186418281, "delta_db": 1.0902093666498889, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.000658897280424305, "output_power": 0.0009048042571219712, "delta_db": 1.3773692027054554, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.296049498403337e-05, "output_power": 7.33645964293739e-05, "delta_db": 0.6641839916789761, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00018191885630194454, "output_power": 0.00019451047702831844, "delta_db": 0.29065281937727794, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012525230408717123, "output_power": 0.00012791748622242752, "delta_db": 0.09144192553873388, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6363544763667003, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0072354084085335, "pitch": null, "output_rms_dbfs": -23.09243151163267, "output_peak": 0.4168886354198507, "reference_rms_dbfs": -24.099666920041205, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.0016740586539115}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 40080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.0001593748762509214, "log_rms_mae_db": 1.590857030153452, "rms_envelope_nrmse": 0.18135847941749125, "envelope_correlation": 0.985586599879198, "rms_gain_error_db": 1.0072593597563921, "reference_mean_power": 0.003898496329889168, "output_mean_power": 0.004916126683523469}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.3730898670333144e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0011160072986886684, "output_power": 0.0015723636924328357, "reference_active": true, "delta_db": 1.4888597204047207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003924601048841299, "output_power": 0.0005258968385710522, "reference_active": true, "delta_db": 1.27105044480039}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.521600297301381e-05, "output_power": 6.758586784218043e-05, "reference_active": true, "delta_db": -1.0066526514917868}], "active_band_count": 3, "active_band_mae_db": 1.2555209388989657}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00014354118626608257, "log_rms_mae_db": 1.064954708474147, "rms_envelope_nrmse": 0.1456151239672922, "envelope_correlation": 0.9947517820149684, "rms_gain_error_db": 1.0073381142276403, "reference_mean_power": 0.00394559944963645, "output_mean_power": 0.004975615427811173}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 2.0339309551628037e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010452646972319804, "output_power": 0.0014744018665112427, "reference_active": true, "delta_db": 1.493895892216826}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020675655639960477, "output_power": 0.00027463827621124025, "reference_active": true, "delta_db": 1.2330177430837972}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.033282667633223e-06, "output_power": 5.418799505335768e-06, "reference_active": true, "delta_db": -0.46650590602546604}], "active_band_count": 3, "active_band_mae_db": 1.0644731804420298}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 0.00014555177137634006, "log_rms_mae_db": 2.3080406296280915, "rms_envelope_nrmse": 0.23086459180385785, "envelope_correlation": 0.9833306562074816, "rms_gain_error_db": 1.345109941948298, "reference_mean_power": 0.0025948122953778335, "output_mean_power": 0.0035368524417987296}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 2.091667884405576e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010721637846984168, "output_power": 0.0015986150270759798, "reference_active": true, "delta_db": 1.7348275740894161}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00045198280410970506, "output_power": 0.0005810516012738205, "reference_active": true, "delta_db": 1.0909279016608564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.681659745338991e-05, "output_power": 4.722297802072625e-05, "reference_active": true, "delta_db": -0.8032185029197315}], "active_band_count": 3, "active_band_mae_db": 1.2096579928900013}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 0.00013266563413733552, "log_rms_mae_db": 1.6345448909845088, "rms_envelope_nrmse": 0.20335928497371822, "envelope_correlation": 0.9904519671688629, "rms_gain_error_db": 1.345135814777238, "reference_mean_power": 0.002626297782108314, "output_mean_power": 0.0035797899824493505}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 1.815151890487515e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010442677388382842, "output_power": 0.0015492242568266095, "reference_active": true, "delta_db": 1.7130242698759153}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002292871431701824, "output_power": 0.0002854294678352752, "reference_active": true, "delta_db": 0.9511910460743455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.238038641621491e-06, "output_power": 6.473888642004514e-06, "reference_active": true, "delta_db": -0.4845567255190578}], "active_band_count": 3, "active_band_mae_db": 1.0495906804897728}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.977119603351369e-05, "log_rms_mae_db": 1.600932053257719, "rms_envelope_nrmse": 0.20052428308302533, "envelope_correlation": 0.968519885615217, "rms_gain_error_db": 0.4964046138318602, "reference_mean_power": 0.0001519473982974274, "output_mean_power": 0.00017034670195421875}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 8.081548208249742e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5556653827814076e-05, "output_power": 5.30472052809759e-05, "reference_active": true, "delta_db": 0.6611069132680634}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1818343242486535e-05, "output_power": 1.4376676476481285e-05, "reference_active": true, "delta_db": 0.8510190063630868}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7000790662746577e-06, "output_power": 3.3905660237757537e-06, "reference_active": true, "delta_db": -0.3793879880097418}], "active_band_count": 3, "active_band_mae_db": 0.6305046358802973}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.109627865595404e-05, "log_rms_mae_db": 0.9880095618142967, "rms_envelope_nrmse": 0.106749164481851, "envelope_correlation": 0.9924547081066234, "rms_gain_error_db": 0.4973829561023534, "reference_mean_power": 0.0001537098951112271, "output_mean_power": 0.00017236144319891824}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 6.838335178235257e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3201238570188224e-05, "output_power": 4.9976301964389126e-05, "reference_active": true, "delta_db": 0.6326791889647}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.064245816738516e-06, "output_power": 7.64661350456131e-06, "reference_active": true, "delta_db": 1.006923421899471}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0299327724151498e-07, "output_power": 1.8206854779549656e-07, "reference_active": true, "delta_db": -0.47246726898988056}], "active_band_count": 3, "active_band_mae_db": 0.7040232932846838}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 500, "rms_floor": 4.882145378793849e-05, "log_rms_mae_db": 1.6875200000757473, "rms_envelope_nrmse": 0.1985046336472933, "envelope_correlation": 0.9718148081136394, "rms_gain_error_db": 0.19087335088355672, "reference_mean_power": 0.00013710154501992013, "output_mean_power": 0.00014326156266708504}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 500, "frequency_resolution_hz": 0.4, "power_floor": 1.0763544842397476e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.5103901227352814e-05, "output_power": 4.738889723269753e-05, "reference_active": true, "delta_db": 0.21462495060069783}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.11132679073469e-05, "output_power": 3.53031775137145e-05, "reference_active": true, "delta_db": 0.5486816787010014}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.435692144349255e-06, "output_power": 5.075647937898467e-06, "reference_active": true, "delta_db": -0.2976336156652892}], "active_band_count": 3, "active_band_mae_db": 0.3536467483223295}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 40080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 494, "rms_floor": 4.3950747306848176e-05, "log_rms_mae_db": 1.1811602286700673, "rms_envelope_nrmse": 0.11385417380360048, "envelope_correlation": 0.990306628989457, "rms_gain_error_db": 0.1908469769418394, "reference_mean_power": 0.00013864857515666164, "output_mean_power": 0.0001448772215616186}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 494, "frequency_resolution_hz": 0.4048582995951417, "power_floor": 9.21368112936246e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.381631867120006e-05, "output_power": 4.635424225178155e-05, "reference_active": true, "delta_db": 0.2445359959391174}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.911163628753255e-05, "output_power": 2.1138349547592477e-05, "reference_active": true, "delta_db": 0.4377320341001577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5242776229186804e-07, "output_power": 3.208193267128413e-07, "reference_active": true, "delta_db": -0.4080958932936784}], "active_band_count": 3, "active_band_mae_db": 0.3634546411109845}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8926781311942733, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4e8f847ac3bc6081e8951f8eed637d4a80594ef6cf5a7074c040d10cec15a62e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.968519885615217, "standard_deviation": null, "minimum": 0.968519885615217, "maximum": 0.968519885615217, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.600932053257719, "standard_deviation": null, "minimum": 1.600932053257719, "maximum": 1.600932053257719, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6305046358802973, "standard_deviation": null, "minimum": 0.6305046358802973, "maximum": 0.6305046358802973, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20052428308302533, "standard_deviation": null, "minimum": 0.20052428308302533, "maximum": 0.20052428308302533, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9924547081066234, "standard_deviation": null, "minimum": 0.9924547081066234, "maximum": 0.9924547081066234, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9880095618142967, "standard_deviation": null, "minimum": 0.9880095618142967, "maximum": 0.9880095618142967, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7040232932846838, "standard_deviation": null, "minimum": 0.7040232932846838, "maximum": 0.7040232932846838, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.106749164481851, "standard_deviation": null, "minimum": 0.106749164481851, "maximum": 0.106749164481851, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9718148081136394, "standard_deviation": null, "minimum": 0.9718148081136394, "maximum": 0.9718148081136394, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6875200000757473, "standard_deviation": null, "minimum": 1.6875200000757473, "maximum": 1.6875200000757473, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3536467483223295, "standard_deviation": null, "minimum": 0.3536467483223295, "maximum": 0.3536467483223295, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1985046336472933, "standard_deviation": null, "minimum": 0.1985046336472933, "maximum": 0.1985046336472933, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.990306628989457, "standard_deviation": null, "minimum": 0.990306628989457, "maximum": 0.990306628989457, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1811602286700673, "standard_deviation": null, "minimum": 1.1811602286700673, "maximum": 1.1811602286700673, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3634546411109845, "standard_deviation": null, "minimum": 0.3634546411109845, "maximum": 0.3634546411109845, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11385417380360048, "standard_deviation": null, "minimum": 0.11385417380360048, "maximum": 0.11385417380360048, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9833306562074816, "standard_deviation": null, "minimum": 0.9833306562074816, "maximum": 0.9833306562074816, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3080406296280915, "standard_deviation": null, "minimum": 2.3080406296280915, "maximum": 2.3080406296280915, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.2096579928900013, "standard_deviation": null, "minimum": 1.2096579928900013, "maximum": 1.2096579928900013, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23086459180385785, "standard_deviation": null, "minimum": 0.23086459180385785, "maximum": 0.23086459180385785, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9904519671688629, "standard_deviation": null, "minimum": 0.9904519671688629, "maximum": 0.9904519671688629, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6345448909845088, "standard_deviation": null, "minimum": 1.6345448909845088, "maximum": 1.6345448909845088, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.0495906804897728, "standard_deviation": null, "minimum": 1.0495906804897728, "maximum": 1.0495906804897728, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20335928497371822, "standard_deviation": null, "minimum": 0.20335928497371822, "maximum": 0.20335928497371822, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6363544763667003, "standard_deviation": null, "minimum": 0.6363544763667003, "maximum": 0.6363544763667003, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.0016740586539115, "standard_deviation": null, "minimum": 6.0016740586539115, "maximum": 6.0016740586539115, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4168886354198507, "standard_deviation": null, "minimum": 0.4168886354198507, "maximum": 0.4168886354198507, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0072354084085335, "standard_deviation": null, "minimum": 1.0072354084085335, "maximum": 1.0072354084085335, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.985586599879198, "standard_deviation": null, "minimum": 0.985586599879198, "maximum": 0.985586599879198, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.590857030153452, "standard_deviation": null, "minimum": 1.590857030153452, "maximum": 1.590857030153452, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.2555209388989657, "standard_deviation": null, "minimum": 1.2555209388989657, "maximum": 1.2555209388989657, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18135847941749125, "standard_deviation": null, "minimum": 0.18135847941749125, "maximum": 0.18135847941749125, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9947517820149684, "standard_deviation": null, "minimum": 0.9947517820149684, "maximum": 0.9947517820149684, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.064954708474147, "standard_deviation": null, "minimum": 1.064954708474147, "maximum": 1.064954708474147, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.0644731804420298, "standard_deviation": null, "minimum": 1.0644731804420298, "maximum": 1.0644731804420298, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1456151239672922, "standard_deviation": null, "minimum": 0.1456151239672922, "maximum": 0.1456151239672922, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_6829"}, {"case": "librispeech_6829-68769-0041", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2808361053466797, "predicted_mos": 3.2808361053466797, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 42560, "input_sample_rate": 16000, "input_seconds": 2.66, "input_float32_sha256": "a76a5303e43c59a2624d7185e748887fc0ead688cf1aa40800b44685779b1988", "internal_samples": 42560, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.8544986001215875}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.3817324164474896, "全帯域_10ms_rms_envelope_nrmse": 0.1749660962446044, "全帯域_10ms_envelope_correlation": 0.9867094158026204, "全帯域_10ms_modulation_mae_db": 0.9927547921543153, "全帯域_40ms_log_rms_mae_db": 1.0173752857179321, "全帯域_40ms_rms_envelope_nrmse": 0.14866554327918494, "全帯域_40ms_envelope_correlation": 0.9936872354348942, "全帯域_40ms_modulation_mae_db": 0.977730769639992, "300–1000 Hz_10ms_log_rms_mae_db": 1.7215039757173545, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21211078691592444, "300–1000 Hz_10ms_envelope_correlation": 0.9886658226097402, "300–1000 Hz_10ms_modulation_mae_db": 1.2230390440230963, "300–1000 Hz_40ms_log_rms_mae_db": 1.2933271155940511, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.19229284809201513, "300–1000 Hz_40ms_envelope_correlation": 0.9938607861186055, "300–1000 Hz_40ms_modulation_mae_db": 1.345444706278342, "1000–3000 Hz_10ms_log_rms_mae_db": 1.544876021708366, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1822008366776408, "1000–3000 Hz_10ms_envelope_correlation": 0.9781536520167178, "1000–3000 Hz_10ms_modulation_mae_db": 0.6575757930006798, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9784144147760401, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09854518864615447, "1000–3000 Hz_40ms_envelope_correlation": 0.9951604030165425, "1000–3000 Hz_40ms_modulation_mae_db": 0.7637271066823873, "3000–6000 Hz_10ms_log_rms_mae_db": 1.508333839074879, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20617298704244355, "3000–6000 Hz_10ms_envelope_correlation": 0.972600447109902, "3000–6000 Hz_10ms_modulation_mae_db": 0.917591907577093, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8938744980173275, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.09836879914050749, "3000–6000 Hz_40ms_envelope_correlation": 0.993888140259325, "3000–6000 Hz_40ms_modulation_mae_db": 1.1802388798086982, "mr_stft_db": 5.738466547433681, "band_mae_db": 0.6451923684985648, "rms_gain_error_db": 1.0159973886777973, "output_peak": 0.6444663938580104}, "full_scale": {"sample_count": 42560, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 42560, "output_samples": 42560, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.161190571178658, "spectral_convergence": 0.5576398967175402, "spectral_convergence_status": "ok", "amplitude_floor": 1.6028753096638316e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.279030847852485, "spectral_convergence": 0.3126541508769492, "spectral_convergence_status": "ok", "amplitude_floor": 1.0331614165752405e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7751782232699, "spectral_convergence": 0.3537500180067513, "spectral_convergence_status": "ok", "amplitude_floor": 7.831529815953997e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042273796895084704, "output_power": 0.0003783098872699982, "delta_db": -0.48223564610823444, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0017345552824327432, "output_power": 0.002118242881011665, "delta_db": 0.8678760960137561, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017389853785610774, "output_power": 0.002534507683714714, "delta_db": 1.635976817283042, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00033443608489121105, "output_power": 0.0003825862432254505, "delta_db": 0.5841621922819261, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00017248781948922851, "output_power": 0.0001803723430634169, "delta_db": 0.19411514792112558, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00012677719609633915, "output_power": 0.0001299331518559048, "delta_db": 0.1067883113833043, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6451923684985648, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0159973886777973, "pitch": null, "output_rms_dbfs": -22.423040023833643, "output_peak": 0.6444663938580104, "reference_rms_dbfs": -23.43903741251144, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.738466547433681}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 42560, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00020029577381909937, "log_rms_mae_db": 1.3817324164474896, "rms_envelope_nrmse": 0.1749660962446044, "envelope_correlation": 0.9867094158026204, "rms_gain_error_db": 1.0160003488892042, "reference_mean_power": 0.004538506592765079, "output_mean_power": 0.005734730397196418}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.8045928558777289e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008527253032911584, "output_power": 0.001168220036213544, "reference_active": true, "delta_db": 1.3671550039411193}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002893325022312121, "output_power": 0.000381807462841123, "reference_active": true, "delta_db": 1.204471899415045}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.722708282094668e-05, "output_power": 8.853673619372205e-05, "reference_active": true, "delta_db": -0.40663747310678167}], "active_band_count": 3, "active_band_mae_db": 0.9927547921543153}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00017887965336918488, "log_rms_mae_db": 1.0173752857179321, "rms_envelope_nrmse": 0.14866554327918494, "envelope_correlation": 0.9936872354348942, "rms_gain_error_db": 1.016005333093572, "reference_mean_power": 0.004590346759257262, "output_mean_power": 0.005800240842948021}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.4591224648550268e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007420820227306516, "output_power": 0.001008143003469778, "reference_active": true, "delta_db": 1.3307022971752716}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015779167985385838, "output_power": 0.00022507851879090805, "reference_active": true, "delta_db": 1.5424994875765314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.56670863086132e-06, "output_power": 6.476624142007865e-06, "reference_active": true, "delta_db": -0.05999052416817298}], "active_band_count": 3, "active_band_mae_db": 0.977730769639992}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00017771570730226456, "log_rms_mae_db": 1.7215039757173545, "rms_envelope_nrmse": 0.21211078691592444, "envelope_correlation": 0.9886658226097402, "rms_gain_error_db": 1.3678118282691167, "reference_mean_power": 0.003163634497093627, "output_mean_power": 0.004334784236508436}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.820001870942524e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009255585566014781, "output_power": 0.0013144557955014865, "reference_active": true, "delta_db": 1.523420854461077}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002598152375676334, "output_power": 0.0003662568050431776, "reference_active": true, "delta_db": 1.4912108475984518}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.760488032618478e-05, "output_power": 6.674834323619242e-05, "reference_active": true, "delta_db": -0.6544854300097601}], "active_band_count": 3, "active_band_mae_db": 1.2230390440230963}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 0.00016184521030355344, "log_rms_mae_db": 1.2933271155940511, "rms_envelope_nrmse": 0.19229284809201513, "envelope_correlation": 0.9938607861186055, "rms_gain_error_db": 1.3678136717778264, "reference_mean_power": 0.0031997824540876, "output_mean_power": 0.004384315711161192}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.5556561049407657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008585656786559776, "output_power": 0.0011954654157866605, "reference_active": true, "delta_db": 1.4376349323561128}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00015662214176101382, "output_power": 0.0002238070679300566, "reference_active": true, "delta_db": 1.5502063910883264}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.505375823689367e-06, "output_power": 5.895534218525788e-06, "reference_active": true, "delta_db": -1.0484927953905867}], "active_band_count": 3, "active_band_mae_db": 1.345444706278342}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 0.00011375403588925618, "log_rms_mae_db": 1.544876021708366, "rms_envelope_nrmse": 0.1822008366776408, "envelope_correlation": 0.9781536520167178, "rms_gain_error_db": 0.520168014449453, "reference_mean_power": 0.0003949518827357503, "output_mean_power": 0.00044520598077375545}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 1.6254954818630519e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.926011413944617e-05, "output_power": 8.971303255981865e-05, "reference_active": true, "delta_db": 0.5380084392323219}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0135009616582686e-05, "output_power": 1.9844500130614438e-05, "reference_active": true, "delta_db": -0.06311677564747985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.142336744382203e-06, "output_power": 5.208109490827191e-06, "reference_active": true, "delta_db": -1.3716021641222378}], "active_band_count": 3, "active_band_mae_db": 0.6575757930006798}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 8.70060383669589e-05, "log_rms_mae_db": 0.9784144147760401, "rms_envelope_nrmse": 0.09854518864615447, "envelope_correlation": 0.9951604030165425, "rms_gain_error_db": 0.5202089300952979, "reference_mean_power": 0.00039945208139827864, "output_mean_power": 0.00045028303168069935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.3661376304062792e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.270190105592004e-05, "output_power": 8.192901973873114e-05, "reference_active": true, "delta_db": 0.5189199144205305}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.983801499648517e-06, "output_power": 9.919809840763463e-06, "reference_active": true, "delta_db": -0.027925907136793035}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.054433356878425e-07, "output_power": 3.382508719937717e-07, "reference_active": true, "delta_db": -1.7443354984898385}], "active_band_count": 3, "active_band_mae_db": 0.7637271066823873}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 531, "rms_floor": 6.751108516993363e-05, "log_rms_mae_db": 1.508333839074879, "rms_envelope_nrmse": 0.20617298704244355, "envelope_correlation": 0.972600447109902, "rms_gain_error_db": 0.2504264464230353, "reference_mean_power": 0.00015129473773616968, "output_mean_power": 0.0001602752516686998}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 531, "frequency_resolution_hz": 0.3766478342749529, "power_floor": 2.0948865520528105e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010579126587837114, "output_power": 0.00010662450012860628, "reference_active": true, "delta_db": 0.034071942952710994}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.588508061007275e-05, "output_power": 1.534108726516441e-05, "reference_active": true, "delta_db": -0.15133282678339866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.6030646339801794e-06, "output_power": 3.10233388904901e-06, "reference_active": true, "delta_db": -2.5673709529951694}], "active_band_count": 3, "active_band_mae_db": 0.917591907577093}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 42560, "uncovered_tail_samples": 0, "metrics": {"frame_count": 525, "rms_floor": 5.7873331101151596e-05, "log_rms_mae_db": 0.8938744980173275, "rms_envelope_nrmse": 0.09836879914050749, "envelope_correlation": 0.993888140259325, "rms_gain_error_db": 0.2504389280224664, "reference_mean_power": 0.0001530185286286403, "output_mean_power": 0.00016210182877445347}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 525, "frequency_resolution_hz": 0.38095238095238093, "power_floor": 1.9708112174484802e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.70769534187325e-05, "output_power": 9.769520509659404e-05, "reference_active": true, "delta_db": 0.02757110625711859}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.36755214697274e-06, "output_power": 7.103853895182893e-06, "reference_active": true, "delta_db": -0.15829197699383446}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2457588916034453e-07, "output_power": 1.4991010207615255e-07, "reference_active": true, "delta_db": -3.3548535561751414}], "active_band_count": 3, "active_band_mae_db": 1.1802388798086982}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9397997609474298, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c7546101c6be29207166923f5d261734914c2e92917c277d33ba730657974f06", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9781536520167178, "standard_deviation": null, "minimum": 0.9781536520167178, "maximum": 0.9781536520167178, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.544876021708366, "standard_deviation": null, "minimum": 1.544876021708366, "maximum": 1.544876021708366, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6575757930006798, "standard_deviation": null, "minimum": 0.6575757930006798, "maximum": 0.6575757930006798, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1822008366776408, "standard_deviation": null, "minimum": 0.1822008366776408, "maximum": 0.1822008366776408, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951604030165425, "standard_deviation": null, "minimum": 0.9951604030165425, "maximum": 0.9951604030165425, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9784144147760401, "standard_deviation": null, "minimum": 0.9784144147760401, "maximum": 0.9784144147760401, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7637271066823873, "standard_deviation": null, "minimum": 0.7637271066823873, "maximum": 0.7637271066823873, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09854518864615447, "standard_deviation": null, "minimum": 0.09854518864615447, "maximum": 0.09854518864615447, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.972600447109902, "standard_deviation": null, "minimum": 0.972600447109902, "maximum": 0.972600447109902, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.508333839074879, "standard_deviation": null, "minimum": 1.508333839074879, "maximum": 1.508333839074879, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.917591907577093, "standard_deviation": null, "minimum": 0.917591907577093, "maximum": 0.917591907577093, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20617298704244355, "standard_deviation": null, "minimum": 0.20617298704244355, "maximum": 0.20617298704244355, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993888140259325, "standard_deviation": null, "minimum": 0.993888140259325, "maximum": 0.993888140259325, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8938744980173275, "standard_deviation": null, "minimum": 0.8938744980173275, "maximum": 0.8938744980173275, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1802388798086982, "standard_deviation": null, "minimum": 1.1802388798086982, "maximum": 1.1802388798086982, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09836879914050749, "standard_deviation": null, "minimum": 0.09836879914050749, "maximum": 0.09836879914050749, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9886658226097402, "standard_deviation": null, "minimum": 0.9886658226097402, "maximum": 0.9886658226097402, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7215039757173545, "standard_deviation": null, "minimum": 1.7215039757173545, "maximum": 1.7215039757173545, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.2230390440230963, "standard_deviation": null, "minimum": 1.2230390440230963, "maximum": 1.2230390440230963, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21211078691592444, "standard_deviation": null, "minimum": 0.21211078691592444, "maximum": 0.21211078691592444, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9938607861186055, "standard_deviation": null, "minimum": 0.9938607861186055, "maximum": 0.9938607861186055, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2933271155940511, "standard_deviation": null, "minimum": 1.2933271155940511, "maximum": 1.2933271155940511, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.345444706278342, "standard_deviation": null, "minimum": 1.345444706278342, "maximum": 1.345444706278342, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19229284809201513, "standard_deviation": null, "minimum": 0.19229284809201513, "maximum": 0.19229284809201513, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6451923684985648, "standard_deviation": null, "minimum": 0.6451923684985648, "maximum": 0.6451923684985648, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.738466547433681, "standard_deviation": null, "minimum": 5.738466547433681, "maximum": 5.738466547433681, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6444663938580104, "standard_deviation": null, "minimum": 0.6444663938580104, "maximum": 0.6444663938580104, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0159973886777973, "standard_deviation": null, "minimum": 1.0159973886777973, "maximum": 1.0159973886777973, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9867094158026204, "standard_deviation": null, "minimum": 0.9867094158026204, "maximum": 0.9867094158026204, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.3817324164474896, "standard_deviation": null, "minimum": 1.3817324164474896, "maximum": 1.3817324164474896, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9927547921543153, "standard_deviation": null, "minimum": 0.9927547921543153, "maximum": 0.9927547921543153, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1749660962446044, "standard_deviation": null, "minimum": 0.1749660962446044, "maximum": 0.1749660962446044, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9936872354348942, "standard_deviation": null, "minimum": 0.9936872354348942, "maximum": 0.9936872354348942, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0173752857179321, "standard_deviation": null, "minimum": 1.0173752857179321, "maximum": 1.0173752857179321, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.977730769639992, "standard_deviation": null, "minimum": 0.977730769639992, "maximum": 0.977730769639992, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.14866554327918494, "standard_deviation": null, "minimum": 0.14866554327918494, "maximum": 0.14866554327918494, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_6829"}, {"case": "librispeech_6930-75918-0013", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.300865650177002, "predicted_mos": 2.300865650177002, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 47040, "input_sample_rate": 16000, "input_seconds": 2.94, "input_float32_sha256": "f0483bd879a4001e518c807231b15e287d99139549abed37ff0a60cbc1df19fe", "internal_samples": 47040, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.418881600257009}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.8432308050872157, "全帯域_10ms_rms_envelope_nrmse": 0.20758081856716543, "全帯域_10ms_envelope_correlation": 0.9593224027141256, "全帯域_10ms_modulation_mae_db": 0.23736513270892542, "全帯域_40ms_log_rms_mae_db": 2.160587003580561, "全帯域_40ms_rms_envelope_nrmse": 0.12336057579312243, "全帯域_40ms_envelope_correlation": 0.985534974481588, "全帯域_40ms_modulation_mae_db": 0.3360616485477335, "300–1000 Hz_10ms_log_rms_mae_db": 2.0346627801689388, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2410738439334644, "300–1000 Hz_10ms_envelope_correlation": 0.9686066785823185, "300–1000 Hz_10ms_modulation_mae_db": 1.840678293018742, "300–1000 Hz_40ms_log_rms_mae_db": 1.5750578391708938, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.16815988407183383, "300–1000 Hz_40ms_envelope_correlation": 0.9891416363425017, "300–1000 Hz_40ms_modulation_mae_db": 1.6111004879477473, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6176894119838314, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22071338267926954, "1000–3000 Hz_10ms_envelope_correlation": 0.9665460567221872, "1000–3000 Hz_10ms_modulation_mae_db": 1.169677962771437, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0939865929113697, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14514054516068867, "1000–3000 Hz_40ms_envelope_correlation": 0.9879823811090027, "1000–3000 Hz_40ms_modulation_mae_db": 0.9856839544631373, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9115895887282575, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.28030253962403007, "3000–6000 Hz_10ms_envelope_correlation": 0.9425441540180864, "3000–6000 Hz_10ms_modulation_mae_db": 0.5769474294464001, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2259206099420463, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.16469638648214022, "3000–6000 Hz_40ms_envelope_correlation": 0.9787878336727078, "3000–6000 Hz_40ms_modulation_mae_db": 0.23784652064444464, "mr_stft_db": 6.023231837087702, "band_mae_db": 0.7961230719946414, "rms_gain_error_db": 0.426967863543967, "output_peak": 0.33108023979388235}, "full_scale": {"sample_count": 47040, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 47040, "output_samples": 47040, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.76184835533729, "spectral_convergence": 0.7227221148652648, "spectral_convergence_status": "ok", "amplitude_floor": 8.904239335448439e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.418780277452281, "spectral_convergence": 0.5382488041830503, "spectral_convergence_status": "ok", "amplitude_floor": 4.6801685305432975e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.889066878473536, "spectral_convergence": 0.592574680595908, "spectral_convergence_status": "ok", "amplitude_floor": 2.078367899935203e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 4.736571532646871e-05, "output_power": 3.466037026863795e-05, "delta_db": -1.3563090262335165, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00015972486768670453, "output_power": 0.0001928369781762833, "delta_db": 0.8181778027365992, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 3.942775929744416e-05, "output_power": 4.645108405557849e-05, "delta_db": 0.7119375726140473, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 6.533823001233142e-05, "output_power": 8.220854502470547e-05, "delta_db": 0.9974959644608616, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0005191422513822901, "output_power": 0.0005514150373553833, "delta_db": 0.2619222915029012, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017487193556096138, "output_power": 0.00020221328966570917, "delta_db": 0.6308957744199216, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7961230719946414, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.426967863543967, "pitch": null, "output_rms_dbfs": -29.547610302847893, "output_peak": 0.33108023979388235, "reference_rms_dbfs": -29.974578166391858, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.023231837087702}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 47040, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001280716783512989, "log_rms_mae_db": 2.8432308050872157, "rms_envelope_nrmse": 0.20758081856716543, "envelope_correlation": 0.9593224027141256, "rms_gain_error_db": 0.42703861324515147, "reference_mean_power": 0.0010075555569028208, "output_mean_power": 0.0011116622649710306}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 5.126669358054452e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020945501880404742, "output_power": 0.0002320646953213464, "reference_active": true, "delta_db": 0.4451830393497236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010240530560263733, "output_power": 0.00010775183017873083, "reference_active": true, "delta_db": 0.22102197492274708}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.779418868232472e-05, "output_power": 2.8089437332087382e-05, "reference_active": true, "delta_db": 0.04589038385430558}], "active_band_count": 3, "active_band_mae_db": 0.23736513270892542}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 9.465799671406843e-05, "log_rms_mae_db": 2.160587003580561, "rms_envelope_nrmse": 0.12336057579312243, "envelope_correlation": 0.985534974481588, "rms_gain_error_db": 0.4274584209576173, "reference_mean_power": 0.001017821671515535, "output_mean_power": 0.0011230976945758168}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 4.336511567866126e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018218362838522706, "output_power": 0.00019790774801500097, "reference_active": true, "delta_db": 0.359534497348014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6513405903536415e-05, "output_power": 5.968289483944526e-05, "reference_active": true, "delta_db": 0.23698398061975043}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.038210274571303e-06, "output_power": 1.8538830757691773e-06, "reference_active": true, "delta_db": -0.41166646767543613}], "active_band_count": 3, "active_band_mae_db": 0.3360616485477335}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.812061629421404e-05, "log_rms_mae_db": 2.0346627801689388, "rms_envelope_nrmse": 0.2410738439334644, "envelope_correlation": 0.9686066785823185, "rms_gain_error_db": 0.9723107977570462, "reference_mean_power": 0.00014923818668565985, "output_mean_power": 0.0001866856960186829}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 7.990988889403065e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7339937941838396e-05, "output_power": 5.6511048614327944e-05, "reference_active": true, "delta_db": 0.769056818304027}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.462900675807451e-06, "output_power": 7.198439081171739e-06, "reference_active": true, "delta_db": 0.46810852084216514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2961437099516358e-06, "output_power": 3.4764830046014085e-06, "reference_active": true, "delta_db": 4.284869539910034}], "active_band_count": 3, "active_band_mae_db": 1.840678293018742}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.488253980944051e-05, "log_rms_mae_db": 1.5750578391708938, "rms_envelope_nrmse": 0.16815988407183383, "envelope_correlation": 0.9891416363425017, "rms_gain_error_db": 0.9723066443723837, "reference_mean_power": 0.00015077911609921791, "output_mean_power": 0.00018861310191120353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 7.46020344966024e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.50390789043777e-05, "output_power": 5.421660339562213e-05, "reference_active": true, "delta_db": 0.8054280591072451}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3028613860270352e-06, "output_power": 3.7578166986407042e-06, "reference_active": true, "delta_db": 0.5604524429657678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.84375886973483e-08, "output_power": 1.9650743594369793e-07, "reference_active": true, "delta_db": 3.467420961770229}], "active_band_count": 3, "active_band_mae_db": 1.6111004879477473}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 4.5882621361725546e-05, "log_rms_mae_db": 1.6176894119838314, "rms_envelope_nrmse": 0.22071338267926954, "envelope_correlation": 0.9665460567221872, "rms_gain_error_db": 0.7326035805185471, "reference_mean_power": 0.0001648439551682589, "output_mean_power": 0.00019513419639715713}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 1.0813916284336453e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.944970033970189e-05, "output_power": 6.0877304333700134e-05, "reference_active": true, "delta_db": 0.9029174933166493}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1861966099880283e-05, "output_power": 1.3054897477874162e-05, "reference_active": true, "delta_db": 0.41616787054761517}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.815426369634166e-06, "output_power": 2.908304626813109e-06, "reference_active": true, "delta_db": -2.1899485244500467}], "active_band_count": 3, "active_band_mae_db": 1.169677962771437}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 4.018253092677299e-05, "log_rms_mae_db": 1.0939865929113697, "rms_envelope_nrmse": 0.14514054516068867, "envelope_correlation": 0.9879823811090027, "rms_gain_error_db": 0.7326149697675238, "reference_mean_power": 0.00016654362747028494, "output_mean_power": 0.00019714670219139144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 9.377377349368422e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.3529785361500865e-05, "output_power": 5.342677329404472e-05, "reference_active": true, "delta_db": 0.8897242006720021}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.594238028592104e-06, "output_power": 5.816797848264664e-06, "reference_active": true, "delta_db": 0.16943029807975596}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1488780558631956e-07, "output_power": 2.034070989954205e-07, "reference_active": true, "delta_db": -1.8978973646376538}], "active_band_count": 3, "active_band_mae_db": 0.9856839544631373}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 587, "rms_floor": 0.0001195123420429498, "log_rms_mae_db": 1.9115895887282575, "rms_envelope_nrmse": 0.28030253962403007, "envelope_correlation": 0.9425441540180864, "rms_gain_error_db": 0.33706471450956427, "reference_mean_power": 0.0005186016960204273, "output_mean_power": 0.000560454556317368}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 587, "frequency_resolution_hz": 0.34071550255536626, "power_floor": 3.8692936095348015e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017018269823121574, "output_power": 0.00018406782770226282, "reference_active": true, "delta_db": 0.34062481895297414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010605303684284744, "output_power": 9.478570583695751e-05, "reference_active": true, "delta_db": -0.4878026063581168}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2588117400162252e-05, "output_power": 2.780491185120478e-05, "reference_active": true, "delta_db": 0.9024148630281094}], "active_band_count": 3, "active_band_mae_db": 0.5769474294464001}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 47040, "uncovered_tail_samples": 0, "metrics": {"frame_count": 581, "rms_floor": 8.500195655960054e-05, "log_rms_mae_db": 1.2259206099420463, "rms_envelope_nrmse": 0.16469638648214022, "envelope_correlation": 0.9787878336727078, "rms_gain_error_db": 0.33706578346527916, "reference_mean_power": 0.0005239212873916433, "output_mean_power": 0.0005662035955484237}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 581, "frequency_resolution_hz": 0.3442340791738382, "power_floor": 3.3612492298744645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016844288389443048, "output_power": 0.00017800233556820367, "reference_active": true, "delta_db": 0.239730323971234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.87289448022482e-05, "output_power": 5.335072885191633e-05, "reference_active": true, "delta_db": -0.41711840911966}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9018720843055598e-06, "output_power": 1.8772072421135204e-06, "reference_active": true, "delta_db": -0.05669082884243979}], "active_band_count": 3, "active_band_mae_db": 0.23784652064444464}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8971918366993207, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "942755e9a3121e4750c0344114d598fe9cf84862b50cc87e21ca226b61d63d2c", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9665460567221872, "standard_deviation": null, "minimum": 0.9665460567221872, "maximum": 0.9665460567221872, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6176894119838314, "standard_deviation": null, "minimum": 1.6176894119838314, "maximum": 1.6176894119838314, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.169677962771437, "standard_deviation": null, "minimum": 1.169677962771437, "maximum": 1.169677962771437, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22071338267926954, "standard_deviation": null, "minimum": 0.22071338267926954, "maximum": 0.22071338267926954, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9879823811090027, "standard_deviation": null, "minimum": 0.9879823811090027, "maximum": 0.9879823811090027, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0939865929113697, "standard_deviation": null, "minimum": 1.0939865929113697, "maximum": 1.0939865929113697, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9856839544631373, "standard_deviation": null, "minimum": 0.9856839544631373, "maximum": 0.9856839544631373, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14514054516068867, "standard_deviation": null, "minimum": 0.14514054516068867, "maximum": 0.14514054516068867, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9425441540180864, "standard_deviation": null, "minimum": 0.9425441540180864, "maximum": 0.9425441540180864, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9115895887282575, "standard_deviation": null, "minimum": 1.9115895887282575, "maximum": 1.9115895887282575, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5769474294464001, "standard_deviation": null, "minimum": 0.5769474294464001, "maximum": 0.5769474294464001, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.28030253962403007, "standard_deviation": null, "minimum": 0.28030253962403007, "maximum": 0.28030253962403007, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9787878336727078, "standard_deviation": null, "minimum": 0.9787878336727078, "maximum": 0.9787878336727078, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2259206099420463, "standard_deviation": null, "minimum": 1.2259206099420463, "maximum": 1.2259206099420463, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.23784652064444464, "standard_deviation": null, "minimum": 0.23784652064444464, "maximum": 0.23784652064444464, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16469638648214022, "standard_deviation": null, "minimum": 0.16469638648214022, "maximum": 0.16469638648214022, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9686066785823185, "standard_deviation": null, "minimum": 0.9686066785823185, "maximum": 0.9686066785823185, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.0346627801689388, "standard_deviation": null, "minimum": 2.0346627801689388, "maximum": 2.0346627801689388, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.840678293018742, "standard_deviation": null, "minimum": 1.840678293018742, "maximum": 1.840678293018742, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2410738439334644, "standard_deviation": null, "minimum": 0.2410738439334644, "maximum": 0.2410738439334644, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9891416363425017, "standard_deviation": null, "minimum": 0.9891416363425017, "maximum": 0.9891416363425017, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5750578391708938, "standard_deviation": null, "minimum": 1.5750578391708938, "maximum": 1.5750578391708938, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.6111004879477473, "standard_deviation": null, "minimum": 1.6111004879477473, "maximum": 1.6111004879477473, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16815988407183383, "standard_deviation": null, "minimum": 0.16815988407183383, "maximum": 0.16815988407183383, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7961230719946414, "standard_deviation": null, "minimum": 0.7961230719946414, "maximum": 0.7961230719946414, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.023231837087702, "standard_deviation": null, "minimum": 6.023231837087702, "maximum": 6.023231837087702, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.33108023979388235, "standard_deviation": null, "minimum": 0.33108023979388235, "maximum": 0.33108023979388235, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.426967863543967, "standard_deviation": null, "minimum": 0.426967863543967, "maximum": 0.426967863543967, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9593224027141256, "standard_deviation": null, "minimum": 0.9593224027141256, "maximum": 0.9593224027141256, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.8432308050872157, "standard_deviation": null, "minimum": 2.8432308050872157, "maximum": 2.8432308050872157, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.23736513270892542, "standard_deviation": null, "minimum": 0.23736513270892542, "maximum": 0.23736513270892542, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20758081856716543, "standard_deviation": null, "minimum": 0.20758081856716543, "maximum": 0.20758081856716543, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.985534974481588, "standard_deviation": null, "minimum": 0.985534974481588, "maximum": 0.985534974481588, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.160587003580561, "standard_deviation": null, "minimum": 2.160587003580561, "maximum": 2.160587003580561, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.3360616485477335, "standard_deviation": null, "minimum": 0.3360616485477335, "maximum": 0.3360616485477335, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12336057579312243, "standard_deviation": null, "minimum": 0.12336057579312243, "maximum": 0.12336057579312243, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_6930"}, {"case": "librispeech_6930-76324-0024", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.287475347518921, "predicted_mos": 2.287475347518921, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 64800, "input_sample_rate": 16000, "input_seconds": 4.05, "input_float32_sha256": "864e2ec73bb84468dd63cf5f4e7a46bb9e30850cbaab1eabb9e65a998342ba7a", "internal_samples": 64800, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.955427600070834}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.4864408151222235, "全帯域_10ms_rms_envelope_nrmse": 0.2638696521734203, "全帯域_10ms_envelope_correlation": 0.9234289822125812, "全帯域_10ms_modulation_mae_db": 0.6565569298174421, "全帯域_40ms_log_rms_mae_db": 1.7735464752060963, "全帯域_40ms_rms_envelope_nrmse": 0.18173058363956504, "全帯域_40ms_envelope_correlation": 0.9599242163924414, "全帯域_40ms_modulation_mae_db": 0.43252110630330654, "300–1000 Hz_10ms_log_rms_mae_db": 2.6060929895530327, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22146101316955558, "300–1000 Hz_10ms_envelope_correlation": 0.9698441990686015, "300–1000 Hz_10ms_modulation_mae_db": 0.8020434471770833, "300–1000 Hz_40ms_log_rms_mae_db": 2.1161864633894405, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1665185198712064, "300–1000 Hz_40ms_envelope_correlation": 0.9865249878494065, "300–1000 Hz_40ms_modulation_mae_db": 0.7544929428528224, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7439051163132657, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.21454343692217068, "1000–3000 Hz_10ms_envelope_correlation": 0.9631825873038894, "1000–3000 Hz_10ms_modulation_mae_db": 1.7585907986328095, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0960747266943511, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13560061940967724, "1000–3000 Hz_40ms_envelope_correlation": 0.9855838222076359, "1000–3000 Hz_40ms_modulation_mae_db": 2.1292738922654615, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8143019993261367, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25296693704630185, "3000–6000 Hz_10ms_envelope_correlation": 0.9562166726854741, "3000–6000 Hz_10ms_modulation_mae_db": 1.8602511979606187, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0592939982319112, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14636897603790644, "3000–6000 Hz_40ms_envelope_correlation": 0.9858700183060279, "3000–6000 Hz_40ms_modulation_mae_db": 1.5426113328638043, "mr_stft_db": 6.230049564523231, "band_mae_db": 1.6846444325658174, "rms_gain_error_db": -0.15398467363722204, "output_peak": 0.48952640467168435}, "full_scale": {"sample_count": 64800, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 64800, "output_samples": 64800, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.010349366926333, "spectral_convergence": 0.745011086145682, "spectral_convergence_status": "ok", "amplitude_floor": 1.7331847370499167e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.637423334776363, "spectral_convergence": 0.6504404691246636, "spectral_convergence_status": "ok", "amplitude_floor": 1.0846095570828337e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.042375991866998, "spectral_convergence": 0.6626341712933435, "spectral_convergence_status": "ok", "amplitude_floor": 4.5533844599041527e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00024118742205032635, "output_power": 5.616372205025228e-05, "delta_db": -6.328987741628618, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 9.191122965591724e-05, "output_power": 0.0001106597141503982, "delta_db": 0.8062096758279995, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 7.435432994216677e-05, "output_power": 9.664853194404579e-05, "delta_db": 1.1388899738666864, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 2.4154339925686723e-05, "output_power": 3.0185664144337907e-05, "delta_db": 0.9680556176895412, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006210431709364446, "output_power": 0.0007244925102492388, "delta_db": 0.6691210953397018, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00023011417890304625, "output_power": 0.00021992931939253624, "delta_db": -0.1966024910423567, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.6846444325658174, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": -0.15398467363722204, "pitch": null, "output_rms_dbfs": -29.07251473270231, "output_peak": 0.48952640467168435, "reference_rms_dbfs": -28.918530059065088, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.230049564523231}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 64800, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.0001629307662014554, "log_rms_mae_db": 2.4864408151222235, "rms_envelope_nrmse": 0.2638696521734203, "envelope_correlation": 0.9234289822125812, "rms_gain_error_db": -0.15348335028969215, "reference_mean_power": 0.0012840985730840112, "output_mean_power": 0.0012395099813651096}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 7.3397819245367e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003947037290166488, "output_power": 0.000344879339826222, "reference_active": true, "delta_db": -0.5860405095762263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.980542214464163e-05, "output_power": 7.088712973224569e-05, "reference_active": true, "delta_db": 0.06678234021184798}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6036857037353872e-05, "output_power": 3.525927291657056e-05, "reference_active": true, "delta_db": 1.3168479396642518}], "active_band_count": 3, "active_band_mae_db": 0.6565569298174421}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.000144561944924798, "log_rms_mae_db": 1.7735464752060963, "rms_envelope_nrmse": 0.18173058363956504, "envelope_correlation": 0.9599242163924414, "rms_gain_error_db": -0.15307747342396813, "reference_mean_power": 0.0012928029739092497, "output_mean_power": 0.0012480287646845954}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 6.614408925582191e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037454976712341086, "output_power": 0.00033031787097451045, "reference_active": true, "delta_db": -0.5457746070320175}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9110632304140427e-05, "output_power": 3.125660903581479e-05, "reference_active": true, "delta_db": 0.30890221804311546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1315453531682465e-06, "output_power": 2.360387070326373e-06, "reference_active": true, "delta_db": 0.44288649383478657}], "active_band_count": 3, "active_band_mae_db": 0.43252110630330654}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 3.733123977436472e-05, "log_rms_mae_db": 2.6060929895530327, "rms_envelope_nrmse": 0.22146101316955558, "envelope_correlation": 0.9698441990686015, "rms_gain_error_db": 0.9474988459139831, "reference_mean_power": 0.00013419941652038683, "output_mean_power": 0.00016691697757215018}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.2377163404532544e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0020888510118017e-05, "output_power": 2.6495022191830418e-05, "reference_active": true, "delta_db": 1.2168094041580744}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.868636044401026e-06, "output_power": 9.721915979211018e-06, "reference_active": true, "delta_db": 0.9185240546978324}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3701828329102724e-06, "output_power": 1.4583380598008766e-06, "reference_active": true, "delta_db": 0.27079688267534346}], "active_band_count": 3, "active_band_mae_db": 0.8020434471770833}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.100732475570721e-05, "log_rms_mae_db": 2.1161864633894405, "rms_envelope_nrmse": 0.1665185198712064, "envelope_correlation": 0.9865249878494065, "rms_gain_error_db": 0.9475063871031348, "reference_mean_power": 0.00013520032575819994, "output_mean_power": 0.00016816219856805959}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 4.6212903051863556e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.815754516308902e-05, "output_power": 2.3769478742539537e-05, "reference_active": true, "delta_db": 1.1696252483502207}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8683197262456875e-06, "output_power": 4.712612882397011e-06, "reference_active": true, "delta_db": 0.8573940368949875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2646920446881671e-07, "output_power": 1.3354595589922896e-07, "reference_active": true, "delta_db": 0.23645954331325936}], "active_band_count": 3, "active_band_mae_db": 0.7544929428528224}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 4.662083951365485e-05, "log_rms_mae_db": 1.7439051163132657, "rms_envelope_nrmse": 0.21454343692217068, "envelope_correlation": 0.9631825873038894, "rms_gain_error_db": 0.4687061373299383, "reference_mean_power": 0.00017506320579060826, "output_mean_power": 0.00019501386549580465}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 9.028764248682407e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.2773698500231605e-05, "output_power": 4.4038541008163365e-05, "reference_active": true, "delta_db": 0.1265611852364913}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.603092418254862e-06, "output_power": 1.1607069507673203e-05, "reference_active": true, "delta_db": 2.449752093825134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4970630790605404e-06, "output_power": 6.511025222824701e-06, "reference_active": true, "delta_db": 2.699459116836804}], "active_band_count": 3, "active_band_mae_db": 1.7585907986328095}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 3.913562930844258e-05, "log_rms_mae_db": 1.0960747266943511, "rms_envelope_nrmse": 0.13560061940967724, "envelope_correlation": 0.9855838222076359, "rms_gain_error_db": 0.4691553851683189, "reference_mean_power": 0.00017624479575510055, "output_mean_power": 0.00019635042258893428}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 7.899096535479963e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.7672515224694995e-05, "output_power": 3.993995910382857e-05, "reference_active": true, "delta_db": 0.2538299877899615}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.050253319949336e-06, "output_power": 6.358578111861148e-06, "reference_active": true, "delta_db": 3.190241022506803}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.55667878122912e-07, "output_power": 5.035600102867086e-07, "reference_active": true, "delta_db": 2.94375066649962}], "active_band_count": 3, "active_band_mae_db": 2.1292738922654615}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 809, "rms_floor": 0.00015276264453937597, "log_rms_mae_db": 1.8143019993261367, "rms_envelope_nrmse": 0.25296693704630185, "envelope_correlation": 0.9562166726854741, "rms_gain_error_db": 0.4586583567442716, "reference_mean_power": 0.0006070254152411961, "output_mean_power": 0.0006746409678846521}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 809, "frequency_resolution_hz": 0.24721878862793573, "power_floor": 5.389907119886337e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002863305566547835, "output_power": 0.0003057942662258463, "reference_active": true, "delta_db": 0.2856164036462163}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.608243104449444e-05, "output_power": 5.472267022741236e-05, "reference_active": true, "delta_db": 1.8087149056941565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0810295137794261e-05, "output_power": 2.4125692020441856e-05, "reference_active": true, "delta_db": 3.4864222845414834}], "active_band_count": 3, "active_band_mae_db": 1.8602511979606187}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 64800, "uncovered_tail_samples": 0, "metrics": {"frame_count": 803, "rms_floor": 0.00013215266595876854, "log_rms_mae_db": 1.0592939982319112, "rms_envelope_nrmse": 0.14636897603790644, "envelope_correlation": 0.9858700183060279, "rms_gain_error_db": 0.4595608484470571, "reference_mean_power": 0.0006109513084873906, "output_mean_power": 0.0006791452759260106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 803, "frequency_resolution_hz": 0.24906600249066005, "power_floor": 5.083168365591421e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002672957849080471, "output_power": 0.00029777133078247217, "reference_active": true, "delta_db": 0.46890771606237275}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4768224147952787e-05, "output_power": 2.181106962985774e-05, "reference_active": true, "delta_db": 1.6934868886971506}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.446503663842908e-07, "output_power": 1.4901188691805672e-06, "reference_active": true, "delta_db": 2.4654393938318897}], "active_band_count": 3, "active_band_mae_db": 1.5426113328638043}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8465789416638737, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d51f86c417bae4cab08e26dc8d3ea4c89ffe40d395741181806759e1058c64b2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9631825873038894, "standard_deviation": null, "minimum": 0.9631825873038894, "maximum": 0.9631825873038894, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7439051163132657, "standard_deviation": null, "minimum": 1.7439051163132657, "maximum": 1.7439051163132657, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.7585907986328095, "standard_deviation": null, "minimum": 1.7585907986328095, "maximum": 1.7585907986328095, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21454343692217068, "standard_deviation": null, "minimum": 0.21454343692217068, "maximum": 0.21454343692217068, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9855838222076359, "standard_deviation": null, "minimum": 0.9855838222076359, "maximum": 0.9855838222076359, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0960747266943511, "standard_deviation": null, "minimum": 1.0960747266943511, "maximum": 1.0960747266943511, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 2.1292738922654615, "standard_deviation": null, "minimum": 2.1292738922654615, "maximum": 2.1292738922654615, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13560061940967724, "standard_deviation": null, "minimum": 0.13560061940967724, "maximum": 0.13560061940967724, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9562166726854741, "standard_deviation": null, "minimum": 0.9562166726854741, "maximum": 0.9562166726854741, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8143019993261367, "standard_deviation": null, "minimum": 1.8143019993261367, "maximum": 1.8143019993261367, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.8602511979606187, "standard_deviation": null, "minimum": 1.8602511979606187, "maximum": 1.8602511979606187, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25296693704630185, "standard_deviation": null, "minimum": 0.25296693704630185, "maximum": 0.25296693704630185, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9858700183060279, "standard_deviation": null, "minimum": 0.9858700183060279, "maximum": 0.9858700183060279, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0592939982319112, "standard_deviation": null, "minimum": 1.0592939982319112, "maximum": 1.0592939982319112, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5426113328638043, "standard_deviation": null, "minimum": 1.5426113328638043, "maximum": 1.5426113328638043, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14636897603790644, "standard_deviation": null, "minimum": 0.14636897603790644, "maximum": 0.14636897603790644, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9698441990686015, "standard_deviation": null, "minimum": 0.9698441990686015, "maximum": 0.9698441990686015, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.6060929895530327, "standard_deviation": null, "minimum": 2.6060929895530327, "maximum": 2.6060929895530327, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8020434471770833, "standard_deviation": null, "minimum": 0.8020434471770833, "maximum": 0.8020434471770833, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22146101316955558, "standard_deviation": null, "minimum": 0.22146101316955558, "maximum": 0.22146101316955558, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9865249878494065, "standard_deviation": null, "minimum": 0.9865249878494065, "maximum": 0.9865249878494065, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.1161864633894405, "standard_deviation": null, "minimum": 2.1161864633894405, "maximum": 2.1161864633894405, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7544929428528224, "standard_deviation": null, "minimum": 0.7544929428528224, "maximum": 0.7544929428528224, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1665185198712064, "standard_deviation": null, "minimum": 0.1665185198712064, "maximum": 0.1665185198712064, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.6846444325658174, "standard_deviation": null, "minimum": 1.6846444325658174, "maximum": 1.6846444325658174, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.230049564523231, "standard_deviation": null, "minimum": 6.230049564523231, "maximum": 6.230049564523231, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.48952640467168435, "standard_deviation": null, "minimum": 0.48952640467168435, "maximum": 0.48952640467168435, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": -0.15398467363722204, "standard_deviation": null, "minimum": -0.15398467363722204, "maximum": -0.15398467363722204, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9234289822125812, "standard_deviation": null, "minimum": 0.9234289822125812, "maximum": 0.9234289822125812, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.4864408151222235, "standard_deviation": null, "minimum": 2.4864408151222235, "maximum": 2.4864408151222235, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6565569298174421, "standard_deviation": null, "minimum": 0.6565569298174421, "maximum": 0.6565569298174421, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2638696521734203, "standard_deviation": null, "minimum": 0.2638696521734203, "maximum": 0.2638696521734203, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9599242163924414, "standard_deviation": null, "minimum": 0.9599242163924414, "maximum": 0.9599242163924414, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.7735464752060963, "standard_deviation": null, "minimum": 1.7735464752060963, "maximum": 1.7735464752060963, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.43252110630330654, "standard_deviation": null, "minimum": 0.43252110630330654, "maximum": 0.43252110630330654, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18173058363956504, "standard_deviation": null, "minimum": 0.18173058363956504, "maximum": 0.18173058363956504, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_6930"}, {"case": "librispeech_6930-81414-0026", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.176114320755005, "predicted_mos": 2.176114320755005, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49200, "input_sample_rate": 16000, "input_seconds": 3.075, "input_float32_sha256": "376663be52fa670832da53238ea7d15cc4542d0b0f921fa2a6951eb03b32b6a0", "internal_samples": 49200, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.8771635997109115}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.0834930749604834, "全帯域_10ms_rms_envelope_nrmse": 0.2079807152596612, "全帯域_10ms_envelope_correlation": 0.9591691653199365, "全帯域_10ms_modulation_mae_db": 1.9049822574813007, "全帯域_40ms_log_rms_mae_db": 1.5159283879273804, "全帯域_40ms_rms_envelope_nrmse": 0.14234441878769907, "全帯域_40ms_envelope_correlation": 0.9858760414902662, "全帯域_40ms_modulation_mae_db": 1.5410849344818294, "300–1000 Hz_10ms_log_rms_mae_db": 2.227553885241236, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.37136883697412554, "300–1000 Hz_10ms_envelope_correlation": 0.925632681937948, "300–1000 Hz_10ms_modulation_mae_db": 5.691104274320305, "300–1000 Hz_40ms_log_rms_mae_db": 1.7281434905795818, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.28141072362596203, "300–1000 Hz_40ms_envelope_correlation": 0.9689348701801322, "300–1000 Hz_40ms_modulation_mae_db": 5.670230934464446, "1000–3000 Hz_10ms_log_rms_mae_db": 1.547882113435803, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1968483364575148, "1000–3000 Hz_10ms_envelope_correlation": 0.9701595290163786, "1000–3000 Hz_10ms_modulation_mae_db": 1.9272233969022727, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9924262761877923, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11086316323166295, "1000–3000 Hz_40ms_envelope_correlation": 0.9927289158543845, "1000–3000 Hz_40ms_modulation_mae_db": 1.8898864874313857, "3000–6000 Hz_10ms_log_rms_mae_db": 1.558498035035606, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2163083360969513, "3000–6000 Hz_10ms_envelope_correlation": 0.9563015278896213, "3000–6000 Hz_10ms_modulation_mae_db": 1.1874030234770736, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9157123662326077, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12173208282461825, "3000–6000 Hz_40ms_envelope_correlation": 0.987275336286109, "3000–6000 Hz_40ms_modulation_mae_db": 0.9591567466332093, "mr_stft_db": 6.121747538152451, "band_mae_db": 1.0772752440552387, "rms_gain_error_db": 0.8060389123248317, "output_peak": 0.5004980393634809}, "full_scale": {"sample_count": 49200, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49200, "output_samples": 49200, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.679674742519985, "spectral_convergence": 0.6676745423384394, "spectral_convergence_status": "ok", "amplitude_floor": 6.39186387874099e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.442526034808792, "spectral_convergence": 0.5309518553762409, "spectral_convergence_status": "ok", "amplitude_floor": 4.1641566181737356e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.2430418371285725, "spectral_convergence": 0.6301489097298127, "spectral_convergence_status": "ok", "amplitude_floor": 3.471921787916526e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.471560290376807e-05, "output_power": 5.1562871047641254e-05, "delta_db": -1.6107421416274974, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0002820835097806379, "output_power": 0.00042941152935459964, "delta_db": 1.824960020836359, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026780543872196867, "output_power": 0.0003696956976175542, "delta_db": 1.4002500381591831, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00028630139823372175, "output_power": 0.0003313145037002372, "delta_db": 0.634169790471394, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0006167999147673381, "output_power": 0.0006840050104697166, "delta_db": 0.4491497799349797, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00037968124815517825, "output_power": 0.00043038499130739253, "delta_db": 0.5443796933020177, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.0772752440552387, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8060389123248317, "pitch": null, "output_rms_dbfs": -26.3895726481001, "output_peak": 0.5004980393634809, "reference_rms_dbfs": -27.195611560424933, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.121747538152451}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49200, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 0.00010992699921133752, "log_rms_mae_db": 2.0834930749604834, "rms_envelope_nrmse": 0.2079807152596612, "envelope_correlation": 0.9591691653199365, "rms_gain_error_db": 0.8060657276890737, "reference_mean_power": 0.0019104810338505282, "output_mean_power": 0.002300113692851935}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 4.701235881229352e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003075331225832077, "output_power": 0.00041400364285874, "reference_active": true, "delta_db": 1.2911226460357845}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8607043791082874e-05, "output_power": 4.5425757354429035e-05, "reference_active": true, "delta_db": 2.0082919596809283}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1458430393639735e-05, "output_power": 1.9983812560801974e-05, "reference_active": true, "delta_db": 2.4155321667271896}], "active_band_count": 3, "active_band_mae_db": 1.9049822574813007}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 9.464213695185308e-05, "log_rms_mae_db": 1.5159283879273804, "rms_envelope_nrmse": 0.14234441878769907, "envelope_correlation": 0.9858760414902662, "rms_gain_error_db": 0.806610985475994, "reference_mean_power": 0.0019290839559002245, "output_mean_power": 0.002322802194067244}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 4.127760176364045e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002739478020838986, "output_power": 0.0003704975096543689, "reference_active": true, "delta_db": 1.3111747277220553}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7486141765201905e-05, "output_power": 2.3124800171464977e-05, "reference_active": true, "delta_db": 1.2138399353718488}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.8391125647823e-07, "output_power": 9.466095063668232e-07, "reference_active": true, "delta_db": 2.0982401403515842}], "active_band_count": 3, "active_band_mae_db": 1.5410849344818294}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 6.478025356347348e-05, "log_rms_mae_db": 2.227553885241236, "rms_envelope_nrmse": 0.37136883697412554, "envelope_correlation": 0.925632681937948, "rms_gain_error_db": 1.6272544786957703, "reference_mean_power": 0.0004974887922363912, "output_mean_power": 0.0007236169795969163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.1951630543463553e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001563474072286115, "output_power": 0.000234382955553112, "reference_active": true, "delta_db": 1.7583534282016846}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.160597194093242e-06, "output_power": 3.551415144231637e-05, "reference_active": true, "delta_db": 6.386795006414449}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0536955073753773e-06, "output_power": 1.6045471811190248e-05, "reference_active": true, "delta_db": 8.928164388344783}], "active_band_count": 3, "active_band_mae_db": 5.691104274320305}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 5.647638237474489e-05, "log_rms_mae_db": 1.7281434905795818, "rms_envelope_nrmse": 0.28141072362596203, "envelope_correlation": 0.9689348701801322, "rms_gain_error_db": 1.6272353749276245, "reference_mean_power": 0.00050239739349007, "output_mean_power": 0.000730753518381456}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.0265825460069485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014180308081562638, "output_power": 0.00022565810779943455, "reference_active": true, "delta_db": 2.0176527571448446}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.4691763208394e-06, "output_power": 1.7648350700385024e-05, "reference_active": true, "delta_db": 5.964766362281969}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.310040798443831e-08, "output_power": 7.4435301624995e-07, "reference_active": true, "delta_db": 9.028273683966525}], "active_band_count": 3, "active_band_mae_db": 5.670230934464446}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 7.855020595588598e-05, "log_rms_mae_db": 1.547882113435803, "rms_envelope_nrmse": 0.1968483364575148, "envelope_correlation": 0.9701595290163786, "rms_gain_error_db": 0.5508770161682643, "reference_mean_power": 0.0004940955013379756, "output_mean_power": 0.0005609169983802789}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 1.3474634060369984e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.364408948380063e-05, "output_power": 9.42875931162803e-05, "reference_active": true, "delta_db": 0.02974178329557673}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.996833569277459e-06, "output_power": 1.0716290925873285e-05, "reference_active": true, "delta_db": 1.8514295166736572}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2622375016010212e-06, "output_power": 5.553772511095843e-06, "reference_active": true, "delta_db": 3.900498890737585}], "active_band_count": 3, "active_band_mae_db": 1.9272233969022727}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 6.30584220600245e-05, "log_rms_mae_db": 0.9924262761877923, "rms_envelope_nrmse": 0.11086316323166295, "envelope_correlation": 0.9927289158543845, "rms_gain_error_db": 0.5508776006356509, "reference_mean_power": 0.0004989699589215838, "output_mean_power": 0.0005664507540511602}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 1.243238840477423e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.827683631948728e-05, "output_power": 8.926634274300587e-05, "reference_active": true, "delta_db": 0.048409815729467454}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8441959267659476e-06, "output_power": 5.645830926360089e-06, "reference_active": true, "delta_db": 1.6692235414126175}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.483409815447214e-08, "output_power": 2.355955749247841e-07, "reference_active": true, "delta_db": 3.952026105152072}], "active_band_count": 3, "active_band_mae_db": 1.8898864874313857}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 614, "rms_floor": 9.512970769085293e-05, "log_rms_mae_db": 1.558498035035606, "rms_envelope_nrmse": 0.2163083360969513, "envelope_correlation": 0.9563015278896213, "rms_gain_error_db": 0.4869005977990233, "reference_mean_power": 0.0007033343720650757, "output_mean_power": 0.0007867774459002883}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 614, "frequency_resolution_hz": 0.3257328990228013, "power_floor": 2.9605352394438525e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001169604983627503, "output_power": 0.00013584232838092532, "reference_active": true, "delta_db": 0.6499590678541561}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1203242070906313e-05, "output_power": 3.0537884156124795e-05, "reference_active": true, "delta_db": 1.5843667170950444}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.096712651632301e-05, "output_power": 1.4889536614723737e-05, "reference_active": true, "delta_db": 1.3278832854820202}], "active_band_count": 3, "active_band_mae_db": 1.1874030234770736}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49200, "uncovered_tail_samples": 0, "metrics": {"frame_count": 608, "rms_floor": 7.035164526339387e-05, "log_rms_mae_db": 0.9157123662326077, "rms_envelope_nrmse": 0.12173208282461825, "envelope_correlation": 0.987275336286109, "rms_gain_error_db": 0.4869014813048001, "reference_mean_power": 0.0007102740547296548, "output_mean_power": 0.0007945406090613826}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 608, "frequency_resolution_hz": 0.32894736842105265, "power_floor": 2.6373521771123283e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010173798550296795, "output_power": 0.00012064857851081269, "reference_active": true, "delta_db": 0.7403907540264596}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.290709427246265e-05, "output_power": 1.6076970898452245e-05, "reference_active": true, "delta_db": 0.9537574350937421}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.316189457175965e-07, "output_power": 8.294445415843659e-07, "reference_active": true, "delta_db": 1.183322050779426}], "active_band_count": 3, "active_band_mae_db": 0.9591567466332093}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.811054382174257, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e9b96b5c10f0cd21e591ae953b157a0e0ed85412fa5f3c684fad72c7f9653665", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9701595290163786, "standard_deviation": null, "minimum": 0.9701595290163786, "maximum": 0.9701595290163786, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.547882113435803, "standard_deviation": null, "minimum": 1.547882113435803, "maximum": 1.547882113435803, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.9272233969022727, "standard_deviation": null, "minimum": 1.9272233969022727, "maximum": 1.9272233969022727, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1968483364575148, "standard_deviation": null, "minimum": 0.1968483364575148, "maximum": 0.1968483364575148, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9927289158543845, "standard_deviation": null, "minimum": 0.9927289158543845, "maximum": 0.9927289158543845, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9924262761877923, "standard_deviation": null, "minimum": 0.9924262761877923, "maximum": 0.9924262761877923, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.8898864874313857, "standard_deviation": null, "minimum": 1.8898864874313857, "maximum": 1.8898864874313857, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11086316323166295, "standard_deviation": null, "minimum": 0.11086316323166295, "maximum": 0.11086316323166295, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9563015278896213, "standard_deviation": null, "minimum": 0.9563015278896213, "maximum": 0.9563015278896213, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.558498035035606, "standard_deviation": null, "minimum": 1.558498035035606, "maximum": 1.558498035035606, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1874030234770736, "standard_deviation": null, "minimum": 1.1874030234770736, "maximum": 1.1874030234770736, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2163083360969513, "standard_deviation": null, "minimum": 0.2163083360969513, "maximum": 0.2163083360969513, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.987275336286109, "standard_deviation": null, "minimum": 0.987275336286109, "maximum": 0.987275336286109, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9157123662326077, "standard_deviation": null, "minimum": 0.9157123662326077, "maximum": 0.9157123662326077, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9591567466332093, "standard_deviation": null, "minimum": 0.9591567466332093, "maximum": 0.9591567466332093, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12173208282461825, "standard_deviation": null, "minimum": 0.12173208282461825, "maximum": 0.12173208282461825, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.925632681937948, "standard_deviation": null, "minimum": 0.925632681937948, "maximum": 0.925632681937948, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.227553885241236, "standard_deviation": null, "minimum": 2.227553885241236, "maximum": 2.227553885241236, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 5.691104274320305, "standard_deviation": null, "minimum": 5.691104274320305, "maximum": 5.691104274320305, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.37136883697412554, "standard_deviation": null, "minimum": 0.37136883697412554, "maximum": 0.37136883697412554, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9689348701801322, "standard_deviation": null, "minimum": 0.9689348701801322, "maximum": 0.9689348701801322, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7281434905795818, "standard_deviation": null, "minimum": 1.7281434905795818, "maximum": 1.7281434905795818, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 5.670230934464446, "standard_deviation": null, "minimum": 5.670230934464446, "maximum": 5.670230934464446, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.28141072362596203, "standard_deviation": null, "minimum": 0.28141072362596203, "maximum": 0.28141072362596203, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.0772752440552387, "standard_deviation": null, "minimum": 1.0772752440552387, "maximum": 1.0772752440552387, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.121747538152451, "standard_deviation": null, "minimum": 6.121747538152451, "maximum": 6.121747538152451, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5004980393634809, "standard_deviation": null, "minimum": 0.5004980393634809, "maximum": 0.5004980393634809, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8060389123248317, "standard_deviation": null, "minimum": 0.8060389123248317, "maximum": 0.8060389123248317, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9591691653199365, "standard_deviation": null, "minimum": 0.9591691653199365, "maximum": 0.9591691653199365, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.0834930749604834, "standard_deviation": null, "minimum": 2.0834930749604834, "maximum": 2.0834930749604834, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.9049822574813007, "standard_deviation": null, "minimum": 1.9049822574813007, "maximum": 1.9049822574813007, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2079807152596612, "standard_deviation": null, "minimum": 0.2079807152596612, "maximum": 0.2079807152596612, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9858760414902662, "standard_deviation": null, "minimum": 0.9858760414902662, "maximum": 0.9858760414902662, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.5159283879273804, "standard_deviation": null, "minimum": 1.5159283879273804, "maximum": 1.5159283879273804, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.5410849344818294, "standard_deviation": null, "minimum": 1.5410849344818294, "maximum": 1.5410849344818294, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.14234441878769907, "standard_deviation": null, "minimum": 0.14234441878769907, "maximum": 0.14234441878769907, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_6930"}, {"case": "librispeech_7021-79740-0002", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.0683696269989014, "predicted_mos": 3.0683696269989014, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "cf3c23698da232f7def5a53f5ef5ab47897164909a950ddfbd493d58824d7d69", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.394103199709207}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6253652440094974, "全帯域_10ms_rms_envelope_nrmse": 0.20026323947906555, "全帯域_10ms_envelope_correlation": 0.9800816754970373, "全帯域_10ms_modulation_mae_db": 0.7146201443704085, "全帯域_40ms_log_rms_mae_db": 1.1978004541481007, "全帯域_40ms_rms_envelope_nrmse": 0.1639865522943413, "全帯域_40ms_envelope_correlation": 0.9901113065892115, "全帯域_40ms_modulation_mae_db": 1.1894796757544752, "300–1000 Hz_10ms_log_rms_mae_db": 1.9060516096212892, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.3133798719870052, "300–1000 Hz_10ms_envelope_correlation": 0.9728950255022101, "300–1000 Hz_10ms_modulation_mae_db": 1.0919951404102044, "300–1000 Hz_40ms_log_rms_mae_db": 1.4832983207197425, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.27139175887037825, "300–1000 Hz_40ms_envelope_correlation": 0.9859930246767363, "300–1000 Hz_40ms_modulation_mae_db": 1.1428430011042965, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6534785222984816, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.229736074140684, "1000–3000 Hz_10ms_envelope_correlation": 0.9731055426442256, "1000–3000 Hz_10ms_modulation_mae_db": 1.0286673505925061, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2773430309317861, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1403803785877823, "1000–3000 Hz_40ms_envelope_correlation": 0.9929211652050484, "1000–3000 Hz_40ms_modulation_mae_db": 0.8905864481791189, "3000–6000 Hz_10ms_log_rms_mae_db": 1.700556996520391, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25762211775020805, "3000–6000 Hz_10ms_envelope_correlation": 0.9714531154174011, "3000–6000 Hz_10ms_modulation_mae_db": 1.5347430372615012, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2421378065289348, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17952020553607215, "3000–6000 Hz_40ms_envelope_correlation": 0.9899783062152666, "3000–6000 Hz_40ms_modulation_mae_db": 1.6683239254870978, "mr_stft_db": 5.352003029362746, "band_mae_db": 0.9299532730427366, "rms_gain_error_db": 1.0003359546726527, "output_peak": 0.6586828823428705}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.202090568362617, "spectral_convergence": 0.633411337759037, "spectral_convergence_status": "ok", "amplitude_floor": 1.8465213976940698e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.909530511905556, "spectral_convergence": 0.30870955036742664, "spectral_convergence_status": "ok", "amplitude_floor": 1.4089022879434144e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 4.9443880078200655, "spectral_convergence": 0.31187067585556844, "spectral_convergence_status": "ok", "amplitude_floor": 1.0260349239920231e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002262162256014784, "output_power": 0.002637091529321745, "delta_db": 0.6660145179284516, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0014828754084214857, "output_power": 0.002062395032348791, "delta_db": 1.4326719090822706, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00026062691889949753, "output_power": 0.0003635516987805289, "delta_db": 1.4454690844832918, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00020447377727008734, "output_power": 0.0002455597875096733, "delta_db": 0.7951962906114711, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 4.078753769921038e-05, "output_power": 4.458378825386148e-05, "delta_db": 0.38649479208917453, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.800369964531977e-05, "output_power": 3.408809373175813e-05, "delta_db": 0.8538730440617601, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9299532730427366, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0003359546726527, "pitch": null, "output_rms_dbfs": -22.686312635204523, "output_peak": 0.6586828823428705, "reference_rms_dbfs": -23.686648589877173, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.352003029362746}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00022610981418459577, "log_rms_mae_db": 1.6253652440094974, "rms_envelope_nrmse": 0.20026323947906555, "envelope_correlation": 0.9800816754970373, "rms_gain_error_db": 1.002150635231271, "reference_mean_power": 0.00428070632223443, "output_mean_power": 0.005391759318835322}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.20587323032393e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001500467490243704, "output_power": 0.0018645961852928705, "reference_active": true, "delta_db": 0.9435820112211125}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013253274491379155, "output_power": 0.0001456661782679382, "reference_active": true, "delta_db": 0.41035533296246657}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.272532196136399e-05, "output_power": 2.7282968341317766e-05, "reference_active": true, "delta_db": -0.7899230889276465}], "active_band_count": 3, "active_band_mae_db": 0.7146201443704085}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00019856877416737668, "log_rms_mae_db": 1.1978004541481007, "rms_envelope_nrmse": 0.1639865522943413, "envelope_correlation": 0.9901113065892115, "rms_gain_error_db": 1.0053942022321252, "reference_mean_power": 0.004296662851053013, "output_mean_power": 0.005415900752148461}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.009968768189794e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014002280466658247, "output_power": 0.0017372916244696426, "reference_active": true, "delta_db": 0.9367395356059578}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3360463722840526e-05, "output_power": 6.037980849412245e-05, "reference_active": true, "delta_db": 0.5367213613286828}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.498078024970164e-06, "output_power": 1.542084549786438e-06, "reference_active": true, "delta_db": -2.094978130328785}], "active_band_count": 3, "active_band_mae_db": 1.1894796757544752}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.000212457290437723, "log_rms_mae_db": 1.9060516096212892, "rms_envelope_nrmse": 0.3133798719870052, "envelope_correlation": 0.9728950255022101, "rms_gain_error_db": 1.7261921131325533, "reference_mean_power": 0.0012391682631200216, "output_mean_power": 0.0018439514971891658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.921692896601052e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00036532110371103373, "output_power": 0.0004965533201188196, "reference_active": true, "delta_db": 1.3329112900037474}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.3180096390035205e-05, "output_power": 6.097694806370754e-05, "reference_active": true, "delta_db": 1.4988207638342923}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1750484667690236e-05, "output_power": 1.3016107203651884e-05, "reference_active": true, "delta_db": 0.44425336739257354}], "active_band_count": 3, "active_band_mae_db": 1.0919951404102044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018128574308840577, "log_rms_mae_db": 1.4832983207197425, "rms_envelope_nrmse": 0.27139175887037825, "envelope_correlation": 0.9859930246767363, "rms_gain_error_db": 1.728300686205927, "reference_mean_power": 0.0012444705639415084, "output_mean_power": 0.001852740935264415}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.573537659747316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003540640123239247, "output_power": 0.00048209527426686186, "reference_active": true, "delta_db": 1.3405108777530896}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0107771514146375e-05, "output_power": 2.7491112067896065e-05, "reference_active": true, "delta_db": 1.3582836656963355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.973740035567712e-07, "output_power": 6.740445886520897e-07, "reference_active": true, "delta_db": -0.7297344598634643}], "active_band_count": 3, "active_band_mae_db": 1.1428430011042965}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001386433646872303, "log_rms_mae_db": 1.6534785222984816, "rms_envelope_nrmse": 0.229736074140684, "envelope_correlation": 0.9731055426442256, "rms_gain_error_db": 0.7744083024385402, "reference_mean_power": 0.00021050674147203463, "output_mean_power": 0.00025159779978959105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9020912968770603e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.015318920526352e-05, "output_power": 0.00010224380858877274, "reference_active": true, "delta_db": 1.0571621149140382}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8603151739641485e-05, "output_power": 4.814002329476142e-05, "reference_active": true, "delta_db": 0.9588353216529587}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0682238748218467e-05, "output_power": 1.616579568747897e-05, "reference_active": true, "delta_db": -1.0700046152105216}], "active_band_count": 3, "active_band_mae_db": 1.0286673505925061}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.950462169412164e-05, "log_rms_mae_db": 1.2773430309317861, "rms_envelope_nrmse": 0.1403803785877823, "envelope_correlation": 0.9929211652050484, "rms_gain_error_db": 0.7785710788046009, "reference_mean_power": 0.0002110528457073582, "output_mean_power": 0.00025249240566689525}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7773203833688661e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.04144461796749e-05, "output_power": 0.0001120859311782138, "reference_active": true, "delta_db": 0.933132777276092}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.4052463298203573e-05, "output_power": 2.7795697418723396e-05, "reference_active": true, "delta_db": 0.6281801471725534}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4632154850806674e-06, "output_power": 1.1330880167508561e-06, "reference_active": true, "delta_db": -1.110446420088711}], "active_band_count": 3, "active_band_mae_db": 0.8905864481791189}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.6125957709065377e-05, "log_rms_mae_db": 1.700556996520391, "rms_envelope_nrmse": 0.25762211775020805, "envelope_correlation": 0.9714531154174011, "rms_gain_error_db": 0.9895585267001741, "reference_mean_power": 2.4404568133397883e-05, "output_mean_power": 3.064975302936684e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.103592290864487e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1042852972405107e-05, "output_power": 1.368872847223873e-05, "reference_active": true, "delta_db": 0.9328181908320167}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8700267025557585e-06, "output_power": 5.177917849382669e-06, "reference_active": true, "delta_db": 1.2644119423342741}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.131168166122288e-06, "output_power": 1.968915474851152e-06, "reference_active": true, "delta_db": 2.406998978618213}], "active_band_count": 3, "active_band_mae_db": 1.5347430372615012}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.642002234154929e-05, "log_rms_mae_db": 1.2421378065289348, "rms_envelope_nrmse": 0.17952020553607215, "envelope_correlation": 0.9899783062152666, "rms_gain_error_db": 0.9887979541162848, "reference_mean_power": 2.4522755214627932e-05, "output_mean_power": 3.0792791293734895e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9657184130662428e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.1868339578364968e-05, "output_power": 1.5206440317043895e-05, "reference_active": true, "delta_db": 1.076375978367616}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9915566770914906e-06, "output_power": 2.77413797465947e-06, "reference_active": true, "delta_db": 1.4393538699295485}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.546174499350975e-08, "output_power": 1.5159886471691862e-07, "reference_active": true, "delta_db": 2.4892419281641294}], "active_band_count": 3, "active_band_mae_db": 1.6683239254870978}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9135670355247533, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3fe816b33e5eb851ed2b90ead98523c6064bb630aec23ec41669db4f046cb3ed", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9731055426442256, "standard_deviation": null, "minimum": 0.9731055426442256, "maximum": 0.9731055426442256, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6534785222984816, "standard_deviation": null, "minimum": 1.6534785222984816, "maximum": 1.6534785222984816, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0286673505925061, "standard_deviation": null, "minimum": 1.0286673505925061, "maximum": 1.0286673505925061, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.229736074140684, "standard_deviation": null, "minimum": 0.229736074140684, "maximum": 0.229736074140684, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9929211652050484, "standard_deviation": null, "minimum": 0.9929211652050484, "maximum": 0.9929211652050484, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2773430309317861, "standard_deviation": null, "minimum": 1.2773430309317861, "maximum": 1.2773430309317861, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8905864481791189, "standard_deviation": null, "minimum": 0.8905864481791189, "maximum": 0.8905864481791189, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1403803785877823, "standard_deviation": null, "minimum": 0.1403803785877823, "maximum": 0.1403803785877823, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9714531154174011, "standard_deviation": null, "minimum": 0.9714531154174011, "maximum": 0.9714531154174011, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.700556996520391, "standard_deviation": null, "minimum": 1.700556996520391, "maximum": 1.700556996520391, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5347430372615012, "standard_deviation": null, "minimum": 1.5347430372615012, "maximum": 1.5347430372615012, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25762211775020805, "standard_deviation": null, "minimum": 0.25762211775020805, "maximum": 0.25762211775020805, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9899783062152666, "standard_deviation": null, "minimum": 0.9899783062152666, "maximum": 0.9899783062152666, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2421378065289348, "standard_deviation": null, "minimum": 1.2421378065289348, "maximum": 1.2421378065289348, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.6683239254870978, "standard_deviation": null, "minimum": 1.6683239254870978, "maximum": 1.6683239254870978, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17952020553607215, "standard_deviation": null, "minimum": 0.17952020553607215, "maximum": 0.17952020553607215, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9728950255022101, "standard_deviation": null, "minimum": 0.9728950255022101, "maximum": 0.9728950255022101, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9060516096212892, "standard_deviation": null, "minimum": 1.9060516096212892, "maximum": 1.9060516096212892, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0919951404102044, "standard_deviation": null, "minimum": 1.0919951404102044, "maximum": 1.0919951404102044, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3133798719870052, "standard_deviation": null, "minimum": 0.3133798719870052, "maximum": 0.3133798719870052, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9859930246767363, "standard_deviation": null, "minimum": 0.9859930246767363, "maximum": 0.9859930246767363, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4832983207197425, "standard_deviation": null, "minimum": 1.4832983207197425, "maximum": 1.4832983207197425, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1428430011042965, "standard_deviation": null, "minimum": 1.1428430011042965, "maximum": 1.1428430011042965, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.27139175887037825, "standard_deviation": null, "minimum": 0.27139175887037825, "maximum": 0.27139175887037825, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9299532730427366, "standard_deviation": null, "minimum": 0.9299532730427366, "maximum": 0.9299532730427366, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.352003029362746, "standard_deviation": null, "minimum": 5.352003029362746, "maximum": 5.352003029362746, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6586828823428705, "standard_deviation": null, "minimum": 0.6586828823428705, "maximum": 0.6586828823428705, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0003359546726527, "standard_deviation": null, "minimum": 1.0003359546726527, "maximum": 1.0003359546726527, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9800816754970373, "standard_deviation": null, "minimum": 0.9800816754970373, "maximum": 0.9800816754970373, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6253652440094974, "standard_deviation": null, "minimum": 1.6253652440094974, "maximum": 1.6253652440094974, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7146201443704085, "standard_deviation": null, "minimum": 0.7146201443704085, "maximum": 0.7146201443704085, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20026323947906555, "standard_deviation": null, "minimum": 0.20026323947906555, "maximum": 0.20026323947906555, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9901113065892115, "standard_deviation": null, "minimum": 0.9901113065892115, "maximum": 0.9901113065892115, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1978004541481007, "standard_deviation": null, "minimum": 1.1978004541481007, "maximum": 1.1978004541481007, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1894796757544752, "standard_deviation": null, "minimum": 1.1894796757544752, "maximum": 1.1894796757544752, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1639865522943413, "standard_deviation": null, "minimum": 0.1639865522943413, "maximum": 0.1639865522943413, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7021"}, {"case": "librispeech_7021-79740-0011", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.456448793411255, "predicted_mos": 3.456448793411255, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "0b0b7429c8327c3435eebf64bbb745f752fcc49e357c4781ec03122c05cdf3d3", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.972521599847823}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4901808559623813, "全帯域_10ms_rms_envelope_nrmse": 0.1670748013120895, "全帯域_10ms_envelope_correlation": 0.9827015504783592, "全帯域_10ms_modulation_mae_db": 0.6019163930785646, "全帯域_40ms_log_rms_mae_db": 1.1774236404507479, "全帯域_40ms_rms_envelope_nrmse": 0.12217508607437991, "全帯域_40ms_envelope_correlation": 0.9940118846078752, "全帯域_40ms_modulation_mae_db": 1.005093747233684, "300–1000 Hz_10ms_log_rms_mae_db": 1.815973760259867, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2254158714342678, "300–1000 Hz_10ms_envelope_correlation": 0.9779168494511317, "300–1000 Hz_10ms_modulation_mae_db": 0.7202713556062726, "300–1000 Hz_40ms_log_rms_mae_db": 1.5564175674676661, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.18490911006221095, "300–1000 Hz_40ms_envelope_correlation": 0.9881071015974113, "300–1000 Hz_40ms_modulation_mae_db": 0.9243484975838197, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5029995916108105, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.22364623654912585, "1000–3000 Hz_10ms_envelope_correlation": 0.9817995690441031, "1000–3000 Hz_10ms_modulation_mae_db": 2.176925682666356, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2195707385609844, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16385403411815141, "1000–3000 Hz_40ms_envelope_correlation": 0.9942922114734009, "1000–3000 Hz_40ms_modulation_mae_db": 2.2317425397962007, "3000–6000 Hz_10ms_log_rms_mae_db": 1.3253910574966186, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.19307246287656749, "3000–6000 Hz_10ms_envelope_correlation": 0.9797145446976581, "3000–6000 Hz_10ms_modulation_mae_db": 0.4005707614615687, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8769607010741707, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1122067189746195, "3000–6000 Hz_40ms_envelope_correlation": 0.9937458622409119, "3000–6000 Hz_40ms_modulation_mae_db": 0.6904178034739067, "mr_stft_db": 5.418336543001818, "band_mae_db": 0.7574639904132576, "rms_gain_error_db": 0.7636606366101841, "output_peak": 0.58411816608275}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.046519029609278, "spectral_convergence": 0.6238685387023037, "spectral_convergence_status": "ok", "amplitude_floor": 1.6898412267444875e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.965709954046959, "spectral_convergence": 0.262343159901242, "spectral_convergence_status": "ok", "amplitude_floor": 1.0945626100175482e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.242780645349217, "spectral_convergence": 0.296691971431143, "spectral_convergence_status": "ok", "amplitude_floor": 6.58943811507865e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017588563541266742, "output_power": 0.002051048131839004, "delta_db": 0.6674547999686758, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0012071674288590822, "output_power": 0.0014938411992232552, "delta_db": 0.9253692381710793, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003927549262280726, "output_power": 0.0004355320566168863, "delta_db": 0.4489848505273369, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003207071656425, "output_power": 0.0004061126636509165, "delta_db": 1.0253786854471663, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.4115036625953016e-05, "output_power": 6.807332633368333e-05, "delta_db": 0.9965901575947989, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.305762815858983e-05, "output_power": 4.8100625502218337e-05, "delta_db": 0.4810062107704893, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7574639904132576, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7636606366101841, "pitch": null, "output_rms_dbfs": -23.465262157564794, "output_peak": 0.58411816608275, "reference_rms_dbfs": -24.228922794174977, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.418336543001818}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002198976712894503, "log_rms_mae_db": 1.4901808559623813, "rms_envelope_nrmse": 0.1670748013120895, "envelope_correlation": 0.9827015504783592, "rms_gain_error_db": 0.767215735249514, "reference_mean_power": 0.0037766562988771884, "output_mean_power": 0.004506392717920301}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.069519300574134e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006799849644412746, "output_power": 0.0007693275559764181, "reference_active": true, "delta_db": 0.5361197852563475}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022275189153607582, "output_power": 0.0002728451665692656, "reference_active": true, "delta_db": 0.8809486408360148}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.238302101062001e-05, "output_power": 3.8754651347206e-05, "reference_active": true, "delta_db": -0.38868075314333145}], "active_band_count": 3, "active_band_mae_db": 0.6019163930785646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018113949472991762, "log_rms_mae_db": 1.1774236404507479, "rms_envelope_nrmse": 0.12217508607437991, "envelope_correlation": 0.9940118846078752, "rms_gain_error_db": 0.7735115333853326, "reference_mean_power": 0.003780325824366865, "output_mean_power": 0.004517315111354591}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9093794448028363e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006511405512347742, "output_power": 0.0007340087504104628, "reference_active": true, "delta_db": 0.5202649449457326}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001166725476043267, "output_power": 0.00014706299695510917, "reference_active": true, "delta_db": 1.0053473107876982}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.399065253982206e-06, "output_power": 2.4120849752931653e-06, "reference_active": true, "delta_db": -1.4896689859676215}], "active_band_count": 3, "active_band_mae_db": 1.005093747233684}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013526658282762358, "log_rms_mae_db": 1.815973760259867, "rms_envelope_nrmse": 0.2254158714342678, "envelope_correlation": 0.9779168494511317, "rms_gain_error_db": 1.0667673459617324, "reference_mean_power": 0.001047203990993015, "output_mean_power": 0.0013387763238130012}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.584657585149468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003504540347149448, "output_power": 0.00041675200237141323, "reference_active": true, "delta_db": 0.7524663085950906}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.5576057457372e-05, "output_power": 0.00012697382299454207, "reference_active": true, "delta_db": 1.2336508398330412}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9623586773434205e-05, "output_power": 2.042904579233072e-05, "reference_active": true, "delta_db": 0.17469691839068624}], "active_band_count": 3, "active_band_mae_db": 0.7202713556062726}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012031399971827572, "log_rms_mae_db": 1.5564175674676661, "rms_envelope_nrmse": 0.18490911006221095, "envelope_correlation": 0.9881071015974113, "rms_gain_error_db": 1.0676131658761743, "reference_mean_power": 0.0010510881038568295, "output_mean_power": 0.001344003616657548}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.061243701565635e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00034327462417827815, "output_power": 0.0004060271844838297, "reference_active": true, "delta_db": 0.7291341124935782}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.625199841508274e-05, "output_power": 6.288889656374349e-05, "reference_active": true, "delta_db": 1.3344347266056042}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5417688576161987e-06, "output_power": 1.3093977903150149e-06, "reference_active": true, "delta_db": -0.7094766536522765}], "active_band_count": 3, "active_band_mae_db": 0.9243484975838197}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00013601220330575488, "log_rms_mae_db": 1.5029995916108105, "rms_envelope_nrmse": 0.22364623654912585, "envelope_correlation": 0.9817995690441031, "rms_gain_error_db": 1.0552075837552588, "reference_mean_power": 0.00024923569921471797, "output_mean_power": 0.00031778325335476656}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.70738105242738e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001701759171968848, "output_power": 0.0002176786226221162, "reference_active": true, "delta_db": 1.0691768075008352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6462817859538857e-05, "output_power": 3.558234862785841e-05, "reference_active": true, "delta_db": 1.285985228736099}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.097841074264338e-06, "output_power": 8.102529762149912e-06, "reference_active": true, "delta_db": 4.175615011762134}], "active_band_count": 3, "active_band_mae_db": 2.176925682666356}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011511226838452259, "log_rms_mae_db": 1.2195707385609844, "rms_envelope_nrmse": 0.16385403411815141, "envelope_correlation": 0.9942922114734009, "rms_gain_error_db": 1.0583687851921888, "reference_mean_power": 0.00024963838926192667, "output_mean_power": 0.00031852846611280753}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.6223780118670214e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00016948627965662844, "output_power": 0.00021963405030418458, "reference_active": true, "delta_db": 1.1256512381168156}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1767373976886565e-05, "output_power": 1.6357160364094768e-05, "reference_active": true, "delta_db": 1.4302835557296103}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8952968721807594e-07, "output_power": 4.915939488861636e-07, "reference_active": true, "delta_db": 4.139292825542177}], "active_band_count": 3, "active_band_mae_db": 2.2317425397962007}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.3800826555528964e-05, "log_rms_mae_db": 1.3253910574966186, "rms_envelope_nrmse": 0.19307246287656749, "envelope_correlation": 0.9797145446976581, "rms_gain_error_db": 0.3879232152035663, "reference_mean_power": 4.104766861392576e-05, "output_mean_power": 4.488289038806404e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.890895346126465e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.549789437651056e-06, "output_power": 1.058020355107289e-05, "reference_active": true, "delta_db": 0.44500227137152654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.275656140671785e-06, "output_power": 3.126167615668859e-06, "reference_active": true, "delta_db": -0.2028604590620951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.939546621487433e-07, "output_power": 5.228394392619589e-07, "reference_active": true, "delta_db": -0.5538495539510844}], "active_band_count": 3, "active_band_mae_db": 0.4005707614615687}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.772714769720092e-05, "log_rms_mae_db": 0.8769607010741707, "rms_envelope_nrmse": 0.1122067189746195, "envelope_correlation": 0.9937458622409119, "rms_gain_error_db": 0.3886439529025397, "reference_mean_power": 4.122071638798201e-05, "output_mean_power": 4.5079587204527746e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.778938094011469e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.936665760227185e-06, "output_power": 1.0888215331157044e-05, "reference_active": true, "delta_db": 0.39716019423757964}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7465237791634235e-06, "output_power": 1.6293450253748756e-06, "reference_active": true, "delta_db": -0.30161444007168514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8043700755065915e-08, "output_power": 3.5025841305599036e-08, "reference_active": true, "delta_db": -1.3724787761124553}], "active_band_count": 3, "active_band_mae_db": 0.6904178034739067}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9326563023506298, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "f890d14d2f54491d9a96f4bb58b10814396be0cfddf3fabede650f8ba401c5fb", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9817995690441031, "standard_deviation": null, "minimum": 0.9817995690441031, "maximum": 0.9817995690441031, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5029995916108105, "standard_deviation": null, "minimum": 1.5029995916108105, "maximum": 1.5029995916108105, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 2.176925682666356, "standard_deviation": null, "minimum": 2.176925682666356, "maximum": 2.176925682666356, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22364623654912585, "standard_deviation": null, "minimum": 0.22364623654912585, "maximum": 0.22364623654912585, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9942922114734009, "standard_deviation": null, "minimum": 0.9942922114734009, "maximum": 0.9942922114734009, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2195707385609844, "standard_deviation": null, "minimum": 1.2195707385609844, "maximum": 1.2195707385609844, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 2.2317425397962007, "standard_deviation": null, "minimum": 2.2317425397962007, "maximum": 2.2317425397962007, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16385403411815141, "standard_deviation": null, "minimum": 0.16385403411815141, "maximum": 0.16385403411815141, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9797145446976581, "standard_deviation": null, "minimum": 0.9797145446976581, "maximum": 0.9797145446976581, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.3253910574966186, "standard_deviation": null, "minimum": 1.3253910574966186, "maximum": 1.3253910574966186, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4005707614615687, "standard_deviation": null, "minimum": 0.4005707614615687, "maximum": 0.4005707614615687, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19307246287656749, "standard_deviation": null, "minimum": 0.19307246287656749, "maximum": 0.19307246287656749, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9937458622409119, "standard_deviation": null, "minimum": 0.9937458622409119, "maximum": 0.9937458622409119, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8769607010741707, "standard_deviation": null, "minimum": 0.8769607010741707, "maximum": 0.8769607010741707, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6904178034739067, "standard_deviation": null, "minimum": 0.6904178034739067, "maximum": 0.6904178034739067, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1122067189746195, "standard_deviation": null, "minimum": 0.1122067189746195, "maximum": 0.1122067189746195, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9779168494511317, "standard_deviation": null, "minimum": 0.9779168494511317, "maximum": 0.9779168494511317, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.815973760259867, "standard_deviation": null, "minimum": 1.815973760259867, "maximum": 1.815973760259867, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.7202713556062726, "standard_deviation": null, "minimum": 0.7202713556062726, "maximum": 0.7202713556062726, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2254158714342678, "standard_deviation": null, "minimum": 0.2254158714342678, "maximum": 0.2254158714342678, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9881071015974113, "standard_deviation": null, "minimum": 0.9881071015974113, "maximum": 0.9881071015974113, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5564175674676661, "standard_deviation": null, "minimum": 1.5564175674676661, "maximum": 1.5564175674676661, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9243484975838197, "standard_deviation": null, "minimum": 0.9243484975838197, "maximum": 0.9243484975838197, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18490911006221095, "standard_deviation": null, "minimum": 0.18490911006221095, "maximum": 0.18490911006221095, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7574639904132576, "standard_deviation": null, "minimum": 0.7574639904132576, "maximum": 0.7574639904132576, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.418336543001818, "standard_deviation": null, "minimum": 5.418336543001818, "maximum": 5.418336543001818, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.58411816608275, "standard_deviation": null, "minimum": 0.58411816608275, "maximum": 0.58411816608275, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7636606366101841, "standard_deviation": null, "minimum": 0.7636606366101841, "maximum": 0.7636606366101841, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9827015504783592, "standard_deviation": null, "minimum": 0.9827015504783592, "maximum": 0.9827015504783592, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4901808559623813, "standard_deviation": null, "minimum": 1.4901808559623813, "maximum": 1.4901808559623813, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6019163930785646, "standard_deviation": null, "minimum": 0.6019163930785646, "maximum": 0.6019163930785646, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1670748013120895, "standard_deviation": null, "minimum": 0.1670748013120895, "maximum": 0.1670748013120895, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9940118846078752, "standard_deviation": null, "minimum": 0.9940118846078752, "maximum": 0.9940118846078752, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1774236404507479, "standard_deviation": null, "minimum": 1.1774236404507479, "maximum": 1.1774236404507479, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.005093747233684, "standard_deviation": null, "minimum": 1.005093747233684, "maximum": 1.005093747233684, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12217508607437991, "standard_deviation": null, "minimum": 0.12217508607437991, "maximum": 0.12217508607437991, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7021"}, {"case": "librispeech_7021-85628-0025", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.387463092803955, "predicted_mos": 3.387463092803955, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 44400, "input_sample_rate": 16000, "input_seconds": 2.775, "input_float32_sha256": "0b1cc36f10bab5ada3eda5a0cbb0f3cb6e9105bbc5712a22ede5df76e42dc3eb", "internal_samples": 44400, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.198673400096595}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.071754788309549, "全帯域_10ms_rms_envelope_nrmse": 0.15165677101954703, "全帯域_10ms_envelope_correlation": 0.9867139342258024, "全帯域_10ms_modulation_mae_db": 0.6788409283307445, "全帯域_40ms_log_rms_mae_db": 1.6991262765129582, "全帯域_40ms_rms_envelope_nrmse": 0.1054341308824508, "全帯域_40ms_envelope_correlation": 0.9958380556748785, "全帯域_40ms_modulation_mae_db": 0.9386844972291467, "300–1000 Hz_10ms_log_rms_mae_db": 1.7089222448874513, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21709973708612498, "300–1000 Hz_10ms_envelope_correlation": 0.9766740875718397, "300–1000 Hz_10ms_modulation_mae_db": 0.9494147706773077, "300–1000 Hz_40ms_log_rms_mae_db": 1.6088895622853514, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.15656784875983468, "300–1000 Hz_40ms_envelope_correlation": 0.9904984175151293, "300–1000 Hz_40ms_modulation_mae_db": 1.3285153735268604, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5234721972130754, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2127448739554468, "1000–3000 Hz_10ms_envelope_correlation": 0.9765633849797124, "1000–3000 Hz_10ms_modulation_mae_db": 0.9786015661228409, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0566909476969757, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1222601847808516, "1000–3000 Hz_40ms_envelope_correlation": 0.9945539195453164, "1000–3000 Hz_40ms_modulation_mae_db": 1.3011149911957764, "3000–6000 Hz_10ms_log_rms_mae_db": 1.770383596715316, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25786750773534567, "3000–6000 Hz_10ms_envelope_correlation": 0.9519602799341975, "3000–6000 Hz_10ms_modulation_mae_db": 0.11477115108363806, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0595955614593693, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14600664641858846, "3000–6000 Hz_40ms_envelope_correlation": 0.9837597427919645, "3000–6000 Hz_40ms_modulation_mae_db": 0.5047780241741828, "mr_stft_db": 5.830085953624703, "band_mae_db": 0.5324164044091749, "rms_gain_error_db": 0.6547792997361889, "output_peak": 0.5308888280545734}, "full_scale": {"sample_count": 44400, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 44400, "output_samples": 44400, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.515129971225314, "spectral_convergence": 0.7012848971555158, "spectral_convergence_status": "ok", "amplitude_floor": 1.585777823535548e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.488828884341858, "spectral_convergence": 0.24866968926304273, "spectral_convergence_status": "ok", "amplitude_floor": 7.694479165248034e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.486299005306936, "spectral_convergence": 0.22873590239719424, "spectral_convergence_status": "ok", "amplitude_floor": 6.411814451459315e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00186841781908955, "output_power": 0.0021633928817568066, "delta_db": 0.6366139590913099, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005255584463676, "output_power": 0.0006392077862880938, "delta_db": 0.8502103642233907, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00025377475865110864, "output_power": 0.0002722611938006627, "delta_db": 0.30537321046349786, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00014708318852606447, "output_power": 0.0001762398003547029, "delta_db": 0.7854095607001983, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 9.522695850195022e-06, "output_power": 1.0725775410818425e-05, "delta_db": 0.5166878552372708, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.071083630879648e-06, "output_power": 5.932610932944943e-06, "delta_db": -0.10020347673938115, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5324164044091749, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.6547792997361889, "pitch": null, "output_rms_dbfs": -24.857495913315105, "output_peak": 0.5308888280545734, "reference_rms_dbfs": -25.512275213051296, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.830085953624703}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 44400, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 0.00014622417526559514, "log_rms_mae_db": 2.071754788309549, "rms_envelope_nrmse": 0.15165677101954703, "envelope_correlation": 0.9867139342258024, "rms_gain_error_db": 0.6547803559777103, "reference_mean_power": 0.0028155003489198547, "output_mean_power": 0.0032736603708197263}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.019039396656961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008418483111337408, "output_power": 0.0009870613175423139, "reference_active": true, "delta_db": 0.6911028747319298}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001428627689371398, "output_power": 0.00014289175575934017, "reference_active": true, "delta_db": 0.0008810930980771341}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9192224262112575e-05, "output_power": 1.40822298317249e-05, "reference_active": true, "delta_db": -1.3445388171622268}], "active_band_count": 3, "active_band_mae_db": 0.6788409283307445}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 0.00013893685565863478, "log_rms_mae_db": 1.6991262765129582, "rms_envelope_nrmse": 0.1054341308824508, "envelope_correlation": 0.9958380556748785, "rms_gain_error_db": 0.6547833469559188, "reference_mean_power": 0.0028463248833832306, "output_mean_power": 0.0033095031915049097}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.8422021834325858e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007582285194051116, "output_power": 0.0008877528510593034, "reference_active": true, "delta_db": 0.6849196012746388}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661169958011628e-05, "output_power": 6.882943392378583e-05, "reference_active": true, "delta_db": 0.14223683077300647}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7101796636835213e-06, "output_power": 1.0818125910561132e-06, "reference_active": true, "delta_db": -1.988897059639795}], "active_band_count": 3, "active_band_mae_db": 0.9386844972291467}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 8.753945313428826e-05, "log_rms_mae_db": 1.7089222448874513, "rms_envelope_nrmse": 0.21709973708612498, "envelope_correlation": 0.9766740875718397, "rms_gain_error_db": 0.8438144956782806, "reference_mean_power": 0.0005201241069420424, "output_mean_power": 0.0006316673547711772}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 5.057137989146479e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002701505367125327, "output_power": 0.00035193488742687947, "reference_active": true, "delta_db": 1.1485648603502039}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.4045222297363644e-05, "output_power": 5.800359078426831e-05, "reference_active": true, "delta_db": 0.3069757242363516}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.5635511670477814e-06, "output_power": 5.488521584505999e-06, "reference_active": true, "delta_db": -1.3927037274453675}], "active_band_count": 3, "active_band_mae_db": 0.9494147706773077}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 7.440703161891069e-05, "log_rms_mae_db": 1.6088895622853514, "rms_envelope_nrmse": 0.15656784875983468, "envelope_correlation": 0.9904984175151293, "rms_gain_error_db": 0.8438141547734485, "reference_mean_power": 0.0005258188868038473, "output_mean_power": 0.0006385833588774156}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 4.6537380366659205e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00025071266719842557, "output_power": 0.00032427235402894094, "reference_active": true, "delta_db": 1.1173364716011367}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3806665370442115e-05, "output_power": 2.7624699059982e-05, "reference_active": true, "delta_db": 0.6459898807849623}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0797322752712393e-06, "output_power": 6.472828593401904e-07, "reference_active": true, "delta_db": -2.222219768194482}], "active_band_count": 3, "active_band_mae_db": 1.3285153735268604}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 5.3294712554339956e-05, "log_rms_mae_db": 1.5234721972130754, "rms_envelope_nrmse": 0.2127448739554468, "envelope_correlation": 0.9765633849797124, "rms_gain_error_db": 0.6687917422752822, "reference_mean_power": 0.00010733282194912887, "output_mean_power": 0.00012520213134255016}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 1.1016693525679884e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.607161402154175e-05, "output_power": 7.79830846533667e-05, "reference_active": true, "delta_db": 0.7198549384842303}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0221078826476055e-05, "output_power": 9.753344092344463e-06, "reference_active": true, "delta_db": -0.20343191499453167}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.376144897878774e-06, "output_power": 2.124072288422348e-06, "reference_active": true, "delta_db": -2.0125178448897607}], "active_band_count": 3, "active_band_mae_db": 0.9786015661228409}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 4.1357257962539876e-05, "log_rms_mae_db": 1.0566909476969757, "rms_envelope_nrmse": 0.1222601847808516, "envelope_correlation": 0.9945539195453164, "rms_gain_error_db": 0.6687917402808985, "reference_mean_power": 0.00010850799286328448, "output_mean_power": 0.00012657295057783528}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 1.0153781665834753e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.269855261867839e-05, "output_power": 7.261924070326254e-05, "reference_active": true, "delta_db": 0.6379418831697776}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.20737764206477e-06, "output_power": 4.098518362545976e-06, "reference_active": true, "delta_db": -0.11384609593106636}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.246816673622798e-07, "output_power": 1.0874566130941362e-07, "reference_active": true, "delta_db": -3.1515569944864854}], "active_band_count": 3, "active_band_mae_db": 1.3011149911957764}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 554, "rms_floor": 1.102648736164134e-05, "log_rms_mae_db": 1.770383596715316, "rms_envelope_nrmse": 0.25786750773534567, "envelope_correlation": 0.9519602799341975, "rms_gain_error_db": 0.24707249318724556, "reference_mean_power": 5.352135557413054e-06, "output_mean_power": 5.665449254358131e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 554, "frequency_resolution_hz": 0.36101083032490977, "power_floor": 2.71214230899019e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.798500894794614e-06, "output_power": 1.9110743679736767e-06, "reference_active": true, "delta_db": 0.263669293902171}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2650926706890884e-07, "output_power": 4.3181876183681314e-07, "reference_active": true, "delta_db": 0.05373035981207642}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0854303858213565e-07, "output_power": 1.0921778262986023e-07, "reference_active": true, "delta_db": 0.026913799536666776}], "active_band_count": 3, "active_band_mae_db": 0.11477115108363806}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 44400, "uncovered_tail_samples": 0, "metrics": {"frame_count": 548, "rms_floor": 9.954651742580568e-06, "log_rms_mae_db": 1.0595955614593693, "rms_envelope_nrmse": 0.14600664641858846, "envelope_correlation": 0.9837597427919645, "rms_gain_error_db": 0.24707162799978383, "reference_mean_power": 5.410729753830906e-06, "output_mean_power": 5.727472410537767e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 548, "frequency_resolution_hz": 0.36496350364963503, "power_floor": 2.394714459074953e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.814627629314412e-06, "output_power": 1.9158002486749197e-06, "reference_active": true, "delta_db": 0.23562706185321625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0097168111890166e-07, "output_power": 2.0155641259799255e-07, "reference_active": true, "delta_db": 0.012617545823633461}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.411981302222167e-09, "output_power": 7.031890660816385e-09, "reference_active": true, "delta_db": -1.2660894648456988}], "active_band_count": 3, "active_band_mae_db": 0.5047780241741828}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9389630332594973, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "ec75066ce8a7be86057c9d213bd309652eb8d4065173750cf29bd00c15748e03", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9765633849797124, "standard_deviation": null, "minimum": 0.9765633849797124, "maximum": 0.9765633849797124, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5234721972130754, "standard_deviation": null, "minimum": 1.5234721972130754, "maximum": 1.5234721972130754, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9786015661228409, "standard_deviation": null, "minimum": 0.9786015661228409, "maximum": 0.9786015661228409, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2127448739554468, "standard_deviation": null, "minimum": 0.2127448739554468, "maximum": 0.2127448739554468, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945539195453164, "standard_deviation": null, "minimum": 0.9945539195453164, "maximum": 0.9945539195453164, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0566909476969757, "standard_deviation": null, "minimum": 1.0566909476969757, "maximum": 1.0566909476969757, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.3011149911957764, "standard_deviation": null, "minimum": 1.3011149911957764, "maximum": 1.3011149911957764, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1222601847808516, "standard_deviation": null, "minimum": 0.1222601847808516, "maximum": 0.1222601847808516, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9519602799341975, "standard_deviation": null, "minimum": 0.9519602799341975, "maximum": 0.9519602799341975, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.770383596715316, "standard_deviation": null, "minimum": 1.770383596715316, "maximum": 1.770383596715316, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.11477115108363806, "standard_deviation": null, "minimum": 0.11477115108363806, "maximum": 0.11477115108363806, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25786750773534567, "standard_deviation": null, "minimum": 0.25786750773534567, "maximum": 0.25786750773534567, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9837597427919645, "standard_deviation": null, "minimum": 0.9837597427919645, "maximum": 0.9837597427919645, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0595955614593693, "standard_deviation": null, "minimum": 1.0595955614593693, "maximum": 1.0595955614593693, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5047780241741828, "standard_deviation": null, "minimum": 0.5047780241741828, "maximum": 0.5047780241741828, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14600664641858846, "standard_deviation": null, "minimum": 0.14600664641858846, "maximum": 0.14600664641858846, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9766740875718397, "standard_deviation": null, "minimum": 0.9766740875718397, "maximum": 0.9766740875718397, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7089222448874513, "standard_deviation": null, "minimum": 1.7089222448874513, "maximum": 1.7089222448874513, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9494147706773077, "standard_deviation": null, "minimum": 0.9494147706773077, "maximum": 0.9494147706773077, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21709973708612498, "standard_deviation": null, "minimum": 0.21709973708612498, "maximum": 0.21709973708612498, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9904984175151293, "standard_deviation": null, "minimum": 0.9904984175151293, "maximum": 0.9904984175151293, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6088895622853514, "standard_deviation": null, "minimum": 1.6088895622853514, "maximum": 1.6088895622853514, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3285153735268604, "standard_deviation": null, "minimum": 1.3285153735268604, "maximum": 1.3285153735268604, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15656784875983468, "standard_deviation": null, "minimum": 0.15656784875983468, "maximum": 0.15656784875983468, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5324164044091749, "standard_deviation": null, "minimum": 0.5324164044091749, "maximum": 0.5324164044091749, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.830085953624703, "standard_deviation": null, "minimum": 5.830085953624703, "maximum": 5.830085953624703, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5308888280545734, "standard_deviation": null, "minimum": 0.5308888280545734, "maximum": 0.5308888280545734, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.6547792997361889, "standard_deviation": null, "minimum": 0.6547792997361889, "maximum": 0.6547792997361889, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9867139342258024, "standard_deviation": null, "minimum": 0.9867139342258024, "maximum": 0.9867139342258024, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.071754788309549, "standard_deviation": null, "minimum": 2.071754788309549, "maximum": 2.071754788309549, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6788409283307445, "standard_deviation": null, "minimum": 0.6788409283307445, "maximum": 0.6788409283307445, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15165677101954703, "standard_deviation": null, "minimum": 0.15165677101954703, "maximum": 0.15165677101954703, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9958380556748785, "standard_deviation": null, "minimum": 0.9958380556748785, "maximum": 0.9958380556748785, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.6991262765129582, "standard_deviation": null, "minimum": 1.6991262765129582, "maximum": 1.6991262765129582, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9386844972291467, "standard_deviation": null, "minimum": 0.9386844972291467, "maximum": 0.9386844972291467, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1054341308824508, "standard_deviation": null, "minimum": 0.1054341308824508, "maximum": 0.1054341308824508, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7021"}, {"case": "librispeech_7127-75946-0002", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.726134777069092, "predicted_mos": 2.726134777069092, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7836ab4b8cffdf378597947c9ed883e0a10ba98b1839f39fd4abb32a67c39b34", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.2010995000600815}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 4.6210881301316915, "全帯域_10ms_rms_envelope_nrmse": 0.20610808518297108, "全帯域_10ms_envelope_correlation": 0.9822859411839131, "全帯域_10ms_modulation_mae_db": 0.7477470467575219, "全帯域_40ms_log_rms_mae_db": 3.997645379581835, "全帯域_40ms_rms_envelope_nrmse": 0.1676608443935688, "全帯域_40ms_envelope_correlation": 0.9936815045605981, "全帯域_40ms_modulation_mae_db": 0.9206605327050669, "300–1000 Hz_10ms_log_rms_mae_db": 2.01984013206352, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.26534088036329095, "300–1000 Hz_10ms_envelope_correlation": 0.9811282191042677, "300–1000 Hz_10ms_modulation_mae_db": 1.14331143263901, "300–1000 Hz_40ms_log_rms_mae_db": 1.6374307017125493, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.23431383150619955, "300–1000 Hz_40ms_envelope_correlation": 0.9902942711807218, "300–1000 Hz_40ms_modulation_mae_db": 1.546166509217061, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6215903858489904, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19596383577392573, "1000–3000 Hz_10ms_envelope_correlation": 0.9811546152292867, "1000–3000 Hz_10ms_modulation_mae_db": 0.5580944728391398, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0125027854792736, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13079746836428754, "1000–3000 Hz_40ms_envelope_correlation": 0.994883463206526, "1000–3000 Hz_40ms_modulation_mae_db": 0.5375374647234561, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9376318186353663, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21831496934531788, "3000–6000 Hz_10ms_envelope_correlation": 0.9722005872696613, "3000–6000 Hz_10ms_modulation_mae_db": 1.220472161662063, "3000–6000 Hz_40ms_log_rms_mae_db": 1.104354221164951, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12929837404542158, "3000–6000 Hz_40ms_envelope_correlation": 0.9908393349931197, "3000–6000 Hz_40ms_modulation_mae_db": 1.5986703243962943, "mr_stft_db": 6.057166478858757, "band_mae_db": 0.8242981535466515, "rms_gain_error_db": 1.1345153996479036, "output_peak": 0.5931671303575784}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.032570978960316, "spectral_convergence": 0.6340833153035208, "spectral_convergence_status": "ok", "amplitude_floor": 1.3425996786840756e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.472021987553306, "spectral_convergence": 0.33497341724180185, "spectral_convergence_status": "ok", "amplitude_floor": 1.1106411258473627e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.666906470062646, "spectral_convergence": 0.36127781980606377, "spectral_convergence_status": "ok", "amplitude_floor": 1.0520318003394391e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00033395677419724207, "output_power": 0.00036324804771543825, "delta_db": 0.36513031448241284, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008500397987721807, "output_power": 0.001144430281435919, "delta_db": 1.2915008083403618, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006268367014596879, "output_power": 0.0009208162889856446, "delta_db": 1.6701857663778075, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003695716856827319, "output_power": 0.00045660795252961106, "delta_db": 0.9184478028490118, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013451399439061329, "output_power": 0.00014246418698742432, "delta_db": 0.24938234755247884, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00011413253897640505, "output_power": 0.0001266262270352548, "delta_db": 0.4511418816778353, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8242981535466515, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1345153996479036, "pitch": null, "output_rms_dbfs": -25.01111738520865, "output_peak": 0.5931671303575784, "reference_rms_dbfs": -26.14563278485655, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.057166478858757}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018099058448312497, "log_rms_mae_db": 4.6210881301316915, "rms_envelope_nrmse": 0.20610808518297108, "envelope_correlation": 0.9822859411839131, "rms_gain_error_db": 1.1345231818196146, "reference_mean_power": 0.002431072908198481, "output_mean_power": 0.00315682350638213}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.757963657265947e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003340894048273399, "output_power": 0.00042695048708507886, "reference_active": true, "delta_db": 1.0651481053466973}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010057920506755965, "output_power": 0.00011872629342934597, "reference_active": true, "delta_db": 0.7203871074397752}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.918423836914529e-05, "output_power": 1.7265293649410704e-05, "reference_active": true, "delta_db": -0.4577059274860933}], "active_band_count": 3, "active_band_mae_db": 0.7477470467575219}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016835992502135956, "log_rms_mae_db": 3.997645379581835, "rms_envelope_nrmse": 0.1676608443935688, "envelope_correlation": 0.9936815045605981, "rms_gain_error_db": 1.1346959176702782, "reference_mean_power": 0.002443196853891242, "output_mean_power": 0.0031726930134294688}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 7.921742652983223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003035779346246488, "output_power": 0.00038970266122313567, "reference_active": true, "delta_db": 1.0846316959361393}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.67325523785903e-05, "output_power": 5.7059982667363324e-05, "reference_active": true, "delta_db": 0.8671213398813384}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.222466914385115e-06, "output_power": 1.0144117188787547e-06, "reference_active": true, "delta_db": -0.8102285622977229}], "active_band_count": 3, "active_band_mae_db": 0.9206605327050669}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015437420870825604, "log_rms_mae_db": 2.01984013206352, "rms_envelope_nrmse": 0.26534088036329095, "envelope_correlation": 0.9811282191042677, "rms_gain_error_db": 1.5892829539235478, "reference_mean_power": 0.0008784594297426283, "output_mean_power": 0.0012666306837826624}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.115238295382225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001974652277483934, "output_power": 0.0002821702968853363, "reference_active": true, "delta_db": 1.5502066464927655}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.173113277871683e-05, "output_power": 6.90659389542311e-05, "reference_active": true, "delta_db": 1.2551193199402166}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0925495372248406e-05, "output_power": 9.461942494290632e-06, "reference_active": true, "delta_db": -0.6246083314840472}], "active_band_count": 3, "active_band_mae_db": 1.14331143263901}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012941193693393203, "log_rms_mae_db": 1.6374307017125493, "rms_envelope_nrmse": 0.23431383150619955, "envelope_correlation": 0.9902942711807218, "rms_gain_error_db": 1.5892824547133286, "reference_mean_power": 0.000882875660341548, "output_mean_power": 0.0012729981999715628}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.7308162041776265e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018289387957124914, "output_power": 0.00026007711035322823, "reference_active": true, "delta_db": 1.5290295885812415}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5176917167074748e-05, "output_power": 3.352453898102618e-05, "reference_active": true, "delta_db": 1.2436026322481124}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.42204663895331e-07, "output_power": 6.131384038686019e-07, "reference_active": true, "delta_db": -1.8658673068218286}], "active_band_count": 3, "active_band_mae_db": 1.546166509217061}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.92340600003817e-05, "log_rms_mae_db": 1.6215903858489904, "rms_envelope_nrmse": 0.19596383577392573, "envelope_correlation": 0.9811546152292867, "rms_gain_error_db": 0.8268981885928844, "reference_mean_power": 0.000424662888055856, "output_mean_power": 0.0005137290538295247}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.0384779490480225e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014012639512762435, "output_power": 0.00015898968103636855, "reference_active": true, "delta_db": 0.5484898851538988}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6627820389923143e-05, "output_power": 4.0288091084533536e-05, "reference_active": true, "delta_db": 0.41365614403909595}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.517673552823662e-06, "output_power": 7.22948319291949e-06, "reference_active": true, "delta_db": -0.7121373893244244}], "active_band_count": 3, "active_band_mae_db": 0.5580944728391398}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.48214657797314e-05, "log_rms_mae_db": 1.0125027854792736, "rms_envelope_nrmse": 0.13079746836428754, "envelope_correlation": 0.994883463206526, "rms_gain_error_db": 0.826900812381597, "reference_mean_power": 0.0004267980543892402, "output_mean_power": 0.0005163123486643269}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.813635389636035e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013656501467579018, "output_power": 0.00015362942245111938, "reference_active": true, "delta_db": 0.5113494216646899}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7671441991318854e-05, "output_power": 1.977712770545634e-05, "reference_active": true, "delta_db": 0.4889122846247208}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.379326657430796e-07, "output_power": 4.6718925809463475e-07, "reference_active": true, "delta_db": -0.6123506878809576}], "active_band_count": 3, "active_band_mae_db": 0.5375374647234561}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.51121938209953e-05, "log_rms_mae_db": 1.9376318186353663, "rms_envelope_nrmse": 0.21831496934531788, "envelope_correlation": 0.9722005872696613, "rms_gain_error_db": 0.41898682660526937, "reference_mean_power": 0.00015588925116025816, "output_mean_power": 0.00017167808232776074}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2103456000035232e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4374604191221866e-05, "output_power": 6.168113071715458e-05, "reference_active": true, "delta_db": 0.5475621783550658}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5832401077981513e-05, "output_power": 1.8319557608128604e-05, "reference_active": true, "delta_db": 0.6336819864629608}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.903828543673632e-06, "output_power": 2.2053295824376325e-06, "reference_active": true, "delta_db": -2.4801723201681627}], "active_band_count": 3, "active_band_mae_db": 1.220472161662063}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.091623719145545e-05, "log_rms_mae_db": 1.104354221164951, "rms_envelope_nrmse": 0.12929837404542158, "envelope_correlation": 0.9908393349931197, "rms_gain_error_db": 0.41898693352519717, "reference_mean_power": 0.00015667279701186957, "output_mean_power": 0.00017254099179977116}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1374269590951097e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.419500527522151e-05, "output_power": 6.164514758066824e-05, "reference_active": true, "delta_db": 0.5593963373828735}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.661491941713126e-06, "output_power": 8.132448311765658e-06, "reference_active": true, "delta_db": 0.8664980480130213}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7948760162947e-07, "output_power": 1.2863255265459265e-07, "reference_active": true, "delta_db": -3.370116587792988}], "active_band_count": 3, "active_band_mae_db": 1.5986703243962943}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8807831511062547, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "27c5cabc1d1ff7b3f8f75fdae8e9274ba4d750a61a102ea27a30c3512b9e8b6d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9811546152292867, "standard_deviation": null, "minimum": 0.9811546152292867, "maximum": 0.9811546152292867, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6215903858489904, "standard_deviation": null, "minimum": 1.6215903858489904, "maximum": 1.6215903858489904, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5580944728391398, "standard_deviation": null, "minimum": 0.5580944728391398, "maximum": 0.5580944728391398, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19596383577392573, "standard_deviation": null, "minimum": 0.19596383577392573, "maximum": 0.19596383577392573, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.994883463206526, "standard_deviation": null, "minimum": 0.994883463206526, "maximum": 0.994883463206526, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0125027854792736, "standard_deviation": null, "minimum": 1.0125027854792736, "maximum": 1.0125027854792736, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5375374647234561, "standard_deviation": null, "minimum": 0.5375374647234561, "maximum": 0.5375374647234561, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13079746836428754, "standard_deviation": null, "minimum": 0.13079746836428754, "maximum": 0.13079746836428754, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9722005872696613, "standard_deviation": null, "minimum": 0.9722005872696613, "maximum": 0.9722005872696613, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9376318186353663, "standard_deviation": null, "minimum": 1.9376318186353663, "maximum": 1.9376318186353663, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.220472161662063, "standard_deviation": null, "minimum": 1.220472161662063, "maximum": 1.220472161662063, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21831496934531788, "standard_deviation": null, "minimum": 0.21831496934531788, "maximum": 0.21831496934531788, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9908393349931197, "standard_deviation": null, "minimum": 0.9908393349931197, "maximum": 0.9908393349931197, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.104354221164951, "standard_deviation": null, "minimum": 1.104354221164951, "maximum": 1.104354221164951, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5986703243962943, "standard_deviation": null, "minimum": 1.5986703243962943, "maximum": 1.5986703243962943, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12929837404542158, "standard_deviation": null, "minimum": 0.12929837404542158, "maximum": 0.12929837404542158, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9811282191042677, "standard_deviation": null, "minimum": 0.9811282191042677, "maximum": 0.9811282191042677, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.01984013206352, "standard_deviation": null, "minimum": 2.01984013206352, "maximum": 2.01984013206352, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.14331143263901, "standard_deviation": null, "minimum": 1.14331143263901, "maximum": 1.14331143263901, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26534088036329095, "standard_deviation": null, "minimum": 0.26534088036329095, "maximum": 0.26534088036329095, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9902942711807218, "standard_deviation": null, "minimum": 0.9902942711807218, "maximum": 0.9902942711807218, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6374307017125493, "standard_deviation": null, "minimum": 1.6374307017125493, "maximum": 1.6374307017125493, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.546166509217061, "standard_deviation": null, "minimum": 1.546166509217061, "maximum": 1.546166509217061, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23431383150619955, "standard_deviation": null, "minimum": 0.23431383150619955, "maximum": 0.23431383150619955, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8242981535466515, "standard_deviation": null, "minimum": 0.8242981535466515, "maximum": 0.8242981535466515, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.057166478858757, "standard_deviation": null, "minimum": 6.057166478858757, "maximum": 6.057166478858757, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5931671303575784, "standard_deviation": null, "minimum": 0.5931671303575784, "maximum": 0.5931671303575784, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1345153996479036, "standard_deviation": null, "minimum": 1.1345153996479036, "maximum": 1.1345153996479036, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9822859411839131, "standard_deviation": null, "minimum": 0.9822859411839131, "maximum": 0.9822859411839131, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 4.6210881301316915, "standard_deviation": null, "minimum": 4.6210881301316915, "maximum": 4.6210881301316915, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7477470467575219, "standard_deviation": null, "minimum": 0.7477470467575219, "maximum": 0.7477470467575219, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20610808518297108, "standard_deviation": null, "minimum": 0.20610808518297108, "maximum": 0.20610808518297108, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9936815045605981, "standard_deviation": null, "minimum": 0.9936815045605981, "maximum": 0.9936815045605981, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 3.997645379581835, "standard_deviation": null, "minimum": 3.997645379581835, "maximum": 3.997645379581835, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9206605327050669, "standard_deviation": null, "minimum": 0.9206605327050669, "maximum": 0.9206605327050669, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1676608443935688, "standard_deviation": null, "minimum": 0.1676608443935688, "maximum": 0.1676608443935688, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7127"}, {"case": "librispeech_7127-75946-0029", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.6047606468200684, "predicted_mos": 2.6047606468200684, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "af3086f49e9778ff57caae1d389f6c8a7c971d8eac3516edbec9676f5231467c", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.87786820018664}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 3.879589153657085, "全帯域_10ms_rms_envelope_nrmse": 0.20034160700327064, "全帯域_10ms_envelope_correlation": 0.9796625409809712, "全帯域_10ms_modulation_mae_db": 1.1577196180091454, "全帯域_40ms_log_rms_mae_db": 3.376398915114064, "全帯域_40ms_rms_envelope_nrmse": 0.16154638983444322, "全帯域_40ms_envelope_correlation": 0.9924175192475236, "全帯域_40ms_modulation_mae_db": 1.1893427539064527, "300–1000 Hz_10ms_log_rms_mae_db": 2.1426688003433054, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.25914155680702333, "300–1000 Hz_10ms_envelope_correlation": 0.9854309448922784, "300–1000 Hz_10ms_modulation_mae_db": 1.4465440857189957, "300–1000 Hz_40ms_log_rms_mae_db": 1.8319093927174845, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.23560917387062322, "300–1000 Hz_40ms_envelope_correlation": 0.9929434121413091, "300–1000 Hz_40ms_modulation_mae_db": 1.3405495022549125, "1000–3000 Hz_10ms_log_rms_mae_db": 1.3834976609052585, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16258591213582763, "1000–3000 Hz_10ms_envelope_correlation": 0.9819416105169766, "1000–3000 Hz_10ms_modulation_mae_db": 0.19843115538254844, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8154725851310001, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07603438303090712, "1000–3000 Hz_40ms_envelope_correlation": 0.9970401230659381, "1000–3000 Hz_40ms_modulation_mae_db": 0.2459540716877049, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8671853737438524, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23080094961925846, "3000–6000 Hz_10ms_envelope_correlation": 0.9643945523824875, "3000–6000 Hz_10ms_modulation_mae_db": 0.3498615014620308, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1259408072170858, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13177379696664898, "3000–6000 Hz_40ms_envelope_correlation": 0.9879748923442728, "3000–6000 Hz_40ms_modulation_mae_db": 0.5949702904147139, "mr_stft_db": 5.942419981435037, "band_mae_db": 0.7617883727139488, "rms_gain_error_db": 1.0884760753081, "output_peak": 0.5784166230179498}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.946962015392204, "spectral_convergence": 0.6515601352045498, "spectral_convergence_status": "ok", "amplitude_floor": 1.016885378852613e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.231393385814618, "spectral_convergence": 0.32685001125818686, "spectral_convergence_status": "ok", "amplitude_floor": 6.009137305261044e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.648904543098291, "spectral_convergence": 0.3584908572688995, "spectral_convergence_status": "ok", "amplitude_floor": 5.349766097366943e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003378185892142299, "output_power": 0.0003552985410393232, "delta_db": 0.21909880245755275, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0005186258824964237, "output_power": 0.0007362419139085793, "delta_db": 1.5216635116128248, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004467018619566432, "output_power": 0.0006749680269302863, "delta_db": 1.7926543817986182, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002054215054829419, "output_power": 0.00022559121779965247, "delta_db": 0.4067628095187785, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00019049740954433386, "output_power": 0.00019855936620792033, "delta_db": 0.18001303557948584, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.3523016958893975e-05, "output_power": 4.8280611495484245e-05, "delta_db": 0.4505376953164331, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7617883727139488, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0884760753081, "pitch": null, "output_rms_dbfs": -26.499576072398053, "output_peak": 0.5784166230179498, "reference_rms_dbfs": -27.58805214770615, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.942419981435037}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001220285373493014, "log_rms_mae_db": 3.879589153657085, "rms_envelope_nrmse": 0.20034160700327064, "envelope_correlation": 0.9796625409809712, "rms_gain_error_db": 1.0887386107910633, "reference_mean_power": 0.0017439355594579833, "output_mean_power": 0.0022408061835538203}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.14297331662316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004136999026359357, "output_power": 0.000585092109924495, "reference_active": true, "delta_db": 1.5053882292411804}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.498679838909322e-05, "output_power": 0.00011442495847776782, "reference_active": true, "delta_db": 1.29169294637066}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.720897303077822e-05, "output_power": 2.0107715035380554e-05, "reference_active": true, "delta_db": 0.6760776784155959}], "active_band_count": 3, "active_band_mae_db": 1.1577196180091454}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00011225274665232883, "log_rms_mae_db": 3.376398915114064, "rms_envelope_nrmse": 0.16154638983444322, "envelope_correlation": 0.9924175192475236, "rms_gain_error_db": 1.0897953418528707, "reference_mean_power": 0.0017512823148994152, "output_mean_power": 0.002250793727602629}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.129560410778096e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037542654743209525, "output_power": 0.0005338584965653974, "reference_active": true, "delta_db": 1.5290117956494726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.98989275491616e-05, "output_power": 5.5962738675603964e-05, "reference_active": true, "delta_db": 1.469377372590054}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1683754663794263e-06, "output_power": 1.3321291704025768e-06, "reference_active": true, "delta_db": 0.5696390934798314}], "active_band_count": 3, "active_band_mae_db": 1.1893427539064527}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.832439224759049e-05, "log_rms_mae_db": 2.1426688003433054, "rms_envelope_nrmse": 0.25914155680702333, "envelope_correlation": 0.9854309448922784, "rms_gain_error_db": 1.667564430128952, "reference_mean_power": 0.0005656706566344367, "output_mean_power": 0.0008304626290859264}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3997327154946766e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018576385507894965, "output_power": 0.00028862729378403926, "reference_active": true, "delta_db": 1.9137618215362981}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.516179020835253e-05, "output_power": 5.065251420283139e-05, "reference_active": true, "delta_db": 1.585300285989191}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.50699704718427e-06, "output_power": 9.110102429708086e-06, "reference_active": true, "delta_db": 0.8405701496314977}], "active_band_count": 3, "active_band_mae_db": 1.4465440857189957}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.746258281389908e-05, "log_rms_mae_db": 1.8319093927174845, "rms_envelope_nrmse": 0.23560917387062322, "envelope_correlation": 0.9929434121413091, "rms_gain_error_db": 1.6699132781124, "reference_mean_power": 0.0005675319915431874, "output_mean_power": 0.0008336460089943556}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.076532458167812e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017606095149491155, "output_power": 0.0002732788606080975, "reference_active": true, "delta_db": 1.9094299368649692}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7577844172032785e-05, "output_power": 2.5553794548152746e-05, "reference_active": true, "delta_db": 1.6248978863170098}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.080515782562606e-07, "output_power": 5.683814206459898e-07, "reference_active": true, "delta_db": 0.487320683582759}], "active_band_count": 3, "active_band_mae_db": 1.3405495022549125}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.679296131568703e-05, "log_rms_mae_db": 1.3834976609052585, "rms_envelope_nrmse": 0.16258591213582763, "envelope_correlation": 0.9819416105169766, "rms_gain_error_db": 0.37879377232722156, "reference_mean_power": 0.00025821561430829635, "output_mean_power": 0.0002817486720331419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9030353181622983e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.270493557085603e-05, "output_power": 9.780087973401302e-05, "reference_active": true, "delta_db": 0.2323990493984349}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6852701179097796e-05, "output_power": 1.8059800969818623e-05, "reference_active": true, "delta_db": 0.30043439578374764}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.066518069212071e-06, "output_power": 4.125425117911654e-06, "reference_active": true, "delta_db": 0.06246002096546282}], "active_band_count": 3, "active_band_mae_db": 0.19843115538254844}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.121002038281381e-05, "log_rms_mae_db": 0.8154725851310001, "rms_envelope_nrmse": 0.07603438303090712, "envelope_correlation": 0.9970401230659381, "rms_gain_error_db": 0.3786724351625105, "reference_mean_power": 0.0002594441947231775, "output_mean_power": 0.00028308131277775196}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7574651639406604e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.903046567125758e-05, "output_power": 9.35896673696241e-05, "reference_active": true, "delta_db": 0.21689258790749663}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.408774632836393e-06, "output_power": 9.144811520697207e-06, "reference_active": true, "delta_db": 0.3644204594656456}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6645360300144737e-07, "output_power": 2.762336067191359e-07, "reference_active": true, "delta_db": 0.15654916768997243}], "active_band_count": 3, "active_band_mae_db": 0.2459540716877049}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.318455452770038e-05, "log_rms_mae_db": 1.8671853737438524, "rms_envelope_nrmse": 0.23080094961925846, "envelope_correlation": 0.9643945523824875, "rms_gain_error_db": 0.24730349548143232, "reference_mean_power": 0.0001492769552865986, "output_mean_power": 0.00015802402447623255}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.0711725750134131e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.027413944353342e-05, "output_power": 5.7310776445996506e-05, "reference_active": true, "delta_db": -0.21894726192561448}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2776930907508557e-05, "output_power": 1.1179939819306548e-05, "reference_active": true, "delta_db": -0.5798708052494546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.16053005381717e-06, "output_power": 2.039311968985795e-06, "reference_active": true, "delta_db": -0.2507664372110235}], "active_band_count": 3, "active_band_mae_db": 0.3498615014620308}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.04724555086843e-05, "log_rms_mae_db": 1.1259408072170858, "rms_envelope_nrmse": 0.13177379696664898, "envelope_correlation": 0.9879748923442728, "rms_gain_error_db": 0.24751418900964106, "reference_mean_power": 0.00014998968981604926, "output_mean_power": 0.00015878622573804355}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.934560164313186e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.89522015257842e-05, "output_power": 5.5200173708478964e-05, "reference_active": true, "delta_db": -0.28559583730693394}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.6946349724579084e-06, "output_power": 5.41265757426159e-06, "reference_active": true, "delta_db": -0.2205533811657493}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8905823102195032e-07, "output_power": 1.4083786235351564e-07, "reference_active": true, "delta_db": -1.2787616527714585}], "active_band_count": 3, "active_band_mae_db": 0.5949702904147139}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8893358885515297, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "3e218a8caad2ef802f13f63dd591502798b5724a3a1a55061c0775db3318b832", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9819416105169766, "standard_deviation": null, "minimum": 0.9819416105169766, "maximum": 0.9819416105169766, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.3834976609052585, "standard_deviation": null, "minimum": 1.3834976609052585, "maximum": 1.3834976609052585, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.19843115538254844, "standard_deviation": null, "minimum": 0.19843115538254844, "maximum": 0.19843115538254844, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16258591213582763, "standard_deviation": null, "minimum": 0.16258591213582763, "maximum": 0.16258591213582763, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9970401230659381, "standard_deviation": null, "minimum": 0.9970401230659381, "maximum": 0.9970401230659381, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8154725851310001, "standard_deviation": null, "minimum": 0.8154725851310001, "maximum": 0.8154725851310001, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.2459540716877049, "standard_deviation": null, "minimum": 0.2459540716877049, "maximum": 0.2459540716877049, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07603438303090712, "standard_deviation": null, "minimum": 0.07603438303090712, "maximum": 0.07603438303090712, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9643945523824875, "standard_deviation": null, "minimum": 0.9643945523824875, "maximum": 0.9643945523824875, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8671853737438524, "standard_deviation": null, "minimum": 1.8671853737438524, "maximum": 1.8671853737438524, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3498615014620308, "standard_deviation": null, "minimum": 0.3498615014620308, "maximum": 0.3498615014620308, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23080094961925846, "standard_deviation": null, "minimum": 0.23080094961925846, "maximum": 0.23080094961925846, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9879748923442728, "standard_deviation": null, "minimum": 0.9879748923442728, "maximum": 0.9879748923442728, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1259408072170858, "standard_deviation": null, "minimum": 1.1259408072170858, "maximum": 1.1259408072170858, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.5949702904147139, "standard_deviation": null, "minimum": 0.5949702904147139, "maximum": 0.5949702904147139, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13177379696664898, "standard_deviation": null, "minimum": 0.13177379696664898, "maximum": 0.13177379696664898, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9854309448922784, "standard_deviation": null, "minimum": 0.9854309448922784, "maximum": 0.9854309448922784, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1426688003433054, "standard_deviation": null, "minimum": 2.1426688003433054, "maximum": 2.1426688003433054, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4465440857189957, "standard_deviation": null, "minimum": 1.4465440857189957, "maximum": 1.4465440857189957, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25914155680702333, "standard_deviation": null, "minimum": 0.25914155680702333, "maximum": 0.25914155680702333, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9929434121413091, "standard_deviation": null, "minimum": 0.9929434121413091, "maximum": 0.9929434121413091, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.8319093927174845, "standard_deviation": null, "minimum": 1.8319093927174845, "maximum": 1.8319093927174845, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3405495022549125, "standard_deviation": null, "minimum": 1.3405495022549125, "maximum": 1.3405495022549125, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23560917387062322, "standard_deviation": null, "minimum": 0.23560917387062322, "maximum": 0.23560917387062322, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7617883727139488, "standard_deviation": null, "minimum": 0.7617883727139488, "maximum": 0.7617883727139488, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.942419981435037, "standard_deviation": null, "minimum": 5.942419981435037, "maximum": 5.942419981435037, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5784166230179498, "standard_deviation": null, "minimum": 0.5784166230179498, "maximum": 0.5784166230179498, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0884760753081, "standard_deviation": null, "minimum": 1.0884760753081, "maximum": 1.0884760753081, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9796625409809712, "standard_deviation": null, "minimum": 0.9796625409809712, "maximum": 0.9796625409809712, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 3.879589153657085, "standard_deviation": null, "minimum": 3.879589153657085, "maximum": 3.879589153657085, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1577196180091454, "standard_deviation": null, "minimum": 1.1577196180091454, "maximum": 1.1577196180091454, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20034160700327064, "standard_deviation": null, "minimum": 0.20034160700327064, "maximum": 0.20034160700327064, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9924175192475236, "standard_deviation": null, "minimum": 0.9924175192475236, "maximum": 0.9924175192475236, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 3.376398915114064, "standard_deviation": null, "minimum": 3.376398915114064, "maximum": 3.376398915114064, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1893427539064527, "standard_deviation": null, "minimum": 1.1893427539064527, "maximum": 1.1893427539064527, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16154638983444322, "standard_deviation": null, "minimum": 0.16154638983444322, "maximum": 0.16154638983444322, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7127"}, {"case": "librispeech_7127-75947-0011", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.1728482246398926, "predicted_mos": 3.1728482246398926, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 57920, "input_sample_rate": 16000, "input_seconds": 3.62, "input_float32_sha256": "dc4433b24bdc02af49d2e820db31ebf9704c979701a05eb4b04a8434de3279ec", "internal_samples": 57920, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.6072970000095665}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.084677646249098, "全帯域_10ms_rms_envelope_nrmse": 0.2355041157980394, "全帯域_10ms_envelope_correlation": 0.9828163028103004, "全帯域_10ms_modulation_mae_db": 0.9835128011264941, "全帯域_40ms_log_rms_mae_db": 1.9172555935060738, "全帯域_40ms_rms_envelope_nrmse": 0.20456135843958376, "全帯域_40ms_envelope_correlation": 0.9929872166902689, "全帯域_40ms_modulation_mae_db": 1.071717286878175, "300–1000 Hz_10ms_log_rms_mae_db": 1.62728808452163, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2816895104876399, "300–1000 Hz_10ms_envelope_correlation": 0.9845686276927159, "300–1000 Hz_10ms_modulation_mae_db": 1.6615238100067113, "300–1000 Hz_40ms_log_rms_mae_db": 1.2160563516746878, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2581613937465063, "300–1000 Hz_40ms_envelope_correlation": 0.9921747849253814, "300–1000 Hz_40ms_modulation_mae_db": 1.7034468883779563, "1000–3000 Hz_10ms_log_rms_mae_db": 1.493105383284407, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1645680407056072, "1000–3000 Hz_10ms_envelope_correlation": 0.9867859302913825, "1000–3000 Hz_10ms_modulation_mae_db": 0.8032377503773014, "1000–3000 Hz_40ms_log_rms_mae_db": 0.999774954916521, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.1102824736691875, "1000–3000 Hz_40ms_envelope_correlation": 0.9972517846599399, "1000–3000 Hz_40ms_modulation_mae_db": 0.5979816524808069, "3000–6000 Hz_10ms_log_rms_mae_db": 1.9970055792427943, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2219150969129665, "3000–6000 Hz_10ms_envelope_correlation": 0.9714659440390534, "3000–6000 Hz_10ms_modulation_mae_db": 0.5684124422947141, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0926963135125127, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12810895658347093, "3000–6000 Hz_40ms_envelope_correlation": 0.9933390978015396, "3000–6000 Hz_40ms_modulation_mae_db": 0.7885085448819282, "mr_stft_db": 5.924144954677993, "band_mae_db": 0.9329299060721112, "rms_gain_error_db": 1.4421856116631762, "output_peak": 0.5878861636006801}, "full_scale": {"sample_count": 57920, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 57920, "output_samples": 57920, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.950785784964542, "spectral_convergence": 0.6435109021070385, "spectral_convergence_status": "ok", "amplitude_floor": 1.9579592332350965e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.193520192573237, "spectral_convergence": 0.3531833744075315, "spectral_convergence_status": "ok", "amplitude_floor": 1.206925486771671e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6281288864961985, "spectral_convergence": 0.4196121819655526, "spectral_convergence_status": "ok", "amplitude_floor": 6.395923898397892e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00042562388767747, "output_power": 0.0004337340762329805, "delta_db": 0.0819754947415807, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001322032073723835, "output_power": 0.0020644224814055216, "delta_db": 1.935565881936798, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006256947892797981, "output_power": 0.0009504175670529882, "delta_db": 1.8155191676393114, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002997481206139804, "output_power": 0.00037114191351091726, "delta_db": 0.9278353372290228, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003961082609243847, "output_power": 0.00046168375781901745, "delta_db": 0.6653069639720415, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 7.769256628656105e-06, "output_power": 7.468644843209699e-06, "delta_db": -0.1713765909139139, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9329299060721112, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.4421856116631762, "pitch": null, "output_rms_dbfs": -23.676572751181002, "output_peak": 0.5878861636006801, "reference_rms_dbfs": -25.11875836284418, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.924144954677993}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 57920, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.00020370994921487513, "log_rms_mae_db": 2.084677646249098, "rms_envelope_nrmse": 0.2355041157980394, "envelope_correlation": 0.9828163028103004, "rms_gain_error_db": 1.442186038737359, "reference_mean_power": 0.003081231895334411, "output_mean_power": 0.004294800437521164}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 1.3803128421607645e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003568424263155193, "output_power": 0.0005558110593919377, "reference_active": true, "delta_db": 1.9245070036160077}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014048200587458577, "output_power": 0.00016628958151659266, "reference_active": true, "delta_db": 0.7324434068343546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7025084230927585e-05, "output_power": 1.591220883612844e-05, "reference_active": true, "delta_db": -0.2935879929291197}], "active_band_count": 3, "active_band_mae_db": 0.9835128011264941}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.00013932647455664127, "log_rms_mae_db": 1.9172555935060738, "rms_envelope_nrmse": 0.20456135843958376, "envelope_correlation": 0.9929872166902689, "rms_gain_error_db": 1.4421905053569664, "reference_mean_power": 0.0031070091895842884, "output_mean_power": 0.004330734785332041}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 1.2870957120852427e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031478645736461515, "output_power": 0.0004903524618138582, "reference_active": true, "delta_db": 1.9249231999621828}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.928842911180188e-05, "output_power": 8.425761123354809e-05, "reference_active": true, "delta_db": 0.8494842715114259}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4070650648317675e-06, "output_power": 1.2712756875357788e-06, "reference_active": true, "delta_db": -0.4407443891609166}], "active_band_count": 3, "active_band_mae_db": 1.071717286878175}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 0.0001235612100218032, "log_rms_mae_db": 1.62728808452163, "rms_envelope_nrmse": 0.2816895104876399, "envelope_correlation": 0.9845686276927159, "rms_gain_error_db": 1.8129672633409821, "reference_mean_power": 0.0011039001556015875, "output_mean_power": 0.0016758167266904811}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 4.707425731708226e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018855725502559516, "output_power": 0.00029363851826753743, "reference_active": true, "delta_db": 1.9236977670670286}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8855557263471545e-05, "output_power": 9.857491509579419e-05, "reference_active": true, "delta_db": 2.239789355814314}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.265121532076037e-06, "output_power": 5.1527591285095505e-06, "reference_active": true, "delta_db": 0.8210843071387921}], "active_band_count": 3, "active_band_mae_db": 1.6615238100067113}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 0.0001123986305082509, "log_rms_mae_db": 1.2160563516746878, "rms_envelope_nrmse": 0.2581613937465063, "envelope_correlation": 0.9921747849253814, "rms_gain_error_db": 1.8129675159404282, "reference_mean_power": 0.0011131375427710934, "output_mean_power": 0.001689839984454736}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 4.401493945076042e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00017713633289459933, "output_power": 0.00027732482973269766, "reference_active": true, "delta_db": 1.946811054208658}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7898955892530673e-05, "output_power": 5.0085484486556016e-05, "reference_active": true, "delta_db": 2.541239288674226}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.0644420897956173e-07, "output_power": 3.521858539142648e-07, "reference_active": true, "delta_db": -0.6222903222509844}], "active_band_count": 3, "active_band_mae_db": 1.7034468883779563}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 8.851421864498589e-05, "log_rms_mae_db": 1.493105383284407, "rms_envelope_nrmse": 0.1645680407056072, "envelope_correlation": 0.9867859302913825, "rms_gain_error_db": 0.7633653593281476, "reference_mean_power": 0.000506505585091631, "output_mean_power": 0.0006038384650236885}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 2.2793822474038483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010569673324300439, "output_power": 0.00012414943825367035, "reference_active": true, "delta_db": 0.6988319399266649}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5368696588833574e-05, "output_power": 1.5162154284367466e-05, "reference_active": true, "delta_db": -0.05876125229323354}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.905327294990813e-06, "output_power": 2.669595311422772e-06, "reference_active": true, "delta_db": -1.6521200589120055}], "active_band_count": 3, "active_band_mae_db": 0.8032377503773014}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 8.08613786301264e-05, "log_rms_mae_db": 0.999774954916521, "rms_envelope_nrmse": 0.1102824736691875, "envelope_correlation": 0.9972517846599399, "rms_gain_error_db": 0.7633692087830815, "reference_mean_power": 0.0005107405079369301, "output_mean_power": 0.0006088877334749198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 2.1269882857538688e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.701418164823122e-05, "output_power": 0.000114012820282315, "reference_active": true, "delta_db": 0.7011846415736354}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.796784971798785e-06, "output_power": 6.52425577834179e-06, "reference_active": true, "delta_db": -0.17772551595462666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1410692673718103e-07, "output_power": 1.7343120345872349e-07, "reference_active": true, "delta_db": -0.9150347999141586}], "active_band_count": 3, "active_band_mae_db": 0.5979816524808069}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 723, "rms_floor": 5.208111658419103e-05, "log_rms_mae_db": 1.9970055792427943, "rms_envelope_nrmse": 0.2219150969129665, "envelope_correlation": 0.9714659440390534, "rms_gain_error_db": 0.6936505663681676, "reference_mean_power": 0.0001411400109581116, "output_mean_power": 0.00016558279284297714}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 723, "frequency_resolution_hz": 0.276625172890733, "power_floor": 8.582618674096986e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.483177751159981e-05, "output_power": 4.075537224437905e-05, "reference_active": true, "delta_db": 0.6820922653098516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4244491686748372e-05, "output_power": 1.723854965596217e-05, "reference_active": true, "delta_db": 0.828537680998596}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3188700900305924e-06, "output_power": 2.2172551933750504e-06, "reference_active": true, "delta_db": -0.1946073805756946}], "active_band_count": 3, "active_band_mae_db": 0.5684124422947141}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 57920, "uncovered_tail_samples": 0, "metrics": {"frame_count": 717, "rms_floor": 4.6774501347655295e-05, "log_rms_mae_db": 1.0926963135125127, "rms_envelope_nrmse": 0.12810895658347093, "envelope_correlation": 0.9933390978015396, "rms_gain_error_db": 0.6936747902782536, "reference_mean_power": 0.0001423195946220122, "output_mean_power": 0.0001669675894022144}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 717, "frequency_resolution_hz": 0.2789400278940028, "power_floor": 7.830084455082624e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.343144196598961e-05, "output_power": 3.913539127137714e-05, "reference_active": true, "delta_db": 0.6841457120692035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.21519729141624e-06, "output_power": 9.058298816198614e-06, "reference_active": true, "delta_db": 0.9879843225564895}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.774546333615459e-07, "output_power": 1.5126825023411973e-07, "reference_active": true, "delta_db": -0.6933956000200918}], "active_band_count": 3, "active_band_mae_db": 0.7885085448819282}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9158642014842729, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "da5eb61df60d3e219807b6be0055fef5fe2f83c662874b37074ade5a6f8cad77", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9867859302913825, "standard_deviation": null, "minimum": 0.9867859302913825, "maximum": 0.9867859302913825, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.493105383284407, "standard_deviation": null, "minimum": 1.493105383284407, "maximum": 1.493105383284407, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8032377503773014, "standard_deviation": null, "minimum": 0.8032377503773014, "maximum": 0.8032377503773014, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1645680407056072, "standard_deviation": null, "minimum": 0.1645680407056072, "maximum": 0.1645680407056072, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9972517846599399, "standard_deviation": null, "minimum": 0.9972517846599399, "maximum": 0.9972517846599399, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.999774954916521, "standard_deviation": null, "minimum": 0.999774954916521, "maximum": 0.999774954916521, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5979816524808069, "standard_deviation": null, "minimum": 0.5979816524808069, "maximum": 0.5979816524808069, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1102824736691875, "standard_deviation": null, "minimum": 0.1102824736691875, "maximum": 0.1102824736691875, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9714659440390534, "standard_deviation": null, "minimum": 0.9714659440390534, "maximum": 0.9714659440390534, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.9970055792427943, "standard_deviation": null, "minimum": 1.9970055792427943, "maximum": 1.9970055792427943, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.5684124422947141, "standard_deviation": null, "minimum": 0.5684124422947141, "maximum": 0.5684124422947141, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2219150969129665, "standard_deviation": null, "minimum": 0.2219150969129665, "maximum": 0.2219150969129665, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9933390978015396, "standard_deviation": null, "minimum": 0.9933390978015396, "maximum": 0.9933390978015396, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0926963135125127, "standard_deviation": null, "minimum": 1.0926963135125127, "maximum": 1.0926963135125127, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7885085448819282, "standard_deviation": null, "minimum": 0.7885085448819282, "maximum": 0.7885085448819282, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12810895658347093, "standard_deviation": null, "minimum": 0.12810895658347093, "maximum": 0.12810895658347093, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9845686276927159, "standard_deviation": null, "minimum": 0.9845686276927159, "maximum": 0.9845686276927159, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.62728808452163, "standard_deviation": null, "minimum": 1.62728808452163, "maximum": 1.62728808452163, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.6615238100067113, "standard_deviation": null, "minimum": 1.6615238100067113, "maximum": 1.6615238100067113, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2816895104876399, "standard_deviation": null, "minimum": 0.2816895104876399, "maximum": 0.2816895104876399, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9921747849253814, "standard_deviation": null, "minimum": 0.9921747849253814, "maximum": 0.9921747849253814, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2160563516746878, "standard_deviation": null, "minimum": 1.2160563516746878, "maximum": 1.2160563516746878, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7034468883779563, "standard_deviation": null, "minimum": 1.7034468883779563, "maximum": 1.7034468883779563, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2581613937465063, "standard_deviation": null, "minimum": 0.2581613937465063, "maximum": 0.2581613937465063, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9329299060721112, "standard_deviation": null, "minimum": 0.9329299060721112, "maximum": 0.9329299060721112, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.924144954677993, "standard_deviation": null, "minimum": 5.924144954677993, "maximum": 5.924144954677993, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5878861636006801, "standard_deviation": null, "minimum": 0.5878861636006801, "maximum": 0.5878861636006801, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.4421856116631762, "standard_deviation": null, "minimum": 1.4421856116631762, "maximum": 1.4421856116631762, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9828163028103004, "standard_deviation": null, "minimum": 0.9828163028103004, "maximum": 0.9828163028103004, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.084677646249098, "standard_deviation": null, "minimum": 2.084677646249098, "maximum": 2.084677646249098, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.9835128011264941, "standard_deviation": null, "minimum": 0.9835128011264941, "maximum": 0.9835128011264941, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2355041157980394, "standard_deviation": null, "minimum": 0.2355041157980394, "maximum": 0.2355041157980394, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9929872166902689, "standard_deviation": null, "minimum": 0.9929872166902689, "maximum": 0.9929872166902689, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.9172555935060738, "standard_deviation": null, "minimum": 1.9172555935060738, "maximum": 1.9172555935060738, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.071717286878175, "standard_deviation": null, "minimum": 1.071717286878175, "maximum": 1.071717286878175, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.20456135843958376, "standard_deviation": null, "minimum": 0.20456135843958376, "maximum": 0.20456135843958376, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7127"}, {"case": "librispeech_7176-88083-0002", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9395484924316406, "predicted_mos": 2.9395484924316406, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "80a95ff960c51bb73136f8789fe9bfece732794a9860782ba4382dacb08dd429", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.020312400069088}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7036528653932677, "全帯域_10ms_rms_envelope_nrmse": 0.14611571786206864, "全帯域_10ms_envelope_correlation": 0.9810616811690431, "全帯域_10ms_modulation_mae_db": 0.8806991822445469, "全帯域_40ms_log_rms_mae_db": 1.0897116299156187, "全帯域_40ms_rms_envelope_nrmse": 0.09233097746936984, "全帯域_40ms_envelope_correlation": 0.9933190778097922, "全帯域_40ms_modulation_mae_db": 1.1709250439985255, "300–1000 Hz_10ms_log_rms_mae_db": 2.106813006484751, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.1514309550276871, "300–1000 Hz_10ms_envelope_correlation": 0.9861595978398503, "300–1000 Hz_10ms_modulation_mae_db": 0.828284051400343, "300–1000 Hz_40ms_log_rms_mae_db": 1.428941800665007, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11135105694112518, "300–1000 Hz_40ms_envelope_correlation": 0.9938615891526512, "300–1000 Hz_40ms_modulation_mae_db": 0.9911992549624459, "1000–3000 Hz_10ms_log_rms_mae_db": 1.9711400542617372, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16511277535956537, "1000–3000 Hz_10ms_envelope_correlation": 0.9813351549456654, "1000–3000 Hz_10ms_modulation_mae_db": 1.205846164334218, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2563281137094342, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09662353423998711, "1000–3000 Hz_40ms_envelope_correlation": 0.9945484364127548, "1000–3000 Hz_40ms_modulation_mae_db": 1.4170314439899456, "3000–6000 Hz_10ms_log_rms_mae_db": 2.186174043813906, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2467976102319727, "3000–6000 Hz_10ms_envelope_correlation": 0.9552827942982528, "3000–6000 Hz_10ms_modulation_mae_db": 1.4174913847729542, "3000–6000 Hz_40ms_log_rms_mae_db": 1.3796151678969664, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15532604003653158, "3000–6000 Hz_40ms_envelope_correlation": 0.9807387155341741, "3000–6000 Hz_40ms_modulation_mae_db": 2.0601110212490137, "mr_stft_db": 5.984483191457509, "band_mae_db": 0.37366191794822107, "rms_gain_error_db": 0.4256580131434407, "output_peak": 0.76652939881511}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.190991665123206, "spectral_convergence": 0.7502853744954924, "spectral_convergence_status": "ok", "amplitude_floor": 1.9341972091070217e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.29461816610997, "spectral_convergence": 0.258632355422962, "spectral_convergence_status": "ok", "amplitude_floor": 1.0469896576426271e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.467839743139349, "spectral_convergence": 0.2322365122804659, "spectral_convergence_status": "ok", "amplitude_floor": 8.978224937278799e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002340108897553237, "output_power": 0.002377329651024998, "delta_db": 0.06853339211237981, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003418531859998669, "output_power": 0.003989425595575747, "delta_db": 0.6707073792611175, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009669299209010629, "output_power": 0.0010591012597066327, "delta_db": 0.3954248524445748, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00016643839195516676, "output_power": 0.00018746946762307555, "delta_db": 0.5167703487249877, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.655994855693747e-05, "output_power": 8.042011259640526e-05, "delta_db": 0.21363043871511567, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.076442519269658e-05, "output_power": 3.7375827197987205e-05, "delta_db": -0.37690509643115105, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.37366191794822107, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.4256580131434407, "pitch": null, "output_rms_dbfs": -21.11757478192309, "output_peak": 0.76652939881511, "reference_rms_dbfs": -21.54323279506653, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.984483191457509}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019906122792766575, "log_rms_mae_db": 1.7036528653932677, "rms_envelope_nrmse": 0.14611571786206864, "envelope_correlation": 0.9810616811690431, "rms_gain_error_db": 0.427129007905684, "reference_mean_power": 0.007012794178240996, "output_mean_power": 0.007737559355738389}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.192818590388021e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015356197181510699, "output_power": 0.0017074103224306101, "reference_active": true, "delta_db": 0.4605422264564401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003622112164404606, "output_power": 0.0003483491710400303, "reference_active": true, "delta_db": -0.16947113409103393}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.688762087391039e-05, "output_power": 2.9501885839478157e-05, "reference_active": true, "delta_db": -2.0120841861861667}], "active_band_count": 3, "active_band_mae_db": 0.8806991822445469}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018229956701751592, "log_rms_mae_db": 1.0897116299156187, "rms_envelope_nrmse": 0.09233097746936984, "envelope_correlation": 0.9933190778097922, "rms_gain_error_db": 0.43242249410455047, "reference_mean_power": 0.0070179714231847195, "output_mean_power": 0.007752715461317587}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.8443307374696506e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0014276197130393435, "output_power": 0.0015812417709962928, "reference_active": true, "delta_db": 0.44385742191346694}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018503527061359352, "output_power": 0.00017734839368401594, "reference_active": true, "delta_db": -0.1842726036227577}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.327921114523828e-06, "output_power": 2.227485192008973e-06, "reference_active": true, "delta_db": -2.8846451064593515}], "active_band_count": 3, "active_band_mae_db": 1.1709250439985255}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00015227518148493586, "log_rms_mae_db": 2.106813006484751, "rms_envelope_nrmse": 0.1514309550276871, "envelope_correlation": 0.9861595978398503, "rms_gain_error_db": 0.6016847716536038, "reference_mean_power": 0.0024461505543329222, "output_mean_power": 0.002809646360864923}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3707367109300066e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00065188242251555, "output_power": 0.0007404417961586716, "reference_active": true, "delta_db": 0.5532165485467234}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.000159373128948768, "output_power": 0.00016177645493697056, "reference_active": true, "delta_db": 0.06500215178778293}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8816561344170712e-05, "output_power": 1.2242691762189227e-05, "reference_active": true, "delta_db": -1.8666334538665228}], "active_band_count": 3, "active_band_mae_db": 0.828284051400343}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013579343349222632, "log_rms_mae_db": 1.428941800665007, "rms_envelope_nrmse": 0.11135105694112518, "envelope_correlation": 0.9938615891526512, "rms_gain_error_db": 0.6036397380115467, "reference_mean_power": 0.0024424650949128645, "output_mean_power": 0.0028066763798650658}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.250982515304928e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006212562593255464, "output_power": 0.0006970508759005207, "reference_active": true, "delta_db": 0.49993699862567464}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.001783099426803e-05, "output_power": 8.02608710860589e-05, "reference_active": true, "delta_db": 0.013170939154800651}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.744470580402989e-06, "output_power": 9.899532071916287e-07, "reference_active": true, "delta_db": -2.4604898271068625}], "active_band_count": 3, "active_band_mae_db": 0.9911992549624459}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 7.560305347439803e-05, "log_rms_mae_db": 1.9711400542617372, "rms_envelope_nrmse": 0.16511277535956537, "envelope_correlation": 0.9813351549456654, "rms_gain_error_db": 0.45557372921564643, "reference_mean_power": 0.0001982579107701744, "output_mean_power": 0.00022018508552182003}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.44612649270932e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.480334346914858e-05, "output_power": 3.940081643225581e-05, "reference_active": true, "delta_db": 0.5388425349948487}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0911626587448635e-05, "output_power": 1.0399101125931728e-05, "reference_active": true, "delta_db": -0.20893693827151338}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.7887330599139967e-06, "output_power": 1.440227314434897e-06, "reference_active": true, "delta_db": -2.869759019736292}], "active_band_count": 3, "active_band_mae_db": 1.205846164334218}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.513956828007693e-05, "log_rms_mae_db": 1.2563281137094342, "rms_envelope_nrmse": 0.09662353423998711, "envelope_correlation": 0.9945484364127548, "rms_gain_error_db": 0.4570744122812084, "reference_mean_power": 0.00019869265760544437, "output_mean_power": 0.00022074417890805844}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.64113687643322e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.340875190885911e-05, "output_power": 3.676668356077343e-05, "reference_active": true, "delta_db": 0.4159420579961173}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.757829899772919e-06, "output_power": 5.172192920739031e-06, "reference_active": true, "delta_db": -0.46584115099812373}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2467549734271334e-07, "output_power": 1.0342477849095805e-07, "reference_active": true, "delta_db": -3.3693111229755957}], "active_band_count": 3, "active_band_mae_db": 1.4170314439899456}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.7519381942341847e-05, "log_rms_mae_db": 2.186174043813906, "rms_envelope_nrmse": 0.2467976102319727, "envelope_correlation": 0.9552827942982528, "rms_gain_error_db": -0.153463331765433, "reference_mean_power": 3.309162956541551e-05, "output_mean_power": 3.19427145806875e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.646064632792655e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.7250471980521725e-06, "output_power": 5.82714271180619e-06, "reference_active": true, "delta_db": -0.6223968220076468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3903630336351795e-06, "output_power": 2.1166134780126286e-06, "reference_active": true, "delta_db": -0.5282230666480575}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2031554280127943e-06, "output_power": 5.890284994809656e-07, "reference_active": true, "delta_db": -3.101854265663158}], "active_band_count": 3, "active_band_mae_db": 1.4174913847729542}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2421508904297867e-05, "log_rms_mae_db": 1.3796151678969664, "rms_envelope_nrmse": 0.15532604003653158, "envelope_correlation": 0.9807387155341741, "rms_gain_error_db": -0.15372093937703535, "reference_mean_power": 3.3247421347363265e-05, "output_mean_power": 3.209119380745715e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.522637378578784e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 6.9685546500066154e-06, "output_power": 5.871910953467318e-06, "reference_active": true, "delta_db": -0.7436324933460277}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.482924821402799e-06, "output_power": 1.1415438441588992e-06, "reference_active": true, "delta_db": -1.1362653809172882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.259963132949643e-08, "output_power": 3.440057792047528e-08, "reference_active": true, "delta_db": -4.300435189483726}], "active_band_count": 3, "active_band_mae_db": 2.0601110212490137}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9234571210634901, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "73e217734490cf9250ff2c5bf66fcd6f8e96735d8697ad912bfbfdedca22c88a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9813351549456654, "standard_deviation": null, "minimum": 0.9813351549456654, "maximum": 0.9813351549456654, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.9711400542617372, "standard_deviation": null, "minimum": 1.9711400542617372, "maximum": 1.9711400542617372, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.205846164334218, "standard_deviation": null, "minimum": 1.205846164334218, "maximum": 1.205846164334218, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16511277535956537, "standard_deviation": null, "minimum": 0.16511277535956537, "maximum": 0.16511277535956537, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9945484364127548, "standard_deviation": null, "minimum": 0.9945484364127548, "maximum": 0.9945484364127548, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2563281137094342, "standard_deviation": null, "minimum": 1.2563281137094342, "maximum": 1.2563281137094342, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.4170314439899456, "standard_deviation": null, "minimum": 1.4170314439899456, "maximum": 1.4170314439899456, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09662353423998711, "standard_deviation": null, "minimum": 0.09662353423998711, "maximum": 0.09662353423998711, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9552827942982528, "standard_deviation": null, "minimum": 0.9552827942982528, "maximum": 0.9552827942982528, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.186174043813906, "standard_deviation": null, "minimum": 2.186174043813906, "maximum": 2.186174043813906, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.4174913847729542, "standard_deviation": null, "minimum": 1.4174913847729542, "maximum": 1.4174913847729542, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2467976102319727, "standard_deviation": null, "minimum": 0.2467976102319727, "maximum": 0.2467976102319727, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9807387155341741, "standard_deviation": null, "minimum": 0.9807387155341741, "maximum": 0.9807387155341741, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.3796151678969664, "standard_deviation": null, "minimum": 1.3796151678969664, "maximum": 1.3796151678969664, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.0601110212490137, "standard_deviation": null, "minimum": 2.0601110212490137, "maximum": 2.0601110212490137, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15532604003653158, "standard_deviation": null, "minimum": 0.15532604003653158, "maximum": 0.15532604003653158, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9861595978398503, "standard_deviation": null, "minimum": 0.9861595978398503, "maximum": 0.9861595978398503, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.106813006484751, "standard_deviation": null, "minimum": 2.106813006484751, "maximum": 2.106813006484751, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.828284051400343, "standard_deviation": null, "minimum": 0.828284051400343, "maximum": 0.828284051400343, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1514309550276871, "standard_deviation": null, "minimum": 0.1514309550276871, "maximum": 0.1514309550276871, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9938615891526512, "standard_deviation": null, "minimum": 0.9938615891526512, "maximum": 0.9938615891526512, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.428941800665007, "standard_deviation": null, "minimum": 1.428941800665007, "maximum": 1.428941800665007, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9911992549624459, "standard_deviation": null, "minimum": 0.9911992549624459, "maximum": 0.9911992549624459, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11135105694112518, "standard_deviation": null, "minimum": 0.11135105694112518, "maximum": 0.11135105694112518, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.37366191794822107, "standard_deviation": null, "minimum": 0.37366191794822107, "maximum": 0.37366191794822107, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.984483191457509, "standard_deviation": null, "minimum": 5.984483191457509, "maximum": 5.984483191457509, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.76652939881511, "standard_deviation": null, "minimum": 0.76652939881511, "maximum": 0.76652939881511, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.4256580131434407, "standard_deviation": null, "minimum": 0.4256580131434407, "maximum": 0.4256580131434407, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9810616811690431, "standard_deviation": null, "minimum": 0.9810616811690431, "maximum": 0.9810616811690431, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7036528653932677, "standard_deviation": null, "minimum": 1.7036528653932677, "maximum": 1.7036528653932677, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8806991822445469, "standard_deviation": null, "minimum": 0.8806991822445469, "maximum": 0.8806991822445469, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.14611571786206864, "standard_deviation": null, "minimum": 0.14611571786206864, "maximum": 0.14611571786206864, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9933190778097922, "standard_deviation": null, "minimum": 0.9933190778097922, "maximum": 0.9933190778097922, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0897116299156187, "standard_deviation": null, "minimum": 1.0897116299156187, "maximum": 1.0897116299156187, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1709250439985255, "standard_deviation": null, "minimum": 1.1709250439985255, "maximum": 1.1709250439985255, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.09233097746936984, "standard_deviation": null, "minimum": 0.09233097746936984, "maximum": 0.09233097746936984, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7176"}, {"case": "librispeech_7176-88083-0004", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.5959885120391846, "predicted_mos": 3.5959885120391846, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "050bd611d4e172e08397511c70504a8b01512bd10d4cefae99eaf18a1d7fdd6f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.9310487001203}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.8388092499690918, "全帯域_10ms_rms_envelope_nrmse": 0.15163748143601802, "全帯域_10ms_envelope_correlation": 0.9781838643809772, "全帯域_10ms_modulation_mae_db": 0.20546884055033596, "全帯域_40ms_log_rms_mae_db": 1.0763114096848156, "全帯域_40ms_rms_envelope_nrmse": 0.08757434822070266, "全帯域_40ms_envelope_correlation": 0.9938537550880164, "全帯域_40ms_modulation_mae_db": 0.09758645749951263, "300–1000 Hz_10ms_log_rms_mae_db": 2.138691814503766, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.17568617000091696, "300–1000 Hz_10ms_envelope_correlation": 0.9810281568196958, "300–1000 Hz_10ms_modulation_mae_db": 0.47310329615842156, "300–1000 Hz_40ms_log_rms_mae_db": 1.4180234688403714, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1263351361531652, "300–1000 Hz_40ms_envelope_correlation": 0.9922696946897344, "300–1000 Hz_40ms_modulation_mae_db": 0.6061772458484286, "1000–3000 Hz_10ms_log_rms_mae_db": 2.0241434551048605, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17780436424500057, "1000–3000 Hz_10ms_envelope_correlation": 0.9771552118878191, "1000–3000 Hz_10ms_modulation_mae_db": 0.3379016106121377, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2666680654733145, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09978448597392826, "1000–3000 Hz_40ms_envelope_correlation": 0.9934258053266132, "1000–3000 Hz_40ms_modulation_mae_db": 0.4415482356551685, "3000–6000 Hz_10ms_log_rms_mae_db": 2.346315060367808, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.30293599591688647, "3000–6000 Hz_10ms_envelope_correlation": 0.9416753415126962, "3000–6000 Hz_10ms_modulation_mae_db": 0.4679821017112336, "3000–6000 Hz_40ms_log_rms_mae_db": 1.4856102180196087, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18734449255816862, "3000–6000 Hz_40ms_envelope_correlation": 0.9767708792341104, "3000–6000 Hz_40ms_modulation_mae_db": 0.4320209320332844, "mr_stft_db": 6.123788612640484, "band_mae_db": 0.36644090663254686, "rms_gain_error_db": 0.4273083503309456, "output_peak": 0.7748992106208392}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.335858127912934, "spectral_convergence": 0.7385788047006525, "spectral_convergence_status": "ok", "amplitude_floor": 2.016382236479878e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.490433581512877, "spectral_convergence": 0.2628268617446415, "spectral_convergence_status": "ok", "amplitude_floor": 1.164576113083389e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.545074128495644, "spectral_convergence": 0.24728031465647848, "spectral_convergence_status": "ok", "amplitude_floor": 9.482830334434223e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0029188205763026483, "output_power": 0.0029887237439952504, "delta_db": 0.10278374692052354, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003712420548895462, "output_power": 0.004290262756559795, "delta_db": 0.62826723403582, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0009931195946472759, "output_power": 0.0011398159653602982, "delta_db": 0.5983318504797603, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003469926396566426, "output_power": 0.0003868403330219288, "delta_db": 0.4721148577705073, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013263724937728663, "output_power": 0.00014087141709786963, "delta_db": 0.2615737648637453, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.438064081917431e-05, "output_power": 6.64221140111989e-05, "delta_db": 0.13557398572492457, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.36644090663254686, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.4273083503309456, "pitch": null, "output_rms_dbfs": -20.451336969028112, "output_peak": 0.7748992106208392, "reference_rms_dbfs": -20.87864531935906, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.123788612640484}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023582673937264833, "log_rms_mae_db": 1.8388092499690918, "rms_envelope_nrmse": 0.15163748143601802, "envelope_correlation": 0.9781838643809772, "rms_gain_error_db": 0.42731015582766985, "reference_mean_power": 0.008175179133340451, "output_mean_power": 0.009020451861843429}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.4118743138857815e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017924334098852407, "output_power": 0.0018808700414070954, "reference_active": true, "delta_db": 0.20915758732613116}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00033647515979494424, "output_power": 0.00032099752337116114, "reference_active": true, "delta_db": -0.20451326392301694}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.450579873055231e-05, "output_power": 5.7110541137332894e-05, "reference_active": true, "delta_db": 0.20273567040185975}], "active_band_count": 3, "active_band_mae_db": 0.20546884055033596}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002175646104262597, "log_rms_mae_db": 1.0763114096848156, "rms_envelope_nrmse": 0.08757434822070266, "envelope_correlation": 0.9938537550880164, "rms_gain_error_db": 0.42727399165607, "reference_mean_power": 0.008216199965234223, "output_mean_power": 0.009065638552194546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.000498094679175e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016116003385924513, "output_power": 0.001692638495524684, "reference_active": true, "delta_db": 0.21306863811239468}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017199808112094875, "output_power": 0.00017157922929519646, "reference_active": true, "delta_db": -0.01058889055006074}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.5420165509687798e-06, "output_power": 3.486104503045195e-06, "reference_active": true, "delta_db": -0.06910184383608246}], "active_band_count": 3, "active_band_mae_db": 0.09758645749951263}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019245421130728438, "log_rms_mae_db": 2.138691814503766, "rms_envelope_nrmse": 0.17568617000091696, "envelope_correlation": 0.9810281568196958, "rms_gain_error_db": 0.7189174831330951, "reference_mean_power": 0.0024199693169718968, "output_mean_power": 0.002855627843607351}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3235427499409267e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006913978889533925, "output_power": 0.000784872125769568, "reference_active": true, "delta_db": 0.550708564908693}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018527392767596858, "output_power": 0.0001994599196051293, "reference_active": true, "delta_db": 0.32041331309552523}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5604035138941515e-05, "output_power": 2.2567820221155524e-05, "reference_active": true, "delta_db": -0.5481880104710463}], "active_band_count": 3, "active_band_mae_db": 0.47310329615842156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017723694424605393, "log_rms_mae_db": 1.4180234688403714, "rms_envelope_nrmse": 0.1263351361531652, "envelope_correlation": 0.9922696946897344, "rms_gain_error_db": 0.7189153956662603, "reference_mean_power": 0.0024321306730885443, "output_mean_power": 0.0028699771862259163}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1677626980235007e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006540469587465989, "output_power": 0.0007364585975138135, "reference_active": true, "delta_db": 0.5153940597069251}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.291953991407944e-05, "output_power": 0.00010371124961358806, "reference_active": true, "delta_db": 0.47718816345550374}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.049791938534597e-06, "output_power": 1.694785849992738e-06, "reference_active": true, "delta_db": -0.8259495143828571}], "active_band_count": 3, "active_band_mae_db": 0.6061772458484286}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.001235221787968e-05, "log_rms_mae_db": 2.0241434551048605, "rms_envelope_nrmse": 0.17780436424500057, "envelope_correlation": 0.9771552118878191, "rms_gain_error_db": 0.4153622953290456, "reference_mean_power": 0.0003849600012580844, "output_mean_power": 0.0004235959865228766}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.5531826233320727e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012212996791699553, "output_power": 0.00013057626623169083, "reference_active": true, "delta_db": 0.29042002983088894}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.773842605512037e-05, "output_power": 2.982618946830641e-05, "reference_active": true, "delta_db": 0.31515957824819574}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.9827801273791894e-06, "output_power": 6.356467705806499e-06, "reference_active": true, "delta_db": -0.4081252237573283}], "active_band_count": 3, "active_band_mae_db": 0.3379016106121377}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.393121399816863e-05, "log_rms_mae_db": 1.2666680654733145, "rms_envelope_nrmse": 0.09978448597392826, "envelope_correlation": 0.9934258053266132, "rms_gain_error_db": 0.4153627788609523, "reference_mean_power": 0.0003868956457514561, "output_mean_power": 0.00042572594673391143}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2949024312838256e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011569002203634676, "output_power": 0.0001243365035076768, "reference_active": true, "delta_db": 0.31302746522774155}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4800082503354946e-05, "output_power": 1.637747311979485e-05, "reference_active": true, "delta_db": 0.4398275891582889}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.803144704877755e-07, "output_power": 5.08726586496205e-07, "reference_active": true, "delta_db": -0.571789652579475}], "active_band_count": 3, "active_band_mae_db": 0.4415482356551685}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.314726805023413e-05, "log_rms_mae_db": 2.346315060367808, "rms_envelope_nrmse": 0.30293599591688647, "envelope_correlation": 0.9416753415126962, "rms_gain_error_db": 0.4036287371666594, "reference_mean_power": 8.262304385157997e-05, "output_mean_power": 9.067009362093294e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.2225580529948354e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.126031645648209e-05, "output_power": 2.2106305659598253e-05, "reference_active": true, "delta_db": 0.16946445992803777}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.562486047803945e-06, "output_power": 1.1652504241246865e-05, "reference_active": true, "delta_db": 0.8584845517551282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.527384735880018e-06, "output_power": 3.2348422138142444e-06, "reference_active": true, "delta_db": -0.37599729345053484}], "active_band_count": 3, "active_band_mae_db": 0.4679821017112336}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6400870566820594e-05, "log_rms_mae_db": 1.4856102180196087, "rms_envelope_nrmse": 0.18734449255816862, "envelope_correlation": 0.9767708792341104, "rms_gain_error_db": 0.4036283035944688, "reference_mean_power": 8.30385039555115e-05, "output_mean_power": 9.112600825569561e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.767621782464082e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.197465758003415e-05, "output_power": 2.3560393033702814e-05, "reference_active": true, "delta_db": 0.3026041465432131}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.865612603723312e-06, "output_power": 6.438938919700059e-06, "reference_active": true, "delta_db": 0.4050092888625334}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.828054895066371e-07, "output_power": 2.469692592760996e-07, "reference_active": true, "delta_db": -0.5884493606941068}], "active_band_count": 3, "active_band_mae_db": 0.4320209320332844}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9226557112486001, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "97186ed66271ee87678e098451807fdb1a0a8f1acf1937c8491a037a1a6f2d0d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9771552118878191, "standard_deviation": null, "minimum": 0.9771552118878191, "maximum": 0.9771552118878191, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.0241434551048605, "standard_deviation": null, "minimum": 2.0241434551048605, "maximum": 2.0241434551048605, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.3379016106121377, "standard_deviation": null, "minimum": 0.3379016106121377, "maximum": 0.3379016106121377, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17780436424500057, "standard_deviation": null, "minimum": 0.17780436424500057, "maximum": 0.17780436424500057, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9934258053266132, "standard_deviation": null, "minimum": 0.9934258053266132, "maximum": 0.9934258053266132, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2666680654733145, "standard_deviation": null, "minimum": 1.2666680654733145, "maximum": 1.2666680654733145, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.4415482356551685, "standard_deviation": null, "minimum": 0.4415482356551685, "maximum": 0.4415482356551685, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09978448597392826, "standard_deviation": null, "minimum": 0.09978448597392826, "maximum": 0.09978448597392826, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9416753415126962, "standard_deviation": null, "minimum": 0.9416753415126962, "maximum": 0.9416753415126962, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.346315060367808, "standard_deviation": null, "minimum": 2.346315060367808, "maximum": 2.346315060367808, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.4679821017112336, "standard_deviation": null, "minimum": 0.4679821017112336, "maximum": 0.4679821017112336, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30293599591688647, "standard_deviation": null, "minimum": 0.30293599591688647, "maximum": 0.30293599591688647, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9767708792341104, "standard_deviation": null, "minimum": 0.9767708792341104, "maximum": 0.9767708792341104, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.4856102180196087, "standard_deviation": null, "minimum": 1.4856102180196087, "maximum": 1.4856102180196087, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.4320209320332844, "standard_deviation": null, "minimum": 0.4320209320332844, "maximum": 0.4320209320332844, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18734449255816862, "standard_deviation": null, "minimum": 0.18734449255816862, "maximum": 0.18734449255816862, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9810281568196958, "standard_deviation": null, "minimum": 0.9810281568196958, "maximum": 0.9810281568196958, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.138691814503766, "standard_deviation": null, "minimum": 2.138691814503766, "maximum": 2.138691814503766, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.47310329615842156, "standard_deviation": null, "minimum": 0.47310329615842156, "maximum": 0.47310329615842156, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17568617000091696, "standard_deviation": null, "minimum": 0.17568617000091696, "maximum": 0.17568617000091696, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9922696946897344, "standard_deviation": null, "minimum": 0.9922696946897344, "maximum": 0.9922696946897344, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4180234688403714, "standard_deviation": null, "minimum": 1.4180234688403714, "maximum": 1.4180234688403714, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6061772458484286, "standard_deviation": null, "minimum": 0.6061772458484286, "maximum": 0.6061772458484286, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1263351361531652, "standard_deviation": null, "minimum": 0.1263351361531652, "maximum": 0.1263351361531652, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.36644090663254686, "standard_deviation": null, "minimum": 0.36644090663254686, "maximum": 0.36644090663254686, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.123788612640484, "standard_deviation": null, "minimum": 6.123788612640484, "maximum": 6.123788612640484, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7748992106208392, "standard_deviation": null, "minimum": 0.7748992106208392, "maximum": 0.7748992106208392, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.4273083503309456, "standard_deviation": null, "minimum": 0.4273083503309456, "maximum": 0.4273083503309456, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9781838643809772, "standard_deviation": null, "minimum": 0.9781838643809772, "maximum": 0.9781838643809772, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.8388092499690918, "standard_deviation": null, "minimum": 1.8388092499690918, "maximum": 1.8388092499690918, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.20546884055033596, "standard_deviation": null, "minimum": 0.20546884055033596, "maximum": 0.20546884055033596, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15163748143601802, "standard_deviation": null, "minimum": 0.15163748143601802, "maximum": 0.15163748143601802, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9938537550880164, "standard_deviation": null, "minimum": 0.9938537550880164, "maximum": 0.9938537550880164, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0763114096848156, "standard_deviation": null, "minimum": 1.0763114096848156, "maximum": 1.0763114096848156, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.09758645749951263, "standard_deviation": null, "minimum": 0.09758645749951263, "maximum": 0.09758645749951263, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08757434822070266, "standard_deviation": null, "minimum": 0.08757434822070266, "maximum": 0.08757434822070266, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7176"}, {"case": "librispeech_7176-88083-0026", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.072894334793091, "predicted_mos": 3.072894334793091, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 88480, "input_sample_rate": 16000, "input_seconds": 5.53, "input_float32_sha256": "7fed0bed13623c00b50d776013f62b10b3d0a279a1e996065083c68fb85f27b9", "internal_samples": 88480, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.712158599868417}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9276043214367375, "全帯域_10ms_rms_envelope_nrmse": 0.1495118187643159, "全帯域_10ms_envelope_correlation": 0.9803575600338925, "全帯域_10ms_modulation_mae_db": 0.8064827347531779, "全帯域_40ms_log_rms_mae_db": 1.2248420811940868, "全帯域_40ms_rms_envelope_nrmse": 0.08846566536398857, "全帯域_40ms_envelope_correlation": 0.9937224322412479, "全帯域_40ms_modulation_mae_db": 1.1059609797948633, "300–1000 Hz_10ms_log_rms_mae_db": 2.487926406483803, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18531901048517468, "300–1000 Hz_10ms_envelope_correlation": 0.9782573937020033, "300–1000 Hz_10ms_modulation_mae_db": 1.1073436497539784, "300–1000 Hz_40ms_log_rms_mae_db": 1.7386900252548312, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.12081625249071663, "300–1000 Hz_40ms_envelope_correlation": 0.9918425915585528, "300–1000 Hz_40ms_modulation_mae_db": 1.2674401055832947, "1000–3000 Hz_10ms_log_rms_mae_db": 2.187210650325183, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.19671424512819233, "1000–3000 Hz_10ms_envelope_correlation": 0.9728884785113232, "1000–3000 Hz_10ms_modulation_mae_db": 0.8984666153018362, "1000–3000 Hz_40ms_log_rms_mae_db": 1.3689331094616566, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10621576143291325, "1000–3000 Hz_40ms_envelope_correlation": 0.9918860429458589, "1000–3000 Hz_40ms_modulation_mae_db": 1.1541840084645925, "3000–6000 Hz_10ms_log_rms_mae_db": 2.7068681174312603, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2825993624534353, "3000–6000 Hz_10ms_envelope_correlation": 0.9493890515357651, "3000–6000 Hz_10ms_modulation_mae_db": 0.7984020443577137, "3000–6000 Hz_40ms_log_rms_mae_db": 1.8260586788034725, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17674554733192818, "3000–6000 Hz_40ms_envelope_correlation": 0.9801415997992353, "3000–6000 Hz_40ms_modulation_mae_db": 1.1257743983489967, "mr_stft_db": 6.314328008633088, "band_mae_db": 0.37408928468164593, "rms_gain_error_db": 0.3925743113461682, "output_peak": 0.7719099371952937}, "full_scale": {"sample_count": 88480, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 88480, "output_samples": 88480, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.7438633393658876, "spectral_convergence": 0.8182027336320425, "spectral_convergence_status": "ok", "amplitude_floor": 2.1848847242256272e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.622572773009961, "spectral_convergence": 0.29751304514920657, "spectral_convergence_status": "ok", "amplitude_floor": 1.2036494394056804e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.576547913523418, "spectral_convergence": 0.2721837120096839, "spectral_convergence_status": "ok", "amplitude_floor": 7.153950066191215e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.002200971739696214, "output_power": 0.002347697800486997, "delta_db": 0.28027726792727126, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002201298666996261, "output_power": 0.0024538185811573588, "delta_db": 0.47163480165667265, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0011542423166681864, "output_power": 0.0012857878866515365, "delta_db": 0.46872337669186104, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00019781669780676847, "output_power": 0.00021081586180944975, "delta_db": 0.27640336312442626, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.802033486419172e-05, "output_power": 6.455705023090295e-05, "delta_db": 0.46363447189919527, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.6630669693501258e-05, "output_power": 1.7753990398484786e-05, "delta_db": 0.2838624267904495, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.37408928468164593, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.3925743113461682, "pitch": null, "output_rms_dbfs": -21.951499719459797, "output_peak": 0.7719099371952937, "reference_rms_dbfs": -22.34407403080596, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.314328008633088}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 88480, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00024519594682554, "log_rms_mae_db": 1.9276043214367375, "rms_envelope_nrmse": 0.1495118187643159, "envelope_correlation": 0.9803575600338925, "rms_gain_error_db": 0.39258407266213624, "reference_mean_power": 0.00583423429533283, "output_mean_power": 0.006386196436488821}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.472320009859619e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0016537460501756134, "output_power": 0.0017591755254516969, "reference_active": true, "delta_db": 0.26840354344835166}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0003778384775464376, "output_power": 0.00035725498072882995, "reference_active": true, "delta_db": -0.24327890343228134}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.8517105494706513e-05, "output_power": 3.126931067667591e-05, "reference_active": true, "delta_db": -1.9077657573789009}], "active_band_count": 3, "active_band_mae_db": 0.8064827347531779}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00020002070942310007, "log_rms_mae_db": 1.2248420811940868, "rms_envelope_nrmse": 0.08846566536398857, "envelope_correlation": 0.9937224322412479, "rms_gain_error_db": 0.3926025388733516, "reference_mean_power": 0.005865946925483087, "output_mean_power": 0.006420936620096956}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.084203911038529e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015941324207372327, "output_power": 0.0016865724348692174, "reference_active": true, "delta_db": 0.24480603639714202}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00018248750335661006, "output_power": 0.00017816714214157443, "reference_active": true, "delta_db": -0.10405515735621237}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.552526527817181e-06, "output_power": 2.298001514468247e-06, "reference_active": true, "delta_db": -2.969021745631235}], "active_band_count": 3, "active_band_mae_db": 1.1059609797948633}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 0.00018690652054173922, "log_rms_mae_db": 2.487926406483803, "rms_envelope_nrmse": 0.18531901048517468, "envelope_correlation": 0.9782573937020033, "rms_gain_error_db": 0.5615201015343801, "reference_mean_power": 0.0020929267743018004, "output_mean_power": 0.0023818041300946585}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 8.903808595053404e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005237826202169963, "output_power": 0.0005785268709570448, "reference_active": true, "delta_db": 0.43172451644579896}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020282121471265822, "output_power": 0.00020517982403537926, "reference_active": true, "delta_db": 0.050212736114456445}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.57818052122161e-05, "output_power": 1.3406146260234769e-05, "reference_active": true, "delta_db": -2.84009369670168}], "active_band_count": 3, "active_band_mae_db": 1.1073436497539784}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 0.00016950242057071555, "log_rms_mae_db": 1.7386900252548312, "rms_envelope_nrmse": 0.12081625249071663, "envelope_correlation": 0.9918425915585528, "rms_gain_error_db": 0.5615499850495628, "reference_mean_power": 0.002104315784196243, "output_mean_power": 0.002394781592547718}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 7.681995008528509e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000538193996495516, "output_power": 0.0005854222363901506, "reference_active": true, "delta_db": 0.3653036549916944}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010305287620923108, "output_power": 0.00010505017873565325, "reference_active": true, "delta_db": 0.08336678206847227}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.715218896542221e-06, "output_power": 1.2544110166781784e-06, "reference_active": true, "delta_db": -3.3536498796897174}], "active_band_count": 3, "active_band_mae_db": 1.2674401055832947}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 7.550458878286076e-05, "log_rms_mae_db": 2.187210650325183, "rms_envelope_nrmse": 0.19671424512819233, "envelope_correlation": 0.9728884785113232, "rms_gain_error_db": 0.32243053282515544, "reference_mean_power": 0.00021315330376123098, "output_mean_power": 0.00022958056568478567}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 1.6207032137826829e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010202948507054384, "output_power": 0.0001118427523524601, "reference_active": true, "delta_db": 0.39882151310521585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.692530062160877e-05, "output_power": 3.483097836742227e-05, "reference_active": true, "delta_db": -0.2535836621140245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.752197568216375e-06, "output_power": 4.2183797557282625e-06, "reference_active": true, "delta_db": -2.0429946706862685}], "active_band_count": 3, "active_band_mae_db": 0.8984666153018362}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 5.938009261210461e-05, "log_rms_mae_db": 1.3689331094616566, "rms_envelope_nrmse": 0.10621576143291325, "envelope_correlation": 0.9918860429458589, "rms_gain_error_db": 0.32277578053732076, "reference_mean_power": 0.00021425147153668813, "output_mean_power": 0.0002307817124096892}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 1.4064708995422735e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010396272320640846, "output_power": 0.00011185654729337645, "reference_active": true, "delta_db": 0.3178376279643681}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97231105181685e-05, "output_power": 1.8211037526860748e-05, "reference_active": true, "delta_db": -0.346407189831855}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.307963255185344e-07, "output_power": 3.31175677983845e-07, "reference_active": true, "delta_db": -2.7983072075975546}], "active_band_count": 3, "active_band_mae_db": 1.1541840084645925}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1105, "rms_floor": 4.4291934004628345e-05, "log_rms_mae_db": 2.7068681174312603, "rms_envelope_nrmse": 0.2825993624534353, "envelope_correlation": 0.9493890515357651, "rms_gain_error_db": 0.5767201644261278, "reference_mean_power": 2.6417005173666327e-05, "output_mean_power": 3.0168630624582495e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1105, "frequency_resolution_hz": 0.18099547511312217, "power_floor": 2.694560310630941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.502841409588563e-05, "output_power": 1.8338982374519964e-05, "reference_active": true, "delta_db": 0.8646207991994828}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.7026465908103935e-06, "output_power": 7.329519578149837e-06, "reference_active": true, "delta_db": 0.38829188205467713}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.011540321330325e-07, "output_power": 6.927390745304745e-07, "reference_active": true, "delta_db": -1.1422934518189807}], "active_band_count": 3, "active_band_mae_db": 0.7984020443577137}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 88480, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1099, "rms_floor": 3.2524867537670264e-05, "log_rms_mae_db": 1.8260586788034725, "rms_envelope_nrmse": 0.17674554733192818, "envelope_correlation": 0.9801415997992353, "rms_gain_error_db": 0.5773455716001847, "reference_mean_power": 2.65459352580292e-05, "output_mean_power": 3.0320236746591035e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1099, "frequency_resolution_hz": 0.18198362147406733, "power_floor": 2.4345030346700597e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.6238306479668564e-05, "output_power": 1.9655648015606524e-05, "reference_active": true, "delta_db": 0.8294663242996421}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6930369104619132e-06, "output_power": 3.898584062949806e-06, "reference_active": true, "delta_db": 0.23523254826010265}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.897644209500044e-08, "output_power": 4.636979037089567e-08, "reference_active": true, "delta_db": -2.3126243224872454}], "active_band_count": 3, "active_band_mae_db": 1.1257743983489967}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.913831536489437, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "86ae8233b7a71d9f9e83872e7fbe5b5799dfabf5d00d19d71f6af56445753e5a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9728884785113232, "standard_deviation": null, "minimum": 0.9728884785113232, "maximum": 0.9728884785113232, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 2.187210650325183, "standard_deviation": null, "minimum": 2.187210650325183, "maximum": 2.187210650325183, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8984666153018362, "standard_deviation": null, "minimum": 0.8984666153018362, "maximum": 0.8984666153018362, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19671424512819233, "standard_deviation": null, "minimum": 0.19671424512819233, "maximum": 0.19671424512819233, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9918860429458589, "standard_deviation": null, "minimum": 0.9918860429458589, "maximum": 0.9918860429458589, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.3689331094616566, "standard_deviation": null, "minimum": 1.3689331094616566, "maximum": 1.3689331094616566, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1541840084645925, "standard_deviation": null, "minimum": 1.1541840084645925, "maximum": 1.1541840084645925, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10621576143291325, "standard_deviation": null, "minimum": 0.10621576143291325, "maximum": 0.10621576143291325, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9493890515357651, "standard_deviation": null, "minimum": 0.9493890515357651, "maximum": 0.9493890515357651, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 2.7068681174312603, "standard_deviation": null, "minimum": 2.7068681174312603, "maximum": 2.7068681174312603, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.7984020443577137, "standard_deviation": null, "minimum": 0.7984020443577137, "maximum": 0.7984020443577137, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2825993624534353, "standard_deviation": null, "minimum": 0.2825993624534353, "maximum": 0.2825993624534353, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9801415997992353, "standard_deviation": null, "minimum": 0.9801415997992353, "maximum": 0.9801415997992353, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.8260586788034725, "standard_deviation": null, "minimum": 1.8260586788034725, "maximum": 1.8260586788034725, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1257743983489967, "standard_deviation": null, "minimum": 1.1257743983489967, "maximum": 1.1257743983489967, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17674554733192818, "standard_deviation": null, "minimum": 0.17674554733192818, "maximum": 0.17674554733192818, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9782573937020033, "standard_deviation": null, "minimum": 0.9782573937020033, "maximum": 0.9782573937020033, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.487926406483803, "standard_deviation": null, "minimum": 2.487926406483803, "maximum": 2.487926406483803, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1073436497539784, "standard_deviation": null, "minimum": 1.1073436497539784, "maximum": 1.1073436497539784, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18531901048517468, "standard_deviation": null, "minimum": 0.18531901048517468, "maximum": 0.18531901048517468, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9918425915585528, "standard_deviation": null, "minimum": 0.9918425915585528, "maximum": 0.9918425915585528, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7386900252548312, "standard_deviation": null, "minimum": 1.7386900252548312, "maximum": 1.7386900252548312, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.2674401055832947, "standard_deviation": null, "minimum": 1.2674401055832947, "maximum": 1.2674401055832947, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12081625249071663, "standard_deviation": null, "minimum": 0.12081625249071663, "maximum": 0.12081625249071663, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.37408928468164593, "standard_deviation": null, "minimum": 0.37408928468164593, "maximum": 0.37408928468164593, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.314328008633088, "standard_deviation": null, "minimum": 6.314328008633088, "maximum": 6.314328008633088, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7719099371952937, "standard_deviation": null, "minimum": 0.7719099371952937, "maximum": 0.7719099371952937, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.3925743113461682, "standard_deviation": null, "minimum": 0.3925743113461682, "maximum": 0.3925743113461682, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9803575600338925, "standard_deviation": null, "minimum": 0.9803575600338925, "maximum": 0.9803575600338925, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.9276043214367375, "standard_deviation": null, "minimum": 1.9276043214367375, "maximum": 1.9276043214367375, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8064827347531779, "standard_deviation": null, "minimum": 0.8064827347531779, "maximum": 0.8064827347531779, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1495118187643159, "standard_deviation": null, "minimum": 0.1495118187643159, "maximum": 0.1495118187643159, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9937224322412479, "standard_deviation": null, "minimum": 0.9937224322412479, "maximum": 0.9937224322412479, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2248420811940868, "standard_deviation": null, "minimum": 1.2248420811940868, "maximum": 1.2248420811940868, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1059609797948633, "standard_deviation": null, "minimum": 1.1059609797948633, "maximum": 1.1059609797948633, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.08846566536398857, "standard_deviation": null, "minimum": 0.08846566536398857, "maximum": 0.08846566536398857, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7176"}, {"case": "librispeech_7729-102255-0028", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.7588675022125244, "predicted_mos": 3.7588675022125244, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "39d71d910d0da49953239b694bca2c39d56434d748a31fd512c7659801342d23", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.0991254998371005}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6601090358104291, "全帯域_10ms_rms_envelope_nrmse": 0.20466116694975073, "全帯域_10ms_envelope_correlation": 0.979122473967667, "全帯域_10ms_modulation_mae_db": 1.2029926322378708, "全帯域_40ms_log_rms_mae_db": 1.30887187898085, "全帯域_40ms_rms_envelope_nrmse": 0.16855055362631455, "全帯域_40ms_envelope_correlation": 0.9909245831259014, "全帯域_40ms_modulation_mae_db": 1.4717075496970564, "300–1000 Hz_10ms_log_rms_mae_db": 1.9620395429379418, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2443025210468253, "300–1000 Hz_10ms_envelope_correlation": 0.974902361681151, "300–1000 Hz_10ms_modulation_mae_db": 1.3791250533251909, "300–1000 Hz_40ms_log_rms_mae_db": 1.6686776741565852, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2047777573259028, "300–1000 Hz_40ms_envelope_correlation": 0.9875703865491752, "300–1000 Hz_40ms_modulation_mae_db": 1.5361642898867889, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5707796311489282, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16833384546262567, "1000–3000 Hz_10ms_envelope_correlation": 0.9852101888608014, "1000–3000 Hz_10ms_modulation_mae_db": 0.8812197833154659, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1064393946370021, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11721404620262826, "1000–3000 Hz_40ms_envelope_correlation": 0.9950705169439644, "1000–3000 Hz_40ms_modulation_mae_db": 1.0196469287637488, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7962420613757475, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21787040759160547, "3000–6000 Hz_10ms_envelope_correlation": 0.9687621173306247, "3000–6000 Hz_10ms_modulation_mae_db": 0.6824264884457065, "3000–6000 Hz_40ms_log_rms_mae_db": 1.095480108361247, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1386751721292226, "3000–6000 Hz_40ms_envelope_correlation": 0.9875907254044548, "3000–6000 Hz_40ms_modulation_mae_db": 0.8535392899431814, "mr_stft_db": 5.940152038208287, "band_mae_db": 0.8154634075777238, "rms_gain_error_db": 1.1261006779290623, "output_peak": 0.8516773440140493}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.670805330059281, "spectral_convergence": 0.678258957473733, "spectral_convergence_status": "ok", "amplitude_floor": 1.7851227833840954e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.36953864433175, "spectral_convergence": 0.345536776814476, "spectral_convergence_status": "ok", "amplitude_floor": 1.2048994709482952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.780112140233828, "spectral_convergence": 0.3192076540345787, "spectral_convergence_status": "ok", "amplitude_floor": 9.468117719284587e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005960090393190935, "output_power": 0.0007420532055990076, "delta_db": 0.9518219905506322, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030564977634421338, "output_power": 0.0040254217895372, "delta_db": 1.195873107584396, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0008630746056010172, "output_power": 0.0011470482984540392, "delta_db": 1.2353336653365157, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003160541845765363, "output_power": 0.0003807919103801189, "delta_db": 0.8092616878409498, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001143536135409032, "output_power": 0.0001211570845016164, "delta_db": 0.2509892127744032, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.420642909107181e-05, "output_power": 3.793654863617948e-05, "delta_db": 0.44950078137944594, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8154634075777238, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1261006779290623, "pitch": null, "output_rms_dbfs": -21.90143528848978, "output_peak": 0.8516773440140493, "reference_rms_dbfs": -23.02753596641884, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.940152038208287}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023539858364162597, "log_rms_mae_db": 1.6601090358104291, "rms_envelope_nrmse": 0.20466116694975073, "envelope_correlation": 0.979122473967667, "rms_gain_error_db": 1.126943319006409, "reference_mean_power": 0.004983368761309487, "output_mean_power": 0.00645977449399905}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0209735395978665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009552928696150586, "output_power": 0.0012384886149440191, "reference_active": true, "delta_db": 1.1275548245278795}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014624552237472608, "output_power": 0.00017334075826982206, "reference_active": true, "delta_db": 0.7381811394377882}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.510798757145359e-05, "output_power": 1.680690370246072e-05, "reference_active": true, "delta_db": -1.7432419327479443}], "active_band_count": 3, "active_band_mae_db": 1.2029926322378708}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002083747898533542, "log_rms_mae_db": 1.30887187898085, "rms_envelope_nrmse": 0.16855055362631455, "envelope_correlation": 0.9909245831259014, "rms_gain_error_db": 1.130014078333908, "reference_mean_power": 0.004999205734592124, "output_mean_power": 0.006484887072037198}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8665599913970187e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008900977663526511, "output_power": 0.0011489192503735823, "reference_active": true, "delta_db": 1.1085179498914868}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.407197015766335e-05, "output_power": 8.808749890124791e-05, "reference_active": true, "delta_db": 0.7526038296875782}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8378799869222006e-06, "output_power": 1.0207445309879587e-06, "reference_active": true, "delta_db": -2.5540008695121044}], "active_band_count": 3, "active_band_mae_db": 1.4717075496970564}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019850808969518194, "log_rms_mae_db": 1.9620395429379418, "rms_envelope_nrmse": 0.2443025210468253, "envelope_correlation": 0.974902361681151, "rms_gain_error_db": 1.3196896323158307, "reference_mean_power": 0.003121339390070273, "output_mean_power": 0.004229703808038275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3811041144776226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007780170156378303, "output_power": 0.0010493111598848031, "reference_active": true, "delta_db": 1.2991519645921743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001204320773242115, "output_power": 0.0001509285805936962, "reference_active": true, "delta_db": 0.9802931021463959}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.313510568503667e-05, "output_power": 1.50826786616084e-05, "reference_active": true, "delta_db": -1.8579300932370026}], "active_band_count": 3, "active_band_mae_db": 1.3791250533251909}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00016710142967257498, "log_rms_mae_db": 1.6686776741565852, "rms_envelope_nrmse": 0.2047777573259028, "envelope_correlation": 0.9875703865491752, "rms_gain_error_db": 1.3213346612287613, "reference_mean_power": 0.003133268891618282, "output_mean_power": 0.004247477950043486}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2683362614683906e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007382464262949856, "output_power": 0.0009874803727039018, "reference_active": true, "delta_db": 1.2632711897878721}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.8108189032543396e-05, "output_power": 7.267205591452773e-05, "reference_active": true, "delta_db": 0.9713010602008232}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.916170233531451e-06, "output_power": 1.1092821572273255e-06, "reference_active": true, "delta_db": -2.373920619671672}], "active_band_count": 3, "active_band_mae_db": 1.5361642898867889}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012483338724103395, "log_rms_mae_db": 1.5707796311489282, "rms_envelope_nrmse": 0.16833384546262567, "envelope_correlation": 0.9852101888608014, "rms_gain_error_db": 0.6953922193122029, "reference_mean_power": 0.00037365748369267646, "output_mean_power": 0.00043854373061838806}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.54909855359075e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.931499876625785e-05, "output_power": 9.790543330296087e-05, "reference_active": true, "delta_db": 0.39882397261450125}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4977670974468328e-05, "output_power": 1.720920278762212e-05, "reference_active": true, "delta_db": 0.6031646631266501}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.3989606403687574e-06, "output_power": 2.329051351670262e-06, "reference_active": true, "delta_db": -1.6416707142052465}], "active_band_count": 3, "active_band_mae_db": 0.8812197833154659}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00010596065945345535, "log_rms_mae_db": 1.1064393946370021, "rms_envelope_nrmse": 0.11721404620262826, "envelope_correlation": 0.9950705169439644, "rms_gain_error_db": 0.69590542442931, "reference_mean_power": 0.00037540546219302315, "output_mean_power": 0.00044064731665724205}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.4355698091451528e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 8.577569295489126e-05, "output_power": 9.298866627341155e-05, "reference_active": true, "delta_db": 0.35065783445492593}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.83374089374301e-06, "output_power": 9.241979867130206e-06, "reference_active": true, "delta_db": 0.7179581524313399}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.527528900320414e-07, "output_power": 1.5983196919322348e-07, "reference_active": true, "delta_db": -1.9903247994049806}], "active_band_count": 3, "active_band_mae_db": 1.0196469287637488}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.3292043050207156e-05, "log_rms_mae_db": 1.7962420613757475, "rms_envelope_nrmse": 0.21787040759160547, "envelope_correlation": 0.9687621173306247, "rms_gain_error_db": 0.4098662113453058, "reference_mean_power": 5.712588023845408e-05, "output_mean_power": 6.277974194375874e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.279319333258023e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4391078150208969e-05, "output_power": 1.503003114433959e-05, "reference_active": true, "delta_db": 0.18866548895298785}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.344791688657969e-06, "output_power": 4.6136160963933385e-06, "reference_active": true, "delta_db": 0.2607249460616184}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.199993664659886e-07, "output_power": 1.1846837185935053e-06, "reference_active": true, "delta_db": 1.5978890303225133}], "active_band_count": 3, "active_band_mae_db": 0.6824264884457065}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.870517765122232e-05, "log_rms_mae_db": 1.095480108361247, "rms_envelope_nrmse": 0.1386751721292226, "envelope_correlation": 0.9875907254044548, "rms_gain_error_db": 0.4098895264501168, "reference_mean_power": 5.74115094616415e-05, "output_mean_power": 6.309397917705383e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.057417179151539e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3970684782124775e-05, "output_power": 1.4885245510954966e-05, "reference_active": true, "delta_db": 0.27538308251619137}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.241188995835654e-06, "output_power": 2.4036848909037113e-06, "reference_active": true, "delta_db": 0.30399052154715345}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.11366674713116e-08, "output_power": 8.069689947204597e-08, "reference_active": true, "delta_db": 1.981244265766199}], "active_band_count": 3, "active_band_mae_db": 0.8535392899431814}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9048022475197683, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d9cd66f518e247fbceb705f13a96a64554ed39684e0b0c65712ea0598f910ba2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9852101888608014, "standard_deviation": null, "minimum": 0.9852101888608014, "maximum": 0.9852101888608014, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5707796311489282, "standard_deviation": null, "minimum": 1.5707796311489282, "maximum": 1.5707796311489282, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8812197833154659, "standard_deviation": null, "minimum": 0.8812197833154659, "maximum": 0.8812197833154659, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16833384546262567, "standard_deviation": null, "minimum": 0.16833384546262567, "maximum": 0.16833384546262567, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9950705169439644, "standard_deviation": null, "minimum": 0.9950705169439644, "maximum": 0.9950705169439644, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1064393946370021, "standard_deviation": null, "minimum": 1.1064393946370021, "maximum": 1.1064393946370021, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.0196469287637488, "standard_deviation": null, "minimum": 1.0196469287637488, "maximum": 1.0196469287637488, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11721404620262826, "standard_deviation": null, "minimum": 0.11721404620262826, "maximum": 0.11721404620262826, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9687621173306247, "standard_deviation": null, "minimum": 0.9687621173306247, "maximum": 0.9687621173306247, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7962420613757475, "standard_deviation": null, "minimum": 1.7962420613757475, "maximum": 1.7962420613757475, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.6824264884457065, "standard_deviation": null, "minimum": 0.6824264884457065, "maximum": 0.6824264884457065, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21787040759160547, "standard_deviation": null, "minimum": 0.21787040759160547, "maximum": 0.21787040759160547, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9875907254044548, "standard_deviation": null, "minimum": 0.9875907254044548, "maximum": 0.9875907254044548, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.095480108361247, "standard_deviation": null, "minimum": 1.095480108361247, "maximum": 1.095480108361247, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.8535392899431814, "standard_deviation": null, "minimum": 0.8535392899431814, "maximum": 0.8535392899431814, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1386751721292226, "standard_deviation": null, "minimum": 0.1386751721292226, "maximum": 0.1386751721292226, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.974902361681151, "standard_deviation": null, "minimum": 0.974902361681151, "maximum": 0.974902361681151, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9620395429379418, "standard_deviation": null, "minimum": 1.9620395429379418, "maximum": 1.9620395429379418, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3791250533251909, "standard_deviation": null, "minimum": 1.3791250533251909, "maximum": 1.3791250533251909, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2443025210468253, "standard_deviation": null, "minimum": 0.2443025210468253, "maximum": 0.2443025210468253, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9875703865491752, "standard_deviation": null, "minimum": 0.9875703865491752, "maximum": 0.9875703865491752, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6686776741565852, "standard_deviation": null, "minimum": 1.6686776741565852, "maximum": 1.6686776741565852, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.5361642898867889, "standard_deviation": null, "minimum": 1.5361642898867889, "maximum": 1.5361642898867889, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2047777573259028, "standard_deviation": null, "minimum": 0.2047777573259028, "maximum": 0.2047777573259028, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8154634075777238, "standard_deviation": null, "minimum": 0.8154634075777238, "maximum": 0.8154634075777238, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.940152038208287, "standard_deviation": null, "minimum": 5.940152038208287, "maximum": 5.940152038208287, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.8516773440140493, "standard_deviation": null, "minimum": 0.8516773440140493, "maximum": 0.8516773440140493, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1261006779290623, "standard_deviation": null, "minimum": 1.1261006779290623, "maximum": 1.1261006779290623, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.979122473967667, "standard_deviation": null, "minimum": 0.979122473967667, "maximum": 0.979122473967667, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6601090358104291, "standard_deviation": null, "minimum": 1.6601090358104291, "maximum": 1.6601090358104291, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.2029926322378708, "standard_deviation": null, "minimum": 1.2029926322378708, "maximum": 1.2029926322378708, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20466116694975073, "standard_deviation": null, "minimum": 0.20466116694975073, "maximum": 0.20466116694975073, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9909245831259014, "standard_deviation": null, "minimum": 0.9909245831259014, "maximum": 0.9909245831259014, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.30887187898085, "standard_deviation": null, "minimum": 1.30887187898085, "maximum": 1.30887187898085, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4717075496970564, "standard_deviation": null, "minimum": 1.4717075496970564, "maximum": 1.4717075496970564, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16855055362631455, "standard_deviation": null, "minimum": 0.16855055362631455, "maximum": 0.16855055362631455, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7729"}, {"case": "librispeech_7729-102255-0038", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 4.036701202392578, "predicted_mos": 4.036701202392578, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "030e3aed6f0f10c95c33fbd2bc79184342e27c45417f0aab1199cde7722ff371", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.818293099757284}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.9166411178240697, "全帯域_10ms_rms_envelope_nrmse": 0.22319637384141933, "全帯域_10ms_envelope_correlation": 0.9716402562695815, "全帯域_10ms_modulation_mae_db": 1.007840166886917, "全帯域_40ms_log_rms_mae_db": 1.3050683288036866, "全帯域_40ms_rms_envelope_nrmse": 0.16703624049086527, "全帯域_40ms_envelope_correlation": 0.9901045886862889, "全帯域_40ms_modulation_mae_db": 1.2167617306244893, "300–1000 Hz_10ms_log_rms_mae_db": 2.1707922460053384, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.26320680351726916, "300–1000 Hz_10ms_envelope_correlation": 0.9719936887827282, "300–1000 Hz_10ms_modulation_mae_db": 1.7313942419960782, "300–1000 Hz_40ms_log_rms_mae_db": 1.7062223525423375, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.21176688715949474, "300–1000 Hz_40ms_envelope_correlation": 0.9882781419066393, "300–1000 Hz_40ms_modulation_mae_db": 1.667911076370648, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7854187102096295, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17771003510855587, "1000–3000 Hz_10ms_envelope_correlation": 0.9819905168396089, "1000–3000 Hz_10ms_modulation_mae_db": 0.9826634072662795, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1174054171754966, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10430099198693632, "1000–3000 Hz_40ms_envelope_correlation": 0.9955386488114761, "1000–3000 Hz_40ms_modulation_mae_db": 1.1495515701910846, "3000–6000 Hz_10ms_log_rms_mae_db": 1.945634853562762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21630975284147272, "3000–6000 Hz_10ms_envelope_correlation": 0.9687294863371113, "3000–6000 Hz_10ms_modulation_mae_db": 1.5839024349890563, "3000–6000 Hz_40ms_log_rms_mae_db": 1.248197049357732, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12116096209762708, "3000–6000 Hz_40ms_envelope_correlation": 0.9902388350657423, "3000–6000 Hz_40ms_modulation_mae_db": 1.5418482283697958, "mr_stft_db": 6.121113908168262, "band_mae_db": 0.6488272692043354, "rms_gain_error_db": 1.0679422709879423, "output_peak": 0.823267041877092}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.066528782618345, "spectral_convergence": 0.8188246778939143, "spectral_convergence_status": "ok", "amplitude_floor": 1.6152428194526273e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5496590879039625, "spectral_convergence": 0.4041573943195765, "spectral_convergence_status": "ok", "amplitude_floor": 1.0855275118818902e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7471538539824785, "spectral_convergence": 0.3166722474806706, "spectral_convergence_status": "ok", "amplitude_floor": 7.682277294566263e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006930713321042831, "output_power": 0.0007883543747727688, "delta_db": 0.5594354671851046, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002559270540127681, "output_power": 0.00347393042531389, "delta_db": 1.3270491877508164, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007234666478039212, "output_power": 0.000879819586225526, "delta_db": 0.8497511114455712, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002266743638240336, "output_power": 0.00026195191647543527, "delta_db": 0.6281917462370544, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.48754354732022e-05, "output_power": 6.187330013194391e-05, "delta_db": -0.20577005474252086, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.432671244156477e-05, "output_power": 4.77465643718908e-05, "delta_db": 0.322766047864945, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6488272692043354, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0679422709879423, "pitch": null, "output_rms_dbfs": -22.58558744738831, "output_peak": 0.823267041877092, "reference_rms_dbfs": -23.65352971837625, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.121113908168262}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019088238002709922, "log_rms_mae_db": 1.9166411178240697, "rms_envelope_nrmse": 0.22319637384141933, "envelope_correlation": 0.9716402562695815, "rms_gain_error_db": 1.0679423981837803, "reference_mean_power": 0.004315280923443156, "output_mean_power": 0.005518274666230007}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.8761677888507434e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007125355736933842, "output_power": 0.00095727048598279, "reference_active": true, "delta_db": 1.282281176098946}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011942531791187452, "output_power": 0.0001402260982641169, "reference_active": true, "delta_db": 0.6973244407555163}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5392584995669358e-05, "output_power": 1.996711992845223e-05, "reference_active": true, "delta_db": -1.0439148838062886}], "active_band_count": 3, "active_band_mae_db": 1.007840166886917}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018007320980187037, "log_rms_mae_db": 1.3050683288036866, "rms_envelope_nrmse": 0.16703624049086527, "envelope_correlation": 0.9901045886862889, "rms_gain_error_db": 1.0679432185325515, "reference_mean_power": 0.004336982504671315, "output_mean_power": 0.005546027159921268}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7431428852385876e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006439047386291532, "output_power": 0.000861352173468495, "reference_active": true, "delta_db": 1.2635913275160437}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.115354721039444e-05, "output_power": 5.9513669418805054e-05, "reference_active": true, "delta_db": 0.6574097330329378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9654159892814645e-06, "output_power": 1.3198545433070416e-06, "reference_active": true, "delta_db": -1.7292841313244864}], "active_band_count": 3, "active_band_mae_db": 1.2167617306244893}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017279669295308916, "log_rms_mae_db": 2.1707922460053384, "rms_envelope_nrmse": 0.26320680351726916, "envelope_correlation": 0.9719936887827282, "rms_gain_error_db": 1.3665406149792083, "reference_mean_power": 0.0026069520032741206, "output_mean_power": 0.003570977361216946}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.2087875823984064e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005529275902426358, "output_power": 0.0008039330819201093, "reference_active": true, "delta_db": 1.625516391767635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.648535673845476e-05, "output_power": 0.00013518379117077937, "reference_active": true, "delta_db": 1.464632149621816}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.705863634087677e-05, "output_power": 1.0508503965877624e-05, "reference_active": true, "delta_db": -2.1040341845987833}], "active_band_count": 3, "active_band_mae_db": 1.7313942419960782}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014230358058689408, "log_rms_mae_db": 1.7062223525423375, "rms_envelope_nrmse": 0.21176688715949474, "envelope_correlation": 0.9882781419066393, "rms_gain_error_db": 1.3665406595950595, "reference_mean_power": 0.0026200630509491097, "output_mean_power": 0.003588936782981596}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.1215946456987494e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005050917134712818, "output_power": 0.0007290149116194545, "reference_active": true, "delta_db": 1.593661681351159}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.798176008445663e-05, "output_power": 5.274876978062476e-05, "reference_active": true, "delta_db": 1.426372491840127}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3776930382412419e-06, "output_power": 8.725344001000966e-07, "reference_active": true, "delta_db": -1.9836990559206582}], "active_band_count": 3, "active_band_mae_db": 1.667911076370648}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.058549471038879e-05, "log_rms_mae_db": 1.7854187102096295, "rms_envelope_nrmse": 0.17771003510855587, "envelope_correlation": 0.9819905168396089, "rms_gain_error_db": 0.5570522258331477, "reference_mean_power": 0.0002448429483315042, "output_mean_power": 0.0002783510238492626}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7588735706755028e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.377351846027054e-05, "output_power": 0.00010480306845780521, "reference_active": true, "delta_db": 0.48293786843515996}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.692404819271496e-06, "output_power": 9.35897412496906e-06, "reference_active": true, "delta_db": -0.15203298298357754}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.8112011913022686e-06, "output_power": 2.2374840005603086e-06, "reference_active": true, "delta_db": -2.3130193703801014}], "active_band_count": 3, "active_band_mae_db": 0.9826634072662795}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.399057463688353e-05, "log_rms_mae_db": 1.1174054171754966, "rms_envelope_nrmse": 0.10430099198693632, "envelope_correlation": 0.9955386488114761, "rms_gain_error_db": 0.5570548027774609, "reference_mean_power": 0.0002460738466011548, "output_mean_power": 0.00027975054316695856}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6710797473991227e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.068314304092372e-05, "output_power": 0.00010022149380585214, "reference_active": true, "delta_db": 0.43434307447619236}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.419207768284792e-06, "output_power": 4.236025145520559e-06, "reference_active": true, "delta_db": -0.18385890857965825}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.686031483587781e-07, "output_power": 1.3997994742967759e-07, "reference_active": true, "delta_db": -2.8304527275174034}], "active_band_count": 3, "active_band_mae_db": 1.1495515701910846}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.748604525835054e-05, "log_rms_mae_db": 1.945634853562762, "rms_envelope_nrmse": 0.21630975284147272, "envelope_correlation": 0.9687294863371113, "rms_gain_error_db": -0.25644339959164697, "reference_mean_power": 6.169396290931914e-05, "output_mean_power": 5.8156508882763656e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.220333375078007e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9572001812371347e-05, "output_power": 1.778487630840207e-05, "reference_active": true, "delta_db": -0.41584398235346265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.501529531199527e-06, "output_power": 5.998025984720418e-06, "reference_active": true, "delta_db": -1.5148872485460732}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.287934963899872e-07, "output_power": 4.850891471536426e-07, "reference_active": true, "delta_db": -2.820976074067633}], "active_band_count": 3, "active_band_mae_db": 1.5839024349890563}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.636845752844561e-05, "log_rms_mae_db": 1.248197049357732, "rms_envelope_nrmse": 0.12116096209762708, "envelope_correlation": 0.9902388350657423, "rms_gain_error_db": -0.2564425323117407, "reference_mean_power": 6.200418008997283e-05, "output_mean_power": 5.844895027368683e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.99689742660592e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0058245625049685e-05, "output_power": 1.7739188512327907e-05, "reference_active": true, "delta_db": -0.533591962597982}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.381270388794365e-06, "output_power": 2.9490757302537826e-06, "reference_active": true, "delta_db": -1.719141312784418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.024894795639882e-07, "output_power": 5.934691481807742e-08, "reference_active": true, "delta_db": -2.372811409726988}], "active_band_count": 3, "active_band_mae_db": 1.5418482283697958}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8957996802797924, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "8e8eaa2a4bfc54b9f27201fcfda930cc3ff8d80a8473bcb5109f9b312b105ac2", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9819905168396089, "standard_deviation": null, "minimum": 0.9819905168396089, "maximum": 0.9819905168396089, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7854187102096295, "standard_deviation": null, "minimum": 1.7854187102096295, "maximum": 1.7854187102096295, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9826634072662795, "standard_deviation": null, "minimum": 0.9826634072662795, "maximum": 0.9826634072662795, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17771003510855587, "standard_deviation": null, "minimum": 0.17771003510855587, "maximum": 0.17771003510855587, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9955386488114761, "standard_deviation": null, "minimum": 0.9955386488114761, "maximum": 0.9955386488114761, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1174054171754966, "standard_deviation": null, "minimum": 1.1174054171754966, "maximum": 1.1174054171754966, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1495515701910846, "standard_deviation": null, "minimum": 1.1495515701910846, "maximum": 1.1495515701910846, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10430099198693632, "standard_deviation": null, "minimum": 0.10430099198693632, "maximum": 0.10430099198693632, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9687294863371113, "standard_deviation": null, "minimum": 0.9687294863371113, "maximum": 0.9687294863371113, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.945634853562762, "standard_deviation": null, "minimum": 1.945634853562762, "maximum": 1.945634853562762, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5839024349890563, "standard_deviation": null, "minimum": 1.5839024349890563, "maximum": 1.5839024349890563, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21630975284147272, "standard_deviation": null, "minimum": 0.21630975284147272, "maximum": 0.21630975284147272, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9902388350657423, "standard_deviation": null, "minimum": 0.9902388350657423, "maximum": 0.9902388350657423, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.248197049357732, "standard_deviation": null, "minimum": 1.248197049357732, "maximum": 1.248197049357732, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.5418482283697958, "standard_deviation": null, "minimum": 1.5418482283697958, "maximum": 1.5418482283697958, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12116096209762708, "standard_deviation": null, "minimum": 0.12116096209762708, "maximum": 0.12116096209762708, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9719936887827282, "standard_deviation": null, "minimum": 0.9719936887827282, "maximum": 0.9719936887827282, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1707922460053384, "standard_deviation": null, "minimum": 2.1707922460053384, "maximum": 2.1707922460053384, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.7313942419960782, "standard_deviation": null, "minimum": 1.7313942419960782, "maximum": 1.7313942419960782, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26320680351726916, "standard_deviation": null, "minimum": 0.26320680351726916, "maximum": 0.26320680351726916, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9882781419066393, "standard_deviation": null, "minimum": 0.9882781419066393, "maximum": 0.9882781419066393, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7062223525423375, "standard_deviation": null, "minimum": 1.7062223525423375, "maximum": 1.7062223525423375, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.667911076370648, "standard_deviation": null, "minimum": 1.667911076370648, "maximum": 1.667911076370648, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.21176688715949474, "standard_deviation": null, "minimum": 0.21176688715949474, "maximum": 0.21176688715949474, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6488272692043354, "standard_deviation": null, "minimum": 0.6488272692043354, "maximum": 0.6488272692043354, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.121113908168262, "standard_deviation": null, "minimum": 6.121113908168262, "maximum": 6.121113908168262, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.823267041877092, "standard_deviation": null, "minimum": 0.823267041877092, "maximum": 0.823267041877092, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0679422709879423, "standard_deviation": null, "minimum": 1.0679422709879423, "maximum": 1.0679422709879423, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9716402562695815, "standard_deviation": null, "minimum": 0.9716402562695815, "maximum": 0.9716402562695815, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.9166411178240697, "standard_deviation": null, "minimum": 1.9166411178240697, "maximum": 1.9166411178240697, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.007840166886917, "standard_deviation": null, "minimum": 1.007840166886917, "maximum": 1.007840166886917, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.22319637384141933, "standard_deviation": null, "minimum": 0.22319637384141933, "maximum": 0.22319637384141933, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9901045886862889, "standard_deviation": null, "minimum": 0.9901045886862889, "maximum": 0.9901045886862889, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.3050683288036866, "standard_deviation": null, "minimum": 1.3050683288036866, "maximum": 1.3050683288036866, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.2167617306244893, "standard_deviation": null, "minimum": 1.2167617306244893, "maximum": 1.2167617306244893, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.16703624049086527, "standard_deviation": null, "minimum": 0.16703624049086527, "maximum": 0.16703624049086527, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7729"}, {"case": "librispeech_7729-102255-0042", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.2990152835845947, "predicted_mos": 3.2990152835845947, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6a4a93a03daa9aa7b481eb45c184f3d0d5a86fbe32a46f51c1463fa01c25c15b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.7046345002017915}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.683723203470434, "全帯域_10ms_rms_envelope_nrmse": 0.16775747271496322, "全帯域_10ms_envelope_correlation": 0.9837680060569017, "全帯域_10ms_modulation_mae_db": 0.7359412120795382, "全帯域_40ms_log_rms_mae_db": 1.1578350486113893, "全帯域_40ms_rms_envelope_nrmse": 0.12340131684371916, "全帯域_40ms_envelope_correlation": 0.9947410507100278, "全帯域_40ms_modulation_mae_db": 1.1707930291903665, "300–1000 Hz_10ms_log_rms_mae_db": 1.6824051327982348, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.19446051932908592, "300–1000 Hz_10ms_envelope_correlation": 0.9828384417658632, "300–1000 Hz_10ms_modulation_mae_db": 1.3046464961691004, "300–1000 Hz_40ms_log_rms_mae_db": 1.1668549116829519, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.14999933083105796, "300–1000 Hz_40ms_envelope_correlation": 0.9936359153335587, "300–1000 Hz_40ms_modulation_mae_db": 1.4621056664045977, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7256042888667515, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1742786887369514, "1000–3000 Hz_10ms_envelope_correlation": 0.983036166267795, "1000–3000 Hz_10ms_modulation_mae_db": 0.8656514294205852, "1000–3000 Hz_40ms_log_rms_mae_db": 1.111647207665029, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11412189888398126, "1000–3000 Hz_40ms_envelope_correlation": 0.9946762711254634, "1000–3000 Hz_40ms_modulation_mae_db": 1.1452524521383642, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7840749781987477, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20149073046732024, "3000–6000 Hz_10ms_envelope_correlation": 0.9747618135009416, "3000–6000 Hz_10ms_modulation_mae_db": 0.9764044358640226, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1367567144978294, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.1224787107289743, "3000–6000 Hz_40ms_envelope_correlation": 0.9907892938550387, "3000–6000 Hz_40ms_modulation_mae_db": 1.1806100060314426, "mr_stft_db": 6.002721294072482, "band_mae_db": 0.6067897633700001, "rms_gain_error_db": 0.8066017840020329, "output_peak": 0.6658339559928478}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.76148009471371, "spectral_convergence": 0.6527374437520252, "spectral_convergence_status": "ok", "amplitude_floor": 2.172633843844252e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.474020161411563, "spectral_convergence": 0.3178488463154496, "spectral_convergence_status": "ok", "amplitude_floor": 1.076658885105748e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.772663626092173, "spectral_convergence": 0.31151718882482227, "spectral_convergence_status": "ok", "amplitude_floor": 9.276051433709775e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0006687421878497317, "output_power": 0.0007595642641763849, "delta_db": 0.5530580225486001, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0031496712549154435, "output_power": 0.0038598216515606984, "delta_db": 0.8830201102954217, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0007488394009410867, "output_power": 0.0009092181046358572, "delta_db": 0.8427938777302278, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002473939509459554, "output_power": 0.00029001007253936833, "delta_db": 0.6902400549591986, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011342358064728732, "output_power": 0.00011982758762316644, "delta_db": 0.2385346283451677, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.620279276522665e-05, "output_power": 5.10478453910068e-05, "delta_db": 0.4330918863413845, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6067897633700001, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8066017840020329, "pitch": null, "output_rms_dbfs": -22.226101902184357, "output_peak": 0.6658339559928478, "reference_rms_dbfs": -23.03270368618639, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.002721294072482}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023158079660483383, "log_rms_mae_db": 1.683723203470434, "rms_envelope_nrmse": 0.16775747271496322, "envelope_correlation": 0.9837680060569017, "rms_gain_error_db": 0.8102887325784335, "reference_mean_power": 0.004973550834679421, "output_mean_power": 0.005993705974161057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.9210188012408756e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006213961937331268, "output_power": 0.0007454231276296727, "reference_active": true, "delta_db": 0.7903427440118361}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00026574930428793645, "output_power": 0.0002963791874962704, "reference_active": true, "delta_db": 0.47375566864067914}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2739241814293584e-05, "output_power": 2.6344872149678138e-05, "reference_active": true, "delta_db": -0.9437252235860994}], "active_band_count": 3, "active_band_mae_db": 0.7359412120795382}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000211279420280134, "log_rms_mae_db": 1.1578350486113893, "rms_envelope_nrmse": 0.12340131684371916, "envelope_correlation": 0.9947410507100278, "rms_gain_error_db": 0.8121361758122272, "reference_mean_power": 0.0049805093532051875, "output_mean_power": 0.006004645566687104}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.7395423077013874e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005644300449164411, "output_power": 0.0006732835486335715, "reference_active": true, "delta_db": 0.7658787938877272}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013986675109572895, "output_power": 0.0001572815381271245, "reference_active": true, "delta_db": 0.5096326071137687}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.497453327177472e-06, "output_power": 1.492143571262282e-06, "reference_active": true, "delta_db": -2.2368676865696036}], "active_band_count": 3, "active_band_mae_db": 1.1707930291903665}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020019856693346264, "log_rms_mae_db": 1.6824051327982348, "rms_envelope_nrmse": 0.19446051932908592, "envelope_correlation": 0.9828384417658632, "rms_gain_error_db": 0.9835731221782316, "reference_mean_power": 0.0028106313477950454, "output_mean_power": 0.003525016855445496}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3651110338311905e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00046335294017358356, "output_power": 0.0005719680313851686, "reference_active": true, "delta_db": 0.9145983259825619}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00024122957535154213, "output_power": 0.0002791897087232211, "reference_active": true, "delta_db": 0.6346885329009759}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.948878321904196e-05, "output_power": 1.7107698582009938e-05, "reference_active": true, "delta_db": -2.3646526296237638}], "active_band_count": 3, "active_band_mae_db": 1.3046464961691004}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018155803316515472, "log_rms_mae_db": 1.1668549116829519, "rms_envelope_nrmse": 0.14999933083105796, "envelope_correlation": 0.9936359153335587, "rms_gain_error_db": 0.9846258920467095, "reference_mean_power": 0.002819303754025505, "output_mean_power": 0.0035367507869056747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2479023732657914e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004464610508434935, "output_power": 0.0005408617452982075, "reference_active": true, "delta_db": 0.8330268811128881}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001291081626943261, "output_power": 0.0001498881449997221, "reference_active": true, "delta_db": 0.6481358401721504}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4462923050468237e-06, "output_power": 1.2531207501357237e-06, "reference_active": true, "delta_db": -2.9051542779287542}], "active_band_count": 3, "active_band_mae_db": 1.4621056664045977}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.07685175719321e-05, "log_rms_mae_db": 1.7256042888667515, "rms_envelope_nrmse": 0.1742786887369514, "envelope_correlation": 0.983036166267795, "rms_gain_error_db": 0.6389363937571045, "reference_mean_power": 0.0002950078039052426, "output_mean_power": 0.0003417646531411952}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.385626430084388e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.894665895613538e-05, "output_power": 0.0001151699364844035, "reference_active": true, "delta_db": 0.6593799297174936}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.27901607459427e-05, "output_power": 2.413885932194097e-05, "reference_active": true, "delta_db": 0.24969355325333634}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.304247232293777e-06, "output_power": 2.918160189335087e-06, "reference_active": true, "delta_db": -1.6878808052909255}], "active_band_count": 3, "active_band_mae_db": 0.8656514294205852}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.03385602019198e-05, "log_rms_mae_db": 1.111647207665029, "rms_envelope_nrmse": 0.11412189888398126, "envelope_correlation": 0.9946762711254634, "rms_gain_error_db": 0.6357563836119627, "reference_mean_power": 0.0002923716436469955, "output_mean_power": 0.00033846275690657386}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2489747353423383e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.236418785147827e-05, "output_power": 0.0001065497427189184, "reference_active": true, "delta_db": 0.6204878949754861}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2735133445304299e-05, "output_power": 1.329222863834041e-05, "reference_active": true, "delta_db": 0.18594302843024768}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6865666311394477e-07, "output_power": 1.4664422778312071e-07, "reference_active": true, "delta_db": -2.629326433009359}], "active_band_count": 3, "active_band_mae_db": 1.1452524521383642}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 4.524569706134418e-05, "log_rms_mae_db": 1.7840749781987477, "rms_envelope_nrmse": 0.20149073046732024, "envelope_correlation": 0.9747618135009416, "rms_gain_error_db": 0.30753870073301653, "reference_mean_power": 6.736853190004857e-05, "output_mean_power": 7.231209656722776e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.630234817861519e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0381998821210514e-05, "output_power": 2.4233235423568915e-05, "reference_active": true, "delta_db": 0.7516462932259865}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.955680541944287e-06, "output_power": 4.591928827932049e-06, "reference_active": true, "delta_db": 0.647739371292398}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.594012590914479e-07, "output_power": 9.378481722810654e-07, "reference_active": true, "delta_db": 1.5298276430736832}], "active_band_count": 3, "active_band_mae_db": 0.9764044358640226}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.8717196486707034e-05, "log_rms_mae_db": 1.1367567144978294, "rms_envelope_nrmse": 0.1224787107289743, "envelope_correlation": 0.9907892938550387, "rms_gain_error_db": 0.30885446390945237, "reference_mean_power": 6.764325505404543e-05, "output_mean_power": 7.262897991274646e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.430902640057432e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.9875215171126693e-05, "output_power": 2.358356884221753e-05, "reference_active": true, "delta_db": 0.7429768735424253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.8203988319997446e-06, "output_power": 2.1100348444438443e-06, "reference_active": true, "delta_db": 0.6412307897437381}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.685144573008789e-08, "output_power": 6.056430891395242e-08, "reference_active": true, "delta_db": 2.1576223548081646}], "active_band_count": 3, "active_band_mae_db": 1.1806100060314426}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8823078886515939, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a5b3924900a58d764a10c7aa709c8ebbca9d51fc10514aa134f2d3afc1a16608", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.983036166267795, "standard_deviation": null, "minimum": 0.983036166267795, "maximum": 0.983036166267795, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7256042888667515, "standard_deviation": null, "minimum": 1.7256042888667515, "maximum": 1.7256042888667515, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8656514294205852, "standard_deviation": null, "minimum": 0.8656514294205852, "maximum": 0.8656514294205852, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1742786887369514, "standard_deviation": null, "minimum": 0.1742786887369514, "maximum": 0.1742786887369514, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946762711254634, "standard_deviation": null, "minimum": 0.9946762711254634, "maximum": 0.9946762711254634, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.111647207665029, "standard_deviation": null, "minimum": 1.111647207665029, "maximum": 1.111647207665029, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1452524521383642, "standard_deviation": null, "minimum": 1.1452524521383642, "maximum": 1.1452524521383642, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11412189888398126, "standard_deviation": null, "minimum": 0.11412189888398126, "maximum": 0.11412189888398126, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9747618135009416, "standard_deviation": null, "minimum": 0.9747618135009416, "maximum": 0.9747618135009416, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7840749781987477, "standard_deviation": null, "minimum": 1.7840749781987477, "maximum": 1.7840749781987477, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9764044358640226, "standard_deviation": null, "minimum": 0.9764044358640226, "maximum": 0.9764044358640226, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20149073046732024, "standard_deviation": null, "minimum": 0.20149073046732024, "maximum": 0.20149073046732024, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9907892938550387, "standard_deviation": null, "minimum": 0.9907892938550387, "maximum": 0.9907892938550387, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1367567144978294, "standard_deviation": null, "minimum": 1.1367567144978294, "maximum": 1.1367567144978294, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.1806100060314426, "standard_deviation": null, "minimum": 1.1806100060314426, "maximum": 1.1806100060314426, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1224787107289743, "standard_deviation": null, "minimum": 0.1224787107289743, "maximum": 0.1224787107289743, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9828384417658632, "standard_deviation": null, "minimum": 0.9828384417658632, "maximum": 0.9828384417658632, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.6824051327982348, "standard_deviation": null, "minimum": 1.6824051327982348, "maximum": 1.6824051327982348, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3046464961691004, "standard_deviation": null, "minimum": 1.3046464961691004, "maximum": 1.3046464961691004, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.19446051932908592, "standard_deviation": null, "minimum": 0.19446051932908592, "maximum": 0.19446051932908592, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9936359153335587, "standard_deviation": null, "minimum": 0.9936359153335587, "maximum": 0.9936359153335587, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.1668549116829519, "standard_deviation": null, "minimum": 1.1668549116829519, "maximum": 1.1668549116829519, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.4621056664045977, "standard_deviation": null, "minimum": 1.4621056664045977, "maximum": 1.4621056664045977, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14999933083105796, "standard_deviation": null, "minimum": 0.14999933083105796, "maximum": 0.14999933083105796, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6067897633700001, "standard_deviation": null, "minimum": 0.6067897633700001, "maximum": 0.6067897633700001, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.002721294072482, "standard_deviation": null, "minimum": 6.002721294072482, "maximum": 6.002721294072482, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6658339559928478, "standard_deviation": null, "minimum": 0.6658339559928478, "maximum": 0.6658339559928478, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8066017840020329, "standard_deviation": null, "minimum": 0.8066017840020329, "maximum": 0.8066017840020329, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9837680060569017, "standard_deviation": null, "minimum": 0.9837680060569017, "maximum": 0.9837680060569017, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.683723203470434, "standard_deviation": null, "minimum": 1.683723203470434, "maximum": 1.683723203470434, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7359412120795382, "standard_deviation": null, "minimum": 0.7359412120795382, "maximum": 0.7359412120795382, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16775747271496322, "standard_deviation": null, "minimum": 0.16775747271496322, "maximum": 0.16775747271496322, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9947410507100278, "standard_deviation": null, "minimum": 0.9947410507100278, "maximum": 0.9947410507100278, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.1578350486113893, "standard_deviation": null, "minimum": 1.1578350486113893, "maximum": 1.1578350486113893, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.1707930291903665, "standard_deviation": null, "minimum": 1.1707930291903665, "maximum": 1.1707930291903665, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12340131684371916, "standard_deviation": null, "minimum": 0.12340131684371916, "maximum": 0.12340131684371916, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_7729"}, {"case": "librispeech_8224-274381-0010", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9537792205810547, "predicted_mos": 2.9537792205810547, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "6528d2893c602c19815a8c3d8b86d416d7da079d998224a8d75d8e0963f01adc", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.741473900154233}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.8319160971349353, "全帯域_10ms_rms_envelope_nrmse": 0.201052937070744, "全帯域_10ms_envelope_correlation": 0.9776522642885185, "全帯域_10ms_modulation_mae_db": 0.8770829659082814, "全帯域_40ms_log_rms_mae_db": 2.3235164652318385, "全帯域_40ms_rms_envelope_nrmse": 0.15532638044517622, "全帯域_40ms_envelope_correlation": 0.9904694000429302, "全帯域_40ms_modulation_mae_db": 0.920198369071971, "300–1000 Hz_10ms_log_rms_mae_db": 2.2048595127714252, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.26333059900122124, "300–1000 Hz_10ms_envelope_correlation": 0.9803890625079897, "300–1000 Hz_10ms_modulation_mae_db": 1.5220920573048542, "300–1000 Hz_40ms_log_rms_mae_db": 1.6505714963649762, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.22944855816462562, "300–1000 Hz_40ms_envelope_correlation": 0.9902903972819874, "300–1000 Hz_40ms_modulation_mae_db": 1.5185671827818348, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6906153417240728, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20926854187013483, "1000–3000 Hz_10ms_envelope_correlation": 0.9784319744154764, "1000–3000 Hz_10ms_modulation_mae_db": 1.0555202880002885, "1000–3000 Hz_40ms_log_rms_mae_db": 1.165680578709219, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.14446226471244508, "1000–3000 Hz_40ms_envelope_correlation": 0.9941183603322851, "1000–3000 Hz_40ms_modulation_mae_db": 1.1630764347032077, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7469228945363806, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2121884763637945, "3000–6000 Hz_10ms_envelope_correlation": 0.9737186564441751, "3000–6000 Hz_10ms_modulation_mae_db": 0.12750828849463344, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0844278967454075, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11419998709925515, "3000–6000 Hz_40ms_envelope_correlation": 0.9926929170917761, "3000–6000 Hz_40ms_modulation_mae_db": 0.2576375586136619, "mr_stft_db": 5.962885655193323, "band_mae_db": 0.7770708733567617, "rms_gain_error_db": 0.9056786639329478, "output_peak": 0.6652808282001511}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.79521931724402, "spectral_convergence": 0.6647203384169174, "spectral_convergence_status": "ok", "amplitude_floor": 1.642409573690628e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.405950839786369, "spectral_convergence": 0.2916263607773621, "spectral_convergence_status": "ok", "amplitude_floor": 1.497573454010957e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.687486808549579, "spectral_convergence": 0.30656088231098283, "spectral_convergence_status": "ok", "amplitude_floor": 1.2731241516974033e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0012546591027172457, "output_power": 0.001211210480719566, "delta_db": -0.15306121653379037, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0029841932734943235, "output_power": 0.003981049094355825, "delta_db": 1.2517058622676691, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0004877853564370709, "output_power": 0.0006590006468104781, "delta_db": 1.3065708242157168, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013442175178747268, "output_power": 0.00017052070167406508, "delta_db": 1.0330756021545828, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00010124396480278944, "output_power": 0.00012003651336549146, "delta_db": 0.7394422804986647, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001060495609474016, "output_power": 0.00010177753526388272, "delta_db": -0.17856945447014733, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7770708733567617, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.9056786639329478, "pitch": null, "output_rms_dbfs": -22.045652777826696, "output_peak": 0.6652808282001511, "reference_rms_dbfs": -22.951331441759642, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.962885655193323}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024147266762860616, "log_rms_mae_db": 2.8319160971349353, "rms_envelope_nrmse": 0.201052937070744, "envelope_correlation": 0.9776522642885185, "rms_gain_error_db": 0.905687879882643, "reference_mean_power": 0.005072567967533502, "output_mean_power": 0.006248800547672776}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.4164286663529534e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010865583816620668, "output_power": 0.0014412072773613011, "reference_active": true, "delta_db": 1.226733801437761}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020269542962956985, "output_power": 0.000252948533326971, "reference_active": true, "delta_db": 0.9618820923499843}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.69603101954121e-05, "output_power": 2.985301403373734e-05, "reference_active": true, "delta_db": 0.4426330039370987}], "active_band_count": 3, "active_band_mae_db": 0.8770829659082814}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.000225157536036158, "log_rms_mae_db": 2.3235164652318385, "rms_envelope_nrmse": 0.15532638044517622, "envelope_correlation": 0.9904694000429302, "rms_gain_error_db": 0.9057180225417711, "reference_mean_power": 0.005098028148503288, "output_mean_power": 0.00628020805139523}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.2603245746946083e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010232290392639372, "output_power": 0.001368720089079462, "reference_active": true, "delta_db": 1.263417846286036}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010544090587602268, "output_power": 0.00012792479686758542, "reference_active": true, "delta_db": 0.8394560759975366}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0049541105469593e-06, "output_power": 2.33279489055264e-06, "reference_active": true, "delta_db": 0.6577211849323408}], "active_band_count": 3, "active_band_mae_db": 0.920198369071971}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00020992712290116374, "log_rms_mae_db": 2.2048595127714252, "rms_envelope_nrmse": 0.26333059900122124, "envelope_correlation": 0.9803890625079897, "rms_gain_error_db": 1.5340662875048032, "reference_mean_power": 0.0023388664575994473, "output_mean_power": 0.0033297532718121192}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4073975659228857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007410848450397675, "output_power": 0.0010773550214508282, "reference_active": true, "delta_db": 1.6249090803897674}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011531333502346333, "output_power": 0.00017740867584027377, "reference_active": true, "delta_db": 1.8709532165030112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6543285086796136e-05, "output_power": 2.0744992593353004e-05, "reference_active": true, "delta_db": -1.0704138750217846}], "active_band_count": 3, "active_band_mae_db": 1.5220920573048542}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002036893389343094, "log_rms_mae_db": 1.6505714963649762, "rms_envelope_nrmse": 0.22944855816462562, "envelope_correlation": 0.9902903972819874, "rms_gain_error_db": 1.5340645972780083, "reference_mean_power": 0.0023506239289763434, "output_mean_power": 0.003346490624420106}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.3312664615669961e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007215176046332657, "output_power": 0.001047915999692062, "reference_active": true, "delta_db": 1.6207953915629147}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5162124576630795e-05, "output_power": 8.096439859065101e-05, "reference_active": true, "delta_db": 1.6665310951936962}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1332065410684048e-06, "output_power": 1.5929252649078882e-06, "reference_active": true, "delta_db": -1.2683750615888931}], "active_band_count": 3, "active_band_mae_db": 1.5185671827818348}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 8.160365671559225e-05, "log_rms_mae_db": 1.6906153417240728, "rms_envelope_nrmse": 0.20926854187013483, "envelope_correlation": 0.9784319744154764, "rms_gain_error_db": 0.937735581423623, "reference_mean_power": 0.00021265544148097967, "output_mean_power": 0.0002639064824836291}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.181263558889223e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.569393841260949e-05, "output_power": 7.107408240256252e-05, "reference_active": true, "delta_db": 1.0590333123445401}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4402699100691371e-05, "output_power": 1.801827149139068e-05, "reference_active": true, "delta_db": 0.9726923883017407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.927958039351209e-06, "output_power": 3.8023173818583936e-06, "reference_active": true, "delta_db": 1.1348351633545848}], "active_band_count": 3, "active_band_mae_db": 1.0555202880002885}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.199537921297776e-05, "log_rms_mae_db": 1.165680578709219, "rms_envelope_nrmse": 0.14446226471244508, "envelope_correlation": 0.9941183603322851, "rms_gain_error_db": 0.9378201279448074, "reference_mean_power": 0.00021371687213348275, "output_mean_power": 0.00026522888661996663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0829999638899264e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.5369039282181144e-05, "output_power": 7.091490461933168e-05, "reference_active": true, "delta_db": 1.0747053504645743}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.495407348960159e-06, "output_power": 8.099326291945082e-06, "reference_active": true, "delta_db": 0.9584250310053222}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.109761653644897e-07, "output_power": 2.95014448780455e-07, "reference_active": true, "delta_db": 1.456098922639726}], "active_band_count": 3, "active_band_mae_db": 1.1630764347032077}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.3802525725767476e-05, "log_rms_mae_db": 1.7469228945363806, "rms_envelope_nrmse": 0.2121884763637945, "envelope_correlation": 0.9737186564441751, "rms_gain_error_db": 0.31920647594468987, "reference_mean_power": 6.940550281225073e-05, "output_mean_power": 7.469895949474797e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.994304097313729e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.68316376202847e-05, "output_power": 3.840432794012142e-05, "reference_active": true, "delta_db": 0.18159140512270244}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.3496621932801275e-06, "output_power": 5.230058328762576e-06, "reference_active": true, "delta_db": -0.09819826767110808}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1665818120524947e-06, "output_power": 1.1945070585523317e-06, "reference_active": true, "delta_db": 0.10273519269008977}], "active_band_count": 3, "active_band_mae_db": 0.12750828849463344}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.808436973199796e-05, "log_rms_mae_db": 1.0844278967454075, "rms_envelope_nrmse": 0.11419998709925515, "envelope_correlation": 0.9926929170917761, "rms_gain_error_db": 0.31921680828100124, "reference_mean_power": 6.97538862423984e-05, "output_mean_power": 7.507409223124229e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.7257017907786166e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.6073966740876416e-05, "output_power": 3.778458493454962e-05, "reference_active": true, "delta_db": 0.20120755297403592}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3003381054168396e-06, "output_power": 2.218961891635251e-06, "reference_active": true, "delta_db": -0.15641829897628992}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.80849015690717e-08, "output_power": 8.592029304079125e-08, "reference_active": true, "delta_db": 0.4152868238906599}], "active_band_count": 3, "active_band_mae_db": 0.2576375586136619}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8835289695767539, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7b12c3be2db393e646f1f850c9fb51cfbcdc580ccaba9f5e9df666b98fb74080", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9784319744154764, "standard_deviation": null, "minimum": 0.9784319744154764, "maximum": 0.9784319744154764, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6906153417240728, "standard_deviation": null, "minimum": 1.6906153417240728, "maximum": 1.6906153417240728, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.0555202880002885, "standard_deviation": null, "minimum": 1.0555202880002885, "maximum": 1.0555202880002885, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20926854187013483, "standard_deviation": null, "minimum": 0.20926854187013483, "maximum": 0.20926854187013483, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9941183603322851, "standard_deviation": null, "minimum": 0.9941183603322851, "maximum": 0.9941183603322851, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.165680578709219, "standard_deviation": null, "minimum": 1.165680578709219, "maximum": 1.165680578709219, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1630764347032077, "standard_deviation": null, "minimum": 1.1630764347032077, "maximum": 1.1630764347032077, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14446226471244508, "standard_deviation": null, "minimum": 0.14446226471244508, "maximum": 0.14446226471244508, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9737186564441751, "standard_deviation": null, "minimum": 0.9737186564441751, "maximum": 0.9737186564441751, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7469228945363806, "standard_deviation": null, "minimum": 1.7469228945363806, "maximum": 1.7469228945363806, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.12750828849463344, "standard_deviation": null, "minimum": 0.12750828849463344, "maximum": 0.12750828849463344, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2121884763637945, "standard_deviation": null, "minimum": 0.2121884763637945, "maximum": 0.2121884763637945, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9926929170917761, "standard_deviation": null, "minimum": 0.9926929170917761, "maximum": 0.9926929170917761, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0844278967454075, "standard_deviation": null, "minimum": 1.0844278967454075, "maximum": 1.0844278967454075, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.2576375586136619, "standard_deviation": null, "minimum": 0.2576375586136619, "maximum": 0.2576375586136619, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11419998709925515, "standard_deviation": null, "minimum": 0.11419998709925515, "maximum": 0.11419998709925515, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9803890625079897, "standard_deviation": null, "minimum": 0.9803890625079897, "maximum": 0.9803890625079897, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.2048595127714252, "standard_deviation": null, "minimum": 2.2048595127714252, "maximum": 2.2048595127714252, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.5220920573048542, "standard_deviation": null, "minimum": 1.5220920573048542, "maximum": 1.5220920573048542, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26333059900122124, "standard_deviation": null, "minimum": 0.26333059900122124, "maximum": 0.26333059900122124, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9902903972819874, "standard_deviation": null, "minimum": 0.9902903972819874, "maximum": 0.9902903972819874, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6505714963649762, "standard_deviation": null, "minimum": 1.6505714963649762, "maximum": 1.6505714963649762, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.5185671827818348, "standard_deviation": null, "minimum": 1.5185671827818348, "maximum": 1.5185671827818348, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.22944855816462562, "standard_deviation": null, "minimum": 0.22944855816462562, "maximum": 0.22944855816462562, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7770708733567617, "standard_deviation": null, "minimum": 0.7770708733567617, "maximum": 0.7770708733567617, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.962885655193323, "standard_deviation": null, "minimum": 5.962885655193323, "maximum": 5.962885655193323, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6652808282001511, "standard_deviation": null, "minimum": 0.6652808282001511, "maximum": 0.6652808282001511, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.9056786639329478, "standard_deviation": null, "minimum": 0.9056786639329478, "maximum": 0.9056786639329478, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9776522642885185, "standard_deviation": null, "minimum": 0.9776522642885185, "maximum": 0.9776522642885185, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.8319160971349353, "standard_deviation": null, "minimum": 2.8319160971349353, "maximum": 2.8319160971349353, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.8770829659082814, "standard_deviation": null, "minimum": 0.8770829659082814, "maximum": 0.8770829659082814, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.201052937070744, "standard_deviation": null, "minimum": 0.201052937070744, "maximum": 0.201052937070744, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9904694000429302, "standard_deviation": null, "minimum": 0.9904694000429302, "maximum": 0.9904694000429302, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.3235164652318385, "standard_deviation": null, "minimum": 2.3235164652318385, "maximum": 2.3235164652318385, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.920198369071971, "standard_deviation": null, "minimum": 0.920198369071971, "maximum": 0.920198369071971, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.15532638044517622, "standard_deviation": null, "minimum": 0.15532638044517622, "maximum": 0.15532638044517622, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8224"}, {"case": "librispeech_8224-274381-0013", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.826406478881836, "predicted_mos": 2.826406478881836, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "f00a9e1537fe278a4132f11cc5da2321aca65f21f1af84121a82cf4ecab7181e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.287586099933833}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.8217710493851986, "全帯域_10ms_rms_envelope_nrmse": 0.27366195945868604, "全帯域_10ms_envelope_correlation": 0.9708885250162833, "全帯域_10ms_modulation_mae_db": 1.711599025009604, "全帯域_40ms_log_rms_mae_db": 2.501116010805872, "全帯域_40ms_rms_envelope_nrmse": 0.2506773485543524, "全帯域_40ms_envelope_correlation": 0.979839212618042, "全帯域_40ms_modulation_mae_db": 1.499223614782359, "300–1000 Hz_10ms_log_rms_mae_db": 2.3804336929325283, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.31655151720728675, "300–1000 Hz_10ms_envelope_correlation": 0.9839304244127245, "300–1000 Hz_10ms_modulation_mae_db": 2.5506834015807223, "300–1000 Hz_40ms_log_rms_mae_db": 1.9894365605564033, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.29967340934066306, "300–1000 Hz_40ms_envelope_correlation": 0.9895961235510081, "300–1000 Hz_40ms_modulation_mae_db": 2.5445556764828012, "1000–3000 Hz_10ms_log_rms_mae_db": 1.567944787720355, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18138271771298545, "1000–3000 Hz_10ms_envelope_correlation": 0.9808362162221954, "1000–3000 Hz_10ms_modulation_mae_db": 0.9614049938193755, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9536830633457832, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11230806160256354, "1000–3000 Hz_40ms_envelope_correlation": 0.9946218279186113, "1000–3000 Hz_40ms_modulation_mae_db": 0.9582958462693695, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6954304018409938, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.15548487228436714, "3000–6000 Hz_10ms_envelope_correlation": 0.9862618746529078, "3000–6000 Hz_10ms_modulation_mae_db": 0.1995876885023915, "3000–6000 Hz_40ms_log_rms_mae_db": 1.052244396899177, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.08383491340873458, "3000–6000 Hz_40ms_envelope_correlation": 0.9959813799298064, "3000–6000 Hz_40ms_modulation_mae_db": 0.11164499499796247, "mr_stft_db": 5.856288987290405, "band_mae_db": 0.8850025473121857, "rms_gain_error_db": 1.4743871565951896, "output_peak": 0.6800981461495669}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.60638636388873, "spectral_convergence": 0.5696102621085177, "spectral_convergence_status": "ok", "amplitude_floor": 2.0096490980240517e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.3649265375673725, "spectral_convergence": 0.37669905612028726, "spectral_convergence_status": "ok", "amplitude_floor": 1.509105853911261e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.5975540604151135, "spectral_convergence": 0.3763665002379943, "spectral_convergence_status": "ok", "amplitude_floor": 1.2442618586634314e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0017656001839806658, "output_power": 0.0015268761734887248, "delta_db": -0.630885473113954, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.005544968065872631, "output_power": 0.00882927072278646, "delta_db": 2.020257840064069, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005978914392629673, "output_power": 0.0008902213413312314, "delta_db": 1.7287566629602489, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.426968976514186e-05, "output_power": 9.785048924550332e-05, "delta_db": 0.6489160635135574, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 6.40147911812626e-05, "output_power": 6.79533387635475e-05, "delta_db": 0.2593046671767545, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00024101533193812626, "output_power": 0.00023980333185283698, "delta_db": -0.021894577044529816, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8850025473121857, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.4743871565951896, "pitch": null, "output_rms_dbfs": -19.33600441033302, "output_peak": 0.6800981461495669, "reference_rms_dbfs": -20.810391566928214, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.856288987290405}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00026421160622171636, "log_rms_mae_db": 2.8217710493851986, "rms_envelope_nrmse": 0.27366195945868604, "envelope_correlation": 0.9708885250162833, "rms_gain_error_db": 1.474698951338259, "reference_mean_power": 0.008303652784602253, "output_mean_power": 0.011661088089650152}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.437173268209281e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017543654513432086, "output_power": 0.0029941798867075585, "reference_active": true, "delta_db": 2.3215782247044556}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013368683738807852, "output_power": 0.00020676901016237657, "reference_active": true, "delta_db": 1.8939679935320455}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.235741942938941e-05, "output_power": 8.941454179978213e-05, "reference_active": true, "delta_db": 0.9192508567923114}], "active_band_count": 3, "active_band_mae_db": 1.711599025009604}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022872947944797367, "log_rms_mae_db": 2.501116010805872, "rms_envelope_nrmse": 0.2506773485543524, "envelope_correlation": 0.979839212618042, "rms_gain_error_db": 1.4760677480616562, "reference_mean_power": 0.008338532782066458, "output_mean_power": 0.0117137625257536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.164744003117204e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015799393891106446, "output_power": 0.002756522844987037, "reference_active": true, "delta_db": 2.4172116956937484}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.374442374979845e-05, "output_power": 8.985630528408134e-05, "reference_active": true, "delta_db": 1.4910635783324881}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.293312028286094e-06, "output_power": 3.772006542786292e-06, "reference_active": true, "delta_db": 0.58939557032084}], "active_band_count": 3, "active_band_mae_db": 1.499223614782359}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00024243770094671305, "log_rms_mae_db": 2.3804336929325283, "rms_envelope_nrmse": 0.31655151720728675, "envelope_correlation": 0.9839304244127245, "rms_gain_error_db": 2.0722853136277513, "reference_mean_power": 0.003959461759548422, "output_mean_power": 0.006380646495981558}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.015177817247118e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001136879031892713, "output_power": 0.001954476371412011, "reference_active": true, "delta_db": 2.3531616779184774}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001232393146611943, "output_power": 0.00022447900853500696, "reference_active": true, "delta_db": 2.6042646100106333}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0364107060899577e-05, "output_power": 5.647066547954918e-05, "reference_active": true, "delta_db": 2.694623916813057}], "active_band_count": 3, "active_band_mae_db": 2.5506834015807223}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00020477863430336455, "log_rms_mae_db": 1.9894365605564033, "rms_envelope_nrmse": 0.29967340934066306, "envelope_correlation": 0.9895961235510081, "rms_gain_error_db": 2.0722789192056856, "reference_mean_power": 0.003979363820398493, "output_mean_power": 0.0064127090938144735}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.8942412803353574e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010558685473009428, "output_power": 0.0018293054229235298, "reference_active": true, "delta_db": 2.386763687257585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.091449836890923e-05, "output_power": 9.20625890448689e-05, "reference_active": true, "delta_db": 2.572417149731931}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8755256559270567e-06, "output_power": 3.471935341472423e-06, "reference_active": true, "delta_db": 2.6744861924588874}], "active_band_count": 3, "active_band_mae_db": 2.5445556764828012}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.5081112731014556e-05, "log_rms_mae_db": 1.567944787720355, "rms_envelope_nrmse": 0.18138271771298545, "envelope_correlation": 0.9808362162221954, "rms_gain_error_db": 0.618052764013428, "reference_mean_power": 0.00010911551831198975, "output_mean_power": 0.00012580323132868815}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 6.836309277153812e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.5256828161206645e-05, "output_power": 2.9057618049386863e-05, "reference_active": true, "delta_db": 0.608812013842444}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.88342602252987e-06, "output_power": 1.2557329243460778e-05, "reference_active": true, "delta_db": 1.0398976534700022}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.436572483663522e-06, "output_power": 4.090483557488742e-06, "reference_active": true, "delta_db": -1.2355053141456802}], "active_band_count": 3, "active_band_mae_db": 0.9614049938193755}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.4434083199144116e-05, "log_rms_mae_db": 0.9536830633457832, "rms_envelope_nrmse": 0.11230806160256354, "envelope_correlation": 0.9946218279186113, "rms_gain_error_db": 0.617810190726701, "reference_mean_power": 0.0001096329653054558, "output_mean_power": 0.00012639275489108748}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.313909407445067e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.3642314296732694e-05, "output_power": 2.7075287945451453e-05, "reference_active": true, "delta_db": 0.5888309757074967}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.77067085832153e-06, "output_power": 7.687880881793055e-06, "reference_active": true, "delta_db": 0.5517494361726234}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.651384635522461e-07, "output_power": 3.1199780690329576e-07, "reference_active": true, "delta_db": -1.7343071269279884}], "active_band_count": 3, "active_band_mae_db": 0.9582958462693695}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010577651466639849, "log_rms_mae_db": 1.6954304018409938, "rms_envelope_nrmse": 0.15548487228436714, "envelope_correlation": 0.9862618746529078, "rms_gain_error_db": -0.030405061430864445, "reference_mean_power": 0.00018020158938811219, "output_mean_power": 0.00017894439965714053}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.705326923064497e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.613613486254781e-05, "output_power": 1.7500184838889147e-05, "reference_active": true, "delta_db": 0.3524312081168589}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.942687703713047e-06, "output_power": 8.976856963612131e-06, "reference_active": true, "delta_db": 0.01656240633972337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.043858273628482e-06, "output_power": 9.900670169401077e-07, "reference_active": true, "delta_db": -0.2297694510505922}], "active_band_count": 3, "active_band_mae_db": 0.1995876885023915}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.602886913996206e-05, "log_rms_mae_db": 1.052244396899177, "rms_envelope_nrmse": 0.08383491340873458, "envelope_correlation": 0.9959813799298064, "rms_gain_error_db": -0.028983112181655397, "reference_mean_power": 0.0001775816262598491, "output_mean_power": 0.0001764004617599854}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.444034789789241e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7728340440048457e-05, "output_power": 1.9072643171493803e-05, "reference_active": true, "delta_db": 0.31742800720453107}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.955564320473026e-06, "output_power": 4.943477773858545e-06, "reference_active": true, "delta_db": -0.010605315258532061}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.260796671427193e-08, "output_power": 9.275525319539413e-08, "reference_active": true, "delta_db": 0.00690166253082426}], "active_band_count": 3, "active_band_mae_db": 0.11164499499796247}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8909324659603844, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "a09cabace0f8d9b1ba8a56aeeb2f27ff0a9671d0a4bbaa476628191d2f574a27", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9808362162221954, "standard_deviation": null, "minimum": 0.9808362162221954, "maximum": 0.9808362162221954, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.567944787720355, "standard_deviation": null, "minimum": 1.567944787720355, "maximum": 1.567944787720355, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9614049938193755, "standard_deviation": null, "minimum": 0.9614049938193755, "maximum": 0.9614049938193755, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18138271771298545, "standard_deviation": null, "minimum": 0.18138271771298545, "maximum": 0.18138271771298545, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9946218279186113, "standard_deviation": null, "minimum": 0.9946218279186113, "maximum": 0.9946218279186113, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9536830633457832, "standard_deviation": null, "minimum": 0.9536830633457832, "maximum": 0.9536830633457832, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9582958462693695, "standard_deviation": null, "minimum": 0.9582958462693695, "maximum": 0.9582958462693695, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11230806160256354, "standard_deviation": null, "minimum": 0.11230806160256354, "maximum": 0.11230806160256354, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9862618746529078, "standard_deviation": null, "minimum": 0.9862618746529078, "maximum": 0.9862618746529078, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6954304018409938, "standard_deviation": null, "minimum": 1.6954304018409938, "maximum": 1.6954304018409938, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.1995876885023915, "standard_deviation": null, "minimum": 0.1995876885023915, "maximum": 0.1995876885023915, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15548487228436714, "standard_deviation": null, "minimum": 0.15548487228436714, "maximum": 0.15548487228436714, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9959813799298064, "standard_deviation": null, "minimum": 0.9959813799298064, "maximum": 0.9959813799298064, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.052244396899177, "standard_deviation": null, "minimum": 1.052244396899177, "maximum": 1.052244396899177, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.11164499499796247, "standard_deviation": null, "minimum": 0.11164499499796247, "maximum": 0.11164499499796247, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08383491340873458, "standard_deviation": null, "minimum": 0.08383491340873458, "maximum": 0.08383491340873458, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9839304244127245, "standard_deviation": null, "minimum": 0.9839304244127245, "maximum": 0.9839304244127245, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.3804336929325283, "standard_deviation": null, "minimum": 2.3804336929325283, "maximum": 2.3804336929325283, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 2.5506834015807223, "standard_deviation": null, "minimum": 2.5506834015807223, "maximum": 2.5506834015807223, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.31655151720728675, "standard_deviation": null, "minimum": 0.31655151720728675, "maximum": 0.31655151720728675, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9895961235510081, "standard_deviation": null, "minimum": 0.9895961235510081, "maximum": 0.9895961235510081, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.9894365605564033, "standard_deviation": null, "minimum": 1.9894365605564033, "maximum": 1.9894365605564033, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 2.5445556764828012, "standard_deviation": null, "minimum": 2.5445556764828012, "maximum": 2.5445556764828012, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.29967340934066306, "standard_deviation": null, "minimum": 0.29967340934066306, "maximum": 0.29967340934066306, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8850025473121857, "standard_deviation": null, "minimum": 0.8850025473121857, "maximum": 0.8850025473121857, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.856288987290405, "standard_deviation": null, "minimum": 5.856288987290405, "maximum": 5.856288987290405, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6800981461495669, "standard_deviation": null, "minimum": 0.6800981461495669, "maximum": 0.6800981461495669, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.4743871565951896, "standard_deviation": null, "minimum": 1.4743871565951896, "maximum": 1.4743871565951896, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9708885250162833, "standard_deviation": null, "minimum": 0.9708885250162833, "maximum": 0.9708885250162833, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.8217710493851986, "standard_deviation": null, "minimum": 2.8217710493851986, "maximum": 2.8217710493851986, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.711599025009604, "standard_deviation": null, "minimum": 1.711599025009604, "maximum": 1.711599025009604, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.27366195945868604, "standard_deviation": null, "minimum": 0.27366195945868604, "maximum": 0.27366195945868604, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.979839212618042, "standard_deviation": null, "minimum": 0.979839212618042, "maximum": 0.979839212618042, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.501116010805872, "standard_deviation": null, "minimum": 2.501116010805872, "maximum": 2.501116010805872, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.499223614782359, "standard_deviation": null, "minimum": 1.499223614782359, "maximum": 1.499223614782359, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2506773485543524, "standard_deviation": null, "minimum": 0.2506773485543524, "maximum": 0.2506773485543524, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8224"}, {"case": "librispeech_8224-274384-0010", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.280862808227539, "predicted_mos": 3.280862808227539, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "97f4fac9aa21c397bbcf8668076295cf81b1a89ad1f378dec60aabf0f435d623", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.222847899887711}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.2047002621551672, "全帯域_10ms_rms_envelope_nrmse": 0.23665546560794443, "全帯域_10ms_envelope_correlation": 0.9818564062873011, "全帯域_10ms_modulation_mae_db": 1.906036188122614, "全帯域_40ms_log_rms_mae_db": 1.8313495215961113, "全帯域_40ms_rms_envelope_nrmse": 0.21101425688220046, "全帯域_40ms_envelope_correlation": 0.9907189922883355, "全帯域_40ms_modulation_mae_db": 1.9242439894349441, "300–1000 Hz_10ms_log_rms_mae_db": 2.100889643481308, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.3109655969544412, "300–1000 Hz_10ms_envelope_correlation": 0.9851761433515538, "300–1000 Hz_10ms_modulation_mae_db": 1.9440989004744178, "300–1000 Hz_40ms_log_rms_mae_db": 1.7237641811746616, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.29126595174434217, "300–1000 Hz_40ms_envelope_correlation": 0.9916361378212016, "300–1000 Hz_40ms_modulation_mae_db": 1.7686912329307434, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5235354074956906, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1880151979019074, "1000–3000 Hz_10ms_envelope_correlation": 0.9841806945830366, "1000–3000 Hz_10ms_modulation_mae_db": 0.511533529950758, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0557156983069929, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13224573094214648, "1000–3000 Hz_40ms_envelope_correlation": 0.99599478184487, "1000–3000 Hz_40ms_modulation_mae_db": 0.7363288487407829, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7622687513267303, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2670357189936135, "3000–6000 Hz_10ms_envelope_correlation": 0.9604474165890825, "3000–6000 Hz_10ms_modulation_mae_db": 1.854405580831339, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1244893262953766, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17368861756417894, "3000–6000 Hz_40ms_envelope_correlation": 0.9836626059895586, "3000–6000 Hz_40ms_modulation_mae_db": 1.976152925889794, "mr_stft_db": 5.7333472286970375, "band_mae_db": 1.020196977526353, "rms_gain_error_db": 1.4434154096581164, "output_peak": 0.46455676852500616}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.377063783257145, "spectral_convergence": 0.5579109753496361, "spectral_convergence_status": "ok", "amplitude_floor": 1.597965137573199e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.348589426346893, "spectral_convergence": 0.33304075586831045, "spectral_convergence_status": "ok", "amplitude_floor": 1.057274363712373e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.474388476487074, "spectral_convergence": 0.3503170410573812, "spectral_convergence_status": "ok", "amplitude_floor": 9.008040625007043e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007434014661781762, "output_power": 0.0007021194995700559, "delta_db": -0.24812378681945957, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.002448126060340392, "output_power": 0.0037551440091402556, "delta_db": 1.8579281978475395, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00039049712732524606, "output_power": 0.000593712747062275, "delta_db": 1.8195853000793738, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012090645510899839, "output_power": 0.0001499054780424907, "delta_db": 0.9336801554561185, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.957546973001399e-05, "output_power": 7.839875831650345e-05, "delta_db": 1.1924170908146448, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.142800821939945e-05, "output_power": 8.013626158051142e-05, "delta_db": -0.06944733414098272, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.020196977526353, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.4434154096581164, "pitch": null, "output_rms_dbfs": -22.7088247046149, "output_peak": 0.46455676852500616, "reference_rms_dbfs": -24.152240114273017, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.7333472286970375}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00019565573818943504, "log_rms_mae_db": 2.2047002621551672, "rms_envelope_nrmse": 0.23665546560794443, "envelope_correlation": 0.9818564062873011, "rms_gain_error_db": 1.4433728437530569, "reference_mean_power": 0.0038470996951946897, "output_mean_power": 0.0053637771434644835}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.743468430119212e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007829242029289011, "output_power": 0.0011599216500902626, "reference_active": true, "delta_db": 1.7070893583734883}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011656588263872551, "output_power": 0.00017977517486903542, "reference_active": true, "delta_db": 1.881582633549066}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.15404555717803e-05, "output_power": 5.150052060758545e-05, "reference_active": true, "delta_db": 2.129436572445288}], "active_band_count": 3, "active_band_mae_db": 1.906036188122614}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00018472557440126405, "log_rms_mae_db": 1.8313495215961113, "rms_envelope_nrmse": 0.21101425688220046, "envelope_correlation": 0.9907189922883355, "rms_gain_error_db": 1.4437094695213468, "reference_mean_power": 0.0038659686805529734, "output_mean_power": 0.005390502829021122}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.614770154461247e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007106100759473788, "output_power": 0.0010574996970057395, "reference_active": true, "delta_db": 1.7264889042826548}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.7305685337334634e-05, "output_power": 9.319198441808864e-05, "reference_active": true, "delta_db": 2.1118084887686877}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9649709160477995e-06, "output_power": 3.067606039940802e-06, "reference_active": true, "delta_db": 1.9344345752534893}], "active_band_count": 3, "active_band_mae_db": 1.9242439894349441}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00016965929826439602, "log_rms_mae_db": 2.100889643481308, "rms_envelope_nrmse": 0.3109655969544412, "envelope_correlation": 0.9851761433515538, "rms_gain_error_db": 2.053654777258363, "reference_mean_power": 0.0021633278035333714, "output_mean_power": 0.0034712653217319777}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3458386409803657e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007177587670080827, "output_power": 0.0011412971949304447, "reference_active": true, "delta_db": 2.014202440056867}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.048912255781678e-05, "output_power": 0.00014379318201165843, "reference_active": true, "delta_db": 2.0114191732962112}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.109541579796594e-05, "output_power": 4.71372104026417e-05, "reference_active": true, "delta_db": 1.806675088070175}], "active_band_count": 3, "active_band_mae_db": 1.9440989004744178}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015157137475553504, "log_rms_mae_db": 1.7237641811746616, "rms_envelope_nrmse": 0.29126595174434217, "envelope_correlation": 0.9916361378212016, "rms_gain_error_db": 2.053617505363915, "reference_mean_power": 0.002174161455071234, "output_mean_power": 0.003488619006495358}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2731603714689836e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006867133707128568, "output_power": 0.0010874890099030498, "reference_active": true, "delta_db": 1.9964937325027612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.067232594302272e-05, "output_power": 6.414156759920124e-05, "reference_active": true, "delta_db": 1.9784056046855591}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.08557261679534e-06, "output_power": 2.833627457637285e-06, "reference_active": true, "delta_db": 1.3311743616039096}], "active_band_count": 3, "active_band_mae_db": 1.7686912329307434}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 5.465903254113391e-05, "log_rms_mae_db": 1.5235354074956906, "rms_envelope_nrmse": 0.1880151979019074, "envelope_correlation": 0.9841806945830366, "rms_gain_error_db": 0.8965880917620659, "reference_mean_power": 0.00014077271984820836, "output_mean_power": 0.0001730522741094816}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.622290255773331e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.707421130487826e-05, "output_power": 4.496199877686936e-05, "reference_active": true, "delta_db": 0.8377368859498623}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.08772682349557e-06, "output_power": 5.7534744346894936e-06, "reference_active": true, "delta_db": 0.5340640307438818}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8011808387082025e-06, "output_power": 1.869981394542611e-06, "reference_active": true, "delta_db": 0.16279967315852983}], "active_band_count": 3, "active_band_mae_db": 0.511533529950758}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.107059424790359e-05, "log_rms_mae_db": 1.0557156983069929, "rms_envelope_nrmse": 0.13224573094214648, "envelope_correlation": 0.99599478184487, "rms_gain_error_db": 0.8969506296323986, "reference_mean_power": 0.00014139647469838974, "output_mean_power": 0.00017383356815826843}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.069193023932203e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.382591725297068e-05, "output_power": 4.126225427160248e-05, "reference_active": true, "delta_db": 0.8630336886360669}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.1244897853414024e-06, "output_power": 2.3450911930164417e-06, "reference_active": true, "delta_db": 0.4290508836825654}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.400225448356044e-07, "output_power": 1.133725119068775e-07, "reference_active": true, "delta_db": -0.9169019739037163}], "active_band_count": 3, "active_band_mae_db": 0.7363288487407829}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.428054041295646e-05, "log_rms_mae_db": 1.7622687513267303, "rms_envelope_nrmse": 0.2670357189936135, "envelope_correlation": 0.9604474165890825, "rms_gain_error_db": 0.48138096395047186, "reference_mean_power": 8.952225188868405e-05, "output_mean_power": 0.00010001591086661383}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.314971416941356e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.877588546016404e-05, "output_power": 6.563641347412585e-05, "reference_active": true, "delta_db": 0.4794566166821987}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.04956306951581e-06, "output_power": 1.378316863906624e-05, "reference_active": true, "delta_db": 1.8272145871065564}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.5294615410314917e-06, "output_power": 5.35405261169236e-06, "reference_active": true, "delta_db": 3.2565455387052626}], "active_band_count": 3, "active_band_mae_db": 1.854405580831339}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.5452516433588795e-05, "log_rms_mae_db": 1.1244893262953766, "rms_envelope_nrmse": 0.17368861756417894, "envelope_correlation": 0.9836626059895586, "rms_gain_error_db": 0.48135863304013876, "reference_mean_power": 8.996899751083782e-05, "output_mean_power": 0.0001005145064799291}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.951283845645369e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.884212028117595e-05, "output_power": 6.859759475786422e-05, "reference_active": true, "delta_db": 0.6662057483604183}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.436071463262977e-06, "output_power": 6.891702654473578e-06, "reference_active": true, "delta_db": 1.9132799751813143}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7711408251558852e-07, "output_power": 3.8295734111045847e-07, "reference_active": true, "delta_db": 3.34897305412765}], "active_band_count": 3, "active_band_mae_db": 1.976152925889794}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9237629648743544, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "0fe6174d4d2d2c5bf1ec55af38c3e7ba8d9681322f301d9c4cdefd6c7b89d4a5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9841806945830366, "standard_deviation": null, "minimum": 0.9841806945830366, "maximum": 0.9841806945830366, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5235354074956906, "standard_deviation": null, "minimum": 1.5235354074956906, "maximum": 1.5235354074956906, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.511533529950758, "standard_deviation": null, "minimum": 0.511533529950758, "maximum": 0.511533529950758, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1880151979019074, "standard_deviation": null, "minimum": 0.1880151979019074, "maximum": 0.1880151979019074, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.99599478184487, "standard_deviation": null, "minimum": 0.99599478184487, "maximum": 0.99599478184487, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0557156983069929, "standard_deviation": null, "minimum": 1.0557156983069929, "maximum": 1.0557156983069929, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.7363288487407829, "standard_deviation": null, "minimum": 0.7363288487407829, "maximum": 0.7363288487407829, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13224573094214648, "standard_deviation": null, "minimum": 0.13224573094214648, "maximum": 0.13224573094214648, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9604474165890825, "standard_deviation": null, "minimum": 0.9604474165890825, "maximum": 0.9604474165890825, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7622687513267303, "standard_deviation": null, "minimum": 1.7622687513267303, "maximum": 1.7622687513267303, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.854405580831339, "standard_deviation": null, "minimum": 1.854405580831339, "maximum": 1.854405580831339, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2670357189936135, "standard_deviation": null, "minimum": 0.2670357189936135, "maximum": 0.2670357189936135, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9836626059895586, "standard_deviation": null, "minimum": 0.9836626059895586, "maximum": 0.9836626059895586, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1244893262953766, "standard_deviation": null, "minimum": 1.1244893262953766, "maximum": 1.1244893262953766, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.976152925889794, "standard_deviation": null, "minimum": 1.976152925889794, "maximum": 1.976152925889794, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17368861756417894, "standard_deviation": null, "minimum": 0.17368861756417894, "maximum": 0.17368861756417894, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9851761433515538, "standard_deviation": null, "minimum": 0.9851761433515538, "maximum": 0.9851761433515538, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.100889643481308, "standard_deviation": null, "minimum": 2.100889643481308, "maximum": 2.100889643481308, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.9440989004744178, "standard_deviation": null, "minimum": 1.9440989004744178, "maximum": 1.9440989004744178, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3109655969544412, "standard_deviation": null, "minimum": 0.3109655969544412, "maximum": 0.3109655969544412, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9916361378212016, "standard_deviation": null, "minimum": 0.9916361378212016, "maximum": 0.9916361378212016, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7237641811746616, "standard_deviation": null, "minimum": 1.7237641811746616, "maximum": 1.7237641811746616, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.7686912329307434, "standard_deviation": null, "minimum": 1.7686912329307434, "maximum": 1.7686912329307434, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.29126595174434217, "standard_deviation": null, "minimum": 0.29126595174434217, "maximum": 0.29126595174434217, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.020196977526353, "standard_deviation": null, "minimum": 1.020196977526353, "maximum": 1.020196977526353, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.7333472286970375, "standard_deviation": null, "minimum": 5.7333472286970375, "maximum": 5.7333472286970375, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.46455676852500616, "standard_deviation": null, "minimum": 0.46455676852500616, "maximum": 0.46455676852500616, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.4434154096581164, "standard_deviation": null, "minimum": 1.4434154096581164, "maximum": 1.4434154096581164, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9818564062873011, "standard_deviation": null, "minimum": 0.9818564062873011, "maximum": 0.9818564062873011, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.2047002621551672, "standard_deviation": null, "minimum": 2.2047002621551672, "maximum": 2.2047002621551672, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.906036188122614, "standard_deviation": null, "minimum": 1.906036188122614, "maximum": 1.906036188122614, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.23665546560794443, "standard_deviation": null, "minimum": 0.23665546560794443, "maximum": 0.23665546560794443, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9907189922883355, "standard_deviation": null, "minimum": 0.9907189922883355, "maximum": 0.9907189922883355, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.8313495215961113, "standard_deviation": null, "minimum": 1.8313495215961113, "maximum": 1.8313495215961113, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.9242439894349441, "standard_deviation": null, "minimum": 1.9242439894349441, "maximum": 1.9242439894349441, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.21101425688220046, "standard_deviation": null, "minimum": 0.21101425688220046, "maximum": 0.21101425688220046, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8224"}, {"case": "librispeech_8230-279154-0018", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.1220879554748535, "predicted_mos": 3.1220879554748535, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "ed6f3ab0ee5bd4c63b5a98a5751d9b3b1d3b6f4001bad2440e595f5ceb1fb69b", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.939187599811703}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.3186777074781104, "全帯域_10ms_rms_envelope_nrmse": 0.16382597402688795, "全帯域_10ms_envelope_correlation": 0.9838834840690979, "全帯域_10ms_modulation_mae_db": 0.7648044857704778, "全帯域_40ms_log_rms_mae_db": 1.7408358182829347, "全帯域_40ms_rms_envelope_nrmse": 0.12439106857872376, "全帯域_40ms_envelope_correlation": 0.9937682986395849, "全帯域_40ms_modulation_mae_db": 0.9716732484345748, "300–1000 Hz_10ms_log_rms_mae_db": 2.060453580461512, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.22061886463093847, "300–1000 Hz_10ms_envelope_correlation": 0.9870059237005536, "300–1000 Hz_10ms_modulation_mae_db": 0.9144456152883311, "300–1000 Hz_40ms_log_rms_mae_db": 1.5539693585386658, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.1919649390465119, "300–1000 Hz_40ms_envelope_correlation": 0.9942707307162046, "300–1000 Hz_40ms_modulation_mae_db": 1.2648120005897834, "1000–3000 Hz_10ms_log_rms_mae_db": 1.622514464884628, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1978970289150519, "1000–3000 Hz_10ms_envelope_correlation": 0.9756762547816191, "1000–3000 Hz_10ms_modulation_mae_db": 0.43825332837200975, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9876171712076833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11063618823988755, "1000–3000 Hz_40ms_envelope_correlation": 0.9951429419986519, "1000–3000 Hz_40ms_modulation_mae_db": 0.6005505948528793, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8402740606247703, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20709424316672445, "3000–6000 Hz_10ms_envelope_correlation": 0.9752478170418504, "3000–6000 Hz_10ms_modulation_mae_db": 0.9330780889107816, "3000–6000 Hz_40ms_log_rms_mae_db": 1.160367954082952, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13199225537658416, "3000–6000 Hz_40ms_envelope_correlation": 0.9898983225392048, "3000–6000 Hz_40ms_modulation_mae_db": 0.9989188958068652, "mr_stft_db": 5.829727916452735, "band_mae_db": 0.5861325325438403, "rms_gain_error_db": 0.7718709337439512, "output_peak": 0.662478991500702}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.699345403156156, "spectral_convergence": 0.6019827557408435, "spectral_convergence_status": "ok", "amplitude_floor": 1.9267471569597447e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.31802085475512, "spectral_convergence": 0.2722516158795406, "spectral_convergence_status": "ok", "amplitude_floor": 1.271916808390449e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.471817491446927, "spectral_convergence": 0.2678510000673518, "spectral_convergence_status": "ok", "amplitude_floor": 1.1644398478092437e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001373055391651405, "output_power": 0.0013580590062820814, "delta_db": -0.047694178431614484, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0033897235254818038, "output_power": 0.004306339884264175, "delta_db": 1.039440266973919, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0002841140646523882, "output_power": 0.0003728626474993854, "delta_db": 1.1805614557346062, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.671358771236327e-05, "output_power": 0.00010121440972981572, "delta_db": 0.6715519148276872, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6620686952686656e-05, "output_power": 3.0087555103490536e-05, "delta_db": 0.5316764031925146, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.533501052724457e-05, "output_power": 2.5604021705732017e-05, "delta_db": 0.04587097610270054, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5861325325438403, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7718709337439512, "pitch": null, "output_rms_dbfs": -22.080170530930758, "output_peak": 0.662478991500702, "reference_rms_dbfs": -22.85204146467471, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.829727916452735}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023735521522065317, "log_rms_mae_db": 2.3186777074781104, "rms_envelope_nrmse": 0.16382597402688795, "envelope_correlation": 0.9838834840690979, "rms_gain_error_db": 0.7719177631347134, "reference_mean_power": 0.0051897925018650195, "output_mean_power": 0.006199287396809079}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6580095170174236e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00099786392963893, "output_power": 0.0012879688016467373, "reference_active": true, "delta_db": 1.108340190574188}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00025981544310873085, "output_power": 0.0002615456626159833, "reference_active": true, "delta_db": 0.0288256080038477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.167033379867013e-05, "output_power": 3.1922890887743944e-05, "reference_active": true, "delta_db": -1.1572476587333975}], "active_band_count": 3, "active_band_mae_db": 0.7648044857704778}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022053657198928817, "log_rms_mae_db": 1.7408358182829347, "rms_envelope_nrmse": 0.12439106857872376, "envelope_correlation": 0.9937682986395849, "rms_gain_error_db": 0.7717709249898761, "reference_mean_power": 0.0052155458817555326, "output_mean_power": 0.006229839568331458}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.3916979728964525e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008985624617991317, "output_power": 0.0011710967225779297, "reference_active": true, "delta_db": 1.1504449394568612}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001435836821660598, "output_power": 0.00013566441687678198, "reference_active": true, "delta_db": -0.2463913406269866}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.3776224720397365e-06, "output_power": 1.676195036500778e-06, "reference_active": true, "delta_db": -1.5181834652198765}], "active_band_count": 3, "active_band_mae_db": 0.9716732484345748}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00018677846007938983, "log_rms_mae_db": 2.060453580461512, "rms_envelope_nrmse": 0.22061886463093847, "envelope_correlation": 0.9870059237005536, "rms_gain_error_db": 1.343550914994174, "reference_mean_power": 0.0023293656178310037, "output_mean_power": 0.0031738964411509216}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.5191576623571145e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006490500325829311, "output_power": 0.000889581361316049, "reference_active": true, "delta_db": 1.3690749885335263}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00013228128531435308, "output_power": 0.00015858357589114603, "reference_active": true, "delta_db": 0.7875980040975584}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.716163060824064e-05, "output_power": 1.4993125094877321e-05, "reference_active": true, "delta_db": -0.5866638532339085}], "active_band_count": 3, "active_band_mae_db": 0.9144456152883311}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00017927232752442356, "log_rms_mae_db": 1.5539693585386658, "rms_envelope_nrmse": 0.1919649390465119, "envelope_correlation": 0.9942707307162046, "rms_gain_error_db": 1.342924503565916, "reference_mean_power": 0.002341010543733861, "output_mean_power": 0.003189303284200073}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.424681480871225e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006148042140063018, "output_power": 0.0008332785325415808, "reference_active": true, "delta_db": 1.3205335780560041}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.176867672883116e-05, "output_power": 7.260589674041736e-05, "reference_active": true, "delta_db": 0.7020359613194317}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.094779276972058e-06, "output_power": 7.280143198173375e-07, "reference_active": true, "delta_db": -1.7718664623939142}], "active_band_count": 3, "active_band_mae_db": 1.2648120005897834}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.042588377582574e-05, "log_rms_mae_db": 1.622514464884628, "rms_envelope_nrmse": 0.1978970289150519, "envelope_correlation": 0.9756762547816191, "rms_gain_error_db": 0.6573239822962498, "reference_mean_power": 9.196537764000522e-05, "output_mean_power": 0.00010699334288618796}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3593984658429913e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.0478584079151807e-05, "output_power": 2.337516271565805e-05, "reference_active": true, "delta_db": 0.5745471714797111}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.015542002555929e-06, "output_power": 6.587381038672524e-06, "reference_active": true, "delta_db": 0.39438021847691024}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6245635384567205e-06, "output_power": 1.5002147884498026e-06, "reference_active": true, "delta_db": -0.3458325951594078}], "active_band_count": 3, "active_band_mae_db": 0.43825332837200975}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.6202941099492215e-05, "log_rms_mae_db": 0.9876171712076833, "rms_envelope_nrmse": 0.11063618823988755, "envelope_correlation": 0.9951429419986519, "rms_gain_error_db": 0.6574110532871842, "reference_mean_power": 9.242233001846982e-05, "output_mean_power": 0.00010752712114235185}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9023164132494484e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.934413521065944e-05, "output_power": 2.1948453614474197e-05, "reference_active": true, "delta_db": 0.5485460815834294}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.243887754162173e-06, "output_power": 3.509074024515668e-06, "reference_active": true, "delta_db": 0.3412671159846493}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3065359453687705e-07, "output_power": 1.0591013066024393e-07, "reference_active": true, "delta_db": -0.9118385869905592}], "active_band_count": 3, "active_band_mae_db": 0.6005505948528793}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.535256840646245e-05, "log_rms_mae_db": 1.8402740606247703, "rms_envelope_nrmse": 0.20709424316672445, "envelope_correlation": 0.9752478170418504, "rms_gain_error_db": 0.28688705704582573, "reference_mean_power": 2.9383590737924492e-05, "output_mean_power": 3.139016309032394e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.113292932890967e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.65042891211228e-06, "output_power": 1.0688481865359641e-05, "reference_active": true, "delta_db": 0.4436940879306879}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.01875786446688e-06, "output_power": 4.8779974749007176e-06, "reference_active": true, "delta_db": 0.8414973100763546}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.253460098772285e-07, "output_power": 1.0278967980522954e-06, "reference_active": true, "delta_db": 1.5140428687253025}], "active_band_count": 3, "active_band_mae_db": 0.9330780889107816}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.9034899521347454e-05, "log_rms_mae_db": 1.160367954082952, "rms_envelope_nrmse": 0.13199225537658416, "envelope_correlation": 0.9898983225392048, "rms_gain_error_db": 0.2868811984463524, "reference_mean_power": 2.9530847534057242e-05, "output_mean_power": 3.154743333011198e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9758537203809177e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.950101722919118e-06, "output_power": 1.1323867787097097e-05, "reference_active": true, "delta_db": 0.5616726938272225}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.0870738801319347e-06, "output_power": 2.586347109715747e-06, "reference_active": true, "delta_db": 0.9314898757141561}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.171675766600786e-08, "output_power": 8.724941300592552e-08, "reference_active": true, "delta_db": 1.503594117879217}], "active_band_count": 3, "active_band_mae_db": 0.9989188958068652}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9265732292065638, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "cde39b97fb82f1521c07def9238fbd4205e9d988b32d967ee7cce6c87dc9725a", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9756762547816191, "standard_deviation": null, "minimum": 0.9756762547816191, "maximum": 0.9756762547816191, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.622514464884628, "standard_deviation": null, "minimum": 1.622514464884628, "maximum": 1.622514464884628, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.43825332837200975, "standard_deviation": null, "minimum": 0.43825332837200975, "maximum": 0.43825332837200975, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1978970289150519, "standard_deviation": null, "minimum": 0.1978970289150519, "maximum": 0.1978970289150519, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9951429419986519, "standard_deviation": null, "minimum": 0.9951429419986519, "maximum": 0.9951429419986519, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9876171712076833, "standard_deviation": null, "minimum": 0.9876171712076833, "maximum": 0.9876171712076833, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6005505948528793, "standard_deviation": null, "minimum": 0.6005505948528793, "maximum": 0.6005505948528793, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11063618823988755, "standard_deviation": null, "minimum": 0.11063618823988755, "maximum": 0.11063618823988755, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9752478170418504, "standard_deviation": null, "minimum": 0.9752478170418504, "maximum": 0.9752478170418504, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8402740606247703, "standard_deviation": null, "minimum": 1.8402740606247703, "maximum": 1.8402740606247703, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.9330780889107816, "standard_deviation": null, "minimum": 0.9330780889107816, "maximum": 0.9330780889107816, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20709424316672445, "standard_deviation": null, "minimum": 0.20709424316672445, "maximum": 0.20709424316672445, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9898983225392048, "standard_deviation": null, "minimum": 0.9898983225392048, "maximum": 0.9898983225392048, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.160367954082952, "standard_deviation": null, "minimum": 1.160367954082952, "maximum": 1.160367954082952, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.9989188958068652, "standard_deviation": null, "minimum": 0.9989188958068652, "maximum": 0.9989188958068652, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13199225537658416, "standard_deviation": null, "minimum": 0.13199225537658416, "maximum": 0.13199225537658416, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9870059237005536, "standard_deviation": null, "minimum": 0.9870059237005536, "maximum": 0.9870059237005536, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.060453580461512, "standard_deviation": null, "minimum": 2.060453580461512, "maximum": 2.060453580461512, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9144456152883311, "standard_deviation": null, "minimum": 0.9144456152883311, "maximum": 0.9144456152883311, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22061886463093847, "standard_deviation": null, "minimum": 0.22061886463093847, "maximum": 0.22061886463093847, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9942707307162046, "standard_deviation": null, "minimum": 0.9942707307162046, "maximum": 0.9942707307162046, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.5539693585386658, "standard_deviation": null, "minimum": 1.5539693585386658, "maximum": 1.5539693585386658, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.2648120005897834, "standard_deviation": null, "minimum": 1.2648120005897834, "maximum": 1.2648120005897834, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.1919649390465119, "standard_deviation": null, "minimum": 0.1919649390465119, "maximum": 0.1919649390465119, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5861325325438403, "standard_deviation": null, "minimum": 0.5861325325438403, "maximum": 0.5861325325438403, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.829727916452735, "standard_deviation": null, "minimum": 5.829727916452735, "maximum": 5.829727916452735, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.662478991500702, "standard_deviation": null, "minimum": 0.662478991500702, "maximum": 0.662478991500702, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7718709337439512, "standard_deviation": null, "minimum": 0.7718709337439512, "maximum": 0.7718709337439512, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9838834840690979, "standard_deviation": null, "minimum": 0.9838834840690979, "maximum": 0.9838834840690979, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.3186777074781104, "standard_deviation": null, "minimum": 2.3186777074781104, "maximum": 2.3186777074781104, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7648044857704778, "standard_deviation": null, "minimum": 0.7648044857704778, "maximum": 0.7648044857704778, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.16382597402688795, "standard_deviation": null, "minimum": 0.16382597402688795, "maximum": 0.16382597402688795, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9937682986395849, "standard_deviation": null, "minimum": 0.9937682986395849, "maximum": 0.9937682986395849, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.7408358182829347, "standard_deviation": null, "minimum": 1.7408358182829347, "maximum": 1.7408358182829347, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9716732484345748, "standard_deviation": null, "minimum": 0.9716732484345748, "maximum": 0.9716732484345748, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.12439106857872376, "standard_deviation": null, "minimum": 0.12439106857872376, "maximum": 0.12439106857872376, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8230"}, {"case": "librispeech_8230-279154-0032", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.9454026222229004, "predicted_mos": 2.9454026222229004, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 62080, "input_sample_rate": 16000, "input_seconds": 3.88, "input_float32_sha256": "481318497ae49a95404c9e227aebc216367ea07165f300d0d86784e4e279ab50", "internal_samples": 62080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.833313399925828}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.5831390668176994, "全帯域_10ms_rms_envelope_nrmse": 0.20230689727336898, "全帯域_10ms_envelope_correlation": 0.9848468648475821, "全帯域_10ms_modulation_mae_db": 0.4738061678902646, "全帯域_40ms_log_rms_mae_db": 2.067808856070089, "全帯域_40ms_rms_envelope_nrmse": 0.17423834957834317, "全帯域_40ms_envelope_correlation": 0.9919559243051319, "全帯域_40ms_modulation_mae_db": 0.7121325872388827, "300–1000 Hz_10ms_log_rms_mae_db": 2.043938237397909, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.26645762382461224, "300–1000 Hz_10ms_envelope_correlation": 0.9863117363905703, "300–1000 Hz_10ms_modulation_mae_db": 1.2729598643822722, "300–1000 Hz_40ms_log_rms_mae_db": 1.4452506910216802, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.24436980399278851, "300–1000 Hz_40ms_envelope_correlation": 0.9923133403731909, "300–1000 Hz_40ms_modulation_mae_db": 1.3340996066005442, "1000–3000 Hz_10ms_log_rms_mae_db": 1.517041295834021, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.13525206689973998, "1000–3000 Hz_10ms_envelope_correlation": 0.992312260217533, "1000–3000 Hz_10ms_modulation_mae_db": 0.5293601435606649, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8840590929394287, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10050466559804377, "1000–3000 Hz_40ms_envelope_correlation": 0.9974494825392479, "1000–3000 Hz_40ms_modulation_mae_db": 0.6494917848664121, "3000–6000 Hz_10ms_log_rms_mae_db": 1.970204402383078, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3237793703796474, "3000–6000 Hz_10ms_envelope_correlation": 0.9404751444794677, "3000–6000 Hz_10ms_modulation_mae_db": 1.7387175030464768, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1921278109513103, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2132758246210347, "3000–6000 Hz_40ms_envelope_correlation": 0.9757494327629044, "3000–6000 Hz_40ms_modulation_mae_db": 1.8629533365101711, "mr_stft_db": 5.766269208146407, "band_mae_db": 0.6563898307374396, "rms_gain_error_db": 1.0891160835823213, "output_peak": 0.8028650976410031}, "full_scale": {"sample_count": 62080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 62080, "output_samples": 62080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.6329692470362, "spectral_convergence": 0.5563457152824055, "spectral_convergence_status": "ok", "amplitude_floor": 2.580715109285093e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.241541778158783, "spectral_convergence": 0.28807034241645463, "spectral_convergence_status": "ok", "amplitude_floor": 1.9335078654426385e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.424296599244235, "spectral_convergence": 0.29643182959015346, "spectral_convergence_status": "ok", "amplitude_floor": 1.5577782358148932e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0007926248856121452, "output_power": 0.0007781591287639895, "delta_db": -0.07999286893162649, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.003328913174136213, "output_power": 0.00463139976274194, "delta_db": 1.4340980117991808, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00045950159700284816, "output_power": 0.0005039758862859409, "delta_db": 0.4012273215194314, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00012582167228939335, "output_power": 0.00014631392051850449, "delta_db": 0.6553019442458806, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 3.488237987418478e-05, "output_power": 4.0553725187844714e-05, "delta_db": 0.6542464653339559, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.708544414240077e-05, "output_power": 7.906340911569365e-05, "delta_db": 0.7134723725945623, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6563898307374396, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.0891160835823213, "pitch": null, "output_rms_dbfs": -22.090490421328806, "output_peak": 0.8028650976410031, "reference_rms_dbfs": -23.179606504911124, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.766269208146407}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 62080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.00032928880743725145, "log_rms_mae_db": 2.5831390668176994, "rms_envelope_nrmse": 0.20230689727336898, "envelope_correlation": 0.9848468648475821, "rms_gain_error_db": 1.089116973019508, "reference_mean_power": 0.004815032927514685, "output_mean_power": 0.006187439403142572}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.902918899290217e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008077922828075312, "output_power": 0.0009635333753931985, "reference_active": true, "delta_db": 0.7656706273159765}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00023728667544934464, "output_power": 0.00026434504789634605, "reference_active": true, "delta_db": 0.4689782735608555}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.1018322593991373e-05, "output_power": 2.0133584099467346e-05, "reference_active": true, "delta_db": -0.1867696027939619}], "active_band_count": 3, "active_band_mae_db": 0.4738061678902646}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0003041508818572066, "log_rms_mae_db": 2.067808856070089, "rms_envelope_nrmse": 0.17423834957834317, "envelope_correlation": 0.9919559243051319, "rms_gain_error_db": 1.0891318399106846, "reference_mean_power": 0.004852582666148238, "output_mean_power": 0.006235713115640536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.7242336092637014e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007426996521093883, "output_power": 0.0008768868116753079, "reference_active": true, "delta_db": 0.7213031787722944}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00010890208813537604, "output_power": 0.00013297533952693746, "reference_active": true, "delta_db": 0.8673490069701337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4730800018618504e-06, "output_power": 1.298529367571497e-06, "reference_active": true, "delta_db": -0.5477455759742198}], "active_band_count": 3, "active_band_mae_db": 0.7121325872388827}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 0.0002965570632263201, "log_rms_mae_db": 2.043938237397909, "rms_envelope_nrmse": 0.26645762382461224, "envelope_correlation": 0.9863117363905703, "rms_gain_error_db": 1.6649051629421383, "reference_mean_power": 0.002818219366063788, "output_mean_power": 0.004134902859687831}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 1.2520173458640518e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00044507178368220307, "output_power": 0.0005939540695761323, "reference_active": true, "delta_db": 1.2532280023126146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00017793142616898364, "output_power": 0.00022084664051853623, "reference_active": true, "delta_db": 0.9383813747515137}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.6359305994808858e-05, "output_power": 1.1247030016204634e-05, "reference_active": true, "delta_db": -1.6272702160826884}], "active_band_count": 3, "active_band_mae_db": 1.2729598643822722}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 0.0002772405007925501, "log_rms_mae_db": 1.4452506910216802, "rms_envelope_nrmse": 0.24436980399278851, "envelope_correlation": 0.9923133403731909, "rms_gain_error_db": 1.6649043825195058, "reference_mean_power": 0.0028402077631144488, "output_mean_power": 0.004167163575996238}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 1.1541839932451843e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134314303572302, "output_power": 0.0005400678002147662, "reference_active": true, "delta_db": 1.1604479467941187}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.103778868784845e-05, "output_power": 0.0001094320631456317, "reference_active": true, "delta_db": 1.3045700540609984}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.3787553243482739e-06, "output_power": 9.677409545591165e-07, "reference_active": true, "delta_db": -1.5372808189465155}], "active_band_count": 3, "active_band_mae_db": 1.3340996066005442}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 6.247605344635176e-05, "log_rms_mae_db": 1.517041295834021, "rms_envelope_nrmse": 0.13525206689973998, "envelope_correlation": 0.992312260217533, "rms_gain_error_db": 0.660562120260191, "reference_mean_power": 0.00014490237542137544, "output_mean_power": 0.00016870646171726546}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 6.946266094979904e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.0294766814481675e-05, "output_power": 3.496304048637779e-05, "reference_active": true, "delta_db": 0.6224157911486455}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.606273878647975e-06, "output_power": 8.559126766312554e-06, "reference_active": true, "delta_db": -0.02385703967929483}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.077953085521739e-06, "output_power": 8.677989093880479e-07, "reference_active": true, "delta_db": -0.9418075998540542}], "active_band_count": 3, "active_band_mae_db": 0.5293601435606649}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 5.585752960561645e-05, "log_rms_mae_db": 0.8840590929394287, "rms_envelope_nrmse": 0.10050466559804377, "envelope_correlation": 0.9974494825392479, "rms_gain_error_db": 0.6605622732231359, "reference_mean_power": 0.00014603285132237434, "output_mean_power": 0.00017002265447455295}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 6.45401857377675e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.9246286179300554e-05, "output_power": 3.332416888303666e-05, "reference_active": true, "delta_db": 0.5668860144166513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6077467608233e-06, "output_power": 3.711524650789787e-06, "reference_active": true, "delta_db": 0.12316303912894935}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.605305892606176e-08, "output_power": 6.440560774609524e-08, "reference_active": true, "delta_db": -1.2584263010536358}], "active_band_count": 3, "active_band_mae_db": 0.6494917848664121}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 775, "rms_floor": 3.703669227124254e-05, "log_rms_mae_db": 1.970204402383078, "rms_envelope_nrmse": 0.3237793703796474, "envelope_correlation": 0.9404751444794677, "rms_gain_error_db": 0.7682853610252921, "reference_mean_power": 5.2354216813681856e-05, "output_mean_power": 6.248563725562522e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 775, "frequency_resolution_hz": 0.25806451612903225, "power_floor": 3.452458222743894e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.744222722260824e-05, "output_power": 2.172056162033708e-05, "reference_active": true, "delta_db": 0.9526911063723695}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.482818619951907e-06, "output_power": 1.0193518769107774e-05, "reference_active": true, "delta_db": 1.3425890865858503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0100258421135412e-06, "output_power": 1.978881175569656e-06, "reference_active": true, "delta_db": 2.9208723161812107}], "active_band_count": 3, "active_band_mae_db": 1.7387175030464768}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 62080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 769, "rms_floor": 2.970153096248883e-05, "log_rms_mae_db": 1.1921278109513103, "rms_envelope_nrmse": 0.2132758246210347, "envelope_correlation": 0.9757494327629044, "rms_gain_error_db": 0.7682856761120894, "reference_mean_power": 5.276267962923645e-05, "output_mean_power": 6.297314905646949e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 769, "frequency_resolution_hz": 0.2600780234070221, "power_floor": 3.130411452597279e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.7491368881094015e-05, "output_power": 2.2612131887260617e-05, "reference_active": true, "delta_db": 1.1151771099860233}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.132236890482271e-06, "output_power": 5.719328839520787e-06, "reference_active": true, "delta_db": 1.4115985699155933}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.718357196867276e-08, "output_power": 1.3597889991019546e-07, "reference_active": true, "delta_db": 3.0620843296288975}], "active_band_count": 3, "active_band_mae_db": 1.8629533365101711}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9082183611663129, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "df09c66acf0546201d618c466d0d63b241b6c0c6dd6d3480872c0fa7c16b370d", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.992312260217533, "standard_deviation": null, "minimum": 0.992312260217533, "maximum": 0.992312260217533, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.517041295834021, "standard_deviation": null, "minimum": 1.517041295834021, "maximum": 1.517041295834021, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5293601435606649, "standard_deviation": null, "minimum": 0.5293601435606649, "maximum": 0.5293601435606649, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.13525206689973998, "standard_deviation": null, "minimum": 0.13525206689973998, "maximum": 0.13525206689973998, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9974494825392479, "standard_deviation": null, "minimum": 0.9974494825392479, "maximum": 0.9974494825392479, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8840590929394287, "standard_deviation": null, "minimum": 0.8840590929394287, "maximum": 0.8840590929394287, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6494917848664121, "standard_deviation": null, "minimum": 0.6494917848664121, "maximum": 0.6494917848664121, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10050466559804377, "standard_deviation": null, "minimum": 0.10050466559804377, "maximum": 0.10050466559804377, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9404751444794677, "standard_deviation": null, "minimum": 0.9404751444794677, "maximum": 0.9404751444794677, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.970204402383078, "standard_deviation": null, "minimum": 1.970204402383078, "maximum": 1.970204402383078, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7387175030464768, "standard_deviation": null, "minimum": 1.7387175030464768, "maximum": 1.7387175030464768, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3237793703796474, "standard_deviation": null, "minimum": 0.3237793703796474, "maximum": 0.3237793703796474, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9757494327629044, "standard_deviation": null, "minimum": 0.9757494327629044, "maximum": 0.9757494327629044, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1921278109513103, "standard_deviation": null, "minimum": 1.1921278109513103, "maximum": 1.1921278109513103, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.8629533365101711, "standard_deviation": null, "minimum": 1.8629533365101711, "maximum": 1.8629533365101711, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2132758246210347, "standard_deviation": null, "minimum": 0.2132758246210347, "maximum": 0.2132758246210347, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9863117363905703, "standard_deviation": null, "minimum": 0.9863117363905703, "maximum": 0.9863117363905703, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.043938237397909, "standard_deviation": null, "minimum": 2.043938237397909, "maximum": 2.043938237397909, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.2729598643822722, "standard_deviation": null, "minimum": 1.2729598643822722, "maximum": 1.2729598643822722, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26645762382461224, "standard_deviation": null, "minimum": 0.26645762382461224, "maximum": 0.26645762382461224, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9923133403731909, "standard_deviation": null, "minimum": 0.9923133403731909, "maximum": 0.9923133403731909, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4452506910216802, "standard_deviation": null, "minimum": 1.4452506910216802, "maximum": 1.4452506910216802, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3340996066005442, "standard_deviation": null, "minimum": 1.3340996066005442, "maximum": 1.3340996066005442, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.24436980399278851, "standard_deviation": null, "minimum": 0.24436980399278851, "maximum": 0.24436980399278851, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6563898307374396, "standard_deviation": null, "minimum": 0.6563898307374396, "maximum": 0.6563898307374396, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.766269208146407, "standard_deviation": null, "minimum": 5.766269208146407, "maximum": 5.766269208146407, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.8028650976410031, "standard_deviation": null, "minimum": 0.8028650976410031, "maximum": 0.8028650976410031, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.0891160835823213, "standard_deviation": null, "minimum": 1.0891160835823213, "maximum": 1.0891160835823213, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9848468648475821, "standard_deviation": null, "minimum": 0.9848468648475821, "maximum": 0.9848468648475821, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.5831390668176994, "standard_deviation": null, "minimum": 2.5831390668176994, "maximum": 2.5831390668176994, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.4738061678902646, "standard_deviation": null, "minimum": 0.4738061678902646, "maximum": 0.4738061678902646, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20230689727336898, "standard_deviation": null, "minimum": 0.20230689727336898, "maximum": 0.20230689727336898, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9919559243051319, "standard_deviation": null, "minimum": 0.9919559243051319, "maximum": 0.9919559243051319, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 2.067808856070089, "standard_deviation": null, "minimum": 2.067808856070089, "maximum": 2.067808856070089, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.7121325872388827, "standard_deviation": null, "minimum": 0.7121325872388827, "maximum": 0.7121325872388827, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.17423834957834317, "standard_deviation": null, "minimum": 0.17423834957834317, "maximum": 0.17423834957834317, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8230"}, {"case": "librispeech_8230-279154-0033", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.416494846343994, "predicted_mos": 3.416494846343994, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "42580971b8fbc3a1b017c899bc2978ce202379569b078f77c1520286e38d57f9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.8887652000412345}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.094450566708371, "全帯域_10ms_rms_envelope_nrmse": 0.15791147178829054, "全帯域_10ms_envelope_correlation": 0.9768023747879895, "全帯域_10ms_modulation_mae_db": 0.1061671434034049, "全帯域_40ms_log_rms_mae_db": 1.424234715042344, "全帯域_40ms_rms_envelope_nrmse": 0.07863643574353049, "全帯域_40ms_envelope_correlation": 0.9955717616441553, "全帯域_40ms_modulation_mae_db": 0.26971612300328146, "300–1000 Hz_10ms_log_rms_mae_db": 1.9816943516995644, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18465854669513537, "300–1000 Hz_10ms_envelope_correlation": 0.9754691158661338, "300–1000 Hz_10ms_modulation_mae_db": 0.43826145080589024, "300–1000 Hz_40ms_log_rms_mae_db": 1.578371998659191, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.10738720957888152, "300–1000 Hz_40ms_envelope_correlation": 0.9933709333698885, "300–1000 Hz_40ms_modulation_mae_db": 0.771556337255546, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4995352133353437, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.1876759447873549, "1000–3000 Hz_10ms_envelope_correlation": 0.9764320943416785, "1000–3000 Hz_10ms_modulation_mae_db": 0.8355032125347543, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0305282183751228, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.09410540979269044, "1000–3000 Hz_40ms_envelope_correlation": 0.9956284517022947, "1000–3000 Hz_40ms_modulation_mae_db": 0.9939580138751486, "3000–6000 Hz_10ms_log_rms_mae_db": 1.8172974159271145, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.2258404609298131, "3000–6000 Hz_10ms_envelope_correlation": 0.9662856823378818, "3000–6000 Hz_10ms_modulation_mae_db": 0.29632906089777683, "3000–6000 Hz_40ms_log_rms_mae_db": 1.1098975560204496, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14128813905721518, "3000–6000 Hz_40ms_envelope_correlation": 0.9870504941066685, "3000–6000 Hz_40ms_modulation_mae_db": 0.47458109830409984, "mr_stft_db": 5.95523823673635, "band_mae_db": 0.4173500200643881, "rms_gain_error_db": 0.39925821053619615, "output_peak": 1.063525851640955}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 6, "at_or_above_full_scale_fraction": 6.25e-05, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 6.25e-05, "longest_full_scale_run_samples": 3, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 7.171231274073988, "spectral_convergence": 0.7519602971100211, "spectral_convergence_status": "ok", "amplitude_floor": 2.5091999427961322e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.1733057303561845, "spectral_convergence": 0.26621110628411093, "spectral_convergence_status": "ok", "amplitude_floor": 1.1904358876245626e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.52117770577888, "spectral_convergence": 0.2766406959358577, "spectral_convergence_status": "ok", "amplitude_floor": 8.234850781853941e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.001792776939061309, "output_power": 0.0018116457934691126, "delta_db": 0.04547032806498066, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0038916933601920887, "output_power": 0.004430411827133885, "delta_db": 0.5630548436755509, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002833146281043118, "output_power": 0.0030771756791773074, "delta_db": 0.35883292684068063, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00044617096434630785, "output_power": 0.000506057186706771, "delta_db": 0.5469829271128, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 5.1707528478122786e-05, "output_power": 5.682251221397221e-05, "delta_db": 0.40966650848800185, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 3.2019368729027054e-05, "output_power": 3.6595022966296736e-05, "delta_db": 0.5800925862043148, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.4173500200643881, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.39925821053619615, "pitch": null, "output_rms_dbfs": -20.035448939347816, "output_peak": 1.063525851640955, "reference_rms_dbfs": -20.434707149884012, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.95523823673635}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0003179454955178215, "log_rms_mae_db": 2.094450566708371, "rms_envelope_nrmse": 0.15791147178829054, "envelope_correlation": 0.9768023747879895, "rms_gain_error_db": 0.39925930113391883, "reference_mean_power": 0.009055057750136907, "output_mean_power": 0.009926980174981571}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.4928013782154505e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.002203466496672066, "output_power": 0.0023110141836606684, "reference_active": true, "delta_db": 0.2069615941070863}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0002920043469136947, "output_power": 0.00029192635139710165, "reference_active": true, "delta_db": -0.0011601726975130585}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.410659541472349e-05, "output_power": 7.601421726293768e-05, "reference_active": true, "delta_db": 0.11037966340561536}], "active_band_count": 3, "active_band_mae_db": 0.1061671434034049}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002752533875730688, "log_rms_mae_db": 1.424234715042344, "rms_envelope_nrmse": 0.07863643574353049, "envelope_correlation": 0.9955717616441553, "rms_gain_error_db": 0.39919658321584484, "reference_mean_power": 0.009100511091100074, "output_mean_power": 0.009976666195518827}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.115339465742665e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001970497325835552, "output_power": 0.0020821950251581323, "reference_active": true, "delta_db": 0.23945554692866206}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014191960918623396, "output_power": 0.00014190759252944717, "reference_active": true, "delta_db": -0.0003677426057119772}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.976197244793298e-06, "output_power": 3.4876702121526095e-06, "reference_active": true, "delta_db": -0.5693250794754703}], "active_band_count": 3, "active_band_mae_db": 0.26971612300328146}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028904791382782896, "log_rms_mae_db": 1.9816943516995644, "rms_envelope_nrmse": 0.18465854669513537, "envelope_correlation": 0.9754691158661338, "rms_gain_error_db": 0.5271343220175604, "reference_mean_power": 0.00500094100859156, "output_mean_power": 0.00564631578849136}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 3.3504006393160654e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.001701857204603173, "output_power": 0.0018402593538099355, "reference_active": true, "delta_db": 0.3395591634807748}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00029604441935142547, "output_power": 0.0002804069315946413, "reference_active": true, "delta_db": -0.23568133647578465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.422561443844994e-05, "output_power": 6.260361560076372e-05, "reference_active": true, "delta_db": -0.7395438524611113}], "active_band_count": 3, "active_band_mae_db": 0.43826145080589024}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002500254339817177, "log_rms_mae_db": 1.578371998659191, "rms_envelope_nrmse": 0.10738720957888152, "envelope_correlation": 0.9933709333698885, "rms_gain_error_db": 0.5270900076724756, "reference_mean_power": 0.00502600851498449, "output_mean_power": 0.005674560371220585}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.0757929049004627e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0015562264230646289, "output_power": 0.0016838877522311938, "reference_active": true, "delta_db": 0.3424035320807812}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001439985744956168, "output_power": 0.00013414985906161982, "reference_active": true, "delta_db": -0.3076797211030057}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.7762361828128385e-06, "output_power": 3.255571354222975e-06, "reference_active": true, "delta_db": -1.664585758582851}], "active_band_count": 3, "active_band_mae_db": 0.771556337255546}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00010495143398294122, "log_rms_mae_db": 1.4995352133353437, "rms_envelope_nrmse": 0.1876759447873549, "envelope_correlation": 0.9764320943416785, "rms_gain_error_db": 0.4837479728927338, "reference_mean_power": 0.0004146271168721916, "output_mean_power": 0.00046348160167399607}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.0723455289073609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010117430747443104, "output_power": 0.00011239931082996126, "reference_active": true, "delta_db": 0.45693408007081737}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2721317302684375e-05, "output_power": 1.3491654181575606e-05, "reference_active": true, "delta_db": 0.25533115627132846}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6903995233241656e-06, "output_power": 4.066691401630496e-06, "reference_active": true, "delta_db": 1.794244401262117}], "active_band_count": 3, "active_band_mae_db": 0.8355032125347543}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.780703381807892e-05, "log_rms_mae_db": 1.0305282183751228, "rms_envelope_nrmse": 0.09410540979269044, "envelope_correlation": 0.9956284517022947, "rms_gain_error_db": 0.48372424172644196, "reference_mean_power": 0.0004167096352122121, "output_mean_power": 0.00046580695266758233}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.9169450940862202e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 9.13104680727751e-05, "output_power": 0.00010145527918656351, "reference_active": true, "delta_db": 0.457540812514555}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.750575155696485e-06, "output_power": 6.4027692313029555e-06, "reference_active": true, "delta_db": 0.466565655335364}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0115145688994735e-07, "output_power": 3.2307246089381195e-07, "reference_active": true, "delta_db": 2.0577675737755268}], "active_band_count": 3, "active_band_mae_db": 0.9939580138751486}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 2.4613391249073974e-05, "log_rms_mae_db": 1.8172974159271145, "rms_envelope_nrmse": 0.2258404609298131, "envelope_correlation": 0.9662856823378818, "rms_gain_error_db": 0.4752830302576928, "reference_mean_power": 2.714804754964509e-05, "output_mean_power": 3.028774248987207e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3627974020623764e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.8436688435136e-06, "output_power": 5.05221942548214e-06, "reference_active": true, "delta_db": 0.18307761120683758}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.21772084491833e-06, "output_power": 1.9903322296227693e-06, "reference_active": true, "delta_db": -0.4698130290892355}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.650596314175208e-07, "output_power": 4.4045240817312094e-07, "reference_active": true, "delta_db": -0.23609654239725741}], "active_band_count": 3, "active_band_mae_db": 0.29632906089777683}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 2.2268327401512992e-05, "log_rms_mae_db": 1.1098975560204496, "rms_envelope_nrmse": 0.14128813905721518, "envelope_correlation": 0.9870504941066685, "rms_gain_error_db": 0.4752740756715307, "reference_mean_power": 2.7284479980131438e-05, "output_mean_power": 3.04398906845905e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.2818152654967464e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.7409144698248785e-06, "output_power": 4.8816835566937855e-06, "reference_active": true, "delta_db": 0.1270750356266035}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1748833712187593e-06, "output_power": 1.0761099026607273e-06, "reference_active": true, "delta_db": -0.3813812912135852}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.424838593381526e-08, "output_power": 2.7740316510812613e-08, "reference_active": true, "delta_db": -0.9152869680721107}], "active_band_count": 3, "active_band_mae_db": 0.47458109830409984}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9016164563730352, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "e0426eb106abfe4e41ff2458866a95584a4ce5f95f5fc50f94b14576fbcf6bd3", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9764320943416785, "standard_deviation": null, "minimum": 0.9764320943416785, "maximum": 0.9764320943416785, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4995352133353437, "standard_deviation": null, "minimum": 1.4995352133353437, "maximum": 1.4995352133353437, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.8355032125347543, "standard_deviation": null, "minimum": 0.8355032125347543, "maximum": 0.8355032125347543, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.1876759447873549, "standard_deviation": null, "minimum": 0.1876759447873549, "maximum": 0.1876759447873549, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9956284517022947, "standard_deviation": null, "minimum": 0.9956284517022947, "maximum": 0.9956284517022947, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0305282183751228, "standard_deviation": null, "minimum": 1.0305282183751228, "maximum": 1.0305282183751228, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9939580138751486, "standard_deviation": null, "minimum": 0.9939580138751486, "maximum": 0.9939580138751486, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.09410540979269044, "standard_deviation": null, "minimum": 0.09410540979269044, "maximum": 0.09410540979269044, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9662856823378818, "standard_deviation": null, "minimum": 0.9662856823378818, "maximum": 0.9662856823378818, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.8172974159271145, "standard_deviation": null, "minimum": 1.8172974159271145, "maximum": 1.8172974159271145, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.29632906089777683, "standard_deviation": null, "minimum": 0.29632906089777683, "maximum": 0.29632906089777683, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2258404609298131, "standard_deviation": null, "minimum": 0.2258404609298131, "maximum": 0.2258404609298131, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9870504941066685, "standard_deviation": null, "minimum": 0.9870504941066685, "maximum": 0.9870504941066685, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.1098975560204496, "standard_deviation": null, "minimum": 1.1098975560204496, "maximum": 1.1098975560204496, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.47458109830409984, "standard_deviation": null, "minimum": 0.47458109830409984, "maximum": 0.47458109830409984, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14128813905721518, "standard_deviation": null, "minimum": 0.14128813905721518, "maximum": 0.14128813905721518, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9754691158661338, "standard_deviation": null, "minimum": 0.9754691158661338, "maximum": 0.9754691158661338, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9816943516995644, "standard_deviation": null, "minimum": 1.9816943516995644, "maximum": 1.9816943516995644, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.43826145080589024, "standard_deviation": null, "minimum": 0.43826145080589024, "maximum": 0.43826145080589024, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18465854669513537, "standard_deviation": null, "minimum": 0.18465854669513537, "maximum": 0.18465854669513537, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9933709333698885, "standard_deviation": null, "minimum": 0.9933709333698885, "maximum": 0.9933709333698885, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.578371998659191, "standard_deviation": null, "minimum": 1.578371998659191, "maximum": 1.578371998659191, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.771556337255546, "standard_deviation": null, "minimum": 0.771556337255546, "maximum": 0.771556337255546, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10738720957888152, "standard_deviation": null, "minimum": 0.10738720957888152, "maximum": 0.10738720957888152, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.4173500200643881, "standard_deviation": null, "minimum": 0.4173500200643881, "maximum": 0.4173500200643881, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.95523823673635, "standard_deviation": null, "minimum": 5.95523823673635, "maximum": 5.95523823673635, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.063525851640955, "standard_deviation": null, "minimum": 1.063525851640955, "maximum": 1.063525851640955, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.39925821053619615, "standard_deviation": null, "minimum": 0.39925821053619615, "maximum": 0.39925821053619615, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9768023747879895, "standard_deviation": null, "minimum": 0.9768023747879895, "maximum": 0.9768023747879895, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.094450566708371, "standard_deviation": null, "minimum": 2.094450566708371, "maximum": 2.094450566708371, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.1061671434034049, "standard_deviation": null, "minimum": 0.1061671434034049, "maximum": 0.1061671434034049, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.15791147178829054, "standard_deviation": null, "minimum": 0.15791147178829054, "maximum": 0.15791147178829054, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9955717616441553, "standard_deviation": null, "minimum": 0.9955717616441553, "maximum": 0.9955717616441553, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.424234715042344, "standard_deviation": null, "minimum": 1.424234715042344, "maximum": 1.424234715042344, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.26971612300328146, "standard_deviation": null, "minimum": 0.26971612300328146, "maximum": 0.26971612300328146, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.07863643574353049, "standard_deviation": null, "minimum": 0.07863643574353049, "maximum": 0.07863643574353049, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8230"}, {"case": "librispeech_8455-210777-0010", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3585898876190186, "predicted_mos": 3.3585898876190186, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 49600, "input_sample_rate": 16000, "input_seconds": 3.1, "input_float32_sha256": "22d4811789bb3ff35caf1df5cc445821501fe9fe2da309fb2477d8fb01d0c7d2", "internal_samples": 49600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.883679000195116}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.448631455412157, "全帯域_10ms_rms_envelope_nrmse": 0.1538553018244409, "全帯域_10ms_envelope_correlation": 0.9887346747889803, "全帯域_10ms_modulation_mae_db": 0.6106184463393615, "全帯域_40ms_log_rms_mae_db": 1.8948864583257103, "全帯域_40ms_rms_envelope_nrmse": 0.1258150504486902, "全帯域_40ms_envelope_correlation": 0.9944077537953412, "全帯域_40ms_modulation_mae_db": 0.4810834099060452, "300–1000 Hz_10ms_log_rms_mae_db": 2.258126004262772, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.20055710753712044, "300–1000 Hz_10ms_envelope_correlation": 0.9860917232526509, "300–1000 Hz_10ms_modulation_mae_db": 0.9825699821312064, "300–1000 Hz_40ms_log_rms_mae_db": 1.6494754361090742, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17582222400797654, "300–1000 Hz_40ms_envelope_correlation": 0.9912342432437339, "300–1000 Hz_40ms_modulation_mae_db": 1.1565200588353939, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4051770968499233, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.09781953207443657, "1000–3000 Hz_10ms_envelope_correlation": 0.9953925726463543, "1000–3000 Hz_10ms_modulation_mae_db": 0.5117020980038107, "1000–3000 Hz_40ms_log_rms_mae_db": 0.957101813168861, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.07036480618817144, "1000–3000 Hz_40ms_envelope_correlation": 0.9980760030770128, "1000–3000 Hz_40ms_modulation_mae_db": 0.6677824731647203, "3000–6000 Hz_10ms_log_rms_mae_db": 1.592605735673243, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.23864891530228358, "3000–6000 Hz_10ms_envelope_correlation": 0.970084697559398, "3000–6000 Hz_10ms_modulation_mae_db": 2.3037422311056317, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0719694283885033, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.17034759059872434, "3000–6000 Hz_40ms_envelope_correlation": 0.9867347119860457, "3000–6000 Hz_40ms_modulation_mae_db": 2.238477909858861, "mr_stft_db": 5.988326759297469, "band_mae_db": 0.7151790658788556, "rms_gain_error_db": 0.773032709405308, "output_peak": 1.227513994034603}, "full_scale": {"sample_count": 49600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 10, "at_or_above_full_scale_fraction": 0.00020161290322580645, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.00020161290322580645, "longest_full_scale_run_samples": 2, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 49600, "output_samples": 49600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.5446109162833634, "spectral_convergence": 0.5364611218881057, "spectral_convergence_status": "ok", "amplitude_floor": 1.4924605057201793e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.580113017593372, "spectral_convergence": 0.2996003132374375, "spectral_convergence_status": "ok", "amplitude_floor": 8.127484488587521e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.840256344015672, "spectral_convergence": 0.33776025821736555, "spectral_convergence_status": "ok", "amplitude_floor": 5.495951303547364e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008185024372796483, "output_power": 0.0009398905828345717, "delta_db": 0.6005732127935105, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0004929662452993321, "output_power": 0.0006121777095339363, "delta_db": 0.9406032911882533, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.001304221454098428, "output_power": 0.001686818801905969, "delta_db": 1.1171709321694054, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007846801909808332, "output_power": 0.0008574975773958291, "delta_db": 0.38540212634786325, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0003122815376510157, "output_power": 0.00033422546886530135, "delta_db": 0.2949323207574409, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0001061689290478197, "output_power": 0.00013220159282089, "delta_db": 0.95239251201666, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7151790658788556, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.773032709405308, "pitch": null, "output_rms_dbfs": -23.40767450326878, "output_peak": 1.227513994034603, "reference_rms_dbfs": -24.180707212674086, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.988326759297469}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 49600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.0002690918479937382, "log_rms_mae_db": 2.448631455412157, "rms_envelope_nrmse": 0.1538553018244409, "envelope_correlation": 0.9887346747889803, "rms_gain_error_db": 0.7730344030066271, "reference_mean_power": 0.0038249880205223005, "output_mean_power": 0.00457018225409776}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 2.1583633605855604e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008154057878800693, "output_power": 0.0009336782835510449, "reference_active": true, "delta_db": 0.5882346783855265}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.087495849252009e-05, "output_power": 0.00010282098730483307, "reference_active": true, "delta_db": 0.5363754421997011}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.599429116228097e-05, "output_power": 3.059162176490909e-05, "reference_active": true, "delta_db": 0.7072452184328565}], "active_band_count": 3, "active_band_mae_db": 0.6106184463393615}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.0002456731948415248, "log_rms_mae_db": 1.8948864583257103, "rms_envelope_nrmse": 0.1258150504486902, "envelope_correlation": 0.9944077537953412, "rms_gain_error_db": 0.7730399444672125, "reference_mean_power": 0.0038624153279842384, "output_mean_power": 0.00461490713658075}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 2.0216648233438394e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007562450561775054, "output_power": 0.0008575416053960621, "reference_active": true, "delta_db": 0.5459265137263223}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.1399930780791825e-05, "output_power": 4.8064930129858204e-05, "reference_active": true, "delta_db": 0.648287003347377}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0921358976760216e-06, "output_power": 1.156592451443814e-06, "reference_active": true, "delta_db": 0.24903671264443605}], "active_band_count": 3, "active_band_mae_db": 0.4810834099060452}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00017000794064843354, "log_rms_mae_db": 2.258126004262772, "rms_envelope_nrmse": 0.20055710753712044, "envelope_correlation": 0.9860917232526509, "rms_gain_error_db": 1.0490467076062115, "reference_mean_power": 0.001461911046007658, "output_mean_power": 0.001861339605514937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.2056298552980857e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006038941034712422, "output_power": 0.0007621637151505055, "reference_active": true, "delta_db": 1.0108748007278958}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.14042207191201e-05, "output_power": 8.09056005488797e-05, "reference_active": true, "delta_db": 0.5425470215295288}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.020128231954835e-05, "output_power": 1.4653795472953206e-05, "reference_active": true, "delta_db": -1.3942881241361946}], "active_band_count": 3, "active_band_mae_db": 0.9825699821312064}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00015858554889566586, "log_rms_mae_db": 1.6494754361090742, "rms_envelope_nrmse": 0.17582222400797654, "envelope_correlation": 0.9912342432437339, "rms_gain_error_db": 1.0490448429913575, "reference_mean_power": 0.0014762175799508475, "output_mean_power": 0.0018795542148059572}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.148107626412446e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000571116387197927, "output_power": 0.000712162665622803, "reference_active": true, "delta_db": 0.9585458083116692}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.637477969141665e-05, "output_power": 4.123717336359485e-05, "reference_active": true, "delta_db": 0.5448851691368465}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.172790453481755e-06, "output_power": 7.4577445302304e-07, "reference_active": true, "delta_db": -1.966129199057666}], "active_band_count": 3, "active_band_mae_db": 1.1565200588353939}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 0.00019136572750236326, "log_rms_mae_db": 1.4051770968499233, "rms_envelope_nrmse": 0.09781953207443657, "envelope_correlation": 0.9953925726463543, "rms_gain_error_db": 0.35405077364236476, "reference_mean_power": 0.0009651549113833473, "output_mean_power": 0.0010471336171072746}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 7.99902404373172e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003359743599084894, "output_power": 0.00034549909110777506, "reference_active": true, "delta_db": 0.12140774026702014}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9966531082920368e-05, "output_power": 3.585942305595363e-05, "reference_active": true, "delta_db": 0.779668258467183}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.661292025781732e-06, "output_power": 5.756459369784859e-06, "reference_active": true, "delta_db": -0.634030295277229}], "active_band_count": 3, "active_band_mae_db": 0.5117020980038107}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 0.00016758862552795117, "log_rms_mae_db": 0.957101813168861, "rms_envelope_nrmse": 0.07036480618817144, "envelope_correlation": 0.9980760030770128, "rms_gain_error_db": 0.3540497430341645, "reference_mean_power": 0.0009746009072787481, "output_mean_power": 0.001057381689776798}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 7.731649944214162e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00031280752850392277, "output_power": 0.00031948068485394333, "reference_active": true, "delta_db": 0.09167409827515516}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5385931815426458e-05, "output_power": 1.95058502011885e-05, "reference_active": true, "delta_db": 1.030410811684562}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1192579703922553e-07, "output_power": 2.5463907178803237e-07, "reference_active": true, "delta_db": -0.8812625095344435}], "active_band_count": 3, "active_band_mae_db": 0.6677824731647203}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 619, "rms_floor": 5.61189760356505e-05, "log_rms_mae_db": 1.592605735673243, "rms_envelope_nrmse": 0.23864891530228358, "envelope_correlation": 0.970084697559398, "rms_gain_error_db": 0.771114922210921, "reference_mean_power": 0.00014982208037105655, "output_mean_power": 0.00017893171126065512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 619, "frequency_resolution_hz": 0.32310177705977383, "power_floor": 1.5994961899965245e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.519717012759751e-05, "output_power": 6.904041505783354e-05, "reference_active": true, "delta_db": 0.9718658090824028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.5683811710113524e-05, "output_power": 2.1068630083266238e-05, "reference_active": true, "delta_db": 1.2818467819405859}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.1546072478197644e-06, "output_power": 9.21927367032875e-06, "reference_active": true, "delta_db": 4.657514102293906}], "active_band_count": 3, "active_band_mae_db": 2.3037422311056317}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 49600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 613, "rms_floor": 4.51769085602287e-05, "log_rms_mae_db": 1.0719694283885033, "rms_envelope_nrmse": 0.17034759059872434, "envelope_correlation": 0.9867347119860457, "rms_gain_error_db": 0.7711154446228385, "reference_mean_power": 0.0001512883740521334, "output_mean_power": 0.00018068291971534592}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 613, "frequency_resolution_hz": 0.3262642740619902, "power_floor": 1.4928738695410609e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.9695472955753735e-05, "output_power": 6.514908699476932e-05, "reference_active": true, "delta_db": 1.1759150567657475}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.547268075821193e-06, "output_power": 1.1176834612365997e-05, "reference_active": true, "delta_db": 1.1649149921783681}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7628269396099532e-07, "output_power": 4.826919398232403e-07, "reference_active": true, "delta_db": 4.374603680632468}], "active_band_count": 3, "active_band_mae_db": 2.238477909858861}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8881489188067627, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "38cc3cfff70f2919785b465155a87850f9fde710f349dc247a558f291352484e", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9953925726463543, "standard_deviation": null, "minimum": 0.9953925726463543, "maximum": 0.9953925726463543, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4051770968499233, "standard_deviation": null, "minimum": 1.4051770968499233, "maximum": 1.4051770968499233, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5117020980038107, "standard_deviation": null, "minimum": 0.5117020980038107, "maximum": 0.5117020980038107, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.09781953207443657, "standard_deviation": null, "minimum": 0.09781953207443657, "maximum": 0.09781953207443657, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9980760030770128, "standard_deviation": null, "minimum": 0.9980760030770128, "maximum": 0.9980760030770128, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.957101813168861, "standard_deviation": null, "minimum": 0.957101813168861, "maximum": 0.957101813168861, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.6677824731647203, "standard_deviation": null, "minimum": 0.6677824731647203, "maximum": 0.6677824731647203, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.07036480618817144, "standard_deviation": null, "minimum": 0.07036480618817144, "maximum": 0.07036480618817144, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.970084697559398, "standard_deviation": null, "minimum": 0.970084697559398, "maximum": 0.970084697559398, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.592605735673243, "standard_deviation": null, "minimum": 1.592605735673243, "maximum": 1.592605735673243, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 2.3037422311056317, "standard_deviation": null, "minimum": 2.3037422311056317, "maximum": 2.3037422311056317, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.23864891530228358, "standard_deviation": null, "minimum": 0.23864891530228358, "maximum": 0.23864891530228358, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9867347119860457, "standard_deviation": null, "minimum": 0.9867347119860457, "maximum": 0.9867347119860457, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0719694283885033, "standard_deviation": null, "minimum": 1.0719694283885033, "maximum": 1.0719694283885033, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 2.238477909858861, "standard_deviation": null, "minimum": 2.238477909858861, "maximum": 2.238477909858861, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17034759059872434, "standard_deviation": null, "minimum": 0.17034759059872434, "maximum": 0.17034759059872434, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9860917232526509, "standard_deviation": null, "minimum": 0.9860917232526509, "maximum": 0.9860917232526509, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.258126004262772, "standard_deviation": null, "minimum": 2.258126004262772, "maximum": 2.258126004262772, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9825699821312064, "standard_deviation": null, "minimum": 0.9825699821312064, "maximum": 0.9825699821312064, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20055710753712044, "standard_deviation": null, "minimum": 0.20055710753712044, "maximum": 0.20055710753712044, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9912342432437339, "standard_deviation": null, "minimum": 0.9912342432437339, "maximum": 0.9912342432437339, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6494754361090742, "standard_deviation": null, "minimum": 1.6494754361090742, "maximum": 1.6494754361090742, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.1565200588353939, "standard_deviation": null, "minimum": 1.1565200588353939, "maximum": 1.1565200588353939, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17582222400797654, "standard_deviation": null, "minimum": 0.17582222400797654, "maximum": 0.17582222400797654, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7151790658788556, "standard_deviation": null, "minimum": 0.7151790658788556, "maximum": 0.7151790658788556, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.988326759297469, "standard_deviation": null, "minimum": 5.988326759297469, "maximum": 5.988326759297469, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.227513994034603, "standard_deviation": null, "minimum": 1.227513994034603, "maximum": 1.227513994034603, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.773032709405308, "standard_deviation": null, "minimum": 0.773032709405308, "maximum": 0.773032709405308, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9887346747889803, "standard_deviation": null, "minimum": 0.9887346747889803, "maximum": 0.9887346747889803, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.448631455412157, "standard_deviation": null, "minimum": 2.448631455412157, "maximum": 2.448631455412157, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6106184463393615, "standard_deviation": null, "minimum": 0.6106184463393615, "maximum": 0.6106184463393615, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.1538553018244409, "standard_deviation": null, "minimum": 0.1538553018244409, "maximum": 0.1538553018244409, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9944077537953412, "standard_deviation": null, "minimum": 0.9944077537953412, "maximum": 0.9944077537953412, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.8948864583257103, "standard_deviation": null, "minimum": 1.8948864583257103, "maximum": 1.8948864583257103, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.4810834099060452, "standard_deviation": null, "minimum": 0.4810834099060452, "maximum": 0.4810834099060452, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1258150504486902, "standard_deviation": null, "minimum": 0.1258150504486902, "maximum": 0.1258150504486902, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8455"}, {"case": "librispeech_8455-210777-0049", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.40759539604187, "predicted_mos": 3.40759539604187, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 65760, "input_sample_rate": 16000, "input_seconds": 4.11, "input_float32_sha256": "7af6278863ad7924ed2e725ae1728cddfce4f80c00562ded9e637156105cea1a", "internal_samples": 65760, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.792592000216246}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.12807702052811, "全帯域_10ms_rms_envelope_nrmse": 0.18076374796525757, "全帯域_10ms_envelope_correlation": 0.9857243837754187, "全帯域_10ms_modulation_mae_db": 0.7286058322556141, "全帯域_40ms_log_rms_mae_db": 1.6271977153792943, "全帯域_40ms_rms_envelope_nrmse": 0.13424458020789648, "全帯域_40ms_envelope_correlation": 0.9960237656652949, "全帯域_40ms_modulation_mae_db": 0.734785133223086, "300–1000 Hz_10ms_log_rms_mae_db": 2.154195450925897, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21862498030456118, "300–1000 Hz_10ms_envelope_correlation": 0.9836925476794804, "300–1000 Hz_10ms_modulation_mae_db": 1.5521478810834137, "300–1000 Hz_40ms_log_rms_mae_db": 1.6590289894914092, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.17341214770997096, "300–1000 Hz_40ms_envelope_correlation": 0.9937048570535489, "300–1000 Hz_40ms_modulation_mae_db": 1.8336080541740973, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6628069091149198, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2341004290869693, "1000–3000 Hz_10ms_envelope_correlation": 0.9822062870753473, "1000–3000 Hz_10ms_modulation_mae_db": 1.6347284299522276, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1528844685991082, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.17946141072404684, "1000–3000 Hz_40ms_envelope_correlation": 0.9949304189027364, "1000–3000 Hz_40ms_modulation_mae_db": 1.711151889868326, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5778912152972762, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.24943713018990588, "3000–6000 Hz_10ms_envelope_correlation": 0.9622396237926956, "3000–6000 Hz_10ms_modulation_mae_db": 1.7915634819247916, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9915136038007897, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.149307288404339, "3000–6000 Hz_40ms_envelope_correlation": 0.9859480380482514, "3000–6000 Hz_40ms_modulation_mae_db": 1.9860153973911345, "mr_stft_db": 6.025744568160448, "band_mae_db": 0.6953520476499824, "rms_gain_error_db": 0.9204929237787431, "output_peak": 0.6862236482246062}, "full_scale": {"sample_count": 65760, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 65760, "output_samples": 65760, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.673380664847031, "spectral_convergence": 0.7041350811388264, "spectral_convergence_status": "ok", "amplitude_floor": 1.6529112798953643e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.543999591959759, "spectral_convergence": 0.3064734500663521, "spectral_convergence_status": "ok", "amplitude_floor": 1.0394789070711673e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.859853447674554, "spectral_convergence": 0.3046197734999119, "spectral_convergence_status": "ok", "amplitude_floor": 8.202976886987361e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0008081818875357184, "output_power": 0.0009240137407829131, "delta_db": 0.5816931647456675, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0008371388318692298, "output_power": 0.001050783592643004, "delta_db": 0.9871579515842939, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0012771265272884344, "output_power": 0.001624046002931929, "delta_db": 1.0436440120675774, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003871568194673651, "output_power": 0.0005206197847574426, "delta_db": 1.2863375485516568, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001322399975596578, "output_power": 0.0001393832177341878, "delta_db": 0.22847653664982484, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.960478257270517e-05, "output_power": 1.9808077992578783e-05, "delta_db": 0.044803072300873444, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6953520476499824, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.9204929237787431, "pitch": null, "output_rms_dbfs": -23.68692789683275, "output_peak": 0.6862236482246062, "reference_rms_dbfs": -24.60742082061149, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.025744568160448}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 65760, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00020667206862955213, "log_rms_mae_db": 2.12807702052811, "rms_envelope_nrmse": 0.18076374796525757, "envelope_correlation": 0.9857243837754187, "rms_gain_error_db": 0.9204984868688217, "reference_mean_power": 0.003465660170380573, "output_mean_power": 0.004283865470245962}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 2.300680394398606e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009732960931433549, "output_power": 0.0012864586000545075, "reference_active": true, "delta_db": 1.21150834498059}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.170752162168669e-05, "output_power": 9.823993897003313e-05, "reference_active": true, "delta_db": 0.8002604645737221}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.07791133597542e-05, "output_power": 2.1628772923747907e-05, "reference_active": true, "delta_db": 0.17404868721253025}], "active_band_count": 3, "active_band_mae_db": 0.7286058322556141}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00018941444655877122, "log_rms_mae_db": 1.6271977153792943, "rms_envelope_nrmse": 0.13424458020789648, "envelope_correlation": 0.9960237656652949, "rms_gain_error_db": 0.9205151017445377, "reference_mean_power": 0.003491154294575285, "output_mean_power": 0.004315394993941419}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 2.1871865789273095e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008996086963166702, "output_power": 0.0011949832787163275, "reference_active": true, "delta_db": 1.233081832748331}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.8292682982878605e-05, "output_power": 4.590490707692709e-05, "reference_active": true, "delta_db": 0.7874331628230323}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.1507764875014336e-06, "output_power": 1.1030798196970334e-06, "reference_active": true, "delta_db": -0.18384040409789457}], "active_band_count": 3, "active_band_mae_db": 0.734785133223086}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.0001875778806757321, "log_rms_mae_db": 2.154195450925897, "rms_envelope_nrmse": 0.21862498030456118, "envelope_correlation": 0.9836925476794804, "rms_gain_error_db": 1.135522129335874, "reference_mean_power": 0.0016622862050937872, "output_mean_power": 0.0021590267028568687}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 1.477365390744301e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007590270471846978, "output_power": 0.001014318444448963, "reference_active": true, "delta_db": 1.2591707097405134}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.577918818080543e-05, "output_power": 8.524584188949775e-05, "reference_active": true, "delta_db": 1.8420101503693642}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4662085193890421e-05, "output_power": 1.0248716039436178e-05, "reference_active": true, "delta_db": -1.555262783140363}], "active_band_count": 3, "active_band_mae_db": 1.5521478810834137}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 0.00016891518288067464, "log_rms_mae_db": 1.6590289894914092, "rms_envelope_nrmse": 0.17341214770997096, "envelope_correlation": 0.9937048570535489, "rms_gain_error_db": 1.1355207416195865, "reference_mean_power": 0.001674521427963224, "output_mean_power": 0.002174917478818734}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 1.422523294383883e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007062170888212691, "output_power": 0.0009501200866951327, "reference_active": true, "delta_db": 1.2884027742570399}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.54990695717255e-05, "output_power": 4.0145582249274845e-05, "reference_active": true, "delta_db": 1.9711342718806342}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.502016839919997e-07, "output_power": 5.671358280431944e-07, "reference_active": true, "delta_db": -2.2412871163846186}], "active_band_count": 3, "active_band_mae_db": 1.8336080541740973}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 0.00010800393678736649, "log_rms_mae_db": 1.6628069091149198, "rms_envelope_nrmse": 0.2341004290869693, "envelope_correlation": 0.9822062870753473, "rms_gain_error_db": 1.2415110259353364, "reference_mean_power": 0.0004089931043607793, "output_mean_power": 0.0005443360386524281}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 4.3909884369057556e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019183671841687967, "output_power": 0.00026396370677212784, "reference_active": true, "delta_db": 1.3861248169271976}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4527862162762071e-05, "output_power": 2.247518983265735e-05, "reference_active": true, "delta_db": 1.8950165790742737}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.245905579917347e-06, "output_power": 1.1982370581118085e-05, "reference_active": true, "delta_db": 1.6230438938552114}], "active_band_count": 3, "active_band_mae_db": 1.6347284299522276}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 8.815867833009974e-05, "log_rms_mae_db": 1.1528844685991082, "rms_envelope_nrmse": 0.17946141072404684, "envelope_correlation": 0.9949304189027364, "rms_gain_error_db": 1.2415122017186273, "reference_mean_power": 0.0004120037826983822, "output_mean_power": 0.000548343151289819}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 4.2438330741066236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00018201492488703638, "output_power": 0.00025210376981118593, "reference_active": true, "delta_db": 1.4147233911963943}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.209318154228534e-06, "output_power": 1.0710223540795601e-05, "reference_active": true, "delta_db": 1.7190434361487235}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.163481758524237e-07, "output_power": 5.013421495498987e-07, "reference_active": true, "delta_db": 1.9996888422598602}], "active_band_count": 3, "active_band_mae_db": 1.711151889868326}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 821, "rms_floor": 8.95305280668978e-05, "log_rms_mae_db": 1.5778912152972762, "rms_envelope_nrmse": 0.24943713018990588, "envelope_correlation": 0.9622396237926956, "rms_gain_error_db": -0.05137454212542389, "reference_mean_power": 8.45768795231113e-05, "output_mean_power": 8.358227800951813e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 821, "frequency_resolution_hz": 0.24360535931790497, "power_floor": 3.626876619728336e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.450635181337156e-05, "output_power": 1.2797904763589866e-05, "reference_active": true, "delta_db": -0.5441933190721814}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.817293180171653e-06, "output_power": 5.1118789522749655e-06, "reference_active": true, "delta_db": -1.8447583875345122}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5650085525655336e-06, "output_power": 7.869422112965975e-07, "reference_active": true, "delta_db": -2.9857387391676813}], "active_band_count": 3, "active_band_mae_db": 1.7915634819247916}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 65760, "uncovered_tail_samples": 0, "metrics": {"frame_count": 815, "rms_floor": 5.6185612647784654e-05, "log_rms_mae_db": 0.9915136038007897, "rms_envelope_nrmse": 0.149307288404339, "envelope_correlation": 0.9859480380482514, "rms_gain_error_db": -0.051375601091855204, "reference_mean_power": 8.519942751057583e-05, "output_mean_power": 8.419748446854312e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 815, "frequency_resolution_hz": 0.24539877300613497, "power_floor": 3.342409970638096e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.5109841043230135e-05, "output_power": 1.2730416171755959e-05, "reference_active": true, "delta_db": -0.744172940840635}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.127479817162529e-06, "output_power": 2.586636463235518e-06, "reference_active": true, "delta_db": -2.0294956286387245}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2069317938633786e-07, "output_power": 5.797556178655054e-08, "reference_active": true, "delta_db": -3.1843776226940443}], "active_band_count": 3, "active_band_mae_db": 1.9860153973911345}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8942489045973983, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "c078f5bb4da9d5cf5a4c876975ca7dcca3d0f6d463b440ca63a1e7e8213116cd", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9822062870753473, "standard_deviation": null, "minimum": 0.9822062870753473, "maximum": 0.9822062870753473, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6628069091149198, "standard_deviation": null, "minimum": 1.6628069091149198, "maximum": 1.6628069091149198, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.6347284299522276, "standard_deviation": null, "minimum": 1.6347284299522276, "maximum": 1.6347284299522276, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2341004290869693, "standard_deviation": null, "minimum": 0.2341004290869693, "maximum": 0.2341004290869693, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9949304189027364, "standard_deviation": null, "minimum": 0.9949304189027364, "maximum": 0.9949304189027364, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1528844685991082, "standard_deviation": null, "minimum": 1.1528844685991082, "maximum": 1.1528844685991082, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.711151889868326, "standard_deviation": null, "minimum": 1.711151889868326, "maximum": 1.711151889868326, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17946141072404684, "standard_deviation": null, "minimum": 0.17946141072404684, "maximum": 0.17946141072404684, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9622396237926956, "standard_deviation": null, "minimum": 0.9622396237926956, "maximum": 0.9622396237926956, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5778912152972762, "standard_deviation": null, "minimum": 1.5778912152972762, "maximum": 1.5778912152972762, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.7915634819247916, "standard_deviation": null, "minimum": 1.7915634819247916, "maximum": 1.7915634819247916, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.24943713018990588, "standard_deviation": null, "minimum": 0.24943713018990588, "maximum": 0.24943713018990588, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9859480380482514, "standard_deviation": null, "minimum": 0.9859480380482514, "maximum": 0.9859480380482514, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9915136038007897, "standard_deviation": null, "minimum": 0.9915136038007897, "maximum": 0.9915136038007897, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.9860153973911345, "standard_deviation": null, "minimum": 1.9860153973911345, "maximum": 1.9860153973911345, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.149307288404339, "standard_deviation": null, "minimum": 0.149307288404339, "maximum": 0.149307288404339, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9836925476794804, "standard_deviation": null, "minimum": 0.9836925476794804, "maximum": 0.9836925476794804, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.154195450925897, "standard_deviation": null, "minimum": 2.154195450925897, "maximum": 2.154195450925897, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.5521478810834137, "standard_deviation": null, "minimum": 1.5521478810834137, "maximum": 1.5521478810834137, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21862498030456118, "standard_deviation": null, "minimum": 0.21862498030456118, "maximum": 0.21862498030456118, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9937048570535489, "standard_deviation": null, "minimum": 0.9937048570535489, "maximum": 0.9937048570535489, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.6590289894914092, "standard_deviation": null, "minimum": 1.6590289894914092, "maximum": 1.6590289894914092, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.8336080541740973, "standard_deviation": null, "minimum": 1.8336080541740973, "maximum": 1.8336080541740973, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.17341214770997096, "standard_deviation": null, "minimum": 0.17341214770997096, "maximum": 0.17341214770997096, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6953520476499824, "standard_deviation": null, "minimum": 0.6953520476499824, "maximum": 0.6953520476499824, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.025744568160448, "standard_deviation": null, "minimum": 6.025744568160448, "maximum": 6.025744568160448, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6862236482246062, "standard_deviation": null, "minimum": 0.6862236482246062, "maximum": 0.6862236482246062, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.9204929237787431, "standard_deviation": null, "minimum": 0.9204929237787431, "maximum": 0.9204929237787431, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9857243837754187, "standard_deviation": null, "minimum": 0.9857243837754187, "maximum": 0.9857243837754187, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.12807702052811, "standard_deviation": null, "minimum": 2.12807702052811, "maximum": 2.12807702052811, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7286058322556141, "standard_deviation": null, "minimum": 0.7286058322556141, "maximum": 0.7286058322556141, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.18076374796525757, "standard_deviation": null, "minimum": 0.18076374796525757, "maximum": 0.18076374796525757, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9960237656652949, "standard_deviation": null, "minimum": 0.9960237656652949, "maximum": 0.9960237656652949, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.6271977153792943, "standard_deviation": null, "minimum": 1.6271977153792943, "maximum": 1.6271977153792943, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.734785133223086, "standard_deviation": null, "minimum": 0.734785133223086, "maximum": 0.734785133223086, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.13424458020789648, "standard_deviation": null, "minimum": 0.13424458020789648, "maximum": 0.13424458020789648, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8455"}, {"case": "librispeech_8455-210777-0060", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.48962140083313, "predicted_mos": 3.48962140083313, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "3357c01a2d3f93316405491f6dd4058f8a8e4742799b47a54173e76147055fc9", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.875670500099659}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.794493793263051, "全帯域_10ms_rms_envelope_nrmse": 0.177563374421457, "全帯域_10ms_envelope_correlation": 0.9815914479795114, "全帯域_10ms_modulation_mae_db": 0.44648435503580836, "全帯域_40ms_log_rms_mae_db": 1.319855926635497, "全帯域_40ms_rms_envelope_nrmse": 0.13288163990623325, "全帯域_40ms_envelope_correlation": 0.9943149328513303, "全帯域_40ms_modulation_mae_db": 0.35115285650846323, "300–1000 Hz_10ms_log_rms_mae_db": 1.814950400093778, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2329502365722995, "300–1000 Hz_10ms_envelope_correlation": 0.9813080741293166, "300–1000 Hz_10ms_modulation_mae_db": 0.768608823546372, "300–1000 Hz_40ms_log_rms_mae_db": 1.2785302512444026, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.19294281208066275, "300–1000 Hz_40ms_envelope_correlation": 0.9921373665309129, "300–1000 Hz_40ms_modulation_mae_db": 0.6935545933845827, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5485936263165545, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.178088745019671, "1000–3000 Hz_10ms_envelope_correlation": 0.9850006241868663, "1000–3000 Hz_10ms_modulation_mae_db": 0.6346530727394849, "1000–3000 Hz_40ms_log_rms_mae_db": 0.997643867709442, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12130727453099292, "1000–3000 Hz_40ms_envelope_correlation": 0.9959546544412291, "1000–3000 Hz_40ms_modulation_mae_db": 0.694333063049866, "3000–6000 Hz_10ms_log_rms_mae_db": 1.814498330477143, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.21442535672599883, "3000–6000 Hz_10ms_envelope_correlation": 0.9696866860944658, "3000–6000 Hz_10ms_modulation_mae_db": 0.8832846234009252, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0682921227794901, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12240725143921456, "3000–6000 Hz_40ms_envelope_correlation": 0.9900201446601663, "3000–6000 Hz_40ms_modulation_mae_db": 0.7684554514730272, "mr_stft_db": 5.810484850244961, "band_mae_db": 0.6698095555743446, "rms_gain_error_db": 0.8869581087862015, "output_peak": 1.2815085810222646}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 17, "at_or_above_full_scale_fraction": 0.00017708333333333335, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0001875, "longest_full_scale_run_samples": 3, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.660290897600118, "spectral_convergence": 0.6730334145282195, "spectral_convergence_status": "ok", "amplitude_floor": 2.2239575932173076e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.12984055665154, "spectral_convergence": 0.28117637590557126, "spectral_convergence_status": "ok", "amplitude_floor": 1.640120571472942e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.6413230964832275, "spectral_convergence": 0.303129358459228, "spectral_convergence_status": "ok", "amplitude_floor": 1.3411038650227006e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0016608291202231148, "output_power": 0.0017608666383487428, "delta_db": 0.2540151436698585, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.001899519997387598, "output_power": 0.002251839210383453, "delta_db": 0.7389350706661436, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.002727484270807038, "output_power": 0.0037516360971979077, "delta_db": 1.384584511435619, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0007661816102082716, "output_power": 0.0009303935625612581, "delta_db": 0.8433497293366993, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.000247481283821024, "output_power": 0.0002688160515910837, "delta_db": 0.35912837546868465, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 6.389711909369124e-05, "output_power": 7.069126088215182e-05, "delta_db": 0.4388445028690625, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.6698095555743446, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.8869581087862015, "pitch": null, "output_rms_dbfs": -20.441082407186375, "output_peak": 1.2815085810222646, "reference_rms_dbfs": -21.328040515972575, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.810484850244961}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00031196729436307625, "log_rms_mae_db": 1.794493793263051, "rms_envelope_nrmse": 0.177563374421457, "envelope_correlation": 0.9815914479795114, "rms_gain_error_db": 0.8871918196462789, "reference_mean_power": 0.00737113406357093, "output_mean_power": 0.009041770761077202}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.772697507541185e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010414087648688783, "output_power": 0.001244591234526995, "reference_active": true, "delta_db": 0.7740550945421834}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001272580988246754, "output_power": 0.00013746426879433627, "reference_active": true, "delta_db": 0.3350439550056797}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4550125163182702e-05, "output_power": 2.5887440407984734e-05, "reference_active": true, "delta_db": 0.23035401555956195}], "active_band_count": 3, "active_band_mae_db": 0.44648435503580836}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.0002933828481800726, "log_rms_mae_db": 1.319855926635497, "rms_envelope_nrmse": 0.13288163990623325, "envelope_correlation": 0.9943149328513303, "rms_gain_error_db": 0.8879134343076834, "reference_mean_power": 0.007404046443755548, "output_mean_power": 0.00908365178980512}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.623838097183376e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009238145152881926, "output_power": 0.001111063521849311, "reference_active": true, "delta_db": 0.8015410743700295}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.0904498001922555e-05, "output_power": 6.156765580321386e-05, "reference_active": true, "delta_db": 0.047032501605919}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.8155579307174739e-06, "output_power": 1.7318952119579976e-06, "reference_active": true, "delta_db": -0.20488499354944126}], "active_band_count": 3, "active_band_mae_db": 0.35115285650846323}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00027755394478816403, "log_rms_mae_db": 1.814950400093778, "rms_envelope_nrmse": 0.2329502365722995, "envelope_correlation": 0.9813080741293166, "rms_gain_error_db": 1.2828315210196226, "reference_mean_power": 0.003209968762055962, "output_mean_power": 0.004313044690219449}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.3498130551666851e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006624323096451465, "output_power": 0.0008897923647979773, "reference_active": true, "delta_db": 1.2814716811389872}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.187213693276335e-05, "output_power": 8.654785253795391e-05, "reference_active": true, "delta_db": 0.8069573877023666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.2558954262897386e-05, "output_power": 1.19457587204343e-05, "reference_active": true, "delta_db": -0.2173974017977626}], "active_band_count": 3, "active_band_mae_db": 0.768608823546372}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025197665775423135, "log_rms_mae_db": 1.2785302512444026, "rms_envelope_nrmse": 0.19294281208066275, "envelope_correlation": 0.9921373665309129, "rms_gain_error_db": 1.2829147331398234, "reference_mean_power": 0.0032255924683410116, "output_mean_power": 0.00433412038019151}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.287517318914353e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006130037756394711, "output_power": 0.0008238090004874433, "reference_active": true, "delta_db": 1.2836338305867545}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.4809964114174026e-05, "output_power": 3.9600664054835676e-05, "reference_active": true, "delta_db": 0.559988930079113}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.243410741089396e-07, "output_power": 8.752420000147462e-07, "reference_active": true, "delta_db": -0.2370410194878806}], "active_band_count": 3, "active_band_mae_db": 0.6935545933845827}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017139073768520627, "log_rms_mae_db": 1.5485936263165545, "rms_envelope_nrmse": 0.178088745019671, "envelope_correlation": 0.9850006241868663, "rms_gain_error_db": 0.7810023762142764, "reference_mean_power": 0.0008077036407054461, "output_mean_power": 0.0009668348093057285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.727013621661235e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00024375238055045323, "output_power": 0.00029632785165002436, "reference_active": true, "delta_db": 0.8482360661023859}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.980069758207981e-05, "output_power": 3.7644322527620145e-05, "reference_active": true, "delta_db": 1.0147305542574665}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.347390229447928e-06, "output_power": 7.278365260570325e-06, "reference_active": true, "delta_db": -0.04099259785860238}], "active_band_count": 3, "active_band_mae_db": 0.6346530727394849}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015289190214292506, "log_rms_mae_db": 0.997643867709442, "rms_envelope_nrmse": 0.12130727453099292, "envelope_correlation": 0.9959546544412291, "rms_gain_error_db": 0.7810971843198606, "reference_mean_power": 0.0008114918980720446, "output_mean_power": 0.0009713906225198752}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.4774424074646414e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022848364185803233, "output_power": 0.0002784993037860849, "reference_active": true, "delta_db": 0.8596900134861352}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.2508904456720676e-05, "output_power": 1.5774168875160866e-05, "reference_active": true, "delta_db": 1.0072721060446985}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.746475959138801e-07, "output_power": 6.039560013557684e-07, "reference_active": true, "delta_db": 0.21603706961876415}], "active_band_count": 3, "active_band_mae_db": 0.694333063049866}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.106734387114872e-05, "log_rms_mae_db": 1.814498330477143, "rms_envelope_nrmse": 0.21442535672599883, "envelope_correlation": 0.9696866860944658, "rms_gain_error_db": 0.29820498256379424, "reference_mean_power": 0.0001388420753372346, "output_mean_power": 0.00014871048667237645}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.164366134779655e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.596504488102423e-05, "output_power": 5.2677675371657515e-05, "reference_active": true, "delta_db": 0.5919891310085563}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.50200265064541e-05, "output_power": 1.6373958543940257e-05, "reference_active": true, "delta_db": 0.3748298740962501}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4297410894022624e-06, "output_power": 3.5798389137504484e-06, "reference_active": true, "delta_db": 1.683034865097969}], "active_band_count": 3, "active_band_mae_db": 0.8832846234009252}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 4.91579507177982e-05, "log_rms_mae_db": 1.0682921227794901, "rms_envelope_nrmse": 0.12240725143921456, "envelope_correlation": 0.9900201446601663, "rms_gain_error_db": 0.2981030539579228, "reference_mean_power": 0.00013952517981136655, "output_mean_power": 0.00014943863647897277}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0998934517742833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.388750580480172e-05, "output_power": 5.0986047398474166e-05, "reference_active": true, "delta_db": 0.6511044545907487}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.1884324982671665e-06, "output_power": 8.534569598964991e-06, "reference_active": true, "delta_db": 0.7454742546870001}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0513859982120003e-07, "output_power": 2.528867944482008e-07, "reference_active": true, "delta_db": 0.9087876451413325}], "active_band_count": 3, "active_band_mae_db": 0.7684554514730272}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9205848416190375, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "7ededef9303710872f32370b9354928f217646b38107f67ddffed3b92594aeba", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9850006241868663, "standard_deviation": null, "minimum": 0.9850006241868663, "maximum": 0.9850006241868663, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5485936263165545, "standard_deviation": null, "minimum": 1.5485936263165545, "maximum": 1.5485936263165545, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6346530727394849, "standard_deviation": null, "minimum": 0.6346530727394849, "maximum": 0.6346530727394849, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.178088745019671, "standard_deviation": null, "minimum": 0.178088745019671, "maximum": 0.178088745019671, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9959546544412291, "standard_deviation": null, "minimum": 0.9959546544412291, "maximum": 0.9959546544412291, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.997643867709442, "standard_deviation": null, "minimum": 0.997643867709442, "maximum": 0.997643867709442, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.694333063049866, "standard_deviation": null, "minimum": 0.694333063049866, "maximum": 0.694333063049866, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12130727453099292, "standard_deviation": null, "minimum": 0.12130727453099292, "maximum": 0.12130727453099292, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9696866860944658, "standard_deviation": null, "minimum": 0.9696866860944658, "maximum": 0.9696866860944658, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.814498330477143, "standard_deviation": null, "minimum": 1.814498330477143, "maximum": 1.814498330477143, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.8832846234009252, "standard_deviation": null, "minimum": 0.8832846234009252, "maximum": 0.8832846234009252, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21442535672599883, "standard_deviation": null, "minimum": 0.21442535672599883, "maximum": 0.21442535672599883, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9900201446601663, "standard_deviation": null, "minimum": 0.9900201446601663, "maximum": 0.9900201446601663, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0682921227794901, "standard_deviation": null, "minimum": 1.0682921227794901, "maximum": 1.0682921227794901, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.7684554514730272, "standard_deviation": null, "minimum": 0.7684554514730272, "maximum": 0.7684554514730272, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12240725143921456, "standard_deviation": null, "minimum": 0.12240725143921456, "maximum": 0.12240725143921456, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9813080741293166, "standard_deviation": null, "minimum": 0.9813080741293166, "maximum": 0.9813080741293166, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.814950400093778, "standard_deviation": null, "minimum": 1.814950400093778, "maximum": 1.814950400093778, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.768608823546372, "standard_deviation": null, "minimum": 0.768608823546372, "maximum": 0.768608823546372, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2329502365722995, "standard_deviation": null, "minimum": 0.2329502365722995, "maximum": 0.2329502365722995, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9921373665309129, "standard_deviation": null, "minimum": 0.9921373665309129, "maximum": 0.9921373665309129, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2785302512444026, "standard_deviation": null, "minimum": 1.2785302512444026, "maximum": 1.2785302512444026, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.6935545933845827, "standard_deviation": null, "minimum": 0.6935545933845827, "maximum": 0.6935545933845827, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.19294281208066275, "standard_deviation": null, "minimum": 0.19294281208066275, "maximum": 0.19294281208066275, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.6698095555743446, "standard_deviation": null, "minimum": 0.6698095555743446, "maximum": 0.6698095555743446, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.810484850244961, "standard_deviation": null, "minimum": 5.810484850244961, "maximum": 5.810484850244961, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 1.2815085810222646, "standard_deviation": null, "minimum": 1.2815085810222646, "maximum": 1.2815085810222646, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.8869581087862015, "standard_deviation": null, "minimum": 0.8869581087862015, "maximum": 0.8869581087862015, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9815914479795114, "standard_deviation": null, "minimum": 0.9815914479795114, "maximum": 0.9815914479795114, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.794493793263051, "standard_deviation": null, "minimum": 1.794493793263051, "maximum": 1.794493793263051, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.44648435503580836, "standard_deviation": null, "minimum": 0.44648435503580836, "maximum": 0.44648435503580836, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.177563374421457, "standard_deviation": null, "minimum": 0.177563374421457, "maximum": 0.177563374421457, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9943149328513303, "standard_deviation": null, "minimum": 0.9943149328513303, "maximum": 0.9943149328513303, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.319855926635497, "standard_deviation": null, "minimum": 1.319855926635497, "maximum": 1.319855926635497, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.35115285650846323, "standard_deviation": null, "minimum": 0.35115285650846323, "maximum": 0.35115285650846323, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.13288163990623325, "standard_deviation": null, "minimum": 0.13288163990623325, "maximum": 0.13288163990623325, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8455"}, {"case": "librispeech_8463-294828-0003", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.012377977371216, "predicted_mos": 3.012377977371216, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "9056f4bb9e16e8c0cbe226b8f8f547a67e2fb993a0975e2ad9d4c8f005b10536", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.87315890006721}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5236446610602021, "全帯域_10ms_rms_envelope_nrmse": 0.17142823187405234, "全帯域_10ms_envelope_correlation": 0.9799318008680328, "全帯域_10ms_modulation_mae_db": 0.3705004616520015, "全帯域_40ms_log_rms_mae_db": 1.0221685671134961, "全帯域_40ms_rms_envelope_nrmse": 0.10822210423058422, "全帯域_40ms_envelope_correlation": 0.9962101271429572, "全帯域_40ms_modulation_mae_db": 0.6887291287763334, "300–1000 Hz_10ms_log_rms_mae_db": 1.7665813446613192, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.16694283541185026, "300–1000 Hz_10ms_envelope_correlation": 0.9861527401280722, "300–1000 Hz_10ms_modulation_mae_db": 0.8903696298871044, "300–1000 Hz_40ms_log_rms_mae_db": 1.208812932615596, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.11585959285313456, "300–1000 Hz_40ms_envelope_correlation": 0.997125140834917, "300–1000 Hz_40ms_modulation_mae_db": 0.648454512022035, "1000–3000 Hz_10ms_log_rms_mae_db": 1.702832850282129, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.189303152684545, "1000–3000 Hz_10ms_envelope_correlation": 0.979547158331353, "1000–3000 Hz_10ms_modulation_mae_db": 0.5771742437631704, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0679009353575315, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.12265998686937944, "1000–3000 Hz_40ms_envelope_correlation": 0.9948166938814251, "1000–3000 Hz_40ms_modulation_mae_db": 0.9588738861315038, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7020294069173367, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25863970993582525, "3000–6000 Hz_10ms_envelope_correlation": 0.9442519113088492, "3000–6000 Hz_10ms_modulation_mae_db": 0.37450295023528174, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0060731127146598, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.14394050861750501, "3000–6000 Hz_40ms_envelope_correlation": 0.9814429805565716, "3000–6000 Hz_40ms_modulation_mae_db": 0.35515425510182297, "mr_stft_db": 6.124215959760284, "band_mae_db": 0.5306839440003494, "rms_gain_error_db": 0.7451442987877752, "output_peak": 0.5944408725674648}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.8091230980658, "spectral_convergence": 0.7011570319205566, "spectral_convergence_status": "ok", "amplitude_floor": 1.2198198453862723e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.5533114022419925, "spectral_convergence": 0.2819325205706382, "spectral_convergence_status": "ok", "amplitude_floor": 7.927151445605076e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.0102133789730585, "spectral_convergence": 0.3389209196480838, "spectral_convergence_status": "ok", "amplitude_floor": 6.454878545778745e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00013983470011737857, "output_power": 0.00014393419852163614, "delta_db": 0.12549038446704985, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00042159782923140474, "output_power": 0.0004902618559918196, "delta_db": 0.6552973893309161, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0017365691446010552, "output_power": 0.0021177293964695218, "delta_db": 0.8617838495254417, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0006597865121791135, "output_power": 0.000791774319253402, "delta_db": 0.7919797840562753, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.0001856647778551521, "output_power": 0.00020178764704786533, "delta_db": 0.36165053959436966, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.000145726779943178, "output_power": 0.00015934173449235189, "delta_db": 0.3879017170280436, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5306839440003494, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 0.7451442987877752, "pitch": null, "output_rms_dbfs": -24.08397963083441, "output_peak": 0.5944408725674648, "reference_rms_dbfs": -24.82912392962218, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.124215959760284}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00017345649672524565, "log_rms_mae_db": 1.5236446610602021, "rms_envelope_nrmse": 0.17142823187405234, "envelope_correlation": 0.9799318008680328, "rms_gain_error_db": 0.7453353490476131, "reference_mean_power": 0.003291322552458033, "output_mean_power": 0.003907544931488105}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.4377088226614287e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006890430591645026, "output_power": 0.0008152606923119308, "reference_active": true, "delta_db": 0.7305014101471494}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.716540294963627e-05, "output_power": 0.00010424601146594632, "reference_active": true, "delta_db": 0.30547791466130286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.855953967229281e-05, "output_power": 2.806719404023523e-05, "reference_active": true, "delta_db": -0.07552206014755213}], "active_band_count": 3, "active_band_mae_db": 0.3705004616520015}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00015649665907765232, "log_rms_mae_db": 1.0221685671134961, "rms_envelope_nrmse": 0.10822210423058422, "envelope_correlation": 0.9962101271429572, "rms_gain_error_db": 0.7463559545561507, "reference_mean_power": 0.003302205038848876, "output_mean_power": 0.003921386336002406}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.335462563779621e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006377316706455328, "output_power": 0.0007611332778821319, "reference_active": true, "delta_db": 0.768227251655952}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.741968368909027e-05, "output_power": 5.0494698426518966e-05, "reference_active": true, "delta_db": 0.272871300630102}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7622617238454117e-06, "output_power": 1.3917509457123778e-06, "reference_active": true, "delta_db": -1.0250888340429463}], "active_band_count": 3, "active_band_mae_db": 0.6887291287763334}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00014746435244218596, "log_rms_mae_db": 1.7665813446613192, "rms_envelope_nrmse": 0.16694283541185026, "envelope_correlation": 0.9861527401280722, "rms_gain_error_db": 0.8268385090186308, "reference_mean_power": 0.001906578877331925, "output_mean_power": 0.0023064212362344898}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 8.936060821906154e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045087668804334914, "output_power": 0.0005415867852636378, "reference_active": true, "delta_db": 0.7961027771546565}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.461610811295139e-05, "output_power": 6.486621308149656e-05, "reference_active": true, "delta_db": 0.7469779483108514}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.1974528143292685e-06, "output_power": 8.035558094983678e-06, "reference_active": true, "delta_db": 1.1280281641958054}], "active_band_count": 3, "active_band_mae_db": 0.8903696298871044}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00013246473438636124, "log_rms_mae_db": 1.208812932615596, "rms_envelope_nrmse": 0.11585959285313456, "envelope_correlation": 0.997125140834917, "rms_gain_error_db": 0.8277146104263545, "reference_mean_power": 0.0019129188838057705, "output_mean_power": 0.0023145577192146076}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 8.520895123968316e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004268128119473447, "output_power": 0.0005154477544171478, "reference_active": true, "delta_db": 0.819472044533093}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.526792747799959e-05, "output_power": 3.178424556439718e-05, "reference_active": true, "delta_db": 0.9964228566352964}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.02211562944459e-07, "output_power": 5.174085144871557e-07, "reference_active": true, "delta_db": 0.12946863489771565}], "active_band_count": 3, "active_band_mae_db": 0.648454512022035}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.50803654710503e-05, "log_rms_mae_db": 1.702832850282129, "rms_envelope_nrmse": 0.189303152684545, "envelope_correlation": 0.979547158331353, "rms_gain_error_db": 0.7747847854122976, "reference_mean_power": 0.0007310287661738595, "output_mean_power": 0.0008738018209889523}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.3756975679714823e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00020778711952139343, "output_power": 0.00025208204576928527, "reference_active": true, "delta_db": 0.8392329204017133}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.101230284766401e-05, "output_power": 4.486899075238267e-05, "reference_active": true, "delta_db": 0.39032145286157055}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.417981343795666e-05, "output_power": 1.2632045494569324e-05, "reference_active": true, "delta_db": -0.5019683580262272}], "active_band_count": 3, "active_band_mae_db": 0.5771742437631704}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.582207074769948e-05, "log_rms_mae_db": 1.0679009353575315, "rms_envelope_nrmse": 0.12265998686937944, "envelope_correlation": 0.9948166938814251, "rms_gain_error_db": 0.775824618741483, "reference_mean_power": 0.000733846607513954, "output_mean_power": 0.0008773800452057353}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.9980296656839734e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00019557456921819033, "output_power": 0.00023709129821787356, "reference_active": true, "delta_db": 0.8360323236813726}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.6985486820992048e-05, "output_power": 1.801858004010783e-05, "reference_active": true, "delta_db": 0.2564256458027402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.92201473631864e-07, "output_power": 6.57935770755813e-07, "reference_active": true, "delta_db": -1.784163688910399}], "active_band_count": 3, "active_band_mae_db": 0.9588738861315038}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00011116745339746556, "log_rms_mae_db": 1.7020294069173367, "rms_envelope_nrmse": 0.25863970993582525, "envelope_correlation": 0.9442519113088492, "rms_gain_error_db": 0.3788185010165622, "reference_mean_power": 0.00017647199152424748, "output_mean_power": 0.00019255625770929645}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.117031117126669e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.030239739448223e-05, "output_power": 4.3023904924809014e-05, "reference_active": true, "delta_db": 0.28378944181145654}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.8894394185439072e-05, "output_power": 3.122059223726778e-05, "reference_active": true, "delta_db": 0.3362754387035991}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 7.877164672472975e-06, "output_power": 7.014965306261474e-06, "reference_active": true, "delta_db": -0.5034439701907896}], "active_band_count": 3, "active_band_mae_db": 0.37450295023528174}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 6.815724618424621e-05, "log_rms_mae_db": 1.0060731127146598, "rms_envelope_nrmse": 0.14394050861750501, "envelope_correlation": 0.9814429805565716, "rms_gain_error_db": 0.3790676975352059, "reference_mean_power": 0.0001772156182174664, "output_mean_power": 0.00019337875678608737}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.0001638026368737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.251019833717301e-05, "output_power": 4.5265220690845746e-05, "reference_active": true, "delta_db": 0.2727151088449306}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.608059024163373e-05, "output_power": 1.7103893455990004e-05, "reference_active": true, "delta_db": 0.26792996974507666}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.447224316204941e-07, "output_power": 5.71335239826712e-07, "reference_active": true, "delta_db": -0.5248176867154617}], "active_band_count": 3, "active_band_mae_db": 0.35515425510182297}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9225730788048278, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "67ecfcf90ff939dbbee725e2a6ac6fd177c3ea33fe562e3a6a94126a9f43f3ce", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.979547158331353, "standard_deviation": null, "minimum": 0.979547158331353, "maximum": 0.979547158331353, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.702832850282129, "standard_deviation": null, "minimum": 1.702832850282129, "maximum": 1.702832850282129, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5771742437631704, "standard_deviation": null, "minimum": 0.5771742437631704, "maximum": 0.5771742437631704, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.189303152684545, "standard_deviation": null, "minimum": 0.189303152684545, "maximum": 0.189303152684545, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9948166938814251, "standard_deviation": null, "minimum": 0.9948166938814251, "maximum": 0.9948166938814251, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0679009353575315, "standard_deviation": null, "minimum": 1.0679009353575315, "maximum": 1.0679009353575315, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.9588738861315038, "standard_deviation": null, "minimum": 0.9588738861315038, "maximum": 0.9588738861315038, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12265998686937944, "standard_deviation": null, "minimum": 0.12265998686937944, "maximum": 0.12265998686937944, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9442519113088492, "standard_deviation": null, "minimum": 0.9442519113088492, "maximum": 0.9442519113088492, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7020294069173367, "standard_deviation": null, "minimum": 1.7020294069173367, "maximum": 1.7020294069173367, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.37450295023528174, "standard_deviation": null, "minimum": 0.37450295023528174, "maximum": 0.37450295023528174, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25863970993582525, "standard_deviation": null, "minimum": 0.25863970993582525, "maximum": 0.25863970993582525, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9814429805565716, "standard_deviation": null, "minimum": 0.9814429805565716, "maximum": 0.9814429805565716, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0060731127146598, "standard_deviation": null, "minimum": 1.0060731127146598, "maximum": 1.0060731127146598, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.35515425510182297, "standard_deviation": null, "minimum": 0.35515425510182297, "maximum": 0.35515425510182297, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.14394050861750501, "standard_deviation": null, "minimum": 0.14394050861750501, "maximum": 0.14394050861750501, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9861527401280722, "standard_deviation": null, "minimum": 0.9861527401280722, "maximum": 0.9861527401280722, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.7665813446613192, "standard_deviation": null, "minimum": 1.7665813446613192, "maximum": 1.7665813446613192, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.8903696298871044, "standard_deviation": null, "minimum": 0.8903696298871044, "maximum": 0.8903696298871044, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16694283541185026, "standard_deviation": null, "minimum": 0.16694283541185026, "maximum": 0.16694283541185026, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.997125140834917, "standard_deviation": null, "minimum": 0.997125140834917, "maximum": 0.997125140834917, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.208812932615596, "standard_deviation": null, "minimum": 1.208812932615596, "maximum": 1.208812932615596, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.648454512022035, "standard_deviation": null, "minimum": 0.648454512022035, "maximum": 0.648454512022035, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11585959285313456, "standard_deviation": null, "minimum": 0.11585959285313456, "maximum": 0.11585959285313456, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5306839440003494, "standard_deviation": null, "minimum": 0.5306839440003494, "maximum": 0.5306839440003494, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.124215959760284, "standard_deviation": null, "minimum": 6.124215959760284, "maximum": 6.124215959760284, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5944408725674648, "standard_deviation": null, "minimum": 0.5944408725674648, "maximum": 0.5944408725674648, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 0.7451442987877752, "standard_deviation": null, "minimum": 0.7451442987877752, "maximum": 0.7451442987877752, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9799318008680328, "standard_deviation": null, "minimum": 0.9799318008680328, "maximum": 0.9799318008680328, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5236446610602021, "standard_deviation": null, "minimum": 1.5236446610602021, "maximum": 1.5236446610602021, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.3705004616520015, "standard_deviation": null, "minimum": 0.3705004616520015, "maximum": 0.3705004616520015, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.17142823187405234, "standard_deviation": null, "minimum": 0.17142823187405234, "maximum": 0.17142823187405234, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9962101271429572, "standard_deviation": null, "minimum": 0.9962101271429572, "maximum": 0.9962101271429572, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0221685671134961, "standard_deviation": null, "minimum": 1.0221685671134961, "maximum": 1.0221685671134961, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.6887291287763334, "standard_deviation": null, "minimum": 0.6887291287763334, "maximum": 0.6887291287763334, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.10822210423058422, "standard_deviation": null, "minimum": 0.10822210423058422, "maximum": 0.10822210423058422, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8463"}, {"case": "librispeech_8463-294828-0014", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.5060665607452393, "predicted_mos": 2.5060665607452393, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 91600, "input_sample_rate": 16000, "input_seconds": 5.725, "input_float32_sha256": "8b4194f2787ff23c79820eab24786f456ff18f439959b1997164cd6390cff6c7", "internal_samples": 91600, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.819706300273538}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.4640708430781826, "全帯域_10ms_rms_envelope_nrmse": 0.21469430070934115, "全帯域_10ms_envelope_correlation": 0.9818651986913043, "全帯域_10ms_modulation_mae_db": 1.1200287191786045, "全帯域_40ms_log_rms_mae_db": 1.0157828873182702, "全帯域_40ms_rms_envelope_nrmse": 0.18395366255337822, "全帯域_40ms_envelope_correlation": 0.990804143222257, "全帯域_40ms_modulation_mae_db": 0.8851381360208261, "300–1000 Hz_10ms_log_rms_mae_db": 2.010409220265798, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.26013496293978255, "300–1000 Hz_10ms_envelope_correlation": 0.9810028851758157, "300–1000 Hz_10ms_modulation_mae_db": 1.351115855570171, "300–1000 Hz_40ms_log_rms_mae_db": 1.4192049392611916, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2332102741303438, "300–1000 Hz_40ms_envelope_correlation": 0.9885102609572981, "300–1000 Hz_40ms_modulation_mae_db": 0.9738000802374401, "1000–3000 Hz_10ms_log_rms_mae_db": 1.592638047849142, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.17865200434276346, "1000–3000 Hz_10ms_envelope_correlation": 0.9835958523225675, "1000–3000 Hz_10ms_modulation_mae_db": 0.9056116157648, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0291795803794, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.13245377008905504, "1000–3000 Hz_40ms_envelope_correlation": 0.9935242501545223, "1000–3000 Hz_40ms_modulation_mae_db": 0.8109053651989043, "3000–6000 Hz_10ms_log_rms_mae_db": 1.51950766045723, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20293902610696335, "3000–6000 Hz_10ms_envelope_correlation": 0.9668025600305126, "3000–6000 Hz_10ms_modulation_mae_db": 0.3332622507606557, "3000–6000 Hz_40ms_log_rms_mae_db": 0.917375177519951, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11152158244358662, "3000–6000 Hz_40ms_envelope_correlation": 0.9899284229033766, "3000–6000 Hz_40ms_modulation_mae_db": 0.3390904930909813, "mr_stft_db": 5.991455000218569, "band_mae_db": 0.8121074964081214, "rms_gain_error_db": 1.2113866762231438, "output_peak": 0.8770002141121868}, "full_scale": {"sample_count": 91600, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 91600, "output_samples": 91600, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.4481139697429795, "spectral_convergence": 0.5804882888303009, "spectral_convergence_status": "ok", "amplitude_floor": 1.4112644553101518e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.553046050206782, "spectral_convergence": 0.3449605559581625, "spectral_convergence_status": "ok", "amplitude_floor": 1.1975531593072457e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.973204980705947, "spectral_convergence": 0.38909505954522117, "spectral_convergence_status": "ok", "amplitude_floor": 1.114415465542996e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 5.564075826256522e-05, "output_power": 5.4843946895732156e-05, "delta_db": -0.06264338200772059, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0010000129923479963, "output_power": 0.0015651269740905576, "delta_db": 1.945439338613471, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0018259392170112694, "output_power": 0.0023706945356810104, "delta_db": 1.1338928230211098, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0005127219920465177, "output_power": 0.0006307478321511915, "delta_db": 0.8997382062739352, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00033320627903786135, "output_power": 0.00037274522262132313, "delta_db": 0.4869890913538695, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.0002817706196064759, "output_power": 0.0003049930343303427, "delta_db": 0.34394213717862254, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.8121074964081214, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.2113866762231438, "pitch": null, "output_rms_dbfs": -22.75793660302499, "output_peak": 0.8770002141121868, "reference_rms_dbfs": -23.96932327924813, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.991455000218569}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 91600, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00020825955187282386, "log_rms_mae_db": 1.4640708430781826, "rms_envelope_nrmse": 0.21469430070934115, "envelope_correlation": 0.9818651986913043, "rms_gain_error_db": 1.2114199753049988, "reference_mean_power": 0.004012756910556895, "output_mean_power": 0.005303772032218026}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 2.2120307480210933e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008889321025104637, "output_power": 0.0011695786340234153, "reference_active": true, "delta_db": 1.191608355458025}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00022980725896418495, "output_power": 0.0002817539440894789, "reference_active": true, "delta_db": 0.8850626145689313}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1131932747773025e-05, "output_power": 5.527396711203074e-05, "reference_active": true, "delta_db": 1.2834151875088566}], "active_band_count": 3, "active_band_mae_db": 1.1200287191786045}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.0001984160011662782, "log_rms_mae_db": 1.0157828873182702, "rms_envelope_nrmse": 0.18395366255337822, "envelope_correlation": 0.990804143222257, "rms_gain_error_db": 1.2114710852445119, "reference_mean_power": 0.0040337408532874226, "output_mean_power": 0.005331569835149423}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 2.0115882799069226e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008024963383362849, "output_power": 0.0010581679938402179, "reference_active": true, "delta_db": 1.2011156192661028}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011463438169489824, "output_power": 0.00013600366106279773, "reference_active": true, "delta_db": 0.7423570637989632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.4945953602736586e-06, "output_power": 4.1170967084595385e-06, "reference_active": true, "delta_db": 0.7119417249974127}], "active_band_count": 3, "active_band_mae_db": 0.8851381360208261}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 0.00019054595180557928, "log_rms_mae_db": 2.010409220265798, "rms_envelope_nrmse": 0.26013496293978255, "envelope_correlation": 0.9810028851758157, "rms_gain_error_db": 1.5187151387551248, "reference_mean_power": 0.0025757888112863028, "output_mean_power": 0.0036541112573314654}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.7595886306573375e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007279305516452219, "output_power": 0.0009759231139519667, "reference_active": true, "delta_db": 1.2732565669736333}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00019010480993234813, "output_power": 0.0002514840647094048, "reference_active": true, "delta_db": 1.2151736589303286}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6294378423488108e-05, "output_power": 3.770115393160344e-05, "reference_active": true, "delta_db": 1.564917340806551}], "active_band_count": 3, "active_band_mae_db": 1.351115855570171}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 0.00018019065505981728, "log_rms_mae_db": 1.4192049392611916, "rms_envelope_nrmse": 0.2332102741303438, "envelope_correlation": 0.9885102609572981, "rms_gain_error_db": 1.5187206617173123, "reference_mean_power": 0.0025893611436675294, "output_mean_power": 0.0036733701519298108}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.6269545971301712e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0006780414214466183, "output_power": 0.0009074465443340739, "reference_active": true, "delta_db": 1.2656482547479961}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.548437806715927e-05, "output_power": 0.00012311626658699666, "reference_active": true, "delta_db": 1.103831136661907}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9327510117604047e-06, "output_power": 3.3301767025128627e-06, "reference_active": true, "delta_db": 0.5519208493024177}], "active_band_count": 3, "active_band_mae_db": 0.9738000802374401}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.327720993461315e-05, "log_rms_mae_db": 1.592638047849142, "rms_envelope_nrmse": 0.17865200434276346, "envelope_correlation": 0.9835958523225675, "rms_gain_error_db": 0.7911111921503435, "reference_mean_power": 0.0007122971869413385, "output_mean_power": 0.0008546186153564298}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 4.689927289528067e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00023362430343497353, "output_power": 0.0003067363701594709, "reference_active": true, "delta_db": 1.1824725442494246}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.797902128989837e-05, "output_power": 4.772553877559875e-05, "reference_active": true, "delta_db": 0.9920707097942252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.042184778585592e-05, "output_power": 1.1807933135142693e-05, "reference_active": true, "delta_db": 0.5422915932507499}], "active_band_count": 3, "active_band_mae_db": 0.9056116157648}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 8.747046012456969e-05, "log_rms_mae_db": 1.0291795803794, "rms_envelope_nrmse": 0.13245377008905504, "envelope_correlation": 0.9935242501545223, "rms_gain_error_db": 0.7911800955032845, "reference_mean_power": 0.0007159674763911086, "output_mean_power": 0.0008590358805676279}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 4.328159646639031e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00021700970165722394, "output_power": 0.0002864904755991844, "reference_active": true, "delta_db": 1.2063103849084384}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.702508733600319e-05, "output_power": 2.098588323295441e-05, "reference_active": true, "delta_db": 0.9083790363332567}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.679451250427792e-07, "output_power": 9.338883025123994e-07, "reference_active": true, "delta_db": 0.31802667435501764}], "active_band_count": 3, "active_band_mae_db": 0.8109053651989043}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1144, "rms_floor": 9.039773028293347e-05, "log_rms_mae_db": 1.51950766045723, "rms_envelope_nrmse": 0.20293902610696335, "envelope_correlation": 0.9668025600305126, "rms_gain_error_db": 0.36182099188676786, "reference_mean_power": 0.0003237352885604671, "output_mean_power": 0.00035186181709966557}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1144, "frequency_resolution_hz": 0.17482517482517484, "power_floor": 1.2480235329901466e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.70846061617512e-05, "output_power": 5.956298315059155e-05, "reference_active": true, "delta_db": 0.18457431700765864}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9595191589145483e-05, "output_power": 1.9501439850142726e-05, "reference_active": true, "delta_db": -0.020828362679861594}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.074204859780537e-06, "output_power": 6.092652440073391e-06, "reference_active": true, "delta_db": 0.7943840725944469}], "active_band_count": 3, "active_band_mae_db": 0.3332622507606557}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 91600, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1138, "rms_floor": 6.399293414479695e-05, "log_rms_mae_db": 0.917375177519951, "rms_envelope_nrmse": 0.11152158244358662, "envelope_correlation": 0.9899284229033766, "rms_gain_error_db": 0.3619656838220604, "reference_mean_power": 0.0003253818281247376, "output_mean_power": 0.00035366319273055286}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1138, "frequency_resolution_hz": 0.17574692442882248, "power_floor": 1.1005868407421181e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.4497636322434906e-05, "output_power": 5.714769160189083e-05, "reference_active": true, "delta_db": 0.2062102597357329}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 9.385752117634814e-06, "output_power": 9.229132769453822e-06, "reference_active": true, "delta_db": -0.0730818634315053}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7591737704275347e-07, "output_power": 4.4554373102525274e-07, "reference_active": true, "delta_db": 0.737979356105706}], "active_band_count": 3, "active_band_mae_db": 0.3390904930909813}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8767157136081088, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "62c9b6b56b2ff7cc3ad81895200569a5e1c9250955032a45875a82f8427b9fed", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9835958523225675, "standard_deviation": null, "minimum": 0.9835958523225675, "maximum": 0.9835958523225675, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.592638047849142, "standard_deviation": null, "minimum": 1.592638047849142, "maximum": 1.592638047849142, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9056116157648, "standard_deviation": null, "minimum": 0.9056116157648, "maximum": 0.9056116157648, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.17865200434276346, "standard_deviation": null, "minimum": 0.17865200434276346, "maximum": 0.17865200434276346, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9935242501545223, "standard_deviation": null, "minimum": 0.9935242501545223, "maximum": 0.9935242501545223, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0291795803794, "standard_deviation": null, "minimum": 1.0291795803794, "maximum": 1.0291795803794, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8109053651989043, "standard_deviation": null, "minimum": 0.8109053651989043, "maximum": 0.8109053651989043, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13245377008905504, "standard_deviation": null, "minimum": 0.13245377008905504, "maximum": 0.13245377008905504, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9668025600305126, "standard_deviation": null, "minimum": 0.9668025600305126, "maximum": 0.9668025600305126, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.51950766045723, "standard_deviation": null, "minimum": 1.51950766045723, "maximum": 1.51950766045723, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.3332622507606557, "standard_deviation": null, "minimum": 0.3332622507606557, "maximum": 0.3332622507606557, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20293902610696335, "standard_deviation": null, "minimum": 0.20293902610696335, "maximum": 0.20293902610696335, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9899284229033766, "standard_deviation": null, "minimum": 0.9899284229033766, "maximum": 0.9899284229033766, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.917375177519951, "standard_deviation": null, "minimum": 0.917375177519951, "maximum": 0.917375177519951, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.3390904930909813, "standard_deviation": null, "minimum": 0.3390904930909813, "maximum": 0.3390904930909813, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11152158244358662, "standard_deviation": null, "minimum": 0.11152158244358662, "maximum": 0.11152158244358662, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9810028851758157, "standard_deviation": null, "minimum": 0.9810028851758157, "maximum": 0.9810028851758157, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.010409220265798, "standard_deviation": null, "minimum": 2.010409220265798, "maximum": 2.010409220265798, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.351115855570171, "standard_deviation": null, "minimum": 1.351115855570171, "maximum": 1.351115855570171, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26013496293978255, "standard_deviation": null, "minimum": 0.26013496293978255, "maximum": 0.26013496293978255, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9885102609572981, "standard_deviation": null, "minimum": 0.9885102609572981, "maximum": 0.9885102609572981, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4192049392611916, "standard_deviation": null, "minimum": 1.4192049392611916, "maximum": 1.4192049392611916, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.9738000802374401, "standard_deviation": null, "minimum": 0.9738000802374401, "maximum": 0.9738000802374401, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2332102741303438, "standard_deviation": null, "minimum": 0.2332102741303438, "maximum": 0.2332102741303438, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.8121074964081214, "standard_deviation": null, "minimum": 0.8121074964081214, "maximum": 0.8121074964081214, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.991455000218569, "standard_deviation": null, "minimum": 5.991455000218569, "maximum": 5.991455000218569, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.8770002141121868, "standard_deviation": null, "minimum": 0.8770002141121868, "maximum": 0.8770002141121868, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.2113866762231438, "standard_deviation": null, "minimum": 1.2113866762231438, "maximum": 1.2113866762231438, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9818651986913043, "standard_deviation": null, "minimum": 0.9818651986913043, "maximum": 0.9818651986913043, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.4640708430781826, "standard_deviation": null, "minimum": 1.4640708430781826, "maximum": 1.4640708430781826, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1200287191786045, "standard_deviation": null, "minimum": 1.1200287191786045, "maximum": 1.1200287191786045, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.21469430070934115, "standard_deviation": null, "minimum": 0.21469430070934115, "maximum": 0.21469430070934115, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.990804143222257, "standard_deviation": null, "minimum": 0.990804143222257, "maximum": 0.990804143222257, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0157828873182702, "standard_deviation": null, "minimum": 1.0157828873182702, "maximum": 1.0157828873182702, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8851381360208261, "standard_deviation": null, "minimum": 0.8851381360208261, "maximum": 0.8851381360208261, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18395366255337822, "standard_deviation": null, "minimum": 0.18395366255337822, "maximum": 0.18395366255337822, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8463"}, {"case": "librispeech_8463-294828-0022", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 2.3326575756073, "predicted_mos": 2.3326575756073, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 37680, "input_sample_rate": 16000, "input_seconds": 2.355, "input_float32_sha256": "5955a93b001af01d432105d82df26708b492882f7cbe27023771613c4f315354", "internal_samples": 37680, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 2, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.666546799708158}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.6200256682503889, "全帯域_10ms_rms_envelope_nrmse": 0.3042558276866359, "全帯域_10ms_envelope_correlation": 0.9724610855436365, "全帯域_10ms_modulation_mae_db": 1.1964639220055318, "全帯域_40ms_log_rms_mae_db": 1.0743947760848152, "全帯域_40ms_rms_envelope_nrmse": 0.2773221788566475, "全帯域_40ms_envelope_correlation": 0.9831352422940586, "全帯域_40ms_modulation_mae_db": 1.279738273243825, "300–1000 Hz_10ms_log_rms_mae_db": 1.9118014212956669, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.3652173006527835, "300–1000 Hz_10ms_envelope_correlation": 0.975215084386892, "300–1000 Hz_10ms_modulation_mae_db": 1.1803782039838173, "300–1000 Hz_40ms_log_rms_mae_db": 1.2989529287066335, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.340016039661847, "300–1000 Hz_40ms_envelope_correlation": 0.9852055724576568, "300–1000 Hz_40ms_modulation_mae_db": 1.4207467587168106, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6875921021009275, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.29666019830372853, "1000–3000 Hz_10ms_envelope_correlation": 0.9552423159950124, "1000–3000 Hz_10ms_modulation_mae_db": 0.5645627007053551, "1000–3000 Hz_40ms_log_rms_mae_db": 1.1707058669518695, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.23989021817165138, "1000–3000 Hz_40ms_envelope_correlation": 0.9741926506791787, "1000–3000 Hz_40ms_modulation_mae_db": 0.8628417244878235, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5970829781063138, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20263772388301798, "3000–6000 Hz_10ms_envelope_correlation": 0.9741481469131148, "3000–6000 Hz_10ms_modulation_mae_db": 1.1779495765578039, "3000–6000 Hz_40ms_log_rms_mae_db": 0.8333132794484215, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.11590051689310092, "3000–6000 Hz_40ms_envelope_correlation": 0.993009203185563, "3000–6000 Hz_40ms_modulation_mae_db": 1.3385620860127265, "mr_stft_db": 6.177529818292499, "band_mae_db": 1.025407540784968, "rms_gain_error_db": 1.7689697786535783, "output_peak": 0.6762865567355035}, "full_scale": {"sample_count": 37680, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 37680, "output_samples": 37680, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.758161778246726, "spectral_convergence": 0.6045270392334, "spectral_convergence_status": "ok", "amplitude_floor": 1.2997261015401668e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6428231288047295, "spectral_convergence": 0.4673221363990902, "spectral_convergence_status": "ok", "amplitude_floor": 1.1046780177944747e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 6.131604547826044, "spectral_convergence": 0.5390908020786719, "spectral_convergence_status": "ok", "amplitude_floor": 9.696590055390529e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 7.751691690693756e-05, "output_power": 8.49896168717522e-05, "delta_db": 0.39969380349779826, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0016017256752476519, "output_power": 0.0028850798901164606, "delta_db": 2.5556970635039358, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006838056632308089, "output_power": 0.000823913768734273, "delta_db": 0.8094906730542506, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0003361315194748372, "output_power": 0.0005188300358974166, "delta_db": 1.8851587142121167, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00035354415012418854, "output_power": 0.00035690937379852656, "delta_db": 0.0411429869345667, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00021020525507420865, "output_power": 0.0002337597689996526, "delta_db": 0.4612620035071411, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 1.025407540784968, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.7689697786535783, "pitch": null, "output_rms_dbfs": -23.094953743484588, "output_peak": 0.6762865567355035, "reference_rms_dbfs": -24.863923522138165, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.177529818292499}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 37680, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00018834951637393047, "log_rms_mae_db": 1.6200256682503889, "rms_envelope_nrmse": 0.3042558276866359, "envelope_correlation": 0.9724610855436365, "rms_gain_error_db": 1.7689824241067025, "reference_mean_power": 0.00326985830085679, "output_mean_power": 0.004913909744170282}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 9.479011996203747e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0005558024994761568, "output_power": 0.0006936621539285496, "reference_active": true, "delta_db": 0.9622750463063274}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00011456565813557757, "output_power": 0.00015485717789773699, "reference_active": true, "delta_db": 1.3087688615523632}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.821714301732997e-05, "output_power": 2.8211321597834576e-05, "reference_active": true, "delta_db": -1.318347858157905}], "active_band_count": 3, "active_band_mae_db": 1.1964639220055318}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00017471448177334785, "log_rms_mae_db": 1.0743947760848152, "rms_envelope_nrmse": 0.2773221788566475, "envelope_correlation": 0.9831352422940586, "rms_gain_error_db": 1.769018959092815, "reference_mean_power": 0.0033120735889998944, "output_mean_power": 0.004977392321234755}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 7.763101583949748e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004881183543297566, "output_power": 0.0006143226533750841, "reference_active": true, "delta_db": 0.9987139198785029}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.883974713754685e-05, "output_power": 8.093874183862553e-05, "reference_active": true, "delta_db": 1.38485651479599}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7824879244994614e-06, "output_power": 1.2748590106781897e-06, "reference_active": true, "delta_db": -1.455644385056982}], "active_band_count": 3, "active_band_mae_db": 1.279738273243825}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.00017587844894922502, "log_rms_mae_db": 1.9118014212956669, "rms_envelope_nrmse": 0.3652173006527835, "envelope_correlation": 0.975215084386892, "rms_gain_error_db": 2.242045959226004, "reference_mean_power": 0.002090015296145697, "output_mean_power": 0.0035023057770604527}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 6.927253047980857e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041075223634869497, "output_power": 0.0005750345781731612, "reference_active": true, "delta_db": 1.4611402397862396}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.156045953139752e-05, "output_power": 0.0001260951543448646, "reference_active": true, "delta_db": 1.8921873354905836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2217147797459886e-05, "output_power": 2.3198986266834363e-05, "reference_active": true, "delta_db": 0.18780703667462872}], "active_band_count": 3, "active_band_mae_db": 1.1803782039838173}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 0.00016316895420677254, "log_rms_mae_db": 1.2989529287066335, "rms_envelope_nrmse": 0.340016039661847, "envelope_correlation": 0.9852055724576568, "rms_gain_error_db": 2.2420504730033772, "reference_mean_power": 0.002117036355483854, "output_mean_power": 0.0035475895214254014}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 5.84659163450443e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037072755253225205, "output_power": 0.0005257623831423534, "reference_active": true, "delta_db": 1.5173464623602873}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.09979708848843e-05, "output_power": 6.447386843532954e-05, "reference_active": true, "delta_db": 1.9662136599558941}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.223097583568823e-06, "output_power": 1.4632856833985043e-06, "reference_active": true, "delta_db": 0.7786801538342503}], "active_band_count": 3, "active_band_mae_db": 1.4207467587168106}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 0.0001038394087983997, "log_rms_mae_db": 1.6875921021009275, "rms_envelope_nrmse": 0.29666019830372853, "envelope_correlation": 0.9552423159950124, "rms_gain_error_db": 1.206667882273426, "reference_mean_power": 0.0005521302842545539, "output_mean_power": 0.0007289678200343362}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 2.3748882259523475e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000136307073642641, "output_power": 0.00015387103386906972, "reference_active": true, "delta_db": 0.5263847775653905}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.293151120982887e-05, "output_power": 3.628301201424076e-05, "reference_active": true, "delta_db": 0.4209167188571875}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.586128694071059e-05, "output_power": 1.3356718137341174e-05, "reference_active": true, "delta_db": -0.7463866056934874}], "active_band_count": 3, "active_band_mae_db": 0.5645627007053551}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 8.776193320418374e-05, "log_rms_mae_db": 1.1707058669518695, "rms_envelope_nrmse": 0.23989021817165138, "envelope_correlation": 0.9741926506791787, "rms_gain_error_db": 1.2067532204695368, "reference_mean_power": 0.0005592400075632049, "output_mean_power": 0.0007383691702968171}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.9436382041482247e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012615724913962246, "output_power": 0.00014233614208283097, "reference_active": true, "delta_db": 0.5240297978816414}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.519569809811637e-05, "output_power": 1.6174819090527017e-05, "reference_active": true, "delta_db": 0.271187753578169}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.080674445150989e-07, "output_power": 6.008794392635894e-07, "reference_active": true, "delta_db": -1.7933076220036601}], "active_band_count": 3, "active_band_mae_db": 0.8628417244878235}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 470, "rms_floor": 6.310171076800209e-05, "log_rms_mae_db": 1.5970829781063138, "rms_envelope_nrmse": 0.20263772388301798, "envelope_correlation": 0.9741481469131148, "rms_gain_error_db": 0.5375850091375077, "reference_mean_power": 0.00026303793349419837, "output_mean_power": 0.0002976986634488692}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 470, "frequency_resolution_hz": 0.425531914893617, "power_floor": 1.3356760090821468e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.8507771208283054e-05, "output_power": 6.873663355584437e-05, "reference_active": true, "delta_db": 0.6997470388667525}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1981263949052112e-05, "output_power": 1.4123401232409188e-05, "reference_active": true, "delta_db": 0.714366613375095}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.256106518668102e-06, "output_power": 5.681393055161618e-06, "reference_active": true, "delta_db": -2.119735077431564}], "active_band_count": 3, "active_band_mae_db": 1.1779495765578039}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 37680, "uncovered_tail_samples": 0, "metrics": {"frame_count": 464, "rms_floor": 5.861628173421479e-05, "log_rms_mae_db": 0.8333132794484215, "rms_envelope_nrmse": 0.11590051689310092, "envelope_correlation": 0.993009203185563, "rms_gain_error_db": 0.5376616339005067, "reference_mean_power": 0.00026641634082957415, "output_mean_power": 0.0003015275663189937}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 464, "frequency_resolution_hz": 0.4310344827586207, "power_floor": 1.157398705126346e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 5.177699204483558e-05, "output_power": 6.04135012715303e-05, "reference_active": true, "delta_db": 0.6699718930773914}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.036512028004977e-06, "output_power": 8.006933407018196e-06, "reference_active": true, "delta_db": 0.5610878215846722}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.588547226531691e-07, "output_power": 2.943311796129687e-07, "reference_active": true, "delta_db": -2.784626543376116}], "active_band_count": 3, "active_band_mae_db": 1.3385620860127265}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8457275385286233, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5b29f2df469751e1940a235f38468f6706bbdb65a494847714e56ceb2ff1fe01", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9552423159950124, "standard_deviation": null, "minimum": 0.9552423159950124, "maximum": 0.9552423159950124, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6875921021009275, "standard_deviation": null, "minimum": 1.6875921021009275, "maximum": 1.6875921021009275, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5645627007053551, "standard_deviation": null, "minimum": 0.5645627007053551, "maximum": 0.5645627007053551, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.29666019830372853, "standard_deviation": null, "minimum": 0.29666019830372853, "maximum": 0.29666019830372853, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9741926506791787, "standard_deviation": null, "minimum": 0.9741926506791787, "maximum": 0.9741926506791787, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.1707058669518695, "standard_deviation": null, "minimum": 1.1707058669518695, "maximum": 1.1707058669518695, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8628417244878235, "standard_deviation": null, "minimum": 0.8628417244878235, "maximum": 0.8628417244878235, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23989021817165138, "standard_deviation": null, "minimum": 0.23989021817165138, "maximum": 0.23989021817165138, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9741481469131148, "standard_deviation": null, "minimum": 0.9741481469131148, "maximum": 0.9741481469131148, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5970829781063138, "standard_deviation": null, "minimum": 1.5970829781063138, "maximum": 1.5970829781063138, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1779495765578039, "standard_deviation": null, "minimum": 1.1779495765578039, "maximum": 1.1779495765578039, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20263772388301798, "standard_deviation": null, "minimum": 0.20263772388301798, "maximum": 0.20263772388301798, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.993009203185563, "standard_deviation": null, "minimum": 0.993009203185563, "maximum": 0.993009203185563, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.8333132794484215, "standard_deviation": null, "minimum": 0.8333132794484215, "maximum": 0.8333132794484215, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.3385620860127265, "standard_deviation": null, "minimum": 1.3385620860127265, "maximum": 1.3385620860127265, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11590051689310092, "standard_deviation": null, "minimum": 0.11590051689310092, "maximum": 0.11590051689310092, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.975215084386892, "standard_deviation": null, "minimum": 0.975215084386892, "maximum": 0.975215084386892, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.9118014212956669, "standard_deviation": null, "minimum": 1.9118014212956669, "maximum": 1.9118014212956669, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.1803782039838173, "standard_deviation": null, "minimum": 1.1803782039838173, "maximum": 1.1803782039838173, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3652173006527835, "standard_deviation": null, "minimum": 0.3652173006527835, "maximum": 0.3652173006527835, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9852055724576568, "standard_deviation": null, "minimum": 0.9852055724576568, "maximum": 0.9852055724576568, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.2989529287066335, "standard_deviation": null, "minimum": 1.2989529287066335, "maximum": 1.2989529287066335, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.4207467587168106, "standard_deviation": null, "minimum": 1.4207467587168106, "maximum": 1.4207467587168106, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.340016039661847, "standard_deviation": null, "minimum": 0.340016039661847, "maximum": 0.340016039661847, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 1.025407540784968, "standard_deviation": null, "minimum": 1.025407540784968, "maximum": 1.025407540784968, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.177529818292499, "standard_deviation": null, "minimum": 6.177529818292499, "maximum": 6.177529818292499, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.6762865567355035, "standard_deviation": null, "minimum": 0.6762865567355035, "maximum": 0.6762865567355035, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.7689697786535783, "standard_deviation": null, "minimum": 1.7689697786535783, "maximum": 1.7689697786535783, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9724610855436365, "standard_deviation": null, "minimum": 0.9724610855436365, "maximum": 0.9724610855436365, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.6200256682503889, "standard_deviation": null, "minimum": 1.6200256682503889, "maximum": 1.6200256682503889, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1964639220055318, "standard_deviation": null, "minimum": 1.1964639220055318, "maximum": 1.1964639220055318, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.3042558276866359, "standard_deviation": null, "minimum": 0.3042558276866359, "maximum": 0.3042558276866359, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9831352422940586, "standard_deviation": null, "minimum": 0.9831352422940586, "maximum": 0.9831352422940586, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.0743947760848152, "standard_deviation": null, "minimum": 1.0743947760848152, "maximum": 1.0743947760848152, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.279738273243825, "standard_deviation": null, "minimum": 1.279738273243825, "maximum": 1.279738273243825, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.2773221788566475, "standard_deviation": null, "minimum": 0.2773221788566475, "maximum": 0.2773221788566475, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8463"}, {"case": "librispeech_8555-284447-0007", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.571512460708618, "predicted_mos": 3.571512460708618, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "eeff9350105017fb293c011047488d3d8c9c67323a7405f5dd28d2ea78835c5f", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.268222600221634}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5649914197551655, "全帯域_10ms_rms_envelope_nrmse": 0.19638867173788743, "全帯域_10ms_envelope_correlation": 0.9904047708760947, "全帯域_10ms_modulation_mae_db": 1.1483305875414156, "全帯域_40ms_log_rms_mae_db": 1.2509287066657233, "全帯域_40ms_rms_envelope_nrmse": 0.18642707199475042, "全帯域_40ms_envelope_correlation": 0.9936938369823728, "全帯域_40ms_modulation_mae_db": 1.4601272116499502, "300–1000 Hz_10ms_log_rms_mae_db": 1.775545304495761, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.21651530221651538, "300–1000 Hz_10ms_envelope_correlation": 0.9906709102700179, "300–1000 Hz_10ms_modulation_mae_db": 1.3307284086081306, "300–1000 Hz_40ms_log_rms_mae_db": 1.4874337204681787, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.20742650602076923, "300–1000 Hz_40ms_envelope_correlation": 0.9932923046334164, "300–1000 Hz_40ms_modulation_mae_db": 1.3883334294189884, "1000–3000 Hz_10ms_log_rms_mae_db": 1.56259020338132, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2158175012265078, "1000–3000 Hz_10ms_envelope_correlation": 0.9881939209088763, "1000–3000 Hz_10ms_modulation_mae_db": 1.1919534122869906, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2167126864597833, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.18783565582975403, "1000–3000 Hz_40ms_envelope_correlation": 0.9954937397246504, "1000–3000 Hz_40ms_modulation_mae_db": 1.1768671208003842, "3000–6000 Hz_10ms_log_rms_mae_db": 1.6905223314500204, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.231672337099674, "3000–6000 Hz_10ms_envelope_correlation": 0.9656670580108018, "3000–6000 Hz_10ms_modulation_mae_db": 0.27087378995043915, "3000–6000 Hz_40ms_log_rms_mae_db": 1.2754098874219484, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.15081407255453066, "3000–6000 Hz_40ms_envelope_correlation": 0.9852105739151154, "3000–6000 Hz_40ms_modulation_mae_db": 0.2297853944046191, "mr_stft_db": 5.667120952769104, "band_mae_db": 0.9741783692115304, "rms_gain_error_db": 1.3548419559678533, "output_peak": 0.32128162623458384}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.929902809483129, "spectral_convergence": 0.464107496390147, "spectral_convergence_status": "ok", "amplitude_floor": 9.671435620644006e-07}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.332616529889924, "spectral_convergence": 0.24922524971595664, "spectral_convergence_status": "ok", "amplitude_floor": 9.322400769888516e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.738843518934261, "spectral_convergence": 0.27979850348498747, "spectral_convergence_status": "ok", "amplitude_floor": 9.453157582642776e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00016624637287256398, "output_power": 0.00015805553182928937, "delta_db": -0.21942478301022916, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013640231584289317, "output_power": 0.0018653936806772892, "delta_db": 1.3594875733785148, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005975940179232537, "output_power": 0.0008897353705737777, "delta_db": 1.7285461444620527, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 1.8168186983374496e-05, "output_power": 2.3908278232699018e-05, "delta_db": 1.1923671038781338, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.463537251271816e-05, "output_power": 3.1300619610449633e-05, "delta_db": 1.0399380104507399, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 1.0343523773544458e-05, "output_power": 1.1096836238340984e-05, "delta_db": 0.30530660008951227, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9741783692115304, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.3548419559678533, "pitch": null, "output_rms_dbfs": -25.25858021620303, "output_peak": 0.32128162623458384, "reference_rms_dbfs": -26.613422172170882, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.667120952769104}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0001446111949915804, "log_rms_mae_db": 1.5649914197551655, "rms_envelope_nrmse": 0.19638867173788743, "envelope_correlation": 0.9904047708760947, "rms_gain_error_db": 1.3548434933617362, "reference_mean_power": 0.002182828397481346, "output_mean_power": 0.0029819746312682695}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.790920693994897e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003230276115117931, "output_power": 0.00044598546594811064, "reference_active": true, "delta_db": 1.4008105966832427}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.477236494661316e-05, "output_power": 7.092060164984396e-05, "reference_active": true, "delta_db": 1.997623761448874}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.548345752003965e-06, "output_power": 9.65125692687106e-06, "reference_active": true, "delta_db": 0.046557404492130014}], "active_band_count": 3, "active_band_mae_db": 1.1483305875414156}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00014111493703772434, "log_rms_mae_db": 1.2509287066657233, "rms_envelope_nrmse": 0.18642707199475042, "envelope_correlation": 0.9936938369823728, "rms_gain_error_db": 1.3548456186993652, "reference_mean_power": 0.002193693032291648, "output_mean_power": 0.002996818338954457}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.190420923463322e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00028912450266098784, "output_power": 0.0004034368265998639, "reference_active": true, "delta_db": 1.4469064031524197}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.2978791346744758e-05, "output_power": 3.8580518163406815e-05, "reference_active": true, "delta_db": 2.250408745740418}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.072854493661425e-07, "output_power": 5.189030541829579e-07, "reference_active": true, "delta_db": -0.6830664860570128}], "active_band_count": 3, "active_band_mae_db": 1.4601272116499502}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012712872255296568, "log_rms_mae_db": 1.775545304495761, "rms_envelope_nrmse": 0.21651530221651538, "envelope_correlation": 0.9906709102700179, "rms_gain_error_db": 1.4942071024822416, "reference_mean_power": 0.0014813690704446063, "output_mean_power": 0.0020897001938805057}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 7.409297669270013e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003163260619212235, "output_power": 0.00044048559507724525, "reference_active": true, "delta_db": 1.437967358607103}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.6701975061022124e-05, "output_power": 6.034223291493876e-05, "reference_active": true, "delta_db": 2.159319410695369}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.301526036454895e-06, "output_power": 6.901374274556975e-06, "reference_active": true, "delta_db": 0.39489845652191985}], "active_band_count": 3, "active_band_mae_db": 1.3307284086081306}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00012343588785799389, "log_rms_mae_db": 1.4874337204681787, "rms_envelope_nrmse": 0.20742650602076923, "envelope_correlation": 0.9932923046334164, "rms_gain_error_db": 1.4941743970792267, "reference_mean_power": 0.0014887936512772218, "output_mean_power": 0.0021001578977750536}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 6.99328908394483e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002875949770086079, "output_power": 0.0004025266691064102, "reference_active": true, "delta_db": 1.4601336291163438}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.97680708793848e-05, "output_power": 3.399133133270092e-05, "reference_active": true, "delta_db": 2.3540388527347273}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.520482053574084e-07, "output_power": 4.169672724738306e-07, "reference_active": true, "delta_db": -0.35082780640589406}], "active_band_count": 3, "active_band_mae_db": 1.3883334294189884}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.4084091617788044e-05, "log_rms_mae_db": 1.56259020338132, "rms_envelope_nrmse": 0.2158175012265078, "envelope_correlation": 0.9881939209088763, "rms_gain_error_db": 1.359257739433482, "reference_mean_power": 3.6694438258439554e-05, "output_mean_power": 5.017946394909428e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.6589829493109366e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3179349061495639e-05, "output_power": 1.9089691379995816e-05, "reference_active": true, "delta_db": 1.6090494664187176}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.9890642401085437e-06, "output_power": 4.481479343936178e-06, "reference_active": true, "delta_db": 1.7588615013908893}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.136200949313584e-07, "output_power": 7.755802466068596e-07, "reference_active": true, "delta_db": -0.20794926905136485}], "active_band_count": 3, "active_band_mae_db": 1.1919534122869906}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.151462172211685e-05, "log_rms_mae_db": 1.2167126864597833, "rms_envelope_nrmse": 0.18783565582975403, "envelope_correlation": 0.9954937397246504, "rms_gain_error_db": 1.3593141499392833, "reference_mean_power": 3.687647047180041e-05, "output_mean_power": 5.0429047124684134e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.4709890047082147e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.2743108138238886e-05, "output_power": 1.8556797271105173e-05, "reference_active": true, "delta_db": 1.6322765464935365}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.4131557808728274e-06, "output_power": 2.1281519016239902e-06, "reference_active": true, "delta_db": 1.7781258402849678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.7131268313465536e-08, "output_power": 5.873456587013568e-08, "reference_active": true, "delta_db": 0.12019897562264842}], "active_band_count": 3, "active_band_mae_db": 1.1768671208003842}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 1.703026630874025e-05, "log_rms_mae_db": 1.6905223314500204, "rms_envelope_nrmse": 0.231672337099674, "envelope_correlation": 0.9656670580108018, "rms_gain_error_db": 0.4629698985437035, "reference_mean_power": 1.1971789360385644e-05, "output_mean_power": 1.3318522748174519e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.043568625201122e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.248382965898838e-06, "output_power": 4.809509334735795e-06, "reference_active": true, "delta_db": 0.5387711314445836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.589400168019184e-06, "output_power": 2.584635756790689e-06, "reference_active": true, "delta_db": -0.007998236274181413}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.486231293405906e-07, "output_power": 6.1010871825843e-07, "reference_active": true, "delta_db": -0.26585200213255256}], "active_band_count": 3, "active_band_mae_db": 0.27087378995043915}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 1.3691778537856407e-05, "log_rms_mae_db": 1.2754098874219484, "rms_envelope_nrmse": 0.15081407255453066, "envelope_correlation": 0.9852105739151154, "rms_gain_error_db": 0.4629756507036428, "reference_mean_power": 1.2031928101806957e-05, "output_mean_power": 1.3385444360050027e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.457061965350652e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.507382710758712e-06, "output_power": 4.992463551726345e-06, "reference_active": true, "delta_db": 0.4439046840762513}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3712782786962106e-06, "output_power": 1.3078264160019565e-06, "reference_active": true, "delta_db": -0.20575491588561473}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.274779912953651e-08, "output_power": 4.2358844218525036e-08, "reference_active": true, "delta_db": -0.039696583251991324}], "active_band_count": 3, "active_band_mae_db": 0.2297853944046191}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.943394270667172, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d7b5d69084188c0435c3f430f8fe5fc63b5c045fe855cbebe96c990492b88450", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9881939209088763, "standard_deviation": null, "minimum": 0.9881939209088763, "maximum": 0.9881939209088763, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.56259020338132, "standard_deviation": null, "minimum": 1.56259020338132, "maximum": 1.56259020338132, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.1919534122869906, "standard_deviation": null, "minimum": 1.1919534122869906, "maximum": 1.1919534122869906, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2158175012265078, "standard_deviation": null, "minimum": 0.2158175012265078, "maximum": 0.2158175012265078, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9954937397246504, "standard_deviation": null, "minimum": 0.9954937397246504, "maximum": 0.9954937397246504, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2167126864597833, "standard_deviation": null, "minimum": 1.2167126864597833, "maximum": 1.2167126864597833, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1768671208003842, "standard_deviation": null, "minimum": 1.1768671208003842, "maximum": 1.1768671208003842, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18783565582975403, "standard_deviation": null, "minimum": 0.18783565582975403, "maximum": 0.18783565582975403, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9656670580108018, "standard_deviation": null, "minimum": 0.9656670580108018, "maximum": 0.9656670580108018, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.6905223314500204, "standard_deviation": null, "minimum": 1.6905223314500204, "maximum": 1.6905223314500204, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.27087378995043915, "standard_deviation": null, "minimum": 0.27087378995043915, "maximum": 0.27087378995043915, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.231672337099674, "standard_deviation": null, "minimum": 0.231672337099674, "maximum": 0.231672337099674, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9852105739151154, "standard_deviation": null, "minimum": 0.9852105739151154, "maximum": 0.9852105739151154, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.2754098874219484, "standard_deviation": null, "minimum": 1.2754098874219484, "maximum": 1.2754098874219484, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.2297853944046191, "standard_deviation": null, "minimum": 0.2297853944046191, "maximum": 0.2297853944046191, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.15081407255453066, "standard_deviation": null, "minimum": 0.15081407255453066, "maximum": 0.15081407255453066, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9906709102700179, "standard_deviation": null, "minimum": 0.9906709102700179, "maximum": 0.9906709102700179, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.775545304495761, "standard_deviation": null, "minimum": 1.775545304495761, "maximum": 1.775545304495761, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.3307284086081306, "standard_deviation": null, "minimum": 1.3307284086081306, "maximum": 1.3307284086081306, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.21651530221651538, "standard_deviation": null, "minimum": 0.21651530221651538, "maximum": 0.21651530221651538, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9932923046334164, "standard_deviation": null, "minimum": 0.9932923046334164, "maximum": 0.9932923046334164, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.4874337204681787, "standard_deviation": null, "minimum": 1.4874337204681787, "maximum": 1.4874337204681787, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.3883334294189884, "standard_deviation": null, "minimum": 1.3883334294189884, "maximum": 1.3883334294189884, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20742650602076923, "standard_deviation": null, "minimum": 0.20742650602076923, "maximum": 0.20742650602076923, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9741783692115304, "standard_deviation": null, "minimum": 0.9741783692115304, "maximum": 0.9741783692115304, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.667120952769104, "standard_deviation": null, "minimum": 5.667120952769104, "maximum": 5.667120952769104, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.32128162623458384, "standard_deviation": null, "minimum": 0.32128162623458384, "maximum": 0.32128162623458384, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.3548419559678533, "standard_deviation": null, "minimum": 1.3548419559678533, "maximum": 1.3548419559678533, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9904047708760947, "standard_deviation": null, "minimum": 0.9904047708760947, "maximum": 0.9904047708760947, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5649914197551655, "standard_deviation": null, "minimum": 1.5649914197551655, "maximum": 1.5649914197551655, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.1483305875414156, "standard_deviation": null, "minimum": 1.1483305875414156, "maximum": 1.1483305875414156, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.19638867173788743, "standard_deviation": null, "minimum": 0.19638867173788743, "maximum": 0.19638867173788743, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9936938369823728, "standard_deviation": null, "minimum": 0.9936938369823728, "maximum": 0.9936938369823728, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2509287066657233, "standard_deviation": null, "minimum": 1.2509287066657233, "maximum": 1.2509287066657233, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.4601272116499502, "standard_deviation": null, "minimum": 1.4601272116499502, "maximum": 1.4601272116499502, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18642707199475042, "standard_deviation": null, "minimum": 0.18642707199475042, "maximum": 0.18642707199475042, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8555"}, {"case": "librispeech_8555-284449-0020", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.3007524013519287, "predicted_mos": 3.3007524013519287, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 81520, "input_sample_rate": 16000, "input_seconds": 5.095, "input_float32_sha256": "c3110b8da1a8b1404c4d1182a22b36da8b793cf98e73f986fcf9d358e3f81e5a", "internal_samples": 81520, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 6.020552099682391}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.783984169277373, "全帯域_10ms_rms_envelope_nrmse": 0.24775304556080704, "全帯域_10ms_envelope_correlation": 0.980944885182259, "全帯域_10ms_modulation_mae_db": 1.3836747890195644, "全帯域_40ms_log_rms_mae_db": 1.346598655160742, "全帯域_40ms_rms_envelope_nrmse": 0.22924964866794306, "全帯域_40ms_envelope_correlation": 0.9879650014936765, "全帯域_40ms_modulation_mae_db": 1.2101891417992745, "300–1000 Hz_10ms_log_rms_mae_db": 2.1687225784021495, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.30623611370597587, "300–1000 Hz_10ms_envelope_correlation": 0.9806161866605547, "300–1000 Hz_10ms_modulation_mae_db": 1.6820352815509592, "300–1000 Hz_40ms_log_rms_mae_db": 1.7733782653175585, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.2874542077583534, "300–1000 Hz_40ms_envelope_correlation": 0.9872218937781846, "300–1000 Hz_40ms_modulation_mae_db": 1.417020397596551, "1000–3000 Hz_10ms_log_rms_mae_db": 1.5755559192509774, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.15676129576022707, "1000–3000 Hz_10ms_envelope_correlation": 0.9889503630547015, "1000–3000 Hz_10ms_modulation_mae_db": 0.5910356899373248, "1000–3000 Hz_40ms_log_rms_mae_db": 1.134747938692485, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.11564980110385331, "1000–3000 Hz_40ms_envelope_correlation": 0.9966604254347202, "1000–3000 Hz_40ms_modulation_mae_db": 0.444231862505573, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5494038876063725, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.20043924124037352, "3000–6000 Hz_10ms_envelope_correlation": 0.9692854914141983, "3000–6000 Hz_10ms_modulation_mae_db": 0.44705044644867886, "3000–6000 Hz_40ms_log_rms_mae_db": 1.0914030685853784, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12344620690300134, "3000–6000 Hz_40ms_envelope_correlation": 0.9877654848871655, "3000–6000 Hz_40ms_modulation_mae_db": 0.6189726791574555, "mr_stft_db": 5.803612797244251, "band_mae_db": 0.892767018012869, "rms_gain_error_db": 1.5917683860714877, "output_peak": 0.38249752165123496}, "full_scale": {"sample_count": 81520, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 81520, "output_samples": 81520, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.003218820537246, "spectral_convergence": 0.5066001314845803, "spectral_convergence_status": "ok", "amplitude_floor": 1.0222158521079025e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.430578231179816, "spectral_convergence": 0.3261518894717548, "spectral_convergence_status": "ok", "amplitude_floor": 7.564059658380168e-07}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.977041340015694, "spectral_convergence": 0.3594573624223621, "spectral_convergence_status": "ok", "amplitude_floor": 6.35834815641291e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 8.932339290337003e-05, "output_power": 8.899760861692365e-05, "delta_db": -0.015868739419850855, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0011480494900330098, "output_power": 0.0017071251566834653, "delta_db": 1.7230475227719042, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0005518546578574125, "output_power": 0.0008242334089871747, "delta_db": 1.7422550147145335, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 9.085077476738271e-05, "output_power": 0.0001071782381454939, "delta_db": 0.7177797831118728, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.0710168905088768e-05, "output_power": 2.5446545022770276e-05, "delta_db": 0.8944518394426018, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 2.8163123695227053e-05, "output_power": 2.992269759311657e-05, "delta_db": 0.2631992086164496, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.892767018012869, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.5917683860714877, "pitch": null, "output_rms_dbfs": -25.55501828809646, "output_peak": 0.38249752165123496, "reference_rms_dbfs": -27.14678667416795, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.803612797244251}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 81520, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.00013747273046535536, "log_rms_mae_db": 1.783984169277373, "rms_envelope_nrmse": 0.24775304556080704, "envelope_correlation": 0.980944885182259, "rms_gain_error_db": 1.5917686262406208, "reference_mean_power": 0.0019308462257719779, "output_mean_power": 0.0027856371790020065}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 6.283614364116833e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00041565087617635905, "output_power": 0.0006054455124943794, "reference_active": true, "delta_db": 1.6334636427726998}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.15171169386713e-05, "output_power": 0.00012272057629817134, "reference_active": true, "delta_db": 1.7766857472999904}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.519660659304176e-05, "output_power": 2.9883368680519055e-05, "reference_active": true, "delta_db": 0.7408749769860029}], "active_band_count": 3, "active_band_mae_db": 1.3836747890195644}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012452330782545057, "log_rms_mae_db": 1.346598655160742, "rms_envelope_nrmse": 0.22924964866794306, "envelope_correlation": 0.9879650014936765, "rms_gain_error_db": 1.5917690528155493, "reference_mean_power": 0.0019422928952156904, "output_mean_power": 0.002802151595835275}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.257129910319434e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0003707029677887353, "output_power": 0.0005443210706890332, "reference_active": true, "delta_db": 1.6682908326241135}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 4.2635280867150454e-05, "output_power": 6.522343784311792e-05, "reference_active": true, "delta_db": 1.846345578567571}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.869258984918021e-06, "output_power": 1.8200209009668398e-06, "reference_active": true, "delta_db": -0.11593101420613869}], "active_band_count": 3, "active_band_mae_db": 1.2101891417992745}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 0.0001332544033329483, "log_rms_mae_db": 2.1687225784021495, "rms_envelope_nrmse": 0.30623611370597587, "envelope_correlation": 0.9806161866605547, "rms_gain_error_db": 1.9821899306370085, "reference_mean_power": 0.00116782876738011, "output_mean_power": 0.0018433090790952174}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 5.939192180731074e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004134913428254024, "output_power": 0.0006241585953082164, "reference_active": true, "delta_db": 1.7882853431463361}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.995066727507785e-05, "output_power": 0.00011514246304245119, "reference_active": true, "delta_db": 2.1644365364507565}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.9671389178533067e-05, "output_power": 2.5303082240964793e-05, "reference_active": true, "delta_db": 1.0933839650557848}], "active_band_count": 3, "active_band_mae_db": 1.6820352815509592}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 0.00012167319573637772, "log_rms_mae_db": 1.7733782653175585, "rms_envelope_nrmse": 0.2874542077583534, "envelope_correlation": 0.9872218937781846, "rms_gain_error_db": 1.982190483965534, "reference_mean_power": 0.0011747523056415098, "output_mean_power": 0.001854237476466315}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 5.189128511363634e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00037925439414056246, "output_power": 0.0005737460069923408, "reference_active": true, "delta_db": 1.7978905472916331}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.625302583005344e-05, "output_power": 5.990949829590282e-05, "reference_active": true, "delta_db": 2.1815142220006676}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7116807182635842e-06, "output_power": 1.6078928948683887e-06, "reference_active": true, "delta_db": -0.27165642349735164}], "active_band_count": 3, "active_band_mae_db": 1.417020397596551}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 4.762512740172365e-05, "log_rms_mae_db": 1.5755559192509774, "rms_envelope_nrmse": 0.15676129576022707, "envelope_correlation": 0.9889503630547015, "rms_gain_error_db": 0.7877509657642372, "reference_mean_power": 0.00010302415332771722, "output_mean_power": 0.00012351342098939805}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 3.1359809483594164e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.4569632781045495e-05, "output_power": 1.700272660304169e-05, "reference_active": true, "delta_db": 0.6706996582308284}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.842890886644979e-06, "output_power": 4.246974047825141e-06, "reference_active": true, "delta_db": 0.43421553663266244}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.113490153313903e-06, "output_power": 1.298690215097119e-06, "reference_active": true, "delta_db": 0.6681918749484838}], "active_band_count": 3, "active_band_mae_db": 0.5910356899373248}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 4.327669210129085e-05, "log_rms_mae_db": 1.134747938692485, "rms_envelope_nrmse": 0.11564980110385331, "envelope_correlation": 0.9966604254347202, "rms_gain_error_db": 0.7877509597812503, "reference_mean_power": 0.00010363449577271808, "output_mean_power": 0.0001242451471263092}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 2.7762525532534898e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 1.3720292251105881e-05, "output_power": 1.5956593315933514e-05, "reference_active": true, "delta_db": 0.6557681413929268}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.7263230443470765e-06, "output_power": 1.939477586393505e-06, "reference_active": true, "delta_db": 0.5056269714788798}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 9.698769643844551e-08, "output_power": 9.323663628931331e-08, "reference_active": true, "delta_db": -0.17130047464491227}], "active_band_count": 3, "active_band_mae_db": 0.444231862505573}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1018, "rms_floor": 1.6513968336769973e-05, "log_rms_mae_db": 1.5494038876063725, "rms_envelope_nrmse": 0.20043924124037352, "envelope_correlation": 0.9692854914141983, "rms_gain_error_db": 0.3058799838635345, "reference_mean_power": 8.517905641506151e-06, "output_mean_power": 9.139466034328131e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1018, "frequency_resolution_hz": 0.19646365422396858, "power_floor": 4.8245079516023535e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.4329635827636286e-06, "output_power": 2.6708989638448267e-06, "reference_active": true, "delta_db": 0.4052185126868413}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.1930715407234243e-06, "output_power": 1.1838487934370235e-06, "reference_active": true, "delta_db": -0.0337025040170951}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.14650683405709e-07, "output_power": 2.556263665729713e-07, "reference_active": true, "delta_db": -0.9022303226421002}], "active_band_count": 3, "active_band_mae_db": 0.44705044644867886}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 81520, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1012, "rms_floor": 1.2321016663379126e-05, "log_rms_mae_db": 1.0914030685853784, "rms_envelope_nrmse": 0.12344620690300134, "envelope_correlation": 0.9877654848871655, "rms_gain_error_db": 0.3058859279139701, "reference_mean_power": 8.568334809688713e-06, "output_mean_power": 9.193587654082154e-06}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1012, "frequency_resolution_hz": 0.19762845849802368, "power_floor": 4.130413403311084e-12, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.518595705395907e-06, "output_power": 2.692812803916622e-06, "reference_active": true, "delta_db": 0.2904770515744727}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.297939167002702e-07, "output_power": 6.248840900214466e-07, "reference_active": true, "delta_db": -0.03398994015360736}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9684693476821222e-08, "output_power": 2.085871086823172e-08, "reference_active": true, "delta_db": -1.5324510457442864}], "active_band_count": 3, "active_band_mae_db": 0.6189726791574555}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9433514509163576, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "4e0076eb0f6c7dca01b6d25ab253b98e0422e9138dbd0c4b26b1154d87e9f834", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9889503630547015, "standard_deviation": null, "minimum": 0.9889503630547015, "maximum": 0.9889503630547015, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.5755559192509774, "standard_deviation": null, "minimum": 1.5755559192509774, "maximum": 1.5755559192509774, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.5910356899373248, "standard_deviation": null, "minimum": 0.5910356899373248, "maximum": 0.5910356899373248, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.15676129576022707, "standard_deviation": null, "minimum": 0.15676129576022707, "maximum": 0.15676129576022707, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9966604254347202, "standard_deviation": null, "minimum": 0.9966604254347202, "maximum": 0.9966604254347202, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.134747938692485, "standard_deviation": null, "minimum": 1.134747938692485, "maximum": 1.134747938692485, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.444231862505573, "standard_deviation": null, "minimum": 0.444231862505573, "maximum": 0.444231862505573, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.11564980110385331, "standard_deviation": null, "minimum": 0.11564980110385331, "maximum": 0.11564980110385331, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9692854914141983, "standard_deviation": null, "minimum": 0.9692854914141983, "maximum": 0.9692854914141983, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5494038876063725, "standard_deviation": null, "minimum": 1.5494038876063725, "maximum": 1.5494038876063725, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 0.44705044644867886, "standard_deviation": null, "minimum": 0.44705044644867886, "maximum": 0.44705044644867886, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20043924124037352, "standard_deviation": null, "minimum": 0.20043924124037352, "maximum": 0.20043924124037352, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9877654848871655, "standard_deviation": null, "minimum": 0.9877654848871655, "maximum": 0.9877654848871655, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.0914030685853784, "standard_deviation": null, "minimum": 1.0914030685853784, "maximum": 1.0914030685853784, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 0.6189726791574555, "standard_deviation": null, "minimum": 0.6189726791574555, "maximum": 0.6189726791574555, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12344620690300134, "standard_deviation": null, "minimum": 0.12344620690300134, "maximum": 0.12344620690300134, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9806161866605547, "standard_deviation": null, "minimum": 0.9806161866605547, "maximum": 0.9806161866605547, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.1687225784021495, "standard_deviation": null, "minimum": 2.1687225784021495, "maximum": 2.1687225784021495, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.6820352815509592, "standard_deviation": null, "minimum": 1.6820352815509592, "maximum": 1.6820352815509592, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.30623611370597587, "standard_deviation": null, "minimum": 0.30623611370597587, "maximum": 0.30623611370597587, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9872218937781846, "standard_deviation": null, "minimum": 0.9872218937781846, "maximum": 0.9872218937781846, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.7733782653175585, "standard_deviation": null, "minimum": 1.7733782653175585, "maximum": 1.7733782653175585, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.417020397596551, "standard_deviation": null, "minimum": 1.417020397596551, "maximum": 1.417020397596551, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2874542077583534, "standard_deviation": null, "minimum": 0.2874542077583534, "maximum": 0.2874542077583534, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.892767018012869, "standard_deviation": null, "minimum": 0.892767018012869, "maximum": 0.892767018012869, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.803612797244251, "standard_deviation": null, "minimum": 5.803612797244251, "maximum": 5.803612797244251, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.38249752165123496, "standard_deviation": null, "minimum": 0.38249752165123496, "maximum": 0.38249752165123496, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.5917683860714877, "standard_deviation": null, "minimum": 1.5917683860714877, "maximum": 1.5917683860714877, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.980944885182259, "standard_deviation": null, "minimum": 0.980944885182259, "maximum": 0.980944885182259, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.783984169277373, "standard_deviation": null, "minimum": 1.783984169277373, "maximum": 1.783984169277373, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 1.3836747890195644, "standard_deviation": null, "minimum": 1.3836747890195644, "maximum": 1.3836747890195644, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.24775304556080704, "standard_deviation": null, "minimum": 0.24775304556080704, "maximum": 0.24775304556080704, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9879650014936765, "standard_deviation": null, "minimum": 0.9879650014936765, "maximum": 0.9879650014936765, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.346598655160742, "standard_deviation": null, "minimum": 1.346598655160742, "maximum": 1.346598655160742, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.2101891417992745, "standard_deviation": null, "minimum": 1.2101891417992745, "maximum": 1.2101891417992745, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.22924964866794306, "standard_deviation": null, "minimum": 0.22924964866794306, "maximum": 0.22924964866794306, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8555"}, {"case": "librispeech_8555-292519-0001", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.007436990737915, "predicted_mos": 3.007436990737915, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "7067e4fd2e1a138ee9d1b31171150f4bed71cab230d760ec930c4187662db0c7", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.715358799789101}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.5205177182636056, "全帯域_10ms_rms_envelope_nrmse": 0.20175403568561406, "全帯域_10ms_envelope_correlation": 0.9930198256134725, "全帯域_10ms_modulation_mae_db": 0.738813533237355, "全帯域_40ms_log_rms_mae_db": 1.177677402549068, "全帯域_40ms_rms_envelope_nrmse": 0.1949492301770257, "全帯域_40ms_envelope_correlation": 0.9949141433092601, "全帯域_40ms_modulation_mae_db": 0.8628631356574218, "300–1000 Hz_10ms_log_rms_mae_db": 1.3465436722719972, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.18214148592747467, "300–1000 Hz_10ms_envelope_correlation": 0.9850168012138498, "300–1000 Hz_10ms_modulation_mae_db": 0.845742939714068, "300–1000 Hz_40ms_log_rms_mae_db": 1.114755718137367, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.16733742896031445, "300–1000 Hz_40ms_envelope_correlation": 0.9882528844150241, "300–1000 Hz_40ms_modulation_mae_db": 0.7349840429883517, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6359374243736187, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.20737929995395182, "1000–3000 Hz_10ms_envelope_correlation": 0.9832303308949031, "1000–3000 Hz_10ms_modulation_mae_db": 1.2010263753376702, "1000–3000 Hz_40ms_log_rms_mae_db": 1.2765229014856214, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.16809939546497057, "1000–3000 Hz_40ms_envelope_correlation": 0.9920635499425473, "1000–3000 Hz_40ms_modulation_mae_db": 1.1642848678797932, "3000–6000 Hz_10ms_log_rms_mae_db": 1.7311947311171136, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.3061399759847174, "3000–6000 Hz_10ms_envelope_correlation": 0.9594670767009792, "3000–6000 Hz_10ms_modulation_mae_db": 1.5495301603365785, "3000–6000 Hz_40ms_log_rms_mae_db": 1.334024298643551, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.18312673337089674, "3000–6000 Hz_40ms_envelope_correlation": 0.9907869934956324, "3000–6000 Hz_40ms_modulation_mae_db": 1.238865272578706, "mr_stft_db": 5.309227031417108, "band_mae_db": 0.9514812455862258, "rms_gain_error_db": 1.4002418968984176, "output_peak": 0.4642065753997796}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 5.926474390480636, "spectral_convergence": 0.3985858027621418, "spectral_convergence_status": "ok", "amplitude_floor": 1.7050603882585003e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 4.981010317075589, "spectral_convergence": 0.23217793704799364, "spectral_convergence_status": "ok", "amplitude_floor": 1.5194442893884604e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.020196386695099, "spectral_convergence": 0.24702172838114855, "spectral_convergence_status": "ok", "amplitude_floor": 1.5327487413055003e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00047003850181382925, "output_power": 0.0005307297345386302, "delta_db": 0.5273998665006852, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.004513086919531444, "output_power": 0.006442145210208384, "delta_db": 1.5455681200883564, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0003251325154635926, "output_power": 0.00037688916199972775, "delta_db": 0.6415324478381923, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 8.51153215194187e-05, "output_power": 0.00010607362369737303, "delta_db": 0.9559966150094636, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 2.6288307106843834e-05, "output_power": 3.7668481153938394e-05, "delta_db": 1.5621548953110473, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 4.4880972570108e-05, "output_power": 5.0082478399063087e-05, "delta_db": 0.4762355287696099, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.9514812455862258, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.4002418968984176, "pitch": null, "output_rms_dbfs": -21.22421999291719, "output_peak": 0.4642065753997796, "reference_rms_dbfs": -22.624461889815606, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.309227031417108}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00023914198234423542, "log_rms_mae_db": 1.5205177182636056, "rms_envelope_nrmse": 0.20175403568561406, "envelope_correlation": 0.9930198256134725, "rms_gain_error_db": 1.399463338333968, "reference_mean_power": 0.005456771858924342, "output_mean_power": 0.0075315112763410024}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.071388012833875e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00045567084702756786, "output_power": 0.0005811284181890116, "reference_active": true, "delta_db": 1.0562086959371424}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.44246106516196e-05, "output_power": 7.894295807778481e-05, "reference_active": true, "delta_db": 0.8826159326305495}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.7099940640562705e-05, "output_power": 1.8228721085912148e-05, "reference_active": true, "delta_db": 0.277615971144373}], "active_band_count": 3, "active_band_mae_db": 0.738813533237355}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00022747909354312384, "log_rms_mae_db": 1.177677402549068, "rms_envelope_nrmse": 0.1949492301770257, "envelope_correlation": 0.9949141433092601, "rms_gain_error_db": 1.3985318244684963, "reference_mean_power": 0.005416920568028677, "output_mean_power": 0.007474904521984966}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.96418119219355e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.000410225191407457, "output_power": 0.000531518834346661, "reference_active": true, "delta_db": 1.1249633201472644}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.7757195967845804e-05, "output_power": 3.4135163053005374e-05, "reference_active": true, "delta_db": 0.8982639011495788}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0255605769228834e-06, "output_power": 1.1681471037833231e-06, "reference_active": true, "delta_db": 0.5653621856754223}], "active_band_count": 3, "active_band_mae_db": 0.8628631356574218}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.442623236865579e-05, "log_rms_mae_db": 1.3465436722719972, "rms_envelope_nrmse": 0.18214148592747467, "envelope_correlation": 0.9850168012138498, "rms_gain_error_db": 0.9377810907033304, "reference_mean_power": 0.0007990709247977912, "output_mean_power": 0.000991661465850312}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.944736222458957e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00022092122499658314, "output_power": 0.00024564929459430367, "reference_active": true, "delta_db": 0.46078078655647253}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.3665816077667355e-05, "output_power": 3.387991958785096e-05, "reference_active": true, "delta_db": 1.5582088587543388}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.5979447501439136e-06, "output_power": 6.307431245899107e-06, "reference_active": true, "delta_db": 0.5182391738313926}], "active_band_count": 3, "active_band_mae_db": 0.845742939714068}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 8.226666373654229e-05, "log_rms_mae_db": 1.114755718137367, "rms_envelope_nrmse": 0.16733742896031445, "envelope_correlation": 0.9882528844150241, "rms_gain_error_db": 0.9308463999672217, "reference_mean_power": 0.0008012273487091395, "output_mean_power": 0.0009927511638266188}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.73006991688152e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0002092887788100005, "output_power": 0.00023415704690526982, "reference_active": true, "delta_db": 0.48761288405441666}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0995901908054441e-05, "output_power": 1.5655977532735058e-05, "reference_active": true, "delta_db": 1.5344933246561785}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.1412698290887665e-07, "output_power": 3.970534156084766e-07, "reference_active": true, "delta_db": -0.1828459202544601}], "active_band_count": 3, "active_band_mae_db": 0.7349840429883517}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 3.9464514079665034e-05, "log_rms_mae_db": 1.6359374243736187, "rms_envelope_nrmse": 0.20737929995395182, "envelope_correlation": 0.9832303308949031, "rms_gain_error_db": 1.042158278112956, "reference_mean_power": 7.576176061225114e-05, "output_mean_power": 9.630878110309473e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.472428410826198e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.044189308875811e-05, "output_power": 3.9972018449835934e-05, "reference_active": true, "delta_db": 1.1828442280414762}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.3472076435688925e-06, "output_power": 4.328432497038205e-06, "reference_active": true, "delta_db": 1.116479943853018}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.0111506850143103e-06, "output_power": 7.489285470735252e-07, "reference_active": true, "delta_db": -1.3037549541185167}], "active_band_count": 3, "active_band_mae_db": 1.2010263753376702}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 3.608084029396227e-05, "log_rms_mae_db": 1.2765229014856214, "rms_envelope_nrmse": 0.16809939546497057, "envelope_correlation": 0.9920635499425473, "rms_gain_error_db": 1.0423882993546574, "reference_mean_power": 7.611087404608628e-05, "output_mean_power": 9.675770065915118e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 5.164443748057941e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 2.887302600627445e-05, "output_power": 3.7960395392867525e-05, "reference_active": true, "delta_db": 1.1883842533062714}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.617498060844069e-06, "output_power": 1.8461952355377467e-06, "reference_active": true, "delta_db": 0.574338571698477}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 8.602390292027545e-08, "output_power": 5.7757177900558785e-08, "reference_active": true, "delta_db": -1.730131778634631}], "active_band_count": 3, "active_band_mae_db": 1.1642848678797932}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 6.198647979957052e-05, "log_rms_mae_db": 1.7311947311171136, "rms_envelope_nrmse": 0.3061399759847174, "envelope_correlation": 0.9594670767009792, "rms_gain_error_db": 1.0472759151366922, "reference_mean_power": 5.064740774931045e-05, "output_mean_power": 6.445918551206493e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 9.700608322524163e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.136344717386399e-05, "output_power": 5.3132411979103066e-05, "reference_active": true, "delta_db": 1.0874280631698334}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.648317567219929e-06, "output_power": 8.869807652790993e-06, "reference_active": true, "delta_db": 1.2520244596979402}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.888415633999808e-06, "output_power": 6.617386123882276e-06, "reference_active": true, "delta_db": 2.309137958141962}], "active_band_count": 3, "active_band_mae_db": 1.5495301603365785}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 5.0010568231240365e-05, "log_rms_mae_db": 1.334024298643551, "rms_envelope_nrmse": 0.18312673337089674, "envelope_correlation": 0.9907869934956324, "rms_gain_error_db": 1.047541571987147, "reference_mean_power": 5.089623620868825e-05, "output_mean_power": 6.477983306569675e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 9.379392951970345e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.821321246572876e-05, "output_power": 5.218917604186669e-05, "reference_active": true, "delta_db": 1.3536689129911172}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.5895785128125065e-06, "output_power": 4.1427892716408525e-06, "reference_active": true, "delta_db": 0.6224938605740664}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.2481135158983729e-07, "output_power": 3.356305936278388e-07, "reference_active": true, "delta_db": 1.740433044170934}], "active_band_count": 3, "active_band_mae_db": 1.238865272578706}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9584127086010172, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "d0cef45c7534a7812fb919e5fee194e60f8fa510c434db33feb21e0863a2dc94", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9832303308949031, "standard_deviation": null, "minimum": 0.9832303308949031, "maximum": 0.9832303308949031, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6359374243736187, "standard_deviation": null, "minimum": 1.6359374243736187, "maximum": 1.6359374243736187, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 1.2010263753376702, "standard_deviation": null, "minimum": 1.2010263753376702, "maximum": 1.2010263753376702, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.20737929995395182, "standard_deviation": null, "minimum": 0.20737929995395182, "maximum": 0.20737929995395182, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9920635499425473, "standard_deviation": null, "minimum": 0.9920635499425473, "maximum": 0.9920635499425473, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.2765229014856214, "standard_deviation": null, "minimum": 1.2765229014856214, "maximum": 1.2765229014856214, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 1.1642848678797932, "standard_deviation": null, "minimum": 1.1642848678797932, "maximum": 1.1642848678797932, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16809939546497057, "standard_deviation": null, "minimum": 0.16809939546497057, "maximum": 0.16809939546497057, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9594670767009792, "standard_deviation": null, "minimum": 0.9594670767009792, "maximum": 0.9594670767009792, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.7311947311171136, "standard_deviation": null, "minimum": 1.7311947311171136, "maximum": 1.7311947311171136, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.5495301603365785, "standard_deviation": null, "minimum": 1.5495301603365785, "maximum": 1.5495301603365785, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.3061399759847174, "standard_deviation": null, "minimum": 0.3061399759847174, "maximum": 0.3061399759847174, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9907869934956324, "standard_deviation": null, "minimum": 0.9907869934956324, "maximum": 0.9907869934956324, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.334024298643551, "standard_deviation": null, "minimum": 1.334024298643551, "maximum": 1.334024298643551, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.238865272578706, "standard_deviation": null, "minimum": 1.238865272578706, "maximum": 1.238865272578706, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.18312673337089674, "standard_deviation": null, "minimum": 0.18312673337089674, "maximum": 0.18312673337089674, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9850168012138498, "standard_deviation": null, "minimum": 0.9850168012138498, "maximum": 0.9850168012138498, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 1.3465436722719972, "standard_deviation": null, "minimum": 1.3465436722719972, "maximum": 1.3465436722719972, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.845742939714068, "standard_deviation": null, "minimum": 0.845742939714068, "maximum": 0.845742939714068, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18214148592747467, "standard_deviation": null, "minimum": 0.18214148592747467, "maximum": 0.18214148592747467, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9882528844150241, "standard_deviation": null, "minimum": 0.9882528844150241, "maximum": 0.9882528844150241, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.114755718137367, "standard_deviation": null, "minimum": 1.114755718137367, "maximum": 1.114755718137367, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 0.7349840429883517, "standard_deviation": null, "minimum": 0.7349840429883517, "maximum": 0.7349840429883517, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.16733742896031445, "standard_deviation": null, "minimum": 0.16733742896031445, "maximum": 0.16733742896031445, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.9514812455862258, "standard_deviation": null, "minimum": 0.9514812455862258, "maximum": 0.9514812455862258, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.309227031417108, "standard_deviation": null, "minimum": 5.309227031417108, "maximum": 5.309227031417108, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.4642065753997796, "standard_deviation": null, "minimum": 0.4642065753997796, "maximum": 0.4642065753997796, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.4002418968984176, "standard_deviation": null, "minimum": 1.4002418968984176, "maximum": 1.4002418968984176, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9930198256134725, "standard_deviation": null, "minimum": 0.9930198256134725, "maximum": 0.9930198256134725, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.5205177182636056, "standard_deviation": null, "minimum": 1.5205177182636056, "maximum": 1.5205177182636056, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.738813533237355, "standard_deviation": null, "minimum": 0.738813533237355, "maximum": 0.738813533237355, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.20175403568561406, "standard_deviation": null, "minimum": 0.20175403568561406, "maximum": 0.20175403568561406, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9949141433092601, "standard_deviation": null, "minimum": 0.9949141433092601, "maximum": 0.9949141433092601, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.177677402549068, "standard_deviation": null, "minimum": 1.177677402549068, "maximum": 1.177677402549068, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.8628631356574218, "standard_deviation": null, "minimum": 0.8628631356574218, "maximum": 0.8628631356574218, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.1949492301770257, "standard_deviation": null, "minimum": 0.1949492301770257, "maximum": 0.1949492301770257, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_8555"}, {"case": "librispeech_908-157963-0018", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.6532111167907715, "predicted_mos": 3.6532111167907715, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 68080, "input_sample_rate": 16000, "input_seconds": 4.255, "input_float32_sha256": "74b45bb9f83e2f6258026a59a4515a7b05039f52dfcd914e9e356bc75f9b1437", "internal_samples": 68080, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.865496900398284}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 1.7822974367765472, "全帯域_10ms_rms_envelope_nrmse": 0.22190546542619438, "全帯域_10ms_envelope_correlation": 0.9733331899620722, "全帯域_10ms_modulation_mae_db": 0.6175067282708718, "全帯域_40ms_log_rms_mae_db": 1.2093509246904914, "全帯域_40ms_rms_envelope_nrmse": 0.18323657438577168, "全帯域_40ms_envelope_correlation": 0.9862196843337133, "全帯域_40ms_modulation_mae_db": 0.9103917129109659, "300–1000 Hz_10ms_log_rms_mae_db": 2.2932703807923547, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.26870644972306235, "300–1000 Hz_10ms_envelope_correlation": 0.9753541382732982, "300–1000 Hz_10ms_modulation_mae_db": 1.0657518980566663, "300–1000 Hz_40ms_log_rms_mae_db": 1.807105191075618, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.23374804198160223, "300–1000 Hz_40ms_envelope_correlation": 0.9864202736908206, "300–1000 Hz_40ms_modulation_mae_db": 1.06551931183617, "1000–3000 Hz_10ms_log_rms_mae_db": 1.4966666675220412, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.16540447043076592, "1000–3000 Hz_10ms_envelope_correlation": 0.9818107865136483, "1000–3000 Hz_10ms_modulation_mae_db": 0.6637640087940612, "1000–3000 Hz_40ms_log_rms_mae_db": 0.8642887687693223, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.0921956483686506, "1000–3000 Hz_40ms_envelope_correlation": 0.9954433381837292, "1000–3000 Hz_40ms_modulation_mae_db": 0.8206898242674371, "3000–6000 Hz_10ms_log_rms_mae_db": 1.380654180401841, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.22528105921620473, "3000–6000 Hz_10ms_envelope_correlation": 0.9650620090976764, "3000–6000 Hz_10ms_modulation_mae_db": 1.1175685652456524, "3000–6000 Hz_40ms_log_rms_mae_db": 0.7793188887846726, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.12591457901070402, "3000–6000 Hz_40ms_envelope_correlation": 0.9891191990548233, "3000–6000 Hz_40ms_modulation_mae_db": 1.2004981761466509, "mr_stft_db": 5.969440866496261, "band_mae_db": 0.5800179109362701, "rms_gain_error_db": 1.1019287192349572, "output_peak": 0.5799483117919135}, "full_scale": {"sample_count": 68080, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 68080, "output_samples": 68080, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.740078264417789, "spectral_convergence": 0.6888821176953284, "spectral_convergence_status": "ok", "amplitude_floor": 2.0850976744763412e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.357202134532526, "spectral_convergence": 0.3378846769216813, "spectral_convergence_status": "ok", "amplitude_floor": 1.4381271492089952e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.811042200538471, "spectral_convergence": 0.3385439719516781, "spectral_convergence_status": "ok", "amplitude_floor": 9.411979934974602e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0005432476915893821, "output_power": 0.0005715919958868664, "delta_db": 0.2208824909361639, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0030286578008489277, "output_power": 0.0041629071844751925, "delta_db": 1.3814652158656915, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00038180407470981095, "output_power": 0.0004626242234170632, "delta_db": 0.833878100933745, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.0002532984259990213, "output_power": 0.00028319828096637833, "delta_db": 0.484581217422655, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00013717204322211298, "output_power": 0.0001471915368839726, "delta_db": 0.30617232316759735, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 8.29065897621083e-05, "output_power": 7.821251292102488e-05, "delta_db": -0.2531281172917675, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5800179109362701, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.1019287192349572, "pitch": null, "output_rms_dbfs": -22.436891080441743, "output_peak": 0.5799483117919135, "reference_rms_dbfs": -23.538819799676702, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.969440866496261}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 68080, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00026228408676881105, "log_rms_mae_db": 1.7822974367765472, "rms_envelope_nrmse": 0.22190546542619438, "envelope_correlation": 0.9733331899620722, "rms_gain_error_db": 1.101934872630548, "reference_mean_power": 0.004432288013980724, "output_mean_power": 0.0057124374903631285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 1.6030570640082068e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0008005952323264084, "output_power": 0.000987848141203906, "reference_active": true, "delta_db": 0.9127718750711523}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00020547999455019126, "output_power": 0.0002499331892399063, "reference_active": true, "delta_db": 0.8505438544931048}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.989786705158196e-05, "output_power": 3.9086720502988894e-05, "reference_active": true, "delta_db": -0.0892044552483586}], "active_band_count": 3, "active_band_mae_db": 0.6175067282708718}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00021831174911177267, "log_rms_mae_db": 1.2093509246904914, "rms_envelope_nrmse": 0.18323657438577168, "envelope_correlation": 0.9862196843337133, "rms_gain_error_db": 1.1019780854075403, "reference_mean_power": 0.004463740704660867, "output_mean_power": 0.005753031704087342}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 1.4245155353761532e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0007405152195784611, "output_power": 0.000917368937422536, "reference_active": true, "delta_db": 0.9301004174546075}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001021109367695941, "output_power": 0.0001262282709120129, "reference_active": true, "delta_db": 0.9208437280010207}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.434051823082556e-06, "output_power": 1.9874980955091392e-06, "reference_active": true, "delta_db": -0.8802309932772696}], "active_band_count": 3, "active_band_mae_db": 0.9103917129109659}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 0.00023386465365005232, "log_rms_mae_db": 2.2932703807923547, "rms_envelope_nrmse": 0.26870644972306235, "envelope_correlation": 0.9753541382732982, "rms_gain_error_db": 1.5011378128677455, "reference_mean_power": 0.002112021838569365, "output_mean_power": 0.0029840918451277663}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 9.681983603336737e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00048709120906038954, "output_power": 0.0006561575577477188, "reference_active": true, "delta_db": 1.2939784391067597}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00014674692890756196, "output_power": 0.00019555766660576536, "reference_active": true, "delta_db": 1.2470582544000466}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.8361558797573682e-05, "output_power": 2.4384182306098215e-05, "reference_active": true, "delta_db": -0.6562190006631925}], "active_band_count": 3, "active_band_mae_db": 1.0657518980566663}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 0.00020000043006212289, "log_rms_mae_db": 1.807105191075618, "rms_envelope_nrmse": 0.23374804198160223, "envelope_correlation": 0.9864202736908206, "rms_gain_error_db": 1.5011351125677006, "reference_mean_power": 0.002127032769351675, "output_mean_power": 0.0030052990356977827}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.787606965551091e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0004705886462934124, "output_power": 0.0006303786594246886, "reference_active": true, "delta_db": 1.2696005739193148}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 7.123456963821095e-05, "output_power": 9.420976595912232e-05, "reference_active": true, "delta_db": 1.2140512003130206}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5465229502268847e-06, "output_power": 1.3123985840132453e-06, "reference_active": true, "delta_db": -0.7129061612761755}], "active_band_count": 3, "active_band_mae_db": 1.06551931183617}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 7.720600674415526e-05, "log_rms_mae_db": 1.4966666675220412, "rms_envelope_nrmse": 0.16540447043076592, "envelope_correlation": 0.9818107865136483, "rms_gain_error_db": 0.44756925133398395, "reference_mean_power": 0.00033128809071676906, "output_mean_power": 0.0003672507988059747}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 2.2485323271997473e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00014154142624946028, "output_power": 0.00015729804953702246, "reference_active": true, "delta_db": 0.45839770013791536}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.6917189688068767e-05, "output_power": 2.8394607257184038e-05, "reference_active": true, "delta_db": 0.23206151037486678}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.319943953798946e-06, "output_power": 4.68414084664547e-06, "reference_active": true, "delta_db": -1.3008328158694014}], "active_band_count": 3, "active_band_mae_db": 0.6637640087940612}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 6.30102051319373e-05, "log_rms_mae_db": 0.8642887687693223, "rms_envelope_nrmse": 0.0921956483686506, "envelope_correlation": 0.9954433381837292, "rms_gain_error_db": 0.4475687318804871, "reference_mean_power": 0.0003336427865109939, "output_mean_power": 0.00036986106244024674}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 2.017085687613918e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00013578457159768264, "output_power": 0.00014942610311874494, "reference_active": true, "delta_db": 0.41576044220145836}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.310083041117988e-05, "output_power": 1.3820653164741707e-05, "reference_active": true, "delta_db": 0.23229743518851917}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.7328752869332565e-07, "output_power": 2.458342113867934e-07, "reference_active": true, "delta_db": -1.8140115954123337}], "active_band_count": 3, "active_band_mae_db": 0.8206898242674371}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 850, "rms_floor": 5.974138539374114e-05, "log_rms_mae_db": 1.380654180401841, "rms_envelope_nrmse": 0.22528105921620473, "envelope_correlation": 0.9650620090976764, "rms_gain_error_db": -0.26056916928333695, "reference_mean_power": 8.138813397574021e-05, "output_mean_power": 7.664859075075176e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 850, "frequency_resolution_hz": 0.23529411764705882, "power_floor": 8.567729341398199e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 4.0300307346839795e-05, "output_power": 3.840466628041743e-05, "reference_active": true, "delta_db": -0.20924362627204296}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.809515102528064e-06, "output_power": 4.883204728454252e-06, "reference_active": true, "delta_db": -0.754349525730762}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.571956353556118e-06, "output_power": 2.0606056635520373e-06, "reference_active": true, "delta_db": -2.3891125437341523}], "active_band_count": 3, "active_band_mae_db": 1.1175685652456524}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 68080, "uncovered_tail_samples": 0, "metrics": {"frame_count": 844, "rms_floor": 4.362120322552387e-05, "log_rms_mae_db": 0.7793188887846726, "rms_envelope_nrmse": 0.12591457901070402, "envelope_correlation": 0.9891191990548233, "rms_gain_error_db": -0.26057038538785604, "reference_mean_power": 8.196654666810843e-05, "output_mean_power": 7.71932986378556e-05}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 844, "frequency_resolution_hz": 0.23696682464454977, "power_floor": 8.099518469769939e-11, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 3.90961297245674e-05, "output_power": 3.688451153873458e-05, "reference_active": true, "delta_db": -0.252897305772518}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.1532747685680723e-06, "output_power": 2.534295654143633e-06, "reference_active": true, "delta_db": -0.9490453689652525}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.4027458558124955e-07, "output_power": 8.072792858155021e-08, "reference_active": true, "delta_db": -2.3995518537021825}], "active_band_count": 3, "active_band_mae_db": 1.2004981761466509}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8911805860455717, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "04eaed8ecab135db3cb1a6f8a90d326821e253804e4fb8442b8fa43a246e3dd5", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9818107865136483, "standard_deviation": null, "minimum": 0.9818107865136483, "maximum": 0.9818107865136483, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.4966666675220412, "standard_deviation": null, "minimum": 1.4966666675220412, "maximum": 1.4966666675220412, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.6637640087940612, "standard_deviation": null, "minimum": 0.6637640087940612, "maximum": 0.6637640087940612, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.16540447043076592, "standard_deviation": null, "minimum": 0.16540447043076592, "maximum": 0.16540447043076592, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9954433381837292, "standard_deviation": null, "minimum": 0.9954433381837292, "maximum": 0.9954433381837292, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.8642887687693223, "standard_deviation": null, "minimum": 0.8642887687693223, "maximum": 0.8642887687693223, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8206898242674371, "standard_deviation": null, "minimum": 0.8206898242674371, "maximum": 0.8206898242674371, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.0921956483686506, "standard_deviation": null, "minimum": 0.0921956483686506, "maximum": 0.0921956483686506, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9650620090976764, "standard_deviation": null, "minimum": 0.9650620090976764, "maximum": 0.9650620090976764, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.380654180401841, "standard_deviation": null, "minimum": 1.380654180401841, "maximum": 1.380654180401841, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1175685652456524, "standard_deviation": null, "minimum": 1.1175685652456524, "maximum": 1.1175685652456524, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.22528105921620473, "standard_deviation": null, "minimum": 0.22528105921620473, "maximum": 0.22528105921620473, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9891191990548233, "standard_deviation": null, "minimum": 0.9891191990548233, "maximum": 0.9891191990548233, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.7793188887846726, "standard_deviation": null, "minimum": 0.7793188887846726, "maximum": 0.7793188887846726, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2004981761466509, "standard_deviation": null, "minimum": 1.2004981761466509, "maximum": 1.2004981761466509, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.12591457901070402, "standard_deviation": null, "minimum": 0.12591457901070402, "maximum": 0.12591457901070402, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9753541382732982, "standard_deviation": null, "minimum": 0.9753541382732982, "maximum": 0.9753541382732982, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.2932703807923547, "standard_deviation": null, "minimum": 2.2932703807923547, "maximum": 2.2932703807923547, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.0657518980566663, "standard_deviation": null, "minimum": 1.0657518980566663, "maximum": 1.0657518980566663, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26870644972306235, "standard_deviation": null, "minimum": 0.26870644972306235, "maximum": 0.26870644972306235, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9864202736908206, "standard_deviation": null, "minimum": 0.9864202736908206, "maximum": 0.9864202736908206, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 1.807105191075618, "standard_deviation": null, "minimum": 1.807105191075618, "maximum": 1.807105191075618, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.06551931183617, "standard_deviation": null, "minimum": 1.06551931183617, "maximum": 1.06551931183617, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.23374804198160223, "standard_deviation": null, "minimum": 0.23374804198160223, "maximum": 0.23374804198160223, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5800179109362701, "standard_deviation": null, "minimum": 0.5800179109362701, "maximum": 0.5800179109362701, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.969440866496261, "standard_deviation": null, "minimum": 5.969440866496261, "maximum": 5.969440866496261, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5799483117919135, "standard_deviation": null, "minimum": 0.5799483117919135, "maximum": 0.5799483117919135, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.1019287192349572, "standard_deviation": null, "minimum": 1.1019287192349572, "maximum": 1.1019287192349572, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9733331899620722, "standard_deviation": null, "minimum": 0.9733331899620722, "maximum": 0.9733331899620722, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 1.7822974367765472, "standard_deviation": null, "minimum": 1.7822974367765472, "maximum": 1.7822974367765472, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.6175067282708718, "standard_deviation": null, "minimum": 0.6175067282708718, "maximum": 0.6175067282708718, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.22190546542619438, "standard_deviation": null, "minimum": 0.22190546542619438, "maximum": 0.22190546542619438, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9862196843337133, "standard_deviation": null, "minimum": 0.9862196843337133, "maximum": 0.9862196843337133, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.2093509246904914, "standard_deviation": null, "minimum": 1.2093509246904914, "maximum": 1.2093509246904914, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9103917129109659, "standard_deviation": null, "minimum": 0.9103917129109659, "maximum": 0.9103917129109659, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.18323657438577168, "standard_deviation": null, "minimum": 0.18323657438577168, "maximum": 0.18323657438577168, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_908"}, {"case": "librispeech_908-31957-0002", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.475159168243408, "predicted_mos": 3.475159168243408, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 76640, "input_sample_rate": 16000, "input_seconds": 4.79, "input_float32_sha256": "73946d5431ebb4ce815116f42137d0a19e7b1aa736bedd8711776379cbfbff1c", "internal_samples": 76640, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.651145099662244}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.599287864857621, "全帯域_10ms_rms_envelope_nrmse": 0.2367013365808326, "全帯域_10ms_envelope_correlation": 0.9689693385754906, "全帯域_10ms_modulation_mae_db": 0.7736629594629741, "全帯域_40ms_log_rms_mae_db": 1.9104514215771695, "全帯域_40ms_rms_envelope_nrmse": 0.17297098955234855, "全帯域_40ms_envelope_correlation": 0.9901672984910975, "全帯域_40ms_modulation_mae_db": 1.3014967713118997, "300–1000 Hz_10ms_log_rms_mae_db": 2.858751507536637, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.26880550700122025, "300–1000 Hz_10ms_envelope_correlation": 0.973319576977453, "300–1000 Hz_10ms_modulation_mae_db": 1.4367227564231253, "300–1000 Hz_40ms_log_rms_mae_db": 2.330265616767495, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.20749726302903504, "300–1000 Hz_40ms_envelope_correlation": 0.991181747677247, "300–1000 Hz_40ms_modulation_mae_db": 1.6675503304568708, "1000–3000 Hz_10ms_log_rms_mae_db": 1.7129682306158205, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.2031613587727969, "1000–3000 Hz_10ms_envelope_correlation": 0.9746184485186903, "1000–3000 Hz_10ms_modulation_mae_db": 0.4888465519211967, "1000–3000 Hz_40ms_log_rms_mae_db": 0.9243801721489946, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.08879059174070227, "1000–3000 Hz_40ms_envelope_correlation": 0.9962566031241921, "1000–3000 Hz_40ms_modulation_mae_db": 0.5691162945453407, "3000–6000 Hz_10ms_log_rms_mae_db": 1.4259964008955446, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.39068453640662376, "3000–6000 Hz_10ms_envelope_correlation": 0.9255674027812328, "3000–6000 Hz_10ms_modulation_mae_db": 1.1496765944910161, "3000–6000 Hz_40ms_log_rms_mae_db": 0.9829236163535209, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.2358175033884462, "3000–6000 Hz_40ms_envelope_correlation": 0.9750886291470557, "3000–6000 Hz_40ms_modulation_mae_db": 1.0484845108267584, "mr_stft_db": 5.957040239440043, "band_mae_db": 0.7935953644920337, "rms_gain_error_db": 1.107190596227508, "output_peak": 0.5856033752631841}, "full_scale": {"sample_count": 76640, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 76640, "output_samples": 76640, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.764509582158525, "spectral_convergence": 0.679450961388336, "spectral_convergence_status": "ok", "amplitude_floor": 1.4500777065259659e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.371765803646904, "spectral_convergence": 0.3797123055683112, "spectral_convergence_status": "ok", "amplitude_floor": 1.1426990309647766e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.7348453325147, "spectral_convergence": 0.37583556003240015, "spectral_convergence_status": "ok", "amplitude_floor": 7.641152847432082e-07}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.0003890931416418852, "output_power": 0.0004164571757506739, "delta_db": 0.29516773810288044, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.0013826376510160613, "output_power": 0.0017818475022709267, "delta_db": 1.1016215350049858, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.0006763873367355198, "output_power": 0.001011182984870634, "delta_db": 1.7463428496805173, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00010128235718512909, "output_power": 0.00011099527928502932, "delta_db": 0.39770707998177635, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 7.238121331616846e-05, "output_power": 8.098709404837844e-05, "delta_db": 0.4878995718636791, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00017231931162072514, "output_power": 0.00020399395227874978, "delta_db": 0.7328334123183639, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.7935953644920337, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.107190596227508, "pitch": null, "output_rms_dbfs": -24.4303883782929, "output_peak": 0.5856033752631841, "reference_rms_dbfs": -25.537578974520407, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 5.957040239440043}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 76640, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00018226521943209611, "log_rms_mae_db": 2.599287864857621, "rms_envelope_nrmse": 0.2367013365808326, "envelope_correlation": 0.9689693385754906, "rms_gain_error_db": 1.1071983968393426, "reference_mean_power": 0.0027970153309965396, "output_mean_power": 0.003609231063255439}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.029329446119484e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010840604762391139, "output_power": 0.0015098544749346295, "reference_active": true, "delta_db": 1.438815797189657}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001764324474949557, "output_power": 0.00021526002412064512, "reference_active": true, "delta_db": 0.8638492584809264}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.741521666729804e-05, "output_power": 4.761569434494349e-05, "reference_active": true, "delta_db": 0.01832382271833867}], "active_band_count": 3, "active_band_mae_db": 0.7736629594629741}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001676463927845064, "log_rms_mae_db": 1.9104514215771695, "rms_envelope_nrmse": 0.17297098955234855, "envelope_correlation": 0.9901672984910975, "rms_gain_error_db": 1.1072199144051766, "reference_mean_power": 0.0028146416719843875, "output_mean_power": 0.003631993852263739}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.889944424347358e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0010453379933036452, "output_power": 0.0014598507304053416, "reference_active": true, "delta_db": 1.4505171616518664}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.750212405766423e-05, "output_power": 8.65497645477228e-05, "reference_active": true, "delta_db": 1.0794845196006386}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.9228813706071663e-06, "output_power": 2.1299152642489045e-06, "reference_active": true, "delta_db": -1.3744886326831938}], "active_band_count": 3, "active_band_mae_db": 1.3014967713118997}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00017586243391937327, "log_rms_mae_db": 2.858751507536637, "rms_envelope_nrmse": 0.26880550700122025, "envelope_correlation": 0.973319576977453, "rms_gain_error_db": 1.3775349210059173, "reference_mean_power": 0.0014437040494073058, "output_mean_power": 0.0019825843183565285}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 1.5100208608044745e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009242573938618176, "output_power": 0.0013771008106218657, "reference_active": true, "delta_db": 1.7317280041248746}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001371496531392223, "output_power": 0.00018437385874518314, "reference_active": true, "delta_db": 1.2850463151161076}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.2803310723594984e-05, "output_power": 2.4354447558449066e-05, "reference_active": true, "delta_db": -1.2933939500283933}], "active_band_count": 3, "active_band_mae_db": 1.4367227564231253}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 0.0001539617967365963, "log_rms_mae_db": 2.330265616767495, "rms_envelope_nrmse": 0.20749726302903504, "envelope_correlation": 0.991181747677247, "rms_gain_error_db": 1.3775306368135647, "reference_mean_power": 0.0014528118102568712, "output_mean_power": 0.0019950896947100276}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 1.4301118553576922e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0009152950896245062, "output_power": 0.0013510177059748545, "reference_active": true, "delta_db": 1.6909990830881427}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.331274162853222e-05, "output_power": 7.601641512009597e-05, "reference_active": true, "delta_db": 1.5407636786971282}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.0543911334413026e-06, "output_power": 1.366451774812356e-06, "reference_active": true, "delta_db": -1.7708882295853412}], "active_band_count": 3, "active_band_mae_db": 1.6675503304568708}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 8.12616340215947e-05, "log_rms_mae_db": 1.7129682306158205, "rms_envelope_nrmse": 0.2031613587727969, "envelope_correlation": 0.9746184485186903, "rms_gain_error_db": 0.40750592631894755, "reference_mean_power": 0.00015412407568599862, "output_mean_power": 0.00016928601339798115}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.103412885748503e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011785014167056138, "output_power": 0.00012510251154463262, "reference_active": true, "delta_db": 0.2593591969371625}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.031054512020625e-06, "output_power": 8.009992688904906e-06, "reference_active": true, "delta_db": -0.011404540738240128}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.6793453202028428e-06, "output_power": 2.0344691009801813e-06, "reference_active": true, "delta_db": -1.1957759180881875}], "active_band_count": 3, "active_band_mae_db": 0.4888465519211967}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 6.921465297386044e-05, "log_rms_mae_db": 0.9243801721489946, "rms_envelope_nrmse": 0.08879059174070227, "envelope_correlation": 0.9962566031241921, "rms_gain_error_db": 0.40750732257925887, "reference_mean_power": 0.00015509566252518478, "output_mean_power": 0.00017035323474622432}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.0373126402115645e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00011306875179582822, "output_power": 0.00012010639528124551, "reference_active": true, "delta_db": 0.2622353493800251}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 3.441071013913094e-06, "output_power": 3.68726603790611e-06, "reference_active": true, "delta_db": 0.3001083796353337}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.5029543761566567e-07, "output_power": 1.1546364991422062e-07, "reference_active": true, "delta_db": -1.1450051546206634}], "active_band_count": 3, "active_band_mae_db": 0.5691162945453407}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 957, "rms_floor": 0.00013414824565394521, "log_rms_mae_db": 1.4259964008955446, "rms_envelope_nrmse": 0.39068453640662376, "envelope_correlation": 0.9255674027812328, "rms_gain_error_db": 0.7672294051188897, "reference_mean_power": 0.0001730034207748911, "output_mean_power": 0.00020643229031193965}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 957, "frequency_resolution_hz": 0.2089864158829676, "power_floor": 2.1522167661145593e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0001202536722048893, "output_power": 0.00014725285366421923, "reference_active": true, "delta_db": 0.8796537251858955}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 2.5558902161138425e-05, "output_power": 3.5292618297472724e-05, "reference_active": true, "delta_db": 1.4014168358991252}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.530968744127951e-05, "output_power": 2.0033749845217635e-05, "reference_active": true, "delta_db": 1.1679592223880277}], "active_band_count": 3, "active_band_mae_db": 1.1496765944910161}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 76640, "uncovered_tail_samples": 0, "metrics": {"frame_count": 951, "rms_floor": 8.410768197622129e-05, "log_rms_mae_db": 0.9829236163535209, "rms_envelope_nrmse": 0.2358175033884462, "envelope_correlation": 0.9750886291470557, "rms_gain_error_db": 0.7672305450901327, "reference_mean_power": 0.00017409446216015204, "output_mean_power": 0.00020773420448486247}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 951, "frequency_resolution_hz": 0.2103049421661409, "power_floor": 2.085538754133615e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00012977220791713187, "output_power": 0.0001622093545221934, "reference_active": true, "delta_db": 0.9689420242325959}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.3304937813307329e-05, "output_power": 1.8571911026687627e-05, "reference_active": true, "delta_db": 1.4484374538558407}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 1.046486321108266e-06, "output_power": 1.2374879006447013e-06, "reference_active": true, "delta_db": 0.7280740543918384}], "active_band_count": 3, "active_band_mae_db": 1.0484845108267584}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.9092308236787714, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "2494e4c48dde17fa4a35388189e4fcc3f15c7a201c53e724c9dce3d707124cbe", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9746184485186903, "standard_deviation": null, "minimum": 0.9746184485186903, "maximum": 0.9746184485186903, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.7129682306158205, "standard_deviation": null, "minimum": 1.7129682306158205, "maximum": 1.7129682306158205, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.4888465519211967, "standard_deviation": null, "minimum": 0.4888465519211967, "maximum": 0.4888465519211967, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2031613587727969, "standard_deviation": null, "minimum": 0.2031613587727969, "maximum": 0.2031613587727969, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9962566031241921, "standard_deviation": null, "minimum": 0.9962566031241921, "maximum": 0.9962566031241921, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 0.9243801721489946, "standard_deviation": null, "minimum": 0.9243801721489946, "maximum": 0.9243801721489946, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.5691162945453407, "standard_deviation": null, "minimum": 0.5691162945453407, "maximum": 0.5691162945453407, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.08879059174070227, "standard_deviation": null, "minimum": 0.08879059174070227, "maximum": 0.08879059174070227, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9255674027812328, "standard_deviation": null, "minimum": 0.9255674027812328, "maximum": 0.9255674027812328, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.4259964008955446, "standard_deviation": null, "minimum": 1.4259964008955446, "maximum": 1.4259964008955446, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.1496765944910161, "standard_deviation": null, "minimum": 1.1496765944910161, "maximum": 1.1496765944910161, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.39068453640662376, "standard_deviation": null, "minimum": 0.39068453640662376, "maximum": 0.39068453640662376, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9750886291470557, "standard_deviation": null, "minimum": 0.9750886291470557, "maximum": 0.9750886291470557, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 0.9829236163535209, "standard_deviation": null, "minimum": 0.9829236163535209, "maximum": 0.9829236163535209, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.0484845108267584, "standard_deviation": null, "minimum": 1.0484845108267584, "maximum": 1.0484845108267584, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.2358175033884462, "standard_deviation": null, "minimum": 0.2358175033884462, "maximum": 0.2358175033884462, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.973319576977453, "standard_deviation": null, "minimum": 0.973319576977453, "maximum": 0.973319576977453, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.858751507536637, "standard_deviation": null, "minimum": 2.858751507536637, "maximum": 2.858751507536637, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 1.4367227564231253, "standard_deviation": null, "minimum": 1.4367227564231253, "maximum": 1.4367227564231253, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.26880550700122025, "standard_deviation": null, "minimum": 0.26880550700122025, "maximum": 0.26880550700122025, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.991181747677247, "standard_deviation": null, "minimum": 0.991181747677247, "maximum": 0.991181747677247, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.330265616767495, "standard_deviation": null, "minimum": 2.330265616767495, "maximum": 2.330265616767495, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.6675503304568708, "standard_deviation": null, "minimum": 1.6675503304568708, "maximum": 1.6675503304568708, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.20749726302903504, "standard_deviation": null, "minimum": 0.20749726302903504, "maximum": 0.20749726302903504, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.7935953644920337, "standard_deviation": null, "minimum": 0.7935953644920337, "maximum": 0.7935953644920337, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 5.957040239440043, "standard_deviation": null, "minimum": 5.957040239440043, "maximum": 5.957040239440043, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.5856033752631841, "standard_deviation": null, "minimum": 0.5856033752631841, "maximum": 0.5856033752631841, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.107190596227508, "standard_deviation": null, "minimum": 1.107190596227508, "maximum": 1.107190596227508, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9689693385754906, "standard_deviation": null, "minimum": 0.9689693385754906, "maximum": 0.9689693385754906, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.599287864857621, "standard_deviation": null, "minimum": 2.599287864857621, "maximum": 2.599287864857621, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7736629594629741, "standard_deviation": null, "minimum": 0.7736629594629741, "maximum": 0.7736629594629741, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.2367013365808326, "standard_deviation": null, "minimum": 0.2367013365808326, "maximum": 0.2367013365808326, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9901672984910975, "standard_deviation": null, "minimum": 0.9901672984910975, "maximum": 0.9901672984910975, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.9104514215771695, "standard_deviation": null, "minimum": 1.9104514215771695, "maximum": 1.9104514215771695, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 1.3014967713118997, "standard_deviation": null, "minimum": 1.3014967713118997, "maximum": 1.3014967713118997, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.17297098955234855, "standard_deviation": null, "minimum": 0.17297098955234855, "maximum": 0.17297098955234855, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_908"}, {"case": "librispeech_908-31957-0016", "method": "world_harvest", "status": "measured", "mos": {"status": "measured", "value": 3.080683469772339, "predicted_mos": 3.080683469772339, "model": "UTMOSv2", "language": "en", "style": "normal", "domain_validation": "未校正の探索的予測", "extraction_seed": 20261012, "repetitions": 1, "input_samples": 96000, "input_sample_rate": 16000, "input_seconds": 6.0, "input_float32_sha256": "2d7810cd578b733cbf80fe212dce724691a69460f153f61269757d1c84b1bc5e", "internal_samples": 96000, "ssl_processed_samples": 48000, "spectrogram_processed_samples_per_frame": 22400, "ssl_tile_repetitions": 1, "spectrogram_tile_repetitions": 1, "reason": null, "seconds": 5.821125200018287}, "seed_diagnostics": [{"seed": 0, "status": "measured", "values": {"全帯域_10ms_log_rms_mae_db": 2.431871285626157, "全帯域_10ms_rms_envelope_nrmse": 0.266247632701962, "全帯域_10ms_envelope_correlation": 0.9803825720413651, "全帯域_10ms_modulation_mae_db": 0.7914563372673818, "全帯域_40ms_log_rms_mae_db": 1.814962833075714, "全帯域_40ms_rms_envelope_nrmse": 0.23405594711930808, "全帯域_40ms_envelope_correlation": 0.9897727538420816, "全帯域_40ms_modulation_mae_db": 0.9171145845548577, "300–1000 Hz_10ms_log_rms_mae_db": 2.619175689120357, "300–1000 Hz_10ms_rms_envelope_nrmse": 0.2988151295415368, "300–1000 Hz_10ms_envelope_correlation": 0.9846259242993173, "300–1000 Hz_10ms_modulation_mae_db": 0.9723270051482388, "300–1000 Hz_40ms_log_rms_mae_db": 2.1006558063384126, "300–1000 Hz_40ms_rms_envelope_nrmse": 0.27246859791137207, "300–1000 Hz_40ms_envelope_correlation": 0.9919425505533596, "300–1000 Hz_40ms_modulation_mae_db": 1.2639734356993262, "1000–3000 Hz_10ms_log_rms_mae_db": 1.6784415545738454, "1000–3000 Hz_10ms_rms_envelope_nrmse": 0.18470808604302258, "1000–3000 Hz_10ms_envelope_correlation": 0.9791697480613677, "1000–3000 Hz_10ms_modulation_mae_db": 0.9243400877586493, "1000–3000 Hz_40ms_log_rms_mae_db": 1.0105306103097722, "1000–3000 Hz_40ms_rms_envelope_nrmse": 0.10562912771268516, "1000–3000 Hz_40ms_envelope_correlation": 0.9940205135900725, "1000–3000 Hz_40ms_modulation_mae_db": 0.8701733177990739, "3000–6000 Hz_10ms_log_rms_mae_db": 1.5939442546596658, "3000–6000 Hz_10ms_rms_envelope_nrmse": 0.25961404134817584, "3000–6000 Hz_10ms_envelope_correlation": 0.9606797236753232, "3000–6000 Hz_10ms_modulation_mae_db": 1.332441295449329, "3000–6000 Hz_40ms_log_rms_mae_db": 1.06864482482299, "3000–6000 Hz_40ms_rms_envelope_nrmse": 0.13200318092770597, "3000–6000 Hz_40ms_envelope_correlation": 0.9896682697725787, "3000–6000 Hz_40ms_modulation_mae_db": 1.2886663609801738, "mr_stft_db": 6.012100740912978, "band_mae_db": 0.5853400676623898, "rms_gain_error_db": 1.5210648729912122, "output_peak": 0.7985527260762748}, "full_scale": {"sample_count": 96000, "full_scale_amplitude": 1.0, "at_or_above_full_scale_samples": 0, "at_or_above_full_scale_fraction": 0.0, "near_full_scale_amplitude": 0.999, "near_full_scale_fraction": 0.0, "longest_full_scale_run_samples": 0, "meaning": "固定振幅尺度の閾値超過。録音時のクリッピング発生の証明ではない"}}], "measurement": {"status": "ok", "reference_samples": 96000, "output_samples": 96000, "length_delta_samples": 0, "sample_rate": 16000, "reference_finite": true, "output_finite": true, "alignment": "none", "gain_normalization": "none", "mr_stft": [{"window_ms": 5, "window_samples": 80, "hop_samples": 20, "log_magnitude_mae_db": 6.692053780853801, "spectral_convergence": 0.6456037075548051, "spectral_convergence_status": "ok", "amplitude_floor": 2.3135422827537554e-06}, {"window_ms": 20, "window_samples": 320, "hop_samples": 80, "log_magnitude_mae_db": 5.6231198612865425, "spectral_convergence": 0.37659550209405973, "spectral_convergence_status": "ok", "amplitude_floor": 1.7511633455315343e-06}, {"window_ms": 80, "window_samples": 1280, "hop_samples": 320, "log_magnitude_mae_db": 5.721128580598592, "spectral_convergence": 0.3812850073201859, "spectral_convergence_status": "ok", "amplitude_floor": 1.6157176909573571e-06}], "band_energy": {"bands": [{"low_hz": 0.0, "high_hz": 200.0, "reference_power": 0.00044579673236332844, "output_power": 0.0004499572836506465, "delta_db": 0.04034405489636742, "reference_active": true}, {"low_hz": 200.0, "high_hz": 500.0, "reference_power": 0.00284687947566611, "output_power": 0.004502438901486964, "delta_db": 1.990787466939631, "reference_active": true}, {"low_hz": 500.0, "high_hz": 1000.0, "reference_power": 0.00047388905696308834, "output_power": 0.0005647192171806318, "delta_db": 0.7615588676929679, "reference_active": true}, {"low_hz": 1000.0, "high_hz": 2000.0, "reference_power": 0.00013682287067652582, "output_power": 0.00015287484511066577, "delta_db": 0.4817733192366599, "reference_active": true}, {"low_hz": 2000.0, "high_hz": 4000.0, "reference_power": 0.00011121195656671286, "output_power": 0.00011358231604616619, "delta_db": 0.09159238678159187, "reference_active": true}, {"low_hz": 4000.0, "high_hz": 8000.0, "reference_power": 0.00022124255174898303, "output_power": 0.00022880583198141494, "delta_db": 0.1459843104271199, "reference_active": true}], "power_floor": 1e-12, "active_band_count": 6, "band_count": 6, "active_band_mae_db": 0.5853400676623898, "active_policy": "reference power > max(1e-12, total reference power * 1e-6)", "status": "ok"}, "rms_gain_error_db": 1.5210648729912122, "pitch": null, "output_rms_dbfs": -22.209536944378705, "output_peak": 0.7985527260762748, "reference_rms_dbfs": -23.730601817369916, "rms_gain_status": "ok", "mr_stft_log_magnitude_mae_db": 6.012100740912978}, "temporal": {"schema": "temporal-diagnostics-v2", "status": "measured", "sample_rate": 16000, "samples": 96000, "alignment": "none", "normalization": "none", "bands": [{"name": "全帯域", "status": "measured", "filter": {"kind": "none"}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00028051035499693775, "log_rms_mae_db": 2.431871285626157, "rms_envelope_nrmse": 0.266247632701962, "envelope_correlation": 0.9803825720413651, "rms_gain_error_db": 1.521072007086322, "reference_mean_power": 0.004239368434074771, "output_mean_power": 0.0060173928060227085}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 5.0558356273882855e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0018964431225846456, "output_power": 0.0030552034541025696, "reference_active": true, "delta_db": 2.0710031432729585}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.0001936564415781366, "output_power": 0.00020092910381652745, "reference_active": true, "delta_db": 0.16010899744898344}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.246230567335058e-05, "output_power": 6.0435528486997715e-05, "reference_active": true, "delta_db": -0.14325687108020366}], "active_band_count": 3, "active_band_mae_db": 0.7914563372673818}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00026381677718961605, "log_rms_mae_db": 1.814962833075714, "rms_envelope_nrmse": 0.23405594711930808, "envelope_correlation": 0.9897727538420816, "rms_gain_error_db": 1.5210928916815436, "reference_mean_power": 0.004260663575263385, "output_mean_power": 0.006047648378649134}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 4.853150694908015e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017547087963691286, "output_power": 0.002854818925931767, "reference_active": true, "delta_db": 2.113735140258372}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 8.102196929549433e-05, "output_power": 7.931480167195654e-05, "reference_active": true, "delta_db": -0.09248552197408436}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.0378089105429984e-06, "output_power": 2.67946533303886e-06, "reference_active": true, "delta_db": -0.5451230914321168}], "active_band_count": 3, "active_band_mae_db": 0.9171145845548577}}]}, {"name": "300–1000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [300.0, 1000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.0002736722962808348, "log_rms_mae_db": 2.619175689120357, "rms_envelope_nrmse": 0.2988151295415368, "envelope_correlation": 0.9846259242993173, "rms_gain_error_db": 1.8724703623378818, "reference_mean_power": 0.0024581880191069405, "output_mean_power": 0.0037832246776315767}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 4.077905340514435e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0019022535341174436, "output_power": 0.003052307862463563, "reference_active": true, "delta_db": 2.0535993575482827}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 0.00012849688146439824, "output_power": 0.00011929850095703904, "reference_active": true, "delta_db": -0.322576011651836}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 4.095154000220668e-05, "output_power": 3.6156773913806674e-05, "reference_active": true, "delta_db": -0.5408056462445977}], "active_band_count": 3, "active_band_mae_db": 0.9723270051482388}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 0.00025339024948019116, "log_rms_mae_db": 2.1006558063384126, "rms_envelope_nrmse": 0.27246859791137207, "envelope_correlation": 0.9919425505533596, "rms_gain_error_db": 1.8724704729027843, "reference_mean_power": 0.002470549817529662, "output_mean_power": 0.003802249950898262}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 3.981046082661891e-09, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.0017987268874748684, "output_power": 0.002879842295447877, "reference_active": true, "delta_db": 2.0440347921710584}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 5.5079379851283876e-05, "output_power": 4.54725981633921e-05, "reference_active": true, "delta_db": -0.8323927265680378}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.4721075346991203e-06, "output_power": 2.002248972440859e-06, "reference_active": true, "delta_db": -0.9154927883588825}], "active_band_count": 3, "active_band_mae_db": 1.2639734356993262}}]}, {"name": "1000–3000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [1000.0, 3000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 9.470771529403075e-05, "log_rms_mae_db": 1.6784415545738454, "rms_envelope_nrmse": 0.18470808604302258, "envelope_correlation": 0.9791697480613677, "rms_gain_error_db": 0.42420331543751305, "reference_mean_power": 0.00019492290635367423, "output_mean_power": 0.00021492315624179083}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 1.7570632979474236e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.555963644612216e-05, "output_power": 8.398123117128914e-05, "reference_active": true, "delta_db": 0.45892377484529395}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.293127801504832e-05, "output_power": 1.4131144293749765e-05, "reference_active": true, "delta_db": 0.3853588212542258}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 6.505165714035109e-06, "output_power": 4.172387134914859e-06, "reference_active": true, "delta_db": -1.9287376671764282}], "active_band_count": 3, "active_band_mae_db": 0.9243400877586493}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 7.885515861839929e-05, "log_rms_mae_db": 1.0105306103097722, "rms_envelope_nrmse": 0.10562912771268516, "envelope_correlation": 0.9940205135900725, "rms_gain_error_db": 0.4242056321549592, "reference_mean_power": 0.00019590175284412906, "output_mean_power": 0.00021600255343104055}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 1.6226758883463992e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 7.225076970752032e-05, "output_power": 7.919718667737464e-05, "reference_active": true, "delta_db": 0.39867276302351146}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 6.071619770568594e-06, "output_power": 6.937233473194833e-06, "reference_active": true, "delta_db": 0.5788174450515503}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 3.6400879277669946e-07, "output_power": 2.4992453905680516e-07, "reference_active": true, "delta_db": -1.6330297453221598}], "active_band_count": 3, "active_band_mae_db": 0.8701733177990739}}]}, {"name": "3000–6000 Hz", "status": "measured", "filter": {"kind": "butterworth-forward-backward", "prototype_order": 4, "edges_hz": [3000.0, 6000.0], "pad_type": "even", "pad_samples": 27}, "windows": [{"window_ms": 10.0, "window_samples": 160, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1199, "rms_floor": 0.00012408236865898127, "log_rms_mae_db": 1.5939442546596658, "rms_envelope_nrmse": 0.25961404134817584, "envelope_correlation": 0.9606797236753232, "rms_gain_error_db": -0.05389329451030983, "reference_mean_power": 0.00018911407706020183, "output_mean_power": 0.00018678178786101114}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1199, "frequency_resolution_hz": 0.16680567139282734, "power_floor": 2.8583468534278485e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010082037892943086, "output_power": 9.214536819452572e-05, "reference_active": true, "delta_db": -0.3907481565103524}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.9440078487669303e-05, "output_power": 2.5241161323459726e-05, "reference_active": true, "delta_db": 1.1341131851498867}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 5.009136082695238e-06, "output_power": 8.851341252417328e-06, "reference_active": true, "delta_db": 2.4724625446877475}], "active_band_count": 3, "active_band_mae_db": 1.332441295449329}}, {"window_ms": 40.0, "window_samples": 640, "hop_samples": 80, "hop_ms": 5.0, "status": "measured", "covered_start_sample": 0, "covered_end_sample": 96000, "uncovered_tail_samples": 0, "metrics": {"frame_count": 1193, "rms_floor": 9.970743476594613e-05, "log_rms_mae_db": 1.06864482482299, "rms_envelope_nrmse": 0.13200318092770597, "envelope_correlation": 0.9896682697725787, "rms_gain_error_db": -0.053891170749772135, "reference_mean_power": 0.00019006483009233416, "output_mean_power": 0.00018772090732918078}, "modulation_bands": {"status": "measured", "envelope_rate_hz": 200.0, "frames": 1193, "frequency_resolution_hz": 0.16764459346186086, "power_floor": 2.7984096251410197e-10, "window": "hann", "detrend": "constant", "gain_normalization": "none", "bands": [{"low_hz": 2.0, "high_hz": 8.0, "reference_power": 0.00010012152579842322, "output_power": 9.45867014077349e-05, "reference_active": true, "delta_db": -0.24697379144155038}, {"low_hz": 8.0, "high_hz": 20.0, "reference_power": 1.0693666398121268e-05, "output_power": 1.3804962672064767e-05, "reference_active": true, "delta_db": 1.1090860503297097}, {"low_hz": 20.0, "high_hz": 50.0, "reference_power": 2.760055744312906e-07, "output_power": 4.919395931636145e-07, "reference_active": true, "delta_db": 2.509939241169261}], "active_band_count": 3, "active_band_mae_db": 1.2886663609801738}}]}], "event": null, "silent_intervals": null, "limitations": ["有限長雑音の包絡差は独立seedでもゼロにならない", "帯域フィルタの端点過渡は測定に含むがイベント測定には使わない", "RMS包絡は有声・無声や知覚品質の正解測定ではない"]}, "estoi": {"value": 0.8688847109847203, "library": "pystoi", "version": "0.4.1", "extended": true, "reference_required": true, "internal_sample_rate": 10000, "evaluation_seed": 0, "internal_silence_removal_db": 40, "warnings": [], "normalization": "公式実装の行・列正規化", "scope": "明瞭性の推定。MOS・自然性・話者類似度ではない", "status": "measured"}, "waveform_sha256": "5da310be931cb854801accc942ebcfa9319a1a735009bdc00e6ef96530de9ae1", "seed_summary": {"1000–3000 Hz_10ms_envelope_correlation": {"mean": 0.9791697480613677, "standard_deviation": null, "minimum": 0.9791697480613677, "maximum": 0.9791697480613677, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_log_rms_mae_db": {"mean": 1.6784415545738454, "standard_deviation": null, "minimum": 1.6784415545738454, "maximum": 1.6784415545738454, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_modulation_mae_db": {"mean": 0.9243400877586493, "standard_deviation": null, "minimum": 0.9243400877586493, "maximum": 0.9243400877586493, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.18470808604302258, "standard_deviation": null, "minimum": 0.18470808604302258, "maximum": 0.18470808604302258, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_envelope_correlation": {"mean": 0.9940205135900725, "standard_deviation": null, "minimum": 0.9940205135900725, "maximum": 0.9940205135900725, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_log_rms_mae_db": {"mean": 1.0105306103097722, "standard_deviation": null, "minimum": 1.0105306103097722, "maximum": 1.0105306103097722, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_modulation_mae_db": {"mean": 0.8701733177990739, "standard_deviation": null, "minimum": 0.8701733177990739, "maximum": 0.8701733177990739, "valid_count": 1, "planned_count": 1}, "1000–3000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.10562912771268516, "standard_deviation": null, "minimum": 0.10562912771268516, "maximum": 0.10562912771268516, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_envelope_correlation": {"mean": 0.9606797236753232, "standard_deviation": null, "minimum": 0.9606797236753232, "maximum": 0.9606797236753232, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_log_rms_mae_db": {"mean": 1.5939442546596658, "standard_deviation": null, "minimum": 1.5939442546596658, "maximum": 1.5939442546596658, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_modulation_mae_db": {"mean": 1.332441295449329, "standard_deviation": null, "minimum": 1.332441295449329, "maximum": 1.332441295449329, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.25961404134817584, "standard_deviation": null, "minimum": 0.25961404134817584, "maximum": 0.25961404134817584, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_envelope_correlation": {"mean": 0.9896682697725787, "standard_deviation": null, "minimum": 0.9896682697725787, "maximum": 0.9896682697725787, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_log_rms_mae_db": {"mean": 1.06864482482299, "standard_deviation": null, "minimum": 1.06864482482299, "maximum": 1.06864482482299, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_modulation_mae_db": {"mean": 1.2886663609801738, "standard_deviation": null, "minimum": 1.2886663609801738, "maximum": 1.2886663609801738, "valid_count": 1, "planned_count": 1}, "3000–6000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.13200318092770597, "standard_deviation": null, "minimum": 0.13200318092770597, "maximum": 0.13200318092770597, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_envelope_correlation": {"mean": 0.9846259242993173, "standard_deviation": null, "minimum": 0.9846259242993173, "maximum": 0.9846259242993173, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_log_rms_mae_db": {"mean": 2.619175689120357, "standard_deviation": null, "minimum": 2.619175689120357, "maximum": 2.619175689120357, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_modulation_mae_db": {"mean": 0.9723270051482388, "standard_deviation": null, "minimum": 0.9723270051482388, "maximum": 0.9723270051482388, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_10ms_rms_envelope_nrmse": {"mean": 0.2988151295415368, "standard_deviation": null, "minimum": 0.2988151295415368, "maximum": 0.2988151295415368, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_envelope_correlation": {"mean": 0.9919425505533596, "standard_deviation": null, "minimum": 0.9919425505533596, "maximum": 0.9919425505533596, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_log_rms_mae_db": {"mean": 2.1006558063384126, "standard_deviation": null, "minimum": 2.1006558063384126, "maximum": 2.1006558063384126, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_modulation_mae_db": {"mean": 1.2639734356993262, "standard_deviation": null, "minimum": 1.2639734356993262, "maximum": 1.2639734356993262, "valid_count": 1, "planned_count": 1}, "300–1000 Hz_40ms_rms_envelope_nrmse": {"mean": 0.27246859791137207, "standard_deviation": null, "minimum": 0.27246859791137207, "maximum": 0.27246859791137207, "valid_count": 1, "planned_count": 1}, "band_mae_db": {"mean": 0.5853400676623898, "standard_deviation": null, "minimum": 0.5853400676623898, "maximum": 0.5853400676623898, "valid_count": 1, "planned_count": 1}, "mr_stft_db": {"mean": 6.012100740912978, "standard_deviation": null, "minimum": 6.012100740912978, "maximum": 6.012100740912978, "valid_count": 1, "planned_count": 1}, "output_peak": {"mean": 0.7985527260762748, "standard_deviation": null, "minimum": 0.7985527260762748, "maximum": 0.7985527260762748, "valid_count": 1, "planned_count": 1}, "rms_gain_error_db": {"mean": 1.5210648729912122, "standard_deviation": null, "minimum": 1.5210648729912122, "maximum": 1.5210648729912122, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_envelope_correlation": {"mean": 0.9803825720413651, "standard_deviation": null, "minimum": 0.9803825720413651, "maximum": 0.9803825720413651, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_log_rms_mae_db": {"mean": 2.431871285626157, "standard_deviation": null, "minimum": 2.431871285626157, "maximum": 2.431871285626157, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_modulation_mae_db": {"mean": 0.7914563372673818, "standard_deviation": null, "minimum": 0.7914563372673818, "maximum": 0.7914563372673818, "valid_count": 1, "planned_count": 1}, "全帯域_10ms_rms_envelope_nrmse": {"mean": 0.266247632701962, "standard_deviation": null, "minimum": 0.266247632701962, "maximum": 0.266247632701962, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_envelope_correlation": {"mean": 0.9897727538420816, "standard_deviation": null, "minimum": 0.9897727538420816, "maximum": 0.9897727538420816, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_log_rms_mae_db": {"mean": 1.814962833075714, "standard_deviation": null, "minimum": 1.814962833075714, "maximum": 1.814962833075714, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_modulation_mae_db": {"mean": 0.9171145845548577, "standard_deviation": null, "minimum": 0.9171145845548577, "maximum": 0.9171145845548577, "valid_count": 1, "planned_count": 1}, "全帯域_40ms_rms_envelope_nrmse": {"mean": 0.23405594711930808, "standard_deviation": null, "minimum": 0.23405594711930808, "maximum": 0.23405594711930808, "valid_count": 1, "planned_count": 1}}, "speaker": "librispeech_908"}]}}
